This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Other format: | [Raw text] |
Hi, The problem here is that we sometimes could get a GIMPLE_MODIFY_STMT so the use of TREE_OPERAND is wrong and could cause an ICE. This comes from C++ code where there is a TARGET_EXPR and the statement has been already changed to a GIMPLE_MODIFY_STMT so we would ICE if the vector constructor was a constant. Committed to the 4.3 branch as obvious after a boostrap/test on i386-darwin8.11 with no regressions. Note this does not effect the trunk as the gimplifier has changed and GIMPLE_MODIFY_STMT has since been removed. Thanks, Andrew Pinski ChangeLog: * gimplifier.c (gimplify_init_constructor): For constant vector CONSTRUCTORs use GENERIC_TREE_OPERAND instead of TREE_OPERAND.
Attachment:
fixpr38565.diff.txt
Description: Text document
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |