Aren
/
NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
3,294 Commits 606 Branches 0 Tags 3.1 GiB
Commit Graph

233 Commits

Author SHA1 Message Date
cgd 299ff91b14 BSDI official patch #14: 
SUMMARY:
    Here is a patch for a kernel hang that can be provoked with a write
    or send of a negative amount.  The talk program is capable of exercising
    this bug.  This patch also includes a fix for a bug that caused data
    to be delivered to TCP in smaller chunks than desired, and which caused
    TCP to send a short packet when starting up.  Finally, there is a bug
    fix for MSG_PEEK with an oobmark pending.
 1993-10-26 22:36:25 +00:00
cgd 889b512ea4 if you try to allocate a buffer larger than MAXBSIZE, panic. 1993-10-26 19:38:55 +00:00
mycroft 118cdf4851 Clean up deleted files. 1993-10-26 13:15:17 +00:00
cgd 13ca4d98c7 it's correct to return EACCES if FS is mounted MNT_NOEXEC. 1993-10-26 01:42:56 +00:00
cgd c7582cbd4f use copyinstr rather than strcpy to copy in the saved shell-script 
name from userland.  From Markus Wild, mw@lamp
 1993-10-26 00:32:29 +00:00
cgd ce7108c729 slightly clean up ws's original patch to this file for the sense 
of wait vs. nowait.  this patch from torek.
 1993-10-22 02:48:35 +00:00
cgd f784b878c2 ACCOUNTING define doesn't exist any more... 1993-10-20 23:04:33 +00:00
cgd 9ceb615f82 add some timeout_t typedefs 1993-10-20 00:53:34 +00:00
cgd fc8ee7437c knf'ify, clean up some, move sysacct() args out of function declaration. 1993-10-20 00:45:34 +00:00
cgd 287eb9a50e delete my '/dev/acct'-style accounting in favor of the version written by 
Mark Tinguely (tinguely@plains.NoDak.edu).  It's the 'standard' way
of doing accounting, rather than a wacky one.
 1993-10-20 00:16:40 +00:00
cgd 2ff9ed5060 fix my last change; for some reason i thought that 'p' was defined 
in these functions.  use curproc instead.
 1993-10-19 23:41:44 +00:00
cgd ddd610e422 from Mark Tinguely, with mods by davidg and then fially fixed by cgd. 
do the space-used integration, over time, for processes.
 1993-10-19 23:37:50 +00:00
cgd dfaa02a09b pay for block i/o. slightly different than how done by Mark Tinguely. 1993-10-19 02:47:56 +00:00
cgd dd818c7cd0 get rid of __main() -- it's going into libkern 1993-10-15 06:16:22 +00:00
deraadt c52686ddb3 from magnum branch: nbqd() was broken 1993-10-13 03:15:57 +00:00
cgd a84055b552 use the ALIGN macro to align the stack, rather than rounding up 
by a (char *).  the former is safe, even on strange machines
(read: sparc).
 1993-10-06 07:27:58 +00:00
cgd 714fa4716b changed the Debugger() call, which not all kernels have, to panic(), 
but only when DIAGNOSTIC is defined.
 1993-10-06 07:24:21 +00:00
cgd 772436479d new, improved, and rationally-implemented vfs_bio. no more serious 
structural changes should happen, as it now does the right thing
w.r.t. buffer resizing and having lots of buffers vs. relatively
little buffer space.  Ports can now "do the standard thing", re:
nbuf and bufpages, which is make nbuf = bufpages by default.
 1993-10-04 04:56:28 +00:00
mycroft 9a1fcf008b Update after fixing syscalls.master. 1993-10-02 09:55:09 +00:00
mycroft c05111e593 Put back the $'s in the RCS id that the last commit removed. 1993-10-02 09:51:40 +00:00
cgd e27f6e3ea0 patch from Christoph Badura <bad@flatlin.ka.sub.org> to fix credential 
use by read-ahead blocks.  This fixes those weak NFS authentication
messages, and allows us to use BSDI NFS servers again...
 1993-10-01 21:47:11 +00:00
cgd 1d0f22a29a convert to use the buffers which are (now) statically allocated at 
startup in machdep.c...  buffers are now *never* allocated after boot.
currently, the limitation that says bufpages must cover nbuf*MAXBSIZE
still exists, but this is one step closer to removing that limitation.
 1993-09-29 04:49:15 +00:00
cgd 4a21fe3648 ifdef out the ttyfree()'s, so that fill_eproc doesn't panic when 
a process's session still holds a ref to a tty which has been deallocated
and reused.
 1993-09-29 02:36:21 +00:00
cgd 94ac210002 rewrite biodone to the spec in the daemon book, and to account for 
the fact that buffers with B_CALL set shouldn't be brelse()'d.
 1993-09-21 00:40:02 +00:00
brezak 1905040ad2 Add sysarch() syscall. 1993-09-19 18:55:38 +00:00
cgd 9c32bc786a make allproc be volatile, and cast things accordingly. 
suggested by torek, because CSRG had problems with reordering
of assignments to allproc leading to strange panics from kernels
compiled with gcc2...
 1993-09-15 22:30:32 +00:00
mycroft 40222e846f Fix from Noriyuki Soda <soda@sra.co.jp>: 
recvmsg(2) always returns -1 with errno==EMSGSIZE, when trying
        to pass file descriptors through UNIX domain socket.
 1993-09-14 03:35:08 +00:00
glass 0c9ec05f20 check return codes on copyout()s, panic if they fail. 1993-09-12 07:56:22 +00:00
jtc 070f4a7f50 Removed functions moved to libkern: strcpy, strcmp, strlen, ffs, etc. 1993-09-11 00:08:03 +00:00
phil de69711383 These changes are due to a mismatch with user functions prototyped as 
func(short) and the fact the the kernel uses full ints.  This caused
problems on the pc532 port.  These fixes take the good 16 bits passed
by the user program and converts them into the correct form for the
kernel.
 1993-09-09 22:03:59 +00:00
mycroft 64540d3533 Patch from David Greenman to reduce CPU usage during network transmit. 1993-09-08 21:12:49 +00:00
ws 053f138dae Changes to VFS readdir semantics 
NFS changes for better cookie support
ISOFS changes for better Rockridge support and support for generation numbers
 1993-09-07 15:40:14 +00:00
mycroft 526dbecdb2 A little more cleanup; 0 --> FALSE and 1 --> TRUE. 1993-09-06 02:06:34 +00:00
mycroft cc054a83e7 De-ANSIfy and re-KNFify. 1993-09-06 02:03:03 +00:00
mycroft 29aea147e8 Add \n to end of error message. 1993-09-05 22:07:31 +00:00
sef b477be52eb Yet more of the ptrace() reorg; now ptrace_setregs() and ptrace_getregs() 
are present, along with PT_GETREGS and PT_SETREGS ptrace commands.
 1993-09-05 03:53:52 +00:00
cgd 15359c9f59 enable loadable execution handlers. need to do something better re: 
how much of the file to pass in the exec package, but i think the
solution to that will be to pass e.g. a disk block's worth, or whatever;
if exec handlers really need more, they've got the damned vnode.
 1993-09-05 01:36:28 +00:00
cgd 83aeee1bc0 there now exists an "exec switch". break the a.out code into another 
module, and tidy up the interface so that other object formats are
more easily supportable.
 1993-09-05 01:33:35 +00:00
cgd a05bd46763 patch from Szabolcs Sziget <pink@bagira.fsz.bme.hu> to load lkm device 
drivers properly.
 1993-09-05 00:55:57 +00:00
sef fddf2c7605 ptrace_single_step() and ptrace_set_pc() should return errors if 
necessary.  (Mainly because the SPARC can't easily single step, so
it should return EINVAL, and then ptrace() should return that to the
user.)
 1993-09-04 08:46:36 +00:00
cgd 3aa99f85dc better ptrace() support from Sean Eric Fagan <sef@kithrup.com> 1993-09-04 05:32:18 +00:00
cgd 6a474f29e2 get rid of maxdmap, and seperate MAXDSIZ and MAXSSIZ in rlimit checking. 1993-09-04 00:37:54 +00:00
jtc 9c00bc072d Include systm.h to get prototypes (and possibly inlines) of *max functions. 
Change mbinit() to match prototype.
 1993-09-04 00:01:43 +00:00
cgd 491d6b4b9d fix problem where exec would lose on shell scripts with symlinks in 
their names.  fix from "James W. Dolter" <jdolter@sawtooth.eecs.umich.edu>.
test script:

#!/bin/csh
mkdir /tmp/foodir
cat > /tmp/foodir/testscript <<EOF
#!/bin/sh
echo "Executing script \$1"
EOF
chmod 755 /tmp/foodir/testscript
(cd /tmp ; ln -s foodir foosym)
echo "Trying full path (w/o symlink) to /tmp/foodir/testscript"
/tmp/foodir/testscript "with out symlink"
echo "Trying full path (w/ symlink) to /tmp/foosym/testscript"
/tmp/foosym/testscript "with symlink"
echo "End of Test"
 1993-09-01 21:43:50 +00:00
deraadt 28c37a0949 fixed a little /lib/cpp boo-boo 1993-08-31 13:46:31 +00:00
cgd d365662684 ok, initialize that errno, to make theo and gcc2 happy... 1993-08-31 01:52:34 +00:00
cgd 518f7c78d6 a program being exec'd *must* cause *some* exec commands to be invoked; 
if it doesn't there's a problem in the kernel, because a program
with no exec commands run will have no address space except the stack,
and i don't think it's valid to have a "run from stack only" exec type,
so panic.  if need for that case comes up later, it can be changed...
 1993-08-31 01:46:36 +00:00
deraadt 6fce70ed63 blast it, i forgot two ('s 1993-08-30 06:25:10 +00:00
deraadt 9aed375ba3 tty XXstart() routines return void 1993-08-29 13:46:31 +00:00
deraadt bab3a9ce58 added two __P() macros 1993-08-29 13:16:04 +00:00