When I execute command ‘$ nvprof --query-events’, among the events, I see the following:
l2_subp0_read_tex_sector_queries: Number of read requests from Texture cache to slice 0 of L2 cache. This increments by 1 for each 32-byte access.
l2_subp1_read_tex_sector_queries: Number of read requests from Texture cache to slice 1 of L2 cache. This increments by 1 for each 32-byte access.
So what are the slice 0/1 of L2 cache? How do they improve/effect performance?
Also, can someone clarify that ‘system memory’ refers to GPU DRAMs, correct?