summaryrefslogtreecommitdiff
path: root/drivers/firmware/qemu_fw_cfg.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/firmware/qemu_fw_cfg.c')
-rw-r--r--drivers/firmware/qemu_fw_cfg.c291
1 files changed, 230 insertions, 61 deletions
diff --git a/drivers/firmware/qemu_fw_cfg.c b/drivers/firmware/qemu_fw_cfg.c
index a41b572eeeb1..14fedbeca724 100644
--- a/drivers/firmware/qemu_fw_cfg.c
+++ b/drivers/firmware/qemu_fw_cfg.c
@@ -10,20 +10,21 @@
* and select subsets of aarch64), a Device Tree node (on arm), or using
* a kernel module (or command line) parameter with the following syntax:
*
- * [qemu_fw_cfg.]ioport=<size>@<base>[:<ctrl_off>:<data_off>]
+ * [qemu_fw_cfg.]ioport=<size>@<base>[:<ctrl_off>:<data_off>[:<dma_off>]]
* or
- * [qemu_fw_cfg.]mmio=<size>@<base>[:<ctrl_off>:<data_off>]
+ * [qemu_fw_cfg.]mmio=<size>@<base>[:<ctrl_off>:<data_off>[:<dma_off>]]
*
* where:
* <size> := size of ioport or mmio range
* <base> := physical base address of ioport or mmio range
* <ctrl_off> := (optional) offset of control register
* <data_off> := (optional) offset of data register
+ * <dma_off> := (optional) offset of dma register
*
* e.g.:
- * qemu_fw_cfg.ioport=2@0x510:0:1 (the default on x86)
+ * qemu_fw_cfg.ioport=12@0x510:0:1:4 (the default on x86)
* or
- * qemu_fw_cfg.mmio=0xA@0x9020000:8:0 (the default on arm)
+ * qemu_fw_cfg.mmio=16@0x9020000:8:0:16 (the default on arm)
*/
#include <linux/module.h>
@@ -32,29 +33,17 @@
#include <linux/slab.h>
#include <linux/io.h>
#include <linux/ioport.h>
+#include <uapi/linux/qemu_fw_cfg.h>
+#include <linux/delay.h>
+#include <linux/crash_dump.h>
+#include <linux/crash_core.h>
MODULE_AUTHOR("Gabriel L. Somlo <somlo@cmu.edu>");
MODULE_DESCRIPTION("QEMU fw_cfg sysfs support");
MODULE_LICENSE("GPL");
-/* selector key values for "well-known" fw_cfg entries */
-#define FW_CFG_SIGNATURE 0x00
-#define FW_CFG_ID 0x01
-#define FW_CFG_FILE_DIR 0x19
-
-/* size in bytes of fw_cfg signature */
-#define FW_CFG_SIG_SIZE 4
-
-/* fw_cfg "file name" is up to 56 characters (including terminating nul) */
-#define FW_CFG_MAX_FILE_PATH 56
-
-/* fw_cfg file directory entry type */
-struct fw_cfg_file {
- u32 size;
- u16 select;
- u16 reserved;
- char name[FW_CFG_MAX_FILE_PATH];
-};
+/* fw_cfg revision attribute, in /sys/firmware/qemu_fw_cfg top-level dir. */
+static u32 fw_cfg_rev;
/* fw_cfg device i/o register addresses */
static bool fw_cfg_is_mmio;
@@ -63,19 +52,83 @@ static resource_size_t fw_cfg_p_size;
static void __iomem *fw_cfg_dev_base;
static void __iomem *fw_cfg_reg_ctrl;
static void __iomem *fw_cfg_reg_data;
+static void __iomem *fw_cfg_reg_dma;
/* atomic access to fw_cfg device (potentially slow i/o, so using mutex) */
static DEFINE_MUTEX(fw_cfg_dev_lock);
/* pick appropriate endianness for selector key */
-static inline u16 fw_cfg_sel_endianness(u16 key)
+static void fw_cfg_sel_endianness(u16 key)
+{
+ if (fw_cfg_is_mmio)
+ iowrite16be(key, fw_cfg_reg_ctrl);
+ else
+ iowrite16(key, fw_cfg_reg_ctrl);
+}
+
+#ifdef CONFIG_CRASH_CORE
+static inline bool fw_cfg_dma_enabled(void)
+{
+ return (fw_cfg_rev & FW_CFG_VERSION_DMA) && fw_cfg_reg_dma;
+}
+
+/* qemu fw_cfg device is sync today, but spec says it may become async */
+static void fw_cfg_wait_for_control(struct fw_cfg_dma_access *d)
{
- return fw_cfg_is_mmio ? cpu_to_be16(key) : cpu_to_le16(key);
+ for (;;) {
+ u32 ctrl = be32_to_cpu(READ_ONCE(d->control));
+
+ /* do not reorder the read to d->control */
+ rmb();
+ if ((ctrl & ~FW_CFG_DMA_CTL_ERROR) == 0)
+ return;
+
+ cpu_relax();
+ }
+}
+
+static ssize_t fw_cfg_dma_transfer(void *address, u32 length, u32 control)
+{
+ phys_addr_t dma;
+ struct fw_cfg_dma_access *d = NULL;
+ ssize_t ret = length;
+
+ d = kmalloc(sizeof(*d), GFP_KERNEL);
+ if (!d) {
+ ret = -ENOMEM;
+ goto end;
+ }
+
+ /* fw_cfg device does not need IOMMU protection, so use physical addresses */
+ *d = (struct fw_cfg_dma_access) {
+ .address = cpu_to_be64(address ? virt_to_phys(address) : 0),
+ .length = cpu_to_be32(length),
+ .control = cpu_to_be32(control)
+ };
+
+ dma = virt_to_phys(d);
+
+ iowrite32be((u64)dma >> 32, fw_cfg_reg_dma);
+ /* force memory to sync before notifying device via MMIO */
+ wmb();
+ iowrite32be(dma, fw_cfg_reg_dma + 4);
+
+ fw_cfg_wait_for_control(d);
+
+ if (be32_to_cpu(READ_ONCE(d->control)) & FW_CFG_DMA_CTL_ERROR) {
+ ret = -EIO;
+ }
+
+end:
+ kfree(d);
+
+ return ret;
}
+#endif
/* read chunk of given fw_cfg blob (caller responsible for sanity-check) */
-static inline void fw_cfg_read_blob(u16 key,
- void *buf, loff_t pos, size_t count)
+static ssize_t fw_cfg_read_blob(u16 key,
+ void *buf, loff_t pos, size_t count)
{
u32 glk = -1U;
acpi_status status;
@@ -88,18 +141,60 @@ static inline void fw_cfg_read_blob(u16 key,
/* Should never get here */
WARN(1, "fw_cfg_read_blob: Failed to lock ACPI!\n");
memset(buf, 0, count);
- return;
+ return -EINVAL;
}
mutex_lock(&fw_cfg_dev_lock);
- iowrite16(fw_cfg_sel_endianness(key), fw_cfg_reg_ctrl);
+ fw_cfg_sel_endianness(key);
while (pos-- > 0)
ioread8(fw_cfg_reg_data);
ioread8_rep(fw_cfg_reg_data, buf, count);
mutex_unlock(&fw_cfg_dev_lock);
acpi_release_global_lock(glk);
+ return count;
+}
+
+#ifdef CONFIG_CRASH_CORE
+/* write chunk of given fw_cfg blob (caller responsible for sanity-check) */
+static ssize_t fw_cfg_write_blob(u16 key,
+ void *buf, loff_t pos, size_t count)
+{
+ u32 glk = -1U;
+ acpi_status status;
+ ssize_t ret = count;
+
+ /* If we have ACPI, ensure mutual exclusion against any potential
+ * device access by the firmware, e.g. via AML methods:
+ */
+ status = acpi_acquire_global_lock(ACPI_WAIT_FOREVER, &glk);
+ if (ACPI_FAILURE(status) && status != AE_NOT_CONFIGURED) {
+ /* Should never get here */
+ WARN(1, "%s: Failed to lock ACPI!\n", __func__);
+ return -EINVAL;
+ }
+
+ mutex_lock(&fw_cfg_dev_lock);
+ if (pos == 0) {
+ ret = fw_cfg_dma_transfer(buf, count, key << 16
+ | FW_CFG_DMA_CTL_SELECT
+ | FW_CFG_DMA_CTL_WRITE);
+ } else {
+ fw_cfg_sel_endianness(key);
+ ret = fw_cfg_dma_transfer(NULL, pos, FW_CFG_DMA_CTL_SKIP);
+ if (ret < 0)
+ goto end;
+ ret = fw_cfg_dma_transfer(buf, count, FW_CFG_DMA_CTL_WRITE);
+ }
+
+end:
+ mutex_unlock(&fw_cfg_dev_lock);
+
+ acpi_release_global_lock(glk);
+
+ return ret;
}
+#endif /* CONFIG_CRASH_CORE */
/* clean up fw_cfg device i/o */
static void fw_cfg_io_cleanup(void)
@@ -118,12 +213,14 @@ static void fw_cfg_io_cleanup(void)
# if (defined(CONFIG_ARM) || defined(CONFIG_ARM64))
# define FW_CFG_CTRL_OFF 0x08
# define FW_CFG_DATA_OFF 0x00
+# define FW_CFG_DMA_OFF 0x10
# elif (defined(CONFIG_PPC_PMAC) || defined(CONFIG_SPARC32)) /* ppc/mac,sun4m */
# define FW_CFG_CTRL_OFF 0x00
# define FW_CFG_DATA_OFF 0x02
# elif (defined(CONFIG_X86) || defined(CONFIG_SPARC64)) /* x86, sun4u */
# define FW_CFG_CTRL_OFF 0x00
# define FW_CFG_DATA_OFF 0x01
+# define FW_CFG_DMA_OFF 0x04
# else
# error "QEMU FW_CFG not available on this architecture!"
# endif
@@ -133,7 +230,7 @@ static void fw_cfg_io_cleanup(void)
static int fw_cfg_do_platform_probe(struct platform_device *pdev)
{
char sig[FW_CFG_SIG_SIZE];
- struct resource *range, *ctrl, *data;
+ struct resource *range, *ctrl, *data, *dma;
/* acquire i/o range details */
fw_cfg_is_mmio = false;
@@ -170,6 +267,7 @@ static int fw_cfg_do_platform_probe(struct platform_device *pdev)
/* were custom register offsets provided (e.g. on the command line)? */
ctrl = platform_get_resource_byname(pdev, IORESOURCE_REG, "ctrl");
data = platform_get_resource_byname(pdev, IORESOURCE_REG, "data");
+ dma = platform_get_resource_byname(pdev, IORESOURCE_REG, "dma");
if (ctrl && data) {
fw_cfg_reg_ctrl = fw_cfg_dev_base + ctrl->start;
fw_cfg_reg_data = fw_cfg_dev_base + data->start;
@@ -179,9 +277,17 @@ static int fw_cfg_do_platform_probe(struct platform_device *pdev)
fw_cfg_reg_data = fw_cfg_dev_base + FW_CFG_DATA_OFF;
}
+ if (dma)
+ fw_cfg_reg_dma = fw_cfg_dev_base + dma->start;
+#ifdef FW_CFG_DMA_OFF
+ else
+ fw_cfg_reg_dma = fw_cfg_dev_base + FW_CFG_DMA_OFF;
+#endif
+
/* verify fw_cfg device signature */
- fw_cfg_read_blob(FW_CFG_SIGNATURE, sig, 0, FW_CFG_SIG_SIZE);
- if (memcmp(sig, "QEMU", FW_CFG_SIG_SIZE) != 0) {
+ if (fw_cfg_read_blob(FW_CFG_SIGNATURE, sig,
+ 0, FW_CFG_SIG_SIZE) < 0 ||
+ memcmp(sig, "QEMU", FW_CFG_SIG_SIZE) != 0) {
fw_cfg_io_cleanup();
return -ENODEV;
}
@@ -189,9 +295,6 @@ static int fw_cfg_do_platform_probe(struct platform_device *pdev)
return 0;
}
-/* fw_cfg revision attribute, in /sys/firmware/qemu_fw_cfg top-level dir. */
