[Bug target/65710] [4.9/5 Regression] Thumb1 ICE caused by no register to spill

yroux at gcc dot gnu.org gcc-bugzilla@gcc.gnu.org
Fri Apr 10 13:16:00 GMT 2015


--- Comment #11 from Yvan Roux <yroux at gcc dot gnu.org> ---
Re-introducing the condition that stop updating lra_constraint_new_regno_start
after switching off inheritance and remat, seems to fix the problem:
gcc.target/arm/pr65647-2.c is ok and the reduced test case from Terry builds as
well, at least with my quick test on 4.9 branch.  I'll do it on trunk and pass
a full validation.  

Is it the correct fix for you Vlad ?

More information about the Gcc-bugs mailing list