----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.m5sim.org/r/897/#review1626 -----------------------------------------------------------
src/arch/x86/isa/insts/general_purpose/input_output/general_io.py <http://reviews.m5sim.org/r/897/#comment2106> You're changing the behavior of these microops, and that's not ok because these are used to talk to devices. These are nonspeculative on purpose, and I'm pretty sure making them speculative breaks the O3 FS regression. Have you run the regressions? - Gabe On 2011-10-30 18:22:08, Nilay Vaish wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.m5sim.org/r/897/ > ----------------------------------------------------------- > > (Updated 2011-10-30 18:22:08) > > > Review request for Default. > > > Summary > ------- > > X86 ISA: Change definitions of locked instructions > This patch is for changing the defintion of locked instructions. These > should behave as memory barriers and should be executed in a non speculative > fashion. > > > Diffs > ----- > > src/arch/x86/isa/insts/general_purpose/input_output/general_io.py > dd77c8d0a93e > src/arch/x86/isa/insts/general_purpose/input_output/string_io.py > dd77c8d0a93e > src/arch/x86/isa/microops/ldstop.isa dd77c8d0a93e > > Diff: http://reviews.m5sim.org/r/897/diff > > > Testing > ------- > > > Thanks, > > Nilay > > _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
