[Mesa-dev] [PATCH 2/3] anv/device: keep track of 'device lost' state

[Jason Ekstrand]

On Thu, Mar 23, 2017 at 2:32 AM, Iago Toral Quiroga <itoral at igalia.com>
wrote:

> The Vulkan specs say:
>
>    "A logical device may become lost because of hardware errors, execution
>     timeouts, power management events and/or platform-specific events. This
>     may cause pending and future command execution to fail and cause
> hardware
>     resources to be corrupted. When this happens, certain commands will
>     return VK_ERROR_DEVICE_LOST (see Error Codes for a list of such
> commands).
>     After any such event, the logical device is considered lost. It is not
>     possible to reset the logical device to a non-lost state, however the
> lost
>     state is specific to a logical device (VkDevice), and the corresponding
>     physical device (VkPhysicalDevice) may be otherwise unaffected. In some
>     cases, the physical device may also be lost, and attempting to create a
>     new logical device will fail, returning VK_ERROR_DEVICE_LOST."
>
> This means that we need to track if a logical device has been lost so we
> can
> have the commands referenced by the spec return VK_ERROR_DEVICE_LOST
> immediately.
> ---
>  src/intel/vulkan/anv_device.c  | 5 +++++
>  src/intel/vulkan/anv_private.h | 1 +
>  2 files changed, 6 insertions(+)
>
> diff --git a/src/intel/vulkan/anv_device.c b/src/intel/vulkan/anv_device.c
> index a11cb32..19bac84 100644
> --- a/src/intel/vulkan/anv_device.c
> +++ b/src/intel/vulkan/anv_device.c
> @@ -929,6 +929,7 @@ anv_device_submit_simple_batch(struct anv_device
> *device,
>     ret = anv_gem_wait(device, bo.gem_handle, &timeout);
>     if (ret != 0) {
>        /* We don't know the real error. */
> +      device->lost = true;
>        result = vk_errorf(VK_ERROR_DEVICE_LOST, "execbuf2 failed: %m");
>        goto fail;
>     }
> @@ -973,6 +974,7 @@ VkResult anv_CreateDevice(
>     device->_loader_data.loaderMagic = ICD_LOADER_MAGIC;
>     device->instance = physical_device->instance;
>     device->chipset_id = physical_device->chipset_id;
> +   device->lost = false;
>
>     if (pAllocator)
>        device->alloc = *pAllocator;
> @@ -1250,6 +1252,7 @@ anv_device_execbuf(struct anv_device *device,
>     int ret = anv_gem_execbuffer(device, execbuf);
>     if (ret != 0) {
>        /* We don't know the real error. */
> +      device->lost = true;
>        return vk_errorf(VK_ERROR_DEVICE_LOST, "execbuf2 failed: %m");
>     }
>
> @@ -1339,6 +1342,7 @@ out:
>         * submit the same job again to this device.
>         */
>        result = VK_ERROR_DEVICE_LOST;
> +      device->lost = true;
>
>        /* If we return VK_ERROR_DEVICE LOST here, we need to ensure that
>         * vkWaitForFences() and vkGetFenceStatus() return a valid result
> @@ -1865,6 +1869,7 @@ VkResult anv_WaitForFences(
>                 return VK_TIMEOUT;
>              } else if (ret == -1) {
>                 /* We don't know the real error. */
> +                device->lost = true;
>                 return vk_errorf(VK_ERROR_DEVICE_LOST, "gem wait failed:
> %m");
>              } else {
>                 fence->state = ANV_FENCE_STATE_SIGNALED;
> diff --git a/src/intel/vulkan/anv_private.h b/src/intel/vulkan/anv_
> private.h
> index fd82ce9..68f7359 100644
> --- a/src/intel/vulkan/anv_private.h
> +++ b/src/intel/vulkan/anv_private.h
> @@ -619,6 +619,7 @@ struct anv_device {
>
>      pthread_mutex_t                             mutex;
>      pthread_cond_t                              queue_submit;
> +    bool                                        lost;
>  };
>
>  static void inline
> --
> 2.7.4
>
> _______________________________________________
> 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/20170323/2b6f80b7/attachment-0001.html>