]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
21 years agoGrap the ATA lock on all channels before suspend, this makes certain
sos [Sun, 4 May 2003 09:34:14 +0000 (09:34 +0000)]
Grap the ATA lock on all channels before suspend, this makes certain
that we have no outstanding ops in transit, which would cause problems
on resume.

21 years agoOn getting status print the stripesize on RAID0's.
sos [Sun, 4 May 2003 09:28:43 +0000 (09:28 +0000)]
On getting status print the stripesize on RAID0's.

21 years agoFix a bunch of typos and grammatical errors.
murray [Sun, 4 May 2003 08:23:24 +0000 (08:23 +0000)]
Fix a bunch of typos and grammatical errors.

PR: docs/40234
Submitted by: Chris Pepper <pepper@rockefeller.edu>  (mostly)
MFC After: 3 days

21 years agoEliminate the separate malloc type for the sparing table.
scottl [Sun, 4 May 2003 07:41:07 +0000 (07:41 +0000)]
Eliminate the separate malloc type for the sparing table.

21 years agoAdd a missing __inline. Strange that gcc never complained about it.
scottl [Sun, 4 May 2003 07:40:29 +0000 (07:40 +0000)]
Add a missing __inline.  Strange that gcc never complained about it.
Implement udf_readlblks() in terms of RDSECTOR.

21 years agoCorrectly calculate the size of the extent that should be read in
scottl [Sun, 4 May 2003 07:39:11 +0000 (07:39 +0000)]
Correctly calculate the size of the extent that should be read in
udf_readatoffset().  This should fixe problems with reading udf filesystems
created with mkisofs.

21 years agoForgot to update string and signal tables when some of the trap types
jake [Sun, 4 May 2003 07:21:04 +0000 (07:21 +0000)]
Forgot to update string and signal tables when some of the trap types
changed.

21 years agoAvoid a lock-order reversal and implement vm_object locking
alc [Sun, 4 May 2003 06:56:27 +0000 (06:56 +0000)]
Avoid a lock-order reversal and implement vm_object locking
in vm_pageout_page_free().

21 years agoDocument the login-backoff and login-retries capabilities.
murray [Sun, 4 May 2003 06:43:24 +0000 (06:43 +0000)]
Document the login-backoff and login-retries capabilities.

PR: docs/51397
MFC After: 3 days

21 years agoDocument a workaround for getting past the groff infinite loop problem
imp [Sun, 4 May 2003 06:26:30 +0000 (06:26 +0000)]
Document a workaround for getting past the groff infinite loop problem
when your userland is older tahn your kernel and Makefile.inc1 therefore
thinks that you don't need to rebuild groff, when in fact you do.  This
workaround is to set OSRELDATE to 500110 before building, but only if you
have a new enough Makefile.inc1.

I've confirmed this on my laptop, which was suffering from this problem.
Since it only hits power users, we'll naively assume that such users are
smart enough to read UPDATING.

21 years agoRemove man page reference for gated, as it isn't in the base system
murray [Sun, 4 May 2003 06:17:38 +0000 (06:17 +0000)]
Remove man page reference for gated, as it isn't in the base system
nor is it available from the ports tree.

21 years agoNote that gated is a commercial routing product now, and remove
murray [Sun, 4 May 2003 06:13:25 +0000 (06:13 +0000)]
Note that gated is a commercial routing product now, and remove
references to it's man page, which is almost never installed on a
FreeBSD system.  The information about using this command with gated
has been retained.  I have just made it clear that gated is not a part
of FreeBSD.

PR: docs/51407

21 years agoGrab some of the latest TI cardbus bridge IDs. Inspired by a similar
imp [Sun, 4 May 2003 06:10:05 +0000 (06:10 +0000)]
Grab some of the latest TI cardbus bridge IDs.  Inspired by a similar
commit to NetBSD.  I'll add these to various tables soon.

21 years agoMy previous commit broke builds for restricted namespaces. Add
mbr [Sun, 4 May 2003 05:57:50 +0000 (05:57 +0000)]
My previous commit broke builds for restricted namespaces. Add
a underscore to fix this.

Reviewed by: peter

21 years agoForced commit. Grr: I used the wrong text before. Here's the correct
grog [Sun, 4 May 2003 05:20:56 +0000 (05:20 +0000)]
Forced commit.  Grr: I used the wrong text before.  Here's the correct
information.

Rewrite of startup code:

vinum_scandisk:

  1.  Hand a string with device names, not a table and length.  This
      is the way we get it at system startup time.  This also means
      that we don't insist on a /dev in the names.
  2.  Allocate drive list dynamically.
  3.  If passed a null drive list, use kern.disks as a list to search.

21 years agoUpdate declaration of vinum_scandisk (only one parameter).
grog [Sun, 4 May 2003 05:19:48 +0000 (05:19 +0000)]
Update declaration of vinum_scandisk (only one parameter).

