[Qemu-devel] [PULL 0/4] test and debug devices

[Qemu-devel] [PULL 0/4] test and debug devices

[Gerd Hoffmann]

  Hi,

This patch series improves the debug devices we have.  isa-debugcon is
switched over to the memory api.  isa-debug-exit is added and replaces
the hard-coded bochs debug port.  pc-testdev (used by kvm-unit-tests) is
added.

please pull,
  Gerd

The following changes since commit 25bbf61e4bacd1e4fa4115ffcf151051b9d6608e:

  pty: unbreak libvirt (2013-01-03 12:53:41 -0600)

are available in the git repository at:
  git://git.kraxel.org/qemu testdev.1

Gerd Hoffmann (3):
      switch debugcon to memory api
      add isa-debug-exit device.
      pc: remove bochs bios debug ports

Lucas Meneghel Rodrigues (1):
      hw: Add test device for unittests execution

 hw/debugcon.c         |   31 +++++++--
 hw/debugexit.c        |   75 ++++++++++++++++++++
 hw/i386/Makefile.objs |    3 +-
 hw/pc-testdev.c       |  182 +++++++++++++++++++++++++++++++++++++++++++++++++
 hw/pc.c               |   41 -----------
 5 files changed, 283 insertions(+), 49 deletions(-)
 create mode 100644 hw/debugexit.c
 create mode 100644 hw/pc-testdev.c

[Qemu-devel] [PATCH 1/4] switch debugcon to memory api

[Gerd Hoffmann]

Also some QOM glue while being at it.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
---
 hw/debugcon.c |   31 ++++++++++++++++++++++++-------
 1 files changed, 24 insertions(+), 7 deletions(-)

diff --git a/hw/debugcon.c b/hw/debugcon.c
index 14f83f1..e8a855e 100644
--- a/hw/debugcon.c
+++ b/hw/debugcon.c
@@ -29,20 +29,27 @@
 #include "isa.h"
 #include "pc.h"
 
+#define TYPE_ISA_DEBUGCON_DEVICE "isa-debugcon"
+#define ISA_DEBUGCON_DEVICE(obj) \
+     OBJECT_CHECK(ISADebugconState, (obj), TYPE_ISA_DEBUGCON_DEVICE)
+
 //#define DEBUG_DEBUGCON
 
 typedef struct DebugconState {
+    MemoryRegion io;
     CharDriverState *chr;
     uint32_t readback;
 } DebugconState;
 
 typedef struct ISADebugconState {
-    ISADevice dev;
+    ISADevice parent_obj;
+
     uint32_t iobase;
     DebugconState state;
 } ISADebugconState;
 
-static void debugcon_ioport_write(void *opaque, uint32_t addr, uint32_t val)
+static void debugcon_ioport_write(void *opaque, hwaddr addr, uint64_t val,
+                                  unsigned width)
 {
     DebugconState *s = opaque;
     unsigned char ch = val;
@@ -55,7 +62,7 @@ static void debugcon_ioport_write(void *opaque, uint32_t addr, uint32_t val)
 }
 
 
-static uint32_t debugcon_ioport_read(void *opaque, uint32_t addr)
+static uint64_t debugcon_ioport_read(void *opaque, hwaddr addr, unsigned width)
 {
     DebugconState *s = opaque;
 
@@ -66,6 +73,14 @@ static uint32_t debugcon_ioport_read(void *opaque, uint32_t addr)
     return s->readback;
 }
 
+static const MemoryRegionOps debugcon_ops = {
+    .read = debugcon_ioport_read,
+    .write = debugcon_ioport_write,
+    .valid.min_access_size = 1,
+    .valid.max_access_size = 1,
+    .endianness = DEVICE_LITTLE_ENDIAN,
+};
+
 static void debugcon_init_core(DebugconState *s)
 {
     if (!s->chr) {
@@ -78,12 +93,14 @@ static void debugcon_init_core(DebugconState *s)
 
 static int debugcon_isa_initfn(ISADevice *dev)
 {
-    ISADebugconState *isa = DO_UPCAST(ISADebugconState, dev, dev);
+    ISADebugconState *isa = ISA_DEBUGCON_DEVICE(dev);
     DebugconState *s = &isa->state;
 
     debugcon_init_core(s);
-    register_ioport_write(isa->iobase, 1, 1, debugcon_ioport_write, s);
-    register_ioport_read(isa->iobase, 1, 1, debugcon_ioport_read, s);
+    memory_region_init_io(&s->io, &debugcon_ops, s,
+                          TYPE_ISA_DEBUGCON_DEVICE, 1);
+    memory_region_add_subregion(isa_address_space_io(dev),
+                                isa->iobase, &s->io);
     return 0;
 }
 
@@ -103,7 +120,7 @@ static void debugcon_isa_class_initfn(ObjectClass *klass, void *data)
 }
 
 static TypeInfo debugcon_isa_info = {
-    .name          = "isa-debugcon",
+    .name          = TYPE_ISA_DEBUGCON_DEVICE,
     .parent        = TYPE_ISA_DEVICE,
     .instance_size = sizeof(ISADebugconState),
     .class_init    = debugcon_isa_class_initfn,
-- 
1.7.1

Re: [Qemu-devel] [PATCH 1/4] switch debugcon to memory api

[Andreas Färber]

Am 04.01.2013 09:16, schrieb Gerd Hoffmann:
> Also some QOM glue while being at it.
> 
> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
> ---
>  hw/debugcon.c |   31 ++++++++++++++++++++++++-------
>  1 files changed, 24 insertions(+), 7 deletions(-)
> 
> diff --git a/hw/debugcon.c b/hw/debugcon.c
> index 14f83f1..e8a855e 100644
> --- a/hw/debugcon.c
> +++ b/hw/debugcon.c
> @@ -29,20 +29,27 @@
>  #include "isa.h"
>  #include "pc.h"
>  
> +#define TYPE_ISA_DEBUGCON_DEVICE "isa-debugcon"
> +#define ISA_DEBUGCON_DEVICE(obj) \
> +     OBJECT_CHECK(ISADebugconState, (obj), TYPE_ISA_DEBUGCON_DEVICE)

Note that my previous QOM'ification RFC used the more verbose
[TYPE_]ISA_DEBUG_CONSOLE to avoid the "debugcon" abbreviation. Care to
cleanup?

Otherwise looks good, thanks, and would've got a Reviewed-by had you not
rushed to send a PULL.

Andreas

-- 
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg

[Qemu-devel] [PATCH 2/4] add isa-debug-exit device.

[Gerd Hoffmann]

When present it makes qemu exit on any write.
Mapped to port 0x501 by default.

Without this patch Anthony doesn't allow me to
remove the bochs bios debug ports because his
test suite uses this.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
---
 hw/debugexit.c        |   75 +++++++++++++++++++++++++++++++++++++++++++++++++
 hw/i386/Makefile.objs |    2 +-
 2 files changed, 76 insertions(+), 1 deletions(-)
 create mode 100644 hw/debugexit.c

diff --git a/hw/debugexit.c b/hw/debugexit.c
new file mode 100644
index 0000000..90642eb
--- /dev/null
+++ b/hw/debugexit.c
@@ -0,0 +1,75 @@
+/*
+ * debug exit port emulation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 or
+ * (at your option) any later version.
+ */
+
+#include "hw.h"
+#include "isa.h"
+
+#define TYPE_ISA_DEBUG_EXIT_DEVICE "isa-debug-exit"
+#define ISA_DEBUG_EXIT_DEVICE(obj) \
+     OBJECT_CHECK(ISADebugExitState, (obj), TYPE_ISA_DEBUG_EXIT_DEVICE)
+
+typedef struct ISADebugExitState {
+    ISADevice parent_obj;
+
+    uint32_t iobase;
+    uint32_t iosize;
+    MemoryRegion io;
+} ISADebugExitState;
+
+static void debug_exit_write(void *opaque, hwaddr addr, uint64_t val,
+                             unsigned width)
+{
+    exit((val << 1) | 1);
+}
+
+static const MemoryRegionOps debug_exit_ops = {
+    .write = debug_exit_write,
+    .valid.min_access_size = 1,
+    .valid.max_access_size = 4,
+    .endianness = DEVICE_LITTLE_ENDIAN,
+};
+
+static int debug_exit_initfn(ISADevice *dev)
+{
+    ISADebugExitState *isa = ISA_DEBUG_EXIT_DEVICE(dev);
+
+    memory_region_init_io(&isa->io, &debug_exit_ops, isa,
+                          TYPE_ISA_DEBUG_EXIT_DEVICE, isa->iosize);
+    memory_region_add_subregion(isa_address_space_io(dev),
+                                isa->iobase, &isa->io);
+    return 0;
+}
+
+static Property debug_exit_properties[] = {
+    DEFINE_PROP_HEX32("iobase", ISADebugExitState, iobase, 0x501),
+    DEFINE_PROP_HEX32("iosize", ISADebugExitState, iosize, 0x02),
+    DEFINE_PROP_END_OF_LIST(),
+};
+
+static void debug_exit_class_initfn(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    ISADeviceClass *ic = ISA_DEVICE_CLASS(klass);
+    ic->init = debug_exit_initfn;
+    dc->props = debug_exit_properties;
+}
+
+static TypeInfo debug_exit_info = {
+    .name          = TYPE_ISA_DEBUG_EXIT_DEVICE,
+    .parent        = TYPE_ISA_DEVICE,
+    .instance_size = sizeof(ISADebugExitState),
+    .class_init    = debug_exit_class_initfn,
+};
+
+static void debug_exit_register_types(void)
+{
+    type_register_static(&debug_exit_info);
+}
+
+type_init(debug_exit_register_types)
diff --git a/hw/i386/Makefile.objs b/hw/i386/Makefile.objs
index 257f3c1..2ba04db 100644
--- a/hw/i386/Makefile.objs
+++ b/hw/i386/Makefile.objs
@@ -3,7 +3,7 @@ obj-y += apic_common.o apic.o kvmvapic.o
 obj-y += sga.o ioapic_common.o ioapic.o piix_pci.o
 obj-y += vmport.o
 obj-y += pci/pci-hotplug.o smbios.o wdt_ib700.o
-obj-y += debugcon.o multiboot.o
+obj-y += debugcon.o debugexit.o multiboot.o
 obj-y += pc_piix.o
 obj-y += pc_sysfw.o
 obj-y += lpc_ich9.o q35.o pc_q35.o
-- 
1.7.1

[Qemu-devel] [PATCH 3/4] hw: Add test device for unittests execution

[Gerd Hoffmann]

From: Lucas Meneghel Rodrigues <lmr@redhat.com>

Add a test device which supports the kvmctl ioports,
so one can run the KVM unittest suite.

Intended Usage:

qemu-system-x86_64 -nographic \
    -device pc-testdev \
    -device isa-debug-exit,iobase=0xf4,iosize=0x04 \
    -kernel /path/to/kvm/unittests/msr.flat

Where msr.flat is one of the KVM unittests, present on a
separate repo,

git://git.kernel.org/pub/scm/virt/kvm/kvm-unit-tests.git

[ kraxel: more memory api + qom fixes ]

CC: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Lucas Meneghel Rodrigues <lmr@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
---
 hw/i386/Makefile.objs |    1 +
 hw/pc-testdev.c       |  182 +++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 183 insertions(+), 0 deletions(-)
 create mode 100644 hw/pc-testdev.c

diff --git a/hw/i386/Makefile.objs b/hw/i386/Makefile.objs
index 2ba04db..025803a 100644
--- a/hw/i386/Makefile.objs
+++ b/hw/i386/Makefile.objs
@@ -12,5 +12,6 @@ obj-$(CONFIG_XEN_PCI_PASSTHROUGH) += xen-host-pci-device.o
 obj-$(CONFIG_XEN_PCI_PASSTHROUGH) += xen_pt.o xen_pt_config_init.o xen_pt_msi.o
 obj-y += kvm/
 obj-$(CONFIG_SPICE) += qxl.o qxl-logger.o qxl-render.o
+obj-y += pc-testdev.o
 
 obj-y := $(addprefix ../,$(obj-y))
diff --git a/hw/pc-testdev.c b/hw/pc-testdev.c
new file mode 100644
index 0000000..620c86c
--- /dev/null
+++ b/hw/pc-testdev.c
@@ -0,0 +1,182 @@
+/*
+ * QEMU x86 ISA testdev
+ *
+ * Copyright (c) 2012 Avi Kivity, Gerd Hoffmann, Marcelo Tosatti
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+/*
+ * This device is used to test KVM features specific to the x86 port, such
+ * as emulation, power management, interrupt routing, among others. It's meant
+ * to be used like:
+ *
+ * qemu-system-x86_64 -device pc-testdev -serial stdio \
+ * -device isa-debug-exit,iobase=0xf4,iosize=0x4 \
+ * -kernel /home/lmr/Code/virt-test.git/kvm/unittests/msr.flat
+ *
+ * Where msr.flat is one of the KVM unittests, present on a separate repo,
+ * git://git.kernel.org/pub/scm/virt/kvm/kvm-unit-tests.git
+*/
+
+#include <sys/mman.h>
+#include "hw.h"
+#include "qdev.h"
+#include "isa.h"
+
+#define IOMEM_LEN    0x10000
+
+typedef struct PCTestdev {
+    ISADevice parent_obj;
+
+    MemoryRegion ioport;
+    MemoryRegion flush;
+    MemoryRegion irq;
+    MemoryRegion iomem;
+    uint32_t ioport_data;
+    char iomem_buf[IOMEM_LEN];
+} PCTestdev;
+
+#define TYPE_TESTDEV "pc-testdev"
+#define TESTDEV(obj) \
+     OBJECT_CHECK(struct PCTestdev, (obj), TYPE_TESTDEV)
+
+static void test_irq_line(void *opaque, hwaddr addr, uint64_t data,
+                          unsigned len)
+{
+    struct PCTestdev *dev = opaque;
+    struct ISADevice *isa = ISA_DEVICE(dev);
+
+    qemu_set_irq(isa_get_irq(isa, addr), !!data);
+}
+
+static const MemoryRegionOps test_irq_ops = {
+    .write = test_irq_line,
+    .valid.min_access_size = 1,
+    .valid.max_access_size = 1,
+    .endianness = DEVICE_LITTLE_ENDIAN,
+};
+
+static void test_ioport_write(void *opaque, hwaddr addr, uint64_t data,
+                              unsigned len)
+{
+    struct PCTestdev *dev = opaque;
+    dev->ioport_data = data;
+}
+
+static uint64_t test_ioport_read(void *opaque, hwaddr addr, unsigned len)
+{
+    struct PCTestdev *dev = opaque;
+    return dev->ioport_data;
+}
+
+static const MemoryRegionOps test_ioport_ops = {
+    .read = test_ioport_read,
+    .write = test_ioport_write,
+    .endianness = DEVICE_LITTLE_ENDIAN,
+};
+
+static void test_flush_page(void *opaque, hwaddr addr, uint64_t data,
+                            unsigned len)
+{
+    hwaddr page = 4096;
+    void *a = cpu_physical_memory_map(data & ~0xffful, &page, 0);
+
+    /* We might not be able to get the full page, only mprotect what we actually
+       have mapped */
+    mprotect(a, page, PROT_NONE);
+    mprotect(a, page, PROT_READ|PROT_WRITE);
+    cpu_physical_memory_unmap(a, page, 0, 0);
+}
+
+static const MemoryRegionOps test_flush_ops = {
+    .write = test_flush_page,
+    .valid.min_access_size = 4,
+    .valid.max_access_size = 4,
+    .endianness = DEVICE_LITTLE_ENDIAN,
+};
+
+static uint64_t test_iomem_read(void *opaque, hwaddr addr, unsigned len)
+{
+    struct PCTestdev *dev = opaque;
+    uint64_t ret = 0;
+    memcpy(&ret, &dev->iomem_buf[addr], len);
+    ret = le64_to_cpu(ret);
+
+    return ret;
+}
+
+static void test_iomem_write(void *opaque, hwaddr addr, uint64_t val,
+                             unsigned len)
+{
+    struct PCTestdev *dev = opaque;
+    val = cpu_to_le64(val);
+    memcpy(&dev->iomem_buf[addr], &val, len);
+    dev->iomem_buf[addr] = val;
+}
+
+static const MemoryRegionOps test_iomem_ops = {
+    .read = test_iomem_read,
+    .write = test_iomem_write,
+    .endianness = DEVICE_LITTLE_ENDIAN,
+};
+
+static int init_test_device(ISADevice *isa)
+{
+    struct PCTestdev *dev = TESTDEV(isa);
+    MemoryRegion *mem = isa_address_space(isa);
+    MemoryRegion *io = isa_address_space_io(isa);
+
+    memory_region_init_io(&dev->ioport, &test_ioport_ops, dev,
+                          "pc-testdev-ioport", 4);
+    memory_region_init_io(&dev->flush, &test_flush_ops, dev,
+                          "pc-testdev-flush-page", 4);
+    memory_region_init_io(&dev->irq, &test_irq_ops, dev,
+                          "pc-testdev-irq-line", 24);
+    memory_region_init_io(&dev->iomem, &test_iomem_ops, dev,
+                          "pc-testdev-iomem", IOMEM_LEN);
+
+    memory_region_add_subregion(io,  0xe0,       &dev->ioport);
+    memory_region_add_subregion(io,  0xe4,       &dev->flush);
+    memory_region_add_subregion(io,  0x2000,     &dev->irq);
+    memory_region_add_subregion(mem, 0xff000000, &dev->iomem);
+
+    return 0;
+}
+
+static void testdev_class_init(ObjectClass *klass, void *data)
+{
+    ISADeviceClass *k = ISA_DEVICE_CLASS(klass);
+
+    k->init = init_test_device;
+}
+
+static TypeInfo testdev_info = {
+    .name           = TYPE_TESTDEV,
+    .parent         = TYPE_ISA_DEVICE,
+    .instance_size  = sizeof(struct PCTestdev),
+    .class_init     = testdev_class_init,
+};
+
+static void testdev_register_types(void)
+{
+    type_register_static(&testdev_info);
+}
+
+type_init(testdev_register_types)
-- 
1.7.1

Re: [Qemu-devel] [PATCH 3/4] hw: Add test device for unittests execution

[Stefan Weil]

> From: Lucas Meneghel Rodrigues <lmr@redhat.com>
>
> Add a test device which supports the kvmctl ioports,
> so one can run the KVM unittest suite.
>
> Intended Usage:
>
> qemu-system-x86_64 -nographic \
>     -device pc-testdev \
>     -device isa-debug-exit,iobase=0xf4,iosize=0x04 \
>     -kernel /path/to/kvm/unittests/msr.flat
>
> Where msr.flat is one of the KVM unittests, present on a
> separate repo,
>
> git://git.kernel.org/pub/scm/virt/kvm/kvm-unit-tests.git
>
> [ kraxel: more memory api + qom fixes ]
>
> CC: Paolo Bonzini <pbonzini@redhat.com>
> Signed-off-by: Alexander Graf <agraf@suse.de>
> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
> Signed-off-by: Lucas Meneghel Rodrigues <lmr@redhat.com>
> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
> ---
>  hw/i386/Makefile.objs |    1 +
>  hw/pc-testdev.c       |  182
> +++++++++++++++++++++++++++++++++++++++++++++++++
>  2 files changed, 183 insertions(+), 0 deletions(-)
>  create mode 100644 hw/pc-testdev.c


This patch breaks QEMU git master for MinGW.

  CC    i386-softmmu/hw/i386/../pc-testdev.o
/qemu/hw/i386/../pc-testdev.c:38:22: warning: sys/mman.h: Datei oder
Verzeichnis nicht gefunden
/qemu/hw/i386/../pc-testdev.c: In function ‘test_flush_page’:
/qemu/hw/i386/../pc-testdev.c:103: warning: implicit declaration of
function ‘mprotect’
...

Should hw/pc-testdev.c be only compiled when CONFIG_KVM is defined?

Regards,

Stefan

Re: [Qemu-devel] [PATCH 3/4] hw: Add test device for unittests execution

[Alexander Graf]

Am 04.01.2013 um 22:44 schrieb "Stefan Weil" <sw@weilnetz.de>:

>> From: Lucas Meneghel Rodrigues <lmr@redhat.com>
>> 
>> Add a test device which supports the kvmctl ioports,
>> so one can run the KVM unittest suite.
>> 
>> Intended Usage:
>> 
>> qemu-system-x86_64 -nographic \
>>    -device pc-testdev \
>>    -device isa-debug-exit,iobase=0xf4,iosize=0x04 \
>>    -kernel /path/to/kvm/unittests/msr.flat
>> 
>> Where msr.flat is one of the KVM unittests, present on a
>> separate repo,
>> 
>> git://git.kernel.org/pub/scm/virt/kvm/kvm-unit-tests.git
>> 
>> [ kraxel: more memory api + qom fixes ]
>> 
>> CC: Paolo Bonzini <pbonzini@redhat.com>
>> Signed-off-by: Alexander Graf <agraf@suse.de>
>> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
>> Signed-off-by: Lucas Meneghel Rodrigues <lmr@redhat.com>
>> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
>> ---
>> hw/i386/Makefile.objs |    1 +
>> hw/pc-testdev.c       |  182
>> +++++++++++++++++++++++++++++++++++++++++++++++++
>> 2 files changed, 183 insertions(+), 0 deletions(-)
>> create mode 100644 hw/pc-testdev.c
> 
> 
> This patch breaks QEMU git master for MinGW.
> 
>  CC    i386-softmmu/hw/i386/../pc-testdev.o
> /qemu/hw/i386/../pc-testdev.c:38:22: warning: sys/mman.h: Datei oder
> Verzeichnis nicht gefunden
> /qemu/hw/i386/../pc-testdev.c: In function ‘test_flush_page’:
> /qemu/hw/i386/../pc-testdev.c:103: warning: implicit declaration of
> function ‘mprotect’
> ...
> 
> Should hw/pc-testdev.c be only compiled when CONFIG_KVM is defined?

It should work fine with tcg too. How about guarding the mprotect with an #ifdef __linux__?

Alex

> 
> Regards,
> 
> Stefan
> 
> 

Re: [Qemu-devel] [PATCH 3/4] hw: Add test device for unittests execution

[Andreas Färber]

Am 04.01.2013 09:16, schrieb Gerd Hoffmann:
> From: Lucas Meneghel Rodrigues <lmr@redhat.com>
> 
> Add a test device which supports the kvmctl ioports,
> so one can run the KVM unittest suite.
> 
> Intended Usage:
> 
> qemu-system-x86_64 -nographic \
>     -device pc-testdev \
>     -device isa-debug-exit,iobase=0xf4,iosize=0x04 \
>     -kernel /path/to/kvm/unittests/msr.flat
> 
> Where msr.flat is one of the KVM unittests, present on a
> separate repo,
> 
> git://git.kernel.org/pub/scm/virt/kvm/kvm-unit-tests.git
> 
> [ kraxel: more memory api + qom fixes ]
> 
> CC: Paolo Bonzini <pbonzini@redhat.com>
> Signed-off-by: Alexander Graf <agraf@suse.de>
> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
> Signed-off-by: Lucas Meneghel Rodrigues <lmr@redhat.com>
> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
[...]
> diff --git a/hw/pc-testdev.c b/hw/pc-testdev.c
> new file mode 100644
> index 0000000..620c86c
> --- /dev/null
> +++ b/hw/pc-testdev.c
[...]
> +typedef struct PCTestdev {
> +    ISADevice parent_obj;
> +
> +    MemoryRegion ioport;
> +    MemoryRegion flush;
> +    MemoryRegion irq;
> +    MemoryRegion iomem;
> +    uint32_t ioport_data;
> +    char iomem_buf[IOMEM_LEN];
> +} PCTestdev;
> +
> +#define TYPE_TESTDEV "pc-testdev"
> +#define TESTDEV(obj) \
> +     OBJECT_CHECK(struct PCTestdev, (obj), TYPE_TESTDEV)

You define a typedef above but ignore it here and everywhere below. I'm
surprised that Anthony didn't complain - struct is an implementation
detail of today's QOM classes in lack of C++/etc. language support, and
even qdev used the typedefs.

Lucas/Gerd, can one of you please clean this up in a follow-up?

I only noticed this because I had to touch some of these lines rebasing
my ISA QOM realize queue - maybe some Perl wizard can add a
checkpatch.pl rule to catch this?

Thanks,
Andreas

> +
> +static void test_irq_line(void *opaque, hwaddr addr, uint64_t data,
> +                          unsigned len)
> +{
> +    struct PCTestdev *dev = opaque;
> +    struct ISADevice *isa = ISA_DEVICE(dev);
> +
> +    qemu_set_irq(isa_get_irq(isa, addr), !!data);
> +}
> +
> +static const MemoryRegionOps test_irq_ops = {
> +    .write = test_irq_line,
> +    .valid.min_access_size = 1,
> +    .valid.max_access_size = 1,
> +    .endianness = DEVICE_LITTLE_ENDIAN,
> +};
> +
> +static void test_ioport_write(void *opaque, hwaddr addr, uint64_t data,
> +                              unsigned len)
> +{
> +    struct PCTestdev *dev = opaque;
> +    dev->ioport_data = data;
> +}
> +
> +static uint64_t test_ioport_read(void *opaque, hwaddr addr, unsigned len)
> +{
> +    struct PCTestdev *dev = opaque;
> +    return dev->ioport_data;
> +}
> +
> +static const MemoryRegionOps test_ioport_ops = {
> +    .read = test_ioport_read,
> +    .write = test_ioport_write,
> +    .endianness = DEVICE_LITTLE_ENDIAN,
> +};
> +
> +static void test_flush_page(void *opaque, hwaddr addr, uint64_t data,
> +                            unsigned len)
> +{
> +    hwaddr page = 4096;
> +    void *a = cpu_physical_memory_map(data & ~0xffful, &page, 0);
> +
> +    /* We might not be able to get the full page, only mprotect what we actually
> +       have mapped */
> +    mprotect(a, page, PROT_NONE);
> +    mprotect(a, page, PROT_READ|PROT_WRITE);
> +    cpu_physical_memory_unmap(a, page, 0, 0);
> +}
> +
> +static const MemoryRegionOps test_flush_ops = {
> +    .write = test_flush_page,
> +    .valid.min_access_size = 4,
> +    .valid.max_access_size = 4,
> +    .endianness = DEVICE_LITTLE_ENDIAN,
> +};
> +
> +static uint64_t test_iomem_read(void *opaque, hwaddr addr, unsigned len)
> +{
> +    struct PCTestdev *dev = opaque;
> +    uint64_t ret = 0;
> +    memcpy(&ret, &dev->iomem_buf[addr], len);
> +    ret = le64_to_cpu(ret);
> +
> +    return ret;
> +}
> +
> +static void test_iomem_write(void *opaque, hwaddr addr, uint64_t val,
> +                             unsigned len)
> +{
> +    struct PCTestdev *dev = opaque;
> +    val = cpu_to_le64(val);
> +    memcpy(&dev->iomem_buf[addr], &val, len);
> +    dev->iomem_buf[addr] = val;
> +}
> +
> +static const MemoryRegionOps test_iomem_ops = {
> +    .read = test_iomem_read,
> +    .write = test_iomem_write,
> +    .endianness = DEVICE_LITTLE_ENDIAN,
> +};
> +
> +static int init_test_device(ISADevice *isa)
> +{
> +    struct PCTestdev *dev = TESTDEV(isa);
> +    MemoryRegion *mem = isa_address_space(isa);
> +    MemoryRegion *io = isa_address_space_io(isa);
> +
> +    memory_region_init_io(&dev->ioport, &test_ioport_ops, dev,
> +                          "pc-testdev-ioport", 4);
> +    memory_region_init_io(&dev->flush, &test_flush_ops, dev,
> +                          "pc-testdev-flush-page", 4);
> +    memory_region_init_io(&dev->irq, &test_irq_ops, dev,
> +                          "pc-testdev-irq-line", 24);
> +    memory_region_init_io(&dev->iomem, &test_iomem_ops, dev,
> +                          "pc-testdev-iomem", IOMEM_LEN);
> +
> +    memory_region_add_subregion(io,  0xe0,       &dev->ioport);
> +    memory_region_add_subregion(io,  0xe4,       &dev->flush);
> +    memory_region_add_subregion(io,  0x2000,     &dev->irq);
> +    memory_region_add_subregion(mem, 0xff000000, &dev->iomem);
> +
> +    return 0;
> +}
> +
> +static void testdev_class_init(ObjectClass *klass, void *data)
> +{
> +    ISADeviceClass *k = ISA_DEVICE_CLASS(klass);
> +
> +    k->init = init_test_device;
> +}
> +
> +static TypeInfo testdev_info = {

static const

> +    .name           = TYPE_TESTDEV,
> +    .parent         = TYPE_ISA_DEVICE,
> +    .instance_size  = sizeof(struct PCTestdev),
> +    .class_init     = testdev_class_init,
> +};
> +
> +static void testdev_register_types(void)
> +{
> +    type_register_static(&testdev_info);
> +}
> +
> +type_init(testdev_register_types)


-- 
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg

[Qemu-devel] [PATCH 4/4] pc: remove bochs bios debug ports

[Gerd Hoffmann]

Prehistoric leftover, zap it.  We poweroff via acpi these days.

And having a port (0x501,0x502) where any random guest write will make
qemu exit -- with no way to turn it off -- is a bad joke anyway.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
---
 hw/pc.c |   41 -----------------------------------------
 1 files changed, 0 insertions(+), 41 deletions(-)

diff --git a/hw/pc.c b/hw/pc.c
index 71902e2..4879e4f 100644
--- a/hw/pc.c
+++ b/hw/pc.c
@@ -524,34 +524,6 @@ static void handle_a20_line_change(void *opaque, int irq, int level)
     cpu_x86_set_a20(cpu, level);
 }
 
-/***********************************************************/
-/* Bochs BIOS debug ports */
-
-static void bochs_bios_write(void *opaque, uint32_t addr, uint32_t val)
-{
-    static const char shutdown_str[8] = "Shutdown";
-    static int shutdown_index = 0;
-
-    switch(addr) {
-    case 0x8900:
-        /* same as Bochs power off */
-        if (val == shutdown_str[shutdown_index]) {
-            shutdown_index++;
-            if (shutdown_index == 8) {
-                shutdown_index = 0;
-                qemu_system_shutdown_request();
-            }
-        } else {
-            shutdown_index = 0;
-        }
-        break;
-
-    case 0x501:
-    case 0x502:
-        exit((val << 1) | 1);
-    }
-}
-
 int e820_add_entry(uint64_t address, uint64_t length, uint32_t type)
 {
     int index = le32_to_cpu(e820_table.count);
@@ -569,14 +541,6 @@ int e820_add_entry(uint64_t address, uint64_t length, uint32_t type)
     return index;
 }
 
-static const MemoryRegionPortio bochs_bios_portio_list[] = {
-    { 0x500, 1, 1, .write = bochs_bios_write, }, /* 0x500 */
-    { 0x501, 1, 1, .write = bochs_bios_write, }, /* 0x501 */
-    { 0x501, 2, 2, .write = bochs_bios_write, }, /* 0x501 */
-    { 0x8900, 1, 1, .write = bochs_bios_write, }, /* 0x8900 */
-    PORTIO_END_OF_LIST(),
-};
-
 static void *bochs_bios_init(void)
 {
     void *fw_cfg;
@@ -584,11 +548,6 @@ static void *bochs_bios_init(void)
     size_t smbios_len;
     uint64_t *numa_fw_cfg;
     int i, j;
-    PortioList *bochs_bios_port_list = g_new(PortioList, 1);
-
-    portio_list_init(bochs_bios_port_list, bochs_bios_portio_list,
-                     NULL, "bochs-bios");
-    portio_list_add(bochs_bios_port_list, get_system_io(), 0x0);
 
     fw_cfg = fw_cfg_init(BIOS_CFG_IOPORT, BIOS_CFG_IOPORT + 1, 0, 0);
 
-- 
1.7.1

Re: [Qemu-devel] [PULL 0/4] test and debug devices

[Andreas Färber]

Am 04.01.2013 09:16, schrieb Gerd Hoffmann:
>   Hi,
> 
> This patch series improves the debug devices we have.  isa-debugcon is
> switched over to the memory api.  isa-debug-exit is added and replaces
> the hard-coded bochs debug port.  pc-testdev (used by kvm-unit-tests) is
> added.

Missing some CCs, and I was actually expecting another PATCH series
first... as brought up before it is refreshing an uncommitted series by
Hervé without any mention thereof in the commit message and I thought
there was an unanswered question to Anthony on both Lucas' and Hervé's
series?

Andreas

> 
> please pull,
>   Gerd
> 
> The following changes since commit 25bbf61e4bacd1e4fa4115ffcf151051b9d6608e:
> 
>   pty: unbreak libvirt (2013-01-03 12:53:41 -0600)
> 
> are available in the git repository at:
>   git://git.kraxel.org/qemu testdev.1
> 
> Gerd Hoffmann (3):
>       switch debugcon to memory api
>       add isa-debug-exit device.
>       pc: remove bochs bios debug ports
> 
> Lucas Meneghel Rodrigues (1):
>       hw: Add test device for unittests execution
> 
>  hw/debugcon.c         |   31 +++++++--
>  hw/debugexit.c        |   75 ++++++++++++++++++++
>  hw/i386/Makefile.objs |    3 +-
>  hw/pc-testdev.c       |  182 +++++++++++++++++++++++++++++++++++++++++++++++++
>  hw/pc.c               |   41 -----------
>  5 files changed, 283 insertions(+), 49 deletions(-)
>  create mode 100644 hw/debugexit.c
>  create mode 100644 hw/pc-testdev.c

-- 
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg

Re: [Qemu-devel] [PULL 0/4] test and debug devices

[Gerd Hoffmann]

  Hi,

>> This patch series improves the debug devices we have.  isa-debugcon is
>> switched over to the memory api.  isa-debug-exit is added and replaces
>> the hard-coded bochs debug port.  pc-testdev (used by kvm-unit-tests) is
>> added.
> 
> Missing some CCs, and I was actually expecting another PATCH series
> first... as brought up before it is refreshing an uncommitted series by
> Hervé without any mention thereof in the commit message

Hmm?  The isa-debug-exit patch is written by myself.  After hacking it
up I've noticed Herve wrote a simliar one.  But I didn't grab Herve's
patch.  I've picked up + updated Lucas' testdev patch, and that is
clearly visible in the commit message.

> and I thought
> there was an unanswered question to Anthony on both Lucas' and Hervé's
> series?

The only outstanding question I'm aware of is whenever anthony does a
outb or outw on the bochs debug port in his test suite.  My
isa-debug-exit device handles both cases, so we should be fine here, and
I'm sure Anthony will actually try before committing the series ;)

What else is unanswered?

cheers,
  Gerd

Re: [Qemu-devel] [PULL 0/4] test and debug devices

[Andreas Färber]

Hi Gerd,

Am 04.01.2013 13:57, schrieb Gerd Hoffmann:
>>> This patch series improves the debug devices we have.  isa-debugcon is
>>> switched over to the memory api.  isa-debug-exit is added and replaces
>>> the hard-coded bochs debug port.  pc-testdev (used by kvm-unit-tests) is
>>> added.
>>
>> Missing some CCs, and I was actually expecting another PATCH series
>> first... as brought up before it is refreshing an uncommitted series by
>> Hervé without any mention thereof in the commit message
> 
> Hmm?  The isa-debug-exit patch is written by myself.  After hacking it
> up I've noticed Herve wrote a simliar one.  But I didn't grab Herve's
> patch.  I've picked up + updated Lucas' testdev patch, and that is
> clearly visible in the commit message.
> 
>> and I thought
>> there was an unanswered question to Anthony on both Lucas' and Hervé's
>> series?
> 
> The only outstanding question I'm aware of is whenever anthony does a
> outb or outw on the bochs debug port in his test suite.  My
> isa-debug-exit device handles both cases, so we should be fine here, and
> I'm sure Anthony will actually try before committing the series ;)
> 
> What else is unanswered?

I believe the original unanswered question was whether it is okay for
Anthony to add -device foo for his testing now that the default port was
changed to what he was relying on or whether he wants it to
automatically get instantiated for machine backwards compatibility
unless --no-defaults (or whatever it's called) is used. Your series,
just like Hervé's that got stopped by Anthony, simple moves things into
a standalone device in 2/4 and I don't see any pc-1.3 handling in 4/4.

Cheers,
Andreas

-- 
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg

Re: [Qemu-devel] [PULL 0/4] test and debug devices

[Anthony Liguori]

Gerd Hoffmann <kraxel@redhat.com> writes:

>   Hi,
>
>>> This patch series improves the debug devices we have.  isa-debugcon is
>>> switched over to the memory api.  isa-debug-exit is added and replaces
>>> the hard-coded bochs debug port.  pc-testdev (used by kvm-unit-tests) is
>>> added.
>> 
>> Missing some CCs, and I was actually expecting another PATCH series
>> first... as brought up before it is refreshing an uncommitted series by
>> Hervé without any mention thereof in the commit message
>
> Hmm?  The isa-debug-exit patch is written by myself.  After hacking it
> up I've noticed Herve wrote a simliar one.  But I didn't grab Herve's
> patch.  I've picked up + updated Lucas' testdev patch, and that is
> clearly visible in the commit message.
>
>> and I thought
>> there was an unanswered question to Anthony on both Lucas' and Hervé's
>> series?
>
> The only outstanding question I'm aware of is whenever anthony does a
> outb or outw on the bochs debug port in his test suite.

outb.

Regards,

Anthony Liguori

> My
> isa-debug-exit device handles both cases, so we should be fine here, and
> I'm sure Anthony will actually try before committing the series ;)
>
> What else is unanswered?
>
> cheers,
>   Gerd

Re: [Qemu-devel] [PULL 0/4] test and debug devices

[Anthony Liguori]

Pulled, thanks.

N.B.  This note may be extraneous because the pull request was sent by a
version of git older than 1.7.9 making the pull request ambigious.  Please
consider upgrading to a newer version of git.

Regards,

Anthony Liguori