Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
153,991 Commits 606 Branches 0 Tags 3.1 GiB
80c18fce28
Commit Graph

153991 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
yamt
a6f37bf1a0 wm(4): Add IPv6 Rx tcp/udp checksum offloading. [yamt 20061116] 
tcp(4), ifconfig(8), wm(4): Add TCPv6 hardware transmit segment
		offload support. [yamt 20061124]
 2006-11-23 20:15:29 +00:00
elad
8f6d3d35da Always #undef CG14_MAP_REGS, since it opens a security hole (as the comment 
says) and should be used only for debugging.

Okay mouse@, christos@.
 2006-11-23 19:49:41 +00:00
yamt
809ec70bcf implement ipv6 TSO. 
partly from Matthias Scheler.  tested by him.
 2006-11-23 19:41:58 +00:00
martin
22b629efe2 Fix an obvious copy&pasto and make it compile with GEM_DEBUG. 2006-11-23 18:41:25 +00:00
rpaulo
785b8e0981 WARNING: input date is in the past! 2006-11-23 18:02:15 +00:00
pooka
c9aa28f463 * document open, close and access 
* nomenclature improvements
 2006-11-23 17:45:11 +00:00
elad
3dc874e6b2 Add note about Segvguard interface/implementation being experimental and 
with the potential to change in future releases.
 2006-11-23 17:24:36 +00:00
pooka
4b5bf45935 const coherence 2006-11-23 16:44:28 +00:00
tnozaki
068d17c461 c99 istyle initializer, pointed by uebayasi-san. thanks a lot! 2006-11-23 13:59:03 +00:00
martin
fba603e221 Note crunched dhclient change with -fno-strict-aliasing hack. 2006-11-23 13:24:28 +00:00
elad
6d74a5fdc7 Document Veriexec strict levels (hm, maybe we should have veriexec(8)?) 
as requested by Adam Hamsik.
 2006-11-23 13:23:22 +00:00
martin
ca4e1f8a92 If we need -fno-strict-aliasing (see hack "gcc4 pointer sign and strict 
aliasing problems", usr.sbin/dhcp/Makefile.inc rev 1.20), we need it
on the crunched binaries too.
 2006-11-23 13:22:18 +00:00
elad
f1e710a4d8 Document flaw in veriexec_rawchk(). This is what I added to the comment: 
* XXX: This is bogus. There's an obvious race condition between the time
 * XXX: the disk is open for writing, in which an attacker can access a
 * XXX: monitored file to get its signature cached again, and when the raw
 * XXX: file is overwritten on disk.
 * XXX:
 * XXX: To solve this, we need something like the following:
 * XXX:         open raw disk:
 * XXX:           - raise refcount,
 * XXX:           - invalidate fingerprints,
 * XXX:           - mark all entries with "no cache" flag
 * XXX:
 * XXX:         veriexec_verify:
 * XXX:           - if "no cache", don't cache evaluation result
 * XXX:
 * XXX:         close raw disk:
 * XXX:           - lower refcount,
 * XXX:           - if refcount == 0, remove "no cache" flag from all entries
 2006-11-23 13:11:29 +00:00
martin
be008341cd Hopping through a void* is not enough to get rid of alignment requirements. 
Do a memcpy to a properly aligned local variable instead.
 2006-11-23 13:07:17 +00:00
martin
6a3c1d7aeb When checking wether we are debugging, look for -g in ${CFLAGS} instead 
of ${COPTS}. The latter explicitly includes the former and ${DBG}, where
the -g most likely is hidden.
 2006-11-23 12:05:26 +00:00
tron
9506122aab Backout accidental commit which broke kernel builds. 2006-11-23 09:43:56 +00:00
he
e004bf7e3b This file needs to know the size of "struct device", so include 
<sys/device.h> here.  This should allow the sparc GENERIC_SUN4U and
sparc64 kernels to build sata_subr.c again.
 2006-11-23 08:44:42 +00:00
mrg
d010dedb55 bump version to NetBSD nb2 20061121 2006-11-23 05:43:40 +00:00
tnozaki
51d316fde4 add GB12345 csmapper files. 2006-11-23 05:20:04 +00:00
rpaulo
74c9505238 Add pseudo-device etherip. 2006-11-23 04:22:02 +00:00
rpaulo
74b7608af9 Remove extra prototype. 2006-11-23 04:20:54 +00:00
rpaulo
370be902f0 WARNING: date is in the future! 2006-11-23 04:12:51 +00:00
rpaulo
5423539f94 New EtherIP driver based on tap(4) and gif(4) by Hans Rosenfeld. 
Notable changes:
	* Fixes PR 34268.
	* Separates the code from gif(4) (which is more cleaner).
	* Allows the usage of STP (Spanning Tree Protocol).
	* Removed EtherIP implementation from gif(4)/tap(4).

Some input from Christos.
 2006-11-23 04:07:07 +00:00
tnozaki
dec2eb4c9d 1. move GB*.esdb from i18n/esdb/MISC to i18n/esdb/GB directory. 
2. add iconv support for Chinese Tradisional GB12345.
 2006-11-23 03:25:24 +00:00
hubertf
1ee989e6be Don't try to handle symlinks on versions of the SSH File Transfer Protocol 
that don't support it, e.g. the Solaris 9/x86 sshd (SSH-2.0-Sun_SSH_1.0;
..._1.1 from Solaris 10/x86 is fine)

OK'd by pooka@
 2006-11-23 01:51:15 +00:00
tnozaki
c8ab6b2b31 add missing rcsid. 2006-11-22 23:47:21 +00:00
tnozaki
45e156694b oops, forgot to remove unused enum. 2006-11-22 23:44:17 +00:00
tnozaki
8033a5b008 1. add iconv support for following encodings: 
Chinese Simplefied
        HZ, HZ8 - 7/8bit stateful encoding, see RFC1842,1843. (libHZ)
        zW      - 7bit stateful encoding, see RFC1842. (libZW)
2. add citrus_prop.[ch] - parser for encoding module's init parameter strings.
 2006-11-22 23:38:25 +00:00
dogcow
ec21dc99eb #include <sys/device.h> to fix build failure on sparc64. 
(why only sparc64? no idea.)
 2006-11-22 22:14:23 +00:00
dsl
503ea65cd7 If we only want .OBJDIR printed, there is no point parsing the default 
make rules, nor reading Makefile.
It is rather a shame we can't stop .depend being parsed as well.
Quite possibly one could argue that nothing in .depend should affect the
output of 'make -V xxxx'.
 2006-11-22 22:04:57 +00:00
christos
aac134252a From defend dot the dot world at gmail dot com: Sanitize environment variables 
for set{u,g}id binaries, so that in case they are playing with set{u,g}id
and exec'ing other binaries they don't get affected by the
LD_{PRELOAD,DEBUG,LIBRARY_PATH} environment setup. We leave LD_BIND_NOW alone.
There are no binaries affected in the base system.
 2006-11-22 21:46:36 +00:00
drochner
30cf9aac05 -fix interpretation of the poll interval in interrupt descriptors for 
USB2/highspeed: This is an exponent now, not milliseconds.
 (There are wrong example descriptors for hubs in the USB2 spec; tolerate
  invalid values and assume 256ms which is good for hubs, just in case
  someone blindly copied from the book.)
-being here, fix an obvious memory leak
 2006-11-22 21:10:36 +00:00
elad
2664dc174c Add missing items to the securelevel implications list. 
Now it is complete.
 2006-11-22 21:00:30 +00:00
elad
539403d985 Prevent iopl/ioperm in securelevel 1, not 2. 2006-11-22 20:57:52 +00:00
martin
2d95ef4c0f Remove CRUNCHGEN_FLAGS, probably left over from a previous hack to work 
around bugs in an older compiler. Nowadays the default -Os works fine.
 2006-11-22 20:54:16 +00:00
tnozaki
9b1aff8258 add fallback mapping. 2006-11-22 20:22:30 +00:00
tnozaki
5bda830543 fix memory leak. 2006-11-22 20:11:03 +00:00
riz
2adcd99325 Add the nbinclude directory to the include path, so that 
<hppa/elf_machdep.h> (and possibly others) can be found.
 2006-11-22 18:37:38 +00:00
drochner
421a2aa25d regen 2006-11-22 17:58:39 +00:00
drochner
a46bdb5784 add vendor "O2 Micro" 2006-11-22 17:57:40 +00:00
bouyer
6c04cd11cb Cosmetic: remove one ":" from the port status string, it looks better this 
way.
 2006-11-22 17:51:02 +00:00
bouyer
dcf3bce253 In sata_reset_interface() return only DET bits of SStatus, as the 
return value is compared against SStatus_DET_*
 2006-11-22 17:49:12 +00:00
rpaulo
7420eb145a Change pickup to fifo. Spotted by Havard Eidnes. 2006-11-22 17:46:46 +00:00
christos
4ca73ce810 Don't expose _ss functions for 4.0. Requested by core. 2006-11-22 17:23:24 +00:00
reed
57c4a8a6b0 Check that -w width is not above maximum. (It already checks for zero or 
negative.) Using width above DWIDTH may cause overflow as noted by Gruzicki
Wlodek on bugtraq.

While here replace one use of 132 with DWIDTH.
 2006-11-22 16:15:42 +00:00
elad
72438de6ef Remove redundant securelevel check; this is already done in procfs_rw() 
and we can't get here (procfs_control()) without being there first.

Pointed out by yamt@.
 2006-11-22 15:48:11 +00:00
riz
2f0a27492f Make sys/exec_elf.h and machine/elf_machdep.h available to host 
tools, and allow paxctl to use them when compiled as a host
tool.  This fixes the build on MacOS X (and perhaps other platforms)
 2006-11-22 15:08:47 +00:00
tsutsui
90ee4817fb Umm, fix typo. 2006-11-22 14:32:28 +00:00
elad
054d9c15fe Remove KAUTH_PROCESS_DEBUGINIT, should have never gotten into the tree in 
the first place.
 2006-11-22 13:59:27 +00:00
christos
91ad430f6c adjust_limits takes p, not l again 2006-11-22 13:56:09 +00:00