[llvm] r207937 - AVX-512: minor change in rndscale intrinsic

[Demikhovsky, Elena]

I'm aligning llvm intrinsics with GCC built-ins. The change is really minor, related to semantics and not to correctness.
I'll try to give more info next time.

-  Elena

-----Original Message-----
From: Philip Reames [mailto:listmail at philipreames.com] 
Sent: Monday, May 05, 2014 19:26
To: Demikhovsky, Elena; llvm-commits at cs.uiuc.edu
Subject: Re: [llvm] r207937 - AVX-512: minor change in rndscale intrinsic

What is the motivation for this change?  Not objecting to the change (no context), but the submission comment needs more detail.

Philip

On 05/04/2014 06:35 AM, Elena Demikhovsky wrote:
> Author: delena
> Date: Sun May  4 08:35:37 2014
> New Revision: 207937
>
> URL: http://llvm.org/viewvc/llvm-project?rev=207937&view=rev
> Log:
> AVX-512: minor change in rndscale intrinsic
>
> Modified:
>      llvm/trunk/lib/Target/X86/X86InstrAVX512.td
>      llvm/trunk/test/CodeGen/X86/avx512-intrinsics.ll
>
> Modified: llvm/trunk/lib/Target/X86/X86InstrAVX512.td
> URL: 
> http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/X86/X86Instr
> AVX512.td?rev=207937&r1=207936&r2=207937&view=diff
> ======================================================================
> ========
> --- llvm/trunk/lib/Target/X86/X86InstrAVX512.td (original)
> +++ llvm/trunk/lib/Target/X86/X86InstrAVX512.td Sun May  4 08:35:37 
> +++ 2014
> @@ -1384,6 +1384,15 @@ def: Pat<(v8i64 (int_x86_avx512_mask_loa
>                    (bc_v8i64 (v16i32 immAllZerosV)), GR8:$mask)),
>          (VMOVDQU64rmkz (v8i1 (COPY_TO_REGCLASS GR8:$mask, VK8WM)), 
> addr:$ptr)>;
>   
> +def: Pat<(int_x86_avx512_mask_storeu_d_512 addr:$ptr, (v16i32 VR512:$src),
> +          GR16:$mask),
> +         (VMOVDQU32mrk addr:$ptr, (v16i1 (COPY_TO_REGCLASS GR16:$mask, VK16WM)),
> +            VR512:$src)>;
> +def: Pat<(int_x86_avx512_mask_storeu_q_512 addr:$ptr, (v8i64 VR512:$src),
> +          GR8:$mask),
> +         (VMOVDQU64mrk addr:$ptr, (v8i1 (COPY_TO_REGCLASS GR8:$mask, VK8WM)),
> +            VR512:$src)>;
> +
>   let AddedComplexity = 20 in {
>   def : Pat<(v8i64 (vselect VK8WM:$mask, (v8i64 VR512:$src),
>                              (bc_v8i64 (v16i32 immAllZerosV)))), @@ 
> -3754,7 +3763,7 @@ defm VRNDSCALEPSZ : avx512_rndscale<0x08
>                                   EVEX_CD8<32, CD8VF>;
>   
>   def : Pat<(v16f32 (int_x86_avx512_mask_rndscale_ps_512 (v16f32 VR512:$src1),
> -                   imm:$src2, (bc_v16f32 (v16i32 immAllZerosV)), (i16 -1),
> +                   imm:$src2, (v16f32 VR512:$src1), (i16 -1),
>                      FROUND_CURRENT)),
>                      (VRNDSCALEPSZr VR512:$src1, imm:$src2)>;
>   
> @@ -3764,7 +3773,7 @@ defm VRNDSCALEPDZ : avx512_rndscale<0x09
>                                   VEX_W, EVEX_CD8<64, CD8VF>;
>   
>   def : Pat<(v8f64 (int_x86_avx512_mask_rndscale_pd_512 (v8f64 VR512:$src1),
> -                  imm:$src2, (bc_v8f64 (v16i32 immAllZerosV)), (i8 -1),
> +                  imm:$src2, (v8f64 VR512:$src1), (i8 -1),
>                     FROUND_CURRENT)),
>                      (VRNDSCALEPDZr VR512:$src1, imm:$src2)>;
>   
>
> Modified: llvm/trunk/test/CodeGen/X86/avx512-intrinsics.ll
> URL: 
> http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/avx512
> -intrinsics.ll?rev=207937&r1=207936&r2=207937&view=diff
> ======================================================================
> ========
> --- llvm/trunk/test/CodeGen/X86/avx512-intrinsics.ll (original)
> +++ llvm/trunk/test/CodeGen/X86/avx512-intrinsics.ll Sun May  4 
> +++ 08:35:37 2014
> @@ -78,7 +78,7 @@ declare <8 x double> @llvm.x86.avx512.ma
>   
>   define <8 x double> @test7(<8 x double> %a) {
>   ; CHECK: vrndscalepd {{.*}}encoding: 
> [0x62,0xf3,0xfd,0x48,0x09,0xc0,0x0b]
> -  %res = call <8 x double> @llvm.x86.avx512.mask.rndscale.pd.512(<8 x 
> double> %a, i32 11, <8 x double> zeroinitializer, i8 -1, i32 4)
> +  %res = call <8 x double> @llvm.x86.avx512.mask.rndscale.pd.512(<8 x 
> + double> %a, i32 11, <8 x double> %a, i8 -1, i32 4)
>     ret <8 x double>%res
>   }
>   
> @@ -86,7 +86,7 @@ declare <16 x float> @llvm.x86.avx512.ma
>   
>   define <16 x float> @test8(<16 x float> %a) {
>   ; CHECK: vrndscaleps {{.*}}encoding: 
> [0x62,0xf3,0x7d,0x48,0x08,0xc0,0x0b]
> -  %res = call <16 x float> @llvm.x86.avx512.mask.rndscale.ps.512(<16 
> x float> %a, i32 11, <16 x float> zeroinitializer, i16 -1, i32 4)
> +  %res = call <16 x float> @llvm.x86.avx512.mask.rndscale.ps.512(<16 
> + x float> %a, i32 11, <16 x float> %a, i16 -1, i32 4)
>     ret <16 x float>%res
>   }
>   
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits

---------------------------------------------------------------------
Intel Israel (74) Limited

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.