aboutsummaryrefslogtreecommitdiff
path: root/target/riscv/pmu.c
AgeCommit message (Collapse)Author
2024-08-06target/riscv: Add asserts for out-of-bound accessAtish Patra
Coverity complained about the possible out-of-bounds access with counter_virt/counter_virt_prev because these two arrays are accessed with privilege mode. However, these two arrays are accessed only when virt is enabled. Thus, the privilege mode can't be M mode. Add the asserts anyways to detect any wrong usage of these arrays in the future. Suggested-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Atish Patra <atishp@rivosinc.com> Fixes: Coverity CID 1558459 Fixes: Coverity CID 1558462 Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20240724-fixes-v1-1-4a64596b0d64@rivosinc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2024-07-18target/riscv: Do not setup pmu timer if OF is disabledAtish Patra
The timer is setup function is invoked in both hpmcounter write and mcountinhibit write path. If the OF bit set, the LCOFI interrupt is disabled. There is no benefitting in setting up the qemu timer until LCOFI is cleared to indicate that interrupts can be fired again. Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Signed-off-by: Atish Patra <atishp@rivosinc.com> Message-ID: <20240711-smcntrpmf_v7-v8-12-b7c38ae7b263@rivosinc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2024-07-18target/riscv: More accurately model priv mode filtering.Rajnesh Kanwal
In case of programmable counters configured to count inst/cycles we often end-up with counter not incrementing at all from kernel's perspective. For example: - Kernel configures hpm3 to count instructions and sets hpmcounter to -10000 and all modes except U mode are inhibited. - In QEMU we configure a timer to expire after ~10000 instructions. - Problem is, it's often the case that kernel might not even schedule Umode task and we hit the timer callback in QEMU. - In the timer callback we inject the interrupt into kernel, kernel runs the handler and reads hpmcounter3 value. - Given QEMU maintains individual counters to count for each privilege mode, and given umode never ran, the umode counter didn't increment and QEMU returns same value as was programmed by the kernel when starting the counter. - Kernel checks for overflow using previous and current value of the counter and reprograms the counter given there wasn't an overflow as per the counter value. (Which itself is a problem. We have QEMU telling kernel that counter3 overflowed but the counter value returned by QEMU doesn't seem to reflect that.). This change makes sure that timer is reprogrammed from the handler if the counter didn't overflow based on the counter value. Second, this change makes sure that whenever the counter is read, it's value is updated to reflect the latest count. Signed-off-by: Rajnesh Kanwal <rkanwal@rivosinc.com> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Message-ID: <20240711-smcntrpmf_v7-v8-11-b7c38ae7b263@rivosinc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2024-07-18target/riscv: Start counters from both mhpmcounter and mcountinhibitRajnesh Kanwal
Currently we start timer counter from write_mhpmcounter path only without checking for mcountinhibit bit. This changes adds mcountinhibit check and also programs the counter from write_mcountinhibit as well. When a counter is stopped using mcountinhibit we simply update the value of the counter based on current host ticks and save it for future reads. We don't need to disable running timer as pmu_timer_trigger_irq will discard the interrupt if the counter has been inhibited. Signed-off-by: Rajnesh Kanwal <rkanwal@rivosinc.com> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Message-ID: <20240711-smcntrpmf_v7-v8-10-b7c38ae7b263@rivosinc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2024-07-18target/riscv: Implement privilege mode filtering for cycle/instretAtish Patra
Privilege mode filtering can also be emulated for cycle/instret by tracking host_ticks/icount during each privilege mode switch. This patch implements that for both cycle/instret and mhpmcounters. The first one requires Smcntrpmf while the other one requires Sscofpmf to be enabled. The cycle/instret are still computed using host ticks when icount is not enabled. Otherwise, they are computed using raw icount which is more accurate in icount mode. Co-Developed-by: Rajnesh Kanwal <rkanwal@rivosinc.com> Signed-off-by: Rajnesh Kanwal <rkanwal@rivosinc.com> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Acked-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Atish Patra <atishp@rivosinc.com> Message-ID: <20240711-smcntrpmf_v7-v8-7-b7c38ae7b263@rivosinc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2023-11-07target/riscv: Add "pmu-mask" property to replace "pmu-num"Rob Bradford
Using a mask instead of the number of PMU devices supports the accurate emulation of platforms that have a discontinuous set of PMU counters. The "pmu-num" property now generates a warning when used by the user on the command line. Rather than storing the value for "pmu-num" convert it directly to the mask if it is specified (overwriting the default "pmu-mask" value) likewise the value is calculated from the mask if the property value is obtained. In the unusual situation that both "pmu-mask" and "pmu-num" are provided then then the order on the command line determines which takes precedence (later overwriting earlier.) Signed-off-by: Rob Bradford <rbradford@rivosinc.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-ID: <20231031154000.18134-5-rbradford@rivosinc.com> [Changes by AF - Fixup ext_zihpm logic after rebase ] Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2023-11-07target/riscv: Use existing PMU counter mask in FDT generationRob Bradford
