[lfs-dev] Test failures in something close to current LFS

Ken Moffat zarniwhoop at ntlworld.com
Fri Jan 30 14:59:17 PST 2015


On Thu, Jan 29, 2015 at 12:23:56AM +0000, Ken Moffat wrote:
> 
> gcc: 125 unexpected failures in g++, 658 in gcc, 22 in libstdc++
> instead of the usual handful of failures.  The last time I saw those
> sorts of numbers was a little while before my AmigaOne expired.
> 
 I wonder if some of it was because my tmpfs for /tmp was too small.
I thought I was using 6GB, but I was on my 7.6 system when I set
that up, the November system I used to build the new one was still
using only 2GB for /tmp.

 I've now rebuilt gcc on the completed system, again using only 2GB
for the tmpfs.  This time, the results were a little better - only
83 unexpected failures in gcc, and 42 in g++, none in libstdc++.
In both cases, I used -j4.

 Looking at the gcc errors, in the gcc3/ directory I see 65
unexpected failures starting in c11-atomic-3.c, at a line which
relates t pointer arithmetic.  In the torture tests, in
limits-exprparen.c I see a segfault (ICE) on the first test of each
different optimization level, and they all fail - I think that is
normal for me.

 I could not find any details for the g++ tests, and they were all
over the place.

 Looking at my syslog, I see a series of segfaults -probably the
torture tests, perhaps a little more.  At the end of the tests there
was still 500MB free in /tmp.

 I've built over 300 regular packages on this system (I exclude Xorg
headers, docbook catalogues, fonts, perl modules becasue those do
not necessarily do any compilation.  During my scripted builds I had
_one_ segfault (cups), but it was ok at the second attempt.

 So, for the moment I'm just going to flag this up as "looks
serious, unexplained, but so far I do not see any problems in the
resulting system."

 And for what it is worth, 6GB is not enough to build firefox in
/tmp - I had to drop down to 4GB to get enough space for the link,
as well as building on a disk (the build in 6GB was using over 5GB
plus whatever the link wanted).

 Also, although some of the build has felt faster in /tmp -
particularly untarring and rm -rf the build directories, for the
actual build times I see almost-identical times.  That matches
Bruce's reply to my original post.

ĸen
-- 
Nanny Ogg usually went to bed early. After all, she was an old lady.
Sometimes she went to bed as early as 6 a.m.


More information about the lfs-dev mailing list