Re: [PATCH v2] sunrpc: simplify dprintk() macros and cleanup redundant debug guards

On Thu, Mar 12, 2026 at 11:54:15PM +0800, Sean Chang wrote:

On Thu, Mar 12, 2026 at 5:47 AM Andy Shevchenko
[off-list ref] wrote:

quoted

On Tue, Mar 03, 2026 at 10:07:25PM +0800, Sean Chang wrote:

quoted

Following David Laight's suggestion, simplify the macro definitions by
removing the unnecessary 'fmt' argument and using no_printk(__VA_ARGS__)
directly.

Verification with .lst files under -O2 confirms that the compiler
successfully performs "dead code elimination". Even when variables
(like char buf[] in nfsfh.c) or static helper functions (like
nlmdbg_cookie2a() in svclock.c) are declared without #ifdef, they are
completely optimized out (no stack allocation, no symbol references in
the final executable) as they are only referenced within no_printk().

Does this patch fixes also 202603110038.P6d14oxa-lkp@intel.com?

Regarding the LKP report:
I have reproduced the Sparse warning (void vs int mismatch) locally.
To resolve this, I'll use the do-while(0) form in v3 to ensure the macro
always evaluates to void:
-# define dfprintk(fac, ...)            no_printk(__VA_ARGS__)
-# define dfprintk_rcu(fac, ...)        no_printk(__VA_ARGS__)
+# define dfprintk(fac, ...)            do { no_printk(__VA_ARGS__); } while (0)
+# define dfprintk_rcu(fac, ...)        do { no_printk(__VA_ARGS__); } while (0)

Wouldn't be better to drop ({}) in nfs_error*() macros?

-- 
With Best Regards,
Andy Shevchenko