ĵ254354 166 275 292 287 256 360 248 292 318 330 138 851 237 238 266 147 399 425 169 482 167 196 166 106 359 204 136 1244 106 170 315 241 246 132 290 154 1547 150 255 197 159 133 198 138 261 1235 102 140 187 1561 203 126 144 288 115 957 337 117 117 144 157 134 286 944 268 232 172 245 270 393 502 273 242 185 258 786 2945 1385 1133 183 107 113 164 124 158 235 273 485 220 160 160 222 119 240 239 156 346 204 276 140 132 215 165 200 406 164 310 297 324 209 543 127 218 1059 147 623 361 158 148 139 132 144 201 565 408 882 198 388 295 118 154 306 175 330 127 264 259 239 353 118 142 163 212 353 213 116 332 236 801 259 154 498 140 286 741 207 166 110 198 126 149 128 2539 210 303 308 395 K 10 svn:author V 6 tuexen K 8 svn:date V 27 2013-08-15T04:35:25.768706Z K 7 svn:log V 179 MFC r254338: Don't send uninitialized memory (two instances of 4 bytes) in every cookie on the wire. This bug was reported in https://bugzilla.mozilla.org/show_bug.cgi?id=905080 END K 10 svn:author V 6 tuexen K 8 svn:date V 27 2013-08-15T05:14:20.454686Z K 7 svn:log V 196 MFC r254352: Don't send uninitialized memory (two instances of 4 bytes) in every cookie on the wire. This bug was reported in https://bugzilla.mozilla.org/show_bug.cgi?id=905080 Approved by: re@ END K 10 svn:author V 7 glebius K 8 svn:date V 27 2013-08-15T07:54:31.033800Z K 7 svn:log V 190 Make sendfile() a method in the struct fileops. Currently only vnode backed file descriptors have this method implemented. Reviewed by: kib Sponsored by: Nginx, Inc. Sponsored by: Netflix END K 10 svn:author V 3 jlh K 8 svn:date V 27 2013-08-15T08:12:16.665574Z K 7 svn:log V 163 MFC r253853: Include /usr/local/etc/libmap.d/ by default. PR: 180568 Reviewed by: bapt Obtained from: kib Approved by: re (delphij) END K 10 svn:author V 3 jlh K 8 svn:date V 27 2013-08-15T08:21:00.442325Z K 7 svn:log V 267 Belatedly bump __FreeBSD_version for libc being an ld script. This should have been done in r251668, on June 12, 2013. This will have no practical consequences, besides having -lssp_nonshared appearing twice on the command-line for systems built in this time frame. END K 10 svn:author V 3 mav K 8 svn:date V 27 2013-08-15T10:17:20.711952Z K 7 svn:log V 155 Take SIM lock when executing async handler if lock was owned when registered. This should be partial workaround for rare SIMs that want to receive events. END K 10 svn:author V 3 gjb K 8 svn:date V 27 2013-08-15T10:31:31.147684Z K 7 svn:log V 199 MFC r254265: Make sure bootonly.iso for -BETAs and -RCs use the releases/ directory on the FTP mirrors to fetch distributions, since these are always pushed to releases/ during the release cycle. END K 10 svn:author V 3 gjb K 8 svn:date V 27 2013-08-15T10:38:10.953051Z K 7 svn:log V 225 MFC r254265: Make sure bootonly.iso for -BETAs and -RCs use the releases/ directory on the FTP mirrors to fetch distributions, since these are always pushed to releases/ during the release cycle. Approved by: re (marius) END K 10 svn:author V 7 attilio K 8 svn:date V 27 2013-08-15T11:01:25.867614Z K 7 svn:log V 233 On the recovery path for vm_page_alloc(), if a page had been requested wired, unwind back the wiring bits otherwise we can end up freeing a page that is considered wired. Sponsored by: EMC / Isilon storage division Reported by: alc END K 10 svn:author V 3 mav K 8 svn:date V 27 2013-08-15T11:41:53.158272Z K 7 svn:log V 46 Make polling work in new locking environment. END K 10 svn:author V 6 scottl K 8 svn:date V 27 2013-08-15T12:19:16.193385Z K 7 svn:log V 755 Merge r254306: Update PCI drivers to no longer look at the MEMIO-enabled bit in the PCI command register. The lazy BAR allocation code in FreeBSD sometimes disables this bit when it detects a range conflict, and will re-enable it on demand when a driver allocates the BAR. Thus, the bit is no longer a reliable indication of capability, and should not be checked. This results in the elimination of a lot of code from drivers, and also gives the opportunity to simplify a lot of drivers to use a helper API to set the busmaster enable bit. This changes fixes some recent reports of disk controllers and their associated drives/enclosures disappearing during boot. Submitted by: jhb Reviewed by: jfv, marius, adrian, achim Approved by: re END K 10 svn:author V 3 mav K 8 svn:date V 27 2013-08-15T12:55:57.462774Z K 7 svn:log V 144 Remove check for zero sized BIOs. I don't remember why have I added it, but if such BIOs exist, they probably could be completed at GEOM layer. END K 10 svn:author V 2 ae K 8 svn:date V 27 2013-08-15T13:09:58.949845Z K 7 svn:log V 146 MFC r254092: Make the check for number of entries less strict. Some partitioning tools can create GPT with number of entries less than 128. END K 10 svn:author V 2 ae K 8 svn:date V 27 2013-08-15T14:33:09.606400Z K 7 svn:log V 174 Merge r254366: Make the check for number of entries less strict. Some partitioning tools can create GPT with number of entries less than 128. Approved by: re (marius) END K 10 svn:author V 3 gjb K 8 svn:date V 27 2013-08-15T14:41:39.250044Z K 7 svn:log V 55 Update releng/9.2 to -RC2. Approved by: re (implicit) END K 10 svn:author V 3 mav K 8 svn:date V 27 2013-08-15T15:12:14.503573Z K 7 svn:log V 306 Make da(4) and ada(4) drivers call biodone() without holding any locks. That both reduces lock hold time improving performance, and makes devices reenterable, that may be needed for direct-dispatching GEOM. As side effect this removes limit on number of BIO_DELETEs aggregated in single ATA TRIM command. END K 10 svn:author V 6 dteske K 8 svn:date V 27 2013-08-15T15:34:26.153224Z K 7 svn:log V 329 MFS9 SVN r254331: Add optional support for default override of standard setup; but only if corresponding functions are provided. If override function does not exist, boot remains unmodified. This patch should not result in any changes. This includes changes to take advantage of the new functionality. Approved by: re (marius) END K 10 svn:author V 7 nemysis K 8 svn:date V 27 2013-08-15T16:03:09.142524Z K 7 svn:log V 73 - Add myself in committers-ports.dot Approved by: miwi/wxs/wg (mentors) END K 10 svn:author V 3 ken K 8 svn:date V 27 2013-08-15T16:41:27.202554Z K 7 svn:log V 389 Export the maxio field in the CAM XPT_PATH_INQ CCB in the isp(4) driver. This tells consumers up the stack the maximum I/O size that the controller can handle. The I/O size is bounded by the number of scatter/gather segments the controller can handle and the page size. For an amd64 system, it works out to around 5MB. Reviewed by: mjacob MFC after: 3 days Sponsored by: Spectra Logic END K 10 svn:author V 6 brooks K 8 svn:date V 27 2013-08-15T17:21:06.970896Z K 7 svn:log V 72 Call set_i8254_freq with MODE_STOP (0) rather than a magic number of 0. END K 10 svn:author V 6 brooks K 8 svn:date V 27 2013-08-15T17:44:44.489219Z K 7 svn:log V 100 Use an ANSI C definition of initializecpucache() to match the declaration and the rest of the file. END K 10 svn:author V 3 mav K 8 svn:date V 27 2013-08-15T19:00:36.227107Z K 7 svn:log V 74 Introduce new per-target lock to protect list of LUNs reported by device. END K 10 svn:author V 3 ray K 8 svn:date V 27 2013-08-15T19:32:08.842128Z K 7 svn:log V 14 MFC @r254374. END K 10 svn:author V 3 mav K 8 svn:date V 27 2013-08-15T19:46:31.029393Z K 7 svn:log V 266 Remove remnants of per-device periph allocation queues. It was axed out during previous locking attempt and theoretically could be restored now, but since we removed per-SIM allocation queue for good, it gives quite small real benefit to recomplicate things again. END K 10 svn:author V 5 trasz K 8 svn:date V 27 2013-08-15T20:00:32.482169Z K 7 svn:log V 109 Turn comments about locking into actual lock assertions. Reviewed by: ken Tested by: ken MFC after: 1 month END K 10 svn:author V 4 jkim K 8 svn:date V 27 2013-08-15T20:03:22.405129Z K 7 svn:log V 43 Avoid potential redefinition of the macro. END K 10 svn:author V 8 cperciva K 8 svn:date V 27 2013-08-15T20:19:17.879209Z K 7 svn:log V 1145 Change the queue of locks in kern_rangelock.c from holding lock requests in the order that they arrive, to holding (a) granted write lock requests, followed by (b) granted read lock requests, followed by (c) ungranted requests, in order of arrival. This changes the stopping condition for iterating through granted locks to see if a new request can be granted: When considering a read lock request, we can stop iterating as soon as we see a read lock request, since anything after that point is either a granted read lock request or a request which has not yet been granted. (For write lock requests, we must still compare against all granted lock requests.) For workloads with R parallel reads and W parallel writes, this improves the time spent from O((R+W)^2) to O(W*(R+W)); i.e., heavy parallel-read workloads become significantly more scalable. No statistically significant change in buildworld time has been measured, but synthetic tests of parallel 'dd > /dev/null' and 'openssl enc >/dev/null' with the input file cached yield dramatic (up to 10x) improvement with high (up to 128 processes) levels of parallelism. Reviewed by: kib END K 10 svn:author V 3 mav K 8 svn:date V 27 2013-08-15T20:28:13.451953Z K 7 svn:log V 14 MFC @ r254380 END K 10 svn:author V 3 jfv K 8 svn:date V 27 2013-08-15T20:33:17.644699Z K 7 svn:log V 78 MFC r254008 Make the fallback from MSIX to MSI interrupt usage more graceful. END K 10 svn:author V 3 jfv K 8 svn:date V 27 2013-08-15T21:06:38.240449Z K 7 svn:log V 222 MFC r254262 Further improve the msix setup, make sure pci_alloc_msix() gives us the vectors we requested, and fall back to MSI when not, also release any allocated resources before the fallback. END K 10 svn:author V 4 jkim K 8 svn:date V 27 2013-08-15T21:09:05.597229Z K 7 svn:log V 147 Simplify check for CMPXCHG8B instruction. Note CMPXCHG8B instruction is always available for Rise mP6 processors although it is not set by CPUID. END K 10 svn:author V 3 jfv K 8 svn:date V 27 2013-08-15T21:24:43.814324Z K 7 svn:log V 153 MFC r254264 Make the igb_mq_start() routine use TRYLOCK rather than simple queuing, there were latency and other problems with the latter. END K 10 svn:author V 2 np K 8 svn:date V 27 2013-08-15T21:48:29.673050Z K 7 svn:log V 41 Flush inactive LRO entries periodically. END K 10 svn:author V 4 jeff K 8 svn:date V 27 2013-08-15T22:29:49.641953Z K 7 svn:log V 196 - Fix bug in r254304. Use the ACTIVE pq count for the active list processing, not inactive. This was the result of a bad merge. Reported by: pho Sponsored by: EMC / Isilon Storage Division END K 10 svn:author V 6 jilles K 8 svn:date V 27 2013-08-15T22:33:27.619739Z K 7 svn:log V 59 sigsuspend(2): Add xrefs to pselect(2) and sigwait-alikes. END K 10 svn:author V 3 ken K 8 svn:date V 27 2013-08-15T22:52:39.635338Z K 7 svn:log V 1453 Change the way that unmapped I/O capability is advertised. The previous method was to set the D_UNMAPPED_IO flag in the cdevsw for the driver. The problem with this is that in many cases (e.g. sa(4)) there may be some instances of the driver that can handle unmapped I/O and some that can't. The isp(4) driver can handle unmapped I/O, but the esp(4) driver currently cannot. The cdevsw is shared among all driver instances. So instead of setting a flag on the cdevsw, set a flag on the cdev. This allows drivers to indicate support for unmapped I/O on a per-instance basis. sys/conf.h: Remove the D_UNMAPPED_IO cdevsw flag and replace it with an SI_UNMAPPED cdev flag. kern_physio.c: Look at the cdev SI_UNMAPPED flag to determine whether or not a particular driver can handle unmapped I/O. geom_dev.c: Set the SI_UNMAPPED flag for all GEOM cdevs. Since GEOM will create a temporary mapping when needed, setting SI_UNMAPPED unconditionally will work. Remove the D_UNMAPPED_IO flag. nvme_ns.c: Set the SI_UNMAPPED flag on cdevs created here if NVME_UNMAPPED_BIO_SUPPORT is enabled. vfs_aio.c: In aio_qphysio(), check the SI_UNMAPPED flag on a cdev instead of the D_UNMAPPED_IO flag on the cdevsw. sys/param.h: Bump __FreeBSD_version to 1000045 for the switch from setting the D_UNMAPPED_IO flag in the cdevsw to setting SI_UNMAPPED in the cdev. Reviewed by: kib, jimharris MFC after: 1 week Sponsored by: Spectra Logic END K 10 svn:author V 4 jkim K 8 svn:date V 27 2013-08-15T23:58:16.583975Z K 7 svn:log V 57 MFC: r254379 Avoid potential redefinition of the macro. END K 10 svn:author V 4 jkim K 8 svn:date V 27 2013-08-16T00:00:36.694084Z K 7 svn:log V 161 MFC: r254384 Simplify check for CMPXCHG8B instruction. Note CMPXCHG8B instruction is always available for Rise mP6 processors although it is not set by CPUID. END K 10 svn:author V 4 jkim K 8 svn:date V 27 2013-08-16T00:04:58.286956Z K 7 svn:log V 103 - Optimize atomic_cmpset_64_i386(). - Move all 64-bit atomic functions to atomic.h. Suggested by: bde END K 10 svn:author V 4 jkim K 8 svn:date V 27 2013-08-16T00:14:52.250501Z K 7 svn:log V 66 Restore atomic_cmpset_int() for CPU_DISABLE_CMPXCHG case as well. END K 10 svn:author V 4 jkim K 8 svn:date V 27 2013-08-16T00:24:34.320044Z K 7 svn:log V 40 Micro-optimize atomic_cmpset_64_i386(). END K 10 svn:author V 6 grehan K 8 svn:date V 27 2013-08-16T00:35:20.869657Z K 7 svn:log V 102 Fix ordering of legacy IRQ reservations. Submitted by: Jeremiah Lott jlott at averesystems dot com END K 10 svn:author V 5 markj K 8 svn:date V 27 2013-08-16T03:41:41.735867Z K 7 svn:log V 44 Use strdup(9) instead of reimplementing it. END K 10 svn:author V 3 gjb K 8 svn:date V 27 2013-08-16T05:30:13.678312Z K 7 svn:log V 168 Mark the makefs(8) '-p' flag as deprecated in preference for the '-Z' flag for compatibility with NetBSD. Submitted by: Eric van Gyzen (via stable@) MFC after: 3 days END K 10 svn:author V 7 davidxu K 8 svn:date V 27 2013-08-16T06:40:12.730016Z K 7 svn:log V 1137 MFC r239347, 240295, 240296 and 253325: r239347 | davidxu | 2012-08-17 10:26:31 +0800 (Fri, 17 Aug 2012) | 7 lines Implement syscall clock_getcpuclockid2, so we can get a clock id for process, thread or others we want to support. Use the syscall to implement POSIX API clock_getcpuclock and pthread_getcpuclockid. PR: 168417 ------------------------------------------------------------------------ r240295 | davidxu | 2012-09-10 13:00:29 +0800 (Mon, 10 Sep 2012) | 2 lines Add missing prototype for clock_getcpuclockid. ------------------------------------------------------------------------ r240296 | davidxu | 2012-09-10 13:09:39 +0800 (Mon, 10 Sep 2012) | 2 lines Process CPU-Time Clocks option is supported, define _POSIX_CPUTIME. ------------------------------------------------------------------------ r253325 | kib | 2013-07-14 03:32:50 +0800 (Sun, 14 Jul 2013) | 6 lines Allow to call clock_gettime() on the clock id for zombie process. Reported by: Petr Salinger PR: threads/180496 Sponsored by: The FreeBSD Foundation END K 10 svn:author V 7 davidxu K 8 svn:date V 27 2013-08-16T06:42:24.676062Z K 7 svn:log V 7 Regen. END K 10 svn:author V 7 glebius K 8 svn:date V 27 2013-08-16T07:02:17.091093Z K 7 svn:log V 44 Provide UTF-8 version of russian calendars. END K 10 svn:author V 2 cy K 8 svn:date V 27 2013-08-16T07:03:28.648702Z K 7 svn:log V 96 Include three new files that should be in vendor-sys branch too. Approved by: glebius (Mentor) END K 10 svn:author V 5 erwin K 8 svn:date V 27 2013-08-16T07:11:13.140874Z K 7 svn:log V 1465 MFC 253983, 253984: Update Bind to 9.8.5-P2 New Features Adds a new configuration option, "check-spf"; valid values are "warn" (default) and "ignore". When set to "warn", checks SPF and TXT records in spf format, warning if either resource record type occurs without a corresponding record of the other resource record type. [RT #33355] Adds support for Uniform Resource Identifier (URI) resource records. [RT #23386] Adds support for the EUI48 and EUI64 RR types. [RT #33082] Adds support for the RFC 6742 ILNP record types (NID, LP, L32, and L64). [RT #31836] Feature Changes Changes timing of when slave zones send NOTIFY messages after loading a new copy of the zone. They now send the NOTIFY before writing the zone data to disk. This will result in quicker propagation of updates in multi-level server structures. [RT #27242] "named -V" can now report a source ID string. (This is will be of most interest to developers and troubleshooters). The source ID for ISC's production versions of BIND is defined in the "srcid" file in the build tree and is normally set to the most recent git hash. [RT #31494] Response Policy Zone performance enhancements. New "response-policy" option "min-ns-dots". "nsip" and "nsdname" now enabled by default with RPZ. [RT #32251] Approved by: delphij (mentor) Sponsored by: DK Hostmaster A/S END K 10 svn:author V 2 cy K 8 svn:date V 27 2013-08-16T07:16:26.377977Z K 7 svn:log V 111 Tag three new files that should have been in vendor-sys in r254219 and r254221. Approved by: glebius (Mentor) END K 10 svn:author V 5 kevlo K 8 svn:date V 27 2013-08-16T07:42:06.849741Z K 7 svn:log V 32 Bring datasheet URL up to date. END K 10 svn:author V 3 mav K 8 svn:date V 27 2013-08-16T09:12:42.416220Z K 7 svn:log V 52 Add creation of taskqueue thread missed at r254214. END K 10 svn:author V 3 mav K 8 svn:date V 27 2013-08-16T09:28:09.410516Z K 7 svn:log V 195 Add safety belt in case of CCB memory allocation failed: queue the periph to taskqueue where it can safely sleep waiting for memory. That is better then just quit, leaving periph stuck forever. END K 10 svn:author V 3 des K 8 svn:date V 27 2013-08-16T10:53:36.984223Z K 7 svn:log V 23 GC unused source file. END K 10 svn:author V 3 mav K 8 svn:date V 27 2013-08-16T12:25:02.817710Z K 7 svn:log V 864 Make first steps toward direct BIO dispatch in GEOM: - Define flags, declaring that specific consumer/provider is capable of sending requests/replies (respectively) directly, i,e. doesn't hold any locks and so reenterable, and/or able to directly receive replies/requests, i.e. doesn't depend on GEOM up/down threads semantics. As result, GEOM will make direct calls only if both caller and callee are cpable of it in each case. - Define disk(9) flag to declare that disk is capable of direct request completion, and use it for da(4) and ada(4) drivers. Make GEOM DISK to pass that flag to its provider and also assume that any disk is capable of receiveing requests directly. - Mark GEOM DEV as capable of both direct send and receive. - Make md(4) declare both direct send and receive after adding mutex to serialize its statistics update on request path. END K 10 svn:author V 6 jilles K 8 svn:date V 27 2013-08-16T13:10:30.573187Z K 7 svn:log V 241 Add dup3(), based on F_DUP2FD and F_DUP2FD_CLOEXEC fcntls. I removed functionality not proposed for POSIX in Austin group issue #411. A man page (my own) and test cases will follow in later commits. PR: 176233 Submitted by: Jukka Ukkonen END K 10 svn:author V 6 jilles K 8 svn:date V 27 2013-08-16T13:16:27.579184Z K 7 svn:log V 22 Add man page dup3(3). END K 10 svn:author V 6 jilles K 8 svn:date V 27 2013-08-16T13:16:55.845794Z K 7 svn:log V 22 Add tests for dup3(). END K 10 svn:author V 6 jilles K 8 svn:date V 27 2013-08-16T13:48:11.432981Z K 7 svn:log V 49 sh: Add test for the non-standard jobid builtin. END K 10 svn:author V 6 jilles K 8 svn:date V 27 2013-08-16T13:56:43.223291Z K 7 svn:log V 62 sh: Recognize "--" as end of options in bg/fg/jobid builtins. END K 10 svn:author V 6 jilles K 8 svn:date V 27 2013-08-16T14:06:29.205137Z K 7 svn:log V 39 pselect(2): Add xref to sigsuspend(2). END K 10 svn:author V 3 kib K 8 svn:date V 27 2013-08-16T14:22:20.885408Z K 7 svn:log V 193 Restore the previous sendfile(2) behaviour on the block devices. Provide valid .fo_sendfile method for several missed struct fileops. Reviewed by: glebius Sponsored by: The FreeBSD Foundation END K 10 svn:author V 3 ken K 8 svn:date V 27 2013-08-16T16:14:32.891278Z K 7 svn:log V 851 Add unmapped I/O and larger I/O support to the sa(4) driver. We now pay attention to the maxio field in the XPT_PATH_INQ CCB, and if it is set, propagate it up to physio via the si_iosize_max field in the cdev structure. We also now pay attention to the PIM_UNMAPPED capability bit in the XPT_PATH_INQ CCB, and set the new SI_UNMAPPED cdev flag when the underlying SIM supports unmapped I/O. scsi_sa.c: Add unmapped I/O support and propagate the SIM's maximum I/O size up. Adjust scsi_tape_read_write() in the same way that scsi_read_write() was changed to support unmapped I/O. We overload the readop parameter with bits that tell us whether it's an unmapped I/O, and we need to set the CAM_DATA_BIO CCB flag. This change should be backwards compatible in source and binary forms. MFC after: 1 week Sponsored by: Spectra Logic END K 10 svn:author V 3 sjg K 8 svn:date V 27 2013-08-16T16:26:23.167206Z K 7 svn:log V 175 When we need to build using the in-tree make, switch at the earliest opportunity. In the case of fmake vs bmake, this helps ensure correct load handling. Reviewed by: obrien END K 10 svn:author V 3 mav K 8 svn:date V 27 2013-08-16T16:55:39.420674Z K 7 svn:log V 139 Move random things done by g_down thread before executing request inside g_io_check() and don't forget to call it also on direct dispatch. END K 10 svn:author V 3 sjg K 8 svn:date V 27 2013-08-16T18:47:18.144159Z K 7 svn:log V 80 Add .MAKE.ALWAYS_PASS_JOB_QUEUE knob (default yes) for backwards compatability. END K 10 svn:author V 7 delphij K 8 svn:date V 27 2013-08-16T19:23:44.405493Z K 7 svn:log V 148 Update vendor-sys/illumos/dist to illumos-gate 14133:4caa3904e52a: Illumos ZFS issues: 4039 zfs_rename()/zfs_link() needs stronger test for XDEV END K 10 svn:author V 7 delphij K 8 svn:date V 27 2013-08-16T19:27:29.220402Z K 7 svn:log V 173 Update vendor/illumos/dist and vendor-sys/illumos/dist to illumos-gate 14137:f7523d207d74: Illumos ZFS issues: 3996 want a libzfs_core API to rollback to latest snapshot END K 10 svn:author V 7 delphij K 8 svn:date V 27 2013-08-16T19:33:36.380060Z K 7 svn:log V 296 Update vendor/illumos/dist and vendor-sys/illumos/dist to illumos-gate 14143:42d090a37218: Illumos DTrace issues: 3089 want ::typedef 3094 libctf should support removing a dynamic type 3095 libctf does not validate arrays correctly 3096 libctf does not validate function types correctly END K 10 svn:author V 3 ian K 8 svn:date V 27 2013-08-16T19:40:00.997220Z K 7 svn:log V 409 When the timeout clock is based on the SD clock, the timeout counter has to be recalculated every time the SD clock frequency changes. Also, tidy up the counter calculation... it makes no sense to calculate a value one larger than the limit, then whine that it's too large and truncate it to the limit. If the BROKEN_TIMEOUT quirk is set, don't calculate the counter at all, just set it to the limit value. END K 10 svn:author V 3 ian K 8 svn:date V 27 2013-08-16T19:44:49.829873Z K 7 svn:log V 180 Add named constants for 8-bit bus support. The sdhci and mmc drivers don't have support for this yet, but some low-level hardware is ready for it when the higher layers catch up. END K 10 svn:author V 3 ian K 8 svn:date V 27 2013-08-16T20:22:57.665331Z K 7 svn:log V 149 Print the card relative address in hex, because that's what all the other debugging output does (when it appears in command arguments, for example). END K 10 svn:author V 6 jilles K 8 svn:date V 27 2013-08-16T20:24:41.670635Z K 7 svn:log V 90 sh: Remove unnecessary reset functions. These are already handled by exception handlers. END K 10 svn:author V 3 ian K 8 svn:date V 27 2013-08-16T20:32:56.803635Z K 7 svn:log V 165 During card identification, run the bus at 400KHz, not the minimum speed the bus claims to be capable of. The 400KHz speed is dictated by the SD and MMC standards. END K 10 svn:author V 2 np K 8 svn:date V 27 2013-08-16T20:35:03.620578Z K 7 svn:log V 694 Implement support for rx buffer packing. This is a T4 and T5 chip feature which lets the chip deliver multiple Ethernet frames in a single buffer. This is more efficient within the chip, in the driver, and reduces wastage of space in rx buffers. - Always allocate rx buffers from the jumbop zone, no matter what the MTU is. Do not use the normal cluster refcounting mechanism. - Reserve space for an mbuf and a refcount in the cluster itself and let the chip DMA multiple frames in the rest. - Use the embedded mbuf for the first frame and allocate mbufs on the fly for any additional frames delivered in the cluster. Each of these mbufs has a reference on the underlying cluster. END K 10 svn:author V 2 np K 8 svn:date V 27 2013-08-16T21:04:58.714108Z K 7 svn:log V 2852 Enable buffer packing by default for T5 cards. At this point I consistently get full 40G line rate for a single TCP connection (both 1500 MTU and 9000 MTU, both single direction and bidirectional 40G tx + 40G rx simultaneously) out of the box. For example, this is on a recent HEAD with changes from this branch. No tuning. sysctl.conf is empty and loader.conf has just one line setting a high hw.intr_storm_threshold. No special buffer sizes or r/w sizes are given to the benchmark programs either. # sysctl hw.model hw.model: Intel(R) Xeon(R) CPU E5-1620 0 @ 3.60GHz # iperf -v iperf version 2.0.5 (08 Jul 2010) pthreads /* MTU 1500. */ # iperf -c ha1 -t60 ------------------------------------------------------------ Client connecting to ha1, TCP port 5001 TCP window size: 32.5 KByte (default) ------------------------------------------------------------ [ 3] local 10.3.166.121 port 32906 connected with 10.3.166.122 port 5001 [ ID] Interval Transfer Bandwidth [ 3] 0.0-60.0 sec 263 GBytes 37.6 Gbits/sec # iperf -c ha1 -t60 -d ------------------------------------------------------------ Server listening on TCP port 5001 TCP window size: 64.0 KByte (default) ------------------------------------------------------------ ------------------------------------------------------------ Client connecting to ha1, TCP port 5001 TCP window size: 265 KByte (default) ------------------------------------------------------------ [ 5] local 10.3.166.121 port 54353 connected with 10.3.166.122 port 5001 [ 4] local 10.3.166.121 port 5001 connected with 10.3.166.122 port 33091 [ ID] Interval Transfer Bandwidth [ 4] 0.0-60.0 sec 262 GBytes 37.5 Gbits/sec [ 5] 0.0-60.0 sec 263 GBytes 37.6 Gbits/sec /* MTU 9000. */ # iperf -c ha0 -t60 ------------------------------------------------------------ Client connecting to ha0, TCP port 5001 TCP window size: 35.0 KByte (default) ------------------------------------------------------------ [ 3] local 10.2.166.121 port 44734 connected with 10.2.166.122 port 5001 [ ID] Interval Transfer Bandwidth [ 3] 0.0-60.0 sec 276 GBytes 39.6 Gbits/sec # iperf -c ha0 -t60 -d ------------------------------------------------------------ Server listening on TCP port 5001 TCP window size: 64.0 KByte (default) ------------------------------------------------------------ ------------------------------------------------------------ Client connecting to ha0, TCP port 5001 TCP window size: 291 KByte (default) ------------------------------------------------------------ [ 5] local 10.2.166.121 port 53293 connected with 10.2.166.122 port 5001 [ 4] local 10.2.166.121 port 5001 connected with 10.2.166.122 port 14311 [ ID] Interval Transfer Bandwidth [ 5] 0.0-60.0 sec 276 GBytes 39.5 Gbits/sec [ 4] 0.0-60.0 sec 276 GBytes 39.5 Gbits/sec END K 10 svn:author V 3 jhb K 8 svn:date V 27 2013-08-16T21:13:55.240401Z K 7 svn:log V 1291 Add new mmap(2) flags to permit applications to request specific virtual address alignment of mappings. - MAP_ALIGNED(n) requests a mapping aligned on a boundary of (1 << n). Requests for n >= number of bits in a pointer or less than the size of a page fail with EINVAL. This matches the API provided by NetBSD. - MAP_ALIGNED_SUPER is a special case of MAP_ALIGNED. It can be used to optimize the chances of using large pages. By default it will align the mapping on a large page boundary (the system is free to choose any large page size to align to that seems best for the mapping request). However, if the object being mapped is already using large pages, then it will align the virtual mapping to match the existing large pages in the object instead. - Internally, VMFS_ALIGNED_SPACE is now renamed to VMFS_SUPER_SPACE, and VMFS_ALIGNED_SPACE(n) is repurposed for specifying a specific alignment. MAP_ALIGNED(n) maps to using VMFS_ALIGNED_SPACE(n), while MAP_ALIGNED_SUPER maps to VMFS_SUPER_SPACE. - mmap() of a device object now uses VMFS_OPTIMAL_SPACE rather than explicitly using VMFS_SUPER_SPACE. All device objects are forced to use a specific color on creation, so VMFS_OPTIMAL_SPACE is effectively equivalent. Reviewed by: alc MFC after: 1 month END K 10 svn:author V 3 ian K 8 svn:date V 27 2013-08-16T23:05:34.314159Z K 7 svn:log V 1039 Handle command retries for commands originating at the mmc layer, and ensure that all such commands have a non-zero retry count except for those that are expected to fail (for example, because they are used to probe for feature support). While it is possible to pass a retry count down to the hardware driver in the command request structure, no hardware driver currently implements any retry logic. The hardware doesn't know much about the context of a single request, so it makes more sense to handle retries at a layer that does. This adds retry loops to the mmc_wait_for_cmd() and mmc_wait_for_app_cmd() functions. These functions are the gateway from other code within mmc.c to the hardware. App commands are a sequence of two commands and a retry has to rerun both of them in order, so it needs its own retry loop. Retry looping is specifically NOT implemented in mmc_wait_for_request() because it is the gateway for children on the bus, and they have to implement their own retry logic depending on what makes sense for them. END K 10 svn:author V 3 ian K 8 svn:date V 27 2013-08-17T00:19:27.585378Z K 7 svn:log V 91 Consistently init all mmc request, command, and data structures to zero before using them. END K 10 svn:author V 7 delphij K 8 svn:date V 27 2013-08-17T00:25:11.082057Z K 7 svn:log V 11 Fix build. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2013-08-17T01:02:04.351885Z K 7 svn:log V 18 Correct spelling. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2013-08-17T01:14:28.244054Z K 7 svn:log V 69 Log the MAC address of the node in question rather than the pointer. END K 10 svn:author V 6 mjacob K 8 svn:date V 27 2013-08-17T01:17:51.058711Z K 7 svn:log V 29 Correct sentence syntax too. END K 10 svn:author V 3 sjg K 8 svn:date V 27 2013-08-17T04:41:35.173689Z K 7 svn:log V 66 Don't let user specified DESTDIR, break building our chosen make. END K 10 svn:author V 8 hselasky K 8 svn:date V 27 2013-08-17T06:29:45.478218Z K 7 svn:log V 137 Fix some USB controller names according to pciconf output. MFC after: 1 week Submitted by: Dmitry Luhtionov END K 10 svn:author V 3 kib K 8 svn:date V 27 2013-08-17T07:10:01.439615Z K 7 svn:log V 180 Remove the arbitrary binding of the pagedaemon threads to the domains, update the comment accordingly and make it more precise. Requested and reviewed by: jeff (previous version) END K 10 svn:author V 3 hrs K 8 svn:date V 27 2013-08-17T07:12:52.804657Z K 7 svn:log V 392 Unbreak rwhod(8): - It did not work with GENERIC kernel after r250603 because options PROCDESC was required for pdfork(2). It now just uses fork(2) instead when this syscall is not available. - Fix verify(). This function was broken in r250602 because the outermost "()" was removed from the condition !(isalnum() || ispunct()). It prevented hostnames including "-", for example. END K 10 svn:author V 3 hrs K 8 svn:date V 27 2013-08-17T07:14:45.664777Z K 7 svn:log V 127 Return 0 in nbi->expire when la_expire == 0. Conversion from time_uptime to time_second should not be performed in this case. END K 10 svn:author V 3 kib K 8 svn:date V 27 2013-08-17T08:05:32.835319Z K 7 svn:log V 68 MFC r254191: Match malloc(9) calls with free(9), not contigfree(9). END K 10 svn:author V 3 kib K 8 svn:date V 27 2013-08-17T08:08:58.649957Z K 7 svn:log V 68 MFC r254191: Match malloc(9) calls with free(9), not contigfree(9). END K 10 svn:author V 3 kib K 8 svn:date V 27 2013-08-17T08:31:34.342780Z K 7 svn:log V 129 MFC r253969: Do not override the ENOENT error for the empty path, or EFAULT errors from copyins, with the relative lookup check. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-17T14:09:46.335263Z K 7 svn:log V 27 Remove redundant variable. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-17T14:13:45.521618Z K 7 svn:log V 147 In r114945 the line 'nmp = TAILQ_NEXT(mp, mnt_list);' was duplicated. Instead of just removing the duplicate, convert the loop to TAILQ_FOREACH(). END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-17T14:17:13.459809Z K 7 svn:log V 146 Make pdfork(2), pdkill(2) and pdgetpid(2) syscalls available for 32bit binaries running under 64bit kernel. Sponsored by: The FreeBSD Foundation END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-17T14:18:41.019149Z K 7 svn:log V 64 Regenerate after r254447. Sponsored by: The FreeBSD Foundation END K 10 svn:author V 6 andrew K 8 svn:date V 27 2013-08-17T14:36:32.134183Z K 7 svn:log V 250 Silence a warning that is incorrect on ARMv6 and later. In the smull, umull, smlal, and umlal the output registers are allowed to be the same as either input registers, where in ARMv4 and ARMv5 they could only be the same as the last input register. END K 10 svn:author V 6 andrew K 8 svn:date V 27 2013-08-17T14:42:40.119385Z K 7 svn:log V 108 Ensure we set all fpu registers to zero by using the address and size of the union over one of its members. END K 10 svn:author V 6 andrew K 8 svn:date V 27 2013-08-17T14:52:19.048972Z K 7 svn:log V 180 Remove unused FPE code. This is not enabled anywhere as it is the only file I can find containing FAST_FPE. It appears this would not work as want_resched is not defined anywhere. END K 10 svn:author V 6 andrew K 8 svn:date V 27 2013-08-17T14:53:53.628424Z K 7 svn:log V 45 Remove fpe_sp_state as we don't support fpe. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-17T14:55:31.208980Z K 7 svn:log V 40 Correct function name and return value. END K 10 svn:author V 6 andrew K 8 svn:date V 27 2013-08-17T15:09:14.261886Z K 7 svn:log V 119 Remove the ARMFPE option. It is unsupported, and appears to be broken as arm_fpe_core_changecontext is not a function. END K 10 svn:author V 6 andrew K 8 svn:date V 27 2013-08-17T15:21:17.280066Z K 7 svn:log V 70 Remove the armfpe config options. These files don't exist on FreeBSD. END K 10 svn:author V 3 kib K 8 svn:date V 27 2013-08-17T16:42:18.775584Z K 7 svn:log V 107 Fix the mismerge in r254444, use '==' instead of '='. Pointed out by: jilles, John Wehle END K 10 svn:author V 6 bryanv K 8 svn:date V 27 2013-08-17T17:02:43.329187Z K 7 svn:log V 310 Do not use potentially stale thread in kthread_add() When an existing process is provided, the thread selected to use to initialize the new thread could have exited and be reaped. Acquire the proc lock earlier to ensure the thread remains valid. Reviewed by: jhb, julian (previous version) MFC after: 3 days END K 10 svn:author V 6 bryanv K 8 svn:date V 27 2013-08-17T17:09:26.232632Z K 7 svn:log V 69 Move the resetting of the completion queue size to the correct spot. END K 10 svn:author V 3 hrs K 8 svn:date V 27 2013-08-17T17:23:42.916953Z K 7 svn:log V 217 - Use getnameinfo(3) instead of gethostbyaddr(3) or inet_ntop(3). - Fill sin6_scope_id from in6p.sin6_addr.s6_addr[2]. struct inpcb has struct in6_addr for the endpoint addresses, so sin6_scope_id must be filled. END K 10 svn:author V 3 mav K 8 svn:date V 27 2013-08-17T18:38:00.701597Z K 7 svn:log V 204 Remove locking from taskqueue_member(). The list of threads is static during the taskqueue life cycle, so there is no need to protect it, taking quite congested lock several more times for each ZFS I/O. END K 10 svn:author V 6 andrew K 8 svn:date V 27 2013-08-17T18:51:38.234528Z K 7 svn:log V 228 Rename device vfp to option VFP and retire the ARM_VFP_SUPPORT option. This simplifies enabling as previously both options were required to be enabled, now we only need a single option. While here enable VFP on the PandaBoard. END K 10 svn:author V 3 hrs K 8 svn:date V 27 2013-08-17T19:23:35.623869Z K 7 svn:log V 116 - Remove struct ifinfo *iflist. It is no longer used. - Suppress warnings about increase of alignment requirement. END K 10 svn:author V 6 jilles K 8 svn:date V 27 2013-08-17T19:24:58.762789Z K 7 svn:log V 447 libc: Access _logname_valid more efficiently. The variable _logname_valid is not exported via the version script; therefore, change C and i386/amd64 assembler code to remove indirection (which allowed interposition). This makes the code slightly smaller and faster. Also, remove #define PIC_GOT from i386/amd64 in !PIC mode. Without PIC, there is no place containing the address of each variable, so there is no possible definition for PIC_GOT. END K 10 svn:author V 3 ian K 8 svn:date V 27 2013-08-17T19:29:51.108968Z K 7 svn:log V 35 Enable VFP support for BeagleBone. END K 10 svn:author V 6 emaste K 8 svn:date V 27 2013-08-17T19:34:41.970885Z K 7 svn:log V 122 Correct implementation of atomic_flag_test_and_set The function sets the flag and returns the previous value (7.17.8.1). END K 10 svn:author V 4 neel K 8 svn:date V 27 2013-08-17T19:49:08.065297Z K 7 svn:log V 965 Bump up the maximum addressable memory on amd64 systems from 1TB to 4TB. Bump up the KVA size proportionally from 512GB to 2TB. The number of page table pages used by the direct map is now calculated at run time based on 'Maxmem'. This means the small memory systems will not see any additional tax in terms of page table pages for the direct map. However all amd64 systems, regardless of the memory size, will use 3 more pages to accomodate the bump in the KVA size. More details available here: http://lists.freebsd.org/pipermail/freebsd-hackers/2013-June/043015.html http://lists.freebsd.org/pipermail/freebsd-current/2013-July/043143.html Tested with the following configurations: - Sandybridge server with 64GB of memory. - bhyve VM with 64MB of memory. - bhyve VM with a 8GB of memory with the memory segment above 4GB cuddling right up against the 4TB maximum memory limit. Discussed on: hackers@, current@ Submitted by: Chris Torek (torek@torek.net) END K 10 svn:author V 5 markj K 8 svn:date V 27 2013-08-17T21:53:37.766358Z K 7 svn:log V 53 Remove a couple of unused macros. MFC after: 3 days END K 10 svn:author V 5 markj K 8 svn:date V 27 2013-08-17T22:02:26.491621Z K 7 svn:log V 528 Add a "translated type" argument to SDT_PROBE_ARGTYPE() and add some macros which allow one to define SDT probes that specify translated types. The idea is to make it easy to write SDT probe definitions that can work across multiple operating systems. In particular, this makes it possible to port illumos SDT probes to FreeBSD without changing their argument types, so long as the appropriate translators are defined. Then DTrace scripts written for Solaris/illumos will work on FreeBSD without any changes. MFC after: 1 week END K 10 svn:author V 5 markj K 8 svn:date V 27 2013-08-17T22:06:30.535495Z K 7 svn:log V 266 Update the SDT(9) man page with the macros added in 254468. Also change the existing examples to not pass an mbuf as a probe argument. There's no obvious reason to have it there, and it doesn't really jibe with the example added in this revision. MFC after: 1 week END K 10 svn:author V 3 hrs K 8 svn:date V 27 2013-08-17T22:13:26.131481Z K 7 svn:log V 66 Use sysctl(ICMPV6CTL_ND6_DRLIST) instead of SIOCGDRLST_IN6 ioctl. END K 10 svn:author V 5 peter K 8 svn:date V 27 2013-08-17T23:33:36.054498Z K 7 svn:log V 54 Attempt to enable commits on the root for properties. END K 10 svn:author V 5 peter K 8 svn:date V 27 2013-08-17T23:36:50.370762Z K 7 svn:log V 45 Try again - enable commits to root directory END K 10 svn:author V 5 peter K 8 svn:date V 27 2013-08-17T23:37:16.548823Z K 7 svn:log V 38 Enable auto-props for the repository. END K 10 svn:author V 5 peter K 8 svn:date V 27 2013-08-18T01:29:57.300388Z K 7 svn:log V 50 Add *.dts, *.dtsi, bsd.*.mk Submitted by: rpaulo END K 10 svn:author V 6 bryanv K 8 svn:date V 27 2013-08-18T04:22:13.356129Z K 7 svn:log V 105 Do not populate the second Rx ring unless the configuration makes sense (LRO is enabled or a large MTU). END K 10 svn:author V 6 adrian K 8 svn:date V 27 2013-08-18T06:08:52.547882Z K 7 svn:log V 469 Add in missing events for Sandy Bridge Xeon. * Add in MEM_LOAD_UOPS_LLC_HIT_RETIRED for both sandy bridge and sandy bridge Xeon. Right now it only is enabled for Sandy Bridge. * D2/0F is actually a combination rather than a separate counter, so just flip that on for the CPU types that support it. There's an errata for using this on SB Xeon hardware - I've documented it in kern/181346. Tested: * Sandy Bridge * Sandy Bridge Xeon Sponsored by: Netflix, Inc. END K 10 svn:author V 6 emaste K 8 svn:date V 27 2013-08-18T08:18:49.844897Z K 7 svn:log V 312 MFC r239960: Properly enable Clang-style atomics when available. In addition to testing against cxx_atomic, we must check c_atomic. The former is only set when building C++ code. Also use __has_extension instead of __has_feature. This allows us to use the atomics outside of C11. PR: threads/170073 END K 10 svn:author V 6 emaste K 8 svn:date V 27 2013-08-18T08:24:58.275911Z K 7 svn:log V 786 MFC r240970: - Make C11 atomic macros usable in expressions: - Replace do-while statements with void expressions. - Wrap __asm statements in statement expressions. - Make the macros function-like: - Evaluate all arguments exactly once. - Make sure there's a sequence point between evaluation of the arguments and the function body. Arguments should be evaluated before any memory barriers. - Fix use of __atomic_is_lock_free built-in. It requires the address of an atomic variable as second argument. Use this built-in on clang as well because clang's __c11_atomic_is_lock_free only takes the size of the variable into account. - In atomic_exchange_explicit put the barrier before instead of after the __sync_lock_test_and_set call. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-18T09:45:52.603778Z K 7 svn:log V 105 Better organize the filecaps structure, which reduces its size from 32 bytes to 24 bytes on 64bit archs. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-18T10:21:29.986403Z K 7 svn:log V 295 Add process descriptors support to the GENERIC kernel. It is already being used by the tools in base systems and with sandboxing more and more tools the usage should only increase. Submitted by: Mariusz Zaborski Sponsored by: Google Summer of Code 2013 MFC after: 1 month END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-18T10:30:41.504620Z K 7 svn:log V 202 Implement 32bit versions of the cap_ioctls_limit(2) and cap_ioctls_get(2) system calls as unsigned longs have different size on i386 and amd64. Reported by: jilles Sponsored by: The FreeBSD Foundation END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-18T10:31:30.115303Z K 7 svn:log V 26 Regenerate after r254481. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-18T10:33:46.577756Z K 7 svn:log V 62 Make example more correct (errstr is a pointer, not boolean). END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-18T10:38:59.102662Z K 7 svn:log V 213 Consistently use 'af' as an argument name for address family. Now both gethostbyname2(3) and gethostbyaddr(3) use the same argument name. The same argument name is also used in implementations of those functions. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-18T10:44:37.763430Z K 7 svn:log V 83 Make the "FD" column one character wider, so that "trace" can also align properly. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-18T11:25:42.609333Z K 7 svn:log V 237 Cast argument of is*() ctype functions to unsigned char. Without the cast there is ambiguity between 0xFF and -1 (EOF). Suggested by: jilles Submitted by: Mariusz Zaborski Sponsored by: Google Summer of Code 2013 END K 10 svn:author V 6 andrew K 8 svn:date V 27 2013-08-18T11:54:20.666035Z K 7 svn:log V 32 Enable VFP support on EFIKA MX. END K 10 svn:author V 6 jilles K 8 svn:date V 27 2013-08-18T13:25:18.182138Z K 7 svn:log V 168 dup3(3): Replace copyright notice. Although I copied dup(2) to create dup3(3), I removed almost all the non-boilerplate, so dup3(3) is copyright me. Reported by: bjk END K 10 svn:author V 6 jilles K 8 svn:date V 27 2013-08-18T13:27:04.458941Z K 7 svn:log V 163 Disallow opening a POSIX message queue for execute. O_EXEC was formerly ignored, so equivalent to O_RDONLY. Reject O_EXEC with [EINVAL] like the invalid mode 3. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-18T13:34:11.500182Z K 7 svn:log V 146 Move the PAIR32TO64() macro and the RETVAL_HI/RETVAL_LO defines to a header file for use by other .c files. Sponsored by: The FreeBSD Foundation END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-18T13:37:54.984291Z K 7 svn:log V 260 The cap_rights_limit(2) system calls needs a wrapper for 32bit binaries running under 64bit kernels as the 'rights' argument has to be split into two registers or the half of the rights will disappear. Reported by: jilles Sponsored by: The FreeBSD Foundation END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-18T13:38:39.342781Z K 7 svn:log V 26 Regenerate after r254491. END K 10 svn:author V 6 andrew K 8 svn:date V 27 2013-08-18T16:16:36.587996Z K 7 svn:log V 47 Enable VFP on the CubieBoard and CubieBoard 2. END K 10 svn:author V 6 andrew K 8 svn:date V 27 2013-08-18T17:18:52.263651Z K 7 svn:log V 68 Enable VFP in the Versatile PB (QEMU) kernel. Tested on QEMU 1.6.0. END K 10 svn:author V 6 cognet K 8 svn:date V 27 2013-08-18T18:08:12.027743Z K 7 svn:log V 116 Increase the max KVA available for general consumption on the Exynos 5. Submitted by: Ruslan Bukin END K 10 svn:author V 3 ian K 8 svn:date V 27 2013-08-18T19:08:53.990987Z K 7 svn:log V 260 Add a new SDHCI_QUIRK_DONT_SHIFT_RESPONSE for hardware that pre-shifts the response bits the way we do in software. While the hardware is just doing the sensible thing rather than leaving it to the software, it's in violation of the spec by doing so. Grrrr. END K 10 svn:author V 4 tijl K 8 svn:date V 27 2013-08-18T19:37:35.045484Z K 7 svn:log V 119 Change the return type of the fallback implementation of the atomic_compare_exchange_* macros in stdatomic.h to _Bool. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-18T20:06:35.777273Z K 7 svn:log V 24 Remove redundant space. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2013-08-18T20:11:34.601639Z K 7 svn:log V 239 Implement fdclosedir(3) function, which is equivalent to the closedir(3) function, but returns directory file descriptor instead of closing it. Submitted by: Mariusz Zaborski Sponsored by: Google Summer of Code 2013 END K 10 svn:author V 6 adrian K 8 svn:date V 27 2013-08-18T20:40:13.066663Z K 7 svn:log V 140 Return the correct status if ieee80211_ff_check() consumes the mbuf. I broke this when converting the net80211 TX path to use if_transmit. END K 10 svn:author V 3 kib K 8 svn:date V 27 2013-08-18T21:36:22.108716Z K 7 svn:log V 708 When code from r254064 in pmap_ts_referenced() drops pv lock and blocks on a pmap lock, pmap_release() might proceed in parallel and destroy the pmap mutex, since unlocked pv lock allows to remove pv entry owned by the pmap. For now, gate the pmap_release() on write-locked pvh_global_lock. Since pmap_ts_release() does not unlock the global lock, pmap_release() would not destroy pmap mutex until the pmap_ts_referenced() finished. We cannot enter pmap_ts_referenced() and encounter a pv entry for the destroyed pmap if pmap_release() passed the global lock gate, since pmap_remove_pages() would finish earlier. Reported by: jeff, pho Reviewed by: alc Tested by: pho Sponsored by: The FreeBSD Foundation END K 10 svn:author V 6 davide K 8 svn:date V 27 2013-08-18T22:04:52.529619Z K 7 svn:log V 163 Garbage collect the old calloutng repository, now that everything was committed to HEAD. Thanks to Peter Wemm who pointed out to me the existence of 'svn delete'. END K 10 svn:author V 6 davide K 8 svn:date V 27 2013-08-18T22:20:53.978623Z K 7 svn:log V 59 Create a branch for umtx adaptive spinning implementation. END K 10 svn:author V 6 davide K 8 svn:date V 27 2013-08-18T23:12:16.060523Z K 7 svn:log V 402 Introduce a facility to share the thread state between kernel and userland. The API is the same of Solaris' schedctl(), even though the implementation is probably different. There are still some rough edges (e.g. structures are not cleaned up upon thread exit/proc exit) but I'll have some patches to address them and I'm going to fix in the next days after some testing. Discussed with: kib, attilio END K 10 svn:author V 6 davide K 8 svn:date V 27 2013-08-18T23:15:29.294500Z K 7 svn:log V 45 Commit the actual schedctl() implementation. END K 10 svn:author V 6 adrian K 8 svn:date V 27 2013-08-18T23:40:30.190686Z K 7 svn:log V 190 Don't return ENOTSUPP here - the net80211 pluggable ioctl API will treat this as the final item in the linker set and not try others. This stopped the fast frames IOCTLs from being called. END K 10 svn:author V 3 ian K 8 svn:date V 27 2013-08-19T01:29:13.855919Z K 7 svn:log V 648 Allow a hardware driver to pass clock frequencies into the sdhci driver. The sdhci spec says that if the base or timeout clock frequency in the capabilities register is zero, the driver must obtain the frequency "from another source." This change defines that other source to be the low-level hardware driver, which can pre-set the frequencies in slot.max_clk and slot.timeout_clk before calling sdhci_init_slot(). This helps with a growing number of SoCs that have sdhci base clock frequencies that either won't fit into the range allowed by the number of bits available in the capabilities register, or the frequency is runtime- configurable. END K 10 svn:author V 5 peter K 8 svn:date V 27 2013-08-19T04:56:03.570602Z K 7 svn:log V 112 Add the optional ability to run as a different user. Obtained from: Antique freebsd.org cluster archive images END K 10 svn:author V 8 jhibbits K 8 svn:date V 27 2013-08-19T05:10:46.038473Z K 7 svn:log V 69 Fix some ppc64 dtrace bugs, and enable systrace_freebsd32 for ppc64. END K 10 svn:author V 5 peter K 8 svn:date V 27 2013-08-19T05:12:38.926470Z K 7 svn:log V 16 Revert r254508. END K 10 svn:author V 5 peter K 8 svn:date V 27 2013-08-19T05:37:49.938800Z K 7 svn:log V 103 Pass pidfile to bsnmpd if it's been changed (parts cut/pasted from rc.d/rarpd and rc.d/wpa_supplicant) END K 10 svn:author V 6 rpaulo K 8 svn:date V 27 2013-08-19T05:48:42.613134Z K 7 svn:log V 31 Style changes and typos fixed. END K 10 svn:author V 6 emaste K 8 svn:date V 27 2013-08-19T07:00:13.903152Z K 7 svn:log V 54 MFC r251192: Fix misspelling of structure field name. END K 10 svn:author V 6 andrew K 8 svn:date V 27 2013-08-19T08:28:35.367679Z K 7 svn:log V 33 Enable VFP on the Arndale Board. END K 10 svn:author V 5 andre K 8 svn:date V 27 2013-08-19T09:49:51.654705Z K 7 svn:log V 2443 MFC a bundle of commits that bring autotuning to mbufs, maxfiles/sockets and maxusers to the 9-stable branch. It is committed as bundle because these patches build on each other and only provide the functionality in their entirety. Some are bug fixes to aspects of earlier commits. MFC r242029 (alfred): Allow autotune maxusers > 384 on 64 bit machines. MFC r242847 (alfred): Allow maxusers to scale on machines with large address space. MFC r243631 (andre): Base the mbuf related limits on the available physical memory or kernel memory, whichever is lower. The overall mbuf related memory limit must be set so that mbufs (and clusters of various sizes) can't exhaust physical RAM or KVM. At the same time divorce maxfiles from maxusers and set maxfiles to physpages / 8 with a floor based on maxusers. This way busy servers can make use of the significantly increased mbuf limits with a much larger number of open sockets. MFC r243639 (andre): Complete r243631 by applying the remainder of kern_mbuf.c that got lost while merging into the commit tree. MFC r243668 (andre): Using a long is the wrong type to represent the realmem and maxmbufmem variable as they may overflow on i386/PAE and i386 with > 2GB RAM. MFC r243995, r243996, r243997 (pjd): Style cleanups, Make use of the fact that uma_zone_set_max(9) already returns actual limit set. MFC r244080 (andre): Prevent long type overflow of realmem calculation on ILP32 by forcing calculation to be in quad_t space. Fix style issue with second parameter to qmin(). MFC r245469 (alfred): Do not autotune ncallout to be greater than 18508. MFC r245575 (andre): Move the mbuf memory limit calculations from init_param2() to tunable_mbinit() where it is next to where it is used later. MFC r246207 (andre): Remove unused VM_MAX_AUTOTUNE_NMBCLUSTERS define. MFC r249843 (andre): Base the calculation of maxmbufmem in part on kmem_map size instead of kernel_map size to prevent kernel memory exhaustion by mbufs and a subsequent panic on physical page allocation failure. MFC r253204 (andre): Fix style issues, a typo in "kern.ipc.nmbufs" and correctly plave and expose the value of the tunable maxmbufmem as "kern.ipc.maxmbufmem" through sysctl. MFC r253207 (andre): Make use of the fact that uma_zone_set_max(9) already returns the rounded limit making a call to uma_zone_get_max(9) unnecessary. Tested by: alfred (iXsystems) END K 10 svn:author V 5 andre K 8 svn:date V 27 2013-08-19T10:20:20.953450Z K 7 svn:log V 115 Remove unused and incomplete support for delayed fragment checksums from bce(4), bxe(4), mge(4) and ti(4) drivers. END K 10 svn:author V 5 andre K 8 svn:date V 27 2013-08-19T10:30:15.871162Z K 7 svn:log V 208 Remove unused M_FRAG, M_FIRSTFRAG and M_LASTFRAG tagging from ip_fragment(). There wasn't any real driver (and hardware) support for it. Modern hardware does full fragmentation/segmentation offload instead. END K 10 svn:author V 5 andre K 8 svn:date V 27 2013-08-19T10:34:10.768595Z K 7 svn:log V 213 Move ip_reassemble()'s use of the global M_FRAG mbuf flag to a protocol layer specific flag instead. The flag is only relevant while the packet stays in the IP reassembly queue. Discussed with: trociny, glebius END K 10 svn:author V 5 andre K 8 svn:date V 27 2013-08-19T11:08:36.301212Z K 7 svn:log V 300 Move the global M_SKIP_FIREWALL mbuf flags to a protocol layer specific flag instead. The flag is only used within the IP and IPv6 layer 3 protocols. Because some firewall packages treat IPv4 and IPv6 packets the same the flag should have the same value for both. Discussed with: trociny, glebius END