[PATCH] PCI: MSI(X) save/restore for suspend/resume

Add MSI(X) configure sapce save/restore in generic PCI helper.

Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

4 files changed, 246 insertions, 29 deletions

diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index a77e79c8c82..2087a397ef1 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -504,6 +504,201 @@ void pci_scan_msi_device(struct pci_dev *dev)
 		nr_reserved_vectors++;
 }
 
+#ifdef CONFIG_PM
+int pci_save_msi_state(struct pci_dev *dev)
+{
+	int pos, i = 0;
+	u16 control;
+	struct pci_cap_saved_state *save_state;
+	u32 *cap;
+
+	pos = pci_find_capability(dev, PCI_CAP_ID_MSI);
+	if (pos <= 0 || dev->no_msi)
+		return 0;
+
+	pci_read_config_word(dev, msi_control_reg(pos), &control);
+	if (!(control & PCI_MSI_FLAGS_ENABLE))
+		return 0;
+
+	save_state = kzalloc(sizeof(struct pci_cap_saved_state) + sizeof(u32) * 5,
+		GFP_KERNEL);
+	if (!save_state) {
+		printk(KERN_ERR "Out of memory in pci_save_msi_state\n");
+		return -ENOMEM;
+	}
+	cap = &save_state->data[0];
+
+	pci_read_config_dword(dev, pos, &cap[i++]);
+	control = cap[0] >> 16;
+	pci_read_config_dword(dev, pos + PCI_MSI_ADDRESS_LO, &cap[i++]);
+	if (control & PCI_MSI_FLAGS_64BIT) {
+		pci_read_config_dword(dev, pos + PCI_MSI_ADDRESS_HI, &cap[i++]);
+		pci_read_config_dword(dev, pos + PCI_MSI_DATA_64, &cap[i++]);
+	} else
+		pci_read_config_dword(dev, pos + PCI_MSI_DATA_32, &cap[i++]);
+	if (control & PCI_MSI_FLAGS_MASKBIT)
+		pci_read_config_dword(dev, pos + PCI_MSI_MASK_BIT, &cap[i++]);
+	disable_msi_mode(dev, pos, PCI_CAP_ID_MSI);
+	save_state->cap_nr = PCI_CAP_ID_MSI;
+	pci_add_saved_cap(dev, save_state);
+	return 0;
+}
+
+void pci_restore_msi_state(struct pci_dev *dev)
+{
+	int i = 0, pos;
+	u16 control;
+	struct pci_cap_saved_state *save_state;
+	u32 *cap;
+
+	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_MSI);
+	pos = pci_find_capability(dev, PCI_CAP_ID_MSI);
+	if (!save_state || pos <= 0)
+		return;
+	cap = &save_state->data[0];
+
+	control = cap[i++] >> 16;
+	pci_write_config_dword(dev, pos + PCI_MSI_ADDRESS_LO, cap[i++]);
+	if (control & PCI_MSI_FLAGS_64BIT) {
+		pci_write_config_dword(dev, pos + PCI_MSI_ADDRESS_HI, cap[i++]);
+		pci_write_config_dword(dev, pos + PCI_MSI_DATA_64, cap[i++]);
+	} else
+		pci_write_config_dword(dev, pos + PCI_MSI_DATA_32, cap[i++]);
+	if (control & PCI_MSI_FLAGS_MASKBIT)
+		pci_write_config_dword(dev, pos + PCI_MSI_MASK_BIT, cap[i++]);
+	pci_write_config_word(dev, pos + PCI_MSI_FLAGS, control);
+	enable_msi_mode(dev, pos, PCI_CAP_ID_MSI);
+	pci_remove_saved_cap(save_state);
+	kfree(save_state);
+}
+
+int pci_save_msix_state(struct pci_dev *dev)
+{
+	int pos;
+	u16 control;
+	struct pci_cap_saved_state *save_state;
+
+	pos = pci_find_capability(dev, PCI_CAP_ID_MSIX);
+	if (pos <= 0 || dev->no_msi)
+		return 0;
+
+	pci_read_config_word(dev, msi_control_reg(pos), &control);
+	if (!(control & PCI_MSIX_FLAGS_ENABLE))
+		return 0;
+	save_state = kzalloc(sizeof(struct pci_cap_saved_state) + sizeof(u16),
+		GFP_KERNEL);
+	if (!save_state) {
+		printk(KERN_ERR "Out of memory in pci_save_msix_state\n");
+		return -ENOMEM;
+	}
+	*((u16 *)&save_state->data[0]) = control;
+
+	disable_msi_mode(dev, pos, PCI_CAP_ID_MSIX);
+	save_state->cap_nr = PCI_CAP_ID_MSIX;
+	pci_add_saved_cap(dev, save_state);
+	return 0;
+}
+
+void pci_restore_msix_state(struct pci_dev *dev)
+{
+	u16 save;
+	int pos;
+	int vector, head, tail = 0;
+	void __iomem *base;
+	int j;
+	struct msg_address address;
+	struct msg_data data;
+	struct msi_desc *entry;
+	int temp;
+	struct pci_cap_saved_state *save_state;
+
+	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_MSIX);
+	if (!save_state)
+		return;
+	save = *((u16 *)&save_state->data[0]);
+	pci_remove_saved_cap(save_state);
+	kfree(save_state);
+
+	pos = pci_find_capability(dev, PCI_CAP_ID_MSIX);
+	if (pos <= 0)
+		return;
+
+	/* route the table */
+	temp = dev->irq;
+	if (msi_lookup_vector(dev, PCI_CAP_ID_MSIX))
+		return;
+	vector = head = dev->irq;
+	while (head != tail) {
+		entry = msi_desc[vector];
+		base = entry->mask_base;
+		j = entry->msi_attrib.entry_nr;
+
+		msi_address_init(&address);
+		msi_data_init(&data, vector);
+
+		address.lo_address.value &= MSI_ADDRESS_DEST_ID_MASK;
+		address.lo_address.value |= entry->msi_attrib.current_cpu <<
+					MSI_TARGET_CPU_SHIFT;
+
+		writel(address.lo_address.value,
+			base + j * PCI_MSIX_ENTRY_SIZE +
+			PCI_MSIX_ENTRY_LOWER_ADDR_OFFSET);
+		writel(address.hi_address,
+			base + j * PCI_MSIX_ENTRY_SIZE +
+			PCI_MSIX_ENTRY_UPPER_ADDR_OFFSET);
+		writel(*(u32*)&data,
+			base + j * PCI_MSIX_ENTRY_SIZE +
+			PCI_MSIX_ENTRY_DATA_OFFSET);
+
+		tail = msi_desc[vector]->link.tail;
+		vector = tail;
+	}
+	dev->irq = temp;
+
+	pci_write_config_word(dev, msi_control_reg(pos), save);
+	enable_msi_mode(dev, pos, PCI_CAP_ID_MSIX);
+}
+#endif
+
+static void msi_register_init(struct pci_dev *dev, struct msi_desc *entry)
+{
+	struct msg_address address;
+	struct msg_data data;
+	int pos, vector = dev->irq;
+	u16 control;
+
+   	pos = pci_find_capability(dev, PCI_CAP_ID_MSI);
+	pci_read_config_word(dev, msi_control_reg(pos), &control);
+	/* Configure MSI capability structure */
+	msi_address_init(&address);
+	msi_data_init(&data, vector);
+	entry->msi_attrib.current_cpu = ((address.lo_address.u.dest_id >>
+				MSI_TARGET_CPU_SHIFT) & MSI_TARGET_CPU_MASK);
+	pci_write_config_dword(dev, msi_lower_address_reg(pos),
+			address.lo_address.value);
+	if (is_64bit_address(control)) {
+		pci_write_config_dword(dev,
+			msi_upper_address_reg(pos), address.hi_address);
+		pci_write_config_word(dev,
+			msi_data_reg(pos, 1), *((u32*)&data));
+	} else
+		pci_write_config_word(dev,
+			msi_data_reg(pos, 0), *((u32*)&data));
+	if (entry->msi_attrib.maskbit) {
+		unsigned int maskbits, temp;
+		/* All MSIs are unmasked by default, Mask them all */
+		pci_read_config_dword(dev,
+			msi_mask_bits_reg(pos, is_64bit_address(control)),
+			&maskbits);
+		temp = (1 << multi_msi_capable(control));
+		temp = ((temp - 1) & ~temp);
+		maskbits |= temp;
+		pci_write_config_dword(dev,
+			msi_mask_bits_reg(pos, is_64bit_address(control)),
+			maskbits);
+	}
+}
+
 /**
  * msi_capability_init - configure device's MSI capability structure
  * @dev: pointer to the pci_dev data structure of MSI device function
@@ -516,8 +711,6 @@ void pci_scan_msi_device(struct pci_dev *dev)
 static int msi_capability_init(struct pci_dev *dev)
 {
 	struct msi_desc *entry;
-	struct msg_address address;
-	struct msg_data data;
 	int pos, vector;
 	u16 control;
 
@@ -549,33 +742,8 @@ static int msi_capability_init(struct pci_dev *dev)
 	/* Replace with MSI handler */
 	irq_handler_init(PCI_CAP_ID_MSI, vector, entry->msi_attrib.maskbit);
 	/* Configure MSI capability structure */
-	msi_address_init(&address);
-	msi_data_init(&data, vector);
-	entry->msi_attrib.current_cpu = ((address.lo_address.u.dest_id >>
-				MSI_TARGET_CPU_SHIFT) & MSI_TARGET_CPU_MASK);
-	pci_write_config_dword(dev, msi_lower_address_reg(pos),
-			address.lo_address.value);
-	if (is_64bit_address(control)) {
-		pci_write_config_dword(dev,
-			msi_upper_address_reg(pos), address.hi_address);
-		pci_write_config_word(dev,
-			msi_data_reg(pos, 1), *((u32*)&data));
-	} else
-		pci_write_config_word(dev,
-			msi_data_reg(pos, 0), *((u32*)&data));
-	if (entry->msi_attrib.maskbit) {
-		unsigned int maskbits, temp;
-		/* All MSIs are unmasked by default, Mask them all */
-		pci_read_config_dword(dev,
-			msi_mask_bits_reg(pos, is_64bit_address(control)),
-			&maskbits);
-		temp = (1 << multi_msi_capable(control));
-		temp = ((temp - 1) & ~temp);
-		maskbits |= temp;
-		pci_write_config_dword(dev,
-			msi_mask_bits_reg(pos, is_64bit_address(control)),
-			maskbits);
-	}
+	msi_register_init(dev, entry);
+
 	attach_msi_entry(entry, vector);
 	/* Set MSI enabled bits	 */
 	enable_msi_mode(dev, pos, PCI_CAP_ID_MSI);
@@ -731,6 +899,7 @@ int pci_enable_msi(struct pci_dev* dev)
 			vector_irq[dev->irq] = -1;
 			nr_released_vectors--;
 			spin_unlock_irqrestore(&msi_lock, flags);
+			msi_register_init(dev, msi_desc[dev->irq]);
 			enable_msi_mode(dev, pos, PCI_CAP_ID_MSI);
 			return 0;
 		}
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index bea1ad1ad5b..69a617d2182 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -444,6 +444,10 @@ pci_save_state(struct pci_dev *dev)
 	/* XXX: 100% dword access ok here? */
 	for (i = 0; i < 16; i++)
 		pci_read_config_dword(dev, i * 4,&dev->saved_config_space[i]);
+	if ((i = pci_save_msi_state(dev)) != 0)
+		return i;
+	if ((i = pci_save_msix_state(dev)) != 0)
+		return i;
 	return 0;
 }
 
@@ -458,6 +462,8 @@ pci_restore_state(struct pci_dev *dev)
 
 	for (i = 0; i < 16; i++)
 		pci_write_config_dword(dev,i * 4, dev->saved_config_space[i]);
+	pci_restore_msi_state(dev);
+	pci_restore_msix_state(dev);
 	return 0;
 }
 
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index 8f3fb47ea67..30630cbe2fe 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -55,6 +55,17 @@ void pci_no_msi(void);
 static inline void disable_msi_mode(struct pci_dev *dev, int pos, int type) { }
 static inline void pci_no_msi(void) { }
 #endif
+#if defined(CONFIG_PCI_MSI) && defined(CONFIG_PM)
+int pci_save_msi_state(struct pci_dev *dev);
+int pci_save_msix_state(struct pci_dev *dev);
+void pci_restore_msi_state(struct pci_dev *dev);
+void pci_restore_msix_state(struct pci_dev *dev);
+#else
+static inline int pci_save_msi_state(struct pci_dev *dev) { return 0; }
+static inline int pci_save_msix_state(struct pci_dev *dev) { return 0; }
+static inline void pci_restore_msi_state(struct pci_dev *dev) {}
+static inline void pci_restore_msix_state(struct pci_dev *dev) {}
+#endif
 
 extern int pcie_mch_quirk;
 extern struct device_attribute pci_dev_attrs[];
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 0aad5a378e9..15e1675edef 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -100,6 +100,12 @@ enum pci_bus_flags {
 	PCI_BUS_FLAGS_NO_MSI = (pci_bus_flags_t) 1,
 };
 
+struct pci_cap_saved_state {
+	struct hlist_node next;
+	char cap_nr;
+	u32 data[0];
+};
+
 /*
  * The pci_dev structure is used to describe PCI devices.
  */
@@ -159,6 +165,7 @@ struct pci_dev {
 	unsigned int	block_ucfg_access:1;	/* userspace config space access is blocked */
 
 	u32		saved_config_space[16]; /* config space saved at suspend time */
+	struct hlist_head saved_cap_space;
 	struct bin_attribute *rom_attr; /* attribute descriptor for sysfs ROM entry */
 	int rom_attr_enabled;		/* has display of the rom attribute been enabled? */
 	struct bin_attribute *res_attr[DEVICE_COUNT_RESOURCE]; /* sysfs file for resources */
@@ -169,6 +176,30 @@ struct pci_dev {
 #define	to_pci_dev(n) container_of(n, struct pci_dev, dev)
 #define for_each_pci_dev(d) while ((d = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, d)) != NULL)
 
+static inline struct pci_cap_saved_state *pci_find_saved_cap(
+	struct pci_dev *pci_dev,char cap)
+{
+	struct pci_cap_saved_state *tmp;
+	struct hlist_node *pos;
+
+	hlist_for_each_entry(tmp, pos, &pci_dev->saved_cap_space, next) {
+		if (tmp->cap_nr == cap)
+			return tmp;
+	}
+	return NULL;
+}
+
+static inline void pci_add_saved_cap(struct pci_dev *pci_dev,
+	struct pci_cap_saved_state *new_cap)
+{
+	hlist_add_head(&new_cap->next, &pci_dev->saved_cap_space);
+}
+
+static inline void pci_remove_saved_cap(struct pci_cap_saved_state *cap)
+{
+	hlist_del(&cap->next);
+}
+
 /*
  *  For PCI devices, the region numbers are assigned this way:
  *