projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 64fc072) | patch
[CodeGen, X86] Classify vectors <= 32 bits as INTEGER
authorDavid Majnemer <david.majnemer@gmail.com>
Fri, 17 Jul 2015 05:49:13 +0000 (05:49 +0000)
committerDavid Majnemer <david.majnemer@gmail.com>
Fri, 17 Jul 2015 05:49:13 +0000 (05:49 +0000)
commitf8d14dbcfcd8e73903bc778f09b9796f10f45ec8
tree8d9d97c20794c027f8ede1510c097566ae0317d3tree | snapshot
parent64fc0727ffe36ee582af1e70913db49fdf67a3dccommit | diff
[CodeGen, X86] Classify vectors <= 32 bits as INTEGER

We shouldn't crash despite the AMD64 ABI not giving clear guidance as to
how to pass around vector types <= 32 bits.  Instead, classify such
vectors as INTEGER to be compatible with GCC.

This fixes PR24162.

llvm-svn: 242508
clang/lib/CodeGen/TargetInfo.cpp diff | blob | history
clang/test/CodeGenCXX/x86_64-arguments.cpp diff | blob | history
Domain: System / Toolchain;