[Mesa-dev] [PATCH 3/3] anv: bail out if anv_wsi_init() fails

Jason Ekstrand jason at jlekstrand.net
Sun May 29 17:10:48 UTC 2016


On May 29, 2016 9:41 AM, "Emil Velikov" <emil.l.velikov at gmail.com> wrote:
>
> From: Emil Velikov <emil.velikov at collabora.com>
>
> Otherwise we'll end up setting up a device with no winsys integration.

Seems reasonable. Running headless isn't a problem but winsys failing to
initialize basically only happens on out-of-memory so we should fail in
that case.

> Signed-off-by: Emil Velikov <emil.velikov at collabora.com>
> ---
> Hard-coding the rendernode name in anv_physical_device_init() is a bad
> idea really. We could/should be using drmGetDevices() to get info on all
> the devices (master/render/etc. node names, pci location etc.) and apply
> our heuristics on top of that.

Yes, we should come up with something better.  Preferably that doesn't
involve libdrm...

1 and 3 rb

> That can come up as a follow up change.
> ---
>  src/intel/vulkan/anv_device.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/src/intel/vulkan/anv_device.c b/src/intel/vulkan/anv_device.c
> index da7a394..046777d 100644
> --- a/src/intel/vulkan/anv_device.c
> +++ b/src/intel/vulkan/anv_device.c
> @@ -150,7 +150,9 @@ anv_physical_device_init(struct anv_physical_device
*device,
>     device->compiler->shader_debug_log = compiler_debug_log;
>     device->compiler->shader_perf_log = compiler_perf_log;
>
> -   anv_init_wsi(device);
> +   result = anv_init_wsi(device);
> +   if (result != VK_SUCCESS)
> +       goto fail;
>
>     /* XXX: Actually detect bit6 swizzling */
>     isl_device_init(&device->isl_dev, device->info, swizzled);
> --
> 2.8.2
>
> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/mesa-dev/attachments/20160529/1a426303/attachment.html>


More information about the mesa-dev mailing list