Re: mips/mipsel in samtools and pysam
Hi Afif, regarding bug #749367 of htslib I inspected the upstream discussion about the proposed patch[4]. I need to admit that I can not make any safe conclusion whether the mips issue is solved or not. The fact that the build problem remains (OK, according to #749367 there were 5 failures in the beginning the build log [3] now has only 4 failures) seems to me that upstream does not found a final solution. However, this is not release critical since htslib was never available for mips. However, as far as I can see there in tracker[5] are more important problems in samtools affecting several architectures. According to Charles[6] this might be fixed upstream in the next release. I'm not sure how promissing the suggested action (=waiting for upstream) might be. Does anybody have contact to upstream and would find out whether a new release can be expected soon or whether we should try to backport the changes (and would volunteer to do the backporting ;-)). Kind regards Andreas. On Mon, Jul 27, 2015 at 11:30:37PM -0700, Afif Elghraoui wrote: The build logs for the current versions of samtools and pysam (without contributed patches provided for the older release) show that building was successful on mipsel [1-2]. Building on mips has not been attempted for these two packages since htslib has apparently failed to [3] build there. Perhaps mips will work as well if htslib becomes available there. At least for pysam, the provided patches no longer apply cleanly, but maybe they will not be needed anymore.. Regards Afif 1. https://buildd.debian.org/status/package.php?p=python-pysam 2. https://buildd.debian.org/status/package.php?p=samtoolssuite=sid 3. https://buildd.debian.org/status/logs.php?pkg=htslibver=1.2.1-1arch=mips [4] https://github.com/samtools/htslib/pull/99 [5] https://tracker.debian.org/pkg/samtools [6] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=762647#39 -- http://fam-tille.de -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150728105243.gq16...@an3as.eu
Re: mips/mipsel in samtools and pysam
Hi, Andreas, The build logs for the current versions of samtools and pysam (without contributed patches provided for the older release) show that building was successful on mipsel [1-2]. Building on mips has not been attempted for these two packages since htslib has apparently failed to [3] build there. Perhaps mips will work as well if htslib becomes available there. At least for pysam, the provided patches no longer apply cleanly, but maybe they will not be needed anymore.. Regards Afif 1. https://buildd.debian.org/status/package.php?p=python-pysam 2. https://buildd.debian.org/status/package.php?p=samtoolssuite=sid 3. https://buildd.debian.org/status/logs.php?pkg=htslibver=1.2.1-1arch=mips -- Afif Elghraoui | عفيف الغراوي http://afif.ghraoui.name -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/55b7218d.9090...@ghraoui.name
Re: mips/mipsel in samtools and pysam
On Wed, Jul 29, 2015 at 05:24:38AM +0900, Charles Plessy wrote: Why not just patiently waiting for the next upstream release ? The htlslib packages work very well on amd64, and this is where our work is used. Yes, but reverse dependencies are currently removed from testing. Bringing them back to testing is my main motivation. Thus specifying a positive list for samtools that should be revised for next upstream seems to be the most sensible thing to do for me if the effort of backporting is to high. An please do not take this comment negatively: if people can do more that is great, but at the moment, with my reduced free time and an intensely stressful job, I am not even in a position to participate to such an effort occasionally. Perfectly understood and no offense taken. Kind regards Andreas. -- http://fam-tille.de -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150728210312.ge16...@an3as.eu
Re: mips/mipsel in samtools and pysam
Le Tue, Jul 28, 2015 at 12:52:43PM +0200, Andreas Tille a écrit : However, as far as I can see there in tracker[5] are more important problems in samtools affecting several architectures. According to Charles[6] this might be fixed upstream in the next release. I'm not sure how promissing the suggested action (=waiting for upstream) might be. Does anybody have contact to upstream and would find out whether a new release can be expected soon or whether we should try to backport the changes (and would volunteer to do the backporting ;-)). Hi Andreas, life is short, time is precisous, and frankly speaking, MIPS, hurd, hppa, etc. are irrelevant to bioinformatics. Why not just patiently waiting for the next upstream release ? The htlslib packages work very well on amd64, and this is where our work is used. On my side, I do not have time and patience to deal with these porting issues with more than a best effort priority, because there are no users for htlslib on these platforms. If there are volunteers to backport changes in order to go faster than Upstream, please do so but please also remove my name from the Uploaders field. On my side I have no time for this wokflow that multiplies patch management and uploads. An please do not take this comment negatively: if people can do more that is great, but at the moment, with my reduced free time and an intensely stressful job, I am not even in a position to participate to such an effort occasionally. Making bioinformatics viable on alternative architectures is a huge task that is only doable with full dedication. Actually, this is really the kind of task that I think should be only done if hardware vendors or granting agencies would provide a strong support. I am not seeing myself as a pioneer to develop a new market for manufacturers, I am a researcher sharing the tools of today on the machines of today. Have a nice day, Charles -- Charles Plessy Debian Med packaging team, http://www.debian.org/devel/debian-med Tsurumi, Kanagawa, Japan -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150728202354.gb19...@falafel.plessy.net
Re: mips/mipsel in samtools and pysam (was: finalizing pysam 0.8.3)
Hi, Andreas, On الجمعـة 24 تـمـوز 2015 22:06, Andreas Tille wrote: I just notoced that we did not cared about the mips build problem (#753490). Would you check whether the patches apply also to this version and forward them upstream. I did not forget about it and was planning to post an update yesterday to the bug log (CC'd bug report now). This bug requires #753485 (in samtools) to be fixed first. I don't have a chance to look into samtools as of yet. I was also hesitant to apply patches to pysam until I can make sure it passes all its tests. Since the current failures look like python2/python3 transition issues, I think any new failures should stand out, so I will review these pysam patches. Thanks for all your support Regards Afif -- Afif Elghraoui | عفيف الغراوي http://afif.ghraoui.name -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/55b3de4b.9000...@ghraoui.name
