ƒ´M183183 176 175 125 163 149 247 188 119 221 207 254 125 351 194 450 146 437 173 286 541 369 156 164 158 126 1086 122 115 266 387 413 393 352 478 272 217 592 152 104 188 140 228 228 167 136 163 132 150 674 300 155 264 127 120 259 255 511 466 147 127 328 137 124 319 317 1147 750 111 152 608 178 188 752 203 153 292 112 164 170 172 224 116 332 463 168 162 133 223 198 119 513 185 589 221 845 252 752 252 345 367 157 377 385 202 172 153 293 218 229 151 187 1384 338 159 173 155 103 628 187 187 2012 489 494 112 112 97 691 305 430 510 209 209 496 198 594 407 143 271 245 240 417 858 540 175 1109 1109 121 129 511 259 356 173 135 299 141 315 250 226 157 148 179 156 131 142 754 398 406 188 240 562 336 332 332 155 405 336 336 K 10 svn:author V 6 jkoshy K 8 svn:date V 27 2008-09-19T13:17:22.594956Z K 7 svn:log V 80 Add an example illustrating the use sleep(1) for performing timed measurements. END K 10 svn:author V 6 jkoshy K 8 svn:date V 27 2008-09-19T13:21:23.728151Z K 7 svn:log V 30 Keep symbols in sorted order. END K 10 svn:author V 3 kib K 8 svn:date V 27 2008-09-19T13:32:17.530548Z K 7 svn:log V 71 MFC r182193 (libc part): Document WNOWAIT. Approved by: re (kensmith) END K 10 svn:author V 6 jkoshy K 8 svn:date V 27 2008-09-19T14:51:41.365799Z K 7 svn:log V 54 Bring the usage message in sync with the manual page. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2008-09-19T15:09:44.260773Z K 7 svn:log V 151 Cast bd_rcount, bd_dcount, and bd_fcount to intmax_t for printing until the base types for these are synced with 8-CURRENT. Approved by: re(kensmith) END K 10 svn:author V 6 obrien K 8 svn:date V 27 2008-09-19T15:17:32.781842Z K 7 svn:log V 93 Add freebsd32 compat shim for nmount(2). (and quiet some compiler warnings for vfs_donmount) END K 10 svn:author V 6 obrien K 8 svn:date V 27 2008-09-19T15:21:40.264256Z K 7 svn:log V 24 Regenerate for r183188. END K 10 svn:author V 4 jkim K 8 svn:date V 27 2008-09-19T15:25:13.981177Z K 7 svn:log V 127 Clean up DSDT and XSDT correctly instead of FADT when the tables are bad. Submitted by: jhb Tested by: olli MFC after: 3 days END K 10 svn:author V 7 thompsa K 8 svn:date V 27 2008-09-19T15:27:37.824396Z K 7 svn:log V 110 Change the wording to prefer 'forwarding' but still retain the word routing for clarity. Suggested by: dougb END K 10 svn:author V 4 emax K 8 svn:date V 27 2008-09-19T15:52:56.710476Z K 7 svn:log V 160 Move uuid_enc_le, uuid_dec_le, uuid_enc_be and uuid_dec_be symbols into FBSD_1.1 section where they should be. Pointed out by: kib, deischen MFC after: 3 days END K 10 svn:author V 6 obrien K 8 svn:date V 27 2008-09-19T16:14:42.424015Z K 7 svn:log V 30 Rescue is "not yet" for MIPS. END K 10 svn:author V 7 rnoland K 8 svn:date V 27 2008-09-19T19:11:35.395016Z K 7 svn:log V 254 pci_setup_intr() will only enable MSI/MSI-X for direct children. Add methods to vga_pci.c to request on behalf of it's children. This causes vgapci to show up as the interrupt owner in vmstat -i, rather than the child device. Approved by: jhb(mentor) END K 10 svn:author V 3 imp K 8 svn:date V 27 2008-09-19T19:20:05.721723Z K 7 svn:log V 101 crunchgen doesn't work yet on mips. So sysinstall can't be built there. omit it like we do for arm. END K 10 svn:author V 3 imp K 8 svn:date V 27 2008-09-19T19:26:38.379488Z K 7 svn:log V 357 mips_subr.c was in the Juniper code drop, but didn't have any functions used by other code in the tree. As such it was removed from the merged tree until the functions were needed in the future. The file was missing from the FreeBSD import, but it was listed in the files.mips file as being standard. Remove it from there until such time as we need one. END K 10 svn:author V 3 imp K 8 svn:date V 27 2008-09-19T19:46:11.053787Z K 7 svn:log V 54 Add atheros HAL building so the IDT kernel can build. END K 10 svn:author V 7 sobomax K 8 svn:date V 27 2008-09-19T19:49:58.372227Z K 7 svn:log V 340 Backout rev 183181. It appears that I should have been using boot-device of "cd:,\\:tbxi" with properly configured boot.tbxi, instead of booting \boot\loader directly. Rev 183168 could probably stay, since it can be viewed as an anti-foot-shooting measure and has no impact on normal operation. I can revert it as well, if anybody objects. END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2008-09-19T21:12:19.720153Z K 7 svn:log V 79 Fix issue with tom loading by moving cxgb_log_tcb in to tom MFC after: 3 days END K 10 svn:author V 3 zec K 8 svn:date V 27 2008-09-20T09:09:25.653980Z K 7 svn:log V 193 Move #defines for MRT-related constants from net/route.c to net/route.h, because the vnet code will need those constants as well. Reviewed by: bz Approved by: julian (mentor) MFC after: never END K 10 svn:author V 6 marius K 8 svn:date V 27 2008-09-20T11:26:13.362782Z K 7 svn:log V 445 Use the STICK timers only when absolutely necessary, i.e. if a machine consists of CPUs running at different speeds, for driving hardclock as these timers in turn are driven at frequencies as low as 5MHz, resulting in bad granularity compared to the TICK timers. However, don't employ the workaround for the BlackBird erratum #1 when using the TICK timer on machines with cheetah-class CPUs for performance reasons. Reported by: Florian Smeets END K 10 svn:author V 6 marius K 8 svn:date V 27 2008-09-20T11:28:42.382364Z K 7 svn:log V 273 Some of the assembly files depend on v9a/v9b-only instructions so compile these with -mcpu=ultrasparc (which is the hard-coded default of our system compiler), which allows the remainder of the kernel to be compiled with "only" -mcpu=v9 for reference and testing purposes. END K 10 svn:author V 6 marius K 8 svn:date V 27 2008-09-20T11:34:18.632233Z K 7 svn:log V 61 Remove duplicate entry accidentally introduced with r183202. END K 10 svn:author V 6 marius K 8 svn:date V 27 2008-09-20T11:46:07.041357Z K 7 svn:log V 69 Revert this part of r183203 which wasn't meant to be committed, yet. END K 10 svn:author V 3 rik K 8 svn:date V 27 2008-09-20T15:54:22.902295Z K 7 svn:log V 66 Do not do the useless job for an empty table. MFC after: 1 month END K 10 svn:author V 3 rik K 8 svn:date V 27 2008-09-20T16:17:49.046474Z K 7 svn:log V 34 Style(9) the show_nat() function. END K 10 svn:author V 3 alc K 8 svn:date V 27 2008-09-20T16:24:42.222948Z K 7 svn:log V 993 MFamd64 SVN rev 179749 CVS rev 1.620 Reverse the direction of pmap_promote_pde()'s traversal over the specified page table page. The direction of the traversal can matter if pmap_promote_pde() has to remove write access (PG_RW) from a PTE that hasn't been modified (PG_M). In general, if there are two or more such PTEs to choose among, it is better to write protect the one nearer the high end of the page table page rather than the low end. This is because most programs access memory in an ascending direction. The net result of this change is a sometimes significant reduction in the number of failed promotion attempts and the number of pages that are write protected by pmap_promote_pde(). MFamd64 SVN rev 179777 CVS rev 1.621 Tweak the promotion test in pmap_promote_pde(). Specifically, test PG_A before PG_M. This sometimes prevents unnecessary removal of write access from a PTE. Overall, the net result is fewer demotions and promotion failures. END K 10 svn:author V 3 rik K 8 svn:date V 27 2008-09-20T16:46:19.637988Z K 7 svn:log V 30 Remove some unused variables. END K 10 svn:author V 3 rik K 8 svn:date V 27 2008-09-20T19:25:02.477523Z K 7 svn:log V 23 Free allocated memory. END K 10 svn:author V 7 thompsa K 8 svn:date V 27 2008-09-20T19:38:37.910963Z K 7 svn:log V 169 Fix clone destruction, can't use the simple api because that does not remove the ifnet from cloner's list. Expose if_clone_destroyif api to do this. Submitted by: sam END K 10 svn:author V 3 kib K 8 svn:date V 27 2008-09-20T19:43:22.704399Z K 7 svn:log V 294 Initialize birthtime fields in vn_stat() to prevent stat(2) from returning uninitialized birthtime. Most file systems don't initialize birthtime properly in their VOP_GETTATTR(). Submitted by: Jaakko Heinonen Reviewed by: bde Discussed on: freebsd-fs MFC after: 1 month END K 10 svn:author V 3 kib K 8 svn:date V 27 2008-09-20T19:46:45.123402Z K 7 svn:log V 320 Initialize va_flags and va_filerev properly in VOP_GETATTR(). Don't initialize va_vaflags and va_spare because they are not part of the VOP_GETATTR() API. Also don't initialize birthtime to ctime or zero. Submitted by: Jaakko Heinonen Reviewed by: bde Discussed on: freebsd-fs MFC after: 1 month END K 10 svn:author V 3 kib K 8 svn:date V 27 2008-09-20T19:48:24.465774Z K 7 svn:log V 300 Initialize va_rdev to NODEV and va_fsid to VNOVAL before the VOP_GETATTR() call in vn_stat(). Thus if a file system doesn't initialize those fields in VOP_GETATTR() they will have a sane default value. Submitted by: Jaakko Heinonen Discussed on: freebsd-fs MFC after: 1 month END K 10 svn:author V 3 kib K 8 svn:date V 27 2008-09-20T19:49:15.961738Z K 7 svn:log V 259 Initialize va_rdev to NODEV instead of 0 or VNOVAL in VOP_GETATTR(). NODEV is more appropriate when va_rdev doesn't have a meaningful value. Submitted by: Jaakko Heinonen Suggested by: bde Discussed on: freebsd-fs MFC after: 1 month END K 10 svn:author V 3 kib K 8 svn:date V 27 2008-09-20T19:50:52.053758Z K 7 svn:log V 385 fdescfs, devfs, mqueuefs, nfs, portalfs, pseudofs, tmpfs and xfs initialize the vattr structure in VOP_GETATTR() with VATTR_NULL(), vattr_null() or by zeroing it. Remove these to allow preinitialization of fields work in vn_stat(). This is needed to get birthtime initialized correctly. Submitted by: Jaakko Heinonen Discussed on: freebsd-fs MFC after: 1 month END K 10 svn:author V 3 kib K 8 svn:date V 27 2008-09-20T19:56:02.245418Z K 7 svn:log V 179 Allow the d_mmap driver methods to use cdevpriv KPI during verification phase of establishing mapping. Discussed with: rwatson, jhb, rnoland Tested by: rnoland MFC after: 3 days END K 10 svn:author V 7 rodrigc K 8 svn:date V 27 2008-09-20T21:12:54.856557Z K 7 svn:log V 120 does not have a forward declration of getosreldate(), so remove it from this man page. MFC after: 3 days END K 10 svn:author V 4 kaiw K 8 svn:date V 27 2008-09-20T22:10:10.313223Z K 7 svn:log V 498 Add support for option "-M", which is used to operate ar(1) in a script mode like the MRI(Microtec Research Inc.) "librarian" program. Originally this option is provided by Binutils ar(1) to ease the transition for developers who are used to writing "librarian" scripts. We added this option to BSD ar(1) because: 1. Further improve the compatibility with Binutils ar(1). 2. There are still a few software using this -M option. (at least one in our ports collection) Suggested by: rink & erwin END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T00:26:13.951102Z K 7 svn:log V 60 eliminate hardwired lists; use the media type to autoconfig END K 10 svn:author V 4 kaiw K 8 svn:date V 27 2008-09-21T00:27:16.754030Z K 7 svn:log V 11 Fix build. END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T00:28:54.409809Z K 7 svn:log V 96 fix memory smash on lp64 platforms; mostly noticeable in user mode as being unable to associate END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T00:30:33.769281Z K 7 svn:log V 48 fix compilation on 64-bit platform w/ ATH_DEBUG END K 10 svn:author V 7 thompsa K 8 svn:date V 27 2008-09-21T06:01:03.189802Z K 7 svn:log V 131 MFC r183135 Make sure there is at least one port to avoid divide by zero when choosing the tx port. Approved by: re (kensmith) END K 10 svn:author V 7 thompsa K 8 svn:date V 27 2008-09-21T06:01:52.617095Z K 7 svn:log V 131 MFC r183135 Make sure there is at least one port to avoid divide by zero when choosing the tx port. Approved by: re (kensmith) END K 10 svn:author V 3 zec K 8 svn:date V 27 2008-09-21T07:33:33.365371Z K 7 svn:log V 75 Fix error message content. Approved by: julian (mentor) MFC after: 3 days END K 10 svn:author V 5 simon K 8 svn:date V 27 2008-09-21T11:02:23.337182Z K 7 svn:log V 42 Disable keyword expansion on vendor tree. END K 10 svn:author V 5 simon K 8 svn:date V 27 2008-09-21T12:37:21.320009Z K 7 svn:log V 69 Bootstrapping merge history from vendor-crypto/openssl/dist/@182044. END K 10 svn:author V 3 rik K 8 svn:date V 27 2008-09-21T12:54:09.940553Z K 7 svn:log V 40 Move table list to a separate function. END K 10 svn:author V 5 simon K 8 svn:date V 27 2008-09-21T13:18:25.098760Z K 7 svn:log V 56 The vendor area is the proper home for these files now. END K 10 svn:author V 2 ed K 8 svn:date V 27 2008-09-21T14:02:43.970161Z K 7 svn:log V 582 Already initialize the vfs timestamps inside the cdev upon allocation. In the MPSAFE TTY branch I noticed the vfs timestamps inside devfs were allocated with 0, where the getattr() routine bumps the timestamps to boottime if the value is below 3600. The reason why it has been designed like this, is because timestamps during boot are likely to be invalid. This means that device nodes that are created on demand (posix_openpt()) have timestamps with a value of boottime, which is not what we want. Solve this by calling vfs_timestamp() inside devfs_alloc(). Discussed with: kib END K 10 svn:author V 5 simon K 8 svn:date V 27 2008-09-21T14:12:30.083307Z K 7 svn:log V 205 Remove files from vendor tree which were not part of OpenSSL 0.9.8e (last vendor import). The file were removed in different earlier releases, but were not removed from the CVS vendor branch at the time. END K 10 svn:author V 5 simon K 8 svn:date V 27 2008-09-21T14:15:02.714342Z K 7 svn:log V 61 - Change FREEBSD-Xlist so it can be used as input to tar(1). END K 10 svn:author V 5 simon K 8 svn:date V 27 2008-09-21T14:30:38.416778Z K 7 svn:log V 169 - Remove files which aren't in the vendor distribution anymore. - Remove all of include as there is only the openssl subdir with symlinks (which were always removed). END K 10 svn:author V 5 simon K 8 svn:date V 27 2008-09-21T14:56:30.171319Z K 7 svn:log V 33 Vendor import of OpenSSL 0.9.8i. END K 10 svn:author V 7 antoine K 8 svn:date V 27 2008-09-21T16:11:11.650210Z K 7 svn:log V 24 Correct suser(9) entry. END K 10 svn:author V 3 alc K 8 svn:date V 27 2008-09-21T18:01:34.034652Z K 7 svn:log V 166 Prevent an integer overflow in vm_pageout_page_stats() on machines with a large number of physical pages. PR: 126158 Submitted by: Dmitry Tejblum MFC after: 3 days END K 10 svn:author V 5 simon K 8 svn:date V 27 2008-09-21T18:02:00.832974Z K 7 svn:log V 160 Include $NANO_NAME in the completed message. This is nice if you are building multiple NanoBSD images at once to keep track of what is running and what isn't. END K 10 svn:author V 2 ed K 8 svn:date V 27 2008-09-21T18:12:18.143194Z K 7 svn:log V 419 Now that the number of clist consumers have dropped massively, trim down the code to prevent useless waste of space. - Remove support for quote bits. There is not a single driver that needs these bits anymore. This means putc() now accepts a char instead of an int. - Remove the unneeded catq() and nextc() routines. They were only used by the old TTY layer. - Convert the clist code to use ANSI C prototypes. END K 10 svn:author V 2 ed K 8 svn:date V 27 2008-09-21T19:24:15.830461Z K 7 svn:log V 374 Make fstat() on a pseudo-terminal master return sane timestamps. Because pseudo-terminal master file descriptors no longer have a vnode underneath, we have to fill in fstat() values ourselves. Make our implementation somewhat sane by returning the timestamps of the TTY device node that corresponds with our file descriptor. Obtained from: //depot/projects/mpsafettty/... END K 10 svn:author V 3 rik K 8 svn:date V 27 2008-09-21T20:42:42.303005Z K 7 svn:log V 55 Export IPFW_TABLES_MAX value for compiled in defaults. END K 10 svn:author V 3 rik K 8 svn:date V 27 2008-09-21T21:46:56.458696Z K 7 svn:log V 35 Add the check of the table number. END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T22:02:26.916063Z K 7 svn:log V 235 add new build knobs and jigger some existing controls to improve control over the result of buildworld and installworld; this especially helps packaging systems such as nanobsd Reviewed by: various (posted to arch) MFC after: 1 month END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T22:09:18.981445Z K 7 svn:log V 45 o fix spelling of EOSP o correct SMPS values END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T22:16:03.078591Z K 7 svn:log V 32 check for undefined identifiers END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T22:22:28.252928Z K 7 svn:log V 226 Cleanup starting seq# for tx ampdu: o use txa_start to form the addba request and purge txa_seqstart o fill txa_start before calling ic_addba_request to permit drivers to override when they handle seq# generation (e.g. mwl) END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T22:24:07.009040Z K 7 svn:log V 224 Always update ic_lastnonht so we track overlapping legacy bss properly; was doing this only on a change in the protection mode so the advertised protection mode in the beacon would oscillate. Submitted by: Chris Zimmermann END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T23:00:19.115509Z K 7 svn:log V 1053 Cleanup AMPDU handling: For receive: o explicitly tag rx frames w/ M_AMPDU instead of passing frames through the reorder processing according to the node having HT and the frame being QoS data o relax ieee80211_ampdu_reorder asserts to allow any frame to be passed in, unsuitable frames are returned to the caller for normal processing; this permits drivers that cannot inspect the PLCP to mark all data frames as potential ampdu candidates with only a small penalty o add M_AMPDU_MPDU to identify frames resubmitted from the reorder q For transmit: o tag aggregation candidates with M_AMPDU_MPDU o fix the QoS ack policy set in ampdu subframes; we only support immediate BA streams which should be marked for "normal ack" to get implicit block ack behaviour; interestingly certain vendor parts BA'd frames with the 11e BA ack policy set o do not assign a sequence # to aggregation candidates; this must be done when frames are submitted for transmit (NB: this can/will be handled better when aggregation is pulled up to net80211) END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T23:16:19.177056Z K 7 svn:log V 657 Crypto api changes: o don't use the key index to identify when the driver has been asked to allocate a key slot, use an explicit flag; allows drivers to force s/w fallback for entries in the global table o change callback api to allocate driver resources for a crypto key: - de-const the key parameter so drivers can muck with the flags - on callback failure don't automatically try to setup s/w crypto; instead the driver must now mark the key entry for s/w crypto and the caller will re-attach the cipher module NB: api change permits drivers more control over fallback to s/w crypto (e.g. based on a limited number of h/w key slots) END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T23:18:36.668243Z K 7 svn:log V 19 simplify debug msg END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T23:20:04.870010Z K 7 svn:log V 60 change ieee80211_ampdu_stop to take an explicit reason code END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T23:25:33.994450Z K 7 svn:log V 515 Add current transmit parameters for fixed rate handling so drivers don't duplicate this. These are setup according to the role of the node--the bss node for ap and adhoc modes need to use parameters that are the least common denomimator of all nodes in the bss; otherwise we are setting up params for a station joining a bss and we select those according to the capabilities of the station. This stuff needs more work as we do extra work due to having setup in common code paths shared by nodes using both roles. END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T23:31:17.358540Z K 7 svn:log V 86 o widen ni_flags as it's going to be full shortly o shuffle members to minimize holes END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T23:32:48.759130Z K 7 svn:log V 96 don't propagate ht bss occupancy changes to legacy vaps; they won't have an htinfo ie to update END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T23:44:15.517636Z K 7 svn:log V 659 Revamp ht ie handling: o change ieee80211_parse_htcap and ieee80211_parse_htinfo to save only internal state obtained from the ie's; no dynamic state such as ni_chw is altered o add ieee80211_ht_updateparams to parse ht cap+info ie's and update dynamic node state o change ieee80211_ht_node_init to not take an htcap ie that is parsed; instead have the caller make a separate call as one caller wants to parse the ie while another wants to parse both cap+info ie's and update state so can better do this with ieee80211_ht_updateparams These changes fix sta mode state handling where the node's channel width was shifted to ht20/ht40 prematurely. END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-21T23:59:14.358758Z K 7 svn:log V 110 MIMO power save support; still needs callbacks for notifying drivers of dynamic state change in station mode. END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-22T00:10:22.289841Z K 7 svn:log V 61 RIFS support; needs driver callback for dynamic state change END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-22T00:14:50.416563Z K 7 svn:log V 199 Fix handling of shortgi: use the local configuration (and implicitly device capabilities) to decide whether to use short gi. Drivers inspect ni_flags to decide whether to send a frame w/ short sgi. END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-22T00:21:24.937586Z K 7 svn:log V 20 cleanup after merge END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-22T00:22:09.345779Z K 7 svn:log V 72 cleanup bss node setup; syncs w/ p4 state that's been in use for months END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-22T00:24:47.262192Z K 7 svn:log V 78 "ampdudensity -" is too subtle; use "NA" for any and accept it as a parameter END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-22T00:30:26.833664Z K 7 svn:log V 80 MIMO power save and RIFS; while here also update per-node state/flags for ampdu END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2008-09-22T03:21:02.415250Z K 7 svn:log V 123 Unbreak G3 support. G3 processors don't have an L3 cache, so we shouldn't try to program it. Approved by: marcel (mentor) END K 10 svn:author V 8 keramida K 8 svn:date V 27 2008-09-22T04:12:27.993355Z K 7 svn:log V 19 Unbreak the build. END K 10 svn:author V 7 yongari K 8 svn:date V 27 2008-09-22T06:17:21.035899Z K 7 svn:log V 235 Add workaround for occasional packet loss issue of JMC250 A2 when it runs on half-duplex media. While I'm here add register definition for GPREG1. ATM the GPREG1 register is only valid for JMC250 A1/A2. Submitted by: Ethan at JMicron END K 10 svn:author V 7 rwatson K 8 svn:date V 27 2008-09-22T06:44:03.536377Z K 7 svn:log V 366 When invoking the udp_send() from udp6_send() due to use of a v6-mapped IPv4 address, first drop the udbinfo and inpcb locks, which will otherwise be recursed. This leads to a potential minor race, but is preferable to a deadlock when acquiring a read lock after a write lock on the inpcb. MFC after: 3 days Reported by: Norbert Papke , lioux END K 10 svn:author V 6 jkoshy K 8 svn:date V 27 2008-09-22T10:37:02.681402Z K 7 svn:log V 73 Support sparsely numbered CPUs. Requested by: obrien, alfred (long ago) END K 10 svn:author V 6 jkoshy K 8 svn:date V 27 2008-09-22T14:31:24.721849Z K 7 svn:log V 67 Document changes in behaviour due to sparse CPU numbering support. END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-22T15:37:47.218344Z K 7 svn:log V 41 add back regdomain.xml Noticed by: jhay END K 10 svn:author V 4 jkim K 8 svn:date V 27 2008-09-22T15:57:58.948911Z K 7 svn:log V 129 MFC: r183190 Clean up DSDT and XSDT correctly instead of FADT when the tables are bad. Submitted by: jhb Approved by: re (kib) END K 10 svn:author V 6 obrien K 8 svn:date V 27 2008-09-22T16:09:16.477090Z K 7 svn:log V 102 Add freebsd32 compat shims for ioctl(2) MDIOCATTACH, MDIOCDETACH, MDIOCQUERY, and MDIOCLIST requests. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2008-09-22T16:09:43.824114Z K 7 svn:log V 24 Regenerate for r183270. END K 10 svn:author V 2 ed K 8 svn:date V 27 2008-09-22T16:10:12.221090Z K 7 svn:log V 421 Make the make_pseudo_driver.sh shellscript work again. It seems this script was broken because of the SYSINIT changes and a rather awkward variable initialisation. For some reason the make_device_driver.sh script is also broken, related to BUS_SETUP_INTR. I have no experience with FreeBSD's interrupt handling, so I hope someone else is willing to take a look at that shell script. PR: misc/126435 MFC after: 1 month END K 10 svn:author V 6 obrien K 8 svn:date V 27 2008-09-22T16:24:36.448597Z K 7 svn:log V 90 Add freebsd32 compat shims for ioctl(2) CDIOREADTOCHEADER and CDIOREADTOCENTRYS requests. END K 10 svn:author V 2 ed K 8 svn:date V 27 2008-09-22T18:44:09.345297Z K 7 svn:log V 497 Fix style(9) issue in TTY header files: document function argument names. According to style(9), function argument names should only be omitted for prototypes that are exported to userspace. This means we should document the function arguments in the TTY header files, because they are only used in userspace. While there, change the type of the buffer argument of ttydisc_rint_bypass() to `const void *' instead of `char *'. Requested by: attilio Obtained from: //depot/projects/mpsafetty/... END K 10 svn:author V 5 trasz K 8 svn:date V 27 2008-09-22T18:59:24.071823Z K 7 svn:log V 126 Fix usage of mac_vnode_check_open() in linuxulator - last argument should be VREAD, not FREAD. Approved by: rwatson (mentor) END K 10 svn:author V 2 ed K 8 svn:date V 27 2008-09-22T19:25:14.599157Z K 7 svn:log V 753 Introduce a hooks layer for the MPSAFE TTY layer. One of the features that prevented us from fixing some of the TTY consumers to work once again, was an interface that allowed consumers to do the following: - `Sniff' incoming data, which is used by the snp(4) driver. - Take direct control of the input and output paths of a TTY, which is used by ng_tty(4), ppp(4), sl(4), etc. There's no practical advantage in committing a hooks layer without having any consumers. In P4 there is a preliminary port of snp(4) and thompsa@ is busy porting ng_tty(4) to this interface. I already want to have it in the tree, because this may stimulate others to work on the remaining modules. Discussed with: thompsa Obtained from: //depot/projects/mpsafetty/... END K 10 svn:author V 7 thompsa K 8 svn:date V 27 2008-09-22T19:28:18.550059Z K 7 svn:log V 155 Add ng_rmnode_flags() so the caller can pass NG_QUEUE and have the node destroyed asynchronously due to locking or other constraints. Reviewed by: julian END K 10 svn:author V 8 rdivacky K 8 svn:date V 27 2008-09-22T20:19:54.052497Z K 7 svn:log V 654 Merge r182892 from head to stable/7, I had to manually change the code to include "thread" argument to the vn_lock() which got removed in HEAD: Getdents requires padding with 2 bytes instead of 1 byte as with getdents64. The last byte is used for storing the d_type, add this to plain getdents case where it was missing before. Also change the code to use strlcpy instead of plain strcpy. This changes fix the getdents crash we had reports about (hl2 server etc.) PR: kern/117010 MFC after: 1 week Submitted by: Dmitry Chagin (dchagin@) Tested by: MITA Yoshio