* [PATCH 0/3] mailbox: Add Marvell MHU driver
@ 2026-11-19 15:29 Paul Cercueil
2026-11-19 15:29 ` [PATCH 1/3] dt-bindings: mailbox: add Marvell MHU Paul Cercueil
` (2 more replies)
0 siblings, 3 replies; 7+ messages in thread
From: Paul Cercueil @ 2026-11-19 15:29 UTC (permalink / raw)
To: Jassi Brar
Cc: Rob Herring, Krzysztof Kozlowski, Conor Dooley, Sudeep Holla,
Cristian Marussi, Thomas Petazzoni, Sunil Kovvuri Goutham,
linux-kernel, devicetree, arm-scmi, linux-arm-kernel,
Paul Cercueil
Hi Jassi,
This patchset adds support for the mailbox found in the OcteonTX and
OcteonTX2 based SoCs, which is used for SCMI communication.
This is technically a v5, where the v4 was sent in 2023 [1].
Given that years have passed, and that the original authors left Marvell
AFAIK, I figured out it was just better to restart at v1.
The last patch applies to the SCMI driver, it does not have a direct
dependency on the other two and can go through the ARM tree, if you or
the ARM-SCMI maintainers prefer so.
Cheers,
-Paul
[1]: https://lore.kernel.org/all/4397b39c-084b-954b-956a-5168a3c8958c@linaro.org/T/
George Cherian (3):
dt-bindings: mailbox: add Marvell MHU
mailbox: marvell-mhu: add Marvell MHU driver
scmi: driver: Increase timeout for rx response
.../bindings/mailbox/marvell,mhu.yaml | 60 ++++
MAINTAINERS | 7 +
.../firmware/arm_scmi/transports/mailbox.c | 2 +-
drivers/mailbox/Kconfig | 11 +
drivers/mailbox/Makefile | 2 +
drivers/mailbox/marvell_mhu.c | 267 ++++++++++++++++++
6 files changed, 348 insertions(+), 1 deletion(-)
create mode 100644 Documentation/devicetree/bindings/mailbox/marvell,mhu.yaml
create mode 100644 drivers/mailbox/marvell_mhu.c
--
2.47.3
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH 1/3] dt-bindings: mailbox: add Marvell MHU
2026-11-19 15:29 [PATCH 0/3] mailbox: Add Marvell MHU driver Paul Cercueil
@ 2026-11-19 15:29 ` Paul Cercueil
2026-11-19 15:38 ` sashiko-bot
2026-11-19 15:29 ` [PATCH 2/3] mailbox: marvell-mhu: add Marvell MHU driver Paul Cercueil
2026-11-19 15:29 ` [PATCH 3/3] scmi: driver: Increase timeout for rx response Paul Cercueil
2 siblings, 1 reply; 7+ messages in thread
From: Paul Cercueil @ 2026-11-19 15:29 UTC (permalink / raw)
To: Jassi Brar
Cc: Rob Herring, Krzysztof Kozlowski, Conor Dooley, Sudeep Holla,
Cristian Marussi, Thomas Petazzoni, Sunil Kovvuri Goutham,
linux-kernel, devicetree, arm-scmi, linux-arm-kernel,
George Cherian, Sujeet Baranwal, Paul Cercueil
From: George Cherian <gcherian@marvell.com>
Marvell Message Handling Unit is a mailbox controller present in
Marvell OcteonTx and OcteonTX2 SoC family.
Signed-off-by: George Cherian <gcherian@marvell.com>
Co-developed-by: Sujeet Baranwal <sbaranwal@marvell.com>
Signed-off-by: Sujeet Baranwal <sbaranwal@marvell.com>
Signed-off-by: Paul Cercueil <paul.cercueil@bootlin.com>
---
.../bindings/mailbox/marvell,mhu.yaml | 60 +++++++++++++++++++
MAINTAINERS | 6 ++
2 files changed, 66 insertions(+)
create mode 100644 Documentation/devicetree/bindings/mailbox/marvell,mhu.yaml
diff --git a/Documentation/devicetree/bindings/mailbox/marvell,mhu.yaml b/Documentation/devicetree/bindings/mailbox/marvell,mhu.yaml
new file mode 100644
index 000000000000..228607165f85
--- /dev/null
+++ b/Documentation/devicetree/bindings/mailbox/marvell,mhu.yaml
@@ -0,0 +1,60 @@
+# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/mailbox/marvell,mhu.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Marvell Message Handling Unit
+
+maintainers:
+ - Sujeet Baranwal <sbaranwal@marvell.com>
+ - George Cherian <gcherian@marvell.com>
+
+description:
+ The Control-Processors Cluster (CPC) provides Arm-platform specification
+ entities for managing the system. One of the CPC processors is the System
+ Control Processor (SCP). The SCP is responsible, among others, for booting
+ the chip, clock and power initialization, controlling power consumption
+ through DVFS, monitoring temperature sensors and controlling AVS. The SCP,
+ as each XCP, contains mailboxes for software-to-software communications.
+ Mailbox writes cause an interrupt to the local XCP core or to the AP.
+ This driver exposes AP-SCP Message Handling Unit to the system, providing
+ the mailbox communication mechanism to the system, with the intention
+ of plugging into the SCMI framework. It is designed to work with Marvell
+ OcteonTX and OcteonTX2-based platforms.
