47773 Commits

Author SHA1 Message Date
julian
a90f5b9dba Remove the home-grown metadata facility in favour of the now generic
mbuf tags facility. Netgraph modules will all need a recompile.

Submitted by:	Gleb Smirnoff <glebius@cell.sick.ru>
2004-06-30 22:51:29 +00:00
phk
309dbc34d2 Define the tty methods as typedefs.
Change the return type for t_break to void.

Add t_ioctl (more about this later).
2004-06-30 21:38:08 +00:00
imp
e23137428c Remove extra RMAN_RESOURCE_VISISBLE 2004-06-30 21:37:55 +00:00
imp
c74757e280 puc needs to peek into struct resource too. 2004-06-30 21:27:59 +00:00
phk
aaba1b361d This needs __RMAN_RESOURCE_VISIBLE 2004-06-30 21:24:19 +00:00
njl
d87b94c8e4 Add missing function debug trace macros to the new powerres functions.
Reported by:	phk
2004-06-30 20:31:13 +00:00
imp
d698489e91 We need to make resources visible here as well. 2004-06-30 19:24:26 +00:00
imp
9f2638da1f Hide struct resource and struct rman. You must define
__RMAN_RESOURCE_VISIBLE to see inside these now.

Reviewed by: dfr, njl (not njr)
2004-06-30 16:54:10 +00:00
njl
5fc49780bf Move flags into a private ivar so it can't collide with device flags.
Unify the code to disable GPEs with the enable code.  Shutdown is handled
the same way.  ACPI now does all wake/sleep prep for child devices so
now they no longer need to call external functions in the suspend/resume
path.  Add the flags to non-ACPI busses (i.e., pci).
2004-06-30 16:08:03 +00:00
njl
d42a9ac2bf Add a method for referencing/switching power resources based on _PRW.
This brings us into line with the standard, which requires power resources
be enabled when wake is enabled for a given device.  Move the dereferencing
code into its own function, +acpi_pwr_dereference_resource().
2004-06-30 16:02:40 +00:00
njl
cfc63ebbf3 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
njl
0d2007fe4f Diff reduction for style. 2004-06-30 15:10:02 +00:00
imp
b5563d0851 My last commit broke oldcard. Rather than duplicate the lines for
pccarddevs.h, just make it standard for now.  Once oldcard is gone,
we'll revisit.
2004-06-30 14:52:49 +00:00
njl
6e4012e1cd Move acpi_if.m to files.{amd64,i386,ia64}. This should fix the alpha build.
Pointed out by:	gallatin
2004-06-30 14:19:28 +00:00
nyan
de83878d91 The ahb module is not needed for pc98. 2004-06-30 13:09:13 +00:00
sos
a600480444 Remove leftover debug output. 2004-06-30 09:22:47 +00:00
njl
8feb332718 Like usbdevs, use before-depend to ensure ordering.
Glass plaque award:	obrien
2004-06-30 05:21:43 +00:00
njl
3c444fb050 Restore the terminating backslash lost by a typo. 2004-06-30 04:54:39 +00:00
njl
13c866dd86 Make acpi_quirks.h conditional on device acpi. 2004-06-30 04:52:55 +00:00
njl
22c21aa533 Add new quirk code that disables problem BIOS versions. Remove old quirk
code that was never really used.  Print a message when disabling ACPI via
a quirk.  Allow the user to override the blacklist decision by setting
hint.acpi.0.disabled="0".  Add missing AcpiTerminate() calls; they are
needed to clean up if bailing out after AcpiInitializeSubsystem().
2004-06-30 04:49:54 +00:00
njl
f4e4cbe915 Add glue for building acpi_quirk.c 2004-06-30 04:47:31 +00:00
njl
e452f1561c Add the table quirk matching code. It matches rules defined in acpi_quirks
and sets the flags.  It also calls the machdep quirk code first.  This
allows table quirks (or whitelists) to override machdep quirks.
2004-06-30 04:45:32 +00:00
njl
53a70792e9 Add machdep quirks functions. On i386, this disables acpi on systems with
BIOS dates earlier than Jan 1, 1999.  Add prototypes and quirks flags.
2004-06-30 04:42:29 +00:00
njl
dc22443bf3 Complete quirk handling script by matching the in-kernel table format.
Use a union to properly assign values.  Collapse some unnecessary variables
in the revision rules.
2004-06-30 04:40:20 +00:00
imp
30c1715cd9 Add pnpinfo and location information to uhub. We also keep track of
the subdevices of uhub better now to accomplish this.

Submitted by: Bernd Walter
2004-06-30 02:56:24 +00:00
imp
3e05633a44 Include more information about the device in the devadded and
devremoved events.  This reduces the races around these events.  We
now include the pnp info in both.  This lets one do more interesting
thigns with devd on device insertion.

Submitted by: Bernd Walter
2004-06-30 02:46:25 +00:00
peter
39ccf02e18 Fix leftover argument to pmap_unuse_pt(). I committed the wrong diff.
Submmitted by:  Jon Noack <noackjr@alumni.rice.edu>
2004-06-30 01:43:23 +00:00
jhb
faceae6354 Document the MUTEX_WAKE_ALL option and turn it on in LINT. 2004-06-30 01:00:59 +00:00
imp
7091d2d55f Add ahb module for i386 and alpha (the onlt two platforms that have eisa
bus)
2004-06-29 21:20:17 +00:00
imp
fefd81efd9 Add the ahb module for old Adaptec 1742 drivers now that eisa bus
interface is cleaner.
2004-06-29 20:50:24 +00:00
imp
c9bcce22e7 Remove saved_* from dc_softc. They are now no longer needed.
Submitted by: Marius Strobl
2004-06-29 20:38:44 +00:00
sos
0cf7e93085 Update the special handling code for ATA devices to allow usage of
PCI native addressing. That means that if the HW says that using "real"
addresses instead of the hardwired legacy compat ones is allowed, we will
use them.
2004-06-29 20:25:43 +00:00
njl
338dc29a06 Update for acpi_id_probe() 2004-06-29 19:05:20 +00:00
njl
df407d3803 Staticize acpi_MatchHid() and include acpi_if.h 2004-06-29 19:03:50 +00:00
njl
72ad681e84 Use the acpi_id_probe() method instead of acpi_MatchHid(), which is now
static.
2004-06-29 19:02:27 +00:00
njl
2a43f50bb0 Add implementation of the ACPI methods which hands them off to ACPI-CA.
acpi_id_probe() returns NULL for no match or the ID string that matched
if the driver should attach.
2004-06-29 19:00:36 +00:00
njl
ffbd0ede95 Add acpi_if.h dependencies to module build. 2004-06-29 18:59:01 +00:00
njl
116b6466b4 Add acpi methods for HID/CID probing, evaluating objects, and walking the
namespace.  This is to allow decoupling of attachments from ACPI where they
need some functionality when ACPI is present but do not want to require ACPI
to always be loaded.
2004-06-29 18:56:14 +00:00
peter
486f03d1c3 Reduce the size of pv entries by 15%. This saves 1MB of KVA for mapping
pv entries per 1GB of user virtual memory.  (eg: if we had 1GB file was
mmaped into 30 processes, that would theoretically reduce the KVA demand by
30MB for pv entries.  In reality though, we limit pv entries so we don't
have that many at once.)

We used to store the vm_page_t for the page table page.  But we recently
had the pa of the ptp, or can calculate it fairly quickly.  If we wanted
to avoid the shift/mask operation in pmap_pde(), we could recover the
pa but that means we have to store it for a while.

This does not measurably change performance.

Suggested by:  alc
Tested by:  alc
2004-06-29 15:57:05 +00:00
marks
dd192c5d24 Remove 3rd clause from the licence.
Approved by:	njl
2004-06-29 15:46:12 +00:00
njl
7bd198c390 Use a default of the FADT for matching CreatorId quirks. Use XSDT to mean
"match either RSDT or XSDT."
2004-06-29 14:32:06 +00:00
phk
474032642a Fix regression in last commit. 2004-06-29 08:33:58 +00:00
jhb
9c6cf2340f Oops, this didn't make it into my submit before I committed: Defer
creation of the sysctl tree for the turnstile profiling stats until a
SI_SUB_LOCK sysinit.  Doing it in init_turnstiles() is too early as it is
called before mi_startup().
2004-06-29 03:48:49 +00:00
darrenr
d82c378cea Mess from update & merge - don't release the ifnet lock twice, just once
and after we're finished with it.
2004-06-29 03:39:06 +00:00
peter
87fa6f8535 Wrap long line. 2004-06-29 03:13:54 +00:00
jhb
6502f84a50 Add two new kernel options to allow rudimentary profiling of the internal
hash tables used in the sleep queue and turnstile code.  Each option adds
a sysctl tree under debug containing the maximum depth of any bucket in
the hash table as well as a separate node for each bucket (or chain)
containing the current depth and maximum depth for that bucket.
2004-06-29 02:30:12 +00:00
njl
348466ddef Include isa/pnpvar.h and remove a duplicate copy of PNP_EISAID. 2004-06-29 01:33:35 +00:00
njl
bd2a2aff39 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
tegge
16dc5859f2 Initialize result->backing_object_offset before linking result onto the list of
vm objects shadowing source in vm_object_shadow().  This closes a race where
vm_object_collapse() could be called with a partially uninitialized object
argument causing symptoms that looked like hardware problems, e.g.  signal 6,
10, 11 or a /bin/sh busy-waiting for a nonexistant child process.
2004-06-28 20:26:35 +00:00
imp
61364cc869 Remove the setting of the pci config variables on power state changes.
The bus does this now.
2004-06-28 20:26:21 +00:00