Re: svn commit: r285404 - head/sys/compat/cloudabi

2015-07-14 Thread Ed Schouten
Hi Kostik, > Exactly. I think that for programs that are built around pipelines, > where a series of processes have stdin/stdout attached, having > something like SIGPIPE makes sense. So far the focus It looks like I accidentally truncated the last sentence. Sorry about that. "So far the focus w

Re: svn commit: r285404 - head/sys/compat/cloudabi

2015-07-14 Thread Konstantin Belousov
On Tue, Jul 14, 2015 at 12:22:51PM +0200, Ed Schouten wrote: > Ah, okay. Now I think I understand what you're hinting at. So your > proposal is to let cloudabi_sys_proc_raise() only call into > sys_kill(), nothing else. We then reset all signals to their default > behavior at some point during proc

Re: svn commit: r285404 - head/sys/compat/cloudabi

2015-07-14 Thread Ed Schouten
Hi Kostik, 2015-07-11 22:56 GMT+02:00 Konstantin Belousov : >> Bucket 2: The system call could also just fail and return an error >> (MSG_NOSIGPIPE). > SIGPIPE exists to ensure that naive programs do something reasonable > when their stdout suddenly goes away. Or, transposing the PoV, it allows >

Re: svn commit: r285404 - head/sys/compat/cloudabi

2015-07-12 Thread David Chisnall
On 11 Jul 2015, at 21:56, Konstantin Belousov wrote: > >> Bucket 2: The system call could also just fail and return an error >> (MSG_NOSIGPIPE). > SIGPIPE exists to ensure that naive programs do something reasonable > when their stdout suddenly goes away. Or, transposing the PoV, it allows > to w

Re: svn commit: r285404 - head/sys/compat/cloudabi

2015-07-11 Thread Konstantin Belousov
On Sat, Jul 11, 2015 at 10:36:59PM +0200, Ed Schouten wrote: > Hi Kostik, > > 2015-07-11 21:47 GMT+02:00 Konstantin Belousov : > > This is very strange and unmaintanable approach. Also, you cannot handle > > traps this way. > > Yes. That's a pretty good observation. > > Some time ago I thought a

Re: svn commit: r285404 - head/sys/compat/cloudabi

2015-07-11 Thread Ed Schouten
Hi Kostik, 2015-07-11 21:47 GMT+02:00 Konstantin Belousov : > This is very strange and unmaintanable approach. Also, you cannot handle > traps this way. Yes. That's a pretty good observation. Some time ago I thought about this and my view is that signals can be grouped in three buckets: Bucket

Re: svn commit: r285404 - head/sys/compat/cloudabi

2015-07-11 Thread Konstantin Belousov
On Sat, Jul 11, 2015 at 07:41:31PM +, Ed Schouten wrote: > Author: ed > Date: Sat Jul 11 19:41:31 2015 > New Revision: 285404 > URL: https://svnweb.freebsd.org/changeset/base/285404 > > Log: > Implement normal and abnormal process termination. > > CloudABI does not provide an explicit k

svn commit: r285404 - head/sys/compat/cloudabi

2015-07-11 Thread Ed Schouten
Author: ed Date: Sat Jul 11 19:41:31 2015 New Revision: 285404 URL: https://svnweb.freebsd.org/changeset/base/285404 Log: Implement normal and abnormal process termination. CloudABI does not provide an explicit kill() system call, for the reason that there is no access to the global proce