| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
The uverbs CQ creation UAPI allows users to supply their own umem when
creating a CQ. Update mlx4 to support this model while preserving compatibility
with the legacy interface that allocates umem internally.
Link: https://patch.msgid.link/20260213-refactor-umem-v1-13-f3be85847922@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
|
|
Inline the mlx4_ib_get_cq_umem helper function into its two call sites
(mlx4_ib_create_cq and mlx4_alloc_resize_umem) to prepare for the
transition to modern CQ creation interface.
Link: https://patch.msgid.link/20260213-refactor-umem-v1-12-f3be85847922@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
|
|
The uverbs CQ creation UAPI allows users to supply their own umem for a CQ.
Update mlx5 to support this workflow while preserving support for creating
umem through the legacy interface.
Link: https://patch.msgid.link/20260213-refactor-umem-v1-11-f3be85847922@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
|
|
There is no need to maintain separate, nearly empty create_flags and
private_flags fields. Unifying them reduces memory usage.
Link: https://patch.msgid.link/20260213-refactor-umem-v1-10-f3be85847922@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
|
|
Since ib_core now handles validation, the device driver no longer needs
to verify that the CQE count is non‑zero.
Link: https://patch.msgid.link/20260213-refactor-umem-v1-9-f3be85847922@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
|
|
All drivers already ensure that the number of CQEs is at least 1.
Add this validation to the core so drivers no longer need to repeat it.
Future patches converting to the .create_user_cq() interface will remove
the per‑driver checks.
Link: https://patch.msgid.link/20260213-refactor-umem-v1-8-f3be85847922@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
|
|
Ensure that .create_cq_umem() and .create_cq() follow the same API
contract, allowing drivers to be gradually migrated to the umem-aware
CQ management flow.
Link: https://patch.msgid.link/20260213-refactor-umem-v1-7-f3be85847922@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
|
|
Align this code with other locations where efa_free_mapped() depends on the
presence of a valid CPU address, which is guaranteed when qp->rq_size != 0.
Link: https://patch.msgid.link/20260213-refactor-umem-v1-6-f3be85847922@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
|
|
In the current implementation, CQ umem is handled both by ib_core and
the driver. ib_core sometimes creates and destroys it, while the driver
also destroys it.
Store the umem in struct ib_cq and ensure that only ib_core manages
its lifetime, relying solely on its internal reference counter.
Link: https://patch.msgid.link/20260213-refactor-umem-v1-5-f3be85847922@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
|
|
To manage UMEM objects at the core level and reuse the existing
ib_destroy_cq*() flow, move the UMEM files to be built together with
ib_core. Attempting to call ib_umem_release() from verbs.c currently
results in the following error:
depmod: ERROR: Cycle detected: ib_core -> ib_uverbs -> ib_core
depmod: ERROR: Found 2 modules in dependency cycles!
verbs.c:(.text+0x250c): undefined reference to `ib_umem_release'
Link: https://patch.msgid.link/20260213-refactor-umem-v1-4-f3be85847922@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
|
|
The DMA iterator logic was mixed into verbs and umem-specific code,
forcing all users to include rdma/ib_umem.h. Move the block iterator
logic into iter.c and rdma/iter.h so that rdma/ib_umem.h and
rdma/ib_verbs.h can be separated in a follow-up patch.
Link: https://patch.msgid.link/20260213-refactor-umem-v1-1-f3be85847922@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
|
|
In ib_umem_dmabuf_get_pinned_with_dma_device(), the call to
ib_umem_dmabuf_map_pages() can fail. If this occurs, the dmabuf
is immediately unpinned but the umem_dmabuf->pinned flag is still
set. Then, when ib_umem_release() is called, it calls
ib_umem_dmabuf_revoke() which will call dma_buf_unpin() again.
Fix this by removing the immediate unpin upon failure and just let
the ib_umem_release/revoke path handle it. This also ensures the
proper unmap-unpin unwind ordering if the dmabuf_map_pages call
happened to fail due to dma_resv_wait_timeout (and therefore has
a non-NULL umem_dmabuf->sgt).
Fixes: 1e4df4a21c5a ("RDMA/umem: Allow pinned dmabuf umem usage")
Signed-off-by: Jacob Moroni <jmoroni@google.com>
Link: https://patch.msgid.link/20260224234153.1207849-1-jmoroni@google.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
|
|
When listening on wildcard addresses we have a global list for the application
layer rdma_cm_id and for any existing device or any device added in future we
try to listen on any wildcard listener.
