Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update wgpu to 0.20 #2417

Open
wants to merge 1 commit into
base: cosmic-text-0.11
Choose a base branch
from
Open

Update wgpu to 0.20 #2417

wants to merge 1 commit into from

Conversation

hecrj
Copy link
Member

@hecrj hecrj commented May 1, 2024

This PR updates wgpu to 0.20. Depends on #2416.

Unfortunately, this wgpu update doesn't seem to work very well on my end (Arch, X11, RTX 4090, NVIDIA drivers). 0.19 is smooth as butter for me, while in 0.20 the examples feel very sluggish and I constantly get a validation error:

2024-05-01T14:25:29.145657Z ERROR wgpu_hal::vulkan::instance: VALIDATION [VUID-vkAcquireNextImageKHR-semaphore-01779 (0x5717e75b)]
        Validation Error: [ VUID-vkAcquireNextImageKHR-semaphore-01779 ] Object 0: handle = 0x4fac1c0000000032, type = VK_OBJECT_TYPE_SEMAPHORE; | MessageID = 0x5717e75b | vkAcquireNextImageKHR():  Semaphore must not have any pending operations. The Vulkan spec states: If semaphore is not VK_NULL_HANDLE it must not have any uncompleted signal or wait operations pending (https://www.khronos.org/registry/vulkan/specs/1.3-extensions/html/vkspec.html#VUID-vkAcquireNextImageKHR-semaphore-01779)
2024-05-01T14:25:29.145682Z ERROR wgpu_hal::vulkan::instance:   objects: (type: SEMAPHORE, hndl: 0x4fac1c0000000032, name: ?)

I imagine there is something wrong with the release, although maybe I missed something. Let me know if it works for you!

@hecrj hecrj added improvement An internal improvement help wanted Extra attention is needed rendering wgpu change labels May 1, 2024
@hecrj hecrj added this to the 0.13 milestone May 1, 2024
@hecrj hecrj changed the base branch from master to cosmic-text-0.11 May 1, 2024 14:31
@hecrj
Copy link
Member Author

hecrj commented May 1, 2024

Opened an issue in the wgpu repo: gfx-rs/wgpu#5644

@matze
Copy link
Contributor

matze commented May 1, 2024

Works alright without validation errors and sluggish behavior on a Radeon RX 5600 XT but there is a lot more log output that there used to.

@skygrango
Copy link
Contributor

RX 7900XTX no errors like yours

first wgpu-0.20

     Running benches/wgpu.rs (/home/menter/software/iced/target/release/deps/wgpu-2badaebd889b51ec)
Gnuplot not found, using plotters backend
wgpu — canvas (light) time:   [325.68 µs 326.59 µs 327.73 µs]
Found 12 outliers among 100 measurements (12.00%)
  5 (5.00%) high mild
  7 (7.00%) high severe

wgpu — canvas (heavy) time:   [371.11 µs 373.13 µs 375.81 µs]
Found 12 outliers among 100 measurements (12.00%)
  4 (4.00%) high mild
  8 (8.00%) high severe

wgpu - layered text (light)
                        time:   [141.67 µs 142.28 µs 142.98 µs]
Found 10 outliers among 100 measurements (10.00%)
  4 (4.00%) high mild
  6 (6.00%) high severe

wgpu - layered text (heavy)
                        time:   [4.5305 ms 4.5671 ms 4.6064 ms]
Found 2 outliers among 100 measurements (2.00%)
  1 (1.00%) high mild
  1 (1.00%) high severe

second cosmic-text-0.11

     Running benches/wgpu.rs (/home/menter/software/iced/target/release/deps/wgpu-8542c9a313f43b18)
Gnuplot not found, using plotters backend
wgpu — canvas (light) time:   [322.68 µs 323.36 µs 324.16 µs]
                        change: [-8.3115% -3.0241% +2.0292%] (p = 0.30 > 0.05)
                        No change in performance detected.
Found 12 outliers among 100 measurements (12.00%)
  4 (4.00%) high mild
  8 (8.00%) high severe

wgpu — canvas (heavy) time:   [369.82 µs 371.96 µs 374.69 µs]
                        change: [-13.772% -0.7633% +14.241%] (p = 0.92 > 0.05)
                        No change in performance detected.
Found 12 outliers among 100 measurements (12.00%)
  4 (4.00%) high mild
  8 (8.00%) high severe

wgpu - layered text (light)
                        time:   [139.25 µs 139.81 µs 140.45 µs]
                        change: [-6.7075% -1.6502% +3.6626%] (p = 0.58 > 0.05)
                        No change in performance detected.
Found 14 outliers among 100 measurements (14.00%)
  6 (6.00%) high mild
  8 (8.00%) high severe

wgpu - layered text (heavy)
                        time:   [4.3791 ms 4.4079 ms 4.4371 ms]
                        change: [-4.4980% -3.4845% -2.4238%] (p = 0.00 < 0.05)
                        Performance has improved.

wgpu - layered text (heavy)
螢幕截圖_20240502_232802-1

wgpu - layered text (light)
螢幕截圖_20240502_232906-1

wgpu — canvas (heavy)
螢幕截圖_20240502_232940

wgpu — canvas (light)
螢幕截圖_20240502_233015

@hecrj hecrj force-pushed the cosmic-text-0.11 branch 3 times, most recently from 005dd7d to acfa97a Compare May 8, 2024 20:38
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
change help wanted Extra attention is needed improvement An internal improvement rendering wgpu
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants