Related articles |
---|
Info needed (Sparc C compiler w/o register window) clim@ICSI.Berkeley.EDU (1992-04-14) |
Re: Info needed (Sparc C compiler w/o register window) torek@horse.ee.lbl.gov (1992-04-15) |
Re: Info needed (Sparc C compiler w/o register window) pardo@cs.washington.edu (1992-04-16) |
No threads on SPARC? (was Re: Info needed (Sparc C compiler w/o regist mueller@mount.fsu.edu (1992-04-17) |
Newsgroups: | comp.compilers,comp.lang.c,gnu.gcc.help |
From: | pardo@cs.washington.edu (David Keppel) |
Keywords: | registers, sparc |
Organization: | Computer Science & Engineering, U. of Washington, Seattle |
References: | 92-04-058 92-04-065 |
Date: | Thu, 16 Apr 1992 15:28:46 GMT |
>>[I'm no fan of register windows, but is it really that hard to handle the
>>over- and underflow? -John]
torek@horse.ee.lbl.gov (Chris Torek) writes:
>On the SPARCstation, the answer is `yes'. The reason is a bit
>complicated, ...
For an other explanation, with essentially the same conclusion, see my TR
via anonymous ftp from `cs.washington.edu' (128.95.1.4) in
`pub/pardo/sparc-threads.ps.Z'. It's a rather long-winded description of
why you can't write a truly user-space threads package on the SPARC and
how to do a job that's close. It also has an appendix on assigning one
register window to each thread and doing context switches by switching
windows and why that's rarely a win compared to just saving and restoring
registers manually from user space.
;-D oN ( Thread light and bool ) Pardo
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.