Commit Graph

102 Commits

Author SHA1 Message Date
scottl
dbe9aeaa21 Work again to fix the interrupt masking problems. We now recognize
that there are 3 different interrupt enable bits, 2 for different
families of cards, and 1 for when MSI is used.  Also apply a big
hammer backstop for cards that aren't recognized.  This should fix
all of the interrupt issues at boot.
2008-08-29 01:23:16 +00:00
scottl
02392e305a Correctly set the interrupt enable and disable bits. The previous
code interfered with Performant mode and legacy interrupts.  Also
remove a register read operation on the Simplq code that was
effectively a time-wasting no-op.
2008-08-02 13:04:26 +00:00
scottl
c326e0792a A number of significant enhancements to the ciss driver:
1.  The FreeBSD driver was setting an interrupt coalesce delay of 1000us
for reasons that I can only speculate on.  This was hurting everything
from lame sequential I/O "benchmarks" to legitimate filesystem metadata
operations that relied on serialized barrier writes.  One of my
filesystem tests went from 35s to complete down to 6s.

2.  Implemented the Performant transport method.  Without the fix in
(1), I saw almost no difference.  With it, my filesystem tests showed
another 5-10% improvement in speed.  It was hard to measure CPU
utilization in any meaningful way, so it's not clear if there was a
benefit there, though there should have been since the interrupt handler
was reduced from 2 or more PCI reads down to 1.

3.  Implemented MSI-X.  Without any docs on this, I was just taking a
guess, and it appears to only work with the Performant method.  This
could be a programming or understanding mistake on my part.  While this
by itself made almost no difference to performance since the Performant
method already eliminated most of the synchronous reads over the PCI
bus, it did allow the CISS hardware to stop sharing its interrupt with
the USB hardware, which in turn allowed the driver to become decoupled
from the Giant-locked USB driver stack.  This increased performance by
almost 20%.  The MSI-X setup was done with 4 vectors allocated, but only
1 vector used since the performant method was told to only use 1 of 4
queues.  Fiddling with this might make it work with the simpleq method,
not sure.  I did not implement MSI since I have no MSI-specific hardware
in my test lab.

4.  Improved the locking in the driver, trimmed some data structures.
This didn't improve test times in any measurable way, but it does look
like it gave a minor improvement to CPU usage when many
processes/threads were doing I/O in parallel.  Again, this was hard to
accurately test.
2008-07-11 21:20:51 +00:00
ps
2e4980c9ce Add support for the P212, P410, P410i, P411, and P812 HP Smart Array
controllers.

Submitted by:  Scott Benesh at HP
2008-06-10 17:51:51 +00:00
ps
6fea0a35ac Remove a check that didn't allow > 12 byte CDB's to be issued to
ciss.  This should allow volumes > 2TB to work.

Reported by:	Emil Mikulic
2008-05-16 08:27:02 +00:00
iwasaki
9b9b90a013 Add `hw.ciss.nop_message_heartbeat' tunable (default disabled) for
NOP-message polling in ciss_periodic().
Note that setting the tunable to non-zero can be workaround only for
`ADAPTER HEARTBEAT FAILED' problem, and may freeze the system w/o
the problem.

Reviewed by:	scottl
Reported by:	Attila Nagy
MFC after:	3 days
2008-02-10 06:21:52 +00:00
iwasaki
b46ea5a5ec Don't repeat error logging about NOP message sending if
ciss_report_request() return an error (which is most likely data
underrun).

Noticed by:	Mark Atkinson
MFC after:	1 week
2008-01-28 16:21:34 +00:00
iwasaki
9d42dea98f Fix NOP message sending in ciss_periodic() which causes panic with
option INVARIANTS.

Reviewed by:	simokawa
Tested by:	noriyosi_kawano
MFC after:	1 week
2007-11-05 13:54:23 +00:00
julian
51d643caa6 Rename the kthread_xxx (e.g. kthread_create()) calls
to kproc_xxx as they actually make whole processes.
Thos makes way for us to add REAL kthread_create() and friends
that actually make theads. it turns out that most of these
calls actually end up being moved back to the thread version
when it's added. but we need to make this cosmetic change first.

I'd LOVE to do this rename in 7.0  so that we can eventually MFC the
new kthread_xxx() calls.
2007-10-20 23:23:23 +00:00
iwasaki
f337cc48b9 Add NOP-message polling to ciss_periodic().
Disable adapter by detecting adapter is dead.

Tested by:	Masaki YATSU(on RELENG_6)
Reviewed by:	scottl
MFC after:	1 week
2007-10-13 05:45:45 +00:00
scottl
805629b231 Fix a mistake made in the MPSAFE commit that caused CAM to serialize requests
to the controller.
2007-10-12 17:03:41 +00:00
scottl
ff584e70fa Prepare for future integration between CAM and newbus. xpt_bus_register
now takes a device_t to be the parent of the bus that is being created.
Most SIMs have been updated with a reasonable argument, but a few exceptions
just pass NULL for now.  This argument isn't used yet and the newbus
integration likely won't be ready until after 7.0-RELEASE.
2007-06-17 05:55:54 +00:00
scottl
1c9a6e5327 Satisfy witness during shutdown 2007-06-05 05:03:13 +00:00
scottl
2b1170b7f0 Eliminate M_TEMP. 2007-05-14 22:01:03 +00:00
scottl
db629f0c2f Streamline locking in ciss_free() 2007-05-02 04:44:31 +00:00
scottl
de5eff9fe0 MPSAFE ciss driver 2007-05-01 05:13:15 +00:00
scottl
7605c12c0f Remove Giant from CAM. Drivers (SIMs) now register a mutex that CAM will
use to synchornize and protect all data objects that are used for that
SIM.  Drivers that are not yet MPSAFE register Giant and operate as
usual.  RIght now, no drivers are MPSAFE, though a few will be changed
in the coming week as this work settles down.

The driver API has changed, so all CAM drivers will need to be recompiled.
The userland API has not changed, so tools like camcontrol do not need to
be recompiled.
2007-04-15 08:49:19 +00:00
piso
6a2ffa86e5 o break newbus api: add a new argument of type driver_filter_t to
bus_setup_intr()

o add an int return code to all fast handlers

o retire INTR_FAST/IH_FAST

For more info: http://docs.freebsd.org/cgi/getmsg.cgi?fetch=465712+0+current/freebsd-current

Reviewed by: many
Approved by: re@
2007-02-23 12:19:07 +00:00
maxim
996e6b0fcb o ciss.ko depends on cam and pci.
PR:		kern/105989
Submitted by:	nork
MFC after:	1 month
2006-11-30 15:14:17 +00:00
mjacob
19b599de4a 2nd and final commit that moves us to CAM_NEW_TRAN_CODE
as the default.

Reviewed by multitudes.
2006-11-02 00:54:38 +00:00
mjacob
7e1db24f98 The first of 3 major steps to move the CAM layer forward to using
the CAM_NEW_TRAN_CODE that has been in the tree for some years now.

This first step consists solely of adding to or correcting
CAM_NEW_TRAN_CODE pieces in the kernel source tree such
that a both a GENERIC (at least on i386) and a LINT build
with CAM_NEW_TRAN_CODE as an option will compile correctly
and run (at least with some the h/w I have).

After a short settle time, the other pieces (making
CAM_NEW_TRAN_CODE the default and updating libcam
and camcontrol) will be brought in.

This will be an incompatible change in that the size of structures
related to XPT_PATH_INQ and XPT_{GET,SET}_TRAN_SETTINGS change
in both size and content. However, basic system operation and
basic system utilities work well enough with this change.

Reviewed by:	freebsd-scsi and specific stakeholders
2006-10-31 05:53:29 +00:00
ps
7d671abdfe Spin until a request structure is available in the ioctl path. 2006-05-30 06:42:02 +00:00
phk
ef310efff8 Since DELAY() was moved, most <machine/clock.h> #includes have been
unnecessary.
2006-05-16 14:37:58 +00:00
ps
d4df787e9e Free another memory leak when dealing with disk notification. 2006-04-20 03:05:02 +00:00
ps
181e281264 Free some previously leaked memory on module unload. 2006-04-19 19:56:10 +00:00
ps
56b29d3d85 It seems ciss should ignore overrun and underrun on a SCSI INQUIRY
command.  This fixes some weird booting issues on newer versions
of the firmware on the MSA20.

Reported by:	Philippe Pegon <Philippe dot Pegon at crc dot u-strasbg dot fr>
2005-12-16 06:50:55 +00:00
ps
666d51b0ba Update PCI ids to add the E200, E200i, P400, and P400i storage
controllers.  Remove the E400 since it is not a real product.

Submitted by:	HP
2005-11-11 16:45:33 +00:00
ps
66d9331654 There's no reason to check the valence. This allows ciss to work
on the P600.
2005-08-09 20:53:51 +00:00
nyan
0fce92f5c4 Remove bus_{mem,p}io.h and related code for a micro-optimization on i386
and amd64.  The optimization is a trivial on recent machines.

Reviewed by:	-arch (imp, marcel, dfr)
2005-05-29 04:42:30 +00:00
ps
e04ec83e80 Support passthru ioctl commands from 32bit binaries. 2005-05-18 05:31:34 +00:00
ps
9c8b8647e1 Add support for the P600 and name the E400. 2005-04-28 14:40:23 +00:00
ps
b7632e9a5b Provide a way to soft reset a proxy controller such as an MSA20 or
MSA500.  This is useful if you need to reset one of the storage
arrays on reboot.
2005-04-19 06:11:16 +00:00
sam
61bf6bf48d handle ciss_lookup failure
Noticed by:	Coverity Prevent analysis tool
2005-03-29 01:44:17 +00:00
imp
a2e81fc93f Use BUS_PROBE_DEFAULT for pci probe return value 2005-03-05 18:30:12 +00:00
scottl
8919700643 - Remove dead code.
- Protect against negative values as array indexes.

Submitted by: Coverity Prevent analysis tool
2005-02-21 05:17:29 +00:00
ps
da1618943d Remove 6422, V100 and add the P600.
Submitted by:	John Cagle
2005-01-20 22:51:38 +00:00
peter
c86a9c8396 ciss's interrupt handler was missing the INTR_ENTROPY flag. 2004-08-16 23:13:16 +00:00
ps
5bf78408ff Clean up structure packing. 2004-07-29 08:06:22 +00:00
kan
6cabb1fb0d Make __packed attribute apply to structs themselves, not their typedefs.
GCC 3.4.2 issues warning about that.
2004-07-28 06:19:25 +00:00
ps
6213728da1 Name non-physical devices. 2004-07-28 06:18:05 +00:00
ps
069fb92824 Reset the update flag when scanning for new devices. 2004-06-22 23:05:06 +00:00
ps
a1bdd1093e Use the maximum physical bus when rescanning those targets. This
fixes a panic on cards which do not have any drives attached.
2004-06-22 17:39:57 +00:00
ps
a5b5031c52 on media exchange, update/rescan the drives. This allows a volume
in a failed stated to come back on-line without a reboot.
2004-06-21 22:57:10 +00:00
ps
f0bec239ac There is no need to call ciss_media_accept async anymore since all
notify events are handled in a kthread.
2004-06-21 21:56:57 +00:00
scottl
3eaa7bd6e0 Add SCSI passthrough support to CISS. This allows devices like tape drives
that are on a CISS bus to be exported up to CAM and made available as normal
devices.  This will typically add one or two buses to CAM, which will be
numbered starting at 32 to allow room for CISS proxy buses.  Also, the CISS
firmware usually hides disk devices, but these can also be exposed as 'pass'
devices if you set the hw.ciss.expose_hidden_physical tunable.

Sponsored by:	Tape Laboratories, Inc.
MFC After: 3 days
2004-06-21 20:18:40 +00:00
phk
dfd1f7fd50 Do the dreaded s/dev_t/struct cdev */
Bump __FreeBSD_version accordingly.
2004-06-16 09:47:26 +00:00
ps
6e7aa9fe96 If a device is not ONLINE, do not attempt to submit any commands
or I/O to the volume.  This solves a panic when removing a disk in
a RAID 0 volume.

Reported by:	le
2004-06-15 19:40:47 +00:00
ps
c2d8798214 Document the 64bit version of blocks_to_recover for logical drive status.
Pad the struct to 1024 bytes as defined in the firmware spec.
2004-06-05 18:31:04 +00:00
ps
4e0becfbfc Add pci id's for the SmartArray 6422 and V100 controllers. Also
add a whole bunch of pci id's for future controllers.

Submitted by:	John Cagle <first.last@hp.com>
2004-06-05 18:12:56 +00:00
ps
246b19df64 Update logical drive structure 2004-06-04 17:22:13 +00:00