Support Andes PMU extension
v2 cover Support Andes PMU extension
Hi All,
This patch series introduces the Andes PMU extension, which serves the same purpose as Sscofpmf. In this version we use FDT-based probing and the CONFIG_ANDES_CUSTOM_PMU to enable perf sampling and filtering support.
Its non-standard local interrupt is assigned to bit 18 in the custom S-mode local interrupt pending CSR (slip), while the interrupt cause is (256 + 18).
The feature needs the PMU device callbacks in OpenSBI.
The OpenSBI and Linux patches can be found on Andes Technology GitHub
- https://github.com/andestech/opensbi/commits/andes-pmu-support-v2
- https://github.com/andestech/linux/commits/andes-pmu-support-v2
The PMU device tree node used on AX45MP:
- https://github.com/andestech/opensbi/blob/andes-pmu-support-v2/docs/pmu_support.md#example-3
Tested hardware:
- ASUS Tinker-V (RZ/Five, AX45MP single core)
- Andes AE350 (AX45MP quad core)
Locus Wei-Han Chen (1):
riscv: andes: Support symbolic FW and HW raw events
Yu Chien Peter Lin (9):
riscv: errata: Rename defines for Andes irqchip/riscv-intc: Allow large non-standard hwirq number irqchip/riscv-intc: Introduce Andes IRQ chip
riscv: dts: renesas: r9a07g043f: Update compatible string to use Andes
INTC
dt-bindings: riscv: Add andestech,cpu-intc to interrupt-controller
perf: RISC-V: Eliminate redundant IRQ enable/disable operations
perf: RISC-V: Move T-Head PMU to CPU feature alternative framework
perf: RISC-V: Introduce Andes PMU for perf event sampling
riscv: dts: renesas: Add Andes PMU extension
.../devicetree/bindings/riscv/cpus.yaml | 4 +- arch/riscv/Kconfig.errata | 13 -- arch/riscv/boot/dts/renesas/r9a07g043f.dtsi | 4 +- arch/riscv/errata/andes/errata.c | 10 +- arch/riscv/errata/thead/errata.c | 19 --- arch/riscv/include/asm/errata_list.h | 19 +-- arch/riscv/include/asm/hwcap.h | 2 + arch/riscv/include/asm/vendorid_list.h | 2 +- arch/riscv/kernel/alternative.c | 2 +- arch/riscv/kernel/cpufeature.c | 2 + drivers/irqchip/irq-riscv-intc.c | 61 +++++++-- drivers/perf/Kconfig | 27 ++++ drivers/perf/riscv_pmu_sbi.c | 49 ++++++- include/linux/irqchip/irq-riscv-intc.h | 12 ++ .../arch/riscv/andes/ax45/firmware.json | 68 ++++++++++ .../arch/riscv/andes/ax45/instructions.json | 127 ++++++++++++++++++ .../arch/riscv/andes/ax45/memory.json | 57 ++++++++ .../arch/riscv/andes/ax45/microarch.json | 77 +++++++++++ tools/perf/pmu-events/arch/riscv/mapfile.csv | 1 + 19 files changed, 481 insertions(+), 75 deletions(-) create mode 100644 include/linux/irqchip/irq-riscv-intc.h create mode 100644 tools/perf/pmu-events/arch/riscv/andes/ax45/firmware.json create mode 100644 tools/perf/pmu-events/arch/riscv/andes/ax45/instructions.json create mode 100644 tools/perf/pmu-events/arch/riscv/andes/ax45/memory.json create mode 100644 tools/perf/pmu-events/arch/riscv/andes/ax45/microarch.json
Signature · 7 lines
2.34.1 _______________________________________________ linux-riscv mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-riscv
Quoted 10 lines · Yu Chien Peter Lin
Hi All,
This patch series introduces the Andes PMU extension, which serves the same purpose as Sscofpmf. In this version we use FDT-based probing and the
CONFIG_ANDES_CUSTOM_PMUto enable perf sampling and filtering support.Its non-standard local interrupt is assigned to bit 18 in the custom S-mode local interrupt pending CSR (slip), while the interrupt cause is (256 + 18).
Chief, you gotta thread your patchsets. It seems like every mail in this series has been sent individually, which is going to break tools like b4 and the automated testing on patchwork.
Cheers,
Conor.
_______________________________________________
linux-riscv mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-riscv
Quoted 8 lines · Conor Dooley
Quoted 7 lines · Yu Chien Peter Lin
Hi All,
This patch series introduces the Andes PMU extension, which serves the same purpose as Sscofpmf. In this version we use FDT-based probing and the
CONFIG_ANDES_CUSTOM_PMUto enable perf sampling and filtering support.
Quoted 7 lines
Quoted 3 lines
Its non-standard local interrupt is assigned to bit 18 in the custom S-mode local interrupt pending CSR (slip), while the interrupt cause is (256 + 18).
Chief, you gotta thread your patchsets. It seems like every mail in this series has been sent individually, which is going to break tools like b4 and the automated testing on patchwork.
Hi Conor,
Sure, thanks for taking care of this!
Regards,
Peter Lin
Quoted 2 lines
Cheers,
Conor.
_______________________________________________
linux-riscv mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-riscv