ƒ­@19410 229 833 115 120 115 162 441 167 348 204 137 112 274 982 427 310 219 122 135 147 464 200 225 114 127 113 112 129 143 197 199 216 113 112 112 112 112 112 112 112 928 112 112 301 155 410 209 206 277 206 322 407 235 245 425 133 445 364 225 312 114 319 161 223 321 256 481 229 188 168 138 236 186 577 167 293 115 189 246 312 205 110 109 109 127 203 114 129 99 223 109 109 124 126 249 185 150 314 169 197 126 98 141 160 136 109 152 189 146 188 98 504 314 1133 349 175 210 216 164 182 194 167 108 145 150 424 109 171 109 141 135 702 304 177 127 127 198 171 142 451 158 240 184 261 190 220 134 212 121 140 97 117 129 329 165 585 175 172 171 137 219 150 460 153 658 166 166 286 382 109 186 98 140 98 203 259 211 230 227 112 97 124 112 136 121 264 435 131 218 228 152 158 154 213 112 112 367 132 630 127 210 145 207 165 390 493 336 128 222 505 166 554 258 374 372 191 509 900 802 540 310 209 177 239 157 479 134 262 323 169 K 10 svn:author V 5 guido K 8 svn:date V 27 1996-11-04T22:17:20.000000Z K 7 svn:log V 738 New vx driver for: 3COM 3C590 Etherlink III PCI, 3COM 3C595 Fast Etherlink PCI, 3COM 3C592 Etherlink III EISA, 3COM 3C590 Fast Etherlink EISA, 3COM 3C900 Etherlink XL PCI and 3COM 3C905 Fast Etherlink XL PCI. This driver is based on OpenBSD's driver. I modified it to run under FreeBSd and made it actually work usefully. Afterwards, nao@tom-yam.or.jp (HAMADA Naoki) added EISA support as well as early support for 3C900 Etherlink XL PCI and 3C905 Fast Etherlink XL PCI. He also split up the driver in a bus independant and bus dependant parts. Especially the 3c59X support should be pretty stable now. Submitted by: partly nao@tom-yam.or.jp (HAMADA Naoki) Obtained from:partly OpenBSD END K 10 svn:author V 5 wosch K 8 svn:date V 27 1996-11-04T23:50:58.000000Z K 7 svn:log V 21 export PATH variable END K 10 svn:author V 5 wosch K 8 svn:date V 27 1996-11-04T23:54:01.000000Z K 7 svn:log V 26 add PATH variable, use it END K 10 svn:author V 5 wosch K 8 svn:date V 27 1996-11-04T23:57:40.000000Z K 7 svn:log V 21 export PATH variable END K 10 svn:author V 7 smpatel K 8 svn:date V 27 1996-11-05T02:04:37.000000Z K 7 svn:log V 66 Add audio mixer ioctls. Only writing to the mixer is implemented. END K 10 svn:author V 5 dyson K 8 svn:date V 27 1996-11-05T04:19:08.000000Z K 7 svn:log V 346 Vastly improved contigmalloc routine. It does not solve the problem of allocating contiguous buffer memory in general, but make it much more likely to work at boot-up time. The best chance for an LKM-type load of a sound driver is immediately after the mount of the root filesystem. This appears to work for a 64K allocation on an 8MB system. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-05T05:13:39.000000Z K 7 svn:log V 75 Correct the ancient paths here in case someone ever uncomments this again. END K 10 svn:author V 6 msmith K 8 svn:date V 27 1996-11-05T05:52:36.000000Z K 7 svn:log V 252 Protect against PCI devices which may have their 'changed' flag set, or list items which may look like devices but which don't have an isa_device structure attached to them. This _shouldn't_ be possible, but it appears to have been observed-by: Joerg END K 10 svn:author V 5 wosch K 8 svn:date V 27 1996-11-05T07:17:33.000000Z K 7 svn:log V 109 add /sbin to PATH, the location of sysctl(8) may changed in future use shell builtin 'exec' for last command END K 10 svn:author V 5 wosch K 8 svn:date V 27 1996-11-05T07:22:47.000000Z K 7 svn:log V 43 add PATH variable better argument handling END K 10 svn:author V 5 wosch K 8 svn:date V 27 1996-11-05T07:24:48.000000Z K 7 svn:log V 18 add PATH variable END K 10 svn:author V 5 gibbs K 8 svn:date V 27 1996-11-05T07:51:29.000000Z K 7 svn:log V 179 index_untagged_scb should rely solely on the argument passed in SINDEX and not access SCB_TCL directly. This could have caused problems on twin channel adapters. 2.2 Candidate. END K 10 svn:author V 5 gibbs K 8 svn:date V 27 1996-11-05T07:57:29.000000Z K 7 svn:log V 887 Move the include opt_aic7xxx in aic7xxx.h so that all of the driver files get it automatically. The AHC_FORCE_PIO option wasn't having any effect because the PCI probe code didn't include this file. Fix some problems with the new sync and wide negotiation code. First off, go back to async transfers by using a message reject again. The SCSI II and III spec indicate that if a target's response to an initiater does not suit (i.e. its too low), then performing a message reject is the appropriate response. If, on the other hand, the initiator begins the negotiation and we want to go async, we will send back an SDTR message with a 0 period and offset. Also fix a really bad negotiation problem caused by a missing "break". This would usually hit people that had "smart" wide devices that immediately attempt sync negotiation after a successful wide negotiation. 2.2 Candidate. END K 10 svn:author V 5 gibbs K 8 svn:date V 27 1996-11-05T07:59:28.000000Z K 7 svn:log V 332 Clean up some code having to do with the 398X cards. We can't rely on the 7810 being either the last of the first device to be probed, so use a counting scheme instead to determine when one card ends and another begins. There may be a better way to do this by decoding the PCI tag, which I will investigate later. 2.2 Candidate. END K 10 svn:author V 2 dg K 8 svn:date V 27 1996-11-05T08:19:40.000000Z K 7 svn:log V 218 Eliminate an unnecessary synchronous write (and an 8K bcopy+bzero) when truncating/deleting large files. Reviewed by: mckusick, dyson Submitted by: Kirk McKusick , modified for FreeBSD by me. END K 10 svn:author V 5 gibbs K 8 svn:date V 27 1996-11-05T08:39:33.000000Z K 7 svn:log V 124 Add missing parenthesis. That's what I get for having e different versions of this driver in three different trees. END K 10 svn:author V 5 gibbs K 8 svn:date V 27 1996-11-05T08:49:17.000000Z K 7 svn:log V 28 A little 80 column cleanup. END K 10 svn:author V 5 gibbs K 8 svn:date V 27 1996-11-05T09:20:08.000000Z K 7 svn:log V 41 Update to changes in generic SCSI layer. END K 10 svn:author V 3 max K 8 svn:date V 27 1996-11-05T09:41:38.000000Z K 7 svn:log V 55 Added label for the "MAINTAINER on Makefiles" section. END K 10 svn:author V 3 max K 8 svn:date V 27 1996-11-05T09:42:59.000000Z K 7 svn:log V 371 ^^^^ ^^^^^ doesn't make the ".." in typewritten format. And the ascii version of the file shows '..' literally. After looking into linuxdoc dtd, we found there is no way to make ".." in typewritten font. Noticed by: Mitsuru IWASAKI Added reference to "MAINTAINER on Makefiles" section. END K 10 svn:author V 5 asami K 8 svn:date V 27 1996-11-05T11:44:48.000000Z K 7 svn:log V 105 Use &a.hackers; and &a.announce; where applicable. Submitted by: Mitsuru IWASAKI END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-05T19:53:20.000000Z K 7 svn:log V 132 Remove a bogon from version 1.69. Just because we say "use all disk" doesn't mean that our geometry can be flushed down the drain. END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-05T20:07:46.000000Z K 7 svn:log V 22 Merger from -current. END K 10 svn:author V 4 jhay K 8 svn:date V 27 1996-11-05T20:10:37.000000Z K 7 svn:log V 34 Add knobs for IPX. 2.2 candidate? END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-05T20:14:23.000000Z K 7 svn:log V 21 Merge from -current. END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-05T20:18:00.000000Z K 7 svn:log V 20 Merge from -current END K 10 svn:author V 3 jdp K 8 svn:date V 27 1996-11-05T22:27:33.000000Z K 7 svn:log V 37 Fix a spelling error. 