On 4/7/2016 10:14 AM, Sagi Grimberg wrote:
So this scatterlist seems perfectly fine (no gaps). The only thing I
can think of is that we don't have enough space reserved, but I counted
127 pages in the scatterlist when iser reserves 128 (by default).
Just in case I mis-counted, can you (or Max) try with the below
patch:
This works fine for me, but I wonder if the miscounting is in the
block layer?
We communicate the "max_pages" via sg_tablesize and max_hw_sectors.
iSER communicates (or at least supposed to) sg_tablesize=128 and
max_sectors=1024 to scsi (which sets the block queues attributes), that
should be sufficient for handling all the incoming IO. Unless we didn't
account for the first-page offset and the reminder coming at the last
segment (which results in an extra page). The patch should fix the
problem, however in this particular sg list I don't see the violation
(counted 127 pages)...
I counted 129 pages.
I will run some tests in our labs too.
Thanks.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
