* angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each
@ 2010-07-15 17:55 Paul Menzel
2010-07-16 9:03 ` Paul Menzel
0 siblings, 1 reply; 14+ messages in thread
From: Paul Menzel @ 2010-07-15 17:55 UTC (permalink / raw)
To: openembedded-devel
[-- Attachment #1: Type: text/plain, Size: 600 bytes --]
Dear OE folks,
I ran out of space and noticed
/angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4
is 8,1G big.
Some of the executable files in
/angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core/
are 101 MB big.
This regression must have been introduced between this morning 6 a.m.
UTC and now.
I could not reproduce this using the distributions minimal-uclibc or
angstrom-2008.1 with glibc.
If you need more information I could try to downgrade the uclibc
revision.
Thanks,
Paul
[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 205 bytes --]
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each 2010-07-15 17:55 angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each Paul Menzel @ 2010-07-16 9:03 ` Paul Menzel 2010-07-16 15:23 ` Khem Raj 0 siblings, 1 reply; 14+ messages in thread From: Paul Menzel @ 2010-07-16 9:03 UTC (permalink / raw) To: openembedded-devel [-- Attachment #1: Type: text/plain, Size: 4541 bytes --] Dear OE folks, Am Donnerstag, den 15.07.2010, 19:55 +0200 schrieb Paul Menzel: > I ran out of space and noticed > > /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4 > > is 8,1G big. > > Some of the executable files in > > /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core/ > > are 101 MB big. > > This regression must have been introduced between this morning 6 a.m. > UTC and now. > > I could not reproduce this using the distributions minimal-uclibc or > angstrom-2008.1 with glibc. > > If you need more information I could try to downgrade the uclibc > revision. I can still reproduce the problem with angstrom-2008.1, uclibc-git and OE dev branch HEAD from today 9fd5a8cb38a376633eb0bd7981548179b6bea25d. /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core$ ls -lh | sort -k 5 […] -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-annotate -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-apply -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-archive -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-blame -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-branch -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-bundle -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-cherry -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-cherry-pick -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-clean -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-clone -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-commit -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-commit-tree -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-config -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-diff -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-fetch -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-fmt-merge-msg -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-for-each-ref -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-fsck -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-fsck-objects -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-grep -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-hash-object -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-help -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-index-pack -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-init -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-init-db -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-ls-remote -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-mailsplit -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-merge -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-merge-recursive -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-merge-subtree -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-mktag -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-mktree -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-name-rev -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-pack-redundant -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-pack-refs -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-prune -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-push -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-read-tree -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-reset -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-rev-list -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-shortlog -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-show -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-show-branch -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-show-ref -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-stage -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-stripspace -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-symbolic-ref -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-update-index -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-upload-archive -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-verify-tag -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-write-tree […] Could you please check your work directory and see if you have the same problem. Thanks, Paul [-- Attachment #2: This is a digitally signed message part --] [-- Type: application/pgp-signature, Size: 205 bytes --] ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each 2010-07-16 9:03 ` Paul Menzel @ 2010-07-16 15:23 ` Khem Raj 2010-07-16 23:03 ` Koen Kooi 0 siblings, 1 reply; 14+ messages in thread From: Khem Raj @ 2010-07-16 15:23 UTC (permalink / raw) To: openembedded-devel On Fri, Jul 16, 2010 at 2:03 AM, Paul Menzel <paulepanter@users.sourceforge.net> wrote: > Dear OE folks, > > > Am Donnerstag, den 15.07.2010, 19:55 +0200 schrieb Paul Menzel: > >> I ran out of space and noticed >> >> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4 >> >> is 8,1G big. >> >> Some of the executable files in >> >> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core/ >> >> are 101 MB big. its most probably the -g3/-ggdb3 bloat. and this option should be removed from build flags atleast so the native binaries dont bloat. here is a little change that you will need diff --git a/conf/distro/include/angstrom-uclibc.inc b/conf/distro/include/angstrom-uc index 4ca6838..1c7f7f5 100644 --- a/conf/distro/include/angstrom-uclibc.inc +++ b/conf/distro/include/angstrom-uclibc.inc @@ -15,7 +15,7 @@ TARGET_OS = "${@['${TARGET_OS_UC}', 'uclinux-uclibc'][bb.data.getVar #mess with compiler flags to use -Os instead of -O2 #Please see http://free-electrons.com/doc/embedded_linux_optimizations/img47.html for FULL_OPTIMIZATION = "-fexpensive-optimizations -fomit-frame-pointer -frename-register -BUILD_OPTIMIZATION = "-O2 -ggdb3" +BUILD_OPTIMIZATION = "-O2" its ok with other libcs for angstrom as the flag has already been removed from them. Try this and rebake git-native after cleaning it Thx -Khem >> >> This regression must have been introduced between this morning 6 a.m. >> UTC and now. >> >> I could not reproduce this using the distributions minimal-uclibc or >> angstrom-2008.1 with glibc. >> >> If you need more information I could try to downgrade the uclibc >> revision. > > I can still reproduce the problem with angstrom-2008.1, uclibc-git and > OE dev branch HEAD from today 9fd5a8cb38a376633eb0bd7981548179b6bea25d. > > /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core$ ls -lh | sort -k 5 > […] > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-annotate > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-apply > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-archive > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-blame > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-branch > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-bundle > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-cherry > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-cherry-pick > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-clean > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-clone > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-commit > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-commit-tree > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-config > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-diff > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-fetch > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-fmt-merge-msg > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-for-each-ref > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-fsck > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-fsck-objects > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-grep > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-hash-object > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-help > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-index-pack > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-init > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-init-db > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-ls-remote > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-mailsplit > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-merge > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-merge-recursive > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-merge-subtree > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-mktag > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-mktree > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-name-rev > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-pack-redundant > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-pack-refs > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-prune > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-push > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-read-tree > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-reset > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-rev-list > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-shortlog > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-show > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-show-branch > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-show-ref > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-stage > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-stripspace > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-symbolic-ref > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-update-index > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-upload-archive > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-verify-tag > -rwxr-xr-x 1 usern usern 101M 16. Jul 10:51 git-write-tree > […] > > Could you please check your work directory and see if you have the same > problem. > > > Thanks, > > Paul > > _______________________________________________ > Openembedded-devel mailing list > Openembedded-devel@lists.openembedded.org > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel > > ^ permalink raw reply related [flat|nested] 14+ messages in thread
* Re: angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each 2010-07-16 15:23 ` Khem Raj @ 2010-07-16 23:03 ` Koen Kooi 2010-07-17 2:06 ` Khem Raj ` (2 more replies) 0 siblings, 3 replies; 14+ messages in thread From: Koen Kooi @ 2010-07-16 23:03 UTC (permalink / raw) To: openembedded-devel -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 16-07-10 17:23, Khem Raj wrote: > On Fri, Jul 16, 2010 at 2:03 AM, Paul Menzel > <paulepanter@users.sourceforge.net> wrote: >> Dear OE folks, >> >> >> Am Donnerstag, den 15.07.2010, 19:55 +0200 schrieb Paul Menzel: >> >>> I ran out of space and noticed >>> >>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4 >>> >>> is 8,1G big. >>> >>> Some of the executable files in >>> >>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core/ >>> >>> are 101 MB big. > > its most probably the -g3/-ggdb3 bloat. No, it's git not creating (hard)links but copies of the same app. Run an md5sum on the binaries, they will all match. I suggest you stop spreading FUD about -ggdb3, it is not being appriciated. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (Darwin) iD8DBQFMQOVfMkyGM64RGpERAjW7AKCwH5aLnQ5AbEC8PESZf7Ny7psLsQCgt1Ve Z1rBOhQsiOxILfsDfyNLqrE= =OQOZ -----END PGP SIGNATURE----- ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each 2010-07-16 23:03 ` Koen Kooi @ 2010-07-17 2:06 ` Khem Raj 2010-07-17 12:17 ` Koen Kooi 2010-07-17 7:32 ` Khem Raj 2010-07-17 8:35 ` Paul Menzel 2 siblings, 1 reply; 14+ messages in thread From: Khem Raj @ 2010-07-17 2:06 UTC (permalink / raw) To: openembedded-devel On Fri, Jul 16, 2010 at 4:03 PM, Koen Kooi <k.kooi@student.utwente.nl> wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On 16-07-10 17:23, Khem Raj wrote: >> On Fri, Jul 16, 2010 at 2:03 AM, Paul Menzel >> <paulepanter@users.sourceforge.net> wrote: >>> Dear OE folks, >>> >>> >>> Am Donnerstag, den 15.07.2010, 19:55 +0200 schrieb Paul Menzel: >>> >>>> I ran out of space and noticed >>>> >>>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4 >>>> >>>> is 8,1G big. >>>> >>>> Some of the executable files in >>>> >>>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core/ >>>> >>>> are 101 MB big. >> >> its most probably the -g3/-ggdb3 bloat. > > No, it's git not creating (hard)links but copies of the same app. Run an > md5sum on the binaries, they will all match. I suggest you stop > spreading FUD about -ggdb3, it is not being appriciated. Hi Koen Its not FUD and I only speak of what I see with my eyes. I have data to prove it. Compile uclibc with ggdb3 and it ends up with 202M of shared library and without its only ~600K. (with gcc 4.4 and 4.5) I have similar experience with 4.3 but I dont have a build handy to see how big the binaries got. same for git native without -ggdb3 I get 1.4M of git executable and above 100M with -ggdb3 now if you copy 101M of binary 10 times it will highlight the problem even more. I will post complete result of a consol-image build with -ggdb3 and without it for reference and to clear FUD as well. I agree FUD should not be appreciated but facts should. Thanks -Khem > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.5 (Darwin) > > iD8DBQFMQOVfMkyGM64RGpERAjW7AKCwH5aLnQ5AbEC8PESZf7Ny7psLsQCgt1Ve > Z1rBOhQsiOxILfsDfyNLqrE= > =OQOZ > -----END PGP SIGNATURE----- > > > _______________________________________________ > Openembedded-devel mailing list > Openembedded-devel@lists.openembedded.org > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel > ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each 2010-07-17 2:06 ` Khem Raj @ 2010-07-17 12:17 ` Koen Kooi 2010-07-17 17:17 ` Phil Blundell 0 siblings, 1 reply; 14+ messages in thread From: Koen Kooi @ 2010-07-17 12:17 UTC (permalink / raw) To: openembedded-devel -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 17-07-10 04:06, Khem Raj wrote: > On Fri, Jul 16, 2010 at 4:03 PM, Koen Kooi <k.kooi@student.utwente.nl> wrote: > On 16-07-10 17:23, Khem Raj wrote: >>>> On Fri, Jul 16, 2010 at 2:03 AM, Paul Menzel >>>> <paulepanter@users.sourceforge.net> wrote: >>>>> Dear OE folks, >>>>> >>>>> >>>>> Am Donnerstag, den 15.07.2010, 19:55 +0200 schrieb Paul Menzel: >>>>> >>>>>> I ran out of space and noticed >>>>>> >>>>>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4 >>>>>> >>>>>> is 8,1G big. >>>>>> >>>>>> Some of the executable files in >>>>>> >>>>>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core/ >>>>>> >>>>>> are 101 MB big. >>>> >>>> its most probably the -g3/-ggdb3 bloat. > > No, it's git not creating (hard)links but copies of the same app. Run an > md5sum on the binaries, they will all match. I suggest you stop > spreading FUD about -ggdb3, it is not being appriciated. > >> Hi Koen > >> Its not FUD and I only speak of what I see with my eyes. I have data >> to prove it. Compile uclibc with ggdb3 and it ends up with 202M of >> shared library and without its only ~600K. (with gcc 4.4 and 4.5) I >> have similar experience with 4.3 but I dont >> have a build handy to see how big the binaries got. And I was saying that the 8GB of git, if a bug in git, not with gdb3. I know gdb3 makes unstripped executables bigger and don't deny that. I do take offence with you saying that the root cause is -ggdb3, when it's actually git being buggy. And as you have seen, instead of fixing git, people are removing -ggdb3 :( regards, Koen -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (Darwin) iD8DBQFMQZ9yMkyGM64RGpERAvrYAJ0XYJZ40fAfTpWs5eCdEuWZlidzQgCgnDiw sWRwYL9QbYfPDgdyp47pwzk= =hKZ3 -----END PGP SIGNATURE----- ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each 2010-07-17 12:17 ` Koen Kooi @ 2010-07-17 17:17 ` Phil Blundell 0 siblings, 0 replies; 14+ messages in thread From: Phil Blundell @ 2010-07-17 17:17 UTC (permalink / raw) To: openembedded-devel On Sat, 2010-07-17 at 14:17 +0200, Koen Kooi wrote: > And I was saying that the 8GB of git, if a bug in git, not with gdb3. I > know gdb3 makes unstripped executables bigger and don't deny that. > > I do take offence with you saying that the root cause is -ggdb3, when > it's actually git being buggy. And as you have seen, instead of fixing > git, people are removing -ggdb3 :( It may well be true that the 8GB total folder size is due to a deficiency in git. But, if the individual binaries are coming out at 101MB, that's still an increase of something like 10000% compared to an equivalent build without debugging. Obviously it's up to angstrom to make its own policies, but I do wonder how many of its users actually find themselves debugging -native packages on a regular basis and, for those that do, how much benefit they are really gaining from -ggdb3 versus plain -g. p. ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each 2010-07-16 23:03 ` Koen Kooi 2010-07-17 2:06 ` Khem Raj @ 2010-07-17 7:32 ` Khem Raj 2010-07-17 8:35 ` Paul Menzel 2 siblings, 0 replies; 14+ messages in thread From: Khem Raj @ 2010-07-17 7:32 UTC (permalink / raw) To: openembedded-devel [-- Attachment #1: Type: text/plain, Size: 2431 bytes --] On (17/07/10 01:03), Koen Kooi wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On 16-07-10 17:23, Khem Raj wrote: > > On Fri, Jul 16, 2010 at 2:03 AM, Paul Menzel > > <paulepanter@users.sourceforge.net> wrote: > >> Dear OE folks, > >> > >> > >> Am Donnerstag, den 15.07.2010, 19:55 +0200 schrieb Paul Menzel: > >> > >>> I ran out of space and noticed > >>> > >>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4 > >>> > >>> is 8,1G big. > >>> > >>> Some of the executable files in > >>> > >>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core/ > >>> > >>> are 101 MB big. > > > > its most probably the -g3/-ggdb3 bloat. > > No, it's git not creating (hard)links but copies of the same app. Run an > md5sum on the binaries, they will all match. I suggest you stop > spreading FUD about -ggdb3, it is not being appriciated. Hi Koen I built minimal-image/uclibc from scratch on a ubuntu maverick (gcc 4.4.4) host once with -g and secong time using -ggdb3 instead in both FULL_OPTIMIZATION and BUILD_OPTIMIZATION Total tmp size with -g = 15G Total tmp size with -ggdb3 = 70G Total sysroot size with -g = 1.1G (868M is native sysroot ) Total sysroot size with -ggdb3 = 13.4G ( 12.7G is native sysroot ) Total size of cross directory with -g = 158M Total size of cross directory with -gdb3 = 567M Because I did not strip the binaries Total size of deploy with -g = 135M Total size of deploy with -ggdb3 = 521M Attached is a list of files in sysroot which are larger than 2M in size sorted in increasing size order for both -g and -ggdb generated by using find . -type f -size +2000k -exec ls -lh {} \; 2> /dev/null | awk '{ print $NF ": " $5 }' | sort -nk 2,2 btw git-native build tree is 24G which is a problem but if we ignore that tmp still is ~45G for a minimal-image Does that clear the FUD ? :) anyway my point was that users know what to expect. If we put it in minimum requirement to build a distro its fine atleast it sets expectation on how much disk space would be needed. btw. now may I request you for some emperical evidence on "gcc 4.4 is slow on ARMv7" or is it another FUD or hearsay ? :) Thanks -Khem ps: I also noticed that the build with -ggdb3 took noticeable longer time. I wish I had timed it but may be next time. [-- Attachment #2: file-size-with-g.txt --] [-- Type: text/plain, Size: 8440 bytes --] ./x86_64-linux/usr/lib/libcrypto.so.0.9.8: 2.0M ./mipsel-oe-linux-uclibc/usr/lib/libgcrypt.a: 2.1M ./mipsel-oe-linux-uclibc/usr/lib/libgio-2.0.so.0.2400.1: 2.4M ./mipsel-oe-linux-uclibc/usr/lib/libglib-2.0.so.0.2400.1: 2.4M ./x86_64-linux/usr/lib/libncurses.a: 2.5M ./mipsel-oe-linux-uclibc/usr/lib/libgnutls.so.26.14.12: 2.6M ./x86_64-linux/usr/bin/qemu-cris: 2.8M ./x86_64-linux/usr/lib/libgio-2.0.so.0.2400.1: 2.8M ./x86_64-linux/usr/bin/qemu-sh4: 3.0M ./mipsel-oe-linux-uclibc/lib/libuClibc-0.9.32-git.so: 3.1M ./x86_64-linux/usr/bin/qemu-microblaze: 3.1M ./x86_64-linux/usr/bin/qemu-sh4eb: 3.1M ./x86_64-linux/usr/lib/libglib-2.0.so.0.2400.1: 3.1M ./x86_64-linux/usr/lib/libcrypto.a: 3.3M ./x86_64-linux/usr/bin/qemu-m68k: 3.4M ./x86_64-linux/usr/bin/qemu-sparc: 3.4M ./x86_64-linux/usr/bin/qemu-alpha: 3.5M ./x86_64-linux/usr/lib/libxml2.so.2.7.7: 3.9M ./mipsel-oe-linux-uclibc/usr/lib/libgnutls.a: 4.1M ./x86_64-linux/usr/bin/git: 4.2M ./x86_64-linux/usr/bin/git-receive-pack: 4.2M ./x86_64-linux/usr/bin/git-upload-archive: 4.2M ./x86_64-linux/usr/libexec/git-core/git: 4.2M ./x86_64-linux/usr/libexec/git-core/git-add: 4.2M ./x86_64-linux/usr/libexec/git-core/git-annotate: 4.2M ./x86_64-linux/usr/libexec/git-core/git-apply: 4.2M ./x86_64-linux/usr/libexec/git-core/git-archive: 4.2M ./x86_64-linux/usr/libexec/git-core/git-bisect--helper: 4.2M ./x86_64-linux/usr/libexec/git-core/git-blame: 4.2M ./x86_64-linux/usr/libexec/git-core/git-branch: 4.2M ./x86_64-linux/usr/libexec/git-core/git-bundle: 4.2M ./x86_64-linux/usr/libexec/git-core/git-cat-file: 4.2M ./x86_64-linux/usr/libexec/git-core/git-check-attr: 4.2M ./x86_64-linux/usr/libexec/git-core/git-checkout: 4.2M ./x86_64-linux/usr/libexec/git-core/git-checkout-index: 4.2M ./x86_64-linux/usr/libexec/git-core/git-check-ref-format: 4.2M ./x86_64-linux/usr/libexec/git-core/git-cherry: 4.2M ./x86_64-linux/usr/libexec/git-core/git-cherry-pick: 4.2M ./x86_64-linux/usr/libexec/git-core/git-clean: 4.2M ./x86_64-linux/usr/libexec/git-core/git-clone: 4.2M ./x86_64-linux/usr/libexec/git-core/git-commit: 4.2M ./x86_64-linux/usr/libexec/git-core/git-commit-tree: 4.2M ./x86_64-linux/usr/libexec/git-core/git-config: 4.2M ./x86_64-linux/usr/libexec/git-core/git-count-objects: 4.2M ./x86_64-linux/usr/libexec/git-core/git-describe: 4.2M ./x86_64-linux/usr/libexec/git-core/git-diff: 4.2M ./x86_64-linux/usr/libexec/git-core/git-diff-files: 4.2M ./x86_64-linux/usr/libexec/git-core/git-diff-index: 4.2M ./x86_64-linux/usr/libexec/git-core/git-diff-tree: 4.2M ./x86_64-linux/usr/libexec/git-core/git-fast-export: 4.2M ./x86_64-linux/usr/libexec/git-core/git-fetch: 4.2M ./x86_64-linux/usr/libexec/git-core/git-fetch-pack: 4.2M ./x86_64-linux/usr/libexec/git-core/git-fmt-merge-msg: 4.2M ./x86_64-linux/usr/libexec/git-core/git-for-each-ref: 4.2M ./x86_64-linux/usr/libexec/git-core/git-format-patch: 4.2M ./x86_64-linux/usr/libexec/git-core/git-fsck: 4.2M ./x86_64-linux/usr/libexec/git-core/git-fsck-objects: 4.2M ./x86_64-linux/usr/libexec/git-core/git-gc: 4.2M ./x86_64-linux/usr/libexec/git-core/git-get-tar-commit-id: 4.2M ./x86_64-linux/usr/libexec/git-core/git-grep: 4.2M ./x86_64-linux/usr/libexec/git-core/git-hash-object: 4.2M ./x86_64-linux/usr/libexec/git-core/git-help: 4.2M ./x86_64-linux/usr/libexec/git-core/git-index-pack: 4.2M ./x86_64-linux/usr/libexec/git-core/git-init: 4.2M ./x86_64-linux/usr/libexec/git-core/git-init-db: 4.2M ./x86_64-linux/usr/libexec/git-core/git-log: 4.2M ./x86_64-linux/usr/libexec/git-core/git-ls-files: 4.2M ./x86_64-linux/usr/libexec/git-core/git-ls-remote: 4.2M ./x86_64-linux/usr/libexec/git-core/git-ls-tree: 4.2M ./x86_64-linux/usr/libexec/git-core/git-mailinfo: 4.2M ./x86_64-linux/usr/libexec/git-core/git-mailsplit: 4.2M ./x86_64-linux/usr/libexec/git-core/git-merge: 4.2M ./x86_64-linux/usr/libexec/git-core/git-merge-base: 4.2M ./x86_64-linux/usr/libexec/git-core/git-merge-file: 4.2M ./x86_64-linux/usr/libexec/git-core/git-merge-index: 4.2M ./x86_64-linux/usr/libexec/git-core/git-merge-ours: 4.2M ./x86_64-linux/usr/libexec/git-core/git-merge-recursive: 4.2M ./x86_64-linux/usr/libexec/git-core/git-merge-subtree: 4.2M ./x86_64-linux/usr/libexec/git-core/git-merge-tree: 4.2M ./x86_64-linux/usr/libexec/git-core/git-mktag: 4.2M ./x86_64-linux/usr/libexec/git-core/git-mktree: 4.2M ./x86_64-linux/usr/libexec/git-core/git-mv: 4.2M ./x86_64-linux/usr/libexec/git-core/git-name-rev: 4.2M ./x86_64-linux/usr/libexec/git-core/git-pack-objects: 4.2M ./x86_64-linux/usr/libexec/git-core/git-pack-redundant: 4.2M ./x86_64-linux/usr/libexec/git-core/git-pack-refs: 4.2M ./x86_64-linux/usr/libexec/git-core/git-patch-id: 4.2M ./x86_64-linux/usr/libexec/git-core/git-peek-remote: 4.2M ./x86_64-linux/usr/libexec/git-core/git-prune: 4.2M ./x86_64-linux/usr/libexec/git-core/git-prune-packed: 4.2M ./x86_64-linux/usr/libexec/git-core/git-push: 4.2M ./x86_64-linux/usr/libexec/git-core/git-read-tree: 4.2M ./x86_64-linux/usr/libexec/git-core/git-receive-pack: 4.2M ./x86_64-linux/usr/libexec/git-core/git-reflog: 4.2M ./x86_64-linux/usr/libexec/git-core/git-remote: 4.2M ./x86_64-linux/usr/libexec/git-core/git-replace: 4.2M ./x86_64-linux/usr/libexec/git-core/git-repo-config: 4.2M ./x86_64-linux/usr/libexec/git-core/git-rerere: 4.2M ./x86_64-linux/usr/libexec/git-core/git-reset: 4.2M ./x86_64-linux/usr/libexec/git-core/git-revert: 4.2M ./x86_64-linux/usr/libexec/git-core/git-rev-list: 4.2M ./x86_64-linux/usr/libexec/git-core/git-rev-parse: 4.2M ./x86_64-linux/usr/libexec/git-core/git-rm: 4.2M ./x86_64-linux/usr/libexec/git-core/git-send-pack: 4.2M ./x86_64-linux/usr/libexec/git-core/git-shortlog: 4.2M ./x86_64-linux/usr/libexec/git-core/git-show: 4.2M ./x86_64-linux/usr/libexec/git-core/git-show-branch: 4.2M ./x86_64-linux/usr/libexec/git-core/git-show-ref: 4.2M ./x86_64-linux/usr/libexec/git-core/git-stage: 4.2M ./x86_64-linux/usr/libexec/git-core/git-status: 4.2M ./x86_64-linux/usr/libexec/git-core/git-stripspace: 4.2M ./x86_64-linux/usr/libexec/git-core/git-symbolic-ref: 4.2M ./x86_64-linux/usr/libexec/git-core/git-tag: 4.2M ./x86_64-linux/usr/libexec/git-core/git-tar-tree: 4.2M ./x86_64-linux/usr/libexec/git-core/git-unpack-file: 4.2M ./x86_64-linux/usr/libexec/git-core/git-unpack-objects: 4.2M ./x86_64-linux/usr/libexec/git-core/git-update-index: 4.2M ./x86_64-linux/usr/libexec/git-core/git-update-ref: 4.2M ./x86_64-linux/usr/libexec/git-core/git-update-server-info: 4.2M ./x86_64-linux/usr/libexec/git-core/git-upload-archive: 4.2M ./x86_64-linux/usr/libexec/git-core/git-var: 4.2M ./x86_64-linux/usr/libexec/git-core/git-verify-pack: 4.2M ./x86_64-linux/usr/libexec/git-core/git-verify-tag: 4.2M ./x86_64-linux/usr/libexec/git-core/git-whatchanged: 4.2M ./x86_64-linux/usr/libexec/git-core/git-write-tree: 4.2M ./x86_64-linux/usr/bin/qemu-i386: 4.3M ./x86_64-linux/usr/bin/qemu-mipsel: 4.3M ./x86_64-linux/usr/bin/qemu-sparc32plus: 4.3M ./x86_64-linux/usr/bin/qemu-arm: 4.4M ./x86_64-linux/usr/bin/qemu-sparc64: 4.4M ./x86_64-linux/usr/bin/qemu-x86_64: 4.4M ./x86_64-linux/usr/bin/qemu-mips: 4.5M ./x86_64-linux/usr/bin/qemu-armeb: 4.6M ./qemumipsel-oe-linux-uclibc/kernel/vmlinux: 5.1M ./x86_64-linux/usr/lib/libpython2.6.so.1.0: 5.1M ./x86_64-linux/usr/bin/qemu-system-cris: 5.3M ./x86_64-linux/usr/bin/qemu-system-microblaze: 5.3M ./x86_64-linux/usr/bin/qemu-system-m68k: 5.7M ./x86_64-linux/usr/bin/qemu-system-sh4: 5.7M ./x86_64-linux/usr/bin/qemu-system-sh4eb: 5.7M ./x86_64-linux/usr/bin/qemu-system-sparc: 6.1M ./x86_64-linux/usr/bin/qemu-ppc: 6.4M ./x86_64-linux/usr/lib/libxml2.a: 6.6M ./mipsel-oe-linux-uclibc/usr/lib/libc.a: 6.7M ./mipsel-oe-linux-uclibc/usr/lib/libstdc++.a: 6.9M ./x86_64-linux/usr/bin/qemu-ppc64: 7.8M ./x86_64-linux/usr/bin/qemu-ppc64abi32: 7.8M ./x86_64-linux/usr/bin/qemu-system-sparc64: 7.8M ./x86_64-linux/usr/lib/python2.6/config/libpython2.6.a: 8.3M ./x86_64-linux/usr/bin/qemu: 8.5M ./x86_64-linux/usr/bin/qemu-system-mipsel: 8.6M ./x86_64-linux/usr/bin/qemu-system-mips: 8.7M ./x86_64-linux/usr/bin/qemu-system-x86_64: 8.8M ./x86_64-linux/usr/bin/qemu-system-mips64: 9.0M ./x86_64-linux/usr/bin/qemu-system-mips64el: 9.0M ./x86_64-linux/usr/bin/qemu-system-arm: 9.1M ./x86_64-linux/usr/bin/qemu-system-ppc: 10M ./x86_64-linux/usr/bin/qemu-system-ppcemb: 10M ./x86_64-linux/usr/bin/qemu-system-ppc64: 12M [-- Attachment #3: file-size-with-ggdb3.txt --] [-- Type: text/plain, Size: 11371 bytes --] ./x86_64-linux/usr/lib/libcrypto.so.0.9.8: 2.0M ./mipsel-oe-linux-uclibc/usr/lib/libacl.so.1.1.0: 2.1M ./x86_64-linux/usr/bin/dir: 2.1M ./x86_64-linux/usr/bin/ls: 2.1M ./x86_64-linux/usr/bin/rm: 2.1M ./x86_64-linux/usr/bin/stdbuf: 2.1M ./x86_64-linux/usr/bin/vdir: 2.1M ./mipsel-oe-linux-uclibc/lib/librt-0.9.32-git.so: 2.2M ./mipsel-oe-linux-uclibc/usr/lib/libacl.a: 2.3M ./mipsel-oe-linux-uclibc/usr/lib/libgobject-2.0.so.0.2400.1: 2.4M ./x86_64-linux/usr/bin/sort: 2.4M ./x86_64-linux/usr/bin/df: 2.5M ./x86_64-linux/usr/lib/libform.so.5.0.4: 2.5M ./x86_64-linux/usr/lib/libgobject-2.0.so.0.2400.1: 2.5M ./mipsel-oe-linux-uclibc/usr/lib/libform.so.5.0.4: 2.7M ./x86_64-linux/usr/bin/du: 2.8M ./mipsel-oe-linux-uclibc/usr/lib/libform.a: 2.9M ./x86_64-linux/usr/lib/libform.a: 2.9M ./mipsel-oe-linux-uclibc/usr/lib/liblzo.so.1.0.0: 3.1M ./mipsel-oe-linux-uclibc/usr/lib/liblzo.a: 3.3M ./x86_64-linux/usr/lib/libcrypto.a: 3.3M ./x86_64-linux/usr/bin/cp: 3.5M ./x86_64-linux/usr/bin/mv: 3.5M ./x86_64-linux/usr/bin/m4: 3.6M ./x86_64-linux/usr/lib/libopkg.so.0.0.0: 3.7M ./x86_64-linux/usr/bin/install: 3.8M ./x86_64-linux/usr/lib/libgettextsrc-0.18.so: 3.8M ./mipsel-oe-linux-uclibc/usr/lib/libopkg.so.0.0.0: 4.0M ./x86_64-linux/usr/bin/qemu-img: 4.1M ./x86_64-linux/usr/bin/qemu-nbd: 4.1M ./x86_64-linux/usr/bin/qemu-io: 4.2M ./mipsel-oe-linux-uclibc/usr/lib/libopkg.a: 4.3M ./x86_64-linux/usr/lib/libopkg.a: 4.3M ./mipsel-oe-linux-uclibc/lib/libm-0.9.32-git.so: 4.5M ./mipsel-oe-linux-uclibc/usr/lib/libm.a: 4.9M ./x86_64-linux/usr/lib/libelf.so.0.8.13: 5.0M ./qemumipsel-oe-linux-uclibc/kernel/vmlinux: 5.1M ./x86_64-linux/usr/lib/libpython2.6.so.1.0: 5.1M ./x86_64-linux/usr/lib/libelf.a: 5.5M ./x86_64-linux/usr/bin/qemu-cris: 5.7M ./mipsel-oe-linux-uclibc/usr/lib/libgcrypt.so.11.4.4: 5.9M ./x86_64-linux/usr/bin/qemu-sh4: 6.1M ./x86_64-linux/usr/bin/qemu-microblaze: 6.2M ./x86_64-linux/usr/bin/qemu-sparc: 6.2M ./x86_64-linux/usr/bin/qemu-sh4eb: 6.3M ./x86_64-linux/usr/bin/qemu-alpha: 6.5M ./x86_64-linux/usr/bin/qemu-m68k: 6.6M ./x86_64-linux/usr/lib/libgettextpo.so.0.5.0: 6.6M ./mipsel-oe-linux-uclibc/usr/lib/libgcrypt.a: 6.7M ./mipsel-oe-linux-uclibc/usr/lib/libthread_db.a: 6.7M ./mipsel-oe-linux-uclibc/usr/lib/libstdc++.a: 6.9M ./x86_64-linux/usr/bin/qemu-sparc32plus: 7.1M ./x86_64-linux/usr/bin/qemu-sparc64: 7.3M ./x86_64-linux/usr/bin/qemu-mipsel: 7.4M ./x86_64-linux/usr/lib/libgettextpo.a: 7.4M ./x86_64-linux/usr/bin/qemu-mips: 7.6M ./x86_64-linux/usr/bin/qemu-x86_64: 7.6M ./x86_64-linux/usr/bin/qemu-i386: 7.7M ./mipsel-oe-linux-uclibc/usr/lib/libblkid.so.1.1.0: 8.0M ./x86_64-linux/usr/lib/python2.6/config/libpython2.6.a: 8.3M ./mipsel-oe-linux-uclibc/usr/lib/libblkid.a: 8.4M ./x86_64-linux/usr/lib/libxml2.a: 8.4M ./x86_64-linux/usr/lib/libgettextlib-0.18.so: 8.8M ./x86_64-linux/usr/bin/qemu-arm: 9.0M ./x86_64-linux/usr/bin/qemu-armeb: 9.2M ./x86_64-linux/usr/bin/qemu-ppc: 9.6M ./x86_64-linux/usr/lib/libxml2.so.2.7.7: 9.9M ./x86_64-linux/usr/bin/qemu-ppc64: 11M ./x86_64-linux/usr/bin/qemu-ppc64abi32: 11M ./mipsel-oe-linux-uclibc/usr/lib/libgnutls.so.26.14.12: 14M ./x86_64-linux/usr/lib/libglib-2.0.so.0.2400.1: 14M ./mipsel-oe-linux-uclibc/usr/lib/libglib-2.0.so.0.2400.1: 15M ./mipsel-oe-linux-uclibc/usr/lib/libgnutls.a: 16M ./x86_64-linux/usr/lib/libncurses.so.5.0.4: 16M ./mipsel-oe-linux-uclibc/usr/lib/libgio-2.0.so.0.2400.1: 17M ./mipsel-oe-linux-uclibc/usr/lib/libncurses.a: 17M ./mipsel-oe-linux-uclibc/usr/lib/libncurses.so.5.0.4: 17M ./x86_64-linux/usr/lib/libgio-2.0.so.0.2400.1: 17M ./x86_64-linux/usr/lib/libncurses.a: 17M ./mipsel-oe-linux-uclibc/lib/libpthread-0.9.32-git.so: 22M ./mipsel-oe-linux-uclibc/usr/lib/libpthread.a: 23M ./x86_64-linux/usr/bin/qemu-system-cris: 24M ./x86_64-linux/usr/bin/qemu-system-m68k: 24M ./x86_64-linux/usr/bin/qemu-system-microblaze: 24M ./x86_64-linux/usr/bin/qemu-system-sh4: 25M ./x86_64-linux/usr/bin/qemu-system-sh4eb: 25M ./x86_64-linux/usr/bin/qemu-system-sparc: 25M ./x86_64-linux/usr/bin/qemu-system-sparc64: 27M ./x86_64-linux/usr/bin/qemu: 32M ./x86_64-linux/usr/bin/qemu-system-mips: 32M ./x86_64-linux/usr/bin/qemu-system-mips64: 32M ./x86_64-linux/usr/bin/qemu-system-mips64el: 32M ./x86_64-linux/usr/bin/qemu-system-mipsel: 32M ./x86_64-linux/usr/bin/qemu-system-x86_64: 32M ./x86_64-linux/usr/bin/qemu-system-ppc: 34M ./x86_64-linux/usr/bin/qemu-system-ppcemb: 34M ./x86_64-linux/usr/bin/qemu-system-arm: 36M ./x86_64-linux/usr/bin/qemu-system-ppc64: 36M ./x86_64-linux/usr/bin/git-shell: 41M ./x86_64-linux/usr/libexec/git-core/git-daemon: 41M ./x86_64-linux/usr/libexec/git-core/git-http-backend: 41M ./x86_64-linux/usr/libexec/git-core/git-imap-send: 41M ./x86_64-linux/usr/libexec/git-core/git-shell: 41M ./x86_64-linux/usr/bin/git-upload-pack: 42M ./x86_64-linux/usr/libexec/git-core/git-fast-import: 42M ./x86_64-linux/usr/libexec/git-core/git-http-fetch: 42M ./x86_64-linux/usr/libexec/git-core/git-http-push: 42M ./x86_64-linux/usr/libexec/git-core/git-upload-pack: 42M ./x86_64-linux/usr/libexec/git-core/git-remote-ftp: 43M ./x86_64-linux/usr/libexec/git-core/git-remote-ftps: 43M ./x86_64-linux/usr/libexec/git-core/git-remote-http: 43M ./x86_64-linux/usr/libexec/git-core/git-remote-https: 43M ./x86_64-linux/usr/bin/git: 103M ./x86_64-linux/usr/bin/git-receive-pack: 103M ./x86_64-linux/usr/bin/git-upload-archive: 103M ./x86_64-linux/usr/libexec/git-core/git: 103M ./x86_64-linux/usr/libexec/git-core/git-add: 103M ./x86_64-linux/usr/libexec/git-core/git-annotate: 103M ./x86_64-linux/usr/libexec/git-core/git-apply: 103M ./x86_64-linux/usr/libexec/git-core/git-archive: 103M ./x86_64-linux/usr/libexec/git-core/git-bisect--helper: 103M ./x86_64-linux/usr/libexec/git-core/git-blame: 103M ./x86_64-linux/usr/libexec/git-core/git-branch: 103M ./x86_64-linux/usr/libexec/git-core/git-bundle: 103M ./x86_64-linux/usr/libexec/git-core/git-cat-file: 103M ./x86_64-linux/usr/libexec/git-core/git-check-attr: 103M ./x86_64-linux/usr/libexec/git-core/git-checkout: 103M ./x86_64-linux/usr/libexec/git-core/git-checkout-index: 103M ./x86_64-linux/usr/libexec/git-core/git-check-ref-format: 103M ./x86_64-linux/usr/libexec/git-core/git-cherry: 103M ./x86_64-linux/usr/libexec/git-core/git-cherry-pick: 103M ./x86_64-linux/usr/libexec/git-core/git-clean: 103M ./x86_64-linux/usr/libexec/git-core/git-clone: 103M ./x86_64-linux/usr/libexec/git-core/git-commit: 103M ./x86_64-linux/usr/libexec/git-core/git-commit-tree: 103M ./x86_64-linux/usr/libexec/git-core/git-config: 103M ./x86_64-linux/usr/libexec/git-core/git-count-objects: 103M ./x86_64-linux/usr/libexec/git-core/git-describe: 103M ./x86_64-linux/usr/libexec/git-core/git-diff: 103M ./x86_64-linux/usr/libexec/git-core/git-diff-files: 103M ./x86_64-linux/usr/libexec/git-core/git-diff-index: 103M ./x86_64-linux/usr/libexec/git-core/git-diff-tree: 103M ./x86_64-linux/usr/libexec/git-core/git-fast-export: 103M ./x86_64-linux/usr/libexec/git-core/git-fetch: 103M ./x86_64-linux/usr/libexec/git-core/git-fetch-pack: 103M ./x86_64-linux/usr/libexec/git-core/git-fmt-merge-msg: 103M ./x86_64-linux/usr/libexec/git-core/git-for-each-ref: 103M ./x86_64-linux/usr/libexec/git-core/git-format-patch: 103M ./x86_64-linux/usr/libexec/git-core/git-fsck: 103M ./x86_64-linux/usr/libexec/git-core/git-fsck-objects: 103M ./x86_64-linux/usr/libexec/git-core/git-gc: 103M ./x86_64-linux/usr/libexec/git-core/git-get-tar-commit-id: 103M ./x86_64-linux/usr/libexec/git-core/git-grep: 103M ./x86_64-linux/usr/libexec/git-core/git-hash-object: 103M ./x86_64-linux/usr/libexec/git-core/git-help: 103M ./x86_64-linux/usr/libexec/git-core/git-index-pack: 103M ./x86_64-linux/usr/libexec/git-core/git-init: 103M ./x86_64-linux/usr/libexec/git-core/git-init-db: 103M ./x86_64-linux/usr/libexec/git-core/git-log: 103M ./x86_64-linux/usr/libexec/git-core/git-ls-files: 103M ./x86_64-linux/usr/libexec/git-core/git-ls-remote: 103M ./x86_64-linux/usr/libexec/git-core/git-ls-tree: 103M ./x86_64-linux/usr/libexec/git-core/git-mailinfo: 103M ./x86_64-linux/usr/libexec/git-core/git-mailsplit: 103M ./x86_64-linux/usr/libexec/git-core/git-merge: 103M ./x86_64-linux/usr/libexec/git-core/git-merge-base: 103M ./x86_64-linux/usr/libexec/git-core/git-merge-file: 103M ./x86_64-linux/usr/libexec/git-core/git-merge-index: 103M ./x86_64-linux/usr/libexec/git-core/git-merge-ours: 103M ./x86_64-linux/usr/libexec/git-core/git-merge-recursive: 103M ./x86_64-linux/usr/libexec/git-core/git-merge-subtree: 103M ./x86_64-linux/usr/libexec/git-core/git-merge-tree: 103M ./x86_64-linux/usr/libexec/git-core/git-mktag: 103M ./x86_64-linux/usr/libexec/git-core/git-mktree: 103M ./x86_64-linux/usr/libexec/git-core/git-mv: 103M ./x86_64-linux/usr/libexec/git-core/git-name-rev: 103M ./x86_64-linux/usr/libexec/git-core/git-pack-objects: 103M ./x86_64-linux/usr/libexec/git-core/git-pack-redundant: 103M ./x86_64-linux/usr/libexec/git-core/git-pack-refs: 103M ./x86_64-linux/usr/libexec/git-core/git-patch-id: 103M ./x86_64-linux/usr/libexec/git-core/git-peek-remote: 103M ./x86_64-linux/usr/libexec/git-core/git-prune: 103M ./x86_64-linux/usr/libexec/git-core/git-prune-packed: 103M ./x86_64-linux/usr/libexec/git-core/git-push: 103M ./x86_64-linux/usr/libexec/git-core/git-read-tree: 103M ./x86_64-linux/usr/libexec/git-core/git-receive-pack: 103M ./x86_64-linux/usr/libexec/git-core/git-reflog: 103M ./x86_64-linux/usr/libexec/git-core/git-remote: 103M ./x86_64-linux/usr/libexec/git-core/git-replace: 103M ./x86_64-linux/usr/libexec/git-core/git-repo-config: 103M ./x86_64-linux/usr/libexec/git-core/git-rerere: 103M ./x86_64-linux/usr/libexec/git-core/git-reset: 103M ./x86_64-linux/usr/libexec/git-core/git-revert: 103M ./x86_64-linux/usr/libexec/git-core/git-rev-list: 103M ./x86_64-linux/usr/libexec/git-core/git-rev-parse: 103M ./x86_64-linux/usr/libexec/git-core/git-rm: 103M ./x86_64-linux/usr/libexec/git-core/git-send-pack: 103M ./x86_64-linux/usr/libexec/git-core/git-shortlog: 103M ./x86_64-linux/usr/libexec/git-core/git-show: 103M ./x86_64-linux/usr/libexec/git-core/git-show-branch: 103M ./x86_64-linux/usr/libexec/git-core/git-show-ref: 103M ./x86_64-linux/usr/libexec/git-core/git-stage: 103M ./x86_64-linux/usr/libexec/git-core/git-status: 103M ./x86_64-linux/usr/libexec/git-core/git-stripspace: 103M ./x86_64-linux/usr/libexec/git-core/git-symbolic-ref: 103M ./x86_64-linux/usr/libexec/git-core/git-tag: 103M ./x86_64-linux/usr/libexec/git-core/git-tar-tree: 103M ./x86_64-linux/usr/libexec/git-core/git-unpack-file: 103M ./x86_64-linux/usr/libexec/git-core/git-unpack-objects: 103M ./x86_64-linux/usr/libexec/git-core/git-update-index: 103M ./x86_64-linux/usr/libexec/git-core/git-update-ref: 103M ./x86_64-linux/usr/libexec/git-core/git-update-server-info: 103M ./x86_64-linux/usr/libexec/git-core/git-upload-archive: 103M ./x86_64-linux/usr/libexec/git-core/git-var: 103M ./x86_64-linux/usr/libexec/git-core/git-verify-pack: 103M ./x86_64-linux/usr/libexec/git-core/git-verify-tag: 103M ./x86_64-linux/usr/libexec/git-core/git-whatchanged: 103M ./x86_64-linux/usr/libexec/git-core/git-write-tree: 103M ./mipsel-oe-linux-uclibc/lib/libuClibc-0.9.32-git.so: 104M ./mipsel-oe-linux-uclibc/usr/lib/libc.a: 108M ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each 2010-07-16 23:03 ` Koen Kooi 2010-07-17 2:06 ` Khem Raj 2010-07-17 7:32 ` Khem Raj @ 2010-07-17 8:35 ` Paul Menzel 2010-07-17 12:15 ` Koen Kooi 2 siblings, 1 reply; 14+ messages in thread From: Paul Menzel @ 2010-07-17 8:35 UTC (permalink / raw) To: openembedded-devel [-- Attachment #1: Type: text/plain, Size: 1022 bytes --] Am Samstag, den 17.07.2010, 01:03 +0200 schrieb Koen Kooi: > On 16-07-10 17:23, Khem Raj wrote: > > On Fri, Jul 16, 2010 at 2:03 AM, Paul Menzel > > <paulepanter@users.sourceforge.net> wrote: > >> Am Donnerstag, den 15.07.2010, 19:55 +0200 schrieb Paul Menzel: > >> > >>> I ran out of space and noticed > >>> > >>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4 > >>> > >>> is 8,1G big. > >>> > >>> Some of the executable files in > >>> > >>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core/ > >>> > >>> are 101 MB big. > > > > its most probably the -g3/-ggdb3 bloat. > > No, it's git not creating (hard)links but copies of the same app. Run an > md5sum on the binaries, they will all match. I suggest you stop > spreading FUD about -ggdb3, it is not being appriciated. I do not know anything about the technical background. I can only report that removing `-ggdb3` fixed it for me. Thanks, Paul [-- Attachment #2: This is a digitally signed message part --] [-- Type: application/pgp-signature, Size: 205 bytes --] ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each 2010-07-17 8:35 ` Paul Menzel @ 2010-07-17 12:15 ` Koen Kooi 2010-07-18 14:48 ` Paul Menzel 0 siblings, 1 reply; 14+ messages in thread From: Koen Kooi @ 2010-07-17 12:15 UTC (permalink / raw) To: openembedded-devel -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 17-07-10 10:35, Paul Menzel wrote: > Am Samstag, den 17.07.2010, 01:03 +0200 schrieb Koen Kooi: >> On 16-07-10 17:23, Khem Raj wrote: >>> On Fri, Jul 16, 2010 at 2:03 AM, Paul Menzel >>> <paulepanter@users.sourceforge.net> wrote: > >>>> Am Donnerstag, den 15.07.2010, 19:55 +0200 schrieb Paul Menzel: >>>> >>>>> I ran out of space and noticed >>>>> >>>>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4 >>>>> >>>>> is 8,1G big. >>>>> >>>>> Some of the executable files in >>>>> >>>>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core/ >>>>> >>>>> are 101 MB big. >>> >>> its most probably the -g3/-ggdb3 bloat. >> >> No, it's git not creating (hard)links but copies of the same app. Run an >> md5sum on the binaries, they will all match. I suggest you stop >> spreading FUD about -ggdb3, it is not being appriciated. > > I do not know anything about the technical background. I can only report > that removing `-ggdb3` fixed it for me. Are you saying that removing -ggdb3 made git create proper (hard)links? regards. Koen -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (Darwin) iD8DBQFMQZ7PMkyGM64RGpERAilSAJ0Xh7q/84Ls1fu9SiqPiX4FcOS+9ACfRG/L 0sRjnr5z7xFxOVC7oSc1aG8= =MEay -----END PGP SIGNATURE----- ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each 2010-07-17 12:15 ` Koen Kooi @ 2010-07-18 14:48 ` Paul Menzel 2010-07-20 21:14 ` Paul Menzel 0 siblings, 1 reply; 14+ messages in thread From: Paul Menzel @ 2010-07-18 14:48 UTC (permalink / raw) To: openembedded-devel [-- Attachment #1: Type: text/plain, Size: 4958 bytes --] Am Samstag, den 17.07.2010, 14:15 +0200 schrieb Koen Kooi: > On 17-07-10 10:35, Paul Menzel wrote: > > Am Samstag, den 17.07.2010, 01:03 +0200 schrieb Koen Kooi: > >> On 16-07-10 17:23, Khem Raj wrote: > >>> On Fri, Jul 16, 2010 at 2:03 AM, Paul Menzel > >>> <paulepanter@users.sourceforge.net> wrote: > > > >>>> Am Donnerstag, den 15.07.2010, 19:55 +0200 schrieb Paul Menzel: > >>>> > >>>>> I ran out of space and noticed > >>>>> > >>>>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4 > >>>>> > >>>>> is 8,1G big. > >>>>> > >>>>> Some of the executable files in > >>>>> > >>>>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core/ > >>>>> > >>>>> are 101 MB big. > >>> > >>> its most probably the -g3/-ggdb3 bloat. > >> > >> No, it's git not creating (hard)links but copies of the same app. Run an > >> md5sum on the binaries, they will all match. I suggest you stop > >> spreading FUD about -ggdb3, it is not being appriciated. > > > > I do not know anything about the technical background. I can only report > > that removing `-ggdb3` fixed it for me. > > Are you saying that removing -ggdb3 made git create proper (hard)links? No, I just said, that after doing as Khem suggested the available space was big enough. Anyway, I tried to investigate a little and could not find anything related to (hard)link problems. But probably I overlooked something. I head refresh my memory about hard and soft links [1]. 1. As you suggested in your reply I ran `md5sum` on the binaries. They do indeed match on the binaries I posted, but also the inode matches. It looks like that is way it should be. $ echo "Test" > test $ ln test test.link $ md5sum test* 2205e48de5f93c784733ffcca841d2b5 test 2205e48de5f93c784733ffcca841d2b5 test.link $ ls -1i test* 3653430 test 3653430 test.link So I think the hardlinks were created properly. I did not think of this when reporting the problem so I did not run `ls -i` on the listed files and I did not check with `du`. /angstrom-dev/work/i686-linux/git-native-1.7.1.1-r0/image/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core $ md5sum * | sort […] 82c05218d9634923e301a9463e8f9c0d git-shell 85edbd7803242b4895ed33b4fcfa3622 git-remote-ftp 85edbd7803242b4895ed33b4fcfa3622 git-remote-ftps 85edbd7803242b4895ed33b4fcfa3622 git-remote-http 85edbd7803242b4895ed33b4fcfa3622 git-remote-https 89a6ef52f4b7c336fb5fc0601edfd95f git-mergetool 9a7f760ecef6e64f370c907d61ad0ffb git-svn 9b57eb61ce668f2105a61a467f4d6d9b git-difftool ac3472a81bc59a69119b0d027c3f617b git-merge-one-file b14a5bba57d0ae0ccfdb0fceac0eb4ec git-merge-octopus b6899828bc756c5d30398d49dce91deb git-cvsexportcommit b775de2d21923f5d08e0bdcf3f9d31f6 git-mergetool--lib bd786a897bdc214bd5d89f51ad79f152 git-instaweb c8466908a274e211a43d287e510f2f19 git-merge-resolve ce109bd27d9404980bd82ddb529eca90 git-send-email d3da496bd962246727ceaabc1d4b1bdd git-sh-setup d46c9b755b712bf3c27521b3c6e5fd30 git-relink d4b577d9a69d9af883de29d6af4f04c5 git d4b577d9a69d9af883de29d6af4f04c5 git-add d4b577d9a69d9af883de29d6af4f04c5 git-annotate d4b577d9a69d9af883de29d6af4f04c5 git-apply […] /angstrom-dev/work/i686-linux/git-native-1.7.1.1-r0/image/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core $ ls -1i [ … ordered manually to match md5sum output] 3691063 git-shell 3691070 git-remote-ftp 3691070 git-remote-ftps 3691070 git-remote-http 3691070 git-remote-https 3691082 git-mergetool 3691100 git-svn 3691093 git-difftool 3691080 git-merge-one-file 3691079 git-merge-octopus 3691095 git-cvsexportcommit 3691102 git-mergetool--lib 3691101 git-instaweb 3691081 git-merge-resolve 3691099 git-send-email 3691104 git-sh-setup 3691098 git-relink 3691105 git […] What did I miss? Were you seeing something different? 2. You seem to have investigated this problem before and committed a fix for git [2]. git 1.7.0.2: fix hardlinking problems as reported by Eric Brombaugh I could not find Eric’s report on this list or the Web. So I have to test without your patch. But `git-native` does not seem to use your patch, so I guess the behavior should be the one without your patch. Could you please tell me if my observations are correct. Thanks, Paul [1] info coreutils "ln invocation" [2] http://git.openembedded.org/cgit.cgi/openembedded/commit/?id=84de51631e1e7bb76141332807ed2ba9e8df02ae [-- Attachment #2: This is a digitally signed message part --] [-- Type: application/pgp-signature, Size: 205 bytes --] ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each 2010-07-18 14:48 ` Paul Menzel @ 2010-07-20 21:14 ` Paul Menzel 2010-07-20 22:44 ` Khem Raj 0 siblings, 1 reply; 14+ messages in thread From: Paul Menzel @ 2010-07-20 21:14 UTC (permalink / raw) To: openembedded-devel [-- Attachment #1: Type: text/plain, Size: 5376 bytes --] Am Sonntag, den 18.07.2010, 16:48 +0200 schrieb Paul Menzel: > Am Samstag, den 17.07.2010, 14:15 +0200 schrieb Koen Kooi: > > On 17-07-10 10:35, Paul Menzel wrote: > > > Am Samstag, den 17.07.2010, 01:03 +0200 schrieb Koen Kooi: > > >> On 16-07-10 17:23, Khem Raj wrote: > > >>> On Fri, Jul 16, 2010 at 2:03 AM, Paul Menzel > > >>> <paulepanter@users.sourceforge.net> wrote: > > > > > >>>> Am Donnerstag, den 15.07.2010, 19:55 +0200 schrieb Paul Menzel: > > >>>> > > >>>>> I ran out of space and noticed > > >>>>> > > >>>>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4 > > >>>>> > > >>>>> is 8,1G big. > > >>>>> > > >>>>> Some of the executable files in > > >>>>> > > >>>>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core/ > > >>>>> > > >>>>> are 101 MB big. > > >>> > > >>> its most probably the -g3/-ggdb3 bloat. > > >> > > >> No, it's git not creating (hard)links but copies of the same app. Run an > > >> md5sum on the binaries, they will all match. I suggest you stop > > >> spreading FUD about -ggdb3, it is not being appriciated. > > > > > > I do not know anything about the technical background. I can only report > > > that removing `-ggdb3` fixed it for me. > > > > Are you saying that removing -ggdb3 made git create proper (hard)links? > > No, I just said, that after doing as Khem suggested the available space > was big enough. > > Anyway, I tried to investigate a little and could not find anything > related to (hard)link problems. But probably I overlooked something. I > head refresh my memory about hard and soft links [1]. > > 1. As you suggested in your reply I ran `md5sum` on the binaries. They > do indeed match on the binaries I posted, but also the inode matches. It > looks like that is way it should be. > > $ echo "Test" > test > $ ln test test.link > $ md5sum test* > 2205e48de5f93c784733ffcca841d2b5 test > 2205e48de5f93c784733ffcca841d2b5 test.link > $ ls -1i test* > 3653430 test > 3653430 test.link > > So I think the hardlinks were created properly. I did not think of this > when reporting the problem so I did not run `ls -i` on the listed files > and I did not check with `du`. > > /angstrom-dev/work/i686-linux/git-native-1.7.1.1-r0/image/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core $ md5sum * | sort > […] > 82c05218d9634923e301a9463e8f9c0d git-shell > 85edbd7803242b4895ed33b4fcfa3622 git-remote-ftp > 85edbd7803242b4895ed33b4fcfa3622 git-remote-ftps > 85edbd7803242b4895ed33b4fcfa3622 git-remote-http > 85edbd7803242b4895ed33b4fcfa3622 git-remote-https > 89a6ef52f4b7c336fb5fc0601edfd95f git-mergetool > 9a7f760ecef6e64f370c907d61ad0ffb git-svn > 9b57eb61ce668f2105a61a467f4d6d9b git-difftool > ac3472a81bc59a69119b0d027c3f617b git-merge-one-file > b14a5bba57d0ae0ccfdb0fceac0eb4ec git-merge-octopus > b6899828bc756c5d30398d49dce91deb git-cvsexportcommit > b775de2d21923f5d08e0bdcf3f9d31f6 git-mergetool--lib > bd786a897bdc214bd5d89f51ad79f152 git-instaweb > c8466908a274e211a43d287e510f2f19 git-merge-resolve > ce109bd27d9404980bd82ddb529eca90 git-send-email > d3da496bd962246727ceaabc1d4b1bdd git-sh-setup > d46c9b755b712bf3c27521b3c6e5fd30 git-relink > d4b577d9a69d9af883de29d6af4f04c5 git > d4b577d9a69d9af883de29d6af4f04c5 git-add > d4b577d9a69d9af883de29d6af4f04c5 git-annotate > d4b577d9a69d9af883de29d6af4f04c5 git-apply > […] > > /angstrom-dev/work/i686-linux/git-native-1.7.1.1-r0/image/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core $ ls -1i > [ … ordered manually to match md5sum output] > 3691063 git-shell > 3691070 git-remote-ftp > 3691070 git-remote-ftps > 3691070 git-remote-http > 3691070 git-remote-https > 3691082 git-mergetool > 3691100 git-svn > 3691093 git-difftool > 3691080 git-merge-one-file > 3691079 git-merge-octopus > 3691095 git-cvsexportcommit > 3691102 git-mergetool--lib > 3691101 git-instaweb > 3691081 git-merge-resolve > 3691099 git-send-email > 3691104 git-sh-setup > 3691098 git-relink > 3691105 git > […] > > What did I miss? Were you seeing something different? > > 2. You seem to have investigated this problem before and committed a fix > for git [2]. > > git 1.7.0.2: fix hardlinking problems as reported by Eric Brombaugh > > I could not find Eric’s report on this list or the Web. So I have to > test without your patch. But `git-native` does not seem to use your > patch, so I guess the behavior should be the one without your patch. > > Could you please tell me if my observations are correct. I am a little bit lost. Koen, could you give me a pointer, please, so that I can try to continue to solve this issue. Thanks, Paul > [1] info coreutils "ln invocation" > [2] http://git.openembedded.org/cgit.cgi/openembedded/commit/?id=84de51631e1e7bb76141332807ed2ba9e8df02ae [-- Attachment #2: This is a digitally signed message part --] [-- Type: application/pgp-signature, Size: 205 bytes --] ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each 2010-07-20 21:14 ` Paul Menzel @ 2010-07-20 22:44 ` Khem Raj 2010-07-21 10:20 ` Paul Menzel 0 siblings, 1 reply; 14+ messages in thread From: Khem Raj @ 2010-07-20 22:44 UTC (permalink / raw) To: openembedded-devel On Tue, Jul 20, 2010 at 2:14 PM, Paul Menzel <paulepanter@users.sourceforge.net> wrote: > Am Sonntag, den 18.07.2010, 16:48 +0200 schrieb Paul Menzel: >> Am Samstag, den 17.07.2010, 14:15 +0200 schrieb Koen Kooi: >> > On 17-07-10 10:35, Paul Menzel wrote: >> > > Am Samstag, den 17.07.2010, 01:03 +0200 schrieb Koen Kooi: >> > >> On 16-07-10 17:23, Khem Raj wrote: >> > >>> On Fri, Jul 16, 2010 at 2:03 AM, Paul Menzel >> > >>> <paulepanter@users.sourceforge.net> wrote: >> > > >> > >>>> Am Donnerstag, den 15.07.2010, 19:55 +0200 schrieb Paul Menzel: >> > >>>> >> > >>>>> I ran out of space and noticed >> > >>>>> >> > >>>>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4 >> > >>>>> >> > >>>>> is 8,1G big. >> > >>>>> >> > >>>>> Some of the executable files in >> > >>>>> >> > >>>>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core/ >> > >>>>> >> > >>>>> are 101 MB big. >> > >>> >> > >>> its most probably the -g3/-ggdb3 bloat. >> > >> >> > >> No, it's git not creating (hard)links but copies of the same app. Run an >> > >> md5sum on the binaries, they will all match. I suggest you stop >> > >> spreading FUD about -ggdb3, it is not being appriciated. >> > > >> > > I do not know anything about the technical background. I can only report >> > > that removing `-ggdb3` fixed it for me. >> > >> > Are you saying that removing -ggdb3 made git create proper (hard)links? >> >> No, I just said, that after doing as Khem suggested the available space >> was big enough. >> >> Anyway, I tried to investigate a little and could not find anything >> related to (hard)link problems. But probably I overlooked something. I >> head refresh my memory about hard and soft links [1]. >> >> 1. As you suggested in your reply I ran `md5sum` on the binaries. They >> do indeed match on the binaries I posted, but also the inode matches. It >> looks like that is way it should be. >> >> $ echo "Test" > test >> $ ln test test.link >> $ md5sum test* >> 2205e48de5f93c784733ffcca841d2b5 test >> 2205e48de5f93c784733ffcca841d2b5 test.link >> $ ls -1i test* >> 3653430 test >> 3653430 test.link >> >> So I think the hardlinks were created properly. I did not think of this >> when reporting the problem so I did not run `ls -i` on the listed files >> and I did not check with `du`. >> >> /angstrom-dev/work/i686-linux/git-native-1.7.1.1-r0/image/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core $ md5sum * | sort >> […] >> 82c05218d9634923e301a9463e8f9c0d git-shell >> 85edbd7803242b4895ed33b4fcfa3622 git-remote-ftp >> 85edbd7803242b4895ed33b4fcfa3622 git-remote-ftps >> 85edbd7803242b4895ed33b4fcfa3622 git-remote-http >> 85edbd7803242b4895ed33b4fcfa3622 git-remote-https >> 89a6ef52f4b7c336fb5fc0601edfd95f git-mergetool >> 9a7f760ecef6e64f370c907d61ad0ffb git-svn >> 9b57eb61ce668f2105a61a467f4d6d9b git-difftool >> ac3472a81bc59a69119b0d027c3f617b git-merge-one-file >> b14a5bba57d0ae0ccfdb0fceac0eb4ec git-merge-octopus >> b6899828bc756c5d30398d49dce91deb git-cvsexportcommit >> b775de2d21923f5d08e0bdcf3f9d31f6 git-mergetool--lib >> bd786a897bdc214bd5d89f51ad79f152 git-instaweb >> c8466908a274e211a43d287e510f2f19 git-merge-resolve >> ce109bd27d9404980bd82ddb529eca90 git-send-email >> d3da496bd962246727ceaabc1d4b1bdd git-sh-setup >> d46c9b755b712bf3c27521b3c6e5fd30 git-relink >> d4b577d9a69d9af883de29d6af4f04c5 git >> d4b577d9a69d9af883de29d6af4f04c5 git-add >> d4b577d9a69d9af883de29d6af4f04c5 git-annotate >> d4b577d9a69d9af883de29d6af4f04c5 git-apply >> […] >> >> /angstrom-dev/work/i686-linux/git-native-1.7.1.1-r0/image/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core $ ls -1i >> [ … ordered manually to match md5sum output] >> 3691063 git-shell >> 3691070 git-remote-ftp >> 3691070 git-remote-ftps >> 3691070 git-remote-http >> 3691070 git-remote-https >> 3691082 git-mergetool >> 3691100 git-svn >> 3691093 