summaryrefslogtreecommitdiffstats
path: root/arch/arm/mm/flush.c
diff options
context:
space:
mode:
authorNicolas Pitre <nico@fluxnic.net>2010-03-29 21:46:02 +0100
committerRussell King <rmk+kernel@arm.linux.org.uk>2010-04-14 11:11:27 +0100
commit7e5a69e83ba7a0d5917ad830f417cba8b8d6aa72 (patch)
treed8547f21cc0dd6fbc605d5f72c5662f65bbd18cd /arch/arm/mm/flush.c
parent317aa408d69a5b833a116317c18c7e957989ce44 (diff)
downloadblackbird-op-linux-7e5a69e83ba7a0d5917ad830f417cba8b8d6aa72.tar.gz
blackbird-op-linux-7e5a69e83ba7a0d5917ad830f417cba8b8d6aa72.zip
ARM: 6007/1: fix highmem with VIPT cache and DMA
The VIVT cache of a highmem page is always flushed before the page is unmapped. This cache flush is explicit through flush_cache_kmaps() in flush_all_zero_pkmaps(), or through __cpuc_flush_dcache_area() in kunmap_atomic(). There is also an implicit flush of those highmem pages that were part of a process that just terminated making those pages free as the whole VIVT cache has to be flushed on every task switch. Hence unmapped highmem pages need no cache maintenance in that case. However unmapped pages may still be cached with a VIPT cache because the cache is tagged with physical addresses. There is no need for a whole cache flush during task switching for that reason, and despite the explicit cache flushes in flush_all_zero_pkmaps() and kunmap_atomic(), some highmem pages that were mapped in user space end up still cached even when they become unmapped. So, we do have to perform cache maintenance on those unmapped highmem pages in the context of DMA when using a VIPT cache. Unfortunately, it is not possible to perform that cache maintenance using physical addresses as all the L1 cache maintenance coprocessor functions accept virtual addresses only. Therefore we have no choice but to set up a temporary virtual mapping for that purpose. And of course the explicit cache flushing when unmapping a highmem page on a system with a VIPT cache now can go, which should increase performance. While at it, because the code in __flush_dcache_page() has to be modified anyway, let's also make sure the mapped highmem pages are pinned with kmap_high_get() for the duration of the cache maintenance operation. Because kunmap() does unmap highmem pages lazily, it was reported by Gary King <GKing@nvidia.com> that those pages ended up being unmapped during cache maintenance on SMP causing segmentation faults. Signed-off-by: Nicolas Pitre <nico@marvell.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Diffstat (limited to 'arch/arm/mm/flush.c')
-rw-r--r--arch/arm/mm/flush.c25
1 files changed, 15 insertions, 10 deletions
diff --git a/arch/arm/mm/flush.c b/arch/arm/mm/flush.c
index e34f095e2090..c6844cb9b508 100644
--- a/arch/arm/mm/flush.c
+++ b/arch/arm/mm/flush.c
@@ -13,6 +13,7 @@
#include <asm/cacheflush.h>
#include <asm/cachetype.h>
+#include <asm/highmem.h>
#include <asm/smp_plat.h>
#include <asm/system.h>
#include <asm/tlbflush.h>
@@ -152,21 +153,25 @@ void copy_to_user_page(struct vm_area_struct *vma, struct page *page,
void __flush_dcache_page(struct address_space *mapping, struct page *page)
{
- void *addr = page_address(page);
-
/*
* Writeback any data associated with the kernel mapping of this
* page. This ensures that data in the physical page is mutually
* coherent with the kernels mapping.
*/
-#ifdef CONFIG_HIGHMEM
- /*
- * kmap_atomic() doesn't set the page virtual address, and
- * kunmap_atomic() takes care of cache flushing already.
- */
- if (addr)
-#endif
- __cpuc_flush_dcache_area(addr, PAGE_SIZE);
+ if (!PageHighMem(page)) {
+ __cpuc_flush_dcache_area(page_address(page), PAGE_SIZE);
+ } else {
+ void *addr = kmap_high_get(page);
+ if (addr) {
+ __cpuc_flush_dcache_area(addr, PAGE_SIZE);
+ kunmap_high(page);
+ } else if (cache_is_vipt()) {
+ pte_t saved_pte;
+ addr = kmap_high_l1_vipt(page, &saved_pte);
+ __cpuc_flush_dcache_area(addr, PAGE_SIZE);
+ kunmap_high_l1_vipt(page, saved_pte);
+ }
+ }
/*
* If this is a page cache page, and we have an aliasing VIPT cache,
OpenPOWER on IntegriCloud