actually need it. This should improve performance on those versions of Linux
that do not need this.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@2036 d57e44dd-8a1f-0410-8b47-8ef2f437770f
there seems to be a problem with the Linux SCSI stack because when we delete a
LUN on the target, it does not get deleted from Linux. Still investigating.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@1947 d57e44dd-8a1f-0410-8b47-8ef2f437770f
TRACE_DBG("Created tid '%08lX'", (unsigned long)&tr_id[4]);
is wrong. All it will do is print the address of the fifth byte in the tr_id.
It should cast to an unsigned long * and deref it, at least. However, I have
anothe change that prints out the tr_id as a SAS address.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@1934 d57e44dd-8a1f-0410-8b47-8ef2f437770f
use an OUI that I think is not currently in use, so it seems unlikely that we
will clash with anyone else, but you never know. I might have to change it
later.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@1928 d57e44dd-8a1f-0410-8b47-8ef2f437770f
versions of Linux it was an unsigned char, and a value of 260 means 4 after a compiler warning. Added a comment to describe what is going on as well. Since
this is done very infequently it should not cause performance problems and
means that it will not silently fail to work if people try to use it on kernels
earlier than 2.6.26. If I figure out what kernel version the change was made
in we could use an ifdef.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@1852 d57e44dd-8a1f-0410-8b47-8ef2f437770f
- ISCSI preliminary completed commands handling fixes
- Write residuals cleanups and fixes
- Interface to handle residuals and received from initiators data changed
- BIDI cleanups and fixes.
- All BIDI IN fields and functions renamed to OUT to be in line with the in-kernel convention
+
Docs updated
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@1780 d57e44dd-8a1f-0410-8b47-8ef2f437770f
is built via the kbuild infastructure. In kernel version 2.6.24 (commit
f77bf01425b11947eeb3b5b54685212c302741b8) the kbuild variable ccflags-y has
been introduced and the use of EXTRA_CFLAGS in kbuild makefiles has been
deprecated. The patch below converts SCST's in-tree makefiles such that these
use ccflags-y instead of EXTRA_CFLAGS when generating kernel patches for
kernel 2.6.24 or later.
Additionally, this patch removes the "default n" from the file
scst/kernel/in-tree/Kconfig.scst since this statement is superfluous.
The patch below has been tested by verifying the output of the following
command:
for ((v=23;v<=32;v++)); do rm -rf re* && scripts/run-regression-tests 2.6.$v; done
Signed-off-by: Bart Van Assche <bart.vanassche@gmail.com>
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@1447 d57e44dd-8a1f-0410-8b47-8ef2f437770f
complaints:
- code indent should use tabs where possible.
- trailing whitespace.
- else should follow close brace '}'.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@1074 d57e44dd-8a1f-0410-8b47-8ef2f437770f
now report functions that are called with a lock held but that are missing
__acquires() and __releases() annotations. The patch below adds such
annotations and makes the following sparse warnings disappear:
drivers/scst/iscsi-scst/iscsi.c:2861:9: warning: context imbalance in 'iscsi_check_send_delayed_tm_resp' - unexpected unlock
drivers/scst/iscsi-scst/nthread.c:1614:17: warning: context imbalance in 'scst_do_job_wr' - unexpected unlock
drivers/scst/iscsi-scst/iscsi.c:2861:9: warning: context imbalance in 'iscsi_check_send_delayed_tm_resp' - unexpected unlock
drivers/scst/iscsi-scst/nthread.c:1614:17: warning: context imbalance in 'scst_do_job_wr' - unexpected unlock
drivers/scst/scst_local/scst_local.c:554:30: warning: context imbalance in 'scst_local_queuecommand' - unexpected unlock
The patch below has been tested by comparing the output of
"scripts/run-regression-tests 2.6.30.3" before and after the patch was applied.
Signed-off-by: Bart Van Assche <bart.vanassche@gmail.com>
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@982 d57e44dd-8a1f-0410-8b47-8ef2f437770f
can send a request as a non scatterlist. If we get one, we create a single
element scatterlist and hand that on to SCST.
This costs some extra space in the structure I use to keep track of requests
and the done function, but only on those versions for which it is possible.
For real kernels it is scatterlists all the way down young man.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@922 d57e44dd-8a1f-0410-8b47-8ef2f437770f
2.6.24, 2.6.25.4, 2.6.27.x and 2.6.29. I have also tested under 2.6.18 on a
CentOS 5.3 system.
There might still be problems with some 2.6.18 systems, and I will commit more
changes as the need arises.
I have only tested that scst_local builds and loads with these changes. I have
not tested any devices as yet.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@899 d57e44dd-8a1f-0410-8b47-8ef2f437770f
While writing DKMS support for the SCST and iSCST-SCST kernel modules, I ran
into a (common) problem with regard to out-of-tree build detection. A common
method to detect out-of-tree building is by using the KERNELRELEASE variable.
The usage of this variable poses a problem when using DKMS. DKMS itself uses
the KERNELRELEASE variable to specify the kernel for which the driver to
build. A suggested alternative would to use the PATCHLEVEL variable as part
of the detection mechanism instead. I attached a patch reflecting this
suggested change.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@720 d57e44dd-8a1f-0410-8b47-8ef2f437770f
- Docs about limitation of having initiator and target on the same host updated + cleanups
- Minor local thread storage improvements
- TRACE_MGMT_MINOR excluded from the default set of trace flags to not confuse people
- Dedicated kmem_cache for blockio created
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@662 d57e44dd-8a1f-0410-8b47-8ef2f437770f
It, hopefully, fixes compilation problems on CentOS 5.2 as well as on vanilla kernels <2.6.28
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@638 d57e44dd-8a1f-0410-8b47-8ef2f437770f
- In scst_local new experimental compile-time option to force direct processing added
- Attempt in SCST structures to separate read-mostly from read-write data to decrease cache ping-pong between CPUs
- In scst_vdisk new module parameter num_threads added to specify a number of threads for each vdisk/vcdrom. Default is 5.
- Debug logging for failed digests in iSCSI-SCST improved
- Docs updates/cleanups
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@635 d57e44dd-8a1f-0410-8b47-8ef2f437770f
