Merge branch 'x86-platform-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 platform updates from Ingo Molnar:
 "The main change is the addition of SGI/UV4 support"

* 'x86-platform-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (22 commits)
  x86/platform/UV: Fix incorrect nodes and pnodes for cpuless and memoryless nodes
  x86/platform/UV: Remove Obsolete GRU MMR address translation
  x86/platform/UV: Update physical address conversions for UV4
  x86/platform/UV: Build GAM reference tables
  x86/platform/UV: Support UV4 socket address changes
  x86/platform/UV: Add obtaining GAM Range Table from UV BIOS
  x86/platform/UV: Add UV4 addressing discovery function
  x86/platform/UV: Fold blade info into per node hub info structs
  x86/platform/UV: Allocate common per node hub info structs on local node
  x86/platform/UV: Move blade local processor ID to the per cpu info struct
  x86/platform/UV: Move scir info to the per cpu info struct
  x86/platform/UV: Create per cpu info structs to replace per hub info structs
  x86/platform/UV: Update MMIOH setup function to work for both UV3 and UV4
  x86/platform/UV: Clean up redunduncies after merge of UV4 MMR definitions
  x86/platform/UV: Add UV4 Specific MMR definitions
  x86/platform/UV: Prep for UV4 MMR updates
  x86/platform/UV: Add UV MMR Illegal Access Function
  x86/platform/UV: Add UV4 Specific Defines
  x86/platform/UV: Add UV Architecture Defines
  x86/platform/UV: Add Initial UV4 definitions
  ...

Documentation/kernel-parameters.txt

@@ -131,6 +131,7 @@ parameter is applicable:
 			More X86-64 boot options can be found in
 			Documentation/x86/x86_64/boot-options.txt .
 	X86	Either 32-bit or 64-bit x86 (same as X86-32+X86-64)
+	X86_UV	SGI UV support is enabled.
 	XEN	Xen support is enabled
 
 In addition, the following text indicates that the option:
@@ -542,6 +543,13 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 			Format: <int> (must be >=0)
 			Default: 64
 
+	bau=		[X86_UV] Enable the BAU on SGI UV.  The default
+			behavior is to disable the BAU (i.e. bau=0).
+			Format: { "0" | "1" }
+			0 - Disable the BAU.
+			1 - Enable the BAU.
+			unset - Disable the BAU.
+
 	baycom_epp=	[HW,AX25]
 			Format: <io>,<mode>
 

arch/x86/include/asm/bios_ebda.h

@@ -17,27 +17,6 @@ static inline unsigned int get_bios_ebda(void)
 	return address;	/* 0 means none */
 }
 
-/*
- * Return the sanitized length of the EBDA in bytes, if it exists.
- */
-static inline unsigned int get_bios_ebda_length(void)
-{
-	unsigned int address;
-	unsigned int length;
-
-	address = get_bios_ebda();
-	if (!address)
-		return 0;
-
-	/* EBDA length is byte 0 of the EBDA (stored in KiB) */
-	length = *(unsigned char *)phys_to_virt(address);
-	length <<= 10;
-
-	/* Trim the length if it extends beyond 640KiB */
-	length = min_t(unsigned int, (640 * 1024) - address, length);
-	return length;
-}
-
 void reserve_ebda_region(void);
 
 #ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION

arch/x86/include/asm/uv/bios.h

@@ -51,15 +51,66 @@ enum {
 	BIOS_STATUS_UNAVAIL		= -EBUSY
 };
 
+/* Address map parameters */
+struct uv_gam_parameters {
+	u64	mmr_base;
+	u64	gru_base;
+	u8	mmr_shift;	/* Convert PNode to MMR space offset */
+	u8	gru_shift;	/* Convert PNode to GRU space offset */
+	u8	gpa_shift;	/* Size of offset field in GRU phys addr */
+	u8	unused1;
+};
+
+/* UV_TABLE_GAM_RANGE_ENTRY values */
+#define UV_GAM_RANGE_TYPE_UNUSED	0 /* End of table */
+#define UV_GAM_RANGE_TYPE_RAM		1 /* Normal RAM */
+#define UV_GAM_RANGE_TYPE_NVRAM		2 /* Non-volatile memory */
+#define UV_GAM_RANGE_TYPE_NV_WINDOW	3 /* NVMDIMM block window */
+#define UV_GAM_RANGE_TYPE_NV_MAILBOX	4 /* NVMDIMM mailbox */
+#define UV_GAM_RANGE_TYPE_HOLE		5 /* Unused address range */
+#define UV_GAM_RANGE_TYPE_MAX		6
+
+/* The structure stores PA bits 56:26, for 64MB granularity */
+#define UV_GAM_RANGE_SHFT		26		/* 64MB */
+
+struct uv_gam_range_entry {
+	char	type;		/* Entry type: GAM_RANGE_TYPE_UNUSED, etc. */
+	char	unused1;
+	u16	nasid;		/* HNasid */
+	u16	sockid;		/* Socket ID, high bits of APIC ID */
+	u16	pnode;		/* Index to MMR and GRU spaces */
+	u32	pxm;		/* ACPI proximity domain number */
+	u32	limit;		/* PA bits 56:26 (UV_GAM_RANGE_SHFT) */
+};
+
+#define	UV_SYSTAB_SIG			"UVST"
+#define	UV_SYSTAB_VERSION_1		1	/* UV1/2/3 BIOS version */
+#define	UV_SYSTAB_VERSION_UV4		0x400	/* UV4 BIOS base version */
+#define	UV_SYSTAB_VERSION_UV4_1		0x401	/* + gpa_shift */
+#define	UV_SYSTAB_VERSION_UV4_2		0x402	/* + TYPE_NVRAM/WINDOW/MBOX */
+#define	UV_SYSTAB_VERSION_UV4_LATEST	UV_SYSTAB_VERSION_UV4_2
+
+#define	UV_SYSTAB_TYPE_UNUSED		0	/* End of table (offset == 0) */
+#define	UV_SYSTAB_TYPE_GAM_PARAMS	1	/* GAM PARAM conversions */
+#define	UV_SYSTAB_TYPE_GAM_RNG_TBL	2	/* GAM entry table */
+#define	UV_SYSTAB_TYPE_MAX		3
+
 /*
  * The UV system table describes specific firmware
  * capabilities available to the Linux kernel at runtime.
  */
 struct uv_systab {
-	char signature[4];	/* must be "UVST" */
+	char signature[4];	/* must be UV_SYSTAB_SIG */
 	u32 revision;		/* distinguish different firmware revs */
 	u64 function;		/* BIOS runtime callback function ptr */
+	u32 size;		/* systab size (starting with _VERSION_UV4) */
+	struct {
+		u32 type:8;	/* type of entry */
+		u32 offset:24;	/* byte offset from struct start to entry */
+	} entry[1];		/* additional entries follow */
 };
+extern struct uv_systab *uv_systab;
+/* (... end of definitions from UV BIOS ...) */
 
 enum {
 	BIOS_FREQ_BASE_PLATFORM = 0,
@@ -99,7 +150,11 @@ extern s64 uv_bios_change_memprotect(u64, u64, enum uv_memprotect);
 extern s64 uv_bios_reserved_page_pa(u64, u64 *, u64 *, u64 *);
 extern int uv_bios_set_legacy_vga_target(bool decode, int domain, int bus);
 
+#ifdef CONFIG_EFI
 extern void uv_bios_init(void);
+#else
+void uv_bios_init(void) { }
+#endif
 
 extern unsigned long sn_rtc_cycles_per_second;
 extern int uv_type;
@@ -107,7 +162,7 @@ extern long sn_partition_id;
 extern long sn_coherency_id;
 extern long sn_region_size;
 extern long system_serial_number;
-#define partition_coherence_id()	(sn_coherency_id)
+#define uv_partition_coherence_id()	(sn_coherency_id)
 
 extern struct kobject *sgi_uv_kobj;	/* /sys/firmware/sgi_uv */
 

arch/x86/include/asm/uv/uv_bau.h

@@ -598,7 +598,7 @@ struct bau_control {
 	int			timeout_tries;
 	int			ipi_attempts;
 	int			conseccompletes;
-	short			nobau;
+	bool			nobau;
 	short			baudisabled;
 	short			cpu;
 	short			osnode;

arch/x86/platform/uv/bios_uv.c

@@ -21,19 +21,20 @@
 
 #include <linux/efi.h>
 #include <linux/export.h>
+#include <linux/slab.h>
 #include <asm/efi.h>
 #include <linux/io.h>
 #include <asm/uv/bios.h>
 #include <asm/uv/uv_hub.h>
 
-static struct uv_systab uv_systab;
+struct uv_systab *uv_systab;
 
 s64 uv_bios_call(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, u64 a4, u64 a5)
 {
-	struct uv_systab *tab = &uv_systab;
+	struct uv_systab *tab = uv_systab;
 	s64 ret;
 
-	if (!tab->function)
+	if (!tab || !tab->function)
 		/*
 		 * BIOS does not support UV systab
 		 */
@@ -183,34 +184,31 @@ int uv_bios_set_legacy_vga_target(bool decode, int domain, int bus)
 }
 EXPORT_SYMBOL_GPL(uv_bios_set_legacy_vga_target);
 
-
 #ifdef CONFIG_EFI
 void uv_bios_init(void)
 {
-	struct uv_systab *tab;
-
-	if ((efi.uv_systab == EFI_INVALID_TABLE_ADDR) ||
-	    (efi.uv_systab == (unsigned long)NULL)) {
-		printk(KERN_CRIT "No EFI UV System Table.\n");
-		uv_systab.function = (unsigned long)NULL;
+	uv_systab = NULL;
+	if ((efi.uv_systab == EFI_INVALID_TABLE_ADDR) || !efi.uv_systab) {
+		pr_crit("UV: UVsystab: missing\n");
 		return;
 	}
 
-	tab = (struct uv_systab *)ioremap(efi.uv_systab,
-					sizeof(struct uv_systab));
-	if (strncmp(tab->signature, "UVST", 4) != 0)
-		printk(KERN_ERR "bad signature in UV system table!");
-
-	/*
-	 * Copy table to permanent spot for later use.
-	 */
-	memcpy(&uv_systab, tab, sizeof(struct uv_systab));
-	iounmap(tab);
+	uv_systab = ioremap(efi.uv_systab, sizeof(struct uv_systab));
+	if (!uv_systab || strncmp(uv_systab->signature, UV_SYSTAB_SIG, 4)) {
+		pr_err("UV: UVsystab: bad signature!\n");
+		iounmap(uv_systab);
+		return;
+	}
 
-	printk(KERN_INFO "EFI UV System Table Revision %d\n",
-					uv_systab.revision);
+	if (uv_systab->revision >= UV_SYSTAB_VERSION_UV4) {
+		iounmap(uv_systab);
+		uv_systab = ioremap(efi.uv_systab, uv_systab->size);
+		if (!uv_systab) {
+			pr_err("UV: UVsystab: ioremap(%d) failed!\n",
+				uv_systab->size);
+			return;
+		}
+	}
+	pr_info("UV: UVsystab: Revision:%x\n", uv_systab->revision);
 }
-#else	/* !CONFIG_EFI */
-
-void uv_bios_init(void) { }
 #endif

arch/x86/platform/uv/tlb_uv.c

@@ -37,7 +37,7 @@ static int timeout_base_ns[] = {
 };
 
 static int timeout_us;
-static int nobau;
+static bool nobau = true;
 static int nobau_perm;
 static cycles_t congested_cycles;
 
@@ -106,13 +106,28 @@ static char *stat_description[] = {
 	"enable:   number times use of the BAU was re-enabled"
 };
 
-static int __init
-setup_nobau(char *arg)
+static int __init setup_bau(char *arg)
 {
-	nobau = 1;
+	int result;
+
+	if (!arg)
+		return -EINVAL;
+
+	result = strtobool(arg, &nobau);
+	if (result)
+		return result;
+
+	/* we need to flip the logic here, so that bau=y sets nobau to false */
+	nobau = !nobau;
+
+	if (!nobau)
+		pr_info("UV BAU Enabled\n");
+	else
+		pr_info("UV BAU Disabled\n");
+
 	return 0;
 }
-early_param("nobau", setup_nobau);
+early_param("bau", setup_bau);
 
 /* base pnode in this partition */
 static int uv_base_pnode __read_mostly;
@@ -131,10 +146,10 @@ set_bau_on(void)
 		pr_info("BAU not initialized; cannot be turned on\n");
 		return;
 	}
-	nobau = 0;
+	nobau = false;
 	for_each_present_cpu(cpu) {
 		bcp = &per_cpu(bau_control, cpu);
-		bcp->nobau = 0;
+		bcp->nobau = false;
 	}
 	pr_info("BAU turned on\n");
 	return;
@@ -146,10 +161,10 @@ set_bau_off(void)
 	int cpu;
 	struct bau_control *bcp;
 
-	nobau = 1;
+	nobau = true;
 	for_each_present_cpu(cpu) {
 		bcp = &per_cpu(bau_control, cpu);
-		bcp->nobau = 1;
+		bcp->nobau = true;
 	}
 	pr_info("BAU turned off\n");
 	return;
@@ -1886,7 +1901,7 @@ static void __init init_per_cpu_tunables(void)
 		bcp = &per_cpu(bau_control, cpu);
 		bcp->baudisabled		= 0;
 		if (nobau)
-			bcp->nobau		= 1;
+			bcp->nobau		= true;
 		bcp->statp			= &per_cpu(ptcstats, cpu);
 		/* time interval to catch a hardware stay-busy bug */
 		bcp->timeout_interval		= usec_2_cycles(2*timeout_us);
@@ -2025,7 +2040,8 @@ static int scan_sock(struct socket_desc *sdp, struct uvhub_desc *bdp,
 			return 1;
 		}
 		bcp->uvhub_master = *hmasterp;
-		bcp->uvhub_cpu = uv_cpu_hub_info(cpu)->blade_processor_id;
+		bcp->uvhub_cpu = uv_cpu_blade_processor_id(cpu);
+
 		if (bcp->uvhub_cpu >= MAX_CPUS_PER_UVHUB) {
 			printk(KERN_EMERG "%d cpus per uvhub invalid\n",
 				bcp->uvhub_cpu);

arch/x86/platform/uv/uv_sysfs.c

@@ -34,7 +34,7 @@ static ssize_t partition_id_show(struct kobject *kobj,
 static ssize_t coherence_id_show(struct kobject *kobj,
 			struct kobj_attribute *attr, char *buf)
 {
-	return snprintf(buf, PAGE_SIZE, "%ld\n", partition_coherence_id());
+	return snprintf(buf, PAGE_SIZE, "%ld\n", uv_partition_coherence_id());
 }
 
 static struct kobj_attribute partition_id_attr =

arch/x86/platform/uv/uv_time.c

@@ -165,7 +165,7 @@ static __init int uv_rtc_allocate_timers(void)
 	for_each_present_cpu(cpu) {
 		int nid = cpu_to_node(cpu);
 		int bid = uv_cpu_to_blade_id(cpu);
-		int bcpu = uv_cpu_hub_info(cpu)->blade_processor_id;
+		int bcpu = uv_cpu_blade_processor_id(cpu);
 		struct uv_rtc_timer_head *head = blade_info[bid];
 
 		if (!head) {
@@ -226,7 +226,7 @@ static int uv_rtc_set_timer(int cpu, u64 expires)
 	int pnode = uv_cpu_to_pnode(cpu);
 	int bid = uv_cpu_to_blade_id(cpu);
 	struct uv_rtc_timer_head *head = blade_info[bid];
-	int bcpu = uv_cpu_hub_info(cpu)->blade_processor_id;
+	int bcpu = uv_cpu_blade_processor_id(cpu);
 	u64 *t = &head->cpu[bcpu].expires;
 	unsigned long flags;
 	int next_cpu;
@@ -262,7 +262,7 @@ static int uv_rtc_unset_timer(int cpu, int force)
 	int pnode = uv_cpu_to_pnode(cpu);
 	int bid = uv_cpu_to_blade_id(cpu);
 	struct uv_rtc_timer_head *head = blade_info[bid];
-	int bcpu = uv_cpu_hub_info(cpu)->blade_processor_id;
+	int bcpu = uv_cpu_blade_processor_id(cpu);
 	u64 *t = &head->cpu[bcpu].expires;
 	unsigned long flags;
 	int rc = 0;

drivers/misc/sgi-gru/grukservices.c

@@ -718,8 +718,8 @@ static int send_message_queue_full(void *cb, struct gru_message_queue_desc *mqd,
 static int send_message_put_nacked(void *cb, struct gru_message_queue_desc *mqd,
 			void *mesg, int lines)
 {
-	unsigned long m, *val = mesg, gpa, save;
-	int ret;
+	unsigned long m;
+	int ret, loops = 200;	/* experimentally determined */
 
 	m = mqd->mq_gpa + (gru_get_amo_value_head(cb) << 6);
 	if (lines == 2) {
@@ -735,22 +735,28 @@ static int send_message_put_nacked(void *cb, struct gru_message_queue_desc *mqd,
 		return MQE_OK;
 
 	/*
-	 * Send a cross-partition interrupt to the SSI that contains the target
-	 * message queue. Normally, the interrupt is automatically delivered by
-	 * hardware but some error conditions require explicit delivery.
-	 * Use the GRU to deliver the interrupt. Otherwise partition failures
+	 * Send a noop message in order to deliver a cross-partition interrupt
+	 * to the SSI that contains the target message queue. Normally, the
+	 * interrupt is automatically delivered by hardware following mesq
+	 * operations, but some error conditions require explicit delivery.
+	 * The noop message will trigger delivery. Otherwise partition failures
 	 * could cause unrecovered errors.
 	 */
-	gpa = uv_global_gru_mmr_address(mqd->interrupt_pnode, UVH_IPI_INT);
-	save = *val;
-	*val = uv_hub_ipi_value(mqd->interrupt_apicid, mqd->interrupt_vector,
-				dest_Fixed);
-	gru_vstore_phys(cb, gpa, gru_get_tri(mesg), IAA_REGISTER, IMA);
-	ret = gru_wait(cb);
-	*val = save;
-	if (ret != CBS_IDLE)
-		return MQE_UNEXPECTED_CB_ERR;
-	return MQE_OK;
+	do {
+		ret = send_noop_message(cb, mqd, mesg);
+	} while ((ret == MQIE_AGAIN || ret == MQE_CONGESTION) && (loops-- > 0));
+
+	if (ret == MQIE_AGAIN || ret == MQE_CONGESTION) {
+		/*
+		 * Don't indicate to the app to resend the message, as it's
+		 * already been successfully sent.  We simply send an OK
+		 * (rather than fail the send with MQE_UNEXPECTED_CB_ERR),
+		 * assuming that the other side is receiving enough
+		 * interrupts to get this message processed anyway.
+		 */
+		ret = MQE_OK;
+	}
+	return ret;
 }
 
 /*