Skip to contentSkip to conversation

hid: intel-ish-hid: ipc: enable OOB support for EHL

PATCH Applied Zhang Lixu 3 participants 9 messages Dec 16, 2020 – Jan 6, 2021

Subsystems (2)
lore
Share & tools
lore

Zhang Lixu

author
#0 lore raw

cover hid: intel-ish-hid: ipc: enable OOB support for EHL

The EHL (Elkhart Lake) based platforms provide a OOB (Out of band) service, which allows wakup device when the system is in S5 (Soft-Off state). This OOB service can be enabled/disabled from BIOS settings.

These two patches is to enable this feature for EHL platform.

We have tested these patches on both ISH platforms and EHL platforms, it works fine.

Zhang Lixu (2):
hid: intel-ish-hid: ipc: finish power flow for EHL OOB
hid: intel-ish-hid: ipc: Address EHL Sx resume issues

 drivers/hid/intel-ish-hid/ipc/hw-ish.h  |  1 +
 drivers/hid/intel-ish-hid/ipc/ipc.c     | 27 +++++++++++++
 drivers/hid/intel-ish-hid/ipc/pci-ish.c | 54 ++++++++++++++++++++++++-
 3 files changed, 81 insertions(+), 1 deletion(-)
Signature · 1 line
2.25.1
author
#1 lore raw

PATCH 1/2 hid: intel-ish-hid: ipc: finish power flow for EHL OOB

The EHL (Elkhart Lake) based platforms provide a OOB (Out of band) service, which allows wakup device when the system is in S5 (Soft-Off state). This OOB service can be enabled/disabled from BIOS settings. When enabled, the ISH device gets PME wake capability. To enable PME wakeup, driver also needs to enable ACPI GPE bit.

Once wakeup, BIOS will clear the wakeup bit to identify wakeup is successful. So driver need to re-enable it in resume function to keep the next wakeup capability.

Since this feature is only present on EHL, we use EHL PCI device id to enable this feature.

---
 drivers/hid/intel-ish-hid/ipc/pci-ish.c | 48 +++++++++++++++++++++++++
 1 file changed, 48 insertions(+)
drivers/hid/intel-ish-hid/ipc/pci-ish.c +48 -0
index c6d48a8648b7..c9c5488e44cb 100644
@@ -5,6 +5,7 @@
  * Copyright (c) 2014-2016, Intel Corporation.
  */
 
+#include <linux/acpi.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
 #include <linux/kernel.h>
@@ -111,6 +112,42 @@ static inline bool ish_should_leave_d0i3(struct pci_dev *pdev)
 	return !pm_resume_via_firmware() || pdev->device == CHV_DEVICE_ID;
 }
 
+static int enable_gpe(struct device *dev)
+{
+#ifdef CONFIG_ACPI
+	acpi_status acpi_sts;
+	struct acpi_device *adev;
+	struct acpi_device_wakeup *wakeup;
+
+	adev = ACPI_COMPANION(dev);
+	if (!adev) {
+		dev_err(dev, "get acpi handle failed\n");
+		return -ENODEV;
+	}
+	wakeup = &adev->wakeup;
+
+	acpi_sts = acpi_enable_gpe(wakeup->gpe_device, wakeup->gpe_number);
+	if (ACPI_FAILURE(acpi_sts)) {
+		dev_err(dev, "enable ose_gpe failed\n");
+		return -EIO;
+	}
+
+	return 0;
+#else
+	return -ENODEV;
+#endif
+}
+
+static void enable_pme_wake(struct pci_dev *pdev)
+{
+	if ((pci_pme_capable(pdev, PCI_D0) ||
+	     pci_pme_capable(pdev, PCI_D3hot) ||
+	     pci_pme_capable(pdev, PCI_D3cold)) && !enable_gpe(&pdev->dev)) {
+		pci_pme_active(pdev, true);
+		dev_dbg(&pdev->dev, "ish ipc driver pme wake enabled\n");
+	}
+}
+
 /**
  * ish_probe() - PCI driver probe callback
  * @pdev:	pci device
@@ -179,6 +216,10 @@ static int ish_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	init_waitqueue_head(&ishtp->suspend_wait);
 	init_waitqueue_head(&ishtp->resume_wait);
 
+	/* Enable PME for EHL */
+	if (pdev->device == EHL_Ax_DEVICE_ID)
+		enable_pme_wake(pdev);
+
 	ret = ish_init(ishtp);
 	if (ret)
 		return ret;
@@ -317,6 +358,13 @@ static int __maybe_unused ish_resume(struct device *device)
 	struct pci_dev *pdev = to_pci_dev(device);
 	struct ishtp_device *dev = pci_get_drvdata(pdev);
 
+	/* add this to finish power flow for EHL */
+	if (dev->pdev->device == EHL_Ax_DEVICE_ID) {
+		pci_set_power_state(pdev, PCI_D0);
+		enable_pme_wake(pdev);
+		dev_dbg(dev->devc, "set power state to D0 for ehl\n");
+	}
+
 	ish_resume_device = device;
 	dev->resume_flag = 1;
 
Signature · 1 line
2.25.1
maintainer Acked-by on v1 1/2
#2 lore raw

Re: [PATCH 1/2] hid: intel-ish-hid: ipc: finish power flow for EHL OOB

Quoted 12 lines · Zhang Lixu

The EHL (Elkhart Lake) based platforms provide a OOB (Out of band) service, which allows wakup device when the system is in S5 (Soft-Off state). This OOB service can be enabled/disabled from BIOS settings. When
enabled, the ISH device gets PME wake capability. To enable PME wakeup,
driver also needs to enable ACPI GPE bit.

Once wakeup, BIOS will clear the wakeup bit to identify wakeup is successful. So driver need to re-enable it in resume function to keep the next wakeup capability.

Quoted 8 lines

Since this feature is only present on EHL, we use EHL PCI device id to
enable this feature.

Quoted 94 lines

---
 drivers/hid/intel-ish-hid/ipc/pci-ish.c | 48
+++++++++++++++++++++++++

 1 file changed, 48 insertions(+)
drivers/hid/intel-ish-hid/ipc/pci-ish.c +0 -0
index c6d48a8648b7..c9c5488e44cb 100644
@@ -5,6 +5,7 @@
   * Copyright (c) 2014-2016, Intel Corporation.
   */


+#include <linux/acpi.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
 #include <linux/kernel.h>

drivers/hid/intel-ish-hid/ipc/pci-ish.c +0 -0
@@ -111,6 +112,42 @@ static inline bool ish_should_leave_d0i3(struct
 pci_dev *pdev)

        return !pm_resume_via_firmware() || pdev->device == CHV_DEVICE_ID;
 }

drivers/hid/intel-ish-hid/ipc/pci-ish.c +15 -0
+static int enable_gpe(struct device *dev)
+{
+#ifdef CONFIG_ACPI
+       acpi_status acpi_sts;
+       struct acpi_device *adev;
+       struct acpi_device_wakeup *wakeup;
+
+       adev = ACPI_COMPANION(dev);
+       if (!adev) {
+               dev_err(dev, "get acpi handle failed\n");
+               return -ENODEV;
+       }
+       wakeup = &adev->wakeup;
+
+       acpi_sts = acpi_enable_gpe(wakeup->gpe_device, wakeup-
Quoted 1 line

gpe_number);

+       if (ACPI_FAILURE(acpi_sts)) {
+               dev_err(dev, "enable ose_gpe failed\n"); +               return -EIO;
+       }

drivers/hid/intel-ish-hid/ipc/pci-ish.c +12 -0
+
+       return 0;
+#else
+       return -ENODEV;
+#endif
+}
+
+static void enable_pme_wake(struct pci_dev *pdev)
+{
+       if ((pci_pme_capable(pdev, PCI_D0) ||
+            pci_pme_capable(pdev, PCI_D3hot) ||
+            pci_pme_capable(pdev, PCI_D3cold)) && !enable_gpe(&pdev-
Quoted 1 line

dev)) {

+               pci_pme_active(pdev, true);
+               dev_dbg(&pdev->dev, "ish ipc driver pme wake enabled\n");
+       }

drivers/hid/intel-ish-hid/ipc/pci-ish.c +2 -0
+}
+

 /**
  * ish_probe() - PCI driver probe callback
  * @pdev:      pci device

drivers/hid/intel-ish-hid/ipc/pci-ish.c +0 -0
@@ -179,6 +216,10 @@ static int ish_probe(struct pci_dev *pdev, const
 struct pci_device_id *ent)
         init_waitqueue_head(&ishtp->suspend_wait);
         init_waitqueue_head(&ishtp->resume_wait);


+       /* Enable PME for EHL */
+       if (pdev->device == EHL_Ax_DEVICE_ID)
+               enable_pme_wake(pdev);
+
        ret = ish_init(ishtp);
        if (ret)
                return ret;

drivers/hid/intel-ish-hid/ipc/pci-ish.c +0 -0
@@ -317,6 +358,13 @@ static int __maybe_unused ish_resume(struct
 device *device)
         struct pci_dev *pdev = to_pci_dev(device);
         struct ishtp_device *dev = pci_get_drvdata(pdev);


+       /* add this to finish power flow for EHL */ +       if (dev->pdev->device == EHL_Ax_DEVICE_ID) { +               pci_set_power_state(pdev, PCI_D0); +               enable_pme_wake(pdev);
+               dev_dbg(dev->devc, "set power state to D0 for ehl\n");
+       }
+
        ish_resume_device = device; dev->resume_flag = 1;

author
#3 lore raw

PATCH 2/2 hid: intel-ish-hid: ipc: Address EHL Sx resume issues

In reply to Zhang Lixu

When OOB is disabled, FW will be power gated when system is in S3/S4/S5 which is the same behavior with legacy ISH FW.
When OOB is enabled, FW will always power on which is totally different comparing to legacy ISH FW.

So NO_D3 flag is not enough to check FW's status after resume. Here we can use IPC FW status register to check host link status. If it is false, it means FW get reset after power gated, need go through the whole initialization flow; If it is true, it means FW is alive, just set host ready bit to let fw know host is up.

---
 drivers/hid/intel-ish-hid/ipc/hw-ish.h  |  1 +
 drivers/hid/intel-ish-hid/ipc/ipc.c     | 27 +++++++++++++++++++++++++
 drivers/hid/intel-ish-hid/ipc/pci-ish.c |  6 +++++-
 3 files changed, 33 insertions(+), 1 deletion(-)
drivers/hid/intel-ish-hid/ipc/hw-ish.h +1 -0
index 1fb294ca463e..111ad259ba74 100644
@@ -81,5 +81,6 @@ struct ishtp_device *ish_dev_init(struct pci_dev *pdev);
 int ish_hw_start(struct ishtp_device *dev);
 void ish_device_disable(struct ishtp_device *dev);
 int ish_disable_dma(struct ishtp_device *dev);
+void ish_set_host_ready(struct ishtp_device *dev);
 
 #endif /* _ISHTP_HW_ISH_H_ */
drivers/hid/intel-ish-hid/ipc/ipc.c +27 -0
index a45ac7fa417b..47bbeb8b492b 100644
@@ -193,6 +193,33 @@ static void ish_clr_host_rdy(struct ishtp_device *dev)
 	ish_reg_write(dev, IPC_REG_HOST_COMM, host_status);
 }
 
+static bool ish_chk_host_rdy(struct ishtp_device *dev)
+{
+	uint32_t host_status = ish_reg_read(dev, IPC_REG_HOST_COMM);
+
+	return (host_status & IPC_HOSTCOMM_READY_BIT);
+}
+
+/**
+ * ish_set_host_ready() - reconfig ipc host registers
+ * @dev: ishtp device pointer
+ *
+ * Set host to ready state
+ * This API is called in some case:
+ *    fw is still on, but ipc is powered down.
+ *    such as OOB case.
+ *
+ * Return: 0 for success else error fault code
+ */
+void ish_set_host_ready(struct ishtp_device *dev)
+{
+	if (ish_chk_host_rdy(dev))
+		return;
+
+	ish_set_host_rdy(dev);
+	set_host_ready(dev);
+}
+
 /**
  * _ishtp_read_hdr() - Read message header
  * @dev: ISHTP device pointer
drivers/hid/intel-ish-hid/ipc/pci-ish.c +5 -1
index c9c5488e44cb..8cb40696984a 100644
@@ -259,11 +259,15 @@ static void __maybe_unused ish_resume_handler(struct work_struct *work)
 {
 	struct pci_dev *pdev = to_pci_dev(ish_resume_device);
 	struct ishtp_device *dev = pci_get_drvdata(pdev);
+	uint32_t fwsts = dev->ops->get_fw_status(dev);
 	int ret;
 
-	if (ish_should_leave_d0i3(pdev) && !dev->suspend_flag) {
+	if (ish_should_leave_d0i3(pdev) && !dev->suspend_flag
+			&& IPC_IS_ISH_ILUP(fwsts)) {
 		disable_irq_wake(pdev->irq);
 
+		ish_set_host_ready(dev);
+
 		ishtp_send_resume(dev);
 
 		/* Waiting to get resume response */
Signature · 1 line
2.25.1
maintainer Acked-by on v1 2/2
#4 lore raw

Re: [PATCH 2/2] hid: intel-ish-hid: ipc: Address EHL Sx resume issues

Quoted 15 lines · Zhang Lixu

When OOB is disabled, FW will be power gated when system is in S3/S4/S5
which is the same behavior with legacy ISH FW.
When OOB is enabled, FW will always power on which is totally different
comparing to legacy ISH FW.

So NO_D3 flag is not enough to check FW's status after resume. Here we can use IPC FW status register to check host link status. If it is false, it means FW get reset after power gated, need go through
the whole initialization flow; If it is true, it means FW is alive, just
set host ready bit to let fw know host is up.

Quoted 4 lines
Quoted 83 lines

---
 drivers/hid/intel-ish-hid/ipc/hw-ish.h  |  1 + drivers/hid/intel-ish-hid/ipc/ipc.c     | 27
+++++++++++++++++++++++++
 drivers/hid/intel-ish-hid/ipc/pci-ish.c |  6 +++++-

 3 files changed, 33 insertions(+), 1 deletion(-)
drivers/hid/intel-ish-hid/ipc/hw-ish.h +1 -0
index 1fb294ca463e..111ad259ba74 100644
@@ -81,5 +81,6 @@ struct ishtp_device *ish_dev_init(struct pci_dev
 *pdev);
  int ish_hw_start(struct ishtp_device *dev);
  void ish_device_disable(struct ishtp_device *dev);
  int ish_disable_dma(struct ishtp_device *dev);
+void ish_set_host_ready(struct ishtp_device *dev);


 #endif /* _ISHTP_HW_ISH_H_ */

drivers/hid/intel- renamed from drivers/hid/intel-ish-hid/ipc/ipc.c +0 -0

ish-hid/ipc/ipc.c
index a45ac7fa417b..47bbeb8b492b 100644

drivers/hid/intel-ish-hid/ipc/ipc.c +0 -0
@@ -193,6 +193,33 @@ static void ish_clr_host_rdy(struct ishtp_device
 *dev)
         ish_reg_write(dev, IPC_REG_HOST_COMM, host_status);
  }

drivers/hid/intel-ish-hid/ipc/ipc.c +27 -0
+static bool ish_chk_host_rdy(struct ishtp_device *dev)
+{
+       uint32_t host_status = ish_reg_read(dev, IPC_REG_HOST_COMM);
+
+       return (host_status & IPC_HOSTCOMM_READY_BIT);
+}
+
+/**
+ * ish_set_host_ready() - reconfig ipc host registers
+ * @dev: ishtp device pointer
+ *
+ * Set host to ready state
+ * This API is called in some case:
+ *    fw is still on, but ipc is powered down.
+ *    such as OOB case.
+ *
+ * Return: 0 for success else error fault code
+ */
+void ish_set_host_ready(struct ishtp_device *dev)
+{
+       if (ish_chk_host_rdy(dev))
+               return;
+
+       ish_set_host_rdy(dev);
+       set_host_ready(dev);
+}
+

 /**
  * _ishtp_read_hdr() - Read message header
  * @dev: ISHTP device pointer

drivers/hid/intel-ish-hid/ipc/pci-ish.c +1 -0
index c9c5488e44cb..8cb40696984a 100644
@@ -259,11 +259,15 @@ static void __maybe_unused
 ish_resume_handler(struct work_struct *work)
  {
         struct pci_dev *pdev = to_pci_dev(ish_resume_device);
         struct ishtp_device *dev = pci_get_drvdata(pdev);
+       uint32_t fwsts = dev->ops->get_fw_status(dev);
         int ret;


-       if (ish_should_leave_d0i3(pdev) && !dev->suspend_flag) { +       if (ish_should_leave_d0i3(pdev) && !dev->suspend_flag +                       && IPC_IS_ISH_ILUP(fwsts)) { disable_irq_wake(pdev->irq);

+               ish_set_host_ready(dev);
+
                ishtp_send_resume(dev);

                /* Waiting to get resume response */

maintainer
#5 lore raw
In reply to Zhang Lixu
Quoted 9 lines · Zhang Lixu

The EHL (Elkhart Lake) based platforms provide a OOB (Out of band) service, which allows wakup device when the system is in S5 (Soft-Off state). This OOB service can be enabled/disabled from BIOS settings.

These two patches is to enable this feature for EHL platform.

We have tested these patches on both ISH platforms and EHL platforms, it works fine.

Quoted 8 lines

Zhang Lixu (2):
hid: intel-ish-hid: ipc: finish power flow for EHL OOB
hid: intel-ish-hid: ipc: Address EHL Sx resume issues

 drivers/hid/intel-ish-hid/ipc/hw-ish.h  |  1 +
 drivers/hid/intel-ish-hid/ipc/ipc.c     | 27 +++++++++++++
 drivers/hid/intel-ish-hid/ipc/pci-ish.c | 54 ++++++++++++++++++++++++-
 3 files changed, 81 insertions(+), 1 deletion(-)

Srinivas, can I please get your Acked-by / Reviewed-by for this? Thanks,

Signature · 2 lines
Jiri Kosina
SUSE Labs
#6 lore raw
Quoted 18 lines · Jiri Kosina
Quoted 16 lines · Zhang Lixu

The EHL (Elkhart Lake) based platforms provide a OOB (Out of band) service, which allows wakup device when the system is in S5 (Soft- Off
state). This OOB service can be enabled/disabled from BIOS settings.

These two patches is to enable this feature for EHL platform.

We have tested these patches on both ISH platforms and EHL platforms,
it works fine.

Zhang Lixu (2):
  hid: intel-ish-hid: ipc: finish power flow for EHL OOB
  hid: intel-ish-hid: ipc: Address EHL Sx resume issues

Quoted 7 lines
Quoted 5 lines

 drivers/hid/intel-ish-hid/ipc/hw-ish.h  |  1 + drivers/hid/intel-ish-hid/ipc/ipc.c     | 27 +++++++++++++ drivers/hid/intel-ish-hid/ipc/pci-ish.c | 54
++++++++++++++++++++++++-

 3 files changed, 81 insertions(+), 1 deletion(-)

Srinivas, can I please get your Acked-by / Reviewed-by for this?

Acked on individual patches.

Thanks,
Srinivas

Quoted 2 lines

Thanks,

#7 lore raw
Quoted 25 lines · Srinivas Pandruvada
Quoted 24 lines · Jiri Kosina
Quoted 21 lines · Zhang Lixu

The EHL (Elkhart Lake) based platforms provide a OOB (Out of band) service, which allows wakup device when the system is in S5 (Soft- Off
state). This OOB service can be enabled/disabled from BIOS settings.

These two patches is to enable this feature for EHL platform.

We have tested these patches on both ISH platforms and EHL platforms,
it works fine.

Zhang Lixu (2):
  hid: intel-ish-hid: ipc: finish power flow for EHL OOB
  hid: intel-ish-hid: ipc: Address EHL Sx resume issues

 drivers/hid/intel-ish-hid/ipc/hw-ish.h  |  1 + drivers/hid/intel-ish-hid/ipc/ipc.c     | 27 +++++++++++++ drivers/hid/intel-ish-hid/ipc/pci-ish.c | 54
++++++++++++++++++++++++-

 3 files changed, 81 insertions(+), 1 deletion(-)
Quoted 11 lines
Quoted 1 line

Srinivas, can I please get your Acked-by / Reviewed-by for this?

Acked on individual patches.

Thanks,
Srinivas

Quoted 2 lines

Thanks,

Thanks.

maintainer Applied
#8 lore raw
In reply to Zhang Lixu
Quoted 9 lines · Zhang Lixu

The EHL (Elkhart Lake) based platforms provide a OOB (Out of band) service, which allows wakup device when the system is in S5 (Soft-Off state). This OOB service can be enabled/disabled from BIOS settings.

These two patches is to enable this feature for EHL platform.

We have tested these patches on both ISH platforms and EHL platforms, it works fine.

Quoted 8 lines

Zhang Lixu (2):
hid: intel-ish-hid: ipc: finish power flow for EHL OOB
hid: intel-ish-hid: ipc: Address EHL Sx resume issues

 drivers/hid/intel-ish-hid/ipc/hw-ish.h  |  1 +
 drivers/hid/intel-ish-hid/ipc/ipc.c     | 27 +++++++++++++
 drivers/hid/intel-ish-hid/ipc/pci-ish.c | 54 ++++++++++++++++++++++++-
 3 files changed, 81 insertions(+), 1 deletion(-)

Applied to hid.git#for-5.12/intel-ish.

Signature · 2 lines
Jiri Kosina
SUSE Labs