]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
21 years agoCorrect mostly harmless off-by-one error in getdomainname().
Jacques Vidrine [Sat, 13 Sep 2003 17:12:22 +0000 (17:12 +0000)]
Correct mostly harmless off-by-one error in getdomainname().

Reviewed by: imp

21 years agoDocument /var/empty as generic empty directory, not sshd specific
Eivind Eklund [Sat, 13 Sep 2003 16:23:49 +0000 (16:23 +0000)]
Document /var/empty as generic empty directory, not sshd specific

21 years agoasami is no longer our ports lead, so remove a statement saying
Eivind Eklund [Sat, 13 Sep 2003 16:16:31 +0000 (16:16 +0000)]
asami is no longer our ports lead, so remove a statement saying
the ports collection is maintained by asami and the ports team.

21 years agoFix spelling/grammar errors and remove a duplicate
Eivind Eklund [Sat, 13 Sep 2003 15:46:35 +0000 (15:46 +0000)]
Fix spelling/grammar errors and remove a duplicate

21 years agoMerged from sys/isa/fd.c revision 1.258.
Yoshihiro Takahashi [Sat, 13 Sep 2003 12:56:05 +0000 (12:56 +0000)]
Merged from sys/isa/fd.c revision 1.258.

21 years agoThe re driver works on pc98 maybe.
Yoshihiro Takahashi [Sat, 13 Sep 2003 12:51:12 +0000 (12:51 +0000)]
The re driver works on pc98 maybe.

21 years agoDo not assume there is only a space between #define and the macro name
Jens Schweikhardt [Sat, 13 Sep 2003 06:59:22 +0000 (06:59 +0000)]
Do not assume there is only a space between #define and the macro name
when grepping for JOBS. The recent style cleanup replaced the space with
a tab and broke job control detection. Little edits, disastrous consequences.

Submitted by: Peter Edwards <pmedwards@eircom.net>
X-MFC when:     in about 5 weeks with the other sh arithmetic fixes.

21 years agoAdd support for automatic hardware flow control for 16[679]50 UARTs.
Marcel Moolenaar [Sat, 13 Sep 2003 06:25:04 +0000 (06:25 +0000)]
Add support for automatic hardware flow control for 16[679]50 UARTs.
We simply use the detected FIFO size to determine whether we have
a post 16550 UART or not. The support lacks proper serialization of
hardware access for now.

21 years agoConvert vmapbuf() from using pmap_extract() to using
Alan Cox [Sat, 13 Sep 2003 04:29:55 +0000 (04:29 +0000)]
Convert vmapbuf() from using pmap_extract() to using
pmap_extract_and_hold().  Note, however, that GIANT_REQUIRED should not be
removed until all platforms fully implement the "prot" parameter to
pmap_extract_and_hold().

Reviewed by: tegge

21 years agoI forgot whom I got this from- only set single initiator buffered mode
Matt Jacob [Sat, 13 Sep 2003 02:01:56 +0000 (02:01 +0000)]
I forgot whom I got this from- only set single initiator buffered mode
if we've recorded in our softc that we should set it.

21 years agoNetBSD'er with time on their hands (dma->DMA)
Matt Jacob [Sat, 13 Sep 2003 01:59:14 +0000 (01:59 +0000)]
NetBSD'er with time on their hands (dma->DMA)

21 years agoBland synchronization of common source with no effect on FreeBSD whatsomever.
Matt Jacob [Sat, 13 Sep 2003 01:58:51 +0000 (01:58 +0000)]
Bland synchronization of common source with no effect on FreeBSD whatsomever.

21 years agoSome ridiculous target mode botches- like having the wrong definitions
Matt Jacob [Sat, 13 Sep 2003 01:58:26 +0000 (01:58 +0000)]
Some ridiculous target mode botches- like having the wrong definitions
for messages. Some spelling fixes. Some target mode structure cleanups
to reflect reality.

21 years agoSome canonicalization function cleanups based upon some target mode
Matt Jacob [Sat, 13 Sep 2003 01:57:31 +0000 (01:57 +0000)]
Some canonicalization function cleanups based upon some target mode
structure cleanups.

