From patchwork Thu Jan 24 12:28:07 2013 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Cornelia Huck X-Patchwork-Id: 2031521 Return-Path: X-Original-To: patchwork-kvm@patchwork.kernel.org Delivered-To: patchwork-process-083081@patchwork1.kernel.org Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by patchwork1.kernel.org (Postfix) with ESMTP id E7B273FDBC for ; Thu, 24 Jan 2013 12:28:46 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754246Ab3AXM2k (ORCPT ); Thu, 24 Jan 2013 07:28:40 -0500 Received: from e06smtp14.uk.ibm.com ([195.75.94.110]:38130 "EHLO e06smtp14.uk.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753759Ab3AXM2X (ORCPT ); Thu, 24 Jan 2013 07:28:23 -0500 Received: from /spool/local by e06smtp14.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Thu, 24 Jan 2013 12:26:23 -0000 Received: from b06cxnps3074.portsmouth.uk.ibm.com (9.149.109.194) by e06smtp14.uk.ibm.com (192.168.101.144) with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted; Thu, 24 Jan 2013 12:26:21 -0000 Received: from d06av09.portsmouth.uk.ibm.com (d06av09.portsmouth.uk.ibm.com [9.149.37.250]) by b06cxnps3074.portsmouth.uk.ibm.com (8.13.8/8.13.8/NCO v10.0) with ESMTP id r0OCSAHA24182876; Thu, 24 Jan 2013 12:28:10 GMT Received: from d06av09.portsmouth.uk.ibm.com (loopback [127.0.0.1]) by d06av09.portsmouth.uk.ibm.com (8.14.4/8.13.1/NCO v10.0 AVout) with ESMTP id r0OCSGwT026407; Thu, 24 Jan 2013 05:28:18 -0700 Received: from tuxmaker.boeblingen.de.ibm.com (tuxmaker.boeblingen.de.ibm.com [9.152.85.9]) by d06av09.portsmouth.uk.ibm.com (8.14.4/8.13.1/NCO v10.0 AVin) with ESMTP id r0OCSB3Z026110; Thu, 24 Jan 2013 05:28:16 -0700 From: Cornelia Huck To: qemu-devel , KVM , linux-s390 Cc: Marcelo Tosatti , Gleb Natapov , Anthony Liguori , Christian Borntraeger , Carsten Otte , Alexander Graf , Heiko Carstens , Martin Schwidefsky , Sebastian Ott , =?UTF-8?q?Andreas=20F=C3=A4rber?= Subject: [PATCH 07/11] s390: Wire up channel I/O in kvm. Date: Thu, 24 Jan 2013 13:28:07 +0100 Message-Id: <1359030491-46725-8-git-send-email-cornelia.huck@de.ibm.com> X-Mailer: git-send-email 1.7.12.4 In-Reply-To: <1359030491-46725-1-git-send-email-cornelia.huck@de.ibm.com> References: <1359030491-46725-1-git-send-email-cornelia.huck@de.ibm.com> x-cbid: 13012412-1948-0000-0000-0000041F3D70 Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Trigger the code for our virtual css in case of instruction intercepts for I/O instructions. Handle the tsch exit for the subchannel-related part of tsch. Signed-off-by: Cornelia Huck --- target-s390x/cpu.h | 11 +++ target-s390x/kvm.c | 239 ++++++++++++++++++++++++++++++++++++++++++++++++++--- 2 files changed, 237 insertions(+), 13 deletions(-) diff --git a/target-s390x/cpu.h b/target-s390x/cpu.h index 778065c..ce12fa4 100644 --- a/target-s390x/cpu.h +++ b/target-s390x/cpu.h @@ -1058,6 +1058,13 @@ void QEMU_NORETURN runtime_exception(CPUS390XState *env, int excp, #include +#ifdef CONFIG_KVM +void kvm_s390_io_interrupt(S390CPU *cpu, uint16_t subchannel_id, + uint16_t subchannel_nr, uint32_t io_int_parm, + uint32_t io_int_word); +void kvm_s390_crw_mchk(S390CPU *cpu); +void kvm_s390_enable_css_support(S390CPU *cpu); +#else static inline void kvm_s390_io_interrupt(S390CPU *cpu, uint16_t subchannel_id, uint16_t subchannel_nr, @@ -1068,6 +1075,10 @@ static inline void kvm_s390_io_interrupt(S390CPU *cpu, static inline void kvm_s390_crw_mchk(S390CPU *cpu) { } +static inline void kvm_s390_enable_css_support(S390CPU *cpu) +{ +} +#endif static inline void s390_io_interrupt(S390CPU *cpu, uint16_t