Re: Adding current bandwidth to ftp(1)
Stuart Henderson writes: > On 2018/04/09 09:27, Matthias Schmidt wrote: >> Hi Christian, >> >> On 08.04.2018 22:42, Christian Barthel wrote: >> > Hello, >> > >> > i am often sitting behind a very slow internet link and thought, it >> > might be useful to show the current bandwidth while downloading >> > something with ftp(1). >> >> Since I sometimes work with crappy WLANs/connections I very much appreciate >> your patch! > > Do you know you can just send SIGINFO (often bound to ^T) to get the same > information? Hi Stuart, Hi Matthias, thanks for your replies. I've tested it with 80x25 terminals and also checked the bsd.rd installer image. To me, it looks ok. I also know about SIGINFO but in my particular case, it was useful to print the bandwidth all the time. But I guess that most users don't need that information and having SIGINFO/^T is sufficient. Thanks for your comments, Christian
net/if.h missing includes
isn't standalone if __BSD_VISIBLE is defined (the default). This has caused fallout earlier in the ports tree: https://cvsweb.openbsd.org/cgi-bin/cvsweb/ports/net/samba/patches/patch-lib_replace_wscript.diff?r1=1.7&r2=1.8 https://cvsweb.openbsd.org/cgi-bin/cvsweb/~checkout~/ports/net/dnsdist/patches/patch-dnsdist-lua_cc?rev=1.1&content-type=text/plain Went in an mkr on amd64. ok? Index: sys/net/if.h === RCS file: /cvs/src/sys/net/if.h,v retrieving revision 1.192 diff -u -p -r1.192 if.h --- sys/net/if.h19 Feb 2018 04:43:48 - 1.192 +++ sys/net/if.h2 Apr 2018 18:08:03 - @@ -59,6 +59,10 @@ __END_DECLS #if __BSD_VISIBLE +#include +#include +#include + /* * Structure used to query names of interface cloners. */ -- jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE
Re: Eliminate trailing whitespace & typo in chmod
https://en.wikipedia.org/wiki/If_and_only_if As a non-native speaker, it took some years before I realized the use of “iff” is not a typo. Kind regards, Job
Re: Eliminate trailing whitespace & typo in chmod
On 2018/04/13 20:01, llgx...@gmail.com wrote: > - * on the level: follow it iff it's a command line arg. > + * on the level: follow it if it's a command line arg. iff = "if and only if" http://mathworld.wolfram.com/Iff.html https://en.wikipedia.org/wiki/If_and_only_if
Eliminate trailing whitespace & typo in chmod
> > Index: chmod.c > > === > > RCS file: /cvs/src/bin/chmod/chmod.c,v > > retrieving revision 1.42 > > diff -u -p -u -p -r1.42 chmod.c > > --- chmod.c 28 May 2017 08:03:36 - 1.42 > > +++ chmod.c 13 Apr 2018 17:46:36 - > > @@ -253,10 +253,10 @@ done: > > > > /* > > * For -RH, the decision of how to handle symlinks depends > > -* on the level: follow it iff it's a command line arg. > > +* on the level: follow it if it's a command line arg. > > */ > > "iff" is not a typo. Oops, sorry, my bad. I hope the following one is ok: Index: chmod.c === RCS file: /cvs/src/bin/chmod/chmod.c,v retrieving revision 1.42 diff -u -p -u -p -r1.42 chmod.c --- chmod.c 28 May 2017 08:03:36 - 1.42 +++ chmod.c 13 Apr 2018 19:07:58 - @@ -256,7 +256,7 @@ done: * on the level: follow it iff it's a command line arg. */ if (fts_options & FTS_COMFOLLOW) { - atflags = p->fts_level == FTS_ROOTLEVEL ? 0 : + atflags = p->fts_level == FTS_ROOTLEVEL ? 0 : AT_SYMLINK_NOFOLLOW; }
Re: vmd(4) close vmm parent socket
On Fri, Apr 13, 2018 at 11:29:06AM +0200, Martijn van Duren wrote: > Hello tech@, > > Playing with vmd I noticed that a vm process has vmm's socket to the > parent process still open. > > Patch below works for me. > > OK? > > martijn@ > > Index: vmm.c > === > RCS file: /cvs/src/usr.sbin/vmd/vmm.c,v > retrieving revision 1.80 > diff -u -p -r1.80 vmm.c > --- vmm.c 5 Feb 2018 05:01:08 - 1.80 > +++ vmm.c 13 Apr 2018 09:27:14 - > @@ -664,6 +664,7 @@ vmm_start_vm(struct imsg *imsg, uint32_t > } else { > /* Child */ > close(fds[0]); > + close(PROC_PARENT_SOCK_FILENO); > > ret = start_vm(vm, fds[1]); > > sure
Re: Eliminate trailing whitespace & typo in chmod
>Index: chmod.c >=== >RCS file: /cvs/src/bin/chmod/chmod.c,v >retrieving revision 1.42 >diff -u -p -u -p -r1.42 chmod.c >--- chmod.c28 May 2017 08:03:36 - 1.42 >+++ chmod.c13 Apr 2018 17:46:36 - >@@ -253,10 +253,10 @@ done: > > /* >* For -RH, the decision of how to handle symlinks depends >- * on the level: follow it iff it's a command line arg. >+ * on the level: follow it if it's a command line arg. >*/ "iff" is not a typo.
