<meta http-equiv="Content-Type" content="text/html; charset=utf-8"><div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Apr 6, 2016 at 9:32 AM, Halim Amer <span dir="ltr"><<a href="mailto:aamer@anl.gov" target="_blank">aamer@anl.gov</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">This information is insufficient to blame the compiler.<br></blockquote><div><br></div><div><span style="color:rgb(0,0,0);font-family:Tahoma;font-size:13.3333px">This statement sure seems sufficient to me:</span></div><div><span style="color:rgb(0,0,0);font-family:Tahoma;font-size:13.3333px"><br></span></div><div><span style="color:rgb(0,0,0);font-family:Tahoma;font-size:13.3333px">"mpich gets miscompiled when building with CFLAGS='-O2 -ftree-vectorize -g' (but not with just -O2 -g or -O3 -g)."</span></div><div><span style="color:rgb(0,0,0);font-family:Tahoma;font-size:13.3333px"><br></span></div><div><span style="color:rgb(0,0,0);font-family:Tahoma;font-size:13.3333px">If changing compiler flags alone causes MPICH to go from working to nonworking, it's hard not to blame the compiler.</span></div><div><span style="color:rgb(0,0,0);font-family:Tahoma;font-size:13.3333px"><br>Of course, if the app flags changed, then all bets are off.</span></div><div><span style="color:rgb(0,0,0);font-family:Tahoma;font-size:13.3333px"><br></span></div><div><span style="color:rgb(0,0,0);font-family:Tahoma;font-size:13.3333px">Anyways, I concur that using the latest development tree of a compiler isn't a good idea. I'm sure the GCC folks would readily admit that there are bugs in the trunk. It never hurts to file this bug, but if there is no MCVE, it will be hard for anyone to make progress.</span></div><div><span style="color:rgb(0,0,0);font-family:Tahoma;font-size:13.3333px"><br></span></div><div><span style="color:rgb(0,0,0);font-family:Tahoma;font-size:13.3333px">Jeff</span></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
We don't know if the user (application code), MPICH, or gcc is doing something wrong.<br>
<br>
First, you need to use a stable compiler version (not trunk) and a recent MPICH (3.2 not the old 3.1 version).<br>
<br>
Second, you need to make sure the user code is not doing something wrong (e.g. MPI_Alltoall could have gotten an invalid buffer address or count). If you send us a toy program that reproduces the problem, we could help with this.<br>
<br>
--Halim<br>
<a href="http://www.mcs.anl.gov/~aamer" rel="noreferrer" target="_blank">www.mcs.anl.gov/~aamer</a><div><div class="h5"><br>
<br>
On 4/6/16 10:24 AM, VandeVondele Joost wrote:<br>
</div></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div><div class="h5">
Hi,<br>
<br>
when compiling mpich-3.1.2 using gcc trunk (which will be gcc 6.0 in a<br>
couple of days/weeks), I noticed mpich gets miscompiled when building<br>
with CFLAGS='-O2 -ftree-vectorize -g' (but not with just -O2 -g or -O3 -g).<br>
<br>
I believe that the miscompilation happens with the macro<br>
MPIDI_Request_create_sreq since a typical trace looks like:<br>
<br>
Program received signal SIGSEGV: Segmentation fault - invalid memory<br>
reference.<br>
<br>
Backtrace for this error:<br>
#0 0x33ac83269f in ???<br>
#1 0x7f3743953d69 in MPIDI_Isend_self<br>
at src/mpid/ch3/src/mpidi_isend_self.c:34<br>
#2 0x7f374394dfa3 in MPID_Isend<br>
at src/mpid/ch3/src/mpid_isend.c:55<br>
#3 0x7f374390793a in MPIC_Isend<br>
at src/mpi/coll/helper_fns.c:646<br>
#4 0x7f37438628c5 in MPIR_Alltoall_intra<br>
at src/mpi/coll/alltoall.c:520<br>
#5 0x7f374386376e in MPIR_Alltoall<br>
at src/mpi/coll/alltoall.c:726<br>
#6 0x7f3743863827 in MPIR_Alltoall_impl<br>
at src/mpi/coll/alltoall.c:761<br>
#7 0x7f3743863e7a in PMPI_Alltoall<br>
at src/mpi/coll/alltoall.c:884<br>
#8 0x7f37424181bf in pmpi_alltoall_<br>
at src/binding/fortran/mpif_h/alltoallf.c:272<br>
<br>
I don't know mpich well enough to extract a testcase, but it would be<br>
great if somebody could extract that and report to the gcc team (I'm<br>
happy to do the reporting if needed).<br>
<br>
thanks in advance,<br>
<br>
Joost<br>
<br>
<br>
<br></div></div>
_______________________________________________<br>
discuss mailing list <a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a><br>
To manage subscription options or unsubscribe:<br>
<a href="https://lists.mpich.org/mailman/listinfo/discuss" rel="noreferrer" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
<br>
</blockquote>
_______________________________________________<br>
discuss mailing list <a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a><br>
To manage subscription options or unsubscribe:<br>
<a href="https://lists.mpich.org/mailman/listinfo/discuss" rel="noreferrer" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature">Jeff Hammond<br><a href="mailto:jeff.science@gmail.com" target="_blank">jeff.science@gmail.com</a><br><a href="http://jeffhammond.github.io/" target="_blank">http://jeffhammond.github.io/</a></div>
</div></div>