On Thu, 2012-12-06 at 09:45 +0100, Sebastian Andrzej Siewior wrote: > On 12/05/2012 09:53 PM, Nicholas A. Bellinger wrote: > > Changing the commit log here to drop the commit 8f9f44f reference, as I > > still don't see how that change effects the underlying issue with > > highmem. ;) > > You can checkout that commit see that it does not work. Checkout HEAD~1 > and you should see that it works. > Really, it's functionally the exact same rountines in target_core that do the SGL mapping before the conversion of tcm_loop to use target_submit_cmd_map_sgls(). I think this may have been hitting low-memory allocations for the first couple of tests during bisection, until non GFP_DMA started getting used for individual tcm_loop LLD I/O. --nab -- To unsubscribe from this list: send the line "unsubscribe target-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
