nv-tegra-mirror/linux-nvgpu
1
0
Fork 0
mirror of git://nv-tegra.nvidia.com/linux-nvgpu.git synced 2025-12-23 01:50:07 +03:00
Code Issues Packages Projects Releases Wiki Activity

gpu: nvgpu: pg init task hogging cpu

Browse Source 
The Pg init task hogs the kernel by having a wait condition with no timeout
waiting for pg state change, but ps state may not post a change in a long time
depending on runtime conditions, so we get soft-crashes warning spews in the
kernel

We solve this by making the condition wait interruptible

bug 200346134

Change-Id: I8a3349031acc5065b767dc22eec6e5df113d3ad7
Signed-off-by: David Nieto <dmartineznie@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1566545
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
This commit is contained in:
David Nieto
2017-09-22 10:47:12 -07:00
committed by mobile promotions
parent 350bb74859
commit 82ef5f7b3b
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
drivers/gpu/nvgpu/common/pmu/pmu.c
View File

@@ -409,7 +409,7 @@ static int nvgpu_pg_init_task(void *arg)
while (true) { while (true) {
NVGPU_COND_WAIT(&pg_init->wq, NVGPU_COND_WAIT_INTERRUPTIBLE(&pg_init->wq,
(pg_init->state_change == true), 0); (pg_init->state_change == true), 0);
pmu->pg_init.state_change = false; pmu->pg_init.state_change = false;