Re: Shark suggests SSE on code that uses .... SSE
Re: Shark suggests SSE on code that uses .... SSE
- Subject: Re: Shark suggests SSE on code that uses .... SSE
- From: Benson Margulies <email@hidden>
- Date: Tue, 14 Apr 2009 17:11:46 -0400
The wording, if you ask, me, could be a hair clearer, but I guess I'm
splitting a hair.
On Tue, Apr 14, 2009 at 4:43 PM, Rick Altherr <email@hidden> wrote:
>
> On Apr 14, 2009, at 1:08 PM, Benson Margulies wrote:
>
>> 5.1% 0x100410dec movq (%rdx,
>> %rcx, 8), %rdi ! Loop start[6],
>> SSE bt_ape_decoder.cpp:167
>> 0.0% 0x100410df0 movq (%rdi),
>> %rax bt_ape_decoder.cpp:168
>> 0.0% 0x100410df3 movl %r14d,
>> %r8d bt_ape_decoder.cpp:168
>> 0x100410df6 movl %r12d,
>> �x bt_ape_decoder.cpp:168
>> 5.0% 0x100410df9 movl %r13d,
>> �x bt_ape_decoder.cpp:168
>> 0x100410dfc movq
>> -168(%rbp), %rsi bt_ape_decoder.cpp:168
>> 0.0% 0x100410e03 call +32(%rax)
>> bt_ape_decoder.cpp:168
>> 5.1% 0x100410e06 movss %xmm0,
>> -68(%rbp) bt_ape_decoder.cpp:168
>> 0.0% 0x100410e0b movss
>> -68(%rbp), %xmm0 bt_ape_decoder.cpp:168
>> 6.9% 0x100410e10 addss
>> -56(%rbp), %xmm0 bt_ape_decoder.cpp:168
>> 10.2% 0x100410e15 movss %xmm0,
>> -56(%rbp) bt_ape_decoder.cpp:168
>> 6.1% 0x100410e1a movl �x,
>> �x stl_vector.h:166
>> 0.0% 0x100410e1c movq +8(%r15),
>> %rax stl_vector.h:166
>> 0.0% 0x100410e20 movq (%rax),
>> %rdx stl_vector.h:343
>> 0.0% 0x100410e23 movq +8(%rax),
>> %rax stl_vector.h:361
>> 6.0% 0x100410e27 incq %rbx
>> bt_ape_decoder.cpp:403
>> 0.0% 0x100410e2a subq %rdx,
>> %rax bt_ape_matrix.h:166
>> 0.0% 0x100410e2d sarq $3, %rax
>> bt_ape_matrix.h:166
>> 0.0% 0x100410e31 cmpq %rax,
>> %rcx bt_ape_matrix.h:166
>> 6.1% 0x100410e34 jb
>> 0x0000000100410dec
>> <BT_ApeDecoder::forwardViterbi(BT_ApeDecodeState&)+478> Loop
>> end[6] bt_ape_matrix.h:166
>> _______________________________________________
>> Do not post admin requests to the list. They will be ignored.
>> Xcode-users mailing list (email@hidden)
>> Help/Unsubscribe/Update your Subscription:
>>
>> This email sent to email@hidden
>
>
> SSE is being used for scalar operations here. Rather than use x87, the
> compiler defaults to using SSE scalar operations for floating point. Shark
> is suggesting you use vector SSE operations.
>
> --
> Rick Altherr
> Performance Tools
> Development Technologies
> email@hidden
>
>
>
>
>
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Xcode-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden