Nigel
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
241,907 Commits 29 Branches 0 Tags 2.2 GiB
Commit Graph

19 Commits

Author SHA1 Message Date
Eric Christopher abb2b54ad3 After PR28761 use -Wall with -Werror in builtins tests to identify 
possible problems in headers.

llvm-svn: 277696
 2016-08-04 06:02:50 +00:00
Craig Topper 95b61b0544 [X86] Use __builtin_ia32_vec_ext_v4hi and __builtin_ia32_vec_set_v4hi to implement pextrw/pinsertw MMX intrinsics instead of trying to use native IR. 
Without this we end up generating code that doesn't use mmx registers and probably doesn't work well with other mmx intrinsics.

llvm-svn: 274968
 2016-07-09 05:30:41 +00:00
Simon Pilgrim 2d1decf7cb [X86][SSE] Tidied up MMX/SSE/SSE2 builtin tests to the correct test file 
llvm-svn: 269852
 2016-05-17 22:03:31 +00:00
Michael Kuperstein 591278c08d [X86] Add missing m64/int64 conversions 
Define the 64-bit equivalents of _m_to_int and _m_from_int.

Differential Revision: http://reviews.llvm.org/D15572

llvm-svn: 256122
 2015-12-20 12:37:18 +00:00
Simon Pilgrim 068c2ce836 [X86] Stripped backend codegen tests 
As discussed on the ml, backend tests need to be put in llvm/test/CodeGen/X86 as fast-isel tests using IR that is as close to what is generated here as possible.

The llvm tests will (re)added in a future commit

I will update PR24580 on this new plan

llvm-svn: 254594
 2015-12-03 08:45:21 +00:00
Simon Pilgrim 7b8f926465 [X86][MMX] Added MMX IR + assembly codegen builtin tests for some missing cvt intrinsics 
llvm-svn: 253169
 2015-11-15 14:40:31 +00:00
Simon Pilgrim 54cf98eeb2 [X86][MMX] Sorted MMX IR + assembly codegen builtin tests 
Makes it easier to track what tests are missing....

llvm-svn: 253131
 2015-11-14 13:25:06 +00:00
Simon Pilgrim 087b6a7216 [X86][MMX] Added MMX IR + assembly codegen builtin tests 
Improved tests as discussed in PR24580

llvm-svn: 253130
 2015-11-14 12:47:44 +00:00
Eric Christopher cd875efa78 Canonicalize some of the x86 builtin tests and either remove or comment 
about optimization options.

llvm-svn: 250271
 2015-10-14 05:40:21 +00:00
Chandler Carruth cbe6411401 Fix the SSE4 byte sign extension in a cleaner way, and more thoroughly 
test that our intrinsics behave the same under -fsigned-char and
-funsigned-char.

This further testing uncovered that AVX-2 has a broken cmpgt for 8-bit
elements, and has for a long time. This is fixed in the same way as
SSE4 handles the case.

The other ISA extensions currently work correctly because they use
specific instruction intrinsics. As soon as they are rewritten in terms
of generic IR, they will need to add these special casts. I've added the
necessary testing to catch this however, so we shouldn't have to chase
it down again.

I considered changing the core typedef to be signed, but that seems like
a bad idea. Notably, it would be an ABI break if anyone is reaching into
the innards of the intrinsic headers and passing __v16qi on an API
boundary. I can't be completely confident that this wouldn't happen due
to a macro expanding in a lambda, etc., so it seems much better to leave
it alone. It also matches GCC's behavior exactly.

A fun side note is that for both GCC and Clang, -funsigned-char really
does change the semantics of __v16qi. To observe this, consider:

  % cat x.cc
  #include <smmintrin.h>
  #include <iostream>

  int main() {
    __v16qi a = { 1, -1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0};
    __v16qi b = _mm_set1_epi8(-1);
    std::cout << (int)(a / b)[0] << ", " << (int)(a / b)[1] << '\n';
  }
  % clang++ -o x x.cc && ./x
  -1, 1
  % clang++ -funsigned-char -o x x.cc && ./x
  0, 1

However, while this may be surprising, both Clang and GCC agree.

Differential Revision: http://reviews.llvm.org/D13324

llvm-svn: 249097
 2015-10-01 23:40:12 +00:00
Simon Pilgrim fbb8904411 [X86] Remove unnecessary MMX declarations from Intrin.h 
As discussed in PR23648 - the intrinsics _m_from_int, _m_to_int and _m_prefetch are defined in mmintrin.h and prfchwintrin.h so we don't need to in Intrin.h

Added tests for _m_from_int and _m_to_int

D11338 already added a test for _m_prefetch

Differential Revision: http://reviews.llvm.org/D12272

llvm-svn: 245975
 2015-08-25 21:27:46 +00:00
NAKAMURA Takumi 0acd8a7561 clang/test: REQUIRES: s/x86-64-registered-target/x86-registered-target/ 
llvm-svn: 196350
 2013-12-04 03:41:33 +00:00
Galina Kistanova f56b4f6fdd These tests require particular registered targets. Declared as such. 
llvm-svn: 132600
 2011-06-03 22:24:54 +00:00
Bill Wendling 67c5621742 Linux still needs these hacks. 
llvm-svn: 116888
 2010-10-20 00:12:59 +00:00
Bill Wendling 2bb7c59356 Now that mm_malloc.h was rewritten to *not* include errno.h (see 
http://llvm.org/viewvc/llvm-project?rev=116771&view=rev) we can get rid of these
hacks.

llvm-svn: 116853
 2010-10-19 20:08:12 +00:00
Bill Wendling 964f521e85 Erm...this isn't the same as PR6658. 
llvm-svn: 115737
 2010-10-06 03:33:37 +00:00
Bill Wendling 8700d39139 Add same hack to mmx-builtins.c as in Headers/c89.c to get it to pass on Win32. PR6658. 
llvm-svn: 115735
 2010-10-06 03:31:09 +00:00
Bill Wendling a76cdeb783 Do x86_64. 
llvm-svn: 115460
 2010-10-03 08:34:05 +00:00
Bill Wendling 4a58223401 Test to verify that all of the MMX builtins are handled by the front-end. 
llvm-svn: 115459
 2010-10-03 03:39:54 +00:00