Merge remote-tracking branch 'remotes/awilliam/tags/vfio-update-20201101.0' into staging

VFIO update 2020-11-01

 * Migration support (Kirti Wankhede)
 * s390 DMA limiting (Matthew Rosato)
 * zPCI hardware info (Matthew Rosato)
 * Lock guard (Amey Narkhede)
 * Print fixes (Zhengui li)
 * Warning/build fixes

# gpg: Signature made Sun 01 Nov 2020 20:38:10 GMT
# gpg:                using RSA key 239B9B6E3BB08B22
# gpg: Good signature from "Alex Williamson <alex.williamson@redhat.com>" [full]
# gpg:                 aka "Alex Williamson <alex@shazbot.org>" [full]
# gpg:                 aka "Alex Williamson <alwillia@redhat.com>" [full]
# gpg:                 aka "Alex Williamson <alex.l.williamson@gmail.com>" [full]
# Primary key fingerprint: 42F6 C04E 540B D1A9 9E7B  8A90 239B 9B6E 3BB0 8B22

* remotes/awilliam/tags/vfio-update-20201101.0: (32 commits)
  vfio: fix incorrect print type
  hw/vfio: Use lock guard macros
  s390x/pci: get zPCI function info from host
  vfio: Add routine for finding VFIO_DEVICE_GET_INFO capabilities
  s390x/pci: use a PCI Function structure
  s390x/pci: clean up s390 PCI groups
  s390x/pci: use a PCI Group structure
  s390x/pci: create a header dedicated to PCI CLP
  s390x/pci: Honor DMA limits set by vfio
  s390x/pci: Add routine to get the vfio dma available count
  vfio: Find DMA available capability
  vfio: Create shared routine for scanning info capabilities
  s390x/pci: Move header files to include/hw/s390x
  linux-headers: update against 5.10-rc1
  update-linux-headers: Add vfio_zdev.h
  qapi: Add VFIO devices migration stats in Migration stats
  vfio: Make vfio-pci device migration capable
  vfio: Add ioctl to get dirty pages bitmap during dma unmap
  vfio: Dirty page tracking when vIOMMU is enabled
  vfio: Add vfio_listener_log_sync to mark dirty pages
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

1 files changed, 276 insertions, 0 deletions

diff --git a/hw/s390x/s390-pci-vfio.c b/hw/s390x/s390-pci-vfio.c
new file mode 100644
index 0000000000..d5c78063b5
--- /dev/null
+++ b/hw/s390x/s390-pci-vfio.c
@@ -0,0 +1,276 @@
+/*
+ * s390 vfio-pci interfaces
+ *
+ * Copyright 2020 IBM Corp.
+ * Author(s): Matthew Rosato <mjrosato@linux.ibm.com>
+ *
+ * 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/ioctl.h>
+#include <linux/vfio.h>
+#include <linux/vfio_zdev.h>
+
+#include "qemu/osdep.h"
+#include "trace.h"
+#include "hw/s390x/s390-pci-bus.h"
+#include "hw/s390x/s390-pci-clp.h"
+#include "hw/s390x/s390-pci-vfio.h"
+#include "hw/vfio/pci.h"
+#include "hw/vfio/vfio-common.h"
+
+/*
+ * Get the current DMA available count from vfio.  Returns true if vfio is
+ * limiting DMA requests, false otherwise.  The current available count read
+ * from vfio is returned in avail.
+ */
+bool s390_pci_update_dma_avail(int fd, unsigned int *avail)
+{
+    g_autofree struct vfio_iommu_type1_info *info;
+    uint32_t argsz;
+
+    assert(avail);
+
+    argsz = sizeof(struct vfio_iommu_type1_info);
+    info = g_malloc0(argsz);
+
+    /*
+     * If the specified argsz is not large enough to contain all capabilities
+     * it will be updated upon return from the ioctl.  Retry until we have
+     * a big enough buffer to hold the entire capability chain.
+     */
+retry:
+    info->argsz = argsz;
+
+    if (ioctl(fd, VFIO_IOMMU_GET_INFO, info)) {
+        return false;
+    }
+
+    if (info->argsz > argsz) {
+        argsz = info->argsz;
+        info = g_realloc(info, argsz);
+        goto retry;
+    }
+
+    /* If the capability exists, update with the current value */
+    return vfio_get_info_dma_avail(info, avail);
+}
+
+S390PCIDMACount *s390_pci_start_dma_count(S390pciState *s,
+                                          S390PCIBusDevice *pbdev)
+{
+    S390PCIDMACount *cnt;
+    uint32_t avail;
+    VFIOPCIDevice *vpdev = container_of(pbdev->pdev, VFIOPCIDevice, pdev);
+    int id;
+
+    assert(vpdev);
+
+    id = vpdev->vbasedev.group->container->fd;
+
+    if (!s390_pci_update_dma_avail(id, &avail)) {
+        return NULL;
+    }
+
+    QTAILQ_FOREACH(cnt, &s->zpci_dma_limit, link) {
+        if (cnt->id  == id) {
+            cnt->users++;
+            return cnt;
+        }
+    }
+
+    cnt = g_new0(S390PCIDMACount, 1);
+    cnt->id = id;
+    cnt->users = 1;
+    cnt->avail = avail;
+    QTAILQ_INSERT_TAIL(&s->zpci_dma_limit, cnt, link);
+    return cnt;
+}
+
+void s390_pci_end_dma_count(S390pciState *s, S390PCIDMACount *cnt)
+{
+    assert(cnt);
+
+    cnt->users--;
+    if (cnt->users == 0) {
+        QTAILQ_REMOVE(&s->zpci_dma_limit, cnt, link);
+    }
+}
+
+static void s390_pci_read_base(S390PCIBusDevice *pbdev,
+                               struct vfio_device_info *info)
+{
+    struct vfio_info_cap_header *hdr;
+    struct vfio_device_info_cap_zpci_base *cap;
+    VFIOPCIDevice *vpci =  container_of(pbdev->pdev, VFIOPCIDevice, pdev);
+
+    hdr = vfio_get_device_info_cap(info, VFIO_DEVICE_INFO_CAP_ZPCI_BASE);
+
+    /* If capability not provided, just leave the defaults in place */
+    if (hdr == NULL) {
+        trace_s390_pci_clp_cap(vpci->vbasedev.name,
+                               VFIO_DEVICE_INFO_CAP_ZPCI_BASE);
+        return;
+    }
+    cap = (void *) hdr;
+
+    pbdev->zpci_fn.sdma = cap->start_dma;
+    pbdev->zpci_fn.edma = cap->end_dma;
+    pbdev->zpci_fn.pchid = cap->pchid;
+    pbdev->zpci_fn.vfn = cap->vfn;
+    pbdev->zpci_fn.pfgid = cap->gid;
+    /* The following values remain 0 until we support other FMB formats */
+    pbdev->zpci_fn.fmbl = 0;
+    pbdev->zpci_fn.pft = 0;
+}
+
+static void s390_pci_read_group(S390PCIBusDevice *pbdev,
+                                struct vfio_device_info *info)
+{
+    struct vfio_info_cap_header *hdr;
+    struct vfio_device_info_cap_zpci_group *cap;
+    ClpRspQueryPciGrp *resgrp;
+    VFIOPCIDevice *vpci =  container_of(pbdev->pdev, VFIOPCIDevice, pdev);
+
+    hdr = vfio_get_device_info_cap(info, VFIO_DEVICE_INFO_CAP_ZPCI_GROUP);
+
+    /* If capability not provided, just use the default group */
+    if (hdr == NULL) {
+        trace_s390_pci_clp_cap(vpci->vbasedev.name,
+                               VFIO_DEVICE_INFO_CAP_ZPCI_GROUP);
+        pbdev->zpci_fn.pfgid = ZPCI_DEFAULT_FN_GRP;
+        pbdev->pci_group = s390_group_find(ZPCI_DEFAULT_FN_GRP);
+        return;
+    }
+    cap = (void *) hdr;
+
+    /* See if the PCI group is already defined, create if not */
+    pbdev->pci_group = s390_group_find(pbdev->zpci_fn.pfgid);
+
+    if (!pbdev->pci_group) {
+        pbdev->pci_group = s390_group_create(pbdev->zpci_fn.pfgid);
+
+        resgrp = &pbdev->pci_group->zpci_group;
+        if (cap->flags & VFIO_DEVICE_INFO_ZPCI_FLAG_REFRESH) {
+            resgrp->fr = 1;
+        }
+        stq_p(&resgrp->dasm, cap->dasm);
+        stq_p(&resgrp->msia, cap->msi_addr);
+        stw_p(&resgrp->mui, cap->mui);
+        stw_p(&resgrp->i, cap->noi);
+        stw_p(&resgrp->maxstbl, cap->maxstbl);
+        stb_p(&resgrp->version, cap->version);
+    }
+}
+
+static void s390_pci_read_util(S390PCIBusDevice *pbdev,
+                               struct vfio_device_info *info)
+{
+    struct vfio_info_cap_header *hdr;
+    struct vfio_device_info_cap_zpci_util *cap;
+    VFIOPCIDevice *vpci =  container_of(pbdev->pdev, VFIOPCIDevice, pdev);
+
+    hdr = vfio_get_device_info_cap(info, VFIO_DEVICE_INFO_CAP_ZPCI_UTIL);
+
+    /* If capability not provided, just leave the defaults in place */
+    if (hdr == NULL) {
+        trace_s390_pci_clp_cap(vpci->vbasedev.name,
+                               VFIO_DEVICE_INFO_CAP_ZPCI_UTIL);
+        return;
+    }
+    cap = (void *) hdr;
+
+    if (cap->size > CLP_UTIL_STR_LEN) {
+        trace_s390_pci_clp_cap_size(vpci->vbasedev.name, cap->size,
+                                    VFIO_DEVICE_INFO_CAP_ZPCI_UTIL);
+        return;
+    }
+
+    pbdev->zpci_fn.flags |= CLP_RSP_QPCI_MASK_UTIL;
+    memcpy(pbdev->zpci_fn.util_str, cap->util_str, CLP_UTIL_STR_LEN);
+}
+
+static void s390_pci_read_pfip(S390PCIBusDevice *pbdev,
+                               struct vfio_device_info *info)
+{
+    struct vfio_info_cap_header *hdr;
+    struct vfio_device_info_cap_zpci_pfip *cap;
+    VFIOPCIDevice *vpci =  container_of(pbdev->pdev, VFIOPCIDevice, pdev);
+
+    hdr = vfio_get_device_info_cap(info, VFIO_DEVICE_INFO_CAP_ZPCI_PFIP);
+
+    /* If capability not provided, just leave the defaults in place */
+    if (hdr == NULL) {
+        trace_s390_pci_clp_cap(vpci->vbasedev.name,
+                               VFIO_DEVICE_INFO_CAP_ZPCI_PFIP);
+        return;
+    }
+    cap = (void *) hdr;
+
+    if (cap->size > CLP_PFIP_NR_SEGMENTS) {
+        trace_s390_pci_clp_cap_size(vpci->vbasedev.name, cap->size,
+                                    VFIO_DEVICE_INFO_CAP_ZPCI_PFIP);
+        return;
+    }
+
+    memcpy(pbdev->zpci_fn.pfip, cap->pfip, CLP_PFIP_NR_SEGMENTS);
+}
+
+/*
+ * This function will issue the VFIO_DEVICE_GET_INFO ioctl and look for
+ * capabilities that contain information about CLP features provided by the
+ * underlying host.
+ * On entry, defaults have already been placed into the guest CLP response
+ * buffers.  On exit, defaults will have been overwritten for any CLP features
+ * found in the capability chain; defaults will remain for any CLP features not
+ * found in the chain.
+ */
+void s390_pci_get_clp_info(S390PCIBusDevice *pbdev)
+{
+    g_autofree struct vfio_device_info *info;
+    VFIOPCIDevice *vfio_pci;
+    uint32_t argsz;
+    int fd;
+
+    argsz = sizeof(*info);
+    info = g_malloc0(argsz);
+
+    vfio_pci = container_of(pbdev->pdev, VFIOPCIDevice, pdev);
+    fd = vfio_pci->vbasedev.fd;
+
+    /*
+     * If the specified argsz is not large enough to contain all capabilities
+     * it will be updated upon return from the ioctl.  Retry until we have
+     * a big enough buffer to hold the entire capability chain.  On error,
+     * just exit and rely on CLP defaults.
+     */
+retry:
+    info->argsz = argsz;
+
+    if (ioctl(fd, VFIO_DEVICE_GET_INFO, info)) {
+        trace_s390_pci_clp_dev_info(vfio_pci->vbasedev.name);
+        return;
+    }
+
+    if (info->argsz > argsz) {
+        argsz = info->argsz;
+        info = g_realloc(info, argsz);
+        goto retry;
+    }
+
+    /*
+     * Find the CLP features provided and fill in the guest CLP responses.
+     * Always call s390_pci_read_base first as information from this could
+     * determine which function group is used in s390_pci_read_group.
+     * For any feature not found, the default values will remain in the CLP
+     * response.
+     */
+    s390_pci_read_base(pbdev, info);
+    s390_pci_read_group(pbdev, info);
+    s390_pci_read_util(pbdev, info);
+    s390_pci_read_pfip(pbdev, info);
+
+    return;
+}