Hi,
在 2025/08/27 14:07, Paul Menzel 写道:
Dear Kuai,
Thank you for your reply.
Am 27.08.25 um 05:44 schrieb Yu Kuai:
在 2025/08/26 17:52, Paul Menzel 写道:
It’d be great if you could motivate, why a lockless bitmap is needed
> compared to the current implemention.
Se the performance test, old bitmap have global spinlock and is bad with
fast disk.
Yes, but it’s at the end, and not explicitly stated. Should you resend,
it’d be great if you could add that.
If there is no suggestions about functionality, I can add following in
the beginning when I apply this:
Due to known performance issues with md-bitmap and the unreasonable
implementations:
- self-managed IO submitting like filemap_write_page();
- global spin_lock
I have decided not to continue optimizing based on the current bitmap
implementation.
And the same as fixing those typos.
Thanks,
Kuai
[snip the typo part]
How can/should this patch be tested/benchmarked?
There is pending mdadm patch, rfc verion can be used. Will work on
formal version after this set is applied.
Understood. Maybe add an URL to the mdadm patch. (Sorry, should I have
missed it.)
--- a/drivers/md/md-bitmap.h
+++ b/drivers/md/md-bitmap.h
@@ -9,10 +9,26 @@
#define BITMAP_MAGIC 0x6d746962
+/*
+ * version 3 is host-endian order, this is deprecated and not used
for new
+ * array
+ */
+#define BITMAP_MAJOR_LO 3
+#define BITMAP_MAJOR_HOSTENDIAN 3
+/* version 4 is little-endian order, the default value */
+#define BITMAP_MAJOR_HI 4
+/* version 5 is only used for cluster */
+#define BITMAP_MAJOR_CLUSTERED 5
Move this to the header in a separate patch?
I prefer not, old bitmap use this as well.
Hmm, I do not understand the answer, as it’s moved in this patch, why
can’t it be moved in another? But it’s not that important.
Kind regards,
Paul
.
