Matthew Dillon [Fri, 15 Aug 2003 08:32:32 +0000 (08:32 +0000)]
Make modules work again part 1: hook up 'dev'. Note that not all devices
were converted to modules even in the original module build, so coverage
is somewhat spotty.
Matthew Dillon [Fri, 15 Aug 2003 07:34:25 +0000 (07:34 +0000)]
Standardize the syscall generation target to 'sysent'.
Matthew Dillon [Fri, 15 Aug 2003 07:33:35 +0000 (07:33 +0000)]
master makefile for netproto modules, of which there is exactly one
at the moment (ncp).
Matthew Dillon [Fri, 15 Aug 2003 07:32:28 +0000 (07:32 +0000)]
Make modules work again part 1: forgot a file in 'emulation'.
Matthew Dillon [Fri, 15 Aug 2003 07:31:25 +0000 (07:31 +0000)]
Make modules work again part 1: linkup crypto.
Matthew Dillon [Fri, 15 Aug 2003 07:26:16 +0000 (07:26 +0000)]
Make modules work again part 1: linkup vfs, rename Makefile.module files,
add additional Makefile's as appropriate, and fixup junk header file
generation.
Matthew Dillon [Fri, 15 Aug 2003 07:03:12 +0000 (07:03 +0000)]
Make modules work again part 1: linkup "net" and rename Makefile.module files
to Makefile. Still more to come...
Matthew Dillon [Fri, 15 Aug 2003 06:32:58 +0000 (06:32 +0000)]
Make modules work again part 1: linkup emulation/ and change the architecture
softlink scheme so the module build and the direct linked-to-kernel build
can more easily access architecture-specific files. A new config is required.
still more to come...
Matthew Dillon [Fri, 15 Aug 2003 01:19:54 +0000 (01:19 +0000)]
Make modules work again part 1: wire up the module build for bus/
Matthew Dillon [Thu, 14 Aug 2003 23:26:45 +0000 (23:26 +0000)]
Make modules work again part 1: Rename netgraph's Makefile.module to
Makefile and cleanup the subdirectory makefiles to build the netgraph
modules.
Jeffrey Hsu [Thu, 14 Aug 2003 23:11:33 +0000 (23:11 +0000)]
Add statistics to disambiguate how a spurious retransmit was detected.
Jeffrey Hsu [Thu, 14 Aug 2003 23:09:33 +0000 (23:09 +0000)]
Make the logic clear on when to use Eifel detection or fall back
onto the old RTT heuristic.
Move infrequently executed code out of the fast path to avoid
L1 instruction cache pollution.
Add statistics to disambiguate how a spurious retransmit was detected.
Matthew Dillon [Thu, 14 Aug 2003 23:05:11 +0000 (23:05 +0000)]
Add softlinks so ports like sysutils/ffsrecov will compile. Eventually
these and other softlinks will be replaced by the environment code.
Jeffrey Hsu [Thu, 14 Aug 2003 21:08:57 +0000 (21:08 +0000)]
Fix spurious spelling within comments.
Jeroen Ruigrok/asmodai [Thu, 14 Aug 2003 18:31:14 +0000 (18:31 +0000)]
Add README for the documentation directory.
Matthew Dillon [Thu, 14 Aug 2003 18:28:49 +0000 (18:28 +0000)]
Remove FBSDID, move $ rcs id back to a comment field.
Jeroen Ruigrok/asmodai [Thu, 14 Aug 2003 18:27:49 +0000 (18:27 +0000)]
Remove reference to FreeBSD's documentation about their source trees.
It is a whole different tree altogether.
Jeroen Ruigrok/asmodai [Thu, 14 Aug 2003 18:26:32 +0000 (18:26 +0000)]
First stab at our own UPDATING file.
If you make any change worth mentioning to people using the source tree,
feel free to add to the text. Do not worry about spelling, layout or the
like, it will get fixed up shortly afterwards.
Jeroen Ruigrok/asmodai [Thu, 14 Aug 2003 18:10:40 +0000 (18:10 +0000)]
Add, hopefully, support for the ICH5 sound hardware.
David Rhodus [Wed, 13 Aug 2003 22:55:22 +0000 (22:55 +0000)]
* Fix the type of the NULL arg to execl()
* de-__P()
* Explicitly compare the values returned by chdir(2) and
chroot(2) to -1
Obtained from: FreeBSD
Jeffrey Hsu [Wed, 13 Aug 2003 21:17:53 +0000 (21:17 +0000)]
New statistics to keep track of supurious retransmits.
