On 4/10/2021 5:03 PM, David Coe wrote: > Results for AMD Ryzen 4700U running Ubuntu 21.04β kernel 5.11.0-13 > > $ sudo dmesg | grep IOMMU > [ 0.490352] pci 0000:00:00.2: AMD-Vi: IOMMU performance counters supported > [ 0.491985] pci 0000:00:00.2: AMD-Vi: Found IOMMU cap 0x40 > [ 0.493732] perf/amd_iommu: Detected AMD IOMMU #0 (2 banks, 4 counters/bank). > [ 0.793259] AMD-Vi: AMD IOMMUv2 driver by Joerg Roedel <email address hidden> > > .... > > $ sudo perf stat -e 'amd_iommu_0/cmd_processed/, amd_iommu_0/cmd_processed_inv/, amd_iommu_0/ign_rd_wr_mmio_1ff8h/, amd_iommu_0/int_dte_hit/, amd_iommu_0/int_dte_mis/, amd_iommu_0/mem_dte_hit/, amd_iommu_0/mem_dte_mis/, amd_iommu_0/mem_iommu_tlb_pde_hit/, amd_iommu_0/mem_iommu_tlb_pde_mis/, amd_iommu_0/mem_iommu_tlb_pte_hit/, amd_iommu_0/mem_iommu_tlb_pte_mis/, amd_iommu_0/mem_pass_excl/, amd_iommu_0/mem_pass_pretrans/, amd_iommu_0/mem_pass_untrans/, amd_iommu_0/mem_target_abort/, > amd_iommu_0/mem_trans_total/, amd_iommu_0/page_tbl_read_gst/, amd_iommu_0/page_tbl_read_nst/, amd_iommu_0/page_tbl_read_tot/, amd_iommu_0/smi_blk/, amd_iommu_0/smi_recv/, amd_iommu_0/tlb_inv/, amd_iommu_0/vapic_int_guest/, amd_iommu_0/vapic_int_non_guest/' sleep 10 > > Performance counter stats for 'system wide': > > 12 amd_iommu_0/cmd_processed/ (33.28%) > 6 amd_iommu_0/cmd_processed_inv/ (33.32%) > 0 amd_iommu_0/ign_rd_wr_mmio_1ff8h/ (33.36%) > 290 amd_iommu_0/int_dte_hit/ (33.40%) > 20 amd_iommu_0/int_dte_mis/ (33.46%) > 391 amd_iommu_0/mem_dte_hit/ (33.49%) > 3,720 amd_iommu_0/mem_dte_mis/ (33.49%) > 44 amd_iommu_0/mem_iommu_tlb_pde_hit/ (33.46%) > 810 amd_iommu_0/mem_iommu_tlb_pde_mis/ (33.45%) > 35 amd_iommu_0/mem_iommu_tlb_pte_hit/ (33.41%) > 746 amd_iommu_0/mem_iommu_tlb_pte_mis/ (33.37%) > 0 amd_iommu_0/mem_pass_excl/ (33.32%) > 0 amd_iommu_0/mem_pass_pretrans/ (33.28%) > 0 amd_iommu_0/mem_pass_untrans/ (33.28%) > 0 amd_iommu_0/mem_target_abort/ (33.27%) > 715 amd_iommu_0/mem_trans_total/ (33.27%) > 0 amd_iommu_0/page_tbl_read_gst/ (33.28%) > 36 amd_iommu_0/page_tbl_read_nst/ (33.27%) > 36 amd_iommu_0/page_tbl_read_tot/ (33.27%) > 0 amd_iommu_0/smi_blk/ (33.28%) > 0 amd_iommu_0/smi_recv/ (33.26%) > 0 amd_iommu_0/tlb_inv/ (33.23%) > 0 amd_iommu_0/vapic_int_guest/ (33.24%) > 366 amd_iommu_0/vapic_int_non_guest/ (33.27%) > > The immediately obvious difference is the with the enormous count seen on mem_dte_mis on the older Ryzen 2400G. Will do some RTFM but anyone with comments and insight? > > 841,689,151,202,939 amd_iommu_0/mem_dte_mis/ (33.44%) > > Otherwise, all seems to running smoothly (especially for a distribution still in β). Bravo and many thanks all! > That doesn't look correct. Lemme do some more investigation also.
Thanks, Suravee
On 4/10/2021 5:03 PM, David Coe wrote: 0/cmd_processed /, amd_iommu_ 0/cmd_processed _inv/, amd_iommu_ 0/ign_rd_ wr_mmio_ 1ff8h/, amd_iommu_ 0/int_dte_ hit/, amd_iommu_ 0/int_dte_ mis/, amd_iommu_ 0/mem_dte_ hit/, amd_iommu_ 0/mem_dte_ mis/, amd_iommu_ 0/mem_iommu_ tlb_pde_ hit/, amd_iommu_ 0/mem_iommu_ tlb_pde_ mis/, amd_iommu_ 0/mem_iommu_ tlb_pte_ hit/, amd_iommu_ 0/mem_iommu_ tlb_pte_ mis/, amd_iommu_ 0/mem_pass_ excl/, amd_iommu_ 0/mem_pass_ pretrans/ , amd_iommu_ 0/mem_pass_ untrans/ , amd_iommu_ 0/mem_target_ abort/, 0/mem_trans_ total/, amd_iommu_ 0/page_ tbl_read_ gst/, amd_iommu_ 0/page_ tbl_read_ nst/, amd_iommu_ 0/page_ tbl_read_ tot/, amd_iommu_ 0/smi_blk/ , amd_iommu_ 0/smi_recv/ , amd_iommu_ 0/tlb_inv/ , amd_iommu_ 0/vapic_ int_guest/ , amd_iommu_ 0/vapic_ int_non_ guest/' sleep 10 0/cmd_processed / (33.28%) 0/cmd_processed _inv/ (33.32%) 0/ign_rd_ wr_mmio_ 1ff8h/ (33.36%) 0/int_dte_ hit/ (33.40%) 0/int_dte_ mis/ (33.46%) 0/mem_dte_ hit/ (33.49%) 0/mem_dte_ mis/ (33.49%) 0/mem_iommu_ tlb_pde_ hit/ (33.46%) 0/mem_iommu_ tlb_pde_ mis/ (33.45%) 0/mem_iommu_ tlb_pte_ hit/ (33.41%) 0/mem_iommu_ tlb_pte_ mis/ (33.37%) 0/mem_pass_ excl/ (33.32%) 0/mem_pass_ pretrans/ (33.28%) 0/mem_pass_ untrans/ (33.28%) 0/mem_target_ abort/ (33.27%) 0/mem_trans_ total/ (33.27%) 0/page_ tbl_read_ gst/ (33.28%) 0/page_ tbl_read_ nst/ (33.27%) 0/page_ tbl_read_ tot/ (33.27%) 0/smi_blk/ (33.28%) 0/smi_recv/ (33.26%) 0/tlb_inv/ (33.23%) 0/vapic_ int_guest/ (33.24%) 0/vapic_ int_non_ guest/ (33.27%) 0/mem_dte_ mis/ (33.44%)
> Results for AMD Ryzen 4700U running Ubuntu 21.04β kernel 5.11.0-13
>
> $ sudo dmesg | grep IOMMU
> [ 0.490352] pci 0000:00:00.2: AMD-Vi: IOMMU performance counters supported
> [ 0.491985] pci 0000:00:00.2: AMD-Vi: Found IOMMU cap 0x40
> [ 0.493732] perf/amd_iommu: Detected AMD IOMMU #0 (2 banks, 4 counters/bank).
> [ 0.793259] AMD-Vi: AMD IOMMUv2 driver by Joerg Roedel <email address hidden>
>
> ....
>
> $ sudo perf stat -e 'amd_iommu_
> amd_iommu_
>
> Performance counter stats for 'system wide':
>
> 12 amd_iommu_
> 6 amd_iommu_
> 0 amd_iommu_
> 290 amd_iommu_
> 20 amd_iommu_
> 391 amd_iommu_
> 3,720 amd_iommu_
> 44 amd_iommu_
> 810 amd_iommu_
> 35 amd_iommu_
> 746 amd_iommu_
> 0 amd_iommu_
> 0 amd_iommu_
> 0 amd_iommu_
> 0 amd_iommu_
> 715 amd_iommu_
> 0 amd_iommu_
> 36 amd_iommu_
> 36 amd_iommu_
> 0 amd_iommu_
> 0 amd_iommu_
> 0 amd_iommu_
> 0 amd_iommu_
> 366 amd_iommu_
>
> The immediately obvious difference is the with the enormous count seen on mem_dte_mis on the older Ryzen 2400G. Will do some RTFM but anyone with comments and insight?
>
> 841,689,151,202,939 amd_iommu_
>
> Otherwise, all seems to running smoothly (especially for a distribution still in β). Bravo and many thanks all!
>
That doesn't look correct. Lemme do some more investigation also.
Thanks,
Suravee