Hi Aneesh,
On Mon, Oct 27, 2014 at 11:28:41PM +0530, Aneesh Kumar K.V wrote:
> VM_BUG_ON(address & ~HPAGE_PMD_MASK);
> if (pmd_trans_huge(*pmdp)) {
> pmd = pmdp_get_and_clear(vma->vm_mm, address, pmdp);
> } else {
The only problematic path that needs IPI is the below one yes.
> /*
> * khugepaged calls this for normal pmd
> */
> pmd = *pmdp;
> pmd_clear(pmdp);
> /*
> * Wait for all pending hash_page to finish. This is needed
> * in case of subpage collapse. When we collapse normal pages
> * to hugepage, we first clear the pmd, then invalidate all
> * the PTE entries. The assumption here is that any low level
> * page fault will see a none pmd and take the slow path that
> * will wait on mmap_sem. But we could very well be in a
> * hash_page with local ptep pointer value. Such a hash page
> * can result in adding new HPTE entries for normal subpages.
> * That means we could be modifying the page content as we
> * copy them to a huge page. So wait for parallel hash_page
> * to finish before invalidating HPTE entries. We can do this
> * by sending an IPI to all the cpus and executing a dummy
> * function there.
> */
> kick_all_cpus_sync();
>
> We already do an IPI for ppc64.
Agreed, ppc64 is already covered.
sparc/arm seem to be using the generic pmdp_clear_flush implementation
instead, which just calls flush_tlb_range, so perhaps they aren't.
As above, the IPIs are only needed if the *pmd is not transhuge.
Thanks,
Andrea
--
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>