On Tue, 2013-01-15 at 11:33 -0800, Roland Dreier wrote: > On Tue, Jan 15, 2013 at 8:20 AM, Martin Svec <martin.svec@xxxxxxxx> wrote: > > AFAIK, full WRITE_SAME support for IBLOCK backends will be in kernel 3.8. > > The way I understand things, this only works if the underlying block > device supports WRITE SAME (ie is a reasonably high-end array). For > "normal" block devices, there is no emulation of WRITE SAME without > unmap in the target core. This statement is incorrect. The WRITE_SAME w/o UNMAP emulation that went into v3.8-rc1 here: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=commitdiff;h=f6970ad31d42fceb38b5595cbad093a4d0bfcc43 is hardware agnostic, and has no dependency on underlying drive support for UNMAP (eg: block discard). This is because the logic simply translates WRITE_SAME w/o UNMAP operations into struct bio WRITEs for a single received block. --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
