Commit Graph

100 Commits

Author SHA1 Message Date
Scott Long
dc9efde5b7 Destroy mutexes on detach. 2004-08-12 05:31:17 +00:00
Scott Long
ba1d57e7d9 Change FREE() to free() 2004-08-12 05:09:17 +00:00
Scott Long
bb6fe253cc Remove the AAC_LOCK macros. They no longer abstract anything and only
obfuscate the code.  No functional differences.
2004-08-12 05:05:06 +00:00
Scott Long
3576af8fd7 Release the sync fib after the controller has been shut down. This also
releases the I/O lock instead of just leaking it.
2004-08-12 04:47:32 +00:00
Alexander Kabaev
eec256de79 Avoid casts as lvalues. 2004-07-28 06:21:53 +00:00
Poul-Henning Kamp
89c9c53da0 Do the dreaded s/dev_t/struct cdev */
Bump __FreeBSD_version accordingly.
2004-06-16 09:47:26 +00:00
Scott Long
03b5fe51bf Collapse sync fib locking into normal i/o locking. The former didn't
protect the registers so it was trivially possible for a sync command and
i/o command to fight each other and confuse the controller.  Make the
sync fib alloc/release functions inline and remove the somewhat worthless
AAC_SYNC_LOCK_FORCE flag.  Thanks to Adil Katchi for helping me to track
this down in RELENG_4.
2004-06-02 18:15:48 +00:00
Scott Long
614c22b2a2 Commit the correct version of the patch from last night. This fixes an
immediate panic when doing any i/o, and it closes a completion race.
2004-06-01 15:50:11 +00:00
Scott Long
397fa34f51 Collapse aac_map_command() into aac_startio(). Check the AAC_QUEUE_FRZN in
every iteration of aac_startio().  This ensures that a command that is
deferred for lack of resources doesn't immediately get retried in the
aac_startio() loop.  This avoids an almost certain livelock.
2004-06-01 05:32:26 +00:00
Scott Long
d8a0a47347 Remove the 'timeout' argument from aac_wait_command() as it isn't used and
never will be.  Update the XXX comment for this function to accurately reflect
why things are the way they are.
2004-04-14 19:11:29 +00:00
John Baldwin
6074439965 kthread_exit() no longer requires Giant, so don't force callers to acquire
Giant just to call kthread_exit().

Requested by:	many
2004-03-05 22:42:17 +00:00
Scott Long
75fba44b93 Revert the last commit. I don't know what I was thinking, but this change
definitely doesn't help any thing.
2004-02-25 05:41:44 +00:00
Poul-Henning Kamp
dc08ffec87 Device megapatch 4/6:
Introduce d_version field in struct cdevsw, this must always be
initialized to D_VERSION.

Flip sense of D_NOGIANT flag to D_NEEDGIANT, this involves removing
four D_NOGIANT flags and adding 145 D_NEEDGIANT flags.
2004-02-21 21:10:55 +00:00
Scott Long
41a1325df9 Fix a major brain-o. If the command needs to be put on the deferred queue,
take it off of the busy queue first.  This should fix the 'command is on
another queue' panic that showed up recently.
2004-02-21 05:32:40 +00:00
Poul-Henning Kamp
0b7ed341e1 Change the disk(9) API in order to make device removal more robust.
Previously the "struct disk" were owned by the device driver and this
gave us problems when the device disappared and the users of that device
were not immediately disappearing.

Now the struct disk is allocate with a new call, disk_alloc() and owned
by geom_disk and just abandonned by the device driver when disk_create()
is called.

Unfortunately, this results in a ton of "s/\./->/" changes to device
drivers.

Since I'm doing the sweep anyway, a couple of other API improvements
have been carried out at the same time:

The Giant awareness flag has been flipped from DISKFLAG_NOGIANT to
DISKFLAG_NEEDSGIANT

A version number have been added to disk_create() so that we can detect,
report and ignore binary drivers with old ABI in the future.

