Re: PATCH killing dead code and design errors in pre6

David S. Miller (davem@dm.cobaltmicro.com)
Wed, 13 Jan 1999 11:49:39 -0800

If you look at the code path which actually checks for whether
constructor/destructor stuff needs to happen, SLAB is checking
multiple flag bits in a single word at once and branches to a slow
path if any of these features are being used (ctors, dtors, SLAB
debugging, etc.)

The overhead is therefore one instruction and two pointers in an
internal structure which is always hot in the cache anyways. Worth
removing the facility completely? I don't think so.

Later,
David S. Miller
davem@dm.cobaltmicro.com

-
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/