Skip to contentSkip to conversation

autofs: fix a leak in autofs_expire_indirect()

PATCH In review Sasha Levin 2 participants 27 messages Nov 22, 2019 – Dec 6, 2019

Subsystems (13)
lore
Share & tools
lore

Sasha Levin

author maintainer
#0 lore raw

PATCH 1/25 autofs: fix a leak in autofs_expire_indirect()

From: Al Viro <[email protected]>

[ Upstream commit 03ad0d703df75c43f78bd72e16124b5b94a95188 ]

if the second call of should_expire() in there ends up grabbing and returning a new reference to dentry, we need to drop it before continuing.

---
 fs/autofs/expire.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
fs/autofs/expire.c +3 -2
index 28d9c2b1b3bb3..70e9afe589fbf 100644
@@ -501,9 +501,10 @@ static struct dentry *autofs_expire_indirect(struct super_block *sb,
 		 */
 		how &= ~AUTOFS_EXP_LEAVES;
 		found = should_expire(expired, mnt, timeout, how);
-		if (!found || found != expired)
-			/* Something has changed, continue */
+		if (found != expired) { // something has changed, continue
+			dput(found);
 			goto next;
+		}
 
 		if (expired != dentry)
 			dput(dentry);
Signature · 1 line
2.20.1
author maintainer
#5 lore raw

PATCH 6/25 audit_get_nd(): don't unlock parent too early

From: Al Viro <[email protected]>

[ Upstream commit 69924b89687a2923e88cc42144aea27868913d0e ]

if the child has been negative and just went positive under us, we want coherent d_is_positive() and ->d_inode. Don't unlock the parent until we'd done that work...

---
 kernel/audit_watch.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/audit_watch.c +1 -1
index 787c7afdf8294..4f7262eba73d8 100644
@@ -366,12 +366,12 @@ static int audit_get_nd(struct audit_watch *watch, struct path *parent)
 	struct dentry *d = kern_path_locked(watch->path, parent);
 	if (IS_ERR(d))
 		return PTR_ERR(d);
-	inode_unlock(d_backing_inode(parent->dentry));
 	if (d_is_positive(d)) {
 		/* update watch filter fields */
 		watch->dev = d->d_sb->s_dev;
 		watch->ino = d_backing_inode(d)->i_ino;
 	}
+	inode_unlock(d_backing_inode(parent->dentry));
 	dput(d);
 	return 0;
 }
Signature · 1 line
2.20.1
author maintainer
#12 lore raw

PATCH 13/25 drm/sun4i: tcon: Set min division of TCON0_DCLK to 1.

In reply to Sasha Levin

From: Yunhao Tian <[email protected]>

[ Upstream commit 0b8e7bbde5e7e2c419567e1ee29587dae3b78ee3 ]

The datasheet of V3s (and various other chips) wrote that TCON0_DCLK_DIV can be >= 1 if only dclk is used, and must >= 6 if dclk1 or dclk2 is used. As currently neither dclk1 nor dclk2 is used (no writes to these bits), let's set minimal division to 1.

If this minimal division is 6, some common dot clock frequencies can't be produced (e.g. 30MHz will not be possible and will fallback to 25MHz), which is obviously not an expected behaviour.

---
 drivers/gpu/drm/sun4i/sun4i_tcon.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/sun4i/sun4i_tcon.c +1 -1
index 8c31c9ab06f8b..fda1ae12069a7 100644
@@ -423,7 +423,7 @@ static void sun4i_tcon0_mode_set_rgb(struct sun4i_tcon *tcon,
 
 	WARN_ON(!tcon->quirks->has_channel_0);
 
-	tcon->dclk_min_div = 6;
+	tcon->dclk_min_div = 1;
 	tcon->dclk_max_div = 127;
 	sun4i_tcon0_mode_set_common(tcon, mode);
 
Signature · 1 line
2.20.1
author maintainer
#17 lore raw

PATCH 18/25 slcan: Fix memory leak in error path

In reply to Sasha Levin

From: Jouni Hogander <[email protected]>

[ Upstream commit ed50e1600b4483c049ce76e6bd3b665a6a9300ed ]

This patch is fixing memory leak reported by Syzkaller:

BUG: memory leak unreferenced object 0xffff888067f65500 (size 4096):
  comm "syz-executor043", pid 454, jiffies 4294759719 (age 11.930s)
  hex dump (first 32 bytes):
    73 6c 63 61 6e 30 00 00 00 00 00 00 00 00 00 00 slcan0..........
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
  backtrace:
    [<00000000a06eec0d>] __kmalloc+0x18b/0x2c0
    [<0000000083306e66>] kvmalloc_node+0x3a/0xc0
    [<000000006ac27f87>] alloc_netdev_mqs+0x17a/0x1080
    [<0000000061a996c9>] slcan_open+0x3ae/0x9a0
    [<000000001226f0f9>] tty_ldisc_open.isra.1+0x76/0xc0
    [<0000000019289631>] tty_set_ldisc+0x28c/0x5f0
    [<000000004de5a617>] tty_ioctl+0x48d/0x1590
    [<00000000daef496f>] do_vfs_ioctl+0x1c7/0x1510
    [<0000000059068dbc>] ksys_ioctl+0x99/0xb0
    [<000000009a6eb334>] __x64_sys_ioctl+0x78/0xb0
    [<0000000053d0332e>] do_syscall_64+0x16f/0x580
    [<0000000021b83b99>] entry_SYSCALL_64_after_hwframe+0x44/0xa9
    [<000000008ea75434>] 0xffffffffffffffff
Cc Wolfgang Grandegger <[email protected]>
Cc Marc Kleine-Budde <[email protected]>
Cc Lukas Bulwahn <[email protected]>
---
 drivers/net/can/slcan.c | 1 +
 1 file changed, 1 insertion(+)
drivers/net/can/slcan.c +1 -0
index aa97dbc797b6b..5d338b2ac39e1 100644
@@ -613,6 +613,7 @@ static int slcan_open(struct tty_struct *tty)
 	sl->tty = NULL;
 	tty->disc_data = NULL;
 	clear_bit(SLF_INUSE, &sl->flags);
+	free_netdev(sl->dev);
 
 err_exit:
 	rtnl_unlock();
Signature · 1 line
2.20.1
author maintainer
#21 lore raw

PATCH 22/25 i2c: acpi: Force bus speed to 400KHz if a Silead touchscreen is present

In reply to Sasha Levin

From: Hans de Goede <[email protected]>

[ Upstream commit 7574c0db2e68c4d0bae9d415a683bdd8b2a761e9 ]

Many cheap devices use Silead touchscreen controllers. Testing has shown repeatedly that these touchscreen controllers work fine at 400KHz, but for unknown reasons do not work properly at 100KHz. This has been seen on both ARM and x86 devices using totally different i2c controllers.

On some devices the ACPI tables list another device at the same I2C-bus as only being capable of 100KHz, testing has shown that these other devices work fine at 400KHz (as can be expected of any recent I2C hw).

This commit makes i2c_acpi_find_bus_speed() always return 400KHz if a Silead touchscreen controller is present, fixing the touchscreen not working on devices which ACPI tables' wrongly list another device on the same bus as only being capable of 100KHz.

Specifically this fixes the touchscreen on the Jumper EZpad 6 m4 not working.

---
 drivers/i2c/i2c-core-acpi.c | 28 +++++++++++++++++++++++++++-
 1 file changed, 27 insertions(+), 1 deletion(-)
drivers/i2c/i2c-core-acpi.c +27 -1
index 32affd3fa8bd1..559c3b1284d73 100644
@@ -43,6 +43,7 @@ struct i2c_acpi_lookup {
 	int index;
 	u32 speed;
 	u32 min_speed;
+	u32 force_speed;
 };
 
 static int i2c_acpi_fill_info(struct acpi_resource *ares, void *data)
@@ -240,6 +241,19 @@ i2c_acpi_match_device(const struct acpi_device_id *matches,
 	return acpi_match_device(matches, &client->dev);
 }
 
+static const struct acpi_device_id i2c_acpi_force_400khz_device_ids[] = {
+	/*
+	 * These Silead touchscreen controllers only work at 400KHz, for
+	 * some reason they do not work at 100KHz. On some devices the ACPI
+	 * tables list another device at their bus as only being capable
+	 * of 100KHz, testing has shown that these other devices work fine
+	 * at 400KHz (as can be expected of any recent i2c hw) so we force
+	 * the speed of the bus to 400 KHz if a Silead device is present.
+	 */
+	{ "MSSL1680", 0 },
+	{}
+};
+
 static acpi_status i2c_acpi_lookup_speed(acpi_handle handle, u32 level,
 					   void *data, void **return_value)
 {
@@ -258,6 +272,9 @@ static acpi_status i2c_acpi_lookup_speed(acpi_handle handle, u32 level,
 	if (lookup->speed <= lookup->min_speed)
 		lookup->min_speed = lookup->speed;
 
+	if (acpi_match_device_ids(adev, i2c_acpi_force_400khz_device_ids) == 0)
+		lookup->force_speed = 400000;
+
 	return AE_OK;
 }
 
@@ -295,7 +312,16 @@ u32 i2c_acpi_find_bus_speed(struct device *dev)
 		return 0;
 	}
 
-	return lookup.min_speed != UINT_MAX ? lookup.min_speed : 0;
+	if (lookup.force_speed) {
+		if (lookup.force_speed != lookup.min_speed)
+			dev_warn(dev, FW_BUG "DSDT uses known not-working I2C bus speed %d, forcing it to %d\n",
+				 lookup.min_speed, lookup.force_speed);
+		return lookup.force_speed;
+	} else if (lookup.min_speed != UINT_MAX) {
+		return lookup.min_speed;
+	} else {
+		return 0;
+	}
 }
 EXPORT_SYMBOL_GPL(i2c_acpi_find_bus_speed);
 
Signature · 1 line
2.20.1
author maintainer
#22 lore raw

PATCH 23/25 i2c: core: fix use after free in of_i2c_notify

In reply to Sasha Levin

From: Wen Yang <[email protected]>

[ Upstream commit a4c2fec16f5e6a5fee4865e6e0e91e2bc2d10f37 ]

We can't use "adap->dev" after it has been freed.

Fixes 5bf4fa7daea6 ("i2c: break out OF support into separate file")
---
 drivers/i2c/i2c-core-of.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/i2c/i2c-core-of.c +2 -2
index 0f01cdba9d2c6..14d4884996968 100644
@@ -253,14 +253,14 @@ static int of_i2c_notify(struct notifier_block *nb, unsigned long action,
 		}
 
 		client = of_i2c_register_device(adap, rd->dn);
-		put_device(&adap->dev);
-
 		if (IS_ERR(client)) {
 			dev_err(&adap->dev, "failed to create client for '%pOF'\n",
 				 rd->dn);
+			put_device(&adap->dev);
 			of_node_clear_flag(rd->dn, OF_POPULATED);
 			return notifier_from_errno(PTR_ERR(client));
 		}
+		put_device(&adap->dev);
 		break;
 	case OF_RECONFIG_CHANGE_REMOVE:
 		/* already depopulated? */
Signature · 1 line
2.20.1
author maintainer
#1 lore raw

PATCH 2/25 RDMA/hns: Correct the value of HNS_ROCE_HEM_CHUNK_LEN

In reply to Sasha Levin

From: Sirong Wang <[email protected]>

[ Upstream commit 531eb45b3da4267fc2a64233ba256c8ffb02edd2 ]

Size of pointer to buf field of struct hns_roce_hem_chunk should be considered when calculating HNS_ROCE_HEM_CHUNK_LEN, or sg table size will be larger than expected when allocating hem.

Fixes 9a4435375cd1 ("IB/hns: Add driver files for hns RoCE driver")
---
 drivers/infiniband/hw/hns/hns_roce_hem.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/infiniband/hw/hns/hns_roce_hem.h +1 -1
index e8850d59e7804..a94444db3045a 100644
@@ -54,7 +54,7 @@ enum {
 
 #define HNS_ROCE_HEM_CHUNK_LEN	\
 	 ((256 - sizeof(struct list_head) - 2 * sizeof(int)) /	 \
-	 (sizeof(struct scatterlist)))
+	 (sizeof(struct scatterlist) + sizeof(void *)))
 
 #define check_whether_bt_num_3(type, hop_num) \
 	(type < HEM_TYPE_MTT && hop_num == 2)
Signature · 1 line
2.20.1
author maintainer
#3 lore raw

PATCH 4/25 iwlwifi: pcie: don't consider IV len in A-MSDU

In reply to Sasha Levin

From: Mordechay Goodstein <[email protected]>

[ Upstream commit cb1a4badf59275eb7221dcec621e8154917eabd1 ]

From gen2 PN is totally offloaded to hardware (also the space for the IV isn't part of the skb). As you can see in mvm/mac80211.c:3545, the MAC for cipher types CCMP/GCMP doesn't set IEEE80211_KEY_FLAG_PUT_IV_SPACE for gen2 NICs.

This causes all the AMSDU data to be corrupted with cipher enabled.

---
 .../net/wireless/intel/iwlwifi/pcie/tx-gen2.c | 20 +++++++------------
 1 file changed, 7 insertions(+), 13 deletions(-)
drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c +7 -13
index b99f33ff91230..98f4507799be7 100644
@@ -242,27 +242,23 @@ static int iwl_pcie_gen2_build_amsdu(struct iwl_trans *trans,
 	struct ieee80211_hdr *hdr = (void *)skb->data;
 	unsigned int snap_ip_tcp_hdrlen, ip_hdrlen, total_len, hdr_room;
 	unsigned int mss = skb_shinfo(skb)->gso_size;
-	u16 length, iv_len, amsdu_pad;
+	u16 length, amsdu_pad;
 	u8 *start_hdr;
 	struct iwl_tso_hdr_page *hdr_page;
 	struct page **page_ptr;
 	struct tso_t tso;
 
-	/* if the packet is protected, then it must be CCMP or GCMP */
-	iv_len = ieee80211_has_protected(hdr->frame_control) ?
-		IEEE80211_CCMP_HDR_LEN : 0;
-
 	trace_iwlwifi_dev_tx(trans->dev, skb, tfd, sizeof(*tfd),
 			     &dev_cmd->hdr, start_len, 0);
 
 	ip_hdrlen = skb_transport_header(skb) - skb_network_header(skb);
 	snap_ip_tcp_hdrlen = 8 + ip_hdrlen + tcp_hdrlen(skb);
-	total_len = skb->len - snap_ip_tcp_hdrlen - hdr_len - iv_len;
+	total_len = skb->len - snap_ip_tcp_hdrlen - hdr_len;
 	amsdu_pad = 0;
 
 	/* total amount of header we may need for this A-MSDU */
 	hdr_room = DIV_ROUND_UP(total_len, mss) *
-		(3 + snap_ip_tcp_hdrlen + sizeof(struct ethhdr)) + iv_len;
+		(3 + snap_ip_tcp_hdrlen + sizeof(struct ethhdr));
 
 	/* Our device supports 9 segments at most, it will fit in 1 page */
 	hdr_page = get_page_hdr(trans, hdr_room);
@@ -273,14 +269,12 @@ static int iwl_pcie_gen2_build_amsdu(struct iwl_trans *trans,
 	start_hdr = hdr_page->pos;
 	page_ptr = (void *)((u8 *)skb->cb + trans_pcie->page_offs);
 	*page_ptr = hdr_page->page;
-	memcpy(hdr_page->pos, skb->data + hdr_len, iv_len);
-	hdr_page->pos += iv_len;
 
 	/*
-	 * Pull the ieee80211 header + IV to be able to use TSO core,
+	 * Pull the ieee80211 header to be able to use TSO core,
 	 * we will restore it for the tx_status flow.
 	 */
-	skb_pull(skb, hdr_len + iv_len);
+	skb_pull(skb, hdr_len);
 
 	/*
 	 * Remove the length of all the headers that we don't actually
@@ -355,8 +349,8 @@ static int iwl_pcie_gen2_build_amsdu(struct iwl_trans *trans,
 		}
 	}
 
-	/* re -add the WiFi header and IV */
-	skb_push(skb, hdr_len + iv_len);
+	/* re -add the WiFi header */
+	skb_push(skb, hdr_len);
 
 	return 0;
 
Signature · 1 line
2.20.1
author maintainer
#7 lore raw

PATCH 8/25 xfrm: release device reference for invalid state

In reply to Sasha Levin

From: Xiaodong Xu <[email protected]>

[ Upstream commit 4944a4b1077f74d89073624bd286219d2fcbfce3 ]

An ESP packet could be decrypted in async mode if the input handler for this packet returns -EINPROGRESS in xfrm_input(). At this moment the device reference in skb is held. Later xfrm_input() will be invoked again to resume the processing.
If the transform state is still valid it would continue to release the device reference and there won't be a problem; however if the transform state is not valid when async resumption happens, the packet will be dropped while the device reference is still being held.
When the device is deleted for some reason and the reference to this device is not properly released, the kernel will keep logging like:

unregister_netdevice: waiting for ppp2 to become free. Usage count = 1

The issue is observed when running IPsec traffic over a PPPoE device based on a bridge interface. By terminating the PPPoE connection on the server end for multiple times, the PPPoE device on the client side will eventually get stuck on the above warning message.

This patch will check the async mode first and continue to release device reference in async resumption, before it is dropped due to invalid state.

v2: Do not assign address family from outer_mode in the transform if the state is invalid

v3: Release device reference in the error path instead of jumping to resume

Fixes 4ce3dbe397d7b ("xfrm: Fix xfrm_input() to verify state is valid when (encap_type < 0)")
---
 net/xfrm/xfrm_input.c | 3 +++
 1 file changed, 3 insertions(+)
net/xfrm/xfrm_input.c +3 -0
index 790b514f86b62..b0b1294daef17 100644
@@ -246,6 +246,9 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 			else
 				XFRM_INC_STATS(net,
 					       LINUX_MIB_XFRMINSTATEINVALID);
+
+			if (encap_type == -1)
+				dev_put(skb->dev);
 			goto drop;
 		}
 
Signature · 1 line
2.20.1
author maintainer
#8 lore raw

PATCH 9/25 Input: cyttsp4_core - fix use after free bug

In reply to Sasha Levin

From: Pan Bian <[email protected]>

[ Upstream commit 79aae6acbef16f720a7949f8fc6ac69816c79d62 ]

The device md->input is used after it is released. Setting the device data to NULL is unnecessary as the device is never used again. Instead, md->input should be assigned NULL to avoid accessing the freed memory accidently. Besides, checking md->si against NULL is superfluous as it points to a variable address, which cannot be NULL.

---
 drivers/input/touchscreen/cyttsp4_core.c | 7 -------
 1 file changed, 7 deletions(-)
drivers/input/touchscreen/cyttsp4_core.c +0 -7
index 727c3232517cd..c84ee739a8d50 100644
@@ -2000,11 +2000,6 @@ static int cyttsp4_mt_probe(struct cyttsp4 *cd)
 
 	/* get sysinfo */
 	md->si = &cd->sysinfo;
-	if (!md->si) {
-		dev_err(dev, "%s: Fail get sysinfo pointer from core p=%p\n",
-			__func__, md->si);
-		goto error_get_sysinfo;
-	}
 
 	rc = cyttsp4_setup_input_device(cd);
 	if (rc)
@@ -2014,8 +2009,6 @@ static int cyttsp4_mt_probe(struct cyttsp4 *cd)
 
 error_init_input:
 	input_free_device(md->input);
-error_get_sysinfo:
-	input_set_drvdata(md->input, NULL);
 error_alloc_failed:
 	dev_err(dev, "%s failed.\n", __func__);
 	return rc;
Signature · 1 line
2.20.1
author maintainer
#9 lore raw

PATCH 10/25 sched/core: Avoid spurious lock dependencies

In reply to Sasha Levin

From: Peter Zijlstra <[email protected]>

[ Upstream commit ff51ff84d82aea5a889b85f2b9fb3aa2b8691668 ]

While seemingly harmless, __sched_fork() does hrtimer_init(), which, when DEBUG_OBJETS, can end up doing allocations.

This then results in the following lock order:

rq->lock

    zone->lock.rlock
      batched_entropy_u64.lock

Which in turn causes deadlocks when we do wakeups while holding that batched_entropy lock -- as the random code does.

Solve this by moving __sched_fork() out from under rq->lock. This is safe because nothing there relies on rq->lock, as also evident from the other __sched_fork() callsite.

---
 kernel/sched/core.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
kernel/sched/core.c +2 -1
index 78ecdfae25b69..2befd2c4ce9e6 100644
@@ -5413,10 +5413,11 @@ void init_idle(struct task_struct *idle, int cpu)
 	struct rq *rq = cpu_rq(cpu);
 	unsigned long flags;
 
+	__sched_fork(0, idle);
+
 	raw_spin_lock_irqsave(&idle->pi_lock, flags);
 	raw_spin_lock(&rq->lock);
 
-	__sched_fork(0, idle);
 	idle->state = TASK_RUNNING;
 	idle->se.exec_start = sched_clock();
 	idle->flags |= PF_IDLE;
Signature · 1 line
2.20.1
author maintainer
#14 lore raw

PATCH 15/25 net: usb: qmi_wwan: add support for Foxconn T77W968 LTE modules

In reply to Sasha Levin

From: Aleksander Morgado <[email protected]>

[ Upstream commit 802753cb0b141cf5170ab97fe7e79f5ca10d06b0 ]

These are the Foxconn-branded variants of the Dell DW5821e modules, same USB layout as those.

The QMI interface is exposed in USB configuration #1:

P:  Vendor=0489 ProdID=e0b4 Rev=03.18
S:  Manufacturer=FII
S:  Product=T77W968 LTE
S:  SerialNumber=0123456789ABCDEF
C:  #Ifs= 6 Cfg#= 1 Atr=a0 MxPwr=500mA
I:  If#=0x0 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=ff Driver=qmi_wwan
I:  If#=0x1 Alt= 0 #EPs= 1 Cls=03(HID  ) Sub=00 Prot=00 Driver=usbhid
I:  If#=0x2 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=00 Prot=00 Driver=option
I:  If#=0x3 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=00 Prot=00 Driver=option
I:  If#=0x4 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=00 Prot=00 Driver=option
I:  If#=0x5 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
---
 drivers/net/usb/qmi_wwan.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/net/usb/qmi_wwan.c +2 -0
index 9f037c50054df..b55fd76348f9f 100644
@@ -1306,6 +1306,8 @@ static const struct usb_device_id products[] = {
 	{QMI_QUIRK_SET_DTR(0x2c7c, 0x0191, 4)},	/* Quectel EG91 */
 	{QMI_FIXED_INTF(0x2c7c, 0x0296, 4)},	/* Quectel BG96 */
 	{QMI_QUIRK_SET_DTR(0x2cb7, 0x0104, 4)},	/* Fibocom NL678 series */
+	{QMI_FIXED_INTF(0x0489, 0xe0b4, 0)},	/* Foxconn T77W968 LTE */
+	{QMI_FIXED_INTF(0x0489, 0xe0b5, 0)},	/* Foxconn T77W968 LTE with eSIM support*/
 
 	/* 4. Gobi 1000 devices */
 	{QMI_GOBI1K_DEVICE(0x05c6, 0x9212)},	/* Acer Gobi Modem Device */
Signature · 1 line
2.20.1
author maintainer
#16 lore raw

PATCH 17/25 net: cdc_ncm: Signedness bug in cdc_ncm_set_dgram_size()

In reply to Sasha Levin

From: Dan Carpenter <[email protected]>

[ Upstream commit a56dcc6b455830776899ce3686735f1172e12243 ]

This code is supposed to test for negative error codes and partial reads, but because sizeof() is size_t (unsigned) type then negative error codes are type promoted to high positive values and the condition doesn't work as expected.

Fixes 332f989a3b00 ("CDC-NCM: handle incomplete transfer of MTU")
---
 drivers/net/usb/cdc_ncm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/usb/cdc_ncm.c +1 -1
index a57d82ef0f810..1f57a6a2b8a25 100644
@@ -579,7 +579,7 @@ static void cdc_ncm_set_dgram_size(struct usbnet *dev, int new_size)
 	err = usbnet_read_cmd(dev, USB_CDC_GET_MAX_DATAGRAM_SIZE,
 			      USB_TYPE_CLASS | USB_DIR_IN | USB_RECIP_INTERFACE,
 			      0, iface_no, &max_datagram_size, sizeof(max_datagram_size));
-	if (err < sizeof(max_datagram_size)) {
+	if (err != sizeof(max_datagram_size)) {
 		dev_dbg(&dev->intf->dev, "GET_MAX_DATAGRAM_SIZE failed\n");
 		goto out;
 	}
Signature · 1 line
2.20.1
author maintainer
#18 lore raw

PATCH 19/25 rsxx: add missed destroy_workqueue calls in remove

In reply to Sasha Levin

From: Chuhong Yuan <[email protected]>

[ Upstream commit dcb77e4b274b8f13ac6482dfb09160cd2fae9a40 ]

The driver misses calling destroy_workqueue in remove like what is done when probe fails.
Add the missed calls to fix it.

---
 drivers/block/rsxx/core.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/block/rsxx/core.c +2 -0
index f2c631ce793cc..14056dc450642 100644
@@ -1014,8 +1014,10 @@ static void rsxx_pci_remove(struct pci_dev *dev)
 
 	cancel_work_sync(&card->event_work);
 
+	destroy_workqueue(card->event_wq);
 	rsxx_destroy_dev(card);
 	rsxx_dma_destroy(card);
+	destroy_workqueue(card->creg_ctrl.creg_wq);
 
 	spin_lock_irqsave(&card->irq_lock, flags);
 	rsxx_disable_ier_and_isr(card, CR_INTR_ALL);
Signature · 1 line
2.20.1
author maintainer
#19 lore raw

PATCH 20/25 ax88172a: fix information leak on short answers

In reply to Sasha Levin

From: Oliver Neukum <[email protected]>

[ Upstream commit a9a51bd727d141a67b589f375fe69d0e54c4fe22 ]

If a malicious device gives a short MAC it can elicit up to 5 bytes of leaked memory out of the driver. We need to check for ETH_ALEN instead.

---
 drivers/net/usb/ax88172a.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/usb/ax88172a.c +1 -1
index 501576f538546..914cac55a7ae7 100644
@@ -208,7 +208,7 @@ static int ax88172a_bind(struct usbnet *dev, struct usb_interface *intf)
 
 	/* Get the MAC address */
 	ret = asix_read_cmd(dev, AX_CMD_READ_NODE_ID, 0, 0, ETH_ALEN, buf, 0);
-	if (ret < 0) {
+	if (ret < ETH_ALEN) {
 		netdev_err(dev->net, "Failed to read MAC address: %d\n", ret);
 		goto free;
 	}
Signature · 1 line
2.20.1
author maintainer
#26 lore raw

PATCH 25/25 net: gemini: add missed free_netdev

In reply to Sasha Levin

From: Chuhong Yuan <[email protected]>

[ Upstream commit 18d647ae74116bfee38953978501cea2960a0c25 ]

This driver forgets to free allocated netdev in remove like what is done in probe failure.
Add the free to fix it.

---
 drivers/net/ethernet/cortina/gemini.c | 1 +
 1 file changed, 1 insertion(+)
drivers/net/ethernet/cortina/gemini.c +1 -0
index dfd1ad0b1cb94..4af78de0e077a 100644
@@ -2530,6 +2530,7 @@ static int gemini_ethernet_port_remove(struct platform_device *pdev)
 	struct gemini_ethernet_port *port = platform_get_drvdata(pdev);
 
 	gemini_port_remove(port);
+	free_netdev(port->netdev);
 	return 0;
 }
 
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#11 lore raw

PATCH 12/25 ALSA: pcm: Fix stream lock usage in snd_pcm_period_elapsed()

In reply to Sasha Levin

From: paulhsia <[email protected]>

[ Upstream commit f5cdc9d4003a2f66ea57b3edd3e04acc2b1a4439 ]

If the nullity check for substream->runtime is outside of the lock region, it is possible to have a null runtime in the critical section if snd_pcm_detach_substream is called right before the lock.

---
 sound/core/pcm_lib.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
sound/core/pcm_lib.c +6 -2
index 4e6110d778bd2..ad52126d3d22e 100644
@@ -1797,11 +1797,14 @@ void snd_pcm_period_elapsed(struct snd_pcm_substream *substream)
 	struct snd_pcm_runtime *runtime;
 	unsigned long flags;
 
-	if (PCM_RUNTIME_CHECK(substream))
+	if (snd_BUG_ON(!substream))
 		return;
-	runtime = substream->runtime;
 
 	snd_pcm_stream_lock_irqsave(substream, flags);
+	if (PCM_RUNTIME_CHECK(substream))
+		goto _unlock;
+	runtime = substream->runtime;
+
 	if (!snd_pcm_running(substream) ||
 	    snd_pcm_update_hw_ptr0(substream, 1) < 0)
 		goto _end;
@@ -1812,6 +1815,7 @@ void snd_pcm_period_elapsed(struct snd_pcm_substream *substream)
 #endif
  _end:
 	kill_fasync(&runtime->fasync, SIGIO, POLL_IN);
+ _unlock:
 	snd_pcm_stream_unlock_irqrestore(substream, flags);
 }
 EXPORT_SYMBOL(snd_pcm_period_elapsed);
Signature · 6 lines
2.20.1

_______________________________________________
Alsa-devel mailing list
[email protected]
https://mailman.alsa-project.org/mailman/listinfo/alsa-devel
author maintainer
#20 lore raw

PATCH 21/25 net: ep93xx_eth: fix mismatch of request_mem_region in remove

In reply to Sasha Levin

From: Chuhong Yuan <[email protected]>

[ Upstream commit 3df70afe8d33f4977d0e0891bdcfb639320b5257 ]

The driver calls release_resource in remove to match request_mem_region in probe, which is incorrect.
Fix it by using the right one, release_mem_region.

---
 drivers/net/ethernet/cirrus/ep93xx_eth.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/net/ethernet/cirrus/ep93xx_eth.c +3 -2
index 13dfdfca49fc7..edc1d19c9c02e 100644
@@ -767,6 +767,7 @@ static int ep93xx_eth_remove(struct platform_device *pdev)
 {
 	struct net_device *dev;
 	struct ep93xx_priv *ep;
+	struct resource *mem;
 
 	dev = platform_get_drvdata(pdev);
 	if (dev == NULL)
@@ -782,8 +783,8 @@ static int ep93xx_eth_remove(struct platform_device *pdev)
 		iounmap(ep->base_addr);
 
 	if (ep->res != NULL) {
-		release_resource(ep->res);
-		kfree(ep->res);
+		mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+		release_mem_region(mem->start, resource_size(mem));
 	}
 
 	free_netdev(dev);
Signature · 1 line
2.20.1
author maintainer
#23 lore raw

PATCH 24/25 Input: synaptics - enable RMI mode for X1 Extreme 2nd Generation

In reply to Sasha Levin

From: Lyude Paul <[email protected]>

[ Upstream commit 768ea88bcb235ac3a92754bf82afcd3f12200bcc ]

Just got one of these for debugging some unrelated issues, and noticed that Lenovo seems to have gone back to using RMI4 over smbus with Synaptics touchpads on some of their new systems, particularly this one. So, let's enable RMI mode for the X1 Extreme 2nd Generation.

---
 drivers/input/mouse/synaptics.c | 1 +
 1 file changed, 1 insertion(+)
drivers/input/mouse/synaptics.c +1 -0
index 06cebde2422ea..afdb9947d8af9 100644
@@ -180,6 +180,7 @@ static const char * const smbus_pnp_ids[] = {
 	"LEN0096", /* X280 */
 	"LEN0097", /* X280 -> ALPS trackpoint */
 	"LEN009b", /* T580 */
+	"LEN0402", /* X1 Extreme 2nd Generation */
 	"LEN200f", /* T450s */
 	"LEN2054", /* E480 */
 	"LEN2055", /* E580 */
Signature · 1 line
2.20.1
author maintainer
#15 lore raw

PATCH 16/25 slip: Fix memory leak in slip_open error path

In reply to Sasha Levin

From: Jouni Hogander <[email protected]>

[ Upstream commit 3b5a39979dafea9d0cd69c7ae06088f7a84cdafa ]

Driver/net/can/slcan.c is derived from slip.c. Memory leak was detected by Syzkaller in slcan. Same issue exists in slip.c and this patch is addressing the leak in slip.c.

Here is the slcan memory leak trace reported by Syzkaller:

BUG: memory leak unreferenced object 0xffff888067f65500 (size 4096):
  comm "syz-executor043", pid 454, jiffies 4294759719 (age 11.930s)
  hex dump (first 32 bytes):
    73 6c 63 61 6e 30 00 00 00 00 00 00 00 00 00 00 slcan0..........
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
  backtrace:
    [<00000000a06eec0d>] __kmalloc+0x18b/0x2c0
    [<0000000083306e66>] kvmalloc_node+0x3a/0xc0
    [<000000006ac27f87>] alloc_netdev_mqs+0x17a/0x1080
    [<0000000061a996c9>] slcan_open+0x3ae/0x9a0
    [<000000001226f0f9>] tty_ldisc_open.isra.1+0x76/0xc0
    [<0000000019289631>] tty_set_ldisc+0x28c/0x5f0
    [<000000004de5a617>] tty_ioctl+0x48d/0x1590
    [<00000000daef496f>] do_vfs_ioctl+0x1c7/0x1510
    [<0000000059068dbc>] ksys_ioctl+0x99/0xb0
    [<000000009a6eb334>] __x64_sys_ioctl+0x78/0xb0
    [<0000000053d0332e>] do_syscall_64+0x16f/0x580
    [<0000000021b83b99>] entry_SYSCALL_64_after_hwframe+0x44/0xa9
    [<000000008ea75434>] 0xfffffffffffffff
Cc "David S. Miller" <[email protected]>
Cc Oliver Hartkopp <[email protected]>
Cc Lukas Bulwahn <[email protected]>
---
 drivers/net/slip/slip.c | 1 +
 1 file changed, 1 insertion(+)
drivers/net/slip/slip.c +1 -0
index b008266e91eab..a5874059da9d8 100644
@@ -855,6 +855,7 @@ static int slip_open(struct tty_struct *tty)
 	sl->tty = NULL;
 	tty->disc_data = NULL;
 	clear_bit(SLF_INUSE, &sl->flags);
+	free_netdev(sl->dev);
 
 err_exit:
 	rtnl_unlock();
Signature · 1 line
2.20.1
author maintainer
#13 lore raw

PATCH 14/25 selftests: kvm: fix build with glibc >= 2.30

In reply to Sasha Levin

From: Vitaly Kuznetsov <[email protected]>

[ Upstream commit e37f9f139f62deddff90c7298ae3a85026a71067 ]

Glibc-2.30 gained gettid() wrapper, selftests fail to compile:

lib/assert.c:58:14: error: static declaration of ‘gettid’ follows non-static declaration
58 | static pid_t gettid(void)

      |              ^~~~~~
In file included from /usr/include/unistd.h:1170,
                 from include/test_util.h:18,
                 from lib/assert.c:10:
/usr/include/bits/unistd_ext.h:34:16: note: previous declaration of ‘gettid’ was here
   34 | extern __pid_t gettid (void) __THROW;
      |                ^~~~~~
---
 tools/testing/selftests/kvm/lib/assert.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
tools/testing/selftests/kvm/lib/assert.c +2 -2
index cd01144d27c8d..d306677065699 100644
@@ -56,7 +56,7 @@ static void test_dump_stack(void)
 #pragma GCC diagnostic pop
 }
 
-static pid_t gettid(void)
+static pid_t _gettid(void)
 {
 	return syscall(SYS_gettid);
 }
@@ -73,7 +73,7 @@ test_assert(bool exp, const char *exp_str,
 		fprintf(stderr, "==== Test Assertion Failure ====\n"
 			"  %s:%u: %s\n"
 			"  pid=%d tid=%d - %s\n",
-			file, line, exp_str, getpid(), gettid(),
+			file, line, exp_str, getpid(), _gettid(),
 			strerror(errno));
 		test_dump_stack();
 		if (fmt) {
Signature · 1 line
2.20.1
author maintainer
#10 lore raw

PATCH 11/25 perf/core: Consistently fail fork on allocation failures

In reply to Sasha Levin

From: Alexander Shishkin <[email protected]>

[ Upstream commit 697d877849d4b34ab58d7078d6930bad0ef6fc66 ]

Commit:

313ccb9615948 ("perf: Allocate context task_ctx_data for child event")

makes the inherit path skip over the current event in case of task_ctx_data allocation failure. This, however, is inconsistent with allocation failures in perf_event_alloc(), which would abort the fork.

Correct this by returning an error code on task_ctx_data allocation failure and failing the fork in that case.

Cc Arnaldo Carvalho de Melo <[email protected]>
Cc David Ahern <[email protected]>
Cc Jiri Olsa <[email protected]>
Cc Jiri Olsa <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Mark Rutland <[email protected]>
Cc Namhyung Kim <[email protected]>
Cc Stephane Eranian <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc Vince Weaver <[email protected]>
---
 kernel/events/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/events/core.c +1 -1
index 625ba462e5bbd..460d5fd3ec4e4 100644
@@ -11377,7 +11377,7 @@ inherit_event(struct perf_event *parent_event,
 						   GFP_KERNEL);
 		if (!child_ctx->task_ctx_data) {
 			free_event(child_event);
-			return NULL;
+			return ERR_PTR(-ENOMEM);
 		}
 	}
 
Signature · 1 line
2.20.1
author maintainer
#6 lore raw

PATCH 7/25 NFC: nxp-nci: Fix NULL pointer dereference after I2C communication error

In reply to Sasha Levin

From: Stephan Gerhold <[email protected]>

[ Upstream commit a71a29f50de1ef97ab55c151a1598eb12dde379d ]

I2C communication errors (-EREMOTEIO) during the IRQ handler of nxp-nci result in a NULL pointer dereference at the moment:

    BUG: kernel NULL pointer dereference, address: 0000000000000000
    Oops: 0002 [#1] PREEMPT SMP NOPTI
    CPU: 1 PID: 355 Comm: irq/137-nxp-nci Not tainted 5.4.0-rc6 #1
    RIP: 0010:skb_queue_tail+0x25/0x50
    Call Trace:
     nci_recv_frame+0x36/0x90 [nci]
     nxp_nci_i2c_irq_thread_fn+0xd1/0x285 [nxp_nci_i2c]
     ? preempt_count_add+0x68/0xa0
     ? irq_forced_thread_fn+0x80/0x80
     irq_thread_fn+0x20/0x60
     irq_thread+0xee/0x180
     ? wake_threads_waitq+0x30/0x30
     kthread+0xfb/0x130
     ? irq_thread_check_affinity+0xd0/0xd0
     ? kthread_park+0x90/0x90
     ret_from_fork+0x1f/0x40

Afterward the kernel must be rebooted to work properly again.

This happens because it attempts to call nci_recv_frame() with skb == NULL. However, unlike nxp_nci_fw_recv_frame(), nci_recv_frame() does not have any NULL checks for skb, causing the NULL pointer dereference.

Change the code to call only nxp_nci_fw_recv_frame() in case of an error. Make sure to log it so it is obvious that a communication error occurred. The error above then becomes:

    nxp-nci_i2c i2c-NXP1001:00: NFC: Read failed with error -121
    nci: __nci_request: wait_for_completion_interruptible_timeout failed 0
    nxp-nci_i2c i2c-NXP1001:00: NFC: Read failed with error -121
Fixes 6be88670fc59 ("NFC: nxp-nci_i2c: Add I2C support to NXP NCI driver")
---
 drivers/nfc/nxp-nci/i2c.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/nfc/nxp-nci/i2c.c +4 -2
index ba695e392c3b7..0df745cad601a 100644
@@ -236,8 +236,10 @@ static irqreturn_t nxp_nci_i2c_irq_thread_fn(int irq, void *phy_id)
 
 	if (r == -EREMOTEIO) {
 		phy->hard_fault = r;
-		skb = NULL;
-	} else if (r < 0) {
+		if (info->mode == NXP_NCI_MODE_FW)
+			nxp_nci_fw_recv_frame(phy->ndev, NULL);
+	}
+	if (r < 0) {
 		nfc_err(&client->dev, "Read failed with error %d\n", r);
 		goto exit_irq_handled;
 	}
Signature · 1 line
2.20.1
author maintainer
#4 lore raw

PATCH 5/25 exportfs_decode_fh(): negative pinned may become positive without the parent locked

In reply to Sasha Levin

From: Al Viro <[email protected]>

[ Upstream commit a2ece088882666e1dc7113744ac912eb161e3f87 ]

---
 fs/exportfs/expfs.c | 31 +++++++++++++++++++------------
 1 file changed, 19 insertions(+), 12 deletions(-)
fs/exportfs/expfs.c +19 -12
index 63707abcbeb3e..808cae6d5f50f 100644
@@ -517,26 +517,33 @@ struct dentry *exportfs_decode_fh(struct vfsmount *mnt, struct fid *fid,
 		 * inode is actually connected to the parent.
 		 */
 		err = exportfs_get_name(mnt, target_dir, nbuf, result);
-		if (!err) {
-			inode_lock(target_dir->d_inode);
-			nresult = lookup_one_len(nbuf, target_dir,
-						 strlen(nbuf));
-			inode_unlock(target_dir->d_inode);
-			if (!IS_ERR(nresult)) {
-				if (nresult->d_inode) {
-					dput(result);
-					result = nresult;
-				} else
-					dput(nresult);
-			}
+		if (err) {
+			dput(target_dir);
+			goto err_result;
 		}
 
+		inode_lock(target_dir->d_inode);
+		nresult = lookup_one_len(nbuf, target_dir, strlen(nbuf));
+		if (!IS_ERR(nresult)) {
+			if (unlikely(nresult->d_inode != result->d_inode)) {
+				dput(nresult);
+				nresult = ERR_PTR(-ESTALE);
+			}
+		}
+		inode_unlock(target_dir->d_inode);
 		/*
 		 * At this point we are done with the parent, but it's pinned
 		 * by the child dentry anyway.
 		 */
 		dput(target_dir);
 
+		if (IS_ERR(nresult)) {
+			err = PTR_ERR(nresult);
+			goto err_result;
+		}
+		dput(result);
+		result = nresult;
+
 		/*
 		 * And finally make sure the dentry is actually acceptable
 		 * to NFSD.
Signature · 1 line
2.20.1
author maintainer
#2 lore raw

PATCH 3/25 xfrm: Fix memleak on xfrm state destroy

In reply to Sasha Levin

From: Steffen Klassert <[email protected]>

[ Upstream commit 86c6739eda7d2a03f2db30cbee67a5fb81afa8ba ]

We leak the page that we use to create skb page fragments when destroying the xfrm_state. Fix this by dropping a page reference if a page was assigned to the xfrm_state.

Fixes cac2661c53f3 ("esp4: Avoid skb_cow_data whenever possible")
---
 net/xfrm/xfrm_state.c | 2 ++
 1 file changed, 2 insertions(+)
net/xfrm/xfrm_state.c +2 -0
index 11e09eb138d60..47a8ff972a2bf 100644
@@ -456,6 +456,8 @@ static void ___xfrm_state_destroy(struct xfrm_state *x)
 		x->type->destructor(x);
 		xfrm_put_type(x->type);
 	}
+	if (x->xfrag.page)
+		put_page(x->xfrag.page);
 	xfrm_dev_state_free(x);
 	security_xfrm_state_free(x);
 	xfrm_state_free(x);
Signature · 1 line
2.20.1
#24 lore raw

Re: [PATCH AUTOSEL 4.19 24/25] Input: synaptics - enable RMI mode for X1 Extreme 2nd Generation

In reply to Sasha Levin

Hi Sasha,

Quoted 9 lines · Sasha Levin

From: Lyude Paul <[email protected]>

[ Upstream commit 768ea88bcb235ac3a92754bf82afcd3f12200bcc ]

Just got one of these for debugging some unrelated issues, and noticed that Lenovo seems to have gone back to using RMI4 over smbus with Synaptics touchpads on some of their new systems, particularly this one. So, let's enable RMI mode for the X1 Extreme 2nd Generation.

Quoted 4 lines

This will be reverted, do not pick up for stable.

Quoted 19 lines
---
 drivers/input/mouse/synaptics.c | 1 +
 1 file changed, 1 insertion(+)
drivers/input/mouse/synaptics.c +1 -0
index 06cebde2422ea..afdb9947d8af9 100644
@@ -180,6 +180,7 @@ static const char * const smbus_pnp_ids[] = {
 	"LEN0096", /* X280 */
 	"LEN0097", /* X280 -> ALPS trackpoint */
 	"LEN009b", /* T580 */
+	"LEN0402", /* X1 Extreme 2nd Generation */
 	"LEN200f", /* T450s */
 	"LEN2054", /* E480 */
 	"LEN2055", /* E580 */
Signature · 1 line
2.20.1

Thanks.

Signature · 1 line
Dmitry
author maintainer
#25 lore raw

Re: [PATCH AUTOSEL 4.19 24/25] Input: synaptics - enable RMI mode for X1 Extreme 2nd Generation

Quoted 12 lines · Dmitry Torokhov

Hi Sasha,

Quoted 9 lines · Sasha Levin

From: Lyude Paul <[email protected]>

[ Upstream commit 768ea88bcb235ac3a92754bf82afcd3f12200bcc ]

Just got one of these for debugging some unrelated issues, and noticed that Lenovo seems to have gone back to using RMI4 over smbus with Synaptics touchpads on some of their new systems, particularly this one. So, let's enable RMI mode for the X1 Extreme 2nd Generation.

Quoted 6 lines
Quoted 4 lines

This will be reverted, do not pick up for stable.

I've dropped it, thanks!

Signature · 2 lines
Thanks,
Sasha