Manual page update to follow shortly.
2004-02-18 21:36:53 +00:00
Scott Long
ecd1c51f92 Do some small cleanups to comments and remove AACQ_COMPLETE definitions since
the completion queue is long-gone.
2004-02-07 17:40:38 +00:00
Scott Long
4102d44b77 If a command has to be deferred because there are no more resources for it
on the card, unmap it first.  This allows it to be picked up properly when
the queue gets kicked again.  This was the root problem for the lost command
(i.e. stuck in getblk/vinvalb) problem.  While here, panic if commands don't
map correctly instead of just silently ignoring the problem and dropping
command.  Also slow down the dynamic allocation of new commands.

It should be safe to go back into the aac waters.  Thanks to everyone who
suffered through this and provided good feedback.
2004-02-07 10:30:22 +00:00
Scott Long
a32a982d4a - Broaden the scope of locking in aac_command_thread() again to catch some
edge cases in the loop.

- Try to grab a command before dequeueing the bio from the bioq.  The old
  behaviour of requeuing deferred bios to the end of the bioq is arguably
  wrong.  This should be fixed in the future to check the bioq head without
  automatically dequeueing the bio.
2004-02-07 03:26:38 +00:00
Scott Long
9148fa21dd Take the plunge and make this driver be INTR_FAST. This re-arranges the
interrupt handler so that no locks are needed, and schedules the
command completion routine with a taskqueue_fast.  This also corrects the
locking in the command thread and removes the need for operation flags.

Simple load tests show that this is now considerably faster than FreeBSD 4.x
in the SMP case when multiple i/o tasks are running.
2004-01-30 07:04:39 +00:00
Seigo Tanimura
512824f8f7 - Implement selwakeuppri() which allows raising the priority of a
thread being waken up.  The thread waken up can run at a priority as
  high as after tsleep().

- Replace selwakeup()s with selwakeuppri()s and pass appropriate
  priorities.

- Add cv_broadcastpri() which raises the priority of the broadcast
  threads.  Used by selwakeuppri() if collision occurs.

Not objected in:	-arch, -current
2003-11-09 09:17:26 +00:00
Scott Long
3df780cf76 Fix a couple of bugs with AIF handling:
- Correct the logic for the AIF array index pointers so that correct slot is
   always looked at.
 - Copy the full FIB payload size when copying AIF's, not just the first 64
   bytes.

Thanks to Mirapoint, Inc, for pointing these problems out and offering a
solution.
2003-10-17 21:44:06 +00:00
Scott Long
c567467af2 Eliminate the use of a statically assign major number for the aac device. 2003-09-29 14:10:04 +00:00
Scott Long
7753acd255 Correctly wrap the producer queue index when dequeuing commands. This wasn't
a problem for command responses since we rarely ever filled the queue.
However, adapter-initiated commands have a much smaller queue and could
tickle this bug.  It's possible that this might fix the recently reported
problems with the aac-2120s, though I haven't been able to reproduce the
problem locally.

MFC-After: 1 day
2003-09-16 16:07:15 +00:00
Scott Long
8778f63db2 Commands submitted through the management interface won't have scatter/
gather lists.  Stop ignoring them and instead call the callback directly.
This unbreaks the management interface.
2003-09-01 20:44:18 +00:00
Scott Long
0bcbebd6df If ~ chars were pennies, I'd be pennyless. This should fix all of the
'command not in queue' panics.  Also fix a nearby style problem.
2003-09-01 00:20:29 +00:00
David E. O'Brien
aad970f1fe Use __FBSDID().
Also some minor style cleanups.
2003-08-24 17:55:58 +00:00
Scott Long
b88ffdc860 Make aac(4) compile cleanly on 64-bit machines. The code was already 64-bit
safe, but some (unneeded and/or harmless) downcasts were generating warnings.
The driver still is not endian-clean.
2003-08-19 21:31:45 +00:00
Scott Long
cd481291bd Handle the EINPROGRESS case of bus_dmamap_load() for data buffers. 2003-07-09 21:16:23 +00:00
Scott Long
204c0bef67 Fix a missing } that got dropped from the last commit. 2003-07-09 19:30:30 +00:00
Scott Long
4b00f859d0 Add a new quirk for cards that incorrectly interpret the amount of memory
in the system.  This might also have a small performance gain.
2003-07-09 19:19:16 +00:00
Scott Long
f6b1c44d1f Mega busdma API commit.
Add two new arguments to bus_dma_tag_create(): lockfunc and lockfuncarg.
Lockfunc allows a driver to provide a function for managing its locking
semantics while using busdma.  At the moment, this is used for the
asynchronous busdma_swi and callback mechanism.  Two lockfunc implementations
are provided: busdma_lock_mutex() performs standard mutex operations on the
mutex that is specified from lockfuncarg.  dftl_lock() is a panic
implementation and is defaulted to when NULL, NULL are passed to
bus_dma_tag_create().  The only time that NULL, NULL should ever be used is
when the driver ensures that bus_dmamap_load() will not be deferred.
Drivers that do not provide their own locking can pass
busdma_lock_mutex,&Giant args in order to preserve the former behaviour.

