Commit Graph

68 Commits

Author SHA1 Message Date
Nate Lawson
31bef809aa Revert two changes I was testing regarding polling delay. 2005-12-06 14:51:55 +00:00
Nate Lawson
c4a9fa4529 Add KTR support and move some performance debugging variables in the EC
to KTR.  We're reusing the KTR_DEV level.
2005-12-06 14:47:28 +00:00
David E. O'Brien
2a191126de Canonize the include of acpi.h. 2005-09-11 18:39:03 +00:00
Nate Lawson
a6761eb377 If there is a problem during probe, be sure to free up any resources
allocated for it.  The normal exit case handles this correctly so we
use it as well for errors.

Submitted by:	pjd
Obtained from:	Coverity Prevent
2005-05-12 22:35:08 +00:00
Mark Santcroos
b3919c8d96 Use AcpiUtStrupr() instead of strupr() as the latter will disappear in
future versions of acpica.

MFC after:	2 weeks
2005-04-14 06:50:13 +00:00
Nate Lawson
e33bea8d9e Add the acpi_ec_read and write methods. This allows an external driver
(like an EC/SMbus controller) to access the EC address space.  Access
is synchronized by the EcLock/Unlock routines in EcSpaceHandler().

Tested by:	Hans Petter Selasky
2005-03-20 01:27:27 +00:00
Nate Lawson
1395b555de Since the GPE handler is directly called by ACPI-CA and it may have unknown
locks held, specify the ACPI_ISR flag to keep it from acquiring any more
mutexes (which could potentially sleep.)  This should fix "could sleep"
warning messages on the following path:

    msleep()
    AcpiOsWaitSemaphore()
    AcpiUtAcquireMutex()
    AcpiDisableGpe()
    EcGpeHandler()
    AcpiEvGpeDispatch()
    AcpiEvGpeDetect()
    AcpiEvGpeDetect()
    AcpiEvSciXruptHandler()
2005-02-21 23:38:41 +00:00
Warner Losh
098ca2bda9 Start each of the license/copyright comments with /*-, minor shuffle of lines 2005-01-06 01:43:34 +00:00
Nate Lawson
d05fa56bb3 Remove trailing whitespace. 2004-12-27 05:36:47 +00:00
Nate Lawson
f4b7de154d MPSAFE locking
* Use the common serialization macros instead of rolling our own.
* Increase the coverage of the lock in EcSpaceHandler() to cover the entire
  loop to avoid dropping the lock when reading more than one byte.
2004-08-13 06:21:53 +00:00
Nate Lawson
5a66986def Remove duplicate FreeBSD id. 2004-07-02 20:15:53 +00:00
Nate Lawson
99eb9c8b3b Get rid of the strict aliasing error by retrieving the ECDT via a table
header pointer and then casting it to the ecdt pointer.  This fixes the
-O2 build.  I'm unsure what changed recently to reveal this error since
this code has been unchanged for months.
2004-07-02 19:17:27 +00:00
Nate Lawson
b0eefa38f9 Rework the code that waits for a response from the EC. Use an sx lock
instead of a mutex so we do not unblock it in msleep().  If we do this,
another event could occur, resetting the status register since reads
reset it.  While I'm here, remove the backoff approach.  Instead, sleep
in 10 ms chunks for up to the configured timeout using either DELAY (if
we aren't booted yet) or tsleep.

Help from:	dillon
Tested by:	Andrew Thompson andy AT fud.org.nz
2004-07-01 00:51:31 +00:00
Nate Lawson
340a7f6a9b Disable the EC GPE in the shutdown path. This is correct but is not known
to fix any bug.
2004-06-30 16:00:20 +00:00
Nate Lawson
5fcc8a587e Use the acpi_id_probe() method instead of acpi_MatchHid(), which is now
static.
2004-06-29 19:02:27 +00:00
Nate Lawson
636af3826f While booting, use DELAY() for waiting for the EC to respond instead of
msleep().  Until we're finally up, msleep is a no-op.

Explained by:	peter
2004-06-28 20:41:30 +00:00
Nate Lawson
9123341378 Add support to ACPI to manage its own resources. Previously, resource
allocation was passed up to nexus.  Now, we probe sysresource objects and
manage the resources they describe in a local rman pool.  This helps
devices which attach/detach varying resources (like the _CST object) and
module loads/unloads.  The allocation/release routines now check to see if
the resource is described in a child sysresource object and if so,
allocate from the local rman.  Sysresource objects add their resources to
the pool and reserve them upon boot.  This means sysresources need to be
probed before other ACPI devices.

Changes include:
* Add ordering to the child device probe.  The current order is:  system
resource objects, embedded controllers, then everything else.
* Make acpi_MatchHid take a handle instead of a device_t arg.
* Replace acpi_{get,set}_resource with the generic equivalents.
2004-06-13 22:52:30 +00:00
Poul-Henning Kamp
fe12f24bb0 Add missing <sys/module.h> includes 2004-05-30 20:08:47 +00:00
Nate Lawson
f3fc4f8b4a Changes to implement 20040514:
* Add calls to AcpiSetGpeType.  We use wake/run as the type for lid and
button switches since wake-only causes Thinkpads to immediately wake on
the second suspend.  Note that with wake/run, some systems return both
wake and device-specific notifies so we don't register for system notifies
for lid and button switches.
* Remove the hw.acpi.osi_method tunable since it is not needed.
* Always print unknown notifies for all types.
* Add more cleanup for the EC if it fails to attach.
* Use the GPE handle now that we parse it.  This allows GPEs to be defined
in AML GPE blocks.
* Always use ACPI_NOT_ISR since it's ok to acquire a mutex in our thread
which processes queued requests.
2004-05-25 02:47:35 +00:00
Nate Lawson
0025fb0f4f Add support for GPE being a package of { reference, gpe bit }.
Rework the ECDT probe to pass all the parameters in a temporary struct.
Note why we are mostly ok evaluating _GLK so early.
2004-05-14 04:17:56 +00:00
Nate Lawson
c868ac7d12 Add support for retrieving _GLK in the ECDT probe. Now we no longer always
use the global lock at the beginning of the ECDT probe.  Instead, we use
the handle from the ECDT to call _GLK.  Also, unify the device description.
2004-05-13 03:17:08 +00:00
Nate Lawson
b7d13479aa Update the name for edge triggered for the 20040402 import. 2004-04-14 02:20:35 +00:00
Nate Lawson
64278df5e0 Add MODULE_DEPEND entries so some of these drivers can eventually be
loaded separately from ACPI (i.e., embedded use).
2004-04-09 18:14:32 +00:00
Nate Lawson
5f96beb9e0 Convert callers to the new bus_alloc_resource_any(9) API.
Submitted by:	Mark Santcroos <marks@ripe.net>
Reviewed by:	imp, dfr, bde
2004-03-17 17:50:55 +00:00
Nate Lawson
4ed391b8d7 Fix an off-by-one error and rework our EC space handler. Writing to address
0xFF would fail previously as AE_BAD_PARAMETER.  It's unknown if this caused
any actual problems.
2004-03-04 05:58:50 +00:00
Nate Lawson
c310653ea1 Change to acpi_{Get,Set}Integer to provide both methods. Convert all
callers to the new API.

Submitted by:	Mark Santcroos <marks@ripe.net>
2004-03-03 18:34:42 +00:00
Nate Lawson
16844a97c8 Delete the region we are passed if that is the requested operation.
This should fix the problem with removing an address space handler
although we don't currently use that capability so it's unlikely anyone
saw this problem.
2004-01-03 02:01:39 +00:00
Nate Lawson
656b9dd5c3 Consistently print attach messages. 2003-09-26 05:24:55 +00:00
Nate Lawson
c87cfaedf2 Bump the EC timeout from 50 to 100 ms. I believe the underlying issue is
global lock contention as symptoms only appear under heavy load (i.e. the
nightly periodic run).
2003-09-15 21:20:55 +00:00
David E. O'Brien
aad970f1fe Use __FBSDID().
Also some minor style cleanups.
2003-08-24 17:55:58 +00:00
Nate Lawson
a329ebca91 Add and document the hw.acpi.ec.poll_timeout tunable. 2003-07-30 16:22:53 +00:00
Nate Lawson
44904c4020 Add an informational debugging printf of the maximum time spent in
EcEventWait().  If you get AE_HARDWARE_NO_RESPONSE errors, please enable
this info by setting hw.acpi.verbose=1 in loader.conf.
2003-07-25 16:49:46 +00:00
Nate Lawson
c72508e283 Do not call acpi_MatchHid() for all probe cases since it accesses the
namespace.  To compensate for it only being used in the !ECDT case, use
a more robust approach to indicate a device was probed via ECDT by setting
the private ivar to be &acpi_ec_devclass.  Without the acpi_MatchHid() call
now, it might have been possible for a non-EC device to have had its magic
match our previous flag.

Pointed out by:		takawata
2003-07-20 21:12:21 +00:00
Nate Lawson
3a371f32b3 Close a race condition by passing status retrieved via a non-SCI call
to EcGpeQueryHandler on to any waiting threads through the softc.  Similar
behavior was in the original version.

Also:
* Merge EcQuery into EcGpeQueryHandler to simplify locking
* Hold EcLock from the initial read of the CSR down to the wakeup or
  until after the query command has been processed.
* ec_gpebit only needs to be a UINT8
2003-07-20 21:11:32 +00:00
Nate Lawson
f8335e3a97 Add ECDT (ACPI 2.0) support. This allows the EC to be enabled before the
namespace has been evaluated.  Machines with ACPI 2.0 expect this behavior
and have AML which calls EC functions early in the boot process.  If the
ECDT is not available, fall back to original probe behavior.

Other minor changes:
* Add GPE bit and GLK usage to the device announcement
* Always use the global lock in the ECDT case, but potentially downgrade to
  not using it if _GLK is 0 once the namespace is available.  This is
  announced with "Changing GLK from 1 to 0"
* Remove the acpi_object_list definitions which were earlier deprecated

Ideas from:	takawata
2003-07-20 00:48:38 +00:00
Nate Lawson
ff40920e9e Change the msleep part of EcWaitEvent to be a separate loop, fixing a
problem that for some very slow ECs (~6 ms occasionally) causes a
timeout.  Also finish resource cleanup in the error case in attach.

Tested by:	ume
2003-07-15 19:24:36 +00:00
Nate Lawson
6fca9360db Update code to work with 0619 dist
* Use ACPI_BUFFER as the type for AcpiGetObjectInfo
* Remove AcpiEnableEvent/AcpiClearEvent for ACPI_EVENT_FIXED (power/sleep
  buttons) as they are no longer needed
* Change calls to use the new GPE functions
* Add AcpiOs*Lock functions
2003-07-13 22:57:16 +00:00
Nate Lawson
1f04e8f501 Rewrite much of the embedded controller driver.
* Always use polled mode.  The intr approach did not work for many
  controllers and required the hw.acpi.ec.event_driven workaround.
* Only use an edge (not level) triggered GPE handler
* Add sc->ec_mtx for locking operations to a single EC.  There were
  many race conditions earlier between an SCI event and EcRead/Write.
* Use 1 ms as the global lock timeout
* Only acquire global lock if _GLK != 0
* Update EcWaitEvent to use an incremental backoff delay in its
  poll loop.  Wait 50 ms max instead of 10.  Most ECs respond
  in < 5 us (50 us when heavily loaded).  However, some time out
  occasionally even with a 10 ms timeout.  For delays past 1 ms, use
  msleep instead of DELAY to give SCI interrupts a chance to occur.
* Add EcCommand to send a command and wait for the appropriate event.
* The hw.acpi.ec.event_driven tunable is no longer applicable and
  has been removed.

