diff options
author | Daniel Vetter <daniel.vetter@ffwll.ch> | 2015-01-13 01:07:46 +0300 |
---|---|---|
committer | Daniel Vetter <daniel.vetter@ffwll.ch> | 2015-01-13 01:07:46 +0300 |
commit | 0a87a2db485a1456b7427914969c0e8195a1bbda (patch) | |
tree | 8d0186672af22c6ee76118c471881cd66a36502d /drivers/gpu | |
parent | 7226572d8ed48f7e1aa9de5383d919490d6e9a0c (diff) | |
parent | fcf3aac5fc307f0cae429f5844ddc25761662858 (diff) | |
download | linux-0a87a2db485a1456b7427914969c0e8195a1bbda.tar.xz |
Merge tag 'topic/i915-hda-componentized-2015-01-12' into drm-intel-next-queued
Conflicts:
drivers/gpu/drm/i915/intel_runtime_pm.c
Separate branch so that Takashi can also pull just this refactoring
into sound-next.
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Diffstat (limited to 'drivers/gpu')
163 files changed, 7154 insertions, 1134 deletions
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig index 24c2d7caedd5..c3413b6adb17 100644 --- a/drivers/gpu/drm/Kconfig +++ b/drivers/gpu/drm/Kconfig @@ -167,6 +167,8 @@ config DRM_SAVAGE source "drivers/gpu/drm/exynos/Kconfig" +source "drivers/gpu/drm/rockchip/Kconfig" + source "drivers/gpu/drm/vmwgfx/Kconfig" source "drivers/gpu/drm/gma500/Kconfig" diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile index 47d89869c5df..e620807418ea 100644 --- a/drivers/gpu/drm/Makefile +++ b/drivers/gpu/drm/Makefile @@ -37,6 +37,7 @@ obj-$(CONFIG_DRM_MIPI_DSI) += drm_mipi_dsi.o obj-$(CONFIG_DRM_TTM) += ttm/ obj-$(CONFIG_DRM_TDFX) += tdfx/ obj-$(CONFIG_DRM_R128) += r128/ +obj-$(CONFIG_HSA_AMD) += amd/amdkfd/ obj-$(CONFIG_DRM_RADEON)+= radeon/ obj-$(CONFIG_DRM_MGA) += mga/ obj-$(CONFIG_DRM_I810) += i810/ @@ -49,6 +50,7 @@ obj-$(CONFIG_DRM_VMWGFX)+= vmwgfx/ obj-$(CONFIG_DRM_VIA) +=via/ obj-$(CONFIG_DRM_NOUVEAU) +=nouveau/ obj-$(CONFIG_DRM_EXYNOS) +=exynos/ +obj-$(CONFIG_DRM_ROCKCHIP) +=rockchip/ obj-$(CONFIG_DRM_GMA500) += gma500/ obj-$(CONFIG_DRM_UDL) += udl/ obj-$(CONFIG_DRM_AST) += ast/ @@ -66,4 +68,3 @@ obj-$(CONFIG_DRM_IMX) += imx/ obj-y += i2c/ obj-y += panel/ obj-y += bridge/ -obj-$(CONFIG_HSA_AMD) += amd/amdkfd/ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index 102cd36799b1..fcfdf23e1913 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -31,7 +31,6 @@ #include <uapi/linux/kfd_ioctl.h> #include <linux/time.h> #include <linux/mm.h> -#include <linux/uaccess.h> #include <uapi/asm-generic/mman-common.h> #include <asm/processor.h> #include "kfd_priv.h" @@ -102,35 +101,39 @@ struct device *kfd_chardev(void) static int kfd_open(struct inode *inode, struct file *filep) { struct kfd_process *process; + bool is_32bit_user_mode; if (iminor(inode) != 0) return -ENODEV; + is_32bit_user_mode = is_compat_task(); + + if (is_32bit_user_mode == true) { + dev_warn(kfd_device, + "Process %d (32-bit) failed to open /dev/kfd\n" + "32-bit processes are not supported by amdkfd\n", + current->pid); + return -EPERM; + } + process = kfd_create_process(current); if (IS_ERR(process)) return PTR_ERR(process); - process->is_32bit_user_mode = is_compat_task(); - dev_dbg(kfd_device, "process %d opened, compat mode (32 bit) - %d\n", process->pasid, process->is_32bit_user_mode); - kfd_init_apertures(process); - return 0; } -static long kfd_ioctl_get_version(struct file *filep, struct kfd_process *p, - void __user *arg) +static int kfd_ioctl_get_version(struct file *filep, struct kfd_process *p, + void *data) { - struct kfd_ioctl_get_version_args args; + struct kfd_ioctl_get_version_args *args = data; int err = 0; - args.major_version = KFD_IOCTL_MAJOR_VERSION; - args.minor_version = KFD_IOCTL_MINOR_VERSION; - - if (copy_to_user(arg, &args, sizeof(args))) - err = -EFAULT; + args->major_version = KFD_IOCTL_MAJOR_VERSION; + args->minor_version = KFD_IOCTL_MINOR_VERSION; return err; } @@ -214,10 +217,10 @@ static int set_queue_properties_from_user(struct queue_properties *q_properties, return 0; } -static long kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, - void __user *arg) +static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, + void *data) { - struct kfd_ioctl_create_queue_args args; + struct kfd_ioctl_create_queue_args *args = data; struct kfd_dev *dev; int err = 0; unsigned int queue_id; @@ -226,16 +229,13 @@ static long kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, memset(&q_properties, 0, sizeof(struct queue_properties)); - if (copy_from_user(&args, arg, sizeof(args))) - return -EFAULT; - pr_debug("kfd: creating queue ioctl\n"); - err = set_queue_properties_from_user(&q_properties, &args); + err = set_queue_properties_from_user(&q_properties, args); if (err) return err; - dev = kfd_device_by_id(args.gpu_id); + dev = kfd_device_by_id(args->gpu_id); if (dev == NULL) return -EINVAL; @@ -243,7 +243,7 @@ static long kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, pdd = kfd_bind_process_to_device(dev, p); if (IS_ERR(pdd)) { - err = PTR_ERR(pdd); + err = -ESRCH; goto err_bind_process; } @@ -256,33 +256,26 @@ static long kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, if (err != 0) goto err_create_queue; - args.queue_id = queue_id; + args->queue_id = queue_id; /* Return gpu_id as doorbell offset for mmap usage */ - args.doorbell_offset = args.gpu_id << PAGE_SHIFT; - - if (copy_to_user(arg, &args, sizeof(args))) { - err = -EFAULT; - goto err_copy_args_out; - } + args->doorbell_offset = args->gpu_id << PAGE_SHIFT; mutex_unlock(&p->mutex); - pr_debug("kfd: queue id %d was created successfully\n", args.queue_id); + pr_debug("kfd: queue id %d was created successfully\n", args->queue_id); pr_debug("ring buffer address == 0x%016llX\n", - args.ring_base_address); + args->ring_base_address); pr_debug("read ptr address == 0x%016llX\n", - args.read_pointer_address); + args->read_pointer_address); pr_debug("write ptr address == 0x%016llX\n", - args.write_pointer_address); + args->write_pointer_address); return 0; -err_copy_args_out: - pqm_destroy_queue(&p->pqm, queue_id); err_create_queue: err_bind_process: mutex_unlock(&p->mutex); @@ -290,99 +283,90 @@ err_bind_process: } static int kfd_ioctl_destroy_queue(struct file *filp, struct kfd_process *p, - void __user *arg) + void *data) { int retval; - struct kfd_ioctl_destroy_queue_args args; - - if (copy_from_user(&args, arg, sizeof(args))) - return -EFAULT; + struct kfd_ioctl_destroy_queue_args *args = data; pr_debug("kfd: destroying queue id %d for PASID %d\n", - args.queue_id, + args->queue_id, p->pasid); mutex_lock(&p->mutex); - retval = pqm_destroy_queue(&p->pqm, args.queue_id); + retval = pqm_destroy_queue(&p->pqm, args->queue_id); mutex_unlock(&p->mutex); return retval; } static int kfd_ioctl_update_queue(struct file *filp, struct kfd_process *p, - void __user *arg) + void *data) { int retval; - struct kfd_ioctl_update_queue_args args; + struct kfd_ioctl_update_queue_args *args = data; struct queue_properties properties; - if (copy_from_user(&args, arg, sizeof(args))) - return -EFAULT; - - if (args.queue_percentage > KFD_MAX_QUEUE_PERCENTAGE) { + if (args->queue_percentage > KFD_MAX_QUEUE_PERCENTAGE) { pr_err("kfd: queue percentage must be between 0 to KFD_MAX_QUEUE_PERCENTAGE\n"); return -EINVAL; } - if (args.queue_priority > KFD_MAX_QUEUE_PRIORITY) { + if (args->queue_priority > KFD_MAX_QUEUE_PRIORITY) { pr_err("kfd: queue priority must be between 0 to KFD_MAX_QUEUE_PRIORITY\n"); return -EINVAL; } - if ((args.ring_base_address) && + if ((args->ring_base_address) && (!access_ok(VERIFY_WRITE, - (const void __user *) args.ring_base_address, + (const void __user *) args->ring_base_address, sizeof(uint64_t)))) { pr_err("kfd: can't access ring base address\n"); return -EFAULT; } - if (!is_power_of_2(args.ring_size) && (args.ring_size != 0)) { + if (!is_power_of_2(args->ring_size) && (args->ring_size != 0)) { pr_err("kfd: ring size must be a power of 2 or 0\n"); return -EINVAL; } - properties.queue_address = args.ring_base_address; - properties.queue_size = args.ring_size; - properties.queue_percent = args.queue_percentage; - properties.priority = args.queue_priority; + properties.queue_address = args->ring_base_address; + properties.queue_size = args->ring_size; + properties.queue_percent = args->queue_percentage; + properties.priority = args->queue_priority; pr_debug("kfd: updating queue id %d for PASID %d\n", - args.queue_id, p->pasid); + args->queue_id, p->pasid); mutex_lock(&p->mutex); - retval = pqm_update_queue(&p->pqm, args.queue_id, &properties); + retval = pqm_update_queue(&p->pqm, args->queue_id, &properties); mutex_unlock(&p->mutex); return retval; } -static long kfd_ioctl_set_memory_policy(struct file *filep, - struct kfd_process *p, void __user *arg) +static int kfd_ioctl_set_memory_policy(struct file *filep, + struct kfd_process *p, void *data) { - struct kfd_ioctl_set_memory_policy_args args; + struct kfd_ioctl_set_memory_policy_args *args = data; struct kfd_dev *dev; int err = 0; struct kfd_process_device *pdd; enum cache_policy default_policy, alternate_policy; - if (copy_from_user(&args, arg, sizeof(args))) - return -EFAULT; - - if (args.default_policy != KFD_IOC_CACHE_POLICY_COHERENT - && args.default_policy != KFD_IOC_CACHE_POLICY_NONCOHERENT) { + if (args->default_policy != KFD_IOC_CACHE_POLICY_COHERENT + && args->default_policy != KFD_IOC_CACHE_POLICY_NONCOHERENT) { return -EINVAL; } - if (args.alternate_policy != KFD_IOC_CACHE_POLICY_COHERENT - && args.alternate_policy != KFD_IOC_CACHE_POLICY_NONCOHERENT) { + if (args->alternate_policy != KFD_IOC_CACHE_POLICY_COHERENT + && args->alternate_policy != KFD_IOC_CACHE_POLICY_NONCOHERENT) { return -EINVAL; } - dev = kfd_device_by_id(args.gpu_id); + dev = kfd_device_by_id(args->gpu_id); if (dev == NULL) return -EINVAL; @@ -390,23 +374,23 @@ static long kfd_ioctl_set_memory_policy(struct file *filep, pdd = kfd_bind_process_to_device(dev, p); if (IS_ERR(pdd)) { - err = PTR_ERR(pdd); + err = -ESRCH; goto out; } - default_policy = (args.default_policy == KFD_IOC_CACHE_POLICY_COHERENT) + default_policy = (args->default_policy == KFD_IOC_CACHE_POLICY_COHERENT) ? cache_policy_coherent : cache_policy_noncoherent; alternate_policy = - (args.alternate_policy == KFD_IOC_CACHE_POLICY_COHERENT) + (args->alternate_policy == KFD_IOC_CACHE_POLICY_COHERENT) ? cache_policy_coherent : cache_policy_noncoherent; if (!dev->dqm->set_cache_memory_policy(dev->dqm, &pdd->qpd, default_policy, alternate_policy, - (void __user *)args.alternate_aperture_base, - args.alternate_aperture_size)) + (void __user *)args->alternate_aperture_base, + args->alternate_aperture_size)) err = -EINVAL; out: @@ -415,53 +399,44 @@ out: return err; } -static long kfd_ioctl_get_clock_counters(struct file *filep, - struct kfd_process *p, void __user *arg) +static int kfd_ioctl_get_clock_counters(struct file *filep, + struct kfd_process *p, void *data) { - struct kfd_ioctl_get_clock_counters_args args; + struct kfd_ioctl_get_clock_counters_args *args = data; struct kfd_dev *dev; struct timespec time; - if (copy_from_user(&args, arg, sizeof(args))) - return -EFAULT; - - dev = kfd_device_by_id(args.gpu_id); + dev = kfd_device_by_id(args->gpu_id); if (dev == NULL) return -EINVAL; /* Reading GPU clock counter from KGD */ - args.gpu_clock_counter = kfd2kgd->get_gpu_clock_counter(dev->kgd); + args->gpu_clock_counter = kfd2kgd->get_gpu_clock_counter(dev->kgd); /* No access to rdtsc. Using raw monotonic time */ getrawmonotonic(&time); - args.cpu_clock_counter = (uint64_t)timespec_to_ns(&time); + args->cpu_clock_counter = (uint64_t)timespec_to_ns(&time); get_monotonic_boottime(&time); - args.system_clock_counter = (uint64_t)timespec_to_ns(&time); + args->system_clock_counter = (uint64_t)timespec_to_ns(&time); /* Since the counter is in nano-seconds we use 1GHz frequency */ - args.system_clock_freq = 1000000000; - - if (copy_to_user(arg, &args, sizeof(args))) - return -EFAULT; + args->system_clock_freq = 1000000000; return 0; } static int kfd_ioctl_get_process_apertures(struct file *filp, - struct kfd_process *p, void __user *arg) + struct kfd_process *p, void *data) { - struct kfd_ioctl_get_process_apertures_args args; + struct kfd_ioctl_get_process_apertures_args *args = data; struct kfd_process_device_apertures *pAperture; struct kfd_process_device *pdd; dev_dbg(kfd_device, "get apertures for PASID %d", p->pasid); - if (copy_from_user(&args, arg, sizeof(args))) - return -EFAULT; - - args.num_of_nodes = 0; + args->num_of_nodes = 0; mutex_lock(&p->mutex); @@ -470,7 +445,8 @@ static int kfd_ioctl_get_process_apertures(struct file *filp, /* Run over all pdd of the process */ pdd = kfd_get_first_process_device_data(p); do { - pAperture = &args.process_apertures[args.num_of_nodes]; + pAperture = + &args->process_apertures[args->num_of_nodes]; pAperture->gpu_id = pdd->dev->id; pAperture->lds_base = pdd->lds_base; pAperture->lds_limit = pdd->lds_limit; @@ -480,7 +456,7 @@ static int kfd_ioctl_get_process_apertures(struct file *filp, pAperture->scratch_limit = pdd->scratch_limit; dev_dbg(kfd_device, - "node id %u\n", args.num_of_nodes); + "node id %u\n", args->num_of_nodes); dev_dbg(kfd_device, "gpu id %u\n", pdd->dev->id); dev_dbg(kfd_device, @@ -496,80 +472,131 @@ static int kfd_ioctl_get_process_apertures(struct file *filp, dev_dbg(kfd_device, "scratch_limit %llX\n", pdd->scratch_limit); - args.num_of_nodes++; + args->num_of_nodes++; } while ((pdd = kfd_get_next_process_device_data(p, pdd)) != NULL && - (args.num_of_nodes < NUM_OF_SUPPORTED_GPUS)); + (args->num_of_nodes < NUM_OF_SUPPORTED_GPUS)); } mutex_unlock(&p->mutex); - if (copy_to_user(arg, &args, sizeof(args))) - return -EFAULT; - return 0; } +#define AMDKFD_IOCTL_DEF(ioctl, _func, _flags) \ + [_IOC_NR(ioctl)] = {.cmd = ioctl, .func = _func, .flags = _flags, .cmd_drv = 0, .name = #ioctl} + +/** Ioctl table */ +static const struct amdkfd_ioctl_desc amdkfd_ioctls[] = { + AMDKFD_IOCTL_DEF(AMDKFD_IOC_GET_VERSION, + kfd_ioctl_get_version, 0), + + AMDKFD_IOCTL_DEF(AMDKFD_IOC_CREATE_QUEUE, + kfd_ioctl_create_queue, 0), + + AMDKFD_IOCTL_DEF(AMDKFD_IOC_DESTROY_QUEUE, + kfd_ioctl_destroy_queue, 0), + + AMDKFD_IOCTL_DEF(AMDKFD_IOC_SET_MEMORY_POLICY, + kfd_ioctl_set_memory_policy, 0), + + AMDKFD_IOCTL_DEF(AMDKFD_IOC_GET_CLOCK_COUNTERS, + kfd_ioctl_get_clock_counters, 0), + + AMDKFD_IOCTL_DEF(AMDKFD_IOC_GET_PROCESS_APERTURES, + kfd_ioctl_get_process_apertures, 0), + + AMDKFD_IOCTL_DEF(AMDKFD_IOC_UPDATE_QUEUE, + kfd_ioctl_update_queue, 0), +}; + +#define AMDKFD_CORE_IOCTL_COUNT ARRAY_SIZE(amdkfd_ioctls) + static long kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg) { struct kfd_process *process; - long err = -EINVAL; + amdkfd_ioctl_t *func; + const struct amdkfd_ioctl_desc *ioctl = NULL; + unsigned int nr = _IOC_NR(cmd); + char stack_kdata[128]; + char *kdata = NULL; + unsigned int usize, asize; + int retcode = -EINVAL; + + if (nr >= AMDKFD_CORE_IOCTL_COUNT) + goto err_i1; + + if ((nr >= AMDKFD_COMMAND_START) && (nr < AMDKFD_COMMAND_END)) { + u32 amdkfd_size; + + ioctl = &amdkfd_ioctls[nr]; + + amdkfd_size = _IOC_SIZE(ioctl->cmd); + usize = asize = _IOC_SIZE(cmd); + if (amdkfd_size > asize) + asize = amdkfd_size; + + cmd = ioctl->cmd; + } else + goto err_i1; - dev_dbg(kfd_device, - "ioctl cmd 0x%x (#%d), arg 0x%lx\n", - cmd, _IOC_NR(cmd), arg); + dev_dbg(kfd_device, "ioctl cmd 0x%x (#%d), arg 0x%lx\n", cmd, nr, arg); process = kfd_get_process(current); - if (IS_ERR(process)) - return PTR_ERR(process); + if (IS_ERR(process)) { + dev_dbg(kfd_device, "no process\n"); + goto err_i1; + } - switch (cmd) { - case KFD_IOC_GET_VERSION: - err = kfd_ioctl_get_version(filep, process, (void __user *)arg); - break; - case KFD_IOC_CREATE_QUEUE: - err = kfd_ioctl_create_queue(filep, process, - (void __user *)arg); - break; - - case KFD_IOC_DESTROY_QUEUE: - err = kfd_ioctl_destroy_queue(filep, process, - (void __user *)arg); - break; - - case KFD_IOC_SET_MEMORY_POLICY: - err = kfd_ioctl_set_memory_policy(filep, process, - (void __user *)arg); - break; - - case KFD_IOC_GET_CLOCK_COUNTERS: - err = kfd_ioctl_get_clock_counters(filep, process, - (void __user *)arg); - break; - - case KFD_IOC_GET_PROCESS_APERTURES: - err = kfd_ioctl_get_process_apertures(filep, process, - (void __user *)arg); - break; - - case KFD_IOC_UPDATE_QUEUE: - err = kfd_ioctl_update_queue(filep, process, - (void __user *)arg); - break; - - default: - dev_err(kfd_device, - "unknown ioctl cmd 0x%x, arg 0x%lx)\n", - cmd, arg); - err = -EINVAL; - break; + /* Do not trust userspace, use our own definition */ + func = ioctl->func; + + if (unlikely(!func)) { + dev_dbg(kfd_device, "no function\n"); + retcode = -EINVAL; + goto err_i1; } - if (err < 0) - dev_err(kfd_device, - "ioctl error %ld for ioctl cmd 0x%x (#%d)\n", - err, cmd, _IOC_NR(cmd)); + if (cmd & (IOC_IN | IOC_OUT)) { + if (asize <= sizeof(stack_kdata)) { + kdata = stack_kdata; + } else { + kdata = kmalloc(asize, GFP_KERNEL); + if (!kdata) { + retcode = -ENOMEM; + goto err_i1; + } + } + if (asize > usize) + memset(kdata + usize, 0, asize - usize); + } - return err; + if (cmd & IOC_IN) { + if (copy_from_user(kdata, (void __user *)arg, usize) != 0) { + retcode = -EFAULT; + goto err_i1; + } + } else if (cmd & IOC_OUT) { + memset(kdata, 0, usize); + } + + retcode = func(filep, process, kdata); + + if (cmd & IOC_OUT) + if (copy_to_user((void __user *)arg, kdata, usize) != 0) + retcode = -EFAULT; + +err_i1: + if (!ioctl) + dev_dbg(kfd_device, "invalid ioctl: pid=%d, cmd=0x%02x, nr=0x%02x\n", + task_pid_nr(current), cmd, nr); + + if (kdata != stack_kdata) + kfree(kdata); + + if (retcode) + dev_dbg(kfd_device, "ret = %d\n", retcode); + + return retcode; } static int kfd_mmap(struct file *filp, struct vm_area_struct *vma) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 924e90c072e5..9c8961d22360 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -161,6 +161,9 @@ static void deallocate_vmid(struct device_queue_manager *dqm, { int bit = qpd->vmid - KFD_VMID_START_OFFSET; + /* Release the vmid mapping */ + set_pasid_vmid_mapping(dqm, 0, qpd->vmid); + set_bit(bit, (unsigned long *)&dqm->vmid_bitmap); qpd->vmid = 0; q->properties.vmid = 0; @@ -272,6 +275,18 @@ static int create_compute_queue_nocpsch(struct device_queue_manager *dqm, return retval; } + pr_debug("kfd: loading mqd to hqd on pipe (%d) queue (%d)\n", + q->pipe, + q->queue); + + retval = mqd->load_mqd(mqd, q->mqd, q->pipe, + q->queue, q->properties.write_ptr); + if (retval != 0) { + deallocate_hqd(dqm, q); + mqd->uninit_mqd(mqd, q->mqd, q->mqd_mem_obj); + return retval; + } + return 0; } @@ -320,6 +335,7 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q) { int retval; struct mqd_manager *mqd; + bool prev_active = false; BUG_ON(!dqm || !q || !q->mqd); @@ -330,10 +346,18 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q) return -ENOMEM; } - retval = mqd->update_mqd(mqd, q->mqd, &q->properties); if (q->properties.is_active == true) + prev_active = true; + + /* + * + * check active state vs. the previous state + * and modify counter accordingly + */ + retval = mqd->update_mqd(mqd, q->mqd, &q->properties); + if ((q->properties.is_active == true) && (prev_active == false)) dqm->queue_count++; - else + else if ((q->properties.is_active == false) && (prev_active == true)) dqm->queue_count--; if (sched_policy != KFD_SCHED_POLICY_NO_HWS) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c index 66df4da01c29..e64aa99e5e41 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c @@ -299,13 +299,13 @@ int kfd_init_apertures(struct kfd_process *process) struct kfd_dev *dev; struct kfd_process_device *pdd; - mutex_lock(&process->mutex); - /*Iterating over all devices*/ while ((dev = kfd_topology_enum_kfd_devices(id)) != NULL && id < NUM_OF_SUPPORTED_GPUS) { pdd = kfd_get_process_device_data(dev, process, 1); + if (!pdd) + return -1; /* * For 64 bit process aperture will be statically reserved in @@ -348,8 +348,6 @@ int kfd_init_apertures(struct kfd_process *process) id++; } - mutex_unlock(&process->mutex); - return 0; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c index 9abac48de499..935071410724 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c @@ -221,8 +221,14 @@ static int acquire_packet_buffer(struct kernel_queue *kq, queue_size_dwords; if (packet_size_in_dwords >= queue_size_dwords || - packet_size_in_dwords >= available_size) + packet_size_in_dwords >= available_size) { + /* + * make sure calling functions know + * acquire_packet_buffer() failed + */ + *buffer_ptr = NULL; return -ENOMEM; + } if (wptr + packet_size_in_dwords >= queue_size_dwords) { while (wptr > 0) { diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c index adc31474e786..4c3828cf45bf 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c @@ -184,7 +184,7 @@ static bool is_occupied(struct mqd_manager *mm, void *mqd, uint32_t queue_id) { - return kfd2kgd->hqd_is_occupies(mm->dev->kgd, queue_address, + return kfd2kgd->hqd_is_occupied(mm->dev->kgd, queue_address, pipe_id, queue_id); } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c b/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c index 2458ab7c0c6e..4c25ef504f79 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c @@ -32,8 +32,7 @@ int kfd_pasid_init(void) { pasid_limit = max_num_of_processes; - pasid_bitmap = kzalloc(DIV_ROUND_UP(pasid_limit, BITS_PER_BYTE), - GFP_KERNEL); + pasid_bitmap = kcalloc(BITS_TO_LONGS(pasid_limit), sizeof(long), GFP_KERNEL); if (!pasid_bitmap) return -ENOMEM; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index f9fb81e3bb09..a5edb29507e3 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -463,6 +463,24 @@ struct kfd_process { bool is_32bit_user_mode; }; +/** + * Ioctl function type. + * + * \param filep pointer to file structure. + * \param p amdkfd process pointer. + * \param data pointer to arg that was copied from user. + */ +typedef int amdkfd_ioctl_t(struct file *filep, struct kfd_process *p, + void *data); + +struct amdkfd_ioctl_desc { + unsigned int cmd; + int flags; + amdkfd_ioctl_t *func; + unsigned int cmd_drv; + const char *name; +}; + void kfd_process_create_wq(void); void kfd_process_destroy_wq(void); struct kfd_process *kfd_create_process(const struct task_struct *); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index b4f49ac13334..3c76ef05cbcf 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -26,6 +26,8 @@ #include <linux/slab.h> #include <linux/amd-iommu.h> #include <linux/notifier.h> +#include <linux/compat.h> + struct mm_struct; #include "kfd_priv.h" @@ -196,7 +198,7 @@ static void kfd_process_destroy_delayed(struct rcu_head *rcu) mmdrop(p->mm); work = (struct kfd_process_release_work *) - kmalloc(sizeof(struct kfd_process_release_work), GFP_KERNEL); + kmalloc(sizeof(struct kfd_process_release_work), GFP_ATOMIC); if (work) { INIT_WORK((struct work_struct *) work, kfd_process_wq_release); @@ -285,8 +287,15 @@ static struct kfd_process *create_process(const struct task_struct *thread) if (err != 0) goto err_process_pqm_init; + /* init process apertures*/ + process->is_32bit_user_mode = is_compat_task(); + if (kfd_init_apertures(process) != 0) + goto err_init_apretures; + return process; +err_init_apretures: + pqm_uninit(&process->pqm); err_process_pqm_init: hash_del_rcu(&process->kfd_processes); synchronize_rcu(); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index 5733e2859e8a..cca1708fd811 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -700,8 +700,6 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, dev->node_props.simd_per_cu); sysfs_show_32bit_prop(buffer, "max_slots_scratch_cu", dev->node_props.max_slots_scratch_cu); - sysfs_show_32bit_prop(buffer, "engine_id", - dev->node_props.engine_id); sysfs_show_32bit_prop(buffer, "vendor_id", dev->node_props.vendor_id); sysfs_show_32bit_prop(buffer, "device_id", @@ -715,6 +713,12 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, dev->gpu->kgd)); sysfs_show_64bit_prop(buffer, "local_mem_size", kfd2kgd->get_vmem_size(dev->gpu->kgd)); + + sysfs_show_32bit_prop(buffer, "fw_version", + kfd2kgd->get_fw_version( + dev->gpu->kgd, + KGD_ENGINE_MEC1)); + } ret = sysfs_show_32bit_prop(buffer, "max_engine_clk_ccompute", @@ -917,7 +921,7 @@ static int kfd_build_sysfs_node_tree(void) uint32_t i = 0; list_for_each_entry(dev, &topology_device_list, list) { - ret = kfd_build_sysfs_node_entry(dev, 0); + ret = kfd_build_sysfs_node_entry(dev, i); if (ret < 0) return ret; i++; diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h index 9c729dd8dd50..96a512208fad 100644 --- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h @@ -45,6 +45,17 @@ enum kgd_memory_pool { KGD_POOL_FRAMEBUFFER = 3, }; +enum kgd_engine_type { + KGD_ENGINE_PFP = 1, + KGD_ENGINE_ME, + KGD_ENGINE_CE, + KGD_ENGINE_MEC1, + KGD_ENGINE_MEC2, + KGD_ENGINE_RLC, + KGD_ENGINE_SDMA, + KGD_ENGINE_MAX +}; + struct kgd2kfd_shared_resources { /* Bit n == 1 means VMID n is available for KFD. */ unsigned int compute_vmid_bitmap; @@ -137,6 +148,8 @@ struct kgd2kfd_calls { * * @hqd_destroy: Destructs and preempts the queue assigned to that hqd slot. * + * @get_fw_version: Returns FW versions from the header + * * This structure contains function pointers to services that the kgd driver * provides to amdkfd driver. * @@ -170,12 +183,14 @@ struct kfd2kgd_calls { int (*hqd_load)(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id, uint32_t queue_id, uint32_t __user *wptr); - bool (*hqd_is_occupies)(struct kgd_dev *kgd, uint64_t queue_address, + bool (*hqd_is_occupied)(struct kgd_dev *kgd, uint64_t queue_address, uint32_t pipe_id, uint32_t queue_id); int (*hqd_destroy)(struct kgd_dev *kgd, uint32_t reset_type, unsigned int timeout, uint32_t pipe_id, uint32_t queue_id); + uint16_t (*get_fw_version)(struct kgd_dev *kgd, + enum kgd_engine_type type); }; bool kgd2kfd_init(unsigned interface_version, diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c index 908e5316eac4..b01420c84864 100644 --- a/drivers/gpu/drm/armada/armada_drv.c +++ b/drivers/gpu/drm/armada/armada_drv.c @@ -486,7 +486,6 @@ static struct platform_driver armada_drm_platform_driver = { .remove = armada_drm_remove, .driver = { .name = "armada-drm", - .owner = THIS_MODULE, }, .id_table = armada_drm_platform_ids, }; diff --git a/drivers/gpu/drm/armada/armada_gem.c b/drivers/gpu/drm/armada/armada_gem.c index 7496f55611a5..ef5feeecec84 100644 --- a/drivers/gpu/drm/armada/armada_gem.c +++ b/drivers/gpu/drm/armada/armada_gem.c @@ -226,7 +226,7 @@ struct armada_gem_object *armada_gem_alloc_object(struct drm_device *dev, obj->dev_addr = DMA_ERROR_CODE; - mapping = obj->obj.filp->f_path.dentry->d_inode->i_mapping; + mapping = file_inode(obj->obj.filp)->i_mapping; mapping_set_gfp_mask(mapping, GFP_HIGHUSER | __GFP_RECLAIMABLE); DRM_DEBUG_DRIVER("alloc obj %p size %zu\n", obj, size); diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c index 4a78a773151c..bbdbe4721573 100644 --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -61,7 +61,7 @@ drm_atomic_helper_plane_changed(struct drm_atomic_state *state, struct drm_crtc_state *crtc_state; if (plane->state->crtc) { - crtc_state = state->crtc_states[drm_crtc_index(plane->crtc)]; + crtc_state = state->crtc_states[drm_crtc_index(plane->state->crtc)]; if (WARN_ON(!crtc_state)) return; diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index 2985e3ff364f..4d0baa43de58 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -725,6 +725,8 @@ void drm_crtc_cleanup(struct drm_crtc *crtc) WARN_ON(crtc->state && !crtc->funcs->atomic_destroy_state); if (crtc->state && crtc->funcs->atomic_destroy_state) crtc->funcs->atomic_destroy_state(crtc, crtc->state); + + memset(crtc, 0, sizeof(*crtc)); } EXPORT_SYMBOL(drm_crtc_cleanup); @@ -908,6 +910,11 @@ void drm_connector_cleanup(struct drm_connector *connector) struct drm_device *dev = connector->dev; struct drm_display_mode *mode, *t; + if (connector->tile_group) { + drm_mode_put_tile_group(dev, connector->tile_group); + connector->tile_group = NULL; + } + list_for_each_entry_safe(mode, t, &connector->probed_modes, head) drm_mode_remove(connector, mode); @@ -927,6 +934,8 @@ void drm_connector_cleanup(struct drm_connector *connector) if (connector->state && connector->funcs->atomic_destroy_state) connector->funcs->atomic_destroy_state(connector, connector->state); + + memset(connector, 0, sizeof(*connector)); } EXPORT_SYMBOL(drm_connector_cleanup); @@ -1068,6 +1077,8 @@ void drm_bridge_cleanup(struct drm_bridge *bridge) list_del(&bridge->head); dev->mode_config.num_bridge--; drm_modeset_unlock_all(dev); + + memset(bridge, 0, sizeof(*bridge)); } EXPORT_SYMBOL(drm_bridge_cleanup); @@ -1134,10 +1145,11 @@ void drm_encoder_cleanup(struct drm_encoder *encoder) drm_modeset_lock_all(dev); drm_mode_object_put(dev, &encoder->base); kfree(encoder->name); - encoder->name = NULL; list_del(&encoder->head); dev->mode_config.num_encoder--; drm_modeset_unlock_all(dev); + + memset(encoder, 0, sizeof(*encoder)); } EXPORT_SYMBOL(drm_encoder_cleanup); @@ -1257,6 +1269,8 @@ void drm_plane_cleanup(struct drm_plane *plane) WARN_ON(plane->state && !plane->funcs->atomic_destroy_state); if (plane->state && plane->funcs->atomic_destroy_state) plane->funcs->atomic_destroy_state(plane, plane->state); + + memset(plane, 0, sizeof(*plane)); } EXPORT_SYMBOL(drm_plane_cleanup); @@ -1339,6 +1353,11 @@ static int drm_mode_create_standard_connector_properties(struct drm_device *dev) "PATH", 0); dev->mode_config.path_property = dev_path; + dev->mode_config.tile_property = drm_property_create(dev, + DRM_MODE_PROP_BLOB | + DRM_MODE_PROP_IMMUTABLE, + "TILE", 0); + return 0; } @@ -3456,7 +3475,7 @@ void drm_fb_release(struct drm_file *priv) /* * When the file gets released that means no one else can access the fb - * list any more, so no need to grab fpriv->fbs_lock. And we need to to + * list any more, so no need to grab fpriv->fbs_lock. And we need to * avoid upsetting lockdep since the universal cursor code adds a * framebuffer while holding mutex locks. * @@ -4095,6 +4114,52 @@ int drm_mode_connector_set_path_property(struct drm_connector *connector, EXPORT_SYMBOL(drm_mode_connector_set_path_property); /** + * drm_mode_connector_set_tile_property - set tile property on connector + * @connector: connector to set property on. + * + * This looks up the tile information for a connector, and creates a + * property for userspace to parse if it exists. The property is of + * the form of 8 integers using ':' as a separator. + * + * Returns: + * Zero on success, errno on failure. + */ +int drm_mode_connector_set_tile_property(struct drm_connector *connector) +{ + struct drm_device *dev = connector->dev; + int ret, size; + char tile[256]; + + if (connector->tile_blob_ptr) + drm_property_destroy_blob(dev, connector->tile_blob_ptr); + + if (!connector->has_tile) { + connector->tile_blob_ptr = NULL; + ret = drm_object_property_set_value(&connector->base, + dev->mode_config.tile_property, 0); + return ret; + } + + snprintf(tile, 256, "%d:%d:%d:%d:%d:%d:%d:%d", + connector->tile_group->id, connector->tile_is_single_monitor, + connector->num_h_tile, connector->num_v_tile, + connector->tile_h_loc, connector->tile_v_loc, + connector->tile_h_size, connector->tile_v_size); + size = strlen(tile) + 1; + + connector->tile_blob_ptr = drm_property_create_blob(connector->dev, + size, tile); + if (!connector->tile_blob_ptr) + return -EINVAL; + + ret = drm_object_property_set_value(&connector->base, + dev->mode_config.tile_property, + connector->tile_blob_ptr->base.id); + return ret; +} +EXPORT_SYMBOL(drm_mode_connector_set_tile_property); + +/** * drm_mode_connector_update_edid_property - update the edid property of a connector * @connector: drm connector * @edid: new value of the edid property @@ -5164,6 +5229,7 @@ void drm_mode_config_init(struct drm_device *dev) INIT_LIST_HEAD(&dev->mode_config.property_blob_list); INIT_LIST_HEAD(&dev->mode_config.plane_list); idr_init(&dev->mode_config.crtc_idr); + idr_init(&dev->mode_config.tile_idr); drm_modeset_lock_all(dev); drm_mode_create_standard_connector_properties(dev); @@ -5251,6 +5317,7 @@ void drm_mode_config_cleanup(struct drm_device *dev) crtc->funcs->destroy(crtc); } + idr_destroy(&dev->mode_config.tile_idr); idr_destroy(&dev->mode_config.crtc_idr); drm_modeset_lock_fini(&dev->mode_config.connection_mutex); } @@ -5273,3 +5340,100 @@ struct drm_property *drm_mode_create_rotation_property(struct drm_device *dev, supported_rotations); } EXPORT_SYMBOL(drm_mode_create_rotation_property); + +/** + * DOC: Tile group + * + * Tile groups are used to represent tiled monitors with a unique + * integer identifier. Tiled monitors using DisplayID v1.3 have + * a unique 8-byte handle, we store this in a tile group, so we + * have a common identifier for all tiles in a monitor group. + */ +static void drm_tile_group_free(struct kref *kref) +{ + struct drm_tile_group *tg = container_of(kref, struct drm_tile_group, refcount); + struct drm_device *dev = tg->dev; + mutex_lock(&dev->mode_config.idr_mutex); + idr_remove(&dev->mode_config.tile_idr, tg->id); + mutex_unlock(&dev->mode_config.idr_mutex); + kfree(tg); +} + +/** + * drm_mode_put_tile_group - drop a reference to a tile group. + * @dev: DRM device + * @tg: tile group to drop reference to. + * + * drop reference to tile group and free if 0. + */ +void drm_mode_put_tile_group(struct drm_device *dev, + struct drm_tile_group *tg) +{ + kref_put(&tg->refcount, drm_tile_group_free); +} + +/** + * drm_mode_get_tile_group - get a reference to an existing tile group + * @dev: DRM device + * @topology: 8-bytes unique per monitor. + * + * Use the unique bytes to get a reference to an existing tile group. + * + * RETURNS: + * tile group or NULL if not found. + */ +struct drm_tile_group *drm_mode_get_tile_group(struct drm_device *dev, + char topology[8]) +{ + struct drm_tile_group *tg; + int id; + mutex_lock(&dev->mode_config.idr_mutex); + idr_for_each_entry(&dev->mode_config.tile_idr, tg, id) { + if (!memcmp(tg->group_data, topology, 8)) { + if (!kref_get_unless_zero(&tg->refcount)) + tg = NULL; + mutex_unlock(&dev->mode_config.idr_mutex); + return tg; + } + } + mutex_unlock(&dev->mode_config.idr_mutex); + return NULL; +} + +/** + * drm_mode_create_tile_group - create a tile group from a displayid description + * @dev: DRM device + * @topology: 8-bytes unique per monitor. + * + * Create a tile group for the unique monitor, and get a unique + * identifier for the tile group. + * + * RETURNS: + * new tile group or error. + */ +struct drm_tile_group *drm_mode_create_tile_group(struct drm_device *dev, + char topology[8]) +{ + struct drm_tile_group *tg; + int ret; + + tg = kzalloc(sizeof(*tg), GFP_KERNEL); + if (!tg) + return ERR_PTR(-ENOMEM); + + kref_init(&tg->refcount); + memcpy(tg->group_data, topology, 8); + tg->dev = dev; + + mutex_lock(&dev->mode_config.idr_mutex); + ret = idr_alloc(&dev->mode_config.tile_idr, tg, 1, 0, GFP_KERNEL); + if (ret >= 0) { + tg->id = ret; + } else { + kfree(tg); + tg = ERR_PTR(ret); + } + + mutex_unlock(&dev->mode_config.idr_mutex); + return tg; +} diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c index 959e2074b0d4..79968e39c8d0 100644 --- a/drivers/gpu/drm/drm_dp_helper.c +++ b/drivers/gpu/drm/drm_dp_helper.c @@ -186,10 +186,11 @@ static int drm_dp_dpcd_access(struct drm_dp_aux *aux, u8 request, /* * The specification doesn't give any recommendation on how often to - * retry native transactions, so retry 7 times like for I2C-over-AUX - * transactions. + * retry native transactions. We used to retry 7 times like for + * aux i2c transactions but real world devices this wasn't + * sufficient, bump to 32 which makes Dell 4k monitors happier. */ - for (retry = 0; retry < 7; retry++) { + for (retry = 0; retry < 32; retry++) { mutex_lock(&aux->hw_mutex); err = aux->transfer(aux, &msg); diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c index 5682d7e9f1ec..9a5b68717ec8 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/drm_dp_mst_topology.c @@ -839,6 +839,8 @@ static void drm_dp_put_mst_branch_device(struct drm_dp_mst_branch *mstb) static void drm_dp_port_teardown_pdt(struct drm_dp_mst_port *port, int old_pdt) { + struct drm_dp_mst_branch *mstb; + switch (old_pdt) { case DP_PEER_DEVICE_DP_LEGACY_CONV: case DP_PEER_DEVICE_SST_SINK: @@ -846,8 +848,9 @@ static void drm_dp_port_teardown_pdt(struct drm_dp_mst_port *port, int old_pdt) drm_dp_mst_unregister_i2c_bus(&port->aux); break; case DP_PEER_DEVICE_MST_BRANCHING: - drm_dp_put_mst_branch_device(port->mstb); + mstb = port->mstb; port->mstb = NULL; + drm_dp_put_mst_branch_device(mstb); break; } } @@ -858,6 +861,8 @@ static void drm_dp_destroy_port(struct kref *kref) struct drm_dp_mst_topology_mgr *mgr = port->mgr; if (!port->input) { port->vcpi.num_slots = 0; + + kfree(port->cached_edid); if (port->connector) (*port->mgr->cbs->destroy_connector)(mgr, port->connector); drm_dp_port_teardown_pdt(port, port->pdt); @@ -1097,6 +1102,10 @@ static void drm_dp_add_port(struct drm_dp_mst_branch *mstb, char proppath[255]; build_mst_prop_path(port, mstb, proppath, sizeof(proppath)); port->connector = (*mstb->mgr->cbs->add_connector)(mstb->mgr, port, proppath); + + if (port->port_num >= 8) { + port->cached_edid = drm_get_edid(port->connector, &port->aux.ddc); + } } /* put reference to this port */ @@ -2167,7 +2176,8 @@ EXPORT_SYMBOL(drm_dp_mst_hpd_irq); * This returns the current connection state for a port. It validates the * port pointer still exists so the caller doesn't require a reference */ -enum drm_connector_status drm_dp_mst_detect_port(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_port *port) +enum drm_connector_status drm_dp_mst_detect_port(struct drm_connector *connector, + struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_port *port) { enum drm_connector_status status = connector_status_disconnected; @@ -2186,6 +2196,10 @@ enum drm_connector_status drm_dp_mst_detect_port(struct drm_dp_mst_topology_mgr case DP_PEER_DEVICE_SST_SINK: status = connector_status_connected; + /* for logical ports - cache the EDID */ + if (port->port_num >= 8 && !port->cached_edid) { + port->cached_edid = drm_get_edid(connector, &port->aux.ddc); + } break; case DP_PEER_DEVICE_DP_LEGACY_CONV: if (port->ldps) @@ -2217,7 +2231,12 @@ struct edid *drm_dp_mst_get_edid(struct drm_connector *connector, struct drm_dp_ if (!port) return NULL; - edid = drm_get_edid(connector, &port->aux.ddc); + if (port->cached_edid) + edid = drm_edid_duplicate(port->cached_edid); + else + edid = drm_get_edid(connector, &port->aux.ddc); + + drm_mode_connector_set_tile_property(connector); drm_dp_put_port(port); return edid; } diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index a7b5a71856a7..53bc7a628909 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -34,6 +34,7 @@ #include <linux/module.h> #include <drm/drmP.h> #include <drm/drm_edid.h> +#include <drm/drm_displayid.h> #define version_greater(edid, maj, min) \ (((edid)->version > (maj)) || \ @@ -1014,6 +1015,27 @@ module_param_named(edid_fixup, edid_fixup, int, 0400); MODULE_PARM_DESC(edid_fixup, "Minimum number of valid EDID header bytes (0-8, default 6)"); +static void drm_get_displayid(struct drm_connector *connector, + struct edid *edid); + +static int drm_edid_block_checksum(const u8 *raw_edid) +{ + int i; + u8 csum = 0; + for (i = 0; i < EDID_LENGTH; i++) + csum += raw_edid[i]; + + return csum; +} + +static bool drm_edid_is_zero(const u8 *in_edid, int length) +{ + if (memchr_inv(in_edid, 0, length)) + return false; + + return true; +} + /** * drm_edid_block_valid - Sanity check the EDID block (base or extension) * @raw_edid: pointer to raw EDID block @@ -1027,8 +1049,7 @@ MODULE_PARM_DESC(edid_fixup, */ bool drm_edid_block_valid(u8 *raw_edid, int block, bool print_bad_edid) { - int i; - u8 csum = 0; + u8 csum; struct edid *edid = (struct edid *)raw_edid; if (WARN_ON(!raw_edid)) @@ -1048,8 +1069,7 @@ bool drm_edid_block_valid(u8 *raw_edid, int block, bool print_bad_edid) } } - for (i = 0; i < EDID_LENGTH; i++) - csum += raw_edid[i]; + csum = drm_edid_block_checksum(raw_edid); if (csum) { if (print_bad_edid) { DRM_ERROR("EDID checksum is invalid, remainder is %d\n", csum); @@ -1080,9 +1100,13 @@ bool drm_edid_block_valid(u8 *raw_edid, int block, bool print_bad_edid) bad: if (print_bad_edid) { - printk(KERN_ERR "Raw EDID:\n"); - print_hex_dump(KERN_ERR, " \t", DUMP_PREFIX_NONE, 16, 1, + if (drm_edid_is_zero(raw_edid, EDID_LENGTH)) { + printk(KERN_ERR "EDID block is all zeroes\n"); + } else { + printk(KERN_ERR "Raw EDID:\n"); + print_hex_dump(KERN_ERR, " \t", DUMP_PREFIX_NONE, 16, 1, raw_edid, EDID_LENGTH, false); + } } return false; } @@ -1115,7 +1139,7 @@ EXPORT_SYMBOL(drm_edid_is_valid); #define DDC_SEGMENT_ADDR 0x30 /** * drm_do_probe_ddc_edid() - get EDID information via I2C - * @adapter: I2C device adaptor + * @data: I2C device adapter * @buf: EDID data buffer to be filled * @block: 128 byte EDID block to start fetching from * @len: EDID data buffer length to fetch @@ -1176,14 +1200,6 @@ drm_do_probe_ddc_edid(void *data, u8 *buf, unsigned int block, size_t len) return ret == xfers ? 0 : -1; } -static bool drm_edid_is_zero(u8 *in_edid, int length) -{ - if (memchr_inv(in_edid, 0, length)) - return false; - - return true; -} - /** * drm_do_get_edid - get EDID data using a custom EDID block read function * @connector: connector we're probing @@ -1308,10 +1324,15 @@ EXPORT_SYMBOL(drm_probe_ddc); struct edid *drm_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter) { + struct edid *edid; + if (!drm_probe_ddc(adapter)) return NULL; - return drm_do_get_edid(connector, drm_do_probe_ddc_edid, adapter); + edid = drm_do_get_edid(connector, drm_do_probe_ddc_edid, adapter); + if (edid) + drm_get_displayid(connector, edid); + return edid; } EXPORT_SYMBOL(drm_get_edid); @@ -2406,7 +2427,7 @@ add_detailed_modes(struct drm_connector *connector, struct edid *edid, /* * Search EDID for CEA extension block. */ -static u8 *drm_find_cea_extension(struct edid *edid) +static u8 *drm_find_edid_extension(struct edid *edid, int ext_id) { u8 *edid_ext = NULL; int i; @@ -2418,7 +2439,7 @@ static u8 *drm_find_cea_extension(struct edid *edid) /* Find CEA extension */ for (i = 0; i < edid->extensions; i++) { edid_ext = (u8 *)edid + EDID_LENGTH * (i + 1); - if (edid_ext[0] == CEA_EXT) + if (edid_ext[0] == ext_id) break; } @@ -2428,6 +2449,16 @@ static u8 *drm_find_cea_extension(struct edid *edid) return edid_ext; } +static u8 *drm_find_cea_extension(struct edid *edid) +{ + return drm_find_edid_extension(edid, CEA_EXT); +} + +static u8 *drm_find_displayid_extension(struct edid *edid) +{ + return drm_find_edid_extension(edid, DISPLAYID_EXT); +} + /* * Calculate the alternate clock for the CEA mode * (60Hz vs. 59.94Hz etc.) @@ -3888,3 +3919,123 @@ drm_hdmi_vendor_infoframe_from_display_mode(struct hdmi_vendor_infoframe *frame, return 0; } EXPORT_SYMBOL(drm_hdmi_vendor_infoframe_from_display_mode); + +static int drm_parse_display_id(struct drm_connector *connector, + u8 *displayid, int length, + bool is_edid_extension) +{ + /* if this is an EDID extension the first byte will be 0x70 */ + int idx = 0; + struct displayid_hdr *base; + struct displayid_block *block; + u8 csum = 0; + int i; + + if (is_edid_extension) + idx = 1; + + base = (struct displayid_hdr *)&displayid[idx]; + + DRM_DEBUG_KMS("base revision 0x%x, length %d, %d %d\n", + base->rev, base->bytes, base->prod_id, base->ext_count); + + if (base->bytes + 5 > length - idx) + return -EINVAL; + + for (i = idx; i <= base->bytes + 5; i++) { + csum += displayid[i]; + } + if (csum) { + DRM_ERROR("DisplayID checksum invalid, remainder is %d\n", csum); + return -EINVAL; + } + + block = (struct displayid_block *)&displayid[idx + 4]; + DRM_DEBUG_KMS("block id %d, rev %d, len %d\n", + block->tag, block->rev, block->num_bytes); + + switch (block->tag) { + case DATA_BLOCK_TILED_DISPLAY: { + struct displayid_tiled_block *tile = (struct displayid_tiled_block *)block; + + u16 w, h; + u8 tile_v_loc, tile_h_loc; + u8 num_v_tile, num_h_tile; + struct drm_tile_group *tg; + + w = tile->tile_size[0] | tile->tile_size[1] << 8; + h = tile->tile_size[2] | tile->tile_size[3] << 8; + + num_v_tile = (tile->topo[0] & 0xf) | (tile->topo[2] & 0x30); + num_h_tile = (tile->topo[0] >> 4) | ((tile->topo[2] >> 2) & 0x30); + tile_v_loc = (tile->topo[1] & 0xf) | ((tile->topo[2] & 0x3) << 4); + tile_h_loc = (tile->topo[1] >> 4) | (((tile->topo[2] >> 2) & 0x3) << 4); + + connector->has_tile = true; + if (tile->tile_cap & 0x80) + connector->tile_is_single_monitor = true; + + connector->num_h_tile = num_h_tile + 1; + connector->num_v_tile = num_v_tile + 1; + connector->tile_h_loc = tile_h_loc; + connector->tile_v_loc = tile_v_loc; + connector->tile_h_size = w + 1; + connector->tile_v_size = h + 1; + + DRM_DEBUG_KMS("tile cap 0x%x\n", tile->tile_cap); + DRM_DEBUG_KMS("tile_size %d x %d\n", w + 1, h + 1); + DRM_DEBUG_KMS("topo num tiles %dx%d, location %dx%d\n", + num_h_tile + 1, num_v_tile + 1, tile_h_loc, tile_v_loc); + DRM_DEBUG_KMS("vend %c%c%c\n", tile->topology_id[0], tile->topology_id[1], tile->topology_id[2]); + + tg = drm_mode_get_tile_group(connector->dev, tile->topology_id); + if (!tg) { + tg = drm_mode_create_tile_group(connector->dev, tile->topology_id); + } + if (!tg) + return -ENOMEM; + + if (connector->tile_group != tg) { + /* if we haven't got a pointer, + take the reference, drop ref to old tile group */ + if (connector->tile_group) { + drm_mode_put_tile_group(connector->dev, connector->tile_group); + } + connector->tile_group = tg; + } else + /* if same tile group, then release the ref we just took. */ + drm_mode_put_tile_group(connector->dev, tg); + } + break; + default: + printk("unknown displayid tag %d\n", block->tag); + break; + } + return 0; +} + +static void drm_get_displayid(struct drm_connector *connector, + struct edid *edid) +{ + void *displayid = NULL; + int ret; + connector->has_tile = false; + displayid = drm_find_displayid_extension(edid); + if (!displayid) { + /* drop reference to any tile group we had */ + goto out_drop_ref; + } + + ret = drm_parse_display_id(connector, displayid, EDID_LENGTH, true); + if (ret < 0) + goto out_drop_ref; + if (!connector->has_tile) + goto out_drop_ref; + return; +out_drop_ref: + if (connector->tile_group) { + drm_mode_put_tile_group(connector->dev, connector->tile_group); + connector->tile_group = NULL; + } + return; +} diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index 09d47e9ba026..52ce26d6b4fb 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -347,9 +347,18 @@ bool drm_fb_helper_restore_fbdev_mode_unlocked(struct drm_fb_helper *fb_helper) { struct drm_device *dev = fb_helper->dev; bool ret; + bool do_delayed = false; + drm_modeset_lock_all(dev); ret = restore_fbdev_mode(fb_helper); + + do_delayed = fb_helper->delayed_hotplug; + if (do_delayed) + fb_helper->delayed_hotplug = false; drm_modeset_unlock_all(dev); + + if (do_delayed) + drm_fb_helper_hotplug_event(fb_helper); return ret; } EXPORT_SYMBOL(drm_fb_helper_restore_fbdev_mode_unlocked); @@ -888,10 +897,6 @@ int drm_fb_helper_set_par(struct fb_info *info) drm_fb_helper_restore_fbdev_mode_unlocked(fb_helper); - if (fb_helper->delayed_hotplug) { - fb_helper->delayed_hotplug = false; - drm_fb_helper_hotplug_event(fb_helper); - } return 0; } EXPORT_SYMBOL(drm_fb_helper_set_par); @@ -995,19 +1000,21 @@ static int drm_fb_helper_single_fb_probe(struct drm_fb_helper *fb_helper, crtc_count = 0; for (i = 0; i < fb_helper->crtc_count; i++) { struct drm_display_mode *desired_mode; + int x, y; desired_mode = fb_helper->crtc_info[i].desired_mode; - + x = fb_helper->crtc_info[i].x; + y = fb_helper->crtc_info[i].y; if (desired_mode) { if (gamma_size == 0) gamma_size = fb_helper->crtc_info[i].mode_set.crtc->gamma_size; - if (desired_mode->hdisplay < sizes.fb_width) - sizes.fb_width = desired_mode->hdisplay; - if (desired_mode->vdisplay < sizes.fb_height) - sizes.fb_height = desired_mode->vdisplay; - if (desired_mode->hdisplay > sizes.surface_width) - sizes.surface_width = desired_mode->hdisplay; - if (desired_mode->vdisplay > sizes.surface_height) - sizes.surface_height = desired_mode->vdisplay; + if (desired_mode->hdisplay + x < sizes.fb_width) + sizes.fb_width = desired_mode->hdisplay + x; + if (desired_mode->vdisplay + y < sizes.fb_height) + sizes.fb_height = desired_mode->vdisplay + y; + if (desired_mode->hdisplay + x > sizes.surface_width) + sizes.surface_width = desired_mode->hdisplay + x; + if (desired_mode->vdisplay + y > sizes.surface_height) + sizes.surface_height = desired_mode->vdisplay + y; crtc_count++; } } @@ -1307,6 +1314,7 @@ static void drm_enable_connectors(struct drm_fb_helper *fb_helper, static bool drm_target_cloned(struct drm_fb_helper *fb_helper, struct drm_display_mode **modes, + struct drm_fb_offset *offsets, bool *enabled, int width, int height) { int count, i, j; @@ -1378,27 +1386,88 @@ static bool drm_target_cloned(struct drm_fb_helper *fb_helper, return false; } +static int drm_get_tile_offsets(struct drm_fb_helper *fb_helper, + struct drm_display_mode **modes, + struct drm_fb_offset *offsets, + int idx, + int h_idx, int v_idx) +{ + struct drm_fb_helper_connector *fb_helper_conn; + int i; + int hoffset = 0, voffset = 0; + + for (i = 0; i < fb_helper->connector_count; i++) { + fb_helper_conn = fb_helper->connector_info[i]; + if (!fb_helper_conn->connector->has_tile) + continue; + + if (!modes[i] && (h_idx || v_idx)) { + DRM_DEBUG_KMS("no modes for connector tiled %d %d\n", i, + fb_helper_conn->connector->base.id); + continue; + } + if (fb_helper_conn->connector->tile_h_loc < h_idx) + hoffset += modes[i]->hdisplay; + + if (fb_helper_conn->connector->tile_v_loc < v_idx) + voffset += modes[i]->vdisplay; + } + offsets[idx].x = hoffset; + offsets[idx].y = voffset; + DRM_DEBUG_KMS("returned %d %d for %d %d\n", hoffset, voffset, h_idx, v_idx); + return 0; +} + static bool drm_target_preferred(struct drm_fb_helper *fb_helper, struct drm_display_mode **modes, + struct drm_fb_offset *offsets, bool *enabled, int width, int height) { struct drm_fb_helper_connector *fb_helper_conn; int i; - + uint64_t conn_configured = 0, mask; + int tile_pass = 0; + mask = (1 << fb_helper->connector_count) - 1; +retry: for (i = 0; i < fb_helper->connector_count; i++) { fb_helper_conn = fb_helper->connector_info[i]; - if (enabled[i] == false) + if (conn_configured & (1 << i)) continue; + if (enabled[i] == false) { + conn_configured |= (1 << i); + continue; + } + + /* first pass over all the untiled connectors */ + if (tile_pass == 0 && fb_helper_conn->connector->has_tile) + continue; + + if (tile_pass == 1) { + if (fb_helper_conn->connector->tile_h_loc != 0 || + fb_helper_conn->connector->tile_v_loc != 0) + continue; + + } else { + if (fb_helper_conn->connector->tile_h_loc != tile_pass -1 && + fb_helper_conn->connector->tile_v_loc != tile_pass - 1) + /* if this tile_pass doesn't cover any of the tiles - keep going */ + continue; + + /* find the tile offsets for this pass - need + to find all tiles left and above */ + drm_get_tile_offsets(fb_helper, modes, offsets, + i, fb_helper_conn->connector->tile_h_loc, fb_helper_conn->connector->tile_v_loc); + } DRM_DEBUG_KMS("looking for cmdline mode on connector %d\n", fb_helper_conn->connector->base.id); /* got for command line mode first */ modes[i] = drm_pick_cmdline_mode(fb_helper_conn, width, height); if (!modes[i]) { - DRM_DEBUG_KMS("looking for preferred mode on connector %d\n", - fb_helper_conn->connector->base.id); + DRM_DEBUG_KMS("looking for preferred mode on connector %d %d\n", + fb_helper_conn->connector->base.id, fb_helper_conn->connector->tile_group ? fb_helper_conn->connector->tile_group->id : 0); modes[i] = drm_has_preferred_mode(fb_helper_conn, width, height); } /* No preferred modes, pick one off the list */ @@ -1408,6 +1477,12 @@ static bool drm_target_preferred(struct drm_fb_helper *fb_helper, } DRM_DEBUG_KMS("found mode %s\n", modes[i] ? modes[i]->name : "none"); + conn_configured |= (1 << i); + } + + if ((conn_configured & mask) != mask) { + tile_pass++; + goto retry; } return true; } @@ -1497,6 +1572,7 @@ static void drm_setup_crtcs(struct drm_fb_helper *fb_helper) struct drm_device *dev = fb_helper->dev; struct drm_fb_helper_crtc **crtcs; struct drm_display_mode **modes; + struct drm_fb_offset *offsets; struct drm_mode_set *modeset; bool *enabled; int width, height; @@ -1511,9 +1587,11 @@ static void drm_setup_crtcs(struct drm_fb_helper *fb_helper) sizeof(struct drm_fb_helper_crtc *), GFP_KERNEL); modes = kcalloc(dev->mode_config.num_connector, sizeof(struct drm_display_mode *), GFP_KERNEL); + offsets = kcalloc(dev->mode_config.num_connector, + sizeof(struct drm_fb_offset), GFP_KERNEL); enabled = kcalloc(dev->mode_config.num_connector, sizeof(bool), GFP_KERNEL); - if (!crtcs || !modes || !enabled) { + if (!crtcs || !modes || !enabled || !offsets) { DRM_ERROR("Memory allocation failed\n"); goto out; } @@ -1523,14 +1601,16 @@ static void drm_setup_crtcs(struct drm_fb_helper *fb_helper) if (!