griefith/test2
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

Vulkan: Optimize swapchain image selection

Browse Source 
Improve the number of images that are requested in the swapchain. For
mailbox presentation mode we select tripple buffering. For V-Sync
presentation mode we select double buffering.

In future we might want to make the presentation mode and swapchain
images dynamic based on the actual action that the user is performing.
When doing action that benefit from low latency (paint cursor) we should
select mailbox, otherwise we can use fifo to reduce CPU/GPU usage and
safe some energy usage along the way. See #136923 for design task.

Pull Request: https://projects.blender.org/blender/blender/pulls/136922
This commit is contained in:
Jeroen Bakker
2025-04-03 10:30:09 +02:00
parent 28d0bef706
commit d271b69c67
1 changed files with 8 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
intern/ghost/intern/GHOST_ContextVK.cc
View File

@@ -743,6 +743,9 @@ static GHOST_TSuccess selectPresentMode(VkPhysicalDevice device,
vkGetPhysicalDeviceSurfacePresentModesKHR(device, surface, &present_count, presents.data());
/* MAILBOX is the lowest latency V-Sync enabled mode. We will use it if available as it fixes
* some lag on NVIDIA/Intel GPUs. */
/* TODO: select the correct presentation mode based on the actual being performed by the user.
* When low latency is required (paint cursor) we should select mailbox, otherwise we can do FIFO
* to reduce CPU/GPU usage.*/
for (auto present_mode : presents) {
if (present_mode == VK_PRESENT_MODE_MAILBOX_KHR) {
*r_presentMode = present_mode;
@@ -750,7 +753,7 @@ static GHOST_TSuccess selectPresentMode(VkPhysicalDevice device,
}
}
/*FIFO present mode is always available and we (should) prefer it as it will keep the main loop
/* FIFO present mode is always available and we (should) prefer it as it will keep the main loop
* running along the monitor refresh rate. Mailbox and FIFO relaxed can generate a lot of frames
* that will never be displayed. */
*r_presentMode = VK_PRESENT_MODE_FIFO_KHR;
@@ -877,8 +880,10 @@ GHOST_TSuccess GHOST_ContextVK::recreateSwapchain()
}
}
/* Driver can stall if only using minimal image count. */
uint32_t image_count_requested = 4;
/* Use double buffering when using FIFO. Increasing the number of images could stall when doing
* actions that require low latency (paint cursor, UI resizing). MAILBOX prefers tripple
* buffering. */
uint32_t image_count_requested = present_mode == VK_PRESENT_MODE_MAILBOX_KHR ? 3 : 2;
/* NOTE: maxImageCount == 0 means no limit. */
if (capabilities.minImageCount != 0 && image_count_requested < capabilities.minImageCount) {
image_count_requested = capabilities.minImageCount;