Skip to contentSkip to conversation

ALSA: seq: Cover unsubscribe_port() in list_mutex

PATCH In review Sasha Levin 2 participants 11 messages Jun 4, 2019 – Jun 5, 2019

Subsystems (8)
lore
Share & tools
lore

Sasha Levin

author maintainer
#0 lore raw

PATCH 1/10 ALSA: seq: Cover unsubscribe_port() in list_mutex

From: Takashi Iwai <[email protected]>

[ Upstream commit 7c32ae35fbf9cffb7aa3736f44dec10c944ca18e ]

The call of unsubscribe_port() which manages the group count and module refcount from delete_and_unsubscribe_port() looks racy; it's not covered by the group list lock, and it's likely a cause of the reported unbalance at port deletion. Let's move the call inside the group list_mutex to plug the hole.

---
 sound/core/seq/seq_ports.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/core/seq/seq_ports.c +1 -1
index f04714d70bf7..a42e2ce4a726 100644
@@ -550,10 +550,10 @@ static void delete_and_unsubscribe_port(struct snd_seq_client *client,
 		list_del_init(list);
 	grp->exclusive = 0;
 	write_unlock_irq(&grp->list_lock);
-	up_write(&grp->list_mutex);
 
 	if (!empty)
 		unsubscribe_port(client, port, grp, &subs->info, ack);
+	up_write(&grp->list_mutex);
 }
 
 /* connect two ports */
Signature · 1 line
2.20.1
author maintainer
#1 lore raw

PATCH 2/10 driver core: platform: Fix the usage of platform device name(pdev->name)

From: Venkata Narendra Kumar Gutta <[email protected]>

[ Upstream commit edb16da34b084c66763f29bee42b4e6bb33c3d66 ]

Platform core is using pdev->name as the platform device name to do the binding of the devices with the drivers. But, when the platform driver overrides the platform device name with dev_set_name(), the pdev->name is pointing to a location which is freed and becomes an invalid parameter to do the binding match.

use-after-free instance:

[   33.325013] BUG: KASAN: use-after-free in strcmp+0x8c/0xb0
[   33.330646] Read of size 1 at addr ffffffc10beae600 by task modprobe
[   33.339068] CPU: 5 PID: 518 Comm: modprobe Tainted:
			G S      W  O      4.19.30+ #3
[   33.346835] Hardware name: MTP (DT)
[   33.350419] Call trace:
[   33.352941]  dump_backtrace+0x0/0x3b8
[   33.356713]  show_stack+0x24/0x30
[   33.360119]  dump_stack+0x160/0x1d8
[   33.363709]  print_address_description+0x84/0x2e0
[   33.368549]  kasan_report+0x26c/0x2d0
[   33.372322]  __asan_report_load1_noabort+0x2c/0x38
[   33.377248]  strcmp+0x8c/0xb0
[   33.380306]  platform_match+0x70/0x1f8
[   33.384168]  __driver_attach+0x78/0x3a0
[   33.388111]  bus_for_each_dev+0x13c/0x1b8
[   33.392237]  driver_attach+0x4c/0x58
[   33.395910]  bus_add_driver+0x350/0x560
[   33.399854]  driver_register+0x23c/0x328
[   33.403886]  __platform_driver_register+0xd0/0xe0

So, use dev_name(&pdev->dev), which fetches the platform device name from the kobject(dev->kobj->name) of the device instead of the pdev->name.

---
 drivers/base/platform.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/base/platform.c +4 -4
index 065fcc4be263..071dd053a917 100644
@@ -796,7 +796,7 @@ static ssize_t modalias_show(struct device *dev, struct device_attribute *a,
 	if (len != -ENODEV)
 		return len;
 
-	len = snprintf(buf, PAGE_SIZE, "platform:%s\n", pdev->name);
+	len = snprintf(buf, PAGE_SIZE, "platform:%s\n", dev_name(&pdev->dev));
 
 	return (len >= PAGE_SIZE) ? (PAGE_SIZE - 1) : len;
 }
@@ -872,7 +872,7 @@ static int platform_uevent(struct device *dev, struct kobj_uevent_env *env)
 		return rc;
 
 	add_uevent_var(env, "MODALIAS=%s%s", PLATFORM_MODULE_PREFIX,
-			pdev->name);
+			dev_name(&pdev->dev));
 	return 0;
 }
 
@@ -881,7 +881,7 @@ static const struct platform_device_id *platform_match_id(
 			struct platform_device *pdev)
 {
 	while (id->name[0]) {
-		if (strcmp(pdev->name, id->name) == 0) {
+		if (strcmp(dev_name(&pdev->dev), id->name) == 0) {
 			pdev->id_entry = id;
 			return id;
 		}
@@ -925,7 +925,7 @@ static int platform_match(struct device *dev, struct device_driver *drv)
 		return platform_match_id(pdrv->id_table, pdev) != NULL;
 
 	/* fall-back to driver name match */
-	return (strcmp(pdev->name, drv->name) == 0);
+	return (strcmp(dev_name(&pdev->dev), drv->name) == 0);
 }
 
 #ifdef CONFIG_PM_SLEEP
Signature · 1 line
2.20.1
author maintainer
#2 lore raw

PATCH 3/10 Drivers: misc: fix out-of-bounds access in function param_set_kgdbts_var

In reply to Sasha Levin

From: Young Xiao <[email protected]>

[ Upstream commit b281218ad4311a0342a40cb02fb17a363df08b48 ]

There is an out-of-bounds access to "config[len - 1]" array when the variable "len" is zero.

See commit dada6a43b040 ("kgdboc: fix KASAN global-out-of-bounds bug in param_set_kgdboc_var()") for details.

---
 drivers/misc/kgdbts.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/misc/kgdbts.c +2 -2
index 99635dd9dbac..bb3a76ad80da 100644
@@ -1132,7 +1132,7 @@ static void kgdbts_put_char(u8 chr)
 
 static int param_set_kgdbts_var(const char *kmessage, struct kernel_param *kp)
 {
-	int len = strlen(kmessage);
+	size_t len = strlen(kmessage);
 
 	if (len >= MAX_CONFIG_LEN) {
 		printk(KERN_ERR "kgdbts: config string too long\n");
@@ -1152,7 +1152,7 @@ static int param_set_kgdbts_var(const char *kmessage, struct kernel_param *kp)
 
 	strcpy(config, kmessage);
 	/* Chop out \n char as a result of echo */
-	if (config[len - 1] == '\n')
+	if (len && config[len - 1] == '\n')
 		config[len - 1] = '\0';
 
 	/* Go and configure with the new params. */
Signature · 1 line
2.20.1
author maintainer
#3 lore raw

PATCH 4/10 ipc: prevent lockup on alloc_msg and free_msg

In reply to Sasha Levin

From: Li Rongqing <[email protected]>

[ Upstream commit d6a2946a88f524a47cc9b79279667137899db807 ]

msgctl10 of ltp triggers the following lockup When CONFIG_KASAN is enabled on large memory SMP systems, the pages initialization can take a long time, if msgctl10 requests a huge block memory, and it will block rcu scheduler, so release cpu actively.

After adding schedule() in free_msg, free_msg can not be called when holding spinlock, so adding msg to a tmp list, and free it out of spinlock

  rcu: INFO: rcu_preempt detected stalls on CPUs/tasks:
  rcu:     Tasks blocked on level-1 rcu_node (CPUs 16-31): P32505
  rcu:     Tasks blocked on level-1 rcu_node (CPUs 48-63): P34978
  rcu:     (detected by 11, t=35024 jiffies, g=44237529, q=16542267)
  msgctl10        R  running task    21608 32505   2794 0x00000082
  Call Trace:
   preempt_schedule_irq+0x4c/0xb0
   retint_kernel+0x1b/0x2d
  RIP: 0010:__is_insn_slot_addr+0xfb/0x250
  Code: 82 1d 00 48 8b 9b 90 00 00 00 4c 89 f7 49 c1 ee 03 e8 59 83 1d 00 48 b8 00 00 00 00 00 fc ff df 4c 39 eb 48 89 9d 58 ff ff ff <41> c6 04 06 f8 74 66 4c 8d 75 98 4c 89 f1 48 c1 e9 03 48 01 c8 48
  RSP: 0018:ffff88bce041f758 EFLAGS: 00000246 ORIG_RAX: ffffffffffffff13
  RAX: dffffc0000000000 RBX: ffffffff8471bc50 RCX: ffffffff828a2a57
  RDX: dffffc0000000000 RSI: dffffc0000000000 RDI: ffff88bce041f780
  RBP: ffff88bce041f828 R08: ffffed15f3f4c5b3 R09: ffffed15f3f4c5b3
  R10: 0000000000000001 R11: ffffed15f3f4c5b2 R12: 000000318aee9b73
  R13: ffffffff8471bc50 R14: 1ffff1179c083ef0 R15: 1ffff1179c083eec
   kernel_text_address+0xc1/0x100
   __kernel_text_address+0xe/0x30
   unwind_get_return_address+0x2f/0x50
   __save_stack_trace+0x92/0x100
   create_object+0x380/0x650
   __kmalloc+0x14c/0x2b0
   load_msg+0x38/0x1a0
   do_msgsnd+0x19e/0xcf0
   do_syscall_64+0x117/0x400
   entry_SYSCALL_64_after_hwframe+0x49/0xbe

  rcu: INFO: rcu_preempt detected stalls on CPUs/tasks:
  rcu:     Tasks blocked on level-1 rcu_node (CPUs 0-15): P32170
  rcu:     (detected by 14, t=35016 jiffies, g=44237525, q=12423063)
  msgctl10        R  running task    21608 32170  32155 0x00000082
  Call Trace:
   preempt_schedule_irq+0x4c/0xb0
   retint_kernel+0x1b/0x2d
  RIP: 0010:lock_acquire+0x4d/0x340
  Code: 48 81 ec c0 00 00 00 45 89 c6 4d 89 cf 48 8d 6c 24 20 48 89 3c 24 48 8d bb e4 0c 00 00 89 74 24 0c 48 c7 44 24 20 b3 8a b5 41 <48> c1 ed 03 48 c7 44 24 28 b4 25 18 84 48 c7 44 24 30 d0 54 7a 82
  RSP: 0018:ffff88af83417738 EFLAGS: 00000282 ORIG_RAX: ffffffffffffff13
  RAX: dffffc0000000000 RBX: ffff88bd335f3080 RCX: 0000000000000002
  RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff88bd335f3d64
  RBP: ffff88af83417758 R08: 0000000000000000 R09: 0000000000000000
  R10: 0000000000000001 R11: ffffed13f3f745b2 R12: 0000000000000000
  R13: 0000000000000002 R14: 0000000000000000 R15: 0000000000000000
   is_bpf_text_address+0x32/0xe0
   kernel_text_address+0xec/0x100
   __kernel_text_address+0xe/0x30
   unwind_get_return_address+0x2f/0x50
   __save_stack_trace+0x92/0x100
   save_stack+0x32/0xb0
   __kasan_slab_free+0x130/0x180
   kfree+0xfa/0x2d0
   free_msg+0x24/0x50
   do_msgrcv+0x508/0xe60
   do_syscall_64+0x117/0x400
   entry_SYSCALL_64_after_hwframe+0x49/0xbe

Davidlohr said:
"So after releasing the lock, the msg rbtree/list is empty and new
calls will not see those in the newly populated tmp_msg list, and therefore they cannot access the delayed msg freeing pointers, which is good. Also the fact that the node_cache is now freed before the actual messages seems to be harmless as this is wanted for msg_insert() avoiding GFP_ATOMIC allocations, and after releasing the info->lock the thing is freed anyway so it should not change things"

---
 ipc/mqueue.c  | 10 ++++++++--
 ipc/msgutil.c |  6 ++++++
 2 files changed, 14 insertions(+), 2 deletions(-)
ipc/mqueue.c +8 -2
index 5e24eb0ab5dd..6ed74825ab54 100644
@@ -373,7 +373,8 @@ static void mqueue_evict_inode(struct inode *inode)
 	struct user_struct *user;
 	unsigned long mq_bytes, mq_treesize;
 	struct ipc_namespace *ipc_ns;
-	struct msg_msg *msg;
+	struct msg_msg *msg, *nmsg;
+	LIST_HEAD(tmp_msg);
 
 	clear_inode(inode);
 
@@ -384,10 +385,15 @@ static void mqueue_evict_inode(struct inode *inode)
 	info = MQUEUE_I(inode);
 	spin_lock(&info->lock);
 	while ((msg = msg_get(info)) != NULL)
-		free_msg(msg);
+		list_add_tail(&msg->m_list, &tmp_msg);
 	kfree(info->node_cache);
 	spin_unlock(&info->lock);
 
+	list_for_each_entry_safe(msg, nmsg, &tmp_msg, m_list) {
+		list_del(&msg->m_list);
+		free_msg(msg);
+	}
+
 	/* Total amount of bytes accounted for the mqueue */
 	mq_treesize = info->attr.mq_maxmsg * sizeof(struct msg_msg) +
 		min_t(unsigned int, info->attr.mq_maxmsg, MQ_PRIO_MAX) *
ipc/msgutil.c +6 -0
index ed81aafd2392..9467307487f7 100644
@@ -18,6 +18,7 @@
 #include <linux/utsname.h>
 #include <linux/proc_ns.h>
 #include <linux/uaccess.h>
+#include <linux/sched.h>
 
 #include "util.h"
 
@@ -66,6 +67,9 @@ static struct msg_msg *alloc_msg(size_t len)
 	pseg = &msg->next;
 	while (len > 0) {
 		struct msg_msgseg *seg;
+
+		cond_resched();
+
 		alen = min(len, DATALEN_SEG);
 		seg = kmalloc(sizeof(*seg) + alen, GFP_KERNEL);
 		if (seg == NULL)
@@ -178,6 +182,8 @@ void free_msg(struct msg_msg *msg)
 	kfree(msg);
 	while (seg != NULL) {
 		struct msg_msgseg *tmp = seg->next;
+
+		cond_resched();
 		kfree(seg);
 		seg = tmp;
 	}
Signature · 1 line
2.20.1
author maintainer
#4 lore raw

PATCH 5/10 scsi: lpfc: add check for loss of ndlp when sending RRQ

In reply to Sasha Levin

From: James Smart <[email protected]>

[ Upstream commit c8cb261a072c88ca1aff0e804a30db4c7606521b ]

There was a missing qualification of a valid ndlp structure when calling to send an RRQ for an abort. Add the check.

---
 drivers/scsi/lpfc/lpfc_els.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/scsi/lpfc/lpfc_els.c +4 -1
index 398c9a0a5ade..82a690924f5e 100644
@@ -6498,7 +6498,10 @@ int
 lpfc_send_rrq(struct lpfc_hba *phba, struct lpfc_node_rrq *rrq)
 {
 	struct lpfc_nodelist *ndlp = lpfc_findnode_did(rrq->vport,
-							rrq->nlp_DID);
+						       rrq->nlp_DID);
+	if (!ndlp)
+		return 1;
+
 	if (lpfc_test_rrq_active(phba, ndlp, rrq->xritag))
 		return lpfc_issue_els_rrq(rrq->vport, ndlp,
 					 rrq->nlp_DID, rrq);
Signature · 1 line
2.20.1
author maintainer
#5 lore raw

PATCH 6/10 scsi: bnx2fc: fix incorrect cast to u64 on shift operation

In reply to Sasha Levin

From: Colin Ian King <[email protected]>

[ Upstream commit d0c0d902339249c75da85fd9257a86cbb98dfaa5 ]

Currently an int is being shifted and the result is being cast to a u64 which leads to undefined behaviour if the shift is more than 31 bits. Fix this by casting the integer value 1 to u64 before the shift operation.

Addresses-Coverity: ("Bad shift operation")

Fixes 7b594769120b ("[SCSI] bnx2fc: Handle REC_TOV error code from firmware")
---
 drivers/scsi/bnx2fc/bnx2fc_hwi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/bnx2fc/bnx2fc_hwi.c +1 -1
index 28c671b609b2..0c71b69b9f88 100644
@@ -829,7 +829,7 @@ static void bnx2fc_process_unsol_compl(struct bnx2fc_rport *tgt, u16 wqe)
 			((u64)err_entry->data.err_warn_bitmap_hi << 32) |
 			(u64)err_entry->data.err_warn_bitmap_lo;
 		for (i = 0; i < BNX2FC_NUM_ERR_BITS; i++) {
-			if (err_warn_bit_map & (u64) (1 << i)) {
+			if (err_warn_bit_map & ((u64)1 << i)) {
 				err_warn = i;
 				break;
 			}
Signature · 1 line
2.20.1
author maintainer
#6 lore raw

PATCH 7/10 usbnet: ipheth: fix racing condition

In reply to Sasha Levin

From: Bernd Eckstein <[email protected]>

[ Upstream commit 94d250fae48e6f873d8362308f5c4d02cd1b1fd2 ]

Fix a racing condition in ipheth.c that can lead to slow performance.

Bug: In ipheth_tx(), netif_wake_queue() may be called on the callback ipheth_sndbulk_callback(), _before_ netif_stop_queue() is called. When this happens, the queue is stopped longer than it needs to be, thus reducing network performance.

Fix: Move netif_stop_queue() in front of usb_submit_urb(). Now the order is always correct. In case, usb_submit_urb() fails, the queue is woken up again as callback will not fire.

Testing: This racing condition is usually not noticeable, as it has to occur very frequently to slowdown the network. The callback from the USB is usually triggered slow enough, so the situation does not appear. However, on a Ubuntu Linux on VMWare Workstation, running on Windows 10, the we loose the race quite often and the following speedup can be noticed:

Without this patch: Download:  4.10 Mbit/s, Upload:  4.01 Mbit/s
With this patch:    Download: 36.23 Mbit/s, Upload: 17.61 Mbit/s
---
 drivers/net/usb/ipheth.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/usb/ipheth.c +2 -1
index 01f95d192d25..2b16a5fed9de 100644
@@ -437,17 +437,18 @@ static int ipheth_tx(struct sk_buff *skb, struct net_device *net)
 			  dev);
 	dev->tx_urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
 
+	netif_stop_queue(net);
 	retval = usb_submit_urb(dev->tx_urb, GFP_ATOMIC);
 	if (retval) {
 		dev_err(&dev->intf->dev, "%s: usb_submit_urb: %d\n",
 			__func__, retval);
 		dev->net->stats.tx_errors++;
 		dev_kfree_skb_any(skb);
+		netif_wake_queue(net);
 	} else {
 		dev->net->stats.tx_packets++;
 		dev->net->stats.tx_bytes += skb->len;
 		dev_consume_skb_any(skb);
-		netif_stop_queue(net);
 	}
 
 	return NETDEV_TX_OK;
Signature · 1 line
2.20.1
author maintainer
#7 lore raw

PATCH 8/10 usbnet: fix kernel crash after disconnect

In reply to Sasha Levin

From: Kloetzke Jan <[email protected]>

[ Upstream commit ad70411a978d1e6e97b1e341a7bde9a79af0c93d ]

When disconnecting cdc_ncm the kernel sporadically crashes shortly after the disconnect:

  [   57.868812] Unable to handle kernel NULL pointer dereference at virtual address 00000000
  ...
  [   58.006653] PC is at 0x0
  [   58.009202] LR is at call_timer_fn+0xec/0x1b4
  [   58.013567] pc : [<0000000000000000>] lr : [<ffffff80080f5130>] pstate: 00000145
  [   58.020976] sp : ffffff8008003da0
  [   58.024295] x29: ffffff8008003da0 x28: 0000000000000001
  [   58.029618] x27: 000000000000000a x26: 0000000000000100
  [   58.034941] x25: 0000000000000000 x24: ffffff8008003e68
  [   58.040263] x23: 0000000000000000 x22: 0000000000000000
  [   58.045587] x21: 0000000000000000 x20: ffffffc68fac1808
  [   58.050910] x19: 0000000000000100 x18: 0000000000000000
  [   58.056232] x17: 0000007f885aff8c x16: 0000007f883a9f10
  [   58.061556] x15: 0000000000000001 x14: 000000000000006e
  [   58.066878] x13: 0000000000000000 x12: 00000000000000ba
  [   58.072201] x11: ffffffc69ff1db30 x10: 0000000000000020
  [   58.077524] x9 : 8000100008001000 x8 : 0000000000000001
  [   58.082847] x7 : 0000000000000800 x6 : ffffff8008003e70
  [   58.088169] x5 : ffffffc69ff17a28 x4 : 00000000ffff138b
  [   58.093492] x3 : 0000000000000000 x2 : 0000000000000000
  [   58.098814] x1 : 0000000000000000 x0 : 0000000000000000
  ...
  [   58.205800] [<          (null)>]           (null)
  [   58.210521] [<ffffff80080f5298>] expire_timers+0xa0/0x14c
  [   58.215937] [<ffffff80080f542c>] run_timer_softirq+0xe8/0x128
  [   58.221702] [<ffffff8008081120>] __do_softirq+0x298/0x348
  [   58.227118] [<ffffff80080a6304>] irq_exit+0x74/0xbc
  [   58.232009] [<ffffff80080e17dc>] __handle_domain_irq+0x78/0xac
  [   58.237857] [<ffffff8008080cf4>] gic_handle_irq+0x80/0xac
  ...

The crash happens roughly 125..130ms after the disconnect. This correlates with the 'delay' timer that is started on certain USB tx/rx errors in the URB completion handler.

The problem is a race of usbnet_stop() with usbnet_start_xmit(). In usbnet_stop() we call usbnet_terminate_urbs() to cancel all URBs in flight. This only makes sense if no new URBs are submitted concurrently, though. But the usbnet_start_xmit() can run at the same time on another CPU which almost unconditionally submits an URB. The error callback of the new URB will then schedule the timer after it was already stopped.

The fix adds a check if the tx queue is stopped after the tx list lock has been taken. This should reliably prevent the submission of new URBs while usbnet_terminate_urbs() does its job. The same thing is done on the rx side even though it might be safe due to other flags that are checked there.

---
 drivers/net/usb/usbnet.c | 6 ++++++
 1 file changed, 6 insertions(+)
drivers/net/usb/usbnet.c +6 -0
index 9710cf71054a..2502681369cd 100644
@@ -499,6 +499,7 @@ static int rx_submit (struct usbnet *dev, struct urb *urb, gfp_t flags)
 
 	if (netif_running (dev->net) &&
 	    netif_device_present (dev->net) &&
+	    test_bit(EVENT_DEV_OPEN, &dev->flags) &&
 	    !test_bit (EVENT_RX_HALT, &dev->flags) &&
 	    !test_bit (EVENT_DEV_ASLEEP, &dev->flags)) {
 		switch (retval = usb_submit_urb (urb, GFP_ATOMIC)) {
@@ -1385,6 +1386,11 @@ netdev_tx_t usbnet_start_xmit (struct sk_buff *skb,
 		spin_unlock_irqrestore(&dev->txq.lock, flags);
 		goto drop;
 	}
+	if (netif_queue_stopped(net)) {
+		usb_autopm_put_interface_async(dev->intf);
+		spin_unlock_irqrestore(&dev->txq.lock, flags);
+		goto drop;
+	}
 
 #ifdef CONFIG_PM
 	/* if this triggers the device is still a sleep */
Signature · 1 line
2.20.1
author maintainer
#8 lore raw

PATCH 9/10 KVM: x86/pmu: do not mask the value that is written to fixed PMUs

In reply to Sasha Levin

From: Paolo Bonzini <[email protected]>

[ Upstream commit 2924b52117b2812e9633d5ea337333299166d373 ]

According to the SDM, for MSR_IA32_PERFCTR0/1 "the lower-order 32 bits of each MSR may be written with any value, and the high-order 8 bits are sign-extended according to the value of bit 31", but the fixed counters in real hardware are limited to the width of the fixed counters ("bits beyond the width of the fixed-function counter are reserved and must be written as zeros"). Fix KVM to do the same.

---
 arch/x86/kvm/pmu_intel.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)
arch/x86/kvm/pmu_intel.c +8 -5
index 23a7c7ba377a..8fc07ea23344 100644
@@ -235,11 +235,14 @@ static int intel_pmu_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 		}
 		break;
 	default:
-		if ((pmc = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0)) ||
-		    (pmc = get_fixed_pmc(pmu, msr))) {
-			if (!msr_info->host_initiated)
-				data = (s64)(s32)data;
-			pmc->counter += data - pmc_read_counter(pmc);
+		if ((pmc = get_gp_pmc(pmu, msr, MSR_IA32_PERFCTR0))) {
+			if (msr_info->host_initiated)
+				pmc->counter = data;
+			else
+				pmc->counter = (s32)data;
+			return 0;
+		} else if ((pmc = get_fixed_pmc(pmu, msr))) {
+			pmc->counter = data;
 			return 0;
 		} else if ((pmc = get_gp_pmc(pmu, msr, MSR_P6_EVNTSEL0))) {
 			if (data == pmc->eventsel)
Signature · 1 line
2.20.1
author maintainer
#9 lore raw

PATCH 10/10 KVM: s390: fix memory slot handling for KVM_SET_USER_MEMORY_REGION

In reply to Sasha Levin

From: Christian Borntraeger <[email protected]>

[ Upstream commit 19ec166c3f39fe1d3789888a74cc95544ac266d4 ]

kselftests exposed a problem in the s390 handling for memory slots. Right now we only do proper memory slot handling for creation of new memory slots. Neither MOVE, nor DELETION are handled properly. Let us implement those.

---
 arch/s390/kvm/kvm-s390.c | 35 +++++++++++++++++++++--------------
 1 file changed, 21 insertions(+), 14 deletions(-)
arch/s390/kvm/kvm-s390.c +21 -14
index 5ddb1debba95..23911ecfbad6 100644
@@ -2721,21 +2721,28 @@ void kvm_arch_commit_memory_region(struct kvm *kvm,
 				const struct kvm_memory_slot *new,
 				enum kvm_mr_change change)
 {
-	int rc;
-
-	/* If the basics of the memslot do not change, we do not want
-	 * to update the gmap. Every update causes several unnecessary
-	 * segment translation exceptions. This is usually handled just
-	 * fine by the normal fault handler + gmap, but it will also
-	 * cause faults on the prefix page of running guest CPUs.
-	 */
-	if (old->userspace_addr == mem->userspace_addr &&
-	    old->base_gfn * PAGE_SIZE == mem->guest_phys_addr &&
-	    old->npages * PAGE_SIZE == mem->memory_size)
-		return;
+	int rc = 0;
 
-	rc = gmap_map_segment(kvm->arch.gmap, mem->userspace_addr,
-		mem->guest_phys_addr, mem->memory_size);
+	switch (change) {
+	case KVM_MR_DELETE:
+		rc = gmap_unmap_segment(kvm->arch.gmap, old->base_gfn * PAGE_SIZE,
+					old->npages * PAGE_SIZE);
+		break;
+	case KVM_MR_MOVE:
+		rc = gmap_unmap_segment(kvm->arch.gmap, old->base_gfn * PAGE_SIZE,
+					old->npages * PAGE_SIZE);
+		if (rc)
+			break;
+		/* FALLTHROUGH */
+	case KVM_MR_CREATE:
+		rc = gmap_map_segment(kvm->arch.gmap, mem->userspace_addr,
+				      mem->guest_phys_addr, mem->memory_size);
+		break;
+	case KVM_MR_FLAGS_ONLY:
+		break;
+	default:
+		WARN(1, "Unknown KVM MR CHANGE: %d\n", change);
+	}
 	if (rc)
 		pr_warn("failed to commit memory region\n");
 	return;
Signature · 1 line
2.20.1
#10 lore raw

Re: [PATCH AUTOSEL 4.4 10/10] KVM: s390: fix memory slot handling for KVM_SET_USER_MEMORY_REGION

Quoted 8 lines · Sasha Levin

From: Christian Borntraeger <[email protected]>

[ Upstream commit 19ec166c3f39fe1d3789888a74cc95544ac266d4 ]

kselftests exposed a problem in the s390 handling for memory slots. Right now we only do proper memory slot handling for creation of new memory slots. Neither MOVE, nor DELETION are handled properly. Let us implement those.

I gave this a quick spin on 4.4.y and it seems to work fine.

Quoted 56 lines
---
 arch/s390/kvm/kvm-s390.c | 35 +++++++++++++++++++++--------------
 1 file changed, 21 insertions(+), 14 deletions(-)
arch/s390/kvm/kvm-s390.c +21 -14
index 5ddb1debba95..23911ecfbad6 100644
@@ -2721,21 +2721,28 @@ void kvm_arch_commit_memory_region(struct kvm *kvm,
 				const struct kvm_memory_slot *new,
 				enum kvm_mr_change change)
 {
-	int rc;
-
-	/* If the basics of the memslot do not change, we do not want
-	 * to update the gmap. Every update causes several unnecessary
-	 * segment translation exceptions. This is usually handled just
-	 * fine by the normal fault handler + gmap, but it will also
-	 * cause faults on the prefix page of running guest CPUs.
-	 */
-	if (old->userspace_addr == mem->userspace_addr &&
-	    old->base_gfn * PAGE_SIZE == mem->guest_phys_addr &&
-	    old->npages * PAGE_SIZE == mem->memory_size)
-		return;
+	int rc = 0;
 
-	rc = gmap_map_segment(kvm->arch.gmap, mem->userspace_addr,
-		mem->guest_phys_addr, mem->memory_size);
+	switch (change) {
+	case KVM_MR_DELETE:
+		rc = gmap_unmap_segment(kvm->arch.gmap, old->base_gfn * PAGE_SIZE,
+					old->npages * PAGE_SIZE);
+		break;
+	case KVM_MR_MOVE:
+		rc = gmap_unmap_segment(kvm->arch.gmap, old->base_gfn * PAGE_SIZE,
+					old->npages * PAGE_SIZE);
+		if (rc)
+			break;
+		/* FALLTHROUGH */
+	case KVM_MR_CREATE:
+		rc = gmap_map_segment(kvm->arch.gmap, mem->userspace_addr,
+				      mem->guest_phys_addr, mem->memory_size);
+		break;
+	case KVM_MR_FLAGS_ONLY:
+		break;
+	default:
+		WARN(1, "Unknown KVM MR CHANGE: %d\n", change);
+	}
 	if (rc)
 		pr_warn("failed to commit memory region\n");
 	return;