| Message ID | 20250103082814.3850096-1-0x1207@gmail.com (mailing list archive) |
|---|---|
| State | Changes Requested |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | [net-next,v2] page_pool: check for dma_sync_size earlier | expand |
On Fri, 3 Jan 2025 16:28:14 +0800 Furong Xu wrote: > This is a micro optimization, about 0.6% overall performance improvement > has been observed on a Cortex-A53 platform. You need to say more about how you measured this improvement. Is it PPS improvement in a traffic test? Measured using a micro-benchmark? Based on first order metric or by comparing perf output?
diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 9733206d6406..9bb2d2300d0b 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -458,7 +458,7 @@ page_pool_dma_sync_for_device(const struct page_pool *pool, netmem_ref netmem, u32 dma_sync_size) { - if (pool->dma_sync && dma_dev_need_sync(pool->p.dev)) + if (pool->dma_sync && dma_dev_need_sync(pool->p.dev) && dma_sync_size) __page_pool_dma_sync_for_device(pool, netmem, dma_sync_size); }
Setting dma_sync_size to 0 is not illegal, fec_main.c and ravb_main.c already did. We can save a couple of function calls if check for dma_sync_size earlier. This is a micro optimization, about 0.6% overall performance improvement has been observed on a Cortex-A53 platform. Signed-off-by: Furong Xu <0x1207@gmail.com> --- V1 -> V2: Add measurement data about performance improvement in commit message V1: https://lore.kernel.org/r/20241010114019.1734573-1-0x1207@gmail.com --- net/core/page_pool.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)