Re: [parisc-linux] rmap: parisc __flush_dcache_page

[Andrea Arcangeli]

On Thu, Apr 08, 2004 at 04:35:12PM +0100, Hugh Dickins wrote:
> It's not fixable via the i_shared_sem, but we can add another layer

I meant it's unfixable unless we change the VM common code.

> of spin_lock around the i_mmap* list/tree manipulations, one that
> preprocesses away to nothing on all arches but parisc and arm, and
> is used in their __flush_dcache_page.  *Not* the page_table_lock ;)

I'd prefer to use only a spinlock then to carry around two overlapping
locks, the need_resched() check is needed anyways even with preempt in
the real costly paths.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/