ƒ³8209173 184 1801 218 170 139 404 609 186 167 167 167 326 122 196 237 201 323 256 140 483 166 620 276 718 197 182 422 217 183 165 237 150 173 497 148 262 181 216 328 135 266 533 257 536 1569 358 261 551 454 484 441 204 503 233 135 131 182 236 140 366 245 304 380 423 269 131 239 316 220 273 273 333 383 110 187 229 358 160 255 316 216 149 175 186 200 164 168 165 179 211 351 1181 582 305 431 414 122 179 178 996 203 287 1465 456 209 204 228 175 196 211 184 203 236 601 138 121 140 202 196 148 155 432 764 256 148 141 270 392 236 158 365 161 437 237 597 187 176 270 210 542 161 169 334 314 266 249 249 320 135 148 337 270 239 680 320 348 327 191 166 226 189 750 174 239 195 409 393 369 167 150 156 199 218 172 272 162 118 219 317 218 218 115 143 209 246 K 10 svn:author V 3 alc K 8 svn:date V 27 2010-06-14T19:54:19.761660Z K 7 svn:log V 1707 Eliminate checks for a page having a NULL object in vm_pageout_scan() and vm_pageout_page_stats(). These checks were recently introduced by the first page locking commit, r207410, but they are not needed. At the same time, eliminate some redundant accesses to the page's object field. (These accesses should have neen eliminated by r207410.) Make the assertion in vm_page_flag_set() stricter. Specifically, only managed pages should have PG_WRITEABLE set. Add a comment documenting an assertion to vm_page_flag_clear(). It has long been the case that fictitious pages have their wire count permanently set to one. Add comments to vm_page_wire() and vm_page_unwire() documenting this. Add assertions to these functions as well. Update the comment describing vm_page_unwire(). Much of the old comment had little to do with vm_page_unwire(), but a lot to do with _vm_page_deactivate(). Move relevant parts of the old comment to _vm_page_deactivate(). Only pages that belong to an object can be paged out. Therefore, it is pointless for vm_page_unwire() to acquire the page queues lock and enqueue such pages in one of the paging queues. Generally speaking, such pages are immediately freed after the call to vm_page_unwire(). Previously, it was the call to vm_page_free() that reacquired the page queues lock and removed these pages from the paging queues. Now, we will never acquire the page queues lock for this case. (It is also worth noting that since both vm_page_unwire() and vm_page_free() occurred with the page locked, the page daemon never saw the page with its object field set to NULL.) Change the panic with vm_page_unwire() to provide a more precise message. Reviewed by: kib@ END K 10 svn:author V 4 jkim K 8 svn:date V 27 2010-06-14T20:08:26.965238Z K 7 svn:log V 124 Fix ACPI suspend/resume on amd64, which was broken since r208833. We need actual storage for FPU state to save and restore. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-14T21:01:13.461223Z K 7 svn:log V 78 Eliminate dead code. Found by: Coverity Prevent CID: 5158 MFC after: 3 days END K 10 svn:author V 8 rdivacky K 8 svn:date V 27 2010-06-14T21:11:23.748862Z K 7 svn:log V 42 Fix a mismerge and get us closer to head. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-14T21:18:58.632967Z K 7 svn:log V 311 Remove macros that are not really needed. The idea was to have them in case we grow more descriptors, but I'll reconsider readding them once we get there. Passing (a = b) expression to FD_ISSET() is bad idea, as FD_ISSET() evaluates its argument twice. Found by: Coverity Prevent CID: 5243 MFC after: 3 days END K 10 svn:author V 6 tuexen K 8 svn:date V 27 2010-06-14T21:25:07.069989Z K 7 svn:log V 513 * Fix a bug where the length of the ASCONF-ACK was calculated wrong due to using an uninitialized variable. * Fix a bug where a NULL pointer was dereferenced when interfaces come and go at a high rate. * Fix a bug where inps where not deregistered from iterators. * Fix a race condition in freeing an association. * Fix a refcount problem related to the iterator. Each of the above bug results in a panic. It shows up when interfaces come and go at a high rate. Obtained from: rrs (partly) MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-14T21:25:20.315077Z K 7 svn:log V 94 Plug memory leaks. Found by: Coverity Prevent CID: 7052, 7053, 7054, 7055 MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-14T21:33:18.003564Z K 7 svn:log V 75 Plug memory leak. Found by: Coverity Prevent CID: 7051 MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-14T21:37:25.412749Z K 7 svn:log V 75 Plug memory leak. Found by: Coverity Prevent CID: 7056 MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-14T21:41:22.218847Z K 7 svn:log V 75 Plug memory leak. Found by: Coverity Prevent CID: 7057 MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-14T21:44:20.883469Z K 7 svn:log V 233 Initialize gctl_seq for synchronization requests. Reported by: hiroshi@soupacific.com Analysed by: Mikolaj Golub Tested by: hiroshi@soupacific.com, Mikolaj Golub MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-14T21:44:58.157524Z K 7 svn:log V 30 Fix typos. MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-14T21:46:48.289982Z K 7 svn:log V 103 Correct various log messages. Submitted by: Mikolaj Golub MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-14T21:56:24.246809Z K 7 svn:log V 144 BIO_DELETE contains range we want to delete and doesn't provide any useful data, so there is no need to copy it to userland. MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-14T21:58:55.695648Z K 7 svn:log V 108 'unit' can be negative, so use signed type for it. Found by: Coverity Prevent CID: 3731 MFC after: 3 days END K 10 svn:author V 3 ken K 8 svn:date V 27 2010-06-14T22:02:18.437731Z K 7 svn:log V 230 For the target port groups structures, don't allocate the initial element. This makes things easier for target implementations to calculate how many elements they need to allocate. Discussed with: mjacob, gibbs MFC after: 1 week END K 10 svn:author V 4 jkim K 8 svn:date V 27 2010-06-14T23:01:50.968247Z K 7 svn:log V 162 Fix typos that broke duration calculations on protection frames. A similar fix was done for ral(4) long ago and it must be copy-and-paste bugs. Found by: clang END K 10 svn:author V 6 emaste K 8 svn:date V 27 2010-06-14T23:51:35.599434Z K 7 svn:log V 45 Use literal format strings. Found by clang. END K 10 svn:author V 8 rmacklem K 8 svn:date V 27 2010-06-15T00:25:04.148719Z K 7 svn:log V 385 Add MODULE_DEPEND() macros to the experimental NFS client and server so that the modules will load when kernels are built with none of the NFS* configuration options specified. I believe this resolves the problems reported by PR kern/144458 and the email on freebsd-stable@ posted by Dmitry Pryanishnikov on June 13. Tested by: kib PR: kern/144458 Reviewed by: kib MFC after: 1 week END K 10 svn:author V 2 ed K 8 svn:date V 27 2010-06-15T04:47:16.104112Z K 7 svn:log V 75 err() takes a printf format. Submitted by: Pawel Worach Spotted by: clang END K 10 svn:author V 3 avg K 8 svn:date V 27 2010-06-15T07:06:54.769699Z K 7 svn:log V 527 sound/pcm: use non-const string as a value with SYSCTL_STRING Although the sysctls are marked with CTLFLAG_RD and the values will stay immutable, current sysctl implementation stores value pointer in void* type, which means that const qualifier is discarded anyway and some newer compilers complaint about that. We can't use de-const trick in sysctl implementation, because in that case we could miss an opposite situation where a const value is used with CTLFLAG_RW sysctl. Complaint from: gcc 4.4, clang MFC after: 2 weeks END K 10 svn:author V 2 ae K 8 svn:date V 27 2010-06-15T08:53:13.417252Z K 7 svn:log V 184 * Include sys/systm.h for KASSERT() * Remove unneeded includes and comment * Replace home made OFFSETOF() macro with standard offsetof() Pointed out by: bde Approved by: kib (mentor) END K 10 svn:author V 8 netchild K 8 svn:date V 27 2010-06-15T08:58:16.888972Z K 7 svn:log V 620 Add a periodic zfs scrub script. Features: - configurable amount of days between scrubs (default value or per pool) - do not scrub directly after pool creation (respects the configured number of days between scrubs) - do not scrub if a scrub is in progress - tells how to see the status of the scrub - tells how many days since the last scrub if it skips the scrubbing - warns if a non-existent pool is specified explicitely (default: no pools specified -> all currently imported pools are handled) - runs late in the periodic run to not slow down the other periodic daily scripts Discussed on: fs@ END K 10 svn:author V 7 fabient K 8 svn:date V 27 2010-06-15T09:12:31.257915Z K 7 svn:log V 100 Set the comment at the right place for PMC uncore classes. Submitted by: rstone MFC after: 1 month END K 10 svn:author V 3 kib K 8 svn:date V 27 2010-06-15T09:18:27.148636Z K 7 svn:log V 90 Add assert to check that the (current) thread is in critical section. MFC after: 1 month END K 10 svn:author V 3 kib K 8 svn:date V 27 2010-06-15T09:19:33.461085Z K 7 svn:log V 329 Use critical sections instead of disabling local interrupts to ensure the consistency between PCPU fpcurthread and the state of the FPU. Explicitely assert that the calling conventions for fpudrop() are adhered too. In cpu_thread_exit(), add missed critical section entrance. Reviewed by: bde Tested by: pho MFC after: 1 month END K 10 svn:author V 3 kib K 8 svn:date V 27 2010-06-15T09:30:36.288944Z K 7 svn:log V 124 Fix the syscall module name after r205320. Submitted by: Vladislav Movchan MFC after: 1 week END K 10 svn:author V 2 ae K 8 svn:date V 27 2010-06-15T10:01:49.549106Z K 7 svn:log V 92 Make all tests in libdialog compilable. Fix coredump in menu3.c. Approved by: kib (mentor) END K 10 svn:author V 5 edwin K 8 svn:date V 27 2010-06-15T11:37:33.212960Z K 7 svn:log V 71 Make mandoc/mdocml happy. Submitted by: Uli(?) via gmane.comp.time.tz END K 10 svn:author V 8 rdivacky K 8 svn:date V 27 2010-06-15T13:08:03.747019Z K 7 svn:log V 139 Use __noinline instead of __attribute__((noinline)). Remove this from strcmp() as it is used only once. Suggested by: ed, Benjamin Kramer END K 10 svn:author V 8 rdivacky K 8 svn:date V 27 2010-06-15T13:13:12.490751Z K 7 svn:log V 53 make memcpy static and fix the previous commit style END K 10 svn:author V 3 kib K 8 svn:date V 27 2010-06-15T14:59:35.581848Z K 7 svn:log V 81 Rename CRITSECT_ASSERT to CRITICAL_ASSERT. Suggested by: jhb MFC after: 1 month END K 10 svn:author V 2 ed K 8 svn:date V 27 2010-06-15T17:08:03.974761Z K 7 svn:log V 405 Remove unneeded files from the build. I used the following command to determine which source files were unneeded: | for i in `find lib/clang -name '*.o'` | do | MATCHES="`(nm -g --defined-only $i; nm -g --defined-only \ | usr.bin/clang/clang/clang) | sed -e 's/.* //' | \ | sort | uniq -d | wc -l`" | [ $MATCHES -eq 0 ] && echo "$i: unneeded" | done This should slightly improve the build times. END K 10 svn:author V 2 ed K 8 svn:date V 27 2010-06-15T17:28:16.259362Z K 7 svn:log V 57 Remove even two more unneeded files from libllvmsupport. END K 10 svn:author V 3 imp K 8 svn:date V 27 2010-06-15T18:11:37.099810Z K 7 svn:log V 169 make these convenience functions more convenient by accepting all args, not just the first. makes mechanical conversion of old style more forgiving. MFC after: 3 days END K 10 svn:author V 3 kib K 8 svn:date V 27 2010-06-15T18:16:04.653886Z K 7 svn:log V 89 Remove two obsoleted comments, add a note about 32bit compatibility. MFC after: 1 month END K 10 svn:author V 3 imp K 8 svn:date V 27 2010-06-15T18:16:45.027859Z K 7 svn:log V 123 Create a make.conf. not needed for runtime, but some ports want to spam it at compile or install time. MFC after: 3 days END K 10 svn:author V 3 imp K 8 svn:date V 27 2010-06-15T18:18:31.381674Z K 7 svn:log V 235 Allow population of /cfg and /data. Begin the move to making all slice creation overrideable too, but there's a few problems doing that for the duplicated partitions (s1 and s2), so just comment that it needs work. MFC after: 3 days END K 10 svn:author V 3 alc K 8 svn:date V 27 2010-06-15T18:37:31.974756Z K 7 svn:log V 43 Eliminate unnecessary page queues locking. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2010-06-15T18:51:41.878132Z K 7 svn:log V 173 Restore the machine check register banks on resume. For banks being monitored via CMCI, reset the interrupt threshold to 1 on resume. Reviewed by: jkim MFC after: 2 weeks END K 10 svn:author V 3 jhb K 8 svn:date V 27 2010-06-15T19:14:39.727173Z K 7 svn:log V 440 When updating individual CPU's lowest Cx state to use, never set it to a state lower than the lowest one supported by the current CPU. This closes some races with changes to the hw.acpi.cpu_cx_lowest sysctl while Cx states for individual CPUs were changing (e.g. unplugging the AC adapter of a laptop) that could result in panics. Submitted by: Giovanni Trematerra Tested by: David Demelier demelier dot david of gmail MFC after: 3 days END K 10 svn:author V 3 mav K 8 svn:date V 27 2010-06-15T19:19:04.557423Z K 7 svn:log V 164 When Emulate3Button is active, do not set select() timeout in states when it is not needed. No need to kick CPU every 20ms without a purpose. Reviewed by: philip@ END K 10 svn:author V 6 sbruno K 8 svn:date V 27 2010-06-15T19:28:37.765527Z K 7 svn:log V 440 Add a new column to the output of vmstat -z to indicate the number of times the system was forced to sleep when requesting a new allocation. Expand the debugger hook, db_show_uma, to display these results as well. This has proven to be very useful in out of memory situations when it is not known why systems have become sluggish or fail in odd