Ideas from:	Linux
2003-07-10 17:22:46 +00:00
Nate Lawson
464c662efa Whitespace cleanups (verified with MD5 1832ce5ede6c58c844dfb53fd48f0eea).
Also clarify a macro expansion with an XXX.  I will be fixing access to
Acpi Global Lock in a subsequent commit.
2003-06-27 21:57:42 +00:00
Mitsuru IWASAKI
21b5fd3cfd Resolve conflicts arising from the ACPI CA 20021118 import. 2002-11-27 18:09:20 +00:00
Mitsuru IWASAKI
4690674e12 Add new loader tunable hw.acpi.ec.event_driven so that we control
EcWaitEventIntr()'s behavior.
Some EC controllers require this for working properly (including
sleep/wakeup).
2002-10-24 10:56:33 +00:00
John Baldwin
fc0ea94a7d Catch up to changes in acpivar.h to add support for using ACPI on
4-stable systems.

Sponsored by:	The Weather Channel
2002-10-16 17:28:53 +00:00
Takanori Watanabe
4ec593593a Make interrupt driven EC transaction optional. 2002-07-02 03:38:07 +00:00
Peter Wemm
b4a052380f Brutally deal with __func__ being 'const char *' on gcc-3.1. 2002-05-19 06:16:47 +00:00
Mike Smith
dbd0058a4e Match namespace cleanup changes in ACPI CA 20020217 update.
Use ACPI_SUCCESS/ACPI_FAILURE consistently.

The ACPI global lock acquire takes a timeout value.  I'm not sure what
we should do about timeouts on it; a deadlock against this lock is
catastrophic.
2002-02-23 05:26:31 +00:00
Takanori Watanabe
f8372ade81 Fix error handling.
PR:30665
Submitted by:TOMITA Yoshinori <yoshint@flab.fujitsu.co.jp>
2002-01-31 01:23:22 +00:00
Mike Smith
3273b00523 Staticise devclasses and some unnecessarily global variables. 2002-01-08 06:46:01 +00:00
Mitsuru IWASAKI
c573e654b7 Add OS layer ACPI mutex and threading support.
- Temporary fix a bug of Intel ACPI CA core code.
 - Add OS layer ACPI mutex support.  This can be disabled by
   specifying option ACPI_NO_SEMAPHORES.
 - Add ACPI threading support.  Now that we have a dedicate taskqueue for
   ACPI tasks and more ACPI task threads can be created by specifying option
   ACPI_MAX_THREADS.
 - Change acpi_EvaluateIntoBuffer() behavior slightly to reuse given
   caller's buffer unless AE_BUFFER_OVERFLOW occurs.  Also CM battery's
   evaluations were changed to use acpi_EvaluateIntoBuffer().
 - Add new utility function acpi_ConvertBufferToInteger().
 - Add simple locking for CM battery and temperature updating.
 - Fix a minor problem on EC locking.
 - Make the thermal zone polling rate to be changeable.
 - Change minor things on AcpiOsSignal(); in ACPI_SIGNAL_FATAL case,
   entering Debugger is easier to investigate the problem rather than panic.
2001-12-22 16:05:41 +00:00
Mike Smith
76f2b644fd Synch with minor changes in the ACPI CA 20011120 snapshot. 2001-11-28 04:36:29 +00:00
Mitsuru IWASAKI
6971b3c7d1 Cleanups of verbose printing. All the messages for the debugging is
disabled unless verbose flag is set.  Also fix some messages in terms
of English.
The critical messages and error messages in probe/attach routine are
unchanged by this commit.
2001-11-18 18:12:07 +00:00