Excellent. I'm about to leave for the MAA-AMS Joint Meetings, so I
won't get a chance to bench your K10 stuff on a Dunnington until I get
back. I'll report on what speeds I get.
--jason
On Sun, Jan 4, 2009 at 5:44 PM, wrote:
>
> On Sunday 04 January 2009 01:57:05 Jason Martin wrote:
>> Fair
On Sunday 04 January 2009 01:57:05 Jason Martin wrote:
> Fair enough. I don't have any strong opinions on the matter. I
> should probably update the Intel code to include your popcount routine
> since the new Intel cores support the SSE 4.1 instruction for popcount
> on the xmm registers.
>
> --
Fair enough. I don't have any strong opinions on the matter. I
should probably update the Intel code to include your popcount routine
since the new Intel cores support the SSE 4.1 instruction for popcount
on the xmm registers.
--jason
On Sat, Jan 3, 2009 at 8:02 PM, wrote:
>
> On Sunday 04
On Sunday 04 January 2009 00:57:44 ja...@njkfrudils.plus.com wrote:
> On Sunday 04 January 2009 00:36:46 Jason Martin wrote:
> > Alternatively, we could stop trying to identify chips by marketing
> > brands and just use the values returned by CPUID. This would create a
> > lot of duplicated code
On Sunday 04 January 2009 00:36:46 Jason Martin wrote:
> Alternatively, we could stop trying to identify chips by marketing
> brands and just use the values returned by CPUID. This would create a
> lot of duplicated code in sub-directories, but disk space is cheap.
> So, would something like:
>
>
That seems complicated. In order to figure out which assembly files to
modify, one needs to know the CPUID of the processors one wishes to
work on.
Also, given that there are probably 20 odd Intel and 10 AMD
family/model combinations this is a lot of directories and unnecessary
duplication. I thi
Alternatively, we could stop trying to identify chips by marketing
brands and just use the values returned by CPUID. This would create a
lot of duplicated code in sub-directories, but disk space is cheap.
So, would something like:
mpn/x86_64//
work for our configuration?
Jason Worth Martin
Ass
I think that features such as SSE should be tested for after testing
for the main chip core. So under /mpn/x86_64/k8 you'd have directories
for any features not available on all k8's.
Bill.
2009/1/3 mabshoff :
>
>
>
> On Jan 3, 2:25 pm, jason wrote:
>> On Jan 3, 9:00 am, "Bill Hart" wrote:
>
>
On Jan 3, 2:25 pm, jason wrote:
> On Jan 3, 9:00 am, "Bill Hart" wrote:
Hi,
> > The new intel machines. And I don't know if all Dunnington's use the
> > same family/system CPUID etc. So there might be mutiple CPUID's we
> > need to add to config.guess.
>
> > Bill.
>
> We should change the lo
2009/1/3 jason :
>
>
>
> On Jan 3, 9:00 am, "Bill Hart" wrote:
>> The new intel machines. And I don't know if all Dunnington's use the
>> same family/system CPUID etc. So there might be mutiple CPUID's we
>> need to add to config.guess.
>>
>> Bill.
>
> We should change the lowest common denominat
On Jan 3, 9:00 am, "Bill Hart" wrote:
> The new intel machines. And I don't know if all Dunnington's use the
> same family/system CPUID etc. So there might be mutiple CPUID's we
> need to add to config.guess.
>
> Bill.
We should change the lowest common denominator on a x86_64 system to
someth
I've now added support for Phenom in config.guess. This allows MPIR to
build on the CUDA machine. Just check out the latest svn.
Bill.
2009/1/3 Bill Hart :
> The new intel machines. And I don't know if all Dunnington's use the
> same family/system CPUID etc. So there might be mutiple CPUID's we
The new intel machines. And I don't know if all Dunnington's use the
same family/system CPUID etc. So there might be mutiple CPUID's we
need to add to config.guess.
Bill.
On 03/01/2009, mabshoff wrote:
>
>
>
> On Jan 3, 12:15 am, "Bill Hart" wrote:
>
> Hi Bill,
>
>> I'll fix this.
>
> Thanks.
On Jan 3, 12:15 am, "Bill Hart" wrote:
Hi Bill,
> I'll fix this.
Thanks.
> It's just a CPUID problem.
Yep, that is what I assumed.
> We actually already have a
> ticket for it. There are lots of new machines we should fix detection
> for.
Other than the new Dunnington core which machines
I'll fix this. It's just a CPUID problem. We actually already have a
ticket for it. There are lots of new machines we should fix detection
for.
Bill.
2009/1/3 mabshoff :
>
> Hi,
>
> the "new" Jason [I guess we need to find some terminology here :)]
> just made aware of the following problem on t
15 matches
Mail list logo
