> Date: Wed, 10 Jun 2020 20:08:31 +0200
> From: Christian Weisgerber
>
> Next try.
> Optimized versions for kernel ffs(3) on arm64, powerpc, powerpc64.
>
> I have tested arm64; cwen@ has tested powerpc in userland.
> powerpc64 is copied from powerpc.
>
> ok?
ok kettenis@
> Index:
Next try.
Optimized versions for kernel ffs(3) on arm64, powerpc, powerpc64.
I have tested arm64; cwen@ has tested powerpc in userland.
powerpc64 is copied from powerpc.
ok?
Index: lib/libkern/arch/arm64/ffs.S
===
RCS file:
Mark Kettenis wrote:
> > Date: Tue, 9 Jun 2020 23:43:50 +0200
> > From: Christian Weisgerber
> >
> > Mark Kettenis:
> >
> > > Unfortunately that doesn't quite work. At least in my build it
> > > doesn't pick up .c files in the linker/arch directories.
> > >
> > > > Index:
> Date: Tue, 9 Jun 2020 23:43:50 +0200
> From: Christian Weisgerber
>
> Mark Kettenis:
>
> > Unfortunately that doesn't quite work. At least in my build it
> > doesn't pick up .c files in the linker/arch directories.
> >
> > > Index: lib/libkern/arch/arm64/ffs.c
>
> I was certain I had
Mark Kettenis:
> Unfortunately that doesn't quite work. At least in my build it
> doesn't pick up .c files in the linker/arch directories.
>
> > Index: lib/libkern/arch/arm64/ffs.c
I was certain I had checked this, but indeed it doesn't work.
The Makefile rules are generated from
> From: Christian Weisgerber
> Date: Tue, 9 Jun 2020 18:39:45 - (UTC)
>
> Here are optimized ffs(3) implementations for
> * arm64 (superseding the earlier ffs.S)
> * powerpc
> * powerpc64
>
> arm64 tested by myself, powerpc tested by cwen@.
>
> OK?
>
> (Some other archs fell through.
On 2020-06-09, Christian Weisgerber wrote:
> Here are optimized ffs(3) implementations for
> * arm64 (superseding the earlier ffs.S)
> * powerpc
> * powerpc64
> +int ffs(int x)
Oops, I'm going to change that to
int
ffs(int x)
before commit.
--
Christian "naddy" Weisgerber
On Mon, Jun 08, 2020 at 11:16:59PM +0200, Christian Weisgerber wrote:
> I blame dlg@ for making me scrutinize the POWER7 instruction set,
> which led me to the clz instruction, which led me to ffs(). I
> wanted to add this to libc, but then realized the futility because
> the compiler already
Either works for me.
Honestly
- either one is hard to understand without referring to the
docs (and then realizing power numbers bits backwards).
- It is simple enough it has to work or there will be chaos
and I'm not that afraid of more poeple seeing RETGUARD stubs :)
Christian Weisgerber
On Mon, 08 Jun 2020 22:35:42 -, Christian Weisgerber wrote:
> Style question:
> Since this mostly comes down to embedding a single special instruction
> in between normal C operations, I wonder whether I should just do .c
> with asm() instead of .S, and leave all the boilerplate to the
>
On 2020-06-08, Christian Weisgerber wrote:
> More archs to come...
Style question:
Since this mostly comes down to embedding a single special instruction
in between normal C operations, I wonder whether I should just do .c
with asm() instead of .S, and leave all the boilerplate to the
compiler?
Here is an optimized kernel ffs(3) for arm64.
I blame dlg@ for making me scrutinize the POWER7 instruction set,
which led me to the clz instruction, which led me to ffs(). I
wanted to add this to libc, but then realized the futility because
the compiler already inlines its optimized copy of
12 matches
Mail list logo
