Re: AVX support

[Maxime Villard <max%m00nbsd.net@localhost>]

Le 29/12/2016 à 12:36, David Laight a écrit :
> > Back then I had written a patch to switch to a dynamic FPU area - that is,
> > the
> > kernel allocates dynamically an fpu area for each lwp, the first time they
> > use the fpu. But for some reason I've never committed it.
>
> Not much point, it probably saves very little memory and adds a dufficult
> to recover 'malloc fail'.

The point was not to save memory, it was just to avoid collision between FPU
states and the kernel stack.

It does indeed introduce a difficult to recover malloc fail, but it is already
what we're doing on sparc64 [1] - I'm not saying that having a PR_WAITOK in
interrupt context is a particularly good idea, but just that dynamic FPU areas
are already handled this way on other architectures.

> [...]
>
> The AVX save area is only written if the process has used the avx registers.
> There is also an instruction to clear the 'avx used' flag.
> Since the avx regs are 'caller saved' they don't need to be
> preserved by system calls. So system call entry can mark the avx
> registers as 'don't save, restore as zeros'.
> Then if a process sleeps in a system call only the small part of the
> save area that contains the x87 regs would be accessed if/when
> the state is saved.
> On a context switch from a hardware interrupt all the user regs
> need saving - but the kernel stack use is minimal.
> Getting it right with lazy fpu save/restore is 'interesting'
> but not impossible.

That's not what I understood last time I read the specification. I perhaps
missed something obvious; I'll have to recheck.

[1] https://nxr.netbsd.org/xref/src/sys/arch/sparc64/sparc64/trap.c#696