LFS 7.3-rc1

druuna · 02-21-2013, 02:17 PM

Quote:

Originally Posted by stoat

I emailed the lfs-dev mailing list about the gcc test stuff. Armin Krejzi replied saying that he noticed the same things with gcc and thinks it should be okay to continue with those test results (which I did).

Bruce Dubbs also replied saying that he thought the g++ out-of-memory thing might be stack memory instead of total memory. He's wondering if the ulimit command should be changed to "ulimit -s unlimited". druuna tried some larger settings, but not unlimited. I may try that just to find out. He also commented that the libstd++ thing that I posted above seems to be a new normal for this version of gcc.

I just redid gcc with ulimit -s unlimited and the error still shows up.

I ran free with a 10 second delta in another console and noticed no noticeable dip in available memory (which does happen at an earlier stage, but it "recovers").

BTW: Thanks for the info!

@ Keith Hedger: I did not come across this error during its build/check/installation.

stoat · 02-22-2013, 08:52 AM

That link in section 6.17 (GCC-4.7.2) of the 7.3-rc1 book to build logs now works and points to some logs for a core2duo system. The GCC tests have the g++ unresolved testcases with "out-of-memory" and the libstdc++ unexpected failure as already described above.

Keith Hedger · 02-25-2013, 05:55 AM

Just finished installing the basic system and rebooted and everything works fine, now to install X and Xfce ( always fun ! )
Good work and congrats to the LFS team.

stoat · 02-25-2013, 07:15 AM

Of course do what you need to do, but it might be worth waiting a few more days for the final version. In spite of the release candidate building well, it is not in a "package freeze" state and significant changes could happen in the final version.

Keith Hedger · 02-25-2013, 07:24 AM

Actually I script any new install so that I can rebuild easily in case something goes wrong or I need to tweak and the Xorg/Xfce instructions on BLFS are pretty stable anyway and I'm just tweaking my previous scripts for them.

druuna · 02-25-2013, 07:29 AM

I agree with stoat, especially if you want to use it for daily use. I would wait until LFS 7.3 stable is released.

One of the changes that's being talked about: kernel 3.8 instead of 3.7. Another: gcc 4.8 instead of gcc 4.7

Off-topic: Am I the only one that cannot reach (Page not found!) the LFS mailing list archives (dev, support and announce)?

Edit: Keith Hedger replied while I was typing: If you can do it in an easy way I would go for it, maybe you'll find 7.3 related problems that haven't shown yet when building (parts of) BLFS.

Keith Hedger · 02-25-2013, 07:41 AM

Talking of making it easy I just fubar'ed the new system so I'm running the script to rebuild it again.
Of course I won't run it as my main system for a while lots to do first!

stoat · 02-25-2013, 07:43 AM

Quote:

Originally Posted by druuna

Am I the only one that cannot reach (Page not found!) the LFS mailing list archives (dev, support and announce)?

You're not the only one. I asked about that in one of the mailing lists. The reply was...

Quote:

Originally Posted by B.Dubbs

That is something that needs to be re-enabled due to a server transition.

But it apparently is a low priority matter.

pethead · 03-15-2013, 04:42 AM

Applying patch to Perl occurs below

Code:

patch -Npl -i $LFS/sources/perl-5.16.2-libc-1.patch 
patch: **** strip count l is not a number
Segmentation fault (core dumped)

Any patch calls before are ended with result of 'patch: **** strip count l is not a number' only.
It is good?

*Chapter 5.28. Perl-5.16.2
Linux From Scratch - Version 7.3
Chapter 5. Constructing a Temporary System

stoat · 03-15-2013, 08:33 AM

The code in your post has a lower-case letter "L" for the -p option in the patch command (as the error message is telling you). It should be a numeral one. In some fonts it is hard to tell the difference.

P.S.: This right here is why I copy and paste nearly every command from the beginning of the LFS book to the end of building a BLFS system. I often type make or make install, but everything else I copy and paste. I don't view that as a sign of weakness or unworthiness. It's just accurate, IMO.

pethead · 03-15-2013, 08:43 AM

OMG!

That's really!
Thanks.
*But patch has been applied or such not?

stoat · 03-15-2013, 09:07 AM

Quote:

Originally Posted by pethead

But patch has been applied or such not?

I would say the patch was not applied. The patch command also has a message to tell you that it is applying the patch (patching so-and-so-file...). And another message for when you attempt to run the same patch again. And so on. Patch will tell you what it's doing. But for those times when you weren't watching and aren't sure what happened, you can always exit the build directory back to the source directory (cd ..), delete the build directory, untar the tarball again, and start over. I have often done that when something got botched with patch commands or configure options.

pethead · 03-19-2013, 04:19 AM

http://www.linuxfromscratch.org/lfs/...ter06/vim.html

6.62. Vim-7.3

result of
make test

Code:

Test results:
test37 FAILED
test47 FAILED
TEST FAILURE
make[2]: *** [report] Error 1
make[2]: Leaving directory `/root/vim73/src/testdir'
make[1]: *** [test] Error 2
make[1]: Leaving directory `/root/vim73/src'
make: *** [test] Error 2

may I do 'make install' and continue?

stoat · 03-19-2013, 06:09 AM

Compare the test log to the one posted by the developers. Find their logs here..

http://www.linuxfromscratch.org/lfs/build-logs/

To me, what you posted of yours doesn't compare well. I don't usually test vim (or very many others not specified as critical). But if I did see a test like that one, I think I would not proceed until I found out more about it. You can go directly to the developers with that one by posting it on the lfs-support mailing list. Find out more about subscribing and submitting to it here...

http://www.linuxfromscratch.org/mail.html

If those guys don't know, then probably nobody does.

Lennie · 03-20-2013, 02:08 PM

I got this in the test of gcc (6.17):

Code:

		=== libmudflap tests ===


Running target unix
FAIL: libmudflap.c++/pass55-frag.cxx ( -O) execution test
ERROR: tcl error sourcing /sources/lfs/gcc/src/gcc-4.7.2/libmudflap/testsuite/libmudflap.cth/cthfrags.exp.
ERROR: couldn't compile regular expression pattern: out of memory

		=== libmudflap Summary ===

# of expected passes		1255
# of unexpected failures	1

		=== libstdc++ tests ===


Running target unix
FAIL: libstdc++-abi/abi_check

		=== libstdc++ Summary ===

# of expected passes		8241
# of unexpected failures	1
# of expected failures		43
# of unsupported tests		477

I haven't seen the same failure of libstdc++ somewhere else. Is it safe to continue?