mirrors/monado
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/monado/monado.git synced 2025-01-01 12:46:12 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

c/client: Tidy VK compositor

Browse source
This commit is contained in:
Jakob Bornecrantz 2022-03-31 23:32:22 +01:00
parent 3bff3827d9
commit 48fd3e626c
1 changed files with 24 additions and 28 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

52
src/xrt/compositor/client/comp_vk_client.c
View file

@ -14,12 +14,6 @@
#include "comp_vk_client.h"
#include <stdio.h>
#include <stdlib.h>
#include <assert.h>
#define MS_TO_NS(ms) (ms * 1000L * 1000L)
/*!
* Down-cast helper.
@ -671,8 +665,7 @@ client_vk_swapchain_create(struct xrt_compositor *xc,
return XRT_ERROR_FAILED_TO_SUBMIT_VULKAN_COMMANDS;
}
// Prerecord command buffers for swapchain image ownership/layout
// transitions
// Prerecord command buffers for swapchain image ownership/layout transitions
for (uint32_t i = 0; i < xsc->image_count; i++) {
ret = vk_init_cmd_buffer(vk, &sc->acquire[i]);
if (ret != VK_SUCCESS) {
@ -691,22 +684,6 @@ client_vk_swapchain_create(struct xrt_compositor *xc,
.layerCount = VK_REMAINING_ARRAY_LAYERS,
};
/*
* The biggest reason is that VK_IMAGE_LAYOUT_PRESENT_SRC_KHR is
* used here is that this is what hello_xr used to barrier to,
* and it worked on a wide verity of drivers. So it's safe.
*
* There might not be a Vulkan renderer on the other endm
* there could be a OpenGL compositor, heck there could be a X
* server even. On Linux VK_IMAGE_LAYOUT_PRESENT_SRC_KHR is what
* you use if you want to "flush" out all of the pixels to the
* memory buffer that has been shared to you from a X11 server.
*
* This is not what the spec says you should do when it comes to
* external images thou. Instead we should use the queue family
* index `VK_QUEUE_FAMILY_EXTERNAL`. And use semaphores to
* synchronize.
*/
VkImageMemoryBarrier acquire = {
.sType = VK_STRUCTURE_TYPE_IMAGE_MEMORY_BARRIER,
.srcAccessMask = 0,
@ -732,10 +709,29 @@ client_vk_swapchain_create(struct xrt_compositor *xc,
};
//! @todo less conservative pipeline stage masks based on usage
vk->vkCmdPipelineBarrier(sc->acquire[i], VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT,
VK_PIPELINE_STAGE_ALL_COMMANDS_BIT, 0, 0, NULL, 0, NULL, 1, &acquire);
vk->vkCmdPipelineBarrier(sc->release[i], VK_PIPELINE_STAGE_ALL_COMMANDS_BIT,
VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT, 0, 0, NULL, 0, NULL, 1, &release);
vk->vkCmdPipelineBarrier( //
sc->acquire[i], // commandBuffer
VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT, // srcStageMask
VK_PIPELINE_STAGE_ALL_COMMANDS_BIT, // dstStageMask
0, // dependencyFlags
0, // memoryBarrierCount
NULL, // pMemoryBarriers
0, // bufferMemoryBarrierCount
NULL, // pBufferMemoryBarriers
1, // imageMemoryBarrierCount
&acquire); // pImageMemoryBarriers
vk->vkCmdPipelineBarrier( //
sc->release[i], // commandBuffer
VK_PIPELINE_STAGE_ALL_COMMANDS_BIT, // srcStageMask
VK_PIPELINE_STAGE_BOTTOM_OF_PIPE_BIT, // dstStageMask
0, // dependencyFlags
0, // memoryBarrierCount
NULL, // pMemoryBarriers
0, // bufferMemoryBarrierCount
NULL, // pBufferMemoryBarriers
1, // imageMemoryBarrierCount
&release); // pImageMemoryBarriers
ret = vk->vkEndCommandBuffer(sc->acquire[i]);
if (ret != VK_SUCCESS) {