| View previous topic :: View next topic |
| Author |
Message |
Vlad.Sharp
Guru
Joined: 08 Dec 2004
Posts: 337
Location: Cambridgeshire, UK
|
 Posted: Mon May 28, 2007 8:57 am Post subject: Re: Possible fix for kde-3.5.7 with gcc-4.2 on AMD64 |
 |
|
| devsk wrote: | | does 2.17.50.0.16 binutils work for the whole system completely? Did you get any segv or other issues with that version of binutils? I am hesitant to move to unstable (upstream) version of basic things like gcc and binutils. |
Yes, it does  - I'm running ~x86 and hardened and the last couple of upstream versions of binutils have been working great. KDE works, Gnome works, Openoffice 2.2 compiles with gcc 4.2 (albeit a whole 30 minutes slower than previous, but hey?) - but seems to have a major font bug... Firefox works without a hitch, and so does thunderbird.
Other than that, I have ~1042 packages all running fine. (Except the openoffice it seems). |
|
| Back to top |
|
 |
cneil
n00b
Joined: 28 May 2007
Posts: 39
Location: Stillwater, OK
|
| Posted: Mon May 28, 2007 3:29 pm Post subject: Re: Possible fix for kde-3.5.7 with gcc-4.2 on AMD64 |
|
|
| Vlad.Sharp wrote: | | devsk wrote: | | does 2.17.50.0.16 binutils work for the whole system completely? Did you get any segv or other issues with that version of binutils? I am hesitant to move to unstable (upstream) version of basic things like gcc and binutils. |
Yes, it does - I'm running ~x86 and hardened and the last couple of upstream versions of binutils have been working great. KDE works, Gnome works, Openoffice 2.2 compiles with gcc 4.2 (albeit a whole 30 minutes slower than previous, but hey?) - but seems to have a major font bug... Firefox works without a hitch, and so does thunderbird.
Other than that, I have ~1042 packages all running fine. (Except the openoffice it seems). |
You mentioned that openoffice had a "major font bug". Did you mean an invisible font? When I compiled openoffice with gcc-4.2, I couldn't see what I was typing. The cursor would move and I could scroll down and back and see the text. It also crashed when nearly any dialog box opened. Recompiling openoffice (and hunspell and STLport) with gcc-4.1 fixed it for me.
| n0rbi666 wrote: | I have binutils-2.17.50.0.15 and everything is ok, no segfaults.
With 2.17.50.0.16 - when emerging mesa, ld segfaults - so I get back to 2.17.50.0.15 |
Hmmm, I just compiled mesa on both an amd64 box and an x86 box (older athlon-tbird). Mesa compiled fine with binutils-2.17.50.0.16 on both boxes. I've been using the prerelease version of binutils since about .12 or .13 and have not personally experienced a single failure due to binutils in that time. |
|
| Back to top |
|
|
kernelOfTruth
Watchman
Joined: 20 Dec 2005
Posts: 6111
Location: Vienna, Austria; Germany; hello world :)
|
|
| Back to top |
|
|
Vlad.Sharp
Guru
Joined: 08 Dec 2004
Posts: 337
Location: Cambridgeshire, UK
|
| Posted: Mon May 28, 2007 5:00 pm Post subject: Re: Possible fix for kde-3.5.7 with gcc-4.2 on AMD64 |
|
|
| cneil wrote: | | Vlad.Sharp wrote: | | devsk wrote: | | does 2.17.50.0.16 binutils work for the whole system completely? Did you get any segv or other issues with that version of binutils? I am hesitant to move to unstable (upstream) version of basic things like gcc and binutils. |
Yes, it does - I'm running ~x86 and hardened and the last couple of upstream versions of binutils have been working great. KDE works, Gnome works, Openoffice 2.2 compiles with gcc 4.2 (albeit a whole 30 minutes slower than previous, but hey?) - but seems to have a major font bug... Firefox works without a hitch, and so does thunderbird.
Other than that, I have ~1042 packages all running fine. (Except the openoffice it seems). |
You mentioned that openoffice had a "major font bug". Did you mean an invisible font? When I compiled openoffice with gcc-4.2, I couldn't see what I was typing. The cursor would move and I could scroll down and back and see the text. It also crashed when nearly any dialog box opened. Recompiling openoffice (and hunspell and STLport) with gcc-4.1 fixed it for me.
|
Yes, exactly that. I'll file a bug upstream when I get some time - or search for one, in any case.
| kernelOfTruth wrote: | for those of you prefering higher security (= hardening) there's also a gcc 4.2.0 version with hardened support in kevquinn's overlay available
I haven't got it to work on amd64 & hardened yet, though ... |
Hardened works great with gcc 4.2 on ~x86 (running here atm.) But you still have to patch gcc for things like the Xorg pci failure... Gcc 4.2.1 should hopefully fix that. However, kevquinn's overlay rocks! |
|
| Back to top |
|
|
devsk
Advocate
Joined: 24 Oct 2003
Posts: 2998
Location: Bay Area, CA
|
| Posted: Mon May 28, 2007 6:11 pm Post subject: |
|
|
| what does kevquinn's overlay contain for gcc that the in-portage gcc 4.2 doesn't? |
|
| Back to top |
|
|
kernelOfTruth
Watchman
Joined: 20 Dec 2005
Posts: 6111
Location: Vienna, Austria; Germany; hello world :)
|
| Posted: Mon May 28, 2007 6:23 pm Post subject: |
|
|
| devsk wrote: | | what does kevquinn's overlay contain for gcc that the in-portage gcc 4.2 doesn't? |
have a look at output of
gcc-config -l
after having compiled gcc-4.2.0 in hardened, then you'll know what (still) is missing:
- those "profiles" (or how they are called)
example:
| Quote: | gcc-config -l
[1] x86_64-pc-linux-gnu-3.4.4
[2] x86_64-pc-linux-gnu-3.4.4-hardened
[3] x86_64-pc-linux-gnu-3.4.4-hardenednopie
[4] x86_64-pc-linux-gnu-3.4.4-hardenednopiessp
[5] x86_64-pc-linux-gnu-3.4.4-hardenednossp
[6] x86_64-pc-linux-gnu-3.4.6
[7] x86_64-pc-linux-gnu-3.4.6-hardenednopie
[8] x86_64-pc-linux-gnu-3.4.6-hardenednopiessp
[9] x86_64-pc-linux-gnu-3.4.6-hardenednossp
[10] x86_64-pc-linux-gnu-3.4.6-vanilla
[11] x86_64-pc-linux-gnu-4.1.2 *
[12] x86_64-pc-linux-gnu-4.1.2-hardenednopiessp
[13] x86_64-pc-linux-gnu-4.1.2-hardenednossp
[14] x86_64-pc-linux-gnu-4.1.2-vanilla |
_________________
https://github.com/kernelOfTruth/ZFS-for-SystemRescueCD/tree/ZFS-for-SysRescCD-4.9.0
https://github.com/kernelOfTruth/pulseaudio-equalizer-ladspa
Hardcore Gentoo Linux user since 2004 |
|
| Back to top |
|
|
devsk
Advocate
Joined: 24 Oct 2003
Posts: 2998
Location: Bay Area, CA
|
| Posted: Mon May 28, 2007 6:49 pm Post subject: |
|
|
| I meant which patches. I try to restrict the # of overlays on my system. Just curious. Is it just the hardened patches? |
|
| Back to top |
|
|
kernelOfTruth
Watchman
Joined: 20 Dec 2005
Posts: 6111
Location: Vienna, Austria; Germany; hello world :)
|
|
| Back to top |
|
|
devsk
Advocate
Joined: 24 Oct 2003
Posts: 2998
Location: Bay Area, CA
|
| Posted: Mon May 28, 2007 7:32 pm Post subject: |
|
|
| thanks for confirming that! |
|
| Back to top |
|
|
n0rbi666
l33t
Joined: 04 Mar 2005
Posts: 707
Location: \Poland\Krakow
|
| Posted: Mon May 28, 2007 8:35 pm Post subject: Re: Possible fix for kde-3.5.7 with gcc-4.2 on AMD64 |
|
|
| cneil wrote: | | n0rbi666 wrote: | I have binutils-2.17.50.0.15 and everything is ok, no segfaults.
With 2.17.50.0.16 - when emerging mesa, ld segfaults - so I get back to 2.17.50.0.15 |
Hmmm, I just compiled mesa on both an amd64 box and an x86 box (older athlon-tbird). Mesa compiled fine with binutils-2.17.50.0.16 on both boxes. I've been using the prerelease version of binutils since about .12 or .13 and have not personally experienced a single failure due to binutils in that time. |
Hm, maybe because I use mesa from r300-driver overlay (so mesa from git, for r300 radeon). With .16 it segfaults on ld, with .15 it works ok  (BTW I use snapshots from .4 or .5 and didn't noticed any problems so far ) |
|
| Back to top |
|
|
cneil
n00b
Joined: 28 May 2007
Posts: 39
Location: Stillwater, OK
|
| Posted: Mon May 28, 2007 8:57 pm Post subject: Re: Possible fix for kde-3.5.7 with gcc-4.2 on AMD64 |
|
|
| n0rbi666 wrote: | | Hm, maybe because I use mesa from r300-driver overlay (so mesa from git, for r300 radeon). With .16 it segfaults on ld, with .15 it works ok (BTW I use snapshots from .4 or .5 and didn't noticed any problems so far ) |
Maybe, the mesa straight out of portage (mesa-6.5.2-r1) compiled fine with .16.
Neil |
|
| Back to top |
|
|
ToeiRei
Veteran
Joined: 03 Jan 2005
Posts: 1191
Location: Austria
|
| Posted: Mon May 28, 2007 9:23 pm Post subject: |
|
|
Just my status here:
Tried gcc 4.2.0 from the official portage tree - did toolchain, emerge -e system, revdep-rebuild and tried emerge -e world
Restult so far: Building the packages was fine until emerge -e world hit the kdelibs. So I checked my system for strange things; Bad news: SSL/TLS functions broke. So I used my backup packages to get my system back stable.
Rei
_________________
Please stand by - The mailer daemon is busy burning your messages in hell... |
|
| Back to top |
|
|
Uzytkownik
Guru
Joined: 31 Oct 2004
Posts: 399
Location: Bay Area, US
|
| Posted: Mon May 28, 2007 9:25 pm Post subject: |
|
|
Some time ago on this forum was a patch for OpenSSL.
_________________
I've probably left my head... somwhere. Please wait untill I find it. |
|
| Back to top |
|
|
ToeiRei
Veteran
Joined: 03 Jan 2005
Posts: 1191
Location: Austria
|
| Posted: Mon May 28, 2007 9:29 pm Post subject: |
|
|
| Uzytkownik wrote: | | Some time ago on this forum was a patch for OpenSSL. |
in other words, I have to use an overlay for a sane build or wait for gcc 4.2.1?
Rei
_________________
Please stand by - The mailer daemon is busy burning your messages in hell... |
|
| Back to top |
|
|
Uzytkownik
Guru
Joined: 31 Oct 2004
Posts: 399
Location: Bay Area, US
|
| Posted: Mon May 28, 2007 9:31 pm Post subject: |
|
|
No. Have overlay or wait for OpenSSL 0.9.9 if I remember 
IMO it could be created centralized overlay for such things...
_________________
I've probably left my head... somwhere. Please wait untill I find it. |
|
| Back to top |
|
|
zxy
Veteran
Joined: 06 Jan 2006
Posts: 1160
Location: in bed in front of the computer
|
| Posted: Wed May 30, 2007 12:00 am Post subject: |
|
|
I've read a lot in this thread about which packages are failing and which need patches, ...
There is also some comparison to gcc-4.1.2 considering file sizes where 4.2.0 is reportedly creating bigger files (with -O2). Is it the same for -Os?
Did the speed of the programs compiled with 4.2.0 improve against gcc-4.1.2?
Are there any compariosons made with different -O* settings?
_________________
Nature does not hurry, yet everything is accomplished.
Lao Tzu |
|
| Back to top |
|
|
devsk
Advocate
Joined: 24 Oct 2003
Posts: 2998
Location: Bay Area, CA
|
| Posted: Wed May 30, 2007 12:41 am Post subject: |
|
|
I just did a comparison of -O2 and -Os for samba and the 4.5MB difference seems substantial to me:
before:
| Code: | # q size samba
net-fs/samba-3.0.24-r3: 137 files, 53 non-files, 45132.928 KB
|
after:
| Code: | # q size samba
net-fs/samba-3.0.24-r3: 138 files, 53 non-files, 40684.717 KB
|
What's people's experience with 4.1.2 with regards to -O2 vs. -Os? Do they vary in size by that much? |
|
| Back to top |
|
|
Anarchy
Developer
Joined: 29 Jun 2005
Posts: 140
|
| Posted: Wed May 30, 2007 2:03 am Post subject: gcc-4.2.0 regression 64bit |
|
|
| piwacet wrote: | http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30052
So let me see if I understand this: with certain packages on x86_64, gcc-4.2.0 requires lots of ram? This is basically a show-stopper for stabilization on AMD64 if I understand this correctly. Hope it's fixed in 4.2.1. My AMD64 compiled xorg-server-1.2 requiring most of my 1G of main memory and 1G of swap; took forever but did finish. |
I have just opened a bug for the backport of the patch with fortran fix for the ice. You might want to cc yourself on the bug report to keep up with it. https://bugs.gentoo.org/show_bug.cgi?id=180304 |
|
| Back to top |
|
|
devsk
Advocate
Joined: 24 Oct 2003
Posts: 2998
Location: Bay Area, CA
|
| Posted: Thu May 31, 2007 1:29 am Post subject: |
|
|
I have just noticed a very bizarre bug with 4.2.0. Compiling kicker-3.5.7 with latest prerelease LD and GCC 4.2.0, with -Os and -fforce-addr, I get a huge shared object. With -O2 and -fforce-addr, I get correct shared object. With just -Os, I get correct .so as well. Why is -Os and -fforce-addr combo producing such a huge .so?
| Code: |
# x86_64-pc-linux-gnu-g++ -DHAVE_CONFIG_H -I. -I. -I../../.. -I/usr/kde/3.5/include -I/usr/qt/3/include -I. -DQT_THREAD_SUPPORT -D_REENTRANT -ansi -D_XOPEN_SOURCE=500 -D_BSD_SOURCE -DNDEBUG -DNO_DEBUG -O2 -march=k8 -msse2 -msse3 -fforce-addr -pipe -fno-exceptions -fno-check-new -fno-common -fvisibility=hidden -fvisibility-inlines-hidden -DQT_CLEAN_NAMESPACE -DQT_NO_ASCII_CAST -DQT_NO_STL -DQT_NO_COMPAT -DQT_NO_TRANSLATION -MT naughty_panelapplet_la.all_cpp.lo -MD -MP -MF .deps/naughty_panelapplet_la.all_cpp.Tpo -c naughty_panelapplet_la.all_cpp.cpp -fPIC -DPIC -o .libs/naughty_panelapplet_la.all_cpp.o && \rm naughty_panelapplet.la && make >/dev/null && l .libs/naughty_panelapplet.so
-rwxr-xr-x 1 root root 107318 May 30 18:26 .libs/naughty_panelapplet.so*
