Bug#691115: unblock libdvdread/4.2.0+20120521-3
Hi, intrigeri wrote (27 Dec 2012 20:55:46 GMT) : (Trying to make Wheezy better.) Dmitry Smirnov wrote (15 Dec 2012 10:57:10 GMT) : I don't need to -- the original bug reporter (#688574) was not aware of similar cases. He had a perfect example of a problem and I just had a misfortune to discover another case of similar crash. Two questions here: 1. Do you think #688574 is serious enough to be worth fixing in Wheezy? (If so, perhaps the bug priority should be revisited. Also, given you've uploaded a new upstream version to unstable, a fix would probably have to go through t-p-u.) No news a month later, which wouldn't have happened, I guess, if Daniel thought #688574 was RC. So, I suggest the release team closes this unblock request. 2. Any comment on Adam's reasoning about how the fix is sub-optimal, and about his suggestions to improve it? (Related question: what's the upstream status of this patch, by the way?) I'm still curious about this, even if it's now probably unrelated to the matter at hand. Cheers, -- intrigeri | GnuPG key @ https://gaffer.ptitcanardnoir.org/intrigeri/intrigeri.asc | OTR fingerprint @ https://gaffer.ptitcanardnoir.org/intrigeri/otr.asc -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/85obgbabr2@boum.org
Bug#691115: unblock libdvdread/4.2.0+20120521-3
Hi Daniel! (Trying to make Wheezy better.) Dmitry Smirnov wrote (15 Dec 2012 10:57:10 GMT) : I don't need to -- the original bug reporter (#688574) was not aware of similar cases. He had a perfect example of a problem and I just had a misfortune to discover another case of similar crash. Two questions here: 1. Do you think #688574 is serious enough to be worth fixing in Wheezy? (If so, perhaps the bug priority should be revisited. Also, given you've uploaded a new upstream version to unstable, a fix would probably have to go through t-p-u.) 2. Any comment on Adam's reasoning about how the fix is sub-optimal, and about his suggestions to improve it? (Related question: what's the upstream status of this patch, by the way?) Cheers, -- intrigeri | GnuPG key @ https://gaffer.ptitcanardnoir.org/intrigeri/intrigeri.asc | OTR fingerprint @ https://gaffer.ptitcanardnoir.org/intrigeri/otr.asc -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/85sj6r1agd@boum.org
Bug#691115: unblock libdvdread/4.2.0+20120521-3
On Sat, 15 Dec 2012 07:37:50 intrigeri wrote: Waht would you do? If there was a bug I really wanted to see fixed in Wheezy, I would 1. talk to the maintainer and possibly 2. prepare an upload for t-p-u. Thanks, we'll see how it goes. Now I'd prefer to leave it with maintainer. The effects of this bug on a package that is not in testing is hardly relevant to the requested unblock. Please find a more relevant example to illustrate the case :) I don't need to -- the original bug reporter (#688574) was not aware of similar cases. He had a perfect example of a problem and I just had a misfortune to discover another case of similar crash. If two confirmed crashes is not enough for you, how many do you need? ;) If it's worth it, going through t-p-u might be an option. Understood thanks. IMHO Daniel will be the best person to take care of this. Regards, Dmitry. -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/201212152157.10736.only...@member.fsf.org
Bug#691115: unblock libdvdread/4.2.0+20120521-3
Hi, Dmitry Smirnov wrote (12 Dec 2012 22:40:05 GMT) : On Wed, 12 Dec 2012 21:30:14 intrigeri wrote: Dmitry Smirnov wrote (12 Dec 2012 01:16:15 GMT) : There were no reply from maintainer in #688574 so perhaps it would be better to set Daniel as owner of this bug... Please do it if you feel it's useful. Waht would you do? If there was a bug I really wanted to see fixed in Wheezy, I would 1. talk to the maintainer and possibly 2. prepare an upload for t-p-u. Given the crash fixed by 4.2.0+20120521-3 has severity normal, I'm unsure it's worth the effort. I'm not sure if normal is an adequate severity for crash. I've no idea how far the implications go, so I have no opinion on this matter. I'd like to hear the maintainer's opinion. Daniel, what do you think? (You might want to read #691115 first, to get some context.) For example handbrake (not in testing) was unusable (crashing on DVD open) with libdvdread prior to 4.2.0+20120521-3. The effects of this bug on a package that is not in testing is hardly relevant to the requested unblock. Please find a more relevant example to illustrate the case :) Dmitry, you filed the unblock request that is now outdated, what do you think? We can close it if you think that's the right thing to do. What else we can do? If it's worth it, going through t-p-u might be an option. Cheers, -- intrigeri | GnuPG key @ https://gaffer.ptitcanardnoir.org/intrigeri/intrigeri.asc | OTR fingerprint @ https://gaffer.ptitcanardnoir.org/intrigeri/otr.asc -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/85bodwqsip@boum.org
Bug#691115: unblock libdvdread/4.2.0+20120521-3
Hi, Dmitry Smirnov wrote (12 Dec 2012 01:16:15 GMT) : There were no reply from maintainer in #688574 so perhaps it would be better to set Daniel as owner of this bug... Please do it if you feel it's useful. A full new upstream version was uploaded to unstable since then, so an update in testing would now have to go through t-p-u. Given the crash fixed by 4.2.0+20120521-3 has severity normal, I'm unsure it's worth the effort. Dmitry, you filed the unblock request that is now outdated, what do you think? IMHO even if fix is not implemented properly it is still prevent certain crashes which can't be worse than what's in testing right now. I'm not sure this would be worse than what's in testing right now, but let's acknowledge that the fix has potential for future regressions, and does not only bring good: using internal implementation details of other libraries results in code that can break without notice, in the future, when the depended upon library is updated. Cheers, -- intrigeri | GnuPG key @ https://gaffer.ptitcanardnoir.org/intrigeri/intrigeri.asc | OTR fingerprint @ https://gaffer.ptitcanardnoir.org/intrigeri/otr.asc -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/85txrr8sw9@boum.org
Bug#691115: unblock libdvdread/4.2.0+20120521-3
On Wed, 12 Dec 2012 21:30:14 intrigeri wrote: Dmitry Smirnov wrote (12 Dec 2012 01:16:15 GMT) : There were no reply from maintainer in #688574 so perhaps it would be better to set Daniel as owner of this bug... Please do it if you feel it's useful. Waht would you do? A full new upstream version was uploaded to unstable since then, so an update in testing would now have to go through t-p-u. Thank you, I didn't notice that new libdvdread was uploaded. Given the crash fixed by 4.2.0+20120521-3 has severity normal, I'm unsure it's worth the effort. I'm not sure if normal is an adequate severity for crash. For example handbrake (not in testing) was unusable (crashing on DVD open) with libdvdread prior to 4.2.0+20120521-3. Dmitry, you filed the unblock request that is now outdated, what do you think? We can close it if you think that's the right thing to do. What else we can do? IMHO even if fix is not implemented properly it is still prevent certain crashes which can't be worse than what's in testing right now. I'm not sure this would be worse than what's in testing right now, but let's acknowledge that the fix has potential for future regressions, and does not only bring good: using internal implementation details of other libraries results in code that can break without notice, in the future, when the depended upon library is updated. I'm with you. Indeed that's choosing between bad and the worse. However decision making here is simple -- on one hand we have a certain (confirmed) crash when on the other we have potential for regressions that may or may not be manifested. To me crash is too serious to ignore because the whole functionality of the package is compromised. Regards, Dmitry. -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/201212130940.05955.only...@member.fsf.org
Bug#691115: unblock libdvdread/4.2.0+20120521-3
