> On Nov. 21, 2016, 7:07 p.m., Michael LeBeane wrote:
> > src/sim/syscall_emul.hh, line 780
> > <http://reviews.gem5.org/r/3718/diff/1/?file=63648#file63648line780>
> >
> >     Why 1?  Comment would be helpful.

This webpage, https://cygwin.com/ml/glibc-linux/2000-q1/msg00102.html, gives an 
answer about what "mem_unit" is. I didn't add a comment because none of the 
other fields has a comment and this is one of the more useless fields for us to 
know in gem5 (probably).

Turns out that ROCm was calling this in a function call chain and was receiving 
garbage because the value isn't set to anything. "1" is what my desktop was 
giving me when I printed out the structure so I passed it in and the problem 
went away.


- Brandon


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/3718/#review9135
-----------------------------------------------------------


On Nov. 17, 2016, 8:27 p.m., Brandon Potter wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://reviews.gem5.org/r/3718/
> -----------------------------------------------------------
> 
> (Updated Nov. 17, 2016, 8:27 p.m.)
> 
> 
> Review request for Default.
> 
> 
> Repository: gem5
> 
> 
> Description
> -------
> 
> Changeset 11711:96f28447a815
> ---------------------------
> syscall_emul: extend sysinfo system call to include mem_unit
> 
> 
> Diffs
> -----
> 
>   src/sim/syscall_emul.hh c38fcdaa5fe508dbb18cc084e758ad0ce8e2e2f4 
> 
> Diff: http://reviews.gem5.org/r/3718/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Brandon Potter
> 
>

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

Reply via email to