-static u32 fw_cfg_rev;
-
static ssize_t fw_cfg_showrev(struct kobject *k, struct attribute *a, char *buf)
{
return sprintf(buf, "%u\n", fw_cfg_rev);
@@ -208,10 +311,38 @@ static const struct {
/* fw_cfg_sysfs_entry type */
struct fw_cfg_sysfs_entry {
struct kobject kobj;
- struct fw_cfg_file f;
+ u32 size;
+ u16 select;
+ char name[FW_CFG_MAX_FILE_PATH];
struct list_head list;
};
+#ifdef CONFIG_CRASH_CORE
+static ssize_t fw_cfg_write_vmcoreinfo(const struct fw_cfg_file *f)
+{
+ static struct fw_cfg_vmcoreinfo *data;
+ ssize_t ret;
+
+ data = kmalloc(sizeof(struct fw_cfg_vmcoreinfo), GFP_KERNEL);
+ if (!data)
+ return -ENOMEM;
+
+ *data = (struct fw_cfg_vmcoreinfo) {
+ .guest_format = cpu_to_le16(FW_CFG_VMCOREINFO_FORMAT_ELF),
+ .size = cpu_to_le32(VMCOREINFO_NOTE_SIZE),
+ .paddr = cpu_to_le64(paddr_vmcoreinfo_note())
+ };
+ /* spare ourself reading host format support for now since we
+ * don't know what else to format - host may ignore ours
+ */
+ ret = fw_cfg_write_blob(be16_to_cpu(f->select), data,
+ 0, sizeof(struct fw_cfg_vmcoreinfo));
+
+ kfree(data);
+ return ret;
+}
+#endif /* CONFIG_CRASH_CORE */
+
/* get fw_cfg_sysfs_entry from kobject member */
static inline struct fw_cfg_sysfs_entry *to_entry(struct kobject *kobj)
{
@@ -272,17 +403,17 @@ struct fw_cfg_sysfs_attribute fw_cfg_sysfs_attr_##_attr = { \
static ssize_t fw_cfg_sysfs_show_size(struct fw_cfg_sysfs_entry *e, char *buf)
{
- return sprintf(buf, "%u\n", e->f.size);
+ return sprintf(buf, "%u\n", e->size);
}
static ssize_t fw_cfg_sysfs_show_key(struct fw_cfg_sysfs_entry *e, char *buf)
{
- return sprintf(buf, "%u\n", e->f.select);
+ return sprintf(buf, "%u\n", e->select);
}
static ssize_t fw_cfg_sysfs_show_name(struct fw_cfg_sysfs_entry *e, char *buf)
{
- return sprintf(buf, "%s\n", e->f.name);
+ return sprintf(buf, "%s\n", e->name);
}
static FW_CFG_SYSFS_ATTR(size);
@@ -333,14 +464,13 @@ static ssize_t fw_cfg_sysfs_read_raw(struct file *filp, struct kobject *kobj,
{
struct fw_cfg_sysfs_entry *entry = to_entry(kobj);
- if (pos > entry->f.size)
+ if (pos > entry->size)
return -EINVAL;
- if (count > entry->f.size - pos)
- count = entry->f.size - pos;
+ if (count > entry->size - pos)
+ count = entry->size - pos;
- fw_cfg_read_blob(entry->f.select, buf, pos, count);
- return count;
+ return fw_cfg_read_blob(entry->select, buf, pos, count);
}
static struct bin_attribute fw_cfg_sysfs_attr_raw = {
@@ -452,17 +582,28 @@ static int fw_cfg_register_file(const struct fw_cfg_file *f)
int err;
struct fw_cfg_sysfs_entry *entry;
+#ifdef CONFIG_CRASH_CORE
+ if (fw_cfg_dma_enabled() &&
+ strcmp(f->name, FW_CFG_VMCOREINFO_FILENAME) == 0 &&
+ !is_kdump_kernel()) {
+ if (fw_cfg_write_vmcoreinfo(f) < 0)
+ pr_warn("fw_cfg: failed to write vmcoreinfo");
+ }
+#endif
+
/* allocate new entry */
entry = kzalloc(sizeof(*entry), GFP_KERNEL);
if (!entry)
return -ENOMEM;
/* set file entry information */
- memcpy(&entry->f, f, sizeof(struct fw_cfg_file));
+ entry->size = be32_to_cpu(f->size);
+ entry->select = be16_to_cpu(f->select);
+ memcpy(entry->name, f->name, FW_CFG_MAX_FILE_PATH);
/* register entry under "/sys/firmware/qemu_fw_cfg/by_key/" */
err = kobject_init_and_add(&entry->kobj, &fw_cfg_sysfs_entry_ktype,
- fw_cfg_sel_ko, "%d", entry->f.select);
+ fw_cfg_sel_ko, "%d", entry->select);
if (err)
goto err_register;
@@ -472,7 +613,7 @@ static int fw_cfg_register_file(const struct fw_cfg_file *f)
goto err_add_raw;
/* try adding "/sys/firmware/qemu_fw_cfg/by_name/" symlink */
- fw_cfg_build_symlink(fw_cfg_fname_kset, &entry->kobj, entry->f.name);
+ fw_cfg_build_symlink(fw_cfg_fname_kset, &entry->kobj, entry->name);
/* success, add entry to global cache */
fw_cfg_sysfs_cache_enlist(entry);
@@ -489,28 +630,35 @@ err_register:
static int fw_cfg_register_dir_entries(void)
{
int ret = 0;
+ __be32 files_count;
u32 count, i;
struct fw_cfg_file *dir;
size_t dir_size;
- fw_cfg_read_blob(FW_CFG_FILE_DIR, &count, 0, sizeof(count));
- count = be32_to_cpu(count);
+ ret = fw_cfg_read_blob(FW_CFG_FILE_DIR, &files_count,
+ 0, sizeof(files_count));
+ if (ret < 0)
+ return ret;
+
+ count = be32_to_cpu(files_count);
dir_size = count * sizeof(struct fw_cfg_file);
dir = kmalloc(dir_size, GFP_KERNEL);
if (!dir)
return -ENOMEM;
- fw_cfg_read_blob(FW_CFG_FILE_DIR, dir, sizeof(count), dir_size);
+ ret = fw_cfg_read_blob(FW_CFG_FILE_DIR, dir,
+ sizeof(files_count), dir_size);
+ if (ret < 0)
+ goto end;
for (i = 0; i < count; i++) {
- dir[i].size = be32_to_cpu(dir[i].size);
- dir[i].select = be16_to_cpu(dir[i].select);
ret = fw_cfg_register_file(&dir[i]);
if (ret)
break;
}
+end:
kfree(dir);
return ret;
}
@@ -525,6 +673,7 @@ static inline void fw_cfg_kobj_cleanup(struct kobject *kobj)
static int fw_cfg_sysfs_probe(struct platform_device *pdev)
{
int err;
+ __le32 rev;
/* NOTE: If we supported multiple fw_cfg devices, we'd first create
* a subdirectory named after e.g. pdev->id, then hang per-device
@@ -550,8 +699,11 @@ static int fw_cfg_sysfs_probe(struct platform_device *pdev)
goto err_probe;
/* get revision number, add matching top-level attribute */
- fw_cfg_read_blob(FW_CFG_ID, &fw_cfg_rev, 0, sizeof(fw_cfg_rev));
- fw_cfg_rev = le32_to_cpu(fw_cfg_rev);
+ err = fw_cfg_read_blob(FW_CFG_ID, &rev, 0, sizeof(rev));
+ if (err < 0)
+ goto err_probe;
+
+ fw_cfg_rev = le32_to_cpu(rev);
err = sysfs_create_file(fw_cfg_top_ko, &fw_cfg_rev_attr.attr);
if (err)
goto err_rev;
@@ -597,7 +749,7 @@ MODULE_DEVICE_TABLE(of, fw_cfg_sysfs_mmio_match);
#ifdef CONFIG_ACPI
static const struct acpi_device_id fw_cfg_sysfs_acpi_match[] = {
- { "QEMU0002", },
+ { FW_CFG_ACPI_DEVICE_ID, },
{},
};
MODULE_DEVICE_TABLE(acpi, fw_cfg_sysfs_acpi_match);
@@ -629,6 +781,7 @@ static struct platform_device *fw_cfg_cmdline_dev;
/* use special scanf/printf modifier for phys_addr_t, resource_size_t */
#define PH_ADDR_SCAN_FMT "@%" __PHYS_ADDR_PREFIX "i%n" \
":%" __PHYS_ADDR_PREFIX "i" \
+ ":%" __PHYS_ADDR_PREFIX "i%n" \
":%" __PHYS_ADDR_PREFIX "i%n"
#define PH_ADDR_PR_1_FMT "0x%" __PHYS_ADDR_PREFIX "x@" \
@@ -638,12 +791,15 @@ static struct platform_device *fw_cfg_cmdline_dev;
":%" __PHYS_ADDR_PREFIX "u" \
":%" __PHYS_ADDR_PREFIX "u"
+#define PH_ADDR_PR_4_FMT PH_ADDR_PR_3_FMT \
+ ":%" __PHYS_ADDR_PREFIX "u"
+
static int fw_cfg_cmdline_set(const char *arg, const struct kernel_param *kp)
{
- struct resource res[3] = {};
+ struct resource res[4] = {};
char *str;
phys_addr_t base;
- resource_size_t size, ctrl_off, data_off;
+ resource_size_t size, ctrl_off, data_off, dma_off;
int processed, consumed = 0;
/* only one fw_cfg device can exist system-wide, so if one
@@ -659,19 +815,20 @@ static int fw_cfg_cmdline_set(const char *arg, const struct kernel_param *kp)
/* consume "<size>" portion of command line argument */
size = memparse(arg, &str);
- /* get "@<base>[:<ctrl_off>:<data_off>]" chunks */
+ /* get "@<base>[:<ctrl_off>:<data_off>[:<dma_off>]]" chunks */
processed = sscanf(str, PH_ADDR_SCAN_FMT,
&base, &consumed,
- &ctrl_off, &data_off, &consumed);
+ &ctrl_off, &data_off, &consumed,
+ &dma_off, &consumed);
- /* sscanf() must process precisely 1 or 3 chunks:
+ /* sscanf() must process precisely 1, 3 or 4 chunks:
* <base> is mandatory, optionally followed by <ctrl_off>
- * and <data_off>;
+ * and <data_off>, and <dma_off>;
* there must be no extra characters after the last chunk,
* so str[consumed] must be '\0'.
*/
if (str[consumed] ||
- (processed != 1 && processed != 3))
+ (processed != 1 && processed != 3 && processed != 4))
return -EINVAL;
res[0].start = base;
@@ -688,6 +845,11 @@ static int fw_cfg_cmdline_set(const char *arg, const struct kernel_param *kp)
res[2].start = data_off;
res[2].flags = IORESOURCE_REG;
}
+ if (processed > 3) {
+ res[3].name = "dma";
+ res[3].start = dma_off;
+ res[3].flags = IORESOURCE_REG;
+ }
/* "processed" happens to nicely match the number of resources
* we need to pass in to this platform device.
@@ -720,6 +882,13 @@ static int fw_cfg_cmdline_get(char *buf, const struct kernel_param *kp)
fw_cfg_cmdline_dev->resource[0].start,
fw_cfg_cmdline_dev->resource[1].start,
fw_cfg_cmdline_dev->resource[2].start);
+ case 4:
+ return snprintf(buf, PAGE_SIZE, PH_ADDR_PR_4_FMT,
+ resource_size(&fw_cfg_cmdline_dev->resource[0]),
+ fw_cfg_cmdline_dev->resource[0].start,
+ fw_cfg_cmdline_dev->resource[1].start,
+ fw_cfg_cmdline_dev->resource[2].start,
+ fw_cfg_cmdline_dev->resource[3].start);
}
/* Should never get here */
generated by cgit v1.2.3 (git 2.39.0) at 2025-03-06 19:26:51 +0300