[PATCH x86 for review III] [1/29] i386: avoid gcc extension

[PATCH x86 for review III] [1/29] i386: avoid gcc extension

[Andi Kleen]

From: Randy Dunlap <randy.dunlap@oracle.com>

setcc() in math-emu is written as a gcc extension statement expression
macro that returns a value.  However, it's not used that way and it's not
needed like that, so just make it a do-while non-extension macro so that we
don't use an extension when it's not needed.

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 arch/i386/math-emu/status_w.h |    8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

Index: linux/arch/i386/math-emu/status_w.h
===================================================================
--- linux.orig/arch/i386/math-emu/status_w.h
+++ linux/arch/i386/math-emu/status_w.h
@@ -48,9 +48,11 @@
 
 #define status_word() \
   ((partial_status & ~SW_Top & 0xffff) | ((top << SW_Top_Shift) & SW_Top))
-#define setcc(cc) ({ \
-  partial_status &= ~(SW_C0|SW_C1|SW_C2|SW_C3); \
-  partial_status |= (cc) & (SW_C0|SW_C1|SW_C2|SW_C3); })
+static inline void setcc(int cc)
+{
+	partial_status &= ~(SW_C0|SW_C1|SW_C2|SW_C3);
+	partial_status |= (cc) & (SW_C0|SW_C1|SW_C2|SW_C3);
+}
 
 #ifdef PECULIAR_486
    /* Default, this conveys no information, but an 80486 does it. */

[PATCH x86 for review III] [2/29] i386: support Classic MediaGXm

[Andi Kleen]

From: takada <takada@mbf.nifty.com>

I hope to support "classic" MediaGXm in kernel.

The DIR1 register of MediaGXm( or Geode) shows the following values for
identify CPU.  For example, My MediaGXm shows 0x42.

We can read National Semiconductor's datasheet without any NDAs.
  http://www.national.com/pf/GX/GXLV.html

from datasheets:
DIR1
0x30 - 0x33 GXm rev. 1.0 - 2.3
0x34 - 0x4f GXm rev. 2.4 - 3.x
0x5x        GXm rev. 5.0 - 5.4
0x6x        GXLV
0x7x         (unknow)
0x8x	    Gx1

In nsc driver of X, accept 0x30 through 0x82. What will 0x7x mean?

Cc: Jordan Crouse <jordan.crouse@amd.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Andi Kleen <ak@suse.de>

---

 arch/i386/kernel/cpu/cyrix.c |   11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

Index: linux/arch/i386/kernel/cpu/cyrix.c
===================================================================
--- linux.orig/arch/i386/kernel/cpu/cyrix.c
+++ linux/arch/i386/kernel/cpu/cyrix.c
@@ -285,10 +285,15 @@ static void __cpuinit init_cyrix(struct 
 		/* GXm supports extended cpuid levels 'ala' AMD */
 		if (c->cpuid_level == 2) {
 			/* Enable cxMMX extensions (GX1 Datasheet 54) */
-			setCx86(CX86_CCR7, getCx86(CX86_CCR7)|1);
+			setCx86(CX86_CCR7, getCx86(CX86_CCR7) | 1);
 			
-			/* GXlv/GXm/GX1 */
-			if((dir1 >= 0x50 && dir1 <= 0x54) || dir1 >= 0x63)
+			/*
+			 * GXm : 0x30 ... 0x5f GXm  datasheet 51
+			 * GXlv: 0x6x          GXlv datasheet 54
+			 *  ?  : 0x7x
+			 * GX1 : 0x8x          GX1  datasheet 56
+			 */
+			if((0x30 <= dir1 && dir1 <= 0x6f) || (0x80 <=dir1 && dir1 <= 0x8f))
 				geode_configure();
 			get_model_name(c);  /* get CPU marketing name */
 			return;

[PATCH x86 for review III] [3/29] i386: entry.S END/ENDPROC annotations

[Andi Kleen]

From: "Jan Beulich" <jbeulich@novell.com>

Annotate i386/kernel/entry.S with END/ENDPROC to assist disassemblers and
other analysis tools.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 arch/i386/kernel/entry.S |   41 +++++++++++++++++++++++++++++++++++++----
 1 file changed, 37 insertions(+), 4 deletions(-)

Index: linux/arch/i386/kernel/entry.S
===================================================================
--- linux.orig/arch/i386/kernel/entry.S
+++ linux/arch/i386/kernel/entry.S
@@ -227,6 +227,7 @@ ENTRY(ret_from_fork)
 	CFI_ADJUST_CFA_OFFSET -4
 	jmp syscall_exit
 	CFI_ENDPROC
+END(ret_from_fork)
 
 /*
  * Return to user mode is not as complex as all this looks,
@@ -258,6 +259,7 @@ ENTRY(resume_userspace)
 					# int/exception return?
 	jne work_pending
 	jmp restore_all
+END(ret_from_exception)
 
 #ifdef CONFIG_PREEMPT
 ENTRY(resume_kernel)
@@ -272,6 +274,7 @@ need_resched:
 	jz restore_all
 	call preempt_schedule_irq
 	jmp need_resched
+END(resume_kernel)
 #endif
 	CFI_ENDPROC
 
@@ -359,6 +362,7 @@ sysenter_past_esp:
 	.align 4
 	.long 1b,2b
 .popsection
+ENDPROC(sysenter_entry)
 
 	# system call handler stub
 ENTRY(system_call)
@@ -459,6 +463,7 @@ ldt_ss:
 	CFI_ADJUST_CFA_OFFSET -8
 	jmp restore_nocheck
 	CFI_ENDPROC
+ENDPROC(system_call)
 
 	# perform work that needs to be done immediately before resumption
 	ALIGN
@@ -504,6 +509,7 @@ work_notifysig_v86:
 	xorl %edx, %edx
 	call do_notify_resume
 	jmp resume_userspace_sig
+END(work_pending)
 
 	# perform syscall exit tracing
 	ALIGN
@@ -519,6 +525,7 @@ syscall_trace_entry:
 	cmpl $(nr_syscalls), %eax
 	jnae syscall_call
 	jmp syscall_exit
+END(syscall_trace_entry)
 
 	# perform syscall exit tracing
 	ALIGN
@@ -532,6 +539,7 @@ syscall_exit_work:
 	movl $1, %edx
 	call do_syscall_trace
 	jmp resume_userspace
+END(syscall_exit_work)
 	CFI_ENDPROC
 
 	RING0_INT_FRAME			# can't unwind into user space anyway
@@ -542,10 +550,12 @@ syscall_fault:
 	GET_THREAD_INFO(%ebp)
 	movl $-EFAULT,PT_EAX(%esp)
 	jmp resume_userspace
+END(syscall_fault)
 
 syscall_badsys:
 	movl $-ENOSYS,PT_EAX(%esp)
 	jmp resume_userspace
+END(syscall_badsys)
 	CFI_ENDPROC
 
 #define FIXUP_ESPFIX_STACK \
@@ -581,9 +591,9 @@ syscall_badsys:
 ENTRY(interrupt)
 .text
 
-vector=0
 ENTRY(irq_entries_start)
 	RING0_INT_FRAME
+vector=0
 .rept NR_IRQS
 	ALIGN
  .if vector
@@ -592,11 +602,16 @@ ENTRY(irq_entries_start)
 1:	pushl $~(vector)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp common_interrupt
-.data
+ .previous
 	.long 1b
-.text
+ .text
 vector=vector+1
 .endr
+END(irq_entries_start)
+
+.previous
+END(interrupt)
+.previous
 
 /*
  * the CPU automatically disables interrupts when executing an IRQ vector,
@@ -609,6 +624,7 @@ common_interrupt:
 	movl %esp,%eax
 	call do_IRQ
 	jmp ret_from_intr
+ENDPROC(common_interrupt)
 	CFI_ENDPROC
 
 #define BUILD_INTERRUPT(name, nr)	\
@@ -621,7 +637,8 @@ ENTRY(name)				\
 	movl %esp,%eax;			\
 	call smp_/**/name;		\
 	jmp ret_from_intr;		\
-	CFI_ENDPROC
+	CFI_ENDPROC;			\
+ENDPROC(name)
 
 /* The include is where all of the SMP etc. interrupts come from */
 #include "entry_arch.h"
@@ -697,6 +714,7 @@ ENTRY(coprocessor_error)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
+END(coprocessor_error)
 
 ENTRY(simd_coprocessor_error)
 	RING0_INT_FRAME
@@ -706,6 +724,7 @@ ENTRY(simd_coprocessor_error)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
+END(simd_coprocessor_error)
 
 ENTRY(device_not_available)
 	RING0_INT_FRAME
@@ -726,6 +745,7 @@ device_not_available_emulate:
 	CFI_ADJUST_CFA_OFFSET -4
 	jmp ret_from_exception
 	CFI_ENDPROC
+END(device_not_available)
 
 /*
  * Debug traps and NMI can happen at the one SYSENTER instruction
@@ -869,10 +889,12 @@ ENTRY(native_iret)
 	.align 4
 	.long 1b,iret_exc
 .previous
+END(native_iret)
 
 ENTRY(native_irq_enable_sysexit)
 	sti
 	sysexit
+END(native_irq_enable_sysexit)
 #endif
 
 KPROBE_ENTRY(int3)
@@ -895,6 +917,7 @@ ENTRY(overflow)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
+END(overflow)
 
 ENTRY(bounds)
 	RING0_INT_FRAME
@@ -904,6 +927,7 @@ ENTRY(bounds)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
+END(bounds)
 
 ENTRY(invalid_op)
 	RING0_INT_FRAME
@@ -913,6 +937,7 @@ ENTRY(invalid_op)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
+END(invalid_op)
 
 ENTRY(coprocessor_segment_overrun)
 	RING0_INT_FRAME
@@ -922,6 +947,7 @@ ENTRY(coprocessor_segment_overrun)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
+END(coprocessor_segment_overrun)
 
 ENTRY(invalid_TSS)
 	RING0_EC_FRAME
@@ -929,6 +955,7 @@ ENTRY(invalid_TSS)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
+END(invalid_TSS)
 
 ENTRY(segment_not_present)
 	RING0_EC_FRAME
@@ -936,6 +963,7 @@ ENTRY(segment_not_present)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
+END(segment_not_present)
 
 ENTRY(stack_segment)
 	RING0_EC_FRAME
@@ -943,6 +971,7 @@ ENTRY(stack_segment)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
+END(stack_segment)
 
 KPROBE_ENTRY(general_protection)
 	RING0_EC_FRAME
@@ -958,6 +987,7 @@ ENTRY(alignment_check)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
+END(alignment_check)
 
 ENTRY(divide_error)
 	RING0_INT_FRAME
@@ -967,6 +997,7 @@ ENTRY(divide_error)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
+END(divide_error)
 
 #ifdef CONFIG_X86_MCE
 ENTRY(machine_check)
@@ -977,6 +1008,7 @@ ENTRY(machine_check)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
+END(machine_check)
 #endif
 
 ENTRY(spurious_interrupt_bug)
@@ -987,6 +1019,7 @@ ENTRY(spurious_interrupt_bug)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
+END(spurious_interrupt_bug)
 
 ENTRY(kernel_thread_helper)
 	pushl $0		# fake return address for unwinder

[PATCH x86 for review III] [4/29] x86_64: clean up sparsemem memory_present call

[Andi Kleen]

From: "Bob Picco" <bob.picco@hp.com>

Eliminate arch specific memory_present call x86_64 NUMA by utilizing
sparse_memory_present_with_active_regions.

Acked-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Bob Picco <bob.picco@hp.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 arch/x86_64/mm/numa.c |   17 ++---------------
 1 file changed, 2 insertions(+), 15 deletions(-)

Index: linux/arch/x86_64/mm/numa.c
===================================================================
--- linux.orig/arch/x86_64/mm/numa.c
+++ linux/arch/x86_64/mm/numa.c
@@ -460,20 +460,6 @@ unsigned long __init numa_free_all_bootm
 	return pages;
 } 
 
-#ifdef CONFIG_SPARSEMEM
-static void __init arch_sparse_init(void)
-{
-	int i;
-
-	for_each_online_node(i)
-		memory_present(i, node_start_pfn(i), node_end_pfn(i));
-
-	sparse_init();
-}
-#else
-#define arch_sparse_init() do {} while (0)
-#endif
-
 void __init paging_init(void)
 { 
 	int i;
@@ -483,7 +469,8 @@ void __init paging_init(void)
 	max_zone_pfns[ZONE_DMA32] = MAX_DMA32_PFN;
 	max_zone_pfns[ZONE_NORMAL] = end_pfn;
 
-	arch_sparse_init();
+	sparse_memory_present_with_active_regions(MAX_NUMNODES);
+	sparse_init();
 
 	for_each_online_node(i) {
 		setup_node_zones(i); 

[PATCH x86 for review III] [5/29] i386: arch/i386/kernel/alternative.c should #include <asm/bugs.h>

[Andi Kleen]

From: Adrian Bunk <bunk@stusta.de>

Every file should include the headers containing the prototypes for its
global functions.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Andi Kleen <ak@suse.de>

---

 arch/i386/kernel/alternative.c |    1 +
 1 file changed, 1 insertion(+)

Index: linux/arch/i386/kernel/alternative.c
===================================================================
--- linux.orig/arch/i386/kernel/alternative.c
+++ linux/arch/i386/kernel/alternative.c
@@ -4,6 +4,7 @@
 #include <linux/list.h>
 #include <asm/alternative.h>
 #include <asm/sections.h>
+#include <asm/bugs.h>
 
 static int no_replacement    = 0;
 static int smp_alt_once      = 0;

[PATCH x86 for review III] [6/29] i386: Remove unused kernel config option X86_XADD

[Andi Kleen]

From: "Robert P. J. Day" <rpjday@mindspring.com>

Remove the unused kernel config option X86_XADD, which is unused in any
source or header file.

Signed-off-by: Robert P. J. Day <rpjday@mindspring.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 arch/i386/Kconfig.cpu |    5 -----
 1 file changed, 5 deletions(-)

Index: linux/arch/i386/Kconfig.cpu
===================================================================
--- linux.orig/arch/i386/Kconfig.cpu
+++ linux/arch/i386/Kconfig.cpu
@@ -226,11 +226,6 @@ config X86_CMPXCHG
 	depends on !M386
 	default y
 
-config X86_XADD
-	bool
-	depends on !M386
-	default y
-
 config X86_L1_CACHE_SHIFT
 	int
 	default "7" if MPENTIUM4 || X86_GENERIC

[PATCH x86 for review III] [7/29] x86_64: update IO-APIC dest field to 8-bit for xAPIC

[Andi Kleen]

From: Benjamin Romer <benjamin.romer@unisys.com>

On the Unisys ES7000/ONE system, we encountered a problem where performing
a kexec reboot or dump on any cell other than cell 0 causes the system
timer to stop working, resulting in a hang during timer calibration in the
new kernel.

We traced the problem to one line of code in disable_IO_APIC(), which needs
to restore the timer's IO-APIC configuration before rebooting.  The code is
currently using the 4-bit physical destination field, rather than using the
8-bit logical destination field, and it cuts off the upper 4 bits of the
timer's APIC ID.  If we change this to use the logical destination field,
the timer works and we can kexec on the upper cells.  This was tested on
two different cells (0 and 2) in an ES7000/ONE system.

For reference, the relevant Intel xAPIC spec is kept at
ftp://download.intel.com/design/chipsets/e8501/datashts/30962001.pdf,
specifically on page 334.

Signed-off-by: Benjamin M Romer <benjamin.romer@unisys.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 arch/x86_64/kernel/io_apic.c |   24 +++++++++++-------------
 include/asm-x86_64/io_apic.h |   14 ++------------
 2 files changed, 13 insertions(+), 25 deletions(-)

Index: linux/arch/x86_64/kernel/io_apic.c
===================================================================
--- linux.orig/arch/x86_64/kernel/io_apic.c
+++ linux/arch/x86_64/kernel/io_apic.c
@@ -831,7 +831,7 @@ static void __init setup_IO_APIC_irq(int
 	entry.delivery_mode = INT_DELIVERY_MODE;
 	entry.dest_mode = INT_DEST_MODE;
 	entry.mask = 0;				/* enable IRQ */
-	entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
+	entry.dest = cpu_mask_to_apicid(TARGET_CPUS);
 
 	entry.trigger = irq_trigger(idx);
 	entry.polarity = irq_polarity(idx);
@@ -839,7 +839,7 @@ static void __init setup_IO_APIC_irq(int
 	if (irq_trigger(idx)) {
 		entry.trigger = 1;
 		entry.mask = 1;
-		entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
+		entry.dest = cpu_mask_to_apicid(TARGET_CPUS);
 	}
 
 	if (!apic && !IO_APIC_IRQ(irq))
@@ -851,7 +851,7 @@ static void __init setup_IO_APIC_irq(int
 		if (vector < 0)
 			return;
 
-		entry.dest.logical.logical_dest = cpu_mask_to_apicid(mask);
+		entry.dest = cpu_mask_to_apicid(mask);
 		entry.vector = vector;
 
 		ioapic_register_intr(irq, vector, IOAPIC_AUTO);
@@ -920,7 +920,7 @@ static void __init setup_ExtINT_IRQ0_pin
 	 */
 	entry.dest_mode = INT_DEST_MODE;
 	entry.mask = 0;					/* unmask IRQ now */
-	entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
+	entry.dest = cpu_mask_to_apicid(TARGET_CPUS);
 	entry.delivery_mode = INT_DELIVERY_MODE;
 	entry.polarity = 0;
 	entry.trigger = 0;
@@ -1020,18 +1020,17 @@ void __apicdebuginit print_IO_APIC(void)
 
 	printk(KERN_DEBUG ".... IRQ redirection table:\n");
 
-	printk(KERN_DEBUG " NR Log Phy Mask Trig IRR Pol"
-			  " Stat Dest Deli Vect:   \n");
+	printk(KERN_DEBUG " NR Dst Mask Trig IRR Pol"
+			  " Stat Dmod Deli Vect:   \n");
 
 	for (i = 0; i <= reg_01.bits.entries; i++) {
 		struct IO_APIC_route_entry entry;
 
 		entry = ioapic_read_entry(apic, i);
 
-		printk(KERN_DEBUG " %02x %03X %02X  ",
+		printk(KERN_DEBUG " %02x %03X ",
 			i,
-			entry.dest.logical.logical_dest,
-			entry.dest.physical.physical_dest
+			entry.dest
 		);
 
 		printk("%1d    %1d    %1d   %1d   %1d    %1d    %1d    %02X\n",
@@ -1293,8 +1292,7 @@ void disable_IO_APIC(void)
 		entry.dest_mode       = 0; /* Physical */
 		entry.delivery_mode   = dest_ExtINT; /* ExtInt */
 		entry.vector          = 0;
-		entry.dest.physical.physical_dest =
-					GET_APIC_ID(apic_read(APIC_ID));
+		entry.dest          = GET_APIC_ID(apic_read(APIC_ID));
 
 		/*
 		 * Add it to the IO-APIC irq-routing table:
@@ -1556,7 +1554,7 @@ static inline void unlock_ExtINT_logic(v
 
 	entry1.dest_mode = 0;			/* physical delivery */
 	entry1.mask = 0;			/* unmask IRQ now */
-	entry1.dest.physical.physical_dest = hard_smp_processor_id();
+	entry1.dest = hard_smp_processor_id();
 	entry1.delivery_mode = dest_ExtINT;
 	entry1.polarity = entry0.polarity;
 	entry1.trigger = 0;
@@ -2131,7 +2129,7 @@ int io_apic_set_pci_routing (int ioapic,
 
 	entry.delivery_mode = INT_DELIVERY_MODE;
 	entry.dest_mode = INT_DEST_MODE;
-	entry.dest.logical.logical_dest = cpu_mask_to_apicid(mask);
+	entry.dest = cpu_mask_to_apicid(mask);
 	entry.trigger = triggering;
 	entry.polarity = polarity;
 	entry.mask = 1;					 /* Disabled (masked) */
Index: linux/include/asm-x86_64/io_apic.h
===================================================================
--- linux.orig/include/asm-x86_64/io_apic.h
+++ linux/include/asm-x86_64/io_apic.h
@@ -85,18 +85,8 @@ struct IO_APIC_route_entry {
 		mask		:  1,	/* 0: enabled, 1: disabled */
 		__reserved_2	: 15;
 
-	union {		struct { __u32
-					__reserved_1	: 24,
-					physical_dest	:  4,
-					__reserved_2	:  4;
-			} physical;
-
-			struct { __u32
-					__reserved_1	: 24,
-					logical_dest	:  8;
-			} logical;
-	} dest;
-
+	__u32	__reserved_3	: 24,
+		dest		:  8;
 } __attribute__ ((packed));
 
 /*

[PATCH x86 for review III] [8/29] x86_64: avoid warning message livelock

[Andi Kleen]

From: Roland Dreier <rdreier@cisco.com>

I've seen my box paralyzed by an endless spew of

    rtc: lost some interrupts at 1024Hz.

messages on the serial console.  What seems to be happening is that
something real causes an interrupt to be lost and triggers the
message.  But then printing the message to the serial console (from
the hpet interrupt handler) takes more than 1/1024th of a second, and
then some more interrupts are lost, so the message triggers again....

Fix this by adding a printk_ratelimit() before printing the warning.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Andi Kleen <ak@suse.de>

---

 arch/x86_64/kernel/time.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

Index: linux/arch/x86_64/kernel/time.c
===================================================================
--- linux.orig/arch/x86_64/kernel/time.c
+++ linux/arch/x86_64/kernel/time.c
@@ -1226,8 +1226,9 @@ static void hpet_rtc_timer_reinit(void)
 		if (PIE_on)
 			PIE_count += lost_ints;
 
-		printk(KERN_WARNING "rtc: lost some interrupts at %ldHz.\n",
-		       hpet_rtc_int_freq);
+		if (printk_ratelimit())
+			printk(KERN_WARNING "rtc: lost some interrupts at %ldHz.\n",
+			       hpet_rtc_int_freq);
 	}
 }
 

[PATCH x86 for review III] [9/29] x86_64: Minor patch for compilation warning in x86_64 signal code

[Andi Kleen]

From: Evgeniy Polyakov <johnpol@2ka.mipt.ru>

If DEBUG_SIG is enbaled in source code, ia32_signal.c compiles with warning
due to wrong format string.  Attached patch fixes that.  It is quite minor
update, since by default DEBUG_SIG is not enabled and can not be turned on
without code modification.

Signed-off-by: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 arch/x86_64/ia32/ia32_signal.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

Index: linux/arch/x86_64/ia32/ia32_signal.c
===================================================================
--- linux.orig/arch/x86_64/ia32/ia32_signal.c
+++ linux/arch/x86_64/ia32/ia32_signal.c
@@ -495,7 +495,7 @@ int ia32_setup_frame(int sig, struct k_s
 		ptrace_notify(SIGTRAP);
 
 #if DEBUG_SIG
-	printk("SIG deliver (%s:%d): sp=%p pc=%p ra=%p\n",
+	printk("SIG deliver (%s:%d): sp=%p pc=%lx ra=%u\n",
 		current->comm, current->pid, frame, regs->rip, frame->pretcode);
 #endif
 
@@ -601,7 +601,7 @@ int ia32_setup_rt_frame(int sig, struct 
 		ptrace_notify(SIGTRAP);
 
 #if DEBUG_SIG
-	printk("SIG deliver (%s:%d): sp=%p pc=%p ra=%p\n",
+	printk("SIG deliver (%s:%d): sp=%p pc=%lx ra=%u\n",
 		current->comm, current->pid, frame, regs->rip, frame->pretcode);
 #endif
 

[PATCH x86 for review III] [10/29] i386: don't include bugs.h

[Andi Kleen]

From: Andrew Morton <akpm@osdl.org>

That stupid non-inlined-static function in bugs.h causes:

include/asm/bugs.h:186: warning: 'check_bugs' defined but not used

But fortunately the include isn't needed.

Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Andi Kleen <ak@suse.de>

---

 arch/i386/kernel/alternative.c |    1 -
 1 file changed, 1 deletion(-)

Index: linux/arch/i386/kernel/alternative.c
===================================================================
--- linux.orig/arch/i386/kernel/alternative.c
+++ linux/arch/i386/kernel/alternative.c
@@ -4,7 +4,6 @@
 #include <linux/list.h>
 #include <asm/alternative.h>
 #include <asm/sections.h>
-#include <asm/bugs.h>
 
 static int no_replacement    = 0;
 static int smp_alt_once      = 0;

[PATCH x86 for review III] [11/29] i386: add option to show more code in oops reports

[Andi Kleen]

From: Chuck Ebbert <cebbert@redhat.com>

Sometimes developers need to see more object code in an oops report,
e.g. when kernel may be corrupted at runtime.

Add the "code_bytes" option for this.

Signed-off-by: Chuck Ebbert <cebbert@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 Documentation/kernel-parameters.txt |    5 +++++
 arch/i386/kernel/traps.c            |   20 ++++++++++++++++----
 2 files changed, 21 insertions(+), 4 deletions(-)

Index: linux/Documentation/kernel-parameters.txt
===================================================================
--- linux.orig/Documentation/kernel-parameters.txt
+++ linux/Documentation/kernel-parameters.txt
@@ -364,6 +364,11 @@ and is between 256 and 4096 characters. 
 			clocksource is not available, it defaults to PIT.
 			Format: { pit | tsc | cyclone | pmtmr }
 
+	code_bytes	[IA32] How many bytes of object code to print in an
+			oops report.
+			Range: 0 - 8192
+			Default: 64
+
 	disable_8254_timer
 	enable_8254_timer
 			[IA32/X86_64] Disable/Enable interrupt 0 timer routing
Index: linux/arch/i386/kernel/traps.c
===================================================================
--- linux.orig/arch/i386/kernel/traps.c
+++ linux/arch/i386/kernel/traps.c
@@ -94,6 +94,7 @@ asmlinkage void spurious_interrupt_bug(v
 asmlinkage void machine_check(void);
 
 int kstack_depth_to_print = 24;
+static unsigned int code_bytes = 64;
 ATOMIC_NOTIFIER_HEAD(i386die_chain);
 
 int register_die_notifier(struct notifier_block *nb)
@@ -325,7 +326,8 @@ void show_registers(struct pt_regs *regs
 	 */
 	if (in_kernel) {
 		u8 *eip;
-		int code_bytes = 64;
+		unsigned int code_prologue = code_bytes * 43 / 64;
+		unsigned int code_len = code_bytes;
 		unsigned char c;
 
 		printk("\n" KERN_EMERG "Stack: ");
@@ -333,14 +335,14 @@ void show_registers(struct pt_regs *regs
 
 		printk(KERN_EMERG "Code: ");
 
-		eip = (u8 *)regs->eip - 43;
+		eip = (u8 *)regs->eip - code_prologue;
 		if (eip < (u8 *)PAGE_OFFSET ||
 			probe_kernel_address(eip, c)) {
 			/* try starting at EIP */
 			eip = (u8 *)regs->eip;
-			code_bytes = 32;
+			code_len = code_len - code_prologue + 1;
 		}
-		for (i = 0; i < code_bytes; i++, eip++) {
+		for (i = 0; i < code_len; i++, eip++) {
 			if (eip < (u8 *)PAGE_OFFSET ||
 				probe_kernel_address(eip, c)) {
 				printk(" Bad EIP value.");
@@ -1190,3 +1192,13 @@ static int __init kstack_setup(char *s)
 	return 1;
 }
 __setup("kstack=", kstack_setup);
+
+static int __init code_bytes_setup(char *s)
+{
+	code_bytes = simple_strtoul(s, NULL, 0);
+	if (code_bytes > 8192)
+		code_bytes = 8192;
+
+	return 1;
+}
+__setup("code_bytes=", code_bytes_setup);

[PATCH x86 for review III] [12/29] x86_64: 32-bit ptrace mangles sixth system call argument

[Andi Kleen]

From: Jeff Dike <jdike@addtoit.com>

The 32-bit sysenter entry point mangles the sixth system call argument for
both 32-bit and 64-bit ptrace.  In both cases, strace shows the frame
pointer (ebp) as the sixth argument.

Here's a snippet of a 64-bit strace of a 32-bit test program which
calls mmap through sysenter:

mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0xfff00fcc) = 0xfffffffff7f7a000
fstat64(0x1, 0xfff008d8)                = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0xfff0089c) = 0xfffffffff7f79000
write(1, "mmap returns 0xf7f7a000\n", 24mmap returns 0xf7f7a000
) = 24

Here's a 32-bit strace of the same program:

mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0xffc224ec) = 0xf7fcb000
fstat64(1, {st_mode=S_IFCHR|0620, st_rdev=makedev(136, 1), ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0xffc21dbc) = 0xf7fca000
write(1, "mmap returns 0xf7fcb000\n", 24mmap returns 0xf7fcb000
) = 24

The first mmap is the one made by the test - its final argument (the
offset) is 0, but strace shows 0xfff00fcc, which is the value of ebp.
The second is a guilty bystander which is also showing the bug.

The patch below copies %r9 (where the sixth argument has been
stashed) into the RBP slot of pt_regs before syscall_trace_enter is
called.  This fixes ptrace.

To allow a successful return to userspace, the original value of rbp
must be restored.  This is done by storing the current value of rbp
into the RBP slot of pt_regs before the RESTORE_REST.

With this patch, the straces now look like this:

64-bit:

mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xfffffffff7f5a000
fstat64(0x1, 0xff926ee8)                = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xfffffffff7f59000
write(1, "mmap returns 0xf7f5a000\n", 24mmap returns 0xf7f5a000
) = 24

32-bit:

mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xf7fa9000
fstat64(1, {st_mode=S_IFCHR|0620, st_rdev=makedev(136, 1), ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xf7fa8000
write(1, "mmap returns 0xf7fa9000\n", 24mmap returns 0xf7fa9000
) = 24

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 arch/x86_64/ia32/ia32entry.S |   12 ++++++++++++
 1 file changed, 12 insertions(+)

Index: linux/arch/x86_64/ia32/ia32entry.S
===================================================================
--- linux.orig/arch/x86_64/ia32/ia32entry.S
+++ linux/arch/x86_64/ia32/ia32entry.S
@@ -148,11 +148,23 @@ sysenter_do_call:	
 sysenter_tracesys:
 	CFI_RESTORE_STATE
 	SAVE_REST
+	/*
+	 * We need the 6th system call argument to be in regs->rbp at
+	 * this point so that ptrace will see it.  It's in r9 now, so copy
+	 * it to the rbp slot now.
+	 */
+	movq	%r9, RBP(%rsp)
 	CLEAR_RREGS
 	movq	$-ENOSYS,RAX(%rsp)	/* really needed? */
 	movq	%rsp,%rdi        /* &pt_regs -> arg1 */
 	call	syscall_trace_enter
 	LOAD_ARGS ARGOFFSET  /* reload args from stack in case ptrace changed it */
+	/*
+	 * Now, we need the correct value of rbp to be restored.  It
+	 * was never munged, so we can save it to the rbp slot and
+	 * just have it restored.
+	 */
+	movq	%rbp, RBP(%rsp)
 	RESTORE_REST
 	movl	%ebp, %ebp
 	/* no need to do an access_ok check here because rbp has been

[PATCH x86 for review III] [13/29] i386: geode configuration fixes

[Andi Kleen]

From: TAKADA Yoshihito <takada@mbf.nifty.com>

Original code doesn't write back to CCR4 register.  This patch reflects a
value of a register.

Cc: Jordan Crouse <jordan.crouse@amd.com>
Acked-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Andi Kleen <ak@suse.de>

---

 arch/i386/kernel/cpu/cyrix.c |   21 ++++++++++-----------
 1 file changed, 10 insertions(+), 11 deletions(-)

Index: linux/arch/i386/kernel/cpu/cyrix.c
===================================================================
--- linux.orig/arch/i386/kernel/cpu/cyrix.c
+++ linux/arch/i386/kernel/cpu/cyrix.c
@@ -161,19 +161,19 @@ static void __cpuinit set_cx86_inc(void)
 static void __cpuinit geode_configure(void)
 {
 	unsigned long flags;
-	u8 ccr3, ccr4;
+	u8 ccr3;
 	local_irq_save(flags);
 
 	/* Suspend on halt power saving and enable #SUSP pin */
 	setCx86(CX86_CCR2, getCx86(CX86_CCR2) | 0x88);
 
 	ccr3 = getCx86(CX86_CCR3);
-	setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10);	/* Enable */
+	setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10);	/* enable MAPEN */
 	
-	ccr4 = getCx86(CX86_CCR4);
-	ccr4 |= 0x38;		/* FPU fast, DTE cache, Mem bypass */
-	
-	setCx86(CX86_CCR3, ccr3);
+
+	/* FPU fast, DTE cache, Mem bypass */
+	setCx86(CX86_CCR4, getCx86(CX86_CCR4) | 0x38);
+	setCx86(CX86_CCR3, ccr3);			/* disable MAPEN */
 	
 	set_cx86_memwb();
 	set_cx86_reorder();	
@@ -420,15 +420,14 @@ static void __cpuinit cyrix_identify(str
 		
    	        if (dir0 == 5 || dir0 == 3)
    	        {
-			unsigned char ccr3, ccr4;
+			unsigned char ccr3;
 			unsigned long flags;
 			printk(KERN_INFO "Enabling CPUID on Cyrix processor.\n");
 			local_irq_save(flags);
 			ccr3 = getCx86(CX86_CCR3);
-			setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN  */
-			ccr4 = getCx86(CX86_CCR4);
-			setCx86(CX86_CCR4, ccr4 | 0x80);          /* enable cpuid  */
-			setCx86(CX86_CCR3, ccr3);                 /* disable MAPEN */
+			setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10);       /* enable MAPEN  */
+			setCx86(CX86_CCR4, getCx86(CX86_CCR4) | 0x80);  /* enable cpuid  */
+			setCx86(CX86_CCR3, ccr3);                       /* disable MAPEN */
 			local_irq_restore(flags);
 		}
 	}

[PATCH x86 for review III] [15/29] x86_64: Fix wrong gcc check in bitops.h

[Andi Kleen]

gcc 5.0 will likely not have the constraint problem
Signed-off-by: Andi Kleen <ak@suse.de>

---
 include/asm-x86_64/bitops.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Index: linux/include/asm-x86_64/bitops.h
===================================================================
--- linux.orig/include/asm-x86_64/bitops.h
+++ linux/include/asm-x86_64/bitops.h
@@ -7,7 +7,7 @@
 
 #include <asm/alternative.h>
 
-#if __GNUC__ < 4 || __GNUC_MINOR__ < 1
+#if __GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 1)
 /* Technically wrong, but this avoids compilation errors on some gcc
    versions. */
 #define ADDR "=m" (*(volatile long *) addr)

[PATCH x86 for review III] [16/29] i386: Remove fastcall in paravirt.[ch]

[Andi Kleen]

Not needed because fastcall is always default now

Signed-off-by: Andi Kleen <ak@suse.de>

---
 arch/i386/kernel/paravirt.c |  102 ++++++++++++++++-----------------
 include/asm-i386/paravirt.h |  136 ++++++++++++++++++++++----------------------
 2 files changed, 119 insertions(+), 119 deletions(-)

Index: linux/arch/i386/kernel/paravirt.c
===================================================================
--- linux.orig/arch/i386/kernel/paravirt.c
+++ linux/arch/i386/kernel/paravirt.c
@@ -92,7 +92,7 @@ static unsigned native_patch(u8 type, u1
 	return insn_len;
 }
 
-static fastcall unsigned long native_get_debugreg(int regno)
+static unsigned long native_get_debugreg(int regno)
 {
 	unsigned long val = 0; 	/* Damn you, gcc! */
 
@@ -115,7 +115,7 @@ static fastcall unsigned long native_get
 	return val;
 }
 
-static fastcall void native_set_debugreg(int regno, unsigned long value)
+static void native_set_debugreg(int regno, unsigned long value)
 {
 	switch (regno) {
 	case 0:
@@ -146,55 +146,55 @@ void init_IRQ(void)
 	paravirt_ops.init_IRQ();
 }
 
-static fastcall void native_clts(void)
+static void native_clts(void)
 {
 	asm volatile ("clts");
 }
 
-static fastcall unsigned long native_read_cr0(void)
+static unsigned long native_read_cr0(void)
 {
 	unsigned long val;
 	asm volatile("movl %%cr0,%0\n\t" :"=r" (val));
 	return val;
 }
 
-static fastcall void native_write_cr0(unsigned long val)
+static void native_write_cr0(unsigned long val)
 {
 	asm volatile("movl %0,%%cr0": :"r" (val));
 }
 
-static fastcall unsigned long native_read_cr2(void)
+static unsigned long native_read_cr2(void)
 {
 	unsigned long val;
 	asm volatile("movl %%cr2,%0\n\t" :"=r" (val));
 	return val;
 }
 
-static fastcall void native_write_cr2(unsigned long val)
+static void native_write_cr2(unsigned long val)
 {
 	asm volatile("movl %0,%%cr2": :"r" (val));
 }
 
-static fastcall unsigned long native_read_cr3(void)
+static unsigned long native_read_cr3(void)
 {
 	unsigned long val;
 	asm volatile("movl %%cr3,%0\n\t" :"=r" (val));
 	return val;
 }
 
-static fastcall void native_write_cr3(unsigned long val)
+static void native_write_cr3(unsigned long val)
 {
 	asm volatile("movl %0,%%cr3": :"r" (val));
 }
 
-static fastcall unsigned long native_read_cr4(void)
+static unsigned long native_read_cr4(void)
 {
 	unsigned long val;
 	asm volatile("movl %%cr4,%0\n\t" :"=r" (val));
 	return val;
 }
 
-static fastcall unsigned long native_read_cr4_safe(void)
+static unsigned long native_read_cr4_safe(void)
 {
 	unsigned long val;
 	/* This could fault if %cr4 does not exist */
@@ -207,51 +207,51 @@ static fastcall unsigned long native_rea
 	return val;
 }
 
-static fastcall void native_write_cr4(unsigned long val)
+static void native_write_cr4(unsigned long val)
 {
 	asm volatile("movl %0,%%cr4": :"r" (val));
 }
 
-static fastcall unsigned long native_save_fl(void)
+static unsigned long native_save_fl(void)
 {
 	unsigned long f;
 	asm volatile("pushfl ; popl %0":"=g" (f): /* no input */);
 	return f;
 }
 
-static fastcall void native_restore_fl(unsigned long f)
+static void native_restore_fl(unsigned long f)
 {
 	asm volatile("pushl %0 ; popfl": /* no output */
 			     :"g" (f)
 			     :"memory", "cc");
 }
 
-static fastcall void native_irq_disable(void)
+static void native_irq_disable(void)
 {
 	asm volatile("cli": : :"memory");
 }
 
-static fastcall void native_irq_enable(void)
+static void native_irq_enable(void)
 {
 	asm volatile("sti": : :"memory");
 }
 
-static fastcall void native_safe_halt(void)
+static void native_safe_halt(void)
 {
 	asm volatile("sti; hlt": : :"memory");
 }
 
-static fastcall void native_halt(void)
+static void native_halt(void)
 {
 	asm volatile("hlt": : :"memory");
 }
 
-static fastcall void native_wbinvd(void)
+static void native_wbinvd(void)
 {
 	asm volatile("wbinvd": : :"memory");
 }
 
-static fastcall unsigned long long native_read_msr(unsigned int msr, int *err)
+static unsigned long long native_read_msr(unsigned int msr, int *err)
 {
 	unsigned long long val;
 
@@ -270,7 +270,7 @@ static fastcall unsigned long long nativ
 	return val;
 }
 
-static fastcall int native_write_msr(unsigned int msr, unsigned long long val)
+static int native_write_msr(unsigned int msr, unsigned long long val)
 {
 	int err;
 	asm volatile("2: wrmsr ; xorl %0,%0\n"
@@ -288,53 +288,53 @@ static fastcall int native_write_msr(uns
 	return err;
 }
 
-static fastcall unsigned long long native_read_tsc(void)
+static unsigned long long native_read_tsc(void)
 {
 	unsigned long long val;
 	asm volatile("rdtsc" : "=A" (val));
 	return val;
 }
 
-static fastcall unsigned long long native_read_pmc(void)
+static unsigned long long native_read_pmc(void)
 {
 	unsigned long long val;
 	asm volatile("rdpmc" : "=A" (val));
 	return val;
 }
 
-static fastcall void native_load_tr_desc(void)
+static void native_load_tr_desc(void)
 {
 	asm volatile("ltr %w0"::"q" (GDT_ENTRY_TSS*8));
 }
 
-static fastcall void native_load_gdt(const struct Xgt_desc_struct *dtr)
+static void native_load_gdt(const struct Xgt_desc_struct *dtr)
 {
 	asm volatile("lgdt %0"::"m" (*dtr));
 }
 
-static fastcall void native_load_idt(const struct Xgt_desc_struct *dtr)
+static void native_load_idt(const struct Xgt_desc_struct *dtr)
 {
 	asm volatile("lidt %0"::"m" (*dtr));
 }
 
-static fastcall void native_store_gdt(struct Xgt_desc_struct *dtr)
+static void native_store_gdt(struct Xgt_desc_struct *dtr)
 {
 	asm ("sgdt %0":"=m" (*dtr));
 }
 
-static fastcall void native_store_idt(struct Xgt_desc_struct *dtr)
+static void native_store_idt(struct Xgt_desc_struct *dtr)
 {
 	asm ("sidt %0":"=m" (*dtr));
 }
 
-static fastcall unsigned long native_store_tr(void)
+static unsigned long native_store_tr(void)
 {
 	unsigned long tr;
 	asm ("str %0":"=r" (tr));
 	return tr;
 }
 
-static fastcall void native_load_tls(struct thread_struct *t, unsigned int cpu)
+static void native_load_tls(struct thread_struct *t, unsigned int cpu)
 {
 #define C(i) get_cpu_gdt_table(cpu)[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i]
 	C(0); C(1); C(2);
@@ -348,22 +348,22 @@ static inline void native_write_dt_entry
 	lp[1] = entry_high;
 }
 
-static fastcall void native_write_ldt_entry(void *dt, int entrynum, u32 low, u32 high)
+static void native_write_ldt_entry(void *dt, int entrynum, u32 low, u32 high)
 {
 	native_write_dt_entry(dt, entrynum, low, high);
 }
 
-static fastcall void native_write_gdt_entry(void *dt, int entrynum, u32 low, u32 high)
+static void native_write_gdt_entry(void *dt, int entrynum, u32 low, u32 high)
 {
 	native_write_dt_entry(dt, entrynum, low, high);
 }
 
-static fastcall void native_write_idt_entry(void *dt, int entrynum, u32 low, u32 high)
+static void native_write_idt_entry(void *dt, int entrynum, u32 low, u32 high)
 {
 	native_write_dt_entry(dt, entrynum, low, high);
 }
 
-static fastcall void native_load_esp0(struct tss_struct *tss,
+static void native_load_esp0(struct tss_struct *tss,
 				      struct thread_struct *thread)
 {
 	tss->esp0 = thread->esp0;
@@ -375,12 +375,12 @@ static fastcall void native_load_esp0(st
 	}
 }
 
-static fastcall void native_io_delay(void)
+static void native_io_delay(void)
 {
 	asm volatile("outb %al,$0x80");
 }
 
-static fastcall void native_flush_tlb(void)
+static void native_flush_tlb(void)
 {
 	__native_flush_tlb();
 }
@@ -389,49 +389,49 @@ static fastcall void native_flush_tlb(vo
  * Global pages have to be flushed a bit differently. Not a real
  * performance problem because this does not happen often.
  */
-static fastcall void native_flush_tlb_global(void)
+static void native_flush_tlb_global(void)
 {
 	__native_flush_tlb_global();
 }
 
-static fastcall void native_flush_tlb_single(u32 addr)
+static void native_flush_tlb_single(u32 addr)
 {
 	__native_flush_tlb_single(addr);
 }
 
 #ifndef CONFIG_X86_PAE
-static fastcall void native_set_pte(pte_t *ptep, pte_t pteval)
+static void native_set_pte(pte_t *ptep, pte_t pteval)
 {
 	*ptep = pteval;
 }
 
-static fastcall void native_set_pte_at(struct mm_struct *mm, u32 addr, pte_t *ptep, pte_t pteval)
+static void native_set_pte_at(struct mm_struct *mm, u32 addr, pte_t *ptep, pte_t pteval)
 {
 	*ptep = pteval;
 }
 
-static fastcall void native_set_pmd(pmd_t *pmdp, pmd_t pmdval)
+static void native_set_pmd(pmd_t *pmdp, pmd_t pmdval)
 {
 	*pmdp = pmdval;
 }
 
 #else /* CONFIG_X86_PAE */
 
-static fastcall void native_set_pte(pte_t *ptep, pte_t pte)
+static void native_set_pte(pte_t *ptep, pte_t pte)
 {
 	ptep->pte_high = pte.pte_high;
 	smp_wmb();
 	ptep->pte_low = pte.pte_low;
 }
 
-static fastcall void native_set_pte_at(struct mm_struct *mm, u32 addr, pte_t *ptep, pte_t pte)
+static void native_set_pte_at(struct mm_struct *mm, u32 addr, pte_t *ptep, pte_t pte)
 {
 	ptep->pte_high = pte.pte_high;
 	smp_wmb();
 	ptep->pte_low = pte.pte_low;
 }
 
-static fastcall void native_set_pte_present(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte)
+static void native_set_pte_present(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte)
 {
 	ptep->pte_low = 0;
 	smp_wmb();
@@ -440,29 +440,29 @@ static fastcall void native_set_pte_pres
 	ptep->pte_low = pte.pte_low;
 }
 
-static fastcall void native_set_pte_atomic(pte_t *ptep, pte_t pteval)
+static void native_set_pte_atomic(pte_t *ptep, pte_t pteval)
 {
 	set_64bit((unsigned long long *)ptep,pte_val(pteval));
 }
 
-static fastcall void native_set_pmd(pmd_t *pmdp, pmd_t pmdval)
+static void native_set_pmd(pmd_t *pmdp, pmd_t pmdval)
 {
 	set_64bit((unsigned long long *)pmdp,pmd_val(pmdval));
 }
 
-static fastcall void native_set_pud(pud_t *pudp, pud_t pudval)
+static void native_set_pud(pud_t *pudp, pud_t pudval)
 {
 	*pudp = pudval;
 }
 
-static fastcall void native_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+static void native_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
 	ptep->pte_low = 0;
 	smp_wmb();
 	ptep->pte_high = 0;
 }
 
-static fastcall void native_pmd_clear(pmd_t *pmd)
+static void native_pmd_clear(pmd_t *pmd)
 {
 	u32 *tmp = (u32 *)pmd;
 	*tmp = 0;
@@ -472,8 +472,8 @@ static fastcall void native_pmd_clear(pm
 #endif /* CONFIG_X86_PAE */
 
 /* These are in entry.S */
-extern fastcall void native_iret(void);
-extern fastcall void native_irq_enable_sysexit(void);
+extern void native_iret(void);
+extern void native_irq_enable_sysexit(void);
 
 static int __init print_banner(void)
 {
Index: linux/include/asm-i386/paravirt.h
===================================================================
--- linux.orig/include/asm-i386/paravirt.h
+++ linux/include/asm-i386/paravirt.h
@@ -59,102 +59,102 @@ struct paravirt_ops
 	   convention.  This makes it easier to implement inline
 	   assembler replacements. */
 
-	void (fastcall *cpuid)(unsigned int *eax, unsigned int *ebx,
+	void (*cpuid)(unsigned int *eax, unsigned int *ebx,
 		      unsigned int *ecx, unsigned int *edx);
 
-	unsigned long (fastcall *get_debugreg)(int regno);
-	void (fastcall *set_debugreg)(int regno, unsigned long value);
+	unsigned long (*get_debugreg)(int regno);
+	void (*set_debugreg)(int regno, unsigned long value);
 
-	void (fastcall *clts)(void);
+	void (*clts)(void);
 
-	unsigned long (fastcall *read_cr0)(void);
-	void (fastcall *write_cr0)(unsigned long);
+	unsigned long (*read_cr0)(void);
+	void (*write_cr0)(unsigned long);
 
-	unsigned long (fastcall *read_cr2)(void);
-	void (fastcall *write_cr2)(unsigned long);
-
-	unsigned long (fastcall *read_cr3)(void);
-	void (fastcall *write_cr3)(unsigned long);
-
-	unsigned long (fastcall *read_cr4_safe)(void);
-	unsigned long (fastcall *read_cr4)(void);
-	void (fastcall *write_cr4)(unsigned long);
-
-	unsigned long (fastcall *save_fl)(void);
-	void (fastcall *restore_fl)(unsigned long);
-	void (fastcall *irq_disable)(void);
-	void (fastcall *irq_enable)(void);
-	void (fastcall *safe_halt)(void);
-	void (fastcall *halt)(void);
-	void (fastcall *wbinvd)(void);
+	unsigned long (*read_cr2)(void);
+	void (*write_cr2)(unsigned long);
+
+	unsigned long (*read_cr3)(void);
+	void (*write_cr3)(unsigned long);
+
+	unsigned long (*read_cr4_safe)(void);
+	unsigned long (*read_cr4)(void);
+	void (*write_cr4)(unsigned long);
+
+	unsigned long (*save_fl)(void);
+	void (*restore_fl)(unsigned long);
+	void (*irq_disable)(void);
+	void (*irq_enable)(void);
+	void (*safe_halt)(void);
+	void (*halt)(void);
+	void (*wbinvd)(void);
 
 	/* err = 0/-EFAULT.  wrmsr returns 0/-EFAULT. */
-	u64 (fastcall *read_msr)(unsigned int msr, int *err);
-	int (fastcall *write_msr)(unsigned int msr, u64 val);
+	u64 (*read_msr)(unsigned int msr, int *err);
+	int (*write_msr)(unsigned int msr, u64 val);
 
-	u64 (fastcall *read_tsc)(void);
-	u64 (fastcall *read_pmc)(void);
+	u64 (*read_tsc)(void);
+	u64 (*read_pmc)(void);
 
-	void (fastcall *load_tr_desc)(void);
-	void (fastcall *load_gdt)(const struct Xgt_desc_struct *);
-	void (fastcall *load_idt)(const struct Xgt_desc_struct *);
-	void (fastcall *store_gdt)(struct Xgt_desc_struct *);
-	void (fastcall *store_idt)(struct Xgt_desc_struct *);
-	void (fastcall *set_ldt)(const void *desc, unsigned entries);
-	unsigned long (fastcall *store_tr)(void);
-	void (fastcall *load_tls)(struct thread_struct *t, unsigned int cpu);
-	void (fastcall *write_ldt_entry)(void *dt, int entrynum,
+	void (*load_tr_desc)(void);
+	void (*load_gdt)(const struct Xgt_desc_struct *);
+	void (*load_idt)(const struct Xgt_desc_struct *);
+	void (*store_gdt)(struct Xgt_desc_struct *);
+	void (*store_idt)(struct Xgt_desc_struct *);
+	void (*set_ldt)(const void *desc, unsigned entries);
+	unsigned long (*store_tr)(void);
+	void (*load_tls)(struct thread_struct *t, unsigned int cpu);
+	void (*write_ldt_entry)(void *dt, int entrynum,
 					 u32 low, u32 high);
-	void (fastcall *write_gdt_entry)(void *dt, int entrynum,
+	void (*write_gdt_entry)(void *dt, int entrynum,
 					 u32 low, u32 high);
-	void (fastcall *write_idt_entry)(void *dt, int entrynum,
+	void (*write_idt_entry)(void *dt, int entrynum,
 					 u32 low, u32 high);
-	void (fastcall *load_esp0)(struct tss_struct *tss,
+	void (*load_esp0)(struct tss_struct *tss,
 				   struct thread_struct *thread);
 
-	void (fastcall *set_iopl_mask)(unsigned mask);
+	void (*set_iopl_mask)(unsigned mask);
 
-	void (fastcall *io_delay)(void);
+	void (*io_delay)(void);
 	void (*const_udelay)(unsigned long loops);
 
 #ifdef CONFIG_X86_LOCAL_APIC
-	void (fastcall *apic_write)(unsigned long reg, unsigned long v);
-	void (fastcall *apic_write_atomic)(unsigned long reg, unsigned long v);
-	unsigned long (fastcall *apic_read)(unsigned long reg);
+	void (*apic_write)(unsigned long reg, unsigned long v);
+	void (*apic_write_atomic)(unsigned long reg, unsigned long v);
+	unsigned long (*apic_read)(unsigned long reg);
 	void (*setup_boot_clock)(void);
 	void (*setup_secondary_clock)(void);
 #endif
 
-	void (fastcall *flush_tlb_user)(void);
-	void (fastcall *flush_tlb_kernel)(void);
-	void (fastcall *flush_tlb_single)(u32 addr);
-
-	void (fastcall *alloc_pt)(u32 pfn);
-	void (fastcall *alloc_pd)(u32 pfn);
-	void (fastcall *alloc_pd_clone)(u32 pfn, u32 clonepfn, u32 start, u32 count);
-	void (fastcall *release_pt)(u32 pfn);
-	void (fastcall *release_pd)(u32 pfn);
-
-	void (fastcall *set_pte)(pte_t *ptep, pte_t pteval);
-	void (fastcall *set_pte_at)(struct mm_struct *mm, u32 addr, pte_t *ptep, pte_t pteval);
-	void (fastcall *set_pmd)(pmd_t *pmdp, pmd_t pmdval);
-	void (fastcall *pte_update)(struct mm_struct *mm, u32 addr, pte_t *ptep);
-	void (fastcall *pte_update_defer)(struct mm_struct *mm, u32 addr, pte_t *ptep);
+	void (*flush_tlb_user)(void);
+	void (*flush_tlb_kernel)(void);
+	void (*flush_tlb_single)(u32 addr);
+
+	void (*alloc_pt)(u32 pfn);
+	void (*alloc_pd)(u32 pfn);
+	void (*alloc_pd_clone)(u32 pfn, u32 clonepfn, u32 start, u32 count);
+	void (*release_pt)(u32 pfn);
+	void (*release_pd)(u32 pfn);
+
+	void (*set_pte)(pte_t *ptep, pte_t pteval);
+	void (*set_pte_at)(struct mm_struct *mm, u32 addr, pte_t *ptep, pte_t pteval);
+	void (*set_pmd)(pmd_t *pmdp, pmd_t pmdval);
+	void (*pte_update)(struct mm_struct *mm, u32 addr, pte_t *ptep);
+	void (*pte_update_defer)(struct mm_struct *mm, u32 addr, pte_t *ptep);
 #ifdef CONFIG_X86_PAE
-	void (fastcall *set_pte_atomic)(pte_t *ptep, pte_t pteval);
-	void (fastcall *set_pte_present)(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte);
-	void (fastcall *set_pud)(pud_t *pudp, pud_t pudval);
-	void (fastcall *pte_clear)(struct mm_struct *mm, unsigned long addr, pte_t *ptep);
-	void (fastcall *pmd_clear)(pmd_t *pmdp);
+	void (*set_pte_atomic)(pte_t *ptep, pte_t pteval);
+	void (*set_pte_present)(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte);
+	void (*set_pud)(pud_t *pudp, pud_t pudval);
+	void (*pte_clear)(struct mm_struct *mm, unsigned long addr, pte_t *ptep);
+	void (*pmd_clear)(pmd_t *pmdp);
 #endif
 
-	void (fastcall *set_lazy_mode)(int mode);
+	void (*set_lazy_mode)(int mode);
 
 	/* These two are jmp to, not actually called. */
-	void (fastcall *irq_enable_sysexit)(void);
-	void (fastcall *iret)(void);
+	void (*irq_enable_sysexit)(void);
+	void (*iret)(void);
 
-	void (fastcall *startup_ipi_hook)(int phys_apicid, unsigned long start_eip, unsigned long start_esp);
+	void (*startup_ipi_hook)(int phys_apicid, unsigned long start_eip, unsigned long start_esp);
 };
 
 /* Mark a paravirt probe function. */

[PATCH x86 for review III] [17/29] x86: Add new CPUID bits for AMD Family 10 CPUs in /proc/cpuinfo

[Andi Kleen]

Just various new acronyms. The new popcnt bit is in the middle
of Intel space. This looks a little weird, but I've been assured
it's ok.

Also I fixed RDTSCP for i386 which was at the wrong place.

For i386 and x86-64.

Signed-off-by: Andi Kleen <ak@suse.de>

---
 arch/i386/kernel/cpu/proc.c |   14 +++++++++-----
 arch/x86_64/kernel/setup.c  |   14 ++++++++++----
 2 files changed, 19 insertions(+), 9 deletions(-)

Index: linux/arch/x86_64/kernel/setup.c
===================================================================
--- linux.orig/arch/x86_64/kernel/setup.c
+++ linux/arch/x86_64/kernel/setup.c
@@ -942,7 +942,8 @@ static int show_cpuinfo(struct seq_file 
 		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 		NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL,
 		NULL, NULL, NULL, NULL, "nx", NULL, "mmxext", NULL,
-		NULL, "fxsr_opt", NULL, "rdtscp", NULL, "lm", "3dnowext", "3dnow",
+		NULL, "fxsr_opt", "pdpe1gb", "rdtscp", NULL, "lm",
+		"3dnowext", "3dnow",
 
 		/* Transmeta-defined */
 		"recovery", "longrun", NULL, "lrti", NULL, NULL, NULL, NULL,
@@ -960,7 +961,7 @@ static int show_cpuinfo(struct seq_file 
 		/* Intel-defined (#2) */
 		"pni", NULL, NULL, "monitor", "ds_cpl", "vmx", "smx", "est",
 		"tm2", "ssse3", "cid", NULL, NULL, "cx16", "xtpr", NULL,
-		NULL, NULL, "dca", NULL, NULL, NULL, NULL, NULL,
+		NULL, NULL, "dca", NULL, NULL, NULL, NULL, "popcnt",
 		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 
 		/* VIA/Cyrix/Centaur-defined */
@@ -970,8 +971,10 @@ static int show_cpuinfo(struct seq_file 
 		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 
 		/* AMD-defined (#2) */
-		"lahf_lm", "cmp_legacy", "svm", NULL, "cr8_legacy", NULL, NULL, NULL,
-		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+		"lahf_lm", "cmp_legacy", "svm", "extapic", "cr8_legacy",
+		"altmovcr8", "abm", "sse4a",
+		"misalignsse", "3dnowprefetch",
+		"osvw", "ibs", NULL, NULL, NULL, NULL,
 		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 	};
@@ -982,6 +985,9 @@ static int show_cpuinfo(struct seq_file 
 		"ttp",  /* thermal trip */
 		"tm",
 		"stc",
+		"100mhzsteps",
+		"hwpstate",
+		NULL,	/* tsc invariant mapped to constant_tsc */
 		NULL,
 		/* nothing */	/* constant_tsc - moved to flags */
 	};
Index: linux/arch/i386/kernel/cpu/proc.c
===================================================================
--- linux.orig/arch/i386/kernel/cpu/proc.c
+++ linux/arch/i386/kernel/cpu/proc.c
@@ -29,7 +29,7 @@ static int show_cpuinfo(struct seq_file 
 		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 		NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL,
 		NULL, NULL, NULL, "mp", "nx", NULL, "mmxext", NULL,
-		NULL, "fxsr_opt", "rdtscp", NULL, NULL, "lm", "3dnowext", "3dnow",
+		NULL, "fxsr_opt", "pdpe1gb", "rdtscp", NULL, "lm", "3dnowext", "3dnow",
 
 		/* Transmeta-defined */
 		"recovery", "longrun", NULL, "lrti", NULL, NULL, NULL, NULL,
@@ -47,7 +47,7 @@ static int show_cpuinfo(struct seq_file 
 		/* Intel-defined (#2) */
 		"pni", NULL, NULL, "monitor", "ds_cpl", "vmx", "smx", "est",
 		"tm2", "ssse3", "cid", NULL, NULL, "cx16", "xtpr", NULL,
-		NULL, NULL, "dca", NULL, NULL, NULL, NULL, NULL,
+		NULL, NULL, "dca", NULL, NULL, NULL, NULL, "popcnt",
 		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 
 		/* VIA/Cyrix/Centaur-defined */
@@ -57,8 +57,9 @@ static int show_cpuinfo(struct seq_file 
 		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 
 		/* AMD-defined (#2) */
-		"lahf_lm", "cmp_legacy", "svm", NULL, "cr8legacy", NULL, NULL, NULL,
-		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+		"lahf_lm", "cmp_legacy", "svm", "extapic", "cr8legacy", "abm",
+		"sse4a", "misalignsse",
+		"3dnowprefetch", "osvw", "ibs", NULL, NULL, NULL, NULL, NULL,
 		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 	};
@@ -69,8 +70,11 @@ static int show_cpuinfo(struct seq_file 
 		"ttp",  /* thermal trip */
 		"tm",
 		"stc",
+		"100mhzsteps",
+		"hwpstate",
 		NULL,
-		/* nothing */	/* constant_tsc - moved to flags */
+		NULL,	/* constant_tsc - moved to flags */
+		/* nothing */
 	};
 	struct cpuinfo_x86 *c = v;
 	int i, n = c - cpu_data;

[PATCH x86 for review III] [18/29] i386: Add L3 cache support to AMD CPUID4 emulation

[Andi Kleen]

With that an L3 cache is correctly reported in the cache information in /sys

Signed-off-by: Andi Kleen <ak@suse.de>

---
 arch/i386/kernel/cpu/intel_cacheinfo.c |   65 +++++++++++++++++++++++++--------
 1 file changed, 50 insertions(+), 15 deletions(-)

Index: linux/arch/i386/kernel/cpu/intel_cacheinfo.c
===================================================================
--- linux.orig/arch/i386/kernel/cpu/intel_cacheinfo.c
+++ linux/arch/i386/kernel/cpu/intel_cacheinfo.c
@@ -135,7 +135,7 @@ unsigned short			num_cache_leaves;
 
 /* AMD doesn't have CPUID4. Emulate it here to report the same
    information to the user.  This makes some assumptions about the machine:
-   No L3, L2 not shared, no SMT etc. that is currently true on AMD CPUs.
+   L2 not shared, no SMT etc. that is currently true on AMD CPUs.
 
    In theory the TLBs could be reported as fake type (they are in "dummy").
    Maybe later */
@@ -159,11 +159,23 @@ union l2_cache {
 	unsigned val;
 };
 
+union l3_cache {
+	struct {
+		unsigned line_size : 8;
+		unsigned lines_per_tag : 4;
+		unsigned assoc : 4;
+		unsigned res : 2;
+		unsigned size_encoded : 14;
+	};
+	unsigned val;
+};
+
 static const unsigned short assocs[] = {
 	[1] = 1, [2] = 2, [4] = 4, [6] = 8,
 	[8] = 16,
 	[0xf] = 0xffff // ??
-	};
+};
+
 static const unsigned char levels[] = { 1, 1, 2 };
 static const unsigned char types[] = { 1, 2, 3 };
 
@@ -175,37 +187,60 @@ static void __cpuinit amd_cpuid4(int lea
 	unsigned line_size, lines_per_tag, assoc, size_in_kb;
 	union l1_cache l1i, l1d;
 	union l2_cache l2;
+	union l3_cache l3;
+	union l1_cache *l1 = &l1d;
 
 	eax->full = 0;
 	ebx->full = 0;
 	ecx->full = 0;
 
 	cpuid(0x80000005, &dummy, &dummy, &l1d.val, &l1i.val);
-	cpuid(0x80000006, &dummy, &dummy, &l2.val, &dummy);
-
-	if (leaf > 2 || !l1d.val || !l1i.val || !l2.val)
-		return;
+	cpuid(0x80000006, &dummy, &dummy, &l2.val, &l3.val);
 
-	eax->split.is_self_initializing = 1;
-	eax->split.type = types[leaf];
-	eax->split.level = levels[leaf];
-	eax->split.num_threads_sharing = 0;
-	eax->split.num_cores_on_die = current_cpu_data.x86_max_cores - 1;
-
-	if (leaf <= 1) {
-		union l1_cache *l1 = leaf == 0 ? &l1d : &l1i;
+	switch (leaf) {
+	case 1:
+		l1 = &l1i;
+	case 0:
+		if (!l1->val)
+			return;
 		assoc = l1->assoc;
 		line_size = l1->line_size;
 		lines_per_tag = l1->lines_per_tag;
 		size_in_kb = l1->size_in_kb;
-	} else {
+		break;
+	case 2:
+		if (!l2.val)
+			return;
 		assoc = l2.assoc;
 		line_size = l2.line_size;
 		lines_per_tag = l2.lines_per_tag;
 		/* cpu_data has errata corrections for K7 applied */
 		size_in_kb = current_cpu_data.x86_cache_size;
+		break;
+	case 3:
+		if (!l3.val)
+			return;
+		assoc = l3.assoc;
+		line_size = l3.line_size;
+		lines_per_tag = l3.lines_per_tag;
+		switch (l3.size_encoded) {
+		case 4:  size_in_kb = 2 * 1024; break;
+		case 8:  size_in_kb = 4 * 1024; break;
+		case 12: size_in_kb = 6 * 1024; break;
+		default: size_in_kb = 0; break;
+		}
+		break;
+	default:
+		return;
 	}
 
+	eax->split.is_self_initializing = 1;
+	eax->split.type = types[leaf];
+	eax->split.level = levels[leaf];
+	eax->split.num_threads_sharing = 0;
+	eax->split.num_cores_on_die = current_cpu_data.x86_max_cores - 1;
+
+
 	if (assoc == 0xf)
 		eax->split.is_fully_associative = 1;
 	ebx->split.coherency_line_size = line_size - 1;

[PATCH x86 for review III] [19/29] x86: Enable NMI watchdog for AMD Family 0x10 CPUs

[Andi Kleen]

For i386/x86-64.

Straight forward -- just reuse the Family 0xf code.

Signed-off-by: Andi Kleen <ak@suse.de>

---
 arch/i386/kernel/nmi.c   |    6 ++++--
 arch/x86_64/kernel/nmi.c |    2 +-
 2 files changed, 5 insertions(+), 3 deletions(-)

Index: linux/arch/i386/kernel/nmi.c
===================================================================
--- linux.orig/arch/i386/kernel/nmi.c
+++ linux/arch/i386/kernel/nmi.c
@@ -185,7 +185,8 @@ static __cpuinit inline int nmi_known_cp
 {
 	switch (boot_cpu_data.x86_vendor) {
 	case X86_VENDOR_AMD:
-		return ((boot_cpu_data.x86 == 15) || (boot_cpu_data.x86 == 6));
+		return ((boot_cpu_data.x86 == 15) || (boot_cpu_data.x86 == 6)
+			|| (boot_cpu_data.x86 == 16));
 	case X86_VENDOR_INTEL:
 		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
 			return 1;
@@ -817,7 +818,8 @@ void setup_apic_nmi_watchdog (void *unus
 	if (nmi_watchdog == NMI_LOCAL_APIC) {
 		switch (boot_cpu_data.x86_vendor) {
 		case X86_VENDOR_AMD:
-			if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15)
+			if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15 &&
+				boot_cpu_data.x86 != 16)
 				return;
 			if (!setup_k7_watchdog())
 				return;
Index: linux/arch/x86_64/kernel/nmi.c
===================================================================
--- linux.orig/arch/x86_64/kernel/nmi.c
+++ linux/arch/x86_64/kernel/nmi.c
@@ -172,7 +172,7 @@ static __cpuinit inline int nmi_known_cp
 {
 	switch (boot_cpu_data.x86_vendor) {
 	case X86_VENDOR_AMD:
-		return boot_cpu_data.x86 == 15;
+		return boot_cpu_data.x86 == 15 || boot_cpu_data.x86 == 16;
 	case X86_VENDOR_INTEL:
 		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
 			return 1;

[PATCH x86 for review III] [20/29] i386: Fix warning in microcode.c

[Andi Kleen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 622 bytes --]


Fix bogus gcc warning

linux/arch/i386/kernel/microcode.c:387: warning: ‘new_mc’ may be used uninitialized in this function


Signed-off-by: Andi Kleen <ak@suse.de>

---
 arch/i386/kernel/microcode.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Index: linux/arch/i386/kernel/microcode.c
===================================================================
--- linux.orig/arch/i386/kernel/microcode.c
+++ linux/arch/i386/kernel/microcode.c
@@ -384,7 +384,7 @@ static int do_microcode_update (void)
 {
 	long cursor = 0;
 	int error = 0;
-	void *new_mc;
+	void *new_mc = NULL;
 	int cpu;
 	cpumask_t old;
 

[PATCH x86 for review III] [21/29] i386: Fix warning in cpu initialization

[Andi Kleen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 796 bytes --]


Fix bogus warning

linux/arch/i386/kernel/cpu/transmeta.c:12: warning: ‘cpu_freq’ may be used uninitialized in this function

Signed-off-by: Andi Kleen <ak@suse.de>

---
 arch/i386/kernel/cpu/transmeta.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Index: linux/arch/i386/kernel/cpu/transmeta.c
===================================================================
--- linux.orig/arch/i386/kernel/cpu/transmeta.c
+++ linux/arch/i386/kernel/cpu/transmeta.c
@@ -9,7 +9,7 @@ static void __cpuinit init_transmeta(str
 {
 	unsigned int cap_mask, uk, max, dummy;
 	unsigned int cms_rev1, cms_rev2;
-	unsigned int cpu_rev, cpu_freq, cpu_flags, new_cpu_rev;
+	unsigned int cpu_rev, cpu_freq = 0, cpu_flags, new_cpu_rev;
 	char cpu_info[65];
 
 	get_model_name(c);	/* Same as AMD/Cyrix */

[PATCH x86 for review III] [23/29] x86: Don't require the vDSO for handling a.out signals

[Andi Kleen]

and in other strange binfmts. vDSO is not necessarily mapped there.

Signed-off-by: Andi Kleen <ak@suse.de>

---
 arch/i386/kernel/signal.c      |    6 +++++-
 arch/x86_64/ia32/ia32_signal.c |    7 ++++++-
 fs/binfmt_elf.c                |    3 ++-
 include/linux/binfmts.h        |    1 +
 4 files changed, 14 insertions(+), 3 deletions(-)

Index: linux/arch/i386/kernel/signal.c
===================================================================
--- linux.orig/arch/i386/kernel/signal.c
+++ linux/arch/i386/kernel/signal.c
@@ -21,6 +21,7 @@
 #include <linux/suspend.h>
 #include <linux/ptrace.h>
 #include <linux/elf.h>
+#include <linux/binfmts.h>
 #include <asm/processor.h>
 #include <asm/ucontext.h>
 #include <asm/uaccess.h>
@@ -349,7 +350,10 @@ static int setup_frame(int sig, struct k
 			goto give_sigsegv;
 	}
 
-	restorer = (void *)VDSO_SYM(&__kernel_sigreturn);
+	if (current->binfmt->hasvdso)
+		restorer = (void *)VDSO_SYM(&__kernel_sigreturn);
+	else
+		restorer = (void *)&frame->retcode;
 	if (ka->sa.sa_flags & SA_RESTORER)
 		restorer = ka->sa.sa_restorer;
 
Index: linux/arch/x86_64/ia32/ia32_signal.c
===================================================================
--- linux.orig/arch/x86_64/ia32/ia32_signal.c
+++ linux/arch/x86_64/ia32/ia32_signal.c
@@ -21,6 +21,7 @@
 #include <linux/stddef.h>
 #include <linux/personality.h>
 #include <linux/compat.h>
+#include <linux/binfmts.h>
 #include <asm/ucontext.h>
 #include <asm/uaccess.h>
 #include <asm/i387.h>
@@ -449,7 +450,11 @@ int ia32_setup_frame(int sig, struct k_s
 
 	/* Return stub is in 32bit vsyscall page */
 	{ 
-		void __user *restorer = VSYSCALL32_SIGRETURN; 
+		void __user *restorer;
+		if (current->binfmt->hasvdso)
+			restorer = VSYSCALL32_SIGRETURN;
+		else
+			restorer = (void *)&frame->retcode;
 		if (ka->sa.sa_flags & SA_RESTORER)
 			restorer = ka->sa.sa_restorer;       
 		err |= __put_user(ptr_to_compat(restorer), &frame->pretcode);
Index: linux/fs/binfmt_elf.c
===================================================================
--- linux.orig/fs/binfmt_elf.c
+++ linux/fs/binfmt_elf.c
@@ -76,7 +76,8 @@ static struct linux_binfmt elf_format = 
 		.load_binary	= load_elf_binary,
 		.load_shlib	= load_elf_library,
 		.core_dump	= elf_core_dump,
-		.min_coredump	= ELF_EXEC_PAGESIZE
+		.min_coredump	= ELF_EXEC_PAGESIZE,
+		.hasvdso	= 1
 };
 
 #define BAD_ADDR(x) ((unsigned long)(x) >= TASK_SIZE)
Index: linux/include/linux/binfmts.h
===================================================================
--- linux.orig/include/linux/binfmts.h
+++ linux/include/linux/binfmts.h
@@ -59,6 +59,7 @@ struct linux_binfmt {
 	int (*load_shlib)(struct file *);
 	int (*core_dump)(long signr, struct pt_regs * regs, struct file * file);
 	unsigned long min_coredump;	/* minimal dump size */
+	int hasvdso;
 };
 
 extern int register_binfmt(struct linux_binfmt *);

[PATCH x86 for review III] [24/29] x86_64: -mm merge plans for 2.6.21

[Andi Kleen]

From: Ralf Baechle <ralf@linux-mips.org>
On Sat, Feb 10, 2007 at 11:22:05AM +0100, Heiko Carstens wrote:

> Which remembers me that I think that MIPS is using the non-compat version
> of sys_epoll_pwait for compat syscalls. But maybe MIPS doesn't need a compat
> syscall for some reason. Dunno.

Which reminds me that x86_64 i386 compat doesn't wire up sys_epoll_pwait ;-)

  Ralf

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andi Kleen <ak@suse.de>

---
 arch/x86_64/ia32/ia32entry.S |    1 +
 1 file changed, 1 insertion(+)

Index: linux/arch/x86_64/ia32/ia32entry.S
===================================================================
--- linux.orig/arch/x86_64/ia32/ia32entry.S
+++ linux/arch/x86_64/ia32/ia32entry.S
@@ -730,4 +730,5 @@ ia32_sys_call_table:
 	.quad compat_sys_vmsplice
 	.quad compat_sys_move_pages
 	.quad sys_getcpu
+	.quad sys_epoll_pwait
 ia32_syscall_end:		

[PATCH x86 for review III] [25/29] i386: paravirt unhandled fallthrough

[Andi Kleen]

From: Rusty Russell <rusty@rustcorp.com.au>
The current code simply calls "start_kernel" directly if we're under a
hypervisor and no paravirt_ops backend wants us, because paravirt.c
registers that as a backend.

This was always a vain hope; start_kernel won't get far without setup.
It's also impossible for paravirt_ops backends which don't sit in the
arch/i386/kernel directory: they can't link before paravirt.o anyway.

Keep it simple: if we pass all the registered paravirt probes, BUG().

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>

---
 arch/i386/kernel/Makefile   |    2 --
 arch/i386/kernel/head.S     |    7 ++++++-
 arch/i386/kernel/paravirt.c |    3 ---
 3 files changed, 6 insertions(+), 6 deletions(-)

Index: linux/arch/i386/kernel/Makefile
===================================================================
--- linux.orig/arch/i386/kernel/Makefile
+++ linux/arch/i386/kernel/Makefile
@@ -41,8 +41,6 @@ obj-$(CONFIG_HPET_TIMER) 	+= hpet.o
 obj-$(CONFIG_K8_NB)		+= k8.o
 
 obj-$(CONFIG_VMI)		+= vmi.o vmitime.o
-
-# Make sure this is linked after any other paravirt_ops structs: see head.S
 obj-$(CONFIG_PARAVIRT)		+= paravirt.o
 
 EXTRA_AFLAGS   := -traditional
Index: linux/arch/i386/kernel/head.S
===================================================================
--- linux.orig/arch/i386/kernel/head.S
+++ linux/arch/i386/kernel/head.S
@@ -513,10 +513,11 @@ startup_paravirt:
 	pushl	%ecx
 	pushl	%eax
 
-	/* paravirt.o is last in link, and that probe fn never returns */
 	pushl	$__start_paravirtprobe
 1:
 	movl	0(%esp), %eax
+	cmpl	$__stop_paravirtprobe, %eax
+	je	unhandled_paravirt
 	pushl	(%eax)
 	movl	8(%esp), %eax
 	call	*(%esp)
@@ -528,6 +529,10 @@ startup_paravirt:
 
 	addl	$4, (%esp)
 	jmp	1b
+
+unhandled_paravirt:
+	/* Nothing wanted us: we're screwed. */
+	ud2
 #endif
 
 /*
Index: linux/arch/i386/kernel/paravirt.c
===================================================================
--- linux.orig/arch/i386/kernel/paravirt.c
+++ linux/arch/i386/kernel/paravirt.c
@@ -482,9 +482,6 @@ static int __init print_banner(void)
 }
 core_initcall(print_banner);
 
-/* We simply declare start_kernel to be the paravirt probe of last resort. */
-paravirt_probe(start_kernel);
-
 struct paravirt_ops paravirt_ops = {
 	.name = "bare hardware",
 	.paravirt_enabled = 0,

[PATCH x86 for review III] [26/29] i386: Move mce_disabled to asm/mce.h

[Andi Kleen]

From: Rusty Russell <rusty@rustcorp.com.au>
Allows external actors to disable mce.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>

===================================================================
---
 arch/i386/kernel/cpu/mcheck/mce.h |    2 +-
 include/asm-i386/mce.h            |    2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

Index: linux/arch/i386/kernel/cpu/mcheck/mce.h
===================================================================
--- linux.orig/arch/i386/kernel/cpu/mcheck/mce.h
+++ linux/arch/i386/kernel/cpu/mcheck/mce.h
@@ -1,4 +1,5 @@
 #include <linux/init.h>
+#include <asm/mce.h>
 
 void amd_mcheck_init(struct cpuinfo_x86 *c);
 void intel_p4_mcheck_init(struct cpuinfo_x86 *c);
@@ -9,6 +10,5 @@ void winchip_mcheck_init(struct cpuinfo_
 /* Call the installed machine check handler for this CPU setup. */
 extern fastcall void (*machine_check_vector)(struct pt_regs *, long error_code);
 
-extern int mce_disabled;
 extern int nr_mce_banks;
 
Index: linux/include/asm-i386/mce.h
===================================================================
--- linux.orig/include/asm-i386/mce.h
+++ linux/include/asm-i386/mce.h
@@ -3,3 +3,5 @@ extern void mcheck_init(struct cpuinfo_x
 #else
 #define mcheck_init(c) do {} while(0)
 #endif
+
+extern int mce_disabled;

[PATCH x86 for review III] [27/29] i386: Rename cpu_gdt_descr and remove extern declaration from smpboot.c

[Andi Kleen]

From: Rusty Russell <rusty@rustcorp.com.au>
When I implemented the DECLARE_PER_CPU(var) macros, I was careful that
people couldn't use "var" in a non-percpu context, by prepending
percpu__.  I never considered that this would allow them to overload
the same name for a per-cpu and a non-percpu variable.

It is only one of many horrors in the i386 boot code, but let's rename
the non-perpcu cpu_gdt_descr to early_gdt_descr (not boot_gdt_descr,
that's something else...)

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>

===================================================================
---
 arch/i386/kernel/head.S    |    6 +++---
 arch/i386/kernel/smpboot.c |    1 -
 include/asm-i386/desc.h    |    2 +-
 3 files changed, 4 insertions(+), 5 deletions(-)

Index: linux/arch/i386/kernel/head.S
===================================================================
--- linux.orig/arch/i386/kernel/head.S
+++ linux/arch/i386/kernel/head.S
@@ -319,7 +319,7 @@ is386:	movl $2,%ecx		# set MP
 
 	call check_x87
 	call setup_pda
-	lgdt cpu_gdt_descr
+	lgdt early_gdt_descr
 	lidt idt_descr
 	ljmp $(__KERNEL_CS),$1f
 1:	movl $(__KERNEL_DS),%eax	# reload all the segment registers
@@ -375,7 +375,7 @@ ENTRY(setup_pda)
 	movl start_pda, %eax
 
 	/* slot the PDA address into the GDT */
-	mov cpu_gdt_descr+2, %ecx
+	mov early_gdt_descr+2, %ecx
 	mov %ax, (__KERNEL_PDA+0+2)(%ecx)		/* base & 0x0000ffff */
 	shr $16, %eax
 	mov %al, (__KERNEL_PDA+4+0)(%ecx)		/* base & 0x00ff0000 */
@@ -597,7 +597,7 @@ idt_descr:
 
 # boot GDT descriptor (later on used by CPU#0):
 	.word 0				# 32 bit align gdt_desc.address
-ENTRY(cpu_gdt_descr)
+ENTRY(early_gdt_descr)
 	.word GDT_ENTRIES*8-1
 	.long cpu_gdt_table
 
Index: linux/arch/i386/kernel/smpboot.c
===================================================================
--- linux.orig/arch/i386/kernel/smpboot.c
+++ linux/arch/i386/kernel/smpboot.c
@@ -622,7 +622,6 @@ extern struct {
 	unsigned short ss;
 } stack_start;
 extern struct i386_pda *start_pda;
-extern struct Xgt_desc_struct cpu_gdt_descr;
 
 #ifdef CONFIG_NUMA
 
Index: linux/include/asm-i386/desc.h
===================================================================
--- linux.orig/include/asm-i386/desc.h
+++ linux/include/asm-i386/desc.h
@@ -22,7 +22,7 @@ struct Xgt_desc_struct {
 
 extern struct Xgt_desc_struct idt_descr;
 DECLARE_PER_CPU(struct Xgt_desc_struct, cpu_gdt_descr);
-
+extern struct Xgt_desc_struct early_gdt_descr;
 
 static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
 {

[PATCH x86 for review III] [28/29] i386: Remove extern declaration from mm/discontig.c, put in header.

[Andi Kleen]

From: Rusty Russell <rusty@rustcorp.com.au>
Extern declarations belong in headers.  Times, they are a'changin.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>

===================================================================
---
 arch/i386/mm/discontig.c |    1 -
 include/asm-i386/setup.h |    2 ++
 2 files changed, 2 insertions(+), 1 deletion(-)

Index: linux/arch/i386/mm/discontig.c
===================================================================
--- linux.orig/arch/i386/mm/discontig.c
+++ linux/arch/i386/mm/discontig.c
@@ -101,7 +101,6 @@ extern void find_max_pfn(void);
 extern void add_one_highpage_init(struct page *, int, int);
 
 extern struct e820map e820;
-extern unsigned long init_pg_tables_end;
 extern unsigned long highend_pfn, highstart_pfn;
 extern unsigned long max_low_pfn;
 extern unsigned long totalram_pages;
Index: linux/include/asm-i386/setup.h
===================================================================
--- linux.orig/include/asm-i386/setup.h
+++ linux/include/asm-i386/setup.h
@@ -77,6 +77,8 @@ int __init sanitize_e820_map(struct e820
 void __init add_memory_region(unsigned long long start,
 			      unsigned long long size, int type);
 
+extern unsigned long init_pg_tables_end;
+
 #endif /* __ASSEMBLY__ */
 
 #endif  /*  __KERNEL__  */

[PATCH x86 for review III] [29/29] x86: Unify pcspeaker platform device code between i386/x86-64

[Andi Kleen]

Trivial cleanup.

Only change is that it is always compiled in now on x86-64 like on i386.


Signed-off-by: Andi Kleen <ak@suse.de>

---
 arch/i386/kernel/Makefile    |    1 +
 arch/i386/kernel/pcspeaker.c |   20 ++++++++++++++++++++
 arch/i386/kernel/setup.c     |   26 --------------------------
 arch/x86_64/kernel/Makefile  |    2 ++
 arch/x86_64/kernel/setup.c   |   20 --------------------
 5 files changed, 23 insertions(+), 46 deletions(-)

Index: linux/arch/i386/kernel/Makefile
===================================================================
--- linux.orig/arch/i386/kernel/Makefile
+++ linux/arch/i386/kernel/Makefile
@@ -42,6 +42,7 @@ obj-$(CONFIG_K8_NB)		+= k8.o
 
 obj-$(CONFIG_VMI)		+= vmi.o vmitime.o
 obj-$(CONFIG_PARAVIRT)		+= paravirt.o
+obj-y				+= pcspeaker.o
 
 EXTRA_AFLAGS   := -traditional
 
Index: linux/arch/i386/kernel/pcspeaker.c
===================================================================
--- /dev/null
+++ linux/arch/i386/kernel/pcspeaker.c
@@ -0,0 +1,20 @@
+#include <linux/platform_device.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+
+static __init int add_pcspkr(void)
+{
+	struct platform_device *pd;
+	int ret;
+
+	pd = platform_device_alloc("pcspkr", -1);
+	if (!pd)
+		return -ENOMEM;
+
+	ret = platform_device_add(pd);
+	if (ret)
+		platform_device_put(pd);
+
+	return ret;
+}
+device_initcall(add_pcspkr);
Index: linux/arch/i386/kernel/setup.c
===================================================================
--- linux.orig/arch/i386/kernel/setup.c
+++ linux/arch/i386/kernel/setup.c
@@ -33,7 +33,6 @@
 #include <linux/initrd.h>
 #include <linux/bootmem.h>
 #include <linux/seq_file.h>
-#include <linux/platform_device.h>
 #include <linux/console.h>
 #include <linux/mca.h>
 #include <linux/root_dev.h>
@@ -660,28 +659,3 @@ void __init setup_arch(char **cmdline_p)
 #endif
 	tsc_init();
 }
-
-static __init int add_pcspkr(void)
-{
-	struct platform_device *pd;
-	int ret;
-
-	pd = platform_device_alloc("pcspkr", -1);
-	if (!pd)
-		return -ENOMEM;
-
-	ret = platform_device_add(pd);
-	if (ret)
-		platform_device_put(pd);
-
-	return ret;
-}
-device_initcall(add_pcspkr);
-
-/*
- * Local Variables:
- * mode:c
- * c-file-style:"k&r"
- * c-basic-offset:8
- * End:
- */
Index: linux/arch/x86_64/kernel/Makefile
===================================================================
--- linux.orig/arch/x86_64/kernel/Makefile
+++ linux/arch/x86_64/kernel/Makefile
@@ -42,6 +42,7 @@ obj-$(CONFIG_PCI)		+= early-quirks.o
 
 obj-y				+= topology.o
 obj-y				+= intel_cacheinfo.o
+obj-y				+= pcspeaker.o
 
 CFLAGS_vsyscall.o		:= $(PROFILING) -g0
 
@@ -55,3 +56,4 @@ quirks-y			+= ../../i386/kernel/quirks.o
 i8237-y				+= ../../i386/kernel/i8237.o
 msr-$(subst m,y,$(CONFIG_X86_MSR))  += ../../i386/kernel/msr.o
 alternative-y			+= ../../i386/kernel/alternative.o
+pcspeaker-y			+= ../../i386/kernel/pcspeaker.o
Index: linux/arch/x86_64/kernel/setup.c
===================================================================
--- linux.orig/arch/x86_64/kernel/setup.c
+++ linux/arch/x86_64/kernel/setup.c
@@ -1104,23 +1104,3 @@ struct seq_operations cpuinfo_op = {
 	.stop =	c_stop,
 	.show =	show_cpuinfo,
 };
-
-#if defined(CONFIG_INPUT_PCSPKR) || defined(CONFIG_INPUT_PCSPKR_MODULE)
-#include <linux/platform_device.h>
-static __init int add_pcspkr(void)
-{
-	struct platform_device *pd;
-	int ret;
-
-	pd = platform_device_alloc("pcspkr", -1);
-	if (!pd)
-		return -ENOMEM;
-
-	ret = platform_device_add(pd);
-	if (ret)
-		platform_device_put(pd);
-
-	return ret;
-}
-device_initcall(add_pcspkr);
-#endif

Re: [PATCH x86 for review III] [10/29] i386: don't include bugs.h

[Jörn Engel]

On Mon, 12 February 2007 17:51:30 +0100, Andi Kleen wrote:
> 
> From: Andrew Morton <akpm@osdl.org>
> 
> That stupid non-inlined-static function in bugs.h causes:
> 
> include/asm/bugs.h:186: warning: 'check_bugs' defined but not used
> 
> But fortunately the include isn't needed.
> 
> Cc: Andi Kleen <ak@muc.de>
> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
> Signed-off-by: Andi Kleen <ak@suse.de>
> 
> ---
> 
>  arch/i386/kernel/alternative.c |    1 -
>  1 file changed, 1 deletion(-)
> 
> Index: linux/arch/i386/kernel/alternative.c
> ===================================================================
> --- linux.orig/arch/i386/kernel/alternative.c
> +++ linux/arch/i386/kernel/alternative.c
> @@ -4,7 +4,6 @@
>  #include <linux/list.h>
>  #include <asm/alternative.h>
>  #include <asm/sections.h>
> -#include <asm/bugs.h>
>  
>  static int no_replacement    = 0;
>  static int smp_alt_once      = 0;

Didn't your patchset also include a near-identical patch from Adrian
Bunk (with - and + exchanged)?

Jörn

-- 
Courage is not the absence of fear, but rather the judgement that
something else is more important than fear.
-- Ambrose Redmoon

Re: [PATCH x86 for review III] [10/29] i386: don't include bugs.h

[Andi Kleen]

> > Index: linux/arch/i386/kernel/alternative.c
> > ===================================================================
> > --- linux.orig/arch/i386/kernel/alternative.c
> > +++ linux/arch/i386/kernel/alternative.c
> > @@ -4,7 +4,6 @@
> >  #include <linux/list.h>
> >  #include <asm/alternative.h>
> >  #include <asm/sections.h>
> > -#include <asm/bugs.h>
> >  
> >  static int no_replacement    = 0;
> >  static int smp_alt_once      = 0;
> 
> Didn't your patchset also include a near-identical patch from Adrian
> Bunk (with - and + exchanged)?

Good catch. I will remove both patches.

-Andi

Re: [PATCH x86 for review III] [1/29] i386: avoid gcc extension

[Jochen Voß]

Hi,

On 12/02/07, Andi Kleen <ak@suse.de> wrote:
> setcc() in math-emu is written as a gcc extension statement expression
> macro that returns a value.  However, it's not used that way and it's not
> needed like that, so just make it a do-while non-extension macro so that we
> don't use an extension when it's not needed.

This description ...

> -#define setcc(cc) ({ \
> -  partial_status &= ~(SW_C0|SW_C1|SW_C2|SW_C3); \
> -  partial_status |= (cc) & (SW_C0|SW_C1|SW_C2|SW_C3); })
> +static inline void setcc(int cc)
> +{
> +       partial_status &= ~(SW_C0|SW_C1|SW_C2|SW_C3);
> +       partial_status |= (cc) & (SW_C0|SW_C1|SW_C2|SW_C3);
> +}

... seems to contradict the implementation.  No "do-while" here.

I hope this helps,
Jochen
-- 
http://seehuhn.de/

Re: [PATCH x86 for review III] [7/29] x86_64: update IO-APIC dest field to 8-bit for xAPIC

[Eric W. Biederman]

Andi Kleen <ak@suse.de> writes:

> From: Benjamin Romer <benjamin.romer@unisys.com>
>
> On the Unisys ES7000/ONE system, we encountered a problem where performing
> a kexec reboot or dump on any cell other than cell 0 causes the system
> timer to stop working, resulting in a hang during timer calibration in the
> new kernel.
>
> We traced the problem to one line of code in disable_IO_APIC(), which needs
> to restore the timer's IO-APIC configuration before rebooting.  The code is
> currently using the 4-bit physical destination field, rather than using the
> 8-bit logical destination field, and it cuts off the upper 4 bits of the
> timer's APIC ID.  If we change this to use the logical destination field,
> the timer works and we can kexec on the upper cells.  This was tested on
> two different cells (0 and 2) in an ES7000/ONE system.
>
> For reference, the relevant Intel xAPIC spec is kept at
> ftp://download.intel.com/design/chipsets/e8501/datashts/30962001.pdf,
> specifically on page 334.
>
> Signed-off-by: Benjamin M Romer <benjamin.romer@unisys.com>
> Signed-off-by: Andi Kleen <ak@suse.de>
> Cc: Andi Kleen <ak@suse.de>
> Cc: "Eric W. Biederman" <ebiederm@xmission.com>
> Cc: Vivek Goyal <vgoyal@in.ibm.com>
> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

Acked-by: Eric W. Biederman" <ebiederm@xmission.com>

Re: [PATCH x86 for review III] [1/29] i386: avoid gcc extension

[Randy Dunlap]

On Mon, 12 Feb 2007 19:43:36 +0000 Jochen Voß wrote:

> Hi,
> 
> On 12/02/07, Andi Kleen <ak@suse.de> wrote:
> > setcc() in math-emu is written as a gcc extension statement expression
> > macro that returns a value.  However, it's not used that way and it's not
> > needed like that, so just make it a do-while non-extension macro so that we
> > don't use an extension when it's not needed.
> 
> This description ...
> 
> ... seems to contradict the implementation.  No "do-while" here.
> 
> I hope this helps,

Yes, thanks.  First version was do-while, latter was inline.

Here's an updated patch/description.

---
From: Randy Dunlap <randy.dunlap@oracle.com>

setcc() in math-emu is written as a gcc extension statement expression
macro that returns a value.  However, it's not used that way and it's
not needed like that, so just make it an inline void so that
we don't use an extension when it's not needed.

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
---
 arch/i386/math-emu/status_w.h |    8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

--- linux-2620-rc2.orig/arch/i386/math-emu/status_w.h
+++ linux-2620-rc2/arch/i386/math-emu/status_w.h
@@ -48,9 +48,11 @@
 
 #define status_word() \
   ((partial_status & ~SW_Top & 0xffff) | ((top << SW_Top_Shift) & SW_Top))
-#define setcc(cc) ({ \
-  partial_status &= ~(SW_C0|SW_C1|SW_C2|SW_C3); \
-  partial_status |= (cc) & (SW_C0|SW_C1|SW_C2|SW_C3); })
+static inline void setcc(int cc)
+{
+	partial_status &= ~(SW_C0|SW_C1|SW_C2|SW_C3);
+	partial_status |= (cc) & (SW_C0|SW_C1|SW_C2|SW_C3);
+}
 
 #ifdef PECULIAR_486
    /* Default, this conveys no information, but an 80486 does it. */

Re: [PATCH x86 for review III] [12/29] x86_64: 32-bit ptrace mangles sixth system call argument

[Chuck Ebbert]

Andi Kleen wrote:
> From: Jeff Dike <jdike@addtoit.com>
> The patch below copies %r9 (where the sixth argument has been
> stashed) into the RBP slot of pt_regs before syscall_trace_enter is
> called.  This fixes ptrace.
> 
> To allow a successful return to userspace, the original value of rbp
> must be restored.  This is done by storing the current value of rbp
> into the RBP slot of pt_regs before the RESTORE_REST.
> 
> --- linux.orig/arch/x86_64/ia32/ia32entry.S
> +++ linux/arch/x86_64/ia32/ia32entry.S
> @@ -148,11 +148,23 @@ sysenter_do_call:	
>  sysenter_tracesys:
>  	CFI_RESTORE_STATE
>  	SAVE_REST
> +	/*
> +	 * We need the 6th system call argument to be in regs->rbp at
> +	 * this point so that ptrace will see it.  It's in r9 now, so copy
> +	 * it to the rbp slot now.
> +	 */
> +	movq	%r9, RBP(%rsp)
>  	CLEAR_RREGS
>  	movq	$-ENOSYS,RAX(%rsp)	/* really needed? */
>  	movq	%rsp,%rdi        /* &pt_regs -> arg1 */
>  	call	syscall_trace_enter
>  	LOAD_ARGS ARGOFFSET  /* reload args from stack in case ptrace changed it */
> +	/*
> +	 * Now, we need the correct value of rbp to be restored.  It
> +	 * was never munged, so we can save it to the rbp slot and
> +	 * just have it restored.
> +	 */
> +	movq	%rbp, RBP(%rsp)
>  	RESTORE_REST
>  	movl	%ebp, %ebp
>  	/* no need to do an access_ok check here because rbp has been

Didn't we have problems with this exact approach before? This one was dropped
because it caused 32-bit programs to crash:
http://www2.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.13/2.6.13-mm2/broken-out/x86-64-ptrace-ia32-bp-fix.patch

See:
	http://lkml.org/lkml/2005/09/16/261

for the one I proposed instead.

And in any case doesn't cstar_tracesys() need the same fix?

Re: [PATCH x86 for review III] [17/29] x86: Add new CPUID bits for AMD Family 10 CPUs in /proc/cpuinfo

[Chuck Ebbert]

Andi Kleen wrote:
> Just various new acronyms. The new popcnt bit is in the middle
> of Intel space. This looks a little weird, but I've been assured
> it's ok.
> 
> Also I fixed RDTSCP for i386 which was at the wrong place.
> 
> For i386 and x86-64.
> 
> Signed-off-by: Andi Kleen <ak@suse.de>
> 
> ---
>  arch/i386/kernel/cpu/proc.c |   14 +++++++++-----
>  arch/x86_64/kernel/setup.c  |   14 ++++++++++----
>  2 files changed, 19 insertions(+), 9 deletions(-)
> 
> Index: linux/arch/x86_64/kernel/setup.c
> ===================================================================
> --- linux.orig/arch/x86_64/kernel/setup.c
> +++ linux/arch/x86_64/kernel/setup.c
> @@ -942,7 +942,8 @@ static int show_cpuinfo(struct seq_file 
>  		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
>  		NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL,
>  		NULL, NULL, NULL, NULL, "nx", NULL, "mmxext", NULL,
> -		NULL, "fxsr_opt", NULL, "rdtscp", NULL, "lm", "3dnowext", "3dnow",
> +		NULL, "fxsr_opt", "pdpe1gb", "rdtscp", NULL, "lm",
> +		"3dnowext", "3dnow",
>  
>  		/* Transmeta-defined */
>  		"recovery", "longrun", NULL, "lrti", NULL, NULL, NULL, NULL,
> @@ -960,7 +961,7 @@ static int show_cpuinfo(struct seq_file 
>  		/* Intel-defined (#2) */
>  		"pni", NULL, NULL, "monitor", "ds_cpl", "vmx", "smx", "est",
>  		"tm2", "ssse3", "cid", NULL, NULL, "cx16", "xtpr", NULL,
> -		NULL, NULL, "dca", NULL, NULL, NULL, NULL, NULL,
> +		NULL, NULL, "dca", NULL, NULL, NULL, NULL, "popcnt",
>  		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
>  
>  		/* VIA/Cyrix/Centaur-defined */
> @@ -970,8 +971,10 @@ static int show_cpuinfo(struct seq_file 
>  		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
>  
>  		/* AMD-defined (#2) */
> -		"lahf_lm", "cmp_legacy", "svm", NULL, "cr8_legacy", NULL, NULL, NULL,
> -		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
> +		"lahf_lm", "cmp_legacy", "svm", "extapic", "cr8_legacy",
> +		"altmovcr8", "abm", "sse4a",
> +		"misalignsse", "3dnowprefetch",
> +		"osvw", "ibs", NULL, NULL, NULL, NULL,
>  		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
>  		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
>  	};
> @@ -982,6 +985,9 @@ static int show_cpuinfo(struct seq_file 
>  		"ttp",  /* thermal trip */
>  		"tm",
>  		"stc",
> +		"100mhzsteps",
> +		"hwpstate",
> +		NULL,	/* tsc invariant mapped to constant_tsc */
>  		NULL,
>  		/* nothing */	/* constant_tsc - moved to flags */
>  	};
> Index: linux/arch/i386/kernel/cpu/proc.c
> ===================================================================
> --- linux.orig/arch/i386/kernel/cpu/proc.c
> +++ linux/arch/i386/kernel/cpu/proc.c
> @@ -29,7 +29,7 @@ static int show_cpuinfo(struct seq_file 
>  		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
>  		NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL,
>  		NULL, NULL, NULL, "mp", "nx", NULL, "mmxext", NULL,
> -		NULL, "fxsr_opt", "rdtscp", NULL, NULL, "lm", "3dnowext", "3dnow",
> +		NULL, "fxsr_opt", "pdpe1gb", "rdtscp", NULL, "lm", "3dnowext", "3dnow",
>  
>  		/* Transmeta-defined */
>  		"recovery", "longrun", NULL, "lrti", NULL, NULL, NULL, NULL,
> @@ -47,7 +47,7 @@ static int show_cpuinfo(struct seq_file 
>  		/* Intel-defined (#2) */
>  		"pni", NULL, NULL, "monitor", "ds_cpl", "vmx", "smx", "est",
>  		"tm2", "ssse3", "cid", NULL, NULL, "cx16", "xtpr", NULL,
> -		NULL, NULL, "dca", NULL, NULL, NULL, NULL, NULL,
> +		NULL, NULL, "dca", NULL, NULL, NULL, NULL, "popcnt",
>  		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
>  
>  		/* VIA/Cyrix/Centaur-defined */
> @@ -57,8 +57,9 @@ static int show_cpuinfo(struct seq_file 
>  		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
>  
>  		/* AMD-defined (#2) */
> -		"lahf_lm", "cmp_legacy", "svm", NULL, "cr8legacy", NULL, NULL, NULL,
> -		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
> +		"lahf_lm", "cmp_legacy", "svm", "extapic", "cr8legacy", "abm",
> +		"sse4a", "misalignsse",
> +		"3dnowprefetch", "osvw", "ibs", NULL, NULL, NULL, NULL, NULL,
>  		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
>  		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
>  	};
> @@ -69,8 +70,11 @@ static int show_cpuinfo(struct seq_file 
>  		"ttp",  /* thermal trip */
>  		"tm",
>  		"stc",
> +		"100mhzsteps",
> +		"hwpstate",
>  		NULL,
> -		/* nothing */	/* constant_tsc - moved to flags */
> +		NULL,	/* constant_tsc - moved to flags */
> +		/* nothing */
>  	};
>  	struct cpuinfo_x86 *c = v;
>  	int i, n = c - cpu_data;
> -

Since we seem to have become the place where all these are collected,
shouldn't we document what they mean? I've got some this machine that
I'd like to know more about:

dts ss ds_cpl est tm2 ssse3 xtpr

Re: [PATCH x86 for review III] [17/29] x86: Add new CPUID bits for AMD Family 10 CPUs in /proc/cpuinfo

[Andi Kleen]

> 
> Since we seem to have become the place where all these are collected,
> shouldn't we document what they mean?

This is not a data sheet and the CPU vendors have fine tech writers
working on this already.

Also I don't really want to encourage people to hack this without
at least taking a basic look at the documentation first.

> I've got some this machine that 
> I'd like to know more about:
> 
> dts ss ds_cpl est tm2 ssse3 xtpr

All documented in Intel documentation. But it's all relatively boring stuff

dts = debug trace store
ss = self snoop
ds_cpl = cpl qualified debug store
est = enhanced speedstep
tm2 = thermal monitor 2
ssse3 = more sse3 instructions
xtpr = tpr register chipset update control msr

-Andi

Re: [PATCH x86 for review III] [12/29] x86_64: 32-bit ptrace mangles sixth system call argument

[Andi Kleen]

> Didn't we have problems with this exact approach before? This one was dropped
> because it caused 32-bit programs to crash:
> http://www2.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.13/2.6.13-mm2/broken-out/x86-64-ptrace-ia32-bp-fix.patch
> 
> See:
> 	http://lkml.org/lkml/2005/09/16/261
> 
> for the one I proposed instead.

Yes you're right. I dropped it for now.

-Andi

Re: [PATCH x86 for review III] [17/29] x86: Add new CPUID bits for AMD Family 10 CPUs in /proc/cpuinfo

[Chuck Ebbert]

Andi Kleen wrote:

>> Since we seem to have become the place where all these are collected,
>> shouldn't we document what they mean?
> 
> This is not a data sheet and the CPU vendors have fine tech writers
> working on this already.
> 
> Also I don't really want to encourage people to hack this without
> at least taking a basic look at the documentation first.
> 
>> I've got some this machine that 
>> I'd like to know more about:
>>
>> dts ss ds_cpl est tm2 ssse3 xtpr
> 
> All documented in Intel documentation. But it's all relatively boring stuff
> 
> dts = debug trace store
> ss = self snoop
> ds_cpl = cpl qualified debug store
> est = enhanced speedstep
> tm2 = thermal monitor 2
> ssse3 = more sse3 instructions
> xtpr = tpr register chipset update control msr
> 

That's exactly what I was asking for.  Our abbreviations are too cryptic
to let someone go look up the meanings without a lot of trouble.

A one-liner for each in the comments would be just enough.

Re: [PATCH x86 for review III] [17/29] x86: Add new CPUID bits for AMD Family 10 CPUs in /proc/cpuinfo

[Andi Kleen]

> A one-liner for each in the comments would be just enough.

No. RTFM.

-Andi

Re: [PATCH x86 for review III] [17/29] x86: Add new CPUID bits for AMD Family 10 CPUs in /proc/cpuinfo

[Chuck Ebbert]

Andi Kleen wrote:
>> A one-liner for each in the comments would be just enough.
> 
> No. RTFM.

Sure, I'll get right on that:

./AMD:
total 16216
-rw-rw-r-- 1 cebbert cebbert 2485489 Nov 22 13:49 24592.pdf
-rw-rw-r-- 1 cebbert cebbert 3537793 Nov 22 13:49 24593.pdf
-rw-rw-r-- 1 cebbert cebbert 2212151 Nov 22 13:49 24594.pdf
-rw-rw-r-- 1 cebbert cebbert 3474511 Nov 22 13:49 25112.pdf
-rw-rw-r-- 1 cebbert cebbert 1992762 Nov 22 13:49 26568.pdf
-rw-rw-r-- 1 cebbert cebbert 1558189 Nov 22 13:49 26569.pdf
-rw-rw-r-- 1 cebbert cebbert  616422 Nov 22 13:49 34434.pdf
-rw-rw-r-- 1 cebbert cebbert  646779 Nov 22 13:49 40555.pdf

./Intel:
total 17284
-rw-rw-r-- 1 cebbert cebbert 3199114 Nov  8 16:58 248966.pdf
-rw-rw-r-- 1 cebbert cebbert  585350 Oct 30 18:44 25204618.pdf
-rw-rw-r-- 1 cebbert cebbert 3311769 Nov 16 14:49 253665.pdf
-rw-rw-r-- 1 cebbert cebbert 2515357 Nov 16 14:49 253666.pdf
-rw-rw-r-- 1 cebbert cebbert 2093479 Nov 16 14:49 253667.pdf
-rw-rw-r-- 1 cebbert cebbert 3274186 Nov 16 14:49 253668.pdf
-rw-rw-r-- 1 cebbert cebbert 2645170 Nov 16 14:50 253669.pdf

:-)

Re: [PATCH x86 for review III] [1/29] i386: avoid gcc extension

[Arnd Bergmann]

On Monday 12 February 2007 17:51, Andi Kleen wrote:
> setcc() in math-emu is written as a gcc extension statement expression
> macro that returns a value.  However, it's not used that way and it's not
> needed like that, so just make it a do-while non-extension macro so that we
> don't use an extension when it's not needed.
> 

The patch looks good but it doesn't match the description any more, since
you now use a function...

Re: [PATCH x86 for review III] [1/29] i386: avoid gcc extension

[Randy Dunlap]

On Tue, 13 Feb 2007 16:11:14 +0100 Arnd Bergmann wrote:

> On Monday 12 February 2007 17:51, Andi Kleen wrote:
> > setcc() in math-emu is written as a gcc extension statement expression
> > macro that returns a value.  However, it's not used that way and it's not
> > needed like that, so just make it a do-while non-extension macro so that we
> > don't use an extension when it's not needed.
> > 
> 
> The patch looks good but it doesn't match the description any more, since
> you now use a function...

Hi Arnd,
Someone else had that same comment, so I sent a corrected patch
(description change only) to that mail.

Thanks,
---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***

Re: [PATCH x86 for review III] [1/29] i386: avoid gcc extension

[Andi Kleen]

On Tuesday 13 February 2007 17:10, Randy Dunlap wrote:
> On Tue, 13 Feb 2007 16:11:14 +0100 Arnd Bergmann wrote:
> 
> > On Monday 12 February 2007 17:51, Andi Kleen wrote:
> > > setcc() in math-emu is written as a gcc extension statement expression
> > > macro that returns a value.  However, it's not used that way and it's not
> > > needed like that, so just make it a do-while non-extension macro so that we
> > > don't use an extension when it's not needed.
> > > 
> > 
> > The patch looks good but it doesn't match the description any more, since
> > you now use a function...
> 
> Hi Arnd,
> Someone else had that same comment, so I sent a corrected patch
> (description change only) to that mail.

I had already fixed it up

-Andi

Re: [PATCH x86 for review III] [12/29] x86_64: 32-bit ptrace mangles sixth system call argument

[Jeff Dike]

On Mon, Feb 12, 2007 at 05:04:54PM -0500, Chuck Ebbert wrote:
> See:
> 	http://lkml.org/lkml/2005/09/16/261
> 
> for the one I proposed instead.

That works for me - can we get this into mainline?

				Jeff

-- 
Work email - jdike at linux dot intel dot com