* [Qemu-devel] changeset b0cb0a66d broken - missing QEMU_OPTION_enable_hax
@ 2017-01-23 19:32 James Hanley
2017-01-23 19:41 ` Peter Maydell
0 siblings, 1 reply; 5+ messages in thread
From: James Hanley @ 2017-01-23 19:32 UTC (permalink / raw)
To: qemu-devel, Vincent Palatin
Changeset fails to compile:
qemu/vl.c:3683:18: error: ‘QEMU_OPTION_enable_hax’ undeclared (first use in
this function)
case QEMU_OPTION_enable_hax:
^
I assume that there is details in
./qemu-options.def
./qemu-options.hx
that were not included with the commit?
http://git.qemu.org/?p=qemu.git;a=commit;h=b0cb0a66d6d535112aa513568ef21dcb1ad283ed
-Jim
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [Qemu-devel] changeset b0cb0a66d broken - missing QEMU_OPTION_enable_hax
2017-01-23 19:32 [Qemu-devel] changeset b0cb0a66d broken - missing QEMU_OPTION_enable_hax James Hanley
@ 2017-01-23 19:41 ` Peter Maydell
2017-01-23 19:55 ` James Hanley
2017-01-24 17:28 ` Michael S. Tsirkin
0 siblings, 2 replies; 5+ messages in thread
From: Peter Maydell @ 2017-01-23 19:41 UTC (permalink / raw)
To: James Hanley; +Cc: qemu-devel, Vincent Palatin
On 23 January 2017 at 19:32, James Hanley <jhanley@dgtlrift.com> wrote:
> Changeset fails to compile:
>
> qemu/vl.c:3683:18: error: ‘QEMU_OPTION_enable_hax’ undeclared (first use in
> this function)
> case QEMU_OPTION_enable_hax:
> ^
> I assume that there is details in
>
> ./qemu-options.def
> ./qemu-options.hx
>
> that were not included with the commit?
Your source tree probably has a stale qemu-options.def in it
from an attempt to do an in-tree build at some point in the
past (and now you are doing out-of-tree builds). If you do a
'make distclean' in the source tree this will get rid of the
stale junk and the build should work. (Or you can just git
clone a fresh tree; or attempt to more carefully remove
individual stale stuff, but that risks leaving some other
stale file around.)
(The change to qemu-options.hx is sufficient, because we
generate qemu-options.def from it in the build tree; but
if you have a stale .def file in the source tree then
gcc will pick that one up before the one in the build tree;
which works until something requires a change that postdates
the stale stuff.)
thanks
-- PMM
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [Qemu-devel] changeset b0cb0a66d broken - missing QEMU_OPTION_enable_hax
2017-01-23 19:41 ` Peter Maydell
@ 2017-01-23 19:55 ` James Hanley
2017-01-24 17:28 ` Michael S. Tsirkin
1 sibling, 0 replies; 5+ messages in thread
From: James Hanley @ 2017-01-23 19:55 UTC (permalink / raw)
To: Peter Maydell; +Cc: qemu-devel, Vincent Palatin
That did it - Thanks, and my apologies to Vincent for the false alarm.
-Jim
On Mon, Jan 23, 2017 at 2:41 PM, Peter Maydell <peter.maydell@linaro.org>
wrote:
> On 23 January 2017 at 19:32, James Hanley <jhanley@dgtlrift.com> wrote:
> > Changeset fails to compile:
> >
> > qemu/vl.c:3683:18: error: ‘QEMU_OPTION_enable_hax’ undeclared (first use
> in
> > this function)
> > case QEMU_OPTION_enable_hax:
> > ^
> > I assume that there is details in
> >
> > ./qemu-options.def
> > ./qemu-options.hx
> >
> > that were not included with the commit?
>
> Your source tree probably has a stale qemu-options.def in it
> from an attempt to do an in-tree build at some point in the
> past (and now you are doing out-of-tree builds). If you do a
> 'make distclean' in the source tree this will get rid of the
> stale junk and the build should work. (Or you can just git
> clone a fresh tree; or attempt to more carefully remove
> individual stale stuff, but that risks leaving some other
> stale file around.)
>
> (The change to qemu-options.hx is sufficient, because we
> generate qemu-options.def from it in the build tree; but
> if you have a stale .def file in the source tree then
> gcc will pick that one up before the one in the build tree;
> which works until something requires a change that postdates
> the stale stuff.)
>
> thanks
> -- PMM
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [Qemu-devel] changeset b0cb0a66d broken - missing QEMU_OPTION_enable_hax
2017-01-23 19:41 ` Peter Maydell
2017-01-23 19:55 ` James Hanley
@ 2017-01-24 17:28 ` Michael S. Tsirkin
2017-01-24 17:56 ` Peter Maydell
1 sibling, 1 reply; 5+ messages in thread
From: Michael S. Tsirkin @ 2017-01-24 17:28 UTC (permalink / raw)
To: Peter Maydell; +Cc: James Hanley, Vincent Palatin, qemu-devel
On Mon, Jan 23, 2017 at 07:41:48PM +0000, Peter Maydell wrote:
> On 23 January 2017 at 19:32, James Hanley <jhanley@dgtlrift.com> wrote:
> > Changeset fails to compile:
> >
> > qemu/vl.c:3683:18: error: ‘QEMU_OPTION_enable_hax’ undeclared (first use in
> > this function)
> > case QEMU_OPTION_enable_hax:
> > ^
> > I assume that there is details in
> >
> > ./qemu-options.def
> > ./qemu-options.hx
> >
> > that were not included with the commit?
>
> Your source tree probably has a stale qemu-options.def in it
> from an attempt to do an in-tree build at some point in the
> past (and now you are doing out-of-tree builds). If you do a
> 'make distclean' in the source tree this will get rid of the
> stale junk and the build should work. (Or you can just git
> clone a fresh tree; or attempt to more carefully remove
> individual stale stuff, but that risks leaving some other
> stale file around.)
>
> (The change to qemu-options.hx is sufficient, because we
> generate qemu-options.def from it in the build tree; but
> if you have a stale .def file in the source tree then
> gcc will pick that one up before the one in the build tree;
> which works until something requires a change that postdates
> the stale stuff.)
>
> thanks
> -- PMM
Maybe we should change the include order to always prefer
the build tree files if there?
An alternative is to do what Linux does, create a
file in tree when doing a build, and then make an
out of tree one fail and suggest distclean.
--
MST
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [Qemu-devel] changeset b0cb0a66d broken - missing QEMU_OPTION_enable_hax
2017-01-24 17:28 ` Michael S. Tsirkin
@ 2017-01-24 17:56 ` Peter Maydell
0 siblings, 0 replies; 5+ messages in thread
From: Peter Maydell @ 2017-01-24 17:56 UTC (permalink / raw)
To: Michael S. Tsirkin; +Cc: James Hanley, Vincent Palatin, qemu-devel
On 24 January 2017 at 17:28, Michael S. Tsirkin <mst@redhat.com> wrote:
> Maybe we should change the include order to always prefer
> the build tree files if there?
If we can do that it would be nice.
> An alternative is to do what Linux does, create a
> file in tree when doing a build, and then make an
> out of tree one fail and suggest distclean.
We already try to do that -- Makefile looks for
config-host.mak in the source tree and suggests a
rune involving distclean. The check isn't perfect,
though (and nor is the Linux kernel's -- I lost an
hour or so the other week to a problem that turned
out to be stale junk in the kernel source tree that I
fixed with a 'make mrproper').
thanks
-- PMM
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2017-01-24 17:56 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-01-23 19:32 [Qemu-devel] changeset b0cb0a66d broken - missing QEMU_OPTION_enable_hax James Hanley
2017-01-23 19:41 ` Peter Maydell
2017-01-23 19:55 ` James Hanley
2017-01-24 17:28 ` Michael S. Tsirkin
2017-01-24 17:56 ` Peter Maydell
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).