This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Fix gcc.target/mips/extend-1.c
Adam Nemet <anemet@caviumnetworks.com> writes:
> Richard Sandiford writes:
> And when omitting octeon in addition:
>
> isa=!octeon -mbranch-likely
>
> should select at least a -mips2. This however does not work just like:
>
> isa<=4 -mbranch-likely
>
> would not work on mips1 since having specified the ISA "overrides" the effect
> of the other flags in terms of the selected architecture.
Right.
> Subtle, maybe it would be good to have a comment about this.
Well, the case above is one in which you'd need to specify both
upper and lower bounds for the ISA level. That's actually something
I'd wondered about when writing the current form of mips.exp,
but there was no immediate need for it, and things seemed complicated
enough already.
> Otherwise I have no problem with your version of the patch.
Thanks, applied.
Richard