[PATCH] x86-64 updates

Lots of x86-64 updates. Merge with 2.4 and NUMA works now. Also reenabled
the preemptive kernel. And some other bug fixes.
IOMMU disabled by default now because it has problems.

 - Add more CONFIG options for device driver debugging and iommu
   force/debug.  (don't enable iommu force currently)
 - Some S3/ACPI fixes/cleanups from Pavel.
 - Set MSG_COMPAT_* in msg_flags for networking 32bit emulation.
   This unfortunately still doesn't fix the fd passing problems.
 - Sync PCI IOMMU code with 2.4 (minor fixes, flush less often)
 - Really fix UP compilation (Pavel)
 - Reenable preempt
 - Fix CONFIG_DISCONTIGMEM bootup and enable.  Still needs more tuning.
 - Fix some bugs in topology discovery and clean code up.
 - Don't put unwind tables into object files
 - Some kernel debugging hooks
 - Move CPU detection into early real mode code to better interact with
   vesafb consoles
 - Initialize mode in real mode character output
 - New 32bit FPU signal save/restore
 - Various fixes in FPU handling in ptrace
 - Fix security holes in ptrace (32bit and 64bit)
 - Fix serial ioctl (including security hole)
 - Add bluetooth ioctls to 32bit emu (from sparc64)
 - Correctly enable si_val in queued signals in 32bit emulation
 - Rework SEM_STAT emulation.  LTP still fails unfortunately.
 - Fix error case in msg* emulation
 - Fix debug register access from ptrace (Michal Ludvig, me)
 - Fix handling of NULL arguments in 32bit execve
 - Fix some error cases for 32bit readv/writev (LTP still complains)
 - Remove rate control from unimplemented syscall warnings
 - Fix error message for missing aperture
 - Turn some APIC printks into Dprintk to make the bootup more quiet
 - Some fixes for no APIC (probably still broken), add disableapic
   option (untested)
 - Sync K8 MCE handler with 2.4.  Should work a lot better now.
 - Remove never used KDB hooks
 - Fix buffer overflow in command line copying
 - Merge from i386: use separate status word for lazy FPU state
 - Don't force the IOMMU for dma masks < 4GB.
 - Print backtrace in Sysrq-T (from Andrea)
 - Merge from i386: fix FPU race in fork.
 - Disable NX mode by default for now
 - Rewrite dump_pagetable
 - Fix off by one bug in ioremap (i386 merge)
 - Merge from i386: handle VIA pci bridge bugs
 - Disable NUMA ACPI support (no SRAT support yet)
 - Fix aio 32bit emulation
 - Increase 32bit address space to nearly 4GB
 - Add exit_group syscall
 - Fix TLS setting in clone (Ulrich Drepper)

15 files changed, 83 insertions, 70 deletions

diff --git a/include/asm-x86_64/apic.h b/include/asm-x86_64/apic.h
index e15d95de1f69..bce8e2252907 100644
--- a/include/asm-x86_64/apic.h
+++ b/include/asm-x86_64/apic.h
@@ -75,7 +75,7 @@ extern void smp_local_timer_interrupt (struct pt_regs * regs);
 extern void setup_boot_APIC_clock (void);
 extern void setup_secondary_APIC_clock (void);
 extern void setup_apic_nmi_watchdog (void);
-extern inline void nmi_watchdog_tick (struct pt_regs * regs);
+extern void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
 extern int APIC_init_uniprocessor (void);
 extern void disable_APIC_timer(void);
 extern void enable_APIC_timer(void);
diff --git a/include/asm-x86_64/debugreg.h b/include/asm-x86_64/debugreg.h
index 81c8f097e4d6..2c4fe65e69ac 100644
--- a/include/asm-x86_64/debugreg.h
+++ b/include/asm-x86_64/debugreg.h
@@ -58,7 +58,7 @@
    We can slow the instruction pipeline for instructions coming via the
    gdt or the ldt if we want to.  I am not sure why this is an advantage */
 
-#define DR_CONTROL_RESERVED (0xFC00) /* Reserved by Intel */
+#define DR_CONTROL_RESERVED (0xFFFFFFFFFC00) /* Reserved by Intel */
 #define DR_LOCAL_SLOWDOWN (0x100)   /* Local slow the pipeline */
 #define DR_GLOBAL_SLOWDOWN (0x200)  /* Global slow the pipeline */
 
diff --git a/include/asm-x86_64/desc.h b/include/asm-x86_64/desc.h
index 33d997b78ecc..c871c9e5aecb 100644
--- a/include/asm-x86_64/desc.h
+++ b/include/asm-x86_64/desc.h
@@ -8,6 +8,7 @@
 #ifndef __ASSEMBLY__
 
 #include <asm/segment.h>
+#include <asm/mmu.h>
 
 // 8 byte segment descriptor
 struct desc_struct { 
diff --git a/include/asm-x86_64/hdreg.h b/include/asm-x86_64/hdreg.h
index 6ca37a9778de..52a2ed303fcf 100644
--- a/include/asm-x86_64/hdreg.h
+++ b/include/asm-x86_64/hdreg.h
@@ -7,6 +7,4 @@
 #ifndef __ASMx86_64_HDREG_H
 #define __ASMx86_64_HDREG_H
 
-typedef unsigned long ide_ioreg_t;
-
 #endif /* __ASMx86_64_HDREG_H */
diff --git a/include/asm-x86_64/i387.h b/include/asm-x86_64/i387.h
index 09e18af42ae9..e5bc31f19001 100644
--- a/include/asm-x86_64/i387.h
+++ b/include/asm-x86_64/i387.h
@@ -19,15 +19,15 @@
 #include <asm/thread_info.h>
 
 extern void fpu_init(void);
-extern void init_fpu(void);
-int save_i387(struct _fpstate *buf);
+extern void init_fpu(struct task_struct *child);
+extern int save_i387(struct _fpstate *buf);
 
 static inline int need_signal_i387(struct task_struct *me) 
 { 
 	if (!me->used_math)
 		return 0;
 	me->used_math = 0; 
-	if (me->thread_info->flags & _TIF_USEDFPU)
+	if (me->thread_info->status & TS_USEDFPU)
 		return 0;
 	return 1;
 } 
@@ -39,14 +39,14 @@ static inline int need_signal_i387(struct task_struct *me)
 #define kernel_fpu_end() stts()
 
 #define unlazy_fpu(tsk) do { \
-	if ((tsk)->thread_info->flags & _TIF_USEDFPU) \
+	if ((tsk)->thread_info->status & TS_USEDFPU) \
 		save_init_fpu(tsk); \
 } while (0)
 
 #define clear_fpu(tsk) do { \
-	if ((tsk)->thread_info->flags & _TIF_USEDFPU) {		\
+	if ((tsk)->thread_info->status & TS_USEDFPU) {		\
 		asm volatile("fwait");				\
-		(tsk)->thread_info->flags &= ~_TIF_USEDFPU;	\
+		(tsk)->thread_info->status &= ~TS_USEDFPU;	\
 		stts();						\
 	}							\
 } while (0)
@@ -114,11 +114,11 @@ static inline int save_i387_checking(struct i387_fxsave_struct *fx)
 
 static inline void kernel_fpu_begin(void)
 {
-	struct task_struct *me = current;
-	if (test_tsk_thread_flag(me,TIF_USEDFPU)) {
-		asm volatile("fxsave %0 ; fnclex"
-			      : "=m" (me->thread.i387.fxsave));
-		clear_tsk_thread_flag(me, TIF_USEDFPU);
+	struct thread_info *me = current_thread_info();
+	if (me->status & TS_USEDFPU) { 
+		asm volatile("rex64 ; fxsave %0 ; fnclex"
+			      : "=m" (me->task->thread.i387.fxsave));
+		me->status &= ~TS_USEDFPU;
 		return;
 	}
 	clts();
@@ -128,7 +128,7 @@ static inline void save_init_fpu( struct task_struct *tsk )
 {
 	asm volatile( "fxsave %0 ; fnclex"
 		      : "=m" (tsk->thread.i387.fxsave));
-	tsk->thread_info->flags &= ~TIF_USEDFPU;
+	tsk->thread_info->status &= ~TS_USEDFPU;
 	stts();
 }
 
@@ -140,18 +140,4 @@ static inline int restore_i387(struct _fpstate *buf)
 	return restore_fpu_checking((struct i387_fxsave_struct *)buf);
 }
 
-
-static inline void empty_fpu(struct task_struct *child)
-{
-	if (!child->used_math) {
-		/* Simulate an empty FPU. */
-		memset(&child->thread.i387.fxsave,0,sizeof(struct i387_fxsave_struct));
-		child->thread.i387.fxsave.cwd = 0x037f; 
-		child->thread.i387.fxsave.swd = 0;
-		child->thread.i387.fxsave.twd = 0; 
-		child->thread.i387.fxsave.mxcsr = 0x1f80;
-	}
-	child->used_math = 1; 
-}		
-
 #endif /* __ASM_X86_64_I387_H */
diff --git a/include/asm-x86_64/kdebug.h b/include/asm-x86_64/kdebug.h
index 745262828a27..316d01ac6eef 100644
--- a/include/asm-x86_64/kdebug.h
+++ b/include/asm-x86_64/kdebug.h
@@ -9,8 +9,13 @@ struct die_args {
 	struct pt_regs *regs;
 	const char *str;
 	long err; 
+	int trapnr;
+	int signr;
 }; 
 
+/* Note - you should never unregister because that can race with NMIs.
+   If you really want to do it first unregister - then synchronize_kernel - then free. 
+  */
 extern struct notifier_block *die_chain;
 
 /* Grossly misnamed. */
@@ -21,15 +26,16 @@ enum die_val {
 	DIE_PANIC,
 	DIE_NMI,
 	DIE_DIE,
+	DIE_NMIWATCHDOG,
+	DIE_KERNELDEBUG,
+	DIE_TRAP,
+	DIE_GPF,
 	DIE_CALL,
-	DIE_CPUINIT,	/* not really a die, but .. */
-	DIE_TRAPINIT,	/* not really a die, but .. */
-	DIE_STOP, 
 }; 
 	
-static inline int notify_die(enum die_val val,char *str,struct pt_regs *regs,long err)
+static inline int notify_die(enum die_val val,char *str,struct pt_regs *regs,long err,int trap, int sig)
 { 
-	struct die_args args = { regs: regs, str: str, err: err }; 
+	struct die_args args = { .regs=regs, .str=str, .err=err, .trapnr=trap,.signr=sig }; 
 	return notifier_call_chain(&die_chain, val, &args); 
 } 
 
diff --git a/include/asm-x86_64/numa.h b/include/asm-x86_64/numa.h
index 7686e4dfd9f4..340e71cfc538 100644
--- a/include/asm-x86_64/numa.h
+++ b/include/asm-x86_64/numa.h
@@ -8,13 +8,11 @@ struct node {
 	u64 start,end; 
 };
 
-#define for_all_nodes(x) for ((x) = 0; (x) <= maxnode; (x)++) \
+#define for_all_nodes(x) for ((x) = 0; (x) < numnodes; (x)++) \
 				if ((1UL << (x)) & nodes_present)
 
-#define early_for_all_nodes(n) \
-	for (n=0; n<MAXNODE;n++) if (nodes[n].start!=nodes[n].end)
 
-extern int compute_hash_shift(struct node *nodes, int numnodes, u64 maxmem);
+extern int compute_hash_shift(struct node *nodes);
 extern unsigned long nodes_present;
 
 #define ZONE_ALIGN (1UL << (MAX_ORDER+PAGE_SHIFT))
diff --git a/include/asm-x86_64/pci.h b/include/asm-x86_64/pci.h
index 74594c63e38c..d94c514a06ab 100644
--- a/include/asm-x86_64/pci.h
+++ b/include/asm-x86_64/pci.h
@@ -44,8 +44,7 @@ int pcibios_set_irq_routing(struct pci_dev *dev, int pin, int irq);
 
 struct pci_dev;
 
-extern int iommu_setup(char *opt, char **end);
-
+extern int iommu_setup(char *opt);
 extern void pci_iommu_init(void);
 
 /* Allocate and map kernel buffer using consistent mode DMA for a device.
@@ -77,10 +76,11 @@ extern void pci_free_consistent(struct pci_dev *hwdev, size_t size,
  * Once the device is given the dma address, the device owns this memory
  * until either pci_unmap_single or pci_dma_sync_single is performed.
  */
-extern dma_addr_t pci_map_single(struct pci_dev *hwdev, void *ptr,
-			  size_t size, int direction);
+extern dma_addr_t __pci_map_single(struct pci_dev *hwdev, void *ptr,
+			  size_t size, int direction, int flush);
+
 
-extern void pci_unmap_single(struct pci_dev *hwdev, dma_addr_t addr,
+void pci_unmap_single(struct pci_dev *hwdev, dma_addr_t addr,
 				   size_t size, int direction);
 
 /*
@@ -118,12 +118,16 @@ static inline void pci_dma_sync_sg(struct pci_dev *hwdev,
 	BUG_ON(direction == PCI_DMA_NONE); 
 } 
 
-#define PCI_DMA_BUS_IS_PHYS	0
+/* The PCI address space does equal the physical memory
+ * address space.  The networking and block device layers use
+ * this boolean for bounce buffer decisions.
+ */
+#define PCI_DMA_BUS_IS_PHYS	(0)
 
 
 #else
-static inline dma_addr_t pci_map_single(struct pci_dev *hwdev, void *ptr,
-					size_t size, int direction)
+static inline dma_addr_t __pci_map_single(struct pci_dev *hwdev, void *ptr,
+					size_t size, int direction, int flush)
 {
 	dma_addr_t addr; 
 
@@ -210,6 +214,11 @@ extern int pci_map_sg(struct pci_dev *hwdev, struct scatterlist *sg,
 extern void pci_unmap_sg(struct pci_dev *hwdev, struct scatterlist *sg,
 			 int nents, int direction);
 
+static inline dma_addr_t pci_map_single(struct pci_dev *hwdev, void *ptr,
+			  size_t size, int direction)
+{
+	return __pci_map_single(hwdev,ptr,size,direction,1); 
+}
 
 #define pci_unmap_page pci_unmap_single
 
diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index 7aa012007c7b..4d31ca2fef05 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -283,6 +283,7 @@ static inline int pmd_large(pmd_t pte) {
 #define pml4_page(pml4) ((unsigned long) __va(pml4_val(pml4) & PTE_MASK))
 #define pml4_index(address) ((address >> PML4_SHIFT) & (PTRS_PER_PML4-1))
 #define pml4_offset_k(address) (init_level4_pgt + pml4_index(address))
+#define pml4_present(pml4) (pml4_val(pml4) & _PAGE_PRESENT)
 #define mk_kernel_pml4(address) ((pml4_t){ (address) | _KERNPG_TABLE })
 #define level3_offset_k(dir, address) ((pgd_t *) pml4_page(*(dir)) + pgd_index(address))
 
diff --git a/include/asm-x86_64/processor.h b/include/asm-x86_64/processor.h
index 705ccf54580c..49d3c6ee414d 100644
--- a/include/asm-x86_64/processor.h
+++ b/include/asm-x86_64/processor.h
@@ -269,7 +269,7 @@ struct mm_struct;
 extern void release_thread(struct task_struct *);
 
 /* Prepare to copy thread state - unlazy all lazy status */
-#define prepare_to_copy(tsk)	do { } while (0)
+extern void prepare_to_copy(struct task_struct *tsk);
 
 /*
  * create a kernel thread without removing it from tasklists
@@ -308,8 +308,8 @@ extern inline void sync_core(void)
 #define ARCH_HAS_PREFETCHW
 #define ARCH_HAS_SPINLOCK_PREFETCH
 
-#define prefetch(x) __builtin_prefetch((x),0)
-#define prefetchw(x) __builtin_prefetch((x),1)
+#define prefetch(x) __builtin_prefetch((x),0,1)
+#define prefetchw(x) __builtin_prefetch((x),1,1)
 #define spin_lock_prefetch(x)  prefetchw(x)
 #define cpu_relax()   rep_nop()
 
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index 785d812fe2e8..2ced019ad2ed 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -6,6 +6,7 @@
 /* misc architecture specific prototypes */
 
 struct cpuinfo_x86; 
+struct pt_regs;
 
 extern void get_cpu_vendor(struct cpuinfo_x86*);
 extern void start_kernel(void);
@@ -41,6 +42,8 @@ extern void free_bootmem_generic(unsigned long phys, unsigned len);
 extern unsigned long end_pfn_map; 
 
 extern void show_stack(unsigned long * rsp);
+extern void show_trace(unsigned long * rsp);
+extern void show_registers(struct pt_regs *regs);
 
 extern void exception_table_check(void);
 
diff --git a/include/asm-x86_64/suspend.h b/include/asm-x86_64/suspend.h
index bccd85bb8dec..cc193889eefa 100644
--- a/include/asm-x86_64/suspend.h
+++ b/include/asm-x86_64/suspend.h
@@ -11,7 +11,7 @@ arch_prepare_suspend(void)
 {
 }
 
-/* image of the saved processor state */
+/* Image of the saved processor state. If you touch this, fix acpi_wakeup.S. */
 struct saved_context {
   	u16 ds, es, fs, gs, ss;
 	unsigned long gs_base, gs_kernel_base, fs_base;
diff --git a/include/asm-x86_64/system.h b/include/asm-x86_64/system.h
index 439c801f015b..3d3846fc9e04 100644
--- a/include/asm-x86_64/system.h
+++ b/include/asm-x86_64/system.h
@@ -22,18 +22,18 @@
 struct save_context_frame { 
 	unsigned long rbp; 
 	unsigned long rbx;
+	unsigned long r11;
+	unsigned long r10;
+	unsigned long r9;
+	unsigned long r8;
 	unsigned long rcx;
 	unsigned long rdx;	
-	unsigned long rsi;
-	unsigned long rdi; 
 	unsigned long r15;
 	unsigned long r14;
 	unsigned long r13;
 	unsigned long r12;
-	unsigned long r11;
-	unsigned long r10;
-	unsigned long r9;
-	unsigned long r8; 
+	unsigned long rdi;
+	unsigned long rsi;
 }; 
 
 /* frame pointer must be last for get_wchan */
@@ -43,19 +43,20 @@ struct save_context_frame {
    rbp needs to be always explicitely saved because gcc cannot clobber the
    frame pointer and the scheduler is compiled with frame pointers. -AK */
 #define SAVE_CONTEXT \
-	__PUSH(r8) __PUSH(r9) __PUSH(r10) __PUSH(r11) __PUSH(r12) __PUSH(r13) \
-	__PUSH(r14) __PUSH(r15) \
-	__PUSH(rdi) __PUSH(rsi) \
-	__PUSH(rdx) __PUSH(rcx) \
+	__PUSH(rsi) __PUSH(rdi) \
+    __PUSH(r12) __PUSH(r13) __PUSH(r14) __PUSH(r15)  \
+	__PUSH(rdx) __PUSH(rcx) __PUSH(r8) __PUSH(r9) __PUSH(r10) __PUSH(r11)  \
 	__PUSH(rbx) __PUSH(rbp) 
 #define RESTORE_CONTEXT \
 	__POP(rbp) __POP(rbx) \
-	__POP(rcx) __POP(rdx) \
-	__POP(rsi) __POP(rdi) \
-	__POP(r15) __POP(r14) __POP(r13) __POP(r12) __POP(r11) __POP(r10) \
-	__POP(r9) __POP(r8)
+	__POP(r11) __POP(r10) __POP(r9) __POP(r8) __POP(rcx) __POP(rdx) \
+	__POP(r15) __POP(r14) __POP(r13) __POP(r12) \
+	__POP(rdi) __POP(rsi)
 
 /* RED-PEN: pipeline stall on ret because it is not predicted */
+/* RED-PEN: the register saving could be optimized */
+/* frame pointer must be last for get_wchan */
+
 #define switch_to(prev,next,last) \
 	asm volatile(SAVE_CONTEXT						    \
 		     "movq %%rsp,%[prevrsp]\n\t"				    \
diff --git a/include/asm-x86_64/thread_info.h b/include/asm-x86_64/thread_info.h
index f841e925bef0..11e8d21989c7 100644
--- a/include/asm-x86_64/thread_info.h
+++ b/include/asm-x86_64/thread_info.h
@@ -27,6 +27,7 @@ struct thread_info {
 	struct task_struct	*task;		/* main task structure */
 	struct exec_domain	*exec_domain;	/* execution domain */
 	__u32			flags;		/* low level flags */
+	__u32			status;		/* thread synchronous flags */
 	__u32			cpu;		/* current CPU */
 	int 			preempt_count;
 
@@ -100,16 +101,14 @@ static inline struct thread_info *stack_thread_info(void)
 #define TIF_SIGPENDING		2	/* signal pending */
 #define TIF_NEED_RESCHED	3	/* rescheduling necessary */
 #define TIF_SINGLESTEP		4	/* reenable singlestep on user return*/
-#define TIF_USEDFPU		16	/* FPU was used by this task this quantum */
-#define TIF_POLLING_NRFLAG	17	/* true if poll_idle() is polling TIF_NEED_RESCHED */
-#define TIF_IA32		18	/* 32bit process */ 
+#define TIF_POLLING_NRFLAG	16	/* true if poll_idle() is polling TIF_NEED_RESCHED */
+#define TIF_IA32		17	/* 32bit process */ 
 
 #define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
 #define _TIF_NOTIFY_RESUME	(1<<TIF_NOTIFY_RESUME)
 #define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
 #define _TIF_SINGLESTEP		(1<<TIF_SINGLESTEP)
 #define _TIF_NEED_RESCHED	(1<<TIF_NEED_RESCHED)
-#define _TIF_USEDFPU		(1<<TIF_USEDFPU)
 #define _TIF_POLLING_NRFLAG	(1<<TIF_POLLING_NRFLAG)
 #define _TIF_IA32		(1<<TIF_IA32)
 
@@ -118,6 +117,15 @@ static inline struct thread_info *stack_thread_info(void)
 
 #define PREEMPT_ACTIVE     0x4000000
 
+/*
+ * Thread-synchronous status.
+ *
+ * This is different from the flags in that nobody else
+ * ever touches our thread-synchronous status, so we don't
+ * have to worry about atomic accesses.
+ */
+#define TS_USEDFPU		0x0001	/* FPU was used by this task this quantum (SMP) */
+
 #endif /* __KERNEL__ */
 
 #endif /* _ASM_THREAD_INFO_H */
diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h
index 2c2d2e3fd5f8..7702073edc1c 100644
--- a/include/asm-x86_64/unistd.h
+++ b/include/asm-x86_64/unistd.h
@@ -520,8 +520,10 @@ __SYSCALL(__NR_clock_gettime, sys_clock_gettime)
 __SYSCALL(__NR_clock_getres, sys_clock_getres)
 #define __NR_clock_nanosleep	230
 __SYSCALL(__NR_clock_nanosleep, sys_clock_nanosleep)
+#define __NR_exit_group		231
+__SYSCALL(__NR_exit_group, sys_exit_group)
 
-#define __NR_syscall_max __NR_clock_nanosleep
+#define __NR_syscall_max __NR_exit_group
 #ifndef __NO_STUBS
 
 /* user-visible error numbers are in the range -1 - -4095 */