On Tue, Jul 3, 2012 at 12:48 PM, Uros Bizjak <ubiz...@gmail.com> wrote: > On Tue, Jul 3, 2012 at 12:37 PM, Richard Guenther > <richard.guent...@gmail.com> wrote: > >>>>> Based on the observation above, the patch is OK for mainline, but >>>>> please also handle "rep nop" case. >>>> >>>> Here's the new version of the patch that does that. Note that someone >>>> needs to commit this for me, since I am not empowered to do it myself. >>>> >>>> gcc/ >>>> 2012-07-02 Roland McGrath <mcgra...@google.com> >>>> >>>> * configure.ac (HAVE_AS_IX86_REP_LOCK_PREFIX): Also require that >>>> the >>>> assembler accept 'rep bsf ...', 'rep bsr ...', 'rep ret', and 'rep >>>> nop'. >>>> * configure: Regenerated. >>>> * config/i386/i386.md (simple_return_internal_long): Use %; >>>> (ctz<mode>2): Likewise. >>>> (*pause): Likewise. >>> >>> The patch is OK. >>> >>> I have committed it to the mainline SVN. >> >> Given that the latest binutils release (2.22) does not support 'rep >> ret', does that >> not pessimize people with just "partial" 'rep *' support? > > No, gcc just emits "rep; ret". Using older binutils, this is just > cosmetic change in the asm dumps. IIUC, Roland worries that *future* > binutils could insert something after the "rep" in "rep;INSN" combo.
Ah, thanks for the clarification. I've run into this because if you just re-build in an already configured tree the cache variable is used and the test is not re-done and you get unsupported rep ret ... Richard. > Uros.
