Gabriel Krisman Bertazi <krisman@xxxxxxxxxxxxx> wrote: > After the two previous RFCs this is an attempt to get the ball spinning > again. > > The problem I'm trying to solve is providing an interface to monitor > filesystem errors. This patchset includes an example implementation of > ext4 error notification. This goes obviously on top of the watch_queue > mechanism. Thanks for picking this up and advancing it. > Regarding the buffer overrun issue that would require fsinfo or another > method to expose counters, I think they can be added at a later date > with no change to what this patch series attempts to do, therefore I'm > proposing we don't wait for fsinfo before getting this merged. That's fine, but anyone wanting to use this will need to be aware that overruns are a problem. David
