Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
256,687 Commits 606 Branches 0 Tags
Commit Graph

256687 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
ozaki-r
2c1d530904 Dedup common codes in error paths (NFCI) 2018-02-14 09:13:03 +00:00
ozaki-r
b4596d0446 Fix mbuf leaks on error paths 
Pointed out by maxv@
 2018-02-14 08:59:23 +00:00
maxv
d33dbf4b86 Use .Cm instead of .Li, same as arp.8. 2018-02-14 06:52:41 +00:00
maxv
f574a1384a Re-make ip6_nexthdr global, it will be used in soon-to-be-added code... 2018-02-14 05:29:39 +00:00
maxv
239cec4999 Revert my two last changes in this file. They are apparently causing 
problems with racoon, I'll investigate this later.
 2018-02-14 05:24:44 +00:00
ozaki-r
4e4255942a Use cpu_softintr_p 2018-02-14 03:56:26 +00:00
christos
298caec6c7 need -lrt for shm_{open,close} 2018-02-14 03:37:39 +00:00
christos
4c0f6a5b74 depend on librt for shm_{open,close} 2018-02-14 02:12:35 +00:00
christos
74bd282433 Asan needs bits and pieces of ubsan now; go figure. 2018-02-14 02:12:07 +00:00
christos
d637e5b6a0 our __csa_atexit does not like NULL dso; use atexit(3) instead. 2018-02-14 02:11:22 +00:00
christos
be74da7a3a enable addr2line 2018-02-14 02:10:32 +00:00
christos
8ac5b43509 - enable netbsd 
- add cast to (uptr) for _Unwind_GetIP
 2018-02-14 02:10:06 +00:00
christos
82081465bf Provide SHAR384 flavors: The OpenSSL SHA512_Final implementation is re-used 
to implement SHA384_Final, but does the right thing by checking the internally
kept digest size. The NetBSD libc implementation provides two entry points
that write a different size digest, so if we use the SHA512_Final from here
we end up overwriting memory.
 2018-02-13 22:35:47 +00:00
christos
0f39d79ef1 Provide the means to use the libc sha2 internally instead of the openssl 
implementation; this does not really matter since their structs are larger
than ours, but it helps when we want to verify that we are not using any
of the openssl code.
 2018-02-13 22:32:10 +00:00
christos
ba6745c33a Add a (void *) cast to make things compile with our own sha2 header 2018-02-13 22:29:14 +00:00
sjg
5984bab77d Avoid unused variable warning 2018-02-13 19:37:30 +00:00
christos
ab7c594717 add missing test files added in the vendor branch. 2018-02-13 17:39:38 +00:00
maxv
995aacb845 Make the arpresolve branch more readable, fix typo, fix XXX (which I 
added), add missing pserialize_read_exit (which I forgot).
 2018-02-13 15:21:59 +00:00
maxv
78024c919d Mmh. Add a missing check: if ARP was disabled on the interface, don't 
process ARP packets. Otherwise the kernel will add ARP entries even if

	ifconfig wm0 -arp

was entered.
 2018-02-13 14:50:28 +00:00
uwe
f2874cb24e Delete 'define dbcool {}' line commented out in previous. It's bogus 
and the gcc6 correctly complains about that when it compiles ioconf.c

PR toolchain/53023
 2018-02-13 13:19:30 +00:00
hannken
6e4615fb0f Treat an inode with "mode == 0" and "blocks != 0" as partially allocated 
and clear it as ffs_newvnode() tests for "blocks == 0".
 2018-02-13 11:20:08 +00:00
maxv
b46d58304b Remove KERNEL_LOCK around the MPLS code. It's not needed, since we're only 
touching the tag of the mbuf - the tag belongs only to the mbuf, and the
mbuf is not shared.

ok knakahara@
 2018-02-13 10:50:38 +00:00
maxv
9f3f1d81a9 Be tougher: 
* In arpintr(), don't allow IEEE1394 packets on non-IEEE1394 interfaces.