(fb_helper->funcs->initial_config && fb_helper->funcs->initial_config(fb_helper, crtcs, modes, + offsets, enabled, width, height))) { memset(modes, 0, dev->mode_config.num_connector*sizeof(modes[0])); memset(crtcs, 0, dev->mode_config.num_connector*sizeof(crtcs[0])); + memset(offsets, 0, dev->mode_config.num_connector*sizeof(offsets[0])); - if (!drm_target_cloned(fb_helper, - modes, enabled, width, height) && - !drm_target_preferred(fb_helper, - modes, enabled, width, height)) + if (!drm_target_cloned(fb_helper, modes, offsets, + enabled, width, height) && + !drm_target_preferred(fb_helper, modes, offsets, + enabled, width, height)) DRM_ERROR("Unable to find initial modes\n"); DRM_DEBUG_KMS("picking CRTCs for %dx%d config\n", @@ -1550,18 +1630,23 @@ static void drm_setup_crtcs(struct drm_fb_helper *fb_helper) for (i = 0; i < fb_helper->connector_count; i++) { struct drm_display_mode *mode = modes[i]; struct drm_fb_helper_crtc *fb_crtc = crtcs[i]; + struct drm_fb_offset *offset = &offsets[i]; modeset = &fb_crtc->mode_set; if (mode && fb_crtc) { - DRM_DEBUG_KMS("desired mode %s set on crtc %d\n", - mode->name, fb_crtc->mode_set.crtc->base.id); + DRM_DEBUG_KMS("desired mode %s set on crtc %d (%d,%d)\n", + mode->name, fb_crtc->mode_set.crtc->base.id, offset->x, offset->y); fb_crtc->desired_mode = mode; + fb_crtc->x = offset->x; + fb_crtc->y = offset->y; if (modeset->mode) drm_mode_destroy(dev, modeset->mode); modeset->mode = drm_mode_duplicate(dev, fb_crtc->desired_mode); modeset->connectors[modeset->num_connectors++] = fb_helper->connector_info[i]->connector; modeset->fb = fb_helper->fb; + modeset->x = offset->x; + modeset->y = offset->y; } } @@ -1578,6 +1663,7 @@ static void drm_setup_crtcs(struct drm_fb_helper *fb_helper) out: kfree(crtcs); kfree(modes); + kfree(offsets); kfree(enabled); } diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c index 91e1105f2800..0b9514b6cd64 100644 --- a/drivers/gpu/drm/drm_fops.c +++ b/drivers/gpu/drm/drm_fops.c @@ -527,6 +527,7 @@ ssize_t drm_read(struct file *filp, char __user *buffer, if (copy_to_user(buffer + total, e->event, e->event->length)) { total = -EFAULT; + e->destroy(e); break; } diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c index 0e47df4ef24e..4d79dad9d44f 100644 --- a/drivers/gpu/drm/drm_irq.c +++ b/drivers/gpu/drm/drm_irq.c @@ -166,7 +166,7 @@ static void vblank_disable_and_save(struct drm_device *dev, int crtc) spin_lock_irqsave(&dev->vblank_time_lock, irqflags); /* - * If the vblank interrupt was already disbled update the count + * If the vblank interrupt was already disabled update the count * and timestamp to maintain the appearance that the counter * has been ticking all along until this time. This makes the * count account for the entire time between drm_vblank_on() and @@ -830,6 +830,8 @@ drm_get_last_vbltimestamp(struct drm_device *dev, int crtc, * vblank events since the system was booted, including lost events due to * modesetting activity. * + * This is the legacy version of drm_crtc_vblank_count(). + * * Returns: * The software vblank counter. */ @@ -844,6 +846,25 @@ u32 drm_vblank_count(struct drm_device *dev, int crtc) EXPORT_SYMBOL(drm_vblank_count); /** + * drm_crtc_vblank_count - retrieve "cooked" vblank counter value + * @crtc: which counter to retrieve + * + * Fetches the "cooked" vblank count value that represents the number of + * vblank events since the system was booted, including lost events due to + * modesetting activity. + * + * This is the native KMS version of drm_vblank_count(). + * + * Returns: + * The software vblank counter. + */ +u32 drm_crtc_vblank_count(struct drm_crtc *crtc) +{ + return drm_vblank_count(crtc->dev, drm_crtc_index(crtc)); +} +EXPORT_SYMBOL(drm_crtc_vblank_count); + +/** * drm_vblank_count_and_time - retrieve "cooked" vblank counter value * and the system timestamp corresponding to that vblank counter value. * @@ -904,6 +925,8 @@ static void send_vblank_event(struct drm_device *dev, * * Updates sequence # and timestamp on event, and sends it to userspace. * Caller must hold event lock. + * + * This is the legacy version of drm_crtc_send_vblank_event(). */ void drm_send_vblank_event(struct drm_device *dev, int crtc, struct drm_pending_vblank_event *e) @@ -923,6 +946,23 @@ void drm_send_vblank_event(struct drm_device *dev, int crtc, EXPORT_SYMBOL(drm_send_vblank_event); /** + * drm_crtc_send_vblank_event - helper to send vblank event after pageflip + * @crtc: the source CRTC of the vblank event + * @e: the event to send + * + * Updates sequence # and timestamp on event, and sends it to userspace. + * Caller must hold event lock. + * + * This is the native KMS version of drm_send_vblank_event(). + */ +void drm_crtc_send_vblank_event(struct drm_crtc *crtc, + struct drm_pending_vblank_event *e) +{ + drm_send_vblank_event(crtc->dev, drm_crtc_index(crtc), e); +} +EXPORT_SYMBOL(drm_crtc_send_vblank_event); + +/** * drm_vblank_enable - enable the vblank interrupt on a CRTC * @dev: DRM device * @crtc: CRTC in question @@ -1594,6 +1634,8 @@ static void drm_handle_vblank_events(struct drm_device *dev, int crtc) * * Drivers should call this routine in their vblank interrupt handlers to * update the vblank counter and send any signals that may be pending. + * + * This is the legacy version of drm_crtc_handle_vblank(). */ bool drm_handle_vblank(struct drm_device *dev, int crtc) { @@ -1670,3 +1712,21 @@ bool drm_handle_vblank(struct drm_device *dev, int crtc) return true; } EXPORT_SYMBOL(drm_handle_vblank); + +/** + * drm_crtc_handle_vblank - handle a vblank event + * @crtc: where this event occurred + * + * Drivers should call this routine in their vblank interrupt handlers to + * update the vblank counter and send any signals that may be pending. + * + * This is the native KMS version of drm_handle_vblank(). + * + * Returns: + * True if the event was successfully handled, false on failure. + */ +bool drm_crtc_handle_vblank(struct drm_crtc *crtc) +{ + return drm_handle_vblank(crtc->dev, drm_crtc_index(crtc)); +} +EXPORT_SYMBOL(drm_crtc_handle_vblank); diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c index 25ba3628960a..121470a83d1a 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.c +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c @@ -618,7 +618,6 @@ static struct platform_driver exynos_drm_platform_driver = { .probe = exynos_drm_platform_probe, .remove = exynos_drm_platform_remove, .driver = { - .owner = THIS_MODULE, .name = "exynos-drm", .pm = &exynos_drm_pm_ops, }, diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c index 68d38eb6774d..835b6af00970 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c @@ -1817,7 +1817,7 @@ static int fimc_resume(struct device *dev) } #endif -#ifdef CONFIG_PM_RUNTIME +#ifdef CONFIG_PM static int fimc_runtime_suspend(struct device *dev) { struct fimc_context *ctx = get_fimc_context(dev); diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c index 6ff8599f6cbf..81a250830808 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c +++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c @@ -1540,7 +1540,7 @@ static int g2d_resume(struct device *dev) } #endif -#ifdef CONFIG_PM_RUNTIME +#ifdef CONFIG_PM static int g2d_runtime_suspend(struct device *dev) { struct g2d_data *g2d = dev_get_drvdata(dev); diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c index c6a013fc321c..0261468c8019 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c @@ -1764,7 +1764,7 @@ static int gsc_resume(struct device *dev) } #endif -#ifdef CONFIG_PM_RUNTIME +#ifdef CONFIG_PM static int gsc_runtime_suspend(struct device *dev) { struct gsc_context *ctx = get_gsc_context(dev); diff --git a/drivers/gpu/drm/exynos/exynos_drm_rotator.c b/drivers/gpu/drm/exynos/exynos_drm_rotator.c index b6a37d4f5b13..425e70625388 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_rotator.c +++ b/drivers/gpu/drm/exynos/exynos_drm_rotator.c @@ -822,7 +822,7 @@ static int rotator_resume(struct device *dev) } #endif -#ifdef CONFIG_PM_RUNTIME +#ifdef CONFIG_PM static int rotator_runtime_suspend(struct device *dev) { struct rot_context *rot = dev_get_drvdata(dev); diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index 8cbff3010e1c..2447de36de44 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -833,6 +833,8 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) intel_runtime_pm_enable(dev_priv); + i915_audio_component_init(dev_priv); + return 0; out_power_well: @@ -873,6 +875,8 @@ int i915_driver_unload(struct drm_device *dev) struct drm_i915_private *dev_priv = dev->dev_private; int ret; + i915_audio_component_cleanup(dev_priv); + ret = i915_gem_suspend(dev); if (ret) { DRM_ERROR("failed to idle hardware: %d\n", ret); diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 95bc829184eb..308774f42079 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -706,11 +706,12 @@ static int i915_drm_resume(struct drm_device *dev) dev_priv->display.hpd_irq_setup(dev); spin_unlock_irq(&dev_priv->irq_lock); - intel_dp_mst_resume(dev); drm_modeset_lock_all(dev); intel_modeset_setup_hw_state(dev, true); drm_modeset_unlock_all(dev); + intel_dp_mst_resume(dev); + /* * ... but also need to make sure that hotplug processing * doesn't cause havoc. Like in the driver load code we don't @@ -810,6 +811,8 @@ int i915_reset(struct drm_device *dev) if (!i915.reset) return 0; + intel_reset_gt_powersave(dev); + mutex_lock(&dev->struct_mutex); i915_gem_reset(dev); @@ -881,7 +884,7 @@ int i915_reset(struct drm_device *dev) * of re-init after reset. */ if (INTEL_INFO(dev)->gen > 5) - intel_reset_gt_powersave(dev); + intel_enable_gt_powersave(dev); } else { mutex_unlock(&dev->struct_mutex); } @@ -939,8 +942,7 @@ static int i915_pm_suspend(struct device *dev) static int i915_pm_suspend_late(struct device *dev) { - struct pci_dev *pdev = to_pci_dev(dev); - struct drm_device *drm_dev = pci_get_drvdata(pdev); + struct drm_device *drm_dev = dev_to_i915(dev)->dev; /* * We have a suspedn ordering issue with the snd-hda driver also @@ -959,8 +961,7 @@ static int i915_pm_suspend_late(struct device *dev) static int i915_pm_resume_early(struct device *dev) { - struct pci_dev *pdev = to_pci_dev(dev); - struct drm_device *drm_dev = pci_get_drvdata(pdev); + struct drm_device *drm_dev = dev_to_i915(dev)->dev; if (drm_dev->switch_power_state == DRM_SWITCH_POWER_OFF) return 0; @@ -970,8 +971,7 @@ static int i915_pm_resume_early(struct device *dev) static int i915_pm_resume(struct device *dev) { - struct pci_dev *pdev = to_pci_dev(dev); - struct drm_device *drm_dev = pci_get_drvdata(pdev); + struct drm_device *drm_dev = dev_to_i915(dev)->dev; if (drm_dev->switch_power_state == DRM_SWITCH_POWER_OFF) return 0; @@ -1588,7 +1588,7 @@ static struct drm_driver driver = { .gem_prime_import = i915_gem_prime_import, .dumb_create = i915_gem_dumb_create, - .dumb_map_offset = i915_gem_dumb_map_offset, + .dumb_map_offset = i915_gem_mmap_gtt, .dumb_destroy = drm_gem_dumb_destroy, .ioctls = i915_ioctls, .fops = &i915_driver_fops, diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 71016f5192c5..e008fa0c58da 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -970,6 +970,7 @@ struct i915_suspend_saved_registers { u32 savePIPEB_LINK_N1; u32 saveMCHBAR_RENDER_STANDBY; u32 savePCH_PORT_HOTPLUG; + u16 saveGCDGMBUS; }; struct vlv_s0ix_state { @@ -1771,6 +1772,9 @@ struct drm_i915_private { struct drm_property *broadcast_rgb_property; struct drm_property *force_audio_property; + /* hda/i915 audio component */ + bool audio_component_registered; + uint32_t hw_context_size; struct list_head context_list; @@ -1829,8 +1833,6 @@ struct drm_i915_private { */ struct workqueue_struct *dp_wq; - uint32_t bios_vgacntr; - /* Abstract the submission mechanism (legacy ringbuffer or execlists) away */ struct { int (*do_execbuf)(struct drm_device *dev, struct drm_file *file, @@ -1858,6 +1860,11 @@ static inline struct drm_i915_private *to_i915(const struct drm_device *dev) return dev->dev_private; } +static inline struct drm_i915_private *dev_to_i915(struct device *dev) +{ + return to_i915(dev_get_drvdata(dev)); +} + /* Iterate over initialised rings */ #define for_each_ring(ring__, dev_priv__, i__) \ for ((i__) = 0; (i__) < I915_NUM_RINGS; (i__)++) \ @@ -2633,9 +2640,8 @@ void i915_vma_move_to_active(struct i915_vma *vma, int i915_gem_dumb_create(struct drm_file *file_priv, struct drm_device *dev, struct drm_mode_create_dumb *args); -int i915_gem_dumb_map_offset(struct drm_file *file_priv, - struct drm_device *dev, uint32_t handle, - uint64_t *offset); +int i915_gem_mmap_gtt(struct drm_file *file_priv, struct drm_device *dev, + uint32_t handle, uint64_t *offset); /** * Returns true if seq1 is later than seq2. */ @@ -3218,6 +3224,11 @@ static inline unsigned long msecs_to_jiffies_timeout(const unsigned int m) return min_t(unsigned long, MAX_JIFFY_OFFSET, j + 1); } +static inline unsigned long nsecs_to_jiffies_timeout(const u64 n) +{ + return min_t(u64, MAX_JIFFY_OFFSET, nsecs_to_jiffies64(n) + 1); +} + static inline unsigned long timespec_to_jiffies_timeout(const struct timespec *value) { diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 3f6ca46a1dfe..9f430f77a520 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -395,7 +395,6 @@ static int i915_gem_create(struct drm_file *file, struct drm_device *dev, uint64_t size, - bool dumb, uint32_t *handle_p) { struct drm_i915_gem_object *obj; @@ -411,7 +410,6 @@ i915_gem_create(struct drm_file *file, if (obj == NULL) return -ENOMEM; - obj->base.dumb = dumb; ret = drm_gem_handle_create(file, &obj->base, &handle); /* drop reference from allocate - handle holds it now */ drm_gem_object_unreference_unlocked(&obj->base); @@ -431,7 +429,7 @@ i915_gem_dumb_create(struct drm_file *file, args->pitch = ALIGN(args->width * DIV_ROUND_UP(args->bpp, 8), 64); args->size = args->pitch * args->height; return i915_gem_create(file, dev, - args->size, true, &args->handle); + args->size, &args->handle); } /** @@ -444,7 +442,7 @@ i915_gem_create_ioctl(struct drm_device *dev, void *data, struct drm_i915_gem_create *args = data; return i915_gem_create(file, dev, - args->size, false, &args->handle); + args->size, &args->handle); } static inline int @@ -1044,6 +1042,7 @@ int i915_gem_pwrite_ioctl(struct drm_device *dev, void *data, struct drm_file *file) { + struct drm_i915_private *dev_priv = dev->dev_private; struct drm_i915_gem_pwrite *args = data; struct drm_i915_gem_object *obj; int ret; @@ -1063,9 +1062,11 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data, return -EFAULT; } + intel_runtime_pm_get(dev_priv); + ret = i915_mutex_lock_interruptible(dev); if (ret) - return ret; + goto put_rpm; obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle)); if (&obj->base == NULL) { @@ -1117,6 +1118,9 @@ out: drm_gem_object_unreference(&obj->base); unlock: mutex_unlock(&dev->struct_mutex); +put_rpm: + intel_runtime_pm_put(dev_priv); + return ret; } @@ -1220,7 +1224,8 @@ int __i915_wait_request(struct drm_i915_gem_request *req, if (i915_gem_request_completed(req, true)) return 0; - timeout_expire = timeout ? jiffies + nsecs_to_jiffies((u64)*timeout) : 0; + timeout_expire = timeout ? + jiffies + nsecs_to_jiffies_timeout((u64)*timeout) : 0; if (INTEL_INFO(dev)->gen >= 6 && ring->id == RCS && can_wait_boost(file_priv)) { gen6_rps_boost(dev_priv); @@ -1296,6 +1301,16 @@ int __i915_wait_request(struct drm_i915_gem_request *req, s64 tres = *timeout - (now - before); *timeout = tres < 0 ? 0 : tres; + + /* + * Apparently ktime isn't accurate enough and occasionally has a + * bit of mismatch in the jiffies<->nsecs<->ktime loop. So patch + * things up to make the test happy. We allow up to 1 jiffy. + * + * This is a regrssion from the timespec->ktime conversion. + */ + if (ret == -ETIME && *timeout < jiffies_to_usecs(1)*1000) + *timeout = 0; } return ret; @@ -1840,10 +1855,10 @@ static void i915_gem_object_free_mmap_offset(struct drm_i915_gem_object *obj) drm_gem_free_mmap_offset(&obj->base); } -static int +int i915_gem_mmap_gtt(struct drm_file *file, struct drm_device *dev, - uint32_t handle, bool dumb, + uint32_t handle, uint64_t *offset) { struct drm_i915_private *dev_priv = dev->dev_private; @@ -1860,13 +1875,6 @@ i915_gem_mmap_gtt(struct drm_file *file, goto unlock; } - /* - * We don't allow dumb mmaps on objects created using another - * interface. - */ - WARN_ONCE(dumb && !(obj->base.dumb || obj->base.import_attach), - "Illegal dumb map of accelerated buffer.\n"); - if (obj->base.size > dev_priv->gtt.mappable_end) { ret = -E2BIG; goto out; @@ -1891,15 +1899,6 @@ unlock: return ret; } -int -i915_gem_dumb_map_offset(struct drm_file *file, - struct drm_device *dev, - uint32_t handle, - uint64_t *offset) -{ - return i915_gem_mmap_gtt(file, dev, handle, true, offset); -} - /** * i915_gem_mmap_gtt_ioctl - prepare an object for GTT mmap'ing * @dev: DRM device @@ -1921,7 +1920,7 @@ i915_gem_mmap_gtt_ioctl(struct drm_device *dev, void *data, { struct drm_i915_gem_mmap_gtt *args = data; - return i915_gem_mmap_gtt(file, dev, args->handle, false, &args->offset); + return i915_gem_mmap_gtt(file, dev, args->handle, &args->offset); } static inline int diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c index b775ed4a189c..8603bf48d3ee 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.c +++ b/drivers/gpu/drm/i915/i915_gem_context.c @@ -486,7 +486,12 @@ mi_set_context(struct intel_engine_cs *ring, u32 hw_flags) { u32 flags = hw_flags | MI_MM_SPACE_GTT; - int ret; + const int num_rings = + /* Use an extended w/a on ivb+ if signalling from other rings */ + i915_semaphore_is_enabled(ring->dev) ? + hweight32(INTEL_INFO(ring->dev)->ring_mask) - 1 : + 0; + int len, i, ret; /* w/a: If Flush TLB Invalidation Mode is enabled, driver must do a TLB * invalidation prior to MI_SET_CONTEXT. On GEN6 we don't set the value @@ -503,15 +508,31 @@ mi_set_context(struct intel_engine_cs *ring, if (!IS_HASWELL(ring->dev) && INTEL_INFO(ring->dev)->gen < 8) flags |= (MI_SAVE_EXT_STATE_EN | MI_RESTORE_EXT_STATE_EN); - ret = intel_ring_begin(ring, 6); + + len = 4; + if (INTEL_INFO(ring->dev)->gen >= 7) + len += 2 + (num_rings ? 4*num_rings + 2 : 0); + + ret = intel_ring_begin(ring, len); if (ret) return ret; /* WaProgramMiArbOnOffAroundMiSetContext:ivb,vlv,hsw,bdw,chv */ - if (INTEL_INFO(ring->dev)->gen >= 7) + if (INTEL_INFO(ring->dev)->gen >= 7) { intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_DISABLE); - else - intel_ring_emit(ring, MI_NOOP); + if (num_rings) { + struct intel_engine_cs *signaller; + + intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(num_rings)); + for_each_ring(signaller, to_i915(ring->dev), i) { + if (signaller == ring) + continue; + + intel_ring_emit(ring, RING_PSMI_CTL(signaller->mmio_base)); + intel_ring_emit(ring, _MASKED_BIT_ENABLE(GEN6_PSMI_SLEEP_MSG_DISABLE)); + } + } + } intel_ring_emit(ring, MI_NOOP); intel_ring_emit(ring, MI_SET_CONTEXT); @@ -523,10 +544,21 @@ mi_set_context(struct intel_engine_cs *ring, */ intel_ring_emit(ring, MI_NOOP); - if (INTEL_INFO(ring->dev)->gen >= 7) + if (INTEL_INFO(ring->dev)->gen >= 7) { + if (num_rings) { + struct intel_engine_cs *signaller; + + intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(num_rings)); + for_each_ring(signaller, to_i915(ring->dev), i) { + if (signaller == ring) + continue; + + intel_ring_emit(ring, RING_PSMI_CTL(signaller->mmio_base)); + intel_ring_emit(ring, _MASKED_BIT_DISABLE(GEN6_PSMI_SLEEP_MSG_DISABLE)); + } + } intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_ENABLE); - else - intel_ring_emit(ring, MI_NOOP); + } intel_ring_advance(ring); diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index 6c21c59ed71d..e3ef17783765 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -122,9 +122,6 @@ eb_lookup_vmas(struct eb_vmas *eb, goto err; } - WARN_ONCE(obj->base.dumb, - "GPU use of dumb buffer is illegal.\n"); - drm_gem_object_reference(&obj->base); list_add_tail(&obj->obj_exec_link, &objects); } diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c index c38891892547..a2045848bd1a 100644 --- a/drivers/gpu/drm/i915/i915_gem_stolen.c +++ b/drivers/gpu/drm/i915/i915_gem_stolen.c @@ -137,7 +137,11 @@ static unsigned long i915_stolen_to_physical(struct drm_device *dev) r = devm_request_mem_region(dev->dev, base + 1, dev_priv->gtt.stolen_size - 1, "Graphics Stolen Memory"); - if (r == NULL) { + /* + * GEN3 firmware likes to smash pci bridges into the stolen + * range. Apparently this works. + */ + if (r == NULL && !IS_GEN3(dev)) { DRM_ERROR("conflict detected with stolen region: [0x%08x - 0x%08x]\n", base, base + (uint32_t)dev_priv->gtt.stolen_size); base = 0; diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index e3dd2d62c992..aa3180cf2921 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -285,10 +285,14 @@ void gen6_enable_rps_interrupts(struct drm_device *dev) struct drm_i915_private *dev_priv = dev->dev_private; spin_lock_irq(&dev_priv->irq_lock); + WARN_ON(dev_priv->rps.pm_iir); WARN_ON(I915_READ(gen6_pm_iir(dev_priv)) & dev_priv->pm_rps_events); dev_priv->rps.interrupts_enabled = true; + I915_WRITE(gen6_pm_ier(dev_priv), I915_READ(gen6_pm_ier(dev_priv)) | + dev_priv->pm_rps_events); gen6_enable_pm_irq(dev_priv, dev_priv->pm_rps_events); + spin_unlock_irq(&dev_priv->irq_lock); } @@ -3313,8 +3317,10 @@ static void gen5_gt_irq_postinstall(struct drm_device *dev) GEN5_IRQ_INIT(GT, dev_priv->gt_irq_mask, gt_irqs); if (INTEL_INFO(dev)->gen >= 6) { - pm_irqs |= dev_priv->pm_rps_events; - + /* + * RPS interrupts will get enabled/disabled on demand when RPS + * itself is enabled/disabled. + */ if (HAS_VEBOX(dev)) pm_irqs |= PM_VEBOX_USER_INTERRUPT; @@ -3526,7 +3532,11 @@ static void gen8_gt_irq_postinstall(struct drm_i915_private *dev_priv) dev_priv->pm_irq_mask = 0xffffffff; GEN8_IRQ_INIT_NDX(GT, 0, ~gt_interrupts[0], gt_interrupts[0]); GEN8_IRQ_INIT_NDX(GT, 1, ~gt_interrupts[1], gt_interrupts[1]); - GEN8_IRQ_INIT_NDX(GT, 2, dev_priv->pm_irq_mask, dev_priv->pm_rps_events); + /* + * RPS interrupts will get enabled/disabled on demand when RPS itself + * is enabled/disabled. + */ + GEN8_IRQ_INIT_NDX(GT, 2, dev_priv->pm_irq_mask, 0); GEN8_IRQ_INIT_NDX(GT, 3, ~gt_interrupts[3], gt_interrupts[3]); } @@ -3615,7 +3625,7 @@ static void vlv_display_irq_uninstall(struct drm_i915_private *dev_priv) vlv_display_irq_reset(dev_priv); - dev_priv->irq_mask = 0; + dev_priv->irq_mask = ~0; } static void valleyview_irq_uninstall(struct drm_device *dev) @@ -3721,8 +3731,6 @@ static bool i8xx_handle_vblank(struct drm_device *dev, if ((iir & flip_pending) == 0) goto check_page_flip; - intel_prepare_page_flip(dev, plane); - /* We detect FlipDone by looking for the change in PendingFlip from '1' * to '0' on the following vblank, i.e. IIR has the Pendingflip * asserted following the MI_DISPLAY_FLIP, but ISR is deasserted, hence @@ -3732,6 +3740,7 @@ static bool i8xx_handle_vblank(struct drm_device *dev, if (I915_READ16(ISR) & flip_pending) goto check_page_flip; + intel_prepare_page_flip(dev, plane); intel_finish_page_flip(dev, pipe); return true; @@ -3903,8 +3912,6 @@ static bool i915_handle_vblank(struct drm_device *dev, if ((iir & flip_pending) == 0) goto check_page_flip; - intel_prepare_page_flip(dev, plane); - /* We detect FlipDone by looking for the change in PendingFlip from '1' * to '0' on the following vblank, i.e. IIR has the Pendingflip * asserted following the MI_DISPLAY_FLIP, but ISR is deasserted, hence @@ -3914,6 +3921,7 @@ static bool i915_handle_vblank(struct drm_device *dev, if (I915_READ(ISR) & flip_pending) goto check_page_flip; + intel_prepare_page_flip(dev, plane); intel_finish_page_flip(dev, pipe); return true; diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 0cb0067af4bb..0f32fd1a9d10 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -34,8 +34,19 @@ #define _PORT3(port, a, b, c) ((port) == PORT_A ? (a) : \ (port) == PORT_B ? (b) : (c)) -#define _MASKED_BIT_ENABLE(a) (((a) << 16) | (a)) -#define _MASKED_BIT_DISABLE(a) ((a) << 16) +#define _MASKED_FIELD(mask, value) ({ \ + if (__builtin_constant_p(mask)) \ + BUILD_BUG_ON_MSG(((mask) & 0xffff0000), "Incorrect mask"); \ + if (__builtin_constant_p(value)) \ + BUILD_BUG_ON_MSG((value) & 0xffff0000, "Incorrect value"); \ + if (__builtin_constant_p(mask) && __builtin_constant_p(value)) \ + BUILD_BUG_ON_MSG((value) & ~(mask), \ + "Incorrect value for mask"); \ + (mask) << 16 | (value); }) +#define _MASKED_BIT_ENABLE(a) ({ typeof(a) _a = (a); _MASKED_FIELD(_a, _a); }) +#define _MASKED_BIT_DISABLE(a) (_MASKED_FIELD((a), 0)) + + /* PCI config space */ @@ -76,6 +87,7 @@ #define I915_GC_RENDER_CLOCK_166_MHZ (0 << 0) #define I915_GC_RENDER_CLOCK_200_MHZ (1 << 0) #define I915_GC_RENDER_CLOCK_333_MHZ (4 << 0) +#define GCDGMBUS 0xcc #define PCI_LBPC 0xf4 /* legacy/combination backlight modes, also called LBB */ @@ -389,6 +401,7 @@ #define PIPE_CONTROL_STORE_DATA_INDEX (1<<21) #define PIPE_CONTROL_CS_STALL (1<<20) #define PIPE_CONTROL_TLB_INVALIDATE (1<<18) +#define PIPE_CONTROL_MEDIA_STATE_CLEAR (1<<16) #define PIPE_CONTROL_QW_WRITE (1<<14) #define PIPE_CONTROL_POST_SYNC_OP_MASK (3<<14) #define PIPE_CONTROL_DEPTH_STALL (1<<13) @@ -1123,6 +1136,7 @@ enum punit_power_well { #define GEN6_VERSYNC (RING_SYNC_1(VEBOX_RING_BASE)) #define GEN6_VEVSYNC (RING_SYNC_2(VEBOX_RING_BASE)) #define GEN6_NOSYNC 0 +#define RING_PSMI_CTL(base) ((base)+0x50) #define RING_MAX_IDLE(base) ((base)+0x54) #define RING_HWS_PGA(base) ((base)+0x80) #define RING_HWS_PGA_GEN6(base) ((base)+0x2080) @@ -1289,7 +1303,7 @@ enum punit_power_well { #define GEN6_WIZ_HASHING_8x8 GEN6_WIZ_HASHING(0, 0) #define GEN6_WIZ_HASHING_8x4 GEN6_WIZ_HASHING(0, 1) #define GEN6_WIZ_HASHING_16x4 GEN6_WIZ_HASHING(1, 0) -#define GEN6_WIZ_HASHING_MASK (GEN6_WIZ_HASHING(1, 1) << 16) +#define GEN6_WIZ_HASHING_MASK GEN6_WIZ_HASHING(1, 1) #define GEN6_TD_FOUR_ROW_DISPATCH_DISABLE (1 << 5) #define GFX_MODE 0x02520 @@ -1453,6 +1467,7 @@ enum punit_power_well { #define GEN6_BLITTER_FBC_NOTIFY (1<<3) #define GEN6_RC_SLEEP_PSMI_CONTROL 0x2050 +#define GEN6_PSMI_SLEEP_MSG_DISABLE (1 << 0) #define GEN8_RC_SEMA_IDLE_MSG_DISABLE (1 << 12) #define GEN8_FF_DOP_CLOCK_GATE_DISABLE (1<<10) diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c index 1e4999dd3ed5..9f19ed38cdc3 100644 --- a/drivers/gpu/drm/i915/i915_suspend.c +++ b/drivers/gpu/drm/i915/i915_suspend.c @@ -303,6 +303,10 @@ int i915_save_state(struct drm_device *dev) } } + if (IS_GEN4(dev)) + pci_read_config_word(dev->pdev, GCDGMBUS, + &dev_priv->regfile.saveGCDGMBUS); + /* Cache mode state */ if (INTEL_INFO(dev)->gen < 7) dev_priv->regfile.saveCACHE_MODE_0 = I915_READ(CACHE_MODE_0); @@ -331,6 +335,10 @@ int i915_restore_state(struct drm_device *dev) mutex_lock(&dev->struct_mutex); i915_gem_restore_fences(dev); + + if (IS_GEN4(dev)) + pci_write_config_word(dev->pdev, GCDGMBUS, + dev_priv->regfile.saveGCDGMBUS); i915_restore_display(dev); if (!drm_core_check_feature(dev, DRIVER_MODESET)) { diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c index 2c7ed5cb29c0..ee41b882e71a 100644 --- a/drivers/gpu/drm/i915/intel_audio.c +++ b/drivers/gpu/drm/i915/intel_audio.c @@ -22,6 +22,9 @@ */ #include <linux/kernel.h> +#include <linux/component.h> +#include <drm/i915_component.h> +#include "intel_drv.h" #include <drm/drmP.h> #include <drm/drm_edid.h> @@ -461,3 +464,110 @@ void intel_init_audio(struct drm_device *dev) dev_priv->display.audio_codec_disable = ilk_audio_codec_disable; } } + +static void i915_audio_component_get_power(struct device *dev) +{ + intel_display_power_get(dev_to_i915(dev), POWER_DOMAIN_AUDIO); +} + +static void i915_audio_component_put_power(struct device *dev) +{ + intel_display_power_put(dev_to_i915(dev), POWER_DOMAIN_AUDIO); +} + +/* Get CDCLK in kHz */ +static int i915_audio_component_get_cdclk_freq(struct device *dev) +{ + struct drm_i915_private *dev_priv = dev_to_i915(dev); + int ret; + + if (WARN_ON_ONCE(!HAS_DDI(dev_priv))) + return -ENODEV; + + intel_display_power_get(dev_priv, POWER_DOMAIN_AUDIO); + ret = intel_ddi_get_cdclk_freq(dev_priv); + intel_display_power_put(dev_priv, POWER_DOMAIN_AUDIO); + + return ret; +} + +static const struct i915_audio_component_ops i915_audio_component_ops = { + .owner = THIS_MODULE, + .get_power = i915_audio_component_get_power, + .put_power = i915_audio_component_put_power, + .get_cdclk_freq = i915_audio_component_get_cdclk_freq, +}; + +static int i915_audio_component_bind(struct device *i915_dev, + struct device *hda_dev, void *data) +{ + struct i915_audio_component *acomp = data; + + if (WARN_ON(acomp->ops || acomp->dev)) + return -EEXIST; + + acomp->ops = &i915_audio_component_ops; + acomp->dev = i915_dev; + + return 0; +} + +static void i915_audio_component_unbind(struct device *i915_dev, + struct device *hda_dev, void *data) +{ + struct i915_audio_component *acomp = data; + + acomp->ops = NULL; + acomp->dev = NULL; +} + +static const struct component_ops i915_audio_component_bind_ops = { + .bind = i915_audio_component_bind, + .unbind = i915_audio_component_unbind, +}; + +/** + * i915_audio_component_init - initialize and register the audio component + * @dev_priv: i915 device instance + * + * This will register with the component framework a child component which + * will bind dynamically to the snd_hda_intel driver's corresponding master + * component when the latter is registered. During binding the child + * initializes an instance of struct i915_audio_component which it receives + * from the master. The master can then start to use the interface defined by + * this struct. Each side can break the binding at any point by deregistering + * its own component after which each side's component unbind callback is + * called. + * + * We ignore any error during registration and continue with reduced + * functionality (i.e. without HDMI audio). + */ +void i915_audio_component_init(struct drm_i915_private *dev_priv) +{ + int ret; + + ret = component_add(dev_priv->dev->dev, &i915_audio_component_bind_ops); + if (ret < 0) { + DRM_ERROR("failed to add audio component (%d)\n", ret); + /* continue with reduced functionality */ + return; + } + + dev_priv->audio_component_registered = true; +} + +/** + * i915_audio_component_cleanup - deregister the audio component + * @dev_priv: i915 device instance + * + * Deregisters the audio component, breaking any existing binding to the + * corresponding snd_hda_intel driver's master component. + */ +void i915_audio_component_cleanup(struct drm_i915_private *dev_priv) +{ + if (!dev_priv->audio_component_registered) + return; + + component_del(dev_priv->dev->dev, &i915_audio_component_bind_ops); + dev_priv->audio_component_registered = false; +} diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index e224820ea5a4..a340f51c790a 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -4500,7 +4500,6 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc) ironlake_fdi_disable(crtc); ironlake_disable_pch_transcoder(dev_priv, pipe); - intel_set_pch_fifo_underrun_reporting(dev_priv, pipe, true); if (HAS_PCH_CPT(dev)) { /* disable TRANS_DP_CTL */ @@ -4571,8 +4570,6 @@ static void haswell_crtc_disable(struct drm_crtc *crtc) if (intel_crtc->config.has_pch_encoder) { lpt_disable_pch_transcoder(dev_priv); - intel_set_pch_fifo_underrun_reporting(dev_priv, TRANSCODER_A, - true); intel_ddi_fdi_disable(crtc); } @@ -11467,10 +11464,12 @@ static int intel_crtc_set_config(struct drm_mode_set *set) to_intel_crtc(set->crtc)->config.has_audio) config->mode_changed = true; - /* Force mode sets for any infoframe stuff */ - if (pipe_config->has_infoframe || - to_intel_crtc(set->crtc)->config.has_infoframe) - config->mode_changed = true; + /* + * Note we have an issue here with infoframes: current code + * only updates them on the full mode set path per hw + * requirements. So here we should be checking for any + * required changes and forcing a mode set. + */ } /* set_mode will free it in the mode_changed case */ @@ -12957,11 +12956,7 @@ static void i915_disable_vga(struct drm_device *dev) vga_put(dev->pdev, VGA_RSRC_LEGACY_IO); udelay(300); - /* - * Fujitsu-Siemens Lifebook S6010 (830) has problems resuming - * from S3 without preserving (some of?) the other bits. - */ - I915_WRITE(vga_reg, dev_priv->bios_vgacntr | VGA_DISP_DISABLE); + I915_WRITE(vga_reg, VGA_DISP_DISABLE); POSTING_READ(vga_reg); } @@ -13046,8 +13041,6 @@ void intel_modeset_init(struct drm_device *dev) intel_shared_dpll_init(dev); - /* save the BIOS value before clobbering it */ - dev_priv->bios_vgacntr = I915_READ(i915_vgacntrl_reg(dev)); /* Just disable it once at startup */ i915_disable_vga(dev); intel_setup_outputs(dev); diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c index bfe359506377..7f8c6a66680a 100644 --- a/drivers/gpu/drm/i915/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/intel_dp_mst.c @@ -283,7 +283,7 @@ intel_dp_mst_detect(struct drm_connector *connector, bool force) struct intel_connector *intel_connector = to_intel_connector(connector); struct intel_dp *intel_dp = intel_connector->mst_port; - return drm_dp_mst_detect_port(&intel_dp->mst_mgr, intel_connector->port); + return drm_dp_mst_detect_port(connector, &intel_dp->mst_mgr, intel_connector->port); } static int @@ -414,6 +414,8 @@ static struct drm_connector *intel_dp_add_mst_connector(struct drm_dp_mst_topolo intel_dp_add_properties(intel_dp, connector); drm_object_attach_property(&connector->base, dev->mode_config.path_property, 0); + drm_object_attach_property(&connector->base, dev->mode_config.tile_property, 0); + drm_mode_connector_set_path_property(connector, pathprop); drm_reinit_primary_mode_group(dev); mutex_lock(&dev->mode_config.mutex); diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 588b618ab668..1043a1e3a569 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -873,6 +873,8 @@ void intel_fb_obj_flush(struct drm_i915_gem_object *obj, bool retire); void intel_init_audio(struct drm_device *dev); void intel_audio_codec_enable(struct intel_encoder *encoder); void intel_audio_codec_disable(struct intel_encoder *encoder); +void i915_audio_component_init(struct drm_i915_private *dev_priv); +void i915_audio_component_cleanup(struct drm_i915_private *dev_priv); /* intel_display.c */ bool intel_has_pending_fb_unpin(struct drm_device *dev); diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c index f2183b554cbc..850cf7d6578c 100644 --- a/drivers/gpu/drm/i915/intel_fbdev.c +++ b/drivers/gpu/drm/i915/intel_fbdev.c @@ -324,6 +324,7 @@ intel_fb_helper_crtc(struct drm_fb_helper *fb_helper, struct drm_crtc *crtc) static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper, struct drm_fb_helper_crtc **crtcs, struct drm_display_mode **modes, + struct drm_fb_offset *offsets, bool *enabled, int width, int height) { struct drm_device *dev = fb_helper->dev; @@ -332,6 +333,8 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper, bool fallback = true; int num_connectors_enabled = 0; int num_connectors_detected = 0; + uint64_t conn_configured = 0, mask; + int pass = 0; save_enabled = kcalloc(dev->mode_config.num_connector, sizeof(bool), GFP_KERNEL); @@ -339,7 +342,8 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper, return false; memcpy(save_enabled, enabled, dev->mode_config.num_connector); - + mask = (1 << fb_helper->connector_count) - 1; +retry: for (i = 0; i < fb_helper->connector_count; i++) { struct drm_fb_helper_connector *fb_conn; struct drm_connector *connector; @@ -349,12 +353,19 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper, fb_conn = fb_helper->connector_info[i]; connector = fb_conn->connector; + if (conn_configured & (1 << i)) + continue; + + if (pass == 0 && !connector->has_tile) + continue; + if (connector->status == connector_status_connected) num_connectors_detected++; if (!enabled[i]) { DRM_DEBUG_KMS("connector %s not enabled, skipping\n", connector->name); + conn_configured |= (1 << i); continue; } @@ -373,6 +384,7 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper, DRM_DEBUG_KMS("connector %s has no encoder or crtc, skipping\n", connector->name); enabled[i] = false; + conn_configured |= (1 << i); continue; } @@ -400,8 +412,8 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper, /* try for preferred next */ if (!modes[i]) { - DRM_DEBUG_KMS("looking for preferred mode on connector %s\n", - connector->name); + DRM_DEBUG_KMS("looking for preferred mode on connector %s %d\n", + connector->name, connector->has_tile); modes[i] = drm_has_preferred_mode(fb_conn, width, height); } @@ -444,6 +456,12 @@ static bool intel_fb_initial_config(struct drm_fb_helper *fb_helper, modes[i]->flags & DRM_MODE_FLAG_INTERLACE ? "i" :""); fallback = false; + conn_configured |= (1 << i); + } + + if ((conn_configured & mask) != mask) { + pass++; + goto retry; } /* diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c index c03d457a5150..14654d628ca4 100644 --- a/drivers/gpu/drm/i915/intel_lvds.c +++ b/drivers/gpu/drm/i915/intel_lvds.c @@ -899,6 +899,17 @@ void intel_lvds_init(struct drm_device *dev) int pipe; u8 pin; + /* + * Unlock registers and just leave them unlocked. Do this before + * checking quirk lists to avoid bogus WARNINGs. + */ + if (HAS_PCH_SPLIT(dev)) { + I915_WRITE(PCH_PP_CONTROL, + I915_READ(PCH_PP_CONTROL) | PANEL_UNLOCK_REGS); + } else { + I915_WRITE(PP_CONTROL, + I915_READ(PP_CONTROL) | PANEL_UNLOCK_REGS); + } if (!intel_lvds_supported(dev)) return; @@ -1097,17 +1108,6 @@ out: lvds_encoder->a3_power = I915_READ(lvds_encoder->reg) & LVDS_A3_POWER_MASK; - /* - * Unlock registers and just - * leave them unlocked - */ - if (HAS_PCH_SPLIT(dev)) { - I915_WRITE(PCH_PP_CONTROL, - I915_READ(PCH_PP_CONTROL) | PANEL_UNLOCK_REGS); - } else { - I915_WRITE(PP_CONTROL, - I915_READ(PP_CONTROL) | PANEL_UNLOCK_REGS); - } lvds_connector->lid_notifier.notifier_call = intel_lid_notify; if (acpi_lid_notifier_register(&lvds_connector->lid_notifier)) { DRM_DEBUG_KMS("lid notifier registration failed\n"); diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 091860432f01..3ba446a69ecd 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -5455,6 +5455,20 @@ void intel_cleanup_gt_powersave(struct drm_device *dev) valleyview_cleanup_gt_powersave(dev); } +static void gen6_suspend_rps(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + flush_delayed_work(&dev_priv->rps.delayed_resume_work); + + /* + * TODO: disable RPS interrupts on GEN9+ too once RPS support + * is added for it. + */ + if (INTEL_INFO(dev)->gen < 9) + gen6_disable_rps_interrupts(dev); +} + /** * intel_suspend_gt_powersave - suspend PM work and helper threads * @dev: drm device @@ -5470,14 +5484,7 @@ void intel_suspend_gt_powersave(struct drm_device *dev) if (INTEL_INFO(dev)->gen < 6) return; - flush_delayed_work(&dev_priv->rps.delayed_resume_work); - - /* - * TODO: disable RPS interrupts on GEN9+ too once RPS support - * is added for it. - */ - if (INTEL_INFO(dev)->gen < 9) - gen6_disable_rps_interrupts(dev); + gen6_suspend_rps(dev); /* Force GPU to min freq during suspend */ gen6_rps_idle(dev_priv); @@ -5580,8 +5587,11 @@ void intel_reset_gt_powersave(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; + if (INTEL_INFO(dev)->gen < 6) + return; + + gen6_suspend_rps(dev); dev_priv->rps.enabled = false; - intel_enable_gt_powersave(dev); } static void ibx_init_clock_gating(struct drm_device *dev) @@ -5772,7 +5782,7 @@ static void gen6_init_clock_gating(struct drm_device *dev) * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM). */ I915_WRITE(GEN6_GT_MODE, - GEN6_WIZ_HASHING_MASK | GEN6_WIZ_HASHING_16x4); + _MASKED_FIELD(GEN6_WIZ_HASHING_MASK, GEN6_WIZ_HASHING_16x4)); ilk_init_lp_watermarks(dev); @@ -5970,7 +5980,7 @@ static void haswell_init_clock_gating(struct drm_device *dev) * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM). */ I915_WRITE(GEN7_GT_MODE, - GEN6_WIZ_HASHING_MASK | GEN6_WIZ_HASHING_16x4); + _MASKED_FIELD(GEN6_WIZ_HASHING_MASK, GEN6_WIZ_HASHING_16x4)); /* WaSampleCChickenBitEnable:hsw */ I915_WRITE(HALF_SLICE_CHICKEN3, @@ -6071,7 +6081,7 @@ static void ivybridge_init_clock_gating(struct drm_device *dev) * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM). */ I915_WRITE(GEN7_GT_MODE, - GEN6_WIZ_HASHING_MASK | GEN6_WIZ_HASHING_16x4); + _MASKED_FIELD(GEN6_WIZ_HASHING_MASK, GEN6_WIZ_HASHING_16x4)); snpcr = I915_READ(GEN6_MBCUNIT_SNPCR); snpcr &= ~GEN6_MBC_SNPCR_MASK; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 3cad32a80108..12a36f0ca53d 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -373,12 +373,15 @@ gen7_render_ring_flush(struct intel_engine_cs *ring, flags |= PIPE_CONTROL_VF_CACHE_INVALIDATE; flags |= PIPE_CONTROL_CONST_CACHE_INVALIDATE; flags |= PIPE_CONTROL_STATE_CACHE_INVALIDATE; + flags |= PIPE_CONTROL_MEDIA_STATE_CLEAR; /* * TLB invalidate requires a post-sync write. */ flags |= PIPE_CONTROL_QW_WRITE; flags |= PIPE_CONTROL_GLOBAL_GTT_IVB; + flags |= PIPE_CONTROL_STALL_AT_SCOREBOARD; + /* Workaround: we must issue a pipe_control with CS-stall bit * set before a pipe_control command that has the state cache * invalidate bit set. */ @@ -727,7 +730,7 @@ static int intel_rcs_ctx_init(struct intel_engine_cs *ring, } static int wa_add(struct drm_i915_private *dev_priv, - const u32 addr, const u32 val, const u32 mask) + const u32 addr, const u32 mask, const u32 val) { const u32 idx = dev_priv->workarounds.count; @@ -743,22 +746,25 @@ static int wa_add(struct drm_i915_private *dev_priv, return 0; } -#define WA_REG(addr, val, mask) { \ - const int r = wa_add(dev_priv, (addr), (val), (mask)); \ +#define WA_REG(addr, mask, val) { \ + const int r = wa_add(dev_priv, (addr), (mask), (val)); \ if (r) \ return r; \ } #define WA_SET_BIT_MASKED(addr, mask) \ - WA_REG(addr, _MASKED_BIT_ENABLE(mask), (mask) & 0xffff) + WA_REG(addr, (mask), _MASKED_BIT_ENABLE(mask)) #define WA_CLR_BIT_MASKED(addr, mask) \ - WA_REG(addr, _MASKED_BIT_DISABLE(mask), (mask) & 0xffff) + WA_REG(addr, (mask), _MASKED_BIT_DISABLE(mask)) + +#define WA_SET_FIELD_MASKED(addr, mask, value) \ + WA_REG(addr, mask, _MASKED_FIELD(mask, value)) -#define WA_SET_BIT(addr, mask) WA_REG(addr, I915_READ(addr) | (mask), mask) -#define WA_CLR_BIT(addr, mask) WA_REG(addr, I915_READ(addr) & ~(mask), mask) +#define WA_SET_BIT(addr, mask) WA_REG(addr, mask, I915_READ(addr) | (mask)) +#define WA_CLR_BIT(addr, mask) WA_REG(addr, mask, I915_READ(addr) & ~(mask)) -#define WA_WRITE(addr, val) WA_REG(addr, val, 0xffffffff) +#define WA_WRITE(addr, val) WA_REG(addr, 0xffffffff, val) static int bdw_init_workarounds(struct intel_engine_cs *ring) { @@ -802,8 +808,9 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring) * disable bit, which we don't touch here, but it's good * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM). */ - WA_SET_BIT_MASKED(GEN7_GT_MODE, - GEN6_WIZ_HASHING_MASK | GEN6_WIZ_HASHING_16x4); + WA_SET_FIELD_MASKED(GEN7_GT_MODE, + GEN6_WIZ_HASHING_MASK, + GEN6_WIZ_HASHING_16x4); return 0; } diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index 6aa3a81df485..8bf7bb4a12bc 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -31,7 +31,6 @@ #include "i915_drv.h" #include "intel_drv.h" -#include <drm/i915_powerwell.h> /** * DOC: runtime pm @@ -50,8 +49,6 @@ * present for a given platform. */ -static struct i915_power_domains *hsw_pwr; - #define for_each_power_well(i, power_well, domain_mask, power_domains) \ for (i = 0; \ i < (power_domains)->power_well_count && \ @@ -615,29 +612,6 @@ static void chv_pipe_power_well_disable(struct drm_i915_private *dev_priv, vlv_power_sequencer_reset(dev_priv); } -static void check_power_well_state(struct drm_i915_private *dev_priv, - struct i915_power_well *power_well) -{ - bool enabled = power_well->ops->is_enabled(dev_priv, power_well); - - if (power_well->always_on || !i915.disable_power_well) { - if (!enabled) - goto mismatch; - - return; - } - - if (enabled != (power_well->count > 0)) - goto mismatch; - - return; - -mismatch: - I915_STATE_WARN(1, "state mismatch for '%s' (always_on %d hw state %d use-count %d disable_power_well %d\n", - power_well->name, power_well->always_on, enabled, - power_well->count, i915.disable_power_well); -} - /** * intel_display_power_get - grab a power domain reference * @dev_priv: i915 device instance @@ -669,8 +643,6 @@ void intel_display_power_get(struct drm_i915_private *dev_priv, power_well->ops->enable(dev_priv, power_well); power_well->hw_enabled = true; } - - check_power_well_state(dev_priv, power_well); } power_domains->domain_use_count[domain]++; @@ -709,8 +681,6 @@ void intel_display_power_put(struct drm_i915_private *dev_priv, power_well->hw_enabled = false; power_well->ops->disable(dev_priv, power_well); } - - check_power_well_state(dev_priv, power_well); } mutex_unlock(&power_domains->lock); @@ -1098,10 +1068,8 @@ int intel_power_domains_init(struct drm_i915_private *dev_priv) */ if (IS_HASWELL(dev_priv->dev)) { set_power_wells(power_domains, hsw_power_wells); - hsw_pwr = power_domains; } else if (IS_BROADWELL(dev_priv->dev)) { set_power_wells(power_domains, bdw_power_wells); - hsw_pwr = power_domains; } else if (IS_CHERRYVIEW(dev_priv->dev)) { set_power_wells(power_domains, chv_power_wells); } else if (IS_VALLEYVIEW(dev_priv->dev)) { @@ -1145,8 +1113,6 @@ void intel_power_domains_fini(struct drm_i915_private *dev_priv) * the power well is not enabled, so just enable it in case * we're going to unload/reload. */ intel_display_set_init_power(dev_priv, true); - - hsw_pwr = NULL; } static void intel_power_domains_resume(struct drm_i915_private *dev_priv) @@ -1355,52 +1321,3 @@ void intel_runtime_pm_enable(struct drm_i915_private *dev_priv) pm_runtime_put_autosuspend(device); } -/* Display audio driver power well request */ -int i915_request_power_well(void) -{ - struct drm_i915_private *dev_priv; - - if (!hsw_pwr) - return -ENODEV; - - dev_priv = container_of(hsw_pwr, struct drm_i915_private, - power_domains); - intel_display_power_get(dev_priv, POWER_DOMAIN_AUDIO); - return 0; -} -EXPORT_SYMBOL_GPL(i915_request_power_well); - -/* Display audio driver power well release */ -int i915_release_power_well(void) -{ - struct drm_i915_private *dev_priv; - - if (!hsw_pwr) - return -ENODEV; - - dev_priv = container_of(hsw_pwr, struct drm_i915_private, - power_domains); - intel_display_power_put(dev_priv, POWER_DOMAIN_AUDIO); - return 0; -} -EXPORT_SYMBOL_GPL(i915_release_power_well); - -/* - * Private interface for the audio driver to get CDCLK in kHz. - * - * Caller must request power well using i915_request_power_well() prior to - * making the call. - */ -int i915_get_cdclk_freq(void) -{ - struct drm_i915_private *dev_priv; - - if (!hsw_pwr) - return -ENODEV; - - dev_priv = container_of(hsw_pwr, struct drm_i915_private, - power_domains); - - return intel_ddi_get_cdclk_freq(dev_priv); -} -EXPORT_SYMBOL_GPL(i915_get_cdclk_freq); diff --git a/drivers/gpu/drm/imx/Kconfig b/drivers/gpu/drm/imx/Kconfig index 82fb758a29bc..ab31848e92cf 100644 --- a/drivers/gpu/drm/imx/Kconfig +++ b/drivers/gpu/drm/imx/Kconfig @@ -6,6 +6,7 @@ config DRM_IMX select DRM_GEM_CMA_HELPER select DRM_KMS_CMA_HELPER depends on DRM && (ARCH_MXC || ARCH_MULTIPLATFORM) + depends on IMX_IPUV3_CORE help enable i.MX graphics support @@ -40,11 +41,11 @@ config DRM_IMX_LDB found on i.MX53 and i.MX6 processors. config DRM_IMX_IPUV3 - tristate "DRM Support for i.MX IPUv3" + tristate depends on DRM_IMX depends on IMX_IPUV3_CORE - help - Choose this if you have a i.MX5 or i.MX6 processor. + default y if DRM_IMX=y + default m if DRM_IMX=m config DRM_IMX_HDMI tristate "Freescale i.MX DRM HDMI" diff --git a/drivers/gpu/drm/imx/imx-drm-core.c b/drivers/gpu/drm/imx/imx-drm-core.c index 2f8007241734..b250130debc8 100644 --- a/drivers/gpu/drm/imx/imx-drm-core.c +++ b/drivers/gpu/drm/imx/imx-drm-core.c @@ -30,8 +30,6 @@ #define MAX_CRTC 4 -struct imx_drm_crtc; - struct imx_drm_component { struct device_node *of_node; struct list_head list; @@ -633,7 +631,8 @@ static int imx_drm_platform_probe(struct platform_device *pdev) continue; } - component_match_add(&pdev->dev, &match, compare_of, remote); + component_match_add(&pdev->dev, &match, compare_of, + remote); of_node_put(remote); } of_node_put(port); @@ -692,7 +691,6 @@ static struct platform_driver imx_drm_pdrv = { .probe = imx_drm_platform_probe, .remove = imx_drm_platform_remove, .driver = { - .owner = THIS_MODULE, .name = "imx-drm", .pm = &imx_drm_pm_ops, .of_match_table = imx_drm_dt_ids, diff --git a/drivers/gpu/drm/imx/imx-hdmi.c b/drivers/gpu/drm/imx/imx-hdmi.c index aaec6b2cdf56..ddc53e039530 100644 --- a/drivers/gpu/drm/imx/imx-hdmi.c +++ b/drivers/gpu/drm/imx/imx-hdmi.c @@ -1754,7 +1754,6 @@ static struct platform_driver imx_hdmi_driver = { .remove = imx_hdmi_platform_remove, .driver = { .name = "imx-hdmi", - .owner = THIS_MODULE, .of_match_table = imx_hdmi_dt_ids, }, }; diff --git a/drivers/gpu/drm/imx/imx-ldb.c b/drivers/gpu/drm/imx/imx-ldb.c index 4662e00b456a..c60460043e24 100644 --- a/drivers/gpu/drm/imx/imx-ldb.c +++ b/drivers/gpu/drm/imx/imx-ldb.c @@ -11,11 +11,6 @@ * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, - * MA 02110-1301, USA. */ #include <linux/module.h> @@ -604,7 +599,6 @@ static struct platform_driver imx_ldb_driver = { .driver = { .of_match_table = imx_ldb_dt_ids, .name = DRIVER_NAME, - .owner = THIS_MODULE, }, }; diff --git a/drivers/gpu/drm/imx/imx-tve.c b/drivers/gpu/drm/imx/imx-tve.c index 42c651be6c20..a729f4f7074c 100644 --- a/drivers/gpu/drm/imx/imx-tve.c +++ b/drivers/gpu/drm/imx/imx-tve.c @@ -11,11 +11,6 @@ * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, - * MA 02110-1301, USA. */ #include <linux/clk.h> @@ -665,7 +660,8 @@ static int imx_tve_bind(struct device *dev, struct device *master, void *data) ret = regmap_read(tve->regmap, TVE_COM_CONF_REG, &val); if (ret < 0) { - dev_err(dev, "failed to read configuration register: %d\n", ret); + dev_err(dev, "failed to read configuration register: %d\n", + ret); return ret; } if (val != 0x00100000) { @@ -724,7 +720,6 @@ static struct platform_driver imx_tve_driver = { .driver = { .of_match_table = imx_tve_dt_ids, .name = "imx-tve", - .owner = THIS_MODULE, }, }; diff --git a/drivers/gpu/drm/imx/ipuv3-crtc.c b/drivers/gpu/drm/imx/ipuv3-crtc.c index 11e84a251773..ebee59cb96d8 100644 --- a/drivers/gpu/drm/imx/ipuv3-crtc.c +++ b/drivers/gpu/drm/imx/ipuv3-crtc.c @@ -11,11 +11,6 @@ * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, - * MA 02110-1301, USA. */ #include <linux/component.h> #include <linux/module.h> diff --git a/drivers/gpu/drm/imx/ipuv3-plane.c b/drivers/gpu/drm/imx/ipuv3-plane.c index 944962b692bb..6987e16fe99b 100644 --- a/drivers/gpu/drm/imx/ipuv3-plane.c +++ b/drivers/gpu/drm/imx/ipuv3-plane.c @@ -64,6 +64,7 @@ int ipu_plane_set_base(struct ipu_plane *ipu_plane, struct drm_framebuffer *fb, { struct drm_gem_cma_object *cma_obj; unsigned long eba; + int active; cma_obj = drm_fb_cma_get_gem_obj(fb, 0); if (!cma_obj) { @@ -74,12 +75,17 @@ int ipu_plane_set_base(struct ipu_plane *ipu_plane, struct drm_framebuffer *fb, dev_dbg(ipu_plane->base.dev->dev, "phys = %pad, x = %d, y = %d", &cma_obj->paddr, x, y); - ipu_cpmem_set_stride(ipu_plane->ipu_ch, fb->pitches[0]); - eba = cma_obj->paddr + fb->offsets[0] + fb->pitches[0] * y + (fb->bits_per_pixel >> 3) * x; - ipu_cpmem_set_buffer(ipu_plane->ipu_ch, 0, eba); - ipu_cpmem_set_buffer(ipu_plane->ipu_ch, 1, eba); + + if (ipu_plane->enabled) { + active = ipu_idmac_get_current_buffer(ipu_plane->ipu_ch); + ipu_cpmem_set_buffer(ipu_plane->ipu_ch, !active, eba); + ipu_idmac_select_buffer(ipu_plane->ipu_ch, !active); + } else { + ipu_cpmem_set_buffer(ipu_plane->ipu_ch, 0, eba); + ipu_cpmem_set_buffer(ipu_plane->ipu_ch, 1, eba); + } /* cache offsets for subsequent pageflips */ ipu_plane->x = x; @@ -137,6 +143,18 @@ int ipu_plane_mode_set(struct ipu_plane *ipu_plane, struct drm_crtc *crtc, if (crtc_h < 2) return -EINVAL; + /* + * since we cannot touch active IDMAC channels, we do not support + * resizing the enabled plane or changing its format + */ + if (ipu_plane->enabled) { + if (src_w != ipu_plane->w || src_h != ipu_plane->h || + fb->pixel_format != ipu_plane->base.fb->pixel_format) + return -EINVAL; + + return ipu_plane_set_base(ipu_plane, fb, src_x, src_y); + } + switch (ipu_plane->dp_flow) { case IPU_DP_FLOW_SYNC_BG: ret = ipu_dp_setup_channel(ipu_plane->dp, @@ -148,14 +166,22 @@ int ipu_plane_mode_set(struct ipu_plane *ipu_plane, struct drm_crtc *crtc, ret); return ret; } - ipu_dp_set_global_alpha(ipu_plane->dp, 1, 0, 1); + ipu_dp_set_global_alpha(ipu_plane->dp, true, 0, true); break; case IPU_DP_FLOW_SYNC_FG: ipu_dp_setup_channel(ipu_plane->dp, ipu_drm_fourcc_to_colorspace(fb->pixel_format), IPUV3_COLORSPACE_UNKNOWN); ipu_dp_set_window_pos(ipu_plane->dp, crtc_x, crtc_y); - break; + /* Enable local alpha on partial plane */ + switch (fb->pixel_format) { + case DRM_FORMAT_ARGB8888: + case DRM_FORMAT_ABGR8888: + ipu_dp_set_global_alpha(ipu_plane->dp, false, 0, false); + break; + default: + break; + } } ret = ipu_dmfc_init_channel(ipu_plane->dmfc, crtc_w); @@ -181,11 +207,16 @@ int ipu_plane_mode_set(struct ipu_plane *ipu_plane, struct drm_crtc *crtc, return ret; } ipu_cpmem_set_high_priority(ipu_plane->ipu_ch); + ipu_idmac_set_double_buffer(ipu_plane->ipu_ch, 1); + ipu_cpmem_set_stride(ipu_plane->ipu_ch, fb->pitches[0]); ret = ipu_plane_set_base(ipu_plane, fb, src_x, src_y); if (ret < 0) return ret; + ipu_plane->w = src_w; + ipu_plane->h = src_h; + return 0; } diff --git a/drivers/gpu/drm/imx/ipuv3-plane.h b/drivers/gpu/drm/imx/ipuv3-plane.h index c0aae5bcb5d4..af125fb40ef5 100644 --- a/drivers/gpu/drm/imx/ipuv3-plane.h +++ b/drivers/gpu/drm/imx/ipuv3-plane.h @@ -26,6 +26,8 @@ struct ipu_plane { int x; int y; + int w; + int h; bool enabled; }; diff --git a/drivers/gpu/drm/imx/parallel-display.c b/drivers/gpu/drm/imx/parallel-display.c index 015a454b87e1..796c3c1c170a 100644 --- a/drivers/gpu/drm/imx/parallel-display.c +++ b/drivers/gpu/drm/imx/parallel-display.c @@ -11,11 +11,6 @@ * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, - * MA 02110-1301, USA. */ #include <linux/component.h> @@ -128,6 +123,10 @@ static void imx_pd_encoder_prepare(struct drm_encoder *encoder) static void imx_pd_encoder_commit(struct drm_encoder *encoder) { + struct imx_parallel_display *imxpd = enc_to_imxpd(encoder); + + drm_panel_prepare(imxpd->panel); + drm_panel_enable(imxpd->panel); } static void imx_pd_encoder_mode_set(struct drm_encoder *encoder, @@ -138,6 +137,10 @@ static void imx_pd_encoder_mode_set(struct drm_encoder *encoder, static void imx_pd_encoder_disable(struct drm_encoder *encoder) { + struct imx_parallel_display *imxpd = enc_to_imxpd(encoder); + + drm_panel_disable(imxpd->panel); + drm_panel_unprepare(imxpd->panel); } static struct drm_connector_funcs imx_pd_connector_funcs = { @@ -284,7 +287,6 @@ static struct platform_driver imx_pd_driver = { .driver = { .of_match_table = imx_pd_dt_ids, .name = "imx-parallel-display", - .owner = THIS_MODULE, }, }; diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index aa873048308b..94a5bee69fe7 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -386,9 +386,7 @@ void adreno_gpu_cleanup(struct adreno_gpu *gpu) msm_gem_put_iova(gpu->memptrs_bo, gpu->base.id); drm_gem_object_unreference(gpu->memptrs_bo); } - if (gpu->pm4) - release_firmware(gpu->pm4); - if (gpu->pfp) - release_firmware(gpu->pfp); + release_firmware(gpu->pm4); + release_firmware(gpu->pfp); msm_gpu_cleanup(&gpu->base); } diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c index fbebb0405d76..b4e70e0e3cfa 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c @@ -141,6 +141,15 @@ static int hpd_enable(struct hdmi_connector *hdmi_connector) uint32_t hpd_ctrl; int i, ret; + for (i = 0; i < config->hpd_reg_cnt; i++) { + ret = regulator_enable(hdmi->hpd_regs[i]); + if (ret) { + dev_err(dev->dev, "failed to enable hpd regulator: %s (%d)\n", + config->hpd_reg_names[i], ret); + goto fail; + } + } + ret = gpio_config(hdmi, true); if (ret) { dev_err(dev->dev, "failed to configure GPIOs: %d\n", ret); @@ -164,15 +173,6 @@ static int hpd_enable(struct hdmi_connector *hdmi_connector) } } - for (i = 0; i < config->hpd_reg_cnt; i++) { - ret = regulator_enable(hdmi->hpd_regs[i]); - if (ret) { - dev_err(dev->dev, "failed to enable hpd regulator: %s (%d)\n", - config->hpd_reg_names[i], ret); - goto fail; - } - } - hdmi_set_mode(hdmi, false); phy->funcs->reset(phy); hdmi_set_mode(hdmi, true); @@ -200,7 +200,7 @@ fail: return ret; } -static int hdp_disable(struct hdmi_connector *hdmi_connector) +static void hdp_disable(struct hdmi_connector *hdmi_connector) { struct hdmi *hdmi = hdmi_connector->hdmi; const struct hdmi_platform_config *config = hdmi->config; @@ -212,28 +212,19 @@ static int hdp_disable(struct hdmi_connector *hdmi_connector) hdmi_set_mode(hdmi, false); - for (i = 0; i < config->hpd_reg_cnt; i++) { - ret = regulator_disable(hdmi->hpd_regs[i]); - if (ret) { - dev_err(dev->dev, "failed to disable hpd regulator: %s (%d)\n", - config->hpd_reg_names[i], ret); - goto fail; - } - } - for (i = 0; i < config->hpd_clk_cnt; i++) clk_disable_unprepare(hdmi->hpd_clks[i]); ret = gpio_config(hdmi, false); - if (ret) { - dev_err(dev->dev, "failed to unconfigure GPIOs: %d\n", ret); - goto fail; - } - - return 0; + if (ret) + dev_warn(dev->dev, "failed to unconfigure GPIOs: %d\n", ret); -fail: - return ret; + for (i = 0; i < config->hpd_reg_cnt; i++) { + ret = regulator_disable(hdmi->hpd_regs[i]); + if (ret) + dev_warn(dev->dev, "failed to disable hpd regulator: %s (%d)\n", + config->hpd_reg_names[i], ret); + } } static void @@ -260,11 +251,11 @@ void hdmi_connector_irq(struct drm_connector *connector) (hpd_int_status & HDMI_HPD_INT_STATUS_INT)) { bool detected = !!(hpd_int_status & HDMI_HPD_INT_STATUS_CABLE_DETECTED); - DBG("status=%04x, ctrl=%04x", hpd_int_status, hpd_int_ctrl); - - /* ack the irq: */ + /* ack & disable (temporarily) HPD events: */ hdmi_write(hdmi, REG_HDMI_HPD_INT_CTRL, - hpd_int_ctrl | HDMI_HPD_INT_CTRL_INT_ACK); + HDMI_HPD_INT_CTRL_INT_ACK); + + DBG("status=%04x, ctrl=%04x", hpd_int_status, hpd_int_ctrl); /* detect disconnect if we are connected or visa versa: */ hpd_int_ctrl = HDMI_HPD_INT_CTRL_INT_EN; diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c index a7672e100d8b..3449213f1e76 100644 --- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c +++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c @@ -331,17 +331,8 @@ static int mdp4_crtc_atomic_check(struct drm_crtc *crtc, struct drm_crtc_state *state) { struct mdp4_crtc *mdp4_crtc = to_mdp4_crtc(crtc); - struct drm_device *dev = crtc->dev; - DBG("%s: check", mdp4_crtc->name); - - if (mdp4_crtc->event) { - dev_err(dev->dev, "already pending flip!\n"); - return -EBUSY; - } - // TODO anything else to check? - return 0; } @@ -357,7 +348,7 @@ static void mdp4_crtc_atomic_flush(struct drm_crtc *crtc) struct drm_device *dev = crtc->dev; unsigned long flags; - DBG("%s: flush", mdp4_crtc->name); + DBG("%s: event: %p", mdp4_crtc->name, crtc->state->event); WARN_ON(mdp4_crtc->event); diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c index 0e9a2e3a82d7..f021f960a8a2 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c @@ -303,11 +303,6 @@ static int mdp5_crtc_atomic_check(struct drm_crtc *crtc, DBG("%s: check", mdp5_crtc->name); - if (mdp5_crtc->event) { - dev_err(dev->dev, "already pending flip!\n"); - return -EBUSY; - } - /* request a free CTL, if none is already allocated for this CRTC */ if (state->enable && !mdp5_crtc->ctl) { mdp5_crtc->ctl = mdp5_ctlm_request(mdp5_kms->ctlm, crtc); @@ -364,7 +359,7 @@ static void mdp5_crtc_atomic_flush(struct drm_crtc *crtc) struct drm_device *dev = crtc->dev; unsigned long flags; - DBG("%s: flush", mdp5_crtc->name); + DBG("%s: event: %p", mdp5_crtc->name, crtc->state->event); WARN_ON(mdp5_crtc->event); @@ -460,10 +455,7 @@ void mdp5_crtc_set_intf(struct drm_crtc *crtc, int intf, /* now that we know what irq's we want: */ mdp5_crtc->err.irqmask = intf2err(intf); mdp5_crtc->vblank.irqmask = intf2vblank(intf); - - /* when called from modeset_init(), skip the rest until later: */ - if (!mdp5_kms) - return; + mdp_irq_update(&mdp5_kms->base); spin_lock_irqsave(&mdp5_kms->resource_lock, flags); intf_sel = mdp5_read(mdp5_kms, REG_MDP5_DISP_INTF_SEL); diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c index a11f1b80c488..9f01a4f21af2 100644 --- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c +++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c @@ -216,17 +216,7 @@ static int modeset_init(struct mdp5_kms *mdp5_kms) goto fail; } - /* NOTE: the vsync and error irq's are actually associated with - * the INTF/encoder.. the easiest way to deal with this (ie. what - * we do now) is assume a fixed relationship between crtc's and - * encoders. I'm not sure if there is ever a need to more freely - * assign crtcs to encoders, but if there is then we need to take - * care of error and vblank irq's that the crtc has registered, - * and also update user-requested vblank_mask. - */ - encoder->possible_crtcs = BIT(0); - mdp5_crtc_set_intf(priv->crtcs[0], 3, INTF_HDMI); - + encoder->possible_crtcs = (1 << priv->num_crtcs) - 1;; priv->encoders[priv->num_encoders++] = encoder; /* Construct bridge/connector for HDMI: */ diff --git a/drivers/gpu/drm/msm/mdp/mdp_kms.c b/drivers/gpu/drm/msm/mdp/mdp_kms.c index 03455b64a245..2a731722d840 100644 --- a/drivers/gpu/drm/msm/mdp/mdp_kms.c +++ b/drivers/gpu/drm/msm/mdp/mdp_kms.c @@ -42,7 +42,10 @@ static void update_irq(struct mdp_kms *mdp_kms) mdp_kms->funcs->set_irqmask(mdp_kms, irqmask); } -static void update_irq_unlocked(struct mdp_kms *mdp_kms) +/* if an mdp_irq's irqmask has changed, such as when mdp5 crtc<->encoder + * link changes, this must be called to figure out the new global irqmask + */ +void mdp_irq_update(struct mdp_kms *mdp_kms) { unsigned long flags; spin_lock_irqsave(&list_lock, flags); @@ -122,7 +125,7 @@ void mdp_irq_register(struct mdp_kms *mdp_kms, struct mdp_irq *irq) spin_unlock_irqrestore(&list_lock, flags); if (needs_update) - update_irq_unlocked(mdp_kms); + mdp_irq_update(mdp_kms); } void mdp_irq_unregister(struct mdp_kms *mdp_kms, struct mdp_irq *irq) @@ -141,5 +144,5 @@ void mdp_irq_unregister(struct mdp_kms *mdp_kms, struct mdp_irq *irq) spin_unlock_irqrestore(&list_lock, flags); if (needs_update) - update_irq_unlocked(mdp_kms); + mdp_irq_update(mdp_kms); } diff --git a/drivers/gpu/drm/msm/mdp/mdp_kms.h b/drivers/gpu/drm/msm/mdp/mdp_kms.h index 99557b5ad4fd..b268ce95d394 100644 --- a/drivers/gpu/drm/msm/mdp/mdp_kms.h +++ b/drivers/gpu/drm/msm/mdp/mdp_kms.h @@ -75,7 +75,7 @@ void mdp_update_vblank_mask(struct mdp_kms *mdp_kms, uint32_t mask, bool enable) void mdp_irq_wait(struct mdp_kms *mdp_kms, uint32_t irqmask); void mdp_irq_register(struct mdp_kms *mdp_kms, struct mdp_irq *irq); void mdp_irq_unregister(struct mdp_kms *mdp_kms, struct mdp_irq *irq); - +void mdp_irq_update(struct mdp_kms *mdp_kms); /* * pixel format helpers: diff --git a/drivers/gpu/drm/msm/msm_atomic.c b/drivers/gpu/drm/msm/msm_atomic.c index f0de412e13dc..191968256c58 100644 --- a/drivers/gpu/drm/msm/msm_atomic.c +++ b/drivers/gpu/drm/msm/msm_atomic.c @@ -23,10 +23,41 @@ struct msm_commit { struct drm_atomic_state *state; uint32_t fence; struct msm_fence_cb fence_cb; + uint32_t crtc_mask; }; static void fence_cb(struct msm_fence_cb *cb); +/* block until specified crtcs are no longer pending update, and + * atomically mark them as pending update + */ +static int start_atomic(struct msm_drm_private *priv, uint32_t crtc_mask) +{ + int ret; + + spin_lock(&priv->pending_crtcs_event.lock); + ret = wait_event_interruptible_locked(priv->pending_crtcs_event, + !(priv->pending_crtcs & crtc_mask)); + if (ret == 0) { + DBG("start: %08x", crtc_mask); + priv->pending_crtcs |= crtc_mask; + } + spin_unlock(&priv->pending_crtcs_event.lock); + + return ret; +} + +/* clear specified crtcs (no longer pending update) + */ +static void end_atomic(struct msm_drm_private *priv, uint32_t crtc_mask) +{ + spin_lock(&priv->pending_crtcs_event.lock); + DBG("end: %08x", crtc_mask); + priv->pending_crtcs &= ~crtc_mask; + wake_up_all_locked(&priv->pending_crtcs_event); + spin_unlock(&priv->pending_crtcs_event.lock); +} + static struct msm_commit *new_commit(struct drm_atomic_state *state) { struct msm_commit *c = kzalloc(sizeof(*c), GFP_KERNEL); @@ -58,12 +89,27 @@ static void complete_commit(struct msm_commit *c) drm_atomic_helper_commit_post_planes(dev, state); + /* NOTE: _wait_for_vblanks() only waits for vblank on + * enabled CRTCs. So we end up faulting when disabling + * due to (potentially) unref'ing the outgoing fb's + * before the vblank when the disable has latched. + * + * But if it did wait on disabled (or newly disabled) + * CRTCs, that would be racy (ie. we could have missed + * the irq. We need some way to poll for pipe shut + * down. Or just live with occasionally hitting the + * timeout in the CRTC disable path (which really should + * not be critical path) + */ + drm_atomic_helper_wait_for_vblanks(dev, state); drm_atomic_helper_cleanup_planes(dev, state); drm_atomic_state_free(state); + end_atomic(dev->dev_private, c->crtc_mask); + kfree(c); } @@ -97,8 +143,9 @@ static void add_fb(struct msm_commit *c, struct drm_framebuffer *fb) int msm_atomic_commit(struct drm_device *dev, struct drm_atomic_state *state, bool async) { - struct msm_commit *c; int nplanes = dev->mode_config.num_total_plane; + int ncrtcs = dev->mode_config.num_crtc; + struct msm_commit *c; int i, ret; ret = drm_atomic_helper_prepare_planes(dev, state); @@ -106,6 +153,18 @@ int msm_atomic_commit(struct drm_device *dev, return ret; c = new_commit(state); + if (!c) + return -ENOMEM; + + /* + * Figure out what crtcs we have: + */ + for (i = 0; i < ncrtcs; i++) { + struct drm_crtc *crtc = state->crtcs[i]; + if (!crtc) + continue; + c->crtc_mask |= (1 << drm_crtc_index(crtc)); + } /* * Figure out what fence to wait for: @@ -122,6 +181,14 @@ int msm_atomic_commit(struct drm_device *dev, } /* + * Wait for pending updates on any of the same crtc's and then + * mark our set of crtc's as busy: + */ + ret = start_atomic(dev->dev_private, c->crtc_mask); + if (ret) + return ret; + + /* * This is the point of no return - everything below never fails except * when the hw goes bonghits. Which means we can commit the new state on * the software side now. diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index d3b791b7ddef..9a61546a0b05 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -193,6 +193,7 @@ static int msm_load(struct drm_device *dev, unsigned long flags) priv->wq = alloc_ordered_workqueue("msm", 0); init_waitqueue_head(&priv->fence_event); + init_waitqueue_head(&priv->pending_crtcs_event); INIT_LIST_HEAD(&priv->inactive_list); INIT_LIST_HEAD(&priv->fence_cbs); @@ -1012,7 +1013,6 @@ static struct platform_driver msm_platform_driver = { .probe = msm_pdev_probe, .remove = msm_pdev_remove, .driver = { - .owner = THIS_MODULE, .name = "msm", .of_match_table = dt_match, .pm = &msm_pm_ops, diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h index 136303818436..b69ef2d5a26c 100644 --- a/drivers/gpu/drm/msm/msm_drv.h +++ b/drivers/gpu/drm/msm/msm_drv.h @@ -96,6 +96,10 @@ struct msm_drm_private { /* callbacks deferred until bo is inactive: */ struct list_head fence_cbs; + /* crtcs pending async atomic updates: */ + uint32_t pending_crtcs; + wait_queue_head_t pending_crtcs_event; + /* registered MMUs: */ unsigned int num_mmus; struct msm_mmu *mmus[NUM_DOMAINS]; diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c index 94d55e526b4e..1f3af13ccede 100644 --- a/drivers/gpu/drm/msm/msm_fbdev.c +++ b/drivers/gpu/drm/msm/msm_fbdev.c @@ -190,8 +190,7 @@ fail_unlock: fail: if (ret) { - if (fbi) - framebuffer_release(fbi); + framebuffer_release(fbi); if (fb) { drm_framebuffer_unregister_private(fb); drm_framebuffer_remove(fb); diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index 4a6f0e49d5b5..49dea4fb55ac 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -535,8 +535,7 @@ void msm_gem_free_object(struct drm_gem_object *obj) drm_free_large(msm_obj->pages); } else { - if (msm_obj->vaddr) - vunmap(msm_obj->vaddr); + vunmap(msm_obj->vaddr); put_pages(obj); } diff --git a/drivers/gpu/drm/nouveau/core/core/event.c b/drivers/gpu/drm/nouveau/core/core/event.c index ff2b434b3db4..760947e380c9 100644 --- a/drivers/gpu/drm/nouveau/core/core/event.c +++ b/drivers/gpu/drm/nouveau/core/core/event.c @@ -26,7 +26,7 @@ void nvkm_event_put(struct nvkm_event *event, u32 types, int index) { - BUG_ON(!spin_is_locked(&event->refs_lock)); + assert_spin_locked(&event->refs_lock); while (types) { int type = __ffs(types); types &= ~(1 << type); if (--event->refs[index * event->types_nr + type] == 0) { @@ -39,7 +39,7 @@ nvkm_event_put(struct nvkm_event *event, u32 types, int index) void nvkm_event_get(struct nvkm_event *event, u32 types, int index) { - BUG_ON(!spin_is_locked(&event->refs_lock)); + assert_spin_locked(&event->refs_lock); while (types) { int type = __ffs(types); types &= ~(1 << type); if (++event->refs[index * event->types_nr + type] == 1) { diff --git a/drivers/gpu/drm/nouveau/core/core/notify.c b/drivers/gpu/drm/nouveau/core/core/notify.c index d1bcde55e9d7..839a32577680 100644 --- a/drivers/gpu/drm/nouveau/core/core/notify.c +++ b/drivers/gpu/drm/nouveau/core/core/notify.c @@ -98,7 +98,7 @@ nvkm_notify_send(struct nvkm_notify *notify, void *data, u32 size) struct nvkm_event *event = notify->event; unsigned long flags; - BUG_ON(!spin_is_locked(&event->list_lock)); + assert_spin_locked(&event->list_lock); BUG_ON(size != notify->size); spin_lock_irqsave(&event->refs_lock, flags); diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c index cd05677ad4b7..72a40f95d048 100644 --- a/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c +++ b/drivers/gpu/drm/nouveau/core/engine/device/nvc0.c @@ -218,7 +218,6 @@ nvc0_identify(struct nouveau_device *device) device->oclass[NVDEV_ENGINE_BSP ] = &nvc0_bsp_oclass; device->oclass[NVDEV_ENGINE_PPP ] = &nvc0_ppp_oclass; device->oclass[NVDEV_ENGINE_COPY0 ] = &nvc0_copy0_oclass; - device->oclass[NVDEV_ENGINE_COPY1 ] = &nvc0_copy1_oclass; device->oclass[NVDEV_ENGINE_DISP ] = nva3_disp_oclass; device->oclass[NVDEV_ENGINE_PERFMON] = &nvc0_perfmon_oclass; break; diff --git a/drivers/gpu/drm/nouveau/core/engine/device/nve0.c b/drivers/gpu/drm/nouveau/core/engine/device/nve0.c index 674da1f095b2..732922690653 100644 --- a/drivers/gpu/drm/nouveau/core/engine/device/nve0.c +++ b/drivers/gpu/drm/nouveau/core/engine/device/nve0.c @@ -249,6 +249,39 @@ nve0_identify(struct nouveau_device *device) device->oclass[NVDEV_ENGINE_PPP ] = &nvc0_ppp_oclass; device->oclass[NVDEV_ENGINE_PERFMON] = &nvf0_perfmon_oclass; break; + case 0x106: + device->cname = "GK208B"; + device->oclass[NVDEV_SUBDEV_VBIOS ] = &nouveau_bios_oclass; + device->oclass[NVDEV_SUBDEV_GPIO ] = nve0_gpio_oclass; + device->oclass[NVDEV_SUBDEV_I2C ] = nve0_i2c_oclass; + device->oclass[NVDEV_SUBDEV_FUSE ] = &gf100_fuse_oclass; + device->oclass[NVDEV_SUBDEV_CLOCK ] = &nve0_clock_oclass; + device->oclass[NVDEV_SUBDEV_THERM ] = &nvd0_therm_oclass; + device->oclass[NVDEV_SUBDEV_MXM ] = &nv50_mxm_oclass; + device->oclass[NVDEV_SUBDEV_DEVINIT] = nvc0_devinit_oclass; + device->oclass[NVDEV_SUBDEV_MC ] = gk20a_mc_oclass; + device->oclass[NVDEV_SUBDEV_BUS ] = nvc0_bus_oclass; + device->oclass[NVDEV_SUBDEV_TIMER ] = &nv04_timer_oclass; + device->oclass[NVDEV_SUBDEV_FB ] = nve0_fb_oclass; + device->oclass[NVDEV_SUBDEV_LTC ] = gk104_ltc_oclass; + device->oclass[NVDEV_SUBDEV_IBUS ] = &nve0_ibus_oclass; + device->oclass[NVDEV_SUBDEV_INSTMEM] = nv50_instmem_oclass; + device->oclass[NVDEV_SUBDEV_VM ] = &nvc0_vmmgr_oclass; + device->oclass[NVDEV_SUBDEV_BAR ] = &nvc0_bar_oclass; + device->oclass[NVDEV_SUBDEV_PWR ] = nv108_pwr_oclass; + device->oclass[NVDEV_SUBDEV_VOLT ] = &nv40_volt_oclass; + device->oclass[NVDEV_ENGINE_DMAOBJ ] = nvd0_dmaeng_oclass; + device->oclass[NVDEV_ENGINE_FIFO ] = nv108_fifo_oclass; + device->oclass[NVDEV_ENGINE_SW ] = nvc0_software_oclass; + device->oclass[NVDEV_ENGINE_GR ] = nv108_graph_oclass; + device->oclass[NVDEV_ENGINE_DISP ] = nvf0_disp_oclass; + device->oclass[NVDEV_ENGINE_COPY0 ] = &nve0_copy0_oclass; + device->oclass[NVDEV_ENGINE_COPY1 ] = &nve0_copy1_oclass; + device->oclass[NVDEV_ENGINE_COPY2 ] = &nve0_copy2_oclass; + device->oclass[NVDEV_ENGINE_BSP ] = &nve0_bsp_oclass; + device->oclass[NVDEV_ENGINE_VP ] = &nve0_vp_oclass; + device->oclass[NVDEV_ENGINE_PPP ] = &nvc0_ppp_oclass; + break; case 0x108: device->cname = "GK208"; device->oclass[NVDEV_SUBDEV_VBIOS ] = &nouveau_bios_oclass; diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c index 5ae6a43893b5..1931057f9962 100644 --- a/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c +++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nv04.c @@ -551,8 +551,8 @@ nv04_fifo_intr(struct nouveau_subdev *subdev) } if (status & 0x40000000) { - nouveau_fifo_uevent(&priv->base); nv_wr32(priv, 0x002100, 0x40000000); + nouveau_fifo_uevent(&priv->base); status &= ~0x40000000; } } diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c index 1fe1f8fbda0c..074d434c3077 100644 --- a/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c +++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nvc0.c @@ -740,6 +740,8 @@ nvc0_fifo_intr_engine_unit(struct nvc0_fifo_priv *priv, int engn) u32 inte = nv_rd32(priv, 0x002628); u32 unkn; + nv_wr32(priv, 0x0025a8 + (engn * 0x04), intr); + for (unkn = 0; unkn < 8; unkn++) { u32 ints = (intr >> (unkn * 0x04)) & inte; if (ints & 0x1) { @@ -751,8 +753,6 @@ nvc0_fifo_intr_engine_unit(struct nvc0_fifo_priv *priv, int engn) nv_mask(priv, 0x002628, ints, 0); } } - - nv_wr32(priv, 0x0025a8 + (engn * 0x04), intr); } static void diff --git a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c index fc9ef663f25a..6a8db7c80bd1 100644 --- a/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c +++ b/drivers/gpu/drm/nouveau/core/engine/fifo/nve0.c @@ -982,8 +982,8 @@ nve0_fifo_intr(struct nouveau_subdev *subdev) } if (stat & 0x80000000) { - nve0_fifo_intr_engine(priv); nv_wr32(priv, 0x002100, 0x80000000); + nve0_fifo_intr_engine(priv); stat &= ~0x80000000; } diff --git a/drivers/gpu/drm/nouveau/core/subdev/bios/shadowramin.c b/drivers/gpu/drm/nouveau/core/subdev/bios/shadowramin.c index 5e58bba0dd5c..a7a890fad1e5 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/bios/shadowramin.c +++ b/drivers/gpu/drm/nouveau/core/subdev/bios/shadowramin.c @@ -44,8 +44,10 @@ static void pramin_fini(void *data) { struct priv *priv = data; - nv_wr32(priv->bios, 0x001700, priv->bar0); - kfree(priv); + if (priv) { + nv_wr32(priv->bios, 0x001700, priv->bar0); + kfree(priv); + } } static void * diff --git a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvaa.c b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvaa.c index 00f2ca7e44a5..033a8e999497 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvaa.c +++ b/drivers/gpu/drm/nouveau/core/subdev/fb/ramnvaa.c @@ -24,34 +24,71 @@ #include "nv50.h" +struct nvaa_ram_priv { + struct nouveau_ram base; + u64 poller_base; +}; + static int nvaa_ram_ctor(struct nouveau_object *parent, struct nouveau_object *engine, struct nouveau_oclass *oclass, void *data, u32 datasize, struct nouveau_object **pobject) { - const u32 rsvd_head = ( 256 * 1024) >> 12; /* vga memory */ - const u32 rsvd_tail = (1024 * 1024) >> 12; /* vbios etc */ + u32 rsvd_head = ( 256 * 1024); /* vga memory */ + u32 rsvd_tail = (1024 * 1024); /* vbios etc */ struct nouveau_fb *pfb = nouveau_fb(parent); - struct nouveau_ram *ram; + struct nvaa_ram_priv *priv; int ret; - ret = nouveau_ram_create(parent, engine, oclass, &ram); - *pobject = nv_object(ram); + ret = nouveau_ram_create(parent, engine, oclass, &priv); + *pobject = nv_object(priv); if (ret) return ret; - ram->size = nv_rd32(pfb, 0x10020c); - ram->size = (ram->size & 0xffffff00) | ((ram->size & 0x000000ff) << 32); + priv->base.type = NV_MEM_TYPE_STOLEN; + priv->base.stolen = (u64)nv_rd32(pfb, 0x100e10) << 12; + priv->base.size = (u64)nv_rd32(pfb, 0x100e14) << 12; - ret = nouveau_mm_init(&pfb->vram, rsvd_head, (ram->size >> 12) - - (rsvd_head + rsvd_tail), 1); + rsvd_tail += 0x1000; + priv->poller_base = priv->base.size - rsvd_tail; + + ret = nouveau_mm_init(&pfb->vram, rsvd_head >> 12, + (priv->base.size - (rsvd_head + rsvd_tail)) >> 12, + 1); if (ret) return ret; - ram->type = NV_MEM_TYPE_STOLEN; - ram->stolen = (u64)nv_rd32(pfb, 0x100e10) << 12; - ram->get = nv50_ram_get; - ram->put = nv50_ram_put; + priv->base.get = nv50_ram_get; + priv->base.put = nv50_ram_put; + return 0; +} + +static int +nvaa_ram_init(struct nouveau_object *object) +{ + struct nouveau_fb *pfb = nouveau_fb(object); + struct nvaa_ram_priv *priv = (void *)object; + int ret; + u64 dniso, hostnb, flush; + + ret = nouveau_ram_init(&priv->base); + if (ret) + return ret; + + dniso = ((priv->base.size - (priv->poller_base + 0x00)) >> 5) - 1; + hostnb = ((priv->base.size - (priv->poller_base + 0x20)) >> 5) - 1; + flush = ((priv->base.size - (priv->poller_base + 0x40)) >> 5) - 1; + + /* Enable NISO poller for various clients and set their associated + * read address, only for MCP77/78 and MCP79/7A. (fd#25701) + */ + nv_wr32(pfb, 0x100c18, dniso); + nv_mask(pfb, 0x100c14, 0x00000000, 0x00000001); + nv_wr32(pfb, 0x100c1c, hostnb); + nv_mask(pfb, 0x100c14, 0x00000000, 0x00000002); + nv_wr32(pfb, 0x100c24, flush); + nv_mask(pfb, 0x100c14, 0x00000000, 0x00010000); + return 0; } @@ -60,7 +97,7 @@ nvaa_ram_oclass = { .ofuncs = &(struct nouveau_ofuncs) { .ctor = nvaa_ram_ctor, .dtor = _nouveau_ram_dtor, - .init = _nouveau_ram_init, + .init = nvaa_ram_init, .fini = _nouveau_ram_fini, }, }; diff --git a/drivers/gpu/drm/nouveau/core/subdev/mc/nv4c.c b/drivers/gpu/drm/nouveau/core/subdev/mc/nv4c.c index a75c35ccf25c..165401c4045c 100644 --- a/drivers/gpu/drm/nouveau/core/subdev/mc/nv4c.c +++ b/drivers/gpu/drm/nouveau/core/subdev/mc/nv4c.c @@ -24,13 +24,6 @@ #include "nv04.h" -static void -nv4c_mc_msi_rearm(struct nouveau_mc *pmc) -{ - struct nv04_mc_priv *priv = (void *)pmc; - nv_wr08(priv, 0x088050, 0xff); -} - struct nouveau_oclass * nv4c_mc_oclass = &(struct nouveau_mc_oclass) { .base.handle = NV_SUBDEV(MC, 0x4c), @@ -41,5 +34,4 @@ nv4c_mc_oclass = &(struct nouveau_mc_oclass) { .fini = _nouveau_mc_fini, }, .intr = nv04_mc_intr, - .msi_rearm = nv4c_mc_msi_rearm, }.base; diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index 21ec561edc99..bba2960d3dfb 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -1572,8 +1572,10 @@ nouveau_ttm_tt_unpopulate(struct ttm_tt *ttm) * so use the DMA API for them. */ if (!nv_device_is_cpu_coherent(device) && - ttm->caching_state == tt_uncached) + ttm->caching_state == tt_uncached) { ttm_dma_unpopulate(ttm_dma, dev->dev); + return; + } #if __OS_HAS_AGP if (drm->agp.stat == ENABLED) { diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c index 5d93902a91ab..f8042433752b 100644 --- a/drivers/gpu/drm/nouveau/nouveau_display.c +++ b/drivers/gpu/drm/nouveau/nouveau_display.c @@ -876,7 +876,6 @@ nouveau_display_dumb_create(struct drm_file *file_priv, struct drm_device *dev, if (ret) return ret; - bo->gem.dumb = true; ret = drm_gem_handle_create(file_priv, &bo->gem, &args->handle); drm_gem_object_unreference_unlocked(&bo->gem); return ret; @@ -892,14 +891,6 @@ nouveau_display_dumb_map_offset(struct drm_file *file_priv, gem = drm_gem_object_lookup(dev, file_priv, handle); if (gem) { struct nouveau_bo *bo = nouveau_gem_object(gem); - - /* - * We don't allow dumb mmaps on objects created using another - * interface. - */ - WARN_ONCE(!(gem->dumb || gem->import_attach), - "Illegal dumb map of accelerated buffer.\n"); - *poffset = drm_vma_node_offset_addr(&bo->bo.vma_node); drm_gem_object_unreference_unlocked(gem); return 0; diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index afb93bb72f97..65910e3aed0c 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -664,7 +664,6 @@ nouveau_pmops_suspend(struct device *dev) pci_save_state(pdev); pci_disable_device(pdev); - pci_ignore_hotplug(pdev); pci_set_power_state(pdev, PCI_D3hot); return 0; } @@ -732,6 +731,7 @@ nouveau_pmops_runtime_suspend(struct device *dev) ret = nouveau_do_suspend(drm_dev, true); pci_save_state(pdev); pci_disable_device(pdev); + pci_ignore_hotplug(pdev); pci_set_power_state(pdev, PCI_D3cold); drm_dev->switch_power_state = DRM_SWITCH_POWER_DYNAMIC_OFF; return ret; diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c index 515cd9aebb99..f32a434724e3 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fence.c +++ b/drivers/gpu/drm/nouveau/nouveau_fence.c @@ -52,20 +52,24 @@ nouveau_fctx(struct nouveau_fence *fence) return container_of(fence->base.lock, struct nouveau_fence_chan, lock); } -static void +static int nouveau_fence_signal(struct nouveau_fence *fence) { + int drop = 0; + fence_signal_locked(&fence->base); list_del(&fence->head); + rcu_assign_pointer(fence->channel, NULL); if (test_bit(FENCE_FLAG_USER_BITS, &fence->base.flags)) { struct nouveau_fence_chan *fctx = nouveau_fctx(fence); if (!--fctx->notify_ref) - nvif_notify_put(&fctx->notify); + drop = 1; } fence_put(&fence->base); + return drop; } static struct nouveau_fence * @@ -88,16 +92,23 @@ nouveau_fence_context_del(struct nouveau_fence_chan *fctx) { struct nouveau_fence *fence; - nvif_notify_fini(&fctx->notify); - spin_lock_irq(&fctx->lock); while (!list_empty(&fctx->pending)) { fence = list_entry(fctx->pending.next, typeof(*fence), head); - nouveau_fence_signal(fence); - fence->channel = NULL; + if (nouveau_fence_signal(fence)) + nvif_notify_put(&fctx->notify); } spin_unlock_irq(&fctx->lock); + + nvif_notify_fini(&fctx->notify); + fctx->dead = 1; + + /* + * Ensure that all accesses to fence->channel complete before freeing + * the channel. + */ + synchronize_rcu(); } static void @@ -112,21 +123,23 @@ nouveau_fence_context_free(struct nouveau_fence_chan *fctx) kref_put(&fctx->fence_ref, nouveau_fence_context_put); } -static void +static int nouveau_fence_update(struct nouveau_channel *chan, struct nouveau_fence_chan *fctx) { struct nouveau_fence *fence; - + int drop = 0; u32 seq = fctx->read(chan); while (!list_empty(&fctx->pending)) { fence = list_entry(fctx->pending.next, typeof(*fence), head); if ((int)(seq - fence->base.seqno) < 0) - return; + break; - nouveau_fence_signal(fence); + drop |= nouveau_fence_signal(fence); } + + return drop; } static int @@ -135,18 +148,21 @@ nouveau_fence_wait_uevent_handler(struct nvif_notify *notify) struct nouveau_fence_chan *fctx = container_of(notify, typeof(*fctx), notify); unsigned long flags; + int ret = NVIF_NOTIFY_KEEP; spin_lock_irqsave(&fctx->lock, flags); if (!list_empty(&fctx->pending)) { struct nouveau_fence *fence; + struct nouveau_channel *chan; fence = list_entry(fctx->pending.next, typeof(*fence), head); - nouveau_fence_update(fence->channel, fctx); + chan = rcu_dereference_protected(fence->channel, lockdep_is_held(&fctx->lock)); + if (nouveau_fence_update(fence->channel, fctx)) + ret = NVIF_NOTIFY_DROP; } spin_unlock_irqrestore(&fctx->lock, flags); - /* Always return keep here. NVIF refcount is handled with nouveau_fence_update */ - return NVIF_NOTIFY_KEEP; + return ret; } void @@ -262,7 +278,10 @@ nouveau_fence_emit(struct nouveau_fence *fence, struct nouveau_channel *chan) if (!ret) { fence_get(&fence->base); spin_lock_irq(&fctx->lock); - nouveau_fence_update(chan, fctx); + + if (nouveau_fence_update(chan, fctx)) + nvif_notify_put(&fctx->notify); + list_add_tail(&fence->head, &fctx->pending); spin_unlock_irq(&fctx->lock); } @@ -276,13 +295,16 @@ nouveau_fence_done(struct nouveau_fence *fence) if (fence->base.ops == &nouveau_fence_ops_legacy || fence->base.ops == &nouveau_fence_ops_uevent) { struct nouveau_fence_chan *fctx = nouveau_fctx(fence); + struct nouveau_channel *chan; unsigned long flags; if (test_bit(FENCE_FLAG_SIGNALED_BIT, &fence->base.flags)) return true; spin_lock_irqsave(&fctx->lock, flags); - nouveau_fence_update(fence->channel, fctx); + chan = rcu_dereference_protected(fence->channel, lockdep_is_held(&fctx->lock)); + if (chan && nouveau_fence_update(chan, fctx)) + nvif_notify_put(&fctx->notify); spin_unlock_irqrestore(&fctx->lock, flags); } return fence_is_signaled(&fence->base); @@ -387,12 +409,18 @@ nouveau_fence_sync(struct nouveau_bo *nvbo, struct nouveau_channel *chan, bool e if (fence && (!exclusive || !fobj || !fobj->shared_count)) { struct nouveau_channel *prev = NULL; + bool must_wait = true; f = nouveau_local_fence(fence, chan->drm); - if (f) - prev = f->channel; + if (f) { + rcu_read_lock(); + prev = rcu_dereference(f->channel); + if (prev && (prev == chan || fctx->sync(f, prev, chan) == 0)) + must_wait = false; + rcu_read_unlock(); + } - if (!prev || (prev != chan && (ret = fctx->sync(f, prev, chan)))) + if (must_wait) ret = fence_wait(fence, intr); return ret; @@ -403,19 +431,22 @@ nouveau_fence_sync(struct nouveau_bo *nvbo, struct nouveau_channel *chan, bool e for (i = 0; i < fobj->shared_count && !ret; ++i) { struct nouveau_channel *prev = NULL; + bool must_wait = true; fence = rcu_dereference_protected(fobj->shared[i], reservation_object_held(resv)); f = nouveau_local_fence(fence, chan->drm); - if (f) - prev = f->channel; + if (f) { + rcu_read_lock(); + prev = rcu_dereference(f->channel); + if (prev && (prev == chan || fctx->sync(f, prev, chan) == 0)) + must_wait = false; + rcu_read_unlock(); + } - if (!prev || (prev != chan && (ret = fctx->sync(f, prev, chan)))) + if (must_wait) ret = fence_wait(fence, intr); - - if (ret) - break; } return ret; @@ -463,7 +494,7 @@ static const char *nouveau_fence_get_timeline_name(struct fence *f) struct nouveau_fence *fence = from_fence(f); struct nouveau_fence_chan *fctx = nouveau_fctx(fence); - return fence->channel ? fctx->name : "dead channel"; + return !fctx->dead ? fctx->name : "dead channel"; } /* @@ -476,9 +507,16 @@ static bool nouveau_fence_is_signaled(struct fence *f) { struct nouveau_fence *fence = from_fence(f); struct nouveau_fence_chan *fctx = nouveau_fctx(fence); - struct nouveau_channel *chan = fence->channel; + struct nouveau_channel *chan; + bool ret = false; + + rcu_read_lock(); + chan = rcu_dereference(fence->channel); + if (chan) + ret = (int)(fctx->read(chan) - fence->base.seqno) >= 0; + rcu_read_unlock(); - return (int)(fctx->read(chan) - fence->base.seqno) >= 0; + return ret; } static bool nouveau_fence_no_signaling(struct fence *f) diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.h b/drivers/gpu/drm/nouveau/nouveau_fence.h index 943b0b17b1fc..96e461c6f68f 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fence.h +++ b/drivers/gpu/drm/nouveau/nouveau_fence.h @@ -14,7 +14,7 @@ struct nouveau_fence { bool sysmem; - struct nouveau_channel *channel; + struct nouveau_channel __rcu *channel; unsigned long timeout; }; @@ -47,7 +47,7 @@ struct nouveau_fence_chan { char name[32]; struct nvif_notify notify; - int notify_ref; + int notify_ref, dead; }; struct nouveau_fence_priv { diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c index 28d51a22a4bf..bf0f9e21d714 100644 --- a/drivers/gpu/drm/nouveau/nouveau_gem.c +++ b/drivers/gpu/drm/nouveau/nouveau_gem.c @@ -36,7 +36,14 @@ void nouveau_gem_object_del(struct drm_gem_object *gem) { struct nouveau_bo *nvbo = nouveau_gem_object(gem); + struct nouveau_drm *drm = nouveau_bdev(nvbo->bo.bdev); struct ttm_buffer_object *bo = &nvbo->bo; + struct device *dev = drm->dev->dev; + int ret; + + ret = pm_runtime_get_sync(dev); + if (WARN_ON(ret < 0 && ret != -EACCES)) + return; if (gem->import_attach) drm_prime_gem_destroy(gem, nvbo->bo.sg); @@ -46,6 +53,9 @@ nouveau_gem_object_del(struct drm_gem_object *gem) /* reset filp so nouveau_bo_del_ttm() can test for it */ gem->filp = NULL; ttm_bo_unref(&bo); + + pm_runtime_mark_last_busy(dev); + pm_runtime_put_autosuspend(dev); } int @@ -53,7 +63,9 @@ nouveau_gem_object_open(struct drm_gem_object *gem, struct drm_file *file_priv) { struct nouveau_cli *cli = nouveau_cli(file_priv); struct nouveau_bo *nvbo = nouveau_gem_object(gem); + struct nouveau_drm *drm = nouveau_bdev(nvbo->bo.bdev); struct nouveau_vma *vma; + struct device *dev = drm->dev->dev; int ret; if (!cli->vm) @@ -71,11 +83,16 @@ nouveau_gem_object_open(struct drm_gem_object *gem, struct drm_file *file_priv) goto out; } + ret = pm_runtime_get_sync(dev); + if (ret < 0 && ret != -EACCES) + goto out; + ret = nouveau_bo_vma_add(nvbo, cli->vm, vma); - if (ret) { + if (ret) kfree(vma); - goto out; - } + + pm_runtime_mark_last_busy(dev); + pm_runtime_put_autosuspend(dev); } else { vma->refcount++; } @@ -129,6 +146,8 @@ nouveau_gem_object_close(struct drm_gem_object *gem, struct drm_file *file_priv) { struct nouveau_cli *cli = nouveau_cli(file_priv); struct nouveau_bo *nvbo = nouveau_gem_object(gem); + struct nouveau_drm *drm = nouveau_bdev(nvbo->bo.bdev); + struct device *dev = drm->dev->dev; struct nouveau_vma *vma; int ret; @@ -141,8 +160,14 @@ nouveau_gem_object_close(struct drm_gem_object *gem, struct drm_file *file_priv) vma = nouveau_bo_vma_find(nvbo, cli->vm); if (vma) { - if (--vma->refcount == 0) - nouveau_gem_object_unmap(nvbo, vma); + if (--vma->refcount == 0) { + ret = pm_runtime_get_sync(dev); + if (!WARN_ON(ret < 0 && ret != -EACCES)) { + nouveau_gem_object_unmap(nvbo, vma); + pm_runtime_mark_last_busy(dev); + pm_runtime_put_autosuspend(dev); + } + } } ttm_bo_unreserve(&nvbo->bo); } @@ -444,9 +469,6 @@ validate_list(struct nouveau_channel *chan, struct nouveau_cli *cli, list_for_each_entry(nvbo, list, entry) { struct drm_nouveau_gem_pushbuf_bo *b = &pbbo[nvbo->pbbo_index]; - WARN_ONCE(nvbo->gem.dumb, - "GPU use of dumb buffer is illegal.\n"); - ret = nouveau_gem_set_domain(&nvbo->gem, b->read_domains, b->write_domains, b->valid_domains); diff --git a/drivers/gpu/drm/nouveau/nouveau_ttm.c b/drivers/gpu/drm/nouveau/nouveau_ttm.c index 753a6def61e7..3d1cfcb96b6b 100644 --- a/drivers/gpu/drm/nouveau/nouveau_ttm.c +++ b/drivers/gpu/drm/nouveau/nouveau_ttm.c @@ -28,6 +28,7 @@ #include "nouveau_ttm.h" #include "nouveau_gem.h" +#include "drm_legacy.h" static int nouveau_vram_manager_init(struct ttm_mem_type_manager *man, unsigned long psize) { @@ -281,7 +282,7 @@ nouveau_ttm_mmap(struct file *filp, struct vm_area_struct *vma) struct nouveau_drm *drm = nouveau_drm(file_priv->minor->dev); if (unlikely(vma->vm_pgoff < DRM_FILE_PAGE_OFFSET)) - return -EINVAL; + return drm_legacy_mmap(filp, vma); return ttm_bo_mmap(filp, vma, &drm->ttm.bdev); } diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c index 862ba03c236c..8241ed9b353c 100644 --- a/drivers/gpu/drm/omapdrm/omap_drv.c +++ b/drivers/gpu/drm/omapdrm/omap_drv.c @@ -718,7 +718,6 @@ static const struct dev_pm_ops omapdrm_pm_ops = { static struct platform_driver pdev = { .driver = { .name = DRIVER_NAME, - .owner = THIS_MODULE, #ifdef CONFIG_PM .pm = &omapdrm_pm_ops, #endif diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c index c4b6167a8bf3..e95385bf8356 100644 --- a/drivers/gpu/drm/panel/panel-simple.c +++ b/drivers/gpu/drm/panel/panel-simple.c @@ -807,7 +807,6 @@ static void panel_simple_platform_shutdown(struct platform_device *pdev) static struct platform_driver panel_simple_platform_driver = { .driver = { .name = "panel-simple", - .owner = THIS_MODULE, .of_match_table = platform_of_match, }, .probe = panel_simple_platform_probe, diff --git a/drivers/gpu/drm/qxl/qxl_release.c b/drivers/gpu/drm/qxl/qxl_release.c index 446e71ca36cb..d9b25684ac98 100644 --- a/drivers/gpu/drm/qxl/qxl_release.c +++ b/drivers/gpu/drm/qxl/qxl_release.c @@ -264,7 +264,8 @@ int qxl_release_reserve_list(struct qxl_release *release, bool no_intr) if (list_is_singular(&release->bos)) return 0; - ret = ttm_eu_reserve_buffers(&release->ticket, &release->bos, !no_intr); + ret = ttm_eu_reserve_buffers(&release->ticket, &release->bos, + !no_intr, NULL); if (ret) return ret; diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index 30d242b25078..ed644a4f6f57 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -1851,10 +1851,9 @@ static int radeon_atom_pick_pll(struct drm_crtc *crtc) return pll; } /* otherwise, pick one of the plls */ - if ((rdev->family == CHIP_KAVERI) || - (rdev->family == CHIP_KABINI) || + if ((rdev->family == CHIP_KABINI) || (rdev->family == CHIP_MULLINS)) { - /* KB/KV/ML has PPLL1 and PPLL2 */ + /* KB/ML has PPLL1 and PPLL2 */ pll_in_use = radeon_get_pll_use_mask(crtc); if (!(pll_in_use & (1 << ATOM_PPLL2))) return ATOM_PPLL2; @@ -1863,7 +1862,7 @@ static int radeon_atom_pick_pll(struct drm_crtc *crtc) DRM_ERROR("unable to allocate a PPLL\n"); return ATOM_PPLL_INVALID; } else { - /* CI has PPLL0, PPLL1, and PPLL2 */ + /* CI/KV has PPLL0, PPLL1, and PPLL2 */ pll_in_use = radeon_get_pll_use_mask(crtc); if (!(pll_in_use & (1 << ATOM_PPLL2))) return ATOM_PPLL2; @@ -2039,6 +2038,7 @@ int atombios_crtc_mode_set(struct drm_crtc *crtc, atombios_crtc_set_base(crtc, x, y, old_fb); atombios_overscan_setup(crtc, mode, adjusted_mode); atombios_scaler_setup(crtc); + radeon_cursor_reset(crtc); /* update the hw version fpr dpm */ radeon_crtc->hw_mode = *adjusted_mode; @@ -2154,6 +2154,7 @@ static void atombios_crtc_disable(struct drm_crtc *crtc) case ATOM_PPLL0: /* disable the ppll */ if ((rdev->family == CHIP_ARUBA) || + (rdev->family == CHIP_KAVERI) || (rdev->family == CHIP_BONAIRE) || (rdev->family == CHIP_HAWAII)) atombios_crtc_program_pll(crtc, radeon_crtc->crtc_id, radeon_crtc->pll_id, diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c index 11ba9d21b89b..db42a670f995 100644 --- a/drivers/gpu/drm/radeon/atombios_dp.c +++ b/drivers/gpu/drm/radeon/atombios_dp.c @@ -492,6 +492,10 @@ int radeon_dp_mode_valid_helper(struct drm_connector *connector, struct radeon_connector_atom_dig *dig_connector; int dp_clock; + if ((mode->clock > 340000) && + (!radeon_connector_is_dp12_capable(connector))) + return MODE_CLOCK_HIGH; + if (!radeon_connector->con_priv) return MODE_CLOCK_HIGH; dig_connector = radeon_connector->con_priv; diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c index 3f898d020ae6..f373a81ba3d5 100644 --- a/drivers/gpu/drm/radeon/ci_dpm.c +++ b/drivers/gpu/drm/radeon/ci_dpm.c @@ -937,7 +937,7 @@ static void ci_fan_ctrl_set_static_mode(struct radeon_device *rdev, u32 mode) tmp |= TMIN(0); WREG32_SMC(CG_FDO_CTRL2, tmp); - tmp = RREG32_SMC(CG_FDO_CTRL2) & FDO_PWM_MODE_MASK; + tmp = RREG32_SMC(CG_FDO_CTRL2) & ~FDO_PWM_MODE_MASK; tmp |= FDO_PWM_MODE(mode); WREG32_SMC(CG_FDO_CTRL2, tmp); } @@ -1162,7 +1162,7 @@ static int ci_fan_ctrl_set_fan_speed_rpm(struct radeon_device *rdev, tmp |= TARGET_PERIOD(tach_period); WREG32_SMC(CG_TACH_CTRL, tmp); - ci_fan_ctrl_set_static_mode(rdev, FDO_PWM_MODE_STATIC); + ci_fan_ctrl_set_static_mode(rdev, FDO_PWM_MODE_STATIC_RPM); return 0; } @@ -1178,7 +1178,7 @@ static void ci_fan_ctrl_set_default_mode(struct radeon_device *rdev) tmp |= FDO_PWM_MODE(pi->fan_ctrl_default_mode); WREG32_SMC(CG_FDO_CTRL2, tmp); - tmp = RREG32_SMC(CG_FDO_CTRL2) & TMIN_MASK; + tmp = RREG32_SMC(CG_FDO_CTRL2) & ~TMIN_MASK; tmp |= TMIN(pi->t_min); WREG32_SMC(CG_FDO_CTRL2, tmp); pi->fan_ctrl_is_in_default_mode = true; @@ -5849,7 +5849,6 @@ int ci_dpm_init(struct radeon_device *rdev) rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac; pi->fan_ctrl_is_in_default_mode = true; - rdev->pm.dpm.fan.ucode_fan_control = false; return 0; } diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h index e4e88ca8b82e..03003f8a6de6 100644 --- a/drivers/gpu/drm/radeon/cikd.h +++ b/drivers/gpu/drm/radeon/cikd.h @@ -213,18 +213,18 @@ #define CG_FDO_CTRL0 0xC0300064 #define FDO_STATIC_DUTY(x) ((x) << 0) -#define FDO_STATIC_DUTY_MASK 0x0000000F +#define FDO_STATIC_DUTY_MASK 0x000000FF #define FDO_STATIC_DUTY_SHIFT 0 #define CG_FDO_CTRL1 0xC0300068 #define FMAX_DUTY100(x) ((x) << 0) -#define FMAX_DUTY100_MASK 0x0000000F +#define FMAX_DUTY100_MASK 0x000000FF #define FMAX_DUTY100_SHIFT 0 #define CG_FDO_CTRL2 0xC030006C #define TMIN(x) ((x) << 0) -#define TMIN_MASK 0x0000000F +#define TMIN_MASK 0x000000FF #define TMIN_SHIFT 0 #define FDO_PWM_MODE(x) ((x) << 11) -#define FDO_PWM_MODE_MASK (3 << 11) +#define FDO_PWM_MODE_MASK (7 << 11) #define FDO_PWM_MODE_SHIFT 11 #define TACH_PWM_RESP_RATE(x) ((x) << 25) #define TACH_PWM_RESP_RATE_MASK (0x7f << 25) @@ -2156,4 +2156,6 @@ #define ATC_VM_APERTURE1_HIGH_ADDR 0x330Cu #define ATC_VM_APERTURE1_LOW_ADDR 0x3304u +#define IH_VMID_0_LUT 0x3D40u + #endif diff --git a/drivers/gpu/drm/radeon/dce3_1_afmt.c b/drivers/gpu/drm/radeon/dce3_1_afmt.c index 2fe8cfc966d9..bafdf92a5732 100644 --- a/drivers/gpu/drm/radeon/dce3_1_afmt.c +++ b/drivers/gpu/drm/radeon/dce3_1_afmt.c @@ -103,7 +103,7 @@ static void dce3_2_afmt_write_sad_regs(struct drm_encoder *encoder) } sad_count = drm_edid_to_sad(radeon_connector->edid, &sads); - if (sad_count < 0) { + if (sad_count <= 0) { DRM_ERROR("Couldn't read SADs: %d\n", sad_count); return; } diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c index 5c8b358f9fba..924b1b7ab455 100644 --- a/drivers/gpu/drm/radeon/evergreen_cs.c +++ b/drivers/gpu/drm/radeon/evergreen_cs.c @@ -35,7 +35,7 @@ #define MIN(a,b) (((a)<(b))?