Re: [PATCH v3 2/5] mm/cma: add cma_pages_valid to determine if pages are in CMA

On 10/5/21 1:45 AM, Oscar Salvador wrote:

On Fri, Oct 01, 2021 at 10:52:07AM -0700, Mike Kravetz wrote:

quoted

+bool cma_pages_valid(struct cma *cma, const struct page *pages,
+		     unsigned long count)
+{
+	unsigned long pfn;
+
+	if (!cma || !pages)
+		return false;
+
+	pfn = page_to_pfn(pages);
+
+	if (pfn < cma->base_pfn || pfn >= cma->base_pfn + cma->count)
+		return false;
+
+	return true;
+}
+
 /**
  * cma_release() - release allocated pages
  * @cma:   Contiguous memory region for which the allocation is performed.

@@ -539,16 +555,13 @@ bool cma_release(struct cma *cma, const struct page *pages,
 {
 	unsigned long pfn;
 
-	if (!cma || !pages)
+	if (!cma_pages_valid(cma, pages, count))
 		return false;
 
 	pr_debug("%s(page %p, count %lu)\n", __func__, (void *)pages, count);
 
 	pfn = page_to_pfn(pages);
 
-	if (pfn < cma->base_pfn || pfn >= cma->base_pfn + cma->count)
-		return false;
-

After this patch, the timing of printing the debug statement changes as we back
off earlier.
You might want to point that out in the changelog in case someone wonders why.

When making this change, I did not want to duplicate that debug
statement.  However, duplicated code only becomes an issue if
CONFIG_DEBUG.  So, duplication should be acceptable.

I will make the debug statements function as before.
-- 
Mike Kravetz