[PATCH v2] MIPS: mm: Use the bitmap API to allocate bitmaps

Christophe JAILLET posted 1 patch 3 years, 9 months ago
arch/mips/mm/context.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
[PATCH v2] MIPS: mm: Use the bitmap API to allocate bitmaps
Posted by Christophe JAILLET 3 years, 9 months ago
Use bitmap_zalloc() instead of hand-writing them.

It is less verbose and it improves the semantic.

While at it, turn a bitmap_clear() into an equivalent bitmap_zero(). It is
also less verbose.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
---
v1 --> v2: don't speak about bitmap_free() in the log message (Sergey Shtylyov)
---
 arch/mips/mm/context.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/arch/mips/mm/context.c b/arch/mips/mm/context.c
index b25564090939..966f40066f03 100644
--- a/arch/mips/mm/context.c
+++ b/arch/mips/mm/context.c
@@ -67,7 +67,7 @@ static void flush_context(void)
 	int cpu;
 
 	/* Update the list of reserved MMIDs and the MMID bitmap */
-	bitmap_clear(mmid_map, 0, num_mmids);
+	bitmap_zero(mmid_map, num_mmids);
 
 	/* Reserve an MMID for kmap/wired entries */
 	__set_bit(MMID_KERNEL_WIRED, mmid_map);
@@ -277,8 +277,7 @@ static int mmid_init(void)
 	WARN_ON(num_mmids <= num_possible_cpus());
 
 	atomic64_set(&mmid_version, asid_first_version(0));
-	mmid_map = kcalloc(BITS_TO_LONGS(num_mmids), sizeof(*mmid_map),
-			   GFP_KERNEL);
+	mmid_map = bitmap_zalloc(num_mmids, GFP_KERNEL);
 	if (!mmid_map)
 		panic("Failed to allocate bitmap for %u MMIDs\n", num_mmids);
 
-- 
2.34.1
Re: [PATCH v2] MIPS: mm: Use the bitmap API to allocate bitmaps
Posted by Thomas Bogendoerfer 3 years, 9 months ago
On Tue, Jul 05, 2022 at 10:56:51PM +0200, Christophe JAILLET wrote:
> Use bitmap_zalloc() instead of hand-writing them.
> 
> It is less verbose and it improves the semantic.
> 
> While at it, turn a bitmap_clear() into an equivalent bitmap_zero(). It is
> also less verbose.
> 
> Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> ---
> v1 --> v2: don't speak about bitmap_free() in the log message (Sergey Shtylyov)
> ---
>  arch/mips/mm/context.c | 5 ++---
>  1 file changed, 2 insertions(+), 3 deletions(-)
> 
> diff --git a/arch/mips/mm/context.c b/arch/mips/mm/context.c
> index b25564090939..966f40066f03 100644
> --- a/arch/mips/mm/context.c
> +++ b/arch/mips/mm/context.c
> @@ -67,7 +67,7 @@ static void flush_context(void)
>  	int cpu;
>  
>  	/* Update the list of reserved MMIDs and the MMID bitmap */
> -	bitmap_clear(mmid_map, 0, num_mmids);
> +	bitmap_zero(mmid_map, num_mmids);
>  
>  	/* Reserve an MMID for kmap/wired entries */
>  	__set_bit(MMID_KERNEL_WIRED, mmid_map);
> @@ -277,8 +277,7 @@ static int mmid_init(void)
>  	WARN_ON(num_mmids <= num_possible_cpus());
>  
>  	atomic64_set(&mmid_version, asid_first_version(0));
> -	mmid_map = kcalloc(BITS_TO_LONGS(num_mmids), sizeof(*mmid_map),
> -			   GFP_KERNEL);
> +	mmid_map = bitmap_zalloc(num_mmids, GFP_KERNEL);
>  	if (!mmid_map)
>  		panic("Failed to allocate bitmap for %u MMIDs\n", num_mmids);
>  
> -- 
> 2.34.1

applied to mips-next.

Thomas.

-- 
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea.                                                [ RFC1925, 2.3 ]