(a):(b)) int r600_dma_cs_next_reloc(struct radeon_cs_parser *p, - struct radeon_cs_reloc **cs_reloc); + struct radeon_bo_list **cs_reloc); struct evergreen_cs_track { u32 group_size; u32 nbanks; @@ -1094,7 +1094,7 @@ static int evergreen_cs_parse_packet0(struct radeon_cs_parser *p, static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) { struct evergreen_cs_track *track = (struct evergreen_cs_track *)p->track; - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; u32 last_reg; u32 m, i, tmp, *ib; int r; @@ -1792,7 +1792,7 @@ static bool evergreen_is_safe_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) static int evergreen_packet3_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt) { - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct evergreen_cs_track *track; volatile u32 *ib; unsigned idx; @@ -2661,7 +2661,7 @@ int evergreen_cs_parse(struct radeon_cs_parser *p) p->track = NULL; return r; } - } while (p->idx < p->chunks[p->chunk_ib_idx].length_dw); + } while (p->idx < p->chunk_ib->length_dw); #if 0 for (r = 0; r < p->ib.length_dw; r++) { printk(KERN_INFO "%05d 0x%08X\n", r, p->ib.ptr[r]); @@ -2684,8 +2684,8 @@ int evergreen_cs_parse(struct radeon_cs_parser *p) **/ int evergreen_dma_cs_parse(struct radeon_cs_parser *p) { - struct radeon_cs_chunk *ib_chunk = &p->chunks[p->chunk_ib_idx]; - struct radeon_cs_reloc *src_reloc, *dst_reloc, *dst2_reloc; + struct radeon_cs_chunk *ib_chunk = p->chunk_ib; + struct radeon_bo_list *src_reloc, *dst_reloc, *dst2_reloc; u32 header, cmd, count, sub_cmd; volatile u32 *ib = p->ib.ptr; u32 idx; @@ -3100,7 +3100,7 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p) DRM_ERROR("Unknown packet type %d at %d !\n", cmd, idx); return -EINVAL; } - } while (p->idx < p->chunks[p->chunk_ib_idx].length_dw); + } while (p->idx < p->chunk_ib->length_dw); #if 0 for (r = 0; r < p->ib->length_dw; r++) { printk(KERN_INFO "%05d 0x%08X\n", r, p->ib.ptr[r]); diff --git a/drivers/gpu/drm/radeon/kv_dpm.c b/drivers/gpu/drm/radeon/kv_dpm.c index 9b42001295ba..e3e9c10cfba9 100644 --- a/drivers/gpu/drm/radeon/kv_dpm.c +++ b/drivers/gpu/drm/radeon/kv_dpm.c @@ -2745,13 +2745,11 @@ int kv_dpm_init(struct radeon_device *rdev) pi->enable_auto_thermal_throttling = true; pi->disable_nb_ps3_in_battery = false; if (radeon_bapm == -1) { - /* There are stability issues reported on with - * bapm enabled on an asrock system. - */ - if (rdev->pdev->subsystem_vendor == 0x1849) - pi->bapm_enable = false; - else + /* only enable bapm on KB, ML by default */ + if (rdev->family == CHIP_KABINI || rdev->family == CHIP_MULLINS) pi->bapm_enable = true; + else + pi->bapm_enable = false; } else if (radeon_bapm == 0) { pi->bapm_enable = false; } else { diff --git a/drivers/gpu/drm/radeon/mkregtable.c b/drivers/gpu/drm/radeon/mkregtable.c index 4a85bb644e24..b928c17bdeed 100644 --- a/drivers/gpu/drm/radeon/mkregtable.c +++ b/drivers/gpu/drm/radeon/mkregtable.c @@ -347,7 +347,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_entry - get the struct for this entry * @ptr: the &struct list_head pointer. * @type: the type of the struct this is embedded in. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. */ #define list_entry(ptr, type, member) \ container_of(ptr, type, member) @@ -356,7 +356,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_first_entry - get the first element from a list * @ptr: the list head to take the element from. * @type: the type of the struct this is embedded in. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Note, that list is expected to be not empty. */ @@ -406,7 +406,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_for_each_entry - iterate over list of given type * @pos: the type * to use as a loop cursor. * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. */ #define list_for_each_entry(pos, head, member) \ for (pos = list_entry((head)->next, typeof(*pos), member); \ @@ -417,7 +417,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_for_each_entry_reverse - iterate backwards over list of given type. * @pos: the type * to use as a loop cursor. * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. */ #define list_for_each_entry_reverse(pos, head, member) \ for (pos = list_entry((head)->prev, typeof(*pos), member); \ @@ -428,7 +428,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_prepare_entry - prepare a pos entry for use in list_for_each_entry_continue() * @pos: the type * to use as a start point * @head: the head of the list - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Prepares a pos entry for use as a start point in list_for_each_entry_continue(). */ @@ -439,7 +439,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_for_each_entry_continue - continue iteration over list of given type * @pos: the type * to use as a loop cursor. * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Continue to iterate over list of given type, continuing after * the current position. @@ -453,7 +453,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_for_each_entry_continue_reverse - iterate backwards from the given point * @pos: the type * to use as a loop cursor. * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Start to iterate over list of given type backwards, continuing after * the current position. @@ -467,7 +467,7 @@ static inline void list_splice_tail_init(struct list_head *list, * list_for_each_entry_from - iterate over list of given type from the current point * @pos: the type * to use as a loop cursor. * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Iterate over list of given type, continuing from current position. */ @@ -480,7 +480,7 @@ static inline void list_splice_tail_init(struct list_head *list, * @pos: the type * to use as a loop cursor. * @n: another type * to use as temporary storage * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. */ #define list_for_each_entry_safe(pos, n, head, member) \ for (pos = list_entry((head)->next, typeof(*pos), member), \ @@ -493,7 +493,7 @@ static inline void list_splice_tail_init(struct list_head *list, * @pos: the type * to use as a loop cursor. * @n: another type * to use as temporary storage * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Iterate over list of given type, continuing after current point, * safe against removal of list entry. @@ -509,7 +509,7 @@ static inline void list_splice_tail_init(struct list_head *list, * @pos: the type * to use as a loop cursor. * @n: another type * to use as temporary storage * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Iterate over list of given type from current point, safe against * removal of list entry. @@ -524,7 +524,7 @@ static inline void list_splice_tail_init(struct list_head *list, * @pos: the type * to use as a loop cursor. * @n: another type * to use as temporary storage * @head: the head for your list. - * @member: the name of the list_struct within the struct. + * @member: the name of the list_head within the struct. * * Iterate backwards over list of given type, safe against removal * of list entry. diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index b53b31a7b76f..74f06d540591 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -1254,7 +1254,7 @@ int r100_reloc_pitch_offset(struct radeon_cs_parser *p, int r; u32 tile_flags = 0; u32 tmp; - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; u32 value; r = radeon_cs_packet_next_reloc(p, &reloc, 0); @@ -1293,7 +1293,7 @@ int r100_packet3_load_vbpntr(struct radeon_cs_parser *p, int idx) { unsigned c, i; - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct r100_cs_track *track; int r = 0; volatile uint32_t *ib; @@ -1542,7 +1542,7 @@ static int r100_packet0_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt, unsigned idx, unsigned reg) { - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct r100_cs_track *track; volatile uint32_t *ib; uint32_t tmp; @@ -1901,7 +1901,7 @@ int r100_cs_track_check_pkt3_indx_buffer(struct radeon_cs_parser *p, static int r100_packet3_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt) { - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct r100_cs_track *track; unsigned idx; volatile uint32_t *ib; @@ -2061,7 +2061,7 @@ int r100_cs_parse(struct radeon_cs_parser *p) } if (r) return r; - } while (p->idx < p->chunks[p->chunk_ib_idx].length_dw); + } while (p->idx < p->chunk_ib->length_dw); return 0; } diff --git a/drivers/gpu/drm/radeon/r200.c b/drivers/gpu/drm/radeon/r200.c index 732d4938aab7..c70e6d5bcd19 100644 --- a/drivers/gpu/drm/radeon/r200.c +++ b/drivers/gpu/drm/radeon/r200.c @@ -146,7 +146,7 @@ int r200_packet0_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt, unsigned idx, unsigned reg) { - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct r100_cs_track *track; volatile uint32_t *ib; uint32_t tmp; diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c index 1bc4704034ce..064ad5569cca 100644 --- a/drivers/gpu/drm/radeon/r300.c +++ b/drivers/gpu/drm/radeon/r300.c @@ -598,7 +598,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt, unsigned idx, unsigned reg) { - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct r100_cs_track *track; volatile uint32_t *ib; uint32_t tmp, tile_flags = 0; @@ -1142,7 +1142,7 @@ fail: static int r300_packet3_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt) { - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct r100_cs_track *track; volatile uint32_t *ib; unsigned idx; @@ -1283,7 +1283,7 @@ int r300_cs_parse(struct radeon_cs_parser *p) if (r) { return r; } - } while (p->idx < p->chunks[p->chunk_ib_idx].length_dw); + } while (p->idx < p->chunk_ib->length_dw); return 0; } diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c index c47537a1ddba..acc1f99c84d9 100644 --- a/drivers/gpu/drm/radeon/r600_cs.c +++ b/drivers/gpu/drm/radeon/r600_cs.c @@ -969,7 +969,7 @@ static int r600_cs_parse_packet0(struct radeon_cs_parser *p, static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) { struct r600_cs_track *track = (struct r600_cs_track *)p->track; - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; u32 m, i, tmp, *ib; int r; @@ -1626,7 +1626,7 @@ static bool r600_is_safe_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) static int r600_packet3_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt) { - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; struct r600_cs_track *track; volatile u32 *ib; unsigned idx; @@ -2316,7 +2316,7 @@ int r600_cs_parse(struct radeon_cs_parser *p) p->track = NULL; return r; } - } while (p->idx < p->chunks[p->chunk_ib_idx].length_dw); + } while (p->idx < p->chunk_ib->length_dw); #if 0 for (r = 0; r < p->ib.length_dw; r++) { printk(KERN_INFO "%05d 0x%08X\n", r, p->ib.ptr[r]); @@ -2351,10 +2351,10 @@ static void r600_cs_parser_fini(struct radeon_cs_parser *parser, int error) static int r600_cs_parser_relocs_legacy(struct radeon_cs_parser *p) { - if (p->chunk_relocs_idx == -1) { + if (p->chunk_relocs == NULL) { return 0; } - p->relocs = kzalloc(sizeof(struct radeon_cs_reloc), GFP_KERNEL); + p->relocs = kzalloc(sizeof(struct radeon_bo_list), GFP_KERNEL); if (p->relocs == NULL) { return -ENOMEM; } @@ -2398,7 +2398,7 @@ int r600_cs_legacy(struct drm_device *dev, void *data, struct drm_file *filp, /* Copy the packet into the IB, the parser will read from the * input memory (cached) and write to the IB (which can be * uncached). */ - ib_chunk = &parser.chunks[parser.chunk_ib_idx]; + ib_chunk = parser.chunk_ib; parser.ib.length_dw = ib_chunk->length_dw; *l = parser.ib.length_dw; if (copy_from_user(ib, ib_chunk->user_ptr, ib_chunk->length_dw * 4)) { @@ -2435,24 +2435,24 @@ void r600_cs_legacy_init(void) * GPU offset using the provided start. **/ int r600_dma_cs_next_reloc(struct radeon_cs_parser *p, - struct radeon_cs_reloc **cs_reloc) + struct radeon_bo_list **cs_reloc) { struct radeon_cs_chunk *relocs_chunk; unsigned idx; *cs_reloc = NULL; - if (p->chunk_relocs_idx == -1) { + if (p->chunk_relocs == NULL) { DRM_ERROR("No relocation chunk !\n"); return -EINVAL; } - relocs_chunk = &p->chunks[p->chunk_relocs_idx]; + relocs_chunk = p->chunk_relocs; idx = p->dma_reloc_idx; if (idx >= p->nrelocs) { DRM_ERROR("Relocs at %d after relocations chunk end %d !\n", idx, p->nrelocs); return -EINVAL; } - *cs_reloc = p->relocs_ptr[idx]; + *cs_reloc = &p->relocs[idx]; p->dma_reloc_idx++; return 0; } @@ -2472,8 +2472,8 @@ int r600_dma_cs_next_reloc(struct radeon_cs_parser *p, **/ int r600_dma_cs_parse(struct radeon_cs_parser *p) { - struct radeon_cs_chunk *ib_chunk = &p->chunks[p->chunk_ib_idx]; - struct radeon_cs_reloc *src_reloc, *dst_reloc; + struct radeon_cs_chunk *ib_chunk = p->chunk_ib; + struct radeon_bo_list *src_reloc, *dst_reloc; u32 header, cmd, count, tiled; volatile u32 *ib = p->ib.ptr; u32 idx, idx_value; @@ -2619,7 +2619,7 @@ int r600_dma_cs_parse(struct radeon_cs_parser *p) DRM_ERROR("Unknown packet type %d at %d !\n", cmd, idx); return -EINVAL; } - } while (p->idx < p->chunks[p->chunk_ib_idx].length_dw); + } while (p->idx < p->chunk_ib->length_dw); #if 0 for (r = 0; r < p->ib->length_dw; r++) { printk(KERN_INFO "%05d 0x%08X\n", r, p->ib.ptr[r]); diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 3207bb60715e..54529b837afa 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -450,6 +450,15 @@ struct radeon_mman { #endif }; +struct radeon_bo_list { + struct radeon_bo *robj; + struct ttm_validate_buffer tv; + uint64_t gpu_offset; + unsigned prefered_domains; + unsigned allowed_domains; + uint32_t tiling_flags; +}; + /* bo virtual address in a specific vm */ struct radeon_bo_va { /* protected by bo being reserved */ @@ -920,6 +929,9 @@ struct radeon_vm { struct rb_root va; + /* protecting invalidated and freed */ + spinlock_t status_lock; + /* BOs moved, but not yet updated in the PT */ struct list_head invalidated; @@ -1044,19 +1056,7 @@ void cayman_dma_fini(struct radeon_device *rdev); /* * CS. */ -struct radeon_cs_reloc { - struct drm_gem_object *gobj; - struct radeon_bo *robj; - struct ttm_validate_buffer tv; - uint64_t gpu_offset; - unsigned prefered_domains; - unsigned allowed_domains; - uint32_t tiling_flags; - uint32_t handle; -}; - struct radeon_cs_chunk { - uint32_t chunk_id; uint32_t length_dw; uint32_t *kdata; void __user *user_ptr; @@ -1074,16 +1074,15 @@ struct radeon_cs_parser { unsigned idx; /* relocations */ unsigned nrelocs; - struct radeon_cs_reloc *relocs; - struct radeon_cs_reloc **relocs_ptr; - struct radeon_cs_reloc *vm_bos; + struct radeon_bo_list *relocs; + struct radeon_bo_list *vm_bos; struct list_head validated; unsigned dma_reloc_idx; /* indices of various chunks */ - int chunk_ib_idx; - int chunk_relocs_idx; - int chunk_flags_idx; - int chunk_const_ib_idx; + struct radeon_cs_chunk *chunk_ib; + struct radeon_cs_chunk *chunk_relocs; + struct radeon_cs_chunk *chunk_flags; + struct radeon_cs_chunk *chunk_const_ib; struct radeon_ib ib; struct radeon_ib const_ib; void *track; @@ -1097,7 +1096,7 @@ struct radeon_cs_parser { static inline u32 radeon_get_ib_value(struct radeon_cs_parser *p, int idx) { - struct radeon_cs_chunk *ibc = &p->chunks[p->chunk_ib_idx]; + struct radeon_cs_chunk *ibc = p->chunk_ib; if (ibc->kdata) return ibc->kdata[idx]; @@ -2975,7 +2974,7 @@ int radeon_vm_manager_init(struct radeon_device *rdev); void radeon_vm_manager_fini(struct radeon_device *rdev); int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm); void radeon_vm_fini(struct radeon_device *rdev, struct radeon_vm *vm); -struct radeon_cs_reloc *radeon_vm_get_bos(struct radeon_device *rdev, +struct radeon_bo_list *radeon_vm_get_bos(struct radeon_device *rdev, struct radeon_vm *vm, struct list_head *head); struct radeon_fence *radeon_vm_grab_id(struct radeon_device *rdev, @@ -3089,7 +3088,7 @@ bool radeon_cs_packet_next_is_pkt3_nop(struct radeon_cs_parser *p); void radeon_cs_dump_packet(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt); int radeon_cs_packet_next_reloc(struct radeon_cs_parser *p, - struct radeon_cs_reloc **cs_reloc, + struct radeon_bo_list **cs_reloc, int nomm); int r600_cs_common_vline_parse(struct radeon_cs_parser *p, uint32_t *vline_start_end, diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c index 75f22e5e999f..c830863bc98a 100644 --- a/drivers/gpu/drm/radeon/radeon_cs.c +++ b/drivers/gpu/drm/radeon/radeon_cs.c @@ -77,22 +77,18 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p) struct drm_device *ddev = p->rdev->ddev; struct radeon_cs_chunk *chunk; struct radeon_cs_buckets buckets; - unsigned i, j; - bool duplicate, need_mmap_lock = false; + unsigned i; + bool need_mmap_lock = false; int r; - if (p->chunk_relocs_idx == -1) { + if (p->chunk_relocs == NULL) { return 0; } - chunk = &p->chunks[p->chunk_relocs_idx]; + chunk = p->chunk_relocs; p->dma_reloc_idx = 0; /* FIXME: we assume that each relocs use 4 dwords */ p->nrelocs = chunk->length_dw / 4; - p->relocs_ptr = kcalloc(p->nrelocs, sizeof(void *), GFP_KERNEL); - if (p->relocs_ptr == NULL) { - return -ENOMEM; - } - p->relocs = kcalloc(p->nrelocs, sizeof(struct radeon_cs_reloc), GFP_KERNEL); + p->relocs = kcalloc(p->nrelocs, sizeof(struct radeon_bo_list), GFP_KERNEL); if (p->relocs == NULL) { return -ENOMEM; } @@ -101,31 +97,17 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p) for (i = 0; i < p->nrelocs; i++) { struct drm_radeon_cs_reloc *r; + struct drm_gem_object *gobj; unsigned priority; - duplicate = false; r = (struct drm_radeon_cs_reloc *)&chunk->kdata[i*4]; - for (j = 0; j < i; j++) { - if (r->handle == p->relocs[j].handle) { - p->relocs_ptr[i] = &p->relocs[j]; - duplicate = true; - break; - } - } - if (duplicate) { - p->relocs[i].handle = 0; - continue; - } - - p->relocs[i].gobj = drm_gem_object_lookup(ddev, p->filp, - r->handle); - if (p->relocs[i].gobj == NULL) { + gobj = drm_gem_object_lookup(ddev, p->filp, r->handle); + if (gobj == NULL) { DRM_ERROR("gem object lookup failed 0x%x\n", r->handle); return -ENOENT; } - p->relocs_ptr[i] = &p->relocs[i]; - p->relocs[i].robj = gem_to_radeon_bo(p->relocs[i].gobj); + p->relocs[i].robj = gem_to_radeon_bo(gobj); /* The userspace buffer priorities are from 0 to 15. A higher * number means the buffer is more important. @@ -184,7 +166,6 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p) p->relocs[i].tv.bo = &p->relocs[i].robj->tbo; p->relocs[i].tv.shared = !r->write_domain; - p->relocs[i].handle = r->handle; radeon_cs_buckets_add(&buckets, &p->relocs[i].tv.head, priority); @@ -251,22 +232,19 @@ static int radeon_cs_get_ring(struct radeon_cs_parser *p, u32 ring, s32 priority static int radeon_cs_sync_rings(struct radeon_cs_parser *p) { - int i, r = 0; + struct radeon_bo_list *reloc; + int r; - for (i = 0; i < p->nrelocs; i++) { + list_for_each_entry(reloc, &p->validated, tv.head) { struct reservation_object *resv; - if (!p->relocs[i].robj) - continue; - - resv = p->relocs[i].robj->tbo.resv; + resv = reloc->robj->tbo.resv; r = radeon_sync_resv(p->rdev, &p->ib.sync, resv, - p->relocs[i].tv.shared); - + reloc->tv.shared); if (r) - break; + return r; } - return r; + return 0; } /* XXX: note that this is called from the legacy UMS CS ioctl as well */ @@ -286,10 +264,10 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) p->idx = 0; p->ib.sa_bo = NULL; p->const_ib.sa_bo = NULL; - p->chunk_ib_idx = -1; - p->chunk_relocs_idx = -1; - p->chunk_flags_idx = -1; - p->chunk_const_ib_idx = -1; + p->chunk_ib = NULL; + p->chunk_relocs = NULL; + p->chunk_flags = NULL; + p->chunk_const_ib = NULL; p->chunks_array = kcalloc(cs->num_chunks, sizeof(uint64_t), GFP_KERNEL); if (p->chunks_array == NULL) { return -ENOMEM; @@ -316,24 +294,23 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) return -EFAULT; } p->chunks[i].length_dw = user_chunk.length_dw; - p->chunks[i].chunk_id = user_chunk.chunk_id; - if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_RELOCS) { - p->chunk_relocs_idx = i; + if (user_chunk.chunk_id == RADEON_CHUNK_ID_RELOCS) { + p->chunk_relocs = &p->chunks[i]; } - if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_IB) { - p->chunk_ib_idx = i; + if (user_chunk.chunk_id == RADEON_CHUNK_ID_IB) { + p->chunk_ib = &p->chunks[i]; /* zero length IB isn't useful */ if (p->chunks[i].length_dw == 0) return -EINVAL; } - if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_CONST_IB) { - p->chunk_const_ib_idx = i; + if (user_chunk.chunk_id == RADEON_CHUNK_ID_CONST_IB) { + p->chunk_const_ib = &p->chunks[i]; /* zero length CONST IB isn't useful */ if (p->chunks[i].length_dw == 0) return -EINVAL; } - if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_FLAGS) { - p->chunk_flags_idx = i; + if (user_chunk.chunk_id == RADEON_CHUNK_ID_FLAGS) { + p->chunk_flags = &p->chunks[i]; /* zero length flags aren't useful */ if (p->chunks[i].length_dw == 0) return -EINVAL; @@ -342,10 +319,10 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) size = p->chunks[i].length_dw; cdata = (void __user *)(unsigned long)user_chunk.chunk_data; p->chunks[i].user_ptr = cdata; - if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_CONST_IB) + if (user_chunk.chunk_id == RADEON_CHUNK_ID_CONST_IB) continue; - if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_IB) { + if (user_chunk.chunk_id == RADEON_CHUNK_ID_IB) { if (!p->rdev || !(p->rdev->flags & RADEON_IS_AGP)) continue; } @@ -358,7 +335,7 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) if (copy_from_user(p->chunks[i].kdata, cdata, size)) { return -EFAULT; } - if (p->chunks[i].chunk_id == RADEON_CHUNK_ID_FLAGS) { + if (user_chunk.chunk_id == RADEON_CHUNK_ID_FLAGS) { p->cs_flags = p->chunks[i].kdata[0]; if (p->chunks[i].length_dw > 1) ring = p->chunks[i].kdata[1]; @@ -399,8 +376,8 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) static int cmp_size_smaller_first(void *priv, struct list_head *a, struct list_head *b) { - struct radeon_cs_reloc *la = list_entry(a, struct radeon_cs_reloc, tv.head); - struct radeon_cs_reloc *lb = list_entry(b, struct radeon_cs_reloc, tv.head); + struct radeon_bo_list *la = list_entry(a, struct radeon_bo_list, tv.head); + struct radeon_bo_list *lb = list_entry(b, struct radeon_bo_list, tv.head); /* Sort A before B if A is smaller. */ return (int)la->robj->tbo.num_pages - (int)lb->robj->tbo.num_pages; @@ -441,13 +418,15 @@ static void radeon_cs_parser_fini(struct radeon_cs_parser *parser, int error, bo if (parser->relocs != NULL) { for (i = 0; i < parser->nrelocs; i++) { - if (parser->relocs[i].gobj) - drm_gem_object_unreference_unlocked(parser->relocs[i].gobj); + struct radeon_bo *bo = parser->relocs[i].robj; + if (bo == NULL) + continue; + + drm_gem_object_unreference_unlocked(&bo->gem_base); } } kfree(parser->track); kfree(parser->relocs); - kfree(parser->relocs_ptr); drm_free_large(parser->vm_bos); for (i = 0; i < parser->nchunks; i++) drm_free_large(parser->chunks[i].kdata); @@ -462,7 +441,7 @@ static int radeon_cs_ib_chunk(struct radeon_device *rdev, { int r; - if (parser->chunk_ib_idx == -1) + if (parser->chunk_ib == NULL) return 0; if (parser->cs_flags & RADEON_CS_USE_VM) @@ -505,9 +484,6 @@ static int radeon_bo_vm_update_pte(struct radeon_cs_parser *p, if (r) return r; - radeon_sync_resv(p->rdev, &p->ib.sync, vm->page_directory->tbo.resv, - true); - r = radeon_vm_clear_freed(rdev, vm); if (r) return r; @@ -525,10 +501,6 @@ static int radeon_bo_vm_update_pte(struct radeon_cs_parser *p, for (i = 0; i < p->nrelocs; i++) { struct radeon_bo *bo; - /* ignore duplicates */ - if (p->relocs_ptr[i] != &p->relocs[i]) - continue; - bo = p->relocs[i].robj; bo_va = radeon_vm_bo_find(vm, bo); if (bo_va == NULL) { @@ -553,7 +525,7 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev, struct radeon_vm *vm = &fpriv->vm; int r; - if (parser->chunk_ib_idx == -1) + if (parser->chunk_ib == NULL) return 0; if ((parser->cs_flags & RADEON_CS_USE_VM) == 0) return 0; @@ -587,7 +559,7 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev, } if ((rdev->family >= CHIP_TAHITI) && - (parser->chunk_const_ib_idx != -1)) { + (parser->chunk_const_ib != NULL)) { r = radeon_ib_schedule(rdev, &parser->ib, &parser->const_ib, true); } else { r = radeon_ib_schedule(rdev, &parser->ib, NULL, true); @@ -614,7 +586,7 @@ static int radeon_cs_ib_fill(struct radeon_device *rdev, struct radeon_cs_parser struct radeon_vm *vm = NULL; int r; - if (parser->chunk_ib_idx == -1) + if (parser->chunk_ib == NULL) return 0; if (parser->cs_flags & RADEON_CS_USE_VM) { @@ -622,8 +594,8 @@ static int radeon_cs_ib_fill(struct radeon_device *rdev, struct radeon_cs_parser vm = &fpriv->vm; if ((rdev->family >= CHIP_TAHITI) && - (parser->chunk_const_ib_idx != -1)) { - ib_chunk = &parser->chunks[parser->chunk_const_ib_idx]; + (parser->chunk_const_ib != NULL)) { + ib_chunk = parser->chunk_const_ib; if (ib_chunk->length_dw > RADEON_IB_VM_MAX_SIZE) { DRM_ERROR("cs IB CONST too big: %d\n", ib_chunk->length_dw); return -EINVAL; @@ -642,13 +614,13 @@ static int radeon_cs_ib_fill(struct radeon_device *rdev, struct radeon_cs_parser return -EFAULT; } - ib_chunk = &parser->chunks[parser->chunk_ib_idx]; + ib_chunk = parser->chunk_ib; if (ib_chunk->length_dw > RADEON_IB_VM_MAX_SIZE) { DRM_ERROR("cs IB too big: %d\n", ib_chunk->length_dw); return -EINVAL; } } - ib_chunk = &parser->chunks[parser->chunk_ib_idx]; + ib_chunk = parser->chunk_ib; r = radeon_ib_get(rdev, parser->ring, &parser->ib, vm, ib_chunk->length_dw * 4); @@ -740,7 +712,7 @@ int radeon_cs_packet_parse(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt, unsigned idx) { - struct radeon_cs_chunk *ib_chunk = &p->chunks[p->chunk_ib_idx]; + struct radeon_cs_chunk *ib_chunk = p->chunk_ib; struct radeon_device *rdev = p->rdev; uint32_t header; @@ -834,7 +806,7 @@ void radeon_cs_dump_packet(struct radeon_cs_parser *p, * GPU offset using the provided start. **/ int radeon_cs_packet_next_reloc(struct radeon_cs_parser *p, - struct radeon_cs_reloc **cs_reloc, + struct radeon_bo_list **cs_reloc, int nomm) { struct radeon_cs_chunk *relocs_chunk; @@ -842,12 +814,12 @@ int radeon_cs_packet_next_reloc(struct radeon_cs_parser *p, unsigned idx; int r; - if (p->chunk_relocs_idx == -1) { + if (p->chunk_relocs == NULL) { DRM_ERROR("No relocation chunk !\n"); return -EINVAL; } *cs_reloc = NULL; - relocs_chunk = &p->chunks[p->chunk_relocs_idx]; + relocs_chunk = p->chunk_relocs; r = radeon_cs_packet_parse(p, &p3reloc, p->idx); if (r) return r; @@ -873,6 +845,6 @@ int radeon_cs_packet_next_reloc(struct radeon_cs_parser *p, (u64)relocs_chunk->kdata[idx + 3] << 32; (*cs_reloc)->gpu_offset |= relocs_chunk->kdata[idx + 0]; } else - *cs_reloc = p->relocs_ptr[(idx / 4)]; + *cs_reloc = &p->relocs[(idx / 4)]; return 0; } diff --git a/drivers/gpu/drm/radeon/radeon_cursor.c b/drivers/gpu/drm/radeon/radeon_cursor.c index 85f38ee11888..45e54060ee97 100644 --- a/drivers/gpu/drm/radeon/radeon_cursor.c +++ b/drivers/gpu/drm/radeon/radeon_cursor.c @@ -227,11 +227,24 @@ int radeon_crtc_cursor_move(struct drm_crtc *crtc, return ret; } -static void radeon_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj, - uint64_t gpu_addr, int hot_x, int hot_y) +static int radeon_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj) { struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); struct radeon_device *rdev = crtc->dev->dev_private; + struct radeon_bo *robj = gem_to_radeon_bo(obj); + uint64_t gpu_addr; + int ret; + + ret = radeon_bo_reserve(robj, false); + if (unlikely(ret != 0)) + goto fail; + /* Only 27 bit offset for legacy cursor */ + ret = radeon_bo_pin_restricted(robj, RADEON_GEM_DOMAIN_VRAM, + ASIC_IS_AVIVO(rdev) ? 0 : 1 << 27, + &gpu_addr); + radeon_bo_unreserve(robj); + if (ret) + goto fail; if (ASIC_IS_DCE4(rdev)) { WREG32(EVERGREEN_CUR_SURFACE_ADDRESS_HIGH + radeon_crtc->crtc_offset, @@ -253,18 +266,12 @@ static void radeon_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj, WREG32(RADEON_CUR_OFFSET + radeon_crtc->crtc_offset, radeon_crtc->legacy_cursor_offset); } - if (hot_x != radeon_crtc->cursor_hot_x || - hot_y != radeon_crtc->cursor_hot_y) { - int x, y; - - x = radeon_crtc->cursor_x + radeon_crtc->cursor_hot_x - hot_x; - y = radeon_crtc->cursor_y + radeon_crtc->cursor_hot_y - hot_y; + return 0; - radeon_cursor_move_locked(crtc, x, y); +fail: + drm_gem_object_unreference_unlocked(obj); - radeon_crtc->cursor_hot_x = hot_x; - radeon_crtc->cursor_hot_y = hot_y; - } + return ret; } int radeon_crtc_cursor_set2(struct drm_crtc *crtc, @@ -276,10 +283,7 @@ int radeon_crtc_cursor_set2(struct drm_crtc *crtc, int32_t hot_y) { struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); - struct radeon_device *rdev = crtc->dev->dev_private; struct drm_gem_object *obj; - struct radeon_bo *robj; - uint64_t gpu_addr; int ret; if (!handle) { @@ -301,41 +305,76 @@ int radeon_crtc_cursor_set2(struct drm_crtc *crtc, return -ENOENT; } - robj = gem_to_radeon_bo(obj); - ret = radeon_bo_reserve(robj, false); - if (unlikely(ret != 0)) - goto fail; - /* Only 27 bit offset for legacy cursor */ - ret = radeon_bo_pin_restricted(robj, RADEON_GEM_DOMAIN_VRAM, - ASIC_IS_AVIVO(rdev) ? 0 : 1 << 27, - &gpu_addr); - radeon_bo_unreserve(robj); - if (ret) - goto fail; - radeon_crtc->cursor_width = width; radeon_crtc->cursor_height = height; radeon_lock_cursor(crtc, true); - radeon_set_cursor(crtc, obj, gpu_addr, hot_x, hot_y); - radeon_show_cursor(crtc); + + if (hot_x != radeon_crtc->cursor_hot_x || + hot_y != radeon_crtc->cursor_hot_y) { + int x, y; + + x = radeon_crtc->cursor_x + radeon_crtc->cursor_hot_x - hot_x; + y = radeon_crtc->cursor_y + radeon_crtc->cursor_hot_y - hot_y; + + radeon_cursor_move_locked(crtc, x, y); + + radeon_crtc->cursor_hot_x = hot_x; + radeon_crtc->cursor_hot_y = hot_y; + } + + ret = radeon_set_cursor(crtc, obj); + + if (ret) + DRM_ERROR("radeon_set_cursor returned %d, not changing cursor\n", + ret); + else + radeon_show_cursor(crtc); + radeon_lock_cursor(crtc, false); unpin: if (radeon_crtc->cursor_bo) { - robj = gem_to_radeon_bo(radeon_crtc->cursor_bo); + struct radeon_bo *robj = gem_to_radeon_bo(radeon_crtc->cursor_bo); ret = radeon_bo_reserve(robj, false); if (likely(ret == 0)) { radeon_bo_unpin(robj); radeon_bo_unreserve(robj); } - drm_gem_object_unreference_unlocked(radeon_crtc->cursor_bo); + if (radeon_crtc->cursor_bo != obj) + drm_gem_object_unreference_unlocked(radeon_crtc->cursor_bo); } radeon_crtc->cursor_bo = obj; return 0; -fail: - drm_gem_object_unreference_unlocked(obj); +} - return ret; +/** + * radeon_cursor_reset - Re-set the current cursor, if any. + * + * @crtc: drm crtc + * + * If the CRTC passed in currently has a cursor assigned, this function + * makes sure it's visible. + */ +void radeon_cursor_reset(struct drm_crtc *crtc) +{ + struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); + int ret; + + if (radeon_crtc->cursor_bo) { + radeon_lock_cursor(crtc, true); + + radeon_cursor_move_locked(crtc, radeon_crtc->cursor_x, + radeon_crtc->cursor_y); + + ret = radeon_set_cursor(crtc, radeon_crtc->cursor_bo); + if (ret) + DRM_ERROR("radeon_set_cursor returned %d, not showing " + "cursor\n", ret); + else + radeon_show_cursor(crtc); + + radeon_lock_cursor(crtc, false); + } } diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c index 0ea1db83d573..29b9220ec399 100644 --- a/drivers/gpu/drm/radeon/radeon_fb.c +++ b/drivers/gpu/drm/radeon/radeon_fb.c @@ -48,10 +48,40 @@ struct radeon_fbdev { struct radeon_device *rdev; }; +/** + * radeon_fb_helper_set_par - Hide cursor on CRTCs used by fbdev. + * + * @info: fbdev info + * + * This function hides the cursor on all CRTCs used by fbdev. + */ +static int radeon_fb_helper_set_par(struct fb_info *info) +{ + int ret; + + ret = drm_fb_helper_set_par(info); + + /* XXX: with universal plane support fbdev will automatically disable + * all non-primary planes (including the cursor) + */ + if (ret == 0) { + struct drm_fb_helper *fb_helper = info->par; + int i; + + for (i = 0; i < fb_helper->crtc_count; i++) { + struct drm_crtc *crtc = fb_helper->crtc_info[i].mode_set.crtc; + + radeon_crtc_cursor_set2(crtc, NULL, 0, 0, 0, 0, 0); + } + } + + return ret; +} + static struct fb_ops radeonfb_ops = { .owner = THIS_MODULE, .fb_check_var = drm_fb_helper_check_var, - .fb_set_par = drm_fb_helper_set_par, + .fb_set_par = radeon_fb_helper_set_par, .fb_fillrect = cfb_fillrect, .fb_copyarea = cfb_copyarea, .fb_imageblit = cfb_imageblit, diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c index 12cfaeac1205..a46f73737994 100644 --- a/drivers/gpu/drm/radeon/radeon_gem.c +++ b/drivers/gpu/drm/radeon/radeon_gem.c @@ -394,10 +394,9 @@ int radeon_gem_set_domain_ioctl(struct drm_device *dev, void *data, return r; } -static int radeon_mode_mmap(struct drm_file *filp, - struct drm_device *dev, - uint32_t handle, bool dumb, - uint64_t *offset_p) +int radeon_mode_dumb_mmap(struct drm_file *filp, + struct drm_device *dev, + uint32_t handle, uint64_t *offset_p) { struct drm_gem_object *gobj; struct radeon_bo *robj; @@ -406,14 +405,6 @@ static int radeon_mode_mmap(struct drm_file *filp, if (gobj == NULL) { return -ENOENT; } - - /* - * We don't allow dumb mmaps on objects created using another - * interface. - */ - WARN_ONCE(dumb && !(gobj->dumb || gobj->import_attach), - "Illegal dumb map of GPU buffer.\n"); - robj = gem_to_radeon_bo(gobj); if (radeon_ttm_tt_has_userptr(robj->tbo.ttm)) { drm_gem_object_unreference_unlocked(gobj); @@ -424,20 +415,12 @@ static int radeon_mode_mmap(struct drm_file *filp, return 0; } -int radeon_mode_dumb_mmap(struct drm_file *filp, - struct drm_device *dev, - uint32_t handle, uint64_t *offset_p) -{ - return radeon_mode_mmap(filp, dev, handle, true, offset_p); -} - int radeon_gem_mmap_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) { struct drm_radeon_gem_mmap *args = data; - return radeon_mode_mmap(filp, dev, args->handle, false, - &args->addr_ptr); + return radeon_mode_dumb_mmap(filp, dev, args->handle, &args->addr_ptr); } int radeon_gem_busy_ioctl(struct drm_device *dev, void *data, @@ -548,7 +531,7 @@ static void radeon_gem_va_update_vm(struct radeon_device *rdev, struct radeon_bo_va *bo_va) { struct ttm_validate_buffer tv, *entry; - struct radeon_cs_reloc *vm_bos; + struct radeon_bo_list *vm_bos; struct ww_acquire_ctx ticket; struct list_head list; unsigned domain; @@ -564,7 +547,7 @@ static void radeon_gem_va_update_vm(struct radeon_device *rdev, if (!vm_bos) return; - r = ttm_eu_reserve_buffers(&ticket, &list, true); + r = ttm_eu_reserve_buffers(&ticket, &list, true, NULL); if (r) goto error_free; @@ -763,7 +746,6 @@ int radeon_mode_dumb_create(struct drm_file *file_priv, return -ENOMEM; r = drm_gem_handle_create(file_priv, gobj, &handle); - gobj->dumb = true; /* drop reference from allocate - handle holds it now */ drm_gem_object_unreference_unlocked(gobj); if (r) { diff --git a/drivers/gpu/drm/radeon/radeon_kfd.c b/drivers/gpu/drm/radeon/radeon_kfd.c index 065d02068ec3..8bf87f1203cc 100644 --- a/drivers/gpu/drm/radeon/radeon_kfd.c +++ b/drivers/gpu/drm/radeon/radeon_kfd.c @@ -28,6 +28,8 @@ #include "cikd.h" #include "cik_reg.h" #include "radeon_kfd.h" +#include "radeon_ucode.h" +#include <linux/firmware.h> #define CIK_PIPE_PER_MEC (4) @@ -49,6 +51,7 @@ static uint64_t get_vmem_size(struct kgd_dev *kgd); static uint64_t get_gpu_clock_counter(struct kgd_dev *kgd); static uint32_t get_max_engine_clock_in_mhz(struct kgd_dev *kgd); +static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type); /* * Register access functions @@ -69,7 +72,7 @@ static int kgd_init_pipeline(struct kgd_dev *kgd, uint32_t pipe_id, static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id, uint32_t queue_id, uint32_t __user *wptr); -static bool kgd_hqd_is_occupies(struct kgd_dev *kgd, uint64_t queue_address, +static bool kgd_hqd_is_occupied(struct kgd_dev *kgd, uint64_t queue_address, uint32_t pipe_id, uint32_t queue_id); static int kgd_hqd_destroy(struct kgd_dev *kgd, uint32_t reset_type, @@ -89,14 +92,16 @@ static const struct kfd2kgd_calls kfd2kgd = { .init_memory = kgd_init_memory, .init_pipeline = kgd_init_pipeline, .hqd_load = kgd_hqd_load, - .hqd_is_occupies = kgd_hqd_is_occupies, + .hqd_is_occupied = kgd_hqd_is_occupied, .hqd_destroy = kgd_hqd_destroy, + .get_fw_version = get_fw_version }; static const struct kgd2kfd_calls *kgd2kfd; bool radeon_kfd_init(void) { +#if defined(CONFIG_HSA_AMD_MODULE) bool (*kgd2kfd_init_p)(unsigned, const struct kfd2kgd_calls*, const struct kgd2kfd_calls**); @@ -113,6 +118,17 @@ bool radeon_kfd_init(void) } return true; +#elif defined(CONFIG_HSA_AMD) + if (!kgd2kfd_init(KFD_INTERFACE_VERSION, &kfd2kgd, &kgd2kfd)) { + kgd2kfd = NULL; + + return false; + } + + return true; +#else + return false; +#endif } void radeon_kfd_fini(void) @@ -374,6 +390,10 @@ static int kgd_set_pasid_vmid_mapping(struct kgd_dev *kgd, unsigned int pasid, cpu_relax(); write_register(kgd, ATC_VMID_PASID_MAPPING_UPDATE_STATUS, 1U << vmid); + /* Mapping vmid to pasid also for IH block */ + write_register(kgd, IH_VMID_0_LUT + vmid * sizeof(uint32_t), + pasid_mapping); + return 0; } @@ -513,7 +533,7 @@ static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id, return 0; } -static bool kgd_hqd_is_occupies(struct kgd_dev *kgd, uint64_t queue_address, +static bool kgd_hqd_is_occupied(struct kgd_dev *kgd, uint64_t queue_address, uint32_t pipe_id, uint32_t queue_id) { uint32_t act; @@ -552,6 +572,7 @@ static int kgd_hqd_destroy(struct kgd_dev *kgd, uint32_t reset_type, if (timeout == 0) { pr_err("kfd: cp queue preemption time out (%dms)\n", temp); + release_queue(kgd); return -ETIME; } msleep(20); @@ -561,3 +582,52 @@ static int kgd_hqd_destroy(struct kgd_dev *kgd, uint32_t reset_type, release_queue(kgd); return 0; } + +static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type) +{ + struct radeon_device *rdev = (struct radeon_device *) kgd; + const union radeon_firmware_header *hdr; + + BUG_ON(kgd == NULL || rdev->mec_fw == NULL); + + switch (type) { + case KGD_ENGINE_PFP: + hdr = (const union radeon_firmware_header *) rdev->pfp_fw->data; + break; + + case KGD_ENGINE_ME: + hdr = (const union radeon_firmware_header *) rdev->me_fw->data; + break; + + case KGD_ENGINE_CE: + hdr = (const union radeon_firmware_header *) rdev->ce_fw->data; + break; + + case KGD_ENGINE_MEC1: + hdr = (const union radeon_firmware_header *) rdev->mec_fw->data; + break; + + case KGD_ENGINE_MEC2: + hdr = (const union radeon_firmware_header *) + rdev->mec2_fw->data; + break; + + case KGD_ENGINE_RLC: + hdr = (const union radeon_firmware_header *) rdev->rlc_fw->data; + break; + + case KGD_ENGINE_SDMA: + hdr = (const union radeon_firmware_header *) + rdev->sdma_fw->data; + break; + + default: + return 0; + } + + if (hdr == NULL) + return 0; + + /* Only 12 bit in use*/ + return hdr->common.ucode_version; +} diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c index f4dd26ae33e5..3cf9c1fa6475 100644 --- a/drivers/gpu/drm/radeon/radeon_kms.c +++ b/drivers/gpu/drm/radeon/radeon_kms.c @@ -800,6 +800,8 @@ int radeon_get_vblank_timestamp_kms(struct drm_device *dev, int crtc, /* Get associated drm_crtc: */ drmcrtc = &rdev->mode_info.crtcs[crtc]->base; + if (!drmcrtc) + return -EINVAL; /* Helper routine in DRM core does all the work: */ return drm_calc_vbltimestamp_from_scanoutpos(dev, crtc, max_error, diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c index cafb1ccf2ec3..678b4386540d 100644 --- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c +++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c @@ -1054,6 +1054,7 @@ static int radeon_crtc_mode_set(struct drm_crtc *crtc, DRM_ERROR("Mode need scaling but only first crtc can do that.\n"); } } + radeon_cursor_reset(crtc); return 0; } diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h index f3d87cdd5c9d..390db897f322 100644 --- a/drivers/gpu/drm/radeon/radeon_mode.h +++ b/drivers/gpu/drm/radeon/radeon_mode.h @@ -818,6 +818,7 @@ extern int radeon_crtc_cursor_set2(struct drm_crtc *crtc, int32_t hot_y); extern int radeon_crtc_cursor_move(struct drm_crtc *crtc, int x, int y); +extern void radeon_cursor_reset(struct drm_crtc *crtc); extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int flags, diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c index 87b00d902bf7..86fc56434b28 100644 --- a/drivers/gpu/drm/radeon/radeon_object.c +++ b/drivers/gpu/drm/radeon/radeon_object.c @@ -233,6 +233,13 @@ int radeon_bo_create(struct radeon_device *rdev, if (!(rdev->flags & RADEON_IS_PCIE)) bo->flags &= ~(RADEON_GEM_GTT_WC | RADEON_GEM_GTT_UC); +#ifdef CONFIG_X86_32 + /* XXX: Write-combined CPU mappings of GTT seem broken on 32-bit + * See https://bugs.freedesktop.org/show_bug.cgi?id=84627 + */ + bo->flags &= ~RADEON_GEM_GTT_WC; +#endif + radeon_ttm_placement_from_domain(bo, domain); /* Kernel allocation are uninterruptible */ down_read(&rdev->pm.mclk_lock); @@ -502,28 +509,26 @@ int radeon_bo_list_validate(struct radeon_device *rdev, struct ww_acquire_ctx *ticket, struct list_head *head, int ring) { - struct radeon_cs_reloc *lobj; - struct radeon_bo *bo; + struct radeon_bo_list *lobj; + struct list_head duplicates; int r; u64 bytes_moved = 0, initial_bytes_moved; u64 bytes_moved_threshold = radeon_bo_get_threshold_for_moves(rdev); - r = ttm_eu_reserve_buffers(ticket, head, true); + INIT_LIST_HEAD(&duplicates); + r = ttm_eu_reserve_buffers(ticket, head, true, &duplicates); if (unlikely(r != 0)) { return r; } list_for_each_entry(lobj, head, tv.head) { - bo = lobj->robj; + struct radeon_bo *bo = lobj->robj; if (!bo->pin_count) { u32 domain = lobj->prefered_domains; u32 allowed = lobj->allowed_domains; u32 current_domain = radeon_mem_type_to_domain(bo->tbo.mem.mem_type); - WARN_ONCE(bo->gem_base.dumb, - "GPU use of dumb buffer is illegal.\n"); - /* Check if this buffer will be moved and don't move it * if we have moved too many buffers for this IB already. * @@ -562,6 +567,12 @@ int radeon_bo_list_validate(struct radeon_device *rdev, lobj->gpu_offset = radeon_bo_gpu_offset(bo); lobj->tiling_flags = bo->tiling_flags; } + + list_for_each_entry(lobj, &duplicates, tv.head) { + lobj->gpu_offset = radeon_bo_gpu_offset(lobj->robj); + lobj->tiling_flags = lobj->robj->tiling_flags; + } + return 0; } diff --git a/drivers/gpu/drm/radeon/radeon_state.c b/drivers/gpu/drm/radeon/radeon_state.c index 535403e0c8a2..15aee723db77 100644 --- a/drivers/gpu/drm/radeon/radeon_state.c +++ b/drivers/gpu/drm/radeon/radeon_state.c @@ -1703,7 +1703,7 @@ static int radeon_cp_dispatch_texture(struct drm_device * dev, u32 format; u32 *buffer; const u8 __user *data; - int size, dwords, tex_width, blit_width, spitch; + unsigned int size, dwords, tex_width, blit_width, spitch; u32 height; int i; u32 texpitch, microtile; diff --git a/drivers/gpu/drm/radeon/radeon_trace.h b/drivers/gpu/drm/radeon/radeon_trace.h index 9db74a96ef61..ce075cb08cb2 100644 --- a/drivers/gpu/drm/radeon/radeon_trace.h +++ b/drivers/gpu/drm/radeon/radeon_trace.h @@ -38,7 +38,7 @@ TRACE_EVENT(radeon_cs, TP_fast_assign( __entry->ring = p->ring; - __entry->dw = p->chunks[p->chunk_ib_idx].length_dw; + __entry->dw = p->chunk_ib->length_dw; __entry->fences = radeon_fence_count_emitted( p->rdev, p->ring); ), diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index cbe7b32d181c..d02aa1d0f588 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -196,7 +196,7 @@ static void radeon_evict_flags(struct ttm_buffer_object *bo, rbo = container_of(bo, struct radeon_bo, tbo); switch (bo->mem.mem_type) { case TTM_PL_VRAM: - if (rbo->rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready == false) + if (rbo->rdev->ring[radeon_copy_ring_index(rbo->rdev)].ready == false) radeon_ttm_placement_from_domain(rbo, RADEON_GEM_DOMAIN_CPU); else if (rbo->rdev->mc.visible_vram_size < rbo->rdev->mc.real_vram_size && bo->mem.start < (rbo->rdev->mc.visible_vram_size >> PAGE_SHIFT)) { diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c index 11b662469253..c10b2aec6450 100644 --- a/drivers/gpu/drm/radeon/radeon_uvd.c +++ b/drivers/gpu/drm/radeon/radeon_uvd.c @@ -488,12 +488,12 @@ static int radeon_uvd_cs_reloc(struct radeon_cs_parser *p, unsigned buf_sizes[], bool *has_msg_cmd) { struct radeon_cs_chunk *relocs_chunk; - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; unsigned idx, cmd, offset; uint64_t start, end; int r; - relocs_chunk = &p->chunks[p->chunk_relocs_idx]; + relocs_chunk = p->chunk_relocs; offset = radeon_get_ib_value(p, data0); idx = radeon_get_ib_value(p, data1); if (idx >= relocs_chunk->length_dw) { @@ -502,7 +502,7 @@ static int radeon_uvd_cs_reloc(struct radeon_cs_parser *p, return -EINVAL; } - reloc = p->relocs_ptr[(idx / 4)]; + reloc = &p->relocs[(idx / 4)]; start = reloc->gpu_offset; end = start + radeon_bo_size(reloc->robj); start += offset; @@ -610,13 +610,13 @@ int radeon_uvd_cs_parse(struct radeon_cs_parser *p) [0x00000003] = 2048, }; - if (p->chunks[p->chunk_ib_idx].length_dw % 16) { + if (p->chunk_ib->length_dw % 16) { DRM_ERROR("UVD IB length (%d) not 16 dwords aligned!\n", - p->chunks[p->chunk_ib_idx].length_dw); + p->chunk_ib->length_dw); return -EINVAL; } - if (p->chunk_relocs_idx == -1) { + if (p->chunk_relocs == NULL) { DRM_ERROR("No relocation chunk !\n"); return -EINVAL; } @@ -640,7 +640,7 @@ int radeon_uvd_cs_parse(struct radeon_cs_parser *p) DRM_ERROR("Unknown packet type %d !\n", pkt.type); return -EINVAL; } - } while (p->idx < p->chunks[p->chunk_ib_idx].length_dw); + } while (p->idx < p->chunk_ib->length_dw); if (!has_msg_cmd) { DRM_ERROR("UVD-IBs need a msg command!