[PATCH] mm: memory_hotplug: no need to check res twice in add_memory

[PATCH] mm: memory_hotplug: no need to check res twice in add_memory

[Sasha Levin]

Remove one redundant check of res.

Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
---
 mm/memory_hotplug.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 962e353..4082244 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -898,8 +898,7 @@ error:
 	/* rollback pgdat allocation and others */
 	if (new_pgdat)
 		rollback_node_hotadd(nid, pgdat);
-	if (res)
-		release_memory_resource(res);
+	release_memory_resource(res);
 
 out:
 	unlock_memory_hotplug();
-- 
1.8.0

--
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>

[PATCH] mm/sparse: don't check return value of alloc_bootmem calls

[Sasha Levin]

These calls will panic if they can't allocate the memory, so we can
assume that we get actual memory back. This simplifies some functions
and removes unneeded checks.

Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
---
 mm/sparse.c | 18 ++++++------------
 1 file changed, 6 insertions(+), 12 deletions(-)

diff --git a/mm/sparse.c b/mm/sparse.c
index 6b5fb76..ae64d6e 100644
--- a/mm/sparse.c
+++ b/mm/sparse.c
@@ -403,15 +403,13 @@ void __init sparse_mem_maps_populate_node(struct page **map_map,
 	size = PAGE_ALIGN(size);
 	map = __alloc_bootmem_node_high(NODE_DATA(nodeid), size * map_count,
 					 PAGE_SIZE, __pa(MAX_DMA_ADDRESS));
-	if (map) {
-		for (pnum = pnum_begin; pnum < pnum_end; pnum++) {
-			if (!present_section_nr(pnum))
-				continue;
-			map_map[pnum] = map;
-			map += size;
-		}
-		return;
+	for (pnum = pnum_begin; pnum < pnum_end; pnum++) {
+		if (!present_section_nr(pnum))
+			continue;
+		map_map[pnum] = map;
+		map += size;
 	}
+	return;
 
 	/* fallback */
 	for (pnum = pnum_begin; pnum < pnum_end; pnum++) {
@@ -497,8 +495,6 @@ void __init sparse_init(void)
 	 */
 	size = sizeof(unsigned long *) * NR_MEM_SECTIONS;
 	usemap_map = alloc_bootmem(size);
-	if (!usemap_map)
-		panic("can not allocate usemap_map\n");
 
 	for (pnum = 0; pnum < NR_MEM_SECTIONS; pnum++) {
 		struct mem_section *ms;
@@ -538,8 +534,6 @@ void __init sparse_init(void)
 #ifdef CONFIG_SPARSEMEM_ALLOC_MEM_MAP_TOGETHER
 	size2 = sizeof(struct page *) * NR_MEM_SECTIONS;
 	map_map = alloc_bootmem(size2);
-	if (!map_map)
-		panic("can not allocate map_map\n");
 
 	for (pnum = 0; pnum < NR_MEM_SECTIONS; pnum++) {
 		struct mem_section *ms;
-- 
1.8.0

--
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>

Re: [PATCH] mm/sparse: don't check return value of alloc_bootmem calls

[David Rientjes]

On Thu, 20 Dec 2012, Sasha Levin wrote:

> diff --git a/mm/sparse.c b/mm/sparse.c
> index 6b5fb76..ae64d6e 100644
> --- a/mm/sparse.c
> +++ b/mm/sparse.c
> @@ -403,15 +403,13 @@ void __init sparse_mem_maps_populate_node(struct page **map_map,
>  	size = PAGE_ALIGN(size);
>  	map = __alloc_bootmem_node_high(NODE_DATA(nodeid), size * map_count,
>  					 PAGE_SIZE, __pa(MAX_DMA_ADDRESS));
> -	if (map) {
> -		for (pnum = pnum_begin; pnum < pnum_end; pnum++) {
> -			if (!present_section_nr(pnum))
> -				continue;
> -			map_map[pnum] = map;
> -			map += size;
> -		}
> -		return;
> +	for (pnum = pnum_begin; pnum < pnum_end; pnum++) {
> +		if (!present_section_nr(pnum))
> +			continue;
> +		map_map[pnum] = map;
> +		map += size;
>  	}
> +	return;
>  
>  	/* fallback */
>  	for (pnum = pnum_begin; pnum < pnum_end; pnum++) {

That's not true when slab_is_available() and why would you possibly add a 
return statement right before fallback code in such cases?