ways. Reviewed by: rwatson alc Approved by: scottl (mentor) peter Obtained from: Yahoo Inc. END K 10 svn:author V 4 jkim K 8 svn:date V 27 2010-06-15T19:28:44.342340Z K 7 svn:log V 1474 Implement flexible BPF timestamping framework. - Allow setting format, resolution and accuracy of BPF time stamps per listener. Previously, we were only able to use microtime(9). Now we can set various resolutions and accuracies with ioctl(2) BIOCSTSTAMP command. Similarly, we can get the current resolution and accuracy with BIOCGTSTAMP command. Document all supported options in bpf(4) and their uses. - Introduce new time stamp 'struct bpf_ts' and header 'struct bpf_xhdr'. The new time stamp has both 64-bit second and fractional parts. bpf_xhdr has this time stamp instead of 'struct timeval' for bh_tstamp. The new structures let us use bh_tstamp of same size on both 32-bit and 64-bit platforms without adding additional shims for 32-bit binaries. On 64-bit platforms, size of BPF header does not change compared to bpf_hdr as its members are already all 64-bit long. On 32-bit platforms, the size may increase by 8 bytes. For backward compatibility, struct bpf_hdr with struct timeval is still the default header unless new time stamp format is explicitly requested. However, the behaviour may change in the future and all relevant code is wrapped around "#ifdef BURN_BRIDGES" for now. - Add experimental support for tagging mbufs with time stamps from a lower layer, e.g., device driver. Currently, mbuf_tags(9) is used to tag mbufs. The time stamps must be uptime in 'struct bintime' format as binuptime(9) and getbinuptime(9) do. Reviewed by: net@ END K 10 svn:author V 6 jilles K 8 svn:date V 27 2010-06-15T21:00:53.323531Z K 7 svn:log V 262 libedit: Fix a bug that could make completion listings incomplete. The element matches[0] is the common prefix and is not counted in len, so subtracting 1 is not needed. A counter for the number of matches per line was incremented twice. Submitted by: Guy Yur END K 10 svn:author V 3 jfv K 8 svn:date V 27 2010-06-15T21:11:51.099556Z K 7 svn:log V 168 Change to have legacy interrupts use the same handler had a flaw, thanks to John Baldwin for finding it. Change which queue legacy tasks are enqueued on. MFC: soonest END K 10 svn:author V 6 jilles K 8 svn:date V 27 2010-06-15T21:34:57.534333Z K 7 svn:log V 455 libedit: Allow simple quoting in filename completion. The completer recognizes characters escaped with backslashes as being literal parts of a word, and adds backslashes to avoid almost all misinterpretation. In particular, filenames containing spaces can be completed correctly. For bug compatibility with the NetBSD version, the improved completion function has a new name, _el_fn_sh_complete, and _el_fn_complete is unchanged. Submitted by: Guy Yur END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-15T21:46:07.257575Z K 7 svn:log V 353 Implement a page table spilling mechanism to allow the kernel to continue when user-space applications have filled all the slots in a PTEG. This relies on VM gracefully remapping pages when it gets faults on pages it thinks are mapped already, which it seems to do in limited forkbomb-based testing. Comments by alc appreciated before merging to HEAD. END K 10 svn:author V 6 jilles K 8 svn:date V 27 2010-06-15T21:58:40.504446Z K 7 svn:log V 388 sh: Add filename completion. This uses the new libedit completion function with quoting support. Unlike NetBSD, there is no 'set +o tabcomplete' option to disable completion. I do not see any reason for such a special treatment, as completion is rather useful and it is possible to do bind ^I ed-insert to disable completion and insert a tab character instead. Submitted by: Guy Yur END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-15T22:01:38.037062Z K 7 svn:log V 340 Modify the console mouse pointer drawing routine to use single-byte writes instead of 4-byte ones. Because the mouse pointer can start part way through a character cell, 4-byte memory operations are not necessarily aligned, triggering a fatal alignment exception when the console pointer was moved on PowerPC G5 systems. MFC after: 3 days END K 10 svn:author V 6 cognet K 8 svn:date V 27 2010-06-15T22:16:02.250344Z K 7 svn:log V 108 Turn off cache if there's more than one kernel mapping, and one is writable. Submitted by: Mark Tinguely END K 10 svn:author V 6 jilles K 8 svn:date V 27 2010-06-15T22:23:21.565992Z K 7 svn:log V 407 libedit: Reduce surprising behaviour with filename completion some more: * Quote '*', '?' and '['. While it may be more useful to expand them to matching pathnames, this at least matches with the completion we do. * '@' is a regular character for filenames. Some other shells do @ completion but we do not. * Prefix names starting with '-' and '+' with './' so they are not seen as options. END K 10 svn:author V 3 imp K 8 svn:date V 27 2010-06-15T23:55:42.558529Z K 7 svn:log V 140 Allow boot.config to have different values than just -h. Submitted by: bsdrp by way of freenas and olivier cochard-labbe MFC after: 3 days END K 10 svn:author V 3 alc K 8 svn:date V 27 2010-06-16T00:41:21.611167Z K 7 svn:log V 43 Eliminate unnecessary page queues locking. END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-16T02:50:16.938459Z K 7 svn:log V 31 IFC to pull in syscons changes END K 10 svn:author V 3 avg K 8 svn:date V 27 2010-06-16T07:52:44.936206Z K 7 svn:log V 90 device.hints: do install when WITHOUT_BOOT is set Discussed with: imp MFC after: 2 weeks END K 10 svn:author V 3 avg K 8 svn:date V 27 2010-06-16T08:20:01.346253Z K 7 svn:log V 143 device.hints: do not install when WITHOUT_BOOT is set Forced commit to fix commit message in r209228. Discussed with: imp MFC after: 2 weeks END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-16T12:42:20.999455Z K 7 svn:log V 48 Remove redundant assignment. MFC after: 3 days END K 10 svn:author V 8 jchandra K 8 svn:date V 27 2010-06-16T12:55:14.627717Z K 7 svn:log V 268 Merge jmallett@'s n64 work into HEAD - changeset 1. Update libc assembly code to use macros that work on both o32 and n64. Merge string functions from NetBSD. The changes are from http://svn.freebsd.org/base/user/jmallett/octeon Approved by: rrs (mentor), jmallett END K 10 svn:author V 3 raj K 8 svn:date V 27 2010-06-16T14:10:39.308410Z K 7 svn:log V 152 Move ARM nexus rman initialization to attach routine. This fixes a panic, which started to trigger after r209129 cleanup. Submitted by: Andrew Turner END K 10 svn:author V 8 jchandra K 8 svn:date V 27 2010-06-16T14:13:36.586477Z K 7 svn:log V 206 Merge jmallett@'s n64 work into HEAD - changeset 2 Update libc Makefiles. Add makecontext implementation. Changes from http://svn.freebsd.org/base/user/jmallett/octeon Approved by: rrs(mentor), jmallett END K 10 svn:author V 3 mav K 8 svn:date V 27 2010-06-16T15:09:45.568560Z K 7 svn:log V 287 Freq sysctls are quite heavy due to set of malloc()/free() calls. Avoid reading current frequency on every period. Instead do it only after changing and periodically from time to time if somebody else change it. Also dynamically decrease sampling frequency up to 4 times on inactivity, END K 10 svn:author V 6 brucec K 8 svn:date V 27 2010-06-16T15:40:13.918970Z K 7 svn:log V 327 * Allow partial MB/GB values to be entered in the slice and label editors. * Don't strdup the name when calling deviceRegister because the string is copied within new_device. * Use a subtype of 165, not 3, when creating a slice in noninteractive mode. PR: bin/135333 PR: bin/66350 Approved by: rrs (mentor) MFC after: 1 month END K 10 svn:author V 6 brucec K 8 svn:date V 27 2010-06-16T15:49:17.641340Z K 7 svn:log V 173 Call free and freeaddrinfo before exiting. PR: bin/144730 PR: bin/144974 Submitted by: Earl R. Lapus Approved by: rrs (mentor) MFC after: 1 month END K 10 svn:author V 2 jh K 8 svn:date V 27 2010-06-16T16:02:17.935076Z K 7 svn:log V 40 Correct the function name in a KASSERT. END K 10 svn:author V 3 jfv K 8 svn:date V 27 2010-06-16T16:37:36.387984Z K 7 svn:log V 146 Changes from John Baldwin adding to last commit, change rxeof api for poll friendliness, and eliminate unnecessary link tasklet use. Thanks John! END K 10 svn:author V 8 jchandra K 8 svn:date V 27 2010-06-16T16:39:01.665133Z K 7 svn:log V 218 Merge jmallett@'s n64 work into HEAD - changeset 3 Update rtld MIPS code to use assembly macros that work on o32/n64. Changes from http://svn.freebsd.org/base/user/jmallett/octeon Approved by: rrs(mentor), jmallett END K 10 svn:author V 3 raj K 8 svn:date V 27 2010-06-16T16:52:35.810481Z K 7 svn:log V 127 Use awk alone for extracting FDT_DTS_FILE instead of grep/cut, which is simpler and more robust. Submitted by: Andrzej Tobola END K 10 svn:author V 3 gnn K 8 svn:date V 27 2010-06-16T17:36:53.181744Z K 7 svn:log V 180 Move statistics into the sysctl tree making it easier to find and use them. Add previously hidden statistics, some of which include interrupt and host/card communication counters. END K 10 svn:author V 3 gnn K 8 svn:date V 27 2010-06-16T20:57:41.909652Z K 7 svn:log V 180 Move statistics into the sysctl tree making it easier to find and use them. Add previously hidden statistics, some of which include interrupt and host/card communication counters. END K 10 svn:author V 8 jchandra K 8 svn:date V 27 2010-06-17T05:03:01.955509Z K 7 svn:log V 235 Merge jmallett@'s n64 work into HEAD - changeset 4 Re-write tlb operations in C with a simpler API. Update callers to use the new API. Changes from http://svn.freebsd.org/base/user/jmallett/octeon Approved by: rrs(mentor), jmallett END K 10 svn:author V 2 ed K 8 svn:date V 27 2010-06-17T08:49:31.364482Z K 7 svn:log V 291 Remove the unit argument from the recently added make_dev_p(). New code that creates character devices shouldn't use device unit numbers, but only si_drv[12] to hold pointer to per-device data. Make this function more future proof by removing the unit number argument. Discussed with: kib END K 10 svn:author V 2 ae K 8 svn:date V 27 2010-06-17T09:36:23.941877Z K 7 svn:log V 19 Rewrap long lines. END K 10 svn:author V 2 ae K 8 svn:date V 27 2010-06-17T09:43:07.845101Z K 7 svn:log V 96 Clean types. Replace "unsigned long long" to "off_t". Add a temporary make knob LIBDIALOG_PATH. END K 10 svn:author V 3 avg K 8 svn:date V 27 2010-06-17T10:15:13.724416Z K 7 svn:log V 136 lock_profile_release_lock: do not compare unsigned with zero Found by: Coverity Prevent CID: 3660 Reviewed by: jhb MFC after: 2 weeks END K 10 svn:author V 3 mav K 8 svn:date V 27 2010-06-17T11:54:49.783552Z K 7 svn:log V 265 Merge COUNT_XINVLTLB_HITS and COUNT_IPIS kernel options from i386 to amd64. This information can be very valuable for CPU sleep-time (and respectively idle power consumption) optimization. Add counters for timer-related IPIs. Reviewed by: jhb@ (previous version) END K 10 svn:author V 5 gabor K 8 svn:date V 27 2010-06-17T12:05:47.062507Z K 7 svn:log V 66 - Add c99(7) manual page - Add MLINKS to c.7, c79.7, c89.7, c90.7 END K 10 svn:author V 8 netchild K 8 svn:date V 27 2010-06-17T12:25:47.430860Z K 7 svn:log V 157 - add the zfs scrub script - move the zfs status script into the MK_ZFS conditional to respect WITHOUT_ZFS Noticed by: Andrzej Tobola END K 10 svn:author V 2 ae K 8 svn:date V 27 2010-06-17T12:30:41.274876Z K 7 svn:log V 224 Use getmntopts() from sbin/mount for parsing fstab mntops. Add dump frequency and pass number fields to the struct de_fs. Properly initialize mount options checkboxes of "Select Mount Options" dialog. Wrap some long lines. END K 10 svn:author V 3 kib K 8 svn:date V 27 2010-06-17T12:35:17.252677Z K 7 svn:log V 123 In the ia32_{get,set}_fpcontext(), use fpu{get,set}userregs instead of fpu{get,set}regs. Noted by: bde MFC after: 1 month END K 10 svn:author V 8 netchild K 8 svn:date V 27 2010-06-17T12:37:50.478813Z K 7 svn:log V 52 Add the ZFS periodic daily scripts to the ZFS part. END K 10 svn:author V 5 gabor K 8 svn:date V 27 2010-06-17T13:59:41.449459Z K 7 svn:log V 81 - Fix typo, it should have been c78.7 Submitted by: pluknet END K 10 svn:author V 5 gabor K 8 svn:date V 27 2010-06-17T14:37:47.718441Z K 7 svn:log V 92 - Update GCC reference from 3.3 to 4.2 [1] - Add reference to c99(7) Submitted by: stefanf END K 10 svn:author V 2 jh K 8 svn:date V 27 2010-06-17T16:12:06.619753Z K 7 svn:log V 108 - Fix compilation of the subr_unit.c user space test program. - Use %zu for size_t in a few format strings. END K 10 svn:author V 8 kensmith K 8 svn:date V 27 2010-06-17T17:06:52.770471Z K 7 svn:log V 67 Release the code freeze on stable/8. Approved by: core (implicit) END K 10 svn:author V 6 rpaulo K 8 svn:date V 27 2010-06-17T17:34:45.612928Z K 7 svn:log V 73 Make DTrace syscall provider work again by including opt_kdtrace.h here. END K 10 svn:author V 3 jfv K 8 svn:date V 27 2010-06-17T17:38:39.146007Z K 7 svn:log V 73 Two stats were duplicated, thanks to Andrew Boyer for pointing this out. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-17T17:39:51.686883Z K 7 svn:log V 87 Backout r207970 for now, it can lead to deadlocks. Reported by: kan MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-17T17:41:42.162512Z K 7 svn:log V 118 Turn off UMA allocations on all archs by default. It isn't stable even on amd64. Reported by: many MFC after: 3 days END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-17T19:06:11.579315Z K 7 svn:log V 258 MFC r209186,r209187: r209186: BIO_DELETE contains range we want to delete and doesn't provide any useful data, so there is no need to copy it to userland. r209187: 'unit' can be negative, so use signed type for it. Found by: Coverity Prevent CID: 3731 END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-17T19:17:31.219591Z K 7 svn:log V 1087 MFC r209175,r209177,r209179,r209180,r209181,r209182,r209183,r209184,r209185: r209175: Eliminate dead code. Found by: Coverity Prevent CID: 5158 r209177: Remove macros that are not really needed. The idea was to have them in case we grow more descriptors, but I'll reconsider readding them once we get there. Passing (a = b) expression to FD_ISSET() is bad idea, as FD_ISSET() evaluates its argument twice. Found by: Coverity Prevent CID: 5243 r209179: Plug memory leaks. Found by: Coverity Prevent CID: 7052, 7053, 7054, 7055 r209180: Plug memory leak. Found by: Coverity Prevent CID: 7051 r209181: Plug memory leak. Found by: Coverity Prevent CID: 7056 r209182: Plug memory leak. Found by: Coverity Prevent CID: 7057 r209183: Initialize gctl_seq for synchronization requests. Reported by: hiroshi@soupacific.com Analysed by: Mikolaj Golub Tested by: hiroshi@soupacific.com, Mikolaj Golub r209184: Fix typos. r209185: Correct various log messages. Submitted by: Mikolaj Golub END K 10 svn:author V 6 tuexen K 8 svn:date V 27 2010-06-17T19:17:50.429907Z K 7 svn:log V 486 MFC 209178 * Fix a bug where the length of the ASCONF-ACK was calculated wrong due to using an uninitialized variable. * Fix a bug where a NULL pointer was dereferenced when interfaces come and go at a high rate. * Fix a bug where inps where not deregistered from iterators. * Fix a race condition in freeing an association. * Fix a refcount problem related to the iterator. Each of the above bug results in a panic. It shows up when interfaces come and go at a high rate. END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-17T19:25:05.387894Z K 7 svn:log V 212 MFC r209260,r209261: r209260: Backout r207970 for now, it can lead to deadlocks. Reported by: kan r209261: Turn off UMA allocations on all archs by default. It isn't stable even on amd64. Reported by: many END K 10 svn:author V 6 sbruno K 8 svn:date V 27 2010-06-17T19:28:56.253366Z K 7 svn:log V 335 Expand man page to document the fact that mptutil/mpt doesn't support RAID volumes in excess of 2TB. Document workaround via geom or zfs Modified submitter's original patch to reference why this is broken and what to do to work around the issue. Submitted by: hubert@tournier.org PR: bin/147572 Reviewed by: jhb MFC after: 2 weeks END K 10 svn:author V 6 mlaier K 8 svn:date V 27 2010-06-17T19:46:36.552362Z K 7 svn:log V 318 MFC r209063: Cache the last result from if_indextoname for printing. This speeds up "arp -an" when using a lot of aliases (on a single interface). A better fix would include a better interface for if_indextoname than getting the whole address list from the kernel just to find the one index->name mapping. END K 10 svn:author V 7 delphij K 8 svn:date V 27 2010-06-17T19:48:03.463604Z K 7 svn:log V 26 Remove an unused comment. END K 10 svn:author V 6 marius K 8 svn:date V 27 2010-06-17T20:10:39.730310Z K 7 svn:log V 84 MFC: r209138 Update a branch missed in r207537 (committed to stable/8 in r207890). END K 10 svn:author V 6 marius K 8 svn:date V 27 2010-06-17T20:10:41.823494Z K 7 svn:log V 83 MFC: r209138 Update a branch missed in r207537 (commited to stable/7 in r207891). END K 10 svn:author V 6 mlaier K 8 svn:date V 27 2010-06-17T20:38:18.323627Z K 7 svn:log V 900 MFC r209063: Cache the last result from if_indextoname for printing. This speeds up "arp -an" when using a lot of aliases (on a single interface). A better fix would include a better interface for if_indextoname than getting the whole address list from the kernel just to find the one index->name mapping. > Description of fields to fill in above: 76 columns --| > PR: If a GNATS PR is affected by the change. > Submitted by: If someone else sent in the change. > Reviewed by: If someone else reviewed your modification. > Approved by: If you needed approval for this commit. > Obtained from: If the change is from a third party. > MFC after: N [day[s]|week[s]|month[s]]. Request a reminder email. > Security: Vulnerability reference (one per line) or description. > Empty fields above will be automatically removed. _M arp M arp/arp.c END K 10 svn:author V 6 marius K 8 svn:date V 27 2010-06-17T20:53:56.296429Z K 7 svn:log V 107 MFC: r209138 Update a branch missed in r207537 (committed to stable/8 in r207890). Approved by: re (kib) END K 10 svn:author V 5 randi K 8 svn:date V 27 2010-06-17T21:17:35.061168Z K 7 svn:log V 192 MFC r209004: Fix uninitialized variables that cause a crash when the network is initialized and sysinstall is not running as init. Submitted by: Nick Mills Approved by: cperciva (mentor) END K 10 svn:author V 2 mm K 8 svn:date V 27 2010-06-17T22:38:23.606391Z K 7 svn:log V 1372 MFC r209093-r209101: MFC r209093: Fix unable to remove a file over NFS after hitting refquota limit OpenSolaris onnv rev: 8890:8c2bd5f17bf2 OpenSolaris Bug ID: 6798878 MFC r209094: Fix zfs destroy fails to free object in open context, stops up txg train OpenSolaris onnv rev: 9409:9dc3f17354ed OpenSolaris Bug ID: 6809683 MFC r209095: Fix incomplete resilvering after disk replacement (raidz) OpenSolaris onnv rev: 9434:3bebded7c76a OpenSolaris Bug ID: 6794570 MFC r209096: Fix vdev_probe() starvation brings txg train to a screeching halt OpenSolaris onnv rev: 9722:e3866bad4e96 OpenSolaris Bug ID: 6844069 MFC r209097: Fix ZFS panic deadlock: cycle in blocking chain via zfs_zget OpenSolaris onnv rev: 9774:0bb234ab2287 OpenSolaris Bug ID: 6788152 MFC r209098: Fix zpool resilver stalls with spa_scrub_thread in a 3 way deadlock OpenSolaris onnv rev: 9997:174d75a29a1c OpenSolaris Bug ID: 6843235 MFC r209099: Fix possible zfs panic on zpool import OpenSolaris onnv rev: 10040:38b25aeeaf7a OpenSolaris Bug ID: 6857012 MFC r209100: Fix panic in zfs_getsecattr OpenSolaris onnv rev: 10295:f7a18a1e9610 OpenSolaris Bug ID: 6870564 MFC r209101: Fix arc_read_done may try to byteswap undefined data (sparc related) OpenSolaris onnv rev: 10839:cf83b553a2ab OpenSolaris Bug ID: 6836714 Obtained from: OpenSolaris (multiple Bug IDs) Approved by: pjd, delphij (mentor) END K 10 svn:author V 2 mm K 8 svn:date V 27 2010-06-17T22:47:44.542084Z K 7 svn:log V 364 Import latest ARC change from OpenSolaris: - large ghost eviction causes high write latency - arc_adjust might adjust MRU unnecessarily - arc_adapt can lead to wild arc_p adjustment OpenSolaris onnv-revision: 12636:13b5d698941e Submitted by: avg Approved by: pjd, delphij (mentor) Obtained from: OpenSolaris (Bug ID 6950219, 6953403, 6951024) MFC after: 1 month END K 10 svn:author V 6 sbruno K 8 svn:date V 27 2010-06-18T01:17:16.943333Z K 7 svn:log V 113 Much closer approximation of the kernel's calculation of this value. Reviewed by: alc Obtained from: Yahoo Inc. END K 10 svn:author V 6 qingli K 8 svn:date V 27 2010-06-18T03:31:33.439325Z K 7 svn:log V 108 MFC r208553 This patch fixes the problem where proxy ARP entries cannot be added over the if_ng interface. END K 10 svn:author V 6 mjacob K 8 svn:date V 27 2010-06-18T03:59:07.105294Z K 7 svn:log V 132 This is an MFC of 208082 Make sure to check that the active provider pointer points to something before dereferencing the pointer. END K 10 svn:author V 6 mjacob K 8 svn:date V 27 2010-06-18T04:01:59.552459Z K 7 svn:log V 80 This is an MFC of 208101 Yet another potential dereference of a dead provider. END K 10 svn:author V 6 mjacob K 8 svn:date V 27 2010-06-18T04:43:22.733883Z K 7 svn:log V 100 This is an MFC of 207933 Deal sensibly with more than 26 sg devices. It isn't a complete solution. END K 10 svn:author V 6 mjacob K 8 svn:date V 27 2010-06-18T04:45:06.775303Z K 7 svn:log V 115 This is an MFC of 207938 Pick up the right change, not it's close cousin. The one previously submitted was wrong. END K 10 svn:author V 6 mjacob K 8 svn:date V 27 2010-06-18T04:51:17.799460Z K 7 svn:log V 89 This is an MFC of 207543 Print IR_RESYNC updates informatively. Obtained from: pluknet END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-18T06:24:40.057311Z K 7 svn:log V 110 MFC r208770: opensolaris_kmem.c is already part of opensolaris.ko which zfs.ko depends on. Reported by: avg END K 10 svn:author V 5 brian K 8 svn:date V 27 2010-06-18T07:19:30.064885Z K 7 svn:log V 141 MFC r208728 - Fix stuttering sequences and reverse ranges PR: 123635 Submitted by: Ulrich Spörlein, uqs at spoerlein dot net END K 10 svn:author V 5 dougb K 8 svn:date V 27 2010-06-18T07:37:50.299488Z K 7 svn:log V 506 Catch up with ports version 2.30: If the user specified --no-confirm don't build the list. With a lot of ports to update it can get quite large. If there are a very large number of ports installed the grep in iport_from_origin() can fail with "Argument list too long." It's also theoretically possible that the grep would fail for other reasons. So if it does, fall back to 'pkg_info -O'. My method is faster which is why I'm still trying it first, but this is a sort of "belt and suspenders" situation. END K 10 svn:author V 5 dougb K 8 svn:date V 27 2010-06-18T08:11:52.029240Z K 7 svn:log V 44 Add the AAAA address for i.root-servers.net END K 10 svn:author V 5 dougb K 8 svn:date V 27 2010-06-18T08:38:56.254776Z K 7 svn:log V 27 Unset build_l in pm_make() END K 10 svn:author V 5 dougb K 8 svn:date V 27 2010-06-18T08:51:20.080234Z K 7 svn:log V 46 build_l also needs to be unset in pm_make_s() END K 10 svn:author V 6 tuexen K 8 svn:date V 27 2010-06-18T09:01:44.541238Z K 7 svn:log V 106 Fix a rece condition in the shutdown handling. The race condition resulted in a panic. MFC after: 3 days END K 10 svn:author V 3 kib K 8 svn:date V 27 2010-06-18T09:25:00.880545Z K 7 svn:log V 103 MFC r208340: When waiting for the busy page, do not unlock the object unless unlock cannot be avoided. END K 10 svn:author V 3 kib K 8 svn:date V 27 2010-06-18T09:29:08.649700Z K 7 svn:log V 56 MFC r208565: Remove a check that should be always true. END K 10 svn:author V 3 kib K 8 svn:date V 27 2010-06-18T09:34:31.747136Z K 7 svn:log V 63 MFC r208604: Clarify a potential issue in get_fpcontext() use. END K 10 svn:author V 3 kib K 8 svn:date V 27 2010-06-18T09:40:01.170679Z K 7 svn:log V 339 MFC r208773: Sometimes vnodes share the lock despite being different vnodes on different mount points, e.g. the nullfs vnode and the covered vnode from the lower filesystem. In this case, existing assertion in vop_rename_pre() may be triggered. Check for vnode locks equiality instead of the vnodes itself to not trip over the situation. END K 10 svn:author V 3 kib K 8 svn:date V 27 2010-06-18T11:09:51.007898Z K 7 svn:log V 671 Often reported issue with newer ld is: error in /usr/lib/crtendS.o(.eh_frame); no .eh_frame_hdr table will be created. The issue is that crtend is compiled with unwind table, and also it places the special CIE into the .eh_frame indicating the end of section, that is located before generated unwind table. New ld has assertion that verifies that closing CIE is indeed the last CIE, causing the crypting message to be issued, and refusing to generate dwarf unwind. Add -fno-asynchronous-unwind-tables to disable unwind table generation for crtbegin/crtend. While there, disable omitting the frame pointer [1]. Requested by: kan [1] Reviewed by: kan MFC after: 2 weeks END K 10 svn:author V 3 kib K 8 svn:date V 27 2010-06-18T11:10:39.670089Z K 7 svn:log V 163 Add unwind annotations to the asm part of crt1 on i386. Terminate the process with SIGTRAP if _start1() unexpectedly returns. Reviewed by: kan MFC after: 2 weeks END K 10 svn:author V 2 ae K 8 svn:date V 27 2010-06-18T11:18:03.599305Z K 7 svn:log V 57 Save ufsid in UFS's private structure. Some style fixes. END K 10 svn:author V 2 ae K 8 svn:date V 27 2010-06-18T11:36:55.037353Z K 7 svn:log V 50 Add GEOM NOP class to list of discovered devices. END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-18T14:06:27.413535Z K 7 svn:log V 169 Provide for multiple, cascaded PICs on PowerPC systems, and extend the OFW interrupt map interface to also return the device's interrupt parent. MFC after: 8.1-RELEASE END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-18T14:16:24.957063Z K 7 svn:log V 291 Change the default interrupt polarity on PowerPC systems from high to low. On Apple systems at least, all the level interrupts are wired active low. Before this change, our PIC programming only worked because Apple hardware ignores the interrupt polarity bit on all interrupts except IRQ 0. END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-18T14:17:45.823362Z K 7 svn:log V 135 Following r209299, level interrupts are low by default on PPC, so remove the hack here to reprogram the interrupt for K2 SATA devices. END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-18T14:20:54.214858Z K 7 svn:log V 58 Revert changes accidentally committed as part of r209298. END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-18T14:28:57.480757Z K 7 svn:log V 264 Add support for the Keywest I2C controller in Apple uninorth northbridges. Although the Keywest registers have only 1 byte of content, they are secretly 4-byte registers, which became apparent from them moving on the big-endian Uninorth version of the controller. END K 10 svn:author V 3 kan K 8 svn:date V 27 2010-06-18T15:17:38.219290Z K 7 svn:log V 69 MFC r208763: Plug possible memory leak. Found by: Coverity Prevent END K 10 svn:author V 3 kan K 8 svn:date V 27 2010-06-18T15:25:57.987757Z K 7 svn:log V 344 MFC r208994: Do not require pos parameter to be zero in MAP_ANONYMOUS mmap requests in Linux emulation layer. Linux seems to only require that pos is page-aligned, but otherwise ignores it. Default FreeBSD mmap parameter checking is too strict to allow some Linux binaries to run. tsMuxeR is one example of such a binary. Discussed with: jhb END K 10 svn:author V 3 kan K 8 svn:date V 27 2010-06-18T16:07:24.891313Z K 7 svn:log V 144 Do not allow EOF token to be put back into input buffer. This reimplements previous change from r20930 in more generic way. MFC after: 1 week END K 10 svn:author V 3 kan K 8 svn:date V 27 2010-06-18T16:14:37.818054Z K 7 svn:log V 504 MFC r207578: Do not encode more than CTF_MAX_VLEN(1023) enum members. CTF can not represent enums with more than CTF_MAX_VLEN members, but ctfconvert will happily ignore that limitation and create CTF section no other tool can interpret. This change is different from similar change from upstream, which just returns an error if big enum is encountered. Doing that means that every FreeBSD kernel with compiled in hwpmc will have no useable CTF information due to pmc_event enum having 1236+ members. END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-18T16:29:03.065930Z K 7 svn:log V 87 Missed change to sun4v while adding iparent lookup to the OFW interrupt map interface. END K 10 svn:author V 3 jfv K 8 svn:date V 27 2010-06-18T16:36:57.111769Z K 7 svn:log V 84 MFC of ixgbe driver version 2.2.1 -mostly stability fixes MFC to RELENG 8.1 asap END K 10 svn:author V 3 jfv K 8 svn:date V 27 2010-06-18T17:22:08.694210Z K 7 svn:log V 177 MFC of E1000 drivers: Stability fixes from M.Tuexen Cool stats changes from gnn Changes for POLL friendliness from John Baldwin Thanks everyone! MFC to RELENG8.1 asap END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-18T17:39:56.222822Z K 7 svn:log V 109 Add MSI support for PCI devices attached to the CPC925 and CPC945 bridges found in Apple and IBM G5 systems. END K 10 svn:author V 7 delphij K 8 svn:date V 27 2010-06-18T18:15:40.538731Z K 7 svn:log V 445 Fix two regressions introduced by GNU cpio 2.8 import: cpio/src/copyout.c: Old behavior is not to strip leading / from symbol link target. By default cpio will replace the symbol link rather than following it so this should not be a risk. cpio/src/util.c: Zero out rdev_{maj,min} for files where they are not applicable. This is a direct commit since GNU cpio has been removed from -HEAD. Sorry for the breakage... Reported by: sbruno END K 10 svn:author V 3 mav K 8 svn:date V 27 2010-06-18T18:18:03.972834Z K 7 svn:log V 69 Do not print first digits of IRQ number if whole number doesn't fit. END K 10 svn:author V 3 kib K 8 svn:date V 27 2010-06-18T18:51:09.753044Z K 7 svn:log V 77 Only enable kdtrace hook in the LINT on the architectures that implement it. END K 10 svn:author V 8 jchandra K 8 svn:date V 27 2010-06-18T20:07:30.847611Z K 7 svn:log V 236 Merge jmallett@'s n64 work into HEAD - changeset 5 Remove unnecessary locking and sched_pin() call while creating a temporary mapping. Changes from http://svn.freebsd.org/base/user/jmallett/octeon Approved by: rrs (mentor), jmallett END K 10 svn:author V 5 randi K 8 svn:date V 27 2010-06-18T21:10:56.291222Z K 7 svn:log V 219 MFC r209273: Fix uninitialized variables that cause a crash when the network is initialized and sysinstall is not running as init. Submitted by: Nick Mills Approved by: cperciva (mentor) Approved by: re (kensmith) END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-18T21:24:17.168059Z K 7 svn:log V 165 Missed commit in r209310: the IRQ number in INTR_VEC() should have parantheses around it to allow arithmetic expressions to be passed. Submitted by: Andreas Tobler END K 10 svn:author V 6 mjacob K 8 svn:date V 27 2010-06-18T21:37:27.541917Z K 7 svn:log V 153 This is an MFC of 208761 Various minor and not so minor fixes suggested by Coverity. In at least one case, it's amazing that target mode worked at all. END K 10 svn:author V 8 jchandra K 8 svn:date V 27 2010-06-18T21:40:32.880350Z K 7 svn:log V 151 Add support for newer XLS chips/boards in the GMAC driver. Improved processor id code in board.h, remove unnecessary macros. Approved by: rrs(mentor) END K 10 svn:author V 3 pjd K 8 svn:date V 27 2010-06-18T22:06:49.467582Z K 7 svn:log V 227 MFC r209265: r209260: Backout r207970 for now, it can lead to deadlocks. Reported by: kan r209261: Turn off UMA allocations on all archs by default. It isn't stable even on amd64. Reported by: many Approved by: re (kib) END K 10 svn:author V 3 alc K 8 svn:date V 27 2010-06-18T22:12:12.691678Z K 7 svn:log V 43 Eliminate unnecessary page queues locking. END K 10 svn:author V 3 alc K 8 svn:date V 27 2010-06-18T23:14:16.459056Z K 7 svn:log V 56 Catch up with the page and page queues locking changes. END K 10 svn:author V 6 mjacob K 8 svn:date V 27 2010-06-19T00:37:14.412925Z K 7 svn:log V 241 This is an MFC of 208808 I was getting panics in sleepq_add for the second sleep in isp_kthread. I don't know why- but it occurred to me in looking at the second sleep is that all I want is a pause- not an actual sleep. So do that instead. END K 10 svn:author V 6 mjacob K 8 svn:date V 27 2010-06-19T00:39:19.084093Z K 7 svn:log V 174 This is an MFC of 208809 Make the internal target > SPC2 (so REPORT LUNS can be tested). Give the NIL inquiry data real values other than just plain 0x7f in the first byte. END K 10 svn:author V 6 mjacob K 8 svn:date V 27 2010-06-19T00:41:29.240674Z K 7 svn:log V 143 This is an MFC of 208849 Be more specific about which CDB length we're going to use. Not really a likely bug but we might as well be clearer. END K 10 svn:author V 8 lstewart K 8 svn:date V 27 2010-06-19T02:30:10.396527Z K 7 svn:log V 582 - Rename the internal for loop iterator to "_i" to avoid potential shadowing of external variables named "i". The "_" prefix is reserved for infrastructure type code and is therefore not expected to be used by normal code likely to call DPCPU_SUM(). [1] - Change DPCPU_SUM to return the sum rather than calculate and assign it internally. Usage is now: "sum = DPCPU_SUM(dpcpu_var, member_to_sum);" [2] - Fix some style nits. [3] Sponsored by: FreeBSD Foundation Suggested by: bde [3], mdf [1], kib [1,2], pjd [1,3] Reviewed by: kib MFC after: 1 week (instead of r209119) END K 10 svn:author V 6 marcel K 8 svn:date V 27 2010-06-19T04:57:58.578055Z K 7 svn:log V 224 MFC revs 209026 and 209085: o Bump MAX_BPAGES from 256 to 1024. o Synchronize the kernel entry on all CPUs with the use of the ptc.g instruction on a single CPU by implementing a bare-bones readers- writer lock. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2010-06-19T05:35:47.885901Z K 7 svn:log V 252 MFC revs 209026 and 209085: o Bump MAX_BPAGES from 256 to 1024. o Synchronize the kernel entry on all CPUs with the use of the ptc.g instruction on a single CPU by implementing a bare-bones readers- writer lock. Approved by: re (kensmith) END K 10 svn:author V 3 mav K 8 svn:date V 27 2010-06-19T08:36:12.846297Z K 7 svn:log V 234 While we indeed can't precisely measure time spent in C1, we can consider measured interval as upper bound. It should be more precise then just assuming hz/2. For idle CPU it should be quite precise, for busy - not worse then before. END K 10 svn:author V 5 brian K 8 svn:date V 27 2010-06-19T08:42:29.106014Z K 7 svn:log V 97 Add a missing linefeed PR: 147337 Submitted by: cyberleo at cyberleo dot net MFC after: 1 week END K 10 svn:author V 3 mav K 8 svn:date V 27 2010-06-19T08:46:17.829220Z K 7 svn:log V 74 Oops! Add " / hz" missed in r209328. Assume interrupt rate hz/2, not 1/2. END K 10 svn:author V 5 brian K 8 svn:date V 27 2010-06-19T09:21:34.145466Z K 7 svn:log V 131 Remove vestiges of 'slip'. PR: 145648 Submitted by: alexbestms at wwu dot de and spam at rm-rf dot kiev dot ua MFC after: 1 week END K 10 svn:author V 5 brian K 8 svn:date V 27 2010-06-19T09:33:11.063472Z K 7 svn:log V 95 Remove an unnecessary cat. PR: 145447 Submitted by: u at netbeisser dot de MFC after: 1 week END K 10 svn:author V 8 jmallett K 8 svn:date V 27 2010-06-19T09:50:04.934158Z K 7 svn:log V 652 Working towards making the cryptocteon code compile and work on FreeBSD. Some convoluted use of scatterlists has been replaced with direct use of pointers into a linear buffer since it appeared that only a single scatterlist entry was being used; I intend to make that use iovecs instead in the near future. The crypto device front-end is mostly commented out and is not compiled. Actual enabling and disabling of use of the crypto coprocessor within the kernel is unimplemented in this commit, but is trivial for kernel-only use. We need a mechanism that allows use of the coprocessor from userland at some point, but that's a stretch goal for me. END K 10 svn:author V 8 jmallett K 8 svn:date V 27 2010-06-19T10:06:46.539441Z K 7 svn:log V 77 Merge from head, mostly pick up jchandra's merging of things from my branch. END K 10 svn:author V 8 jmallett K 8 svn:date V 27 2010-06-19T10:14:43.512073Z K 7 svn:log V 141 Mechanically update PTE bit/field names to start with PTE_ rather than PG_ per discussion on mips@ and on specific recommendation from alc@. END K 10 svn:author V 8 jmallett K 8 svn:date V 27 2010-06-19T10:23:23.454483Z K 7 svn:log V 98 Add notes about comments I need to update and attempt to explain the wired bit (and mostly fail.) END K 10 svn:author V 6 jilles K 8 svn:date V 27 2010-06-19T10:33:04.993394Z K 7 svn:log V 313 sh: Fix compilation with -DNO_HISTORY. The LINENO code uses snprintf() and relied on "myhistedit.h" to pull in the necessary . Compiling with -DNO_HISTORY disables all editing and history support and allows linking without -ledit -ltermcap. This may be useful for embedded systems. MFC after: 2 weeks END K 10 svn:author V 6 adrian K 8 svn:date V 27 2010-06-19T12:12:39.233227Z K 7 svn:log V 297 Extend the AR71XX watchdog debugging and data. * Add some per-device sysctl entries which record the watchdog state - whether it is armed; whether the last reboot was due to the watchdog. * Add a per-device sysctl debug flag to enable logging watchdog arming/ disarming. Reviewed by: gonzo@ END K 10 svn:author V 3 mav K 8 svn:date V 27 2010-06-19T13:09:42.648152Z K 7 svn:log V 276 Core i5, same as previously Core2Duo, found to not set P-state for single core lower then set on other cores. Do not try to test P-states on attach on SMP systems. It is hopeless now and will just pollute verbose logs. If needed, check still can be forced via loader tunable. END K 10 svn:author V 3 mav K 8 svn:date V 27 2010-06-19T13:41:17.534946Z K 7 svn:log V 75 Report transport type in XPT_PATH_INQ. PR: i386/147929 MFC after: 3 days END K 10 svn:author V 3 mav K 8 svn:date V 27 2010-06-19T13:42:14.044883Z K 7 svn:log V 58 Report transport type in XPT_PATH_INQ. MFC after: 3 days END K 10 svn:author V 5 gavin K 8 svn:date V 27 2010-06-19T16:24:55.347582Z K 7 svn:log V 62 Variable 'getmap' is unused. Found by: clang static analyzer END K 10 svn:author V 5 gavin K 8 svn:date V 27 2010-06-19T16:29:23.712773Z K 7 svn:log V 104 Remove dead assignments, we overwrite the variable almost immediately. Found by: clang static analyzer END K 10 svn:author V 5 gavin K 8 svn:date V 27 2010-06-19T17:04:01.474072Z K 7 svn:log V 123 Initialise the "trials" variable to zero earlier in case we unexpectedly error out early. Found by: clang static analyzer END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-19T17:21:48.837254Z K 7 svn:log V 72 IFC @ 209344. This reduces diffs on this branch to actually only ppc64. END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-19T17:28:35.263863Z K 7 svn:log V 171 Moved cpu_thread_swapin/out() back to booke/aim dependent vm_machdep.c, since they relate to VM operations and could be used to optimize KVA use of the SLB on 64-bit AIM. END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-19T18:07:15.794538Z K 7 svn:log V 62 Fix a typo that kept systems with > 1 GB of RAM from booting. END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-19T18:13:31.267741Z K 7 svn:log V 18 Fix SVN mismerge. END K 10 svn:author V 3 uqs K 8 svn:date V 27 2010-06-19T18:38:03.240997Z K 7 svn:log V 126 MFH r200600: Clarify what's meant by NUL for getline.3 PR: docs/141125 Submitted by: Jeremy Huddleston END K 10 svn:author V 3 uqs K 8 svn:date V 27 2010-06-19T18:52:37.689080Z K 7 svn:log V 224 mdoc nitpicking for gsched.8 - remove stray argument [1] - remove stray whitespace - use canonical wording for the HISTORY section PR: docs/147119 [1] Submitted by: Alexander Best [1] MFC after: 1 week END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-19T19:09:08.764493Z K 7 svn:log V 117 Fix a bug where the kernel could panic for large numbers of spilled mappings due to looking in the wrong PVO bucket. END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-19T19:45:08.895805Z K 7 svn:log V 117 The VM tells us which pages are wired, we don't need to wire the whole kernel against PTE spills. Suggested by: mdf END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-19T20:11:06.339479Z K 7 svn:log V 15 Missed a file. END K 10 svn:author V 10 nwhitehorn K 8 svn:date V 27 2010-06-19T20:34:37.022287Z K 7 svn:log V 43 Change a silent failure to a loud failure. END K 10 svn:author V 8 jmallett K 8 svn:date V 27 2010-06-19T21:10:13.423315Z K 7 svn:log V 111 o) Make compilable on FreeBSD. o) Add functions requested by hps@ to manage the toggle bit in the host driver. END K 10 svn:author V 8 jmallett K 8 svn:date V 27 2010-06-19T21:11:06.590444Z K 7 svn:log V 148 Add octusb driver from hps@ which uses the Simple Executive code to provide a simple host-mode controller. Only compile-tested. Submitted by: hps END