21 years agoRewrite of startup code:
grog [Sun, 4 May 2003 05:16:02 +0000 (05:16 +0000)]
Rewrite of startup code:

Add ioctl VINUM_READCONFIG which implements both the "read" and
"start" commands in vinum(8).  Aim for marginally better error
messages when something goes wrong.

21 years agoRewrite of startup code:
grog [Sun, 4 May 2003 05:15:32 +0000 (05:15 +0000)]
Rewrite of startup code:

Add ioctl VINUM_READCONFIG which implements both the "read" and
"start" commands in vinum(8).

21 years agoRewrite of startup code:
grog [Sun, 4 May 2003 05:15:07 +0000 (05:15 +0000)]
Rewrite of startup code:

parse_config: Remove kw_read case.  It's not used any more.

parse_user_config: Remove check for failed read.

21 years agoRewrite of startup code:
grog [Sun, 4 May 2003 05:13:45 +0000 (05:13 +0000)]
Rewrite of startup code:

vinumattach: Move the startup code to vinum_scandisk.

21 years agoRewrite of startup code:
grog [Sun, 4 May 2003 05:13:17 +0000 (05:13 +0000)]
Rewrite of startup code:

vinum_read:
  Accept 0 parameters, hand an empty string to the kernel if so.
  Use new ioctl VINUM_READCONFIG, not VINUM_CREATE, to read config.

vinum_start: When passed 0 parameters, drop the devstat kludge and
     call vinum_read instead.  It's no longer necessary to distinguish
     between "start" and "read" if no parameters are passed, and
     probably one of them should go away.

21 years agoMany developers run with userland != to kernel. While this isn't
imp [Sun, 4 May 2003 04:14:10 +0000 (04:14 +0000)]
Many developers run with userland != to kernel.  While this isn't
supported, it usually works for months at a time.  Allow these people
to override the OSRELDATE of their installed world when things don't
match and the exact OSRELDATE matters and is different than the
kernel.  Now that Makefile.inc1 depends more and more about which date
you have to optimize the pieces it builds, it may be necessary to
pessimize things if its guesses are wrong.

If OSRELDATE is already set, we won't fork the sysctl to find out what
the kernel's date is.

Developers on IRC suggested that they run mismatches all the time as
well.

Reviewed by: obrien

21 years agoImplement the node cache as a hash table.
scottl [Sun, 4 May 2003 03:40:11 +0000 (03:40 +0000)]
Implement the node cache as a hash table.

21 years agoUse __FBSDID vs. rcsid[]. Also protect sccs[] and copyright[] from GCC 3.3.
obrien [Sun, 4 May 2003 02:54:49 +0000 (02:54 +0000)]
Use __FBSDID vs. rcsid[].  Also protect sccs[] and copyright[] from GCC 3.3.

21 years agoUse __FBSDID vs. rcsid[].
obrien [Sun, 4 May 2003 02:51:42 +0000 (02:51 +0000)]
Use __FBSDID vs. rcsid[].

21 years agoSomething I missed in my tree.
ps [Sun, 4 May 2003 02:03:20 +0000 (02:03 +0000)]
Something I missed in my tree.
If the media changes, poke the DSP on the 5401/5411 Broadcom PHY's.

Obtained from: NetBSD

21 years agoKSE is useful for pc98.
nyan [Sun, 4 May 2003 02:02:48 +0000 (02:02 +0000)]
KSE is useful for pc98.

21 years agoUse __FBSDID vs. rcsid[].
obrien [Sun, 4 May 2003 00:59:13 +0000 (00:59 +0000)]
Use __FBSDID vs. rcsid[].

21 years agoFix signed/unsigned comparison warnings.
obrien [Sun, 4 May 2003 00:56:00 +0000 (00:56 +0000)]
Fix signed/unsigned comparison warnings.

21 years agoFix a sign/unsigned comparison.
obrien [Sun, 4 May 2003 00:43:39 +0000 (00:43 +0000)]
Fix a sign/unsigned comparison.

21 years agoCorrect spelling errors.
gibbs [Sun, 4 May 2003 00:20:07 +0000 (00:20 +0000)]
Correct spelling errors.

Switch to handling bad SCSI status as a sequencer interrupt
instead of having the kernel proccess these failures via
the completion queue.  This is done because:

 o The old scheme required us to pause the sequencer and clear
   critical sections for each SCB.  It seems that these pause
   actions, if coincident with a sequencer FIFO interrupt, would
   result in a FIFO interrupt getting lost or directing to the
   wrong FIFO.  This caused hangs when the driver was stressed
   under high "queue full" loads.
 o The completion code assumed that it was always called with
   the sequencer running.  This may not be the case in timeout
   processing where completions occur manually via
   ahd_pause_and_flushwork().
 o With this scheme, the extra expense of clearing critical
   sections is avoided since the sequencer will only self pause
   once all pending selections have cleared and it is not in
   a critical section.

  aic79xx.c
Add code to handle the new BAD_SCB_STATUS sequencer
interrupt code.  This just redirects the SCB through
the already existing ahd_complete_scb() code path.
Remove code in ahd_handle_scsi_status() that paused
the sequencer, made sure that no selections where
pending, and cleared critical sections.  Bad
status SCBs are now only processed when all of these
conditions are true.

  aic79xx.reg:
Add the BAD_SCB_STATUS sequencer interrupt code.

  aic79xx.seq:
When completing an SCB upload to the host, if
we are doing this because the SCB contains non-zero
SCSI status, defer completing the SCB until there
are no pending selection events.  When completing
these SCBs, use the new BAD_SCB_STATUS sequencer
interrupt.  For all other uploaded SCBs (currently
only for underruns), the SCB is completed via the
normal done queue.  Additionally, keep the SCB that
is currently being uploaded on the COMPLETE_DMA_SCB
list until the dma is completed, not just until the
DMA is started.  This ensures that the DMA is restarted
properly should the host disable the DMA transfer for
some reason.

In our RevA workaround for Maxtor drives, guard against
the host pausing us while trying to pause I/O until the
first data-valid REQ by clearing the current snapshot
so that we can tell if the transfer has completed prior
to us noticing the REQINIT status.

In cfg4data_intr, shave off an instruction before getting
the data path running by adding an entrypoint to the
overrun handler to also increment the FIFO use count.

In the overrun handler, be sure to clear our LONGJMP
address in both exit paths.

Perform a few sequencer optimizations.

  aic79xx.c:
Print the full path from the SCB when a packetized
status overrun occurs.

Remove references to LONGJMP_SCB which is being
removed from firmware usage.

Print the new SCB_FIFO_USE_COUNT field in the
per-SCB section of ahd_dump_card_state().  The
SCB_TAG field is now re-used by the sequencer,
so it no longer makes sense to reference this
field in the kernel driver.

  aic79xx.h:
Re-arrange fields in the hardware SCB from largest
size type to smallest.  This makes it easier to
move fields without changing field alignment.

The hardware scb tag field is now down near the
"spare" portion of the SCB to facilitate reuse
by the sequencer.

  aic79xx.reg:
Remove LONGJMP_ADDR.

Rearrange SCB fields to match aic79xx.h.
Add SCB_FIFO_USE_COUNT as the first byte
of the SCB_TAG field.

  aic79xx.seq:
Add a per-SCB "Fifos in use count" field and use
it to determine when it is safe (all data posted)
to deliver status back to the host.  The old method
involved polling one or both FIFOs to verify that
the current task did not have pending data.  This
makes running down the GSFIFO very cheap, so we
will empty the GSFIFO in one idle loop pass in
all cases.

Use this simplification of the completion process
to prune down the data FIFO teardown sequencer for
packetized transfers.  Much more code is now shared
between the data residual and transfer complete cases.

Correct some issues in the packetized status handler.
It used to be possible to CLRCHN our FIFO before status
had fully transferred to the host.  We also failed to
handle NONPACKREQ phases that could occur should a CRC
error occur during transmission of the status data packet.

Correct a few big endian issues:

  aic79xx.c:
  aic79xx_inline.h:
  aic79xx_pci.c:
  aic79xx_osm.c:
o Always get the SCB's tag via the SCB_GET_TAG acccessor
o Add missing use of byte swapping macros when touching
  hscb fields.
o Don't double swap SEEPROM data when it is printed.
  Correct a big-endian bug.  We cannot assign a
o When assigning a 32bit LE variable to a 64bit LE
  variable, we must be explict about how the words
  of the 64bit LE variable are initialized.  Cast to
  (uint32_t*) to do this.

aic79xx.c:
In ahd_clear_critical_section(), hit CRLSCSIINT
after restoring the interrupt masks to avoid what
appears to be a glitch on SCSIINT.  Any real SCSIINT
status will be persistent and will immidiately
reset SCSIINT.  This clear should only get rid of
spurious SCSIINTs.

This glitch was the cause of the "Unexpected PKT busfree"
status that occurred under high queue full loads

Call ahd_fini_scbdata() after shutdown so that
any ahd_chip_init() routine that might access
SCB data will not access free'd memory.

Reset the bus on an IOERR since the chip doesn't
seem to reset to the new voltage level without
this.

Change offset calculation for scatter gather maps
so that the calculation is correct if an integral
multiple of sg lists does not fit in the allocation
size.

Adjust bus dma tag for data buffers based on 39BIT
addressing flag in our softc.

Use the QFREEZE count to simplify ahd_pause_and_flushworkd().
We can thus rely on the sequencer eventually clearing ENSELO.

