Please ignore the link.
Sonic >-----Original Message----- >From: Zhang, Sonic >Sent: Thursday, July 29, 2010 3:36 PM >To: 'Mike Frysinger'; [email protected] >Cc: [email protected] >Subject: RE: [Linux-kernel-commits] [9020] >trunk/arch/blackfin/mach-bf561/atomic.S: Bug[#6044] Forgot >CSYNC in get_core_lock_noflush > >http://docs.blackfin.uclinux.org/doku.php?id=power_management_support > > >Sonic > >>-----Original Message----- >>From: Mike Frysinger [mailto:[email protected]] >>Sent: Wednesday, July 28, 2010 11:18 PM >>To: [email protected] >>Cc: [email protected] >>Subject: Re: [Linux-kernel-commits] [9020] >>trunk/arch/blackfin/mach-bf561/atomic.S: Bug[#6044] Forgot CSYNC in >>get_core_lock_noflush >> >>On Wed, Jul 28, 2010 at 06:53, <[email protected]> wrote: >>> Modified: trunk/arch/blackfin/mach-bf561/atomic.S (9019 => 9020) >>> >>> @@ -71,6 +71,8 @@ >>> SSYNC(r2); >>> jump .Lretry_corelock_noflush >>> .Ldone_corelock_noflush: >>> + nop; >>> + CSYNC(r2); >> >>why the NOP ? the CSYNC() macro already takes care of padding things >>correctly. >>-mike >> > >I also think NOP is no necessary. But, the SMP kgdb single >step fails without this nop. > _______________________________________________ Linux-kernel-commits mailing list [email protected] https://blackfin.uclinux.org/mailman/listinfo/linux-kernel-commits
