Merge tag 'pull-request-2025-03-07' of https://gitlab.com/thuth/qemu into staging

* Bug fixes and some small improvements for functional tests
* Improve performance of s390x PCI passthrough devices with relaxed translation

# -----BEGIN PGP SIGNATURE-----
#
# iQJFBAABCAAvFiEEJ7iIR+7gJQEY8+q5LtnXdP5wLbUFAmfK3dsRHHRodXRoQHJl
# ZGhhdC5jb20ACgkQLtnXdP5wLbUfDQ/8CopnzCKGKFhyM5skrHbhDbUVbul6yV4L
# kIOo7N8OlrNcQB90bj+Udy+mUANHjkmSiBa5lJ/78ej4DFS6CxeVgrl1fSEl36xn
# GjWDwSUiN8pG1O4YtnDqWVTBieGSzbkQr1jHgpeAnvv08s+TtmudP1T8IznWU2v9
# FqD78SdebZ0Kua+ksBgMxwkHd6VMw13vsu6KuT9VBhie40LcDrFOuG8RDz/qo4IO
# Yg9s1Bqcy7Wa4+0ldMXS1plSdIqJBtVc/HDTg1QwH994b4Lvr7ffrFZmuVcd2dbE
# XKQ5jAMOYZqWdlXszkyd8moYGhmevCkQlALhpnbHixnfakfFYX0wTiJB6oCthFQ0
# It0J/ntNsCmJiIHNbPLzsJ1pE5+ureRnGbxVe05n+zfm8MaXL6s4nSdZzHyp8n43
# UZQqVzK55Q34K9O0qoUdCdBCjMKS9v5u95jjJo8+nc8sJoeQTssOoiixwB/E4y21
# 9qSh7CbDjQK4zwuzQ7jKD603zAJH6ivvsHXlMBMXJFBiSMCAoQQ1vyou6yRHswRr
# gLHDwiWUx8SX8ckbbJ/+Zo9+T8JBMvC5hNYG8VoAtlTQusG4bHSbKdPNNH0eHsEp
# f7RlZPRizkcK3w0Nj+u4kXdnnex3QLLSgnyAYq7zEl6V+mho8KqaBezkO7wQDHZy
# +GW5ignQ1Gs=
# =CyiZ
# -----END PGP SIGNATURE-----
# gpg: Signature made Fri 07 Mar 2025 19:51:55 HKT
# gpg:                using RSA key 27B88847EEE0250118F3EAB92ED9D774FE702DB5
# gpg:                issuer "[email protected]"
# gpg: Good signature from "Thomas Huth <[email protected]>" [full]
# gpg:                 aka "Thomas Huth <[email protected]>" [full]
# gpg:                 aka "Thomas Huth <[email protected]>" [full]
# gpg:                 aka "Thomas Huth <[email protected]>" [unknown]
# Primary key fingerprint: 27B8 8847 EEE0 2501 18F3  EAB9 2ED9 D774 FE70 2DB5

* tag 'pull-request-2025-03-07' of https://gitlab.com/thuth/qemu:
  s390x/pci: indicate QEMU supports relaxed translation for passthrough
  s390x/pci: add support for guests that request direct mapping
  MAINTAINERS: Add docs/devel/testing/functional.rst to the functional section
  doc: add missing 'Asset' type in function test doc
  tests/functional/test_virtio_balloon: Only use KVM for running this test
  tests/functional: fix race in virtio balloon test
  tests/functional: Increase the timeout of the mips64el_replay test
  tests/functional/test_mips_malta: Add a network test via the pcnet NIC
  tests/functional: Move the code for testing HTTP downloads to a common function
  tests/functional: stop output from zstd command when uncompressing
  tests/functional: drop unused 'get_tag' method
  tests/functional: skip memaddr tests on 32-bit builds
  tests/functional: reduce tuxrun maxmem to work on 32-bit hosts
  tests/functional: set 'qemu_bin' as an object level field
  tests/functional: remove unused 'bin_prefix' variable

Signed-off-by: Stefan Hajnoczi <[email protected]>

Author: stefanhaRH

MAINTAINERS

@@ -4236,6 +4236,7 @@ Functional testing framework
 M: Thomas Huth <[email protected]>
 R: Philippe Mathieu-Daudé <[email protected]>
 R: Daniel P. Berrange <[email protected]>
+F: docs/devel/testing/functional.rst
 F: tests/functional/qemu_test/
 
 Windows Hosted Continuous Integration

docs/devel/testing/functional.rst

@@ -173,7 +173,7 @@ QEMU binary selection
 ^^^^^^^^^^^^^^^^^^^^^
 
 The QEMU binary used for the ``self.vm`` QEMUMachine instance will
-primarily depend on the value of the ``qemu_bin`` class attribute.
+primarily depend on the value of the ``qemu_bin`` instance attribute.
 If it is not explicitly set by the test code, its default value will
 be the result the QEMU_TEST_QEMU_BINARY environment variable.
 
@@ -251,7 +251,7 @@ Many functional tests download assets (e.g. Linux kernels, initrds,
 firmware images, etc.) from the internet to be able to run tests with
 them. This imposes additional challenges to the test framework.
 
-First there is the the problem that some people might not have an
+First there is the problem that some people might not have an
 unconstrained internet connection, so such tests should not be run by
 default when running ``make check``. To accomplish this situation,
 the tests that download files should only be added to the "thorough"
@@ -274,7 +274,9 @@ the tests are run. This pre-caching is done with the qemu_test.Asset
 class. To use it in your test, declare an asset in your test class with
 its URL and SHA256 checksum like this::
 
-    ASSET_somename = (
+    from qemu_test import Asset
+
+    ASSET_somename = Asset(
         ('https://www.qemu.org/assets/images/qemu_head_200.png'),
         '34b74cad46ea28a2966c1d04e102510daf1fd73e6582b6b74523940d5da029dd')
 

hw/s390x/s390-pci-bus.c

@@ -18,6 +18,8 @@
 #include "hw/s390x/s390-pci-inst.h"
 #include "hw/s390x/s390-pci-kvm.h"
 #include "hw/s390x/s390-pci-vfio.h"
+#include "hw/s390x/s390-virtio-ccw.h"
+#include "hw/boards.h"
 #include "hw/pci/pci_bus.h"
 #include "hw/qdev-properties.h"
 #include "hw/pci/pci_bridge.h"
@@ -724,12 +726,42 @@ void s390_pci_iommu_enable(S390PCIIOMMU *iommu)
     g_free(name);
 }
 
+void s390_pci_iommu_direct_map_enable(S390PCIIOMMU *iommu)
+{
+    MachineState *ms = MACHINE(qdev_get_machine());
+    S390CcwMachineState *s390ms = S390_CCW_MACHINE(ms);
+
+    /*
+     * For direct-mapping we must map the entire guest address space.  Rather
+     * than using an iommu, create a memory region alias that maps GPA X to
+     * IOVA X + SDMA.  VFIO will handle pinning via its memory listener.
+     */
+    g_autofree char *name = g_strdup_printf("iommu-dm-s390-%04x",
+                                            iommu->pbdev->uid);
+
+    iommu->dm_mr = g_malloc0(sizeof(*iommu->dm_mr));
+    memory_region_init_alias(iommu->dm_mr, OBJECT(&iommu->mr), name,
+                             get_system_memory(), 0,
+                             s390_get_memory_limit(s390ms));
+    iommu->enabled = true;
+    memory_region_add_subregion(&iommu->mr, iommu->pbdev->zpci_fn.sdma,
+                                iommu->dm_mr);
+}
+
 void s390_pci_iommu_disable(S390PCIIOMMU *iommu)
 {
     iommu->enabled = false;
     g_hash_table_remove_all(iommu->iotlb);
-    memory_region_del_subregion(&iommu->mr, MEMORY_REGION(&iommu->iommu_mr));
-    object_unparent(OBJECT(&iommu->iommu_mr));
+    if (iommu->dm_mr) {
+        memory_region_del_subregion(&iommu->mr, iommu->dm_mr);
+        object_unparent(OBJECT(iommu->dm_mr));
+        g_free(iommu->dm_mr);
+        iommu->dm_mr = NULL;
+    } else {
+        memory_region_del_subregion(&iommu->mr,
+                                    MEMORY_REGION(&iommu->iommu_mr));
+        object_unparent(OBJECT(&iommu->iommu_mr));
+    }
 }
 
 static void s390_pci_iommu_free(S390pciState *s, PCIBus *bus, int32_t devfn)
@@ -1145,6 +1177,7 @@ static void s390_pcihost_plug(HotplugHandler *hotplug_dev, DeviceState *dev,
             /* Always intercept emulated devices */
             pbdev->interp = false;
             pbdev->forwarding_assist = false;
+            pbdev->rtr_avail = false;
         }
 
         if (s390_pci_msix_init(pbdev) && !pbdev->interp) {
@@ -1511,6 +1544,8 @@ static const Property s390_pci_device_properties[] = {
     DEFINE_PROP_BOOL("interpret", S390PCIBusDevice, interp, true),
     DEFINE_PROP_BOOL("forwarding-assist", S390PCIBusDevice, forwarding_assist,
                      true),
+    DEFINE_PROP_BOOL("relaxed-translation", S390PCIBusDevice, rtr_avail,
+                     true),
 };
 
 static const VMStateDescription s390_pci_device_vmstate = {

hw/s390x/s390-pci-inst.c

@@ -16,6 +16,7 @@
 #include "exec/memory.h"
 #include "qemu/error-report.h"
 #include "system/hw_accel.h"
+#include "hw/boards.h"
 #include "hw/pci/pci_device.h"
 #include "hw/s390x/s390-pci-inst.h"
 #include "hw/s390x/s390-pci-bus.h"
@@ -1008,17 +1009,25 @@ static int reg_ioat(CPUS390XState *env, S390PCIBusDevice *pbdev, ZpciFib fib,
     }
 
     /* currently we only support designation type 1 with translation */
-    if (!(dt == ZPCI_IOTA_RTTO && t)) {
+    if (t && dt != ZPCI_IOTA_RTTO) {
         error_report("unsupported ioat dt %d t %d", dt, t);
         s390_program_interrupt(env, PGM_OPERAND, ra);
         return -EINVAL;
+    } else if (!t && !pbdev->rtr_avail) {
+        error_report("relaxed translation not allowed");
+        s390_program_interrupt(env, PGM_OPERAND, ra);
+        return -EINVAL;
     }
 
     iommu->pba = pba;
     iommu->pal = pal;
     iommu->g_iota = g_iota;
 
-    s390_pci_iommu_enable(iommu);
+    if (t) {
+        s390_pci_iommu_enable(iommu);
+    } else {
+        s390_pci_iommu_direct_map_enable(iommu);
+    }
 
     return 0;
 }

hw/s390x/s390-pci-vfio.c

@@ -131,13 +131,28 @@ static void s390_pci_read_base(S390PCIBusDevice *pbdev,
     /* Store function type separately for type-specific behavior */
     pbdev->pft = cap->pft;
 
+    /*
+     * If the device is a passthrough ISM device, disallow relaxed
+     * translation.
+     */
+    if (pbdev->pft == ZPCI_PFT_ISM) {
+        pbdev->rtr_avail = false;
+    }
+
     /*
      * If appropriate, reduce the size of the supported DMA aperture reported
-     * to the guest based upon the vfio DMA limit.
+     * to the guest based upon the vfio DMA limit.  This is applicable for
+     * devices that are guaranteed to not use relaxed translation.  If the
+     * device is capable of relaxed translation then we must advertise the
+     * full aperture.  In this case, if translation is used then we will
+     * rely on the vfio DMA limit counting and use RPCIT CC1 / status 16
+     * to request that the guest free DMA mappings as necessary.
      */
-    vfio_size = pbdev->iommu->max_dma_limit << TARGET_PAGE_BITS;
-    if (vfio_size > 0 && vfio_size < cap->end_dma - cap->start_dma + 1) {
-        pbdev->zpci_fn.edma = cap->start_dma + vfio_size - 1;
+    if (!pbdev->rtr_avail) {
+        vfio_size = pbdev->iommu->max_dma_limit << TARGET_PAGE_BITS;
+        if (vfio_size > 0 && vfio_size < cap->end_dma - cap->start_dma + 1) {
+            pbdev->zpci_fn.edma = cap->start_dma + vfio_size - 1;
+        }
     }
 }
 
@@ -223,8 +238,11 @@ static void s390_pci_read_group(S390PCIBusDevice *pbdev,
         pbdev->pci_group = s390_group_create(pbdev->zpci_fn.pfgid, start_gid);
 
         resgrp = &pbdev->pci_group->zpci_group;
+        if (pbdev->rtr_avail) {
+            resgrp->fr |= CLP_RSP_QPCIG_MASK_RTR;
+        }
         if (cap->flags & VFIO_DEVICE_INFO_ZPCI_FLAG_REFRESH) {
-            resgrp->fr = 1;
+            resgrp->fr |= CLP_RSP_QPCIG_MASK_REFRESH;
         }
         resgrp->dasm = cap->dasm;
         resgrp->msia = cap->msi_addr;

hw/s390x/s390-virtio-ccw.c

@@ -936,8 +936,13 @@ static void ccw_machine_9_2_instance_options(MachineState *machine)
 
 static void ccw_machine_9_2_class_options(MachineClass *mc)
 {
+    static GlobalProperty compat[] = {
+        { TYPE_S390_PCI_DEVICE, "relaxed-translation", "off", },
+    };
+
     ccw_machine_10_0_class_options(mc);
     compat_props_add(mc->compat_props, hw_compat_9_2, hw_compat_9_2_len);
+    compat_props_add(mc->compat_props, compat, G_N_ELEMENTS(compat));
 }
 DEFINE_CCW_MACHINE(9, 2);
 

include/hw/s390x/s390-pci-bus.h

@@ -277,6 +277,7 @@ struct S390PCIIOMMU {
     AddressSpace as;
     MemoryRegion mr;
     IOMMUMemoryRegion iommu_mr;
+    MemoryRegion *dm_mr;
     bool enabled;
     uint64_t g_iota;
     uint64_t pba;
@@ -362,6 +363,7 @@ struct S390PCIBusDevice {
     bool interp;
     bool forwarding_assist;
     bool aif;
+    bool rtr_avail;
     QTAILQ_ENTRY(S390PCIBusDevice) link;
 };
 
@@ -389,6 +391,7 @@ int pci_chsc_sei_nt2_have_event(void);
 void s390_pci_sclp_configure(SCCB *sccb);
 void s390_pci_sclp_deconfigure(SCCB *sccb);
 void s390_pci_iommu_enable(S390PCIIOMMU *iommu);
+void s390_pci_iommu_direct_map_enable(S390PCIIOMMU *iommu);
 void s390_pci_iommu_disable(S390PCIIOMMU *iommu);
 void s390_pci_generate_error_event(uint16_t pec, uint32_t fh, uint32_t fid,
                                    uint64_t faddr, uint32_t e);

include/hw/s390x/s390-pci-clp.h

@@ -158,6 +158,7 @@ typedef struct ClpRspQueryPciGrp {
 #define CLP_RSP_QPCIG_MASK_NOI 0xfff
     uint16_t i;
     uint8_t version;
+#define CLP_RSP_QPCIG_MASK_RTR     0x20
 #define CLP_RSP_QPCIG_MASK_FRAME   0x2
 #define CLP_RSP_QPCIG_MASK_REFRESH 0x1
     uint8_t fr;

tests/functional/meson.build

@@ -36,6 +36,7 @@ test_timeouts = {
   'intel_iommu': 300,
   'mips_malta' : 120,
   'mipsel_replay' : 480,
+  'mips64el_replay' : 180,
   'netdev_ethtool' : 180,
   'ppc_40p' : 240,
   'ppc64_hv' : 1000,

tests/functional/qemu_test/linuxkernel.py

@@ -3,8 +3,12 @@
 # This work is licensed under the terms of the GNU GPL, version 2 or
 # later.  See the COPYING file in the top-level directory.
 
+import hashlib
+import urllib.request
+
+from .cmd import wait_for_console_pattern, exec_command_and_wait_for_pattern
 from .testcase import QemuSystemTest
-from .cmd import wait_for_console_pattern
+from .utils import get_usernet_hostfwd_port
 
 
 class LinuxKernelTest(QemuSystemTest):
@@ -26,3 +30,23 @@ def launch_kernel(self, kernel, initrd=None, dtb=None, console_index=0,
         self.vm.launch()
         if wait_for:
                 self.wait_for_console_pattern(wait_for)
+
+    def check_http_download(self, filename, hashsum, guestport=8080,
+                            pythoncmd='python3 -m http.server'):
+        exec_command_and_wait_for_pattern(self,
+                        f'{pythoncmd} {guestport} & sleep 1',
+                        f'Serving HTTP on 0.0.0.0 port {guestport}')
+        hl = hashlib.sha256()
+        hostport = get_usernet_hostfwd_port(self.vm)
+        url = f'http://localhost:{hostport}{filename}'
+        self.log.info(f'Downloading {url} ...')
+        with urllib.request.urlopen(url) as response:
+            while True:
+                chunk = response.read(1 << 20)
+                if not chunk:
+                    break
+                hl.update(chunk)
+
+        digest = hl.hexdigest()
+        self.log.info(f'sha256sum of download is {digest}.')
+        self.assertEqual(digest, hashsum)