‚ê7121847 153 303 299 209 588 170 203 627 354 126 204 156 226 120 262 164 125 221 517 294 237 162 1279 321 134 548 210 221 439 223 97 260 187 107 111 156 168 408 488 350 219 287 1563 193 565 255 556 368 213 258 548 159 162 137 298 131 162 145 564 400 253 300 162 175 153 220 728 159 171 310 249 284 133 164 188 260 234 676 210 211 233 148 118 445 520 325 178 327 146 239 204 216 331 326 329 752 107 198 305 150 243 132 239 168 187 399 177 225 249 123 147 106 231 105 496 105 112 113 132 146 122 239 182 124 253 317 246 404 249 419 185 227 182 135 140 272 1129 250 187 718 1999 212 143 143 233 307 624 282 407 384 1257 141 394 128 K 10 svn:author V 3 kan K 8 svn:date V 27 2003-11-01T05:51:54.000000Z K 7 svn:log V 210 Temporarily undo parts of the stuct mount locking commit by jeff. It is unsafe to hold a mutex across vput/vrele calls. This will be redone when a better locking strategy is agreed upon. Discussed with: jeff END K 10 svn:author V 4 bmah K 8 svn:date V 27 2003-11-01T06:08:52.000000Z K 7 svn:log V 205 More trimmage of the hardware notes: sf(4), tx(4), wb(4), nge(4), ste(4), sk(4), le(4), lge(4), txp(4), bge(4). These are all Ethernet drivers whose manual pages already list specific supported devices. END K 10 svn:author V 5 silby K 8 svn:date V 27 2003-11-01T07:06:04.000000Z K 7 svn:log V 114 Fix a few style glitches in the previous commit and make the tunable error message more brief. Suggested by: bde END K 10 svn:author V 5 silby K 8 svn:date V 27 2003-11-01T07:30:08.000000Z K 7 svn:log V 493 - Add a new function tcp_twrecycleable, which tells us if the ISN which we will generate for a given ip/port tuple has advanced far enough for the time_wait socket in question to be safely recycled. - Have in_pcblookup_local use tcp_twrecycleable to determine if time_Wait sockets which are hogging local ports can be safely freed. This change preserves proper TIME_WAIT behavior under normal circumstances while allowing for safe and fast recycling whenever ephemeral port space is scarce. END K 10 svn:author V 3 tjr K 8 svn:date V 27 2003-11-01T08:18:18.000000Z K 7 svn:log V 78 Fix a typo that caused the optimized single-byte locale path not to be taken. END K 10 svn:author V 3 tjr K 8 svn:date V 27 2003-11-01T08:20:58.000000Z K 7 svn:log V 110 Remove TODO comment about creating a macro version of towctrans(). Remove unnecessary inclusion of . END K 10 svn:author V 3 bde K 8 svn:date V 27 2003-11-01T08:43:54.000000Z K 7 svn:log V 534 Fixed breakage of my world in rev.1.30. #include instead of depending on namespace pollution in for the declarations of struct timeval and utimes(). Fixed some style bugs in rev.1.30 and some nearby style bugs (mainly unsorting and missing or extra blank lines). Removed a wrong comment that was obtained from NetBSD in rev.1.14. It said that chflags() reset the times that were set "above" by utimes(), but utimes wasn't "above" in FreeBSD until rev.1.30, and chflags() does't actually reset the times. END K 10 svn:author V 3 alc K 8 svn:date V 27 2003-11-01T08:57:26.000000Z K 7 svn:log V 261 - Modify swap_pager_copy() and its callers such that the source and destination objects are locked on entry and exit. Add comments to the callers noting that the locks can be released by swap_pager_copy(). - Remove several instances of GIANT_REQUIRED. END K 10 svn:author V 3 sos K 8 svn:date V 27 2003-11-01T09:30:15.000000Z K 7 svn:log V 34 Fix cable detection on AMD chips. END K 10 svn:author V 5 markm K 8 svn:date V 27 2003-11-01T09:31:54.000000Z K 7 svn:log V 109 Shorten the code by removing one "do-nothing" function, replacing it with nullop(), which is in kern_conf.c. END K 10 svn:author V 3 sos K 8 svn:date V 27 2003-11-01T09:44:33.000000Z K 7 svn:log V 64 GEOM'ify atapi-cd. Original patch by phk, subtle changes by me. END K 10 svn:author V 3 dfr K 8 svn:date V 27 2003-11-01T12:45:03.000000Z K 7 svn:log V 133 Make the cardbus driver a derived class of the pci driver. In theory, this should allow many of the pci methods to be re-staticised. END K 10 svn:author V 3 kan K 8 svn:date V 27 2003-11-01T15:04:50.000000Z K 7 svn:log V 28 Remove now unused variable. END K 10 svn:author V 7 trhodes K 8 svn:date V 27 2003-11-01T15:58:06.000000Z K 7 svn:log V 165 The copywrite is not a 'static char', remove the #ifdefs and move the copywrite up into the commented out 'copywrite' section. Include sys/linker.h for kldload(3). END K 10 svn:author V 7 trhodes K 8 svn:date V 27 2003-11-01T16:57:19.000000Z K 7 svn:log V 68 Revert the first part of my previous change. Requested by: wollman END K 10 svn:author V 6 scottl K 8 svn:date V 27 2003-11-01T18:44:51.000000Z K 7 svn:log V 30 MFC: Remove erroneous PCI Id. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-01T20:04:12.000000Z K 7 svn:log V 125 o Compile-out "wizard" mode on ia64. o Do not set bootblocks on ia64. It's not even a functionality in libdisk on ia64. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-01T20:14:06.000000Z K 7 svn:log V 421 o Compile-out "wizard" mode on ia64. o Also allow swap and filesystem partitions outside a freebsd slice. This is typically the case for GPT. o Allow chunks of type "whole" to be displayed at the top. This is to allow a GPT disk to be labeled. We need a slice out of which we can make partitions, but a GPT disk doesn't have slices. For GPT disks a chunk of type "whole" can then be used as a placeholder. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-01T20:17:11.000000Z K 7 svn:log V 198 o Do not define WITH_SYSCONS and WITH_MICE on ia64. We cannot have a generic kernel with syscons. o Do not define WITH_SLICES on ia64. We only label because we're going to create GPT disks. END K 10 svn:author V 3 alc K 8 svn:date V 27 2003-11-01T23:06:41.000000Z K 7 svn:log V 144 - Increase the scope of two vm object locks in vm_object_collapse(). - Remove the acquisition and release of Giant from vm_object_coalesce(). END K 10 svn:author V 6 brooks K 8 svn:date V 27 2003-11-01T23:24:38.000000Z K 7 svn:log V 67 Do if_xname conversion. I missed this one because it's pc98 only. END K 10 svn:author V 4 jeff K 8 svn:date V 27 2003-11-02T03:36:33.000000Z K 7 svn:log V 1184 - Change sched_interact_update() to only accept slp+runtime values between 0 and SCHED_SLP_RUN_MAX * 2. This allows us to simplify the algorithm quite a bit. Before, it dealt with arbitrary values which required us to do nasty integer division tricks that didn't quite work out correctly. - Chnage sched_wakeup() to detect conditions where the slp+runtime could exceed SCHED_SLP_RUN_MAX * 2. This can happen if we go to sleep for longer than 6 seconds. In this case, we'll just clear the runtime and set the sleep time to the max. - Define a new function, sched_interact_fork() which updates the slp+runtime of a newly forked thread. We want to limit the amount of history retained from the parent so that we learn the child's behavior quickly. We don't, however want to decay it to nothing. Previously, we would simply divide each parameter by 100 whenever we forked. After a few forks the values would reach 0 and tasks would not be considered interactive. - Add another KTR entry, cleanup some existing entries. - Remove a useless sched_interact_update() from sched_priority(). This is already done by the callers that require it. END K 10 svn:author V 4 jeff K 8 svn:date V 27 2003-11-02T03:49:32.000000Z K 7 svn:log V 227 - Remove uses of PRIO_TOTAL and replace them with SCHED_PRI_NRESV - SCHED_PRI_NRESV does not have the off by one error in PRIO_TOTAL so we do not have to account for it in the few places that we use it. Requested by: bde END K 10 svn:author V 4 jeff K 8 svn:date V 27 2003-11-02T03:50:16.000000Z K 7 svn:log V 41 - Remove PRIO_TOTAL. Submitted by: bde END K 10 svn:author V 4 jeff K 8 svn:date V 27 2003-11-02T04:10:15.000000Z K 7 svn:log V 454 - Rename SCHED_PRI_NTHRESH to SCHED_SLICE_NTHRESH since it is only used in slice assignment. Add a comment describing what it does. - Remove a stale XXX comment, the nice should not impact the interactivity, nice adjustments only effect non-interactive tasks in ULE. - Don't allow nice -20 tasks to totally starve nice 0 tasks. Give them at least SCHED_SLICE_MIN ticks. We still allow nice 0 tasks to starve nice +20 tasks as intended. END K 10 svn:author V 4 jeff K 8 svn:date V 27 2003-11-02T04:25:59.000000Z K 7 svn:log V 116 - In sched_prio() only force us onto the current queue if our priority is being elevated (numerically smaller). END K 10 svn:author V 3 tjr K 8 svn:date V 27 2003-11-02T04:28:58.000000Z K 7 svn:log V 128 MFC rev. 1.20: Fix a warning that showed up on 64 bit systems. It was actually a real bug that had been swept under the carpet. END K 10 svn:author V 3 kan K 8 svn:date V 27 2003-11-02T04:52:53.000000Z K 7 svn:log V 346 Take care not to call vput if thread used in corresponding vget wasn't curthread, i.e. when we receive a thread pointer to use as a function argument. Use VOP_UNLOCK/vrele in these cases. The only case there td != curthread known at the moment is boot() calling sync with thread0 pointer. This fixes the panic on shutdown people have reported. END K 10 svn:author V 3 imp K 8 svn:date V 27 2003-11-02T05:33:12.000000Z K 7 svn:log V 130 The '574 family works (I had a bad dongle which is why I thought it might not be). The modem combo cards do not (confirmed today) END K 10 svn:author V 2 ru K 8 svn:date V 27 2003-11-02T06:47:39.000000Z K 7 svn:log V 7 Style. END K 10 svn:author V 2 ru K 8 svn:date V 27 2003-11-02T06:56:52.000000Z K 7 svn:log V 168 From ChangeLog: : 2003-07-06 Ruslan Ermilov : : * src/roff/nroff/nroff.sh: Add option -d for completeness. : * src/roff/nroff/nroff.man: Updated. END K 10 svn:author V 4 bmah K 8 svn:date V 27 2003-11-02T07:07:08.000000Z K 7 svn:log V 94 Bump BASE variable definition for 4.9. Pointed out by: Kent Hauser END K 10 svn:author V 2 ru K 8 svn:date V 27 2003-11-02T07:25:24.000000Z K 7 svn:log V 16 Sort encodings. END K 10 svn:author V 2 ru K 8 svn:date V 27 2003-11-02T07:27:40.000000Z K 7 svn:log V 20 MFV: The -d option. END K 10 svn:author V 2 ru K 8 svn:date V 27 2003-11-02T07:31:44.000000Z K 7 svn:log V 65 DNS should not necessarily be named(8), tweak the comment a bit. END K 10 svn:author V 4 bmah K 8 svn:date V 27 2003-11-02T07:41:03.000000Z K 7 svn:log V 75 Trim more devices from the hardware list: ida(4), ciss(4), iir(4), ie(4). END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-02T07:44:59.000000Z K 7 svn:log V 312 o In Print_Chunk(), don't print the address of the chunk on ia64. It doesn't have any meaning and only results in lines longer than 80 characters. o In Delete_Chunk2(), also look for chunks of type "part" under chunks of type "whole" on ia64. They're not only under chunks of type "freebsd" there. END K 10 svn:author V 5 silby K 8 svn:date V 27 2003-11-02T07:47:03.000000Z K 7 svn:log V 393 Add an additional check to the tcp_twrecycleable function; I had previously only considered the send sequence space. Unfortunately, some OSes (windows) still use a random positive increments scheme for their syn-ack ISNs, so I must consider receive sequence space as well. The value of 250000 bytes / second for Microsoft's ISN rate of increase was determined by testing with an XP machine. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-02T07:58:19.000000Z K 7 svn:log V 254 Turn Track_Aligned(), Prev_Track_Aligned(), Next_Track_Aligned(), Cyl_Aligned(), Prev_Cyl_Aligned() and Next_Cyl_Aligned() into tautologies on ia64. GPT removes all notion of tracks, heads and sectors per track, so there are no alignment considerations. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-02T08:01:18.000000Z K 7 svn:log V 123 Compile-out support for allfreebsd, dedicate, sanitize and bios on ia64. These commands are PC specific and not supported. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-02T08:07:25.000000Z K 7 svn:log V 191 Do not fill in d_ncylinders, d_ntracks and d_nsectors in the disklabel on ia64. Not only do we not have a disklabel by default, we also do not have a notion of cylinders, tracks and sectors. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-02T08:39:08.000000Z K 7 svn:log V 1466 o Move Int_Open_Disk() from disk.c to open_disk.c for use by all platforms except ia64 and use Int_Open_Disk() in open_ia64_disk.c on ia64. We need to know more than GEOM can provide us so we're forced to read from the disk. Move uuid_type() to open_ia64_disk.c and remove all references on non-ia64. o Pass the GEOM conftxt to Int_Open_Disk() so that only Open_Disk() needs to know about GEOM and libdisk can more easily be used with media not handled by GEOM. o Create an ia64 specific definiton of struct disk on ia64, because we don't need/have most of the fields other platforms need and other fields not applicable on platforms other than ia64. o Do not compile change.c on ia64. It's too PC specific. o In Fixup_Names() in create_chunk.c, try all partition numbers that are valid for the GPT disk. We have the total number of partitions that can be allocated in the disk structure on ia64. Also, use the GPT partition naming if we're creating one under a chunk of type "whole". It's a GPT partition in that case. o In Create_Chunk(), compile-out the PC specific code on ia64 that checks BIOS geometry restrictions. o In Debug_Disk() in disk.c, dump the ia64 specific fields. o Save the partition index in the chunk on ia64 so that we can preserve it when we write the data back to disk. This avoids that partitions get moved around or swapped after installing FreeBSD, which may render a disk unusable. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-02T08:41:18.000000Z K 7 svn:log V 98 Turn Write_Disk() into a stub for now. It needs to be rewritten to write out a GPT and not a MBR. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-02T08:58:57.000000Z K 7 svn:log V 469 o Add PART_EFI so that we can use it instead of PART_FAT on ia64 to better deal with the fact that we need an EFI partition and that we need to have a mountpoint for it. o When creating a new partition, add EFI to the list of types the user can select from. This makes it easy to create an EFI. o Do not include wizard.c on ia64. o The user cannot create a partition on ia64 that's a multiple of the cylinder size. We don't have a notion of cyclinders. END K 10 svn:author V 3 ume K 8 svn:date V 27 2003-11-02T09:13:33.000000Z K 7 svn:log V 162 - do not quit from key_sendup() even if writes to non-target socket fails. - remove an unneeded function. - fix pfkey stat. - fix comment. Obtained from: KAME END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-02T09:19:07.000000Z K 7 svn:log V 460 Remove headers copied from i386 and either useless or wrong on ia64. An example of useless is bios.h. An example of wrong is msdos.h (due to the use of long for 32-bit fields). display.h cannot be removed because it's used by syscons. That header however has no platform dependency and shouldn't really be here. Removal if these headers may cause build failures in the ports tree. It's the ports that need fixing in that case. Tested with: buildworld, LINT END K 10 svn:author V 3 tjr K 8 svn:date V 27 2003-11-02T10:09:33.000000Z K 7 svn:log V 275 Convert the Big5, EUC, MSKanji and UTF-8 encoding methods to implement mbrtowc() and wcrtomb() directly. GB18030, GBK and UTF2 are left unconverted; GB18030 will be done eventually, but GBK and UTF2 may just be removed, as they are subsets of GB18030 and UTF-8 respectively. END K 10 svn:author V 3 ume K 8 svn:date V 27 2003-11-02T10:49:47.000000Z K 7 svn:log V 120 cleanup secasvar management; refcnt is key.c responsibility; alloc/free is keydb.c responsibility. Obtained from: KAME END K 10 svn:author V 5 markm K 8 svn:date V 27 2003-11-02T10:55:16.000000Z K 7 svn:log V 163 Make sure we get all user-written input. This simplifies the code considerably. Submitted by: (forgotten) [I'll happily acknowledge the submitter if he owns up!] END K 10 svn:author V 4 jeff K 8 svn:date V 27 2003-11-02T10:56:48.000000Z K 7 svn:log V 454 - Remove the ksq_loads[] array. We are only interested in three counts, the total load, the timeshare load, and the number of threads that can be migrated to another cpu. Account for these seperately. - Introduce a KSE_CAN_MIGRATE() macro which determines whether or not a KSE can be migrated to another CPU. Currently, this only checks to see if we're an interrupt handler. Eventually this will also be used to support CPU binding. END K 10 svn:author V 3 ume K 8 svn:date V 27 2003-11-02T11:26:42.000000Z K 7 svn:log V 67 make debugging with "setkey -x" a lot easier. Obtained from: KAME END K 10 svn:author V 3 ume K 8 svn:date V 27 2003-11-02T11:43:07.000000Z K 7 svn:log V 70 exit(3) with negative value does not make sense' Obtained from: KAME END K 10 svn:author V 3 ume K 8 svn:date V 27 2003-11-02T12:28:04.000000Z K 7 svn:log V 45 correct %d/%u mismatch. Obtained from: KAME END K 10 svn:author V 4 ceri K 8 svn:date V 27 2003-11-02T17:35:33.000000Z K 7 svn:log V 204 Since we do not honour the Confidential: field, and as an attempt to decrease the number of misfiled PR submissions in general, do not allow users to send a PR that contains errors. Discussed with: core END K 10 svn:author V 3 ume K 8 svn:date V 27 2003-11-02T19:09:29.000000Z K 7 svn:log V 39 rename variables. Obtained from: KAME END K 10 svn:author V 3 imp K 8 svn:date V 27 2003-11-02T20:08:58.000000Z K 7 svn:log V 70 MFP4: o define EP_W2_ALT_EEPROM. we don't use it yet, but may soon. END K 10 svn:author V 3 imp K 8 svn:date V 27 2003-11-02T20:10:09.000000Z K 7 svn:log V 53 MFp4: o Minor tweak to error processing in ep_alloc. END K 10 svn:author V 3 imp K 8 svn:date V 27 2003-11-02T20:13:39.000000Z K 7 svn:log V 471 MFp4: o Remove redundant $FreeBSD$ o Better comments about ep_get_macaddr. o remove one tab in a switch statement (style only) o Recognize ID 0x0035 as the device ID for the 3CXEM556 that I have. This makes the 3CXEM556 work for me. Not 100% sure this is the assigned ID, as I don't have the datasheets for this part, but it does work and get the correct ethrnet address. o Comment about the whole fake IRQ 3 thing. some need it, some don't, all work with it. END K 10 svn:author V 3 imp K 8 svn:date V 27 2003-11-02T20:18:19.000000Z K 7 svn:log V 307 MFp4: o Fix MFC cards. We were bogusly setting CCR_IOBASE[01] and CCR_IOLIMIT. now when we activate the resource, we adjust these for MFC cards, per the spec. o Change type of pf_mfc_* to be bus_addr_t, which is more correct than long. This makes my 3C362D/3C363D and 3CXEM556 cards work! Woo Hoo! END K 10 svn:author V 3 imp K 8 svn:date V 27 2003-11-02T20:20:28.000000Z K 7 svn:log V 160 Only the 3C1 is now known not to work. All my other cards work with this driver. I've not tested the '556B or the '556BI cards, however, as I don't own them. END K 10 svn:author V 3 alc K 8 svn:date V 27 2003-11-02T21:30:10.000000Z K 7 svn:log V 207 - Introduce and use vm_object_reference_locked(). Unlike vm_object_reference(), this function must not be used to reanimate dead vm objects. This restriction simplifies locking. Reviewed by: tegge END K 10 svn:author V 2 ru K 8 svn:date V 27 2003-11-02T21:43:29.000000Z K 7 svn:log V 71 Correct the spelling of m_devget() in the sysctl variable description. END K 10 svn:author V 2 ru K 8 svn:date V 27 2003-11-02T21:46:11.000000Z K 7 svn:log V 84 Removed (after the env(1)/printenv(1) split) extraneous arguments to the .Nm macro. END K 10 svn:author V 3 sos K 8 svn:date V 27 2003-11-02T22:04:53.000000Z K 7 svn:log V 61 On ATA control commands return the registers in the request. END K 10 svn:author V 5 markm K 8 svn:date V 27 2003-11-02T22:13:36.000000Z K 7 svn:log V 125 Use a better way to take NO_LPR into account, and correctly check the NOCRYPT macro when looking at crypto. Reminded by: ru END K 10 svn:author V 3 sos K 8 svn:date V 27 2003-11-02T22:24:47.000000Z K 7 svn:log V 635 Fix burning of CD's that got broken by the GEOM'ification. GEOM was not designed to handle media that does not have a size. Blank CD's are of that type, so cheat and set the media size to -1. This allows burning to work, but makes GEOM issue outofrange reads that makes the ATAPI subsystem spew out a few warnings. GEOM should be tought about this. GEOM was not designed to handle changing the sectorsize between opens. Writing multitack CD's with both audio and data tracks needs to change sector size on the fly. We cheat here and stuff the current sectorsize into GEOM private internals. GEOM should grow some clean way for this. END K 10 svn:author V 3 alc K 8 svn:date V 27 2003-11-02T22:52:42.000000Z K 7 svn:log V 67 - Increase the scope of two vm object locks in vm_object_split(). END K 10 svn:author V 2 se K 8 svn:date V 27 2003-11-02T23:12:08.000000Z K 7 svn:log V 80 Set exit code to 1 in case at least one of the input files could not be opened. END K 10 svn:author V 3 tjr K 8 svn:date V 27 2003-11-02T23:20:24.000000Z K 7 svn:log V 217 Change the buffer length test in NEEDSP() so that it does not subtract one unsigned number from another potentially smaller one, leading to wraparound (and heap corruption, eventually). PR: 58813 MFC after: 2 weeks END K 10 svn:author V 3 rsm K 8 svn:date V 27 2003-11-02T23:29:33.000000Z K 7 svn:log V 156 Constify a few more strings that were hidden behind XE_DEBUG define. Remove XE_VENDOR_ID_* defines in favour of PCMCIA_VENDOR_*. Reviewed by: imp (mentor) END K 10 svn:author V 3 imp K 8 svn:date V 27 2003-11-02T23:30:22.000000Z K 7 svn:log V 191 Separate out the old 3CXEM556 models (10Mbps only) from the faster 3CCFEM556 models (10/100Mbps). I have only the former, and none of the latter, so document that the latter might not work. END K 10 svn:author V 3 imp K 8 svn:date V 27 2003-11-03T00:46:08.000000Z K 7 svn:log V 41 Both CC and CX versions of the 556 exist END K 10 svn:author V 3 alc K 8 svn:date V 27 2003-11-03T00:59:54.000000Z K 7 svn:log V 72 - Increase the scope of the source object lock in vm_map_copy_entry(). END K 10 svn:author V 3 imp K 8 svn:date V 27 2003-11-03T01:30:03.000000Z K 7 svn:log V 96 ia64 tenderbox hates what I did wrt a printf, so cast the snot out of it and hope for the best. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-03T03:18:34.000000Z K 7 svn:log V 164 Rewrite Write_Disk() so that it creates a GPT. Note that the code is basicly untested, but the guts is all there. I need to free up a disk before I give it a spin. END K 10 svn:author V 3 sam K 8 svn:date V 27 2003-11-03T03:22:39.000000Z K 7 svn:log V 141 Correct rev 1.56 which (incorrectly) reversed the test used to decide if in_pcbpurgeif0 should be invoked. Supported by: FreeBSD Foundation END K 10 svn:author V 4 jeff K 8 svn:date V 27 2003-11-03T03:27:22.000000Z K 7 svn:log V 582 - Remove kseq_find(), we no longer scan other cpu's run queues when we go idle. They figure out that we're idle fast enough that the cache pollution introduces by scanning their run queue is more expensive than waiting a little longer. - Add kseq_setidle() to mark us as being idle. Use this in place of kseq_find(). - Remove kseq_load_highest(), kseq_find() was the only consumer of this interface. kseq_balance() has it's own customized version that finds the lowest and highest loads simultaneously. Continuously told that this would be faster by: terry END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-03T04:25:55.000000Z K 7 svn:log V 114 Build zzz(8) on ia64. It first checks ACPI before it checks for APM. It will therefore always suspend using ACPI. END K 10 svn:author V 3 kan K 8 svn:date V 27 2003-11-03T04:46:19.000000Z K 7 svn:log V 118 Use VOP_UNLOCK/vrele instead of vput. td was erecived as a parameter and one cannot be sure it is equal to curthread. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-03T05:04:09.000000Z K 7 svn:log V 137 Add a bogus definition of __va_list for use by lint. Make it visible only when lint is defined to protect builds with non-GNU compilers. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-03T05:09:57.000000Z K 7 svn:log V 53 The size of a long double is 16 bytes, not 12 bytes. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-03T05:14:47.000000Z K 7 svn:log V 23 Build lint(1) on ia64. END K 10 svn:author V 3 sam K 8 svn:date V 27 2003-11-03T06:11:44.000000Z K 7 svn:log V 352 Remove bogus RTFREE that was added in rev 1.47. The rmx code operates directly on the radix tree and does not hold any routing table refernces. This fixes the reference counting problems that manifested itself as a panic during unmount of filesystems that were mounted by NFS over an interface that had been removed. Supported by: FreeBSD Foundation END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-03T06:24:48.000000Z K 7 svn:log V 424 Fix two bugs in the calculation of the last LBA of the GPT covered part of the disk. The first appears to be a typo and instead of dividing the media size with the sector size, we multiplied. The second is an off-by-1 error that's the result of mixing up count and index. The code in question is only applicable for virgin disks and is used to create the "whole" chunk, which covers only the GPT usable portion of the disk. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-03T06:52:30.000000Z K 7 svn:log V 229 Don't divide the start and end of the chunk by the sector size when filling in the GPT entry. Both are already in sector numbers (LBA) and exactly what we need for the entry. We now write a structurally correct GPT partitioning. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-03T07:10:53.000000Z K 7 svn:log V 83 Remove alpha-bitops.h now that it's unused. It's in the attic if we want it again. END K 10 svn:author V 6 marcel K 8 svn:date V 27 2003-11-03T08:04:04.000000Z K 7 svn:log V 231 Handle unaligned 4-byte loads. While in the neighborhood, remove the cr.isr sanity check. We actually encounter insanities, which very likely means that the insanity check itself is insane. Remove an empty comment while I'm at it. END K 10 svn:author V 5 harti K 8 svn:date V 27 2003-11-03T09:05:49.000000Z K 7 svn:log V 52 Virgin import of NgATM shared kernel/user part 0.91 END K 10 svn:author V 5 harti K 8 svn:date V 27 2003-11-03T09:05:49.000000Z K 7 svn:log V 144 This commit was generated by cvs2svn to compensate for changes in r121934, which included commits to RCS files with non-trunk default branches. END K 10 svn:author V 5 harti K 8 svn:date V 27 2003-11-03T09:11:53.000000Z K 7 svn:log V 109 These files are generated by the genfiles script. See FREEBSD-upgrade for instructions how to upgrade these. END K 10 svn:author V 5 harti K 8 svn:date V 27 2003-11-03T09:15:31.000000Z K 7 svn:log V 121 Explain how to handle the generated files. Explain that the import must actually be done into the netnatm sub-directory. END K 10 svn:author V 5 harti K 8 svn:date V 27 2003-11-03T09:18:52.000000Z K 7 svn:log V 236 Put the address handling, traffic descripto handling and the message encoding and decoding stuff into the base module. All of this is accessed by several of the NgATM modules and putting this into atmbase reduceds the memory footprint. END K 10 svn:author V 3 dfr K 8 svn:date V 27 2003-11-03T09:22:18.000000Z K 7 svn:log V 233 Remove explicit cardbus attachments from drivers where this is identical to the pci attachment. Cardbus is a derived class of pci so all pci drivers are automatically available for matching against cardbus devices. Reviewed by: imp END K 10 svn:author V 3 des K 8 svn:date V 27 2003-11-03T09:25:02.000000Z K 7 svn:log V 236 Move sysctl declarations next to the corresponding tunable declarations. Add a sysctl declaration for hw.ata.atapi_dma, which had gone MIA (though setting it in loader.conf still worked, it was not visible at runtime) Approved by: sos END K 10 svn:author V 3 phk K 8 svn:date V 27 2003-11-03T09:47:30.000000Z K 7 svn:log V 659 Add a generic LED driver for flashing lamps. The hardware driver decides the name under /dev/led and provides the function to turn the lamp on/off. All leds are serviced by a single timeout which runs at a basic rate of hz/10. The LED is controlled by ascii strings as follows. 0 Turn off. 1 Turn on. f Flash: _- f2 Flash: __-- f3 Flash: ___--- f4...f9 etc. d%d Digits. "d12": -__________-_-______________________________ s%s String, roll your own: 'a-j' gives on for (1...10)/10 sec. 