Hi David,
On 18 July 2018 at 01:12, David Riley <davidri...@chromium.org> wrote:
Commit message here should explain why this is needed. Is the current
kms_swrast usage failing/crashing somewhere, etc.
> Signed-off-by: David Riley <davidri...@chromium.org>
> ---
> src/egl/drivers/dri2/platform_surfaceless.c | 28 +++++++++++++++++----
> 1 file changed, 23 insertions(+), 5 deletions(-)
>
> diff --git a/src/egl/drivers/dri2/platform_surfaceless.c
> b/src/egl/drivers/dri2/platform_surfaceless.c
> index a0348a5e95..f5fe7119c6 100644
> --- a/src/egl/drivers/dri2/platform_surfaceless.c
> +++ b/src/egl/drivers/dri2/platform_surfaceless.c
> @@ -260,6 +260,13 @@ static const __DRIimageLoaderExtension
> image_loader_extension = {
> .flushFrontBuffer = surfaceless_flush_front_buffer,
> };
>
> +static const __DRIswrastLoaderExtension swrast_loader_extension = {
> + .base = { __DRI_SWRAST_LOADER, 1 },
> + .getDrawableInfo = NULL,
> + .putImage = NULL,
> + .getImage = NULL,
> +};
> +
> #define DRM_RENDER_DEV_NAME "%s/renderD%d"
>
> static const __DRIextension *image_loader_extensions[] = {
> @@ -269,6 +276,14 @@ static const __DRIextension *image_loader_extensions[] =
> {
> NULL,
> };
>
> +static const __DRIextension *swrast_loader_extensions[] = {
> + &swrast_loader_extension.base,
> + &image_loader_extension.base,
> + &image_lookup_extension.base,
> + &use_invalidate.base,
How did you compiler this list. Gut suggests that at least one of
those should not be here.
Doesn't this break the existing kms_swrast usage?
>
> dri2_dpy->fd = fd;
> - if (dri2_load_driver_dri3(dpy))
> + if (dri2_load_driver_dri3(dpy)) {
> return true;
> + }
Unnecessary change.
HTH
Emil
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
