121625 Commits

Author SHA1 Message Date
simokawa
2bdb850e9b MFC: rev 1.59
Teardown interrupt only when sc->ih is not NULL.
2007-04-03 22:41:43 +00:00
simokawa
2b4fbe39ba MFC: rev 1.85
- Don't call fw_busreset() in firewire_attach().
2007-04-03 22:39:20 +00:00
stefanf
edd291ee53 MFC the latest synchronisation with the NetBSD code. 2007-04-03 20:47:25 +00:00
julian
21cbc159a9 Revert the MFC if 1.112 ( 1.106.2.27 )
It turns out that 1.112 has bug in -current which is faithfully
reproduced in releng-6 with the MFC.
2007-04-03 09:46:52 +00:00
julian
0288e45287 Complete reversal of MFC:
I will investigate this further to possbly re-MFC it but for now
play safe and revert this line too.
2007-04-03 08:29:31 +00:00
julian
d9980da194 Revert one of the MFCs from Friday as it produces an
unacceptable ABI change. I will re-MFC this when I have tested a version
that brings back the desirable changes but leaves the ABI the same.
2007-04-03 08:16:05 +00:00
bms
ecb163fc29 MFC:
Merge latest ifmcstat from HEAD.

Obtained from:	KAME (with many FreeBSD changes)
2007-04-03 00:39:34 +00:00
ume
144ec955e7 MFC 1.305: Turn default address selection on by default. Now,
when ipv6_enable="NO", an IPv4 address is preferred for a
destination address.
2007-04-02 09:23:29 +00:00
alc
5edafb93fe MFC
Retire pmap_track_modified().
2007-04-02 06:09:43 +00:00
alc
f73f289da3 MFC revision 1.179
Eliminate unnecessary PG_BUSY tests.
2007-04-02 05:43:38 +00:00
yongari
19c0ef9c62 MFC if_sk.c, rev. 1.129 to RELENG_6
Fix incorrect busy check for PHY write operation.
  While I'm here remove unnecessary return statement.
2007-04-02 04:30:23 +00:00
scottl
37ec55c28d MFC: Freeze the simq not the devq if we run out of command slots. This fixes
the last round of instability in the rev 13/14 driver.
2007-04-02 03:33:29 +00:00
yongari
e17b429190 MFC if_sk.c, rev 1.132 if_skreg.h, rev 1.40 to RELENG_6
Remove incomplete Yukon II support code which was added in if_sk.c, rev 1.123.
  Yukon II users should use msk(4).
2007-04-02 02:04:13 +00:00
yongari
499debd933 Enable PCI MSI support for msk(4). 2007-04-02 01:22:31 +00:00
alc
9211a664d1 MFC revisions 1.170 and 1.171
Reimplement the reclamation of PV entries.
2007-04-02 00:38:37 +00:00
bms
127ead92a0 MFC:
Add a sysctl net.link.tap.up_on_open which defaults to zero; when it
  is non-zero, tap(4) instances will be marked IFF_UP on attach.

  1.71 +6 -1 src/sys/net/if_tap.c
  1.19 +8 -1 src/share/man/man4/tap.4

PR:		110383
Requested by:	Frank Behrens
2007-04-01 22:02:30 +00:00
alc
f27a453111 MFC revision 1.155
Introduce the function pmap_enter_object().
2007-04-01 21:43:54 +00:00
alc
a87be90bc2 MFC revision 1.182
Change pmap_protect() so that execute access can be removed without
  simultaneously removing write access.
2007-04-01 19:38:18 +00:00
mjacob
cfe8535079 MFC of constification, cleanups of pre-7.X compiles, additions of
SYNC_REG memory barriers and so on.
2007-04-01 17:21:19 +00:00
mjacob
e3534c136f MFC of constification && some bad nvram flags. 2007-04-01 17:15:33 +00:00
mjacob
7e2e12449f MFC of some constification, error message spelling errors, splitting
FC and SPI default set/get into separate functions. Also, stop looking
for interrupts at the end of isp_starts- this seems to send some 23XX
cards into the weeds.
2007-04-01 17:14:25 +00:00
alc
9c0d35ee33 MFC revisions 1.175 and 1.177
Introduce the function pmap_enter_object().
2007-04-01 05:59:12 +00:00
thompsa
edf58557bc MFC r1.134
Correct the name of the 'proto' command.
2007-04-01 02:09:47 +00:00
alc
e33b5985b3 MFC revision 1.168
Remove the remaining spl*() calls.  Add some assertions.  Eliminate some
  excessive white space.

MFC revision 1.174
  Retire pmap_track_modified().
2007-04-01 01:34:31 +00:00
sam
804648e546 MFC: add IEEE80211_KEY_UNDEFINED and use it instead of local defs
Obtained from:	netbsd
2007-03-31 22:49:55 +00:00
sam
7f0211f809 MFC 1.102: move __inline before type to appease gcc 4.x
Obtained from:	netbsd
2007-03-31 22:42:45 +00:00
sam
a9e655b0b3 MFC 1.99: clear/reclaim challenge text when switching auth mode and
operating as an ap

Obtained from:	Atheros
2007-03-31 22:34:57 +00:00
sam
da9f4c9540 MFC 1.101: discard deauth+disassoc frames that are not destined for us;
these can be received when the interface is in promisc mode

Obtained from:	netbsd
2007-03-31 22:31:36 +00:00
alc
393b70416f MFC revision 1.559
Introduce the function pmap_enter_object().

MFC revision 1.562 (in part)
  Change pmap_enter_quick_locked() to fail rather than wait if it is
  unable to allocate a page table page.  This prevents a race between
  pmap_enter_object() and the page daemon.  Specifically, an inactive
  page that is a successor to the page that was given to
  pmap_enter_quick_locked() might become a cache page while
  pmap_enter_quick_locked() waits and later pmap_enter_object() maps
  the cache page violating the invariant that cache pages are never
  mapped.  Similarly, change
  pmap_enter_quick_locked() to call pmap_try_insert_pv_entry() rather
  than pmap_insert_entry().  Generally speaking,
  pmap_enter_quick_locked() is used to create speculative mappings.  So,
  it should not try hard to allocate memory if free memory is scarce.

  Add an assertion that the object containing m_start is locked in
  pmap_enter_object().  Remove a similar assertion from
  pmap_enter_quick_locked() because that function no longer accesses the
  containing object.

  Remove a stale comment.
2007-03-31 22:13:26 +00:00
sam
492b3a44e5 MFC 1.11: allow net80211 to fillin rate sets 2007-03-31 22:03:04 +00:00
sam
e6c2d53a22 MFC 1.10: allow net80211 to fillin rate sets 2007-03-31 22:02:42 +00:00
sam
718c304e1c MFC 1.49: allow net80211 to fillin rate sets 2007-03-31 22:00:44 +00:00
sam
09bb88ed66 MFC 1.51: allow net80211 to fillin rate sets 2007-03-31 21:57:27 +00:00
sam
5844ed10bf sync w/ HEAD 2007-03-31 21:55:11 +00:00
sam
0442eb3fd8 MFC: expose IEEE80211_DISCARD, IEEE80211_DISCARD_IE, and
IEEE80211_DISCARD_MAC so they can be used within
     net80211 but outside ieee80211_input.c
2007-03-31 21:53:08 +00:00
sam
33d139239f MFC 1.106: improve debug msg for ie's that are too short 2007-03-31 21:50:53 +00:00
sam
e35d16a13a MFC: keep tx/rx seq #'s for non-QoS traffic separate from QoS; stations
aren't supposed mix traffic but if they did frames might be mis-handled

Obtained from:	Atheros
2007-03-31 21:49:28 +00:00
sam
b73801d433 MFC 1.104: split check for rate set mismatch on assoc req away from
check for pure 11g mode so we can give meaningful diagnostic msgs
2007-03-31 21:45:55 +00:00
sam
2822b1cc98 MFC 1.103: disallow re-associate with a slot time mismatch
Obtained from:	Atheros
2007-03-31 21:44:41 +00:00
sam
c1059823eb MFC 1.36: change ieee80211_mhz2ieee to use the PSB mapping when the
frequency falls in the proper place, not when we're handed
	  a 1/2 or 1/4-rate channel
2007-03-31 21:42:21 +00:00
alc
e3a059531c MFC revision 1.76
Introduce the function pmap_enter_object().
2007-03-31 20:58:52 +00:00
jhb
00a9c64c1f Missed updating these with the amd64/i386 MSI MFC.
Reported by:	Derkjan de Haan
2007-03-31 20:21:14 +00:00
alc
4e067a3bb6 MFC revision 1.552
In general, bits in the page directory entry (PDE) and the page table
  entry (PTE) have the same meaning.  The exception to this rule is the
  eighth bit (0x080).  It is the PS bit in a PDE and the PAT bit in a
  PTE.  This change avoids the possibility that pmap_enter() confuses a
  PAT bit with a PS bit, avoiding a panic().

  Eliminate a diagnostic printf() from the i386 pmap_enter() that serves
  no current purpose, i.e., I've seen no bug reports in the last two
  years that are helped by this printf().
2007-03-31 19:20:23 +00:00
sam
e04dc0c1b8 remove dead code
Coverity CID:	1082
Found with:	Coverity Prevent(tm)
2007-03-31 19:11:19 +00:00
julian
e0e2c2dd5a unbreak build by puting back definition used by ipfw(8)
the real fix is to just remove the offending lines there but I don't
have time to check that now. I'll do it right on Tuesday.
2007-03-31 16:44:28 +00:00
jhb
c6b8fdd801 MFC: Simple single-message MSI support. 2007-03-31 15:25:11 +00:00
jhb
80b8da4bdb MFC: PCI MSI support for amd64 and i386. 2007-03-31 15:23:21 +00:00
jhb
e0eee19fe2 Bump __FreeBSD_version for PCI MSI support. 2007-03-31 14:51:46 +00:00
jhb
8d05413dd2 MFC: MI support for PCI Message Signalled Interrupts (MSI). For more
details see the relevant commit logs in HEAD.  The notable differences
between this patch and what is in HEAD now are that pci_remap_msix() is not
currently implemented in this patch (I need to change the API in HEAD, so
I'm not going to MFC it until its API is stable) and MSI and MSI-X are not
enabled by default.  To enable MSI and MSI-X support, one must set the
'hw.pci.msi_enable' and/or 'hw.pci.msix_enable' loader tunables.
2007-03-31 14:50:51 +00:00
julian
d1bfd0194e MFC: 1.161
include the opt_diver.h to get the banner massage right.

This brings ip_fw2.c up to date with -current with the exception of:
 1.113, 114, 125, 126, 130, 138, 140, 152-156, 158-160
All of which I do not believe to be MFC candidates (with the
possible exception of 140 which has a risk factor I don't
want to put in RELENG_6). 138 is also a possibility but requires more work
in other files.
I need to do the userland ipfw too to add some new features, but, not today..
2007-03-31 01:51:29 +00:00