Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc

Pull powerpc updates from Ben Herrenschmidt:
 "So here's my next branch for powerpc.  A bit late as I was on vacation
  last week.  It's mostly the same stuff that was in next already, I
  just added two patches today which are the wiring up of lockref for
  powerpc, which for some reason fell through the cracks last time and
  is trivial.

  The highlights are, in addition to a bunch of bug fixes:

   - Reworked Machine Check handling on kernels running without a
     hypervisor (or acting as a hypervisor).  Provides hooks to handle
     some errors in real mode such as TLB errors, handle SLB errors,
     etc...

   - Support for retrieving memory error information from the service
     processor on IBM servers running without a hypervisor and routing
     them to the memory poison infrastructure.

   - _PAGE_NUMA support on server processors

   - 32-bit BookE relocatable kernel support

   - FSL e6500 hardware tablewalk support

   - A bunch of new/revived board support

   - FSL e6500 deeper idle states and altivec powerdown support

  You'll notice a generic mm change here, it has been acked by the
  relevant authorities and is a pre-req for our _PAGE_NUMA support"

* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc: (121 commits)
  powerpc: Implement arch_spin_is_locked() using arch_spin_value_unlocked()
  powerpc: Add support for the optimised lockref implementation
  powerpc/powernv: Call OPAL sync before kexec'ing
  powerpc/eeh: Escalate error on non-existing PE
  powerpc/eeh: Handle multiple EEH errors
  powerpc: Fix transactional FP/VMX/VSX unavailable handlers
  powerpc: Don't corrupt transactional state when using FP/VMX in kernel
  powerpc: Reclaim two unused thread_info flag bits
  powerpc: Fix races with irq_work
  Move precessing of MCE queued event out from syscall exit path.
  pseries/cpuidle: Remove redundant call to ppc64_runlatch_off() in cpu idle routines
  powerpc: Make add_system_ram_resources() __init
  powerpc: add SATA_MV to ppc64_defconfig
  powerpc/powernv: Increase candidate fw image size
  powerpc: Add debug checks to catch invalid cpu-to-node mappings
  powerpc: Fix the setup of CPU-to-Node mappings during CPU online
  powerpc/iommu: Don't detach device without IOMMU group
  powerpc/eeh: Hotplug improvement
  powerpc/eeh: Call opal_pci_reinit() on powernv for restoring config space
  powerpc/eeh: Add restore_config operation
  ...

1 files changed, 236 insertions, 30 deletions

diff --git a/arch/powerpc/kernel/head_fsl_booke.S b/arch/powerpc/kernel/head_fsl_booke.S
index f45726a1d96..b497188a94a 100644
--- a/arch/powerpc/kernel/head_fsl_booke.S
+++ b/arch/powerpc/kernel/head_fsl_booke.S
@@ -65,29 +65,78 @@ _ENTRY(_start);
 	nop
 
 	/* Translate device tree address to physical, save in r30/r31 */
-	mfmsr	r16
-	mfspr	r17,SPRN_PID
-	rlwinm	r17,r17,16,0x3fff0000	/* turn PID into MAS6[SPID] */
-	rlwimi	r17,r16,28,0x00000001	/* turn MSR[DS] into MAS6[SAS] */
-	mtspr	SPRN_MAS6,r17
-
-	tlbsx	0,r3			/* must succeed */
-
-	mfspr	r16,SPRN_MAS1
-	mfspr	r20,SPRN_MAS3
-	rlwinm	r17,r16,25,0x1f		/* r17 = log2(page size) */
-	li	r18,1024
-	slw	r18,r18,r17		/* r18 = page size */
-	addi	r18,r18,-1
-	and	r19,r3,r18		/* r19 = page offset */
-	andc	r31,r20,r18		/* r31 = page base */
-	or	r31,r31,r19		/* r31 = devtree phys addr */
-	mfspr	r30,SPRN_MAS7
+	bl	get_phys_addr
+	mr	r30,r3
+	mr	r31,r4
 
 	li	r25,0			/* phys kernel start (low) */
 	li	r24,0			/* CPU number */
 	li	r23,0			/* phys kernel start (high) */
 
+#ifdef CONFIG_RELOCATABLE
+	LOAD_REG_ADDR_PIC(r3, _stext)	/* Get our current runtime base */
+
+	/* Translate _stext address to physical, save in r23/r25 */
+	bl	get_phys_addr
+	mr	r23,r3
+	mr	r25,r4
+
+	bl	0f
+0:	mflr	r8
+	addis	r3,r8,(is_second_reloc - 0b)@ha
+	lwz	r19,(is_second_reloc - 0b)@l(r3)
+
+	/* Check if this is the second relocation. */
+	cmpwi	r19,1
+	bne	1f
+
+	/*
+	 * For the second relocation, we already get the real memstart_addr
+	 * from device tree. So we will map PAGE_OFFSET to memstart_addr,
+	 * then the virtual address of start kernel should be:
+	 *          PAGE_OFFSET + (kernstart_addr - memstart_addr)
+	 * Since the offset between kernstart_addr and memstart_addr should
+	 * never be beyond 1G, so we can just use the lower 32bit of them
+	 * for the calculation.
+	 */
+	lis	r3,PAGE_OFFSET@h
+
+	addis	r4,r8,(kernstart_addr - 0b)@ha
+	addi	r4,r4,(kernstart_addr - 0b)@l
+	lwz	r5,4(r4)
+
+	addis	r6,r8,(memstart_addr - 0b)@ha
+	addi	r6,r6,(memstart_addr - 0b)@l
+	lwz	r7,4(r6)
+
+	subf	r5,r7,r5
+	add	r3,r3,r5
+	b	2f
+
+1:
+	/*
+	 * We have the runtime (virutal) address of our base.
+	 * We calculate our shift of offset from a 64M page.
+	 * We could map the 64M page we belong to at PAGE_OFFSET and
+	 * get going from there.
+	 */
+	lis	r4,KERNELBASE@h
+	ori	r4,r4,KERNELBASE@l
+	rlwinm	r6,r25,0,0x3ffffff		/* r6 = PHYS_START % 64M */
+	rlwinm	r5,r4,0,0x3ffffff		/* r5 = KERNELBASE % 64M */
+	subf	r3,r5,r6			/* r3 = r6 - r5 */
+	add	r3,r4,r3			/* Required Virtual Address */
+
+2:	bl	relocate
+
+	/*
+	 * For the second relocation, we already set the right tlb entries
+	 * for the kernel space, so skip the code in fsl_booke_entry_mapping.S
+	*/
+	cmpwi	r19,1
+	beq	set_ivor
+#endif
+
 /* We try to not make any assumptions about how the boot loader
  * setup or used the TLBs.  We invalidate all mappings from the
  * boot loader and load a single entry in TLB1[0] to map the
@@ -113,6 +162,7 @@ _ENTRY(__early_start)
 #include "fsl_booke_entry_mapping.S"
 #undef ENTRY_MAPPING_BOOT_SETUP
 
+set_ivor:
 	/* Establish the interrupt vector offsets */
 	SET_IVOR(0,  CriticalInput);
 	SET_IVOR(1,  MachineCheck);
@@ -166,8 +216,7 @@ _ENTRY(__early_start)
 	/* Check to see if we're the second processor, and jump
 	 * to the secondary_start code if so
 	 */
-	lis	r24, boot_cpuid@h
-	ori	r24, r24, boot_cpuid@l
+	LOAD_REG_ADDR_PIC(r24, boot_cpuid)
 	lwz	r24, 0(r24)
 	cmpwi	r24, -1
 	mfspr   r24,SPRN_PIR
@@ -197,6 +246,18 @@ _ENTRY(__early_start)
 
 	bl	early_init
 
+#ifdef CONFIG_RELOCATABLE
+	mr	r3,r30
+	mr	r4,r31
+#ifdef CONFIG_PHYS_64BIT
+	mr	r5,r23
+	mr	r6,r25
+#else
+	mr	r5,r25
+#endif
+	bl	relocate_init
+#endif
+
 #ifdef CONFIG_DYNAMIC_MEMSTART
 	lis	r3,kernstart_addr@ha
 	la	r3,kernstart_addr@l(r3)
@@ -856,6 +917,33 @@ KernelSPE:
 #endif /* CONFIG_SPE */
 
 /*
+ * Translate the effec addr in r3 to phys addr. The phys addr will be put
+ * into r3(higher 32bit) and r4(lower 32bit)
+ */
+get_phys_addr:
+	mfmsr	r8
+	mfspr	r9,SPRN_PID
+	rlwinm	r9,r9,16,0x3fff0000	/* turn PID into MAS6[SPID] */
+	rlwimi	r9,r8,28,0x00000001	/* turn MSR[DS] into MAS6[SAS] */
+	mtspr	SPRN_MAS6,r9
+
+	tlbsx	0,r3			/* must succeed */
+
+	mfspr	r8,SPRN_MAS1
+	mfspr	r12,SPRN_MAS3
+	rlwinm	r9,r8,25,0x1f		/* r9 = log2(page size) */
+	li	r10,1024
+	slw	r10,r10,r9		/* r10 = page size */
+	addi	r10,r10,-1
+	and	r11,r3,r10		/* r11 = page offset */
+	andc	r4,r12,r10		/* r4 = page base */
+	or	r4,r4,r11		/* r4 = devtree phys addr */
+#ifdef CONFIG_PHYS_64BIT
+	mfspr	r3,SPRN_MAS7
+#endif
+	blr
+
+/*
  * Global functions
  */
 
@@ -1057,24 +1145,36 @@ _GLOBAL(__flush_disable_L1)
 /* When we get here, r24 needs to hold the CPU # */
 	.globl __secondary_start
 __secondary_start:
-	lis	r3,__secondary_hold_acknowledge@h
-	ori	r3,r3,__secondary_hold_acknowledge@l
-	stw	r24,0(r3)
-
-	li	r3,0
-	mr	r4,r24		/* Why? */
-	bl	call_setup_cpu
-
-	lis	r3,tlbcam_index@ha
-	lwz	r3,tlbcam_index@l(r3)
+	LOAD_REG_ADDR_PIC(r3, tlbcam_index)
+	lwz	r3,0(r3)
 	mtctr	r3
 	li	r26,0		/* r26 safe? */
 
+	bl	switch_to_as1
+	mr	r27,r3		/* tlb entry */
 	/* Load each CAM entry */
 1:	mr	r3,r26
 	bl	loadcam_entry
 	addi	r26,r26,1
 	bdnz	1b
+	mr	r3,r27		/* tlb entry */
+	LOAD_REG_ADDR_PIC(r4, memstart_addr)
+	lwz	r4,0(r4)
+	mr	r5,r25		/* phys kernel start */
+	rlwinm	r5,r5,0,~0x3ffffff	/* aligned 64M */
+	subf	r4,r5,r4	/* memstart_addr - phys kernel start */
+	li	r5,0		/* no device tree */
+	li	r6,0		/* not boot cpu */
+	bl	restore_to_as0
+
+
+	lis	r3,__secondary_hold_acknowledge@h
+	ori	r3,r3,__secondary_hold_acknowledge@l
+	stw	r24,0(r3)
+
+	li	r3,0
+	mr	r4,r24		/* Why? */
+	bl	call_setup_cpu
 
 	/* get current_thread_info and current */
 	lis	r1,secondary_ti@ha
@@ -1111,6 +1211,112 @@ __secondary_hold_acknowledge:
 #endif
 
 /*
+ * Create a tlb entry with the same effective and physical address as
+ * the tlb entry used by the current running code. But set the TS to 1.
+ * Then switch to the address space 1. It will return with the r3 set to
+ * the ESEL of the new created tlb.
+ */
+_GLOBAL(switch_to_as1)
+	mflr	r5
+
+	/* Find a entry not used */
+	mfspr	r3,SPRN_TLB1CFG
+	andi.	r3,r3,0xfff
+	mfspr	r4,SPRN_PID
+	rlwinm	r4,r4,16,0x3fff0000	/* turn PID into MAS6[SPID] */
+	mtspr	SPRN_MAS6,r4
+1:	lis	r4,0x1000		/* Set MAS0(TLBSEL) = 1 */
+	addi	r3,r3,-1
+	rlwimi	r4,r3,16,4,15		/* Setup MAS0 = TLBSEL | ESEL(r3) */
+	mtspr	SPRN_MAS0,r4
+	tlbre
+	mfspr	r4,SPRN_MAS1
+	andis.	r4,r4,MAS1_VALID@h
+	bne	1b
+
+	/* Get the tlb entry used by the current running code */
+	bl	0f
+0:	mflr	r4
+	tlbsx	0,r4
+
+	mfspr	r4,SPRN_MAS1
+	ori	r4,r4,MAS1_TS		/* Set the TS = 1 */
+	mtspr	SPRN_MAS1,r4
+
+	mfspr	r4,SPRN_MAS0
+	rlwinm	r4,r4,0,~MAS0_ESEL_MASK
+	rlwimi	r4,r3,16,4,15		/* Setup MAS0 = TLBSEL | ESEL(r3) */
+	mtspr	SPRN_MAS0,r4
+	tlbwe
+	isync
+	sync
+
+	mfmsr	r4
+	ori	r4,r4,MSR_IS | MSR_DS
+	mtspr	SPRN_SRR0,r5
+	mtspr	SPRN_SRR1,r4
+	sync
+	rfi
+
+/*
+ * Restore to the address space 0 and also invalidate the tlb entry created
+ * by switch_to_as1.
+ * r3 - the tlb entry which should be invalidated
+ * r4 - __pa(PAGE_OFFSET in AS1) - __pa(PAGE_OFFSET in AS0)
+ * r5 - device tree virtual address. If r4 is 0, r5 is ignored.
+ * r6 - boot cpu
+*/
+_GLOBAL(restore_to_as0)
+	mflr	r0
+
+	bl	0f
+0:	mflr	r9
+	addi	r9,r9,1f - 0b
+
+	/*
+	 * We may map the PAGE_OFFSET in AS0 to a different physical address,
+	 * so we need calculate the right jump and device tree address based
+	 * on the offset passed by r4.
+	 */
+	add	r9,r9,r4
+	add	r5,r5,r4
+	add	r0,r0,r4
+
+2:	mfmsr	r7
+	li	r8,(MSR_IS | MSR_DS)
+	andc	r7,r7,r8
+
+	mtspr	SPRN_SRR0,r9
+	mtspr	SPRN_SRR1,r7
+	sync
+	rfi
+
+	/* Invalidate the temporary tlb entry for AS1 */
+1:	lis	r9,0x1000		/* Set MAS0(TLBSEL) = 1 */
+	rlwimi	r9,r3,16,4,15		/* Setup MAS0 = TLBSEL | ESEL(r3) */
+	mtspr	SPRN_MAS0,r9
+	tlbre
+	mfspr	r9,SPRN_MAS1
+	rlwinm	r9,r9,0,2,31		/* Clear MAS1 Valid and IPPROT */
+	mtspr	SPRN_MAS1,r9
+	tlbwe
+	isync
+
+	cmpwi	r4,0
+	cmpwi	cr1,r6,0
+	cror	eq,4*cr1+eq,eq
+	bne	3f			/* offset != 0 && is_boot_cpu */
+	mtlr	r0
+	blr
+
+	/*
+	 * The PAGE_OFFSET will map to a different physical address,
+	 * jump to _start to do another relocation again.
+	*/
+3:	mr	r3,r5
+	bl	_start
+
+/*
  * We put a few things here that have to be page-aligned. This stuff
  * goes at the beginning of the data segment, which is page-aligned.
  */