When the listener has a restricted_node_type we should prevent listening on
devices with a different node type.
While there fix the documentation comment of rdma_restrict_node_type()
to include rdma_resolve_addr() instead of having rdma_bind_addr() twice.
Fixes: a760e80e90f5 ("RDMA/core: introduce rdma_restrict_node_type()")
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Leon Romanovsky <leon@kernel.org>
Cc: Steve French <smfrench@gmail.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: linux-rdma@vger.kernel.org
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Link: https://patch.msgid.link/20260224165951.3582093-2-metze@samba.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
|
|
If stolen memory cannot be allocated for the fbdev because of the
preference for fbc, have an info about that in the log.
v2: log text changed
Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Acked-by: Mika Kahola <mika.kahola@intel.com>
Link: https://patch.msgid.link/20260220170908.201422-7-vinod.govindapillai@intel.com
|
|
If stolen memory cannot be allocated for the fbdev and initial plane
bo because of the preference for fbc, have an info about that in
the log.
v2: log text changed
Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Acked-by: Mika Kahola <mika.kahola@intel.com>
Link: https://patch.msgid.link/20260220170908.201422-6-vinod.govindapillai@intel.com
|
|
Pull the "should we keep the bios fb in stolen?" logic into
into a helper function, same as was done for i915. Gives us
a single place where to tweak the heuristics.
v2: changes related to rebase and consolidated other conditions
for the stolen preference into this single function (Vinod)
v3: avoid including intel_display_core.h (Jani Nikula)
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Acked-by: Mika Kahola <mika.kahola@intel.com>
Link: https://patch.msgid.link/20260220170908.201422-5-vinod.govindapillai@intel.com
|
|
Consolidate the "should we allocate fbdev fb in stolen?"
check into a helper function. Makes it easier to change the
heuristics without having to change so many places.
v2: rebase related changes and consolidate all the prefer
stolen conditions into a single function (Vinod)
v3: avoid including intel_display_core.h (Jani)
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Acked-by: Mika Kahola <mika.kahola@intel.com>
Link: https://patch.msgid.link/20260220170908.201422-4-vinod.govindapillai@intel.com
|
|
Remove the remaining usage of dev_priv in intel_fbdev_fb.c
and use i915 instead.
Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://patch.msgid.link/20260220170908.201422-3-vinod.govindapillai@intel.com
|
|
Looks like stolen->size is in bytes, not pages. Remove the
bogus PAGE_SHIFT stuff.
Also for some random reason xe rejects the FB if it takes up
exactly half of stolen, whereas i915 allows it to be used
in that case. Adjust xe to follow the i915 rule for consistency.
v2: rebase related updates
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Acked-by: Mika Kahola <mika.kahola@intel.com>
Link: https://patch.msgid.link/20260220170908.201422-2-vinod.govindapillai@intel.com
|
|
Sync with v7.0-rc1 which contains a few treewide changes affecting i915.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
|
|
On QP creation we validate the requested max inline size is supported by
the device. Use the new extended max inline size instead of the old one
to support actual max inline available.
Reviewed-by: Michael Margolin <mrgolin@amazon.com>
Signed-off-by: Yonatan Nachum <ynachum@amazon.com>
Link: https://patch.msgid.link/20260217112304.36849-4-ynachum@amazon.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
|
|
Add new extended max inline query and report the new value to userspace.
Reviewed-by: Firas Jahjah <firasj@amazon.com>
Reviewed-by: Michael Margolin <mrgolin@amazon.com>
Signed-off-by: Yonatan Nachum <ynachum@amazon.com>
Link: https://patch.msgid.link/20260217112304.36849-3-ynachum@amazon.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
|
|
As preparation for adding a second queue attributes query, change the
name of the existing queue attributes.
Reviewed-by: Michael Margolin <mrgolin@amazon.com>
Signed-off-by: Yonatan Nachum <ynachum@amazon.com>
Link: https://patch.msgid.link/20260217112304.36849-2-ynachum@amazon.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
|
|
Once a device is successfully registered, set the "registered" flag to
true. After that point, all failures jump to the out_unlock label to
unwind the flow via coresight_unregister().
Since failure handling is unified, the comment about resource release
for the etm_perf_add_symlink_sink() failure is no need, remove it.