git-difftool >> 3691080 git-merge-one-file >> 3691079 git-merge-octopus >> 3691095 git-cvsexportcommit >> 3691102 git-mergetool--lib >> 3691101 git-instaweb >> 3691081 git-merge-resolve >> 3691099 git-send-email >> 3691104 git-sh-setup >> 3691098 git-relink >> 3691105 git >> […] >> >> What did I miss? Were you seeing something different? >> >> 2. You seem to have investigated this problem before and committed a fix >> for git [2]. >> >> git 1.7.0.2: fix hardlinking problems as reported by Eric Brombaugh >> >> I could not find Eric’s report on this list or the Web. So I have to >> test without your patch. But `git-native` does not seem to use your >> patch, so I guess the behavior should be the one without your patch. >> >> Could you please tell me if my observations are correct. > > I am a little bit lost. Koen, could you give me a pointer, please, so > that I can try to continue to solve this issue. > Koen fixed it for target git recipe I guess you need to port that fix to git-native as well. if you want to use git-native built with OE. > > Thanks, > > Paul > > >> [1] info coreutils "ln invocation" >> [2] http://git.openembedded.org/cgit.cgi/openembedded/commit/?id=84de51631e1e7bb76141332807ed2ba9e8df02ae > > _______________________________________________ > Openembedded-devel mailing list > Openembedded-devel@lists.openembedded.org > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel > > ^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each 2010-07-20 22:44 ` Khem Raj @ 2010-07-21 10:20 ` Paul Menzel 0 siblings, 0 replies; 14+ messages in thread From: Paul Menzel @ 2010-07-21 10:20 UTC (permalink / raw) To: openembedded-devel [-- Attachment #1: Type: text/plain, Size: 6159 bytes --] Am Dienstag, den 20.07.2010, 15:44 -0700 schrieb Khem Raj: > On Tue, Jul 20, 2010 at 2:14 PM, Paul Menzel > <paulepanter@users.sourceforge.net> wrote: > > Am Sonntag, den 18.07.2010, 16:48 +0200 schrieb Paul Menzel: > >> Am Samstag, den 17.07.2010, 14:15 +0200 schrieb Koen Kooi: > >> > On 17-07-10 10:35, Paul Menzel wrote: > >> > > Am Samstag, den 17.07.2010, 01:03 +0200 schrieb Koen Kooi: > >> > >> On 16-07-10 17:23, Khem Raj wrote: > >> > >>> On Fri, Jul 16, 2010 at 2:03 AM, Paul Menzel > >> > >>> <paulepanter@users.sourceforge.net> wrote: > >> > > > >> > >>>> Am Donnerstag, den 15.07.2010, 19:55 +0200 schrieb Paul Menzel: > >> > >>>> > >> > >>>>> I ran out of space and noticed > >> > >>>>> > >> > >>>>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4 > >> > >>>>> > >> > >>>>> is 8,1G big. > >> > >>>>> > >> > >>>>> Some of the executable files in > >> > >>>>> > >> > >>>>> /angstrom-dev/work/i686-linux/git-native-1.7.0.2-r4/sysroot-destdir/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core/ > >> > >>>>> > >> > >>>>> are 101 MB big. > >> > >>> > >> > >>> its most probably the -g3/-ggdb3 bloat. > >> > >> > >> > >> No, it's git not creating (hard)links but copies of the same app. Run an > >> > >> md5sum on the binaries, they will all match. I suggest you stop > >> > >> spreading FUD about -ggdb3, it is not being appriciated. > >> > > > >> > > I do not know anything about the technical background. I can only report > >> > > that removing `-ggdb3` fixed it for me. > >> > > >> > Are you saying that removing -ggdb3 made git create proper (hard)links? > >> > >> No, I just said, that after doing as Khem suggested the available space > >> was big enough. > >> > >> Anyway, I tried to investigate a little and could not find anything > >> related to (hard)link problems. But probably I overlooked something. I > >> head refresh my memory about hard and soft links [1]. > >> > >> 1. As you suggested in your reply I ran `md5sum` on the binaries. They > >> do indeed match on the binaries I posted, but also the inode matches. It > >> looks like that is way it should be. > >> > >> $ echo "Test" > test > >> $ ln test test.link > >> $ md5sum test* > >> 2205e48de5f93c784733ffcca841d2b5 test > >> 2205e48de5f93c784733ffcca841d2b5 test.link > >> $ ls -1i test* > >> 3653430 test > >> 3653430 test.link > >> > >> So I think the hardlinks were created properly. I did not think of this > >> when reporting the problem so I did not run `ls -i` on the listed files > >> and I did not check with `du`. > >> > >> /angstrom-dev/work/i686-linux/git-native-1.7.1.1-r0/image/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core $ md5sum * | sort > >> […] > >> 82c05218d9634923e301a9463e8f9c0d git-shell > >> 85edbd7803242b4895ed33b4fcfa3622 git-remote-ftp > >> 85edbd7803242b4895ed33b4fcfa3622 git-remote-ftps > >> 85edbd7803242b4895ed33b4fcfa3622 git-remote-http > >> 85edbd7803242b4895ed33b4fcfa3622 git-remote-https > >> 89a6ef52f4b7c336fb5fc0601edfd95f git-mergetool > >> 9a7f760ecef6e64f370c907d61ad0ffb git-svn > >> 9b57eb61ce668f2105a61a467f4d6d9b git-difftool > >> ac3472a81bc59a69119b0d027c3f617b git-merge-one-file > >> b14a5bba57d0ae0ccfdb0fceac0eb4ec git-merge-octopus > >> b6899828bc756c5d30398d49dce91deb git-cvsexportcommit > >> b775de2d21923f5d08e0bdcf3f9d31f6 git-mergetool--lib > >> bd786a897bdc214bd5d89f51ad79f152 git-instaweb > >> c8466908a274e211a43d287e510f2f19 git-merge-resolve > >> ce109bd27d9404980bd82ddb529eca90 git-send-email > >> d3da496bd962246727ceaabc1d4b1bdd git-sh-setup > >> d46c9b755b712bf3c27521b3c6e5fd30 git-relink > >> d4b577d9a69d9af883de29d6af4f04c5 git > >> d4b577d9a69d9af883de29d6af4f04c5 git-add > >> d4b577d9a69d9af883de29d6af4f04c5 git-annotate > >> d4b577d9a69d9af883de29d6af4f04c5 git-apply > >> […] > >> > >> /angstrom-dev/work/i686-linux/git-native-1.7.1.1-r0/image/angstrom-dev/sysroots/i686-linux/usr/libexec/git-core $ ls -1i > >> [ … ordered manually to match md5sum output] > >> 3691063 git-shell > >> 3691070 git-remote-ftp > >> 3691070 git-remote-ftps > >> 3691070 git-remote-http > >> 3691070 git-remote-https > >> 3691082 git-mergetool > >> 3691100 git-svn > >> 3691093 git-difftool > >> 3691080 git-merge-one-file > >> 3691079 git-merge-octopus > >> 3691095 git-cvsexportcommit > >> 3691102 git-mergetool--lib > >> 3691101 git-instaweb > >> 3691081 git-merge-resolve > >> 3691099 git-send-email > >> 3691104 git-sh-setup > >> 3691098 git-relink > >> 3691105 git > >> […] > >> > >> What did I miss? Were you seeing something different? > >> > >> 2. You seem to have investigated this problem before and committed a fix > >> for git [2]. > >> > >> git 1.7.0.2: fix hardlinking problems as reported by Eric Brombaugh > >> > >> I could not find Eric’s report on this list or the Web. So I have to > >> test without your patch. But `git-native` does not seem to use your > >> patch, so I guess the behavior should be the one without your patch. > >> > >> Could you please tell me if my observations are correct. > > > > I am a little bit lost. Koen, could you give me a pointer, please, so > > that I can try to continue to solve this issue. > > Koen fixed it for target git recipe I guess you need to port that fix > to git-native as well. > if you want to use git-native built with OE. But as far as I see, the hardlinks were created correctly in git-native for me. So I do not know what I am supposed to do. Thanks, Paul > >> [1] info coreutils "ln invocation" > >> [2] http://git.openembedded.org/cgit.cgi/openembedded/commit/?id=84de51631e1e7bb76141332807ed2ba9e8df02ae [-- Attachment #2: This is a digitally signed message part --] [-- Type: application/pgp-signature, Size: 205 bytes --] ^ permalink raw reply [flat|nested] 14+ messages in thread
end of thread, other threads:[~2010-07-21 10:20 UTC | newest] Thread overview: 14+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2010-07-15 17:55 angstrom-2008.1, uclibc: git-native: sysroot-destdir 6, 4 GB big, since some executables are 100 MB each Paul Menzel 2010-07-16 9:03 ` Paul Menzel 2010-07-16 15:23 ` Khem Raj 2010-07-16 23:03 ` Koen Kooi 2010-07-17 2:06 ` Khem Raj 2010-07-17 12:17 ` Koen Kooi 2010-07-17 17:17 ` Phil Blundell 2010-07-17 7:32 ` Khem Raj 2010-07-17 8:35 ` Paul Menzel 2010-07-17 12:15 ` Koen Kooi 2010-07-18 14:48 ` Paul Menzel 2010-07-20 21:14 ` Paul Menzel 2010-07-20 22:44 ` Khem Raj 2010-07-21 10:20 ` Paul Menzel
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.