zakarumych / gpu-descriptor Goto Github PK

Backend agnostic descriptor allocator for Vulkan-like APIs

License: Other

Rust 100.00%

gpu-descriptor's Introduction

gpu-descriptor

Library for Vulkan-like APIs to allocated descriptor sets from descriptor pools fast, with least overhead and zero fragmentation.

Straightforward usage:

use gpu_descriptor::DescriptorAllocator;

let mut allocator = DescriptorAllocator::new(max_update_after_bind_descriptors_in_all_pools); // Limit as dictated by API for selected hardware

let result = allocator.allocate(
    device, // Implementation of `gpu_descriptor::DescriptorDevice`. Comes from plugins.
    layout, // Descriptor set layout recognized by device's type.
    flags,  // Flags specified when layout was created.
    layout_descriptor_count, // Descriptors count in the layout.
    count, // count of sets to allocated.
);

License

Licensed under either of

Apache License, Version 2.0, (license/APACHE or http://www.apache.org/licenses/LICENSE-2.0)
MIT license (license/MIT or http://opensource.org/licenses/MIT)

at your option.

Contributions

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in the work by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.

Support me on Patreon

gpu-descriptor's People

Contributors

Stargazers

Watchers

Forkers

kvark hellbutcher xmac94x friz64 madoshakalaka icodein daxpedda waywardmonkeys wackbyte atouchet erichdongubler

gpu-descriptor's Issues

Handling of empty layouts

See gfx-rs/wgpu#1390 (comment)
Is this supposed to be handled by this library?

Resetting the whole pool without freeing individual descriptors

How can I use the DescriptorAllocator without freeing individual descriptors? I would like to just drop them and then reset the whole pool. Current API seems to assume all descriptor sets are individually freed, instead.

Release with new version of `hashbrown`, again

As in #24 there is another new hashbrown (that is causing multiple other old things to be brought in as well), can another patch release be done and released with modern hashbrown?

Error should implement Error

Division by zero in new_pool_size

Lines like this should check for 0:

max_sets = (u32::MAX / self.size.sampler).min(max_sets);

LICENSE files in package subdirs

Background:

Hello, the Fuchsia project vendors crates from crates.io, and in order to do so we require explicit license files alongside the source code. Here is the policy: https://fuchsia.dev/fuchsia-src/contribute/governance/policy/open-source-licensing-policies?hl=en#licenses_and_tracking . In particular, reading the SPDX package/license field from the crate's Cargo.toml is not good enough.

Request:

Could you please add LICENSE-* files to the package subdirectories, specifically gpu-descriptor (although you might as well do it for all package dirs that are uploaded to crates.io)?

We are currently using gpu-descriptor 0.2.2. Since crates.io doesn't allow re-uploading the same version with a modified crate, it seems the right thing to do is to upload a new crate with version 0.2.3.

I'd appreciate if you're able to do this, since I have to file similar issues for many other crate dependencies. But if you don't have the bandwidth to address this issue, please let me know and I'll find time to submit a pull request (although of course I won't be able to upload anything to crates.io)

Purpose of this crate?

Looking at the API, it has a notion of descriptor pools. Users are expected to create a descriptor pool with certain capacity, and then allocate/free descriptors from that pool. How does it help versus just using gfx-hal directly?
My expectation was that this crate would abstract away the management of descriptor pools , but it's not doing that. Compare to gfx-descriptor (and rendy-descriptor): https://docs.rs/gfx-descriptor/0.2.0/gfx_descriptor/struct.DescriptorAllocator.html