Signed-off-by: Leo Yan <leo.yan@arm.com>
Reviewed-by: James Clark <james.clark@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20260209-arm_coresight_refactor_dev_register-v4-8-62d6042f76f7@arm.com
|
|
Separate the failure handling path from the successful case. Use the
'out_unlock' label only for failure handling.
Reviewed-by: James Clark <james.clark@linaro.org>
Signed-off-by: Leo Yan <leo.yan@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20260209-arm_coresight_refactor_dev_register-v4-7-62d6042f76f7@arm.com
|
|
Move the sink device type checks into etm_perf_add_symlink_sink(), and
return -EOPNOTSUPP for unsupported devices. This simplifies the
registration flow to invoke etm_perf_add_symlink_sink() unconditionally.
Reviewed-by: James Clark <james.clark@linaro.org>
Signed-off-by: Leo Yan <leo.yan@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20260209-arm_coresight_refactor_dev_register-v4-6-62d6042f76f7@arm.com
|
|
Move the sysfs connection group cleanup into coresight_remove_conns(),
so that the driver removes connections and related sysfs resources in
one go.
As side effect, the csdev argument to coresight_release_platform_data()
is no longer needed; adjust the code for this.
Reviewed-by: James Clark <james.clark@linaro.org>
Signed-off-by: Leo Yan <leo.yan@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20260209-arm_coresight_refactor_dev_register-v4-5-62d6042f76f7@arm.com
|
|
To use a central place for releasing connections, move the output
connection sysfs link cleanup into coresight_remove_conns().
Also update the comments accordingly.
Reviewed-by: James Clark <james.clark@linaro.org>
Signed-off-by: Leo Yan <leo.yan@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20260209-arm_coresight_refactor_dev_register-v4-4-62d6042f76f7@arm.com
|
|
The device registration is protected by CoreSight mutex to ensure the
atomic operations when adding a device onto bus. One the other hand,
the locking is absent when unregister a device.
Use mutex to ensure atomicity on device unregistration. During
unregistration, unbinding the associated CTI device is not included in
the locking region, as CTI has its own locking mechanism.
Fixes: 8c1d3f79d9ca ("coresight: core: Fix coresight device probe failure issue")
Reviewed-by: James Clark <james.clark@linaro.org>
Signed-off-by: Leo Yan <leo.yan@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20260209-arm_coresight_refactor_dev_register-v4-3-62d6042f76f7@arm.com
|
|
The parent device's reference count is incremented before allocating the
sink ID map. If the allocation fails, the reference count is not
decremented, preventing proper cleanup.
Fix this by incrementing the reference count only after the sink ID map
is successfully allocated.
Fixes: 5ad628a76176 ("coresight: Use per-sink trace ID maps for Perf sessions")
Reviewed-by: James Clark <james.clark@linaro.org>
Signed-off-by: Leo Yan <leo.yan@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20260209-arm_coresight_refactor_dev_register-v4-2-62d6042f76f7@arm.com
|
|
The memory leak detector reports:
echo clear > /sys/kernel/debug/kmemleak
modprobe coresight_funnel
rmmod coresight_funnel
# Scan memory leak and report it
echo scan > /sys/kernel/debug/kmemleak
cat /sys/kernel/debug/kmemleak
unreferenced object 0xffff0008020c7200 (size 64):
comm "modprobe", pid 410, jiffies 4295333721
hex dump (first 32 bytes):
d8 da fe 7e 09 00 ff ff e8 2e ff 7e 09 00 ff ff ...~.......~....
b0 6c ff 7e 09 00 ff ff 30 83 00 7f 09 00 ff ff .l.~....0.......
backtrace (crc 4116a690):
kmemleak_alloc+0xd8/0xf8
__kmalloc_node_track_caller_noprof+0x2c8/0x6f0
krealloc_node_align_noprof+0x13c/0x2c8
coresight_alloc_device_name+0xe4/0x158 [coresight]
0xffffd327ecef8394
0xffffd327ecef85ec
amba_probe+0x118/0x1c8
really_probe+0xc8/0x3f0
__driver_probe_device+0x88/0x190
driver_probe_device+0x44/0x120
__driver_attach+0x100/0x238
bus_for_each_dev+0x84/0xf0
driver_attach+0x2c/0x40
bus_add_driver+0x128/0x258
driver_register+0x64/0x138
__amba_driver_register+0x2c/0x48
The memory leak is caused by not freeing the device list that maintains
device indices.
This device list preserves stable device indices across unbind and
rebind device operations, so it does not share the same lifetime as a
device instances and must only be freed when the module is unloaded.
Some modules do not implement a module exit callback because they are
registered using module_platform_driver(). As a result, the device
list cannot be released during module exit for those modules.
Fix this by moving the device list into the core layer. As a general
solution, instead of maintaining a static list in each driver, drivers
now allocate device lists via coresight_allocate_device_list() and
device indices via coresight_allocate_device_idx().
The list is released only when the core module is unloaded by calling
coresight_release_device_list(), avoiding the leak.
Fixes: 0f5f9b6ba9e1 ("coresight: Use platform agnostic names")
Reviewed-by: James Clark <james.clark@linaro.org>
Signed-off-by: Leo Yan <leo.yan@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20260209-arm_coresight_refactor_dev_register-v4-1-62d6042f76f7@arm.com
|
|
During the device remove process, the device is reset, causing the
configuration registers to go back to their default state, which is
zero. As the driver is checking if the event log support was enabled
before deallocating, it will fail if a reset happened before.
Do not check if the support was enabled, the check for 'idxd->evl'
being valid (only allocated if the HW capability is available) is
enough.
Fixes: 244da66cda35 ("dmaengine: idxd: setup event log configuration")
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Link: https://patch.msgid.link/20260121-idxd-fix-flr-on-kernel-queues-v3-v3-10-7ed70658a9d1@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
It can happen that when the cdev .release() is called, the driver
already called ida_destroy(). Move ida_free() to the _del() path.
We see with DEBUG_KOBJECT_RELEASE enabled and forcing an early PCI
unbind.
Fixes: 04922b7445a1 ("dmaengine: idxd: fix cdev setup and free device lifetime issues")
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Link: https://patch.msgid.link/20260121-idxd-fix-flr-on-kernel-queues-v3-v3-9-7ed70658a9d1@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
idxd_wq_disable_cleanup() which is called from the reset path for a
workqueue, sets the wq type to NONE, which for other parts of the
driver mean that the wq is empty (all its resources were released).
Only set the wq type to NONE after its resources are released.
Fixes: da32b28c95a7 ("dmaengine: idxd: cleanup workqueue config after disabling")
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Link: https://patch.msgid.link/20260121-idxd-fix-flr-on-kernel-queues-v3-v3-8-7ed70658a9d1@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
The workqueue associated with an DSA/IAA device is not released when
the object is freed.
Fixes: 47c16ac27d4c ("dmaengine: idxd: fix idxd conf_dev 'struct device' lifetime")
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Link: https://patch.msgid.link/20260121-idxd-fix-flr-on-kernel-queues-v3-v3-7-7ed70658a9d1@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
When the dmaengine "core" asks the driver to synchronize, send a Drain
operation to the device workqueue, which will wait for the already
submitted operations to finish.
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Link: https://patch.msgid.link/20260121-idxd-fix-flr-on-kernel-queues-v3-v3-6-7ed70658a9d1@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
When used as a dmaengine, the DMA "core" might ask the driver to
terminate all pending requests, when that happens, flush all pending
descriptors.
In this context, flush means removing the requests from the pending
lists, so even if they are completed after, the user is not notified.
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Link: https://patch.msgid.link/20260121-idxd-fix-flr-on-kernel-queues-v3-v3-5-7ed70658a9d1@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
When a Function Level Reset (FLR) happens, terminate the pending
descriptors that were issued by in-kernel users and disable the
interrupts associated with those. They will be re-enabled after FLR
finishes.
idxd_wq_flush_desc() is declared on idxd.h because it's going to be
used in by the DMA backend in a future patch.
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Link: https://patch.msgid.link/20260121-idxd-fix-flr-on-kernel-queues-v3-v3-4-7ed70658a9d1@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
In the case that the first Function Level Reset (FLR) concludes
correctly, but in the second FLR the scratch area for the saved
configuration cannot be allocated, it's possible for a invalid memory
access to happen.
Always set the deallocated scratch area to NULL after FLR completes.
Fixes: 98d187a98903 ("dmaengine: idxd: Enable Function Level Reset (FLR) for halt")
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Link: https://patch.msgid.link/20260121-idxd-fix-flr-on-kernel-queues-v3-v3-3-7ed70658a9d1@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
If reporting errors to the event log is not supported by the hardware,
and an error that causes Function Level Reset (FLR) is received, the
driver will try to restore the event log even if it was not allocated.
Also, only try to free the event log if it was properly allocated.
Fixes: 6078a315aec1 ("dmaengine: idxd: Add idxd_device_config_save() and idxd_device_config_restore() helpers")
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Link: https://patch.msgid.link/20260121-idxd-fix-flr-on-kernel-queues-v3-v3-2-7ed70658a9d1@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
Move the check for IDXD_FLAG_CONFIGURABLE and the locking to "inside"
idxd_device_config(), as this is common to all callers, and the one
that wasn't holding the lock was an error (that was causing the
lockdep warning).
Suggested-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Link: https://patch.msgid.link/20260121-idxd-fix-flr-on-kernel-queues-v3-v3-1-7ed70658a9d1@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
efi_free_boot_services() frees memory occupied by EFI_BOOT_SERVICES_CODE
and EFI_BOOT_SERVICES_DATA using memblock_free_late().
There are two issue with that: memblock_free_late() should be used for
memory allocated with memblock_alloc() while the memory reserved with
memblock_reserve() should be freed with free_reserved_area().
More acutely, with CONFIG_DEFERRED_STRUCT_PAGE_INIT=y
efi_free_boot_services() is called before deferred initialization of the
memory map is complete.
Benjamin Herrenschmidt reports that this causes a leak of ~140MB of
RAM on EC2 t3a.nano instances which only have 512MB or RAM.
If the freed memory resides in the areas that memory map for them is
still uninitialized, they won't be actually freed because
memblock_free_late() calls memblock_free_pages() and the latter skips
uninitialized pages.
Using free_reserved_area() at this point is also problematic because
__free_page() accesses the buddy of the freed page and that again might
end up in uninitialized part of the memory map.
Delaying the entire efi_free_boot_services() could be problematic
because in addition to freeing boot services memory it updates
efi.memmap without any synchronization and that's undesirable late in
boot when there is concurrency.
More robust approach is to only defer freeing of the EFI boot services
memory.
Split efi_free_boot_services() in two. First efi_unmap_boot_services()
collects ranges that should be freed into an array then
efi_free_boot_services() later frees them after deferred init is complete.
Link: https://lore.kernel.org/all/ec2aaef14783869b3be6e3c253b2dcbf67dbc12a.camel@kernel.crashing.org
Fixes: 916f676f8dc0 ("x86, efi: Retain boot service code until after switching to virtual mode")
Cc: <stable@vger.kernel.org>
Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Reviewed-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
|
|
When using MSI (not MSI-X) with multiple IRQs, the MSI data value
must be unique per vector to ensure correct interrupt delivery.
Currently, the driver fails to increment the MSI data per vector,
causing interrupts to be misrouted.
Fix this by caching the base MSI data and adjusting each vector's
data accordingly during IRQ setup.
Fixes: e63d79d1ff04 ("dmaengine: dw-edma: Add Synopsys DesignWare eDMA IP core driver")
Signed-off-by: Shenghui Shi <brody.shi@m2semi.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20260209103726.414-1-brody.shi@m2semi.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
checkpatch.pl --strict reports a CHECK warning in dw-axi-dmac-platform.c:
CHECK: Alignment should match open parenthesis
This warning occurs when multi-line function calls or expressions have
continuation lines that don't properly align with the opening parenthesis
position.
This patch fixes all instances in dw-axi-dmac-platform.c where continuation
lines were indented with an inconsistent number of spaces/tabs that neither
matched the parenthesis column nor followed a standard indent pattern.
Proper alignment improves code readability and maintainability by making
parameter lists visually consistent across the kernel codebase.
Fixes: 1fe20f1b8454 ("dmaengine: Introduce DW AXI DMAC driver")
Fixes: e32634f466a9 ("dma: dw-axi-dmac: support per channel interrupt")
Signed-off-by: Khairul Anuar Romli <karom.9560@gmail.com>
Link: https://patch.msgid.link/20260202060224.12616-2-karom.9560@gmail.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
Interrupt emulation can assert the dw-edma IRQ line without updating the
DONE/ABORT bits. With the shared read/write/common IRQ handlers, the
driver cannot reliably distinguish such an emulated interrupt from a
real one and leaving a level IRQ asserted may wedge the line.
