Commit Graph

161 Commits

Author SHA1 Message Date
Craig Topper d7e5b21914 [X86] Remove extra escaped new lines in intrinsic headers left over from an earlier conversion away from a macro. NFC
llvm-svn: 286756
2016-11-13 07:26:31 +00:00
Craig Topper 2c8f49e67b [AVX-512] Use scalar vfmsub/vfnmsub mask3 intrinsics instead of inverting the mask argument of a vfmadd intrinsic.
Summary: Inverting the mask argument does not reflect the intended semantics of the intrinsic.

Reviewers: igorb, delena

Subscribers: cfe-commits

Differential Revision: https://reviews.llvm.org/D26019

llvm-svn: 286733
2016-11-12 23:24:34 +00:00
Craig Topper 1a44193afd [AVX-512] Convert the rest of the masked shift by immediate and by single element builtins over to the newly added unmasked builtins and a select.
This should also fix PR30691 since the new builtins are handled like the legacy builtins in the backend.

llvm-svn: 286714
2016-11-12 07:16:59 +00:00
Ayman Musa e60a41ca28 [X86][AVX512][Clang] Add support for mask_{move|store|load}_s{s/d} and int2mask/mask2int intrinsics.
Differential Revision: https://reviews.llvm.org/D26021

llvm-svn: 286229
2016-11-08 12:00:30 +00:00
Craig Topper 08bf53ffda [AVX-512] Remove masked vector insert builtins and replace with native shufflevectors and selects.
Unfortunately, the backend currently doesn't fold masks into the instructions correctly when they come from these shufflevectors. I'll work on that in a future commit.

llvm-svn: 285667
2016-11-01 05:47:56 +00:00
Craig Topper 93ffabd28d [AVX-512] Remove masked vector extract builtins and replace with native shufflevectors and selects.
Unfortunately, the backend currently doesn't fold masks into the instructions correctly when they come from these shufflevectors. I'll work on that in a future commit.

llvm-svn: 285540
2016-10-31 04:30:56 +00:00
Michael Zuckerman d343697f1e Fixing "type" issue for (epi32)
and replaceing hardcoded inf with clang builtin inf "__builtin_inff()" for float ({max|min}_{pd|ps})

llvm-svn: 285519
2016-10-30 14:54:05 +00:00
Michael Zuckerman 25eb420233 [X86][AVX512][Clang][Intrinsics][reduce] Adding missing reduce (max|min) intrinsics to Clang .
After LGTM and Check-all 

Vector-reduction arithmetic accepts vectors as inputs and produces 
scalars as outputs.This class of vector operation forms the basis 
of many scientific computations. In vector-reduction arithmetic, 
the evaluation off is independent of the order of the input elements of V.

Reviewer: 1. craig.topper 
          2. igorb

Differential Revision: https://reviews.llvm.org/D25988

llvm-svn: 285493
2016-10-29 10:29:20 +00:00
Michael Zuckerman edd99eb07a 1. Fixing small types issue (PD|PS) (reduce) .
2. Cosmetic changes

llvm-svn: 285405
2016-10-28 15:16:03 +00:00
Craig Topper f202365910 [AVX-512] Fix the operand order for all calls to __builtin_ia32_vfmaddss3_mask.
Summary: The preserved input should be the first argument and the vector inputs should be in the same order as the intrinsics it is used to implement.

Reviewers: igorb, delena

Subscribers: cfe-commits

Differential Revision: https://reviews.llvm.org/D25902

llvm-svn: 285175
2016-10-26 05:35:38 +00:00
Michael Zuckerman facb37cabf [X86][AVX512][Clang][Intrinsics][reduce] Adding missing reduce (Operators: +,*,&&,||) intrinsics to Clang
Committed after LGTM and check-all

 
Vector-reduction arithmetic accepts vectors as inputs and produces scalars as outputs.
This class of vector operation forms the basis of many scientific computations.
In vector-reduction arithmetic, the evaluation off is independent of the order of the input elements of V.
 
Used bisection method. At each step, we partition the vector with previous
step in half, and the operation is performed on its two halves.
This takes log2(n) steps where n is the number of elements in the vector.
 
Reviwer: 1. igorb
         2. craig.topper    
Differential Revision: https://reviews.llvm.org/D25527

llvm-svn: 285054
2016-10-25 07:56:04 +00:00
Michael Zuckerman 33bd5b235b revert r284963
because new test file is failing in some OS.
test/CodeGen/avx512-reduceIntrin.c

llvm-svn: 284967
2016-10-24 11:30:23 +00:00
Michael Zuckerman 98cb041891 [X86][AVX512][Clang][Intrinsics][reduce] Adding missing reduce (Operators: +,*,&&,||) intrinsics to Clang
Committed after LGTM and check-all

Vector-reduction arithmetic accepts vectors as inputs and produces scalars as outputs.
This class of vector operation forms the basis of many scientific computations.
In vector-reduction arithmetic, the evaluation off is independent of the order of the input elements of V.

Used bisection method. At each step, we partition the vector with previous
step in half, and the operation is performed on its two halves.
This takes log2(n) steps where n is the number of elements in the vector.

Differential Revision: https://reviews.llvm.org/D25527

llvm-svn: 284963
2016-10-24 10:53:20 +00:00
Craig Topper 0c5da26572 [AVX-512] Replace 512-bit pmovzx/sx builtins with native IR.
llvm-svn: 284936
2016-10-23 07:35:47 +00:00
Michael Zuckerman 9e43ccfe68 [Clang][AVX512][BuiltIn]Adding missing intrinsics move_{sd|ss} to clang
Differential Revision: http://reviews.llvm.org/D21021

llvm-svn: 283314
2016-10-05 12:56:06 +00:00
Craig Topper c4a8228bcc [AVX-512] Use native IR for masked 512-bit add/sub/mul/div ps/pd intrinsics when rounding mode isn't used.
llvm-svn: 283073
2016-10-02 17:43:00 +00:00
Ayman Musa 17a2819b05 Update to commit r282488, fix the buildboot failure.
llvm-svn: 282492
2016-09-27 15:37:31 +00:00
Ayman Musa 2e250e8845 [avx512] Add aliases to some missing avx512 intrinsics.
Differential Revision:https: //reviews.llvm.org/D24961

llvm-svn: 282488
2016-09-27 14:06:32 +00:00
Craig Topper f43e4a1728 [AVX-512] Remove masked integer mullo builtins and replace with native IR.
llvm-svn: 280597
2016-09-03 19:19:49 +00:00
Craig Topper 0e18976b8d [AVX-512] Remove masked integer add/sub builtins and replace with native IR.
llvm-svn: 280596
2016-09-03 18:29:35 +00:00
Asaf Badouh 356bb76809 [X86][AVX512F] minor fix of the parameter names
add "__" prefix

Bug 28842 https://llvm.org/bugs/show_bug.cgi?id=29040

Differential Revision: https://reviews.llvm.org/D23753


 

llvm-svn: 279392
2016-08-21 07:56:47 +00:00
Asaf Badouh 2f344b788c [AVX512] integer comparisions enumeration.
fix Bug 28842 https://llvm.org/bugs/show_bug.cgi?id=28842

Differential Revision: https://reviews.llvm.org/D22212

 

llvm-svn: 277955
2016-08-07 10:43:04 +00:00
Eric Christopher b638558e12 Remove unused variable.
Fixes PR28761.

llvm-svn: 277221
2016-07-29 22:11:11 +00:00
Craig Topper 45db56c375 [X86] Add missing __x86_64__ qualifiers on a bunch of intrinsics that assume 64-bit GPRs are available.
Usages of these intrinsics in a 32-bit build results in assertions in the backend.

llvm-svn: 276249
2016-07-21 07:38:39 +00:00
Asaf Badouh a0b6f8fb56 [X86][AVX512F] minor fix of the parameter names
add "__" prefix

llvm-svn: 275384
2016-07-14 08:40:30 +00:00
Craig Topper 4d61a3c2d8 [AVX512] Replace masked AND/OR/XOR intrinsics with native code and remove the builtins.
llvm-svn: 275049
2016-07-11 06:14:18 +00:00
Craig Topper 6e76fb61a7 [X86] Use __butilin_shufflevector for 512-bit shufps intrinsics.
llvm-svn: 275012
2016-07-10 05:57:21 +00:00
Simon Pilgrim f5a8837e1b [X86][AVX512] Converted the VBROADCAST intrinsics to generic IR
llvm-svn: 274544
2016-07-05 12:59:33 +00:00
Asaf Badouh 136332888a [X86][AVX512F] add float/double abs intrinsics
add abs intrinsics that use native LLVM-IR.
change _mm512_mask[z]_and_epi{32|64} to use select intrinsic

Differential Revision: http://reviews.llvm.org/D21973

llvm-svn: 274542
2016-07-05 12:24:14 +00:00
Asaf Badouh f9cdb8de7a [AVX512] minor fix in sqrt{ss|sd} intrinsics arguments
Differential Revision: http://reviews.llvm.org/D21988

llvm-svn: 274541
2016-07-05 11:36:21 +00:00
Craig Topper 2a383c9273 [X86] Use undefined instead of setzero in shufflevector based intrinsics when the second source is unused. Rewrite immediate extractions in shuffle intrinsics to be in ((c >> x) & y) form instead of ((c & z) >> x). This way only x varies between each use instead of having to vary x and z.
llvm-svn: 274525
2016-07-04 22:18:01 +00:00
Simon Pilgrim 427154db2a [X86][AVX512] Converted the VSHUFPD intrinsics to generic IR
llvm-svn: 274523
2016-07-04 21:30:47 +00:00
Simon Pilgrim 30db811526 [X86][AVX512] Converted the VPERMPD/VPERMQ intrinsics to generic IR
llvm-svn: 274502
2016-07-04 13:34:44 +00:00
Simon Pilgrim 17388f2569 [X86][AVX512] Converted the VPERMILPD/VPERMILPS intrinsics to generic IR
llvm-svn: 274492
2016-07-04 11:06:15 +00:00
Simon Pilgrim 275d721485 [X86][AVX512] Converted the MOVDDUP/MOVSLDUP/MOVSHDUP masked intrinsics to generic IR
llvm companion patch imminent

