Commit Graph

3440 Commits

Author SHA1 Message Date
christos dd40a9f6f6 include new compat siginfo header. 2005-09-24 17:29:47 +00:00
christos 5c03cd2255 more header cleanup. 2005-09-24 17:27:44 +00:00
christos ca3a193432 factor out some of the osockaddr, omsghdr code. 2005-09-24 15:51:03 +00:00
chs 33359b3121 need to include sys/socket.h here now, sys/mount.h no longer does it for us. 2005-09-24 15:42:58 +00:00
christos 03ec972e58 maplen was uninitialized. 2005-09-19 02:47:15 +00:00
christos d359ae5c43 MINSIGSTKSZ on linux is different depending on the platform and usually
smaller than ours.
2005-09-19 02:46:49 +00:00
christos e20c28140a Add compat signal headers. 2005-09-18 14:30:15 +00:00
yamt a0bdf5ef81 irix_sys_swapctl: follow uvm_swap_stats changes. 2005-09-17 14:52:55 +00:00
yamt 209478cccf protect p_nrlwps by sched_lock. 2005-09-17 14:35:51 +00:00
christos 29f3692018 add <compat/sys/signal*.h> 2005-09-16 17:09:52 +00:00
christos c315b31f19 Fix a comment. 2005-09-15 03:00:16 +00:00
he 70fe9e01e8 This now needs <compat/sys/signal.h> to build. 2005-09-14 13:49:50 +00:00
he 87e3d67bce This file now needs <compat/sys/stat.h> to build. 2005-09-14 13:49:27 +00:00
christos ff66f5797e need linux_syscallargs.h to compile, and remove inline function. 2005-09-13 22:08:21 +00:00
christos c83d281198 Forgot to reverse the conditional test. 2005-09-13 16:35:24 +00:00
christos 58b392c018 make this compile again. 2005-09-13 03:23:09 +00:00
christos 6acf4b5ca8 add compat netbsd32 context. 2005-09-13 03:22:54 +00:00
christos 063b880cf0 compat code reorg. 2005-09-13 01:42:32 +00:00
christos 49110e63f0 Range checks are not optional. The omission of this one, allows the user
to corrupt the heap and/or crash the kernel (Christer Oeberg).
2005-09-12 22:17:49 +00:00
christos 6a2a989266 variable name mistake again. 2005-09-12 22:16:56 +00:00
christos d9f67be7fe - Avoid overflow by checking the count argument (Christer Oeberg).
- While I am here, don't leak fds either.
2005-09-12 20:54:38 +00:00
jmmv ec93365612 Initial addition of tmpfs, an efficient memory file-system. This project
was developed as part of Google's Summer of Code 2005 program.  This
change adds the kernel code, the mount_tmpfs utility, a regression test
suite and does all other related changes to integrate these.

The file-system is still *experimental*.  Therefore, it is disabled by
default in all kernels.  However, as typically done, a commented-out
entry is added in them to ease its setup.

Note that I haven't commited the required mountd(8) changes to be able
to export tmpfs file-systems because NFS support is still very unstable
and because, before enabling it, I'd like to do some other changes.

OK'ed by my project mentor, William Studenmund (wrstuden@).
2005-09-10 19:20:48 +00:00
rpaulo 257017665e Add a SYSCTL_SETUP_PROTO() so that it compiles fine in the lkm. 2005-09-02 14:34:00 +00:00
he eb89cd94e9 Now that <ufs/ufsmount.h> includes <ufs/extattr.h>, we also need
<sys/vnode.h> and <sys/vnode_if.h> to bring the definition of struct
vop_getextattr_args into scope.  Fixes build problem for arc.
2005-08-31 16:01:56 +00:00
he 54e6da6105 Change from using sys___{,f,l}stat13() to sys___{,f,l}stat30() and
associated data structures, following the change to 64-bit inodes.
2005-08-22 10:57:04 +00:00
yamt 0ed4057082 linux_machdepioctl: remove duplicated FILE_USE/UNUSE. 2005-08-21 13:13:50 +00:00
kent 868a42c866 fix a compilation problem on NetBSD/amd64 2005-08-20 15:02:36 +00:00
yamt e4ab81774d compat_30_sys_getdents: netbsd 1.2 -> 3.0 in comments/messages. 2005-08-19 13:17:37 +00:00
yamt 34dab5e58c compat_30_sys_getdents: don't leak kernel stack garbage to userland. 2005-08-19 13:16:56 +00:00
christos 4ce7529a64 change a hard-coded 0xf to _DIRENT_ALIGN (thanks yamt) 2005-08-19 06:01:00 +00:00
christos b041fac92d Fix lossage I created with the 64 bit ino_t change. 2005-08-19 04:24:38 +00:00
christos 758a209d23 64 bit inode changes. 2005-08-19 02:03:49 +00:00
christos c0ee8a4184 Fix systeminfo.
1. return length is the string length even if the string would not fit.
2. add SI_ISALIST
3. on 32 bit emulation, don't return sparc64 as the arch!
2005-07-28 03:50:52 +00:00
cube ae35d4a4e9 Regen (__sigtimedwait(2)). 2005-07-23 22:04:12 +00:00
cube 73fe747f11 Implement __sigtimedwait(2). 2005-07-23 22:03:45 +00:00
cube a6b803a79f Constify conversion functions. 2005-07-23 21:51:29 +00:00
cube 21472d491b Regen (timer_create(2) and friends). 2005-07-23 18:57:07 +00:00
cube 23a1110cbd Implement the timer_create(2) family of syscalls. 2005-07-23 18:56:15 +00:00
cube ee24060ae4 Remove duplication of code for netbsd32_wait4() by using stackgap(9). 2005-07-22 22:46:29 +00:00
cube 606552e84c Rework sunos32_execv{,e} after removal of netbsd32_execve2.
XXX COMPAT_SUNOS32 needs a SUNOS32PTR64() macro.
2005-07-13 12:49:32 +00:00
cube 86f97cc446 Do the CHECK_ALT_EXIST dance again in netbsd32_execve(). It was lost in
previous commit.
2005-07-13 11:55:19 +00:00
cube e13e6c14c7 netbsd32_execve2() doesn't exist anymore. 2005-07-13 11:53:57 +00:00
christos 00a78ed413 Delete define for sunos_syscall_intern. 2005-07-12 22:45:11 +00:00
martin e14ba58310 When returning 64bit values (from netbsd32_lseek) fix up the return
value vector (of 32bit values, most likely) via a MD macro.
2005-07-12 15:06:17 +00:00
cube e937141b42 Regen (*xattr(2)). 2005-07-12 07:46:19 +00:00
cube 728e5e01bd Add the *xattr(2) family of syscalls. Just like the *extattr(2) family,
they're waiting for an actual FS back-end to be tested, but are still
straightforward enough.
2005-07-12 07:45:34 +00:00
cube 0477abe883 Regen (rasctl(2)). 2005-07-11 20:18:52 +00:00
cube 241e7db4d5 Add rasctl(2). Thanks to the regression tests for that syscall, I was able
to discover the bugs in netbsd32_setitimer() and netbsd32_execve().
2005-07-11 20:18:05 +00:00
cube 355bcba766 Split sys_execve() and add execve1() that does most of the work, and takes
as an argument a function that will retrieve an element of the pointer
arrays in user space.  This allows COMPAT_NETBSD32 to share the code for
the emulated version of execve(2), and fixes various issues that came from
the slow drift between the two implementations.

Note:  when splitting up a syscall function, I'll use two different ways
       of naming the resulting helper function.  If it stills does
       copyin/out operations, it will be named <syscall>1().  If it does
       not (as it was the case for get/setitimer), it will be named
       do<syscall>.
2005-07-11 20:15:26 +00:00
cube 08fd92bac5 Split sys_getitimer and sys_setitimer to make it possible to share the
relevant code with the COMPAT_NETBSD32 version, and make the latter use
the new functions.

This fixes netbsd32_setitimer() which had drifted from the native syscall
and did not work properly anymore.
2005-07-11 19:50:42 +00:00