Re: [PATCH v3] dma-buf: system_heap: Set allocation orders for larger page sizes

[Pankaj Patil <pankaj.patil@oss.qualcomm.com>]

On 4/18/2025 1:49 AM, Juan Yescas wrote:
> This change sets the allocation orders for the different page sizes
> (4k, 16k, 64k) based on PAGE_SHIFT. Before this change, the orders
> for large page sizes were calculated incorrectly, this caused system
> heap to allocate from 2% to 4% more memory on 16KiB page size kernels.
> 
> This change was tested on 4k/16k page size kernels.
> 
> Signed-off-by: Juan Yescas <jyescas@google.com>
> Acked-by: John Stultz <jstultz@google.com>
> Reviewed-by: T.J. Mercier <tjmercier@google.com>
> ---
> 
> Changes in v2:
>   - Add John's Acked-by tag.
>   - Add TJ's Reviewed-by tag
>   - Use dma-buf: system_heap: in the subject since this is specific to the
>     system heap, as per TJ.
>   - Remove extra space in if statement.
> 
> Changes in v3:
>   - Remove defines for the orders as per Christian.
>   - Update the order array for 64k page sizes as per TJ
> 
>  drivers/dma-buf/heaps/system_heap.c | 21 ++++++++++++++++++---
>  1 file changed, 18 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/dma-buf/heaps/system_heap.c b/drivers/dma-buf/heaps/system_heap.c
> index 26d5dc89ea16..8c10c0f2b473 100644
> --- a/drivers/dma-buf/heaps/system_heap.c
> +++ b/drivers/dma-buf/heaps/system_heap.c
> @@ -18,6 +18,7 @@
>  #include <linux/mm.h>
>  #include <linux/module.h>
>  #include <linux/scatterlist.h>
> +#include <linux/sizes.h>
>  #include <linux/slab.h>
>  #include <linux/vmalloc.h>
>  
> @@ -44,14 +45,28 @@ struct dma_heap_attachment {
>  #define HIGH_ORDER_GFP  (((GFP_HIGHUSER | __GFP_ZERO | __GFP_NOWARN \
>  				| __GFP_NORETRY) & ~__GFP_RECLAIM) \
>  				| __GFP_COMP)
> -static gfp_t order_flags[] = {HIGH_ORDER_GFP, HIGH_ORDER_GFP, LOW_ORDER_GFP};
>  /*
>   * The selection of the orders used for allocation (1MB, 64K, 4K) is designed
>   * to match with the sizes often found in IOMMUs. Using order 4 pages instead
>   * of order 0 pages can significantly improve the performance of many IOMMUs
>   * by reducing TLB pressure and time spent updating page tables.
> + *
> + * Note: When the order is 0, the minimum allocation is PAGE_SIZE. The possible
> + * page sizes for ARM devices could be 4K, 16K and 64K.
> + */
> +#if (PAGE_SIZE == SZ_64K)
> +static gfp_t order_flags[] = {HIGH_ORDER_GFP, LOW_ORDER_GFP};
> +/* 1MiB allocs are calculated with (2 ^ (20 - PAGE_SHIFT)) * PAGE_SIZE = 1MiB */
> +static const unsigned int orders[] = {20 - PAGE_SHIFT, 0};
> +#else
> +static gfp_t order_flags[] = {HIGH_ORDER_GFP, HIGH_ORDER_GFP, LOW_ORDER_GFP};
> +/*
> + * 1MiB allocs are calculated with (2 ^ (20 - PAGE_SHIFT)) * PAGE_SIZE = 1MiB
> + * 64KiB allocs are calculated with (2 ^ (16 - PAGE_SHIFT)) * PAGE_SIZE = 64KiB
>   */
> -static const unsigned int orders[] = {8, 4, 0};
> +static const unsigned int orders[] = {20 - PAGE_SHIFT, 16 - PAGE_SHIFT, 0};
> +#endif
> +
>  #define NUM_ORDERS ARRAY_SIZE(orders)
>  
>  static struct sg_table *dup_sg_table(struct sg_table *table)
> @@ -318,7 +333,7 @@ static struct page *alloc_largest_available(unsigned long size,
>  	int i;
>  
>  	for (i = 0; i < NUM_ORDERS; i++) {
> -		if (size <  (PAGE_SIZE << orders[i]))
> +		if (size < (PAGE_SIZE << orders[i]))
>  			continue;
>  		if (max_order < orders[i])
>  			continue;

Hello Juan,
I've posted a respin of the patch by extending upon your work,
Can you please review?
Aplogies for adding signed-off by, will drop it in v2.

Link -
https://lore.kernel.org/all/20260921-dma_buf_improvement-v1-1-23d82e771dec@oss.qualcomm.com/