| Message ID | 1416559920-7779-1-git-send-email-oded.gabbay@amd.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c index 5055fc9..9abac48 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c @@ -25,6 +25,7 @@ #include <linux/mutex.h> #include <linux/slab.h> #include <linux/printk.h> +#include <linux/sched.h> #include "kfd_kernel_queue.h" #include "kfd_priv.h" #include "kfd_device_queue_manager.h" @@ -274,7 +275,7 @@ static int sync_with_hw(struct kernel_queue *kq, unsigned long timeout_ms) *kq->wptr_kernel, *kq->rptr_kernel); return -ETIME; } - cpu_relax(); + schedule(); } return 0;
amdkfd uses cpu_relax() in its sync_with_hw() function. Because cpu_relax() is defined as 'REP; NOP' on x86_64, it will block the CPU from servicing IOMMU PPR requests. This may cause a deadlock, because sync_with_hw() won't be completed until the PPR request has been served. Therefore, we need to use schedule() instead of cpu_relax() as it is the minimum requirement to allow other threads to execute. Signed-off-by: Oded Gabbay <oded.gabbay@amd.com> --- drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)