'A-J' gives on for (1...10)/10 sec. 'sAaAbBa': _-_--__- m%s Morse '.' dot '-' dash ' ' letter space '\n' word space My mdoc skills do not reach to express that. END K 10 svn:author V 3 phk K 8 svn:date V 27 2003-11-03T10:19:33.000000Z K 7 svn:log V 15 Free major#100 END K 10 svn:author V 3 phk K 8 svn:date V 27 2003-11-03T10:34:40.000000Z K 7 svn:log V 105 Introduce new CPU_SOEKRIS option to tell soekris hardware from other hardware based on similar chipsets. END K 10 svn:author V 3 phk K 8 svn:date V 27 2003-11-03T11:03:40.000000Z K 7 svn:log V 212 Change /dev/soekris-errled to be /dev/led/error and make it conditional on CPU_SOEKRIS. Note the subtle change in semantfics for 'f%d' flash instruction and the new morse facility (see details in dev/led/led.c) END K 10 svn:author V 3 phk K 8 svn:date V 27 2003-11-03T11:05:43.000000Z K 7 svn:log V 58 Add a '-l' option suitable for use with /dev/led/* lamps. END K 10 svn:author V 5 harti K 8 svn:date V 27 2003-11-03T12:05:58.000000Z K 7 svn:log V 148 Put address handling, traffic descriptor handling and message encoding and decoding into the atmbase module when compiled directly into the kernel. END K 10 svn:author V 5 harti K 8 svn:date V 27 2003-11-03T12:08:41.000000Z K 7 svn:log V 38 Virgin import of NgATM user part 0.91 END K 10 svn:author V 5 harti K 8 svn:date V 27 2003-11-03T12:08:41.000000Z K 7 svn:log V 144 This commit was generated by cvs2svn to compensate for changes in r121947, which included commits to RCS files with non-trunk default branches. END K 10 svn:author V 5 harti K 8 svn:date V 27 2003-11-03T12:26:47.000000Z K 7 svn:log V 74 Add a sub-directory for the NgATM message encoding/decoding header files. END K 10 svn:author V 5 harti K 8 svn:date V 27 2003-11-03T12:27:22.000000Z K 7 svn:log V 93 Build address handling, traffic descriptor handling, encoding and decoding into the library. END K 10 svn:author V 5 harti K 8 svn:date V 27 2003-11-03T12:41:27.000000Z K 7 svn:log V 304 Terminate the execl() argument list with (char *)NULL instead of 0. The later is wrong and may fail on sizeof(char *) != sizeof(int) machines. The patch has already been sent to and accepted by the diffutils author, but the file is already off the vendor branch, so correct it here in the meanwhile too. END K 10 svn:author V 7 rushani K 8 svn:date V 27 2003-11-03T12:48:41.000000Z K 7 svn:log V 81 Catch up with the English version: 1.184 -> 1.188 hardware/common/dev.sgml END K 10 svn:author V 8 simokawa K 8 svn:date V 27 2003-11-03T13:01:23.000000Z K 7 svn:log V 127 - Change driver name to fix if_xname breakage in the previous revision. - Don't call device_get_unit() twice. Tested by: nork END K 10 svn:author V 7 rushani K 8 svn:date V 27 2003-11-03T14:02:26.000000Z K 7 svn:log V 152 Merge the following from the English version: 1.1.2.113 -> 1.1.2.117 errata/article.sgml 1.22.2.385 -> 1.22.2.388 relnotes/common/new.sgml END K 10 svn:author V 6 obrien K 8 svn:date V 27 2003-11-03T14:59:05.000000Z K 7 svn:log V 28 Add AMD Features NX and LM. END K 10 svn:author V 3 phk K 8 svn:date V 27 2003-11-03T15:45:42.000000Z K 7 svn:log V 55 Get word spacing right in morse mode. Fix a prototype. END K 10 svn:author V 3 phk K 8 svn:date V 27 2003-11-03T15:58:58.000000Z K 7 svn:log V 14 Fix prototypo END K 10 svn:author V 3 imp K 8 svn:date V 27 2003-11-03T16:04:31.000000Z K 7 svn:log V 138 o Use %j and uintmax_t rather than uint64_t for casting. o use more proper bus_size_t for iosize rather than bus_addr_t. Noticed by: bde END K 10 svn:author V 3 imp K 8 svn:date V 27 2003-11-03T16:05:46.000000Z K 7 svn:log V 13 Sync to 1.78 END K 10 svn:author V 3 imp K 8 svn:date V 27 2003-11-03T16:09:17.000000Z K 7 svn:log V 403 Add intel EtherExpress PRO PCMCIA card ID. Well, I don't know if it is really EtherExpress or EEPro or what, but it does appear in a couple of ethernet cards that have appeared recently on ebay. Silicom appears to make these cards, and they have the 82595TX chipset in them, and sometimes uarts. The ex driver needs some work to support these cards, but I thought I'd get the device into pccarddevs. END K 10 svn:author V 3 imp K 8 svn:date V 27 2003-11-03T16:12:27.000000Z K 7 svn:log V 13 Sync to 1.79 END K 10 svn:author V 3 des K 8 svn:date V 27 2003-11-03T16:14:45.000000Z K 7 svn:log V 20 Whitespace cleanup. END K 10 svn:author V 6 murray K 8 svn:date V 27 2003-11-03T16:45:49.000000Z K 7 svn:log V 18 MFC: PRC locales. END K 10 svn:author V 3 mux K 8 svn:date V 27 2003-11-03T17:03:01.000000Z K 7 svn:log V 40 Change a bogus -n parameter to echo(1). END K 10 svn:author V 3 phk K 8 svn:date V 27 2003-11-03T17:06:10.000000Z K 7 svn:log V 54 Thanks to the kind mdoc help of Sergey (osa@): led.4 END K 10 svn:author V 5 mikeh K 8 svn:date V 27 2003-11-03T17:12:09.000000Z K 7 svn:log V 28 Import latest CVS lukemftp. END K 10 svn:author V 5 mikeh K 8 svn:date V 27 2003-11-03T17:12:09.000000Z K 7 svn:log V 144 This commit was generated by cvs2svn to compensate for changes in r121966, which included commits to RCS files with non-trunk default branches. END K 10 svn:author V 7 cvs2svn K 8 svn:date V 27 2003-11-03T17:12:10.000000Z K 7 svn:log V 86 This commit was manufactured by cvs2svn to create tag 'lukemftp-vendor-nbsd_20031103'. END K 10 svn:author V 3 phk K 8 svn:date V 27 2003-11-03T17:55:02.000000Z K 7 svn:log V 32 More mdocery from Sergey (osa@) END K 10 svn:author V 7 rwatson K 8 svn:date V 27 2003-11-03T17:58:23.000000Z K 7 svn:log V 156 Unlock pipe mutex when failing MAC pipe ioctl access control check. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories END K 10 svn:author V 7 rwatson K 8 svn:date V 27 2003-11-03T18:01:38.000000Z K 7 svn:log V 220 Remove comment about desire for eventual explicit labeling of ICMP header copy made on input path: this is now handled differently. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories END K 10 svn:author V 7 rwatson K 8 svn:date V 27 2003-11-03T18:03:05.000000Z K 7 svn:log V 149 Note that when ip_output() is called from ip_forward(), it will already have its options inserted, so the opt argument to ip_output() must be NULL. END K 10 svn:author V 7 iedowse K 8 svn:date V 27 2003-11-03T19:45:05.000000Z K 7 svn:log V 307 Override the root server address if an IP address is specified in the root path. This is reported to make non-PXE netbooting, such as is used on sparc64 systems, work correctly when the TFTP server is not the same as the root server. PR: kern/57328 Submitted by: Per Kristian Hove END K 10 svn:author V 6 anholt K 8 svn:date V 27 2003-11-03T20:44:00.000000Z K 7 svn:log V 153 Change the DRM_ERROR about authenticator not found back to DRM_DEBUG. It's noisier than I expected, and I don't have the time to actually get it fixed. END K 10 svn:author V 7 rwatson K 8 svn:date V 27 2003-11-03T21:00:16.000000Z K 7 svn:log V 322 When printing ACLs, truncate user and group names if they're too long, rather than generating an error. This is consistent with other tools printing user and group names, and means you can read the ACL using our tools rather than being up a creek. PR: 56991 Submitted by: Michael Bretterklieber END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T21:00:34.000000Z K 7 svn:log V 93 Remove soon to be obsolete file to break kernel build while the new interrupt code comes in. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T21:06:54.000000Z K 7 svn:log V 134 Revert the critical section implementation to disable interrupts via cli/sti now that we support many more than 32 interrupt sources. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T21:09:17.000000Z K 7 svn:log V 90 Define IDTVEC() and TRAP() assembly macros so that they can be shared with several files. END K 10 svn:author V 2 ru K 8 svn:date V 27 2003-11-03T21:09:47.000000Z K 7 svn:log V 44 Don't be so chatty about building includes. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T21:10:17.000000Z K 7 svn:log V 48 Move the NMI handling code out to its own file. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T21:12:04.000000Z K 7 svn:log V 179 - Always use 256 IDT entries since it is now a runtime decison as to how many entries we use. - Add a constant IDT_IO_INTS for the first IDT entry used for device interrupts. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T21:25:52.000000Z K 7 svn:log V 1035 New device interrupt code. This defines an interrupt source abstraction that provides methods via a PIC driver to do things like mask a source, unmask a source, enable it when the first interrupt handler is added, etc. The interrupt code provides a table of interrupt sources indexed by IRQ numbers, or vectors. These vectors are what new-bus uses for its IRQ resources and for bus_setup_intr()/bus_teardown_intr(). The interrupt code then maps that vector a given interrupt source object. When an interrupt comes in, the low-level interrupt code looks up the interrupt source for the source that triggered the interrupt and hands it off to this code to execute the appropriate handlers. By having an interrupt source abstraction, this allows us to have different types of interrupt source providers within the shared IRQ address space. For example, IRQ 0 may map to pin 0 of the master 8259A PIC, IRQs 1 through 60 may map to pins on various I/O APICs, and IRQs 120 through 128 may map to MSI interrupts for various PCI devices. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T21:28:54.000000Z K 7 svn:log V 157 Allocate space for the intrcnt array. This array is managed in the interrupt code layer as interrupt sources are added and handlers added to those sources. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T21:30:00.000000Z K 7 svn:log V 95 Add a per-thread variable for saving the state of eflags to support the critical section code. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T21:34:45.000000Z K 7 svn:log V 625 Add the new atpic(4) driver for the 8259A master and slave PICs. By default we provide 16 interrupt sources for IRQs 0 through 15. However, if the I/O APIC driver has already registered sources for any of those IRQs then we will silently fail to register our own source for that IRQ. Note that i386/isa/icu.h is now specific to the 8259A and no longer contains any info relevant to APICs. Also note that fast interrupts no longer use a separate entry point. Instead, both fast and threaded interrupts share the same entry point which merely looks up the appropriate source and passes control to intr_execute_handlers(). END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T21:53:38.000000Z K 7 svn:log V 1905 New APIC support code: - The apic interrupt entry points have been rewritten so that each entry point can serve 32 different vectors. When the entry is executed, it uses one of the 32-bit ISR registers to determine which vector in its assigned range was triggered. Thus, the apic code can support 159 different interrupt vectors with only 5 entry points. - We now always to disable the local APIC to work around an errata in certain PPros and then re-enable it again if we decide to use the APICs to route interrupts. - We no longer map IO APICs or local APICs using special page table entries. Instead, we just use pmap_mapdev(). We also no longer export the virtual address of the local APIC as a global symbol to the rest of the system, but only in local_apic.c. To aid this, the APIC ID of each CPU is exported as a per-CPU variable. - Interrupt sources are provided for each intpin on each IO APIC. Currently, each source is given a unique interrupt vector meaning that PCI interrupts are not shared on most machines with an I/O APIC. That mapping for interrupt sources to interrupt vectors is up to the APIC enumerator driver however. - We no longer probe to see if we need to use mixed mode to route IRQ 0, instead we always use mixed mode to route IRQ 0 for now. This can be disabled via the 'NO_MIXED_MODE' kernel option. - The npx(4) driver now always probes to see if a built-in FPU is present since this test can now be performed with the new APIC code. However, an SMP kernel will panic if there is more than one CPU and a built-in FPU is not found. - PCI interrupts are now properly routed when using APICs to route interrupts, so remove the hack to psuedo-route interrupts when the intpin register was read. - The apic.h header was moved to apicreg.h and a new apicvar.h header that declares the APIs used by the new APIC code was added. END K 10 svn:author V 3 imp K 8 svn:date V 27 2003-11-03T21:54:24.000000Z K 7 svn:log V 119 Fix two small style nits pointed out by bde: Remove spaces after cast and indent continued line 4 spaces instead of 2. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T22:06:35.000000Z K 7 svn:log V 51 Enable PCI interrupt routing for i386 SMP kernels. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T22:07:21.000000Z K 7 svn:log V 51 Update names of entry points for interrupt frames. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T22:08:52.000000Z K 7 svn:log V 140 - Export doreti as a global symbol. - Don't include isa/vector.s. Each PIC driver's entry points now live in their own standalone files. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T22:12:37.000000Z K 7 svn:log V 214 Add the MP Table APIC enumerator. This code uses the BIOS MP Table to enumerate I/O APICs as well as local APICs. It also provides Host-PCI and PCI-PCI bridge drivers to use the MP Table to route PCI interrupts. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T22:17:44.000000Z K 7 svn:log V 531 Add the ACPI MADT table APIC enumerator. This code uses the ACPI Multiple APIC Descriptor Table to enumerate both I/O APICs and local APICs. ACPI does not embed PCI interrupt routing information in the MADT like the MP Table does. Instead, ACPI stores the PCI interrupt routing information in the _PRT object under each PCI bus device. The MADT table simply provides hints about which interrupt vectors map to which I/O APICs. Thus when using ACPI, the existing ACPI PCI bridge drivers are sufficient to route PCI interrupts. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T22:18:57.000000Z K 7 svn:log V 189 - Update includes. - Use intr_suspend/resume() callouts to the interrupt code layer which suspends and resumes all the known interrupt sources instead of calling icu_reinit() directly. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T22:20:50.000000Z K 7 svn:log V 314 - Remove explicit enabling of the BSP's APIC in the APIC_IO case and the slave pin on the master PIC in the !APIC_IO case. The PIC drivers now manage these details internally. - Remove an spl0() that hasn't done anything since SMPng was first committed. - Update some comments that have rotted since SMPng. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T22:22:04.000000Z K 7 svn:log V 291 Don't probe PnP BIOS devices for PICs for now to avoid problems with those devices claiming resources that they don't actually use. The PIC drivers only register valid interrupt sources, so we don't need to rely on these drivers to claim invalid IRQs to prevent their use by other drivers. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T22:32:04.000000Z K 7 svn:log V 1163 New i386 SMP code: - The MP code no longer knows anything specific about an MP Table. Instead, the local APIC code adds CPUs via the cpu_add() function when a local APIC is enumerated by an APIC enumerator. - Don't divide the argument to mp_bootaddress() by 1024 just so that we can turn around and mulitply it by 1024 again. - We no longer panic if SMP is enabled but we are booted on a UP machine. - init_secondary(), the asm code between init_secondary() and ap_init() in mpboot.s and ap_init() have all been merged together in C into init_secondary(). - We now use the cpuid feature bits to determine if we should enable PSE, PGE, or VME on each AP. - Due to the change in the implementation of critical sections, acquire the SMP TLB mutex around a slightly larger chunk of code for TLB shootdowns. - Remove some of the debug code from the original SMP implementation that is no longer used or no longer applies to the new APIC code. - Use a temporary hack to disable the ACPI module until the SMP code has been further reorganized to allow ACPI to work as a module again. - Add a DDB command to dump the interesting contents of the IDT. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T22:34:53.000000Z K 7 svn:log V 49 Catch up to i386 interrupt and SMP code changes. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T22:36:43.000000Z K 7 svn:log V 301 - Always allocate the maximum size for the IRQ resource manager. Ideally we would manage this better by having the interrupt code add each interrupt vector to the resource map when each source is registered. - Use the new interrupt code API for registering and tearing down interrupt handlers. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2003-11-03T22:37:28.000000Z K 7 svn:log V 36 Catch up to interrupt code changes. END