Revision 9c674815d346305068b27bf03b5e86b659a1b111 authored by Christoph Hellwig on 18 April 2016, 21:06:28 UTC, committed by Doug Ledford on 05 May 2016, 16:41:24 UTC
iSER currently has a couple places that set max_sectors in either the host
template or SCSI host, and all of them get it wrong.

This patch instead uses a single assignment that (hopefully) gets it right:
the max_sectors value must be derived from the number of segments in the
FR or FMR structure, but actually be one lower than the page size multiplied
by the number of sectors, as it has to handle the case of non-aligned I/O.

Without this I get trivial to reproduce hangs when running xfstests
(on XFS) over iSER to Linux targets.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Max Gurtovoy <maxg@mellanox.com>
Acked-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Doug Ledford <dledford@redhat.com>
1 parent 4c8bb95
History
File Mode Size
power
regmap
Kconfig -rw-r--r-- 11.1 KB
Makefile -rw-r--r-- 967 bytes
attribute_container.c -rw-r--r-- 12.0 KB
base.h -rw-r--r-- 5.3 KB
bus.c -rw-r--r-- 32.0 KB
cacheinfo.c -rw-r--r-- 14.2 KB
class.c -rw-r--r-- 14.9 KB
component.c -rw-r--r-- 11.8 KB
container.c -rw-r--r-- 968 bytes
core.c -rw-r--r-- 58.6 KB
cpu.c -rw-r--r-- 11.9 KB
dd.c -rw-r--r-- 22.8 KB
devcoredump.c -rw-r--r-- 7.9 KB
devres.c -rw-r--r-- 24.4 KB
devtmpfs.c -rw-r--r-- 9.0 KB
dma-coherent.c -rw-r--r-- 8.8 KB
dma-contiguous.c -rw-r--r-- 8.0 KB
dma-mapping.c -rw-r--r-- 8.2 KB
driver.c -rw-r--r-- 5.6 KB
firmware.c -rw-r--r-- 614 bytes
firmware_class.c -rw-r--r-- 39.5 KB
hypervisor.c -rw-r--r-- 544 bytes
init.c -rw-r--r-- 776 bytes
isa.c -rw-r--r-- 3.7 KB
map.c -rw-r--r-- 3.3 KB
memory.c -rw-r--r-- 19.7 KB
module.c -rw-r--r-- 1.9 KB
node.c -rw-r--r-- 18.4 KB
pinctrl.c -rw-r--r-- 2.5 KB
platform-msi.c -rw-r--r-- 10.9 KB
platform.c -rw-r--r-- 36.1 KB
property.c -rw-r--r-- 31.1 KB
soc.c -rw-r--r-- 3.9 KB
syscore.c -rw-r--r-- 3.2 KB
topology.c -rw-r--r-- 4.1 KB
transport_class.c -rw-r--r-- 9.4 KB

back to top