Re: [PATCH 1/2] erofs: using macro instead of definition of log functions

[Gao Xiang]

Hi Hao,

On 2024/10/16 23:24, Gou Hao wrote:

No functional change intended.

Signed-off-by: Gou Hao <gou...@uniontech.com>
---
  fs/erofs/super.c | 51 ++++++++++++++++++------------------------------
  1 file changed, 19 insertions(+), 32 deletions(-)

diff --git a/fs/erofs/super.c b/fs/erofs/super.c
index 666873f745da..b04f888c8123 100644
--- a/fs/erofs/super.c
+++ b/fs/erofs/super.c
@@ -18,39 +18,26 @@

static struct kmem_cache *erofs_inode_cachep __read_mostly; -void _erofs_err(struct super_block *sb, const char *func, const char *fmt, ...)

-{
-       struct va_format vaf;
-       va_list args;
-
-       va_start(args, fmt);
-
-       vaf.fmt = fmt;
-       vaf.va = &args;
-
-       if (sb)
-               pr_err("(device %s): %s: %pV", sb->s_id, func, &vaf);
-       else
-               pr_err("%s: %pV", func, &vaf);
-       va_end(args);
-}
-
-void _erofs_info(struct super_block *sb, const char *func, const char *fmt, 
...)
-{
-       struct va_format vaf;
-       va_list args;
-
-       va_start(args, fmt);
-
-       vaf.fmt = fmt;
-       vaf.va = &args;
+#define _erofs_log_def(name) \
+       void _erofs_##name(struct super_block *sb, const char *func, const char 
*fmt, ...) \
+       { \
+               struct va_format vaf; \
+               va_list args; \
+               \
+               va_start(args, (fmt)); \
+               \
+               vaf.fmt = (fmt); \
+               vaf.va = &args; \
+               \
+               if ((sb)) \
+                       pr_##name("(device %s): %s: %pV", (sb)->s_id, (func), 
&vaf); \
+               else \
+                       pr_##name("%s: %pV", (func), &vaf); \
+               va_end(args); \
+       }

Thanks for the patch!

Although code simplicity is quite important for EROFS, but
I'm not sure introducing unnecessary macro definitions (which
can be avoided) is better for code readability.

I wonder if we can put this into another way, like the current
_btrfs_printk() and _f2fs_printk() if we really need to work
on this.

Thanks,
Gao Xiang