Approved by: kib (mentor) Approved by: re (kensmith) END K 10 svn:author V 3 bms K 8 svn:date V 27 2008-09-22T20:21:39.761809Z K 7 svn:log V 159 Mark the first slice of a NanoBSD image 'active' by default. This fixes USB boot (not yet merged to HEAD) with 3 flavours of BIOS I've seen. Approved by: phk END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-22T20:53:22.068776Z K 7 svn:log V 252 Close a race between concurrent calls to ufsdirhash_recycle() and ufsdirhash_free() introduced in my last commit by removing the dirhash about to be free'd in ufsdirhash_free() from the global dirhash list before dropping the sx lock. Tested by: kris END K 10 svn:author V 6 marius K 8 svn:date V 27 2008-09-22T20:55:19.225726Z K 7 svn:log V 271 - Don't read the identifier string from the VPD if there's no chip containing an Ethernet address fitted as this is yet another thing that fails in that case in order to avoid the one second delay until pci_read_vpd_reg() times out. - Const'ify the bge_devs array. END K 10 svn:author V 7 rnoland K 8 svn:date V 27 2008-09-22T21:53:00.333647Z K 7 svn:log V 61 Add my entry to committers-src.dot Approved by: jhb(mentor) END K 10 svn:author V 4 emax K 8 svn:date V 27 2008-09-22T22:08:43.856485Z K 7 svn:log V 283 Use Giant for kbdmux(4) locking. This is to workaround the problem where interrupt handlers may race with kbdmux(4) in polling mode. PR: kern/127446 Reported by: Eygene Ryabinkin rea-fbsd at codelabs dot ru Tested by: Eygene Ryabinkin rea-fbsd at codelabs dot ru MFC after: 1 week END K 10 svn:author V 3 bms K 8 svn:date V 27 2008-09-22T23:56:36.576991Z K 7 svn:log V 292 * Add USB boot support. * Allow the image name to be renamed via NANO_IMGNAME. * Propagate TARGET_ARCH into src top level make targets explicitly to support cross-building. * Increase the default size of NanoBSD media from 488MB to 584MB to accomodate a -CURRENT world. Reviewed by: phk END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2008-09-23T01:27:19.647908Z K 7 svn:log V 107 - Track number of times that the transmit queue overflowed - Trivial whitespace cleanup MFC after: 3 days END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2008-09-23T01:55:36.562752Z K 7 svn:log V 78 - Fix flag check - Fix adaptive thread sleep - set oactive when queue is full END K 10 svn:author V 6 marcel K 8 svn:date V 27 2008-09-23T02:09:00.771498Z K 7 svn:log V 58 Nathan Whitehorn (nwhitehorn) is flying solo from now on. END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2008-09-23T02:12:47.183662Z K 7 svn:log V 192 Change the DBDMA API to allow DBDMA registers in a subregion of a resource. This is necessary to allow future support of DMA for the various Apple on-board ATA controllers. MFC after: 1 week END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2008-09-23T02:22:24.388809Z K 7 svn:log V 123 Allow cxgb to be unified across versions by making newer features conditional Submitted by: Chelsio Inc MFC after: 3 days END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2008-09-23T03:02:57.799010Z K 7 svn:log V 128 In preparation for PowerPC G5 support, allow PVO objects to contain page table entries for both the 32-bit and 64-bit AIM MMUs. END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2008-09-23T03:13:33.309461Z K 7 svn:log V 57 Update to xen specific files for SMP MFC after: 1 month END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2008-09-23T03:16:54.695297Z K 7 svn:log V 93 Update cxgb include paths to not require prefixing with dev/cxgb Submitted by: Chelsio Inc. END K 10 svn:author V 3 rik K 8 svn:date V 27 2008-09-23T10:36:37.876406Z K 7 svn:log V 1290 MFH: 182818, 182823, 182825, 183012 - IPFW_DEFAULT_RULE related fixes. New Revision: 182818 URL: http://svn.freebsd.org/changeset/base/182818 Log: Export the IPFW_DEFAULT_RULE outside ip_fw2.c. This number in not only the default rule number but also the maximum rule number. User space software such as ipfw and natd should be aware of its value. The software that already includes ip_fw.h should use the defined value. All other a expected to use sysctl (as discussed on net@). MFC after: 5 days. Discussed on: net@ Modified: head/sys/netinet/ip_fw.h head/sys/netinet/ip_fw2.c New Revision: 182823 URL: http://svn.freebsd.org/changeset/base/182823 Log: Use IPFW_DEFAULT_RULE instead of hardcoded value since now it is available. MFC after: 5 days. Modified: head/sbin/ipfw/ipfw2.c New Revision: 182825 URL: http://svn.freebsd.org/changeset/base/182825 Log: Check rule numbers against maximum value to avoid rules cleanup due to overflow. MFC after: 5 days. Modified: head/sbin/natd/natd.c New Revision: 183012 URL: http://svn.freebsd.org/changeset/base/183012 Log: Make the commet for the default rule number more clear. Submitted by: yar@ Modified: head/sys/netinet/ip_fw.h Approved by: re (kensmith) END K 10 svn:author V 3 kib K 8 svn:date V 27 2008-09-23T11:56:07.548592Z K 7 svn:log V 245 MFC r180607 (by jeff): Check whether we've recorded this tick in ts_ticks on another cpu in sched_tick() to prevent multiple increments for one tick. This pushes the value out of range and breaks priority calculation. Approved by: re (rwatson) END K 10 svn:author V 6 matteo K 8 svn:date V 27 2008-09-23T12:17:20.653775Z K 7 svn:log V 64 MFC rev 1.5 (r182853) Use %d to print numfibs Approved by: re END K 10 svn:author V 2 ru K 8 svn:date V 27 2008-09-23T13:12:51.866151Z K 7 svn:log V 82 Add missing library dependencies. PR: bin/127573 Submitted by: Eygene Ryabinkin END K 10 svn:author V 6 obrien K 8 svn:date V 27 2008-09-23T14:25:38.437707Z K 7 svn:log V 60 Reverse if() logic to improve readability. Reviewed by: ru END K 10 svn:author V 6 obrien K 8 svn:date V 27 2008-09-23T14:25:56.682853Z K 7 svn:log V 9 style(9) END K 10 svn:author V 6 obrien K 8 svn:date V 27 2008-09-23T14:45:10.918478Z K 7 svn:log V 532 The kernel implemented 'memcmp' is an alias for 'bcmp'. However, memcmp and bcmp are not the same thing. 'man bcmp' states that the return is "non-zero" if the two byte strings are not identical. Where as, 'man memcmp' states that the return is the "difference between the first two differing bytes (treated as unsigned char values" if the two byte strings are not identical. So provide a proper memcmp(9), but it is a C implementation not a tuned assembly implementation. Therefore bcmp(9) should be preferred over memcmp(9). END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-23T15:57:39.769389Z K 7 svn:log V 95 MFC: Fix vnode interlocking in QUOTA lock upgrade case in ufs_access(). Approved by: re (kib) END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-23T15:58:41.226857Z K 7 svn:log V 95 MFC: Fix vnode interlocking in QUOTA lock upgrade case in ufs_access(). Approved by: re (kib) END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-23T15:59:38.011218Z K 7 svn:log V 1918 MFC: 179623 Essentially, neither madvise(..., MADV_DONTNEED) nor madvise(..., MADV_FREE) work. (Moreover, I don't believe that they have ever worked as intended.) The explanation is fairly simple. Both MADV_DONTNEED and MADV_FREE perform vm_page_dontneed() on each page within the range given to madvise(). This function moves the page to the inactive queue. Specifically, if the page is clean, it is moved to the head of the inactive queue where it is first in line for processing by the page daemon. On the other hand, if it is dirty, it is placed at the tail. Let's further examine the case in which the page is clean. Recall that the page is at the head of the line for processing by the page daemon. The expectation of vm_page_dontneed()'s author was that the page would be transferred from the inactive queue to the cache queue by the page daemon. (Once the page is in the cache queue, it is, in effect, free, that is, it can be reallocated to a new vm object by vm_page_alloc() if it isn't reactivated quickly enough by a user of the old vm object.) The trouble is that nowhere in the execution of either MADV_DONTNEED or MADV_FREE is either the machine-independent reference flag (PG_REFERENCED) or the reference bit in any page table entry (PTE) mapping the page cleared. Consequently, the immediate reaction of the page daemon is to reactivate the page because it is referenced. In effect, the madvise() was for naught. The case in which the page was dirty is not too different. Instead of being laundered, the page is reactivated. Note: The essential difference between MADV_DONTNEED and MADV_FREE is that MADV_FREE clears a page's dirty field. So, MADV_FREE is always executing the clean case above. This revision changes vm_page_dontneed() to clear both the machine- independent reference flag (PG_REFERENCED) and the reference bit in all PTEs mapping the page. Approved by: re (kib), alc END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-23T16:03:32.021735Z K 7 svn:log V 396 MFC: 179296 To date, our implementation of munmap(2) has required that the entirety of the specified range be mapped. Specifically, it has returned EINVAL if the entire range is not mapped. There is not, however, any basis for this in either SuSv2 or our own man page. Moreover, neither Linux nor Solaris impose this requirement. This revision removes this requirement. Approved by: re (kib) END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-23T16:04:17.775076Z K 7 svn:log V 401 MFC: 179296 To date, our implementation of munmap(2) has required that the entirety of the specified range be mapped. Specifically, it has returned EINVAL if the entire range is not mapped. There is not, however, any basis for this in either SuSv2 or our own man page. Moreover, neither Linux nor Solaris impose this requirement. This revision removes this requirement. Approved by: re (kib), alc END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-23T16:11:15.508356Z K 7 svn:log V 20 add missing options END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-23T16:15:42.373169Z K 7 svn:log V 20 add missing options END K 10 svn:author V 3 sam K 8 svn:date V 27 2008-09-23T16:18:15.848672Z K 7 svn:log V 6 regen END K 10 svn:author V 2 ed K 8 svn:date V 27 2008-09-23T17:12:25.492765Z K 7 svn:log V 599 Track state to determine if the associated TTY device node has been used. It turns out our old TTY layer (and other implementations) block when you read() on a PTY master device of which the slave device node has not been opened yet. Our new implementation just returned 0. This caused applications like telnetd to die in a very subtle way (when child processes would open the TTY later than the first call to select()). Introduce a new flag called PTS_FINISHED, which indicates whether we should block or bail out of a read() or write() occurs. Reported by: Claude Buisson END K 10 svn:author V 3 kib K 8 svn:date V 27 2008-09-23T17:16:24.319154Z K 7 svn:log V 212 MFC r182959: Remove warning about static LDT segment allocation. Applications continue using it after ~7 years since warning was introduced, and there is no reason to discourage them. Approved by: re (kensmith) END K 10 svn:author V 3 kib K 8 svn:date V 27 2008-09-23T17:23:59.840490Z K 7 svn:log V 337 MFC r182960 (with XEN-related chunks removed): The user_ldt_alloc() function shall return with dt_lock locked. The user_ldt_free() function shall return with dt_lock unlocked. Error handling code in both functions do not handle this, fix it by doing necessary lock/unlock. While there, fix minor style nits. Approved by: re (kensmith) END K 10 svn:author V 3 kib K 8 svn:date V 27 2008-09-23T17:41:15.368291Z K 7 svn:log V 417 MFC r182961: When doing rfork(0), i.e. separating curproc VM from any other user of the same vmspace, decrement the reference count of the shared LDT instead of a newly-made copy. Code factually removed LDT from the process that did rfork(0). Introduce user_ldt_deref() function that does decrement of refcount for the struct proc_ldt, and call it in the rfork(0) case on the shared LDT. Approved by: re (kensmith) END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-23T18:33:11.956372Z K 7 svn:log V 116 MFC: Add a CTASSERT() that KERNBASE is valid and update the KVA_PAGES comments to cover PAE. Approved by: re (kib) END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-23T18:33:36.263292Z K 7 svn:log V 116 MFC: Add a CTASSERT() that KERNBASE is valid and update the KVA_PAGES comments to cover PAE. Approved by: re (kib) END K 10 svn:author V 5 simon K 8 svn:date V 27 2008-09-23T18:42:35.503631Z K 7 svn:log V 401 - Change all "echo #..." into using a progress print function to make it possible to make NanoBSD output more quite or verbose. The default output should remain mostly unchanged. [1] - Add missing shift for -i. - Clean up usage() so it's now (mostly) sorted alphabetically. - Make command line argument handling more consistent in the code and remove redundant semicolons. Reviwed by: phk [1] END K 10 svn:author V 5 simon K 8 svn:date V 27 2008-09-23T18:54:56.101345Z K 7 svn:log V 103 In cust_install_files() we ignore CVS directories. In a similar way now also ignore .svn directories. END K 10 svn:author V 7 rwatson K 8 svn:date V 27 2008-09-23T21:29:22.093894Z K 7 svn:log V 497 Merge r183265 from head to stable/7: When invoking the udp_send() from udp6_send() due to use of a v6-mapped IPv4 address, first drop the udbinfo and inpcb locks, which will otherwise be recursed. This leads to a potential minor race, but is preferable to a deadlock when acquiring a read lock after a write lock on the inpcb. MFC after: 3 days Reported by: Norbert Papke , lioux Approved by: re (kib) Tested by: Norbert Papke , lioux END K 10 svn:author V 7 sobomax K 8 svn:date V 27 2008-09-23T22:16:23.067118Z K 7 svn:log V 310 Improve rev 183168, so that if /chosen/stdout is connected to the serial port by OF the syscons won't take over console. Only attach syscons to "screen" if /chosen/stdout is not connected, which could be the case when loader(8) is booted directly from the OF. This fixes Marcel's Xserver. Reported by: marcel END K 10 svn:author V 7 delphij K 8 svn:date V 27 2008-09-24T00:04:51.714685Z K 7 svn:log V 47 Remove spurious duplicated defination of sock. END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2008-09-24T00:28:46.647235Z K 7 svn:log V 170 Allow the cacheline size on PowerPC to be set at runtime. This is essential for supporting 64-bit CPUs, which often have 128-byte cache lines instead of the standard 32. END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2008-09-24T01:06:11.692634Z K 7 svn:log V 144 Fix some ifdef logic in the libc syncicache() bit that I somehow missed in the last commit. Now it really can use a runtime-set cacheline size. END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2008-09-24T01:19:08.334730Z K 7 svn:log V 145 - Remove default NIC dependency on ulp headers - make toe module build dependent on kernel support Submitted by: Chelsio Inc. MFC after: 1 week END K 10 svn:author V 3 kib K 8 svn:date V 27 2008-09-24T10:14:37.973067Z K 7 svn:log V 324 Change the static struct sysentvec and struct Elf_Brandinfo initializers to the C99 style. At least, it is easier to read sysent definitions that way, and search for the actual instances of sigcode etc. Explicitely initialize sysentvec.sv_maxssiz that was missed in most sysvecs. No objection from: jhb MFC after: 1 month END K 10 svn:author V 7 rwatson K 8 svn:date V 27 2008-09-24T11:07:03.112732Z K 7 svn:log V 761 When dropping a packet and issuing a reset during TCP segment handling, unconditionally drop the tcbinfo lock (after all, we assert it lines before), but call tcp_dropwithreset() under both inpcb and inpcbinfo locks only if we pass in an tcpcb. Otherwise, if the pointer is NULL, firewall code may later recurse the global tcbinfo lock trying to look up an inpcb. This is an instance where a layering violation leads not only potentially to code reentrace and recursion, but also to lock recursion, and was revealed by the conversion to rwlocks because acquiring a read lock on an rwlock already held with a write lock is forbidden. When these locks were mutexes, they simply recursed. Reported by: Stefan Ehmann MFC after: 3 days END K 10 svn:author V 2 ed K 8 svn:date V 27 2008-09-24T11:16:09.802531Z K 7 svn:log V 448 Fix a crash when calling tty_rel_free() while draining during closure. Yesterday I got two reports of potential crashes, related to TTY deallocation during device closure. When a thread is in TF_OPENCLOSE, draining its output upon closure, we should not allow calls to tty_rel_free() to happen at the same time. This could cause the TTY to be torn down twice. PR: kern/127561 Reported by: KOIE Hidetaka Discussed with: thompsa END K 10 svn:author V 2 ru K 8 svn:date V 27 2008-09-24T15:18:27.112468Z K 7 svn:log V 84 Allow a jail's IP alias to be created with an arbitrary netmask. MFC after: 3 days END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-24T17:49:26.148164Z K 7 svn:log V 1015 MFC: 176526,176596 - Correct a long-standing error in vm_object_deallocate(). Specifically, only anonymous default (OBJT_DEFAULT) and swap (OBJT_SWAP) objects should ever have OBJ_ONEMAPPING set. However, vm_object_deallocate() was setting it on device (OBJT_DEVICE) objects. As a result, vm_object_page_remove() could be called on a device object and if that occurred pmap_remove_all() would be called on the device object's pages. However, a device object's pages are fictitious, and fictitious pages do not have an initialized pv list (struct md_page). - Correct a long-standing error in vm_object_page_remove(). Specifically, pmap_remove_all() must not be called on fictitious pages. To date, fictitious pages have been allocated from zeroed memory, effectively hiding this problem because the fictitious pages appear to have an empty pv list. - Rewrite the comments describing vm_object_page_remove() to better describe what it does. Add an assertion. Approved by: re (kib), alc END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-24T17:49:54.807832Z K 7 svn:log V 1015 MFC: 176526,176596 - Correct a long-standing error in vm_object_deallocate(). Specifically, only anonymous default (OBJT_DEFAULT) and swap (OBJT_SWAP) objects should ever have OBJ_ONEMAPPING set. However, vm_object_deallocate() was setting it on device (OBJT_DEVICE) objects. As a result, vm_object_page_remove() could be called on a device object and if that occurred pmap_remove_all() would be called on the device object's pages. However, a device object's pages are fictitious, and fictitious pages do not have an initialized pv list (struct md_page). - Correct a long-standing error in vm_object_page_remove(). Specifically, pmap_remove_all() must not be called on fictitious pages. To date, fictitious pages have been allocated from zeroed memory, effectively hiding this problem because the fictitious pages appear to have an empty pv list. - Rewrite the comments describing vm_object_page_remove() to better describe what it does. Add an assertion. Approved by: re (kib), alc END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-24T18:46:39.226019Z K 7 svn:log V 29 Add multiple include guards. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-24T18:47:24.851639Z K 7 svn:log V 37 Update description of witness_watch. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-24T18:51:33.787392Z K 7 svn:log V 418 Part 1 of making shared lookups more resilient with respect to forced unmounts. When we upgrade a vnode lock from shared to exclusive during a name cache lookup, fail the lookup with EBADF if the vnode is invalidated while we are waiting for the exclusive lock. Also, for correctness (though I'm not sure it can occur in practice), downgrade an exclusively locked vnode if it should be share locked. Tested by: pho END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-24T18:53:04.164571Z K 7 svn:log V 166 Enable shared lookups on UFS. There are some remaining issues with forced unmounts, but those are in the VFS lookup code are not UFS specific. Tested by: pho, kris END K 10 svn:author V 2 ed K 8 svn:date V 27 2008-09-24T19:43:09.847591Z K 7 svn:log V 264 Add another TTY locking migration macro: tty_getlock(). Some subsystems (HPS-USB) like to lock down the TTY through a more generic approach, namely the regular mtx(9) macro's. Allow the TTY lock to be obtained through the new macro. Discussed with: hps, thompsa END K 10 svn:author V 6 marius K 8 svn:date V 27 2008-09-24T20:41:35.461429Z K 7 svn:log V 78 Add back ar(1) which was accidentally decoupled in r183242. Approved by: sam END K 10 svn:author V 3 des K 8 svn:date V 27 2008-09-24T21:07:11.460046Z K 7 svn:log V 43 Undocument dead option. MFC after: 3 days END K 10 svn:author V 3 des K 8 svn:date V 27 2008-09-24T21:17:14.260682Z K 7 svn:log V 206 Fix "xmalloc: zero size" bug when talking to older servers. This patch has been submitted but not yet committed upstream. Obtained from: https://bugzilla.mindrot.org/show_bug.cgi?id=1496 MFC after: 1 week END K 10 svn:author V 3 des K 8 svn:date V 27 2008-09-24T21:20:44.856785Z K 7 svn:log V 49 MFV "xmalloc: zero size" fix. MFC after: 1 week END K 10 svn:author V 6 marius K 8 svn:date V 27 2008-09-24T21:26:46.265355Z K 7 svn:log V 219 - Use bus_{read,write}_*(9) instead of bus_space_{read,write}_*(9) etc and take advantage of rman_get_rid(9) in order to save some softc members. - Provide and consume module dependency information for lebuffer(4). END K 10 svn:author V 3 alc K 8 svn:date V 27 2008-09-25T06:11:58.651385Z K 7 svn:log V 157 MFC r183236 Prevent an integer overflow in vm_pageout_page_stats() on machines with a large number of physical pages. PR: 126158 Approved by: re (kib) END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2008-09-25T06:46:28.529104Z K 7 svn:log V 131 fix insta-panic: - determine which ext_arg offsets to use based on the version number Submitted by: Chelsio Inc. MFC after: 1 day END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2008-09-25T07:01:31.737579Z K 7 svn:log V 63 partial update to interface headers to 3.2 MFC after: 1 month END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2008-09-25T07:02:00.042169Z K 7 svn:log V 54 reflect header change in netfront MFC after: 1 month END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2008-09-25T07:03:09.997613Z K 7 svn:log V 85 Make nkpt dependent on the size of the initial memory allocation MFC after: 1 month END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2008-09-25T07:05:17.698370Z K 7 svn:log V 62 add NPGPTD_SHIFT for the nkpt calculation MFC after: 1 month END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2008-09-25T07:09:50.466648Z K 7 svn:log V 37 add ipi mapping MFC after: 1 month END K 10 svn:author V 5 kmacy K 8 svn:date V 27 2008-09-25T07:11:04.089187Z K 7 svn:log V 48 add initial ipi definitions MFC after: 1 month END K 10 svn:author V 7 yongari K 8 svn:date V 27 2008-09-25T07:31:45.783482Z K 7 svn:log V 657 Add another hardware bug workaround for Yukon II controllers that have hardware ram buffer. The silicon bug seem to be triggered by pause frames if receive buffer is not aligned on FIFO word(8 bytes). To workaround the issue, make sure to align Rx buffers on 8 bytes. Unfortunately this workaround requires yet another Rx fixup for strict alignment architecture machines to align IP header. For newer hardwares that lacks ram buffer may not have this bug so check number of available ram buffer size to see the existence of ram buffer. Reported by: Ian Freislich (ianf clue dot co dot za), das Tested by: Ian Freislich (ianf clue dot co dot za) END K 10 svn:author V 8 dwmalone K 8 svn:date V 27 2008-09-25T09:28:18.712369Z K 7 svn:log V 300 Add a flag, -T, that tells syslogd to always replace the timestamp on messages from the network. We already replace malformatted timestamps and this option lets us replace timestamps that are correctly formatted but wrong. PR: 120891 Submitted by: Thomas Vogt MFC after: 1 week END K 10 svn:author V 3 phk K 8 svn:date V 27 2008-09-25T12:00:56.494228Z K 7 svn:log V 313 Make the ubsa(4) work with Huawei Exxx (tested with E169) 3G radio devices: Kick the device into the right mode if it comes up as a flash-disk. Set the buffers to a sensible 1024 bytes instead of a far too small default. Don't attempt to change speed, baud, parity and such, the device does not understand it. END K 10 svn:author V 3 phk K 8 svn:date V 27 2008-09-25T12:16:18.592238Z K 7 svn:log V 96 Tell umass to leave Huawei radio devices alone so ubsa(4) will be able to kick them into shape. END K 10 svn:author V 4 nyan K 8 svn:date V 27 2008-09-25T12:25:54.270931Z K 7 svn:log V 146 - Add libbsdxml and libsbuf required by bsdlabel. - Remove the mknod from pc98 fixit and also the ftp from pc98 small fixit to avoid disk full. END K 10 svn:author V 8 dwmalone K 8 svn:date V 27 2008-09-25T12:35:01.958600Z K 7 svn:log V 464 Some people's 6to4 routers seem to have been blowing up because of the unlocked route caching in if_stf. Add a mutex that protects access to cached route. This seemed to fix problems for Pekka Savola. Nick Sayer had similar problems, and in his case completly disabling the route cache seemed to help. Add a sysctl net.link.stf.route_cache that can be used to turn off route caching in if_stf. PR: 122283 MFC after: 2 weeks Tested by: Pekka Savola, Nick Sayer. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-25T13:42:19.186658Z K 7 svn:log V 243 - Don't do a WITNESS_SAVE() on the interlock if it is Giant in the condition variable wait routines. DROP_GIANT() already manages that state in the Giant interlock case. - Assert that Giant is held when it is passed as a sleep interlock. END K 10 svn:author V 8 dwmalone K 8 svn:date V 27 2008-09-25T14:24:07.596679Z K 7 svn:log V 234 MFC: Lockf was exiting with status 1 if the command did not exit normally. This is easy to confuse with the actual exit status of the program. Instead exit with EX_SOFTWARE if the command doesn't exit normally. Approved by: re (kib) END K 10 svn:author V 8 dwmalone K 8 svn:date V 27 2008-09-25T14:27:00.716264Z K 7 svn:log V 234 MFC: Lockf was exiting with status 1 if the command did not exit normally. This is easy to confuse with the actual exit status of the program. Instead exit with EX_SOFTWARE if the command doesn't exit normally. Approved by: re (kib) END K 10 svn:author V 7 thompsa K 8 svn:date V 27 2008-09-25T16:46:36.619451Z K 7 svn:log V 59 Use db_printsym to lookup and print the function pointers. END K 10 svn:author V 7 rwatson K 8 svn:date V 27 2008-09-25T17:26:54.306481Z K 7 svn:log V 308 As a follow-on to r183323, correct another case where ip_output() was called without an inpcb pointer despite holding the tcbinfo global lock, which lead to a deadlock or panic when ipfw tried to further acquire it recursively. Reported by: Stefan Ehmann MFC after: 3 days END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-25T18:37:08.953848Z K 7 svn:log V 243 MFC: 178280 - Clean up the code that checks the types of address so that it is done by understandable macros. - Fix the bug that prevented the system from responding on interfaces with link local addresses assigned. Approved by: re (gnn) END K 10 svn:author V 3 jhb K 8 svn:date V 27 2008-09-25T18:38:04.949172Z K 7 svn:log V 243 MFC: 178280 - Clean up the code that checks the types of address so that it is done by understandable macros. - Fix the bug that prevented the system from responding on interfaces with link local addresses assigned. Approved by: re (gnn) END