21 years agoNo time like the present to turn back on SMP locking.
Matt Jacob [Sat, 13 Sep 2003 01:56:45 +0000 (01:56 +0000)]
No time like the present to turn back on SMP locking.

21 years agoReport correct active vs. nvram node/port WWNs in an ioctl.
Matt Jacob [Sat, 13 Sep 2003 01:56:24 +0000 (01:56 +0000)]
Report correct active vs. nvram node/port WWNs in an ioctl.

21 years agoOn reset, make sure that we have some parameters set correctly. This
Matt Jacob [Sat, 13 Sep 2003 01:55:44 +0000 (01:55 +0000)]
On reset, make sure that we have some parameters set correctly. This
fixes a longstanding issue WRT resetting the chip after startup- it
would fail if we were connected as an F-port to a switch. If we
were connected as an F-port, we got assigned a hard loop ID of 255,
which is really a bogus loop id. Then when we turned around to
reset ourselves, the firmware would reject the ICB_INIT request
because the loop id was bogus. *sputter*

Minor fixlet from somebody in NetBSD with too much time on their
hands (dma -> DMA).

21 years agoMove an overly verbose message under #ifdef CODA_VERBOSE.
Tim J. Robbins [Sat, 13 Sep 2003 01:13:56 +0000 (01:13 +0000)]
Move an overly verbose message under #ifdef CODA_VERBOSE.

21 years agomdoc(7): Fix common mistakes made in the SEE ALSO section.
Ruslan Ermilov [Fri, 12 Sep 2003 21:54:11 +0000 (21:54 +0000)]
mdoc(7): Fix common mistakes made in the SEE ALSO section.

21 years agoWhen determining the device class to use for the serial console, check
Thomas Moestl [Fri, 12 Sep 2003 20:13:23 +0000 (20:13 +0000)]
When determining the device class to use for the serial console, check
the "compatible" property too in the ns8250 case. This gets the serial
console to work on Blade 100s, where the device name is just "serial".

Reviewed by: marcel

21 years agoHandle ISA devices in OF_decode_addr(), with the same code that is
Thomas Moestl [Fri, 12 Sep 2003 20:04:29 +0000 (20:04 +0000)]
Handle ISA devices in OF_decode_addr(), with the same code that is
used in the EBus case.

21 years agoRemove double usage of performance, and combine it to one.
Hiten Pandya [Fri, 12 Sep 2003 19:20:05 +0000 (19:20 +0000)]
Remove double usage of performance, and combine it to one.

21 years agoBump __FreeBSD_cc_version for (1) 5.1 (post-mortem) (2) -pthread changes.
David E. O'Brien [Fri, 12 Sep 2003 19:19:10 +0000 (19:19 +0000)]
Bump __FreeBSD_cc_version for (1) 5.1 (post-mortem) (2) -pthread changes.

21 years agoUsing prefix=/usr will put these files in /usr/gnats, not /usr/ports.
Ceri Davies [Fri, 12 Sep 2003 19:02:34 +0000 (19:02 +0000)]
Using prefix=/usr will put these files in /usr/gnats, not /usr/ports.

21 years agoName the vnode method vectors consistently with the rest of the filesystems.
Poul-Henning Kamp [Fri, 12 Sep 2003 16:44:40 +0000 (16:44 +0000)]
Name the vnode method vectors consistently with the rest of the filesystems.

This improves the output of src/tools/tools/vop_table

21 years agoAdd the ports/dns and ports/polish categories.
Ceri Davies [Fri, 12 Sep 2003 09:09:49 +0000 (09:09 +0000)]
Add the ports/dns and ports/polish categories.

21 years agoUnbreak "make release" by moving bfe and re driver into 2nd floppy.
Jun Kuriyama [Fri, 12 Sep 2003 08:20:47 +0000 (08:20 +0000)]
Unbreak "make release" by moving bfe and re driver into 2nd floppy.

21 years agopipe_build_write_buffer() only requires read access of the page that it
Alan Cox [Fri, 12 Sep 2003 07:13:15 +0000 (07:13 +0000)]
pipe_build_write_buffer() only requires read access of the page that it
obtains from pmap_extract_and_hold().

21 years agoAdd a new parameter to pmap_extract_and_hold() that is needed to eliminate
Alan Cox [Fri, 12 Sep 2003 07:07:49 +0000 (07:07 +0000)]
Add a new parameter to pmap_extract_and_hold() that is needed to eliminate
Giant from vmapbuf().

Idea from: tegge

21 years agoMove the parse8601 and parseDWM routines into a new ptime.c file. The
Garance A Drosehn [Fri, 12 Sep 2003 01:33:37 +0000 (01:33 +0000)]
Move the parse8601 and parseDWM routines into a new ptime.c file.  The
only code-change is to add a "next_time" parameter to both routines (and
that is not used yet).  A later update will make "next_time" more useful.

MFC after: 20 days

21 years agoDelay most of the adapter initilization until after interrupts are
Paul Saab [Thu, 11 Sep 2003 23:30:28 +0000 (23:30 +0000)]
Delay most of the adapter initilization until after interrupts are
enabled.

Tested by: Richard Puga <puga@mauibuilt.com>

21 years agoAdd support for using uart(4) for pulse capturing for the Pulse Per
Marcel Moolenaar [Thu, 11 Sep 2003 23:06:42 +0000 (23:06 +0000)]
Add support for using uart(4) for pulse capturing for the Pulse Per
Second (PPS) timing interface. The support is non-optional and by
default uses the DCD line signal as the pulse input. A compile-time
option (UART_PPS_ON_CTS) can be used to have uart(4) use the CTS line
signal.

Include <sys/timepps.h> in uart_bus.h to avoid having to add the
inclusion of that header in all source files.

Reviewed by: phk

21 years agoFix a bunch of off-by-one errors in the range checking code.
Ruslan Ermilov [Thu, 11 Sep 2003 21:40:21 +0000 (21:40 +0000)]
Fix a bunch of off-by-one errors in the range checking code.

21 years agoDisable the use of cloning use in floppy and CD drivers.
Poul-Henning Kamp [Thu, 11 Sep 2003 19:27:24 +0000 (19:27 +0000)]
Disable the use of cloning use in floppy and CD drivers.

This commit puts the relevant code snippets under #ifdef GONE_IN_5
(rather than #ifndef BURN_BRIDGES) thereby disabling the code now.

The code wil be entirely removed before 5.2 unless we find reasons
why this would be a bad idea.

Approach suggested by: imp

21 years ago(atapi_action, case XPT_PATH_INQ): Handle properly the case of
Thomas Quinot [Thu, 11 Sep 2003 17:34:47 +0000 (17:34 +0000)]
(atapi_action, case XPT_PATH_INQ): Handle properly the case of
CAM_TARGET_WILDCARD (target_id_t is an unsigned integer type).
Fixes boot-time crash on Alpha.

Reviewed by: roberto
Approved by: re (murray)
MFC after: 1 week

21 years agoRemove the unrequired -bi from the newaliases line. Note in the commit log
Tom Rhodes [Thu, 11 Sep 2003 16:27:16 +0000 (16:27 +0000)]
Remove the unrequired -bi from the newaliases line.  Note in the commit log
that the last change should have read: exim_enable="YES" in the changes listing.

Discussed with: ceri

21 years agoUpdate for bfe(4) and re(4).
Bruce A. Mah [Thu, 11 Sep 2003 16:03:46 +0000 (16:03 +0000)]
Update for bfe(4) and re(4).

Trim my(4) cards and use manpage entity.

21 years agoNew release note: bfe(4).
Bruce A. Mah [Thu, 11 Sep 2003 16:00:51 +0000 (16:00 +0000)]
New release note:  bfe(4).

Modified release note:  Recent rl(4) additions were moved to the new
re(4) driver.

21 years agoAdd support of the DIOCGSECTORSIZE & DIOCGMEDIASIZE ioctl's
Søren Schmidt [Thu, 11 Sep 2003 15:59:16 +0000 (15:59 +0000)]
Add support of the DIOCGSECTORSIZE & DIOCGMEDIASIZE ioctl's
so that newfs works on dvd-rw/dvd-ram again..

21 years agoMFi386 revisions 1.570 and 1.571.
Yoshihiro Takahashi [Thu, 11 Sep 2003 14:07:02 +0000 (14:07 +0000)]
MFi386 revisions 1.570 and 1.571.

21 years agoMFi386: revision 1.204.
Yoshihiro Takahashi [Thu, 11 Sep 2003 14:05:06 +0000 (14:05 +0000)]
MFi386: revision 1.204.

21 years agoAdd missing .El
Christian Weisgerber [Thu, 11 Sep 2003 13:45:29 +0000 (13:45 +0000)]
Add missing .El

Approved by: ru

21 years agoMFi386: revision 1.1093.
Yoshihiro Takahashi [Thu, 11 Sep 2003 13:44:48 +0000 (13:44 +0000)]
MFi386: revision 1.1093.

21 years agoMerged from sys/isa/fd.c revisions 1.252, 1.253, 1.254, 1.255 and 1.257.
Yoshihiro Takahashi [Thu, 11 Sep 2003 13:35:35 +0000 (13:35 +0000)]
Merged from sys/isa/fd.c revisions 1.252, 1.253, 1.254, 1.255 and 1.257.

21 years agoClean up dead code.
Bill Paul [Thu, 11 Sep 2003 08:28:38 +0000 (08:28 +0000)]
Clean up dead code.

21 years agoToggle the interface on and off before starting the diag test. This
Bill Paul [Thu, 11 Sep 2003 07:54:16 +0000 (07:54 +0000)]
Toggle the interface on and off before starting the diag test. This
seems to be necessary for the 8139C+ under certain circumstances, and
doesn't appear to hurt the other chips. (In the failure case, the
packet would be sent through the TX DMA ring but not get echoed
back. I suspect this has something to do with the link state changing
unexpectedly.)

21 years ago- For the 8169 chips, read the station address by forcing an EEPROM
Bill Paul [Thu, 11 Sep 2003 06:56:46 +0000 (06:56 +0000)]
- For the 8169 chips, read the station address by forcing an EEPROM
  autoload and then copying the contends of the station address
  registers. For some reason, reading the EEPROM on the 8169S doesn't
  work right. This gets around the problem, and allows us to read
  the station address correctly on the 8169S.

- Insert a delay after initiating packet transmition in re_diag() to
  allow lots of time for the frame to echo back to the host, and wait
  for both the 'RX complete' and 'timeout expired' bits in the ISR
  register to be set.

- Deal more intelligently with the fact that the frame length
  field in the RX descriptor is a different width on the 8139C+
  than it is on the 8169/8169S/8110S

- For the 8169, you have to set bit 17 in the TX config register
  to enter digital loopback mode, but for the 8139C+, you have to
  set both bits 17 and 18. Take this into account so that re_diag()
  works properly for both types of chips.

21 years agoRemove symorder. It's almost useless now that we have ELF kernels and
Tim J. Robbins [Thu, 11 Sep 2003 05:58:21 +0000 (05:58 +0000)]
Remove symorder. It's almost useless now that we have ELF kernels and
no a.out toolchain.

21 years agoMinor commentary cleanup, since I didn't understand the comments that
Warner Losh [Thu, 11 Sep 2003 04:22:28 +0000 (04:22 +0000)]
Minor commentary cleanup, since I didn't understand the comments that
I wrote.

21 years agoFix compile on pc98. Maybe this is correct.
Warner Losh [Thu, 11 Sep 2003 04:14:26 +0000 (04:14 +0000)]
Fix compile on pc98.  Maybe this is correct.

21 years agoRe-enable VLAN_MTU capability for this driver. (Got reverted when
Bill Paul [Thu, 11 Sep 2003 04:05:01 +0000 (04:05 +0000)]
Re-enable VLAN_MTU capability for this driver. (Got reverted when
I pulled out the C+/8169 bits.)

21 years agoAdd a PHY driver to support the built-in gigE PHY in the 8169S/8110S
Bill Paul [Thu, 11 Sep 2003 03:53:46 +0000 (03:53 +0000)]
Add a PHY driver to support the built-in gigE PHY in the 8169S/8110S
ethernet chips. This driver is pretty simple, however it contains
special DSP initialization code which is needed in order to get
the chip to negotiate a gigE link. (This special initialization
may not be needed in subsequent chip revs.) Also:

- Fix typo in if_rlreg.h (RL_GMEDIASTAT_1000MPS -> RL_GMEDIASTAT_1000MBPS)

- Deal with shared interrupts in re_intr(): if interface isn't up,
  return.

