Re: [PATCH v2 1/3] argo: lower level of noisy connection-refused log

[dmukhin@xxxxxxxx]

On Fri, May 22, 2026 at 08:36:58PM +0300, Mykola Kvach wrote:
> Hi Denis,
> 
> Just one follow-up after sending my Reviewed-by.
> 
> The patches look fine to me and my tag still stands, but there are a
> couple of side effects worth considering here.
> 
> On Fri, May 22, 2026 at 7:53 PM <dmukhin@xxxxxxxx> wrote:
> >
> > From: Denis Mukhin <dmukhin@xxxxxxxx>
> >
> > Lower the log level of the "connection refused" log line, as it can
> > spam the logs when a dom0 service using the Argo hypercall tries to
> > communicate with a domain that is still starting up.
> >
> > Signed-off-by: Denis Mukhin <dmukhin@xxxxxxxx>
> > ---
> > Changes since v1:
> > - dropped duplicate "argo: " prefix
> > ---
> >  xen/common/argo.c | 7 +++----
> >  1 file changed, 3 insertions(+), 4 deletions(-)
> >
> > diff --git a/xen/common/argo.c b/xen/common/argo.c
> > index 28626e00a8cb..98a3db7fd070 100644
> > --- a/xen/common/argo.c
> > +++ b/xen/common/argo.c
> > @@ -2034,10 +2034,9 @@ sendv(struct domain *src_d, xen_argo_addr_t 
> > *src_addr,
> >                                          src_id.domain_id);
> >      if ( !ring_info )
> >      {
> > -        gprintk(XENLOG_ERR,
> > -                "argo: vm%u connection refused, src (vm%u:%x) dst 
> > (vm%u:%x)\n",
> > -                current->domain->domain_id, src_id.domain_id, src_id.aport,
> > -                dst_addr->domain_id, dst_addr->aport);
> > +        argo_dprintk("vm%u connection refused, src (vm%u:%x) dst 
> > (vm%u:%x)\n",
> 
> First, switching from gprintk() to argo_dprintk() means the message will
> no longer include the "%pv" context printed by gprintk(), i.e. the
> current vCPU/domain context. If that context is still useful for this
> message, it may need to be added explicitly.

I think %pv context is useful for debugging, I will add another small
patch to re-wire argo_dprintk() to gprintk().

> 
> Second, this also changes when the message is printed. gprintk() is a
> regular printk() wrapper with a guest prefix and the requested log level,
> while argo_dprintk() is compiled as a no-op unless ARGO_DEBUG is enabled.
> So the change is not only lowering the log level from error to debug, but
> also making the message depend on ARGO_DEBUG.
> 
> This may be intended, but in that case it might be worth mentioning this
> in the commit message.

Yes, this specific logline should be fine to be compiled out, the error
is propaged back to the caller domain via XEN_ARGO_OP_sendv hypercall.

I will update the commit message.

Thanks!

> 
> ~Mykola
>