On 06/08/2018 02:52 PM, Philippe Mathieu-Daudé wrote: > On 06/04/2018 07:28 AM, Peter Maydell wrote: >> On 1 June 2018 at 17:26, Michael S. Tsirkin <m...@redhat.com> wrote: >>> The following changes since commit 63b88968f139b6a77f2f81e6f1eedf70c0170a85: >>> >>> intel-iommu: rework the page walk logic (2018-05-23 17:34:05 +0300) >>> >>> are available in the Git repository at: >>> >>> git://git.kernel.org/pub/scm/virt/kvm/mst/qemu.git tags/for_upstream >>> >>> for you to fetch changes up to 25b1d45a1975fd8624c37b5bf42e8502ccf53460: >>> >>> vhost-blk: turn on pre-defined RO feature bit (2018-06-01 19:20:38 +0300) >>> >>> ---------------------------------------------------------------- >>> acpi, vhost, misc: fixes, features >>> >>> vDPA support, fix to vhost blk RO bit handling, some include path >>> cleanups, NFIT ACPI table. >>> >>> Signed-off-by: Michael S. Tsirkin <m...@redhat.com> >>> >>> ---------------------------------------------------------------- >> >> Applied, thanks. > > Not sure this is a problem, but something experienced after rebasing: > > > #1 > > /source/qemu/qemu-nbd.c: In function ‘version’: > /source/qemu/qemu-nbd.c:133:6: error: expected ‘)’ before > ‘QEMU_FULL_VERSION’ > "%s " QEMU_FULL_VERSION "\n" > ^~~~~~~~~~~~~~~~~~ > ) > /source/qemu/qemu-nbd.c:133:3: error: format ‘%s’ expects a matching > ‘char *’ argument [-Werror=format=] > "%s " QEMU_FULL_VERSION "\n" > ~^ > cc1: all warnings being treated as errors > > It took me a while to figure this out, probably because today is Friday. > > $ git status --ignored show me I had a "qemu-version.h" generated long > time ago in the source directory. > > - I only do out-of-tree builds. > > - sometimes I run "make help/docker/vm-test" and friends in the source > directory, and my 'make' is aliased as 'make -jN'. > I doubt those make calls generate this file, as ./configure generates > it, so I probably messed with a ./configure in source dir. > > Now, building out-of-tree dirs was finding the old /qemu-version.h > before the generated one in current dir. > > I was not noticing it due to the /qemu-version.h entry in /.gitignore . > > > #2 > > CC vl.o > /source/qemu/vl.c: In function ‘main’: > /source/qemu/vl.c:3052:18: error: ‘QEMU_OPTION_blockdev’ undeclared > (first use in this function); did you mean ‘QEMU_OPTION_clock’? > case QEMU_OPTION_blockdev: > ^~~~~~~~~~~~~~~~~~~~ > QEMU_OPTION_clock > /source/qemu/vl.c:3052:18: note: each undeclared identifier is reported > only once for each function it appears in > /source/qemu/vl.c:3171:18: error: ‘QEMU_OPTION_nic’ undeclared (first > use in this function); did you mean ‘QEMU_OPTION_rtc’? > case QEMU_OPTION_nic: > ^~~~~~~~~~~~~~~ > QEMU_OPTION_rtc > /source/qemu/vl.c:3574:18: error: ‘QEMU_OPTION_preconfig’ undeclared > (first use in this function); did you mean ‘QEMU_OPTION_writeconfig’? > case QEMU_OPTION_preconfig: > ^~~~~~~~~~~~~~~~~~~~~ > QEMU_OPTION_writeconfig > > This is the same story, but with /qemu-options.def (also in .gitignore). > > > So 'rm ${srcdir}/qemu-version.h ${srcdir}/qemu-options.def' fixed my > problems. > > Should we add a check for this kind of mistakes when running out-of-tree > builds? I'll try something.
Oops, doesn't seem related to this merge...