llvm-svn: 274442
2016-07-02 17:16:25 +00:00
Craig Topper 79f53ca0b5 [AVX512] Replace masked unpack builtins with shufflevector and selects.
llvm-svn: 273533
2016-06-23 06:36:42 +00:00
Michael Zuckerman 223676d2cc [Clang][AVX512][intrinsics] Adding missing intrinsics div_pd and div_ps
Differential Revision: http://reviews.llvm.org/D20626

llvm-svn: 272658
2016-06-14 12:38:58 +00:00
Asaf Badouh 880f0c252b [X86][AVX512F] bugfix - sqrtps should get __mask16 as mask parameter
CR: Michael Zuckerman
llvm-svn: 272549
2016-06-13 15:15:57 +00:00
Simon Pilgrim beca5f295c [Clang][X86] Convert non-temporal store builtins to generic __builtin_nontemporal_store in headers
We can now use __builtin_nontemporal_store instead of target specific builtins for naturally aligned nontemporal stores which avoids the need for handling in CGBuiltin.cpp

The scalar integer nontemporal (unaligned) store builtins will have to wait as __builtin_nontemporal_store currently assumes natural alignment and doesn't accept the 'packed struct' trick that we use for normal unaligned load/stores.

The nontemporal loads require further backend support before we can safely convert them to __builtin_nontemporal_load

Differential Revision: http://reviews.llvm.org/D21272

llvm-svn: 272540
2016-06-13 09:57:52 +00:00
Craig Topper 7cc9263ec2 [AVX512] Implement masked and 512-bit pshufd intrinsics directly with __builtin_shufflevector and __builtin_ia32_select.
llvm-svn: 272467
2016-06-11 12:50:19 +00:00
Craig Topper 3a0c7260f4 [X86] Add void to the argument list of intrinsics that don't take arguments since empty argument list mean something else in C.
llvm-svn: 272244
2016-06-09 05:14:28 +00:00
Igor Breger aadb876200 [AVX512] Emit select instruction instead of using x86 specific instrinsics.
This will allow us to remove the x86 instrinics from the backend.

Differential Revision: http://reviews.llvm.org/D21060

llvm-svn: 272141
2016-06-08 13:59:20 +00:00
Craig Topper f3efec65bb [AVX512] Reformat macro intrinsics, ensure arguments have proper typecasts, ensure result is typecasted back to the generic types.
llvm-svn: 272119
2016-06-08 06:08:07 +00:00
Craig Topper 605894985f [X86] Put parentheses around macro arguments in intrinsics.
llvm-svn: 272118
2016-06-08 06:08:04 +00:00
Michael Zuckerman 95721ac863 [Clang][AVX512]Adding set4 intrinsics
Differential Revision: http://reviews.llvm.org/D20866

llvm-svn: 271835
2016-06-05 15:43:30 +00:00
Michael Zuckerman f36f6eb036 [Clang][AVX512][Intrinsics] Adding two definitions _mm512_setzero and _mm512_setzero_epi32
Differential Revision: http://reviews.llvm.org/D20871

llvm-svn: 271832
2016-06-05 15:12:52 +00:00
Craig Topper 6a77b62640 [X86] Use unsigned types for vector arithmetic in intrinsics to avoid undefined behavior for signed integer overflow.
This is really only needed for addition, subtraction, and multiplication, but I did the bitwise ops too for overall consistency. Clang currently doesn't set NSW for signed vector operations so the undefined behavior shouldn't happen today.

llvm-svn: 271778
2016-06-04 05:43:41 +00:00
Asaf Badouh 89f657611c [X86][AVX512] add intrinsics of Scalar FP to integer
Differential Revision: http://reviews.llvm.org/D20861

llvm-svn: 271499
2016-06-02 08:11:35 +00:00
Michael Zuckerman 6170c15fc6 [Clang][Intrinsics][avx512] Continue Adding round cvt to clang
And remove trailing spaces in intrinsic f test
Differential Revision: http://reviews.llvm.org/D20810

llvm-svn: 271398
2016-06-01 14:41:41 +00:00
Michael Zuckerman e6aa66a53d [Clang][Intrinsics][avx512] Adding round intrinsics fot max/min/sqrt instruction set to clang
Differential Revision: http://reviews.llvm.org/D20812

llvm-svn: 271373
2016-06-01 08:34:03 +00:00