[SPARC64]: Enable MSI on sun4u Fire PCI-E controllers.

The support code is identical to the hypervisor sun4v stuff,
just replacing the hypervisor calls with register reads and
writes in the Fire controller.

Signed-off-by: David S. Miller <davem@davemloft.net>

arch/sparc64/kernel/irq.c

@@ -406,6 +406,18 @@ static void sun4v_irq_disable(unsigned int virt_irq)
 }
 
 #ifdef CONFIG_PCI_MSI
+static void sun4u_msi_enable(unsigned int virt_irq)
+{
+	sun4u_irq_enable(virt_irq);
+	unmask_msi_irq(virt_irq);
+}
+
+static void sun4u_msi_disable(unsigned int virt_irq)
+{
+	mask_msi_irq(virt_irq);
+	sun4u_irq_disable(virt_irq);
+}
+
 static void sun4v_msi_enable(unsigned int virt_irq)
 {
 	sun4v_irq_enable(virt_irq);
@@ -583,6 +595,17 @@ static struct irq_chip sun4v_irq_ack = {
 };
 
 #ifdef CONFIG_PCI_MSI
+static struct irq_chip sun4u_msi = {
+	.typename	= "sun4u+msi",
+	.mask		= mask_msi_irq,
+	.unmask		= unmask_msi_irq,
+	.enable		= sun4u_msi_enable,
+	.disable	= sun4u_msi_disable,
+	.ack		= run_pre_handler,
+	.end		= sun4u_irq_end,
+	.set_affinity	= sun4u_set_affinity,
+};
+
 static struct irq_chip sun4v_msi = {
 	.typename	= "sun4v+msi",
 	.mask		= mask_msi_irq,
@@ -628,6 +651,7 @@ void irq_install_pre_handler(int virt_irq,
 	    chip == &sun4v_irq_ack ||
 	    chip == &sun4v_virq_ack
 #ifdef CONFIG_PCI_MSI
+	    || chip == &sun4u_msi
 	    || chip == &sun4v_msi
 #endif
 	    )
@@ -789,6 +813,53 @@ void sun4v_destroy_msi(unsigned int virt_irq)
 {
 	virt_irq_free(virt_irq);
 }
+
+unsigned int sun4u_build_msi(u32 portid, unsigned int *virt_irq_p,
+			     unsigned int msi_start, unsigned int msi_end,
+			     unsigned long imap_base, unsigned long iclr_base)
+{
+	struct ino_bucket *bucket;
+	struct irq_handler_data *data;
+	unsigned long sysino;
+	unsigned int devino;
+
+	/* Find a free devino in the given range.  */
+	for (devino = msi_start; devino < msi_end; devino++) {
+		sysino = (portid << 6) | devino;
+		bucket = &ivector_table[sysino];
+		if (!bucket->virt_irq)
+			break;
+	}
+	if (devino >= msi_end)
+		return -ENOSPC;
+
+	sysino = (portid << 6) | devino;
+	bucket = &ivector_table[sysino];
+	bucket->virt_irq = virt_irq_alloc(__irq(bucket));
+	*virt_irq_p = bucket->virt_irq;
+	set_irq_chip(bucket->virt_irq, &sun4u_msi);
+
+	data = get_irq_chip_data(bucket->virt_irq);
+	if (unlikely(data))
+		return devino;
+
+	data = kzalloc(sizeof(struct irq_handler_data), GFP_ATOMIC);
+	if (unlikely(!data)) {
+		virt_irq_free(*virt_irq_p);
+		return -ENOMEM;
+	}
+	set_irq_chip_data(bucket->virt_irq, data);
+
+	data->imap = (imap_base + (devino * 0x8UL));
+	data->iclr = (iclr_base + (devino * 0x8UL));
+
+	return devino;
+}
+
+void sun4u_destroy_msi(unsigned int virt_irq)
+{
+	virt_irq_free(virt_irq);
+}
 #endif
 
 void ack_bad_irq(unsigned int virt_irq)

arch/sparc64/kernel/pci_fire.c

@@ -6,9 +6,12 @@
 #include <linux/pci.h>
 #include <linux/slab.h>
 #include <linux/init.h>
+#include <linux/msi.h>
+#include <linux/irq.h>
 
 #include <asm/oplib.h>
 #include <asm/prom.h>
+#include <asm/irq.h>
 
 #include "pci_impl.h"
 
@@ -84,6 +87,440 @@ static int pci_fire_pbm_iommu_init(struct pci_pbm_info *pbm)
 	return 0;
 }
 
+#ifdef CONFIG_PCI_MSI
+struct pci_msiq_entry {
+	u64		word0;
+#define MSIQ_WORD0_RESV			0x8000000000000000UL
+#define MSIQ_WORD0_FMT_TYPE		0x7f00000000000000UL
+#define MSIQ_WORD0_FMT_TYPE_SHIFT	56
+#define MSIQ_WORD0_LEN			0x00ffc00000000000UL
+#define MSIQ_WORD0_LEN_SHIFT		46
+#define MSIQ_WORD0_ADDR0		0x00003fff00000000UL
+#define MSIQ_WORD0_ADDR0_SHIFT		32
+#define MSIQ_WORD0_RID			0x00000000ffff0000UL
+#define MSIQ_WORD0_RID_SHIFT		16
+#define MSIQ_WORD0_DATA0		0x000000000000ffffUL
+#define MSIQ_WORD0_DATA0_SHIFT		0
+
+#define MSIQ_TYPE_MSG			0x6
+#define MSIQ_TYPE_MSI32			0xb
+#define MSIQ_TYPE_MSI64			0xf
+
+	u64		word1;
+#define MSIQ_WORD1_ADDR1		0xffffffffffff0000UL
+#define MSIQ_WORD1_ADDR1_SHIFT		16
+#define MSIQ_WORD1_DATA1		0x000000000000ffffUL
+#define MSIQ_WORD1_DATA1_SHIFT		0
+
+	u64		resv[6];
+};
+
+/* All MSI registers are offset from pbm->pbm_regs */
+#define EVENT_QUEUE_BASE_ADDR_REG	0x010000UL
+#define  EVENT_QUEUE_BASE_ADDR_ALL_ONES	0xfffc000000000000UL
+
+#define EVENT_QUEUE_CONTROL_SET(EQ)	(0x011000UL + (EQ) * 0x8UL)
+#define  EVENT_QUEUE_CONTROL_SET_OFLOW	0x0200000000000000UL
+#define  EVENT_QUEUE_CONTROL_SET_EN	0x0000100000000000UL
+
+#define EVENT_QUEUE_CONTROL_CLEAR(EQ)	(0x011200UL + (EQ) * 0x8UL)
+#define  EVENT_QUEUE_CONTROL_CLEAR_OF	0x0200000000000000UL
+#define  EVENT_QUEUE_CONTROL_CLEAR_E2I	0x0000800000000000UL
+#define  EVENT_QUEUE_CONTROL_CLEAR_DIS	0x0000100000000000UL
+
+#define EVENT_QUEUE_STATE(EQ)		(0x011400UL + (EQ) * 0x8UL)
+#define  EVENT_QUEUE_STATE_MASK		0x0000000000000007UL
+#define  EVENT_QUEUE_STATE_IDLE		0x0000000000000001UL
+#define  EVENT_QUEUE_STATE_ACTIVE	0x0000000000000002UL
+#define  EVENT_QUEUE_STATE_ERROR	0x0000000000000004UL
+
+#define EVENT_QUEUE_TAIL(EQ)		(0x011600UL + (EQ) * 0x8UL)
+#define  EVENT_QUEUE_TAIL_OFLOW		0x0200000000000000UL
+#define  EVENT_QUEUE_TAIL_VAL		0x000000000000007fUL
+
+#define EVENT_QUEUE_HEAD(EQ)		(0x011800UL + (EQ) * 0x8UL)
+#define  EVENT_QUEUE_HEAD_VAL		0x000000000000007fUL
+
+#define MSI_MAP(MSI)			(0x020000UL + (MSI) * 0x8UL)
+#define  MSI_MAP_VALID			0x8000000000000000UL
+#define  MSI_MAP_EQWR_N			0x4000000000000000UL
+#define  MSI_MAP_EQNUM			0x000000000000003fUL
+
+#define MSI_CLEAR(MSI)			(0x028000UL + (MSI) * 0x8UL)
+#define  MSI_CLEAR_EQWR_N		0x4000000000000000UL
+
+#define IMONDO_DATA0			0x02C000UL
+#define  IMONDO_DATA0_DATA		0xffffffffffffffc0UL
+
+#define IMONDO_DATA1			0x02C008UL
+#define  IMONDO_DATA1_DATA		0xffffffffffffffffUL
+
+#define MSI_32BIT_ADDR			0x034000UL
+#define  MSI_32BIT_ADDR_VAL		0x00000000ffff0000UL
+
+#define MSI_64BIT_ADDR			0x034008UL
+#define  MSI_64BIT_ADDR_VAL		0xffffffffffff0000UL
+
+/* For now this just runs as a pre-handler for the real interrupt handler.
+ * So we just walk through the queue and ACK all the entries, update the
+ * head pointer, and return.
+ *
+ * In the longer term it would be nice to do something more integrated
+ * wherein we can pass in some of this MSI info to the drivers.  This
+ * would be most useful for PCIe fabric error messages, although we could
+ * invoke those directly from the loop here in order to pass the info around.
+ */
+static void pci_msi_prehandler(unsigned int ino, void *data1, void *data2)
+{
+	unsigned long msiqid, orig_head, head, type_fmt, type;
+	struct pci_pbm_info *pbm = data1;
+	struct pci_msiq_entry *base, *ep;
+
+	msiqid = (unsigned long) data2;
+
+	head = fire_read(pbm->pbm_regs + EVENT_QUEUE_HEAD(msiqid));
+
+	orig_head = head;
+	base = (pbm->msi_queues + ((msiqid - pbm->msiq_first) * 8192));
+	ep = &base[head];
+	while ((ep->word0 & MSIQ_WORD0_FMT_TYPE) != 0) {
+		unsigned long msi_num;
+
+		type_fmt = ((ep->word0 & MSIQ_WORD0_FMT_TYPE) >>
+			    MSIQ_WORD0_FMT_TYPE_SHIFT);
+		type = (type_fmt >>3);
+		if (unlikely(type != MSIQ_TYPE_MSI32 &&
+			     type != MSIQ_TYPE_MSI64))
+			goto bad_type;
+
+		msi_num = ((ep->word0 & MSIQ_WORD0_DATA0) >>
+			   MSIQ_WORD0_DATA0_SHIFT);
+
+		fire_write(pbm->pbm_regs + MSI_CLEAR(msi_num),
+			   MSI_CLEAR_EQWR_N);
+
+		/* Clear the entry.  */
+		ep->word0 &= ~MSIQ_WORD0_FMT_TYPE;
+
+		/* Go to next entry in ring.  */
+		head++;
+		if (head >= pbm->msiq_ent_count)
+			head = 0;
+		ep = &base[head];
+	}
+
+	if (likely(head != orig_head)) {
+		/* ACK entries by updating head pointer.  */
+		fire_write(pbm->pbm_regs +
+			   EVENT_QUEUE_HEAD(msiqid),
+			   head);
+	}
+	return;
+
+bad_type:
+	printk(KERN_EMERG "MSI: Entry has bad type %lx\n", type);
+	return;
+}
+
+static int msi_bitmap_alloc(struct pci_pbm_info *pbm)
+{
+	unsigned long size, bits_per_ulong;
+
+	bits_per_ulong = sizeof(unsigned long) * 8;
+	size = (pbm->msi_num + (bits_per_ulong - 1)) & ~(bits_per_ulong - 1);
+	size /= 8;
+	BUG_ON(size % sizeof(unsigned long));
+
+	pbm->msi_bitmap = kzalloc(size, GFP_KERNEL);
+	if (!pbm->msi_bitmap)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void msi_bitmap_free(struct pci_pbm_info *pbm)
+{
+	kfree(pbm->msi_bitmap);
+	pbm->msi_bitmap = NULL;
+}
+
+static int msi_queue_alloc(struct pci_pbm_info *pbm)
+{
+	unsigned long pages, order, i;
+
+	order = get_order(512 * 1024);
+	pages = __get_free_pages(GFP_KERNEL | __GFP_COMP, order);
+	if (pages == 0UL) {
+		printk(KERN_ERR "MSI: Cannot allocate MSI queues (o=%lu).\n",
+		       order);
+		return -ENOMEM;
+	}
+	memset((char *)pages, 0, PAGE_SIZE << order);
+	pbm->msi_queues = (void *) pages;
+
+	fire_write(pbm->pbm_regs + EVENT_QUEUE_BASE_ADDR_REG,
+		   (EVENT_QUEUE_BASE_ADDR_ALL_ONES |
+		    __pa(pbm->msi_queues)));
+
+	fire_write(pbm->pbm_regs + IMONDO_DATA0,
+		   pbm->portid << 6);
+	fire_write(pbm->pbm_regs + IMONDO_DATA1, 0);
+
+	fire_write(pbm->pbm_regs + MSI_32BIT_ADDR,
+		   pbm->msi32_start);
+	fire_write(pbm->pbm_regs + MSI_64BIT_ADDR,
+		   pbm->msi64_start);
+
+	for (i = 0; i < pbm->msiq_num; i++) {
+		fire_write(pbm->pbm_regs + EVENT_QUEUE_HEAD(i), 0);
+		fire_write(pbm->pbm_regs + EVENT_QUEUE_TAIL(i), 0);
+	}
+
+	return 0;
+}
+
+static int alloc_msi(struct pci_pbm_info *pbm)
+{
+	int i;
+
+	for (i = 0; i < pbm->msi_num; i++) {
+		if (!test_and_set_bit(i, pbm->msi_bitmap))
+			return i + pbm->msi_first;
+	}
+
+	return -ENOENT;
+}
+
+static void free_msi(struct pci_pbm_info *pbm, int msi_num)
+{
+	msi_num -= pbm->msi_first;
+	clear_bit(msi_num, pbm->msi_bitmap);
+}
+
+static int pci_setup_msi_irq(unsigned int *virt_irq_p,
+			     struct pci_dev *pdev,
+			     struct msi_desc *entry)
+{
+	struct pci_pbm_info *pbm = pdev->dev.archdata.host_controller;
+	unsigned long devino, msiqid, cregs, imap_off;
+	struct msi_msg msg;
+	int msi_num, err;
+	u64 val;
+
+	*virt_irq_p = 0;
+
+	msi_num = alloc_msi(pbm);
+	if (msi_num < 0)
+		return msi_num;
+
+	cregs = (unsigned long) pbm->pbm_regs;
+
+	err = sun4u_build_msi(pbm->portid, virt_irq_p,
+			      pbm->msiq_first_devino,
+			      (pbm->msiq_first_devino +
+			       pbm->msiq_num),
+			      cregs + 0x001000UL,
+			      cregs + 0x001400UL);
+	if (err < 0)
+		goto out_err;
+	devino = err;
+
+	imap_off = 0x001000UL + (devino * 0x8UL);
+
+	val = fire_read(pbm->pbm_regs + imap_off);
+	val |= (1UL << 63) | (1UL << 6);
+	fire_write(pbm->pbm_regs + imap_off, val);
+
+	msiqid = ((devino - pbm->msiq_first_devino) +
+		  pbm->msiq_first);
+
+	fire_write(pbm->pbm_regs +
+		   EVENT_QUEUE_CONTROL_SET(msiqid),
+		   EVENT_QUEUE_CONTROL_SET_EN);
+
+	val = fire_read(pbm->pbm_regs + MSI_MAP(msi_num));
+	val &= ~(MSI_MAP_EQNUM);
+	val |= msiqid;
+	fire_write(pbm->pbm_regs + MSI_MAP(msi_num), val);
+
+	fire_write(pbm->pbm_regs + MSI_CLEAR(msi_num),
+		   MSI_CLEAR_EQWR_N);
+
+	val = fire_read(pbm->pbm_regs + MSI_MAP(msi_num));
+	val |= MSI_MAP_VALID;
+	fire_write(pbm->pbm_regs + MSI_MAP(msi_num), val);
+
+	sparc64_set_msi(*virt_irq_p, msi_num);
+
+	if (entry->msi_attrib.is_64) {
+		msg.address_hi = pbm->msi64_start >> 32;
+		msg.address_lo = pbm->msi64_start & 0xffffffff;
+	} else {
+		msg.address_hi = 0;
+		msg.address_lo = pbm->msi32_start;
+	}
+	msg.data = msi_num;
+
+	set_irq_msi(*virt_irq_p, entry);
+	write_msi_msg(*virt_irq_p, &msg);
+
+	irq_install_pre_handler(*virt_irq_p,
+				pci_msi_prehandler,
+				pbm, (void *) msiqid);
+
+	return 0;
+
+out_err:
+	free_msi(pbm, msi_num);
+	return err;
+}
+
+static void pci_teardown_msi_irq(unsigned int virt_irq,
+				 struct pci_dev *pdev)
+{
+	struct pci_pbm_info *pbm = pdev->dev.archdata.host_controller;
+	unsigned long msiqid, msi_num;
+	u64 val;
+
+	msi_num = sparc64_get_msi(virt_irq);
+
+	val = fire_read(pbm->pbm_regs + MSI_MAP(msi_num));
+
+	msiqid = (val & MSI_MAP_EQNUM);
+
+	val &= ~MSI_MAP_VALID;
+	fire_write(pbm->pbm_regs + MSI_MAP(msi_num), val);
+
+	fire_write(pbm->pbm_regs + EVENT_QUEUE_CONTROL_CLEAR(msiqid),
+		   EVENT_QUEUE_CONTROL_CLEAR_DIS);
+
+	free_msi(pbm, msi_num);
+
+	/* The sun4u_destroy_msi() will liberate the devino and thus the MSIQ
+	 * allocation.
+	 */
+	sun4u_destroy_msi(virt_irq);
+}
+
+static void pci_fire_msi_init(struct pci_pbm_info *pbm)
+{
+	const u32 *val;
+	int len;
+
+	val = of_get_property(pbm->prom_node, "#msi-eqs", &len);
+	if (!val || len != 4)
+		goto no_msi;
+	pbm->msiq_num = *val;
+	if (pbm->msiq_num) {
+		const struct msiq_prop {
+			u32 first_msiq;
+			u32 num_msiq;
+			u32 first_devino;
+		} *mqp;
+		const struct msi_range_prop {
+			u32 first_msi;
+			u32 num_msi;
+		} *mrng;
+		const struct addr_range_prop {
+			u32 msi32_high;
+			u32 msi32_low;
+			u32 msi32_len;
+			u32 msi64_high;
+			u32 msi64_low;
+			u32 msi64_len;
+		} *arng;
+
+		val = of_get_property(pbm->prom_node, "msi-eq-size", &len);
+		if (!val || len != 4)
+			goto no_msi;
+
+		pbm->msiq_ent_count = *val;
+
+		mqp = of_get_property(pbm->prom_node,
+				      "msi-eq-to-devino", &len);
+		if (!mqp)
+			mqp = of_get_property(pbm->prom_node,
+					      "msi-eq-devino", &len);
+		if (!mqp || len != sizeof(struct msiq_prop))
+			goto no_msi;
+
+		pbm->msiq_first = mqp->first_msiq;
+		pbm->msiq_first_devino = mqp->first_devino;
+
+		val = of_get_property(pbm->prom_node, "#msi", &len);
+		if (!val || len != 4)
+			goto no_msi;
+		pbm->msi_num = *val;
+
+		mrng = of_get_property(pbm->prom_node, "msi-ranges", &len);
+		if (!mrng || len != sizeof(struct msi_range_prop))
+			goto no_msi;
+		pbm->msi_first = mrng->first_msi;
+
+		val = of_get_property(pbm->prom_node, "msi-data-mask", &len);
+		if (!val || len != 4)
+			goto no_msi;
+		pbm->msi_data_mask = *val;
+
+		val = of_get_property(pbm->prom_node, "msix-data-width", &len);
+		if (!val || len != 4)
+			goto no_msi;
+		pbm->msix_data_width = *val;
+
+		arng = of_get_property(pbm->prom_node, "msi-address-ranges",
+				       &len);
+		if (!arng || len != sizeof(struct addr_range_prop))
+			goto no_msi;
+		pbm->msi32_start = ((u64)arng->msi32_high << 32) |
+			(u64) arng->msi32_low;
+		pbm->msi64_start = ((u64)arng->msi64_high << 32) |
+			(u64) arng->msi64_low;
+		pbm->msi32_len = arng->msi32_len;
+		pbm->msi64_len = arng->msi64_len;
+
+		if (msi_bitmap_alloc(pbm))
+			goto no_msi;
+
+		if (msi_queue_alloc(pbm)) {
+			msi_bitmap_free(pbm);
+			goto no_msi;
+		}
+
+		printk(KERN_INFO "%s: MSI Queue first[%u] num[%u] count[%u] "
+		       "devino[0x%x]\n",
+		       pbm->name,
+		       pbm->msiq_first, pbm->msiq_num,
+		       pbm->msiq_ent_count,
+		       pbm->msiq_first_devino);
+		printk(KERN_INFO "%s: MSI first[%u] num[%u] mask[0x%x] "
+		       "width[%u]\n",
+		       pbm->name,
+		       pbm->msi_first, pbm->msi_num, pbm->msi_data_mask,
+		       pbm->msix_data_width);
+		printk(KERN_INFO "%s: MSI addr32[0x%lx:0x%x] "
+		       "addr64[0x%lx:0x%x]\n",
+		       pbm->name,
+		       pbm->msi32_start, pbm->msi32_len,
+		       pbm->msi64_start, pbm->msi64_len);
+		printk(KERN_INFO "%s: MSI queues at RA [%016lx]\n",
+		       pbm->name,
+		       __pa(pbm->msi_queues));
+	}
+	pbm->setup_msi_irq = pci_setup_msi_irq;
+	pbm->teardown_msi_irq = pci_teardown_msi_irq;
+
+	return;
+
+no_msi:
+	pbm->msiq_num = 0;
+	printk(KERN_INFO "%s: No MSI support.\n", pbm->name);
+}
+#else /* CONFIG_PCI_MSI */
+static void pci_fire_msi_init(struct pci_pbm_info *pbm)
+{
+}
+#endif /* !(CONFIG_PCI_MSI) */
+
 /* Based at pbm->controller_regs */
 #define FIRE_PARITY_CONTROL	0x470010UL
 #define  FIRE_PARITY_ENAB	0x8000000000000000UL
@@ -176,6 +613,7 @@ static int pci_fire_pbm_init(struct pci_controller_info *p,
 {
 	const struct linux_prom64_registers *regs;
 	struct pci_pbm_info *pbm;
+	int err;
 
 	if ((portid & 1) == 0)
 		pbm = &p->pbm_A;
@@ -208,7 +646,13 @@ static int pci_fire_pbm_init(struct pci_controller_info *p,
 
 	pci_fire_hw_init(pbm);
 
-	return pci_fire_pbm_iommu_init(pbm);
+	err = pci_fire_pbm_iommu_init(pbm);
+	if (err)
+		return err;
+
+	pci_fire_msi_init(pbm);
+
+	return 0;
 }
 
 static inline int portid_compare(u32 x, u32 y)

include/asm-sparc64/irq.h

@@ -51,6 +51,12 @@ extern unsigned int sun4v_build_msi(u32 devhandle, unsigned int *virt_irq_p,
 				    unsigned int msi_devino_start,
 				    unsigned int msi_devino_end);
 extern void sun4v_destroy_msi(unsigned int virt_irq);
+extern unsigned int sun4u_build_msi(u32 portid, unsigned int *virt_irq_p,
+				    unsigned int msi_devino_start,
+				    unsigned int msi_devino_end,
+				    unsigned long imap_base,
+				    unsigned long iclr_base);
+extern void sun4u_destroy_msi(unsigned int virt_irq);
 extern unsigned int sbus_build_irq(void *sbus, unsigned int ino);
 
 extern void sparc64_set_msi(unsigned int virt_irq, u32 msi);