Re: [xen-unstable test] 12828: regressions - FAIL

[Olaf Hering <olaf@aepfle.de>]

On Thu, May 10, Ian Campbell wrote:

> On Thu, 2012-05-10 at 14:01 +0100, Ian Campbell wrote:
> > On Thu, 2012-05-10 at 12:58 +0100, xen.org wrote:
> > > flight 12828 xen-unstable real [real]
> > > http://www.chiark.greenend.org.uk/~xensrcts/logs/12828/
> > > 
> > > Regressions :-(
> > > 
> > > Tests which did not succeed and are blocking,
> > > including tests which could not be run:
> > >  build-i386                    4 xen-build                 fail REGR. vs. 12827
> > 
> > gcc  -O1 -fno-omit-frame-pointer -m64 -g -fno-strict-aliasing -std=gnu99 -Wall -Wstrict-prototypes -Wdeclaration-after-statement   -D__XEN_TOOLS__ -MMD -MF .img2qcow.o.d  -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -fno-optimize-sibling-calls -Werror -g -Wno-unused -fno-strict-aliasing -I../include -I../drivers -I/home/osstest/build.12828.build-amd64/xen-unstable/tools/blktap2/drivers/../../../tools/libxc -I/home/osstest/build.12828.build-amd64/xen-unstable/tools/blktap2/drivers/../../../tools/include -D_GNU_SOURCE -DUSE_NFS_LOCKS -fPIC  -c -o img2qcow.o img2qcow.c 
> > gcc  -O1 -fno-omit-frame-pointer -m64 -g -fno-strict-aliasing -std=gnu99 -Wall -Wstrict-prototypes -Wdeclaration-after-statement   -D__XEN_TOOLS__ -MMD -MF .qcow2raw.o.d  -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -fno-optimize-sibling-calls -Werror -g -Wno-unused -fno-strict-aliasing -I../include -I../drivers -I/home/osstest/build.12828.build-amd64/xen-unstable/tools/blktap2/drivers/../../../tools/libxc -I/home/osstest/build.12828.build-amd64/xen-unstable/tools/blktap2/drivers/../../../tools/include -D_GNU_SOURCE -DUSE_NFS_LOCKS -fPIC  -c -o qcow2raw.o qcow2raw.c 
> > cc1: warnings being treated as errors
> > block-remus.c: In function 'ramdisk_flush':
> > block-remus.c:508: error: 'buf' may be used uninitialized in this function
> > make[5]: *** [block-remus.o] Error 1
> > make[5]: *** Waiting for unfinished jobs....
> > 
> > I presume this is 25289:27d63b9f111a "blktap2: Do not build with -O0" or
> > one of the followup patches?
> > 
> > Ah:
> > 
> > > version targeted for testing:
> > >  xen                  27d63b9f111a
> > 
> > So before all the fixups -- I suppose this will pass on the next run.
> 
> Keir says this isn't one he has fixed. I can't repro it either (and I
> run Squeeze, just like the build system) but by inspection this ought to
> solve it.

Looks like a compiler issue, since merge_requests() seems to correctly
return a non-null value when it does not touch the passed pointer to
buf. So your simple change looks like a correct workaround.


In general I think Makefiles should rather not mess with -W* and -O*
options like this, and use only the global flags if possible.


Olaf