 * In revarpinput(), kick IEEE1394 packets right away. They are not
   supported.
 2018-02-13 10:47:41 +00:00
maxv
66dc46f13b Same change as rev1.258, but this time in revarpinput: use m_pullup. 2018-02-13 10:31:01 +00:00
maxv
cc65a52d20 Minor stylistic changes, and use C99 types. 2018-02-13 10:20:50 +00:00
martin
1ff701ef77 Adapt to ar_tha() changes in if_arp.h rev1.31. 2018-02-13 10:17:04 +00:00
maxv
d8195ae763 Replace dead code by KASSERT. 2018-02-13 10:05:05 +00:00
mrg
a54e8b26f9 obsolete some GCC 4.8 files. 2018-02-13 10:02:12 +00:00
martin
24e4f9aca2 Fix copy & pasto (dsa code vs. rsa code) in previous, fixes PR lib/53012 
and recentish sshfs test failures.
 2018-02-13 09:51:33 +00:00
mrg
fa27b654ce use -fno-delete-null-pointer-checks for kernels. 2018-02-13 09:50:35 +00:00
maxv
afc364f7fa Put time_second and time_uptime in different cache lines, probably saves 
us some false sharing.
 2018-02-13 09:26:17 +00:00
maxv
e8a9ae2459 Don't force ARPHRD_IEEE1394 on IEEE1394 interfaces. If it's not there, then 
kick the packet. And do this earlier.
 2018-02-13 08:51:37 +00:00
maxv
e296dd2e51 Define ar_* as inlined functions, not as macros. Makes it easier to 
understand why ARPHRD_IEEE1394 needs to be handled with care - it doesn't
have ar_tha.
 2018-02-13 08:43:26 +00:00
maxv
a79a3b36b5 Use only one label, clearer. 2018-02-13 08:20:12 +00:00
maxv
86df9cc3e2 Fix three things in arpintr(): 
* mtod can't return NULL.

 * It is wrong to kick the packet if m->m_len < arplen. While this check
   always returns false for native Ethernet interfaces, it may not if the
   frame is encapsulated in EtherIP/L2TP. Use m_pullup instead.

 * Remove XXX, it is fine. Reduce the indentation level afterwards.
 2018-02-13 07:51:24 +00:00
maxv
0bb6dee8f3 Style, no functional change. 2018-02-13 07:44:25 +00:00
pgoyette
7a4c4c3326 Disable the 'define dbcool {}' line - it's not needed, and causes 
problems with config(1) and gcc6.  See PR toolchain/530223 for more
info.
 2018-02-13 07:21:19 +00:00
maxv
4770de59d1 Remove double declaration; 'ddb_regs' is already declared as a macro 
in db_machdep.h if MULTIPROCESSOR is on, and the macro has higher
priority.

Don't declare 'ddb_regs' locally in this case, because it is misleading.
Part of PR/52964.
 2018-02-13 06:44:13 +00:00
mrg
b549c2f1b5 note that openssl and GCC may upset your builds. 2018-02-13 04:36:00 +00:00
ozaki-r
2a3f19c32a Define ddb_regp only if MULTIPROCESSOR (NFC) 2018-02-13 04:11:28 +00:00
ozaki-r
98aad07956 Fix NULL pointer dereference via ddb_regs 
ddb_regs can be *ddb_regp (see db_machdep.h) so ddb_regp should be NULL-ed
after dereference to ddb_regs.

Also dbreg should be restored to ddb_regp because ddb_regp can be changed
by db_mach_cpu during db_trap.

Fix PR 52964
Helped by nonaka@
 2018-02-13 04:10:41 +00:00
christos
6769d26ead Unconditionally print the trap like we do for amdt64 (Dimitris Karagkasidis) 2018-02-13 01:05:18 +00:00
sevan
9b1213ca1f Spelling 2018-02-13 00:34:11 +00:00
joerg
e9096488b2 Add a new option CISS_NO_INTERRUPT_HACK for driving ciss(4) by callouts. 
This is intended as workaround for certain Xen issues with dom0 kernels
and will likely want to have a high HZ value as well for decent
performance.
 2018-02-12 23:11:00 +00:00
jakllsch
3f96c239a8 Fix __siglongjmp14(). 
Fixes SIGINT causing ksh to "longjmp botch", presumably due to incorrect
magic number.
cvs: ----------------------------------------------------------------------
 2018-02-12 22:31:04 +00:00
joerg
61efc2e9d5 Simplify and make the GCC check more precise at the same time. 2018-02-12 22:18:36 +00:00
sjg
9d28f87b89 Do not treat .info as warning for -W 
Reported by: lwhsu at FreeBSD.org
 2018-02-12 21:38:09 +00:00
adam
0f74ee8d8c Fix building with MKGCC=no 2018-02-12 20:47:09 +00:00
ginsbach
78c930635f Add CSDL 2018-02-12 20:43:17 +00:00
christos
c98f924910 Keep a pointer to the interface of the multicast membership, because the 
multicast element itself might go away in in_delmulti (but the interface
can't because we hold the lock). From ozaki-r@
 2018-02-12 18:19:12 +00:00