Bugzilla Bug 69512

On ppc64 If 010_all_long2int.patch is applied, it cannot decode normally.
Please do not apply on ppc64.

Note:
010_all_long2int.patch is contained in mpg123-0.59s-gentoo-1.1.tar.bz2.


Reproducible: Always
Steps to Reproduce:

1. ACCEPT_KEYWORDS=~ppc64 emerge mpg123
2. mpg123 -w test.wav test.mp3
3. play a wav file.

Created an attachment (id=42914) [edit]
patch for mpg123-0.59s-r6.ebuild


mpg123-0.59s-r6 has the problem which plays only a noise.
The appended patch fixes the problem.

emerge info output:

Portage 2.0.50-r11 (default-ppc64-2004.2, gcc-3.4.1, glibc-2.3.4.20040808-r1, 2.6.8-gentoo-r3)
=================================================================
System uname: 2.6.8-gentoo-r3 ppc64 PPC970, altivec supported
Gentoo Base System version 1.4.16
Autoconf: sys-devel/autoconf-2.59-r5
Automake: sys-devel/automake-1.8.5-r1
ACCEPT_KEYWORDS="ppc64"
AUTOCLEAN="yes"
CFLAGS="-O2 -pipe"
CHOST="powerpc64-unknown-linux-gnu"
COMPILER=""
CONFIG_PROTECT="/etc /usr/X11R6/lib/X11/xkb /usr/kde/2/share/config /usr/kde/3.3/env /usr/kde/3.3/share/config /usr/kde/3.3/shutdown /usr/kde/3/share/config /usr/share/config /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi/ /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-O2 -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoaddcvs buildpkg ccache noclean sandbox"
GENTOO_MIRRORS="http://mirror.gentoo.gr.jp/"
MAKEOPTS="-j5"
PKGDIR="/var/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/portage"
SYNC="rsync://rsync.jp.gentoo.org/gentoo-portage"
USE="X Xaw3d aalib alsa altivec apache2 arts berkdb bitmap-fonts cdr cjk crypt cups dga dvd dvdr emacs encode esd f77 flash foomaticdb gdbm gif gnome gpm gtk gtk2 imap imlib joystick jpeg kde libwww mad mitshm motif mozilla mpeg mule ncurses nls nogcj oggvorbis opengl oss pam pdflib perl png ppc64 python qt quicktime readline samba sdl slang ssl tcpd tetex tiff truetype usb xml xml2 xmms xprint xv zlib"

Created an attachment (id=43167) [edit]
patch for mpg123-0.59s-r6.ebuild

mpg123-0.59s-r6 has the problem which plays only a noise.
The appended patch fixes the problem.
And in the case of PPC64, 010_all_long2int.patch is delete. 

Use EPATCH_EXCLUDE rather than removing the patch... but this is not the proper
correction...  that patch is correct, but if it causes you problems, then it
means that there are more log/int bugs in the source that need to be fixed.

I wonder if this is related to bug #56765.  Can youu try different output
options to see if this is limited to just a couple please?

Thank you for your hint.
I created the new patch to source.

I think that "mpg123 -s hoge.mp3" is working normally on PPC64.
Because, PPC64 is the architecture of big endian and the output raw data is also byte order of big endian.
(mpg123 is output by the host byte order. Please see man if necessary.)


The procedure which I checked is as follows.

1. mpg123 -s test.mp3 > mpg123.raw
2. oggenc -r --raw-endianness 1 mpg123.raw
3. ogg123 mpg123.ogg

Also, mpg321 seems output by the byte order of little endian.

1. mpg321 -s test.mp3 > mpg321.raw
2. oggenc -r --raw-endianness 0 mpg321.raw
3. ogg123 mpg321.ogg


Note:
About option raw-endianness of oggenc...
When set to 1 : big endian
When set to 0 : little endian

Please see man for details.


Thank you.

Created an attachment (id=43206) [edit]
patch for mpg123-0.59s-r6.ebuild

Created an attachment (id=43207) [edit]
patch for mpg123 on ppc64

A file name is mpg123-0.59s-ppc64.patch .

ok, I'll take a look at that on my archs (x86, amd64, sparc)... well... just
amd64 as it changes nothing on the other two...

commited that patch.

@eradicator: Does this also affect amd64? If yes and you fixed that, please close.

Markus

fixed in cvs.

corsair, please make sure to test an ebuild before committing.  Your patch name didn't match what was being used by the ebuild  (you committed ${PF}-... whereas youu epatched ${P}-...)

I revbumped instead and kept it stable on amd64 and ppc64.

eradicator: 
Although I tried the new version -r7, the following errors occurred.
Would you upload the new patch?

Thank you.


>>> Downloading http://dev.gentoo.org/~eradicator/mpg123/mpg123-0.59s-gentoo-1.2.tar.bz2
--16:27:44--  http://dev.gentoo.org/%7Eeradicator/mpg123/mpg123-0.59s-gentoo-1.2.tar.bz2
           => `/usr/portage/distfiles/mpg123-0.59s-gentoo-1.2.tar.bz2'
Resolving dev.gentoo.org... 156.56.111.197
Connecting to dev.gentoo.org[156.56.111.197]:80... connected.
HTTP request sent, awaiting response... 404 Not Found
16:27:44 ERROR 404: Not Found.

!!! Couldn't download mpg123-0.59s-gentoo-1.2.tar.bz2. Aborting.

damn.. I tested it, but I think didn't copied the ebuild (to update it in CVS)
or that is to say I modified it once more.. :-/ sry..

Have to leave now.. school. I'll fix that this afternoon.

Markus

yuta: sorry about that... should be there now.

eradicator: 
Thank you!
I tried the new version and it confirmed that these problems did not occur.

So this bug is finaly closed?

Markus

yes

The problem of the sound in a wide range package was solved by Bug #71119 on
PPC64.
Consequently, the necessity of using the code for big endian contained in this
package itself was lost.

And the problem has occurred again now owing to the code.

Created an attachment (id=43923) [edit]
patch for 001_all_Makefile.patch

This patch solves the above-mentioned problem.


eradicator:
I am sorry to trouble you, but could you apply this patch?

fixed in cvs.

closing