> On Oct. 16, 2015, 7:19 p.m., Nilay Vaish wrote:
> > src/arch/x86/isa/insts/simd128/floating_point/data_transfer/move.py, line 84
> > <http://reviews.gem5.org/r/3065/diff/1/?file=49142#file49142line84>
> >
> >     Is there a reason to use disp instead of DISPLACEMENT?  As far as I can 
> > tell, 
> >     it should not matter, which makes me wonder why we have three different 
> > names: disp, DISPLACMENT and adjustedDisp.
> >     I think all of them translate to the displacement field in the 
> > instruction.

As far as I know you are right, they are all the same.  I just copied the 
movaps/movapd code which uses DISPLACEMENT in the first uop and disp in the 
second.  I don't know if there's a reason for that or if it's just arbitrary, 
but it doesn't make sense to me to make this version different.


- Steve


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/3065/#review7384
-----------------------------------------------------------


On Oct. 6, 2015, 8 p.m., Steve Reinhardt wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://reviews.gem5.org/r/3065/
> -----------------------------------------------------------
> 
> (Updated Oct. 6, 2015, 8 p.m.)
> 
> 
> Review request for Default.
> 
> 
> Repository: gem5
> 
> 
> Description
> -------
> 
> Changeset 11059:fc3d1684cbd1
> ---------------------------
> x86: implement movntps/movntpd SSE insts
> 
> These are non-temporal packed SSE stores.  For now, we are not
> supporting the cache hints, but are implementing them the same
> as the non-hinting movaps/movapd instructions.
> 
> Also got rid of the placeholder file that was intended for these insts
> (simd128/floating_point/data_transfer/move_non_temporal.py) since it
> seemed more logical to put them in with the other moves.
> 
> 
> Diffs
> -----
> 
>   src/arch/x86/SConscript 842f56345a421244a7a8988a5bc4fb1cfbf409ef 
>   src/arch/x86/isa/decoder/two_byte_opcodes.isa 
> 842f56345a421244a7a8988a5bc4fb1cfbf409ef 
>   src/arch/x86/isa/insts/simd128/floating_point/data_transfer/__init__.py 
> 842f56345a421244a7a8988a5bc4fb1cfbf409ef 
>   src/arch/x86/isa/insts/simd128/floating_point/data_transfer/move.py 
> 842f56345a421244a7a8988a5bc4fb1cfbf409ef 
>   
> src/arch/x86/isa/insts/simd128/floating_point/data_transfer/move_non_temporal.py
>  842f56345a421244a7a8988a5bc4fb1cfbf409ef 
> 
> Diff: http://reviews.gem5.org/r/3065/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Steve Reinhardt
> 
>

_______________________________________________
gem5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/gem5-dev

Reply via email to