Rework of the X86 irq stack handling:

The irq stack switching was moved out of the ASM entry code in course of
   the entry code consolidation. It ended up being suboptimal in various
   ways.
 
   - Make the stack switching inline so the stackpointer manipulation is not
     longer at an easy to find place.
 
   - Get rid of the unnecessary indirect call.
 
   - Avoid the double stack switching in interrupt return and reuse the
     interrupt stack for softirq handling.
 
   - A objtool fix for CONFIG_FRAME_POINTER=y builds where it got confused
     about the stack pointer manipulation.
 -----BEGIN PGP SIGNATURE-----
 
 iQJHBAABCgAxFiEEQp8+kY+LLUocC4bMphj1TA10mKEFAmA21OcTHHRnbHhAbGlu
 dXRyb25peC5kZQAKCRCmGPVMDXSYoaX0D/9S0ud6oqbsIvI8LwhvYub63a2cjKP9
 liHAJ7xwMYYVwzf0skwsPb/QE6+onCzdq0upJkgG/gEYm2KbiaMWZ4GgHdj0O7ER
 qXKJONDd36AGxSEdaVzLY5kPuD/mkomGk5QdaZaTmjruthkNzg4y/N2wXUBIMZR0
 FdpSpp5fGspSZCn/DXDx6FjClwpLI53VclvDs6DcZ2DIBA0K+F/cSLb1UQoDLE1U
 hxGeuNa+GhKeeZ5C+q5giho1+ukbwtjMW9WnKHAVNiStjm0uzdqq7ERGi/REvkcB
 LY62u5uOSW1zIBMmzUjDDQEqvypB0iFxFCpN8g9sieZjA0zkaUioRTQyR+YIQ8Cp
 l8LLir0dVQivR1bHghHDKQJUpdw/4zvDj4mMH10XHqbcOtIxJDOJHC5D00ridsAz
 OK0RlbAJBl9FTdLNfdVReBCoehYAO8oefeyMAG12nZeSh5XVUWl238rvzmzIYNhG
 cEtkSx2wIUNEA+uSuI+xvfmwpxL7voTGvqmiRDCAFxyO7Bl/GBu9OEBFA1eOvHB+
 +wTmPDMswRetQNh4QCRXzk1JzP1Wk5CobUL9iinCWFoTJmnsPPSOWlosN6ewaNXt
 kYFpRLy5xt9EP7dlfgBSjiRlthDhTdMrFjD5bsy1vdm1w7HKUo82lHa4O8Hq3PHS
 tinKICUqRsbjig==
 =Sqr1
 -----END PGP SIGNATURE-----

Merge tag 'x86-entry-2021-02-24' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 irq entry updates from Thomas Gleixner:
 "The irq stack switching was moved out of the ASM entry code in course
  of the entry code consolidation. It ended up being suboptimal in
  various ways.

  This reworks the X86 irq stack handling:

   - Make the stack switching inline so the stackpointer manipulation is
     not longer at an easy to find place.

   - Get rid of the unnecessary indirect call.

   - Avoid the double stack switching in interrupt return and reuse the
     interrupt stack for softirq handling.

   - A objtool fix for CONFIG_FRAME_POINTER=y builds where it got
     confused about the stack pointer manipulation"

* tag 'x86-entry-2021-02-24' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  objtool: Fix stack-swizzle for FRAME_POINTER=y
  um: Enforce the usage of asm-generic/softirq_stack.h
  x86/softirq/64: Inline do_softirq_own_stack()
  softirq: Move do_softirq_own_stack() to generic asm header
  softirq: Move __ARCH_HAS_DO_SOFTIRQ to Kconfig
  x86: Select CONFIG_HAVE_IRQ_EXIT_ON_IRQ_STACK
  x86/softirq: Remove indirection in do_softirq_own_stack()
  x86/entry: Use run_sysvec_on_irqstack_cond() for XEN upcall
  x86/entry: Convert device interrupts to inline stack switching
  x86/entry: Convert system vectors to irq stack macro
  x86/irq: Provide macro for inlining irq stack switching
  x86/apic: Split out spurious handling code
  x86/irq/64: Adjust the per CPU irq stack pointer by 8
  x86/irq: Sanitize irq stack tracking
  x86/entry: Fix instrumentation annotation

arch/Kconfig

@@ -821,6 +821,12 @@ config HAVE_IRQ_EXIT_ON_IRQ_STACK
 	  This spares a stack switch and improves cache usage on softirq
 	  processing.
 
+config HAVE_SOFTIRQ_ON_OWN_STACK
+	bool
+	help
+	  Architecture provides a function to run __do_softirq() on a
+	  seperate stack.
+
 config PGTABLE_LEVELS
 	int
 	default 2

arch/parisc/Kconfig

@@ -63,6 +63,7 @@ config PARISC
 	select HAVE_FTRACE_MCOUNT_RECORD if HAVE_DYNAMIC_FTRACE
 	select HAVE_KPROBES_ON_FTRACE
 	select HAVE_DYNAMIC_FTRACE_WITH_REGS
+	select HAVE_SOFTIRQ_ON_OWN_STACK if IRQSTACKS
 	select SET_FS
 
 	help

arch/parisc/include/asm/hardirq.h

@@ -12,10 +12,6 @@
 #include <linux/threads.h>
 #include <linux/irq.h>
 
-#ifdef CONFIG_IRQSTACKS
-#define __ARCH_HAS_DO_SOFTIRQ
-#endif
-
 typedef struct {
 	unsigned int __softirq_pending;
 	unsigned int kernel_stack_usage;

arch/parisc/kernel/irq.c

@@ -17,6 +17,7 @@
 #include <linux/types.h>
 #include <asm/io.h>
 
+#include <asm/softirq_stack.h>
 #include <asm/smp.h>
 #include <asm/ldcw.h>
 

arch/powerpc/Kconfig

@@ -235,6 +235,7 @@ config PPC
 	select MMU_GATHER_PAGE_SIZE
 	select HAVE_REGS_AND_STACK_ACCESS_API
 	select HAVE_RELIABLE_STACKTRACE		if PPC_BOOK3S_64 && CPU_LITTLE_ENDIAN
+	select HAVE_SOFTIRQ_ON_OWN_STACK
 	select HAVE_SYSCALL_TRACEPOINTS
 	select HAVE_VIRT_CPU_ACCOUNTING
 	select HAVE_IRQ_TIME_ACCOUNTING

arch/powerpc/include/asm/irq.h

@@ -37,8 +37,6 @@ extern int distribute_irqs;
 
 struct pt_regs;
 
-#define __ARCH_HAS_DO_SOFTIRQ
-
 #if defined(CONFIG_BOOKE) || defined(CONFIG_40x)
 /*
  * Per-cpu stacks for handling critical, debug and machine check

arch/powerpc/kernel/irq.c

@@ -66,6 +66,7 @@
 #include <asm/livepatch.h>
 #include <asm/asm-prototypes.h>
 #include <asm/hw_irq.h>
+#include <asm/softirq_stack.h>
 
 #ifdef CONFIG_PPC64
 #include <asm/paca.h>

arch/s390/Kconfig

@@ -184,6 +184,7 @@ config S390
 	select HAVE_REGS_AND_STACK_ACCESS_API
 	select HAVE_RELIABLE_STACKTRACE
 	select HAVE_RSEQ
+	select HAVE_SOFTIRQ_ON_OWN_STACK
 	select HAVE_SYSCALL_TRACEPOINTS
 	select HAVE_VIRT_CPU_ACCOUNTING
 	select HAVE_VIRT_CPU_ACCOUNTING_IDLE

arch/s390/include/asm/hardirq.h

@@ -18,7 +18,6 @@
 #define or_softirq_pending(x)  (S390_lowcore.softirq_pending |= (x))
 
 #define __ARCH_IRQ_STAT
-#define __ARCH_HAS_DO_SOFTIRQ
 #define __ARCH_IRQ_EXIT_IRQS_DISABLED
 
 static inline void ack_bad_irq(unsigned int irq)

arch/s390/kernel/irq.c

@@ -28,6 +28,7 @@
 #include <asm/irq.h>
 #include <asm/hw_irq.h>
 #include <asm/stacktrace.h>
+#include <asm/softirq_stack.h>
 #include "entry.h"
 
 DEFINE_PER_CPU_SHARED_ALIGNED(struct irq_stat, irq_stat);

arch/sh/Kconfig

@@ -54,6 +54,7 @@ config SUPERH
 	select HAVE_PERF_EVENTS
 	select HAVE_REGS_AND_STACK_ACCESS_API
 	select HAVE_UID16
+	select HAVE_SOFTIRQ_ON_OWN_STACK if IRQSTACKS
 	select HAVE_STACKPROTECTOR
 	select HAVE_SYSCALL_TRACEPOINTS
 	select IRQ_FORCED_THREADING

arch/sh/include/asm/irq.h

@@ -51,7 +51,6 @@ asmlinkage int do_IRQ(unsigned int irq, struct pt_regs *regs);
 #ifdef CONFIG_IRQSTACKS
 extern void irq_ctx_init(int cpu);
 extern void irq_ctx_exit(int cpu);
-# define __ARCH_HAS_DO_SOFTIRQ
 #else
 # define irq_ctx_init(cpu) do { } while (0)
 # define irq_ctx_exit(cpu) do { } while (0)

arch/sh/kernel/irq.c

@@ -20,6 +20,7 @@
 #include <linux/uaccess.h>
 #include <asm/thread_info.h>
 #include <cpu/mmu_context.h>
+#include <asm/softirq_stack.h>
 
 atomic_t irq_err_count;
 

arch/sparc/Kconfig

@@ -96,6 +96,7 @@ config SPARC64
 	select ARCH_HAS_PTE_SPECIAL
 	select PCI_DOMAINS if PCI
 	select ARCH_HAS_GIGANTIC_PAGE
+	select HAVE_SOFTIRQ_ON_OWN_STACK
 
 config ARCH_PROC_KCORE_TEXT
 	def_bool y

arch/sparc/include/asm/irq_64.h

@@ -93,7 +93,6 @@ void arch_trigger_cpumask_backtrace(const struct cpumask *mask,
 
 extern void *hardirq_stack[NR_CPUS];
 extern void *softirq_stack[NR_CPUS];
-#define __ARCH_HAS_DO_SOFTIRQ
 
 #define NO_IRQ		0xffffffff
 

arch/sparc/kernel/irq_64.c

@@ -42,6 +42,7 @@
 #include <asm/head.h>
 #include <asm/hypervisor.h>
 #include <asm/cacheflush.h>
+#include <asm/softirq_stack.h>
 
 #include "entry.h"
 #include "cpumap.h"

arch/um/include/asm/Kbuild

@@ -20,6 +20,7 @@ generic-y += param.h
 generic-y += pci.h
 generic-y += percpu.h
 generic-y += preempt.h
+generic-y += softirq_stack.h
 generic-y += switch_to.h
 generic-y += topology.h
 generic-y += trace_clock.h

arch/x86/Kconfig

@@ -191,6 +191,7 @@ config X86
 	select HAVE_HW_BREAKPOINT
 	select HAVE_IDE
 	select HAVE_IOREMAP_PROT
+	select HAVE_IRQ_EXIT_ON_IRQ_STACK	if X86_64
 	select HAVE_IRQ_TIME_ACCOUNTING
 	select HAVE_KERNEL_BZIP2
 	select HAVE_KERNEL_GZIP
@@ -223,6 +224,7 @@ config X86
 	select HAVE_REGS_AND_STACK_ACCESS_API
 	select HAVE_RELIABLE_STACKTRACE		if X86_64 && (UNWINDER_FRAME_POINTER || UNWINDER_ORC) && STACK_VALIDATION
 	select HAVE_FUNCTION_ARG_ACCESS_API
+	select HAVE_SOFTIRQ_ON_OWN_STACK
 	select HAVE_STACKPROTECTOR		if CC_HAS_SANE_STACKPROTECTOR
 	select HAVE_STACK_VALIDATION		if X86_64
 	select HAVE_STATIC_CALL

arch/x86/entry/common.c

@@ -249,30 +249,23 @@ static __always_inline bool get_and_clear_inhcall(void) { return false; }
 static __always_inline void restore_inhcall(bool inhcall) { }
 #endif
 
-static void __xen_pv_evtchn_do_upcall(void)
+static void __xen_pv_evtchn_do_upcall(struct pt_regs *regs)
 {
-	irq_enter_rcu();
+	struct pt_regs *old_regs = set_irq_regs(regs);
+
 	inc_irq_stat(irq_hv_callback_count);
 
 	xen_hvm_evtchn_do_upcall();
 
-	irq_exit_rcu();
+	set_irq_regs(old_regs);
 }
 
 __visible noinstr void xen_pv_evtchn_do_upcall(struct pt_regs *regs)
 {
-	struct pt_regs *old_regs;
+	irqentry_state_t state = irqentry_enter(regs);
 	bool inhcall;
-	irqentry_state_t state;
 
-	state = irqentry_enter(regs);
+	run_sysvec_on_irqstack_cond(__xen_pv_evtchn_do_upcall, regs);
-	old_regs = set_irq_regs(regs);
-
-	instrumentation_begin();
-	run_on_irqstack_cond(__xen_pv_evtchn_do_upcall, regs);
-	instrumentation_begin();
-
-	set_irq_regs(old_regs);
 
 	inhcall = get_and_clear_inhcall();
 	if (inhcall && !WARN_ON_ONCE(state.exit_rcu)) {

arch/x86/entry/entry_64.S

@@ -754,47 +754,6 @@ SYM_CODE_START_LOCAL_NOALIGN(.Lbad_gs)
 SYM_CODE_END(.Lbad_gs)
 	.previous
 
-/*
- * rdi: New stack pointer points to the top word of the stack
- * rsi: Function pointer
- * rdx: Function argument (can be NULL if none)
- */
-SYM_FUNC_START(asm_call_on_stack)
-SYM_INNER_LABEL(asm_call_sysvec_on_stack, SYM_L_GLOBAL)
-SYM_INNER_LABEL(asm_call_irq_on_stack, SYM_L_GLOBAL)
-	/*
-	 * Save the frame pointer unconditionally. This allows the ORC
-	 * unwinder to handle the stack switch.
-	 */
-	pushq		%rbp
-	mov		%rsp, %rbp
-
-	/*
-	 * The unwinder relies on the word at the top of the new stack
-	 * page linking back to the previous RSP.
-	 */
-	mov		%rsp, (%rdi)
-	mov		%rdi, %rsp
-	/* Move the argument to the right place */
-	mov		%rdx, %rdi
-
-1:
-	.pushsection .discard.instr_begin
-	.long 1b - .
-	.popsection
-
-	CALL_NOSPEC	rsi
-
-2:
-	.pushsection .discard.instr_end
-	.long 2b - .
-	.popsection
-
-	/* Restore the previous stack pointer from RBP. */
-	leaveq
-	ret
-SYM_FUNC_END(asm_call_on_stack)
-
 #ifdef CONFIG_XEN_PV
 /*
  * A note on the "critical region" in our callback handler.

arch/x86/include/asm/idtentry.h

@@ -187,23 +187,22 @@ __visible noinstr void func(struct pt_regs *regs, unsigned long error_code)
  * has to be done in the function body if necessary.
  */
 #define DEFINE_IDTENTRY_IRQ(func)					\
-static __always_inline void __##func(struct pt_regs *regs, u8 vector);	\
+static void __##func(struct pt_regs *regs, u32 vector);			\
 									\
 __visible noinstr void func(struct pt_regs *regs,			\
 			    unsigned long error_code)			\
 {									\
 	irqentry_state_t state = irqentry_enter(regs);			\
+	u32 vector = (u32)(u8)error_code;				\
 									\
 	instrumentation_begin();					\
-	irq_enter_rcu();						\
 	kvm_set_cpu_l1tf_flush_l1d();					\
-	__##func (regs, (u8)error_code);				\
+	run_irq_on_irqstack_cond(__##func, regs, vector);		\
-	irq_exit_rcu();							\
 	instrumentation_end();						\
 	irqentry_exit(regs, state);					\
 }									\
 									\
-static __always_inline void __##func(struct pt_regs *regs, u8 vector)
+static noinline void __##func(struct pt_regs *regs, u32 vector)
 
 /**
  * DECLARE_IDTENTRY_SYSVEC - Declare functions for system vector entry points
@@ -237,10 +236,8 @@ __visible noinstr void func(struct pt_regs *regs)			\
 	irqentry_state_t state = irqentry_enter(regs);			\
 									\
 	instrumentation_begin();					\
-	irq_enter_rcu();						\
 	kvm_set_cpu_l1tf_flush_l1d();					\
 	run_sysvec_on_irqstack_cond(__##func, regs);			\
-	irq_exit_rcu();							\
 	instrumentation_end();						\
 	irqentry_exit(regs, state);					\
 }									\

arch/x86/include/asm/irq.h

@@ -25,8 +25,6 @@ static inline int irq_canonicalize(int irq)
 
 extern int irq_init_percpu_irqstack(unsigned int cpu);
 
-#define __ARCH_HAS_DO_SOFTIRQ
-
 struct irq_desc;
 
 extern void fixup_irqs(void);

arch/x86/include/asm/irq_stack.h

@@ -7,100 +7,217 @@
 #include <asm/processor.h>
 
 #ifdef CONFIG_X86_64
-static __always_inline bool irqstack_active(void)
+
-{
+/*
-	return __this_cpu_read(irq_count) != -1;
+ * Macro to inline switching to an interrupt stack and invoking function
+ * calls from there. The following rules apply:
+ *
+ * - Ordering:
+ *
+ *   1. Write the stack pointer into the top most place of the irq
+ *	stack. This ensures that the various unwinders can link back to the
+ *	original stack.
+ *
+ *   2. Switch the stack pointer to the top of the irq stack.
+ *
+ *   3. Invoke whatever needs to be done (@asm_call argument)
+ *
+ *   4. Pop the original stack pointer from the top of the irq stack
+ *	which brings it back to the original stack where it left off.
+ *
+ * - Function invocation:
+ *
+ *   To allow flexible usage of the macro, the actual function code including
+ *   the store of the arguments in the call ABI registers is handed in via
+ *   the @asm_call argument.
+ *
+ * - Local variables:
+ *
+ *   @tos:
+ *	The @tos variable holds a pointer to the top of the irq stack and
+ *	_must_ be allocated in a non-callee saved register as this is a
+ *	restriction coming from objtool.
+ *
+ *	Note, that (tos) is both in input and output constraints to ensure
+ *	that the compiler does not assume that R11 is left untouched in
+ *	case this macro is used in some place where the per cpu interrupt
+ *	stack pointer is used again afterwards
+ *
+ * - Function arguments:
+ *	The function argument(s), if any, have to be defined in register
+ *	variables at the place where this is invoked. Storing the
+ *	argument(s) in the proper register(s) is part of the @asm_call
+ *
+ * - Constraints:
+ *
+ *   The constraints have to be done very carefully because the compiler
+ *   does not know about the assembly call.
+ *
+ *   output:
+ *     As documented already above the @tos variable is required to be in
+ *     the output constraints to make the compiler aware that R11 cannot be
+ *     reused after the asm() statement.
+ *
+ *     For builds with CONFIG_UNWIND_FRAME_POINTER ASM_CALL_CONSTRAINT is
+ *     required as well as this prevents certain creative GCC variants from
+ *     misplacing the ASM code.
+ *
+ *  input:
+ *    - func:
+ *	  Immediate, which tells the compiler that the function is referenced.
+ *
+ *    - tos:
+ *	  Register. The actual register is defined by the variable declaration.
+ *
+ *    - function arguments:
+ *	  The constraints are handed in via the 'argconstr' argument list. They
+ *	  describe the register arguments which are used in @asm_call.
+ *
+ *  clobbers:
+ *     Function calls can clobber anything except the callee-saved
+ *     registers. Tell the compiler.
+ */
+#define call_on_irqstack(func, asm_call, argconstr...)			\
+{									\
+	register void *tos asm("r11");					\
+									\
+	tos = ((void *)__this_cpu_read(hardirq_stack_ptr));		\
+									\
+	asm_inline volatile(						\
+	"movq	%%rsp, (%[tos])				\n"		\
+	"movq	%[tos], %%rsp				\n"		\
+									\
+	asm_call							\
+									\
+	"popq	%%rsp					\n"		\
+									\
+	: "+r" (tos), ASM_CALL_CONSTRAINT				\
+	: [__func] "i" (func), [tos] "r" (tos) argconstr		\
+	: "cc", "rax", "rcx", "rdx", "rsi", "rdi", "r8", "r9", "r10",	\
+	  "memory"							\
+	);								\
 }
 
-void asm_call_on_stack(void *sp, void (*func)(void), void *arg);
+/* Macros to assert type correctness for run_*_on_irqstack macros */
-void asm_call_sysvec_on_stack(void *sp, void (*func)(struct pt_regs *regs),
+#define assert_function_type(func, proto)				\
-			      struct pt_regs *regs);
+	static_assert(__builtin_types_compatible_p(typeof(&func), proto))
-void asm_call_irq_on_stack(void *sp, void (*func)(struct irq_desc *desc),
-			   struct irq_desc *desc);
 
-static __always_inline void __run_on_irqstack(void (*func)(void))
+#define assert_arg_type(arg, proto)					\
-{
+	static_assert(__builtin_types_compatible_p(typeof(arg), proto))
-	void *tos = __this_cpu_read(hardirq_stack_ptr);
 
-	__this_cpu_add(irq_count, 1);
+/*
-	asm_call_on_stack(tos - 8, func, NULL);
+ * Macro to invoke system vector and device interrupt C handlers.
-	__this_cpu_sub(irq_count, 1);
+ */
+#define call_on_irqstack_cond(func, regs, asm_call, constr, c_args...)	\
+{									\
+	/*								\
+	 * User mode entry and interrupt on the irq stack do not	\
+	 * switch stacks. If from user mode the task stack is empty.	\
+	 */								\
+	if (user_mode(regs) || __this_cpu_read(hardirq_stack_inuse)) {	\
+		irq_enter_rcu();					\
+		func(c_args);						\
+		irq_exit_rcu();						\
+	} else {							\
+		/*							\
+		 * Mark the irq stack inuse _before_ and unmark _after_	\
+		 * switching stacks. Interrupts are disabled in both	\
+		 * places. Invoke the stack switch macro with the call	\
+		 * sequence which matches the above direct invocation.	\
+		 */							\
+		__this_cpu_write(hardirq_stack_inuse, true);		\
+		call_on_irqstack(func, asm_call, constr);		\
+		__this_cpu_write(hardirq_stack_inuse, false);		\
+	}								\
 }
 
-static __always_inline void
+/*
-__run_sysvec_on_irqstack(void (*func)(struct pt_regs *regs),
+ * Function call sequence for __call_on_irqstack() for system vectors.
-			 struct pt_regs *regs)
+ *
-{
+ * Note that irq_enter_rcu() and irq_exit_rcu() do not use the input
-	void *tos = __this_cpu_read(hardirq_stack_ptr);
+ * mechanism because these functions are global and cannot be optimized out
+ * when compiling a particular source file which uses one of these macros.
+ *
+ * The argument (regs) does not need to be pushed or stashed in a callee
+ * saved register to be safe vs. the irq_enter_rcu() call because the
+ * clobbers already prevent the compiler from storing it in a callee
+ * clobbered register. As the compiler has to preserve @regs for the final
+ * call to idtentry_exit() anyway, it's likely that it does not cause extra
+ * effort for this asm magic.
+ */
+#define ASM_CALL_SYSVEC							\
+	"call irq_enter_rcu				\n"		\
+	"movq	%[arg1], %%rdi				\n"		\
+	"call %P[__func]				\n"		\
+	"call irq_exit_rcu				\n"
 
-	__this_cpu_add(irq_count, 1);
+#define SYSVEC_CONSTRAINTS	, [arg1] "r" (regs)
-	asm_call_sysvec_on_stack(tos - 8, func, regs);
+
-	__this_cpu_sub(irq_count, 1);
+#define run_sysvec_on_irqstack_cond(func, regs)				\
+{									\
+	assert_function_type(func, void (*)(struct pt_regs *));		\
+	assert_arg_type(regs, struct pt_regs *);			\
+									\
+	call_on_irqstack_cond(func, regs, ASM_CALL_SYSVEC,		\
+			      SYSVEC_CONSTRAINTS, regs);		\
 }
 
-static __always_inline void
+/*
-__run_irq_on_irqstack(void (*func)(struct irq_desc *desc),
+ * As in ASM_CALL_SYSVEC above the clobbers force the compiler to store
-		      struct irq_desc *desc)
+ * @regs and @vector in callee saved registers.
-{
+ */
-	void *tos = __this_cpu_read(hardirq_stack_ptr);
+#define ASM_CALL_IRQ							\
+	"call irq_enter_rcu				\n"		\
+	"movq	%[arg1], %%rdi				\n"		\
+	"movl	%[arg2], %%esi				\n"		\
+	"call %P[__func]				\n"		\
+	"call irq_exit_rcu				\n"
 
-	__this_cpu_add(irq_count, 1);
+#define IRQ_CONSTRAINTS	, [arg1] "r" (regs), [arg2] "r" (vector)
-	asm_call_irq_on_stack(tos - 8, func, desc);
+
-	__this_cpu_sub(irq_count, 1);
+#define run_irq_on_irqstack_cond(func, regs, vector)			\
+{									\
+	assert_function_type(func, void (*)(struct pt_regs *, u32));	\
+	assert_arg_type(regs, struct pt_regs *);			\
+	assert_arg_type(vector, u32);					\
+									\
+	call_on_irqstack_cond(func, regs, ASM_CALL_IRQ,			\
+			      IRQ_CONSTRAINTS, regs, vector);		\
+}
+
+#define ASM_CALL_SOFTIRQ						\
+	"call %P[__func]				\n"
+
+/*
+ * Macro to invoke __do_softirq on the irq stack. This is only called from
+ * task context when bottom halfs are about to be reenabled and soft
+ * interrupts are pending to be processed. The interrupt stack cannot be in
+ * use here.
+ */
+#define do_softirq_own_stack()						\
+{									\
+	__this_cpu_write(hardirq_stack_inuse, true);			\
+	call_on_irqstack(__do_softirq, ASM_CALL_SOFTIRQ);		\
+	__this_cpu_write(hardirq_stack_inuse, false);			\
 }
 
 #else /* CONFIG_X86_64 */
-static inline bool irqstack_active(void) { return false; }
+/* System vector handlers always run on the stack they interrupted. */
-static inline void __run_on_irqstack(void (*func)(void)) { }
+#define run_sysvec_on_irqstack_cond(func, regs)				\
-static inline void __run_sysvec_on_irqstack(void (*func)(struct pt_regs *regs),
+{									\
-					    struct pt_regs *regs) { }
+	irq_enter_rcu();						\
-static inline void __run_irq_on_irqstack(void (*func)(struct irq_desc *desc),
+	func(regs);							\
-					 struct irq_desc *desc) { }
+	irq_exit_rcu();							\
+}
+
+/* Switches to the irq stack within func() */
+#define run_irq_on_irqstack_cond(func, regs, vector)			\
+{									\
+	irq_enter_rcu();						\
+	func(regs, vector);						\
+	irq_exit_rcu();							\
+}
+
 #endif /* !CONFIG_X86_64 */
 
-static __always_inline bool irq_needs_irq_stack(struct pt_regs *regs)
-{
-	if (IS_ENABLED(CONFIG_X86_32))
-		return false;
-	if (!regs)
-		return !irqstack_active();
-	return !user_mode(regs) && !irqstack_active();
-}
-
-
-static __always_inline void run_on_irqstack_cond(void (*func)(void),
-						 struct pt_regs *regs)
-{
-	lockdep_assert_irqs_disabled();
-
-	if (irq_needs_irq_stack(regs))
-		__run_on_irqstack(func);
-	else
-		func();
-}
-
-static __always_inline void
-run_sysvec_on_irqstack_cond(void (*func)(struct pt_regs *regs),
-			    struct pt_regs *regs)
-{
-	lockdep_assert_irqs_disabled();
-
-	if (irq_needs_irq_stack(regs))
-		__run_sysvec_on_irqstack(func, regs);
-	else
-		func(regs);
-}
-
-static __always_inline void
-run_irq_on_irqstack_cond(void (*func)(struct irq_desc *desc), struct irq_desc *desc,
-			 struct pt_regs *regs)
-{
-	lockdep_assert_irqs_disabled();
-
-	if (irq_needs_irq_stack(regs))
-		__run_irq_on_irqstack(func, desc);
-	else
-		func(desc);
-}
-
 #endif

arch/x86/include/asm/processor.h

@@ -426,8 +426,6 @@ struct irq_stack {
 	char		stack[IRQ_STACK_SIZE];
 } __aligned(IRQ_STACK_SIZE);
 
-DECLARE_PER_CPU(struct irq_stack *, hardirq_stack_ptr);
-
 #ifdef CONFIG_X86_32
 DECLARE_PER_CPU(unsigned long, cpu_current_top_of_stack);
 #else
@@ -454,7 +452,8 @@ static inline unsigned long cpu_kernelmode_gs_base(int cpu)
 	return (unsigned long)per_cpu(fixed_percpu_data.gs_base, cpu);
 }
 
-DECLARE_PER_CPU(unsigned int, irq_count);
+DECLARE_PER_CPU(void *, hardirq_stack_ptr);
+DECLARE_PER_CPU(bool, hardirq_stack_inuse);
 extern asmlinkage void ignore_sysret(void);
 
 /* Save actual FS/GS selectors and bases to current->thread */
@@ -473,9 +472,9 @@ struct stack_canary {
 };
 DECLARE_PER_CPU_ALIGNED(struct stack_canary, stack_canary);
 #endif
-/* Per CPU softirq stack pointer */
+DECLARE_PER_CPU(struct irq_stack *, hardirq_stack_ptr);
 DECLARE_PER_CPU(struct irq_stack *, softirq_stack_ptr);
-#endif	/* X86_64 */
+#endif	/* !X86_64 */
 
 extern unsigned int fpu_kernel_xstate_size;
 extern unsigned int fpu_user_xstate_size;

arch/x86/include/asm/softirq_stack.h

@@ -0,0 +1,11 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _ASM_X86_SOFTIRQ_STACK_H
+#define _ASM_X86_SOFTIRQ_STACK_H
+
+#ifdef CONFIG_X86_64
+# include <asm/irq_stack.h>
+#else
+# include <asm-generic/softirq_stack.h>
+#endif
+
+#endif

arch/x86/kernel/apic/apic.c

@@ -2138,18 +2138,11 @@ void __init register_lapic_address(unsigned long address)
  * Local APIC interrupts
  */
 
-/**
+/*
- * spurious_interrupt - Catch all for interrupts raised on unused vectors
+ * Common handling code for spurious_interrupt and spurious_vector entry
- * @regs:	Pointer to pt_regs on stack
+ * points below. No point in allowing the compiler to inline it twice.
- * @vector:	The vector number
- *
- * This is invoked from ASM entry code to catch all interrupts which
- * trigger on an entry which is routed to the common_spurious idtentry
- * point.
- *
- * Also called from sysvec_spurious_apic_interrupt().
  */
-DEFINE_IDTENTRY_IRQ(spurious_interrupt)
+static noinline void handle_spurious_interrupt(u8 vector)
 {
 	u32 v;
 
@@ -2184,9 +2177,23 @@ out:
 	trace_spurious_apic_exit(vector);
 }
 
+/**
+ * spurious_interrupt - Catch all for interrupts raised on unused vectors
+ * @regs:	Pointer to pt_regs on stack
+ * @vector:	The vector number
+ *
+ * This is invoked from ASM entry code to catch all interrupts which
+ * trigger on an entry which is routed to the common_spurious idtentry
+ * point.
+ */
+DEFINE_IDTENTRY_IRQ(spurious_interrupt)
+{
+	handle_spurious_interrupt(vector);
+}
+
 DEFINE_IDTENTRY_SYSVEC(sysvec_spurious_apic_interrupt)
 {
-	__spurious_interrupt(regs, SPURIOUS_APIC_VECTOR);
+	handle_spurious_interrupt(SPURIOUS_APIC_VECTOR);
 }
 
 /*

arch/x86/kernel/cpu/common.c

@@ -1742,8 +1742,8 @@ DEFINE_PER_CPU(struct task_struct *, current_task) ____cacheline_aligned =
 	&init_task;
 EXPORT_PER_CPU_SYMBOL(current_task);
 
-DEFINE_PER_CPU(struct irq_stack *, hardirq_stack_ptr);
+DEFINE_PER_CPU(void *, hardirq_stack_ptr);
-DEFINE_PER_CPU(unsigned int, irq_count) __visible = -1;
+DEFINE_PER_CPU(bool, hardirq_stack_inuse);
 
 DEFINE_PER_CPU(int, __preempt_count) = INIT_PREEMPT_COUNT;
 EXPORT_PER_CPU_SYMBOL(__preempt_count);

arch/x86/kernel/dumpstack_64.c

@@ -129,11 +129,20 @@ static __always_inline bool in_exception_stack(unsigned long *stack, struct stac
 static __always_inline bool in_irq_stack(unsigned long *stack, struct stack_info *info)
 {
 	unsigned long *end = (unsigned long *)this_cpu_read(hardirq_stack_ptr);
-	unsigned long *begin = end - (IRQ_STACK_SIZE / sizeof(long));
+	unsigned long *begin;
 
 	/*
-	 * This is a software stack, so 'end' can be a valid stack pointer.
+	 * @end points directly to the top most stack entry to avoid a -8
-	 * It just means the stack is empty.
+	 * adjustment in the stack switch hotpath. Adjust it back before
+	 * calculating @begin.
+	 */
+	end++;
+	begin = end - (IRQ_STACK_SIZE / sizeof(long));
+
+	/*
+	 * Due to the switching logic RSP can never be == @end because the
+	 * final operation is 'popq %rsp' which means after that RSP points
+	 * to the original stack and not to @end.
 	 */
 	if (stack < begin || stack >= end)
 		return false;
@@ -143,8 +152,9 @@ static __always_inline bool in_irq_stack(unsigned long *stack, struct stack_info
 	info->end	= end;
 
 	/*
-	 * The next stack pointer is the first thing pushed by the entry code
+	 * The next stack pointer is stored at the top of the irq stack
-	 * after switching to the irq stack.
+	 * before switching to the irq stack. Actual stack entries are all
+	 * below that.
 	 */
 	info->next_sp = (unsigned long *)*(end - 1);
 

arch/x86/kernel/irq.c

@@ -228,7 +228,7 @@ static __always_inline void handle_irq(struct irq_desc *desc,
 				       struct pt_regs *regs)
 {
 	if (IS_ENABLED(CONFIG_X86_64))
-		run_irq_on_irqstack_cond(desc->handle_irq, desc, regs);
+		generic_handle_irq_desc(desc);
 	else
 		__handle_irq(desc, regs);
 }

arch/x86/kernel/irq_32.c

@@ -22,6 +22,7 @@
 
 #include <asm/apic.h>
 #include <asm/nospec-branch.h>
+#include <asm/softirq_stack.h>
 
 #ifdef CONFIG_DEBUG_STACKOVERFLOW
 

arch/x86/kernel/irq_64.c

@@ -20,6 +20,7 @@
 #include <linux/sched/task_stack.h>
 
 #include <asm/cpu_entry_area.h>
+#include <asm/softirq_stack.h>
 #include <asm/irq_stack.h>
 #include <asm/io_apic.h>
 #include <asm/apic.h>
@@ -48,7 +49,8 @@ static int map_irq_stack(unsigned int cpu)
 	if (!va)
 		return -ENOMEM;
 
-	per_cpu(hardirq_stack_ptr, cpu) = va + IRQ_STACK_SIZE;
+	/* Store actual TOS to avoid adjustment in the hotpath */
+	per_cpu(hardirq_stack_ptr, cpu) = va + IRQ_STACK_SIZE - 8;
 	return 0;
 }
 #else
@@ -60,7 +62,8 @@ static int map_irq_stack(unsigned int cpu)
 {
 	void *va = per_cpu_ptr(&irq_stack_backing_store, cpu);
 
-	per_cpu(hardirq_stack_ptr, cpu) = va + IRQ_STACK_SIZE;
+	/* Store actual TOS to avoid adjustment in the hotpath */
+	per_cpu(hardirq_stack_ptr, cpu) = va + IRQ_STACK_SIZE - 8;
 	return 0;
 }
 #endif
@@ -71,8 +74,3 @@ int irq_init_percpu_irqstack(unsigned int cpu)
 		return 0;
 	return map_irq_stack(cpu);
 }
-
-void do_softirq_own_stack(void)
-{
-	run_on_irqstack_cond(__do_softirq, NULL);
-}

arch/x86/kernel/process_64.c

@@ -539,7 +539,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	int cpu = smp_processor_id();
 
 	WARN_ON_ONCE(IS_ENABLED(CONFIG_DEBUG_ENTRY) &&
-		     this_cpu_read(irq_count) != -1);
+		     this_cpu_read(hardirq_stack_inuse));
 
 	if (!test_thread_flag(TIF_NEED_FPU_LOAD))
 		switch_fpu_prepare(prev_fpu, cpu);

include/asm-generic/Kbuild

@@ -50,6 +50,7 @@ mandatory-y += sections.h
 mandatory-y += serial.h
 mandatory-y += shmparam.h
 mandatory-y += simd.h
+mandatory-y += softirq_stack.h
 mandatory-y += switch_to.h
 mandatory-y += timex.h
 mandatory-y += tlbflush.h

include/asm-generic/softirq_stack.h

@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+#ifndef __ASM_GENERIC_SOFTIRQ_STACK_H
+#define __ASM_GENERIC_SOFTIRQ_STACK_H
+
+#ifdef CONFIG_HAVE_SOFTIRQ_ON_OWN_STACK
+void do_softirq_own_stack(void);
+#else
+static inline void do_softirq_own_stack(void)
+{
+	__do_softirq();
+}
+#endif
+
+#endif

include/linux/interrupt.h

@@ -569,15 +569,6 @@ struct softirq_action
 asmlinkage void do_softirq(void);
 asmlinkage void __do_softirq(void);
 
-#ifdef __ARCH_HAS_DO_SOFTIRQ
-void do_softirq_own_stack(void);
-#else
-static inline void do_softirq_own_stack(void)
-{
-	__do_softirq();
-}
-#endif
-
 extern void open_softirq(int nr, void (*action)(struct softirq_action *));
 extern void softirq_init(void);
 extern void __raise_softirq_irqoff(unsigned int nr);

kernel/softirq.c

@@ -26,6 +26,8 @@
 #include <linux/tick.h>
 #include <linux/irq.h>
 
+#include <asm/softirq_stack.h>
+
 #define CREATE_TRACE_POINTS
 #include <trace/events/irq.h>
 

tools/objtool/check.c

@@ -2084,6 +2084,20 @@ static int update_cfi_state(struct instruction *insn, struct cfi_state *cfi,
 				}
 			}
 
+			else if (op->dest.reg == CFI_SP &&
+				 cfi->vals[op->src.reg].base == CFI_SP_INDIRECT &&
+				 cfi->vals[op->src.reg].offset == cfa->offset) {
+
+				/*
+				 * The same stack swizzle case 2) as above. But
+				 * because we can't change cfa->base, case 3)
+				 * will become a regular POP. Pretend we're a
+				 * PUSH so things don't go unbalanced.
+				 */
+				cfi->stack_size += 8;
+			}
+
+
 			break;
 
 		case OP_SRC_ADD: