Re: [PATCH v2] dma-mapping: benchmark: pretend DMA is transmitting

[Christoph Hellwig]

On Fri, Feb 05, 2021 at 10:52:37AM +0000, Song Bao Hua (Barry Song) wrote:
> I assume there is no need to keep the same size with 5.11-rc, so
> could change the struct to:
> 
> struct map_benchmark {
> 	__u64 avg_map_100ns; /* average map latency in 100ns */
> 	__u64 map_stddev; /* standard deviation of map latency */
> 	__u64 avg_unmap_100ns; /* as above */
> 	__u64 unmap_stddev;
> 	__u32 threads; /* how many threads will do map/unmap in parallel */
> 	__u32 seconds; /* how long the test will last */
> 	__s32 node; /* which numa node this benchmark will run on */
> 	__u32 dma_bits; /* DMA addressing capability */
> 	__u32 dma_dir; /* DMA data direction */
> 	__u8 expansion[84]; /* For future use */
> };
> 
> This won't increase size on 64bit system, but it increases 4bytes
> on 32bits system comparing to 5.11-rc. How do you think about it?

Yes, that sounds good.  Please send me a two patch series with the
first one changing the alignment, and the second adding the delay.
I'll send the first one off to Linus ASAP then.