From: Fabiano Rosas <farosas@suse.de>
To: "Philippe Mathieu-Daudé" <philmd@linaro.org>, qemu-devel@nongnu.org
Cc: "Paolo Bonzini" <pbonzini@redhat.com>,
"Tyrone Ting" <kfting@nuvoton.com>,
"Alex Bennée" <alex.bennee@linaro.org>,
"Manos Pitsidianakis" <manos.pitsidianakis@linaro.org>,
"Eduardo Habkost" <eduardo@habkost.net>,
"Joel Stanley" <joel@jms.id.au>,
"Alistair Francis" <alistair@alistair23.me>,
"Anton Johansson" <anjo@rev.ng>,
"Andrey Smirnov" <andrew.smirnov@gmail.com>,
"Peter Maydell" <peter.maydell@linaro.org>,
"Hao Wu" <wuhaotsh@google.com>, "Cédric Le Goater" <clg@kaod.org>,
"Jean-Christophe Dubois" <jcd@tribudubois.net>,
"Igor Mitsyanko" <i.mitsyanko@gmail.com>,
"Edgar E. Iglesias" <edgar.iglesias@gmail.com>,
"Andrew Jeffery" <andrew@codeconstruct.com.au>,
"Rob Herring" <robh@kernel.org>,
qemu-arm@nongnu.org,
"Mark Cave-Ayland" <mark.cave-ayland@ilande.co.uk>,
"Philippe Mathieu-Daudé" <philmd@linaro.org>
Subject: Re: [PATCH 14/33] hw/cpu/arm: Handle 'has_el2/3' properties once in MPCore parent
Date: Fri, 12 Jan 2024 18:33:13 -0300 [thread overview]
Message-ID: <87y1cudyp2.fsf@suse.de> (raw)
In-Reply-To: <20231212162935.42910-15-philmd@linaro.org>
Philippe Mathieu-Daudé <philmd@linaro.org> writes:
> Move the 'has_el2' and 'has_el3' properties to the abstract
> QOM parent.
>
> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
> ---
> include/hw/cpu/cortex_mpcore.h | 5 +++++
> hw/arm/exynos4210.c | 10 ++++++++--
> hw/arm/vexpress.c | 6 ++++++
> hw/arm/xilinx_zynq.c | 6 ++++++
> hw/cpu/a15mpcore.c | 18 ++++++------------
> hw/cpu/a9mpcore.c | 5 +----
> hw/cpu/cortex_mpcore.c | 3 +++
> 7 files changed, 35 insertions(+), 18 deletions(-)
>
> diff --git a/include/hw/cpu/cortex_mpcore.h b/include/hw/cpu/cortex_mpcore.h
> index 0e7cca9e93..387552468c 100644
> --- a/include/hw/cpu/cortex_mpcore.h
> +++ b/include/hw/cpu/cortex_mpcore.h
> @@ -30,6 +30,8 @@
> * QEMU interface:
> * + QOM property "num-cores" which set the number of cores present in
> * the cluster.
> + * + QOM properties "cpu-has-el3", "cpu-has-el2" which set whether the CPUs
> + * have the exception level features present.
> */
> #define TYPE_CORTEX_MPCORE_PRIV "cortex_mpcore_priv"
> OBJECT_DECLARE_TYPE(CortexMPPrivState, CortexMPPrivClass, CORTEX_MPCORE_PRIV)
> @@ -53,6 +55,9 @@ struct CortexMPPrivState {
>
> /* Properties */
> uint32_t num_cores;
> +
> + bool cpu_has_el3;
> + bool cpu_has_el2;
> };
>
> #define TYPE_A9MPCORE_PRIV "a9mpcore_priv"
> diff --git a/hw/arm/exynos4210.c b/hw/arm/exynos4210.c
> index ea1364499d..7386a8fe57 100644
> --- a/hw/arm/exynos4210.c
> +++ b/hw/arm/exynos4210.c
> @@ -548,7 +548,7 @@ static void exynos4210_realize(DeviceState *socdev, Error **errp)
> Exynos4210State *s = EXYNOS4210_SOC(socdev);
> MemoryRegion *system_mem = get_system_memory();
> SysBusDevice *busdev;
> - DeviceState *dev, *uart[4], *pl330[3];
> + DeviceState *dev, *mpdev, *uart[4], *pl330[3];
> int i, n;
>
> for (n = 0; n < EXYNOS4210_NCPUS; n++) {
> @@ -582,7 +582,13 @@ static void exynos4210_realize(DeviceState *socdev, Error **errp)
> }
>
> /* Private memory region and Internal GIC */
> - qdev_prop_set_uint32(DEVICE(&s->a9mpcore), "num-cores", EXYNOS4210_NCPUS);
> + mpdev = DEVICE(&s->a9mpcore);
> + qdev_prop_set_uint32(mpdev, "num-cores", EXYNOS4210_NCPUS);
> + /*
> + * By default A9 CPUs have EL3 enabled. This board does not currently
> + * support EL3 so the CPU EL3 property is disabled before realization.
> + */
> + qdev_prop_set_bit(mpdev, "cpu-has-el3", false);
> busdev = SYS_BUS_DEVICE(&s->a9mpcore);
> sysbus_realize(busdev, &error_fatal);
> sysbus_mmio_map(busdev, 0, EXYNOS4210_SMP_PRIVATE_BASE_ADDR);
> diff --git a/hw/arm/vexpress.c b/hw/arm/vexpress.c
> index a320d1c181..294b6f15f2 100644
> --- a/hw/arm/vexpress.c
> +++ b/hw/arm/vexpress.c
> @@ -239,6 +239,12 @@ static void init_cpus(MachineState *ms, const char *cpu_type,
> * wires itself up to the CPU's generic_timer gpio out lines.
> */
> dev = qdev_new(privdev);
> + if (!secure) {
> + qdev_prop_set_bit(dev, "cpu-has-el3", false);
> + }
> + if (!virt) {
> + qdev_prop_set_bit(dev, "cpu-has-el2", false);
> + }
> qdev_prop_set_uint32(dev, "num-cpu", smp_cpus);
> busdev = SYS_BUS_DEVICE(dev);
> sysbus_realize_and_unref(busdev, &error_fatal);
> diff --git a/hw/arm/xilinx_zynq.c b/hw/arm/xilinx_zynq.c
> index fc9c927d09..28430dcfba 100644
> --- a/hw/arm/xilinx_zynq.c
> +++ b/hw/arm/xilinx_zynq.c
> @@ -240,6 +240,12 @@ static void zynq_init(MachineState *machine)
>
> dev = qdev_new(TYPE_A9MPCORE_PRIV);
> qdev_prop_set_uint32(dev, "num-cpu", 1);
> + /*
> + * By default A9 CPUs have EL3 enabled. This board does not
> + * currently support EL3 so the CPU EL3 property is disabled before
> + * realization.
> + */
> + qdev_prop_set_bit(dev, "cpu-has-el3", false);
> busdev = SYS_BUS_DEVICE(dev);
> sysbus_realize_and_unref(busdev, &error_fatal);
> sysbus_mmio_map(busdev, 0, MPCORE_PERIPHBASE);
> diff --git a/hw/cpu/a15mpcore.c b/hw/cpu/a15mpcore.c
> index 16874426e1..a16544fdde 100644
> --- a/hw/cpu/a15mpcore.c
> +++ b/hw/cpu/a15mpcore.c
> @@ -52,9 +52,6 @@ static void a15mp_priv_realize(DeviceState *dev, Error **errp)
> SysBusDevice *gicsbd;
> Error *local_err = NULL;
> int i;
> - bool has_el3;
> - bool has_el2 = false;
> - Object *cpuobj;
>
> cc->parent_realize(dev, &local_err);
> if (local_err) {
> @@ -70,14 +67,11 @@ static void a15mp_priv_realize(DeviceState *dev, Error **errp)
> /* Make the GIC's TZ support match the CPUs. We assume that
> * either all the CPUs have TZ, or none do.
> */
> - cpuobj = OBJECT(qemu_get_cpu(0));
> - has_el3 = object_property_find(cpuobj, "has_el3") &&
> - object_property_get_bool(cpuobj, "has_el3", &error_abort);
> - qdev_prop_set_bit(gicdev, "has-security-extensions", has_el3);
> + qdev_prop_set_bit(gicdev, "has-security-extensions",
> + c->cpu_has_el3);
> /* Similarly for virtualization support */
> - has_el2 = object_property_find(cpuobj, "has_el2") &&
> - object_property_get_bool(cpuobj, "has_el2", &error_abort);
> - qdev_prop_set_bit(gicdev, "has-virtualization-extensions", has_el2);
> + qdev_prop_set_bit(gicdev, "has-virtualization-extensions",
> + c->cpu_has_el2);
> }
>
> if (!sysbus_realize(SYS_BUS_DEVICE(&s->gic), errp)) {
> @@ -112,7 +106,7 @@ static void a15mp_priv_realize(DeviceState *dev, Error **errp)
> qdev_get_gpio_in(gicdev,
> ppibase + timer_irq[irq]));
> }
> - if (has_el2) {
> + if (c->cpu_has_el2) {
> /* Connect the GIC maintenance interrupt to PPI ID 25 */
> sysbus_connect_irq(SYS_BUS_DEVICE(gicdev), i + 4 * c->num_cores,
> qdev_get_gpio_in(gicdev, ppibase + 25));
> @@ -134,7 +128,7 @@ static void a15mp_priv_realize(DeviceState *dev, Error **errp)
> sysbus_mmio_get_region(gicsbd, 0));
> memory_region_add_subregion(&c->container, 0x2000,
> sysbus_mmio_get_region(gicsbd, 1));
> - if (has_el2) {
> + if (c->cpu_has_el2) {
> memory_region_add_subregion(&c->container, 0x4000,
> sysbus_mmio_get_region(gicsbd, 2));
> memory_region_add_subregion(&c->container, 0x6000,
> diff --git a/hw/cpu/a9mpcore.c b/hw/cpu/a9mpcore.c
> index 9c138f4442..54949314f8 100644
> --- a/hw/cpu/a9mpcore.c
> +++ b/hw/cpu/a9mpcore.c
> @@ -51,7 +51,6 @@ static void a9mp_priv_realize(DeviceState *dev, Error **errp)
> SysBusDevice *scubusdev, *gicbusdev, *gtimerbusdev, *mptimerbusdev,
> *wdtbusdev;
> Error *local_err = NULL;
> - bool has_el3;
> CPUState *cpu0;
> Object *cpuobj;
>
> @@ -86,9 +85,7 @@ static void a9mp_priv_realize(DeviceState *dev, Error **errp)
> /* Make the GIC's TZ support match the CPUs. We assume that
> * either all the CPUs have TZ, or none do.
> */
> - has_el3 = object_property_find(cpuobj, "has_el3") &&
> - object_property_get_bool(cpuobj, "has_el3", &error_abort);
> - qdev_prop_set_bit(gicdev, "has-security-extensions", has_el3);
> + qdev_prop_set_bit(gicdev, "has-security-extensions", c->cpu_has_el3);
>
> if (!sysbus_realize(SYS_BUS_DEVICE(&s->gic), errp)) {
> return;
> diff --git a/hw/cpu/cortex_mpcore.c b/hw/cpu/cortex_mpcore.c
> index d7ea633e4e..549b81f708 100644
> --- a/hw/cpu/cortex_mpcore.c
> +++ b/hw/cpu/cortex_mpcore.c
> @@ -27,6 +27,9 @@ static Property cortex_mpcore_priv_properties[] = {
> DEFINE_PROP_UINT32("num-cores", CortexMPPrivState, num_cores, 1),
> DEFINE_PROP_UINT32("num-cpu", CortexMPPrivState, num_cores, 1), /* alias */
>
> + DEFINE_PROP_BOOL("cpu-has-el3", CortexMPPrivState, cpu_has_el3, true),
> + DEFINE_PROP_BOOL("cpu-has-el2", CortexMPPrivState, cpu_has_el2, false),
Are we missing setting cpu_has_el2 somewhere else? This^ results in fewer
cpregs being registered and is what breaks migration.
You can test with:
$ (echo "migrate file:migfile"; echo "quit") | ./qemu-system-arm -M ast2600-evb -monitor stdio
$ ./scripts/analyze-migration.py -f migfile | grep array_len
Before series:
$ ./scripts/analyze-migration.py -f migfile | grep array_len
"cpreg_vmstate_array_len": "0x0000010a",
"cpreg_vmstate_array_len": "0x0000010a",
After series:
$ ./scripts/analyze-migration.py -f migfile | grep array_len
"cpreg_vmstate_array_len": "0x000000df",
"cpreg_vmstate_array_len": "0x000000df",
next prev parent reply other threads:[~2024-01-12 21:34 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-12-12 16:29 [PATCH 00/33] hw/cpu/arm: Remove one use of qemu_get_cpu() in A7/A15 MPCore priv Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 01/33] hw/arm/boot: Propagate vCPU to arm_load_dtb() Philippe Mathieu-Daudé
2024-01-02 13:51 ` Cédric Le Goater
2023-12-12 16:29 ` [PATCH 02/33] hw/arm/fsl-imx6: Add a local 'gic' variable Philippe Mathieu-Daudé
2024-01-02 13:52 ` Cédric Le Goater
2023-12-12 16:29 ` [PATCH 03/33] hw/arm/fsl-imx6ul: " Philippe Mathieu-Daudé
2024-01-02 13:52 ` Cédric Le Goater
2023-12-12 16:29 ` [PATCH 04/33] hw/arm/fsl-imx7: " Philippe Mathieu-Daudé
2024-01-02 13:53 ` Cédric Le Goater
2023-12-12 16:29 ` [PATCH 05/33] hw/cpu: Remove dead Kconfig Philippe Mathieu-Daudé
2024-01-02 13:53 ` Cédric Le Goater
2023-12-12 16:29 ` [PATCH 06/33] hw/cpu/arm: Rename 'busdev' -> 'gicsbd' in a15mp_priv_realize() Philippe Mathieu-Daudé
2024-01-02 13:54 ` Cédric Le Goater
2023-12-12 16:29 ` [PATCH 07/33] hw/cpu/arm: Alias 'num-cpu' property on TYPE_REALVIEW_MPCORE Philippe Mathieu-Daudé
2024-01-02 13:57 ` Cédric Le Goater
2023-12-12 16:29 ` [PATCH 08/33] hw/cpu/arm: Declare CPU QOM types using DEFINE_TYPES() macro Philippe Mathieu-Daudé
2024-01-02 13:57 ` Cédric Le Goater
2023-12-12 16:29 ` [PATCH 09/33] hw/cpu/arm: Merge {a9mpcore.h, a15mpcore.h} as cortex_mpcore.h Philippe Mathieu-Daudé
2024-01-02 14:00 ` Cédric Le Goater
2023-12-12 16:29 ` [PATCH 10/33] hw/cpu/arm: Introduce abstract CORTEX_MPCORE_PRIV QOM type Philippe Mathieu-Daudé
2024-01-02 14:23 ` Cédric Le Goater
2023-12-12 16:29 ` [PATCH 11/33] hw/cpu/arm: Have A9MPCORE/A15MPCORE inheritate common CORTEX_MPCORE_PRIV Philippe Mathieu-Daudé
2024-01-02 14:23 ` Cédric Le Goater
2023-12-12 16:29 ` [PATCH 12/33] hw/cpu/arm: Create MPCore container in QOM parent Philippe Mathieu-Daudé
2024-01-02 14:23 ` Cédric Le Goater
2023-12-12 16:29 ` [PATCH 13/33] hw/cpu/arm: Handle 'num_cores' property once in MPCore parent Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 14/33] hw/cpu/arm: Handle 'has_el2/3' properties " Philippe Mathieu-Daudé
2024-01-12 21:33 ` Fabiano Rosas [this message]
2024-01-16 16:25 ` Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 15/33] hw/cpu/arm: Handle 'gic-irq' property " Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 16/33] hw/cpu/arm: Handle GIC " Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 17/33] hw/cpu/arm: Document more properties of CORTEX_MPCORE_PRIV QOM type Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 18/33] hw/cpu/arm: Replace A15MPPrivState by CortexMPPrivState Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 19/33] hw/cpu/arm: Introduce TYPE_A7MPCORE_PRIV for Cortex-A7 MPCore Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 20/33] hw/cpu/arm: Consolidate check on max GIC spi supported Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 21/33] hw/cpu/arm: Create CPUs once in MPCore parent Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 22/33] hw/arm/aspeed_ast2600: Let the A7MPcore create/wire the CPU cores Philippe Mathieu-Daudé
2024-01-02 0:11 ` Andrew Jeffery
2023-12-12 16:29 ` [PATCH 23/33] hw/arm/exynos4210: Let the A9MPcore " Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 24/33] hw/arm/fsl-imx6: " Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 25/33] hw/arm/fsl-imx6ul: Let the A7MPcore " Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 26/33] hw/arm/fsl-imx7: " Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 27/33] hw/arm/highbank: Let the A9/A15MPcore " Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 28/33] hw/arm/vexpress: " Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 29/33] hw/arm/xilinx_zynq: Let the A9MPcore " Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 30/33] hw/arm/npcm7xx: " Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 31/33] hw/cpu/a9mpcore: Remove legacy code Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 32/33] hw/cpu/arm: Remove 'num-cpu' property alias Philippe Mathieu-Daudé
2023-12-12 16:29 ` [PATCH 33/33] hw/cpu/arm: Remove use of qemu_get_cpu() in A7/A15 realize() Philippe Mathieu-Daudé
2023-12-26 11:17 ` [PATCH 00/33] hw/cpu/arm: Remove one use of qemu_get_cpu() in A7/A15 MPCore priv Philippe Mathieu-Daudé
2024-01-02 14:55 ` Cédric Le Goater
2024-01-02 16:15 ` Philippe Mathieu-Daudé
2024-01-02 16:41 ` Cédric Le Goater
2024-01-03 9:19 ` Philippe Mathieu-Daudé
2024-01-03 19:53 ` Fabiano Rosas
2024-01-09 15:02 ` Cédric Le Goater
2024-01-09 17:40 ` Fabiano Rosas
2024-01-09 18:06 ` Cédric Le Goater
2024-01-09 20:21 ` Fabiano Rosas
2024-01-09 21:22 ` Philippe Mathieu-Daudé
2024-01-10 3:36 ` Peter Xu
2024-01-10 6:03 ` Markus Armbruster
2024-01-10 6:26 ` Peter Xu
2024-01-10 8:09 ` Markus Armbruster
2024-01-10 8:44 ` Peter Xu
2024-01-12 9:03 ` Cédric Le Goater
2024-01-10 13:19 ` Fabiano Rosas
2024-01-10 13:54 ` Markus Armbruster
2024-01-12 10:26 ` Cédric Le Goater
2024-01-12 19:54 ` Fabiano Rosas
2024-01-15 9:04 ` Cédric Le Goater
2024-01-12 8:41 ` Cédric Le Goater
2024-01-09 21:07 ` Philippe Mathieu-Daudé
2024-01-09 21:09 ` Philippe Mathieu-Daudé
2024-01-12 8:00 ` Cédric Le Goater
2024-01-12 7:29 ` Cédric Le Goater
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87y1cudyp2.fsf@suse.de \
--to=farosas@suse.de \
--cc=alex.bennee@linaro.org \
--cc=alistair@alistair23.me \
--cc=andrew.smirnov@gmail.com \
--cc=andrew@codeconstruct.com.au \
--cc=anjo@rev.ng \
--cc=clg@kaod.org \
--cc=edgar.iglesias@gmail.com \
--cc=eduardo@habkost.net \
--cc=i.mitsyanko@gmail.com \
--cc=jcd@tribudubois.net \
--cc=joel@jms.id.au \
--cc=kfting@nuvoton.com \
--cc=manos.pitsidianakis@linaro.org \
--cc=mark.cave-ayland@ilande.co.uk \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=philmd@linaro.org \
--cc=qemu-arm@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=robh@kernel.org \
--cc=wuhaotsh@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).