Merge tag 's390-5.5-2' of git://git.kernel.org/pub/scm/linux/kernel/g…

…it/s390/linux

Pull more s390 updates from Vasily Gorbik:

 - Make stack unwinder reliable and suitable for livepatching. Add
   unwinder testing module.

 - Fixes for CALL_ON_STACK helper used for stack switching.

 - Fix unwinding from bpf code.

 - Fix getcpu and remove compat support in vdso code.

 - Fix address space control registers initialization.

 - Save KASLR offset for early dumps.

 - Handle new FILTERED_BY_HYPERVISOR reply code in crypto code.

 - Minor perf code cleanup and potential memory leak fix.

 - Add couple of error messages for corner cases during PCI device
   creation.

* tag 's390-5.5-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (33 commits)
  s390: remove compat vdso code
  s390/livepatch: Implement reliable stack tracing for the consistency model
  s390/unwind: add stack pointer alignment sanity checks
  s390/unwind: filter out unreliable bogus %r14
  s390/unwind: start unwinding from reliable state
  s390/test_unwind: add program check context tests
  s390/test_unwind: add irq context tests
  s390/test_unwind: print verbose unwinding results
  s390/test_unwind: add CALL_ON_STACK tests
  s390: fix register clobbering in CALL_ON_STACK
  s390/test_unwind: require that unwinding ended successfully
  s390/unwind: add a test for the internal API
  s390/unwind: always inline get_stack_pointer
  s390/pci: add error message on device number limit
  s390/pci: add error message for UID collision
  s390/cpum_sf: Check for SDBT and SDB consistency
  s390/cpum_sf: Use TEAR_REG macro consistantly
  s390/cpum_sf: Remove unnecessary check for pending SDBs
  s390/cpum_sf: Replace function name in debug statements
  s390/kaslr: store KASLR offset for early dumps
  ...

arch/s390/Kconfig

@@ -170,6 +170,7 @@ config S390
 	select HAVE_PERF_EVENTS
 	select HAVE_RCU_TABLE_FREE
 	select HAVE_REGS_AND_STACK_ACCESS_API
+	select HAVE_RELIABLE_STACKTRACE
 	select HAVE_RSEQ
 	select HAVE_SYSCALL_TRACEPOINTS
 	select HAVE_VIRT_CPU_ACCOUNTING
@@ -426,9 +427,6 @@ config COMPAT
 	  (and some other stuff like libraries and such) is needed for
 	  executing 31 bit applications.  It is safe to say "Y".
 
-config COMPAT_VDSO
-	def_bool COMPAT && !CC_IS_CLANG
-
 config SYSVIPC_COMPAT
 	def_bool y if COMPAT && SYSVIPC
 
@@ -1018,3 +1016,17 @@ config S390_GUEST
 	  the KVM hypervisor.
 
 endmenu
+
+menu "Selftests"
+
+config S390_UNWIND_SELFTEST
+	def_tristate n
+	prompt "Test unwind functions"
+	help
+	  This option enables s390 specific stack unwinder testing kernel
+	  module. This option is not useful for distributions or general
+	  kernels, but only for kernel developers working on architecture code.
+
+	  Say N if you are unsure.
+
+endmenu

arch/s390/Makefile

@@ -157,7 +157,6 @@ zfcpdump:
 
 vdso_install:
 	$(Q)$(MAKE) $(build)=arch/$(ARCH)/kernel/vdso64 $@
-	$(Q)$(MAKE) $(build)=arch/$(ARCH)/kernel/vdso32 $@
 
 archclean:
 	$(Q)$(MAKE) $(clean)=$(boot)

arch/s390/boot/startup.c

@@ -170,6 +170,11 @@ void startup_kernel(void)
 		handle_relocs(__kaslr_offset);
 
 	if (__kaslr_offset) {
+		/*
+		 * Save KASLR offset for early dumps, before vmcore_info is set.
+		 * Mark as uneven to distinguish from real vmcore_info pointer.
+		 */
+		S390_lowcore.vmcore_info = __kaslr_offset | 0x1UL;
 		/* Clear non-relocated kernel */
 		if (IS_ENABLED(CONFIG_KERNEL_UNCOMPRESSED))
 			memset(img, 0, vmlinux.image_size);

arch/s390/include/asm/cpu_mf.h

@@ -313,7 +313,7 @@ static inline unsigned long *trailer_entry_ptr(unsigned long v)
 	return (unsigned long *) ret;
 }
 
-/* Return if the entry in the sample data block table (sdbt)
+/* Return true if the entry in the sample data block table (sdbt)
  * is a link to the next sdbt */
 static inline int is_link_entry(unsigned long *s)
 {

arch/s390/include/asm/perf_event.h

@@ -12,6 +12,7 @@
 
 #include <linux/perf_event.h>
 #include <linux/device.h>
+#include <asm/stacktrace.h>
 
 /* Per-CPU flags for PMU states */
 #define PMU_F_RESERVED			0x1000
@@ -73,4 +74,10 @@ struct perf_sf_sde_regs {
 #define SDB_FULL_BLOCKS(hwc)	(SAMPL_FLAGS(hwc) & PERF_CPUM_SF_FULL_BLOCKS)
 #define SAMPLE_FREQ_MODE(hwc)	(SAMPL_FLAGS(hwc) & PERF_CPUM_SF_FREQ_MODE)
 
+#define perf_arch_fetch_caller_regs(regs, __ip) do {			\
+	(regs)->psw.addr = (__ip);					\
+	(regs)->gprs[15] = (unsigned long)__builtin_frame_address(0) -	\
+		offsetof(struct stack_frame, back_chain);		\
+} while (0)
+
 #endif /* _ASM_S390_PERF_EVENT_H */

arch/s390/include/asm/processor.h

@@ -310,7 +310,7 @@ void enabled_wait(void);
 /*
  * Function to drop a processor into disabled wait state
  */
-static inline void __noreturn disabled_wait(void)
+static __always_inline void __noreturn disabled_wait(void)
 {
 	psw_t psw;
 

arch/s390/include/asm/stacktrace.h

@@ -33,12 +33,12 @@ static inline bool on_stack(struct stack_info *info,
 	return addr >= info->begin && addr + len <= info->end;
 }
 
-static inline unsigned long get_stack_pointer(struct task_struct *task,
-					      struct pt_regs *regs)
+static __always_inline unsigned long get_stack_pointer(struct task_struct *task,
+						       struct pt_regs *regs)
 {
 	if (regs)
 		return (unsigned long) kernel_stack_pointer(regs);
-	if (!task || task == current)
+	if (task == current)
 		return current_stack_pointer();
 	return (unsigned long) task->thread.ksp;
 }
@@ -62,6 +62,17 @@ struct stack_frame {
 };
 #endif
 
+/*
+ * Unlike current_stack_pointer() which simply returns current value of %r15
+ * current_frame_address() returns function stack frame address, which matches
+ * %r15 upon function invocation. It may differ from %r15 later if function
+ * allocates stack for local variables or new stack frame to call other
+ * functions.
+ */
+#define current_frame_address()						\
+	((unsigned long)__builtin_frame_address(0) -			\
+	 offsetof(struct stack_frame, back_chain))
+
 #define CALL_ARGS_0()							\
 	register unsigned long r2 asm("2")
 #define CALL_ARGS_1(arg1)						\
@@ -95,20 +106,33 @@ struct stack_frame {
 
 #define CALL_ON_STACK(fn, stack, nr, args...)				\
 ({									\
+	unsigned long frame = current_frame_address();			\
 	CALL_ARGS_##nr(args);						\
 	unsigned long prev;						\
 									\
 	asm volatile(							\
 		"	la	%[_prev],0(15)\n"			\
-		"	la	15,0(%[_stack])\n"			\
-		"	stg	%[_prev],%[_bc](15)\n"			\
+		"	lg	15,%[_stack]\n"				\
+		"	stg	%[_frame],%[_bc](15)\n"			\
 		"	brasl	14,%[_fn]\n"				\
 		"	la	15,0(%[_prev])\n"			\
 		: [_prev] "=&a" (prev), CALL_FMT_##nr			\
-		  [_stack] "a" (stack),					\
+		  [_stack] "R" (stack),					\
 		  [_bc] "i" (offsetof(struct stack_frame, back_chain)),	\
+		  [_frame] "d" (frame),					\
 		  [_fn] "X" (fn) : CALL_CLOBBER_##nr);			\
 	r2;								\
 })
 
+#define CALL_ON_STACK_NORETURN(fn, stack)				\
+({									\
+	asm volatile(							\
+		"	la	15,0(%[_stack])\n"			\
+		"	xc	%[_bc](8,15),%[_bc](15)\n"		\
+		"	brasl	14,%[_fn]\n"				\
+		::[_bc] "i" (offsetof(struct stack_frame, back_chain)),	\
+		  [_stack] "a" (stack), [_fn] "X" (fn));		\
+	BUG();								\
+})
+
 #endif /* _ASM_S390_STACKTRACE_H */

arch/s390/include/asm/unwind.h

@@ -35,7 +35,6 @@ struct unwind_state {
 	struct task_struct *task;
 	struct pt_regs *regs;
 	unsigned long sp, ip;
-	bool reuse_sp;
 	int graph_idx;
 	bool reliable;
 	bool error;
@@ -59,10 +58,11 @@ static inline bool unwind_error(struct unwind_state *state)
 static inline void unwind_start(struct unwind_state *state,
 				struct task_struct *task,
 				struct pt_regs *regs,
-				unsigned long sp)
+				unsigned long first_frame)
 {
-	sp = sp ? : get_stack_pointer(task, regs);
-	__unwind_start(state, task, regs, sp);
+	task = task ?: current;
+	first_frame = first_frame ?: get_stack_pointer(task, regs);
+	__unwind_start(state, task, regs, first_frame);
 }
 
 static inline struct pt_regs *unwind_get_entry_regs(struct unwind_state *state)

arch/s390/include/asm/vdso.h

@@ -41,8 +41,17 @@ struct vdso_data {
 struct vdso_per_cpu_data {
 	__u64 ectg_timer_base;
 	__u64 ectg_user_time;
-	__u32 cpu_nr;
-	__u32 node_id;
+	/*
+	 * Note: node_id and cpu_nr must be at adjacent memory locations.
+	 * VDSO userspace must read both values with a single instruction.
+	 */
+	union {
+		__u64 getcpu_val;
+		struct {
+			__u32 node_id;
+			__u32 cpu_nr;
+		};
+	};
 };
 
 extern struct vdso_data *vdso_data;

arch/s390/kernel/Makefile

@@ -81,4 +81,3 @@ obj-$(CONFIG_TRACEPOINTS)	+= trace.o
 
 # vdso
 obj-y				+= vdso64/
-obj-$(CONFIG_COMPAT_VDSO)	+= vdso32/

arch/s390/kernel/asm-offsets.c

@@ -78,8 +78,7 @@ int main(void)
 	OFFSET(__VDSO_TS_END, vdso_data, ts_end);
 	OFFSET(__VDSO_ECTG_BASE, vdso_per_cpu_data, ectg_timer_base);
 	OFFSET(__VDSO_ECTG_USER, vdso_per_cpu_data, ectg_user_time);
-	OFFSET(__VDSO_CPU_NR, vdso_per_cpu_data, cpu_nr);
-	OFFSET(__VDSO_NODE_ID, vdso_per_cpu_data, node_id);
+	OFFSET(__VDSO_GETCPU_VAL, vdso_per_cpu_data, getcpu_val);
 	BLANK();
 	/* constants used by the vdso */
 	DEFINE(__CLOCK_REALTIME, CLOCK_REALTIME);

arch/s390/kernel/dumpstack.c

@@ -38,6 +38,7 @@ const char *stack_type_name(enum stack_type type)
 		return "unknown";
 	}
 }
+EXPORT_SYMBOL_GPL(stack_type_name);
 
 static inline bool in_stack(unsigned long sp, struct stack_info *info,
 			    enum stack_type type, unsigned long low,
@@ -93,7 +94,9 @@ int get_stack_info(unsigned long sp, struct task_struct *task,
 	if (!sp)
 		goto unknown;
 
-	task = task ? : current;
+	/* Sanity check: ABI requires SP to be aligned 8 bytes. */
+	if (sp & 0x7)
+		goto unknown;
 
 	/* Check per-task stack */
 	if (in_task_stack(sp, task, info))
@@ -128,8 +131,6 @@ void show_stack(struct task_struct *task, unsigned long *stack)
 	struct unwind_state state;
 
 	printk("Call Trace:\n");
-	if (!task)
-		task = current;
 	unwind_for_each_frame(&state, task, NULL, (unsigned long) stack)
 		printk(state.reliable ? " [<%016lx>] %pSR \n" :
 					"([<%016lx>] %pSR)\n",

arch/s390/kernel/head64.S

@@ -31,7 +31,7 @@ ENTRY(startup_continue)
 #
 	larl	%r14,init_task
 	stg	%r14,__LC_CURRENT
-	larl	%r15,init_thread_union+THREAD_SIZE-STACK_FRAME_OVERHEAD
+	larl	%r15,init_thread_union+THREAD_SIZE-STACK_FRAME_OVERHEAD-__PT_SIZE
 #ifdef CONFIG_KASAN
 	brasl	%r14,kasan_early_init
 #endif

arch/s390/kernel/machine_kexec.c

@@ -164,7 +164,9 @@ static bool kdump_csum_valid(struct kimage *image)
 #ifdef CONFIG_CRASH_DUMP
 	int rc;
 
+	preempt_disable();
 	rc = CALL_ON_STACK(do_start_kdump, S390_lowcore.nodat_stack, 1, image);
+	preempt_enable();
 	return rc == 0;
 #else
 	return false;
@@ -254,10 +256,10 @@ void arch_crash_save_vmcoreinfo(void)
 	VMCOREINFO_SYMBOL(lowcore_ptr);
 	VMCOREINFO_SYMBOL(high_memory);
 	VMCOREINFO_LENGTH(lowcore_ptr, NR_CPUS);
-	mem_assign_absolute(S390_lowcore.vmcore_info, paddr_vmcoreinfo_note());
 	vmcoreinfo_append_str("SDMA=%lx\n", __sdma);
 	vmcoreinfo_append_str("EDMA=%lx\n", __edma);
 	vmcoreinfo_append_str("KERNELOFFSET=%lx\n", kaslr_offset());
+	mem_assign_absolute(S390_lowcore.vmcore_info, paddr_vmcoreinfo_note());
 }
 
 void machine_shutdown(void)