Archives
Archives
| From: | Jack Lloyd <lloyd@×××××××××.net> | ||
|---|---|---|---|
| To: | gentoo-amd64@l.g.o | ||
| Subject: | Re: [gentoo-amd64] Re: Unexpected side effect of GCC 4 | ||
| Date: | Mon, 06 Nov 2006 12:51:17 | ||
| Message-Id: | 20061106124912.GI5921@randombit.net | ||
| In Reply to: | Re: [gentoo-amd64] Re: Unexpected side effect of GCC 4 by "Hemmann | ||
| 1 | On Mon, Nov 06, 2006 at 01:05:31PM +0100, Hemmann, Volker Armin wrote: |
| 2 | |
| 3 | > lets ask the other way round. Why should it speed up anything to have X>number |
| 4 | > of cores? Instead of a single thread per core, compiling happily, you have |
| 5 | > two or more competing for one core and regularly kick out each others data |
| 6 | > from the cache. |
| 7 | |
| 8 | To account for I/O wait states |
| 9 | -- |
| 10 | gentoo-amd64@g.o mailing list |
| Subject | Author |
|---|---|
| Re: [gentoo-amd64] Re: Unexpected side effect of GCC 4 | "Hemmann |