dragonfly.git
17 years agoMerge from FreeBSD:
Hiten Pandya [Mon, 9 Feb 2004 14:03:41 +0000 (14:03 +0000)]
Merge from FreeBSD:

Skip all devices whose header type  is > 2; '0' is normal, '1'
is a PCI Bridge and '2' is a Cardbus bridge.  Those are PCI 2.2
defines.

17 years agoremove unwanted __DragonFly__ >= tests
Robert Garrett [Mon, 9 Feb 2004 12:03:36 +0000 (12:03 +0000)]
remove unwanted __DragonFly__ >= tests

17 years ago__FreeBSD__ -> __DragonFly__
Robert Garrett [Sun, 8 Feb 2004 10:59:22 +0000 (10:59 +0000)]
__FreeBSD__ -> __DragonFly__

17 years agoThe last commit was noticed by Yonetani Tomokazu
Robert Garrett [Sun, 8 Feb 2004 10:54:25 +0000 (10:54 +0000)]
The last commit was noticed by Yonetani Tomokazu

17 years agoFix a couple of typos that slipped in
Robert Garrett [Sun, 8 Feb 2004 10:51:18 +0000 (10:51 +0000)]
Fix a couple of typos that slipped in

17 years agoPatch this file to allow the crossbuild to work again
Robert Garrett [Sun, 8 Feb 2004 10:48:30 +0000 (10:48 +0000)]
Patch this file to allow the crossbuild to work again
if all my tests are correct, anyway.

thanks expec to Ed Germain

17 years agoMerge from FreeBSD:
Hiten Pandya [Sun, 8 Feb 2004 07:10:46 +0000 (07:10 +0000)]
Merge from FreeBSD:

Replace an outb() with a DELAY(1) during the test for configuration
mechanism #1.  This fixes probing for PCI on some systems.

17 years agoMerge from FreeBSD:
Hiten Pandya [Sun, 8 Feb 2004 07:03:17 +0000 (07:03 +0000)]
Merge from FreeBSD:

Do not set the No_CRC bit in the Mode Control Register.

17 years agoMerge from FreeBSD:
Hiten Pandya [Sun, 8 Feb 2004 06:51:54 +0000 (06:51 +0000)]
Merge from FreeBSD:

By not setting No CRC in the Mode Control Register, we must
also reduce the size of the packet by 4 bytes to remvoe the
Ethernet CRC.

17 years agoMerge from FreeBSD:
Hiten Pandya [Sun, 8 Feb 2004 06:47:35 +0000 (06:47 +0000)]
Merge from FreeBSD:

Add support for `Allied Telesis SIC-AT' boards.

17 years agoMerge from FreeBSD:
Hiten Pandya [Sun, 8 Feb 2004 06:01:16 +0000 (06:01 +0000)]
Merge from FreeBSD:

Don't use the cached inp->inp_route unless it is marked
as RTF_UP; this fixes a panic when an interface is detached
(in event of suspend/resume).

17 years agoMerge from FreeBSD:
Hiten Pandya [Sun, 8 Feb 2004 05:56:10 +0000 (05:56 +0000)]
Merge from FreeBSD:

Set fs_ronly to the correct value in ffs_reload() when
reloading the file system super block after fsck has
repaired the file system.

17 years agoMerge from FreeBSD:
Hiten Pandya [Sun, 8 Feb 2004 05:44:58 +0000 (05:44 +0000)]
Merge from FreeBSD:

Return a size related error, instead of return 0 when *NOT*
creating a core dump due to resource limits specifying the
size as 0.  Otherwise waitpid() will incorrectly return a
status indicating that a core dump was created.

17 years agoMerge from FreeBSD:
Hiten Pandya [Sun, 8 Feb 2004 05:27:42 +0000 (05:27 +0000)]
Merge from FreeBSD:

Provide better return values for nfs_read();

17 years agoCosmetic code cleanup.
Jeffrey Hsu [Sun, 8 Feb 2004 00:11:02 +0000 (00:11 +0000)]
Cosmetic code cleanup.

17 years agoAdd a tunable hw.pci_disable_bios_route to work around broken PCI-BIOSes.
Joerg Sonnenberger [Sat, 7 Feb 2004 15:56:58 +0000 (15:56 +0000)]
Add a tunable hw.pci_disable_bios_route to work around broken PCI-BIOSes.

Many BIOS implementations don't route an interrupt to cardbus bridges
by default and some doesn't provide a working call. This results in
panics on boot up. For the OLDCARD pcic, it was possible to force the
interrupt-less operation, but NEWCARD doesn't have such an option and
will never have.

