On Wed, Feb 25, 2026 at 10:12:36PM +0800, David Yang wrote:

On Wed, Feb 25, 2026 at 9:52 PM Andrew Lunn [off-list ref] wrote:

quoted

quoted

-/* Some registers, like VLANn_CTRL, should always be written in 64-bit, even if
- * you are to write only the lower / upper 32 bits.
+/* Some registers, like VLANn_CTRL, should always be written in chunks, even if
+ * you only want to write parts of 32 bits.
  *
- * There is no such restriction for reading, but we still provide 64-bit read
- * wrappers so that we always handle u64 values.
+ * There is no such restriction for reading, but we still provide multi-chunk
+ * read wrappers so that we can handle them consistently.
  */

-static int yt921x_reg64_read(struct yt921x_priv *priv, u32 reg, u64 *valp)
+static int
+yt921x_longreg_read(struct yt921x_priv *priv, u32 reg, u32 *vals,
+                 unsigned int span)

Just looking at this, it is not obvious what units span is. Many bulk
read operations takes bytes, even if the underlying code works in
words. I would probably rename span to something like num_u32.

I would also keep yt921x_reg64_read, yt921x_reg64_write,
yt921x_reg64_update_bits() and just makes them wrappers which call
yt921x_longreg_read().

I don't think keeping reg64_* would make it better. There will be more
96-bit regs as subsequent series progressing. It would be more messy
to have three types of regs: u32, u64, and u32[].

The lower level helper takes a u32[], but your higher level is always
yt921x_reg32_read(), yt921x_reg64_read(), or yt921x_reg96_read(). That
makes it obvious what size you are reading, and keeps your patch
small.

      Andrew