\n"); diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c index 9e85757d5599..976fe432f4e2 100644 --- a/drivers/gpu/drm/radeon/radeon_vce.c +++ b/drivers/gpu/drm/radeon/radeon_vce.c @@ -453,11 +453,11 @@ int radeon_vce_cs_reloc(struct radeon_cs_parser *p, int lo, int hi, unsigned size) { struct radeon_cs_chunk *relocs_chunk; - struct radeon_cs_reloc *reloc; + struct radeon_bo_list *reloc; uint64_t start, end, offset; unsigned idx; - relocs_chunk = &p->chunks[p->chunk_relocs_idx]; + relocs_chunk = p->chunk_relocs; offset = radeon_get_ib_value(p, lo); idx = radeon_get_ib_value(p, hi); @@ -467,7 +467,7 @@ int radeon_vce_cs_reloc(struct radeon_cs_parser *p, int lo, int hi, return -EINVAL; } - reloc = p->relocs_ptr[(idx / 4)]; + reloc = &p->relocs[(idx / 4)]; start = reloc->gpu_offset; end = start + radeon_bo_size(reloc->robj); start += offset; @@ -534,7 +534,7 @@ int radeon_vce_cs_parse(struct radeon_cs_parser *p) uint32_t *size = &tmp; int i, r; - while (p->idx < p->chunks[p->chunk_ib_idx].length_dw) { + while (p->idx < p->chunk_ib->length_dw) { uint32_t len = radeon_get_ib_value(p, p->idx); uint32_t cmd = radeon_get_ib_value(p, p->idx + 1); diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c index 0b10f3a03ce2..cde48c42b30a 100644 --- a/drivers/gpu/drm/radeon/radeon_vm.c +++ b/drivers/gpu/drm/radeon/radeon_vm.c @@ -125,41 +125,37 @@ void radeon_vm_manager_fini(struct radeon_device *rdev) * Add the page directory to the list of BOs to * validate for command submission (cayman+). */ -struct radeon_cs_reloc *radeon_vm_get_bos(struct radeon_device *rdev, +struct radeon_bo_list *radeon_vm_get_bos(struct radeon_device *rdev, struct radeon_vm *vm, struct list_head *head) { - struct radeon_cs_reloc *list; + struct radeon_bo_list *list; unsigned i, idx; list = drm_malloc_ab(vm->max_pde_used + 2, - sizeof(struct radeon_cs_reloc)); + sizeof(struct radeon_bo_list)); if (!list) return NULL; /* add the vm page table to the list */ - list[0].gobj = NULL; list[0].robj = vm->page_directory; list[0].prefered_domains = RADEON_GEM_DOMAIN_VRAM; list[0].allowed_domains = RADEON_GEM_DOMAIN_VRAM; list[0].tv.bo = &vm->page_directory->tbo; list[0].tv.shared = true; list[0].tiling_flags = 0; - list[0].handle = 0; list_add(&list[0].tv.head, head); for (i = 0, idx = 1; i <= vm->max_pde_used; i++) { if (!vm->page_tables[i].bo) continue; - list[idx].gobj = NULL; list[idx].robj = vm->page_tables[i].bo; list[idx].prefered_domains = RADEON_GEM_DOMAIN_VRAM; list[idx].allowed_domains = RADEON_GEM_DOMAIN_VRAM; list[idx].tv.bo = &list[idx].robj->tbo; list[idx].tv.shared = true; list[idx].tiling_flags = 0; - list[idx].handle = 0; list_add(&list[idx++].tv.head, head); } @@ -491,7 +487,9 @@ int radeon_vm_bo_set_addr(struct radeon_device *rdev, tmp->vm = vm; tmp->addr = bo_va->addr; tmp->bo = radeon_bo_ref(bo_va->bo); + spin_lock(&vm->status_lock); list_add(&tmp->vm_status, &vm->freed); + spin_unlock(&vm->status_lock); } interval_tree_remove(&bo_va->it, &vm->va); @@ -802,11 +800,11 @@ static void radeon_vm_frag_ptes(struct radeon_device *rdev, * * Global and local mutex must be locked! */ -static void radeon_vm_update_ptes(struct radeon_device *rdev, - struct radeon_vm *vm, - struct radeon_ib *ib, - uint64_t start, uint64_t end, - uint64_t dst, uint32_t flags) +static int radeon_vm_update_ptes(struct radeon_device *rdev, + struct radeon_vm *vm, + struct radeon_ib *ib, + uint64_t start, uint64_t end, + uint64_t dst, uint32_t flags) { uint64_t mask = RADEON_VM_PTE_COUNT - 1; uint64_t last_pte = ~0, last_dst = ~0; @@ -819,8 +817,12 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev, struct radeon_bo *pt = vm->page_tables[pt_idx].bo; unsigned nptes; uint64_t pte; + int r; radeon_sync_resv(rdev, &ib->sync, pt->tbo.resv, true); + r = reservation_object_reserve_shared(pt->tbo.resv); + if (r) + return r; if ((addr & ~mask) == (end & ~mask)) nptes = end - addr; @@ -854,6 +856,8 @@ static void radeon_vm_update_ptes(struct radeon_device *rdev, last_pte + 8 * count, last_dst, flags); } + + return 0; } /** @@ -878,7 +882,7 @@ static void radeon_vm_fence_pts(struct radeon_vm *vm, end >>= radeon_vm_block_size; for (i = start; i <= end; ++i) - radeon_bo_fence(vm->page_tables[i].bo, fence, false); + radeon_bo_fence(vm->page_tables[i].bo, fence, true); } /** @@ -911,7 +915,9 @@ int radeon_vm_bo_update(struct radeon_device *rdev, return -EINVAL; } + spin_lock(&vm->status_lock); list_del_init(&bo_va->vm_status); + spin_unlock(&vm->status_lock); bo_va->flags &= ~RADEON_VM_PAGE_VALID; bo_va->flags &= ~RADEON_VM_PAGE_SYSTEM; @@ -987,9 +993,13 @@ int radeon_vm_bo_update(struct radeon_device *rdev, radeon_sync_fence(&ib.sync, vm->ids[i].last_id_use); } - radeon_vm_update_ptes(rdev, vm, &ib, bo_va->it.start, - bo_va->it.last + 1, addr, - radeon_vm_page_flags(bo_va->flags)); + r = radeon_vm_update_ptes(rdev, vm, &ib, bo_va->it.start, + bo_va->it.last + 1, addr, + radeon_vm_page_flags(bo_va->flags)); + if (r) { + radeon_ib_free(rdev, &ib); + return r; + } radeon_asic_vm_pad_ib(rdev, &ib); WARN_ON(ib.length_dw > ndw); @@ -1022,17 +1032,25 @@ int radeon_vm_bo_update(struct radeon_device *rdev, int radeon_vm_clear_freed(struct radeon_device *rdev, struct radeon_vm *vm) { - struct radeon_bo_va *bo_va, *tmp; + struct radeon_bo_va *bo_va; int r; - list_for_each_entry_safe(bo_va, tmp, &vm->freed, vm_status) { + spin_lock(&vm->status_lock); + while (!list_empty(&vm->freed)) { + bo_va = list_first_entry(&vm->freed, + struct radeon_bo_va, vm_status); + spin_unlock(&vm->status_lock); + r = radeon_vm_bo_update(rdev, bo_va, NULL); radeon_bo_unref(&bo_va->bo); radeon_fence_unref(&bo_va->last_pt_update); kfree(bo_va); if (r) return r; + + spin_lock(&vm->status_lock); } + spin_unlock(&vm->status_lock); return 0; } @@ -1051,14 +1069,23 @@ int radeon_vm_clear_freed(struct radeon_device *rdev, int radeon_vm_clear_invalids(struct radeon_device *rdev, struct radeon_vm *vm) { - struct radeon_bo_va *bo_va, *tmp; + struct radeon_bo_va *bo_va; int r; - list_for_each_entry_safe(bo_va, tmp, &vm->invalidated, vm_status) { + spin_lock(&vm->status_lock); + while (!list_empty(&vm->invalidated)) { + bo_va = list_first_entry(&vm->invalidated, + struct radeon_bo_va, vm_status); + spin_unlock(&vm->status_lock); + r = radeon_vm_bo_update(rdev, bo_va, NULL); if (r) return r; + + spin_lock(&vm->status_lock); } + spin_unlock(&vm->status_lock); + return 0; } @@ -1081,6 +1108,7 @@ void radeon_vm_bo_rmv(struct radeon_device *rdev, mutex_lock(&vm->mutex); interval_tree_remove(&bo_va->it, &vm->va); + spin_lock(&vm->status_lock); list_del(&bo_va->vm_status); if (bo_va->addr) { @@ -1090,6 +1118,7 @@ void radeon_vm_bo_rmv(struct radeon_device *rdev, radeon_fence_unref(&bo_va->last_pt_update); kfree(bo_va); } + spin_unlock(&vm->status_lock); mutex_unlock(&vm->mutex); } @@ -1110,10 +1139,10 @@ void radeon_vm_bo_invalidate(struct radeon_device *rdev, list_for_each_entry(bo_va, &bo->va, bo_list) { if (bo_va->addr) { - mutex_lock(&bo_va->vm->mutex); + spin_lock(&bo_va->vm->status_lock); list_del(&bo_va->vm_status); list_add(&bo_va->vm_status, &bo_va->vm->invalidated); - mutex_unlock(&bo_va->vm->mutex); + spin_unlock(&bo_va->vm->status_lock); } } } @@ -1141,6 +1170,7 @@ int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm) } mutex_init(&vm->mutex); vm->va = RB_ROOT; + spin_lock_init(&vm->status_lock); INIT_LIST_HEAD(&vm->invalidated); INIT_LIST_HEAD(&vm->freed); diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index cf4c420b5572..32e354b8b0ab 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -5893,7 +5893,7 @@ static void si_fan_ctrl_set_static_mode(struct radeon_device *rdev, u32 mode) tmp |= TMIN(0); WREG32(CG_FDO_CTRL2, tmp); - tmp = RREG32(CG_FDO_CTRL2) & FDO_PWM_MODE_MASK; + tmp = RREG32(CG_FDO_CTRL2) & ~FDO_PWM_MODE_MASK; tmp |= FDO_PWM_MODE(mode); WREG32(CG_FDO_CTRL2, tmp); } @@ -6098,7 +6098,7 @@ static int si_fan_ctrl_set_fan_speed_rpm(struct radeon_device *rdev, tmp |= TARGET_PERIOD(tach_period); WREG32(CG_TACH_CTRL, tmp); - si_fan_ctrl_set_static_mode(rdev, FDO_PWM_MODE_STATIC); + si_fan_ctrl_set_static_mode(rdev, FDO_PWM_MODE_STATIC_RPM); return 0; } @@ -6114,7 +6114,7 @@ static void si_fan_ctrl_set_default_mode(struct radeon_device *rdev) tmp |= FDO_PWM_MODE(si_pi->fan_ctrl_default_mode); WREG32(CG_FDO_CTRL2, tmp); - tmp = RREG32(CG_FDO_CTRL2) & TMIN_MASK; + tmp = RREG32(CG_FDO_CTRL2) & ~TMIN_MASK; tmp |= TMIN(si_pi->t_min); WREG32(CG_FDO_CTRL2, tmp); si_pi->fan_ctrl_is_in_default_mode = true; diff --git a/drivers/gpu/drm/radeon/sid.h b/drivers/gpu/drm/radeon/sid.h index c549c16a4fe4..4069be89e585 100644 --- a/drivers/gpu/drm/radeon/sid.h +++ b/drivers/gpu/drm/radeon/sid.h @@ -208,18 +208,18 @@ #define CG_FDO_CTRL0 0x754 #define FDO_STATIC_DUTY(x) ((x) << 0) -#define FDO_STATIC_DUTY_MASK 0x0000000F +#define FDO_STATIC_DUTY_MASK 0x000000FF #define FDO_STATIC_DUTY_SHIFT 0 #define CG_FDO_CTRL1 0x758 #define FMAX_DUTY100(x) ((x) << 0) -#define FMAX_DUTY100_MASK 0x0000000F +#define FMAX_DUTY100_MASK 0x000000FF #define FMAX_DUTY100_SHIFT 0 #define CG_FDO_CTRL2 0x75C #define TMIN(x) ((x) << 0) -#define TMIN_MASK 0x0000000F +#define TMIN_MASK 0x000000FF #define TMIN_SHIFT 0 #define FDO_PWM_MODE(x) ((x) << 11) -#define FDO_PWM_MODE_MASK (3 << 11) +#define FDO_PWM_MODE_MASK (7 << 11) #define FDO_PWM_MODE_SHIFT 11 #define TACH_PWM_RESP_RATE(x) ((x) << 25) #define TACH_PWM_RESP_RATE_MASK (0x7f << 25) diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.c b/drivers/gpu/drm/rcar-du/rcar_du_drv.c index 967ae8f20233..7bfa09cf18d5 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_drv.c +++ b/drivers/gpu/drm/rcar-du/rcar_du_drv.c @@ -328,7 +328,6 @@ static struct platform_driver rcar_du_platform_driver = { .probe = rcar_du_probe, .remove = rcar_du_remove, .driver = { - .owner = THIS_MODULE, .name = "rcar-du", .pm = &rcar_du_pm_ops, .of_match_table = rcar_du_of_table, diff --git a/drivers/gpu/drm/rockchip/Kconfig b/drivers/gpu/drm/rockchip/Kconfig new file mode 100644 index 000000000000..ca9f085efa92 --- /dev/null +++ b/drivers/gpu/drm/rockchip/Kconfig @@ -0,0 +1,17 @@ +config DRM_ROCKCHIP + tristate "DRM Support for Rockchip" + depends on DRM && ROCKCHIP_IOMMU + select DRM_KMS_HELPER + select DRM_KMS_FB_HELPER + select DRM_PANEL + select FB_CFB_FILLRECT + select FB_CFB_COPYAREA + select FB_CFB_IMAGEBLIT + select VT_HW_CONSOLE_BINDING if FRAMEBUFFER_CONSOLE + select VIDEOMODE_HELPERS + help + Choose this option if you have a Rockchip soc chipset. + This driver provides kernel mode setting and buffer + management to userspace. This driver does not provide + 2D or 3D acceleration; acceleration is performed by other + IP found on the SoC. diff --git a/drivers/gpu/drm/rockchip/Makefile b/drivers/gpu/drm/rockchip/Makefile new file mode 100644 index 000000000000..2cb0672f57ed --- /dev/null +++ b/drivers/gpu/drm/rockchip/Makefile @@ -0,0 +1,8 @@ +# +# Makefile for the drm device driver. This driver provides support for the +# Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher. + +rockchipdrm-y := rockchip_drm_drv.o rockchip_drm_fb.o rockchip_drm_fbdev.o \ + rockchip_drm_gem.o + +obj-$(CONFIG_DRM_ROCKCHIP) += rockchipdrm.o rockchip_drm_vop.o diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c new file mode 100644 index 000000000000..a798c7c71f91 --- /dev/null +++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c @@ -0,0 +1,551 @@ +/* + * Copyright (C) Fuzhou Rockchip Electronics Co.Ltd + * Author:Mark Yao <mark.yao@rock-chips.com> + * + * based on exynos_drm_drv.c + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <asm/dma-iommu.h> + +#include <drm/drmP.h> +#include <drm/drm_crtc_helper.h> +#include <drm/drm_fb_helper.h> +#include <linux/dma-mapping.h> +#include <linux/pm_runtime.h> +#include <linux/of_graph.h> +#include <linux/component.h> + +#include "rockchip_drm_drv.h" +#include "rockchip_drm_fb.h" +#include "rockchip_drm_fbdev.h" +#include "rockchip_drm_gem.h" + +#define DRIVER_NAME "rockchip" +#define DRIVER_DESC "RockChip Soc DRM" +#define DRIVER_DATE "20140818" +#define DRIVER_MAJOR 1 +#define DRIVER_MINOR 0 + +/* + * Attach a (component) device to the shared drm dma mapping from master drm + * device. This is used by the VOPs to map GEM buffers to a common DMA + * mapping. + */ +int rockchip_drm_dma_attach_device(struct drm_device *drm_dev, + struct device *dev) +{ + struct dma_iommu_mapping *mapping = drm_dev->dev->archdata.mapping; + int ret; + + ret = dma_set_coherent_mask(dev, DMA_BIT_MASK(32)); + if (ret) + return ret; + + dma_set_max_seg_size(dev, DMA_BIT_MASK(32)); + + return arm_iommu_attach_device(dev, mapping); +} +EXPORT_SYMBOL_GPL(rockchip_drm_dma_attach_device); + +void rockchip_drm_dma_detach_device(struct drm_device *drm_dev, + struct device *dev) +{ + arm_iommu_detach_device(dev); +} +EXPORT_SYMBOL_GPL(rockchip_drm_dma_detach_device); + +int rockchip_register_crtc_funcs(struct drm_device *dev, + const struct rockchip_crtc_funcs *crtc_funcs, + int pipe) +{ + struct rockchip_drm_private *priv = dev->dev_private; + + if (pipe > ROCKCHIP_MAX_CRTC) + return -EINVAL; + + priv->crtc_funcs[pipe] = crtc_funcs; + + return 0; +} +EXPORT_SYMBOL_GPL(rockchip_register_crtc_funcs); + +void rockchip_unregister_crtc_funcs(struct drm_device *dev, int pipe) +{ + struct rockchip_drm_private *priv = dev->dev_private; + + if (pipe > ROCKCHIP_MAX_CRTC) + return; + + priv->crtc_funcs[pipe] = NULL; +} +EXPORT_SYMBOL_GPL(rockchip_unregister_crtc_funcs); + +static struct drm_crtc *rockchip_crtc_from_pipe(struct drm_device *drm, + int pipe) +{ + struct drm_crtc *crtc; + int i = 0; + + list_for_each_entry(crtc, &drm->mode_config.crtc_list, head) + if (i++ == pipe) + return crtc; + + return NULL; +} + +static int rockchip_drm_crtc_enable_vblank(struct drm_device *dev, int pipe) +{ + struct rockchip_drm_private *priv = dev->dev_private; + struct drm_crtc *crtc = rockchip_crtc_from_pipe(dev, pipe); + + if (crtc && priv->crtc_funcs[pipe] && + priv->crtc_funcs[pipe]->enable_vblank) + return priv->crtc_funcs[pipe]->enable_vblank(crtc); + + return 0; +} + +static void rockchip_drm_crtc_disable_vblank(struct drm_device *dev, int pipe) +{ + struct rockchip_drm_private *priv = dev->dev_private; + struct drm_crtc *crtc = rockchip_crtc_from_pipe(dev, pipe); + + if (crtc && priv->crtc_funcs[pipe] && + priv->crtc_funcs[pipe]->enable_vblank) + priv->crtc_funcs[pipe]->disable_vblank(crtc); +} + +static int rockchip_drm_load(struct drm_device *drm_dev, unsigned long flags) +{ + struct rockchip_drm_private *private; + struct dma_iommu_mapping *mapping; + struct device *dev = drm_dev->dev; + int ret; + + private = devm_kzalloc(drm_dev->dev, sizeof(*private), GFP_KERNEL); + if (!private) + return -ENOMEM; + + drm_dev->dev_private = private; + + drm_mode_config_init(drm_dev); + + rockchip_drm_mode_config_init(drm_dev); + + dev->dma_parms = devm_kzalloc(dev, sizeof(*dev->dma_parms), + GFP_KERNEL); + if (!dev->dma_parms) { + ret = -ENOMEM; + goto err_config_cleanup; + } + + /* TODO(djkurtz): fetch the mapping start/size from somewhere */ + mapping = arm_iommu_create_mapping(&platform_bus_type, 0x00000000, + SZ_2G); + if (IS_ERR(mapping)) { + ret = PTR_ERR(mapping); + goto err_config_cleanup; + } + + ret = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(32)); + if (ret) + goto err_release_mapping; + + dma_set_max_seg_size(dev, DMA_BIT_MASK(32)); + + ret = arm_iommu_attach_device(dev, mapping); + if (ret) + goto err_release_mapping; + + /* Try to bind all sub drivers. */ + ret = component_bind_all(dev, drm_dev); + if (ret) + goto err_detach_device; + + /* init kms poll for handling hpd */ + drm_kms_helper_poll_init(drm_dev); + + /* + * enable drm irq mode. + * - with irq_enabled = true, we can use the vblank feature. + */ + drm_dev->irq_enabled = true; + + ret = drm_vblank_init(drm_dev, ROCKCHIP_MAX_CRTC); + if (ret) + goto err_kms_helper_poll_fini; + + /* + * with vblank_disable_allowed = true, vblank interrupt will be disabled + * by drm timer once a current process gives up ownership of + * vblank event.(after drm_vblank_put function is called) + */ + drm_dev->vblank_disable_allowed = true; + + ret = rockchip_drm_fbdev_init(drm_dev); + if (ret) + goto err_vblank_cleanup; + + return 0; +err_vblank_cleanup: + drm_vblank_cleanup(drm_dev); +err_kms_helper_poll_fini: + drm_kms_helper_poll_fini(drm_dev); + component_unbind_all(dev, drm_dev); +err_detach_device: + arm_iommu_detach_device(dev); +err_release_mapping: + arm_iommu_release_mapping(dev->archdata.mapping); +err_config_cleanup: + drm_mode_config_cleanup(drm_dev); + drm_dev->dev_private = NULL; + return ret; +} + +static int rockchip_drm_unload(struct drm_device *drm_dev) +{ + struct device *dev = drm_dev->dev; + + rockchip_drm_fbdev_fini(drm_dev); + drm_vblank_cleanup(drm_dev); + drm_kms_helper_poll_fini(drm_dev); + component_unbind_all(dev, drm_dev); + arm_iommu_detach_device(dev); + arm_iommu_release_mapping(dev->archdata.mapping); + drm_mode_config_cleanup(drm_dev); + drm_dev->dev_private = NULL; + + return 0; +} + +void rockchip_drm_lastclose(struct drm_device *dev) +{ + struct rockchip_drm_private *priv = dev->dev_private; + + drm_fb_helper_restore_fbdev_mode_unlocked(&priv->fbdev_helper); +} + +static const struct file_operations rockchip_drm_driver_fops = { + .owner = THIS_MODULE, + .open = drm_open, + .mmap = rockchip_gem_mmap, + .poll = drm_poll, + .read = drm_read, + .unlocked_ioctl = drm_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = drm_compat_ioctl, +#endif + .release = drm_release, +}; + +const struct vm_operations_struct rockchip_drm_vm_ops = { + .open = drm_gem_vm_open, + .close = drm_gem_vm_close, +}; + +static struct drm_driver rockchip_drm_driver = { + .driver_features = DRIVER_MODESET | DRIVER_GEM | DRIVER_PRIME, + .load = rockchip_drm_load, + .unload = rockchip_drm_unload, + .lastclose = rockchip_drm_lastclose, + .get_vblank_counter = drm_vblank_count, + .enable_vblank = rockchip_drm_crtc_enable_vblank, + .disable_vblank = rockchip_drm_crtc_disable_vblank, + .gem_vm_ops = &rockchip_drm_vm_ops, + .gem_free_object = rockchip_gem_free_object, + .dumb_create = rockchip_gem_dumb_create, + .dumb_map_offset = rockchip_gem_dumb_map_offset, + .dumb_destroy = drm_gem_dumb_destroy, + .prime_handle_to_fd = drm_gem_prime_handle_to_fd, + .prime_fd_to_handle = drm_gem_prime_fd_to_handle, + .gem_prime_import = drm_gem_prime_import, + .gem_prime_export = drm_gem_prime_export, + .gem_prime_get_sg_table = rockchip_gem_prime_get_sg_table, + .gem_prime_vmap = rockchip_gem_prime_vmap, + .gem_prime_vunmap = rockchip_gem_prime_vunmap, + .gem_prime_mmap = rockchip_gem_mmap_buf, + .fops = &rockchip_drm_driver_fops, + .name = DRIVER_NAME, + .desc = DRIVER_DESC, + .date = DRIVER_DATE, + .major = DRIVER_MAJOR, + .minor = DRIVER_MINOR, +}; + +#ifdef CONFIG_PM_SLEEP +static int rockchip_drm_sys_suspend(struct device *dev) +{ + struct drm_device *drm = dev_get_drvdata(dev); + struct drm_connector *connector; + + if (!drm) + return 0; + + drm_modeset_lock_all(drm); + list_for_each_entry(connector, &drm->mode_config.connector_list, head) { + int old_dpms = connector->dpms; + + if (connector->funcs->dpms) + connector->funcs->dpms(connector, DRM_MODE_DPMS_OFF); + + /* Set the old mode back to the connector for resume */ + connector->dpms = old_dpms; + } + drm_modeset_unlock_all(drm); + + return 0; +} + +static int rockchip_drm_sys_resume(struct device *dev) +{ + struct drm_device *drm = dev_get_drvdata(dev); + struct drm_connector *connector; + enum drm_connector_status status; + bool changed = false; + + if (!drm) + return 0; + + drm_modeset_lock_all(drm); + list_for_each_entry(connector, &drm->mode_config.connector_list, head) { + int desired_mode = connector->dpms; + + /* + * at suspend time, we save dpms to connector->dpms, + * restore the old_dpms, and at current time, the connector + * dpms status must be DRM_MODE_DPMS_OFF. + */ + connector->dpms = DRM_MODE_DPMS_OFF; + + /* + * If the connector has been disconnected during suspend, + * disconnect it from the encoder and leave it off. We'll notify + * userspace at the end. + */ + if (desired_mode == DRM_MODE_DPMS_ON) { + status = connector->funcs->detect(connector, true); + if (status == connector_status_disconnected) { + connector->encoder = NULL; + connector->status = status; + changed = true; + continue; + } + } + if (connector->funcs->dpms) + connector->funcs->dpms(connector, desired_mode); + } + drm_modeset_unlock_all(drm); + + drm_helper_resume_force_mode(drm); + + if (changed) + drm_kms_helper_hotplug_event(drm); + + return 0; +} +#endif + +static const struct dev_pm_ops rockchip_drm_pm_ops = { + SET_SYSTEM_SLEEP_PM_OPS(rockchip_drm_sys_suspend, + rockchip_drm_sys_resume) +}; + +/* + * @node: device tree node containing encoder input ports + * @encoder: drm_encoder + */ +int rockchip_drm_encoder_get_mux_id(struct device_node *node, + struct drm_encoder *encoder) +{ + struct device_node *ep = NULL; + struct drm_crtc *crtc = encoder->crtc; + struct of_endpoint endpoint; + struct device_node *port; + int ret; + + if (!node || !crtc) + return -EINVAL; + + do { + ep = of_graph_get_next_endpoint(node, ep); + if (!ep) + break; + + port = of_graph_get_remote_port(ep); + of_node_put(port); + if (port == crtc->port) { + ret = of_graph_parse_endpoint(ep, &endpoint); + return ret ?: endpoint.id; + } + } while (ep); + + return -EINVAL; +} + +static int compare_of(struct device *dev, void *data) +{ + struct device_node *np = data; + + return dev->of_node == np; +} + +static void rockchip_add_endpoints(struct device *dev, + struct component_match **match, + struct device_node *port) +{ + struct device_node *ep, *remote; + + for_each_child_of_node(port, ep) { + remote = of_graph_get_remote_port_parent(ep); + if (!remote || !of_device_is_available(remote)) { + of_node_put(remote); + continue; + } else if (!of_device_is_available(remote->parent)) { + dev_warn(dev, "parent device of %s is not available\n", + remote->full_name); + of_node_put(remote); + continue; + } + + component_match_add(dev, match, compare_of, remote); + of_node_put(remote); + } +} + +static int rockchip_drm_bind(struct device *dev) +{ + struct drm_device *drm; + int ret; + + drm = drm_dev_alloc(&rockchip_drm_driver, dev); + if (!drm) + return -ENOMEM; + + ret = drm_dev_set_unique(drm, "%s", dev_name(dev)); + if (ret) + goto err_free; + + ret = drm_dev_register(drm, 0); + if (ret) + goto err_free; + + dev_set_drvdata(dev, drm); + + return 0; + +err_free: + drm_dev_unref(drm); + return ret; +} + +static void rockchip_drm_unbind(struct device *dev) +{ + struct drm_device *drm = dev_get_drvdata(dev); + + drm_dev_unregister(drm); + drm_dev_unref(drm); + dev_set_drvdata(dev, NULL); +} + +static const struct component_master_ops rockchip_drm_ops = { + .bind = rockchip_drm_bind, + .unbind = rockchip_drm_unbind, +}; + +static int rockchip_drm_platform_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct component_match *match = NULL; + struct device_node *np = dev->of_node; + struct device_node *port; + int i; + + if (!np) + return -ENODEV; + /* + * Bind the crtc ports first, so that + * drm_of_find_possible_crtcs called from encoder .bind callbacks + * works as expected. + */ + for (i = 0;; i++) { + port = of_parse_phandle(np, "ports", i); + if (!port) + break; + + if (!of_device_is_available(port->parent)) { + of_node_put(port); + continue; + } + + component_match_add(dev, &match, compare_of, port->parent); + of_node_put(port); + } + + if (i == 0) { + dev_err(dev, "missing 'ports' property\n"); + return -ENODEV; + } + + if (!match) { + dev_err(dev, "No available vop found for display-subsystem.\n"); + return -ENODEV; + } + /* + * For each bound crtc, bind the encoders attached to its + * remote endpoint. + */ + for (i = 0;; i++) { + port = of_parse_phandle(np, "ports", i); + if (!port) + break; + + if (!of_device_is_available(port->parent)) { + of_node_put(port); + continue; + } + + rockchip_add_endpoints(dev, &match, port); + of_node_put(port); + } + + return component_master_add_with_match(dev, &rockchip_drm_ops, match); +} + +static int rockchip_drm_platform_remove(struct platform_device *pdev) +{ + component_master_del(&pdev->dev, &rockchip_drm_ops); + + return 0; +} + +static const struct of_device_id rockchip_drm_dt_ids[] = { + { .compatible = "rockchip,display-subsystem", }, + { /* sentinel */ }, +}; +MODULE_DEVICE_TABLE(of, rockchip_drm_dt_ids); + +static struct platform_driver rockchip_drm_platform_driver = { + .probe = rockchip_drm_platform_probe, + .remove = rockchip_drm_platform_remove, + .driver = { + .owner = THIS_MODULE, + .name = "rockchip-drm", + .of_match_table = rockchip_drm_dt_ids, + .pm = &rockchip_drm_pm_ops, + }, +}; + +module_platform_driver(rockchip_drm_platform_driver); + +MODULE_AUTHOR("Mark Yao <mark.yao@rock-chips.com>"); +MODULE_DESCRIPTION("ROCKCHIP DRM Driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.h b/drivers/gpu/drm/rockchip/rockchip_drm_drv.h new file mode 100644 index 000000000000..dc4e5f03ac79 --- /dev/null +++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.h @@ -0,0 +1,68 @@ +/* + * Copyright (C) Fuzhou Rockchip Electronics Co.Ltd + * Author:Mark Yao <mark.yao@rock-chips.com> + * + * based on exynos_drm_drv.h + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#ifndef _ROCKCHIP_DRM_DRV_H +#define _ROCKCHIP_DRM_DRV_H + +#include <drm/drm_fb_helper.h> +#include <drm/drm_gem.h> + +#include <linux/module.h> +#include <linux/component.h> + +#define ROCKCHIP_MAX_FB_BUFFER 3 +#define ROCKCHIP_MAX_CONNECTOR 2 +#define ROCKCHIP_MAX_CRTC 2 + +struct drm_device; +struct drm_connector; + +/* + * Rockchip drm private crtc funcs. + * @enable_vblank: enable crtc vblank irq. + * @disable_vblank: disable crtc vblank irq. + */ +struct rockchip_crtc_funcs { + int (*enable_vblank)(struct drm_crtc *crtc); + void (*disable_vblank)(struct drm_crtc *crtc); +}; + +/* + * Rockchip drm private structure. + * + * @crtc: array of enabled CRTCs, used to map from "pipe" to drm_crtc. + * @num_pipe: number of pipes for this device. + */ +struct rockchip_drm_private { + struct drm_fb_helper fbdev_helper; + struct drm_gem_object *fbdev_bo; + const struct rockchip_crtc_funcs *crtc_funcs[ROCKCHIP_MAX_CRTC]; +}; + +int rockchip_register_crtc_funcs(struct drm_device *dev, + const struct rockchip_crtc_funcs *crtc_funcs, + int pipe); +void rockchip_unregister_crtc_funcs(struct drm_device *dev, int pipe); +int rockchip_drm_encoder_get_mux_id(struct device_node *node, + struct drm_encoder *encoder); +int rockchip_drm_crtc_mode_config(struct drm_crtc *crtc, int connector_type, + int out_mode); +int rockchip_drm_dma_attach_device(struct drm_device *drm_dev, + struct device *dev); +void rockchip_drm_dma_detach_device(struct drm_device *drm_dev, + struct device *dev); + +#endif /* _ROCKCHIP_DRM_DRV_H_ */ diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_fb.c b/drivers/gpu/drm/rockchip/rockchip_drm_fb.c new file mode 100644 index 000000000000..77d52893d40f --- /dev/null +++ b/drivers/gpu/drm/rockchip/rockchip_drm_fb.c @@ -0,0 +1,201 @@ +/* + * Copyright (C) Fuzhou Rockchip Electronics Co.Ltd + * Author:Mark Yao <mark.yao@rock-chips.com> + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <linux/kernel.h> +#include <drm/drm.h> +#include <drm/drmP.h> +#include <drm/drm_fb_helper.h> +#include <drm/drm_crtc_helper.h> + +#include "rockchip_drm_drv.h" +#include "rockchip_drm_gem.h" + +#define to_rockchip_fb(x) container_of(x, struct rockchip_drm_fb, fb) + +struct rockchip_drm_fb { + struct drm_framebuffer fb; + struct drm_gem_object *obj[ROCKCHIP_MAX_FB_BUFFER]; +}; + +struct drm_gem_object *rockchip_fb_get_gem_obj(struct drm_framebuffer *fb, + unsigned int plane) +{ + struct rockchip_drm_fb *rk_fb = to_rockchip_fb(fb); + + if (plane >= ROCKCHIP_MAX_FB_BUFFER) + return NULL; + + return rk_fb->obj[plane]; +} +EXPORT_SYMBOL_GPL(rockchip_fb_get_gem_obj); + +static void rockchip_drm_fb_destroy(struct drm_framebuffer *fb) +{ + struct rockchip_drm_fb *rockchip_fb = to_rockchip_fb(fb); + struct drm_gem_object *obj; + int i; + + for (i = 0; i < ROCKCHIP_MAX_FB_BUFFER; i++) { + obj = rockchip_fb->obj[i]; + if (obj) + drm_gem_object_unreference_unlocked(obj); + } + + drm_framebuffer_cleanup(fb); + kfree(rockchip_fb); +} + +static int rockchip_drm_fb_create_handle(struct drm_framebuffer *fb, + struct drm_file *file_priv, + unsigned int *handle) +{ + struct rockchip_drm_fb *rockchip_fb = to_rockchip_fb(fb); + + return drm_gem_handle_create(file_priv, + rockchip_fb->obj[0], handle); +} + +static struct drm_framebuffer_funcs rockchip_drm_fb_funcs = { + .destroy = rockchip_drm_fb_destroy, + .create_handle = rockchip_drm_fb_create_handle, +}; + +static struct rockchip_drm_fb * +rockchip_fb_alloc(struct drm_device *dev, struct drm_mode_fb_cmd2 *mode_cmd, + struct drm_gem_object **obj, unsigned int num_planes) +{ + struct rockchip_drm_fb *rockchip_fb; + int ret; + int i; + + rockchip_fb = kzalloc(sizeof(*rockchip_fb), GFP_KERNEL); + if (!rockchip_fb) + return ERR_PTR(-ENOMEM); + + drm_helper_mode_fill_fb_struct(&rockchip_fb->fb, mode_cmd); + + for (i = 0; i < num_planes; i++) + rockchip_fb->obj[i] = obj[i]; + + ret = drm_framebuffer_init(dev, &rockchip_fb->fb, + &rockchip_drm_fb_funcs); + if (ret) { + dev_err(dev->dev, "Failed to initialize framebuffer: %d\n", + ret); + kfree(rockchip_fb); + return ERR_PTR(ret); + } + + return rockchip_fb; +} + +static struct drm_framebuffer * +rockchip_user_fb_create(struct drm_device *dev, struct drm_file *file_priv, + struct drm_mode_fb_cmd2 *mode_cmd) +{ + struct rockchip_drm_fb *rockchip_fb; + struct drm_gem_object *objs[ROCKCHIP_MAX_FB_BUFFER]; + struct drm_gem_object *obj; + unsigned int hsub; + unsigned int vsub; + int num_planes; + int ret; + int i; + + hsub = drm_format_horz_chroma_subsampling(mode_cmd->pixel_format); + vsub = drm_format_vert_chroma_subsampling(mode_cmd->pixel_format); + num_planes = min(drm_format_num_planes(mode_cmd->pixel_format), + ROCKCHIP_MAX_FB_BUFFER); + + for (i = 0; i < num_planes; i++) { + unsigned int width = mode_cmd->width / (i ? hsub : 1); + unsigned int height = mode_cmd->height / (i ? vsub : 1); + unsigned int min_size; + + obj = drm_gem_object_lookup(dev, file_priv, + mode_cmd->handles[i]); + if (!obj) { + dev_err(dev->dev, "Failed to lookup GEM object\n"); + ret = -ENXIO; + goto err_gem_object_unreference; + } + + min_size = (height - 1) * mode_cmd->pitches[i] + + mode_cmd->offsets[i] + + width * drm_format_plane_cpp(mode_cmd->pixel_format, i); + + if (obj->size < min_size) { + drm_gem_object_unreference_unlocked(obj); + ret = -EINVAL; + goto err_gem_object_unreference; + } + objs[i] = obj; + } + + rockchip_fb = rockchip_fb_alloc(dev, mode_cmd, objs, i); + if (IS_ERR(rockchip_fb)) { + ret = PTR_ERR(rockchip_fb); + goto err_gem_object_unreference; + } + + return &rockchip_fb->fb; + +err_gem_object_unreference: + for (i--; i >= 0; i--) + drm_gem_object_unreference_unlocked(objs[i]); + return ERR_PTR(ret); +} + +static void rockchip_drm_output_poll_changed(struct drm_device *dev) +{ + struct rockchip_drm_private *private = dev->dev_private; + struct drm_fb_helper *fb_helper = &private->fbdev_helper; + + drm_fb_helper_hotplug_event(fb_helper); +} + +static const struct drm_mode_config_funcs rockchip_drm_mode_config_funcs = { + .fb_create = rockchip_user_fb_create, + .output_poll_changed = rockchip_drm_output_poll_changed, +}; + +struct drm_framebuffer * +rockchip_drm_framebuffer_init(struct drm_device *dev, + struct drm_mode_fb_cmd2 *mode_cmd, + struct drm_gem_object *obj) +{ + struct rockchip_drm_fb *rockchip_fb; + + rockchip_fb = rockchip_fb_alloc(dev, mode_cmd, &obj, 1); + if (IS_ERR(rockchip_fb)) + return NULL; + + return &rockchip_fb->fb; +} + +void rockchip_drm_mode_config_init(struct drm_device *dev) +{ + dev->mode_config.min_width = 0; + dev->mode_config.min_height = 0; + + /* + * set max width and height as default value(4096x4096). + * this value would be used to check framebuffer size limitation + * at drm_mode_addfb(). + */ + dev->mode_config.max_width = 4096; + dev->mode_config.max_height = 4096; + + dev->mode_config.funcs = &rockchip_drm_mode_config_funcs; +} diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_fb.h b/drivers/gpu/drm/rockchip/rockchip_drm_fb.h new file mode 100644 index 000000000000..09574d48226f --- /dev/null +++ b/drivers/gpu/drm/rockchip/rockchip_drm_fb.h @@ -0,0 +1,28 @@ +/* + * Copyright (C) Fuzhou Rockchip Electronics Co.Ltd + * Author:Mark Yao <mark.yao@rock-chips.com> + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#ifndef _ROCKCHIP_DRM_FB_H +#define _ROCKCHIP_DRM_FB_H + +struct drm_framebuffer * +rockchip_drm_framebuffer_init(struct drm_device *dev, + struct drm_mode_fb_cmd2 *mode_cmd, + struct drm_gem_object *obj); +void rockchip_drm_framebuffer_fini(struct drm_framebuffer *fb); + +void rockchip_drm_mode_config_init(struct drm_device *dev); + +struct drm_gem_object *rockchip_fb_get_gem_obj(struct drm_framebuffer *fb, + unsigned int plane); +#endif /* _ROCKCHIP_DRM_FB_H */ diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c b/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c new file mode 100644 index 000000000000..a5d889a8716b --- /dev/null +++ b/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.c @@ -0,0 +1,210 @@ +/* + * Copyright (C) Fuzhou Rockchip Electronics Co.Ltd + * Author:Mark Yao <mark.yao@rock-chips.com> + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <drm/drm.h> +#include <drm/drmP.h> +#include <drm/drm_fb_helper.h> +#include <drm/drm_crtc_helper.h> + +#include "rockchip_drm_drv.h" +#include "rockchip_drm_gem.h" +#include "rockchip_drm_fb.h" + +#define PREFERRED_BPP 32 +#define to_drm_private(x) \ + container_of(x, struct rockchip_drm_private, fbdev_helper) + +static int rockchip_fbdev_mmap(struct fb_info *info, + struct vm_area_struct *vma) +{ + struct drm_fb_helper *helper = info->par; + struct rockchip_drm_private *private = to_drm_private(helper); + + return rockchip_gem_mmap_buf(private->fbdev_bo, vma); +} + +static struct fb_ops rockchip_drm_fbdev_ops = { + .owner = THIS_MODULE, + .fb_mmap = rockchip_fbdev_mmap, + .fb_fillrect = cfb_fillrect, + .fb_copyarea = cfb_copyarea, + .fb_imageblit = cfb_imageblit, + .fb_check_var = drm_fb_helper_check_var, + .fb_set_par = drm_fb_helper_set_par, + .fb_blank = drm_fb_helper_blank, + .fb_pan_display = drm_fb_helper_pan_display, + .fb_setcmap = drm_fb_helper_setcmap, +}; + +static int rockchip_drm_fbdev_create(struct drm_fb_helper *helper, + struct drm_fb_helper_surface_size *sizes) +{ + struct rockchip_drm_private *private = to_drm_private(helper); + struct drm_mode_fb_cmd2 mode_cmd = { 0 }; + struct drm_device *dev = helper->dev; + struct rockchip_gem_object *rk_obj; + struct drm_framebuffer *fb; + unsigned int bytes_per_pixel; + unsigned long offset; + struct fb_info *fbi; + size_t size; + int ret; + + bytes_per_pixel = DIV_ROUND_UP(sizes->surface_bpp, 8); + + mode_cmd.width = sizes->surface_width; + mode_cmd.height = sizes->surface_height; + mode_cmd.pitches[0] = sizes->surface_width * bytes_per_pixel; + mode_cmd.pixel_format = drm_mode_legacy_fb_format(sizes->surface_bpp, + sizes->surface_depth); + + size = mode_cmd.pitches[0] * mode_cmd.height; + + rk_obj = rockchip_gem_create_object(dev, size); + if (IS_ERR(rk_obj)) + return -ENOMEM; + + private->fbdev_bo = &rk_obj->base; + + fbi = framebuffer_alloc(0, dev->dev); + if (!fbi) { + dev_err(dev->dev, "Failed to allocate framebuffer info.\n"); + ret = -ENOMEM; + goto err_rockchip_gem_free_object; + } + + helper->fb = rockchip_drm_framebuffer_init(dev, &mode_cmd, + private->fbdev_bo); + if (IS_ERR(helper->fb)) { + dev_err(dev->dev, "Failed to allocate DRM framebuffer.\n"); + ret = PTR_ERR(helper->fb); + goto err_framebuffer_release; + } + + helper->fbdev = fbi; + + fbi->par = helper; + fbi->flags = FBINFO_FLAG_DEFAULT; + fbi->fbops = &rockchip_drm_fbdev_ops; + + ret = fb_alloc_cmap(&fbi->cmap, 256, 0); + if (ret) { + dev_err(dev->dev, "Failed to allocate color map.\n"); + goto err_drm_framebuffer_unref; + } + + fb = helper->fb; + drm_fb_helper_fill_fix(fbi, fb->pitches[0], fb->depth); + drm_fb_helper_fill_var(fbi, helper, fb->width, fb->height); + + offset = fbi->var.xoffset * bytes_per_pixel; + offset += fbi->var.yoffset * fb->pitches[0]; + + dev->mode_config.fb_base = 0; + fbi->screen_base = rk_obj->kvaddr + offset; + fbi->screen_size = rk_obj->base.size; + fbi->fix.smem_len = rk_obj->base.size; + + DRM_DEBUG_KMS("FB [%dx%d]-%d kvaddr=%p offset=%ld size=%d\n", + fb->width, fb->height, fb->depth, rk_obj->kvaddr, + offset, size); + return 0; + +err_drm_framebuffer_unref: + drm_framebuffer_unreference(helper->fb); +err_framebuffer_release: + framebuffer_release(fbi); +err_rockchip_gem_free_object: + rockchip_gem_free_object(&rk_obj->base); + return ret; +} + +static const struct drm_fb_helper_funcs rockchip_drm_fb_helper_funcs = { + .fb_probe = rockchip_drm_fbdev_create, +}; + +int rockchip_drm_fbdev_init(struct drm_device *dev) +{ + struct rockchip_drm_private *private = dev->dev_private; + struct drm_fb_helper *helper; + unsigned int num_crtc; + int ret; + + if (!dev->mode_config.num_crtc || !dev->mode_config.num_connector) + return -EINVAL; + + num_crtc = dev->mode_config.num_crtc; + + helper = &private->fbdev_helper; + + drm_fb_helper_prepare(dev, helper, &rockchip_drm_fb_helper_funcs); + + ret = drm_fb_helper_init(dev, helper, num_crtc, ROCKCHIP_MAX_CONNECTOR); + if (ret < 0) { + dev_err(dev->dev, "Failed to initialize drm fb helper - %d.\n", + ret); + return ret; + } + + ret = drm_fb_helper_single_add_all_connectors(helper); + if (ret < 0) { + dev_err(dev->dev, "Failed to add connectors - %d.\n", ret); + goto err_drm_fb_helper_fini; + } + + /* disable all the possible outputs/crtcs before entering KMS mode */ + drm_helper_disable_unused_functions(dev); + + ret = drm_fb_helper_initial_config(helper, PREFERRED_BPP); + if (ret < 0) { + dev_err(dev->dev, "Failed to set initial hw config - %d.\n", + ret); + goto err_drm_fb_helper_fini; + } + + return 0; + +err_drm_fb_helper_fini: + drm_fb_helper_fini(helper); + return ret; +} + +void rockchip_drm_fbdev_fini(struct drm_device *dev) +{ + struct rockchip_drm_private *private = dev->dev_private; + struct drm_fb_helper *helper; + + helper = &private->fbdev_helper; + + if (helper->fbdev) { + struct fb_info *info; + int ret; + + info = helper->fbdev; + ret = unregister_framebuffer(info); + if (ret < 0) + DRM_DEBUG_KMS("failed unregister_framebuffer() - %d\n", + ret); + + if (info->cmap.len) + fb_dealloc_cmap(&info->cmap); + + framebuffer_release(info); + } + + if (helper->fb) + drm_framebuffer_unreference(helper->fb); + + drm_fb_helper_fini(helper); +} diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.h b/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.h new file mode 100644 index 000000000000..50432e9b5b37 --- /dev/null +++ b/drivers/gpu/drm/rockchip/rockchip_drm_fbdev.h @@ -0,0 +1,21 @@ +/* + * Copyright (C) Fuzhou Rockchip Electronics Co.Ltd + * Author:Mark Yao <mark.yao@rock-chips.com> + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#ifndef _ROCKCHIP_DRM_FBDEV_H +#define _ROCKCHIP_DRM_FBDEV_H + +int rockchip_drm_fbdev_init(struct drm_device *dev); +void rockchip_drm_fbdev_fini(struct drm_device *dev); + +#endif /* _ROCKCHIP_DRM_FBDEV_H */ diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c new file mode 100644 index 000000000000..bc98a227dc76 --- /dev/null +++ b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c @@ -0,0 +1,294 @@ +/* + * Copyright (C) Fuzhou Rockchip Electronics Co.Ltd + * Author:Mark Yao <mark.yao@rock-chips.com> + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <drm/drm.h> +#include <drm/drmP.h> +#include <drm/drm_gem.h> +#include <drm/drm_vma_manager.h> + +#include <linux/dma-attrs.h> + +#include "rockchip_drm_drv.h" +#include "rockchip_drm_gem.h" + +static int rockchip_gem_alloc_buf(struct rockchip_gem_object *rk_obj) +{ + struct drm_gem_object *obj = &rk_obj->base; + struct drm_device *drm = obj->dev; + + init_dma_attrs(&rk_obj->dma_attrs); + dma_set_attr(DMA_ATTR_WRITE_COMBINE, &rk_obj->dma_attrs); + + /* TODO(djkurtz): Use DMA_ATTR_NO_KERNEL_MAPPING except for fbdev */ + rk_obj->kvaddr = dma_alloc_attrs(drm->dev, obj->size, + &rk_obj->dma_addr, GFP_KERNEL, + &rk_obj->dma_attrs); + if (IS_ERR(rk_obj->kvaddr)) { + int ret = PTR_ERR(rk_obj->kvaddr); + + DRM_ERROR("failed to allocate %#x byte dma buffer, %d", + obj->size, ret); + return ret; + } + + return 0; +} + +static void rockchip_gem_free_buf(struct rockchip_gem_object *rk_obj) +{ + struct drm_gem_object *obj = &rk_obj->base; + struct drm_device *drm = obj->dev; + + dma_free_attrs(drm->dev, obj->size, rk_obj->kvaddr, rk_obj->dma_addr, + &rk_obj->dma_attrs); +} + +int rockchip_gem_mmap_buf(struct drm_gem_object *obj, + struct vm_area_struct *vma) +{ + struct rockchip_gem_object *rk_obj = to_rockchip_obj(obj); + struct drm_device *drm = obj->dev; + unsigned long vm_size; + + vma->vm_flags |= VM_IO | VM_DONTEXPAND | VM_DONTDUMP; + vm_size = vma->vm_end - vma->vm_start; + + if (vm_size > obj->size) + return -EINVAL; + + return dma_mmap_attrs(drm->dev, vma, rk_obj->kvaddr, rk_obj->dma_addr, + obj->size, &rk_obj->dma_attrs); +} + +/* drm driver mmap file operations */ +int rockchip_gem_mmap(struct file *filp, struct vm_area_struct *vma) +{ + struct drm_file *priv = filp->private_data; + struct drm_device *dev = priv->minor->dev; + struct drm_gem_object *obj; + struct drm_vma_offset_node *node; + int ret; + + if (drm_device_is_unplugged(dev)) + return -ENODEV; + + mutex_lock(&dev->struct_mutex); + + node = drm_vma_offset_exact_lookup(dev->vma_offset_manager, + vma->vm_pgoff, + vma_pages(vma)); + if (!node) { + mutex_unlock(&dev->struct_mutex); + DRM_ERROR("failed to find vma node.\n"); + return -EINVAL; + } else if (!drm_vma_node_is_allowed(node, filp)) { + mutex_unlock(&dev->struct_mutex); + return -EACCES; + } + + obj = container_of(node, struct drm_gem_object, vma_node); + ret = rockchip_gem_mmap_buf(obj, vma); + + mutex_unlock(&dev->struct_mutex); + + return ret; +} + +struct rockchip_gem_object * + rockchip_gem_create_object(struct drm_device *drm, unsigned int size) +{ + struct rockchip_gem_object *rk_obj; + struct drm_gem_object *obj; + int ret; + + size = round_up(size, PAGE_SIZE); + + rk_obj = kzalloc(sizeof(*rk_obj), GFP_KERNEL); + if (!rk_obj) + return ERR_PTR(-ENOMEM); + + obj = &rk_obj->base; + + drm_gem_private_object_init(drm, obj, size); + + ret = rockchip_gem_alloc_buf(rk_obj); + if (ret) + goto err_free_rk_obj; + + return rk_obj; + +err_free_rk_obj: + kfree(rk_obj); + return ERR_PTR(ret); +} + +/* + * rockchip_gem_free_object - (struct drm_driver)->gem_free_object callback + * function + */ +void rockchip_gem_free_object(struct drm_gem_object *obj) +{ + struct rockchip_gem_object *rk_obj; + + drm_gem_free_mmap_offset(obj); + + rk_obj = to_rockchip_obj(obj); + + rockchip_gem_free_buf(rk_obj); + + kfree(rk_obj); +} + +/* + * rockchip_gem_create_with_handle - allocate an object with the given + * size and create a gem handle on it + * + * returns a struct rockchip_gem_object* on success or ERR_PTR values + * on failure. + */ +static struct rockchip_gem_object * +rockchip_gem_create_with_handle(struct drm_file *file_priv, + struct drm_device *drm, unsigned int size, + unsigned int *handle) +{ + struct rockchip_gem_object *rk_obj; + struct drm_gem_object *obj; + int ret; + + rk_obj = rockchip_gem_create_object(drm, size); + if (IS_ERR(rk_obj)) + return ERR_CAST(rk_obj); + + obj = &rk_obj->base; + + /* + * allocate a id of idr table where the obj is registered + * and handle has the id what user can see. + */ + ret = drm_gem_handle_create(file_priv, obj, handle); + if (ret) + goto err_handle_create; + + /* drop reference from allocate - handle holds it now. */ + drm_gem_object_unreference_unlocked(obj); + + return rk_obj; + +err_handle_create: + rockchip_gem_free_object(obj); + + return ERR_PTR(ret); +} + +int rockchip_gem_dumb_map_offset(struct drm_file *file_priv, + struct drm_device *dev, uint32_t handle, + uint64_t *offset) +{ + struct drm_gem_object *obj; + int ret; + + mutex_lock(&dev->struct_mutex); + + obj = drm_gem_object_lookup(dev, file_priv, handle); + if (!obj) { + DRM_ERROR("failed to lookup gem object.\n"); + ret = -EINVAL; + goto unlock; + } + + ret = drm_gem_create_mmap_offset(obj); + if (ret) + goto out; + + *offset = drm_vma_node_offset_addr(&obj->vma_node); + DRM_DEBUG_KMS("offset = 0x%llx\n", *offset); + +out: + drm_gem_object_unreference(obj); +unlock: + mutex_unlock(&dev->struct_mutex); + return ret; +} + +/* + * rockchip_gem_dumb_create - (struct drm_driver)->dumb_create callback + * function + * + * This aligns the pitch and size arguments to the minimum required. wrap + * this into your own function if you need bigger alignment. + */ +int rockchip_gem_dumb_create(struct drm_file *file_priv, + struct drm_device *dev, + struct drm_mode_create_dumb *args) +{ + struct rockchip_gem_object *rk_obj; + int min_pitch = DIV_ROUND_UP(args->width * args->bpp, 8); + + /* + * align to 64 bytes since Mali requires it. + */ + min_pitch = ALIGN(min_pitch, 64); + + if (args->pitch < min_pitch) + args->pitch = min_pitch; + + if (args->size < args->pitch * args->height) + args->size = args->pitch * args->height; + + rk_obj = rockchip_gem_create_with_handle(file_priv, dev, args->size, + &args->handle); + + return PTR_ERR_OR_ZERO(rk_obj); +} + +/* + * Allocate a sg_table for this GEM object. + * Note: Both the table's contents, and the sg_table itself must be freed by + * the caller. + * Returns a pointer to the newly allocated sg_table, or an ERR_PTR() error. + */ +struct sg_table *rockchip_gem_prime_get_sg_table(struct drm_gem_object *obj) +{ + struct rockchip_gem_object *rk_obj = to_rockchip_obj(obj); + struct drm_device *drm = obj->dev; + struct sg_table *sgt; + int ret; + + sgt = kzalloc(sizeof(*sgt), GFP_KERNEL); + if (!sgt) + return ERR_PTR(-ENOMEM); + + ret = dma_get_sgtable_attrs(drm->dev, sgt, rk_obj->kvaddr, + rk_obj->dma_addr, obj->size, + &rk_obj->dma_attrs); + if (ret) { + DRM_ERROR("failed to allocate sgt, %d\n", ret); + kfree(sgt); + return ERR_PTR(ret); + } + + return sgt; +} + +void *rockchip_gem_prime_vmap(struct drm_gem_object *obj) +{ + struct rockchip_gem_object *rk_obj = to_rockchip_obj(obj); + + return rk_obj->kvaddr; +} + +void rockchip_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr) +{ + /* Nothing to do */ +} diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_gem.h b/drivers/gpu/drm/rockchip/rockchip_drm_gem.h new file mode 100644 index 000000000000..67bcebe90003 --- /dev/null +++ b/drivers/gpu/drm/rockchip/rockchip_drm_gem.h @@ -0,0 +1,54 @@ +/* + * Copyright (C) Fuzhou Rockchip Electronics Co.Ltd + * Author:Mark Yao <mark.yao@rock-chips.com> + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#ifndef _ROCKCHIP_DRM_GEM_H +#define _ROCKCHIP_DRM_GEM_H + +#define to_rockchip_obj(x) container_of(x, struct rockchip_gem_object, base) + +struct rockchip_gem_object { + struct drm_gem_object base; + unsigned int flags; + + void *kvaddr; + dma_addr_t dma_addr; + struct dma_attrs dma_attrs; +}; + +struct sg_table *rockchip_gem_prime_get_sg_table(struct drm_gem_object *obj); +struct drm_gem_object * +rockchip_gem_prime_import_sg_table(struct drm_device *dev, size_t size, + struct sg_table *sgt); +void *rockchip_gem_prime_vmap(struct drm_gem_object *obj); +void rockchip_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr); + +/* drm driver mmap file operations */ +int rockchip_gem_mmap(struct file *filp, struct vm_area_struct *vma); + +/* mmap a gem object to userspace. */ +int rockchip_gem_mmap_buf(struct drm_gem_object *obj, + struct vm_area_struct *vma); + +struct rockchip_gem_object * + rockchip_gem_create_object(struct drm_device *drm, unsigned int size); + +void rockchip_gem_free_object(struct drm_gem_object *obj); + +int rockchip_gem_dumb_create(struct drm_file *file_priv, + struct drm_device *dev, + struct drm_mode_create_dumb *args); +int rockchip_gem_dumb_map_offset(struct drm_file *file_priv, + struct drm_device *dev, uint32_t handle, + uint64_t *offset); +#endif /* _ROCKCHIP_DRM_GEM_H */ diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c new file mode 100644 index 000000000000..e7ca25b3fb38 --- /dev/null +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c @@ -0,0 +1,1455 @@ +/* + * Copyright (C) Fuzhou Rockchip Electronics Co.Ltd + * Author:Mark Yao <mark.yao@rock-chips.com> + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <drm/drm.h> +#include <drm/drmP.h> +#include <drm/drm_crtc.h> +#include <drm/drm_crtc_helper.h> +#include <drm/drm_plane_helper.h> + +#include <linux/kernel.h> +#include <linux/platform_device.h> +#include <linux/clk.h> +#include <linux/of.h> +#include <linux/of_device.h> +#include <linux/pm_runtime.h> +#include <linux/component.h> + +#include <linux/reset.h> +#include <linux/delay.h> + +#include "rockchip_drm_drv.h" +#include "rockchip_drm_gem.h" +#include "rockchip_drm_fb.h" +#include "rockchip_drm_vop.h" + +#define VOP_REG(off, _mask, s) \ + {.offset = off, \ + .mask = _mask, \ + .shift = s,} + +#define __REG_SET_RELAXED(x, off, mask, shift, v) \ + vop_mask_write_relaxed(x, off, (mask) << shift, (v) << shift) +#define __REG_SET_NORMAL(x, off, mask, shift, v) \ + vop_mask_write(x, off, (mask) << shift, (v) << shift) + +#define REG_SET(x, base, reg, v, mode) \ + __REG_SET_##mode(x, base + reg.offset, reg.mask, reg.shift, v) + +#define VOP_WIN_SET(x, win, name, v) \ + REG_SET(x, win->base, win->phy->name, v, RELAXED) +#define VOP_CTRL_SET(x, name, v) \ + REG_SET(x, 0, (x)->data->ctrl->name, v, NORMAL) + +#define VOP_WIN_GET(x, win, name) \ + vop_read_reg(x, win->base, &win->phy->name) + +#define VOP_WIN_GET_YRGBADDR(vop, win) \ + vop_readl(vop, win->base + win->phy->yrgb_mst.offset) + +#define to_vop(x) container_of(x, struct vop, crtc) +#define to_vop_win(x) container_of(x, struct vop_win, base) + +struct vop_win_state { + struct list_head head; + struct drm_framebuffer *fb; + dma_addr_t yrgb_mst; + struct drm_pending_vblank_event *event; +}; + +struct vop_win { + struct drm_plane base; + const struct vop_win_data *data; + struct vop *vop; + + struct list_head pending; + struct vop_win_state *active; +}; + +struct vop { + struct drm_crtc crtc; + struct device *dev; + struct drm_device *drm_dev; + unsigned int dpms; + + int connector_type; + int connector_out_mode; + + /* mutex vsync_ work */ + struct mutex vsync_mutex; + bool vsync_work_pending; + + const struct vop_data *data; + + uint32_t *regsbak; + void __iomem *regs; + + /* physical map length of vop register */ + uint32_t len; + + /* one time only one process allowed to config the register */ + spinlock_t reg_lock; + /* lock vop irq reg */ + spinlock_t irq_lock; + + unsigned int irq; + + /* vop AHP clk */ + struct clk *hclk; + /* vop dclk */ + struct clk *dclk; + /* vop share memory frequency */ + struct clk *aclk; + + /* vop dclk reset */ + struct reset_control *dclk_rst; + + int pipe; + + struct vop_win win[]; +}; + +enum vop_data_format { + VOP_FMT_ARGB8888 = 0, + VOP_FMT_RGB888, + VOP_FMT_RGB565, + VOP_FMT_YUV420SP = 4, + VOP_FMT_YUV422SP, + VOP_FMT_YUV444SP, +}; + +struct vop_reg_data { + uint32_t offset; + uint32_t value; +}; + +struct vop_reg { + uint32_t offset; + uint32_t shift; + uint32_t mask; +}; + +struct vop_ctrl { + struct vop_reg standby; + struct vop_reg data_blank; + struct vop_reg gate_en; + struct vop_reg mmu_en; + struct vop_reg rgb_en; + struct vop_reg edp_en; + struct vop_reg hdmi_en; + struct vop_reg mipi_en; + struct vop_reg out_mode; + struct vop_reg dither_down; + struct vop_reg dither_up; + struct vop_reg pin_pol; + + struct vop_reg htotal_pw; + struct vop_reg hact_st_end; + struct vop_reg vtotal_pw; + struct vop_reg vact_st_end; + struct vop_reg hpost_st_end; + struct vop_reg vpost_st_end; +}; + +struct vop_win_phy { + const uint32_t *data_formats; + uint32_t nformats; + + struct vop_reg enable; + struct vop_reg format; + struct vop_reg act_info; + struct vop_reg dsp_info; + struct vop_reg dsp_st; + struct vop_reg yrgb_mst; + struct vop_reg uv_mst; + struct vop_reg yrgb_vir; + struct vop_reg uv_vir; + + struct vop_reg dst_alpha_ctl; + struct vop_reg src_alpha_ctl; +}; + +struct vop_win_data { + uint32_t base; + const struct vop_win_phy *phy; + enum drm_plane_type type; +}; + +struct vop_data { + const struct vop_reg_data *init_table; + unsigned int table_size; + const struct vop_ctrl *ctrl; + const struct vop_win_data *win; + unsigned int win_size; +}; + +static const uint32_t formats_01[] = { + DRM_FORMAT_XRGB8888, + DRM_FORMAT_ARGB8888, + DRM_FORMAT_RGB888, + DRM_FORMAT_RGB565, + DRM_FORMAT_NV12, + DRM_FORMAT_NV16, + DRM_FORMAT_NV24, +}; + +static const uint32_t formats_234[] = { + DRM_FORMAT_XRGB8888, + DRM_FORMAT_ARGB8888, + DRM_FORMAT_RGB888, + DRM_FORMAT_RGB565, +}; + +static const struct vop_win_phy win01_data = { + .data_formats = formats_01, + .nformats = ARRAY_SIZE(formats_01), + .enable = VOP_REG(WIN0_CTRL0, 0x1, 0), + .format = VOP_REG(WIN0_CTRL0, 0x7, 1), + .act_info = VOP_REG(WIN0_ACT_INFO, 0x1fff1fff, 0), + .dsp_info = VOP_REG(WIN0_DSP_INFO, 0x0fff0fff, 0), + .dsp_st = VOP_REG(WIN0_DSP_ST, 0x1fff1fff, 0), + .yrgb_mst = VOP_REG(WIN0_YRGB_MST, 0xffffffff, 0), + .uv_mst = VOP_REG(WIN0_CBR_MST, 0xffffffff, 0), + .yrgb_vir = VOP_REG(WIN0_VIR, 0x3fff, 0), + .uv_vir = VOP_REG(WIN0_VIR, 0x3fff, 16), + .src_alpha_ctl = VOP_REG(WIN0_SRC_ALPHA_CTRL, 0xff, 0), + .dst_alpha_ctl = VOP_REG(WIN0_DST_ALPHA_CTRL, 0xff, 0), +}; + +static const struct vop_win_phy win23_data = { + .data_formats = formats_234, + .nformats = ARRAY_SIZE(formats_234), + .enable = VOP_REG(WIN2_CTRL0, 0x1, 0), + .format = VOP_REG(WIN2_CTRL0, 0x7, 1), + .dsp_info = VOP_REG(WIN2_DSP_INFO0, 0x0fff0fff, 0), + .dsp_st = VOP_REG(WIN2_DSP_ST0, 0x1fff1fff, 0), + .yrgb_mst = VOP_REG(WIN2_MST0, 0xffffffff, 0), + .yrgb_vir = VOP_REG(WIN2_VIR0_1, 0x1fff, 0), + .src_alpha_ctl = VOP_REG(WIN2_SRC_ALPHA_CTRL, 0xff, 0), + .dst_alpha_ctl = VOP_REG(WIN2_DST_ALPHA_CTRL, 0xff, 0), +}; + +static const struct vop_win_phy cursor_data = { + .data_formats = formats_234, + .nformats = ARRAY_SIZE(formats_234), + .enable = VOP_REG(HWC_CTRL0, 0x1, 0), + .format = VOP_REG(HWC_CTRL0, 0x7, 1), + .dsp_st = VOP_REG(HWC_DSP_ST, 0x1fff1fff, 0), + .yrgb_mst = VOP_REG(HWC_MST, 0xffffffff, 0), +}; + +static const struct vop_ctrl ctrl_data = { + .standby = VOP_REG(SYS_CTRL, 0x1, 22), + .gate_en = VOP_REG(SYS_CTRL, 0x1, 23), + .mmu_en = VOP_REG(SYS_CTRL, 0x1, 20), + .rgb_en = VOP_REG(SYS_CTRL, 0x1, 12), + .hdmi_en = VOP_REG(SYS_CTRL, 0x1, 13), + .edp_en = VOP_REG(SYS_CTRL, 0x1, 14), + .mipi_en = VOP_REG(SYS_CTRL, 0x1, 15), + .dither_down = VOP_REG(DSP_CTRL1, 0xf, 1), + .dither_up = VOP_REG(DSP_CTRL1, 0x1, 6), + .data_blank = VOP_REG(DSP_CTRL0, 0x1, 19), + .out_mode = VOP_REG(DSP_CTRL0, 0xf, 0), + .pin_pol = VOP_REG(DSP_CTRL0, 0xf, 4), + .htotal_pw = VOP_REG(DSP_HTOTAL_HS_END, 0x1fff1fff, 0), + .hact_st_end = VOP_REG(DSP_HACT_ST_END, 0x1fff1fff, 0), + .vtotal_pw = VOP_REG(DSP_VTOTAL_VS_END, 0x1fff1fff, 0), + .vact_st_end = VOP_REG(DSP_VACT_ST_END, 0x1fff1fff, 0), + .hpost_st_end = VOP_REG(POST_DSP_HACT_INFO, 0x1fff1fff, 0), + .vpost_st_end = VOP_REG(POST_DSP_VACT_INFO, 0x1fff1fff, 0), +}; + +static const struct vop_reg_data vop_init_reg_table[] = { + {SYS_CTRL, 0x00c00000}, + {DSP_CTRL0, 0x00000000}, + {WIN0_CTRL0, 0x00000080}, + {WIN1_CTRL0, 0x00000080}, +}; + +/* + * Note: rk3288 has a dedicated 'cursor' window, however, that window requires + * special support to get alpha blending working. For now, just use overlay + * window 1 for the drm cursor. + */ +static const struct vop_win_data rk3288_vop_win_data[] = { + { .base = 0x00, .phy = &win01_data, .type = DRM_PLANE_TYPE_PRIMARY }, + { .base = 0x40, .phy = &win01_data, .type = DRM_PLANE_TYPE_CURSOR }, + { .base = 0x00, .phy = &win23_data, .type = DRM_PLANE_TYPE_OVERLAY }, + { .base = 0x50, .phy = &win23_data, .type = DRM_PLANE_TYPE_OVERLAY }, + { .base = 0x00, .phy = &cursor_data, .type = DRM_PLANE_TYPE_OVERLAY }, +}; + +static const struct vop_data rk3288_vop = { + .init_table = vop_init_reg_table, + .table_size = ARRAY_SIZE(vop_init_reg_table), + .ctrl = &ctrl_data, + .win = rk3288_vop_win_data, + .win_size = ARRAY_SIZE(rk3288_vop_win_data), +}; + +static const struct of_device_id vop_driver_dt_match[] = { + { .compatible = "rockchip,rk3288-vop", + .data = &rk3288_vop }, + {}, +}; + +static inline void vop_writel(struct vop *vop, uint32_t offset, uint32_t v) +{ + writel(v, vop->regs + offset); + vop->regsbak[offset >> 2] = v; +} + +static inline uint32_t vop_readl(struct vop *vop, uint32_t offset) +{ + return readl(vop->regs + offset); +} + +static inline uint32_t vop_read_reg(struct vop *vop, uint32_t base, + const struct vop_reg *reg) +{ + return (vop_readl(vop, base + reg->offset) >> reg->shift) & reg->mask; +} + +static inline void vop_cfg_done(struct vop *vop) +{ + writel(0x01, vop->regs + REG_CFG_DONE); +} + +static inline void vop_mask_write(struct vop *vop, uint32_t offset, + uint32_t mask, uint32_t v) +{ + if (mask) { + uint32_t cached_val = vop->regsbak[offset >> 2]; + + cached_val = (cached_val & ~mask) | v; + writel(cached_val, vop->regs + offset); + vop->regsbak[offset >> 2] = cached_val; + } +} + +static inline void vop_mask_write_relaxed(struct vop *vop, uint32_t offset, + uint32_t mask, uint32_t v) +{ + if (mask) { + uint32_t cached_val = vop->regsbak[offset >> 2]; + + cached_val = (cached_val & ~mask) | v; + writel_relaxed(cached_val, vop->regs + offset); + vop->regsbak[offset >> 2] = cached_val; + } +} + +static enum vop_data_format vop_convert_format(uint32_t format) +{ + switch (format) { + case DRM_FORMAT_XRGB8888: + case DRM_FORMAT_ARGB8888: + return VOP_FMT_ARGB8888; + case DRM_FORMAT_RGB888: + return VOP_FMT_RGB888; + case DRM_FORMAT_RGB565: + return VOP_FMT_RGB565; + case DRM_FORMAT_NV12: + return VOP_FMT_YUV420SP; + case DRM_FORMAT_NV16: + return VOP_FMT_YUV422SP; + case DRM_FORMAT_NV24: + return VOP_FMT_YUV444SP; + default: + DRM_ERROR("unsupport format[%08x]\n", format); + return -EINVAL; + } +} + +static bool is_alpha_support(uint32_t format) +{ + switch (format) { + case DRM_FORMAT_ARGB8888: + return true; + default: + return false; + } +} + +static void vop_enable(struct drm_crtc *crtc) +{ + struct vop *vop = to_vop(crtc); + int ret; + + ret = clk_enable(vop->hclk); + if (ret < 0) { + dev_err(vop->dev, "failed to enable hclk - %d\n", ret); + return; + } + + ret = clk_enable(vop->dclk); + if (ret < 0) { + dev_err(vop->dev, "failed to enable dclk - %d\n", ret); + goto err_disable_hclk; + } + + ret = clk_enable(vop->aclk); + if (ret < 0) { + dev_err(vop->dev, "failed to enable aclk - %d\n", ret); + goto err_disable_dclk; + } + + /* + * Slave iommu shares power, irq and clock with vop. It was associated + * automatically with this master device via common driver code. + * Now that we have enabled the clock we attach it to the shared drm + * mapping. + */ + ret = rockchip_drm_dma_attach_device(vop->drm_dev, vop->dev); + if (ret) { + dev_err(vop->dev, "failed to attach dma mapping, %d\n", ret); + goto err_disable_aclk; + } + + spin_lock(&vop->reg_lock); + + VOP_CTRL_SET(vop, standby, 0); + + spin_unlock(&vop->reg_lock); + + enable_irq(vop->irq); + + drm_vblank_on(vop->drm_dev, vop->pipe); + + return; + +err_disable_aclk: + clk_disable(vop->aclk); +err_disable_dclk: + clk_disable(vop->dclk); +err_disable_hclk: + clk_disable(vop->hclk); +} + +static void vop_disable(struct drm_crtc *crtc) +{ + struct vop *vop = to_vop(crtc); + + drm_vblank_off(crtc->dev, vop->pipe); + + disable_irq(vop->irq); + + /* + * TODO: Since standby doesn't take effect until the next vblank, + * when we turn off dclk below, the vop is probably still active. + */ + spin_lock(&vop->reg_lock); + + VOP_CTRL_SET(vop, standby, 1); + + spin_unlock(&vop->reg_lock); + /* + * disable dclk to stop frame scan, so we can safely detach iommu, + */ + clk_disable(vop->dclk); + + rockchip_drm_dma_detach_device(vop->drm_dev, vop->dev); + + clk_disable(vop->aclk); + clk_disable(vop->hclk); +} + +/* + * Caller must hold vsync_mutex. + */ +static struct drm_framebuffer *vop_win_last_pending_fb(struct vop_win *vop_win) +{ + struct vop_win_state *last; + struct vop_win_state *active = vop_win->active; + + if (list_empty(&vop_win->pending)) + return active ? active->fb : NULL; + + last = list_last_entry(&vop_win->pending, struct vop_win_state, head); + return last ? last->fb : NULL; +} + +/* + * Caller must hold vsync_mutex. + */ +static int vop_win_queue_fb(struct vop_win *vop_win, + struct drm_framebuffer *fb, dma_addr_t yrgb_mst, + struct drm_pending_vblank_event *event) +{ + struct vop_win_state *state; + + state = kzalloc(sizeof(*state), GFP_KERNEL); + if (!state) + return -ENOMEM; + + state->fb = fb; + state->yrgb_mst = yrgb_mst; + state->event = event; + + list_add_tail(&state->head, &vop_win->pending); + + return 0; +} + +static int vop_update_plane_event(struct drm_plane *plane, + struct drm_crtc *crtc, + struct drm_framebuffer *fb, int crtc_x, + int crtc_y, unsigned int crtc_w, + unsigned int crtc_h, uint32_t src_x, + uint32_t src_y, uint32_t src_w, + uint32_t src_h, + struct drm_pending_vblank_event *event) +{ + struct vop_win *vop_win = to_vop_win(plane); + const struct vop_win_data *win = vop_win->data; + struct vop *vop = to_vop(crtc); + struct drm_gem_object *obj; + struct rockchip_gem_object *rk_obj; + unsigned long offset; + unsigned int actual_w; + unsigned int actual_h; + unsigned int dsp_stx; + unsigned int dsp_sty; + unsigned int y_vir_stride; + dma_addr_t yrgb_mst; + enum vop_data_format format; + uint32_t val; + bool is_alpha; + bool visible; + int ret; + struct drm_rect dest = { + .x1 = crtc_x, + .y1 = crtc_y, + .x2 = crtc_x + crtc_w, + .y2 = crtc_y + crtc_h, + }; + struct drm_rect src = { + /* 16.16 fixed point */ + .x1 = src_x, + .y1 = src_y, + .x2 = src_x + src_w, + .y2 = src_y + src_h, + }; + const struct drm_rect clip = { + .x2 = crtc->mode.hdisplay, + .y2 = crtc->mode.vdisplay, + }; + bool can_position = plane->type != DRM_PLANE_TYPE_PRIMARY; + + ret = drm_plane_helper_check_update(plane, crtc, fb, + &src, &dest, &clip, + DRM_PLANE_HELPER_NO_SCALING, + DRM_PLANE_HELPER_NO_SCALING, + can_position, false, &visible); + if (ret) + return ret; + + if (!visible) + return 0; + + is_alpha = is_alpha_support(fb->pixel_format); + format = vop_convert_format(fb->pixel_format); + if (format < 0) + return format; + + obj = rockchip_fb_get_gem_obj(fb, 0); + if (!obj) { + DRM_ERROR("fail to get rockchip gem object from framebuffer\n"); + return -EINVAL; + } + + rk_obj = to_rockchip_obj(obj); + + actual_w = (src.x2 - src.x1) >> 16; + actual_h = (src.y2 - src.y1) >> 16; + crtc_x = max(0, crtc_x); + crtc_y = max(0, crtc_y); + + dsp_stx = crtc_x + crtc->mode.htotal - crtc->mode.hsync_start; + dsp_sty = crtc_y + crtc->mode.vtotal - crtc->mode.vsync_start; + + offset = (src.x1 >> 16) * (fb->bits_per_pixel >> 3); + offset += (src.y1 >> 16) * fb->pitches[0]; + yrgb_mst = rk_obj->dma_addr + offset; + + y_vir_stride = fb->pitches[0] / (fb->bits_per_pixel >> 3); + + /* + * If this plane update changes the plane's framebuffer, (or more + * precisely, if this update has a different framebuffer than the last + * update), enqueue it so we can track when it completes. + * + * Only when we discover that this update has completed, can we + * unreference any previous framebuffers. + */ + mutex_lock(&vop->vsync_mutex); + if (fb != vop_win_last_pending_fb(vop_win)) { + ret = drm_vblank_get(plane->dev, vop->pipe); + if (ret) { + DRM_ERROR("failed to get vblank, %d\n", ret); + mutex_unlock(&vop->vsync_mutex); + return ret; + } + + drm_framebuffer_reference(fb); + + ret = vop_win_queue_fb(vop_win, fb, yrgb_mst, event); + if (ret) { + drm_vblank_put(plane->dev, vop->pipe); + mutex_unlock(&vop->vsync_mutex); + return ret; + } + + vop->vsync_work_pending = true; + } + mutex_unlock(&vop->vsync_mutex); + + spin_lock(&vop->reg_lock); + + VOP_WIN_SET(vop, win, format, format); + VOP_WIN_SET(vop, win, yrgb_vir, y_vir_stride); + VOP_WIN_SET(vop, win, yrgb_mst, yrgb_mst); + val = (actual_h - 1) << 16; + val |= (actual_w - 1) & 0xffff; + VOP_WIN_SET(vop, win, act_info, val); + VOP_WIN_SET(vop, win, dsp_info, val); + val = (dsp_sty - 1) << 16; + val |= (dsp_stx - 1) & 0xffff; + VOP_WIN_SET(vop, win, dsp_st, val); + + if (is_alpha) { + VOP_WIN_SET(vop, win, dst_alpha_ctl, + DST_FACTOR_M0(ALPHA_SRC_INVERSE)); + val = SRC_ALPHA_EN(1) | SRC_COLOR_M0(ALPHA_SRC_PRE_MUL) | + SRC_ALPHA_M0(ALPHA_STRAIGHT) | + SRC_BLEND_M0(ALPHA_PER_PIX) | + SRC_ALPHA_CAL_M0(ALPHA_NO_SATURATION) | + SRC_FACTOR_M0(ALPHA_ONE); + VOP_WIN_SET(vop, win, src_alpha_ctl, val); + } else { + VOP_WIN_SET(vop, win, src_alpha_ctl, SRC_ALPHA_EN(0)); + } + + VOP_WIN_SET(vop, win, enable, 1); + + vop_cfg_done(vop); + spin_unlock(&vop->reg_lock); + + return 0; +} + +static int vop_update_plane(struct drm_plane *plane, struct drm_crtc *crtc, + struct drm_framebuffer *fb, int crtc_x, int crtc_y, + unsigned int crtc_w, unsigned int crtc_h, + uint32_t src_x, uint32_t src_y, uint32_t src_w, + uint32_t src_h) +{ + return vop_update_plane_event(plane, crtc, fb, crtc_x, crtc_y, crtc_w, + crtc_h, src_x, src_y, src_w, src_h, + NULL); +} + +static int vop_update_primary_plane(struct drm_crtc *crtc, + struct drm_pending_vblank_event *event) +{ + unsigned int crtc_w, crtc_h; + + crtc_w = crtc->primary->fb->width - crtc->x; + crtc_h = crtc->primary->fb->height - crtc->y; + + return vop_update_plane_event(crtc->primary, crtc, crtc->primary->fb, + 0, 0, crtc_w, crtc_h, crtc->x << 16, + crtc->y << 16, crtc_w << 16, + crtc_h << 16, event); +} + +static int vop_disable_plane(struct drm_plane *plane) +{ + struct vop_win *vop_win = to_vop_win(plane); + const struct vop_win_data *win = vop_win->data; + struct vop *vop; + int ret; + + if (!plane->crtc) + return 0; + + vop = to_vop(plane->crtc); + + ret = drm_vblank_get(plane->dev, vop->pipe); + if (ret) { + DRM_ERROR("failed to get vblank, %d\n", ret); + return ret; + } + + mutex_lock(&vop->vsync_mutex); + + ret = vop_win_queue_fb(vop_win, NULL, 0, NULL); + if (ret) { + drm_vblank_put(plane->dev, vop->pipe); + mutex_unlock(&vop->vsync_mutex); + return ret; + } + + vop->vsync_work_pending = true; + mutex_unlock(&vop->vsync_mutex); + + spin_lock(&vop->reg_lock); + VOP_WIN_SET(vop, win, enable, 0); + vop_cfg_done(vop); + spin_unlock(&vop->reg_lock); + + return 0; +} + +static void vop_plane_destroy(struct drm_plane *plane) +{ + vop_disable_plane(plane); + drm_plane_cleanup(plane); +} + +static const struct drm_plane_funcs vop_plane_funcs = { + .update_plane = vop_update_plane, + .disable_plane = vop_disable_plane, + .destroy = vop_plane_destroy, +}; + +int rockchip_drm_crtc_mode_config(struct drm_crtc *crtc, + int connector_type, + int out_mode) +{ + struct vop *vop = to_vop(crtc); + + vop->connector_type = connector_type; + vop->connector_out_mode = out_mode; + + return 0; +} + +static int vop_crtc_enable_vblank(struct drm_crtc *crtc) +{ + struct vop *vop = to_vop(crtc); + unsigned long flags; + + if (vop->dpms != DRM_MODE_DPMS_ON) + return -EPERM; + + spin_lock_irqsave(&vop->irq_lock, flags); + + vop_mask_write(vop, INTR_CTRL0, FS_INTR_MASK, FS_INTR_EN(1)); + + spin_unlock_irqrestore(&vop->irq_lock, flags); + + return 0; +} + +static void vop_crtc_disable_vblank(struct drm_crtc *crtc) +{ + struct vop *vop = to_vop(crtc); + unsigned long flags; + + if (vop->dpms != DRM_MODE_DPMS_ON) + return; + spin_lock_irqsave(&vop->irq_lock, flags); + vop_mask_write(vop, INTR_CTRL0, FS_INTR_MASK, FS_INTR_EN(0)); + spin_unlock_irqrestore(&vop->irq_lock, flags); +} + +static const struct rockchip_crtc_funcs private_crtc_funcs = { + .enable_vblank = vop_crtc_enable_vblank, + .disable_vblank = vop_crtc_disable_vblank, +}; + +static void vop_crtc_dpms(struct drm_crtc *crtc, int mode) +{ + struct vop *vop = to_vop(crtc); + + DRM_DEBUG_KMS("crtc[%d] mode[%d]\n", crtc->base.id, mode); + + if (vop->dpms == mode) { + DRM_DEBUG_KMS("desired dpms mode is same as previous one.\n"); + return; + } + + switch (mode) { + case DRM_MODE_DPMS_ON: + vop_enable(crtc); + break; + case DRM_MODE_DPMS_STANDBY: + case DRM_MODE_DPMS_SUSPEND: + case DRM_MODE_DPMS_OFF: + vop_disable(crtc); + break; + default: + DRM_DEBUG_KMS("unspecified mode %d\n", mode); + break; + } + + vop->dpms = mode; +} + +static void vop_crtc_prepare(struct drm_crtc *crtc) +{ + vop_crtc_dpms(crtc, DRM_MODE_DPMS_ON); +} + +static bool vop_crtc_mode_fixup(struct drm_crtc *crtc, + const struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + if (adjusted_mode->htotal == 0 || adjusted_mode->vtotal == 0) + return false; + + return true; +} + +static int vop_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y, + struct drm_framebuffer *old_fb) +{ + int ret; + + crtc->x = x; + crtc->y = y; + + ret = vop_update_primary_plane(crtc, NULL); + if (ret < 0) { + DRM_ERROR("fail to update plane\n"); + return ret; + } + + return 0; +} + +static int vop_crtc_mode_set(struct drm_crtc *crtc, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode, + int x, int y, struct drm_framebuffer *fb) +{ + struct vop *vop = to_vop(crtc); + u16 hsync_len = adjusted_mode->hsync_end - adjusted_mode->hsync_start; + u16 hdisplay = adjusted_mode->hdisplay; + u16 htotal = adjusted_mode->htotal; + u16 hact_st = adjusted_mode->htotal - adjusted_mode->hsync_start; + u16 hact_end = hact_st + hdisplay; + u16 vdisplay = adjusted_mode->vdisplay; + u16 vtotal = adjusted_mode->vtotal; + u16 vsync_len = adjusted_mode->vsync_end - adjusted_mode->vsync_start; + u16 vact_st = adjusted_mode->vtotal - adjusted_mode->vsync_start; + u16 vact_end = vact_st + vdisplay; + int ret; + uint32_t val; + + /* + * disable dclk to stop frame scan, so that we can safe config mode and + * enable iommu. + */ + clk_disable(vop->dclk); + + switch (vop->connector_type) { + case DRM_MODE_CONNECTOR_LVDS: + VOP_CTRL_SET(vop, rgb_en, 1); + break; + case DRM_MODE_CONNECTOR_eDP: + VOP_CTRL_SET(vop, edp_en, 1); + break; + case DRM_MODE_CONNECTOR_HDMIA: + VOP_CTRL_SET(vop, hdmi_en, 1); + break; + default: + DRM_ERROR("unsupport connector_type[%d]\n", + vop->connector_type); + return -EINVAL; + }; + VOP_CTRL_SET(vop, out_mode, vop->connector_out_mode); + + val = 0x8; + val |= (adjusted_mode->flags & DRM_MODE_FLAG_NHSYNC) ? 1 : 0; + val |= (adjusted_mode->flags & DRM_MODE_FLAG_NVSYNC) ? (1 << 1) : 0; + VOP_CTRL_SET(vop, pin_pol, val); + + VOP_CTRL_SET(vop, htotal_pw, (htotal << 16) | hsync_len); + val = hact_st << 16; + val |= hact_end; + VOP_CTRL_SET(vop, hact_st_end, val); + VOP_CTRL_SET(vop, hpost_st_end, val); + + VOP_CTRL_SET(vop, vtotal_pw, (vtotal << 16) | vsync_len); + val = vact_st << 16; + val |= vact_end; + VOP_CTRL_SET(vop, vact_st_end, val); + VOP_CTRL_SET(vop, vpost_st_end, val); + + ret = vop_crtc_mode_set_base(crtc, x, y, fb); + if (ret) + return ret; + + /* + * reset dclk, take all mode config affect, so the clk would run in + * correct frame. + */ + reset_control_assert(vop->dclk_rst); + usleep_range(10, 20); + reset_control_deassert(vop->dclk_rst); + + clk_set_rate(vop->dclk, adjusted_mode->clock * 1000); + ret = clk_enable(vop->dclk); + if (ret < 0) { + dev_err(vop->dev, "failed to enable dclk - %d\n", ret); + return ret; + } + + return 0; +} + +static void vop_crtc_commit(struct drm_crtc *crtc) +{ +} + +static const struct drm_crtc_helper_funcs vop_crtc_helper_funcs = { + .dpms = vop_crtc_dpms, + .prepare = vop_crtc_prepare, + .mode_fixup = vop_crtc_mode_fixup, + .mode_set = vop_crtc_mode_set, + .mode_set_base = vop_crtc_mode_set_base, + .commit = vop_crtc_commit, +}; + +static int vop_crtc_page_flip(struct drm_crtc *crtc, + struct drm_framebuffer *fb, + struct drm_pending_vblank_event *event, + uint32_t page_flip_flags) +{ + struct vop *vop = to_vop(crtc); + struct drm_framebuffer *old_fb = crtc->primary->fb; + int ret; + + /* when the page flip is requested, crtc's dpms should be on */ + if (vop->dpms > DRM_MODE_DPMS_ON) { + DRM_DEBUG("failed page flip request at dpms[%d].\n", vop->dpms); + return 0; + } + + crtc->primary->fb = fb; + + ret = vop_update_primary_plane(crtc, event); + if (ret) + crtc->primary->fb = old_fb; + + return ret; +} + +static void vop_win_state_complete(struct vop_win *vop_win, + struct vop_win_state *state) +{ + struct vop *vop = vop_win->vop; + struct drm_crtc *crtc = &vop->crtc; + struct drm_device *drm = crtc->dev; + unsigned long flags; + + if (state->event) { + spin_lock_irqsave(&drm->event_lock, flags); + drm_send_vblank_event(drm, -1, state->event); + spin_unlock_irqrestore(&drm->event_lock, flags); + } + + list_del(&state->head); + drm_vblank_put(crtc->dev, vop->pipe); +} + +static void vop_crtc_destroy(struct drm_crtc *crtc) +{ + drm_crtc_cleanup(crtc); +} + +static const struct drm_crtc_funcs vop_crtc_funcs = { + .set_config = drm_crtc_helper_set_config, + .page_flip = vop_crtc_page_flip, + .destroy = vop_crtc_destroy, +}; + +static bool vop_win_state_is_active(struct vop_win *vop_win, + struct vop_win_state *state) +{ + bool active = false; + + if (state->fb) { + dma_addr_t yrgb_mst; + + /* check yrgb_mst to tell if pending_fb is now front */ + yrgb_mst = VOP_WIN_GET_YRGBADDR(vop_win->vop, vop_win->data); + + active = (yrgb_mst == state->yrgb_mst); + } else { + bool enabled; + + /* if enable bit is clear, plane is now disabled */ + enabled = VOP_WIN_GET(vop_win->vop, vop_win->data, enable); + + active = (enabled == 0); + } + + return active; +} + +static void vop_win_state_destroy(struct vop_win_state *state) +{ + struct drm_framebuffer *fb = state->fb; + + if (fb) + drm_framebuffer_unreference(fb); + + kfree(state); +} + +static void vop_win_update_state(struct vop_win *vop_win) +{ + struct vop_win_state *state, *n, *new_active = NULL; + + /* Check if any pending states are now active */ + list_for_each_entry(state, &vop_win->pending, head) + if (vop_win_state_is_active(vop_win, state)) { + new_active = state; + break; + } + + if (!new_active) + return; + + /* + * Destroy any 'skipped' pending states - states that were queued + * before the newly active state. + */ + list_for_each_entry_safe(state, n, &vop_win->pending, head) { + if (state == new_active) + break; + vop_win_state_complete(vop_win, state); + vop_win_state_destroy(state); + } + + vop_win_state_complete(vop_win, new_active); + + if (vop_win->active) + vop_win_state_destroy(vop_win->active); + vop_win->active = new_active; +} + +static bool vop_win_has_pending_state(struct vop_win *vop_win) +{ + return !list_empty(&vop_win->pending); +} + +static irqreturn_t vop_isr_thread(int irq, void *data) +{ + struct vop *vop = data; + const struct vop_data *vop_data = vop->data; + unsigned int i; + + mutex_lock(&vop->vsync_mutex); + + if (!vop->vsync_work_pending) + goto done; + + vop->vsync_work_pending = false; + + for (i = 0; i < vop_data->win_size; i++) { + struct vop_win *vop_win = &vop->win[i]; + + vop_win_update_state(vop_win); + if (vop_win_has_pending_state(vop_win)) + vop->vsync_work_pending = true; + } + +done: + mutex_unlock(&vop->vsync_mutex); + + return IRQ_HANDLED; +} + +static irqreturn_t vop_isr(int irq, void *data) +{ + struct vop *vop = data; + uint32_t intr0_reg, active_irqs; + unsigned long flags; + + /* + * INTR_CTRL0 register has interrupt status, enable and clear bits, we + * must hold irq_lock to avoid a race with enable/disable_vblank(). + */ + spin_lock_irqsave(&vop->irq_lock, flags); + intr0_reg = vop_readl(vop, INTR_CTRL0); + active_irqs = intr0_reg & INTR_MASK; + /* Clear all active interrupt sources */ + if (active_irqs) + vop_writel(vop, INTR_CTRL0, + intr0_reg | (active_irqs << INTR_CLR_SHIFT)); + spin_unlock_irqrestore(&vop->irq_lock, flags); + + /* This is expected for vop iommu irqs, since the irq is shared */ + if (!active_irqs) + return IRQ_NONE; + + /* Only Frame Start Interrupt is enabled; other irqs are spurious. */ + if (!(active_irqs & FS_INTR)) { + DRM_ERROR("Unknown VOP IRQs: %#02x\n", active_irqs); + return IRQ_NONE; + } + + drm_handle_vblank(vop->drm_dev, vop->pipe); + + return (vop->vsync_work_pending) ? IRQ_WAKE_THREAD : IRQ_HANDLED; +} + +static int vop_create_crtc(struct vop *vop) +{ + const struct vop_data *vop_data = vop->data; + struct device *dev = vop->dev; + struct drm_device *drm_dev = vop->drm_dev; + struct drm_plane *primary = NULL, *cursor = NULL, *plane; + struct drm_crtc *crtc = &vop->crtc; + struct device_node *port; + int ret; + int i; + + /* + * Create drm_plane for primary and cursor planes first, since we need + * to pass them to drm_crtc_init_with_planes, which sets the + * "possible_crtcs" to the newly initialized crtc. + */ + for (i = 0; i < vop_data->win_size; i++) { + struct vop_win *vop_win = &vop->win[i]; + const struct vop_win_data *win_data = vop_win->data; + + if (win_data->type != DRM_PLANE_TYPE_PRIMARY && + win_data->type != DRM_PLANE_TYPE_CURSOR) + continue; + + ret = drm_universal_plane_init(vop->drm_dev, &vop_win->base, + 0, &vop_plane_funcs, + win_data->phy->data_formats, + win_data->phy->nformats, + win_data->type); + if (ret) { + DRM_ERROR("failed to initialize plane\n"); + goto err_cleanup_planes; + } + + plane = &vop_win->base; + if (plane->type == DRM_PLANE_TYPE_PRIMARY) + primary = plane; + else if (plane->type == DRM_PLANE_TYPE_CURSOR) + cursor = plane; + } + + ret = drm_crtc_init_with_planes(drm_dev, crtc, primary, cursor, + &vop_crtc_funcs); + if (ret) + return ret; + + drm_crtc_helper_add(crtc, &vop_crtc_helper_funcs); + + /* + * Create drm_planes for overlay windows with possible_crtcs restricted + * to the newly created crtc. + */ + for (i = 0; i < vop_data->win_size; i++) { + struct vop_win *vop_win = &vop->win[i]; + const struct vop_win_data *win_data = vop_win->data; + unsigned long possible_crtcs = 1 << drm_crtc_index(crtc); + + if (win_data->type != DRM_PLANE_TYPE_OVERLAY) + continue; + + ret = drm_universal_plane_init(vop->drm_dev, &vop_win->base, + possible_crtcs, + &vop_plane_funcs, + win_data->phy->data_formats, + win_data->phy->nformats, + win_data->type); + if (ret) { + DRM_ERROR("failed to initialize overlay plane\n"); + goto err_cleanup_crtc; + } + } + + port = of_get_child_by_name(dev->of_node, "port"); + if (!port) { + DRM_ERROR("no port node found in %s\n", + dev->of_node->full_name); + goto err_cleanup_crtc; + } + + crtc->port = port; + vop->pipe = drm_crtc_index(crtc); + rockchip_register_crtc_funcs(drm_dev, &private_crtc_funcs, vop->pipe); + + return 0; + +err_cleanup_crtc: + drm_crtc_cleanup(crtc); +err_cleanup_planes: + list_for_each_entry(plane, &drm_dev->mode_config.plane_list, head) + drm_plane_cleanup(plane); + return ret; +} + +static void vop_destroy_crtc(struct vop *vop) +{ + struct drm_crtc *crtc = &vop->crtc; + + rockchip_unregister_crtc_funcs(vop->drm_dev, vop->pipe); + of_node_put(crtc->port); + drm_crtc_cleanup(crtc); +} + +static int vop_initial(struct vop *vop) +{ + const struct vop_data *vop_data = vop->data; + const struct vop_reg_data *init_table = vop_data->init_table; + struct reset_control *ahb_rst; + int i, ret; + + vop->hclk = devm_clk_get(vop->dev, "hclk_vop"); + if (IS_ERR(vop->hclk)) { + dev_err(vop->dev, "failed to get hclk source\n"); + return PTR_ERR(vop->hclk); + } + vop->aclk = devm_clk_get(vop->dev, "aclk_vop"); + if (IS_ERR(vop->aclk)) { + dev_err(vop->dev, "failed to get aclk source\n"); + return PTR_ERR(vop->aclk); + } + vop->dclk = devm_clk_get(vop->dev, "dclk_vop"); + if (IS_ERR(vop->dclk)) { + dev_err(vop->dev, "failed to get dclk source\n"); + return PTR_ERR(vop->dclk); + } + + ret = clk_prepare(vop->hclk); + if (ret < 0) { + dev_err(vop->dev, "failed to prepare hclk\n"); + return ret; + } + + ret = clk_prepare(vop->dclk); + if (ret < 0) { + dev_err(vop->dev, "failed to prepare dclk\n"); + goto err_unprepare_hclk; + } + + ret = clk_prepare(vop->aclk); + if (ret < 0) { + dev_err(vop->dev, "failed to prepare aclk\n"); + goto err_unprepare_dclk; + } + + /* + * enable hclk, so that we can config vop register. + */ + ret = clk_enable(vop->hclk); + if (ret < 0) { + dev_err(vop->dev, "failed to prepare aclk\n"); + goto err_unprepare_aclk; + } + /* + * do hclk_reset, reset all vop registers. + */ + ahb_rst = devm_reset_control_get(vop->dev, "ahb"); + if (IS_ERR(ahb_rst)) { + dev_err(vop->dev, "failed to get ahb reset\n"); + ret = PTR_ERR(ahb_rst); + goto err_disable_hclk; + } + reset_control_assert(ahb_rst); + usleep_range(10, 20); + reset_control_deassert(ahb_rst); + + memcpy(vop->regsbak, vop->regs, vop->len); + + for (i = 0; i < vop_data->table_size; i++) + vop_writel(vop, init_table[i].offset, init_table[i].value); + + for (i = 0; i < vop_data->win_size; i++) { + const struct vop_win_data *win = &vop_data->win[i]; + + VOP_WIN_SET(vop, win, enable, 0); + } + + vop_cfg_done(vop); + + /* + * do dclk_reset, let all config take affect. + */ + vop->dclk_rst = devm_reset_control_get(vop->dev, "dclk"); + if (IS_ERR(vop->dclk_rst)) { + dev_err(vop->dev, "failed to get dclk reset\n"); + ret = PTR_ERR(vop->dclk_rst); + goto err_unprepare_aclk; + } + reset_control_assert(vop->dclk_rst); + usleep_range(10, 20); + reset_control_deassert(vop->dclk_rst); + + clk_disable(vop->hclk); + + vop->dpms = DRM_MODE_DPMS_OFF; + + return 0; + +err_disable_hclk: + clk_disable(vop->hclk); +err_unprepare_aclk: + clk_unprepare(vop->aclk); +err_unprepare_dclk: + clk_unprepare(vop->dclk); +err_unprepare_hclk: + clk_unprepare(vop->hclk); + return ret; +} + +/* + * Initialize the vop->win array elements. + */ +static void vop_win_init(struct vop *vop) +{ + const struct vop_data *vop_data = vop->data; + unsigned int i; + + for (i = 0; i < vop_data->win_size; i++) { + struct vop_win *vop_win = &vop->win[i]; + const struct vop_win_data *win_data = &vop_data->win[i]; + + vop_win->data = win_data; + vop_win->vop = vop; + INIT_LIST_HEAD(&vop_win->pending); + } +} + +static int vop_bind(struct device *dev, struct device *master, void *data) +{ + struct platform_device *pdev = to_platform_device(dev); + const struct of_device_id *of_id; + const struct vop_data *vop_data; + struct drm_device *drm_dev = data; + struct vop *vop; + struct resource *res; + size_t alloc_size; + int ret; + + of_id = of_match_device(vop_driver_dt_match, dev); + vop_data = of_id->data; + if (!vop_data) + return -ENODEV; + + /* Allocate vop struct and its vop_win array */ + alloc_size = sizeof(*vop) + sizeof(*vop->win) * vop_data->win_size; + vop = devm_kzalloc(dev, alloc_size, GFP_KERNEL); + if (!vop) + return -ENOMEM; + + vop->dev = dev; + vop->data = vop_data; + vop->drm_dev = drm_dev; + dev_set_drvdata(dev, vop); + + vop_win_init(vop); + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + vop->len = resource_size(res); + vop->regs = devm_ioremap_resource(dev, res); + if (IS_ERR(vop->regs)) + return PTR_ERR(vop->regs); + + vop->regsbak = devm_kzalloc(dev, vop->len, GFP_KERNEL); + if (!vop->regsbak) + return -ENOMEM; + + ret = vop_initial(vop); + if (ret < 0) { + dev_err(&pdev->dev, "cannot initial vop dev - err %d\n", ret); + return ret; + } + + vop->irq = platform_get_irq(pdev, 0); + if (vop->irq < 0) { + dev_err(dev, "cannot find irq for vop\n"); + return vop->irq; + } + + spin_lock_init(&vop->reg_lock); + spin_lock_init(&vop->irq_lock); + + mutex_init(&vop->vsync_mutex); + + ret = devm_request_threaded_irq(dev, vop->irq, vop_isr, vop_isr_thread, + IRQF_SHARED, dev_name(dev), vop); + if (ret) + return ret; + + /* IRQ is initially disabled; it gets enabled in power_on */ + disable_irq(vop->irq); + + ret = vop_create_crtc(vop); + if (ret) + return ret; + + pm_runtime_enable(&pdev->dev); + return 0; +} + +static void vop_unbind(struct device *dev, struct device *master, void *data) +{ + struct vop *vop = dev_get_drvdata(dev); + + pm_runtime_disable(dev); + vop_destroy_crtc(vop); +} + +static const struct component_ops vop_component_ops = { + .bind = vop_bind, + .unbind = vop_unbind, +}; + +static int vop_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + + if (!dev->of_node) { + dev_err(dev, "can't find vop devices\n"); + return -ENODEV; + } + + return component_add(dev, &vop_component_ops); +} + +static int vop_remove(struct platform_device *pdev) +{ + component_del(&pdev->dev, &vop_component_ops); + + return 0; +} + +struct platform_driver vop_platform_driver = { + .probe = vop_probe, + .remove = vop_remove, + .driver = { + .name = "rockchip-vop", + .owner = THIS_MODULE, + .of_match_table = of_match_ptr(vop_driver_dt_match), + }, +}; + +module_platform_driver(vop_platform_driver); + +MODULE_AUTHOR("Mark Yao <mark.yao@rock-chips.com>"); +MODULE_DESCRIPTION("ROCKCHIP VOP Driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.h b/drivers/gpu/drm/rockchip/rockchip_drm_vop.h new file mode 100644 index 000000000000..63e9b3a084c5 --- /dev/null +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.h @@ -0,0 +1,201 @@ +/* + * Copyright (C) Fuzhou Rockchip Electronics Co.Ltd + * Author:Mark Yao <mark.yao@rock-chips.com> + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#ifndef _ROCKCHIP_DRM_VOP_H +#define _ROCKCHIP_DRM_VOP_H + +/* register definition */ +#define REG_CFG_DONE 0x0000 +#define VERSION_INFO 0x0004 +#define SYS_CTRL 0x0008 +#define SYS_CTRL1 0x000c +#define DSP_CTRL0 0x0010 +#define DSP_CTRL1 0x0014 +#define DSP_BG 0x0018 +#define MCU_CTRL 0x001c +#define INTR_CTRL0 0x0020 +#define INTR_CTRL1 0x0024 +#define WIN0_CTRL0 0x0030 +#define WIN0_CTRL1 0x0034 +#define WIN0_COLOR_KEY 0x0038 +#define WIN0_VIR 0x003c +#define WIN0_YRGB_MST 0x0040 +#define WIN0_CBR_MST 0x0044 +#define WIN0_ACT_INFO 0x0048 +#define WIN0_DSP_INFO 0x004c +#define WIN0_DSP_ST 0x0050 +#define WIN0_SCL_FACTOR_YRGB 0x0054 +#define WIN0_SCL_FACTOR_CBR 0x0058 +#define WIN0_SCL_OFFSET 0x005c +#define WIN0_SRC_ALPHA_CTRL 0x0060 +#define WIN0_DST_ALPHA_CTRL 0x0064 +#define WIN0_FADING_CTRL 0x0068 +/* win1 register */ +#define WIN1_CTRL0 0x0070 +#define WIN1_CTRL1 0x0074 +#define WIN1_COLOR_KEY 0x0078 +#define WIN1_VIR 0x007c +#define WIN1_YRGB_MST 0x0080 +#define WIN1_CBR_MST 0x0084 +#define WIN1_ACT_INFO 0x0088 +#define WIN1_DSP_INFO 0x008c +#define WIN1_DSP_ST 0x0090 +#define WIN1_SCL_FACTOR_YRGB 0x0094 +#define WIN1_SCL_FACTOR_CBR 0x0098 +#define WIN1_SCL_OFFSET 0x009c +#define WIN1_SRC_ALPHA_CTRL 0x00a0 +#define WIN1_DST_ALPHA_CTRL 0x00a4 +#define WIN1_FADING_CTRL 0x00a8 +/* win2 register */ +#define WIN2_CTRL0 0x00b0 +#define WIN2_CTRL1 0x00b4 +#define WIN2_VIR0_1 0x00b8 +#define WIN2_VIR2_3 0x00bc +#define WIN2_MST0 0x00c0 +#define WIN2_DSP_INFO0 0x00c4 +#define WIN2_DSP_ST0 0x00c8 +#define WIN2_COLOR_KEY 0x00cc +#define WIN2_MST1 0x00d0 +#define WIN2_DSP_INFO1 0x00d4 +#define WIN2_DSP_ST1 0x00d8 +#define WIN2_SRC_ALPHA_CTRL 0x00dc +#define WIN2_MST2 0x00e0 +#define WIN2_DSP_INFO2 0x00e4 +#define WIN2_DSP_ST2 0x00e8 +#define WIN2_DST_ALPHA_CTRL 0x00ec +#define WIN2_MST3 0x00f0 +#define WIN2_DSP_INFO3 0x00f4 +#define WIN2_DSP_ST3 0x00f8 +#define WIN2_FADING_CTRL 0x00fc +/* win3 register */ +#define WIN3_CTRL0 0x0100 +#define WIN3_CTRL1 0x0104 +#define WIN3_VIR0_1 0x0108 +#define WIN3_VIR2_3 0x010c +#define WIN3_MST0 0x0110 +#define WIN3_DSP_INFO0 0x0114 +#define WIN3_DSP_ST0 0x0118 +#define WIN3_COLOR_KEY 0x011c +#define WIN3_MST1 0x0120 +#define WIN3_DSP_INFO1 0x0124 +#define WIN3_DSP_ST1 0x0128 +#define WIN3_SRC_ALPHA_CTRL 0x012c +#define WIN3_MST2 0x0130 +#define WIN3_DSP_INFO2 0x0134 +#define WIN3_DSP_ST2 0x0138 +#define WIN3_DST_ALPHA_CTRL 0x013c +#define WIN3_MST3 0x0140 +#define WIN3_DSP_INFO3 0x0144 +#define WIN3_DSP_ST3 0x0148 +#define WIN3_FADING_CTRL 0x014c +/* hwc register */ +#define HWC_CTRL0 0x0150 +#define HWC_CTRL1 0x0154 +#define HWC_MST 0x0158 +#define HWC_DSP_ST 0x015c +#define HWC_SRC_ALPHA_CTRL 0x0160 +#define HWC_DST_ALPHA_CTRL 0x0164 +#define HWC_FADING_CTRL 0x0168 +/* post process register */ +#define POST_DSP_HACT_INFO 0x0170 +#define POST_DSP_VACT_INFO 0x0174 +#define POST_SCL_FACTOR_YRGB 0x0178 +#define POST_SCL_CTRL 0x0180 +#define POST_DSP_VACT_INFO_F1 0x0184 +#define DSP_HTOTAL_HS_END 0x0188 +#define DSP_HACT_ST_END 0x018c +#define DSP_VTOTAL_VS_END 0x0190 +#define DSP_VACT_ST_END 0x0194 +#define DSP_VS_ST_END_F1 0x0198 +#define DSP_VACT_ST_END_F1 0x019c +/* register definition end */ + +/* interrupt define */ +#define DSP_HOLD_VALID_INTR (1 << 0) +#define FS_INTR (1 << 1) +#define LINE_FLAG_INTR (1 << 2) +#define BUS_ERROR_INTR (1 << 3) + +#define INTR_MASK (DSP_HOLD_VALID_INTR | FS_INTR | \ + LINE_FLAG_INTR | BUS_ERROR_INTR) + +#define DSP_HOLD_VALID_INTR_EN(x) ((x) << 4) +#define FS_INTR_EN(x) ((x) << 5) +#define LINE_FLAG_INTR_EN(x) ((x) << 6) +#define BUS_ERROR_INTR_EN(x) ((x) << 7) +#define DSP_HOLD_VALID_INTR_MASK (1 << 4) +#define FS_INTR_MASK (1 << 5) +#define LINE_FLAG_INTR_MASK (1 << 6) +#define BUS_ERROR_INTR_MASK (1 << 7) + +#define INTR_CLR_SHIFT 8 +#define DSP_HOLD_VALID_INTR_CLR (1 << (INTR_CLR_SHIFT + 0)) +#define FS_INTR_CLR (1 << (INTR_CLR_SHIFT + 1)) +#define LINE_FLAG_INTR_CLR (1 << (INTR_CLR_SHIFT + 2)) +#define BUS_ERROR_INTR_CLR (1 << (INTR_CLR_SHIFT + 3)) + +#define DSP_LINE_NUM(x) (((x) & 0x1fff) << 12) +#define DSP_LINE_NUM_MASK (0x1fff << 12) + +/* src alpha ctrl define */ +#define SRC_FADING_VALUE(x) (((x) & 0xff) << 24) +#define SRC_GLOBAL_ALPHA(x) (((x) & 0xff) << 16) +#define SRC_FACTOR_M0(x) (((x) & 0x7) << 6) +#define SRC_ALPHA_CAL_M0(x) (((x) & 0x1) << 5) +#define SRC_BLEND_M0(x) (((x) & 0x3) << 3) +#define SRC_ALPHA_M0(x) (((x) & 0x1) << 2) +#define SRC_COLOR_M0(x) (((x) & 0x1) << 1) +#define SRC_ALPHA_EN(x) (((x) & 0x1) << 0) +/* dst alpha ctrl define */ +#define DST_FACTOR_M0(x) (((x) & 0x7) << 6) + +/* + * display output interface supported by rockchip lcdc + */ +#define ROCKCHIP_OUT_MODE_P888 0 +#define ROCKCHIP_OUT_MODE_P666 1 +#define ROCKCHIP_OUT_MODE_P565 2 +/* for use special outface */ +#define ROCKCHIP_OUT_MODE_AAAA 15 + +enum alpha_mode { + ALPHA_STRAIGHT, + ALPHA_INVERSE, +}; + +enum global_blend_mode { + ALPHA_GLOBAL, + ALPHA_PER_PIX, + ALPHA_PER_PIX_GLOBAL, +}; + +enum alpha_cal_mode { + ALPHA_SATURATION, + ALPHA_NO_SATURATION, +}; + +enum color_mode { + ALPHA_SRC_PRE_MUL, + ALPHA_SRC_NO_PRE_MUL, +}; + +enum factor_mode { + ALPHA_ZERO, + ALPHA_ONE, + ALPHA_SRC, + ALPHA_SRC_INVERSE, + ALPHA_SRC_GLOBAL, +}; + +#endif /* _ROCKCHIP_DRM_VOP_H */ diff --git a/drivers/gpu/drm/shmobile/shmob_drm_drv.c b/drivers/gpu/drm/shmobile/shmob_drm_drv.c index e62cbde81e50..666321de7b99 100644 --- a/drivers/gpu/drm/shmobile/shmob_drm_drv.c +++ b/drivers/gpu/drm/shmobile/shmob_drm_drv.c @@ -348,7 +348,6 @@ static struct platform_driver shmob_drm_platform_driver = { .probe = shmob_drm_probe, .remove = shmob_drm_remove, .driver = { - .owner = THIS_MODULE, .name = "shmob-drm", .pm = &shmob_drm_pm_ops, }, diff --git a/drivers/gpu/drm/sti/Kconfig b/drivers/gpu/drm/sti/Kconfig index ae8850f3e63b..d6d6b705b8c1 100644 --- a/drivers/gpu/drm/sti/Kconfig +++ b/drivers/gpu/drm/sti/Kconfig @@ -5,6 +5,7 @@ config DRM_STI select DRM_KMS_HELPER select DRM_GEM_CMA_HELPER select DRM_KMS_CMA_HELPER + select FW_LOADER_USER_HELPER_FALLBACK help Choose this option to enable DRM on STM stiH41x chipset diff --git a/drivers/gpu/drm/sti/Makefile b/drivers/gpu/drm/sti/Makefile index 04ac2ceef27f..6ba9d27c1b90 100644 --- a/drivers/gpu/drm/sti/Makefile +++ b/drivers/gpu/drm/sti/Makefile @@ -3,6 +3,7 @@ sticompositor-y := \ sti_mixer.o \ sti_gdp.o \ sti_vid.o \ + sti_cursor.o \ sti_compositor.o \ sti_drm_crtc.o \ sti_drm_plane.o @@ -18,4 +19,5 @@ obj-$(CONFIG_DRM_STI) = \ sti_hda.o \ sti_tvout.o \ sticompositor.o \ - sti_drm_drv.o
\ No newline at end of file + sti_hqvdp.o \ + sti_drm_drv.o diff --git a/drivers/gpu/drm/sti/sti_compositor.c b/drivers/gpu/drm/sti/sti_compositor.c index 390d93e9a06c..43215d3020fb 100644 --- a/drivers/gpu/drm/sti/sti_compositor.c +++ b/drivers/gpu/drm/sti/sti_compositor.c @@ -24,14 +24,16 @@ * stiH407 compositor properties */ struct sti_compositor_data stih407_compositor_data = { - .nb_subdev = 6, + .nb_subdev = 8, .subdev_desc = { + {STI_CURSOR_SUBDEV, (int)STI_CURSOR, 0x000}, {STI_GPD_SUBDEV, (int)STI_GDP_0, 0x100}, {STI_GPD_SUBDEV, (int)STI_GDP_1, 0x200}, {STI_GPD_SUBDEV, (int)STI_GDP_2, 0x300}, {STI_GPD_SUBDEV, (int)STI_GDP_3, 0x400}, {STI_VID_SUBDEV, (int)STI_VID_0, 0x700}, - {STI_MIXER_MAIN_SUBDEV, STI_MIXER_MAIN, 0xC00} + {STI_MIXER_MAIN_SUBDEV, STI_MIXER_MAIN, 0xC00}, + {STI_MIXER_AUX_SUBDEV, STI_MIXER_AUX, 0xD00}, }, }; @@ -67,11 +69,11 @@ static int sti_compositor_init_subdev(struct sti_compositor *compo, break; case STI_GPD_SUBDEV: case STI_VID_SUBDEV: + case STI_CURSOR_SUBDEV: compo->layer[layer_id++] = sti_layer_create(compo->dev, desc[i].id, compo->regs + desc[i].offset); break; - /* case STI_CURSOR_SUBDEV : TODO */ default: DRM_ERROR("Unknow subdev compoment type\n"); return 1; @@ -102,33 +104,35 @@ static int sti_compositor_bind(struct device *dev, struct device *master, enum sti_layer_type type = desc & STI_LAYER_TYPE_MASK; enum drm_plane_type plane_type = DRM_PLANE_TYPE_OVERLAY; - if (compo->mixer[crtc]) + if (crtc < compo->nb_mixers) plane_type = DRM_PLANE_TYPE_PRIMARY; switch (type) { case STI_CUR: cursor = sti_drm_plane_init(drm_dev, compo->layer[i], - (1 << crtc) - 1, - DRM_PLANE_TYPE_CURSOR); + 1, DRM_PLANE_TYPE_CURSOR); break; case STI_GDP: case STI_VID: primary = sti_drm_plane_init(drm_dev, compo->layer[i], - (1 << crtc) - 1, plane_type); + (1 << compo->nb_mixers) - 1, + plane_type); plane++; break; case STI_BCK: + case STI_VDP: break; } /* The first planes are reserved for primary planes*/ - if (compo->mixer[crtc]) { + if (crtc < compo->nb_mixers && primary) { sti_drm_crtc_init(drm_dev, compo->mixer[crtc], primary, cursor); crtc++; cursor = NULL; + primary = NULL; } } } @@ -267,7 +271,6 @@ static int sti_compositor_remove(struct platform_device *pdev) static struct platform_driver sti_compositor_driver = { .driver = { .name = "sti-compositor", - .owner = THIS_MODULE, .of_match_table = compositor_of_match, }, .probe = sti_compositor_probe, diff --git a/drivers/gpu/drm/sti/sti_compositor.h b/drivers/gpu/drm/sti/sti_compositor.h index 3ea19db72e0f..019eb44c62cc 100644 --- a/drivers/gpu/drm/sti/sti_compositor.h +++ b/drivers/gpu/drm/sti/sti_compositor.h @@ -64,7 +64,6 @@ struct sti_compositor_data { * @layer: array of layers * @nb_mixers: number of mixers for this compositor * @nb_layers: number of layers (GDP,VID,...) for this compositor - * @enable: true if compositor is enable else false * @vtg_vblank_nb: callback for VTG VSYNC notification */ struct sti_compositor { @@ -83,7 +82,6 @@ struct sti_compositor { struct sti_layer *layer[STI_MAX_LAYER]; int nb_mixers; int nb_layers; - bool enable; struct notifier_block vtg_vblank_nb; }; diff --git a/drivers/gpu/drm/sti/sti_cursor.c b/drivers/gpu/drm/sti/sti_cursor.c new file mode 100644 index 000000000000..010eaee60bf7 --- /dev/null +++ b/drivers/gpu/drm/sti/sti_cursor.c @@ -0,0 +1,242 @@ +/* + * Copyright (C) STMicroelectronics SA 2014 + * Authors: Vincent Abriou <vincent.abriou@st.com> + * Fabien Dessenne <fabien.dessenne@st.com> + * for STMicroelectronics. + * License terms: GNU General Public License (GPL), version 2 + */ +#include <drm/drmP.h> + +#include "sti_cursor.h" +#include "sti_layer.h" +#include "sti_vtg.h" + +/* Registers */ +#define CUR_CTL 0x00 +#define CUR_VPO 0x0C +#define CUR_PML 0x14 +#define CUR_PMP 0x18 +#define CUR_SIZE 0x1C +#define CUR_CML 0x20 +#define CUR_AWS 0x28 +#define CUR_AWE 0x2C + +#define CUR_CTL_CLUT_UPDATE BIT(1) + +#define STI_CURS_MIN_SIZE 1 +#define STI_CURS_MAX_SIZE 128 + +/* + * pixmap dma buffer stucture + * + * @paddr: physical address + * @size: buffer size + * @base: virtual address + */ +struct dma_pixmap { + dma_addr_t paddr; + size_t size; + void *base; +}; + +/** + * STI Cursor structure + * + * @layer: layer structure + * @width: cursor width + * @height: cursor height + * @clut: color look up table + * @clut_paddr: color look up table physical address + * @pixmap: pixmap dma buffer (clut8-format cursor) + */ +struct sti_cursor { + struct sti_layer layer; + unsigned int width; + unsigned int height; + unsigned short *clut; + dma_addr_t clut_paddr; + struct dma_pixmap pixmap; +}; + +static const uint32_t cursor_supported_formats[] = { + DRM_FORMAT_ARGB8888, +}; + +#define to_sti_cursor(x) container_of(x, struct sti_cursor, layer) + +static const uint32_t *sti_cursor_get_formats(struct sti_layer *layer) +{ + return cursor_supported_formats; +} + +static unsigned int sti_cursor_get_nb_formats(struct sti_layer *layer) +{ + return ARRAY_SIZE(cursor_supported_formats); +} + +static void sti_cursor_argb8888_to_clut8(struct sti_layer *layer) +{ + struct sti_cursor *cursor = to_sti_cursor(layer); + u32 *src = layer->vaddr; + u8 *dst = cursor->pixmap.base; + unsigned int i, j; + u32 a, r, g, b; + + for (i = 0; i < cursor->height; i++) { + for (j = 0; j < cursor->width; j++) { + /* Pick the 2 higher bits of each component */ + a = (*src >> 30) & 3; + r = (*src >> 22) & 3; + g = (*src >> 14) & 3; + b = (*src >> 6) & 3; + *dst = a << 6 | r << 4 | g << 2 | b; + src++; + dst++; + } + } +} + +static int sti_cursor_prepare_layer(struct sti_layer *layer, bool first_prepare) +{ + struct sti_cursor *cursor = to_sti_cursor(layer); + struct drm_display_mode *mode = layer->mode; + u32 y, x; + u32 val; + + DRM_DEBUG_DRIVER("\n"); + + dev_dbg(layer->dev, "%s %s\n", __func__, sti_layer_to_str(layer)); + + if (layer->src_w < STI_CURS_MIN_SIZE || + layer->src_h < STI_CURS_MIN_SIZE || + layer->src_w > STI_CURS_MAX_SIZE || + layer->src_h > STI_CURS_MAX_SIZE) { + DRM_ERROR("Invalid cursor size (%dx%d)\n", + layer->src_w, layer->src_h); + return -EINVAL; + } + + /* If the cursor size has changed, re-allocated the pixmap */ + if (!cursor->pixmap.base || + (cursor->width != layer->src_w) || + (cursor->height != layer->src_h)) { + cursor->width = layer->src_w; + cursor->height = layer->src_h; + + if (cursor->pixmap.base) + dma_free_writecombine(layer->dev, + cursor->pixmap.size, + cursor->pixmap.base, + cursor->pixmap.paddr); + + cursor->pixmap.size = cursor->width * cursor->height; + + cursor->pixmap.base = dma_alloc_writecombine(layer->dev, + cursor->pixmap.size, + &cursor->pixmap.paddr, + GFP_KERNEL | GFP_DMA); + if (!cursor->pixmap.base) { + DRM_ERROR("Failed to allocate memory for pixmap\n"); + return -ENOMEM; + } + } + + /* Convert ARGB8888 to CLUT8 */ + sti_cursor_argb8888_to_clut8(layer); + + /* AWS and AWE depend on the mode */ + y = sti_vtg_get_line_number(*mode, 0); + x = sti_vtg_get_pixel_number(*mode, 0); + val = y << 16 | x; + writel(val, layer->regs + CUR_AWS); + y = sti_vtg_get_line_number(*mode, mode->vdisplay - 1); + x = sti_vtg_get_pixel_number(*mode, mode->hdisplay - 1); + val = y << 16 | x; + writel(val, layer->regs + CUR_AWE); + + if (first_prepare) { + /* Set and fetch CLUT */ + writel(cursor->clut_paddr, layer->regs + CUR_CML); + writel(CUR_CTL_CLUT_UPDATE, layer->regs + CUR_CTL); + } + + return 0; +} + +static int sti_cursor_commit_layer(struct sti_layer *layer) +{ + struct sti_cursor *cursor = to_sti_cursor(layer); + struct drm_display_mode *mode = layer->mode; + u32 ydo, xdo; + + dev_dbg(layer->dev, "%s %s\n", __func__, sti_layer_to_str(layer)); + + /* Set memory location, size, and position */ + writel(cursor->pixmap.paddr, layer->regs + CUR_PML); + writel(cursor->width, layer->regs + CUR_PMP); + writel(cursor->height << 16 | cursor->width, layer->regs + CUR_SIZE); + + ydo = sti_vtg_get_line_number(*mode, layer->dst_y); + xdo = sti_vtg_get_pixel_number(*mode, layer->dst_y); + writel((ydo << 16) | xdo, layer->regs + CUR_VPO); + + return 0; +} + +static int sti_cursor_disable_layer(struct sti_layer *layer) +{ + return 0; +} + +static void sti_cursor_init(struct sti_layer *layer) +{ + struct sti_cursor *cursor = to_sti_cursor(layer); + unsigned short *base = cursor->clut; + unsigned int a, r, g, b; + + /* Assign CLUT values, ARGB444 format */ + for (a = 0; a < 4; a++) + for (r = 0; r < 4; r++) + for (g = 0; g < 4; g++) + for (b = 0; b < 4; b++) + *base++ = (a * 5) << 12 | + (r * 5) << 8 | + (g * 5) << 4 | + (b * 5); +} + +static const struct sti_layer_funcs cursor_ops = { + .get_formats = sti_cursor_get_formats, + .get_nb_formats = sti_cursor_get_nb_formats, + .init = sti_cursor_init, + .prepare = sti_cursor_prepare_layer, + .commit = sti_cursor_commit_layer, + .disable = sti_cursor_disable_layer, +}; + +struct sti_layer *sti_cursor_create(struct device *dev) +{ + struct sti_cursor *cursor; + + cursor = devm_kzalloc(dev, sizeof(*cursor), GFP_KERNEL); + if (!cursor) { + DRM_ERROR("Failed to allocate memory for cursor\n"); + return NULL; + } + + /* Allocate clut buffer */ + cursor->clut = dma_alloc_writecombine(dev, + 0x100 * sizeof(unsigned short), + &cursor->clut_paddr, + GFP_KERNEL | GFP_DMA); + + if (!cursor->clut) { + DRM_ERROR("Failed to allocate memory for cursor clut\n"); + devm_kfree(dev, cursor); + return NULL; + } + + cursor->layer.ops = &cursor_ops; + + return (struct sti_layer *)cursor; +} diff --git a/drivers/gpu/drm/sti/sti_cursor.h b/drivers/gpu/drm/sti/sti_cursor.h new file mode 100644 index 000000000000..3c9827404f27 --- /dev/null +++ b/drivers/gpu/drm/sti/sti_cursor.h @@ -0,0 +1,12 @@ +/* + * Copyright (C) STMicroelectronics SA 2013 + * Authors: Vincent Abriou <vincent.abriou@st.com> for STMicroelectronics. + * License terms: GNU General Public License (GPL), version 2 + */ + +#ifndef _STI_CURSOR_H_ +#define _STI_CURSOR_H_ + +struct sti_layer *sti_cursor_create(struct device *dev); + +#endif diff --git a/drivers/gpu/drm/sti/sti_drm_crtc.c b/drivers/gpu/drm/sti/sti_drm_crtc.c index 36a1ad3c4823..4c651c200f20 100644 --- a/drivers/gpu/drm/sti/sti_drm_crtc.c +++ b/drivers/gpu/drm/sti/sti_drm_crtc.c @@ -28,7 +28,7 @@ static void sti_drm_crtc_prepare(struct drm_crtc *crtc) struct device *dev = mixer->dev; struct sti_compositor *compo = dev_get_drvdata(dev); - compo->enable = true; + mixer->enabled = true; /* Prepare and enable the compo IP clock */ if (mixer->id == STI_MIXER_MAIN) { @@ -38,6 +38,8 @@ static void sti_drm_crtc_prepare(struct drm_crtc *crtc) if (clk_prepare_enable(compo->clk_compo_aux)) DRM_INFO("Failed to prepare/enable compo_aux clk\n"); } + + sti_mixer_clear_all_layers(mixer); } static void sti_drm_crtc_commit(struct drm_crtc *crtc) @@ -62,6 +64,8 @@ static void sti_drm_crtc_commit(struct drm_crtc *crtc) /* Enable layer on mixer */ if (sti_mixer_set_layer_status(mixer, layer, true)) DRM_ERROR("Can not enable layer at mixer\n"); + + drm_crtc_vblank_on(crtc); } static bool sti_drm_crtc_mode_fixup(struct drm_crtc *crtc, @@ -144,7 +148,8 @@ sti_drm_crtc_mode_set(struct drm_crtc *crtc, struct drm_display_mode *mode, w = crtc->primary->fb->width - x; h = crtc->primary->fb->height - y; - return sti_layer_prepare(layer, crtc->primary->fb, &crtc->mode, + return sti_layer_prepare(layer, crtc, + crtc->primary->fb, &crtc->mode, mixer->id, 0, 0, w, h, x, y, w, h); } @@ -171,7 +176,8 @@ static int sti_drm_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y, w = crtc->primary->fb->width - crtc->x; h = crtc->primary->fb->height - crtc->y; - ret = sti_layer_prepare(layer, crtc->primary->fb, &crtc->mode, + ret = sti_layer_prepare(layer, crtc, + crtc->primary->fb, &crtc->mode, mixer->id, 0, 0, w, h, crtc->x, crtc->y, w, h); if (ret) { @@ -196,7 +202,7 @@ static void sti_drm_crtc_disable(struct drm_crtc *crtc) struct sti_compositor *compo = dev_get_drvdata(dev); struct sti_layer *layer; - if (!compo->enable) + if (!mixer->enabled) return; DRM_DEBUG_KMS("CRTC:%d (%s)\n", crtc->base.id, sti_mixer_to_str(mixer)); @@ -222,7 +228,7 @@ static void sti_drm_crtc_disable(struct drm_crtc *crtc) /* Then disable layer itself */ sti_layer_disable(layer); - drm_vblank_off(crtc->dev, mixer->id); + drm_crtc_vblank_off(crtc); /* Disable pixel clock and compo IP clocks */ if (mixer->id == STI_MIXER_MAIN) { @@ -233,7 +239,7 @@ static void sti_drm_crtc_disable(struct drm_crtc *crtc) clk_disable_unprepare(compo->clk_compo_aux); } - compo->enable = false; + mixer->enabled = false; } static struct drm_crtc_helper_funcs sti_crtc_helper_funcs = { @@ -364,7 +370,6 @@ void sti_drm_crtc_disable_vblank(struct drm_device *dev, int crtc) struct sti_drm_private *priv = dev->dev_private; struct sti_compositor *compo = priv->compo; struct notifier_block *vtg_vblank_nb = &compo->vtg_vblank_nb; - unsigned long flags; DRM_DEBUG_DRIVER("\n"); @@ -373,13 +378,10 @@ void sti_drm_crtc_disable_vblank(struct drm_device *dev, int crtc) DRM_DEBUG_DRIVER("Warning: cannot unregister VTG notifier\n"); /* free the resources of the pending requests */ - spin_lock_irqsave(&dev->event_lock, flags); if (compo->mixer[crtc]->pending_event) { drm_vblank_put(dev, crtc); compo->mixer[crtc]->pending_event = NULL; } - spin_unlock_irqrestore(&dev->event_lock, flags); - } EXPORT_SYMBOL(sti_drm_crtc_disable_vblank); @@ -399,6 +401,7 @@ bool sti_drm_crtc_is_main(struct drm_crtc *crtc) return false; } +EXPORT_SYMBOL(sti_drm_crtc_is_main); int sti_drm_crtc_init(struct drm_device *drm_dev, struct sti_mixer *mixer, struct drm_plane *primary, struct drm_plane *cursor) diff --git a/drivers/gpu/drm/sti/sti_drm_drv.c b/drivers/gpu/drm/sti/sti_drm_drv.c index 223d93c3a05d..5239fa121726 100644 --- a/drivers/gpu/drm/sti/sti_drm_drv.c +++ b/drivers/gpu/drm/sti/sti_drm_drv.c @@ -67,8 +67,12 @@ static int sti_drm_load(struct drm_device *dev, unsigned long flags) sti_drm_mode_config_init(dev); ret = component_bind_all(dev->dev, dev); - if (ret) + if (ret) { + drm_kms_helper_poll_fini(dev); + drm_mode_config_cleanup(dev); + kfree(private); return ret; + } drm_helper_disable_unused_functions(dev); @@ -184,7 +188,6 @@ static struct platform_driver sti_drm_master_driver = { .probe = sti_drm_master_probe, .remove = sti_drm_master_remove, .driver = { - .owner = THIS_MODULE, .name = DRIVER_NAME "__master", }, }; @@ -228,7 +231,6 @@ static struct platform_driver sti_drm_platform_driver = { .probe = sti_drm_platform_probe, .remove = sti_drm_platform_remove, .driver = { - .owner = THIS_MODULE, .name = DRIVER_NAME, .of_match_table = sti_drm_dt_ids, }, diff --git a/drivers/gpu/drm/sti/sti_drm_plane.c b/drivers/gpu/drm/sti/sti_drm_plane.c index f4118d4cac22..bb6a29339e10 100644 --- a/drivers/gpu/drm/sti/sti_drm_plane.c +++ b/drivers/gpu/drm/sti/sti_drm_plane.c @@ -45,7 +45,8 @@ sti_drm_update_plane(struct drm_plane *plane, struct drm_crtc *crtc, } /* src_x are in 16.16 format. */ - res = sti_layer_prepare(layer, fb, &crtc->mode, mixer->id, + res = sti_layer_prepare(layer, crtc, fb, + &crtc->mode, mixer->id, crtc_x, crtc_y, crtc_w, crtc_h, src_x >> 16, src_y >> 16, src_w >> 16, src_h >> 16); @@ -193,3 +194,4 @@ struct drm_plane *sti_drm_plane_init(struct drm_device *dev, return &layer->plane; } +EXPORT_SYMBOL(sti_drm_plane_init); diff --git a/drivers/gpu/drm/sti/sti_gdp.c b/drivers/gpu/drm/sti/sti_gdp.c index 4e30b74559f5..32448d1d1e8f 100644 --- a/drivers/gpu/drm/sti/sti_gdp.c +++ b/drivers/gpu/drm/sti/sti_gdp.c @@ -73,7 +73,9 @@ struct sti_gdp_node { struct sti_gdp_node_list { struct sti_gdp_node *top_field; + dma_addr_t top_field_paddr; struct sti_gdp_node *btm_field; + dma_addr_t btm_field_paddr; }; /** @@ -81,6 +83,8 @@ struct sti_gdp_node_list { * * @layer: layer structure * @clk_pix: pixel clock for the current gdp + * @clk_main_parent: gdp parent clock if main path used + * @clk_aux_parent: gdp parent clock if aux path used * @vtg_field_nb: callback for VTG FIELD (top or bottom) notification * @is_curr_top: true if the current node processed is the top field * @node_list: array of node list @@ -88,6 +92,8 @@ struct sti_gdp_node_list { struct sti_gdp { struct sti_layer layer; struct clk *clk_pix; + struct clk *clk_main_parent; + struct clk *clk_aux_parent; struct notifier_block vtg_field_nb; bool is_curr_top; struct sti_gdp_node_list node_list[GDP_NODE_NB_BANK]; @@ -168,7 +174,6 @@ static int sti_gdp_get_alpharange(int format) static struct sti_gdp_node_list *sti_gdp_get_free_nodes(struct sti_layer *layer) { int hw_nvn; - void *virt_nvn; struct sti_gdp *gdp = to_sti_gdp(layer); unsigned int i; @@ -176,11 +181,9 @@ static struct sti_gdp_node_list *sti_gdp_get_free_nodes(struct sti_layer *layer) if (!hw_nvn) goto end; - virt_nvn = dma_to_virt(layer->dev, (dma_addr_t) hw_nvn); - for (i = 0; i < GDP_NODE_NB_BANK; i++) - if ((virt_nvn != gdp->node_list[i].btm_field) && - (virt_nvn != gdp->node_list[i].top_field)) + if ((hw_nvn != gdp->node_list[i].btm_field_paddr) && + (hw_nvn != gdp->node_list[i].top_field_paddr)) return &gdp->node_list[i]; /* in hazardious cases restart with the first node */ @@ -204,7 +207,6 @@ static struct sti_gdp_node_list *sti_gdp_get_current_nodes(struct sti_layer *layer) { int hw_nvn; - void *virt_nvn; struct sti_gdp *gdp = to_sti_gdp(layer); unsigned int i; @@ -212,11 +214,9 @@ struct sti_gdp_node_list *sti_gdp_get_current_nodes(struct sti_layer *layer) if (!hw_nvn) goto end; - virt_nvn = dma_to_virt(layer->dev, (dma_addr_t) hw_nvn); - for (i = 0; i < GDP_NODE_NB_BANK; i++) - if ((virt_nvn == gdp->node_list[i].btm_field) || - (virt_nvn == gdp->node_list[i].top_field)) + if ((hw_nvn == gdp->node_list[i].btm_field_paddr) || + (hw_nvn == gdp->node_list[i].top_field_paddr)) return &gdp->node_list[i]; end: @@ -292,8 +292,8 @@ static int sti_gdp_prepare_layer(struct sti_layer *layer, bool first_prepare) /* Same content and chained together */ memcpy(btm_field, top_field, sizeof(*btm_field)); - top_field->gam_gdp_nvn = virt_to_dma(dev, btm_field); - btm_field->gam_gdp_nvn = virt_to_dma(dev, top_field); + top_field->gam_gdp_nvn = list->btm_field_paddr; + btm_field->gam_gdp_nvn = list->top_field_paddr; /* Interlaced mode */ if (layer->mode->flags & DRM_MODE_FLAG_INTERLACE) @@ -311,6 +311,17 @@ static int sti_gdp_prepare_layer(struct sti_layer *layer, bool first_prepare) /* Set and enable gdp clock */ if (gdp->clk_pix) { + struct clk *clkp; + /* According to the mixer used, the gdp pixel clock + * should have a different parent clock. */ + if (layer->mixer_id == STI_MIXER_MAIN) + clkp = gdp->clk_main_parent; + else + clkp = gdp->clk_aux_parent; + + if (clkp) + clk_set_parent(gdp->clk_pix, clkp); + res = clk_set_rate(gdp->clk_pix, rate); if (res < 0) { DRM_ERROR("Cannot set rate (%dHz) for gdp\n", @@ -349,8 +360,8 @@ static int sti_gdp_commit_layer(struct sti_layer *layer) struct sti_gdp_node *updated_top_node = updated_list->top_field; struct sti_gdp_node *updated_btm_node = updated_list->btm_field; struct sti_gdp *gdp = to_sti_gdp(layer); - u32 dma_updated_top = virt_to_dma(layer->dev, updated_top_node); - u32 dma_updated_btm = virt_to_dma(layer->dev, updated_btm_node); + u32 dma_updated_top = updated_list->top_field_paddr; + u32 dma_updated_btm = updated_list->btm_field_paddr; struct sti_gdp_node_list *curr_list = sti_gdp_get_current_nodes(layer); dev_dbg(layer->dev, "%s %s top/btm_node:0x%p/0x%p\n", __func__, @@ -461,16 +472,16 @@ static void sti_gdp_init(struct sti_layer *layer) { struct sti_gdp *gdp = to_sti_gdp(layer); struct device_node *np = layer->dev->of_node; - dma_addr_t dma; + dma_addr_t dma_addr; void *base; unsigned int i, size; /* Allocate all the nodes within a single memory page */ size = sizeof(struct sti_gdp_node) * GDP_NODE_PER_FIELD * GDP_NODE_NB_BANK; - base = dma_alloc_writecombine(layer->dev, - size, &dma, GFP_KERNEL | GFP_DMA); + size, &dma_addr, GFP_KERNEL | GFP_DMA); + if (!base) { DRM_ERROR("Failed to allocate memory for GDP node\n"); return; @@ -478,21 +489,26 @@ static void sti_gdp_init(struct sti_layer *layer) memset(base, 0, size); for (i = 0; i < GDP_NODE_NB_BANK; i++) { - if (virt_to_dma(layer->dev, base) & 0xF) { + if (dma_addr & 0xF) { DRM_ERROR("Mem alignment failed\n"); return; } gdp->node_list[i].top_field = base; + gdp->node_list[i].top_field_paddr = dma_addr; + DRM_DEBUG_DRIVER("node[%d].top_field=%p\n", i, base); base += sizeof(struct sti_gdp_node); + dma_addr += sizeof(struct sti_gdp_node); - if (virt_to_dma(layer->dev, base) & 0xF) { + if (dma_addr & 0xF) { DRM_ERROR("Mem alignment failed\n"); return; } gdp->node_list[i].btm_field = base; + gdp->node_list[i].btm_field_paddr = dma_addr; DRM_DEBUG_DRIVER("node[%d].btm_field=%p\n", i, base); base += sizeof(struct sti_gdp_node); + dma_addr += sizeof(struct sti_gdp_node); } if (of_device_is_compatible(np, "st,stih407-compositor")) { @@ -520,6 +536,14 @@ static void sti_gdp_init(struct sti_layer *layer) gdp->clk_pix = devm_clk_get(layer->dev, clk_name); if (IS_ERR(gdp->clk_pix)) DRM_ERROR("Cannot get %s clock\n", clk_name); + + gdp->clk_main_parent = devm_clk_get(layer->dev, "main_parent"); + if (IS_ERR(gdp->clk_main_parent)) + DRM_ERROR("Cannot get main_parent clock\n"); + + gdp->clk_aux_parent = devm_clk_get(layer->dev, "aux_parent"); + if (IS_ERR(gdp->clk_aux_parent)) + DRM_ERROR("Cannot get aux_parent clock\n"); } } diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c index b22968c08d1f..d032e024b0b8 100644 --- a/drivers/gpu/drm/sti/sti_hdmi.c +++ b/drivers/gpu/drm/sti/sti_hdmi.c @@ -130,8 +130,7 @@ static irqreturn_t hdmi_irq_thread(int irq, void *arg) /* Hot plug/unplug IRQ */ if (hdmi->irq_status & HDMI_INT_HOT_PLUG) { - /* read gpio to get the status */ - hdmi->hpd = gpio_get_value(hdmi->hpd_gpio); + hdmi->hpd = readl(hdmi->regs + HDMI_STA) & HDMI_STA_HOT_PLUG; if (hdmi->drm_dev) drm_helper_hpd_irq_event(hdmi->drm_dev); } @@ -273,31 +272,32 @@ static int hdmi_avi_infoframe_config(struct sti_hdmi *hdmi) hdmi_write(hdmi, val, HDMI_SW_DI_CFG); /* Infoframe header */ - val = buffer[0x0]; - val |= buffer[0x1] << 8; - val |= buffer[0x2] << 16; + val = buffer[0]; + val |= buffer[1] << 8; + val |= buffer[2] << 16; hdmi_write(hdmi, val, HDMI_SW_DI_N_HEAD_WORD(HDMI_IFRAME_SLOT_AVI)); /* Infoframe packet bytes */ - val = frame[0x0]; - val |= frame[0x1] << 8; - val |= frame[0x2] << 16; - val |= frame[0x3] << 24; + val = buffer[3]; + val |= *(frame++) << 8; + val |= *(frame++) << 16; + val |= *(frame++) << 24; hdmi_write(hdmi, val, HDMI_SW_DI_N_PKT_WORD0(HDMI_IFRAME_SLOT_AVI)); - val = frame[0x4]; - val |= frame[0x5] << 8; - val |= frame[0x6] << 16; - val |= frame[0x7] << 24; + val = *(frame++); + val |= *(frame++) << 8; + val |= *(frame++) << 16; + val |= *(frame++) << 24; hdmi_write(hdmi, val, HDMI_SW_DI_N_PKT_WORD1(HDMI_IFRAME_SLOT_AVI)); - val = frame[0x8]; - val |= frame[0x9] << 8; - val |= frame[0xA] << 16; - val |= frame[0xB] << 24; + val = *(frame++); + val |= *(frame++) << 8; + val |= *(frame++) << 16; + val |= *(frame++) << 24; hdmi_write(hdmi, val, HDMI_SW_DI_N_PKT_WORD2(HDMI_IFRAME_SLOT_AVI)); - val = frame[0xC]; + val = *(frame++); + val |= *(frame) << 8; hdmi_write(hdmi, val, HDMI_SW_DI_N_PKT_WORD3(HDMI_IFRAME_SLOT_AVI)); /* Enable transmission slot for AVI infoframe @@ -480,17 +480,15 @@ static const struct drm_bridge_funcs sti_hdmi_bridge_funcs = { static int sti_hdmi_connector_get_modes(struct drm_connector *connector) { - struct i2c_adapter *i2c_adap; + struct sti_hdmi_connector *hdmi_connector + = to_sti_hdmi_connector(connector); + struct sti_hdmi *hdmi = hdmi_connector->hdmi; struct edid *edid; int count; DRM_DEBUG_DRIVER("\n"); - i2c_adap = i2c_get_adapter(1); - if (!i2c_adap) - goto fail; - - edid = drm_get_edid(connector, i2c_adap); + edid = drm_get_edid(connector, hdmi->ddc_adapt); if (!edid) goto fail; @@ -603,29 +601,38 @@ static int sti_hdmi_bind(struct device *dev, struct device *master, void *data) struct sti_hdmi_connector *connector; struct drm_connector *drm_connector; struct drm_bridge *bridge; - struct i2c_adapter *i2c_adap; + struct device_node *ddc; int err; - i2c_adap = i2c_get_adapter(1); - if (!i2c_adap) - return -EPROBE_DEFER; + ddc = of_parse_phandle(dev->of_node, "ddc", 0); + if (ddc) { + hdmi->ddc_adapt = of_find_i2c_adapter_by_node(ddc); + if (!hdmi->ddc_adapt) { + err = -EPROBE_DEFER; + of_node_put(ddc); + return err; + } + + of_node_put(ddc); + } /* Set the drm device handle */ hdmi->drm_dev = drm_dev; encoder = sti_hdmi_find_encoder(drm_dev); if (!encoder) - return -ENOMEM; + goto err_adapt; connector = devm_kzalloc(dev, sizeof(*connector), GFP_KERNEL); if (!connector) - return -ENOMEM; + goto err_adapt; + connector->hdmi = hdmi; bridge = devm_kzalloc(dev, sizeof(*bridge), GFP_KERNEL); if (!bridge) - return -ENOMEM; + goto err_adapt; bridge->driver_private = hdmi; drm_bridge_init(drm_dev, bridge, &sti_hdmi_bridge_funcs); @@ -662,6 +669,8 @@ err_sysfs: err_connector: drm_bridge_cleanup(bridge); drm_connector_cleanup(drm_connector); +err_adapt: + put_device(&hdmi->ddc_adapt->dev); return -EINVAL; } @@ -757,13 +766,7 @@ static int sti_hdmi_probe(struct platform_device *pdev) return PTR_ERR(hdmi->clk_audio); } - hdmi->hpd_gpio = of_get_named_gpio(np, "hdmi,hpd-gpio", 0); - if (hdmi->hpd_gpio < 0) { - DRM_ERROR("Failed to get hdmi hpd-gpio\n"); - return -EIO; - } - - hdmi->hpd = gpio_get_value(hdmi->hpd_gpio); + hdmi->hpd = readl(hdmi->regs + HDMI_STA) & HDMI_STA_HOT_PLUG; init_waitqueue_head(&hdmi->wait_event); @@ -788,6 +791,11 @@ static int sti_hdmi_probe(struct platform_device *pdev) static int sti_hdmi_remove(struct platform_device *pdev) { + struct sti_hdmi *hdmi = dev_get_drvdata(&pdev->dev); + + if (hdmi->ddc_adapt) + put_device(&hdmi->ddc_adapt->dev); + component_del(&pdev->dev, &sti_hdmi_ops); return 0; } diff --git a/drivers/gpu/drm/sti/sti_hdmi.h b/drivers/gpu/drm/sti/sti_hdmi.h index 61bec6557ceb..3d22390e1f3b 100644 --- a/drivers/gpu/drm/sti/sti_hdmi.h +++ b/drivers/gpu/drm/sti/sti_hdmi.h @@ -14,6 +14,9 @@ #define HDMI_STA 0x0010 #define HDMI_STA_DLL_LCK BIT(5) +#define HDMI_STA_HOT_PLUG_SHIFT 4 +#define HDMI_STA_HOT_PLUG (1 << HDMI_STA_HOT_PLUG_SHIFT) + struct sti_hdmi; struct hdmi_phy_ops { @@ -37,7 +40,6 @@ struct hdmi_phy_ops { * @irq_status: interrupt status register * @phy_ops: phy start/stop operations * @enabled: true if hdmi is enabled else false - * @hpd_gpio: hdmi hot plug detect gpio number * @hpd: hot plug detect status * @wait_event: wait event * @event_received: wait event status @@ -57,11 +59,11 @@ struct sti_hdmi { u32 irq_status; struct hdmi_phy_ops *phy_ops; bool enabled; - int hpd_gpio; bool hpd; wait_queue_head_t wait_event; bool event_received; struct reset_control *reset; + struct i2c_adapter *ddc_adapt; }; u32 hdmi_read(struct sti_hdmi *hdmi, int offset); diff --git a/drivers/gpu/drm/sti/sti_hqvdp.c b/drivers/gpu/drm/sti/sti_hqvdp.c new file mode 100644 index 000000000000..f3db05dab0ab --- /dev/null +++ b/drivers/gpu/drm/sti/sti_hqvdp.c @@ -0,0 +1,1073 @@ +/* + * Copyright (C) STMicroelectronics SA 2014 + * Authors: Fabien Dessenne <fabien.dessenne@st.com> for STMicroelectronics. + * License terms: GNU General Public License (GPL), version 2 + */ + +#include <linux/clk.h> +#include <linux/component.h> +#include <linux/firmware.h> +#include <linux/module.h> +#include <linux/platform_device.h> +#include <linux/reset.h> + +#include <drm/drmP.h> + +#include "sti_drm_plane.h" +#include "sti_hqvdp.h" +#include "sti_hqvdp_lut.h" +#include "sti_layer.h" +#include "sti_vtg.h" + +/* Firmware name */ +#define HQVDP_FMW_NAME "hqvdp-stih407.bin" + +/* Regs address */ +#define HQVDP_DMEM 0x00000000 /* 0x00000000 */ +#define HQVDP_PMEM 0x00040000 /* 0x00040000 */ +#define HQVDP_RD_PLUG 0x000E0000 /* 0x000E0000 */ +#define HQVDP_RD_PLUG_CONTROL (HQVDP_RD_PLUG + 0x1000) /* 0x000E1000 */ +#define HQVDP_RD_PLUG_PAGE_SIZE (HQVDP_RD_PLUG + 0x1004) /* 0x000E1004 */ +#define HQVDP_RD_PLUG_MIN_OPC (HQVDP_RD_PLUG + 0x1008) /* 0x000E1008 */ +#define HQVDP_RD_PLUG_MAX_OPC (HQVDP_RD_PLUG + 0x100C) /* 0x000E100C */ +#define HQVDP_RD_PLUG_MAX_CHK (HQVDP_RD_PLUG + 0x1010) /* 0x000E1010 */ +#define HQVDP_RD_PLUG_MAX_MSG (HQVDP_RD_PLUG + 0x1014) /* 0x000E1014 */ +#define HQVDP_RD_PLUG_MIN_SPACE (HQVDP_RD_PLUG + 0x1018) /* 0x000E1018 */ +#define HQVDP_WR_PLUG 0x000E2000 /* 0x000E2000 */ +#define HQVDP_WR_PLUG_CONTROL (HQVDP_WR_PLUG + 0x1000) /* 0x000E3000 */ +#define HQVDP_WR_PLUG_PAGE_SIZE (HQVDP_WR_PLUG + 0x1004) /* 0x000E3004 */ +#define HQVDP_WR_PLUG_MIN_OPC (HQVDP_WR_PLUG + 0x1008) /* 0x000E3008 */ +#define HQVDP_WR_PLUG_MAX_OPC (HQVDP_WR_PLUG + 0x100C) /* 0x000E300C */ +#define HQVDP_WR_PLUG_MAX_CHK (HQVDP_WR_PLUG + 0x1010) /* 0x000E3010 */ +#define HQVDP_WR_PLUG_MAX_MSG (HQVDP_WR_PLUG + 0x1014) /* 0x000E3014 */ +#define HQVDP_WR_PLUG_MIN_SPACE (HQVDP_WR_PLUG + 0x1018) /* 0x000E3018 */ +#define HQVDP_MBX 0x000E4000 /* 0x000E4000 */ +#define HQVDP_MBX_IRQ_TO_XP70 (HQVDP_MBX + 0x0000) /* 0x000E4000 */ +#define HQVDP_MBX_INFO_HOST (HQVDP_MBX + 0x0004) /* 0x000E4004 */ +#define HQVDP_MBX_IRQ_TO_HOST (HQVDP_MBX + 0x0008) /* 0x000E4008 */ +#define HQVDP_MBX_INFO_XP70 (HQVDP_MBX + 0x000C) /* 0x000E400C */ +#define HQVDP_MBX_SW_RESET_CTRL (HQVDP_MBX + 0x0010) /* 0x000E4010 */ +#define HQVDP_MBX_STARTUP_CTRL1 (HQVDP_MBX + 0x0014) /* 0x000E4014 */ +#define HQVDP_MBX_STARTUP_CTRL2 (HQVDP_MBX + 0x0018) /* 0x000E4018 */ +#define HQVDP_MBX_GP_STATUS (HQVDP_MBX + 0x001C) /* 0x000E401C */ +#define HQVDP_MBX_NEXT_CMD (HQVDP_MBX + 0x0020) /* 0x000E4020 */ +#define HQVDP_MBX_CURRENT_CMD (HQVDP_MBX + 0x0024) /* 0x000E4024 */ +#define HQVDP_MBX_SOFT_VSYNC (HQVDP_MBX + 0x0028) /* 0x000E4028 */ + +/* Plugs config */ +#define PLUG_CONTROL_ENABLE 0x00000001 +#define PLUG_PAGE_SIZE_256 0x00000002 +#define PLUG_MIN_OPC_8 0x00000003 +#define PLUG_MAX_OPC_64 0x00000006 +#define PLUG_MAX_CHK_2X 0x00000001 +#define PLUG_MAX_MSG_1X 0x00000000 +#define PLUG_MIN_SPACE_1 0x00000000 + +/* SW reset CTRL */ +#define SW_RESET_CTRL_FULL BIT(0) +#define SW_RESET_CTRL_CORE BIT(1) + +/* Startup ctrl 1 */ +#define STARTUP_CTRL1_RST_DONE BIT(0) +#define STARTUP_CTRL1_AUTH_IDLE BIT(2) + +/* Startup ctrl 2 */ +#define STARTUP_CTRL2_FETCH_EN BIT(1) + +/* Info xP70 */ +#define INFO_XP70_FW_READY BIT(15) +#define INFO_XP70_FW_PROCESSING BIT(14) +#define INFO_XP70_FW_INITQUEUES BIT(13) + +/* SOFT_VSYNC */ +#define SOFT_VSYNC_HW 0x00000000 +#define SOFT_VSYNC_SW_CMD 0x00000001 +#define SOFT_VSYNC_SW_CTRL_IRQ 0x00000003 + +/* Reset & boot poll config */ +#define POLL_MAX_ATTEMPT 50 +#define POLL_DELAY_MS 20 + +#define SCALE_FACTOR 8192 +#define SCALE_MAX_FOR_LEG_LUT_F 4096 +#define SCALE_MAX_FOR_LEG_LUT_E 4915 +#define SCALE_MAX_FOR_LEG_LUT_D 6654 +#define SCALE_MAX_FOR_LEG_LUT_C 8192 + +enum sti_hvsrc_orient { + HVSRC_HORI, + HVSRC_VERT +}; + +/* Command structures */ +struct sti_hqvdp_top { + u32 config; + u32 mem_format; + u32 current_luma; + u32 current_enh_luma; + u32 current_right_luma; + u32 current_enh_right_luma; + u32 current_chroma; + u32 current_enh_chroma; + u32 current_right_chroma; + u32 current_enh_right_chroma; + u32 output_luma; + u32 output_chroma; + u32 luma_src_pitch; + u32 luma_enh_src_pitch; + u32 luma_right_src_pitch; + u32 luma_enh_right_src_pitch; + u32 chroma_src_pitch; + u32 chroma_enh_src_pitch; + u32 chroma_right_src_pitch; + u32 chroma_enh_right_src_pitch; + u32 luma_processed_pitch; + u32 chroma_processed_pitch; + u32 input_frame_size; + u32 input_viewport_ori; + u32 input_viewport_ori_right; + u32 input_viewport_size; + u32 left_view_border_width; + u32 right_view_border_width; + u32 left_view_3d_offset_width; + u32 right_view_3d_offset_width; + u32 side_stripe_color; + u32 crc_reset_ctrl; +}; + +/* Configs for interlaced : no IT, no pass thru, 3 fields */ +#define TOP_CONFIG_INTER_BTM 0x00000000 +#define TOP_CONFIG_INTER_TOP 0x00000002 + +/* Config for progressive : no IT, no pass thru, 3 fields */ +#define TOP_CONFIG_PROGRESSIVE 0x00000001 + +/* Default MemFormat: in=420_raster_dual out=444_raster;opaque Mem2Tv mode */ +#define TOP_MEM_FORMAT_DFLT 0x00018060 + +/* Min/Max size */ +#define MAX_WIDTH 0x1FFF +#define MAX_HEIGHT 0x0FFF +#define MIN_WIDTH 0x0030 +#define MIN_HEIGHT 0x0010 + +struct sti_hqvdp_vc1re { + u32 ctrl_prv_csdi; + u32 ctrl_cur_csdi; + u32 ctrl_nxt_csdi; + u32 ctrl_cur_fmd; + u32 ctrl_nxt_fmd; +}; + +struct sti_hqvdp_fmd { + u32 config; + u32 viewport_ori; + u32 viewport_size; + u32 next_next_luma; + u32 next_next_right_luma; + u32 next_next_next_luma; + u32 next_next_next_right_luma; + u32 threshold_scd; + u32 threshold_rfd; + u32 threshold_move; + u32 threshold_cfd; +}; + +struct sti_hqvdp_csdi { + u32 config; + u32 config2; + u32 dcdi_config; + u32 prev_luma; + u32 prev_enh_luma; + u32 prev_right_luma; + u32 prev_enh_right_luma; + u32 next_luma; + u32 next_enh_luma; + u32 next_right_luma; + u32 next_enh_right_luma; + u32 prev_chroma; + u32 prev_enh_chroma; + u32 prev_right_chroma; + u32 prev_enh_right_chroma; + u32 next_chroma; + u32 next_enh_chroma; + u32 next_right_chroma; + u32 next_enh_right_chroma; + u32 prev_motion; + u32 prev_right_motion; + u32 cur_motion; + u32 cur_right_motion; + u32 next_motion; + u32 next_right_motion; +}; + +/* Config for progressive: by pass */ +#define CSDI_CONFIG_PROG 0x00000000 +/* Config for directional deinterlacing without motion */ +#define CSDI_CONFIG_INTER_DIR 0x00000016 +/* Additional configs for fader, blender, motion,... deinterlace algorithms */ +#define CSDI_CONFIG2_DFLT 0x000001B3 +#define CSDI_DCDI_CONFIG_DFLT 0x00203803 + +struct sti_hqvdp_hvsrc { + u32 hor_panoramic_ctrl; + u32 output_picture_size; + u32 init_horizontal; + u32 init_vertical; + u32 param_ctrl; + u32 yh_coef[NB_COEF]; + u32 ch_coef[NB_COEF]; + u32 yv_coef[NB_COEF]; + u32 cv_coef[NB_COEF]; + u32 hori_shift; + u32 vert_shift; +}; + +/* Default ParamCtrl: all controls enabled */ +#define HVSRC_PARAM_CTRL_DFLT 0xFFFFFFFF + +struct sti_hqvdp_iqi { + u32 config; + u32 demo_wind_size; + u32 pk_config; + u32 coeff0_coeff1; + u32 coeff2_coeff3; + u32 coeff4; + u32 pk_lut; + u32 pk_gain; + u32 pk_coring_level; + u32 cti_config; + u32 le_config; + u32 le_lut[64]; + u32 con_bri; + u32 sat_gain; + u32 pxf_conf; + u32 default_color; +}; + +/* Default Config : IQI bypassed */ +#define IQI_CONFIG_DFLT 0x00000001 +/* Default Contrast & Brightness gain = 256 */ +#define IQI_CON_BRI_DFLT 0x00000100 +/* Default Saturation gain = 256 */ +#define IQI_SAT_GAIN_DFLT 0x00000100 +/* Default PxfConf : P2I bypassed */ +#define IQI_PXF_CONF_DFLT 0x00000001 + +struct sti_hqvdp_top_status { + u32 processing_time; + u32 input_y_crc; + u32 input_uv_crc; +}; + +struct sti_hqvdp_fmd_status { + u32 fmd_repeat_move_status; + u32 fmd_scene_count_status; + u32 cfd_sum; + u32 field_sum; + u32 next_y_fmd_crc; + u32 next_next_y_fmd_crc; + u32 next_next_next_y_fmd_crc; +}; + +struct sti_hqvdp_csdi_status { + u32 prev_y_csdi_crc; + u32 cur_y_csdi_crc; + u32 next_y_csdi_crc; + u32 prev_uv_csdi_crc; + u32 cur_uv_csdi_crc; + u32 next_uv_csdi_crc; + u32 y_csdi_crc; + u32 uv_csdi_crc; + u32 uv_cup_crc; + u32 mot_csdi_crc; + u32 mot_cur_csdi_crc; + u32 mot_prev_csdi_crc; +}; + +struct sti_hqvdp_hvsrc_status { + u32 y_hvsrc_crc; + u32 u_hvsrc_crc; + u32 v_hvsrc_crc; +}; + +struct sti_hqvdp_iqi_status { + u32 pxf_it_status; + u32 y_iqi_crc; + u32 u_iqi_crc; + u32 v_iqi_crc; +}; + +/* Main commands. We use 2 commands one being processed by the firmware, one + * ready to be fetched upon next Vsync*/ +#define NB_VDP_CMD 2 + +struct sti_hqvdp_cmd { + struct sti_hqvdp_top top; + struct sti_hqvdp_vc1re vc1re; + struct sti_hqvdp_fmd fmd; + struct sti_hqvdp_csdi csdi; + struct sti_hqvdp_hvsrc hvsrc; + struct sti_hqvdp_iqi iqi; + struct sti_hqvdp_top_status top_status; + struct sti_hqvdp_fmd_status fmd_status; + struct sti_hqvdp_csdi_status csdi_status; + struct sti_hqvdp_hvsrc_status hvsrc_status; + struct sti_hqvdp_iqi_status iqi_status; +}; + +/* + * STI HQVDP structure + * + * @dev: driver device + * @drm_dev: the drm device + * @regs: registers + * @layer: layer structure for hqvdp it self + * @vid_plane: VID plug used as link with compositor IP + * @clk: IP clock + * @clk_pix_main: pix main clock + * @reset: reset control + * @vtg_nb: notifier to handle VTG Vsync + * @btm_field_pending: is there any bottom field (interlaced frame) to display + * @curr_field_count: number of field updates + * @last_field_count: number of field updates since last fps measure + * @hqvdp_cmd: buffer of commands + * @hqvdp_cmd_paddr: physical address of hqvdp_cmd + * @vtg: vtg for main data path + */ +struct sti_hqvdp { + struct device *dev; + struct drm_device *drm_dev; + void __iomem *regs; + struct sti_layer layer; + struct drm_plane *vid_plane; + struct clk *clk; + struct clk *clk_pix_main; + struct reset_control *reset; + struct notifier_block vtg_nb; + bool btm_field_pending; + unsigned int curr_field_count; + unsigned int last_field_count; + void *hqvdp_cmd; + dma_addr_t hqvdp_cmd_paddr; + struct sti_vtg *vtg; +}; + +#define to_sti_hqvdp(x) container_of(x, struct sti_hqvdp, layer) + +static const uint32_t hqvdp_supported_formats[] = { + DRM_FORMAT_NV12, +}; + +static const uint32_t *sti_hqvdp_get_formats(struct sti_layer *layer) +{ + return hqvdp_supported_formats; +} + +static unsigned int sti_hqvdp_get_nb_formats(struct sti_layer *layer) +{ + return ARRAY_SIZE(hqvdp_supported_formats); +} + +/** + * sti_hqvdp_get_free_cmd + * @hqvdp: hqvdp structure + * + * Look for a hqvdp_cmd that is not being used (or about to be used) by the FW. + * + * RETURNS: + * the offset of the command to be used. + * -1 in error cases + */ +static int sti_hqvdp_get_free_cmd(struct sti_hqvdp *hqvdp) +{ + int curr_cmd, next_cmd; + dma_addr_t cmd = hqvdp->hqvdp_cmd_paddr; + int i; + + curr_cmd = readl(hqvdp->regs + HQVDP_MBX_CURRENT_CMD); + next_cmd = readl(hqvdp->regs + HQVDP_MBX_NEXT_CMD); + + for (i = 0; i < NB_VDP_CMD; i++) { + if ((cmd != curr_cmd) && (cmd != next_cmd)) + return i * sizeof(struct sti_hqvdp_cmd); + cmd += sizeof(struct sti_hqvdp_cmd); + } + + return -1; +} + +/** + * sti_hqvdp_get_curr_cmd + * @hqvdp: hqvdp structure + * + * Look for the hqvdp_cmd that is being used by the FW. + * + * RETURNS: + * the offset of the command to be used. + * -1 in error cases + */ +static int sti_hqvdp_get_curr_cmd(struct sti_hqvdp *hqvdp) +{ + int curr_cmd; + dma_addr_t cmd = hqvdp->hqvdp_cmd_paddr; + unsigned int i; + + curr_cmd = readl(hqvdp->regs + HQVDP_MBX_CURRENT_CMD); + + for (i = 0; i < NB_VDP_CMD; i++) { + if (cmd == curr_cmd) + return i * sizeof(struct sti_hqvdp_cmd); + + cmd += sizeof(struct sti_hqvdp_cmd); + } + + return -1; +} + +/** + * sti_hqvdp_update_hvsrc + * @orient: horizontal or vertical + * @scale: scaling/zoom factor + * @hvsrc: the structure containing the LUT coef + * + * Update the Y and C Lut coef, as well as the shift param + * + * RETURNS: + * None. + */ +static void sti_hqvdp_update_hvsrc(enum sti_hvsrc_orient orient, int scale, + struct sti_hqvdp_hvsrc *hvsrc) +{ + const int *coef_c, *coef_y; + int shift_c, shift_y; + + /* Get the appropriate coef tables */ + if (scale < SCALE_MAX_FOR_LEG_LUT_F) { + coef_y = coef_lut_f_y_legacy; + coef_c = coef_lut_f_c_legacy; + shift_y = SHIFT_LUT_F_Y_LEGACY; + shift_c = SHIFT_LUT_F_C_LEGACY; + } else if (scale < SCALE_MAX_FOR_LEG_LUT_E) { + coef_y = coef_lut_e_y_legacy; + coef_c = coef_lut_e_c_legacy; + shift_y = SHIFT_LUT_E_Y_LEGACY; + shift_c = SHIFT_LUT_E_C_LEGACY; + } else if (scale < SCALE_MAX_FOR_LEG_LUT_D) { + coef_y = coef_lut_d_y_legacy; + coef_c = coef_lut_d_c_legacy; + shift_y = SHIFT_LUT_D_Y_LEGACY; + shift_c = SHIFT_LUT_D_C_LEGACY; + } else if (scale < SCALE_MAX_FOR_LEG_LUT_C) { + coef_y = coef_lut_c_y_legacy; + coef_c = coef_lut_c_c_legacy; + shift_y = SHIFT_LUT_C_Y_LEGACY; + shift_c = SHIFT_LUT_C_C_LEGACY; + } else if (scale == SCALE_MAX_FOR_LEG_LUT_C) { + coef_y = coef_c = coef_lut_b; + shift_y = shift_c = SHIFT_LUT_B; + } else { + coef_y = coef_c = coef_lut_a_legacy; + shift_y = shift_c = SHIFT_LUT_A_LEGACY; + } + + if (orient == HVSRC_HORI) { + hvsrc->hori_shift = (shift_c << 16) | shift_y; + memcpy(hvsrc->yh_coef, coef_y, sizeof(hvsrc->yh_coef)); + memcpy(hvsrc->ch_coef, coef_c, sizeof(hvsrc->ch_coef)); + } else { + hvsrc->vert_shift = (shift_c << 16) | shift_y; + memcpy(hvsrc->yv_coef, coef_y, sizeof(hvsrc->yv_coef)); + memcpy(hvsrc->cv_coef, coef_c, sizeof(hvsrc->cv_coef)); + } +} + +/** + * sti_hqvdp_check_hw_scaling + * @layer: hqvdp layer + * + * Check if the HW is able to perform the scaling request + * The firmware scaling limitation is "CEIL(1/Zy) <= FLOOR(LFW)" where: + * Zy = OutputHeight / InputHeight + * LFW = (Tx * IPClock) / (MaxNbCycles * Cp) + * Tx : Total video mode horizontal resolution + * IPClock : HQVDP IP clock (Mhz) + * MaxNbCycles: max(InputWidth, OutputWidth) + * Cp: Video mode pixel clock (Mhz) + * + * RETURNS: + * True if the HW can scale. + */ +static bool sti_hqvdp_check_hw_scaling(struct sti_layer *layer) +{ + struct sti_hqvdp *hqvdp = to_sti_hqvdp(layer); + unsigned long lfw; + unsigned int inv_zy; + + lfw = layer->mode->htotal * (clk_get_rate(hqvdp->clk) / 1000000); + lfw /= max(layer->src_w, layer->dst_w) * layer->mode->clock / 1000; + + inv_zy = DIV_ROUND_UP(layer->src_h, layer->dst_h); + + return (inv_zy <= lfw) ? true : false; +} + +/** + * sti_hqvdp_prepare_layer + * @layer: hqvdp layer + * @first_prepare: true if it is the first time this function is called + * + * Prepares a command for the firmware + * + * RETURNS: + * 0 on success. + */ +static int sti_hqvdp_prepare_layer(struct sti_layer *layer, bool first_prepare) +{ + struct sti_hqvdp *hqvdp = to_sti_hqvdp(layer); + struct sti_hqvdp_cmd *cmd; + int scale_h, scale_v; + int cmd_offset; + + dev_dbg(hqvdp->dev, "%s %s\n", __func__, sti_layer_to_str(layer)); + + /* prepare and commit VID plane */ + hqvdp->vid_plane->funcs->update_plane(hqvdp->vid_plane, + layer->crtc, layer->fb, + layer->dst_x, layer->dst_y, + layer->dst_w, layer->dst_h, + layer->src_x, layer->src_y, + layer->src_w, layer->src_h); + + cmd_offset = sti_hqvdp_get_free_cmd(hqvdp); + if (cmd_offset == -1) { + DRM_ERROR("No available hqvdp_cmd now\n"); + return -EBUSY; + } + cmd = hqvdp->hqvdp_cmd + cmd_offset; + + if (!sti_hqvdp_check_hw_scaling(layer)) { + DRM_ERROR("Scaling beyond HW capabilities\n"); + return -EINVAL; + } + + /* Static parameters, defaulting to progressive mode */ + cmd->top.config = TOP_CONFIG_PROGRESSIVE; + cmd->top.mem_format = TOP_MEM_FORMAT_DFLT; + cmd->hvsrc.param_ctrl = HVSRC_PARAM_CTRL_DFLT; + cmd->csdi.config = CSDI_CONFIG_PROG; + + /* VC1RE, FMD bypassed : keep everything set to 0 + * IQI/P2I bypassed */ + cmd->iqi.config = IQI_CONFIG_DFLT; + cmd->iqi.con_bri = IQI_CON_BRI_DFLT; + cmd->iqi.sat_gain = IQI_SAT_GAIN_DFLT; + cmd->iqi.pxf_conf = IQI_PXF_CONF_DFLT; + + /* Buffer planes address */ + cmd->top.current_luma = (u32) layer->paddr + layer->offsets[0]; + cmd->top.current_chroma = (u32) layer->paddr + layer->offsets[1]; + + /* Pitches */ + cmd->top.luma_processed_pitch = cmd->top.luma_src_pitch = + layer->pitches[0]; + cmd->top.chroma_processed_pitch = cmd->top.chroma_src_pitch = + layer->pitches[1]; + + /* Input / output size + * Align to upper even value */ + layer->dst_w = ALIGN(layer->dst_w, 2); + layer->dst_h = ALIGN(layer->dst_h, 2); + + if ((layer->src_w > MAX_WIDTH) || (layer->src_w < MIN_WIDTH) || + (layer->src_h > MAX_HEIGHT) || (layer->src_h < MIN_HEIGHT) || + (layer->dst_w > MAX_WIDTH) || (layer->dst_w < MIN_WIDTH) || + (layer->dst_h > MAX_HEIGHT) || (layer->dst_h < MIN_HEIGHT)) { + DRM_ERROR("Invalid in/out size %dx%d -> %dx%d\n", + layer->src_w, layer->src_h, + layer->dst_w, layer->dst_h); + return -EINVAL; + } + cmd->top.input_viewport_size = cmd->top.input_frame_size = + layer->src_h << 16 | layer->src_w; + cmd->hvsrc.output_picture_size = layer->dst_h << 16 | layer->dst_w; + cmd->top.input_viewport_ori = layer->src_y << 16 | layer->src_x; + + /* Handle interlaced */ + if (layer->fb->flags & DRM_MODE_FB_INTERLACED) { + /* Top field to display */ + cmd->top.config = TOP_CONFIG_INTER_TOP; + + /* Update pitches and vert size */ + cmd->top.input_frame_size = (layer->src_h / 2) << 16 | + layer->src_w; + cmd->top.luma_processed_pitch *= 2; + cmd->top.luma_src_pitch *= 2; + cmd->top.chroma_processed_pitch *= 2; + cmd->top.chroma_src_pitch *= 2; + + /* Enable directional deinterlacing processing */ + cmd->csdi.config = CSDI_CONFIG_INTER_DIR; + cmd->csdi.config2 = CSDI_CONFIG2_DFLT; + cmd->csdi.dcdi_config = CSDI_DCDI_CONFIG_DFLT; + } + + /* Update hvsrc lut coef */ + scale_h = SCALE_FACTOR * layer->dst_w / layer->src_w; + sti_hqvdp_update_hvsrc(HVSRC_HORI, scale_h, &cmd->hvsrc); + + scale_v = SCALE_FACTOR * layer->dst_h / layer->src_h; + sti_hqvdp_update_hvsrc(HVSRC_VERT, scale_v, &cmd->hvsrc); + + if (first_prepare) { + /* Prevent VTG shutdown */ + if (clk_prepare_enable(hqvdp->clk_pix_main)) { + DRM_ERROR("Failed to prepare/enable pix main clk\n"); + return -ENXIO; + } + + /* Register VTG Vsync callback to handle bottom fields */ + if ((layer->fb->flags & DRM_MODE_FB_INTERLACED) && + sti_vtg_register_client(hqvdp->vtg, + &hqvdp->vtg_nb, layer->mixer_id)) { + DRM_ERROR("Cannot register VTG notifier\n"); + return -ENXIO; + } + } + + return 0; +} + +static int sti_hqvdp_commit_layer(struct sti_layer *layer) +{ + struct sti_hqvdp *hqvdp = to_sti_hqvdp(layer); + int cmd_offset; + + dev_dbg(hqvdp->dev, "%s %s\n", __func__, sti_layer_to_str(layer)); + + cmd_offset = sti_hqvdp_get_free_cmd(hqvdp); + if (cmd_offset == -1) { + DRM_ERROR("No available hqvdp_cmd now\n"); + return -EBUSY; + } + + writel(hqvdp->hqvdp_cmd_paddr + cmd_offset, + hqvdp->regs + HQVDP_MBX_NEXT_CMD); + + hqvdp->curr_field_count++; + + /* Interlaced : get ready to display the bottom field at next Vsync */ + if (layer->fb->flags & DRM_MODE_FB_INTERLACED) + hqvdp->btm_field_pending = true; + + dev_dbg(hqvdp->dev, "%s Posted command:0x%x\n", + __func__, hqvdp->hqvdp_cmd_paddr + cmd_offset); + + return 0; +} + +static int sti_hqvdp_disable_layer(struct sti_layer *layer) +{ + struct sti_hqvdp *hqvdp = to_sti_hqvdp(layer); + int i; + + DRM_DEBUG_DRIVER("%s\n", sti_layer_to_str(layer)); + + /* Unregister VTG Vsync callback */ + if ((layer->fb->flags & DRM_MODE_FB_INTERLACED) && + sti_vtg_unregister_client(hqvdp->vtg, &hqvdp->vtg_nb)) + DRM_DEBUG_DRIVER("Warning: cannot unregister VTG notifier\n"); + + /* Set next cmd to NULL */ + writel(0, hqvdp->regs + HQVDP_MBX_NEXT_CMD); + + for (i = 0; i < POLL_MAX_ATTEMPT; i++) { + if (readl(hqvdp->regs + HQVDP_MBX_INFO_XP70) + & INFO_XP70_FW_READY) + break; + msleep(POLL_DELAY_MS); + } + + /* VTG can stop now */ + clk_disable_unprepare(hqvdp->clk_pix_main); + + if (i == POLL_MAX_ATTEMPT) { + DRM_ERROR("XP70 could not revert to idle\n"); + return -ENXIO; + } + + /* disable VID plane */ + hqvdp->vid_plane->funcs->disable_plane(hqvdp->vid_plane); + + return 0; +} + +/** + * sti_vdp_vtg_cb + * @nb: notifier block + * @evt: event message + * @data: private data + * + * Handle VTG Vsync event, display pending bottom field + * + * RETURNS: + * 0 on success. + */ +int sti_hqvdp_vtg_cb(struct notifier_block *nb, unsigned long evt, void *data) +{ + struct sti_hqvdp *hqvdp = container_of(nb, struct sti_hqvdp, vtg_nb); + int btm_cmd_offset, top_cmd_offest; + struct sti_hqvdp_cmd *btm_cmd, *top_cmd; + + if ((evt != VTG_TOP_FIELD_EVENT) && (evt != VTG_BOTTOM_FIELD_EVENT)) { + DRM_DEBUG_DRIVER("Unknown event\n"); + return 0; + } + + if (hqvdp->btm_field_pending) { + /* Create the btm field command from the current one */ + btm_cmd_offset = sti_hqvdp_get_free_cmd(hqvdp); + top_cmd_offest = sti_hqvdp_get_curr_cmd(hqvdp); + if ((btm_cmd_offset == -1) || (top_cmd_offest == -1)) { + DRM_ERROR("Cannot get cmds, skip btm field\n"); + return -EBUSY; + } + + btm_cmd = hqvdp->hqvdp_cmd + btm_cmd_offset; + top_cmd = hqvdp->hqvdp_cmd + top_cmd_offest; + + memcpy(btm_cmd, top_cmd, sizeof(*btm_cmd)); + + btm_cmd->top.config = TOP_CONFIG_INTER_BTM; + btm_cmd->top.current_luma += + btm_cmd->top.luma_src_pitch / 2; + btm_cmd->top.current_chroma += + btm_cmd->top.chroma_src_pitch / 2; + + /* Post the command to mailbox */ + writel(hqvdp->hqvdp_cmd_paddr + btm_cmd_offset, + hqvdp->regs + HQVDP_MBX_NEXT_CMD); + + hqvdp->curr_field_count++; + hqvdp->btm_field_pending = false; + + dev_dbg(hqvdp->dev, "%s Posted command:0x%x\n", + __func__, hqvdp->hqvdp_cmd_paddr); + } + + return 0; +} + +static struct drm_plane *sti_hqvdp_find_vid(struct drm_device *dev, int id) +{ + struct drm_plane *plane; + + list_for_each_entry(plane, &dev->mode_config.plane_list, head) { + struct sti_layer *layer = to_sti_layer(plane); + + if (layer->desc == id) + return plane; + } + + return NULL; +} + +static void sti_hqvd_init(struct sti_layer *layer) +{ + struct sti_hqvdp *hqvdp = to_sti_hqvdp(layer); + int size; + + /* find the plane macthing with vid 0 */ + hqvdp->vid_plane = sti_hqvdp_find_vid(hqvdp->drm_dev, STI_VID_0); + if (!hqvdp->vid_plane) { + DRM_ERROR("Cannot find Main video layer\n"); + return; + } + + hqvdp->vtg_nb.notifier_call = sti_hqvdp_vtg_cb; + + /* Allocate memory for the VDP commands */ + size = NB_VDP_CMD * sizeof(struct sti_hqvdp_cmd); + hqvdp->hqvdp_cmd = dma_alloc_writecombine(hqvdp->dev, size, + &hqvdp->hqvdp_cmd_paddr, + GFP_KERNEL | GFP_DMA); + if (!hqvdp->hqvdp_cmd) { + DRM_ERROR("Failed to allocate memory for VDP cmd\n"); + return; + } + + memset(hqvdp->hqvdp_cmd, 0, size); +} + +static const struct sti_layer_funcs hqvdp_ops = { + .get_formats = sti_hqvdp_get_formats, + .get_nb_formats = sti_hqvdp_get_nb_formats, + .init = sti_hqvd_init, + .prepare = sti_hqvdp_prepare_layer, + .commit = sti_hqvdp_commit_layer, + .disable = sti_hqvdp_disable_layer, +}; + +struct sti_layer *sti_hqvdp_create(struct device *dev) +{ + struct sti_hqvdp *hqvdp = dev_get_drvdata(dev); + + hqvdp->layer.ops = &hqvdp_ops; + + return &hqvdp->layer; +} +EXPORT_SYMBOL(sti_hqvdp_create); + +static void sti_hqvdp_init_plugs(struct sti_hqvdp *hqvdp) +{ + /* Configure Plugs (same for RD & WR) */ + writel(PLUG_PAGE_SIZE_256, hqvdp->regs + HQVDP_RD_PLUG_PAGE_SIZE); + writel(PLUG_MIN_OPC_8, hqvdp->regs + HQVDP_RD_PLUG_MIN_OPC); + writel(PLUG_MAX_OPC_64, hqvdp->regs + HQVDP_RD_PLUG_MAX_OPC); + writel(PLUG_MAX_CHK_2X, hqvdp->regs + HQVDP_RD_PLUG_MAX_CHK); + writel(PLUG_MAX_MSG_1X, hqvdp->regs + HQVDP_RD_PLUG_MAX_MSG); + writel(PLUG_MIN_SPACE_1, hqvdp->regs + HQVDP_RD_PLUG_MIN_SPACE); + writel(PLUG_CONTROL_ENABLE, hqvdp->regs + HQVDP_RD_PLUG_CONTROL); + + writel(PLUG_PAGE_SIZE_256, hqvdp->regs + HQVDP_WR_PLUG_PAGE_SIZE); + writel(PLUG_MIN_OPC_8, hqvdp->regs + HQVDP_WR_PLUG_MIN_OPC); + writel(PLUG_MAX_OPC_64, hqvdp->regs + HQVDP_WR_PLUG_MAX_OPC); + writel(PLUG_MAX_CHK_2X, hqvdp->regs + HQVDP_WR_PLUG_MAX_CHK); + writel(PLUG_MAX_MSG_1X, hqvdp->regs + HQVDP_WR_PLUG_MAX_MSG); + writel(PLUG_MIN_SPACE_1, hqvdp->regs + HQVDP_WR_PLUG_MIN_SPACE); + writel(PLUG_CONTROL_ENABLE, hqvdp->regs + HQVDP_WR_PLUG_CONTROL); +} + +/** + * sti_hqvdp_start_xp70 + * @firmware: firmware found + * @ctxt: hqvdp structure + * + * Run the xP70 initialization sequence + */ +static void sti_hqvdp_start_xp70(const struct firmware *firmware, void *ctxt) +{ + struct sti_hqvdp *hqvdp = ctxt; + u32 *fw_rd_plug, *fw_wr_plug, *fw_pmem, *fw_dmem; + u8 *data; + int i; + struct fw_header { + int rd_size; + int wr_size; + int pmem_size; + int dmem_size; + } *header; + + DRM_DEBUG_DRIVER("\n"); + /* Check firmware parts */ + if (!firmware) { + DRM_ERROR("Firmware not available\n"); + return; + } + + header = (struct fw_header *) firmware->data; + if (firmware->size < sizeof(*header)) { + DRM_ERROR("Invalid firmware size (%d)\n", firmware->size); + goto out; + } + if ((sizeof(*header) + header->rd_size + header->wr_size + + header->pmem_size + header->dmem_size) != firmware->size) { + DRM_ERROR("Invalid fmw structure (%d+%d+%d+%d+%d != %d)\n", + sizeof(*header), header->rd_size, header->wr_size, + header->pmem_size, header->dmem_size, + firmware->size); + goto out; + } + + data = (u8 *) firmware->data; + data += sizeof(*header); + fw_rd_plug = (void *) data; + data += header->rd_size; + fw_wr_plug = (void *) data; + data += header->wr_size; + fw_pmem = (void *) data; + data += header->pmem_size; + fw_dmem = (void *) data; + + /* Enable clock */ + if (clk_prepare_enable(hqvdp->clk)) + DRM_ERROR("Failed to prepare/enable HQVDP clk\n"); + + /* Reset */ + writel(SW_RESET_CTRL_FULL, hqvdp->regs + HQVDP_MBX_SW_RESET_CTRL); + + for (i = 0; i < POLL_MAX_ATTEMPT; i++) { + if (readl(hqvdp->regs + HQVDP_MBX_STARTUP_CTRL1) + & STARTUP_CTRL1_RST_DONE) + break; + msleep(POLL_DELAY_MS); + } + if (i == POLL_MAX_ATTEMPT) { + DRM_ERROR("Could not reset\n"); + goto out; + } + + /* Init Read & Write plugs */ + for (i = 0; i < header->rd_size / 4; i++) + writel(fw_rd_plug[i], hqvdp->regs + HQVDP_RD_PLUG + i * 4); + for (i = 0; i < header->wr_size / 4; i++) + writel(fw_wr_plug[i], hqvdp->regs + HQVDP_WR_PLUG + i * 4); + + sti_hqvdp_init_plugs(hqvdp); + + /* Authorize Idle Mode */ + writel(STARTUP_CTRL1_AUTH_IDLE, hqvdp->regs + HQVDP_MBX_STARTUP_CTRL1); + + /* Prevent VTG interruption during the boot */ + writel(SOFT_VSYNC_SW_CTRL_IRQ, hqvdp->regs + HQVDP_MBX_SOFT_VSYNC); + writel(0, hqvdp->regs + HQVDP_MBX_NEXT_CMD); + + /* Download PMEM & DMEM */ + for (i = 0; i < header->pmem_size / 4; i++) + writel(fw_pmem[i], hqvdp->regs + HQVDP_PMEM + i * 4); + for (i = 0; i < header->dmem_size / 4; i++) + writel(fw_dmem[i], hqvdp->regs + HQVDP_DMEM + i * 4); + + /* Enable fetch */ + writel(STARTUP_CTRL2_FETCH_EN, hqvdp->regs + HQVDP_MBX_STARTUP_CTRL2); + + /* Wait end of boot */ + for (i = 0; i < POLL_MAX_ATTEMPT; i++) { + if (readl(hqvdp->regs + HQVDP_MBX_INFO_XP70) + & INFO_XP70_FW_READY) + break; + msleep(POLL_DELAY_MS); + } + if (i == POLL_MAX_ATTEMPT) { + DRM_ERROR("Could not boot\n"); + goto out; + } + + /* Launch Vsync */ + writel(SOFT_VSYNC_HW, hqvdp->regs + HQVDP_MBX_SOFT_VSYNC); + + DRM_INFO("HQVDP XP70 started\n"); +out: + release_firmware(firmware); +} + +int sti_hqvdp_bind(struct device *dev, struct device *master, void *data) +{ + struct sti_hqvdp *hqvdp = dev_get_drvdata(dev); + struct drm_device *drm_dev = data; + struct sti_layer *layer; + int err; + + DRM_DEBUG_DRIVER("\n"); + + hqvdp->drm_dev = drm_dev; + + /* Request for firmware */ + err = request_firmware_nowait(THIS_MODULE, FW_ACTION_HOTPLUG, + HQVDP_FMW_NAME, hqvdp->dev, + GFP_KERNEL, hqvdp, sti_hqvdp_start_xp70); + if (err) { + DRM_ERROR("Can't get HQVDP firmware\n"); + return err; + } + + layer = sti_layer_create(hqvdp->dev, STI_HQVDP_0, hqvdp->regs); + if (!layer) { + DRM_ERROR("Can't create HQVDP plane\n"); + return -ENOMEM; + } + + sti_drm_plane_init(drm_dev, layer, 1, DRM_PLANE_TYPE_OVERLAY); + + return 0; +} + +static void sti_hqvdp_unbind(struct device *dev, + struct device *master, void *data) +{ + /* do nothing */ +} + +static const struct component_ops sti_hqvdp_ops = { + .bind = sti_hqvdp_bind, + .unbind = sti_hqvdp_unbind, +}; + +static int sti_hqvdp_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct device_node *vtg_np; + struct sti_hqvdp *hqvdp; + struct resource *res; + + DRM_DEBUG_DRIVER("\n"); + + hqvdp = devm_kzalloc(dev, sizeof(*hqvdp), GFP_KERNEL); + if (!hqvdp) { + DRM_ERROR("Failed to allocate HQVDP context\n"); + return -ENOMEM; + } + + hqvdp->dev = dev; + + /* Get Memory resources */ + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (res == NULL) { + DRM_ERROR("Get memory resource failed\n"); + return -ENXIO; + } + hqvdp->regs = devm_ioremap(dev, res->start, resource_size(res)); + if (hqvdp->regs == NULL) { + DRM_ERROR("Register mapping failed\n"); + return -ENXIO; + } + + /* Get clock resources */ + hqvdp->clk = devm_clk_get(dev, "hqvdp"); + hqvdp->clk_pix_main = devm_clk_get(dev, "pix_main"); + if (IS_ERR(hqvdp->clk) || IS_ERR(hqvdp->clk)) { + DRM_ERROR("Cannot get clocks\n"); + return -ENXIO; + } + + /* Get reset resources */ + hqvdp->reset = devm_reset_control_get(dev, "hqvdp"); + if (!IS_ERR(hqvdp->reset)) + reset_control_deassert(hqvdp->reset); + + vtg_np = of_parse_phandle(pdev->dev.of_node, "st,vtg", 0); + if (vtg_np) + hqvdp->vtg = of_vtg_find(vtg_np); + + platform_set_drvdata(pdev, hqvdp); + + return component_add(&pdev->dev, &sti_hqvdp_ops); +} + +static int sti_hqvdp_remove(struct platform_device *pdev) +{ + component_del(&pdev->dev, &sti_hqvdp_ops); + return 0; +} + +static struct of_device_id hqvdp_of_match[] = { + { .compatible = "st,stih407-hqvdp", }, + { /* end node */ } +}; +MODULE_DEVICE_TABLE(of, hqvdp_of_match); + +struct platform_driver sti_hqvdp_driver = { + .driver = { + .name = "sti-hqvdp", + .owner = THIS_MODULE, + .of_match_table = hqvdp_of_match, + }, + .probe = sti_hqvdp_probe, + .remove = sti_hqvdp_remove, +}; + +module_platform_driver(sti_hqvdp_driver); + +MODULE_AUTHOR("Benjamin Gaignard <benjamin.gaignard@st.com>"); +MODULE_DESCRIPTION("STMicroelectronics SoC DRM driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/gpu/drm/sti/sti_hqvdp.h b/drivers/gpu/drm/sti/sti_hqvdp.h new file mode 100644 index 000000000000..cd5ecd0a6dea --- /dev/null +++ b/drivers/gpu/drm/sti/sti_hqvdp.h @@ -0,0 +1,12 @@ +/* + * Copyright (C) STMicroelectronics SA 2014 + * Authors: Fabien Dessenne <fabien.dessenne@st.com> for STMicroelectronics. + * License terms: GNU General Public License (GPL), version 2 + */ + +#ifndef _STI_HQVDP_H_ +#define _STI_HQVDP_H_ + +struct sti_layer *sti_hqvdp_create(struct device *dev); + +#endif diff --git a/drivers/gpu/drm/sti/sti_hqvdp_lut.h b/drivers/gpu/drm/sti/sti_hqvdp_lut.h new file mode 100644 index 000000000000..619af7f4384e --- /dev/null +++ b/drivers/gpu/drm/sti/sti_hqvdp_lut.h @@ -0,0 +1,373 @@ +/* + * Copyright (C) STMicroelectronics SA 2014 + * Authors: Fabien Dessenne <fabien.dessenne@st.com> for STMicroelectronics. + * License terms: GNU General Public License (GPL), version 2 + */ + +#ifndef _STI_HQVDP_LUT_H_ +#define _STI_HQVDP_LUT_H_ + +#define NB_COEF 128 + +#define SHIFT_LUT_A_LEGACY 8 +#define SHIFT_LUT_B 8 +#define SHIFT_LUT_C_Y_LEGACY 8 +#define SHIFT_LUT_C_C_LEGACY 8 +#define SHIFT_LUT_D_Y_LEGACY 8 +#define SHIFT_LUT_D_C_LEGACY 8 +#define SHIFT_LUT_E_Y_LEGACY 8 +#define SHIFT_LUT_E_C_LEGACY 8 +#define SHIFT_LUT_F_Y_LEGACY 8 +#define SHIFT_LUT_F_C_LEGACY 8 + +static const u32 coef_lut_a_legacy[NB_COEF] = { + 0x0000ffff, 0x00010000, 0x000100ff, 0x00000000, + 0x00000000, 0x00050000, 0xfffc00ff, 0x00000000, + 0x00000000, 0x00090000, 0xfff900fe, 0x00000000, + 0x00000000, 0x0010ffff, 0xfff600fb, 0x00000000, + 0x00000000, 0x0017fffe, 0xfff400f7, 0x00000000, + 0x00000000, 0x001ffffd, 0xfff200f2, 0x00000000, + 0x00000000, 0x0027fffc, 0xfff100ec, 0x00000000, + 0x00000000, 0x0030fffb, 0xfff000e5, 0x00000000, + 0x00000000, 0x003afffa, 0xffee00de, 0x00000000, + 0x00000000, 0x0044fff9, 0xffed00d6, 0x00000000, + 0x00000000, 0x004efff8, 0xffed00cd, 0x00000000, + 0x00000000, 0x0059fff6, 0xffed00c4, 0x00000000, + 0x00000000, 0x0064fff5, 0xffed00ba, 0x00000000, + 0x00000000, 0x006ffff3, 0xffee00b0, 0x00000000, + 0x00000000, 0x007afff2, 0xffee00a6, 0x00000000, + 0x00000000, 0x0085fff1, 0xffef009b, 0x00000000, + 0x00000000, 0x0090fff0, 0xfff00090, 0x00000000, + 0x00000000, 0x009bffef, 0xfff10085, 0x00000000, + 0x00000000, 0x00a6ffee, 0xfff2007a, 0x00000000, + 0x00000000, 0x00b0ffee, 0xfff3006f, 0x00000000, + 0x00000000, 0x00baffed, 0xfff50064, 0x00000000, + 0x00000000, 0x00c4ffed, 0xfff60059, 0x00000000, + 0x00000000, 0x00cdffed, 0xfff8004e, 0x00000000, + 0x00000000, 0x00d6ffed, 0xfff90044, 0x00000000, + 0x00000000, 0x00deffee, 0xfffa003a, 0x00000000, + 0x00000000, 0x00e5fff0, 0xfffb0030, 0x00000000, + 0x00000000, 0x00ecfff1, 0xfffc0027, 0x00000000, + 0x00000000, 0x00f2fff2, 0xfffd001f, 0x00000000, + 0x00000000, 0x00f7fff4, 0xfffe0017, 0x00000000, + 0x00000000, 0x00fbfff6, 0xffff0010, 0x00000000, + 0x00000000, 0x00fefff9, 0x00000009, 