fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2025-12-22 11:20:11 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

tu/kgsl: Remove zero CB queue submission fast path

Browse source 
The fast path for kgsl_queue_submit when there are no command buffers
and only sync objects led to breakage for two reasons:
* The fast path was not properly handling duplication of the merged sync
  object assigned to signalled `kgsl_syncobj`(s), which could lead to
  multiple `kgsl_syncobj`s owning the same FD and consequently issues
  such as double close of that FD leading to UB. This is fixed by moving
  to the slow path as it always produces a timestamp sync object which
  can be trivially duplicated.
* The Vulkan specification requires that drivers strictly follow the
  order of submission of command buffers and consequently the order of
  semaphore signal/wait operations. Since no submission was being made
  to the kernel, subsequent submissions could be executed without waiting
  for wait/signal operations from previous submissions to complete.

As both of these issues are fixed by moving to the slow path, this patch
removes the fast path in favor of the more correct slow path.

Signed-off-by: Mark Collins <mark@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33894>
This commit is contained in:
Mark Collins 2025-03-05 08:59:55 +00:00 • committed by Marge Bot
parent 87a90a3b38
commit 0342d34bdb
1 changed files with 0 additions and 39 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

39
src/freedreno/vulkan/tu_knl_kgsl.cc
View file

@ -1089,45 +1089,6 @@ kgsl_queue_submit(struct tu_queue *queue, void *_submit,
uint64_t start_ts = tu_perfetto_begin_submit();
#endif
if (submit->commands.size == 0) {
const struct kgsl_syncobj *wait_semaphores[wait_count + 1];
for (uint32_t i = 0; i < wait_count; i++) {
wait_semaphores[i] = &container_of(waits[i].sync,
struct vk_kgsl_syncobj, vk)
->syncobj;
}
struct kgsl_syncobj last_submit_sync;
if (queue->fence >= 0)
last_submit_sync = (struct kgsl_syncobj) {
.state = KGSL_SYNCOBJ_STATE_TS,
.queue = queue,
.timestamp = queue->fence,
};
else
last_submit_sync = (struct kgsl_syncobj) {
.state = KGSL_SYNCOBJ_STATE_SIGNALED,
};
wait_semaphores[wait_count] = &last_submit_sync;
struct kgsl_syncobj wait_sync =
kgsl_syncobj_merge(wait_semaphores, wait_count + 1);
assert(wait_sync.state !=
KGSL_SYNCOBJ_STATE_UNSIGNALED); // Would wait forever
for (uint32_t i = 0; i < signal_count; i++) {
struct kgsl_syncobj *signal_sync =
&container_of(signals[i].sync, struct vk_kgsl_syncobj, vk)
->syncobj;
kgsl_syncobj_reset(signal_sync);
*signal_sync = wait_sync;
}
return VK_SUCCESS;
}
VkResult result = VK_SUCCESS;
if (u_trace_submission_data) {