- Fix another bug in re_gmii_writereg() (properly apply data field mask)

- Allow PHY driver to read the RL_GMEDIASTAT register via the
  re_gmii_readreg() register (this is register needed to determine
  real time link/media status).

21 years agoDeal with the LOGIN_NAME_MAX issue in the NetBSD->FreeBSD
David E. O'Brien [Thu, 11 Sep 2003 03:28:21 +0000 (03:28 +0000)]
Deal with the LOGIN_NAME_MAX issue in the NetBSD->FreeBSD
translation^H^H^Hhack layer.

21 years agoAdd a system notify handler in addition to the device notify handler.
Nate Lawson [Thu, 11 Sep 2003 03:17:33 +0000 (03:17 +0000)]
Add a system notify handler in addition to the device notify handler.
At least some Toshiba notebooks use a Notify of 0 or 1 for this.

PR:
Submitted by: Hiroyuki Aizu <aizu@navi.org>

21 years agoReorder a couple of KASSERTS to give more sensible messages.
Poul-Henning Kamp [Thu, 11 Sep 2003 00:49:02 +0000 (00:49 +0000)]
Reorder a couple of KASSERTS to give more sensible messages.

Found by: GEOM 101 class of '03

21 years agoChanged the ttyd entries to ttyu, which correspond to the device nodes
Jake Burkholder [Thu, 11 Sep 2003 00:14:17 +0000 (00:14 +0000)]
Changed the ttyd entries to ttyu, which correspond to the device nodes
created by uart(4).

21 years agoAdd dumping of the ECDT table.
Nate Lawson [Wed, 10 Sep 2003 23:52:12 +0000 (23:52 +0000)]
Add dumping of the ECDT table.

Courtesy of: USENIX hall track

21 years agoRewrite the SAPIC initialization to always program the RTEs with what
Marcel Moolenaar [Wed, 10 Sep 2003 22:49:38 +0000 (22:49 +0000)]
Rewrite the SAPIC initialization to always program the RTEs with what
we think is the correct trigger mode and polarity. This allows us to
implement BUS_CONFIG_INTR() as an update of the RTE in question.
Consequently, we can trust the RTE when we enable an interrupt and
avoids that we need to know about the trigger mode and polarity at
that time.

21 years agoExtend the ACPI resource handling to make use of the BUS_CONFIG_INTR()
Marcel Moolenaar [Wed, 10 Sep 2003 22:06:41 +0000 (22:06 +0000)]
Extend the ACPI resource handling to make use of the BUS_CONFIG_INTR()
method. This is necessary on ia64 where it's known that serial interfaces
described in the ACPI namespace may not have the well-known IRQs assigned
to them. This confuses us in thinking they are PCI based interrupts and
wrongly program the APIC.

21 years agoo Workaround a bug where my T23 reports that it is ACPI 2.0x compatible
Nate Lawson [Wed, 10 Sep 2003 22:00:45 +0000 (22:00 +0000)]
o Workaround a bug where my T23 reports that it is ACPI 2.0x compatible
but has invalid 64 bit pointers for FACS and DSDT.
o Finish work to print all of the FADT and FACS.
o Resort the comment generating functions.  Submitted by: marcel

Courtesy of: BSDcon back wall

21 years agoIntroduce BUS_CONFIG_INTR(). The method allows devices to tell parents
Marcel Moolenaar [Wed, 10 Sep 2003 21:37:10 +0000 (21:37 +0000)]
Introduce BUS_CONFIG_INTR(). The method allows devices to tell parents
about interrupt trigger mode and interrupt polarity. This allows ACPI
for example to pass interrupt resource information up the hierarchy.
The default implementation of the method therefore is to pass the
request to the parent.

Reviewed by: jhb, njl

21 years agoWith the exim port upgrade, modify sysinstall(8):
Tom Rhodes [Wed, 10 Sep 2003 20:55:09 +0000 (20:55 +0000)]
With the exim port upgrade, modify sysinstall(8):

- Add 'enable_exim="YES"' to rc.conf(5)
- Use the default exim configuration file from the port
- When using sendmail, disable some more scripts that use sendmail specific
  parameters
- Have sysinstall tweak mailer.conf(5) substitution
- Use 'N' flag for newsyslog(8)

Submitted by: Oliver Eikemeier <eikemeier@fillmore-labs.com>
Reviewed by: sheldonh, simon
Tested by: myself (trhodes) and submitter

21 years agocompaq WL110 is a lucent card
Warner Losh [Wed, 10 Sep 2003 19:34:51 +0000 (19:34 +0000)]
compaq WL110 is a lucent card

Submitted by Jeremy Bingham

21 years agomdoc(7): Properly mark C headers.
Ruslan Ermilov [Wed, 10 Sep 2003 19:24:35 +0000 (19:24 +0000)]
mdoc(7): Properly mark C headers.

21 years agoDocument the fact that send(2) can return EPIPE (like when a socket is not
Ollivier Robert [Wed, 10 Sep 2003 19:19:49 +0000 (19:19 +0000)]
Document the fact that send(2) can return EPIPE (like when a socket is not
connected).

PR: docs/56683
Submitted by: Chris S.J. Peron <maneo@bsdpro.com>
MFC after: 3 days

21 years agoRelent and back out rev 1.15.
David E. O'Brien [Wed, 10 Sep 2003 19:08:16 +0000 (19:08 +0000)]
Relent and back out rev 1.15.

21 years agoAdd a dependancy on 'nbsd2fbsd.h'
David E. O'Brien [Wed, 10 Sep 2003 19:03:48 +0000 (19:03 +0000)]
Add a dependancy on 'nbsd2fbsd.h'

21 years agoSort 'bge' correctly.
David E. O'Brien [Wed, 10 Sep 2003 18:54:59 +0000 (18:54 +0000)]
Sort 'bge' correctly.

21 years agomdoc(7): Moved #include directives into the SYNOPSIS section.
Ruslan Ermilov [Wed, 10 Sep 2003 18:47:07 +0000 (18:47 +0000)]
mdoc(7): Moved #include directives into the SYNOPSIS section.

21 years agoTypo.
Ruslan Ermilov [Wed, 10 Sep 2003 18:45:26 +0000 (18:45 +0000)]
Typo.

21 years agoIn case vmapbuf() fails, release all of the held resources.
Alan Cox [Wed, 10 Sep 2003 18:23:43 +0000 (18:23 +0000)]
In case vmapbuf() fails, release all of the held resources.

Submitted by: tegge

21 years agoFix asynchronous physio breakage introduced in rev 1.163.
Hidetoshi Shimokawa [Wed, 10 Sep 2003 15:48:51 +0000 (15:48 +0000)]
Fix asynchronous physio breakage introduced in rev 1.163.
We cannnot use bp->b_caller2 because DEV_STRATEGY will overwrite it.

21 years agoFix bug in re_gmii_writewreg(): we don't need to be screening out certain
Bill Paul [Wed, 10 Sep 2003 15:14:46 +0000 (15:14 +0000)]
Fix bug in re_gmii_writewreg(): we don't need to be screening out certain
PHY addresses here. (The chip will only let you talk to one PHY anyway.)

21 years agoTeach rl(4) about new hwrev codes.
Bill Paul [Wed, 10 Sep 2003 15:12:55 +0000 (15:12 +0000)]
Teach rl(4) about new hwrev codes.

21 years agoTry a bit harder to probe disks that doesn't quite set BUSY right.
Søren Schmidt [Wed, 10 Sep 2003 09:57:16 +0000 (09:57 +0000)]
Try a bit harder to probe disks that doesn't quite set BUSY right.

21 years agomdoc(7): There cannot be a subsection inside a list.
Ruslan Ermilov [Wed, 10 Sep 2003 08:24:33 +0000 (08:24 +0000)]
mdoc(7): There cannot be a subsection inside a list.

Reported by: naddy

21 years agoUpdate hardware revision table. 0x04000000 appears to be the revision
Bill Paul [Wed, 10 Sep 2003 07:21:43 +0000 (07:21 +0000)]
Update hardware revision table. 0x04000000 appears to be the revision
for the 8169S, according to my sample board. The RealTek Linux driver
mentions 0x00800000. I'm assigning this to the 8110S until I get
more info on it. (The (preliminary) RealTek docs only say that 8169S/8110S
chips will have some combination of those two bits set, but doesn't say
exactly what bit combination goes with which chip variant.)

21 years agoWhitespace.
John Baldwin [Wed, 10 Sep 2003 07:09:16 +0000 (07:09 +0000)]
Whitespace.

21 years agoDocument the alternate way of matching MAC addresses: by a bitmask.
Peter Pentchev [Wed, 10 Sep 2003 06:41:16 +0000 (06:41 +0000)]
Document the alternate way of matching MAC addresses: by a bitmask.

PR: 56021
Submitted by: Glen Gibb <grg@ridley.unimelb.edu.au>
MFC after: 1 month

21 years agoMove the definitions for ACPI MADT table entries not present in the ACPICA
John Baldwin [Wed, 10 Sep 2003 06:32:27 +0000 (06:32 +0000)]
Move the definitions for ACPI MADT table entries not present in the ACPICA
distribution to a MI header so it can be shared with other architectures.

21 years agoWe represent PCI intpin's two different ways. One is the way that the
John Baldwin [Wed, 10 Sep 2003 06:00:53 +0000 (06:00 +0000)]
We represent PCI intpin's two different ways.  One is the way that the
intpin register is expressed in hardware where 0 means none, 1 means INTA,
2 INTB, etc.  The other way is commonly used in loops where 0 means INTA,
1 means INTB, etc.  The matchpin argument to pci_cfgintr_search() is
supposed to be the first form, but we passsed in a loop index of the
second.  This fix adds one to the loop index to convert to the first form.

Reported by: Pavlin Radoslavov <pavlin@icir.org>

21 years agoFinish an earlier commit:
John Baldwin [Wed, 10 Sep 2003 05:29:30 +0000 (05:29 +0000)]
Finish an earlier commit:

Add a acpi_SetDefaultIntrModel() method to allow drivers to set the
interrupt model prior to the acpi0 device being probed and attached.

21 years agoIf we failed to size the Rx FIFO, assume the worst. This however
Marcel Moolenaar [Wed, 10 Sep 2003 05:01:08 +0000 (05:01 +0000)]
If we failed to size the Rx FIFO, assume the worst. This however
is not a size of 1. Since we already know there is a FIFO, we can
safely assume that it is at least 16 bytes. Note that all this is
mostly academic anyway. We don't use the size of the Rx FIFO
currently. If we add support for hardware flow control, we only
care about Rx FIFO sizes larger than 16.

21 years agoMove an annoying printf() call that gets triggered every time an
Tim J. Robbins [Wed, 10 Sep 2003 01:41:15 +0000 (01:41 +0000)]
Move an annoying printf() call that gets triggered every time an
operation is interrupted (with ^C or ^Z) under CODA_VERBOSE.

21 years agoRemove an XXX comment by using the per CPU mask added after this comment
John Baldwin [Wed, 10 Sep 2003 01:36:48 +0000 (01:36 +0000)]
Remove an XXX comment by using the per CPU mask added after this comment
was added.

21 years agoAdd a acpi_SetDefaultIntrModel() method to allow drivers to set the
John Baldwin [Wed, 10 Sep 2003 01:14:42 +0000 (01:14 +0000)]
Add a acpi_SetDefaultIntrModel() method to allow drivers to set the
interrupt model prior to the acpi0 device being probed and attached.

21 years agoadd ELSA Vianect WLAN (Marco Wertejuk)
Warner Losh [Wed, 10 Sep 2003 01:14:07 +0000 (01:14 +0000)]
add ELSA Vianect WLAN (Marco Wertejuk)
benq awl100 (David Leemans)

ObTerminalRoomCommit: done!

21 years agoFix a typo.
John Baldwin [Wed, 10 Sep 2003 01:11:58 +0000 (01:11 +0000)]
Fix a typo.

21 years agoAdd comments to the members of the timecounter struct similar to other
John Baldwin [Wed, 10 Sep 2003 01:10:24 +0000 (01:10 +0000)]
Add comments to the members of the timecounter struct similar to other
timecounters.

21 years agoUpdate the license on this file to be a bit more sane.
John Baldwin [Wed, 10 Sep 2003 01:09:32 +0000 (01:09 +0000)]
Update the license on this file to be a bit more sane.

21 years agoAdd constants for entries in the IDT and use those instead of magic
John Baldwin [Wed, 10 Sep 2003 01:07:04 +0000 (01:07 +0000)]
Add constants for entries in the IDT and use those instead of magic
numbers.

21 years agoAdd /usr/X11R6/man to the optional manpath.
Ruslan Ermilov [Wed, 10 Sep 2003 00:07:28 +0000 (00:07 +0000)]
Add /usr/X11R6/man to the optional manpath.

PR: conf/56626
Submitted by: Oliver Eikemeier

21 years agoFixed -Wpointer-arith warning.
Ruslan Ermilov [Tue, 9 Sep 2003 23:50:57 +0000 (23:50 +0000)]
Fixed -Wpointer-arith warning.

Submitted by: Stefan Farfeleder
PR: bin/56653

21 years agoCatch up with mdoc.local list of libraries.
Ruslan Ermilov [Tue, 9 Sep 2003 23:27:03 +0000 (23:27 +0000)]
Catch up with mdoc.local list of libraries.

21 years agoCatch up with mdoc.local list of libraries.
Ruslan Ermilov [Tue, 9 Sep 2003 23:19:03 +0000 (23:19 +0000)]
Catch up with mdoc.local list of libraries.

21 years agoOriginal pthread_once code has memory leak if pthread_once_t is used in
David Xu [Tue, 9 Sep 2003 22:38:12 +0000 (22:38 +0000)]
Original pthread_once code has memory leak if pthread_once_t is used in
a shared library or any other dyanmic allocated data block, once
pthread_once_t is initialized, a mutex is allocated, if we unload the
shared library or free those data block, then there is no way to deallocate
the mutex, result is memory leak.
To fix this problem, we don't use mutex field in pthread_once_t, instead,
we use its state field and an internal mutex and conditional variable in
libkse to do any synchronization, we introduce a third state IN_PROGRESS to
wait if another thread is already in invoking init_routine().
Also while I am here, make pthread_once() conformed to pthread cancellation
point specification.

Reviewed by: deischen

21 years agoSwitch dotrim() to take advantage of the 'struct conf_entry' that
Garance A Drosehn [Tue, 9 Sep 2003 21:14:05 +0000 (21:14 +0000)]
Switch dotrim() to take advantage of the 'struct conf_entry' that
is already passed in, instead of having the caller copy values from
that struct into additional parameters.

MFC after: 22 days

21 years agoReduce the annoying compiler warnings that pop up when compiling with
Garance A Drosehn [Tue, 9 Sep 2003 20:29:26 +0000 (20:29 +0000)]
Reduce the annoying compiler warnings that pop up when compiling with
gcc 3.3.x and -Wshadow.  Just renames 'log' variables to be 'logname'.

MFC after: 22 days

21 years agoClean up get/set_mcontext() and get/set_fpcontext(). These are operated
Peter Wemm [Tue, 9 Sep 2003 19:32:09 +0000 (19:32 +0000)]
Clean up get/set_mcontext() and get/set_fpcontext().  These are operated
on data structures on the kernel stack which are guaranteed to be 16 byte
aligned by gcc, the amd64 ABI and __aligned(16).

Ensire the tss_rsp0 initial stack pointer is 16 byte aligned in case
sizeof(pcb) becomes odd at some point.  This is convenient for the
interrupt handler case because the ring crossing pushes cause the
required odd alignment before the call to the C code.

Have fast_syscall add an additional 8 bytes to ensure that the trapframe
has the correct odd alignment for the call to C code.  Note that there are
no checks to make sure that the trapframe size is appropriate for this.

This makes get/setfpcontext work properly (finally).  You get a GPF in
kernel mode if any of this is botched without the alignment fixup code
that is apparently needed on i386.

21 years agoFix some broken comments.
Dag-Erling Smørgrav [Tue, 9 Sep 2003 19:22:55 +0000 (19:22 +0000)]
Fix some broken comments.