| commit | fe3f6208a00f39df1c20c9e18bd55576ee1a1bd9 | |
| author | Rusty Russell <rusty@rustcorp.com.au> | |
| Thu, 9 Oct 2003 00:42:12 +0000 (8 17:42 -0700) | ||
| committer | Linus Torvalds <torvalds@home.osdl.org> | |
| Thu, 9 Oct 2003 00:42:12 +0000 (8 17:42 -0700) | ||
| tree | 2744e009396ebf8ee1d89fbfce39f6759d0f4e0f | treesnapshot (tar.gz zip) |
| parent | b6e31b8e1e3b9fb29adcef9b1dde2b9865494c7c | commitdiff |
[PATCH] More barriers in module code.
Paul McKenney convinced me that there's no *guarantee* that all archs
will refuse the speculate the atomic ops above the state test, eg:
CPU0 (stopref_set_state) CPU1 (stopref)
atomic_set(&ack, 0); if (state == XXX)
wmb(); atomic_inc(&ack);
state = XXX;
Certainly Alpha needs a rmb() inside stopref to guarantee it sees
the same ordering.
Paul McKenney convinced me that there's no *guarantee* that all archs
will refuse the speculate the atomic ops above the state test, eg:
CPU0 (stopref_set_state) CPU1 (stopref)
atomic_set(&ack, 0); if (state == XXX)
wmb(); atomic_inc(&ack);
state = XXX;
Certainly Alpha needs a rmb() inside stopref to guarantee it sees
the same ordering.
| kernel/module.c | diffblobblamehistory |