* [PATCH v2 0/2] KVM/arm64: add fp/simd lazy switch support @ 2015-10-06 2:33 Mario Smarduch 2015-10-06 2:33 ` [PATCH v2 1/2] add hooks for armv8 fp/simd lazy switch Mario Smarduch 2015-10-06 2:33 ` [PATCH v2 2/2] enable " Mario Smarduch 0 siblings, 2 replies; 5+ messages in thread From: Mario Smarduch @ 2015-10-06 2:33 UTC (permalink / raw) To: linux-arm-kernel This patch series is a followup to the armv7 fp/simd lazy switch implementation, uses similar approach and depends on the series - https://lists.cs.columbia.edu/pipermail/kvmarm/2015-September/016567.html Patches are based on 4.3-rc2 commit 1f93e4a96c91093 Patches are based on earlier arm64 fp/simd optimization work - https://lists.cs.columbia.edu/pipermail/kvmarm/2015-July/015748.html And subsequent fixes by Marc and Christoffer at KVM Forum hackathon to handle 32-bit guest on 64 bit host - https://lists.cs.columbia.edu/pipermail/kvmarm/2015-August/016128.html The patch series have been tested on Foundation Model arm64/arm64 and arm32/arm64. The test program used can be found here https://github.com/mjsmar/arm-arm64-fpsimd-test Launched upto 16 instances on 4-way Guest and another 16 on the host (both cases 1mS sleep), ran overnight. Changes v1->v2: - Tested arm32/arm64 - rebased to 4.3-rc2 - changed a couple register accesses from 64 to 32 bit Mario Smarduch (2): add hooks for armv8 fp/simd lazy switch enable armv8 fp/simd lazy switch arch/arm/kvm/arm.c | 2 -- arch/arm64/include/asm/kvm_asm.h | 1 + arch/arm64/include/asm/kvm_host.h | 3 ++ arch/arm64/kernel/asm-offsets.c | 1 + arch/arm64/kvm/hyp.S | 59 ++++++++++++++++++++++++++------------- 5 files changed, 45 insertions(+), 21 deletions(-) -- 1.9.1 ^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH v2 1/2] add hooks for armv8 fp/simd lazy switch 2015-10-06 2:33 [PATCH v2 0/2] KVM/arm64: add fp/simd lazy switch support Mario Smarduch @ 2015-10-06 2:33 ` Mario Smarduch 2015-10-26 14:32 ` Christoffer Dall 2015-10-06 2:33 ` [PATCH v2 2/2] enable " Mario Smarduch 1 sibling, 1 reply; 5+ messages in thread From: Mario Smarduch @ 2015-10-06 2:33 UTC (permalink / raw) To: linux-arm-kernel This patch adds hooks to support fp/simd lazy switch. A vcpu flag to track fp/simd state, and flag offset in vcpu structure. Signed-off-by: Mario Smarduch <m.smarduch@samsung.com> --- arch/arm64/include/asm/kvm_host.h | 3 +++ arch/arm64/kernel/asm-offsets.c | 1 + 2 files changed, 4 insertions(+) diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h index 4562459..03f25d0 100644 --- a/arch/arm64/include/asm/kvm_host.h +++ b/arch/arm64/include/asm/kvm_host.h @@ -157,6 +157,9 @@ struct kvm_vcpu_arch { /* Interrupt related fields */ u64 irq_lines; /* IRQ and FIQ levels */ + /* Track fp/simd lazy switch */ + u32 vfp_lazy; + /* Cache some mmu pages needed inside spinlock regions */ struct kvm_mmu_memory_cache mmu_page_cache; diff --git a/arch/arm64/kernel/asm-offsets.c b/arch/arm64/kernel/asm-offsets.c index 8d89cf8..8311da4 100644 --- a/arch/arm64/kernel/asm-offsets.c +++ b/arch/arm64/kernel/asm-offsets.c @@ -124,6 +124,7 @@ int main(void) DEFINE(VCPU_HCR_EL2, offsetof(struct kvm_vcpu, arch.hcr_el2)); DEFINE(VCPU_MDCR_EL2, offsetof(struct kvm_vcpu, arch.mdcr_el2)); DEFINE(VCPU_IRQ_LINES, offsetof(struct kvm_vcpu, arch.irq_lines)); + DEFINE(VCPU_VFP_LAZY, offsetof(struct kvm_vcpu, arch.vfp_lazy)); DEFINE(VCPU_HOST_CONTEXT, offsetof(struct kvm_vcpu, arch.host_cpu_context)); DEFINE(VCPU_HOST_DEBUG_STATE, offsetof(struct kvm_vcpu, arch.host_debug_state)); DEFINE(VCPU_TIMER_CNTV_CTL, offsetof(struct kvm_vcpu, arch.timer_cpu.cntv_ctl)); -- 1.9.1 ^ permalink raw reply related [flat|nested] 5+ messages in thread
* [PATCH v2 1/2] add hooks for armv8 fp/simd lazy switch 2015-10-06 2:33 ` [PATCH v2 1/2] add hooks for armv8 fp/simd lazy switch Mario Smarduch @ 2015-10-26 14:32 ` Christoffer Dall 0 siblings, 0 replies; 5+ messages in thread From: Christoffer Dall @ 2015-10-26 14:32 UTC (permalink / raw) To: linux-arm-kernel On Mon, Oct 05, 2015 at 07:33:13PM -0700, Mario Smarduch wrote: > This patch adds hooks to support fp/simd lazy switch. A vcpu flag to track > fp/simd state, and flag offset in vcpu structure. > > Signed-off-by: Mario Smarduch <m.smarduch@samsung.com> > --- > arch/arm64/include/asm/kvm_host.h | 3 +++ > arch/arm64/kernel/asm-offsets.c | 1 + > 2 files changed, 4 insertions(+) > > diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h > index 4562459..03f25d0 100644 > --- a/arch/arm64/include/asm/kvm_host.h > +++ b/arch/arm64/include/asm/kvm_host.h > @@ -157,6 +157,9 @@ struct kvm_vcpu_arch { > /* Interrupt related fields */ > u64 irq_lines; /* IRQ and FIQ levels */ > > + /* Track fp/simd lazy switch */ > + u32 vfp_lazy; > + like for the 32-bit patch set I think this could have a more meaningful name with more clearly defined semantics, e.g. vfp_host_dirty. You could also call it vfp_flags like the debug flags and define a specific flag for your use... Thanks, -Christoffer > /* Cache some mmu pages needed inside spinlock regions */ > struct kvm_mmu_memory_cache mmu_page_cache; > > diff --git a/arch/arm64/kernel/asm-offsets.c b/arch/arm64/kernel/asm-offsets.c > index 8d89cf8..8311da4 100644 > --- a/arch/arm64/kernel/asm-offsets.c > +++ b/arch/arm64/kernel/asm-offsets.c > @@ -124,6 +124,7 @@ int main(void) > DEFINE(VCPU_HCR_EL2, offsetof(struct kvm_vcpu, arch.hcr_el2)); > DEFINE(VCPU_MDCR_EL2, offsetof(struct kvm_vcpu, arch.mdcr_el2)); > DEFINE(VCPU_IRQ_LINES, offsetof(struct kvm_vcpu, arch.irq_lines)); > + DEFINE(VCPU_VFP_LAZY, offsetof(struct kvm_vcpu, arch.vfp_lazy)); > DEFINE(VCPU_HOST_CONTEXT, offsetof(struct kvm_vcpu, arch.host_cpu_context)); > DEFINE(VCPU_HOST_DEBUG_STATE, offsetof(struct kvm_vcpu, arch.host_debug_state)); > DEFINE(VCPU_TIMER_CNTV_CTL, offsetof(struct kvm_vcpu, arch.timer_cpu.cntv_ctl)); > -- > 1.9.1 > ^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH v2 2/2] enable armv8 fp/simd lazy switch 2015-10-06 2:33 [PATCH v2 0/2] KVM/arm64: add fp/simd lazy switch support Mario Smarduch 2015-10-06 2:33 ` [PATCH v2 1/2] add hooks for armv8 fp/simd lazy switch Mario Smarduch @ 2015-10-06 2:33 ` Mario Smarduch 2015-10-26 14:32 ` Christoffer Dall 1 sibling, 1 reply; 5+ messages in thread From: Mario Smarduch @ 2015-10-06 2:33 UTC (permalink / raw) To: linux-arm-kernel This patch enables arm64 lazy fp/simd switch. Removes the ARM constraint, and follows the same approach as armv7 version - found