> On 2011-02-26 08:36:58, Steve Reinhardt wrote: > > I won't vouch for the surrounding code (I don't get the difference between > > the oldRead, read, and write packets, for starters), but this one specific > > line that's changed looks good to me :-)
oldRead is the read that just returned data, read and write are the packets that should go out to continue with the walk. The writes are to update accessed and dirty bits, and the new read is to access the next entry. I also missed an instance of ReadExReq which I just added to the patch. - Gabe ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.m5sim.org/r/514/#review905 ----------------------------------------------------------- On 2011-02-26 16:21:29, Gabe Black wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.m5sim.org/r/514/ > ----------------------------------------------------------- > > (Updated 2011-02-26 16:21:29) > > > Review request for Default, Ali Saidi, Gabe Black, Steve Reinhardt, and > Nathan Binkert. > > > Summary > ------- > > X86: Use regular read requests in the walker instead of read exclusive. > > > Diffs > ----- > > src/arch/x86/pagetable_walker.cc ac1bd3d1aa54 > > Diff: http://reviews.m5sim.org/r/514/diff > > > Testing > ------- > > > Thanks, > > Gabe > >
_______________________________________________ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev