Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
167,155 Commits 606 Branches 0 Tags 3.1 GiB
e4b37a874e
Commit Graph

4205 Commits

Author SHA1 Message Date
dsl
36fb6d9db7 Revert previous. 2007-12-17 21:48:23 +00:00
dsl
a098759aae creat64 and open64 just called the creat/open function. 
Change to do it here.
 2007-12-17 21:39:17 +00:00
dsl
686e359d32 Regen without any 'const' members of the syscall argument structures. 2007-12-17 20:47:31 +00:00
dsl
bf434e38c7 Remove all the 'const'. Although the fields might be 32bit pointers to 
constant data, the 'const' applies to the field itself - which is stupid.
 2007-12-17 20:45:29 +00:00
njoly
c7d62a5250 Fix linux32_sys_wait4 by doing its own jobw instead off calling 
linux_sys_wait4. It worked except when child resources were requested,
because 32/64 bits struct rusage differs in size ...

While here, update linux32_sys_waitpid to use linux32_sys_wait4 with a
NULL rusage.
 2007-12-17 18:10:37 +00:00
njoly
e78a322c4a s/netbsd32_void */netbsd32_caddr_t/ 2007-12-15 14:08:16 +00:00
dsl
63f7948f02 Call sys_open() not bsd_sys_open(). 2007-12-13 07:54:22 +00:00
uebayasi
d9547ee3a0 Regen. 2007-12-13 05:19:08 +00:00
dsl
fc10f67856 Remove the wrapper functions for file operations that do nothing now 
that namei() checks the alternate path.
 2007-12-12 22:07:58 +00:00
dsl
ef3588e13f Regen, reformmatted, and without wrappers for file operations. 2007-12-12 22:06:19 +00:00
dsl
0f63820911 No need to wrap filename function to get the alt path stuff. 2007-12-12 22:04:30 +00:00
dsl
f9ade37fe1 compat_file.[ch] are no longer used 2007-12-12 21:37:26 +00:00
dsl
9ae4eb70ff Regen 2007-12-12 21:29:43 +00:00
dsl
e26d16aac3 No need for any of the bsd_sys_foo() calls since they only wrapped 
CHECK_ALT_XXX() for filenames - which got killed earlier in the year.
Substitute the same functions as in compat_file.c
 2007-12-12 21:24:36 +00:00
dsl
0f47845a4f Add a definition of compat_12_stat_conv() here so I can kill compat_file.h 2007-12-12 21:17:57 +00:00
lukem
4e1d22cc0b fix the __KERNEL_RCSID 2007-12-11 23:32:44 +00:00
lukem
2b8be6f723 use __KERNEL_RCSID() 2007-12-11 13:22:50 +00:00
cube
d97f33f6eb Regen syscall tables for mach binary compatibility. Per request from dsl@. 2007-12-11 08:06:46 +00:00
dsl
be33a1a204 Increase the maximum number of syscall args. 
sys_mach_sys_msg_overwrite_trap() has been overwriting stack!
 2007-12-10 22:32:15 +00:00
dogcow
9ad45b39e6 more NDINIT fallout. not completely compile-tested. 2007-12-09 13:34:23 +00:00
wiz
9c131e8d5c Remove last argument from NDINIT call, following 
"Remove cn_lwp from struct componentname.  curlwp should be used
from on.  The NDINIT() macro no longer takes the lwp parameter and
associates the credentials of the calling thread with the namei
structure." from pooka.
 2007-12-09 12:50:29 +00:00
pooka
db06a930e6 Remove cn_lwp from struct componentname. curlwp should be used 
from on.  The NDINIT() macro no longer takes the lwp parameter and
associates the credentials of the calling thread with the namei
structure.
 2007-12-08 19:29:36 +00:00
dsl
28bae79b27 ANSIfy most of the function definitions in sys/compat (but not ndis). 
All by the magic of sed ...
 2007-12-08 18:35:53 +00:00
ad
9c8beda3c8 Reg... hang on, why don't we build this at compile time?!? 2007-12-07 22:42:11 +00:00
ad
bca34dfe7a Add: _lwp_setname, _lwp_getname, _lwp_ctl. 2007-12-07 22:41:05 +00:00
ad
937533671d lockmgr -> rwlock 2007-12-06 14:53:35 +00:00
dyoung
adde197943 Use IFADDR_EMPTY(). 2007-12-05 22:51:01 +00:00
yamt
65e70dc43b include mutex.h. 2007-12-05 12:01:24 +00:00
ad
73b6db8481 Make it compile. 2007-12-05 08:45:30 +00:00
ad
a0ca06de3d lockmgr -> mutex 2007-12-05 08:39:46 +00:00
ad
78090667c0 Kill ref to lockmgr in comments (for grep). 2007-12-05 08:34:41 +00:00
ad
a5bdf04ea6 lockmgr -> rwlock 2007-12-05 08:33:23 +00:00
dyoung
5a740a2a51 Extract common code into subroutine svr4_count_ifnum(). 
Don't open-code queue(3) macros (x = ifnet.tqh_first; y =
x.if_list.tqe_next).  Instead, use the macros themselves.

Use IFNET_FOREACH() and IFADDR_FOREACH().
 2007-12-05 01:10:47 +00:00
dyoung
cabaa89d1f Don't open-code queue(3) macros (x = ifnet.tqh_first; y = 
x.if_list.tqe_next).  Instead, use the macros themselves.

Use IFNET_FOREACH() and IFADDR_FOREACH().
 2007-12-05 01:06:23 +00:00
dyoung
b86d03582a Use IFADDR_FOREACH(). 2007-12-05 01:03:30 +00:00
dsl
cfc18fc34b Remove all the __P from here, including the other other mismatched one 
that my sed script left behind and 'he' didn't notice.
 2007-12-05 00:31:01 +00:00
he
5b4fada2e1 Re-balance a parenthesis. 2007-12-04 22:43:50 +00:00
dsl
f2af9174b9 Remove all the __P 2007-12-04 18:40:07 +00:00
ad
86b20896d3 Use system-provided atomic ops. 2007-11-29 14:29:29 +00:00
pooka
e49789dd46 more l removal 2007-11-28 18:50:10 +00:00
ad
16b547ec7f Please can we nuke this code? 2007-11-28 18:45:03 +00:00
he
756f91194a Follow up the removal of the lwp argument to VOP_ functions. 
In some cases, use curlwp instead of the now missing argument.
 2007-11-28 10:34:37 +00:00
dogcow
3c143b9440 more lwp cacking fallout 2007-11-27 23:30:56 +00:00
ad
23589af2fb lwp arg to VOPs is gone. 2007-11-27 22:12:24 +00:00
dogcow
6c09f19fb4 and one more lwp removal victim 2007-11-27 11:25:29 +00:00
dogcow
3465a006e9 more VFS_STATVFS(x,y,z) fallout; change them to VFS_STATVFS(x,y). (hi, pooka!) 2007-11-27 09:47:15 +00:00
he
0bfa260fce Remove a now-unused variable. 2007-11-26 23:16:15 +00:00
pooka
61e8303e9d Remove the "struct lwp *" argument from all VFS and VOP interfaces. 
The general trend is to remove it from all kernel interfaces and
this is a start.  In case the calling lwp is desired, curlwp should
be used.

quick consensus on tech-kern
 2007-11-26 19:01:26 +00:00
elad
abb7851f4d Refactor time modification checks and place them in the secmodel code. 
okay christos@
 2007-11-25 00:35:26 +00:00
christos
3714528495 - move the save context before the signal reset (Arto Huusko) 
- set the sigcontext cr2
 2007-11-24 23:52:56 +00:00
elad
1d139a3075 Modify hpux_sys_stime_6x() to call settime(), like everyone else do. 
okay christos@.
 2007-11-24 20:43:34 +00:00
njoly
204ebc94bf Regen. 2007-11-16 12:58:12 +00:00
njoly
601a0e5173 Update, for linux_sys_readlink() removal. 2007-11-16 12:56:50 +00:00
njoly
cb572617af Regen, for syscalls cleanup. 2007-11-16 12:53:56 +00:00
njoly
6412bda281 Cleanup. Remove a few linux syscalls definitions, now identical to 
native ones (with stackgap and ALT_CHECK_xxx removal). No functional
changes expected.
 2007-11-16 12:51:54 +00:00
dsl
300078900a regen 2007-11-12 22:23:29 +00:00
dsl
9bb5de8a76 syscall() needs to be 'NOARGS INDIR'. 