Eliminate trailing whitespace & typo in chmod
Index: chmod.c === RCS file: /cvs/src/bin/chmod/chmod.c,v retrieving revision 1.42 diff -u -p -u -p -r1.42 chmod.c --- chmod.c 28 May 2017 08:03:36 - 1.42 +++ chmod.c 13 Apr 2018 17:46:36 - @@ -253,10 +253,10 @@ done: /* * For -RH, the decision of how to handle symlinks depends -* on the level: follow it iff it's a command line arg. +* on the level: follow it if it's a command line arg. */ if (fts_options & FTS_COMFOLLOW) { - atflags = p->fts_level == FTS_ROOTLEVEL ? 0 : + atflags = p->fts_level == FTS_ROOTLEVEL ? 0 : AT_SYMLINK_NOFOLLOW; }
Re: ksh: count $SECONDS with monotonic clock
On Fri, 13 Apr 2018 19:42:51 +0200, Jeremie Courreges-Anglas wrote: > The diff uses timespecsub: Ah, OK. That's fine then. - todd
Re: ksh: count $SECONDS with monotonic clock
On Fri, Apr 13 2018, "Todd C. Miller" wrote: > On Fri, 13 Apr 2018 11:11:04 -0500, Scott Cheloha wrote: > >> So that $SECONDS advances uniformly, independent of the system clock. > > Why are you including sys/time.h? For struct timespect you only > need time.h which is already included. > > In general, you only need sys/time.h for struct timeval or for some > of the time-related macros. The diff uses timespecsub: @@ -921,8 +922,13 @@ getspec(struct tbl *vp) * has been set - don't do anything in this case * (see initcoms[] in main.c). */ - if (vp->flag & ISSET) - setint(vp, (int64_t)(time(NULL) - seconds)); + if (vp->flag & ISSET) { + struct timespec difference, now; + + clock_gettime(CLOCK_MONOTONIC, &now); + timespecsub(&now, &seconds, &difference); + setint(vp, (int64_t)difference.tv_sec); + } vp->flag |= SPECIAL; break; case V_RANDOM: -- jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE
Re: ksh: count $SECONDS with monotonic clock
On Fri, 13 Apr 2018 11:11:04 -0500, Scott Cheloha wrote: > So that $SECONDS advances uniformly, independent of the system clock. Why are you including sys/time.h? For struct timespect you only need time.h which is already included. In general, you only need sys/time.h for struct timeval or for some of the time-related macros. - todd
Re: ksh: count $SECONDS with monotonic clock
On Fri, Apr 13 2018, Scott Cheloha wrote: > So that $SECONDS advances uniformly, independent of the system clock. > > ok? ok jca@ > -- > Scott Cheloha > > P.S. Similar change forthcoming for $MAILCHECK. > > Index: bin/ksh/var.c > === > RCS file: /cvs/src/bin/ksh/var.c,v > retrieving revision 1.67 > diff -u -p -r1.67 var.c > --- bin/ksh/var.c 9 Apr 2018 17:53:36 - 1.67 > +++ bin/ksh/var.c 13 Apr 2018 16:01:08 - > @@ -1,6 +1,7 @@ > /* $OpenBSD: var.c,v 1.67 2018/04/09 17:53:36 tobias Exp $ */ > > #include > +#include > > #include > #include > @@ -908,7 +909,7 @@ unspecial(const char *name) > ktdelete(tp); > } > > -static time_t seconds;/* time SECONDS last set */ > +static struct timespec seconds; /* time SECONDS last set */ > static int user_lineno;/* what user set $LINENO to */ > > static void > @@ -921,8 +922,13 @@ getspec(struct tbl *vp) >* has been set - don't do anything in this case >* (see initcoms[] in main.c). >*/ > - if (vp->flag & ISSET) > - setint(vp, (int64_t)(time(NULL) - seconds)); > + if (vp->flag & ISSET) { > + struct timespec difference, now; > + > + clock_gettime(CLOCK_MONOTONIC, &now); > + timespecsub(&now, &seconds, &difference); > + setint(vp, (int64_t)difference.tv_sec); > + } > vp->flag |= SPECIAL; > break; > case V_RANDOM: > @@ -1036,7 +1042,8 @@ setspec(struct tbl *vp) > break; > case V_SECONDS: > vp->flag &= ~SPECIAL; > - seconds = time(NULL) - intval(vp); > + clock_gettime(CLOCK_MONOTONIC, &seconds); > + seconds.tv_sec -= intval(vp); > vp->flag |= SPECIAL; > break; > case V_TMOUT: > -- jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE
ksh: count $SECONDS with monotonic clock
So that $SECONDS advances uniformly, independent of the system clock. ok? -- Scott Cheloha P.S. Similar change forthcoming for $MAILCHECK. Index: bin/ksh/var.c === RCS file: /cvs/src/bin/ksh/var.c,v retrieving revision 1.67 diff -u -p -r1.67 var.c --- bin/ksh/var.c 9 Apr 2018 17:53:36 - 1.67 +++ bin/ksh/var.c 13 Apr 2018 16:01:08 - @@ -1,6 +1,7 @@ /* $OpenBSD: var.c,v 1.67 2018/04/09 17:53:36 tobias Exp $ */ #include +#include #include #include @@ -908,7 +909,7 @@ unspecial(const char *name) ktdelete(tp); } -static time_t seconds;/* time SECONDS last set */ +static struct timespec seconds; /* time SECONDS last set */ static int user_lineno;/* what user set $LINENO to */ static void @@ -921,8 +922,13 @@ getspec(struct tbl *vp) * has been set - don't do anything in this case * (see initcoms[] in main.c). */ - if (vp->flag & ISSET) - setint(vp, (int64_t)(time(NULL) - seconds)); + if (vp->flag & ISSET) { + struct timespec difference, now; + + clock_gettime(CLOCK_MONOTONIC, &now); + timespecsub(&now, &seconds, &difference); + setint(vp, (int64_t)difference.tv_sec); + } vp->flag |= SPECIAL; break; case V_RANDOM: @@ -1036,7 +1042,8 @@ setspec(struct tbl *vp) break; case V_SECONDS: vp->flag &= ~SPECIAL; - seconds = time(NULL) - intval(vp); + clock_gettime(CLOCK_MONOTONIC, &seconds); + seconds.tv_sec -= intval(vp); vp->flag |= SPECIAL; break; case V_TMOUT:
Re: vmd(4) close vmm parent socket
On Fri, Apr 13, 2018 at 11:29:06AM +0200, Martijn van Duren wrote: > Hello tech@, > > Playing with vmd I noticed that a vm process has vmm's socket to the > parent process still open. > > Patch below works for me. > > OK? > > martijn@ Nice find. Ok ccardenas@ +--+ Carlos > > Index: vmm.c > === > RCS file: /cvs/src/usr.sbin/vmd/vmm.c,v > retrieving revision 1.80 > diff -u -p -r1.80 vmm.c > --- vmm.c 5 Feb 2018 05:01:08 - 1.80 > +++ vmm.c 13 Apr 2018 09:27:14 - > @@ -664,6 +664,7 @@ vmm_start_vm(struct imsg *imsg, uint32_t > } else { > /* Child */ > close(fds[0]); > + close(PROC_PARENT_SOCK_FILENO); > > ret = start_vm(vm, fds[1]); > >
Re: syspatch(8) exit code
On Fri, Apr 13, 2018 at 07:37:24AM -0600, Theo de Raadt wrote: > No. There are no patches yet, and it can't handle an empty directory. > > I am waiting for aja and robert to decide what to do about it. Otherwise > we'll have thus fuss at the start of every new release. > > Come on guys, make a decision! For 6.3 it will be "fixed" once there's a syspatch. For 6.4 I will fix it at p2k18. -- Antoine
Re: diff for usr.bin/mg/fileio.c
On Thu, Apr 12, 2018 at 07:04:23PM +0200, Han Boetes wrote: > Yes this also works for me. > Awesome, any OKs? Diff again for convenience: diff --git fileio.c fileio.c index 0987f6f30de..339088f5e2d 100644 --- fileio.c +++ fileio.c @@ -703,7 +703,7 @@ expandtilde(const char *fn) struct stat statbuf; const char *cp; char user[LOGIN_NAME_MAX], path[NFILEN]; - char*un, *ret; + char*ret; size_t ulen, plen; path[0] = '\0'; @@ -722,16 +722,13 @@ expandtilde(const char *fn) return (NULL); return(ret); } - if (ulen == 0) { /* ~/ or ~ */ - if ((un = getlogin()) != NULL) - (void)strlcpy(user, un, sizeof(user)); - else - user[0] = '\0'; - } else { /* ~user/ or ~user */ + if (ulen == 0) /* ~/ or ~ */ + pw = getpwuid(geteuid()); + else { /* ~user/ or ~user */ memcpy(user, &fn[1], ulen); user[ulen] = '\0'; + pw = getpwnam(user); } - pw = getpwnam(user); if (pw != NULL) { plen = strlcpy(path, pw->pw_dir, sizeof(path)); if (plen == 0 || path[plen - 1] != '/') { -- I'm not entirely sure you are real.
Re: syspatch(8) exit code
No. There are no patches yet, and it can't handle an empty directory. I am waiting for aja and robert to decide what to do about it. Otherwise we'll have thus fuss at the start of every new release. Come on guys, make a decision! >It looks like in OpenBSD 6.3-release, syspatch(8) changed exit code and now >when running alone or with -c flag, it exit with status code 1. > >On 6.2 exit code was 0 > >63# syspatch;echo $? >1 > >Is it correct ? > >_ >Zbyszek ŻóÅkiewski > >
syspatch(8) exit code
Hi! It looks like in OpenBSD 6.3-release, syspatch(8) changed exit code and now when running alone or with -c flag, it exit with status code 1. On 6.2 exit code was 0 63# syspatch;echo $? 1 Is it correct ? _ Zbyszek Żółkiewski
Re: ifconfig,route,netstat: s/tableid/rtable/ for consistency
>On 2018/04/13 14:38, Klemens Nanni wrote: >> I propose to use `rtable' exclusively to ease searching and improve >> consistency as that's the wording already used across the majority of >> manuals including rtabe(4) and pf.conf(5) for example. > >There's not a lot in it, but tableid is slightly more common (and to >my eye, looks a bit nicer as a description of an argument..) - changing >from one to the other mostly feels like deckchair-rearranging though. > >There is a related problem that I noticed from looking at this though, >sshd_config uses "rdomain" where it's actually talking about an rtable. I think it should be rtable or rdomain in every specific place. You are suggesting a rtable field name is tableid. Is a rdomain field now name rdomainid? Doesn't make sense to me. The concept should be the field name, and use one name rather than two, giving people the least amount of variation.
Re: ifconfig,route,netstat: s/tableid/rtable/ for consistency
On 2018/04/13 14:38, Klemens Nanni wrote: > I propose to use `rtable' exclusively to ease searching and improve > consistency as that's the wording already used across the majority of > manuals including rtabe(4) and pf.conf(5) for example. There's not a lot in it, but tableid is slightly more common (and to my eye, looks a bit nicer as a description of an argument..) - changing from one to the other mostly feels like deckchair-rearranging though. There is a related problem that I noticed from looking at this though, sshd_config uses "rdomain" where it's actually talking about an rtable.
Re: ifconfig,route,netstat: s/tableid/rtable/ for consistency
On Thu, Apr 12, 2018 at 10:07:58AM +0200, Peter Hessler wrote: > On 2018 Apr 11 (Wed) at 23:01:45 +0200 (+0200), Klemens Nanni wrote: > :On Wed, Apr 11, 2018 at 09:28:03AM +0200, Peter Hessler wrote: > :> No, all of these uses are correct as-is. > :`tableid' surely isn't wrong, but using the argument name across manuals > :seems nicer to me. > : > > No, they are different things. Different names help with the concept. > > > :Or is there any real difference between `tableid' and `rtable' I'm not > :aware of? > : > > rtables are layer 3. > > rdomains are layer 2 (aka, arp and ndp lookups). > > You can have multiple rtables within an rdomain. An interface can only > be a member of a single rdomain at a time. Maybe my first mail wasn't clear enough: I'm talking about routing *tables* only. Specifically, how they are referred to as `rtable' and `tableid' across different manual pages. I propose to use `rtable' exclusively to ease searching and improve consistency as that's the wording already used across the majority of manuals including rtabe(4) and pf.conf(5) for example.
vmd(4) close vmm parent socket
Hello tech@, Playing with vmd I noticed that a vm process has vmm's socket to the parent process still open. Patch below works for me. OK? martijn@ Index: vmm.c === RCS file: /cvs/src/usr.sbin/vmd/vmm.c,v retrieving revision 1.80 diff -u -p -r1.80 vmm.c --- vmm.c 5 Feb 2018 05:01:08 - 1.80 +++ vmm.c 13 Apr 2018 09:27:14 - @@ -664,6 +664,7 @@ vmm_start_vm(struct imsg *imsg, uint32_t } else { /* Child */ close(fds[0]); + close(PROC_PARENT_SOCK_FILENO); ret = start_vm(vm, fds[1]);