Re: [Bug] Killing processes under new forking mechanism

[Yves Pelletier <ypelletier@cmc.doe.ca>]

On Sat, 16 Sep 1995, Frank Bartels wrote:

> Are you sure sending TERM or HUP would not have done the job?

I did try a plain "kill" (that's TERM?) before resorting to "kill -9". I
did not try HUP. 

> 
> This may also be a bug in your version of `make'. I've seen this
> problem before and not only under MiNT.

That is possible. I use the binary distribution of "make" from the MiNT
directory at Umich. I have run several different config scripts from
various packages, and only one has given me the trouble I described.
That's the make script to build smail. (Anyone putting a binary
distribution together? :-)

> 
> BTW: I did not see any comments to the no-more-blocking-fork-patch.
> Does it work for you? I still had no time to try it...
> 

I'm not sure if I should be noticing anything at the surface, but if
nothing else, the patched kernel is perfectly stable. I have mainly been
running configure scripts and compiling GNU utils. But I have also run
MTOS and a few GEM applications. I have not yet seen applications that
behave differently than before (better or worse) under a non-blocking
fork. 

Yves
___________________________________________________________________________
"You don't need to mention it. No thanks. I'm a Government Man." db/be