Bug#284356: SONAME bumping and d-i
On Thu, Dec 23, 2004 at 05:45:50PM -0700, dann frazier wrote: $ ls -l /usr/src/kernel-patches/all/2.4.27/debian/series/2.4.27-7 -rw--- 1 root root 344 Dec 21 19:39 /usr/src/kernel-patches/all/2.4.27/debian/series/2.4.27-7 Are those perms correct? builds fine on ia64 otherwise
Bug#284356: SONAME bumping and d-i
On Wed, 2004-12-22 at 22:39 +0900, Horms wrote: [...] http://debian.vergenet.net/pending/ Andreas, let me know when you are ready to go with 2.6. Hre's Johnny! http://www.acm.rpi.edu/~dilinger/kernel-image-2.6.8-i386/ http://www.acm.rpi.edu/~dilinger/kernel-image-2.6.9-i386/ -- Andres Salomon [EMAIL PROTECTED] signature.asc Description: This is a digitally signed message part
Bug#284356: SONAME bumping and d-i
Horms wrote: Hi, I have put together the packages that I would like to upload as kernel-source-2.4.27 2.4.27-7 and kernel-image-2.4.27-i386 2.4.27-7. This includes increasing the SONAME to 2. Could anyone who is interesetd please take a look. Builds fine on mips. Thiemo
Bug#284356: SONAME bumping and d-i
On Wed, Dec 22, 2004 at 10:39:58PM +0900, Horms wrote: Hi, I have put together the packages that I would like to upload as kernel-source-2.4.27 2.4.27-7 and kernel-image-2.4.27-i386 2.4.27-7. This includes increasing the SONAME to 2. Could anyone who is interesetd please take a look. fyi, I noticed this in my build log: tar jxf /usr/src/kernel-source-2.4.27.tar.bz2 cd kernel-source-2.4.27; \ /usr/src/kernel-patches/all/2.4.27/apply/debian 2.4.27-6 tac: /usr/src/kernel-patches/all/2.4.27/debian/series/2.4.27-7: Permission denied -- 2.4.27-7 fully unapplied. $ ls -l /usr/src/kernel-patches/all/2.4.27/debian/series/2.4.27-7 -rw--- 1 root root 344 Dec 21 19:39 /usr/src/kernel-patches/all/2.4.27/debian/series/2.4.27-7 Are those perms correct?
Bug#284356: SONAME bumping and d-i
On Mon, Dec 20, 2004 at 12:22:46AM -0500, Andres Salomon wrote: On Mon, 2004-12-20 at 12:23 +0900, Horms wrote: On Fri, Dec 17, 2004 at 02:32:13PM -0500, Andres Salomon wrote: On Fri, 17 Dec 2004 12:22:25 +0900, Horms wrote: [...] A sounds fine from my point of view, as it involves the least effort. Although as I discovered this morning, there seems to be a second ABI change relating to the same patch, so we should get that out ASAP. http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=284356#msg68 Hm, you're referring to the addition of termios_sem (which was already added to 2.6), or something else? Yes, that is the one. Joeyh doesn't seem to care which we do (he said it's the same amount of effort for him both ways, he just asked that whatever we decide be consistent for 2.4 and 2.6. So, I'm going to recommend reverting the tty locking patch and bumping the SONAME. For 2.4, this makes sense since the kernel w/ the broken ABI hasn't propogated to sarge yet. For 2.6, this is a bit more painful, since the broken ABI is already in sarge, but.. *shrug*. Andres and I discussed this on IRC and as the SONAME=2 packages are produced by the same source package as the SONAME=1 packages once the SONAME=2 are accepted the SONAME=1 will go away. In a nutshell this means the update SONAME=1 then upload SONAME=2 approach doesn't help any more than just uploading SONAME=2. So we have more or less decided to do that for both 2.4 and 2.6 and we are going to work on making that happen. -- Horms
Bug#284356: SONAME bumping and d-i
On Fri, Dec 17, 2004 at 02:32:13PM -0500, Andres Salomon wrote: On Fri, 17 Dec 2004 12:22:25 +0900, Horms wrote: [...] A sounds fine from my point of view, as it involves the least effort. Although as I discovered this morning, there seems to be a second ABI change relating to the same patch, so we should get that out ASAP. http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=284356#msg68 Hm, you're referring to the addition of termios_sem (which was already added to 2.6), or something else? Yes, that is the one. -- Horms
Bug#284356: SONAME bumping and d-i
On Mon, 2004-12-20 at 12:23 +0900, Horms wrote: On Fri, Dec 17, 2004 at 02:32:13PM -0500, Andres Salomon wrote: On Fri, 17 Dec 2004 12:22:25 +0900, Horms wrote: [...] A sounds fine from my point of view, as it involves the least effort. Although as I discovered this morning, there seems to be a second ABI change relating to the same patch, so we should get that out ASAP. http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=284356#msg68 Hm, you're referring to the addition of termios_sem (which was already added to 2.6), or something else? Yes, that is the one. Joeyh doesn't seem to care which we do (he said it's the same amount of effort for him both ways, he just asked that whatever we decide be consistent for 2.4 and 2.6. So, I'm going to recommend reverting the tty locking patch and bumping the SONAME. For 2.4, this makes sense since the kernel w/ the broken ABI hasn't propogated to sarge yet. For 2.6, this is a bit more painful, since the broken ABI is already in sarge, but.. *shrug*. -- Andres Salomon [EMAIL PROTECTED] signature.asc Description: This is a digitally signed message part
Bug#284356: SONAME bumping and d-i
On Fri, 17 Dec 2004 12:22:25 +0900, Horms wrote: [...] A sounds fine from my point of view, as it involves the least effort. Although as I discovered this morning, there seems to be a second ABI change relating to the same patch, so we should get that out ASAP. http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=284356#msg68 Hm, you're referring to the addition of termios_sem (which was already added to 2.6), or something else?
Bug#284356: SONAME bumping and d-i
Hi Joey, On Thu, Dec 16, 2004 at 06:33:31PM -0500, Joey Hess wrote: Looking at how these proposed fixes would affect d-i and existing rc2 images: a. If the SONAME is left unchanged and the new ABI remains, and things are updated to use the new ABI: - Installs from a rc2 netinst CD will keep working, but you'll get a kernel with the old ABI. Installs of third-party (ie, alsa) kernel modules that use the new ABI will then fail. Until you upgrade your kernel.. - Installs from a rc2 businesscard CD will be ok. - The rc2 netboot and floppy images will stop working once udebs built with the new ABI reach sarge. - We won't need to make any other changes to d-i aside from putting those udebs in sarge and rebuilding the d-i images. - However we won't be able to easily/widely test netboot or floppy installs using the new ABI kernels until the udebs reach sarge. The daily builds would need to be hacked to pull udebs from sid to do any significant testing. b. If the SONAME is increased and the ABI changes reverted for -1: It is my understading that releasing -2 packages would result in the -1 packages being removed from the debian archive as regardless of weather the binary packages are -1, -2, or -X, the source package is infact kernel-image-2.4.27-arch. Are you proposing that we should add a new source package, kernel-image-2.4.27-arch-2 which produces SONAME=2 binary packages and leave kernel-image-2.4.27-arch providing the SONAME=1 packages. Or are you suggesting that the -1 binary packages will just hang around in the archive for a while? - All rc2 images will keep working until/unless the udebs from the -1 kernels are removed from sarge, when the netboot and floppy images will break. - We'll need to build new udebs for the -2 kernels, while keeping the udebs from the -1 kernels. This will either mean some ugly linux-kernel-di packages that build both from one package, or the even more ugly splitting off of a linux-kernel-di-i386-2 for the -2 kernels. Nasty nasty nasty. - Changes will be needed in debian-cd to drop the -1 stuff from CDs to avoid space issues. - At the moment I think that d-i/anna will do the right thing WRT using the -2 udebs if the -2 kernel is running. However we've never been in this situation before, so something could fail. - Changes will be needed in rootskel to install the -2 kernel images. Some arches may also need base-installer changes. - Not sure how this affects third party modules in debian, do they have to build modules for both kernel SONAMEs? Do they drop -1 debs? If so rc2 gets subtly broken. - Simply increasing the number of udebs in sarge with -2 kernel udebs will change d-i's memory usage, which could break 20 mb installs. I think we have a big gap before we need to worry about 32 mb installs. Still we'd need to retest everything for lowmem again, or remove the -1 kernel udebs before the next d-i release. I keep seeming to come up with new issues with approach b. Leaning toward a.. A sounds fine from my point of view, as it involves the least effort. Although as I discovered this morning, there seems to be a second ABI change relating to the same patch, so we should get that out ASAP. http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=284356#msg68 -- Horms