In ahd_abort_scbs(), fix a bug that could potentially
corrupt sequencer state.  The saved SCB was being
restored in the SCSI mode instead of the saved mode.
It turns out that the SCB did not need to be saved at all
as the scbptr is already restored by all subroutines
called during this function that modify that register.

aic79xx.c:
aic79xx.h:
aic79xx_pci.c:
Add support for parsing the seeprom vital product
data.  The VPD data are currently unused.

aic79xx.h:
aic79xx.seq:
aic79xx_pci.c:
Add a firmware workaround to make the LED blink
brighter during packetized operations on the H2A.

aic79xx_inline.h:
The host does not use timer interrupts, so don't
gate our decision on whether or not to unpause
the sequencer on whether or not a timer interrupt
is pending.

21 years agoImplement the write twice to send producer index workaround for
ps [Sun, 4 May 2003 00:07:21 +0000 (00:07 +0000)]
Implement the write twice to send producer index workaround for
BCM5700 chips as described in the Broadcom Errata documents.

Obtained from: NetBSD and Broadcom Errata documentation

21 years agoaic7xxx.c:
gibbs [Sat, 3 May 2003 23:55:38 +0000 (23:55 +0000)]
aic7xxx.c:
aic7xxx.h:
Split out core chip initialization into ahc_chip_init().
This will allow us to reset the chip correctly at times
other than initial chip setup.

aic7770.c
aic7xxx_pci.c:
Flesh out bus chip init methods for our two
bus attachments and use these, in addition to
bus suspend/resume hooks to get the core in
better shape for handling these events.

When disabling PCI parity error checking, use FAILDIS.
Although the chip docs indicate that clearing PERRESPEN
should also work, it does not.

Auto-disable pci parity error checking after informing
the user of AHC_PCI_TARGET_PERR_THRESH number of parity
errors observed as a target.

aic7xxx.h:
aic7xxx_pci.c
aic7770.c
aic7xxx.c
Add the instruction_ram_size softc field.

Remove the now unused stack_size softc field.

Modify ahc_loadseq to return a failure code
and to actually check the downloaded instruction
count against the limit set in our softc.

Modify callers of ahc_loadseq to handle load
failures as appropriate.

Set instruction RAM sizes for each chip type.

aic7xxx_pci.c:
Add some delay in the aic785X termination
control code.  This may fix problems with
the 2930.

Be consistent in how we access config space
registers.  16bit registers are accessed using
16bit ops.

aic7xxx.c:
Correct spelling errors.

Have ahc_force_renegotiation() take a devinfo as is done
in the U320 driver.  Use this argument to correct a bug
in the selection timeout handler where we forced a renegotiation
with the last device that had set SAVED_SCSIID.  SAVED_SCSIID
is only updated once a selection is *sucessfull* and so is
stale for any selection timeout.

Cleanup the setup of the devinfo for busfree events.  We
now use this devinfo for a call to ahc_force_renegotiation()
at the bottom of the routine, so it must be initialized in
all cases.

In ahc_pause_and_flushwork(), adjust the loop so that it
will exit in the hot-eject case even if the INT_PEND mask
is something other than 0xFF (as it is in this driver).

Correct a wrapping string constant.

Call ahc_fini_scbdata() after shutdown so that
any ahc_chip_init() routine that might access
SCB data will not access free'd memory.

Correctly setup our buffer tag to indicate that 39bit
addressing is available if in 39bit addressing mode.

Rearrange some variable declarations based on
type size.

aic7xxx.c
aic7xxx.h:
aic7xxx.reg:
Consistently use MAX_OFFSET for the user max syncrate
set from non-volatile storage.  This ensures that the
offset does not conflict with AHC_OFFSET_UNKNOWN.

Change AHC_OFFSET_UNKNOWN to 0xFF.  This is
a value that the curr->offset can never be,
unlike '0' which we previously used.  This
fixes code that only checks for a non-zero
offset to determine if a sync negotiation
is required since it will fire in the unknown
case even if the goal is async.

Change MAX_OFFSET to 0x7f which is the max
offset U160 aic7xxx controllers can negotiate.
This ensures that curr->offset will not
match AHC_OFFSET_UNKNOWN.

aic7xxx_inline.h:
Have our inline interrupt handler return with a value
indicating whether we serviced a real interrupt.  This
is required for Linux support.

Return earlier if the interrupt is not for us.

21 years agoAdd rue(4) devices.
bmah [Sat, 3 May 2003 23:47:21 +0000 (23:47 +0000)]
Add rue(4) devices.

21 years agoNew release notes: rue(4), MI passwd databases, libkse.
bmah [Sat, 3 May 2003 23:46:28 +0000 (23:46 +0000)]
New release notes:  rue(4), MI passwd databases, libkse.

Modified release notes:  Update DRM note and move from kernel section
to contrib section, add arch attribute for libthr note.

21 years agohc_eisa.c:
gibbs [Sat, 3 May 2003 23:27:57 +0000 (23:27 +0000)]
hc_eisa.c:
ahc_pci.c:
ahd_pci.c:
        Don't bogusly use the BUS_DMA_ALLOCNOW flag in our psuedo
        parent bus dma tag.

21 years agoNo, forget about rev 1.24. My system wasn't virgin enough when I tested that.
obrien [Sat, 3 May 2003 23:15:09 +0000 (23:15 +0000)]
No, forget about rev 1.24.  My system wasn't virgin enough when I tested that.
Restore rev 1.23.

21 years agoFix c99 victim: the accepted character '0 most now be types as '0'.
marcel [Sat, 3 May 2003 23:05:16 +0000 (23:05 +0000)]
Fix c99 victim: the accepted character '0 most now be types as '0'.

21 years ago- Change the short hand representation of the various ASIC revisions
ps [Sat, 3 May 2003 22:58:45 +0000 (22:58 +0000)]
- Change the short hand representation of the various ASIC revisions
- Implement the ONEDMA_AT_ONCE workaround as described in the
  5703/5704 eratta documents.
  Obtained from: NetBSD & Broadcom documentation

21 years agoFix whitespace nit...no content changes.
bmah [Sat, 3 May 2003 22:18:49 +0000 (22:18 +0000)]
Fix whitespace nit...no content changes.

21 years agoUpdate "boilerplate", fix wording and markup of some notes.
bmah [Sat, 3 May 2003 22:18:24 +0000 (22:18 +0000)]
Update "boilerplate", fix wording and markup of some notes.

21 years agoGeneral update of the Early Adopter's Guide. Too many changes to
bmah [Sat, 3 May 2003 22:14:55 +0000 (22:14 +0000)]
General update of the Early Adopter's Guide.  Too many changes to
list here, but they basically fall into two categories:  1) Make
the document less 5.0-centric (and more applicable to the state of any
5.X before the RELENG_5 branch) and 2) Update for the current state of
5.X, in anticipation of re-issuing this document as a part of the 5.1
release documentation.

21 years agoMFV:
ru [Sat, 3 May 2003 22:02:44 +0000 (22:02 +0000)]
MFV:

- Do not emit multiple empty lines in the CRT mode.
- Fixed the Nd macro to use the right dash.

21 years agoUse __FBSDID over rcsid[].
obrien [Sat, 3 May 2003 21:31:08 +0000 (21:31 +0000)]
Use __FBSDID over rcsid[].

21 years agoThis is WARNS=2 clean.
obrien [Sat, 3 May 2003 21:30:34 +0000 (21:30 +0000)]
This is WARNS=2 clean.

21 years agoUse __FBSDID over rcsid[]. Protect copyright[] where needed.
obrien [Sat, 3 May 2003 21:06:42 +0000 (21:06 +0000)]
Use __FBSDID over rcsid[].  Protect copyright[] where needed.

21 years agoFix long constant usage for i386.
markm [Sat, 3 May 2003 20:52:48 +0000 (20:52 +0000)]
Fix long constant usage for i386.

Tested by: Joe Marcus Clarke <marcus@marcuscom.com>

21 years agoLock the vm_object on entry to vm_object_vndeallocate().
alc [Sat, 3 May 2003 20:28:26 +0000 (20:28 +0000)]
Lock the vm_object on entry to vm_object_vndeallocate().

21 years agoFix a broken reference to locale(5) and point to re_format(7) too for an
keramida [Sat, 3 May 2003 20:26:11 +0000 (20:26 +0000)]
Fix a broken reference to locale(5) and point to re_format(7) too for an
explanation of `collating elements'.

21 years agoForget about rev 1.23. -$ is now spelled -fno-dollars-in-identifiers.
obrien [Sat, 3 May 2003 19:57:36 +0000 (19:57 +0000)]
Forget about rev 1.23.  -$ is now spelled -fno-dollars-in-identifiers.

Submitted by: kan

21 years agoGCC 3.3's cpplib does not impliment -$ (forbid the use of $ in identifiers).
obrien [Sat, 3 May 2003 19:50:41 +0000 (19:50 +0000)]
GCC 3.3's cpplib does not impliment -$ (forbid the use of $ in identifiers).

21 years agoFix the misspelling of DPADD, then fix "make checkdpadd".
ru [Sat, 3 May 2003 19:49:05 +0000 (19:49 +0000)]
Fix the misspelling of DPADD, then fix "make checkdpadd".

21 years agoMake GCC 3.3 happy with rcsid[], sccs[], and copyright[].
obrien [Sat, 3 May 2003 19:44:46 +0000 (19:44 +0000)]
Make GCC 3.3 happy with rcsid[], sccs[], and copyright[].

21 years agoUse UFS1 for the floppies instead of UFS2
scottl [Sat, 3 May 2003 19:23:10 +0000 (19:23 +0000)]
Use UFS1 for the floppies instead of UFS2

21 years agoThe author wanted stringification, not concatenation.
obrien [Sat, 3 May 2003 19:22:47 +0000 (19:22 +0000)]
The author wanted stringification, not concatenation.

21 years agoDon't shaddow exp(3).
obrien [Sat, 3 May 2003 19:09:11 +0000 (19:09 +0000)]
Don't shaddow exp(3).

21 years ago- Move bge_phy_hack into the phy code and implement the various DSP
ps [Sat, 3 May 2003 19:06:50 +0000 (19:06 +0000)]
- Move bge_phy_hack into the phy code and implement the various DSP
  patch workarounds for each phy revision.
  Obtained from: NetBSD & Broadcom Linux driver

- Disable AUTOPOLL when accessing the PHY as it may cause PCI errors.
  Obtained from: NetBSD

- Check the UPDATED bit in the status block so the driver knows
  that the status block as indeed changed since the last access.
  Broadcom documentation states drivers should unset the UPDATED/CHANGED
  bits after reading them.

- When changing media types, first loop the phy then set the media.
  Broadcom documentation and Linux drivers do this and I observed
  much better handling of link after this change.

- Broadcom documentation states that for 1000BaseT operation,
  autonegotiation must be enabled.  Fix hard coding of media so that
  the driver only advertises 1000BaseT as the supported media type
  and enable autonegotition.

- Only set Master/Slave on the 5701.
  Obtained from Broadcom Linux driver.

21 years agoUse __FBSDID() to quiet GCC 3.3 warnings.
obrien [Sat, 3 May 2003 18:41:59 +0000 (18:41 +0000)]
Use __FBSDID() to quiet GCC 3.3 warnings.

21 years agoFix locking on reinit.
sos [Sat, 3 May 2003 18:28:43 +0000 (18:28 +0000)]
Fix locking on reinit.

21 years agoAdd the pdc20268r as candidate for the TX test.
sos [Sat, 3 May 2003 18:26:00 +0000 (18:26 +0000)]
Add the pdc20268r as candidate for the TX test.

21 years agoUse C99 for the kernel.
obrien [Sat, 3 May 2003 18:05:56 +0000 (18:05 +0000)]
Use C99 for the kernel.

21 years agoThis commit was generated by cvs2svn to compensate for changes in r114584,
imp [Sat, 3 May 2003 17:16:40 +0000 (17:16 +0000)]
This commit was generated by cvs2svn to compensate for changes in r114584,
which included commits to RCS files with non-trunk default branches.

21 years agoFix halt -p. I've done this with an import because kochi-san reports
imp [Sat, 3 May 2003 17:16:40 +0000 (17:16 +0000)]
Fix halt -p.  I've done this with an import because kochi-san reports
this is fixed in a newer version of ACPICA and I don't want to take
this off the vendor branch for a trivial reason.  This patch was
applied to NetBSD by kochi-san, who also posted the patch to
acpi-jp@jp.freebsd.org.

# My Dell Inspiron 8000 now powers off!

Submitted by: takayoshi kochi-san kochi at netbsd dot org

21 years agoFix a bazillion warnings. This makes almost the whole of src/bin/*
markm [Sat, 3 May 2003 16:39:34 +0000 (16:39 +0000)]
Fix a bazillion warnings. This makes almost the whole of src/bin/*
WARNS=6, std=c99 clean.

Tested on: i386, alpha

21 years agoWARNS cleaning for the Alpha. On alpha, size_t is a long, and it
markm [Sat, 3 May 2003 16:02:52 +0000 (16:02 +0000)]
WARNS cleaning for the Alpha. On alpha, size_t is a long, and it
solicits a warning when used for the '*' in printf("%*d"). Cast
to u_int for universal use.

21 years agoDuplicate "from".
keramida [Sat, 3 May 2003 15:55:23 +0000 (15:55 +0000)]
Duplicate "from".

PR: docs/51724
Submitted by: Christopher Nehren <apeiron@comcast.net

21 years agoRemove some games/ cruft that is no longer of relevance.
markm [Sat, 3 May 2003 15:48:12 +0000 (15:48 +0000)]
Remove some games/ cruft that is no longer of relevance.

21 years agoFix a shedload of warnings, some memory leaks and clean up WARNS
markm [Sat, 3 May 2003 10:21:26 +0000 (10:21 +0000)]
Fix a shedload of warnings, some memory leaks and clean up WARNS
and lint. This is now WARNS=9, std=c99 clean on i386.

21 years agoRegen.
akiyama [Sat, 3 May 2003 10:19:43 +0000 (10:19 +0000)]
Regen.

21 years agoAdd RealTek RTL8150 USB to fast Ethernet controller driver.
akiyama [Sat, 3 May 2003 10:16:56 +0000 (10:16 +0000)]
Add RealTek RTL8150 USB to fast Ethernet controller driver.
This driver now supports the Melco LUA-KTX and the GREEN HOUSE
GH-USB100B.

Reviewed by: imp
MFC after: 2 weeks

21 years agoFix format warning. This is WARNS=9, std=c99 clean on i386.
markm [Sat, 3 May 2003 10:16:51 +0000 (10:16 +0000)]
Fix format warning. This is WARNS=9, std=c99 clean on i386.

21 years agoFix warnings. This is now WARNS=9, std=c99 clean in i386.
markm [Sat, 3 May 2003 10:14:26 +0000 (10:14 +0000)]
Fix warnings. This is now WARNS=9, std=c99 clean in i386.
Use return(0) for main() instead of exit(0). Makes it a
wee bit smaller.

21 years agoMore axe-work:
phk [Sat, 3 May 2003 09:58:20 +0000 (09:58 +0000)]
More axe-work:

Hide all the historical fields of the label, unless people ask for them with -A,
set them to intelligently chosen defaults otherwise.

Distill the manual page to remove inaccuracies, misundertandings and obsolete
information.  It can probably still be done better but now at least it is
not misinforming people.

21 years agoSlightly tweak the rc update text, since things changed slightly after
dougb [Sat, 3 May 2003 09:44:25 +0000 (09:44 +0000)]
Slightly tweak the rc update text, since things changed slightly after
I sent the first draft to Bruce. Remove rc.sendmail from the list, since
it was a mistake on my part to remove it from the base, which I've now
corrected. Add the rc.${MACHINE_ARCH} for arch's besides i386.

While I'm here, do a completely trivial re-capitalization of rcNG. :)

Thanks to Bruce for cleaning up my humble first draft.

21 years agoAdded OpenBSD 3.3
wosch [Sat, 3 May 2003 09:18:25 +0000 (09:18 +0000)]
Added OpenBSD 3.3

Submitted by: itojun@iijlab.net

21 years agoSome minor remodelling with a large axe.
phk [Sat, 3 May 2003 09:02:27 +0000 (09:02 +0000)]
Some minor remodelling with a large axe.

21 years ago - Revert kern/vfs_subr.c revision 1.444. The vm_object's size isn't
alc [Sat, 3 May 2003 08:09:24 +0000 (08:09 +0000)]
 - Revert kern/vfs_subr.c revision 1.444.  The vm_object's size isn't
   trustworthy for vnode-backed objects.
 - Restore the old behavior of vm_object_page_remove() when the end
   of the given range is zero.  Add a comment to vm_object_page_remove()
   regarding this behavior.

Reported by: iedowse

21 years agoUse new geom.ctl based OAM instead of ioctls.
phk [Sat, 3 May 2003 08:04:24 +0000 (08:04 +0000)]
Use new geom.ctl based OAM instead of ioctls.
Various cleanup.

21 years agoAdd a OAM interface for changing the label and writing the boot code.
phk [Sat, 3 May 2003 08:01:34 +0000 (08:01 +0000)]
Add a OAM interface for changing the label and writing the boot code.

21 years ago- Fix breakage on PAE enabled kernel
pdeuskar [Sat, 3 May 2003 07:35:47 +0000 (07:35 +0000)]
- Fix breakage on PAE enabled kernel
- Don't use vtophys when you can get physical address using bus_dma API

Submitted by: jake (Jake Burkholder)

21 years agoremove unused variables.
phk [Sat, 3 May 2003 07:12:15 +0000 (07:12 +0000)]
remove unused variables.

Spotted by: dougb

21 years agoPreserve directory structure in PRESERVED_FILES_DIR to avoid
dougb [Sat, 3 May 2003 06:35:19 +0000 (06:35 +0000)]
Preserve directory structure in PRESERVED_FILES_DIR to avoid
filename collisions.

Submitted by: des

21 years agoMove a declaration to its proper place.
alc [Sat, 3 May 2003 04:21:16 +0000 (04:21 +0000)]
Move a declaration to its proper place.

21 years agoSpell cpu_switch correctly.
peter [Sat, 3 May 2003 03:30:29 +0000 (03:30 +0000)]
Spell cpu_switch correctly.

21 years agoForce PnP devices to set SMIT mode.
nyan [Sat, 3 May 2003 02:04:58 +0000 (02:04 +0000)]
Force PnP devices to set SMIT mode.

Submitted by: chi@bd.mbn.or.jp (Chiharu Shibata)

21 years agoFix off-by-one bug.
nyan [Sat, 3 May 2003 01:47:38 +0000 (01:47 +0000)]
Fix off-by-one bug.

Submitted by: chi@bd.mbn.or.jp (Chiharu Shibata)

21 years agoRepocopy *.s to *.S
peter [Sat, 3 May 2003 00:21:43 +0000 (00:21 +0000)]
Repocopy *.s to *.S

21 years agoFix three problems in large (>2TB) device handling:
ken [Sat, 3 May 2003 00:21:40 +0000 (00:21 +0000)]
Fix three problems in large (>2TB) device handling:
 - Make sure we don't release the READ CAPACITY CCB twice
 - If we have a device that needs a 16 byte READ CAPACITY command, make
   sure we call xpt_schedule() so we can get a CCB.
 - Don't unlock the peripheral until we're fully probed.

Many thanks to Julian Elischer for providing hardware and testing this.

Tested by: julian

21 years agoRename amd64/*.s to amd64/*.S
peter [Sat, 3 May 2003 00:19:42 +0000 (00:19 +0000)]
Rename amd64/*.s to amd64/*.S

21 years agobsd_disklabel_le_dec() takes an extra argument now.
phk [Fri, 2 May 2003 22:46:44 +0000 (22:46 +0000)]
bsd_disklabel_le_dec() takes an extra argument now.

21 years agoMake bsd_disklabel_le_enc calculate the checksum and fill it in.
phk [Fri, 2 May 2003 22:46:13 +0000 (22:46 +0000)]
Make bsd_disklabel_le_enc calculate the checksum and fill it in.
(If there is a legitimate need to correctly encode and pack a
disklabel with an invalid checksum custom tools can be built for
that.)

Make bsd_disklabel_le_dec() validate the magics, number of partitions
(against a new parameter) and the checksum.

Vastly simplify the logic of the GEOM::BSD class implementation:

Let g_bsd_modify() always take a byte-stream label.

This simplifies all users, except the ioctl's which now have to
convert to a byte-stream first.  Their loss.

g_bsd_modify() is called with topology held now, and it returns
with it held.

Always update the md5sum in g_bsd_modify(), otherwise the check
is no use after the first modification of the label.  Make the
MD5 over the bytestream version of the label.

Move the rawoffset hack to g_bsd_modify() and remove all the
inram/ondisk conversions.

Don't configure hotspots in g_bsd_modify(), do it in taste instead,
we do not support moving the label to a different location on the
fly anyway.

This passes all current regression tests.

21 years agoHook rc.sendmail back up, it was removed prematurely
dougb [Fri, 2 May 2003 22:27:31 +0000 (22:27 +0000)]
Hook rc.sendmail back up, it was removed prematurely

21 years ago- Bus DMA'fy the driver
pdeuskar [Fri, 2 May 2003 21:17:08 +0000 (21:17 +0000)]
- Bus DMA'fy the driver
- Use htole* macros where appropriate so that the driver could work on non-x86 architectures
- Use m_getcl() instead of MGETHDR/MCLGET macros
Submitted by: sam (Sam Leffler)

21 years agoOption KADB does not exist. It came from alpha, where it still exists.
marcel [Fri, 2 May 2003 20:34:15 +0000 (20:34 +0000)]
Option KADB does not exist. It came from alpha, where it still exists.

21 years agoVastly simplify architecture handling: Use properties of label as
phk [Fri, 2 May 2003 20:14:48 +0000 (20:14 +0000)]
Vastly simplify architecture handling:  Use properties of label as
recorded in global variables, rather than checks on the architecture.

Drop horribly code to handle MBR/PC98's embedded in the BSD label area.
If you need to have an MBR or PC98 on your disk, you should not overlap
it with a BSDLABEL, if you don't need it, this code is nothing but trouble.

21 years agoDefault the location to the compiled for architecture if no -m arg specified.
phk [Fri, 2 May 2003 20:02:11 +0000 (20:02 +0000)]
Default the location to the compiled for architecture if no -m arg specified.

21 years agoAdd three global variables which contain the location, size and a flag
phk [Fri, 2 May 2003 19:58:08 +0000 (19:58 +0000)]
Add three global variables which contain the location, size and a flag
for the alpha checksum, and set them depending on the specified architecture

Don't look for disklabels every 16 bytes, look the only place they should
be for the current architecture.

Always read the label from the raw disk and decode it into struct
disklabel rather than trust a cast from random addresses.

When writing to the raw disk, encode the label properly.

21 years agoPull in geom_bsd_enc.c from sys/geom to encode and decode our labels.
phk [Fri, 2 May 2003 19:54:18 +0000 (19:54 +0000)]
Pull in geom_bsd_enc.c from sys/geom to encode and decode our labels.

21 years agoPull in bcopy() prototype from <string.h> when compiled in userland.
phk [Fri, 2 May 2003 19:53:55 +0000 (19:53 +0000)]
Pull in bcopy() prototype from <string.h> when compiled in userland.

21 years agoAdd untested support for the Broadcom 5704.
ps [Fri, 2 May 2003 19:53:40 +0000 (19:53 +0000)]
Add untested support for the Broadcom 5704.

Obtained from: NetBSD