[PATCH v3 0/5] remoteproc: imx_rproc: Support i.MX95

[PATCH v3 0/5] remoteproc: imx_rproc: Support i.MX95

[Peng Fan (OSS)]

This patchset is pick up a previous patchset [1] with rebased on
next-20251030, and some changes applied.
 - Add runtime ops to separate cpu ops and lmm ops
 - added more comments
 - moved some check imx_sm_rproc_detect_mode() from imx_rproc_sm_prepare().

No changes to the dt-binding patch, so R-b/A-b are kept.
More info could be found in commit message of each patch and below.

[1]https://lore.kernel.org/linux-remoteproc/20250821-imx95-rproc-1-v5-0-e93191dfac51@nxp.com/

i.MX95 features a Cortex-M33 core, six Cortex-A55 cores, and
one Cortex-M7 core. The System Control Management Interface(SCMI)
firmware runs on the M33 core. The i.MX95 SCMI firmware named System
Manager(SM) includes vendor extension protocols, Logical Machine
Management(LMM) protocol and CPU protocol and etc.

There are three cases for M7:
(1) M7 in a separate Logical Machine(LM) that Linux couldn't control it.
(2) M7 in a separate Logical Machine that Linux could control it using
    LMM protocol
(3) M7 runs in same Logical Machine as A55, so Linux could control it
    using CPU protocol

In patch 3, Use LMM and CPU protocol to manage M7. More info could be
found in the patch commit log
    Current setup relies on pre-Linux software(U-Boot) to do
M7 TCM ECC initialization. In future, we could add the support in Linux

Patchset was tested with below boot images when the patchset based on next-20251030:
imx-boot-variant-rpmsg-imx95-19x19-lpddr5-evk-sd.bin-flash_lpboot_sm_a55 (Use LMM protocol)
imx-boot-variant-alt-imx95-19x19-lpddr5-evk-sd.bin-flash_alt (Use CPU protocol)
imx-boot-imx95-19x19-lpddr5-evk-sd.bin-flash_all (M7 not under A55 control)

Also tested i.MX8MP/8ULP-EVK.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
---
Changes in v3:
- Add R-b for patch 2,3,5
- Patch 4:
  Rename imx_rproc_ops_sm to imx_rproc_ops_sm_lmm.
  Add more comments in code
  Simplify if check in imx_rproc_sm_lmm_prepare()
  Update commit log
  Use dev_err for failure path
- Link to v2: https://lore.kernel.org/linux-remoteproc/20251031-imx95-rproc-2025-10-31-v2-0-758b2e547a55@nxp.com/T/#t

Changes in v2:
- Patch 2: Change to const pointer for ops, update commit log and add R-b from Frank
- Patch 3: New patch, introduce prepare ops
- Patch 4: Rename imx_rproc_sm_prepare to imx_rproc_sm_lmm_prepare
           drop IMX_RPROC_FLAGS_SM_CPU_OP and IMX_RPROC_FLAGS_SM_LMM_OP
           Add a bool for dcfg->lmid and info->lmid check.
           Provide separate plat ops for lmm and cpu setup.
           Move LM permission check to a separate function.
- Patch 5: Drop macro IMX95_M7_CPUID and IMX95_M7_LMID
- Link to v1: https://lore.kernel.org/r/20251028-imx95-rproc-2025-10-28-v1-0-ce9e7db9edcb@nxp.com

---
Peng Fan (5):
      dt-bindings: remoteproc: fsl,imx-rproc: Add support for i.MX95
      remoteproc: imx_rproc: Add runtime ops copy to support dynamic behavior
      remoteproc: imx_rproc: Introduce prepare ops for imx_rproc_dcfg
      remoteproc: imx_rproc: Add support for System Manager API
      remoteproc: imx_rproc: Add support for i.MX95

 .../bindings/remoteproc/fsl,imx-rproc.yaml         |   1 +
 drivers/remoteproc/Kconfig                         |   2 +
 drivers/remoteproc/imx_rproc.c                     | 245 +++++++++++++++++++--
 drivers/remoteproc/imx_rproc.h                     |   4 +
 4 files changed, 238 insertions(+), 14 deletions(-)
---
base-commit: 88cbd8ac379cf5ce68b7efcfd4d1484a6871ee0b
change-id: 20251120-imx95-rproc-2025-11-20-3b9369670d6a

Best regards,
-- 
Peng Fan <peng.fan@nxp.com>

[PATCH v3 1/5] dt-bindings: remoteproc: fsl,imx-rproc: Add support for i.MX95

[Peng Fan (OSS)]

From: Peng Fan <peng.fan@nxp.com>

Add compatible string for the Cortex-M7 core in i.MX95

Reviewed-by: Frank Li <Frank.Li@nxp.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
---
 Documentation/devicetree/bindings/remoteproc/fsl,imx-rproc.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/remoteproc/fsl,imx-rproc.yaml b/Documentation/devicetree/bindings/remoteproc/fsl,imx-rproc.yaml
index 57d75acb0b5e52ca49d1361176fdebc18a0bf7a2..ce8ec0119469c8fc0979a192b6e3d3a03108d7d2 100644
--- a/Documentation/devicetree/bindings/remoteproc/fsl,imx-rproc.yaml
+++ b/Documentation/devicetree/bindings/remoteproc/fsl,imx-rproc.yaml
@@ -28,6 +28,7 @@ properties:
       - fsl,imx8qxp-cm4
       - fsl,imx8ulp-cm33
       - fsl,imx93-cm33
+      - fsl,imx95-cm7
 
   clocks:
     maxItems: 1

-- 
2.37.1

[PATCH v3 2/5] remoteproc: imx_rproc: Add runtime ops copy to support dynamic behavior

[Peng Fan (OSS)]

From: Peng Fan <peng.fan@nxp.com>

Structure imx_rproc_dcfg contains a const pointer to imx_rproc_plat_ops,
which defines the start/stop/detect_mode operations for a remote processor.
To preserve the const correctness of the static configuration while
allowing runtime modification of ops behavior, introduce a new
imx_rproc_plat_ops member in struct imx_rproc named `ops`.

During initialization, dcfg->ops is assigned to priv->ops.
Enable the driver to safely override ops at runtime without affecting the
original const configuration.

Improve flexibility for platforms that require dynamic operation switching
(e.g. i.MX95 Logical Machine ops and CPU ops).

Reviewed-by: Frank Li <Frank.Li@nxp.com>
Reviewed-by: Daniel Baluta <daniel.baluta@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
---
 drivers/remoteproc/imx_rproc.c | 27 +++++++++++++--------------
 1 file changed, 13 insertions(+), 14 deletions(-)

diff --git a/drivers/remoteproc/imx_rproc.c b/drivers/remoteproc/imx_rproc.c
index 02e155e967942d745de4ccd96f9008e4211f9b36..82cb8c7866ccb79061c46068d2039f38feebbb3c 100644
--- a/drivers/remoteproc/imx_rproc.c
+++ b/drivers/remoteproc/imx_rproc.c
@@ -116,6 +116,7 @@ struct imx_rproc {
 	u32				entry;		/* cpu start address */
 	u32				core_index;
 	struct dev_pm_domain_list	*pd_list;
+	const struct imx_rproc_plat_ops	*ops;
 };
 
 static const struct imx_rproc_att imx_rproc_att_imx93[] = {
@@ -315,7 +316,6 @@ static int imx_rproc_scu_api_start(struct rproc *rproc)
 static int imx_rproc_start(struct rproc *rproc)
 {
 	struct imx_rproc *priv = rproc->priv;
-	const struct imx_rproc_dcfg *dcfg = priv->dcfg;
 	struct device *dev = priv->dev;
 	int ret;
 
@@ -323,10 +323,10 @@ static int imx_rproc_start(struct rproc *rproc)
 	if (ret)
 		return ret;
 
-	if (!dcfg->ops || !dcfg->ops->start)
+	if (!priv->ops || !priv->ops->start)
 		return -EOPNOTSUPP;
 
-	ret = dcfg->ops->start(rproc);
+	ret = priv->ops->start(rproc);
 	if (ret)
 		dev_err(dev, "Failed to enable remote core!\n");
 
@@ -372,14 +372,13 @@ static int imx_rproc_scu_api_stop(struct rproc *rproc)
 static int imx_rproc_stop(struct rproc *rproc)
 {
 	struct imx_rproc *priv = rproc->priv;
-	const struct imx_rproc_dcfg *dcfg = priv->dcfg;
 	struct device *dev = priv->dev;
 	int ret;
 
-	if (!dcfg->ops || !dcfg->ops->stop)
+	if (!priv->ops || !priv->ops->stop)
 		return -EOPNOTSUPP;
 
-	ret = dcfg->ops->stop(rproc);
+	ret = priv->ops->stop(rproc);
 	if (ret)
 		dev_err(dev, "Failed to stop remote core\n");
 	else
@@ -590,12 +589,11 @@ static int imx_rproc_scu_api_detach(struct rproc *rproc)
 static int imx_rproc_detach(struct rproc *rproc)
 {
 	struct imx_rproc *priv = rproc->priv;
-	const struct imx_rproc_dcfg *dcfg = priv->dcfg;
 
-	if (!dcfg->ops || !dcfg->ops->detach)
+	if (!priv->ops || !priv->ops->detach)
 		return -EOPNOTSUPP;
 
-	return dcfg->ops->detach(rproc);
+	return priv->ops->detach(rproc);
 }
 
 static struct resource_table *imx_rproc_get_loaded_rsc_table(struct rproc *rproc, size_t *table_sz)
@@ -995,18 +993,16 @@ static int imx_rproc_scu_api_detect_mode(struct rproc *rproc)
 
 static int imx_rproc_detect_mode(struct imx_rproc *priv)
 {
-	const struct imx_rproc_dcfg *dcfg = priv->dcfg;
-
 	/*
 	 * To i.MX{7,8} ULP, Linux is under control of RTOS, no need
-	 * dcfg->ops or dcfg->ops->detect_mode, it is state RPROC_DETACHED.
+	 * priv->ops or priv->ops->detect_mode, it is state RPROC_DETACHED.
 	 */
-	if (!dcfg->ops || !dcfg->ops->detect_mode) {
+	if (!priv->ops || !priv->ops->detect_mode) {
 		priv->rproc->state = RPROC_DETACHED;
 		return 0;
 	}
 
-	return dcfg->ops->detect_mode(priv->rproc);
+	return priv->ops->detect_mode(priv->rproc);
 }
 
 static int imx_rproc_sys_off_handler(struct sys_off_data *data)
@@ -1056,6 +1052,9 @@ static int imx_rproc_probe(struct platform_device *pdev)
 	priv->dcfg = dcfg;
 	priv->dev = dev;
 
+	if (dcfg->ops)
+		priv->ops = dcfg->ops;
+
 	dev_set_drvdata(dev, rproc);
 	priv->workqueue = create_workqueue(dev_name(dev));
 	if (!priv->workqueue) {

-- 
2.37.1

[PATCH v3 3/5] remoteproc: imx_rproc: Introduce prepare ops for imx_rproc_dcfg

[Peng Fan (OSS)]

From: Peng Fan <peng.fan@nxp.com>

Allow each platform to provide its own prepare operations, preparing
for i.MX95 LMM and CPU ops support.

No functional changes.

Reviewed-by: Daniel Baluta <daniel.baluta@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
---
 drivers/remoteproc/imx_rproc.c | 5 ++++-
 drivers/remoteproc/imx_rproc.h | 1 +
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/remoteproc/imx_rproc.c b/drivers/remoteproc/imx_rproc.c
index 82cb8c7866ccb79061c46068d2039f38feebbb3c..232eb91e0b5dc2432131b1c140d6688b073fea1d 100644
--- a/drivers/remoteproc/imx_rproc.c
+++ b/drivers/remoteproc/imx_rproc.c
@@ -532,7 +532,10 @@ static int imx_rproc_prepare(struct rproc *rproc)
 		rproc_add_carveout(rproc, mem);
 	}
 
-	return  0;
+	if (priv->ops && priv->ops->prepare)
+		return priv->ops->prepare(rproc);
+
+	return 0;
 }
 
 static int imx_rproc_parse_fw(struct rproc *rproc, const struct firmware *fw)
diff --git a/drivers/remoteproc/imx_rproc.h b/drivers/remoteproc/imx_rproc.h
index 1b2d9f4d6d19dcdc215be97f7e2ab3488281916a..37417568a0ade2ae4d6a4e3d0f139ea52b185254 100644
--- a/drivers/remoteproc/imx_rproc.h
+++ b/drivers/remoteproc/imx_rproc.h
@@ -24,6 +24,7 @@ struct imx_rproc_plat_ops {
 	int (*stop)(struct rproc *rproc);
 	int (*detach)(struct rproc *rproc);
 	int (*detect_mode)(struct rproc *rproc);
+	int (*prepare)(struct rproc *rproc);
 };
 
 struct imx_rproc_dcfg {

-- 
2.37.1

[PATCH v3 4/5] remoteproc: imx_rproc: Add support for System Manager API

[Peng Fan (OSS)]

From: Peng Fan <peng.fan@nxp.com>

i.MX95 features a Cortex-M33 core, six Cortex-A55 cores, and
one Cortex-M7 core. The System Control Management Interface(SCMI)
firmware runs on the M33 core. The i.MX95 SCMI firmware named System
Manager(SM) includes vendor extension protocols, Logical Machine
Management(LMM) protocol and CPU protocol and etc.

Depending on SM configuration, M7 can be used as follows:
 (1) M7 in a separate Logical Machine (LM) from A55 cores, that Linux
     can't control
 (2) M7 in a separate LM from A55 cores that Linux can control using LMM
     protocol.
 (3) M7 runs in same Logical Machine as A55 cores, so Linux can control it
     using CPU protocol

So extend the driver to using LMM and CPU protocol to manage the M7 core.
 - Compare linux LM ID(got using scmi_imx_lmm_info) and M7 LM ID(the ID
   is fixed as 1 in SM firmware if M7 is in a separate LM),
   if Linux LM ID equals M7 LM ID(linux and M7 in same LM), use CPU
   protocol to start/stop. Otherwise, use LMM protocol to start/stop.
   Whether using CPU or LMM protocol to start/stop, the M7 status
   detection could use CPU protocol to detect started or not. So
   in imx_rproc_detect_mode, use scmi_imx_cpu_started to check the
   status of M7.
 - For above case (1) and (2), Use SCMI_IMX_LMM_POWER_ON to detect whether
   the M7 LM is under control of A55 LM.
 - For above case , after using SCMI_IMX_LMM_POWER_ON to check
   permission, SCMI_IMX_LMM_SHUTDOWN API should be called to shutdown
   the M7 LM to save power only when M7 LM is going to be started by
   remoteproc framework. Otherwise bypass SCMI_IMX_LMM_SHUTDOWN API if
   M7 LM is started before booting Linux.

Current setup relies on pre-Linux software(U-Boot) to do M7 TCM ECC
initialization. In future, we could add the support in Linux to decouple
U-Boot and Linux.

Signed-off-by: Peng Fan <peng.fan@nxp.com>
---
 drivers/remoteproc/Kconfig     |   2 +
 drivers/remoteproc/imx_rproc.c | 193 +++++++++++++++++++++++++++++++++++++++++
 drivers/remoteproc/imx_rproc.h |   3 +
 3 files changed, 198 insertions(+)

diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig
index 48a0d3a69ed08057716f1e7ea950899f60bbe0cf..ee54436fea5ad08a9c198ce74d44ce7a9aa206de 100644
--- a/drivers/remoteproc/Kconfig
+++ b/drivers/remoteproc/Kconfig
@@ -27,6 +27,8 @@ config IMX_REMOTEPROC
 	tristate "i.MX remoteproc support"
 	depends on ARCH_MXC
 	depends on HAVE_ARM_SMCCC
+	depends on IMX_SCMI_CPU_DRV || !IMX_SCMI_CPU_DRV
+	depends on IMX_SCMI_LMM_DRV || !IMX_SCMI_LMM_DRV
 	select MAILBOX
 	help
 	  Say y here to support iMX's remote processors via the remote
diff --git a/drivers/remoteproc/imx_rproc.c b/drivers/remoteproc/imx_rproc.c
index 232eb91e0b5dc2432131b1c140d6688b073fea1d..2ef5789f6f2a00d9159cd54e67ae9b2b96f798e8 100644
--- a/drivers/remoteproc/imx_rproc.c
+++ b/drivers/remoteproc/imx_rproc.c
@@ -8,6 +8,7 @@
 #include <linux/clk.h>
 #include <linux/err.h>
 #include <linux/firmware/imx/sci.h>
+#include <linux/firmware/imx/sm.h>
 #include <linux/interrupt.h>
 #include <linux/kernel.h>
 #include <linux/mailbox_client.h>
@@ -22,6 +23,7 @@
 #include <linux/reboot.h>
 #include <linux/regmap.h>
 #include <linux/remoteproc.h>
+#include <linux/scmi_imx_protocol.h>
 #include <linux/workqueue.h>
 
 #include "imx_rproc.h"
@@ -92,8 +94,12 @@ struct imx_rproc_mem {
 #define ATT_CORE_MASK   0xffff
 #define ATT_CORE(I)     BIT((I))
 
+/* Linux has permission to handle the Logical Machine of remote cores */
+#define IMX_RPROC_FLAGS_SM_LMM_AVAIL	BIT(0)
+
 static int imx_rproc_xtr_mbox_init(struct rproc *rproc, bool tx_block);
 static void imx_rproc_free_mbox(void *data);
+static int imx_rproc_sm_detect_mode(struct rproc *rproc);
 
 struct imx_rproc {
 	struct device			*dev;
@@ -117,6 +123,11 @@ struct imx_rproc {
 	u32				core_index;
 	struct dev_pm_domain_list	*pd_list;
 	const struct imx_rproc_plat_ops	*ops;
+	/*
+	 * For i.MX System Manager based systems
+	 * BIT 0: IMX_RPROC_FLAGS_SM_LMM_AVAIL(RPROC LM is under Linux control )
+	 */
+	u32				flags;
 };
 
 static const struct imx_rproc_att imx_rproc_att_imx93[] = {
@@ -313,6 +324,44 @@ static int imx_rproc_scu_api_start(struct rproc *rproc)
 	return imx_sc_pm_cpu_start(priv->ipc_handle, priv->rsrc_id, true, priv->entry);
 }
 
+static int imx_rproc_sm_cpu_start(struct rproc *rproc)
+{
+	struct imx_rproc *priv = rproc->priv;
+	const struct imx_rproc_dcfg *dcfg = priv->dcfg;
+	int ret;
+
+	ret = scmi_imx_cpu_reset_vector_set(dcfg->cpuid, 0, true, false, false);
+	if (ret) {
+		dev_err(priv->dev, "Failed to set reset vector cpuid(%u): %d\n", dcfg->cpuid, ret);
+		return ret;
+	}
+
+	return scmi_imx_cpu_start(dcfg->cpuid, true);
+}
+
+static int imx_rproc_sm_lmm_start(struct rproc *rproc)
+{
+	struct imx_rproc *priv = rproc->priv;
+	const struct imx_rproc_dcfg *dcfg = priv->dcfg;
+	struct device *dev = priv->dev;
+	int ret;
+
+	ret = scmi_imx_lmm_reset_vector_set(dcfg->lmid, dcfg->cpuid, 0, 0);
+	if (ret) {
+		dev_err(dev, "Failed to set reset vector lmid(%u), cpuid(%u): %d\n",
+			dcfg->lmid, dcfg->cpuid, ret);
+		return ret;
+	}
+
+	ret = scmi_imx_lmm_operation(dcfg->lmid, SCMI_IMX_LMM_BOOT, 0);
+	if (ret) {
+		dev_err(dev, "Failed to boot lmm(%d): %d\n", dcfg->lmid, ret);
+		return ret;
+	}
+
+	return 0;
+}
+
 static int imx_rproc_start(struct rproc *rproc)
 {
 	struct imx_rproc *priv = rproc->priv;
@@ -369,6 +418,25 @@ static int imx_rproc_scu_api_stop(struct rproc *rproc)
 	return imx_sc_pm_cpu_start(priv->ipc_handle, priv->rsrc_id, false, priv->entry);
 }
 
+static int imx_rproc_sm_cpu_stop(struct rproc *rproc)
+{
+	struct imx_rproc *priv = rproc->priv;
+	const struct imx_rproc_dcfg *dcfg = priv->dcfg;
+
+	return scmi_imx_cpu_start(dcfg->cpuid, false);
+}
+
+static int imx_rproc_sm_lmm_stop(struct rproc *rproc)
+{
+	struct imx_rproc *priv = rproc->priv;
+	const struct imx_rproc_dcfg *dcfg = priv->dcfg;
+
+	if (!(priv->flags & IMX_RPROC_FLAGS_SM_LMM_AVAIL))
+		return -EACCES;
+
+	return scmi_imx_lmm_operation(dcfg->lmid, SCMI_IMX_LMM_SHUTDOWN, 0);
+}
+
 static int imx_rproc_stop(struct rproc *rproc)
 {
 	struct imx_rproc *priv = rproc->priv;
@@ -485,6 +553,33 @@ static int imx_rproc_mem_release(struct rproc *rproc,
 	return 0;
 }
 
+static int imx_rproc_sm_lmm_prepare(struct rproc *rproc)
+{
+	struct imx_rproc *priv = rproc->priv;
+	const struct imx_rproc_dcfg *dcfg = priv->dcfg;
+	int ret;
+
+	/*
+	 * IMX_RPROC_FLAGS_SM_LMM_AVAIL not set indicates Linux is not able
+	 * to start/stop rproc LM, then if rproc is not in detached state,
+	 * prepare should fail. If in detached state, this is in rproc_attach()
+	 * path.
+	 */
+	if (!(priv->flags & IMX_RPROC_FLAGS_SM_LMM_AVAIL))
+		return rproc->state == RPROC_DETACHED ? 0 : -EACCES;
+
+	/* Power on the Logical Machine to make sure TCM is available. */
+	ret = scmi_imx_lmm_operation(dcfg->lmid, SCMI_IMX_LMM_POWER_ON, 0);
+	if (ret) {
+		dev_err(priv->dev, "Failed to power on lmm(%d): %d\n", dcfg->lmid, ret);
+		return ret;
+	}
+
+	dev_info(priv->dev, "lmm(%d) powered on by Linux\n", dcfg->lmid);
+
+	return 0;
+}
+
 static int imx_rproc_prepare(struct rproc *rproc)
 {
 	struct imx_rproc *priv = rproc->priv;
@@ -994,6 +1089,104 @@ static int imx_rproc_scu_api_detect_mode(struct rproc *rproc)
 	return 0;
 }
 
+static const struct imx_rproc_plat_ops imx_rproc_ops_sm_lmm = {
+	.detect_mode	= imx_rproc_sm_detect_mode,
+	.prepare	= imx_rproc_sm_lmm_prepare,
+	.start		= imx_rproc_sm_lmm_start,
+	.stop		= imx_rproc_sm_lmm_stop,
+};
+
+static const struct imx_rproc_plat_ops imx_rproc_ops_sm_cpu = {
+	.detect_mode	= imx_rproc_sm_detect_mode,
+	.start		= imx_rproc_sm_cpu_start,
+	.stop		= imx_rproc_sm_cpu_stop,
+};
+
+static int imx_rproc_sm_lmm_check(struct rproc *rproc, bool started)
+{
+	struct imx_rproc *priv = rproc->priv;
+	const struct imx_rproc_dcfg *dcfg = priv->dcfg;
+	struct device *dev = priv->dev;
+	int ret;
+
+	/*
+	 * Use power on to do permission check. If rproc is in different LM,
+	 * and linux has permission to handle the LM, set IMX_RPROC_FLAGS_SM_LMM_AVAIL.
+	 */
+	ret = scmi_imx_lmm_operation(dcfg->lmid, SCMI_IMX_LMM_POWER_ON, 0);
+	if (ret != 0) {
+		if (ret == -EACCES) {
+			/*
+			 * rproc LM is booted before Linux and not under Linux Control, so only
+			 * do IPC between RPROC and Linux, not return failure
+			 */
+			dev_info(dev, "lmm(%d) not under Linux Control\n", dcfg->lmid);
+			return 0;
+		}
+
+		dev_err(dev, "power on lmm(%d) failed: %d\n", dcfg->lmid, ret);
+		return ret;
+	}
+
+	priv->flags |= IMX_RPROC_FLAGS_SM_LMM_AVAIL;
+
+	/* rproc was started before boot Linux and under control of Linux, directly return */
+	if (started)
+		return 0;
+
+	/* else shutdown the LM to save power */
+	ret = scmi_imx_lmm_operation(dcfg->lmid, SCMI_IMX_LMM_SHUTDOWN, 0);
+	if (ret) {
+		priv->flags &= ~IMX_RPROC_FLAGS_SM_LMM_AVAIL;
+		dev_err(dev, "shutdown lmm(%d) failed: %d\n", dcfg->lmid, ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int imx_rproc_sm_detect_mode(struct rproc *rproc)
+{
+	struct imx_rproc *priv = rproc->priv;
+	const struct imx_rproc_dcfg *dcfg = priv->dcfg;
+	struct device *dev = priv->dev;
+	struct scmi_imx_lmm_info info;
+	bool started = false;
+	int ret;
+
+	ret = scmi_imx_cpu_started(dcfg->cpuid, &started);
+	if (ret) {
+		dev_err(dev, "Failed to detect cpu(%d) status: %d\n", dcfg->cpuid, ret);
+		return ret;
+	}
+
+	if (started)
+		priv->rproc->state = RPROC_DETACHED;
+
+	/* Get current Linux Logical Machine ID */
+	ret = scmi_imx_lmm_info(LMM_ID_DISCOVER, &info);
+	if (ret) {
+		dev_err(dev, "Failed to get current LMM ID err: %d\n", ret);
+		return ret;
+	}
+
+	/*
+	 * Check whether rproc is in the same LM as host core(running Linux)
+	 * If yes, use CPU protocol API to manage rproc.
+	 * If no, use Logical Machine API to manage rproc.
+	 */
+	if (dcfg->lmid == info.lmid) {
+		priv->ops = &imx_rproc_ops_sm_cpu;
+		dev_info(dev, "Using CPU Protocol OPS\n");
+		return 0;
+	}
+
+	priv->ops = &imx_rproc_ops_sm_lmm;
+	dev_info(dev, "Using LMM Protocol OPS\n");
+
+	return imx_rproc_sm_lmm_check(rproc, started);
+}
+
 static int imx_rproc_detect_mode(struct imx_rproc *priv)
 {
 	/*
diff --git a/drivers/remoteproc/imx_rproc.h b/drivers/remoteproc/imx_rproc.h
index 37417568a0ade2ae4d6a4e3d0f139ea52b185254..d37e6f90548cec727b4aeb874680b42af85bdbb4 100644
--- a/drivers/remoteproc/imx_rproc.h
+++ b/drivers/remoteproc/imx_rproc.h
@@ -38,6 +38,9 @@ struct imx_rproc_dcfg {
 	size_t				att_size;
 	u32				flags;
 	const struct imx_rproc_plat_ops	*ops;
+	/* For System Manager(SM) based SoCs */
+	u32				cpuid; /* ID of the remote core */
+	u32				lmid;  /* ID of the Logcial Machine */
 };
 
 #endif /* _IMX_RPROC_H */

-- 
2.37.1

[PATCH v3 5/5] remoteproc: imx_rproc: Add support for i.MX95

[Peng Fan (OSS)]

From: Peng Fan <peng.fan@nxp.com>

Add imx_rproc_cfg_imx95_m7 and address(TCM and DDR) mapping.
Add i.MX95 of_device_id entry.

Reviewed-by: Daniel Baluta <daniel.baluta@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
---
 drivers/remoteproc/imx_rproc.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/drivers/remoteproc/imx_rproc.c b/drivers/remoteproc/imx_rproc.c
index 2ef5789f6f2a00d9159cd54e67ae9b2b96f798e8..2ff32f01c804d3a120468d0e2e8832c5c67dabb9 100644
--- a/drivers/remoteproc/imx_rproc.c
+++ b/drivers/remoteproc/imx_rproc.c
@@ -130,6 +130,18 @@ struct imx_rproc {
 	u32				flags;
 };
 
+static const struct imx_rproc_att imx_rproc_att_imx95_m7[] = {
+	/* dev addr , sys addr  , size	    , flags */
+	/* TCM CODE NON-SECURE */
+	{ 0x00000000, 0x203C0000, 0x00040000, ATT_OWN | ATT_IOMEM },
+
+	/* TCM SYS NON-SECURE*/
+	{ 0x20000000, 0x20400000, 0x00040000, ATT_OWN | ATT_IOMEM },
+
+	/* DDR */
+	{ 0x80000000, 0x80000000, 0x50000000, 0 },
+};
+
 static const struct imx_rproc_att imx_rproc_att_imx93[] = {
 	/* dev addr , sys addr  , size	    , flags */
 	/* TCM CODE NON-SECURE */
@@ -1445,6 +1457,15 @@ static const struct imx_rproc_dcfg imx_rproc_cfg_imx93 = {
 	.flags		= IMX_RPROC_NEED_CLKS,
 };
 
+static const struct imx_rproc_dcfg imx_rproc_cfg_imx95_m7 = {
+	.att		= imx_rproc_att_imx95_m7,
+	.att_size	= ARRAY_SIZE(imx_rproc_att_imx95_m7),
+	.ops		= &imx_rproc_ops_sm_lmm,
+	/* Must align with System Manager Firmware */
+	.cpuid		= 1, /* Use 1 as cpu id for M7 core */
+	.lmid		= 1, /* Use 1 as Logical Machine ID where M7 resides */
+};
+
 static const struct of_device_id imx_rproc_of_match[] = {
 	{ .compatible = "fsl,imx7ulp-cm4", .data = &imx_rproc_cfg_imx7ulp },
 	{ .compatible = "fsl,imx7d-cm4", .data = &imx_rproc_cfg_imx7d },
@@ -1459,6 +1480,7 @@ static const struct of_device_id imx_rproc_of_match[] = {
 	{ .compatible = "fsl,imx8qm-cm4", .data = &imx_rproc_cfg_imx8qm },
 	{ .compatible = "fsl,imx8ulp-cm33", .data = &imx_rproc_cfg_imx8ulp },
 	{ .compatible = "fsl,imx93-cm33", .data = &imx_rproc_cfg_imx93 },
+	{ .compatible = "fsl,imx95-cm7", .data = &imx_rproc_cfg_imx95_m7 },
 	{},
 };
 MODULE_DEVICE_TABLE(of, imx_rproc_of_match);

-- 
2.37.1

Re: [PATCH v3 0/5] remoteproc: imx_rproc: Support i.MX95

[Mathieu Poirier]

Good morning,

Please note that due to other patchsets that need to be reviewed,
Linux Plumbers and the upcoming 2-week December holidays in the
western hemisphere, I do not foresee being able to look at this
patchset before January.

Thanks,
Mathieu

On Fri, 21 Nov 2025 at 17:58, Peng Fan (OSS) <peng.fan@oss.nxp.com> wrote:
>
> This patchset is pick up a previous patchset [1] with rebased on
> next-20251030, and some changes applied.
>  - Add runtime ops to separate cpu ops and lmm ops
>  - added more comments
>  - moved some check imx_sm_rproc_detect_mode() from imx_rproc_sm_prepare().
>
> No changes to the dt-binding patch, so R-b/A-b are kept.
> More info could be found in commit message of each patch and below.
>
> [1]https://lore.kernel.org/linux-remoteproc/20250821-imx95-rproc-1-v5-0-e93191dfac51@nxp.com/
>
> i.MX95 features a Cortex-M33 core, six Cortex-A55 cores, and
> one Cortex-M7 core. The System Control Management Interface(SCMI)
> firmware runs on the M33 core. The i.MX95 SCMI firmware named System
> Manager(SM) includes vendor extension protocols, Logical Machine
> Management(LMM) protocol and CPU protocol and etc.
>
> There are three cases for M7:
> (1) M7 in a separate Logical Machine(LM) that Linux couldn't control it.
> (2) M7 in a separate Logical Machine that Linux could control it using
>     LMM protocol
> (3) M7 runs in same Logical Machine as A55, so Linux could control it
>     using CPU protocol
>
> In patch 3, Use LMM and CPU protocol to manage M7. More info could be
> found in the patch commit log
>     Current setup relies on pre-Linux software(U-Boot) to do
> M7 TCM ECC initialization. In future, we could add the support in Linux
>
> Patchset was tested with below boot images when the patchset based on next-20251030:
> imx-boot-variant-rpmsg-imx95-19x19-lpddr5-evk-sd.bin-flash_lpboot_sm_a55 (Use LMM protocol)
> imx-boot-variant-alt-imx95-19x19-lpddr5-evk-sd.bin-flash_alt (Use CPU protocol)
> imx-boot-imx95-19x19-lpddr5-evk-sd.bin-flash_all (M7 not under A55 control)
>
> Also tested i.MX8MP/8ULP-EVK.
>
> Signed-off-by: Peng Fan <peng.fan@nxp.com>
> ---
> Changes in v3:
> - Add R-b for patch 2,3,5
> - Patch 4:
>   Rename imx_rproc_ops_sm to imx_rproc_ops_sm_lmm.
>   Add more comments in code
>   Simplify if check in imx_rproc_sm_lmm_prepare()
>   Update commit log
>   Use dev_err for failure path
> - Link to v2: https://lore.kernel.org/linux-remoteproc/20251031-imx95-rproc-2025-10-31-v2-0-758b2e547a55@nxp.com/T/#t
>
> Changes in v2:
> - Patch 2: Change to const pointer for ops, update commit log and add R-b from Frank
> - Patch 3: New patch, introduce prepare ops
> - Patch 4: Rename imx_rproc_sm_prepare to imx_rproc_sm_lmm_prepare
>            drop IMX_RPROC_FLAGS_SM_CPU_OP and IMX_RPROC_FLAGS_SM_LMM_OP
>            Add a bool for dcfg->lmid and info->lmid check.
>            Provide separate plat ops for lmm and cpu setup.
>            Move LM permission check to a separate function.
> - Patch 5: Drop macro IMX95_M7_CPUID and IMX95_M7_LMID
> - Link to v1: https://lore.kernel.org/r/20251028-imx95-rproc-2025-10-28-v1-0-ce9e7db9edcb@nxp.com
>
> ---
> Peng Fan (5):
>       dt-bindings: remoteproc: fsl,imx-rproc: Add support for i.MX95
>       remoteproc: imx_rproc: Add runtime ops copy to support dynamic behavior
>       remoteproc: imx_rproc: Introduce prepare ops for imx_rproc_dcfg
>       remoteproc: imx_rproc: Add support for System Manager API
>       remoteproc: imx_rproc: Add support for i.MX95
>
>  .../bindings/remoteproc/fsl,imx-rproc.yaml         |   1 +
>  drivers/remoteproc/Kconfig                         |   2 +
>  drivers/remoteproc/imx_rproc.c                     | 245 +++++++++++++++++++--
>  drivers/remoteproc/imx_rproc.h                     |   4 +
>  4 files changed, 238 insertions(+), 14 deletions(-)
> ---
> base-commit: 88cbd8ac379cf5ce68b7efcfd4d1484a6871ee0b
> change-id: 20251120-imx95-rproc-2025-11-20-3b9369670d6a
>
> Best regards,
> --
> Peng Fan <peng.fan@nxp.com>
>

Re: [PATCH v3 4/5] remoteproc: imx_rproc: Add support for System Manager API

[Frank Li]

On Sat, Nov 22, 2025 at 08:58:19AM +0800, Peng Fan (OSS) wrote:
> From: Peng Fan <peng.fan@nxp.com>
>
> i.MX95 features a Cortex-M33 core, six Cortex-A55 cores, and
> one Cortex-M7 core. The System Control Management Interface(SCMI)
> firmware runs on the M33 core. The i.MX95 SCMI firmware named System
> Manager(SM) includes vendor extension protocols, Logical Machine
> Management(LMM) protocol and CPU protocol and etc.
>
> Depending on SM configuration, M7 can be used as follows:
>  (1) M7 in a separate Logical Machine (LM) from A55 cores, that Linux
>      can't control
>  (2) M7 in a separate LM from A55 cores that Linux can control using LMM
>      protocol.
>  (3) M7 runs in same Logical Machine as A55 cores, so Linux can control it
>      using CPU protocol
>
> So extend the driver to using LMM and CPU protocol to manage the M7 core.
>  - Compare linux LM ID(got using scmi_imx_lmm_info) and M7 LM ID(the ID
>    is fixed as 1 in SM firmware if M7 is in a separate LM),
>    if Linux LM ID equals M7 LM ID(linux and M7 in same LM), use CPU
>    protocol to start/stop. Otherwise, use LMM protocol to start/stop.
>    Whether using CPU or LMM protocol to start/stop, the M7 status
>    detection could use CPU protocol to detect started or not. So
>    in imx_rproc_detect_mode, use scmi_imx_cpu_started to check the
>    status of M7.
>  - For above case (1) and (2), Use SCMI_IMX_LMM_POWER_ON to detect whether
>    the M7 LM is under control of A55 LM.
>  - For above case , after using SCMI_IMX_LMM_POWER_ON to check
>    permission, SCMI_IMX_LMM_SHUTDOWN API should be called to shutdown
>    the M7 LM to save power only when M7 LM is going to be started by
>    remoteproc framework. Otherwise bypass SCMI_IMX_LMM_SHUTDOWN API if
>    M7 LM is started before booting Linux.
>
> Current setup relies on pre-Linux software(U-Boot) to do M7 TCM ECC
> initialization. In future, we could add the support in Linux to decouple
> U-Boot and Linux.
>
> Signed-off-by: Peng Fan <peng.fan@nxp.com>
> ---
>  drivers/remoteproc/Kconfig     |   2 +
>  drivers/remoteproc/imx_rproc.c | 193 +++++++++++++++++++++++++++++++++++++++++
>  drivers/remoteproc/imx_rproc.h |   3 +
>  3 files changed, 198 insertions(+)
>
...
> +
> +static int imx_rproc_sm_lmm_check(struct rproc *rproc, bool started)
> +{
> +	struct imx_rproc *priv = rproc->priv;
> +	const struct imx_rproc_dcfg *dcfg = priv->dcfg;
> +	struct device *dev = priv->dev;
> +	int ret;
> +
> +	/*
> +	 * Use power on to do permission check. If rproc is in different LM,
> +	 * and linux has permission to handle the LM, set IMX_RPROC_FLAGS_SM_LMM_AVAIL.
> +	 */
> +	ret = scmi_imx_lmm_operation(dcfg->lmid, SCMI_IMX_LMM_POWER_ON, 0);
> +	if (ret != 0) {

if (ret)

> +		if (ret == -EACCES) {
> +			/*
> +			 * rproc LM is booted before Linux and not under Linux Control, so only
> +			 * do IPC between RPROC and Linux, not return failure
> +			 */
> +			dev_info(dev, "lmm(%d) not under Linux Control\n", dcfg->lmid);
> +			return 0;
> +		}
> +
> +		dev_err(dev, "power on lmm(%d) failed: %d\n", dcfg->lmid, ret);
> +		return ret;
> +	}
> +
> +	priv->flags |= IMX_RPROC_FLAGS_SM_LMM_AVAIL;
> +
> +	/* rproc was started before boot Linux and under control of Linux, directly return */
> +	if (started)
> +		return 0;

should it be before scmi_imx_lmm_operation()? If started, supposed need
check scmi_imx_lmm_operation()?

Frank
> +
> +	/* else shutdown the LM to save power */
> +	ret = scmi_imx_lmm_operation(dcfg->lmid, SCMI_IMX_LMM_SHUTDOWN, 0);
> +	if (ret) {
> +		priv->flags &= ~IMX_RPROC_FLAGS_SM_LMM_AVAIL;
> +		dev_err(dev, "shutdown lmm(%d) failed: %d\n", dcfg->lmid, ret);
> +		return ret;
> +	}
> +
> +	return 0;
> +}
> +
...
>  };
>
>  #endif /* _IMX_RPROC_H */
>
> --
> 2.37.1
>

Re: [PATCH v3 4/5] remoteproc: imx_rproc: Add support for System Manager API

[Peng Fan]

On Mon, Nov 24, 2025 at 12:28:00PM -0500, Frank Li wrote:
>On Sat, Nov 22, 2025 at 08:58:19AM +0800, Peng Fan (OSS) wrote:
>> From: Peng Fan <peng.fan@nxp.com>
>>
>> i.MX95 features a Cortex-M33 core, six Cortex-A55 cores, and
>> one Cortex-M7 core. The System Control Management Interface(SCMI)
>> firmware runs on the M33 core. The i.MX95 SCMI firmware named System
>> Manager(SM) includes vendor extension protocols, Logical Machine
>> Management(LMM) protocol and CPU protocol and etc.
>>
>> Depending on SM configuration, M7 can be used as follows:
>>  (1) M7 in a separate Logical Machine (LM) from A55 cores, that Linux
>>      can't control
>>  (2) M7 in a separate LM from A55 cores that Linux can control using LMM
>>      protocol.
>>  (3) M7 runs in same Logical Machine as A55 cores, so Linux can control it
>>      using CPU protocol
>>
>> So extend the driver to using LMM and CPU protocol to manage the M7 core.
>>  - Compare linux LM ID(got using scmi_imx_lmm_info) and M7 LM ID(the ID
>>    is fixed as 1 in SM firmware if M7 is in a separate LM),
>>    if Linux LM ID equals M7 LM ID(linux and M7 in same LM), use CPU
>>    protocol to start/stop. Otherwise, use LMM protocol to start/stop.
>>    Whether using CPU or LMM protocol to start/stop, the M7 status
>>    detection could use CPU protocol to detect started or not. So
>>    in imx_rproc_detect_mode, use scmi_imx_cpu_started to check the
>>    status of M7.
>>  - For above case (1) and (2), Use SCMI_IMX_LMM_POWER_ON to detect whether
>>    the M7 LM is under control of A55 LM.
>>  - For above case , after using SCMI_IMX_LMM_POWER_ON to check
>>    permission, SCMI_IMX_LMM_SHUTDOWN API should be called to shutdown
>>    the M7 LM to save power only when M7 LM is going to be started by
>>    remoteproc framework. Otherwise bypass SCMI_IMX_LMM_SHUTDOWN API if
>>    M7 LM is started before booting Linux.
>>
>> Current setup relies on pre-Linux software(U-Boot) to do M7 TCM ECC
>> initialization. In future, we could add the support in Linux to decouple
>> U-Boot and Linux.
>>
>> Signed-off-by: Peng Fan <peng.fan@nxp.com>
>> ---
>>  drivers/remoteproc/Kconfig     |   2 +
>>  drivers/remoteproc/imx_rproc.c | 193 +++++++++++++++++++++++++++++++++++++++++
>>  drivers/remoteproc/imx_rproc.h |   3 +
>>  3 files changed, 198 insertions(+)
>>
>...
>> +
>> +static int imx_rproc_sm_lmm_check(struct rproc *rproc, bool started)
>> +{
>> +	struct imx_rproc *priv = rproc->priv;
>> +	const struct imx_rproc_dcfg *dcfg = priv->dcfg;
>> +	struct device *dev = priv->dev;
>> +	int ret;
>> +
>> +	/*
>> +	 * Use power on to do permission check. If rproc is in different LM,
>> +	 * and linux has permission to handle the LM, set IMX_RPROC_FLAGS_SM_LMM_AVAIL.
>> +	 */
>> +	ret = scmi_imx_lmm_operation(dcfg->lmid, SCMI_IMX_LMM_POWER_ON, 0);
>> +	if (ret != 0) {
>
>if (ret)

Fix in next version.

>
>> +		if (ret == -EACCES) {
>> +			/*
>> +			 * rproc LM is booted before Linux and not under Linux Control, so only
>> +			 * do IPC between RPROC and Linux, not return failure
>> +			 */
>> +			dev_info(dev, "lmm(%d) not under Linux Control\n", dcfg->lmid);
>> +			return 0;
>> +		}
>> +
>> +		dev_err(dev, "power on lmm(%d) failed: %d\n", dcfg->lmid, ret);
>> +		return ret;
>> +	}
>> +
>> +	priv->flags |= IMX_RPROC_FLAGS_SM_LMM_AVAIL;
>> +
>> +	/* rproc was started before boot Linux and under control of Linux, directly return */
>> +	if (started)
>> +		return 0;
>
>should it be before scmi_imx_lmm_operation()? If started, supposed need
>check scmi_imx_lmm_operation()?

No. We need to detect whether M7 is under control of Linux before return.

Thanks,
Peng

>
>Frank
>> +
>> +	/* else shutdown the LM to save power */
>> +	ret = scmi_imx_lmm_operation(dcfg->lmid, SCMI_IMX_LMM_SHUTDOWN, 0);
>> +	if (ret) {
>> +		priv->flags &= ~IMX_RPROC_FLAGS_SM_LMM_AVAIL;
>> +		dev_err(dev, "shutdown lmm(%d) failed: %d\n", dcfg->lmid, ret);
>> +		return ret;
>> +	}
>> +
>> +	return 0;
>> +}
>> +
>...
>>  };
>>
>>  #endif /* _IMX_RPROC_H */
>>
>> --
>> 2.37.1
>>