Hi, Russell,
Russell King - ARM Linux <linux@xxxxxxxxxxxxxxx> writes:
> On Sun, Feb 11, 2018 at 02:43:39PM +0800, Huang, Ying wrote:
[snip]
>>
>>
>> if page is an anonymous page in swap cache, "mapping &&
>> !mapping_mapped()" will be true, so we will delay flushing. But if my
>> understanding of the code were correct, we should call
>> flush_kernel_dcache() because the kernel may access the page during
>> swapping in/out.
>>
>> The code in other architectures follow the similar logic. Would it be
>> better for page_mapping() here to return NULL for anonymous pages even
>> if they are in swap cache? Of course we need to change the function
>> name. page_file_mapping() appears a good name, but that has been used
>> already. Any suggestion?
>
> flush_dcache_page() does nothing for anonymous pages (see cachetlb.txt,
> it's only defined to do anything for page cache pages.)
>
> flush_anon_page() deals with anonymous pages.
Thanks for your information! But I found this isn't followed exactly in
the code. For example, in get_mergeable_page() in mm/ksm.c,
if (PageAnon(page)) {
flush_anon_page(vma, page, addr);
flush_dcache_page(page);
} else {
put_page(page);
flush_dcache_page() is called for anonymous pages too.
Best Regards,
Huang, Ying
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>
