It's an interrupt gate, not a call gate; the distinction is subtle.
Incidentally, *NEITHER* is "TSS based"; that would be a task gate.
Now, very recent Intel processors (Pentium II and newer) contain a new
SYSENTER instruction would would be a *lot* more efficient, but would
of course require wholesale libc changes.
> 2) I don't see the connection between kernel entry and TSS based task
> switch. Mingo already has patches that replace the current TSS method
> with software based context switchs.
Yes, task switches (NOT context switches) use the TSS. There are some
things that could make a non-TSS based system cheaper (mainly avoiding
to load all the segment registers) but the difference ain't that
great. Mingo's patches would eliminate the use of the hardware task
switch; the main reason isn't speed but the annoying 4000-process
limit that imposes, as well as better handling of the I/O bitmap.
> 3) I'm not sure about Terrys definition of "efficient", but if he means
> required CPU cycles I thought that there were not that many left in
> Linux that FreeBSD could be better. Has anybody hard numbers for both
> kernels?
No hard numbers, but Terry seems to be letting his advocacy get away
of the facts in this particular case, I'm sorry to say.
-hpa
--
PGP: 2047/2A960705 BA 03 D3 2C 14 A8 A8 BD 1E DF FE 69 EE 35 BD 74
See http://www.zytor.com/~hpa/ for web page and full PGP public key
I am Bahá'í -- ask me about it or see http://www.bahai.org/
"To love another person is to see the face of God." -- Les Misérables
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html