Merge tag 's390-5.1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux

Pull s390 fixes from Martin Schwidefsky:
"Improvements and bug fixes for 5.1-rc2:

- Fix early free of the channel program in vfio

- On AP device removal make sure that all messages are flushed with
the driver still attached that queued the message

- Limit brk randomization to 32MB to reduce the chance that the heap
of ld.so is placed after the main stack

- Add a rolling average for the steal time of a CPU, this will be
needed for KVM to decide when to do busy waiting

- Fix a warning in the CPU-MF code

- Add a notification handler for AP configuration change to react
faster to new AP devices"

* tag 's390-5.1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
s390/cpumf: Fix warning from check_processor_id
zcrypt: handle AP Info notification from CHSC SEI command
vfio: ccw: only free cp on final interrupt
s390/vtime: steal time exponential moving average
s390/zcrypt: revisit ap device remove procedure
s390: limit brk randomization to 32MB

arch/s390/include/asm/ap.h

···
 	return reg1;
 }
 
+/*
+ * Interface to tell the AP bus code that a configuration
+ * change has happened. The bus code should at least do
+ * an ap bus resource rescan.
+ */
+#if IS_ENABLED(CONFIG_ZCRYPT)
+void ap_bus_cfg_chg(void);
+#else
+static inline void ap_bus_cfg_chg(void){};
+#endif
+
 #endif /* _ASM_S390_AP_H_ */

arch/s390/include/asm/elf.h

···
 
 /*
  * Cache aliasing on the latest machines calls for a mapping granularity
- * of 512KB. For 64-bit processes use a 512KB alignment and a randomization
- * of up to 1GB. For 31-bit processes the virtual address space is limited,
- * use no alignment and limit the randomization to 8MB.
+ * of 512KB for the anonymous mapping base. For 64-bit processes use a
+ * 512KB alignment and a randomization of up to 1GB. For 31-bit processes
+ * the virtual address space is limited, use no alignment and limit the
+ * randomization to 8MB.
+ * For the additional randomization of the program break use 32MB for
+ * 64-bit and 8MB for 31-bit.
  */
-#define BRK_RND_MASK	(is_compat_task() ? 0x7ffUL : 0x3ffffUL)
+#define BRK_RND_MASK	(is_compat_task() ? 0x7ffUL : 0x1fffUL)
 #define MMAP_RND_MASK	(is_compat_task() ? 0x7ffUL : 0x3ff80UL)
 #define MMAP_ALIGN_MASK	(is_compat_task() ? 0 : 0x7fUL)
 #define STACK_RND_MASK	MMAP_RND_MASK

arch/s390/include/asm/lowcore.h

···
 	__u64	hardirq_timer;			/* 0x02e8 */
 	__u64	softirq_timer;			/* 0x02f0 */
 	__u64	steal_timer;			/* 0x02f8 */
-	__u64	last_update_timer;		/* 0x0300 */
-	__u64	last_update_clock;		/* 0x0308 */
-	__u64	int_clock;			/* 0x0310 */
-	__u64	mcck_clock;			/* 0x0318 */
-	__u64	clock_comparator;		/* 0x0320 */
-	__u64	boot_clock[2];			/* 0x0328 */
+	__u64	avg_steal_timer;		/* 0x0300 */
+	__u64	last_update_timer;		/* 0x0308 */
+	__u64	last_update_clock;		/* 0x0310 */
+	__u64	int_clock;			/* 0x0318*/
+	__u64	mcck_clock;			/* 0x0320 */
+	__u64	clock_comparator;		/* 0x0328 */
+	__u64	boot_clock[2];			/* 0x0330 */
 
 	/* Current process. */
-	__u64	current_task;			/* 0x0338 */
-	__u64	kernel_stack;			/* 0x0340 */
+	__u64	current_task;			/* 0x0340 */
+	__u64	kernel_stack;			/* 0x0348 */
 
 	/* Interrupt, DAT-off and restartstack. */
-	__u64	async_stack;			/* 0x0348 */
-	__u64	nodat_stack;			/* 0x0350 */
-	__u64	restart_stack;			/* 0x0358 */
+	__u64	async_stack;			/* 0x0350 */
+	__u64	nodat_stack;			/* 0x0358 */
+	__u64	restart_stack;			/* 0x0360 */
 
 	/* Restart function and parameter. */
-	__u64	restart_fn;			/* 0x0360 */
-	__u64	restart_data;			/* 0x0368 */
-	__u64	restart_source;			/* 0x0370 */
+	__u64	restart_fn;			/* 0x0368 */
+	__u64	restart_data;			/* 0x0370 */
+	__u64	restart_source;			/* 0x0378 */
 
 	/* Address space pointer. */
-	__u64	kernel_asce;			/* 0x0378 */
-	__u64	user_asce;			/* 0x0380 */
-	__u64	vdso_asce;			/* 0x0388 */
+	__u64	kernel_asce;			/* 0x0380 */
+	__u64	user_asce;			/* 0x0388 */
+	__u64	vdso_asce;			/* 0x0390 */
 
 	/*
 	 * The lpp and current_pid fields form a
 	 * 64-bit value that is set as program
 	 * parameter with the LPP instruction.
 	 */
-	__u32	lpp;				/* 0x0390 */
-	__u32	current_pid;			/* 0x0394 */
+	__u32	lpp;				/* 0x0398 */
+	__u32	current_pid;			/* 0x039c */
 
 	/* SMP info area */
-	__u32	cpu_nr;				/* 0x0398 */
-	__u32	softirq_pending;		/* 0x039c */
-	__u32	preempt_count;			/* 0x03a0 */
-	__u32	spinlock_lockval;		/* 0x03a4 */
-	__u32	spinlock_index;			/* 0x03a8 */
-	__u32	fpu_flags;			/* 0x03ac */
-	__u64	percpu_offset;			/* 0x03b0 */
-	__u64	vdso_per_cpu_data;		/* 0x03b8 */
-	__u64	machine_flags;			/* 0x03c0 */
-	__u64	gmap;				/* 0x03c8 */
-	__u8	pad_0x03d0[0x0400-0x03d0];	/* 0x03d0 */
+	__u32	cpu_nr;				/* 0x03a0 */
+	__u32	softirq_pending;		/* 0x03a4 */
+	__u32	preempt_count;			/* 0x03a8 */
+	__u32	spinlock_lockval;		/* 0x03ac */
+	__u32	spinlock_index;			/* 0x03b0 */
+	__u32	fpu_flags;			/* 0x03b4 */
+	__u64	percpu_offset;			/* 0x03b8 */
+	__u64	vdso_per_cpu_data;		/* 0x03c0 */
+	__u64	machine_flags;			/* 0x03c8 */
+	__u64	gmap;				/* 0x03d0 */
+	__u8	pad_0x03d8[0x0400-0x03d8];	/* 0x03d8 */
 
 	/* br %r1 trampoline */
 	__u16	br_r1_trampoline;		/* 0x0400 */

arch/s390/kernel/perf_cpum_cf_diag.c

···
  */
 static int __hw_perf_event_init(struct perf_event *event)
 {
-	struct cpu_cf_events *cpuhw = this_cpu_ptr(&cpu_cf_events);
 	struct perf_event_attr *attr = &event->attr;
+	struct cpu_cf_events *cpuhw;
 	enum cpumf_ctr_set i;
 	int err = 0;
 
-	debug_sprintf_event(cf_diag_dbg, 5,
-			    "%s event %p cpu %d authorized %#x\n", __func__,
-			    event, event->cpu, cpuhw->info.auth_ctl);
+	debug_sprintf_event(cf_diag_dbg, 5, "%s event %p cpu %d\n", __func__,
+			    event, event->cpu);
 
 	event->hw.config = attr->config;
 	event->hw.config_base = 0;
-	local64_set(&event->count, 0);
 
-	/* Add all authorized counter sets to config_base */
+	/* Add all authorized counter sets to config_base. The
+	 * the hardware init function is either called per-cpu or just once
+	 * for all CPUS (event->cpu == -1).  This depends on the whether
+	 * counting is started for all CPUs or on a per workload base where
+	 * the perf event moves from one CPU to another CPU.
+	 * Checking the authorization on any CPU is fine as the hardware
+	 * applies the same authorization settings to all CPUs.
+	 */
+	cpuhw = &get_cpu_var(cpu_cf_events);
 	for (i = CPUMF_CTR_SET_BASIC; i < CPUMF_CTR_SET_MAX; ++i)
 		if (cpuhw->info.auth_ctl & cpumf_ctr_ctl[i])
 			event->hw.config_base |= cpumf_ctr_ctl[i];
+	put_cpu_var(cpu_cf_events);
 
 	/* No authorized counter sets, nothing to count/sample */
 	if (!event->hw.config_base) {

arch/s390/kernel/smp.c

···
 	lc->percpu_offset = __per_cpu_offset[cpu];
 	lc->kernel_asce = S390_lowcore.kernel_asce;
 	lc->machine_flags = S390_lowcore.machine_flags;
-	lc->user_timer = lc->system_timer = lc->steal_timer = 0;
+	lc->user_timer = lc->system_timer =
+		lc->steal_timer = lc->avg_steal_timer = 0;
 	__ctl_store(lc->cregs_save_area, 0, 15);
 	save_access_regs((unsigned int *) lc->access_regs_save_area);
 	memcpy(lc->stfle_fac_list, S390_lowcore.stfle_fac_list,

arch/s390/kernel/vtime.c

···
  */
 static int do_account_vtime(struct task_struct *tsk)
 {
-	u64 timer, clock, user, guest, system, hardirq, softirq, steal;
+	u64 timer, clock, user, guest, system, hardirq, softirq;
 
 	timer = S390_lowcore.last_update_timer;
 	clock = S390_lowcore.last_update_clock;
···
 	if (softirq)
 		account_system_index_scaled(tsk, softirq, CPUTIME_SOFTIRQ);
 
-	steal = S390_lowcore.steal_timer;
-	if ((s64) steal > 0) {
-		S390_lowcore.steal_timer = 0;
-		account_steal_time(cputime_to_nsecs(steal));
-	}
-
 	return virt_timer_forward(user + guest + system + hardirq + softirq);
 }
 
···
  */
 void vtime_flush(struct task_struct *tsk)
 {
+	u64 steal, avg_steal;
+
 	if (do_account_vtime(tsk))
 		virt_timer_expire();
+
+	steal = S390_lowcore.steal_timer;
+	avg_steal = S390_lowcore.avg_steal_timer / 2;
+	if ((s64) steal > 0) {
+		S390_lowcore.steal_timer = 0;
+		account_steal_time(steal);
+		avg_steal += steal;
+	}
+	S390_lowcore.avg_steal_timer = avg_steal;
 }
 
 /*

drivers/s390/cio/chsc.c

···
 #include <asm/crw.h>
 #include <asm/isc.h>
 #include <asm/ebcdic.h>
+#include <asm/ap.h>
 
 #include "css.h"
 #include "cio.h"
···
 			      " failed (rc=%d).\n", ret);
 }
 
+static void chsc_process_sei_ap_cfg_chg(struct chsc_sei_nt0_area *sei_area)
+{
+	CIO_CRW_EVENT(3, "chsc: ap config changed\n");
+	if (sei_area->rs != 5)
+		return;
+
+	ap_bus_cfg_chg();
+}
+
 static void chsc_process_sei_nt2(struct chsc_sei_nt2_area *sei_area)
 {
 	switch (sei_area->cc) {
···
 		break;
 	case 2: /* i/o resource accessibility */
 		chsc_process_sei_res_acc(sei_area);
+		break;
+	case 3: /* ap config changed */
+		chsc_process_sei_ap_cfg_chg(sei_area);
 		break;
 	case 7: /* channel-path-availability information */
 		chsc_process_sei_chp_avail(sei_area);

drivers/s390/cio/vfio_ccw_drv.c

···
 {
 	struct vfio_ccw_private *private;
 	struct irb *irb;
+	bool is_final;
 
 	private = container_of(work, struct vfio_ccw_private, io_work);
 	irb = &private->irb;
 
+	is_final = !(scsw_actl(&irb->scsw) &
+		     (SCSW_ACTL_DEVACT | SCSW_ACTL_SCHACT));
 	if (scsw_is_solicited(&irb->scsw)) {
 		cp_update_scsw(&private->cp, &irb->scsw);
-		cp_free(&private->cp);
+		if (is_final)
+			cp_free(&private->cp);
 	}
 	memcpy(private->io_region->irb_area, irb, sizeof(*irb));
 
 	if (private->io_trigger)
 		eventfd_signal(private->io_trigger, 1);
 
-	if (private->mdev)
+	if (private->mdev && is_final)
 		private->state = VFIO_CCW_STATE_IDLE;
 }
 

drivers/s390/crypto/ap_bus.c

···
 	struct ap_device *ap_dev = to_ap_dev(dev);
 	struct ap_driver *ap_drv = ap_dev->drv;
 
+	/* prepare ap queue device removal */
 	if (is_queue_dev(dev))
-		ap_queue_remove(to_ap_queue(dev));
+		ap_queue_prepare_remove(to_ap_queue(dev));
+
+	/* driver's chance to clean up gracefully */
 	if (ap_drv->remove)
 		ap_drv->remove(ap_dev);
+
+	/* now do the ap queue device remove */
+	if (is_queue_dev(dev))
+		ap_queue_remove(to_ap_queue(dev));
 
