[PATCH 2/3] s390: Virtual channel subsystem support.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Provide a mechanism for qemu to provide fully virtual subchannels to
the guest. In the KVM case, this relies on the kernel's css support
for I/O and machine check interrupt handling. The !KVM case handles
interrupts on its own.

Signed-off-by: Cornelia Huck <cornelia.huck@xxxxxxxxxx>
---
 hw/s390x/Makefile.objs     |    1 +
 hw/s390x/css.c             | 1209 ++++++++++++++++++++++++++++++++++++++++++++
 hw/s390x/css.h             |   90 ++++
 target-s390x/Makefile.objs |    2 +-
 target-s390x/cpu.h         |  232 +++++++++
 target-s390x/helper.c      |  146 ++++++
 target-s390x/ioinst.c      |  737 +++++++++++++++++++++++++++
 target-s390x/ioinst.h      |  213 ++++++++
 target-s390x/kvm.c         |  251 ++++++++-
 target-s390x/misc_helper.c |    6 +-
 10 files changed, 2872 insertions(+), 15 deletions(-)
 create mode 100644 hw/s390x/css.c
 create mode 100644 hw/s390x/css.h
 create mode 100644 target-s390x/ioinst.c
 create mode 100644 target-s390x/ioinst.h

diff --git a/hw/s390x/Makefile.objs b/hw/s390x/Makefile.objs
index 096dfcd..378b099 100644
--- a/hw/s390x/Makefile.objs
+++ b/hw/s390x/Makefile.objs
@@ -4,3 +4,4 @@ obj-y := $(addprefix ../,$(obj-y))
 obj-y += sclp.o
 obj-y += event-facility.o
 obj-y += sclpquiesce.o sclpconsole.o
+obj-y += css.o
diff --git a/hw/s390x/css.c b/hw/s390x/css.c
new file mode 100644
index 0000000..9adffb3
--- /dev/null
+++ b/hw/s390x/css.c
@@ -0,0 +1,1209 @@
+/*
+ * Channel subsystem base support.
+ *
+ * Copyright 2012 IBM Corp.
+ * Author(s): Cornelia Huck <cornelia.huck@xxxxxxxxxx>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or (at
+ * your option) any later version. See the COPYING file in the top-level
+ * directory.
+ */
+
+#include "qemu-thread.h"
+#include "qemu-queue.h"
+#include <hw/qdev.h>
+#include "bitops.h"
+#include "kvm.h"
+#include "cpu.h"
+#include "ioinst.h"
+#include "css.h"
+#include "virtio-ccw.h"
+
+typedef struct CrwContainer {
+    CRW crw;
+    QTAILQ_ENTRY(CrwContainer) sibling;
+} CrwContainer;
+
+typedef struct ChpInfo {
+    uint8_t in_use;
+    uint8_t type;
+    uint8_t is_virtual;
+} ChpInfo;
+
+typedef struct SubchSet {
+    SubchDev *sch[MAX_SCHID + 1];
+    unsigned long schids_used[BITS_TO_LONGS(MAX_SCHID + 1)];
+    unsigned long devnos_used[BITS_TO_LONGS(MAX_SCHID + 1)];
+} SubchSet;
+
+typedef struct CssImage {
+    SubchSet *sch_set[MAX_SSID + 1];
+    ChpInfo chpids[MAX_CHPID + 1];
+} CssImage;
+
+typedef struct ChannelSubSys {
+    QTAILQ_HEAD(, CrwContainer) pending_crws;
+    bool do_crw_mchk;
+    bool crws_lost;
+    uint8_t max_cssid;
+    uint8_t max_ssid;
+    bool chnmon_active;
+    uint64_t chnmon_area;
+    CssImage *css[MAX_CSSID + 1];
+    uint8_t default_cssid;
+} ChannelSubSys;
+
+static ChannelSubSys *channel_subsys;
+
+int css_create_css_image(uint8_t cssid, bool default_image)
+{
+    if (cssid > MAX_CSSID) {
+        return -EINVAL;
+    }
+    if (channel_subsys->css[cssid]) {
+        return -EBUSY;
+    }
+    channel_subsys->css[cssid] = g_try_malloc0(sizeof(CssImage));
+    if (!channel_subsys->css[cssid]) {
+        return -ENOMEM;
+    }
+    if (default_image) {
+        channel_subsys->default_cssid = cssid;
+    }
+    return 0;
+}
+
+static void css_write_phys_pmcw(uint64_t addr, PMCW *pmcw)
+{
+    int i;
+    uint32_t offset = 0;
+    struct copy_pmcw {
+        uint32_t intparm;
+        uint16_t flags;
+        uint16_t devno;
+        uint8_t lpm;
+        uint8_t pnom;
+        uint8_t lpum;
+        uint8_t pim;
+        uint16_t mbi;
+        uint8_t pom;
+        uint8_t pam;
+        uint8_t chpid[8];
+        uint32_t chars;
+    } *copy;
+
+    copy = (struct copy_pmcw *)pmcw;
+    stl_phys(addr + offset, copy->intparm);
+    offset += sizeof(copy->intparm);
+    stw_phys(addr + offset, copy->flags);
+    offset += sizeof(copy->flags);
+    stw_phys(addr + offset, copy->devno);
+    offset += sizeof(copy->devno);
+    stb_phys(addr + offset, copy->lpm);
+    offset += sizeof(copy->lpm);
+    stb_phys(addr + offset, copy->pnom);
+    offset += sizeof(copy->pnom);
+    stb_phys(addr + offset, copy->lpum);
+    offset += sizeof(copy->lpum);
+    stb_phys(addr + offset, copy->pim);
+    offset += sizeof(copy->pim);
+    stw_phys(addr + offset, copy->mbi);
+    offset += sizeof(copy->mbi);
+    stb_phys(addr + offset, copy->pom);
+    offset += sizeof(copy->pom);
+    stb_phys(addr + offset, copy->pam);
+    offset += sizeof(copy->pam);
+    for (i = 0; i < 8; i++) {
+        stb_phys(addr + offset, copy->chpid[i]);
+        offset += sizeof(copy->chpid[i]);
+    }
+    stl_phys(addr + offset, copy->chars);
+}
+
+static void css_write_phys_scsw(uint64_t addr, SCSW *scsw)
+{
+    uint32_t offset = 0;
+    struct copy_scsw {
+        uint32_t flags;
+        uint32_t cpa;
+        uint8_t dstat;
+        uint8_t cstat;
+        uint16_t count;
+    } *copy;
+
+    copy = (struct copy_scsw *)scsw;
+    stl_phys(addr + offset, copy->flags);
+    offset += sizeof(copy->flags);
+    stl_phys(addr + offset, copy->cpa);
+    offset += sizeof(copy->cpa);
+    stb_phys(addr + offset, copy->dstat);
+    offset += sizeof(copy->dstat);
+    stb_phys(addr + offset, copy->cstat);
+    offset += sizeof(copy->cstat);
+    stw_phys(addr + offset, copy->count);
+}
+
+static void css_inject_io_interrupt(SubchDev *sch)
+{
+    S390CPU *cpu = s390_cpu_addr2state(0);
+
+    s390_io_interrupt(&cpu->env,
+                      channel_subsys->max_cssid > 0 ?
+                      (sch->cssid << 8) | (1 << 3) | (sch->ssid << 1) | 1 :
+                      (sch->ssid << 1) | 1,
+                      sch->schid,
+                      sch->curr_status.pmcw.intparm,
+                      (0x80 >> sch->curr_status.pmcw.isc) << 24);
+}
+
+void css_conditional_io_interrupt(SubchDev *sch)
+{
+    /*
+     * If the subchannel is not currently status pending, make it pending
+     * with alert status.
+     */
+    if (sch && !(sch->curr_status.scsw.stctl & SCSW_STCTL_STATUS_PEND)) {
+        S390CPU *cpu = s390_cpu_addr2state(0);
+
+        sch->curr_status.scsw.stctl =
+            SCSW_STCTL_ALERT | SCSW_STCTL_STATUS_PEND;
+        /* Inject an I/O interrupt. */
+        s390_io_interrupt(&cpu->env,
+                          channel_subsys->max_cssid > 0 ?
+                          (sch->cssid << 8) | (1 << 3) | (sch->ssid << 1) | 1 :
+                          (sch->ssid << 1) | 1,
+                          sch->schid,
+                          sch->curr_status.pmcw.intparm,
+                          (0x80 >> sch->curr_status.pmcw.isc) << 24);
+    }
+}
+
+static void sch_handle_clear_func(SubchDev *sch)
+{
+    PMCW *p = &sch->curr_status.pmcw;
+    SCSW *s = &sch->curr_status.scsw;
+    int path;
+
+    /* Path management: In our simple css, we always choose the only path. */
+    path = 0x80;
+
+    /* Reset values prior to 'issueing the clear signal'. */
+    p->lpum = 0;
+    p->pom = 0xff;
+    s->pno = 0;
+
+    /* We always 'attempt to issue the clear signal', and we always succeed. */
+    sch->orb = NULL;
+    sch->channel_prog = NULL;
+    sch->last_cmd = NULL;
+    s->actl &= ~SCSW_ACTL_CLEAR_PEND;
+    s->stctl |= SCSW_STCTL_STATUS_PEND;
+
+    s->dstat = 0;
+    s->cstat = 0;
+    p->lpum = path;
+
+}
+
+static void sch_handle_halt_func(SubchDev *sch)
+{
+
+    PMCW *p = &sch->curr_status.pmcw;
+    SCSW *s = &sch->curr_status.scsw;
+    int path;
+
+    /* Path management: In our simple css, we always choose the only path. */
+    path = 0x80;
+
+    /* We always 'attempt to issue the halt signal', and we always succeed. */
+    sch->orb = NULL;
+    sch->channel_prog = NULL;
+    sch->last_cmd = NULL;
+    s->actl &= ~SCSW_ACTL_HALT_PEND;
+    s->stctl |= SCSW_STCTL_STATUS_PEND;
+
+    if ((s->actl & (SCSW_ACTL_SUBCH_ACTIVE | SCSW_ACTL_DEVICE_ACTIVE)) ||
+        !((s->actl & SCSW_ACTL_START_PEND) ||
+          (s->actl & SCSW_ACTL_SUSP))) {
+        s->dstat = SCSW_DSTAT_DEVICE_END;
+    }
+    s->cstat = 0;
+    p->lpum = path;
+
+}
+
+static int css_interpret_ccw(SubchDev *sch, CCW1 *ccw)
+{
+    int ret;
+    bool check_len;
+    int len;
+    int i;
+
+    if (!ccw) {
+        return -EIO;
+    }
+
+    /* Check for invalid command codes. */
+    if ((ccw->cmd_code & 0x0f) == 0) {
+        return -EINVAL;
+    }
+    if (((ccw->cmd_code & 0x0f) == CCW_CMD_TIC) &&
+        ((ccw->cmd_code & 0xf0) != 0)) {
+        return -EINVAL;
+    }
+
+    if (ccw->flags & CCW_FLAG_SUSPEND) {
+        return -ERESTART;
+    }
+
+    check_len = !((ccw->flags & CCW_FLAG_SLI) && !(ccw->flags & CCW_FLAG_DC));
+
+    /* Look at the command. */
+    switch (ccw->cmd_code) {
+    case CCW_CMD_NOOP:
+        /* Nothing to do. */
+        ret = 0;
+        break;
+    case CCW_CMD_BASIC_SENSE:
+        if (check_len) {
+            if (ccw->count != sizeof(sch->sense_data)) {
+                ret = -EINVAL;
+                break;
+            }
+        }
+        len = MIN(ccw->count, sizeof(sch->sense_data));
+        cpu_physical_memory_write(ccw->cda, sch->sense_data, len);
+        sch->curr_status.scsw.count = ccw->count - len;
+        memset(sch->sense_data, 0, sizeof(sch->sense_data));
+        ret = 0;
+        break;
+    case CCW_CMD_SENSE_ID:
+    {
+        uint8_t sense_bytes[256];
+
+        /* Sense ID information is device specific. */
+        memcpy(sense_bytes, &sch->id, sizeof(sense_bytes));
+        if (check_len) {
+            if (ccw->count != sizeof(sense_bytes)) {
+                ret = -EINVAL;
+                break;
+            }
+        }
+        len = MIN(ccw->count, sizeof(sense_bytes));
+        /*
+         * Only indicate 0xff in the first sense byte if we actually
+         * have enough place to store at least bytes 0-3.
+         */
+        if (len >= 4) {
+            stb_phys(ccw->cda, 0xff);
+        } else {
+            stb_phys(ccw->cda, 0);
+        }
+        i = 1;
+        for (i = 1; i < len - 1; i++) {
+            stb_phys(ccw->cda + i, sense_bytes[i]);
+        }
+        sch->curr_status.scsw.count = ccw->count - len;
+        ret = 0;
+        break;
+    }
+    case CCW_CMD_TIC:
+        if (sch->last_cmd->cmd_code == CCW_CMD_TIC) {
+            ret = -EINVAL;
+            break;
+        }
+        if (ccw->flags & (CCW_FLAG_CC | CCW_FLAG_DC)) {
+            ret = -EINVAL;
+            break;
+        }
+        sch->channel_prog = qemu_get_ram_ptr(ccw->cda);
+        ret = sch->channel_prog ? -EAGAIN : -EFAULT;
+        break;
+    default:
+        if (sch->ccw_cb) {
+            /* Handle device specific commands. */
+            ret = sch->ccw_cb(sch, ccw);
+        } else {
+            ret = -EOPNOTSUPP;
+        }
+        break;
+    }
+    sch->last_cmd = ccw;
+    if (ret == 0) {
+        if (ccw->flags & CCW_FLAG_CC) {
+            sch->channel_prog += 8;
+            ret = -EAGAIN;
+        }
+    }
+
+    return ret;
+}
+
+static void sch_handle_start_func(SubchDev *sch)
+{
+
+    PMCW *p = &sch->curr_status.pmcw;
+    SCSW *s = &sch->curr_status.scsw;
+    ORB *orb = sch->orb;
+    int path;
+    int ret;
+
+    /* Path management: In our simple css, we always choose the only path. */
+    path = 0x80;
+
+    if (!s->actl & SCSW_ACTL_SUSP) {
+        /* Look at the orb and try to execute the channel program. */
+        p->intparm = orb->intparm;
+        if (!(orb->lpm & path)) {
+            /* Generate a deferred cc 3 condition. */
+            s->cc = 3;
+            s->stctl = (SCSW_STCTL_ALERT | SCSW_STCTL_STATUS_PEND);
+            return;
+        }
+    } else {
+        s->actl &= ~(SCSW_ACTL_SUSP | SCSW_ACTL_RESUME_PEND);
+    }
+    sch->last_cmd = NULL;
+    do {
+        ret = css_interpret_ccw(sch, sch->channel_prog);
+        switch (ret) {
+        case -EAGAIN:
+            /* ccw chain, continue processing */
+            break;
+        case 0:
+            /* success */
+            s->actl &= ~SCSW_ACTL_START_PEND;
+            s->stctl = SCSW_STCTL_PRIMARY | SCSW_STCTL_SECONDARY |
+                SCSW_STCTL_STATUS_PEND;
+            s->dstat = SCSW_DSTAT_CHANNEL_END | SCSW_DSTAT_DEVICE_END;
+            break;
+        case -EOPNOTSUPP:
+            /* unsupported command, generate unit check (command reject) */
+            s->actl &= ~SCSW_ACTL_START_PEND;
+            s->dstat = SCSW_DSTAT_UNIT_CHECK;
+            /* Set sense bit 0 in ecw0. */
+            sch->sense_data[0] = 0x80;
+            s->stctl = SCSW_STCTL_PRIMARY | SCSW_STCTL_SECONDARY |
+                SCSW_STCTL_ALERT | SCSW_STCTL_STATUS_PEND;
+            break;
+        case -EFAULT:
+            /* memory problem, generate channel data check */
+            s->actl &= ~SCSW_ACTL_START_PEND;
+            s->cstat = SCSW_CSTAT_DATA_CHECK;
+            s->stctl = SCSW_STCTL_PRIMARY | SCSW_STCTL_SECONDARY |
+                SCSW_STCTL_ALERT | SCSW_STCTL_STATUS_PEND;
+            break;
+        case -EBUSY:
+            /* subchannel busy, generate deferred cc 1 */
+            s->cc = 1;
+            s->stctl = SCSW_STCTL_ALERT | SCSW_STCTL_STATUS_PEND;
+            break;
+        case -ERESTART:
+            /* channel program has been suspended */
+            s->actl &= ~SCSW_ACTL_START_PEND;
+            s->actl |= SCSW_ACTL_SUSP;
+            break;
+        default:
+            /* error, generate channel program check */
+            s->actl &= ~SCSW_ACTL_START_PEND;
+            s->cstat = SCSW_CSTAT_PROG_CHECK;
+            s->stctl = SCSW_STCTL_PRIMARY | SCSW_STCTL_SECONDARY |
+                SCSW_STCTL_ALERT | SCSW_STCTL_STATUS_PEND;
+            break;
+        }
+    } while (ret == -EAGAIN);
+
+}
+
+/*
+ * On real machines, this would run asynchronously to the main vcpus.
+ * We might want to make some parts of the ssch handling (interpreting
+ * read/writes) asynchronous later on if we start supporting more than
+ * our current very simple devices.
+ */
+static void do_subchannel_work(SubchDev *sch)
+{
+
+    SCSW *s = &sch->curr_status.scsw;
+
+    if (s->fctl & SCSW_FCTL_CLEAR_FUNC) {
+        sch_handle_clear_func(sch);
+    } else if (s->fctl & SCSW_FCTL_HALT_FUNC) {
+        sch_handle_halt_func(sch);
+    } else if (s->fctl & SCSW_FCTL_START_FUNC) {
+        sch_handle_start_func(sch);
+    } else {
+        /* Cannot happen. */
+        return;
+    }
+    css_inject_io_interrupt(sch);
+}
+
+int css_do_stsch(SubchDev *sch, uint64_t addr)
+{
+    int i;
+    uint32_t offset = 0;
+
+    /* Use current status. */
+    css_write_phys_pmcw(addr, &sch->curr_status.pmcw);
+    offset += sizeof(PMCW);
+    css_write_phys_scsw(addr + offset, &sch->curr_status.scsw);
+    offset += sizeof(SCSW);
+    stq_phys(addr + offset, sch->curr_status.mba);
+    offset += sizeof(sch->curr_status.mba);
+    for (i = 0; i < 4; i++) {
+        stb_phys(addr + offset, sch->curr_status.mda[i]);
+        offset += sizeof(sch->curr_status.mda[i]);
+    }
+    return 0;
+}
+
+int css_do_msch(SubchDev *sch, SCHIB *schib)
+{
+    SCSW *s = &sch->curr_status.scsw;
+    PMCW *p = &sch->curr_status.pmcw;
+    int ret;
+
+    if (!sch->curr_status.pmcw.dnv) {
+        ret = 0;
+        goto out;
+    }
+
+    if (s->stctl & SCSW_STCTL_STATUS_PEND) {
+        ret = -EINPROGRESS;
+        goto out;
+    }
+
+    if (s->fctl &
+        (SCSW_FCTL_START_FUNC|SCSW_FCTL_HALT_FUNC|SCSW_FCTL_CLEAR_FUNC)) {
+        ret = -EBUSY;
+        goto out;
+    }
+
+    /* Only update the program-modifiable fields. */
+    p->ena = schib->pmcw.ena;
+    p->intparm = schib->pmcw.intparm;
+    p->isc = schib->pmcw.isc;
+    p->mp = schib->pmcw.mp;
+    p->lpm = schib->pmcw.lpm;
+    p->pom = schib->pmcw.pom;
+    p->lm = schib->pmcw.lm;
+    p->csense = schib->pmcw.csense;
+
+    p->mme = schib->pmcw.mme;
+    p->mbi = schib->pmcw.mbi;
+    p->mbfc = schib->pmcw.mbfc;
+    sch->curr_status.mba = schib->mba;
+
+    ret = 0;
+
+out:
+    return ret;
+}
+
+int css_do_xsch(SubchDev *sch)
+{
+    SCSW *s = &sch->curr_status.scsw;
+    PMCW *p = &sch->curr_status.pmcw;
+    int ret;
+
+    if (!p->dnv || !p->ena) {
+        ret = -ENODEV;
+        goto out;
+    }
+
+    if (!s->fctl || (s->fctl != SCSW_FCTL_START_FUNC) ||
+        (!(s->actl &
+           (SCSW_ACTL_RESUME_PEND | SCSW_ACTL_START_PEND | SCSW_ACTL_SUSP))) ||
+        (s->actl & SCSW_ACTL_SUBCH_ACTIVE)) {
+        ret = -EINPROGRESS;
+        goto out;
+    }
+
+    if (s->stctl != 0) {
+        ret = -EBUSY;
+        goto out;
+    }
+
+    /* Cancel the current operation. */
+    s->fctl &= ~SCSW_FCTL_START_FUNC;
+    s->actl &= ~(SCSW_ACTL_RESUME_PEND|SCSW_ACTL_START_PEND|SCSW_ACTL_SUSP);
+    sch->channel_prog = NULL;
+    sch->last_cmd = NULL;
+    sch->orb = NULL;
+    s->dstat = 0;
+    s->cstat = 0;
+    ret = 0;
+
+out:
+    return ret;
+}
+
+int css_do_csch(SubchDev *sch)
+{
+    SCSW *s = &sch->curr_status.scsw;
+    PMCW *p = &sch->curr_status.pmcw;
+    int ret;
+
+    if (!p->dnv || !p->ena) {
+        ret = -ENODEV;
+        goto out;
+    }
+
+    /* Trigger the clear function. */
+    s->fctl = SCSW_FCTL_CLEAR_FUNC;
+    s->actl = SCSW_ACTL_CLEAR_PEND;
+
+    do_subchannel_work(sch);
+    ret = 0;
+
+out:
+    return ret;
+}
+
+int css_do_hsch(SubchDev *sch)
+{
+    SCSW *s = &sch->curr_status.scsw;
+    PMCW *p = &sch->curr_status.pmcw;
+    int ret;
+
+    if (!p->dnv || !p->ena) {
+        ret = -ENODEV;
+        goto out;
+    }
+
+    if ((s->stctl == SCSW_STCTL_STATUS_PEND) ||
+        (s->stctl & (SCSW_STCTL_PRIMARY |
+                     SCSW_STCTL_SECONDARY |
+                     SCSW_STCTL_ALERT))) {
+        ret = -EINPROGRESS;
+        goto out;
+    }
+
+    if (s->fctl & (SCSW_FCTL_HALT_FUNC | SCSW_FCTL_CLEAR_FUNC)) {
+        ret = -EBUSY;
+        goto out;
+    }
+
+    /* Trigger the halt function. */
+    s->fctl |= SCSW_FCTL_HALT_FUNC;
+    s->fctl &= ~SCSW_FCTL_START_FUNC;
+    if ((s->actl == (SCSW_ACTL_SUBCH_ACTIVE | SCSW_ACTL_DEVICE_ACTIVE)) &&
+        (s->stctl == SCSW_STCTL_INTERMEDIATE)) {
+        s->stctl &= ~SCSW_STCTL_STATUS_PEND;
+    }
+    s->actl |= SCSW_ACTL_HALT_PEND;
+
+    do_subchannel_work(sch);
+    ret = 0;
+
+out:
+    return ret;
+}
+
+static void css_update_chnmon(SubchDev *sch)
+{
+    if (!sch->curr_status.pmcw.mme) {
+        /* Not active. */
+        return;
+    }
+    if (sch->curr_status.pmcw.mbfc) {
+        /* Format 1, per-subchannel area. */
+        struct cmbe *cmbe;
+
+        cmbe = qemu_get_ram_ptr(sch->curr_status.mba);
+        if (cmbe) {
+            cmbe->ssch_rsch_count++;
+        }
+    } else {
+        /* Format 0, global area. */
+        struct cmb *cmb;
+        uint32_t offset;
+
+        offset = sch->curr_status.pmcw.mbi << 5;
+        cmb = qemu_get_ram_ptr(channel_subsys->chnmon_area + offset);
+        if (cmb) {
+            cmb->ssch_rsch_count++;
+        }
+    }
+}
+
+int css_do_ssch(SubchDev *sch, ORB *orb)
+{
+    SCSW *s = &sch->curr_status.scsw;
+    PMCW *p = &sch->curr_status.pmcw;
+    int ret;
+
+    if (!p->dnv || !p->ena) {
+        ret = -ENODEV;
+        goto out;
+    }
+
+    if (s->stctl & SCSW_STCTL_STATUS_PEND) {
+        ret = -EINPROGRESS;
+        goto out;
+    }
+
+    if (s->fctl & (SCSW_FCTL_START_FUNC |
+                   SCSW_FCTL_HALT_FUNC |
+                   SCSW_FCTL_CLEAR_FUNC)) {
+        ret = -EBUSY;
+        goto out;
+    }
+
+    /* If monitoring is active, update counter. */
+    if (channel_subsys->chnmon_active) {
+        css_update_chnmon(sch);
+    }
+    sch->orb = orb;
+    sch->channel_prog = qemu_get_ram_ptr(orb->cpa);
+    /* Trigger the start function. */
+    s->fctl |= SCSW_FCTL_START_FUNC;
+    s->actl |= SCSW_ACTL_START_PEND;
+    s->pno = 0;
+
+    do_subchannel_work(sch);
+    ret = 0;
+
+out:
+    return ret;
+}
+
+int css_do_tsch(SubchDev *sch, uint64_t addr)
+{
+    SCSW *s = &sch->curr_status.scsw;
+    PMCW *p = &sch->curr_status.pmcw;
+    uint8_t stctl;
+    uint8_t fctl;
+    uint8_t actl;
+    IRB irb;
+    int ret;
+    int i;
+    uint32_t offset = 0;
+
+    if (!p->dnv || !p->ena) {
+        ret = 3;
+        goto out;
+    }
+
+    stctl = s->stctl;
+    fctl = s->fctl;
+    actl = s->actl;
+
+    /* Prepare the irb for the guest. */
+    memset(&irb, 0, sizeof(IRB));
+
+    /* Copy scsw from current status. */
+    memcpy(&irb.scsw, s, sizeof(SCSW));
+    if (stctl & SCSW_STCTL_STATUS_PEND) {
+        if (s->cstat & (SCSW_CSTAT_DATA_CHECK |
+                        SCSW_CSTAT_CHN_CTRL_CHK |
+                        SCSW_CSTAT_INTF_CTRL_CHK)) {
+            irb.scsw.eswf = 1;
+            irb.esw[0] = 0x04804000;
+        } else {
+            irb.esw[0] = 0x00800000;
+        }
+        /* If a unit check is pending, copy sense data. */
+        if ((s->dstat & SCSW_DSTAT_UNIT_CHECK) && p->csense) {
+            irb.scsw.eswf = 1;
+            irb.scsw.ectl = 1;
+            memcpy(irb.ecw, sch->sense_data, sizeof(sch->sense_data));
+            irb.esw[1] = 0x02000000 | (sizeof(sch->sense_data) << 8);
+        }
+    }
+    /* Store the irb to the guest. */
+    css_write_phys_scsw(addr + offset, &irb.scsw);
+    offset += sizeof(SCSW);
+    for (i = 0; i < 5; i++) {
+        stl_phys(addr + offset, irb.esw[i]);
+        offset += sizeof(irb.esw[i]);
+    }
+    for (i = 0; i < 8; i++) {
+        stl_phys(addr + offset, irb.ecw[i]);
+        offset += sizeof(irb.ecw[i]);
+    }
+    for (i = 0; i < 8; i++) {
+        stl_phys(addr + offset, irb.emw[i]);
+        offset += sizeof(irb.emw[i]);
+    }
+
+    /* Clear conditions on subchannel, if applicable. */
+    if (stctl & SCSW_STCTL_STATUS_PEND) {
+        s->stctl = 0;
+        if ((stctl != (SCSW_STCTL_INTERMEDIATE | SCSW_STCTL_STATUS_PEND)) ||
+            ((fctl & SCSW_FCTL_HALT_FUNC) &&
+             (actl & SCSW_ACTL_SUSP))) {
+            s->fctl = 0;
+        }
+        if (stctl != (SCSW_STCTL_INTERMEDIATE | SCSW_STCTL_STATUS_PEND)) {
+            s->pno = 0;
+            s->actl &= ~(SCSW_ACTL_RESUME_PEND |
+                         SCSW_ACTL_START_PEND |
+                         SCSW_ACTL_HALT_PEND |
+                         SCSW_ACTL_CLEAR_PEND |
+                         SCSW_ACTL_SUSP);
+        } else {
+            if ((actl & SCSW_ACTL_SUSP) &&
+                (fctl & SCSW_FCTL_START_FUNC)) {
+                s->pno = 0;
+                if (fctl & SCSW_FCTL_HALT_FUNC) {
+                    s->actl &= ~(SCSW_ACTL_RESUME_PEND |
+                                 SCSW_ACTL_START_PEND |
+                                 SCSW_ACTL_HALT_PEND |
+                                 SCSW_ACTL_CLEAR_PEND |
+                                 SCSW_ACTL_SUSP);
+                } else {
+                    s->actl &= ~SCSW_ACTL_RESUME_PEND;
+                }
+            }
+        }
+        /* Clear pending sense data. */
+        if (p->csense) {
+            memset(sch->sense_data, 0 , sizeof(sch->sense_data));
+        }
+    }
+
+    ret = ((stctl & SCSW_STCTL_STATUS_PEND) == 0);
+
+out:
+    return ret;
+}
+
+int css_do_stcrw(uint64_t addr)
+{
+    CrwContainer *crw_cont;
+    int ret;
+
+    crw_cont = QTAILQ_FIRST(&channel_subsys->pending_crws);
+    if (crw_cont) {
+        QTAILQ_REMOVE(&channel_subsys->pending_crws, crw_cont, sibling);
+        stl_phys(addr, *(uint32_t *)&crw_cont->crw);
+        g_free(crw_cont);
+        ret = 0;
+    } else {
+        /* List was empty, turn crw machine checks on again. */
+        stl_phys(addr, 0);
+        channel_subsys->do_crw_mchk = true;
+        ret = 1;
+    }
+
+    return ret;
+}
+
+int css_do_tpi(uint64_t addr, int lowcore)
+{
+    /* No pending interrupts for !KVM. */
+    return 0;
+ }
+
+int css_collect_chp_desc(int m, uint8_t cssid, uint8_t f_chpid, uint8_t l_chpid,
+                         int rfmt, void *buf)
+{
+    int i, desc_size;
+    uint32_t words[8];
+    CssImage *css;
+
+    if (!m && !cssid) {
+        css = channel_subsys->css[channel_subsys->default_cssid];
+    } else {
+        css = channel_subsys->css[cssid];
+    }
+    if (!css) {
+        return 0;
+    }
+    desc_size = 0;
+    for (i = f_chpid; i <= l_chpid; i++) {
+        if (css->chpids[i].in_use) {
+            if (rfmt == 0) {
+                words[0] = 0x80000000 | (css->chpids[i].type << 8) | i;
+                words[1] = 0;
+                memcpy(buf + desc_size, words, 8);
+                desc_size += 8;
+            } else if (rfmt == 1) {
+                words[0] = 0x80000000 | (css->chpids[i].type << 8) | i;
+                words[1] = 0;
+                words[2] = 0;
+                words[3] = 0;
+                words[4] = 0;
+                words[5] = 0;
+                words[6] = 0;
+                words[7] = 0;
+                memcpy(buf + desc_size, words, 32);
+                desc_size += 32;
+            }
+        }
+    }
+    return desc_size;
+}
+
+void css_do_schm(uint8_t mbk, int update, int dct, uint64_t mbo)
+{
+    /* dct is currently ignored (not really meaningful for our devices) */
+    /* TODO: Don't ignore mbk. */
+    if (update && !channel_subsys->chnmon_active) {
+        /* Enable measuring. */
+        channel_subsys->chnmon_area = mbo;
+        channel_subsys->chnmon_active = true;
+    }
+    if (!update && channel_subsys->chnmon_active) {
+        /* Disable measuring. */
+        channel_subsys->chnmon_area = 0;
+        channel_subsys->chnmon_active = false;
+    }
+}
+
+int css_do_rsch(SubchDev *sch)
+{
+    SCSW *s = &sch->curr_status.scsw;
+    PMCW *p = &sch->curr_status.pmcw;
+    int ret;
+
+    if (!p->dnv || !p->ena) {
+        ret = -ENODEV;
+        goto out;
+    }
+
+    if (s->stctl & SCSW_STCTL_STATUS_PEND) {
+        ret = -EINPROGRESS;
+        goto out;
+    }
+
+    if ((s->fctl != SCSW_FCTL_START_FUNC) ||
+        (s->actl & SCSW_ACTL_RESUME_PEND) ||
+        (!(s->actl & SCSW_ACTL_SUSP))) {
+        ret = -EINVAL;
+        goto out;
+    }
+
+    /* If monitoring is active, update counter. */
+    if (channel_subsys->chnmon_active) {
+        css_update_chnmon(sch);
+    }
+
+    s->actl |= SCSW_ACTL_RESUME_PEND;
+    do_subchannel_work(sch);
+    ret = 0;
+
+out:
+    return ret;
+}
+
+int css_do_rchp(uint8_t cssid, uint8_t chpid)
+{
+    uint8_t real_cssid;
+
+    if (cssid > channel_subsys->max_cssid) {
+        return -EINVAL;
+    }
+    if (channel_subsys->max_cssid == 0) {
+        real_cssid = channel_subsys->default_cssid;
+    } else {
+        real_cssid = cssid;
+    }
+    if (!channel_subsys->css[real_cssid]) {
+        return -EINVAL;
+    }
+
+    if (!channel_subsys->css[real_cssid]->chpids[chpid].in_use) {
+        return -ENODEV;
+    }
+
+    if (!channel_subsys->css[real_cssid]->chpids[chpid].is_virtual) {
+        fprintf(stderr,
+                "rchp unsupported for non-virtual chpid %x.%02x!\n",
+                real_cssid, chpid);
+        return -ENODEV;
+    }
+
+    /* We don't really use a channel path, so we're done here. */
+    css_queue_crw(CRW_RSC_CHP, CRW_ERC_INIT,
+                  channel_subsys->max_cssid > 0 ? 1 : 0, chpid);
+    if (channel_subsys->max_cssid > 0) {
+        css_queue_crw(CRW_RSC_CHP, CRW_ERC_INIT, 0, real_cssid << 8);
+    }
+    return 0;
+}
+
+bool css_schid_final(uint8_t cssid, uint8_t ssid, uint16_t schid)
+{
+    SubchSet *set;
+
+    if (cssid > MAX_CSSID || ssid > MAX_SSID || !channel_subsys->css[cssid] ||
+        !channel_subsys->css[cssid]->sch_set[ssid]) {
+        return true;
+    }
+    set = channel_subsys->css[cssid]->sch_set[ssid];
+    return schid > find_last_bit(set->schids_used,
+                                 (MAX_SCHID + 1) / sizeof(unsigned long));
+}
+
+static int css_add_virtual_chpid(uint8_t cssid, uint8_t chpid, uint8_t type)
+{
+    CssImage *css;
+
+    if (cssid > MAX_CSSID) {
+        return -EINVAL;
+    }
+    css = channel_subsys->css[cssid];
+    if (!css) {
+        return -EINVAL;
+    }
+    if (css->chpids[chpid].in_use) {
+        return -EEXIST;
+    }
+    css->chpids[chpid].in_use = 1;
+    css->chpids[chpid].type = type;
+    css->chpids[chpid].is_virtual = 1;
+
+    css_generate_chp_crws(cssid, chpid);
+
+    return 0;
+}
+
+void css_sch_build_virtual_schib(SubchDev *sch, uint8_t chpid, uint8_t type)
+{
+    PMCW *p = &sch->curr_status.pmcw;
+    SCSW *s = &sch->curr_status.scsw;
+    int i;
+    CssImage *css = channel_subsys->css[sch->cssid];
+
+    assert(css != NULL);
+    memset(p, 0, sizeof(PMCW));
+    p->dnv = 1;
+    p->dev = sch->devno;
+    /* single path */
+    p->pim = 0x80;
+    p->pom = 0xff;
+    p->pam = 0x80;
+    p->chpid[0] = chpid;
+    if (!css->chpids[chpid].in_use) {
+        css_add_virtual_chpid(sch->cssid, chpid, type);
+    }
+
+    memset(s, 0, sizeof(SCSW));
+    sch->curr_status.mba = 0;
+    for (i = 0; i < 4; i++) {
+        sch->curr_status.mda[i] = 0;
+    }
+}
+
+SubchDev *css_find_subch(uint8_t m, uint8_t cssid, uint8_t ssid, uint16_t schid)
+{
+    uint8_t real_cssid;
+
+    real_cssid = (!m && (cssid == 0)) ? channel_subsys->default_cssid : cssid;
+
+    if (!channel_subsys->css[real_cssid]) {
+        return NULL;
+    }
+
+    if (!channel_subsys->css[real_cssid]->sch_set[ssid]) {
+        return NULL;
+    }
+
+    return channel_subsys->css[real_cssid]->sch_set[ssid]->sch[schid];
+}
+
+bool css_subch_visible(SubchDev *sch)
+{
+    if (sch->ssid > channel_subsys->max_ssid) {
+        return false;
+    }
+
+    if (sch->cssid != channel_subsys->default_cssid) {
+        return (channel_subsys->max_cssid > 0);
+    }
+
+    return true;
+}
+
+bool css_present(uint8_t cssid)
+{
+    return (channel_subsys->css[cssid] != NULL);
+}
+
+bool css_devno_used(uint8_t cssid, uint8_t ssid, uint16_t devno)
+{
+    if (!channel_subsys->css[cssid]) {
+        return false;
+    }
+    if (!channel_subsys->css[cssid]->sch_set[ssid]) {
+        return false;
+    }
+
+    return !!test_bit(devno,
+                      channel_subsys->css[cssid]->sch_set[ssid]->devnos_used);
+}
+
+void css_subch_assign(uint8_t cssid, uint8_t ssid, uint16_t schid,
+                      uint16_t devno, SubchDev *sch)
+{
+    CssImage *css;
+    SubchSet *s_set;
+
+    if (!channel_subsys->css[cssid]) {
+        fprintf(stderr,
+                "Suspicious call to %s (%x.%x.%04x) for non-existing css!\n",
+                __func__, cssid, ssid, schid);
+        return;
+    }
+    css = channel_subsys->css[cssid];
+
+    if (!css->sch_set[ssid]) {
+        css->sch_set[ssid] = g_malloc0(sizeof(SubchSet));
+    }
+    s_set = css->sch_set[ssid];
+
+    s_set->sch[schid] = sch;
+    if (sch) {
+        set_bit(schid, s_set->schids_used);
+        set_bit(devno, s_set->devnos_used);
+    } else {
+        clear_bit(schid, s_set->schids_used);
+        clear_bit(schid, s_set->devnos_used);
+    }
+}
+
+void css_queue_crw(uint8_t rsc, uint8_t erc, int chain, uint16_t rsid)
+{
+    CrwContainer *crw_cont;
+
+    /* TODO: Maybe use a static crw pool? */
+    crw_cont = g_try_malloc0(sizeof(CrwContainer));
+    if (!crw_cont) {
+        channel_subsys->crws_lost = true;
+        return;
+    }
+    crw_cont->crw.rsc = rsc;
+    crw_cont->crw.erc = erc;
+    crw_cont->crw.c = chain;
+    crw_cont->crw.rsid = rsid;
+    crw_cont->crw.r = channel_subsys->crws_lost ? 1 : 0;
+    channel_subsys->crws_lost = false;
+
+    QTAILQ_INSERT_TAIL(&channel_subsys->pending_crws, crw_cont, sibling);
+
+    if (channel_subsys->do_crw_mchk) {
+        S390CPU *cpu = s390_cpu_addr2state(0);
+
+        channel_subsys->do_crw_mchk = false;
+        /* Inject crw pending machine check. */
+        s390_crw_mchk(&cpu->env);
+    }
+}
+
+void css_generate_sch_crws(uint8_t cssid, uint8_t ssid, uint16_t schid,
+                           int hotplugged, int add)
+{
+    uint8_t guest_cssid;
+    bool chain_crw;
+
+    if (add && !hotplugged) {
+        return;
+    }
+    if (channel_subsys->max_cssid == 0) {
+        /* Default cssid shows up as 0. */
+        guest_cssid = (cssid == channel_subsys->default_cssid) ? 0 : cssid;
+    } else {
+        /* Show real cssid to the guest. */
+        guest_cssid = cssid;
+    }
+    /*
+     * Only notify for higher subchannel sets/channel subsystems if the
+     * guest has enabled it.
+     */
+    if ((ssid > channel_subsys->max_ssid) ||
+        (guest_cssid > channel_subsys->max_cssid) ||
+        ((channel_subsys->max_cssid == 0) &&
+         (cssid != channel_subsys->default_cssid))) {
+        return;
+    }
+    chain_crw = (channel_subsys->max_ssid > 0) ||
+            (channel_subsys->max_cssid > 0);
+    css_queue_crw(CRW_RSC_SUBCH, CRW_ERC_IPI, chain_crw ? 1 : 0, schid);
+    if (chain_crw) {
+        css_queue_crw(CRW_RSC_SUBCH, CRW_ERC_IPI, 0,
+                      (guest_cssid << 8) | (ssid << 4));
+    }
+}
+
+void css_generate_chp_crws(uint8_t cssid, uint8_t chpid)
+{
+    /* TODO */
+}
+
+int css_enable_mcsse(void)
+{
+    channel_subsys->max_cssid = MAX_CSSID;
+    return 0;
+}
+
+int css_enable_mss(void)
+{
+    channel_subsys->max_ssid = MAX_SSID;
+    return 0;
+}
+
+static void css_init(void)
+{
+    channel_subsys = g_malloc0(sizeof(*channel_subsys));
+    QTAILQ_INIT(&channel_subsys->pending_crws);
+    channel_subsys->do_crw_mchk = true;
+    channel_subsys->crws_lost = false;
+    channel_subsys->chnmon_active = false;
+}
+machine_init(css_init);
+
+void css_reset_sch(SubchDev *sch)
+{
+    PMCW *p = &sch->curr_status.pmcw;
+
+    p->intparm = 0;
+    p->isc = 0;
+    p->ena = 0;
+    p->lm = 0;
+    p->mme = 0;
+    p->mp = 0;
+    p->tf = 0;
+    p->dnv = 1;
+    p->dev = sch->devno;
+    p->pim = 0x80;
+    p->lpm = p->pim;
+    p->pnom = 0;
+    p->lpum = 0;
+    p->mbi = 0;
+    p->pom = 0xff;
+    p->pam = 0x80;
+    p->mbfc = 0;
+    p->xmwme = 0;
+    p->csense = 0;
+
+    memset(&sch->curr_status.scsw, 0, sizeof(sch->curr_status.scsw));
+    sch->curr_status.mba = 0;
+
+    sch->channel_prog = NULL;
+    sch->last_cmd = NULL;
+    sch->orb = NULL;
+}
+
+void css_reset(void)
+{
+    CrwContainer *crw_cont;
+
+    /* Clean up monitoring. */
+    channel_subsys->chnmon_active = false;
+    channel_subsys->chnmon_area = 0;
+
+    /* Clear pending CRWs. */
+    while ((crw_cont = QTAILQ_FIRST(&channel_subsys->pending_crws))) {
+        QTAILQ_REMOVE(&channel_subsys->pending_crws, crw_cont, sibling);
+        g_free(crw_cont);
+    }
+    channel_subsys->do_crw_mchk = true;
+    channel_subsys->crws_lost = false;
+
+    /* Reset maximum ids. */
+    channel_subsys->max_cssid = 0;
+    channel_subsys->max_ssid = 0;
+}
diff --git a/hw/s390x/css.h b/hw/s390x/css.h
new file mode 100644
index 0000000..638b801
--- /dev/null
+++ b/hw/s390x/css.h
@@ -0,0 +1,90 @@
+/*
+ * Channel subsystem structures and definitions.
+ *
+ * Copyright 2012 IBM Corp.
+ * Author(s): Cornelia Huck <cornelia.huck@xxxxxxxxxx>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or (at
+ * your option) any later version. See the COPYING file in the top-level
+ * directory.
+ */
+
+#ifndef CSS_H
+#define CSS_H
+
+#include "ioinst.h"
+
+/* Channel subsystem constants. */
+#define MAX_SCHID 65535
+#define MAX_SSID 3
+#define MAX_CSSID 254 /* 255 is reserved */
+#define MAX_CHPID 255
+
+#define MAX_CIWS 62
+
+typedef struct SenseId {
+    /* common part */
+    uint8_t reserved;        /* always 0x'FF' */
+    uint16_t cu_type;        /* control unit type */
+    uint8_t cu_model;        /* control unit model */
+    uint16_t dev_type;       /* device type */
+    uint8_t dev_model;       /* device model */
+    uint8_t unused;          /* padding byte */
+    /* extended part */
+    uint32_t ciw[MAX_CIWS];  /* variable # of CIWs */
+} QEMU_PACKED SenseId;
+
+/* Channel measurements, from linux/drivers/s390/cio/cmf.c. */
+struct cmb {
+    uint16_t ssch_rsch_count;
+    uint16_t sample_count;
+    uint32_t device_connect_time;
+    uint32_t function_pending_time;
+    uint32_t device_disconnect_time;
+    uint32_t control_unit_queuing_time;
+    uint32_t device_active_only_time;
+    uint32_t reserved[2];
+};
+
+struct cmbe {
+    uint32_t ssch_rsch_count;
+    uint32_t sample_count;
+    uint32_t device_connect_time;
+    uint32_t function_pending_time;
+    uint32_t device_disconnect_time;
+    uint32_t control_unit_queuing_time;
+    uint32_t device_active_only_time;
+    uint32_t device_busy_time;
+    uint32_t initial_command_response_time;
+    uint32_t reserved[7];
+};
+
+struct SubchDev {
+    /* channel-subsystem related things: */
+    uint8_t cssid;
+    uint8_t ssid;
+    uint16_t schid;
+    uint16_t devno;
+    SCHIB curr_status;
+    uint8_t sense_data[32];
+    CCW1 *channel_prog;
+    CCW1 *last_cmd;
+    ORB *orb;
+    /* transport-provided data: */
+    int (*ccw_cb) (SubchDev *, CCW1 *);
+    SenseId id;
+    void *driver_data;
+};
+
+typedef SubchDev *(*css_subch_cb_func)(uint8_t m, uint8_t cssid, uint8_t ssid,
+                                       uint16_t schid);
+int css_create_css_image(uint8_t cssid, bool default_image);
+bool css_devno_used(uint8_t cssid, uint8_t ssid, uint16_t devno);
+void css_subch_assign(uint8_t cssid, uint8_t ssid, uint16_t schid,
+                      uint16_t devno, SubchDev *sch);
+void css_sch_build_virtual_schib(SubchDev *sch, uint8_t chpid, uint8_t type);
+void css_reset(void);
+void css_reset_sch(SubchDev *sch);
+void css_queue_crw(uint8_t rsc, uint8_t erc, int chain, uint16_t rsid);
+
+#endif
diff --git a/target-s390x/Makefile.objs b/target-s390x/Makefile.objs
index e728abf..3afb0b7 100644
--- a/target-s390x/Makefile.objs
+++ b/target-s390x/Makefile.objs
@@ -1,4 +1,4 @@
 obj-y += translate.o helper.o cpu.o interrupt.o
 obj-y += int_helper.o fpu_helper.o cc_helper.o mem_helper.o misc_helper.o
-obj-$(CONFIG_SOFTMMU) += machine.o
+obj-$(CONFIG_SOFTMMU) += machine.o ioinst.o
 obj-$(CONFIG_KVM) += kvm.o
diff --git a/target-s390x/cpu.h b/target-s390x/cpu.h
index 5be6e83..ecf44cd 100644
--- a/target-s390x/cpu.h
+++ b/target-s390x/cpu.h
@@ -47,6 +47,11 @@
 #define MMU_USER_IDX 1
 
 #define MAX_EXT_QUEUE 16
+#define MAX_IO_QUEUE 16
+#define MAX_MCHK_QUEUE 16
+
+#define PSW_MCHK_MASK 0x0004000000000000
+#define PSW_IO_MASK 0x0200000000000000
 
 typedef struct PSW {
     uint64_t mask;
@@ -59,6 +64,17 @@ typedef struct ExtQueue {
     uint32_t param64;
 } ExtQueue;
 
+typedef struct IOQueue {
+    uint16_t id;
+    uint16_t nr;
+    uint32_t parm;
+    uint32_t word;
+} IOQueue;
+
+typedef struct MchkQueue {
+    uint16_t type;
+} MchkQueue;
+
 typedef struct CPUS390XState {
     uint64_t regs[16];	/* GP registers */
 
@@ -88,8 +104,16 @@ typedef struct CPUS390XState {
 
     int pending_int;
     ExtQueue ext_queue[MAX_EXT_QUEUE];
+    IOQueue io_queue[MAX_IO_QUEUE][8];
+    MchkQueue mchk_queue[MAX_MCHK_QUEUE];
 
     int ext_index;
+    int io_index[8];
+    int mchk_index;
+
+    uint64_t ckc;
+    uint64_t cputm;
+    uint32_t todpr;
 
     CPU_COMMON
 
@@ -103,6 +127,8 @@ typedef struct CPUS390XState {
     QEMUTimer *tod_timer;
 
     QEMUTimer *cpu_timer;
+
+    void *chsc_page;
 } CPUS390XState;
 
 #include "cpu-qom.h"
@@ -339,6 +365,112 @@ static inline unsigned s390_del_running_cpu(CPUS390XState *env)
 void cpu_lock(void);
 void cpu_unlock(void);
 
+typedef struct SubchDev SubchDev;
+typedef struct SCHIB SCHIB;
+typedef struct ORB ORB;
+
+#ifndef CONFIG_USER_ONLY
+SubchDev *css_find_subch(uint8_t m, uint8_t cssid, uint8_t ssid,
+                         uint16_t schid);
+bool css_subch_visible(SubchDev *sch);
+void css_conditional_io_interrupt(SubchDev *sch);
+int css_do_stsch(SubchDev *sch, uint64_t addr);
+bool css_schid_final(uint8_t cssid, uint8_t ssid, uint16_t schid);
+int css_do_msch(SubchDev *sch, SCHIB *schib);
+int css_do_xsch(SubchDev *sch);
+int css_do_csch(SubchDev *sch);
+int css_do_hsch(SubchDev *sch);
+int css_do_ssch(SubchDev *sch, ORB *orb);
+int css_do_tsch(SubchDev *sch, uint64_t addr);
+int css_do_stcrw(uint64_t addr);
+int css_do_tpi(uint64_t addr, int lowcore);
+int css_collect_chp_desc(int m, uint8_t cssid, uint8_t f_chpid, uint8_t l_chpid,
+                         int rfmt, void *buf);
+void css_do_schm(uint8_t mbk, int update, int dct, uint64_t mbo);
+int css_enable_mcsse(void);
+int css_enable_mss(void);
+int css_do_rsch(SubchDev *sch);
+int css_do_rchp(uint8_t cssid, uint8_t chpid);
+bool css_present(uint8_t cssid);
+#else
+static inline SubchDev *css_find_subch(uint8_t m, uint8_t cssid, uint8_t ssid,
+                                       uint16_t schid)
+{
+    return NULL;
+}
+static inline bool css_subch_visible(SubchDev *sch)
+{
+    return false;
+}
+static inline void css_conditional_io_interrupt(SubchDev *sch)
+{
+}
+static inline int css_do_stsch(SubchDev *sch, uint64_t addr)
+{
+    return -ENODEV;
+}
+static inline bool css_schid_final(uint8_t cssid, uint8_t ssid, uint16_t schid)
+{
+    return true;
+}
+static inline int css_do_msch(SubchDev *sch, SCHIB *schib)
+{
+    return -ENODEV;
+}
+static inline int css_do_xsch(SubchDev *sch)
+{
+    return -ENODEV;
+}
+static inline int css_do_csch(SubchDev *sch)
+{
+    return -ENODEV;
+}
+static inline int css_do_hsch(SubchDev *sch)
+{
+    return -ENODEV;
+}
+static inline int css_do_ssch(SubchDev *sch, ORB *orb)
+{
+    return -ENODEV;
+}
+static inline int css_do_tsch(SubchDev *sch, uint64_t addr)
+{
+    return -ENODEV;
+}
+static inline int css_do_stcrw(uint64_t addr)
+{
+    return 1;
+}
+static inline int css_do_tpi(uint64_t addr, int lowcore)
+{
+    return 0;
+}
+static inline int css_collect_chp_desc(int m, uint8_t cssid, uint8_t f_chpid,
+                                       int rfmt, uint8_t l_chpid, void *buf)
+{
+    return 0;
+}
+static inline void css_do_schm(uint8_t mbk, int update, int dct, uint64_t mbo)
+{
+}
+static inline int css_enable_mss(void)
+{
+    return -EINVAL;
+}
+static inline int css_do_rsch(SubchDev *sch)
+{
+    return -ENODEV;
+}
+static inline int css_do_rchp(uint8_t cssid, uint8_t chpid)
+{
+    return -ENODEV;
+}
+static inline bool css_present(uint8_t cssid)
+{
+    return false;
+}
+#endif
+
 static inline void cpu_set_tls(CPUS390XState *env, target_ulong newtls)
 {
     env->aregs[0] = newtls >> 32;
@@ -364,12 +496,16 @@ static inline void cpu_set_tls(CPUS390XState *env, target_ulong newtls)
 #define EXCP_EXT 1 /* external interrupt */
 #define EXCP_SVC 2 /* supervisor call (syscall) */
 #define EXCP_PGM 3 /* program interruption */
+#define EXCP_IO  7 /* I/O interrupt */
+#define EXCP_MCHK 8 /* machine check */
 
 #endif /* CONFIG_USER_ONLY */
 
 #define INTERRUPT_EXT        (1 << 0)
 #define INTERRUPT_TOD        (1 << 1)
 #define INTERRUPT_CPUTIMER   (1 << 2)
+#define INTERRUPT_IO         (1 << 3)
+#define INTERRUPT_MCHK       (1 << 4)
 
 /* Program Status Word.  */
 #define S390_PSWM_REGNUM 0
@@ -977,6 +1113,45 @@ static inline void cpu_inject_ext(CPUS390XState *env, uint32_t code, uint32_t pa
     cpu_interrupt(env, CPU_INTERRUPT_HARD);
 }
 
+static inline void cpu_inject_io(CPUS390XState *env, uint16_t subchannel_id,
+                                 uint16_t subchannel_number,
+                                 uint32_t io_int_parm, uint32_t io_int_word)
+{
+    int isc = ffs(io_int_word << 2) - 1;
+
+    if (env->io_index[isc] == MAX_IO_QUEUE - 1) {
+        /* ugh - can't queue anymore. Let's drop. */
+        return;
+    }
+
+    env->io_index[isc]++;
+    assert(env->io_index[isc] < MAX_IO_QUEUE);
+
+    env->io_queue[env->io_index[isc]][isc].id = subchannel_id;
+    env->io_queue[env->io_index[isc]][isc].nr = subchannel_number;
+    env->io_queue[env->io_index[isc]][isc].parm = io_int_parm;
+    env->io_queue[env->io_index[isc]][isc].word = io_int_word;
+
+    env->pending_int |= INTERRUPT_IO;
+    cpu_interrupt(env, CPU_INTERRUPT_HARD);
+}
+
+static inline void cpu_inject_crw_mchk(CPUS390XState *env)
+{
+    if (env->mchk_index == MAX_MCHK_QUEUE - 1) {
+        /* ugh - can't queue anymore. Let's drop. */
+        return;
+    }
+
+    env->mchk_index++;
+    assert(env->mchk_index < MAX_MCHK_QUEUE);
+
+    env->mchk_queue[env->mchk_index].type = 1;
+
+    env->pending_int |= INTERRUPT_MCHK;
+    cpu_interrupt(env, CPU_INTERRUPT_HARD);
+}
+
 static inline bool cpu_has_work(CPUS390XState *env)
 {
     return (env->interrupt_request & CPU_INTERRUPT_HARD) &&
@@ -996,5 +1171,62 @@ uint32_t set_cc_nz_f64(float64 v);
 
 /* misc_helper.c */
 void program_interrupt(CPUS390XState *env, uint32_t code, int ilc);
+int css_handle_sch_io(uint32_t sch_id, uint8_t func, uint64_t orb, void *scsw,
+                      void *pmcw);
+void css_generate_sch_crws(uint8_t cssid, uint8_t ssid, uint16_t schid,
+                           int hotplugged, int add);
+void css_generate_chp_crws(uint8_t cssid, uint8_t chpid);
+void css_inject_io(uint8_t cssid, uint8_t ssid, uint16_t schid, uint8_t isc,
+                   uint32_t intparm, int unsolicited);
+#ifdef CONFIG_KVM
+int kvm_s390_io_interrupt(CPUS390XState *env, uint16_t subchannel_id,
+                          uint16_t subchannel_nr, uint32_t io_int_parm,
+                          uint32_t io_int_word);
+int kvm_s390_crw_mchk(CPUS390XState *env);
+void kvm_s390_enable_css_support(CPUS390XState *env);
+#else
+static inline int kvm_s390_io_interrupt(CPUS390XState *env,
+                                        uint16_t subchannel_id,
+                                        uint16_t subchannel_nr,
+                                        uint32_t io_int_parm,
+                                        uint32_t io_int_word)
+{
+    return -EOPNOTSUPP;
+}
+static inline int kvm_s390_crw_mchk(CPUS390XState *env)
+{
+    return -EOPNOTSUPP;
+}
+static inline void kvm_s390_enable_css_support(CPUS390XState *env)
+{
+}
+#endif
+
+static inline void s390_io_interrupt(CPUS390XState *env,
+                                     uint16_t subchannel_id,
+                                     uint16_t subchannel_nr,
+                                     uint32_t io_int_parm,
+                                     uint32_t io_int_word)
+{
+    int ret;
+
+    ret = kvm_s390_io_interrupt(env, subchannel_id, subchannel_nr, io_int_parm,
+                                io_int_word);
+    if (ret == -EOPNOTSUPP) {
+        cpu_inject_io(env, subchannel_id, subchannel_nr, io_int_parm,
+                      io_int_word);
+    }
+}
+
+static inline void s390_crw_mchk(CPUS390XState *env)
+{
+    int ret;
+
+    ret = kvm_s390_crw_mchk(env);
+
+    if (ret == -EOPNOTSUPP) {
+            cpu_inject_crw_mchk(env);
+    }
+}
 
 #endif
diff --git a/target-s390x/helper.c b/target-s390x/helper.c
index b7b812a..8e3930a 100644
--- a/target-s390x/helper.c
+++ b/target-s390x/helper.c
@@ -574,12 +574,145 @@ static void do_ext_interrupt(CPUS390XState *env)
     load_psw(env, mask, addr);
 }
 
+static void do_io_interrupt(CPUS390XState *env)
+{
+    uint64_t mask, addr;
+    LowCore *lowcore;
+    hwaddr len = TARGET_PAGE_SIZE;
+    IOQueue *q;
+    uint8_t isc;
+    int disable = 1;
+    int found = 0;
+
+    if (!(env->psw.mask & PSW_MASK_IO)) {
+        cpu_abort(env, "I/O int w/o I/O mask\n");
+    }
+
+
+    for (isc = 0; isc < 8; isc++) {
+        if (env->io_index[isc] < 0) {
+            continue;
+        }
+        if (env->io_index[isc] > MAX_IO_QUEUE) {
+            cpu_abort(env, "I/O queue overrun for isc %d: %d\n",
+                      isc, env->io_index[isc]);
+        }
+
+        q = &env->io_queue[env->io_index[isc]][isc];
+        if (!(env->cregs[6] & q->word)) {
+            disable = 0;
+            continue;
+        }
+        found = 1;
+        lowcore = cpu_physical_memory_map(env->psa, &len, 1);
+
+        lowcore->subchannel_id = cpu_to_be16(q->id);
+        lowcore->subchannel_nr = cpu_to_be16(q->nr);
+        lowcore->io_int_parm = cpu_to_be32(q->parm);
+        lowcore->io_int_word = cpu_to_be32(q->word);
+        lowcore->io_old_psw.mask = cpu_to_be64(get_psw_mask(env));
+        lowcore->io_old_psw.addr = cpu_to_be64(env->psw.addr);
+        mask = be64_to_cpu(lowcore->io_new_psw.mask);
+        addr = be64_to_cpu(lowcore->io_new_psw.addr);
+
+        cpu_physical_memory_unmap(lowcore, len, 1, len);
+
+        env->io_index[isc]--;
+        if (env->io_index >= 0) {
+            disable = 0;
+        }
+        break;
+    }
+
+    if (disable) {
+        env->pending_int &= ~INTERRUPT_IO;
+    }
+    if (found) {
+        DPRINTF("%s: %" PRIx64 " %" PRIx64 "\n", __func__,
+                env->psw.mask, env->psw.addr);
+
+        load_psw(env, mask, addr);
+    }
+}
+
+static void do_mchk_interrupt(CPUS390XState *env)
+{
+    uint64_t mask, addr;
+    LowCore *lowcore;
+    hwaddr len = TARGET_PAGE_SIZE;
+    MchkQueue *q;
+    int i;
+
+    if (!(env->psw.mask & PSW_MASK_MCHECK)) {
+        cpu_abort(env, "Machine check w/o mchk mask\n");
+    }
+
+    if (env->mchk_index < 0 || env->mchk_index > MAX_MCHK_QUEUE) {
+        cpu_abort(env, "Mchk queue overrun: %d\n", env->mchk_index);
+    }
+
+    q = &env->mchk_queue[env->mchk_index];
+
+    if (q->type != 1) {
+        /* Don't know how to handle this... */
+        cpu_abort(env, "Unknown machine check type %d\n", q->type);
+    }
+    if (!(env->cregs[14] & (1 << 28))) {
+        /* CRW machine checks disabled */
+        return;
+    }
+
+    lowcore = cpu_physical_memory_map(env->psa, &len, 1);
+
+    for (i = 0; i < 16; i++) {
+        lowcore->floating_pt_save_area[i] = cpu_to_be64(env->fregs[i].ll);
+        lowcore->gpregs_save_area[i] = cpu_to_be64(env->regs[i]);
+        lowcore->access_regs_save_area[i] = cpu_to_be32(env->aregs[i]);
+        lowcore->cregs_save_area[i] = cpu_to_be64(env->cregs[i]);
+    }
+    lowcore->prefixreg_save_area = cpu_to_be32(env->psa);
+    lowcore->fpt_creg_save_area = cpu_to_be32(env->fpc);
+    lowcore->tod_progreg_save_area = cpu_to_be32(env->todpr);
+    lowcore->cpu_timer_save_area[0] = cpu_to_be32(env->cputm >> 32);
+    lowcore->cpu_timer_save_area[1] =
+        cpu_to_be32(env->cputm & 0x00000000ffffffff);
+    lowcore->clock_comp_save_area[0] = cpu_to_be32(env->ckc >> 32);
+    lowcore->clock_comp_save_area[1] =
+        cpu_to_be32(env->ckc & 0x00000000ffffffff);
+
+    lowcore->mcck_interruption_code[0] = cpu_to_be32(0x00400f1d);
+    lowcore->mcck_interruption_code[1] = cpu_to_be32(0x40330000);
+    lowcore->mcck_old_psw.mask = cpu_to_be64(get_psw_mask(env));
+    lowcore->mcck_old_psw.addr = cpu_to_be64(env->psw.addr);
+    mask = be64_to_cpu(lowcore->mcck_new_psw.mask);
+    addr = be64_to_cpu(lowcore->mcck_new_psw.addr);
+
+    cpu_physical_memory_unmap(lowcore, len, 1, len);
+
+    env->mchk_index--;
+    if (env->mchk_index == -1) {
+        env->pending_int &= ~INTERRUPT_MCHK;
+    }
+
+    DPRINTF("%s: %" PRIx64 " %" PRIx64 "\n", __func__,
+            env->psw.mask, env->psw.addr);
+
+    load_psw(env, mask, addr);
+}
+
 void do_interrupt(CPUS390XState *env)
 {
     qemu_log_mask(CPU_LOG_INT, "%s: %d at pc=%" PRIx64 "\n",
                   __func__, env->exception_index, env->psw.addr);
 
     s390_add_running_cpu(env);
+    /* handle machine checks */
+    if ((env->psw.mask & PSW_MASK_MCHECK) &&
+        (env->exception_index == -1)) {
+        if (env->pending_int & INTERRUPT_MCHK) {
+            env->exception_index = EXCP_MCHK;
+        }
+    }
     /* handle external interrupts */
     if ((env->psw.mask & PSW_MASK_EXT) &&
         env->exception_index == -1) {
@@ -598,6 +731,13 @@ void do_interrupt(CPUS390XState *env)
             env->pending_int &= ~INTERRUPT_TOD;
         }
     }
+    /* handle I/O interrupts */
+    if ((env->psw.mask & PSW_MASK_IO) &&
+        (env->exception_index == -1)) {
+        if (env->pending_int & INTERRUPT_IO) {
+            env->exception_index = EXCP_IO;
+        }
+    }
 
     switch (env->exception_index) {
     case EXCP_PGM:
@@ -609,6 +749,12 @@ void do_interrupt(CPUS390XState *env)
     case EXCP_EXT:
         do_ext_interrupt(env);
         break;
+    case EXCP_IO:
+        do_io_interrupt(env);
+        break;
+    case EXCP_MCHK:
+        do_mchk_interrupt(env);
+        break;
     }
     env->exception_index = -1;
 
diff --git a/target-s390x/ioinst.c b/target-s390x/ioinst.c
new file mode 100644
index 0000000..6356681
--- /dev/null
+++ b/target-s390x/ioinst.c
@@ -0,0 +1,737 @@
+/*
+ * I/O instructions for S/390
+ *
+ * Copyright 2012 IBM Corp.
+ * Author(s): Cornelia Huck <cornelia.huck@xxxxxxxxxx>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or (at
+ * your option) any later version. See the COPYING file in the top-level
+ * directory.
+ */
+
+#include <sys/types.h>
+#include <sys/ioctl.h>
+#include <sys/mman.h>
+
+#include "cpu.h"
+#include "ioinst.h"
+
+#ifdef DEBUG_IOINST
+#define dprintf(fmt, ...) \
+    do { fprintf(stderr, fmt, ## __VA_ARGS__); } while (0)
+#else
+#define dprintf(fmt, ...) \
+    do { } while (0)
+#endif
+
+/* Special handling for the prefix page. */
+static void *s390_get_address(CPUS390XState *env, ram_addr_t guest_addr)
+{
+    if (guest_addr < 8192) {
+        guest_addr += env->psa;
+    } else if ((env->psa <= guest_addr) && (guest_addr < env->psa + 8192)) {
+        guest_addr -= env->psa;
+    }
+
+    return qemu_get_ram_ptr(guest_addr);
+}
+
+int ioinst_disassemble_sch_ident(uint32_t value, int *m, int *cssid, int *ssid,
+                                 int *schid)
+{
+    if (!(value & IOINST_SCHID_ONE)) {
+        return -EINVAL;
+    }
+    if (!(value & IOINST_SCHID_M)) {
+        if (value & IOINST_SCHID_CSSID) {
+            return -EINVAL;
+        }
+        *cssid = 0;
+        *m = 0;
+    } else {
+        *cssid = (value & IOINST_SCHID_CSSID) >> 24;
+        *m = 1;
+    }
+    *ssid = (value & IOINST_SCHID_SSID) >> 17;
+    *schid = value & IOINST_SCHID_NR;
+    return 0;
+}
+
+int ioinst_handle_xsch(CPUS390XState *env, uint64_t reg1)
+{
+    int cssid, ssid, schid, m;
+    SubchDev *sch;
+    int ret = -ENODEV;
+    int cc;
+
+    if (ioinst_disassemble_sch_ident(reg1, &m, &cssid, &ssid, &schid)) {
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+    dprintf("IOINST: xsch (%x.%x.%04x)\n", cssid, ssid, schid);
+    sch = css_find_subch(m, cssid, ssid, schid);
+    if (sch && css_subch_visible(sch)) {
+        ret = css_do_xsch(sch);
+    }
+    switch (ret) {
+    case -ENODEV:
+        cc = 3;
+        break;
+    case -EBUSY:
+        cc = 2;
+        break;
+    case 0:
+        cc = 0;
+        break;
+    default:
+        cc = 1;
+        break;
+    }
+
+    return cc;
+}
+
+int ioinst_handle_csch(CPUS390XState *env, uint64_t reg1)
+{
+    int cssid, ssid, schid, m;
+    SubchDev *sch;
+    int ret = -ENODEV;
+    int cc;
+
+    if (ioinst_disassemble_sch_ident(reg1, &m, &cssid, &ssid, &schid)) {
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+    dprintf("IOINST: csch (%x.%x.%04x)\n", cssid, ssid, schid);
+    sch = css_find_subch(m, cssid, ssid, schid);
+    if (sch && css_subch_visible(sch)) {
+        ret = css_do_csch(sch);
+    }
+    if (ret == -ENODEV) {
+        cc = 3;
+    } else {
+        cc = 0;
+    }
+    return cc;
+}
+
+int ioinst_handle_hsch(CPUS390XState *env, uint64_t reg1)
+{
+    int cssid, ssid, schid, m;
+    SubchDev *sch;
+    int ret = -ENODEV;
+    int cc;
+
+    if (ioinst_disassemble_sch_ident(reg1, &m, &cssid, &ssid, &schid)) {
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+    dprintf("IOINST: hsch (%x.%x.%04x)\n", cssid, ssid, schid);
+    sch = css_find_subch(m, cssid, ssid, schid);
+    if (sch && css_subch_visible(sch)) {
+        ret = css_do_hsch(sch);
+    }
+    switch (ret) {
+    case -ENODEV:
+        cc = 3;
+        break;
+    case -EBUSY:
+        cc = 2;
+        break;
+    case 0:
+        cc = 0;
+        break;
+    default:
+        cc = 1;
+        break;
+    }
+
+    return cc;
+}
+
+static int ioinst_schib_valid(SCHIB *schib)
+{
+    if ((schib->pmcw.zeroes0 & 0x3) != 0) {
+        return 0;
+    }
+    if ((schib->pmcw.zeroes1 != 0) || (schib->pmcw.zeroes2 != 0)) {
+        return 0;
+    }
+    /* Disallow extended measurements for now. */
+    if (schib->pmcw.xmwme) {
+        return 0;
+    }
+    return 1;
+}
+
+int ioinst_handle_msch(CPUS390XState *env, uint64_t reg1, uint32_t ipb)
+{
+    int cssid, ssid, schid, m;
+    SubchDev *sch;
+    SCHIB *schib;
+    uint64_t addr;
+    int ret = -ENODEV;
+    int cc;
+
+    if (ioinst_disassemble_sch_ident(reg1, &m, &cssid, &ssid, &schid)) {
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+    dprintf("IOINST: msch (%x.%x.%04x)\n", cssid, ssid, schid);
+    addr = ipb >> 28;
+    if (addr > 0) {
+        addr = env->regs[addr];
+    }
+    addr += (ipb & 0xfff0000) >> 16;
+    schib = s390_get_address(env, addr);
+    if (!schib) {
+        program_interrupt(env, PGM_SPECIFICATION, 2);
+        return -EIO;
+    }
+    if (!ioinst_schib_valid(schib)) {
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+    sch = css_find_subch(m, cssid, ssid, schid);
+    if (sch && css_subch_visible(sch)) {
+        ret = css_do_msch(sch, schib);
+    }
+    switch (ret) {
+    case -ENODEV:
+        cc = 3;
+        break;
+    case -EBUSY:
+        cc = 2;
+        break;
+    case 0:
+        cc = 0;
+        break;
+    default:
+        cc = 1;
+        break;
+    }
+
+    return cc;
+}
+
+static int ioinst_orb_valid(ORB *orb)
+{
+    if (orb->zero0 != 0) {
+        return 0;
+    }
+    if (orb->zero1 != 0) {
+        return 0;
+    }
+    if ((orb->cpa & 0x80000000) != 0) {
+        return 0;
+    }
+    return 1;
+}
+
+int ioinst_handle_ssch(CPUS390XState *env, uint64_t reg1, uint32_t ipb)
+{
+    int cssid, ssid, schid, m;
+    SubchDev *sch;
+    ORB *orb;
+    uint64_t addr;
+    int ret = -ENODEV;
+    int cc;
+
+    if (ioinst_disassemble_sch_ident(reg1, &m, &cssid, &ssid, &schid)) {
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+    dprintf("IOINST: ssch (%x.%x.%04x)\n", cssid, ssid, schid);
+    addr = ipb >> 28;
+    if (addr > 0) {
+        addr = env->regs[addr];
+    }
+    addr += (ipb & 0xfff0000) >> 16;
+    orb = s390_get_address(env, addr);
+    if (!orb) {
+        program_interrupt(env, PGM_SPECIFICATION, 2);
+        return -EIO;
+    }
+    if (!ioinst_orb_valid(orb)) {
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+    sch = css_find_subch(m, cssid, ssid, schid);
+    if (sch && css_subch_visible(sch)) {
+        ret = css_do_ssch(sch, orb);
+    }
+    switch (ret) {
+    case -ENODEV:
+        cc = 3;
+        break;
+    case -EBUSY:
+        cc = 2;
+        break;
+    case 0:
+        cc = 0;
+        break;
+    default:
+        cc = 1;
+        break;
+    }
+
+    return cc;
+}
+
+int ioinst_handle_stcrw(CPUS390XState *env, uint32_t ipb)
+{
+    CRW *crw;
+    uint64_t addr;
+    int cc;
+
+    addr = ipb >> 28;
+    if (addr > 0) {
+        addr = env->regs[addr];
+    }
+    addr += (ipb & 0xfff0000) >> 16;
+    crw = s390_get_address(env, addr);
+    if (!crw) {
+        program_interrupt(env, PGM_SPECIFICATION, 2);
+        return -EIO;
+    }
+    if (addr < 8192) {
+        addr += env->psa;
+    } else if ((env->psa <= addr) && (addr < env->psa + 8192)) {
+        addr -= env->psa;
+    }
+    cc = css_do_stcrw(addr);
+    /* 0 - crw stored, 1 - zeroes stored */
+    return cc;
+}
+
+int ioinst_handle_stsch(CPUS390XState *env, uint64_t reg1, uint32_t ipb)
+{
+    int cssid, ssid, schid, m;
+    SubchDev *sch;
+    uint64_t addr;
+    int cc;
+
+    if (ioinst_disassemble_sch_ident(reg1, &m, &cssid, &ssid, &schid)) {
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+    dprintf("IOINST: stsch (%x.%x.%04x)\n", cssid, ssid, schid);
+    addr = ipb >> 28;
+    if (addr > 0) {
+        addr = env->regs[addr];
+    }
+    addr += (ipb & 0xfff0000) >> 16;
+    if (addr < 8192) {
+        addr += env->psa;
+    } else if ((env->psa <= addr) && (addr < env->psa + 8192)) {
+        addr -= env->psa;
+    }
+    if (!qemu_get_ram_ptr(addr)) {
+        program_interrupt(env, PGM_SPECIFICATION, 2);
+        return -EIO;
+    }
+    sch = css_find_subch(m, cssid, ssid, schid);
+    if (sch) {
+        if (css_subch_visible(sch)) {
+            css_do_stsch(sch, addr);
+            cc = 0;
+        } else {
+            /* Indicate no more subchannels in this css/ss */
+            cc = 3;
+        }
+    } else {
+        if (css_schid_final(cssid, ssid, schid)) {
+            cc = 3; /* No more subchannels in this css/ss */
+        } else {
+            int i;
+
+            /* Store an empty schib. */
+            for (i = 0; i < sizeof(SCHIB); i++) {
+                stb_phys(addr + i, 0);
+            }
+            cc = 0;
+        }
+    }
+    return cc;
+}
+
+int ioinst_handle_tsch(CPUS390XState *env, uint64_t reg1, uint32_t ipb)
+{
+    int cssid, ssid, schid, m;
+    SubchDev *sch;
+    IRB *irb;
+    uint64_t addr;
+    int ret = -ENODEV;
+    int cc;
+
+    if (ioinst_disassemble_sch_ident(reg1, &m, &cssid, &ssid, &schid)) {
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+    dprintf("IOINST: tsch (%x.%x.%04x)\n", cssid, ssid, schid);
+    addr = ipb >> 28;
+    if (addr > 0) {
+        addr = env->regs[addr];
+    }
+    addr += (ipb & 0xfff0000) >> 16;
+    irb = s390_get_address(env, addr);
+    if (!irb) {
+        program_interrupt(env, PGM_SPECIFICATION, 2);
+        return -EIO;
+    }
+    sch = css_find_subch(m, cssid, ssid, schid);
+    if (sch && css_subch_visible(sch)) {
+        if (addr < 8192) {
+            addr += env->psa;
+        } else if ((env->psa <= addr) && (addr < env->psa + 8192)) {
+            addr -= env->psa;
+        }
+        ret = css_do_tsch(sch, addr);
+        /* 0 - status pending, 1 - not status pending */
+        cc = ret;
+    } else {
+        cc = 3;
+    }
+    return cc;
+}
+
+typedef struct ChscReq {
+    uint16_t len;
+    uint16_t command;
+    uint32_t param0;
+    uint32_t param1;
+    uint32_t param2;
+} QEMU_PACKED ChscReq;
+
+typedef struct ChscResp {
+    uint16_t len;
+    uint16_t code;
+    uint32_t param;
+    char data[0];
+} QEMU_PACKED ChscResp;
+
+#define CHSC_SCPD 0x0002
+#define CHSC_SCSC 0x0010
+#define CHSC_SDA  0x0031
+
+static void ioinst_handle_chsc_scpd(ChscReq *req, ChscResp *res)
+{
+    uint16_t resp_code;
+    int rfmt;
+    uint16_t cssid;
+    uint8_t f_chpid, l_chpid;
+    int desc_size;
+    int m;
+
+    rfmt = (req->param0 & 0x00000f00) >> 8;
+    if ((rfmt == 0) ||  (rfmt == 1)) {
+        rfmt = (req->param0 & 0x10000000) >> 28;
+    }
+    if ((req->len != 0x0010) || (req->param0 & 0xc000f000) ||
+        (req->param1 & 0xffffff00) || req->param2) {
+        resp_code = 0x0003;
+        goto out_err;
+    }
+    if (req->param0 & 0x0f000000) {
+        resp_code = 0x0007;
+        goto out_err;
+    }
+    cssid = (req->param0 & 0x00ff0000) >> 16;
+    m = req->param0 & 0x20000000;
+    if (cssid != 0) {
+        if (!m || !css_present(cssid)) {
+            resp_code = 0x0008;
+            goto out_err;
+        }
+    }
+    f_chpid = req->param0 & 0x000000ff;
+    l_chpid = req->param1 & 0x000000ff;
+    if (l_chpid < f_chpid) {
+        resp_code = 0x0003;
+        goto out_err;
+    }
+    desc_size = css_collect_chp_desc(m, cssid, f_chpid, l_chpid, rfmt,
+                                     &res->data);
+    res->code = 0x0001;
+    res->len = 8 + desc_size;
+    res->param = rfmt;
+    return;
+
+  out_err:
+    res->code = resp_code;
+    res->len = 8;
+    res->param = rfmt;
+}
+
+static void ioinst_handle_chsc_scsc(ChscReq *req, ChscResp *res)
+{
+    uint8_t cssid;
+    uint16_t resp_code;
+    uint32_t general_chars[510];
+    uint32_t chsc_chars[508];
+
+    if (req->param0 & 0x000f0000) {
+        resp_code = 0x0007;
+        goto out_err;
+    }
+    cssid = (req->param0 & 0x0000ff00) >> 8;
+    if (cssid != 0) {
+        if (!(req->param0 & 0x20000000) || !css_present(cssid)) {
+            resp_code = 0x0008;
+            goto out_err;
+        }
+    }
+    if ((req->param0 & 0xdff000ff) || req->param1 || req->param2) {
+        resp_code = 0x0003;
+        goto out_err;
+    }
+    res->code = 0x0001;
+    res->len = 4080;
+    res->param = 0;
+
+    memset(general_chars, 0, sizeof(general_chars));
+    memset(chsc_chars, 0, sizeof(chsc_chars));
+
+    general_chars[0] = 0x03000000;
+    general_chars[1] = 0x00059000;
+
+    chsc_chars[0] = 0x40000000;
+    chsc_chars[3] = 0x00040000;
+
+    memcpy(res->data, general_chars, sizeof(general_chars));
+    memcpy(res->data + sizeof(general_chars), chsc_chars, sizeof(chsc_chars));
+    return;
+
+  out_err:
+    res->code = resp_code;
+    res->len = 8;
+    res->param = 0;
+}
+
+#define CHSC_SDA_OC_MCSSE 0x0
+#define CHSC_SDA_OC_MSS 0x2
+static void ioinst_handle_chsc_sda(ChscReq *req, ChscResp *res)
+{
+    uint16_t resp_code = 0x0001;
+    uint16_t oc;
+    int ret;
+
+    if ((req->len != 0x0400) || (req->param0 & 0xf0ff0000)) {
+        resp_code = 0x0003;
+        goto out;
+    }
+
+    if (req->param0 & 0x0f000000) {
+        resp_code = 0x0007;
+        goto out;
+    }
+
+    oc = req->param0 & 0x0000ffff;
+    switch (oc) {
+    case CHSC_SDA_OC_MCSSE:
+        ret = css_enable_mcsse();
+        if (ret == -EINVAL) {
+            resp_code = 0x0101;
+            goto out;
+        }
+        break;
+    case CHSC_SDA_OC_MSS:
+        ret = css_enable_mss();
+        if (ret == -EINVAL) {
+            resp_code = 0x0101;
+            goto out;
+        }
+        break;
+    default:
+        resp_code = 0x0003;
+        goto out;
+    }
+
+out:
+    res->code = resp_code;
+    res->len = 8;
+    res->param = 0;
+}
+
+static void ioinst_handle_chsc_unimplemented(ChscResp *res)
+{
+    res->len = 8;
+    res->code = 0x0004;
+    res->param = 0;
+}
+
+int ioinst_handle_chsc(CPUS390XState *env, uint32_t ipb)
+{
+    ChscReq *req;
+    ChscResp *res;
+    uint64_t addr;
+    int reg;
+
+    dprintf("%s\n", "IOINST: CHSC");
+    reg = (ipb >> 20) & 0x00f;
+    addr = env->regs[reg];
+    req = s390_get_address(env, addr);
+    if (!req) {
+        program_interrupt(env, PGM_SPECIFICATION, 2);
+        return -EIO;
+    }
+    if (!env->chsc_page) {
+        env->chsc_page = g_malloc0(TARGET_PAGE_SIZE);
+    } else {
+        memset(env->chsc_page, 0, TARGET_PAGE_SIZE);
+    }
+    res = env->chsc_page;
+    dprintf("IOINST: CHSC: command 0x%04x, len=0x%04x\n",
+            req->command, req->len);
+    switch (req->command) {
+    case CHSC_SCSC:
+        ioinst_handle_chsc_scsc(req, res);
+        break;
+    case CHSC_SCPD:
+        ioinst_handle_chsc_scpd(req, res);
+        break;
+    case CHSC_SDA:
+        ioinst_handle_chsc_sda(req, res);
+        break;
+    default:
+        ioinst_handle_chsc_unimplemented(res);
+        break;
+    }
+    if (addr < 8192) {
+        addr += env->psa;
+    } else if ((env->psa <= addr) && (addr < env->psa + 8192)) {
+        addr -= env->psa;
+    }
+    cpu_physical_memory_write(addr + req->len, res, res->len);
+    return 0;
+}
+
+int ioinst_handle_tpi(CPUS390XState *env, uint32_t ipb)
+{
+    uint64_t addr;
+    int lowcore;
+
+    dprintf("%s\n", "IOINST: tpi");
+    addr = ipb >> 28;
+    if (addr > 0) {
+        addr = env->regs[addr];
+    }
+    addr += (ipb & 0xfff0000) >> 16;
+    lowcore = addr ? 0 : 1;
+    if (addr < 8192) {
+        addr += env->psa;
+    } else if ((env->psa <= addr) && (addr < env->psa + 8192)) {
+        addr -= env->psa;
+    }
+    return css_do_tpi(addr, lowcore);
+}
+
+int ioinst_handle_schm(CPUS390XState *env, uint64_t reg1, uint64_t reg2,
+                       uint32_t ipb)
+{
+    uint8_t mbk;
+    int update;
+    int dct;
+
+    dprintf("%s\n", "IOINST: schm");
+
+    if (reg1 & 0x000000000ffffffc) {
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+
+    mbk = (reg1 & 0x00000000f0000000) >> 28;
+    update = (reg1 & 0x0000000000000002) >> 1;
+    dct = reg1 & 0x0000000000000001;
+
+    if (update && (reg2 & 0x0000000000000fff)) {
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+
+    css_do_schm(mbk, update, dct, update ? reg2 : 0);
+
+    return 0;
+}
+
+int ioinst_handle_rsch(CPUS390XState *env, uint64_t reg1)
+{
+    int cssid, ssid, schid, m;
+    SubchDev *sch;
+    int ret = -ENODEV;
+    int cc;
+
+    if (ioinst_disassemble_sch_ident(reg1, &m, &cssid, &ssid, &schid)) {
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+    dprintf("IOINST: rsch (%x.%x.%04x)\n", cssid, ssid, schid);
+    sch = css_find_subch(m, cssid, ssid, schid);
+    if (sch && css_subch_visible(sch)) {
+        ret = css_do_rsch(sch);
+    }
+    switch (ret) {
+    case -ENODEV:
+        cc = 3;
+        break;
+    case -EINVAL:
+        cc = 2;
+        break;
+    case 0:
+        cc = 0;
+        break;
+    default:
+        cc = 1;
+        break;
+    }
+
+    return cc;
+
+}
+
+int ioinst_handle_rchp(CPUS390XState *env, uint64_t reg1)
+{
+    int cc;
+    uint8_t cssid;
+    uint8_t chpid;
+    int ret;
+
+    if (reg1 & 0xff00ff00) {
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+
+    cssid = (reg1 >> 16) & 0xff;
+    chpid = reg1 & 0xff;
+    dprintf("IOINST: rchp (%x.%02x)\n", cssid, chpid);
+
+    ret = css_do_rchp(cssid, chpid);
+
+    switch (ret) {
+    case -ENODEV:
+        cc = 3;
+        break;
+    case -EBUSY:
+        cc = 2;
+        break;
+    case 0:
+        cc = 0;
+        break;
+    default:
+        /* Invalid channel subsystem. */
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+
+    return cc;
+}
+
+int ioinst_handle_sal(CPUS390XState *env, uint64_t reg1)
+{
+    /* We do not provide address limit checking, so let's suppress it. */
+    if (env->regs[1] & 0x000000008000ffff) {
+        program_interrupt(env, PGM_OPERAND, 2);
+        return -EIO;
+    }
+    return 0;
+}
diff --git a/target-s390x/ioinst.h b/target-s390x/ioinst.h
new file mode 100644
index 0000000..9810fc5
--- /dev/null
+++ b/target-s390x/ioinst.h
@@ -0,0 +1,213 @@
+/*
+ * S/390 channel I/O instructions
+ *
+ * Copyright 2012 IBM Corp.
+ * Author(s): Cornelia Huck <cornelia.huck@xxxxxxxxxx>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or (at
+ * your option) any later version. See the COPYING file in the top-level
+ * directory.
+*/
+
+#ifndef IOINST_S390X_H
+#define IOINST_S390X_H
+/*
+ * Channel I/O related definitions, as defined in the Principles
+ * Of Operation (and taken from the Linux implementation).
+ */
+
+/* subchannel status word (command mode only) */
+typedef struct SCSW {
+    uint32_t key:4;
+    uint32_t sctl:1;
+    uint32_t eswf:1;
+    uint32_t cc:2;
+    uint32_t fmt:1;
+    uint32_t pfch:1;
+    uint32_t isic:1;
+    uint32_t alcc:1;
+    uint32_t ssi:1;
+    uint32_t zcc:1;
+    uint32_t ectl:1;
+    uint32_t pno:1;
+    uint32_t res:1;
+    uint32_t fctl:3;
+    uint32_t actl:7;
+    uint32_t stctl:5;
+    uint32_t cpa;
+    uint32_t dstat:8;
+    uint32_t cstat:8;
+    uint32_t count:16;
+} SCSW;
+
+/* path management control word */
+typedef struct PMCW {
+    uint32_t intparm;
+    uint32_t qf:1;
+    uint32_t w:1;
+    uint32_t isc:3;
+    uint32_t zeroes0:3;
+    uint32_t ena:1;
+    uint32_t lm:2;
+    uint32_t mme:2;
+    uint32_t mp:1;
+    uint32_t tf:1;
+    uint32_t dnv:1;
+    uint32_t dev:16;
+    uint8_t  lpm;
+    uint8_t  pnom;
+    uint8_t  lpum;
+    uint8_t  pim;
+    uint16_t mbi;
+    uint8_t  pom;
+    uint8_t  pam;
+    uint8_t  chpid[8];
+    uint32_t zeroes1:8;
+    uint32_t st:3;
+    uint32_t zeroes2:18;
+    uint32_t mbfc:1;
+    uint32_t xmwme:1;
+    uint32_t csense:1;
+} PMCW;
+
+/* subchannel information block */
+struct SCHIB {
+    PMCW pmcw;
+    SCSW scsw;
+    uint64_t mba;
+    uint8_t mda[4];
+};
+
+/* interruption response block */
+typedef struct IRB {
+    SCSW scsw;
+    uint32_t esw[5];
+    uint32_t ecw[8];
+    uint32_t emw[8];
+} IRB;
+
+/* operation request block */
+struct ORB {
+    uint32_t intparm;
+    uint32_t key:4;
+    uint32_t spnd:1;
+    uint32_t str:1;
+    uint32_t mod:1;
+    uint32_t sync:1;
+    uint32_t fmt:1;
+    uint32_t pfch:1;
+    uint32_t isic:1;
+    uint32_t alcc:1;
+    uint32_t ssic:1;
+    uint32_t zero0:1;
+    uint32_t c64:1;
+    uint32_t i2k:1;
+    uint32_t lpm:8;
+    uint32_t ils:1;
+    uint32_t midaw:1;
+    uint32_t zero1:5;
+    uint32_t orbx:1;
+    uint32_t cpa;
+};
+
+/* channel command word (type 1) */
+typedef struct CCW1 {
+    uint8_t cmd_code;
+    uint8_t flags;
+    uint16_t count;
+    uint32_t cda;
+} CCW1;
+
+#define CCW_FLAG_DC              0x80
+#define CCW_FLAG_CC              0x40
+#define CCW_FLAG_SLI             0x20
+#define CCW_FLAG_SKIP            0x10
+#define CCW_FLAG_PCI             0x08
+#define CCW_FLAG_IDA             0x04
+#define CCW_FLAG_SUSPEND         0x02
+
+#define CCW_CMD_NOOP             0x03
+#define CCW_CMD_BASIC_SENSE      0x04
+#define CCW_CMD_TIC              0x08
+#define CCW_CMD_SENSE_ID         0xe4
+
+#define SCSW_FCTL_CLEAR_FUNC     0x1
+#define SCSW_FCTL_HALT_FUNC      0x2
+#define SCSW_FCTL_START_FUNC     0x4
+
+#define SCSW_ACTL_SUSP           0x1
+#define SCSW_ACTL_DEVICE_ACTIVE  0x2
+#define SCSW_ACTL_SUBCH_ACTIVE   0x4
+#define SCSW_ACTL_CLEAR_PEND     0x8
+#define SCSW_ACTL_HALT_PEND      0x10
+#define SCSW_ACTL_START_PEND     0x20
+#define SCSW_ACTL_RESUME_PEND    0x40
+
+#define SCSW_STCTL_STATUS_PEND   0x1
+#define SCSW_STCTL_SECONDARY     0x2
+#define SCSW_STCTL_PRIMARY       0x4
+#define SCSW_STCTL_INTERMEDIATE  0x8
+#define SCSW_STCTL_ALERT         0x10
+
+#define SCSW_DSTAT_ATTENTION     0x80
+#define SCSW_DSTAT_STAT_MOD      0x40
+#define SCSW_DSTAT_CU_END        0x20
+#define SCSW_DSTAT_BUSY          0x10
+#define SCSW_DSTAT_CHANNEL_END   0x08
+#define SCSW_DSTAT_DEVICE_END    0x04
+#define SCSW_DSTAT_UNIT_CHECK    0x02
+#define SCSW_DSTAT_UNIT_EXCEP    0x01
+
+#define SCSW_CSTAT_PCI           0x80
+#define SCSW_CSTAT_INCORR_LEN    0x40
+#define SCSW_CSTAT_PROG_CHECK    0x20
+#define SCSW_CSTAT_PROT_CHECK    0x10
+#define SCSW_CSTAT_DATA_CHECK    0x08
+#define SCSW_CSTAT_CHN_CTRL_CHK  0x04
+#define SCSW_CSTAT_INTF_CTRL_CHK 0x02
+#define SCSW_CSTAT_CHAIN_CHECK   0x01
+
+typedef struct CRW {
+    uint16_t zero0:1;
+    uint16_t s:1;
+    uint16_t r:1;
+    uint16_t c:1;
+    uint16_t rsc:4;
+    uint16_t a:1;
+    uint16_t zero1:1;
+    uint16_t erc:6;
+    uint16_t rsid;
+} CRW;
+
+#define CRW_ERC_INIT 0x02
+#define CRW_ERC_IPI  0x04
+
+#define CRW_RSC_SUBCH 0x3
+#define CRW_RSC_CHP   0x4
+
+/* schid disintegration */
+#define IOINST_SCHID_ONE   0x00010000
+#define IOINST_SCHID_M     0x00080000
+#define IOINST_SCHID_CSSID 0xff000000
+#define IOINST_SCHID_SSID  0x00060000
+#define IOINST_SCHID_NR    0x0000ffff
+
+int ioinst_disassemble_sch_ident(uint32_t value, int *m, int *cssid, int *ssid,
+                                 int *schid);
+int ioinst_handle_xsch(CPUS390XState *env, uint64_t reg1);
+int ioinst_handle_csch(CPUS390XState *env, uint64_t reg1);
+int ioinst_handle_hsch(CPUS390XState *env, uint64_t reg1);
+int ioinst_handle_msch(CPUS390XState *env, uint64_t reg1, uint32_t ipb);
+int ioinst_handle_ssch(CPUS390XState *env, uint64_t reg1, uint32_t ipb);
+int ioinst_handle_stcrw(CPUS390XState *env, uint32_t ipb);
+int ioinst_handle_stsch(CPUS390XState *env, uint64_t reg1, uint32_t ipb);
+int ioinst_handle_tsch(CPUS390XState *env, uint64_t reg1, uint32_t ipb);
+int ioinst_handle_chsc(CPUS390XState *env, uint32_t ipb);
+int ioinst_handle_tpi(CPUS390XState *env, uint32_t ipb);
+int ioinst_handle_schm(CPUS390XState *env, uint64_t reg1, uint64_t reg2,
+                       uint32_t ipb);
+int ioinst_handle_rsch(CPUS390XState *env, uint64_t reg1);
+int ioinst_handle_rchp(CPUS390XState *env, uint64_t reg1);
+int ioinst_handle_sal(CPUS390XState *env, uint64_t reg1);
+
+#endif
diff --git a/target-s390x/kvm.c b/target-s390x/kvm.c
index a66ac43..b53391e 100644
--- a/target-s390x/kvm.c
+++ b/target-s390x/kvm.c
@@ -26,10 +26,13 @@
 
 #include "qemu-common.h"
 #include "qemu-timer.h"
+#include "qemu-thread.h"
 #include "sysemu.h"
 #include "kvm.h"
 #include "cpu.h"
 #include "device_tree.h"
+#include "trace.h"
+#include "ioinst.h"
 
 /* #define DEBUG_KVM */
 
@@ -43,9 +46,27 @@
 
 #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 DIAG_KVM_HYPERCALL              0x500
 #define DIAG_KVM_BREAKPOINT             0x501
 
@@ -350,10 +371,120 @@ static int kvm_sclp_service_call(CPUS390XState *env, struct kvm_run *run,
     return 0;
 }
 
-static int handle_priv(CPUS390XState *env, struct kvm_run *run, uint8_t ipa1)
+static int kvm_handle_css_inst(CPUS390XState *env, struct kvm_run *run,
+                               uint8_t ipa0, uint8_t ipa1, uint8_t ipb)
+{
+    int r = 0;
+    int no_cc = 0;
+
+    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(env, 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;
+
+    switch (ipa0) {
+    case 0xb2:
+        if (((ipa1 >= 0x30) && (ipa1 <= 0x3c)) ||
+            (ipa1 == 0x5f) ||
+            (ipa1 == 0x74) ||
+            (ipa1 == 0x76)) {
+            ret = 1;
+        }
+        break;
+    case 0xb9:
+        if (ipa1 == 0x9c) {
+            ret = 1;
+        }
+        break;
+    case 0xeb:
+        if (ipb == 0x8a) {
+            ret = 1;
+        }
+        break;
+    }
+
+    return ret;
+}
+
+static int handle_priv(CPUS390XState *env, 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) {
@@ -361,8 +492,16 @@ static int handle_priv(CPUS390XState *env, struct kvm_run *run, uint8_t ipa1)
             r = kvm_sclp_service_call(env, 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(env, run, ipa0, ipa1, ipb);
+                if (r == -1) {
+                    setcc(env, 3);
+                    r = 0;
+                }
+            } else {
+                dprintf("KVM: unknown PRIV: 0x%x\n", ipa1);
+                r = -1;
+            }
             break;
     }
 
@@ -500,15 +639,17 @@ static int handle_instruction(CPUS390XState *env, 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(env, run, ipa1);
-            break;
-        case IPA0_DIAG:
-            r = handle_diag(env, run, ipb_code);
-            break;
-        case IPA0_SIGP:
-            r = handle_sigp(env, run, ipa1);
-            break;
+    case IPA0_B2:
+    case IPA0_B9:
+    case IPA0_EB:
+        r = handle_priv(env, run, ipa0 >> 8, ipa1);
+        break;
+    case IPA0_DIAG:
+        r = handle_diag(env, run, ipb_code);
+        break;
+    case IPA0_SIGP:
+        r = handle_sigp(env, run, ipa1);
+        break;
     }
 
     if (r < 0) {
@@ -565,6 +706,38 @@ static int handle_intercept(CPUS390XState *env)
     return r;
 }
 
+static int handle_tsch(CPUS390XState *env, struct kvm_run *run, int dequeued,
+                       uint16_t subchannel_id, uint16_t subchannel_nr,
+                       uint32_t io_int_parm, uint32_t io_int_word)
+{
+    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(env, ret);
+        ret = 0;
+    } else if (ret < -1) {
+        /*
+         * Failure.
+         * If an I/O interrupt had been dequeued, we have to reinject it.
+         */
+        if (dequeued) {
+            uint32_t type = ((subchannel_id & 0xff00) << 24) |
+                ((subchannel_id & 0x00060) << 22) | (subchannel_nr << 16);
+
+            kvm_s390_interrupt_internal(env, 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(CPUS390XState *env, struct kvm_run *run)
 {
     int ret = 0;
@@ -576,6 +749,13 @@ int kvm_arch_handle_exit(CPUS390XState *env, struct kvm_run *run)
         case KVM_EXIT_S390_RESET:
             qemu_system_reset_request();
             break;
+        case KVM_EXIT_S390_TSCH:
+            ret = handle_tsch(env, run, run->s390_tsch.dequeued,
+                              run->s390_tsch.subchannel_id,
+                              run->s390_tsch.subchannel_nr,
+                              run->s390_tsch.io_int_parm,
+                              run->s390_tsch.io_int_word);
+            break;
         default:
             fprintf(stderr, "Unknown KVM exit: %d\n", run->exit_reason);
             break;
@@ -601,3 +781,48 @@ int kvm_arch_on_sigbus(int code, void *addr)
 {
     return 1;
 }
+
+int kvm_s390_io_interrupt(CPUS390XState *env, uint16_t subchannel_id,
+                          uint16_t subchannel_nr, uint32_t io_int_parm,
+                          uint32_t io_int_word)
+{
+    uint32_t type;
+
+    if (!kvm_enabled()) {
+        return -EOPNOTSUPP;
+    }
+
+    type = ((subchannel_id & 0xff00) << 24) |
+        ((subchannel_id & 0x00060) << 22) | (subchannel_nr << 16);
+    kvm_s390_interrupt_internal(env, type,
+                                ((uint32_t)subchannel_id << 16) | subchannel_nr,
+                                ((uint64_t)io_int_parm << 32) | io_int_word, 1);
+    return 0;
+}
+
+int kvm_s390_crw_mchk(CPUS390XState *env)
+{
+    if (!kvm_enabled()) {
+        return -EOPNOTSUPP;
+    }
+
+    kvm_s390_interrupt_internal(env, KVM_S390_MCHK, 1 << 28,
+                                0x00400f1d40330000, 1);
+    return 0;
+}
+
+void kvm_s390_enable_css_support(CPUS390XState *env)
+{
+    struct kvm_enable_cap cap = {};
+    int r;
+
+    /* Activate host kernel channel subsystem support. */
+    if (kvm_enabled()) {
+        /* One CPU has to run */
+        s390_add_running_cpu(env);
+
+        cap.cap = KVM_CAP_S390_CSS_SUPPORT;
+        r = kvm_vcpu_ioctl(env, KVM_ENABLE_CAP, &cap);
+        assert(r == 0);
+    }
+}
diff --git a/target-s390x/misc_helper.c b/target-s390x/misc_helper.c
index 38d8f2a..cd4bca1 100644
--- a/target-s390x/misc_helper.c
+++ b/target-s390x/misc_helper.c
@@ -49,12 +49,12 @@ void HELPER(exception)(CPUS390XState *env, uint32_t excp)
     cpu_loop_exit(env);
 }
 
-#ifndef CONFIG_USER_ONLY
 void program_interrupt(CPUS390XState *env, uint32_t code, int ilc)
 {
     qemu_log_mask(CPU_LOG_INT, "program interrupt at %#" PRIx64 "\n",
                   env->psw.addr);
 
+#ifndef CONFIG_USER_ONLY
     if (kvm_enabled()) {
 #ifdef CONFIG_KVM
         kvm_s390_interrupt(env, KVM_S390_PROGRAM_INT, code);
@@ -65,8 +65,12 @@ void program_interrupt(CPUS390XState *env, uint32_t code, int ilc)
         env->exception_index = EXCP_PGM;
         cpu_loop_exit(env);
     }
+#else
+    cpu_abort(env, "Program check %x\n", code);
+#endif
 }
 
+#ifndef CONFIG_USER_ONLY
 /* SCLP service call */
 uint32_t HELPER(servc)(CPUS390XState *env, uint32_t r1, uint64_t r2)
 {
-- 
1.7.12.4

--
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


[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]
  Powered by Linux