s390/cpumf: rework program parameter setting to detect guest samples

The program parameter can be used to mark hardware samples with
some token.  Previously, it was used to mark guest samples only.

Improve the program parameter doubleword by combining two parts,
the leftmost LPP part and the rightmost PID part.  Set the PID
part for processes by using the task PID.
To distinguish host and guest samples for the kernel (PID part
is zero), the guest must always set the program paramater to a
non-zero value.  Use the leftmost bit in the LPP part of the
program parameter to be able to detect guest kernel samples.

[[email protected]]: Split __LC_CURRENT and introduced
__LC_LPP. Corrected __LC_CURRENT users and adjusted assembler parts.
And updated the commit message accordingly.

Signed-off-by: Christian Borntraeger <[email protected]>
Signed-off-by: Hendrik Brueckner <[email protected]>
Reviewed-by: Heiko Carstens <[email protected]>
Signed-off-by: Martin Schwidefsky <[email protected]>

Author: borntraeger

arch/s390/include/asm/lowcore.h

@@ -132,7 +132,14 @@ struct _lowcore {
 	/* Address space pointer. */
 	__u64	kernel_asce;			/* 0x0358 */
 	__u64	user_asce;			/* 0x0360 */
-	__u64	current_pid;			/* 0x0368 */
+
+	/*
+	 * The lpp and current_pid fields form a
+	 * 64-bit value that is set as program
+	 * parameter with the LPP instruction.
+	 */
+	__u32	lpp;				/* 0x0368 */
+	__u32	current_pid;			/* 0x036c */
 
 	/* SMP info area */
 	__u32	cpu_nr;				/* 0x0370 */

arch/s390/include/asm/setup.h

@@ -34,6 +34,9 @@
 #define MACHINE_FLAG_VX		_BITUL(18)
 #define MACHINE_FLAG_CAD	_BITUL(19)
 
+#define LPP_MAGIC		_BITUL(31)
+#define LPP_PFAULT_PID_MASK	_AC(0xffffffff, UL)
+
 #ifndef __ASSEMBLY__
 
 #include <asm/lowcore.h>

arch/s390/kernel/asm-offsets.c

@@ -163,6 +163,7 @@ int main(void)
 	OFFSET(__LC_RESTART_DATA, _lowcore, restart_data);
 	OFFSET(__LC_RESTART_SOURCE, _lowcore, restart_source);
 	OFFSET(__LC_USER_ASCE, _lowcore, user_asce);
+	OFFSET(__LC_LPP, _lowcore, lpp);
 	OFFSET(__LC_CURRENT_PID, _lowcore, current_pid);
 	OFFSET(__LC_PERCPU_OFFSET, _lowcore, percpu_offset);
 	OFFSET(__LC_VDSO_PER_CPU, _lowcore, vdso_per_cpu_data);

arch/s390/kernel/entry.S

@@ -187,8 +187,11 @@ ENTRY(__switch_to)
 	stg	%r15,__LC_KERNEL_STACK		# store end of kernel stack
 	lg	%r15,__THREAD_ksp(%r1)		# load kernel stack of next
 	lctl	%c4,%c4,__TASK_pid(%r3)		# load pid to control reg. 4
-	mvc	__LC_CURRENT_PID+4(4,%r0),__TASK_pid(%r3) # store pid of next
+	mvc	__LC_CURRENT_PID(4,%r0),__TASK_pid(%r3) # store pid of next
 	lmg	%r6,%r15,__SF_GPRS(%r15)	# load gprs of next task
+	TSTMSK	__LC_MACHINE_FLAGS,MACHINE_FLAG_LPP
+	bzr	%r14
+	.insn	s,0xb2800000,__LC_LPP		# set program parameter
 	br	%r14
 
 .L__critical_start:
@@ -203,7 +206,7 @@ ENTRY(sie64a)
 	stmg	%r6,%r14,__SF_GPRS(%r15)	# save kernel registers
 	stg	%r2,__SF_EMPTY(%r15)		# save control block pointer
 	stg	%r3,__SF_EMPTY+8(%r15)		# save guest register save area
-	xc	__SF_EMPTY+16(16,%r15),__SF_EMPTY+16(%r15) # host id & reason
+	xc	__SF_EMPTY+16(8,%r15),__SF_EMPTY+16(%r15) # reason code = 0
 	TSTMSK	__LC_CPU_FLAGS,_CIF_FPU		# load guest fp/vx registers ?
 	jno	.Lsie_load_guest_gprs
 	brasl	%r14,load_fpu_regs		# load guest fp/vx regs
@@ -220,14 +223,7 @@ ENTRY(sie64a)
 	jnz	.Lsie_skip
 	TSTMSK	__LC_CPU_FLAGS,_CIF_FPU
 	jo	.Lsie_skip			# exit if fp/vx regs changed
-	TSTMSK	__LC_MACHINE_FLAGS,MACHINE_FLAG_LPP
-	jz	.Lsie_enter
-	.insn	s,0xb2800000,__LC_CURRENT_PID	# set guest id to pid
-.Lsie_enter:
 	sie	0(%r14)
-	TSTMSK	__LC_MACHINE_FLAGS,MACHINE_FLAG_LPP
-	jz	.Lsie_skip
-	.insn	s,0xb2800000,__SF_EMPTY+16(%r15)# set host id
 .Lsie_skip:
 	ni	__SIE_PROG0C+3(%r14),0xfe	# no longer in SIE
 	lctlg	%c1,%c1,__LC_USER_ASCE		# load primary asce
@@ -244,11 +240,11 @@ sie_exit:
 	lg	%r14,__SF_EMPTY+8(%r15)		# load guest register save area
 	stmg	%r0,%r13,0(%r14)		# save guest gprs 0-13
 	lmg	%r6,%r14,__SF_GPRS(%r15)	# restore kernel registers
-	lg	%r2,__SF_EMPTY+24(%r15)		# return exit reason code
+	lg	%r2,__SF_EMPTY+16(%r15)		# return exit reason code
 	br	%r14
 .Lsie_fault:
 	lghi	%r14,-EFAULT
-	stg	%r14,__SF_EMPTY+24(%r15)	# set exit reason code
+	stg	%r14,__SF_EMPTY+16(%r15)	# set exit reason code
 	j	sie_exit
 
 	EX_TABLE(.Lrewind_pad,.Lsie_fault)
@@ -938,7 +934,10 @@ ENTRY(mcck_int_handler)
 # PSW restart interrupt handler
 #
 ENTRY(restart_int_handler)
-	stg	%r15,__LC_SAVE_AREA_RESTART
+	TSTMSK	__LC_MACHINE_FLAGS,MACHINE_FLAG_LPP
+	jz	0f
+	.insn	s,0xb2800000,__LC_LPP
+0:	stg	%r15,__LC_SAVE_AREA_RESTART
 	lg	%r15,__LC_RESTART_STACK
 	aghi	%r15,-__PT_SIZE			# create pt_regs on stack
 	xc	0(__PT_SIZE,%r15),0(%r15)
@@ -1042,10 +1041,7 @@ cleanup_critical:
 
 .Lcleanup_sie:
 	lg	%r9,__SF_EMPTY(%r15)		# get control block pointer
-	TSTMSK	__LC_MACHINE_FLAGS,MACHINE_FLAG_LPP
-	jz	0f
-	.insn	s,0xb2800000,__SF_EMPTY+16(%r15)# set host id
-0:	ni	__SIE_PROG0C+3(%r9),0xfe	# no longer in SIE
+	ni	__SIE_PROG0C+3(%r9),0xfe	# no longer in SIE
 	lctlg	%c1,%c1,__LC_USER_ASCE		# load primary asce
 	larl	%r9,sie_exit			# skip forward to sie_exit
 	br	%r14

arch/s390/kernel/head64.S

@@ -16,7 +16,12 @@
 
 __HEAD
 ENTRY(startup_continue)
-	larl	%r1,sched_clock_base_cc
+	tm	__LC_STFL_FAC_LIST+6,0x80	# LPP available ?
+	jz	0f
+	xc	__LC_LPP+1(7,0),__LC_LPP+1	# clear lpp and current_pid
+	mvi	__LC_LPP,0x80			#   and set LPP_MAGIC
+	.insn	s,0xb2800000,__LC_LPP		# load program parameter
+0:	larl	%r1,sched_clock_base_cc
 	mvc	0(8,%r1),__LC_LAST_UPDATE_CLOCK
 	larl	%r13,.LPG1		# get base
 	lctlg	%c0,%c15,.Lctl-.LPG1(%r13)	# load control registers

arch/s390/kernel/perf_cpum_sf.c

@@ -1019,11 +1019,13 @@ static int perf_push_sample(struct perf_event *event, struct sf_raw_sample *sfr)
 		break;
 	}
 
-	/* The host-program-parameter (hpp) contains the pid of
-	 * the CPU thread as set by sie64a() in entry.S.
-	 * If non-zero assume a guest sample.
+	/*
+	 * A non-zero guest program parameter indicates a guest
+	 * sample.
+	 * Note that some early samples might be misaccounted to
+	 * the host.
 	 */
-	if (sfr->basic.hpp)
+	if (sfr->basic.gpp)
 		sde_regs->in_guest = 1;
 
 	overflow = 0;

arch/s390/kernel/smp.c

@@ -262,6 +262,8 @@ static void pcpu_attach_task(struct pcpu *pcpu, struct task_struct *tsk)
 		+ THREAD_SIZE - STACK_FRAME_OVERHEAD - sizeof(struct pt_regs);
 	lc->thread_info = (unsigned long) task_thread_info(tsk);
 	lc->current_task = (unsigned long) tsk;
+	lc->lpp = LPP_MAGIC;
+	lc->current_pid = tsk->pid;
 	lc->user_timer = ti->user_timer;
 	lc->system_timer = ti->system_timer;
 	lc->steal_timer = 0;

arch/s390/mm/fault.c

@@ -590,7 +590,7 @@ int pfault_init(void)
 		.reffcode = 0,
 		.refdwlen = 5,
 		.refversn = 2,
-		.refgaddr = __LC_CURRENT_PID,
+		.refgaddr = __LC_LPP,
 		.refselmk = 1ULL << 48,
 		.refcmpmk = 1ULL << 48,
 		.reserved = __PF_RES_FIELD };
@@ -649,7 +649,7 @@ static void pfault_interrupt(struct ext_code ext_code,
 		return;
 	inc_irq_stat(IRQEXT_PFL);
 	/* Get the token (= pid of the affected task). */
-	pid = param64;
+	pid = param64 & LPP_PFAULT_PID_MASK;
 	rcu_read_lock();
 	tsk = find_task_by_pid_ns(pid, &init_pid_ns);
 	if (tsk)