 	/* Remove queue/card from list of active queues/cards */
 	spin_lock_bh(&ap_list_lock);
···
 	flush_work(&ap_scan_work);
 }
 EXPORT_SYMBOL(ap_bus_force_rescan);
+
+/*
+* A config change has happened, force an ap bus rescan.
+*/
+void ap_bus_cfg_chg(void)
+{
+	AP_DBF(DBF_INFO, "%s config change, forcing bus rescan\n", __func__);
+
+	ap_bus_force_rescan();
+}
 
 /*
  * hex2bitmap() - parse hex mask string and set bitmap.

drivers/s390/crypto/ap_bus.h

···
 	AP_STATE_WORKING,
 	AP_STATE_QUEUE_FULL,
 	AP_STATE_SUSPEND_WAIT,
+	AP_STATE_REMOVE,	/* about to be removed from driver */
 	AP_STATE_UNBOUND,	/* momentary not bound to a driver */
 	AP_STATE_BORKED,	/* broken */
 	NR_AP_STATES
···
 
 void ap_queue_init_reply(struct ap_queue *aq, struct ap_message *ap_msg);
 struct ap_queue *ap_queue_create(ap_qid_t qid, int device_type);
+void ap_queue_prepare_remove(struct ap_queue *aq);
 void ap_queue_remove(struct ap_queue *aq);
 void ap_queue_suspend(struct ap_device *ap_dev);
 void ap_queue_resume(struct ap_device *ap_dev);

drivers/s390/crypto/ap_queue.c

···
 		[AP_EVENT_POLL] = ap_sm_suspend_read,
 		[AP_EVENT_TIMEOUT] = ap_sm_nop,
 	},
+	[AP_STATE_REMOVE] = {
+		[AP_EVENT_POLL] = ap_sm_nop,
+		[AP_EVENT_TIMEOUT] = ap_sm_nop,
+	},
 	[AP_STATE_UNBOUND] = {
 		[AP_EVENT_POLL] = ap_sm_nop,
 		[AP_EVENT_TIMEOUT] = ap_sm_nop,
···
 }
 EXPORT_SYMBOL(ap_flush_queue);
 
+void ap_queue_prepare_remove(struct ap_queue *aq)
+{
+	spin_lock_bh(&aq->lock);
+	/* flush queue */
+	__ap_flush_queue(aq);
+	/* set REMOVE state to prevent new messages are queued in */
+	aq->state = AP_STATE_REMOVE;
+	del_timer_sync(&aq->timeout);
+	spin_unlock_bh(&aq->lock);
+}
+
 void ap_queue_remove(struct ap_queue *aq)
 {
-	ap_flush_queue(aq);
-	del_timer_sync(&aq->timeout);
-
-	/* reset with zero, also clears irq registration */
+	/*
+	 * all messages have been flushed and the state is
+	 * AP_STATE_REMOVE. Now reset with zero which also
+	 * clears the irq registration and move the state
+	 * to AP_STATE_UNBOUND to signal that this queue
+	 * is not used by any driver currently.
+	 */
 	spin_lock_bh(&aq->lock);
 	ap_zapq(aq->qid);
 	aq->state = AP_STATE_UNBOUND;
 	spin_unlock_bh(&aq->lock);
 }
-EXPORT_SYMBOL(ap_queue_remove);
 
 void ap_queue_reinit_state(struct ap_queue *aq)
 {
···
 	ap_wait(ap_sm_event(aq, AP_EVENT_POLL));
 	spin_unlock_bh(&aq->lock);
 }
-EXPORT_SYMBOL(ap_queue_reinit_state);

drivers/s390/crypto/zcrypt_api.c

···
 
 static inline struct zcrypt_queue *zcrypt_pick_queue(struct zcrypt_card *zc,
 						     struct zcrypt_queue *zq,
+						     struct module **pmod,
 						     unsigned int weight)
 {
 	if (!zq || !try_module_get(zq->queue->ap_dev.drv->driver.owner))
···
 	atomic_add(weight, &zc->load);
 	atomic_add(weight, &zq->load);
 	zq->request_count++;
+	*pmod = zq->queue->ap_dev.drv->driver.owner;
 	return zq;
 }
 
 static inline void zcrypt_drop_queue(struct zcrypt_card *zc,
 				     struct zcrypt_queue *zq,
+				     struct module *mod,
 				     unsigned int weight)
 {
-	struct module *mod = zq->queue->ap_dev.drv->driver.owner;
-
 	zq->request_count--;
 	atomic_sub(weight, &zc->load);
 	atomic_sub(weight, &zq->load);
···
 	unsigned int weight, pref_weight;
 	unsigned int func_code;
 	int qid = 0, rc = -ENODEV;
+	struct module *mod;
 
 	trace_s390_zcrypt_req(mex, TP_ICARSAMODEXPO);
 
