There's a better solution than falling back to buffered io
There's a better solution than falling back to buffered io
Posted Nov 12, 2025 19:35 UTC (Wed) by quwenruo_suse (subscriber, #124148)In reply to: There's a better solution than falling back to buffered io by koverstreet
Parent article: The intersection of unstable pages and direct I/O
I doubt. As I tried both, bouncing and falling back to buffered IO on btrfs, no observable difference.
The huge performance drop in my previous observations is caused by unoptimized checksum implementation (kvm64 has no hardware accelerated CRC32C).
Posted Nov 12, 2025 19:57 UTC (Wed)
by koverstreet (✭ supporter ✭, #4296)
[Link]
There's a better solution than falling back to buffered io
and you thought that test would be representative...?