During the FDT generation use the existing mask containing the enabled counters rather then generating a new one. Using the existing mask will support the use of discontinuous counters. Signed-off-by: Rob Bradford <rbradford@rivosinc.com> Reviewed-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Atish Patra <atishp@rivosinc.com> Message-ID: <20231031154000.18134-4-rbradford@rivosinc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2023-11-07target/riscv: Propagate error from PMU setupRob Bradford
More closely follow the QEMU style by returning an Error and propagating it there is an error relating to the PMU setup. Further simplify the function by removing the num_counters parameter as this is available from the passed in cpu pointer. Signed-off-by: Rob Bradford <rbradford@rivosinc.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com> Reviewed-by: Atish Patra <atishp@rivosinc.com> Message-ID: <20231031154000.18134-2-rbradford@rivosinc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2023-08-31target/riscv/pmu: Restrict 'qemu/log.h' include to sourcePhilippe Mathieu-Daudé
Declarations from "riscv/pmu.h" don't need anything from "qemu/log.h", reduce it's inclusion to the source. Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-Id: <20230828221314.18435-3-philmd@linaro.org>
2023-05-05target/riscv: Fix lines with over 80 charactersWeiwei Li
Fix lines with over 80 characters for both code and comments. Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com> Acked-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Message-Id: <20230405085813.40643-5-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2023-05-05target/riscv: Remove riscv_cpu_virt_enabled()Weiwei Li
Directly use env->virt_enabled instead. Suggested-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com> Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Message-Id: <20230405085813.40643-2-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2023-05-05target/riscv: fix invalid riscv,event-to-mhpmcounters entryConor Dooley
dt-validate complains: > soc: pmu: {'riscv,event-to-mhpmcounters': > [[1, 1, 524281], [2, 2, 524284], [65561, 65561, 524280], > [65563, 65563, 524280], [65569, 65569, 524280], [0, 0, 0], [0, 0]], > pmu: riscv,event-to-mhpmcounters:6: [0, 0] is too short There are bogus 0 entries added at the end, of which one is of insufficient length. This happens because only 15 of fdt_event_ctr_map[]'s 20 elements are populated & qemu_fdt_setprop() is called using the size of the array. Reduce the array to 15 elements to make the error go away. Signed-off-by: Conor Dooley <conor.dooley@microchip.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-Id: <20230404173333.35179-1-conor@kernel.org> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2023-05-05target/riscv: Simplify type conversion for CPURISCVStateWeiwei Li
Use CPURISCVState as argument directly in riscv_cpu_update_mip and riscv_timer_write_timecmp, since type converts from CPURISCVState to RISCVCPU in many caller of them and then back to CPURISCVState in them. Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Message-Id: <20230309071329.45932-4-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2023-05-05target/riscv: Simplify getting RISCVCPU pointer from envWeiwei Li
Use env_archcpu() to get RISCVCPU pointer from env directly. Signed-off-by: Weiwei Li <liweiwei@iscas.ac.cn> Signed-off-by: Junqiang Wang <wangjunqiang@iscas.ac.cn> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230309071329.45932-3-liweiwei@iscas.ac.cn> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2022-09-07hw/riscv: virt: Add PMU DT node to the device treeAtish Patra
Qemu virt machine can support few cache events and cycle/instret counters. It also supports counter overflow for these events. Add a DT node so that OpenSBI/Linux kernel is aware of the virt machine capabilities. There are some dummy nodes added for testing as well. Acked-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Atish Patra <atish.patra@wdc.com> Signed-off-by: Atish Patra <atishp@rivosinc.com> Message-Id: <20220824221701.41932-5-atishp@rivosinc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2022-09-07target/riscv: Add sscofpmf extension supportAtish Patra
The Sscofpmf ('Ss' for Privileged arch and Supervisor-level extensions, and 'cofpmf' for Count OverFlow and Privilege Mode Filtering) extension allows the perf to handle overflow interrupts and filtering support. This patch provides a framework for programmable counters to leverage the extension. As the extension doesn't have any provision for the overflow bit for fixed counters, the fixed events can also be monitoring using programmable counters. The underlying counters for cycle and instruction counters are always running. Thus, a separate timer device is programmed to handle the overflow. Tested-by: Heiko Stuebner <heiko@sntech.de> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Atish Patra <atish.patra@wdc.com> Signed-off-by: Atish Patra <atishp@rivosinc.com> Message-Id: <20220824221701.41932-2-atishp@rivosinc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
2022-07-03target/riscv: Support mcycle/minstret write operationAtish Patra
mcycle/minstret are actually WARL registers and can be written with any given value. With SBI PMU extension, it will be used to store a initial value provided from supervisor OS. The Qemu also need prohibit the counter increment if mcountinhibit is set. Support mcycle/minstret through generic counter infrastructure. Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Atish Patra <atish.patra@wdc.com> Signed-off-by: Atish Patra <atishp@rivosinc.com> Message-Id: <20220620231603.2547260-8-atishp@rivosinc.com> Signed-off-by: Alistair Francis <alistair.francis@wdc.com>