subchannel_id, diff --git a/target-s390x/kvm.c b/target-s390x/kvm.c index add6a58..e876406 100644 --- a/target-s390x/kvm.c +++ b/target-s390x/kvm.c @@ -47,9 +47,29 @@ #define IPA0_DIAG 0x8300 #define IPA0_SIGP 0xae00 -#define IPA0_PRIV 0xb200 +#define IPA0_B2 0xb200 +#define IPA0_B9 0xb900 +#define IPA0_EB 0xeb00 #define PRIV_SCLP_CALL 0x20 +#define PRIV_CSCH 0x30 +#define PRIV_HSCH 0x31 +#define PRIV_MSCH 0x32 +#define PRIV_SSCH 0x33 +#define PRIV_STSCH 0x34 +#define PRIV_TSCH 0x35 +#define PRIV_TPI 0x36 +#define PRIV_SAL 0x37 +#define PRIV_RSCH 0x38 +#define PRIV_STCRW 0x39 +#define PRIV_STCPS 0x3a +#define PRIV_RCHP 0x3b +#define PRIV_SCHM 0x3c +#define PRIV_CHSC 0x5f +#define PRIV_SIGA 0x74 +#define PRIV_XSCH 0x76 +#define PRIV_SQBS 0x8a +#define PRIV_EQBS 0x9c #define DIAG_KVM_HYPERCALL 0x500 #define DIAG_KVM_BREAKPOINT 0x501 @@ -375,10 +395,123 @@ static int kvm_sclp_service_call(S390CPU *cpu, struct kvm_run *run, return 0; } -static int handle_priv(S390CPU *cpu, struct kvm_run *run, uint8_t ipa1) +static int kvm_handle_css_inst(S390CPU *cpu, struct kvm_run *run, + uint8_t ipa0, uint8_t ipa1, uint8_t ipb) +{ + int r = 0; + int no_cc = 0; + CPUS390XState *env = &cpu->env; + + if (ipa0 != 0xb2) { + /* Not handled for now. */ + return -1; + } + cpu_synchronize_state(env); + switch (ipa1) { + case PRIV_XSCH: + r = ioinst_handle_xsch(env, env->regs[1]); + break; + case PRIV_CSCH: + r = ioinst_handle_csch(env, env->regs[1]); + break; + case PRIV_HSCH: + r = ioinst_handle_hsch(env, env->regs[1]); + break; + case PRIV_MSCH: + r = ioinst_handle_msch(env, env->regs[1], run->s390_sieic.ipb); + break; + case PRIV_SSCH: + r = ioinst_handle_ssch(env, env->regs[1], run->s390_sieic.ipb); + break; + case PRIV_STCRW: + r = ioinst_handle_stcrw(env, run->s390_sieic.ipb); + break; + case PRIV_STSCH: + r = ioinst_handle_stsch(env, env->regs[1], run->s390_sieic.ipb); + break; + case PRIV_TSCH: + /* We should only get tsch via KVM_EXIT_S390_TSCH. */ + fprintf(stderr, "Spurious tsch intercept\n"); + break; + case PRIV_CHSC: + r = ioinst_handle_chsc(env, run->s390_sieic.ipb); + break; + case PRIV_TPI: + /* This should have been handled by kvm already. */ + fprintf(stderr, "Spurious tpi intercept\n"); + break; + case PRIV_SCHM: + no_cc = 1; + r = ioinst_handle_schm(env, env->regs[1], env->regs[2], + run->s390_sieic.ipb); + break; + case PRIV_RSCH: + r = ioinst_handle_rsch(env, env->regs[1]); + break; + case PRIV_RCHP: + r = ioinst_handle_rchp(env, env->regs[1]); + break; + case PRIV_STCPS: + /* We do not provide this instruction, it is suppressed. */ + no_cc = 1; + r = 0; + break; + case PRIV_SAL: + no_cc = 1; + r = ioinst_handle_sal(env, env->regs[1]); + break; + default: + r = -1; + break; + } + + if (r >= 0) { + if (!no_cc) { + setcc(cpu, r); + } + r = 0; + } else if (r < -1) { + r = 0; + } + return r; +} + +static int is_ioinst(uint8_t ipa0, uint8_t ipa1, uint8_t ipb) +{ + int ret = 0; + uint16_t ipa = (ipa0 << 8) | ipa1; + + switch (ipa) { + case IPA0_B2 | PRIV_CSCH: + case IPA0_B2 | PRIV_HSCH: + case IPA0_B2 | PRIV_MSCH: + case IPA0_B2 | PRIV_SSCH: + case IPA0_B2 | PRIV_STSCH: + case IPA0_B2 | PRIV_TPI: + case IPA0_B2 | PRIV_SAL: + case IPA0_B2 | PRIV_RSCH: + case IPA0_B2 | PRIV_STCRW: + case IPA0_B2 | PRIV_STCPS: + case IPA0_B2 | PRIV_RCHP: + case IPA0_B2 | PRIV_SCHM: + case IPA0_B2 | PRIV_CHSC: + case IPA0_B2 | PRIV_SIGA: + case IPA0_B2 | PRIV_XSCH: + case IPA0_B9 | PRIV_EQBS: + case IPA0_EB | PRIV_SQBS: + ret = 1; + break; + } + + return ret; +} + +static int handle_priv(S390CPU *cpu, struct kvm_run *run, + uint8_t ipa0, uint8_t ipa1) { int r = 0; uint16_t ipbh0 = (run->s390_sieic.ipb & 0xffff0000) >> 16; + uint8_t ipb = run->s390_sieic.ipb & 0xff; dprintf("KVM: PRIV: %d\n", ipa1); switch (ipa1) { @@ -386,8 +519,16 @@ static int handle_priv(S390CPU *cpu, struct kvm_run *run, uint8_t ipa1) r = kvm_sclp_service_call(cpu, run, ipbh0); break; default: - dprintf("KVM: unknown PRIV: 0x%x\n", ipa1); - r = -1; + if (is_ioinst(ipa0, ipa1, ipb)) { + r = kvm_handle_css_inst(cpu, run, ipa0, ipa1, ipb); + if (r == -1) { + setcc(cpu, 3); + r = 0; + } + } else { + dprintf("KVM: unknown PRIV: 0x%x\n", ipa1); + r = -1; + } break; } @@ -528,15 +669,17 @@ static int handle_instruction(S390CPU *cpu, struct kvm_run *run) dprintf("handle_instruction 0x%x 0x%x\n", run->s390_sieic.ipa, run->s390_sieic.ipb); switch (ipa0) { - case IPA0_PRIV: - r = handle_priv(cpu, run, ipa1); - break; - case IPA0_DIAG: - r = handle_diag(env, run, ipb_code); - break; - case IPA0_SIGP: - r = handle_sigp(cpu, run, ipa1); - break; + case IPA0_B2: + case IPA0_B9: + case IPA0_EB: + r = handle_priv(cpu, run, ipa0 >> 8, ipa1); + break; + case IPA0_DIAG: + r = handle_diag(env, run, ipb_code); + break; + case IPA0_SIGP: + r = handle_sigp(cpu, run, ipa1); + break; } if (r < 0) { @@ -595,6 +738,43 @@ static int handle_intercept(S390CPU *cpu) return r; } +static int handle_tsch(S390CPU *cpu) +{ + CPUS390XState *env = &cpu->env; + CPUState *cs = CPU(cpu); + struct kvm_run *run = cs->kvm_run; + int ret; + + cpu_synchronize_state(env); + ret = ioinst_handle_tsch(env, env->regs[1], run->s390_tsch.ipb); + if (ret >= 0) { + /* Success; set condition code. */ + setcc(cpu, ret); + ret = 0; + } else if (ret < -1) { + /* + * Failure. + * If an I/O interrupt had been dequeued, we have to reinject it. + */ + if (run->s390_tsch.dequeued) { + uint16_t subchannel_id = run->s390_tsch.subchannel_id; + uint16_t subchannel_nr = run->s390_tsch.subchannel_nr; + uint32_t io_int_parm = run->s390_tsch.io_int_parm; + uint32_t io_int_word = run->s390_tsch.io_int_word; + uint32_t type = ((subchannel_id & 0xff00) << 24) | + ((subchannel_id & 0x00060) << 22) | (subchannel_nr << 16); + + kvm_s390_interrupt_internal(cpu, type, + ((uint32_t)subchannel_id << 16) + | subchannel_nr, + ((uint64_t)io_int_parm << 32) + | io_int_word, 1); + } + ret = 0; + } + return ret; +} + int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run) { S390CPU *cpu = S390_CPU(cs); @@ -607,6 +787,9 @@ int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run) case KVM_EXIT_S390_RESET: qemu_system_reset_request(); break; + case KVM_EXIT_S390_TSCH: + ret = handle_tsch(cpu); + break; default: fprintf(stderr, "Unknown KVM exit: %d\n", run->exit_reason); break; @@ -632,3 +815,33 @@ int kvm_arch_on_sigbus(int code, void *addr) { return 1; } + +void kvm_s390_io_interrupt(S390CPU *cpu, uint16_t subchannel_id, + uint16_t subchannel_nr, uint32_t io_int_parm, + uint32_t io_int_word) +{ + uint32_t type; + + type = ((subchannel_id & 0xff00) << 24) | + ((subchannel_id & 0x00060) << 22) | (subchannel_nr << 16); + kvm_s390_interrupt_internal(cpu, type, + ((uint32_t)subchannel_id << 16) | subchannel_nr, + ((uint64_t)io_int_parm << 32) | io_int_word, 1); +} + +void kvm_s390_crw_mchk(S390CPU *cpu) +{ + kvm_s390_interrupt_internal(cpu, KVM_S390_MCHK, 1 << 28, + 0x00400f1d40330000, 1); +} + +void kvm_s390_enable_css_support(S390CPU *cpu) +{ + struct kvm_enable_cap cap = {}; + int r; + + /* Activate host kernel channel subsystem support. */ + cap.cap = KVM_CAP_S390_CSS_SUPPORT; + r = kvm_vcpu_ioctl(CPU(cpu), KVM_ENABLE_CAP, &cap); + assert(r == 0); +}