here. https://lists.cs.columbia.edu/pipermail/kvmarm/2015-September/016567.html To summarize - provided the guest accesses fp/simd unit we limit number of fp/simd context switches to two per vCPU execution schedule. Signed-off-by: Mario Smarduch <m.smarduch@samsung.com> --- arch/arm/kvm/arm.c | 2 -- arch/arm64/include/asm/kvm_asm.h | 1 + arch/arm64/kvm/hyp.S | 59 +++++++++++++++++++++++++++------------- 3 files changed, 41 insertions(+), 21 deletions(-) diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c index 1b1f9e9..fe609f1 100644 --- a/arch/arm/kvm/arm.c +++ b/arch/arm/kvm/arm.c @@ -112,12 +112,10 @@ void kvm_arch_check_processor_compat(void *rtn) */ static void kvm_switch_fp_regs(struct kvm_vcpu *vcpu) { -#ifdef CONFIG_ARM if (vcpu->arch.vfp_lazy == 1) { kvm_call_hyp(__kvm_restore_host_vfp_state, vcpu); vcpu->arch.vfp_lazy = 0; } -#endif } /** diff --git a/arch/arm64/include/asm/kvm_asm.h b/arch/arm64/include/asm/kvm_asm.h index 5e37710..83dcac5 100644 --- a/arch/arm64/include/asm/kvm_asm.h +++ b/arch/arm64/include/asm/kvm_asm.h @@ -117,6 +117,7 @@ extern char __kvm_hyp_vector[]; extern void __kvm_flush_vm_context(void); extern void __kvm_tlb_flush_vmid_ipa(struct kvm *kvm, phys_addr_t ipa); extern void __kvm_tlb_flush_vmid(struct kvm *kvm); +extern void __kvm_restore_host_vfp_state(struct kvm_vcpu *vcpu); extern int __kvm_vcpu_run(struct kvm_vcpu *vcpu); diff --git a/arch/arm64/kvm/hyp.S b/arch/arm64/kvm/hyp.S index e583613..ea99f66 100644 --- a/arch/arm64/kvm/hyp.S +++ b/arch/arm64/kvm/hyp.S @@ -385,14 +385,6 @@ tbz \tmp, #KVM_ARM64_DEBUG_DIRTY_SHIFT, \target .endm -/* - * Branch to target if CPTR_EL2.TFP bit is set (VFP/SIMD trapping enabled) - */ -.macro skip_fpsimd_state tmp, target - mrs \tmp, cptr_el2 - tbnz \tmp, #CPTR_EL2_TFP_SHIFT, \target -.endm - .macro compute_debug_state target // Compute debug state: If any of KDE, MDE or KVM_ARM64_DEBUG_DIRTY // is set, we do a full save/restore cycle and disable trapping. @@ -433,10 +425,6 @@ mrs x5, ifsr32_el2 stp x4, x5, [x3] - skip_fpsimd_state x8, 2f - mrs x6, fpexc32_el2 - str x6, [x3, #16] -2: skip_debug_state x8, 1f mrs x7, dbgvcr32_el2 str x7, [x3, #24] @@ -481,8 +469,15 @@ isb 99: msr hcr_el2, x2 - mov x2, #CPTR_EL2_TTA + + mov x2, #0 + ldr w3, [x0, #VCPU_VFP_LAZY] + tbnz w3, #0, 98f + orr x2, x2, #CPTR_EL2_TFP +98: + orr x2, x2, #CPTR_EL2_TTA + msr cptr_el2, x2 mov x2, #(1 << 15) // Trap CP15 Cr=15 @@ -669,14 +664,12 @@ __restore_debug: ret __save_fpsimd: - skip_fpsimd_state x3, 1f save_fpsimd -1: ret + ret __restore_fpsimd: - skip_fpsimd_state x3, 1f restore_fpsimd -1: ret + ret switch_to_guest_fpsimd: push x4, lr @@ -688,6 +681,9 @@ switch_to_guest_fpsimd: mrs x0, tpidr_el2 + mov w2, #1 + str w2, [x0, #VCPU_VFP_LAZY] + ldr x2, [x0, #VCPU_HOST_CONTEXT] kern_hyp_va x2 bl __save_fpsimd @@ -763,7 +759,6 @@ __kvm_vcpu_return: add x2, x0, #VCPU_CONTEXT save_guest_regs - bl __save_fpsimd bl __save_sysregs skip_debug_state x3, 1f @@ -784,7 +779,6 @@ __kvm_vcpu_return: kern_hyp_va x2 bl __restore_sysregs - bl __restore_fpsimd /* Clear FPSIMD and Trace trapping */ msr cptr_el2, xzr @@ -863,6 +857,33 @@ ENTRY(__kvm_flush_vm_context) ret ENDPROC(__kvm_flush_vm_context) +/** + * kvm_switch_fp_regs() - switch guest/host VFP/SIMD registers + * @vcpu: pointer to vcpu structure. + * + */ +ENTRY(__kvm_restore_host_vfp_state) + push x4, lr + + kern_hyp_va x0 + add x2, x0, #VCPU_CONTEXT + + // Load Guest HCR, determine if guest is 32 or 64 bit + ldr x3, [x0, #VCPU_HCR_EL2] + tbnz x3, #HCR_RW_SHIFT, 1f + mrs x4, fpexc32_el2 + str x4, [x2, #CPU_SYSREG_OFFSET(FPEXC32_EL2)] +1: + bl __save_fpsimd + + ldr x2, [x0, #VCPU_HOST_CONTEXT] + kern_hyp_va x2 + bl __restore_fpsimd + + pop x4, lr + ret +ENDPROC(__kvm_restore_host_vfp_state) + __kvm_hyp_panic: // Guess the context by looking@VTTBR: // If zero, then we're already a host. -- 1.9.1 ^ permalink raw reply related [flat|nested] 5+ messages in thread
* [PATCH v2 2/2] enable armv8 fp/simd lazy switch 2015-10-06 2:33 ` [PATCH v2 2/2] enable " Mario Smarduch @ 2015-10-26 14:32 ` Christoffer Dall 0 siblings, 0 replies; 5+ messages in thread From: Christoffer Dall @ 2015-10-26 14:32 UTC (permalink / raw) To: linux-arm-kernel On Mon, Oct 05, 2015 at 07:33:14PM -0700, Mario Smarduch wrote: > This patch enables arm64 lazy fp/simd switch. Removes the ARM constraint, > and follows the same approach as armv7 version - found here. > > https://lists.cs.columbia.edu/pipermail/kvmarm/2015-September/016567.html > > To summarize - provided the guest accesses fp/simd unit we limit number > of fp/simd context switches to two per vCPU execution schedule. > > Signed-off-by: Mario Smarduch <m.smarduch@samsung.com> > --- > arch/arm/kvm/arm.c | 2 -- > arch/arm64/include/asm/kvm_asm.h | 1 + > arch/arm64/kvm/hyp.S | 59 +++++++++++++++++++++++++++------------- > 3 files changed, 41 insertions(+), 21 deletions(-) > > diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c > index 1b1f9e9..fe609f1 100644 > --- a/arch/arm/kvm/arm.c > +++ b/arch/arm/kvm/arm.c > @@ -112,12 +112,10 @@ void kvm_arch_check_processor_compat(void *rtn) > */ > static void kvm_switch_fp_regs(struct kvm_vcpu *vcpu) > { > -#ifdef CONFIG_ARM > if (vcpu->arch.vfp_lazy == 1) { > kvm_call_hyp(__kvm_restore_host_vfp_state, vcpu); > vcpu->arch.vfp_lazy = 0; > } > -#endif I'm assuming you'll rework this like for the 32-bit series to just do this in EL1, yes? > } > > /** > diff --git a/arch/arm64/include/asm/kvm_asm.h b/arch/arm64/include/asm/kvm_asm.h > index 5e37710..83dcac5 100644 > --- a/arch/arm64/include/asm/kvm_asm.h > +++ b/arch/arm64/include/asm/kvm_asm.h > @@ -117,6 +117,7 @@ extern char __kvm_hyp_vector[]; > extern void __kvm_flush_vm_context(void); > extern void __kvm_tlb_flush_vmid_ipa(struct kvm *kvm, phys_addr_t ipa); > extern void __kvm_tlb_flush_vmid(struct kvm *kvm); > +extern void __kvm_restore_host_vfp_state(struct kvm_vcpu *vcpu); > > extern int __kvm_vcpu_run(struct kvm_vcpu *vcpu); > > diff --git a/arch/arm64/kvm/hyp.S b/arch/arm64/kvm/hyp.S > index e583613..ea99f66 100644 > --- a/arch/arm64/kvm/hyp.S > +++ b/arch/arm64/kvm/hyp.S > @@ -385,14 +385,6 @@ > tbz \tmp, #KVM_ARM64_DEBUG_DIRTY_SHIFT, \target > .endm > > -/* > - * Branch to target if CPTR_EL2.TFP bit is set (VFP/SIMD trapping enabled) > - */ > -.macro skip_fpsimd_state tmp, target > - mrs \tmp, cptr_el2 > - tbnz \tmp, #CPTR_EL2_TFP_SHIFT, \target > -.endm > - > .macro compute_debug_state target > // Compute debug state: If any of KDE, MDE or KVM_ARM64_DEBUG_DIRTY > // is set, we do a full save/restore cycle and disable trapping. > @@ -433,10 +425,6 @@ > mrs x5, ifsr32_el2 > stp x4, x5, [x3] > > - skip_fpsimd_state x8, 2f > - mrs x6, fpexc32_el2 > - str x6, [x3, #16] > -2: > skip_debug_state x8, 1f > mrs x7, dbgvcr32_el2 > str x7, [x3, #24] > @@ -481,8 +469,15 @@ > isb > 99: > msr hcr_el2, x2 > - mov x2, #CPTR_EL2_TTA > + > + mov x2, #0 > + ldr w3, [x0, #VCPU_VFP_LAZY] > + tbnz w3, #0, 98f > + > orr x2, x2, #CPTR_EL2_TFP > +98: > + orr x2, x2, #CPTR_EL2_TTA > + this can be reworked more simply by using an extra register and leaving the initial mov x2, #CPTR_EL2_TTA alone and just skip setting the TFP bit if we've already switched the state. > msr cptr_el2, x2 > > mov x2, #(1 << 15) // Trap CP15 Cr=15 > @@ -669,14 +664,12 @@ __restore_debug: > ret > > __save_fpsimd: > - skip_fpsimd_state x3, 1f > save_fpsimd > -1: ret > + ret > > __restore_fpsimd: > - skip_fpsimd_state x3, 1f > restore_fpsimd > -1: ret > + ret > > switch_to_guest_fpsimd: > push x4, lr > @@ -688,6 +681,9 @@ switch_to_guest_fpsimd: > > mrs x0, tpidr_el2 > > + mov w2, #1 > + str w2, [x0, #VCPU_VFP_LAZY] > + > ldr x2, [x0, #VCPU_HOST_CONTEXT] > kern_hyp_va x2 > bl __save_fpsimd > @@ -763,7 +759,6 @@ __kvm_vcpu_return: > add x2, x0, #VCPU_CONTEXT > > save_guest_regs > - bl __save_fpsimd > bl __save_sysregs > > skip_debug_state x3, 1f > @@ -784,7 +779,6 @@ __kvm_vcpu_return: > kern_hyp_va x2 > > bl __restore_sysregs > - bl __restore_fpsimd > /* Clear FPSIMD and Trace trapping */ > msr cptr_el2, xzr > > @@ -863,6 +857,33 @@ ENTRY(__kvm_flush_vm_context) > ret > ENDPROC(__kvm_flush_vm_context) > > +/** > + * kvm_switch_fp_regs() - switch guest/host VFP/SIMD registers > + * @vcpu: pointer to vcpu structure. > + * > + */ > +ENTRY(__kvm_restore_host_vfp_state) > + push x4, lr even when you move this to el1, I don't think you need to do the additional store of the registers as this is going to be called from C-code and you should have plenty of caller-save registers, if I read the aarch64 PCS correctly. > + > + kern_hyp_va x0 > + add x2, x0, #VCPU_CONTEXT > + > + // Load Guest HCR, determine if guest is 32 or 64 bit > + ldr x3, [x0, #VCPU_HCR_EL2] > + tbnz x3, #HCR_RW_SHIFT, 1f > + mrs x4, fpexc32_el2 > + str x4, [x2, #CPU_SYSREG_OFFSET(FPEXC32_EL2)] > +1: > + bl __save_fpsimd > + > + ldr x2, [x0, #VCPU_HOST_CONTEXT] > + kern_hyp_va x2 > + bl __restore_fpsimd > + > + pop x4, lr > + ret > +ENDPROC(__kvm_restore_host_vfp_state) > + > __kvm_hyp_panic: > // Guess the context by looking at VTTBR: > // If zero, then we're already a host. > -- > 1.9.1 > Otherwise this looks reasonably good. Thanks, -Christoffer ^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2015-10-26 14:32 UTC | newest] Thread overview: 5+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2015-10-06 2:33 [PATCH v2 0/2] KVM/arm64: add fp/simd lazy switch support Mario Smarduch 2015-10-06 2:33 ` [PATCH v2 1/2] add hooks for armv8 fp/simd lazy switch Mario Smarduch 2015-10-26 14:32 ` Christoffer Dall 2015-10-06 2:33 ` [PATCH v2 2/2] enable " Mario Smarduch 2015-10-26 14:32 ` Christoffer Dall
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).