On Fri, 18 Nov 2011 13:17:49 +0100 Herbert Poetzl <herbert@13thfloor.at> wrote: > On Thu, Nov 17, 2011 at 11:46:41PM -0600, Corey Wright wrote: > > On Fri, 18 Nov 2011 03:32:33 +0100 > > Herbert Poetzl <herbert@13thfloor.at> wrote: > > >> On Thu, Nov 17, 2011 at 03:13:18PM +0100, Herbert Poetzl wrote: > >>> On Thu, Nov 17, 2011 at 02:24:19PM +0100, Sebastian Steinhuber wrote: > >>>> Hi! > >>>> I committed patch-3.0.9-vs2.3.2 to vanilla kernel. The sublevel > >>>> changed form 9 to 7 afterwards because of line 167: > >>>> +SUBLEVEL = 7 > > >>> yes, thanks for reporting, my scripts are obviously > >>> not used to the 3.x kernels yet, i.e. I need to do some > >>> adjustments there ... > > >>>> Btw, similar issue with patch-3.1.1-vs2.3.2 leading to a > >>>> 3.1.11-vs2.3.2 numbering: > >>>> line 167: +EXTRAVERSION = 1-vs2.3.2 > > >>> I updated the patches in place with a fixed version > >>> for now > > >>>> But the compilation of 3.0.9-vs2.3.2 stopped with errors. The full > >>>> output of 'make all' is appended to the message. > > >>>> include/linux/vserver/limit_int.h: In function ???__vx_acc_cres???: > >>>> include/linux/vserver/limit_int.h:16:1: error: ???vs_debug_cres??? > >>>> undeclared (first use in this function) > >>>> include/linux/vserver/limit_int.h:16:1: note: each undeclared identifier > >>>> is reported only once for each function it appears in > >>>> include/linux/vserver/limit_int.h: In function ???__vx_add_cres???: > >>>> include/linux/vserver/limit_int.h:33:1: error: ???vs_debug_cres??? > >>>> undeclared (first use in this function) > >>>> include/linux/vserver/limit_int.h: In function ???__vx_cres_avail???: > >>>> include/linux/vserver/limit_int.h:84:1: error: ???vs_debug_limit??? > >>>> undeclared (first use in this function) > >>>> make[2]: *** [arch/x86/mm/init.o] Fehler 1 > >>>> make[1]: *** [arch/x86/mm] Fehler 2 > >>>> make: *** [arch/x86] Fehler 2 > > >>>> I'm guessing that there might be a bug. I'm on x86-64, > >>>> debian/unstable with patches from > >>>> http://vserver.13thfloor.at/Experimental/. > >>>> Any hints? Thanks in advance. > > >>> could you upload your .config file somewhere and > >>> provide the url (via ML or on irc) please? > > >> issue should be fixed in vs2.3.2.1, no need to upload > >> your .config anymore ... > > > but the "SUBLEVEL = 7" in Makefile is back. i noticed it when > > diffing my .config against my .config.old after doing a "make > > oldconfig". the attached 3.0.9 patch fixes it. > > yep, I obviously forgot to change that for real ... > > > the "'vs_debug_*' undeclared" compilation errors are gone, but > > they are now in patch-3.1.1-vs2.3.2.1.diff. see the attached > > build log for the exact compilation errors and the attached > > 3.1.1 patch that works around them (identical to my 3.0.9 patch > > from yesterday). > > correct, 3.1.1 is missing this patch: > http://vserver.13thfloor.at/ExperimentalT/delta-debug-fix01.diff > > I'll once again updated both in place after fixing > my build scripts and adding the missing patch ... > > > I have not had a chance to test the resulting kernel. > > please let us know how it performs once you find > the time ... both kernels (3.0.9-vs2.3.2.1 & 3.1.1-vs2.3.2.1) compiled and executed (testme.sh & testfs.sh) without error. well testfs.sh failed on tests 203, 213, & 231 (quotas) for ext4, but: 1. i don't (yet) use ext4 1. i don't (currently) use quotas 2. i figure it's a known issue as it goes back to at least 2.6.35 corey -- undefined@pobox.com > > thanks for your continued work on the linux 3.x patches. > > you're welcome and thanks in advance for testing! > > best, > Herbert > > > corey > > -- > > undefined@pobox.com > > >> have fun, > >> Herbert > > >>> thanks in advance, > >>> Herbert > > >>>> Best regards, > >>>> Sebastian