ƒ­{225674 181 292 373 196 315 376 146 416 129 298 438 203 1976 267 198 198 106 99 379 227 144 151 343 132 143 223 517 286 263 124 207 1044 1514 401 907 627 784 274 200 154 183 235 125 922 149 134 281 2258 226 111 198 816 148 235 207 254 99 266 217 217 781 187 241 185 232 367 229 307 146 121 208 142 129 528 171 109 135 166 207 178 296 115 152 136 250 133 161 185 104 127 220 225 108 133 318 117 114 247 184 153 160 183 257 141 306 439 146 195 199 267 191 197 164 176 182 144 175 286 418 126 436 536 110 130 230 794 180 408 167 266 181 760 319 333 536 178 729 145 468 280 103 141 198 163 121 189 820 608 226 173 117 214 123 110 360 431 99 460 132 222 121 121 125 95 538 214 631 499 323 622 142 296 224 149 990 161 235 465 686 149 200 167 K 10 svn:author V 3 jhb K 8 svn:date V 27 2011-09-19T16:28:36.868934Z K 7 svn:log V 199 MFC 225341: Clear the mountprog variable after each mountfs() call so that mountprog options don't leak over into subsequent mounts listed in /etc/fstab. While here, fix a memory leak in debug mode. END K 10 svn:author V 7 attilio K 8 svn:date V 27 2011-09-19T18:29:15.141469Z K 7 svn:log V 276 It is safe to initialize locks even on early boot (and it is the same thing all the other architectures already do) thus just initialize kernel_pmap in pmap_bootstrap(). Reported by: alc Reviewed by: alc, marius Tested by: flo, marius Approved by: re (kib) MFC after: 1 week END K 10 svn:author V 6 tuexen K 8 svn:date V 27 2011-09-19T21:47:20.824350Z K 7 svn:log V 100 Cleanup the iterator code, remove code that is never executed. Approved by: re MFC after: 1 month. END K 10 svn:author V 3 kib K 8 svn:date V 27 2011-09-19T22:35:54.023765Z K 7 svn:log V 222 Do not overallocate on the stack. Threaded code might use custom stack size. Reported by: many Tested by: Jeremie Le Hen (previous version) Reviewed by: jilles Approved by: re (bz) MFC after: 2 weeks END K 10 svn:author V 5 gabor K 8 svn:date V 27 2011-09-19T22:49:36.361524Z K 7 svn:log V 281 - Fix a trivial bug in iconv. When there is no space to perform the conversion, conversion must fail and errno must be set to E2BIG. PR: standards/160673 Submitted by: Henning Petersen Reviewed by: pluknet Approved by: re (kib), delphij (mentor) END K 10 svn:author V 6 obrien K 8 svn:date V 27 2011-09-19T23:49:59.598280Z K 7 svn:log V 51 free() filemon structs free list at module unload. END K 10 svn:author V 3 hrs K 8 svn:date V 27 2011-09-20T00:19:36.090755Z K 7 svn:log V 323 Remove RA timer on an interface with !IFF_UP actively after starting to send clean-up RA messages for shutting down. The RA timers could prevent the rtadvd daemon from shutting down because ra_output() just ignored !IFF_UP interfaces and TRANSITIVE->UNCONFIGURED state transition never happend due to it. Spotted by: kib END K 10 svn:author V 6 obrien K 8 svn:date V 27 2011-09-20T00:23:59.821568Z K 7 svn:log V 34 Correct typo in test script name. END K 10 svn:author V 3 hrs K 8 svn:date V 27 2011-09-20T00:29:17.132117Z K 7 svn:log V 205 Copy ip6po_minmtu and ip6po_prefer_tempaddr in ip6_copypktopts(). This fixes inconsistency when options are specified by both setsockopt() and ancillary data types. PR: kern/158307 Approved by: re (bz) END K 10 svn:author V 3 hrs K 8 svn:date V 27 2011-09-20T00:32:30.154200Z K 7 svn:log V 345 Remove RA timer on an interface with !IFF_UP actively after starting to send clean-up RA messages for shutting down. The RA timers could prevent the rtadvd daemon from shutting down because ra_output() just ignored !IFF_UP interfaces and TRANSITIVE->UNCONFIGURED state transition never happened due to it. Spotted by: kib Approved by: re (bz) END K 10 svn:author V 3 hrs K 8 svn:date V 27 2011-09-20T00:37:35.305878Z K 7 svn:log V 110 Use REQUIRE: line to reorder rc.d/bridge instead of BEFORE: line. Pointed out by: dougb Approved by: re (bz) END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-20T04:20:55.598997Z K 7 svn:log V 1879 Undo the BAR TX stuff (as it's broken); correct what the new LH edge of the BAW should be. The BAR TX stuff I committed earlier is just plain wrong. What should happen is this (in ADDBA mode): * If a software retransmit completely fails, all TX for that given TID should be paused until all existing TXes for said TID are completed (failed or not); * Once all the hardware-queued transmissions are completed, the left-hand edge of the BAW (from the TX point of view) should be correct. Ie, there may be further packets queued in the software TX queue, but the left-hand edge of the BAW will point to one frame past the last hardware-queued frame (again, successful or not, it doesn't matter.) * At that point, a BAR should be sent reflecting what the new BAW left hand edge is. * Once that's been received, TX on the given TID should resume. * If it fails, net80211 will cancel the addba session for us; we just need to resume TX'ing on the TID anyway (newly downgraded to non-aggregation.) Now, the things to keep in mind when implementing the correct-er solution: * There may be plenty of outstanding frames on the hardware TX queue, so you can't just send one BAR for each of those. * Whilst the BAR is being TXed, the node may be reset; this SHOULD correctly be handled (ie, cancelling the frame will cause net80211 to reschedule another one) * You can't just call pause() and resume() like I was doing, as multiple outstanding frames may be on the hardware queue (and multiples ones may fail.) A new method of marking a TID as "BAR pending" needs to be added first. That way the TID can be paused and unpaused a total of once, no matter how many outstanding frames fail before the TID has no more frames on the hardware queue. * Finally, this all only works if tid->hwq_depth is completely accurate. This needs to be verified! END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-20T04:30:23.950547Z K 7 svn:log V 171 Manually set the channel when using monitor mode - the firmware doesn't select it automatically. Submitted by: nox Reviewed by: bschmidt Approved by: re PR: kern/160815 END K 10 svn:author V 8 brueffer K 8 svn:date V 27 2011-09-20T04:59:34.782464Z K 7 svn:log V 100 MFC: r225533 Improve the sleep_delay sysctl description by specifying which unit the number is in. END K 10 svn:author V 8 brueffer K 8 svn:date V 27 2011-09-20T05:02:48.265026Z K 7 svn:log V 100 MFC: r225533 Improve the sleep_delay sysctl description by specifying which unit the number is in. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-20T07:19:39.003567Z K 7 svn:log V 11 Whitespace END K 10 svn:author V 7 attilio K 8 svn:date V 27 2011-09-20T08:01:43.939066Z K 7 svn:log V 4 MFC END K 10 svn:author V 7 attilio K 8 svn:date V 27 2011-09-20T08:11:07.496989Z K 7 svn:log V 282 Replace the vm_page_queue_lock mis-usage with a global MD page rwlock that protects the TTE list and page coloring bits. Additively, bring out of any locking some calles to vm_page_aflag_clear() as it is no longer necessary now. Discussed with: alc Reviewed by: alc Tested by: flo END K 10 svn:author V 3 kib K 8 svn:date V 27 2011-09-20T08:34:15.069314Z K 7 svn:log V 134 MFC r225534: Do not try to change the mode or ownership of the root of the mountpoint when newly established mdmfs mount is readonly. END K 10 svn:author V 5 gabor K 8 svn:date V 27 2011-09-20T12:46:07.380263Z K 7 svn:log V 50 - Clean up static function to have less arguments END K 10 svn:author V 5 gabor K 8 svn:date V 27 2011-09-20T13:04:52.752044Z K 7 svn:log V 57 - Dot is not special when dealing with a literal pattern END K 10 svn:author V 8 hselasky K 8 svn:date V 27 2011-09-20T14:17:58.236694Z K 7 svn:log V 245 Avoid starting the USB transfer if an error is already pending. This change fixes a race in device side mode during clear-stall from host, which can cause data to be sent too early on the given endpoint. Approved by: re (kib) MFC after: 1 week END K 10 svn:author V 6 obrien K 8 svn:date V 27 2011-09-20T18:49:19.864469Z K 7 svn:log V 37 Report the acutal filename on error. END K 10 svn:author V 6 grehan K 8 svn:date V 27 2011-09-20T19:07:56.893386Z K 7 svn:log V 48 Create project branch for virtio driver import. END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2011-09-20T20:27:26.030016Z K 7 svn:log V 128 Make KBI changes required for future MFCing of inpcb rtentry / llentry caching. Reviewed by: rwatson, bz Approved by: re (kib) END K 10 svn:author V 3 kib K 8 svn:date V 27 2011-09-20T21:49:54.807851Z K 7 svn:log V 424 Restore the writing of the .bss sections of the dsos (not the main executable) after r190885. The whole region for the dso is mmaped with MAP_NOCORE flag, doing only mprotect(2) over .bss prevented it from writing .bss to core files. Revert the optimization of using mprotect(2) to establish .bss, overlap the section with mmap(2). Reported by: attilio Reviewed by: attilio, emaste Approved by: re (bz) MFC after: 2 weeks END K 10 svn:author V 3 kib K 8 svn:date V 27 2011-09-20T21:53:26.983868Z K 7 svn:log V 193 Use nowait sync request for a vnode when doing softdep cleanup. We possibly own the unrelated vnode lock, doing waiting sync causes deadlocks. Reported and tested by: pho Approved by: re (bz) END K 10 svn:author V 5 gabor K 8 svn:date V 27 2011-09-20T21:53:46.967972Z K 7 svn:log V 168 - Remove a not very necessary sanity check and rely on the offets when REG_STARTEND is set. It is better to let the program segfault than hiding hard to find bugs. END K 10 svn:author V 5 gabor K 8 svn:date V 27 2011-09-20T21:54:43.136295Z K 7 svn:log V 30 - Merge improvements from TRE END K 10 svn:author V 6 obrien K 8 svn:date V 27 2011-09-20T23:22:58.278509Z K 7 svn:log V 111 Things opened R/W should also produce an R record instead of treating an open for O_RDWR the same as O_WRONLY. END K 10 svn:author V 5 gibbs K 8 svn:date V 27 2011-09-20T23:44:34.994937Z K 7 svn:log V 949 Properly handle suspend/resume events in the Xen device framework. Sponsored by: BQ Internet sys/xen/xenbus/xenbusb.c: o In xenbusb_resume(), publish the state transition of the resuming device into XenbusStateIntiailising so that the remote peer can see it. Recording the state locally is not sufficient to trigger a re-connect sequence. o In xenbusb_resume(), defer new-bus resume processing until after the remote peer's XenStore address has been updated. The drivers may need to refer to this information during resume processing. sys/xen/xenbus/xenbusb_back.c: sys/xen/xenbus/xenbusb_front.c: Register xenbusb_resume() rather than bus_generic_resume() as the handler for device_resume events. sys/xen/xenstore/xenstore.c: o Fix grammer in a comment. o In xs_suspend(), pass suspend events on to the child devices (e.g. xenbusb_front/back, that are attached to the XenStore. Approved by: re MFC after: 1 week END K 10 svn:author V 5 gibbs K 8 svn:date V 27 2011-09-21T00:02:44.970827Z K 7 svn:log V 1418 Add suspend/resume support to the Xen blkfront driver. Sponsored by: BQ Internet sys/dev/xen/blkfront/block.h: sys/dev/xen/blkfront/blkfront.c: Remove now unused blkif_vdev_t from the blkfront soft. sys/dev/xen/blkfront/blkfront.c: o In blkfront_suspend(), indicate the desire to suspend by changing the softc connected state to SUSPENDED, and then wait for any I/O pending on the remote peer to drain. Cancel suspend processing if I/O does not drain within 30 seconds. o Enable and update blkfront_resume(). Since I/O is drained prior to the suspension of the VM, the complicated recovery process performed by other Xen blkfront implementations is avoided. We simply tear down the connection to our old peer, and then re-connect. o In blkif_initialize(), fix a resource leak and botched return if we cannot allocate shadow memory for our requests. o In blkfront_backend_changed(), correct our response to the XenbusStateInitialised state. This state indicates that our backend peer has published sufficient data for blkfront to publish ring information and other XenStore data, not that a connection can occur. Blkfront now will only perform connection processing in response to the XenbusStateConnected state. This corrects an issue where blkfront connected before the backend was ready during resume processing. Approved by: re MFC after: 1 week END K 10 svn:author V 5 gibbs K 8 svn:date V 27 2011-09-21T00:06:02.250448Z K 7 svn:log V 306 [ Forced commit. Actual changes accidentally included in r225704 ] sys/dev/xen/control/control.c: Fix locking violations in Xen HVM suspend processing and have it perform similar actions to those performed during an ACPI triggered suspend. Sponsored by: BQ Internet Approved by: re MFC after: 1 week END K 10 svn:author V 5 gibbs K 8 svn:date V 27 2011-09-21T00:08:25.263854Z K 7 svn:log V 812 Correct suspend/resume support in the Netfront driver. Sponsored by: BQ Internet sys/dev/xen/netfront/netfront.c: o Implement netfront_suspend(), a specialized suspend handler for the netfront driver. This routine simply disables the carrier so the driver is idle during system suspend processing. o Fix a leak when re-initializing LRO during a link reset. o In netif_release_tx_bufs(), when cleaning up the grant references for our TX ring, use gnttab_end_foreign_access_ref instead of attempting to grant the page again. o In netif_release_tx_bufs(), we do not track mbufs associated with mbuf chains, but instead just free each mbuf directly. Use m_free(), not m_freem(), to avoid double frees of mbufs. o Refactor some code to enhance clarity. Approved by: re MFC after: 1 week END K 10 svn:author V 5 gibbs K 8 svn:date V 27 2011-09-21T00:13:04.200743Z K 7 svn:log V 532 Modify the netfront driver so it can successfully attach to PV devices with the ioemu attribute set. sys/dev/xen/netfront/netfront.c: o If a mac address for the interface cannot be found in the front-side XenStore tree, look for an entry in the back-side tree. With ioemu devices, the emulator does not populate the front side tree and neither does Xend. o Return an error rather than panic when an attach attempt fails. Reported by: Janne Snabb (fix inspired by patch provided) PR: kern/154302 Approved by: re END K 10 svn:author V 5 gibbs K 8 svn:date V 27 2011-09-21T00:15:29.360198Z K 7 svn:log V 689 Update netfront so that it queries and honors published back-end features. sys/dev/xen/netfront/netfront.c: o Add xn_query_features() which reads the XenStore and records the TSO, LRO, and chained ring-request support of the backend. o Rename xn_configure_lro() to xn_configure_features() and use this routine to manage the setup of TSO, LRO, and checksum offload. o In create_netdev(), initialize if_capabilities and if_hwassist to the capabilities found on all backends. Delegate configuration of if_capenable and the TSO flag if if_hwassist to xn_configure_features(). Reported by: Hugo Silva (fix inspired by patch provided) Approved by: re MFC after: 1 week END K 10 svn:author V 6 grehan K 8 svn:date V 27 2011-09-21T03:37:12.215011Z K 7 svn:log V 178 Initial import of virtio base, PCI front-end, and net/block/balloon drivers. Tested on KVM, Qemu and BHyve. Submitted by: Bryan Venteicher bryanv at daemoninthecloset dot org END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-21T06:35:05.196133Z K 7 svn:log V 104 Undo an autosleep hack that I never finished porting. I'll revisit this when it's time to do autosleep. END K 10 svn:author V 8 hselasky K 8 svn:date V 27 2011-09-21T07:31:16.933180Z K 7 svn:log V 57 MFC r225556: Reduce USB memory usage during enumeration. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-21T13:15:18.062515Z K 7 svn:log V 88 Save and restore the associate id on a reset. This is a candidate to merge into -HEAD. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-21T13:19:45.611600Z K 7 svn:log V 139 Don't use the multicast key search key method in STA mode - just use it in adhoc/hostap mode for now. Obtained from: Atheros, Linux ath9k END K 10 svn:author V 5 gabor K 8 svn:date V 27 2011-09-21T13:22:36.459973Z K 7 svn:log V 31 - Do not allow trailing escape END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-21T16:47:08.488310Z K 7 svn:log V 826 Add a temporary workaround to filter out some of the STA issues I'm seeing in busy 11n tests. I'm occasionally seeing frames destined for other STA's pop up from the AR9160 NIC. But I'm not seeing it from an AR9280 NIC which is also in STA mode. What would thus happen: * the frame would most likely end up entering via ieee80211_input_all(); * It would be punted to the bss node of each VAP; * It would attempt to be decrypted, and may result in the rsc being bumped to a high value and thus causing subsequent frames to fail to be RX'ed due to a suspected replay attack; * It may also end up bumping the TID sequence number, causing subsequent valid frames to also be dropped for being out of sequence. It's quite possible that this is all the result of some subtle descriptor corruption in the ath driver.... END K 10 svn:author V 5 gabor K 8 svn:date V 27 2011-09-21T19:48:27.760004Z K 7 svn:log V 55 - Add forgotten files Reported by: aakuusta@gmail.com END K 10 svn:author V 5 gabor K 8 svn:date V 27 2011-09-21T19:50:26.682901Z K 7 svn:log V 40 - Clean up that is actually unnecessary END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-21T23:38:15.889258Z K 7 svn:log V 185 Return rx clear for the extension channel when requested instead of incorrectly returning rx clear for the control channel. This is a merge candidate to -HEAD. Obtained from: Atheros END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-21T23:48:19.990910Z K 7 svn:log V 2161 Work around some issues seen with the AR9160 when RX'ing aggregate frames. It seems that the AR9160 in STA mode will occasionally pass A-MPDU subframes up to the driver which aren't destined for it. There's nothing obviously wrong - it's not occuring during scanning, AR_RX_FILTER is correct, AID is now being set correct, the STA ID and BSSID values are correct. It only occurs on aggregate frame RX. I've not seen it occur on normal MPDU RX. This commit: * backs out some previous debugging that snuck in; * adds some local debugging (which I'll make more useful, but still leave commented out by default) to dump descriptors which aren't for us; * removes the use of the crypto key cache id for determining the peer, as it's possible a valid looking frame will come in on a key id that is active. As an example of the last one: R[ 0] (DS.V:0xa0a13780 DS.P:0xa13780) L:00a137e0 D:045fb800 * 00000000 00000800 8d2a282a 000b806c 3e60da94 00000103 052c163d 00000000 00000000 00000000 80030143 FRDS 00:1c:c0:43:87:2b->d4:9a:20:5a:db:0c(00:1b:b1:58:f6:f0) data QoS [TID 0] WEP [IV ca 9c 00 e0 08 00 KID 0] 70M +5 8842 2c00 d49a 205a db0c 001b b158 f6f0 001c c043 872b c0c6 0000 c0ca ca9c 0020 e008 0000 aec7 45d1 e2b8 b6c2 cb04 ef91 f063 9fc6 8f61 f491 fc5e db4b cd0e a2c7 5ccc 9aaa ced2 0024 dcd9 d744 f9ba 0e2e d09d f43e 0f87 7ae6 5ff0 191c c7fb 2e38 0bc7 f78c 0f87 17e8 f63d fe71 That frame is to a STA address that isn't me. It's marked as valid and completed. The EVM values are all whacked though (see how they're all marked as 0x00000000 in the receive status) even though the received rate is an MCS rate (0x8d, high byte of the third status word.) This frame ends up being bumped to the BSS node for the VAP; it updates the CCMP crypto state: ccmp_decap: tid=0, keyix=4, wk_rxkeyix=4 (local debugging) .. and that causes subsequent frames to fail. This is only going to be merged back to -HEAD if I discover it's a hardware issue rather than a driver bug somewhere else. Finally, I've not yet seen this surface on Merlin (AR9280.) I've not yet tested the AR5416 but if it's a hardware bug, it's likely it has the same issue. END K 10 svn:author V 7 delphij K 8 svn:date V 27 2011-09-22T01:13:40.570468Z K 7 svn:log V 129 MFC r225452: Sync pf.os with OpenBSD: - Update OpenBSD fingerprints through OpenBSD 4.9 - Fix typos. Obtained from: OpenBSD END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-22T02:07:29.152845Z K 7 svn:log V 16 Add another WAR END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-22T03:03:40.115633Z K 7 svn:log V 102 Remove an unneeded check; tidy it up a bit, add comments about where things may still go pear-shaped. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-22T07:29:45.470485Z K 7 svn:log V 720 Free the mbuf before recycling the current rx ath_buf when doing a radiotap. ath_rxbuf_init() doesn't free/realloc existing mbufs (if they do exist), it just passes the buffer pointer/length to the HAL rx descriptor setup code. When doing radiotap however, the mbuf length is overridden before passing it off to radiotap/bpf. This means that the buffer will stay bein potentially shorter until it's next used for a valid frame. I was seeing short frames show up in packet traces, along with seemingly invalid frame contents, probe requests and other random stuff that didn't make sense. These only show up when doing a background radiotap/tcpdump. This patch fixes this behaviour. This is a merge candidate to HEAD. END K 10 svn:author V 8 jchandra K 8 svn:date V 27 2011-09-22T08:24:33.013432Z K 7 svn:log V 51 Delete the XLP merge branch, as the merge is done. END K 10 svn:author V 3 kib K 8 svn:date V 27 2011-09-22T08:39:20.388610Z K 7 svn:log V 142 MFC r225582: Use the proper dynamic tls block to calculate the tls variable address in case tls data generation was updated. PR: misc/160721 END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-22T10:52:14.125706Z K 7 svn:log V 111 Revert the STA keycache multicast entry fix for now; just so I can tinker with some other things that need it. END K 10 svn:author V 7 attilio K 8 svn:date V 27 2011-09-22T11:07:11.802092Z K 7 svn:log V 157 MFC r225516: Return ENOSPC rather than ENXIO when dump_write() cannot proceed in order to correctl deal with consumers. Sponsored by: Sandvine Incorporated END K 10 svn:author V 7 attilio K 8 svn:date V 27 2011-09-22T12:53:27.855398Z K 7 svn:log V 4 MFC END K 10 svn:author V 5 simon K 8 svn:date V 27 2011-09-22T17:51:09.208918Z K 7 svn:log V 171 - Add 9.0 (stable and releng) branches to svn2cvs mapping. [1] - Checkpoint current list of hacks mainly to handle support for 'Replaced' files. Requested by: kensmith END K 10 svn:author V 2 jh K 8 svn:date V 27 2011-09-22T17:59:33.535587Z K 7 svn:log V 125 MFC r225587: Expose "log" in the default devfs rules. /etc/rc.d/jail creates /dev/log as a symbolic link. PR: conf/160711 END K 10 svn:author V 2 jh K 8 svn:date V 27 2011-09-22T18:21:15.099303Z K 7 svn:log V 125 MFC r225587: Expose "log" in the default devfs rules. /etc/rc.d/jail creates /dev/log as a symbolic link. PR: conf/160711 END K 10 svn:author V 5 gabor K 8 svn:date V 27 2011-09-22T22:08:09.270562Z K 7 svn:log V 686 MFC r202992: Cache failing and opened catalogs in catopen() and related functions. Continuous catopen() calls cause 4 failig stat(2) each, which means a lot of overhead. It is also a good idea to keep the opened catalogs in the memory to speed up further catopen() calls to the same catalog since these catalogs are not big at all. In this case, we count references and only free() the allocated space when the reference count reaches 0. The reads and writes to the cache are syncronized with an rwlock when these functions are called from a threaded program. MFC r202993, r203174, r203719, r204110: Small fixes and style nits for the above change. Approved by: delphij (mentor) END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-22T22:52:24.059326Z K 7 svn:log V 92 Now that I've found the root cause missing TX interrupt issue, revert this previous commit. END K 10 svn:author V 8 kensmith K 8 svn:date V 27 2011-09-23T00:50:36.255486Z K 7 svn:log V 143 Add stable/9 to approvers and paths in preparation for creating stable/9 as part of the 9.0-RELEASE release cycle. Approved by: re (implicit) END K 10 svn:author V 8 kensmith K 8 svn:date V 27 2011-09-23T00:51:37.052201Z K 7 svn:log V 88 Copy head to stable/9 as part of 9.0-RELEASE release cycle. Approved by: re (implicit) END K 10 svn:author V 8 kensmith K 8 svn:date V 27 2011-09-23T02:58:59.363333Z K 7 svn:log V 134 Do the change from 8 to 9 that was supposed to follow copying the line. It's been one of those nights... Approved by: re (implicit) END K 10 svn:author V 8 lstewart K 8 svn:date V 27 2011-09-23T05:28:09.349305Z K 7 svn:log V 269 MFC r225583: Rename the cc.4 and cc.9 modular congestion control related man pages to mod_cc.4 and mod_cc.9 respectively to avoid any possible confusion with the cc.1 gcc man page. Update references to these man pages where required. Requested by: Grenville Armitage END K 10 svn:author V 8 lstewart K 8 svn:date V 27 2011-09-23T05:35:24.244835Z K 7 svn:log V 131 Follow up to r225738 to fix the mismerge of share/man/man4/Makefile. This is an intentional direct commit to the 8-STABLE branch. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-23T13:53:28.938825Z K 7 svn:log V 211 Re-introduce the FIFO packet scheduling code. This makes TX packet scheduling slightly fairer in hostap mode where multiple nodes and multiple TIDs may be actively TX'ing. Obtained from: Linux ath9k, Atheros END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-23T14:55:26.814120Z K 7 svn:log V 51 Revert these values; the ath9k values work better. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-23T15:32:03.299458Z K 7 svn:log V 26 Use fast taskqueues here. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-23T15:36:52.698924Z K 7 svn:log V 112 For now, break out the beacon processing task into a separate task so it doesn't interrupt TX or RX completion. END K 10 svn:author V 8 kensmith K 8 svn:date V 27 2011-09-23T16:37:28.618470Z K 7 svn:log V 45 Ready for BETA3. Approved by: re (implicit) END K 10 svn:author V 5 gabor K 8 svn:date V 27 2011-09-23T21:01:44.785213Z K 7 svn:log V 35 - Minor improvements to the manual END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-24T07:33:29.846992Z K 7 svn:log V 432 Change how RXEOL/RXORN is handled. * If the condition is hit on chips where it isn't fatal, don't restart the PCU by stopping/starting; this introduces quite long pauses and it doesn't look like it's needed (at least for AR9160 hostap.) * Add some debugging to see what's going on. It turns out that I either see a full RX queue (512 packets) or an empty RX queue (0); I'll have to do some further poking to see why that is. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-25T02:41:50.975314Z K 7 svn:log V 76 Begin to flesh out the random crypto related issues I've seen crop up here. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-25T07:38:33.621707Z K 7 svn:log V 14 Update things END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-25T07:38:59.504640Z K 7 svn:log V 40 Add the TID to the replay debug message END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-25T07:39:19.702107Z K 7 svn:log V 71 Add a reminder to check whether this should be also done for HT nodes. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-25T07:43:07.471183Z K 7 svn:log V 111 First cut at the AR5416 blockack/aggregation bug. It doesn't do the reset; it just prints out a debug message. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-25T07:49:03.021909Z K 7 svn:log V 83 Add some further debugging to get a handle on when this BAW tracking error occurs. END K 10 svn:author V 7 wkoszek K 8 svn:date V 27 2011-09-25T11:17:34.469378Z K 7 svn:log V 199 Please welcome Jakub Wojciech Klama (jceel@) as a new src committer. Jakub will be working on FreeBSD/arm and all sorts of embedded stuff. cognet@ and I will be mentoring Jakub. Approved by: core END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-25T15:07:16.377008Z K 7 svn:log V 20 Update with comment END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-25T15:07:50.909287Z K 7 svn:log V 57 Add code to blank the tx aggregation statistics as well. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-25T15:28:09.297119Z K 7 svn:log V 41 Disable TX interrupt mitigation for now. END K 10 svn:author V 8 kensmith K 8 svn:date V 27 2011-09-26T02:27:04.679360Z K 7 svn:log V 152 Shift head from 9.0-CURRENT to 10.0-CURRENT in preparation for releasing it from the 9.0-RELEASE release cycle code freeze. Approved by: re (implicit) END K 10 svn:author V 4 neel K 8 svn:date V 27 2011-09-26T07:05:40.477566Z K 7 svn:log V 40 Kernel configuration for a bhyve guest. END K 10 svn:author V 8 hselasky K 8 svn:date V 27 2011-09-26T07:38:01.923893Z K 7 svn:log V 64 MFC r225659: Implement missing USB debug information functions. END K 10 svn:author V 7 attilio K 8 svn:date V 27 2011-09-26T10:12:16.847473Z K 7 svn:log V 89 MFC r225675: Initialize the pmap lock for the kernel pmap when it is really appropriate. END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-26T11:35:34.573414Z K 7 svn:log V 12 Fix MANDIR. END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-26T11:39:12.483832Z K 7 svn:log V 35 Fix usage string and use sysexits. END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-26T11:40:28.274286Z K 7 svn:log V 127 Make -r the default. Change the default minsize to bsize so we can tell if the logical and physical block sizes are the same. END K 10 svn:author V 8 kensmith K 8 svn:date V 27 2011-09-26T11:57:48.980485Z K 7 svn:log V 127 Forgot to add "RELENG_8" to list of CVS tags. Submitted by: Mamontov Roman Approved by: re (implicit) END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-26T12:08:15.276526Z K 7 svn:log V 16 Add a man page. END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-26T12:20:33.286058Z K 7 svn:log V 41 Add an option to open the device O_SYNC. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-26T15:26:59.131598Z K 7 svn:log V 222 Update if_de(4) to fix bugs - a combination of bitrot and busdma changes. This commit is a merge of 197465, 197464, 197463, 197461, 177937 and 170389. Submitted by: Andreas Longwitz Reviewed by: jhb END K 10 svn:author V 4 jkim K 8 svn:date V 27 2011-09-26T21:40:21.494164Z K 7 svn:log V 24 Import ACPICA 20110922. END K 10 svn:author V 4 jkim K 8 svn:date V 27 2011-09-26T21:42:22.794984Z K 7 svn:log V 21 Tag ACPICA 20110922. END K 10 svn:author V 8 kensmith K 8 svn:date V 27 2011-09-27T03:20:52.829769Z K 7 svn:log V 149 Release head from code freeze now that stable/9 has been created and that will be what eventually becomes 9.0-RELEASE. Approved by: core (implicit) END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-27T03:27:54.512398Z K 7 svn:log V 89 Add some KTR probes to help me figure out interrupt and RXEOL related issues are hiding. END K 10 svn:author V 3 rmh K 8 svn:date V 27 2011-09-27T05:30:52.942969Z K 7 svn:log V 61 Add myself to committers-src.dot. Approved by: kib (mentor) END K 10 svn:author V 3 pjd K 8 svn:date V 27 2011-09-27T06:43:51.557432Z K 7 svn:log V 68 Ensure that pjdlog functions don't modify errno. MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2011-09-27T06:53:30.385976Z K 7 svn:log V 91 Mikolaj needs no mentor anymore. Thanks to Kostik for the help with co-mentoring Mikolaj. END K 10 svn:author V 5 gabor K 8 svn:date V 27 2011-09-27T07:00:24.628363Z K 7 svn:log V 162 - Move misplaced reference [1] - Sort references [1] - s/Since/Because/ [2] Submitted by: arundel [1], Chad Perrin [2] MFC after: 3 days END K 10 svn:author V 7 pluknet K 8 svn:date V 27 2011-09-27T07:05:41.748314Z K 7 svn:log V 45 Fix typo in OSIOCGIFADDR. MFC after: 3 days END K 10 svn:author V 8 hselasky K 8 svn:date V 27 2011-09-27T07:06:02.600966Z K 7 svn:log V 208 Add quirks for some USB mass storage devices which doesn't respond after trying to query the synchronize cache support. Submitted by: Keith White PR: usb/160911 Approved by: re (kensmith) MFC after: 1 week END K 10 svn:author V 4 stas K 8 svn:date V 27 2011-09-27T07:14:12.793656Z K 7 svn:log V 345 - Add missing interdependencies to kerberos libraries. Some of the kerberos libraries were not linked properly (missing dependencies), which causes 3rd party applications linking to fail when --as-needed ld flag is used. I also added the --no-undefined ld(1) flag to make sure that there're no missing dependencies. MFC after: 3 days END K 10 svn:author V 8 hselasky K 8 svn:date V 27 2011-09-27T07:19:01.986160Z K 7 svn:log V 49 MFC r225695: Fix minor race in device side mode. END K 10 svn:author V 4 stas K 8 svn:date V 27 2011-09-27T07:33:04.063416Z K 7 svn:log V 101 - MFC r225354: - Fix NULL pointer dereference when a packet of uneven size is being transmitted. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2011-09-27T07:52:39.960173Z K 7 svn:log V 106 No need to use KEEP_ERRNO() macro around pjdlog functions, as they don't modify errno. MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2011-09-27T07:57:15.145005Z K 7 svn:log V 174 Prefer PJDLOG_ASSERT() and PJDLOG_ABORT() over assert() and abort(). pjdlog versions will log problem to syslog when application is running in background. MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2011-09-27T07:59:10.994163Z K 7 svn:log V 99 Correct two mistakes when converting asserts to PJDLOG_ASSERT()/PJDLOG_ABORT(). MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2011-09-27T08:04:01.861199Z K 7 svn:log V 104 - Convert some impossible conditions into assertions. - Add missing 'if' in comment. MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2011-09-27T08:21:00.912668Z K 7 svn:log V 72 Prefer PJDLOG_ASSERT()/PJDLOG_ABORT() over assert(). MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2011-09-27T08:26:09.961318Z K 7 svn:log V 84 No need to wrap pjdlog functions around with KEEP_ERRNO() macro. MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2011-09-27T08:50:37.389736Z K 7 svn:log V 90 Use PJDLOG_ASSERT() and PJDLOG_ABORT() everywhere instead of assert(). MFC after: 3 days END K 10 svn:author V 3 mav K 8 svn:date V 27 2011-09-27T09:30:20.669693Z K 7 svn:log V 52 Fix interrupt counters dumping on SW_WATCHDOG fire. END K 10 svn:author V 3 mav K 8 svn:date V 27 2011-09-27T09:32:34.735637Z K 7 svn:log V 83 Add one more ID for the Marvell 88SE9128 6Gbps SATA controller. MFC after: 3 days END K 10 svn:author V 3 kib K 8 svn:date V 27 2011-09-27T12:14:43.803811Z K 7 svn:log V 193 Install ciss(4) ioctl header (together with other .h files from sys/dev/ciss). PR: kern/109813 Discussued with: Alex Samorukov (smartmontools maintainer) MFC after: 1 week END K 10 svn:author V 3 kib K 8 svn:date V 27 2011-09-27T13:17:02.587561Z K 7 svn:log V 325 Do not deliver SIGTRAP on exec as the normal signal, use ptracestop() on syscall exit path. Otherwise, if SIGTRAP is ignored, that tdsendsignal() do not want to deliver the signal, and debugger never get a notification of exec. Found and tested by: Anton Yuzhaninov Discussed with: jhb MFC after: 2 weeks END K 10 svn:author V 6 cognet K 8 svn:date V 27 2011-09-27T13:20:41.768526Z K 7 svn:log V 31 property changes from svnmerge END K 10 svn:author V 2 bz K 8 svn:date V 27 2011-09-27T13:27:17.097768Z K 7 svn:log V 344 Unbreak no-ip and no-inet6 module builds with ipfw. For now continue to build the ip_fw_pfil.c hooks and ipfw even in case of no-ip under the assumption that the private L2 hook (which hopefully eventually will be a pfil hook as well) can still be useful. Allow building the module without inet as well. Glanced at by: jhb MFC after: 3 days END K 10 svn:author V 7 attilio K 8 svn:date V 27 2011-09-27T13:42:11.755762Z K 7 svn:log V 439 Revert r225372: wdog_kern_pat() acquires eventhandler mutex, thus it cannot work in kernel context (from where kdb_trap() runs). The right way to fix this is both offering the cpu-stop-on-panic-and-skip-locking logic and also a context for KDB to officially run. We can re-enable this (or a similar) improvement when these 2 patches hit the tree. Sponsored by: Sandvine Incorporated Discussed with: emaste, rstone MFC after: immediately END K 10 svn:author V 6 cognet K 8 svn:date V 27 2011-09-27T13:47:23.541177Z K 7 svn:log V 15 Attempt to MFC END K 10 svn:author V 7 attilio K 8 svn:date V 27 2011-09-27T13:50:30.287063Z K 7 svn:log V 34 MFC r225794: Backout r225372 MFC. END K 10 svn:author V 7 attilio K 8 svn:date V 27 2011-09-27T14:00:32.722081Z K 7 svn:log V 133 Backout r225372. Eventhandler mutex cannot be acquired in kernel context. Sponsored by: Sandvine Incorporated Approved by: re (kib) END K 10 svn:author V 6 emaste K 8 svn:date V 27 2011-09-27T14:28:07.876483Z K 7 svn:log V 698 Improve battery capacity reporting When a status pointer is passed in mfi_dcmd_command does not return an errno (if the ioctl is successful), so move the test for NO_HW_PRESENT outside of the error case. This should fix incorrect reporting for systems with a dead or no battery. Additionally, handle error codes other than NO_HW_PRESENT by omitting the battery capacity display. LSI's supercap-based parts (CV series) report their data using the same interface as battery-based parts, except that they do not include the capacity stats (state of charge, cumulative charge cycles, etc.) Reviewd by: jhb, bz Tested by: pluknet@, Garrett Cooper PR: bin/160581 MFC after: 1 week END K 10 svn:author V 3 mav K 8 svn:date V 27 2011-09-27T15:08:59.167990Z K 7 svn:log V 88 MFC r225788: Fix interrupt counters dumping on SW_WATCHDOG fire. Approved by: re (kib) END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-27T15:57:13.347448Z K 7 svn:log V 315 Followup to r225599: the fseek() was a no-op since the file was opened in append mode. Open it in read-write mode instead. Also move the fseek up one level to cover the (unlikely but not impossible) case where the server accepts ranges but does not send a Content-Size header. PR: bin/117277 MFC after: 3 weeks END K 10 svn:author V 4 jkim K 8 svn:date V 27 2011-09-27T16:33:17.922206Z K 7 svn:log V 74 Avoid accidental conflicts with C++ operator keywords. MFC after: 1 week END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-27T16:57:14.180694Z K 7 svn:log V 173 Lengthen the progress bar to cover the entire width of the numbers which will replace it, change the indicator character from '|' to 'O', and increase the update frequency. END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-27T17:04:13.129663Z K 7 svn:log V 89 MFH r225800: finally fix resume mode with non-cooperating server. Approved by: re (kib) END K 10 svn:author V 6 emaste K 8 svn:date V 27 2011-09-27T17:11:09.623549Z K 7 svn:log V 664 MFC r225798: Improve battery capacity reporting When a status pointer is passed in mfi_dcmd_command does not return an errno (if the ioctl is successful), so move the test for NO_HW_PRESENT outside of the error case. This should fix incorrect reporting for systems with a dead or no battery. Additionally, handle error codes other than NO_HW_PRESENT by omitting the battery capacity display. LSI's supercap-based parts (CV series) report their data using the same interface as battery-based parts, except that they do not include the capacity stats (state of charge, cumulative charge cycles, etc.) PR: bin/160581 Approved by: re END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-27T17:11:31.341144Z K 7 svn:log V 226 Use fseeko() instead of fseek(). The rest of the code is off_t-aware, but the use of fseek() means fetch(1) can't correctly resume a transfer that was interrupted past the 2 GB mark. Pointed out by: ache@ MFC after: 3 weeks END K 10 svn:author V 8 mckusick K 8 svn:date V 27 2011-09-27T17:34:02.116419Z K 7 svn:log V 235 This update eliminates the system hang reported in kern/160662 when taking a snapshot on a filesystem running with journaled soft updates. Reported by: Hans Ottevanger Fix verified by: Hans Ottevanger PR: kern/160662 END K 10 svn:author V 8 mckusick K 8 svn:date V 27 2011-09-27T17:41:48.506358Z K 7 svn:log V 438 This update eliminates a lock-order reversal warning discovered whle tracking down the system hang reported in kern/160662 and corrected in revision 225806. The LOR is not the cause of the system hang and indeed cannot cause an actual deadlock. However, it can be easily eliminated by defering the acquisition of a buflock until after all the vnode locks have been acquired. Reported by: Hans Ottevanger PR: kern/160662 END K 10 svn:author V 8 schweikh K 8 svn:date V 27 2011-09-27T17:54:10.471210Z K 7 svn:log V 81 Fix grammar. PR: 140457 Submitted by: jeremyhu AT apple.com MFC after: 2 weeks END K 10 svn:author V 7 trociny K 8 svn:date V 27 2011-09-27T18:14:04.584603Z K 7 svn:log V 632 When script(1) reads EOF from input it starts spinning on zero-byte reads eating 100% CPU. Fix this by skipping select on STDIN after reading EOF -- permanently if STDIN is not terminal and for one second if it is. Also after reading EOF from STDIN we have to pass it to the program being scripted. The previous approach was to write zero bytes into the pseudo-terminal. This does not work because zero-byte write does not have any effect on read. Fix this by sending VEOF instead. Submitted by: Ronald Klop Discussed with: kib, Chris Torek Approved by: kib MFC after: 1 week END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-27T18:23:58.622907Z K 7 svn:log V 53 Long overdue: make passive mode the default for ftp. END K 10 svn:author V 4 stas K 8 svn:date V 27 2011-09-27T18:40:13.492675Z K 7 svn:log V 374 - MFC r225778: - Add missing interdependencies to kerberos libraries. Some of the kerberos libraries were not linked properly (missing dependencies), which causes 3rd party applications linking to fail when --as-needed ld flag is used. I also added the --no-undefined ld(1) flag to make sure that there're no missing dependencies. Approved by: re (kib) END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-27T18:42:09.102942Z K 7 svn:log V 187 Think first, commit second. 1. Allow the caller to select active mode. 2. Fix the envar logic so it *always* overrides the caller's flags. 3. Document the change from active to passive. END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-27T18:53:35.648088Z K 7 svn:log V 11 Bump date. END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-27T18:57:26.469972Z K 7 svn:log V 49 Update copyright dates and strip my middle name. END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-27T19:02:44.727966Z K 7 svn:log V 105 Document the fact that passive mode is now the default. Update copyright dates and strip my middle name. END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-27T19:08:27.939656Z K 7 svn:log V 71 MFH r225805: use fseeko(2) instead of fseek(2). Approved by: re (kib) END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-27T19:54:58.212053Z K 7 svn:log V 29 Don't force active mode FTP. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-28T02:54:42.354445Z K 7 svn:log V 94 Update the default AIFS value for hostap mode. Obtained from: Linux ath9k, Atheros reference END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-28T03:03:23.171528Z K 7 svn:log V 724 The AR5212 setup path (also used by the AR5416 code) configures a local variable with a beacon interval of 100 TU. This never gets modified if the beacon interval configuration changes. This may have been correct in earlier times, but with the advent of staggered beacons (which default to 1 / ATH_BCBUF beacon interval, so 25 TU here) this value is incorrect. It is used to configure the default CABQ readytime. So here, the cabq was being configured to be much greater than the target beacon timer (TBTT.) The driver should be configuring a cabq readytime value rather then leaving it to the HAL to choose sensible defaults. This should be done in the future - I'm simply trying to ensure sensible defaults are chosen. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-28T03:05:04.738015Z K 7 svn:log V 512 Change the default CABQ time to be 70% of the beacon interval, rather than the whole beacon interval. The reference driver and Linux ath9k both choose 80% of the beacon interval and they do it in the driver rather than the HAL (Ath reference) or ath9k_hw (ath9k.) This quietens stuck beacon conditions on my AR9220/AR9280 based NICs when a lot of burst broadcast/multicast traffic is going on. It doesn't seem to annoy the earlier MACs as much as the AR9280 and later one. Obtained from: Linux ath9k, Atheros END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-28T03:07:51.898416Z K 7 svn:log V 130 Fix lock order to be correcter. Nothing else locks these two queues (cabq, avp mcastq), but it should be consistent and correct. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-28T03:11:51.895042Z K 7 svn:log V 78 Don't bother triggering the cabq queue if it's empty. Obtained from: Atheros END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-28T07:46:28.592011Z K 7 svn:log V 25 Revert autoprops damage. END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-28T08:13:01.302894Z K 7 svn:log V 121 More autoprops damage, compounded by the fact that some source files, for some reason, had the x bit set in the tarball. END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-28T08:14:41.880210Z K 7 svn:log V 31 Vendor import of OpenSSH 5.9p1 END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-28T08:19:45.078252Z K 7 svn:log V 18 Tag OpenSSH 5.9p1 END K 10 svn:author V 2 bz K 8 svn:date V 27 2011-09-28T08:47:17.722492Z K 7 svn:log V 268 Fix handling of corrupt compress(1)ed data. [11:04] Add missing length checks on unix socket addresses. [11:05] Approved by: so (cperciva) Approved by: re (kensmith) Security: FreeBSD-SA-11:04.compress Security: CVE-2011-2895 [11:04] Security: FreeBSD-SA-11:05.unix END K 10 svn:author V 2 mm K 8 svn:date V 27 2011-09-28T11:57:10.160174Z K 7 svn:log V 339 Remove assertion that prevents zfs rename of datasets with mountpoint=none or mountpoint=legacy that have children datasets. This also fixes dataset rename when receiving incremental snapshots as reported on freebsd-fs@ This assertion was made triggerable by opensolaris change #10196. PR: bin/160400 Reviewed by: pjd MFC after: 1 week END K 10 svn:author V 7 attilio K 8 svn:date V 27 2011-09-28T12:13:15.903116Z K 7 svn:log V 4 MFC END K 10 svn:author V 3 pjd K 8 svn:date V 27 2011-09-28T13:08:51.921299Z K 7 svn:log V 367 After every activemap change flush disk's write cache, so that write reordering won't make the actual write to be committed before marking the coresponding extent as dirty. It can be disabled in configuration file. If BIO_FLUSH is not supported by the underlying file system we log a warning and never send BIO_FLUSH again to that GEOM provider. MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2011-09-28T13:13:43.491487Z K 7 svn:log V 40 Break a bit earlier. MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2011-09-28T13:19:47.403278Z K 7 svn:log V 129 If the underlying provider doesn't support BIO_FLUSH, log it only once and don't bother trying in the future. MFC after: 3 days END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-28T13:21:12.751835Z K 7 svn:log V 29 Forgot to remove dead files. END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-28T13:21:53.033650Z K 7 svn:log V 29 Forgot to remove dead files. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2011-09-28T13:25:27.633335Z K 7 svn:log V 33 Correct typo. MFC after: 3 days END K 10 svn:author V 3 mav K 8 svn:date V 27 2011-09-28T13:27:31.947938Z K 7 svn:log V 4 MFC END K 10 svn:author V 2 bz K 8 svn:date V 27 2011-09-28T13:48:36.678408Z K 7 svn:log V 446 Pass the fibnum where we need filtering of the message on the rtsock allowing routing daemons to filter routing updates on an rtsock per FIB. Adjust raw_input() and split it into wrapper and a new function taking an optional callback argument even though we only have one consumer [1] to keep the hackish flags local to rtsock.c. PR: kern/134931 Submitted by: multiple (see PR) Suggested by: rwatson [1] Reviewed by: rwatson MFC after: 3 days END K 10 svn:author V 3 kib K 8 svn:date V 27 2011-09-28T14:51:28.859491Z K 7 svn:log V 121 Use the explicitly-sized types for the dirty and valid masks. Requested by: attilio Reviewed by: alc MFC after: 2 weeks END K 10 svn:author V 3 mav K 8 svn:date V 27 2011-09-28T14:52:25.666006Z K 7 svn:log V 538 MFprojects/hid: Import the rest of HID improvements from the branch: - improve report descriptor parser in libusbhid to handle several kinds of reports same time; - add to the libusbhid API two functions wrapping respective kernel IOCTLs for reading and writing reports; - tune uhid IOCTL interface to allow reading and writing arbitrary report, when multiple supported by the device; - teach usbhidctl to set output and feature reports; - make usbhidaction support all the same item names as bhidctl. Sponsored by: iXsystems, inc. END K 10 svn:author V 3 kib K 8 svn:date V 27 2011-09-28T14:57:50.718139Z K 7 svn:log V 406 Use the trick of performing the atomic operation on the contained aligned word to handle the dirty mask updates in vm_page_clear_dirty_mask(). Remove the vm page queue lock around vm_page_dirty() call in vm_fault_hold() the sole purpose of which was to protect dirty on architectures which does not provide short or byte-wide atomics. Reviewed by: alc, attilio Tested by: flo (sparc64) MFC after: 2 weeks END K 10 svn:author V 3 kib K 8 svn:date V 27 2011-09-28T15:01:20.824568Z K 7 svn:log V 230 Remove locking of the vm page queues from several pmaps, which only protected the dirty mask updates. The dirty mask updates are handled by atomics after the r225840. Submitted by: alc Tested by: flo (sparc64) MFC after: 2 weeks END K 10 svn:author V 6 adrian K 8 svn:date V 27 2011-09-28T15:36:08.770655Z K 7 svn:log V 526 Flip back to the atheros defaults for now. Now that I've figured out why I'm getting performance issues on these mips boards (interrupts occuring in a race window between critical_enter and "wait", and thus being delayed until the next interrupt), this provides better RX performance under heavy UDP load. This doesn't fix the issue - it just means that under heavy UDP load, I'm not getting 1 RX interrupt every 2ms. This turns out not to be often enough in some occasional situations. Some more research is needed though.. END K 10 svn:author V 3 kib K 8 svn:date V 27 2011-09-28T16:12:15.166958Z K 7 svn:log V 50 Fix grammar. Submitted by: bf MFC after: 2 weeks END K 10 svn:author V 5 crees K 8 svn:date V 27 2011-09-28T17:03:49.531734Z K 7 svn:log V 201 Include limits.h instead of sys/limits.h to improve portability. PR: bin/150772 Submitted by: Derrick Brashear Reviewed by: Garrett Cooper Approved by: cognet END K 10 svn:author V 7 delphij K 8 svn:date V 27 2011-09-28T18:03:53.501556Z K 7 svn:log V 127 Sync RCS id with NetBSD, this was intentionally omitted from the advisory in order to reduce patchset size. MFC after: 3 days END K 10 svn:author V 6 eadler K 8 svn:date V 27 2011-09-28T18:49:37.137921Z K 7 svn:log V 54 - add myself to calendar Approved by: sahil (mentor) END K 10 svn:author V 2 ed K 8 svn:date V 27 2011-09-28T18:53:36.761475Z K 7 svn:log V 898 Get rid of major/minor number distinction. As of FreeBSD 6, devices can only be opened through devfs. These device nodes don't have major and minor numbers anymore. The st_rdev field in struct stat is simply based a copy of st_ino. Simply display device numbers as hexadecimal, using "%#jx". This is allowed by POSIX, since it explicitly states things like the following (example taken from ls(1)): "If the file is a character special or block special file, the size of the file may be replaced with implementation-defined information associated with the device in question." This makes the output of these commands more compact. For example, ls(1) now uses approximately four columns less. While there, simplify the column length calculation from ls(1) by calling snprintf() with a NULL buffer. Don't be afraid; if needed one can still obtain individual major/minor numbers using stat(1). END K 10 svn:author V 6 eadler K 8 svn:date V 27 2011-09-28T18:56:02.574145Z K 7 svn:log V 66 - add myself to committers-ports.dot Approved by: sahil (mentor) END K 10 svn:author V 7 delphij K 8 svn:date V 27 2011-09-28T19:01:15.259370Z K 7 svn:log V 138 Test if the interface is afif in dhcpif() and syncdhcpif(), as done in ipv6_autoconfif. Reviewed by: hrs (freebsd-rc@) MFC after: 1 week END K 10 svn:author V 8 mckusick K 8 svn:date V 27 2011-09-28T19:36:21.638767Z K 7 svn:log V 367 MFC: r225806: This update eliminates the system hang reported in kern/160662 when taking a snapshot on a filesystem running with journaled soft updates. As journaled soft updates first appeared in 9.0, this will be the only MFC of this change. Approved by: re (kib) Reported by: Hans Ottevanger Fix verified by: Hans Ottevanger PR: kern/160662 END K 10 svn:author V 8 mckusick K 8 svn:date V 27 2011-09-28T19:38:47.736880Z K 7 svn:log V 588 MFC r225807: This update eliminates a lock-order reversal warning discovered whle tracking down the system hang reported in kern/160662 and corrected in revision 225806 (MFC'ed as 225850). The LOR is not the cause of the system hang and indeed cannot cause an actual deadlock. However, it can be easily eliminated by defering the acquisition of a buflock until after all the vnode locks have been acquired. As journaled soft updates first appeared in 9.0, this will be the only MFC of this change. Approved by: re (kib) Reported by: Hans Ottevanger PR: kern/160662 END K 10 svn:author V 3 des K 8 svn:date V 27 2011-09-28T20:42:22.090005Z K 7 svn:log V 57 Belatedly regenerate after application of the HPN patch. END K 10 svn:author V 6 eadler K 8 svn:date V 27 2011-09-28T20:46:52.381363Z K 7 svn:log V 104 - fix whitespace issue in calendar - add a n after \ Submitted by: brueffer Approved by: bapt (mentor) END K 10 svn:author V 6 eadler K 8 svn:date V 27 2011-09-28T21:42:13.502648Z K 7 svn:log V 72 - I am not the oldest committer alive (yet) Approved by: bapt (mentor) END