Thread (13 messages) 13 messages, 3 authors, 2017-01-02

Re: [PATCH] mm: cma: print allocation failure reason and bitmap status

From: Michal Nazarewicz <hidden>
Date: 2017-01-01 21:59:49
Also in: lkml

On Fri, Dec 30 2016, Michal Hocko wrote:
On Fri 30-12-16 16:24:46, Jaewon Kim wrote:
[...]
quoted
quoted
From 7577cc94da3af27907aa6eec590d2ef51e4b9d80 Mon Sep 17 00:00:00 2001
From: Jaewon Kim <redacted>
Date: Thu, 29 Dec 2016 11:00:16 +0900
Subject: [PATCH] mm: cma: print allocation failure reason and bitmap status

There are many reasons of CMA allocation failure such as EBUSY, ENOMEM, EINTR.
But we did not know error reason so far. This patch prints the error value.

Additionally if CONFIG_CMA_DEBUG is enabled, this patch shows bitmap status to
know available pages. Actually CMA internally try all available regions because
some regions can be failed because of EBUSY. Bitmap status is useful to know in
detail on both ENONEM and EBUSY;
 ENOMEM: not tried at all because of no available region
         it could be too small total region or could be fragmentation issue
 EBUSY:  tried some region but all failed

This is an ENOMEM example with this patch.
[   13.250961]  [1:   Binder:715_1:  846] cma: cma_alloc: alloc failed, req-size: 256 pages, ret: -12
Avabile pages also will be shown if CONFIG_CMA_DEBUG is enabled
[   13.251052]  [1:   Binder:715_1:  846] cma: number of available pages: 4@572+7@585+7@601+8@632+38@730+166@1114+127@1921=>357 pages, total: 2048 pages
please mention how to interpret this information.

some more style suggestions below
quoted
Signed-off-by: Jaewon Kim <redacted>
Acked-by: Michal Nazarewicz <redacted>
quoted
---
 mm/cma.c | 29 ++++++++++++++++++++++++++++-
 1 file changed, 28 insertions(+), 1 deletion(-)
diff --git a/mm/cma.c b/mm/cma.c
index c960459..1bcd9db 100644
--- a/mm/cma.c
+++ b/mm/cma.c
@@ -369,7 +369,7 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
     unsigned long start = 0;
     unsigned long bitmap_maxno, bitmap_no, bitmap_count;
     struct page *page = NULL;
-    int ret;
+    int ret = -ENOMEM;
 
     if (!cma || !cma->count)
         return NULL;
@@ -427,6 +427,33 @@ struct page *cma_alloc(struct cma *cma, size_t count, unsigned int align)
     trace_cma_alloc(pfn, page, count, align);
 
     pr_debug("%s(): returned %p\n", __func__, page);
This line should be moved after the ‘if (ret != 0)’ block, i.e. just
before return.
quoted
+
+    if (ret != 0)
you can simply do
	if (!ret) {

		pr_info("%s: alloc failed, req-size: %zu pages, ret: %d\n",
			__func__, count, ret);
		debug_show_cma_areas();
	}

	return page;

static void debug_show_cma_areas(void)
{
#ifdef CONFIG_CMA_DEBUG
	unsigned int nr, nr_total = 0;
	unsigned long next_set_bit;

	mutex_lock(&cma->lock);
	pr_info("number of available pages: ");
	start = 0;
	for (;;) {
		bitmap_no = find_next_zero_bit(cma->bitmap, cma->count, start);
		if (bitmap_no >= cma->count)
		break;
		next_set_bit = find_next_bit(cma->bitmap, cma->count, bitmap_no);
		nr = next_set_bit - bitmap_no;
		pr_cont("%s%u@%lu", nr_total ? "+" : "", nr, bitmap_no);
		nr_total += nr;
		start = bitmap_no + nr;
	}
	pr_cont("=>%u pages, total: %lu pages\n", nr_total, cma->count);
Perhaps:
	pr_cont("=> %u free of %lu total pages\n", nr_total, cma->count);
or shorter (but more cryptic):
	pr_cont("=> %u/%lu pages\n", nr_total, cma->count);
	mutex_unlock(&cma->lock);
#endif
}
Actually, Linux style is more like:

#ifdef CONFIG_CMA_DEBUG
static void cma_debug_show_areas()
{
	…
}
#else
static inline void cma_debug_show_areas() { }
#endif
-- 
Michal Hocko
SUSE Labs
-- 
Best regards
ミハウ “𝓶𝓲𝓷𝓪86” ナザレヴイツ
«If at first you don’t succeed, give up skydiving»

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help