17 years agoRemove stale text in front of the DragonFly and FreeBSD keywords
Eirik Nygaard [Sat, 7 Feb 2004 15:41:13 +0000 (15:41 +0000)]
Remove stale text in front of the DragonFly and FreeBSD keywords

17 years agoAdjust infrastructure for NEWCARD
Joerg Sonnenberger [Fri, 6 Feb 2004 23:09:36 +0000 (23:09 +0000)]
Adjust infrastructure for NEWCARD

17 years agoDe-K&R-ify source, remove register keywords.
Joerg Sonnenberger [Fri, 6 Feb 2004 22:36:50 +0000 (22:36 +0000)]
De-K&R-ify source, remove register keywords.

Patch submited by Chris Pressey <cpressey@catseye.mine.nu>

Additionally cleanup some prototyps.

17 years agoAllows option to be specified on the command line when mount with -a.
Joerg Sonnenberger [Fri, 6 Feb 2004 22:11:48 +0000 (22:11 +0000)]
Allows option to be specified on the command line when mount with -a.

Patch submitted by Chis Pressey <cpressey@catseye.mine.nu>, backported
from FreeBSD.

17 years agoRemove the pca driver from the build for now, due to the revamp
Hiten Pandya [Fri, 6 Feb 2004 15:31:58 +0000 (15:31 +0000)]
Remove the pca driver from the build for now, due to the revamp
done recently on the timer/clock code. [1]

I have put an XXX comment, which means that we must add it back
once the PCM driver (pca) has been adjusted with the new timer
code.

Submitted by: YONETANI Tomokazu <qhwt+dragonfly-bugs@les.ath.cx> [1]

17 years agoFix build of the PPS driver.
Hiten Pandya [Fri, 6 Feb 2004 15:26:20 +0000 (15:26 +0000)]
Fix build of the PPS driver.

Submitted by: YONETANI Tomokazu <qhwt+dragonfly-bugs@les.ath.cx>

17 years ago__FreeBSD__ -> __DragonFly__
Robert Garrett [Fri, 6 Feb 2004 09:17:41 +0000 (09:17 +0000)]
__FreeBSD__ -> __DragonFly__

17 years agoadd code to test for single files, instead of just full paths
Robert Garrett [Fri, 6 Feb 2004 08:49:55 +0000 (08:49 +0000)]
add code to test for single files, instead of just full paths

17 years agoundefine __FreeBSD__ it is no longer needed, since the kernel include
Robert Garrett [Fri, 6 Feb 2004 03:23:50 +0000 (03:23 +0000)]
undefine __FreeBSD__ it is no longer needed, since the kernel include
has been fixed

17 years ago__FreeBSD__ -> __DragonFly__
Robert Garrett [Fri, 6 Feb 2004 03:20:47 +0000 (03:20 +0000)]
__FreeBSD__ -> __DragonFly__

17 years ago__FreeBSD__ -> __DragonFly__
Robert Garrett [Thu, 5 Feb 2004 21:03:37 +0000 (21:03 +0000)]
__FreeBSD__ -> __DragonFly__

17 years agoFix bug when doing backquote expansion.
Joerg Sonnenberger [Thu, 5 Feb 2004 19:07:24 +0000 (19:07 +0000)]
Fix bug when doing backquote expansion.

Noted and fixed by Andrew Atrens <atrens@nortelnetworks.com>

17 years agonectar 2004/02/05 10:00:35 PST
David Rhodus [Thu, 5 Feb 2004 18:43:22 +0000 (18:43 +0000)]
nectar      2004/02/05 10:00:35 PST

  FreeBSD src repository

  Modified files:
    sys/kern             sysv_shm.c
  Log:
  Correct a reference counting bug in shmat(2).  If vm_map_find(9)
  failed, the reference count for the virtual memory object referenced
  by the specified shared memory segment would have been erroneously
  incremented.

  Reported by:    Joost Pol <joost@pine.nl>

17 years agoSecond part of the firewire sync. Add defined(__DragonFly__) or
Joerg Sonnenberger [Thu, 5 Feb 2004 17:51:44 +0000 (17:51 +0000)]
Second part of the firewire sync. Add defined(__DragonFly__) or
defined(__FreeBSD__) to all checks of __FreeBSD_version to prepare
the kernel for GCC 3.3.

17 years agoAdd missing sbp.h to complete the firewire sync
Joerg Sonnenberger [Thu, 5 Feb 2004 14:32:53 +0000 (14:32 +0000)]
Add missing sbp.h to complete the firewire sync

17 years agoSync DragonFly and FreeBSD-current's FireWire driver.
Joerg Sonnenberger [Thu, 5 Feb 2004 13:32:08 +0000 (13:32 +0000)]
Sync DragonFly and FreeBSD-current's FireWire driver.

Patch submitted by Hidetoshi Shimokawa <simokawa@sat.t.u-tokyo.ac.jp>

Modified to leave the sbp hack inplace.

17 years agoUpdate style(9) to reflect current code practise.
Joerg Sonnenberger [Thu, 5 Feb 2004 11:51:54 +0000 (11:51 +0000)]
Update style(9) to reflect current code practise.

Patch submitted by Chris Pressey <cpressey@catseye.mine.nu>

17 years agoAllow world to build with gcc3, basically force openssl to use gettimeofday
Robert Garrett [Thu, 5 Feb 2004 04:22:22 +0000 (04:22 +0000)]
Allow world to build with gcc3, basically force openssl to use gettimeofday
instead of ftime()

17 years agoDisable -fformat-extensions temporarily
Robert Garrett [Wed, 4 Feb 2004 21:09:17 +0000 (21:09 +0000)]
Disable -fformat-extensions temporarily

17 years agoFix various warnings.
Joerg Sonnenberger [Wed, 4 Feb 2004 17:40:01 +0000 (17:40 +0000)]
Fix various warnings.

Patch submitted by Simon "corecode" Schubert.

17 years agomore __FreeBSD__ and __FreeBSD__ -> __DragonFly__ work.
Robert Garrett [Wed, 4 Feb 2004 15:39:54 +0000 (15:39 +0000)]
more __FreeBSD__ and __FreeBSD__ -> __DragonFly__ work.

with this commit, world now builds with gcc3,
however this is not completed, until the kernel
files are cleaned, and the __FreeBSD__ definitions removed,
as well as contrib being fixed.

17 years agodefine __FreeBSD__ for things that use kernel source that needs it for now
Robert Garrett [Wed, 4 Feb 2004 14:36:11 +0000 (14:36 +0000)]
define __FreeBSD__ for things that use kernel source that needs it for now

17 years ago__FreeBSD__ defines
Robert Garrett [Wed, 4 Feb 2004 14:34:36 +0000 (14:34 +0000)]
__FreeBSD__ defines

17 years agoDefine __FreeBSD__ in the make file for ipmon
Robert Garrett [Wed, 4 Feb 2004 01:02:14 +0000 (01:02 +0000)]
Define __FreeBSD__ in the make file for ipmon

17 years ago__FreeBSD__ -> __DragonFly__
Robert Garrett [Wed, 4 Feb 2004 00:52:55 +0000 (00:52 +0000)]
__FreeBSD__ -> __DragonFly__

17 years agodhclient needs __FreeBSD__ defined
Robert Garrett [Wed, 4 Feb 2004 00:48:39 +0000 (00:48 +0000)]
dhclient needs __FreeBSD__ defined

17 years ago__FreeBSD__ -> __DragonFly__
Robert Garrett [Tue, 3 Feb 2004 23:24:50 +0000 (23:24 +0000)]
__FreeBSD__ -> __DragonFly__

17 years agoMore __FreeBSD__ -> __DragonFly__ translation
Matthew Dillon [Tue, 3 Feb 2004 19:22:59 +0000 (19:22 +0000)]
More __FreeBSD__ -> __DragonFly__ translation

17 years agoThis should fix the dependancy loop for sure.
Matthew Dillon [Tue, 3 Feb 2004 19:06:06 +0000 (19:06 +0000)]
This should fix the dependancy loop for sure.

17 years agoFake __FreeBSD__ for sendmail contrib use.
Matthew Dillon [Tue, 3 Feb 2004 18:54:01 +0000 (18:54 +0000)]
Fake __FreeBSD__ for sendmail contrib use.

17 years agoTry to clean up more of GCC's xrealloc and xmalloc wrapper mess so
Matthew Dillon [Tue, 3 Feb 2004 18:06:06 +0000 (18:06 +0000)]
Try to clean up more of GCC's xrealloc and xmalloc wrapper mess so
buildworld builds.

17 years agogrep needs __FreeBSD__ defined
Robert Garrett [Tue, 3 Feb 2004 10:14:15 +0000 (10:14 +0000)]
grep needs __FreeBSD__ defined

17 years agodefine __FreeBSD__ for diff
Robert Garrett [Tue, 3 Feb 2004 10:04:04 +0000 (10:04 +0000)]
define __FreeBSD__ for diff

17 years agoAdd warning about memory usage to resident.conf and examples to both files
Robert Garrett [Tue, 3 Feb 2004 09:21:23 +0000 (09:21 +0000)]
Add warning about memory usage to resident.conf and examples to both files

17 years agochange FreeBSD in examples to DragonFly
Robert Garrett [Tue, 3 Feb 2004 08:43:41 +0000 (08:43 +0000)]
change FreeBSD in examples to DragonFly

17 years agoUpdate a few macros to print DragonFly instead of FreeBSD
Robert Garrett [Tue, 3 Feb 2004 08:37:33 +0000 (08:37 +0000)]
Update a few macros to print DragonFly instead of FreeBSD

17 years agoRemove obsolete __FreeBSD_version checks, add __DragonFly__ tests to
Matthew Dillon [Tue, 3 Feb 2004 07:34:10 +0000 (07:34 +0000)]
Remove obsolete __FreeBSD_version checks, add __DragonFly__ tests to
existing __FreeBSD__ checks.

17 years agoAdd DragonFly partition type and change FreeBSD in warnings to DragonFly
Robert Garrett [Tue, 3 Feb 2004 07:13:10 +0000 (07:13 +0000)]
Add DragonFly partition type and change FreeBSD in warnings to DragonFly

17 years agoChange __FreeBSD__ -> __DragonFly__
Matthew Dillon [Tue, 3 Feb 2004 07:11:53 +0000 (07:11 +0000)]
Change __FreeBSD__ -> __DragonFly__

17 years agoFake __FreeBSD__ for various contrib/ code that needs it.
Matthew Dillon [Tue, 3 Feb 2004 06:26:32 +0000 (06:26 +0000)]
Fake __FreeBSD__ for various contrib/ code that needs it.

17 years agoAdjust osreldate.h to only set __FreeBSD_version if __FreeBSD__ is set,
Matthew Dillon [Tue, 3 Feb 2004 05:51:51 +0000 (05:51 +0000)]
Adjust osreldate.h to only set __FreeBSD_version if __FreeBSD__ is set,
so we don't pollute the namespace when only __DragonFly__ is set.
__DragonFly_version is always set.

17 years agoForgot the contribited by line
Robert Garrett [Tue, 3 Feb 2004 05:26:48 +0000 (05:26 +0000)]
Forgot the contribited by line

Contributed by:
   Dylan Reinhold

17 years agoAdd -DSETPWENT_VOID to CFLAGS to work around a __FreeBSD__ conditional in
Matthew Dillon [Tue, 3 Feb 2004 05:12:51 +0000 (05:12 +0000)]
Add -DSETPWENT_VOID to CFLAGS to work around a __FreeBSD__ conditional in
contrib/bind.

17 years agoupdate this file correctly
Robert Garrett [Tue, 3 Feb 2004 04:50:47 +0000 (04:50 +0000)]
update this file correctly

I munged the patch the first time round

17 years agoMake the 'realclean' target respect KERNCONF.
Matthew Dillon [Tue, 3 Feb 2004 04:26:01 +0000 (04:26 +0000)]
Make the 'realclean' target respect KERNCONF.

Submitted-by: Peter Kadau <peter.kadau@tuebingen.mpg.de>
17 years agoAdjust __FreeBSD__ -> __DragonFly_
Matthew Dillon [Tue, 3 Feb 2004 04:23:25 +0000 (04:23 +0000)]
Adjust __FreeBSD__ -> __DragonFly_

17 years agoUndo the xmalloc->malloc optimization FreeBSD made in certain cases
Matthew Dillon [Tue, 3 Feb 2004 03:47:12 +0000 (03:47 +0000)]
Undo the xmalloc->malloc optimization FreeBSD made in certain cases
where xmalloc gets defined to malloc which gets defined to xmalloc which
gets defined to malloc again which gets defined to ... and so on.
Culprits: parser files, gcc/system.h, dragonfly-native.h, and various
Makefile's.

Add [DragonFly] to gcc3's version ident.

17 years agoUpdate the GCC3 infrastructure Stage 1/2. This commit generates the basic
Matthew Dillon [Mon, 2 Feb 2004 23:35:12 +0000 (23:35 +0000)]
Update the GCC3 infrastructure Stage 1/2.  This commit generates the basic
infrastructure needed to support /usr/src/contrib/gcc-3.3 (a pristine gcc
dist) intead of /usr/src/contrib/gcc3 (munged version taken from FreeBSD-5).

This stage does not have all the fbsd/dfly specific extensions, such as
-fformat-extensions.

17 years agoAMD64 infrastructure stage 1/99. This is just a preliminary commit. Many
Matthew Dillon [Mon, 2 Feb 2004 08:05:56 +0000 (08:05 +0000)]
AMD64 infrastructure stage 1/99.  This is just a preliminary commit.  Many
of these files are being brought in from FreeBSD-5.  The noteable exception
is sys/amd64/pmap.h, which contains some #if 0'd FreeBSD-5 stuff but has
the beginnings of a new MMU infrastructure.  NOTHING WILL BUILD YET!

The DFly 64 bit AMD work will have a lot of similarities to the FreeBSD-5
work, but the PMAP/MMU/VM subsystem support is going to be written from
scratch and diverge wildly from F-5.

In particular, it is my intent to entirely do-away with %fs for per-cpu
storage references and instead use the top level PML4 page table (512GB
per entry, 256TB total VM) as a fixed per-cpu entity.  Each cpu will get
its own PML4 table.  Initially only one entry (512GB of VM) in PML4 will
be used to represent UVM.  This is the entry that the thread switch code
changes out.  Other entries will be used:

    PML4 TABLE (each cpu gets its own table)

* 1 switched entry 512GB of UVM (expandable to more)
* 1 static entry 512GB of KVM
* 1 static entry to map physical memory
* 1 per-cpu entry to map the per-cpu globaldata space.
* Additional entries for recursive page tables, third party
  process address space caches (userspace<->userspace copying),
  and other features.

This should greatly simplify the PMAP code... no copies of the 'high' kernel
PDE entries, for example, and we still have the flexibility of switching out
the whole PML4 table at a later date without interfering with the basic
feature set.

17 years agobinutils214 stage 2/4 (continued). remove /usr/bin/gcc{2,3} from
Matthew Dillon [Mon, 2 Feb 2004 07:32:30 +0000 (07:32 +0000)]
binutils214 stage 2/4 (continued).  remove /usr/bin/gcc{2,3} from
BSD.usr.dist.

17 years agoMerge from vendor branch BINUTILS:
Matthew Dillon [Mon, 2 Feb 2004 07:29:28 +0000 (07:29 +0000)]
Merge from vendor branch BINUTILS:
Bring GNU binutils-2.14.  This commit is an exact copy of the contents of
binutils-2.14.tar.gz.  No files have been modified.  Files that we do not
currently used have been deleted and two files, README.DRAGONFLY and
README.DELETED have been added.

17 years agoBring GNU binutils-2.14. This commit is an exact copy of the contents of
Matthew Dillon [Mon, 2 Feb 2004 07:29:28 +0000 (07:29 +0000)]
Bring GNU binutils-2.14.  This commit is an exact copy of the contents of
binutils-2.14.tar.gz.  No files have been modified.  Files that we do not
currently used have been deleted and two files, README.DRAGONFLY and
README.DELETED have been added.

17 years agobinutils214 stage 2/4.
Matthew Dillon [Mon, 2 Feb 2004 05:43:16 +0000 (05:43 +0000)]
binutils214 stage 2/4.

 * Normalize binutils212 and binutils214.  Remove /usr/bin/gcc{2,3}/*
   and move those binaries to /usr/libexec/gcc{2,3}.  Adjust objformat
   to suit.

 * Hack up contrib/gcc3 to accomodoate library path requirements and fix
   STANDARD_BINDIR_PREFIX.  Add a missing trailing '/' in STANDARD_EXEC_PREFIX.

 * Adjust /usr/src/etc/Makefile (upgrade_etc) to remove old GCC and BINUTILS
   junk.

 * Cleanup etc/mtree/BSD.usr.dist

 * Only build gcc2 for non-cross-compiled i386 native builds.  Do not
   build libc_r for non-i386 architectures for now.  Allow on sysctl based
   setproctitle() updates for non-i386 architectures for now.

 * Move ldscripts from /usr/libdata/gcc{2,3}/ldscripts into
   /usr/libexec/binutilsXXX/ldscripts and remove /usr/libdata/gcc{2,3}.

 * GCC vendor obrien->dillon, change TARGET_OS in various places from
   freebsd -> dragonfly.

 * Consolidate config.h differences for cross builds in the Makefiles so
   we need only have one config.h for binutils214/as.

 * Cleanup and fix default target BFD for the assembler and linker when
   cross-building.

 * Build both the 32 and 64 bit support BFD's for binutils214 even if
   just doing a native build.

 * Remove a bunch of alpha junk (incomplete).

 * Note gnu/usr.bin/binutils214/libbfd/dfly_targmatch.h contains aliases
   to match dragonfly BFD targets.  Some are probably unnecessary but I
   had to move on.

 * Bring in a bunch of AMD64 related files (incomplete).

 * Update rpcgen to locate cpp with the current path instead of /usr/bin/cpp.
   Add rpcgen to the buildworld bootstrap tools.

Obtained-From: A good chunk of the amd64-specific files were adapted from
FreeBSD-5 (authored by Peter Wemm and others).

17 years agoDon't use wildcards in csh expansions to avoid early termination of
Matthew Dillon [Sun, 1 Feb 2004 19:51:45 +0000 (19:51 +0000)]
Don't use wildcards in csh expansions to avoid early termination of
the upgrade.

17 years agoAdd much needed scripts to /etc/rc.d
Robert Garrett [Sun, 1 Feb 2004 10:25:48 +0000 (10:25 +0000)]
Add much needed scripts to /etc/rc.d

resident
takes a list of pathnames or filenames from /etc/resident.conf
and makes them resident

varsym
takes a list of variable definitions from /etc/varsym.conf
and makes them system wide varsyms

these are currently off by default

thanks to Eirik Nygaard for his assistance

17 years agobinutils214 stage 1/4. Bring in the build infrastructure (left untied from
Matthew Dillon [Sun, 1 Feb 2004 08:53:05 +0000 (08:53 +0000)]
binutils214 stage 1/4.  Bring in the build infrastructure (left untied from
system).  This infrastructure can handle i386 but the amd64 code is not
yet complete.

17 years agoonly update motd if the first word of the first line is Dragonfly
Robert Garrett [Sun, 1 Feb 2004 03:32:08 +0000 (03:32 +0000)]
only update motd if the first word of the first line is Dragonfly

Submitted by:

Dylan Reinhold

17 years agoReadd ending '\' so it compiles the cross build tools
Eirik Nygaard [Sat, 31 Jan 2004 18:08:03 +0000 (18:08 +0000)]
Readd ending '\' so it compiles the cross build tools

17 years agoBuild lwkt_process_ipiq_frame only for the kernel, since
Joerg Sonnenberger [Sat, 31 Jan 2004 17:14:40 +0000 (17:14 +0000)]
Build lwkt_process_ipiq_frame only for the kernel, since
struct intrframe doesn't exist in userspace and breaks libcaps.

17 years agoAdd getopt_long from NetBSD
Joerg Sonnenberger [Sat, 31 Jan 2004 13:38:48 +0000 (13:38 +0000)]
Add getopt_long from NetBSD

17 years agoRemove genassym and gensetdefs from the cross build as well.
Eirik Nygaard [Sat, 31 Jan 2004 11:56:26 +0000 (11:56 +0000)]
Remove genassym and gensetdefs from the cross build as well.

17 years agoMake gcc3 the default for non-i386 architectures, leave gcc2 the default
Matthew Dillon [Sat, 31 Jan 2004 09:14:24 +0000 (09:14 +0000)]
Make gcc3 the default for non-i386 architectures, leave gcc2 the default
for i386 architectures.

17 years agoremove genassym and gensetdefs in the upgrade_etc target.
Matthew Dillon [Sat, 31 Jan 2004 08:59:01 +0000 (08:59 +0000)]
remove genassym and gensetdefs in the upgrade_etc target.

17 years agoRemove genassym and gensetdefs. These programs are obsolete and no longer
Matthew Dillon [Sat, 31 Jan 2004 08:56:18 +0000 (08:56 +0000)]
Remove genassym and gensetdefs.  These programs are obsolete and no longer
used.

17 years agoCut our umbilical cord from mother FreeBSD.
Jeroen Ruigrok/asmodai [Sat, 31 Jan 2004 08:52:16 +0000 (08:52 +0000)]
Cut our umbilical cord from mother FreeBSD.
Reword some sentences.
Add some new directories.

17 years agoMAINTAINER lines in Makefile's are no longer applicable, remove them.
Matthew Dillon [Sat, 31 Jan 2004 06:56:46 +0000 (06:56 +0000)]
MAINTAINER lines in Makefile's are no longer applicable, remove them.

17 years agoRemove limitations on the 'machine' specification.
Matthew Dillon [Sat, 31 Jan 2004 03:26:56 +0000 (03:26 +0000)]
Remove limitations on the 'machine' specification.

Loosely-based-on: FreeBSD-5.x

17 years ago* Sync pages with FreeBSD 4.x
David Rhodus [Sat, 31 Jan 2004 01:09:19 +0000 (01:09 +0000)]
* Sync pages with FreeBSD 4.x

Sent in by: Christian Brueffer

17 years agoUpdate miscellanious firewire manual pages from FreeBSD-4.x.
Matthew Dillon [Fri, 30 Jan 2004 22:38:47 +0000 (22:38 +0000)]
Update miscellanious firewire manual pages from FreeBSD-4.x.

Submitted-by: Christian Brueffer <chris@unixpages.org>
Taken-from: FreeBSD-4.x

17 years agoClean up the sys_nsig verses NSIG code to handle additional fault
Matthew Dillon [Fri, 30 Jan 2004 18:21:18 +0000 (18:21 +0000)]
Clean up the sys_nsig verses NSIG code to handle additional fault
cases and reformat 'trap -l' output to handle the additional signals.

Report-by: Nate Lawson <nate@root.org>, Bruce Evans <bde@zeta.org.au>
17 years ago* Add kern_systimer.c to the kernel build process.
David Rhodus [Fri, 30 Jan 2004 13:23:55 +0000 (13:23 +0000)]
* Add kern_systimer.c to the kernel build process.

17 years agoUpdate to latest version.
Jeroen Ruigrok/asmodai [Fri, 30 Jan 2004 11:31:17 +0000 (11:31 +0000)]
Update to latest version.

Submitted/requested by: Xin LI <delphij@frontfree.net>

17 years agoThis commit represents a major revamping of the clock interrupt and timebase
Matthew Dillon [Fri, 30 Jan 2004 06:20:08 +0000 (06:20 +0000)]
This commit represents a major revamping of the clock interrupt and timebase
infrastructure in DragonFly.

(fix syntax error in conditionalized section of kern_ntptime.c)

17 years agoThis commit represents a major revamping of the clock interrupt and timebase
Matthew Dillon [Fri, 30 Jan 2004 06:18:28 +0000 (06:18 +0000)]
This commit represents a major revamping of the clock interrupt and timebase
infrastructure in DragonFly.

(missing nfs files)

17 years agoThis commit represents a major revamping of the clock interrupt and timebase
Matthew Dillon [Fri, 30 Jan 2004 05:42:18 +0000 (05:42 +0000)]
This commit represents a major revamping of the clock interrupt and timebase
infrastructure in DragonFly.

* Rip out the existing 8254 timer 0 code, and also disable the use of
  Timer 2 (which means that the PC speaker will no longer go beep).  Timer 0
  used to represent a periodic interrupt and a great deal of code was in
  place to attempt to obtain a timebase off of that periodic interrupt.

  Timer 0 is now used in software retriggerable one-shot mode to produce
  variable-delay interrupts.  A new hardware interrupt clock abstraction
  called SYSTIMERS has been introduced which allows threads to register
  periodic or one-shot interrupt/IPI callbacks at approximately 1uS
  granularity.

  Timer 2 is now set in continuous periodic mode with a period of 65536
  and provides the timebase for the system, abstracted to 32 bits.

  All the old platform-integrated hardclock() and statclock() code has
  been rewritten.  The old IPI forwarding code has been #if 0'd out and
  will soon be entirely removed (the systimer abstraction takes care of
  multi-cpu registrations now).  The architecture-specific clkintr() now
  simply calls an entry point into the systimer and provides a Timer 0
  reload and Timer 2 timebase function API.

* On both UP and SMP systems, cpus register systimer interrupts for the Hz
  interrupt, the stat interrupt, and the scheduler round-robin interrupt.
  The abstraction is carefully designed to allow multiple interrupts occuring
  at the same time to be processed in a single hardware interrupt.  While
  we currently use IPI's to distribute requested interrupts from other cpu's,
  the intent is to use the abstraction to take advantage of per-cpu timers
  when available (e.g. on the LAPIC) in the future.

  systimer interrupts run OUTSIDE THE MP LOCK.  Entry points may be called
  from the hard interrupt or via an IPI message (IPI messages have always
  run outside the MP lock).

* Rip out timecounters and disable alternative timecounter code for other
  time sources.  This is temporary.  Eventually other time sources, such as
  the TSC, will be reintegrated as independant, parallel-running entities.
  There will be no 'time switching' per-say, subsystems will be able to
  select which timebase they wish to use.  It is desireable to reintegrate
  at least the TSC to improve [get]{micro,nano}[up]time() performance.

  WARNING: PPS events may not work properly.  They were not removed, but
  they have not been retested with the new code either.

* Remove spl protection around [get]{micro,nano}[up]time() calls, they are
  now internally protected.

* Use uptime instead of realtime in certain CAM timeout tests

* Remove struct clockframe.  Use struct intrframe everywhere where clockframe
  used to be used.

* Replace most splstatclock() protections with crit_*() protections, because
  such protections must now also protect against IPI messaging interrupts.

* Add fields to the per-cpu globaldata structure to access timebase related
  information using only a critical section rather then a mutex.  However,
  the 8254 Timer 2 access code still uses spin locks.  More work needs to
  be done here, the 'realtime' correction is still done in a single global
  'struct timespec basetime' structure.

* Remove the CLKINTR_PENDING icu and apic interrupt hacks.

* Augment the IPI Messaging code to make an intrframe available to callbacks.

* Document 8254 timing modes in i386/sai/timerreg.h.  Note that at the
  moment we assume an 8254 instead of an 8253 as we are using TIMER_SWSTROBE
  mode.  This may or may not have to be changed to an 8253 mode.

* Integrate the NTP correction code into the new timebase subsystem.

* Separate boottime from basettime.  Once boottime is believed to be stable
  it is no longer effected by NTP or other time corrections.

CAVETS:

* PC speaker no longer works

* Profiling interrupt rate not increased (it needs work to be
  made operational on a per-cpu basis rather then system-wide).

* The native timebase API is function-based, but currently hardwired.

* There might or might not be issues with 486 systems due to the
  timer mode I am using.

17 years agoThis should hopefully fix current issues with bootstrap buildworlds from
Matthew Dillon [Fri, 30 Jan 2004 02:35:02 +0000 (02:35 +0000)]
This should hopefully fix current issues with bootstrap buildworlds from
4.x and older versions of DFly, and with user-specified CCVER and CPUTYPE
environment variables.

Explicitly override CCVER in buildworld in a backwards compatible fashion
when building gcc2's and gcc3's startup libraries.  Also override CCVER
in gnu/lib/gcc{2,3}/Makefile when bootstrapping from older DFly or FBsd
systems which do not implement the '.makeenv' make directive.

In bsd.init.mk detect when CCVER is changed by a Makefile or Makefile.inc and
rerun bsd.cpu.mk to update the _CPUCFLAGS set, which is assigned in bsd.cpu.mk.
We relay on the fact that ${} variables are not resolved until rules accessing
them are actually run.

Clean up objformat (no operational changes).

17 years agoSet a variable indicating that we support .makeenv, so we can conditionalize
Matthew Dillon [Thu, 29 Jan 2004 22:40:36 +0000 (22:40 +0000)]
Set a variable indicating that we support .makeenv, so we can conditionalize
its use.

17 years agooops, undo part of the last commit. cpp, f77 need their gccX frontends in
Matthew Dillon [Thu, 29 Jan 2004 02:26:46 +0000 (02:26 +0000)]
oops, undo part of the last commit. cpp, f77 need their gccX frontends in
/usr/bin/gccX

17 years agoAdd back a directory path that buildworld needs, remove /usr/libexec/gcc2
Matthew Dillon [Thu, 29 Jan 2004 01:56:05 +0000 (01:56 +0000)]
Add back a directory path that buildworld needs, remove /usr/libexec/gcc2
from mtree.

17 years agohave upgrade / upgrade_etc remove /usr/libexec/gcc2 now as well, it has
Matthew Dillon [Thu, 29 Jan 2004 01:43:47 +0000 (01:43 +0000)]
have upgrade / upgrade_etc remove /usr/libexec/gcc2 now as well, it has
been moved to /usr/libexec/binutils212.

17 years agoRemove the remainder of Makefile.sub handling. Make objformat aware of the
Matthew Dillon [Thu, 29 Jan 2004 01:39:48 +0000 (01:39 +0000)]
Remove the remainder of Makefile.sub handling.  Make objformat aware of the
new binutils location for cpp and f77.  Make contrib/gcc aware of the new
binutils locations as specified in /usr/src/gnu/usr.bin/cc/Makefile.inc.
This is hardwired at the moment.

This fixes dependancies on the old /usr/libexec/gcc2 directory.

17 years agoRename .sub files back to .inc.
Matthew Dillon [Wed, 28 Jan 2004 21:26:55 +0000 (21:26 +0000)]
Rename .sub files back to .inc.

17 years agoRename Makefile.sub back to Makefile.inc to fix cross builds.
Matthew Dillon [Wed, 28 Jan 2004 19:27:01 +0000 (19:27 +0000)]
Rename Makefile.sub back to Makefile.inc to fix cross builds.

17 years agoUndo some of the previously made changes to deal with cross build issues
Matthew Dillon [Wed, 28 Jan 2004 19:24:14 +0000 (19:24 +0000)]
Undo some of the previously made changes to deal with cross build issues
from 4.x and older versions of DFly.

sys.mk should *NOT* include bsd.init.mk.  That is way, way too early and
blows up .inc files, e.g. such as those used by sys/boot.

Rename .sub back to .inc in sys/boot (more on the way) so it will cross build
from 4.x.