FAILED: libqemu-i386-softmmu.a.p/hw_xen_xen_pt.c.o
...
/home/user/.work/qemu/hw/xen/xen_pt.c: In function ‘xen_pt_need_gsi’:
/home/user/.work/qemu/hw/xen/xen_pt.c:783:9: error: implicit declaration of function ‘error_report’; did you mean ‘error_report_err’? [-Wimplicit-function-declaration]
  783 |         error_report("Cannot open %s: %s", guest_type, strerror(errno));
      |         ^~~~~~~~~~~~
      |         error_report_err
/home/user/.work/qemu/hw/xen/xen_pt.c:783:9: error: nested extern declaration of ‘error_report’ [-Werror=nested-externs]
cc1: all warnings being treated as errors

Fixes: 012842c ("log: make '-msg timestamp=on' apply to all qemu_log usage")
Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
---
 hw/xen/xen_pt.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/hw/xen/xen_pt.c b/hw/xen/xen_pt.c
index 9d16644d82e..006b5b55f24 100644
--- a/hw/xen/xen_pt.c
+++ b/hw/xen/xen_pt.c
@@ -54,6 +54,7 @@
 
 #include "qemu/osdep.h"
 #include "qapi/error.h"
+#include "qemu/error-report.h"
 #include <sys/ioctl.h>
 
 #include "hw/pci/pci.h"
-- 
2.47.2

On Fri, Jul 25, 2025 at 09:57:49AM -0700, Pierrick Bouvier wrote:
> FAILED: libqemu-i386-softmmu.a.p/hw_xen_xen_pt.c.o
> ...
> /home/user/.work/qemu/hw/xen/xen_pt.c: In function ‘xen_pt_need_gsi’:
> /home/user/.work/qemu/hw/xen/xen_pt.c:783:9: error: implicit declaration of function ‘error_report’; did you mean ‘error_report_err’? [-Wimplicit-function-declaration]
>   783 |         error_report("Cannot open %s: %s", guest_type, strerror(errno));
>       |         ^~~~~~~~~~~~
>       |         error_report_err
> /home/user/.work/qemu/hw/xen/xen_pt.c:783:9: error: nested extern declaration of ‘error_report’ [-Werror=nested-externs]
> cc1: all warnings being treated as errors
> 
> Fixes: 012842c ("log: make '-msg timestamp=on' apply to all qemu_log usage")

The problem pre-dated this, so this should be

 Fixes: cfcacbab38e (xen/passthrough: use gsi to map pirq when dom0 is PVH)

See the prior posting of this fix last week:

  https://lists.nongnu.org/archive/html/qemu-devel/2025-07/msg04831.html

> Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
> ---
>  hw/xen/xen_pt.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/hw/xen/xen_pt.c b/hw/xen/xen_pt.c
> index 9d16644d82e..006b5b55f24 100644
> --- a/hw/xen/xen_pt.c
> +++ b/hw/xen/xen_pt.c
> @@ -54,6 +54,7 @@
>  
>  #include "qemu/osdep.h"
>  #include "qapi/error.h"
> +#include "qemu/error-report.h"
>  #include <sys/ioctl.h>
>  
>  #include "hw/pci/pci.h"
> -- 
> 2.47.2
> 

With regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|

On 7/25/25 10:22 AM, Daniel P. Berrangé wrote:
> On Fri, Jul 25, 2025 at 09:57:49AM -0700, Pierrick Bouvier wrote:
>> FAILED: libqemu-i386-softmmu.a.p/hw_xen_xen_pt.c.o
>> ...
>> /home/user/.work/qemu/hw/xen/xen_pt.c: In function ‘xen_pt_need_gsi’:
>> /home/user/.work/qemu/hw/xen/xen_pt.c:783:9: error: implicit declaration of function ‘error_report’; did you mean ‘error_report_err’? [-Wimplicit-function-declaration]
>>    783 |         error_report("Cannot open %s: %s", guest_type, strerror(errno));
>>        |         ^~~~~~~~~~~~
>>        |         error_report_err
>> /home/user/.work/qemu/hw/xen/xen_pt.c:783:9: error: nested extern declaration of ‘error_report’ [-Werror=nested-externs]
>> cc1: all warnings being treated as errors
>>
>> Fixes: 012842c ("log: make '-msg timestamp=on' apply to all qemu_log usage")
> 
> The problem pre-dated this, so this should be
> 
>   Fixes: cfcacbab38e (xen/passthrough: use gsi to map pirq when dom0 is PVH)
> 
> See the prior posting of this fix last week:
> 
>    https://lists.nongnu.org/archive/html/qemu-devel/2025-07/msg04831.html
>

While true, this was not a problem as error-report.h was still 
implicitely included (as Markus pointed). We can debate where the 
regression comes from, but from git bisect point of view, this is the 
trace one, at least with the simple "./configure --enable-debug".

I don't mind changing the fixed commit, or even removing any Fix line at 
all, but we just need something upstream merged as soon as possible to 
not break everyone build.

>> Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
>> ---
>>   hw/xen/xen_pt.c | 1 +
>>   1 file changed, 1 insertion(+)
>>
>> diff --git a/hw/xen/xen_pt.c b/hw/xen/xen_pt.c
>> index 9d16644d82e..006b5b55f24 100644
>> --- a/hw/xen/xen_pt.c
>> +++ b/hw/xen/xen_pt.c
>> @@ -54,6 +54,7 @@
>>   
>>   #include "qemu/osdep.h"
>>   #include "qapi/error.h"
>> +#include "qemu/error-report.h"
>>   #include <sys/ioctl.h>
>>   
>>   #include "hw/pci/pci.h"
>> -- 
>> 2.47.2
>>
> 
> With regards,
> Daniel

On Fri, Jul 25, 2025 at 10:49:58AM -0700, Pierrick Bouvier wrote:
> On 7/25/25 10:22 AM, Daniel P. Berrangé wrote:
> > On Fri, Jul 25, 2025 at 09:57:49AM -0700, Pierrick Bouvier wrote:
> > > FAILED: libqemu-i386-softmmu.a.p/hw_xen_xen_pt.c.o
> > > ...
> > > /home/user/.work/qemu/hw/xen/xen_pt.c: In function ‘xen_pt_need_gsi’:
> > > /home/user/.work/qemu/hw/xen/xen_pt.c:783:9: error: implicit declaration of function ‘error_report’; did you mean ‘error_report_err’? [-Wimplicit-function-declaration]
> > >    783 |         error_report("Cannot open %s: %s", guest_type, strerror(errno));
> > >        |         ^~~~~~~~~~~~
> > >        |         error_report_err
> > > /home/user/.work/qemu/hw/xen/xen_pt.c:783:9: error: nested extern declaration of ‘error_report’ [-Werror=nested-externs]
> > > cc1: all warnings being treated as errors
> > > 
> > > Fixes: 012842c ("log: make '-msg timestamp=on' apply to all qemu_log usage")
> > 
> > The problem pre-dated this, so this should be
> > 
> >   Fixes: cfcacbab38e (xen/passthrough: use gsi to map pirq when dom0 is PVH)
> > 
> > See the prior posting of this fix last week:
> > 
> >    https://lists.nongnu.org/archive/html/qemu-devel/2025-07/msg04831.html
> > 
> 
> While true, this was not a problem as error-report.h was still implicitely
> included (as Markus pointed). We can debate where the regression comes from,
> but from git bisect point of view, this is the trace one, at least with the
> simple "./configure --enable-debug".

It was /only/ implicitly included if you had the 'log' trace backend
enabled - other configurations missed it, hence the build failures
we got in Fedora from commit cfcacbab38e.


With regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|