0x00000000, + 0x00000000, 0x00fffffc, 0x00000005, 0x00000000 +}; + +static const u32 coef_lut_b[NB_COEF] = { + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000, + 0x00000000, 0x00000000, 0x00000100, 0x00000000 +}; + +static const u32 coef_lut_c_y_legacy[NB_COEF] = { + 0x00060004, 0x0038ffe1, 0x003800be, 0x0006ffe1, + 0x00050005, 0x0042ffe1, 0x003800b3, 0x0007ffe1, + 0x00040006, 0x0046ffe1, 0x003300b2, 0x0008ffe2, + 0x00030007, 0x004cffe1, 0x002e00b1, 0x0008ffe2, + 0x00020006, 0x0051ffe2, 0x002900b0, 0x0009ffe3, + 0x00010008, 0x0056ffe2, 0x002400ae, 0x0009ffe4, + 0xffff0008, 0x005cffe3, 0x001f00ad, 0x000affe4, + 0xfffe0008, 0x0062ffe4, 0x001a00ab, 0x000affe5, + 0xfffd000a, 0x0066ffe5, 0x001500a8, 0x000bffe6, + 0xfffc0009, 0x006bffe7, 0x001100a5, 0x000bffe8, + 0xfffa000a, 0x0070ffe8, 0x000d00a3, 0x000bffe9, + 0xfff9000b, 0x0076ffea, 0x0008009f, 0x000bffea, + 0xfff7000b, 0x007affec, 0x0005009b, 0x000cffec, + 0xfff6000b, 0x007effef, 0x00010098, 0x000cffed, + 0xfff4000b, 0x0084fff1, 0xfffd0095, 0x000cffee, + 0xfff3000b, 0x0088fff4, 0xfffa0090, 0x000cfff0, + 0xfff1000b, 0x008dfff7, 0xfff7008d, 0x000bfff1, + 0xfff0000c, 0x0090fffa, 0xfff40088, 0x000bfff3, + 0xffee000c, 0x0095fffd, 0xfff10084, 0x000bfff4, + 0xffed000c, 0x00980001, 0xffef007e, 0x000bfff6, + 0xffec000c, 0x009b0005, 0xffec007a, 0x000bfff7, + 0xffea000b, 0x009f0008, 0xffea0076, 0x000bfff9, + 0xffe9000b, 0x00a3000d, 0xffe80070, 0x000afffa, + 0xffe8000b, 0x00a50011, 0xffe7006b, 0x0009fffc, + 0xffe6000b, 0x00a80015, 0xffe50066, 0x000afffd, + 0xffe5000a, 0x00ab001a, 0xffe40062, 0x0008fffe, + 0xffe4000a, 0x00ad001f, 0xffe3005c, 0x0008ffff, + 0xffe40009, 0x00ae0024, 0xffe20056, 0x00080001, + 0xffe30009, 0x00b00029, 0xffe20051, 0x00060002, + 0xffe20008, 0x00b1002e, 0xffe1004c, 0x00070003, + 0xffe20008, 0x00b20033, 0xffe10046, 0x00060004, + 0xffe10007, 0x00b30038, 0xffe10042, 0x00050005 +}; + +static const u32 coef_lut_c_c_legacy[NB_COEF] = { + 0x0001fff3, 0x003afffb, 0x003a00a1, 0x0001fffb, + 0x0001fff5, 0x0041fffb, 0x0038009a, 0x0001fffb, + 0x0001fff5, 0x0046fffb, 0x00340099, 0x0001fffb, + 0x0001fff7, 0x0049fffb, 0x00300098, 0x0001fffb, + 0x0001fff9, 0x004cfffb, 0x002d0096, 0x0001fffb, + 0x0001fffa, 0x004ffffc, 0x00290095, 0x0001fffb, + 0x0001fff9, 0x0054fffd, 0x00250093, 0x0001fffc, + 0x0001fffa, 0x0058fffd, 0x00220092, 0x0000fffc, + 0x0001fffb, 0x005bfffe, 0x001f0090, 0x0000fffc, + 0x0001fffd, 0x005effff, 0x001c008c, 0x0000fffd, + 0x0001fffd, 0x00620000, 0x0019008a, 0x0000fffd, + 0x0001fffe, 0x00660001, 0x00160088, 0xfffffffd, + 0x0000fffe, 0x006a0003, 0x00130085, 0xfffffffe, + 0x0000fffe, 0x006e0004, 0x00100083, 0xfffffffe, + 0x0000fffe, 0x00710006, 0x000e007f, 0xffffffff, + 0x0000fffe, 0x00750008, 0x000c007c, 0xfffeffff, + 0xfffffffe, 0x0079000a, 0x000a0079, 0xfffeffff, + 0xfffffffe, 0x007c000c, 0x00080075, 0xfffe0000, + 0xffffffff, 0x007f000e, 0x00060071, 0xfffe0000, + 0xfffeffff, 0x00830010, 0x0004006e, 0xfffe0000, + 0xfffeffff, 0x00850013, 0x0003006a, 0xfffe0000, + 0xfffdffff, 0x00880016, 0x00010066, 0xfffe0001, + 0xfffd0000, 0x008a0019, 0x00000062, 0xfffd0001, + 0xfffd0000, 0x008c001c, 0xffff005e, 0xfffd0001, + 0xfffc0000, 0x0090001f, 0xfffe005b, 0xfffb0001, + 0xfffc0000, 0x00920022, 0xfffd0058, 0xfffa0001, + 0xfffc0001, 0x00930025, 0xfffd0054, 0xfff90001, + 0xfffb0001, 0x00950029, 0xfffc004f, 0xfffa0001, + 0xfffb0001, 0x0096002d, 0xfffb004c, 0xfff90001, + 0xfffb0001, 0x00980030, 0xfffb0049, 0xfff70001, + 0xfffb0001, 0x00990034, 0xfffb0046, 0xfff50001, + 0xfffb0001, 0x009a0038, 0xfffb0041, 0xfff50001 +}; + +static const u32 coef_lut_d_y_legacy[NB_COEF] = { + 0xfff80009, 0x0046ffec, 0x004600a3, 0xfff8ffec, + 0xfff70009, 0x004effed, 0x0044009d, 0xfff9ffeb, + 0xfff6000a, 0x0052ffee, 0x003f009d, 0xfffaffea, + 0xfff50009, 0x0057ffef, 0x003b009d, 0xfffbffe9, + 0xfff50008, 0x005bfff0, 0x0037009c, 0xfffcffe9, + 0xfff40008, 0x005ffff2, 0x0033009b, 0xfffcffe9, + 0xfff30007, 0x0064fff3, 0x002f009b, 0xfffdffe8, + 0xfff20007, 0x0068fff5, 0x002b0099, 0xfffeffe8, + 0xfff10008, 0x006bfff7, 0x00270097, 0xffffffe8, + 0xfff00007, 0x006ffff9, 0x00230097, 0xffffffe8, + 0xffef0006, 0x0073fffb, 0x00200095, 0x0000ffe8, + 0xffee0005, 0x0077fffe, 0x001c0093, 0x0000ffe9, + 0xffee0005, 0x007a0000, 0x00180091, 0x0001ffe9, + 0xffed0005, 0x007d0003, 0x0015008e, 0x0002ffe9, + 0xffec0005, 0x00800006, 0x0012008b, 0x0002ffea, + 0xffeb0004, 0x00840008, 0x000e008a, 0x0003ffea, + 0xffeb0003, 0x0087000b, 0x000b0087, 0x0003ffeb, + 0xffea0003, 0x008a000e, 0x00080084, 0x0004ffeb, + 0xffea0002, 0x008b0012, 0x00060080, 0x0005ffec, + 0xffe90002, 0x008e0015, 0x0003007d, 0x0005ffed, + 0xffe90001, 0x00910018, 0x0000007a, 0x0005ffee, + 0xffe90000, 0x0093001c, 0xfffe0077, 0x0005ffee, + 0xffe80000, 0x00950020, 0xfffb0073, 0x0006ffef, + 0xffe8ffff, 0x00970023, 0xfff9006f, 0x0007fff0, + 0xffe8ffff, 0x00970027, 0xfff7006b, 0x0008fff1, + 0xffe8fffe, 0x0099002b, 0xfff50068, 0x0007fff2, + 0xffe8fffd, 0x009b002f, 0xfff30064, 0x0007fff3, + 0xffe9fffc, 0x009b0033, 0xfff2005f, 0x0008fff4, + 0xffe9fffc, 0x009c0037, 0xfff0005b, 0x0008fff5, + 0xffe9fffb, 0x009d003b, 0xffef0057, 0x0009fff5, + 0xffeafffa, 0x009d003f, 0xffee0052, 0x000afff6, + 0xffebfff9, 0x009d0044, 0xffed004e, 0x0009fff7 +}; + +static const u32 coef_lut_d_c_legacy[NB_COEF] = { + 0xfffeffff, 0x003fffff, 0x003f0089, 0xfffeffff, + 0xfffe0000, 0x00460000, 0x0042007d, 0xfffffffe, + 0xfffe0000, 0x00490001, 0x003f007d, 0xfffffffd, + 0xfffd0001, 0x004b0002, 0x003c007d, 0x0000fffc, + 0xfffd0001, 0x004e0003, 0x0039007c, 0x0000fffc, + 0xfffc0001, 0x00510005, 0x0036007c, 0x0000fffb, + 0xfffc0001, 0x00540006, 0x0033007b, 0x0001fffa, + 0xfffc0003, 0x00550008, 0x00310078, 0x0001fffa, + 0xfffb0003, 0x00580009, 0x002e0078, 0x0001fffa, + 0xfffb0002, 0x005b000b, 0x002b0077, 0x0002fff9, + 0xfffa0003, 0x005e000d, 0x00280075, 0x0002fff9, + 0xfffa0002, 0x0060000f, 0x00260074, 0x0002fff9, + 0xfffa0004, 0x00610011, 0x00230072, 0x0002fff9, + 0xfffa0004, 0x00640013, 0x00200070, 0x0002fff9, + 0xfff90004, 0x00660015, 0x001e006e, 0x0003fff9, + 0xfff90004, 0x00680017, 0x001c006c, 0x0003fff9, + 0xfff90003, 0x006b0019, 0x0019006b, 0x0003fff9, + 0xfff90003, 0x006c001c, 0x00170068, 0x0004fff9, + 0xfff90003, 0x006e001e, 0x00150066, 0x0004fff9, + 0xfff90002, 0x00700020, 0x00130064, 0x0004fffa, + 0xfff90002, 0x00720023, 0x00110061, 0x0004fffa, + 0xfff90002, 0x00740026, 0x000f0060, 0x0002fffa, + 0xfff90002, 0x00750028, 0x000d005e, 0x0003fffa, + 0xfff90002, 0x0077002b, 0x000b005b, 0x0002fffb, + 0xfffa0001, 0x0078002e, 0x00090058, 0x0003fffb, + 0xfffa0001, 0x00780031, 0x00080055, 0x0003fffc, + 0xfffa0001, 0x007b0033, 0x00060054, 0x0001fffc, + 0xfffb0000, 0x007c0036, 0x00050051, 0x0001fffc, + 0xfffc0000, 0x007c0039, 0x0003004e, 0x0001fffd, + 0xfffc0000, 0x007d003c, 0x0002004b, 0x0001fffd, + 0xfffdffff, 0x007d003f, 0x00010049, 0x0000fffe, + 0xfffeffff, 0x007d0042, 0x00000046, 0x0000fffe +}; + +static const u32 coef_lut_e_y_legacy[NB_COEF] = { + 0xfff10001, 0x00490004, 0x00490083, 0xfff10004, + 0xfff10000, 0x00500006, 0x004b007b, 0xfff10002, + 0xfff10000, 0x00530007, 0x0048007b, 0xfff10001, + 0xfff10000, 0x00550009, 0x0046007a, 0xfff10000, + 0xfff1fffe, 0x0058000b, 0x0043007b, 0xfff2fffe, + 0xfff1ffff, 0x005a000d, 0x0040007a, 0xfff2fffd, + 0xfff1fffd, 0x005d000f, 0x003e007a, 0xfff2fffc, + 0xfff1fffd, 0x005f0011, 0x003b0079, 0xfff3fffb, + 0xfff1fffc, 0x00610013, 0x00390079, 0xfff3fffa, + 0xfff1fffb, 0x00640015, 0x00360079, 0xfff3fff9, + 0xfff1fffa, 0x00660017, 0x00340078, 0xfff4fff8, + 0xfff1fffb, 0x00680019, 0x00310077, 0xfff4fff7, + 0xfff2fff9, 0x006a001b, 0x002f0076, 0xfff5fff6, + 0xfff2fff9, 0x006c001e, 0x002c0075, 0xfff5fff5, + 0xfff2fff9, 0x006d0020, 0x002a0073, 0xfff6fff5, + 0xfff3fff7, 0x00700022, 0x00270073, 0xfff6fff4, + 0xfff3fff7, 0x00710025, 0x00250071, 0xfff7fff3, + 0xfff4fff6, 0x00730027, 0x00220070, 0xfff7fff3, + 0xfff5fff6, 0x0073002a, 0x0020006d, 0xfff9fff2, + 0xfff5fff5, 0x0075002c, 0x001e006c, 0xfff9fff2, + 0xfff6fff5, 0x0076002f, 0x001b006a, 0xfff9fff2, + 0xfff7fff4, 0x00770031, 0x00190068, 0xfffbfff1, + 0xfff8fff4, 0x00780034, 0x00170066, 0xfffafff1, + 0xfff9fff3, 0x00790036, 0x00150064, 0xfffbfff1, + 0xfffafff3, 0x00790039, 0x00130061, 0xfffcfff1, + 0xfffbfff3, 0x0079003b, 0x0011005f, 0xfffdfff1, + 0xfffcfff2, 0x007a003e, 0x000f005d, 0xfffdfff1, + 0xfffdfff2, 0x007a0040, 0x000d005a, 0xfffffff1, + 0xfffefff2, 0x007b0043, 0x000b0058, 0xfffefff1, + 0x0000fff1, 0x007a0046, 0x00090055, 0x0000fff1, + 0x0001fff1, 0x007b0048, 0x00070053, 0x0000fff1, + 0x0002fff1, 0x007b004b, 0x00060050, 0x0000fff1 +}; + +static const u32 coef_lut_e_c_legacy[NB_COEF] = { + 0xfffa0001, 0x003f0010, 0x003f006d, 0xfffa0010, + 0xfffb0002, 0x00440011, 0x00440062, 0xfffa000e, + 0xfffb0001, 0x00460013, 0x00420062, 0xfffa000d, + 0xfffb0000, 0x00480014, 0x00410062, 0xfffa000c, + 0xfffb0001, 0x00490015, 0x003f0061, 0xfffb000b, + 0xfffb0000, 0x004b0017, 0x003d0061, 0xfffb000a, + 0xfffb0000, 0x004d0018, 0x003b0062, 0xfffb0008, + 0xfffcffff, 0x004f001a, 0x00390061, 0xfffb0007, + 0xfffc0000, 0x004f001c, 0x00380060, 0xfffb0006, + 0xfffcffff, 0x0052001d, 0x00360060, 0xfffb0005, + 0xfffdfffe, 0x0053001f, 0x00340060, 0xfffb0004, + 0xfffdfffe, 0x00540021, 0x0032005e, 0xfffc0004, + 0xfffeffff, 0x00550022, 0x0030005d, 0xfffc0003, + 0xfffeffff, 0x00560024, 0x002f005c, 0xfffc0002, + 0xfffffffd, 0x00580026, 0x002d005c, 0xfffc0001, + 0xfffffffd, 0x005a0027, 0x002b005c, 0xfffc0000, + 0x0000fffd, 0x005a0029, 0x0029005a, 0xfffd0000, + 0x0000fffc, 0x005c002b, 0x0027005a, 0xfffdffff, + 0x0001fffc, 0x005c002d, 0x00260058, 0xfffdffff, + 0x0002fffc, 0x005c002f, 0x00240056, 0xfffffffe, + 0x0003fffc, 0x005d0030, 0x00220055, 0xfffffffe, + 0x0004fffc, 0x005e0032, 0x00210054, 0xfffefffd, + 0x0004fffb, 0x00600034, 0x001f0053, 0xfffefffd, + 0x0005fffb, 0x00600036, 0x001d0052, 0xfffffffc, + 0x0006fffb, 0x00600038, 0x001c004f, 0x0000fffc, + 0x0007fffb, 0x00610039, 0x001a004f, 0xfffffffc, + 0x0008fffb, 0x0062003b, 0x0018004d, 0x0000fffb, + 0x000afffb, 0x0061003d, 0x0017004b, 0x0000fffb, + 0x000bfffb, 0x0061003f, 0x00150049, 0x0001fffb, + 0x000cfffa, 0x00620041, 0x00140048, 0x0000fffb, + 0x000dfffa, 0x00620042, 0x00130046, 0x0001fffb, + 0x000efffa, 0x00620044, 0x00110044, 0x0002fffb +}; + +static const u32 coef_lut_f_y_legacy[NB_COEF] = { + 0xfff6fff0, 0x00490012, 0x0049006e, 0xfff60012, + 0xfff7fff1, 0x004e0013, 0x00490068, 0xfff60010, + 0xfff7fff2, 0x004f0015, 0x00470067, 0xfff6000f, + 0xfff7fff5, 0x004f0017, 0x00450065, 0xfff6000e, + 0xfff8fff5, 0x00500018, 0x00440065, 0xfff6000c, + 0xfff8fff6, 0x0051001a, 0x00420064, 0xfff6000b, + 0xfff8fff6, 0x0052001c, 0x00400064, 0xfff6000a, + 0xfff9fff6, 0x0054001d, 0x003e0064, 0xfff60008, + 0xfff9fff8, 0x0054001f, 0x003c0063, 0xfff60007, + 0xfffafff8, 0x00550021, 0x003a0062, 0xfff60006, + 0xfffbfff7, 0x00560022, 0x00390062, 0xfff60005, + 0xfffbfff8, 0x00570024, 0x00370061, 0xfff60004, + 0xfffcfff8, 0x00580026, 0x00350060, 0xfff60003, + 0xfffdfff8, 0x00590028, 0x0033005f, 0xfff60002, + 0xfffdfff7, 0x005b002a, 0x0031005f, 0xfff60001, + 0xfffefff7, 0x005c002c, 0x002f005e, 0xfff60000, + 0xfffffff6, 0x005e002d, 0x002d005e, 0xfff6ffff, + 0x0000fff6, 0x005e002f, 0x002c005c, 0xfff7fffe, + 0x0001fff6, 0x005f0031, 0x002a005b, 0xfff7fffd, + 0x0002fff6, 0x005f0033, 0x00280059, 0xfff8fffd, + 0x0003fff6, 0x00600035, 0x00260058, 0xfff8fffc, + 0x0004fff6, 0x00610037, 0x00240057, 0xfff8fffb, + 0x0005fff6, 0x00620039, 0x00220056, 0xfff7fffb, + 0x0006fff6, 0x0062003a, 0x00210055, 0xfff8fffa, + 0x0007fff6, 0x0063003c, 0x001f0054, 0xfff8fff9, + 0x0008fff6, 0x0064003e, 0x001d0054, 0xfff6fff9, + 0x000afff6, 0x00640040, 0x001c0052, 0xfff6fff8, + 0x000bfff6, 0x00640042, 0x001a0051, 0xfff6fff8, + 0x000cfff6, 0x00650044, 0x00180050, 0xfff5fff8, + 0x000efff6, 0x00650045, 0x0017004f, 0xfff5fff7, + 0x000ffff6, 0x00670047, 0x0015004f, 0xfff2fff7, + 0x0010fff6, 0x00680049, 0x0013004e, 0xfff1fff7 +}; + +static const u32 coef_lut_f_c_legacy[NB_COEF] = { + 0x0000fffb, 0x003a001a, 0x003a005d, 0x0000001a, + 0x0001fffb, 0x003f001b, 0x00400051, 0x00000019, + 0x0001fffc, 0x0040001c, 0x003f0051, 0x00000017, + 0x0002fffb, 0x0042001d, 0x003e0051, 0xffff0016, + 0x0002fffb, 0x0043001e, 0x003d0051, 0xffff0015, + 0x0003fffc, 0x00430020, 0x003b0050, 0xffff0014, + 0x0003fffb, 0x00450021, 0x003a0051, 0xfffe0013, + 0x0004fffc, 0x00450022, 0x00390050, 0xfffe0012, + 0x0005fffc, 0x00460023, 0x0038004f, 0xfffe0011, + 0x0005fffb, 0x00480025, 0x00360050, 0xfffd0010, + 0x0006fffc, 0x00480026, 0x0035004f, 0xfffd000f, + 0x0006fffc, 0x00490027, 0x0034004f, 0xfffd000e, + 0x0007fffd, 0x00490028, 0x0033004e, 0xfffd000d, + 0x0008fffc, 0x004a002a, 0x0031004d, 0xfffd000d, + 0x0009fffd, 0x004a002b, 0x0030004d, 0xfffc000c, + 0x0009fffc, 0x004c002c, 0x002f004d, 0xfffc000b, + 0x000afffc, 0x004c002e, 0x002e004c, 0xfffc000a, + 0x000bfffc, 0x004d002f, 0x002c004c, 0xfffc0009, + 0x000cfffc, 0x004d0030, 0x002b004a, 0xfffd0009, + 0x000dfffd, 0x004d0031, 0x002a004a, 0xfffc0008, + 0x000dfffd, 0x004e0033, 0x00280049, 0xfffd0007, + 0x000efffd, 0x004f0034, 0x00270049, 0xfffc0006, + 0x000ffffd, 0x004f0035, 0x00260048, 0xfffc0006, + 0x0010fffd, 0x00500036, 0x00250048, 0xfffb0005, + 0x0011fffe, 0x004f0038, 0x00230046, 0xfffc0005, + 0x0012fffe, 0x00500039, 0x00220045, 0xfffc0004, + 0x0013fffe, 0x0051003a, 0x00210045, 0xfffb0003, + 0x0014ffff, 0x0050003b, 0x00200043, 0xfffc0003, + 0x0015ffff, 0x0051003d, 0x001e0043, 0xfffb0002, + 0x0016ffff, 0x0051003e, 0x001d0042, 0xfffb0002, + 0x00170000, 0x0051003f, 0x001c0040, 0xfffc0001, + 0x00190000, 0x00510040, 0x001b003f, 0xfffb0001 +}; + +#endif diff --git a/drivers/gpu/drm/sti/sti_layer.c b/drivers/gpu/drm/sti/sti_layer.c index 06a587c4f1bb..899104f9d4bc 100644 --- a/drivers/gpu/drm/sti/sti_layer.c +++ b/drivers/gpu/drm/sti/sti_layer.c @@ -11,7 +11,9 @@ #include <drm/drm_fb_cma_helper.h> #include "sti_compositor.h" +#include "sti_cursor.h" #include "sti_gdp.h" +#include "sti_hqvdp.h" #include "sti_layer.h" #include "sti_vid.h" @@ -32,10 +34,13 @@ const char *sti_layer_to_str(struct sti_layer *layer) return "VID1"; case STI_CURSOR: return "CURSOR"; + case STI_HQVDP_0: + return "HQVDP0"; default: return "<UNKNOWN LAYER>"; } } +EXPORT_SYMBOL(sti_layer_to_str); struct sti_layer *sti_layer_create(struct device *dev, int desc, void __iomem *baseaddr) @@ -50,6 +55,12 @@ struct sti_layer *sti_layer_create(struct device *dev, int desc, case STI_VID: layer = sti_vid_create(dev); break; + case STI_CUR: + layer = sti_cursor_create(dev); + break; + case STI_VDP: + layer = sti_hqvdp_create(dev); + break; } if (!layer) { @@ -67,8 +78,11 @@ struct sti_layer *sti_layer_create(struct device *dev, int desc, return layer; } +EXPORT_SYMBOL(sti_layer_create); -int sti_layer_prepare(struct sti_layer *layer, struct drm_framebuffer *fb, +int sti_layer_prepare(struct sti_layer *layer, + struct drm_crtc *crtc, + struct drm_framebuffer *fb, struct drm_display_mode *mode, int mixer_id, int dest_x, int dest_y, int dest_w, int dest_h, int src_x, int src_y, int src_w, int src_h) @@ -88,6 +102,7 @@ int sti_layer_prepare(struct sti_layer *layer, struct drm_framebuffer *fb, return 1; } + layer->crtc = crtc; layer->fb = fb; layer->mode = mode; layer->mixer_id = mixer_id; @@ -100,6 +115,7 @@ int sti_layer_prepare(struct sti_layer *layer, struct drm_framebuffer *fb, layer->src_w = src_w; layer->src_h = src_h; layer->format = fb->pixel_format; + layer->vaddr = cma_obj->vaddr; layer->paddr = cma_obj->paddr; for (i = 0; i < 4; i++) { layer->pitches[i] = fb->pitches[i]; diff --git a/drivers/gpu/drm/sti/sti_layer.h b/drivers/gpu/drm/sti/sti_layer.h index 198c3774cc12..ceff497f557e 100644 --- a/drivers/gpu/drm/sti/sti_layer.h +++ b/drivers/gpu/drm/sti/sti_layer.h @@ -22,7 +22,8 @@ enum sti_layer_type { STI_GDP = 1 << STI_LAYER_TYPE_SHIFT, STI_VID = 2 << STI_LAYER_TYPE_SHIFT, STI_CUR = 3 << STI_LAYER_TYPE_SHIFT, - STI_BCK = 4 << STI_LAYER_TYPE_SHIFT + STI_BCK = 4 << STI_LAYER_TYPE_SHIFT, + STI_VDP = 5 << STI_LAYER_TYPE_SHIFT }; enum sti_layer_id_of_type { @@ -39,6 +40,7 @@ enum sti_layer_desc { STI_GDP_3 = STI_GDP | STI_ID_3, STI_VID_0 = STI_VID | STI_ID_0, STI_VID_1 = STI_VID | STI_ID_1, + STI_HQVDP_0 = STI_VDP | STI_ID_0, STI_CURSOR = STI_CUR, STI_BACK = STI_BCK }; @@ -67,6 +69,7 @@ struct sti_layer_funcs { * * @plane: drm plane it is bound to (if any) * @fb: drm fb it is bound to + * @crtc: crtc it is bound to * @mode: display mode * @desc: layer type & id * @device: driver device @@ -82,11 +85,13 @@ struct sti_layer_funcs { * @format: format * @pitches: pitch of 'planes' (eg: Y, U, V) * @offsets: offset of 'planes' + * @vaddr: virtual address of the input buffer * @paddr: physical address of the input buffer */ struct sti_layer { struct drm_plane plane; struct drm_framebuffer *fb; + struct drm_crtc *crtc; struct drm_display_mode *mode; enum sti_layer_desc desc; struct device *dev; @@ -102,12 +107,15 @@ struct sti_layer { uint32_t format; unsigned int pitches[4]; unsigned int offsets[4]; + void *vaddr; dma_addr_t paddr; }; struct sti_layer *sti_layer_create(struct device *dev, int desc, void __iomem *baseaddr); -int sti_layer_prepare(struct sti_layer *layer, struct drm_framebuffer *fb, +int sti_layer_prepare(struct sti_layer *layer, + struct drm_crtc *crtc, + struct drm_framebuffer *fb, struct drm_display_mode *mode, int mixer_id, int dest_x, int dest_y, diff --git a/drivers/gpu/drm/sti/sti_mixer.c b/drivers/gpu/drm/sti/sti_mixer.c index 79f369db9fb6..13a4b84deab6 100644 --- a/drivers/gpu/drm/sti/sti_mixer.c +++ b/drivers/gpu/drm/sti/sti_mixer.c @@ -45,6 +45,7 @@ static const u32 mixerColorSpaceMatIdentity[] = { #define GAM_CTL_GDP1_MASK BIT(4) #define GAM_CTL_GDP2_MASK BIT(5) #define GAM_CTL_GDP3_MASK BIT(6) +#define GAM_CTL_CURSOR_MASK BIT(9) const char *sti_mixer_to_str(struct sti_mixer *mixer) { @@ -122,11 +123,15 @@ int sti_mixer_set_layer_depth(struct sti_mixer *mixer, struct sti_layer *layer) layer_id = GAM_DEPTH_GDP3_ID; break; case STI_VID_0: + case STI_HQVDP_0: layer_id = GAM_DEPTH_VID0_ID; break; case STI_VID_1: layer_id = GAM_DEPTH_VID1_ID; break; + case STI_CURSOR: + /* no need to set depth for cursor */ + return 0; default: DRM_ERROR("Unknown layer %d\n", layer->desc); return 1; @@ -185,9 +190,12 @@ static u32 sti_mixer_get_layer_mask(struct sti_layer *layer) case STI_GDP_3: return GAM_CTL_GDP3_MASK; case STI_VID_0: + case STI_HQVDP_0: return GAM_CTL_VID0_MASK; case STI_VID_1: return GAM_CTL_VID1_MASK; + case STI_CURSOR: + return GAM_CTL_CURSOR_MASK; default: return 0; } @@ -215,6 +223,15 @@ int sti_mixer_set_layer_status(struct sti_mixer *mixer, return 0; } +void sti_mixer_clear_all_layers(struct sti_mixer *mixer) +{ + u32 val; + + DRM_DEBUG_DRIVER("%s clear all layer\n", sti_mixer_to_str(mixer)); + val = sti_mixer_reg_read(mixer, GAM_MIXER_CTL) & 0xFFFF0000; + sti_mixer_reg_write(mixer, GAM_MIXER_CTL, val); +} + void sti_mixer_set_matrix(struct sti_mixer *mixer) { unsigned int i; diff --git a/drivers/gpu/drm/sti/sti_mixer.h b/drivers/gpu/drm/sti/sti_mixer.h index 874372102e52..b97282182908 100644 --- a/drivers/gpu/drm/sti/sti_mixer.h +++ b/drivers/gpu/drm/sti/sti_mixer.h @@ -23,6 +23,7 @@ * @id: id of the mixer * @drm_crtc: crtc object link to the mixer * @pending_event: set if a flip event is pending on crtc + * @enabled: to know if the mixer is active or not */ struct sti_mixer { struct device *dev; @@ -30,6 +31,7 @@ struct sti_mixer { int id; struct drm_crtc drm_crtc; struct drm_pending_vblank_event *pending_event; + bool enabled; }; const char *sti_mixer_to_str(struct sti_mixer *mixer); @@ -39,6 +41,7 @@ struct sti_mixer *sti_mixer_create(struct device *dev, int id, int sti_mixer_set_layer_status(struct sti_mixer *mixer, struct sti_layer *layer, bool status); +void sti_mixer_clear_all_layers(struct sti_mixer *mixer); int sti_mixer_set_layer_depth(struct sti_mixer *mixer, struct sti_layer *layer); int sti_mixer_active_video_area(struct sti_mixer *mixer, struct drm_display_mode *mode); diff --git a/drivers/gpu/drm/sti/sti_tvout.c b/drivers/gpu/drm/sti/sti_tvout.c index b8afe490356a..cb924aa2b321 100644 --- a/drivers/gpu/drm/sti/sti_tvout.c +++ b/drivers/gpu/drm/sti/sti_tvout.c @@ -16,6 +16,8 @@ #include <drm/drmP.h> #include <drm/drm_crtc_helper.h> +#include "sti_drm_crtc.h" + /* glue registers */ #define TVO_CSC_MAIN_M0 0x000 #define TVO_CSC_MAIN_M1 0x004 @@ -96,7 +98,7 @@ #define TVO_SYNC_HD_DCS_SHIFT 8 -#define ENCODER_MAIN_CRTC_MASK BIT(0) +#define ENCODER_CRTC_MASK (BIT(0) | BIT(1)) /* enum listing the supported output data format */ enum sti_tvout_video_out_type { @@ -149,14 +151,15 @@ static void tvout_write(struct sti_tvout *tvout, u32 val, int offset) * Set the clipping mode of a VIP * * @tvout: tvout structure + * @reg: register to set * @cr_r: * @y_g: * @cb_b: */ -static void tvout_vip_set_color_order(struct sti_tvout *tvout, +static void tvout_vip_set_color_order(struct sti_tvout *tvout, int reg, u32 cr_r, u32 y_g, u32 cb_b) { - u32 val = tvout_read(tvout, TVO_VIP_HDMI); + u32 val = tvout_read(tvout, reg); val &= ~(TVO_VIP_REORDER_MASK << TVO_VIP_REORDER_R_SHIFT); val &= ~(TVO_VIP_REORDER_MASK << TVO_VIP_REORDER_G_SHIFT); @@ -165,52 +168,58 @@ static void tvout_vip_set_color_order(struct sti_tvout *tvout, val |= y_g << TVO_VIP_REORDER_G_SHIFT; val |= cb_b << TVO_VIP_REORDER_B_SHIFT; - tvout_write(tvout, val, TVO_VIP_HDMI); + tvout_write(tvout, val, reg); } /** * Set the clipping mode of a VIP * * @tvout: tvout structure + * @reg: register to set * @range: clipping range */ -static void tvout_vip_set_clip_mode(struct sti_tvout *tvout, u32 range) +static void tvout_vip_set_clip_mode(struct sti_tvout *tvout, int reg, u32 range) { - u32 val = tvout_read(tvout, TVO_VIP_HDMI); + u32 val = tvout_read(tvout, reg); val &= ~(TVO_VIP_CLIP_MASK << TVO_VIP_CLIP_SHIFT); val |= range << TVO_VIP_CLIP_SHIFT; - tvout_write(tvout, val, TVO_VIP_HDMI); + tvout_write(tvout, val, reg); } /** * Set the rounded value of a VIP * * @tvout: tvout structure + * @reg: register to set * @rnd: rounded val per component */ -static void tvout_vip_set_rnd(struct sti_tvout *tvout, u32 rnd) +static void tvout_vip_set_rnd(struct sti_tvout *tvout, int reg, u32 rnd) { - u32 val = tvout_read(tvout, TVO_VIP_HDMI); + u32 val = tvout_read(tvout, reg); val &= ~(TVO_VIP_RND_MASK << TVO_VIP_RND_SHIFT); val |= rnd << TVO_VIP_RND_SHIFT; - tvout_write(tvout, val, TVO_VIP_HDMI); + tvout_write(tvout, val, reg); } /** * Select the VIP input * * @tvout: tvout structure + * @reg: register to set + * @main_path: main or auxiliary path + * @sel_input_logic_inverted: need to invert the logic * @sel_input: selected_input (main/aux + conv) */ static void tvout_vip_set_sel_input(struct sti_tvout *tvout, + int reg, bool main_path, bool sel_input_logic_inverted, enum sti_tvout_video_out_type video_out) { u32 sel_input; - u32 val = tvout_read(tvout, TVO_VIP_HDMI); + u32 val = tvout_read(tvout, reg); if (main_path) sel_input = TVO_VIP_SEL_INPUT_MAIN; @@ -232,22 +241,24 @@ static void tvout_vip_set_sel_input(struct sti_tvout *tvout, val &= ~TVO_VIP_SEL_INPUT_MASK; val |= sel_input; - tvout_write(tvout, val, TVO_VIP_HDMI); + tvout_write(tvout, val, reg); } /** * Select the input video signed or unsigned * * @tvout: tvout structure + * @reg: register to set * @in_vid_signed: used video input format */ -static void tvout_vip_set_in_vid_fmt(struct sti_tvout *tvout, u32 in_vid_fmt) +static void tvout_vip_set_in_vid_fmt(struct sti_tvout *tvout, + int reg, u32 in_vid_fmt) { - u32 val = tvout_read(tvout, TVO_VIP_HDMI); + u32 val = tvout_read(tvout, reg); val &= ~TVO_IN_FMT_SIGNED; val |= in_vid_fmt; - tvout_write(tvout, val, TVO_MAIN_IN_VID_FORMAT); + tvout_write(tvout, val, reg); } /** @@ -261,6 +272,7 @@ static void tvout_hdmi_start(struct sti_tvout *tvout, bool main_path) { struct device_node *node = tvout->dev->of_node; bool sel_input_logic_inverted = false; + u32 tvo_in_vid_format; dev_dbg(tvout->dev, "%s\n", __func__); @@ -268,33 +280,36 @@ static void tvout_hdmi_start(struct sti_tvout *tvout, bool main_path) DRM_DEBUG_DRIVER("main vip for hdmi\n"); /* select the input sync for hdmi = VTG set 1 */ tvout_write(tvout, TVO_SYNC_MAIN_VTG_SET_1, TVO_HDMI_SYNC_SEL); + tvo_in_vid_format = TVO_MAIN_IN_VID_FORMAT; } else { DRM_DEBUG_DRIVER("aux vip for hdmi\n"); /* select the input sync for hdmi = VTG set 1 */ tvout_write(tvout, TVO_SYNC_AUX_VTG_SET_1, TVO_HDMI_SYNC_SEL); + tvo_in_vid_format = TVO_AUX_IN_VID_FORMAT; } /* set color channel order */ - tvout_vip_set_color_order(tvout, + tvout_vip_set_color_order(tvout, TVO_VIP_HDMI, TVO_VIP_REORDER_CR_R_SEL, TVO_VIP_REORDER_Y_G_SEL, TVO_VIP_REORDER_CB_B_SEL); /* set clipping mode (Limited range RGB/Y) */ - tvout_vip_set_clip_mode(tvout, TVO_VIP_CLIP_LIMITED_RANGE_RGB_Y); + tvout_vip_set_clip_mode(tvout, TVO_VIP_HDMI, + TVO_VIP_CLIP_LIMITED_RANGE_RGB_Y); /* set round mode (rounded to 8-bit per component) */ - tvout_vip_set_rnd(tvout, TVO_VIP_RND_8BIT_ROUNDED); + tvout_vip_set_rnd(tvout, TVO_VIP_HDMI, TVO_VIP_RND_8BIT_ROUNDED); if (of_device_is_compatible(node, "st,stih407-tvout")) { /* set input video format */ - tvout_vip_set_in_vid_fmt(tvout->regs + TVO_MAIN_IN_VID_FORMAT, - TVO_IN_FMT_SIGNED); + tvout_vip_set_in_vid_fmt(tvout, tvo_in_vid_format, + TVO_IN_FMT_SIGNED); sel_input_logic_inverted = true; } /* input selection */ - tvout_vip_set_sel_input(tvout, main_path, + tvout_vip_set_sel_input(tvout, TVO_VIP_HDMI, main_path, sel_input_logic_inverted, STI_TVOUT_VIDEO_OUT_RGB); } @@ -309,48 +324,47 @@ static void tvout_hda_start(struct sti_tvout *tvout, bool main_path) { struct device_node *node = tvout->dev->of_node; bool sel_input_logic_inverted = false; + u32 tvo_in_vid_format; + int val; dev_dbg(tvout->dev, "%s\n", __func__); - if (!main_path) { - DRM_ERROR("HD Analog on aux not implemented\n"); - return; + if (main_path) { + val = TVO_SYNC_MAIN_VTG_SET_2 << TVO_SYNC_HD_DCS_SHIFT; + val |= TVO_SYNC_MAIN_VTG_SET_3; + tvout_write(tvout, val, TVO_HD_SYNC_SEL); + tvo_in_vid_format = TVO_MAIN_IN_VID_FORMAT; + } else { + val = TVO_SYNC_AUX_VTG_SET_2 << TVO_SYNC_HD_DCS_SHIFT; + val |= TVO_SYNC_AUX_VTG_SET_3; + tvout_write(tvout, val, TVO_HD_SYNC_SEL); + tvo_in_vid_format = TVO_AUX_IN_VID_FORMAT; } - DRM_DEBUG_DRIVER("main vip for HDF\n"); - /* set color channel order */ - tvout_vip_set_color_order(tvout->regs + TVO_VIP_HDF, + tvout_vip_set_color_order(tvout, TVO_VIP_HDF, TVO_VIP_REORDER_CR_R_SEL, TVO_VIP_REORDER_Y_G_SEL, TVO_VIP_REORDER_CB_B_SEL); - /* set clipping mode (Limited range RGB/Y) */ - tvout_vip_set_clip_mode(tvout->regs + TVO_VIP_HDF, - TVO_VIP_CLIP_LIMITED_RANGE_CB_CR); + /* set clipping mode (EAV/SAV clipping) */ + tvout_vip_set_clip_mode(tvout, TVO_VIP_HDF, TVO_VIP_CLIP_EAV_SAV); /* set round mode (rounded to 10-bit per component) */ - tvout_vip_set_rnd(tvout->regs + TVO_VIP_HDF, TVO_VIP_RND_10BIT_ROUNDED); + tvout_vip_set_rnd(tvout, TVO_VIP_HDF, TVO_VIP_RND_10BIT_ROUNDED); if (of_device_is_compatible(node, "st,stih407-tvout")) { /* set input video format */ - tvout_vip_set_in_vid_fmt(tvout, TVO_IN_FMT_SIGNED); + tvout_vip_set_in_vid_fmt(tvout, + tvo_in_vid_format, TVO_IN_FMT_SIGNED); sel_input_logic_inverted = true; } /* Input selection */ - tvout_vip_set_sel_input(tvout->regs + TVO_VIP_HDF, - main_path, + tvout_vip_set_sel_input(tvout, TVO_VIP_HDF, main_path, sel_input_logic_inverted, STI_TVOUT_VIDEO_OUT_YUV); - /* select the input sync for HD analog = VTG set 3 - * and HD DCS = VTG set 2 */ - tvout_write(tvout, - (TVO_SYNC_MAIN_VTG_SET_2 << TVO_SYNC_HD_DCS_SHIFT) - | TVO_SYNC_MAIN_VTG_SET_3, - TVO_HD_SYNC_SEL); - /* power up HD DAC */ tvout_write(tvout, 0, TVO_HD_DAC_CFG_OFF); } @@ -392,7 +406,7 @@ static void sti_hda_encoder_commit(struct drm_encoder *encoder) { struct sti_tvout *tvout = to_sti_tvout(encoder); - tvout_hda_start(tvout, true); + tvout_hda_start(tvout, sti_drm_crtc_is_main(encoder->crtc)); } static void sti_hda_encoder_disable(struct drm_encoder *encoder) @@ -429,7 +443,7 @@ static struct drm_encoder *sti_tvout_create_hda_encoder(struct drm_device *dev, drm_encoder = (struct drm_encoder *) encoder; - drm_encoder->possible_crtcs = ENCODER_MAIN_CRTC_MASK; + drm_encoder->possible_crtcs = ENCODER_CRTC_MASK; drm_encoder->possible_clones = 1 << 0; drm_encoder_init(dev, drm_encoder, @@ -444,7 +458,7 @@ static void sti_hdmi_encoder_commit(struct drm_encoder *encoder) { struct sti_tvout *tvout = to_sti_tvout(encoder); - tvout_hdmi_start(tvout, true); + tvout_hdmi_start(tvout, sti_drm_crtc_is_main(encoder->crtc)); } static void sti_hdmi_encoder_disable(struct drm_encoder *encoder) @@ -478,7 +492,7 @@ static struct drm_encoder *sti_tvout_create_hdmi_encoder(struct drm_device *dev, drm_encoder = (struct drm_encoder *) encoder; - drm_encoder->possible_crtcs = ENCODER_MAIN_CRTC_MASK; + drm_encoder->possible_crtcs = ENCODER_CRTC_MASK; drm_encoder->possible_clones = 1 << 1; drm_encoder_init(dev, drm_encoder, diff --git a/drivers/gpu/drm/sti/sti_vtg.c b/drivers/gpu/drm/sti/sti_vtg.c index 740d6e347a62..9564f2568e2c 100644 --- a/drivers/gpu/drm/sti/sti_vtg.c +++ b/drivers/gpu/drm/sti/sti_vtg.c @@ -51,10 +51,19 @@ #define VTG_TOP_V_HD_3 0x010C #define VTG_BOT_V_HD_3 0x0110 +#define VTG_H_HD_4 0x0120 +#define VTG_TOP_V_VD_4 0x0124 +#define VTG_BOT_V_VD_4 0x0128 +#define VTG_TOP_V_HD_4 0x012c +#define VTG_BOT_V_HD_4 0x0130 + #define VTG_IRQ_BOTTOM BIT(0) #define VTG_IRQ_TOP BIT(1) #define VTG_IRQ_MASK (VTG_IRQ_TOP | VTG_IRQ_BOTTOM) +/* Delay introduced by the HDMI in nb of pixel */ +#define HDMI_DELAY (6) + /* delay introduced by the Arbitrary Waveform Generator in nb of pixels */ #define AWG_DELAY_HD (-9) #define AWG_DELAY_ED (-8) @@ -133,10 +142,10 @@ static void vtg_set_mode(struct sti_vtg *vtg, writel(tmp, vtg->regs + VTG_VID_TFS); writel(tmp, vtg->regs + VTG_VID_BFS); - /* prepare VTG set 1 and 2 for HDMI and VTG set 3 for HD DAC */ - tmp = (mode->hsync_end - mode->hsync_start) << 16; + /* prepare VTG set 1 for HDMI */ + tmp = (mode->hsync_end - mode->hsync_start + HDMI_DELAY) << 16; + tmp |= HDMI_DELAY; writel(tmp, vtg->regs + VTG_H_HD_1); - writel(tmp, vtg->regs + VTG_H_HD_2); tmp = (mode->vsync_end - mode->vsync_start + 1) << 16; tmp |= 1; @@ -146,6 +155,11 @@ static void vtg_set_mode(struct sti_vtg *vtg, writel(0, vtg->regs + VTG_BOT_V_HD_1); /* prepare VTG set 2 for for HD DCS */ + tmp = (mode->hsync_end - mode->hsync_start) << 16; + writel(tmp, vtg->regs + VTG_H_HD_2); + + tmp = (mode->vsync_end - mode->vsync_start + 1) << 16; + tmp |= 1; writel(tmp, vtg->regs + VTG_TOP_V_VD_2); writel(tmp, vtg->regs + VTG_BOT_V_VD_2); writel(0, vtg->regs + VTG_TOP_V_HD_2); @@ -166,6 +180,17 @@ static void vtg_set_mode(struct sti_vtg *vtg, writel(tmp, vtg->regs + VTG_TOP_V_HD_3); writel(tmp, vtg->regs + VTG_BOT_V_HD_3); + /* Prepare VTG set 4 for DVO */ + tmp = (mode->hsync_end - mode->hsync_start) << 16; + writel(tmp, vtg->regs + VTG_H_HD_4); + + tmp = (mode->vsync_end - mode->vsync_start + 1) << 16; + tmp |= 1; + writel(tmp, vtg->regs + VTG_TOP_V_VD_4); + writel(tmp, vtg->regs + VTG_BOT_V_VD_4); + writel(0, vtg->regs + VTG_TOP_V_HD_4); + writel(0, vtg->regs + VTG_BOT_V_HD_4); + /* mode */ writel(type, vtg->regs + VTG_MODE); } diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c index 3367960286a6..978993fa3a36 100644 --- a/drivers/gpu/drm/tegra/dc.c +++ b/drivers/gpu/drm/tegra/dc.c @@ -168,7 +168,7 @@ static int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index, const struct tegra_dc_window *window) { unsigned h_offset, v_offset, h_size, v_size, h_dda, v_dda, bpp; - unsigned long value; + unsigned long value, flags; bool yuv, planar; /* @@ -181,6 +181,8 @@ static int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index, else bpp = planar ? 1 : 2; + spin_lock_irqsave(&dc->lock, flags); + value = WINDOW_A_SELECT << index; tegra_dc_writel(dc, value, DC_CMD_DISPLAY_WINDOW_HEADER); @@ -273,6 +275,7 @@ static int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index, case TEGRA_BO_TILING_MODE_BLOCK: DRM_ERROR("hardware doesn't support block linear mode\n"); + spin_unlock_irqrestore(&dc->lock, flags); return -EINVAL; } @@ -331,6 +334,8 @@ static int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index, tegra_dc_window_commit(dc, index); + spin_unlock_irqrestore(&dc->lock, flags); + return 0; } @@ -338,11 +343,14 @@ static int tegra_window_plane_disable(struct drm_plane *plane) { struct tegra_dc *dc = to_tegra_dc(plane->crtc); struct tegra_plane *p = to_tegra_plane(plane); + unsigned long flags; u32 value; if (!plane->crtc) return 0; + spin_lock_irqsave(&dc->lock, flags); + value = WINDOW_A_SELECT << p->index; tegra_dc_writel(dc, value, DC_CMD_DISPLAY_WINDOW_HEADER); @@ -352,6 +360,8 @@ static int tegra_window_plane_disable(struct drm_plane *plane) tegra_dc_window_commit(dc, p->index); + spin_unlock_irqrestore(&dc->lock, flags); + return 0; } @@ -699,14 +709,16 @@ static int tegra_dc_set_base(struct tegra_dc *dc, int x, int y, struct tegra_bo *bo = tegra_fb_get_plane(fb, 0); unsigned int h_offset = 0, v_offset = 0; struct tegra_bo_tiling tiling; + unsigned long value, flags; unsigned int format, swap; - unsigned long value; int err; err = tegra_fb_get_tiling(fb, &tiling); if (err < 0) return err; + spin_lock_irqsave(&dc->lock, flags); + tegra_dc_writel(dc, WINDOW_A_SELECT, DC_CMD_DISPLAY_WINDOW_HEADER); value = fb->offsets[0] + y * fb->pitches[0] + @@ -752,6 +764,7 @@ static int tegra_dc_set_base(struct tegra_dc *dc, int x, int y, case TEGRA_BO_TILING_MODE_BLOCK: DRM_ERROR("hardware doesn't support block linear mode\n"); + spin_unlock_irqrestore(&dc->lock, flags); return -EINVAL; } @@ -778,6 +791,8 @@ static int tegra_dc_set_base(struct tegra_dc *dc, int x, int y, tegra_dc_writel(dc, value << 8, DC_CMD_STATE_CONTROL); tegra_dc_writel(dc, value, DC_CMD_STATE_CONTROL); + spin_unlock_irqrestore(&dc->lock, flags); + return 0; } @@ -814,23 +829,32 @@ static void tegra_dc_finish_page_flip(struct tegra_dc *dc) unsigned long flags, base; struct tegra_bo *bo; - if (!dc->event) + spin_lock_irqsave(&drm->event_lock, flags); + + if (!dc->event) { + spin_unlock_irqrestore(&drm->event_lock, flags); return; + } bo = tegra_fb_get_plane(crtc->primary->fb, 0); + spin_lock_irqsave(&dc->lock, flags); + /* check if new start address has been latched */ + tegra_dc_writel(dc, WINDOW_A_SELECT, DC_CMD_DISPLAY_WINDOW_HEADER); tegra_dc_writel(dc, READ_MUX, DC_CMD_STATE_ACCESS); base = tegra_dc_readl(dc, DC_WINBUF_START_ADDR); tegra_dc_writel(dc, 0, DC_CMD_STATE_ACCESS); + spin_unlock_irqrestore(&dc->lock, flags); + if (base == bo->paddr + crtc->primary->fb->offsets[0]) { - spin_lock_irqsave(&drm->event_lock, flags); - drm_send_vblank_event(drm, dc->pipe, dc->event); - drm_vblank_put(drm, dc->pipe); + drm_crtc_send_vblank_event(crtc, dc->event); + drm_crtc_vblank_put(crtc); dc->event = NULL; - spin_unlock_irqrestore(&drm->event_lock, flags); } + + spin_unlock_irqrestore(&drm->event_lock, flags); } void tegra_dc_cancel_page_flip(struct drm_crtc *crtc, struct drm_file *file) @@ -843,7 +867,7 @@ void tegra_dc_cancel_page_flip(struct drm_crtc *crtc, struct drm_file *file) if (dc->event && dc->event->base.file_priv == file) { dc->event->base.destroy(&dc->event->base); - drm_vblank_put(drm, dc->pipe); + drm_crtc_vblank_put(crtc); dc->event = NULL; } @@ -853,16 +877,16 @@ void tegra_dc_cancel_page_flip(struct drm_crtc *crtc, struct drm_file *file) static int tegra_dc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb, struct drm_pending_vblank_event *event, uint32_t page_flip_flags) { + unsigned int pipe = drm_crtc_index(crtc); struct tegra_dc *dc = to_tegra_dc(crtc); - struct drm_device *drm = crtc->dev; if (dc->event) return -EBUSY; if (event) { - event->pipe = dc->pipe; + event->pipe = pipe; dc->event = event; - drm_vblank_get(drm, dc->pipe); + drm_crtc_vblank_get(crtc); } tegra_dc_set_base(dc, 0, 0, fb); @@ -1127,7 +1151,7 @@ static irqreturn_t tegra_dc_irq(int irq, void *data) /* dev_dbg(dc->dev, "%s(): vertical blank\n", __func__); */ - drm_handle_vblank(dc->base.dev, dc->pipe); + drm_crtc_handle_vblank(&dc->base); tegra_dc_finish_page_flip(dc); } diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c index e549afeece1f..d4f827593dfa 100644 --- a/drivers/gpu/drm/tegra/drm.c +++ b/drivers/gpu/drm/tegra/drm.c @@ -694,24 +694,28 @@ static const struct file_operations tegra_drm_fops = { .llseek = noop_llseek, }; -static struct drm_crtc *tegra_crtc_from_pipe(struct drm_device *drm, int pipe) +static struct drm_crtc *tegra_crtc_from_pipe(struct drm_device *drm, + unsigned int pipe) { struct drm_crtc *crtc; list_for_each_entry(crtc, &drm->mode_config.crtc_list, head) { - struct tegra_dc *dc = to_tegra_dc(crtc); - - if (dc->pipe == pipe) + if (pipe == drm_crtc_index(crtc)) return crtc; } return NULL; } -static u32 tegra_drm_get_vblank_counter(struct drm_device *dev, int crtc) +static u32 tegra_drm_get_vblank_counter(struct drm_device *drm, int pipe) { + struct drm_crtc *crtc = tegra_crtc_from_pipe(drm, pipe); + + if (!crtc) + return 0; + /* TODO: implement real hardware counter using syncpoints */ - return drm_vblank_count(dev, crtc); + return drm_crtc_vblank_count(crtc); } static int tegra_drm_enable_vblank(struct drm_device *drm, int pipe) diff --git a/drivers/gpu/drm/tegra/gem.c b/drivers/gpu/drm/tegra/gem.c index da32086cbeaf..8777b7f75791 100644 --- a/drivers/gpu/drm/tegra/gem.c +++ b/drivers/gpu/drm/tegra/gem.c @@ -216,32 +216,58 @@ static void tegra_bo_free(struct drm_device *drm, struct tegra_bo *bo) } } -static int tegra_bo_get_pages(struct drm_device *drm, struct tegra_bo *bo, - size_t size) +static int tegra_bo_get_pages(struct drm_device *drm, struct tegra_bo *bo) { + struct scatterlist *s; + struct sg_table *sgt; + unsigned int i; + bo->pages = drm_gem_get_pages(&bo->gem); if (IS_ERR(bo->pages)) return PTR_ERR(bo->pages); - bo->num_pages = size >> PAGE_SHIFT; - - bo->sgt = drm_prime_pages_to_sg(bo->pages, bo->num_pages); - if (IS_ERR(bo->sgt)) { - drm_gem_put_pages(&bo->gem, bo->pages, false, false); - return PTR_ERR(bo->sgt); + bo->num_pages = bo->gem.size >> PAGE_SHIFT; + + sgt = drm_prime_pages_to_sg(bo->pages, bo->num_pages); + if (IS_ERR(sgt)) + goto put_pages; + + /* + * Fake up the SG table so that dma_map_sg() can be used to flush the + * pages associated with it. Note that this relies on the fact that + * the DMA API doesn't hook into IOMMU on Tegra, therefore mapping is + * only cache maintenance. + * + * TODO: Replace this by drm_clflash_sg() once it can be implemented + * without relying on symbols that are not exported. + */ + for_each_sg(sgt->sgl, s, sgt->nents, i) + sg_dma_address(s) = sg_phys(s); + + if (dma_map_sg(drm->dev, sgt->sgl, sgt->nents, DMA_TO_DEVICE) == 0) { + sgt = ERR_PTR(-ENOMEM); + goto release_sgt; } + bo->sgt = sgt; + return 0; + +release_sgt: + sg_free_table(sgt); + kfree(sgt); +put_pages: + drm_gem_put_pages(&bo->gem, bo->pages, false, false); + return PTR_ERR(sgt); } -static int tegra_bo_alloc(struct drm_device *drm, struct tegra_bo *bo, - size_t size) +static int tegra_bo_alloc(struct drm_device *drm, struct tegra_bo *bo) { struct tegra_drm *tegra = drm->dev_private; int err; if (tegra->domain) { - err = tegra_bo_get_pages(drm, bo, size); + err = tegra_bo_get_pages(drm, bo); if (err < 0) return err; @@ -251,6 +277,8 @@ static int tegra_bo_alloc(struct drm_device *drm, struct tegra_bo *bo, return err; } } else { + size_t size = bo->gem.size; + bo->vaddr = dma_alloc_writecombine(drm->dev, size, &bo->paddr, GFP_KERNEL | __GFP_NOWARN); if (!bo->vaddr) { @@ -274,7 +302,7 @@ struct tegra_bo *tegra_bo_create(struct drm_device *drm, size_t size, if (IS_ERR(bo)) return bo; - err = tegra_bo_alloc(drm, bo, size); + err = tegra_bo_alloc(drm, bo); if (err < 0) goto release; diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.c b/drivers/gpu/drm/tilcdc/tilcdc_drv.c index d56d3f8b8d6b..095fca91525c 100644 --- a/drivers/gpu/drm/tilcdc/tilcdc_drv.c +++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.c @@ -644,7 +644,6 @@ static struct platform_driver tilcdc_platform_driver = { .probe = tilcdc_pdev_probe, .remove = tilcdc_pdev_remove, .driver = { - .owner = THIS_MODULE, .name = "tilcdc", .pm = &tilcdc_pm_ops, .of_match_table = tilcdc_of_match, diff --git a/drivers/gpu/drm/ttm/ttm_execbuf_util.c b/drivers/gpu/drm/ttm/ttm_execbuf_util.c index 8ce508e76208..3820ae97a030 100644 --- a/drivers/gpu/drm/ttm/ttm_execbuf_util.c +++ b/drivers/gpu/drm/ttm/ttm_execbuf_util.c @@ -93,7 +93,8 @@ EXPORT_SYMBOL(ttm_eu_backoff_reservation); */ int ttm_eu_reserve_buffers(struct ww_acquire_ctx *ticket, - struct list_head *list, bool intr) + struct list_head *list, bool intr, + struct list_head *dups) { struct ttm_bo_global *glob; struct ttm_validate_buffer *entry; @@ -117,6 +118,13 @@ int ttm_eu_reserve_buffers(struct ww_acquire_ctx *ticket, __ttm_bo_unreserve(bo); ret = -EBUSY; + + } else if (ret == -EALREADY && dups) { + struct ttm_validate_buffer *safe = entry; + entry = list_prev_entry(entry, head); + list_del(&safe->head); + list_add(&safe->head, dups); + continue; } if (!ret) { diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index db7621828bc7..7b5d22110f25 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -1062,8 +1062,12 @@ static long vmw_generic_ioctl(struct file *filp, unsigned int cmd, vmaster = vmw_master_check(dev, file_priv, flags); if (unlikely(IS_ERR(vmaster))) { - DRM_INFO("IOCTL ERROR %d\n", nr); - return PTR_ERR(vmaster); + ret = PTR_ERR(vmaster); + + if (ret != -ERESTARTSYS) + DRM_INFO("IOCTL ERROR Command %d, Error %ld.\n", + nr, ret); + return ret; } ret = ioctl_func(filp, cmd, arg); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 596cd6dafd33..33176d05db35 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -2487,7 +2487,8 @@ int vmw_execbuf_process(struct drm_file *file_priv, if (unlikely(ret != 0)) goto out_err_nores; - ret = ttm_eu_reserve_buffers(&ticket, &sw_context->validate_nodes, true); + ret = ttm_eu_reserve_buffers(&ticket, &sw_context->validate_nodes, + true, NULL); if (unlikely(ret != 0)) goto out_err; @@ -2677,7 +2678,8 @@ void __vmw_execbuf_release_pinned_bo(struct vmw_private *dev_priv, query_val.shared = false; list_add_tail(&query_val.head, &validate_list); - ret = ttm_eu_reserve_buffers(&ticket, &validate_list, false); + ret = ttm_eu_reserve_buffers(&ticket, &validate_list, + false, NULL); if (unlikely(ret != 0)) { vmw_execbuf_unpin_panic(dev_priv); goto out_no_reserve; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c index 197164fd7803..b7594cb758af 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c @@ -545,35 +545,19 @@ void vmw_fence_obj_flush(struct vmw_fence_obj *fence) static void vmw_fence_destroy(struct vmw_fence_obj *fence) { - struct vmw_fence_manager *fman = fman_from_fence(fence); - fence_free(&fence->base); - - /* - * Free kernel space accounting. - */ - ttm_mem_global_free(vmw_mem_glob(fman->dev_priv), - fman->fence_size); } int vmw_fence_create(struct vmw_fence_manager *fman, uint32_t seqno, struct vmw_fence_obj **p_fence) { - struct ttm_mem_global *mem_glob = vmw_mem_glob(fman->dev_priv); struct vmw_fence_obj *fence; int ret; - ret = ttm_mem_global_alloc(mem_glob, fman->fence_size, - false, false); - if (unlikely(ret != 0)) - return ret; - fence = kzalloc(sizeof(*fence), GFP_KERNEL); - if (unlikely(fence == NULL)) { - ret = -ENOMEM; - goto out_no_object; - } + if (unlikely(fence == NULL)) + return -ENOMEM; ret = vmw_fence_obj_init(fman, fence, seqno, vmw_fence_destroy); @@ -585,8 +569,6 @@ int vmw_fence_create(struct vmw_fence_manager *fman, out_err_init: kfree(fence); -out_no_object: - ttm_mem_global_free(mem_glob, fman->fence_size); return ret; } @@ -1105,6 +1087,8 @@ static int vmw_event_fence_action_create(struct drm_file *file_priv, if (ret != 0) goto out_no_queue; + return 0; + out_no_queue: event->base.destroy(&event->base); out_no_event: @@ -1180,17 +1164,10 @@ int vmw_fence_event_ioctl(struct drm_device *dev, void *data, BUG_ON(fence == NULL); - if (arg->flags & DRM_VMW_FE_FLAG_REQ_TIME) - ret = vmw_event_fence_action_create(file_priv, fence, - arg->flags, - arg->user_data, - true); - else - ret = vmw_event_fence_action_create(file_priv, fence, - arg->flags, - arg->user_data, - true); - + ret = vmw_event_fence_action_create(file_priv, fence, + arg->flags, + arg->user_data, + true); if (unlikely(ret != 0)) { if (ret != -ERESTARTSYS) DRM_ERROR("Failed to attach event to fence.\n"); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c index 026de7cea0f6..210ef15b1d09 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c @@ -1222,7 +1222,7 @@ vmw_resource_check_buffer(struct vmw_resource *res, val_buf->bo = ttm_bo_reference(&res->backup->base); val_buf->shared = false; list_add_tail(&val_buf->head, &val_list); - ret = ttm_eu_reserve_buffers(NULL, &val_list, interruptible); + ret = ttm_eu_reserve_buffers(NULL, &val_list, interruptible, NULL); if (unlikely(ret != 0)) goto out_no_reserve; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c index 8719fb3cccc9..6a4584a43aa6 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c @@ -198,7 +198,7 @@ static int vmw_gb_shader_bind(struct vmw_resource *res, cmd->header.size = sizeof(cmd->body); cmd->body.shid = res->id; cmd->body.mobid = bo->mem.start; - cmd->body.offsetInBytes = 0; + cmd->body.offsetInBytes = res->backup_offset; res->backup_dirty = false; vmw_fifo_commit(dev_priv, sizeof(*cmd)); diff --git a/drivers/gpu/ipu-v3/ipu-csi.c b/drivers/gpu/ipu-v3/ipu-csi.c index d6f56471bd2a..752cdd2da89a 100644 --- a/drivers/gpu/ipu-v3/ipu-csi.c +++ b/drivers/gpu/ipu-v3/ipu-csi.c @@ -227,83 +227,83 @@ static int ipu_csi_set_testgen_mclk(struct ipu_csi *csi, u32 pixel_clk, static int mbus_code_to_bus_cfg(struct ipu_csi_bus_config *cfg, u32 mbus_code) { switch (mbus_code) { - case V4L2_MBUS_FMT_BGR565_2X8_BE: - case V4L2_MBUS_FMT_BGR565_2X8_LE: - case V4L2_MBUS_FMT_RGB565_2X8_BE: - case V4L2_MBUS_FMT_RGB565_2X8_LE: + case MEDIA_BUS_FMT_BGR565_2X8_BE: + case MEDIA_BUS_FMT_BGR565_2X8_LE: + case MEDIA_BUS_FMT_RGB565_2X8_BE: + case MEDIA_BUS_FMT_RGB565_2X8_LE: cfg->data_fmt = CSI_SENS_CONF_DATA_FMT_RGB565; cfg->mipi_dt = MIPI_DT_RGB565; cfg->data_width = IPU_CSI_DATA_WIDTH_8; break; - case V4L2_MBUS_FMT_RGB444_2X8_PADHI_BE: - case V4L2_MBUS_FMT_RGB444_2X8_PADHI_LE: + case MEDIA_BUS_FMT_RGB444_2X8_PADHI_BE: + case MEDIA_BUS_FMT_RGB444_2X8_PADHI_LE: cfg->data_fmt = CSI_SENS_CONF_DATA_FMT_RGB444; cfg->mipi_dt = MIPI_DT_RGB444; cfg->data_width = IPU_CSI_DATA_WIDTH_8; break; - case V4L2_MBUS_FMT_RGB555_2X8_PADHI_BE: - case V4L2_MBUS_FMT_RGB555_2X8_PADHI_LE: + case MEDIA_BUS_FMT_RGB555_2X8_PADHI_BE: + case MEDIA_BUS_FMT_RGB555_2X8_PADHI_LE: cfg->data_fmt = CSI_SENS_CONF_DATA_FMT_RGB555; cfg->mipi_dt = MIPI_DT_RGB555; cfg->data_width = IPU_CSI_DATA_WIDTH_8; break; - case V4L2_MBUS_FMT_UYVY8_2X8: + case MEDIA_BUS_FMT_UYVY8_2X8: cfg->data_fmt = CSI_SENS_CONF_DATA_FMT_YUV422_UYVY; cfg->mipi_dt = MIPI_DT_YUV422; cfg->data_width = IPU_CSI_DATA_WIDTH_8; break; - case V4L2_MBUS_FMT_YUYV8_2X8: + case MEDIA_BUS_FMT_YUYV8_2X8: cfg->data_fmt = CSI_SENS_CONF_DATA_FMT_YUV422_YUYV; cfg->mipi_dt = MIPI_DT_YUV422; cfg->data_width = IPU_CSI_DATA_WIDTH_8; break; - case V4L2_MBUS_FMT_UYVY8_1X16: + case MEDIA_BUS_FMT_UYVY8_1X16: cfg->data_fmt = CSI_SENS_CONF_DATA_FMT_YUV422_UYVY; cfg->mipi_dt = MIPI_DT_YUV422; cfg->data_width = IPU_CSI_DATA_WIDTH_16; break; - case V4L2_MBUS_FMT_YUYV8_1X16: + case MEDIA_BUS_FMT_YUYV8_1X16: cfg->data_fmt = CSI_SENS_CONF_DATA_FMT_YUV422_YUYV; cfg->mipi_dt = MIPI_DT_YUV422; cfg->data_width = IPU_CSI_DATA_WIDTH_16; break; - case V4L2_MBUS_FMT_SBGGR8_1X8: - case V4L2_MBUS_FMT_SGBRG8_1X8: - case V4L2_MBUS_FMT_SGRBG8_1X8: - case V4L2_MBUS_FMT_SRGGB8_1X8: + case MEDIA_BUS_FMT_SBGGR8_1X8: + case MEDIA_BUS_FMT_SGBRG8_1X8: + case MEDIA_BUS_FMT_SGRBG8_1X8: + case MEDIA_BUS_FMT_SRGGB8_1X8: cfg->data_fmt = CSI_SENS_CONF_DATA_FMT_BAYER; cfg->mipi_dt = MIPI_DT_RAW8; cfg->data_width = IPU_CSI_DATA_WIDTH_8; break; - case V4L2_MBUS_FMT_SBGGR10_DPCM8_1X8: - case V4L2_MBUS_FMT_SGBRG10_DPCM8_1X8: - case V4L2_MBUS_FMT_SGRBG10_DPCM8_1X8: - case V4L2_MBUS_FMT_SRGGB10_DPCM8_1X8: - case V4L2_MBUS_FMT_SBGGR10_2X8_PADHI_BE: - case V4L2_MBUS_FMT_SBGGR10_2X8_PADHI_LE: - case V4L2_MBUS_FMT_SBGGR10_2X8_PADLO_BE: - case V4L2_MBUS_FMT_SBGGR10_2X8_PADLO_LE: + case MEDIA_BUS_FMT_SBGGR10_DPCM8_1X8: + case MEDIA_BUS_FMT_SGBRG10_DPCM8_1X8: + case MEDIA_BUS_FMT_SGRBG10_DPCM8_1X8: + case MEDIA_BUS_FMT_SRGGB10_DPCM8_1X8: + case MEDIA_BUS_FMT_SBGGR10_2X8_PADHI_BE: + case MEDIA_BUS_FMT_SBGGR10_2X8_PADHI_LE: + case MEDIA_BUS_FMT_SBGGR10_2X8_PADLO_BE: + case MEDIA_BUS_FMT_SBGGR10_2X8_PADLO_LE: cfg->data_fmt = CSI_SENS_CONF_DATA_FMT_BAYER; cfg->mipi_dt = MIPI_DT_RAW10; cfg->data_width = IPU_CSI_DATA_WIDTH_8; break; - case V4L2_MBUS_FMT_SBGGR10_1X10: - case V4L2_MBUS_FMT_SGBRG10_1X10: - case V4L2_MBUS_FMT_SGRBG10_1X10: - case V4L2_MBUS_FMT_SRGGB10_1X10: + case MEDIA_BUS_FMT_SBGGR10_1X10: + case MEDIA_BUS_FMT_SGBRG10_1X10: + case MEDIA_BUS_FMT_SGRBG10_1X10: + case MEDIA_BUS_FMT_SRGGB10_1X10: cfg->data_fmt = CSI_SENS_CONF_DATA_FMT_BAYER; cfg->mipi_dt = MIPI_DT_RAW10; cfg->data_width = IPU_CSI_DATA_WIDTH_10; break; - case V4L2_MBUS_FMT_SBGGR12_1X12: - case V4L2_MBUS_FMT_SGBRG12_1X12: - case V4L2_MBUS_FMT_SGRBG12_1X12: - case V4L2_MBUS_FMT_SRGGB12_1X12: + case MEDIA_BUS_FMT_SBGGR12_1X12: + case MEDIA_BUS_FMT_SGBRG12_1X12: + case MEDIA_BUS_FMT_SGRBG12_1X12: + case MEDIA_BUS_FMT_SRGGB12_1X12: cfg->data_fmt = CSI_SENS_CONF_DATA_FMT_BAYER; cfg->mipi_dt = MIPI_DT_RAW12; cfg->data_width = IPU_CSI_DATA_WIDTH_12; break; - case V4L2_MBUS_FMT_JPEG_1X8: + case MEDIA_BUS_FMT_JPEG_1X8: /* TODO */ cfg->data_fmt = CSI_SENS_CONF_DATA_FMT_JPEG; cfg->mipi_dt = MIPI_DT_RAW8; |