From patchwork Mon Feb 9 12:33:45 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Christian Borntraeger X-Patchwork-Id: 5801091 Return-Path: X-Original-To: patchwork-kvm@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork1.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork1.web.kernel.org (Postfix) with ESMTP id E7D0F9F336 for ; Mon, 9 Feb 2015 12:33:57 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id ADA962011E for ; Mon, 9 Feb 2015 12:33:56 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 2EE152011B for ; Mon, 9 Feb 2015 12:33:55 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760114AbbBIMdu (ORCPT ); Mon, 9 Feb 2015 07:33:50 -0500 Received: from e06smtp17.uk.ibm.com ([195.75.94.113]:48153 "EHLO e06smtp17.uk.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1760003AbbBIMdc (ORCPT ); Mon, 9 Feb 2015 07:33:32 -0500 Received: from /spool/local by e06smtp17.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Mon, 9 Feb 2015 12:33:31 -0000 Received: from d06dlp02.portsmouth.uk.ibm.com (9.149.20.14) by e06smtp17.uk.ibm.com (192.168.101.147) with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted; Mon, 9 Feb 2015 12:33:29 -0000 Received: from b06cxnps3074.portsmouth.uk.ibm.com (d06relay09.portsmouth.uk.ibm.com [9.149.109.194]) by d06dlp02.portsmouth.uk.ibm.com (Postfix) with ESMTP id 3FD87219004D; Mon, 9 Feb 2015 12:33:24 +0000 (GMT) Received: from d06av12.portsmouth.uk.ibm.com (d06av12.portsmouth.uk.ibm.com [9.149.37.247]) by b06cxnps3074.portsmouth.uk.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id t19CXS4c11272464; Mon, 9 Feb 2015 12:33:28 GMT Received: from d06av12.portsmouth.uk.ibm.com (localhost [127.0.0.1]) by d06av12.portsmouth.uk.ibm.com (8.14.4/8.14.4/NCO v10.0 AVout) with ESMTP id t19CXQ0t029762; Mon, 9 Feb 2015 05:33:28 -0700 Received: from tuxmaker.boeblingen.de.ibm.com (tuxmaker.boeblingen.de.ibm.com [9.152.85.9]) by d06av12.portsmouth.uk.ibm.com (8.14.4/8.14.4/NCO v10.0 AVin) with ESMTP id t19CXQSb029743; Mon, 9 Feb 2015 05:33:26 -0700 Received: by tuxmaker.boeblingen.de.ibm.com (Postfix, from userid 25651) id 4205D122442F; Mon, 9 Feb 2015 13:33:26 +0100 (CET) From: Christian Borntraeger To: Paolo Bonzini Cc: KVM , Cornelia Huck , Jens Freimann , linux-s390 , Alexander Graf , Michael Mueller , Christian Borntraeger Subject: [GIT PULL 5/7] KVM: s390: use facilities and cpu_id per KVM Date: Mon, 9 Feb 2015 13:33:45 +0100 Message-Id: <1423485227-51721-6-git-send-email-borntraeger@de.ibm.com> X-Mailer: git-send-email 1.9.3 In-Reply-To: <1423485227-51721-1-git-send-email-borntraeger@de.ibm.com> References: <1423485227-51721-1-git-send-email-borntraeger@de.ibm.com> X-TM-AS-MML: disable X-Content-Scanned: Fidelis XPS MAILER x-cbid: 15020912-0029-0000-0000-00000349D806 Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org X-Spam-Status: No, score=-6.9 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_HI, T_RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP From: Michael Mueller The patch introduces facilities and cpu_ids per virtual machine. Different virtual machines may want to expose different facilities and cpu ids to the guest, so let's make them per-vm instead of global. Signed-off-by: Michael Mueller Reviewed-by: Cornelia Huck Reviewed-by: David Hildenbrand Signed-off-by: Christian Borntraeger --- arch/s390/include/asm/kvm_host.h | 21 +++++++++ arch/s390/kvm/gaccess.c | 4 +- arch/s390/kvm/kvm-s390.c | 92 +++++++++++++++++++++++++--------------- arch/s390/kvm/kvm-s390.h | 13 ++++-- arch/s390/kvm/priv.c | 13 ++++-- 5 files changed, 99 insertions(+), 44 deletions(-) diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h index 77ae014..79dc3b0 100644 --- a/arch/s390/include/asm/kvm_host.h +++ b/arch/s390/include/asm/kvm_host.h @@ -506,6 +506,26 @@ struct s390_io_adapter { #define MAX_S390_IO_ADAPTERS ((MAX_ISC + 1) * 8) #define MAX_S390_ADAPTER_MAPS 256 +/* maximum size of facilities and facility mask is 2k bytes */ +#define S390_ARCH_FAC_LIST_SIZE_BYTE (1<<11) +#define S390_ARCH_FAC_LIST_SIZE_U64 \ + (S390_ARCH_FAC_LIST_SIZE_BYTE / sizeof(u64)) +#define S390_ARCH_FAC_MASK_SIZE_BYTE S390_ARCH_FAC_LIST_SIZE_BYTE +#define S390_ARCH_FAC_MASK_SIZE_U64 \ + (S390_ARCH_FAC_MASK_SIZE_BYTE / sizeof(u64)) + +struct s390_model_fac { + /* facilities used in SIE context */ + __u64 sie[S390_ARCH_FAC_LIST_SIZE_U64]; + /* subset enabled by kvm */ + __u64 kvm[S390_ARCH_FAC_LIST_SIZE_U64]; +}; + +struct kvm_s390_cpu_model { + struct s390_model_fac *fac; + struct cpuid cpu_id; +}; + struct kvm_s390_crypto { struct kvm_s390_crypto_cb *crycb; __u32 crycbd; @@ -536,6 +556,7 @@ struct kvm_arch{ int ipte_lock_count; struct mutex ipte_mutex; spinlock_t start_stop_lock; + struct kvm_s390_cpu_model model; struct kvm_s390_crypto crypto; u64 epoch; }; diff --git a/arch/s390/kvm/gaccess.c b/arch/s390/kvm/gaccess.c index 8a1be90..267523c 100644 --- a/arch/s390/kvm/gaccess.c +++ b/arch/s390/kvm/gaccess.c @@ -357,8 +357,8 @@ static unsigned long guest_translate(struct kvm_vcpu *vcpu, unsigned long gva, union asce asce; ctlreg0.val = vcpu->arch.sie_block->gcr[0]; - edat1 = ctlreg0.edat && test_vfacility(8); - edat2 = edat1 && test_vfacility(78); + edat1 = ctlreg0.edat && test_kvm_facility(vcpu->kvm, 8); + edat2 = edat1 && test_kvm_facility(vcpu->kvm, 78); asce.val = get_vcpu_asce(vcpu); if (asce.r) goto real_address; diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index deac473..8c538a1 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -30,7 +30,6 @@ #include #include #include -#include #include #include "kvm-s390.h" #include "gaccess.h" @@ -100,15 +99,20 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { NULL } }; -unsigned long *vfacilities; -static struct gmap_notifier gmap_notifier; +/* upper facilities limit for kvm */ +unsigned long kvm_s390_fac_list_mask[] = { + 0xff82fffbf4fc2000UL, + 0x005c000000000000UL, +}; -/* test availability of vfacility */ -int test_vfacility(unsigned long nr) +unsigned long kvm_s390_fac_list_mask_size(void) { - return __test_facility(nr, (void *) vfacilities); + BUILD_BUG_ON(ARRAY_SIZE(kvm_s390_fac_list_mask) > S390_ARCH_FAC_MASK_SIZE_U64); + return ARRAY_SIZE(kvm_s390_fac_list_mask); } +static struct gmap_notifier gmap_notifier; + /* Section: not file related */ int kvm_arch_hardware_enable(void) { @@ -351,7 +355,7 @@ static int kvm_s390_vm_set_crypto(struct kvm *kvm, struct kvm_device_attr *attr) struct kvm_vcpu *vcpu; int i; - if (!test_vfacility(76)) + if (!test_kvm_facility(kvm, 76)) return -EINVAL; mutex_lock(&kvm->lock); @@ -700,9 +704,15 @@ static void kvm_s390_set_crycb_format(struct kvm *kvm) kvm->arch.crypto.crycbd |= CRYCB_FORMAT1; } +static void kvm_s390_get_cpu_id(struct cpuid *cpu_id) +{ + get_cpu_id(cpu_id); + cpu_id->version = 0xff; +} + static int kvm_s390_crypto_init(struct kvm *kvm) { - if (!test_vfacility(76)) + if (!test_kvm_facility(kvm, 76)) return 0; kvm->arch.crypto.crycb = kzalloc(sizeof(*kvm->arch.crypto.crycb), @@ -721,7 +731,7 @@ static int kvm_s390_crypto_init(struct kvm *kvm) int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) { - int rc; + int i, rc; char debug_name[16]; static unsigned long sca_offset; @@ -756,6 +766,34 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) if (!kvm->arch.dbf) goto out_nodbf; + /* + * The architectural maximum amount of facilities is 16 kbit. To store + * this amount, 2 kbyte of memory is required. Thus we need a full + * page to hold the active copy (arch.model.fac->sie) and the current + * facilities set (arch.model.fac->kvm). Its address size has to be + * 31 bits and word aligned. + */ + kvm->arch.model.fac = + (struct s390_model_fac *) get_zeroed_page(GFP_KERNEL | GFP_DMA); + if (!kvm->arch.model.fac) + goto out_nofac; + + memcpy(kvm->arch.model.fac->kvm, S390_lowcore.stfle_fac_list, + S390_ARCH_FAC_LIST_SIZE_U64); + + /* + * Apply the kvm facility mask to limit the kvm supported/tolerated + * facility list. + */ + for (i = 0; i < S390_ARCH_FAC_LIST_SIZE_U64; i++) { + if (i < kvm_s390_fac_list_mask_size()) + kvm->arch.model.fac->kvm[i] &= kvm_s390_fac_list_mask[i]; + else + kvm->arch.model.fac->kvm[i] = 0UL; + } + + kvm_s390_get_cpu_id(&kvm->arch.model.cpu_id); + if (kvm_s390_crypto_init(kvm) < 0) goto out_crypto; @@ -787,6 +825,8 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) out_nogmap: kfree(kvm->arch.crypto.crycb); out_crypto: + free_page((unsigned long)kvm->arch.model.fac); +out_nofac: debug_unregister(kvm->arch.dbf); out_nodbf: free_page((unsigned long)(kvm->arch.sca)); @@ -839,6 +879,7 @@ static void kvm_free_vcpus(struct kvm *kvm) void kvm_arch_destroy_vm(struct kvm *kvm) { kvm_free_vcpus(kvm); + free_page((unsigned long)kvm->arch.model.fac); free_page((unsigned long)(kvm->arch.sca)); debug_unregister(kvm->arch.dbf); kfree(kvm->arch.crypto.crycb); @@ -934,7 +975,7 @@ void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu) static void kvm_s390_vcpu_crypto_setup(struct kvm_vcpu *vcpu) { - if (!test_vfacility(76)) + if (!test_kvm_facility(vcpu->kvm, 76)) return; vcpu->arch.sie_block->ecb3 &= ~(ECB3_AES | ECB3_DEA); @@ -973,7 +1014,7 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) CPUSTAT_STOPPED | CPUSTAT_GED); vcpu->arch.sie_block->ecb = 6; - if (test_vfacility(50) && test_vfacility(73)) + if (test_kvm_facility(vcpu->kvm, 50) && test_kvm_facility(vcpu->kvm, 73)) vcpu->arch.sie_block->ecb |= 0x10; vcpu->arch.sie_block->ecb2 = 8; @@ -982,7 +1023,6 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) vcpu->arch.sie_block->eca |= 1; if (sclp_has_sigpif()) vcpu->arch.sie_block->eca |= 0x10000000U; - vcpu->arch.sie_block->fac = (int) (long) vfacilities; vcpu->arch.sie_block->ictl |= ICTL_ISKE | ICTL_SSKE | ICTL_RRBE | ICTL_TPROT; @@ -993,8 +1033,10 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) } hrtimer_init(&vcpu->arch.ckc_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); vcpu->arch.ckc_timer.function = kvm_s390_idle_wakeup; - get_cpu_id(&vcpu->arch.cpu_id); - vcpu->arch.cpu_id.version = 0xff; + + vcpu->arch.cpu_id = vcpu->kvm->arch.model.cpu_id; + memcpy(vcpu->kvm->arch.model.fac->sie, vcpu->kvm->arch.model.fac->kvm, + S390_ARCH_FAC_LIST_SIZE_BYTE); kvm_s390_vcpu_crypto_setup(vcpu); @@ -1038,6 +1080,7 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, vcpu->arch.sie_block->scaol = (__u32)(__u64)kvm->arch.sca; set_bit(63 - id, (unsigned long *) &kvm->arch.sca->mcn); } + vcpu->arch.sie_block->fac = (int) (long) kvm->arch.model.fac->sie; spin_lock_init(&vcpu->arch.local_int.lock); vcpu->arch.local_int.float_int = &kvm->arch.float_int; @@ -2103,30 +2146,11 @@ void kvm_arch_commit_memory_region(struct kvm *kvm, static int __init kvm_s390_init(void) { - int ret; - ret = kvm_init(NULL, sizeof(struct kvm_vcpu), 0, THIS_MODULE); - if (ret) - return ret; - - /* - * guests can ask for up to 255+1 double words, we need a full page - * to hold the maximum amount of facilities. On the other hand, we - * only set facilities that are known to work in KVM. - */ - vfacilities = (unsigned long *) get_zeroed_page(GFP_KERNEL|GFP_DMA); - if (!vfacilities) { - kvm_exit(); - return -ENOMEM; - } - memcpy(vfacilities, S390_lowcore.stfle_fac_list, 16); - vfacilities[0] &= 0xff82fffbf4fc2000UL; - vfacilities[1] &= 0x005c000000000000UL; - return 0; + return kvm_init(NULL, sizeof(struct kvm_vcpu), 0, THIS_MODULE); } static void __exit kvm_s390_exit(void) { - free_page((unsigned long) vfacilities); kvm_exit(); } diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h index c22dce8..985c211 100644 --- a/arch/s390/kvm/kvm-s390.h +++ b/arch/s390/kvm/kvm-s390.h @@ -18,12 +18,10 @@ #include #include #include +#include typedef int (*intercept_handler_t)(struct kvm_vcpu *vcpu); -/* declare vfacilities extern */ -extern unsigned long *vfacilities; - /* Transactional Memory Execution related macros */ #define IS_TE_ENABLED(vcpu) ((vcpu->arch.sie_block->ecb & 0x10)) #define TDB_FORMAT1 1 @@ -127,6 +125,12 @@ static inline void kvm_s390_set_psw_cc(struct kvm_vcpu *vcpu, unsigned long cc) vcpu->arch.sie_block->gpsw.mask |= cc << 44; } +/* test availability of facility in a kvm intance */ +static inline int test_kvm_facility(struct kvm *kvm, unsigned long nr) +{ + return __test_facility(nr, kvm->arch.model.fac->kvm); +} + /* are cpu states controlled by user space */ static inline int kvm_s390_user_cpu_state_ctrl(struct kvm *kvm) { @@ -183,7 +187,8 @@ int kvm_s390_vcpu_setup_cmma(struct kvm_vcpu *vcpu); void kvm_s390_vcpu_unsetup_cmma(struct kvm_vcpu *vcpu); /* is cmma enabled */ bool kvm_s390_cmma_enabled(struct kvm *kvm); -int test_vfacility(unsigned long nr); +unsigned long kvm_s390_fac_list_mask_size(void); +extern unsigned long kvm_s390_fac_list_mask[]; /* implemented in diag.c */ int kvm_s390_handle_diag(struct kvm_vcpu *vcpu); diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c index 1be578d..bdd9b5b 100644 --- a/arch/s390/kvm/priv.c +++ b/arch/s390/kvm/priv.c @@ -337,19 +337,24 @@ static int handle_io_inst(struct kvm_vcpu *vcpu) static int handle_stfl(struct kvm_vcpu *vcpu) { int rc; + unsigned int fac; vcpu->stat.instruction_stfl++; if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); + /* + * We need to shift the lower 32 facility bits (bit 0-31) from a u64 + * into a u32 memory representation. They will remain bits 0-31. + */ + fac = *vcpu->kvm->arch.model.fac->sie >> 32; rc = write_guest_lc(vcpu, offsetof(struct _lowcore, stfl_fac_list), - vfacilities, 4); + &fac, sizeof(fac)); if (rc) return rc; - VCPU_EVENT(vcpu, 5, "store facility list value %x", - *(unsigned int *) vfacilities); - trace_kvm_s390_handle_stfl(vcpu, *(unsigned int *) vfacilities); + VCPU_EVENT(vcpu, 5, "store facility list value %x", fac); + trace_kvm_s390_handle_stfl(vcpu, fac); return 0; }