From: Atish Patra <atishp@rivosinc.com> To: linux-kernel@vger.kernel.org Cc: Atish Patra <atishp@rivosinc.com>, Anup Patel <anup@brainfault.org>, Andrew Jones <ajones@ventanamicro.com>, Atish Patra <atishp@atishpatra.org>, Guo Ren <guoren@kernel.org>, kvm-riscv@lists.infradead.org, kvm@vger.kernel.org, linux-riscv@lists.infradead.org, Mark Rutland <mark.rutland@arm.com>, Palmer Dabbelt <palmer@dabbelt.com>, Paul Walmsley <paul.walmsley@sifive.com>, Sergey Matyukevich <sergey.matyukevich@syntacore.com>, Eric Lin <eric.lin@sifive.com>, Will Deacon <will@kernel.org> Subject: [PATCH v2 09/11] RISC-V: KVM: Implement trap & emulate for hpmcounters Date: Thu, 15 Dec 2022 09:00:44 -0800 [thread overview] Message-ID: <20221215170046.2010255-10-atishp@rivosinc.com> (raw) In-Reply-To: <20221215170046.2010255-1-atishp@rivosinc.com> As the KVM guests only see the virtual PMU counters, all hpmcounter access should trap and KVM emulates the read access on behalf of guests. Signed-off-by: Atish Patra <atishp@rivosinc.com> --- arch/riscv/include/asm/kvm_vcpu_pmu.h | 16 ++++++++++ arch/riscv/kvm/vcpu_insn.c | 4 ++- arch/riscv/kvm/vcpu_pmu.c | 44 ++++++++++++++++++++++++++- 3 files changed, 62 insertions(+), 2 deletions(-) diff --git a/arch/riscv/include/asm/kvm_vcpu_pmu.h b/arch/riscv/include/asm/kvm_vcpu_pmu.h index 6a8c0f7..7a9a8e6 100644 --- a/arch/riscv/include/asm/kvm_vcpu_pmu.h +++ b/arch/riscv/include/asm/kvm_vcpu_pmu.h @@ -43,6 +43,19 @@ struct kvm_pmu { #define vcpu_to_pmu(vcpu) (&(vcpu)->arch.pmu) #define pmu_to_vcpu(pmu) (container_of((pmu), struct kvm_vcpu, arch.pmu)) +#if defined(CONFIG_32BIT) +#define KVM_RISCV_VCPU_HPMCOUNTER_CSR_FUNCS \ +{ .base = CSR_CYCLEH, .count = 31, .func = kvm_riscv_vcpu_pmu_read_hpm }, \ +{ .base = CSR_CYCLE, .count = 31, .func = kvm_riscv_vcpu_pmu_read_hpm }, +#else +#define KVM_RISCV_VCPU_HPMCOUNTER_CSR_FUNCS \ +{ .base = CSR_CYCLE, .count = 31, .func = kvm_riscv_vcpu_pmu_read_hpm }, +#endif + +int kvm_riscv_vcpu_pmu_read_hpm(struct kvm_vcpu *vcpu, unsigned int csr_num, + unsigned long *val, unsigned long new_val, + unsigned long wr_mask); + int kvm_riscv_vcpu_pmu_num_ctrs(struct kvm_vcpu *vcpu, struct kvm_vcpu_sbi_ext_data *edata); int kvm_riscv_vcpu_pmu_ctr_info(struct kvm_vcpu *vcpu, unsigned long cidx, struct kvm_vcpu_sbi_ext_data *edata); @@ -65,6 +78,9 @@ void kvm_riscv_vcpu_pmu_reset(struct kvm_vcpu *vcpu); #else struct kvm_pmu { }; +#define KVM_RISCV_VCPU_HPMCOUNTER_CSR_FUNCS \ +{ .base = 0, .count = 0, .func = NULL }, + static inline int kvm_riscv_vcpu_pmu_init(struct kvm_vcpu *vcpu) { diff --git a/arch/riscv/kvm/vcpu_insn.c b/arch/riscv/kvm/vcpu_insn.c index 1ff2649..f689337 100644 --- a/arch/riscv/kvm/vcpu_insn.c +++ b/arch/riscv/kvm/vcpu_insn.c @@ -213,7 +213,9 @@ struct csr_func { unsigned long wr_mask); }; -static const struct csr_func csr_funcs[] = {}; +static const struct csr_func csr_funcs[] = { + KVM_RISCV_VCPU_HPMCOUNTER_CSR_FUNCS +}; /** * kvm_riscv_vcpu_csr_return -- Handle CSR read/write after user space diff --git a/arch/riscv/kvm/vcpu_pmu.c b/arch/riscv/kvm/vcpu_pmu.c index 0f0748f1..53c4163 100644 --- a/arch/riscv/kvm/vcpu_pmu.c +++ b/arch/riscv/kvm/vcpu_pmu.c @@ -17,6 +17,43 @@ #define kvm_pmu_num_counters(pmu) ((pmu)->num_hw_ctrs + (pmu)->num_fw_ctrs) +static int pmu_ctr_read(struct kvm_vcpu *vcpu, unsigned long cidx, + unsigned long *out_val) +{ + struct kvm_pmu *kvpmu = vcpu_to_pmu(vcpu); + struct kvm_pmc *pmc; + u64 enabled, running; + + pmc = &kvpmu->pmc[cidx]; + if (!pmc->perf_event) + return -EINVAL; + + pmc->counter_val += perf_event_read_value(pmc->perf_event, &enabled, &running); + *out_val = pmc->counter_val; + + return 0; +} + +int kvm_riscv_vcpu_pmu_read_hpm(struct kvm_vcpu *vcpu, unsigned int csr_num, + unsigned long *val, unsigned long new_val, + unsigned long wr_mask) +{ + struct kvm_pmu *kvpmu = vcpu_to_pmu(vcpu); + int cidx, ret = KVM_INSN_CONTINUE_NEXT_SEPC; + + if (!kvpmu || !kvpmu->init_done) + return KVM_INSN_EXIT_TO_USER_SPACE; + + if (wr_mask) + return KVM_INSN_ILLEGAL_TRAP; + cidx = csr_num - CSR_CYCLE; + + if (pmu_ctr_read(vcpu, cidx, val) < 0) + return KVM_INSN_EXIT_TO_USER_SPACE; + + return ret; +} + int kvm_riscv_vcpu_pmu_num_ctrs(struct kvm_vcpu *vcpu, struct kvm_vcpu_sbi_ext_data *edata) { struct kvm_pmu *kvpmu = vcpu_to_pmu(vcpu); @@ -69,7 +106,12 @@ int kvm_riscv_vcpu_pmu_ctr_cfg_match(struct kvm_vcpu *vcpu, unsigned long ctr_ba int kvm_riscv_vcpu_pmu_ctr_read(struct kvm_vcpu *vcpu, unsigned long cidx, struct kvm_vcpu_sbi_ext_data *edata) { - /* TODO */ + int ret; + + ret = pmu_ctr_read(vcpu, cidx, &edata->out_val); + if (ret == -EINVAL) + edata->err_val = SBI_ERR_INVALID_PARAM; + return 0; } -- 2.25.1 _______________________________________________ linux-riscv mailing list linux-riscv@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-riscv
WARNING: multiple messages have this Message-ID (diff)
From: Atish Patra <atishp@rivosinc.com> To: linux-kernel@vger.kernel.org Cc: Atish Patra <atishp@rivosinc.com>, Anup Patel <anup@brainfault.org>, Andrew Jones <ajones@ventanamicro.com>, Atish Patra <atishp@atishpatra.org>, Guo Ren <guoren@kernel.org>, kvm-riscv@lists.infradead.org, kvm@vger.kernel.org, linux-riscv@lists.infradead.org, Mark Rutland <mark.rutland@arm.com>, Palmer Dabbelt <palmer@dabbelt.com>, Paul Walmsley <paul.walmsley@sifive.com>, Sergey Matyukevich <sergey.matyukevich@syntacore.com>, Eric Lin <eric.lin@sifive.com>, Will Deacon <will@kernel.org> Subject: [PATCH v2 09/11] RISC-V: KVM: Implement trap & emulate for hpmcounters Date: Thu, 15 Dec 2022 09:00:44 -0800 [thread overview] Message-ID: <20221215170046.2010255-10-atishp@rivosinc.com> (raw) In-Reply-To: <20221215170046.2010255-1-atishp@rivosinc.com> As the KVM guests only see the virtual PMU counters, all hpmcounter access should trap and KVM emulates the read access on behalf of guests. Signed-off-by: Atish Patra <atishp@rivosinc.com> --- arch/riscv/include/asm/kvm_vcpu_pmu.h | 16 ++++++++++ arch/riscv/kvm/vcpu_insn.c | 4 ++- arch/riscv/kvm/vcpu_pmu.c | 44 ++++++++++++++++++++++++++- 3 files changed, 62 insertions(+), 2 deletions(-) diff --git a/arch/riscv/include/asm/kvm_vcpu_pmu.h b/arch/riscv/include/asm/kvm_vcpu_pmu.h index 6a8c0f7..7a9a8e6 100644 --- a/arch/riscv/include/asm/kvm_vcpu_pmu.h +++ b/arch/riscv/include/asm/kvm_vcpu_pmu.h @@ -43,6 +43,19 @@ struct kvm_pmu { #define vcpu_to_pmu(vcpu) (&(vcpu)->arch.pmu) #define pmu_to_vcpu(pmu) (container_of((pmu), struct kvm_vcpu, arch.pmu)) +#if defined(CONFIG_32BIT) +#define KVM_RISCV_VCPU_HPMCOUNTER_CSR_FUNCS \ +{ .base = CSR_CYCLEH, .count = 31, .func = kvm_riscv_vcpu_pmu_read_hpm }, \ +{ .base = CSR_CYCLE, .count = 31, .func = kvm_riscv_vcpu_pmu_read_hpm }, +#else +#define KVM_RISCV_VCPU_HPMCOUNTER_CSR_FUNCS \ +{ .base = CSR_CYCLE, .count = 31, .func = kvm_riscv_vcpu_pmu_read_hpm }, +#endif + +int kvm_riscv_vcpu_pmu_read_hpm(struct kvm_vcpu *vcpu, unsigned int csr_num, + unsigned long *val, unsigned long new_val, + unsigned long wr_mask); + int kvm_riscv_vcpu_pmu_num_ctrs(struct kvm_vcpu *vcpu, struct kvm_vcpu_sbi_ext_data *edata); int kvm_riscv_vcpu_pmu_ctr_info(struct kvm_vcpu *vcpu, unsigned long cidx, struct kvm_vcpu_sbi_ext_data *edata); @@ -65,6 +78,9 @@ void kvm_riscv_vcpu_pmu_reset(struct kvm_vcpu *vcpu); #else struct kvm_pmu { }; +#define KVM_RISCV_VCPU_HPMCOUNTER_CSR_FUNCS \ +{ .base = 0, .count = 0, .func = NULL }, + static inline int kvm_riscv_vcpu_pmu_init(struct kvm_vcpu *vcpu) { diff --git a/arch/riscv/kvm/vcpu_insn.c b/arch/riscv/kvm/vcpu_insn.c index 1ff2649..f689337 100644 --- a/arch/riscv/kvm/vcpu_insn.c +++ b/arch/riscv/kvm/vcpu_insn.c @@ -213,7 +213,9 @@ struct csr_func { unsigned long wr_mask); }; -static const struct csr_func csr_funcs[] = {}; +static const struct csr_func csr_funcs[] = { + KVM_RISCV_VCPU_HPMCOUNTER_CSR_FUNCS +}; /** * kvm_riscv_vcpu_csr_return -- Handle CSR read/write after user space diff --git a/arch/riscv/kvm/vcpu_pmu.c b/arch/riscv/kvm/vcpu_pmu.c index 0f0748f1..53c4163 100644 --- a/arch/riscv/kvm/vcpu_pmu.c +++ b/arch/riscv/kvm/vcpu_pmu.c @@ -17,6 +17,43 @@ #define kvm_pmu_num_counters(pmu) ((pmu)->num_hw_ctrs + (pmu)->num_fw_ctrs) +static int pmu_ctr_read(struct kvm_vcpu *vcpu, unsigned long cidx, + unsigned long *out_val) +{ + struct kvm_pmu *kvpmu = vcpu_to_pmu(vcpu); + struct kvm_pmc *pmc; + u64 enabled, running; + + pmc = &kvpmu->pmc[cidx]; + if (!pmc->perf_event) + return -EINVAL; + + pmc->counter_val += perf_event_read_value(pmc->perf_event, &enabled, &running); + *out_val = pmc->counter_val; + + return 0; +} + +int kvm_riscv_vcpu_pmu_read_hpm(struct kvm_vcpu *vcpu, unsigned int csr_num, + unsigned long *val, unsigned long new_val, + unsigned long wr_mask) +{ + struct kvm_pmu *kvpmu = vcpu_to_pmu(vcpu); + int cidx, ret = KVM_INSN_CONTINUE_NEXT_SEPC; + + if (!kvpmu || !kvpmu->init_done) + return KVM_INSN_EXIT_TO_USER_SPACE; + + if (wr_mask) + return KVM_INSN_ILLEGAL_TRAP; + cidx = csr_num - CSR_CYCLE; + + if (pmu_ctr_read(vcpu, cidx, val) < 0) + return KVM_INSN_EXIT_TO_USER_SPACE; + + return ret; +} + int kvm_riscv_vcpu_pmu_num_ctrs(struct kvm_vcpu *vcpu, struct kvm_vcpu_sbi_ext_data *edata) { struct kvm_pmu *kvpmu = vcpu_to_pmu(vcpu); @@ -69,7 +106,12 @@ int kvm_riscv_vcpu_pmu_ctr_cfg_match(struct kvm_vcpu *vcpu, unsigned long ctr_ba int kvm_riscv_vcpu_pmu_ctr_read(struct kvm_vcpu *vcpu, unsigned long cidx, struct kvm_vcpu_sbi_ext_data *edata) { - /* TODO */ + int ret; + + ret = pmu_ctr_read(vcpu, cidx, &edata->out_val); + if (ret == -EINVAL) + edata->err_val = SBI_ERR_INVALID_PARAM; + return 0; } -- 2.25.1
next prev parent reply other threads:[~2022-12-15 17:01 UTC|newest] Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top 2022-12-15 17:00 [PATCH v2 00/11] KVM perf support Atish Patra 2022-12-15 17:00 ` Atish Patra 2022-12-15 17:00 ` [PATCH v2 01/11] RISC-V: Define helper functions expose hpm counter width and count Atish Patra 2022-12-15 17:00 ` Atish Patra 2023-01-12 10:06 ` Andrew Jones 2023-01-12 10:06 ` Andrew Jones 2023-01-12 18:18 ` Atish Kumar Patra 2023-01-12 18:18 ` Atish Kumar Patra 2023-01-13 7:22 ` Andrew Jones 2023-01-13 7:22 ` Andrew Jones 2023-01-24 20:41 ` Atish Patra 2023-01-24 20:41 ` Atish Patra 2022-12-15 17:00 ` [PATCH v2 02/11] RISC-V: KVM: Define a probe function for SBI extension data structures Atish Patra 2022-12-15 17:00 ` Atish Patra 2023-01-12 10:21 ` Andrew Jones 2023-01-12 10:21 ` Andrew Jones 2023-01-12 18:19 ` Atish Kumar Patra 2023-01-12 18:19 ` Atish Kumar Patra 2022-12-15 17:00 ` [PATCH v2 03/11] RISC-V: KVM: Return correct code for hsm stop function Atish Patra 2022-12-15 17:00 ` Atish Patra 2023-01-12 10:28 ` Andrew Jones 2023-01-12 10:28 ` Andrew Jones 2023-01-12 18:25 ` Atish Kumar Patra 2023-01-12 18:25 ` Atish Kumar Patra 2023-01-13 7:25 ` Andrew Jones 2023-01-13 7:25 ` Andrew Jones 2022-12-15 17:00 ` [PATCH v2 04/11] RISC-V: KVM: Modify SBI extension handler to return SBI error code Atish Patra 2022-12-15 17:00 ` Atish Patra 2023-01-12 11:04 ` Andrew Jones 2023-01-12 11:04 ` Andrew Jones 2023-01-12 18:47 ` Atish Kumar Patra 2023-01-12 18:47 ` Atish Kumar Patra 2023-01-13 7:42 ` Andrew Jones 2023-01-13 7:42 ` Andrew Jones 2022-12-15 17:00 ` [PATCH v2 05/11] RISC-V: KVM: Improve privilege mode filtering for perf Atish Patra 2022-12-15 17:00 ` Atish Patra 2022-12-15 20:17 ` Conor Dooley 2022-12-15 20:17 ` Conor Dooley 2022-12-15 21:10 ` Atish Kumar Patra 2022-12-15 21:10 ` Atish Kumar Patra 2022-12-15 17:00 ` [PATCH v2 06/11] RISC-V: KVM: Add skeleton support " Atish Patra 2022-12-15 17:00 ` Atish Patra 2023-01-12 15:10 ` Andrew Jones 2023-01-12 15:10 ` Andrew Jones 2023-01-12 18:09 ` Atish Kumar Patra 2023-01-12 18:09 ` Atish Kumar Patra 2022-12-15 17:00 ` [PATCH v2 07/11] RISC-V: KVM: Add SBI PMU extension support Atish Patra 2022-12-15 17:00 ` Atish Patra 2023-01-12 15:29 ` Andrew Jones 2023-01-12 15:29 ` Andrew Jones 2023-01-12 18:04 ` Atish Kumar Patra 2023-01-12 18:04 ` Atish Kumar Patra 2022-12-15 17:00 ` [PATCH v2 08/11] RISC-V: KVM: Disable all hpmcounter access for VS/VU mode Atish Patra 2022-12-15 17:00 ` Atish Patra 2023-01-12 15:47 ` Andrew Jones 2023-01-12 15:47 ` Andrew Jones 2022-12-15 17:00 ` Atish Patra [this message] 2022-12-15 17:00 ` [PATCH v2 09/11] RISC-V: KVM: Implement trap & emulate for hpmcounters Atish Patra 2023-01-13 11:47 ` Andrew Jones 2023-01-13 11:47 ` Andrew Jones 2022-12-15 17:00 ` [PATCH v2 10/11] RISC-V: KVM: Implement perf support without sampling Atish Patra 2022-12-15 17:00 ` Atish Patra 2023-01-13 11:45 ` Andrew Jones 2023-01-13 11:45 ` Andrew Jones 2023-01-23 7:23 ` Andrew Jones 2023-01-23 7:23 ` Andrew Jones 2023-01-26 0:50 ` Atish Patra 2023-01-26 0:50 ` Atish Patra 2022-12-15 17:00 ` [PATCH v2 11/11] RISC-V: KVM: Implement firmware events Atish Patra 2022-12-15 17:00 ` Atish Patra 2023-01-13 12:08 ` Andrew Jones 2023-01-13 12:08 ` Andrew Jones 2023-01-26 3:08 ` Atish Patra 2023-01-26 3:08 ` Atish Patra
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20221215170046.2010255-10-atishp@rivosinc.com \ --to=atishp@rivosinc.com \ --cc=ajones@ventanamicro.com \ --cc=anup@brainfault.org \ --cc=atishp@atishpatra.org \ --cc=eric.lin@sifive.com \ --cc=guoren@kernel.org \ --cc=kvm-riscv@lists.infradead.org \ --cc=kvm@vger.kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-riscv@lists.infradead.org \ --cc=mark.rutland@arm.com \ --cc=palmer@dabbelt.com \ --cc=paul.walmsley@sifive.com \ --cc=sergey.matyukevich@syntacore.com \ --cc=will@kernel.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.