···
 			pref_weight = weight;
 		}
 	}
-	pref_zq = zcrypt_pick_queue(pref_zc, pref_zq, weight);
+	pref_zq = zcrypt_pick_queue(pref_zc, pref_zq, &mod, weight);
 	spin_unlock(&zcrypt_list_lock);
 
 	if (!pref_zq) {
···
 	rc = pref_zq->ops->rsa_modexpo(pref_zq, mex);
 
 	spin_lock(&zcrypt_list_lock);
-	zcrypt_drop_queue(pref_zc, pref_zq, weight);
+	zcrypt_drop_queue(pref_zc, pref_zq, mod, weight);
 	spin_unlock(&zcrypt_list_lock);
 
 out:
···
 	unsigned int weight, pref_weight;
 	unsigned int func_code;
 	int qid = 0, rc = -ENODEV;
+	struct module *mod;
 
 	trace_s390_zcrypt_req(crt, TP_ICARSACRT);
 
···
 			pref_weight = weight;
 		}
 	}
-	pref_zq = zcrypt_pick_queue(pref_zc, pref_zq, weight);
+	pref_zq = zcrypt_pick_queue(pref_zc, pref_zq, &mod, weight);
 	spin_unlock(&zcrypt_list_lock);
 
 	if (!pref_zq) {
···
 	rc = pref_zq->ops->rsa_modexpo_crt(pref_zq, crt);
 
 	spin_lock(&zcrypt_list_lock);
-	zcrypt_drop_queue(pref_zc, pref_zq, weight);
+	zcrypt_drop_queue(pref_zc, pref_zq, mod, weight);
 	spin_unlock(&zcrypt_list_lock);
 
 out:
···
 	unsigned int func_code;
 	unsigned short *domain;
 	int qid = 0, rc = -ENODEV;
+	struct module *mod;
 
 	trace_s390_zcrypt_req(xcRB, TB_ZSECSENDCPRB);
 
···
 			pref_weight = weight;
 		}
 	}
-	pref_zq = zcrypt_pick_queue(pref_zc, pref_zq, weight);
+	pref_zq = zcrypt_pick_queue(pref_zc, pref_zq, &mod, weight);
 	spin_unlock(&zcrypt_list_lock);
 
 	if (!pref_zq) {
···
 	rc = pref_zq->ops->send_cprb(pref_zq, xcRB, &ap_msg);
 
 	spin_lock(&zcrypt_list_lock);
-	zcrypt_drop_queue(pref_zc, pref_zq, weight);
+	zcrypt_drop_queue(pref_zc, pref_zq, mod, weight);
 	spin_unlock(&zcrypt_list_lock);
 
 out:
···
 	unsigned int func_code;
 	struct ap_message ap_msg;
 	int qid = 0, rc = -ENODEV;
+	struct module *mod;
 
 	trace_s390_zcrypt_req(xcrb, TP_ZSENDEP11CPRB);
 
···
 			pref_weight = weight;
 		}
 	}
-	pref_zq = zcrypt_pick_queue(pref_zc, pref_zq, weight);
+	pref_zq = zcrypt_pick_queue(pref_zc, pref_zq, &mod, weight);
 	spin_unlock(&zcrypt_list_lock);
 
 	if (!pref_zq) {
···
 	rc = pref_zq->ops->send_ep11_cprb(pref_zq, xcrb, &ap_msg);
 
 	spin_lock(&zcrypt_list_lock);
-	zcrypt_drop_queue(pref_zc, pref_zq, weight);
+	zcrypt_drop_queue(pref_zc, pref_zq, mod, weight);
 	spin_unlock(&zcrypt_list_lock);
 
 out_free:
···
 	struct ap_message ap_msg;
 	unsigned int domain;
 	int qid = 0, rc = -ENODEV;
+	struct module *mod;
 
 	trace_s390_zcrypt_req(buffer, TP_HWRNGCPRB);
 
···
 			pref_weight = weight;
 		}
 	}
-	pref_zq = zcrypt_pick_queue(pref_zc, pref_zq, weight);
+	pref_zq = zcrypt_pick_queue(pref_zc, pref_zq, &mod, weight);
 	spin_unlock(&zcrypt_list_lock);
 
 	if (!pref_zq) {
···
 	rc = pref_zq->ops->rng(pref_zq, buffer, &ap_msg);
 
 	spin_lock(&zcrypt_list_lock);
-	zcrypt_drop_queue(pref_zc, pref_zq, weight);
+	zcrypt_drop_queue(pref_zc, pref_zq, mod, weight);
 	spin_unlock(&zcrypt_list_lock);
 
 out: