[PATCH -tip v3 1/3] x86/percpu: Correct PER_CPU_VAR usage to include symbol and its addend

[PATCH -tip v3 1/3] x86/percpu: Correct PER_CPU_VAR usage to include symbol and its addend

[Uros Bizjak]

PER_CPU_VAR macro should be applied to a symbol and its addend.
Inconsistent usage is currently harmless, but needs to be corrected
before %rip-relative addressing is introduced to PER_CPU_VAR macro.

No functional changes intended.

Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
---
 arch/x86/entry/calling.h  | 2 +-
 arch/x86/entry/entry_32.S | 2 +-
 arch/x86/entry/entry_64.S | 2 +-
 arch/x86/kernel/head_64.S | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/arch/x86/entry/calling.h b/arch/x86/entry/calling.h
index f6907627172b..47368ab0bda0 100644
--- a/arch/x86/entry/calling.h
+++ b/arch/x86/entry/calling.h
@@ -173,7 +173,7 @@ For 32-bit we have the following conventions - kernel is built with
 .endm
 
 #define THIS_CPU_user_pcid_flush_mask   \
-	PER_CPU_VAR(cpu_tlbstate) + TLB_STATE_user_pcid_flush_mask
+	PER_CPU_VAR(cpu_tlbstate + TLB_STATE_user_pcid_flush_mask)
 
 .macro SWITCH_TO_USER_CR3_NOSTACK scratch_reg:req scratch_reg2:req
 	ALTERNATIVE "jmp .Lend_\@", "", X86_FEATURE_PTI
diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
index c73047bf9f4b..4e295798638b 100644
--- a/arch/x86/entry/entry_32.S
+++ b/arch/x86/entry/entry_32.S
@@ -305,7 +305,7 @@
 .macro CHECK_AND_APPLY_ESPFIX
 #ifdef CONFIG_X86_ESPFIX32
 #define GDT_ESPFIX_OFFSET (GDT_ENTRY_ESPFIX_SS * 8)
-#define GDT_ESPFIX_SS PER_CPU_VAR(gdt_page) + GDT_ESPFIX_OFFSET
+#define GDT_ESPFIX_SS PER_CPU_VAR(gdt_page + GDT_ESPFIX_OFFSET)
 
 	ALTERNATIVE	"jmp .Lend_\@", "", X86_BUG_ESPFIX
 
diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index de6469dffe3a..1a88ad8a7b48 100644
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -190,7 +190,7 @@ SYM_FUNC_START(__switch_to_asm)
 
 #ifdef CONFIG_STACKPROTECTOR
 	movq	TASK_stack_canary(%rsi), %rbx
-	movq	%rbx, PER_CPU_VAR(fixed_percpu_data) + FIXED_stack_canary
+	movq	%rbx, PER_CPU_VAR(fixed_percpu_data + FIXED_stack_canary)
 #endif
 
 	/*
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index 086a2c3aaaa0..3dcabbc49149 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -448,7 +448,7 @@ SYM_CODE_START(soft_restart_cpu)
 	UNWIND_HINT_END_OF_STACK
 
 	/* Find the idle task stack */
-	movq	PER_CPU_VAR(pcpu_hot) + X86_current_task, %rcx
+	movq	PER_CPU_VAR(pcpu_hot + X86_current_task), %rcx
 	movq	TASK_threadsp(%rcx), %rsp
 
 	jmp	.Ljump_to_C_code
-- 
2.41.0

[PATCH -tip v3 2/3] x86/percpu, xen: Correct PER_CPU_VAR usage to include symbol and its addend

[Uros Bizjak]

PER_CPU_VAR macro should be applied to a symbol and its addend.
Inconsistent usage is currently harmless, but needs to be corrected
before %rip-relative addressing is introduced to PER_CPU_VAR macro.

No functional changes intended.

Cc: Juergen Gross <jgross@suse.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
---
 arch/x86/xen/xen-asm.S | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/arch/x86/xen/xen-asm.S b/arch/x86/xen/xen-asm.S
index 9e5e68008785..448958ddbaf8 100644
--- a/arch/x86/xen/xen-asm.S
+++ b/arch/x86/xen/xen-asm.S
@@ -28,7 +28,7 @@
  * non-zero.
  */
 SYM_FUNC_START(xen_irq_disable_direct)
-	movb $1, PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_mask
+	movb $1, PER_CPU_VAR(xen_vcpu_info + XEN_vcpu_info_mask)
 	RET
 SYM_FUNC_END(xen_irq_disable_direct)
 
@@ -69,7 +69,7 @@ SYM_FUNC_END(check_events)
 SYM_FUNC_START(xen_irq_enable_direct)
 	FRAME_BEGIN
 	/* Unmask events */
-	movb $0, PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_mask
+	movb $0, PER_CPU_VAR(xen_vcpu_info + XEN_vcpu_info_mask)
 
 	/*
 	 * Preempt here doesn't matter because that will deal with any
@@ -78,7 +78,7 @@ SYM_FUNC_START(xen_irq_enable_direct)
 	 */
 
 	/* Test for pending */
-	testb $0xff, PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_pending
+	testb $0xff, PER_CPU_VAR(xen_vcpu_info + XEN_vcpu_info_pending)
 	jz 1f
 
 	call check_events
@@ -97,7 +97,7 @@ SYM_FUNC_END(xen_irq_enable_direct)
  * x86 use opposite senses (mask vs enable).
  */
 SYM_FUNC_START(xen_save_fl_direct)
-	testb $0xff, PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_mask
+	testb $0xff, PER_CPU_VAR(xen_vcpu_info + XEN_vcpu_info_mask)
 	setz %ah
 	addb %ah, %ah
 	RET
@@ -113,7 +113,7 @@ SYM_FUNC_END(xen_read_cr2);
 
 SYM_FUNC_START(xen_read_cr2_direct)
 	FRAME_BEGIN
-	_ASM_MOV PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_arch_cr2, %_ASM_AX
+	_ASM_MOV PER_CPU_VAR(xen_vcpu_info + XEN_vcpu_info_arch_cr2), %_ASM_AX
 	FRAME_END
 	RET
 SYM_FUNC_END(xen_read_cr2_direct);
-- 
2.41.0

[PATCH -tip v3 3/3] x86/percpu: Introduce %rip-relative addressing to PER_CPU_VAR

[Uros Bizjak]

Introduce x86_64 %rip-relative addressing to PER_CPU_VAR macro.
Instructions using %rip-relative address operand are one byte shorter
than their absolute address counterparts and are also compatible with
position independent executable (-fpie) build. The patch reduces
code size of a test kernel build by 150 bytes.

PER_CPU_VAR macro is intended to be applied to a symbol and should
not be used with register operands. Introduce new __percpu macro and
use it in cmpxchg{8,16}b_emu.S instead.

Also add a missing function comment to this_cpu_cmpxchg8b_emu.

No functional changes intended.

Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
--
v2: Introduce PER_CPU_ARG macro to conditionally enable
    segment registers in cmpxchg{8,16}b_emu.S for CONFIG_SMP.
v3: Introduce __percpu macro instead of PER_CPU_ARG (hpa).
---
 arch/x86/include/asm/percpu.h | 12 ++++++++----
 arch/x86/lib/cmpxchg16b_emu.S | 12 ++++++------
 arch/x86/lib/cmpxchg8b_emu.S  | 30 +++++++++++++++++++++---------
 3 files changed, 35 insertions(+), 19 deletions(-)

diff --git a/arch/x86/include/asm/percpu.h b/arch/x86/include/asm/percpu.h
index 54746903b8c3..02f1780f02f5 100644
--- a/arch/x86/include/asm/percpu.h
+++ b/arch/x86/include/asm/percpu.h
@@ -4,17 +4,21 @@
 
 #ifdef CONFIG_X86_64
 #define __percpu_seg		gs
+#define __percpu_rel		(%rip)
 #else
 #define __percpu_seg		fs
+#define __percpu_rel
 #endif
 
 #ifdef __ASSEMBLY__
 
 #ifdef CONFIG_SMP
-#define PER_CPU_VAR(var)	%__percpu_seg:var
-#else /* ! SMP */
-#define PER_CPU_VAR(var)	var
-#endif	/* SMP */
+#define __percpu		%__percpu_seg:
+#else
+#define __percpu
+#endif
+
+#define PER_CPU_VAR(var)	__percpu(var)__percpu_rel
 
 #ifdef CONFIG_X86_64_SMP
 #define INIT_PER_CPU_VAR(var)  init_per_cpu__##var
diff --git a/arch/x86/lib/cmpxchg16b_emu.S b/arch/x86/lib/cmpxchg16b_emu.S
index 6962df315793..4fb44894ad87 100644
--- a/arch/x86/lib/cmpxchg16b_emu.S
+++ b/arch/x86/lib/cmpxchg16b_emu.S
@@ -23,14 +23,14 @@ SYM_FUNC_START(this_cpu_cmpxchg16b_emu)
 	cli
 
 	/* if (*ptr == old) */
-	cmpq	PER_CPU_VAR(0(%rsi)), %rax
+	cmpq	__percpu (%rsi), %rax
 	jne	.Lnot_same
-	cmpq	PER_CPU_VAR(8(%rsi)), %rdx
+	cmpq	__percpu 8(%rsi), %rdx
 	jne	.Lnot_same
 
 	/* *ptr = new */
-	movq	%rbx, PER_CPU_VAR(0(%rsi))
-	movq	%rcx, PER_CPU_VAR(8(%rsi))
+	movq	%rbx, __percpu (%rsi)
+	movq	%rcx, __percpu 8(%rsi)
 
 	/* set ZF in EFLAGS to indicate success */
 	orl	$X86_EFLAGS_ZF, (%rsp)
@@ -42,8 +42,8 @@ SYM_FUNC_START(this_cpu_cmpxchg16b_emu)
 	/* *ptr != old */
 
 	/* old = *ptr */
-	movq	PER_CPU_VAR(0(%rsi)), %rax
-	movq	PER_CPU_VAR(8(%rsi)), %rdx
+	movq	__percpu (%rsi), %rax
+	movq	__percpu 8(%rsi), %rdx
 
 	/* clear ZF in EFLAGS to indicate failure */
 	andl	$(~X86_EFLAGS_ZF), (%rsp)
diff --git a/arch/x86/lib/cmpxchg8b_emu.S b/arch/x86/lib/cmpxchg8b_emu.S
index 873e4ef23e49..1c96be769adc 100644
--- a/arch/x86/lib/cmpxchg8b_emu.S
+++ b/arch/x86/lib/cmpxchg8b_emu.S
@@ -24,12 +24,12 @@ SYM_FUNC_START(cmpxchg8b_emu)
 	pushfl
 	cli
 
-	cmpl	0(%esi), %eax
+	cmpl	(%esi), %eax
 	jne	.Lnot_same
 	cmpl	4(%esi), %edx
 	jne	.Lnot_same
 
-	movl	%ebx, 0(%esi)
+	movl	%ebx, (%esi)
 	movl	%ecx, 4(%esi)
 
 	orl	$X86_EFLAGS_ZF, (%esp)
@@ -38,7 +38,7 @@ SYM_FUNC_START(cmpxchg8b_emu)
 	RET
 
 .Lnot_same:
-	movl	0(%esi), %eax
+	movl	(%esi), %eax
 	movl	4(%esi), %edx
 
 	andl	$(~X86_EFLAGS_ZF), (%esp)
@@ -53,18 +53,30 @@ EXPORT_SYMBOL(cmpxchg8b_emu)
 
 #ifndef CONFIG_UML
 
+/*
+ * Emulate 'cmpxchg8b %fs:(%rsi)'
+ *
+ * Inputs:
+ * %esi : memory location to compare
+ * %eax : low 32 bits of old value
+ * %edx : high 32 bits of old value
+ * %ebx : low 32 bits of new value
+ * %ecx : high 32 bits of new value
+ *
+ * Notably this is not LOCK prefixed and is not safe against NMIs
+ */
 SYM_FUNC_START(this_cpu_cmpxchg8b_emu)
 
 	pushfl
 	cli
 
-	cmpl	PER_CPU_VAR(0(%esi)), %eax
+	cmpl	__percpu (%esi), %eax
 	jne	.Lnot_same2
-	cmpl	PER_CPU_VAR(4(%esi)), %edx
+	cmpl	__percpu 4(%esi), %edx
 	jne	.Lnot_same2
 
-	movl	%ebx, PER_CPU_VAR(0(%esi))
-	movl	%ecx, PER_CPU_VAR(4(%esi))
+	movl	%ebx, __percpu (%esi)
+	movl	%ecx, __percpu 4(%esi)
 
 	orl	$X86_EFLAGS_ZF, (%esp)
 
@@ -72,8 +84,8 @@ SYM_FUNC_START(this_cpu_cmpxchg8b_emu)
 	RET
 
 .Lnot_same2:
-	movl	PER_CPU_VAR(0(%esi)), %eax
-	movl	PER_CPU_VAR(4(%esi)), %edx
+	movl	__percpu (%esi), %eax
+	movl	__percpu 4(%esi), %edx
 
 	andl	$(~X86_EFLAGS_ZF), (%esp)
 
-- 
2.41.0