Actually I wonder if this code could use the standard syscall table ?
 2007-11-12 22:23:01 +00:00
christos
2e3ee66dc7 regen 2007-11-11 18:28:18 +00:00
dsl
f36a9ec28b Fix builds for ports that don't define __HAVE_MINIMAL_EMUL (eg sparc64). 
Fallout from a change that made #define names for syscalls (etc) start
NETBSD32 (not netbsd32).
 2007-11-10 18:55:42 +00:00
dsl
ddd89ca02b regen 2007-11-10 13:24:27 +00:00
dsl
b6b07a31fd Mark 'syscall' as 'NOARGS INDIR" 2007-11-10 13:23:18 +00:00
dsl
2cabf5c1f0 Remove the pecoff system table defines, they aren't needed since the 
emulation uses a user-space library and normal netbsd calls.
 2007-11-10 09:41:01 +00:00
njoly
d16de24e5a Make linux32_sys_oldolduname use LINUX_UNAME_ARCH instead of machine, 
like other uname syscalls. While here, remove unneeded
LINUX_UNAME_ARCH #ifdef/#endif checks.
 2007-11-09 17:43:17 +00:00
dsl
ee077208d9 Regen 2007-11-09 15:10:27 +00:00
dsl
3107f9cd42 Add arguments to sys_syscall and sys___syscall. 
Use MI wrappers for them in netbsd32.
Change prefix for constants from netbsd32 to NETBSD32 (used in #defines etc).
 2007-11-09 15:05:33 +00:00
njoly
23e9f9fa6f Replace strncpy with strlcpy in compat linux and linux32 uname, 
old_uname and oldold_uname syscalls to ensure that all *utsname fields
are always NUL terminated.
 2007-11-08 21:07:23 +00:00
njoly
577cb10855 Regen for 64-bit linux syscalls removal. 2007-11-07 00:25:39 +00:00
njoly
b32d95bfbf Remove remaining 64-bit compat linux syscalls linux_sys_xxx() (except 
for ones without arguments), and replace them by their 32-bit
equivalent linux32_sys_xxxx().
 2007-11-07 00:24:29 +00:00
ad
d831186d55 Merge scheduler changes from the vmlocking branch. All discussed on 
tech-kern:

- Invert priority space so that zero is the lowest priority. Rearrange
  number and type of priority levels into bands. Add new bands like
  'kernel real time'.
- Ignore the priority level passed to tsleep. Compute priority for
  sleep dynamically.
- For SCHED_4BSD, make priority adjustment per-LWP, not per-process.
 2007-11-06 00:42:39 +00:00
rmind
6364fdeba5 - Replace lockmgr with mutex; 
- Use condvars instead of tsleep/wakeup;
- Replace malloc with kmem;
- Use itimespecfix/tstohz;

Reviewed by <ad>.
 2007-11-04 11:10:34 +00:00
oster
77002eb972 This should be a mutex_enter() instead of a mutex_exit(). 
Fixes:
Mutex error: mutex_vector_exit: assertion failed: MUTEX_OWNER(mtx->mtx_owner) == curthread
...
mutex_abort(c0aba23c,c07bf8b0,c09c9164,bfbfee34,10) at netbsd:mutex_abort+0x36
mutex_vector_exit(c0aba23c,10,c0aba23c,ca8efc6c,ca8f8540) at netbsd:mutex_vector_exit+0xe2
compat_20_sys_getfsstat(ca8f8540,cb243c48,cb243c68,805eb24,805e000) at netbsd:compat_20_sys_getfsstat+0x1bf
...
 2007-11-01 03:49:52 +00:00
dsl
36acfb90ed regen 2007-10-31 22:39:17 +00:00
dsl
16c3c9c541 This file must use the netbsd32_xxx() system calls, not the sys_xxx() ones. 
The latter actually function (due to luck) for calls with one argument,
but will fail badly if more than one is required.
Noticed as an error in the ktrace outut by Nicolas Joly, reported on
tech-kern.
 2007-10-31 22:35:34 +00:00
njoly
beab9df119 Regen for syscalls cleanup. 2007-10-31 21:06:19 +00:00
njoly
6f36d09e87 compat linux32 syscalls cleanup. 
With stackgap and CHECK_ALT_xxx removal, some linux32 and netbsd32
syscalls are now identical.
To avoid code duplication, remove the linux32 definition and use the
netbsd32 one (no functional change).
 2007-10-31 21:04:02 +00:00
njoly
99e027abbb Remove extra return 2007-10-31 09:16:55 +00:00
njoly
d57ae6a62d Regen for chown additions. 2007-10-27 09:17:50 +00:00
njoly
97151f15f1 Add missing chown's syscalls family members. 
This makes chown(1)/chgrp(1) works under compat linux32.

Problem reported and fix tested by Arto Huusko. Thanks.
 2007-10-27 09:16:24 +00:00
pooka
ea84bd1c76 vfs_getopsbyname("ngs") -> vfs_getopsbyname("nfs"). 'nuff said 2007-10-23 15:21:31 +00:00
njoly
a478f23b9e Add compat_linux and exec_linux_elf lkm support for amd64: 
- Add needed COMPAT_OSSAUDIO to GENERIC.
- Add missing includes needed by linux_syscallargs.h.
- Add lkm building.
 2007-10-19 18:52:09 +00:00
ad
a2a3828545 machine/{bus,cpu,intr}.h -> sys/{bus,cpu,intr}.h 2007-10-19 11:59:34 +00:00
he
167a62205a Follow up the removal of the lwp_t* argument to do_filereadv() and 
do_filewritev().
 2007-10-11 13:51:12 +00:00
ad
53b26d467e mountlist_slock is now a mutex. 2007-10-10 22:00:53 +00:00
ad
99b0354fb8 v_flag -> v_iflag 2007-10-10 21:59:11 +00:00
ad
7dad9f7391 Merge from vmlocking: 
- Split vnode::v_flag into three fields, depending on field locking.
- simple_lock -> kmutex in a few places.
- Fix some simple locking problems.
 2007-10-10 20:42:20 +00:00
ad
62fdbc4b9a LOCK_ASSERT -> KASSERT 2007-10-08 18:07:24 +00:00
ad
451aacda90 Merge file descriptor locking, cwdi locking and cross-call changes 
from the vmlocking branch.
 2007-10-08 15:12:05 +00:00
ad
74dc12ed8b fxsave() is a function, don't re-use the name. 2007-10-03 10:54:16 +00:00
he
525420f3c8 Remove a now-unused local variable. 2007-09-30 10:35:16 +00:00
dsl
2e20a70dbf Change the way p->p_limit (and hence p->p_rlimit) is locked. 
Should fix PR/36939 and make the rlimit code MP safe.
Posted for comment to tech-kern (non received!)

The p_limit field (for a process) is only be changed once (on the first
  write), and a reference to the old structure is kept (for code paths
  that have cached the pointer).
Only p->p_limit is now locked by p->p_mutex, and since the referenced memory
  will not go away, is only needed if the pointer is to be changed.
The contents of 'struct plimit' are all locked by pl_mutex, except that the
  code doesn't bother to acquire it for reads (which are basically atomic).
Add FORK_SHARELIMIT that causes fork1() to share the limits between parent
  and child, use it for the IRIX_PR_SULIMIT.
Fix borked test for both IRIX_PR_SUMASK and IRIX_PR_SDIR being set.
 2007-09-29 12:22:30 +00:00
dsl
2b7c33c183 Use netbsd32_uint64 instead of fsblkcnt_t and fsfilcnt_t in order to get 
the correct alignment (4 bytes for i386) for the whole structure.
 2007-09-29 10:47:45 +00:00
dsl
4bacff8183 Rename members of 'struct plimit' so that the fields are 'pl_xxx' and 
no longer have the same names as members of 'struct proc'.
 2007-09-21 19:19:20 +00:00
mlelstv
4bb7dae272 Implement OSS_SNDCTL_DSP_GETODELAY and provide a no-op function 
for OSS_SNDCTL_DSP_PROFILE.
 2007-09-18 19:49:53 +00:00
dsl
06b0a1bdab Define netbsd32_uint64 for 64bit integers with the alignment requirement 
of the corresponding 32bit architecture.
Use it for the 64bit items in netbsd32_statvfs so that the structure
doesn't collect 8byte alignment (and 4 bytes of trailing padding).
