On Thursday 19 March 2009 18:43:44 Gonzalo Tornaria wrote:
> On Thu, Mar 19, 2009 at 2:35 PM, Jason Moxham
wrote:
> >> Yes, except config.guess may change it (but it shouldn't, as we
> >> discussed before this causes compilation failure, e.g. in a kvm
> >> virtual cpu which reports wrong cpuid).
On Thu, Mar 19, 2009 at 2:35 PM, Jason Moxham wrote:
>> Yes, except config.guess may change it (but it shouldn't, as we
>> discussed before this causes compilation failure, e.g. in a kvm
>> virtual cpu which reports wrong cpuid).
>>
>
> My understanding is that config.guess is to "tweek" the exac
On Thu, Mar 19, 2009 at 2:32 PM, Jason Moxham wrote:
>
> On Thursday 19 March 2009 14:31:55 Gonzalo Tornaria wrote:
>> On Wed, Mar 18, 2009 at 10:44 PM, Bill Hart
> wrote:
>> > Wait, how does it currently decide which ABI to use? Does configure
>> > decide that aside from what config.guess says.
On Thursday 19 March 2009 14:31:55 Gonzalo Tornaria wrote:
> On Wed, Mar 18, 2009 at 10:44 PM, Bill Hart
wrote:
> > Wait, how does it currently decide which ABI to use? Does configure
> > decide that aside from what config.guess says.
>
> Yes, except config.guess may change it (but it shouldn't,
On Thursday 19 March 2009 14:31:55 Gonzalo Tornaria wrote:
> On Wed, Mar 18, 2009 at 10:44 PM, Bill Hart
wrote:
> > Wait, how does it currently decide which ABI to use? Does configure
> > decide that aside from what config.guess says.
>
> Yes, except config.guess may change it (but it shouldn't,
On Mar 19, 2:37 pm, Bill Hart wrote:
> By the way, are you sure it is possible to have lahf and non-lahf
> netburst 64 bit chips?
>
> I thought only very early p4's had this problem, i.e. all 32 bits.
Hi Bill
This is definitely a problem on some 64-bit processors.
Brian
--~--~-~
By the way, are you sure it is possible to have lahf and non-lahf
netburst 64 bit chips?
I thought only very early p4's had this problem, i.e. all 32 bits.
Bill.
2009/3/19 Jason Moxham :
>
>
> A summary of the new proposed x86_64 directorys
>
> base directory when all we know is that it is 64bi
On Wed, Mar 18, 2009 at 10:44 PM, Bill Hart wrote:
> Wait, how does it currently decide which ABI to use? Does configure
> decide that aside from what config.guess says.
Yes, except config.guess may change it (but it shouldn't, as we
discussed before this causes compilation failure, e.g. in a kv
Aw come on, AMD hasn't done too badly. At least now they have "a
portfolio of products suited to a range of needs" or something like
that. That's much better than the kx naming system.
Anyhow, given the justification you already gave, these names seem good to me.
Bill.
2009/3/19 Jason Moxham :
A summary of the new proposed x86_64 directorys
base directory when all we know is that it is 64bit x86 and for any code that
is common to all x86_64 chips
x86_64
For the AMD chips AKA Opteron=K8 Phenom=k10
x86_64/k8
x86_64/k8/k10
For the Intel core2 chips AKA merom=core2-65nm dunnington=pen
> By the time we are in config.guess , we allready know that we are either 32 or
> 64
>
Pff, gosh I'm dumb sometimes. Of course we do.
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
"mpir-devel" group.
To post to t
On Thursday 19 March 2009 01:44:19 Bill Hart wrote:
> 2009/3/19 Jason Moxham :
> > On Tuesday 17 March 2009 18:40:30 Bill Hart wrote:
> >> Jason,
> >>
> >> I had a look over this and it looks brilliant. This really deals with
> >> a number of tickets, which is great!
> >>
> >> Some comments:
> >>
2009/3/19 Jason Moxham :
>
> On Tuesday 17 March 2009 18:40:30 Bill Hart wrote:
>> Jason,
>>
>> I had a look over this and it looks brilliant. This really deals with
>> a number of tickets, which is great!
>>
>> Some comments:
>>
>> The name athlon seems inconsistent with k5, k6, k62, k63, k8, k10
On Tuesday 17 March 2009 18:40:30 Bill Hart wrote:
> Jason,
>
> I had a look over this and it looks brilliant. This really deals with
> a number of tickets, which is great!
>
> Some comments:
>
> The name athlon seems inconsistent with k5, k6, k62, k63, k8, k10. It
> seems to me that Athlon and k7
On Tue, Mar 17, 2009 at 3:40 PM, Bill Hart wrote:
> I also wonder if there is a way to distinguish intel family 15 and 22?
family=6 for core2 always (family=15 is p4)
model=15 for "normal" core2 and xeon
model=22 (=16h), for celeron (I assume core2-celeron)
model=23 (=17h), for "extreme" core2,
One of the issues which will confront us soon is the fact that Intel
have stopped upping clock rates and have started introducing more
cores, and AMD have followed suit.
I predict that eventually we are going to want to start distinguishing
these. Perhaps we need:
k8, k8x2, , core, corex2, core2
For amd we'd have to have:
K5, K6 (mmx), k62 (k6-2, mmx, 3dnow), k63 (k6-III, mmx, 3dnow), k7
(mmx, 3dnow), k7sse, k8 (all sse2), k8sse3, k10, geode?
It looks to me like k9 never materialised, and if it did, it
represents dual core AMD k8's. k11 is supposedly Turion Ultra.
Bill.
2009/3/17 Bill
Jason,
I had a look over this and it looks brilliant. This really deals with
a number of tickets, which is great!
Some comments:
The name athlon seems inconsistent with k5, k6, k62, k63, k8, k10. It
seems to me that Athlon and k7 are synonymous, so shouldn't we use the
latter?
I also wonder if
18 matches
Mail list logo