2.2 Candidate. END K 10 svn:author V 3 jdp K 8 svn:date V 27 1996-11-05T22:29:52.000000Z K 7 svn:log V 51 Merge from HEAD 1.23 -> 1.24: Fix spelling error. END K 10 svn:author V 5 wosch K 8 svn:date V 27 1996-11-05T23:35:16.000000Z K 7 svn:log V 102 Not hang in the downcase routine. Submitted by: Trevor Blackwell tlb@viaweb.com 2.1.6/2.2 Candidate END K 10 svn:author V 4 ache K 8 svn:date V 27 1996-11-06T00:12:14.000000Z K 7 svn:log V 105 Restore 8bit stripped down in all lines Without this change all messages looks very funny 2.2 candidate END K 10 svn:author V 3 max K 8 svn:date V 27 1996-11-06T06:03:17.000000Z K 7 svn:log V 123 It's not too good idea to define nonexistent category in the sample Makefile. (There used to be ``CATEGORIES+=printing''.) END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-06T08:43:35.000000Z K 7 svn:log V 21 Merge from -current. END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-06T08:46:44.000000Z K 7 svn:log V 20 Merge from -current END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-06T09:25:51.000000Z K 7 svn:log V 20 Merge from -current END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-06T09:39:45.000000Z K 7 svn:log V 20 Merge from -current END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-06T10:08:33.000000Z K 7 svn:log V 20 Merge from -current END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-06T10:14:01.000000Z K 7 svn:log V 20 Merge from -current END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-06T10:23:48.000000Z K 7 svn:log V 20 Merge from -current END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-06T10:33:28.000000Z K 7 svn:log V 20 Merge from -current END K 10 svn:author V 3 dfr K 8 svn:date V 27 1996-11-06T10:53:16.000000Z K 7 svn:log V 835 Improve the queuing algorithms used by NFS' asynchronous i/o. The existing mechanism uses a global queue for some buffers and the vp->b_dirtyblkhd queue for others. This turns sequential writes into randomly ordered writes to the server, affecting both read and write performance. The existing mechanism also copes badly with hung servers, tending to block accesses to other servers when all the iods are waiting for a hung server. The new mechanism uses a queue for each mount point. All asynchronous i/o goes through this queue which preserves the ordering of requests. A simple mechanism ensures that the iods are shared out fairly between active mount points. This removes the sysctl variable vfs.nfs.dwrite since the new queueing mechanism removes the old delayed write code completely. This should go into the 2.2 branch. END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-06T11:19:51.000000Z K 7 svn:log V 20 Merge from -current END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-06T11:31:07.000000Z K 7 svn:log V 20 Merge from -current END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T12:42:11.000000Z K 7 svn:log V 206 Disable the `worm' driver for 2.1.6. It doesn't work with the popluar HP drives, and now that 2.2 is around the corner, there's no need to also include the required patch to the SCSI subsystem into 2.1.6. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T13:29:42.000000Z K 7 svn:log V 61 Minor bugfix in the cdburn example (one rtprio was missing). END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T13:33:53.000000Z K 7 svn:log V 315 Updates and bugfixes to the worm driver: . also detect the Phlips CDD2000; it's software-compatible with the HP part Submitted by: cau@cc.gatech.edu (Carlos Ugarte) . correct the blocksize handling for CD-DA tracks, and fix multitrack handling Submitted by: nsayer@quack.kfu.com (Nick Sayer) 2.2 candidates! END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T13:35:27.000000Z K 7 svn:log V 114 Update the man page: the Philips CDD2000 is now supported. 2.2 candidate, iff the driver update has been merged. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T13:41:08.000000Z K 7 svn:log V 111 Added Nick Sayer , for his extensive testing and submitting bugfixes to the CD-R driver. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-06T13:52:57.000000Z K 7 svn:log V 184 Since Joerg reminded me, make a small change to this file in order to allow the user to also say: burncd mybigimage dummy To enable a dummy burn. Saves on blanks during testing. :) END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-06T13:54:35.000000Z K 7 svn:log V 113 Amend the README to mention "dummy" feature. Nobody will read the README, but at least I can say it's doc'd. :) END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-06T14:08:39.000000Z K 7 svn:log V 229 Allow fdisk to be driven from a configuration file, making a 3rd-party utility for front-ending its operation more of a possibility. 2.2-RELEASE candiate. Closes PR#1960 Submitted-By: Darryl Okahata END K 10 svn:author V 3 bde K 8 svn:date V 27 1996-11-06T14:52:20.000000Z K 7 svn:log V 314 Rewrote the COMMENT about the bogus options CHILD_MAX and OPEN_MAX. These will be renamed. Fixed comments about unsupported network protocols. ncr0 is a controller, not a device. This make no difference. Added undocumented options DEVFS_ROOT, I586_CTR_GUPROF and I586_PMC_GUPROF. Sorted undocumented options. END K 10 svn:author V 3 bde K 8 svn:date V 27 1996-11-06T14:56:11.000000Z K 7 svn:log V 142 Compile genassym without -static (undo rev.1.20). The default can be depended on now, and linux_genassym has depended on it for a long time. END K 10 svn:author V 3 bde K 8 svn:date V 27 1996-11-06T15:13:41.000000Z K 7 svn:log V 152 Compile linux_genassym with the same options as genassym. ${PARAM} and - were missingUKERNEL. This was harmless until I declared the kernel's main(). END K 10 svn:author V 3 bde K 8 svn:date V 27 1996-11-06T17:02:38.000000Z K 7 svn:log V 332 Count only hardware interrupts in cnt.v_intr, so that the individual hardware interrupt counts add up to the total. Previously, software interrupts generated by splz() were counted in the total. These software interrupts seem to be very rare - there have apparently been 0 of them on freefall among the last 352448857 interrupts. END K 10 svn:author V 3 jdp K 8 svn:date V 27 1996-11-06T17:05:30.000000Z K 7 svn:log V 41 Bring libz into the RELENG_2_1_0 branch. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T17:31:14.000000Z K 7 svn:log V 350 Fix the sequence at odopen() time so the driver actually notices if a medium with another size is being inserted. Right now, this case was broken and led to a situation where a medium could only be replaced with another one of the same size. Closes PR #kern/1830: Can't mount optical disk... Submitted by: akiyama@kme.mei.co.jp (Shunsuke Akiyama) END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T17:34:25.000000Z K 7 svn:log V 269 Fix my broken config entries for the `od' driver in the !NEW_SCSICONF case. I've accidentally copied over the NEW_SCSICONF entries there, which was totally bogus. Closes PR #kern/1829: KONICA MO drive config... Submitted by: akiyama@kme.mei.co.jp (Shunsuke Akiyama) END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T19:02:18.000000Z K 7 svn:log V 130 Merge the changes from rev 1.12 of fdisk.c and 1.7 of fdisk.8: allow a script-driven automated configuration. Reviewed by: joerg END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T19:12:01.000000Z K 7 svn:log V 217 Merge the change from rev. 1.23: make the `od' driver work again with media of different size. (I dare to commit this myself to the 2.2 branch since i think there's nobody else around who could even test it at all.) END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-06T19:15:30.000000Z K 7 svn:log V 22 Update to XFree86 3.2 END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T19:21:53.000000Z K 7 svn:log V 224 Merge the fixes from rev 1.30 worm.c and 1.65 scsiconf.c: . support for the Philips CDD2000 . fix the CD-DA blocksize handling and multitrack handling (Again, i'm certainly the most knowledgable committer for this stuff.) END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T19:32:31.000000Z K 7 svn:log V 67 Merge from HEAD: mention that we also support the Philips CDD2000. END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-06T19:43:10.000000Z K 7 svn:log V 130 when you try to run iostat is fails with invalid dk_ndrive 0 Reviewed by: phk Submitted by: John-Mark Gurney END K 10 svn:author V 3 pst K 8 svn:date V 27 1996-11-06T22:27:04.000000Z K 7 svn:log V 228 auth.debug should not be sent to root realtime news.notice info should not be sent to /var/log/messages, as news has its own set of logs and notice is overused by inn added entries for newsservers (but they're commented out) END K 10 svn:author V 3 pst K 8 svn:date V 27 1996-11-06T22:30:29.000000Z K 7 svn:log V 163 The standard .mc files shipped with the new sendmail have an extra divert(0) at the top. Make our freebsd-supplied files match the standard files. Apply-to: 2.2 END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T22:46:26.000000Z K 7 svn:log V 386 Add hooks to allow the build of a vi without Tcl or Perl APIs in an environment where either or both are normally to be built. The next commit to release/Makefile will use these hooks for the creation of the `fixit' floppy. (Failing to find a better method, we might also consider pulling them into -current. By now, they just live in the 2.2 branch only.) Agreed by: jkh,phk,peter END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T22:48:03.000000Z K 7 svn:log V 134 Fix the name of our vi inside the fixit_crunch file. Activate the hook to build the vi on the fixit floppy without Tcl or Perl APIs. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T22:49:49.000000Z K 7 svn:log V 94 My MDA screen finally bothered me enough to create an ISO 8859-1 -> IBM cp437 screenmap file. END K 10 svn:author V 7 cvs2svn K 8 svn:date V 27 1996-11-06T22:49:50.000000Z K 7 svn:log V 72 This commit was manufactured by cvs2svn to create branch 'RELENG_2_1_0'. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-06T23:42:04.000000Z K 7 svn:log V 44 Merge the syslog.conf changes from rev 1.6. END K 10 svn:author V 6 msmith K 8 svn:date V 27 1996-11-07T01:13:37.000000Z K 7 svn:log V 140 Add the 'piix' device, to get the right flags for it. This appears to fix Joerg's freeze. Definitely a 2.2 candidate. Reviewed by: joerg END K 10 svn:author V 4 alex K 8 svn:date V 27 1996-11-07T01:46:47.000000Z K 7 svn:log V 93 Style changes: $(...) --> ${...}, consistent placement of -c. Suggested by: bde (via steve) END K 10 svn:author V 3 bde K 8 svn:date V 27 1996-11-07T03:05:27.000000Z K 7 svn:log V 484 Don't switch from fast interrupt handlers to normal interrupt handlers if interrupts are nested more than a few (3) deep. This only reduces the maximum nesting level by 1 with the standard drivers unless there is a related bug somewhere, but can't hurt much (the worst case is returning to hoggish interrupt handler like wdintr(), but such interrupt handlers hurt anyway). Fixed a previously harmless race incrementing the interrupt nesting level. This should be in 2.1.6 and 2.2. END K 10 svn:author V 3 max K 8 svn:date V 27 1996-11-07T03:59:05.000000Z K 7 svn:log V 75 Replaced `$' in the sample Makefiles with $. Suggested by: jfieber END K 10 svn:author V 5 asami K 8 svn:date V 27 1996-11-07T04:10:55.000000Z K 7 svn:log V 198 Remove comment about offset, it is no longer necessary. Add mention of partition types in disklabel. Add disklabel(8) to "see also" section. Reminded by: Chris Timmons END K 10 svn:author V 5 asami K 8 svn:date V 27 1996-11-07T04:14:22.000000Z K 7 svn:log V 21 Merge with -current. END K 10 svn:author V 5 asami K 8 svn:date V 27 1996-11-07T05:37:05.000000Z K 7 svn:log V 95 Merge rev. 1.2 from HEAD. Make xargs take into account the bytes occupied by the environment. END K 10 svn:author V 5 gibbs K 8 svn:date V 27 1996-11-07T06:39:44.000000Z K 7 svn:log V 151 Bzero the kernel scb array after it is allocated otherwise the control byte used on the first transaction on an SCB is indeterminate. Spaces -> tabs. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T08:03:29.000000Z K 7 svn:log V 219 Joerg's changes to add screen font and screenmap setting. Some changes of my own to make screen saver configuration a little more sane, and also make it easier to get to the keyboard/screen setup from the options menu. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-07T08:52:16.000000Z K 7 svn:log V 110 Merge the fix from rev 1.65: make the "piix" device known (and hidden), so it cannot accidentally be removed. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-07T08:53:52.000000Z K 7 svn:log V 16 Update to HEAD. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T09:08:23.000000Z K 7 svn:log V 17 Merge from HEAD. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T09:18:15.000000Z K 7 svn:log V 17 Merge from HEAD. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T09:26:32.000000Z K 7 svn:log V 35 Bring new screen map in from HEAD. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T09:27:23.000000Z K 7 svn:log V 110 If not on a tty, don't do "--more--" processing. Someday, I will get rogomatic to work with this version. :) END K 10 svn:author V 5 asami K 8 svn:date V 27 1996-11-07T12:32:05.000000Z K 7 svn:log V 20 Add mbone category. END K 10 svn:author V 5 asami K 8 svn:date V 27 1996-11-07T12:38:48.000000Z K 7 svn:log V 35 Add ports-chinese and ports-mbone. END K 10 svn:author V 5 wosch K 8 svn:date V 27 1996-11-07T12:44:43.000000Z K 7 svn:log V 6 merge END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T14:17:09.000000Z K 7 svn:log V 130 Properly clear FTP site selection info when network init fails; the site name may have been wrong, or need to be specified by IP. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T14:19:31.000000Z K 7 svn:log V 17 Merge from head. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T14:20:08.000000Z K 7 svn:log V 17 Merge from HEAD. END K 10 svn:author V 5 markm K 8 svn:date V 27 1996-11-07T14:41:20.000000Z K 7 svn:log V 30 Remove references to TELNET*. END K 10 svn:author V 5 markm K 8 svn:date V 27 1996-11-07T14:42:57.000000Z K 7 svn:log V 32 Secure telnet is now in eBones. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-07T14:44:01.000000Z K 7 svn:log V 154 Fix the message buffer mapping. This actually allows to increase the message buffer size in . Reviewed by: davidg,joerg Submitted by: bde END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-07T14:45:41.000000Z K 7 svn:log V 91 Merge Bruce's fixes for the message buffer mapping. Reviewed by: davidg Submitted by: bde END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-07T14:47:07.000000Z K 7 svn:log V 56 Merge Bruce's fast interrupt handler fix from rev 1.23. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-07T14:50:28.000000Z K 7 svn:log V 219 Merge my vi hack^H^H^H^Hframework into HEAD, so the various revisions should be in sync again. It's an env variable only, hence it's harmless and ignored without the actual changes in vi's Makefile. Suggested by: jkh END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-07T14:53:02.000000Z K 7 svn:log V 75 Activate the ISO 8859-1 -> CP437 screeenmap also in 2.2. Reviewed by: jkh END K 10 svn:author V 3 max K 8 svn:date V 27 1996-11-07T15:07:00.000000Z K 7 svn:log V 104 Since we use $ in our sample Makefiles, we can just say $Id$. Suggested by: asami END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T15:35:48.000000Z K 7 svn:log V 34 Complete switch-over to XF86 3.2. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T15:36:38.000000Z K 7 svn:log V 7 Merge. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T15:45:08.000000Z K 7 svn:log V 49 Install an evil work-around for /etc/sendmail.cw END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T15:45:59.000000Z K 7 svn:log V 68 Install an evil work-around for /etc/sendmail.cw [same as from 2.2] END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T16:34:12.000000Z K 7 svn:log V 44 This warning is too annoying now. Kill it. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T16:34:37.000000Z K 7 svn:log V 17 Merge from HEAD. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T16:40:10.000000Z K 7 svn:log V 60 Make sure current_chunk is 0 each time we enter the editor. END K 10 svn:author V 3 bde K 8 svn:date V 27 1996-11-07T17:53:06.000000Z K 7 svn:log V 97 Brought in fixes from rev.1.146 - don't write to the divisor latch registers unnecessarily, etc. END K 10 svn:author V 3 max K 8 svn:date V 27 1996-11-07T18:13:57.000000Z K 7 svn:log V 54 Typo, MATSHITA -> MATSUSHITA. 2.2 & 2.1.6 candidate. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T18:30:59.000000Z K 7 svn:log V 96 More cosmetic tweaks, get the prefixes adjusted properly to XFree86 3.2's new packaging scheme. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-07T18:31:38.000000Z K 7 svn:log V 7 Merge. END K 10 svn:author V 5 wpaul K 8 svn:date V 27 1996-11-08T01:42:02.000000Z K 7 svn:log V 409 - Avoid possible SEGVs: never call clnt_destroy() without checking for NULL RPC client handles. This should hopefully fix the problems Satoshi reported on -current. - Add socket descriptor sanity checks to _yp_unbind(). - Fix yp_order() so that it handles the RPC_PROCUNAVAIL error gracefully. NIS+ in YP compat mode doesn't support the YPPROC_ORDER procedure. This is a 2.2 candidate with bells on. END K 10 svn:author V 5 wpaul K 8 svn:date V 27 1996-11-08T02:06:30.000000Z K 7 svn:log V 219 - Make _yp_unbind() check for a NULL RPC client handle before calling clnt_destroy(). - Let yp_order() handle RPC_PROCUNAVAIL that might be returned from NIS+ servers in YP compat mode gracefully. 2.1.6 candidate END K 10 svn:author V 3 jdp K 8 svn:date V 27 1996-11-08T02:12:40.000000Z K 7 svn:log V 1039 Fixed a bug in the handling of the directories in the search path that is stored in the hints file. If that search path contained a non-existent directory (one, say, that had been removed), and "ldconfig -m /a/perfectly/good/directory" was run, ldconfig returned an error status without printing an error message. This caused some confusing bombs when installing ports, in particular. I changed it so that non-existent directories from the stored search path are silently ignored. Only non-existent directories named explicitly on the command line are treated as errors. Also, a diagnostic is printed if and only if an error status is returned. In an unrelated fix, ldconfig now silently ignores any directories named on the command line when the "-r" option is given. Formerly, these directories incorrectly made their way into the "search directories" line of the listing. It really should be an error to specify directories together with "-r", but I don't have time to fix the manual page in that way right now. 2.2 Candidate. END K 10 svn:author V 5 asami K 8 svn:date V 27 1996-11-08T02:38:44.000000Z K 7 svn:log V 254 Remove option I586_FAST_BCOPY. The code will be included by default if I586_CPU is defined. Note there is a runtime check so the code won't be run for non-Pentium CPUs anyway. 2.2 candidate, this code has been tested for almost half year in -current. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T05:21:01.000000Z K 7 svn:log V 83 Yikes! Add rawboot. Thanks-for-the-reminder-to: Don Lewis END K 10 svn:author V 5 peter K 8 svn:date V 27 1996-11-08T05:22:50.000000Z K 7 svn:log V 115 make the /etc/sendmail.cw file optional by default. Definately should go into 2.2.. Requested by: lots of people END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T05:38:27.000000Z K 7 svn:log V 123 1. Eliminate the sendmail.cw bogon again. Peter fixed it correctly. 2. Preload the ldconfig cache before calling X setup. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T05:39:46.000000Z K 7 svn:log V 72 It slices! It dices! It surges! It merges! NOW how much would you pay? END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T06:05:11.000000Z K 7 svn:log V 90 Remove an extraneous malloc() definition. Submitted-By: Don Lewis END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T06:08:10.000000Z K 7 svn:log V 101 Build c++rt0.o before libg++, just as in -current. Pointed-Out-By: Don Lewis END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T06:09:44.000000Z K 7 svn:log V 75 include distribute target. Submitted-By: Don Lewis END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T06:33:38.000000Z K 7 svn:log V 16 Add bsd.kern.mk END K 10 svn:author V 2 tg K 8 svn:date V 27 1996-11-08T08:10:57.000000Z K 7 svn:log V 54 Add Mats Lofkvist for his elk port update (ages ago). END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T09:30:44.000000Z K 7 svn:log V 58 Adapt to gcc 2.6.3 and bring one change across from HEAD. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T12:56:36.000000Z K 7 svn:log V 331 Wow, make the vmstat display actually usable! This is a cosmetic fix, but low-impact and worth the comeasurate risk. I [jkh] can't even test it in -current since systat's vmstat display is completely broken there at the moment, printing out: "The alternate system clock has died!" Submitted-By: Heiko Blume END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T13:09:27.000000Z K 7 svn:log V 17 Merge with 2_1_0 END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T13:26:32.000000Z K 7 svn:log V 79 Another distribute target fix. Submitted-By: Don Lewis END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T13:27:00.000000Z K 7 svn:log V 17 Merge from 2_1_0 END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T13:39:42.000000Z K 7 svn:log V 49 Merge revision 1.23 from HEAD. Requested-By: bde END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T18:31:47.000000Z K 7 svn:log V 43 Bring in asprintf*() prototypes from HEAD. END K 10 svn:author V 3 bde K 8 svn:date V 27 1996-11-08T18:50:09.000000Z K 7 svn:log V 609 Removed gratuitous differences between ext2_readwrite.c and ufs_readwrite.c. This fixes several bugs and one missing feature: - cluster_read() was needlessly used for reading files of size exactly 1 block. - EFAULT errors for read didn't terminate the loop. This was probably harmless. - IO_VMIO handling was missing near line 275. I don't know what this does. - B_CLUSTEROK was only set if (doclusterwrite) nead line 293. This was harmless, if only because another bug prevents doclusterwrite from being 0. - MNT_NOATIME wasn't implemented. This should be in 2.2, of course. Reviewed by: davidg END K 10 svn:author V 3 bde K 8 svn:date V 27 1996-11-08T19:06:34.000000Z K 7 svn:log V 211 Fixed spacefree calculation in ext2_direnter(). This bug sometimes caused panics. This should be in 2.2, of course. Submitted by: davidg Obtained from: bouyer@antioche.ibp.fr (Manuel BOUYER) (fix for NetBSD) END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-08T19:09:45.000000Z K 7 svn:log V 85 Remove a mkdir in doTARBALL which, as far as I can tell, is not needed for anything. END K 10 svn:author V 5 guido K 8 svn:date V 27 1996-11-08T21:40:16.000000Z K 7 svn:log V 33 Retrofit rwhod security patches. END K 10 svn:author V 5 guido K 8 svn:date V 27 1996-11-08T21:49:52.000000Z K 7 svn:log V 33 Retrofit rwhod security patches. END K 10 svn:author V 5 guido K 8 svn:date V 27 1996-11-08T21:56:21.000000Z K 7 svn:log V 103 Add add Naoki Hamada to the contributors, because of his work on the vx driver. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-08T23:20:38.000000Z K 7 svn:log V 77 Merge from rev 1.16: fix the ldconfig handling for non-existant directories. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-08T23:21:17.000000Z K 7 svn:log V 48 Merge the fix from rev 1.25: sanity checks etc. END K 10 svn:author V 2 se K 8 svn:date V 27 1996-11-08T23:46:04.000000Z K 7 svn:log V 359 Only access the specific interrupt status registers if required. This follows more closely the suggestions in the latest NCR docs, and has been running on my system for weeks with no problem. It does improve the quality of diagnostic messages and does allow to better understand the sequence of events in case of an error. This should go into 2.2 and 2.1.6. END K 10 svn:author V 6 obrien K 8 svn:date V 27 1996-11-08T23:57:04.000000Z K 7 svn:log V 63 Added Oly (oly@world.std.com) for his port of lout and xripple END K 10 svn:author V 5 asami K 8 svn:date V 27 1996-11-09T00:18:25.000000Z K 7 svn:log V 145 (1) Update (2) Don't depend on BOOTSEG (3) Change BOOTSEG from 0x9000 to 0x1000 Should be in 2.2. Submitted by: FreeBSD(98) Development Team END K 10 svn:author V 5 asami K 8 svn:date V 27 1996-11-09T00:39:59.000000Z K 7 svn:log V 90 Re-sync with -current. Should be in 2.2. Submitted by: The FreeBSD(98) Development Team END K 10 svn:author V 7 jfieber K 8 svn:date V 27 1996-11-09T02:04:05.000000Z K 7 svn:log V 164 A skeletal, but functional Docbook to HTML translation spec. The Docbook DTD will follow once I investigate how "contrib" works and if it should be imported there. END K 10 svn:author V 3 max K 8 svn:date V 27 1996-11-09T02:44:11.000000Z K 7 svn:log V 98 Back out my previous change. It probably broke the support for the device. Pointed-Out by: joerg END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-09T03:03:47.000000Z K 7 svn:log V 125 Fix a syntax error in the Belgian keymap. It caused a wrong mapping and a superfluous `i' on stdout (lex' default rule...). END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-09T03:05:40.000000Z K 7 svn:log V 40 Merge from rev 1.3: fix a syntax error. END K 10 svn:author V 5 dyson K 8 svn:date V 27 1996-11-09T03:54:25.000000Z K 7 svn:log V 117 Fix an ordering bug -- pmap_remove_pages should be called BEFORE vm_map_remove, not after... 2.2-RELEASE candidate. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T08:41:50.000000Z K 7 svn:log V 29 Merge r1.83 Requested-By: se END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T09:23:09.000000Z K 7 svn:log V 48 Add an entry for minimum - that's a lot easier. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T09:23:25.000000Z K 7 svn:log V 6 Merge END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T09:24:30.000000Z K 7 svn:log V 25 Purely cosmetic cleanup. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T09:26:44.000000Z K 7 svn:log V 37 Merge in Bruce's cleanups from HEAD. END K 10 svn:author V 3 bde K 8 svn:date V 27 1996-11-09T10:25:04.000000Z K 7 svn:log V 236 Fixed lookup of ".." in checkpath. It always failed, so renames of directories to a different parent directory always failed. This bug was caused by 4.4Lite2 changing the directory format and ext2fs not keeping up. Should be in 2.2. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-09T10:42:28.000000Z K 7 svn:log V 71 Merge the fix from rev 1.48: fix an ordering bug. Requested by: dyson END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-09T11:57:40.000000Z K 7 svn:log V 490 If mounting the fixit floppy fails in the first place, warn the user but make a second attempt using MNT_FORCE, just in case it has been unclean from a previous crash. That's dangerous, but far better than keeping the despaired user standing in the rain... (Experienced admins can still fsck it then, and remount. Others will either totally crash, or incidentally succeed, without much further help possible...) Btw., mount(2) misses the description of MNT_FORCE for the mount syscall. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-09T12:00:15.000000Z K 7 svn:log V 81 Merge from HEAD: try mounting the fixit floppy with MNT_FORCE if all else fails. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-09T12:08:57.000000Z K 7 svn:log V 78 Merge from the 2.2 branch: call our `vi' by its right name. Detected by: jkh END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-09T12:13:56.000000Z K 7 svn:log V 77 Merge from rev 1.6: fix rename for different parent dirs. Requested by: bde END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T13:52:34.000000Z K 7 svn:log V 45 Merge list from HEAD as requested by Justin. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-09T16:16:19.000000Z K 7 svn:log V 124 Implement a -L option that dumps the compiled keymap (as C code) to stdout. The next commit to sysinstall will use this... END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-09T16:17:48.000000Z K 7 svn:log V 56 Merge the -L option from HEAD; sysinstall will need it. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-09T16:25:47.000000Z K 7 svn:log V 365 Several improvements to the `fixit' configuration: . Don't gzip the crunched binary by now; it just fits, and execution is a lot faster this way (it's truly demand-paged again). . Add more(1), ft(8), protocols(5), a stripped down services(5). . Improve the .profile, and make sysinstall actually use it again. Still no go for a 4 MB configuration though. :-( END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-09T16:34:21.000000Z K 7 svn:log V 59 Merge from HEAD: various improvements to the fixit floppy. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-09T16:47:08.000000Z K 7 svn:log V 563 Sysinstall now offers the installation of a keymap early in the game! Not all mappings are supported, most languages come only with one encoding since this should be sufficient to get up & running in using sysinstall, and we are already pretty tight on space. (My previous commit has already bumped the boot MFS size by another 50 KB for this.) This feature requires the `kbdcontrol -L' i've just committed. Plain text keymaps and the entire scanner are overkill for sysinstall. Also updated the list of available keymaps while i was at it. Reviewed by: jkh END K 10 svn:author V 7 cvs2svn K 8 svn:date V 27 1996-11-09T16:47:09.000000Z K 7 svn:log V 70 This commit was manufactured by cvs2svn to create branch 'RELENG_2_2'. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-09T16:49:16.000000Z K 7 svn:log V 72 Merge from HEAD: enable the installation of a keymap early in the game. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-09T16:54:33.000000Z K 7 svn:log V 191 Merge from the 2.2 branch: avoid including the API stuff when building vi for the fixit floppy. Perhaps we'll find a more elegant solution some day, by now it has proven to work reasonably. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T18:12:17.000000Z K 7 svn:log V 289 1. Take out console setup in options now that Joerg has moved it into the main menu. 2. Conditionalized a few small things which needed it. 3. Put PC98 X servers in their own menu, there are so many of them now. 4. Rampaged on the menus.c file in general, reformatting and cleaning up. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T18:13:38.000000Z K 7 svn:log V 17 Merge from HEAD. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T19:26:01.000000Z K 7 svn:log V 94 1. Saner ldconfig path. 2. Choose DIST_XF86_CFG in defaults; XF86Setup isn't happy otherwise. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T19:26:19.000000Z K 7 svn:log V 7 Merge. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T19:47:24.000000Z K 7 svn:log V 48 Whups, we need the VGA16 server by default too. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T19:56:25.000000Z K 7 svn:log V 7 Merge. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T20:21:47.000000Z K 7 svn:log V 110 Back out previous change - I don't think it's necessary now that Poul's shown me the real culprit is rawboot. END K 10 svn:author V 3 jdp K 8 svn:date V 27 1996-11-09T20:26:19.000000Z K 7 svn:log V 166 Document that "ldconfig -m" rescans all directories previously entered into the hints file, in addition to the directories named on the command line. 2.2 Candidate. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T20:32:26.000000Z K 7 svn:log V 118 Commit a saner version of this without so much disklabel trickery. I'll stick to the old way of doing things for 2.1. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T20:50:23.000000Z K 7 svn:log V 137 Disgusting kludge to make depend rule work again in the fact of Joerg's keymap.[ch] changes. If there's a better way, I'd like to know. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T21:04:03.000000Z K 7 svn:log V 134 Merge Makefile fix, an old disks.c fix which somehow didn't get in and kill the mysterious extra copy of keymap.c appended to itself. END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-09T21:16:08.000000Z K 7 svn:log V 20 Merge from -current END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-09T21:16:09.000000Z K 7 svn:log V 6 Merge END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-09T21:36:42.000000Z K 7 svn:log V 32 Remove advansys driver from 2.2 END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-09T21:51:14.000000Z K 7 svn:log V 20 Merge from -current END K 10 svn:author V 4 ache K 8 svn:date V 27 1996-11-10T02:21:05.000000Z K 7 svn:log V 43 Add mount_msdos(8) reference 2.2 candidate END K 10 svn:author V 4 ache K 8 svn:date V 27 1996-11-10T02:23:33.000000Z K 7 svn:log V 28 pcfs -> msdos 2.2 candidate END K 10 svn:author V 5 wosch K 8 svn:date V 27 1996-11-10T04:46:58.000000Z K 7 svn:log V 169 Add sendmail option '-oi' -oi Do not take dots on a line by themselves as a message terminator. Now this crontab entry works: * * * * * echo foo; echo .; echo blah END K 10 svn:author V 6 obrien K 8 svn:date V 27 1996-11-10T05:56:43.000000Z K 7 svn:log V 339 Added to MNT_FORCE option description after seeing Bruce's commit message dealing w/the fixit floppy. Also added the MNT_RELOAD, MNT_WANTRDWR, MNT_ASYNC, MNT_NOATIME, MOUNT_UNION flags. Someone might want to check my description of MNT_RELOAD. 2.2-R candidate. Not a 2.1.6-R candidate -- some current flags aren't in 2.1.5-R's version. END K 10 svn:author V 3 hsu K 8 svn:date V 27 1996-11-10T06:13:12.000000Z K 7 svn:log V 39 Remove unneeded #include . END K 10 svn:author V 6 fenner K 8 svn:date V 27 1996-11-10T07:37:24.000000Z K 7 svn:log V 122 Re-enable the TCP SYN-attack protection code. I was the one who didn't understand the socket state flag. 2.2 candidate. END K 10 svn:author V 3 bde K 8 svn:date V 27 1996-11-10T08:30:18.000000Z K 7 svn:log V 135 Initialize interrupt counters. The boot time values were displayed as ***. Should be in 2.2. Pointed out by: /etc/malloc.conf -> AJ END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-10T09:24:53.000000Z K 7 svn:log V 58 Merge David O'Brien's description of the new mount flags. END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-10T09:26:58.000000Z K 7 svn:log V 64 Being here, also add the description of MNT_FORCE for mount(2). END K 10 svn:author V 5 joerg K 8 svn:date V 27 1996-11-10T09:28:20.000000Z K 7 svn:log V 60 Merge from the 2.2 branch: describe MNT_FORCE for mount(2). END K 10 svn:author V 5 asami K 8 svn:date V 27 1996-11-10T11:21:15.000000Z K 7 svn:log V 118 Put ccd in the right place. Sorry about the screwup. Reported by: Andre Albsmeier END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-10T12:03:38.000000Z K 7 svn:log V 20 Merge from -current END K 10 svn:author V 3 phk K 8 svn:date V 27 1996-11-10T12:15:19.000000Z K 7 svn:log V 20 Merge from -current END K 10 svn:author V 2 dg K 8 svn:date V 27 1996-11-10T12:32:35.000000Z K 7 svn:log V 275 Updated to a newer version from Matt Thomas. This is the same version that -current is based on. This has been tested on wcarchive for about 12 hours and in -current for a couple of months. This update should fix many of the compatibility problems that people have reported. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-10T12:45:04.000000Z K 7 svn:log V 40 Bring in Joerg's changes from the HEAD. END K 10 svn:author V 5 peter K 8 svn:date V 27 1996-11-10T13:06:14.000000Z K 7 svn:log V 535 The kerberised network services should only be active in inetd.conf if kerberos is installed. So far as I'm aware, kerberos aware clients detect ECONNREFUSED and (if allowed) fall back to the non-kerberos servers. They do not know how to interpret messages such as "rlogind: unknown option -k". I believe Garrett also mentioned this. Unfortunately, this adds an extra step to bringing up kerberos. It also stops /var/log/messages getting quite so many useless (and confusing) error messages when somebody does a port scan on you. END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-10T13:11:30.000000Z K 7 svn:log V 35 Merge r1.24 from HEAD (good idea). END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-10T13:14:49.000000Z K 7 svn:log V 117 Another merge from HEAD before rolling a 2.1.6 release candidate (Tuesday is still the cut-off date for bug fixes!). END K 10 svn:author V 2 dg K 8 svn:date V 27 1996-11-10T13:36:46.000000Z K 7 svn:log V 54 Put the packet error printf inside #ifdef DIAGNOSTIC. END K 10 svn:author V 5 peter K 8 svn:date V 27 1996-11-10T14:46:50.000000Z K 7 svn:log V 112 Cosmetic bugfix. fetch was modifying it's argv[] strings in place, which caused ps(1) to show strange things.. END K 10 svn:author V 4 paul K 8 svn:date V 27 1996-11-10T15:49:49.000000Z K 7 svn:log V 72 Added a missing \ for installing perl modules. 2.2 and 2.1.6 candidate. END K 10 svn:author V 4 nate K 8 svn:date V 27 1996-11-10T16:44:13.000000Z K 7 svn:log V 296 Allow us to enable the 'XT_KEYBOARD' code using a configuration flag. This allows the user to add modify syscons's configuration flags using UserConfig that will allow older/quirky hardware (most notably older IBM ThinkPad laptops) to work with the standard boot kernel. Inspired by: The Nomads END K 10 svn:author V 4 nate K 8 svn:date V 27 1996-11-10T16:57:43.000000Z K 7 svn:log V 399 From -current: Allow us to enable the 'XT_KEYBOARD' code using a configuration flag. This allows the user to add modify syscons's configuration flags using UserConfig that will allow older/quirky hardware (most notably older IBM ThinkPad laptops) to work with the standard boot kernel. (note: I used the same flag as in -current so the documentation would be the same) Inspired by: The Nomads END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-10T19:12:48.000000Z K 7 svn:log V 243 Incorporate part of the XT_KEYBOARD change that Nate missed. Please be more careful when modifying code in this branch! If my 8 hour release build had died (and it would have, in a few more hours) because of this, I'd have been displeased. :) END K 10 svn:author V 3 jkh K 8 svn:date V 27 1996-11-10T20:44:59.000000Z K 7 svn:log V 36 Whoops, merge the Makefile too! :-} END K 10 svn:author V 6 julian K 8 svn:date V 27 1996-11-10T21:07:27.000000Z K 7 svn:log V 126 Reviewed by: Bill fenner Submitted by: Archie Cobbs (archie@whistle.com) changes to make inetd compile cleaner under -Wall END K 10 svn:author V 6 julian K 8 svn:date V 27 1996-11-10T21:12:44.000000Z K 7 svn:log V 409 Reviewed by: Bill fenner Submitted by: Archie Cobbs (Archie@whistle.com) Changes to allow inted to control the number of servers to start on each service. This is a defence against a denial of service attack in which the system is made unusable by an external party. It also allows the behaviour of small memory systems to be more accuratly predicted, by bounding the extent to which processes can multiply. END K 10 svn:author V 3 jdp K 8 svn:date V 27 1996-11-10T21:51:31.000000Z K 7 svn:log V 74 Correct many errors of commission and omission in the example CVSupfiles. END K 10 svn:author V 6 fenner K 8 svn:date V 27 1996-11-11T03:50:15.000000Z K 7 svn:log V 458 Update to the unreleased mrouted 3.8a . This includes a minor endian-ness fix, Router Alert options on IGMP messages, and a new keyword, "advert_metric", for fine-tuning tunnel metrics. This also includes a new mtrace, which is also unreleased but builds significantly on the experiences of users' troubles with using and understanding mtrace in release 3.8 . (unreleased does not, of course, mean untested!) This is a candidate for both 2.2 and 2.1.6 . END K 10 svn:author V 5 dyson K 8 svn:date V 27 1996-11-11T04:20:19.000000Z K 7 svn:log V 163 Support the PG_G flag on Pentium-Pro processors. This pretty much eliminates the unnecessary unmapping of the kernel during context switches and during invtlb... END K 10 svn:author V 6 fenner K 8 svn:date V 27 1996-11-11T04:56:32.000000Z K 7 svn:log V 278 Add the IP_RECVIF socket option, which supplies a packet's incoming interface using a sockaddr_dl. Fix the other packet-information socket options (SO_TIMESTAMP, IP_RECVDSTADDR) to work for multicast UDP and raw sockets as well. (They previously only worked for unicast UDP). END K 10 svn:author V 5 gibbs K 8 svn:date V 27 1996-11-11T05:16:41.000000Z K 7 svn:log V 277 Fix two problems with SCB Paging. 1) get_free_or_disc_scb was not being passed its argument correctly in one case 2) Add protection in the form of the QOUTQCNT variable to prevent overflowing the QOUTFIFO. This should make SCB Paging work. Really, I mean it now. 8-) END K 10 svn:author V 5 gibbs K 8 svn:date V 27 1996-11-11T05:17:34.000000Z K 7 svn:log V 97 Add the AHC_FORCE_PIO option. Update comment on AHC_SCBPAGING_ENABLE since I think it works now. END K 10 svn:author V 5 gibbs K 8 svn:date V 27 1996-11-11T05:21:27.000000Z K 7 svn:log V 414 Clean up the memory mapped/Programmed I/O stuff so that the driver completely uses one or the other. This required some changes to the ahc_reset() function, and how early the probes had to allocate their softc. Turn the AHC_IN/OUT* macros into inline functions and lowercase their names to indicate this change. Geting AHC_OUTSB to work as a macro doing conditional memory mapped I/O would have been too gross. END K 10 svn:author V 5 gibbs K 8 svn:date V 27 1996-11-11T05:24:46.000000Z K 7 svn:log V 805 Clean up the memory mapped/Programmed I/O stuff so that the driver completely uses one or the other. This required some changes to the ahc_reset() function, and how early the probes had to allocate their softc. Turn the AHC_IN/OUT* macros into inline functions and lowercase their names to indicate this change. Geting AHC_OUTSB to work as a macro doing conditional memory mapped I/O would have been too gross. Stop setting STPWEN in the main driver and let the PCI front end do it instead. It knows better. Add the clearing of the QOUTQCNT variable during command complete processing in the SCB paging case. Go back to doing unconditional retries for the QUEUE FULL status condition. This is really a kludge, but the code to handle it properly is on the SCSI branch and will not make it into 2.2. END K 10 svn:author V 5 gibbs K 8 svn:date V 27 1996-11-11T05:26:14.000000Z K 7 svn:log V 707 Clean up the memory mapped/Programmed I/O stuff so that the driver completely uses one or the other. This required some changes to the ahc_reset() function, and how early the probes had to allocate their softc. Turn the AHC_IN/OUT* macros into inline functions and lowercase their names to indicate this change. Geting AHC_OUTSB to work as a macro doing conditional memory mapped I/O would have been too gross. Be smart about the STPWEN control bit in SCFRCTL1. It should only be set if the low byte of the bus is to be terminated. We figure this out either by "caching" the value left over from the BIOS setup before we reset the card or by using the values stored in the seeprom if it is availible. END K 10 svn:author V 5 asami K 8 svn:date V 27 1996-11-11T06:50:45.000000Z K 7 svn:log V 445 Change "CATEGORIES+=" to "CATEGORIES=" in sample Makefiles. It was "+=" originally because (as I understand) Jordan used a sed script (or was it perl?) to edit all the ports Makefiles automatically and he wanted to make sure multiple CATEGORIES lines (they were inserted after DISTNAME or something, there shouldn't have been multiple of them to begin with but that's another story) won't be stepping on each other's toes. Reminded by: obrien END K 10 svn:author V 4 ache K 8 svn:date V 27 1996-11-11T08:28:47.000000Z K 7 svn:log V 216 kvm_malloc: When malloc fails. don't try to memset NULL pointer, it cause core dump Replace malloc+memset with calloc, theoretically it can do some optimization of zeroing process internally Improve error diagnostic END K 10 svn:author V 3 hsu K 8 svn:date V 27 1996-11-11T09:05:29.000000Z K 7 svn:log V 116 Make pthread_getspecific() compliant with the final IEEE pthreads specification: return parameter passing changed. END K 10 svn:author V 3 hsu K 8 svn:date V 27 1996-11-11T09:07:05.000000Z K 7 svn:log V 85 Moved enum pthread_mutextype to pthread.h. Add pthread_mutexattr_default definition. END K 10 svn:author V 3 hsu K 8 svn:date V 27 1996-11-11T09:08:11.000000Z K 7 svn:log V 146 Add uthread_attr_init.c, uthread_attr_setstacksize.c, uthread_mattr_init.c, uthread_mattr_kind_np.c, uthread_multi_np.c, and uthread_single_np.c. END K 10 svn:author V 3 hsu K 8 svn:date V 27 1996-11-11T09:09:30.000000Z K 7 svn:log V 65 Add pthread_mutexattr_init() and pthread_mutexattr_setkind_np(). END K 10 svn:author V 6 msmith K 8 svn:date V 27 1996-11-11T09:09:37.000000Z K 7 svn:log V 383 Update the database of known devices (people, please consider this when you are adding new drivers...) to match, as best I can tell, majors.i386. Improve behaviour when attempting to save changes for devices that should not be changeable. Now correclty avoids non-device items, PCI devices and devices with no isa_device structure. Submitted by: (observations from) joerg, bde END K 10 svn:author V 3 hsu K 8 svn:date V 27 1996-11-11T09:11:59.000000Z K 7 svn:log V 42 Parameters pthread_getspecific() changed. END K 10 svn:author V 3 hsu K 8 svn:date V 27 1996-11-11T09:14:24.000000Z K 7 svn:log V 169 Parameters for localtime_r() and gmtime_r() definitions changed. Parameters for pthread_getspecific() call changed. pthread_keycreate() renamed to pthread_key_create(). END K 10 svn:author V 3 hsu K 8 svn:date V 27 1996-11-11T09:19:54.000000Z K 7 svn:log V 230 Moved pthread_mutexattr_default inside #ifndef PTHREAD_KERNEL now that we use it in the uthreads implementation. Moved enum pthread_mutextype here from libc_r/uthread/pthread_private.h. Change prototype for pthread_getspecific(). END K 10 svn:author V 3 hsu K 8 svn:date V 27 1996-11-11T09:21:19.000000Z K 7 svn:log V 77 Prototype pthread_mutexattr_getkind_np() and pthread_mutexattr_setkind_np(). END