-85880 120 624 133 128 210 283 385 278 297 211 137 226 203 971 235 137 1059 164 380 131 239 170 138 527 197 531 123 337 198 158 490 155 128 150 124 378 319 223 160 174 347 239 176 248 117 248 256 291 1737 375 218 221 170 239 150 212 212 400 152 326 196 207 201 109 110 226 127 184 138 139 207 193 548 143 279 280 226 135 194 131 122 146 173 170 133 182 183 137 382 164 119 238 328 116 121 131 125 145 226 178 212 203 150 399 122 126 124 237 808 195 210 280 505 373 267 348 142 176 358 1506 129 K 10 svn:author V 7 rwatson K 8 svn:date V 27 2001-11-02T16:41:06.000000Z K 7 svn:log V 527 o Add a comment to p_candebug() noting that the P_INEXEC check should really be moved elsewhere: p_candebug() encapsulates the security policy decision, whereas the P_INEXEC check has to do with "correctness" regarding race conditions, rather than security policy. Example: even if no security protections were enforced (the "uids are advisory" model), removing P_INEXEC could result in incorrect operation due to races on credential evaluation and modification during execve(). Obtained from: TrustedBSD Project END K 10 svn:author V 3 imp K 8 svn:date V 27 2001-11-02T16:42:11.000000Z K 7 svn:log V 41 MFC's NOTES: Fix typo: driver vs drivers END K 10 svn:author V 6 jlemon K 8 svn:date V 27 2001-11-02T16:45:05.000000Z K 7 svn:log V 33 MFC: r1.347; 28800 baud support. END K 10 svn:author V 6 jlemon K 8 svn:date V 27 2001-11-02T16:50:42.000000Z K 7 svn:log V 114 MFC; sync to -current. Brings in ucode support, dynamic sysctls, and expanded fix for dynamic standby mode. END K 10 svn:author V 6 jlemon K 8 svn:date V 27 2001-11-02T17:04:32.000000Z K 7 svn:log V 187 + Fix another possible vn_close race, in the same fashion as r1.95. + Check that the cached vnode type != VBAD before calling devsw(), this can happen if the vnode has been revoked. END K 10 svn:author V 3 imp K 8 svn:date V 27 2001-11-02T17:28:43.000000Z K 7 svn:log V 292 Always set unit number to -1 unless some other unit is specified in the config file. This fixes the breakage caused by the recent change in the behavior of device_add_child for ata (which shows soren's reservations were well founded). Submitted by: OGAWA Takaya END K 10 svn:author V 3 imp K 8 svn:date V 27 2001-11-02T17:31:01.000000Z K 7 svn:log V 185 Print a warning when device_add_child returns NULL. This used to be impossible at this point, but now it apparently is. Grump. Submitted by: OGAWA Takaya END K 10 svn:author V 3 imp K 8 svn:date V 27 2001-11-02T17:33:06.000000Z K 7 svn:log V 204 Don't hide the failure to allocate device behind boot verbose. It is still telling us of real problems so should remain until it stops doing that. Submitted by: OGAWA Takaya END K 10 svn:author V 4 mike K 8 svn:date V 27 2001-11-02T17:42:03.000000Z K 7 svn:log V 117 Rather than just change the arguments to suser() change the function to suser_xxx() as well. Pointy hat to: rwatson END K 10 svn:author V 6 jlemon K 8 svn:date V 27 2001-11-02T17:48:00.000000Z K 7 svn:log V 42 MFC: r1.65; fixes for the if_le.c driver. END K 10 svn:author V 3 phk K 8 svn:date V 27 2001-11-02T17:58:26.000000Z K 7 svn:log V 133 Reserve 378 for the new mount syscall Maxime Henrion is working on. (This is to get us more than 32 mountoptions). END K 10 svn:author V 3 phk K 8 svn:date V 27 2001-11-02T17:59:23.000000Z K 7 svn:log V 110 Add nmount() stub function and regenerate the syscall-glue which should not need to check in generated files. END K 10 svn:author V 4 mike K 8 svn:date V 27 2001-11-02T18:05:43.000000Z K 7 svn:log V 877 o Add new header . o Make a symbolic link to . o Move most of into , as per C99. o Remove . o Adjust includes in sys/types.h and boot/efi/include/ia64/efibind.h to reflect new location of integer types in . o Remove previously symbolicly linked , instead create a new file. o Add MD headers from NetBSD. o Include in , as required by C99; and include in , to fill in the remaining requirements for . o Add additional integer types in and which are included via . Partially obtain from: NetBSD Tested on: alpha, i386 Discussed on: freebsd-standards@bostonradio.org Reviewed by: bde, fenner, obrien, wollman END K 10 svn:author V 5 wpaul K 8 svn:date V 27 2001-11-02T18:34:58.000000Z K 7 svn:log V 140 MFC: fix mbuf leak in ip_output(). (sending packets with IP options which cause ip_output() to return with error leaks one mbuf per packet) END K 10 svn:author V 3 phk K 8 svn:date V 27 2001-11-02T18:35:54.000000Z K 7 svn:log V 45 Add empty shell for nmount syscall (take 2!) END K 10 svn:author V 7 rwatson K 8 svn:date V 27 2001-11-02T18:44:50.000000Z K 7 svn:log V 962 o Introduce group subset test, which limits the ability of a process to debug another process based on their respective {effective,additional, saved,real} gid's. p1 is only permitted to debug p2 if its effective gids (egid + additional groups) are a strict superset of the gids of p2. This implements properly the security test previously incorrectly implemented in kern_ktrace.c, and is consistent with the kernel security policy (although might be slightly confusing for those more familiar with the userland policy). o Restructure p_candebug() logic so that various results are generated comparing uids, gids, credential changes, and then composed in a single check before testing for privilege. These tests encapsulate the "BSD" inter-process debugging policy. Other non-BSD checks remain seperate. Additional comments are added. Submitted by: tmm, rwatson Obtained from: TrustedBSD Project Reviewed by: petef, tmm, rwatson END K 10 svn:author V 3 phk K 8 svn:date V 27 2001-11-02T19:12:06.000000Z K 7 svn:log V 72 Argh! patch added the nmount at the bottom first time around. Take 3! END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T19:42:07.000000Z K 7 svn:log V 284 Add a NOP nl_langinfo(). The purpose is to get an "nl_langinfo" symbol into libc for some minor compatibility with 5-CURRENT binaries. Its external exposure is kept down to a minium. If one ln's libc.so.4 to libc.so.5 a surprising number of 5-CURRENT binaries will run on RELENG_4. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T21:06:08.000000Z K 7 svn:log V 36 Update vendor branch to gawk-3.1.0. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T21:06:08.000000Z K 7 svn:log V 143 This commit was generated by cvs2svn to compensate for changes in r85898, 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 2001-11-02T21:06:09.000000Z K 7 svn:log V 74 This commit was manufactured by cvs2svn to create tag 'awk-vendor-v3_1_0'. END K 10 svn:author V 5 silby K 8 svn:date V 27 2001-11-02T21:11:36.000000Z K 7 svn:log V 44 Remove an extra " that crept into a string. END K 10 svn:author V 7 rwatson K 8 svn:date V 27 2001-11-02T21:11:41.000000Z K 7 svn:log V 430 o Remove the local temporary variable "struct proc *p" from vfs_mount() in vfs_syscalls.c. Although it did save some indirection, many of those savings will be obscured with the impending commit of suser() changes, and the result is increased code complexity. Also, once p->p_ucred and td->td_ucred are distinguished, this will make vfs_mount() use the correct thread credential, rather than the process credential. END K 10 svn:author V 3 imp K 8 svn:date V 27 2001-11-02T21:14:17.000000Z K 7 svn:log V 104 Back out the -w, option strict and our($...). They don't work for me and have broken the kernel build. END K 10 svn:author V 7 rwatson K 8 svn:date V 27 2001-11-02T21:20:05.000000Z K 7 svn:log V 434 o Remove (struct proc *p = td->td_proc) indirection in ipcperm(), as suser_td(td) works as well as suser_xxx(NULL, p->p_ucred, 0); This simplifies upcoming changes to suser(), and causes this code to use the right credential (well, largely) once the td->td_ucred changes are complete. There remains some redundancy and oddness in this code, which should be rethought after the next batch of suser and credential changes. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T21:20:18.000000Z K 7 svn:log V 28 Well spell it "gnuregex.h". END K 10 svn:author V 3 imp K 8 svn:date V 27 2001-11-02T21:26:07.000000Z K 7 svn:log V 244 Better error messages for the cases where device_add_child fails. We should also whine if the old pccardd is used, but that's a little harder than it sounds. This also has the effect of fixing a typo that was in the last version I committed. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T21:26:14.000000Z K 7 svn:log V 102 Merge revs 1.9 (don't choke on long lines), rev 1.3 (use srandom(time() ^ getpid())) into Gawk 3.1.0. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T21:34:12.000000Z K 7 svn:log V 63 Merge revision 1.2 (unspam l10n ranges check) into Gawk 3.1.0. END K 10 svn:author V 3 imp K 8 svn:date V 27 2001-11-02T21:34:20.000000Z K 7 svn:log V 397 Factor the common parts of the Makefile.foo files. This introduces two new files: kern.pre.mk, which contains most of the definitions, and kern.post.mk, which contains most of the rules. I've tested this on i386 and pc98. I have had feedback on the sparc64 port, but no reports from anybody on alpha, ia64 or powerpc. I appologize in advance if I've broken you. Reviewed by: jake, jhb, arch@ END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T21:37:05.000000Z K 7 svn:log V 60 We use the stock 3.1.0 file (other than needing $FreeBSD$). END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T21:39:18.000000Z K 7 svn:log V 33 We use the stock 3.1.0 file now. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T21:40:02.000000Z K 7 svn:log V 55 The bug rev 1.6 fixed was fixed in the vendor sources. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T21:44:42.000000Z K 7 svn:log V 29 We use the stock 3.1.0 file. END K 10 svn:author V 3 imp K 8 svn:date V 27 2001-11-02T21:50:15.000000Z K 7 svn:log V 285 Arrgh. A clean pc98 build failed due to bogons on my part :-(. Fix it by putting back the link of machine to sys/i386/include rather than ../../include (aka sys/pc98/include). I had a stale machine link on my first test. Not sure what the "right" fix is, but this unbreaks things. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T21:57:00.000000Z K 7 svn:log V 223 We use the stock version of this file now, but since someone foolishly took about 15 files off the vendor branch for what turned out to be good reason a Gawk update takes an order of magnitude more effort than it should... END K 10 svn:author V 3 cjc K 8 svn:date V 27 2001-11-02T22:00:17.000000Z K 7 svn:log V 130 Fix a bug I introduced yesterday. People who built world since the previous commit yesterday may wish to check /var/run for junk. END K 10 svn:author V 3 des K 8 svn:date V 27 2001-11-02T23:15:13.000000Z K 7 svn:log V 68 Add the as-yet-unused S_ALLSTOPS which I forgot in previous commit. END K 10 svn:author V 3 bsd K 8 svn:date V 27 2001-11-02T23:20:43.000000Z K 7 svn:log V 82 MFC Rev 1.13, ensure that /var/log/lastlog exists so that login doesn't complain. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T23:41:29.000000Z K 7 svn:log V 251 I'd put down $10 says the author did a last minute tweak to awkgram.y (which had an error in it), and applied it by hand to awkgram.c (getting it right) rather than really generating an new awkgram.c properly using Bison/YACC... Fix his mistake here. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T23:41:29.000000Z K 7 svn:log V 143 This commit was generated by cvs2svn to compensate for changes in r85919, 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 2001-11-02T23:41:30.000000Z K 7 svn:log V 80 This commit was manufactured by cvs2svn to create tag 'awk-vendor-v3_1_0_fixed'. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T23:43:07.000000Z K 7 svn:log V 152 Bring Gawk back. There just isn't any other POSIX compliant AWK out there. The biggest thing missing from Bell-Labs AWK is the character class regexes. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T23:46:57.000000Z K 7 svn:log V 22 Update to Gawk 3.1.0. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-02T23:48:32.000000Z K 7 svn:log V 152 Bring Gawk back. There just isn't any other POSIX compliant AWK out there. The biggest thing missing from Bell-Labs AWK is the character class regexes. END K 10 svn:author V 3 des K 8 svn:date V 27 2001-11-02T23:50:00.000000Z K 7 svn:log V 163 We have a _SIG_VALID() macro, so use it instead of duplicating the test all over the place. Also replace a printf() + panic() with a KASSERT(). Reviewed by: jhb END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T00:33:07.000000Z K 7 svn:log V 196 MFC: Dont hard-code knowledge about where the stack starts by peering into kernel #define's (GRR!!). Use the kern.usrstack sysctl. This stops libc_r programs from exploding when KVM is changed. END K 10 svn:author V 5 luigi K 8 svn:date V 27 2001-11-03T00:36:11.000000Z K 7 svn:log V 1641 Mega-MFC for ipfw/bridge/dummynet features and fixes added over the past couple of months: * merge of ipfw rule descriptor and chain pointer. No functional change, but the internal data structures and code are way more readable; * BillF code to make ipfw/dummynet/bridge KLD'able. NOTA BENE: this still has some rough edges, which are mostly due to bugs in kldload() rather than in this code. * add a new type of dynamic rule that lets you limit the number of simultaneous connections matching certain criteria (with the usual aggregation based on port/address masks) * fix spl*() protection in same parts of the code; This code also includes some minor bugfixes and code cleanup that I will port to CURRENT as soon as i have a chance. I have tested the code as much as i could, but there is really a million combinations so I might have missed some corner case. Please report if you have problem building things. The only thing known not to work is bridge.ko -- it does forward correctly, but packets directed to the bridge itself are only received from one interface (i suspect some missing initialization), and there are some other issues at unloading time. Please use the statically compiled code for the time being. NOTE ON KLD: It appears that kldload/unload is unable to handle the case of (erroneously) loading/unloading a module which is already compiled in. What happens is that load fails, but the module is listed as loaded, and then the system panics if you attempt an unloading of the module. This problem need fixing in the module loading/unloading code, which is not in my area of competence. END K 10 svn:author V 5 luigi K 8 svn:date V 27 2001-11-03T00:42:01.000000Z K 7 svn:log V 280 MFC: entries for kld'able versions of bridge and dummynet. NOTE: i am not sure about the dependency in dummynet/Makefile, but will fix this later. Also, in my setup, smbfs does not compile as a module. I have left the entry in the Makefile, but I believe this need to be fixed. END K 10 svn:author V 6 dillon K 8 svn:date V 27 2001-11-03T00:59:15.000000Z K 7 svn:log V 122 MFC don't let pmap_object_init_pt() exhaust all available free pages (madvise kernel panic fix). Submitted by: ps, peter END K 10 svn:author V 6 dillon K 8 svn:date V 27 2001-11-03T01:08:55.000000Z K 7 svn:log V 125 Implement i386/i386/pmap.c 1.292 for alpha, ia64 (avoid free page exhaustion / kernel panic for certain madvise() scenarios) END K 10 svn:author V 6 dillon K 8 svn:date V 27 2001-11-03T01:10:25.000000Z K 7 svn:log V 75 MFC i386/i386/pmap.c 1.292 (madvise free page exhaustion fixes) for alpha. END K 10 svn:author V 2 ps K 8 svn:date V 27 2001-11-03T01:21:58.000000Z K 7 svn:log V 147 MFC: kern_environment.c rev 1.19, src/sys/sys/systm.h rev 1.155 - Change getenv_quad() to return an int instead of a quad_t - Add getenv_string() END K 10 svn:author V 3 des K 8 svn:date V 27 2001-11-03T01:31:46.000000Z K 7 svn:log V 58 Remove a debugging warnx() that got committed by mistake. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-03T01:33:12.000000Z K 7 svn:log V 116 Revert rev 1.7 which moved us away from POSIX character classes. The community feels our base AWK must handle them. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-03T01:35:07.000000Z K 7 svn:log V 116 Revert rev 1.3 which moved us away from POSIX character classes. The community feels our base AWK must handle them. END K 10 svn:author V 2 ps K 8 svn:date V 27 2001-11-03T01:41:10.000000Z K 7 svn:log V 308 MFC: - Make MAXTSIZ, DFLDSIZ, MAXDSIZ, DFLSSIZ, MAXSSIZ, SGROWSIZ loader tunable. - Change TUNABLE_*_FETCH to have a return value of 0 if the variable was not found or successfully converted and true otherwise. - Use vm_offset_t instead of caddr_t to fix a warning and remove two casts in kern_exec.c. END K 10 svn:author V 2 ps K 8 svn:date V 27 2001-11-03T01:47:31.000000Z K 7 svn:log V 61 MFC: rev 1.954 Add some comments about KVA_PAGES and a test. END K 10 svn:author V 6 obrien K 8 svn:date V 27 2001-11-03T02:14:10.000000Z K 7 svn:log V 230 Tighten up the random seeding a little bit more. getpid() alone only affect only lower bits of seed which would resulte in the same seed for sequences of fast-started awk's resulting the same random sequence. Submitted by: ache END K 10 svn:author V 4 grog K 8 svn:date V 27 2001-11-03T02:39:15.000000Z K 7 svn:log V 102 allocrqg: Don't ever try to write to a NULL pointer. Reported by: Jeff Roberson END K 10 svn:author V 3 des K 8 svn:date V 27 2001-11-03T03:07:09.000000Z K 7 svn:log V 114 Reduce the number of #include dependencies by declaring some of the structs used in pseudofs.h as opaque structs. END K 10 svn:author V 3 des K 8 svn:date V 27 2001-11-03T03:17:16.000000Z K 7 svn:log V 108 Declare struct uio so consumers don't have to #include unless they intend to call proc_rwmem(). END K 10 svn:author V 3 imp K 8 svn:date V 27 2001-11-03T03:23:39.000000Z K 7 svn:log V 17 Note 4.4-RELEASE END K 10 svn:author V 3 imp K 8 svn:date V 27 2001-11-03T03:39:05.000000Z K 7 svn:log V 18 Note dummynet MFC END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T04:55:48.000000Z K 7 svn:log V 131 Add a 'reset' command. This is useful for panics really early before any symbols are loaded. Especially for unattended machines. END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T05:03:00.000000Z K 7 svn:log V 33 Add ICH3 (82801CA/CAM) and 460GX END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T05:07:59.000000Z K 7 svn:log V 90 Add Id for 82801CA (ICH3?). The ich driver seems to work fine on the laptop that had it. END K 10 svn:author V 4 nyan K 8 svn:date V 27 2001-11-03T05:55:31.000000Z K 7 svn:log V 45 MFi386: sys/i386/isa/npx.c revision 1.80.2.3 END K 10 svn:author V 4 nyan K 8 svn:date V 27 2001-11-03T05:56:35.000000Z K 7 svn:log V 46 Merged from sys/isa/sio.c revision 1.291.2.22 END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T06:24:23.000000Z K 7 svn:log V 112 Dont fail if the intermediate directories in ${DESTDIR}/usr/lib/compat/* dont exist. IMHO, this is kinda silly. END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T06:26:15.000000Z K 7 svn:log V 99 Dont fail if ${DESTDIR}${LIBDIR}/ or any of its components dont exist for the rm -f of libresolv.a END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T06:30:15.000000Z K 7 svn:log V 453 Do not ``build'' (ie remove, and put in stub libraries) for libresolv and/or libgnumalloc on anything but i386. The other platforms post-date this mistake. Do not build libc_r for ia64. There are some fundamental issues that need to be resolved (ie: it cannot use setjmp/longjmp for