--
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>

Re: [PATCH] mm/sparse: don't check return value of alloc_bootmem calls

[Sasha Levin]

On 12/20/2012 03:23 PM, David Rientjes wrote:
> On Thu, 20 Dec 2012, Sasha Levin wrote:
> 
>> diff --git a/mm/sparse.c b/mm/sparse.c
>> index 6b5fb76..ae64d6e 100644
>> --- a/mm/sparse.c
>> +++ b/mm/sparse.c
>> @@ -403,15 +403,13 @@ void __init sparse_mem_maps_populate_node(struct page **map_map,
>>  	size = PAGE_ALIGN(size);
>>  	map = __alloc_bootmem_node_high(NODE_DATA(nodeid), size * map_count,
>>  					 PAGE_SIZE, __pa(MAX_DMA_ADDRESS));
>> -	if (map) {
>> -		for (pnum = pnum_begin; pnum < pnum_end; pnum++) {
>> -			if (!present_section_nr(pnum))
>> -				continue;
>> -			map_map[pnum] = map;
>> -			map += size;
>> -		}
>> -		return;
>> +	for (pnum = pnum_begin; pnum < pnum_end; pnum++) {
>> +		if (!present_section_nr(pnum))
>> +			continue;
>> +		map_map[pnum] = map;
>> +		map += size;
>>  	}
>> +	return;
>>  
>>  	/* fallback */
>>  	for (pnum = pnum_begin; pnum < pnum_end; pnum++) {
> 
> That's not true when slab_is_available() and why would you possibly add a 
> return statement right before fallback code in such cases?

So what we really need is to update the documentation of __alloc_bootmem_node, I'll send
a patch that does that instead.

I've dragged the 'return' out of the if(map) condition, I should have removed the fallback
as well and so the return would also be gone, but that's irrelevant now.


Thanks,
Sasha

--
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>

Re: [PATCH] mm/sparse: don't check return value of alloc_bootmem calls

[David Rientjes]

On Thu, 20 Dec 2012, Sasha Levin wrote:

> So what we really need is to update the documentation of __alloc_bootmem_node, I'll send
> a patch that does that instead.
> 

It panics iff slab is not available to allocate from yet, otherwise it's 
just a wrapper around kmalloc().  This emits a warning to the kernel log, 
though, so __alloc_bootmem_node() should certainly not be called that late 
in the boot sequence.

Since __alloc_bootmem_node_nopanic() is the way to avoid the panic, I 
think the change that should be made here so to panic even when the 
kmalloc() fails in __alloc_bootmem_node(), __alloc_bootmem_node_high(), 
and __alloc_bootmem_low_node().

--
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>

[PATCH] slob: use DIV_ROUND_UP where possible

[Sasha Levin]

Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
---
 mm/slob.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/slob.c b/mm/slob.c
index a99fdf7..f729c46 100644
--- a/mm/slob.c
+++ b/mm/slob.c
@@ -122,7 +122,7 @@ static inline void clear_slob_page_free(struct page *sp)
 }
 
 #define SLOB_UNIT sizeof(slob_t)
-#define SLOB_UNITS(size) (((size) + SLOB_UNIT - 1)/SLOB_UNIT)
+#define SLOB_UNITS(size) DIV_ROUND_UP(size, SLOB_UNIT)
 
 /*
  * struct slob_rcu is inserted at the tail of allocated slob blocks, which
-- 
1.8.0

--
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>

Re: [PATCH] slob: use DIV_ROUND_UP where possible

[Christoph Lameter]

Obviously good.

Acked-by: Christoph Lameter <cl@linux.com>

--
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>