PCLMULQDQ is used to accelerate the most time-consuming part of GHASH,
carry-less multiplication. More information about PCLMULQDQ can be
found at:
http://software.intel.com/en-us/articles/carry-less-multiplication-and-its-usage-for-computing-the-gcm-mode/
Because PCLMULQDQ changes XMM state, its
On Wed, Sep 16, 2009 at 09:35:46AM +0800, Huang Ying wrote:
> PCLMULQDQ is used to accelerate the most time-consuming part of GHASH,
> carry-less multiplication. More information about PCLMULQDQ can be
> found at:
>
> http://software.intel.com/en-us/articles/carry-less-multiplication-and-its-usage
On Mon, 19 Oct 2009 11:53:33 +0900 Herbert Xu
wrote:
> On Wed, Sep 16, 2009 at 09:35:46AM +0800, Huang Ying wrote:
> > PCLMULQDQ is used to accelerate the most time-consuming part of GHASH,
> > carry-less multiplication. More information about PCLMULQDQ can be
> > found at:
> >
> > http://softw
On Sat, Oct 31, 2009 at 05:30:15PM -0700, Andrew Morton wrote:
>
> x86_64 allmodconfig, GNU assembler 2.16.1:
>
> arch/x86/crypto/ghash-clmulni-intel_asm.S: Assembler messages:
> arch/x86/crypto/ghash-clmulni-intel_asm.S:103: Error: no such instruction:
> `pshufb %xmm5,%xmm0'
> arch/x86/crypto/g
* Herbert Xu wrote:
> - pshufb BSWAP, DATA
> + # pshufb BSWAP, DATA
> + .byte 0x66, 0x0f, 0x38, 0x00, 0xc5
A cleanup request: mind creating two macros for this PSHUFB MMX/SSE
instruction in arch/x86/include/asm/i387.h, instead of open-coding the
.byte sequences in ~6 places?
( Af
On Mon, Nov 02, 2009 at 08:50:39AM +0100, Ingo Molnar wrote:
>
> A cleanup request: mind creating two macros for this PSHUFB MMX/SSE
> instruction in arch/x86/include/asm/i387.h, instead of open-coding the
> .byte sequences in ~6 places?
I had a go at doing that, but it seems that i387.h isn't
* Herbert Xu wrote:
> On Mon, Nov 02, 2009 at 08:50:39AM +0100, Ingo Molnar wrote:
> >
> > A cleanup request: mind creating two macros for this PSHUFB MMX/SSE
> > instruction in arch/x86/include/asm/i387.h, instead of open-coding the
> > .byte sequences in ~6 places?
>
> I had a go at doing
On Mon, Nov 02, 2009 at 03:32:58PM +0100, Ingo Molnar wrote:
>
> Please use the standard construct and put an #ifndef __ASSEMBLY__ around
> it.
You mean like this?
diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h
index 0b20bbb..e22d237 100644
--- a/arch/x86/include/asm/i38
* Herbert Xu wrote:
> On Mon, Nov 02, 2009 at 03:32:58PM +0100, Ingo Molnar wrote:
> >
> > Please use the standard construct and put an #ifndef __ASSEMBLY__ around
> > it.
>
> You mean like this?
>
> diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h
> index 0b20bbb..e22d
On Mon, 2009-11-02 at 22:32 +0800, Ingo Molnar wrote:
> * Herbert Xu wrote:
>
> > On Mon, Nov 02, 2009 at 08:50:39AM +0100, Ingo Molnar wrote:
> > >
> > > A cleanup request: mind creating two macros for this PSHUFB MMX/SSE
> > > instruction in arch/x86/include/asm/i387.h, instead of open-codin
* Huang Ying wrote:
> On Mon, 2009-11-02 at 22:32 +0800, Ingo Molnar wrote:
> > * Herbert Xu wrote:
> >
> > > On Mon, Nov 02, 2009 at 08:50:39AM +0100, Ingo Molnar wrote:
> > > >
> > > > A cleanup request: mind creating two macros for this PSHUFB MMX/SSE
> > > > instruction in arch/x86/incl
On Mon, Nov 02, 2009 at 04:46:04PM +0100, Ingo Molnar wrote:
>
> Yeah. Or just a single block of:
>
>
> #ifndef __ASSEMBLY__
> ...
> #endif /* __ASSEMBLY__ */
>
> around the C bits - anything outside that is good for assembly as well.
OK I'll throw this into cryptodev:
commit 3b0d65969
On 11/03/2009 01:03 AM, Ingo Molnar wrote:
>>
>> .macro xmm_num opd xmm
>> .ifc \xmm,%xmm0
>> \opd = 0
>> .endif
>> .ifc \xmm,%xmm1
>> \opd = 1
>> .endif
>> .ifc \xmm,%xmm2
>> \opd = 2
>> .endif
>> .ifc \xmm,%xmm3
>> \opd = 3
>> .endif
>> .ifc \xmm,%xmm4
>> \opd = 4
>> .endif
>> .ifc \xmm,%xmm5
>>
13 matches
Mail list logo