| To: | Atsushi Nemoto <anemo@mba.ocn.ne.jp> |
|---|---|
| Subject: | Re: minor c-r4k.c cleanup |
| From: | Ralf Baechle <ralf@linux-mips.org> |
| Date: | Mon, 14 Apr 2003 18:48:33 +0200 |
| Cc: | linux-mips@linux-mips.org |
| In-reply-to: | <20030415.002425.74756927.anemo@mba.ocn.ne.jp>; from anemo@mba.ocn.ne.jp on Tue, Apr 15, 2003 at 12:24:25AM +0900 |
| Original-recipient: | rfc822;linux-mips@linux-mips.org |
| References: | <20030415.002425.74756927.anemo@mba.ocn.ne.jp> |
| Sender: | linux-mips-bounce@linux-mips.org |
| User-agent: | Mutt/1.2.5.1i |
On Tue, Apr 15, 2003 at 12:24:25AM +0900, Atsushi Nemoto wrote: > r4k_dma_cache_xxx should not flush icache ? The dma_* functions of course don't need I-cache coherence. > And I wonder why r4k_flush_pcache_mm (and r4k_flush_pcache_all) does > nothing if cpu_has_dc_aliases was not true. I'm still > investigating... R4000SC / R4400SC caches have the subset property. S-cache cacheops flush the primary caches as well and P-caches must always be a subset of the S-cache or operation is undefined. Ralf |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [patch] Board bus error handler clean-ups, Maciej W. Rozycki |
|---|---|
| Next by Date: | Re: Oddities with CVS Kernels, Memory on Indigo2, Kumba |
| Previous by Thread: | minor c-r4k.c cleanup, Atsushi Nemoto |
| Next by Thread: | Soft floating point on 5K, Dennis Castleman |
| Indexes: | [Date] [Thread] [Top] [All Lists] |