> -----Original Message----- > From: Alexander Graf [mailto:agraf@xxxxxxx] > Sent: Thursday, May 02, 2013 4:35 PM > To: Bhushan Bharat-R65777 > Cc: kvm-ppc@xxxxxxxxxxxxxxx; kvm@xxxxxxxxxxxxxxx; Wood Scott-B07421 > Subject: Re: [PATCH 7/7 v3] KVM: PPC: Add userspace debug stub support > > > On 02.05.2013, at 11:46, Bhushan Bharat-R65777 wrote: > > > > > > >> -----Original Message----- > >> From: Alexander Graf [mailto:agraf@xxxxxxx] > >> Sent: Friday, April 26, 2013 4:46 PM > >> To: Bhushan Bharat-R65777 > >> Cc: kvm-ppc@xxxxxxxxxxxxxxx; kvm@xxxxxxxxxxxxxxx; Wood Scott-B07421; > >> Bhushan > >> Bharat-R65777 > >> Subject: Re: [PATCH 7/7 v3] KVM: PPC: Add userspace debug stub > >> support > >> > >> > >> On 08.04.2013, at 12:32, Bharat Bhushan wrote: > >> > >>> From: Bharat Bhushan <bharat.bhushan@xxxxxxxxxxxxx> > >>> > >>> This patch adds the debug stub support on booke/bookehv. > >>> Now QEMU debug stub can use hw breakpoint, watchpoint and software > >>> breakpoint to debug guest. > >>> > >>> Debug registers are saved/restored on vcpu_put()/vcpu_get(). > >>> Also the debug registers are saved restored only if guest is using > >>> debug resources. > >>> > >>> Currently we do not support debug resource emulation to guest, so > >>> always exit to user space irrespective of user space is expecting > >>> the debug exception or not. This is unexpected event and let us > >>> leave the action on user space. This is similar to what it was > >>> before, only thing is that now we have proper exit state available to user > space. > >>> > >>> Signed-off-by: Bharat Bhushan <bharat.bhushan@xxxxxxxxxxxxx> > >>> --- > >>> arch/powerpc/include/asm/kvm_host.h | 8 + > >>> arch/powerpc/include/uapi/asm/kvm.h | 22 +++- > >>> arch/powerpc/kvm/booke.c | 242 ++++++++++++++++++++++++++++++++- > -- > >>> arch/powerpc/kvm/booke.h | 5 + > >>> 4 files changed, 255 insertions(+), 22 deletions(-) > >>> > >>> diff --git a/arch/powerpc/include/asm/kvm_host.h > >> b/arch/powerpc/include/asm/kvm_host.h > >>> index e34f8fe..b9ad20f 100644 > >>> --- a/arch/powerpc/include/asm/kvm_host.h > >>> +++ b/arch/powerpc/include/asm/kvm_host.h > >>> @@ -505,7 +505,15 @@ struct kvm_vcpu_arch { > >>> u32 mmucfg; > >>> u32 epr; > >>> u32 crit_save; > >>> + > >>> + /* Flag indicating that debug registers are used by guest */ > >>> + bool debug_active; > >>> + /* for save/restore thread->dbcr0 on vcpu run/heavyweight_exit */ > >>> + u32 saved_dbcr0; > >>> + /* guest debug registers*/ > >>> struct kvmppc_booke_debug_reg dbg_reg; > >>> + /* shadow debug registers */ > >>> + struct kvmppc_booke_debug_reg shadow_dbg_reg; > >>> #endif > >>> gpa_t paddr_accessed; > >>> gva_t vaddr_accessed; > >>> diff --git a/arch/powerpc/include/uapi/asm/kvm.h > >> b/arch/powerpc/include/uapi/asm/kvm.h > >>> index c0c38ed..d7ce449 100644 > >>> --- a/arch/powerpc/include/uapi/asm/kvm.h > >>> +++ b/arch/powerpc/include/uapi/asm/kvm.h > >>> @@ -25,6 +25,7 @@ > >>> /* Select powerpc specific features in <linux/kvm.h> */ #define > >>> __KVM_HAVE_SPAPR_TCE #define __KVM_HAVE_PPC_SMT > >>> +#define __KVM_HAVE_GUEST_DEBUG > >>> > >>> struct kvm_regs { > >>> __u64 pc; > >>> @@ -267,7 +268,24 @@ struct kvm_fpu { > >>> __u64 fpr[32]; > >>> }; > >>> > >>> +/* > >>> + * Defines for h/w breakpoint, watchpoint (read, write or both) and > >>> + * software breakpoint. > >>> + * These are used as "type" in KVM_SET_GUEST_DEBUG ioctl and "status" > >>> + * for KVM_DEBUG_EXIT. > >>> + */ > >>> +#define KVMPPC_DEBUG_NONE 0x0 > >>> +#define KVMPPC_DEBUG_BREAKPOINT (1UL << 1) > >>> +#define KVMPPC_DEBUG_WATCH_WRITE (1UL << 2) > >>> +#define KVMPPC_DEBUG_WATCH_READ (1UL << 3) > >>> struct kvm_debug_exit_arch { > >>> + __u64 address; > >>> + /* > >>> + * exiting to userspace because of h/w breakpoint, watchpoint > >>> + * (read, write or both) and software breakpoint. > >>> + */ > >>> + __u32 status; > >>> + __u32 reserved; > >>> }; > >>> > >>> /* for KVM_SET_GUEST_DEBUG */ > >>> @@ -279,10 +297,6 @@ struct kvm_guest_debug_arch { > >>> * Type denotes h/w breakpoint, read watchpoint, write > >>> * watchpoint or watchpoint (both read and write). > >>> */ > >>> -#define KVMPPC_DEBUG_NONE 0x0 > >>> -#define KVMPPC_DEBUG_BREAKPOINT (1UL << 1) > >>> -#define KVMPPC_DEBUG_WATCH_WRITE (1UL << 2) > >>> -#define KVMPPC_DEBUG_WATCH_READ (1UL << 3) > >>> __u32 type; > >>> __u32 reserved; > >>> } bp[16]; > >>> diff --git a/arch/powerpc/kvm/booke.c b/arch/powerpc/kvm/booke.c > >>> index 97ae158..0e93416 100644 > >>> --- a/arch/powerpc/kvm/booke.c > >>> +++ b/arch/powerpc/kvm/booke.c > >>> @@ -133,6 +133,29 @@ static void kvmppc_vcpu_sync_fpu(struct > >>> kvm_vcpu *vcpu) #endif } > >>> > >>> +static void kvmppc_vcpu_sync_debug(struct kvm_vcpu *vcpu) { > >>> + /* Synchronize guest's desire to get debug interrupts into shadow > >>> +MSR */ #ifndef CONFIG_KVM_BOOKE_HV > >>> + vcpu->arch.shadow_msr &= ~MSR_DE; > >>> + vcpu->arch.shadow_msr |= vcpu->arch.shared->msr & MSR_DE; #endif > >>> + > >>> + /* Force enable debug interrupts when user space wants to debug */ > >>> + if (vcpu->guest_debug) { > >>> +#ifdef CONFIG_KVM_BOOKE_HV > >>> + /* > >>> + * Since there is no shadow MSR, sync MSR_DE into the guest > >>> + * visible MSR. > >>> + */ > >>> + vcpu->arch.shared->msr |= MSR_DE; #else > >>> + vcpu->arch.shadow_msr |= MSR_DE; > >>> + vcpu->arch.shared->msr &= ~MSR_DE; #endif > >>> + } > >>> +} > >>> + > >>> /* > >>> * Helper function for "full" MSR writes. No need to call this if > >>> only > >>> * EE/CE/ME/DE/RI are changing. > >>> @@ -150,6 +173,7 @@ void kvmppc_set_msr(struct kvm_vcpu *vcpu, u32 new_msr) > >>> kvmppc_mmu_msr_notify(vcpu, old_msr); > >>> kvmppc_vcpu_sync_spe(vcpu); > >>> kvmppc_vcpu_sync_fpu(vcpu); > >>> + kvmppc_vcpu_sync_debug(vcpu); > >>> } > >>> > >>> static void kvmppc_booke_queue_irqprio(struct kvm_vcpu *vcpu, @@ > >>> -646,6 +670,46 @@ int kvmppc_core_check_requests(struct kvm_vcpu *vcpu) > >>> return r; > >>> } > >>> > >>> +static void kvmppc_load_usespace_gebug(void) > >> > >> ... > > > > Please tell What does "..." mean. Does that mean next comment is about the > above function? > > It means "this one is so obvious I don't even have to write anything here". > Usespace? Gebug? Seriously? :) ehhh, kicking myself > > > > >> > >>> +{ > >>> + switch_booke_debug_regs(¤t->thread); > >>> +} > >>> + > >>> +static void kvmppc_booke_vcpu_load_debug_regs(struct kvm_vcpu > >>> +*vcpu) { > >>> + if (!vcpu->arch.debug_active) > >>> + return; > >>> + > >>> + /* Disable all debug events and clead pending debug events */ > >>> + mtspr(SPRN_DBCR0, 0x0); > >>> + kvmppc_clear_dbsr(); > >>> + > >>> + /* > >>> + * Check whether guest still need debug resource, if not then there > >>> + * is no need to restore guest context. > >>> + */ > >>> + if (!vcpu->arch.shadow_dbg_reg.dbcr0) > >>> + return; > >>> + > >>> + /* Load Guest Context */ > >>> + mtspr(SPRN_DBCR1, vcpu->arch.shadow_dbg_reg.dbcr1); > >>> + mtspr(SPRN_DBCR2, vcpu->arch.shadow_dbg_reg.dbcr2); #ifdef > >>> +CONFIG_KVM_E500MC > >>> + mtspr(SPRN_DBCR4, vcpu->arch.shadow_dbg_reg.dbcr4); > >> > >> You need to make sure DBCR4 is 0 when you leave things back to normal > >> user space. Otherwise guest debug can interfere with host debug. > > > > > > ok > > > >> > >>> +#endif > >>> + mtspr(SPRN_IAC1, vcpu->arch.shadow_dbg_reg.iac[0]); > >>> + mtspr(SPRN_IAC2, vcpu->arch.shadow_dbg_reg.iac[1]); > >>> +#if CONFIG_PPC_ADV_DEBUG_IACS > 2 > >>> + mtspr(SPRN_IAC3, vcpu->arch.shadow_dbg_reg.iac[2]); > >>> + mtspr(SPRN_IAC4, vcpu->arch.shadow_dbg_reg.iac[3]); > >>> +#endif > >>> + mtspr(SPRN_DAC1, vcpu->arch.shadow_dbg_reg.dac[0]); > >>> + mtspr(SPRN_DAC2, vcpu->arch.shadow_dbg_reg.dac[1]); > >>> + > >>> + /* Enable debug events after other debug registers restored */ > >>> + mtspr(SPRN_DBCR0, vcpu->arch.shadow_dbg_reg.dbcr0); } > >> > >> All of the code above looks suspiciously similar to > >> prime_debug_regs();. Can't we somehow reuse that? > > > > I think we can if > > - Save thread->debug_regs in local data structure > > Yes, it can even be on the stack. Thread->denug_regs is not struct , so memcpy() will not work, it is assigning all registers one by one. > > > - Load vcpu->arch->debug_regs in thread->debug_regs > > - Call prime_debug_regs(); > > - Restore thread->debug_regs from local save values in first step > > On heavyweight exit, based on the values on stack, yes. > > > > >> > >>> + > >>> int kvmppc_vcpu_run(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu) > >>> { > >>> int ret, s; > >>> @@ -693,11 +757,25 @@ int kvmppc_vcpu_run(struct kvm_run *kvm_run, > >>> struct > >> kvm_vcpu *vcpu) > >>> kvmppc_load_guest_fp(vcpu); > >>> #endif > >>> > >>> + /* > >>> + * Clear current->thread.dbcr0 so that kernel does not > >>> + * restore h/w registers on context switch in vcpu running state. > >>> + */ > >>> + vcpu->arch.debug_active = 1; > >> > >> = true; > > > > Ok > > > >> > >>> + vcpu->arch.saved_dbcr0 = current->thread.dbcr0; > >>> + current->thread.dbcr0 = 0; > >>> + kvmppc_booke_vcpu_load_debug_regs(vcpu); > >> > >> static void switch_booke_debug_regs(struct thread_struct *new_thread) > >> { > >> if ((current->thread.dbcr0 & DBCR0_IDM) > >> || (new_thread->dbcr0 & DBCR0_IDM)) > >> prime_debug_regs(new_thread); } > >> > >> The kernel will also restore debug state if the process we come from > >> has debugging enabled. Please adjust the comment accordingly. > > > > kvmppc_booke_vcpu_load_debug_regs(vcpu); cleares DBSR and DBCR0, even if > previous process have used debug registers. Is not that sufficient? I do not > think we should load all other registers. > > It's sufficient, but the comment is wrong. ok > > > > >> > >>> + > >>> ret = __kvmppc_vcpu_run(kvm_run, vcpu); > >>> > >>> /* No need for kvm_guest_exit. It's done in handle_exit. > >>> We also get here with interrupts enabled. */ > >>> > >>> + /* Restore thread->dbcr0 */ > >>> + vcpu->arch.debug_active = 0; > >>> + current->thread.dbcr0 = vcpu->arch.saved_dbcr0; > >>> + kvmppc_load_usespace_gebug(); > >>> + > >>> #ifdef CONFIG_PPC_FPU > >>> kvmppc_save_guest_fp(vcpu); > >>> > >>> @@ -753,6 +831,36 @@ static int emulation_exit(struct kvm_run *run, > >>> struct > >> kvm_vcpu *vcpu) > >>> } > >>> } > >>> > >>> +/* > >>> + * Currently we do not support debug resource emulation to guest, > >>> + * so always exit to user space irrespective of user space is > >>> + * expecting the debug exception or not. This is unexpected event > >>> + * and let us leave the action on user space. > >>> + */ > >>> +static int kvmppc_handle_debug(struct kvm_run *run, struct kvm_vcpu > >>> +*vcpu) { > >>> + u32 dbsr = mfspr(SPRN_DBSR); > >>> + > >>> + kvmppc_clear_dbsr(); > >>> + run->debug.arch.status = 0; > >>> + run->debug.arch.address = vcpu->arch.pc; > >>> + > >>> + if (dbsr & (DBSR_IAC1 | DBSR_IAC2 | DBSR_IAC3 | DBSR_IAC4)) { > >>> + run->debug.arch.status |= KVMPPC_DEBUG_BREAKPOINT; > >>> + } else { > >>> + if (dbsr & (DBSR_DAC1W | DBSR_DAC2W)) > >>> + run->debug.arch.status |= KVMPPC_DEBUG_WATCH_WRITE; > >>> + else if (dbsr & (DBSR_DAC1R | DBSR_DAC2R)) > >>> + run->debug.arch.status |= KVMPPC_DEBUG_WATCH_READ; > >>> + if (dbsr & (DBSR_DAC1R | DBSR_DAC1W)) > >>> + run->debug.arch.address = vcpu->arch.shadow_dbg_reg.dac[0]; > >>> + else if (dbsr & (DBSR_DAC2R | DBSR_DAC2W)) > >>> + run->debug.arch.address = vcpu->arch.shadow_dbg_reg.dac[1]; > >>> + } > >>> + > >>> + return RESUME_HOST; > >>> +} > >>> + > >>> static void kvmppc_fill_pt_regs(struct pt_regs *regs) { > >>> ulong r1, ip, msr, lr; > >>> @@ -1112,18 +1220,10 @@ int kvmppc_handle_exit(struct kvm_run *run, > >>> struct > >> kvm_vcpu *vcpu, > >>> } > >>> > >>> case BOOKE_INTERRUPT_DEBUG: { > >>> - u32 dbsr; > >>> - > >>> - vcpu->arch.pc = mfspr(SPRN_CSRR0); > >>> - > >>> - /* clear IAC events in DBSR register */ > >>> - dbsr = mfspr(SPRN_DBSR); > >>> - dbsr &= DBSR_IAC1 | DBSR_IAC2 | DBSR_IAC3 | DBSR_IAC4; > >>> - mtspr(SPRN_DBSR, dbsr); > >>> - > >>> - run->exit_reason = KVM_EXIT_DEBUG; > >>> + r = kvmppc_handle_debug(run, vcpu); > >>> + if (r == RESUME_HOST) > >>> + run->exit_reason = KVM_EXIT_DEBUG; > >>> kvmppc_account_exit(vcpu, DEBUG_EXITS); > >>> - r = RESUME_HOST; > >>> break; > >>> } > >>> > >>> @@ -1174,7 +1274,7 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) > >>> kvmppc_set_msr(vcpu, 0); > >>> > >>> #ifndef CONFIG_KVM_BOOKE_HV > >>> - vcpu->arch.shadow_msr = MSR_USER | MSR_DE | MSR_IS | MSR_DS; > >>> + vcpu->arch.shadow_msr = MSR_USER | MSR_IS | MSR_DS; > >>> vcpu->arch.shadow_pid = 1; > >>> vcpu->arch.shared->msr = 0; > >>> #endif > >>> @@ -1529,12 +1629,6 @@ int kvm_vcpu_ioctl_set_one_reg(struct > >>> kvm_vcpu *vcpu, > >> struct kvm_one_reg *reg) > >>> return r; > >>> } > >>> > >>> -int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu, > >>> - struct kvm_guest_debug *dbg) > >>> -{ > >>> - return -EINVAL; > >>> -} > >>> - > >>> int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct > >>> kvm_fpu *fpu) { > >>> return -ENOTSUPP; > >>> @@ -1640,16 +1734,128 @@ void kvmppc_decrementer_func(unsigned long data) > >>> kvmppc_set_tsr_bits(vcpu, TSR_DIS); } > >>> > >>> +static void kvmppc_booke_vcpu_put_debug_regs(struct kvm_vcpu *vcpu) > >>> +{ > >>> + /* Disable all debug events First */ > >> > >> first > >> > >>> + mtspr(SPRN_DBCR0, 0x0); > >>> + /* Disable pending debug event by clearing DBSR */ > >>> + kvmppc_clear_dbsr(); > >> > >> kvmppc_handle_debug() happens with preemption enabled, no? > > > > Want to clarify, preemption will be enabled on calling local_irq_enable(); in > kvmppc_handle_exit()? > > Yes. Implicitly :). > > > > >> So we can have a > >> debug event that gets cleared on preempt by this. > > > > Should we read the DBSR in before local_irq_enable() in kvmppc_handle_exit()? > > We have to, yes. Otherwise we could get preempted in between and get bogus data > I presume. Ok Thanks -Bharat > > > > >> > >>> +} > >>> + > >>> +int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu, > >>> + struct kvm_guest_debug *dbg) > >>> +{ > >>> + struct kvmppc_booke_debug_reg *dbg_reg; > >>> + int n, b = 0, w = 0; > >>> + const u32 bp_code[] = { > >>> + DBCR0_IAC1 | DBCR0_IDM, > >>> + DBCR0_IAC2 | DBCR0_IDM, > >>> + DBCR0_IAC3 | DBCR0_IDM, > >>> + DBCR0_IAC4 | DBCR0_IDM > >>> + }; > >>> + const u32 wp_code[] = { > >>> + DBCR0_DAC1W | DBCR0_IDM, > >>> + DBCR0_DAC2W | DBCR0_IDM, > >>> + DBCR0_DAC1R | DBCR0_IDM, > >>> + DBCR0_DAC2R | DBCR0_IDM > >>> + }; > >>> + > >>> + if (!(dbg->control & KVM_GUESTDBG_ENABLE)) { > >>> + /* Clear All debug events */ > >>> + vcpu->arch.shadow_dbg_reg.dbcr0 = 0; > >>> + vcpu->guest_debug = 0; > >>> +#ifdef CONFIG_KVM_BOOKE_HV > >>> + /* > >>> + * When user space is not using the debug resources > >>> + * then allow guest to change the MSR.DE. > >>> + */ > >>> + vcpu->arch.shadow_msrp &= ~MSRP_DEP; #endif > >>> + return 0; > >>> + } > >>> + > >>> +#ifdef CONFIG_KVM_BOOKE_HV > >>> + /* > >>> + * When user space is using the debug resource then > >>> + * do not allow guest to change the MSR.DE. > >>> + */ > >>> + vcpu->arch.shadow_msrp &= ~MSRP_DEP; > >> > >> This is supposed to be |= right? > > > > Yes :-) > > > >> > >>> +#endif > >>> + vcpu->guest_debug = dbg->control; > >>> + vcpu->arch.shadow_dbg_reg.dbcr0 = 0; > >>> + /* Set DBCR0_EDM in guest visible DBCR0 register. */ > >>> + vcpu->arch.dbg_reg.dbcr0 = DBCR0_EDM; > >>> + > >>> + if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) > >>> + vcpu->arch.shadow_dbg_reg.dbcr0 |= DBCR0_IDM | DBCR0_IC; > >>> + > >>> + if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)) > >>> + /* Code below handles only HW breakpoints */ > >> > >> Please move the comment out of the one-lined branch. Just put it > >> below the return. > > > > Ok > > > >> > >>> + return 0; > >>> + > >>> + dbg_reg = &(vcpu->arch.shadow_dbg_reg); > >>> + > >>> + /* > >>> + * On BOOKE (e500v2); Set DBCR1 and DBCR2 to allow debug events > >>> + * to occur when MSR.PR is set. > >>> + * On BOOKE-HV (e500mc+); MSR.PR = 0 when guest is running. So we > >>> + * should clear DBCR1 and DBCR2. And EPCR.DUVD is used to control > >>> + * that debug events will not come in hypervisor (GS = 0). > >>> + */ > >>> +#ifdef CONFIG_KVM_BOOKE_HV > >>> + dbg_reg->dbcr1 = 0; > >>> + dbg_reg->dbcr2 = 0; > >>> +#else > >>> + dbg_reg->dbcr1 = DBCR1_IAC1US | DBCR1_IAC2US | DBCR1_IAC3US | > >>> + DBCR1_IAC4US; > >>> + dbg_reg->dbcr2 = DBCR2_DAC1US | DBCR2_DAC2US; #endif > >>> + > >>> + for (n = 0; n < (KVMPPC_BOOKE_IAC_NUM + KVMPPC_BOOKE_DAC_NUM); n++) { > >>> + u32 type = dbg->arch.bp[n].type; > >>> + > >>> + if (!type) > >>> + continue; > >> > >> Scott's comment on why a zero-type is different from any other > >> invalid type is still outstanding I think. > > > > Userspce does following > > - dbg->arch.bp[] is first zero initialized. So dbg->arch.bp[n].type is > > 0 (KVMPPC_DEBUG_NONE) > > - Then set following for valid breakpoints/watchpoints: > > -- dbg->arch.bp[n].type (KVMPPC_DEBUG_BREAKPOINT or KVMPPC_DEBUG_WATCH_WRITE > or KVMPPC_DEBUG_WATCH_READ) > > -- dbg->arch.bp[n].addr > > > > I tried to avoid loop when type is 0 (probably saying type == > KVMPPC_DEBUG_NONE should be more clear). > > That works, yes. > > > if type is invalid then should we return -EINVAL. > > Yes. > > > Alex > -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
