Re: [Linux-kernel-mentees][PATCH v4 1/1] sgi-gru: Remove *pte_lookup functions

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 8/9/19 2:44 AM, Bharath Vedartham wrote:
On Thu, Aug 08, 2019 at 04:21:44PM -0700, John Hubbard wrote:
On 8/8/19 11:55 AM, Bharath Vedartham wrote:
...
  static int gru_vtop(struct gru_thread_state *gts, unsigned long vaddr,
  		    int write, int atomic, unsigned long *gpa, int *pageshift)
  {
  	struct mm_struct *mm = gts->ts_mm;
  	struct vm_area_struct *vma;
  	unsigned long paddr;
-	int ret, ps;
+	int ret;
+	struct page *page;
vma = find_vma(mm, vaddr);
  	if (!vma)
@@ -263,21 +187,33 @@ static int gru_vtop(struct gru_thread_state *gts, unsigned long vaddr,
/*
  	 * Atomic lookup is faster & usually works even if called in non-atomic
-	 * context.
+	 * context. get_user_pages_fast does atomic lookup before falling back to
+	 * slow gup.
  	 */
  	rmb();	/* Must/check ms_range_active before loading PTEs */
-	ret = atomic_pte_lookup(vma, vaddr, write, &paddr, &ps);
-	if (ret) {
-		if (atomic)
+	if (atomic) {
+		ret = __get_user_pages_fast(vaddr, 1, write, &page);
+		if (!ret)
  			goto upm;
-		if (non_atomic_pte_lookup(vma, vaddr, write, &paddr, &ps))
+	} else {
+		ret = get_user_pages_fast(vaddr, 1, write ? FOLL_WRITE : 0, &page);
+		if (!ret)
  			goto inval;
  	}
+
+	paddr = page_to_phys(page);
+	put_user_page(page);
+
+	if (unlikely(is_vm_hugetlb_page(vma)))
+		*pageshift = HPAGE_SHIFT;
+	else
+		*pageshift = PAGE_SHIFT;
+
  	if (is_gru_paddr(paddr))
  		goto inval;
-	paddr = paddr & ~((1UL << ps) - 1);
+	paddr = paddr & ~((1UL << *pageshift) - 1);
  	*gpa = uv_soc_phys_ram_to_gpa(paddr);
-	*pageshift = ps;

Why are you no longer setting *pageshift? There are a couple of callers
that both use this variable.
Hi John,

I did set *pageshift. The if statement above sets *pageshift. ps was
used to retrive the pageshift value when the pte_lookup functions were
present. ps was passed by reference to those functions and set by them.
But here since we are trying to remove those functions, we don't need ps
and we directly set *pageshift to HPAGE_SHIFT or PAGE_SHIFT based on the
type of vma.

Hope this clears things up?


Right you are, sorry for overlooking that. Looks good.

thanks,
--
John Hubbard
NVIDIA




[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux