[PATCH] Flush dcache before writing into page to avoid alias

anfei zhou anfei.zhou at gmail.com
Mon Jan 25 20:01:10 EST 2010


On Tue, Jan 26, 2010 at 3:58 AM, Andrew Morton
<akpm at linux-foundation.org> wrote:
> On Mon, 25 Jan 2010 21:33:08 +0800 anfei <anfei.zhou at gmail.com> wrote:
>
>> Hi Andrew,
>>
>> On Thu, Jan 21, 2010 at 01:07:57PM +0800, anfei zhou wrote:
>> > The cache alias problem will happen if the changes of user shared mapping
>> > is not flushed before copying, then user and kernel mapping may be mapped
>> > into two different cache line, it is impossible to guarantee the coherence
>> > after iov_iter_copy_from_user_atomic.  So the right steps should be:
>> >     flush_dcache_page(page);
>> >     kmap_atomic(page);
>> >     write to page;
>> >     kunmap_atomic(page);
>> >     flush_dcache_page(page);
>> > More precisely, we might create two new APIs flush_dcache_user_page and
>> > flush_dcache_kern_page to replace the two flush_dcache_page accordingly.
>> >
>> > Here is a snippet tested on omap2430 with VIPT cache, and I think it is
>> > not ARM-specific:
>> >     int val = 0x11111111;
>> >     fd = open("abc", O_RDWR);
>> >     addr = mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_SHARED, fd, 0);
>> >     *(addr+0) = 0x44444444;
>> >     tmp = *(addr+0);
>> >     *(addr+1) = 0x77777777;
>> >     write(fd, &val, sizeof(int));
>> >     close(fd);
>> > The results are not always 0x11111111 0x77777777 at the beginning as expected.
>> >
>> Is this a real bug or not necessary to support?
>
> Bug.  If variable `addr' has type int* then the contents of that file
> should be 0x11111111 0x77777777.  You didn't tell us what the contents
> were in the incorrect case, but I guess it doesn't matter.
>
Sorry, I didn't give the details, here is the old thread with more details:
  http://linux.derkeiler.com/Mailing-Lists/Kernel/2010-01/msg07124.html

Regards,
Anfei.
>
>



More information about the linux-arm-kernel mailing list