Hi Geert,
On Thu, 2012-09-06 at 19:18 +0200, Geert Jansen wrote:
> Hi,
>
> i'm doing some testing with WRITE_SAME / UNMAP but i do not have any
> real hardware that supports these commands. I have configured a LIO
> target with a file backend. However when i try to use these commands
> with sg_unmap and sg_write_same, i get the following kernel messages:
>
> [ 2279.544606] UNMAP emulation not supported for: fileio
> [ 2299.480649] WRITE_SAME w/o UNMAP bit not supported for Block
> Discard Emulation
> [ 2303.320805] WRITE_SAME emulation not supported for: fileio
>
> Is there any way i can have it emulate UNMAP and WRITE_SAME? I don't
> require e.g. UNMAP to really unmap a block should the file backing
> store be on an SSD. Basically i'm just looking for a way to test these
> commands from an initiator point of view.
>
You can test UNMAP + WRITE_SAME w/ UNMAP=1 using scsi_debug.ko,
which enables these features with the following module params:
# modprobe scsi_debug lbpu=1 lbpws=1
# sg_inq --page=bl /dev/sdf
VPD INQUIRY: Block limits page (SBC)
Maximum compare and write length: 0 blocks
Optimal transfer length granularity: 1 blocks
Maximum transfer length: 16384 blocks
Optimal transfer length: 64 blocks
Maximum prefetch, xdread, xdwrite transfer length: 0 blocks
Maximum unmap LBA count: 4294967295
Maximum unmap block descriptor count: 256
Optimal unmap granularity: 1
Unmap granularity alignment valid: 0
Unmap granularity alignment: 0
# sg_unmap -vv --lba=0 --num=1 /dev/sdf
open /dev/sdf with flags=0x802
unmap cdb: 42 00 00 00 00 00 00 00 18 00
unmap parameter list:
00 16 00 10 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 01 00 00 00 00
>From there you should be able to import the scsi_debug LUN into IBLOCK +
enable emulate_tp[u,ws]=1, or export directly via pSCSI.
--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