This replaces the 'packed' attribute which wasn't architecture specific
and would cause massive overheads accessing every member of sparc64.
Should allow the MIPS64 port do DTRT.
 2007-09-16 22:35:01 +00:00
gdt
9768560fd7 remove SIOCSIFALIFETIME_IN6 (should have been part of previous commit) 2007-09-11 19:59:18 +00:00
cube
7a185660d8 Some more licence clean up. 2007-09-10 10:54:20 +00:00
cube
c725329722 Remove 3rd clause and my name from all the licences which were only in my 
name.
 2007-09-10 10:35:51 +00:00
dyoung
f59a11881a Constify: LLADDR() -> CLLADDR(). 2007-08-29 22:33:42 +00:00
dyoung
f004b71d98 Constify. 2007-08-26 22:36:35 +00:00
dyoung
e6fc91327f Use satosdl(). 2007-08-26 22:33:25 +00:00
martin
4418b58c3f Remove all ioctls from the conversion list that are explicitly defined 
to use struct oifreq - the magic size changing makes them fail.
 2007-08-20 19:19:33 +00:00
martin
2adca4d368 Sigh - revert previous. Seems we can't avoid the big list, so fill it. 
Now both "ifconfig ... create" and "pppoectl -d" work again with old
userland.
 2007-08-20 17:48:17 +00:00
martin
1bcc2150d3 Put back the previous hack to make compat_cvtcmd() work for all the ioctls 
that use struct ifreq which have not been explicitly versioned.
If someone feels like fixing it with a list aproach, I think below is
a complete list - the one used in the previous version missed a lot of them.

BIOCGETIF
BIOCSETIF
GREDSOCK
GREGADDRD
GREGADDRS
GREGPROTO
GRESADDRD
GRESADDRS
GRESPROTO
GRESSOCK
SIOCADDMULTI
SIOCDELMULTI
SIOCDIFADDR
SIOCDIFADDR_IN6
SIOCDIFPHYADDR
SIOCGDEFIFACE_IN6
SIOCGIFADDR
SIOCGIFADDR_IN6
SIOCGIFAFLAG_IN6
SIOCGIFALIFETIME_IN6
SIOCGIFBRDADDR
SIOCGIFDLT
SIOCGIFDSTADDR
SIOCGIFDSTADDR_IN6
SIOCGIFFLAGS
SIOCGIFGENERIC
SIOCGIFMETRIC
SIOCGIFMTU
SIOCGIFNETMASK
SIOCGIFNETMASK_IN6
SIOCGIFPDSTADDR
SIOCGIFPDSTADDR_IN6
SIOCGIFPSRCADDR
SIOCGIFPSRCADDR_IN6
SIOCGIFSTAT_ICMP6
SIOCGIFSTAT_IN6
SIOCGPVCSIF
SIOCGVH
SIOCIFCREATE
SIOCIFDESTROY
SIOCSDEFIFACE_IN6
SIOCSIFADDR
SIOCSIFADDR_IN6
SIOCSIFALIFETIME_IN6
SIOCSIFBRDADDR
SIOCSIFDSTADDR
SIOCSIFDSTADDR_IN6
SIOCSIFFLAGS
SIOCSIFGENERIC
SIOCSIFMEDIA
SIOCSIFMETRIC
SIOCSIFMTU
SIOCSIFNETMASK
SIOCSIFNETMASK_IN6
SIOCSNDFLUSH_IN6
SIOCSPFXFLUSH_IN6
SIOCSPVCSIF
SIOCSRTRFLUSH_IN6
SIOCSVH
TAPGIFNAME
 2007-08-20 16:52:59 +00:00
skd
d4509b0376 Clean up net compat ioctls, and clean up handling of wireless ioctls. 2007-08-20 04:49:40 +00:00
ad
d3675885a8 Regen. 2007-08-15 12:09:12 +00:00
ad
63c4506184 Changes to make ktrace LKM friendly and reduce ifdef KTRACE. Proposed 
on tech-kern.
 2007-08-15 12:07:23 +00:00
ad
06f7ccf01d Regen. 2007-08-07 19:01:23 +00:00
ad
830ab6bb3c - Fix a bug with _lwp_park() where if the computed wakeup time was under 
1 microsecond into the future, the thread could enter an untimed sleep.
- Change the signature of _lwp_park() to accept an lwpid_t and second
  hint pointer, but do so in a way that remains compatible with older
  pthread libraries. This can be used to wake another thread before the
  calling thread goes asleep, saving at least one syscall + involuntary
  context switch. This turns out to be a fairly large win on the condvar
  benchmarks that I have tried.
- Mark some more syscalls MP safe.
 2007-08-07 19:00:42 +00:00
xtraeme
bd0460e614 <sys/proc.h> is needed to get nprocs... fixes the build on amd64. 2007-07-21 23:39:46 +00:00
he
2eb0397ad6 Fix three identical typos in the previous (fs_fstypename -> f_fstypename). 2007-07-20 13:48:24 +00:00
dsl
f3ade04dc0 include sys/cdefs.h 2007-07-19 22:17:23 +00:00
briggs
883152ac0b Fix typo: mnfromname -> mntfromname. 2007-07-18 13:53:34 +00:00
christos
f25f73fe50 include <compat/sys/mount.h> for MFSNAMELEN 2007-07-17 20:54:45 +00:00
christos
9558d8b2f1 include <compat/sys/mount.h> for MFSNAMELEN. 2007-07-17 20:53:49 +00:00
christos
289334c878 include <compat/sys/mount.h> for MFSNAMELEN 2007-07-17 20:51:15 +00:00
christos
35cb28e6eb kill MFSNAMELEN 2007-07-17 20:41:23 +00:00
christos
18754a7640 kill silly constants. 2007-07-17 20:39:42 +00:00
christos
ecb7144e3a include <compat/sys/mount.h> MFSNAMELEN 2007-07-17 20:36:11 +00:00
christos
66c05241ca include compat/sys/mount.h for MFSNAMELEN, although not used yet. 2007-07-17 20:34:40 +00:00
christos
1b3668b8ef use the compat MFSNAMELEN 2007-07-17 20:33:17 +00:00
christos
515c243672 MFSNAMELEN only belongs in compat. 2007-07-17 20:31:03 +00:00
joerg
e4fc1154a4 Add native mremap system call based on the UVM implementation for 
Linux compat. Add code to enforce alignment of the new location.
Special thanks to wizd for helping with the man page.
 2007-07-17 17:42:07 +00:00
ad
3fc1741f68 <dsl> I suspect my set of test build kernels is longer than ad's :-) 
tsk!
 2007-07-14 17:11:28 +00:00
dsl
0bfa8b4ee7 Adapt for sys_mount() becoming compat_40_sys_mount(), 2007-07-14 15:47:26 +00:00
dsl
df57a27c09 Change calls from sys_mount() to compat_40_sys_mount() 2007-07-14 15:45:48 +00:00
dsl
6b182ed9a7 Put the stackgap code under OPT_SYSTRACE. 2007-07-13 21:04:29 +00:00
dsl
12da34644a Update generated files. 2007-07-13 20:49:07 +00:00
dsl
f02523fb6e Remove all the syscall stubs that were only there to get files lookup 
up in the alternate path.
 2007-07-13 20:48:39 +00:00
dsl
cfc71184cd Make thsi compile again after the recent changes to select/poll. 
It really should be doing things properly though.
 2007-07-13 20:46:04 +00:00
dsl
758f9f5cde Change compat mount code to pass do_sys_mount() kernel resident buffers. 
Possibly the standard nfs code needs teaching how to set the length and
address family in order to support non-netbsd sockaddr.
There are now no active stackgap() calls in the compat tree.
 2007-07-12 19:41:57 +00:00
ad
88ab7da936 Merge some of the less invasive changes from the vmlocking branch: 
- kthread, callout, devsw API changes
- select()/poll() improvements
- miscellaneous MT safety improvements
 2007-07-09 20:51:58 +00:00
pooka
b6caa3be1b fix pasto: sigframe.sf_sc is of type linux_sigcontext 2007-07-08 09:41:00 +00:00
dsl
1cb4e9208d Rip out the support for /dev/ptmx, the netbsd ptm driver will now DTRT. 
It probably ought to have minor 2 (ie do the chown/chmod/revoke on open)
because the SVR4 programs dont know anything about revoke.
 2007-07-05 19:19:24 +00:00
dsl
513b927545 It seems that min(16,16) isn't a compile time constant. 
So don't bother limiting the local arrays to NGROUPS, just use 16.
 2007-07-04 21:59:16 +00:00
dsl
8d3eb2476f Comment out the code that uses the stackgap to issue the WSDISPLAYIO_PUTCMAP 
ioctl in order to change the display colours.
Changing the code to not need the stackgap is rather pervasive, and it isn't
at all clear this is useful effort given the suspected bitrottedness
of compat darwin.
 2007-07-01 20:14:17 +00:00
dsl
813e1c7584 Fix long-term breakage in the definition of 'control' data for linux. 
cmsg->cmsg_len is 'size_t' not 'socklen_t' - so it is 8 bytes on 64bit
platforms instead of 4.  There is also never padding after the header.
Redo linux sendmsg() so that it stands a chance of getting it right.
Redo linux recvmsg() so that it process control data directly from the mbuf
  list.  Allowing it to hack the data without using the stackgap.
 2007-07-01 18:45:36 +00:00
dsl
bf215415ab Fix LINUX_HDIO_GET_IDENTITY so that it returns the entire 512 byte buffer. 
Rely on the code in dev/ata/wd.c doing a relay for the short read needed
for LINUX_HDIO_OBSOLETE_IDENTITY.
 2007-07-01 09:54:16 +00:00
dsl
168c1071b4 Call fp->f_ops->fo_ioctl() directly (not by sys_ioctl() for LINUX_VT_GETMODE 
and LINUX_VT_SETMODE.  One fewer use of the stackgap.
 2007-06-30 22:54:33 +00:00
dsl
a1188737bf Use the new CDIOCREADSUBCHANNEL_BUF and CDIOREADTOCENTRIES_BUF ioctls to 
avoid copying into the stackgap.
 2007-06-30 22:18:17 +00:00
dsl
53e0243f27 Changes to sompat socket function to avoid the dreaded stackgap. 2007-06-30 15:31:49 +00:00
dsl
4be2514b94 Match updates to kauth_cred_set/getgroups. 2007-06-30 13:34:19 +00:00
dsl
a9e905f23c Refactor this to use the new do_sys_xxx() functions from uipc_syscalls.c 
so that it isn't necessary to copy data to/from the stackgap.
Given the nature of the code in this file, it is now probably slightly
  more broken than previously. but nothing serious should be worse!
 2007-06-24 18:09:52 +00:00
dsl
a150e6e4f5 Use do_sys_recvmsg(), so_sys_accept(), do_sys_getsockname() and do_sys_senmsg() 
in order to avoid the stackgap (etc).
Note that since changing the darwin socket address is simply a matter of
  translating the address family and adding sa_len, it can easily be done
  on the mbuf resident address before/after copying to/from userspace.
Simplify the convertion of AF_LOCAL addresses by usingthe user-supplied
  buffer length instead of dowing an unbounded strlen().
Untested - did this work before?
 2007-06-24 18:07:34 +00:00
dsl
0bb6928543 Update to call do_sys_recv() instead of recvit(). 2007-06-24 18:01:48 +00:00
dsl
6d0a5fcf1a Simplify code to get/set 16bit group lists. 
Cloned from ibcs2 copy (maybe I should have worked out somewhere to put this
as common code...)
 2007-06-23 15:27:23 +00:00
dsl
a5b4569076 Read and write the ldt without using the stackgap. 2007-06-23 15:26:16 +00:00
dsl
e773713d08 Comment out support for FREEBSD_PT_READ_U and FREEBSD_PT_WRITE_U. 
It all looks like a rather broken hack.
If nothing else it accesses userspace without copyin/out.
 2007-06-23 15:24:26 +00:00
dsl
23ec3e8012 Use the functions in kern_auth.c for setting/getting group lists. 
All we have to do here is copy the 16bit to 32bit group numbers across.
 2007-06-23 09:11:12 +00:00
dsl
049a22a9db These now need sys/prot.h 2007-06-23 09:09:56 +00:00
dsl
4147586696 There is no need for the stackgap in sys_bind() and sys_connect(). 
If sa_len is zero, believing the size passed to bind/connect seems
better than trying to strlen somthing that might run off the mapped kma.
Verify the address family against the array size before indexing.
 2007-06-17 21:30:11 +00:00