On 20 May 2010 02:58, Jason Moxham <[email protected]> wrote:
> On Thursday 20 May 2010 02:11:11 Bill Hart wrote:
>> There's a new Opteron on the block. family 16 model 8. It's an Opteron
>> 8439 SE. We don't have it in cpuid.c yet.
>>
>
> Looks like another K102 , doesn't look like any differences for us , although
> we dont know until we test it.
Passes make check once the cpuid is put in. See below for bench. This
particular machine is 2.8GHz(owner is in cc). I didn't make tune, but
that shouldn't be needed.
Index: cpuid.c
===================================================================
--- cpuid.c (revision 2928)
+++ cpuid.c (working copy)
@@ -120,7 +120,8 @@
case 16:
if (model == 2) { CPUIS(k10);break;}
if (model == 4) { CPUIS(k102);break;}
- break;
+ if (model == 8) { CPUIS(k102);break;}
+ break;
}
}
else if (strcmp (vendor_string, "CentaurHauls") == 0)
> By the way the difference between what we call the K102 and the K10 , is the
> "same" as the difference between early K8 and late K8's , a slight difference
> in
> L1-data cache and store forwarding , which does effect some functions eg
> copy,addadd,lshiftcom , some more than others
>
>> Also we currently have a symbol in the library, mpn_sqr_n and a
>> #define in mpir.h to rename it mpn_sqr in line with gmp. But that
>> isn't sufficient. It breaks gcc on machines where gcc has been built
>> with dynamic libs. On the face of it this shouldn't affect anyone, but
>> in fact it will affect lots of people. The only solution I see is to
>> rename the library symbol.
>
> we can just export both names , so we are compatible with the new gmp AND the
> old
Yeah, I noticed that.
Anyhow, good idea, but not in mpir.h. It has to be in the library itself.
>
>>
>> Also, GMP defines mpn_copyi and mpn_copyd. On generic x86_64 machines,
>> we don't seem to have any assembly for this. There's no fallback
>> generic C symbol in the lib.
>
> we can the fallback C code that a fat build uses , with the new K102 cpuid
> above there is no current generic x86_64 cpu :)
True.
>
>>
>> Bill.
>
AuthenticAMD Family 16 Model 8 Stepping 0
Six-Core AMD Opteron(tm) Processor 8439 SE
Speed: 0.80 GHz (reported)
Category base
Program multiply (weight 1.00)
128 128 => 62415242
512 512 => 14910433
8192 8192 => 174206
131072 131072 => 3012
2097152 2097152 => 108
128 128 => 57065363
512 512 => 13512577
8192 8192 => 120162
131072 131072 => 2251
2097152 2097152 => 74.3
15000 10000 => 59494
20000 10000 => 46812
30000 10000 => 27429
16777216 512 => 458
16777216 262144 => 13.7 => 37401,46751
Program divide (weight 1.00)
8192 32 => 1654949
8192 64 => 1526756
8192 128 => 926962
8192 4096 => 297892
131072 65536 => 3232
8388608 4194304 => 11.5
8192 8064 => 8473345
16777216 262144 => 4.76 => 31816,39770
Program divexact (weight 1.00)
8192 32 => 1872460
8192 64 => 1853921
8192 128 => 786511
8192 4096 => 360487
131072 65536 => 3813
8388608 4194304 => 11.8
8192 8064 => 14242122
16777216 262144 => 7.84 => 38603,48254
Program gcd (weight 0.50)
128 128 => 2284479
512 512 => 227241
8192 8192 => 5964
131072 131072 => 111
1048576 1048576 => 6.36 => 4656, 5819
Program gcdext (weight 0.50)
128 128 => 988477
512 512 => 175371
8192 8192 => 4609
131072 131072 => 68.3
1048576 1048576 => 4.31 => 2981, 3726
Program root (weight 0.30)
128 5 => 1142240
512 3 => 514009
8192 11 => 158261
131072 3 => 4138
1048576 3 => 190 => 37396,46745
Program fac_ui (weight 0.20)
128 128 => 1588752
1512 1512 => 53386
15000 15000 => 1249
1000010 1000010 => 4.39
2123456 2123456 => 1.63 => 945, 1182 => 18481,23101
Category app
Program rsa (weight 1.00)
512 => 24338
1024 => 5200
2048 => 883 => 4817, 6021
Program pi (weight 1.00)
10000 => 458
100000 => 27.0
1000000 => 1.61 => 27.1, 33.9
Program bpsw (weight 1.00)
1024 => 1138
4096 => 32.4
16384 => 0.840 => 31.4, 39.2
Program wagstaff (weight 1.00)
1024 => 2626
4096 => 105
16384 => 3.36 => 97.5, 122 => 141, 177 => 1617, 2021
--
You received this message because you are subscribed to the Google Groups
"mpir-devel" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/mpir-devel?hl=en.
