[PATCH v2 0/5] powerpc/64: avoid GOT addressing, don't put data in TOC

[PATCH v2 0/5] powerpc/64: avoid GOT addressing, don't put data in TOC

[Nicholas Piggin]

This is a cleaned up set of the initial prep patches from the pcrel
series, dealing with regularising addressing variables from asm
and using helper macros more consistently.

Changes since v1:
- Use "REGS" stack frame marker for 64-bit, leaving 32-bit unchanged.
- Fix PPC64 32-bit boot wrapper compile.

Nicholas Piggin (5):
  powerpc/64: use 32-bit immediate for STACK_FRAME_REGS_MARKER
  powerpc/64: asm use consistent global variable declaration and access
  powerpc/64: switch asm helpers from GOT to TOC relative addressing
  powerpc/64: provide a helper macro to load r2 with the kernel TOC
  powerpc/64e: provide an addressing macro for use with TOC in alternate
    register

 arch/powerpc/boot/opal-calls.S                |  6 ++--
 arch/powerpc/boot/ppc_asm.h                   | 10 ++++++
 arch/powerpc/include/asm/ppc_asm.h            | 20 ++++++++++-
 arch/powerpc/include/asm/ptrace.h             |  4 +--
 arch/powerpc/kernel/entry_32.S                |  6 ++--
 arch/powerpc/kernel/exceptions-64e.S          | 34 ++++++++-----------
 arch/powerpc/kernel/exceptions-64s.S          |  8 ++---
 arch/powerpc/kernel/head_64.S                 | 11 ++----
 arch/powerpc/kernel/interrupt_64.S            | 28 +++++----------
 arch/powerpc/kernel/optprobes_head.S          |  2 +-
 arch/powerpc/kernel/swsusp_asm64.S            | 16 +++------
 arch/powerpc/kernel/trace/ftrace_low.S        |  2 +-
 arch/powerpc/kernel/trace/ftrace_mprofile.S   |  6 ++--
 arch/powerpc/kernel/vector.S                  | 15 ++++----
 arch/powerpc/kvm/book3s_64_entry.S            |  2 +-
 arch/powerpc/kvm/book3s_hv_rmhandlers.S       |  4 +--
 arch/powerpc/kvm/tm.S                         |  2 +-
 arch/powerpc/lib/copypage_64.S                |  7 +---
 arch/powerpc/lib/string_64.S                  |  7 +---
 arch/powerpc/mm/nohash/tlb_low_64e.S          |  4 +--
 arch/powerpc/perf/bhrb.S                      |  2 +-
 .../powerpc/platforms/powernv/opal-wrappers.S |  2 +-
 arch/powerpc/platforms/pseries/hvCall.S       |  4 +--
 arch/powerpc/xmon/spr_access.S                |  4 +--
 24 files changed, 96 insertions(+), 110 deletions(-)

-- 
2.37.2

[PATCH v2 1/5] powerpc/64: use 32-bit immediate for STACK_FRAME_REGS_MARKER

[Nicholas Piggin]

Using a 32-bit constant for this marker allows it to be loaded with
two ALU instructions, like 32-bit. This avoids a TOC entry and a
TOC load that depends on the r2 value that has just been loaded from
the PACA.

This changes the value for 32-bit as well, so both have the same
value in the low 4 bytes and 64-bit has 0xffffffff in the top bytes.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
 arch/powerpc/include/asm/ptrace.h    | 4 ++--
 arch/powerpc/kernel/entry_32.S       | 6 +++---
 arch/powerpc/kernel/exceptions-64e.S | 8 +-------
 arch/powerpc/kernel/exceptions-64s.S | 2 +-
 arch/powerpc/kernel/head_64.S        | 7 -------
 arch/powerpc/kernel/interrupt_64.S   | 6 +++---
 6 files changed, 10 insertions(+), 23 deletions(-)

diff --git a/arch/powerpc/include/asm/ptrace.h b/arch/powerpc/include/asm/ptrace.h
index a03403695cd4..5b496e589d54 100644
--- a/arch/powerpc/include/asm/ptrace.h
+++ b/arch/powerpc/include/asm/ptrace.h
@@ -99,6 +99,8 @@ struct pt_regs
 
 #define STACK_FRAME_WITH_PT_REGS (STACK_FRAME_OVERHEAD + sizeof(struct pt_regs))
 
+#define STACK_FRAME_REGS_MARKER	ASM_CONST(0x72656773)
+
 #ifdef __powerpc64__
 
 /*
@@ -115,7 +117,6 @@ struct pt_regs
 
 #define STACK_FRAME_OVERHEAD	112	/* size of minimum stack frame */
 #define STACK_FRAME_LR_SAVE	2	/* Location of LR in stack frame */
-#define STACK_FRAME_REGS_MARKER	ASM_CONST(0x7265677368657265)
 #define STACK_INT_FRAME_SIZE	(sizeof(struct pt_regs) + \
 				 STACK_FRAME_OVERHEAD + KERNEL_REDZONE_SIZE)
 #define STACK_FRAME_MARKER	12
@@ -136,7 +137,6 @@ struct pt_regs
 #define KERNEL_REDZONE_SIZE	0
 #define STACK_FRAME_OVERHEAD	16	/* size of minimum stack frame */
 #define STACK_FRAME_LR_SAVE	1	/* Location of LR in stack frame */
-#define STACK_FRAME_REGS_MARKER	ASM_CONST(0x72656773)
 #define STACK_INT_FRAME_SIZE	(sizeof(struct pt_regs) + STACK_FRAME_OVERHEAD)
 #define STACK_FRAME_MARKER	2
 #define STACK_FRAME_MIN_SIZE	STACK_FRAME_OVERHEAD
diff --git a/arch/powerpc/kernel/entry_32.S b/arch/powerpc/kernel/entry_32.S
index 1d599df6f169..c2516f982cfa 100644
--- a/arch/powerpc/kernel/entry_32.S
+++ b/arch/powerpc/kernel/entry_32.S
@@ -265,7 +265,7 @@ fast_exception_return:
 	mtcr	r10
 	lwz	r10,_LINK(r11)
 	mtlr	r10
-	/* Clear the exception_marker on the stack to avoid confusing stacktrace */
+	/* Clear the exception marker on the stack to avoid confusing stacktrace */
 	li	r10, 0
 	stw	r10, 8(r11)
 	REST_GPR(10, r11)
@@ -322,7 +322,7 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS)
 	li	r0,0
 
 	/*
-	 * Leaving a stale exception_marker on the stack can confuse
+	 * Leaving a stale exception marker on the stack can confuse
 	 * the reliable stack unwinder later on. Clear it.
 	 */
 	stw	r0,8(r1)
@@ -374,7 +374,7 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS)
 	mtspr	SPRN_XER,r5
 
 	/*
-	 * Leaving a stale exception_marker on the stack can confuse
+	 * Leaving a stale exception marker on the stack can confuse
 	 * the reliable stack unwinder later on. Clear it.
 	 */
 	stw	r0,8(r1)
diff --git a/arch/powerpc/kernel/exceptions-64e.S b/arch/powerpc/kernel/exceptions-64e.S
index 67dc4e3179a0..08a3139079b7 100644
--- a/arch/powerpc/kernel/exceptions-64e.S
+++ b/arch/powerpc/kernel/exceptions-64e.S
@@ -389,7 +389,7 @@ exc_##n##_common:							    \
 	ld	r9,excf+EX_R1(r13);	/* load orig r1 back from PACA */   \
 	lwz	r10,excf+EX_CR(r13);	/* load orig CR back from PACA	*/  \
 	lbz	r11,PACAIRQSOFTMASK(r13); /* get current IRQ softe */	    \
-	ld	r12,exception_marker@toc(r2);				    \
+	LOAD_REG_IMMEDIATE(r12, STACK_FRAME_REGS_MARKER);		    \
 	li	r0,0;							    \
 	std	r3,GPR10(r1);		/* save r10 to stackframe */	    \
 	std	r4,GPR11(r1);		/* save r11 to stackframe */	    \
@@ -470,12 +470,6 @@ exc_##n##_bad_stack:							    \
 	bl	hdlr;							\
 	b	interrupt_return
 
-/* This value is used to mark exception frames on the stack. */
-	.section	".toc","aw"
-exception_marker:
-	.tc	ID_EXC_MARKER[TC],STACK_FRAME_REGS_MARKER
-
-
 /*
  * And here we have the exception vectors !
  */
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 3d0dc133a9ae..9d375ea58add 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -589,7 +589,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
 	li	r9,IVEC
 	std	r9,_TRAP(r1)		/* set trap number		*/
 	li	r10,0
-	ld	r11,exception_marker@toc(r2)
+	LOAD_REG_IMMEDIATE(r11, STACK_FRAME_REGS_MARKER)
 	std	r10,RESULT(r1)		/* clear regs->result		*/
 	std	r11,STACK_FRAME_OVERHEAD-16(r1) /* mark the frame	*/
 .endm
diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index cf2c08902c05..cac3e1b58360 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -192,13 +192,6 @@ __secondary_hold:
 #endif
 CLOSE_FIXED_SECTION(first_256B)
 
-/* This value is used to mark exception frames on the stack. */
-	.section ".toc","aw"
-/* This value is used to mark exception frames on the stack. */
-exception_marker:
-	.tc	ID_EXC_MARKER[TC],STACK_FRAME_REGS_MARKER
-	.previous
-
 /*
  * On server, we include the exception vectors code here as it
  * relies on absolute addressing which is only possible within
diff --git a/arch/powerpc/kernel/interrupt_64.S b/arch/powerpc/kernel/interrupt_64.S
index ce25b28cf418..ee2d2d410c5a 100644
--- a/arch/powerpc/kernel/interrupt_64.S
+++ b/arch/powerpc/kernel/interrupt_64.S
@@ -92,7 +92,7 @@ _ASM_NOKPROBE_SYMBOL(system_call_vectored_\name)
 	std	r11,_TRAP(r1)
 	std	r12,_CCR(r1)
 	addi	r10,r1,STACK_FRAME_OVERHEAD
-	ld	r11,exception_marker@toc(r2)
+	LOAD_REG_IMMEDIATE(r11, STACK_FRAME_REGS_MARKER)
 	std	r11,-16(r10)		/* "regshere" marker */
 
 BEGIN_FTR_SECTION
@@ -276,7 +276,7 @@ END_BTB_FLUSH_SECTION
 	std	r11,_TRAP(r1)
 	std	r12,_CCR(r1)
 	addi	r10,r1,STACK_FRAME_OVERHEAD
-	ld	r11,exception_marker@toc(r2)
+	LOAD_REG_IMMEDIATE(r11, STACK_FRAME_REGS_MARKER)
 	std	r11,-16(r10)		/* "regshere" marker */
 
 #ifdef CONFIG_PPC_BOOK3S
@@ -619,7 +619,7 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS)
 	mtspr	SPRN_XER,r5
 
 	/*
-	 * Leaving a stale exception_marker on the stack can confuse
+	 * Leaving a stale STACK_FRAME_REGS_MARKER on the stack can confuse
 	 * the reliable stack unwinder later on. Clear it.
 	 */
 	std	r0,STACK_FRAME_OVERHEAD-16(r1)
-- 
2.37.2

[PATCH v2 2/5] powerpc/64: asm use consistent global variable declaration and access

[Nicholas Piggin]

Use helper macros to access global variables, and place them in .data
sections rather than in .toc. Putting addresses in TOC is not required
because the kernel is linked with a single TOC.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
 arch/powerpc/boot/opal-calls.S              |  6 +++---
 arch/powerpc/boot/ppc_asm.h                 |  9 +++++++++
 arch/powerpc/kernel/interrupt_64.S          | 10 ----------
 arch/powerpc/kernel/swsusp_asm64.S          | 16 +++++-----------
 arch/powerpc/kernel/trace/ftrace_mprofile.S |  3 +--
 arch/powerpc/kernel/vector.S                | 15 +++++++--------
 arch/powerpc/lib/copypage_64.S              |  7 +------
 arch/powerpc/lib/string_64.S                |  7 +------
 arch/powerpc/perf/bhrb.S                    |  2 +-
 arch/powerpc/platforms/pseries/hvCall.S     |  4 ++--
 arch/powerpc/xmon/spr_access.S              |  4 ++--
 11 files changed, 32 insertions(+), 51 deletions(-)

diff --git a/arch/powerpc/boot/opal-calls.S b/arch/powerpc/boot/opal-calls.S
index ad0e15d930c4..1f2f330a459e 100644
--- a/arch/powerpc/boot/opal-calls.S
+++ b/arch/powerpc/boot/opal-calls.S
@@ -16,7 +16,7 @@ opal_kentry:
 	li	r5, 0
 	li	r6, 0
 	li	r7, 0
-	ld	r11,opal@got(r2)
+	LOAD_REG_ADDR(r11, opal)
 	ld	r8,0(r11)
 	ld	r9,8(r11)
 	bctr
@@ -35,7 +35,7 @@ opal_call:
 	mr	r13,r2
 
 	/* Set opal return address */
-	ld	r11,opal_return@got(r2)
+	LOAD_REG_ADDR(r11, opal_return)
 	mtlr	r11
 	mfmsr	r12
 
@@ -45,7 +45,7 @@ opal_call:
 	mtspr	SPRN_HSRR1,r12
 
 	/* load the opal call entry point and base */
-	ld	r11,opal@got(r2)
+	LOAD_REG_ADDR(r11, opal)
 	ld	r12,8(r11)
 	ld	r2,0(r11)
 	mtspr	SPRN_HSRR0,r12
diff --git a/arch/powerpc/boot/ppc_asm.h b/arch/powerpc/boot/ppc_asm.h
index 192b97523b05..2824a3e32aab 100644
--- a/arch/powerpc/boot/ppc_asm.h
+++ b/arch/powerpc/boot/ppc_asm.h
@@ -84,4 +84,13 @@
 #define MFTBU(dest)			mfspr dest, SPRN_TBRU
 #endif
 
+#ifdef CONFIG_PPC64_BOOT_WRAPPER
+#define LOAD_REG_ADDR(reg,name)			\
+	ld	reg,name@got(r2)
+#else
+#define LOAD_REG_ADDR(reg,name)			\
+	lis	reg,name@ha;			\
+	addi	reg,reg,name@l
+#endif
+
 #endif /* _PPC64_PPC_ASM_H */
diff --git a/arch/powerpc/kernel/interrupt_64.S b/arch/powerpc/kernel/interrupt_64.S
index ee2d2d410c5a..fbfb72a62da9 100644
--- a/arch/powerpc/kernel/interrupt_64.S
+++ b/arch/powerpc/kernel/interrupt_64.S
@@ -13,16 +13,6 @@
 #include <asm/ppc_asm.h>
 #include <asm/ptrace.h>
 
-	.section	".toc","aw"
-SYS_CALL_TABLE:
-	.tc sys_call_table[TC],sys_call_table
-
-#ifdef CONFIG_COMPAT
-COMPAT_SYS_CALL_TABLE:
-	.tc compat_sys_call_table[TC],compat_sys_call_table
-#endif
-	.previous
-
 	.align 7
 
 .macro DEBUG_SRR_VALID srr
diff --git a/arch/powerpc/kernel/swsusp_asm64.S b/arch/powerpc/kernel/swsusp_asm64.S
index 9f1903c7f540..f645652c2654 100644
--- a/arch/powerpc/kernel/swsusp_asm64.S
+++ b/arch/powerpc/kernel/swsusp_asm64.S
@@ -76,16 +76,10 @@
 swsusp_save_area:
 	.space SL_SIZE
 
-	.section ".toc","aw"
-swsusp_save_area_ptr:
-	.tc	swsusp_save_area[TC],swsusp_save_area
-restore_pblist_ptr:
-	.tc	restore_pblist[TC],restore_pblist
-
 	.section .text
 	.align  5
 _GLOBAL(swsusp_arch_suspend)
-	ld	r11,swsusp_save_area_ptr@toc(r2)
+	LOAD_REG_ADDR(r11, swsusp_save_area)
 	SAVE_SPECIAL(LR)
 	SAVE_REGISTER(r1)
 	SAVE_SPECIAL(CR)
@@ -131,7 +125,7 @@ END_FW_FTR_SECTION_IFCLR(FW_FEATURE_LPAR)
 	bl swsusp_save
 
 	/* restore LR */
-	ld	r11,swsusp_save_area_ptr@toc(r2)
+	LOAD_REG_ADDR(r11, swsusp_save_area)
 	RESTORE_SPECIAL(LR)
 	addi	r1,r1,128
 
@@ -145,7 +139,7 @@ BEGIN_FTR_SECTION
 END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
 	sync
 
-	ld	r12,restore_pblist_ptr@toc(r2)
+	LOAD_REG_ADDR(r11, restore_pblist)
 	ld	r12,0(r12)
 
 	cmpdi	r12,0
@@ -187,7 +181,7 @@ nothing_to_copy:
 	tlbia
 #endif
 
-	ld	r11,swsusp_save_area_ptr@toc(r2)
+	LOAD_REG_ADDR(r11, swsusp_save_area)
 
 	RESTORE_SPECIAL(CR)
 
@@ -265,7 +259,7 @@ END_FW_FTR_SECTION_IFCLR(FW_FEATURE_LPAR)
 	bl	do_after_copyback
 	addi	r1,r1,128
 
-	ld	r11,swsusp_save_area_ptr@toc(r2)
+	LOAD_REG_ADDR(r11, swsusp_save_area)
 	RESTORE_SPECIAL(LR)
 
 	li	r3, 0
diff --git a/arch/powerpc/kernel/trace/ftrace_mprofile.S b/arch/powerpc/kernel/trace/ftrace_mprofile.S
index 4fa23e260cab..33fcfb2eaded 100644
--- a/arch/powerpc/kernel/trace/ftrace_mprofile.S
+++ b/arch/powerpc/kernel/trace/ftrace_mprofile.S
@@ -85,8 +85,7 @@
 	std	r2, STK_GOT(r1)
 	ld	r2,PACATOC(r13)	/* get kernel TOC in r2 */
 
-	addis	r3,r2,function_trace_op@toc@ha
-	addi	r3,r3,function_trace_op@toc@l
+	LOAD_REG_ADDR(r3, function_trace_op)
 	ld	r5,0(r3)
 #else
 	lis	r3,function_trace_op@ha
diff --git a/arch/powerpc/kernel/vector.S b/arch/powerpc/kernel/vector.S
index 5cc24d8cce94..5cf64740edb8 100644
--- a/arch/powerpc/kernel/vector.S
+++ b/arch/powerpc/kernel/vector.S
@@ -155,8 +155,8 @@ _GLOBAL(load_up_vsx)
  * usage of floating-point registers.  These routines must be called
  * with preempt disabled.
  */
-#ifdef CONFIG_PPC32
 	.data
+#ifdef CONFIG_PPC32
 fpzero:
 	.long	0
 fpone:
@@ -169,18 +169,17 @@ fphalf:
 	lfs	fr,name@l(r11)
 #else
 
-	.section ".toc","aw"
 fpzero:
-	.tc	FD_0_0[TC],0
+	.quad	0
 fpone:
-	.tc	FD_3ff00000_0[TC],0x3ff0000000000000	/* 1.0 */
+	.quad	0x3ff0000000000000	/* 1.0 */
 fphalf:
-	.tc	FD_3fe00000_0[TC],0x3fe0000000000000	/* 0.5 */
+	.quad	0x3fe0000000000000	/* 0.5 */
 
-#define LDCONST(fr, name)	\
-	lfd	fr,name@toc(r2)
+#define LDCONST(fr, name)		\
+	addis	r11,r2,name@toc@ha;	\
+	lfd	fr,name@toc@l(r11)
 #endif
-
 	.text
 /*
  * Internal routine to enable floating point and set FPSCR to 0.
diff --git a/arch/powerpc/lib/copypage_64.S b/arch/powerpc/lib/copypage_64.S
index d1091b5ee5da..6812cb19d04a 100644
--- a/arch/powerpc/lib/copypage_64.S
+++ b/arch/powerpc/lib/copypage_64.S
@@ -9,11 +9,6 @@
 #include <asm/export.h>
 #include <asm/feature-fixups.h>
 
-        .section        ".toc","aw"
-PPC64_CACHES:
-        .tc             ppc64_caches[TC],ppc64_caches
-        .section        ".text"
-
 _GLOBAL_TOC(copy_page)
 BEGIN_FTR_SECTION
 	lis	r5,PAGE_SIZE@h
@@ -24,7 +19,7 @@ FTR_SECTION_ELSE
 ALT_FTR_SECTION_END_IFCLR(CPU_FTR_VMX_COPY)
 	ori	r5,r5,PAGE_SIZE@l
 BEGIN_FTR_SECTION
-	ld      r10,PPC64_CACHES@toc(r2)
+	LOAD_REG_ADDR(r10, ppc64_caches)
 	lwz	r11,DCACHEL1LOGBLOCKSIZE(r10)	/* log2 of cache block size */
 	lwz     r12,DCACHEL1BLOCKSIZE(r10)	/* get cache block size */
 	li	r9,0
diff --git a/arch/powerpc/lib/string_64.S b/arch/powerpc/lib/string_64.S
index 169872bc0892..df41ce06f86b 100644
--- a/arch/powerpc/lib/string_64.S
+++ b/arch/powerpc/lib/string_64.S
@@ -11,11 +11,6 @@
 #include <asm/asm-offsets.h>
 #include <asm/export.h>
 
-	.section	".toc","aw"
-PPC64_CACHES:
-	.tc		ppc64_caches[TC],ppc64_caches
-	.section	".text"
-
 /**
  * __arch_clear_user: - Zero a block of memory in user space, with less checking.
  * @to:   Destination address, in user space.
@@ -133,7 +128,7 @@ err1;	stb	r0,0(r3)
 	blr
 
 .Llong_clear:
-	ld	r5,PPC64_CACHES@toc(r2)
+	LOAD_REG_ADDR(r5, ppc64_caches)
 
 	bf	cr7*4+0,11f
 err2;	std	r0,0(r3)
diff --git a/arch/powerpc/perf/bhrb.S b/arch/powerpc/perf/bhrb.S
index 1aa3259716b8..47ba05d5ae76 100644
--- a/arch/powerpc/perf/bhrb.S
+++ b/arch/powerpc/perf/bhrb.S
@@ -21,7 +21,7 @@
 _GLOBAL(read_bhrb)
 	cmpldi	r3,31
 	bgt	1f
-	ld	r4,bhrb_table@got(r2)
+	LOAD_REG_ADDR(r4, bhrb_table)
 	sldi	r3,r3,3
 	add	r3,r4,r3
 	mtctr	r3
diff --git a/arch/powerpc/platforms/pseries/hvCall.S b/arch/powerpc/platforms/pseries/hvCall.S
index ab9fc6506861..762eb15d3bd4 100644
--- a/arch/powerpc/platforms/pseries/hvCall.S
+++ b/arch/powerpc/platforms/pseries/hvCall.S
@@ -16,7 +16,7 @@
 #ifdef CONFIG_TRACEPOINTS
 
 #ifndef CONFIG_JUMP_LABEL
-	.section	".toc","aw"
+	.data
 
 	.globl hcall_tracepoint_refcount
 hcall_tracepoint_refcount:
@@ -88,7 +88,7 @@ hcall_tracepoint_refcount:
 BEGIN_FTR_SECTION;						\
 	b	1f;						\
 END_FTR_SECTION(0, 1);						\
-	ld	r12,hcall_tracepoint_refcount@toc(r2);		\
+	LOAD_REG_ADDR(r12, hcall_tracepoint_refcount) ;		\
 	std	r12,32(r1);					\
 	cmpdi	r12,0;						\
 	bne-	LABEL;						\
diff --git a/arch/powerpc/xmon/spr_access.S b/arch/powerpc/xmon/spr_access.S
index 720a52afdd58..c308ddf268fb 100644
--- a/arch/powerpc/xmon/spr_access.S
+++ b/arch/powerpc/xmon/spr_access.S
@@ -4,12 +4,12 @@
 
 /* unsigned long xmon_mfspr(sprn, default_value) */
 _GLOBAL(xmon_mfspr)
-	PPC_LL	r5, .Lmfspr_table@got(r2)
+	LOAD_REG_ADDR(r5, .Lmfspr_table)
 	b	xmon_mxspr
 
 /* void xmon_mtspr(sprn, new_value) */
 _GLOBAL(xmon_mtspr)
-	PPC_LL	r5, .Lmtspr_table@got(r2)
+	LOAD_REG_ADDR(r5, .Lmtspr_table)
 	b	xmon_mxspr
 
 /*
-- 
2.37.2

[PATCH v2 3/5] powerpc/64: switch asm helpers from GOT to TOC relative addressing

[Nicholas Piggin]

There is no need to use GOT addressing within the kernel.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
 arch/powerpc/boot/ppc_asm.h        | 3 ++-
 arch/powerpc/include/asm/ppc_asm.h | 3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/boot/ppc_asm.h b/arch/powerpc/boot/ppc_asm.h
index 2824a3e32aab..a66cfd76fa4d 100644
--- a/arch/powerpc/boot/ppc_asm.h
+++ b/arch/powerpc/boot/ppc_asm.h
@@ -86,7 +86,8 @@
 
 #ifdef CONFIG_PPC64_BOOT_WRAPPER
 #define LOAD_REG_ADDR(reg,name)			\
-	ld	reg,name@got(r2)
+	addis	reg,r2,name@toc@ha;		\
+	addi	reg,reg,name@toc@l
 #else
 #define LOAD_REG_ADDR(reg,name)			\
 	lis	reg,name@ha;			\
diff --git a/arch/powerpc/include/asm/ppc_asm.h b/arch/powerpc/include/asm/ppc_asm.h
index 83c02f5a7f2a..da570b197e82 100644
--- a/arch/powerpc/include/asm/ppc_asm.h
+++ b/arch/powerpc/include/asm/ppc_asm.h
@@ -315,7 +315,8 @@ GLUE(.,name):
 	rldimi	reg, tmp, 32, 0
 
 #define LOAD_REG_ADDR(reg,name)			\
-	ld	reg,name@got(r2)
+	addis	reg,r2,name@toc@ha;		\
+	addi	reg,reg,name@toc@l
 
 #define LOAD_REG_ADDRBASE(reg,name)	LOAD_REG_ADDR(reg,name)
 #define ADDROFF(name)			0
-- 
2.37.2

[PATCH v2 4/5] powerpc/64: provide a helper macro to load r2 with the kernel TOC

[Nicholas Piggin]

A later change stops the kernel using r2 and loads it with a poison
value.  Provide a PACATOC loading abstraction which can hide this
detail.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
 arch/powerpc/include/asm/ppc_asm.h             |  6 ++++++
 arch/powerpc/kernel/exceptions-64e.S           | 12 ++++++------
 arch/powerpc/kernel/exceptions-64s.S           |  6 +++---
 arch/powerpc/kernel/head_64.S                  |  4 ++--
 arch/powerpc/kernel/interrupt_64.S             | 12 ++++++------
 arch/powerpc/kernel/optprobes_head.S           |  2 +-
 arch/powerpc/kernel/trace/ftrace_low.S         |  2 +-
 arch/powerpc/kernel/trace/ftrace_mprofile.S    |  3 +--
 arch/powerpc/kvm/book3s_64_entry.S             |  2 +-
 arch/powerpc/kvm/book3s_hv_rmhandlers.S        |  4 ++--
 arch/powerpc/kvm/tm.S                          |  2 +-
 arch/powerpc/mm/nohash/tlb_low_64e.S           |  2 +-
 arch/powerpc/platforms/powernv/opal-wrappers.S |  2 +-
 13 files changed, 32 insertions(+), 27 deletions(-)

diff --git a/arch/powerpc/include/asm/ppc_asm.h b/arch/powerpc/include/asm/ppc_asm.h
index da570b197e82..17a87ecd82c9 100644
--- a/arch/powerpc/include/asm/ppc_asm.h
+++ b/arch/powerpc/include/asm/ppc_asm.h
@@ -305,6 +305,12 @@ GLUE(.,name):
 
 #ifdef __powerpc64__
 
+#define __LOAD_PACA_TOC(reg)			\
+	ld	reg,PACATOC(r13)
+
+#define LOAD_PACA_TOC()				\
+	__LOAD_PACA_TOC(r2)
+
 #define LOAD_REG_IMMEDIATE(reg, expr) __LOAD_REG_IMMEDIATE reg, expr
 
 #define LOAD_REG_IMMEDIATE_SYM(reg, tmp, expr)	\
diff --git a/arch/powerpc/kernel/exceptions-64e.S b/arch/powerpc/kernel/exceptions-64e.S
index 08a3139079b7..c20b39d011a9 100644
--- a/arch/powerpc/kernel/exceptions-64e.S
+++ b/arch/powerpc/kernel/exceptions-64e.S
@@ -382,7 +382,7 @@ exc_##n##_common:							    \
 	ld	r4,excf+EX_R11(r13);	/* get back r11 */		    \
 	mfspr	r5,scratch;		/* get back r13 */		    \
 	std	r12,GPR12(r1);		/* save r12 in stackframe */	    \
-	ld	r2,PACATOC(r13);	/* get kernel TOC into r2 */	    \
+	LOAD_PACA_TOC();		/* get kernel TOC into r2 */	    \
 	mflr	r6;			/* save LR in stackframe */	    \
 	mfctr	r7;			/* save CTR in stackframe */	    \
 	mfspr	r8,SPRN_XER;		/* save XER in stackframe */	    \
@@ -698,7 +698,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
 	beq+	1f
 
 #ifdef CONFIG_RELOCATABLE
-	ld	r15,PACATOC(r13)
+	__LOAD_PACA_TOC(r15)
 	ld	r14,interrupt_base_book3e@got(r15)
 	ld	r15,__end_interrupts@got(r15)
 	cmpld	cr0,r10,r14
@@ -769,7 +769,7 @@ kernel_dbg_exc:
 	beq+	1f
 
 #ifdef CONFIG_RELOCATABLE
-	ld	r15,PACATOC(r13)
+	__LOAD_PACA_TOC(r15)
 	ld	r14,interrupt_base_book3e@got(r15)
 	ld	r15,__end_interrupts@got(r15)
 	cmpld	cr0,r10,r14
@@ -894,7 +894,7 @@ kernel_dbg_exc:
 
 .macro SEARCH_RESTART_TABLE
 #ifdef CONFIG_RELOCATABLE
-	ld	r11,PACATOC(r13)
+	__LOAD_PACA_TOC(r11)
 	ld	r14,__start___restart_table@got(r11)
 	ld	r15,__stop___restart_table@got(r11)
 #else
@@ -1073,7 +1073,7 @@ bad_stack_book3e:
 	std	r11,0(r1)
 	li	r12,0
 	std	r12,0(r11)
-	ld	r2,PACATOC(r13)
+	LOAD_PACA_TOC()
 1:	addi	r3,r1,STACK_FRAME_OVERHEAD
 	bl	kernel_bad_stack
 	b	1b
@@ -1314,7 +1314,7 @@ a2_tlbinit_after_linear_map:
 
 	/* Now we branch the new virtual address mapped by this entry */
 #ifdef CONFIG_RELOCATABLE
-	ld	r5,PACATOC(r13)
+	__LOAD_PACA_TOC(r5)
 	ld	r3,1f@got(r5)
 #else
 	LOAD_REG_IMMEDIATE_SYM(r3, r5, 1f)
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 9d375ea58add..8e6fd8c6618e 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -580,7 +580,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
 	std	r2,GPR2(r1)		/* save r2 in stackframe	*/
 	SAVE_GPRS(3, 8, r1)		/* save r3 - r8 in stackframe   */
 	mflr	r9			/* Get LR, later save to stack	*/
-	ld	r2,PACATOC(r13)		/* get kernel TOC into r2	*/
+	LOAD_PACA_TOC()			/* get kernel TOC into r2	*/
 	std	r9,_LINK(r1)
 	lbz	r10,PACAIRQSOFTMASK(r13)
 	mfspr	r11,SPRN_XER		/* save XER in stackframe	*/
@@ -610,7 +610,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
 .macro SEARCH_RESTART_TABLE
 #ifdef CONFIG_RELOCATABLE
 	mr	r12,r2
-	ld	r2,PACATOC(r13)
+	LOAD_PACA_TOC()
 	LOAD_REG_ADDR(r9, __start___restart_table)
 	LOAD_REG_ADDR(r10, __stop___restart_table)
 	mr	r2,r12
@@ -640,7 +640,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
 .macro SEARCH_SOFT_MASK_TABLE
 #ifdef CONFIG_RELOCATABLE
 	mr	r12,r2
-	ld	r2,PACATOC(r13)
+	LOAD_PACA_TOC()
 	LOAD_REG_ADDR(r9, __start___soft_mask_table)
 	LOAD_REG_ADDR(r10, __stop___soft_mask_table)
 	mr	r2,r12
diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index cac3e1b58360..80106aaf0b7a 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -841,7 +841,7 @@ __secondary_start:
  * before going into C code.
  */
 start_secondary_prolog:
-	ld	r2,PACATOC(r13)
+	LOAD_PACA_TOC()
 	li	r3,0
 	std	r3,0(r1)		/* Zero the stack frame pointer	*/
 	bl	start_secondary
@@ -981,7 +981,7 @@ start_here_common:
 	std	r1,PACAKSAVE(r13)
 
 	/* Load the TOC (virtual address) */
-	ld	r2,PACATOC(r13)
+	LOAD_PACA_TOC()
 
 	/* Mark interrupts soft and hard disabled (they might be enabled
 	 * in the PACA when doing hotplug)
diff --git a/arch/powerpc/kernel/interrupt_64.S b/arch/powerpc/kernel/interrupt_64.S
index fbfb72a62da9..3205be189e34 100644
--- a/arch/powerpc/kernel/interrupt_64.S
+++ b/arch/powerpc/kernel/interrupt_64.S
@@ -57,7 +57,7 @@ _ASM_NOKPROBE_SYMBOL(system_call_vectored_\name)
 	std	r0,GPR0(r1)
 	std	r10,GPR1(r1)
 	std	r2,GPR2(r1)
-	ld	r2,PACATOC(r13)
+	LOAD_PACA_TOC()
 	mfcr	r12
 	li	r11,0
 	/* Can we avoid saving r3-r8 in common case? */
@@ -185,7 +185,7 @@ syscall_vectored_\name\()_restart:
 _ASM_NOKPROBE_SYMBOL(syscall_vectored_\name\()_restart)
 	GET_PACA(r13)
 	ld	r1,PACA_EXIT_SAVE_R1(r13)
-	ld	r2,PACATOC(r13)
+	LOAD_PACA_TOC()
 	ld	r3,RESULT(r1)
 	addi	r4,r1,STACK_FRAME_OVERHEAD
 	li	r11,IRQS_ALL_DISABLED
@@ -235,7 +235,7 @@ START_BTB_FLUSH_SECTION
 	BTB_FLUSH(r10)
 END_BTB_FLUSH_SECTION
 #endif
-	ld	r2,PACATOC(r13)
+	LOAD_PACA_TOC()
 	mfcr	r12
 	li	r11,0
 	/* Can we avoid saving r3-r8 in common case? */
@@ -378,7 +378,7 @@ syscall_restart:
 _ASM_NOKPROBE_SYMBOL(syscall_restart)
 	GET_PACA(r13)
 	ld	r1,PACA_EXIT_SAVE_R1(r13)
-	ld	r2,PACATOC(r13)
+	LOAD_PACA_TOC()
 	ld	r3,RESULT(r1)
 	addi	r4,r1,STACK_FRAME_OVERHEAD
 	li	r11,IRQS_ALL_DISABLED
@@ -525,7 +525,7 @@ interrupt_return_\srr\()_user_restart:
 _ASM_NOKPROBE_SYMBOL(interrupt_return_\srr\()_user_restart)
 	GET_PACA(r13)
 	ld	r1,PACA_EXIT_SAVE_R1(r13)
-	ld	r2,PACATOC(r13)
+	LOAD_PACA_TOC()
 	addi	r3,r1,STACK_FRAME_OVERHEAD
 	li	r11,IRQS_ALL_DISABLED
 	stb	r11,PACAIRQSOFTMASK(r13)
@@ -658,7 +658,7 @@ interrupt_return_\srr\()_kernel_restart:
 _ASM_NOKPROBE_SYMBOL(interrupt_return_\srr\()_kernel_restart)
 	GET_PACA(r13)
 	ld	r1,PACA_EXIT_SAVE_R1(r13)
-	ld	r2,PACATOC(r13)
+	LOAD_PACA_TOC()
 	addi	r3,r1,STACK_FRAME_OVERHEAD
 	li	r11,IRQS_ALL_DISABLED
 	stb	r11,PACAIRQSOFTMASK(r13)
diff --git a/arch/powerpc/kernel/optprobes_head.S b/arch/powerpc/kernel/optprobes_head.S
index 5c7f0b4b784b..cd4e7bc32609 100644
--- a/arch/powerpc/kernel/optprobes_head.S
+++ b/arch/powerpc/kernel/optprobes_head.S
@@ -73,7 +73,7 @@ optprobe_template_entry:
 	 * further below.
 	 */
 #ifdef CONFIG_PPC64
-	ld	r2,PACATOC(r13)
+	LOAD_PACA_TOC()
 #endif
 
 	.global optprobe_template_op_address
diff --git a/arch/powerpc/kernel/trace/ftrace_low.S b/arch/powerpc/kernel/trace/ftrace_low.S
index 0bddf1fa6636..294d1e05958a 100644
--- a/arch/powerpc/kernel/trace/ftrace_low.S
+++ b/arch/powerpc/kernel/trace/ftrace_low.S
@@ -48,7 +48,7 @@ _GLOBAL(return_to_handler)
 	 * We might be called from a module.
 	 * Switch to our TOC to run inside the core kernel.
 	 */
-	ld	r2, PACATOC(r13)
+	LOAD_PACA_TOC()
 #else
 	stwu	r1, -16(r1)
 	stw	r3, 8(r1)
diff --git a/arch/powerpc/kernel/trace/ftrace_mprofile.S b/arch/powerpc/kernel/trace/ftrace_mprofile.S
index 33fcfb2eaded..d031093bc436 100644
--- a/arch/powerpc/kernel/trace/ftrace_mprofile.S
+++ b/arch/powerpc/kernel/trace/ftrace_mprofile.S
@@ -83,8 +83,7 @@
 #ifdef CONFIG_PPC64
 	/* Save callee's TOC in the ABI compliant location */
 	std	r2, STK_GOT(r1)
-	ld	r2,PACATOC(r13)	/* get kernel TOC in r2 */
-
+	LOAD_PACA_TOC()		/* get kernel TOC in r2 */
 	LOAD_REG_ADDR(r3, function_trace_op)
 	ld	r5,0(r3)
 #else
diff --git a/arch/powerpc/kvm/book3s_64_entry.S b/arch/powerpc/kvm/book3s_64_entry.S
index e43704547a1e..6c2b1d17cb63 100644
--- a/arch/powerpc/kvm/book3s_64_entry.S
+++ b/arch/powerpc/kvm/book3s_64_entry.S
@@ -315,7 +315,7 @@ kvmppc_p9_exit_interrupt:
 	reg = reg + 1
 	.endr
 
-	ld	r2,PACATOC(r13)
+	LOAD_PACA_TOC()
 
 	mflr	r4
 	std	r4,VCPU_LR(r3)
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 7ded202bf995..c984021e62c8 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -1024,7 +1024,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_HAS_PPR)
 
 	/* Restore R1/R2 so we can handle faults */
 	ld	r1, HSTATE_HOST_R1(r13)
-	ld	r2, PACATOC(r13)
+	LOAD_PACA_TOC()
 
 	mfspr	r10, SPRN_SRR0
 	mfspr	r11, SPRN_SRR1
@@ -2727,7 +2727,7 @@ kvmppc_bad_host_intr:
 	std	r4, _CTR(r1)
 	std	r5, _XER(r1)
 	std	r6, SOFTE(r1)
-	ld	r2, PACATOC(r13)
+	LOAD_PACA_TOC()
 	LOAD_REG_IMMEDIATE(3, 0x7265677368657265)
 	std	r3, STACK_FRAME_OVERHEAD-16(r1)
 
diff --git a/arch/powerpc/kvm/tm.S b/arch/powerpc/kvm/tm.S
index 3bf17c854be4..2158f61e317f 100644
--- a/arch/powerpc/kvm/tm.S
+++ b/arch/powerpc/kvm/tm.S
@@ -110,7 +110,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_P9_TM_HV_ASSIST)
 	mtmsrd	r2, 1
 
 	/* Reload TOC pointer. */
-	ld	r2, PACATOC(r13)
+	LOAD_PACA_TOC()
 
 	/* Save all but r0-r2, r9 & r13 */
 	reg = 3
diff --git a/arch/powerpc/mm/nohash/tlb_low_64e.S b/arch/powerpc/mm/nohash/tlb_low_64e.S
index 68ffbfdba894..0e4d9c817382 100644
--- a/arch/powerpc/mm/nohash/tlb_low_64e.S
+++ b/arch/powerpc/mm/nohash/tlb_low_64e.S
@@ -1124,7 +1124,7 @@ tlb_load_linear:
 	 * we only use 1G pages for now. That might have to be changed in a
 	 * final implementation, especially when dealing with hypervisors
 	 */
-	ld	r11,PACATOC(r13)
+	__LOAD_PACA_TOC(r11)
 	ld	r11,linear_map_top@got(r11)
 	ld	r10,0(r11)
 	tovirt(10,10)
diff --git a/arch/powerpc/platforms/powernv/opal-wrappers.S b/arch/powerpc/platforms/powernv/opal-wrappers.S
index e5acc33b3b20..0ed95f753416 100644
--- a/arch/powerpc/platforms/powernv/opal-wrappers.S
+++ b/arch/powerpc/platforms/powernv/opal-wrappers.S
@@ -57,7 +57,7 @@ opal_return:
 	.long 0xa64b7b7d /* mthsrr1 r11				*/
 	.long 0x2402004c /* hrfid				*/
 #endif
-	ld	r2,PACATOC(r13)
+	LOAD_PACA_TOC()
 	ld	r0,PPC_LR_STKOFF(r1)
 	mtlr	r0
 	blr
-- 
2.37.2

[PATCH v2 5/5] powerpc/64e: provide an addressing macro for use with TOC in alternate register

[Nicholas Piggin]

The interrupt entry code carefully saves a minimal number of registers,
so in some places the TOC is required, it is loaded into a different
register, so provide a macro that can supply an alternate TOC register.

This continues to use got addressing because TOC-relative results in
"got/toc optimization is not supported" messages by the linker. Having
r2 be one of the saved registers and using that for TOC addressing may
be the best way to avoid that and switch this to TOC addressing.

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
 arch/powerpc/include/asm/ppc_asm.h   | 11 +++++++++++
 arch/powerpc/kernel/exceptions-64e.S | 14 +++++++-------
 arch/powerpc/mm/nohash/tlb_low_64e.S |  2 +-
 3 files changed, 19 insertions(+), 8 deletions(-)

diff --git a/arch/powerpc/include/asm/ppc_asm.h b/arch/powerpc/include/asm/ppc_asm.h
index 17a87ecd82c9..622aaaec2be9 100644
--- a/arch/powerpc/include/asm/ppc_asm.h
+++ b/arch/powerpc/include/asm/ppc_asm.h
@@ -324,6 +324,17 @@ GLUE(.,name):
 	addis	reg,r2,name@toc@ha;		\
 	addi	reg,reg,name@toc@l
 
+#ifdef CONFIG_PPC_BOOK3E_64
+/*
+ * This is used in register-constrained interrupt handlers. Not to be used
+ * by BOOK3S. ld complains with "got/toc optimization is not supported" if r2
+ * is not used for the TOC offset, so use @got(tocreg). If the interrupt
+ * handlers saved r2 instead, LOAD_REG_ADDR could be used.
+ */
+#define LOAD_REG_ADDR_ALTTOC(reg,tocreg,name)	\
+	ld	reg,name@got(tocreg)
+#endif
+
 #define LOAD_REG_ADDRBASE(reg,name)	LOAD_REG_ADDR(reg,name)
 #define ADDROFF(name)			0
 
diff --git a/arch/powerpc/kernel/exceptions-64e.S b/arch/powerpc/kernel/exceptions-64e.S
index c20b39d011a9..8ac9dfd38874 100644
--- a/arch/powerpc/kernel/exceptions-64e.S
+++ b/arch/powerpc/kernel/exceptions-64e.S
@@ -699,8 +699,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
 
 #ifdef CONFIG_RELOCATABLE
 	__LOAD_PACA_TOC(r15)
-	ld	r14,interrupt_base_book3e@got(r15)
-	ld	r15,__end_interrupts@got(r15)
+	LOAD_REG_ADDR_ALTTOC(r14, r15, interrupt_base_book3e)
+	LOAD_REG_ADDR_ALTTOC(r15, r15, __end_interrupts)
 	cmpld	cr0,r10,r14
 	cmpld	cr1,r10,r15
 #else
@@ -770,8 +770,8 @@ kernel_dbg_exc:
 
 #ifdef CONFIG_RELOCATABLE
 	__LOAD_PACA_TOC(r15)
-	ld	r14,interrupt_base_book3e@got(r15)
-	ld	r15,__end_interrupts@got(r15)
+	LOAD_REG_ADDR_ALTTOC(r14, r15, interrupt_base_book3e)
+	LOAD_REG_ADDR_ALTTOC(r15, r15, __end_interrupts)
 	cmpld	cr0,r10,r14
 	cmpld	cr1,r10,r15
 #else
@@ -895,8 +895,8 @@ kernel_dbg_exc:
 .macro SEARCH_RESTART_TABLE
 #ifdef CONFIG_RELOCATABLE
 	__LOAD_PACA_TOC(r11)
-	ld	r14,__start___restart_table@got(r11)
-	ld	r15,__stop___restart_table@got(r11)
+	LOAD_REG_ADDR_ALTTOC(r14, r11, __start___restart_table)
+	LOAD_REG_ADDR_ALTTOC(r15, r11, __stop___restart_table)
 #else
 	LOAD_REG_IMMEDIATE_SYM(r14, r11, __start___restart_table)
 	LOAD_REG_IMMEDIATE_SYM(r15, r11, __stop___restart_table)
@@ -1315,7 +1315,7 @@ a2_tlbinit_after_linear_map:
 	/* Now we branch the new virtual address mapped by this entry */
 #ifdef CONFIG_RELOCATABLE
 	__LOAD_PACA_TOC(r5)
-	ld	r3,1f@got(r5)
+	LOAD_REG_ADDR_ALTTOC(r3, r5, 1f)
 #else
 	LOAD_REG_IMMEDIATE_SYM(r3, r5, 1f)
 #endif
diff --git a/arch/powerpc/mm/nohash/tlb_low_64e.S b/arch/powerpc/mm/nohash/tlb_low_64e.S
index 0e4d9c817382..2f3095716721 100644
--- a/arch/powerpc/mm/nohash/tlb_low_64e.S
+++ b/arch/powerpc/mm/nohash/tlb_low_64e.S
@@ -1125,7 +1125,7 @@ tlb_load_linear:
 	 * final implementation, especially when dealing with hypervisors
 	 */
 	__LOAD_PACA_TOC(r11)
-	ld	r11,linear_map_top@got(r11)
+	LOAD_REG_ADDR_ALTTOC(r11, r11, linear_map_top)
 	ld	r10,0(r11)
 	tovirt(10,10)
 	cmpld	cr0,r16,r10
-- 
2.37.2

Re: [PATCH v2 1/5] powerpc/64: use 32-bit immediate for STACK_FRAME_REGS_MARKER

[Christophe Leroy]

Le 26/09/2022 à 05:40, Nicholas Piggin a écrit :
> Using a 32-bit constant for this marker allows it to be loaded with
> two ALU instructions, like 32-bit. This avoids a TOC entry and a
> TOC load that depends on the r2 value that has just been loaded from
> the PACA.
> 
> This changes the value for 32-bit as well, so both have the same
> value in the low 4 bytes and 64-bit has 0xffffffff in the top bytes.

The above is wrong now, isn't it ?

> 
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
>   arch/powerpc/include/asm/ptrace.h    | 4 ++--
>   arch/powerpc/kernel/entry_32.S       | 6 +++---
>   arch/powerpc/kernel/exceptions-64e.S | 8 +-------
>   arch/powerpc/kernel/exceptions-64s.S | 2 +-
>   arch/powerpc/kernel/head_64.S        | 7 -------
>   arch/powerpc/kernel/interrupt_64.S   | 6 +++---
>   6 files changed, 10 insertions(+), 23 deletions(-)
> 
> diff --git a/arch/powerpc/include/asm/ptrace.h b/arch/powerpc/include/asm/ptrace.h
> index a03403695cd4..5b496e589d54 100644
> --- a/arch/powerpc/include/asm/ptrace.h
> +++ b/arch/powerpc/include/asm/ptrace.h
> @@ -99,6 +99,8 @@ struct pt_regs
>   
>   #define STACK_FRAME_WITH_PT_REGS (STACK_FRAME_OVERHEAD + sizeof(struct pt_regs))
>   
> +#define STACK_FRAME_REGS_MARKER	ASM_CONST(0x72656773)
> +
>   #ifdef __powerpc64__
>   
>   /*
> @@ -115,7 +117,6 @@ struct pt_regs
>   
>   #define STACK_FRAME_OVERHEAD	112	/* size of minimum stack frame */
>   #define STACK_FRAME_LR_SAVE	2	/* Location of LR in stack frame */
> -#define STACK_FRAME_REGS_MARKER	ASM_CONST(0x7265677368657265)
>   #define STACK_INT_FRAME_SIZE	(sizeof(struct pt_regs) + \
>   				 STACK_FRAME_OVERHEAD + KERNEL_REDZONE_SIZE)
>   #define STACK_FRAME_MARKER	12
> @@ -136,7 +137,6 @@ struct pt_regs
>   #define KERNEL_REDZONE_SIZE	0
>   #define STACK_FRAME_OVERHEAD	16	/* size of minimum stack frame */
>   #define STACK_FRAME_LR_SAVE	1	/* Location of LR in stack frame */
> -#define STACK_FRAME_REGS_MARKER	ASM_CONST(0x72656773)
>   #define STACK_INT_FRAME_SIZE	(sizeof(struct pt_regs) + STACK_FRAME_OVERHEAD)
>   #define STACK_FRAME_MARKER	2
>   #define STACK_FRAME_MIN_SIZE	STACK_FRAME_OVERHEAD
> diff --git a/arch/powerpc/kernel/entry_32.S b/arch/powerpc/kernel/entry_32.S
> index 1d599df6f169..c2516f982cfa 100644
> --- a/arch/powerpc/kernel/entry_32.S
> +++ b/arch/powerpc/kernel/entry_32.S
> @@ -265,7 +265,7 @@ fast_exception_return:
>   	mtcr	r10
>   	lwz	r10,_LINK(r11)
>   	mtlr	r10
> -	/* Clear the exception_marker on the stack to avoid confusing stacktrace */
> +	/* Clear the exception marker on the stack to avoid confusing stacktrace */
>   	li	r10, 0
>   	stw	r10, 8(r11)
>   	REST_GPR(10, r11)
> @@ -322,7 +322,7 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS)
>   	li	r0,0
>   
>   	/*
> -	 * Leaving a stale exception_marker on the stack can confuse
> +	 * Leaving a stale exception marker on the stack can confuse
>   	 * the reliable stack unwinder later on. Clear it.
>   	 */
>   	stw	r0,8(r1)
> @@ -374,7 +374,7 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS)
>   	mtspr	SPRN_XER,r5
>   
>   	/*
> -	 * Leaving a stale exception_marker on the stack can confuse
> +	 * Leaving a stale exception marker on the stack can confuse
>   	 * the reliable stack unwinder later on. Clear it.
>   	 */
>   	stw	r0,8(r1)
> diff --git a/arch/powerpc/kernel/exceptions-64e.S b/arch/powerpc/kernel/exceptions-64e.S
> index 67dc4e3179a0..08a3139079b7 100644
> --- a/arch/powerpc/kernel/exceptions-64e.S
> +++ b/arch/powerpc/kernel/exceptions-64e.S
> @@ -389,7 +389,7 @@ exc_##n##_common:							    \
>   	ld	r9,excf+EX_R1(r13);	/* load orig r1 back from PACA */   \
>   	lwz	r10,excf+EX_CR(r13);	/* load orig CR back from PACA	*/  \
>   	lbz	r11,PACAIRQSOFTMASK(r13); /* get current IRQ softe */	    \
> -	ld	r12,exception_marker@toc(r2);				    \
> +	LOAD_REG_IMMEDIATE(r12, STACK_FRAME_REGS_MARKER);		    \
>   	li	r0,0;							    \
>   	std	r3,GPR10(r1);		/* save r10 to stackframe */	    \
>   	std	r4,GPR11(r1);		/* save r11 to stackframe */	    \
> @@ -470,12 +470,6 @@ exc_##n##_bad_stack:							    \
>   	bl	hdlr;							\
>   	b	interrupt_return
>   
> -/* This value is used to mark exception frames on the stack. */
> -	.section	".toc","aw"
> -exception_marker:
> -	.tc	ID_EXC_MARKER[TC],STACK_FRAME_REGS_MARKER
> -
> -
>   /*
>    * And here we have the exception vectors !
>    */
> diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
> index 3d0dc133a9ae..9d375ea58add 100644
> --- a/arch/powerpc/kernel/exceptions-64s.S
> +++ b/arch/powerpc/kernel/exceptions-64s.S
> @@ -589,7 +589,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_CFAR)
>   	li	r9,IVEC
>   	std	r9,_TRAP(r1)		/* set trap number		*/
>   	li	r10,0
> -	ld	r11,exception_marker@toc(r2)
> +	LOAD_REG_IMMEDIATE(r11, STACK_FRAME_REGS_MARKER)
>   	std	r10,RESULT(r1)		/* clear regs->result		*/
>   	std	r11,STACK_FRAME_OVERHEAD-16(r1) /* mark the frame	*/
>   .endm
> diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
> index cf2c08902c05..cac3e1b58360 100644
> --- a/arch/powerpc/kernel/head_64.S
> +++ b/arch/powerpc/kernel/head_64.S
> @@ -192,13 +192,6 @@ __secondary_hold:
>   #endif
>   CLOSE_FIXED_SECTION(first_256B)
>   
> -/* This value is used to mark exception frames on the stack. */
> -	.section ".toc","aw"
> -/* This value is used to mark exception frames on the stack. */
> -exception_marker:
> -	.tc	ID_EXC_MARKER[TC],STACK_FRAME_REGS_MARKER
> -	.previous
> -
>   /*
>    * On server, we include the exception vectors code here as it
>    * relies on absolute addressing which is only possible within
> diff --git a/arch/powerpc/kernel/interrupt_64.S b/arch/powerpc/kernel/interrupt_64.S
> index ce25b28cf418..ee2d2d410c5a 100644
> --- a/arch/powerpc/kernel/interrupt_64.S
> +++ b/arch/powerpc/kernel/interrupt_64.S
> @@ -92,7 +92,7 @@ _ASM_NOKPROBE_SYMBOL(system_call_vectored_\name)
>   	std	r11,_TRAP(r1)
>   	std	r12,_CCR(r1)
>   	addi	r10,r1,STACK_FRAME_OVERHEAD
> -	ld	r11,exception_marker@toc(r2)
> +	LOAD_REG_IMMEDIATE(r11, STACK_FRAME_REGS_MARKER)
>   	std	r11,-16(r10)		/* "regshere" marker */
>   
>   BEGIN_FTR_SECTION
> @@ -276,7 +276,7 @@ END_BTB_FLUSH_SECTION
>   	std	r11,_TRAP(r1)
>   	std	r12,_CCR(r1)
>   	addi	r10,r1,STACK_FRAME_OVERHEAD
> -	ld	r11,exception_marker@toc(r2)
> +	LOAD_REG_IMMEDIATE(r11, STACK_FRAME_REGS_MARKER)
>   	std	r11,-16(r10)		/* "regshere" marker */
>   
>   #ifdef CONFIG_PPC_BOOK3S
> @@ -619,7 +619,7 @@ ALT_FTR_SECTION_END_IFCLR(CPU_FTR_STCX_CHECKS_ADDRESS)
>   	mtspr	SPRN_XER,r5
>   
>   	/*
> -	 * Leaving a stale exception_marker on the stack can confuse
> +	 * Leaving a stale STACK_FRAME_REGS_MARKER on the stack can confuse
>   	 * the reliable stack unwinder later on. Clear it.
>   	 */
>   	std	r0,STACK_FRAME_OVERHEAD-16(r1)

Re: [PATCH v2 1/5] powerpc/64: use 32-bit immediate for STACK_FRAME_REGS_MARKER

[Michael Ellerman]

Christophe Leroy <christophe.leroy@csgroup.eu> writes:
> Le 26/09/2022 à 05:40, Nicholas Piggin a écrit :
>> Using a 32-bit constant for this marker allows it to be loaded with
>> two ALU instructions, like 32-bit. This avoids a TOC entry and a
>> TOC load that depends on the r2 value that has just been loaded from
>> the PACA.
>> 
>> This changes the value for 32-bit as well, so both have the same
>> value in the low 4 bytes and 64-bit has 0xffffffff in the top bytes.
>
> The above is wrong now, isn't it ?

Yeah, I fixed it when applying.

cheers

Re: [PATCH v2 0/5] powerpc/64: avoid GOT addressing, don't put data in TOC

[Michael Ellerman]

On Mon, 26 Sep 2022 13:40:52 +1000, Nicholas Piggin wrote:
> This is a cleaned up set of the initial prep patches from the pcrel
> series, dealing with regularising addressing variables from asm
> and using helper macros more consistently.
> 
> Changes since v1:
> - Use "REGS" stack frame marker for 64-bit, leaving 32-bit unchanged.
> - Fix PPC64 32-bit boot wrapper compile.
> 
> [...]

Applied to powerpc/next.

[1/5] powerpc/64: use 32-bit immediate for STACK_FRAME_REGS_MARKER
      https://git.kernel.org/powerpc/c/17773afdcd1589c5925a984f512330410cb2ba4f
[2/5] powerpc/64: asm use consistent global variable declaration and access
      https://git.kernel.org/powerpc/c/dab3b8f4fd09c22e8dbb2d9608194c7d52252f33
[3/5] powerpc/64: switch asm helpers from GOT to TOC relative addressing
      https://git.kernel.org/powerpc/c/754f611774e4b9357a944f5b703dd291c85161cf
[4/5] powerpc/64: provide a helper macro to load r2 with the kernel TOC
      https://git.kernel.org/powerpc/c/8e93fb33c84f68db20c0bc2821334a4c54c3e251
[5/5] powerpc/64e: provide an addressing macro for use with TOC in alternate register
      https://git.kernel.org/powerpc/c/3569d84bb26f6f07d426446da3d2c836180f1565

cheers