Skip to contentSkip to conversation

x86/asm changes for v3.4

GIT PULL Ingo Molnar 1 participant 1 message Mar 19, 2012

Subsystems (3)
x86 Generic include/asm header files Locking primitives
lore
Share & tools
lore

Ingo Molnar

author
#0 lore raw

Linus,

Please pull the latest x86-asm-for-linus git tree from:

git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86-asm-for-linus

HEAD: a240ada241dafe290e7532d1ddeb98fdf1419068 x86: Include probe_roms.h in probe_roms.c

out-of-topic modifications in x86-asm-for-linus:
------------------------------------------------
include/asm-generic/getorder.h # b893485: bitops: Add missing parentheses t
# d66acc3: bitops: Optimise get_order() # e0891a9: bitops: Adjust the comment on get

Thanks,

Ingo

------------------>
Alex Shi (1):
x86/numa: Improve internode cache alignment

David Howells (2):
bitops: Adjust the comment on get_order() to describe the size==0 case
bitops: Optimise get_order()

Jan Beulich (10):
x86-64: Slightly shorten copy_page()
x86-64: Fix memset() to support sizes of 4Gb and above
x86-64: Fix memcpy() to support sizes of 4Gb and above
x86-64: Handle byte-wise tail copying in memcpy() without a loop x86/spinlocks: Eliminate TICKET_MASK
x86-64: Fix CFI annotations for NMI nesting code
x86-64: Improve insn scheduling in SAVE_ARGS_IRQ
x86: Tighten dependencies of CPU_SUP_*_32
x86/32: Print control and debug registers for kerenel context
x86: Include probe_roms.h in probe_roms.c

Joerg Roedel (1):
bitops: Add missing parentheses to new get_order macro

Steven Rostedt (1):
x86: Fix the NMI nesting comments

 arch/x86/Kconfig.cpu                  |    5 +-
 arch/x86/include/asm/spinlock.h       |    4 +-
 arch/x86/include/asm/spinlock_types.h |    1 -
 arch/x86/kernel/dumpstack_32.c        |    2 +-
 arch/x86/kernel/entry_64.S            |   71 +++++++++++++++++---------------
 arch/x86/kernel/probe_roms.c          |    1 +
 arch/x86/lib/copy_page_64.S           |   12 ++----
 arch/x86/lib/memcpy_64.S              |   44 +++++++++-----------
 arch/x86/lib/memset_64.S              |   33 +++++++--------
 include/asm-generic/getorder.h        |   53 +++++++++++++++++++++----
 10 files changed, 128 insertions(+), 98 deletions(-)
arch/x86/Kconfig.cpu +2 -3
index 3c57033..706e12e 100644
@@ -303,7 +303,6 @@ config X86_GENERIC
 config X86_INTERNODE_CACHE_SHIFT
 	int
 	default "12" if X86_VSMP
-	default "7" if NUMA
 	default X86_L1_CACHE_SHIFT
 
 config X86_CMPXCHG

@@ -441,7 +440,7 @@ config CPU_SUP_INTEL
 config CPU_SUP_CYRIX_32
 	default y
 	bool "Support Cyrix processors" if PROCESSOR_SELECT
-	depends on !64BIT
+	depends on M386 || M486 || M586 || M586TSC || M586MMX || (EXPERT && !64BIT)
 	---help---
 	  This enables detection, tunings and quirks for Cyrix processors
 

@@ -495,7 +494,7 @@ config CPU_SUP_TRANSMETA_32
 config CPU_SUP_UMC_32
 	default y
 	bool "Support UMC processors" if PROCESSOR_SELECT
-	depends on !64BIT
+	depends on M386 || M486 || (EXPERT && !64BIT)
 	---help---
 	  This enables detection, tunings and quirks for UMC processors
arch/x86/include/asm/spinlock.h +2 -2
index a82c2bf..76bfa2c 100644
@@ -88,14 +88,14 @@ static inline int __ticket_spin_is_locked(arch_spinlock_t *lock)
 {
 	struct __raw_tickets tmp = ACCESS_ONCE(lock->tickets);
 
-	return !!(tmp.tail ^ tmp.head);
+	return tmp.tail != tmp.head;
 }
 
 static inline int __ticket_spin_is_contended(arch_spinlock_t *lock)
 {
 	struct __raw_tickets tmp = ACCESS_ONCE(lock->tickets);
 
-	return ((tmp.tail - tmp.head) & TICKET_MASK) > 1;
+	return (__ticket_t)(tmp.tail - tmp.head) > 1;
 }
 
 #ifndef CONFIG_PARAVIRT_SPINLOCKS
arch/x86/include/asm/spinlock_types.h +0 -1
index 8ebd5df..ad0ad07 100644
@@ -16,7 +16,6 @@ typedef u32 __ticketpair_t;
 #endif
 
 #define TICKET_SHIFT	(sizeof(__ticket_t) * 8)
-#define TICKET_MASK	((__ticket_t)((1 << TICKET_SHIFT) - 1))
 
 typedef struct arch_spinlock {
 	union {
arch/x86/kernel/dumpstack_32.c +1 -1
index c99f9ed..88ec912 100644
@@ -87,7 +87,7 @@ void show_registers(struct pt_regs *regs)
 	int i;
 
 	print_modules();
-	__show_regs(regs, 0);
+	__show_regs(regs, !user_mode_vm(regs));
 
 	printk(KERN_EMERG "Process %.*s (pid: %d, ti=%p task=%p task.ti=%p)\n",
 		TASK_COMM_LEN, current->comm, task_pid_nr(current),
arch/x86/kernel/entry_64.S +38 -33

136 lines not on this page · raw message

arch/x86/kernel/probe_roms.c +1 -0
index 34e06e8..0bc72e2 100644
@@ -12,6 +12,7 @@
 #include <linux/pci.h>
 #include <linux/export.h>
 
+#include <asm/probe_roms.h>
 #include <asm/pci-direct.h>
 #include <asm/e820.h>
 #include <asm/mmzone.h>
arch/x86/lib/copy_page_64.S +4 -8
index 01c805b..6b34d04 100644
@@ -20,14 +20,12 @@ ENDPROC(copy_page_c)
 
 ENTRY(copy_page)
 	CFI_STARTPROC
-	subq	$3*8,%rsp
-	CFI_ADJUST_CFA_OFFSET 3*8
+	subq	$2*8,%rsp
+	CFI_ADJUST_CFA_OFFSET 2*8
 	movq	%rbx,(%rsp)
 	CFI_REL_OFFSET rbx, 0
 	movq	%r12,1*8(%rsp)
 	CFI_REL_OFFSET r12, 1*8
-	movq	%r13,2*8(%rsp)
-	CFI_REL_OFFSET r13, 2*8
 
 	movl	$(4096/64)-5,%ecx
 	.p2align 4

@@ -91,10 +89,8 @@ ENTRY(copy_page)
 	CFI_RESTORE rbx
 	movq	1*8(%rsp),%r12
 	CFI_RESTORE r12
-	movq	2*8(%rsp),%r13
-	CFI_RESTORE r13
-	addq	$3*8,%rsp
-	CFI_ADJUST_CFA_OFFSET -3*8
+	addq	$2*8,%rsp
+	CFI_ADJUST_CFA_OFFSET -2*8
 	ret
 .Lcopy_page_end:
 	CFI_ENDPROC
arch/x86/lib/memcpy_64.S +20 -24

104 lines not on this page · raw message

arch/x86/lib/memset_64.S +15 -18
index 79bd454..2dcb380 100644
@@ -19,16 +19,15 @@
 	.section .altinstr_replacement, "ax", @progbits
 .Lmemset_c:
 	movq %rdi,%r9
-	movl %edx,%r8d
-	andl $7,%r8d
-	movl %edx,%ecx
-	shrl $3,%ecx
+	movq %rdx,%rcx
+	andl $7,%edx
+	shrq $3,%rcx
 	/* expand byte value  */
 	movzbl %sil,%esi
 	movabs $0x0101010101010101,%rax
-	mulq %rsi		/* with rax, clobbers rdx */
+	imulq %rsi,%rax
 	rep stosq
-	movl %r8d,%ecx
+	movl %edx,%ecx
 	rep stosb
 	movq %r9,%rax
 	ret

@@ -50,7 +49,7 @@
 .Lmemset_c_e:
 	movq %rdi,%r9
 	movb %sil,%al
-	movl %edx,%ecx
+	movq %rdx,%rcx
 	rep stosb
 	movq %r9,%rax
 	ret

@@ -61,12 +60,11 @@ ENTRY(memset)
 ENTRY(__memset)
 	CFI_STARTPROC
 	movq %rdi,%r10
-	movq %rdx,%r11
 
 	/* expand byte value  */
 	movzbl %sil,%ecx
 	movabs $0x0101010101010101,%rax
-	mul    %rcx		/* with rax, clobbers rdx */
+	imulq  %rcx,%rax
 
 	/* align dst */
 	movl  %edi,%r9d

@@ -75,13 +73,13 @@ ENTRY(__memset)
 	CFI_REMEMBER_STATE
 .Lafter_bad_alignment:
 
-	movl %r11d,%ecx
-	shrl $6,%ecx
+	movq  %rdx,%rcx
+	shrq  $6,%rcx
 	jz	 .Lhandle_tail
 
 	.p2align 4
 .Lloop_64:
-	decl   %ecx
+	decq  %rcx
 	movq  %rax,(%rdi)
 	movq  %rax,8(%rdi)
 	movq  %rax,16(%rdi)

@@ -97,7 +95,7 @@ ENTRY(__memset)
 	   to predict jump tables. */
 	.p2align 4
 .Lhandle_tail:
-	movl	%r11d,%ecx
+	movl	%edx,%ecx
 	andl    $63&(~7),%ecx
 	jz 		.Lhandle_7
 	shrl	$3,%ecx

@@ -109,12 +107,11 @@ ENTRY(__memset)
 	jnz    .Lloop_8
 
 .Lhandle_7:
-	movl	%r11d,%ecx
-	andl	$7,%ecx
+	andl	$7,%edx
 	jz      .Lende
 	.p2align 4
 .Lloop_1:
-	decl    %ecx
+	decl    %edx
 	movb 	%al,(%rdi)
 	leaq	1(%rdi),%rdi
 	jnz     .Lloop_1

@@ -125,13 +122,13 @@ ENTRY(__memset)
 
 	CFI_RESTORE_STATE
 .Lbad_alignment:
-	cmpq $7,%r11
+	cmpq $7,%rdx
 	jbe	.Lhandle_7
 	movq %rax,(%rdi)	/* unaligned store */
 	movq $8,%r8
 	subq %r9,%r8
 	addq %r8,%rdi
-	subq %r8,%r11
+	subq %r8,%rdx
 	jmp .Lafter_bad_alignment
 .Lfinal:
 	CFI_ENDPROC
include/asm-generic/getorder.h +45 -8

66 lines not on this page · raw message