Hi
Before I attempt to write a specialized code for the tooms (i imagine toom23
and toom33 would be the only ones worth the effort) , we better use what we
already have or can easily extend. The current toom eval phase doesn't take
advantage of sumdiff or addsub etc , rewrite it to do so , shou
The core Python code for handling the mpn/arch directory tree starts with
the default mpn/generic code and then, as it goes down the arch trees (x86w
and x86_64w), it overloads any *.h *.c, *.asm and *.s files based based on
file name matches (i.e. less any extension). The lists of files for a
On Jul 13, 10:54 am, Jason wrote:
> On Wednesday 13 July 2011 10:06:50 Jason wrote:
>
> > Hi
>
> > New karasub/add for nehalem , I did do a re-shuffle and it was pretty much
> > optimal , but the feedin/winddown code killed it ,
This has been and still is a problem , when you get close to the
mac
On Wednesday 13 July 2011 14:49:12 Cactus wrote:
> I now have the Python build configuration working for Visual Studio 2010.
> It scans the assembler directory tree to identify the possible builds,
> allows the user to select one and then constructs the project files needed
> for a Visual Studio 20
On Wednesday 13 July 2011 14:26:02 Jason wrote:
> On Wednesday 13 July 2011 14:01:39 Cactus wrote:
> > Darn, I already did the conversion :-(
> >
> > I don't have enough registers to use only the 32-bit registers so I have
> > to put stuff in r8 and r9 instead. Given this involved prefix opcodes
The existing Visual Studio 2010 IDE build can only build a subset of tne 20+
build configurations (gc, p0, p3, p4, k8, k10, core2, nehalem). The Python
code is run before Visual Studio 2010 and creates a Visual Studio 2010 build
project for the user selected configuration build (i.e. NOT a com
On Wed, Jul 13, 2011 at 9:49 AM, Cactus wrote:
> I now have the Python build configuration working for Visual Studio 2010.
> It scans the assembler directory tree to identify the possible builds,
> allows the user to select one and then constructs the project files needed
> for a Visual Studio 201
I now have the Python build configuration working for Visual Studio 2010.
It scans the assembler directory tree to identify the possible builds,
allows the user to select one and then constructs the project files needed
for a Visual Studio 2010 build.
It relies strictly on the assumption t
On Wednesday 13 July 2011 14:01:39 Cactus wrote:
> Darn, I already did the conversion :-(
>
> I don't have enough registers to use only the 32-bit registers so I have to
> put stuff in r8 and r9 instead. Given this involved prefix opcodes, I am
> wondering what I should o with your coded nops si
Darn, I already did the conversion :-(
I don't have enough registers to use only the 32-bit registers so I have to
put stuff in r8 and r9 instead. Given this involved prefix opcodes, I am
wondering what I should o with your coded nops since any alignment you are
seeking won't be the same wit
On Wednesday 13 July 2011 10:54:53 Jason wrote:
> On Wednesday 13 July 2011 10:06:50 Jason wrote:
> > Hi
> >
> > New karasub/add for nehalem , I did do a re-shuffle and it was pretty
> > much optimal , but the feedin/winddown code killed it , so at the mo
> > this is just the existing K10 code wit
Yep , cheers
Jason
On Wednesday 13 July 2011 12:43:14 Bill Hart wrote:
> I think I have fixed that now. You should be able to do this in future.
>
> On 13 July 2011 11:34, Jason wrote:
> > On Wednesday 29 June 2011 12:43:13 David Cleaver wrote:
> >> On 6/29/2011 6:16 AM, Bill Hart wrote:
> >>
I think I have fixed that now. You should be able to do this in future.
On 13 July 2011 11:34, Jason wrote:
> On Wednesday 29 June 2011 12:43:13 David Cleaver wrote:
>> On 6/29/2011 6:16 AM, Bill Hart wrote:
>> > I think it is sorted now anyhow.
>> >
>> > Bill.
>> >
>> > On 29 June 2011 12:15, Bi
On Wednesday 29 June 2011 12:43:13 David Cleaver wrote:
> On 6/29/2011 6:16 AM, Bill Hart wrote:
> > I think it is sorted now anyhow.
> >
> > Bill.
> >
> > On 29 June 2011 12:15, Bill Hart wrote:
> >> The only way to join a google group is the way google provides.
> >> Namely, go to the webpage
On Wednesday 13 July 2011 10:06:50 Jason wrote:
> Hi
>
> New karasub/add for nehalem , I did do a re-shuffle and it was pretty much
> optimal , but the feedin/winddown code killed it , so at the mo this is
> just the existing K10 code with the inc's replaced by an add and lea's ,
> just do a diff.
Hi
New karasub/add for nehalem , I did do a re-shuffle and it was pretty much
optimal , but the feedin/winddown code killed it , so at the mo this is just
the existing K10 code with the inc's replaced by an add and lea's , just do a
diff.
Going to a 3-way unroll on karasub free's up 2 registers
16 matches
Mail list logo