kettenis [Fri, 22 Aug 2008 13:02:53 +0000 (13:02 +0000)]
Don't install libbfd and libopcodes and the associated headers.
They really are just a way to share common code between gdb and binutils
(which in the GNU world are seperate packages). The developers (and I'm
one of them) make absolutely no effert to maintain a stable API/ABI.
If ports people really need libbfd or libopcodes, they should just
package a suitable version.
ok weingart@, deraadt@
deraadt [Fri, 22 Aug 2008 12:56:51 +0000 (12:56 +0000)]
sync
brad [Fri, 22 Aug 2008 12:02:33 +0000 (12:02 +0000)]
mention machine/cpu.h for the CPU identifiers.
brad [Fri, 22 Aug 2008 11:51:47 +0000 (11:51 +0000)]
mention machine/cpu.h for the CPU identifiers.
kurt [Fri, 22 Aug 2008 10:41:37 +0000 (10:41 +0000)]
Set PIE address range to avoid or minimize mmap pressure due to limited
address space. The space between PAGE_SIZE and the non-pie fixed link
address creates no mmap pressure so use that space for PIE. However on
hppa the non-pie fixed link address is PAGE_SIZE so just use a small range
for PIE to minimize mmap pressure.
okay miod@
sobrado [Fri, 22 Aug 2008 08:52:35 +0000 (08:52 +0000)]
default file transfer type is binary, not ascii;
commands which toggle settings can take an explicit on or off argument
to force the setting appropiately, show these arguments in usage;
synchronize synopsis and usage of commands; spacing; KNF;
other minor documentation tweaks.
written with help by jmc@
ok jmc@ (documentation), martynas@ (type usage and default file transfer type)
jmc [Fri, 22 Aug 2008 06:41:36 +0000 (06:41 +0000)]
- -R before -r in SYNOPSIS/usage()
- avoid marking up `|'
deraadt [Fri, 22 Aug 2008 03:19:02 +0000 (03:19 +0000)]
simplification, and repair findblkmajor for name## lookups; ok krw
deraadt [Fri, 22 Aug 2008 03:12:37 +0000 (03:12 +0000)]
ANSI and other cleanups; ok krw
marco [Fri, 22 Aug 2008 02:00:12 +0000 (02:00 +0000)]
Add a hack to manually kick off rebuilds. This will need some love in the
near future.
ok deraadt
deraadt [Fri, 22 Aug 2008 01:09:11 +0000 (01:09 +0000)]
synbc
deraadt [Fri, 22 Aug 2008 01:09:07 +0000 (01:09 +0000)]
more puc happiness
jmc [Fri, 22 Aug 2008 01:06:31 +0000 (01:06 +0000)]
kill trailing whitespace;
deraadt [Fri, 22 Aug 2008 00:59:34 +0000 (00:59 +0000)]
After spotting a + record, continue scanning and pick up later groups
in the file. Problem spotted by drahn.
ok millert
millert [Fri, 22 Aug 2008 00:56:13 +0000 (00:56 +0000)]
Replace the old algorithm that included the process id as part of the
temporary file name with one that only uses random data. OK deraadt@
bluhm [Fri, 22 Aug 2008 00:48:33 +0000 (00:48 +0000)]
Test the IPv6 address printing of pf_print_host() in net/pf.c.
help and ok mpf
bluhm [Fri, 22 Aug 2008 00:35:08 +0000 (00:35 +0000)]
Make pf_print_host() print IPv6 addresses correctly.
ok mpf
bluhm [Thu, 21 Aug 2008 23:05:30 +0000 (23:05 +0000)]
Assign the ip and ip6 pointers in ipsp_process_packet() only if a
header of the matching address family is available. Especially do
not read ip->ip_off from an IPv6 packet header.
ok markus
simon [Thu, 21 Aug 2008 21:57:45 +0000 (21:57 +0000)]
Add an optional separator to stringize() and use it from do_list() to
match the GNU pkg-config --list-all output.
ok ckuethe@
oga [Thu, 21 Aug 2008 21:43:56 +0000 (21:43 +0000)]
Rip out the remnants of the vblank signal stuff. Nothing uses it, and
it's stupid anyway. Most of this was ifdefed out.
espie [Thu, 21 Aug 2008 21:02:28 +0000 (21:02 +0000)]
Document new gnu extensions, okay jmc@, otto@
espie [Thu, 21 Aug 2008 21:01:47 +0000 (21:01 +0000)]
in gnu-m4 mode, allow definition of the empty macro (yes, autoconf 2.62
actually uses this, bleh)
okay otto@
espie [Thu, 21 Aug 2008 21:01:04 +0000 (21:01 +0000)]
extend format support to cater to recent GNU autoconf
okay otto@, some useful ideas from miod@
espie [Thu, 21 Aug 2008 21:00:14 +0000 (21:00 +0000)]
gnu extension: 0rN:az for baseN numbers.
okay otto@
espie [Thu, 21 Aug 2008 20:59:14 +0000 (20:59 +0000)]
use unsigned char for buffers, allow differentiating between EOF and y".
okay otto@
oga [Thu, 21 Aug 2008 18:23:17 +0000 (18:23 +0000)]
Instead of having a number of malloced arrays for vblank handling, just
put them in a structure and malloc free an array of those, it is nicer
that way and They're all needed at the same time, anyway.
Idea taken from a suggestion on the dri-devel mailing list.
jmc [Thu, 21 Aug 2008 18:16:13 +0000 (18:16 +0000)]
+ECHOK; from Markus Bergkvist
ok deraadt
millert [Thu, 21 Aug 2008 16:54:44 +0000 (16:54 +0000)]
Remove useless code, the kernel will set errno appropriately if an
element in the path does not exist. OK deraadt@ pvalchev@
ragge [Thu, 21 Aug 2008 16:39:54 +0000 (16:39 +0000)]
Add missing dependency for external.h, from Don Hiatt.
ragge [Thu, 21 Aug 2008 16:34:10 +0000 (16:34 +0000)]
From master repo: Accept // inside #if directives.
jmc [Thu, 21 Aug 2008 05:07:04 +0000 (05:07 +0000)]
tweak previous;
jmc [Thu, 21 Aug 2008 05:04:02 +0000 (05:04 +0000)]
- add missing word in re.4
- update re(4) description in cardbus.4
djm [Thu, 21 Aug 2008 04:09:57 +0000 (04:09 +0000)]
allow ForceCommand internal-sftp with arguments. based on patch from
michael.barabanov AT gmail.com; ok markus@
pedro [Thu, 21 Aug 2008 01:17:39 +0000 (01:17 +0000)]
Fix 'if ();' construction introduced in last commit, okay deraadt@
jsg [Wed, 20 Aug 2008 22:55:33 +0000 (22:55 +0000)]
update for 8168/8102E support.
deraadt [Wed, 20 Aug 2008 22:47:12 +0000 (22:47 +0000)]
sync
dlg [Wed, 20 Aug 2008 22:19:49 +0000 (22:19 +0000)]
fix the comment about enabling socket map support
noticed by ajacoutot
pvalchev [Wed, 20 Aug 2008 20:35:52 +0000 (20:35 +0000)]
hppa autoboot instructions for new and old PDC; help & ok miodski@
miod [Wed, 20 Aug 2008 18:59:59 +0000 (18:59 +0000)]
Minimal driver for the VAXstation 35x0/38x0 LEGSS option, currently limited
to 8 bit mode operation, and no color or accelerated features until I can
find documentation about it.
Speed is decent by VAX standards, except for scrolling, which is so
abysmally slow one could see ZZ Top's beards growing while waiting for the
screen to scroll...
miod [Wed, 20 Aug 2008 18:55:24 +0000 (18:55 +0000)]
Do not mask bit 0x80 in dzcngetc() - we run serial consoles in 8N1.
miod [Wed, 20 Aug 2008 18:52:07 +0000 (18:52 +0000)]
Don't copy interesting parts of struct scsi_xfer to a read only internal
ScsiCmd structure. Since we keep a pointer to the scsi_xfer nevertheless,
we can get the information from it directly.
miod [Wed, 20 Aug 2008 18:50:17 +0000 (18:50 +0000)]
Attach led0 on ka60, and display system load on the front panel.
miod [Wed, 20 Aug 2008 18:49:12 +0000 (18:49 +0000)]
If RI_FORCEMONO is set in ri_flags, build ri_devcmap as if it was a 1bpp
frame buffer.
otto [Wed, 20 Aug 2008 18:29:37 +0000 (18:29 +0000)]
mention t3 fails
otto [Wed, 20 Aug 2008 18:28:46 +0000 (18:28 +0000)]
The code to detect multiple applied diffs to empty files I introduced 4
years ago is not correct if the diff has an empty context for other
reasons, so revert that change. Problem found by Joerg Sonnenberger;
ok tedu@ millert@
deraadt [Wed, 20 Aug 2008 18:21:13 +0000 (18:21 +0000)]
re-enable iso building
jsing [Wed, 20 Aug 2008 18:04:44 +0000 (18:04 +0000)]
Add USER_PCICONF so that pcidump can work.
Prompted by jasper@
ok miod@
miod [Wed, 20 Aug 2008 16:31:41 +0000 (16:31 +0000)]
Oops, previous console changes would initialize serial console on lines 0
(e.g. on microVAXes) at 4800 bps as if they were keyboard lines.
ratchov [Wed, 20 Aug 2008 14:22:50 +0000 (14:22 +0000)]
abuf_flush() must be called only from the ``xxx_in()'' routines (ie
triggered by POLLIN condition).
mix_pushzero() is not called from the ``xxx_in()'' chain, but calls
abuf_flush() resulting in bad initialization of the mixer, which
will lead to a deadlock at some point.
ok jakemsr
dlg [Wed, 20 Aug 2008 12:38:41 +0000 (12:38 +0000)]
enable support for socket maps.
ok millert@
thib [Wed, 20 Aug 2008 12:09:23 +0000 (12:09 +0000)]
remove a few unused files that where probably added by mistake;
spotted in a commit too FreeBSD;
ok todd@, oga@, krw@
thib [Wed, 20 Aug 2008 12:08:04 +0000 (12:08 +0000)]
Return EINVAL instead of '1' for invalid kq filters.
From NetBSD;
OK art@, blambert@
mpf [Wed, 20 Aug 2008 09:29:51 +0000 (09:29 +0000)]
Regression tests for gzip(1):
- Test if we detect truncated or corrupted files.
- Test basic functionality
OK millert@, markus@
mpf [Wed, 20 Aug 2008 09:22:02 +0000 (09:22 +0000)]
Use and report errors that inflate(3) can return, instead of
aborting silently, and pretending everything went alright.
This lets gzip(1) (especially gzip -t) detect truncated archives
and curruptions that violate the structure of the zlib format.
Unquiet compress(1) in zless/zmore, so the user has a chance
to see errors on corrupted archives.
OK millert@, markus@
miod [Wed, 20 Aug 2008 04:37:15 +0000 (04:37 +0000)]
Allow devices attaching at root (such as softraid) to be disabled in UKC.
ok marco@ deraadt@
deraadt [Tue, 19 Aug 2008 11:52:15 +0000 (11:52 +0000)]
sync
markus [Tue, 19 Aug 2008 10:37:12 +0000 (10:37 +0000)]
use the actual keysize from the SA to figure out which AES variant
(aes-192, aes-256) is used; ok hshoexer@
pedro [Tue, 19 Aug 2008 09:49:50 +0000 (09:49 +0000)]
When failing to acquire the vnode lock in readdir_with_callback(), don't
try to release it before returning, okay fgsch@ and tedu@.
pedro [Tue, 19 Aug 2008 09:41:12 +0000 (09:41 +0000)]
Look at 'fs_flags' when dumping FFS2 file systems, okay beck@, millert@
and otto@.
kettenis [Tue, 19 Aug 2008 08:26:20 +0000 (08:26 +0000)]
Make it possible to enter ddb from the serial console.
ok miod@, deraadt@
kettenis [Tue, 19 Aug 2008 07:59:19 +0000 (07:59 +0000)]
Make it possible to enter ddb from the serial console.
ok miod@, deraadt@
jmc [Tue, 19 Aug 2008 06:42:25 +0000 (06:42 +0000)]
+D-Link DWA-110 to the rum(4) list;
deraadt [Tue, 19 Aug 2008 04:30:20 +0000 (04:30 +0000)]
sync
deraadt [Tue, 19 Aug 2008 02:37:26 +0000 (02:37 +0000)]
more
deraadt [Tue, 19 Aug 2008 02:34:08 +0000 (02:34 +0000)]
sync
deraadt [Tue, 19 Aug 2008 02:34:04 +0000 (02:34 +0000)]
another rum; Jonathan Lee
deraadt [Tue, 19 Aug 2008 02:02:02 +0000 (02:02 +0000)]
add missing devices to nam2blk[]
miod [Mon, 18 Aug 2008 23:25:53 +0000 (23:25 +0000)]
Forgot to add this file, needing for KA60 support.
miod [Mon, 18 Aug 2008 23:20:43 +0000 (23:20 +0000)]
Now that KA60 support is in, include the proper header files instead of
harcoding defines, and add rpb device type #39 (sii) boot information.
miod [Mon, 18 Aug 2008 23:19:15 +0000 (23:19 +0000)]
Add support for the ``Firefox'' VAXstation 3520/3540/3820/3840 workstations,
currently limited to serial console and a single processor working.
All ``on-board'' devices, including the Q-bus adapter, but except for
the frame buffer, are supported. The machine will boot over the network
or from SCSI devices.
Lots of thanks to Al Kossow for www.bitsavers.org, on which I found the
technical documentation allowing me to complete this port (which was
lacking at the time I got that machine...).
miod [Mon, 18 Aug 2008 23:10:39 +0000 (23:10 +0000)]
Some uglyness to allow both uba_common.h and ubareg.h to be included;
temporary until they get merged for good (or split differently).
miod [Mon, 18 Aug 2008 23:09:42 +0000 (23:09 +0000)]
Increment device interrupt event counter when redispatching an interrupt.
miod [Mon, 18 Aug 2008 23:08:55 +0000 (23:08 +0000)]
Preparing for KA60 support, crank IPL_TTY to level 0x16 (since it is
hardwired this way on KA60...), and compensate in splassert_check() when
serial device interrupts will remain at 0x14 or 0x15.
miod [Mon, 18 Aug 2008 23:07:22 +0000 (23:07 +0000)]
Move common CVAX defines and code (internal cache IPR, SSC layout, machine
check frame format...) to a single place. Get rid of duplicated KAxx_FOO
constants.
miod [Mon, 18 Aug 2008 23:05:37 +0000 (23:05 +0000)]
Make hardclock() a cpu_dep member. Most machines will use the generic
hardclock() here, except for KA46 which needs to reset the diagnostic
timer.
This gives us working clock interrupt counters for vmstat -i in the process.
miod [Mon, 18 Aug 2008 23:04:26 +0000 (23:04 +0000)]
Shuffle the dz console code (both serial and keyboard) to be more independent
of the console device attachment, and to be able to work on system where there
are more than one dz device of the same type, one of it being the console one.
This also allows a different register layout than the dz@ibus one to be used
for the dz console device.
brad [Mon, 18 Aug 2008 22:28:58 +0000 (22:28 +0000)]
regen
brad [Mon, 18 Aug 2008 22:28:36 +0000 (22:28 +0000)]
Add the PCI id for the Broadcom BCM5716 Ethernet controller.
kettenis [Mon, 18 Aug 2008 21:55:01 +0000 (21:55 +0000)]
Add stge(4) and gentbi(4).
ragge [Mon, 18 Aug 2008 20:54:10 +0000 (20:54 +0000)]
Updates from master repo: Fix two bugs reported by Art Grabowski.
brad [Mon, 18 Aug 2008 20:29:37 +0000 (20:29 +0000)]
Add sparc_bus_addr function for schizo to allow CardBus to work.
ok kettenis@
kevlo [Mon, 18 Aug 2008 04:28:18 +0000 (04:28 +0000)]
use M_NOWAIT instead of M_WAITOK to cause malloc() to return NULL
ok mglocker
oga [Mon, 18 Aug 2008 00:01:49 +0000 (00:01 +0000)]
Another dead struct member.
oga [Sun, 17 Aug 2008 23:49:02 +0000 (23:49 +0000)]
kill struct drm_freelist, after removing markbufs and infobufs nothing
touches it.
oga [Sun, 17 Aug 2008 23:36:22 +0000 (23:36 +0000)]
Kill a dead function and a dead struct member.
oga [Sun, 17 Aug 2008 21:31:38 +0000 (21:31 +0000)]
Check that the driver properly initialised before allowing agp_acquire()
to succeed.
Inspired by Tobias Ulmer, thanks!
oga [Sun, 17 Aug 2008 20:49:52 +0000 (20:49 +0000)]
Check the return value of agp_enable() instead of assuming it completed.
apply some knf and other formatting while i'm at it.
problem originally pointed out by Tobias Ulmer, thanks!
oga [Sun, 17 Aug 2008 19:21:36 +0000 (19:21 +0000)]
Kill the infobufs and markbufs calls. Nothing uses them.
ragge [Sun, 17 Aug 2008 18:40:12 +0000 (18:40 +0000)]
Updates from master repo. Can now compile both userland and kernel.
(with some patches...)
oga [Sun, 17 Aug 2008 16:56:14 +0000 (16:56 +0000)]
kill the stupid pci capability defines, we have our own. Don't wrap
pci_read_capability while i'm at it.
oga [Sun, 17 Aug 2008 16:13:10 +0000 (16:13 +0000)]
drm_load() and drm_unload() are a remnant of code meant to be shared
between the bsds. Just put them in attach/detach instead. Kills a little
bit of dead code, too.
kettenis [Sun, 17 Aug 2008 15:55:55 +0000 (15:55 +0000)]
Scale down cpu_clockrate[1] when we scale down the cpu clock frequency such
that delay(9) continues to do the right thing.
oga [Sun, 17 Aug 2008 15:31:39 +0000 (15:31 +0000)]
Kill some redundant ifdefs, they're taken care of elsewhere.
krw [Sun, 17 Aug 2008 15:14:44 +0000 (15:14 +0000)]
Force fdisk to use the same geometry as the miniroot2.5M disktab
entry, the one that the disklabel will use, rather than the vnd
default geometry. Lets the generated miniroot44.fs boot again.
Problem found by Diana Eichart. Suggestions from drahn@.
ok deraadt@
oga [Sun, 17 Aug 2008 14:37:31 +0000 (14:37 +0000)]
Make sure we can't sleep with a spinlock held
kettenis [Sun, 17 Aug 2008 14:25:19 +0000 (14:25 +0000)]
Garbage collect stupid delay loop.
mglocker [Sat, 16 Aug 2008 18:56:07 +0000 (18:56 +0000)]
If dwMaxPayloadTransferSize doesn't exactly match to an existing
endpoint, pick the next higher endpoint bandwidth.
mglocker [Sat, 16 Aug 2008 16:47:32 +0000 (16:47 +0000)]
Spacing.
mglocker [Sat, 16 Aug 2008 16:37:30 +0000 (16:37 +0000)]
Sync ehci isochronous part with NetBSD:
- Fix check for maximum bInterval value.
- Calculate frames/microframes values slightly different (but with mostly
same result finally).
espie [Sat, 16 Aug 2008 12:26:07 +0000 (12:26 +0000)]
fix