sparc64 and powerpc do not provide real busdma_swi functions, so this is
largely a noop on those platforms.  The busdma_swi on is64 is not properly
locked yet, so warnings will be emitted on this platform when busdma
callback deferrals happen.

If anyone gets panics or warnings from dflt_lock() being called, please
let me know right away.

Reviewed by:	tmm, gibbs
2003-07-01 15:52:06 +00:00
Scott Long
795d7dc093 When scanning for changed containers, don't assume that the controller
will respond in a sane manner.

Thanks to Petri Helenius <pete@he.iki.fi> for spotting this and pestering
me to fix it.
2003-06-03 02:10:55 +00:00
Scott Long
149af931ca Don't pass pointers to kernel data structures through 32-bit fields
as 64-bit architectures won't like this.  Use virtual array indexes
instead.  This *should* allow the driver to work on 64-bit platforms,
though it's still not endian clean.
2003-04-28 06:16:20 +00:00
Poul-Henning Kamp
891619a66d Use bioq_flush() to drain a bio queue with a specific error code.
Retain the mistake of not updating the devstat API for now.

Spell bioq_disksort() consistently with the remaining bioq_*().

#include <geom/geom_disk.h> where this is more appropriate.
2003-04-01 15:06:26 +00:00
Scott Long
b85f5808f8 Add the ability to send 64-bit scatter/gather elements to aac cards. This
is enabled when both the size of bus_addr_t > 4 and the card claims support.

Don't wake up the kthread to allocate more commands if we know that we've
already allocated the max number of commands.
2003-03-30 21:47:16 +00:00
Scott Long
a6d35632f5 Begin support for 64-bit address support and workarounds for newer cards:
- Add data structuress for doing 64-bit scatter/gather
	- Move busdma tag creations around so that only the parent is
	  created in aac_pci.c.
	- Retrieve the capabilities word from the firmware before setting
	  up command structures and tags. This allows the driver to decide
	  whether to do 64-bit commands, and if work-arounds are needed for
	  systems with >2GB of RAM.
	- Only enable the SCSI passthrough if it's enabled in the capabilities
	  word in the firmware.

This should fix problems with the 2120S and 2200S cards in systems with more
than 2GB of RAM.  Full 64-bit support is forthcoming.

MFC-After:	1 week
2003-03-26 17:50:11 +00:00
Scott Long
6f420b1756 Now that bus_dmamem_alloc() handles its Giant mutex requirements itself,
don't bother doing the same in the code that calls it.
2003-03-13 22:56:11 +00:00
Poul-Henning Kamp
60794e0478 Centralize the devstat handling for all GEOM disk device drivers
in geom_disk.c.

As a side effect this makes a lot of #include <sys/devicestat.h>
lines not needed and some biofinish() calls can be reduced to
biodone() again.
2003-03-08 08:01:31 +00:00
Poul-Henning Kamp
7ac40f5f59 Gigacommit to improve device-driver source compatibility between
branches:

Initialize struct cdevsw using C99 sparse initializtion and remove
all initializations to default values.

This patch is automatically generated and has been tested by compiling
LINT with all the fields in struct cdevsw in reverse order on alpha,
sparc64 and i386.

Approved by:    re(scottl)
2003-03-03 12:15:54 +00:00
Scott Long
9e2e96d803 The aac driver has evolved enough over the last few months that it no
longer resembles the 4.x version very much.  Garbage collect the legacy
bits.
2003-03-01 05:07:19 +00:00
Scott Long
ae54359608 Bring aac out from under Giant:
- the mutex aac_io_lock protects the main codepaths which handle queues and
  hardware registers.  Only one acquire/release is done in the top-half and
  the taskqueue.  This mutex also applies to the userland command path and
  CAM data path.
- Move the taskqueue to the new Giant-free version.
- Register the disk device with DISKFLAG_NOGIANT so the top-half processing
  runs without Giant.
- Move the dynamic command allocator to the worker thread to avoid locking
  issues with bus_dmamem_alloc().

This gives about 20% improvement in most of my benchmarks.
2003-02-26 04:46:21 +00:00
Scott Long
7874f606d5 Introduce a new taskqueue that runs completely free of Giant, and in
turns runs its tasks free of Giant too.  It is intended that as drivers
become locked down, they will move out of the old, Giant-bound taskqueue
and into this new one.  The old taskqueue has been renamed to
taskqueue_swi_giant, and the new one keeps the name taskqueue_swi.
2003-02-26 03:15:42 +00:00
Scott Long
7540e65e84 Move to 'struct disk*' API
Submitted by:	phk
2003-02-26 02:50:27 +00:00
Scott Long
65ac4ed65e On detach, don't remove the child from our list of children unless it has
successfully detached.
2003-02-20 23:13:48 +00:00
Scott Long
3cbd08b843 The completion queue is no longer used, so nuke its associated code
and data structures.
2003-02-20 08:51:16 +00:00
Scott Long
cb0d64b9b9 Fix a 64-bit bogon. The hardware command structure only has one 32 bit
field for holding driver-dependant data.  Instead of putting the pointer
to the driver command struct in there, take advantage of these structs
being a (virtually) contiguous array and just put the array index in the
field.
2003-02-19 23:33:56 +00:00
Scott Long
9c3a7fce37 Make the aac driver be INTR_MPSAFE. Once the interrupt handler determines
that a command completion happened, all further processing is deferred to
a taskqueue.  The taskqueue itself runs implicetely under Giant, but we
already used a taskqueue for the biodone() processing, so this at least
saves the contesting of Giant in the interrupt handler.
2003-02-19 21:58:34 +00:00
Scott Long
8480cc638e o Move the cleanup of the fib maps into aac_free_commands() so as to
retain symetry with aac_alloc_commans().  Since aac_alloc_commands()
  allocates fib maps and places them onto the fib lists, aac_free_commands()
  should reverse those operations.

o Combine two ifs with the same body with an ||.

o Switch from uintptr_t to uint32_t for fib map load operations.
  The target is a uint32_t so using this type for the map load call
  avoids an extra cast.  uintptr_t should only be used when you need
  an "int sized the same as the machine's poitner size" which is not
  the case here.

o Removed the commented out M_WAITOK flag in the allocation in
  aac_alloc_commands().  The kernel will only block in the allocator
  if it can grow the size of the kernel.  This usually results in a
  page-out which could involve this aac device.  Thus, sleeping here
  could deadlock the machine.  Assuming this operation is occurring outside
  of attach time, we have enough fibs to operate anyway, so waiting for
  fibs to free up is okay if not optimal.

o In aac_alloc_commands(), if we cannot dmamem_alloc additional fib
  space, free the fib map.

o In aac_alloc_commands(), if we cannot create per-command dmamaps, don't
  lose track of the fib map that is mapping all of the commands that we
  have already released into the free pool.  Instead, just cut out of
  the loop and modify aac_free_commands to not attempt to free maps that
  have not been allocated.

o Don't use a magic number when pre-allocating fibs.

o Use PAGE_SIZE to allocate in page sized chunks instead of an
  architecture specific constant.

Submitted by: gibbs
2003-02-19 21:38:29 +00:00
Scott Long
ffb37f33f8 Implement a new dynamic command allocator. FIBs are allocated in 1 page
blocks now, which should eliminate problems with the driver failing to
attach due to insufficient contiguous RAM.  Allow the FIB pool to grow
from the default of 128 to the max of 512 as demand grows.  Also pad the
adapter init struct to work around the 2120/2200 DMA bug now that there
is no longer a FIB slab.
2003-02-10 00:34:24 +00:00
Scott Long
ddb8683e59 Teach the CAM module how to deregister itself so it can be unloaded. 2003-02-06 03:31:02 +00:00