projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 15fda40) | patch
The inst combining of inttoptr into GEP with one index was using the bit size of
authorBill Wendling <isanbard@gmail.com>
Fri, 14 Mar 2008 05:12:19 +0000 (05:12 +0000)
committerBill Wendling <isanbard@gmail.com>
Fri, 14 Mar 2008 05:12:19 +0000 (05:12 +0000)
commitb9d4f8df709bcc56e795f84739d6e7c395974ee1
tree9787abe069418c689da5bb7d6028124fc0c6d8f5tree | snapshot
parent15fda403f36d353b9d4bcfb03377bc74f40dafe5commit | diff
The inst combining of inttoptr into GEP with one index was using the bit size of
the type instead of the byte size. This was causing troublesome mis-compilations.

True to form, this took 2 days to find and is a one-line fix. :-P

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48354 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Transforms/Scalar/InstructionCombining.cpp diff | blob | history
test/Transforms/InstCombine/2008-03-13-IntToPtr.ll [new file with mode: 0644] blob
C/C++ LLVM-based compilers that forbids OOTA behaviors