This is the mail archive of the libc-hacker@sources.redhat.com mailing list for the glibc project.

Note that libc-hacker is a closed list. You may look at the archives of this list, but subscription and posting are not open.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] sparc* bits/atomic.h


> 2003-03-31  Jakub Jelinek  <jakub at redhat dot com>
> 
> 	* sysdeps/sparc/sparc32/bits/atomic.h: New file.
> 	* sysdeps/sparc/sparc32/sparcv9/bits/atomic.h: New file.
> 	* sysdeps/sparc/sparc64/bits/atomic.h: New file.
> 
> 	* sysdeps/sparc/sparc32/atomicity.h: Removed.
> 	* sysdeps/sparc/sparc32/sparcv9/atomicity.h: Removed.
> 	* sysdeps/sparc/sparc64/atomicity.h: Removed.

Thanks, applied.

> sparc32 (not v9) lacks compare and swap, so it is implemented using
> test-and-set and a global lock. IMHO better than no locking at all,
> performance will suck as usually on those aging boxes.

Are there any actual MP boxes using those chips that Linux runs on?  If not
then you could use a software atomic sequence with kernel support like SH does.
e.g., the low bit of the thread register says to reload the PC from some
other register (and clear the flag bit).


Thanks,
Roland


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]