Re: [openib-general] Re: [PATCH][RFC][0/4] InfiniBand userspace verbs implementation

[Timur Tabi <timur.tabi@ammasso.com>]

Timur Tabi wrote:

> When you say "older", what exactly do you mean?  I have different test 
> that normally fails with just get_user_pages(), but it works with 2.6.9 
> and above.  I haven't been able to get any kernel earlier than 2.6.9 to 
> compile or boot properly, so I'm having a hard time narrowing down the 
> actual point when get_user_pages() started working.

I haven't gotten a reply to this question, but I've done my own research, and I think I 
found the answer.  Using my own test of get_user_pages(), it appears that the fix was 
placed in 2.6.7.  However, I would like to know specifically what the fix is. 
Unfortunately, tracking this stuff down is beyond my understanding of the Linux VM.

Assuming that the fix is in try_to_unmap_one(), the only significant change I see between
2.6.6 and 2.6.7 is the addition of this code:

	pgd = pgd_offset(mm, address);
	if (!pgd_present(*pgd))
		goto out_unlock;

	pmd = pmd_offset(pgd, address);
	if (!pmd_present(*pmd))
		goto out_unlock;

	pte = pte_offset_map(pmd, address);
	if (!pte_present(*pte))
		goto out_unmap;

	if (page_to_pfn(page) != pte_pfn(*pte))
		goto out_unmap;

Can anyone tell me if this is the actual fix, or at least a major part of the actual fix?

-- 
Timur Tabi
Staff Software Engineer
timur.tabi@ammasso.com

One thing a Southern boy will never say is,
"I don't think duct tape will fix it."
      -- Ed Smylie, NASA engineer for Apollo 13