Hi, Adam D. Barratt wrote (21 Oct 2012 16:45:39 GMT) : On Mon, 2012-10-22 at 03:09 +1100, Dmitry Smirnov wrote: ++#define __USE_GNU My understanding was that code outside of the C library should never define __USE_GNU, as it's an internal implementation detail. The real problem appears to be that it's not getting implicitly defined via _GNU_SOURCE, because the existing definition is in the wrong place - it should be before the first #include. Ping? Cheers, -- intrigeri | GnuPG key @ https://gaffer.ptitcanardnoir.org/intrigeri/intrigeri.asc | OTR fingerprint @ https://gaffer.ptitcanardnoir.org/intrigeri/otr.asc -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/85vcc89j3c@boum.org
Bug#691115: unblock libdvdread/4.2.0+20120521-3
On Wed, 12 Dec 2012 12:04:23 intrigeri wrote: Ping? There were no reply from maintainer in #688574 so perhaps it would be better to set Daniel as owner of this bug... IMHO even if fix is not implemented properly it is still prevent certain crashes which can't be worse than what's in testing right now. Regards, Dmitry. -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/201212121216.15562.only...@member.fsf.org
Bug#691115: unblock libdvdread/4.2.0+20120521-3
Package: release.debian.org Severity: normal User: release.debian@packages.debian.org Usertags: unblock Please unblock package 'libdvdread4' unblock libdvdread/4.2.0+20120521-3 This update fix crashes (#688574). Full diff is attached. Thank you. Regards, Dmitry. diff --git a/debian/changelog b/debian/changelog index 261169d..aab071b 100644 --- a/debian/changelog +++ b/debian/changelog @@ -0,0 +1,7 @@ +libdvdread (4.2.0+20120521-3) unstable; urgency=low + + * Adding patch from Peter Van Eynde pvane...@debian.org to fix +miscompilation leading to a segfault (Closes: #688574). + + -- Daniel Baumann daniel.baum...@progress-technologies.net Mon, 24 Sep 2012 17:00:28 +0200 + diff --git a/debian/patches/09-segfault.patch b/debian/patches/09-segfault.patch new file mode 100644 index 000..82c16ea --- /dev/null +++ b/debian/patches/09-segfault.patch @@ -0,0 +1,14 @@ +Author: Peter Van Eynde pvane...@debian.org +Description: Fix miscompilation leading to a segfault (Closes: #688574). + +diff -Naurp libdvdread.orig/src/dvd_reader.c libdvdread/src/dvd_reader.c +--- libdvdread.orig/src/dvd_reader.c 2012-09-24 16:55:49.681245164 +0200 libdvdread/src/dvd_reader.c 2012-09-24 16:56:06.845356530 +0200 +@@ -30,6 +30,7 @@ + #include string.h + #include ctype.h + #define _GNU_SOURCE ++#define __USE_GNU + #include unistd.h + #include limits.h + #include dirent.h diff --git a/debian/patches/series b/debian/patches/series index 79e397c..2eb96d6 100644 --- a/debian/patches/series +++ b/debian/patches/series @@ -8,0 +9 @@ +09-segfault.patch signature.asc Description: This is a digitally signed message part.
Bug#691115: unblock libdvdread/4.2.0+20120521-3
Control: tags -1 + moreinfo On Mon, 2012-10-22 at 03:09 +1100, Dmitry Smirnov wrote: Please unblock package 'libdvdread4' unblock libdvdread/4.2.0+20120521-3 This update fix crashes (#688574). + #include string.h + #include ctype.h + #define _GNU_SOURCE ++#define __USE_GNU My understanding was that code outside of the C library should never define __USE_GNU, as it's an internal implementation detail. The real problem appears to be that it's not getting implicitly defined via _GNU_SOURCE, because the existing definition is in the wrong place - it should be before the first #include. Regards, Adam -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1350837939.8831.36.ca...@jacala.jungle.funky-badger.org
Processed: Re: Bug#691115: unblock libdvdread/4.2.0+20120521-3
Processing control commands: tags -1 + moreinfo Bug #691115 [release.debian.org] unblock libdvdread/4.2.0+20120521-3 Added tag(s) moreinfo. -- 691115: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=691115 Debian Bug Tracking System Contact ow...@bugs.debian.org with problems -- To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/handler.s.b691115.135083803326249.transcr...@bugs.debian.org
