linux/kernel/dma
Chao Gao 9e02977bfa dma-direct: avoid redundant memory sync for swiotlb
When we looked into FIO performance with swiotlb enabled in VM, we found
swiotlb_bounce() is always called one more time than expected for each DMA
read request.

It turns out that the bounce buffer is copied to original DMA buffer twice
after the completion of a DMA request (one is done by in
dma_direct_sync_single_for_cpu(), the other by swiotlb_tbl_unmap_single()).
But the content in bounce buffer actually doesn't change between the two
rounds of copy. So, one round of copy is redundant.

Pass DMA_ATTR_SKIP_CPU_SYNC flag to swiotlb_tbl_unmap_single() to
skip the memory copy in it.

This fix increases FIO 64KB sequential read throughput in a guest with
swiotlb=force by 5.6%.

Fixes: 55897af630 ("dma-direct: merge swiotlb_dma_ops into the dma_direct code")
Reported-by: Wang Zhaoyang1 <zhaoyang1.wang@intel.com>
Reported-by: Gao Liang <liang.gao@intel.com>
Signed-off-by: Chao Gao <chao.gao@intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
2022-04-14 06:30:39 +02:00
..
coherent.c dma-mapping: use 'bitmap_zalloc()' when applicable 2021-10-27 08:20:09 +02:00
contiguous.c cma: factor out minimum alignment requirement 2022-03-22 15:57:05 -07:00
debug.c dma-debug: fix return value of __setup handlers 2022-03-03 14:01:45 +03:00
debug.h dma-debug: teach add_dma_entry() about DMA_ATTR_SKIP_CPU_SYNC 2021-10-18 12:46:45 +02:00
direct.c dma-mapping: move pgprot_decrypted out of dma_pgprot 2022-04-01 06:46:51 +02:00
direct.h dma-direct: avoid redundant memory sync for swiotlb 2022-04-14 06:30:39 +02:00
dummy.c dma-mapping: return error code from dma_dummy_map_sg() 2021-08-09 17:13:06 +02:00
Kconfig dma-mapping: remove CONFIG_DMA_REMAP 2022-03-03 14:00:57 +03:00
Makefile dma-mapping: remove CONFIG_DMA_REMAP 2022-03-03 14:00:57 +03:00
map_benchmark.c dma-mapping: benchmark: extract a common header file for map_benchmark definition 2022-03-10 07:41:14 +01:00
mapping.c dma-mapping: move pgprot_decrypted out of dma_pgprot 2022-04-01 06:46:51 +02:00
ops_helpers.c dma-mapping: handle vmalloc addresses in dma_common_{mmap,get_sgtable} 2021-07-16 11:30:26 +02:00
pool.c dma/pool: create dma atomic pool only if dma zone has managed pages 2022-01-15 16:30:29 +02:00
remap.c kernel/dma: remove unnecessary unmap_kernel_range 2021-04-30 11:20:40 -07:00
swiotlb.c dma-mapping updates for Linux 5.18 2022-03-29 08:50:14 -07:00