A CPU job is a type of job that performs operations that requires CPU
intervention. A timestamp query job is a job that calculates the
query timestamp and updates the query availability by signaling a
syncobj. As V3D doesn't provide any mechanism to obtain a timestamp
from the GPU, it is a job that needs CPU intervention.
So, create a user extension for the CPU job that enables the creation
of a timestamp query job. This user extension will allow the creation of
a CPU job that performs the timestamp query calculation and updates the
timestamp BO with the proper value.
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20231130164420.932823-15-mcanal@igalia.com
A CPU job is a type of job that performs operations that requires CPU
intervention. An indirect CSD job is a job that, when executed in the
queue, will map the indirect buffer, read the dispatch parameters, and
submit a regular dispatch. Therefore, it is a job that needs CPU
intervention.
So, create a user extension for the CPU job that enables the creation
of an indirect CSD. This user extension will allow the creation of a CSD
job linked to a CPU job. The CPU job will wait for the indirect CSD job
dependencies and, once they are signaled, it will update the CSD job
parameters.
Co-developed-by: Melissa Wen <mwen@igalia.com>
Signed-off-by: Melissa Wen <mwen@igalia.com>
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20231130164420.932823-14-mcanal@igalia.com
Currently, v3d_get_extensions() only parses multisync data and assigns
it to the `struct v3d_submit_ext`. But, to implement the CPU job with
user extensions, we want v3d_get_extensions() to be able to parse CPU
job data and assign it to the `struct v3d_cpu_job`.
Therefore, allow the function v3d_get_extensions() to use `struct v3d_cpu_job *`
as a parameter. If the `struct v3d_cpu_job *` is assigned to NULL, it means
that the job is a GPU job and CPU job extensions should be rejected.
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20231130164420.932823-10-mcanal@igalia.com
Create a new type of job, a CPU job. A CPU job is a type of job that
performs operations that requires CPU intervention. The overall idea is
to use user extensions to enable different types of CPU job, allowing the
CPU job to perform different operations according to the type of user
extension. The user extension ID identify the type of CPU job that must
be dealt.
Having a CPU job is interesting for synchronization purposes as a CPU
job has a queue like any other V3D job and can be synchoronized by the
multisync extension.
Signed-off-by: Melissa Wen <mwen@igalia.com>
Co-developed-by: Maíra Canal <mcanal@igalia.com>
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20231130164420.932823-9-mcanal@igalia.com
We want to allow the IOCTLs to allocate the job without initiating it.
This will be useful for the CPU job submission IOCTL, as the CPU job has
the need to use information from the user extensions. Currently, the
user extensions are parsed before the job allocation, making it
impossible to fill the CPU job when parsing the user extensions.
Therefore, decouple the job allocation from the job initiation.
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20231130164420.932823-8-mcanal@igalia.com
Currently, job flow control is implemented simply by limiting the number
of jobs in flight. Therefore, a scheduler is initialized with a credit
limit that corresponds to the number of jobs which can be sent to the
hardware.
This implies that for each job, drivers need to account for the maximum
job size possible in order to not overflow the ring buffer.
However, there are drivers, such as Nouveau, where the job size has a
rather large range. For such drivers it can easily happen that job
submissions not even filling the ring by 1% can block subsequent
submissions, which, in the worst case, can lead to the ring run dry.
In order to overcome this issue, allow for tracking the actual job size
instead of the number of jobs. Therefore, add a field to track a job's
credit count, which represents the number of credits a job contributes
to the scheduler's credit limit.
Signed-off-by: Danilo Krummrich <dakr@redhat.com>
Reviewed-by: Luben Tuikov <ltuikov89@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20231110001638.71750-1-dakr@redhat.com
The previous patch exposed the accumulated amount of active time per
client for each V3D queue. But this doesn't provide a global notion of
the GPU usage.
Therefore, provide the accumulated amount of active time for each V3D
queue (BIN, RENDER, CSD, TFU and CACHE_CLEAN), considering all the jobs
submitted to the queue, independent of the client.
This data is exposed through the sysfs interface, so that if the
interface is queried at two different points of time the usage percentage
of each of the queues can be calculated.
Co-developed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Signed-off-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Acked-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230905213416.1290219-3-mcanal@igalia.com
This patch exposes the accumulated amount of active time per client
through the fdinfo infrastructure. The amount of active time is exposed
for each V3D queue: BIN, RENDER, CSD, TFU and CACHE_CLEAN.
In order to calculate the amount of active time per client, a CPU clock
is used through the function local_clock(). The point where the jobs has
started is marked and is finally compared with the time that the job had
finished.
Moreover, the number of jobs submitted to each queue is also exposed on
fdinfo through the identifier "v3d-jobs-<queue>".
Co-developed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Signed-off-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Acked-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230905213416.1290219-3-mcanal@igalia.com
In Xe, the new Intel GPU driver, a choice has made to have a 1 to 1
mapping between a drm_gpu_scheduler and drm_sched_entity. At first this
seems a bit odd but let us explain the reasoning below.
1. In Xe the submission order from multiple drm_sched_entity is not
guaranteed to be the same completion even if targeting the same hardware
engine. This is because in Xe we have a firmware scheduler, the GuC,
which allowed to reorder, timeslice, and preempt submissions. If a using
shared drm_gpu_scheduler across multiple drm_sched_entity, the TDR falls
apart as the TDR expects submission order == completion order. Using a
dedicated drm_gpu_scheduler per drm_sched_entity solve this problem.
2. In Xe submissions are done via programming a ring buffer (circular
buffer), a drm_gpu_scheduler provides a limit on number of jobs, if the
limit of number jobs is set to RING_SIZE / MAX_SIZE_PER_JOB we get flow
control on the ring for free.
A problem with this design is currently a drm_gpu_scheduler uses a
kthread for submission / job cleanup. This doesn't scale if a large
number of drm_gpu_scheduler are used. To work around the scaling issue,
use a worker rather than kthread for submission / job cleanup.
v2:
- (Rob Clark) Fix msm build
- Pass in run work queue
v3:
- (Boris) don't have loop in worker
v4:
- (Tvrtko) break out submit ready, stop, start helpers into own patch
v5:
- (Boris) default to ordered work queue
v6:
- (Luben / checkpatch) fix alignment in msm_ringbuffer.c
- (Luben) s/drm_sched_submit_queue/drm_sched_wqueue_enqueue
- (Luben) Update comment for drm_sched_wqueue_enqueue
- (Luben) Positive check for submit_wq in drm_sched_init
- (Luben) s/alloc_submit_wq/own_submit_wq
v7:
- (Luben) s/drm_sched_wqueue_enqueue/drm_sched_run_job_queue
v8:
- (Luben) Adjust var names / comments
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Link: https://lore.kernel.org/r/20231031032439.1558703-3-matthew.brost@intel.com
Signed-off-by: Luben Tuikov <ltuikov89@gmail.com>
A proposed update to clang's -Wconstant-logical-operand to warn when the
left hand side is a constant shows the following instance in
nsecs_to_jiffies_timeout() when NSEC_PER_SEC is not a multiple of HZ,
such as CONFIG_HZ=300:
In file included from drivers/gpu/drm/v3d/v3d_debugfs.c:12:
drivers/gpu/drm/v3d/v3d_drv.h:343:24: warning: use of logical '&&' with constant operand [-Wconstant-logical-operand]
343 | if (NSEC_PER_SEC % HZ &&
| ~~~~~~~~~~~~~~~~~ ^
drivers/gpu/drm/v3d/v3d_drv.h:343:24: note: use '&' for a bitwise operation
343 | if (NSEC_PER_SEC % HZ &&
| ^~
| &
drivers/gpu/drm/v3d/v3d_drv.h:343:24: note: remove constant to silence this warning
1 warning generated.
Turn this into an explicit comparison against zero to make the
expression a boolean to make it clear this should be a logical check,
not a bitwise one.
Link: https://reviews.llvm.org/D142609
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Maíra Canal <mairacanal@riseup.net>
Link: https://patchwork.freedesktop.org/patch/msgid/20230718-nsecs_to_jiffies_timeout-constant-logical-operand-v1-1-36ed8fc8faea@kernel.org
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is (mostly) ignored
and this typically results in resource leaks. To improve here there is a
quest to make the remove callback return void. In the first step of this
quest all drivers are converted to .remove_new() which already returns
void.
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20230507162616.1368908-52-u.kleine-koenig@pengutronix.de
When v3d_lookup_bos fails to `allocate validated BO pointers`,
job->bo_count was already set to args->bo_count, but job->bo points to
NULL. In this scenario, we must verify that job->bo is not NULL before
iterating on it to proper clean up a job. Also, drm_gem_object_put
already checks that the object passed is not NULL, doing the job->bo[i]
checker redundant.
Signed-off-by: Melissa Wen <mwen@igalia.com>
Reviewed-by: Maíra Canal <mcanal@igalia.com>
Signed-off-by: Melissa Wen <melissa.srw@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20221205135538.3545051-2-mwen@igalia.com
v3d_perfmon_open_file() instantiates a mutex for a particular file
instance, but it never destroys it by calling mutex_destroy() in
v3d_perfmon_close_file().
Similarly, v3d_perfmon_create_ioctl() instantiates a mutex for a
particular perfmon, but it never destroys it by calling mutex_destroy()
in v3d_perfmon_destroy_ioctl().
So, add the missing mutex_destroy on both cases.
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Melissa Wen <melissa.srw@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20221108175425.39819-3-mcanal@igalia.com
Rename "GEM CMA" helpers to "GEM DMA" helpers - considering the
hierarchy of APIs (mm/cma -> dma -> gem dma) calling them "GEM
DMA" seems to be more applicable.
Besides that, commit e57924d4ae ("drm/doc: Task to rename CMA helpers")
requests to rename the CMA helpers and implies that people seem to be
confused about the naming.
In order to do this renaming the following script was used:
```
#!/bin/bash
DIRS="drivers/gpu include/drm Documentation/gpu"
REGEX_SYM_UPPER="[0-9A-Z_\-]"
REGEX_SYM_LOWER="[0-9a-z_\-]"
REGEX_GREP_UPPER="(${REGEX_SYM_UPPER}*)(GEM)_CMA_(${REGEX_SYM_UPPER}*)"
REGEX_GREP_LOWER="(${REGEX_SYM_LOWER}*)(gem)_cma_(${REGEX_SYM_LOWER}*)"
REGEX_SED_UPPER="s/${REGEX_GREP_UPPER}/\1\2_DMA_\3/g"
REGEX_SED_LOWER="s/${REGEX_GREP_LOWER}/\1\2_dma_\3/g"
# Find all upper case 'CMA' symbols and replace them with 'DMA'.
for ff in $(grep -REHl "${REGEX_GREP_UPPER}" $DIRS)
do
sed -i -E "$REGEX_SED_UPPER" $ff
done
# Find all lower case 'cma' symbols and replace them with 'dma'.
for ff in $(grep -REHl "${REGEX_GREP_LOWER}" $DIRS)
do
sed -i -E "$REGEX_SED_LOWER" $ff
done
# Replace all occurrences of 'CMA' / 'cma' in comments and
# documentation files with 'DMA' / 'dma'.
for ff in $(grep -RiHl " cma " $DIRS)
do
sed -i -E "s/ cma / dma /g" $ff
sed -i -E "s/ CMA / DMA /g" $ff
done
# Rename all 'cma_obj's to 'dma_obj'.
for ff in $(grep -RiHl "cma_obj" $DIRS)
do
sed -i -E "s/cma_obj/dma_obj/g" $ff
done
```
Only a few more manual modifications were needed, e.g. reverting the
following modifications in some DRM Kconfig files
- select CMA if HAVE_DMA_CONTIGUOUS
+ select DMA if HAVE_DMA_CONTIGUOUS
as well as manually picking the occurrences of 'CMA'/'cma' in comments and
documentation which relate to "GEM CMA", but not "FB CMA".
Also drivers/gpu/drm/Makefile was fixed up manually after renaming
drm_gem_cma_helper.c to drm_gem_dma_helper.c.
This patch is compile-time tested building a x86_64 kernel with
`make allyesconfig && make drivers/gpu/drm`.
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Danilo Krummrich <dakr@redhat.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com> #drivers/gpu/drm/arm
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20220802000405.949236-4-dakr@redhat.com
idr_init_base(), implemented by commit 6ce711f275 ("idr: Make 1-based
IDRs more efficient"), let us set an arbitrary base other than
idr_init(), which uses base 0.
Since, for this IDR, no ID < 1 is ever requested/allocated, using
idr_init_base(&idr, 1) avoids unnecessary tree walks.
Signed-off-by: Danilo Krummrich <dakr@redhat.com>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Link: https://lore.kernel.org/r/20220701190227.284783-1-dakr@redhat.com
Instead of distingting between shared and exclusive fences specify
the fence usage while adding fences.
Rework all drivers to use this interface instead and deprecate the old one.
v2: some kerneldoc comments suggested by Daniel
v3: fix a missing case in radeon
v4: rebase on nouveau changes, fix lockdep and temporary disable warning
v5: more documentation updates
v6: separate internal dma_resv changes from this patch, avoids to
disable warning temporary, rebase on upstream changes
v7: fix missed case in lima driver, minimize changes to i915_gem_busy_ioctl
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20220407085946.744568-3-christian.koenig@amd.com
linux/string_helpers.h provides a helper to return "yes"/"no" strings.
Replace the open coded versions with str_yes_no(). The places were
identified with the following semantic patch:
@@
expression b;
@@
- b ? "yes" : "no"
+ str_yes_no(b)
Then the includes were added, so we include-what-we-use, and parenthesis
adjusted in drivers/gpu/drm/v3d/v3d_debugfs.c. After the conversion we
still see the same binary sizes:
text data bss dec hex filename
51149 3295 212 54656 d580 virtio/virtio-gpu.ko.old
51149 3295 212 54656 d580 virtio/virtio-gpu.ko
1441491 60340 800 1502631 16eda7 radeon/radeon.ko.old
1441491 60340 800 1502631 16eda7 radeon/radeon.ko
6125369 328538 34000 6487907 62ff63 amd/amdgpu/amdgpu.ko.old
6125369 328538 34000 6487907 62ff63 amd/amdgpu/amdgpu.ko
411986 10490 6176 428652 68a6c drm.ko.old
411986 10490 6176 428652 68a6c drm.ko
98129 1636 264 100029 186bd dp/drm_dp_helper.ko.old
98129 1636 264 100029 186bd dp/drm_dp_helper.ko
1973432 109640 2352 2085424 1fd230 nouveau/nouveau.ko.old
1973432 109640 2352 2085424 1fd230 nouveau/nouveau.ko
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220126093951.1470898-10-lucas.demarchi@intel.com
GEM helper libraries use struct drm_driver.gem_create_object to let
drivers override GEM object allocation. On failure, the call returns
NULL.
Change the semantics to make the calls return a pointer-encoded error.
This aligns the callback with its callers. Fixes the ingenic driver,
which already returns an error pointer.
Also update the callers to handle the involved types more strictly.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Steven Price <steven.price@arm.com>
Acked-by: Maxime Ripard <maxime@cerno.tech>
Link: https://patchwork.freedesktop.org/patch/msgid/20211130095255.26710-1-tzimmermann@suse.de
Change all GEM SHMEM object functions that receive a GEM object
of type struct drm_gem_object to expect an object of type
struct drm_gem_shmem_object instead.
This change reduces the number of upcasts from struct drm_gem_object
by moving them into callers. The C compiler can now verify that the
GEM SHMEM functions are called with the correct type.
For consistency, the patch also renames drm_gem_shmem_free_object to
drm_gem_shmem_free. It further updates documentation for a number of
functions.
v3:
* fix docs for drm_gem_shmem_object_free()
v2:
* mention _object_ callbacks in docs (Daniel)
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20211108093149.7226-4-tzimmermann@suse.de
