Re: [PATCH v2 0/3] sunrpc: Fix `make W=1` build issues

[Andy Shevchenko]

On Wed, Feb 04, 2026 at 09:28:28AM -0500, Jeff Layton wrote:
> On Wed, 2026-02-04 at 10:41 +0100, Andy Shevchenko wrote:
> > Compiler is not happy about unused variables (especially when
> > dprintk() call is defined as no-op). Here is the series to
> > address the issues.
> > 
> > Changelog v2:
> > - added patch to kill RPC_IFDEBUG() macro (LKP, Geert)
> > - united separate patches in the series
> > - collected tags (Geert)
> > 
> > v1: 20260204010402.2149563-1-andriy.shevchenko@xxxxxxxxxxxxxxx
> > v1: 20260204010415.2149607-1-andriy.shevchenko@xxxxxxxxxxxxxxx
> > 
> > Andy Shevchenko (3):
> >   nfs/blocklayout: Fix compilation error (`make W=1`) in
> >     bl_write_pagelist()
> >   sunrpc: Kill RPC_IFDEBUG()
> >   sunrpc: Fix compilation error (`make W=1`) when dprintk() is no-op
> > 
> >  fs/lockd/svclock.c                       |  5 +++++
> >  fs/nfs/blocklayout/blocklayout.c         |  4 +---
> >  fs/nfsd/nfsfh.c                          |  9 +++++---
> >  include/linux/sunrpc/debug.h             | 10 +++++----
> >  net/sunrpc/xprtrdma/svc_rdma_transport.c | 27 ++++++++++++------------
> >  5 files changed, 32 insertions(+), 23 deletions(-)
> 
> These all look like good changes to me. The first patch should go to
> Trond/Anna and Chuck will probably pick up the other two?

As I explained in v1, the error in the first patch may be shadowed if the third
one (and hence second) is taken first. That's why I prefer them to go via
single place.

> Reviewed-by: Jeff Layton <jlayton@xxxxxxxxxx>

Thanks!

-- 
With Best Regards,
Andy Shevchenko