[Qemu-devel] [PULL 0/7] target-arm queue

[Qemu-devel] [PULL 0/7] target-arm queue

[Peter Maydell]

target-arm queue: a few new features, but all minor stuff.

thanks
-- PMM


The following changes since commit ccb0c7e122db72d3a5da798c6414d4912bba828f:

  Merge remote-tracking branch 'remotes/lalrae/tags/mips-20150626' into staging (2015-06-26 11:32:58 +0100)

are available in the git repository at:


  git://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20150626

for you to fetch changes up to 4e2c0b2a4ab810c8989e181a010e75aeaa1c55f3:

  hw/arm/virt: Make block devices default to virtio (2015-06-26 14:22:37 +0100)

----------------------------------------------------------------
target-arm queue:
 * Change the virt board's default interface type for block devices to virtio
 * Improve some error messages that will now be triggered by some incorrect
   but previously worked-by-accident command lines
 * Print ELR if we're doing debug logging of AArch64 exception entry
 * Handle the "completely empty semihosting commandline" correctly for
   softmmu (we already did for linux-user)
 * Add GICv2m description to ACPI tables for virt board
 * Fix some incorrect table revision entries in virt board ACPI tables

----------------------------------------------------------------
Liviu Ionescu (1):
      target-arm: default empty semihosting cmdline

Peter Maydell (3):
      qdev-properties-system: Change set_pointer's parse callback to use Error
      qdev-properties-system: Improve error message for drive assignment conflict
      hw/arm/virt: Make block devices default to virtio

Shannon Zhao (2):
      hw/arm/virt-acpi-build: Fix table revision and some comments
      hw/arm/virt-acpi-build: Add GICv2m description in ACPI MADT table

Soren Brinkmann (1):
      target-arm: A64: Print ELR when taking exceptions

 hw/arm/virt-acpi-build.c         | 22 ++++++++++++++++-----
 hw/arm/virt.c                    |  2 ++
 hw/core/qdev-properties-system.c | 42 +++++++++++++++++++++++++++-------------
 include/hw/acpi/acpi-defs.h      | 12 ++++++++++++
 target-arm/arm-semi.c            | 11 +++++++++--
 target-arm/helper-a64.c          |  2 ++
 6 files changed, 71 insertions(+), 20 deletions(-)

[Qemu-devel] [PULL 1/7] hw/arm/virt-acpi-build: Fix table revision and some comments

[Peter Maydell]

From: Shannon Zhao <shannon.zhao@linaro.org>

The table revision is not the ACPI spec version. Fix the wrong revision
and also some comments.

Signed-off-by: Shannon Zhao <shannon.zhao@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-id: 1433820378-8336-1-git-send-email-zhaoshenglong@huawei.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 hw/arm/virt-acpi-build.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/hw/arm/virt-acpi-build.c b/hw/arm/virt-acpi-build.c
index d5a8b9c..40029dd 100644
--- a/hw/arm/virt-acpi-build.c
+++ b/hw/arm/virt-acpi-build.c
@@ -387,7 +387,7 @@ build_mcfg(GArray *table_data, GArray *linker, VirtGuestInfo *guest_info)
     mcfg->allocation[0].end_bus_number = (memmap[VIRT_PCIE_ECAM].size
                                           / PCIE_MMCFG_SIZE_MIN) - 1;
 
-    build_header(linker, table_data, (void *)mcfg, "MCFG", len, 5);
+    build_header(linker, table_data, (void *)mcfg, "MCFG", len, 1);
 }
 
 /* GTDT */
@@ -413,7 +413,7 @@ build_gtdt(GArray *table_data, GArray *linker)
 
     build_header(linker, table_data,
                  (void *)(table_data->data + gtdt_start), "GTDT",
-                 table_data->len - gtdt_start, 5);
+                 table_data->len - gtdt_start, 2);
 }
 
 /* MADT */
@@ -450,7 +450,7 @@ build_madt(GArray *table_data, GArray *linker, VirtGuestInfo *guest_info,
 
     build_header(linker, table_data,
                  (void *)(table_data->data + madt_start), "APIC",
-                 table_data->len - madt_start, 5);
+                 table_data->len - madt_start, 3);
 }
 
 /* FADT */
@@ -507,7 +507,7 @@ build_dsdt(GArray *table_data, GArray *linker, VirtGuestInfo *guest_info)
     g_array_append_vals(table_data, dsdt->buf->data, dsdt->buf->len);
     build_header(linker, table_data,
         (void *)(table_data->data + table_data->len - dsdt->buf->len),
-        "DSDT", dsdt->buf->len, 5);
+        "DSDT", dsdt->buf->len, 2);
     free_aml_allocator();
 }
 
@@ -545,6 +545,7 @@ void virt_acpi_build(VirtGuestInfo *guest_info, AcpiBuildTables *tables)
      * FADT
      * GTDT
      * MADT
+     * MCFG
      * DSDT
      */
 
@@ -552,7 +553,7 @@ void virt_acpi_build(VirtGuestInfo *guest_info, AcpiBuildTables *tables)
     dsdt = tables_blob->len;
     build_dsdt(tables_blob, tables->linker, guest_info);
 
-    /* FADT MADT GTDT SPCR pointed to by RSDT */
+    /* FADT MADT GTDT MCFG SPCR pointed to by RSDT */
     acpi_add_table(table_offsets, tables_blob);
     build_fadt(tables_blob, tables->linker, dsdt);
 
-- 
1.9.1

[Qemu-devel] [PULL 2/7] hw/arm/virt-acpi-build: Add GICv2m description in ACPI MADT table

[Peter Maydell]

From: Shannon Zhao <shannon.zhao@linaro.org>

Add GICv2m description in ACPI MADT table, so guest can use MSI when
booting with ACPI.

Signed-off-by: Shannon Zhao <zhaoshenglong@huawei.com>
Signed-off-by: Shannon Zhao <shannon.zhao@linaro.org>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Tested-by: Andrew Jones <drjones@redhat.com>
Message-id: 1434676210-2276-1-git-send-email-shannon.zhao@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 hw/arm/virt-acpi-build.c    | 11 +++++++++++
 include/hw/acpi/acpi-defs.h | 12 ++++++++++++
 2 files changed, 23 insertions(+)

diff --git a/hw/arm/virt-acpi-build.c b/hw/arm/virt-acpi-build.c
index 40029dd..f365140 100644
--- a/hw/arm/virt-acpi-build.c
+++ b/hw/arm/virt-acpi-build.c
@@ -423,8 +423,10 @@ build_madt(GArray *table_data, GArray *linker, VirtGuestInfo *guest_info,
 {
     int madt_start = table_data->len;
     const MemMapEntry *memmap = guest_info->memmap;
+    const int *irqmap = guest_info->irqmap;
     AcpiMultipleApicTable *madt;
     AcpiMadtGenericDistributor *gicd;
+    AcpiMadtGenericMsiFrame *gic_msi;
     int i;
 
     madt = acpi_data_push(table_data, sizeof *madt);
@@ -448,6 +450,15 @@ build_madt(GArray *table_data, GArray *linker, VirtGuestInfo *guest_info,
     gicd->length = sizeof(*gicd);
     gicd->base_address = memmap[VIRT_GIC_DIST].base;
 
+    gic_msi = acpi_data_push(table_data, sizeof *gic_msi);
+    gic_msi->type = ACPI_APIC_GENERIC_MSI_FRAME;
+    gic_msi->length = sizeof(*gic_msi);
+    gic_msi->gic_msi_frame_id = 0;
+    gic_msi->base_address = cpu_to_le64(memmap[VIRT_GIC_V2M].base);
+    gic_msi->flags = cpu_to_le32(1);
+    gic_msi->spi_count = cpu_to_le16(NUM_GICV2M_SPIS);
+    gic_msi->spi_base = cpu_to_le16(irqmap[VIRT_GIC_V2M] + ARM_SPI_BASE);
+
     build_header(linker, table_data,
                  (void *)(table_data->data + madt_start), "APIC",
                  table_data->len - madt_start, 3);
diff --git a/include/hw/acpi/acpi-defs.h b/include/hw/acpi/acpi-defs.h
index 7b4bfb7..2b431e6 100644
--- a/include/hw/acpi/acpi-defs.h
+++ b/include/hw/acpi/acpi-defs.h
@@ -372,6 +372,18 @@ struct AcpiMadtGenericDistributor {
 
 typedef struct AcpiMadtGenericDistributor AcpiMadtGenericDistributor;
 
+struct AcpiMadtGenericMsiFrame {
+    ACPI_SUB_HEADER_DEF
+    uint16_t reserved;
+    uint32_t gic_msi_frame_id;
+    uint64_t base_address;
+    uint32_t flags;
+    uint16_t spi_count;
+    uint16_t spi_base;
+} QEMU_PACKED;
+
+typedef struct AcpiMadtGenericMsiFrame AcpiMadtGenericMsiFrame;
+
 /*
  * Generic Timer Description Table (GTDT)
  */
-- 
1.9.1

[Qemu-devel] [PULL 3/7] target-arm: default empty semihosting cmdline

[Peter Maydell]

From: Liviu Ionescu <ilg@livius.net>

If neither explicit semihosting args nor -kernel are used,
make SYS_GET_CMDLINE return an empty string.

Signed-off-by: Liviu Ionescu <ilg@livius.net>
Message-id: AC7B5AFC-06AE-4FAD-9852-B65708E80E09@livius.net
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target-arm/arm-semi.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/target-arm/arm-semi.c b/target-arm/arm-semi.c
index 74a67e9..a2a7369 100644
--- a/target-arm/arm-semi.c
+++ b/target-arm/arm-semi.c
@@ -436,12 +436,19 @@ uint32_t do_arm_semihosting(CPUARMState *env)
             size_t input_size;
             size_t output_size;
             int status = 0;
+#if !defined(CONFIG_USER_ONLY)
+            const char *cmdline;
+#endif
             GET_ARG(0);
             GET_ARG(1);
             input_size = arg1;
             /* Compute the size of the output string.  */
 #if !defined(CONFIG_USER_ONLY)
-            output_size = strlen(semihosting_get_cmdline()) + 1;
+            cmdline = semihosting_get_cmdline();
+            if (cmdline == NULL) {
+                cmdline = ""; /* Default to an empty line. */
+            }
+            output_size = strlen(cmdline) + 1; /* Count terminating 0. */
 #else
             unsigned int i;
 
@@ -472,7 +479,7 @@ uint32_t do_arm_semihosting(CPUARMState *env)
 
             /* Copy the command-line arguments.  */
 #if !defined(CONFIG_USER_ONLY)
-            pstrcpy(output_buffer, output_size, semihosting_get_cmdline());
+            pstrcpy(output_buffer, output_size, cmdline);
 #else
             if (output_size == 1) {
                 /* Empty command-line.  */
-- 
1.9.1

[Qemu-devel] [PULL 4/7] target-arm: A64: Print ELR when taking exceptions

[Peter Maydell]

From: Soren Brinkmann <soren.brinkmann@xilinx.com>

When taking an exception print the content of the exception link
register. This is useful especially for synchronous exceptions because
in that case this registers holds the address of the instruction that
generated the exception.

Signed-off-by: Soren Brinkmann <soren.brinkmann@xilinx.com>
Message-id: 1435036655-16132-1-git-send-email-soren.brinkmann@xilinx.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
---
 target-arm/helper-a64.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/target-arm/helper-a64.c b/target-arm/helper-a64.c
index e30af06..08c95a3 100644
--- a/target-arm/helper-a64.c
+++ b/target-arm/helper-a64.c
@@ -533,6 +533,8 @@ void aarch64_cpu_do_interrupt(CPUState *cs)
 
         env->condexec_bits = 0;
     }
+    qemu_log_mask(CPU_LOG_INT, "...with ELR 0x%" PRIx64 "\n",
+                  env->elr_el[new_el]);
 
     pstate_write(env, PSTATE_DAIF | new_mode);
     env->aarch64 = 1;
-- 
1.9.1

[Qemu-devel] [PULL 5/7] qdev-properties-system: Change set_pointer's parse callback to use Error

[Peter Maydell]

Instead of having set_pointer() call a parse callback which returns
an error number that we then convert to an Error string with
error_set_from_qdev_prop_error(), make the parse callback take an
Error** and set the error itself. This will allow parse routines
to provide more helpful error messages than the generic ones.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-id: 1435068107-12594-2-git-send-email-peter.maydell@linaro.org
---
 hw/core/qdev-properties-system.c | 33 ++++++++++++++++++++-------------
 1 file changed, 20 insertions(+), 13 deletions(-)

diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index aa794ca..c731bc6 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -35,15 +35,15 @@ static void get_pointer(Object *obj, Visitor *v, Property *prop,
 }
 
 static void set_pointer(Object *obj, Visitor *v, Property *prop,
-                        int (*parse)(DeviceState *dev, const char *str,
-                                     void **ptr),
+                        void (*parse)(DeviceState *dev, const char *str,
+                                      void **ptr, const char *propname,
+                                      Error **errp),
                         const char *name, Error **errp)
 {
     DeviceState *dev = DEVICE(obj);
     Error *local_err = NULL;
     void **ptr = qdev_get_prop_ptr(dev, prop);
     char *str;
-    int ret;
 
     if (dev->realized) {
         qdev_prop_set_after_realize(dev, name, errp);
@@ -60,26 +60,29 @@ static void set_pointer(Object *obj, Visitor *v, Property *prop,
         *ptr = NULL;
         return;
     }
-    ret = parse(dev, str, ptr);
-    error_set_from_qdev_prop_error(errp, ret, dev, prop, str);
+    parse(dev, str, ptr, prop->name, errp);
     g_free(str);
 }
 
 /* --- drive --- */
 
-static int parse_drive(DeviceState *dev, const char *str, void **ptr)
+static void parse_drive(DeviceState *dev, const char *str, void **ptr,
+                        const char *propname, Error **errp)
 {
     BlockBackend *blk;
 
     blk = blk_by_name(str);
     if (!blk) {
-        return -ENOENT;
+        error_setg(errp, "Property '%s.%s' can't find value '%s'",
+                   object_get_typename(OBJECT(dev)), propname, str);
+        return;
     }
     if (blk_attach_dev(blk, dev) < 0) {
-        return -EEXIST;
+        error_setg(errp, "Property '%s.%s' can't take value '%s', it's in use",
+                  object_get_typename(OBJECT(dev)), propname, str);
+        return;
     }
     *ptr = blk;
-    return 0;
 }
 
 static void release_drive(Object *obj, const char *name, void *opaque)
@@ -121,17 +124,21 @@ PropertyInfo qdev_prop_drive = {
 
 /* --- character device --- */
 
-static int parse_chr(DeviceState *dev, const char *str, void **ptr)
+static void parse_chr(DeviceState *dev, const char *str, void **ptr,
+                      const char *propname, Error **errp)
 {
     CharDriverState *chr = qemu_chr_find(str);
     if (chr == NULL) {
-        return -ENOENT;
+        error_setg(errp, "Property '%s.%s' can't find value '%s'",
+                   object_get_typename(OBJECT(dev)), propname, str);
+        return;
     }
     if (qemu_chr_fe_claim(chr) != 0) {
-        return -EEXIST;
+        error_setg(errp, "Property '%s.%s' can't take value '%s', it's in use",
+                  object_get_typename(OBJECT(dev)), propname, str);
+        return;
     }
     *ptr = chr;
-    return 0;
 }
 
 static void release_chr(Object *obj, const char *name, void *opaque)
-- 
1.9.1

[Qemu-devel] [PULL 6/7] qdev-properties-system: Improve error message for drive assignment conflict

[Peter Maydell]

If the user forgot if=none on their drive specification they're likely
to get an error message because the drive is assigned once automatically
by QEMU and once by the manual id=/drive= user command line specification.
Improve the error message produced in this case to explicitly guide the
user towards if=none.

We rephrase the "drive conflict but not for an if=something" error as
well to keep the wording in line.

The two cases that change are:

(1) Drive specified as to be auto-connected and also manually connected
(and the board does handle this if= type):

  qemu-system-x86_64 -nodefaults -display none \
     -drive if=scsi,file=tmp.qcow2,id=foo -device ide-hd,drive=foo

Previously:
  qemu-system-x86_64: -device ide-hd,drive=foo: Property 'ide-hd.drive'
  can't take value 'foo', it's in use

Now:
  qemu-system-x86_64: -device ide-hd,drive=foo: Drive 'foo' is already in
  use because it has been automatically connected to another device (did
  you need 'if=none' in the drive options?)

(2) Drive specified to be manually connected in two different ways:

  qemu-system-x86_64 -nodefaults -display none \
   -drive if=none,file=tmp.qcow2,id=foo -device ide-hd,drive=foo \
   -device ide-hd,drive=foo

Previously:
  qemu-system-x86_64: -device ide-hd,drive=foo: Property 'ide-hd.drive'
  can't take value 'foo', it's in use

Now:
  qemu-system-x86_64: -device ide-hd,drive=foo: Drive 'foo' is already in
  use by another device

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-id: 1435068107-12594-3-git-send-email-peter.maydell@linaro.org
---
 hw/core/qdev-properties-system.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/hw/core/qdev-properties-system.c b/hw/core/qdev-properties-system.c
index c731bc6..921e799 100644
--- a/hw/core/qdev-properties-system.c
+++ b/hw/core/qdev-properties-system.c
@@ -78,8 +78,17 @@ static void parse_drive(DeviceState *dev, const char *str, void **ptr,
         return;
     }
     if (blk_attach_dev(blk, dev) < 0) {
-        error_setg(errp, "Property '%s.%s' can't take value '%s', it's in use",
-                  object_get_typename(OBJECT(dev)), propname, str);
+        DriveInfo *dinfo = blk_legacy_dinfo(blk);
+
+        if (dinfo->type != IF_NONE) {
+            error_setg(errp, "Drive '%s' is already in use because "
+                       "it has been automatically connected to another "
+                       "device (did you need 'if=none' in the drive options?)",
+                       str);
+        } else {
+            error_setg(errp, "Drive '%s' is already in use by another device",
+                       str);
+        }
         return;
     }
     *ptr = blk;
-- 
1.9.1

[Qemu-devel] [PULL 7/7] hw/arm/virt: Make block devices default to virtio

[Peter Maydell]

Now we have virtio-pci, we can make the virt board's default block
device type be IF_VIRTIO. This allows users to use simplified
command lines that don't have to explicitly create virtio-pci-blk
devices; the -hda &c very short options now also work.

This means we also need to set no_cdrom to avoid getting a
default cdrom device -- this is needed because the virtio-blk
device will fail if it is connected to a block backend with
no media, which is what the default cdrom device typically is.
Providing a cdrom with media via -cdrom will succeed, but silently
create a device with non-removable medium. this is probably
not really what the user wants, but is the best we can do now.

Note that this change means that some command lines which used
to work (by accident) will stop working. Where a drive was connected
manually to a device but without 'if=none' being specified, we
used to treat this as an IDE drive, which we would then not autoplug
because the board doesn't support IDE. Now we will treat it as a
virtio disk and autoplug it, which means the attempt to use the
drive manually will fail:
  qemu-system-arm: -drive file=img.qcow2,id=foo: Drive 'foo' is already
  in use because it has been automatically connected to another device
  (did you need 'if=none' in the drive options?)
The command line will have to be changed to include 'if=none', as the
error message suggests.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-id: 1435068107-12594-4-git-send-email-peter.maydell@linaro.org
---
 hw/arm/virt.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/hw/arm/virt.c b/hw/arm/virt.c
index 4e78083a..4846892 100644
--- a/hw/arm/virt.c
+++ b/hw/arm/virt.c
@@ -956,6 +956,8 @@ static void virt_class_init(ObjectClass *oc, void *data)
     mc->init = machvirt_init;
     mc->max_cpus = 8;
     mc->has_dynamic_sysbus = true;
+    mc->block_default_type = IF_VIRTIO;
+    mc->no_cdrom = 1;
 }
 
 static const TypeInfo machvirt_info = {
-- 
1.9.1

Re: [Qemu-devel] [PULL 0/7] target-arm queue

[Peter Maydell]

On 26 June 2015 at 14:31, Peter Maydell <peter.maydell@linaro.org> wrote:
> target-arm queue: a few new features, but all minor stuff.
>
> thanks
> -- PMM
>
>
> The following changes since commit ccb0c7e122db72d3a5da798c6414d4912bba828f:
>
>   Merge remote-tracking branch 'remotes/lalrae/tags/mips-20150626' into staging (2015-06-26 11:32:58 +0100)
>
> are available in the git repository at:
>
>
>   git://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20150626
>
> for you to fetch changes up to 4e2c0b2a4ab810c8989e181a010e75aeaa1c55f3:
>
>   hw/arm/virt: Make block devices default to virtio (2015-06-26 14:22:37 +0100)
>
> ----------------------------------------------------------------
> target-arm queue:
>  * Change the virt board's default interface type for block devices to virtio
>  * Improve some error messages that will now be triggered by some incorrect
>    but previously worked-by-accident command lines
>  * Print ELR if we're doing debug logging of AArch64 exception entry
>  * Handle the "completely empty semihosting commandline" correctly for
>    softmmu (we already did for linux-user)
>  * Add GICv2m description to ACPI tables for virt board
>  * Fix some incorrect table revision entries in virt board ACPI tables
>
> ----------------------------------------------------------------

Applied, thanks.

-- PMM