[Qemu-devel] [PULL] pci, vhost fixes

[Qemu-devel] [PULL] pci, vhost fixes

[Michael S. Tsirkin]

The following changes since commit 259d5577ae546de27e25a40710f81aa031660de8:

  lsi: Fix value overflow in request tag processing (2010-05-22 15:32:32 +0200)

are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/mst/qemu.git pci

Isaku Yamahata (4):
      pci: clean up of pci_set_default_subsystem_id().
      pci: add const to pci_is_express(), pci_config_size().
      pci.h: remove unused constants.
      msix: remove duplicated defines.

Michael S. Tsirkin (1):
      vhost: fix build broken by 0bfcd599e3f5c5679cc7d0165a0a1822e2f60de2

 hw/msix.c      |    8 --------
 hw/pci.c       |   12 +++++-------
 hw/pci.h       |   15 ++-------------
 hw/vhost_net.c |    2 +-
 4 files changed, 8 insertions(+), 29 deletions(-)

-- 
MST

Re: [Qemu-devel] [PULL] pci, vhost fixes

[Blue Swirl]

On Sun, May 30, 2010 at 3:50 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
> The following changes since commit 259d5577ae546de27e25a40710f81aa031660de8:
>
>  lsi: Fix value overflow in request tag processing (2010-05-22 15:32:32 +0200)
>
> are available in the git repository at:
>  git://git.kernel.org/pub/scm/linux/kernel/git/mst/qemu.git pci
>
> Isaku Yamahata (4):
>      pci: clean up of pci_set_default_subsystem_id().
>      pci: add const to pci_is_express(), pci_config_size().
>      pci.h: remove unused constants.
>      msix: remove duplicated defines.
>
> Michael S. Tsirkin (1):
>      vhost: fix build broken by 0bfcd599e3f5c5679cc7d0165a0a1822e2f60de2

Nack, see 29f91781bd300b228d9c38b9f6c1e0b6a38f25fb.

Re: [Qemu-devel] [PULL] pci, vhost fixes

[Michael S. Tsirkin]

On Sun, May 30, 2010 at 04:56:11PM +0000, Blue Swirl wrote:
> On Sun, May 30, 2010 at 3:50 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
> > The following changes since commit 259d5577ae546de27e25a40710f81aa031660de8:
> >
> >  lsi: Fix value overflow in request tag processing (2010-05-22 15:32:32 +0200)
> >
> > are available in the git repository at:
> >  git://git.kernel.org/pub/scm/linux/kernel/git/mst/qemu.git pci
> >
> > Isaku Yamahata (4):
> >      pci: clean up of pci_set_default_subsystem_id().
> >      pci: add const to pci_is_express(), pci_config_size().
> >      pci.h: remove unused constants.
> >      msix: remove duplicated defines.
> >
> > Michael S. Tsirkin (1):
> >      vhost: fix build broken by 0bfcd599e3f5c5679cc7d0165a0a1822e2f60de2
> 
> Nack, see 29f91781bd300b228d9c38b9f6c1e0b6a38f25fb.

Did you follow the discussion about that patch before applying?
29f91781bd300b228d9c38b9f6c1e0b6a38f25fb does not make sense
to me, I think it should be reverted.
we have long long value, let us just print with %ll.
casting to uint64 and printing with PRi64 is silly.


-- 
MST

Re: [Qemu-devel] [PULL] pci, vhost fixes

[Blue Swirl]

On Sun, May 30, 2010 at 6:05 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
> On Sun, May 30, 2010 at 04:56:11PM +0000, Blue Swirl wrote:
>> On Sun, May 30, 2010 at 3:50 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
>> > The following changes since commit 259d5577ae546de27e25a40710f81aa031660de8:
>> >
>> >  lsi: Fix value overflow in request tag processing (2010-05-22 15:32:32 +0200)
>> >
>> > are available in the git repository at:
>> >  git://git.kernel.org/pub/scm/linux/kernel/git/mst/qemu.git pci
>> >
>> > Isaku Yamahata (4):
>> >      pci: clean up of pci_set_default_subsystem_id().
>> >      pci: add const to pci_is_express(), pci_config_size().
>> >      pci.h: remove unused constants.
>> >      msix: remove duplicated defines.
>> >
>> > Michael S. Tsirkin (1):
>> >      vhost: fix build broken by 0bfcd599e3f5c5679cc7d0165a0a1822e2f60de2
>>
>> Nack, see 29f91781bd300b228d9c38b9f6c1e0b6a38f25fb.
>
> Did you follow the discussion about that patch before applying?
> 29f91781bd300b228d9c38b9f6c1e0b6a38f25fb does not make sense
> to me, I think it should be reverted.
> we have long long value, let us just print with %ll.
> casting to uint64 and printing with PRi64 is silly.

We have PRI*64 just for this purpose, so let's use them. The
discussion about that was earlier.

Re: [Qemu-devel] [PULL] pci, vhost fixes

[Michael S. Tsirkin]

On Sun, May 30, 2010 at 06:19:40PM +0000, Blue Swirl wrote:
> On Sun, May 30, 2010 at 6:05 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
> > On Sun, May 30, 2010 at 04:56:11PM +0000, Blue Swirl wrote:
> >> On Sun, May 30, 2010 at 3:50 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
> >> > The following changes since commit 259d5577ae546de27e25a40710f81aa031660de8:
> >> >
> >> >  lsi: Fix value overflow in request tag processing (2010-05-22 15:32:32 +0200)
> >> >
> >> > are available in the git repository at:
> >> >  git://git.kernel.org/pub/scm/linux/kernel/git/mst/qemu.git pci
> >> >
> >> > Isaku Yamahata (4):
> >> >      pci: clean up of pci_set_default_subsystem_id().
> >> >      pci: add const to pci_is_express(), pci_config_size().
> >> >      pci.h: remove unused constants.
> >> >      msix: remove duplicated defines.
> >> >
> >> > Michael S. Tsirkin (1):
> >> >      vhost: fix build broken by 0bfcd599e3f5c5679cc7d0165a0a1822e2f60de2
> >>
> >> Nack, see 29f91781bd300b228d9c38b9f6c1e0b6a38f25fb.
> >
> > Did you follow the discussion about that patch before applying?
> > 29f91781bd300b228d9c38b9f6c1e0b6a38f25fb does not make sense
> > to me, I think it should be reverted.
> > we have long long value, let us just print with %ll.
> > casting to uint64 and printing with PRi64 is silly.
> 
> We have PRI*64 just for this purpose, so let's use them. The
> discussion about that was earlier.

Ho hum. As long as enough people got annoyed, there was hope that
debian will upodate their mingw port. I guess if everyone works around
this in code no one'll bother.  Oh well, not important enough to care.

-- 
MST

Re: [Qemu-devel] [PULL] pci, vhost fixes

[Blue Swirl]

On Sun, May 30, 2010 at 6:58 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
> On Sun, May 30, 2010 at 06:19:40PM +0000, Blue Swirl wrote:
>> On Sun, May 30, 2010 at 6:05 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
>> > On Sun, May 30, 2010 at 04:56:11PM +0000, Blue Swirl wrote:
>> >> On Sun, May 30, 2010 at 3:50 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
>> >> > The following changes since commit 259d5577ae546de27e25a40710f81aa031660de8:
>> >> >
>> >> >  lsi: Fix value overflow in request tag processing (2010-05-22 15:32:32 +0200)
>> >> >
>> >> > are available in the git repository at:
>> >> >  git://git.kernel.org/pub/scm/linux/kernel/git/mst/qemu.git pci
>> >> >
>> >> > Isaku Yamahata (4):
>> >> >      pci: clean up of pci_set_default_subsystem_id().
>> >> >      pci: add const to pci_is_express(), pci_config_size().
>> >> >      pci.h: remove unused constants.
>> >> >      msix: remove duplicated defines.
>> >> >
>> >> > Michael S. Tsirkin (1):
>> >> >      vhost: fix build broken by 0bfcd599e3f5c5679cc7d0165a0a1822e2f60de2
>> >>
>> >> Nack, see 29f91781bd300b228d9c38b9f6c1e0b6a38f25fb.
>> >
>> > Did you follow the discussion about that patch before applying?
>> > 29f91781bd300b228d9c38b9f6c1e0b6a38f25fb does not make sense
>> > to me, I think it should be reverted.
>> > we have long long value, let us just print with %ll.
>> > casting to uint64 and printing with PRi64 is silly.
>>
>> We have PRI*64 just for this purpose, so let's use them. The
>> discussion about that was earlier.
>
> Ho hum. As long as enough people got annoyed, there was hope that
> debian will upodate their mingw port. I guess if everyone works around
> this in code no one'll bother.  Oh well, not important enough to care.

To make things even more annoying, mingw gcc from Debian stable
(4.2.1-sjlj) gives no warnings, the warnings came from the updated
compiler (4.4.2) of the future release. So we'll be stuck with the
warnings for a looong time. :-(

[Qemu-devel] Re: [PULL] pci, vhost fixes

[Paolo Bonzini]

On 05/30/2010 08:19 PM, Blue Swirl wrote:
> We have PRI*64 just for this purpose, so let's use them. The
> discussion about that was earlier.

How does mingw matter for vhost-net specific code?

Paolo

[Qemu-devel] Re: [PULL] pci, vhost fixes

[Michael S. Tsirkin]

On Mon, May 31, 2010 at 02:51:26PM +0200, Paolo Bonzini wrote:
> On 05/30/2010 08:19 PM, Blue Swirl wrote:
>> We have PRI*64 just for this purpose, so let's use them. The
>> discussion about that was earlier.
>
> How does mingw matter for vhost-net specific code?
>
> Paolo

Good point. That code is linux-only, so we are safe using %lld.
By the time someone ports vhost to windows, there's a chance
debian will update it's mingw.  Blue Swirl?

-- 
MST

[Qemu-devel] Re: [PULL] pci, vhost fixes

[Blue Swirl]

On Mon, May 31, 2010 at 1:35 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
> On Mon, May 31, 2010 at 02:51:26PM +0200, Paolo Bonzini wrote:
>> On 05/30/2010 08:19 PM, Blue Swirl wrote:
>>> We have PRI*64 just for this purpose, so let's use them. The
>>> discussion about that was earlier.
>>
>> How does mingw matter for vhost-net specific code?
>>
>> Paolo
>
> Good point. That code is linux-only, so we are safe using %lld.
> By the time someone ports vhost to windows, there's a chance
> debian will update it's mingw.  Blue Swirl?

It's a good point, but there's also consistency: now no '%ll*' formats are used.

Re: [Qemu-devel] Re: [PULL] pci, vhost fixes

[Markus Armbruster]

Blue Swirl <blauwirbel@gmail.com> writes:

> On Mon, May 31, 2010 at 1:35 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
>> On Mon, May 31, 2010 at 02:51:26PM +0200, Paolo Bonzini wrote:
>>> On 05/30/2010 08:19 PM, Blue Swirl wrote:
>>>> We have PRI*64 just for this purpose, so let's use them. The
>>>> discussion about that was earlier.
>>>
>>> How does mingw matter for vhost-net specific code?
>>>
>>> Paolo
>>
>> Good point. That code is linux-only, so we are safe using %lld.
>> By the time someone ports vhost to windows, there's a chance
>> debian will update it's mingw.  Blue Swirl?
>
> It's a good point, but there's also consistency: now no '%ll*' formats are used.

If it ain't broke, don't fix it.  Especially not when the fix is ugly.

Re: [Qemu-devel] Re: [PULL] pci, vhost fixes

[Blue Swirl]

On Tue, Jun 1, 2010 at 5:32 AM, Markus Armbruster <armbru@redhat.com> wrote:
> Blue Swirl <blauwirbel@gmail.com> writes:
>
>> On Mon, May 31, 2010 at 1:35 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
>>> On Mon, May 31, 2010 at 02:51:26PM +0200, Paolo Bonzini wrote:
>>>> On 05/30/2010 08:19 PM, Blue Swirl wrote:
>>>>> We have PRI*64 just for this purpose, so let's use them. The
>>>>> discussion about that was earlier.
>>>>
>>>> How does mingw matter for vhost-net specific code?
>>>>
>>>> Paolo
>>>
>>> Good point. That code is linux-only, so we are safe using %lld.
>>> By the time someone ports vhost to windows, there's a chance
>>> debian will update it's mingw.  Blue Swirl?
>>
>> It's a good point, but there's also consistency: now no '%ll*' formats are used.
>
> If it ain't broke, don't fix it.  Especially not when the fix is ugly.

Fully agree. But because beauty is in the eye of the beholder, our
conclusions are probably different. ;-)