+ The mailbox has no other usage than SCMI communication. In case of
+ configurations running without SCMI support it should be disabled.
+
+properties:
+ compatible:
+ enum:
+ - marvell,octeontx-mhu
+ - marvell,octeontx2-mhu
+ - marvell,cn10-mhu
+
+ reg:
+ maxItems: 1
+
+ "#mbox-cells":
+ description: Index of the channel
+ const: 1
+
+required:
+ - compatible
+ - reg
+ - "#mbox-cells"
+
+additionalProperties: false
+
+examples:
+ - |
+ ecam {
+ #address-cells = <3>;
+ #size-cells = <2>;
+ mailbox: mailbox@1c,0 {
+ compatible = "marvell,octeontx2-mhu";
+ reg = <0xe000 0x0 0x0 0 0>;
+ #mbox-cells = <1>;
+ };
+ };
+...
diff --git a/MAINTAINERS b/MAINTAINERS
index 65e8a4b5c90b..b592a62a5c56 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -15959,6 +15959,12 @@ R: Vamsi Attunuru <vattunuru@marvell.com>
S: Supported
F: drivers/pci/hotplug/octep_hp.c
+MARVELL MHU MAILBOX DRIVER
+M: Sujeet Baranwal <sbaranwal@marvell.com>
+M: George Cherian <gcherian@marvell.com>
+S: Maintained
+F: Documentation/devicetree/bindings/mailbox/marvell,mhu.yml
+
MATROX FRAMEBUFFER DRIVER
L: linux-fbdev@vger.kernel.org
S: Orphan
--
2.47.3
^ permalink raw reply related [flat|nested] 7+ messages in thread
* [PATCH 2/3] mailbox: marvell-mhu: add Marvell MHU driver
2026-11-19 15:29 [PATCH 0/3] mailbox: Add Marvell MHU driver Paul Cercueil
2026-11-19 15:29 ` [PATCH 1/3] dt-bindings: mailbox: add Marvell MHU Paul Cercueil
@ 2026-11-19 15:29 ` Paul Cercueil
2026-11-19 15:49 ` sashiko-bot
2026-11-19 15:29 ` [PATCH 3/3] scmi: driver: Increase timeout for rx response Paul Cercueil
2 siblings, 1 reply; 7+ messages in thread
From: Paul Cercueil @ 2026-11-19 15:29 UTC (permalink / raw)
To: Jassi Brar
Cc: Rob Herring, Krzysztof Kozlowski, Conor Dooley, Sudeep Holla,
Cristian Marussi, Thomas Petazzoni, Sunil Kovvuri Goutham,
linux-kernel, devicetree, arm-scmi, linux-arm-kernel,
George Cherian, Sujeet Baranwal, Paul Cercueil
From: George Cherian <gcherian@marvell.com>
Add support for Message Handling Units on Marvell OcteonTx and OcteonTx2
platforms. The module uses mailbox controller interface to provide
communication between SCP and AP using SCMI protocol over a single
shared memory communication channel.
Signed-off-by: George Cherian <gcherian@marvell.com>
Co-developed-by: Sujeet Baranwal <sbaranwal@marvell.com>
Signed-off-by: Sujeet Baranwal <sbaranwal@marvell.com>
Signed-off-by: Paul Cercueil <paul.cercueil@bootlin.com>
---
MAINTAINERS | 1 +
drivers/mailbox/Kconfig | 11 ++
drivers/mailbox/Makefile | 2 +
drivers/mailbox/marvell_mhu.c | 267 ++++++++++++++++++++++++++++++++++
4 files changed, 281 insertions(+)
create mode 100644 drivers/mailbox/marvell_mhu.c
diff --git a/MAINTAINERS b/MAINTAINERS
index b592a62a5c56..63042166c0da 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -15964,6 +15964,7 @@ M: Sujeet Baranwal <sbaranwal@marvell.com>
M: George Cherian <gcherian@marvell.com>
S: Maintained
F: Documentation/devicetree/bindings/mailbox/marvell,mhu.yml
+F: drivers/mailbox/marvell_mhu.c
MATROX FRAMEBUFFER DRIVER
L: linux-fbdev@vger.kernel.org
diff --git a/drivers/mailbox/Kconfig b/drivers/mailbox/Kconfig
index ab869769e278..ce3b70749a18 100644
--- a/drivers/mailbox/Kconfig
+++ b/drivers/mailbox/Kconfig
@@ -195,6 +195,17 @@ config MAILBOX_TEST
Test client to help with testing new Controller driver
implementations.
+config MARVELL_MHU
+ tristate "Marvell Message Handling Unit"
+ depends on (ARM64 || COMPILE_TEST) && 64BIT
+ depends on OF
+ help
+ This driver adds support for Marvell MHU controller.
+ The MHU mailbox is used by SCMI for OcteonTX and OcteonTX2
+ to control various aspects of platform operation like cpufreq
+ or sensors. The controller supports single channel to SCP.
+ Say Y here to build this module. If unsure say N.
+
config POLARFIRE_SOC_MAILBOX
tristate "PolarFire SoC (MPFS) Mailbox"
depends on HAS_IOMEM
diff --git a/drivers/mailbox/Makefile b/drivers/mailbox/Makefile
index 45dc59bb3e7f..83b26e7bbd27 100644
--- a/drivers/mailbox/Makefile
+++ b/drivers/mailbox/Makefile
@@ -49,6 +49,8 @@ obj-$(CONFIG_BCM_PDC_MBOX) += bcm-pdc-mailbox.o
obj-$(CONFIG_BCM_FLEXRM_MBOX) += bcm-flexrm-mailbox.o
+obj-$(CONFIG_MARVELL_MHU) += marvell_mhu.o
+
obj-$(CONFIG_POLARFIRE_SOC_MAILBOX) += mailbox-mpfs.o
obj-$(CONFIG_MCHP_SBI_IPC_MBOX) += mailbox-mchp-ipc-sbi.o
diff --git a/drivers/mailbox/marvell_mhu.c b/drivers/mailbox/marvell_mhu.c
new file mode 100644
index 000000000000..a037fe104195
--- /dev/null
+++ b/drivers/mailbox/marvell_mhu.c
@@ -0,0 +1,267 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Marvell Message Handling Unit
+ *
+ * Copyright (c) 2023, 2026 Marvell
+ */
+
+#define DRIVER_NAME "marvell-mhu"
+
+#define pr_fmt(fmt) DRIVER_NAME ": " fmt
+
+#include <linux/pci.h>
+#include <linux/acpi.h>
+#include <linux/mailbox_controller.h>
+
+#define BAR0 0
+#define SCP_INDEX 0x0
+#define DEV_AP0 0x2
+#define SCP_TO_AP_INTERRUPT 2
+#define INT_VECS_MIN 0
+#define INT_VECS_MAX 3
+
+#define XCPX_DEVY_XCP_MBOX_LINT_OFFSET 0x000E1C00
+#define XCP_TO_DEV_XCP_MBOX_LINT(xcp_core, device_id) \
+ (XCPX_DEVY_XCP_MBOX_LINT_OFFSET | \
+ ((uint64_t)(xcp_core) << 36) | \
+ ((uint64_t)(device_id) << 4))
+
+#define AP0_TO_SCP_MBOX_LINT XCP_TO_DEV_XCP_MBOX_LINT(SCP_INDEX, DEV_AP0)
+
+/*
+ * Doorbell-Register: XCP(0..1)_DEV(0..7)_XCP_MBOX
+ * Communication data from devices to XCP. When written, sets
+ * XCP(0..1)_DEV(0..7)_XCP_MBOX.
+ * PS: it doesn't matter what is written into this register,
+ * Attempting to writing 'anything' would cause an interrupt
+ * to the target!
+ */
+
+#define XCPX_DEVY_XCP_MBOX_OFFSET 0x000E1000
+#define XCP_TO_DEV_XCP_MBOX(xcp_core, device_id) \
+ (XCPX_DEVY_XCP_MBOX_OFFSET | \
+ ((uint64_t)(xcp_core) << 36) | \
+ ((uint64_t)(device_id) << 4))
+
+/* AP0-to-SCP doorbell */
+#define AP0_TO_SCP_MBOX XCP_TO_DEV_XCP_MBOX(SCP_INDEX, DEV_AP0)
+
+/* Register offset: Enable interrupt from SCP to AP */
+#define XCP0_XCP_DEV0_MBOX_RINT_ENA_W1S 0x000D1C40
+#define XCP0_XCP_DEV1_MBOX_RINT_ENA_W1S 0x000D1C50
+#define XCP0_XCP_DEV2_MBOX_RINT_ENA_W1S 0x000D1C60
+#define XCP0_XCP_DEV3_MBOX_RINT_ENA_W1S 0x000D1C70
+
+/* Rx interrupt from SCP to Non-secure AP (linux kernel) */
+#define XCPX_XCP_DEVY_MBOX_RINT_OFFSET 0x000D1C00
+#define XCPX_XCP_DEVY_MBOX_RINT(xcp_core, device_id) \
+ (XCPX_XCP_DEVY_MBOX_RINT_OFFSET | \
+ ((uint64_t)(xcp_core) << 36) | \
+ ((uint64_t)(device_id) << 4))
+
+/* The interrupt status register */
+#define SCP_TO_AP0_MBOX_RINT XCPX_XCP_DEVY_MBOX_RINT(SCP_INDEX, DEV_AP0)
+
+#define XCPX_XCP_DEVY_MBOX_RINT_OFFSET 0x000D1C00
+#define XCPX_XCP_DEVY_MBOX_RINT(xcp_core, device_id) \
+ (XCPX_XCP_DEVY_MBOX_RINT_OFFSET | \
+ ((uint64_t)(xcp_core) << 36) | \
+ ((uint64_t)(device_id) << 4))
+
+#define SCP_TO_AP0_MBOX_RINT XCPX_XCP_DEVY_MBOX_RINT(SCP_INDEX, DEV_AP0)
+#define SCP_TO_DEV0 XCPX_XCP_DEVY_MBOX_RINT(0, 0)
+#define SCP_TO_DEV1 XCPX_XCP_DEVY_MBOX_RINT(0, 1)
+#define SCP_TO_DEV2 XCPX_XCP_DEVY_MBOX_RINT(0, 2)
+#define SCP_TO_DEV3 XCPX_XCP_DEVY_MBOX_RINT(0, 3)
+
+struct mrvl_mhu {
+ /* SCP link information */
+ void __iomem *base; /* tx_reg, rx_reg */
+
+ struct mbox_controller ctrl;
+ struct mbox_chan chan;
+};
+
+static irqreturn_t mrvl_mhu_rx_interrupt(int irq, void *p)
+{
+ struct mrvl_mhu *mhu = p;
+ u64 val;
+
+ /* Read interrupt status register */
+ val = readq_relaxed(mhu->base + SCP_TO_AP0_MBOX_RINT);
+ if (val) {
+ /* Clear the interrupt : Write on clear */
+ writeq_relaxed(1ul, mhu->base + SCP_TO_AP0_MBOX_RINT);
+ /* Poke mbox */
+ mbox_chan_received_data(&mhu->chan, (void *)&val);
+ return IRQ_HANDLED;
+ }
+
+ return IRQ_NONE;
+}
+
+static int mrvl_mhu_send_data(struct mbox_chan *chan, void *data)
+{
+ const u64 doorbell = 0xff; /* Arbitrary value, SCP requires a value */
+ struct mrvl_mhu *mhu = chan->con_priv;
+
+ /* signal SCP */
+ iowrite64(doorbell, mhu->base + AP0_TO_SCP_MBOX);
+
+ return 0;
+}
+
+static bool mrvl_mhu_last_tx_done(struct mbox_chan *chan)
+{
+ struct mrvl_mhu *mhu = chan->con_priv;
+ u64 status;
+
+ status = ioread64(mhu->base + XCPX_XCP_DEVY_MBOX_RINT(0, 2));
+
+ return status != 0;
+}
+
+static int mrvl_mhu_startup(struct mbox_chan *chan)
+{
+ struct mrvl_mhu *mhu = chan->con_priv;
+
+ /* Enable interrupts only if there is client for data */
+ writeq_relaxed(1ul, mhu->base + XCP0_XCP_DEV2_MBOX_RINT_ENA_W1S);
+
+ return 0;
+}
+
+/* Define operations related to mbox */
+static const struct mbox_chan_ops mrvl_mhu_chan_ops = {
+ .startup = mrvl_mhu_startup,
+ .send_data = mrvl_mhu_send_data,
+ .last_tx_done = mrvl_mhu_last_tx_done,
+};
+
+static int mrvl_mhu_setup_irq(struct pci_dev *pdev)
+{
+ struct device *dev;
+ struct mrvl_mhu *mhu;
+ int irq, ret, nvec;
+
+ mhu = pci_get_drvdata(pdev);
+ dev = &pdev->dev;
+
+ nvec = pci_alloc_irq_vectors(pdev, INT_VECS_MIN, INT_VECS_MAX, PCI_IRQ_MSIX);
+ if (nvec < 0)
+ return nvec;
+
+ irq = pci_irq_vector(pdev, SCP_TO_AP_INTERRUPT);
+ if (irq < 0) {
+ ret = irq;
+ goto irq_err;
+ }
+
+ ret = devm_request_irq(dev, irq, mrvl_mhu_rx_interrupt,
+ 0, "mvl-mhu", mhu);
+ if (ret)
+ goto irq_err;
+
+ return 0;
+
+irq_err:
+ pci_free_irq_vectors(pdev);
+
+ return ret;
+}
+
+static const struct of_device_id mrvl_mhu_dt_ids[] = {
+ { .compatible = "marvell,octeontx-mhu" },
+ { .compatible = "marvell,octeontx2-mhu" },
+ { .compatible = "marvell,cn10-mhu" },
+ { },
+};
+MODULE_DEVICE_TABLE(of, mrvl_mhu_dt_ids);
+
+#define PCI_DEVICE_ID_CAVIUM_OCTEON_MHU 0xa067
+static const struct pci_device_id mrvl_mhu_ids[] = {
+ { PCI_VDEVICE(CAVIUM, PCI_DEVICE_ID_CAVIUM_OCTEON_MHU) },
+ { 0 },
+};
+MODULE_DEVICE_TABLE(pci, mrvl_mhu_ids);
+
+static int mrvl_mhu_probe(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+ struct device *dev = &pdev->dev;
+ struct mrvl_mhu *mhu;
+ int ret;
+
+ /* Only support PCI devices with a corresponding OF node. */
+ if (!dev->of_node)
+ return -ENODEV;
+ if (!of_match_node(mrvl_mhu_dt_ids, dev->of_node))
+ return -EINVAL;
+
+ mhu = devm_kzalloc(dev, sizeof(*mhu), GFP_KERNEL);
+ if (!mhu)
+ return -ENOMEM;
+
+ mhu->chan = (struct mbox_chan) {
+ .con_priv = mhu,
+ };
+
+ mhu->ctrl = (struct mbox_controller){
+ .chans = &mhu->chan,
+ .num_chans = 1,
+ .txdone_poll = true,
+ .txpoll_period = 100,
+ .ops = &mrvl_mhu_chan_ops,
+ .dev = dev,
+ };
+
+ pci_set_drvdata(pdev, mhu);
+ ret = pcim_enable_device(pdev);
+ if (ret)
+ return ret;
+
+ ret = pcim_request_region(pdev, BAR0, DRIVER_NAME);
+ if (ret)
+ return ret;
+
+ mhu->base = pcim_iomap(pdev, BAR0, pci_resource_len(pdev, BAR0));
+ if (!mhu->base)
+ return -EINVAL;
+
+ /* Configure IRQ from SCP to AP */
+ ret = mrvl_mhu_setup_irq(pdev);
+ if (ret)
+ return ret;
+
+ ret = mbox_controller_register(&mhu->ctrl);
+ if (ret) {
+ pci_free_irq_vectors(pdev);
+ return ret;
+ }
+
+ return 0;
+}
+
+static void mrvl_mhu_remove(struct pci_dev *pdev)
+{
+ struct mrvl_mhu *mhu = pci_get_drvdata(pdev);
+
+ mbox_controller_unregister(&mhu->ctrl);
+ pci_free_irq_vectors(pdev);
+}
+
+static struct pci_driver mrvl_mhu_driver = {
+ .name = DRIVER_NAME,
+ .id_table = mrvl_mhu_ids,
+ .probe = mrvl_mhu_probe,
+ .remove = mrvl_mhu_remove,
+ .driver = {
+ .name = DRIVER_NAME,
+ .of_match_table = mrvl_mhu_dt_ids,
+ },
+};
+module_pci_driver(mrvl_mhu_driver);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Marvell Message Handling Unit");
+MODULE_AUTHOR("Sujeet Baranwal <sbaranwal@marvell.com>");
+MODULE_AUTHOR("George Cherian <gcherian@marvell.com>");
--
2.47.3
^ permalink raw reply related [flat|nested] 7+ messages in thread
* [PATCH 3/3] scmi: driver: Increase timeout for rx response
2026-11-19 15:29 [PATCH 0/3] mailbox: Add Marvell MHU driver Paul Cercueil
2026-11-19 15:29 ` [PATCH 1/3] dt-bindings: mailbox: add Marvell MHU Paul Cercueil
2026-11-19 15:29 ` [PATCH 2/3] mailbox: marvell-mhu: add Marvell MHU driver Paul Cercueil
@ 2026-11-19 15:29 ` Paul Cercueil
2026-11-19 15:38 ` sashiko-bot
2 siblings, 1 reply; 7+ messages in thread
From: Paul Cercueil @ 2026-11-19 15:29 UTC (permalink / raw)
To: Jassi Brar
Cc: Rob Herring, Krzysztof Kozlowski, Conor Dooley, Sudeep Holla,
Cristian Marussi, Thomas Petazzoni, Sunil Kovvuri Goutham,
linux-kernel, devicetree, arm-scmi, linux-arm-kernel,
George Cherian, Sujeet Baranwal, Paul Cercueil
From: George Cherian <gcherian@marvell.com>
The timeout is way off on the Marvell CN10K SoC, where the firmware
processor can take more than 200ms to answer.
Signed-off-by: George Cherian <gcherian@marvell.com>
Co-developed-by: Sujeet Baranwal <sbaranwal@marvell.com>
Signed-off-by: Sujeet Baranwal <sbaranwal@marvell.com>
Signed-off-by: Paul Cercueil <paul.cercueil@bootlin.com>
---
drivers/firmware/arm_scmi/transports/mailbox.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/firmware/arm_scmi/transports/mailbox.c b/drivers/firmware/arm_scmi/transports/mailbox.c
index 308736c3ead9..78c38d2644aa 100644
--- a/drivers/firmware/arm_scmi/transports/mailbox.c
+++ b/drivers/firmware/arm_scmi/transports/mailbox.c
@@ -380,7 +380,7 @@ static const struct scmi_transport_ops scmi_mailbox_ops = {
static struct scmi_desc scmi_mailbox_desc = {
.ops = &scmi_mailbox_ops,
- .max_rx_timeout_ms = 30, /* We may increase this if required */
+ .max_rx_timeout_ms = 500, /* We may increase this if required */
.max_msg = 20, /* Limited by MBOX_TX_QUEUE_LEN */
.max_msg_size = SCMI_SHMEM_MAX_PAYLOAD_SIZE,
};
--
2.47.3
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH 3/3] scmi: driver: Increase timeout for rx response
2026-11-19 15:29 ` [PATCH 3/3] scmi: driver: Increase timeout for rx response Paul Cercueil
@ 2026-11-19 15:38 ` sashiko-bot
0 siblings, 0 replies; 7+ messages in thread
From: sashiko-bot @ 2026-11-19 15:38 UTC (permalink / raw)
To: Paul Cercueil; +Cc: devicetree, robh, conor+dt
Thank you for your contribution! Sashiko AI review found 1 potential
issue to consider:
- [Medium] scmi: global max_rx_timeout_ms increase instead of per-SoC DT
property
--
commit a587b3cdc57b99607cd614c8d305614227ccea74
Author: George Cherian <gcherian@marvell.com>
scmi: driver: Increase timeout for rx response
This commit increases the default max_rx_timeout_ms in scmi_mailbox_desc
from 30ms to 500ms to accommodate the Marvell CN10K firmware response
time.
> diff --git a/drivers/firmware/arm_scmi/transports/mailbox.c b/drivers/firmware/arm_scmi/transports/mailbox.c
> index 308736c3ead9c..78c38d2644aa0 100644
> --- a/drivers/firmware/arm_scmi/transports/mailbox.c
> +++ b/drivers/firmware/arm_scmi/transports/mailbox.c
> @@ -380,7 +380,7 @@ static const struct scmi_transport_ops scmi_mailbox_ops = {
>
> static struct scmi_desc scmi_mailbox_desc = {
> .ops = &scmi_mailbox_ops,
> - .max_rx_timeout_ms = 30, /* We may increase this if required */
> + .max_rx_timeout_ms = 500, /* We may increase this if required */
> .max_msg = 20, /* Limited by MBOX_TX_QUEUE_LEN */
> .max_msg_size = SCMI_SHMEM_MAX_PAYLOAD_SIZE,
> };
[Severity: Medium]
Should the 500ms timeout for Marvell CN10K be configured via the
"arm,max-rx-timeout-ms" DeviceTree property instead of raising
scmi_mailbox_desc.max_rx_timeout_ms for all mailbox platforms?
scmi_transport_setup() already parses "arm,max-rx-timeout-ms" to override
trans->desc.max_rx_timeout_ms per platform:
drivers/firmware/arm_scmi/driver.c:scmi_transport_setup() {
...
ret = of_property_read_u32(dev->of_node, "arm,max-rx-timeout-ms",
&trans->desc.max_rx_timeout_ms);
...
}
Raising the default in scmi_mailbox_desc affects every mailbox platform,
increasing the worst-case busy-wait in shmem_tx_prepare() (2 *
cinfo->rx_timeout_ms, called from mailbox_send_message() -> msg_submit()
under spin_lock_irqsave(&chan->lock)) from 60ms to 1000ms with interrupts
disabled when a channel is busy, as well as the worst-case atomic busy-spin
in scmi_wait_for_reply() from 30ms to 500ms.
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261119152944.276763-1-paul.cercueil@bootlin.com?part=3
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH 1/3] dt-bindings: mailbox: add Marvell MHU
2026-11-19 15:29 ` [PATCH 1/3] dt-bindings: mailbox: add Marvell MHU Paul Cercueil
@ 2026-11-19 15:38 ` sashiko-bot
0 siblings, 0 replies; 7+ messages in thread
From: sashiko-bot @ 2026-11-19 15:38 UTC (permalink / raw)
To: Paul Cercueil; +Cc: devicetree, robh, conor+dt
Thank you for your contribution! Sashiko AI review found 2 potential
issues to consider:
- [Low] dt-bindings: mailbox: marvell,mhu: broken .yml path and order in
MAINTAINERS
- [Low] dt-bindings: mailbox: marvell,mhu: indentation and style issues
in DTS example
--
commit 4caa249d5ea1d3d9950edc33feb82a1fda934ec0
Author: George Cherian <gcherian@marvell.com>
dt-bindings: mailbox: add Marvell MHU
This commit adds the device tree binding schema and MAINTAINERS entry for
the Marvell Message Handling Unit mailbox controller on OcteonTX and
OcteonTX2 SoCs.
> diff --git a/Documentation/devicetree/bindings/mailbox/marvell,mhu.yaml b/Documentation/devicetree/bindings/mailbox/marvell,mhu.yaml
> new file mode 100644
> index 0000000000000..228607165f850
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/mailbox/marvell,mhu.yaml
> @@ -0,0 +1,60 @@
[ ... ]
> +examples:
> + - |
> + ecam {
> + #address-cells = <3>;
> + #size-cells = <2>;
> + mailbox: mailbox@1c,0 {
> + compatible = "marvell,octeontx2-mhu";
> + reg = <0xe000 0x0 0x0 0 0>;
> + #mbox-cells = <1>;
> + };
> + };
[Severity: Low]
In this DTS example, the child node mailbox@1c,0 is indented by 8 spaces
relative to ecam instead of 4 spaces, is missing a blank line after
#size-cells = <2>;, and defines an unused mailbox: label.
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 65e8a4b5c90b1..b592a62a5c563 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -15959,6 +15959,12 @@ R: Vamsi Attunuru <vattunuru@marvell.com>
> S: Supported
> F: drivers/pci/hotplug/octep_hp.c
>
> +MARVELL MHU MAILBOX DRIVER
> +M: Sujeet Baranwal <sbaranwal@marvell.com>
> +M: George Cherian <gcherian@marvell.com>
> +S: Maintained
> +F: Documentation/devicetree/bindings/mailbox/marvell,mhu.yml
> +
> MATROX FRAMEBUFFER DRIVER
> L: linux-fbdev@vger.kernel.org
> S: Orphan
[Severity: Low]
Should this F: entry end in .yaml instead of .yml to match the file created
by this commit? Because the path does not exist, scripts/get_maintainer.pl
will not match the binding file and --self-test=patterns reports a broken
file pattern. Also, MARVELL MHU MAILBOX DRIVER is placed after MARVELL
OCTEON HOTPLUG DRIVER rather than in alphabetical order between MARVELL
MACCHIATOBIN SUPPORT and MARVELL MV643XX ETHERNET DRIVER.
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261119152944.276763-1-paul.cercueil@bootlin.com?part=1
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH 2/3] mailbox: marvell-mhu: add Marvell MHU driver
2026-11-19 15:29 ` [PATCH 2/3] mailbox: marvell-mhu: add Marvell MHU driver Paul Cercueil
@ 2026-11-19 15:49 ` sashiko-bot
0 siblings, 0 replies; 7+ messages in thread
From: sashiko-bot @ 2026-11-19 15:49 UTC (permalink / raw)
To: Paul Cercueil; +Cc: devicetree, robh, conor+dt
Thank you for your contribution! Sashiko AI review found 5 potential
issues to consider:
- [High] mailbox: marvell-mhu: manual pci_free_irq_vectors() conflicts
with devres
- [High] mailbox: marvell-mhu: missing shutdown op causes NULL deref in
IRQ
- [Medium] mailbox: marvell-mhu: missing PCI dependency in Kconfig
- [Medium] mailbox: marvell-mhu: wrong register and polarity in
mrvl_mhu_last_tx_done()
- [Low] mailbox: marvell-mhu: duplicate and unused register macro
definitions
--
commit d88d0044b75f722104999e16b9dd2c9f50a997d0
Author: George Cherian <gcherian@marvell.com>
mailbox: marvell-mhu: add Marvell MHU driver
This commit adds a PCI mailbox controller driver for the Marvell Message
Handling Unit on OcteonTX and OcteonTX2 platforms.
> diff --git a/drivers/mailbox/Kconfig b/drivers/mailbox/Kconfig
> index ab869769e2789..ce3b70749a182 100644
> --- a/drivers/mailbox/Kconfig
> +++ b/drivers/mailbox/Kconfig
> @@ -195,6 +195,17 @@ config MAILBOX_TEST
> Test client to help with testing new Controller driver
> implementations.
>
> +config MARVELL_MHU
> + tristate "Marvell Message Handling Unit"
> + depends on (ARM64 || COMPILE_TEST) && 64BIT
> + depends on OF
> + help
[Severity: Medium]
Should MARVELL_MHU also depend on PCI? Without CONFIG_PCI, building this
driver fails because module_pci_driver() is only defined under CONFIG_PCI
and pcim_request_region() and pcim_iomap() have no !CONFIG_PCI stubs.
[ ... ]
> diff --git a/drivers/mailbox/marvell_mhu.c b/drivers/mailbox/marvell_mhu.c
> new file mode 100644
> index 0000000000000..a037fe104195b
> --- /dev/null
> +++ b/drivers/mailbox/marvell_mhu.c
[ ... ]
> +/* Register offset: Enable interrupt from SCP to AP */
> +#define XCP0_XCP_DEV0_MBOX_RINT_ENA_W1S 0x000D1C40
> +#define XCP0_XCP_DEV1_MBOX_RINT_ENA_W1S 0x000D1C50
> +#define XCP0_XCP_DEV2_MBOX_RINT_ENA_W1S 0x000D1C60
> +#define XCP0_XCP_DEV3_MBOX_RINT_ENA_W1S 0x000D1C70
> +
> +/* Rx interrupt from SCP to Non-secure AP (linux kernel) */
> +#define XCPX_XCP_DEVY_MBOX_RINT_OFFSET 0x000D1C00
> +#define XCPX_XCP_DEVY_MBOX_RINT(xcp_core, device_id) \
> + (XCPX_XCP_DEVY_MBOX_RINT_OFFSET | \
> + ((uint64_t)(xcp_core) << 36) | \
> + ((uint64_t)(device_id) << 4))
> +
> +/* The interrupt status register */
> +#define SCP_TO_AP0_MBOX_RINT XCPX_XCP_DEVY_MBOX_RINT(SCP_INDEX, DEV_AP0)
> +
> +#define XCPX_XCP_DEVY_MBOX_RINT_OFFSET 0x000D1C00
> +#define XCPX_XCP_DEVY_MBOX_RINT(xcp_core, device_id) \
> + (XCPX_XCP_DEVY_MBOX_RINT_OFFSET | \
> + ((uint64_t)(xcp_core) << 36) | \
> + ((uint64_t)(device_id) << 4))
> +
> +#define SCP_TO_AP0_MBOX_RINT XCPX_XCP_DEVY_MBOX_RINT(SCP_INDEX, DEV_AP0)
> +#define SCP_TO_DEV0 XCPX_XCP_DEVY_MBOX_RINT(0, 0)
> +#define SCP_TO_DEV1 XCPX_XCP_DEVY_MBOX_RINT(0, 1)
> +#define SCP_TO_DEV2 XCPX_XCP_DEVY_MBOX_RINT(0, 2)
> +#define SCP_TO_DEV3 XCPX_XCP_DEVY_MBOX_RINT(0, 3)
[Severity: Low]
This isn't a bug, but XCPX_XCP_DEVY_MBOX_RINT_OFFSET,
XCPX_XCP_DEVY_MBOX_RINT, and SCP_TO_AP0_MBOX_RINT are defined twice here,
and XCP0_XCP_DEV0/1/3_MBOX_RINT_ENA_W1S and SCP_TO_DEV0..3 are unused.
> +
> +struct mrvl_mhu {
> + /* SCP link information */
> + void __iomem *base; /* tx_reg, rx_reg */
> +
> + struct mbox_controller ctrl;
> + struct mbox_chan chan;
> +};
> +
> +static irqreturn_t mrvl_mhu_rx_interrupt(int irq, void *p)
> +{
> + struct mrvl_mhu *mhu = p;
> + u64 val;
> +
> + /* Read interrupt status register */
> + val = readq_relaxed(mhu->base + SCP_TO_AP0_MBOX_RINT);
> + if (val) {
> + /* Clear the interrupt : Write on clear */
> + writeq_relaxed(1ul, mhu->base + SCP_TO_AP0_MBOX_RINT);
> + /* Poke mbox */
> + mbox_chan_received_data(&mhu->chan, (void *)&val);
> + return IRQ_HANDLED;
> + }
> +
> + return IRQ_NONE;
> +}
> +
> +static int mrvl_mhu_send_data(struct mbox_chan *chan, void *data)
> +{
> + const u64 doorbell = 0xff; /* Arbitrary value, SCP requires a value */
> + struct mrvl_mhu *mhu = chan->con_priv;
> +
> + /* signal SCP */
> + iowrite64(doorbell, mhu->base + AP0_TO_SCP_MBOX);
> +
> + return 0;
> +}
> +
> +static bool mrvl_mhu_last_tx_done(struct mbox_chan *chan)
> +{
> + struct mrvl_mhu *mhu = chan->con_priv;
> + u64 status;
> +
> + status = ioread64(mhu->base + XCPX_XCP_DEVY_MBOX_RINT(0, 2));
> +
> + return status != 0;
> +}
[Severity: Medium]
Does mrvl_mhu_last_tx_done() check the wrong register and polarity?
XCPX_XCP_DEVY_MBOX_RINT(0, 2) is SCP_TO_AP0_MBOX_RINT, the SCP-to-AP0 RX
interrupt status register, which mrvl_mhu_rx_interrupt() immediately clears
to 0 in hardirq context when an RX interrupt fires. Consequently, any
client relying on the controller's txdone_poll = true will read 0 and time
out. Meanwhile, the AP0-to-SCP TX interrupt status macro
AP0_TO_SCP_MBOX_LINT defined earlier in this file is unused, and its status
bit is cleared to 0 when the SCP consumes the doorbell (status == 0).
> +
> +static int mrvl_mhu_startup(struct mbox_chan *chan)
> +{
> + struct mrvl_mhu *mhu = chan->con_priv;
> +
> + /* Enable interrupts only if there is client for data */
> + writeq_relaxed(1ul, mhu->base + XCP0_XCP_DEV2_MBOX_RINT_ENA_W1S);
> +
> + return 0;
> +}
> +
> +/* Define operations related to mbox */
> +static const struct mbox_chan_ops mrvl_mhu_chan_ops = {
> + .startup = mrvl_mhu_startup,
> + .send_data = mrvl_mhu_send_data,
> + .last_tx_done = mrvl_mhu_last_tx_done,
> +};
[Severity: High]
Can an SCP interrupt after mbox_free_channel() cause a NULL pointer
dereference in mrvl_mhu_rx_interrupt() -> mbox_chan_received_data()?
mrvl_mhu_startup() enables the hardware interrupt via
XCP0_XCP_DEV2_MBOX_RINT_ENA_W1S, but mrvl_mhu_chan_ops provides no
.shutdown callback to disable it when a client releases the channel (or if
the interrupt was already left enabled before a client binds). Once
mbox_free_channel() sets chan->cl = NULL, an incoming interrupt calls
mbox_chan_received_data(&mhu->chan, ...), which unconditionally
dereferences chan->cl->rx_callback. In addition, mrvl_mhu_rx_interrupt()
clears SCP_TO_AP0_MBOX_RINT before mbox_chan_received_data() finishes
processing the shared memory payload.
[ ... ]
> + pci_set_drvdata(pdev, mhu);
> + ret = pcim_enable_device(pdev);
> + if (ret)
> + return ret;
> +
> + ret = pcim_request_region(pdev, BAR0, DRIVER_NAME);
> + if (ret)
> + return ret;
> +
> + mhu->base = pcim_iomap(pdev, BAR0, pci_resource_len(pdev, BAR0));
> + if (!mhu->base)
> + return -EINVAL;
> +
> + /* Configure IRQ from SCP to AP */
> + ret = mrvl_mhu_setup_irq(pdev);
> + if (ret)
> + return ret;
> +
> + ret = mbox_controller_register(&mhu->ctrl);
> + if (ret) {
> + pci_free_irq_vectors(pdev);
> + return ret;
> + }
> +
> + return 0;
> +}
> +
> +static void mrvl_mhu_remove(struct pci_dev *pdev)
> +{
> + struct mrvl_mhu *mhu = pci_get_drvdata(pdev);
> +
> + mbox_controller_unregister(&mhu->ctrl);
> + pci_free_irq_vectors(pdev);
> +}
[Severity: High]
Does calling pci_free_irq_vectors() here tear down the MSI-X IRQ descriptor
while the devm_request_irq() handler is still registered?
Because mrvl_mhu_probe() calls pcim_enable_device(), pci_alloc_irq_vectors()
is already managed by devres (via pcim_msi_release()). Calling
pci_free_irq_vectors() manually in mrvl_mhu_remove() and on the
mbox_controller_register() failure path frees the IRQ descriptors before
devres runs devm_irq_release(), which triggers a warning when removing the
populated procfs IRQ directory, causes free_irq() to fail to find the
descriptor and leak struct irqaction, and then runs pci_free_irq_vectors()
a second time during devres teardown.
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261119152944.276763-1-paul.cercueil@bootlin.com?part=2
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2026-11-19 15:49 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-19 15:29 [PATCH 0/3] mailbox: Add Marvell MHU driver Paul Cercueil
2026-11-19 15:29 ` [PATCH 1/3] dt-bindings: mailbox: add Marvell MHU Paul Cercueil
2026-11-19 15:38 ` sashiko-bot
2026-11-19 15:29 ` [PATCH 2/3] mailbox: marvell-mhu: add Marvell MHU driver Paul Cercueil
2026-11-19 15:49 ` sashiko-bot
2026-11-19 15:29 ` [PATCH 3/3] scmi: driver: Increase timeout for rx response Paul Cercueil
2026-11-19 15:38 ` sashiko-bot
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox