Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756124AbbBEAxD (ORCPT ); Wed, 4 Feb 2015 19:53:03 -0500 Received: from ozlabs.org ([103.22.144.67]:54459 "EHLO ozlabs.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750818AbbBEAxA (ORCPT ); Wed, 4 Feb 2015 19:53:00 -0500 From: David Gibson To: paulus@samba.org, agraf@suse.de Cc: aik@ozlabs.ru, kvm-ppc@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, David Gibson Subject: [PATCHv3] kvmppc: Implement H_LOGICAL_CI_{LOAD,STORE} in KVM Date: Thu, 5 Feb 2015 11:53:25 +1100 Message-Id: <1423097605-2169-1-git-send-email-david@gibson.dropbear.id.au> X-Mailer: git-send-email 2.1.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 7360 Lines: 240 On POWER, storage caching is usually configured via the MMU - attributes such as cache-inhibited are stored in the TLB and the hashed page table. This makes correctly performing cache inhibited IO accesses awkward when the MMU is turned off (real mode). Some CPU models provide special registers to control the cache attributes of real mode load and stores but this is not at all consistent. This is a problem in particular for SLOF, the firmware used on KVM guests, which runs entirely in real mode, but which needs to do IO to load the kernel. To simplify this qemu implements two special hypercalls, H_LOGICAL_CI_LOAD and H_LOGICAL_CI_STORE which simulate a cache-inhibited load or store to a logical address (aka guest physical address). SLOF uses these for IO. However, because these are implemented within qemu, not the host kernel, these bypass any IO devices emulated within KVM itself. The simplest way to see this problem is to attempt to boot a KVM guest from a virtio-blk device with iothread / dataplane enabled. The iothread code relies on an in kernel implementation of the virtio queue notification, which is not triggered by the IO hcalls, and so the guest will stall in SLOF unable to load the guest OS. This patch addresses this by providing in-kernel implementations of the 2 hypercalls, which correctly scan the KVM IO bus. Any access to an address not handled by the KVM IO bus will cause a VM exit, hitting the qemu implementation as before. Note that a userspace change is also required, in order to enable these new hcall implementations with KVM_CAP_PPC_ENABLE_HCALL. Signed-off-by: David Gibson --- arch/powerpc/include/asm/kvm_book3s.h | 3 ++ arch/powerpc/kvm/book3s.c | 76 +++++++++++++++++++++++++++++++++++ arch/powerpc/kvm/book3s_hv.c | 12 ++++++ arch/powerpc/kvm/book3s_pr_papr.c | 28 +++++++++++++ 4 files changed, 119 insertions(+) v3: - Use __beXX types instead of plain uXX where appropriate - Remove some unhelpful comments v2: - Removed some debugging printk()s that were accidentally left in - Fix endianness; like all PAPR hypercalls, these should always act big-endian, even if the guest is little-endian (in practice this makes no difference, since the only user is SLOF, which is always big-endian) diff --git a/arch/powerpc/include/asm/kvm_book3s.h b/arch/powerpc/include/asm/kvm_book3s.h index 942c7b1..578e550 100644 --- a/arch/powerpc/include/asm/kvm_book3s.h +++ b/arch/powerpc/include/asm/kvm_book3s.h @@ -292,6 +292,9 @@ static inline bool kvmppc_supports_magic_page(struct kvm_vcpu *vcpu) return !is_kvmppc_hv_enabled(vcpu->kvm); } +extern int kvmppc_h_logical_ci_load(struct kvm_vcpu *vcpu); +extern int kvmppc_h_logical_ci_store(struct kvm_vcpu *vcpu); + /* Magic register values loaded into r3 and r4 before the 'sc' assembly * instruction for the OSI hypercalls */ #define OSI_SC_MAGIC_R3 0x113724FA diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c index 888bf46..397f433 100644 --- a/arch/powerpc/kvm/book3s.c +++ b/arch/powerpc/kvm/book3s.c @@ -820,6 +820,82 @@ void kvmppc_core_destroy_vm(struct kvm *kvm) #endif } +int kvmppc_h_logical_ci_load(struct kvm_vcpu *vcpu) +{ + unsigned long size = kvmppc_get_gpr(vcpu, 4); + unsigned long addr = kvmppc_get_gpr(vcpu, 5); + u64 buf; + int ret; + + if (!is_power_of_2(size) || (size > sizeof(buf))) + return H_TOO_HARD; + + ret = kvm_io_bus_read(vcpu->kvm, KVM_MMIO_BUS, addr, size, &buf); + if (ret != 0) + return H_TOO_HARD; + + switch (size) { + case 1: + kvmppc_set_gpr(vcpu, 4, *(u8 *)&buf); + break; + + case 2: + kvmppc_set_gpr(vcpu, 4, be16_to_cpu(*(__be16 *)&buf)); + break; + + case 4: + kvmppc_set_gpr(vcpu, 4, be32_to_cpu(*(__be32 *)&buf)); + break; + + case 8: + kvmppc_set_gpr(vcpu, 4, be64_to_cpu(*(__be64 *)&buf)); + break; + + default: + BUG(); + } + + return H_SUCCESS; +} +EXPORT_SYMBOL_GPL(kvmppc_h_logical_ci_load); + +int kvmppc_h_logical_ci_store(struct kvm_vcpu *vcpu) +{ + unsigned long size = kvmppc_get_gpr(vcpu, 4); + unsigned long addr = kvmppc_get_gpr(vcpu, 5); + unsigned long val = kvmppc_get_gpr(vcpu, 6); + u64 buf; + int ret; + + switch (size) { + case 1: + *(u8 *)&buf = val; + break; + + case 2: + *(__be16 *)&buf = cpu_to_be16(val); + break; + + case 4: + *(__be32 *)&buf = cpu_to_be32(val); + break; + + case 8: + *(__be64 *)&buf = cpu_to_be64(val); + break; + + default: + return H_TOO_HARD; + } + + ret = kvm_io_bus_write(vcpu->kvm, KVM_MMIO_BUS, addr, size, &buf); + if (ret != 0) + return H_TOO_HARD; + + return H_SUCCESS; +} +EXPORT_SYMBOL_GPL(kvmppc_h_logical_ci_store); + int kvmppc_core_check_processor_compat(void) { /* diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c index de4018a..f631b76 100644 --- a/arch/powerpc/kvm/book3s_hv.c +++ b/arch/powerpc/kvm/book3s_hv.c @@ -706,6 +706,16 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu) /* Send the error out to userspace via KVM_RUN */ return rc; + case H_LOGICAL_CI_LOAD: + ret = kvmppc_h_logical_ci_load(vcpu); + if (ret == H_TOO_HARD) + return RESUME_HOST; + break; + case H_LOGICAL_CI_STORE: + ret = kvmppc_h_logical_ci_store(vcpu); + if (ret == H_TOO_HARD) + return RESUME_HOST; + break; case H_SET_MODE: ret = kvmppc_h_set_mode(vcpu, kvmppc_get_gpr(vcpu, 4), kvmppc_get_gpr(vcpu, 5), @@ -740,6 +750,8 @@ static int kvmppc_hcall_impl_hv(unsigned long cmd) case H_CONFER: case H_REGISTER_VPA: case H_SET_MODE: + case H_LOGICAL_CI_LOAD: + case H_LOGICAL_CI_STORE: #ifdef CONFIG_KVM_XICS case H_XIRR: case H_CPPR: diff --git a/arch/powerpc/kvm/book3s_pr_papr.c b/arch/powerpc/kvm/book3s_pr_papr.c index ce3c893..f2c75a1 100644 --- a/arch/powerpc/kvm/book3s_pr_papr.c +++ b/arch/powerpc/kvm/book3s_pr_papr.c @@ -258,6 +258,28 @@ static int kvmppc_h_pr_put_tce(struct kvm_vcpu *vcpu) return EMULATE_DONE; } +static int kvmppc_h_pr_logical_ci_load(struct kvm_vcpu *vcpu) +{ + long rc; + + rc = kvmppc_h_logical_ci_load(vcpu); + if (rc == H_TOO_HARD) + return EMULATE_FAIL; + kvmppc_set_gpr(vcpu, 3, rc); + return EMULATE_DONE; +} + +static int kvmppc_h_pr_logical_ci_store(struct kvm_vcpu *vcpu) +{ + long rc; + + rc = kvmppc_h_logical_ci_store(vcpu); + if (rc == H_TOO_HARD) + return EMULATE_FAIL; + kvmppc_set_gpr(vcpu, 3, rc); + return EMULATE_DONE; +} + static int kvmppc_h_pr_xics_hcall(struct kvm_vcpu *vcpu, u32 cmd) { long rc = kvmppc_xics_hcall(vcpu, cmd); @@ -290,6 +312,10 @@ int kvmppc_h_pr(struct kvm_vcpu *vcpu, unsigned long cmd) clear_bit(KVM_REQ_UNHALT, &vcpu->requests); vcpu->stat.halt_wakeup++; return EMULATE_DONE; + case H_LOGICAL_CI_LOAD: + return kvmppc_h_pr_logical_ci_load(vcpu); + case H_LOGICAL_CI_STORE: + return kvmppc_h_pr_logical_ci_store(vcpu); case H_XIRR: case H_CPPR: case H_EOI: @@ -323,6 +349,8 @@ int kvmppc_hcall_impl_pr(unsigned long cmd) case H_BULK_REMOVE: case H_PUT_TCE: case H_CEDE: + case H_LOGICAL_CI_LOAD: + case H_LOGICAL_CI_STORE: #ifdef CONFIG_KVM_XICS case H_XIRR: case H_CPPR: -- 2.1.0 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/