[lkml]   [2009]   [Mar]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [RFC] atomic highmem kmap page pinning
On Wed, 04 Mar 2009 00:58:13 -0500 (EST) Nicolas Pitre <> wrote:

> I've implemented highmem for ARM. Yes, some ARM machines do have lots
> of memory...
> The problem is that most ARM machines have a non IO coherent cache,
> meaning that the dma_map_* set of functions must clean and/or invalidate
> the affected memory manually. And because the majority of those
> machines have a VIVT cache, the cache maintenance operations must be
> performed using virtual addresses.
> In dma_map_page(), an highmem pages could still be mapped and cached
> even after kunmap() was called on it. As long as highmem pages are
> mapped, page_address(page) is non null and we can use that to
> synchronize the cache.
> It is unlikely but still possible for kmap() to race and recycle the
> obtained virtual address above, and use it for another page though. In
> that case, the new mapping could end up with dirty cache lines for
> another page, and the unsuspecting cache invalidation loop in
> dma_map_page() won't notice resulting in data loss. Hence the need for
> some kind of kmap page pinning which can be used in any context,
> including IRQ context.
> This is a RFC patch implementing the necessary part in the core code, as
> suggested by RMK. Please comment.

Seems harmless enough to me.

> +void *kmap_high_get(struct page *page)
> +{
> + unsigned long vaddr, flags;
> +
> + spin_lock_irqsave(&kmap_lock, flags);
> + vaddr = (unsigned long)page_address(page);
> + if (vaddr) {
> + BUG_ON(pkmap_count[PKMAP_NR(vaddr)] < 1);
> + pkmap_count[PKMAP_NR(vaddr)]++;
> + }
> + spin_unlock_irqrestore(&kmap_lock, flags);
> + return (void*) vaddr;
> +}

We could remove a pile of ugly casts if we changed PKMAP_NR() to take a
void*. Not that this is relevant.

 \ /
  Last update: 2009-03-04 08:43    [W:0.068 / U:2.016 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site