https://gcc.gnu.org/bugzilla/show_bug.cgi?id=119155
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Attachment #60676|0 |1
is obsolete| |
--- Comment #6 from Richard Biener <rguenth at gcc dot gnu.org> ---
Created attachment 60707
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=60707&action=edit
actual patch
This patch passed testing but on arm
(https://patchwork.sourceware.org/project/gcc/patch/20250307133608.a307c13...@imap1.dmz-prg2.suse.org/)
but I cannot reproduce the reported failure with my cross setup and qemu, not
by
shuffling options like -march=armv8-a -mfpu=neon -mthumb -mfloat-abi=hard
so I'm lost.
I'm not sure whether we trap on an unaligned access (the log from the tester
says it recorded a SIGSEGV), and whether I need to configure qemu in any
special way to reproduce this.
