* [PATCH v2 01/12] MAINTAINERS: Add docs/system/arm/xenpvh.rst
2024-08-20 14:29 [PATCH v2 00/12] xen: pvh: Partial QOM:fication with new x86 PVH machine Edgar E. Iglesias
@ 2024-08-20 14:29 ` Edgar E. Iglesias
2024-08-20 14:29 ` [PATCH v2 02/12] hw/arm: xenpvh: Update file header to use SPDX Edgar E. Iglesias
` (10 subsequent siblings)
11 siblings, 0 replies; 19+ messages in thread
From: Edgar E. Iglesias @ 2024-08-20 14:29 UTC (permalink / raw)
To: qemu-devel
Cc: sstabellini, anthony, paul, peter.maydell, alex.bennee,
xenia.ragiadakou, jason.andryuk, edgar.iglesias, xen-devel
From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Acked-by: Stefano Stabellini <sstabellini@kernel.org>
---
MAINTAINERS | 1 +
1 file changed, 1 insertion(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index 3584d6a6c6..c2fb0c2f42 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -559,6 +559,7 @@ F: include/hw/xen/
F: include/sysemu/xen.h
F: include/sysemu/xen-mapcache.h
F: stubs/xen-hw-stub.c
+F: docs/system/arm/xenpvh.rst
Guest CPU Cores (NVMM)
----------------------
--
2.43.0
^ permalink raw reply related [flat|nested] 19+ messages in thread* [PATCH v2 02/12] hw/arm: xenpvh: Update file header to use SPDX
2024-08-20 14:29 [PATCH v2 00/12] xen: pvh: Partial QOM:fication with new x86 PVH machine Edgar E. Iglesias
2024-08-20 14:29 ` [PATCH v2 01/12] MAINTAINERS: Add docs/system/arm/xenpvh.rst Edgar E. Iglesias
@ 2024-08-20 14:29 ` Edgar E. Iglesias
2024-08-20 14:29 ` [PATCH v2 03/12] hw/arm: xenpvh: Tweak machine description Edgar E. Iglesias
` (9 subsequent siblings)
11 siblings, 0 replies; 19+ messages in thread
From: Edgar E. Iglesias @ 2024-08-20 14:29 UTC (permalink / raw)
To: qemu-devel
Cc: sstabellini, anthony, paul, peter.maydell, alex.bennee,
xenia.ragiadakou, jason.andryuk, edgar.iglesias, xen-devel,
qemu-arm
From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
Update file header to use SPDX and remove stray empty
comment line.
No functional changes.
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Acked-by: Stefano Stabellini <sstabellini@kernel.org>
---
hw/arm/xen_arm.c | 19 +------------------
1 file changed, 1 insertion(+), 18 deletions(-)
diff --git a/hw/arm/xen_arm.c b/hw/arm/xen_arm.c
index 6fad829ede..766a194fa1 100644
--- a/hw/arm/xen_arm.c
+++ b/hw/arm/xen_arm.c
@@ -1,24 +1,7 @@
/*
* QEMU ARM Xen PVH Machine
*
- *
- * 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.
+ * SPDX-License-Identifier: MIT
*/
#include "qemu/osdep.h"
--
2.43.0
^ permalink raw reply related [flat|nested] 19+ messages in thread* [PATCH v2 03/12] hw/arm: xenpvh: Tweak machine description
2024-08-20 14:29 [PATCH v2 00/12] xen: pvh: Partial QOM:fication with new x86 PVH machine Edgar E. Iglesias
2024-08-20 14:29 ` [PATCH v2 01/12] MAINTAINERS: Add docs/system/arm/xenpvh.rst Edgar E. Iglesias
2024-08-20 14:29 ` [PATCH v2 02/12] hw/arm: xenpvh: Update file header to use SPDX Edgar E. Iglesias
@ 2024-08-20 14:29 ` Edgar E. Iglesias
2024-08-20 14:29 ` [PATCH v2 04/12] hw/arm: xenpvh: Add support for SMP guests Edgar E. Iglesias
` (8 subsequent siblings)
11 siblings, 0 replies; 19+ messages in thread
From: Edgar E. Iglesias @ 2024-08-20 14:29 UTC (permalink / raw)
To: qemu-devel
Cc: sstabellini, anthony, paul, peter.maydell, alex.bennee,
xenia.ragiadakou, jason.andryuk, edgar.iglesias, xen-devel,
qemu-arm
From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
Tweak machine description to better express that this is
a Xen PVH machine for ARM.
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
---
hw/arm/xen_arm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/arm/xen_arm.c b/hw/arm/xen_arm.c
index 766a194fa1..5f75cc3779 100644
--- a/hw/arm/xen_arm.c
+++ b/hw/arm/xen_arm.c
@@ -216,7 +216,7 @@ static void xen_arm_machine_class_init(ObjectClass *oc, void *data)
{
MachineClass *mc = MACHINE_CLASS(oc);
- mc->desc = "Xen Para-virtualized PC";
+ mc->desc = "Xen PVH ARM machine";
mc->init = xen_arm_init;
mc->max_cpus = 1;
mc->default_machine_opts = "accel=xen";
--
2.43.0
^ permalink raw reply related [flat|nested] 19+ messages in thread* [PATCH v2 04/12] hw/arm: xenpvh: Add support for SMP guests
2024-08-20 14:29 [PATCH v2 00/12] xen: pvh: Partial QOM:fication with new x86 PVH machine Edgar E. Iglesias
` (2 preceding siblings ...)
2024-08-20 14:29 ` [PATCH v2 03/12] hw/arm: xenpvh: Tweak machine description Edgar E. Iglesias
@ 2024-08-20 14:29 ` Edgar E. Iglesias
2024-08-21 0:55 ` Stefano Stabellini
2024-08-20 14:29 ` [PATCH v2 05/12] hw/arm: xenpvh: Remove double-negation in warning Edgar E. Iglesias
` (7 subsequent siblings)
11 siblings, 1 reply; 19+ messages in thread
From: Edgar E. Iglesias @ 2024-08-20 14:29 UTC (permalink / raw)
To: qemu-devel
Cc: sstabellini, anthony, paul, peter.maydell, alex.bennee,
xenia.ragiadakou, jason.andryuk, edgar.iglesias, xen-devel,
qemu-arm
From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
Add SMP support for Xen PVH ARM guests.
Create ms->smp.max_cpus ioreq servers to handle hotplug.
Note that ms->smp.max_cpus will be passed to us by the
user (Xen tools) set to the guests maxvcpus.
The value in mc->max_cpus is an absolute maximum for the
-smp option and won't be used to setup ioreq servers unless
the user explicitly specifies it with -smp.
If the user doesn't pass -smp on the command-line, smp.cpus
and smp.max_cpus will default to 1.
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
---
hw/arm/xen_arm.c | 23 +++++++++++++++++++++--
1 file changed, 21 insertions(+), 2 deletions(-)
diff --git a/hw/arm/xen_arm.c b/hw/arm/xen_arm.c
index 5f75cc3779..fda65d0d8d 100644
--- a/hw/arm/xen_arm.c
+++ b/hw/arm/xen_arm.c
@@ -173,7 +173,7 @@ static void xen_arm_init(MachineState *machine)
xen_init_ram(machine);
- xen_register_ioreq(xam->state, machine->smp.cpus, &xen_memory_listener);
+ xen_register_ioreq(xam->state, machine->smp.max_cpus, &xen_memory_listener);
xen_create_virtio_mmio_devices(xam);
@@ -218,7 +218,26 @@ static void xen_arm_machine_class_init(ObjectClass *oc, void *data)
MachineClass *mc = MACHINE_CLASS(oc);
mc->desc = "Xen PVH ARM machine";
mc->init = xen_arm_init;
- mc->max_cpus = 1;
+
+ /*
+ * mc->max_cpus holds the MAX value allowed in the -smp command-line opts.
+ *
+ * 1. If users don't pass any -smp option:
+ * ms->smp.cpus will default to 1.
+ * ms->smp.max_cpus will default to 1.
+ *
+ * 2. If users pass -smp X:
+ * ms->smp.cpus will be set to X.
+ * ms->smp.max_cpus will also be set to X.
+ *
+ * 3. If users pass -smp X,maxcpus=Y:
+ * ms->smp.cpus will be set to X.
+ * ms->smp.max_cpus will be set to Y.
+ *
+ * In scenarios 2 and 3, if X or Y are set to something larger than
+ * mc->max_cpus, QEMU will bail out with an error message.
+ */
+ mc->max_cpus = GUEST_MAX_VCPUS;
mc->default_machine_opts = "accel=xen";
/* Set explicitly here to make sure that real ram_size is passed */
mc->default_ram_size = 0;
--
2.43.0
^ permalink raw reply related [flat|nested] 19+ messages in thread* Re: [PATCH v2 04/12] hw/arm: xenpvh: Add support for SMP guests
2024-08-20 14:29 ` [PATCH v2 04/12] hw/arm: xenpvh: Add support for SMP guests Edgar E. Iglesias
@ 2024-08-21 0:55 ` Stefano Stabellini
0 siblings, 0 replies; 19+ messages in thread
From: Stefano Stabellini @ 2024-08-21 0:55 UTC (permalink / raw)
To: Edgar E. Iglesias
Cc: qemu-devel, sstabellini, anthony, paul, peter.maydell,
alex.bennee, xenia.ragiadakou, jason.andryuk, edgar.iglesias,
xen-devel, qemu-arm
On Tue, 20 Aug 2024, Edgar E. Iglesias wrote:
> From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
>
> Add SMP support for Xen PVH ARM guests.
> Create ms->smp.max_cpus ioreq servers to handle hotplug.
>
> Note that ms->smp.max_cpus will be passed to us by the
> user (Xen tools) set to the guests maxvcpus.
>
> The value in mc->max_cpus is an absolute maximum for the
> -smp option and won't be used to setup ioreq servers unless
> the user explicitly specifies it with -smp.
>
> If the user doesn't pass -smp on the command-line, smp.cpus
> and smp.max_cpus will default to 1.
>
> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
> ---
> hw/arm/xen_arm.c | 23 +++++++++++++++++++++--
> 1 file changed, 21 insertions(+), 2 deletions(-)
>
> diff --git a/hw/arm/xen_arm.c b/hw/arm/xen_arm.c
> index 5f75cc3779..fda65d0d8d 100644
> --- a/hw/arm/xen_arm.c
> +++ b/hw/arm/xen_arm.c
> @@ -173,7 +173,7 @@ static void xen_arm_init(MachineState *machine)
>
> xen_init_ram(machine);
>
> - xen_register_ioreq(xam->state, machine->smp.cpus, &xen_memory_listener);
> + xen_register_ioreq(xam->state, machine->smp.max_cpus, &xen_memory_listener);
>
> xen_create_virtio_mmio_devices(xam);
>
> @@ -218,7 +218,26 @@ static void xen_arm_machine_class_init(ObjectClass *oc, void *data)
> MachineClass *mc = MACHINE_CLASS(oc);
> mc->desc = "Xen PVH ARM machine";
> mc->init = xen_arm_init;
> - mc->max_cpus = 1;
> +
> + /*
> + * mc->max_cpus holds the MAX value allowed in the -smp command-line opts.
> + *
> + * 1. If users don't pass any -smp option:
> + * ms->smp.cpus will default to 1.
> + * ms->smp.max_cpus will default to 1.
> + *
> + * 2. If users pass -smp X:
> + * ms->smp.cpus will be set to X.
> + * ms->smp.max_cpus will also be set to X.
> + *
> + * 3. If users pass -smp X,maxcpus=Y:
> + * ms->smp.cpus will be set to X.
> + * ms->smp.max_cpus will be set to Y.
> + *
> + * In scenarios 2 and 3, if X or Y are set to something larger than
> + * mc->max_cpus, QEMU will bail out with an error message.
> + */
> + mc->max_cpus = GUEST_MAX_VCPUS;
> mc->default_machine_opts = "accel=xen";
> /* Set explicitly here to make sure that real ram_size is passed */
> mc->default_ram_size = 0;
> --
> 2.43.0
>
^ permalink raw reply [flat|nested] 19+ messages in thread
* [PATCH v2 05/12] hw/arm: xenpvh: Remove double-negation in warning
2024-08-20 14:29 [PATCH v2 00/12] xen: pvh: Partial QOM:fication with new x86 PVH machine Edgar E. Iglesias
` (3 preceding siblings ...)
2024-08-20 14:29 ` [PATCH v2 04/12] hw/arm: xenpvh: Add support for SMP guests Edgar E. Iglesias
@ 2024-08-20 14:29 ` Edgar E. Iglesias
2024-08-21 0:56 ` Stefano Stabellini
2024-08-20 14:29 ` [PATCH v2 06/12] hw/arm: xenpvh: Move stubbed functions to xen-stubs.c Edgar E. Iglesias
` (6 subsequent siblings)
11 siblings, 1 reply; 19+ messages in thread
From: Edgar E. Iglesias @ 2024-08-20 14:29 UTC (permalink / raw)
To: qemu-devel
Cc: sstabellini, anthony, paul, peter.maydell, alex.bennee,
xenia.ragiadakou, jason.andryuk, edgar.iglesias, xen-devel,
qemu-arm
From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
---
hw/arm/xen_arm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/arm/xen_arm.c b/hw/arm/xen_arm.c
index fda65d0d8d..16b3f00992 100644
--- a/hw/arm/xen_arm.c
+++ b/hw/arm/xen_arm.c
@@ -165,7 +165,7 @@ static void xen_arm_init(MachineState *machine)
xam->state = g_new0(XenIOState, 1);
if (machine->ram_size == 0) {
- warn_report("%s non-zero ram size not specified. QEMU machine started"
+ warn_report("%s: ram size not specified. QEMU machine started"
" without IOREQ (no emulated devices including virtio)",
MACHINE_CLASS(object_get_class(OBJECT(machine)))->desc);
return;
--
2.43.0
^ permalink raw reply related [flat|nested] 19+ messages in thread* Re: [PATCH v2 05/12] hw/arm: xenpvh: Remove double-negation in warning
2024-08-20 14:29 ` [PATCH v2 05/12] hw/arm: xenpvh: Remove double-negation in warning Edgar E. Iglesias
@ 2024-08-21 0:56 ` Stefano Stabellini
0 siblings, 0 replies; 19+ messages in thread
From: Stefano Stabellini @ 2024-08-21 0:56 UTC (permalink / raw)
To: Edgar E. Iglesias
Cc: qemu-devel, sstabellini, anthony, paul, peter.maydell,
alex.bennee, xenia.ragiadakou, jason.andryuk, edgar.iglesias,
xen-devel, qemu-arm
On Tue, 20 Aug 2024, Edgar E. Iglesias wrote:
> From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
>
> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
> ---
> hw/arm/xen_arm.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/hw/arm/xen_arm.c b/hw/arm/xen_arm.c
> index fda65d0d8d..16b3f00992 100644
> --- a/hw/arm/xen_arm.c
> +++ b/hw/arm/xen_arm.c
> @@ -165,7 +165,7 @@ static void xen_arm_init(MachineState *machine)
> xam->state = g_new0(XenIOState, 1);
>
> if (machine->ram_size == 0) {
> - warn_report("%s non-zero ram size not specified. QEMU machine started"
> + warn_report("%s: ram size not specified. QEMU machine started"
> " without IOREQ (no emulated devices including virtio)",
> MACHINE_CLASS(object_get_class(OBJECT(machine)))->desc);
> return;
> --
> 2.43.0
>
^ permalink raw reply [flat|nested] 19+ messages in thread
* [PATCH v2 06/12] hw/arm: xenpvh: Move stubbed functions to xen-stubs.c
2024-08-20 14:29 [PATCH v2 00/12] xen: pvh: Partial QOM:fication with new x86 PVH machine Edgar E. Iglesias
` (4 preceding siblings ...)
2024-08-20 14:29 ` [PATCH v2 05/12] hw/arm: xenpvh: Remove double-negation in warning Edgar E. Iglesias
@ 2024-08-20 14:29 ` Edgar E. Iglesias
2024-08-21 1:06 ` Stefano Stabellini
2024-08-20 14:29 ` [PATCH v2 07/12] hw/arm: xenpvh: Break out a common PVH machine Edgar E. Iglesias
` (5 subsequent siblings)
11 siblings, 1 reply; 19+ messages in thread
From: Edgar E. Iglesias @ 2024-08-20 14:29 UTC (permalink / raw)
To: qemu-devel
Cc: sstabellini, anthony, paul, peter.maydell, alex.bennee,
xenia.ragiadakou, jason.andryuk, edgar.iglesias, xen-devel,
qemu-arm
From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
---
hw/arm/meson.build | 5 ++++-
hw/arm/xen-stubs.c | 32 ++++++++++++++++++++++++++++++++
hw/arm/xen_arm.c | 20 --------------------
3 files changed, 36 insertions(+), 21 deletions(-)
create mode 100644 hw/arm/xen-stubs.c
diff --git a/hw/arm/meson.build b/hw/arm/meson.build
index 0c07ab522f..074612b40c 100644
--- a/hw/arm/meson.build
+++ b/hw/arm/meson.build
@@ -59,7 +59,10 @@ arm_ss.add(when: 'CONFIG_FSL_IMX7', if_true: files('fsl-imx7.c', 'mcimx7d-sabre.
arm_ss.add(when: 'CONFIG_ARM_SMMUV3', if_true: files('smmuv3.c'))
arm_ss.add(when: 'CONFIG_FSL_IMX6UL', if_true: files('fsl-imx6ul.c', 'mcimx6ul-evk.c'))
arm_ss.add(when: 'CONFIG_NRF51_SOC', if_true: files('nrf51_soc.c'))
-arm_ss.add(when: 'CONFIG_XEN', if_true: files('xen_arm.c'))
+arm_ss.add(when: 'CONFIG_XEN', if_true: files(
+ 'xen-stubs.c',
+ 'xen_arm.c',
+))
system_ss.add(when: 'CONFIG_ARM_SMMUV3', if_true: files('smmu-common.c'))
system_ss.add(when: 'CONFIG_CHEETAH', if_true: files('palm.c'))
diff --git a/hw/arm/xen-stubs.c b/hw/arm/xen-stubs.c
new file mode 100644
index 0000000000..4ac6a56a96
--- /dev/null
+++ b/hw/arm/xen-stubs.c
@@ -0,0 +1,32 @@
+/*
+ * Stubs for unimplemented Xen functions for ARM.
+ *
+ * SPDX-License-Identifier: MIT
+ */
+
+#include "qemu/osdep.h"
+#include "qemu/error-report.h"
+#include "qapi/qapi-commands-migration.h"
+#include "hw/boards.h"
+#include "sysemu/sysemu.h"
+#include "hw/xen/xen-hvm-common.h"
+#include "hw/xen/arch_hvm.h"
+
+void arch_handle_ioreq(XenIOState *state, ioreq_t *req)
+{
+ hw_error("Invalid ioreq type 0x%x\n", req->type);
+ return;
+}
+
+void arch_xen_set_memory(XenIOState *state, MemoryRegionSection *section,
+ bool add)
+{
+}
+
+void xen_hvm_modified_memory(ram_addr_t start, ram_addr_t length)
+{
+}
+
+void qmp_xen_set_global_dirty_log(bool enable, Error **errp)
+{
+}
diff --git a/hw/arm/xen_arm.c b/hw/arm/xen_arm.c
index 16b3f00992..f0868e7be5 100644
--- a/hw/arm/xen_arm.c
+++ b/hw/arm/xen_arm.c
@@ -115,26 +115,6 @@ static void xen_init_ram(MachineState *machine)
memory_region_add_subregion(sysmem, XEN_GRANT_ADDR_OFF, &xen_grants);
}
-void arch_handle_ioreq(XenIOState *state, ioreq_t *req)
-{
- hw_error("Invalid ioreq type 0x%x\n", req->type);
-
- return;
-}
-
-void arch_xen_set_memory(XenIOState *state, MemoryRegionSection *section,
- bool add)
-{
-}
-
-void xen_hvm_modified_memory(ram_addr_t start, ram_addr_t length)
-{
-}
-
-void qmp_xen_set_global_dirty_log(bool enable, Error **errp)
-{
-}
-
#ifdef CONFIG_TPM
static void xen_enable_tpm(XenArmState *xam)
{
--
2.43.0
^ permalink raw reply related [flat|nested] 19+ messages in thread* Re: [PATCH v2 06/12] hw/arm: xenpvh: Move stubbed functions to xen-stubs.c
2024-08-20 14:29 ` [PATCH v2 06/12] hw/arm: xenpvh: Move stubbed functions to xen-stubs.c Edgar E. Iglesias
@ 2024-08-21 1:06 ` Stefano Stabellini
0 siblings, 0 replies; 19+ messages in thread
From: Stefano Stabellini @ 2024-08-21 1:06 UTC (permalink / raw)
To: Edgar E. Iglesias
Cc: qemu-devel, sstabellini, anthony, paul, peter.maydell,
alex.bennee, xenia.ragiadakou, jason.andryuk, edgar.iglesias,
xen-devel, qemu-arm
On Tue, 20 Aug 2024, Edgar E. Iglesias wrote:
> From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
>
> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
> ---
> hw/arm/meson.build | 5 ++++-
> hw/arm/xen-stubs.c | 32 ++++++++++++++++++++++++++++++++
> hw/arm/xen_arm.c | 20 --------------------
> 3 files changed, 36 insertions(+), 21 deletions(-)
> create mode 100644 hw/arm/xen-stubs.c
>
> diff --git a/hw/arm/meson.build b/hw/arm/meson.build
> index 0c07ab522f..074612b40c 100644
> --- a/hw/arm/meson.build
> +++ b/hw/arm/meson.build
> @@ -59,7 +59,10 @@ arm_ss.add(when: 'CONFIG_FSL_IMX7', if_true: files('fsl-imx7.c', 'mcimx7d-sabre.
> arm_ss.add(when: 'CONFIG_ARM_SMMUV3', if_true: files('smmuv3.c'))
> arm_ss.add(when: 'CONFIG_FSL_IMX6UL', if_true: files('fsl-imx6ul.c', 'mcimx6ul-evk.c'))
> arm_ss.add(when: 'CONFIG_NRF51_SOC', if_true: files('nrf51_soc.c'))
> -arm_ss.add(when: 'CONFIG_XEN', if_true: files('xen_arm.c'))
> +arm_ss.add(when: 'CONFIG_XEN', if_true: files(
> + 'xen-stubs.c',
> + 'xen_arm.c',
> +))
>
> system_ss.add(when: 'CONFIG_ARM_SMMUV3', if_true: files('smmu-common.c'))
> system_ss.add(when: 'CONFIG_CHEETAH', if_true: files('palm.c'))
> diff --git a/hw/arm/xen-stubs.c b/hw/arm/xen-stubs.c
> new file mode 100644
> index 0000000000..4ac6a56a96
> --- /dev/null
> +++ b/hw/arm/xen-stubs.c
> @@ -0,0 +1,32 @@
> +/*
> + * Stubs for unimplemented Xen functions for ARM.
> + *
> + * SPDX-License-Identifier: MIT
> + */
> +
> +#include "qemu/osdep.h"
> +#include "qemu/error-report.h"
> +#include "qapi/qapi-commands-migration.h"
> +#include "hw/boards.h"
> +#include "sysemu/sysemu.h"
> +#include "hw/xen/xen-hvm-common.h"
> +#include "hw/xen/arch_hvm.h"
> +
> +void arch_handle_ioreq(XenIOState *state, ioreq_t *req)
> +{
> + hw_error("Invalid ioreq type 0x%x\n", req->type);
> + return;
> +}
> +
> +void arch_xen_set_memory(XenIOState *state, MemoryRegionSection *section,
> + bool add)
> +{
> +}
> +
> +void xen_hvm_modified_memory(ram_addr_t start, ram_addr_t length)
> +{
> +}
> +
> +void qmp_xen_set_global_dirty_log(bool enable, Error **errp)
> +{
> +}
> diff --git a/hw/arm/xen_arm.c b/hw/arm/xen_arm.c
> index 16b3f00992..f0868e7be5 100644
> --- a/hw/arm/xen_arm.c
> +++ b/hw/arm/xen_arm.c
> @@ -115,26 +115,6 @@ static void xen_init_ram(MachineState *machine)
> memory_region_add_subregion(sysmem, XEN_GRANT_ADDR_OFF, &xen_grants);
> }
>
> -void arch_handle_ioreq(XenIOState *state, ioreq_t *req)
> -{
> - hw_error("Invalid ioreq type 0x%x\n", req->type);
> -
> - return;
> -}
> -
> -void arch_xen_set_memory(XenIOState *state, MemoryRegionSection *section,
> - bool add)
> -{
> -}
> -
> -void xen_hvm_modified_memory(ram_addr_t start, ram_addr_t length)
> -{
> -}
> -
> -void qmp_xen_set_global_dirty_log(bool enable, Error **errp)
> -{
> -}
> -
> #ifdef CONFIG_TPM
> static void xen_enable_tpm(XenArmState *xam)
> {
> --
> 2.43.0
>
^ permalink raw reply [flat|nested] 19+ messages in thread
* [PATCH v2 07/12] hw/arm: xenpvh: Break out a common PVH machine
2024-08-20 14:29 [PATCH v2 00/12] xen: pvh: Partial QOM:fication with new x86 PVH machine Edgar E. Iglesias
` (5 preceding siblings ...)
2024-08-20 14:29 ` [PATCH v2 06/12] hw/arm: xenpvh: Move stubbed functions to xen-stubs.c Edgar E. Iglesias
@ 2024-08-20 14:29 ` Edgar E. Iglesias
2024-08-21 1:16 ` Stefano Stabellini
2024-08-20 14:29 ` [PATCH v2 08/12] hw/arm: xenpvh: Rename xen_arm.c -> xen-pvh.c Edgar E. Iglesias
` (4 subsequent siblings)
11 siblings, 1 reply; 19+ messages in thread
From: Edgar E. Iglesias @ 2024-08-20 14:29 UTC (permalink / raw)
To: qemu-devel
Cc: sstabellini, anthony, paul, peter.maydell, alex.bennee,
xenia.ragiadakou, jason.andryuk, edgar.iglesias, xen-devel,
Edgar E. Iglesias, qemu-arm
From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
Break out a common Xen PVH machine in preparation for
adding a x86 Xen PVH machine.
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
---
hw/arm/trace-events | 5 -
hw/arm/xen_arm.c | 198 +++--------------------
hw/xen/meson.build | 1 +
hw/xen/trace-events | 4 +
hw/xen/xen-pvh-common.c | 275 ++++++++++++++++++++++++++++++++
include/hw/xen/xen-pvh-common.h | 59 +++++++
6 files changed, 358 insertions(+), 184 deletions(-)
create mode 100644 hw/xen/xen-pvh-common.c
create mode 100644 include/hw/xen/xen-pvh-common.h
diff --git a/hw/arm/trace-events b/hw/arm/trace-events
index be6c8f720b..c64ad344bd 100644
--- a/hw/arm/trace-events
+++ b/hw/arm/trace-events
@@ -68,10 +68,5 @@ z2_aer915_send_too_long(int8_t msg) "message too long (%i bytes)"
z2_aer915_send(uint8_t reg, uint8_t value) "reg %d value 0x%02x"
z2_aer915_event(int8_t event, int8_t len) "i2c event =0x%x len=%d bytes"
-# xen_arm.c
-xen_create_virtio_mmio_devices(int i, int irq, uint64_t base) "Created virtio-mmio device %d: irq %d base 0x%"PRIx64
-xen_init_ram(uint64_t machine_ram_size) "Initialized xen ram with size 0x%"PRIx64
-xen_enable_tpm(uint64_t addr) "Connected tpmdev at address 0x%"PRIx64
-
# bcm2838.c
bcm2838_gic_set_irq(int irq, int level) "gic irq:%d lvl:%d"
diff --git a/hw/arm/xen_arm.c b/hw/arm/xen_arm.c
index f0868e7be5..04cb9855af 100644
--- a/hw/arm/xen_arm.c
+++ b/hw/arm/xen_arm.c
@@ -7,44 +7,12 @@
#include "qemu/osdep.h"
#include "qemu/error-report.h"
#include "qapi/qapi-commands-migration.h"
-#include "qapi/visitor.h"
#include "hw/boards.h"
-#include "hw/irq.h"
-#include "hw/sysbus.h"
-#include "sysemu/block-backend.h"
-#include "sysemu/tpm_backend.h"
#include "sysemu/sysemu.h"
-#include "hw/xen/xen-hvm-common.h"
-#include "sysemu/tpm.h"
+#include "hw/xen/xen-pvh-common.h"
#include "hw/xen/arch_hvm.h"
-#include "trace.h"
#define TYPE_XEN_ARM MACHINE_TYPE_NAME("xenpvh")
-OBJECT_DECLARE_SIMPLE_TYPE(XenArmState, XEN_ARM)
-
-static const MemoryListener xen_memory_listener = {
- .region_add = xen_region_add,
- .region_del = xen_region_del,
- .log_start = NULL,
- .log_stop = NULL,
- .log_sync = NULL,
- .log_global_start = NULL,
- .log_global_stop = NULL,
- .priority = MEMORY_LISTENER_PRIORITY_ACCEL,
-};
-
-struct XenArmState {
- /*< private >*/
- MachineState parent;
-
- XenIOState *state;
-
- struct {
- uint64_t tpm_base_addr;
- } cfg;
-};
-
-static MemoryRegion ram_lo, ram_hi;
/*
* VIRTIO_MMIO_DEV_SIZE is imported from tools/libs/light/libxl_arm.c under Xen
@@ -57,147 +25,26 @@ static MemoryRegion ram_lo, ram_hi;
#define NR_VIRTIO_MMIO_DEVICES \
(GUEST_VIRTIO_MMIO_SPI_LAST - GUEST_VIRTIO_MMIO_SPI_FIRST)
-static void xen_set_irq(void *opaque, int irq, int level)
-{
- if (xendevicemodel_set_irq_level(xen_dmod, xen_domid, irq, level)) {
- error_report("xendevicemodel_set_irq_level failed");
- }
-}
-
-static void xen_create_virtio_mmio_devices(XenArmState *xam)
-{
- int i;
-
- for (i = 0; i < NR_VIRTIO_MMIO_DEVICES; i++) {
- hwaddr base = GUEST_VIRTIO_MMIO_BASE + i * VIRTIO_MMIO_DEV_SIZE;
- qemu_irq irq = qemu_allocate_irq(xen_set_irq, NULL,
- GUEST_VIRTIO_MMIO_SPI_FIRST + i);
-
- sysbus_create_simple("virtio-mmio", base, irq);
-
- trace_xen_create_virtio_mmio_devices(i,
- GUEST_VIRTIO_MMIO_SPI_FIRST + i,
- base);
- }
-}
-
-static void xen_init_ram(MachineState *machine)
+static void xen_arm_instance_init(Object *obj)
{
- MemoryRegion *sysmem = get_system_memory();
- ram_addr_t block_len, ram_size[GUEST_RAM_BANKS];
-
- trace_xen_init_ram(machine->ram_size);
- if (machine->ram_size <= GUEST_RAM0_SIZE) {
- ram_size[0] = machine->ram_size;
- ram_size[1] = 0;
- block_len = GUEST_RAM0_BASE + ram_size[0];
- } else {
- ram_size[0] = GUEST_RAM0_SIZE;
- ram_size[1] = machine->ram_size - GUEST_RAM0_SIZE;
- block_len = GUEST_RAM1_BASE + ram_size[1];
- }
+ XenPVHMachineState *s = XEN_PVH_MACHINE(obj);
- memory_region_init_ram(&xen_memory, NULL, "xen.ram", block_len,
- &error_fatal);
+ /* Default values. */
+ s->cfg.ram_low = (MemMapEntry) { GUEST_RAM0_BASE, GUEST_RAM0_SIZE };
+ s->cfg.ram_high = (MemMapEntry) { GUEST_RAM1_BASE, GUEST_RAM1_SIZE };
- memory_region_init_alias(&ram_lo, NULL, "xen.ram.lo", &xen_memory,
- GUEST_RAM0_BASE, ram_size[0]);
- memory_region_add_subregion(sysmem, GUEST_RAM0_BASE, &ram_lo);
- if (ram_size[1] > 0) {
- memory_region_init_alias(&ram_hi, NULL, "xen.ram.hi", &xen_memory,
- GUEST_RAM1_BASE, ram_size[1]);
- memory_region_add_subregion(sysmem, GUEST_RAM1_BASE, &ram_hi);
- }
-
- /* Setup support for grants. */
- memory_region_init_ram(&xen_grants, NULL, "xen.grants", block_len,
- &error_fatal);
- memory_region_add_subregion(sysmem, XEN_GRANT_ADDR_OFF, &xen_grants);
+ s->cfg.virtio_mmio_num = NR_VIRTIO_MMIO_DEVICES;
+ s->cfg.virtio_mmio_irq_base = GUEST_VIRTIO_MMIO_SPI_FIRST;
+ s->cfg.virtio_mmio = (MemMapEntry) { GUEST_VIRTIO_MMIO_BASE,
+ VIRTIO_MMIO_DEV_SIZE };
}
-#ifdef CONFIG_TPM
-static void xen_enable_tpm(XenArmState *xam)
-{
- Error *errp = NULL;
- DeviceState *dev;
- SysBusDevice *busdev;
-
- TPMBackend *be = qemu_find_tpm_be("tpm0");
- if (be == NULL) {
- error_report("Couldn't find tmp0 backend");
- return;
- }
- dev = qdev_new(TYPE_TPM_TIS_SYSBUS);
- object_property_set_link(OBJECT(dev), "tpmdev", OBJECT(be), &errp);
- object_property_set_str(OBJECT(dev), "tpmdev", be->id, &errp);
- busdev = SYS_BUS_DEVICE(dev);
- sysbus_realize_and_unref(busdev, &error_fatal);
- sysbus_mmio_map(busdev, 0, xam->cfg.tpm_base_addr);
-
- trace_xen_enable_tpm(xam->cfg.tpm_base_addr);
-}
-#endif
-
-static void xen_arm_init(MachineState *machine)
-{
- XenArmState *xam = XEN_ARM(machine);
-
- xam->state = g_new0(XenIOState, 1);
-
- if (machine->ram_size == 0) {
- warn_report("%s: ram size not specified. QEMU machine started"
- " without IOREQ (no emulated devices including virtio)",
- MACHINE_CLASS(object_get_class(OBJECT(machine)))->desc);
- return;
- }
-
- xen_init_ram(machine);
-
- xen_register_ioreq(xam->state, machine->smp.max_cpus, &xen_memory_listener);
-
- xen_create_virtio_mmio_devices(xam);
-
-#ifdef CONFIG_TPM
- if (xam->cfg.tpm_base_addr) {
- xen_enable_tpm(xam);
- } else {
- warn_report("tpm-base-addr is not provided. TPM will not be enabled");
- }
-#endif
-}
-
-#ifdef CONFIG_TPM
-static void xen_arm_get_tpm_base_addr(Object *obj, Visitor *v,
- const char *name, void *opaque,
- Error **errp)
-{
- XenArmState *xam = XEN_ARM(obj);
- uint64_t value = xam->cfg.tpm_base_addr;
-
- visit_type_uint64(v, name, &value, errp);
-}
-
-static void xen_arm_set_tpm_base_addr(Object *obj, Visitor *v,
- const char *name, void *opaque,
- Error **errp)
-{
- XenArmState *xam = XEN_ARM(obj);
- uint64_t value;
-
- if (!visit_type_uint64(v, name, &value, errp)) {
- return;
- }
-
- xam->cfg.tpm_base_addr = value;
-}
-#endif
-
static void xen_arm_machine_class_init(ObjectClass *oc, void *data)
{
-
+ XenPVHMachineClass *xpc = XEN_PVH_MACHINE_CLASS(oc);
MachineClass *mc = MACHINE_CLASS(oc);
+
mc->desc = "Xen PVH ARM machine";
- mc->init = xen_arm_init;
/*
* mc->max_cpus holds the MAX value allowed in the -smp command-line opts.
@@ -218,27 +65,20 @@ static void xen_arm_machine_class_init(ObjectClass *oc, void *data)
* mc->max_cpus, QEMU will bail out with an error message.
*/
mc->max_cpus = GUEST_MAX_VCPUS;
- mc->default_machine_opts = "accel=xen";
- /* Set explicitly here to make sure that real ram_size is passed */
- mc->default_ram_size = 0;
-#ifdef CONFIG_TPM
- object_class_property_add(oc, "tpm-base-addr", "uint64_t",
- xen_arm_get_tpm_base_addr,
- xen_arm_set_tpm_base_addr,
- NULL, NULL);
- object_class_property_set_description(oc, "tpm-base-addr",
- "Set Base address for TPM device.");
+ /* List of supported features known to work on PVH ARM. */
+ xpc->has_tpm = true;
+ xpc->has_virtio_mmio = true;
- machine_class_allow_dynamic_sysbus_dev(mc, TYPE_TPM_TIS_SYSBUS);
-#endif
+ xen_pvh_class_setup_common_props(xpc);
}
static const TypeInfo xen_arm_machine_type = {
.name = TYPE_XEN_ARM,
- .parent = TYPE_MACHINE,
+ .parent = TYPE_XEN_PVH_MACHINE,
.class_init = xen_arm_machine_class_init,
- .instance_size = sizeof(XenArmState),
+ .instance_size = sizeof(XenPVHMachineState),
+ .instance_init = xen_arm_instance_init,
};
static void xen_arm_machine_register_types(void)
diff --git a/hw/xen/meson.build b/hw/xen/meson.build
index d887fa9ba4..4a486e3673 100644
--- a/hw/xen/meson.build
+++ b/hw/xen/meson.build
@@ -15,6 +15,7 @@ xen_specific_ss = ss.source_set()
xen_specific_ss.add(files(
'xen-mapcache.c',
'xen-hvm-common.c',
+ 'xen-pvh-common.c',
))
if have_xen_pci_passthrough
xen_specific_ss.add(files(
diff --git a/hw/xen/trace-events b/hw/xen/trace-events
index d1b27f6c11..a07fe41c6d 100644
--- a/hw/xen/trace-events
+++ b/hw/xen/trace-events
@@ -64,6 +64,10 @@ destroy_hvm_domain_cannot_acquire_handle(void) "Cannot acquire xenctrl handle"
destroy_hvm_domain_failed_action(const char *action, int sts, char *errno_s) "xc_domain_shutdown failed to issue %s, sts %d, %s"
destroy_hvm_domain_action(int xen_domid, const char *action) "Issued domain %d %s"
+# xen-pvh-common.c
+xen_create_virtio_mmio_devices(int i, int irq, uint64_t base) "Created virtio-mmio device %d: irq %d base 0x%"PRIx64
+xen_enable_tpm(uint64_t addr) "Connected tpmdev at address 0x%"PRIx64
+
# xen-mapcache.c
xen_map_cache(uint64_t phys_addr) "want 0x%"PRIx64
xen_remap_bucket(uint64_t index) "index 0x%"PRIx64
diff --git a/hw/xen/xen-pvh-common.c b/hw/xen/xen-pvh-common.c
new file mode 100644
index 0000000000..880e8143d7
--- /dev/null
+++ b/hw/xen/xen-pvh-common.c
@@ -0,0 +1,275 @@
+/*
+ * QEMU Xen PVH machine - common code.
+ *
+ * Copyright (c) 2024 Advanced Micro Devices, Inc.
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#include "qemu/osdep.h"
+#include "qemu/error-report.h"
+#include "qapi/error.h"
+#include "qapi/visitor.h"
+#include "hw/boards.h"
+#include "hw/irq.h"
+#include "hw/sysbus.h"
+#include "sysemu/sysemu.h"
+#include "sysemu/tpm.h"
+#include "sysemu/tpm_backend.h"
+#include "hw/xen/xen-pvh-common.h"
+#include "trace.h"
+
+static const MemoryListener xen_memory_listener = {
+ .region_add = xen_region_add,
+ .region_del = xen_region_del,
+ .log_start = NULL,
+ .log_stop = NULL,
+ .log_sync = NULL,
+ .log_global_start = NULL,
+ .log_global_stop = NULL,
+ .priority = MEMORY_LISTENER_PRIORITY_ACCEL,
+};
+
+static void xen_pvh_init_ram(XenPVHMachineState *s,
+ MemoryRegion *sysmem)
+{
+ MachineState *ms = MACHINE(s);
+ ram_addr_t block_len, ram_size[2];
+
+ if (ms->ram_size <= s->cfg.ram_low.size) {
+ ram_size[0] = ms->ram_size;
+ ram_size[1] = 0;
+ block_len = s->cfg.ram_low.base + ram_size[0];
+ } else {
+ ram_size[0] = s->cfg.ram_low.size;
+ ram_size[1] = ms->ram_size - s->cfg.ram_low.size;
+ block_len = s->cfg.ram_high.base + ram_size[1];
+ }
+
+ memory_region_init_ram(&xen_memory, NULL, "xen.ram", block_len,
+ &error_fatal);
+
+ memory_region_init_alias(&s->ram.low, NULL, "xen.ram.lo", &xen_memory,
+ s->cfg.ram_low.base, ram_size[0]);
+ memory_region_add_subregion(sysmem, s->cfg.ram_low.base, &s->ram.low);
+ if (ram_size[1] > 0) {
+ memory_region_init_alias(&s->ram.high, NULL, "xen.ram.hi", &xen_memory,
+ s->cfg.ram_high.base, ram_size[1]);
+ memory_region_add_subregion(sysmem, s->cfg.ram_high.base, &s->ram.high);
+ }
+
+ /* Setup support for grants. */
+ memory_region_init_ram(&xen_grants, NULL, "xen.grants", block_len,
+ &error_fatal);
+ memory_region_add_subregion(sysmem, XEN_GRANT_ADDR_OFF, &xen_grants);
+}
+
+static void xen_set_irq(void *opaque, int irq, int level)
+{
+ if (xendevicemodel_set_irq_level(xen_dmod, xen_domid, irq, level)) {
+ error_report("xendevicemodel_set_irq_level failed");
+ }
+}
+
+static void xen_create_virtio_mmio_devices(XenPVHMachineState *s)
+{
+ int i;
+
+ for (i = 0; i < s->cfg.virtio_mmio_num; i++) {
+ hwaddr base = s->cfg.virtio_mmio.base + i * s->cfg.virtio_mmio.size;
+ qemu_irq irq = qemu_allocate_irq(xen_set_irq, NULL,
+ s->cfg.virtio_mmio_irq_base + i);
+
+ sysbus_create_simple("virtio-mmio", base, irq);
+
+ trace_xen_create_virtio_mmio_devices(i,
+ s->cfg.virtio_mmio_irq_base + i,
+ base);
+ }
+}
+
+#ifdef CONFIG_TPM
+static void xen_enable_tpm(XenPVHMachineState *s)
+{
+ Error *errp = NULL;
+ DeviceState *dev;
+ SysBusDevice *busdev;
+
+ TPMBackend *be = qemu_find_tpm_be("tpm0");
+ if (be == NULL) {
+ error_report("Couldn't find tmp0 backend");
+ return;
+ }
+ dev = qdev_new(TYPE_TPM_TIS_SYSBUS);
+ object_property_set_link(OBJECT(dev), "tpmdev", OBJECT(be), &errp);
+ object_property_set_str(OBJECT(dev), "tpmdev", be->id, &errp);
+ busdev = SYS_BUS_DEVICE(dev);
+ sysbus_realize_and_unref(busdev, &error_fatal);
+ sysbus_mmio_map(busdev, 0, s->cfg.tpm.base);
+
+ trace_xen_enable_tpm(s->cfg.tpm.base);
+}
+#endif
+
+static void xen_pvh_init(MachineState *ms)
+{
+ XenPVHMachineState *s = XEN_PVH_MACHINE(ms);
+ XenPVHMachineClass *xpc = XEN_PVH_MACHINE_GET_CLASS(s);
+ MemoryRegion *sysmem = get_system_memory();
+
+ if (ms->ram_size == 0) {
+ warn_report("%s: ram size not specified. QEMU machine started"
+ " without IOREQ (no emulated devices including virtio)",
+ MACHINE_CLASS(object_get_class(OBJECT(ms)))->desc);
+ return;
+ }
+
+ xen_pvh_init_ram(s, sysmem);
+ xen_register_ioreq(&s->ioreq, ms->smp.max_cpus, &xen_memory_listener);
+
+ if (s->cfg.virtio_mmio_num) {
+ xen_create_virtio_mmio_devices(s);
+ }
+
+#ifdef CONFIG_TPM
+ if (xpc->has_tpm) {
+ if (s->cfg.tpm.base) {
+ xen_enable_tpm(s);
+ } else {
+ warn_report("tpm-base-addr is not set. TPM will not be enabled");
+ }
+ }
+#endif
+
+ /* Call the implementation specific init. */
+ if (xpc->init) {
+ xpc->init(ms);
+ }
+}
+
+#define XEN_PVH_PROP_MEMMAP_SETTER(n, f) \
+static void xen_pvh_set_ ## n ## _ ## f(Object *obj, Visitor *v, \
+ const char *name, void *opaque, \
+ Error **errp) \
+{ \
+ XenPVHMachineState *xp = XEN_PVH_MACHINE(obj); \
+ uint64_t value; \
+ \
+ if (!visit_type_size(v, name, &value, errp)) { \
+ return; \
+ } \
+ xp->cfg.n.f = value; \
+}
+
+#define XEN_PVH_PROP_MEMMAP_GETTER(n, f) \
+static void xen_pvh_get_ ## n ## _ ## f(Object *obj, Visitor *v, \
+ const char *name, void *opaque, \
+ Error **errp) \
+{ \
+ XenPVHMachineState *xp = XEN_PVH_MACHINE(obj); \
+ uint64_t value = xp->cfg.n.f; \
+ \
+ visit_type_uint64(v, name, &value, errp); \
+}
+
+#define XEN_PVH_PROP_MEMMAP_BASE(n) \
+ XEN_PVH_PROP_MEMMAP_SETTER(n, base) \
+ XEN_PVH_PROP_MEMMAP_GETTER(n, base) \
+
+#define XEN_PVH_PROP_MEMMAP_SIZE(n) \
+ XEN_PVH_PROP_MEMMAP_SETTER(n, size) \
+ XEN_PVH_PROP_MEMMAP_GETTER(n, size)
+
+#define XEN_PVH_PROP_MEMMAP(n) \
+ XEN_PVH_PROP_MEMMAP_BASE(n) \
+ XEN_PVH_PROP_MEMMAP_SIZE(n)
+
+XEN_PVH_PROP_MEMMAP(ram_low)
+XEN_PVH_PROP_MEMMAP(ram_high)
+/* TPM only has a base-addr option. */
+XEN_PVH_PROP_MEMMAP_BASE(tpm)
+XEN_PVH_PROP_MEMMAP(virtio_mmio)
+
+void xen_pvh_class_setup_common_props(XenPVHMachineClass *xpc)
+{
+ ObjectClass *oc = OBJECT_CLASS(xpc);
+ MachineClass *mc = MACHINE_CLASS(xpc);
+
+#define OC_MEMMAP_PROP_BASE(c, prop_name, name) \
+do { \
+ object_class_property_add(c, prop_name "-base", "uint64_t", \
+ xen_pvh_get_ ## name ## _base, \
+ xen_pvh_set_ ## name ## _base, NULL, NULL); \
+ object_class_property_set_description(oc, prop_name "-base", \
+ "Set base address for " prop_name); \
+} while (0)
+
+#define OC_MEMMAP_PROP_SIZE(c, prop_name, name) \
+do { \
+ object_class_property_add(c, prop_name "-size", "uint64_t", \
+ xen_pvh_get_ ## name ## _size, \
+ xen_pvh_set_ ## name ## _size, NULL, NULL); \
+ object_class_property_set_description(oc, prop_name "-size", \
+ "Set memory range size for " prop_name); \
+} while (0)
+
+#define OC_MEMMAP_PROP(c, prop_name, name) \
+do { \
+ OC_MEMMAP_PROP_BASE(c, prop_name, name); \
+ OC_MEMMAP_PROP_SIZE(c, prop_name, name); \
+} while (0)
+
+ /*
+ * We provide memmap properties to allow Xen to move things to other
+ * addresses for example when users need to accomodate the memory-map
+ * for 1:1 mapped devices/memory.
+ */
+ OC_MEMMAP_PROP(oc, "ram-low", ram_low);
+ OC_MEMMAP_PROP(oc, "ram-high", ram_high);
+
+ if (xpc->has_virtio_mmio) {
+ OC_MEMMAP_PROP(oc, "virtio-mmio", virtio_mmio);
+ }
+
+#ifdef CONFIG_TPM
+ if (xpc->has_tpm) {
+ object_class_property_add(oc, "tpm-base-addr", "uint64_t",
+ xen_pvh_get_tpm_base,
+ xen_pvh_set_tpm_base,
+ NULL, NULL);
+ object_class_property_set_description(oc, "tpm-base-addr",
+ "Set Base address for TPM device.");
+
+ machine_class_allow_dynamic_sysbus_dev(mc, TYPE_TPM_TIS_SYSBUS);
+ }
+#endif
+}
+
+static void xen_pvh_class_init(ObjectClass *oc, void *data)
+{
+ MachineClass *mc = MACHINE_CLASS(oc);
+
+ mc->init = xen_pvh_init;
+
+ mc->desc = "Xen PVH machine";
+ mc->max_cpus = 1;
+ mc->default_machine_opts = "accel=xen";
+ /* Set to zero to make sure that the real ram size is passed. */
+ mc->default_ram_size = 0;
+}
+
+static const TypeInfo xen_pvh_info = {
+ .name = TYPE_XEN_PVH_MACHINE,
+ .parent = TYPE_MACHINE,
+ .abstract = true,
+ .instance_size = sizeof(XenPVHMachineState),
+ .class_size = sizeof(XenPVHMachineClass),
+ .class_init = xen_pvh_class_init,
+};
+
+static void xen_pvh_register_types(void)
+{
+ type_register_static(&xen_pvh_info);
+}
+
+type_init(xen_pvh_register_types);
diff --git a/include/hw/xen/xen-pvh-common.h b/include/hw/xen/xen-pvh-common.h
new file mode 100644
index 0000000000..77fd98b9fe
--- /dev/null
+++ b/include/hw/xen/xen-pvh-common.h
@@ -0,0 +1,59 @@
+/*
+ * QEMU Xen PVH machine - common code.
+ *
+ * Copyright (c) 2024 Advanced Micro Devices, Inc.
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#ifndef XEN_PVH_COMMON_H__
+#define XEN_PVH_COMMON_H__
+
+#include <assert.h>
+#include "hw/sysbus.h"
+#include "hw/hw.h"
+#include "hw/xen/xen-hvm-common.h"
+#include "hw/pci-host/gpex.h"
+
+#define TYPE_XEN_PVH_MACHINE MACHINE_TYPE_NAME("xen-pvh-base")
+OBJECT_DECLARE_TYPE(XenPVHMachineState, XenPVHMachineClass,
+ XEN_PVH_MACHINE)
+
+struct XenPVHMachineClass {
+ MachineClass parent;
+
+ /* PVH implementation specific init. */
+ void (*init)(MachineState *state);
+
+ /*
+ * Each implementation can optionally enable features that it
+ * supports and are known to work.
+ */
+ bool has_tpm;
+ bool has_virtio_mmio;
+};
+
+struct XenPVHMachineState {
+ /*< private >*/
+ MachineState parent;
+
+ XenIOState ioreq;
+
+ struct {
+ MemoryRegion low;
+ MemoryRegion high;
+ } ram;
+
+ struct {
+ MemMapEntry ram_low, ram_high;
+ MemMapEntry tpm;
+
+ /* Virtio-mmio */
+ MemMapEntry virtio_mmio;
+ uint32_t virtio_mmio_num;
+ uint32_t virtio_mmio_irq_base;
+ } cfg;
+};
+
+void xen_pvh_class_setup_common_props(XenPVHMachineClass *xpc);
+#endif
--
2.43.0
^ permalink raw reply related [flat|nested] 19+ messages in thread* Re: [PATCH v2 07/12] hw/arm: xenpvh: Break out a common PVH machine
2024-08-20 14:29 ` [PATCH v2 07/12] hw/arm: xenpvh: Break out a common PVH machine Edgar E. Iglesias
@ 2024-08-21 1:16 ` Stefano Stabellini
0 siblings, 0 replies; 19+ messages in thread
From: Stefano Stabellini @ 2024-08-21 1:16 UTC (permalink / raw)
To: Edgar E. Iglesias
Cc: qemu-devel, sstabellini, anthony, paul, peter.maydell,
alex.bennee, xenia.ragiadakou, jason.andryuk, edgar.iglesias,
xen-devel, qemu-arm
On Tue, 20 Aug 2024, Edgar E. Iglesias wrote:
> From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
>
> Break out a common Xen PVH machine in preparation for
> adding a x86 Xen PVH machine.
>
> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
> ---
> hw/arm/trace-events | 5 -
> hw/arm/xen_arm.c | 198 +++--------------------
> hw/xen/meson.build | 1 +
> hw/xen/trace-events | 4 +
> hw/xen/xen-pvh-common.c | 275 ++++++++++++++++++++++++++++++++
> include/hw/xen/xen-pvh-common.h | 59 +++++++
> 6 files changed, 358 insertions(+), 184 deletions(-)
> create mode 100644 hw/xen/xen-pvh-common.c
> create mode 100644 include/hw/xen/xen-pvh-common.h
>
> diff --git a/hw/arm/trace-events b/hw/arm/trace-events
> index be6c8f720b..c64ad344bd 100644
> --- a/hw/arm/trace-events
> +++ b/hw/arm/trace-events
> @@ -68,10 +68,5 @@ z2_aer915_send_too_long(int8_t msg) "message too long (%i bytes)"
> z2_aer915_send(uint8_t reg, uint8_t value) "reg %d value 0x%02x"
> z2_aer915_event(int8_t event, int8_t len) "i2c event =0x%x len=%d bytes"
>
> -# xen_arm.c
> -xen_create_virtio_mmio_devices(int i, int irq, uint64_t base) "Created virtio-mmio device %d: irq %d base 0x%"PRIx64
> -xen_init_ram(uint64_t machine_ram_size) "Initialized xen ram with size 0x%"PRIx64
> -xen_enable_tpm(uint64_t addr) "Connected tpmdev at address 0x%"PRIx64
> -
> # bcm2838.c
> bcm2838_gic_set_irq(int irq, int level) "gic irq:%d lvl:%d"
> diff --git a/hw/arm/xen_arm.c b/hw/arm/xen_arm.c
> index f0868e7be5..04cb9855af 100644
> --- a/hw/arm/xen_arm.c
> +++ b/hw/arm/xen_arm.c
> @@ -7,44 +7,12 @@
> #include "qemu/osdep.h"
> #include "qemu/error-report.h"
> #include "qapi/qapi-commands-migration.h"
> -#include "qapi/visitor.h"
> #include "hw/boards.h"
> -#include "hw/irq.h"
> -#include "hw/sysbus.h"
> -#include "sysemu/block-backend.h"
> -#include "sysemu/tpm_backend.h"
> #include "sysemu/sysemu.h"
> -#include "hw/xen/xen-hvm-common.h"
> -#include "sysemu/tpm.h"
> +#include "hw/xen/xen-pvh-common.h"
> #include "hw/xen/arch_hvm.h"
> -#include "trace.h"
>
> #define TYPE_XEN_ARM MACHINE_TYPE_NAME("xenpvh")
> -OBJECT_DECLARE_SIMPLE_TYPE(XenArmState, XEN_ARM)
> -
> -static const MemoryListener xen_memory_listener = {
> - .region_add = xen_region_add,
> - .region_del = xen_region_del,
> - .log_start = NULL,
> - .log_stop = NULL,
> - .log_sync = NULL,
> - .log_global_start = NULL,
> - .log_global_stop = NULL,
> - .priority = MEMORY_LISTENER_PRIORITY_ACCEL,
> -};
> -
> -struct XenArmState {
> - /*< private >*/
> - MachineState parent;
> -
> - XenIOState *state;
> -
> - struct {
> - uint64_t tpm_base_addr;
> - } cfg;
> -};
> -
> -static MemoryRegion ram_lo, ram_hi;
>
> /*
> * VIRTIO_MMIO_DEV_SIZE is imported from tools/libs/light/libxl_arm.c under Xen
> @@ -57,147 +25,26 @@ static MemoryRegion ram_lo, ram_hi;
> #define NR_VIRTIO_MMIO_DEVICES \
> (GUEST_VIRTIO_MMIO_SPI_LAST - GUEST_VIRTIO_MMIO_SPI_FIRST)
>
> -static void xen_set_irq(void *opaque, int irq, int level)
> -{
> - if (xendevicemodel_set_irq_level(xen_dmod, xen_domid, irq, level)) {
> - error_report("xendevicemodel_set_irq_level failed");
> - }
> -}
> -
> -static void xen_create_virtio_mmio_devices(XenArmState *xam)
> -{
> - int i;
> -
> - for (i = 0; i < NR_VIRTIO_MMIO_DEVICES; i++) {
> - hwaddr base = GUEST_VIRTIO_MMIO_BASE + i * VIRTIO_MMIO_DEV_SIZE;
> - qemu_irq irq = qemu_allocate_irq(xen_set_irq, NULL,
> - GUEST_VIRTIO_MMIO_SPI_FIRST + i);
> -
> - sysbus_create_simple("virtio-mmio", base, irq);
> -
> - trace_xen_create_virtio_mmio_devices(i,
> - GUEST_VIRTIO_MMIO_SPI_FIRST + i,
> - base);
> - }
> -}
> -
> -static void xen_init_ram(MachineState *machine)
> +static void xen_arm_instance_init(Object *obj)
> {
> - MemoryRegion *sysmem = get_system_memory();
> - ram_addr_t block_len, ram_size[GUEST_RAM_BANKS];
> -
> - trace_xen_init_ram(machine->ram_size);
> - if (machine->ram_size <= GUEST_RAM0_SIZE) {
> - ram_size[0] = machine->ram_size;
> - ram_size[1] = 0;
> - block_len = GUEST_RAM0_BASE + ram_size[0];
> - } else {
> - ram_size[0] = GUEST_RAM0_SIZE;
> - ram_size[1] = machine->ram_size - GUEST_RAM0_SIZE;
> - block_len = GUEST_RAM1_BASE + ram_size[1];
> - }
> + XenPVHMachineState *s = XEN_PVH_MACHINE(obj);
>
> - memory_region_init_ram(&xen_memory, NULL, "xen.ram", block_len,
> - &error_fatal);
> + /* Default values. */
> + s->cfg.ram_low = (MemMapEntry) { GUEST_RAM0_BASE, GUEST_RAM0_SIZE };
> + s->cfg.ram_high = (MemMapEntry) { GUEST_RAM1_BASE, GUEST_RAM1_SIZE };
>
> - memory_region_init_alias(&ram_lo, NULL, "xen.ram.lo", &xen_memory,
> - GUEST_RAM0_BASE, ram_size[0]);
> - memory_region_add_subregion(sysmem, GUEST_RAM0_BASE, &ram_lo);
> - if (ram_size[1] > 0) {
> - memory_region_init_alias(&ram_hi, NULL, "xen.ram.hi", &xen_memory,
> - GUEST_RAM1_BASE, ram_size[1]);
> - memory_region_add_subregion(sysmem, GUEST_RAM1_BASE, &ram_hi);
> - }
> -
> - /* Setup support for grants. */
> - memory_region_init_ram(&xen_grants, NULL, "xen.grants", block_len,
> - &error_fatal);
> - memory_region_add_subregion(sysmem, XEN_GRANT_ADDR_OFF, &xen_grants);
> + s->cfg.virtio_mmio_num = NR_VIRTIO_MMIO_DEVICES;
> + s->cfg.virtio_mmio_irq_base = GUEST_VIRTIO_MMIO_SPI_FIRST;
> + s->cfg.virtio_mmio = (MemMapEntry) { GUEST_VIRTIO_MMIO_BASE,
> + VIRTIO_MMIO_DEV_SIZE };
> }
>
> -#ifdef CONFIG_TPM
> -static void xen_enable_tpm(XenArmState *xam)
> -{
> - Error *errp = NULL;
> - DeviceState *dev;
> - SysBusDevice *busdev;
> -
> - TPMBackend *be = qemu_find_tpm_be("tpm0");
> - if (be == NULL) {
> - error_report("Couldn't find tmp0 backend");
> - return;
> - }
> - dev = qdev_new(TYPE_TPM_TIS_SYSBUS);
> - object_property_set_link(OBJECT(dev), "tpmdev", OBJECT(be), &errp);
> - object_property_set_str(OBJECT(dev), "tpmdev", be->id, &errp);
> - busdev = SYS_BUS_DEVICE(dev);
> - sysbus_realize_and_unref(busdev, &error_fatal);
> - sysbus_mmio_map(busdev, 0, xam->cfg.tpm_base_addr);
> -
> - trace_xen_enable_tpm(xam->cfg.tpm_base_addr);
> -}
> -#endif
> -
> -static void xen_arm_init(MachineState *machine)
> -{
> - XenArmState *xam = XEN_ARM(machine);
> -
> - xam->state = g_new0(XenIOState, 1);
> -
> - if (machine->ram_size == 0) {
> - warn_report("%s: ram size not specified. QEMU machine started"
> - " without IOREQ (no emulated devices including virtio)",
> - MACHINE_CLASS(object_get_class(OBJECT(machine)))->desc);
> - return;
> - }
> -
> - xen_init_ram(machine);
> -
> - xen_register_ioreq(xam->state, machine->smp.max_cpus, &xen_memory_listener);
> -
> - xen_create_virtio_mmio_devices(xam);
> -
> -#ifdef CONFIG_TPM
> - if (xam->cfg.tpm_base_addr) {
> - xen_enable_tpm(xam);
> - } else {
> - warn_report("tpm-base-addr is not provided. TPM will not be enabled");
> - }
> -#endif
> -}
> -
> -#ifdef CONFIG_TPM
> -static void xen_arm_get_tpm_base_addr(Object *obj, Visitor *v,
> - const char *name, void *opaque,
> - Error **errp)
> -{
> - XenArmState *xam = XEN_ARM(obj);
> - uint64_t value = xam->cfg.tpm_base_addr;
> -
> - visit_type_uint64(v, name, &value, errp);
> -}
> -
> -static void xen_arm_set_tpm_base_addr(Object *obj, Visitor *v,
> - const char *name, void *opaque,
> - Error **errp)
> -{
> - XenArmState *xam = XEN_ARM(obj);
> - uint64_t value;
> -
> - if (!visit_type_uint64(v, name, &value, errp)) {
> - return;
> - }
> -
> - xam->cfg.tpm_base_addr = value;
> -}
> -#endif
> -
> static void xen_arm_machine_class_init(ObjectClass *oc, void *data)
> {
> -
> + XenPVHMachineClass *xpc = XEN_PVH_MACHINE_CLASS(oc);
> MachineClass *mc = MACHINE_CLASS(oc);
> +
> mc->desc = "Xen PVH ARM machine";
> - mc->init = xen_arm_init;
>
> /*
> * mc->max_cpus holds the MAX value allowed in the -smp command-line opts.
> @@ -218,27 +65,20 @@ static void xen_arm_machine_class_init(ObjectClass *oc, void *data)
> * mc->max_cpus, QEMU will bail out with an error message.
> */
> mc->max_cpus = GUEST_MAX_VCPUS;
> - mc->default_machine_opts = "accel=xen";
> - /* Set explicitly here to make sure that real ram_size is passed */
> - mc->default_ram_size = 0;
>
> -#ifdef CONFIG_TPM
> - object_class_property_add(oc, "tpm-base-addr", "uint64_t",
> - xen_arm_get_tpm_base_addr,
> - xen_arm_set_tpm_base_addr,
> - NULL, NULL);
> - object_class_property_set_description(oc, "tpm-base-addr",
> - "Set Base address for TPM device.");
> + /* List of supported features known to work on PVH ARM. */
> + xpc->has_tpm = true;
> + xpc->has_virtio_mmio = true;
>
> - machine_class_allow_dynamic_sysbus_dev(mc, TYPE_TPM_TIS_SYSBUS);
> -#endif
> + xen_pvh_class_setup_common_props(xpc);
> }
>
> static const TypeInfo xen_arm_machine_type = {
> .name = TYPE_XEN_ARM,
> - .parent = TYPE_MACHINE,
> + .parent = TYPE_XEN_PVH_MACHINE,
> .class_init = xen_arm_machine_class_init,
> - .instance_size = sizeof(XenArmState),
> + .instance_size = sizeof(XenPVHMachineState),
> + .instance_init = xen_arm_instance_init,
> };
>
> static void xen_arm_machine_register_types(void)
> diff --git a/hw/xen/meson.build b/hw/xen/meson.build
> index d887fa9ba4..4a486e3673 100644
> --- a/hw/xen/meson.build
> +++ b/hw/xen/meson.build
> @@ -15,6 +15,7 @@ xen_specific_ss = ss.source_set()
> xen_specific_ss.add(files(
> 'xen-mapcache.c',
> 'xen-hvm-common.c',
> + 'xen-pvh-common.c',
> ))
> if have_xen_pci_passthrough
> xen_specific_ss.add(files(
> diff --git a/hw/xen/trace-events b/hw/xen/trace-events
> index d1b27f6c11..a07fe41c6d 100644
> --- a/hw/xen/trace-events
> +++ b/hw/xen/trace-events
> @@ -64,6 +64,10 @@ destroy_hvm_domain_cannot_acquire_handle(void) "Cannot acquire xenctrl handle"
> destroy_hvm_domain_failed_action(const char *action, int sts, char *errno_s) "xc_domain_shutdown failed to issue %s, sts %d, %s"
> destroy_hvm_domain_action(int xen_domid, const char *action) "Issued domain %d %s"
>
> +# xen-pvh-common.c
> +xen_create_virtio_mmio_devices(int i, int irq, uint64_t base) "Created virtio-mmio device %d: irq %d base 0x%"PRIx64
> +xen_enable_tpm(uint64_t addr) "Connected tpmdev at address 0x%"PRIx64
> +
> # xen-mapcache.c
> xen_map_cache(uint64_t phys_addr) "want 0x%"PRIx64
> xen_remap_bucket(uint64_t index) "index 0x%"PRIx64
> diff --git a/hw/xen/xen-pvh-common.c b/hw/xen/xen-pvh-common.c
> new file mode 100644
> index 0000000000..880e8143d7
> --- /dev/null
> +++ b/hw/xen/xen-pvh-common.c
> @@ -0,0 +1,275 @@
> +/*
> + * QEMU Xen PVH machine - common code.
> + *
> + * Copyright (c) 2024 Advanced Micro Devices, Inc.
> + *
> + * SPDX-License-Identifier: GPL-2.0-or-later
> + */
> +
> +#include "qemu/osdep.h"
> +#include "qemu/error-report.h"
> +#include "qapi/error.h"
> +#include "qapi/visitor.h"
> +#include "hw/boards.h"
> +#include "hw/irq.h"
> +#include "hw/sysbus.h"
> +#include "sysemu/sysemu.h"
> +#include "sysemu/tpm.h"
> +#include "sysemu/tpm_backend.h"
> +#include "hw/xen/xen-pvh-common.h"
> +#include "trace.h"
> +
> +static const MemoryListener xen_memory_listener = {
> + .region_add = xen_region_add,
> + .region_del = xen_region_del,
> + .log_start = NULL,
> + .log_stop = NULL,
> + .log_sync = NULL,
> + .log_global_start = NULL,
> + .log_global_stop = NULL,
> + .priority = MEMORY_LISTENER_PRIORITY_ACCEL,
> +};
> +
> +static void xen_pvh_init_ram(XenPVHMachineState *s,
> + MemoryRegion *sysmem)
> +{
> + MachineState *ms = MACHINE(s);
> + ram_addr_t block_len, ram_size[2];
> +
> + if (ms->ram_size <= s->cfg.ram_low.size) {
> + ram_size[0] = ms->ram_size;
> + ram_size[1] = 0;
> + block_len = s->cfg.ram_low.base + ram_size[0];
> + } else {
> + ram_size[0] = s->cfg.ram_low.size;
> + ram_size[1] = ms->ram_size - s->cfg.ram_low.size;
> + block_len = s->cfg.ram_high.base + ram_size[1];
> + }
> +
> + memory_region_init_ram(&xen_memory, NULL, "xen.ram", block_len,
> + &error_fatal);
> +
> + memory_region_init_alias(&s->ram.low, NULL, "xen.ram.lo", &xen_memory,
> + s->cfg.ram_low.base, ram_size[0]);
> + memory_region_add_subregion(sysmem, s->cfg.ram_low.base, &s->ram.low);
> + if (ram_size[1] > 0) {
> + memory_region_init_alias(&s->ram.high, NULL, "xen.ram.hi", &xen_memory,
> + s->cfg.ram_high.base, ram_size[1]);
> + memory_region_add_subregion(sysmem, s->cfg.ram_high.base, &s->ram.high);
> + }
> +
> + /* Setup support for grants. */
> + memory_region_init_ram(&xen_grants, NULL, "xen.grants", block_len,
> + &error_fatal);
> + memory_region_add_subregion(sysmem, XEN_GRANT_ADDR_OFF, &xen_grants);
> +}
> +
> +static void xen_set_irq(void *opaque, int irq, int level)
> +{
> + if (xendevicemodel_set_irq_level(xen_dmod, xen_domid, irq, level)) {
> + error_report("xendevicemodel_set_irq_level failed");
> + }
> +}
> +
> +static void xen_create_virtio_mmio_devices(XenPVHMachineState *s)
> +{
> + int i;
> +
> + for (i = 0; i < s->cfg.virtio_mmio_num; i++) {
> + hwaddr base = s->cfg.virtio_mmio.base + i * s->cfg.virtio_mmio.size;
> + qemu_irq irq = qemu_allocate_irq(xen_set_irq, NULL,
> + s->cfg.virtio_mmio_irq_base + i);
> +
> + sysbus_create_simple("virtio-mmio", base, irq);
> +
> + trace_xen_create_virtio_mmio_devices(i,
> + s->cfg.virtio_mmio_irq_base + i,
> + base);
> + }
> +}
> +
> +#ifdef CONFIG_TPM
> +static void xen_enable_tpm(XenPVHMachineState *s)
> +{
> + Error *errp = NULL;
> + DeviceState *dev;
> + SysBusDevice *busdev;
> +
> + TPMBackend *be = qemu_find_tpm_be("tpm0");
> + if (be == NULL) {
> + error_report("Couldn't find tmp0 backend");
> + return;
> + }
> + dev = qdev_new(TYPE_TPM_TIS_SYSBUS);
> + object_property_set_link(OBJECT(dev), "tpmdev", OBJECT(be), &errp);
> + object_property_set_str(OBJECT(dev), "tpmdev", be->id, &errp);
> + busdev = SYS_BUS_DEVICE(dev);
> + sysbus_realize_and_unref(busdev, &error_fatal);
> + sysbus_mmio_map(busdev, 0, s->cfg.tpm.base);
> +
> + trace_xen_enable_tpm(s->cfg.tpm.base);
> +}
> +#endif
> +
> +static void xen_pvh_init(MachineState *ms)
> +{
> + XenPVHMachineState *s = XEN_PVH_MACHINE(ms);
> + XenPVHMachineClass *xpc = XEN_PVH_MACHINE_GET_CLASS(s);
> + MemoryRegion *sysmem = get_system_memory();
> +
> + if (ms->ram_size == 0) {
> + warn_report("%s: ram size not specified. QEMU machine started"
> + " without IOREQ (no emulated devices including virtio)",
> + MACHINE_CLASS(object_get_class(OBJECT(ms)))->desc);
> + return;
> + }
> +
> + xen_pvh_init_ram(s, sysmem);
> + xen_register_ioreq(&s->ioreq, ms->smp.max_cpus, &xen_memory_listener);
> +
> + if (s->cfg.virtio_mmio_num) {
> + xen_create_virtio_mmio_devices(s);
> + }
> +
> +#ifdef CONFIG_TPM
> + if (xpc->has_tpm) {
> + if (s->cfg.tpm.base) {
> + xen_enable_tpm(s);
> + } else {
> + warn_report("tpm-base-addr is not set. TPM will not be enabled");
> + }
> + }
> +#endif
> +
> + /* Call the implementation specific init. */
> + if (xpc->init) {
> + xpc->init(ms);
> + }
> +}
> +
> +#define XEN_PVH_PROP_MEMMAP_SETTER(n, f) \
> +static void xen_pvh_set_ ## n ## _ ## f(Object *obj, Visitor *v, \
> + const char *name, void *opaque, \
> + Error **errp) \
> +{ \
> + XenPVHMachineState *xp = XEN_PVH_MACHINE(obj); \
> + uint64_t value; \
> + \
> + if (!visit_type_size(v, name, &value, errp)) { \
> + return; \
> + } \
> + xp->cfg.n.f = value; \
> +}
> +
> +#define XEN_PVH_PROP_MEMMAP_GETTER(n, f) \
> +static void xen_pvh_get_ ## n ## _ ## f(Object *obj, Visitor *v, \
> + const char *name, void *opaque, \
> + Error **errp) \
> +{ \
> + XenPVHMachineState *xp = XEN_PVH_MACHINE(obj); \
> + uint64_t value = xp->cfg.n.f; \
> + \
> + visit_type_uint64(v, name, &value, errp); \
> +}
> +
> +#define XEN_PVH_PROP_MEMMAP_BASE(n) \
> + XEN_PVH_PROP_MEMMAP_SETTER(n, base) \
> + XEN_PVH_PROP_MEMMAP_GETTER(n, base) \
> +
> +#define XEN_PVH_PROP_MEMMAP_SIZE(n) \
> + XEN_PVH_PROP_MEMMAP_SETTER(n, size) \
> + XEN_PVH_PROP_MEMMAP_GETTER(n, size)
> +
> +#define XEN_PVH_PROP_MEMMAP(n) \
> + XEN_PVH_PROP_MEMMAP_BASE(n) \
> + XEN_PVH_PROP_MEMMAP_SIZE(n)
> +
> +XEN_PVH_PROP_MEMMAP(ram_low)
> +XEN_PVH_PROP_MEMMAP(ram_high)
> +/* TPM only has a base-addr option. */
> +XEN_PVH_PROP_MEMMAP_BASE(tpm)
> +XEN_PVH_PROP_MEMMAP(virtio_mmio)
> +
> +void xen_pvh_class_setup_common_props(XenPVHMachineClass *xpc)
> +{
> + ObjectClass *oc = OBJECT_CLASS(xpc);
> + MachineClass *mc = MACHINE_CLASS(xpc);
> +
> +#define OC_MEMMAP_PROP_BASE(c, prop_name, name) \
> +do { \
> + object_class_property_add(c, prop_name "-base", "uint64_t", \
> + xen_pvh_get_ ## name ## _base, \
> + xen_pvh_set_ ## name ## _base, NULL, NULL); \
> + object_class_property_set_description(oc, prop_name "-base", \
> + "Set base address for " prop_name); \
> +} while (0)
> +
> +#define OC_MEMMAP_PROP_SIZE(c, prop_name, name) \
> +do { \
> + object_class_property_add(c, prop_name "-size", "uint64_t", \
> + xen_pvh_get_ ## name ## _size, \
> + xen_pvh_set_ ## name ## _size, NULL, NULL); \
> + object_class_property_set_description(oc, prop_name "-size", \
> + "Set memory range size for " prop_name); \
> +} while (0)
> +
> +#define OC_MEMMAP_PROP(c, prop_name, name) \
> +do { \
> + OC_MEMMAP_PROP_BASE(c, prop_name, name); \
> + OC_MEMMAP_PROP_SIZE(c, prop_name, name); \
> +} while (0)
> +
> + /*
> + * We provide memmap properties to allow Xen to move things to other
> + * addresses for example when users need to accomodate the memory-map
> + * for 1:1 mapped devices/memory.
> + */
> + OC_MEMMAP_PROP(oc, "ram-low", ram_low);
> + OC_MEMMAP_PROP(oc, "ram-high", ram_high);
> +
> + if (xpc->has_virtio_mmio) {
> + OC_MEMMAP_PROP(oc, "virtio-mmio", virtio_mmio);
> + }
> +
> +#ifdef CONFIG_TPM
> + if (xpc->has_tpm) {
> + object_class_property_add(oc, "tpm-base-addr", "uint64_t",
> + xen_pvh_get_tpm_base,
> + xen_pvh_set_tpm_base,
> + NULL, NULL);
> + object_class_property_set_description(oc, "tpm-base-addr",
> + "Set Base address for TPM device.");
> +
> + machine_class_allow_dynamic_sysbus_dev(mc, TYPE_TPM_TIS_SYSBUS);
> + }
> +#endif
> +}
> +
> +static void xen_pvh_class_init(ObjectClass *oc, void *data)
> +{
> + MachineClass *mc = MACHINE_CLASS(oc);
> +
> + mc->init = xen_pvh_init;
> +
> + mc->desc = "Xen PVH machine";
> + mc->max_cpus = 1;
> + mc->default_machine_opts = "accel=xen";
> + /* Set to zero to make sure that the real ram size is passed. */
> + mc->default_ram_size = 0;
> +}
> +
> +static const TypeInfo xen_pvh_info = {
> + .name = TYPE_XEN_PVH_MACHINE,
> + .parent = TYPE_MACHINE,
> + .abstract = true,
> + .instance_size = sizeof(XenPVHMachineState),
> + .class_size = sizeof(XenPVHMachineClass),
> + .class_init = xen_pvh_class_init,
> +};
> +
> +static void xen_pvh_register_types(void)
> +{
> + type_register_static(&xen_pvh_info);
> +}
> +
> +type_init(xen_pvh_register_types);
> diff --git a/include/hw/xen/xen-pvh-common.h b/include/hw/xen/xen-pvh-common.h
> new file mode 100644
> index 0000000000..77fd98b9fe
> --- /dev/null
> +++ b/include/hw/xen/xen-pvh-common.h
> @@ -0,0 +1,59 @@
> +/*
> + * QEMU Xen PVH machine - common code.
> + *
> + * Copyright (c) 2024 Advanced Micro Devices, Inc.
> + *
> + * SPDX-License-Identifier: GPL-2.0-or-later
> + */
> +
> +#ifndef XEN_PVH_COMMON_H__
> +#define XEN_PVH_COMMON_H__
> +
> +#include <assert.h>
> +#include "hw/sysbus.h"
> +#include "hw/hw.h"
> +#include "hw/xen/xen-hvm-common.h"
> +#include "hw/pci-host/gpex.h"
> +
> +#define TYPE_XEN_PVH_MACHINE MACHINE_TYPE_NAME("xen-pvh-base")
> +OBJECT_DECLARE_TYPE(XenPVHMachineState, XenPVHMachineClass,
> + XEN_PVH_MACHINE)
> +
> +struct XenPVHMachineClass {
> + MachineClass parent;
> +
> + /* PVH implementation specific init. */
> + void (*init)(MachineState *state);
> +
> + /*
> + * Each implementation can optionally enable features that it
> + * supports and are known to work.
> + */
> + bool has_tpm;
> + bool has_virtio_mmio;
> +};
> +
> +struct XenPVHMachineState {
> + /*< private >*/
> + MachineState parent;
> +
> + XenIOState ioreq;
> +
> + struct {
> + MemoryRegion low;
> + MemoryRegion high;
> + } ram;
> +
> + struct {
> + MemMapEntry ram_low, ram_high;
> + MemMapEntry tpm;
> +
> + /* Virtio-mmio */
> + MemMapEntry virtio_mmio;
> + uint32_t virtio_mmio_num;
> + uint32_t virtio_mmio_irq_base;
> + } cfg;
> +};
> +
> +void xen_pvh_class_setup_common_props(XenPVHMachineClass *xpc);
> +#endif
> --
> 2.43.0
>
^ permalink raw reply [flat|nested] 19+ messages in thread
* [PATCH v2 08/12] hw/arm: xenpvh: Rename xen_arm.c -> xen-pvh.c
2024-08-20 14:29 [PATCH v2 00/12] xen: pvh: Partial QOM:fication with new x86 PVH machine Edgar E. Iglesias
` (6 preceding siblings ...)
2024-08-20 14:29 ` [PATCH v2 07/12] hw/arm: xenpvh: Break out a common PVH machine Edgar E. Iglesias
@ 2024-08-20 14:29 ` Edgar E. Iglesias
2024-08-20 14:29 ` [PATCH v2 09/12] hw/arm: xenpvh: Reverse virtio-mmio creation order Edgar E. Iglesias
` (3 subsequent siblings)
11 siblings, 0 replies; 19+ messages in thread
From: Edgar E. Iglesias @ 2024-08-20 14:29 UTC (permalink / raw)
To: qemu-devel
Cc: sstabellini, anthony, paul, peter.maydell, alex.bennee,
xenia.ragiadakou, jason.andryuk, edgar.iglesias, xen-devel,
qemu-arm
From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
Rename xen_arm.c -> xen-pvh.c to better express that this
is a PVH machine and to align with x86 HVM and future PVH
machine filenames:
hw/i386/xen/xen-hvm.c
hw/i386/xen/xen-pvh.c (in preparation)
No functional changes.
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
---
hw/arm/meson.build | 2 +-
hw/arm/{xen_arm.c => xen-pvh.c} | 0
2 files changed, 1 insertion(+), 1 deletion(-)
rename hw/arm/{xen_arm.c => xen-pvh.c} (100%)
diff --git a/hw/arm/meson.build b/hw/arm/meson.build
index 074612b40c..4059d0be2e 100644
--- a/hw/arm/meson.build
+++ b/hw/arm/meson.build
@@ -61,7 +61,7 @@ arm_ss.add(when: 'CONFIG_FSL_IMX6UL', if_true: files('fsl-imx6ul.c', 'mcimx6ul-e
arm_ss.add(when: 'CONFIG_NRF51_SOC', if_true: files('nrf51_soc.c'))
arm_ss.add(when: 'CONFIG_XEN', if_true: files(
'xen-stubs.c',
- 'xen_arm.c',
+ 'xen-pvh.c',
))
system_ss.add(when: 'CONFIG_ARM_SMMUV3', if_true: files('smmu-common.c'))
diff --git a/hw/arm/xen_arm.c b/hw/arm/xen-pvh.c
similarity index 100%
rename from hw/arm/xen_arm.c
rename to hw/arm/xen-pvh.c
--
2.43.0
^ permalink raw reply related [flat|nested] 19+ messages in thread* [PATCH v2 09/12] hw/arm: xenpvh: Reverse virtio-mmio creation order
2024-08-20 14:29 [PATCH v2 00/12] xen: pvh: Partial QOM:fication with new x86 PVH machine Edgar E. Iglesias
` (7 preceding siblings ...)
2024-08-20 14:29 ` [PATCH v2 08/12] hw/arm: xenpvh: Rename xen_arm.c -> xen-pvh.c Edgar E. Iglesias
@ 2024-08-20 14:29 ` Edgar E. Iglesias
2024-08-20 14:29 ` [PATCH v2 10/12] hw/xen: pvh-common: Add support for creating PCIe/GPEX Edgar E. Iglesias
` (2 subsequent siblings)
11 siblings, 0 replies; 19+ messages in thread
From: Edgar E. Iglesias @ 2024-08-20 14:29 UTC (permalink / raw)
To: qemu-devel
Cc: sstabellini, anthony, paul, peter.maydell, alex.bennee,
xenia.ragiadakou, jason.andryuk, edgar.iglesias, xen-devel,
Edgar E. Iglesias
From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
We've been creating the virtio-mmio devices in forwards order
but since the qbus lists prepend (rather than append) entries,
the virtio busses end up with decreasing base address order.
Xen enables virtio-mmio nodes in forwards order so there's been
a missmatch. So far, we've been working around this with an
out-of-tree patch to Xen.
This reverses the order making sure the virtio busses end up
ordered with increasing base addresses avoiding the need to
patch Xen.
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Acked-by: Stefano Stabellini <sstabellini@kernel.org>
---
hw/xen/xen-pvh-common.c | 13 ++++++++++++-
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/hw/xen/xen-pvh-common.c b/hw/xen/xen-pvh-common.c
index 880e8143d7..295f920442 100644
--- a/hw/xen/xen-pvh-common.c
+++ b/hw/xen/xen-pvh-common.c
@@ -75,7 +75,18 @@ static void xen_create_virtio_mmio_devices(XenPVHMachineState *s)
{
int i;
- for (i = 0; i < s->cfg.virtio_mmio_num; i++) {
+ /*
+ * We create the transports in reverse order. Since qbus_realize()
+ * prepends (not appends) new child buses, the decrementing loop below will
+ * create a list of virtio-mmio buses with increasing base addresses.
+ *
+ * When a -device option is processed from the command line,
+ * qbus_find_recursive() picks the next free virtio-mmio bus in forwards
+ * order.
+ *
+ * This is what the Xen tools expect.
+ */
+ for (i = s->cfg.virtio_mmio_num - 1; i >= 0; i--) {
hwaddr base = s->cfg.virtio_mmio.base + i * s->cfg.virtio_mmio.size;
qemu_irq irq = qemu_allocate_irq(xen_set_irq, NULL,
s->cfg.virtio_mmio_irq_base + i);
--
2.43.0
^ permalink raw reply related [flat|nested] 19+ messages in thread* [PATCH v2 10/12] hw/xen: pvh-common: Add support for creating PCIe/GPEX
2024-08-20 14:29 [PATCH v2 00/12] xen: pvh: Partial QOM:fication with new x86 PVH machine Edgar E. Iglesias
` (8 preceding siblings ...)
2024-08-20 14:29 ` [PATCH v2 09/12] hw/arm: xenpvh: Reverse virtio-mmio creation order Edgar E. Iglesias
@ 2024-08-20 14:29 ` Edgar E. Iglesias
2024-08-21 1:22 ` Stefano Stabellini
2024-08-20 14:29 ` [PATCH v2 11/12] hw/i386/xen: Add a Xen PVH x86 machine Edgar E. Iglesias
2024-08-20 14:29 ` [PATCH v2 12/12] docs/system/i386: xenpvh: Add a basic description Edgar E. Iglesias
11 siblings, 1 reply; 19+ messages in thread
From: Edgar E. Iglesias @ 2024-08-20 14:29 UTC (permalink / raw)
To: qemu-devel
Cc: sstabellini, anthony, paul, peter.maydell, alex.bennee,
xenia.ragiadakou, jason.andryuk, edgar.iglesias, xen-devel,
Edgar E. Iglesias
From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
Add support for optionally creating a PCIe/GPEX controller.
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
---
hw/xen/xen-pvh-common.c | 76 +++++++++++++++++++++++++++++++++
include/hw/xen/xen-pvh-common.h | 29 +++++++++++++
2 files changed, 105 insertions(+)
diff --git a/hw/xen/xen-pvh-common.c b/hw/xen/xen-pvh-common.c
index 295f920442..28d7168446 100644
--- a/hw/xen/xen-pvh-common.c
+++ b/hw/xen/xen-pvh-common.c
@@ -122,6 +122,64 @@ static void xen_enable_tpm(XenPVHMachineState *s)
}
#endif
+/*
+ * We use the GPEX PCIe controller with its internal INTX PCI interrupt
+ * swizzling. This swizzling is emulated in QEMU and routes all INTX
+ * interrupts from endpoints down to only 4 INTX interrupts.
+ * See include/hw/pci/pci.h : pci_swizzle()
+ */
+static inline void xenpvh_gpex_init(XenPVHMachineState *s,
+ XenPVHMachineClass *xpc,
+ MemoryRegion *sysmem)
+{
+ MemoryRegion *ecam_reg;
+ MemoryRegion *mmio_reg;
+ DeviceState *dev;
+ int i;
+
+ object_initialize_child(OBJECT(s), "gpex", &s->pci.gpex,
+ TYPE_GPEX_HOST);
+ dev = DEVICE(&s->pci.gpex);
+ sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), &error_fatal);
+
+ ecam_reg = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0);
+ memory_region_add_subregion(sysmem, s->cfg.pci_ecam.base, ecam_reg);
+
+ mmio_reg = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 1);
+
+ if (s->cfg.pci_mmio.size) {
+ memory_region_init_alias(&s->pci.mmio_alias, OBJECT(dev), "pcie-mmio",
+ mmio_reg,
+ s->cfg.pci_mmio.base, s->cfg.pci_mmio.size);
+ memory_region_add_subregion(sysmem, s->cfg.pci_mmio.base,
+ &s->pci.mmio_alias);
+ }
+
+ if (s->cfg.pci_mmio_high.size) {
+ memory_region_init_alias(&s->pci.mmio_high_alias, OBJECT(dev),
+ "pcie-mmio-high",
+ mmio_reg, s->cfg.pci_mmio_high.base, s->cfg.pci_mmio_high.size);
+ memory_region_add_subregion(sysmem, s->cfg.pci_mmio_high.base,
+ &s->pci.mmio_high_alias);
+ }
+
+ /*
+ * PVH implementations with PCI enabled must provide set_pci_intx_irq()
+ * and optionally an implementation of set_pci_link_route().
+ */
+ assert(xpc->set_pci_intx_irq);
+
+ for (i = 0; i < GPEX_NUM_IRQS; i++) {
+ qemu_irq irq = qemu_allocate_irq(xpc->set_pci_intx_irq, s, i);
+
+ sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, irq);
+ gpex_set_irq_num(GPEX_HOST(dev), i, s->cfg.pci_intx_irq_base + i);
+ if (xpc->set_pci_link_route) {
+ xpc->set_pci_link_route(i, s->cfg.pci_intx_irq_base + i);
+ }
+ }
+}
+
static void xen_pvh_init(MachineState *ms)
{
XenPVHMachineState *s = XEN_PVH_MACHINE(ms);
@@ -152,6 +210,15 @@ static void xen_pvh_init(MachineState *ms)
}
#endif
+ /* Non-zero pci-ecam-size enables PCI. */
+ if (s->cfg.pci_ecam.size) {
+ if (s->cfg.pci_ecam.size != 256 * MiB) {
+ error_report("pci-ecam-size only supports values 0 or 0x10000000");
+ exit(EXIT_FAILURE);
+ }
+ xenpvh_gpex_init(s, xpc, sysmem);
+ }
+
/* Call the implementation specific init. */
if (xpc->init) {
xpc->init(ms);
@@ -200,6 +267,9 @@ XEN_PVH_PROP_MEMMAP(ram_high)
/* TPM only has a base-addr option. */
XEN_PVH_PROP_MEMMAP_BASE(tpm)
XEN_PVH_PROP_MEMMAP(virtio_mmio)
+XEN_PVH_PROP_MEMMAP(pci_ecam)
+XEN_PVH_PROP_MEMMAP(pci_mmio)
+XEN_PVH_PROP_MEMMAP(pci_mmio_high)
void xen_pvh_class_setup_common_props(XenPVHMachineClass *xpc)
{
@@ -242,6 +312,12 @@ do { \
OC_MEMMAP_PROP(oc, "virtio-mmio", virtio_mmio);
}
+ if (xpc->has_pci) {
+ OC_MEMMAP_PROP(oc, "pci-ecam", pci_ecam);
+ OC_MEMMAP_PROP(oc, "pci-mmio", pci_mmio);
+ OC_MEMMAP_PROP(oc, "pci-mmio-high", pci_mmio_high);
+ }
+
#ifdef CONFIG_TPM
if (xpc->has_tpm) {
object_class_property_add(oc, "tpm-base-addr", "uint64_t",
diff --git a/include/hw/xen/xen-pvh-common.h b/include/hw/xen/xen-pvh-common.h
index 77fd98b9fe..bc09eea936 100644
--- a/include/hw/xen/xen-pvh-common.h
+++ b/include/hw/xen/xen-pvh-common.h
@@ -25,10 +25,29 @@ struct XenPVHMachineClass {
/* PVH implementation specific init. */
void (*init)(MachineState *state);
+ /*
+ * set_pci_intx_irq - Deliver INTX irqs to the guest.
+ *
+ * @opaque: pointer to XenPVHMachineState.
+ * @irq: IRQ after swizzling, between 0-3.
+ * @level: IRQ level.
+ */
+ void (*set_pci_intx_irq)(void *opaque, int irq, int level);
+
+ /*
+ * set_pci_link_route: - optional implementation call to setup
+ * routing between INTX IRQ (0 - 3) and GSI's.
+ *
+ * @line: line the INTx line (0 => A .. 3 => B)
+ * @irq: GSI
+ */
+ int (*set_pci_link_route)(uint8_t line, uint8_t irq);
+
/*
* Each implementation can optionally enable features that it
* supports and are known to work.
*/
+ bool has_pci;
bool has_tpm;
bool has_virtio_mmio;
};
@@ -44,6 +63,12 @@ struct XenPVHMachineState {
MemoryRegion high;
} ram;
+ struct {
+ GPEXHost gpex;
+ MemoryRegion mmio_alias;
+ MemoryRegion mmio_high_alias;
+ } pci;
+
struct {
MemMapEntry ram_low, ram_high;
MemMapEntry tpm;
@@ -52,6 +77,10 @@ struct XenPVHMachineState {
MemMapEntry virtio_mmio;
uint32_t virtio_mmio_num;
uint32_t virtio_mmio_irq_base;
+
+ /* PCI */
+ MemMapEntry pci_ecam, pci_mmio, pci_mmio_high;
+ uint32_t pci_intx_irq_base;
} cfg;
};
--
2.43.0
^ permalink raw reply related [flat|nested] 19+ messages in thread* Re: [PATCH v2 10/12] hw/xen: pvh-common: Add support for creating PCIe/GPEX
2024-08-20 14:29 ` [PATCH v2 10/12] hw/xen: pvh-common: Add support for creating PCIe/GPEX Edgar E. Iglesias
@ 2024-08-21 1:22 ` Stefano Stabellini
0 siblings, 0 replies; 19+ messages in thread
From: Stefano Stabellini @ 2024-08-21 1:22 UTC (permalink / raw)
To: Edgar E. Iglesias
Cc: qemu-devel, sstabellini, anthony, paul, peter.maydell,
alex.bennee, xenia.ragiadakou, jason.andryuk, edgar.iglesias,
xen-devel
On Tue, 20 Aug 2024, Edgar E. Iglesias wrote:
> From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
>
> Add support for optionally creating a PCIe/GPEX controller.
>
> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
> ---
> hw/xen/xen-pvh-common.c | 76 +++++++++++++++++++++++++++++++++
> include/hw/xen/xen-pvh-common.h | 29 +++++++++++++
> 2 files changed, 105 insertions(+)
>
> diff --git a/hw/xen/xen-pvh-common.c b/hw/xen/xen-pvh-common.c
> index 295f920442..28d7168446 100644
> --- a/hw/xen/xen-pvh-common.c
> +++ b/hw/xen/xen-pvh-common.c
> @@ -122,6 +122,64 @@ static void xen_enable_tpm(XenPVHMachineState *s)
> }
> #endif
>
> +/*
> + * We use the GPEX PCIe controller with its internal INTX PCI interrupt
> + * swizzling. This swizzling is emulated in QEMU and routes all INTX
> + * interrupts from endpoints down to only 4 INTX interrupts.
> + * See include/hw/pci/pci.h : pci_swizzle()
> + */
> +static inline void xenpvh_gpex_init(XenPVHMachineState *s,
> + XenPVHMachineClass *xpc,
> + MemoryRegion *sysmem)
> +{
> + MemoryRegion *ecam_reg;
> + MemoryRegion *mmio_reg;
> + DeviceState *dev;
> + int i;
> +
> + object_initialize_child(OBJECT(s), "gpex", &s->pci.gpex,
> + TYPE_GPEX_HOST);
> + dev = DEVICE(&s->pci.gpex);
> + sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), &error_fatal);
> +
> + ecam_reg = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0);
> + memory_region_add_subregion(sysmem, s->cfg.pci_ecam.base, ecam_reg);
> +
> + mmio_reg = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 1);
> +
> + if (s->cfg.pci_mmio.size) {
> + memory_region_init_alias(&s->pci.mmio_alias, OBJECT(dev), "pcie-mmio",
> + mmio_reg,
> + s->cfg.pci_mmio.base, s->cfg.pci_mmio.size);
> + memory_region_add_subregion(sysmem, s->cfg.pci_mmio.base,
> + &s->pci.mmio_alias);
> + }
> +
> + if (s->cfg.pci_mmio_high.size) {
> + memory_region_init_alias(&s->pci.mmio_high_alias, OBJECT(dev),
> + "pcie-mmio-high",
> + mmio_reg, s->cfg.pci_mmio_high.base, s->cfg.pci_mmio_high.size);
> + memory_region_add_subregion(sysmem, s->cfg.pci_mmio_high.base,
> + &s->pci.mmio_high_alias);
> + }
> +
> + /*
> + * PVH implementations with PCI enabled must provide set_pci_intx_irq()
> + * and optionally an implementation of set_pci_link_route().
> + */
> + assert(xpc->set_pci_intx_irq);
> +
> + for (i = 0; i < GPEX_NUM_IRQS; i++) {
> + qemu_irq irq = qemu_allocate_irq(xpc->set_pci_intx_irq, s, i);
> +
> + sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, irq);
> + gpex_set_irq_num(GPEX_HOST(dev), i, s->cfg.pci_intx_irq_base + i);
> + if (xpc->set_pci_link_route) {
> + xpc->set_pci_link_route(i, s->cfg.pci_intx_irq_base + i);
> + }
> + }
> +}
> +
> static void xen_pvh_init(MachineState *ms)
> {
> XenPVHMachineState *s = XEN_PVH_MACHINE(ms);
> @@ -152,6 +210,15 @@ static void xen_pvh_init(MachineState *ms)
> }
> #endif
>
> + /* Non-zero pci-ecam-size enables PCI. */
> + if (s->cfg.pci_ecam.size) {
> + if (s->cfg.pci_ecam.size != 256 * MiB) {
> + error_report("pci-ecam-size only supports values 0 or 0x10000000");
> + exit(EXIT_FAILURE);
> + }
> + xenpvh_gpex_init(s, xpc, sysmem);
> + }
> +
> /* Call the implementation specific init. */
> if (xpc->init) {
> xpc->init(ms);
> @@ -200,6 +267,9 @@ XEN_PVH_PROP_MEMMAP(ram_high)
> /* TPM only has a base-addr option. */
> XEN_PVH_PROP_MEMMAP_BASE(tpm)
> XEN_PVH_PROP_MEMMAP(virtio_mmio)
> +XEN_PVH_PROP_MEMMAP(pci_ecam)
> +XEN_PVH_PROP_MEMMAP(pci_mmio)
> +XEN_PVH_PROP_MEMMAP(pci_mmio_high)
>
> void xen_pvh_class_setup_common_props(XenPVHMachineClass *xpc)
> {
> @@ -242,6 +312,12 @@ do { \
> OC_MEMMAP_PROP(oc, "virtio-mmio", virtio_mmio);
> }
>
> + if (xpc->has_pci) {
> + OC_MEMMAP_PROP(oc, "pci-ecam", pci_ecam);
> + OC_MEMMAP_PROP(oc, "pci-mmio", pci_mmio);
> + OC_MEMMAP_PROP(oc, "pci-mmio-high", pci_mmio_high);
> + }
> +
> #ifdef CONFIG_TPM
> if (xpc->has_tpm) {
> object_class_property_add(oc, "tpm-base-addr", "uint64_t",
> diff --git a/include/hw/xen/xen-pvh-common.h b/include/hw/xen/xen-pvh-common.h
> index 77fd98b9fe..bc09eea936 100644
> --- a/include/hw/xen/xen-pvh-common.h
> +++ b/include/hw/xen/xen-pvh-common.h
> @@ -25,10 +25,29 @@ struct XenPVHMachineClass {
> /* PVH implementation specific init. */
> void (*init)(MachineState *state);
>
> + /*
> + * set_pci_intx_irq - Deliver INTX irqs to the guest.
> + *
> + * @opaque: pointer to XenPVHMachineState.
> + * @irq: IRQ after swizzling, between 0-3.
> + * @level: IRQ level.
> + */
> + void (*set_pci_intx_irq)(void *opaque, int irq, int level);
> +
> + /*
> + * set_pci_link_route: - optional implementation call to setup
> + * routing between INTX IRQ (0 - 3) and GSI's.
> + *
> + * @line: line the INTx line (0 => A .. 3 => B)
> + * @irq: GSI
> + */
> + int (*set_pci_link_route)(uint8_t line, uint8_t irq);
> +
> /*
> * Each implementation can optionally enable features that it
> * supports and are known to work.
> */
> + bool has_pci;
> bool has_tpm;
> bool has_virtio_mmio;
> };
> @@ -44,6 +63,12 @@ struct XenPVHMachineState {
> MemoryRegion high;
> } ram;
>
> + struct {
> + GPEXHost gpex;
> + MemoryRegion mmio_alias;
> + MemoryRegion mmio_high_alias;
> + } pci;
> +
> struct {
> MemMapEntry ram_low, ram_high;
> MemMapEntry tpm;
> @@ -52,6 +77,10 @@ struct XenPVHMachineState {
> MemMapEntry virtio_mmio;
> uint32_t virtio_mmio_num;
> uint32_t virtio_mmio_irq_base;
> +
> + /* PCI */
> + MemMapEntry pci_ecam, pci_mmio, pci_mmio_high;
> + uint32_t pci_intx_irq_base;
> } cfg;
> };
>
> --
> 2.43.0
>
^ permalink raw reply [flat|nested] 19+ messages in thread
* [PATCH v2 11/12] hw/i386/xen: Add a Xen PVH x86 machine
2024-08-20 14:29 [PATCH v2 00/12] xen: pvh: Partial QOM:fication with new x86 PVH machine Edgar E. Iglesias
` (9 preceding siblings ...)
2024-08-20 14:29 ` [PATCH v2 10/12] hw/xen: pvh-common: Add support for creating PCIe/GPEX Edgar E. Iglesias
@ 2024-08-20 14:29 ` Edgar E. Iglesias
2024-08-21 1:22 ` Stefano Stabellini
2024-08-20 14:29 ` [PATCH v2 12/12] docs/system/i386: xenpvh: Add a basic description Edgar E. Iglesias
11 siblings, 1 reply; 19+ messages in thread
From: Edgar E. Iglesias @ 2024-08-20 14:29 UTC (permalink / raw)
To: qemu-devel
Cc: sstabellini, anthony, paul, peter.maydell, alex.bennee,
xenia.ragiadakou, jason.andryuk, edgar.iglesias, xen-devel,
Edgar E. Iglesias, Michael S. Tsirkin, Marcel Apfelbaum,
Paolo Bonzini, Richard Henderson, Eduardo Habkost
From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
Add a Xen PVH x86 machine based on the abstract PVH Machine.
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
---
hw/i386/xen/meson.build | 1 +
hw/i386/xen/xen-pvh.c | 121 ++++++++++++++++++++++++++++++++++++++++
2 files changed, 122 insertions(+)
create mode 100644 hw/i386/xen/xen-pvh.c
diff --git a/hw/i386/xen/meson.build b/hw/i386/xen/meson.build
index 3f0df8bc07..c73c62b8e3 100644
--- a/hw/i386/xen/meson.build
+++ b/hw/i386/xen/meson.build
@@ -4,6 +4,7 @@ i386_ss.add(when: 'CONFIG_XEN', if_true: files(
))
i386_ss.add(when: ['CONFIG_XEN', xen], if_true: files(
'xen-hvm.c',
+ 'xen-pvh.c',
))
i386_ss.add(when: 'CONFIG_XEN_BUS', if_true: files(
diff --git a/hw/i386/xen/xen-pvh.c b/hw/i386/xen/xen-pvh.c
new file mode 100644
index 0000000000..45645667e9
--- /dev/null
+++ b/hw/i386/xen/xen-pvh.c
@@ -0,0 +1,121 @@
+/*
+ * QEMU Xen PVH x86 Machine
+ *
+ * Copyright (c) 2024 Advanced Micro Devices, Inc.
+ * Written by Edgar E. Iglesias <edgar.iglesias@amd.com>
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#include "qemu/osdep.h"
+#include "qemu/error-report.h"
+#include "hw/boards.h"
+#include "sysemu/sysemu.h"
+#include "hw/xen/arch_hvm.h"
+#include <xen/hvm/hvm_info_table.h>
+#include "hw/xen/xen-pvh-common.h"
+
+#define TYPE_XEN_PVH_X86 MACHINE_TYPE_NAME("xenpvh")
+OBJECT_DECLARE_SIMPLE_TYPE(XenPVHx86State, XEN_PVH_X86)
+
+struct XenPVHx86State {
+ /*< private >*/
+ XenPVHMachineState parent;
+
+ DeviceState **cpu;
+};
+
+static DeviceState *xen_pvh_cpu_new(MachineState *ms,
+ int64_t apic_id)
+{
+ Object *cpu = object_new(ms->cpu_type);
+
+ object_property_add_child(OBJECT(ms), "cpu[*]", cpu);
+ object_property_set_uint(cpu, "apic-id", apic_id, &error_fatal);
+ qdev_realize(DEVICE(cpu), NULL, &error_fatal);
+ object_unref(cpu);
+
+ return DEVICE(cpu);
+}
+
+static void xen_pvh_init(MachineState *ms)
+{
+ XenPVHx86State *xp = XEN_PVH_X86(ms);
+ int i;
+
+ /* Create dummy cores. This will indirectly create the APIC MSI window. */
+ xp->cpu = g_malloc(sizeof xp->cpu[0] * ms->smp.max_cpus);
+ for (i = 0; i < ms->smp.max_cpus; i++) {
+ xp->cpu[i] = xen_pvh_cpu_new(ms, i);
+ }
+}
+
+static void xen_pvh_instance_init(Object *obj)
+{
+ XenPVHMachineState *s = XEN_PVH_MACHINE(obj);
+
+ /* Default values. */
+ s->cfg.ram_low = (MemMapEntry) { 0x0, 0x80000000U };
+ s->cfg.ram_high = (MemMapEntry) { 0xC000000000ULL, 0x4000000000ULL };
+ s->cfg.pci_intx_irq_base = 16;
+}
+
+/*
+ * Deliver INTX interrupts to Xen guest.
+ */
+static void xen_pvh_set_pci_intx_irq(void *opaque, int irq, int level)
+{
+ /*
+ * Since QEMU emulates all of the swizziling
+ * We don't want Xen to do any additional swizzling in
+ * xen_set_pci_intx_level() so we always set device to 0.
+ */
+ if (xen_set_pci_intx_level(xen_domid, 0, 0, 0, irq, level)) {
+ error_report("xendevicemodel_set_pci_intx_level failed");
+ }
+}
+
+static void xen_pvh_machine_class_init(ObjectClass *oc, void *data)
+{
+ XenPVHMachineClass *xpc = XEN_PVH_MACHINE_CLASS(oc);
+ MachineClass *mc = MACHINE_CLASS(oc);
+
+ mc->desc = "Xen PVH x86 machine";
+ mc->default_cpu_type = TARGET_DEFAULT_CPU_TYPE;
+
+ /* mc->max_cpus holds the MAX value allowed in the -smp cmd-line opts. */
+ mc->max_cpus = HVM_MAX_VCPUS;
+
+ /* We have an implementation specific init to create CPU objects. */
+ xpc->init = xen_pvh_init;
+
+ /*
+ * PCI INTX routing.
+ *
+ * We describe the mapping between the 4 INTX interrupt and GSIs
+ * using xen_set_pci_link_route(). xen_pvh_set_pci_intx_irq is
+ * used to deliver the interrupt.
+ */
+ xpc->set_pci_intx_irq = xen_pvh_set_pci_intx_irq;
+ xpc->set_pci_link_route = xen_set_pci_link_route;
+
+ /* List of supported features known to work on PVH x86. */
+ xpc->has_pci = true;
+
+ xen_pvh_class_setup_common_props(xpc);
+}
+
+static const TypeInfo xen_pvh_x86_machine_type = {
+ .name = TYPE_XEN_PVH_X86,
+ .parent = TYPE_XEN_PVH_MACHINE,
+ .class_init = xen_pvh_machine_class_init,
+ .instance_init = xen_pvh_instance_init,
+ .instance_size = sizeof(XenPVHx86State),
+};
+
+static void xen_pvh_machine_register_types(void)
+{
+ type_register_static(&xen_pvh_x86_machine_type);
+}
+
+type_init(xen_pvh_machine_register_types)
--
2.43.0
^ permalink raw reply related [flat|nested] 19+ messages in thread* Re: [PATCH v2 11/12] hw/i386/xen: Add a Xen PVH x86 machine
2024-08-20 14:29 ` [PATCH v2 11/12] hw/i386/xen: Add a Xen PVH x86 machine Edgar E. Iglesias
@ 2024-08-21 1:22 ` Stefano Stabellini
0 siblings, 0 replies; 19+ messages in thread
From: Stefano Stabellini @ 2024-08-21 1:22 UTC (permalink / raw)
To: Edgar E. Iglesias
Cc: qemu-devel, sstabellini, anthony, paul, peter.maydell,
alex.bennee, xenia.ragiadakou, jason.andryuk, edgar.iglesias,
xen-devel, Michael S. Tsirkin, Marcel Apfelbaum, Paolo Bonzini,
Richard Henderson, Eduardo Habkost
On Tue, 20 Aug 2024, Edgar E. Iglesias wrote:
> From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
>
> Add a Xen PVH x86 machine based on the abstract PVH Machine.
>
> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
> ---
> hw/i386/xen/meson.build | 1 +
> hw/i386/xen/xen-pvh.c | 121 ++++++++++++++++++++++++++++++++++++++++
> 2 files changed, 122 insertions(+)
> create mode 100644 hw/i386/xen/xen-pvh.c
>
> diff --git a/hw/i386/xen/meson.build b/hw/i386/xen/meson.build
> index 3f0df8bc07..c73c62b8e3 100644
> --- a/hw/i386/xen/meson.build
> +++ b/hw/i386/xen/meson.build
> @@ -4,6 +4,7 @@ i386_ss.add(when: 'CONFIG_XEN', if_true: files(
> ))
> i386_ss.add(when: ['CONFIG_XEN', xen], if_true: files(
> 'xen-hvm.c',
> + 'xen-pvh.c',
> ))
>
> i386_ss.add(when: 'CONFIG_XEN_BUS', if_true: files(
> diff --git a/hw/i386/xen/xen-pvh.c b/hw/i386/xen/xen-pvh.c
> new file mode 100644
> index 0000000000..45645667e9
> --- /dev/null
> +++ b/hw/i386/xen/xen-pvh.c
> @@ -0,0 +1,121 @@
> +/*
> + * QEMU Xen PVH x86 Machine
> + *
> + * Copyright (c) 2024 Advanced Micro Devices, Inc.
> + * Written by Edgar E. Iglesias <edgar.iglesias@amd.com>
> + *
> + * SPDX-License-Identifier: GPL-2.0-or-later
> + */
> +
> +#include "qemu/osdep.h"
> +#include "qemu/error-report.h"
> +#include "hw/boards.h"
> +#include "sysemu/sysemu.h"
> +#include "hw/xen/arch_hvm.h"
> +#include <xen/hvm/hvm_info_table.h>
> +#include "hw/xen/xen-pvh-common.h"
> +
> +#define TYPE_XEN_PVH_X86 MACHINE_TYPE_NAME("xenpvh")
> +OBJECT_DECLARE_SIMPLE_TYPE(XenPVHx86State, XEN_PVH_X86)
> +
> +struct XenPVHx86State {
> + /*< private >*/
> + XenPVHMachineState parent;
> +
> + DeviceState **cpu;
> +};
> +
> +static DeviceState *xen_pvh_cpu_new(MachineState *ms,
> + int64_t apic_id)
> +{
> + Object *cpu = object_new(ms->cpu_type);
> +
> + object_property_add_child(OBJECT(ms), "cpu[*]", cpu);
> + object_property_set_uint(cpu, "apic-id", apic_id, &error_fatal);
> + qdev_realize(DEVICE(cpu), NULL, &error_fatal);
> + object_unref(cpu);
> +
> + return DEVICE(cpu);
> +}
> +
> +static void xen_pvh_init(MachineState *ms)
> +{
> + XenPVHx86State *xp = XEN_PVH_X86(ms);
> + int i;
> +
> + /* Create dummy cores. This will indirectly create the APIC MSI window. */
> + xp->cpu = g_malloc(sizeof xp->cpu[0] * ms->smp.max_cpus);
> + for (i = 0; i < ms->smp.max_cpus; i++) {
> + xp->cpu[i] = xen_pvh_cpu_new(ms, i);
> + }
> +}
> +
> +static void xen_pvh_instance_init(Object *obj)
> +{
> + XenPVHMachineState *s = XEN_PVH_MACHINE(obj);
> +
> + /* Default values. */
> + s->cfg.ram_low = (MemMapEntry) { 0x0, 0x80000000U };
> + s->cfg.ram_high = (MemMapEntry) { 0xC000000000ULL, 0x4000000000ULL };
> + s->cfg.pci_intx_irq_base = 16;
> +}
> +
> +/*
> + * Deliver INTX interrupts to Xen guest.
> + */
> +static void xen_pvh_set_pci_intx_irq(void *opaque, int irq, int level)
> +{
> + /*
> + * Since QEMU emulates all of the swizziling
> + * We don't want Xen to do any additional swizzling in
> + * xen_set_pci_intx_level() so we always set device to 0.
> + */
> + if (xen_set_pci_intx_level(xen_domid, 0, 0, 0, irq, level)) {
> + error_report("xendevicemodel_set_pci_intx_level failed");
> + }
> +}
> +
> +static void xen_pvh_machine_class_init(ObjectClass *oc, void *data)
> +{
> + XenPVHMachineClass *xpc = XEN_PVH_MACHINE_CLASS(oc);
> + MachineClass *mc = MACHINE_CLASS(oc);
> +
> + mc->desc = "Xen PVH x86 machine";
> + mc->default_cpu_type = TARGET_DEFAULT_CPU_TYPE;
> +
> + /* mc->max_cpus holds the MAX value allowed in the -smp cmd-line opts. */
> + mc->max_cpus = HVM_MAX_VCPUS;
> +
> + /* We have an implementation specific init to create CPU objects. */
> + xpc->init = xen_pvh_init;
> +
> + /*
> + * PCI INTX routing.
> + *
> + * We describe the mapping between the 4 INTX interrupt and GSIs
> + * using xen_set_pci_link_route(). xen_pvh_set_pci_intx_irq is
> + * used to deliver the interrupt.
> + */
> + xpc->set_pci_intx_irq = xen_pvh_set_pci_intx_irq;
> + xpc->set_pci_link_route = xen_set_pci_link_route;
> +
> + /* List of supported features known to work on PVH x86. */
> + xpc->has_pci = true;
> +
> + xen_pvh_class_setup_common_props(xpc);
> +}
> +
> +static const TypeInfo xen_pvh_x86_machine_type = {
> + .name = TYPE_XEN_PVH_X86,
> + .parent = TYPE_XEN_PVH_MACHINE,
> + .class_init = xen_pvh_machine_class_init,
> + .instance_init = xen_pvh_instance_init,
> + .instance_size = sizeof(XenPVHx86State),
> +};
> +
> +static void xen_pvh_machine_register_types(void)
> +{
> + type_register_static(&xen_pvh_x86_machine_type);
> +}
> +
> +type_init(xen_pvh_machine_register_types)
> --
> 2.43.0
>
^ permalink raw reply [flat|nested] 19+ messages in thread
* [PATCH v2 12/12] docs/system/i386: xenpvh: Add a basic description
2024-08-20 14:29 [PATCH v2 00/12] xen: pvh: Partial QOM:fication with new x86 PVH machine Edgar E. Iglesias
` (10 preceding siblings ...)
2024-08-20 14:29 ` [PATCH v2 11/12] hw/i386/xen: Add a Xen PVH x86 machine Edgar E. Iglesias
@ 2024-08-20 14:29 ` Edgar E. Iglesias
11 siblings, 0 replies; 19+ messages in thread
From: Edgar E. Iglesias @ 2024-08-20 14:29 UTC (permalink / raw)
To: qemu-devel
Cc: sstabellini, anthony, paul, peter.maydell, alex.bennee,
xenia.ragiadakou, jason.andryuk, edgar.iglesias, xen-devel,
Edgar E. Iglesias, Paolo Bonzini
From: "Edgar E. Iglesias" <edgar.iglesias@amd.com>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
---
MAINTAINERS | 1 +
docs/system/i386/xenpvh.rst | 49 +++++++++++++++++++++++++++++++++++++
docs/system/target-i386.rst | 1 +
3 files changed, 51 insertions(+)
create mode 100644 docs/system/i386/xenpvh.rst
diff --git a/MAINTAINERS b/MAINTAINERS
index c2fb0c2f42..c14ac014e2 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -560,6 +560,7 @@ F: include/sysemu/xen.h
F: include/sysemu/xen-mapcache.h
F: stubs/xen-hw-stub.c
F: docs/system/arm/xenpvh.rst
+F: docs/system/i386/xenpvh.rst
Guest CPU Cores (NVMM)
----------------------
diff --git a/docs/system/i386/xenpvh.rst b/docs/system/i386/xenpvh.rst
new file mode 100644
index 0000000000..354250f073
--- /dev/null
+++ b/docs/system/i386/xenpvh.rst
@@ -0,0 +1,49 @@
+Xen PVH machine (``xenpvh``)
+=========================================
+
+Xen supports a spectrum of types of guests that vary in how they depend
+on HW virtualization features, emulation models and paravirtualization.
+PVH is a mode that uses HW virtualization features (like HVM) but tries
+to avoid emulation models and instead use passthrough or
+paravirtualized devices.
+
+QEMU can be used to provide PV virtio devices on an emulated PCIe controller.
+That is the purpose of this minimal machine.
+
+Supported devices
+-----------------
+
+The x86 Xen PVH QEMU machine provide the following devices:
+
+- RAM
+- GPEX host bridge
+- virtio-pci devices
+
+The idea is to only connect virtio-pci devices but in theory any compatible
+PCI device model will work depending on Xen and guest support.
+
+Running
+-------
+
+The Xen tools will typically construct a command-line and launch QEMU
+for you when needed. But here's an example of what it can look like in
+case you need to construct one manually:
+
+.. code-block:: console
+
+ qemu-system-i386 -xen-domid 3 -no-shutdown \
+ -chardev socket,id=libxl-cmd,path=/var/run/xen/qmp-libxl-3,server=on,wait=off \
+ -mon chardev=libxl-cmd,mode=control \
+ -chardev socket,id=libxenstat-cmd,path=/var/run/xen/qmp-libxenstat-3,server=on,wait=off \
+ -mon chardev=libxenstat-cmd,mode=control \
+ -nodefaults \
+ -no-user-config \
+ -xen-attach -name g0 \
+ -vnc none \
+ -display none \
+ -device virtio-net-pci,id=nic0,netdev=net0,mac=00:16:3e:5c:81:78 \
+ -netdev type=tap,id=net0,ifname=vif3.0-emu,br=xenbr0,script=no,downscript=no \
+ -smp 4,maxcpus=4 \
+ -nographic \
+ -machine xenpvh,ram-low-base=0,ram-low-size=2147483648,ram-high-base=4294967296,ram-high-size=2147483648,pci-ecam-base=824633720832,pci-ecam-size=268435456,pci-mmio-base=4026531840,pci-mmio-size=33554432,pci-mmio-high-base=824902156288,pci-mmio-high-size=68719476736 \
+ -m 4096
diff --git a/docs/system/target-i386.rst b/docs/system/target-i386.rst
index 1b8a1f248a..23e84e3ba7 100644
--- a/docs/system/target-i386.rst
+++ b/docs/system/target-i386.rst
@@ -26,6 +26,7 @@ Architectural features
i386/cpu
i386/hyperv
i386/xen
+ i386/xenpvh
i386/kvm-pv
i386/sgx
i386/amd-memory-encryption
--
2.43.0
^ permalink raw reply related [flat|nested] 19+ messages in thread