* Inefficient use of 64-bit addresses in Clang
@ 2019-08-06 12:30 Agner Fog
2019-08-12 9:20 ` Mark Geisert
0 siblings, 1 reply; 4+ messages in thread
From: Agner Fog @ 2019-08-06 12:30 UTC (permalink / raw)
To: cygwin
Clang is using 64-bit absolute addresses when accessing static data in
64-bit mode. This is inefficient because it requires an extra 10-bytes
long instruction for loading an address into a register every time it
needs to access static data. All other compilers use relative addresses.
Example:
> #include <immintrin.h>
>
> __m128d test (__m128d a) {
> Â Â Â __m128d b = _mm_add_pd(a, _mm_set1_pd(1.5));
> Â Â Â __m128d c = _mm_mul_pd(b, _mm_set1_pd(2.5));
> Â Â Â return c;
> }
Assembly output:
> .LCPI0_0:
>    .quad   4609434218613702656    # double 1.5
>    .quad   4609434218613702656    # double 1.5
> .LCPI0_1:
>    .quad   4612811918334230528    # double 2.5
>    .quad   4612811918334230528    # double 2.5
> Â Â Â .text
>    .globl   _Z4testDv2_d
>    .p2align   4, 0x90
> _Z4testDv2_d:Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â # @_Z4testDv2_d
> # BB#0:
>    vmovapd   (%rcx), %xmm0
>    movabsq   $.LCPI0_0, %rax
>    vaddpd   (%rax), %xmm0, %xmm0
>    movabsq   $.LCPI0_1, %rax
>    vmulpd   (%rax), %xmm0, %xmm0
> Â Â Â retq
Linux Clang uses 32-bit relative addresses:
>    vaddpd   .LCPI0_0(%rip), %xmm0, %xmm0
>    vmulpd   .LCPI0_1(%rip), %xmm0, %xmm0
> Â Â Â retq
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: Inefficient use of 64-bit addresses in Clang
2019-08-06 12:30 Inefficient use of 64-bit addresses in Clang Agner Fog
@ 2019-08-12 9:20 ` Mark Geisert
0 siblings, 0 replies; 4+ messages in thread
From: Mark Geisert @ 2019-08-12 9:20 UTC (permalink / raw)
To: cygwin
Agner Fog wrote:
> Clang is using 64-bit absolute addresses when accessing static data in 64-bit
> mode. This is inefficient because it requires an extra 10-bytes long instruction
> for loading an address into a register every time it needs to access static
> data. All other compilers use relative addresses.
>
> Example:
>
>> #include <immintrin.h>
>>
>> __m128d test (__m128d a) {
>> Â Â Â __m128d b = _mm_add_pd(a, _mm_set1_pd(1.5));
>> Â Â Â __m128d c = _mm_mul_pd(b, _mm_set1_pd(2.5));
>> Â Â Â return c;
>> }
>
> Assembly output:
>
>> .LCPI0_0:
>>    .quad   4609434218613702656    # double 1.5
>>    .quad   4609434218613702656    # double 1.5
>> .LCPI0_1:
>>    .quad   4612811918334230528    # double 2.5
>>    .quad   4612811918334230528    # double 2.5
>> Â Â Â .text
>>    .globl   _Z4testDv2_d
>>    .p2align   4, 0x90
>> _Z4testDv2_d:Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â # @_Z4testDv2_d
>> # BB#0:
>>    vmovapd   (%rcx), %xmm0
>>    movabsq   $.LCPI0_0, %rax
>>    vaddpd   (%rax), %xmm0, %xmm0
>>    movabsq   $.LCPI0_1, %rax
>>    vmulpd   (%rax), %xmm0, %xmm0
>> Â Â Â retq
>
> Linux Clang uses 32-bit relative addresses:
>
>>    vaddpd   .LCPI0_0(%rip), %xmm0, %xmm0
>>    vmulpd   .LCPI0_1(%rip), %xmm0, %xmm0
>> Â Â Â retq
This kind of bug report should probably go to the Clang bugtracker at
https://bugs.llvm.org . On that page you could enter "clang cygwin" in the
Quick Search field and see what known issues there are.
Clang is packaged for Cygwin by a very industrious and busy maintainer, but no
development on Clang is being done here (to my knowledge). So something like a
Cygwin packaging error that breaks Clang on Cygwin would be on-topic for this
list, but a problem at code generation level should go to the Clang developers.
..mark
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
^ permalink raw reply [flat|nested] 4+ messages in thread
[parent not found: <1436640898.268382790.1565602671940.JavaMail.root@zimbra54-e10.priv.proxad.net>]
end of thread, other threads:[~2019-08-14 5:51 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-08-06 12:30 Inefficient use of 64-bit addresses in Clang Agner Fog
2019-08-12 9:20 ` Mark Geisert
[not found] <1436640898.268382790.1565602671940.JavaMail.root@zimbra54-e10.priv.proxad.net>
2019-08-12 9:45 ` falk.tannhauser
2019-08-14 5:51 ` Agner Fog
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).