Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Merge from Lite2 (whiteout/unionfs) | peter | 1997-03-11 | 1 | -2/+4 |
| | |||||
* | Merge from Lite2 | peter | 1997-03-11 | 1 | -2/+2 |
| | |||||
* | Merge Lite2 changes | peter | 1997-03-11 | 5 | -50/+98 |
| | |||||
* | merge from Lite2 - realpath() now shares a lot of code with getcwd() | peter | 1997-03-11 | 3 | -287/+158 |
| | | | | and is now in the same file. | ||||
* | Merge from Lite2 onto mainline - | peter | 1997-03-11 | 19 | -278/+186 |
| | | | | | - add undelete() and undelete.2 (requires libc minor bump some time) - man page updates | ||||
* | This commit was generated by cvs2svn to compensate for changes in r23658, | peter | 1997-03-11 | 9 | -37/+373 |
|\ | | | | | | | which included commits to RCS files with non-trunk default branches. | ||||
| * | Import CSRG 4.4BSD-Lite2 lib/libc onto vendor branch | peter | 1997-03-11 | 75 | -662/+1333 |
| | | |||||
* | | Import CSRG 4.4BSD-Lite2 includes onto vendor branch | peter | 1997-03-11 | 1 | -4/+7 |
| | | |||||
* | | Check for overflow of FD_SET | guido | 1997-03-10 | 1 | -1/+6 |
| | | |||||
* | | Fix brain-o in SunOS passwd.adjunct stuff: !strstr(s, "##") is a) bad | wpaul | 1997-03-10 | 1 | -1/+1 |
| | | | | | | | | | | | | style and b) the wrong logic. Should be strstr(s, "##") != NULL. (Note that the passwd.adjunct stuff has not been merged into 2.2 so this bug is not in that branch.) | ||||
* | | Use __ieee754_sqrt() instead of sqrt() internally. Similarly for the | bde | 1997-03-09 | 12 | -33/+33 |
| | | | | | | | | | | | | float versions. Using sqrt() was inefficient. Obtained from: NetBSD | ||||
* | | Include <machine/asm.h> instead of kernel-only <machine/asmacros.h>. | bde | 1997-03-09 | 24 | -48/+48 |
| | | |||||
* | | Clarified the behaviour of dup2(fd1,fd2) when fd1==fd2 and when fd1 is invalid. | mckay | 1997-03-09 | 1 | -4/+21 |
| | | | | | | | | Safe for 2.2! | ||||
* | | The first argument to fts_set was wrong. Part of PR# 2917. | mpp | 1997-03-09 | 1 | -1/+2 |
| | | |||||
* | | Pause() is made obsolete by sigsuspend(2), not sigpause(3). | mpp | 1997-03-09 | 1 | -4/+4 |
| | | | | | | | | Part of PR# 2917. | ||||
* | | Install opie.h in /usr/include | pst | 1997-03-08 | 1 | -1/+5 |
| | | |||||
* | | Allow comments in group database. | wosch | 1997-03-08 | 1 | -2/+2 |
| | | | | | | | | | | | | | | | | The character `#' introduces a comment. Leading spaces and tabs are ignored: '^[ \t]*#.*\n$' Count an empty line - only spaces, tabs or newline - also as a comment. (to be compatibel with password database comments). '^[ \t]*\n$' | ||||
* | | API cleanups (use const char * where appropriate) | pst | 1997-03-08 | 4 | -15/+15 |
| | | | | | | | | Cannidate for: 2.2 (please!) | ||||
* | | Add a missing semi-colon. | mpp | 1997-03-07 | 1 | -1/+1 |
| | | | | | | | | Submitted by: jmg | ||||
* | | Fixed wrong magic numbers in scaling. hypotf() was very broken for large | bde | 1997-03-05 | 1 | -8/+8 |
| | | | | | | | | | | | | | | | | | | | | and small values: hypotf(2.3819765e+38, 2.0416943e+38) was NaN instead of 3.1372484e+38 hypotf(-3.4028235e+38, 3.3886450e+38) was NaN instead of Inf hypotf(-2.8025969e-45, -2.8025969e-45) was 0 instead of 4.2038954e-45 Found by: ucbtest | ||||
* | | Remove words about lower limit needed (from BUGS section), we have | ache | 1997-03-03 | 1 | -7/+0 |
| | | | | | | | | precise limit now | ||||
* | | Fixed handling of input failure by the scanf family. | bde | 1997-03-03 | 1 | -3/+11 |
| | | | | | | | | | | | | | | | | | | | | | | - 0 was returned instead of EOF when an input failure occured while skipping white-space after 0 assignments. This fixes PR2606. The diagnosis in PR2606 is wrong. - EOF was returned instead of 0 when an input failure occurred after zero assignments and nonzero suppressed assignments. - EOF was spelled -1. This should be in 2.2. | ||||
* | | Use .Sq Li \&\e0 rather than NUL to describe the character with no bits | imp | 1997-03-03 | 1 | -3/+7 |
| | | | | | | | | | | set, as suggested by Garrett Wollman. This is more consistant with how things like strncpy are done, as well as harder to confuse NUL and NULL. | ||||
* | | Enabled the Lite2 getvfsbyname(). It's actually named new_getvfsbyname() | bde | 1997-03-03 | 2 | -4/+8 |
| | | | | | | | | | | | | | | for now so that we don't lose library compatibility. Applications should define _NEW_VFSCONF and use getvfsbyname() instead of new_getvfsbyname() if they want the new vfsconf interface. Parts of the old interface (enough to load vfs modules, I hope) are still available. | ||||
