On 08/21/2011 07:38 PM, Ted Ts'o wrote:
On Sat, Aug 20, 2011 at 07:29:42PM -0700, Allison Henderson wrote:+ /* If the range is block aligned, unmap */ + if (range_to_discard == blocksize) { + clear_buffer_dirty(bh); + bh->b_bdev = NULL; + clear_buffer_mapped(bh); + clear_buffer_req(bh); + clear_buffer_new(bh); + clear_buffer_delay(bh); + clear_buffer_unwritten(bh); + clear_buffer_uptodate(bh); + ClearPageUptodate(page);Is this ClearPageUptodate() necessary? If the page is uptodate, and we've zero'ed out the relevant space, why do we need to force the page to read in from disk again? (In fact, if we want to avoid forcing the whole punched region to be written to disk, we'll need to get rid of the ClearPageUptodate() sooner or later --- and I'm not sure I see why it's needed now. Am I missing something?) - Ted
Oh, I see, maybe we can take this out now then. Initially we put this in there when all we were doing was unmapping the buffer heads. The idea was that if we just unmap the buffer heads and set the page not uptodate we wouldnt have to zero the page, but since that didnt work out for the mapped reads, we added the zeroing back in. I'll run some more tests to see if we can take it out. Thx!
Allison Henderson -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
