[Mesa-dev] [PATCH v2 3/3] i965: if DEBUG_OUT_OF_BOUND_CHK is up, check that noise padding for each bo used in batchbuffer is correct

[Chris Wilson]

Quoting kevin.rogovin at intel.com (2017-12-13 10:18:38)
> From: Kevin Rogovin <kevin.rogovin at intel.com>
> 
> v2:
>  Comments indicating that brw_bo_padding_is_good() will do the required
>  waiting for GPU commands to finish
> 
> Signed-off-by: Kevin Rogovin <kevin.rogovin at intel.com>
> ---
>  src/mesa/drivers/dri/i965/intel_batchbuffer.c | 19 +++++++++++++++++++
>  1 file changed, 19 insertions(+)
> 
> diff --git a/src/mesa/drivers/dri/i965/intel_batchbuffer.c b/src/mesa/drivers/dri/i965/intel_batchbuffer.c
> index 91a6506a89..689ae2362a 100644
> --- a/src/mesa/drivers/dri/i965/intel_batchbuffer.c
> +++ b/src/mesa/drivers/dri/i965/intel_batchbuffer.c
> @@ -758,6 +758,7 @@ execbuffer(int fd,
>     };
>  
>     unsigned long cmd = DRM_IOCTL_I915_GEM_EXECBUFFER2;
> +   bool detected_out_of_bounds_write = false;
>  
>     if (in_fence != -1) {
>        execbuf.rsvd2 = in_fence;
> @@ -787,6 +788,24 @@ execbuffer(int fd,
>               batch->validation_list[i].offset);
>           bo->gtt_offset = batch->validation_list[i].offset;
>        }
> +
> +      if (unlikely(INTEL_DEBUG & DEBUG_OUT_OF_BOUND_CHK)) {
> +         /* brw_bo_padding_is_good() performs the necessary
> +          * syncing itself to make sure that the padding read
> +          * is correct.
> +          */

Actually that's not strictly true. Since you only do a pread here, it
will only synchronize against the last declared write to the bo. There's
no guaranteed sync with the last batch for a set of read-only bo.
Similarly, because of no domain tracking, it won't also ensure that the
bo is cache coherent before the read back.
-Chris