thread switching, which isn't likely to be fixed soon. libc_r has to be reimplemented using something like makecontext()/swapcontext() etc in order to work in ia64.) END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T06:31:27.000000Z K 7 svn:log V 49 Put in an attempt at stack trace/unwind records. END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T06:33:08.000000Z K 7 svn:log V 184 Do not fail doing rm -f ${DESTDIR}${LIBDIR}/lib${LIB}.a if the intermediate directories dont exist. (eg: when building standalone includes, using the beforeinstall target in src/lib) END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T07:30:14.000000Z K 7 svn:log V 185 Move i386/{fdisk,cxconfig,nextboot} to src/sbin. Eliminate the i386 subdir. Add fdisk to ia64 SUBDIR list. Add all the previous occupants of the i386 subdir to the i386 SUBDIR list. END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T07:31:21.000000Z K 7 svn:log V 131 Repo copied contents of src/sbin/i386 to src/sbin and set specific SUBDIR lists there. Some of these are used on other platforms. END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T08:19:21.000000Z K 7 svn:log V 41 FreeBSD/tahoe is not likely for a while. END K 10 svn:author V 3 phk K 8 svn:date V 27 2001-11-03T08:20:44.000000Z K 7 svn:log V 101 Add modules for lpt, plip, ppbus, ppi and pps. Submitted by: j mckitrick END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T08:24:00.000000Z K 7 svn:log V 37 FreeBSD/vax is quite some time away. END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T08:32:28.000000Z K 7 svn:log V 28 No FreeBSD/vax here either. END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T08:35:11.000000Z K 7 svn:log V 52 Remove support for FreeBSD/tahoe Submitted by: phk END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T08:38:54.000000Z K 7 svn:log V 79 Remove FreeBSD/vax and FreeBSD/tahoe support. Add $FreeBSD$ Submitted by: phk END K 10 svn:author V 3 non K 8 svn:date V 27 2001-11-03T08:55:34.000000Z K 7 svn:log V 78 Remove PAO3 dependent part. This will not affect either -current nor -stable. END K 10 svn:author V 3 phk K 8 svn:date V 27 2001-11-03T09:28:33.000000Z K 7 svn:log V 41 Clean up the bogus pci.h file we create. END K 10 svn:author V 5 brian K 8 svn:date V 27 2001-11-03T11:34:09.000000Z K 7 svn:log V 88 cmott@scientech.com -> cm@linktel.net Requested by: Charles Mott END K 10 svn:author V 5 brian K 8 svn:date V 27 2001-11-03T11:36:17.000000Z K 7 svn:log V 89 cmott@scientech.com -> cm@linktel.net Requested by: Charless Mott END K 10 svn:author V 5 brian K 8 svn:date V 27 2001-11-03T11:37:13.000000Z K 7 svn:log V 43 MFC: cmott@scientech.com -> cm@linktel.net END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T12:36:16.000000Z K 7 svn:log V 287 Partial reversion of rev 1.138. kill and killpg allow a signal argument of 0. You cannot return EINVAL for signal 0. This broke (in 5 minutes of testing) at least ssh-agent and screen. However, there was a bug in the original code. Signal 128 is not valid. Pointy-hat to: des, jhb END K 10 svn:author V 8 tanimura K 8 svn:date V 27 2001-11-03T12:47:19.000000Z K 7 svn:log V 67 Avoid crash of minor numbers between /dev/midistat and /dev/dspr*. END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T12:58:10.000000Z K 7 svn:log V 25 Signal 128 is not valid. END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T13:17:45.000000Z K 7 svn:log V 143 Argh!. Signal 128 *is* valid. Revert last commit. See _SIG_IDX() in signal.h where the signal number is offset by 1. Damn Jedi mind trick! END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T13:26:15.000000Z K 7 svn:log V 233 _SIG_MAXSIG (128) is the highest legal signal. The arrays are offset by one - see _SIG_IDX(). Revert part of my mis-correction in kern_sig.c (but signal 0 still has to be allowed) and fix _SIG_VALID() (it was rejecting ignal 128). END K 10 svn:author V 5 peter K 8 svn:date V 27 2001-11-03T13:35:07.000000Z K 7 svn:log V 22 Desupport FreeBSD/vax END K 10 svn:author V 3 dfr K 8 svn:date V 27 2001-11-03T15:51:14.000000Z K 7 svn:log V 29 Implement END K 10 svn:author V 3 dfr K 8 svn:date V 27 2001-11-03T15:52:18.000000Z K 7 svn:log V 39 Implement fpsetmask() and fpgetmask(). END K 10 svn:author V 3 dfr K 8 svn:date V 27 2001-11-03T16:06:05.000000Z K 7 svn:log V 33 Clone the alpha config for ia64. END K 10 svn:author V 3 phk K 8 svn:date V 27 2001-11-03T16:38:06.000000Z K 7 svn:log V 53 Correct attribution and fix translation from Danish. END K 10 svn:author V 3 phk K 8 svn:date V 27 2001-11-03T16:43:13.000000Z K 7 svn:log V 133 Separate concatenated fortunes. Remove unattributed duplicate of Frederick Brooks eternal words about the danger of second systems. END K 10 svn:author V 3 phk K 8 svn:date V 27 2001-11-03T16:45:31.000000Z K 7 svn:log V 86 Add PCIC_TI4410 (patch probably from imp@, can't remember figuring this out myself.) END K 10 svn:author V 3 phk K 8 svn:date V 27 2001-11-03T16:53:24.000000Z K 7 svn:log V 119 Use vfs_timestamp() instead of getnanotime() directly. Fix some modes on directories and symlinks. Instructed by: bde END K 10 svn:author V 3 phk K 8 svn:date V 27 2001-11-03T17:00:02.000000Z K 7 svn:log V 110 Use vfs_timestamp() instead of getnanotime(). Add magic stuff copied from ufs_setattr(). Instructed by: bde END K 10 svn:author V 3 phk K 8 svn:date V 27 2001-11-03T17:05:12.000000Z K 7 svn:log V 58 Remove unneeded and #if 0'ed code to register in cdevsw[] END K 10 svn:author V 8 gallatin K 8 svn:date V 27 2001-11-03T17:22:50.000000Z K 7 svn:log V 301 Introduce a boot environment variable (clock_compat_osf1) which can be set to 1 to make FreeBSD and Tru64 coexist peacefully on a dual boot system and not clobber each other's year in the TOY clock. (Tru64 uses an offset 52 years higher than one would expect) Obtained from: NetBSD MFC After: 1 week END K 10 svn:author V 3 imp K 8 svn:date V 27 2001-11-03T17:42:56.000000Z K 7 svn:log V 30 Some words about awk and pain END K 10 svn:author V 8 nobutaka K 8 svn:date V 27 2001-11-03T19:09:41.000000Z K 7 svn:log V 29 Add myself. Reviewd by: knu END K 10 svn:author V 5 brian K 8 svn:date V 27 2001-11-03T19:29:30.000000Z K 7 svn:log V 30 Call CCP protocol -1 ``none'' END K 10 svn:author V 6 dillon K 8 svn:date V 27 2001-11-03T19:48:50.000000Z K 7 svn:log V 141 MFC 1.114 - cluster_rbuild() 'size' must match b_bcount and b_bufsize. Added printf() warnings for now, which will eventually become panics. END K 10 svn:author V 6 dillon K 8 svn:date V 27 2001-11-03T19:59:28.000000Z K 7 svn:log V 712 MFC implement kern.maxvnodes (primarily kern/vfs_subr.c 1.326), sync, and other VM optimizations and fixes. * Add VOBJDIRTY flag to vnode to avoid certain indirections. * Turn the mount structure vnode list into an LRU queue (LIST -> TAILQ) * Attempt to reuse vnodes when we hit the maxvnodes limit * Optimize msync() to avoid unnecessary v_interlocks * Optimize ffs_sync() for same * Fix vm_object double-termination bug * Add KASSERT to ensure that backing object is not freed out from under remaining VM pages. These fixes change the mount structure and will create a binary kld module incompatibility with filesystem kld's, but I deem the scaling issues with 4.x that we are fixing to be more important. END K 10 svn:author V 3 ume K 8 svn:date V 27 2001-11-03T20:47:50.000000Z K 7 svn:log V 102 Unbreak build without INET6. Reported by: Christopher N . Harrell MFC after: 1 week END K 10 svn:author V 3 jhb K 8 svn:date V 27 2001-11-03T21:40:47.000000Z K 7 svn:log V 117 Output a newline at the end of a dump so that there are blank lines between dumps when using breakpoints or tracing. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2001-11-03T21:42:25.000000Z K 7 svn:log V 187 Add support for trace traps by returning from them just as for breakpoint traps rather than halting. Ideally, we should avoid printing the 'BTX halted' message for debug register dumps. END K 10 svn:author V 5 brian K 8 svn:date V 27 2001-11-03T21:45:32.000000Z K 7 svn:log V 410 Add a ``log'' command for logging specific information. Add an ``UPTIME'' variable to indicate the bundle uptime. It's now possible to put something like this in ppp.linkdown for a server setup: MYADDR: log Session closing: User USER, address HISADDR, up UPTIME Fixed some memory leakage with commands that expand words. Made some functions static. Fixed a diagnostic bug (iface add .... SIOCDIFADDR) END K 10 svn:author V 5 brian K 8 svn:date V 27 2001-11-03T21:47:06.000000Z K 7 svn:log V 278 MFC: Add a ``log'' command for logging specific information. Add an ``UPTIME'' variable to indicate the bundle uptime. Fixed some memory leakage with commands that expand words. Made some functions static. Fixed a diagnostic bug (iface add .... SIOCDIFADDR) END K 10 svn:author V 3 jhb K 8 svn:date V 27 2001-11-03T21:53:46.000000Z K 7 svn:log V 174 Add support for outputting multiple lines when dumping memory during the register dump. Change the default to bump 2 lines of output (32 bytes) instead of 1 line (16 byte). END K 10 svn:author V 3 jhb K 8 svn:date V 27 2001-11-03T22:09:30.000000Z K 7 svn:log V 255 Add support for sending messages to the serial console which is helpful when debugging boot problems. It is not on by default but is enabled via the BTX_SERIAL variable. The port and speed can be set via the same variables used by boot2 and the loader. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2001-11-03T22:21:21.000000Z K 7 svn:log V 50 Whoops, missed these bits in the previous commit. END K 10 svn:author V 3 phk K 8 svn:date V 27 2001-11-03T23:21:00.000000Z K 7 svn:log V 84 Don't choke on old sd%d.ctl devices. Tripped over by: Jos Backus END K 10 svn:author V 3 jhb K 8 svn:date V 27 2001-11-04T02:58:39.000000Z K 7 svn:log V 265 Print out 'foo devices:' as the line before displaying a group of devices in 'lsdev' output rather than printing out a pointer to the print function since the user really could care less about the pointer value. Perhaps this was intended to be a debugging printf? END K 10 svn:author V 3 jhb K 8 svn:date V 27 2001-11-04T03:24:16.000000Z K 7 svn:log V 1412 This is a new CD bootstrap utility designed to replace cdldr. According to the El Torito standard for CD booting, a CD may boot in "No emulation" mode without using a floppy image. In this mode, the BIOS loads a program off of the CD into memory and creates a BIOS device using 2048 byte sectors for the CD. According to the standard, this program can be up to 0xFFFF virtual (512-byte) sectors long. The old cdldr depended on this by having the BIOS load the entire loader and the small cdldr stub as one binary similar to pxeboot so that cdldr didn't have to read the CD to find the loader. However, the NT no emulation loader just uses 1 disk sector (4 virtual sectors), so it seems that at least some BIOS writers just did enough to get NT to boot by only loading 1 sector and ignoring the sector count. Thus, while cdldr should have worked in theory, it doesn't in practice. This replacment fits entirely in 1 sector and includes simple ISO 9660 support. It looks for /boot/loader on the CD and loads it up using the BIOS. This allows us to not have to depend on the limited size of floppy images but use a full GENERIC kernel for CD-ROM installs in the future, among other things. This version of cdboot is a bit bloated as it includes some useful debugging routines that people can pull to use in other x86 assembly modules. Even with all the debugging cruft, we still have 272 bytes to spare. END K 10 svn:author V 3 jhb K 8 svn:date V 27 2001-11-04T03:27:06.000000Z K 7 svn:log V 37 Use the new cdboot instead of cdldr. END