Re: [patch] new spinlock variant, spinlock-2.3.30-A4

Andrea Arcangeli (andrea@suse.de)
Sat, 4 Dec 1999 14:53:35 +0100 (CET)


On 4 Dec 1999, Dimitris Michailidis wrote:

>It doesn't work this way. Stores can be forwarded to reads within a CPU
>before they become externally visible. Your suggestion would considerably
>hurt SMP performance.

I only wanted to tell how things should work in hardware to avoid us to
put lock on the bus to prevent the CPU to reorder instruction. Of course I
agree it doesn't work this way otherwise the testcase wouldn't deadlock ;).

Andrea

-
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.tux.org/lkml/