From: Manfred S. <ma...@co...> - 2001-11-05 19:49:42
|
Hubertus Franke wrote: > > You that sounds completely unreasonable. We have not dived that deep > into the magic of vmm and pipe interaction... :-) > Check Documentation/cachetlb.txt. It has nothing to to with vmm and pipe, it's just just stupid design of cpu caches and Linux architecture hooks. Just two quotes from cachetlb.txt: <<<< Admittedly, the author did not think very much when designing this interface. It does not give the architecture enough information about what exactly is going on <<<< And that was fixed now: now the hooks get so many <<<< parameters that it's impossible for pipe_read to supply <<<< efficiently. All the functionality of flush_icache_page can be implemented in flush_dcache_page and update_mmu_cache. In 2.5 the hope is to remove this interface completely. <<<< Obvious, right? icache flushing happens during update_mmu_cache. I usually use kernel/ptrace.c as the reference, because debuggers and strace immediately show missing flush macros and that code is hopefully correct. -- Manfred |