[v1,3/4] mm/memory_hotplug: Make __remove_section() never fail
diff mbox series

Message ID 20190409100148.24703-4-david@redhat.com
State In Next
Commit 3a3a00b781c42cb66019545c0bb8d885530eeea1
Headers show
Series
  • mm/memory_hotplug: Better error handling when removing memory
Related show

Commit Message

David Hildenbrand April 9, 2019, 10:01 a.m. UTC
Let's just warn in case a section is not valid instead of failing to
remove somewhere in the middle of the process, returning an error that
will be mostly ignored by callers.

Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Michal Hocko <mhocko@suse.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Pavel Tatashin <pasha.tatashin@soleen.com>
Cc: Qian Cai <cai@lca.pw>
Cc: Wei Yang <richard.weiyang@gmail.com>
Cc: Arun KS <arunks@codeaurora.org>
Cc: Mathieu Malaterre <malat@debian.org>
Signed-off-by: David Hildenbrand <david@redhat.com>
---
 mm/memory_hotplug.c | 22 +++++++++-------------
 1 file changed, 9 insertions(+), 13 deletions(-)

Comments

Oscar Salvador April 17, 2019, 1:56 p.m. UTC | #1
On Tue, 2019-04-09 at 12:01 +0200, David Hildenbrand wrote:
> Let's just warn in case a section is not valid instead of failing to
> remove somewhere in the middle of the process, returning an error
> that
> will be mostly ignored by callers.
> 
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Oscar Salvador <osalvador@suse.de>
> Cc: Michal Hocko <mhocko@suse.com>
> Cc: David Hildenbrand <david@redhat.com>
> Cc: Pavel Tatashin <pasha.tatashin@soleen.com>
> Cc: Qian Cai <cai@lca.pw>
> Cc: Wei Yang <richard.weiyang@gmail.com>
> Cc: Arun KS <arunks@codeaurora.org>
> Cc: Mathieu Malaterre <malat@debian.org>
> Signed-off-by: David Hildenbrand <david@redhat.com>

Just a nit:

I think this could be combined with patch#2.
The only reason to fail in here is 1) !valid_section 2)
!present_section.
As I stated in patch#2, one cannot be without the other, so makes sense
to rip present_section check from unregister_mem_section() as well.
Then, you could combine both changelogs explaining the whole thing, and
why we do not need the present_section check either.

But the change looks good to me:

Reviewed-by: Oscar Salvador <osalvador@suse.de>

> ---
>  mm/memory_hotplug.c | 22 +++++++++-------------
>  1 file changed, 9 insertions(+), 13 deletions(-)
> 
> diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
> index b0cb05748f99..17a60281c36f 100644
> --- a/mm/memory_hotplug.c
> +++ b/mm/memory_hotplug.c
> @@ -517,15 +517,15 @@ static void __remove_zone(struct zone *zone,
> unsigned long start_pfn)
>  	pgdat_resize_unlock(zone->zone_pgdat, &flags);
>  }
>  
> -static int __remove_section(struct zone *zone, struct mem_section
> *ms,
> -		unsigned long map_offset, struct vmem_altmap
> *altmap)
> +static void __remove_section(struct zone *zone, struct mem_section
> *ms,
> +			     unsigned long map_offset,
> +			     struct vmem_altmap *altmap)
>  {
>  	unsigned long start_pfn;
>  	int scn_nr;
> -	int ret = -EINVAL;
>  
> -	if (!valid_section(ms))
> -		return ret;
> +	if (WARN_ON_ONCE(!valid_section(ms)))
> +		return;
>  
>  	unregister_memory_section(ms);
>  
> @@ -534,7 +534,6 @@ static int __remove_section(struct zone *zone,
> struct mem_section *ms,
>  	__remove_zone(zone, start_pfn);
>  
>  	sparse_remove_one_section(zone, ms, map_offset, altmap);
> -	return 0;
>  }
>  
>  /**
> @@ -554,7 +553,7 @@ int __remove_pages(struct zone *zone, unsigned
> long phys_start_pfn,
>  {
>  	unsigned long i;
>  	unsigned long map_offset = 0;
> -	int sections_to_remove, ret = 0;
> +	int sections_to_remove;
>  
>  	/* In the ZONE_DEVICE case device driver owns the memory
> region */
>  	if (is_dev_zone(zone)) {
> @@ -575,16 +574,13 @@ int __remove_pages(struct zone *zone, unsigned
> long phys_start_pfn,
>  		unsigned long pfn = phys_start_pfn +
> i*PAGES_PER_SECTION;
>  
>  		cond_resched();
> -		ret = __remove_section(zone, __pfn_to_section(pfn),
> map_offset,
> -				altmap);
> +		__remove_section(zone, __pfn_to_section(pfn),
> map_offset,
> +				 altmap);
>  		map_offset = 0;
> -		if (ret)
> -			break;
>  	}
>  
>  	set_zone_contiguous(zone);
> -
> -	return ret;
> +	return 0;
>  }
>  #endif /* CONFIG_MEMORY_HOTREMOVE */
>
David Hildenbrand April 24, 2019, 6:54 a.m. UTC | #2
On 17.04.19 15:56, Oscar Salvador wrote:
> On Tue, 2019-04-09 at 12:01 +0200, David Hildenbrand wrote:
>> Let's just warn in case a section is not valid instead of failing to
>> remove somewhere in the middle of the process, returning an error
>> that
>> will be mostly ignored by callers.
>>
>> Cc: Andrew Morton <akpm@linux-foundation.org>
>> Cc: Oscar Salvador <osalvador@suse.de>
>> Cc: Michal Hocko <mhocko@suse.com>
>> Cc: David Hildenbrand <david@redhat.com>
>> Cc: Pavel Tatashin <pasha.tatashin@soleen.com>
>> Cc: Qian Cai <cai@lca.pw>
>> Cc: Wei Yang <richard.weiyang@gmail.com>
>> Cc: Arun KS <arunks@codeaurora.org>
>> Cc: Mathieu Malaterre <malat@debian.org>
>> Signed-off-by: David Hildenbrand <david@redhat.com>
> 
> Just a nit:
> 
> I think this could be combined with patch#2.
> The only reason to fail in here is 1) !valid_section 2)
> !present_section.
> As I stated in patch#2, one cannot be without the other, so makes sense
> to rip present_section check from unregister_mem_section() as well.
> Then, you could combine both changelogs explaining the whole thing, and
> why we do not need the present_section check either.
> 

If I have to resend the whole thing, I might do that. Otherwise we can
drop the present_section() based on your explanation later.

Thanks!

> But the change looks good to me:
> 
> Reviewed-by: Oscar Salvador <osalvador@suse.de>

Patch
diff mbox series

diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index b0cb05748f99..17a60281c36f 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -517,15 +517,15 @@  static void __remove_zone(struct zone *zone, unsigned long start_pfn)
 	pgdat_resize_unlock(zone->zone_pgdat, &flags);
 }
 
-static int __remove_section(struct zone *zone, struct mem_section *ms,
-		unsigned long map_offset, struct vmem_altmap *altmap)
+static void __remove_section(struct zone *zone, struct mem_section *ms,
+			     unsigned long map_offset,
+			     struct vmem_altmap *altmap)
 {
 	unsigned long start_pfn;
 	int scn_nr;
-	int ret = -EINVAL;
 
-	if (!valid_section(ms))
-		return ret;
+	if (WARN_ON_ONCE(!valid_section(ms)))
+		return;
 
 	unregister_memory_section(ms);
 
@@ -534,7 +534,6 @@  static int __remove_section(struct zone *zone, struct mem_section *ms,
 	__remove_zone(zone, start_pfn);
 
 	sparse_remove_one_section(zone, ms, map_offset, altmap);
-	return 0;
 }
 
 /**
@@ -554,7 +553,7 @@  int __remove_pages(struct zone *zone, unsigned long phys_start_pfn,
 {
 	unsigned long i;
 	unsigned long map_offset = 0;
-	int sections_to_remove, ret = 0;
+	int sections_to_remove;
 
 	/* In the ZONE_DEVICE case device driver owns the memory region */
 	if (is_dev_zone(zone)) {
@@ -575,16 +574,13 @@  int __remove_pages(struct zone *zone, unsigned long phys_start_pfn,
 		unsigned long pfn = phys_start_pfn + i*PAGES_PER_SECTION;
 
 		cond_resched();
-		ret = __remove_section(zone, __pfn_to_section(pfn), map_offset,
-				altmap);
+		__remove_section(zone, __pfn_to_section(pfn), map_offset,
+				 altmap);
 		map_offset = 0;
-		if (ret)
-			break;
 	}
 
 	set_zone_contiguous(zone);
-
-	return ret;
+	return 0;
 }
 #endif /* CONFIG_MEMORY_HOTREMOVE */