Re: 2.1.pre45-5 and 6

David C Niemi (niemi@wauug.erols.com)
Wed, 16 Jul 1997 08:43:37 -0400 (EDT)

On Tue, 15 Jul 1997, Linus Torvalds wrote:
> Ok. I'm still accumulating a few fixes, so I'll make a -7 within the hour.
>
> The worst problem with -6 is that any pipe inodes are irrevocably lost
> after being closed, along with the one page of data that they used to
> have. That's bad if your benchmark opens a lot of pipes (lmbench does that
> for the context switch test - maybe the byte benchmarks do the same?).
>
> Linus

Yes, there are 2 pipe-oriented benchmarks, one for throughput and one for
context-switching, and they are both prior to the hang.

-6 got a little bit farther along, into the 9th Shell1 iteration (i.e.
about 8 minutes farther than -5). No oops this time, just the same style
of hang as the previous revs.

David
Niemi@erols.com 703-810-5538 Reston, Virginia, USA
--- Most operating systems, sometimes even DOS, separate different
--- types of files into different directories. The Windows philosophy:
--- Throw everything into C:\WINDOWS and let God sort it out.