On Thu, Jul 11, 2013 at 11:36 PM, Dave Hansen <dave.hansen@xxxxxxxxx> wrote:
> On 07/11/2013 07:34 PM, Colin Cross wrote:
>> + pages_pinned = get_user_pages(current, mm, page_start_vaddr,
>> + 1, 0, 0, &page, NULL);
>> + if (pages_pinned < 1) {
>> + seq_puts(m, "<fault>]");
>> + return;
>> + }
>> +
>> + kaddr = (const char *)kmap(page);
>> + len = min(max_len, PAGE_SIZE - page_offset);
>> + write_len = strnlen(kaddr + page_offset, len);
>> + seq_write(m, kaddr + page_offset, write_len);
>> + kunmap(page);
>> + put_page(page);
>
> This looks a bit like access_process_vm()? Can you perhaps use it here?
It's a lot like __access_remote_vm, and this pattern is repeated in
many other places in the kernel. I didn't try to reuse any of them
because I wanted to stop reading at a null byte and __access_remote_vm
would read the full NAME_MAX every time. I was also avoiding having
to allocate a NAME_MAX sized buffer to copy into, instead passing the
mapped user page directly to seq_write.
--
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>