https://sourceware.org/bugzilla/show_bug.cgi?id=27167
--- Comment #5 from Matt Jenkins ---
Sure thing. I've retained the -mips5 flag which GCC doesn't seem to use, just
in case there's some other source that flag could come from. It's doing no harm
being there.
--
You are receiving this mail be
https://sourceware.org/bugzilla/show_bug.cgi?id=27167
--- Comment #4 from Matt Jenkins ---
Created attachment 13113
--> https://sourceware.org/bugzilla/attachment.cgi?id=13113&action=edit
Patch for mips flags
--
You are receiving this mail because:
You are on the CC list for the bug.
https://sourceware.org/bugzilla/show_bug.cgi?id=27167
--- Comment #2 from Matt Jenkins ---
Actually, there's a couple more missing. From GCC's `--target-help`:
Known MIPS ISA levels (for use with the -mips option):
1 2 3 32 32r2 32r3 32r5 32r6 4 64 64r2 64r3 64r5 64r6
-mips64r3 and -mips
https://sourceware.org/bugzilla/show_bug.cgi?id=27167
--- Comment #1 from Matt Jenkins ---
On further investigation there is no MIPS32R4, so the -mips32r4 flag could be
removed from that list.
--
You are receiving this mail because:
You are on the CC list for the bug.
: ld
Assignee: unassigned at sourceware dot org
Reporter: matt at majenko dot co.uk
Target Milestone: ---
LD contains code to "ignore" -mips32r2 and -mips32r6 but nothing in between.
This means that GCC can't compile for -arch=m5100 and others that pass
-mips32r
