News for hackers. Stuff that matters.

A comment on AMD responds to Linux kernel maintainer's rejection of AMDGPU patch. Parent comment

Re:AMD responds to Linux kernel maintainer's rejection of AM... by mcpherrinm on Saturday December 10, 2016 @04:59AM (#13144937)
I think the argument here is that stability is gained from having code shared as closely as possible with the Windows codebase, which is where most of the QA work happens, and what the real experts debugging their ASICs use.

From a hardware and driver development standpoint, I think that makes a decent amount of sense. Hardware is often weird and quirky, so making sure you handle all the edge cases in two different codebases is a lot harder than doing it once. Certainly too much work for a handful of people who just want to plug their code into Linux.

Stability here is on the driver/hardware interaction, not the kernel/driver boundary, which is what the kernel folks seem to care about, naturally.

[ Parent | on HN ]
Re:AMD responds to Linux kernel maintainer's rejection of AM... by __s on Saturday December 10, 2016 @09:51AM (#13145855)
Flipside being that if AMD drivers are developed in the open with proper convention then kernel developers will help maintain them, alleviating cost for AMD. If AMD doesn't want the public maintaining their open source driver then they don't want their driver in the kernel
[ Parent | on HN ]
Re:AMD responds to Linux kernel maintainer's rejection of AM... by chris_wot on Saturday December 10, 2016 @05:28AM (#13145011)
Stability for whom? Trying to contort Linux code to follow the practices that work well for Windows does not strike me as either maintainable, or even particularly stable!
[ Parent | on HN ]
Re:AMD responds to Linux kernel maintainer's rejection of AM... by Certhas on Saturday December 10, 2016 @07:21AM (#13145340)
What does this have to do with Windows vs Linux? It's about closely working with hardware engineers.
[ Parent | on HN ]
Re:AMD responds to Linux kernel maintainer's rejection of AM... by chris_wot on Saturday December 10, 2016 @09:03AM (#13145667)
"I think the argument here is that stability is gained from having code shared as closely as possible with the Windows codebase".

HTH.

[ Parent | on HN ]
Re:AMD responds to Linux kernel maintainer's rejection of AM... by brainfire on Saturday December 10, 2016 @09:36AM (#13145793)
You have to read the rest of the comment too. "HTH"
[ Parent | on HN ]
Re:AMD responds to Linux kernel maintainer's rejection of AM... by chris_wot on Sunday December 11, 2016 @04:36AM (#13150340)
Indeed, I did which you seem to have thought I didn't read. So, do you think that bunging in code designed for Windows, which specifically uses a HAL, into Linux, which tries its best not to with the philosophy that the code that goes into Linux should be as clear as possible with as small an abstraction layer as possible is good for Linux, or any of the hundreds of developers or companies relying on the maintainers to keep to this coding and architecture style?

I think not. So no, it didn't help.

[ Parent | on HN ]