> On Feb. 25, 2017, 1:04 a.m., Brad Beckmann wrote: > > src/mem/ruby/system/GPUCoalescer.py, line 50 > > <http://reviews.gem5.org/r/3824/diff/2/?file=66267#file66267line50> > > > > This patch points out that we really should not have an icache or > > IFETCHes as part of the GPUCoalescer. It is designed to only handle data > > accesses from the GPU CUs. > > > > You don't necessarily need to remove this line and make all the other > > necessary changes, but if you don't, we'll get someone from AMD to make > > that change. > > Andreas Sandberg wrote: > I had a quick look at GPUCoalescer.cc and it seems like the icache > pointer is used in the hit callback. I could remove it and add an assert. > However, since I don't know what's going on in this part of the code base, > you're probably better off doing it yourselves.
Sure thing. We'll take care of it. Thanks again for posting this patch and doing the work you have. > On Feb. 25, 2017, 1:04 a.m., Brad Beckmann wrote: > > src/mem/ruby/system/GPUCoalescer.py, line 52 > > <http://reviews.gem5.org/r/3824/diff/2/?file=66267#file66267line52> > > > > Aren't you missing the following line: > > > > max_outstanding_requests = Param.Int(16, > > "max requests (incl. prefetches) outstanding") > > Andreas Sandberg wrote: > That line should already be included from the original version of the > file. Oops...sorry about that. I didn't realize we were already redefining that variable. That actually seems like another problem this patch is fixing. Thanks! - Brad ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/3824/#review9478 ----------------------------------------------------------- On Feb. 22, 2017, 9:28 a.m., Andreas Sandberg wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem5.org/r/3824/ > ----------------------------------------------------------- > > (Updated Feb. 22, 2017, 9:28 a.m.) > > > Review request for Default. > > > Repository: gem5 > > > Description > ------- > > Changeset 11878:b0a4f3815b55 > --------------------------- > gpu-compute: Fix Python/C++ object hierarchy discrepancies > > The GPUCoalescer and the Shader classes have different base classes in > C++ and Python. This causes subtle bugs in SWIG and compilation errors > for PyBind. > > Change-Id: I1ddd2a8ea43f083470538ddfea891347b21d14d8 > Reviewed-by: Andreas Hansson <andreas.hans...@arm.com> > > > Diffs > ----- > > src/gpu-compute/shader.hh ba90ffa751b6 > src/gpu-compute/shader.cc ba90ffa751b6 > src/mem/ruby/system/GPUCoalescer.hh ba90ffa751b6 > src/mem/ruby/system/GPUCoalescer.py ba90ffa751b6 > > Diff: http://reviews.gem5.org/r/3824/diff/ > > > Testing > ------- > > > Thanks, > > Andreas Sandberg > > _______________________________________________ gem5-dev mailing list gem5-dev@gem5.org http://m5sim.org/mailman/listinfo/gem5-dev
