[PATCH v7 0/6] Use CCP driver to handle PSP I2C arbitration

[PATCH v7 0/6] Use CCP driver to handle PSP I2C arbitration

[Mario Limonciello]

The CCP driver now has symbols that can be used by i2c-designware-amdpsp
to handle the communication regarding i2c arbitration with the PSP for
both Cezanne and Mendocino based designs.

Utilize those symbols.

Mario Limonciello (6):
  crypto: ccp: Drop extra doorbell checks
  crypto: ccp: Bump up doorbell debug message to error
  crypto: ccp: Return doorbell status code as an argument
  crypto: ccp: Use lower 8 bytes to communicate with doorbell command
    register
  i2c: designware: Use PCI PSP driver for communication
  i2c: designware: Add doorbell support for Mendocino

 drivers/crypto/ccp/platform-access.c        |  31 +--
 drivers/i2c/busses/Kconfig                  |   4 +-
 drivers/i2c/busses/i2c-designware-amdpsp.c  | 199 +++++---------------
 drivers/i2c/busses/i2c-designware-core.h    |   1 -
 drivers/i2c/busses/i2c-designware-platdrv.c |   1 -
 include/linux/psp-platform-access.h         |   5 +-
 6 files changed, 62 insertions(+), 179 deletions(-)


base-commit: 9117e682b8b79f7b5e2517fd28d42757d3e8b860
-- 
2.34.1

[PATCH v7 1/6] crypto: ccp: Drop extra doorbell checks

[Mario Limonciello]

The doorbell register set used for I2C arbitration is dedicated for this
purpose and there is no need to utilize other safety checks the platform
access register set uses.

Suggested-by: Mark Hasemeyer <markhas@chromium.org>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
---
v6->v7:
 * new patch
---
 drivers/crypto/ccp/platform-access.c | 20 --------------------
 1 file changed, 20 deletions(-)

diff --git a/drivers/crypto/ccp/platform-access.c b/drivers/crypto/ccp/platform-access.c
index b51fb1196932..1ad3a0a512b1 100644
--- a/drivers/crypto/ccp/platform-access.c
+++ b/drivers/crypto/ccp/platform-access.c
@@ -20,14 +20,6 @@
 
 #define PSP_CMD_TIMEOUT_US	(500 * USEC_PER_MSEC)
 
-/* Doorbell shouldn't be ringing */
-static int check_doorbell(u32 __iomem *doorbell)
-{
-	u32 tmp;
-
-	return readl_poll_timeout(doorbell, tmp, tmp != 0, 0, PSP_CMD_TIMEOUT_US);
-}
-
 /* Recovery field should be equal 0 to start sending commands */
 static int check_recovery(u32 __iomem *cmd)
 {
@@ -156,18 +148,6 @@ int psp_ring_platform_doorbell(int msg)
 
 	mutex_lock(&pa_dev->doorbell_mutex);
 
-	if (check_doorbell(button)) {
-		dev_dbg(psp->dev, "doorbell is not ready\n");
-		ret = -EBUSY;
-		goto unlock;
-	}
-
-	if (check_recovery(cmd)) {
-		dev_dbg(psp->dev, "doorbell command in recovery\n");
-		ret = -EBUSY;
-		goto unlock;
-	}
-
 	if (wait_cmd(cmd)) {
 		dev_dbg(psp->dev, "doorbell command not done processing\n");
 		ret = -EBUSY;

base-commit: 9117e682b8b79f7b5e2517fd28d42757d3e8b860
-- 
2.34.1

[PATCH v7 2/6] crypto: ccp: Bump up doorbell debug message to error

[Mario Limonciello]

This is helpful not just for debugging problems, but also for investigating
captured logs later on.

Suggested-by: Grzegorz Bernacki <gjb@semihalf.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
---
 drivers/crypto/ccp/platform-access.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/crypto/ccp/platform-access.c b/drivers/crypto/ccp/platform-access.c
index 1ad3a0a512b1..1cc154a1c6ab 100644
--- a/drivers/crypto/ccp/platform-access.c
+++ b/drivers/crypto/ccp/platform-access.c
@@ -149,7 +149,7 @@ int psp_ring_platform_doorbell(int msg)
 	mutex_lock(&pa_dev->doorbell_mutex);
 
 	if (wait_cmd(cmd)) {
-		dev_dbg(psp->dev, "doorbell command not done processing\n");
+		dev_err(psp->dev, "doorbell command not done processing\n");
 		ret = -EBUSY;
 		goto unlock;
 	}
-- 
2.34.1

[PATCH v7 3/6] crypto: ccp: Return doorbell status code as an argument

[Mario Limonciello]

If the doorbell failed to ring we return -EIO, but the caller can't
determine why it failed.  Pass the reason for the failure in an
argument for caller to investigate.

Suggested-by: Mark Hasemeyer <markhas@chromium.org>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
---
v5->v6:
 * New patch
---
 drivers/crypto/ccp/platform-access.c | 4 +++-
 include/linux/psp-platform-access.h  | 4 ++--
 2 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/crypto/ccp/platform-access.c b/drivers/crypto/ccp/platform-access.c
index 1cc154a1c6ab..48f59ae91692 100644
--- a/drivers/crypto/ccp/platform-access.c
+++ b/drivers/crypto/ccp/platform-access.c
@@ -132,7 +132,7 @@ int psp_send_platform_access_msg(enum psp_platform_access_msg msg,
 }
 EXPORT_SYMBOL_GPL(psp_send_platform_access_msg);
 
-int psp_ring_platform_doorbell(int msg)
+int psp_ring_platform_doorbell(int msg, u32 *result)
 {
 	struct psp_device *psp = psp_get_master_device();
 	struct psp_platform_access_device *pa_dev;
@@ -164,6 +164,8 @@ int psp_ring_platform_doorbell(int msg)
 
 	val = FIELD_GET(PSP_CMDRESP_STS, ioread32(cmd));
 	if (val) {
+		if (result)
+			*result = val;
 		ret = -EIO;
 		goto unlock;
 	}
diff --git a/include/linux/psp-platform-access.h b/include/linux/psp-platform-access.h
index 89df4549fada..1b661341d8f3 100644
--- a/include/linux/psp-platform-access.h
+++ b/include/linux/psp-platform-access.h
@@ -45,9 +45,9 @@ int psp_send_platform_access_msg(enum psp_platform_access_msg, struct psp_reques
  *  -%EBUSY:     mailbox in recovery or in use
  *  -%ENODEV:    driver not bound with PSP device
  *  -%ETIMEDOUT: request timed out
- *  -%EIO:       unknown error (see kernel log)
+ *  -%EIO:       error will be stored in result argument
  */
-int psp_ring_platform_doorbell(int msg);
+int psp_ring_platform_doorbell(int msg, u32 *result);
 
 /**
  * psp_check_platform_access_status() - Checks whether platform features is ready
-- 
2.34.1

[PATCH v7 4/6] crypto: ccp: Use lower 8 bytes to communicate with doorbell command register

[Mario Limonciello]

Unlike other command registers used by the PSP, only the lower 8 bytes are
used for communication for both command and status of the command.

Suggested-by: Mark Hasemeyer <markhas@chromium.org>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
---
v6->v7:
 * New patch
---
 drivers/crypto/ccp/platform-access.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/crypto/ccp/platform-access.c b/drivers/crypto/ccp/platform-access.c
index 48f59ae91692..939c924fc383 100644
--- a/drivers/crypto/ccp/platform-access.c
+++ b/drivers/crypto/ccp/platform-access.c
@@ -19,6 +19,7 @@
 #include "platform-access.h"
 
 #define PSP_CMD_TIMEOUT_US	(500 * USEC_PER_MSEC)
+#define DOORBELL_CMDRESP_STS	GENMASK(7, 0)
 
 /* Recovery field should be equal 0 to start sending commands */
 static int check_recovery(u32 __iomem *cmd)
@@ -154,7 +155,7 @@ int psp_ring_platform_doorbell(int msg, u32 *result)
 		goto unlock;
 	}
 
-	iowrite32(FIELD_PREP(PSP_DRBL_MSG, msg), cmd);
+	iowrite32(FIELD_PREP(DOORBELL_CMDRESP_STS, msg), cmd);
 	iowrite32(PSP_DRBL_RING, button);
 
 	if (wait_cmd(cmd)) {
@@ -162,7 +163,7 @@ int psp_ring_platform_doorbell(int msg, u32 *result)
 		goto unlock;
 	}
 
-	val = FIELD_GET(PSP_CMDRESP_STS, ioread32(cmd));
+	val = FIELD_GET(DOORBELL_CMDRESP_STS, ioread32(cmd));
 	if (val) {
 		if (result)
 			*result = val;
-- 
2.34.1

[PATCH v7 5/6] i2c: designware: Use PCI PSP driver for communication

[Mario Limonciello]

Currently the PSP semaphore communication base address is discovered
by using an MSR that is not architecturally guaranteed for future
platforms.  Also the mailbox that is utilized for communication with
the PSP may have other consumers in the kernel, so it's better to
make all communication go through a single driver.

Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
---
v6->v7:
 * Also imply CRYPTO_DEV_CCP_DD to fix build errors
 * Fix error message acquire/release inversion
v5->v6:
 * Drop now unnecessary asm/msr.h header
 * Fix IS_REACHABLE
 * Drop tags
 * Fix status code handling for Cezanne
v4->v5:
 * Pick up tags
v3->v4:
 * Pick up tags
v1->v2:
 * Fix Kconfig to use imply
 * Use IS_REACHABLE
---
 drivers/i2c/busses/Kconfig                  |   3 +-
 drivers/i2c/busses/i2c-designware-amdpsp.c  | 177 +++-----------------
 drivers/i2c/busses/i2c-designware-core.h    |   1 -
 drivers/i2c/busses/i2c-designware-platdrv.c |   1 -
 include/linux/psp-platform-access.h         |   1 +
 5 files changed, 29 insertions(+), 154 deletions(-)

diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index 25eb4e8fd22f..0d47d4eeb2a4 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig
@@ -566,9 +566,10 @@ config I2C_DESIGNWARE_PLATFORM
 
 config I2C_DESIGNWARE_AMDPSP
 	bool "AMD PSP I2C semaphore support"
-	depends on X86_MSR
 	depends on ACPI
 	depends on I2C_DESIGNWARE_PLATFORM
+	imply CRYPTO_DEV_SP_PSP
+	imply CRYPTO_DEV_CCP_DD
 	help
 	  This driver enables managed host access to the selected I2C bus shared
 	  between AMD CPU and AMD PSP.
diff --git a/drivers/i2c/busses/i2c-designware-amdpsp.c b/drivers/i2c/busses/i2c-designware-amdpsp.c
index 652e6b64bd5f..37d7d8cd05c0 100644
--- a/drivers/i2c/busses/i2c-designware-amdpsp.c
+++ b/drivers/i2c/busses/i2c-designware-amdpsp.c
@@ -1,35 +1,20 @@
 // SPDX-License-Identifier: GPL-2.0
 
-#include <linux/bitfield.h>
-#include <linux/bits.h>
 #include <linux/i2c.h>
-#include <linux/io-64-nonatomic-lo-hi.h>
+#include <linux/psp-platform-access.h>
 #include <linux/psp.h>
-#include <linux/types.h>
 #include <linux/workqueue.h>
 
-#include <asm/msr.h>
-
 #include "i2c-designware-core.h"
 
-#define MSR_AMD_PSP_ADDR	0xc00110a2
-#define PSP_MBOX_OFFSET		0x10570
-#define PSP_CMD_TIMEOUT_US	(500 * USEC_PER_MSEC)
-
 #define PSP_I2C_RESERVATION_TIME_MS 100
 
-#define PSP_I2C_REQ_BUS_CMD		0x64
 #define PSP_I2C_REQ_RETRY_CNT		400
 #define PSP_I2C_REQ_RETRY_DELAY_US	(25 * USEC_PER_MSEC)
 #define PSP_I2C_REQ_STS_OK		0x0
 #define PSP_I2C_REQ_STS_BUS_BUSY	0x1
 #define PSP_I2C_REQ_STS_INV_PARAM	0x3
 
-struct psp_req_buffer_hdr {
-	u32 total_size;
-	u32 status;
-};
-
 enum psp_i2c_req_type {
 	PSP_I2C_REQ_ACQUIRE,
 	PSP_I2C_REQ_RELEASE,
@@ -41,119 +26,12 @@ struct psp_i2c_req {
 	enum psp_i2c_req_type type;
 };
 
-struct psp_mbox {
-	u32 cmd_fields;
-	u64 i2c_req_addr;
-} __packed;
-
 static DEFINE_MUTEX(psp_i2c_access_mutex);
 static unsigned long psp_i2c_sem_acquired;
-static void __iomem *mbox_iomem;
 static u32 psp_i2c_access_count;
 static bool psp_i2c_mbox_fail;
 static struct device *psp_i2c_dev;
 
-/*
- * Implementation of PSP-x86 i2c-arbitration mailbox introduced for AMD Cezanne
- * family of SoCs.
- */
-
-static int psp_get_mbox_addr(unsigned long *mbox_addr)
-{
-	unsigned long long psp_mmio;
-
-	if (rdmsrl_safe(MSR_AMD_PSP_ADDR, &psp_mmio))
-		return -EIO;
-
-	*mbox_addr = (unsigned long)(psp_mmio + PSP_MBOX_OFFSET);
-
-	return 0;
-}
-
-static int psp_mbox_probe(void)
-{
-	unsigned long mbox_addr;
-	int ret;
-
-	ret = psp_get_mbox_addr(&mbox_addr);
-	if (ret)
-		return ret;
-
-	mbox_iomem = ioremap(mbox_addr, sizeof(struct psp_mbox));
-	if (!mbox_iomem)
-		return -ENOMEM;
-
-	return 0;
-}
-
-/* Recovery field should be equal 0 to start sending commands */
-static int psp_check_mbox_recovery(struct psp_mbox __iomem *mbox)
-{
-	u32 tmp;
-
-	tmp = readl(&mbox->cmd_fields);
-
-	return FIELD_GET(PSP_CMDRESP_RECOVERY, tmp);
-}
-
-static int psp_wait_cmd(struct psp_mbox __iomem *mbox)
-{
-	u32 tmp, expected;
-
-	/* Expect mbox_cmd to be cleared and the response bit to be set by PSP */
-	expected = FIELD_PREP(PSP_CMDRESP_RESP, 1);
-
-	/*
-	 * Check for readiness of PSP mailbox in a tight loop in order to
-	 * process further as soon as command was consumed.
-	 */
-	return readl_poll_timeout(&mbox->cmd_fields, tmp, (tmp == expected),
-				  0, PSP_CMD_TIMEOUT_US);
-}
-
-/* Status equal to 0 means that PSP succeed processing command */
-static u32 psp_check_mbox_sts(struct psp_mbox __iomem *mbox)
-{
-	u32 cmd_reg;
-
-	cmd_reg = readl(&mbox->cmd_fields);
-
-	return FIELD_GET(PSP_CMDRESP_STS, cmd_reg);
-}
-
-static int psp_send_cmd(struct psp_i2c_req *req)
-{
-	struct psp_mbox __iomem *mbox = mbox_iomem;
-	phys_addr_t req_addr;
-	u32 cmd_reg;
-
-	if (psp_check_mbox_recovery(mbox))
-		return -EIO;
-
-	if (psp_wait_cmd(mbox))
-		return -EBUSY;
-
-	/*
-	 * Fill mailbox with address of command-response buffer, which will be
-	 * used for sending i2c requests as well as reading status returned by
-	 * PSP. Use physical address of buffer, since PSP will map this region.
-	 */
-	req_addr = __psp_pa((void *)req);
-	writeq(req_addr, &mbox->i2c_req_addr);
-
-	/* Write command register to trigger processing */
-	cmd_reg = FIELD_PREP(PSP_CMDRESP_CMD, PSP_I2C_REQ_BUS_CMD);
-	writel(cmd_reg, &mbox->cmd_fields);
-
-	if (psp_wait_cmd(mbox))
-		return -ETIMEDOUT;
-
-	if (psp_check_mbox_sts(mbox))
-		return -EIO;
-
-	return 0;
-}
-
 /* Helper to verify status returned by PSP */
 static int check_i2c_req_sts(struct psp_i2c_req *req)
 {
@@ -173,22 +51,25 @@ static int check_i2c_req_sts(struct psp_i2c_req *req)
 	}
 }
 
-static int psp_send_check_i2c_req(struct psp_i2c_req *req)
+/*
+ * Errors in x86-PSP i2c-arbitration protocol may occur at two levels:
+ * 1. mailbox communication - PSP is not operational or some IO errors with
+ *    basic communication had happened.
+ * 2. i2c-requests - PSP refuses to grant i2c arbitration to x86 for too long.
+ *
+ * In order to distinguish between these in error handling code all mailbox
+ * communication errors on the first level (from CCP symbols) will be passed
+ * up and if -EIO is returned the second level will be checked.
+ */
+static int psp_send_i2c_req_cezanne(struct psp_i2c_req *req)
 {
-	/*
-	 * Errors in x86-PSP i2c-arbitration protocol may occur at two levels:
-	 * 1. mailbox communication - PSP is not operational or some IO errors
-	 * with basic communication had happened;
-	 * 2. i2c-requests - PSP refuses to grant i2c arbitration to x86 for too
-	 * long.
-	 * In order to distinguish between these two in error handling code, all
-	 * errors on the first level (returned by psp_send_cmd) are shadowed by
-	 * -EIO.
-	 */
-	if (psp_send_cmd(req))
-		return -EIO;
+	int ret;
 
-	return check_i2c_req_sts(req);
+	ret = psp_send_platform_access_msg(PSP_I2C_REQ_BUS_CMD, (struct psp_request *)req);
+	if (ret == -EIO)
+		return check_i2c_req_sts(req);
+
+	return ret;
 }
 
 static int psp_send_i2c_req(enum psp_i2c_req_type i2c_req_type)
@@ -202,11 +83,11 @@ static int psp_send_i2c_req(enum psp_i2c_req_type i2c_req_type)
 	if (!req)
 		return -ENOMEM;
 
-	req->hdr.total_size = sizeof(*req);
+	req->hdr.payload_size = sizeof(*req);
 	req->type = i2c_req_type;
 
 	start = jiffies;
-	ret = read_poll_timeout(psp_send_check_i2c_req, status,
+	ret = read_poll_timeout(psp_send_i2c_req_cezanne, status,
 				(status != -EBUSY),
 				PSP_I2C_REQ_RETRY_DELAY_US,
 				PSP_I2C_REQ_RETRY_CNT * PSP_I2C_REQ_RETRY_DELAY_US,
@@ -214,7 +95,7 @@ static int psp_send_i2c_req(enum psp_i2c_req_type i2c_req_type)
 	if (ret) {
 		dev_err(psp_i2c_dev, "Timed out waiting for PSP to %s I2C bus\n",
 			(i2c_req_type == PSP_I2C_REQ_ACQUIRE) ?
-			"release" : "acquire");
+			"acquire" : "release");
 		goto cleanup;
 	}
 
@@ -381,7 +262,8 @@ static const struct i2c_lock_operations i2c_dw_psp_lock_ops = {
 
 int i2c_dw_amdpsp_probe_lock_support(struct dw_i2c_dev *dev)
 {
-	int ret;
+	if (!IS_REACHABLE(CONFIG_CRYPTO_DEV_CCP_DD))
+		return -ENODEV;
 
 	if (!dev)
 		return -ENODEV;
@@ -393,11 +275,10 @@ int i2c_dw_amdpsp_probe_lock_support(struct dw_i2c_dev *dev)
 	if (psp_i2c_dev)
 		return -EEXIST;
 
-	psp_i2c_dev = dev->dev;
+	if (psp_check_platform_access_status())
+		return -EPROBE_DEFER;
 
-	ret = psp_mbox_probe();
-	if (ret)
-		return ret;
+	psp_i2c_dev = dev->dev;
 
 	dev_info(psp_i2c_dev, "I2C bus managed by AMD PSP\n");
 
@@ -411,9 +292,3 @@ int i2c_dw_amdpsp_probe_lock_support(struct dw_i2c_dev *dev)
 
 	return 0;
 }
-
-/* Unmap area used as a mailbox with PSP */
-void i2c_dw_amdpsp_remove_lock_support(struct dw_i2c_dev *dev)
-{
-	iounmap(mbox_iomem);
-}
diff --git a/drivers/i2c/busses/i2c-designware-core.h b/drivers/i2c/busses/i2c-designware-core.h
index 050d8c63ad3c..c5d87aae39c6 100644
--- a/drivers/i2c/busses/i2c-designware-core.h
+++ b/drivers/i2c/busses/i2c-designware-core.h
@@ -383,7 +383,6 @@ int i2c_dw_baytrail_probe_lock_support(struct dw_i2c_dev *dev);
 
 #if IS_ENABLED(CONFIG_I2C_DESIGNWARE_AMDPSP)
 int i2c_dw_amdpsp_probe_lock_support(struct dw_i2c_dev *dev);
-void i2c_dw_amdpsp_remove_lock_support(struct dw_i2c_dev *dev);
 #endif
 
 int i2c_dw_validate_speed(struct dw_i2c_dev *dev);
diff --git a/drivers/i2c/busses/i2c-designware-platdrv.c b/drivers/i2c/busses/i2c-designware-platdrv.c
index 74182db03a88..89ad88c54754 100644
--- a/drivers/i2c/busses/i2c-designware-platdrv.c
+++ b/drivers/i2c/busses/i2c-designware-platdrv.c
@@ -214,7 +214,6 @@ static const struct i2c_dw_semaphore_callbacks i2c_dw_semaphore_cb_table[] = {
 #ifdef CONFIG_I2C_DESIGNWARE_AMDPSP
 	{
 		.probe = i2c_dw_amdpsp_probe_lock_support,
-		.remove = i2c_dw_amdpsp_remove_lock_support,
 	},
 #endif
 	{}
diff --git a/include/linux/psp-platform-access.h b/include/linux/psp-platform-access.h
index 1b661341d8f3..75da8f5f7ad8 100644
--- a/include/linux/psp-platform-access.h
+++ b/include/linux/psp-platform-access.h
@@ -7,6 +7,7 @@
 
 enum psp_platform_access_msg {
 	PSP_CMD_NONE = 0x0,
+	PSP_I2C_REQ_BUS_CMD = 0x64,
 };
 
 struct psp_req_buffer_hdr {
-- 
2.34.1

[PATCH v7 6/6] i2c: designware: Add doorbell support for Mendocino

[Mario Limonciello]

Mendocino and later platform don't use the platform feature mailbox for
communication for I2C arbitration, they rely upon ringing a doorbell.

Detect the platform by the device ID of the root port and choose the
appropriate method.

Link: https://lore.kernel.org/linux-i2c/20220916131854.687371-3-jsd@semihalf.com/
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
---
v6->v7:
 * Add missing pci_dev_put()
 * s/mendocino/doorbell/
v5->v6:
 * Handle Mendocino busy code like Cezanne
v4->v5:
 * Poll for busy
 * Rename to mendocino
 * Add explicit dependency on PCI
v3->v4:
 * Adjust to use PCI device ID and function pointers instead
v2->v3:
 * Use CPU ID rather than ACPI ID, this will be pushed to a later patch
v1->v2:
 * New patch
---
 drivers/i2c/busses/Kconfig                 |  1 +
 drivers/i2c/busses/i2c-designware-amdpsp.c | 26 +++++++++++++++++++++-
 2 files changed, 26 insertions(+), 1 deletion(-)

diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index 0d47d4eeb2a4..63bc52b85854 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig
@@ -568,6 +568,7 @@ config I2C_DESIGNWARE_AMDPSP
 	bool "AMD PSP I2C semaphore support"
 	depends on ACPI
 	depends on I2C_DESIGNWARE_PLATFORM
+	depends on PCI
 	imply CRYPTO_DEV_SP_PSP
 	imply CRYPTO_DEV_CCP_DD
 	help
diff --git a/drivers/i2c/busses/i2c-designware-amdpsp.c b/drivers/i2c/busses/i2c-designware-amdpsp.c
index 37d7d8cd05c0..f62d7c081f91 100644
--- a/drivers/i2c/busses/i2c-designware-amdpsp.c
+++ b/drivers/i2c/busses/i2c-designware-amdpsp.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 
 #include <linux/i2c.h>
+#include <linux/pci.h>
 #include <linux/psp-platform-access.h>
 #include <linux/psp.h>
 #include <linux/workqueue.h>
@@ -32,6 +33,8 @@ static u32 psp_i2c_access_count;
 static bool psp_i2c_mbox_fail;
 static struct device *psp_i2c_dev;
 
+static int (*_psp_send_i2c_req)(struct psp_i2c_req *req);
+
 /* Helper to verify status returned by PSP */
 static int check_i2c_req_sts(struct psp_i2c_req *req)
 {
@@ -72,6 +75,17 @@ static int psp_send_i2c_req_cezanne(struct psp_i2c_req *req)
 	return ret;
 }
 
+static int psp_send_i2c_req_doorbell(struct psp_i2c_req *req)
+{
+	int ret;
+
+	ret = psp_ring_platform_doorbell(req->type, &req->hdr.status);
+	if (ret == -EIO)
+		return check_i2c_req_sts(req);
+
+	return ret;
+}
+
 static int psp_send_i2c_req(enum psp_i2c_req_type i2c_req_type)
 {
 	struct psp_i2c_req *req;
@@ -87,7 +101,7 @@ static int psp_send_i2c_req(enum psp_i2c_req_type i2c_req_type)
 	req->type = i2c_req_type;
 
 	start = jiffies;
-	ret = read_poll_timeout(psp_send_i2c_req_cezanne, status,
+	ret = read_poll_timeout(_psp_send_i2c_req, status,
 				(status != -EBUSY),
 				PSP_I2C_REQ_RETRY_DELAY_US,
 				PSP_I2C_REQ_RETRY_CNT * PSP_I2C_REQ_RETRY_DELAY_US,
@@ -262,6 +276,8 @@ static const struct i2c_lock_operations i2c_dw_psp_lock_ops = {
 
 int i2c_dw_amdpsp_probe_lock_support(struct dw_i2c_dev *dev)
 {
+	struct pci_dev *rdev;
+
 	if (!IS_REACHABLE(CONFIG_CRYPTO_DEV_CCP_DD))
 		return -ENODEV;
 
@@ -275,6 +291,14 @@ int i2c_dw_amdpsp_probe_lock_support(struct dw_i2c_dev *dev)
 	if (psp_i2c_dev)
 		return -EEXIST;
 
+	/* Cezanne uses platform mailbox, Mendocino and later use doorbell */
+	rdev = pci_get_domain_bus_and_slot(0, 0, PCI_DEVFN(0, 0));
+	if (rdev->device == 0x1630)
+		_psp_send_i2c_req = psp_send_i2c_req_cezanne;
+	else
+		_psp_send_i2c_req = psp_send_i2c_req_doorbell;
+	pci_dev_put(rdev);
+
 	if (psp_check_platform_access_status())
 		return -EPROBE_DEFER;
 
-- 
2.34.1

Re: [PATCH v7 5/6] i2c: designware: Use PCI PSP driver for communication

[Jarkko Nikula]

On 3/30/23 01:07, Mario Limonciello wrote:
> Currently the PSP semaphore communication base address is discovered
> by using an MSR that is not architecturally guaranteed for future
> platforms.  Also the mailbox that is utilized for communication with
> the PSP may have other consumers in the kernel, so it's better to
> make all communication go through a single driver.
> 
> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
> ---
> v6->v7:
>   * Also imply CRYPTO_DEV_CCP_DD to fix build errors
>   * Fix error message acquire/release inversion
> v5->v6:
>   * Drop now unnecessary asm/msr.h header
>   * Fix IS_REACHABLE
>   * Drop tags
>   * Fix status code handling for Cezanne
> v4->v5:
>   * Pick up tags
> v3->v4:
>   * Pick up tags
> v1->v2:
>   * Fix Kconfig to use imply
>   * Use IS_REACHABLE
> ---
>   drivers/i2c/busses/Kconfig                  |   3 +-
>   drivers/i2c/busses/i2c-designware-amdpsp.c  | 177 +++-----------------
>   drivers/i2c/busses/i2c-designware-core.h    |   1 -
>   drivers/i2c/busses/i2c-designware-platdrv.c |   1 -
>   include/linux/psp-platform-access.h         |   1 +
>   5 files changed, 29 insertions(+), 154 deletions(-)
> 
One note below in case there is a need to have another version of if you 
want. Not a show stopper for this.

Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>

> @@ -214,7 +95,7 @@ static int psp_send_i2c_req(enum psp_i2c_req_type i2c_req_type)
>   	if (ret) {
>   		dev_err(psp_i2c_dev, "Timed out waiting for PSP to %s I2C bus\n",
>   			(i2c_req_type == PSP_I2C_REQ_ACQUIRE) ?
> -			"release" : "acquire");
> +			"acquire" : "release");
>   		goto cleanup;
>   	}
>   
This looks like worth of being an own patch. Maybe even for the 
linux-stable. I think it's very useful to have an error message to show 
correct information what operation actually failed.

Re: [PATCH v7 6/6] i2c: designware: Add doorbell support for Mendocino

[Jarkko Nikula]

On 3/30/23 01:07, Mario Limonciello wrote:
> Mendocino and later platform don't use the platform feature mailbox for
> communication for I2C arbitration, they rely upon ringing a doorbell.
> 
> Detect the platform by the device ID of the root port and choose the
> appropriate method.
> 
> Link: https://lore.kernel.org/linux-i2c/20220916131854.687371-3-jsd@semihalf.com/
> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
> ---
> v6->v7:
>   * Add missing pci_dev_put()
>   * s/mendocino/doorbell/
> v5->v6:
>   * Handle Mendocino busy code like Cezanne
> v4->v5:
>   * Poll for busy
>   * Rename to mendocino
>   * Add explicit dependency on PCI
> v3->v4:
>   * Adjust to use PCI device ID and function pointers instead
> v2->v3:
>   * Use CPU ID rather than ACPI ID, this will be pushed to a later patch
> v1->v2:
>   * New patch
> ---
>   drivers/i2c/busses/Kconfig                 |  1 +
>   drivers/i2c/busses/i2c-designware-amdpsp.c | 26 +++++++++++++++++++++-
>   2 files changed, 26 insertions(+), 1 deletion(-)
> 
Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>

Re: [PATCH v7 5/6] i2c: designware: Use PCI PSP driver for communication

[Jan Dąbroś]

pt., 31 mar 2023 o 13:53 Jarkko Nikula <jarkko.nikula@linux.intel.com>
napisał(a):
>
> On 3/30/23 01:07, Mario Limonciello wrote:
> > Currently the PSP semaphore communication base address is discovered
> > by using an MSR that is not architecturally guaranteed for future
> > platforms.  Also the mailbox that is utilized for communication with
> > the PSP may have other consumers in the kernel, so it's better to
> > make all communication go through a single driver.
> >
> > Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
> > ---
> > v6->v7:
> >   * Also imply CRYPTO_DEV_CCP_DD to fix build errors
> >   * Fix error message acquire/release inversion
> > v5->v6:
> >   * Drop now unnecessary asm/msr.h header
> >   * Fix IS_REACHABLE
> >   * Drop tags
> >   * Fix status code handling for Cezanne
> > v4->v5:
> >   * Pick up tags
> > v3->v4:
> >   * Pick up tags
> > v1->v2:
> >   * Fix Kconfig to use imply
> >   * Use IS_REACHABLE
> > ---
> >   drivers/i2c/busses/Kconfig                  |   3 +-
> >   drivers/i2c/busses/i2c-designware-amdpsp.c  | 177 +++-----------------
> >   drivers/i2c/busses/i2c-designware-core.h    |   1 -
> >   drivers/i2c/busses/i2c-designware-platdrv.c |   1 -
> >   include/linux/psp-platform-access.h         |   1 +
> >   5 files changed, 29 insertions(+), 154 deletions(-)
> >
> One note below in case there is a need to have another version of if you
> want. Not a show stopper for this.
>
> Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
>
> > @@ -214,7 +95,7 @@ static int psp_send_i2c_req(enum psp_i2c_req_type i2c_req_type)
> >       if (ret) {
> >               dev_err(psp_i2c_dev, "Timed out waiting for PSP to %s I2C bus\n",
> >                       (i2c_req_type == PSP_I2C_REQ_ACQUIRE) ?
> > -                     "release" : "acquire");
> > +                     "acquire" : "release");
> >               goto cleanup;
> >       }
> >
> This looks like worth of being an own patch. Maybe even for the
> linux-stable. I think it's very useful to have an error message to show
> correct information what operation actually failed.

Msg here is "TImed out waiting for _PSP to_ %s I2C bus" - when x86
wants to ACQUIRE the bus (i2c_req_type == PSP_I2C_REQ_ACQUIRE) then
PSP needs to RELEASE it and vice versa. If you think logic here is not
straightforward and should be adjusted, then we need to change the
whole sentence.

RE: [PATCH v7 5/6] i2c: designware: Use PCI PSP driver for communication

[Limonciello, Mario]

[Public]



> -----Original Message-----
> From: Jan Dąbroś <jsd@semihalf.com>
> Sent: Friday, March 31, 2023 07:04
> To: Jarkko Nikula <jarkko.nikula@linux.intel.com>
> Cc: Limonciello, Mario <Mario.Limonciello@amd.com>; Grzegorz Bernacki
> <gjb@semihalf.com>; Mark Hasemeyer <markhas@chromium.org>; Andy
> Shevchenko <andriy.shevchenko@linux.intel.com>; Mika Westerberg
> <mika.westerberg@linux.intel.com>; Held, Felix <Felix.Held@amd.com>;
> linux-i2c@vger.kernel.org; linux-kernel@vger.kernel.org
> Subject: Re: [PATCH v7 5/6] i2c: designware: Use PCI PSP driver for
> communication
> 
> pt., 31 mar 2023 o 13:53 Jarkko Nikula <jarkko.nikula@linux.intel.com>
> napisał(a):
> >
> > On 3/30/23 01:07, Mario Limonciello wrote:
> > > Currently the PSP semaphore communication base address is discovered
> > > by using an MSR that is not architecturally guaranteed for future
> > > platforms.  Also the mailbox that is utilized for communication with
> > > the PSP may have other consumers in the kernel, so it's better to
> > > make all communication go through a single driver.
> > >
> > > Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
> > > ---
> > > v6->v7:
> > >   * Also imply CRYPTO_DEV_CCP_DD to fix build errors
> > >   * Fix error message acquire/release inversion
> > > v5->v6:
> > >   * Drop now unnecessary asm/msr.h header
> > >   * Fix IS_REACHABLE
> > >   * Drop tags
> > >   * Fix status code handling for Cezanne
> > > v4->v5:
> > >   * Pick up tags
> > > v3->v4:
> > >   * Pick up tags
> > > v1->v2:
> > >   * Fix Kconfig to use imply
> > >   * Use IS_REACHABLE
> > > ---
> > >   drivers/i2c/busses/Kconfig                  |   3 +-
> > >   drivers/i2c/busses/i2c-designware-amdpsp.c  | 177 +++-----------------
> > >   drivers/i2c/busses/i2c-designware-core.h    |   1 -
> > >   drivers/i2c/busses/i2c-designware-platdrv.c |   1 -
> > >   include/linux/psp-platform-access.h         |   1 +
> > >   5 files changed, 29 insertions(+), 154 deletions(-)
> > >
> > One note below in case there is a need to have another version of if you
> > want. Not a show stopper for this.
> >
> > Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
> >
> > > @@ -214,7 +95,7 @@ static int psp_send_i2c_req(enum
> psp_i2c_req_type i2c_req_type)
> > >       if (ret) {
> > >               dev_err(psp_i2c_dev, "Timed out waiting for PSP to %s I2C
> bus\n",
> > >                       (i2c_req_type == PSP_I2C_REQ_ACQUIRE) ?
> > > -                     "release" : "acquire");
> > > +                     "acquire" : "release");
> > >               goto cleanup;
> > >       }
> > >
> > This looks like worth of being an own patch. Maybe even for the
> > linux-stable. I think it's very useful to have an error message to show
> > correct information what operation actually failed.
> 
> Msg here is "TImed out waiting for _PSP to_ %s I2C bus" - when x86
> wants to ACQUIRE the bus (i2c_req_type == PSP_I2C_REQ_ACQUIRE) then
> PSP needs to RELEASE it and vice versa. If you think logic here is not
> straightforward and should be adjusted, then we need to change the
> whole sentence.

Thanks for clarifying it.  We should just drop this hunk then, it was correct
previously.

I'll wait for Mark or Grzegorz testing results to spin again with the tags
And dropping this hunk.

Re: [PATCH v7 1/6] crypto: ccp: Drop extra doorbell checks

[Mark Hasemeyer]

Reviewed-by: Mark Hasemeyer <markhas@chromium.org>

Re: [PATCH v7 3/6] crypto: ccp: Return doorbell status code as an argument

[Mark Hasemeyer]

Reviewed-by: Mark Hasemeyer <markhas@chromium.org>

Re: [PATCH v7 4/6] crypto: ccp: Use lower 8 bytes to communicate with doorbell command register

[Mark Hasemeyer]

> Unlike other command registers used by the PSP, only the lower 8 bytes are
> used for communication for both command and status of the command.
nit; bytes -> bits

Reviewed-by: Mark Hasemeyer <markhas@chromium.org>

Re: [PATCH v7 5/6] i2c: designware: Use PCI PSP driver for communication

[Mark Hasemeyer]

