ROCm-OpenCL-Runtime VS Heady

Its not that they're supporting buggy code, they just downgraded the quality of their implementation significantly. They made the compiler a lot worse when they swapped to rocm

https://github.com/RadeonOpenCompute/ROCm-OpenCL-Runtime/iss... is the tracking issue for it filed a year ago, which appears to be wontfix largely because its a lot of work

OpenCL still unfortunately supports quite a few things that vulkan doesn't, which makes swapping away very difficult for some use cases

There's https://docs.amd.com/bundle/ROCm-Installation_FAQ/page/Frequently_Asked_Questions.html which leads to a page which doesn't list any gpus that I can see, there's https://rocm.docs.amd.com/en/latest/release/gpu_os_support.html which lists "RDNA2" , there's https://github.com/RadeonOpenCompute/ROCm/issues/1698 which is from last year and mentions changing an env for the RX 6600 XT (navi 23) . Not a lot is mentioned in the readme of https://github.com/RadeonOpenCompute/ROCm-OpenCL-Runtime .

$ dnf info rocm-opencl Installed Packages Name : rocm-opencl Version : 5.4.3 Release : 2.fc38 Architecture : x86_64 Size : 1.7 M Source : rocm-opencl-5.4.3-2.fc38.src.rpm Repository : @System From repo : updates Summary : ROCm OpenCL Runtime URL : https://github.com/RadeonOpenCompute/ROCm-OpenCL-Runtime License : MIT Description : ROCm OpenCL language runtime. : Supports offline and in-process/in-memory compilation.

Thank you for subscribing to AMD OpenCL Facts! To unsubscribe please type AARRHGHGH. I'm open for questions though, and the bug report for this is here. For more mildly disgruntled rambling:

Eg this bug about shared cl/gl textures with mipmaps being broken has now breached its first birthday, without even acknowledgement - basic cl/gl functionality here. This bug took a year for a fix to make its way into a public driver. And this fairly performance critical bug is just "wontfix", and also a significant downgrade from their old driver stack

Its definitely workable if you're willing to put in the effort (except for things that are straight up broken in some cases, like device side enqueue), but there are some issues that require.. fairly major workarounds

Their implementation is here: https://github.com/RadeonOpenCompute/ROCm-OpenCL-Runtime

After a lot of moderately annoyed testing, I discovered that the AMD OpenCL implementation is.. rather dumb. If any two kernels share any arguments arguments, it inserts a command barrier between the two, hard-stalling the GPU. After filing a bug, it turns out this is wontfix as well, which is doubly bad. There's no set of flags in OpenCL that you can use to fix this either

What's particularly bizarre, is that with one bug report I filed, they claim to have fixed it internally in april, but.. no public driver has ever been released with the fix. For 7 months? Which is just a bizarre software development process

I've recently made a small update to Heady, which is a C++ library and command-line tool used to create amalgamated single-header libraries from standard C++ source files. In the readme, I also describe how to prepare your source code for header amalgamation using four rules.

If you're interested, I've made an open source utility called Heady that converts normal C++ library code into an amalgamated header file. You obviously have to follow a few rules and mark up your functions appropriately, but I've been using this successfully in my Jinx scripting library for a while now.

In of the first sections in the tutorial and on the Github landing page, I mention that Jinx requires C++ 17, because I do make use of a few C++ 17-specific features. For example, one of those techniques makes it simpler to offer an amalgamated header-only version of the library. I specifically test on the big three: MSVC, Clang, and gcc, on Windows, Mac, and Linux respectively, so I typically don't use features not available on the latest versions of those OSes and their toolchains, or else offer conditional compilation.