Allocate a dedicated, requestable Linux virtual IRQ (db_irq) for
interrupt emulation and attach an irq_chip whose .irq_ack runs the
core-specific deassert sequence (.ack_emulated_irq()). The physical
dw-edma interrupt handlers raise this virtual IRQ via
generic_handle_irq(), ensuring emulated IRQs are always deasserted.
Export the virtual IRQ number (db_irq) and the doorbell register offset
(db_offset) via struct dw_edma_chip so platform users can expose
interrupt emulation as a doorbell.
Without this, a single interrupt-emulation write can leave the level IRQ
line asserted and cause the generic IRQ layer to disable it.
Signed-off-by: Koichiro Den <den@valinux.co.jp>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20260215152216.3393561-3-den@valinux.co.jp
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
DesignWare eDMA instances support "interrupt emulation", where a
software write can assert the IRQ line without setting the normal
DONE/ABORT status bits.
Introduce core callbacks needed to support this feature:
- .ack_emulated_irq(): core-specific sequence to deassert an emulated
IRQ
- .db_offset(): offset from the DMA register base that is suitable as a
host-writable doorbell target for interrupt emulation
Implement both hooks for the v0 register map. For dw-hdma-v0, provide a
stub .db_offset() returning ~0 until the correct offset is known.
The next patch wires these hooks into the dw-edma IRQ path and exports
the doorbell resources to platform users.
Signed-off-by: Koichiro Den <den@valinux.co.jp>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20260215152216.3393561-2-den@valinux.co.jp
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
clk_disable_unprepare() is safe to call with a NULL clk, the
FSL_EDMA_DRV_HAS_CHCLK check is reduntante. Clean up redundant checks.
Suggested-by: Frank Li <Frank.Li@nxp.com>
Signed-off-by: Zhen Ni <zhen.ni@easystack.cn>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20251014024730.751237-1-zhen.ni@easystack.cn
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
Currently, CTI driver releases resource by deferring
cti_device_release() to the device unregistration:
cti_remove()
`> coresight_unregister()
`> cti_remove_assoc_from_csdev()
`> device_unregister()
`> cti_device_release()
`> mutex_lock(&ect_mutex)
`> release CTI resource
`> mutex_unlock(&ect_mutex)
In the above flow, two different CTI release callbacks are involved:
cti_remove_assoc_from_csdev() and cti_device_release(). The former is
used by a CoreSight device to unbind its associated CTI helper device,
while the latter releases resources for the CTI device itself. Since
there is no dependency between them, it is unnecessary to defer the CTI
resource release until device unregistration.
This commit releases the resources directly in cti_remove() and remove
the injected release callback.
Signed-off-by: Leo Yan <leo.yan@arm.com>
Reviewed-by: Mike Leach <mike.leach@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20260224-arm_coresight_refactor_cti_resource_release-v1-1-ff1b2bca9176@arm.com
|
|
Configure only the requested channel when a fixed channel is specified
to avoid modifying other channels unintentionally.
Fix parameter configuration when a fixed DMA channel is requested on
i.MX9 AON domain and i.MX8QM/QXP/DXL platforms. When a client requests
a fixed channel (e.g., channel 6), the driver traverses channels 0-5
and may unintentionally modify their configuration if they are unused.
This leads to issues such as setting the `is_multi_fifo` flag unexpectedly,
causing memcpy tests to fail when using the dmatest tool.
Only affect edma memcpy test when the channel is fixed.
Fixes: 72f5801a4e2b ("dmaengine: fsl-edma: integrate v3 support")
Signed-off-by: Joy Zou <joy.zou@nxp.com>
Cc: stable@vger.kernel.org
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20250917-b4-edma-chanconf-v1-1-886486e02e91@nxp.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
|
|
dma_fence_array_enable_signaling() runs while holding the array
inline_lock and may add callbacks to underlying fences, which takes
their inline_lock.
Since both locks share the same lockdep class, this valid nesting
triggers a recursive locking warning. Assign a distinct lockdep class
to the array inline_lock so lockdep can correctly model the hierarchy.
Fixes: 5943243914b9 ("dma-buf: use inline lock for the dma-fence-array")
Cc: Christian König <christian.koenig@amd.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Cc: Philipp Stanner <phasta@kernel.org>
Cc: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://patch.msgid.link/20260224183922.2256492-2-matthew.brost@intel.com
|
