sort -k bug

hi sort -k has a bug when the start character of the field is specified but the end character is not eg. sort -k 1.2,1 sort returns 'field out of order' but the end character defaults to 0, which signifies the last character of the field and so is definitely not out of order. a check for start c

Re: km_alloc/km_free optimization

> Date: Sun, 28 Dec 2014 20:56:33 +0100 (CET) > From: Mark Kettenis > > P.S. I would not be surprised if this would "fix" the problems landisk > has with the "* 8" pool diff. Currently updating my landisk to > verify. And it seems this does fix the landisk problem. I strongly suspect

km_alloc/km_free optimization

One side-effect of dlg@'s "* 8" pool diff is that some pools now use "large" pool pages that aren't directly mapped on PMAP_DIRECT architecture. In general it's not possible to map "large" pages directly since the individual memory pages are not guaranteed to be contigious. However, the mbuf clus

Re: ps realloc loop

On Sun, Dec 28, 2014 at 6:02 AM, Ted Unangst wrote: ... > Agreed. Here's a different approach. If we need more memory, allocate > the required amount. No realloc since we aren't interested in the > previous contents. Continue caching the previous allocation, but don't > worry about power of two si

Re: syslogd sending via tcp

On Sun, Dec 28, 2014 at 05:33:08PM +0100, Alexander Bluhm wrote: > Jasper tested and found that it only worked on loopback. I have > forgotten to check for EINPROGRESS after connect. So here is a new > diff. > > bluhm Succesfully tested now with a remote logstash host. > Index: privsep.c > ===

Re: syslogd sending via tcp

minimal man page adaption: Index: syslog.conf.5 === RCS file: /cvs/src/usr.sbin/syslogd/syslog.conf.5,v retrieving revision 1.26 diff -u -r1.26 syslog.conf.5 --- syslog.conf.5 25 Aug 2014 20:25:46 - 1.26 +++ syslog.conf

Re: syslogd sending via tcp

Jasper tested and found that it only worked on loopback. I have forgotten to check for EINPROGRESS after connect. So here is a new diff. bluhm Index: privsep.c === RCS file: /data/mirror/openbsd/cvs/src/usr.sbin/syslogd/privsep.c,v

Re: syslogd sending via tcp

On Wed, Oct 29, 2014 at 07:06:53PM +0100, Alexander Bluhm wrote: > I would like to get this in and develop the missing parts in tree. So here is the diff again, I have merged to current and changed a comment. ok? bluhm Index: privsep.c ===

Re: ps realloc loop

On Fri, Dec 26, 2014 at 11:21, Mark Kettenis wrote: >> Date: Fri, 26 Dec 2014 00:18:04 -0500 >> From: Ted Unangst >> >> Theo noticed that the realloc loop in ps is suboptimal. >> >> Mostly style change: overflow checking for "free" courtesy of >> reallocarray. Vageuly wasteful, but ps strings are

Re: mg: remove some more warnings

Hi Brian, this looks right: It is easy to see that curbp->b_fname can never be NULL, and entering dired mode still works for me both from *scratch* and from a file buffer. Yours, Ingo > Index: dired.c > === > RCS file: /cvs/src/u