Messages in this thread | | | Date | Mon, 1 Sep 2003 13:30:49 +0200 (MEST) | From | Geert Uytterhoeven <> | Subject | Re: x86, ARM, PARISC, PPC, MIPS and Sparc folks please run this |
| |
On Mon, 1 Sep 2003, Jamie Lokier wrote: > There is a bug in test_l1_only which I just noticed. It's unlikely, > but if `dummy' happens to have the same L1 cache address as both words > being tested, and it's a 2-way (or less) set-associative cache, then > it will inadvertently flush the cache and say "store buffer not > coherent" when it means to say "cache not coherent". > > Please try the program below, which is the same as before but with > test_l1_only hopefully improved, and it prints some more helpful > numbers.
Results for 68040 with the new version:
cassandra:/tmp# time ./test2 Test separation: 4096 bytes: FAIL - store buffer not coherent Test separation: 8192 bytes: FAIL - store buffer not coherent Test separation: 16384 bytes: FAIL - store buffer not coherent Test separation: 32768 bytes: FAIL - store buffer not coherent Test separation: 65536 bytes: FAIL - store buffer not coherent Test separation: 131072 bytes: FAIL - store buffer not coherent Test separation: 262144 bytes: FAIL - store buffer not coherent Test separation: 524288 bytes: FAIL - store buffer not coherent Test separation: 1048576 bytes: FAIL - store buffer not coherent Test separation: 2097152 bytes: FAIL - store buffer not coherent Test separation: 4194304 bytes: FAIL - store buffer not coherent Test separation: 8388608 bytes: FAIL - store buffer not coherent Test separation: 16777216 bytes: FAIL - store buffer not coherent VM page alias coherency test: failed; will use copy buffers instead
real 0m0.454s user 0m0.090s sys 0m0.210s cassandra:/tmp# cat /proc/cpuinfo CPU: 68040 MMU: 68040 FPU: 68040 Clocking: 24.8MHz BogoMips: 16.53 Calibration: 82688 loops cassandra:/tmp#
New m68k binary at http://home.tvd.be/cr26864/Linux/m68k/jamie_test2.gz
Gr{oetje,eeting}s,
Geert
-- Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that. -- Linus Torvalds
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |