Merge tag 'pr-20150114-x86-entry' of git://git.kernel.org/pub/scm/linux/kernel/git/luto/linux into x86/asm

Pull x86/entry enhancements from Andy Lutomirski:

" This is my accumulated x86 entry work, part 1, for 3.20.  The meat
  of this is an IST rework.  When an IST exception interrupts user
  space, we will handle it on the per-thread kernel stack instead of
  on the IST stack.  This sounds messy, but it actually simplifies the
  IST entry/exit code, because it eliminates some ugly games we used
  to play in order to handle rescheduling, signal delivery, etc on the
  way out of an IST exception.

  The IST rework introduces proper context tracking to IST exception
  handlers.  I haven't seen any bug reports, but the old code could
  have incorrectly treated an IST exception handler as an RCU extended
  quiescent state.

  The memory failure change (included in this pull request with
  Borislav and Tony's permission) eliminates a bunch of code that
  is no longer needed now that user memory failure handlers are
  called in process context.

  Finally, this includes a few on Denys' uncontroversial and Obviously
  Correct (tm) cleanups.

  The IST and memory failure changes have been in -next for a while.

  LKML references:

  IST rework:
  http://lkml.kernel.org/r/cover.1416604491.git.luto@amacapital.net

  Memory failure change:
  http://lkml.kernel.org/r/54ab2ffa301102cd6e@agluck-desk.sc.intel.com

  Denys' cleanups:
  http://lkml.kernel.org/r/1420927210-19738-1-git-send-email-dvlasenk@redhat.com
"

This tree semantically depends on and is based on the following RCU commit:

  734d16801349 ("rcu: Make rcu_nmi_enter() handle nesting")

... and for that reason won't be pushed upstream before the RCU bits hit Linus's tree.

Signed-off-by: Ingo Molnar <mingo@kernel.org>

Documentation/x86/entry_64.txt

@@ -78,9 +78,6 @@ The expensive (paranoid) way is to read back the MSR_GS_BASE value
 	xorl %ebx,%ebx
 1:	ret
 
-and the whole paranoid non-paranoid macro complexity is about whether
-to suffer that RDMSR cost.
-
 If we are at an interrupt or user-trap/gate-alike boundary then we can
 use the faster check: the stack will be a reliable indicator of
 whether SWAPGS was already done: if we see that we are a secondary
@@ -93,6 +90,15 @@ which might have triggered right after a normal entry wrote CS to the
 stack but before we executed SWAPGS, then the only safe way to check
 for GS is the slower method: the RDMSR.
 
-So we try only to mark those entry methods 'paranoid' that absolutely
-need the more expensive check for the GS base - and we generate all
-'normal' entry points with the regular (faster) entry macros.
+Therefore, super-atomic entries (except NMI, which is handled separately)
+must use idtentry with paranoid=1 to handle gsbase correctly.  This
+triggers three main behavior changes:
+
+ - Interrupt entry will use the slower gsbase check.
+ - Interrupt entry from user mode will switch off the IST stack.
+ - Interrupt exit to kernel mode will not attempt to reschedule.
+
+We try to only use IST entries and the paranoid entry code for vectors
+that absolutely need the more expensive check for the GS base - and we
+generate all 'normal' entry points with the regular (faster) paranoid=0
+variant.

Documentation/x86/x86_64/kernel-stacks

@@ -40,9 +40,11 @@ An IST is selected by a non-zero value in the IST field of an
 interrupt-gate descriptor.  When an interrupt occurs and the hardware
 loads such a descriptor, the hardware automatically sets the new stack
 pointer based on the IST value, then invokes the interrupt handler.  If
-software wants to allow nested IST interrupts then the handler must
-adjust the IST values on entry to and exit from the interrupt handler.
-(This is occasionally done, e.g. for debug exceptions.)
+the interrupt came from user mode, then the interrupt handler prologue
+will switch back to the per-thread stack.  If software wants to allow
+nested IST interrupts then the handler must adjust the IST values on
+entry to and exit from the interrupt handler.  (This is occasionally
+done, e.g. for debug exceptions.)
 
 Events with different IST codes (i.e. with different stacks) can be
 nested.  For example, a debug interrupt can safely be interrupted by an

arch/x86/ia32/ia32entry.S

@@ -179,8 +179,8 @@ sysenter_dispatch:
 sysexit_from_sys_call:
 	andl    $~TS_COMPAT,TI_status+THREAD_INFO(%rsp,RIP-ARGOFFSET)
 	/* clear IF, that popfq doesn't enable interrupts early */
-	andl  $~0x200,EFLAGS-R11(%rsp) 
-	movl	RIP-R11(%rsp),%edx		/* User %eip */
+	andl	$~0x200,EFLAGS-ARGOFFSET(%rsp)
+	movl	RIP-ARGOFFSET(%rsp),%edx		/* User %eip */
 	CFI_REGISTER rip,rdx
 	RESTORE_ARGS 0,24,0,0,0,0
 	xorq	%r8,%r8

arch/x86/include/asm/calling.h

@@ -83,7 +83,6 @@ For 32-bit we have the following conventions - kernel is built with
 #define SS		160
 
 #define ARGOFFSET	R11
-#define SWFRAME		ORIG_RAX
 
 	.macro SAVE_ARGS addskip=0, save_rcx=1, save_r891011=1, rax_enosys=0
 	subq  $9*8+\addskip, %rsp

arch/x86/include/asm/mce.h

@@ -190,7 +190,6 @@ enum mcp_flags {
 void machine_check_poll(enum mcp_flags flags, mce_banks_t *b);
 
 int mce_notify_irq(void);
-void mce_notify_process(void);
 
 DECLARE_PER_CPU(struct mce, injectm);
 

arch/x86/include/asm/thread_info.h

@@ -75,7 +75,6 @@ struct thread_info {
 #define TIF_SYSCALL_EMU		6	/* syscall emulation active */
 #define TIF_SYSCALL_AUDIT	7	/* syscall auditing active */
 #define TIF_SECCOMP		8	/* secure computing */
-#define TIF_MCE_NOTIFY		10	/* notify userspace of an MCE */
 #define TIF_USER_RETURN_NOTIFY	11	/* notify kernel of userspace return */
 #define TIF_UPROBE		12	/* breakpointed or singlestepping */
 #define TIF_NOTSC		16	/* TSC is not accessible in userland */
@@ -100,7 +99,6 @@ struct thread_info {
 #define _TIF_SYSCALL_EMU	(1 << TIF_SYSCALL_EMU)
 #define _TIF_SYSCALL_AUDIT	(1 << TIF_SYSCALL_AUDIT)
 #define _TIF_SECCOMP		(1 << TIF_SECCOMP)
-#define _TIF_MCE_NOTIFY		(1 << TIF_MCE_NOTIFY)
 #define _TIF_USER_RETURN_NOTIFY	(1 << TIF_USER_RETURN_NOTIFY)
 #define _TIF_UPROBE		(1 << TIF_UPROBE)
 #define _TIF_NOTSC		(1 << TIF_NOTSC)
@@ -140,7 +138,7 @@ struct thread_info {
 
 /* Only used for 64 bit */
 #define _TIF_DO_NOTIFY_MASK						\
-	(_TIF_SIGPENDING | _TIF_MCE_NOTIFY | _TIF_NOTIFY_RESUME |	\
+	(_TIF_SIGPENDING | _TIF_NOTIFY_RESUME |				\
 	 _TIF_USER_RETURN_NOTIFY | _TIF_UPROBE)
 
 /* flags to check in __switch_to() */
@@ -170,6 +168,17 @@ static inline struct thread_info *current_thread_info(void)
 	return ti;
 }
 
+static inline unsigned long current_stack_pointer(void)
+{
+	unsigned long sp;
+#ifdef CONFIG_X86_64
+	asm("mov %%rsp,%0" : "=g" (sp));
+#else
+	asm("mov %%esp,%0" : "=g" (sp));
+#endif
+	return sp;
+}
+
 #else /* !__ASSEMBLY__ */
 
 /* how to get the thread information struct from ASM */

arch/x86/include/asm/traps.h

@@ -1,6 +1,7 @@
 #ifndef _ASM_X86_TRAPS_H
 #define _ASM_X86_TRAPS_H
 
+#include <linux/context_tracking_state.h>
 #include <linux/kprobes.h>
 
 #include <asm/debugreg.h>
@@ -110,6 +111,11 @@ asmlinkage void smp_thermal_interrupt(void);
 asmlinkage void mce_threshold_interrupt(void);
 #endif
 
+extern enum ctx_state ist_enter(struct pt_regs *regs);
+extern void ist_exit(struct pt_regs *regs, enum ctx_state prev_state);
+extern void ist_begin_non_atomic(struct pt_regs *regs);
+extern void ist_end_non_atomic(void);
+
 /* Interrupts/Exceptions */
 enum {
 	X86_TRAP_DE = 0,	/*  0, Divide-by-zero */

arch/x86/kernel/cpu/mcheck/mce.c

@@ -43,6 +43,7 @@
 #include <linux/export.h>
 
 #include <asm/processor.h>
+#include <asm/traps.h>
 #include <asm/mce.h>
 #include <asm/msr.h>
 
@@ -1002,51 +1003,6 @@ static void mce_clear_state(unsigned long *toclear)
 	}
 }
 
-/*
- * Need to save faulting physical address associated with a process
- * in the machine check handler some place where we can grab it back
- * later in mce_notify_process()
- */
-#define	MCE_INFO_MAX	16
-
-struct mce_info {
-	atomic_t		inuse;
-	struct task_struct	*t;
-	__u64			paddr;
-	int			restartable;
-} mce_info[MCE_INFO_MAX];
-
-static void mce_save_info(__u64 addr, int c)
-{
-	struct mce_info *mi;
-
-	for (mi = mce_info; mi < &mce_info[MCE_INFO_MAX]; mi++) {
-		if (atomic_cmpxchg(&mi->inuse, 0, 1) == 0) {
-			mi->t = current;
-			mi->paddr = addr;
-			mi->restartable = c;
-			return;
-		}
-	}
-
-	mce_panic("Too many concurrent recoverable errors", NULL, NULL);
-}
-
-static struct mce_info *mce_find_info(void)
-{
-	struct mce_info *mi;
-
-	for (mi = mce_info; mi < &mce_info[MCE_INFO_MAX]; mi++)
-		if (atomic_read(&mi->inuse) && mi->t == current)
-			return mi;
-	return NULL;
-}
-
-static void mce_clear_info(struct mce_info *mi)
-{
-	atomic_set(&mi->inuse, 0);
-}
-
 /*
  * The actual machine check handler. This only handles real
  * exceptions when something got corrupted coming in through int 18.
@@ -1063,6 +1019,7 @@ void do_machine_check(struct pt_regs *regs, long error_code)
 {
 	struct mca_config *cfg = &mca_cfg;
 	struct mce m, *final;
+	enum ctx_state prev_state;
 	int i;
 	int worst = 0;
 	int severity;
@@ -1084,6 +1041,10 @@ void do_machine_check(struct pt_regs *regs, long error_code)
 	DECLARE_BITMAP(toclear, MAX_NR_BANKS);
 	DECLARE_BITMAP(valid_banks, MAX_NR_BANKS);
 	char *msg = "Unknown";
+	u64 recover_paddr = ~0ull;
+	int flags = MF_ACTION_REQUIRED;
+
+	prev_state = ist_enter(regs);
 
 	this_cpu_inc(mce_exception_count);
 
@@ -1203,9 +1164,9 @@ void do_machine_check(struct pt_regs *regs, long error_code)
 		if (no_way_out)
 			mce_panic("Fatal machine check on current CPU", &m, msg);
 		if (worst == MCE_AR_SEVERITY) {
-			/* schedule action before return to userland */
-			mce_save_info(m.addr, m.mcgstatus & MCG_STATUS_RIPV);
-			set_thread_flag(TIF_MCE_NOTIFY);
+			recover_paddr = m.addr;
+			if (!(m.mcgstatus & MCG_STATUS_RIPV))
+				flags |= MF_MUST_KILL;
 		} else if (kill_it) {
 			force_sig(SIGBUS, current);
 		}
@@ -1216,6 +1177,27 @@ void do_machine_check(struct pt_regs *regs, long error_code)
 	mce_wrmsrl(MSR_IA32_MCG_STATUS, 0);
 out:
 	sync_core();
+
+	if (recover_paddr == ~0ull)
+		goto done;
+
+	pr_err("Uncorrected hardware memory error in user-access at %llx",
+		 recover_paddr);
+	/*
+	 * We must call memory_failure() here even if the current process is
+	 * doomed. We still need to mark the page as poisoned and alert any
+	 * other users of the page.
+	 */
+	ist_begin_non_atomic(regs);
+	local_irq_enable();
+	if (memory_failure(recover_paddr >> PAGE_SHIFT, MCE_VECTOR, flags) < 0) {
+		pr_err("Memory error not recovered");
+		force_sig(SIGBUS, current);
+	}
+	local_irq_disable();
+	ist_end_non_atomic();
+done:
+	ist_exit(regs, prev_state);
 }
 EXPORT_SYMBOL_GPL(do_machine_check);
 
@@ -1232,42 +1214,6 @@ int memory_failure(unsigned long pfn, int vector, int flags)
 }
 #endif
 
-/*
- * Called in process context that interrupted by MCE and marked with
- * TIF_MCE_NOTIFY, just before returning to erroneous userland.
- * This code is allowed to sleep.
- * Attempt possible recovery such as calling the high level VM handler to
- * process any corrupted pages, and kill/signal current process if required.
- * Action required errors are handled here.
- */
-void mce_notify_process(void)
-{
-	unsigned long pfn;
-	struct mce_info *mi = mce_find_info();
-	int flags = MF_ACTION_REQUIRED;
-
-	if (!mi)
-		mce_panic("Lost physical address for unconsumed uncorrectable error", NULL, NULL);
-	pfn = mi->paddr >> PAGE_SHIFT;
-
-	clear_thread_flag(TIF_MCE_NOTIFY);
-
-	pr_err("Uncorrected hardware memory error in user-access at %llx",
-		 mi->paddr);
-	/*
-	 * We must call memory_failure() here even if the current process is
-	 * doomed. We still need to mark the page as poisoned and alert any
-	 * other users of the page.
-	 */
-	if (!mi->restartable)
-		flags |= MF_MUST_KILL;
-	if (memory_failure(pfn, MCE_VECTOR, flags) < 0) {
-		pr_err("Memory error not recovered");
-		force_sig(SIGBUS, current);
-	}
-	mce_clear_info(mi);
-}
-
 /*
  * Action optional processing happens here (picking up
  * from the list of faulting pages that do_machine_check()

arch/x86/kernel/cpu/mcheck/p5.c

@@ -8,6 +8,7 @@
 #include <linux/smp.h>
 
 #include <asm/processor.h>
+#include <asm/traps.h>
 #include <asm/mce.h>
 #include <asm/msr.h>
 
@@ -17,8 +18,11 @@ int mce_p5_enabled __read_mostly;
 /* Machine check handler for Pentium class Intel CPUs: */
 static void pentium_machine_check(struct pt_regs *regs, long error_code)
 {
+	enum ctx_state prev_state;
 	u32 loaddr, hi, lotype;
 
+	prev_state = ist_enter(regs);
+
 	rdmsr(MSR_IA32_P5_MC_ADDR, loaddr, hi);
 	rdmsr(MSR_IA32_P5_MC_TYPE, lotype, hi);
 
@@ -33,6 +37,8 @@ static void pentium_machine_check(struct pt_regs *regs, long error_code)
 	}
 
 	add_taint(TAINT_MACHINE_CHECK, LOCKDEP_NOW_UNRELIABLE);
+
+	ist_exit(regs, prev_state);
 }
 
 /* Set up machine check reporting for processors with Intel style MCE: */

arch/x86/kernel/cpu/mcheck/winchip.c

@@ -7,14 +7,19 @@
 #include <linux/types.h>
 
 #include <asm/processor.h>
+#include <asm/traps.h>
 #include <asm/mce.h>
 #include <asm/msr.h>
 
 /* Machine check handler for WinChip C6: */
 static void winchip_machine_check(struct pt_regs *regs, long error_code)
 {
+	enum ctx_state prev_state = ist_enter(regs);
+
 	printk(KERN_EMERG "CPU0: Machine Check Exception.\n");
 	add_taint(TAINT_MACHINE_CHECK, LOCKDEP_NOW_UNRELIABLE);
+
+	ist_exit(regs, prev_state);
 }
 
 /* Set up machine check reporting on the Winchip C6 series */

arch/x86/kernel/entry_64.S

@@ -156,27 +156,6 @@ ENDPROC(native_usergs_sysret64)
 	movq \tmp,R11+\offset(%rsp)
 	.endm
 
-	.macro FAKE_STACK_FRAME child_rip
-	/* push in order ss, rsp, eflags, cs, rip */
-	xorl %eax, %eax
-	pushq_cfi $__KERNEL_DS /* ss */
-	/*CFI_REL_OFFSET	ss,0*/
-	pushq_cfi %rax /* rsp */
-	CFI_REL_OFFSET	rsp,0
-	pushq_cfi $(X86_EFLAGS_IF|X86_EFLAGS_FIXED) /* eflags - interrupts on */
-	/*CFI_REL_OFFSET	rflags,0*/
-	pushq_cfi $__KERNEL_CS /* cs */
-	/*CFI_REL_OFFSET	cs,0*/
-	pushq_cfi \child_rip /* rip */
-	CFI_REL_OFFSET	rip,0
-	pushq_cfi %rax /* orig rax */
-	.endm
-
-	.macro UNFAKE_STACK_FRAME
-	addq $8*6, %rsp
-	CFI_ADJUST_CFA_OFFSET	-(6*8)
-	.endm
-
 /*
  * initial frame state for interrupts (and exceptions without error code)
  */
@@ -239,51 +218,6 @@ ENDPROC(native_usergs_sysret64)
 	CFI_REL_OFFSET r15, R15+\offset
 	.endm
 
-/* save partial stack frame */
-	.macro SAVE_ARGS_IRQ
-	cld
-	/* start from rbp in pt_regs and jump over */
-	movq_cfi rdi, (RDI-RBP)
-	movq_cfi rsi, (RSI-RBP)
-	movq_cfi rdx, (RDX-RBP)
-	movq_cfi rcx, (RCX-RBP)
-	movq_cfi rax, (RAX-RBP)
-	movq_cfi  r8,  (R8-RBP)
-	movq_cfi  r9,  (R9-RBP)
-	movq_cfi r10, (R10-RBP)
-	movq_cfi r11, (R11-RBP)
-
-	/* Save rbp so that we can unwind from get_irq_regs() */
-	movq_cfi rbp, 0
-
-	/* Save previous stack value */
-	movq %rsp, %rsi
-
-	leaq -RBP(%rsp),%rdi	/* arg1 for handler */
-	testl $3, CS-RBP(%rsi)
-	je 1f
-	SWAPGS
-	/*
-	 * irq_count is used to check if a CPU is already on an interrupt stack
-	 * or not. While this is essentially redundant with preempt_count it is
-	 * a little cheaper to use a separate counter in the PDA (short of
-	 * moving irq_enter into assembly, which would be too much work)
-	 */
-1:	incl PER_CPU_VAR(irq_count)
-	cmovzq PER_CPU_VAR(irq_stack_ptr),%rsp
-	CFI_DEF_CFA_REGISTER	rsi
-
-	/* Store previous stack value */
-	pushq %rsi
-	CFI_ESCAPE	0x0f /* DW_CFA_def_cfa_expression */, 6, \
-			0x77 /* DW_OP_breg7 */, 0, \
-			0x06 /* DW_OP_deref */, \
-			0x08 /* DW_OP_const1u */, SS+8-RBP, \
-			0x22 /* DW_OP_plus */
-	/* We entered an interrupt context - irqs are off: */
-	TRACE_IRQS_OFF
-	.endm
-
 ENTRY(save_paranoid)
 	XCPT_FRAME 1 RDI+8
 	cld
@@ -627,19 +561,6 @@ END(\label)
 	FORK_LIKE  vfork
 	FIXED_FRAME stub_iopl, sys_iopl
 
-ENTRY(ptregscall_common)
-	DEFAULT_FRAME 1 8	/* offset 8: return address */
-	RESTORE_TOP_OF_STACK %r11, 8
-	movq_cfi_restore R15+8, r15
-	movq_cfi_restore R14+8, r14
-	movq_cfi_restore R13+8, r13
-	movq_cfi_restore R12+8, r12
-	movq_cfi_restore RBP+8, rbp
-	movq_cfi_restore RBX+8, rbx
-	ret $REST_SKIP		/* pop extended registers */
-	CFI_ENDPROC
-END(ptregscall_common)
-
 ENTRY(stub_execve)
 	CFI_STARTPROC
 	addq $8, %rsp
@@ -780,7 +701,48 @@ END(interrupt)
 	/* reserve pt_regs for scratch regs and rbp */
 	subq $ORIG_RAX-RBP, %rsp
 	CFI_ADJUST_CFA_OFFSET ORIG_RAX-RBP
-	SAVE_ARGS_IRQ
+	cld
+	/* start from rbp in pt_regs and jump over */
+	movq_cfi rdi, (RDI-RBP)
+	movq_cfi rsi, (RSI-RBP)
+	movq_cfi rdx, (RDX-RBP)
+	movq_cfi rcx, (RCX-RBP)
+	movq_cfi rax, (RAX-RBP)
+	movq_cfi  r8,  (R8-RBP)
+	movq_cfi  r9,  (R9-RBP)
+	movq_cfi r10, (R10-RBP)
+	movq_cfi r11, (R11-RBP)
+
+	/* Save rbp so that we can unwind from get_irq_regs() */
+	movq_cfi rbp, 0
+
+	/* Save previous stack value */
+	movq %rsp, %rsi
+
+	leaq -RBP(%rsp),%rdi	/* arg1 for handler */
+	testl $3, CS-RBP(%rsi)
+	je 1f
+	SWAPGS
+	/*
+	 * irq_count is used to check if a CPU is already on an interrupt stack
+	 * or not. While this is essentially redundant with preempt_count it is
+	 * a little cheaper to use a separate counter in the PDA (short of
+	 * moving irq_enter into assembly, which would be too much work)
+	 */
+1:	incl PER_CPU_VAR(irq_count)
+	cmovzq PER_CPU_VAR(irq_stack_ptr),%rsp
+	CFI_DEF_CFA_REGISTER	rsi
+
+	/* Store previous stack value */
+	pushq %rsi
+	CFI_ESCAPE	0x0f /* DW_CFA_def_cfa_expression */, 6, \
+			0x77 /* DW_OP_breg7 */, 0, \
+			0x06 /* DW_OP_deref */, \
+			0x08 /* DW_OP_const1u */, SS+8-RBP, \
+			0x22 /* DW_OP_plus */
+	/* We entered an interrupt context - irqs are off: */
+	TRACE_IRQS_OFF
+
 	call \func
 	.endm
 
@@ -1049,6 +1011,11 @@ ENTRY(\sym)
 	CFI_ADJUST_CFA_OFFSET ORIG_RAX-R15
 
 	.if \paranoid
+	.if \paranoid == 1
+	CFI_REMEMBER_STATE
+	testl $3, CS(%rsp)		/* If coming from userspace, switch */
+	jnz 1f				/* stacks. */
+	.endif
 	call save_paranoid
 	.else
 	call error_entry
@@ -1089,6 +1056,36 @@ ENTRY(\sym)
 	jmp error_exit			/* %ebx: no swapgs flag */
 	.endif
 
+	.if \paranoid == 1
+	CFI_RESTORE_STATE
+	/*
+	 * Paranoid entry from userspace.  Switch stacks and treat it
+	 * as a normal entry.  This means that paranoid handlers
+	 * run in real process context if user_mode(regs).
+	 */
+1:
+	call error_entry
+
+	DEFAULT_FRAME 0
+
+	movq %rsp,%rdi			/* pt_regs pointer */
+	call sync_regs
+	movq %rax,%rsp			/* switch stack */
+
+	movq %rsp,%rdi			/* pt_regs pointer */
+
+	.if \has_error_code
+	movq ORIG_RAX(%rsp),%rsi	/* get error code */
+	movq $-1,ORIG_RAX(%rsp)		/* no syscall to restart */
+	.else
+	xorl %esi,%esi			/* no error code */
+	.endif
+
+	call \do_sym
+
+	jmp error_exit			/* %ebx: no swapgs flag */
+	.endif
+
 	CFI_ENDPROC
 END(\sym)
 .endm
@@ -1109,7 +1106,7 @@ idtentry overflow do_overflow has_error_code=0
 idtentry bounds do_bounds has_error_code=0
 idtentry invalid_op do_invalid_op has_error_code=0
 idtentry device_not_available do_device_not_available has_error_code=0
-idtentry double_fault do_double_fault has_error_code=1 paranoid=1
+idtentry double_fault do_double_fault has_error_code=1 paranoid=2
 idtentry coprocessor_segment_overrun do_coprocessor_segment_overrun has_error_code=0
 idtentry invalid_TSS do_invalid_TSS has_error_code=1
 idtentry segment_not_present do_segment_not_present has_error_code=1
@@ -1290,16 +1287,14 @@ idtentry machine_check has_error_code=0 paranoid=1 do_sym=*machine_check_vector(
 #endif
 
 	/*
-	 * "Paranoid" exit path from exception stack.
-	 * Paranoid because this is used by NMIs and cannot take
-	 * any kernel state for granted.
-	 * We don't do kernel preemption checks here, because only
-	 * NMI should be common and it does not enable IRQs and
-	 * cannot get reschedule ticks.
+	 * "Paranoid" exit path from exception stack.  This is invoked
+	 * only on return from non-NMI IST interrupts that came
+	 * from kernel space.
 	 *
-	 * "trace" is 0 for the NMI handler only, because irq-tracing
-	 * is fundamentally NMI-unsafe. (we cannot change the soft and
-	 * hard flags at once, atomically)
+	 * We may be returning to very strange contexts (e.g. very early
+	 * in syscall entry), so checking for preemption here would
+	 * be complicated.  Fortunately, we there's no good reason
+	 * to try to handle preemption here.
 	 */
 
 	/* ebx:	no swapgs flag */
@@ -1309,43 +1304,14 @@ ENTRY(paranoid_exit)
 	TRACE_IRQS_OFF_DEBUG
 	testl %ebx,%ebx				/* swapgs needed? */
 	jnz paranoid_restore
-	testl $3,CS(%rsp)
-	jnz   paranoid_userspace
-paranoid_swapgs:
 	TRACE_IRQS_IRETQ 0
 	SWAPGS_UNSAFE_STACK
 	RESTORE_ALL 8
-	jmp irq_return
+	INTERRUPT_RETURN
 paranoid_restore:
 	TRACE_IRQS_IRETQ_DEBUG 0
 	RESTORE_ALL 8
-	jmp irq_return
-paranoid_userspace:
-	GET_THREAD_INFO(%rcx)
-	movl TI_flags(%rcx),%ebx
-	andl $_TIF_WORK_MASK,%ebx
-	jz paranoid_swapgs
-	movq %rsp,%rdi			/* &pt_regs */
-	call sync_regs
-	movq %rax,%rsp			/* switch stack for scheduling */
-	testl $_TIF_NEED_RESCHED,%ebx
-	jnz paranoid_schedule
-	movl %ebx,%edx			/* arg3: thread flags */
-	TRACE_IRQS_ON
-	ENABLE_INTERRUPTS(CLBR_NONE)
-	xorl %esi,%esi 			/* arg2: oldset */
-	movq %rsp,%rdi 			/* arg1: &pt_regs */
-	call do_notify_resume
-	DISABLE_INTERRUPTS(CLBR_NONE)
-	TRACE_IRQS_OFF
-	jmp paranoid_userspace
-paranoid_schedule:
-	TRACE_IRQS_ON
-	ENABLE_INTERRUPTS(CLBR_ANY)
-	SCHEDULE_USER
-	DISABLE_INTERRUPTS(CLBR_ANY)
-	TRACE_IRQS_OFF
-	jmp paranoid_userspace
+	INTERRUPT_RETURN
 	CFI_ENDPROC
 END(paranoid_exit)
 

arch/x86/kernel/irq_32.c

@@ -69,16 +69,9 @@ static void call_on_stack(void *func, void *stack)
 		     : "memory", "cc", "edx", "ecx", "eax");
 }
 
-/* how to get the current stack pointer from C */
-#define current_stack_pointer ({		\
-	unsigned long sp;			\
-	asm("mov %%esp,%0" : "=g" (sp));	\
-	sp;					\
-})
-
 static inline void *current_stack(void)
 {
-	return (void *)(current_stack_pointer & ~(THREAD_SIZE - 1));
+	return (void *)(current_stack_pointer() & ~(THREAD_SIZE - 1));
 }
 
 static inline int
@@ -103,7 +96,7 @@ execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq)
 
 	/* Save the next esp at the bottom of the stack */
 	prev_esp = (u32 *)irqstk;
-	*prev_esp = current_stack_pointer;
+	*prev_esp = current_stack_pointer();
 
 	if (unlikely(overflow))
 		call_on_stack(print_stack_overflow, isp);
@@ -156,7 +149,7 @@ void do_softirq_own_stack(void)
 
 	/* Push the previous esp onto the stack */
 	prev_esp = (u32 *)irqstk;
-	*prev_esp = current_stack_pointer;
+	*prev_esp = current_stack_pointer();
 
 	call_on_stack(__do_softirq, isp);
 }

arch/x86/kernel/signal.c

@@ -740,12 +740,6 @@ do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags)
 {
 	user_exit();
 
-#ifdef CONFIG_X86_MCE
-	/* notify userspace of pending MCEs */
-	if (thread_info_flags & _TIF_MCE_NOTIFY)
-		mce_notify_process();
-#endif /* CONFIG_X86_64 && CONFIG_X86_MCE */
-
 	if (thread_info_flags & _TIF_UPROBE)
 		uprobe_notify_resume(regs);
 

arch/x86/kernel/traps.c

@@ -108,6 +108,77 @@ static inline void preempt_conditional_cli(struct pt_regs *regs)
 	preempt_count_dec();
 }
 
+enum ctx_state ist_enter(struct pt_regs *regs)
+{
+	/*
+	 * We are atomic because we're on the IST stack (or we're on x86_32,
+	 * in which case we still shouldn't schedule.
+	 */
+	preempt_count_add(HARDIRQ_OFFSET);
+
+	if (user_mode_vm(regs)) {
+		/* Other than that, we're just an exception. */
+		return exception_enter();
+	} else {
+		/*
+		 * We might have interrupted pretty much anything.  In
+		 * fact, if we're a machine check, we can even interrupt
+		 * NMI processing.  We don't want in_nmi() to return true,
+		 * but we need to notify RCU.
+		 */
+		rcu_nmi_enter();
+		return IN_KERNEL;  /* the value is irrelevant. */
+	}
+}
+
+void ist_exit(struct pt_regs *regs, enum ctx_state prev_state)
+{
+	preempt_count_sub(HARDIRQ_OFFSET);
+
+	if (user_mode_vm(regs))
+		return exception_exit(prev_state);
+	else
+		rcu_nmi_exit();
+}
+
+/**
+ * ist_begin_non_atomic() - begin a non-atomic section in an IST exception
+ * @regs:	regs passed to the IST exception handler
+ *
+ * IST exception handlers normally cannot schedule.  As a special
+ * exception, if the exception interrupted userspace code (i.e.
+ * user_mode_vm(regs) would return true) and the exception was not
+ * a double fault, it can be safe to schedule.  ist_begin_non_atomic()
+ * begins a non-atomic section within an ist_enter()/ist_exit() region.
+ * Callers are responsible for enabling interrupts themselves inside
+ * the non-atomic section, and callers must call is_end_non_atomic()
+ * before ist_exit().
+ */
+void ist_begin_non_atomic(struct pt_regs *regs)
+{
+	BUG_ON(!user_mode_vm(regs));
+
+	/*
+	 * Sanity check: we need to be on the normal thread stack.  This
+	 * will catch asm bugs and any attempt to use ist_preempt_enable
+	 * from double_fault.
+	 */
+	BUG_ON(((current_stack_pointer() ^ this_cpu_read_stable(kernel_stack))
+		& ~(THREAD_SIZE - 1)) != 0);
+
+	preempt_count_sub(HARDIRQ_OFFSET);
+}
+
+/**
+ * ist_end_non_atomic() - begin a non-atomic section in an IST exception
+ *
+ * Ends a non-atomic section started with ist_begin_non_atomic().
+ */
+void ist_end_non_atomic(void)
+{
+	preempt_count_add(HARDIRQ_OFFSET);
+}
+
 static nokprobe_inline int
 do_trap_no_signal(struct task_struct *tsk, int trapnr, char *str,
 		  struct pt_regs *regs,	long error_code)
@@ -251,6 +322,8 @@ dotraplinkage void do_double_fault(struct pt_regs *regs, long error_code)
 	 * end up promoting it to a doublefault.  In that case, modify
 	 * the stack to make it look like we just entered the #GP
 	 * handler from user space, similar to bad_iret.
+	 *
+	 * No need for ist_enter here because we don't use RCU.
 	 */
 	if (((long)regs->sp >> PGDIR_SHIFT) == ESPFIX_PGD_ENTRY &&
 		regs->cs == __KERNEL_CS &&
@@ -263,12 +336,12 @@ dotraplinkage void do_double_fault(struct pt_regs *regs, long error_code)
 		normal_regs->orig_ax = 0;  /* Missing (lost) #GP error code */
 		regs->ip = (unsigned long)general_protection;
 		regs->sp = (unsigned long)&normal_regs->orig_ax;
+
 		return;
 	}
 #endif
 
-	exception_enter();
-	/* Return not checked because double check cannot be ignored */
+	ist_enter(regs);  /* Discard prev_state because we won't return. */
 	notify_die(DIE_TRAP, str, regs, error_code, X86_TRAP_DF, SIGSEGV);
 
 	tsk->thread.error_code = error_code;
@@ -434,7 +507,7 @@ dotraplinkage void notrace do_int3(struct pt_regs *regs, long error_code)
 	if (poke_int3_handler(regs))
 		return;
 
-	prev_state = exception_enter();
+	prev_state = ist_enter(regs);
 #ifdef CONFIG_KGDB_LOW_LEVEL_TRAP
 	if (kgdb_ll_trap(DIE_INT3, "int3", regs, error_code, X86_TRAP_BP,
 				SIGTRAP) == NOTIFY_STOP)
@@ -460,33 +533,20 @@ dotraplinkage void notrace do_int3(struct pt_regs *regs, long error_code)
 	preempt_conditional_cli(regs);
 	debug_stack_usage_dec();
 exit:
-	exception_exit(prev_state);
+	ist_exit(regs, prev_state);
 }
 NOKPROBE_SYMBOL(do_int3);
 
 #ifdef CONFIG_X86_64
 /*
- * Help handler running on IST stack to switch back to user stack
- * for scheduling or signal handling. The actual stack switch is done in
- * entry.S
+ * Help handler running on IST stack to switch off the IST stack if the
+ * interrupted code was in user mode. The actual stack switch is done in
+ * entry_64.S
  */
 asmlinkage __visible notrace struct pt_regs *sync_regs(struct pt_regs *eregs)
 {
-	struct pt_regs *regs = eregs;
-	/* Did already sync */
-	if (eregs == (struct pt_regs *)eregs->sp)
-		;
-	/* Exception from user space */
-	else if (user_mode(eregs))
-		regs = task_pt_regs(current);
-	/*
-	 * Exception from kernel and interrupts are enabled. Move to
-	 * kernel process stack.
-	 */
-	else if (eregs->flags & X86_EFLAGS_IF)
-		regs = (struct pt_regs *)(eregs->sp -= sizeof(struct pt_regs));
-	if (eregs != regs)
-		*regs = *eregs;
+	struct pt_regs *regs = task_pt_regs(current);
+	*regs = *eregs;
 	return regs;
 }
 NOKPROBE_SYMBOL(sync_regs);
@@ -554,7 +614,7 @@ dotraplinkage void do_debug(struct pt_regs *regs, long error_code)
 	unsigned long dr6;
 	int si_code;
 
-	prev_state = exception_enter();
+	prev_state = ist_enter(regs);
 
 	get_debugreg(dr6, 6);
 
@@ -629,7 +689,7 @@ dotraplinkage void do_debug(struct pt_regs *regs, long error_code)
 	debug_stack_usage_dec();
 
 exit:
-	exception_exit(prev_state);
+	ist_exit(regs, prev_state);
 }
 NOKPROBE_SYMBOL(do_debug);
 

kernel/rcu/tree.c

@@ -759,39 +759,71 @@ void rcu_irq_enter(void)
 /**
  * rcu_nmi_enter - inform RCU of entry to NMI context
  *
- * If the CPU was idle with dynamic ticks active, and there is no
- * irq handler running, this updates rdtp->dynticks_nmi to let the
- * RCU grace-period handling know that the CPU is active.
+ * If the CPU was idle from RCU's viewpoint, update rdtp->dynticks and
+ * rdtp->dynticks_nmi_nesting to let the RCU grace-period handling know
+ * that the CPU is active.  This implementation permits nested NMIs, as
+ * long as the nesting level does not overflow an int.  (You will probably
+ * run out of stack space first.)
  */
 void rcu_nmi_enter(void)
 {
 	struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks);
+	int incby = 2;
 
-	if (rdtp->dynticks_nmi_nesting == 0 &&
-	    (atomic_read(&rdtp->dynticks) & 0x1))
-		return;
-	rdtp->dynticks_nmi_nesting++;
-	smp_mb__before_atomic();  /* Force delay from prior write. */
-	atomic_inc(&rdtp->dynticks);
-	/* CPUs seeing atomic_inc() must see later RCU read-side crit sects */
-	smp_mb__after_atomic();  /* See above. */
-	WARN_ON_ONCE(!(atomic_read(&rdtp->dynticks) & 0x1));
+	/* Complain about underflow. */
+	WARN_ON_ONCE(rdtp->dynticks_nmi_nesting < 0);
+
+	/*
+	 * If idle from RCU viewpoint, atomically increment ->dynticks
+	 * to mark non-idle and increment ->dynticks_nmi_nesting by one.
+	 * Otherwise, increment ->dynticks_nmi_nesting by two.  This means
+	 * if ->dynticks_nmi_nesting is equal to one, we are guaranteed
+	 * to be in the outermost NMI handler that interrupted an RCU-idle
+	 * period (observation due to Andy Lutomirski).
+	 */
+	if (!(atomic_read(&rdtp->dynticks) & 0x1)) {
+		smp_mb__before_atomic();  /* Force delay from prior write. */
+		atomic_inc(&rdtp->dynticks);
+		/* atomic_inc() before later RCU read-side crit sects */
+		smp_mb__after_atomic();  /* See above. */
+		WARN_ON_ONCE(!(atomic_read(&rdtp->dynticks) & 0x1));
+		incby = 1;
+	}
+	rdtp->dynticks_nmi_nesting += incby;
+	barrier();
 }
 
 /**
  * rcu_nmi_exit - inform RCU of exit from NMI context
  *
- * If the CPU was idle with dynamic ticks active, and there is no
- * irq handler running, this updates rdtp->dynticks_nmi to let the
- * RCU grace-period handling know that the CPU is no longer active.
+ * If we are returning from the outermost NMI handler that interrupted an
+ * RCU-idle period, update rdtp->dynticks and rdtp->dynticks_nmi_nesting
+ * to let the RCU grace-period handling know that the CPU is back to
+ * being RCU-idle.
  */
 void rcu_nmi_exit(void)
 {
 	struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks);
 
-	if (rdtp->dynticks_nmi_nesting == 0 ||
-	    --rdtp->dynticks_nmi_nesting != 0)
+	/*
+	 * Check for ->dynticks_nmi_nesting underflow and bad ->dynticks.
+	 * (We are exiting an NMI handler, so RCU better be paying attention
+	 * to us!)
+	 */
+	WARN_ON_ONCE(rdtp->dynticks_nmi_nesting <= 0);
+	WARN_ON_ONCE(!(atomic_read(&rdtp->dynticks) & 0x1));
+
+	/*
+	 * If the nesting level is not 1, the CPU wasn't RCU-idle, so
+	 * leave it in non-RCU-idle state.
+	 */
+	if (rdtp->dynticks_nmi_nesting != 1) {
+		rdtp->dynticks_nmi_nesting -= 2;
 		return;
+	}
+
+	/* This NMI interrupted an RCU-idle CPU, restore RCU-idleness. */
+	rdtp->dynticks_nmi_nesting = 0;
 	/* CPUs seeing atomic_inc() must see prior RCU read-side crit sects */
 	smp_mb__before_atomic();  /* See above. */
 	atomic_inc(&rdtp->dynticks);