On Mon, Nov 25, 2024 at 2:24 PM Alex Markuze <amarkuze@xxxxxxxxxx> wrote:
> Max, could you add a cap on the retry count to your original patch?
Before I wrote code that's not useful at all: I don't quite get why
retry on buffer overflow is necessary at all. It looks like it once
seemed to be a useful kludge, but then 1b71fe2efa31 ("ceph analog of
cifs build_path_from_dentry() race fix") added the read_seqretry()
check which, to my limited understanding, is a more robust
implementation of rename detection.
Max
