On Wed, Nov 27, 2024 at 9:57 PM Alex Markuze <amarkuze@xxxxxxxxxx> wrote: > You are correct, that is why I'm testing a patch that deals with all > cases where i_size < offset. I don't like that patch because it looks complicated; it obscures the problem and it runs a bunch of code (fscrypt, zero_page_vector) before noticing the problem. My patch is simple and breaks the loop as soon as the new size is known. But I found a bug in my patch: I forgot to call ceph_osdc_put_request(). And while looking at it, I found another (old) leak bug. I'll post two new patches. (I'm trying hard to suppress a rant about C, after fixing several other Ceph leak bugs this week that caused server outages over here.)