* | | Use stricter MAXLOGNAME now | ache | 1997-03-03 | 1 | -1/+1 |
| | | |||||
* | | Reflect current MAXLONGNAME value and specify that null included | ache | 1997-03-03 | 1 | -1/+1 |
| | | |||||
* | | Don't add/subtract 1 to MAXLOGNAME, it is already NUL-terminated | ache | 1997-03-03 | 1 | -2/+2 |
| | | |||||
* | | Change vfc_typenum back to vfc_index in response to bde's | mpp | 1997-03-03 | 1 | -2/+2 |
| | | | | | | | | commit to getvfsent.c rev 1.10. | ||||
* | | Changed vfc_typenum back to vfc_index. The old vfsconf struct is now | bde | 1997-03-03 | 1 | -1/+1 |
| | | | | | | | | visible again, and the new vfsconf struct didn't match reality. | ||||
* | | MAXLOGNAME currently 16, not 12 | ache | 1997-03-02 | 1 | -1/+1 |
| | | |||||
* | | Pendantic change of null to NUL. Also warn that this function does not | imp | 1997-03-02 | 1 | -3/+9 |
| | | | | | | | | NUL terminate in the case of buffer overflow. | ||||
* | | This commit was generated by cvs2svn to compensate for changes in r23291, | bde | 1997-03-02 | 1 | -0/+77 |
|\ \ | |/ | | | | | which included commits to RCS files with non-trunk default branches. | ||||
| * | Attempt to import Lite2's getvfsbyname.c. | bde | 1997-03-02 | 1 | -0/+77 |
| | | |||||
| * | pull this in to avoid fixing these twice | peter | 1996-08-30 | 1 | -5/+6 |
| | | |||||
* | | getnetbyaddr now takes an unsigned long as its first argument. | mpp | 1997-02-28 | 1 | -1/+1 |
| | | |||||
* | | Oops! I accidently commited a change that wasn't ready for prime | mpp | 1997-02-28 | 1 | -15/+0 |
| | | | | | | | | time yet. Revert to rev 1.2. | ||||
* | | The struct vfsconf element vfc_index is now vfs_typenum. | mpp | 1997-02-28 | 2 | -2/+17 |
| | | |||||
* | | Correct the xref section - it was incorrectly using .Fn instead | mpp | 1997-02-28 | 1 | -4/+4 |
| | | | | | | | | of .Xr for the xrefs. | ||||
* | | "infinity" check was missed from login_getcapsize(), add it | ache | 1997-02-27 | 1 | -1/+7 |
| | | | | | | | | One manifestation of this bug: all networking users have coredumpsize=0 | ||||
* | | Buffer overflow from DNS name information which could cause root access | imp | 1997-02-26 | 1 | -1/+2 |
| | | | | | | | | | | | | | | when called from lpd. Reviewed by: jkh, pst Submitted by: Oliver Friedrichs <oliver@secnet.com> | ||||
* | | Use the C library version of log10() instead of the inaccurate formula | bde | 1997-02-24 | 1 | -4/+2 |
| | | | | | | | | | | | | | | log10(x) = log10e * log(x). This fixes some small (one or two ULP) inaccuracies. Found by: ucbtest | ||||
* | | Oops, I thought I was fixing the double precision lg10 in the | bde | 1997-02-24 | 1 | -2/+2 |
| | | | | | | | | | | | | | | | | | | previous revision. Use log10f() instead of log10(). log10f() is currently slightly slower than log10() on P5's, but it is potentially significantly faster. Fixed declaration of the C function used in the (unused?) KR_headers case. | ||||
* | | Revert $FreeBSD$ to $Id$ | peter | 1997-02-22 | 638 | -719/+719 |
| | | |||||
* | | Revert $FreeBSD$ to $Id$ | peter | 1997-02-22 | 4 | -4/+4 |
| | | |||||
* | | Back out an easy $FreeBSD$ | peter | 1997-02-22 | 1 | -1/+1 |
| | | |||||
* | | Split up the Bessel function wrapper files so that most wrapper functions | bde | 1997-02-20 | 13 | -177/+315 |
| | | | | | | | | are in their own file. | ||||
* | | Removed misplaced duplicate of comment about implementation details. | bde | 1997-02-20 | 1 | -22/+0 |
| | | |||||
* | | Compute (1 - x^2) as ((1 - x) * (1 + x)) instead of as (1 - x * x) to | bde | 1997-02-20 | 2 | -11/+23 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | avoid easily avoidable loss of precision when |x| is nearly 1. Extended (64-bit) precision only moves the meaning of "nearly" here. This probably could be done better by splitting up the range into |x| <= 0.5 and |x| > 0.5 like the C version. However, ucbtest does't report any errors in this version. Perhaps the C version should be used anyway. It's only 25% slower now on a P5, provided the C version of sqrt() isn't used, and the C version could be optimized better. Errors checked by: ucbtest | ||||
* | | Typogrammatical error 'with' -> 'when'. | danny | 1997-02-20 | 1 | -1/+1 |
| | | | | | | | | This could be put into 2.2. | ||||
* | | Add forgotten man page link fts_set.3 -> fts.3 | wosch | 1997-02-16 | 1 | -1/+1 |
| | |