HI, Nirmoy,
On Fri, 2024-12-13 at 13:24 +0100, Nirmoy Das wrote:
> Ensure a non-interruptible wait is used when moving a bo to
> XE_PL_SYSTEM. This prevents dma_mappings from being removed
> prematurely
> while a GPU job is still in progress, even if the CPU receives a
> signal during the operation.
>
> Fixes: 75521e8b56e8 ("drm/xe: Perform dma_map when moving system
> buffer objects to TT")
> Cc: Thomas Hellström <thomas.hellstrom@xxxxxxxxxxxxxxx>
> Cc: Matthew Brost <matthew.brost@xxxxxxxxx>
> Cc: Lucas De Marchi <lucas.demarchi@xxxxxxxxx>
> Cc: <stable@xxxxxxxxxxxxxxx> # v6.11+
> Suggested-by: Matthew Auld <matthew.auld@xxxxxxxxx>
> Signed-off-by: Nirmoy Das <nirmoy.das@xxxxxxxxx>
> Reviewed-by: Matthew Auld <matthew.auld@xxxxxxxxx>
For both patches
Reviewed-by: Thomas Hellström <thomas.hellstrom@xxxxxxxxxxxxxxx>
> ---
> drivers/gpu/drm/xe/xe_bo.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
> index 283cd0294570..06931df876ab 100644
> --- a/drivers/gpu/drm/xe/xe_bo.c
> +++ b/drivers/gpu/drm/xe/xe_bo.c
> @@ -733,7 +733,7 @@ static int xe_bo_move(struct ttm_buffer_object
> *ttm_bo, bool evict,
> new_mem->mem_type == XE_PL_SYSTEM) {
> long timeout = dma_resv_wait_timeout(ttm_bo-
> >base.resv,
>
> DMA_RESV_USAGE_BOOKKEEP,
> - true,
> + false,
>
> MAX_SCHEDULE_TIMEOUT);
> if (timeout < 0) {
> ret = timeout;