Jeffrey Hsu [Wed, 13 Aug 2003 18:34:25 +0000 (18:34 +0000)]
Implement the Eifel Dectection Algorithm for TCP (RFC 3522).
Matthew Dillon [Tue, 12 Aug 2003 04:58:23 +0000 (04:58 +0000)]
Syscall messaging 4a: LINT build.
Matthew Dillon [Tue, 12 Aug 2003 02:36:15 +0000 (02:36 +0000)]
Syscall messaging 4: Further expand the kernel-version of the syscall message.
The (in-kernel) syscall message is now arranged:
struct blah_args {
sysmsg
usrmsg
... syscall arguments ...
}
Original system calls copyin() just the arguments and then initialize sysmsg
and go. Syscall messages copyin() usrmsg+arguments and then initialize sysmsg
as appropriate and go.
Further detail work for EASYNC support. Implement td_msgport as a reply port
and start working on an async capability for the nanosleep() system call.
NOTE: Preliminary system call messaging can be tested using the suite of
programs in /usr/src/test/sysmsg.
NOTE: Work is still in progress and you can crash the system, so use of
MSGF_ASYNC for messaging system calls is currently restricted to root.
Also fixed a bug in the syscall module helper code in sys/sysent.h, which
might have been causing the linux problems (or might not have).
All system call headers had to be regenerated to deal with the structural
changes.
Matthew Dillon [Tue, 12 Aug 2003 02:29:44 +0000 (02:29 +0000)]
Add src/test and throw in some of the ad-hoc timing and testing programs
that have been written.
David Rhodus [Mon, 11 Aug 2003 17:50:15 +0000 (17:50 +0000)]
Add or correct range checking of signal numbers in system calls and
ioctls.
David Rhodus [Mon, 11 Aug 2003 17:30:31 +0000 (17:30 +0000)]
Fix a bug which caused signals on YUV images to fail.
Fix broken programming of VSCALE_HI registers in yuv422_prog().
Obtained from: FreeBSD
David Rhodus [Mon, 11 Aug 2003 17:07:30 +0000 (17:07 +0000)]
Fix for insufficient range checking of signal numbers.
In most cases, attempted delivery of a negative or out-of-range signal
number will trigger an assertion failure and panic, thereby crashing
the system.
FreeBSD Security Advisory FreeBSD-SA-03:09.signal
David Rhodus [Mon, 11 Aug 2003 15:40:52 +0000 (15:40 +0000)]
The iBCS2 system call translator for statfs(2) did not check the
did not check the length parameter for validity.
FreeBSD Security Advisory FreeBSD-SA-03:10.ibcs2
Robert Garrett [Mon, 11 Aug 2003 12:44:27 +0000 (12:44 +0000)]
finish fixing my braindeadness..
thanks to David Reese for the netbsd, openbsd and opendarwin cvs files
Robert Garrett [Mon, 11 Aug 2003 12:42:05 +0000 (12:42 +0000)]
Change the names of these files to match the FreeBSD ones
Robert Garrett [Mon, 11 Aug 2003 12:31:55 +0000 (12:31 +0000)]
loose this file, dont need it, don't want it.. by by
Robert Garrett [Mon, 11 Aug 2003 12:17:59 +0000 (12:17 +0000)]
remove old cvs files
Robert Garrett [Mon, 11 Aug 2003 12:14:24 +0000 (12:14 +0000)]
Add netbsd, openbsd, and open darwin supfiles, rename freebsd stuff
Jeroen Ruigrok/asmodai [Mon, 11 Aug 2003 06:17:37 +0000 (06:17 +0000)]
Change __volatile and __const into volatile and const.
We expect the compiler to compile DragonFly to support C90 at minimum.
Robert Garrett [Mon, 11 Aug 2003 03:59:25 +0000 (03:59 +0000)]
Remove -traditional which breaks the keyword signed
Matthew Dillon [Mon, 11 Aug 2003 01:35:52 +0000 (01:35 +0000)]
add cvs-site dist (The layout for the dragonflybsd.org web site).
Matthew Dillon [Mon, 11 Aug 2003 00:25:17 +0000 (00:25 +0000)]
Add the 'wmake' script to /usr/bin and wmake support to Makefile.inc1.
This script allows you to build any element within /usr/src using a
buildworld environment and is extremely useful for debugging or
development work. Instead of having to restart the buildworld
(even with -DNOCLEAN -DNOTOOLS it can take a while) this script can
be used to immediately build whatever piece of the world you are
working on.
Jeroen Ruigrok/asmodai [Sun, 10 Aug 2003 20:03:39 +0000 (20:03 +0000)]
Remove kerberosIV a.k.a. eBones. It has served well, but definately
needs to be succeeded by kerberos5 a.k.a. Heimdal.
Jeroen Ruigrok/asmodai [Sun, 10 Aug 2003 19:10:42 +0000 (19:10 +0000)]
Change __signed to signed.
The former is a very GCC specific use and hardly used.
Approved by: Matt Dillon
Robert Garrett [Sun, 10 Aug 2003 10:11:13 +0000 (10:11 +0000)]
Fix path for nfs mounts, and correct sysctl
nfs now works properly with RCng
Jeroen Ruigrok/asmodai [Sat, 9 Aug 2003 22:00:54 +0000 (22:00 +0000)]
Remove math.h from Makefile, it gets installed through lib/msun/Makefile.
David Rhodus [Sat, 9 Aug 2003 20:54:39 +0000 (20:54 +0000)]
LINT breakage fix.
#include "cx.c"
Jeroen Ruigrok/asmodai [Sat, 9 Aug 2003 20:43:45 +0000 (20:43 +0000)]
Install math.h.
David Rhodus [Sat, 9 Aug 2003 20:26:21 +0000 (20:26 +0000)]
Fix compile error.
s/+/=
Robert Garrett [Sat, 9 Aug 2003 17:50:11 +0000 (17:50 +0000)]
add missing space between libraries -lcrypt-lcrypto should of been -lcrypt -lcrypto
Robert Garrett [Sat, 9 Aug 2003 17:28:46 +0000 (17:28 +0000)]
I think it's about time we added this file to the cvsup examples directory.
perhaps we should also think about removing the freebsd examples?
Jeroen Ruigrok/asmodai [Sat, 9 Aug 2003 14:16:51 +0000 (14:16 +0000)]
Correct spelling of Matt's surname.
Submitted by: Liu Kang <liukang@bjpu.edu.cn>
David Rhodus [Fri, 8 Aug 2003 22:23:55 +0000 (22:23 +0000)]
Allow the caller to get an error directly if we sent the packet immediately.
Obtained from: FreeBSD
David Rhodus [Fri, 8 Aug 2003 22:11:29 +0000 (22:11 +0000)]
Don't map LINUX_POSIX_VDISABLE to _POSIX_VDISABLE and vice versa for
the VMIN and VTIME members of the c_cc array. These members are not
special control characters. By not excluding these members we
changed the noncanonical mode input processing when both members
were 0 on entry (=LINUX_POSIX_VDISABLE) as we would remap them to 255
(=_POSIX_VDISABLE). See termios(4) case A for how that screws up
your terminal I/O.
Obtained from: FreeBSD
Matthew Dillon [Fri, 8 Aug 2003 21:47:49 +0000 (21:47 +0000)]
Add a few missing cratom() calls. In particular the code in kern_exec()
was modifying the svuid of the parent process when a child exec()'d, which
broke the 'man' program (and probably other things).
Reported-by: Jeroen Ruigrok/asmodai <asmodai@wxs.nl>
David Rhodus [Fri, 8 Aug 2003 21:42:06 +0000 (21:42 +0000)]
Fix a problem that occurs when truncating files on NFSv3 mounts: we need
to set np->n_size back to the desired size again after calling
nfs_meta_setsize(), since it could end up in nfs_loadattrcache() getting
called, which would change n_size back to the value it had before the
truncate request was issued. The result of this bug is that the size info
cached in the nfsnode becomes incorrect, lseek(fd, ofs, SEEK_END) seeks
past the end of the file, stat() returns the wrong size, etc.
Obtained from: FreeBSD
Matthew Dillon [Fri, 8 Aug 2003 19:22:45 +0000 (19:22 +0000)]
Fix the buildkernel target.
Robert Garrett [Fri, 8 Aug 2003 11:55:19 +0000 (11:55 +0000)]
add missed sysdb to /etc/rc.d's Makefile
Matthew Dillon [Fri, 8 Aug 2003 04:18:49 +0000 (04:18 +0000)]
Kernel tree reorganization stage 2: Major cvs repository work.
This stage includes source code adjustments to make buildworld work again,
including extensive additional work on the /usr/include hierarchy.
Matthew Dillon [Fri, 8 Aug 2003 04:11:06 +0000 (04:11 +0000)]
Set _THREAD_SAFE preprocessor symbol when the -pthread option is used,
solves a concurrency problem with libc's stdio.h when inline macros
(like putchar()) are used.
Matthew Dillon [Fri, 8 Aug 2003 03:57:35 +0000 (03:57 +0000)]
kernel tree reorganization stage 1a: remove -I- for kmod build
Matthew Dillon [Thu, 7 Aug 2003 21:54:36 +0000 (21:54 +0000)]
kernel tree reorganization stage 1: Major cvs repository work (not logged as
commits) plus a major reworking of the #include's to accomodate the
relocations.
* CVS repository files manually moved. Old directories left intact
and empty (temporary).
* Reorganize all filesystems into vfs/, most devices into dev/,
sub-divide devices by function.
* Begin to move device-specific architecture files to the device
subdirs rather then throwing them all into, e.g. i386/include
* Reorganize files related to system busses, placing the related code
in a new bus/ directory. Also move cam to bus/cam though this may
not have been the best idea in retrospect.
* Reorganize emulation code and place it in a new emulation/ directory.
* Remove the -I- compiler option in order to allow #include file
localization, rename all config generated X.h files to use_X.h to
clean up the conflicts.
* Remove /usr/src/include (or /usr/include) dependancies during the
kernel build, beyond what is normally needed to compile helper
programs.
* Make config create 'machine' softlinks for architecture specific
directories outside of the standard <arch>/include.
* Bump the config rev.
WARNING! after this commit /usr/include and /usr/src/sys/compile/*
should be regenerated from scratch.
Matthew Dillon [Thu, 7 Aug 2003 21:26:08 +0000 (21:26 +0000)]
kernel tree reorganization stage 1: Major cvs repository work (not logged as
commits) plus a major reworking of the #include's to accomodate the
relocations.
* CVS repository files manually moved. Old directories left intact
and empty (temporary).
* Reorganize all filesystems into vfs/, most devices into dev/,
sub-divide devices by function.
* Begin to move device-specific architecture files to the device
subdirs rather then throwing them all into, e.g. i386/include
* Reorganize files related to system busses, placing the related code
in a new bus/ directory. Also move cam to bus/cam though this may
not have been the best idea in retrospect.
* Reorganize emulation code and place it in a new emulation/ directory.
* Remove the -I- compiler option in order to allow #include file
localization, rename all config generated X.h files to use_X.h to
clean up the conflicts.
* Remove /usr/src/include (or /usr/include) dependancies during the
kernel build, beyond what is normally needed to compile helper
programs.
* Make config create 'machine' softlinks for architecture specific
directories outside of the standard <arch>/include.
* Bump the config rev.
WARNING! after this commit /usr/include and /usr/src/sys/compile/*
should be regenerated from scratch.
Matthew Dillon [Thu, 7 Aug 2003 21:22:38 +0000 (21:22 +0000)]
kernel tree reorganization stage 1: Major cvs repository work (not logged as
commits) plus a major reworking of the #include's to accomodate the
relocations.
* CVS repository files manually moved. Old directories left intact
and empty (temporary).
* Reorganize all filesystems into vfs/, most devices into dev/,
sub-divide devices by function.
* Begin to move device-specific architecture files to the device
subdirs rather then throwing them all into, e.g. i386/include
* Reorganize files related to system busses, placing the related code
in a new bus/ directory. Also move cam to bus/cam though this may
not have been the best idea in retrospect.
* Reorganize emulation code and place it in a new emulation/ directory.
* Remove the -I- compiler option in order to allow #include file
localization, rename all config generated X.h files to use_X.h to
clean up the conflicts.
* Remove /usr/src/include (or /usr/include) dependancies during the
kernel build, beyond what is normally needed to compile helper
programs.
* Make config create 'machine' softlinks for architecture specific
directories outside of the standard <arch>/include.
* Bump the config rev.
WARNING! after this commit /usr/include and /usr/src/sys/compile/*
should be regenerated from scratch.
Matthew Dillon [Thu, 7 Aug 2003 21:19:26 +0000 (21:19 +0000)]
kernel tree reorganization stage 1: Major cvs repository work (not logged as
commits) plus a major reworking of the #include's to accomodate the
relocations.
* CVS repository files manually moved. Old directories left intact
and empty (temporary).
* Reorganize all filesystems into vfs/, most devices into dev/,
sub-divide devices by function.
* Begin to move device-specific architecture files to the device
subdirs rather then throwing them all into, e.g. i386/include
* Reorganize files related to system busses, placing the related code
in a new bus/ directory. Also move cam to bus/cam though this may
not have been the best idea in retrospect.
* Reorganize emulation code and place it in a new emulation/ directory.
* Remove the -I- compiler option in order to allow #include file
localization, rename all config generated X.h files to use_X.h to
clean up the conflicts.
* Remove /usr/src/include (or /usr/include) dependancies during the
kernel build, beyond what is normally needed to compile helper
programs.
* Make config create 'machine' softlinks for architecture specific
directories outside of the standard <arch>/include.
* Bump the config rev.
WARNING! after this commit /usr/include and /usr/src/sys/compile/*
should be regenerated from scratch.
Robert Garrett [Wed, 6 Aug 2003 20:23:19 +0000 (20:23 +0000)]
Hardwire i386 instead of using the borken machine variable
Robert Garrett [Wed, 6 Aug 2003 20:22:33 +0000 (20:22 +0000)]
remove unmatched endif
Robert Garrett [Wed, 6 Aug 2003 20:21:37 +0000 (20:21 +0000)]
remove reference to krb4 in sysinstalls man page
Robert Garrett [Wed, 6 Aug 2003 20:20:29 +0000 (20:20 +0000)]
add -crypto to the list of required libraries
Robert Garrett [Wed, 6 Aug 2003 20:19:10 +0000 (20:19 +0000)]
Undo a local change that made it into the makefile
Robert Garrett [Wed, 6 Aug 2003 19:58:46 +0000 (19:58 +0000)]
Change this back
Hiten Pandya [Tue, 5 Aug 2003 18:07:41 +0000 (18:07 +0000)]
Backout my ``proc cleanup'' brain damage.
-- Thy shall not listen to LINT about stupid things.
Robert Garrett [Tue, 5 Aug 2003 17:11:04 +0000 (17:11 +0000)]
Fix conditional that had two &&'s so that make release
will work.
--Needed for RCng testing.
Hiten Pandya [Tue, 5 Aug 2003 14:08:52 +0000 (14:08 +0000)]
Um, ok. I should have slept a little more.
Fix the affects of my braindamage in svr4_fcntl.c
Spotted by: asmodai, robg
Jeroen Ruigrok/asmodai [Tue, 5 Aug 2003 07:45:44 +0000 (07:45 +0000)]
Remove kerberosIV from the build.
Both normal builds and kerberos5 builds have been tested to work.
Jeroen Ruigrok/asmodai [Tue, 5 Aug 2003 06:15:21 +0000 (06:15 +0000)]
Clean up two lint warnings, by adding /* NOTREACHED */ in the appropriate
places.
Sort of syncs us with rev 1.27 of FreeBSD's cat.c.
Matthew Dillon [Sun, 3 Aug 2003 17:27:17 +0000 (17:27 +0000)]
Fix a compile time error. Rename MAX to RAINMAX and get rid of all the
silly #ifdef checks. I think we can safely blame FreeBSD braindamage
for this one :-).
Matthew Dillon [Sun, 3 Aug 2003 17:11:36 +0000 (17:11 +0000)]
Explicitly use an unsigned index for 'which' in shmsys(), msgsys(), and
semsys(), so the implications of the use of a negative index become obvious.
Fix the misdocumented 'which' UAP argument, which is actually signed.
Note: There is no security issue because sizeof() is unsigned but such
non-explicit comparisons are a very dangerous way to code and this fixes
that.
Matthew Dillon [Sun, 3 Aug 2003 16:45:17 +0000 (16:45 +0000)]
Remove NOSECURE which no longer serves a purpose. Note: FreeBSD also removed
NOSECURE.
Submitted-by: Jeroen Ruigrok/asmodai <asmodai@tendra.org>
Hiten Pandya [Sun, 3 Aug 2003 12:29:05 +0000 (12:29 +0000)]
Move the backtrace() function from kern_subr.c to kern_debug.c.
All debugging related kernel functions, and syscalls should
be added into this file.
Discussed with: Matt (about kern_debug.c)
Hiten Pandya [Sun, 3 Aug 2003 11:51:25 +0000 (11:51 +0000)]
DELAY() does not belong in a MD include file, instead, move
it to sys/systm.h where it fits better with the other clock
related functions.
Hiten Pandya [Sun, 3 Aug 2003 11:47:58 +0000 (11:47 +0000)]
Add backtrace() prototype.
Hiten Pandya [Sun, 3 Aug 2003 11:42:28 +0000 (11:42 +0000)]
LINT cleanup: remove redundant ``struct proc *p'' declarations.
Hiten Pandya [Sun, 3 Aug 2003 11:41:12 +0000 (11:41 +0000)]
Define HAVE_PPSRATECHECK macro, now that we have merged the
ppsratecheck() function from FreeBSD.
Prodded by: LINT
Hiten Pandya [Sun, 3 Aug 2003 11:39:57 +0000 (11:39 +0000)]
LINT cleanup: fix MAX difinition.
Hiten Pandya [Sun, 3 Aug 2003 11:38:45 +0000 (11:38 +0000)]
LINT cleanup: add <sys/systm.h> for printf()
Hiten Pandya [Sun, 3 Aug 2003 10:07:41 +0000 (10:07 +0000)]
Use FOREACH_PROC_IN_SYSTEM() throughout.
Matthew Dillon [Fri, 1 Aug 2003 17:46:27 +0000 (17:46 +0000)]
Cleanup remaining tsleep priority issues.
Submitted-by: Jeroen Ruigrok/asmodai <asmodai@wxs.nl>
Matthew Dillon [Fri, 1 Aug 2003 17:14:23 +0000 (17:14 +0000)]
Fix a pointer bug introduced by syscall messaging. semget() should work
properly now.
Submitted-by: qhwt@myrealbox.com
Robert Garrett [Fri, 1 Aug 2003 16:30:41 +0000 (16:30 +0000)]
fix ;; in /etc/rc.subr
caught by:qhwt
Robert Garrett [Fri, 1 Aug 2003 10:58:59 +0000 (10:58 +0000)]
Enable SSE by default, add nob to turn it off
Submitted by: Craig Dooley
Aproved by: Matt Dillon
Robert Garrett [Fri, 1 Aug 2003 09:57:34 +0000 (09:57 +0000)]
Update mtree so man pages quit complaining
Robert Garrett [Fri, 1 Aug 2003 04:37:51 +0000 (04:37 +0000)]
attach rc.subr to the build
Robert Garrett [Fri, 1 Aug 2003 04:36:57 +0000 (04:36 +0000)]
Add rc.subr to our list of most informative manpages
Robert Garrett [Fri, 1 Aug 2003 04:31:03 +0000 (04:31 +0000)]
Update rc manpage to reflect RCng
Robert Garrett [Fri, 1 Aug 2003 04:25:15 +0000 (04:25 +0000)]
Add nsswitch.conf.5 to man page build
Robert Garrett [Fri, 1 Aug 2003 04:23:44 +0000 (04:23 +0000)]
update manpage to reflect changes in RCng
Robert Garrett [Fri, 1 Aug 2003 04:11:23 +0000 (04:11 +0000)]
Bring in the man page for nsswitch.conf this is part of RCng
Robert Garrett [Fri, 1 Aug 2003 03:48:33 +0000 (03:48 +0000)]
Let us not forget network.subr either
Robert Garrett [Fri, 1 Aug 2003 03:47:06 +0000 (03:47 +0000)]
Add rc.subr to /etc/Makefile this fixes a bug where
rc.subr is failing to be installed
Matthew Dillon [Thu, 31 Jul 2003 19:56:59 +0000 (19:56 +0000)]
Clear the BSS earlier in the boot sequence and clean up the comments. This
will allow us to not have to worry about .data vs .bss globals in the future
(especially since gcc 3.x 'optimizes' things like (global) int x = 0 -> bss).
Also note that many boot loaders do not clear the bss.
Taken-from: FreeBSD sys/i386/i386/locore.s 1.172 and 1.174 (bde)
Submitted-by: Craig Dooley <cd5697@albany.edu>
Hiten Pandya [Thu, 31 Jul 2003 12:37:39 +0000 (12:37 +0000)]
Get LINT to build.
This commit is a followup to the removal of <machine/asnames.h>:
Attic/asnames.h, rev 1.12:
Remove pre-ELF underscore prefix and asnames macro hacks.
Reported by: http://ps.catpa.com/tinderbox/
Hiten Pandya [Thu, 31 Jul 2003 12:01:38 +0000 (12:01 +0000)]
Generalise, and remove SI_SUB_VINUM; use SI_SUB_RAID instead.
Hiten Pandya [Thu, 31 Jul 2003 11:03:34 +0000 (11:03 +0000)]
LINT breakage fix, part one:
Changes the remaining calls to disable_intr() to cpu_disable_intr()
Reviewed by: dillon