mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* [PATCH v2] genalloc: Make the avail variable an atomic_long_t
@ 2017-10-26 16:04 sbates
  2017-10-26 16:22 ` Mathieu Desnoyers
  2017-10-27 20:25 ` Daniel Mentz
  0 siblings, 2 replies; 4+ messages in thread
From: sbates @ 2017-10-26 16:04 UTC (permalink / raw)
  To: corbet, sbates, mathieu.desnoyers, danielmentz, akpm,
	will.deacon, linux-kernel, logang

From: Stephen Bates <sbates@raithlin.com>

If the amount of resources allocated to a gen_pool exceeds 2^32 then
the avail atomic overflows and this causes problems when clients try
and borrow resources from the pool. This is only expected to be an
issue on 64 bit systems.

Add the <linux/atomic.h> header to pull in atomic_long* operations. So
that 32 bit systems continue to use atomic32_t but 64 bit systems can
use atomic64_t.

Changes since v1:
  Change atomic64 to atomic_long as per Mathieu
  Added a Reviewed-by tag from Logan

Signed-off-by: Stephen Bates <sbates@raithlin.com>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
---
 include/linux/genalloc.h |  3 ++-
 lib/genalloc.c           | 10 +++++-----
 2 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/include/linux/genalloc.h b/include/linux/genalloc.h
index 6dfec4d..872f930 100644
--- a/include/linux/genalloc.h
+++ b/include/linux/genalloc.h
@@ -32,6 +32,7 @@
 
 #include <linux/types.h>
 #include <linux/spinlock_types.h>
+#include <linux/atomic.h>
 
 struct device;
 struct device_node;
@@ -71,7 +72,7 @@ struct gen_pool {
  */
 struct gen_pool_chunk {
 	struct list_head next_chunk;	/* next chunk in pool */
-	atomic_t avail;
+	atomic_long_t avail;
 	phys_addr_t phys_addr;		/* physical starting address of memory chunk */
 	unsigned long start_addr;	/* start address of memory chunk */
 	unsigned long end_addr;		/* end address of memory chunk (inclusive) */
diff --git a/lib/genalloc.c b/lib/genalloc.c
index 144fe6b..ca06adc 100644
--- a/lib/genalloc.c
+++ b/lib/genalloc.c
@@ -194,7 +194,7 @@ int gen_pool_add_virt(struct gen_pool *pool, unsigned long virt, phys_addr_t phy
 	chunk->phys_addr = phys;
 	chunk->start_addr = virt;
 	chunk->end_addr = virt + size - 1;
-	atomic_set(&chunk->avail, size);
+	atomic_long_set(&chunk->avail, size);
 
 	spin_lock(&pool->lock);
 	list_add_rcu(&chunk->next_chunk, &pool->chunks);
@@ -304,7 +304,7 @@ unsigned long gen_pool_alloc_algo(struct gen_pool *pool, size_t size,
 	nbits = (size + (1UL << order) - 1) >> order;
 	rcu_read_lock();
 	list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk) {
-		if (size > atomic_read(&chunk->avail))
+		if (size > atomic_long_read(&chunk->avail))
 			continue;
 
 		start_bit = 0;
@@ -324,7 +324,7 @@ unsigned long gen_pool_alloc_algo(struct gen_pool *pool, size_t size,
 
 		addr = chunk->start_addr + ((unsigned long)start_bit << order);
 		size = nbits << order;
-		atomic_sub(size, &chunk->avail);
+		atomic_long_sub(size, &chunk->avail);
 		break;
 	}
 	rcu_read_unlock();
@@ -390,7 +390,7 @@ void gen_pool_free(struct gen_pool *pool, unsigned long addr, size_t size)
 			remain = bitmap_clear_ll(chunk->bits, start_bit, nbits);
 			BUG_ON(remain);
 			size = nbits << order;
-			atomic_add(size, &chunk->avail);
+			atomic_long_add(size, &chunk->avail);
 			rcu_read_unlock();
 			return;
 		}
@@ -464,7 +464,7 @@ size_t gen_pool_avail(struct gen_pool *pool)
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk)
-		avail += atomic_read(&chunk->avail);
+		avail += atomic_long_read(&chunk->avail);
 	rcu_read_unlock();
 	return avail;
 }
-- 
2.7.4

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH v2] genalloc: Make the avail variable an atomic_long_t
  2017-10-26 16:04 [PATCH v2] genalloc: Make the avail variable an atomic_long_t sbates
@ 2017-10-26 16:22 ` Mathieu Desnoyers
  2017-10-27 20:25 ` Daniel Mentz
  1 sibling, 0 replies; 4+ messages in thread
From: Mathieu Desnoyers @ 2017-10-26 16:22 UTC (permalink / raw)
  To: Stephen Bates
  Cc: Jonathan Corbet, Daniel Mentz, Andrew Morton, Will Deacon,
	linux-kernel, logang

----- On Oct 26, 2017, at 6:04 PM, Stephen Bates sbates@raithlin.com wrote:

> From: Stephen Bates <sbates@raithlin.com>
> 
> If the amount of resources allocated to a gen_pool exceeds 2^32 then
> the avail atomic overflows and this causes problems when clients try
> and borrow resources from the pool. This is only expected to be an
> issue on 64 bit systems.
> 
> Add the <linux/atomic.h> header to pull in atomic_long* operations. So
> that 32 bit systems continue to use atomic32_t but 64 bit systems can
> use atomic64_t.
> 
> Changes since v1:
>  Change atomic64 to atomic_long as per Mathieu
>  Added a Reviewed-by tag from Logan
> 
> Signed-off-by: Stephen Bates <sbates@raithlin.com>
> Reviewed-by: Logan Gunthorpe <logang@deltatee.com>

Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>

> ---
> include/linux/genalloc.h |  3 ++-
> lib/genalloc.c           | 10 +++++-----
> 2 files changed, 7 insertions(+), 6 deletions(-)
> 
> diff --git a/include/linux/genalloc.h b/include/linux/genalloc.h
> index 6dfec4d..872f930 100644
> --- a/include/linux/genalloc.h
> +++ b/include/linux/genalloc.h
> @@ -32,6 +32,7 @@
> 
> #include <linux/types.h>
> #include <linux/spinlock_types.h>
> +#include <linux/atomic.h>
> 
> struct device;
> struct device_node;
> @@ -71,7 +72,7 @@ struct gen_pool {
>  */
> struct gen_pool_chunk {
> 	struct list_head next_chunk;	/* next chunk in pool */
> -	atomic_t avail;
> +	atomic_long_t avail;
> 	phys_addr_t phys_addr;		/* physical starting address of memory chunk */
> 	unsigned long start_addr;	/* start address of memory chunk */
> 	unsigned long end_addr;		/* end address of memory chunk (inclusive) */
> diff --git a/lib/genalloc.c b/lib/genalloc.c
> index 144fe6b..ca06adc 100644
> --- a/lib/genalloc.c
> +++ b/lib/genalloc.c
> @@ -194,7 +194,7 @@ int gen_pool_add_virt(struct gen_pool *pool, unsigned long
> virt, phys_addr_t phy
> 	chunk->phys_addr = phys;
> 	chunk->start_addr = virt;
> 	chunk->end_addr = virt + size - 1;
> -	atomic_set(&chunk->avail, size);
> +	atomic_long_set(&chunk->avail, size);
> 
> 	spin_lock(&pool->lock);
> 	list_add_rcu(&chunk->next_chunk, &pool->chunks);
> @@ -304,7 +304,7 @@ unsigned long gen_pool_alloc_algo(struct gen_pool *pool,
> size_t size,
> 	nbits = (size + (1UL << order) - 1) >> order;
> 	rcu_read_lock();
> 	list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk) {
> -		if (size > atomic_read(&chunk->avail))
> +		if (size > atomic_long_read(&chunk->avail))
> 			continue;
> 
> 		start_bit = 0;
> @@ -324,7 +324,7 @@ unsigned long gen_pool_alloc_algo(struct gen_pool *pool,
> size_t size,
> 
> 		addr = chunk->start_addr + ((unsigned long)start_bit << order);
> 		size = nbits << order;
> -		atomic_sub(size, &chunk->avail);
> +		atomic_long_sub(size, &chunk->avail);
> 		break;
> 	}
> 	rcu_read_unlock();
> @@ -390,7 +390,7 @@ void gen_pool_free(struct gen_pool *pool, unsigned long
> addr, size_t size)
> 			remain = bitmap_clear_ll(chunk->bits, start_bit, nbits);
> 			BUG_ON(remain);
> 			size = nbits << order;
> -			atomic_add(size, &chunk->avail);
> +			atomic_long_add(size, &chunk->avail);
> 			rcu_read_unlock();
> 			return;
> 		}
> @@ -464,7 +464,7 @@ size_t gen_pool_avail(struct gen_pool *pool)
> 
> 	rcu_read_lock();
> 	list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk)
> -		avail += atomic_read(&chunk->avail);
> +		avail += atomic_long_read(&chunk->avail);
> 	rcu_read_unlock();
> 	return avail;
> }
> --
> 2.7.4

-- 
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH v2] genalloc: Make the avail variable an atomic_long_t
  2017-10-26 16:04 [PATCH v2] genalloc: Make the avail variable an atomic_long_t sbates
  2017-10-26 16:22 ` Mathieu Desnoyers
@ 2017-10-27 20:25 ` Daniel Mentz
  2017-10-30  0:52   ` Stephen  Bates
  1 sibling, 1 reply; 4+ messages in thread
From: Daniel Mentz @ 2017-10-27 20:25 UTC (permalink / raw)
  To: Stephen Bates
  Cc: Jonathan Corbet, Mathieu Desnoyers, Andrew Morton, Will Deacon,
	lkml, logang

On Thu, Oct 26, 2017 at 9:04 AM,  <sbates@raithlin.com> wrote:
> --- a/include/linux/genalloc.h
> +++ b/include/linux/genalloc.h
> @@ -32,6 +32,7 @@
>
>  #include <linux/types.h>
>  #include <linux/spinlock_types.h>
> +#include <linux/atomic.h>

Do we still need #include <linux/types.h> ? For me, it compiles without it.

Reviewed-by: Daniel Mentz <danielmentz@google.com>

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH v2] genalloc: Make the avail variable an atomic_long_t
  2017-10-27 20:25 ` Daniel Mentz
@ 2017-10-30  0:52   ` Stephen  Bates
  0 siblings, 0 replies; 4+ messages in thread
From: Stephen  Bates @ 2017-10-30  0:52 UTC (permalink / raw)
  To: Daniel Mentz
  Cc: Jonathan Corbet, Mathieu Desnoyers, Andrew Morton, Will Deacon,
	lkml, logang


> Do we still need #include <linux/types.h> ? For me, it compiles without it.

Yes we do. Kbuild reported a failure when I tried omitting it (arm-multi_v7_defconfig).

> Reviewed-by: Daniel Mentz danielmentz@google.com

Thanks for the review

Andrew can you look at picking this up or do you want me to respin with that tags from Daniel and Mathieu applied?

Stephen 

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2017-10-30  0:52 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-10-26 16:04 [PATCH v2] genalloc: Make the avail variable an atomic_long_t sbates
2017-10-26 16:22 ` Mathieu Desnoyers
2017-10-27 20:25 ` Daniel Mentz
2017-10-30  0:52   ` Stephen  Bates

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

Powered by JetHome