Bug 14940 - R.bin segmentation fault when byte-compiling package 'base' during compilation with gcc 4.7
R.bin segmentation fault when byte-compiling package 'base' during compilatio...
Status: CLOSED INVALID
Product: R
Classification: Unclassified
Component: Installation
R 2.15.0
ix86 (32-bit) Linux
: P5 enhancement
Assigned To: R-core
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2012-06-07 11:01 UTC by Jauhien Piatlicki
Modified: 2012-06-07 14:10 UTC (History)
0 users

See Also:


Attachments
build.log (138.30 KB, text/x-log)
2012-06-07 11:05 UTC, Jauhien Piatlicki
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Jauhien Piatlicki 2012-06-07 11:01:23 UTC
R.bin segmentation fault when byte-compiling package 'base' during compilation with gcc 4.7 with flag -O3 on Gentoo Linux

1. Download sources of R-2.15.0
2. Compile it with gcc 4.7 with -O3 flag

R.bin segfault when byte-compiling package 'base'.

Build 2012-06-07 on Gentoo Linux
gcc version: 4.7.0
CFLAGS="-O3 -march=native -mtune=native -fomit-frame-pointer -pipe"
computer: i686 Intel(R) Atom(TM) CPU N280 @ 1.66GHz GenuineIntel GNU/Linux

Original bug on Gentoo's Bugzilla: https://bugs.gentoo.org/show_bug.cgi?id=419761
Comment 1 Jauhien Piatlicki 2012-06-07 11:05:02 UTC
Created attachment 1312 [details]
build.log
Comment 2 Jauhien Piatlicki 2012-06-07 13:37:05 UTC
I'm sorry for possible abusing of R's bugzilla, but what is wrong with this bug, so it is invalid?
I've read R's bug howto and reported it according to rules stated there. I just want to know what's wrong to avoid posting invalid bug reports in the future.
Or R just doesn't support -O3 optimization flag?
Thanks for answer.
Comment 3 Peter Dalgaard 2012-06-07 13:59:04 UTC
The bug isn't in R, so nothing we can fix. 

There's nothing particularly wrong with reporting it and having an answer in the bug database, I'd say. The "invalid" category just fits better than "wontfix" or "spam". Perhaps we should consider a "not-our-bug" category.
Comment 4 Jauhien Piatlicki 2012-06-07 14:10:12 UTC
Ok, thanks.
Comment 5 Brian Ripley 2012-06-07 17:53:05 UTC
On 07/06/2012 13:37, r-bugs@r-project.org wrote:
> https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=14940
>
> --- Comment #2 from Jauhien Piatlicki<piatlicki@gmail.com>  2012-06-07 08:37:05 EDT ---
> I'm sorry for possible abusing of R's bugzilla, but what is wrong with this
> bug, so it is invalid?
> I've read R's bug howto and reported it according to rules stated there. I just
> want to know what's wrong to avoid posting invalid bug reports in the future.
> Or R just doesn't support -O3 optimization flag?
> Thanks for answer.


There is no evidence that the bug is in R.  And we happen to be aware 
that for x86_64 Linux, the bug is in gcc.

Maybe 10 times in the gcc4 series, gcc -O3 bugs have caused R builds to 
fail, so we are not likely to believe that this time the bug is in R 
without very detailed evidence (and we will still say use the 
recommended -O2).

-- 
Brian D. Ripley,                  ripley@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595