The previous change caused o3 boot on Alpha/Linux to fail since the max deptag wasn't Ctrl_DepTag + NumMiscRegs but rather needed to also know about the internal processor registers. I think it's good to keep this panic if we are indexing into bogus space, so I've added a Max_DepTag field to every architecture that is set appropriately.
Ali On Tue, 24 Aug 2010 22:27:07 -0000, "Ali Saidi" <[email protected]> wrote: > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.m5sim.org/r/227/ > ----------------------------------------------------------- > > (Updated 2010-08-24 15:27:07.851020) > > > Review request for Default. > > > Summary > ------- > > ARM: fixed a bug in the arm register flattening logic (FP_Base_DepTag was > set too low) > when decoding a srs instruction, invalid mode encoding returns invalid > instruction. > this can happen when garbage instructions are fetched from mispredicted > path > > > Diffs (updated) > ----- > > src/arch/alpha/registers.hh 47d9409b2b7f > src/arch/arm/registers.hh 47d9409b2b7f > src/arch/mips/registers.hh 47d9409b2b7f > src/arch/power/registers.hh 47d9409b2b7f > src/arch/sparc/registers.hh 47d9409b2b7f > src/arch/x86/registers.hh 47d9409b2b7f > src/cpu/o3/rename_impl.hh 47d9409b2b7f > > Diff: http://reviews.m5sim.org/r/227/diff > > > Testing > ------- > > > Thanks, > > Ali > > _______________________________________________ > m5-dev mailing list > [email protected] > http://m5sim.org/mailman/listinfo/m5-dev _______________________________________________ m5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/m5-dev