# x86_64-pc-linux-gnu-g++ -DHAVE_CONFIG_H -I. -I. -I../../.. -I/usr/kde/3.5/include -I/usr/qt/3/include -I. -DQT_THREAD_SUPPORT -D_REENTRANT -ansi -D_XOPEN_SOURCE=500 -D_BSD_SOURCE -DNDEBUG -DNO_DEBUG -Os -march=k8 -msse2 -msse3 -fforce-addr -pipe -fno-exceptions -fno-check-new -fno-common -fvisibility=hidden -fvisibility-inlines-hidden -DQT_CLEAN_NAMESPACE -DQT_NO_ASCII_CAST -DQT_NO_STL -DQT_NO_COMPAT -DQT_NO_TRANSLATION -MT naughty_panelapplet_la.all_cpp.lo -MD -MP -MF .deps/naughty_panelapplet_la.all_cpp.Tpo -c naughty_panelapplet_la.all_cpp.cpp -fPIC -DPIC -o .libs/naughty_panelapplet_la.all_cpp.o && \rm naughty_panelapplet.la && make >/dev/null && l .libs/naughty_panelapplet.so
-rwxr-xr-x 1 root root 2203191 May 30 18:26 .libs/naughty_panelapplet.so*
# x86_64-pc-linux-gnu-g++ -DHAVE_CONFIG_H -I. -I. -I../../.. -I/usr/kde/3.5/include -I/usr/qt/3/include -I. -DQT_THREAD_SUPPORT -D_REENTRANT -ansi -D_XOPEN_SOURCE=500 -D_BSD_SOURCE -DNDEBUG -DNO_DEBUG -Os -march=k8 -msse2 -msse3 -pipe -fno-exceptions -fno-check-new -fno-common -fvisibility=hidden -fvisibility-inlines-hidden -DQT_CLEAN_NAMESPACE -DQT_NO_ASCII_CAST -DQT_NO_STL -DQT_NO_COMPAT -DQT_NO_TRANSLATION -MT naughty_panelapplet_la.all_cpp.lo -MD -MP -MF .deps/naughty_panelapplet_la.all_cpp.Tpo -c naughty_panelapplet_la.all_cpp.cpp -fPIC -DPIC -o .libs/naughty_panelapplet_la.all_cpp.o && \rm naughty_panelapplet.la && make >/dev/null && l .libs/naughty_panelapplet.so
-rwxr-xr-x 1 root root 106039 May 30 18:27 .libs/naughty_panelapplet.so*
|
|
|
| Back to top |
|
|
spielc
Guru
Joined: 20 Apr 2004
Posts: 452
|
| Posted: Thu May 31, 2007 5:20 pm Post subject: |
|
|
I have a short question: Is/Was anybody in here able to compile a version of qt>4.1.x? I'm not... Whenever i tried to compile a version of qt greater than 4.1.4-r2 i get the following compile error:
| Code: |
/usr/lib/gcc/x86_64-pc-linux-gnu/4.2.0-alpha20070117/../../../../x86_64-pc-linux-gnu/bin/ld: .obj/release-shared/qiodevice.o: relocation R_X86_64_PC32 against `QList<QByteArray>::clear()' can not be used when making a shared object; recompile with -fPIC
/usr/lib/gcc/x86_64-pc-linux-gnu/4.2.0-alpha20070117/../../../../x86_64-pc-linux-gnu/bin/ld: final link failed: Bad value
collect2: ld returned 1 exit status
make[1]: *** [../../lib/libQtCore.so.4.3.0] Error 1
make[1]: Leaving directory `/var/tmp/portage/x11-libs/qt-4.3.0_rc1/work/qt-x11-opensource-src-4.3.0rc1/src/corelib'
make: *** [sub-corelib-all-ordered] Error 2
|
That's more or less the only thing that doesn't compile on my machine with gcc-4.2 (well except the well known problem with openssl but that's not really a compilation problem...). Everything else amarok, gimp, gtk... compiles just fine with gcc-4.2 |
|
| Back to top |
|
|
devsk
Advocate
Joined: 24 Oct 2003
Posts: 2998
Location: Bay Area, CA
|
| Posted: Thu May 31, 2007 5:28 pm Post subject: |
|
|
| spielc wrote: | I have a short question: Is/Was anybody in here able to compile a version of qt>4.1.x? I'm not... Whenever i tried to compile a version of qt greater than 4.1.4-r2 i get the following compile error:
| Code: |
/usr/lib/gcc/x86_64-pc-linux-gnu/4.2.0-alpha20070117/../../../../x86_64-pc-linux-gnu/bin/ld: .obj/release-shared/qiodevice.o: relocation R_X86_64_PC32 against `QList<QByteArray>::clear()' can not be used when making a shared object; recompile with -fPIC
/usr/lib/gcc/x86_64-pc-linux-gnu/4.2.0-alpha20070117/../../../../x86_64-pc-linux-gnu/bin/ld: final link failed: Bad value
collect2: ld returned 1 exit status
make[1]: *** [../../lib/libQtCore.so.4.3.0] Error 1
make[1]: Leaving directory `/var/tmp/portage/x11-libs/qt-4.3.0_rc1/work/qt-x11-opensource-src-4.3.0rc1/src/corelib'
make: *** [sub-corelib-all-ordered] Error 2
|
That's more or less the only thing that doesn't compile on my machine with gcc-4.2 (well except the well known problem with openssl but that's not really a compilation problem...). Everything else amarok, gimp, gtk... compiles just fine with gcc-4.2 |
you missed the page 12 of this thread where it is mentioned that the problem goes away when
| Code: |
Upgrading to the latest prerelease version of binutils (currently 2.17.50.0.16)... |
|
|
| Back to top |
|
|
Ishiki
Tux's lil' helper
Joined: 31 Aug 2005
Posts: 86
|
| Posted: Fri Jun 01, 2007 11:01 am Post subject: |
|
|
gcc-4.2.0, binutils-2.17.50.0.16, glibc-2.5-r3
And can't compile kdelibs (3.5.7)
I used various CFLAGS, and it fails even with the basic ones (LDFLAGS cleared as well).
| Code: | /bin/sh ../libtool --silent --tag=CXX --mode=link i686-pc-linux-gnu-g++ -Wno-lo
ng-long -Wundef -ansi -D_XOPEN_SOURCE=500 -D_BSD_SOURCE -Wcast-align -Wchar-subs
cripts -Wall -W -Wpointer-arith -DNDEBUG -DNO_DEBUG -O2 -march=athlon-xp -O2 -fo
mit-frame-pointer -pipe -Wformat-security -Wmissing-format-attribute -Wno-non-vi
rtual-dtor -fno-exceptions -fno-check-new -fno-common -fvisibility=hidden -fvisi
bility-inlines-hidden -DQT_CLEAN_NAMESPACE -DQT_NO_ASCII_CAST -DQT_NO_STL -DQT_
NO_COMPAT -DQT_NO_TRANSLATION -o kunittestmodrunner -L/usr/kde/3.5/lib -L/usr
/qt/3/lib -R /usr/kde/3.5/lib -R /usr/kde/3.5/lib -R /usr/qt/3/lib modrunner
.o libkunittest.la ../kdecore/libkdecore.la
./.libs/libkunittest.so: undefined reference to `std::basic_ostream<char, std::c
har_traits<char> >& std::__ostream_insert<char, std::char_traits<char> >(std::ba
sic_ostream<char, std::char_traits<char> >&, char const*, int)'
collect2: ld returned 1 exit status
make[2]: *** [kunittestmodrunner] B³±d 1
make[2]: Opuszczenie katalogu `/var/tmp/portage/kde-base/kdelibs-3.5.7/work/kdel
ibs-3.5.7/kunittest'
make[1]: *** [all-recursive] B³±d 1
make[1]: Opuszczenie katalogu `/var/tmp/portage/kde-base/kdelibs-3.5.7/work/kdel
ibs-3.5.7'
make: *** [all] B³±d 2 |
"B³±d" is "error"
The rest of the system compiled with:
CFLAGSSAFE="-Os -march=athlon-xp -pipe -fomit-frame-pointer -ftracer"
COREFLAGS="-funit-at-a-time -freorder-blocks -freorder-blocks-and-partition -fmerge-all-constants"
CFLAGSPLUS="-fno-ident -frename-registers -fweb -fgcse-after-reload -ftree-vectorize"
CFLAGS="${CFLAGSSAFE} ${CFLAGSPLUS} ${COREFLAGS}"
CXXFLAGS="${CFLAGS} -fvisibility-inlines-hidden"
LDFLAGS1="-Wl,-O1 -Wl,--as-needed -Wl,-znow"
LDFLAGS="${LDFLAGS1} -Wl,--hash-style=both"
CFLAGS are mostly from CONRAD Install Guide, that is -Os and some from -O2, I also added a few (like -ftree-vectorize which should be fairly stable as a read). |
|
| Back to top |
|
|
pal_gene
n00b
Joined: 15 Nov 2005
Posts: 54
|
| Posted: Fri Jun 01, 2007 11:49 am Post subject: |
|
|
binutils-2.17.50.0.16 fail was resolved on PR 4497's fix for me.
http://sourceware.org/ml/binutils-cvs/2007-05/msg00064.html
but we should wait next alpha-release or use preview release 
_________________
what happen my profile #
[i speak "ENGRISH"!] |
|
| Back to top |
|
|
Ishiki
Tux's lil' helper
Joined: 31 Aug 2005
Posts: 86
|
| Posted: Sat Jun 02, 2007 10:44 am Post subject: |
|
|
| Genkaku wrote: | The rest of the system compiled with:
CFLAGSSAFE="-Os -march=athlon-xp -pipe -fomit-frame-pointer -ftracer"
COREFLAGS="-funit-at-a-time -freorder-blocks -freorder-blocks-and-partition -fmerge-all-constants"
CFLAGSPLUS="-fno-ident -frename-registers -fweb -fgcse-after-reload -ftree-vectorize"
CFLAGS="${CFLAGSSAFE} ${CFLAGSPLUS} ${COREFLAGS}"
CXXFLAGS="${CFLAGS} -fvisibility-inlines-hidden"
LDFLAGS1="-Wl,-O1 -Wl,--as-needed -Wl,-znow"
LDFLAGS="${LDFLAGS1} -Wl,--hash-style=both"
CFLAGS are mostly from CONRAD Install Guide, that is -Os and some from -O2, I also added a few (like -ftree-vectorize which should be fairly stable as a read). |
A lesson learned: do not even use CFLAGS from "Install Guides". |
|
| Back to top |
|
|
milan
n00b
Joined: 04 Dec 2003
Posts: 8
Location: Ljubljana
|
| Posted: Sat Jun 02, 2007 10:55 am Post subject: |
|
|
| zxy wrote: | | I still don't know wether should I stay or should I go with new version. Emerging world takes a few days and you have to be by the computer all the time, if something doesn't compile. |
You don't need to be by the computer to do emerge - world. I use this stupid script for years now:
| Code: |
#! /bin/sh
emerge -e world
#emerge -uND world
while [ $? != 0 ] ; do
emerge --resume --skipfirst
done
|
Then depending on the number of packages and number of cores used to compile you wait few hours or few days.
When finished you can check /var/log/emerge.log or /var/tmp/portage (providing that /var/tmp/portage was cleaned before running the above sctipt) to see which packages didn't compile and deal with them individually |
|
| Back to top |
|
|
|
Unsupported Software
