✗ CI.checkpatch: warning for drm/xe: Fix races on fdinfo (rev4)
Patchwork
patchwork at emeril.freedesktop.org
Mon Nov 4 15:20:51 UTC 2024
== Series Details ==
Series: drm/xe: Fix races on fdinfo (rev4)
URL : https://patchwork.freedesktop.org/series/140544/
State : warning
== Summary ==
+ KERNEL=/kernel
+ git clone https://gitlab.freedesktop.org/drm/maintainer-tools mt
Cloning into 'mt'...
warning: redirecting to https://gitlab.freedesktop.org/drm/maintainer-tools.git/
+ git -C mt rev-list -n1 origin/master
30ab6715fc09baee6cc14cb3c89ad8858688d474
+ cd /kernel
+ git config --global --add safe.directory /kernel
+ git log -n1
commit de9c0b9637c6b132c7f146d6b778e0acff2cc198
Author: Lucas De Marchi <lucas.demarchi at intel.com>
Date: Mon Nov 4 06:38:15 2024 -0800
drm/xe: Sample gpu timestamp closer to exec queues
Move the force_wake_get to the beginning of the function so the gpu
timestamp can be closer to the sampled value for exec queues. This
avoids additional delays waiting for force wake ack which can make the
proportion between cycles/total_cycles fluctuate either to more or to
less than the real value, depending if the delay happened in the first
or second sample of a top-like application:
sample 0:
read_exec_queue_timestamp
<<<< (A)
read_gpu_timestamp
sample 1:
read_exec_queue_timestamp
<<<<< (B)
read_gpu_timestamp
Additional delay in (A) will cause the utilization to be bigger
than it should. Additional delay in (B) will cause the utilization to be
smaller than it should.
With this a LNL system that was failing after ~60 iterations, get to run
to 100 without a failure. This is still not perfect, and it's easy to
introduce errors by just loading the CPU with `stress --cpu $(nproc)` -
the same igt test in this case fails after 2 or 3 iterations.
Although not perfect, this could be viewed as just extracting that ugly
loop to "find the forcewake for any available engine" into a separate
function, and at the same time improving the results a little bit.
Signed-off-by: Lucas De Marchi <lucas.demarchi at intel.com>
+ /mt/dim checkpatch 0a6cc4357ae4d824f909468ca1deed28ae5ac96f drm-intel
59b392093863 drm/xe: Add trace to lrc timestamp update
-:46: WARNING:FILE_PATH_CHANGES: added, moved or deleted file(s), does MAINTAINERS need updating?
#46:
new file mode 100644
-:89: CHECK:OPEN_ENDED_LINE: Lines should not end with a '('
#89: FILE: drivers/gpu/drm/xe/xe_trace_lrc.h:24:
+ TP_STRUCT__entry(
-:97: CHECK:OPEN_ENDED_LINE: Lines should not end with a '('
#97: FILE: drivers/gpu/drm/xe/xe_trace_lrc.h:32:
+ TP_fast_assign(
-:115: CHECK:SPACING: spaces preferred around that '/' (ctx:VxV)
#115: FILE: drivers/gpu/drm/xe/xe_trace_lrc.h:50:
+#define TRACE_INCLUDE_PATH ../../drivers/gpu/drm/xe
^
-:115: CHECK:SPACING: spaces preferred around that '/' (ctx:VxV)
#115: FILE: drivers/gpu/drm/xe/xe_trace_lrc.h:50:
+#define TRACE_INCLUDE_PATH ../../drivers/gpu/drm/xe
^
-:115: CHECK:SPACING: spaces preferred around that '/' (ctx:VxV)
#115: FILE: drivers/gpu/drm/xe/xe_trace_lrc.h:50:
+#define TRACE_INCLUDE_PATH ../../drivers/gpu/drm/xe
^
-:115: CHECK:SPACING: spaces preferred around that '/' (ctx:VxV)
#115: FILE: drivers/gpu/drm/xe/xe_trace_lrc.h:50:
+#define TRACE_INCLUDE_PATH ../../drivers/gpu/drm/xe
^
-:115: CHECK:SPACING: spaces preferred around that '/' (ctx:VxV)
#115: FILE: drivers/gpu/drm/xe/xe_trace_lrc.h:50:
+#define TRACE_INCLUDE_PATH ../../drivers/gpu/drm/xe
^
total: 0 errors, 1 warnings, 7 checks, 82 lines checked
58c3cda26252 drm/xe: Stop accumulating LRC timestamp on job_free
f59a9bcc80e0 drm/xe: Reword exec_queue and vm lock doc
deec55c6f35f drm/xe: Wait on killed exec queues
de9c0b9637c6 drm/xe: Sample gpu timestamp closer to exec queues
More information about the Intel-xe
mailing list