On Wed, Mar 04, 2020 at 07:51:40AM -0800, Christoph Hellwig wrote: > On Wed, Mar 04, 2020 at 06:53:56PM +1100, Dave Chinner wrote: > > From: Dave Chinner <dchinner@xxxxxxxxxx> > > > > To clean up unmount record writing error handling, we need to move > > xlog_state_ioerror() higher up in the file. Also move the setting of > > the XLOG_IO_ERROR state to inside the function. > > > > Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx> > > FYI, I have a pending series that kills xlog_state_ioerror and > XLOG_IO_ERROR. Let me send that out now that Brians fix is in for-next. Can you rebase that on top of this? removing IOERROR is a much more invasive and riskier set of state machine changes compared to what this patchset does. This patchset simplifies some of the error handling that handles IOERROR, too... Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx
