tile: support CONTEXT_TRACKING and thus NOHZ_FULL

Add the TIF_NOHZ flag appropriately.

Add call to user_exit() on entry to do_work_pending() and on entry
to syscalls via do_syscall_trace_enter(), and also the top of
do_syscall_trace_exit() just because it's done in x86.

Add call to user_enter() at the bottom of do_work_pending() once we
have no more work to do before returning to userspace.

Wrap all the trap code in exception_enter() / exception_exit().
Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>

arch/tile/Kconfig

@@ -27,6 +27,7 @@ config TILE
 	select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE
 	select HAVE_DEBUG_STACKOVERFLOW
 	select ARCH_WANT_FRAME_POINTERS
+	select HAVE_CONTEXT_TRACKING
 
 # FIXME: investigate whether we need/want these options.
 #	select HAVE_IOREMAP_PROT

arch/tile/include/asm/thread_info.h

@@ -126,6 +126,7 @@ extern void _cpu_idle(void);
 #define TIF_NOTIFY_RESUME	8	/* callback before returning to user */
 #define TIF_SYSCALL_TRACEPOINT	9	/* syscall tracepoint instrumentation */
 #define TIF_POLLING_NRFLAG	10	/* idle is polling for TIF_NEED_RESCHED */
+#define TIF_NOHZ		11	/* in adaptive nohz mode */
 
 #define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
 #define _TIF_NEED_RESCHED	(1<<TIF_NEED_RESCHED)
@@ -138,14 +139,16 @@ extern void _cpu_idle(void);
 #define _TIF_NOTIFY_RESUME	(1<<TIF_NOTIFY_RESUME)
 #define _TIF_SYSCALL_TRACEPOINT	(1<<TIF_SYSCALL_TRACEPOINT)
 #define _TIF_POLLING_NRFLAG	(1<<TIF_POLLING_NRFLAG)
+#define _TIF_NOHZ		(1<<TIF_NOHZ)
 
 /* Work to do on any return to user space. */
 #define _TIF_ALLWORK_MASK \
-  (_TIF_SIGPENDING|_TIF_NEED_RESCHED|_TIF_SINGLESTEP|\
-   _TIF_ASYNC_TLB|_TIF_NOTIFY_RESUME)
+	(_TIF_SIGPENDING | _TIF_NEED_RESCHED | _TIF_SINGLESTEP | \
+	 _TIF_ASYNC_TLB | _TIF_NOTIFY_RESUME | _TIF_NOHZ)
 
 /* Work to do at syscall entry. */
-#define _TIF_SYSCALL_ENTRY_WORK (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_TRACEPOINT)
+#define _TIF_SYSCALL_ENTRY_WORK \
+	(_TIF_SYSCALL_TRACE | _TIF_SYSCALL_TRACEPOINT | _TIF_NOHZ)
 
 /* Work to do at syscall exit. */
 #define _TIF_SYSCALL_EXIT_WORK (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_TRACEPOINT)

arch/tile/kernel/process.c

@@ -27,6 +27,7 @@
 #include <linux/kernel.h>
 #include <linux/tracehook.h>
 #include <linux/signal.h>
+#include <linux/context_tracking.h>
 #include <asm/stack.h>
 #include <asm/switch_to.h>
 #include <asm/homecache.h>
@@ -474,6 +475,8 @@ int do_work_pending(struct pt_regs *regs, u32 thread_info_flags)
 	if (!user_mode(regs))
 		return 0;
 
+	user_exit();
+
 	/* Enable interrupts; they are disabled again on return to caller. */
 	local_irq_enable();
 
@@ -496,11 +499,12 @@ int do_work_pending(struct pt_regs *regs, u32 thread_info_flags)
 		tracehook_notify_resume(regs);
 		return 1;
 	}
-	if (thread_info_flags & _TIF_SINGLESTEP) {
+	if (thread_info_flags & _TIF_SINGLESTEP)
 		single_step_once(regs);
+
+	user_enter();
+
 	return 0;
-	}
-	panic("work_pending: bad flags %#x\n", thread_info_flags);
 }
 
 unsigned long get_wchan(struct task_struct *p)

arch/tile/kernel/ptrace.c

@@ -22,6 +22,7 @@
 #include <linux/regset.h>
 #include <linux/elf.h>
 #include <linux/tracehook.h>
+#include <linux/context_tracking.h>
 #include <asm/traps.h>
 #include <arch/chip.h>
 
@@ -252,12 +253,21 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
 
 int do_syscall_trace_enter(struct pt_regs *regs)
 {
-	if (test_thread_flag(TIF_SYSCALL_TRACE)) {
+	u32 work = ACCESS_ONCE(current_thread_info()->flags);
+
+	/*
+	 * If TIF_NOHZ is set, we are required to call user_exit() before
+	 * doing anything that could touch RCU.
+	 */
+	if (work & _TIF_NOHZ)
+		user_exit();
+
+	if (work & _TIF_SYSCALL_TRACE) {
 		if (tracehook_report_syscall_entry(regs))
 			regs->regs[TREG_SYSCALL_NR] = -1;
 	}
 
-	if (test_thread_flag(TIF_SYSCALL_TRACEPOINT))
+	if (work & _TIF_SYSCALL_TRACEPOINT)
 		trace_sys_enter(regs, regs->regs[TREG_SYSCALL_NR]);
 
 	return regs->regs[TREG_SYSCALL_NR];
@@ -267,6 +277,12 @@ void do_syscall_trace_exit(struct pt_regs *regs)
 {
 	long errno;
 
+	/*
+	 * We may come here right after calling schedule_user()
+	 * in which case we can be in RCU user mode.
+	 */
+	user_exit();
+
 	/*
 	 * The standard tile calling convention returns the value (or negative
 	 * errno) in r0, and zero (or positive errno) in r1.
@@ -303,5 +319,7 @@ void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs)
 /* Handle synthetic interrupt delivered only by the simulator. */
 void __kprobes do_breakpoint(struct pt_regs* regs, int fault_num)
 {
+	enum ctx_state prev_state = exception_enter();
 	send_sigtrap(current, regs);
+	exception_exit(prev_state);
 }

arch/tile/kernel/single_step.c

@@ -23,6 +23,7 @@
 #include <linux/types.h>
 #include <linux/err.h>
 #include <linux/prctl.h>
+#include <linux/context_tracking.h>
 #include <asm/cacheflush.h>
 #include <asm/traps.h>
 #include <asm/uaccess.h>
@@ -738,6 +739,7 @@ static DEFINE_PER_CPU(unsigned long, ss_saved_pc);
 
 void gx_singlestep_handle(struct pt_regs *regs, int fault_num)
 {
+	enum ctx_state prev_state = exception_enter();
 	unsigned long *ss_pc = this_cpu_ptr(&ss_saved_pc);
 	struct thread_info *info = (void *)current_thread_info();
 	int is_single_step = test_ti_thread_flag(info, TIF_SINGLESTEP);
@@ -754,6 +756,7 @@ void gx_singlestep_handle(struct pt_regs *regs, int fault_num)
 		__insn_mtspr(SPR_SINGLE_STEP_CONTROL_K, control);
 		send_sigtrap(current, regs);
 	}
+	exception_exit(prev_state);
 }
 
 

arch/tile/kernel/traps.c

@@ -20,6 +20,7 @@
 #include <linux/reboot.h>
 #include <linux/uaccess.h>
 #include <linux/ptrace.h>
+#include <linux/context_tracking.h>
 #include <asm/stack.h>
 #include <asm/traps.h>
 #include <asm/setup.h>
@@ -253,6 +254,7 @@ static int do_bpt(struct pt_regs *regs)
 void __kprobes do_trap(struct pt_regs *regs, int fault_num,
 		       unsigned long reason)
 {
+	enum ctx_state prev_state = exception_enter();
 	siginfo_t info = { 0 };
 	int signo, code;
 	unsigned long address = 0;
@@ -261,7 +263,7 @@ void __kprobes do_trap(struct pt_regs *regs, int fault_num,
 
 	/* Handle breakpoints, etc. */
 	if (is_kernel && fault_num == INT_ILL && do_bpt(regs))
-		return;
+		goto done;
 
 	/* Re-enable interrupts, if they were previously enabled. */
 	if (!(regs->flags & PT_FLAGS_DISABLE_IRQ))
@@ -275,7 +277,7 @@ void __kprobes do_trap(struct pt_regs *regs, int fault_num,
 		const char *name;
 		char buf[100];
 		if (fixup_exception(regs))  /* ILL_TRANS or UNALIGN_DATA */
-			return;
+			goto done;
 		if (fault_num >= 0 &&
 		    fault_num < ARRAY_SIZE(int_name) &&
 		    int_name[fault_num] != NULL)
@@ -294,7 +296,6 @@ void __kprobes do_trap(struct pt_regs *regs, int fault_num,
 			 fault_num, name, regs->pc, buf);
 		show_regs(regs);
 		do_exit(SIGKILL);  /* FIXME: implement i386 die() */
-		return;
 	}
 
 	switch (fault_num) {
@@ -308,7 +309,6 @@ void __kprobes do_trap(struct pt_regs *regs, int fault_num,
 			pr_err("Unreadable instruction for INT_ILL: %#lx\n",
 			       regs->pc);
 			do_exit(SIGKILL);
-			return;
 		}
 		if (!special_ill(instr, &signo, &code)) {
 			signo = SIGILL;
@@ -319,7 +319,7 @@ void __kprobes do_trap(struct pt_regs *regs, int fault_num,
 	case INT_GPV:
 #if CHIP_HAS_TILE_DMA()
 		if (retry_gpv(reason))
-			return;
+			goto done;
 #endif
 		/*FALLTHROUGH*/
 	case INT_UDN_ACCESS:
@@ -346,7 +346,7 @@ void __kprobes do_trap(struct pt_regs *regs, int fault_num,
 			if (!state ||
 			    (void __user *)(regs->pc) != state->buffer) {
 				single_step_once(regs);
-				return;
+				goto done;
 			}
 		}
 #endif
@@ -380,7 +380,6 @@ void __kprobes do_trap(struct pt_regs *regs, int fault_num,
 #endif
 	default:
 		panic("Unexpected do_trap interrupt number %d", fault_num);
-		return;
 	}
 
 	info.si_signo = signo;
@@ -391,6 +390,9 @@ void __kprobes do_trap(struct pt_regs *regs, int fault_num,
 	if (signo != SIGTRAP)
 		trace_unhandled_signal("trap", regs, address, signo);
 	force_sig_info(signo, &info, current);
+
+done:
+	exception_exit(prev_state);
 }
 
 void kernel_double_fault(int dummy, ulong pc, ulong lr, ulong sp, ulong r52)

arch/tile/kernel/unaligned.c

@@ -25,6 +25,7 @@
 #include <linux/module.h>
 #include <linux/compat.h>
 #include <linux/prctl.h>
+#include <linux/context_tracking.h>
 #include <asm/cacheflush.h>
 #include <asm/traps.h>
 #include <asm/uaccess.h>
@@ -1448,6 +1449,7 @@ void jit_bundle_gen(struct pt_regs *regs, tilegx_bundle_bits bundle,
 
 void do_unaligned(struct pt_regs *regs, int vecnum)
 {
+	enum ctx_state prev_state = exception_enter();
 	tilegx_bundle_bits __user  *pc;
 	tilegx_bundle_bits bundle;
 	struct thread_info *info = current_thread_info();
@@ -1487,12 +1489,11 @@ void do_unaligned(struct pt_regs *regs, int vecnum)
 						(int)unaligned_fixup,
 						(unsigned long long)regs->ex1,
 						(unsigned long long)regs->pc);
-				return;
-			}
+			} else {
 				/* Not fixable. Go panic. */
 				panic("Unalign exception in Kernel. pc=%lx",
 				      regs->pc);
-			return;
+			}
 		} else {
 			/*
 			 * Try to fix the exception. If we can't, panic the
@@ -1501,8 +1502,8 @@ void do_unaligned(struct pt_regs *regs, int vecnum)
 			bundle = GX_INSN_BSWAP(
 				*((tilegx_bundle_bits *)(regs->pc)));
 			jit_bundle_gen(regs, bundle, align_ctl);
-			return;
 		}
+		goto done;
 	}
 
 	/*
@@ -1526,7 +1527,7 @@ void do_unaligned(struct pt_regs *regs, int vecnum)
 
 		trace_unhandled_signal("unaligned fixup trap", regs, 0, SIGBUS);
 		force_sig_info(info.si_signo, &info, current);
-		return;
+		goto done;
 	}
 
 
@@ -1543,7 +1544,7 @@ void do_unaligned(struct pt_regs *regs, int vecnum)
 		trace_unhandled_signal("segfault in unalign fixup", regs,
 				       (unsigned long)info.si_addr, SIGSEGV);
 		force_sig_info(info.si_signo, &info, current);
-		return;
+		goto done;
 	}
 
 	if (!info->unalign_jit_base) {
@@ -1578,7 +1579,7 @@ void do_unaligned(struct pt_regs *regs, int vecnum)
 
 		if (IS_ERR((void __force *)user_page)) {
 			pr_err("Out of kernel pages trying do_mmap\n");
-			return;
+			goto done;
 		}
 
 		/* Save the address in the thread_info struct */
@@ -1591,6 +1592,9 @@ void do_unaligned(struct pt_regs *regs, int vecnum)
 
 	/* Generate unalign JIT */
 	jit_bundle_gen(regs, GX_INSN_BSWAP(bundle), align_ctl);
+
+done:
+	exception_exit(prev_state);
 }
 
 #endif /* __tilegx__ */

arch/tile/mm/fault.c

@@ -35,6 +35,7 @@
 #include <linux/syscalls.h>
 #include <linux/uaccess.h>
 #include <linux/kdebug.h>
+#include <linux/context_tracking.h>
 
 #include <asm/pgalloc.h>
 #include <asm/sections.h>
@@ -702,6 +703,7 @@ void do_page_fault(struct pt_regs *regs, int fault_num,
 		   unsigned long address, unsigned long write)
 {
 	int is_page_fault;
+	enum ctx_state prev_state = exception_enter();
 
 #ifdef CONFIG_KPROBES
 	/*
@@ -711,7 +713,7 @@ void do_page_fault(struct pt_regs *regs, int fault_num,
 	 */
 	if (notify_die(DIE_PAGE_FAULT, "page fault", regs, -1,
 		       regs->faultnum, SIGSEGV) == NOTIFY_STOP)
-		return;
+		goto done;
 #endif
 
 #ifdef __tilegx__
@@ -750,7 +752,6 @@ void do_page_fault(struct pt_regs *regs, int fault_num,
 				 current->comm, current->pid, pc, address);
 			show_regs(regs);
 			do_group_exit(SIGKILL);
-			return;
 		}
 	}
 #else
@@ -834,12 +835,15 @@ void do_page_fault(struct pt_regs *regs, int fault_num,
 			async->is_fault = is_page_fault;
 			async->is_write = write;
 			async->address = address;
-			return;
+			goto done;
 		}
 	}
 #endif
 
 	handle_page_fault(regs, fault_num, is_page_fault, address, write);
+
+done:
+	exception_exit(prev_state);
 }