Pull percpu-dtc into release branch

arch/ia64/kernel/entry.S

@@ -767,7 +767,7 @@ ENTRY(ia64_leave_syscall)
 	ld8.fill r15=[r3]			// M0|1 restore r15
 	mov b6=r18				// I0   restore b6
 
-	addl r17=THIS_CPU(ia64_phys_stacked_size_p8),r0 // A
+	LOAD_PHYS_STACK_REG_SIZE(r17)
 	mov f9=f0					// F    clear f9
 (pKStk) br.cond.dpnt.many skip_rbs_switch		// B
 
@@ -775,7 +775,6 @@ ENTRY(ia64_leave_syscall)
 	shr.u r18=r19,16		// I0|1 get byte size of existing "dirty" partition
 	cover				// B    add current frame into dirty partition & set cr.ifs
 	;;
-(pUStk) ld4 r17=[r17]			// M0|1 r17 = cpu_data->phys_stacked_size_p8
 	mov r19=ar.bsp			// M2   get new backing store pointer
 	mov f10=f0			// F    clear f10
 
@@ -953,9 +952,7 @@ GLOBAL_ENTRY(ia64_leave_kernel)
 	shr.u r18=r19,16	// get byte size of existing "dirty" partition
 	;;
 	mov r16=ar.bsp		// get existing backing store pointer
-	addl r17=THIS_CPU(ia64_phys_stacked_size_p8),r0
-	;;
-	ld4 r17=[r17]		// r17 = cpu_data->phys_stacked_size_p8
+	LOAD_PHYS_STACK_REG_SIZE(r17)
 (pKStk)	br.cond.dpnt skip_rbs_switch
 
 	/*

arch/ia64/kernel/ivt.S

@@ -374,6 +374,7 @@ ENTRY(alt_dtlb_miss)
 	movl r19=(((1 << IA64_MAX_PHYS_BITS) - 1) & ~0xfff)
 	mov r21=cr.ipsr
 	mov r31=pr
+	mov r24=PERCPU_ADDR
 	;;
 #ifdef CONFIG_DISABLE_VHPT
 	shr.u r22=r16,61			// get the region number into r21
@@ -386,22 +387,30 @@ ENTRY(alt_dtlb_miss)
 (p8)	mov r29=b0				// save b0
 (p8)	br.cond.dptk dtlb_fault
 #endif
+	cmp.ge p10,p11=r16,r24			// access to per_cpu_data?
+	tbit.z p12,p0=r16,61			// access to region 6?
+	mov r25=PERCPU_PAGE_SHIFT << 2
+	mov r26=PERCPU_PAGE_SIZE
+	nop.m 0
+	nop.b 0
+	;;
+(p10)	mov r19=IA64_KR(PER_CPU_DATA)
+(p11)	and r19=r19,r16				// clear non-ppn fields
 	extr.u r23=r21,IA64_PSR_CPL0_BIT,2	// extract psr.cpl
 	and r22=IA64_ISR_CODE_MASK,r20		// get the isr.code field
 	tbit.nz p6,p7=r20,IA64_ISR_SP_BIT	// is speculation bit on?
-	shr.u r18=r16,57			// move address bit 61 to bit 4
-	and r19=r19,r16				// clear ed, reserved bits, and PTE control bits
 	tbit.nz p9,p0=r20,IA64_ISR_NA_BIT	// is non-access bit on?
 	;;
-	andcm r18=0x10,r18	// bit 4=~address-bit(61)
+(p10)	sub r19=r19,r26
+(p10)	mov cr.itir=r25
 	cmp.ne p8,p0=r0,r23
 (p9)	cmp.eq.or.andcm p6,p7=IA64_ISR_CODE_LFETCH,r22	// check isr.code field
+(p12)	dep r17=-1,r17,4,1			// set ma=UC for region 6 addr
 (p8)	br.cond.spnt page_fault
 
 	dep r21=-1,r21,IA64_PSR_ED_BIT,1
-	or r19=r19,r17		// insert PTE control bits into r19
 	;;
-	or r19=r19,r18		// set bit 4 (uncached) if the access was to region 6
+	or r19=r19,r17		// insert PTE control bits into r19
 (p6)	mov cr.ipsr=r21
 	;;
 (p7)	itc.d r19		// insert the TLB entry

arch/ia64/kernel/mca_asm.S

@@ -101,14 +101,6 @@ ia64_do_tlb_purge:
 	;;
 	srlz.d
 	;;
-	// 2. Purge DTR for PERCPU data.
-	movl r16=PERCPU_ADDR
-	mov r18=PERCPU_PAGE_SHIFT<<2
-	;;
-	ptr.d r16,r18
-	;;
-	srlz.d
-	;;
 	// 3. Purge ITR for PAL code.
 	GET_THIS_PADDR(r2, ia64_mca_pal_base)
 	;;
@@ -196,22 +188,6 @@ ia64_reload_tr:
 	srlz.i
 	srlz.d
 	;;
-	// 2. Reload DTR register for PERCPU data.
-	GET_THIS_PADDR(r2, ia64_mca_per_cpu_pte)
-	;;
-	movl r16=PERCPU_ADDR		// vaddr
-	movl r18=PERCPU_PAGE_SHIFT<<2
-	;;
-	mov cr.itir=r18
-	mov cr.ifa=r16
-	;;
-	ld8 r18=[r2]			// load per-CPU PTE
-	mov r16=IA64_TR_PERCPU_DATA;
-	;;
-	itr.d dtr[r16]=r18
-	;;
-	srlz.d
-	;;
 	// 3. Reload ITR for PAL code.
 	GET_THIS_PADDR(r2, ia64_mca_pal_pte)
 	;;

arch/ia64/kernel/patch.c

@@ -195,3 +195,23 @@ ia64_patch_gate (void)
 	ia64_patch_vtop(START(vtop), END(vtop));
 	ia64_patch_mckinley_e9(START(mckinley_e9), END(mckinley_e9));
 }
+
+void ia64_patch_phys_stack_reg(unsigned long val)
+{
+	s32 * offp = (s32 *) __start___phys_stack_reg_patchlist;
+	s32 * end = (s32 *) __end___phys_stack_reg_patchlist;
+	u64 ip, mask, imm;
+
+	/* see instruction format A4: adds r1 = imm13, r3 */
+	mask = (0x3fUL << 27) | (0x7f << 13);
+	imm = (((val >> 7) & 0x3f) << 27) | (val & 0x7f) << 13;
+
+	while (offp < end) {
+		ip = (u64) offp + *offp;
+		ia64_patch(ip, mask, imm);
+		ia64_fc(ip);
+		++offp;
+	}
+	ia64_sync_i();
+	ia64_srlz_i();
+}

arch/ia64/kernel/setup.c

@@ -75,7 +75,6 @@ extern void ia64_setup_printk_clock(void);
 
 DEFINE_PER_CPU(struct cpuinfo_ia64, cpu_info);
 DEFINE_PER_CPU(unsigned long, local_per_cpu_offset);
-DEFINE_PER_CPU(unsigned long, ia64_phys_stacked_size_p8);
 unsigned long ia64_cycles_per_usec;
 struct ia64_boot_param *ia64_boot_param;
 struct screen_info screen_info;
@@ -869,6 +868,7 @@ void __cpuinit
 cpu_init (void)
 {
 	extern void __cpuinit ia64_mmu_init (void *);
+	static unsigned long max_num_phys_stacked = IA64_NUM_PHYS_STACK_REG;
 	unsigned long num_phys_stacked;
 	pal_vm_info_2_u_t vmi;
 	unsigned int max_ctx;
@@ -982,7 +982,10 @@ cpu_init (void)
 		num_phys_stacked = 96;
 	}
 	/* size of physical stacked register partition plus 8 bytes: */
-	__get_cpu_var(ia64_phys_stacked_size_p8) = num_phys_stacked*8 + 8;
+	if (num_phys_stacked > max_num_phys_stacked) {
+		ia64_patch_phys_stack_reg(num_phys_stacked*8 + 8);
+		max_num_phys_stacked = num_phys_stacked;
+	}
 	platform_cpu_init();
 	pm_idle = default_idle;
 }

arch/ia64/kernel/vmlinux.lds.S

@@ -78,6 +78,13 @@ SECTIONS
 	  __stop___mca_table = .;
 	}
 
+  .data.patch.phys_stack_reg : AT(ADDR(.data.patch.phys_stack_reg) - LOAD_OFFSET)
+	{
+	  __start___phys_stack_reg_patchlist = .;
+	  *(.data.patch.phys_stack_reg)
+	  __end___phys_stack_reg_patchlist = .;
+	}
+
   /* Global data */
   _data = .;
 

arch/ia64/mm/init.c

@@ -355,7 +355,7 @@ setup_gate (void)
 void __devinit
 ia64_mmu_init (void *my_cpu_data)
 {
-	unsigned long psr, pta, impl_va_bits;
+	unsigned long pta, impl_va_bits;
 	extern void __devinit tlb_init (void);
 
 #ifdef CONFIG_DISABLE_VHPT
@@ -364,15 +364,6 @@ ia64_mmu_init (void *my_cpu_data)
 #	define VHPT_ENABLE_BIT	1
 #endif
 
-	/* Pin mapping for percpu area into TLB */
-	psr = ia64_clear_ic();
-	ia64_itr(0x2, IA64_TR_PERCPU_DATA, PERCPU_ADDR,
-		 pte_val(pfn_pte(__pa(my_cpu_data) >> PAGE_SHIFT, PAGE_KERNEL)),
-		 PERCPU_PAGE_SHIFT);
-
-	ia64_set_psr(psr);
-	ia64_srlz_i();
-
 	/*
 	 * Check if the virtually mapped linear page table (VMLPT) overlaps with a mapped
 	 * address space.  The IA-64 architecture guarantees that at least 50 bits of

include/asm-ia64/asmmacro.h

@@ -103,6 +103,16 @@
 # define FSYS_RETURN	br.ret.sptk.many b6
 #endif
 
+/*
+ * If physical stack register size is different from DEF_NUM_STACK_REG,
+ * dynamically patch the kernel for correct size.
+ */
+	.section ".data.patch.phys_stack_reg", "a"
+	.previous
+#define LOAD_PHYS_STACK_REG_SIZE(reg)			\
+[1:]	adds reg=IA64_NUM_PHYS_STACK_REG*8+8,r0;	\
+	.xdata4 ".data.patch.phys_stack_reg", 1b-.
+
 /*
  * Up until early 2004, use of .align within a function caused bad unwind info.
  * TEXT_ALIGN(n) expands into ".align n" if a fixed GAS is available or into nothing

include/asm-ia64/kregs.h

@@ -29,8 +29,7 @@
  */
 #define IA64_TR_KERNEL		0	/* itr0, dtr0: maps kernel image (code & data) */
 #define IA64_TR_PALCODE		1	/* itr1: maps PALcode as required by EFI */
-#define IA64_TR_PERCPU_DATA	1	/* dtr1: percpu data */
-#define IA64_TR_CURRENT_STACK	2	/* dtr2: maps kernel's memory- & register-stacks */
+#define IA64_TR_CURRENT_STACK	1	/* dtr1: maps kernel's memory- & register-stacks */
 
 /* Processor status register bits: */
 #define IA64_PSR_BE_BIT		1

include/asm-ia64/patch.h

@@ -20,6 +20,7 @@ extern void ia64_patch_imm60 (u64 insn_addr, u64 val);		/* patch "brl" w/ip-rel
 
 extern void ia64_patch_mckinley_e9 (unsigned long start, unsigned long end);
 extern void ia64_patch_vtop (unsigned long start, unsigned long end);
+extern void ia64_patch_phys_stack_reg(unsigned long val);
 extern void ia64_patch_gate (void);
 
 #endif /* _ASM_IA64_PATCH_H */

include/asm-ia64/processor.h

@@ -19,6 +19,7 @@
 #include <asm/ptrace.h>
 #include <asm/ustack.h>
 
+#define IA64_NUM_PHYS_STACK_REG	96
 #define IA64_NUM_DBG_REGS	8
 
 #define DEFAULT_MAP_BASE	__IA64_UL_CONST(0x2000000000000000)

include/asm-ia64/sections.h

@@ -11,6 +11,7 @@
 extern char __per_cpu_start[], __per_cpu_end[], __phys_per_cpu_start[];
 extern char __start___vtop_patchlist[], __end___vtop_patchlist[];
 extern char __start___mckinley_e9_bundles[], __end___mckinley_e9_bundles[];
+extern char __start___phys_stack_reg_patchlist[], __end___phys_stack_reg_patchlist[];
 extern char __start_gate_section[];
 extern char __start_gate_mckinley_e9_patchlist[], __end_gate_mckinley_e9_patchlist[];
 extern char __start_gate_vtop_patchlist[], __end_gate_vtop_patchlist[];