Reviewed-by: Mark Hasemeyer <markhas@chromium.org>

Re: [PATCH v7 6/6] i2c: designware: Add doorbell support for Mendocino

[Mark Hasemeyer]

Reviewed-by: Mark Hasemeyer <markhas@chromium.org>

Re: [PATCH v7 0/6] Use CCP driver to handle PSP I2C arbitration

[Mark Hasemeyer]

Tested-by: Mark Hasemeyer <markhas@chromium.org>

Re: [PATCH v7 5/6] i2c: designware: Use PCI PSP driver for communication

[kernel test robot]

Hi Mario,

Thank you for the patch! Yet something to improve:

[auto build test ERROR on 9117e682b8b79f7b5e2517fd28d42757d3e8b860]

url:    https://github.com/intel-lab-lkp/linux/commits/Mario-Limonciello/crypto-ccp-Drop-extra-doorbell-checks/20230330-060959
base:   9117e682b8b79f7b5e2517fd28d42757d3e8b860
patch link:    https://lore.kernel.org/r/20230329220753.7741-6-mario.limonciello%40amd.com
patch subject: [PATCH v7 5/6] i2c: designware: Use PCI PSP driver for communication
config: i386-allyesconfig (https://download.01.org/0day-ci/archive/20230401/202304011558.PgIDB9Fr-lkp@intel.com/config)
compiler: gcc-11 (Debian 11.3.0-8) 11.3.0
reproduce (this is a W=1 build):
        # https://github.com/intel-lab-lkp/linux/commit/0bb8c75c296f066a77162eacc6170c0d8fbcf0b6
        git remote add linux-review https://github.com/intel-lab-lkp/linux
        git fetch --no-tags linux-review Mario-Limonciello/crypto-ccp-Drop-extra-doorbell-checks/20230330-060959
        git checkout 0bb8c75c296f066a77162eacc6170c0d8fbcf0b6
        # save the config file
        mkdir build_dir && cp config build_dir/.config
        make W=1 O=build_dir ARCH=i386 olddefconfig
        make W=1 O=build_dir ARCH=i386 SHELL=/bin/bash

If you fix the issue, kindly add following tag where applicable
| Reported-by: kernel test robot <lkp@intel.com>
| Link: https://lore.kernel.org/oe-kbuild-all/202304011558.PgIDB9Fr-lkp@intel.com/

All errors (new ones prefixed by >>):

   ld: drivers/i2c/busses/i2c-designware-amdpsp.o: in function `psp_send_i2c_req':
>> i2c-designware-amdpsp.c:(.text+0x77): undefined reference to `psp_send_platform_access_msg'
   ld: i2c-designware-amdpsp.c:(.text+0xbb): undefined reference to `psp_send_platform_access_msg'
   ld: drivers/i2c/busses/i2c-designware-amdpsp.o: in function `i2c_dw_amdpsp_probe_lock_support':
>> i2c-designware-amdpsp.c:(.text+0x400): undefined reference to `psp_check_platform_access_status'

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests

Re: [PATCH v7 6/6] i2c: designware: Add doorbell support for Mendocino

[kernel test robot]

Hi Mario,

Thank you for the patch! Yet something to improve:

[auto build test ERROR on 9117e682b8b79f7b5e2517fd28d42757d3e8b860]

url:    https://github.com/intel-lab-lkp/linux/commits/Mario-Limonciello/crypto-ccp-Drop-extra-doorbell-checks/20230330-060959
base:   9117e682b8b79f7b5e2517fd28d42757d3e8b860
patch link:    https://lore.kernel.org/r/20230329220753.7741-7-mario.limonciello%40amd.com
patch subject: [PATCH v7 6/6] i2c: designware: Add doorbell support for Mendocino
config: i386-allyesconfig (https://download.01.org/0day-ci/archive/20230402/202304020338.c4EJ0lmB-lkp@intel.com/config)
compiler: gcc-11 (Debian 11.3.0-8) 11.3.0
reproduce (this is a W=1 build):
        # https://github.com/intel-lab-lkp/linux/commit/9696fd5743f2be39fab0ac256fad8dd9eae9930d
        git remote add linux-review https://github.com/intel-lab-lkp/linux
        git fetch --no-tags linux-review Mario-Limonciello/crypto-ccp-Drop-extra-doorbell-checks/20230330-060959
        git checkout 9696fd5743f2be39fab0ac256fad8dd9eae9930d
        # save the config file
        mkdir build_dir && cp config build_dir/.config
        make W=1 O=build_dir ARCH=i386 olddefconfig
        make W=1 O=build_dir ARCH=i386 SHELL=/bin/bash

If you fix the issue, kindly add following tag where applicable
| Reported-by: kernel test robot <lkp@intel.com>
| Link: https://lore.kernel.org/oe-kbuild-all/202304020338.c4EJ0lmB-lkp@intel.com/

All errors (new ones prefixed by >>):

   ld: drivers/i2c/busses/i2c-designware-amdpsp.o: in function `psp_send_i2c_req_doorbell':
>> i2c-designware-amdpsp.c:(.text+0x12): undefined reference to `psp_ring_platform_doorbell'
   ld: drivers/i2c/busses/i2c-designware-amdpsp.o: in function `psp_send_i2c_req_cezanne':
   i2c-designware-amdpsp.c:(.text+0x67): undefined reference to `psp_send_platform_access_msg'
   ld: drivers/i2c/busses/i2c-designware-amdpsp.o: in function `i2c_dw_amdpsp_probe_lock_support':
   i2c-designware-amdpsp.c:(.text+0x49d): undefined reference to `psp_check_platform_access_status'

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests