Skip to contentSkip to conversation

5.1.20-stable review

PATCH In review Greg Kroah-Hartman 10 participants 384 messages Jul 24, 2019 – Jul 26, 2019 Page 2 of 3

Subsystems (8)
Stable branch USB MIPS Crypto API +4 more subsystems
lore
Share & tools
lore
author maintainer
#161 lore raw

PATCH 160/371 gpio: Fix return value mismatch of function gpiod_get_from_of_node()

In reply to Greg Kroah-Hartman

[ Upstream commit 025bf37725f1929542361eef2245df30badf242e ]

In case the requested gpio property is not found in the device tree, some callers of gpiod_get_from_of_node() expect a return value of NULL, others expect -ENOENT.
In particular devm_fwnode_get_index_gpiod_from_child() expects -ENOENT. Currently it gets a NULL, which breaks the loop that tries all gpio_suffixes. The result is that a gpio property is not found, even though it is there.

This patch changes gpiod_get_from_of_node() to return -ENOENT instead of NULL when the requested gpio property is not found in the device tree. Additionally it modifies all calling functions to properly evaluate the return value.

Another approach would be to leave the return value of gpiod_get_from_of_node() as is and fix the bug in devm_fwnode_get_index_gpiod_from_child(). Other callers would still need to be reworked. The effort would be the same as with the chosen solution.

---
 drivers/gpio/gpiolib.c                 | 6 +-----
 drivers/regulator/da9211-regulator.c   | 2 ++
 drivers/regulator/s2mps11.c            | 4 +++-
 drivers/regulator/s5m8767.c            | 4 +++-
 drivers/regulator/tps65090-regulator.c | 7 ++++---
 5 files changed, 13 insertions(+), 10 deletions(-)
drivers/gpio/gpiolib.c +1 -5
index bca3e7740ef6..fd1344056e1d 100644
@@ -4232,8 +4232,7 @@ EXPORT_SYMBOL_GPL(gpiod_get_index);
  *
  * Returns:
  * On successful request the GPIO pin is configured in accordance with
- * provided @dflags. If the node does not have the requested GPIO
- * property, NULL is returned.
+ * provided @dflags.
  *
  * In case of error an ERR_PTR() is returned.
  */
@@ -4255,9 +4254,6 @@ struct gpio_desc *gpiod_get_from_of_node(struct device_node *node,
 					index, &flags);
 
 	if (!desc || IS_ERR(desc)) {
-		/* If it is not there, just return NULL */
-		if (PTR_ERR(desc) == -ENOENT)
-			return NULL;
 		return desc;
 	}
 
drivers/regulator/da9211-regulator.c +2 -0
index 4d7fe4819c1c..4e95e3d0fcd5 100644
@@ -299,6 +299,8 @@ static struct da9211_pdata *da9211_parse_regulators_dt(
 				  0,
 				  GPIOD_OUT_HIGH | GPIOD_FLAGS_BIT_NONEXCLUSIVE,
 				  "da9211-enable");
+		if (IS_ERR(pdata->gpiod_ren[n]))
+			pdata->gpiod_ren[n] = NULL;
 		n++;
 	}
 
drivers/regulator/s2mps11.c +3 -1
index 134c62db36c5..b518a81f75a3 100644
@@ -821,7 +821,9 @@ static void s2mps14_pmic_dt_parse_ext_control_gpio(struct platform_device *pdev,
 				0,
 				GPIOD_OUT_HIGH | GPIOD_FLAGS_BIT_NONEXCLUSIVE,
 				"s2mps11-regulator");
-		if (IS_ERR(gpio[reg])) {
+		if (PTR_ERR(gpio[reg]) == -ENOENT)
+			gpio[reg] = NULL;
+		else if (IS_ERR(gpio[reg])) {
 			dev_err(&pdev->dev, "Failed to get control GPIO for %d/%s\n",
 				reg, rdata[reg].name);
 			continue;
drivers/regulator/s5m8767.c +3 -1
index bb9d1a083299..6ca27e9d5ef7 100644
@@ -574,7 +574,9 @@ static int s5m8767_pmic_dt_parse_pdata(struct platform_device *pdev,
 			0,
 			GPIOD_OUT_HIGH | GPIOD_FLAGS_BIT_NONEXCLUSIVE,
 			"s5m8767");
-		if (IS_ERR(rdata->ext_control_gpiod))
+		if (PTR_ERR(rdata->ext_control_gpiod) == -ENOENT)
+			rdata->ext_control_gpiod = NULL;
+		else if (IS_ERR(rdata->ext_control_gpiod))
 			return PTR_ERR(rdata->ext_control_gpiod);
 
 		rdata->id = i;
drivers/regulator/tps65090-regulator.c +4 -3
index 0614551796a1..f6466db57900 100644
@@ -381,11 +381,12 @@ static struct tps65090_platform_data *tps65090_parse_dt_reg_data(
 								    "dcdc-ext-control-gpios", 0,
 								    gflags,
 								    "tps65090");
-			if (IS_ERR(rpdata->gpiod))
-				return ERR_CAST(rpdata->gpiod);
-			if (!rpdata->gpiod)
+			if (PTR_ERR(rpdata->gpiod) == -ENOENT) {
 				dev_err(&pdev->dev,
 					"could not find DCDC external control GPIO\n");
+				rpdata->gpiod = NULL;
+			} else if (IS_ERR(rpdata->gpiod))
+				return ERR_CAST(rpdata->gpiod);
 		}
 
 		if (of_property_read_u32(tps65090_matches[idx].of_node,
Signature · 1 line
2.20.1
author maintainer
#162 lore raw

PATCH 161/371 net/mlx5: Get vport ACL namespace by vport index

In reply to Greg Kroah-Hartman

[ Upstream commit f53297d67800feb5fafd94abd926c889aefee690 ]

The ingress and egress ACL root namespaces are created per vport and stored into arrays. However, the vport number is not the same as the index. Passing the array index, instead of vport number, to get the correct ingress and egress acl namespace.

Fixes 9b93ab981e3b ("net/mlx5: Separate ingress/egress namespaces for each vport")
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +2 -2
index 8a67fd197b79..16ed6ebd31ee 100644
@@ -950,7 +950,7 @@ static int esw_vport_enable_egress_acl(struct mlx5_eswitch *esw,
 		  vport->vport, MLX5_CAP_ESW_EGRESS_ACL(dev, log_max_ft_size));
 
 	root_ns = mlx5_get_flow_vport_acl_namespace(dev, MLX5_FLOW_NAMESPACE_ESW_EGRESS,
-						    vport->vport);
+			mlx5_eswitch_vport_num_to_index(esw, vport->vport));
 	if (!root_ns) {
 		esw_warn(dev, "Failed to get E-Switch egress flow namespace for vport (%d)\n", vport->vport);
 		return -EOPNOTSUPP;
@@ -1068,7 +1068,7 @@ static int esw_vport_enable_ingress_acl(struct mlx5_eswitch *esw,
 		  vport->vport, MLX5_CAP_ESW_INGRESS_ACL(dev, log_max_ft_size));
 
 	root_ns = mlx5_get_flow_vport_acl_namespace(dev, MLX5_FLOW_NAMESPACE_ESW_INGRESS,
-						    vport->vport);
+			mlx5_eswitch_vport_num_to_index(esw, vport->vport));
 	if (!root_ns) {
 		esw_warn(dev, "Failed to get E-Switch ingress flow namespace for vport (%d)\n", vport->vport);
 		return -EOPNOTSUPP;
Signature · 1 line
2.20.1
author maintainer
#163 lore raw

PATCH 162/371 ixgbe: Check DDM existence in transceiver before access

In reply to Greg Kroah-Hartman

[ Upstream commit 655c91414579d7bb115a4f7898ee726fc18e0984 ]

Some transceivers may comply with SFF-8472 but not implement the Digital Diagnostic Monitoring (DDM) interface described in it. The existence of such area is specified by bit 6 of byte 92, set to 1 if implemented.

Currently, due to not checking this bit ixgbe fails trying to read SFP module's eeprom with the follow message:

ethtool -m enP51p1s0f0
Cannot get Module EEPROM data: Input/output error

Because it fails to read the additional 256 bytes in which it was assumed to exist the DDM data.

This issue was noticed using a Mellanox Passive DAC PN 01FT738. The eeprom data was confirmed by Mellanox as correct and present in other Passive DACs in from other manufacturers.

---
 drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 3 ++-
 drivers/net/ethernet/intel/ixgbe/ixgbe_phy.h     | 1 +
 2 files changed, 3 insertions(+), 1 deletion(-)
drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c +2 -1
index acba067cc15a..7c52ae8ac005 100644
@@ -3226,7 +3226,8 @@ static int ixgbe_get_module_info(struct net_device *dev,
 		page_swap = true;
 	}
 
-	if (sff8472_rev == IXGBE_SFF_SFF_8472_UNSUP || page_swap) {
+	if (sff8472_rev == IXGBE_SFF_SFF_8472_UNSUP || page_swap ||
+	    !(addr_mode & IXGBE_SFF_DDM_IMPLEMENTED)) {
 		/* We have a SFP, but it does not support SFF-8472 */
 		modinfo->type = ETH_MODULE_SFF_8079;
 		modinfo->eeprom_len = ETH_MODULE_SFF_8079_LEN;
drivers/net/ethernet/intel/ixgbe/ixgbe_phy.h +1 -0
index 214b01085718..6544c4539c0d 100644
@@ -45,6 +45,7 @@
 #define IXGBE_SFF_SOFT_RS_SELECT_10G		0x8
 #define IXGBE_SFF_SOFT_RS_SELECT_1G		0x0
 #define IXGBE_SFF_ADDRESSING_MODE		0x4
+#define IXGBE_SFF_DDM_IMPLEMENTED		0x40
 #define IXGBE_SFF_QSFP_DA_ACTIVE_CABLE		0x1
 #define IXGBE_SFF_QSFP_DA_PASSIVE_CABLE		0x8
 #define IXGBE_SFF_QSFP_CONNECTOR_NOT_SEPARABLE	0x23
Signature · 1 line
2.20.1
author maintainer
#164 lore raw

PATCH 163/371 crypto: serpent - mark __serpent_setkey_sbox noinline

In reply to Greg Kroah-Hartman

[ Upstream commit 473971187d6727609951858c63bf12b0307ef015 ]

The same bug that gcc hit in the past is apparently now showing up with clang, which decides to inline __serpent_setkey_sbox:

crypto/serpent_generic.c:268:5: error: stack frame size of 2112 bytes in function '__serpent_setkey' [-Werror,-Wframe-larger-than=]

Marking it 'noinline' reduces the stack usage from 2112 bytes to 192 and 96 bytes, respectively, and seems to generate more useful object code.

Fixes c871c10e4ea7 ("crypto: serpent - improve __serpent_setkey with UBSAN")
---
 crypto/serpent_generic.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
crypto/serpent_generic.c +7 -1
index 7c3382facc82..600bd288881d 100644
@@ -229,7 +229,13 @@
 	x4 ^= x2;					\
 	})
 
-static void __serpent_setkey_sbox(u32 r0, u32 r1, u32 r2, u32 r3, u32 r4, u32 *k)
+/*
+ * both gcc and clang have misoptimized this function in the past,
+ * producing horrible object code from spilling temporary variables
+ * on the stack. Forcing this part out of line avoids that.
+ */
+static noinline void __serpent_setkey_sbox(u32 r0, u32 r1, u32 r2,
+					   u32 r3, u32 r4, u32 *k)
 {
 	k += 100;
 	S3(r3, r4, r0, r1, r2); store_and_load_keys(r1, r2, r4, r3, 28, 24);
Signature · 1 line
2.20.1
author maintainer
#165 lore raw

PATCH 164/371 crypto: asymmetric_keys - select CRYPTO_HASH where needed

In reply to Greg Kroah-Hartman

[ Upstream commit 90acc0653d2bee203174e66d519fbaaa513502de ]

Build testing with some core crypto options disabled revealed a few modules that are missing CRYPTO_HASH:

crypto/asymmetric_keys/x509_public_key.o: In function `x509_get_sig_params':
x509_public_key.c:(.text+0x4c7): undefined reference to crypto_alloc_shash' x509_public_key.c:(.text+0x5e5): undefined reference to crypto_shash_digest' crypto/asymmetric_keys/pkcs7_verify.o: In function `pkcs7_digest.isra.0':
pkcs7_verify.c:(.text+0xab): undefined reference to crypto_alloc_shash' pkcs7_verify.c:(.text+0x1b2): undefined reference to crypto_shash_digest' pkcs7_verify.c:(.text+0x3c1): undefined reference to crypto_shash_update' pkcs7_verify.c:(.text+0x411): undefined reference to crypto_shash_finup'

This normally doesn't show up in randconfig tests because there is a large number of other options that select CRYPTO_HASH.

---
 crypto/asymmetric_keys/Kconfig | 3 +++
 1 file changed, 3 insertions(+)
crypto/asymmetric_keys/Kconfig +3 -0
index be70ca6c85d3..1f1f004dc757 100644
@@ -15,6 +15,7 @@ config ASYMMETRIC_PUBLIC_KEY_SUBTYPE
 	select MPILIB
 	select CRYPTO_HASH_INFO
 	select CRYPTO_AKCIPHER
+	select CRYPTO_HASH
 	help
 	  This option provides support for asymmetric public key type handling.
 	  If signature generation and/or verification are to be used,
@@ -65,6 +66,7 @@ config TPM_KEY_PARSER
 config PKCS7_MESSAGE_PARSER
 	tristate "PKCS#7 message parser"
 	depends on X509_CERTIFICATE_PARSER
+	select CRYPTO_HASH
 	select ASN1
 	select OID_REGISTRY
 	help
@@ -87,6 +89,7 @@ config SIGNED_PE_FILE_VERIFICATION
 	bool "Support for PE file signature verification"
 	depends on PKCS7_MESSAGE_PARSER=y
 	depends on SYSTEM_DATA_VERIFICATION
+	select CRYPTO_HASH
 	select ASN1
 	select OID_REGISTRY
 	help
Signature · 1 line
2.20.1
author maintainer
#166 lore raw

PATCH 165/371 ath9k: correctly handle short radar pulses

In reply to Greg Kroah-Hartman

[ Upstream commit df5c4150501ee7e86383be88f6490d970adcf157 ]

In commit 3c0efb745a17 ("ath9k: discard undersized packets") the lower bound of RX packets was set to 10 (min ACK size) to filter those that would otherwise be treated as invalid at mac80211.

Alas, short radar pulses are reported as PHY_ERROR frames with length set to 3. Therefore their detection stopped working after that commit.

NOTE: ath9k drivers built thereafter will not pass DFS certification.

This extends the criteria for short packets to explicitly handle PHY_ERROR frames.

Fixes 3c0efb745a17 ("ath9k: discard undersized packets")
---
 drivers/net/wireless/ath/ath9k/recv.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/net/wireless/ath/ath9k/recv.c +5 -1
index 4e97f7f3b2a3..06e660858766 100644
@@ -815,6 +815,7 @@ static int ath9k_rx_skb_preprocess(struct ath_softc *sc,
 	struct ath_common *common = ath9k_hw_common(ah);
 	struct ieee80211_hdr *hdr;
 	bool discard_current = sc->rx.discard_next;
+	bool is_phyerr;
 
 	/*
 	 * Discard corrupt descriptors which are marked in
@@ -827,8 +828,11 @@ static int ath9k_rx_skb_preprocess(struct ath_softc *sc,
 
 	/*
 	 * Discard zero-length packets and packets smaller than an ACK
+	 * which are not PHY_ERROR (short radar pulses have a length of 3)
 	 */
-	if (rx_stats->rs_datalen < 10) {
+	is_phyerr = rx_stats->rs_status & ATH9K_RXERR_PHY;
+	if (!rx_stats->rs_datalen ||
+	    (rx_stats->rs_datalen < 10 && !is_phyerr)) {
 		RX_STAT_INC(sc, rx_len_err);
 		goto corrupt;
 	}
Signature · 1 line
2.20.1
author maintainer
#167 lore raw

PATCH 166/371 wil6210: drop old event after wmi_call timeout

In reply to Greg Kroah-Hartman

[ Upstream commit 1a276003111c0404f6bfeffe924c5a21f482428b ]

This change fixes a rare race condition of handling WMI events after wmi_call expires.

wmi_recv_cmd immediately handles an event when reply_buf is defined and a wmi_call is waiting for the event.
However, in case the wmi_call has already timed-out, there will be no waiting/running wmi_call and the event will be queued in WMI queue and will be handled later in wmi_event_handle.
Meanwhile, a new similar wmi_call for the same command and event may be issued. In this case, when handling the queued event we got WARN_ON printed.

Fixing this case as a valid timeout and drop the unexpected event.

---
 drivers/net/wireless/ath/wil6210/wmi.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)
drivers/net/wireless/ath/wil6210/wmi.c +12 -1
index 63116f4b62c7..de52e532c105 100644
@@ -3211,7 +3211,18 @@ static void wmi_event_handle(struct wil6210_priv *wil,
 		/* check if someone waits for this event */
 		if (wil->reply_id && wil->reply_id == id &&
 		    wil->reply_mid == mid) {
-			WARN_ON(wil->reply_buf);
+			if (wil->reply_buf) {
+				/* event received while wmi_call is waiting
+				 * with a buffer. Such event should be handled
+				 * in wmi_recv_cmd function. Handling the event
+				 * here means a previous wmi_call was timeout.
+				 * Drop the event and do not handle it.
+				 */
+				wil_err(wil,
+					"Old event (%d, %s) while wmi_call is waiting. Drop it and Continue waiting\n",
+					id, eventid2name(id));
+				return;
+			}
 
 			wmi_evt_call_handler(vif, id, evt_data,
 					     len - sizeof(*wmi));
Signature · 1 line
2.20.1
author maintainer
#168 lore raw

PATCH 167/371 EDAC: Fix global-out-of-bounds write when setting edac_mc_poll_msec

In reply to Greg Kroah-Hartman

[ Upstream commit d8655e7630dafa88bc37f101640e39c736399771 ]

Commit 9da21b1509d8 ("EDAC: Poll timeout cannot be zero, p2") assumes edac_mc_poll_msec to be unsigned long, but the type of the variable still remained as int. Setting edac_mc_poll_msec can trigger out-of-bounds write.

Reproducer:

  # echo 1001 > /sys/module/edac_core/parameters/edac_mc_poll_msec

KASAN report:

  BUG: KASAN: global-out-of-bounds in edac_set_poll_msec+0x140/0x150
  Write of size 8 at addr ffffffffb91b2d00 by task bash/1996

  CPU: 1 PID: 1996 Comm: bash Not tainted 5.2.0-rc6+ #23
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-2.fc30 04/01/2014
  Call Trace:
   dump_stack+0xca/0x13e
   print_address_description.cold+0x5/0x246
   __kasan_report.cold+0x75/0x9a
   ? edac_set_poll_msec+0x140/0x150
   kasan_report+0xe/0x20
   edac_set_poll_msec+0x140/0x150
   ? dimmdev_location_show+0x30/0x30
   ? vfs_lock_file+0xe0/0xe0
   ? _raw_spin_lock+0x87/0xe0
   param_attr_store+0x1b5/0x310
   ? param_array_set+0x4f0/0x4f0
   module_attr_store+0x58/0x80
   ? module_attr_show+0x80/0x80
   sysfs_kf_write+0x13d/0x1a0
   kernfs_fop_write+0x2bc/0x460
   ? sysfs_kf_bin_read+0x270/0x270
   ? kernfs_notify+0x1f0/0x1f0
   __vfs_write+0x81/0x100
   vfs_write+0x1e1/0x560
   ksys_write+0x126/0x250
   ? __ia32_sys_read+0xb0/0xb0
   ? do_syscall_64+0x1f/0x390
   do_syscall_64+0xc1/0x390
   entry_SYSCALL_64_after_hwframe+0x49/0xbe
  RIP: 0033:0x7fa7caa5e970
  Code: 73 01 c3 48 8b 0d 28 d5 2b 00 f7 d8 64 89 01 48 83 c8 ff c3 66 0f 1f 44 00 00 83 3d 99 2d 2c 00 00 75 10 b8 01 00 00 00 04
  RSP: 002b:00007fff6acfdfe8 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
  RAX: ffffffffffffffda RBX: 0000000000000005 RCX: 00007fa7caa5e970
  RDX: 0000000000000005 RSI: 0000000000e95c08 RDI: 0000000000000001
  RBP: 0000000000e95c08 R08: 00007fa7cad1e760 R09: 00007fa7cb36a700
  R10: 0000000000000073 R11: 0000000000000246 R12: 0000000000000005
  R13: 0000000000000001 R14: 00007fa7cad1d600 R15: 0000000000000005

  The buggy address belongs to the variable:
   edac_mc_poll_msec+0x0/0x40

  Memory state around the buggy address:
   ffffffffb91b2c00: 00 00 00 00 fa fa fa fa 00 00 00 00 fa fa fa fa
   ffffffffb91b2c80: 00 00 00 00 fa fa fa fa 00 00 00 00 fa fa fa fa
  >ffffffffb91b2d00: 04 fa fa fa fa fa fa fa 04 fa fa fa fa fa fa fa
                     ^
   ffffffffb91b2d80: 04 fa fa fa fa fa fa fa 00 00 00 00 00 00 00 00
   ffffffffb91b2e00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Fix it by changing the type of edac_mc_poll_msec to unsigned int. The reason why this patch adopts unsigned int rather than unsigned long is msecs_to_jiffies() assumes arg to be unsigned int. We can avoid integer conversion bugs and unsigned int will be large enough for edac_mc_poll_msec.

Fixes 9da21b1509d8 ("EDAC: Poll timeout cannot be zero, p2")
---
 drivers/edac/edac_mc_sysfs.c | 16 ++++++++--------
 drivers/edac/edac_module.h   |  2 +-
 2 files changed, 9 insertions(+), 9 deletions(-)
drivers/edac/edac_mc_sysfs.c +8 -8
index 7c01e1cc030c..4386ea4b9b5a 100644
@@ -26,7 +26,7 @@
 static int edac_mc_log_ue = 1;
 static int edac_mc_log_ce = 1;
 static int edac_mc_panic_on_ue;
-static int edac_mc_poll_msec = 1000;
+static unsigned int edac_mc_poll_msec = 1000;
 
 /* Getter functions for above */
 int edac_mc_get_log_ue(void)
@@ -45,30 +45,30 @@ int edac_mc_get_panic_on_ue(void)
 }
 
 /* this is temporary */
-int edac_mc_get_poll_msec(void)
+unsigned int edac_mc_get_poll_msec(void)
 {
 	return edac_mc_poll_msec;
 }
 
 static int edac_set_poll_msec(const char *val, const struct kernel_param *kp)
 {
-	unsigned long l;
+	unsigned int i;
 	int ret;
 
 	if (!val)
 		return -EINVAL;
 
-	ret = kstrtoul(val, 0, &l);
+	ret = kstrtouint(val, 0, &i);
 	if (ret)
 		return ret;
 
-	if (l < 1000)
+	if (i < 1000)
 		return -EINVAL;
 
-	*((unsigned long *)kp->arg) = l;
+	*((unsigned int *)kp->arg) = i;
 
 	/* notify edac_mc engine to reset the poll period */
-	edac_mc_reset_delay_period(l);
+	edac_mc_reset_delay_period(i);
 
 	return 0;
 }
@@ -82,7 +82,7 @@ MODULE_PARM_DESC(edac_mc_log_ue,
 module_param(edac_mc_log_ce, int, 0644);
 MODULE_PARM_DESC(edac_mc_log_ce,
 		 "Log correctable error to console: 0=off 1=on");
-module_param_call(edac_mc_poll_msec, edac_set_poll_msec, param_get_int,
+module_param_call(edac_mc_poll_msec, edac_set_poll_msec, param_get_uint,
 		  &edac_mc_poll_msec, 0644);
 MODULE_PARM_DESC(edac_mc_poll_msec, "Polling period in milliseconds");
 
drivers/edac/edac_module.h +1 -1
index dd7d0b509aa3..75528f07abd5 100644
@@ -36,7 +36,7 @@ extern int edac_mc_get_log_ue(void);
 extern int edac_mc_get_log_ce(void);
 extern int edac_mc_get_panic_on_ue(void);
 extern int edac_get_poll_msec(void);
-extern int edac_mc_get_poll_msec(void);
+extern unsigned int edac_mc_get_poll_msec(void);
 
 unsigned edac_dimm_info_location(struct dimm_info *dimm, char *buf,
 				 unsigned len);
Signature · 1 line
2.20.1
author maintainer
#169 lore raw

PATCH 168/371 bcache: check CACHE_SET_IO_DISABLE in allocator code

In reply to Greg Kroah-Hartman

[ Upstream commit e775339e1ae1205b47d94881db124c11385e597c ]

If CACHE_SET_IO_DISABLE of a cache set flag is set by too many I/O errors, currently allocator routines can still continue allocate space which may introduce inconsistent metadata state.

This patch checkes CACHE_SET_IO_DISABLE bit in following allocator routines,
- bch_bucket_alloc()
- __bch_bucket_alloc_set()
Once CACHE_SET_IO_DISABLE is set on cache set, the allocator routines may reject allocation request earlier to avoid potential inconsistent metadata.

---
 drivers/md/bcache/alloc.c | 9 +++++++++
 1 file changed, 9 insertions(+)
drivers/md/bcache/alloc.c +9 -0
index f8986effcb50..6f776823b9ba 100644
@@ -393,6 +393,11 @@ long bch_bucket_alloc(struct cache *ca, unsigned int reserve, bool wait)
 	struct bucket *b;
 	long r;
 
+
+	/* No allocation if CACHE_SET_IO_DISABLE bit is set */
+	if (unlikely(test_bit(CACHE_SET_IO_DISABLE, &ca->set->flags)))
+		return -1;
+
 	/* fastpath */
 	if (fifo_pop(&ca->free[RESERVE_NONE], r) ||
 	    fifo_pop(&ca->free[reserve], r))
@@ -484,6 +489,10 @@ int __bch_bucket_alloc_set(struct cache_set *c, unsigned int reserve,
 {
 	int i;
 
+	/* No allocation if CACHE_SET_IO_DISABLE bit is set */
+	if (unlikely(test_bit(CACHE_SET_IO_DISABLE, &c->flags)))
+		return -1;
+
 	lockdep_assert_held(&c->bucket_lock);
 	BUG_ON(!n || n > c->caches_loaded || n > MAX_CACHES_PER_SET);
 
Signature · 1 line
2.20.1
author maintainer
#170 lore raw

PATCH 169/371 bcache: check CACHE_SET_IO_DISABLE bit in bch_journal()

In reply to Greg Kroah-Hartman

[ Upstream commit 383ff2183ad16a8842d1fbd9dd3e1cbd66813e64 ]

When too many I/O errors happen on cache set and CACHE_SET_IO_DISABLE bit is set, bch_journal() may continue to work because the journaling bkey might be still in write set yet. The caller of bch_journal() may believe the journal still work but the truth is in-memory journal write set won't be written into cache device any more. This behavior may introduce potential inconsistent metadata status.

This patch checks CACHE_SET_IO_DISABLE bit at the head of bch_journal(), if the bit is set, bch_journal() returns NULL immediately to notice caller to know journal does not work.

---
 drivers/md/bcache/journal.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/md/bcache/journal.c +4 -0
index 6c94fa007796..ac3cec56ec19 100644
@@ -810,6 +810,10 @@ atomic_t *bch_journal(struct cache_set *c,
 	struct journal_write *w;
 	atomic_t *ret;
 
+	/* No journaling if CACHE_SET_IO_DISABLE set already */
+	if (unlikely(test_bit(CACHE_SET_IO_DISABLE, &c->flags)))
+		return NULL;
+
 	if (!CACHE_SYNC(&c->sb))
 		return NULL;
 
Signature · 1 line
2.20.1
author maintainer
#171 lore raw

PATCH 170/371 bcache: acquire bch_register_lock later in cached_dev_free()

In reply to Greg Kroah-Hartman

[ Upstream commit 80265d8dfd77792e133793cef44a21323aac2908 ]

When enable lockdep engine, a lockdep warning can be observed when reboot or shutdown system,

[ 3142.764557][    T1] bcache: bcache_reboot() Stopping all devices:
[ 3142.776265][ T2649]
[ 3142.777159][ T2649] ======================================================
[ 3142.780039][ T2649] WARNING: possible circular locking dependency detected
[ 3142.782869][ T2649] 5.2.0-rc4-lp151.20-default+ #1 Tainted: G        W
[ 3142.785684][ T2649] ------------------------------------------------------
[ 3142.788479][ T2649] kworker/3:67/2649 is trying to acquire lock:
[ 3142.790738][ T2649] 00000000aaf02291 ((wq_completion)bcache_writeback_wq){+.+.}, at: flush_workqueue+0x87/0x4c0
[ 3142.794678][ T2649]
[ 3142.794678][ T2649] but task is already holding lock:
[ 3142.797402][ T2649] 000000004fcf89c5 (&bch_register_lock){+.+.}, at: cached_dev_free+0x17/0x120 [bcache]
[ 3142.801462][ T2649]
[ 3142.801462][ T2649] which lock already depends on the new lock.
[ 3142.801462][ T2649]
[ 3142.805277][ T2649]
[ 3142.805277][ T2649] the existing dependency chain (in reverse order) is:
[ 3142.808902][ T2649]
[ 3142.808902][ T2649] -> #2 (&bch_register_lock){+.+.}:
[ 3142.812396][ T2649]        __mutex_lock+0x7a/0x9d0
[ 3142.814184][ T2649]        cached_dev_free+0x17/0x120 [bcache]
[ 3142.816415][ T2649]        process_one_work+0x2a4/0x640
[ 3142.818413][ T2649]        worker_thread+0x39/0x3f0
[ 3142.820276][ T2649]        kthread+0x125/0x140
[ 3142.822061][ T2649]        ret_from_fork+0x3a/0x50
[ 3142.823965][ T2649]
[ 3142.823965][ T2649] -> #1 ((work_completion)(&cl->work)#2){+.+.}:
[ 3142.827244][ T2649]        process_one_work+0x277/0x640
[ 3142.829160][ T2649]        worker_thread+0x39/0x3f0
[ 3142.830958][ T2649]        kthread+0x125/0x140
[ 3142.832674][ T2649]        ret_from_fork+0x3a/0x50
[ 3142.834915][ T2649]
[ 3142.834915][ T2649] -> #0 ((wq_completion)bcache_writeback_wq){+.+.}:
[ 3142.838121][ T2649]        lock_acquire+0xb4/0x1c0
[ 3142.840025][ T2649]        flush_workqueue+0xae/0x4c0
[ 3142.842035][ T2649]        drain_workqueue+0xa9/0x180
[ 3142.844042][ T2649]        destroy_workqueue+0x17/0x250
[ 3142.846142][ T2649]        cached_dev_free+0x52/0x120 [bcache]
[ 3142.848530][ T2649]        process_one_work+0x2a4/0x640
[ 3142.850663][ T2649]        worker_thread+0x39/0x3f0
[ 3142.852464][ T2649]        kthread+0x125/0x140
[ 3142.854106][ T2649]        ret_from_fork+0x3a/0x50
[ 3142.855880][ T2649]
[ 3142.855880][ T2649] other info that might help us debug this:
[ 3142.855880][ T2649]
[ 3142.859663][ T2649] Chain exists of:
[ 3142.859663][ T2649]   (wq_completion)bcache_writeback_wq --> (work_completion)(&cl->work)#2 --> &bch_register_lock
[ 3142.859663][ T2649]
[ 3142.865424][ T2649]  Possible unsafe locking scenario:
[ 3142.865424][ T2649]
[ 3142.868022][ T2649]        CPU0                    CPU1
[ 3142.869885][ T2649]        ----                    ----
[ 3142.871751][ T2649]   lock(&bch_register_lock);
[ 3142.873379][ T2649]                                lock((work_completion)(&cl->work)#2);
[ 3142.876399][ T2649]                                lock(&bch_register_lock);
[ 3142.879727][ T2649]   lock((wq_completion)bcache_writeback_wq);
[ 3142.882064][ T2649]
[ 3142.882064][ T2649]  *** DEADLOCK ***
[ 3142.882064][ T2649]
[ 3142.885060][ T2649] 3 locks held by kworker/3:67/2649:
[ 3142.887245][ T2649]  #0: 00000000e774cdd0 ((wq_completion)events){+.+.}, at: process_one_work+0x21e/0x640
[ 3142.890815][ T2649]  #1: 00000000f7df89da ((work_completion)(&cl->work)#2){+.+.}, at: process_one_work+0x21e/0x640
[ 3142.894884][ T2649]  #2: 000000004fcf89c5 (&bch_register_lock){+.+.}, at: cached_dev_free+0x17/0x120 [bcache]
[ 3142.898797][ T2649]
[ 3142.898797][ T2649] stack backtrace:
[ 3142.900961][ T2649] CPU: 3 PID: 2649 Comm: kworker/3:67 Tainted: G        W         5.2.0-rc4-lp151.20-default+ #1
[ 3142.904789][ T2649] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/13/2018
[ 3142.909168][ T2649] Workqueue: events cached_dev_free [bcache]
[ 3142.911422][ T2649] Call Trace:
[ 3142.912656][ T2649]  dump_stack+0x85/0xcb
[ 3142.914181][ T2649]  print_circular_bug+0x19a/0x1f0
[ 3142.916193][ T2649]  __lock_acquire+0x16cd/0x1850
[ 3142.917936][ T2649]  ? __lock_acquire+0x6a8/0x1850
[ 3142.919704][ T2649]  ? lock_acquire+0xb4/0x1c0
[ 3142.921335][ T2649]  ? find_held_lock+0x34/0xa0
[ 3142.923052][ T2649]  lock_acquire+0xb4/0x1c0
[ 3142.924635][ T2649]  ? flush_workqueue+0x87/0x4c0
[ 3142.926375][ T2649]  flush_workqueue+0xae/0x4c0
[ 3142.928047][ T2649]  ? flush_workqueue+0x87/0x4c0
[ 3142.929824][ T2649]  ? drain_workqueue+0xa9/0x180
[ 3142.931686][ T2649]  drain_workqueue+0xa9/0x180
[ 3142.933534][ T2649]  destroy_workqueue+0x17/0x250
[ 3142.935787][ T2649]  cached_dev_free+0x52/0x120 [bcache]
[ 3142.937795][ T2649]  process_one_work+0x2a4/0x640
[ 3142.939803][ T2649]  worker_thread+0x39/0x3f0
[ 3142.941487][ T2649]  ? process_one_work+0x640/0x640
[ 3142.943389][ T2649]  kthread+0x125/0x140
[ 3142.944894][ T2649]  ? kthread_create_worker_on_cpu+0x70/0x70
[ 3142.947744][ T2649]  ret_from_fork+0x3a/0x50
[ 3142.970358][ T2649] bcache: bcache_device_free() bcache0 stopped

Here is how the deadlock happens.
1) bcache_reboot() calls bcache_device_stop(), then inside bcache_device_stop() BCACHE_DEV_CLOSING bit is set on d->flags. Then closure_queue(&d->cl) is called to invoke cached_dev_flush().
2) In cached_dev_flush(), cached_dev_free() is called by continu_at().
3) In cached_dev_free(), when stopping the writeback kthread of the cached device by kthread_stop(), dc->writeback_thread will be waken up to quite the kthread while-loop, then cached_dev_put() is called in bch_writeback_thread().
4) Calling cached_dev_put() in writeback kthread may drop dc->count to 0, then dc->detach kworker is scheduled, which is initialized as cached_dev_detach_finish().
5) Inside cached_dev_detach_finish(), the last line of code is to call closure_put(&dc->disk.cl), which drops the last reference counter of closrure dc->disk.cl, then the callback cached_dev_flush() gets called.
Now cached_dev_flush() is called for second time in the code path, the first time is in step 2). And again bch_register_lock will be acquired again, and a A-A lock (lockdep terminology) is happening.

The root cause of the above A-A lock is in cached_dev_free(), mutex bch_register_lock is held before stopping writeback kthread and other kworkers. Fortunately now we have variable 'bcache_is_reboot', which may prevent device registration or unregistration during reboot/shutdown time, so it is unncessary to hold bch_register_lock such early now.

This is how this patch fixes the reboot/shutdown time A-A lock issue:
After moving mutex_lock(&bch_register_lock) to a later location where before atomic_read(&dc->running) in cached_dev_free(), such A-A lock problem can be solved without any reboot time registration race.

---
 drivers/md/bcache/super.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/md/bcache/super.c +2 -2
index e489d2459569..cbde1cc1d2bd 100644
@@ -1186,8 +1186,6 @@ static void cached_dev_free(struct closure *cl)
 {
 	struct cached_dev *dc = container_of(cl, struct cached_dev, disk.cl);
 
-	mutex_lock(&bch_register_lock);
-
 	if (test_and_clear_bit(BCACHE_DEV_WB_RUNNING, &dc->disk.flags))
 		cancel_writeback_rate_update_dwork(dc);
 
@@ -1198,6 +1196,8 @@ static void cached_dev_free(struct closure *cl)
 	if (!IS_ERR_OR_NULL(dc->status_update_thread))
 		kthread_stop(dc->status_update_thread);
 
+	mutex_lock(&bch_register_lock);
+
 	if (atomic_read(&dc->running))
 		bd_unlink_disk_holder(dc->bdev, dc->disk.disk);
 	bcache_device_free(&dc->disk);
Signature · 1 line
2.20.1
author maintainer
#172 lore raw

PATCH 171/371 bcache: check c->gc_thread by IS_ERR_OR_NULL in cache_set_flush()

In reply to Greg Kroah-Hartman

[ Upstream commit b387e9b58679c60f5b1e4313939bd4878204fc37 ]

When system memory is in heavy pressure, bch_gc_thread_start() from run_cache_set() may fail due to out of memory. In such condition, c->gc_thread is assigned to -ENOMEM, not NULL pointer. Then in following failure code path bch_cache_set_error(), when cache_set_flush() gets called, the code piece to stop c->gc_thread is broken,

         if (!IS_ERR_OR_NULL(c->gc_thread))
                 kthread_stop(c->gc_thread);

And KASAN catches such NULL pointer deference problem, with the warning information:

[  561.207881] ==================================================================
[  561.207900] BUG: KASAN: null-ptr-deref in kthread_stop+0x3b/0x440
[  561.207904] Write of size 4 at addr 000000000000001c by task kworker/15:1/313
[  561.207913] CPU: 15 PID: 313 Comm: kworker/15:1 Tainted: G        W         5.0.0-vanilla+ #3
[  561.207916] Hardware name: Lenovo ThinkSystem SR650 -[7X05CTO1WW]-/-[7X05CTO1WW]-, BIOS -[IVE136T-2.10]- 03/22/2019
[  561.207935] Workqueue: events cache_set_flush [bcache]
[  561.207940] Call Trace:
[  561.207948]  dump_stack+0x9a/0xeb
[  561.207955]  ? kthread_stop+0x3b/0x440
[  561.207960]  ? kthread_stop+0x3b/0x440
[  561.207965]  kasan_report+0x176/0x192
[  561.207973]  ? kthread_stop+0x3b/0x440
[  561.207981]  kthread_stop+0x3b/0x440
[  561.207995]  cache_set_flush+0xd4/0x6d0 [bcache]
[  561.208008]  process_one_work+0x856/0x1620
[  561.208015]  ? find_held_lock+0x39/0x1d0
[  561.208028]  ? drain_workqueue+0x380/0x380
[  561.208048]  worker_thread+0x87/0xb80
[  561.208058]  ? __kthread_parkme+0xb6/0x180
[  561.208067]  ? process_one_work+0x1620/0x1620
[  561.208072]  kthread+0x326/0x3e0
[  561.208079]  ? kthread_create_worker_on_cpu+0xc0/0xc0
[  561.208090]  ret_from_fork+0x3a/0x50
[  561.208110] ==================================================================
[  561.208113] Disabling lock debugging due to kernel taint
[  561.208115] irq event stamp: 11800231
[  561.208126] hardirqs last  enabled at (11800231): [<ffffffff83008538>] do_syscall_64+0x18/0x410
[  561.208127] BUG: unable to handle kernel NULL pointer dereference at 000000000000001c
[  561.208129] #PF error: [WRITE]
[  561.312253] hardirqs last disabled at (11800230): [<ffffffff830052ff>] trace_hardirqs_off_thunk+0x1a/0x1c
[  561.312259] softirqs last  enabled at (11799832): [<ffffffff850005c7>] __do_softirq+0x5c7/0x8c3
[  561.405975] PGD 0 P4D 0
[  561.442494] softirqs last disabled at (11799821): [<ffffffff831add2c>] irq_exit+0x1ac/0x1e0
[  561.791359] Oops: 0002 [#1] SMP KASAN NOPTI
[  561.791362] CPU: 15 PID: 313 Comm: kworker/15:1 Tainted: G    B   W         5.0.0-vanilla+ #3
[  561.791363] Hardware name: Lenovo ThinkSystem SR650 -[7X05CTO1WW]-/-[7X05CTO1WW]-, BIOS -[IVE136T-2.10]- 03/22/2019
[  561.791371] Workqueue: events cache_set_flush [bcache]
[  561.791374] RIP: 0010:kthread_stop+0x3b/0x440
[  561.791376] Code: 00 00 65 8b 05 26 d5 e0 7c 89 c0 48 0f a3 05 ec aa df 02 0f 82 dc 02 00 00 4c 8d 63 20 be 04 00 00 00 4c 89 e7 e8 65 c5 53 00 <f0> ff 43 20 48 8d 7b 24 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48
[  561.791377] RSP: 0018:ffff88872fc8fd10 EFLAGS: 00010286
[  561.838895] bcache: bch_count_io_errors() nvme0n1: IO error on writing btree.
[  561.838916] bcache: bch_count_io_errors() nvme0n1: IO error on writing btree.
[  561.838934] bcache: bch_count_io_errors() nvme0n1: IO error on writing btree.
[  561.838948] bcache: bch_count_io_errors() nvme0n1: IO error on writing btree.
[  561.838966] bcache: bch_count_io_errors() nvme0n1: IO error on writing btree.
[  561.838979] bcache: bch_count_io_errors() nvme0n1: IO error on writing btree.
[  561.838996] bcache: bch_count_io_errors() nvme0n1: IO error on writing btree.
[  563.067028] RAX: 0000000000000000 RBX: fffffffffffffffc RCX: ffffffff832dd314
[  563.067030] RDX: 0000000000000000 RSI: 0000000000000004 RDI: 0000000000000297
[  563.067032] RBP: ffff88872fc8fe88 R08: fffffbfff0b8213d R09: fffffbfff0b8213d
[  563.067034] R10: 0000000000000001 R11: fffffbfff0b8213c R12: 000000000000001c
[  563.408618] R13: ffff88dc61cc0f68 R14: ffff888102b94900 R15: ffff88dc61cc0f68
[  563.408620] FS:  0000000000000000(0000) GS:ffff888f7dc00000(0000) knlGS:0000000000000000
[  563.408622] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  563.408623] CR2: 000000000000001c CR3: 0000000f48a1a004 CR4: 00000000007606e0
[  563.408625] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[  563.408627] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[  563.904795] bcache: bch_count_io_errors() nvme0n1: IO error on writing btree.
[  563.915796] PKRU: 55555554
[  563.915797] Call Trace:
[  563.915807]  cache_set_flush+0xd4/0x6d0 [bcache]
[  563.915812]  process_one_work+0x856/0x1620
[  564.001226] bcache: bch_count_io_errors() nvme0n1: IO error on writing btree.
[  564.033563]  ? find_held_lock+0x39/0x1d0
[  564.033567]  ? drain_workqueue+0x380/0x380
[  564.033574]  worker_thread+0x87/0xb80
[  564.062823] bcache: bch_count_io_errors() nvme0n1: IO error on writing btree.
[  564.118042]  ? __kthread_parkme+0xb6/0x180
[  564.118046]  ? process_one_work+0x1620/0x1620
[  564.118048]  kthread+0x326/0x3e0
[  564.118050]  ? kthread_create_worker_on_cpu+0xc0/0xc0
[  564.167066] bcache: bch_count_io_errors() nvme0n1: IO error on writing btree.
[  564.252441]  ret_from_fork+0x3a/0x50
[  564.252447] Modules linked in: msr rpcrdma sunrpc rdma_ucm ib_iser ib_umad rdma_cm ib_ipoib i40iw configfs iw_cm ib_cm libiscsi scsi_transport_iscsi mlx4_ib ib_uverbs mlx4_en ib_core nls_iso8859_1 nls_cp437 vfat fat intel_rapl skx_edac x86_pkg_temp_thermal coretemp iTCO_wdt iTCO_vendor_support crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel ses raid0 aesni_intel cdc_ether enclosure usbnet ipmi_ssif joydev aes_x86_64 i40e scsi_transport_sas mii bcache md_mod crypto_simd mei_me ioatdma crc64 ptp cryptd pcspkr i2c_i801 mlx4_core glue_helper pps_core mei lpc_ich dca wmi ipmi_si ipmi_devintf nd_pmem dax_pmem nd_btt ipmi_msghandler device_dax pcc_cpufreq button hid_generic usbhid mgag200 i2c_algo_bit drm_kms_helper syscopyarea sysfillrect xhci_pci sysimgblt fb_sys_fops xhci_hcd ttm megaraid_sas drm usbcore nfit libnvdimm sg dm_multipath dm_mod scsi_dh_rdac scsi_dh_emc scsi_dh_alua efivarfs
[  564.299390] bcache: bch_count_io_errors() nvme0n1: IO error on writing btree.
[  564.348360] CR2: 000000000000001c
[  564.348362] ---[ end trace b7f0e5cc7b2103b0 ]---

Therefore, it is not enough to only check whether c->gc_thread is NULL, we should use IS_ERR_OR_NULL() to check both NULL pointer and error value.

This patch changes the above buggy code piece in this way,

         if (!IS_ERR_OR_NULL(c->gc_thread))
                 kthread_stop(c->gc_thread);
---
 drivers/md/bcache/super.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/md/bcache/super.c +1 -1
index cbde1cc1d2bd..dd36e3dfd00f 100644
@@ -1557,7 +1557,7 @@ static void cache_set_flush(struct closure *cl)
 	kobject_put(&c->internal);
 	kobject_del(&c->kobj);
 
-	if (c->gc_thread)
+	if (!IS_ERR_OR_NULL(c->gc_thread))
 		kthread_stop(c->gc_thread);
 
 	if (!IS_ERR_OR_NULL(c->root))
Signature · 1 line
2.20.1
author maintainer
#173 lore raw

PATCH 172/371 bcache: fix potential deadlock in cached_def_free()

In reply to Greg Kroah-Hartman

[ Upstream commit 7e865eba00a3df2dc8c4746173a8ca1c1c7f042e ]

When enable lockdep and reboot system with a writeback mode bcache device, the following potential deadlock warning is reported by lockdep engine.

[  101.536569][  T401] kworker/2:2/401 is trying to acquire lock:
[  101.538575][  T401] 00000000bbf6e6c7 ((wq_completion)bcache_writeback_wq){+.+.}, at: flush_workqueue+0x87/0x4c0
[  101.542054][  T401]
[  101.542054][  T401] but task is already holding lock:
[  101.544587][  T401] 00000000f5f305b3 ((work_completion)(&cl->work)#2){+.+.}, at: process_one_work+0x21e/0x640
[  101.548386][  T401]
[  101.548386][  T401] which lock already depends on the new lock.
[  101.548386][  T401]
[  101.551874][  T401]
[  101.551874][  T401] the existing dependency chain (in reverse order) is:
[  101.555000][  T401]
[  101.555000][  T401] -> #1 ((work_completion)(&cl->work)#2){+.+.}:
[  101.557860][  T401]        process_one_work+0x277/0x640
[  101.559661][  T401]        worker_thread+0x39/0x3f0
[  101.561340][  T401]        kthread+0x125/0x140
[  101.562963][  T401]        ret_from_fork+0x3a/0x50
[  101.564718][  T401]
[  101.564718][  T401] -> #0 ((wq_completion)bcache_writeback_wq){+.+.}:
[  101.567701][  T401]        lock_acquire+0xb4/0x1c0
[  101.569651][  T401]        flush_workqueue+0xae/0x4c0
[  101.571494][  T401]        drain_workqueue+0xa9/0x180
[  101.573234][  T401]        destroy_workqueue+0x17/0x250
[  101.575109][  T401]        cached_dev_free+0x44/0x120 [bcache]
[  101.577304][  T401]        process_one_work+0x2a4/0x640
[  101.579357][  T401]        worker_thread+0x39/0x3f0
[  101.581055][  T401]        kthread+0x125/0x140
[  101.582709][  T401]        ret_from_fork+0x3a/0x50
[  101.584592][  T401]
[  101.584592][  T401] other info that might help us debug this:
[  101.584592][  T401]
[  101.588355][  T401]  Possible unsafe locking scenario:
[  101.588355][  T401]
[  101.590974][  T401]        CPU0                    CPU1
[  101.592889][  T401]        ----                    ----
[  101.594743][  T401]   lock((work_completion)(&cl->work)#2);
[  101.596785][  T401]                                lock((wq_completion)bcache_writeback_wq);
[  101.600072][  T401]                                lock((work_completion)(&cl->work)#2);
[  101.602971][  T401]   lock((wq_completion)bcache_writeback_wq);
[  101.605255][  T401]
[  101.605255][  T401]  *** DEADLOCK ***
[  101.605255][  T401]
[  101.608310][  T401] 2 locks held by kworker/2:2/401:
[  101.610208][  T401]  #0: 00000000cf2c7d17 ((wq_completion)events){+.+.}, at: process_one_work+0x21e/0x640
[  101.613709][  T401]  #1: 00000000f5f305b3 ((work_completion)(&cl->work)#2){+.+.}, at: process_one_work+0x21e/0x640
[  101.617480][  T401]
[  101.617480][  T401] stack backtrace:
[  101.619539][  T401] CPU: 2 PID: 401 Comm: kworker/2:2 Tainted: G        W         5.2.0-rc4-lp151.20-default+ #1
[  101.623225][  T401] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 04/13/2018
[  101.627210][  T401] Workqueue: events cached_dev_free [bcache]
[  101.629239][  T401] Call Trace:
[  101.630360][  T401]  dump_stack+0x85/0xcb
[  101.631777][  T401]  print_circular_bug+0x19a/0x1f0
[  101.633485][  T401]  __lock_acquire+0x16cd/0x1850
[  101.635184][  T401]  ? __lock_acquire+0x6a8/0x1850
[  101.636863][  T401]  ? lock_acquire+0xb4/0x1c0
[  101.638421][  T401]  ? find_held_lock+0x34/0xa0
[  101.640015][  T401]  lock_acquire+0xb4/0x1c0
[  101.641513][  T401]  ? flush_workqueue+0x87/0x4c0
[  101.643248][  T401]  flush_workqueue+0xae/0x4c0
[  101.644832][  T401]  ? flush_workqueue+0x87/0x4c0
[  101.646476][  T401]  ? drain_workqueue+0xa9/0x180
[  101.648303][  T401]  drain_workqueue+0xa9/0x180
[  101.649867][  T401]  destroy_workqueue+0x17/0x250
[  101.651503][  T401]  cached_dev_free+0x44/0x120 [bcache]
[  101.653328][  T401]  process_one_work+0x2a4/0x640
[  101.655029][  T401]  worker_thread+0x39/0x3f0
[  101.656693][  T401]  ? process_one_work+0x640/0x640
[  101.658501][  T401]  kthread+0x125/0x140
[  101.660012][  T401]  ? kthread_create_worker_on_cpu+0x70/0x70
[  101.661985][  T401]  ret_from_fork+0x3a/0x50
[  101.691318][  T401] bcache: bcache_device_free() bcache0 stopped

Here is how the above potential deadlock may happen in reboot/shutdown code path,
1) bcache_reboot() is called firstly in the reboot/shutdown code path, then in bcache_reboot(), bcache_device_stop() is called.
2) bcache_device_stop() sets BCACHE_DEV_CLOSING on d->falgs, then call closure_queue(&d->cl) to invoke cached_dev_flush(). And in turn cached_dev_flush() calls cached_dev_free() via closure_at()
3) In cached_dev_free(), after stopped writebach kthread dc->writeback_thread, the kwork dc->writeback_write_wq is stopping by destroy_workqueue().
4) Inside destroy_workqueue(), drain_workqueue() is called. Inside drain_workqueue(), flush_workqueue() is called. Then wq->lockdep_map is acquired by lock_map_acquire() in flush_workqueue(). After the lock acquired the rest part of flush_workqueue() just wait for the workqueue to complete.
5) Now we look back at writeback thread routine bch_writeback_thread(), in the main while-loop, write_dirty() is called via continue_at() in read_dirty_submit(), which is called via continue_at() in while-loop level called function read_dirty(). Inside write_dirty() it may be re-called on workqueeu dc->writeback_write_wq via continue_at(). It means when the writeback kthread is stopped in cached_dev_free() there might be still one kworker queued on dc->writeback_write_wq to execute write_dirty() again.
6) Now this kworker is scheduled on dc->writeback_write_wq to run by process_one_work() (which is called by worker_thread()). Before calling the kwork routine, wq->lockdep_map is acquired.
7) But wq->lockdep_map is acquired already in step 4), so a A-A lock (lockdep terminology) scenario happens.

Indeed on multiple cores syatem, the above deadlock is very rare to happen, just as the code comments in process_one_work() says,
2263 * AFAICT there is no possible deadlock scenario between the
2264 * flush_work() and complete() primitives (except for
single-threaded
2265 * workqueues), so hiding them isn't a problem.

But it is still good to fix such lockdep warning, even no one running bcache on single core system.

The fix is simple. This patch solves the above potential deadlock by,
- Do not destroy workqueue dc->writeback_write_wq in cached_dev_free().
- Flush and destroy dc->writeback_write_wq in writebach kthread routine bch_writeback_thread(), where after quit the thread main while-loop and before cached_dev_put() is called.

By this fix, dc->writeback_write_wq will be stopped and destroy before the writeback kthread stopped, so the chance for a A-A locking on wq->lockdep_map is disappeared, such A-A deadlock won't happen any more.

---
 drivers/md/bcache/super.c     | 2 --
 drivers/md/bcache/writeback.c | 4 ++++
 2 files changed, 4 insertions(+), 2 deletions(-)
drivers/md/bcache/super.c +0 -2
index dd36e3dfd00f..91252107b10c 100644
@@ -1191,8 +1191,6 @@ static void cached_dev_free(struct closure *cl)
 
 	if (!IS_ERR_OR_NULL(dc->writeback_thread))
 		kthread_stop(dc->writeback_thread);
-	if (dc->writeback_write_wq)
-		destroy_workqueue(dc->writeback_write_wq);
 	if (!IS_ERR_OR_NULL(dc->status_update_thread))
 		kthread_stop(dc->status_update_thread);
 
drivers/md/bcache/writeback.c +4 -0
index 73f0efac2b9f..df0f4e5a051a 100644
@@ -735,6 +735,10 @@ static int bch_writeback_thread(void *arg)
 		}
 	}
 
+	if (dc->writeback_write_wq) {
+		flush_workqueue(dc->writeback_write_wq);
+		destroy_workqueue(dc->writeback_write_wq);
+	}
 	cached_dev_put(dc);
 	wait_for_kthread_stop();
 
Signature · 1 line
2.20.1
author maintainer
#174 lore raw

PATCH 173/371 net: hns3: fix a -Wformat-nonliteral compile warning

In reply to Greg Kroah-Hartman

[ Upstream commit 18d219b783da61a6cc77581f55fc4af2fa16bc36 ]

When setting -Wformat=2, there is a compiler warning like this:

hclge_main.c:xxx:x: warning: format not a string literal and no format arguments [-Wformat-nonliteral]
strs[i].desc);
^~~~

This patch adds missing format parameter "%s" to snprintf() to fix it.

Fixes 46a3df9f9718 ("Add HNS3 Acceleration Engine & Compatibility Layer Support")
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +1 -2
index 6d4d5a470163..563eefa20003 100644
@@ -550,8 +550,7 @@ static u8 *hclge_comm_get_strings(u32 stringset,
 		return buff;
 
 	for (i = 0; i < size; i++) {
-		snprintf(buff, ETH_GSTRING_LEN,
-			 strs[i].desc);
+		snprintf(buff, ETH_GSTRING_LEN, "%s", strs[i].desc);
 		buff = buff + ETH_GSTRING_LEN;
 	}
 
Signature · 1 line
2.20.1
author maintainer
#175 lore raw

PATCH 174/371 net: hns3: add some error checking in hclge_tm module

In reply to Greg Kroah-Hartman

[ Upstream commit 04f25edb48c441fc278ecc154c270f16966cbb90 ]

When hdev->tx_sch_mode is HCLGE_FLAG_VNET_BASE_SCH_MODE, the hclge_tm_schd_mode_vnet_base_cfg calls hclge_tm_pri_schd_mode_cfg with vport->vport_id as pri_id, which is used as index for hdev->tm_info.tc_info, it will cause out of bound access issue if vport_id is equal to or larger than HNAE3_MAX_TC.

Also hardware only support maximum speed of HCLGE_ETHER_MAX_RATE.

So this patch adds two checks for above cases.

Fixes 848440544b41 ("net: hns3: Add support of TX Scheduler & Shaper to HNS3 driver")
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c +5 -1
index a7bbb6d3091a..0d53062f7bb5 100644
@@ -54,7 +54,8 @@ static int hclge_shaper_para_calc(u32 ir, u8 shaper_level,
 	u32 tick;
 
 	/* Calc tick */
-	if (shaper_level >= HCLGE_SHAPER_LVL_CNT)
+	if (shaper_level >= HCLGE_SHAPER_LVL_CNT ||
+	    ir > HCLGE_ETHER_MAX_RATE)
 		return -EINVAL;
 
 	tick = tick_array[shaper_level];
@@ -1124,6 +1125,9 @@ static int hclge_tm_schd_mode_vnet_base_cfg(struct hclge_vport *vport)
 	int ret;
 	u8 i;
 
+	if (vport->vport_id >= HNAE3_MAX_TC)
+		return -EINVAL;
+
 	ret = hclge_tm_pri_schd_mode_cfg(hdev, vport->vport_id);
 	if (ret)
 		return ret;
Signature · 1 line
2.20.1
author maintainer
#176 lore raw

PATCH 175/371 ath10k: Fix memory leak in qmi

In reply to Greg Kroah-Hartman

[ Upstream commit c709df58832c5f575f0255bea4b09ad477fc62ea ]

Currently the memory allocated for qmi handle is not being freed during de-init which leads to memory leak.

Free the allocated qmi memory in qmi deinit
to avoid memory leak.

Tested HW: WCN3990
Tested FW: WLAN.HL.3.1-01040-QCAHLSWMTPLZ-1

Fixes fda6fee0001e ("ath10k: add QMI message handshake for wcn3990 client")
---
 drivers/net/wireless/ath/ath10k/qmi.c | 1 +
 1 file changed, 1 insertion(+)
drivers/net/wireless/ath/ath10k/qmi.c +1 -0
index a7bc2c70d076..8f8f717a23ee 100644
@@ -1002,6 +1002,7 @@ int ath10k_qmi_deinit(struct ath10k *ar)
 	qmi_handle_release(&qmi->qmi_hdl);
 	cancel_work_sync(&qmi->event_work);
 	destroy_workqueue(qmi->event_wq);
+	kfree(qmi);
 	ar_snoc->qmi = NULL;
 
 	return 0;
Signature · 1 line
2.20.1
author maintainer
#177 lore raw

PATCH 176/371 ath10k: destroy sdio workqueue while remove sdio module

In reply to Greg Kroah-Hartman

[ Upstream commit 3ed39f8e747a7aafeec07bb244f2c3a1bdca5730 ]

The workqueue need to flush and destory while remove sdio module, otherwise it will have thread which is not destory after remove sdio modules.

Tested with QCA6174 SDIO with firmware
WLAN.RMH.4.4.1-00007-QCARMSWP-1.

---
 drivers/net/wireless/ath/ath10k/sdio.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/net/wireless/ath/ath10k/sdio.c +3 -0
index 73ef3e75d199..28bdf0212538 100644
@@ -2081,6 +2081,9 @@ static void ath10k_sdio_remove(struct sdio_func *func)
 	cancel_work_sync(&ar_sdio->wr_async_work);
 	ath10k_core_unregister(ar);
 	ath10k_core_destroy(ar);
+
+	flush_workqueue(ar_sdio->workqueue);
+	destroy_workqueue(ar_sdio->workqueue);
 }
 
 static const struct sdio_device_id ath10k_sdio_devices[] = {
Signature · 1 line
2.20.1
author maintainer
#178 lore raw

PATCH 177/371 net: mvpp2: prs: Dont override the sign bit in SRAM parser shift

In reply to Greg Kroah-Hartman

[ Upstream commit 8ec3ede559956f8ad58db7b57d25ac724bab69e9 ]

The Header Parser allows identifying various fields in the packet headers, used for various kind of filtering and classification steps.

This is a re-entrant process, where the offset in the packet header depends on the previous lookup results. This offset is represented in the SRAM results of the TCAM, as a shift to be operated.

This shift can be negative in some cases, such as in IPv6 parsing.

This commit prevents overriding the sign bit when setting the shift value, which could cause instabilities when parsing IPv6 flows.

Fixes 3f518509dedc ("ethernet: Add new driver for Marvell Armada 375 network unit")
---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_prs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/ethernet/marvell/mvpp2/mvpp2_prs.c +2 -1
index ae2240074d8e..5692c6087bbb 100644
@@ -312,7 +312,8 @@ static void mvpp2_prs_sram_shift_set(struct mvpp2_prs_entry *pe, int shift,
 	}
 
 	/* Set value */
-	pe->sram[MVPP2_BIT_TO_WORD(MVPP2_PRS_SRAM_SHIFT_OFFS)] = shift & MVPP2_PRS_SRAM_SHIFT_MASK;
+	pe->sram[MVPP2_BIT_TO_WORD(MVPP2_PRS_SRAM_SHIFT_OFFS)] |=
+		shift & MVPP2_PRS_SRAM_SHIFT_MASK;
 
 	/* Reset and set operation */
 	mvpp2_prs_sram_bits_clear(pe, MVPP2_PRS_SRAM_OP_SEL_SHIFT_OFFS,
Signature · 1 line
2.20.1
author maintainer
#179 lore raw

PATCH 178/371 igb: clear out skb->tstamp after reading the txtime

In reply to Greg Kroah-Hartman

[ Upstream commit 1e08511d5d01884a3c9070afd52a47799312074a ]

If a packet which is utilizing the launchtime feature (via SO_TXTIME socket option) also requests the hardware transmit timestamp, the hardware timestamp is not delivered to the userspace. This is because the value in skb->tstamp is mistaken as the software timestamp.

Applications, like ptp4l, request a hardware timestamp by setting the SOF_TIMESTAMPING_TX_HARDWARE socket option. Whenever a new timestamp is detected by the driver (this work is done in igb_ptp_tx_work() which calls igb_ptp_tx_hwtstamps() in igb_ptp.c[1]), it will queue the timestamp in the ERR_QUEUE for the userspace to read. When the userspace is ready, it will issue a recvmsg() call to collect this timestamp. The problem is in this recvmsg() call. If the skb->tstamp is not cleared out, it will be interpreted as a software timestamp and the hardware tx timestamp will not be successfully sent to the userspace. Look at skb_is_swtx_tstamp() and the callee function __sock_recv_timestamp() in net/socket.c for more details.

---
 drivers/net/ethernet/intel/igb/igb_main.c | 1 +
 1 file changed, 1 insertion(+)
drivers/net/ethernet/intel/igb/igb_main.c +1 -0
index 580d14b49fda..a725dc709632 100644
@@ -5687,6 +5687,7 @@ static void igb_tx_ctxtdesc(struct igb_ring *tx_ring,
 	 */
 	if (tx_ring->launchtime_enable) {
 		ts = ns_to_timespec64(first->skb->tstamp);
+		first->skb->tstamp = 0;
 		context_desc->seqnum_seed = cpu_to_le32(ts.tv_nsec / 32);
 	} else {
 		context_desc->seqnum_seed = 0;
Signature · 1 line
2.20.1
author maintainer
#180 lore raw

PATCH 179/371 net: hns3: add Asym Pause support to fix autoneg problem

In reply to Greg Kroah-Hartman

[ Upstream commit bc3781edcea017aa1a29abd953b776cdba298ce2 ]

Local device and link partner config auto-negotiation on both, local device config pause frame use as: rx on/tx off, link partner config pause frame use as: rx off/tx on.

We except the result is:
Local device:

Autonegotiate:  on
RX:             on
TX:             off
RX negotiated:  on
TX negotiated:  off

Link partner:

Autonegotiate:  on
RX:             off
TX:             on
RX negotiated:  off
TX negotiated:  on

But actually, the result of Local device and link partner is both:

Autonegotiate:  on
RX:             off
TX:             off
RX negotiated:  off
TX negotiated:  off

The root cause is that the supported flag is has only Pause, reference to the function genphy_config_advert():
static int genphy_config_advert(struct phy_device *phydev)

{
	...
	linkmode_and(phydev->advertising, phydev->advertising,
		     phydev->supported);
	...
}

The pause frame use of link partner is rx off/tx on, so its advertising only set the bit Asym_Pause, and the supported is only set the bit Pause, so the result of linkmode_and(), is rx off/tx off.

This patch adds Asym_Pause to the supported flag to fix it.

---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 1 +
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c | 7 +++++++
 2 files changed, 8 insertions(+)
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +1 -0
index 563eefa20003..14d37c26196b 100644
@@ -889,6 +889,7 @@ static void hclge_parse_copper_link_mode(struct hclge_dev *hdev,
 	linkmode_set_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, supported);
 	linkmode_set_bit(ETHTOOL_LINK_MODE_TP_BIT, supported);
 	linkmode_set_bit(ETHTOOL_LINK_MODE_Pause_BIT, supported);
+	linkmode_set_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT, supported);
 }
 
 static void hclge_parse_link_mode(struct hclge_dev *hdev, u8 speed_ability)
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c +7 -0
index 48eda2c6fdae..71a6f7c734b6 100644
@@ -215,6 +215,13 @@ int hclge_mac_connect_phy(struct hnae3_handle *handle)
 	linkmode_and(phydev->supported, phydev->supported, mask);
 	linkmode_copy(phydev->advertising, phydev->supported);
 
+	/* supported flag is Pause and Asym Pause, but default advertising
+	 * should be rx on, tx on, so need clear Asym Pause in advertising
+	 * flag
+	 */
+	linkmode_clear_bit(ETHTOOL_LINK_MODE_Asym_Pause_BIT,
+			   phydev->advertising);
+
 	return 0;
 }
 
Signature · 1 line
2.20.1
author maintainer
#181 lore raw

PATCH 180/371 ixgbe: Avoid NULL pointer dereference with VF on non-IPsec hw

In reply to Greg Kroah-Hartman

[ Upstream commit 92924064106e410cdc015f1dbfc0499309f9f5b1 ]

An ipsec structure will not be allocated if the hardware does not support offload. Fixes the following Oops:

[  191.045452] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000
[  191.054232] Mem abort info:
[  191.057014]   ESR = 0x96000004
[  191.060057]   Exception class = DABT (current EL), IL = 32 bits
[  191.065963]   SET = 0, FnV = 0
[  191.069004]   EA = 0, S1PTW = 0
[  191.072132] Data abort info:
[  191.074999]   ISV = 0, ISS = 0x00000004
[  191.078822]   CM = 0, WnR = 0
[  191.081780] user pgtable: 4k pages, 48-bit VAs, pgdp = 0000000043d9e467
[  191.088382] [0000000000000000] pgd=0000000000000000
[  191.093252] Internal error: Oops: 96000004 [#1] SMP
[  191.098119] Modules linked in: vhost_net vhost tap vfio_pci vfio_virqfd vfio_iommu_type1 vfio xt_CHECKSUM iptable_mangle ipt_MASQUERADE iptable_nat nf_nat_ipv4 nf_nat xt_conntrack nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 ipt_REJECT nf_reject_ipv4 xt_tcpudp bridge stp llc ebtable_filter devlink ebtables ip6table_filter ip6_tables iptable_filter bpfilter ipmi_ssif nls_iso8859_1 input_leds joydev ipmi_si hns_roce_hw_v2 ipmi_devintf hns_roce ipmi_msghandler cppc_cpufreq sch_fq_codel ib_iser rdma_cm iw_cm ib_cm ib_core iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 ses enclosure btrfs zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor hid_generic usbhid hid raid6_pq libcrc32c raid1 raid0 multipath linear ixgbevf hibmc_drm ttm
[  191.168607]  drm_kms_helper aes_ce_blk aes_ce_cipher syscopyarea crct10dif_ce sysfillrect ghash_ce qla2xxx sysimgblt sha2_ce sha256_arm64 hisi_sas_v3_hw fb_sys_fops sha1_ce uas nvme_fc mpt3sas ixgbe drm hisi_sas_main nvme_fabrics usb_storage hclge scsi_transport_fc ahci libsas hnae3 raid_class libahci xfrm_algo scsi_transport_sas mdio aes_neon_bs aes_neon_blk crypto_simd cryptd aes_arm64
[  191.202952] CPU: 94 PID: 0 Comm: swapper/94 Not tainted 4.19.0-rc1+ #11
[  191.209553] Hardware name: Huawei D06 /D06, BIOS Hisilicon D06 UEFI RC0 - V1.20.01 04/26/2019
[  191.218064] pstate: 20400089 (nzCv daIf +PAN -UAO)
[  191.222873] pc : ixgbe_ipsec_vf_clear+0x60/0xd0 [ixgbe]
[  191.228093] lr : ixgbe_msg_task+0x2d0/0x1088 [ixgbe]
[  191.233044] sp : ffff000009b3bcd0
[  191.236346] x29: ffff000009b3bcd0 x28: 0000000000000000
[  191.241647] x27: ffff000009628000 x26: 0000000000000000
[  191.246946] x25: ffff803f652d7600 x24: 0000000000000004
[  191.252246] x23: ffff803f6a718900 x22: 0000000000000000
[  191.257546] x21: 0000000000000000 x20: 0000000000000000
[  191.262845] x19: 0000000000000000 x18: 0000000000000000
[  191.268144] x17: 0000000000000000 x16: 0000000000000000
[  191.273443] x15: 0000000000000000 x14: 0000000100000026
[  191.278742] x13: 0000000100000025 x12: ffff8a5f7fbe0df0
[  191.284042] x11: 000000010000000b x10: 0000000000000040
[  191.289341] x9 : 0000000000001100 x8 : ffff803f6a824fd8
[  191.294640] x7 : ffff803f6a825098 x6 : 0000000000000001
[  191.299939] x5 : ffff000000f0ffc0 x4 : 0000000000000000
[  191.305238] x3 : ffff000028c00000 x2 : ffff803f652d7600
[  191.310538] x1 : 0000000000000000 x0 : ffff000000f205f0
[  191.315838] Process swapper/94 (pid: 0, stack limit = 0x00000000addfed5a)
[  191.322613] Call trace:
[  191.325055]  ixgbe_ipsec_vf_clear+0x60/0xd0 [ixgbe]
[  191.329927]  ixgbe_msg_task+0x2d0/0x1088 [ixgbe]
[  191.334536]  ixgbe_msix_other+0x274/0x330 [ixgbe]
[  191.339233]  __handle_irq_event_percpu+0x78/0x270
[  191.343924]  handle_irq_event_percpu+0x40/0x98
[  191.348355]  handle_irq_event+0x50/0xa8
[  191.352180]  handle_fasteoi_irq+0xbc/0x148
[  191.356263]  generic_handle_irq+0x34/0x50
[  191.360259]  __handle_domain_irq+0x68/0xc0
[  191.364343]  gic_handle_irq+0x84/0x180
[  191.368079]  el1_irq+0xe8/0x180
[  191.371208]  arch_cpu_idle+0x30/0x1a8
[  191.374860]  do_idle+0x1dc/0x2a0
[  191.378077]  cpu_startup_entry+0x2c/0x30
[  191.381988]  secondary_start_kernel+0x150/0x1e0
[  191.386506] Code: 6b15003f 54000320 f1404a9f 54000060 (79400260)
Fixes eda0333ac2930 ("ixgbe: add VF IPsec management")
---
 drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c +3 -0
index ff85ce5791a3..31629fc7e820 100644
@@ -842,6 +842,9 @@ void ixgbe_ipsec_vf_clear(struct ixgbe_adapter *adapter, u32 vf)
 	struct ixgbe_ipsec *ipsec = adapter->ipsec;
 	int i;
 
+	if (!ipsec)
+		return;
+
 	/* search rx sa table */
 	for (i = 0; i < IXGBE_IPSEC_MAX_SA_COUNT && ipsec->num_rx_sa; i++) {
 		if (!ipsec->rx_tbl[i].used)
Signature · 1 line
2.20.1
author maintainer
#182 lore raw

PATCH 181/371 iwlwifi: mvm: Drop large non sta frames

In reply to Greg Kroah-Hartman

[ Upstream commit ac70499ee97231a418dc1a4d6c9dc102e8f64631 ]

In some buggy scenarios we could possible attempt to transmit frames larger than maximum MSDU size. Since our devices don't know how to handle this, it may result in asserts, hangs etc.
This can happen, for example, when we receive a large multicast frame and try to transmit it back to the air in AP mode.
Since in a legal scenario this should never happen, drop such frames and warn about it.

---
 drivers/net/wireless/intel/iwlwifi/mvm/tx.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/net/wireless/intel/iwlwifi/mvm/tx.c +3 -0
index 0c2aabc842f9..96f8d38ea321 100644
@@ -726,6 +726,9 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 
 	memcpy(&info, skb->cb, sizeof(info));
 
+	if (WARN_ON_ONCE(skb->len > IEEE80211_MAX_DATA_LEN + hdrlen))
+		return -1;
+
 	if (WARN_ON_ONCE(info.flags & IEEE80211_TX_CTL_AMPDU))
 		return -1;
 
Signature · 1 line
2.20.1
author maintainer
#183 lore raw

PATCH 182/371 bpf: fix uapi bpf_prog_info fields alignment

In reply to Greg Kroah-Hartman

[ Upstream commit 0472301a28f6cf53a6bc5783e48a2d0bbff4682f ]

Merge commit 1c8c5a9d38f60 ("Merge
git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next") undid the fix from commit 36f9814a494 ("bpf: fix uapi hole for 32 bit compat applications") by taking the gpl_compatible 1-bit field definition from commit b85fab0e67b162 ("bpf: Add gpl_compatible flag to struct bpf_prog_info") as is. That breaks architectures with 16-bit alignment like m68k. Add 31-bit pad after gpl_compatible to restore alignment of following fields.

Thanks to Dmitry V. Levin his analysis of this bug history.

---
 include/uapi/linux/bpf.h       | 1 +
 tools/include/uapi/linux/bpf.h | 1 +
 2 files changed, 2 insertions(+)
include/uapi/linux/bpf.h +1 -0
index 9d01f4788d3e..9ae3f28ca469 100644
@@ -2871,6 +2871,7 @@ struct bpf_prog_info {
 	char name[BPF_OBJ_NAME_LEN];
 	__u32 ifindex;
 	__u32 gpl_compatible:1;
+	__u32 :31; /* alignment pad */
 	__u64 netns_dev;
 	__u64 netns_ino;
 	__u32 nr_jited_ksyms;
tools/include/uapi/linux/bpf.h +1 -0
index 929c8e537a14..f6ce794c0f36 100644
@@ -2869,6 +2869,7 @@ struct bpf_prog_info {
 	char name[BPF_OBJ_NAME_LEN];
 	__u32 ifindex;
 	__u32 gpl_compatible:1;
+	__u32 :31; /* alignment pad */
 	__u64 netns_dev;
 	__u64 netns_ino;
 	__u32 nr_jited_ksyms;
Signature · 1 line
2.20.1
author maintainer
#184 lore raw

PATCH 183/371 netfilter: Fix remainder of pseudo-header protocol 0

In reply to Greg Kroah-Hartman

[ Upstream commit 5d1549847c76b1ffcf8e388ef4d0f229bdd1d7e8 ]

Since v5.1-rc1, some types of packets do not get unreachable reply with the following iptables setting. Fox example,

$ iptables -A INPUT -p icmp --icmp-type 8 -j REJECT
$ ping 127.0.0.1 -c 1
PING 127.0.0.1 (127.0.0.1) 56(84) bytes of data.
— 127.0.0.1 ping statistics —
1 packets transmitted, 0 received, 100% packet loss, time 0ms

We should have got the following reply from command line, but we did not.

Quoted 1 line

From 127.0.0.1 icmp_seq=1 Destination Port Unreachable

Yi Zhao reported it and narrowed it down to:
7fc38225363d ("netfilter: reject: skip csum verification for protocols that don't support it"),

This is because nf_ip_checksum still expects pseudo-header protocol type 0 for packets that are of neither TCP or UDP, and thus ICMP packets are mistakenly treated as TCP/UDP.

This patch corrects the conditions in nf_ip_checksum and all other places that still call it with protocol 0.

Fixes 7fc38225363d ("netfilter: reject: skip csum verification for protocols that don't support it")
---
 net/netfilter/nf_conntrack_proto_icmp.c | 2 +-
 net/netfilter/nf_nat_proto.c            | 2 +-
 net/netfilter/utils.c                   | 5 +++--
 3 files changed, 5 insertions(+), 4 deletions(-)
net/netfilter/nf_conntrack_proto_icmp.c +1 -1
index 9becac953587..71a84a0517f3 100644
@@ -221,7 +221,7 @@ int nf_conntrack_icmpv4_error(struct nf_conn *tmpl,
 	/* See ip_conntrack_proto_tcp.c */
 	if (state->net->ct.sysctl_checksum &&
 	    state->hook == NF_INET_PRE_ROUTING &&
-	    nf_ip_checksum(skb, state->hook, dataoff, 0)) {
+	    nf_ip_checksum(skb, state->hook, dataoff, IPPROTO_ICMP)) {
 		icmp_error_log(skb, state, "bad hw icmp checksum");
 		return -NF_ACCEPT;
 	}
net/netfilter/nf_nat_proto.c +1 -1
index 62743da3004f..0b0efbb953bf 100644
@@ -567,7 +567,7 @@ int nf_nat_icmp_reply_translation(struct sk_buff *skb,
 
 	if (!skb_make_writable(skb, hdrlen + sizeof(*inside)))
 		return 0;
-	if (nf_ip_checksum(skb, hooknum, hdrlen, 0))
+	if (nf_ip_checksum(skb, hooknum, hdrlen, IPPROTO_ICMP))
 		return 0;
 
 	inside = (void *)skb->data + hdrlen;
net/netfilter/utils.c +3 -2
index 06dc55590441..51b454d8fa9c 100644
@@ -17,7 +17,8 @@ __sum16 nf_ip_checksum(struct sk_buff *skb, unsigned int hook,
 	case CHECKSUM_COMPLETE:
 		if (hook != NF_INET_PRE_ROUTING && hook != NF_INET_LOCAL_IN)
 			break;
-		if ((protocol == 0 && !csum_fold(skb->csum)) ||
+		if ((protocol != IPPROTO_TCP && protocol != IPPROTO_UDP &&
+		    !csum_fold(skb->csum)) ||
 		    !csum_tcpudp_magic(iph->saddr, iph->daddr,
 				       skb->len - dataoff, protocol,
 				       skb->csum)) {
@@ -26,7 +27,7 @@ __sum16 nf_ip_checksum(struct sk_buff *skb, unsigned int hook,
 		}
 		/* fall through */
 	case CHECKSUM_NONE:
-		if (protocol == 0)
+		if (protocol != IPPROTO_TCP && protocol != IPPROTO_UDP)
 			skb->csum = 0;
 		else
 			skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr,
Signature · 1 line
2.20.1
author maintainer
#185 lore raw

PATCH 184/371 iwlwifi: dbg: fix debug monitor stop and restart delays

In reply to Greg Kroah-Hartman

[ Upstream commit fc838c775f35e272e5cc7ef43853f0b55babbe37 ]

The driver should delay only in recording stop flow between writing to DBGC_IN_SAMPLE register and DBGC_OUT_CTRL register. Any other delay is not needed.

Change the following:
1. Remove any unnecessary delays in the flow
2. Increase the delay in the stop recording flow since 100 micro is not enough
3. Use usleep_range instead of delay since the driver is allowed to sleep in this flow.

Fixes 5cfe79c8d92a ("iwlwifi: fw: stop and start debugging using host command")
---
 drivers/net/wireless/intel/iwlwifi/fw/dbg.c | 2 --
 drivers/net/wireless/intel/iwlwifi/fw/dbg.h | 6 ++++--
 2 files changed, 4 insertions(+), 4 deletions(-)
drivers/net/wireless/intel/iwlwifi/fw/dbg.c +0 -2
index d7380016f1c0..c30f626b1602 100644
@@ -2146,8 +2146,6 @@ void iwl_fw_dbg_collect_sync(struct iwl_fw_runtime *fwrt)
 	/* start recording again if the firmware is not crashed */
 	if (!test_bit(STATUS_FW_ERROR, &fwrt->trans->status) &&
 	    fwrt->fw->dbg.dest_tlv) {
-		/* wait before we collect the data till the DBGC stop */
-		udelay(500);
 		iwl_fw_dbg_restart_recording(fwrt, &params);
 	}
 }
drivers/net/wireless/intel/iwlwifi/fw/dbg.h +4 -2
index a199056234d3..97fcd57e17d8 100644
@@ -297,7 +297,10 @@ _iwl_fw_dbg_stop_recording(struct iwl_trans *trans,
 	}
 
 	iwl_write_umac_prph(trans, DBGC_IN_SAMPLE, 0);
-	udelay(100);
+	/* wait for the DBGC to finish writing the internal buffer to DRAM to
+	 * avoid halting the HW while writing
+	 */
+	usleep_range(700, 1000);
 	iwl_write_umac_prph(trans, DBGC_OUT_CTRL, 0);
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 	trans->dbg_rec_on = false;
@@ -327,7 +330,6 @@ _iwl_fw_dbg_restart_recording(struct iwl_trans *trans,
 		iwl_set_bits_prph(trans, MON_BUFF_SAMPLE_CTL, 0x1);
 	} else {
 		iwl_write_umac_prph(trans, DBGC_IN_SAMPLE, params->in_sample);
-		udelay(100);
 		iwl_write_umac_prph(trans, DBGC_OUT_CTRL, params->out_ctrl);
 	}
 }
Signature · 1 line
2.20.1
author maintainer
#186 lore raw

PATCH 185/371 bnxt_en: Disable bus master during PCI shutdown and driver unload.

In reply to Greg Kroah-Hartman

[ Upstream commit c20dc142dd7b2884b8570eeab323bcd4a84294fa ]

Some chips with older firmware can continue to perform DMA read from context memory even after the memory has been freed. In the PCI shutdown method, we need to call pci_disable_device() to shutdown DMA to prevent this DMA before we put the device into D3hot. DMA memory request in D3hot state will generate PCI fatal error. Similarly, in the driver remove method, the context memory should only be freed after DMA has been shutdown for correctness.

Fixes 98f04cf0f1fc ("bnxt_en: Check context memory requirements from firmware.")
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +2 -1
index 30cafe4cdb6e..bf1fd513fa02 100644
@@ -10165,10 +10165,10 @@ static void bnxt_remove_one(struct pci_dev *pdev)
 	bnxt_dcb_free(bp);
 	kfree(bp->edev);
 	bp->edev = NULL;
+	bnxt_cleanup_pci(bp);
 	bnxt_free_ctx_mem(bp);
 	kfree(bp->ctx);
 	bp->ctx = NULL;
-	bnxt_cleanup_pci(bp);
 	bnxt_free_port_stats(bp);
 	free_netdev(dev);
 }
@@ -10730,6 +10730,7 @@ static void bnxt_shutdown(struct pci_dev *pdev)
 
 	if (system_state == SYSTEM_POWER_OFF) {
 		bnxt_clear_int_mode(bp);
+		pci_disable_device(pdev);
 		pci_wake_from_d3(pdev, bp->wol);
 		pci_set_power_state(pdev, PCI_D3hot);
 	}
Signature · 1 line
2.20.1
author maintainer
#187 lore raw

PATCH 186/371 bnxt_en: Fix statistics context reservation logic for RDMA driver.

In reply to Greg Kroah-Hartman

[ Upstream commit d77b1ad8e87dc5a6cd0d9158b097a4817946ca3b ]

The current logic assumes that the RDMA driver uses one statistics context adjacent to the ones used by the network driver. This assumption is not true and the statistics context used by the RDMA driver is tied to its MSIX base vector. This wrong assumption can cause RDMA driver failure after changing ethtool rings on the network side. Fix the statistics reservation logic accordingly.

Fixes 780baad44f0f ("bnxt_en: Reserve 1 stat_ctx for RDMA driver.")
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +11 -6
index bf1fd513fa02..09557bf49bb0 100644
@@ -5481,7 +5481,16 @@ static int bnxt_cp_rings_in_use(struct bnxt *bp)
 
 static int bnxt_get_func_stat_ctxs(struct bnxt *bp)
 {
-	return bp->cp_nr_rings + bnxt_get_ulp_stat_ctxs(bp);
+	int ulp_stat = bnxt_get_ulp_stat_ctxs(bp);
+	int cp = bp->cp_nr_rings;
+
+	if (!ulp_stat)
+		return cp;
+
+	if (bnxt_nq_rings_in_use(bp) > cp + bnxt_get_ulp_msix_num(bp))
+		return bnxt_get_ulp_msix_base(bp) + ulp_stat;
+
+	return cp + ulp_stat;
 }
 
 static bool bnxt_need_reserve_rings(struct bnxt *bp)
@@ -7373,11 +7382,7 @@ unsigned int bnxt_get_avail_cp_rings_for_en(struct bnxt *bp)
 
 unsigned int bnxt_get_avail_stat_ctxs_for_en(struct bnxt *bp)
 {
-	unsigned int stat;
-
-	stat = bnxt_get_max_func_stat_ctxs(bp) - bnxt_get_ulp_stat_ctxs(bp);
-	stat -= bp->cp_nr_rings;
-	return stat;
+	return bnxt_get_max_func_stat_ctxs(bp) - bnxt_get_func_stat_ctxs(bp);
 }
 
 int bnxt_get_avail_msix(struct bnxt *bp, int num)
Signature · 1 line
2.20.1
author maintainer
#188 lore raw

PATCH 187/371 ALSA: hda: Fix a headphone detection issue when using SOF

In reply to Greg Kroah-Hartman

[ Upstream commit 7c2b3629d09ddec810dc4c1d3a6657c32def8f71 ]

To save power, the hda hdmi driver in ASoC invokes snd_hdac_ext_bus_link_put to disable CORB/RIRB buffers DMA if there is no user of bus and invokes snd_hdac_ext_bus_link_get to set up CORB/RIRB buffers when it is used. Unsolicited responses is disabled in snd_hdac_bus_stop_cmd_io called by snd_hdac_ext_bus_link_put , but it is not enabled in snd_hdac_bus_init_cmd_io called by snd_hdac_ext_bus_link_get. So for put-get sequence, Unsolicited responses is disabled and headphone can't be detected by hda codecs.

Now unsolicited responses is only enabled in snd_hdac_bus_reset_link which resets controller. The function is only called for setup of controller. This patch enables Unsolicited responses after RIRB is initialized in snd_hdac_bus_init_cmd_io which works together with snd_hdac_bus_reset_link to set up controller.

Tested legacy hda driver and SOF driver on intel whiskeylake.

---
 sound/hda/hdac_controller.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
sound/hda/hdac_controller.c +2 -3
index b2e9454f5816..6a190f0d2803 100644
@@ -78,6 +78,8 @@ void snd_hdac_bus_init_cmd_io(struct hdac_bus *bus)
 	snd_hdac_chip_writew(bus, RINTCNT, 1);
 	/* enable rirb dma and response irq */
 	snd_hdac_chip_writeb(bus, RIRBCTL, AZX_RBCTL_DMA_EN | AZX_RBCTL_IRQ_EN);
+	/* Accept unsolicited responses */
+	snd_hdac_chip_updatel(bus, GCTL, AZX_GCTL_UNSOL, AZX_GCTL_UNSOL);
 	spin_unlock_irq(&bus->reg_lock);
 }
 EXPORT_SYMBOL_GPL(snd_hdac_bus_init_cmd_io);
@@ -414,9 +416,6 @@ int snd_hdac_bus_reset_link(struct hdac_bus *bus, bool full_reset)
 		return -EBUSY;
 	}
 
-	/* Accept unsolicited responses */
-	snd_hdac_chip_updatel(bus, GCTL, AZX_GCTL_UNSOL, AZX_GCTL_UNSOL);
-
 	/* detect codecs */
 	if (!bus->codec_mask) {
 		bus->codec_mask = snd_hdac_chip_readw(bus, STATESTS);
Signature · 1 line
2.20.1
author maintainer
#189 lore raw

PATCH 188/371 perf stat: Make metric event lookup more robust

In reply to Greg Kroah-Hartman

[ Upstream commit 145c407c808352acd625be793396fd4f33c794f8 ]

After setting up metric groups through the event parser, the metricgroup code looks them up again in the event list.

Make sure we only look up events that haven't been used by some other metric. The data structures currently cannot handle more than one metric per event. This avoids problems with multiple events partially overlapping.

---
 tools/perf/util/stat-shadow.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
tools/perf/util/stat-shadow.c +3 -2
index 83d8094be4fe..e545e2a8ae71 100644
@@ -303,7 +303,7 @@ static struct perf_evsel *perf_stat__find_event(struct perf_evlist *evsel_list,
 	struct perf_evsel *c2;
 
 	evlist__for_each_entry (evsel_list, c2) {
-		if (!strcasecmp(c2->name, name))
+		if (!strcasecmp(c2->name, name) && !c2->collect_stat)
 			return c2;
 	}
 	return NULL;
@@ -342,7 +342,8 @@ void perf_stat__collect_metric_expr(struct perf_evlist *evsel_list)
 			if (leader) {
 				/* Search in group */
 				for_each_group_member (oc, leader) {
-					if (!strcasecmp(oc->name, metric_names[i])) {
+					if (!strcasecmp(oc->name, metric_names[i]) &&
+						!oc->collect_stat) {
 						found = true;
 						break;
 					}
Signature · 1 line
2.20.1
author maintainer
#190 lore raw

PATCH 189/371 perf stat: Fix metrics with --no-merge

In reply to Greg Kroah-Hartman

[ Upstream commit e3a9427323a53ceee540276a74af7706f350d052 ]

Since Fixes: 8c5421c016a4 ("perf pmu: Display pmu name when printing unmerged events in stat") using --no-merge adds the PMU name to the evsel name.

This breaks the metric value lookup because the parser doesn't know about this.

Remove the extra postfixes for the metric evaluation.

Cc Agustin Vega-Frias <[email protected]>
Cc Kan Liang <[email protected]>
Fixes 8c5421c016a4 ("perf pmu: Display pmu name when printing unmerged events in stat")
---
 tools/perf/util/stat-shadow.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)
tools/perf/util/stat-shadow.c +17 -1
index e545e2a8ae71..0ef98e991ade 100644
@@ -723,6 +723,7 @@ static void generic_metric(struct perf_stat_config *config,
 	double ratio;
 	int i;
 	void *ctxp = out->ctx;
+	char *n, *pn;
 
 	expr__ctx_init(&pctx);
 	expr__add_id(&pctx, name, avg);
@@ -742,7 +743,19 @@ static void generic_metric(struct perf_stat_config *config,
 			stats = &v->stats;
 			scale = 1.0;
 		}
-		expr__add_id(&pctx, metric_events[i]->name, avg_stats(stats)*scale);
+
+		n = strdup(metric_events[i]->name);
+		if (!n)
+			return;
+		/*
+		 * This display code with --no-merge adds [cpu] postfixes.
+		 * These are not supported by the parser. Remove everything
+		 * after the space.
+		 */
+		pn = strchr(n, ' ');
+		if (pn)
+			*pn = 0;
+		expr__add_id(&pctx, n, avg_stats(stats)*scale);
 	}
 	if (!metric_events[i]) {
 		const char *p = metric_expr;
@@ -759,6 +772,9 @@ static void generic_metric(struct perf_stat_config *config,
 				     (metric_name ? metric_name : name) : "", 0);
 	} else
 		print_metric(config, ctxp, NULL, NULL, "", 0);
+
+	for (i = 1; i < pctx.num_ids; i++)
+		free((void *)pctx.ids[i].name);
 }
 
 void perf_stat__print_shadow_stats(struct perf_stat_config *config,
Signature · 1 line
2.20.1
author maintainer
#191 lore raw

PATCH 190/371 perf stat: Dont merge events in the same PMU

In reply to Greg Kroah-Hartman

[ Upstream commit 6c5f4e5cb35b4694dc035d91092d23f596ecd06a ]

Event merging is mainly to collapse similar events in lots of different duplicated PMUs.

It can break metric displaying. It's possible for two metrics to have the same event, and when the two events happen in a row the second wouldn't be displayed. This would also not show the second metric.

To avoid this don't merge events in the same PMU. This makes sense, if we have multiple events in the same PMU there is likely some reason for it (e.g. using multiple groups) and we better not merge them.

While in theory it would be possible to construct metrics that have events with the same name in different PMU no current metrics have this problem.

This is the fix for perf stat -M UPI,IPC (needs also another bug fix to completely work)

Cc Kan Liang <[email protected]>
Fixes 430daf2dc7af ("perf stat: Collapse identically named events")
---
 tools/perf/util/stat-display.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
tools/perf/util/stat-display.c +2 -1
index 6d043c78f3c2..9c940242dcbe 100644
@@ -539,7 +539,8 @@ static void collect_all_aliases(struct perf_stat_config *config, struct perf_evs
 		    alias->scale != counter->scale ||
 		    alias->cgrp != counter->cgrp ||
 		    strcmp(alias->unit, counter->unit) ||
-		    perf_evsel__is_clock(alias) != perf_evsel__is_clock(counter))
+		    perf_evsel__is_clock(alias) != perf_evsel__is_clock(counter) ||
+		    !strcmp(alias->pmu_name, counter->pmu_name))
 			break;
 		alias->merged_stat = true;
 		cb(config, alias, data, false);
Signature · 1 line
2.20.1
author maintainer
#192 lore raw

PATCH 191/371 perf stat: Fix group lookup for metric group

In reply to Greg Kroah-Hartman

[ Upstream commit 2f87f33f4226523df9c9cc28f9874ea02fcc3d3f ]

The metric group code tries to find a group it added earlier in the evlist. Fix the lookup to handle groups with partially overlaps correctly. When a sub string match fails and we reset the match, we have to compare the first element again.

I also renamed the find_evsel function to find_evsel_group to make its purpose clearer.

With the earlier changes this fixes:

Before:

% perf stat -M UPI,IPC sleep 1
...

         1,032,922      uops_retired.retire_slots #      1.1 UPI
         1,896,096      inst_retired.any
         1,896,096      inst_retired.any
         1,177,254      cpu_clk_unhalted.thread

After:

% perf stat -M UPI,IPC sleep 1
...

        1,013,193      uops_retired.retire_slots #      1.1 UPI
           932,033      inst_retired.any
           932,033      inst_retired.any          #      0.9 IPC
         1,091,245      cpu_clk_unhalted.thread
Cc Kan Liang <[email protected]>
Fixes b18f3e365019 ("perf stat: Support JSON metrics in perf stat")
---
 tools/perf/util/metricgroup.c | 47 ++++++++++++++++++++++++++---------
 1 file changed, 35 insertions(+), 12 deletions(-)
tools/perf/util/metricgroup.c +35 -12
index b8d864ed4afe..a48895c4324a 100644
@@ -94,26 +94,49 @@ struct egroup {
 	const char *metric_expr;
 };
 
-static struct perf_evsel *find_evsel(struct perf_evlist *perf_evlist,
-				     const char **ids,
-				     int idnum,
-				     struct perf_evsel **metric_events)
+static bool record_evsel(int *ind, struct perf_evsel **start,
+			 int idnum,
+			 struct perf_evsel **metric_events,
+			 struct perf_evsel *ev)
+{
+	metric_events[*ind] = ev;
+	if (*ind == 0)
+		*start = ev;
+	if (++*ind == idnum) {
+		metric_events[*ind] = NULL;
+		return true;
+	}
+	return false;
+}
+
+static struct perf_evsel *find_evsel_group(struct perf_evlist *perf_evlist,
+					   const char **ids,
+					   int idnum,
+					   struct perf_evsel **metric_events)
 {
 	struct perf_evsel *ev, *start = NULL;
 	int ind = 0;
 
 	evlist__for_each_entry (perf_evlist, ev) {
+		if (ev->collect_stat)
+			continue;
 		if (!strcmp(ev->name, ids[ind])) {
-			metric_events[ind] = ev;
-			if (ind == 0)
-				start = ev;
-			if (++ind == idnum) {
-				metric_events[ind] = NULL;
+			if (record_evsel(&ind, &start, idnum,
+					 metric_events, ev))
 				return start;
-			}
 		} else {
+			/*
+			 * We saw some other event that is not
+			 * in our list of events. Discard
+			 * the whole match and start again.
+			 */
 			ind = 0;
 			start = NULL;
+			if (!strcmp(ev->name, ids[ind])) {
+				if (record_evsel(&ind, &start, idnum,
+						 metric_events, ev))
+					return start;
+			}
 		}
 	}
 	/*
@@ -143,8 +166,8 @@ static int metricgroup__setup_events(struct list_head *groups,
 			ret = -ENOMEM;
 			break;
 		}
-		evsel = find_evsel(perf_evlist, eg->ids, eg->idnum,
-				   metric_events);
+		evsel = find_evsel_group(perf_evlist, eg->ids, eg->idnum,
+					 metric_events);
 		if (!evsel) {
 			pr_debug("Cannot resolve %s: %s\n",
 					eg->metric_name, eg->metric_expr);
Signature · 1 line
2.20.1
author maintainer
#193 lore raw

PATCH 192/371 vxlan: do not destroy fdb if register_netdevice() is failed

In reply to Greg Kroah-Hartman

[ Upstream commit 7c31e54aeee517d1318dfc0bde9fa7de75893dc6 ]

__vxlan_dev_create() destroys FDB using specific pointer which indicates a fdb when error occurs.
But that pointer should not be used when register_netdevice() fails because register_netdevice() internally destroys fdb when error occurs.

This patch makes vxlan_fdb_create() to do not link fdb entry to vxlan dev internally.
Instead, a new function vxlan_fdb_insert() is added to link fdb to vxlan dev.

vxlan_fdb_insert() is called after calling register_netdevice(). This routine can avoid situation that ->ndo_uninit() destroys fdb entry in error path of register_netdevice().
Hence, error path of __vxlan_dev_create() routine can have an opportunity to destroy default fdb entry by hand.

Test command

    ip link add bonding_masters type vxlan id 0 group 239.1.1.1 \
	    dev enp0s9 dstport 4789

Splat looks like:

[  213.392816] kasan: GPF could be caused by NULL-ptr deref or user memory access
[  213.401257] general protection fault: 0000 [#1] SMP DEBUG_PAGEALLOC KASAN PTI
[  213.402178] CPU: 0 PID: 1414 Comm: ip Not tainted 5.2.0-rc5+ #256
[  213.402178] RIP: 0010:vxlan_fdb_destroy+0x120/0x220 [vxlan]
[  213.402178] Code: df 48 8b 2b 48 89 fa 48 c1 ea 03 80 3c 02 00 0f 85 06 01 00 00 4c 8b 63 08 48 b8 00 00 00 00 00 fc d
[  213.402178] RSP: 0018:ffff88810cb9f0a0 EFLAGS: 00010202
[  213.402178] RAX: dffffc0000000000 RBX: ffff888101d4a8c8 RCX: 0000000000000000
[  213.402178] RDX: 1bd5a00000000040 RSI: ffff888101d4a8c8 RDI: ffff888101d4a8d0
[  213.402178] RBP: 0000000000000000 R08: fffffbfff22b72d9 R09: 0000000000000000
[  213.402178] R10: 00000000ffffffef R11: 0000000000000000 R12: dead000000000200
[  213.402178] R13: ffff88810cb9f1f8 R14: ffff88810efccda0 R15: ffff88810efccda0
[  213.402178] FS:  00007f7f6621a0c0(0000) GS:ffff88811b000000(0000) knlGS:0000000000000000
[  213.402178] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  213.402178] CR2: 000055746f0807d0 CR3: 00000001123e0000 CR4: 00000000001006f0
[  213.402178] Call Trace:
[  213.402178]  __vxlan_dev_create+0x3a9/0x7d0 [vxlan]
[  213.402178]  ? vxlan_changelink+0x740/0x740 [vxlan]
[  213.402178]  ? rcu_read_unlock+0x60/0x60 [vxlan]
[  213.402178]  ? __kasan_kmalloc.constprop.3+0xa0/0xd0
[  213.402178]  vxlan_newlink+0x8d/0xc0 [vxlan]
[  213.402178]  ? __vxlan_dev_create+0x7d0/0x7d0 [vxlan]
[  213.554119]  ? __netlink_ns_capable+0xc3/0xf0
[  213.554119]  __rtnl_newlink+0xb75/0x1180
[  213.554119]  ? rtnl_link_unregister+0x230/0x230

[ ... ]

Fixes 0241b836732f ("vxlan: fix default fdb entry netlink notify ordering during netdev create")
---
 drivers/net/vxlan.c | 37 +++++++++++++++++++++++++++----------
 1 file changed, 27 insertions(+), 10 deletions(-)
drivers/net/vxlan.c +27 -10
index 38ecb66fb3e9..82c25f07261f 100644
@@ -806,6 +806,14 @@ static struct vxlan_fdb *vxlan_fdb_alloc(struct vxlan_dev *vxlan,
 	return f;
 }
 
+static void vxlan_fdb_insert(struct vxlan_dev *vxlan, const u8 *mac,
+			     __be32 src_vni, struct vxlan_fdb *f)
+{
+	++vxlan->addrcnt;
+	hlist_add_head_rcu(&f->hlist,
+			   vxlan_fdb_head(vxlan, mac, src_vni));
+}
+
 static int vxlan_fdb_create(struct vxlan_dev *vxlan,
 			    const u8 *mac, union vxlan_addr *ip,
 			    __u16 state, __be16 port, __be32 src_vni,
@@ -831,18 +839,13 @@ static int vxlan_fdb_create(struct vxlan_dev *vxlan,
 		return rc;
 	}
 
-	++vxlan->addrcnt;
-	hlist_add_head_rcu(&f->hlist,
-			   vxlan_fdb_head(vxlan, mac, src_vni));
-
 	*fdb = f;
 
 	return 0;
 }
 
-static void vxlan_fdb_free(struct rcu_head *head)
+static void __vxlan_fdb_free(struct vxlan_fdb *f)
 {
-	struct vxlan_fdb *f = container_of(head, struct vxlan_fdb, rcu);
 	struct vxlan_rdst *rd, *nd;
 
 	list_for_each_entry_safe(rd, nd, &f->remotes, list) {
@@ -852,6 +855,13 @@ static void vxlan_fdb_free(struct rcu_head *head)
 	kfree(f);
 }
 
+static void vxlan_fdb_free(struct rcu_head *head)
+{
+	struct vxlan_fdb *f = container_of(head, struct vxlan_fdb, rcu);
+
+	__vxlan_fdb_free(f);
+}
+
 static void vxlan_fdb_destroy(struct vxlan_dev *vxlan, struct vxlan_fdb *f,
 			      bool do_notify, bool swdev_notify)
 {
@@ -979,6 +989,7 @@ static int vxlan_fdb_update_create(struct vxlan_dev *vxlan,
 	if (rc < 0)
 		return rc;
 
+	vxlan_fdb_insert(vxlan, mac, src_vni, f);
 	rc = vxlan_fdb_notify(vxlan, f, first_remote_rtnl(f), RTM_NEWNEIGH,
 			      swdev_notify, extack);
 	if (rc)
@@ -3573,12 +3584,17 @@ static int __vxlan_dev_create(struct net *net, struct net_device *dev,
 	if (err)
 		goto errout;
 
-	/* notify default fdb entry */
 	if (f) {
+		vxlan_fdb_insert(vxlan, all_zeros_mac,
+				 vxlan->default_dst.remote_vni, f);
+
+		/* notify default fdb entry */
 		err = vxlan_fdb_notify(vxlan, f, first_remote_rtnl(f),
 				       RTM_NEWNEIGH, true, extack);
-		if (err)
-			goto errout;
+		if (err) {
+			vxlan_fdb_destroy(vxlan, f, false, false);
+			goto unregister;
+		}
 	}
 
 	list_add(&vxlan->next, &vn->vxlan_list);
@@ -3590,7 +3606,8 @@ static int __vxlan_dev_create(struct net *net, struct net_device *dev,
 	 * destroy the entry by hand here.
 	 */
 	if (f)
-		vxlan_fdb_destroy(vxlan, f, false, false);
+		__vxlan_fdb_free(f);
+unregister:
 	if (unregister)
 		unregister_netdevice(dev);
 	return err;
Signature · 1 line
2.20.1
author maintainer
#194 lore raw

PATCH 193/371 bnx2x: Prevent ptp_task to be rescheduled indefinitely

In reply to Greg Kroah-Hartman

[ Upstream commit 3c91f25c2f72ba6001775a5932857c1d2131c531 ]

Currently bnx2x ptp worker tries to read a register with timestamp information in case of TX packet timestamping and in case it fails, the routine reschedules itself indefinitely. This was reported as a kworker always at 100% of CPU usage, which was narrowed down to be bnx2x ptp_task.

By following the ioctl handler, we could narrow down the problem to an NTP tool (chrony) requesting HW timestamping from bnx2x NIC with RX filter zeroed; this isn't reproducible for example with ptp4l (from linuxptp) since this tool requests a supported RX filter. It seems NIC FW timestamp mechanism cannot work well with RX_FILTER_NONE - driver's PTP filter init routine skips a register write to the adapter if there's not a supported filter request.

This patch addresses the problem of bnx2x ptp thread's everlasting reschedule by retrying the register read 10 times; between the read attempts the thread sleeps for an increasing amount of time starting in 1ms to give FW some time to perform the timestamping. If it still fails after all retries, we bail out in order to prevent an unbound resource consumption from bnx2x.

The patch also adds an ethtool statistic for accounting the skipped TX timestamp packets and it reduces the priority of timestamping error messages to prevent log flooding. The code was tested using both linuxptp and chrony.

---
 .../net/ethernet/broadcom/bnx2x/bnx2x_cmn.c   |  5 ++-
 .../ethernet/broadcom/bnx2x/bnx2x_ethtool.c   |  4 ++-
 .../net/ethernet/broadcom/bnx2x/bnx2x_main.c  | 33 ++++++++++++++-----
 .../net/ethernet/broadcom/bnx2x/bnx2x_stats.h |  3 ++
 4 files changed, 34 insertions(+), 11 deletions(-)
drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +4 -1
index ecb1bd7eb508..78a01880931c 100644
@@ -3858,9 +3858,12 @@ netdev_tx_t bnx2x_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) {
 		if (!(bp->flags & TX_TIMESTAMPING_EN)) {
+			bp->eth_stats.ptp_skip_tx_ts++;
 			BNX2X_ERR("Tx timestamping was not enabled, this packet will not be timestamped\n");
 		} else if (bp->ptp_tx_skb) {
-			BNX2X_ERR("The device supports only a single outstanding packet to timestamp, this packet will not be timestamped\n");
+			bp->eth_stats.ptp_skip_tx_ts++;
+			netdev_err_once(bp->dev,
+					"Device supports only a single outstanding packet to timestamp, this packet won't be timestamped\n");
 		} else {
 			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
 			/* schedule check for Tx timestamp */
drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c +3 -1
index 59f227fcc68b..0e1b884a5344 100644
@@ -182,7 +182,9 @@ static const struct {
 	{ STATS_OFFSET32(driver_filtered_tx_pkt),
 				4, false, "driver_filtered_tx_pkt" },
 	{ STATS_OFFSET32(eee_tx_lpi),
-				4, true, "Tx LPI entry count"}
+				4, true, "Tx LPI entry count"},
+	{ STATS_OFFSET32(ptp_skip_tx_ts),
+				4, false, "ptp_skipped_tx_tstamp" },
 };
 
 #define BNX2X_NUM_STATS		ARRAY_SIZE(bnx2x_stats_arr)
drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +24 -9
index 626b491f7674..7a075f1f1242 100644
@@ -15243,11 +15243,24 @@ static void bnx2x_ptp_task(struct work_struct *work)
 	u32 val_seq;
 	u64 timestamp, ns;
 	struct skb_shared_hwtstamps shhwtstamps;
+	bool bail = true;
+	int i;
+
+	/* FW may take a while to complete timestamping; try a bit and if it's
+	 * still not complete, may indicate an error state - bail out then.
+	 */
+	for (i = 0; i < 10; i++) {
+		/* Read Tx timestamp registers */
+		val_seq = REG_RD(bp, port ? NIG_REG_P1_TLLH_PTP_BUF_SEQID :
+				 NIG_REG_P0_TLLH_PTP_BUF_SEQID);
+		if (val_seq & 0x10000) {
+			bail = false;
+			break;
+		}
+		msleep(1 << i);
+	}
 
-	/* Read Tx timestamp registers */
-	val_seq = REG_RD(bp, port ? NIG_REG_P1_TLLH_PTP_BUF_SEQID :
-			 NIG_REG_P0_TLLH_PTP_BUF_SEQID);
-	if (val_seq & 0x10000) {
+	if (!bail) {
 		/* There is a valid timestamp value */
 		timestamp = REG_RD(bp, port ? NIG_REG_P1_TLLH_PTP_BUF_TS_MSB :
 				   NIG_REG_P0_TLLH_PTP_BUF_TS_MSB);
@@ -15262,16 +15275,18 @@ static void bnx2x_ptp_task(struct work_struct *work)
 		memset(&shhwtstamps, 0, sizeof(shhwtstamps));
 		shhwtstamps.hwtstamp = ns_to_ktime(ns);
 		skb_tstamp_tx(bp->ptp_tx_skb, &shhwtstamps);
-		dev_kfree_skb_any(bp->ptp_tx_skb);
-		bp->ptp_tx_skb = NULL;
 
 		DP(BNX2X_MSG_PTP, "Tx timestamp, timestamp cycles = %llu, ns = %llu\n",
 		   timestamp, ns);
 	} else {
-		DP(BNX2X_MSG_PTP, "There is no valid Tx timestamp yet\n");
-		/* Reschedule to keep checking for a valid timestamp value */
-		schedule_work(&bp->ptp_task);
+		DP(BNX2X_MSG_PTP,
+		   "Tx timestamp is not recorded (register read=%u)\n",
+		   val_seq);
+		bp->eth_stats.ptp_skip_tx_ts++;
 	}
+
+	dev_kfree_skb_any(bp->ptp_tx_skb);
+	bp->ptp_tx_skb = NULL;
 }
 
 void bnx2x_set_rx_ts(struct bnx2x *bp, struct sk_buff *skb)
drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h +3 -0
index b2644ed13d06..d55e63692cf3 100644
@@ -207,6 +207,9 @@ struct bnx2x_eth_stats {
 	u32 driver_filtered_tx_pkt;
 	/* src: Clear-on-Read register; Will not survive PMF Migration */
 	u32 eee_tx_lpi;
+
+	/* PTP */
+	u32 ptp_skip_tx_ts;
 };
 
 struct bnx2x_eth_q_stats {
Signature · 1 line
2.20.1
author maintainer
#195 lore raw

PATCH 194/371 net: usb: asix: init MAC address buffers

In reply to Greg Kroah-Hartman

[ Upstream commit 78226f6eaac80bf30256a33a4926c194ceefdf36 ]

This is for fixing bug KMSAN: uninit-value in ax88772_bind

Tested by
https://groups.google.com/d/msg/syzkaller-bugs/aFQurGotng4/eB_HlNhhCwAJ

Reported-by [email protected]

syzbot found the following crash on:

HEAD commit:    f75e4cfe kmsan: use kmsan_handle_urb() in urb.c
git tree:       kmsan
console output: https://syzkaller.appspot.com/x/log.txt?x=136d720ea00000

kernel config:
https://syzkaller.appspot.com/x/.config?x=602468164ccdc30a
dashboard link:
https://syzkaller.appspot.com/bug?extid=8a3fc6674bbc3978ed4e
compiler: clang version 9.0.0 (/home/glider/llvm/clang
06d00afa61eef8f7f501ebdb4e8612ea43ec2d78)
syz repro:
https://syzkaller.appspot.com/x/repro.syz?x=12788316a00000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=120359aaa00000

==================================================================

BUG: KMSAN: uninit-value in is_valid_ether_addr
include/linux/etherdevice.h:200 [inline]
BUG: KMSAN: uninit-value in asix_set_netdev_dev_addr
drivers/net/usb/asix_devices.c:73 [inline]
BUG: KMSAN: uninit-value in ax88772_bind+0x93d/0x11e0
drivers/net/usb/asix_devices.c:724
CPU: 0 PID: 3348 Comm: kworker/0:2 Not tainted 5.1.0+ #1
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS
Google 01/01/2011
Workqueue: usb_hub_wq hub_event
Call Trace:
  __dump_stack lib/dump_stack.c:77 [inline]
  dump_stack+0x191/0x1f0 lib/dump_stack.c:113
  kmsan_report+0x130/0x2a0 mm/kmsan/kmsan.c:622
  __msan_warning+0x75/0xe0 mm/kmsan/kmsan_instr.c:310
  is_valid_ether_addr include/linux/etherdevice.h:200 [inline]
  asix_set_netdev_dev_addr drivers/net/usb/asix_devices.c:73 [inline]
  ax88772_bind+0x93d/0x11e0 drivers/net/usb/asix_devices.c:724
  usbnet_probe+0x10f5/0x3940 drivers/net/usb/usbnet.c:1728
  usb_probe_interface+0xd66/0x1320 drivers/usb/core/driver.c:361
  really_probe+0xdae/0x1d80 drivers/base/dd.c:513
  driver_probe_device+0x1b3/0x4f0 drivers/base/dd.c:671
  __device_attach_driver+0x5b8/0x790 drivers/base/dd.c:778
  bus_for_each_drv+0x28e/0x3b0 drivers/base/bus.c:454
  __device_attach+0x454/0x730 drivers/base/dd.c:844
  device_initial_probe+0x4a/0x60 drivers/base/dd.c:891
  bus_probe_device+0x137/0x390 drivers/base/bus.c:514
  device_add+0x288d/0x30e0 drivers/base/core.c:2106
  usb_set_configuration+0x30dc/0x3750 drivers/usb/core/message.c:2027
  generic_probe+0xe7/0x280 drivers/usb/core/generic.c:210
  usb_probe_device+0x14c/0x200 drivers/usb/core/driver.c:266
  really_probe+0xdae/0x1d80 drivers/base/dd.c:513
  driver_probe_device+0x1b3/0x4f0 drivers/base/dd.c:671
  __device_attach_driver+0x5b8/0x790 drivers/base/dd.c:778
  bus_for_each_drv+0x28e/0x3b0 drivers/base/bus.c:454
  __device_attach+0x454/0x730 drivers/base/dd.c:844
  device_initial_probe+0x4a/0x60 drivers/base/dd.c:891
  bus_probe_device+0x137/0x390 drivers/base/bus.c:514
  device_add+0x288d/0x30e0 drivers/base/core.c:2106
  usb_new_device+0x23e5/0x2ff0 drivers/usb/core/hub.c:2534
  hub_port_connect drivers/usb/core/hub.c:5089 [inline]
  hub_port_connect_change drivers/usb/core/hub.c:5204 [inline]
  port_event drivers/usb/core/hub.c:5350 [inline]
  hub_event+0x48d1/0x7290 drivers/usb/core/hub.c:5432
  process_one_work+0x1572/0x1f00 kernel/workqueue.c:2269
  process_scheduled_works kernel/workqueue.c:2331 [inline]
  worker_thread+0x189c/0x2460 kernel/workqueue.c:2417
  kthread+0x4b5/0x4f0 kernel/kthread.c:254
  ret_from_fork+0x35/0x40 arch/x86/entry/entry_64.S:355
---
 drivers/net/usb/asix_devices.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/net/usb/asix_devices.c +3 -3
index 3d93993e74da..2eca4168af2f 100644
@@ -238,7 +238,7 @@ static void asix_phy_reset(struct usbnet *dev, unsigned int reset_bits)
 static int ax88172_bind(struct usbnet *dev, struct usb_interface *intf)
 {
 	int ret = 0;
-	u8 buf[ETH_ALEN];
+	u8 buf[ETH_ALEN] = {0};
 	int i;
 	unsigned long gpio_bits = dev->driver_info->data;
 
@@ -689,7 +689,7 @@ static int asix_resume(struct usb_interface *intf)
 static int ax88772_bind(struct usbnet *dev, struct usb_interface *intf)
 {
 	int ret, i;
-	u8 buf[ETH_ALEN], chipcode = 0;
+	u8 buf[ETH_ALEN] = {0}, chipcode = 0;
 	u32 phyid;
 	struct asix_common_private *priv;
 
@@ -1073,7 +1073,7 @@ static const struct net_device_ops ax88178_netdev_ops = {
 static int ax88178_bind(struct usbnet *dev, struct usb_interface *intf)
 {
 	int ret;
-	u8 buf[ETH_ALEN];
+	u8 buf[ETH_ALEN] = {0};
 
 	usbnet_get_endpoints(dev,intf);
 
Signature · 1 line
2.20.1
author maintainer
#196 lore raw

PATCH 195/371 rxrpc: Fix oops in tracepoint

In reply to Greg Kroah-Hartman

[ Upstream commit 99f0eae653b2db64917d0b58099eb51e300b311d ]

If the rxrpc_eproto tracepoint is enabled, an oops will be cause by the trace line that rxrpc_extract_header() tries to emit when a protocol error occurs (typically because the packet is short) because the call argument is NULL.

Fix this by using ?: to assume 0 as the debug_id if call is NULL.

This can then be induced by:

	echo -e '\0\0\0\0\0\0\0\0' | ncat -4u --send-only <addr> 20001

where addr has the following program running on it:

	#include <stdio.h>
	#include <stdlib.h>
	#include <string.h>
	#include <unistd.h>
	#include <sys/socket.h>
	#include <arpa/inet.h>
	#include <linux/rxrpc.h>
	int main(void)
	{
		struct sockaddr_rxrpc srx;
		int fd;
		memset(&srx, 0, sizeof(srx));
		srx.srx_family			= AF_RXRPC;
		srx.srx_service			= 0;
		srx.transport_type		= AF_INET;
		srx.transport_len		= sizeof(srx.transport.sin);
		srx.transport.sin.sin_family	= AF_INET;
		srx.transport.sin.sin_port	= htons(0x4e21);
		fd = socket(AF_RXRPC, SOCK_DGRAM, AF_INET6);
		bind(fd, (struct sockaddr *)&srx, sizeof(srx));
		sleep(20);
		return 0;
	}

It results in the following oops.

	BUG: kernel NULL pointer dereference, address: 0000000000000340
	#PF: supervisor read access in kernel mode
	#PF: error_code(0x0000) - not-present page
	...
	RIP: 0010:trace_event_raw_event_rxrpc_rx_eproto+0x47/0xac
	...
	Call Trace:
	 <IRQ>
	 rxrpc_extract_header+0x86/0x171
	 ? rcu_read_lock_sched_held+0x5d/0x63
	 ? rxrpc_new_skb+0xd4/0x109
	 rxrpc_input_packet+0xef/0x14fc
	 ? rxrpc_input_data+0x986/0x986
	 udp_queue_rcv_one_skb+0xbf/0x3d0
	 udp_unicast_rcv_skb.isra.8+0x64/0x71
	 ip_protocol_deliver_rcu+0xe4/0x1b4
	 ip_local_deliver+0xf0/0x154
	 __netif_receive_skb_one_core+0x50/0x6c
	 netif_receive_skb_internal+0x26b/0x2e9
	 napi_gro_receive+0xf8/0x1da
	 rtl8169_poll+0x303/0x4c4
	 net_rx_action+0x10e/0x333
	 __do_softirq+0x1a5/0x38f
	 irq_exit+0x54/0xc4
	 do_IRQ+0xda/0xf8
	 common_interrupt+0xf/0xf
	 </IRQ>
	 ...
	 ? cpuidle_enter_state+0x23c/0x34d
	 cpuidle_enter+0x2a/0x36
	 do_idle+0x163/0x1ea
	 cpu_startup_entry+0x1d/0x1f
	 start_secondary+0x157/0x172
	 secondary_startup_64+0xa4/0xb0
Fixes a25e21f0bcd2 ("rxrpc, afs: Use debug_ids rather than pointers in traces")
---
 include/trace/events/rxrpc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/trace/events/rxrpc.h +1 -1
index 7b60fd186cfe..77bc53ce419f 100644
@@ -1383,7 +1383,7 @@ TRACE_EVENT(rxrpc_rx_eproto,
 			     ),
 
 	    TP_fast_assign(
-		    __entry->call = call->debug_id;
+		    __entry->call = call ? call->debug_id : 0;
 		    __entry->serial = serial;
 		    __entry->why = why;
 			   ),
Signature · 1 line
2.20.1
author maintainer
#197 lore raw

PATCH 196/371 libbpf: fix GCC8 warning for strncpy

In reply to Greg Kroah-Hartman

[ Upstream commit cdfc7f888c2a355b01308e97c6df108f1c2b64e8 ]

GCC8 started emitting warning about using strncpy with number of bytes exactly equal destination size, which is generally unsafe, as can lead to non-zero terminated string being copied. Use IFNAMSIZ - 1 as number of bytes to ensure name is always zero-terminated.

---
 tools/lib/bpf/xsk.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
tools/lib/bpf/xsk.c +2 -1
index af5f310ecca1..1fe0e1eec738 100644
@@ -336,7 +336,8 @@ static int xsk_get_max_queues(struct xsk_socket *xsk)
 
 	channels.cmd = ETHTOOL_GCHANNELS;
 	ifr.ifr_data = (void *)&channels;
-	strncpy(ifr.ifr_name, xsk->ifname, IFNAMSIZ);
+	strncpy(ifr.ifr_name, xsk->ifname, IFNAMSIZ - 1);
+	ifr.ifr_name[IFNAMSIZ - 1] = '\0';
 	err = ioctl(fd, SIOCETHTOOL, &ifr);
 	if (err && errno != EOPNOTSUPP) {
 		ret = -errno;
Signature · 1 line
2.20.1
author maintainer
#198 lore raw

PATCH 197/371 bpf, libbpf, smatch: Fix potential NULL pointer dereference

In reply to Greg Kroah-Hartman

[ Upstream commit 33bae185f74d49a0d7b1bfaafb8e959efce0f243 ]

Based on the following report from Smatch, fix the potential NULL pointer dereference check:

tools/lib/bpf/libbpf.c:3493
bpf_prog_load_xattr() warn: variable dereferenced before check 'attr' (see line 3483)

3479 int bpf_prog_load_xattr(const struct bpf_prog_load_attr *attr,
3480 struct bpf_object **pobj, int *prog_fd)
3481 {
3482 struct bpf_object_open_attr open_attr = {

  3483                 .file           = attr->file,
  3484                 .prog_type      = attr->prog_type,
                                         ^^^^^^

3485 };

At the head of function, it directly access 'attr' without checking if it's NULL pointer. This patch moves the values assignment after validating 'attr' and 'attr->file'.

---
 tools/lib/bpf/libbpf.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
tools/lib/bpf/libbpf.c +4 -4
index 11c25d9ea431..43dc8a8e9105 100644
@@ -2897,10 +2897,7 @@ int bpf_prog_load(const char *file, enum bpf_prog_type type,
 int bpf_prog_load_xattr(const struct bpf_prog_load_attr *attr,
 			struct bpf_object **pobj, int *prog_fd)
 {
-	struct bpf_object_open_attr open_attr = {
-		.file		= attr->file,
-		.prog_type	= attr->prog_type,
-	};
+	struct bpf_object_open_attr open_attr = {};
 	struct bpf_program *prog, *first_prog = NULL;
 	enum bpf_attach_type expected_attach_type;
 	enum bpf_prog_type prog_type;
@@ -2913,6 +2910,9 @@ int bpf_prog_load_xattr(const struct bpf_prog_load_attr *attr,
 	if (!attr->file)
 		return -EINVAL;
 
+	open_attr.file = attr->file;
+	open_attr.prog_type = attr->prog_type;
+
 	obj = bpf_object__open_xattr(&open_attr);
 	if (IS_ERR_OR_NULL(obj))
 		return -ENOENT;
Signature · 1 line
2.20.1
author maintainer
#199 lore raw

PATCH 198/371 selftests: bpf: fix inlines in test_lwt_seg6local

In reply to Greg Kroah-Hartman

[ Upstream commit 11aca65ec4db09527d3e9b6b41a0615b7da4386b ]

Selftests are reporting this failure in test_lwt_seg6local.sh:

+ ip netns exec ns2 ip -6 route add fb00::6 encap bpf in obj test_lwt_seg6local.o sec encap_srh dev veth2 Error fetching program/map!
Failed to parse eBPF program: Operation not permitted

The problem is __attribute__((always_inline)) alone is not enough to prevent clang from inserting those functions in .text. In that case, .text is not marked as relocateable.

See the output of objdump -h test_lwt_seg6local.o:

Idx Name          Size      VMA               LMA               File off  Algn
  0 .text         00003530  0000000000000000  0000000000000000  00000040  2**3

CONTENTS, ALLOC, LOAD, READONLY, CODE

This causes the iproute bpf loader to fail in bpf_fetch_prog_sec:
bpf_has_call_data returns true but bpf_fetch_prog_relo fails as there's no relocateable .text section in the file.

To fix this, convert to 'static __always_inline'.

v2: Use 'static __always_inline' instead of 'static inline

    __attribute__((always_inline))'
Fixes c99a84eac026 ("selftests/bpf: test for seg6local End.BPF action")
---
 .../testing/selftests/bpf/progs/test_lwt_seg6local.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
tools/testing/selftests/bpf/progs/test_lwt_seg6local.c +6 -6
index 0575751bc1bc..e2f6ed0a583d 100644
@@ -61,7 +61,7 @@ struct sr6_tlv_t {
 	unsigned char value[0];
 } BPF_PACKET_HEADER;
 
-__attribute__((always_inline)) struct ip6_srh_t *get_srh(struct __sk_buff *skb)
+static __always_inline struct ip6_srh_t *get_srh(struct __sk_buff *skb)
 {
 	void *cursor, *data_end;
 	struct ip6_srh_t *srh;
@@ -95,7 +95,7 @@ __attribute__((always_inline)) struct ip6_srh_t *get_srh(struct __sk_buff *skb)
 	return srh;
 }
 
-__attribute__((always_inline))
+static __always_inline
 int update_tlv_pad(struct __sk_buff *skb, uint32_t new_pad,
 		   uint32_t old_pad, uint32_t pad_off)
 {
@@ -125,7 +125,7 @@ int update_tlv_pad(struct __sk_buff *skb, uint32_t new_pad,
 	return 0;
 }
 
-__attribute__((always_inline))
+static __always_inline
 int is_valid_tlv_boundary(struct __sk_buff *skb, struct ip6_srh_t *srh,
 			  uint32_t *tlv_off, uint32_t *pad_size,
 			  uint32_t *pad_off)
@@ -184,7 +184,7 @@ int is_valid_tlv_boundary(struct __sk_buff *skb, struct ip6_srh_t *srh,
 	return 0;
 }
 
-__attribute__((always_inline))
+static __always_inline
 int add_tlv(struct __sk_buff *skb, struct ip6_srh_t *srh, uint32_t tlv_off,
 	    struct sr6_tlv_t *itlv, uint8_t tlv_size)
 {
@@ -228,7 +228,7 @@ int add_tlv(struct __sk_buff *skb, struct ip6_srh_t *srh, uint32_t tlv_off,
 	return update_tlv_pad(skb, new_pad, pad_size, pad_off);
 }
 
-__attribute__((always_inline))
+static __always_inline
 int delete_tlv(struct __sk_buff *skb, struct ip6_srh_t *srh,
 	       uint32_t tlv_off)
 {
@@ -266,7 +266,7 @@ int delete_tlv(struct __sk_buff *skb, struct ip6_srh_t *srh,
 	return update_tlv_pad(skb, new_pad, pad_size, pad_off);
 }
 
-__attribute__((always_inline))
+static __always_inline
 int has_egr_tlv(struct __sk_buff *skb, struct ip6_srh_t *srh)
 {
 	int tlv_offset = sizeof(struct ip6_t) + sizeof(struct ip6_srh_t) +
Signature · 1 line
2.20.1
author maintainer
#200 lore raw

PATCH 199/371 bonding: validate ip header before check IPPROTO_IGMP

In reply to Greg Kroah-Hartman

[ Upstream commit 9d1bc24b52fb8c5d859f9a47084bf1179470e04c ]

bond_xmit_roundrobin() checks for IGMP packets but it parses the IP header even before checking skb->protocol.

We should validate the IP header with pskb_may_pull() before using iph->protocol.

Reported-and-tested-by [email protected]
Fixes a2fd940f4cff ("bonding: fix broken multicast with round-robin mode")
Cc Jay Vosburgh <[email protected]>
Cc Veaceslav Falico <[email protected]>
Cc Andy Gospodarek <[email protected]>
---
 drivers/net/bonding/bond_main.c | 37 ++++++++++++++++++++-------------
 1 file changed, 23 insertions(+), 14 deletions(-)
drivers/net/bonding/bond_main.c +23 -14
index 59e919b92873..7b9a18e36a93 100644
@@ -3866,8 +3866,8 @@ static netdev_tx_t bond_xmit_roundrobin(struct sk_buff *skb,
 					struct net_device *bond_dev)
 {
 	struct bonding *bond = netdev_priv(bond_dev);
-	struct iphdr *iph = ip_hdr(skb);
 	struct slave *slave;
+	int slave_cnt;
 	u32 slave_id;
 
 	/* Start with the curr_active_slave that joined the bond as the
@@ -3876,23 +3876,32 @@ static netdev_tx_t bond_xmit_roundrobin(struct sk_buff *skb,
 	 * send the join/membership reports.  The curr_active_slave found
 	 * will send all of this type of traffic.
 	 */
-	if (iph->protocol == IPPROTO_IGMP && skb->protocol == htons(ETH_P_IP)) {
-		slave = rcu_dereference(bond->curr_active_slave);
-		if (slave)
-			bond_dev_queue_xmit(bond, skb, slave->dev);
-		else
-			bond_xmit_slave_id(bond, skb, 0);
-	} else {
-		int slave_cnt = READ_ONCE(bond->slave_cnt);
+	if (skb->protocol == htons(ETH_P_IP)) {
+		int noff = skb_network_offset(skb);
+		struct iphdr *iph;
 
-		if (likely(slave_cnt)) {
-			slave_id = bond_rr_gen_slave_id(bond);
-			bond_xmit_slave_id(bond, skb, slave_id % slave_cnt);
-		} else {
-			bond_tx_drop(bond_dev, skb);
+		if (unlikely(!pskb_may_pull(skb, noff + sizeof(*iph))))
+			goto non_igmp;
+
+		iph = ip_hdr(skb);
+		if (iph->protocol == IPPROTO_IGMP) {
+			slave = rcu_dereference(bond->curr_active_slave);
+			if (slave)
+				bond_dev_queue_xmit(bond, skb, slave->dev);
+			else
+				bond_xmit_slave_id(bond, skb, 0);
+			return NETDEV_TX_OK;
 		}
 	}
 
+non_igmp:
+	slave_cnt = READ_ONCE(bond->slave_cnt);
+	if (likely(slave_cnt)) {
+		slave_id = bond_rr_gen_slave_id(bond);
+		bond_xmit_slave_id(bond, skb, slave_id % slave_cnt);
+	} else {
+		bond_tx_drop(bond_dev, skb);
+	}
 	return NETDEV_TX_OK;
 }
 
Signature · 1 line
2.20.1
author maintainer
#201 lore raw

PATCH 200/371 gpiolib: Fix references to gpiod_[gs]et_*value_cansleep() variants

In reply to Greg Kroah-Hartman

[ Upstream commit 3285170f28a850638794cdfe712eb6d93e51e706 ]

Commit 372e722ea4dd4ca1 ("gpiolib: use descriptors internally") renamed the functions to use a "gpiod" prefix, and commit 79a9becda8940deb ("gpiolib: export descriptor-based GPIO interface") introduced the "raw" variants, but both changes forgot to update the comments.

Readd a similar reference to gpiod_set_value(), which was accidentally removed by commit 1e77fc82110ac36f ("gpio: Add missing open drain/source handling to gpiod_set_value_cansleep()").

---
 drivers/gpio/gpiolib.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/gpio/gpiolib.c +4 -3
index fd1344056e1d..b8a5c1e3b99d 100644
@@ -3012,7 +3012,7 @@ int gpiod_get_array_value_complex(bool raw, bool can_sleep,
 int gpiod_get_raw_value(const struct gpio_desc *desc)
 {
 	VALIDATE_DESC(desc);
-	/* Should be using gpio_get_value_cansleep() */
+	/* Should be using gpiod_get_raw_value_cansleep() */
 	WARN_ON(desc->gdev->chip->can_sleep);
 	return gpiod_get_raw_value_commit(desc);
 }
@@ -3033,7 +3033,7 @@ int gpiod_get_value(const struct gpio_desc *desc)
 	int value;
 
 	VALIDATE_DESC(desc);
-	/* Should be using gpio_get_value_cansleep() */
+	/* Should be using gpiod_get_value_cansleep() */
 	WARN_ON(desc->gdev->chip->can_sleep);
 
 	value = gpiod_get_raw_value_commit(desc);
@@ -3304,7 +3304,7 @@ int gpiod_set_array_value_complex(bool raw, bool can_sleep,
 void gpiod_set_raw_value(struct gpio_desc *desc, int value)
 {
 	VALIDATE_DESC_VOID(desc);
-	/* Should be using gpiod_set_value_cansleep() */
+	/* Should be using gpiod_set_raw_value_cansleep() */
 	WARN_ON(desc->gdev->chip->can_sleep);
 	gpiod_set_raw_value_commit(desc, value);
 }
@@ -3345,6 +3345,7 @@ static void gpiod_set_value_nocheck(struct gpio_desc *desc, int value)
 void gpiod_set_value(struct gpio_desc *desc, int value)
 {
 	VALIDATE_DESC_VOID(desc);
+	/* Should be using gpiod_set_value_cansleep() */
 	WARN_ON(desc->gdev->chip->can_sleep);
 	gpiod_set_value_nocheck(desc, value);
 }
Signature · 1 line
2.20.1
author maintainer
#202 lore raw

PATCH 201/371 ASoC: audio-graph-card: fix use-after-free in graph_for_each_link

In reply to Greg Kroah-Hartman

[ Upstream commit 1bcc1fd64e4dd903f4d868a9e053986e3b102713 ]

After calling of_node_put() on the codec_ep and codec_port variables, they are still being used, which may result in use-after-free. We fix this issue by calling of_node_put() after the last usage.

Fixes fce9b90c1ab7 ("ASoC: audio-graph-card: cleanup DAI link loop method - step2")
Cc Liam Girdwood <[email protected]>
Cc Mark Brown <[email protected]>
Cc Jaroslav Kysela <[email protected]>
Cc Takashi Iwai <[email protected]>
Cc Kuninori Morimoto <[email protected]>
---
 sound/soc/generic/audio-graph-card.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
sound/soc/generic/audio-graph-card.c +3 -3
index 69bc4848d787..f730830fb36c 100644
@@ -460,9 +460,6 @@ static int graph_for_each_link(struct graph_priv *priv,
 			codec_ep = of_graph_get_remote_endpoint(cpu_ep);
 			codec_port = of_get_parent(codec_ep);
 
-			of_node_put(codec_ep);
-			of_node_put(codec_port);
-
 			/* get convert-xxx property */
 			memset(&adata, 0, sizeof(adata));
 			graph_get_conversion(dev, codec_ep, &adata);
@@ -482,6 +479,9 @@ static int graph_for_each_link(struct graph_priv *priv,
 			else
 				ret = func_noml(priv, cpu_ep, codec_ep, li);
 
+			of_node_put(codec_ep);
+			of_node_put(codec_port);
+
 			if (ret < 0)
 				return ret;
 
Signature · 1 line
2.20.1
author maintainer
#203 lore raw

PATCH 202/371 tools: bpftool: Fix json dump crash on powerpc

In reply to Greg Kroah-Hartman

[ Upstream commit aa52bcbe0e72fac36b1862db08b9c09c4caefae3 ]

Michael reported crash with by bpf program in json mode on powerpc:

  # bpftool prog -p dump jited id 14
  [{
        "name": "0xd00000000a9aa760",
        "insns": [{
                "pc": "0x0",
                "operation": "nop",
                "operands": [null
                ]
            },{
                "pc": "0x4",
                "operation": "nop",
                "operands": [null
                ]
            },{
                "pc": "0x8",
                "operation": "mflr",
  Segmentation fault (core dumped)

The code is assuming char pointers in format, which is not always true at least for powerpc. Fixing this by dumping the whole string into buffer based on its format.

Please note that libopcodes code does not check return values from fprintf callback, but as per Jakub suggestion returning -1 on allocation failure so we do the best effort to propagate the error.

Fixes 107f041212c1 ("tools: bpftool: add JSON output for `bpftool prog dump jited *` command")
---
 tools/bpf/bpftool/jit_disasm.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)
tools/bpf/bpftool/jit_disasm.c +7 -4
index 3ef3093560ba..bfed711258ce 100644
@@ -11,6 +11,8 @@
  * Licensed under the GNU General Public License, version 2.0 (GPLv2)
  */
 
+#define _GNU_SOURCE
+#include <stdio.h>
 #include <stdarg.h>
 #include <stdint.h>
 #include <stdio.h>
@@ -44,11 +46,13 @@ static int fprintf_json(void *out, const char *fmt, ...)
 	char *s;
 
 	va_start(ap, fmt);
+	if (vasprintf(&s, fmt, ap) < 0)
+		return -1;
+	va_end(ap);
+
 	if (!oper_count) {
 		int i;
 
-		s = va_arg(ap, char *);
-
 		/* Strip trailing spaces */
 		i = strlen(s) - 1;
 		while (s[i] == ' ')
@@ -61,11 +65,10 @@ static int fprintf_json(void *out, const char *fmt, ...)
 	} else if (!strcmp(fmt, ",")) {
 		   /* Skip */
 	} else {
-		s = va_arg(ap, char *);
 		jsonw_string(json_wtr, s);
 		oper_count++;
 	}
-	va_end(ap);
+	free(s);
 	return 0;
 }
 
Signature · 1 line
2.20.1
author maintainer
#204 lore raw

PATCH 203/371 net: hns3: enable broadcast promisc mode when initializing VF

In reply to Greg Kroah-Hartman

[ Upstream commit 2d5066fc175ea77a733d84df9ef414b34f311641 ]

For revision 0x20, the broadcast promisc is enabled by firmware, it's unnecessary to enable it when initializing VF.

For revision 0x21, it's necessary to enable broadcast promisc mode when initializing or re-initializing VF, otherwise, it will be unable to send and receive promisc packets.

Fixes f01f5559cac8 ("net: hns3: don't allow vf to enable promisc mode")
---
 .../ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c  | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +11 -3
index 8dd7fef863f6..d7a15d5b6b61 100644
@@ -2425,6 +2425,12 @@ static int hclgevf_reset_hdev(struct hclgevf_dev *hdev)
 		return ret;
 	}
 
+	if (pdev->revision >= 0x21) {
+		ret = hclgevf_set_promisc_mode(hdev, true);
+		if (ret)
+			return ret;
+	}
+
 	dev_info(&hdev->pdev->dev, "Reset done\n");
 
 	return 0;
@@ -2504,9 +2510,11 @@ static int hclgevf_init_hdev(struct hclgevf_dev *hdev)
 	 * firmware makes sure broadcast packets can be accepted.
 	 * For revision 0x21, default to enable broadcast promisc mode.
 	 */
-	ret = hclgevf_set_promisc_mode(hdev, true);
-	if (ret)
-		goto err_config;
+	if (pdev->revision >= 0x21) {
+		ret = hclgevf_set_promisc_mode(hdev, true);
+		if (ret)
+			goto err_config;
+	}
 
 	/* Initialize RSS for this VF */
 	ret = hclgevf_rss_init_hw(hdev);
Signature · 1 line
2.20.1
author maintainer
#205 lore raw

PATCH 204/371 Bluetooth: hci_bcsp: Fix memory leak in rx_skb

In reply to Greg Kroah-Hartman

[ Upstream commit 4ce9146e0370fcd573f0372d9b4e5a211112567c ]

Syzkaller found that it is possible to provoke a memory leak by never freeing rx_skb in struct bcsp_struct.

Fix by freeing in bcsp_close()

---
 drivers/bluetooth/hci_bcsp.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/bluetooth/hci_bcsp.c +5 -0
index 1a7f0c82fb36..66fe1e6dc631 100644
@@ -759,6 +759,11 @@ static int bcsp_close(struct hci_uart *hu)
 	skb_queue_purge(&bcsp->rel);
 	skb_queue_purge(&bcsp->unrel);
 
+	if (bcsp->rx_skb) {
+		kfree_skb(bcsp->rx_skb);
+		bcsp->rx_skb = NULL;
+	}
+
 	kfree(bcsp);
 	return 0;
 }
Signature · 1 line
2.20.1
author maintainer
#206 lore raw

PATCH 205/371 Bluetooth: Add new 13d3:3491 QCA_ROME device

In reply to Greg Kroah-Hartman

[ Upstream commit 44d34af2e4cfd0c5357182f8b43f3e0a1fe30a2e ]

Without the QCA ROME setup routine this adapter fails to establish a SCO connection.

T:  Bus=01 Lev=01 Prnt=01 Port=08 Cnt=01 Dev#=  2 Spd=12  MxCh= 0
D:  Ver= 1.10 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=13d3 ProdID=3491 Rev=00.01
C:  #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=100mA
I:  If#=0x0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
I:  If#=0x1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
---
 drivers/bluetooth/btusb.c | 1 +
 1 file changed, 1 insertion(+)
drivers/bluetooth/btusb.c +1 -0
index 7db48ae65cd2..0e2c86da6479 100644
@@ -279,6 +279,7 @@ static const struct usb_device_id blacklist_table[] = {
 	{ USB_DEVICE(0x04ca, 0x3015), .driver_info = BTUSB_QCA_ROME },
 	{ USB_DEVICE(0x04ca, 0x3016), .driver_info = BTUSB_QCA_ROME },
 	{ USB_DEVICE(0x04ca, 0x301a), .driver_info = BTUSB_QCA_ROME },
+	{ USB_DEVICE(0x13d3, 0x3491), .driver_info = BTUSB_QCA_ROME },
 	{ USB_DEVICE(0x13d3, 0x3496), .driver_info = BTUSB_QCA_ROME },
 
 	/* Broadcom BCM2035 */
Signature · 1 line
2.20.1
author maintainer
#207 lore raw

PATCH 206/371 Bluetooth: Add new 13d3:3501 QCA_ROME device

In reply to Greg Kroah-Hartman

[ Upstream commit 881cec4f6b4da78e54b73c046a60f39315964c7d ]

Without the QCA ROME setup routine this adapter fails to establish a SCO connection.

T:  Bus=01 Lev=01 Prnt=01 Port=04 Cnt=01 Dev#=  2 Spd=12  MxCh= 0
D:  Ver= 1.10 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=13d3 ProdID=3501 Rev=00.01
C:  #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=100mA
I:  If#=0x0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
I:  If#=0x1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
---
 drivers/bluetooth/btusb.c | 1 +
 1 file changed, 1 insertion(+)
drivers/bluetooth/btusb.c +1 -0
index 0e2c86da6479..4c9f11766e82 100644
@@ -281,6 +281,7 @@ static const struct usb_device_id blacklist_table[] = {
 	{ USB_DEVICE(0x04ca, 0x301a), .driver_info = BTUSB_QCA_ROME },
 	{ USB_DEVICE(0x13d3, 0x3491), .driver_info = BTUSB_QCA_ROME },
 	{ USB_DEVICE(0x13d3, 0x3496), .driver_info = BTUSB_QCA_ROME },
+	{ USB_DEVICE(0x13d3, 0x3501), .driver_info = BTUSB_QCA_ROME },
 
 	/* Broadcom BCM2035 */
 	{ USB_DEVICE(0x0a5c, 0x2009), .driver_info = BTUSB_BCM92035 },
Signature · 1 line
2.20.1
author maintainer
#208 lore raw

PATCH 207/371 Bluetooth: 6lowpan: search for destination address in all peers

In reply to Greg Kroah-Hartman

[ Upstream commit b188b03270b7f8568fc714101ce82fbf5e811c5a ]

Handle overlooked case where the target address is assigned to a peer and neither route nor gateway exist.

For one peer, no checks are performed to see if it is meant to receive packets for a given address.

As soon as there is a second peer however, checks are performed to deal with routes and gateways for handling complex setups with multiple hops to a target address.
This logic assumed that no route and no gateway imply that the destination address can not be reached, which is false in case of a direct peer.

---
 net/bluetooth/6lowpan.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)
net/bluetooth/6lowpan.c +10 -4
index a7cd23f00bde..50530561da98 100644
@@ -187,10 +187,16 @@ static inline struct lowpan_peer *peer_lookup_dst(struct lowpan_btle_dev *dev,
 	}
 
 	if (!rt) {
-		nexthop = &lowpan_cb(skb)->gw;
-
-		if (ipv6_addr_any(nexthop))
-			return NULL;
+		if (ipv6_addr_any(&lowpan_cb(skb)->gw)) {
+			/* There is neither route nor gateway,
+			 * probably the destination is a direct peer.
+			 */
+			nexthop = daddr;
+		} else {
+			/* There is a known gateway
+			 */
+			nexthop = &lowpan_cb(skb)->gw;
+		}
 	} else {
 		nexthop = rt6_nexthop(rt, daddr);
 
Signature · 1 line
2.20.1
author maintainer
#209 lore raw

PATCH 208/371 genirq: Update irq stats from NMI handlers

In reply to Greg Kroah-Hartman

[ Upstream commit c09cb1293523dd786ae54a12fd88001542cba2f6 ]

The NMI handlers handle_percpu_devid_fasteoi_nmi() and handle_fasteoi_nmi() do not update the interrupt counts. Due to that the NMI interrupt count does not show up correctly in /proc/interrupts.

Add the statistics and treat the NMI handlers in the same way as per cpu interrupts and prevent them from updating irq_desc::tot_count as this might be corrupted due to concurrency.

[ tglx: Massaged changelog ]

---
 kernel/irq/chip.c    | 4 ++++
 kernel/irq/irqdesc.c | 8 +++++++-
 2 files changed, 11 insertions(+), 1 deletion(-)
kernel/irq/chip.c +4 -0
index 04fe4f989bd8..bfac4d6761b3 100644
@@ -754,6 +754,8 @@ void handle_fasteoi_nmi(struct irq_desc *desc)
 	unsigned int irq = irq_desc_get_irq(desc);
 	irqreturn_t res;
 
+	__kstat_incr_irqs_this_cpu(desc);
+
 	trace_irq_handler_entry(irq, action);
 	/*
 	 * NMIs cannot be shared, there is only one action.
@@ -968,6 +970,8 @@ void handle_percpu_devid_fasteoi_nmi(struct irq_desc *desc)
 	unsigned int irq = irq_desc_get_irq(desc);
 	irqreturn_t res;
 
+	__kstat_incr_irqs_this_cpu(desc);
+
 	trace_irq_handler_entry(irq, action);
 	res = action->handler(irq, raw_cpu_ptr(action->percpu_dev_id));
 	trace_irq_handler_exit(irq, action, res);
kernel/irq/irqdesc.c +7 -1
index 9f8a709337cf..8a93df71673d 100644
@@ -945,6 +945,11 @@ unsigned int kstat_irqs_cpu(unsigned int irq, int cpu)
 			*per_cpu_ptr(desc->kstat_irqs, cpu) : 0;
 }
 
+static bool irq_is_nmi(struct irq_desc *desc)
+{
+	return desc->istate & IRQS_NMI;
+}
+
 /**
  * kstat_irqs - Get the statistics for an interrupt
  * @irq:	The interrupt number
@@ -962,7 +967,8 @@ unsigned int kstat_irqs(unsigned int irq)
 	if (!desc || !desc->kstat_irqs)
 		return 0;
 	if (!irq_settings_is_per_cpu_devid(desc) &&
-	    !irq_settings_is_per_cpu(desc))
+	    !irq_settings_is_per_cpu(desc) &&
+	    !irq_is_nmi(desc))
 	    return desc->tot_count;
 
 	for_each_possible_cpu(cpu)
Signature · 1 line
2.20.1
author maintainer
#210 lore raw

PATCH 209/371 perf tests: Fix record+probe_libc_inet_pton.sh for powerpc64

In reply to Greg Kroah-Hartman

[ Upstream commit bff5a556c149804de29347a88a884d25e4e4e3a2 ]

'probe libc's inet_pton & backtrace it with ping' testcase sometimes fails on powerpc because distro ping binary does not have symbol information and thus it prints "[unknown]" function name in the backtrace.

Accept "[unknown]" as valid function name for powerpc as well.

 # perf test -v "probe libc's inet_pton & backtrace it with ping"

Before:

59: probe libc's inet_pton & backtrace it with ping :
--- start ---
test child forked, pid 79695
ping 79718 [077] 96483.787025: probe_libc:inet_pton: (7fff83a754c8) 7fff83a754c8 __GI___inet_pton+0x8 (/usr/lib64/power9/libc-2.28.so) 7fff83a2b7a0 gaih_inet.constprop.7+0x1020 (/usr/lib64/power9/libc-2.28.so)
7fff83a2c170 getaddrinfo+0x160 (/usr/lib64/power9/libc-2.28.so) 1171830f4 [unknown] (/usr/bin/ping)
FAIL: expected backtrace entry ".*\+0x[[:xdigit:]]+[[:space:]]\(.*/bin/ping.*\)$"
got "1171830f4 [unknown] (/usr/bin/ping)"
test child finished with -1
---- end ----
probe libc's inet_pton & backtrace it with ping: FAILED!

After:

59: probe libc's inet_pton & backtrace it with ping :
--- start ---
test child forked, pid 79085
ping 79108 [045] 96400.214177: probe_libc:inet_pton: (7fffbb9654c8) 7fffbb9654c8 __GI___inet_pton+0x8 (/usr/lib64/power9/libc-2.28.so) 7fffbb91b7a0 gaih_inet.constprop.7+0x1020 (/usr/lib64/power9/libc-2.28.so)
7fffbb91c170 getaddrinfo+0x160 (/usr/lib64/power9/libc-2.28.so) 132e830f4 [unknown] (/usr/bin/ping)
test child finished with 0
---- end ----
probe libc's inet_pton & backtrace it with ping: Ok

Cc Alexander Shishkin <[email protected]>
Cc Hendrik Brueckner <[email protected]>
Cc Jiri Olsa <[email protected]>
Cc Michael Petlan <[email protected]>
Cc Namhyung Kim <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Sandipan Das <[email protected]>
Fixes 1632936480a5 ("perf tests: Fix record+probe_libc_inet_pton.sh without ping's debuginfo")
---
 tools/perf/tests/shell/record+probe_libc_inet_pton.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/tests/shell/record+probe_libc_inet_pton.sh +1 -1
index 61c9f8fc6fa1..58a99a292930 100755
@@ -44,7 +44,7 @@ trace_libc_inet_pton_backtrace() {
 		eventattr='max-stack=4'
 		echo "gaih_inet.*\+0x[[:xdigit:]]+[[:space:]]\($libc\)$" >> $expected
 		echo "getaddrinfo\+0x[[:xdigit:]]+[[:space:]]\($libc\)$" >> $expected
-		echo ".*\+0x[[:xdigit:]]+[[:space:]]\(.*/bin/ping.*\)$" >> $expected
+		echo ".*(\+0x[[:xdigit:]]+|\[unknown\])[[:space:]]\(.*/bin/ping.*\)$" >> $expected
 		;;
 	*)
 		eventattr='max-stack=3'
Signature · 1 line
2.20.1
author maintainer
#211 lore raw

PATCH 210/371 Bluetooth: Check state in l2cap_disconnect_rsp

In reply to Greg Kroah-Hartman

[ Upstream commit 28261da8a26f4915aa257d12d506c6ba179d961f ]

Because of both sides doing L2CAP disconnection at the same time, it was possible to receive L2CAP Disconnection Response with CID that was already freed. That caused problems if CID was already reused and L2CAP Connection Request with same CID was sent out. Before this patch kernel deleted channel context regardless of the state of the channel.

Example where leftover Disconnection Response (frame #402) causes local device to delete L2CAP channel which was not yet connected. This in turn confuses remote device's stack because same CID is re-used without properly disconnecting.

Btmon capture before patch:
** snip **

Quoted 1 line

ACL Data RX: Handle 43 flags 0x02 dlen 8 #394 [hci1] 10.748949

      Channel: 65 len 4 [PSM 3 mode 0] {chan 2}
      RFCOMM: Disconnect (DISC) (0x43)
         Address: 0x03 cr 1 dlci 0x00
         Control: 0x53 poll/final 1
         Length: 0
         FCS: 0xfd
< ACL Data TX: Handle 43 flags 0x00 dlen 8                #395 [hci1] 10.749062
      Channel: 65 len 4 [PSM 3 mode 0] {chan 2}
      RFCOMM: Unnumbered Ack (UA) (0x63)
         Address: 0x03 cr 1 dlci 0x00
         Control: 0x73 poll/final 1
         Length: 0
         FCS: 0xd7

< ACL Data TX: Handle 43 flags 0x00 dlen 12 #396 [hci1] 10.749073
L2CAP: Disconnection Request (0x06) ident 17 len 4
Destination CID: 65
Source CID: 65

Quoted 1 line

HCI Event: Number of Completed Packets (0x13) plen 5 #397 [hci1] 10.752391

Num handles: 1
Handle: 43
Count: 1

Quoted 1 line

HCI Event: Number of Completed Packets (0x13) plen 5 #398 [hci1] 10.753394

Num handles: 1
Handle: 43
Count: 1

Quoted 1 line

ACL Data RX: Handle 43 flags 0x02 dlen 12 #399 [hci1] 10.756499

L2CAP: Disconnection Request (0x06) ident 26 len 4
Destination CID: 65
Source CID: 65
< ACL Data TX: Handle 43 flags 0x00 dlen 12 #400 [hci1] 10.756548
L2CAP: Disconnection Response (0x07) ident 26 len 4
Destination CID: 65
Source CID: 65
< ACL Data TX: Handle 43 flags 0x00 dlen 12 #401 [hci1] 10.757459

      L2CAP: Connection Request (0x02) ident 18 len 4
        PSM: 1 (0x0001)
        Source CID: 65
Quoted 1 line

ACL Data RX: Handle 43 flags 0x02 dlen 12 #402 [hci1] 10.759148

L2CAP: Disconnection Response (0x07) ident 17 len 4
Destination CID: 65
Source CID: 65
= bluetoothd: 00:1E:AB:4C:56:54: error updating services: Input/o.. 10.759447

Quoted 1 line

HCI Event: Number of Completed Packets (0x13) plen 5 #403 [hci1] 10.759386

Num handles: 1
Handle: 43
Count: 1

Quoted 1 line

ACL Data RX: Handle 43 flags 0x02 dlen 12 #404 [hci1] 10.760397

      L2CAP: Connection Request (0x02) ident 27 len 4
        PSM: 3 (0x0003)
        Source CID: 65
< ACL Data TX: Handle 43 flags 0x00 dlen 16               #405 [hci1] 10.760441
      L2CAP: Connection Response (0x03) ident 27 len 8
        Destination CID: 65
        Source CID: 65
        Result: Connection successful (0x0000)
        Status: No further information available (0x0000)

< ACL Data TX: Handle 43 flags 0x00 dlen 27 #406 [hci1] 10.760449
L2CAP: Configure Request (0x04) ident 19 len 19
Destination CID: 65
Flags: 0x0000
Option: Maximum Transmission Unit (0x01) [mandatory]
MTU: 1013
Option: Retransmission and Flow Control (0x04) [mandatory]
Mode: Basic (0x00)
TX window size: 0
Max transmit: 0
Retransmission timeout: 0
Monitor timeout: 0
Maximum PDU size: 0

Quoted 1 line

HCI Event: Number of Completed Packets (0x13) plen 5 #407 [hci1] 10.761399

Num handles: 1
Handle: 43
Count: 1

Quoted 1 line

ACL Data RX: Handle 43 flags 0x02 dlen 16 #408 [hci1] 10.762942

      L2CAP: Connection Response (0x03) ident 18 len 8
        Destination CID: 66
        Source CID: 65
        Result: Connection successful (0x0000)
        Status: No further information available (0x0000)

*snip*

Similar case after the patch:
*snip*

Quoted 1 line

ACL Data RX: Handle 43 flags 0x02 dlen 8 #22702 [hci0] 1664.411056

      Channel: 65 len 4 [PSM 3 mode 0] {chan 3}
      RFCOMM: Disconnect (DISC) (0x43)
         Address: 0x03 cr 1 dlci 0x00
         Control: 0x53 poll/final 1
         Length: 0
         FCS: 0xfd
< ACL Data TX: Handle 43 flags 0x00 dlen 8            #22703 [hci0] 1664.411136
      Channel: 65 len 4 [PSM 3 mode 0] {chan 3}
      RFCOMM: Unnumbered Ack (UA) (0x63)
         Address: 0x03 cr 1 dlci 0x00
         Control: 0x73 poll/final 1
         Length: 0
         FCS: 0xd7

< ACL Data TX: Handle 43 flags 0x00 dlen 12 #22704 [hci0] 1664.411143
L2CAP: Disconnection Request (0x06) ident 11 len 4
Destination CID: 65
Source CID: 65

Quoted 1 line

HCI Event: Number of Completed Pac.. (0x13) plen 5 #22705 [hci0] 1664.414009

Num handles: 1
Handle: 43
Count: 1

Quoted 1 line

HCI Event: Number of Completed Pac.. (0x13) plen 5 #22706 [hci0] 1664.415007

Num handles: 1
Handle: 43
Count: 1

Quoted 1 line

ACL Data RX: Handle 43 flags 0x02 dlen 12 #22707 [hci0] 1664.418674

L2CAP: Disconnection Request (0x06) ident 17 len 4
Destination CID: 65
Source CID: 65
< ACL Data TX: Handle 43 flags 0x00 dlen 12 #22708 [hci0] 1664.418762
L2CAP: Disconnection Response (0x07) ident 17 len 4
Destination CID: 65
Source CID: 65
< ACL Data TX: Handle 43 flags 0x00 dlen 12 #22709 [hci0] 1664.421073

      L2CAP: Connection Request (0x02) ident 12 len 4
        PSM: 1 (0x0001)
        Source CID: 65
Quoted 1 line

ACL Data RX: Handle 43 flags 0x02 dlen 12 #22710 [hci0] 1664.421371

L2CAP: Disconnection Response (0x07) ident 11 len 4
Destination CID: 65
Source CID: 65

Quoted 1 line

HCI Event: Number of Completed Pac.. (0x13) plen 5 #22711 [hci0] 1664.424082

Num handles: 1
Handle: 43
Count: 1

Quoted 1 line

HCI Event: Number of Completed Pac.. (0x13) plen 5 #22712 [hci0] 1664.425040

Num handles: 1
Handle: 43
Count: 1

Quoted 1 line

ACL Data RX: Handle 43 flags 0x02 dlen 12 #22713 [hci0] 1664.426103

      L2CAP: Connection Request (0x02) ident 18 len 4
        PSM: 3 (0x0003)
        Source CID: 65
< ACL Data TX: Handle 43 flags 0x00 dlen 16           #22714 [hci0] 1664.426186
      L2CAP: Connection Response (0x03) ident 18 len 8
        Destination CID: 66
        Source CID: 65
        Result: Connection successful (0x0000)
        Status: No further information available (0x0000)

< ACL Data TX: Handle 43 flags 0x00 dlen 27 #22715 [hci0] 1664.426196
L2CAP: Configure Request (0x04) ident 13 len 19
Destination CID: 65
Flags: 0x0000
Option: Maximum Transmission Unit (0x01) [mandatory]
MTU: 1013
Option: Retransmission and Flow Control (0x04) [mandatory]
Mode: Basic (0x00)
TX window size: 0
Max transmit: 0
Retransmission timeout: 0
Monitor timeout: 0
Maximum PDU size: 0

Quoted 1 line

ACL Data RX: Handle 43 flags 0x02 dlen 16 #22716 [hci0] 1664.428804

      L2CAP: Connection Response (0x03) ident 12 len 8
        Destination CID: 66
        Source CID: 65
        Result: Connection successful (0x0000)
        Status: No further information available (0x0000)

*snip*

Fix is to check that channel is in state BT_DISCONN before deleting the channel.

This bug was found while fuzzing Bluez's OBEX implementation using Synopsys Defensics.

---
 net/bluetooth/l2cap_core.c | 6 ++++++
 1 file changed, 6 insertions(+)
net/bluetooth/l2cap_core.c +6 -0
index 5406d7cd46ad..771e3e17bb6a 100644
@@ -4394,6 +4394,12 @@ static inline int l2cap_disconnect_rsp(struct l2cap_conn *conn,
 
 	l2cap_chan_lock(chan);
 
+	if (chan->state != BT_DISCONN) {
+		l2cap_chan_unlock(chan);
+		mutex_unlock(&conn->chan_lock);
+		return 0;
+	}
+
 	l2cap_chan_hold(chan);
 	l2cap_chan_del(chan, 0);
 
Signature · 1 line
2.20.1
author maintainer
#212 lore raw

PATCH 211/371 Bluetooth: hidp: NUL terminate a string in the compat ioctl

In reply to Greg Kroah-Hartman

[ Upstream commit dcae9052ebb0c5b2614de620323d615fcbfda7f8 ]

This change is similar to commit a1616a5ac99e ("Bluetooth: hidp: fix buffer overflow") but for the compat ioctl. We take a string from the user and forgot to ensure that it's NUL terminated.

I have also changed the strncpy() in to strscpy() in hidp_setup_hid(). The difference is the strncpy() doesn't necessarily NUL terminate the destination string. Either change would fix the problem but it's nice to take a belt and suspenders approach and do both.

---
 net/bluetooth/hidp/core.c | 2 +-
 net/bluetooth/hidp/sock.c | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)
net/bluetooth/hidp/core.c +1 -1
index a442e21f3894..5abd423b55fa 100644
@@ -775,7 +775,7 @@ static int hidp_setup_hid(struct hidp_session *session,
 	hid->version = req->version;
 	hid->country = req->country;
 
-	strncpy(hid->name, req->name, sizeof(hid->name));
+	strscpy(hid->name, req->name, sizeof(hid->name));
 
 	snprintf(hid->phys, sizeof(hid->phys), "%pMR",
 		 &l2cap_pi(session->ctrl_sock->sk)->chan->src);
net/bluetooth/hidp/sock.c +1 -0
index 2151913892ce..03be6a4baef3 100644
@@ -192,6 +192,7 @@ static int hidp_sock_compat_ioctl(struct socket *sock, unsigned int cmd, unsigne
 		ca.version = ca32.version;
 		ca.flags = ca32.flags;
 		ca.idle_to = ca32.idle_to;
+		ca32.name[sizeof(ca32.name) - 1] = '\0';
 		memcpy(ca.name, ca32.name, 128);
 
 		csock = sockfd_lookup(ca.ctrl_sock, &err);
Signature · 1 line
2.20.1
author maintainer
#213 lore raw

PATCH 212/371 gtp: add missing gtp_encap_disable_sock() in gtp_encap_enable()

In reply to Greg Kroah-Hartman

[ Upstream commit e30155fd23c9c141cbe7d99b786e10a83a328837 ]

If an invalid role is sent from user space, gtp_encap_enable() will fail. Then, it should call gtp_encap_disable_sock() but current code doesn't. It makes memory leak.

Fixes 91ed81f9abc7 ("gtp: support SGSN-side tunnels")
---
 drivers/net/gtp.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/net/gtp.c +6 -1
index 7a145172d503..83488f2bf7a0 100644
@@ -847,8 +847,13 @@ static int gtp_encap_enable(struct gtp_dev *gtp, struct nlattr *data[])
 
 	if (data[IFLA_GTP_ROLE]) {
 		role = nla_get_u32(data[IFLA_GTP_ROLE]);
-		if (role > GTP_ROLE_SGSN)
+		if (role > GTP_ROLE_SGSN) {
+			if (sk0)
+				gtp_encap_disable_sock(sk0);
+			if (sk1u)
+				gtp_encap_disable_sock(sk1u);
 			return -EINVAL;
+		}
 	}
 
 	gtp->sk0 = sk0;
Signature · 1 line
2.20.1
author maintainer
#214 lore raw

PATCH 213/371 Bluetooth: validate BLE connection interval updates

In reply to Greg Kroah-Hartman

[ Upstream commit c49a8682fc5d298d44e8d911f4fa14690ea9485e ]

Problem: The Linux Bluetooth stack yields complete control over the BLE connection interval to the remote device.

The Linux Bluetooth stack provides access to the BLE connection interval min and max values through /sys/kernel/debug/bluetooth/hci0/ conn_min_interval and /sys/kernel/debug/bluetooth/hci0/conn_max_interval. These values are used for initial BLE connections, but the remote device has the ability to request a connection parameter update. In the event that the remote side requests to change the connection interval, the Linux kernel currently only validates that the desired value is within the acceptable range in the Bluetooth specification (6 - 3200, corresponding to 7.5ms - 4000ms). There is currently no validation that the desired value requested by the remote device is within the min/max limits specified in the conn_min_interval/conn_max_interval configurations. This essentially leads to Linux yielding complete control over the connection interval to the remote device.

The proposed patch adds a verification step to the connection parameter update mechanism, ensuring that the desired value is within the min/max bounds of the current connection. If the desired value is outside of the current connection min/max values, then the connection parameter update request is rejected and the negative response is returned to the remote device. Recall that the initial connection is established using the local conn_min_interval/conn_max_interval values, so this allows the Linux administrator to retain control over the BLE connection interval.

The one downside that I see is that the current default Linux values for conn_min_interval and conn_max_interval typically correspond to 30ms and 50ms respectively. If this change were accepted, then it is feasible that some devices would no longer be able to negotiate to their desired connection interval values. This might be remedied by setting the default Linux conn_min_interval and conn_max_interval values to the widest supported range (6 - 3200 / 7.5ms - 4000ms). This could lead to the same behavior as the current implementation, where the remote device could request to change the connection interval value to any value that is permitted by the Bluetooth specification, and Linux would accept the desired value.

---
 net/bluetooth/hci_event.c  | 5 +++++
 net/bluetooth/l2cap_core.c | 9 ++++++++-
 2 files changed, 13 insertions(+), 1 deletion(-)
net/bluetooth/hci_event.c +5 -0
index 8b893baf9bbe..31eb0449479b 100644
@@ -5588,6 +5588,11 @@ static void hci_le_remote_conn_param_req_evt(struct hci_dev *hdev,
 		return send_conn_param_neg_reply(hdev, handle,
 						 HCI_ERROR_UNKNOWN_CONN_ID);
 
+	if (min < hcon->le_conn_min_interval ||
+	    max > hcon->le_conn_max_interval)
+		return send_conn_param_neg_reply(hdev, handle,
+						 HCI_ERROR_INVALID_LL_PARAMS);
+
 	if (hci_check_conn_params(min, max, latency, timeout))
 		return send_conn_param_neg_reply(hdev, handle,
 						 HCI_ERROR_INVALID_LL_PARAMS);
net/bluetooth/l2cap_core.c +8 -1
index 771e3e17bb6a..32d2be9d6858 100644
@@ -5297,7 +5297,14 @@ static inline int l2cap_conn_param_update_req(struct l2cap_conn *conn,
 
 	memset(&rsp, 0, sizeof(rsp));
 
-	err = hci_check_conn_params(min, max, latency, to_multiplier);
+	if (min < hcon->le_conn_min_interval ||
+	    max > hcon->le_conn_max_interval) {
+		BT_DBG("requested connection interval exceeds current bounds.");
+		err = -EINVAL;
+	} else {
+		err = hci_check_conn_params(min, max, latency, to_multiplier);
+	}
+
 	if (err)
 		rsp.result = cpu_to_le16(L2CAP_CONN_PARAM_REJECTED);
 	else
Signature · 1 line
2.20.1
author maintainer
#215 lore raw

PATCH 214/371 gtp: fix suspicious RCU usage

In reply to Greg Kroah-Hartman

[ Upstream commit e198987e7dd7d3645a53875151cd6f8fc425b706 ]

gtp_encap_enable_socket() and gtp_encap_destroy() are not protected by rcu_read_lock(). and it's not safe to write sk->sk_user_data. This patch make these functions to use lock_sock() instead of rcu_dereference_sk_user_data().

Test commands:
gtp-link add gtp1

Splat looks like:

[   83.238315] =============================
[   83.239127] WARNING: suspicious RCU usage
[   83.239702] 5.2.0-rc6+ #49 Not tainted
[   83.240268] -----------------------------
[   83.241205] drivers/net/gtp.c:799 suspicious rcu_dereference_check() usage!
[   83.243828]
[   83.243828] other info that might help us debug this:
[   83.243828]
[   83.246325]
[   83.246325] rcu_scheduler_active = 2, debug_locks = 1
[   83.247314] 1 lock held by gtp-link/1008:
[   83.248523]  #0: 0000000017772c7f (rtnl_mutex){+.+.}, at: __rtnl_newlink+0x5f5/0x11b0
[   83.251503]
[   83.251503] stack backtrace:
[   83.252173] CPU: 0 PID: 1008 Comm: gtp-link Not tainted 5.2.0-rc6+ #49
[   83.253271] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006
[   83.254562] Call Trace:
[   83.254995]  dump_stack+0x7c/0xbb
[   83.255567]  gtp_encap_enable_socket+0x2df/0x360 [gtp]
[   83.256415]  ? gtp_find_dev+0x1a0/0x1a0 [gtp]
[   83.257161]  ? memset+0x1f/0x40
[   83.257843]  gtp_newlink+0x90/0xa21 [gtp]
[   83.258497]  ? __netlink_ns_capable+0xc3/0xf0
[   83.259260]  __rtnl_newlink+0xb9f/0x11b0
[   83.260022]  ? rtnl_link_unregister+0x230/0x230

[ ... ]

Fixes 1e3a3abd8b28 ("gtp: make GTP sockets in gtp_newlink optional")
---
 drivers/net/gtp.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/net/gtp.c +6 -2
index 83488f2bf7a0..f45a806b6c06 100644
@@ -293,12 +293,14 @@ static void gtp_encap_destroy(struct sock *sk)
 {
 	struct gtp_dev *gtp;
 
-	gtp = rcu_dereference_sk_user_data(sk);
+	lock_sock(sk);
+	gtp = sk->sk_user_data;
 	if (gtp) {
 		udp_sk(sk)->encap_type = 0;
 		rcu_assign_sk_user_data(sk, NULL);
 		sock_put(sk);
 	}
+	release_sock(sk);
 }
 
 static void gtp_encap_disable_sock(struct sock *sk)
@@ -800,7 +802,8 @@ static struct sock *gtp_encap_enable_socket(int fd, int type,
 		goto out_sock;
 	}
 
-	if (rcu_dereference_sk_user_data(sock->sk)) {
+	lock_sock(sock->sk);
+	if (sock->sk->sk_user_data) {
 		sk = ERR_PTR(-EBUSY);
 		goto out_sock;
 	}
@@ -816,6 +819,7 @@ static struct sock *gtp_encap_enable_socket(int fd, int type,
 	setup_udp_tunnel_sock(sock_net(sock->sk), sock, &tuncfg);
 
 out_sock:
+	release_sock(sock->sk);
 	sockfd_put(sock);
 	return sk;
 }
Signature · 1 line
2.20.1
author maintainer
#216 lore raw

PATCH 215/371 gtp: fix Illegal context switch in RCU read-side critical section.

In reply to Greg Kroah-Hartman

[ Upstream commit 3f167e1921865b379a9becf03828e7202c7b4917 ]

ipv4_pdp_add() is called in RCU read-side critical section. So GFP_KERNEL should not be used in the function.
This patch make ipv4_pdp_add() to use GFP_ATOMIC instead of GFP_KERNEL.

Test commands:
gtp-link add gtp1 &
gtp-tunnel add gtp1 v1 100 200 1.1.1.1 2.2.2.2

Splat looks like:

[  130.618881] =============================
[  130.626382] WARNING: suspicious RCU usage
[  130.626994] 5.2.0-rc6+ #50 Not tainted
[  130.627622] -----------------------------
[  130.628223] ./include/linux/rcupdate.h:266 Illegal context switch in RCU read-side critical section!
[  130.629684]
[  130.629684] other info that might help us debug this:
[  130.629684]
[  130.631022]
[  130.631022] rcu_scheduler_active = 2, debug_locks = 1
[  130.632136] 4 locks held by gtp-tunnel/1025:
[  130.632925]  #0: 000000002b93c8b7 (cb_lock){++++}, at: genl_rcv+0x15/0x40
[  130.634159]  #1: 00000000f17bc999 (genl_mutex){+.+.}, at: genl_rcv_msg+0xfb/0x130
[  130.635487]  #2: 00000000c644ed8e (rtnl_mutex){+.+.}, at: gtp_genl_new_pdp+0x18c/0x1150 [gtp]
[  130.636936]  #3: 0000000007a1cde7 (rcu_read_lock){....}, at: gtp_genl_new_pdp+0x187/0x1150 [gtp]
[  130.638348]
[  130.638348] stack backtrace:
[  130.639062] CPU: 1 PID: 1025 Comm: gtp-tunnel Not tainted 5.2.0-rc6+ #50
[  130.641318] Call Trace:
[  130.641707]  dump_stack+0x7c/0xbb
[  130.642252]  ___might_sleep+0x2c0/0x3b0
[  130.642862]  kmem_cache_alloc_trace+0x1cd/0x2b0
[  130.643591]  gtp_genl_new_pdp+0x6c5/0x1150 [gtp]
[  130.644371]  genl_family_rcv_msg+0x63a/0x1030
[  130.645074]  ? mutex_lock_io_nested+0x1090/0x1090
[  130.645845]  ? genl_unregister_family+0x630/0x630
[  130.646592]  ? debug_show_all_locks+0x2d0/0x2d0
[  130.647293]  ? check_flags.part.40+0x440/0x440
[  130.648099]  genl_rcv_msg+0xa3/0x130

[ ... ]

Fixes 459aa660eb1d ("gtp: add initial driver for datapath of GPRS Tunneling Protocol (GTP-U)")
---
 drivers/net/gtp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/gtp.c +1 -1
index f45a806b6c06..6f1ad7ccaea6 100644
@@ -958,7 +958,7 @@ static int ipv4_pdp_add(struct gtp_dev *gtp, struct sock *sk,
 
 	}
 
-	pctx = kmalloc(sizeof(struct pdp_ctx), GFP_KERNEL);
+	pctx = kmalloc(sizeof(*pctx), GFP_ATOMIC);
 	if (pctx == NULL)
 		return -ENOMEM;
 
Signature · 1 line
2.20.1
author maintainer
#217 lore raw

PATCH 216/371 gtp: fix use-after-free in gtp_encap_destroy()

In reply to Greg Kroah-Hartman

[ Upstream commit 1788b8569f5de27da09087fa3f6580d2aa04cc75 ]

gtp_encap_destroy() is called twice.
1. When interface is deleted.
2. When udp socket is destroyed.
either gtp->sk0 or gtp->sk1u could be freed by sock_put() in gtp_encap_destroy(). so, when gtp_encap_destroy() is called again, it would uses freed sk pointer.

patch makes gtp_encap_destroy() to set either gtp->sk0 or gtp->sk1u to null. in addition, both gtp->sk0 and gtp->sk1u pointer are protected by rtnl_lock. so, rtnl_lock() is added.

Test command:
gtp-link add gtp1 &
killall gtp-link
ip link del gtp1

Splat looks like:

[   83.182767] BUG: KASAN: use-after-free in __lock_acquire+0x3a20/0x46a0
[   83.184128] Read of size 8 at addr ffff8880cc7d5360 by task ip/1008
[   83.185567] CPU: 1 PID: 1008 Comm: ip Not tainted 5.2.0-rc6+ #50
[   83.188469] Call Trace:
[ ... ]
[   83.200126]  lock_acquire+0x141/0x380
[   83.200575]  ? lock_sock_nested+0x3a/0xf0
[   83.201069]  _raw_spin_lock_bh+0x38/0x70
[   83.201551]  ? lock_sock_nested+0x3a/0xf0
[   83.202044]  lock_sock_nested+0x3a/0xf0
[   83.202520]  gtp_encap_destroy+0x18/0xe0 [gtp]
[   83.203065]  gtp_encap_disable.isra.14+0x13/0x50 [gtp]
[   83.203687]  gtp_dellink+0x56/0x170 [gtp]
[   83.204190]  rtnl_delete_link+0xb4/0x100
[ ... ]
[   83.236513] Allocated by task 976:
[   83.236925]  save_stack+0x19/0x80
[   83.237332]  __kasan_kmalloc.constprop.3+0xa0/0xd0
[   83.237894]  kmem_cache_alloc+0xd8/0x280
[   83.238360]  sk_prot_alloc.isra.42+0x50/0x200
[   83.238874]  sk_alloc+0x32/0x940
[   83.239264]  inet_create+0x283/0xc20
[   83.239684]  __sock_create+0x2dd/0x540
[   83.240136]  __sys_socket+0xca/0x1a0
[   83.240550]  __x64_sys_socket+0x6f/0xb0
[   83.240998]  do_syscall_64+0x9c/0x450
[   83.241466]  entry_SYSCALL_64_after_hwframe+0x49/0xbe
[   83.242061]
[   83.242249] Freed by task 0:
[   83.242616]  save_stack+0x19/0x80
[   83.243013]  __kasan_slab_free+0x111/0x150
[   83.243498]  kmem_cache_free+0x89/0x250
[   83.244444]  __sk_destruct+0x38f/0x5a0
[   83.245366]  rcu_core+0x7e9/0x1c20
[   83.245766]  __do_softirq+0x213/0x8fa
Fixes 1e3a3abd8b28 ("gtp: make GTP sockets in gtp_newlink optional")
---
 drivers/net/gtp.c | 17 +++++++++++++++--
 1 file changed, 15 insertions(+), 2 deletions(-)
drivers/net/gtp.c +15 -2
index 6f1ad7ccaea6..61e9b288d2dc 100644
@@ -289,13 +289,17 @@ static int gtp1u_udp_encap_recv(struct gtp_dev *gtp, struct sk_buff *skb)
 	return gtp_rx(pctx, skb, hdrlen, gtp->role);
 }
 
-static void gtp_encap_destroy(struct sock *sk)
+static void __gtp_encap_destroy(struct sock *sk)
 {
 	struct gtp_dev *gtp;
 
 	lock_sock(sk);
 	gtp = sk->sk_user_data;
 	if (gtp) {
+		if (gtp->sk0 == sk)
+			gtp->sk0 = NULL;
+		else
+			gtp->sk1u = NULL;
 		udp_sk(sk)->encap_type = 0;
 		rcu_assign_sk_user_data(sk, NULL);
 		sock_put(sk);
@@ -303,12 +307,19 @@ static void gtp_encap_destroy(struct sock *sk)
 	release_sock(sk);
 }
 
+static void gtp_encap_destroy(struct sock *sk)
+{
+	rtnl_lock();
+	__gtp_encap_destroy(sk);
+	rtnl_unlock();
+}
+
 static void gtp_encap_disable_sock(struct sock *sk)
 {
 	if (!sk)
 		return;
 
-	gtp_encap_destroy(sk);
+	__gtp_encap_destroy(sk);
 }
 
 static void gtp_encap_disable(struct gtp_dev *gtp)
@@ -1047,6 +1058,7 @@ static int gtp_genl_new_pdp(struct sk_buff *skb, struct genl_info *info)
 		return -EINVAL;
 	}
 
+	rtnl_lock();
 	rcu_read_lock();
 
 	gtp = gtp_find_dev(sock_net(skb->sk), info->attrs);
@@ -1071,6 +1083,7 @@ static int gtp_genl_new_pdp(struct sk_buff *skb, struct genl_info *info)
 
 out_unlock:
 	rcu_read_unlock();
+	rtnl_unlock();
 	return err;
 }
 
Signature · 1 line
2.20.1
author maintainer
#218 lore raw

PATCH 217/371 gtp: fix use-after-free in gtp_newlink()

In reply to Greg Kroah-Hartman

[ Upstream commit a2bed90704c68d3763bf24decb1b781a45395de8 ]

Current gtp_newlink() could be called after unregister_pernet_subsys(). gtp_newlink() uses gtp_net but it can be destroyed by unregister_pernet_subsys().
So unregister_pernet_subsys() should be called after rtnl_link_unregister().

Test commands:
#SHELL 1
while :
do

	   for i in {1..5}
	   do
		./gtp-link add gtp$i &
	   done
	   killall gtp-link

done

#SHELL 2
while :
do
modprobe -rv gtp
done

Splat looks like:

[  753.176631] BUG: KASAN: use-after-free in gtp_newlink+0x9b4/0xa5c [gtp]
[  753.177722] Read of size 8 at addr ffff8880d48f2458 by task gtp-link/7126
[  753.179082] CPU: 0 PID: 7126 Comm: gtp-link Tainted: G        W         5.2.0-rc6+ #50
[  753.185801] Call Trace:
[  753.186264]  dump_stack+0x7c/0xbb
[  753.186863]  ? gtp_newlink+0x9b4/0xa5c [gtp]
[  753.187583]  print_address_description+0xc7/0x240
[  753.188382]  ? gtp_newlink+0x9b4/0xa5c [gtp]
[  753.189097]  ? gtp_newlink+0x9b4/0xa5c [gtp]
[  753.189846]  __kasan_report+0x12a/0x16f
[  753.190542]  ? gtp_newlink+0x9b4/0xa5c [gtp]
[  753.191298]  kasan_report+0xe/0x20
[  753.191893]  gtp_newlink+0x9b4/0xa5c [gtp]
[  753.192580]  ? __netlink_ns_capable+0xc3/0xf0
[  753.193370]  __rtnl_newlink+0xb9f/0x11b0
[ ... ]
[  753.241201] Allocated by task 7186:
[  753.241844]  save_stack+0x19/0x80
[  753.242399]  __kasan_kmalloc.constprop.3+0xa0/0xd0
[  753.243192]  __kmalloc+0x13e/0x300
[  753.243764]  ops_init+0xd6/0x350
[  753.244314]  register_pernet_operations+0x249/0x6f0
[ ... ]
[  753.251770] Freed by task 7178:
[  753.252288]  save_stack+0x19/0x80
[  753.252833]  __kasan_slab_free+0x111/0x150
[  753.253962]  kfree+0xc7/0x280
[  753.254509]  ops_free_list.part.11+0x1c4/0x2d0
[  753.255241]  unregister_pernet_operations+0x262/0x390
[ ... ]
[  753.285883] list_add corruption. next->prev should be prev (ffff8880d48f2458), but was ffff8880d497d878. (next.
[  753.287241] ------------[ cut here ]------------
[  753.287794] kernel BUG at lib/list_debug.c:25!
[  753.288364] invalid opcode: 0000 [#1] SMP DEBUG_PAGEALLOC KASAN PTI
[  753.289099] CPU: 0 PID: 7126 Comm: gtp-link Tainted: G    B   W         5.2.0-rc6+ #50
[  753.291036] RIP: 0010:__list_add_valid+0x74/0xd0
[  753.291589] Code: 48 39 da 75 27 48 39 f5 74 36 48 39 dd 74 31 48 83 c4 08 b8 01 00 00 00 5b 5d c3 48 89 d9 48b
[  753.293779] RSP: 0018:ffff8880cae8f398 EFLAGS: 00010286
[  753.294401] RAX: 0000000000000075 RBX: ffff8880d497d878 RCX: 0000000000000000
[  753.296260] RDX: 0000000000000075 RSI: 0000000000000008 RDI: ffffed10195d1e69
[  753.297070] RBP: ffff8880cd250ae0 R08: ffffed101b4bff21 R09: ffffed101b4bff21
[  753.297899] R10: 0000000000000001 R11: ffffed101b4bff20 R12: ffff8880d497d878
[  753.298703] R13: 0000000000000000 R14: ffff8880cd250ae0 R15: ffff8880d48f2458
[  753.299564] FS:  00007f5f79805740(0000) GS:ffff8880da400000(0000) knlGS:0000000000000000
[  753.300533] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  753.301231] CR2: 00007fe8c7ef4f10 CR3: 00000000b71a6006 CR4: 00000000000606f0
[  753.302183] Call Trace:
[  753.302530]  gtp_newlink+0x5f6/0xa5c [gtp]
[  753.303037]  ? __netlink_ns_capable+0xc3/0xf0
[  753.303576]  __rtnl_newlink+0xb9f/0x11b0
[  753.304092]  ? rtnl_link_unregister+0x230/0x230
Fixes 459aa660eb1d ("gtp: add initial driver for datapath of GPRS Tunneling Protocol (GTP-U)")
---
 drivers/net/gtp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/gtp.c +1 -1
index 61e9b288d2dc..d178d5bad7e4 100644
@@ -1385,9 +1385,9 @@ late_initcall(gtp_init);
 
 static void __exit gtp_fini(void)
 {
-	unregister_pernet_subsys(&gtp_net_ops);
 	genl_unregister_family(&gtp_genl_family);
 	rtnl_link_unregister(&gtp_link_ops);
+	unregister_pernet_subsys(&gtp_net_ops);
 
 	pr_info("GTP module unloaded\n");
 }
Signature · 1 line
2.20.1
author maintainer
#219 lore raw

PATCH 218/371 xdp: fix race on generic receive path

In reply to Greg Kroah-Hartman

[ Upstream commit bf0bdd1343efbbf65b4d53aef1fce14acbd79d50 ]

Unlike driver mode, generic xdp receive could be triggered by different threads on different CPU cores at the same time leading to the fill and rx queue breakage. For example, this could happen while sending packets from two processes to the first interface of veth pair while the second part of it is open with AF_XDP socket.

Need to take a lock for each generic receive to avoid race.

Fixes c497176cb2e4 ("xsk: add Rx receive functions and poll support")
---
 include/net/xdp_sock.h |  2 ++
 net/xdp/xsk.c          | 31 ++++++++++++++++++++++---------
 2 files changed, 24 insertions(+), 9 deletions(-)
include/net/xdp_sock.h +2 -0
index d074b6d60f8a..ac3c047d058c 100644
@@ -67,6 +67,8 @@ struct xdp_sock {
 	 * in the SKB destructor callback.
 	 */
 	spinlock_t tx_completion_lock;
+	/* Protects generic receive. */
+	spinlock_t rx_lock;
 	u64 rx_dropped;
 };
 
net/xdp/xsk.c +22 -9
index a14e8864e4fa..5e0637db92ea 100644
@@ -123,13 +123,17 @@ int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 	u64 addr;
 	int err;
 
-	if (xs->dev != xdp->rxq->dev || xs->queue_id != xdp->rxq->queue_index)
-		return -EINVAL;
+	spin_lock_bh(&xs->rx_lock);
+
+	if (xs->dev != xdp->rxq->dev || xs->queue_id != xdp->rxq->queue_index) {
+		err = -EINVAL;
+		goto out_unlock;
+	}
 
 	if (!xskq_peek_addr(xs->umem->fq, &addr) ||
 	    len > xs->umem->chunk_size_nohr - XDP_PACKET_HEADROOM) {
-		xs->rx_dropped++;
-		return -ENOSPC;
+		err = -ENOSPC;
+		goto out_drop;
 	}
 
 	addr += xs->umem->headroom;
@@ -138,13 +142,21 @@ int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 	memcpy(buffer, xdp->data_meta, len + metalen);
 	addr += metalen;
 	err = xskq_produce_batch_desc(xs->rx, addr, len);
-	if (!err) {
-		xskq_discard_addr(xs->umem->fq);
-		xsk_flush(xs);
-		return 0;
-	}
+	if (err)
+		goto out_drop;
+
+	xskq_discard_addr(xs->umem->fq);
+	xskq_produce_flush_desc(xs->rx);
 
+	spin_unlock_bh(&xs->rx_lock);
+
+	xs->sk.sk_data_ready(&xs->sk);
+	return 0;
+
+out_drop:
 	xs->rx_dropped++;
+out_unlock:
+	spin_unlock_bh(&xs->rx_lock);
 	return err;
 }
 
@@ -765,6 +777,7 @@ static int xsk_create(struct net *net, struct socket *sock, int protocol,
 
 	xs = xdp_sk(sk);
 	mutex_init(&xs->mutex);
+	spin_lock_init(&xs->rx_lock);
 	spin_lock_init(&xs->tx_completion_lock);
 
 	mutex_lock(&net->xdp.lock);
Signature · 1 line
2.20.1
author maintainer
#220 lore raw

PATCH 219/371 net: mvmdio: defer probe of orion-mdio if a clock is not ready

In reply to Greg Kroah-Hartman

[ Upstream commit 433a06d7d74e677c40b1148c70c48677ff62fb6b ]

Defer probing of the orion-mdio interface when getting a clock returns EPROBE_DEFER. This avoids locking up the Armada 8k SoC when mdio is used before all clocks have been enabled.

---
 drivers/net/ethernet/marvell/mvmdio.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/net/ethernet/marvell/mvmdio.c +5 -0
index c5dac6bd2be4..903836e334d8 100644
@@ -321,6 +321,10 @@ static int orion_mdio_probe(struct platform_device *pdev)
 
 	for (i = 0; i < ARRAY_SIZE(dev->clk); i++) {
 		dev->clk[i] = of_clk_get(pdev->dev.of_node, i);
+		if (PTR_ERR(dev->clk[i]) == -EPROBE_DEFER) {
+			ret = -EPROBE_DEFER;
+			goto out_clk;
+		}
 		if (IS_ERR(dev->clk[i]))
 			break;
 		clk_prepare_enable(dev->clk[i]);
@@ -362,6 +366,7 @@ static int orion_mdio_probe(struct platform_device *pdev)
 	if (dev->err_interrupt > 0)
 		writel(0, dev->regs + MVMDIO_ERR_INT_MASK);
 
+out_clk:
 	for (i = 0; i < ARRAY_SIZE(dev->clk); i++) {
 		if (IS_ERR(dev->clk[i]))
 			break;
Signature · 1 line
2.20.1
author maintainer
#221 lore raw

PATCH 220/371 net: hns3: fix __QUEUE_STATE_STACK_XOFF not cleared issue

In reply to Greg Kroah-Hartman

[ Upstream commit f96315f2f17e7b2580d2fec7c4d6a706a131d904 ]

When change MTU or other operations, which just calling .reset_notify to do HNAE3_DOWN_CLIENT and HNAE3_UP_CLIENT, then
the netdev_tx_reset_queue() in the hns3_clear_all_ring() will be ignored. So the dev_watchdog() may misdiagnose a TX timeout.

This patch separates netdev_tx_reset_queue() from hns3_clear_all_ring(), and unifies hns3_clear_all_ring() and hns3_force_clear_all_ring into one, since they are doing similar things.

Fixes 3a30964a2eef ("net: hns3: delay ring buffer clearing during reset")
---
 .../net/ethernet/hisilicon/hns3/hns3_enet.c   | 54 +++++++++----------
 1 file changed, 26 insertions(+), 28 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +26 -28
index 7e7c10513d2c..ecf6ad5bdc2d 100644
@@ -27,8 +27,7 @@
 #define hns3_set_field(origin, shift, val)	((origin) |= ((val) << (shift)))
 #define hns3_tx_bd_count(S)	DIV_ROUND_UP(S, HNS3_MAX_BD_SIZE)
 
-static void hns3_clear_all_ring(struct hnae3_handle *h);
-static void hns3_force_clear_all_ring(struct hnae3_handle *h);
+static void hns3_clear_all_ring(struct hnae3_handle *h, bool force);
 static void hns3_remove_hw_addr(struct net_device *netdev);
 
 static const char hns3_driver_name[] = "hns3";
@@ -459,6 +458,20 @@ static int hns3_nic_net_open(struct net_device *netdev)
 	return 0;
 }
 
+static void hns3_reset_tx_queue(struct hnae3_handle *h)
+{
+	struct net_device *ndev = h->kinfo.netdev;
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	struct netdev_queue *dev_queue;
+	u32 i;
+
+	for (i = 0; i < h->kinfo.num_tqps; i++) {
+		dev_queue = netdev_get_tx_queue(ndev,
+						priv->ring_data[i].queue_index);
+		netdev_tx_reset_queue(dev_queue);
+	}
+}
+
 static void hns3_nic_net_down(struct net_device *netdev)
 {
 	struct hns3_nic_priv *priv = netdev_priv(netdev);
@@ -489,7 +502,9 @@ static void hns3_nic_net_down(struct net_device *netdev)
 	 * to disable the ring through firmware when downing the netdev.
 	 */
 	if (!hns3_nic_resetting(netdev))
-		hns3_clear_all_ring(priv->ae_handle);
+		hns3_clear_all_ring(priv->ae_handle, false);
+
+	hns3_reset_tx_queue(priv->ae_handle);
 }
 
 static int hns3_nic_net_stop(struct net_device *netdev)
@@ -3742,7 +3757,7 @@ static void hns3_client_uninit(struct hnae3_handle *handle, bool reset)
 
 	hns3_del_all_fd_rules(netdev, true);
 
-	hns3_force_clear_all_ring(handle);
+	hns3_clear_all_ring(handle, true);
 
 	hns3_uninit_phy(netdev);
 
@@ -3914,43 +3929,26 @@ static void hns3_force_clear_rx_ring(struct hns3_enet_ring *ring)
 	}
 }
 
-static void hns3_force_clear_all_ring(struct hnae3_handle *h)
-{
-	struct net_device *ndev = h->kinfo.netdev;
-	struct hns3_nic_priv *priv = netdev_priv(ndev);
-	struct hns3_enet_ring *ring;
-	u32 i;
-
-	for (i = 0; i < h->kinfo.num_tqps; i++) {
-		ring = priv->ring_data[i].ring;
-		hns3_clear_tx_ring(ring);
-
-		ring = priv->ring_data[i + h->kinfo.num_tqps].ring;
-		hns3_force_clear_rx_ring(ring);
-	}
-}
-
-static void hns3_clear_all_ring(struct hnae3_handle *h)
+static void hns3_clear_all_ring(struct hnae3_handle *h, bool force)
 {
 	struct net_device *ndev = h->kinfo.netdev;
 	struct hns3_nic_priv *priv = netdev_priv(ndev);
 	u32 i;
 
 	for (i = 0; i < h->kinfo.num_tqps; i++) {
-		struct netdev_queue *dev_queue;
 		struct hns3_enet_ring *ring;
 
 		ring = priv->ring_data[i].ring;
 		hns3_clear_tx_ring(ring);
-		dev_queue = netdev_get_tx_queue(ndev,
-						priv->ring_data[i].queue_index);
-		netdev_tx_reset_queue(dev_queue);
 
 		ring = priv->ring_data[i + h->kinfo.num_tqps].ring;
 		/* Continue to clear other rings even if clearing some
 		 * rings failed.
 		 */
-		hns3_clear_rx_ring(ring);
+		if (force)
+			hns3_force_clear_rx_ring(ring);
+		else
+			hns3_clear_rx_ring(ring);
 	}
 }
 
@@ -4153,8 +4151,8 @@ static int hns3_reset_notify_uninit_enet(struct hnae3_handle *handle)
 		return 0;
 	}
 
-	hns3_clear_all_ring(handle);
-	hns3_force_clear_all_ring(handle);
+	hns3_clear_all_ring(handle, true);
+	hns3_reset_tx_queue(priv->ae_handle);
 
 	hns3_nic_uninit_vector_data(priv);
 
Signature · 1 line
2.20.1
author maintainer
#222 lore raw

PATCH 221/371 iavf: fix dereference of null rx_buffer pointer

In reply to Greg Kroah-Hartman

[ Upstream commit 9fe06a51287b2d41baef7ece94df34b5abf19b90 ]

A recent commit efa14c3985828d ("iavf: allow null RX descriptors") added a null pointer sanity check on rx_buffer, however, rx_buffer is being dereferenced before that check, which implies a null pointer dereference bug can potentially occur. Fix this by only dereferencing rx_buffer until after the null pointer check.

Addresses-Coverity: ("Dereference before null check")

---
 drivers/net/ethernet/intel/iavf/iavf_txrx.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/net/ethernet/intel/iavf/iavf_txrx.c +4 -2
index 9cc2a617c9f3..2a261d849d5a 100644
@@ -1296,7 +1296,7 @@ static struct sk_buff *iavf_construct_skb(struct iavf_ring *rx_ring,
 					  struct iavf_rx_buffer *rx_buffer,
 					  unsigned int size)
 {
-	void *va = page_address(rx_buffer->page) + rx_buffer->page_offset;
+	void *va;
 #if (PAGE_SIZE < 8192)
 	unsigned int truesize = iavf_rx_pg_size(rx_ring) / 2;
 #else
@@ -1308,6 +1308,7 @@ static struct sk_buff *iavf_construct_skb(struct iavf_ring *rx_ring,
 	if (!rx_buffer)
 		return NULL;
 	/* prefetch first cache line of first page */
+	va = page_address(rx_buffer->page) + rx_buffer->page_offset;
 	prefetch(va);
 #if L1_CACHE_BYTES < 128
 	prefetch(va + L1_CACHE_BYTES);
@@ -1362,7 +1363,7 @@ static struct sk_buff *iavf_build_skb(struct iavf_ring *rx_ring,
 				      struct iavf_rx_buffer *rx_buffer,
 				      unsigned int size)
 {
-	void *va = page_address(rx_buffer->page) + rx_buffer->page_offset;
+	void *va;
 #if (PAGE_SIZE < 8192)
 	unsigned int truesize = iavf_rx_pg_size(rx_ring) / 2;
 #else
@@ -1374,6 +1375,7 @@ static struct sk_buff *iavf_build_skb(struct iavf_ring *rx_ring,
 	if (!rx_buffer)
 		return NULL;
 	/* prefetch first cache line of first page */
+	va = page_address(rx_buffer->page) + rx_buffer->page_offset;
 	prefetch(va);
 #if L1_CACHE_BYTES < 128
 	prefetch(va + L1_CACHE_BYTES);
Signature · 1 line
2.20.1
author maintainer
#223 lore raw

PATCH 222/371 blk-iolatency: fix STS_AGAIN handling

In reply to Greg Kroah-Hartman

[ Upstream commit c9b3007feca018d3f7061f5d5a14cb00766ffe9b ]

The iolatency controller is based on rq_qos. It increments on rq_qos_throttle() and decrements on either rq_qos_cleanup() or rq_qos_done_bio(). a3fb01ba5af0 fixes the double accounting issue where blk_mq_make_request() may call both rq_qos_cleanup() and rq_qos_done_bio() on REQ_NO_WAIT. So checking STS_AGAIN prevents the double decrement.

The above works upstream as the only way we can get STS_AGAIN is from blk_mq_get_request() failing. The STS_AGAIN handling isn't a real problem as bio_endio() skipping only happens on reserved tag allocation failures which can only be caused by driver bugs and already triggers WARN.

However, the fix creates a not so great dependency on how STS_AGAIN can be propagated. Internally, we (Facebook) carry a patch that kills read ahead if a cgroup is io congested or a fatal signal is pending. This combined with chained bios progagate their bi_status to the parent is not already set can can cause the parent bio to not clean up properly even though it was successful. This consequently leaks the inflight counter and can hang all IOs under that blkg.

To nip the adverse interaction early, this removes the rq_qos_cleanup() callback in iolatency in favor of cleaning up always on the rq_qos_done_bio() path.

Fixes a3fb01ba5af0 ("blk-iolatency: only account submitted bios")
---
 block/blk-iolatency.c | 51 ++++++++++++-------------------------------
 1 file changed, 14 insertions(+), 37 deletions(-)
block/blk-iolatency.c +14 -37
index 58bac44ba78a..072e1edcf83d 100644
@@ -599,10 +599,6 @@ static void blkcg_iolatency_done_bio(struct rq_qos *rqos, struct bio *bio)
 	if (!blkg || !bio_flagged(bio, BIO_TRACKED))
 		return;
 
-	/* We didn't actually submit this bio, don't account it. */
-	if (bio->bi_status == BLK_STS_AGAIN)
-		return;
-
 	iolat = blkg_to_lat(bio->bi_blkg);
 	if (!iolat)
 		return;
@@ -621,40 +617,22 @@ static void blkcg_iolatency_done_bio(struct rq_qos *rqos, struct bio *bio)
 
 		inflight = atomic_dec_return(&rqw->inflight);
 		WARN_ON_ONCE(inflight < 0);
-		if (iolat->min_lat_nsec == 0)
-			goto next;
-		iolatency_record_time(iolat, &bio->bi_issue, now,
-				      issue_as_root);
-		window_start = atomic64_read(&iolat->window_start);
-		if (now > window_start &&
-		    (now - window_start) >= iolat->cur_win_nsec) {
-			if (atomic64_cmpxchg(&iolat->window_start,
-					window_start, now) == window_start)
-				iolatency_check_latencies(iolat, now);
+		/*
+		 * If bi_status is BLK_STS_AGAIN, the bio wasn't actually
+		 * submitted, so do not account for it.
+		 */
+		if (iolat->min_lat_nsec && bio->bi_status != BLK_STS_AGAIN) {
+			iolatency_record_time(iolat, &bio->bi_issue, now,
+					      issue_as_root);
+			window_start = atomic64_read(&iolat->window_start);
+			if (now > window_start &&
+			    (now - window_start) >= iolat->cur_win_nsec) {
+				if (atomic64_cmpxchg(&iolat->window_start,
+					     window_start, now) == window_start)
+					iolatency_check_latencies(iolat, now);
+			}
 		}
-next:
-		wake_up(&rqw->wait);
-		blkg = blkg->parent;
-	}
-}
-
-static void blkcg_iolatency_cleanup(struct rq_qos *rqos, struct bio *bio)
-{
-	struct blkcg_gq *blkg;
-
-	blkg = bio->bi_blkg;
-	while (blkg && blkg->parent) {
-		struct rq_wait *rqw;
-		struct iolatency_grp *iolat;
-
-		iolat = blkg_to_lat(blkg);
-		if (!iolat)
-			goto next;
-
-		rqw = &iolat->rq_wait;
-		atomic_dec(&rqw->inflight);
 		wake_up(&rqw->wait);
-next:
 		blkg = blkg->parent;
 	}
 }
@@ -670,7 +648,6 @@ static void blkcg_iolatency_exit(struct rq_qos *rqos)
 
 static struct rq_qos_ops blkcg_iolatency_ops = {
 	.throttle = blkcg_iolatency_throttle,
-	.cleanup = blkcg_iolatency_cleanup,
 	.done_bio = blkcg_iolatency_done_bio,
 	.exit = blkcg_iolatency_exit,
 };
Signature · 1 line
2.20.1
author maintainer
#224 lore raw

PATCH 223/371 libbpf: fix another GCC8 warning for strncpy

In reply to Greg Kroah-Hartman

[ Upstream commit 763ff0e7d9c72e7094b31e7fb84a859be9325635 ]

Similar issue was fixed in cdfc7f888c2a ("libbpf: fix GCC8 warning for strncpy") already. This one was missed. Fixing now.

---
 tools/lib/bpf/xsk.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
tools/lib/bpf/xsk.c +2 -1
index 1fe0e1eec738..4ecd33ff46ec 100644
@@ -560,7 +560,8 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 		err = -errno;
 		goto out_socket;
 	}
-	strncpy(xsk->ifname, ifname, IFNAMSIZ);
+	strncpy(xsk->ifname, ifname, IFNAMSIZ - 1);
+	xsk->ifname[IFNAMSIZ - 1] = '\0';
 
 	err = xsk_set_xdp_socket_config(&xsk->config, usr_config);
 	if (err)
Signature · 1 line
2.20.1
author maintainer
#225 lore raw

PATCH 224/371 floppy: fix div-by-zero in setup_format_params

In reply to Greg Kroah-Hartman

[ Upstream commit f3554aeb991214cbfafd17d55e2bfddb50282e32 ]

This fixes a divide by zero error in the setup_format_params function of the floppy driver.

Two consecutive ioctls can trigger the bug: The first one should set the drive geometry with such .sect and .rate values for the F_SECT_PER_TRACK to become zero. Next, the floppy format operation should be called.

A floppy disk is not required to be inserted. An unprivileged user could trigger the bug if the device is accessible.

The patch checks F_SECT_PER_TRACK for a non-zero value in the set_geometry function. The proper check should involve a reasonable upper limit for the .sect and .rate fields, but it could change the UAPI.

The patch also checks F_SECT_PER_TRACK in the setup_format_params, and cancels the formatting operation in case of zero.

The bug was found by syzkaller.

---
 drivers/block/floppy.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/block/floppy.c +5 -0
index 95f608d1a098..e2f742e17683 100644
@@ -2119,6 +2119,9 @@ static void setup_format_params(int track)
 	raw_cmd->kernel_data = floppy_track_buffer;
 	raw_cmd->length = 4 * F_SECT_PER_TRACK;
 
+	if (!F_SECT_PER_TRACK)
+		return;
+
 	/* allow for about 30ms for data transport per track */
 	head_shift = (F_SECT_PER_TRACK + 5) / 6;
 
@@ -3231,6 +3234,8 @@ static int set_geometry(unsigned int cmd, struct floppy_struct *g,
 	/* sanity checking for parameters. */
 	if (g->sect <= 0 ||
 	    g->head <= 0 ||
+	    /* check for zero in F_SECT_PER_TRACK */
+	    (unsigned char)((g->sect << 2) >> FD_SIZECODE(g)) == 0 ||
 	    g->track <= 0 || g->track > UDP->tracks >> STRETCH(g) ||
 	    /* check if reserved bits are set */
 	    (g->stretch & ~(FD_STRETCH | FD_SWAPSIDES | FD_SECTBASEMASK)) != 0)
Signature · 1 line
2.20.1
author maintainer
#226 lore raw

PATCH 225/371 floppy: fix out-of-bounds read in next_valid_format

In reply to Greg Kroah-Hartman

[ Upstream commit 5635f897ed83fd539df78e98ba69ee91592f9bb8 ]

This fixes a global out-of-bounds read access in the next_valid_format function of the floppy driver.

The values from autodetect field of the struct floppy_drive_params are used as indices for the floppy_type array in the next_valid_format function 'floppy_type[DP->autodetect[probed_format]].sect'.

To trigger the bug, one could use a value out of range and set the drive parameters with the FDSETDRVPRM ioctl. A floppy disk is not required to be inserted.

CAP_SYS_ADMIN is required to call FDSETDRVPRM.

The patch adds the check for values of the autodetect field to be in the '0 <= x < ARRAY_SIZE(floppy_type)' range of the floppy_type array indices.

The bug was found by syzkaller.

---
 drivers/block/floppy.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)
drivers/block/floppy.c +18 -0
index e2f742e17683..77c527db5134 100644
@@ -3379,6 +3379,20 @@ static int fd_getgeo(struct block_device *bdev, struct hd_geometry *geo)
 	return 0;
 }
 
+static bool valid_floppy_drive_params(const short autodetect[8])
+{
+	size_t floppy_type_size = ARRAY_SIZE(floppy_type);
+	size_t i = 0;
+
+	for (i = 0; i < 8; ++i) {
+		if (autodetect[i] < 0 ||
+		    autodetect[i] >= floppy_type_size)
+			return false;
+	}
+
+	return true;
+}
+
 static int fd_locked_ioctl(struct block_device *bdev, fmode_t mode, unsigned int cmd,
 		    unsigned long param)
 {
@@ -3505,6 +3519,8 @@ static int fd_locked_ioctl(struct block_device *bdev, fmode_t mode, unsigned int
 		SUPBOUND(size, strlen((const char *)outparam) + 1);
 		break;
 	case FDSETDRVPRM:
+		if (!valid_floppy_drive_params(inparam.dp.autodetect))
+			return -EINVAL;
 		*UDP = inparam.dp;
 		break;
 	case FDGETDRVPRM:
@@ -3702,6 +3718,8 @@ static int compat_setdrvprm(int drive,
 		return -EPERM;
 	if (copy_from_user(&v, arg, sizeof(struct compat_floppy_drive_params)))
 		return -EFAULT;
+	if (!valid_floppy_drive_params(v.autodetect))
+		return -EINVAL;
 	mutex_lock(&floppy_mutex);
 	UDP->cmos = v.cmos;
 	UDP->max_dtr = v.max_dtr;
Signature · 1 line
2.20.1
author maintainer
#227 lore raw

PATCH 226/371 floppy: fix invalid pointer dereference in drive_name

In reply to Greg Kroah-Hartman

[ Upstream commit 9b04609b784027968348796a18f601aed9db3789 ]

This fixes the invalid pointer dereference in the drive_name function of the floppy driver.

The native_format field of the struct floppy_drive_params is used as floppy_type array index in the drive_name function. Thus, the field should be checked the same way as the autodetect field.

To trigger the bug, one could use a value out of range and set the drive parameters with the FDSETDRVPRM ioctl. Next, FDGETDRVTYP ioctl should be used to call the drive_name. A floppy disk is not required to be inserted.

CAP_SYS_ADMIN is required to call FDSETDRVPRM.

The patch adds the check for a value of the native_format field to be in the '0 <= x < ARRAY_SIZE(floppy_type)' range of the floppy_type array indices.

The bug was found by syzkaller.

---
 drivers/block/floppy.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)
drivers/block/floppy.c +8 -3
index 77c527db5134..8b443ee1d005 100644
@@ -3379,7 +3379,8 @@ static int fd_getgeo(struct block_device *bdev, struct hd_geometry *geo)
 	return 0;
 }
 
-static bool valid_floppy_drive_params(const short autodetect[8])
+static bool valid_floppy_drive_params(const short autodetect[8],
+		int native_format)
 {
 	size_t floppy_type_size = ARRAY_SIZE(floppy_type);
 	size_t i = 0;
@@ -3390,6 +3391,9 @@ static bool valid_floppy_drive_params(const short autodetect[8])
 			return false;
 	}
 
+	if (native_format < 0 || native_format >= floppy_type_size)
+		return false;
+
 	return true;
 }
 
@@ -3519,7 +3523,8 @@ static int fd_locked_ioctl(struct block_device *bdev, fmode_t mode, unsigned int
 		SUPBOUND(size, strlen((const char *)outparam) + 1);
 		break;
 	case FDSETDRVPRM:
-		if (!valid_floppy_drive_params(inparam.dp.autodetect))
+		if (!valid_floppy_drive_params(inparam.dp.autodetect,
+				inparam.dp.native_format))
 			return -EINVAL;
 		*UDP = inparam.dp;
 		break;
@@ -3718,7 +3723,7 @@ static int compat_setdrvprm(int drive,
 		return -EPERM;
 	if (copy_from_user(&v, arg, sizeof(struct compat_floppy_drive_params)))
 		return -EFAULT;
-	if (!valid_floppy_drive_params(v.autodetect))
+	if (!valid_floppy_drive_params(v.autodetect, v.native_format))
 		return -EINVAL;
 	mutex_lock(&floppy_mutex);
 	UDP->cmos = v.cmos;
Signature · 1 line
2.20.1
author maintainer
#228 lore raw

PATCH 227/371 floppy: fix out-of-bounds read in copy_buffer

In reply to Greg Kroah-Hartman

[ Upstream commit da99466ac243f15fbba65bd261bfc75ffa1532b6 ]

This fixes a global out-of-bounds read access in the copy_buffer function of the floppy driver.

The FDDEFPRM ioctl allows one to set the geometry of a disk. The sect and head fields (unsigned int) of the floppy_drive structure are used to compute the max_sector (int) in the make_raw_rw_request function. It is possible to overflow the max_sector. Next, max_sector is passed to the copy_buffer function and used in one of the memcpy calls.

An unprivileged user could trigger the bug if the device is accessible, but requires a floppy disk to be inserted.

The patch adds the check for the .sect * .head multiplication for not overflowing in the set_geometry function.

The bug was found by syzkaller.

---
 drivers/block/floppy.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/block/floppy.c +4 -2
index 8b443ee1d005..38e5811a045e 100644
@@ -3232,8 +3232,10 @@ static int set_geometry(unsigned int cmd, struct floppy_struct *g,
 	int cnt;
 
 	/* sanity checking for parameters. */
-	if (g->sect <= 0 ||
-	    g->head <= 0 ||
+	if ((int)g->sect <= 0 ||
+	    (int)g->head <= 0 ||
+	    /* check for overflow in max_sector */
+	    (int)(g->sect * g->head) <= 0 ||
 	    /* check for zero in F_SECT_PER_TRACK */
 	    (unsigned char)((g->sect << 2) >> FD_SIZECODE(g)) == 0 ||
 	    g->track <= 0 || g->track > UDP->tracks >> STRETCH(g) ||
Signature · 1 line
2.20.1
author maintainer
#229 lore raw

PATCH 228/371 xen: let alloc_xenballooned_pages() fail if not enough memory free

In reply to Greg Kroah-Hartman

From: Juergen Gross <[email protected]>

commit a1078e821b605813b63bf6bca414a85f804d5c66 upstream.

Instead of trying to allocate pages with GFP_USER in add_ballooned_pages() check the available free memory via si_mem_available(). GFP_USER is far less limiting memory exhaustion than the test via si_mem_available().

This will avoid dom0 running out of memory due to excessive foreign page mappings especially on ARM and on x86 in PVH mode, as those don't have a pre-ballooned area which can be used for foreign mappings.

As the normal ballooning suffers from the same problem don't balloon down more than si_mem_available() pages in one iteration. At the same time limit the default maximum number of retries.

This is part of XSA-300.

---
 drivers/xen/balloon.c |   16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)
drivers/xen/balloon.c +13 -3
@@ -538,8 +538,15 @@ static void balloon_process(struct work_
 				state = reserve_additional_memory();
 		}
 
-		if (credit < 0)
-			state = decrease_reservation(-credit, GFP_BALLOON);
+		if (credit < 0) {
+			long n_pages;
+
+			n_pages = min(-credit, si_mem_available());
+			state = decrease_reservation(n_pages, GFP_BALLOON);
+			if (state == BP_DONE && n_pages != -credit &&
+			    n_pages < totalreserve_pages)
+				state = BP_EAGAIN;
+		}
 
 		state = update_schedule(state);
 
@@ -578,6 +585,9 @@ static int add_ballooned_pages(int nr_pa
 		}
 	}
 
+	if (si_mem_available() < nr_pages)
+		return -ENOMEM;
+
 	st = decrease_reservation(nr_pages, GFP_USER);
 	if (st != BP_DONE)
 		return -ENOMEM;
@@ -710,7 +720,7 @@ static int __init balloon_init(void)
 	balloon_stats.schedule_delay = 1;
 	balloon_stats.max_schedule_delay = 32;
 	balloon_stats.retry_count = 1;
-	balloon_stats.max_retry_count = RETRY_UNLIMITED;
+	balloon_stats.max_retry_count = 4;
 
 #ifdef CONFIG_XEN_BALLOON_MEMORY_HOTPLUG
 	set_online_page_callback(&xen_online_page);
author maintainer
#230 lore raw

PATCH 229/371 scsi: NCR5380: Always re-enable reselection interrupt

In reply to Greg Kroah-Hartman

From: Finn Thain <[email protected]>

commit 57f31326518e98ee4cabf9a04efe00ed57c54147 upstream.

The reselection interrupt gets disabled during selection and must be re-enabled when hostdata->connected becomes NULL. If it isn't re-enabled a disconnected command may time-out or the target may wedge the bus while trying to reselect the host. This can happen after a command is aborted.

Fix this by enabling the reselection interrupt in NCR5380_main() after calls to NCR5380_select() and NCR5380_information_transfer() return.

Cc Michael Schmitz <[email protected]>
Fixes 8b00c3d5d40d ("ncr5380: Implement new eh_abort_handler")
---
 drivers/scsi/NCR5380.c |   12 ++----------
 1 file changed, 2 insertions(+), 10 deletions(-)
drivers/scsi/NCR5380.c +2 -10
@@ -710,6 +710,8 @@ static void NCR5380_main(struct work_str
 			NCR5380_information_transfer(instance);
 			done = 0;
 		}
+		if (!hostdata->connected)
+			NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
 		spin_unlock_irq(&hostdata->lock);
 		if (!done)
 			cond_resched();
@@ -1111,8 +1113,6 @@ static bool NCR5380_select(struct Scsi_H
 		spin_lock_irq(&hostdata->lock);
 		NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
 		NCR5380_reselect(instance);
-		if (!hostdata->connected)
-			NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
 		shost_printk(KERN_ERR, instance, "reselection after won arbitration?\n");
 		goto out;
 	}
@@ -1120,7 +1120,6 @@ static bool NCR5380_select(struct Scsi_H
 	if (err < 0) {
 		spin_lock_irq(&hostdata->lock);
 		NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
-		NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
 
 		/* Can't touch cmd if it has been reclaimed by the scsi ML */
 		if (!hostdata->selecting)
@@ -1158,7 +1157,6 @@ static bool NCR5380_select(struct Scsi_H
 	if (err < 0) {
 		shost_printk(KERN_ERR, instance, "select: REQ timeout\n");
 		NCR5380_write(INITIATOR_COMMAND_REG, ICR_BASE);
-		NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
 		goto out;
 	}
 	if (!hostdata->selecting) {
@@ -1827,9 +1825,6 @@ static void NCR5380_information_transfer
 					 */
 					NCR5380_write(TARGET_COMMAND_REG, 0);
 
-					/* Enable reselect interrupts */
-					NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
-
 					maybe_release_dma_irq(instance);
 					return;
 				case MESSAGE_REJECT:
@@ -1861,8 +1856,6 @@ static void NCR5380_information_transfer
 					 */
 					NCR5380_write(TARGET_COMMAND_REG, 0);
 
-					/* Enable reselect interrupts */
-					NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
 #ifdef SUN3_SCSI_VME
 					dregs->csr |= CSR_DMA_ENABLE;
 #endif
@@ -1965,7 +1958,6 @@ static void NCR5380_information_transfer
 					cmd->result = DID_ERROR << 16;
 					complete_cmd(instance, cmd);
 					maybe_release_dma_irq(instance);
-					NCR5380_write(SELECT_ENABLE_REG, hostdata->id_mask);
 					return;
 				}
 				msgout = NOP;
author maintainer
#231 lore raw

PATCH 230/371 scsi: NCR5380: Handle PDMA failure reliably

In reply to Greg Kroah-Hartman

From: Finn Thain <[email protected]>

commit f9dfed1c785734b95b08d67600e05d2092508ab0 upstream.

A PDMA error is handled in the core driver by setting the device's 'borken' flag and aborting the command. Unfortunately, do_abort() is not dependable. Perform a SCSI bus reset instead, to make sure that the command fails and gets retried.

---
 drivers/scsi/NCR5380.c |    6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
drivers/scsi/NCR5380.c +2 -4
@@ -1762,10 +1762,8 @@ static void NCR5380_information_transfer
 						scmd_printk(KERN_INFO, cmd,
 							"switching to slow handshake\n");
 						cmd->device->borken = 1;
-						sink = 1;
-						do_abort(instance);
-						cmd->result = DID_ERROR << 16;
-						/* XXX - need to source or sink data here, as appropriate */
+						do_reset(instance);
+						bus_reset_cleanup(instance);
 					}
 				} else {
 					/* Transfer a small chunk so that the
author maintainer
#232 lore raw

PATCH 231/371 Revert "scsi: ncr5380: Increase register polling limit"

In reply to Greg Kroah-Hartman

From: Finn Thain <[email protected]>

commit 25fcf94a2fa89dd3e73e965ebb0b38a2a4f72aa4 upstream.

This reverts commit 4822827a69d7cd3bc5a07b7637484ebd2cf88db6.

The purpose of that commit was to suppress a timeout warning message which appeared to be caused by target latency. But suppressing the warning is undesirable as the warning may indicate a messed up transfer count.

Another problem with that commit is that 15 ms is too long to keep interrupts disabled as interrupt latency can cause system clock drift and other problems.

Cc Michael Schmitz <[email protected]>
Fixes 4822827a69d7 ("scsi: ncr5380: Increase register polling limit")
---
 drivers/scsi/NCR5380.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/NCR5380.h +1 -1
@@ -235,7 +235,7 @@ struct NCR5380_cmd {
 #define NCR5380_PIO_CHUNK_SIZE		256
 
 /* Time limit (ms) to poll registers when IRQs are disabled, e.g. during PDMA */
-#define NCR5380_REG_POLL_TIME		15
+#define NCR5380_REG_POLL_TIME		10
 
 static inline struct scsi_cmnd *NCR5380_to_scmd(struct NCR5380_cmd *ncmd_ptr)
 {
author maintainer
#233 lore raw

PATCH 232/371 scsi: core: Fix race on creating sense cache

In reply to Greg Kroah-Hartman

From: Ming Lei <[email protected]>

commit f9b0530fa02e0c73f31a49ef743e8f44eb8e32cc upstream.

When scsi_init_sense_cache(host) is called concurrently from different hosts, each code path may find that no cache has been created and allocate a new one. The lack of locking can lead to potentially overriding a cache allocated by a different host.

Fix the issue by moving 'mutex_lock(&scsi_sense_cache_mutex)' before scsi_select_sense_cache().

Fixes 0a6ac4ee7c21 ("scsi: respect unchecked_isa_dma for blk-mq")
Cc Stable <[email protected]>
Cc Christoph Hellwig <[email protected]>
Cc Hannes Reinecke <[email protected]>
Cc Ewan D. Milne <[email protected]>
---
 drivers/scsi/scsi_lib.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/scsi/scsi_lib.c +3 -3
@@ -71,11 +71,11 @@ int scsi_init_sense_cache(struct Scsi_Ho
 	struct kmem_cache *cache;
 	int ret = 0;
 
+	mutex_lock(&scsi_sense_cache_mutex);
 	cache = scsi_select_sense_cache(shost->unchecked_isa_dma);
 	if (cache)
-		return 0;
+		goto exit;
 
-	mutex_lock(&scsi_sense_cache_mutex);
 	if (shost->unchecked_isa_dma) {
 		scsi_sense_isadma_cache =
 			kmem_cache_create("scsi_sense_cache(DMA)",
@@ -91,7 +91,7 @@ int scsi_init_sense_cache(struct Scsi_Ho
 		if (!scsi_sense_cache)
 			ret = -ENOMEM;
 	}
-
+ exit:
 	mutex_unlock(&scsi_sense_cache_mutex);
 	return ret;
 }
author maintainer
#234 lore raw

PATCH 233/371 scsi: sd_zbc: Fix compilation warning

In reply to Greg Kroah-Hartman

From: Damien Le Moal <[email protected]>

commit 0cdc58580b37a160fac4b884266b8b7cb096f539 upstream.

kbuild test robot gets the following compilation warning using gcc 7.4 cross compilation for c6x (GCC_VERSION=7.4.0 make.cross ARCH=c6x).

In file included from include/asm-generic/bug.h:18:0,

                    from arch/c6x/include/asm/bug.h:12,
                    from include/linux/bug.h:5,
                    from include/linux/thread_info.h:12,
                    from include/asm-generic/current.h:5,
                    from ./arch/c6x/include/generated/asm/current.h:1,
                    from include/linux/sched.h:12,
                    from include/linux/blkdev.h:5,
                    from drivers//scsi/sd_zbc.c:11:

drivers//scsi/sd_zbc.c: In function 'sd_zbc_read_zones':

Quoted 1 line
Quoted 1 line

include/linux/kernel.h:62:48: warning: 'zone_blocks' may be used

uninitialized in this function [-Wmaybe-uninitialized]

    #define __round_mask(x, y) ((__typeof__(x))((y)-1))
                                                   ^
   drivers//scsi/sd_zbc.c:464:6: note: 'zone_blocks' was declared here
     u32 zone_blocks;
         ^~~~~~~~~~~

This is a false-positive report. The variable zone_blocks is always initialized in sd_zbc_check_zones() before use. It is not initialized only and only if sd_zbc_check_zones() fails.

Avoid this warning by initializing the zone_blocks variable to 0.

Fixes 5f832a395859 ("scsi: sd_zbc: Fix sd_zbc_check_zones() error checks")
Cc Stable <[email protected]>
---
 drivers/scsi/sd_zbc.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/sd_zbc.c +1 -1
@@ -431,7 +431,7 @@ int sd_zbc_read_zones(struct scsi_disk *
 {
 	struct gendisk *disk = sdkp->disk;
 	unsigned int nr_zones;
-	u32 zone_blocks;
+	u32 zone_blocks = 0;
 	int ret;
 
 	if (!sd_is_zoned(sdkp))
author maintainer
#235 lore raw

PATCH 234/371 scsi: zfcp: fix request object use-after-free in send path causing seqno errors

In reply to Greg Kroah-Hartman

From: Benjamin Block <[email protected]>

commit b76becde2b84137faa29bbc9a3b20953b5980e48 upstream.

With a recent change to our send path for FSF commands we introduced a possible use-after-free of request-objects, that might further lead to zfcp crafting bad requests, which the FCP channel correctly complains about with an error (FSF_PROT_SEQ_NUMB_ERROR). This error is then handled by an adapter-wide recovery.

The following sequence illustrates the possible use-after-free:

    Send Path:

        int zfcp_fsf_open_port(struct zfcp_erp_action *erp_action)
        {
                struct zfcp_fsf_req *req;
                ...
                spin_lock_irq(&qdio->req_q_lock);
        //                     ^^^^^^^^^^^^^^^^
        //                     protects QDIO queue during sending
                ...
                req = zfcp_fsf_req_create(qdio,
                                          FSF_QTCB_OPEN_PORT_WITH_DID,
                                          SBAL_SFLAGS0_TYPE_READ,
                                          qdio->adapter->pool.erp_req);
        //            ^^^^^^^^^^^^^^^^^^^
        //            allocation of the request-object
                ...
                retval = zfcp_fsf_req_send(req);
                ...
                spin_unlock_irq(&qdio->req_q_lock);
                return retval;
        }

        static int zfcp_fsf_req_send(struct zfcp_fsf_req *req)
        {
                struct zfcp_adapter *adapter = req->adapter;
                struct zfcp_qdio *qdio = adapter->qdio;
                ...
                zfcp_reqlist_add(adapter->req_list, req);
        //      ^^^^^^^^^^^^^^^^
        //      add request to our driver-internal hash-table for tracking
        //      (protected by separate lock req_list->lock)
                ...
                if (zfcp_qdio_send(qdio, &req->qdio_req)) {
        //          ^^^^^^^^^^^^^^
        //          hand-off the request to FCP channel;
        //          the request can complete at any point now
                        ...
                }

                /* Don't increase for unsolicited status */
                if (!zfcp_fsf_req_is_status_read_buffer(req))
        //           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
        //           possible use-after-free
                        adapter->fsf_req_seq_no++;
        //                       ^^^^^^^^^^^^^^^^
        //                       because of the use-after-free we might
        //                       miss this accounting, and as follow-up
        //                       this results in the FCP channel error
        //                       FSF_PROT_SEQ_NUMB_ERROR
                adapter->req_no++;

                return 0;
        }

        static inline bool
        zfcp_fsf_req_is_status_read_buffer(struct zfcp_fsf_req *req)
        {
                return req->qtcb == NULL;
        //             ^^^^^^^^^
        //             possible use-after-free
        }

    Response Path:

        void zfcp_fsf_reqid_check(struct zfcp_qdio *qdio, int sbal_idx)
        {
                ...
                struct zfcp_fsf_req *fsf_req;
                ...
                for (idx = 0; idx < QDIO_MAX_ELEMENTS_PER_BUFFER; idx++) {
                        ...
                        fsf_req = zfcp_reqlist_find_rm(adapter->req_list,
                                                       req_id);
        //                        ^^^^^^^^^^^^^^^^^^^^
        //                        remove request from our driver-internal
        //                        hash-table (lock req_list->lock)
                        ...
                        zfcp_fsf_req_complete(fsf_req);
                }
        }

        static void zfcp_fsf_req_complete(struct zfcp_fsf_req *req)
        {
                ...
                if (likely(req->status & ZFCP_STATUS_FSFREQ_CLEANUP))
                        zfcp_fsf_req_free(req);
        //              ^^^^^^^^^^^^^^^^^
        //              free memory for request-object
                else
                        complete(&req->completion);
        //              ^^^^^^^^
        //              completion notification for code-paths that wait
        //              synchronous for the completion of the request; in
        //              those the memory is freed separately
        }

The result of the use-after-free only affects the send path, and can not lead to any data corruption. In case we miss the sequence-number accounting, because the memory was already re-purposed, the next FSF command will fail with said FCP channel error, and we will recover the whole adapter. This causes no additional errors, but it slows down traffic. There is a slight chance of the same thing happen again recursively after the adapter recovery, but so far this has not been seen.

This was seen under z/VM, where the send path might run on a virtual CPU that gets scheduled away by z/VM, while the return path might still run, and so create the necessary timing. Running with KASAN can also slow down the kernel sufficiently to run into this user-after-free, and then see the report by KASAN.

To fix this, simply pull the test for the sequence-number accounting in front of the hand-off to the FCP channel (this information doesn't change during hand-off), but leave the sequence-number accounting itself where it is.

To make future regressions of the same kind less likely, add comments to all closely related code-paths.

Fixes f9eca0227600 ("scsi: zfcp: drop duplicate fsf_command from zfcp_fsf_req which is also in QTCB header")
---
 drivers/s390/scsi/zfcp_fsf.c |   45 ++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 40 insertions(+), 5 deletions(-)
drivers/s390/scsi/zfcp_fsf.c +40 -5
@@ -11,6 +11,7 @@
 #define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
 
 #include <linux/blktrace_api.h>
+#include <linux/types.h>
 #include <linux/slab.h>
 #include <scsi/fc/fc_els.h>
 #include "zfcp_ext.h"
@@ -741,6 +742,7 @@ static struct zfcp_fsf_req *zfcp_fsf_req
 
 static int zfcp_fsf_req_send(struct zfcp_fsf_req *req)
 {
+	const bool is_srb = zfcp_fsf_req_is_status_read_buffer(req);
 	struct zfcp_adapter *adapter = req->adapter;
 	struct zfcp_qdio *qdio = adapter->qdio;
 	int req_id = req->req_id;
@@ -757,8 +759,20 @@ static int zfcp_fsf_req_send(struct zfcp
 		return -EIO;
 	}
 
+	/*
+	 * NOTE: DO NOT TOUCH ASYNC req PAST THIS POINT.
+	 *	 ONLY TOUCH SYNC req AGAIN ON req->completion.
+	 *
+	 * The request might complete and be freed concurrently at any point
+	 * now. This is not protected by the QDIO-lock (req_q_lock). So any
+	 * uncontrolled access after this might result in an use-after-free bug.
+	 * Only if the request doesn't have ZFCP_STATUS_FSFREQ_CLEANUP set, and
+	 * when it is completed via req->completion, is it safe to use req
+	 * again.
+	 */
+
 	/* Don't increase for unsolicited status */
-	if (!zfcp_fsf_req_is_status_read_buffer(req))
+	if (!is_srb)
 		adapter->fsf_req_seq_no++;
 	adapter->req_no++;
 
@@ -805,6 +819,7 @@ int zfcp_fsf_status_read(struct zfcp_qdi
 	retval = zfcp_fsf_req_send(req);
 	if (retval)
 		goto failed_req_send;
+	/* NOTE: DO NOT TOUCH req PAST THIS POINT! */
 
 	goto out;
 
@@ -914,8 +929,10 @@ struct zfcp_fsf_req *zfcp_fsf_abort_fcp_
 	req->qtcb->bottom.support.req_handle = (u64) old_req_id;
 
 	zfcp_fsf_start_timer(req, ZFCP_FSF_SCSI_ER_TIMEOUT);
-	if (!zfcp_fsf_req_send(req))
+	if (!zfcp_fsf_req_send(req)) {
+		/* NOTE: DO NOT TOUCH req, UNTIL IT COMPLETES! */
 		goto out;
+	}
 
 out_error_free:
 	zfcp_fsf_req_free(req);
@@ -1098,6 +1115,7 @@ int zfcp_fsf_send_ct(struct zfcp_fc_wka_
 	ret = zfcp_fsf_req_send(req);
 	if (ret)
 		goto failed_send;
+	/* NOTE: DO NOT TOUCH req PAST THIS POINT! */
 
 	goto out;
 
@@ -1198,6 +1216,7 @@ int zfcp_fsf_send_els(struct zfcp_adapte
 	ret = zfcp_fsf_req_send(req);
 	if (ret)
 		goto failed_send;
+	/* NOTE: DO NOT TOUCH req PAST THIS POINT! */
 
 	goto out;
 
@@ -1243,6 +1262,7 @@ int zfcp_fsf_exchange_config_data(struct
 		zfcp_fsf_req_free(req);
 		erp_action->fsf_req_id = 0;
 	}
+	/* NOTE: DO NOT TOUCH req PAST THIS POINT! */
 out:
 	spin_unlock_irq(&qdio->req_q_lock);
 	return retval;
@@ -1279,8 +1299,10 @@ int zfcp_fsf_exchange_config_data_sync(s
 	zfcp_fsf_start_timer(req, ZFCP_FSF_REQUEST_TIMEOUT);
 	retval = zfcp_fsf_req_send(req);
 	spin_unlock_irq(&qdio->req_q_lock);
-	if (!retval)
+	if (!retval) {
+		/* NOTE: ONLY TOUCH SYNC req AGAIN ON req->completion. */
 		wait_for_completion(&req->completion);
+	}
 
 	zfcp_fsf_req_free(req);
 	return retval;
@@ -1330,6 +1352,7 @@ int zfcp_fsf_exchange_port_data(struct z
 		zfcp_fsf_req_free(req);
 		erp_action->fsf_req_id = 0;
 	}
+	/* NOTE: DO NOT TOUCH req PAST THIS POINT! */
 out:
 	spin_unlock_irq(&qdio->req_q_lock);
 	return retval;
@@ -1372,8 +1395,10 @@ int zfcp_fsf_exchange_port_data_sync(str
 	retval = zfcp_fsf_req_send(req);
 	spin_unlock_irq(&qdio->req_q_lock);
 
-	if (!retval)
+	if (!retval) {
+		/* NOTE: ONLY TOUCH SYNC req AGAIN ON req->completion. */
 		wait_for_completion(&req->completion);
+	}
 
 	zfcp_fsf_req_free(req);
 
@@ -1493,6 +1518,7 @@ int zfcp_fsf_open_port(struct zfcp_erp_a
 		erp_action->fsf_req_id = 0;
 		put_device(&port->dev);
 	}
+	/* NOTE: DO NOT TOUCH req PAST THIS POINT! */
 out:
 	spin_unlock_irq(&qdio->req_q_lock);
 	return retval;
@@ -1557,6 +1583,7 @@ int zfcp_fsf_close_port(struct zfcp_erp_
 		zfcp_fsf_req_free(req);
 		erp_action->fsf_req_id = 0;
 	}
+	/* NOTE: DO NOT TOUCH req PAST THIS POINT! */
 out:
 	spin_unlock_irq(&qdio->req_q_lock);
 	return retval;
@@ -1626,6 +1653,7 @@ int zfcp_fsf_open_wka_port(struct zfcp_f
 	retval = zfcp_fsf_req_send(req);
 	if (retval)
 		zfcp_fsf_req_free(req);
+	/* NOTE: DO NOT TOUCH req PAST THIS POINT! */
 out:
 	spin_unlock_irq(&qdio->req_q_lock);
 	if (!retval)
@@ -1681,6 +1709,7 @@ int zfcp_fsf_close_wka_port(struct zfcp_
 	retval = zfcp_fsf_req_send(req);
 	if (retval)
 		zfcp_fsf_req_free(req);
+	/* NOTE: DO NOT TOUCH req PAST THIS POINT! */
 out:
 	spin_unlock_irq(&qdio->req_q_lock);
 	if (!retval)
@@ -1776,6 +1805,7 @@ int zfcp_fsf_close_physical_port(struct
 		zfcp_fsf_req_free(req);
 		erp_action->fsf_req_id = 0;
 	}
+	/* NOTE: DO NOT TOUCH req PAST THIS POINT! */
 out:
 	spin_unlock_irq(&qdio->req_q_lock);
 	return retval;
@@ -1899,6 +1929,7 @@ int zfcp_fsf_open_lun(struct zfcp_erp_ac
 		zfcp_fsf_req_free(req);
 		erp_action->fsf_req_id = 0;
 	}
+	/* NOTE: DO NOT TOUCH req PAST THIS POINT! */
 out:
 	spin_unlock_irq(&qdio->req_q_lock);
 	return retval;
@@ -1987,6 +2018,7 @@ int zfcp_fsf_close_lun(struct zfcp_erp_a
 		zfcp_fsf_req_free(req);
 		erp_action->fsf_req_id = 0;
 	}
+	/* NOTE: DO NOT TOUCH req PAST THIS POINT! */
 out:
 	spin_unlock_irq(&qdio->req_q_lock);
 	return retval;
@@ -2299,6 +2331,7 @@ int zfcp_fsf_fcp_cmnd(struct scsi_cmnd *
 	retval = zfcp_fsf_req_send(req);
 	if (unlikely(retval))
 		goto failed_scsi_cmnd;
+	/* NOTE: DO NOT TOUCH req PAST THIS POINT! */
 
 	goto out;
 
@@ -2373,8 +2406,10 @@ struct zfcp_fsf_req *zfcp_fsf_fcp_task_m
 	zfcp_fc_fcp_tm(fcp_cmnd, sdev, tm_flags);
 
 	zfcp_fsf_start_timer(req, ZFCP_FSF_SCSI_ER_TIMEOUT);
-	if (!zfcp_fsf_req_send(req))
+	if (!zfcp_fsf_req_send(req)) {
+		/* NOTE: DO NOT TOUCH req, UNTIL IT COMPLETES! */
 		goto out;
+	}
 
 	zfcp_fsf_req_free(req);
 	req = NULL;
author maintainer
#236 lore raw

PATCH 235/371 scsi: zfcp: fix request object use-after-free in send path causing wrong traces

In reply to Greg Kroah-Hartman

From: Benjamin Block <[email protected]>

commit 106d45f350c7cac876844dc685845cba4ffdb70b upstream.

When tracing instances where we open and close WKA ports, we also pass the request-ID of the respective FSF command.

But after successfully sending the FSF command we must not use the request-object anymore, as this might result in an use-after-free (see "zfcp: fix request object use-after-free in send path causing seqno errors" ).

To fix this add a new variable that caches the request-ID before sending the request. This won't change during the hand-off to the FCP channel, and so it's safe to trace this cached request-ID later, instead of using the request object.

Fixes d27a7cb91960 ("zfcp: trace on request for open and close of WKA port")
Cc <[email protected]> #2.6.38+
---
 drivers/s390/scsi/zfcp_fsf.c |   10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
drivers/s390/scsi/zfcp_fsf.c +8 -2
@@ -1627,6 +1627,7 @@ int zfcp_fsf_open_wka_port(struct zfcp_f
 {
 	struct zfcp_qdio *qdio = wka_port->adapter->qdio;
 	struct zfcp_fsf_req *req;
+	unsigned long req_id = 0;
 	int retval = -EIO;
 
 	spin_lock_irq(&qdio->req_q_lock);
@@ -1649,6 +1650,8 @@ int zfcp_fsf_open_wka_port(struct zfcp_f
 	hton24(req->qtcb->bottom.support.d_id, wka_port->d_id);
 	req->data = wka_port;
 
+	req_id = req->req_id;
+
 	zfcp_fsf_start_timer(req, ZFCP_FSF_REQUEST_TIMEOUT);
 	retval = zfcp_fsf_req_send(req);
 	if (retval)
@@ -1657,7 +1660,7 @@ int zfcp_fsf_open_wka_port(struct zfcp_f
 out:
 	spin_unlock_irq(&qdio->req_q_lock);
 	if (!retval)
-		zfcp_dbf_rec_run_wka("fsowp_1", wka_port, req->req_id);
+		zfcp_dbf_rec_run_wka("fsowp_1", wka_port, req_id);
 	return retval;
 }
 
@@ -1683,6 +1686,7 @@ int zfcp_fsf_close_wka_port(struct zfcp_
 {
 	struct zfcp_qdio *qdio = wka_port->adapter->qdio;
 	struct zfcp_fsf_req *req;
+	unsigned long req_id = 0;
 	int retval = -EIO;
 
 	spin_lock_irq(&qdio->req_q_lock);
@@ -1705,6 +1709,8 @@ int zfcp_fsf_close_wka_port(struct zfcp_
 	req->data = wka_port;
 	req->qtcb->header.port_handle = wka_port->handle;
 
+	req_id = req->req_id;
+
 	zfcp_fsf_start_timer(req, ZFCP_FSF_REQUEST_TIMEOUT);
 	retval = zfcp_fsf_req_send(req);
 	if (retval)
@@ -1713,7 +1719,7 @@ int zfcp_fsf_close_wka_port(struct zfcp_
 out:
 	spin_unlock_irq(&qdio->req_q_lock);
 	if (!retval)
-		zfcp_dbf_rec_run_wka("fscwp_1", wka_port, req->req_id);
+		zfcp_dbf_rec_run_wka("fscwp_1", wka_port, req_id);
 	return retval;
 }
 
author maintainer
#237 lore raw

PATCH 236/371 scsi: megaraid_sas: Fix calculation of target ID

In reply to Greg Kroah-Hartman

From: Shivasharan S <[email protected]>

commit c8f96df5b8e633056b7ebf5d52a9d6fb1b156ce3 upstream.

In megasas_get_target_prop(), driver is incorrectly calculating the target ID for devices with channel 1 and 3. Due to this, firmware will either fail the command (if there is no device with the target id sent from driver) or could return the properties for a target which was not intended. Devices could end up with the wrong queue depth due to this.

Fix target id calculation for channel 1 and 3.

Fixes 96188a89cc6d ("scsi: megaraid_sas: NVME interface target prop added")
---
 drivers/scsi/megaraid/megaraid_sas_base.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/scsi/megaraid/megaraid_sas_base.c +2 -1
@@ -6168,7 +6168,8 @@ megasas_get_target_prop(struct megasas_i
 	int ret;
 	struct megasas_cmd *cmd;
 	struct megasas_dcmd_frame *dcmd;
-	u16 targetId = (sdev->channel % 2) + sdev->id;
+	u16 targetId = ((sdev->channel % 2) * MEGASAS_MAX_DEV_PER_CHANNEL) +
+			sdev->id;
 
 	cmd = megasas_get_cmd(instance);
 
author maintainer
#238 lore raw

PATCH 237/371 scsi: mac_scsi: Increase PIO/PDMA transfer length threshold

In reply to Greg Kroah-Hartman

From: Finn Thain <[email protected]>

commit 7398cee4c3e6aea1ba07a6449e5533ecd0b92cdd upstream.

Some targets introduce delays when handshaking the response to certain commands. For example, a disk may send a 96-byte response to an INQUIRY command (or a 24-byte response to a MODE SENSE command) too slowly.

Apparently the first 12 or 14 bytes are handshaked okay but then the system bus error timeout is reached while transferring the next word.

Since the scsi bus phase hasn't changed, the driver then sets the target borken flag to prevent further PDMA transfers. The driver also logs the warning, "switching to slow handshake".

Raise the PDMA threshold to 512 bytes so that PIO transfers will be used for these commands. This default is sufficiently low that PDMA will still be used for READ and WRITE commands.

The existing threshold (16 bytes) was chosen more or less at random. However, best performance requires the threshold to be as low as possible. Those systems that don't need the PIO workaround at all may benefit from mac_scsi.setup_use_pdma=1

Cc Michael Schmitz <[email protected]>
Cc [email protected] # v4.14+
Fixes 3a0f64bfa907 ("mac_scsi: Fix pseudo DMA implementation")
---
 drivers/scsi/mac_scsi.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/scsi/mac_scsi.c +2 -2
@@ -52,7 +52,7 @@ static int setup_cmd_per_lun = -1;
 module_param(setup_cmd_per_lun, int, 0);
 static int setup_sg_tablesize = -1;
 module_param(setup_sg_tablesize, int, 0);
-static int setup_use_pdma = -1;
+static int setup_use_pdma = 512;
 module_param(setup_use_pdma, int, 0);
 static int setup_hostid = -1;
 module_param(setup_hostid, int, 0);
@@ -305,7 +305,7 @@ static int macscsi_dma_xfer_len(struct N
                                 struct scsi_cmnd *cmd)
 {
 	if (hostdata->flags & FLAG_NO_PSEUDO_DMA ||
-	    cmd->SCp.this_residual < 16)
+	    cmd->SCp.this_residual < setup_use_pdma)
 		return 0;
 
 	return cmd->SCp.this_residual;
author maintainer
#239 lore raw

PATCH 238/371 scsi: mac_scsi: Fix pseudo DMA implementation, take 2

In reply to Greg Kroah-Hartman

From: Finn Thain <[email protected]>

commit 78ff751f8e6a9446e9fb26b2bff0b8d3f8974cbd upstream.

A system bus error during a PDMA transfer can mess up the calculation of the transfer residual (the PDMA handshaking hardware lacks a byte counter). This results in data corruption.

The algorithm in this patch anticipates a bus error by starting each transfer with a MOVE.B instruction. If a bus error is caught the transfer will be retried. If a bus error is caught later in the transfer (for a MOVE.W instruction) the transfer gets failed and subsequent requests for that target will use PIO instead of PDMA.

This avoids the "!REQ and !ACK" error so the severity level of that message is reduced to KERN_DEBUG.

Cc Michael Schmitz <[email protected]>
Cc Geert Uytterhoeven <[email protected]>
Cc [email protected] # v4.14+
Fixes 3a0f64bfa907 ("mac_scsi: Fix pseudo DMA implementation")
---
 drivers/scsi/mac_scsi.c |  369 ++++++++++++++++++++++++++++--------------------
 1 file changed, 217 insertions(+), 152 deletions(-)
drivers/scsi/mac_scsi.c +217 -152
@@ -3,6 +3,8 @@
  *
  * Copyright 1998, Michael Schmitz <[email protected]>
  *
+ * Copyright 2019 Finn Thain
+ *
  * derived in part from:
  */
 /*
@@ -11,6 +13,7 @@
  * Copyright 1995, Russell King
  */
 
+#include <linux/delay.h>
 #include <linux/types.h>
 #include <linux/module.h>
 #include <linux/ioport.h>
@@ -89,101 +92,217 @@ static int __init mac_scsi_setup(char *s
 __setup("mac5380=", mac_scsi_setup);
 #endif /* !MODULE */
 
-/* Pseudo DMA asm originally by Ove Edlund */
+/*
+ * According to "Inside Macintosh: Devices", Mac OS requires disk drivers to
+ * specify the number of bytes between the delays expected from a SCSI target.
+ * This allows the operating system to "prevent bus errors when a target fails
+ * to deliver the next byte within the processor bus error timeout period."
+ * Linux SCSI drivers lack knowledge of the timing behaviour of SCSI targets
+ * so bus errors are unavoidable.
+ *
+ * If a MOVE.B instruction faults, we assume that zero bytes were transferred
+ * and simply retry. That assumption probably depends on target behaviour but
+ * seems to hold up okay. The NOP provides synchronization: without it the
+ * fault can sometimes occur after the program counter has moved past the
+ * offending instruction. Post-increment addressing can't be used.
+ */
+
+#define MOVE_BYTE(operands) \
+	asm volatile ( \
+		"1:     moveb " operands "     \n" \
+		"11:    nop                    \n" \
+		"       addq #1,%0             \n" \
+		"       subq #1,%1             \n" \
+		"40:                           \n" \
+		"                              \n" \
+		".section .fixup,\"ax\"        \n" \
+		".even                         \n" \
+		"90:    movel #1, %2           \n" \
+		"       jra 40b                \n" \
+		".previous                     \n" \
+		"                              \n" \
+		".section __ex_table,\"a\"     \n" \
+		".align  4                     \n" \
+		".long   1b,90b                \n" \
+		".long  11b,90b                \n" \
+		".previous                     \n" \
+		: "+a" (addr), "+r" (n), "+r" (result) : "a" (io))
 
-#define CP_IO_TO_MEM(s,d,n)				\
-__asm__ __volatile__					\
-    ("    cmp.w  #4,%2\n"				\
-     "    bls    8f\n"					\
-     "    move.w %1,%%d0\n"				\
-     "    neg.b  %%d0\n"				\
-     "    and.w  #3,%%d0\n"				\
-     "    sub.w  %%d0,%2\n"				\
-     "    bra    2f\n"					\
-     " 1: move.b (%0),(%1)+\n"				\
-     " 2: dbf    %%d0,1b\n"				\
-     "    move.w %2,%%d0\n"				\
-     "    lsr.w  #5,%%d0\n"				\
-     "    bra    4f\n"					\
-     " 3: move.l (%0),(%1)+\n"				\
-     "31: move.l (%0),(%1)+\n"				\
-     "32: move.l (%0),(%1)+\n"				\
-     "33: move.l (%0),(%1)+\n"				\
-     "34: move.l (%0),(%1)+\n"				\
-     "35: move.l (%0),(%1)+\n"				\
-     "36: move.l (%0),(%1)+\n"				\
-     "37: move.l (%0),(%1)+\n"				\
-     " 4: dbf    %%d0,3b\n"				\
-     "    move.w %2,%%d0\n"				\
-     "    lsr.w  #2,%%d0\n"				\
-     "    and.w  #7,%%d0\n"				\
-     "    bra    6f\n"					\
-     " 5: move.l (%0),(%1)+\n"				\
-     " 6: dbf    %%d0,5b\n"				\
-     "    and.w  #3,%2\n"				\
-     "    bra    8f\n"					\
-     " 7: move.b (%0),(%1)+\n"				\
-     " 8: dbf    %2,7b\n"				\
-     "    moveq.l #0, %2\n"				\
-     " 9: \n"						\
-     ".section .fixup,\"ax\"\n"				\
-     "    .even\n"					\
-     "91: moveq.l #1, %2\n"				\
-     "    jra 9b\n"					\
-     "94: moveq.l #4, %2\n"				\
-     "    jra 9b\n"					\
-     ".previous\n"					\
-     ".section __ex_table,\"a\"\n"			\
-     "   .align 4\n"					\
-     "   .long  1b,91b\n"				\
-     "   .long  3b,94b\n"				\
-     "   .long 31b,94b\n"				\
-     "   .long 32b,94b\n"				\
-     "   .long 33b,94b\n"				\
-     "   .long 34b,94b\n"				\
-     "   .long 35b,94b\n"				\
-     "   .long 36b,94b\n"				\
-     "   .long 37b,94b\n"				\
-     "   .long  5b,94b\n"				\
-     "   .long  7b,91b\n"				\
-     ".previous"					\
-     : "=a"(s), "=a"(d), "=d"(n)			\
-     : "0"(s), "1"(d), "2"(n)				\
-     : "d0")
+/*
+ * If a MOVE.W (or MOVE.L) instruction faults, it cannot be retried because
+ * the residual byte count would be uncertain. In that situation the MOVE_WORD
+ * macro clears n in the fixup section to abort the transfer.
+ */
+
+#define MOVE_WORD(operands) \
+	asm volatile ( \
+		"1:     movew " operands "     \n" \
+		"11:    nop                    \n" \
+		"       subq #2,%1             \n" \
+		"40:                           \n" \
+		"                              \n" \
+		".section .fixup,\"ax\"        \n" \
+		".even                         \n" \
+		"90:    movel #0, %1           \n" \
+		"       movel #2, %2           \n" \
+		"       jra 40b                \n" \
+		".previous                     \n" \
+		"                              \n" \
+		".section __ex_table,\"a\"     \n" \
+		".align  4                     \n" \
+		".long   1b,90b                \n" \
+		".long  11b,90b                \n" \
+		".previous                     \n" \
+		: "+a" (addr), "+r" (n), "+r" (result) : "a" (io))
+
+#define MOVE_16_WORDS(operands) \
+	asm volatile ( \
+		"1:     movew " operands "     \n" \
+		"2:     movew " operands "     \n" \
+		"3:     movew " operands "     \n" \
+		"4:     movew " operands "     \n" \
+		"5:     movew " operands "     \n" \
+		"6:     movew " operands "     \n" \
+		"7:     movew " operands "     \n" \
+		"8:     movew " operands "     \n" \
+		"9:     movew " operands "     \n" \
+		"10:    movew " operands "     \n" \
+		"11:    movew " operands "     \n" \
+		"12:    movew " operands "     \n" \
+		"13:    movew " operands "     \n" \
+		"14:    movew " operands "     \n" \
+		"15:    movew " operands "     \n" \
+		"16:    movew " operands "     \n" \
+		"17:    nop                    \n" \
+		"       subl  #32,%1           \n" \
+		"40:                           \n" \
+		"                              \n" \
+		".section .fixup,\"ax\"        \n" \
+		".even                         \n" \
+		"90:    movel #0, %1           \n" \
+		"       movel #2, %2           \n" \
+		"       jra 40b                \n" \
+		".previous                     \n" \
+		"                              \n" \
+		".section __ex_table,\"a\"     \n" \
+		".align  4                     \n" \
+		".long   1b,90b                \n" \
+		".long   2b,90b                \n" \
+		".long   3b,90b                \n" \
+		".long   4b,90b                \n" \
+		".long   5b,90b                \n" \
+		".long   6b,90b                \n" \
+		".long   7b,90b                \n" \
+		".long   8b,90b                \n" \
+		".long   9b,90b                \n" \
+		".long  10b,90b                \n" \
+		".long  11b,90b                \n" \
+		".long  12b,90b                \n" \
+		".long  13b,90b                \n" \
+		".long  14b,90b                \n" \
+		".long  15b,90b                \n" \
+		".long  16b,90b                \n" \
+		".long  17b,90b                \n" \
+		".previous                     \n" \
+		: "+a" (addr), "+r" (n), "+r" (result) : "a" (io))
+
+#define MAC_PDMA_DELAY		32
+
+static inline int mac_pdma_recv(void __iomem *io, unsigned char *start, int n)
+{
+	unsigned char *addr = start;
+	int result = 0;
+
+	if (n >= 1) {
+		MOVE_BYTE("%3@,%0@");
+		if (result)
+			goto out;
+	}
+	if (n >= 1 && ((unsigned long)addr & 1)) {
+		MOVE_BYTE("%3@,%0@");
+		if (result)
+			goto out;
+	}
+	while (n >= 32)
+		MOVE_16_WORDS("%3@,%0@+");
+	while (n >= 2)
+		MOVE_WORD("%3@,%0@+");
+	if (result)
+		return start - addr; /* Negated to indicate uncertain length */
+	if (n == 1)
+		MOVE_BYTE("%3@,%0@");
+out:
+	return addr - start;
+}
+
+static inline int mac_pdma_send(unsigned char *start, void __iomem *io, int n)
+{
+	unsigned char *addr = start;
+	int result = 0;
+
+	if (n >= 1) {
+		MOVE_BYTE("%0@,%3@");
+		if (result)
+			goto out;
+	}
+	if (n >= 1 && ((unsigned long)addr & 1)) {
+		MOVE_BYTE("%0@,%3@");
+		if (result)
+			goto out;
+	}
+	while (n >= 32)
+		MOVE_16_WORDS("%0@+,%3@");
+	while (n >= 2)
+		MOVE_WORD("%0@+,%3@");
+	if (result)
+		return start - addr; /* Negated to indicate uncertain length */
+	if (n == 1)
+		MOVE_BYTE("%0@,%3@");
+out:
+	return addr - start;
+}
 
 static inline int macscsi_pread(struct NCR5380_hostdata *hostdata,
                                 unsigned char *dst, int len)
 {
 	u8 __iomem *s = hostdata->pdma_io + (INPUT_DATA_REG << 4);
 	unsigned char *d = dst;
-	int n = len;
-	int transferred;
+
+	hostdata->pdma_residual = len;
 
 	while (!NCR5380_poll_politely(hostdata, BUS_AND_STATUS_REG,
 	                              BASR_DRQ | BASR_PHASE_MATCH,
 	                              BASR_DRQ | BASR_PHASE_MATCH, HZ / 64)) {
-		CP_IO_TO_MEM(s, d, n);
+		int bytes;
+
+		bytes = mac_pdma_recv(s, d, min(hostdata->pdma_residual, 512));
 
-		transferred = d - dst - n;
-		hostdata->pdma_residual = len - transferred;
+		if (bytes > 0) {
+			d += bytes;
+			hostdata->pdma_residual -= bytes;
+		}
 
-		/* No bus error. */
-		if (n == 0)
+		if (hostdata->pdma_residual == 0)
 			return 0;
 
-		/* Target changed phase early? */
 		if (NCR5380_poll_politely2(hostdata, STATUS_REG, SR_REQ, SR_REQ,
-		                           BUS_AND_STATUS_REG, BASR_ACK, BASR_ACK, HZ / 64) < 0)
-			scmd_printk(KERN_ERR, hostdata->connected,
+		                           BUS_AND_STATUS_REG, BASR_ACK,
+		                           BASR_ACK, HZ / 64) < 0)
+			scmd_printk(KERN_DEBUG, hostdata->connected,
 			            "%s: !REQ and !ACK\n", __func__);
 		if (!(NCR5380_read(BUS_AND_STATUS_REG) & BASR_PHASE_MATCH))
 			return 0;
 
+		if (bytes == 0)
+			udelay(MAC_PDMA_DELAY);
+
+		if (bytes >= 0)
+			continue;
+
 		dsprintk(NDEBUG_PSEUDO_DMA, hostdata->host,
-		         "%s: bus error (%d/%d)\n", __func__, transferred, len);
+		         "%s: bus error (%d/%d)\n", __func__, d - dst, len);
 		NCR5380_dprint(NDEBUG_PSEUDO_DMA, hostdata->host);
-		d = dst + transferred;
-		n = len - transferred;
+		return -1;
 	}
 
 	scmd_printk(KERN_ERR, hostdata->connected,
@@ -192,93 +311,27 @@ static inline int macscsi_pread(struct N
 	return -1;
 }
 
-
-#define CP_MEM_TO_IO(s,d,n)				\
-__asm__ __volatile__					\
-    ("    cmp.w  #4,%2\n"				\
-     "    bls    8f\n"					\
-     "    move.w %0,%%d0\n"				\
-     "    neg.b  %%d0\n"				\
-     "    and.w  #3,%%d0\n"				\
-     "    sub.w  %%d0,%2\n"				\
-     "    bra    2f\n"					\
-     " 1: move.b (%0)+,(%1)\n"				\
-     " 2: dbf    %%d0,1b\n"				\
-     "    move.w %2,%%d0\n"				\
-     "    lsr.w  #5,%%d0\n"				\
-     "    bra    4f\n"					\
-     " 3: move.l (%0)+,(%1)\n"				\
-     "31: move.l (%0)+,(%1)\n"				\
-     "32: move.l (%0)+,(%1)\n"				\
-     "33: move.l (%0)+,(%1)\n"				\
-     "34: move.l (%0)+,(%1)\n"				\
-     "35: move.l (%0)+,(%1)\n"				\
-     "36: move.l (%0)+,(%1)\n"				\
-     "37: move.l (%0)+,(%1)\n"				\
-     " 4: dbf    %%d0,3b\n"				\
-     "    move.w %2,%%d0\n"				\
-     "    lsr.w  #2,%%d0\n"				\
-     "    and.w  #7,%%d0\n"				\
-     "    bra    6f\n"					\
-     " 5: move.l (%0)+,(%1)\n"				\
-     " 6: dbf    %%d0,5b\n"				\
-     "    and.w  #3,%2\n"				\
-     "    bra    8f\n"					\
-     " 7: move.b (%0)+,(%1)\n"				\
-     " 8: dbf    %2,7b\n"				\
-     "    moveq.l #0, %2\n"				\
-     " 9: \n"						\
-     ".section .fixup,\"ax\"\n"				\
-     "    .even\n"					\
-     "91: moveq.l #1, %2\n"				\
-     "    jra 9b\n"					\
-     "94: moveq.l #4, %2\n"				\
-     "    jra 9b\n"					\
-     ".previous\n"					\
-     ".section __ex_table,\"a\"\n"			\
-     "   .align 4\n"					\
-     "   .long  1b,91b\n"				\
-     "   .long  3b,94b\n"				\
-     "   .long 31b,94b\n"				\
-     "   .long 32b,94b\n"				\
-     "   .long 33b,94b\n"				\
-     "   .long 34b,94b\n"				\
-     "   .long 35b,94b\n"				\
-     "   .long 36b,94b\n"				\
-     "   .long 37b,94b\n"				\
-     "   .long  5b,94b\n"				\
-     "   .long  7b,91b\n"				\
-     ".previous"					\
-     : "=a"(s), "=a"(d), "=d"(n)			\
-     : "0"(s), "1"(d), "2"(n)				\
-     : "d0")
-
 static inline int macscsi_pwrite(struct NCR5380_hostdata *hostdata,
                                  unsigned char *src, int len)
 {
 	unsigned char *s = src;
 	u8 __iomem *d = hostdata->pdma_io + (OUTPUT_DATA_REG << 4);
-	int n = len;
-	int transferred;
+
+	hostdata->pdma_residual = len;
 
 	while (!NCR5380_poll_politely(hostdata, BUS_AND_STATUS_REG,
 	                              BASR_DRQ | BASR_PHASE_MATCH,
 	                              BASR_DRQ | BASR_PHASE_MATCH, HZ / 64)) {
-		CP_MEM_TO_IO(s, d, n);
+		int bytes;
 
-		transferred = s - src - n;
-		hostdata->pdma_residual = len - transferred;
+		bytes = mac_pdma_send(s, d, min(hostdata->pdma_residual, 512));
 
-		/* Target changed phase early? */
-		if (NCR5380_poll_politely2(hostdata, STATUS_REG, SR_REQ, SR_REQ,
-		                           BUS_AND_STATUS_REG, BASR_ACK, BASR_ACK, HZ / 64) < 0)
-			scmd_printk(KERN_ERR, hostdata->connected,
-			            "%s: !REQ and !ACK\n", __func__);
-		if (!(NCR5380_read(BUS_AND_STATUS_REG) & BASR_PHASE_MATCH))
-			return 0;
+		if (bytes > 0) {
+			s += bytes;
+			hostdata->pdma_residual -= bytes;
+		}
 
-		/* No bus error. */
-		if (n == 0) {
+		if (hostdata->pdma_residual == 0) {
 			if (NCR5380_poll_politely(hostdata, TARGET_COMMAND_REG,
 			                          TCR_LAST_BYTE_SENT,
 			                          TCR_LAST_BYTE_SENT, HZ / 64) < 0)
@@ -287,17 +340,29 @@ static inline int macscsi_pwrite(struct
 			return 0;
 		}
 
+		if (NCR5380_poll_politely2(hostdata, STATUS_REG, SR_REQ, SR_REQ,
+		                           BUS_AND_STATUS_REG, BASR_ACK,
+		                           BASR_ACK, HZ / 64) < 0)
+			scmd_printk(KERN_DEBUG, hostdata->connected,
+			            "%s: !REQ and !ACK\n", __func__);
+		if (!(NCR5380_read(BUS_AND_STATUS_REG) & BASR_PHASE_MATCH))
+			return 0;
+
+		if (bytes == 0)
+			udelay(MAC_PDMA_DELAY);
+
+		if (bytes >= 0)
+			continue;
+
 		dsprintk(NDEBUG_PSEUDO_DMA, hostdata->host,
-		         "%s: bus error (%d/%d)\n", __func__, transferred, len);
+		         "%s: bus error (%d/%d)\n", __func__, s - src, len);
 		NCR5380_dprint(NDEBUG_PSEUDO_DMA, hostdata->host);
-		s = src + transferred;
-		n = len - transferred;
+		return -1;
 	}
 
 	scmd_printk(KERN_ERR, hostdata->connected,
 	            "%s: phase mismatch or !DRQ\n", __func__);
 	NCR5380_dprint(NDEBUG_PSEUDO_DMA, hostdata->host);
-
 	return -1;
 }
 
author maintainer
#240 lore raw

PATCH 239/371 crypto: ghash - fix unaligned memory access in ghash_setkey()

In reply to Greg Kroah-Hartman

From: Eric Biggers <[email protected]>

commit 5c6bc4dfa515738149998bb0db2481a4fdead979 upstream.

Changing ghash_mod_init() to be subsys_initcall made it start running before the alignment fault handler has been installed on ARM. In kernel builds where the keys in the ghash test vectors happened to be misaligned in the kernel image, this exposed the longstanding bug that ghash_setkey() is incorrectly casting the key buffer (which can have any alignment) to be128 for passing to gf128mul_init_4k_lle().

Fix this by memcpy()ing the key to a temporary buffer.

Don't fix it by setting an alignmask on the algorithm instead because that would unnecessarily force alignment of the data too.

Fixes 2cdc6899a88e ("crypto: ghash - Add GHASH digest algorithm for GCM")
---
 crypto/ghash-generic.c |    8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
crypto/ghash-generic.c +7 -1
@@ -34,6 +34,7 @@ static int ghash_setkey(struct crypto_sh
 			const u8 *key, unsigned int keylen)
 {
 	struct ghash_ctx *ctx = crypto_shash_ctx(tfm);
+	be128 k;
 
 	if (keylen != GHASH_BLOCK_SIZE) {
 		crypto_shash_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN);
@@ -42,7 +43,12 @@ static int ghash_setkey(struct crypto_sh
 
 	if (ctx->gf128)
 		gf128mul_free_4k(ctx->gf128);
-	ctx->gf128 = gf128mul_init_4k_lle((be128 *)key);
+
+	BUILD_BUG_ON(sizeof(k) != GHASH_BLOCK_SIZE);
+	memcpy(&k, key, GHASH_BLOCK_SIZE); /* avoid violating alignment rules */
+	ctx->gf128 = gf128mul_init_4k_lle(&k);
+	memzero_explicit(&k, GHASH_BLOCK_SIZE);
+
 	if (!ctx->gf128)
 		return -ENOMEM;
 
author maintainer
#241 lore raw

PATCH 240/371 crypto: caam - limit output IV to CBC to work around CTR mode DMA issue

In reply to Greg Kroah-Hartman

From: Ard Biesheuvel <[email protected]>

commit ed527b13d800dd515a9e6c582f0a73eca65b2e1b upstream.

The CAAM driver currently violates an undocumented and slightly controversial requirement imposed by the crypto stack that a buffer referred to by the request structure via its virtual address may not be modified while any scatterlists passed via the same request structure are mapped for inbound DMA.

This may result in errors like

alg: aead: decryption failed on test 1 for gcm_base(ctr-aes-caam,ghash-generic): ret=74
alg: aead: Failed to load transform for gcm(aes): -2

on non-cache coherent systems, due to the fact that the GCM driver passes an IV buffer by virtual address which shares a cacheline with the auth_tag buffer passed via a scatterlist, resulting in corruption of the auth_tag when the IV is updated while the DMA mapping is live.

Since the IV that is returned to the caller is only valid for CBC mode, and given that the in-kernel users of CBC (such as CTS) don't trigger the same issue as the GCM driver, let's just disable the output IV generation for all modes except CBC for the time being.

Fixes 854b06f76879 ("crypto: caam - properly set IV after {en,de}crypt")
Cc Horia Geanta <[email protected]>
Cc Iuliana Prodan <[email protected]>
---
 drivers/crypto/caam/caamalg.c |    9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)
drivers/crypto/caam/caamalg.c +5 -4
@@ -965,6 +965,7 @@ static void skcipher_encrypt_done(struct
 	struct skcipher_request *req = context;
 	struct skcipher_edesc *edesc;
 	struct crypto_skcipher *skcipher = crypto_skcipher_reqtfm(req);
+	struct caam_ctx *ctx = crypto_skcipher_ctx(skcipher);
 	int ivsize = crypto_skcipher_ivsize(skcipher);
 
 #ifdef DEBUG
@@ -989,9 +990,9 @@ static void skcipher_encrypt_done(struct
 
 	/*
 	 * The crypto API expects us to set the IV (req->iv) to the last
-	 * ciphertext block. This is used e.g. by the CTS mode.
+	 * ciphertext block when running in CBC mode.
 	 */
-	if (ivsize)
+	if ((ctx->cdata.algtype & OP_ALG_AAI_MASK) == OP_ALG_AAI_CBC)
 		scatterwalk_map_and_copy(req->iv, req->dst, req->cryptlen -
 					 ivsize, ivsize, 0);
 
@@ -1809,9 +1810,9 @@ static int skcipher_decrypt(struct skcip
 
 	/*
 	 * The crypto API expects us to set the IV (req->iv) to the last
-	 * ciphertext block.
+	 * ciphertext block when running in CBC mode.
 	 */
-	if (ivsize)
+	if ((ctx->cdata.algtype & OP_ALG_AAI_MASK) == OP_ALG_AAI_CBC)
 		scatterwalk_map_and_copy(req->iv, req->src, req->cryptlen -
 					 ivsize, ivsize, 0);
 
author maintainer
#242 lore raw

PATCH 241/371 crypto: ccp - Validate the the error value used to index error messages

In reply to Greg Kroah-Hartman

From: Hook, Gary <[email protected]>

commit 52393d617af7b554f03531e6756facf2ea687d2e upstream.

The error code read from the queue status register is only 6 bits wide, but we need to verify its value is within range before indexing the error messages.

Fixes 81422badb3907 ("crypto: ccp - Make syslog errors human-readable")
---
 drivers/crypto/ccp/ccp-dev.c |   96 ++++++++++++++++++++++---------------------
 drivers/crypto/ccp/ccp-dev.h |    2 
 2 files changed, 52 insertions(+), 46 deletions(-)
drivers/crypto/ccp/ccp-dev.c +51 -45
@@ -35,56 +35,62 @@ struct ccp_tasklet_data {
 };
 
 /* Human-readable error strings */
+#define CCP_MAX_ERROR_CODE	64
 static char *ccp_error_codes[] = {
 	"",
-	"ERR 01: ILLEGAL_ENGINE",
-	"ERR 02: ILLEGAL_KEY_ID",
-	"ERR 03: ILLEGAL_FUNCTION_TYPE",
-	"ERR 04: ILLEGAL_FUNCTION_MODE",
-	"ERR 05: ILLEGAL_FUNCTION_ENCRYPT",
-	"ERR 06: ILLEGAL_FUNCTION_SIZE",
-	"ERR 07: Zlib_MISSING_INIT_EOM",
-	"ERR 08: ILLEGAL_FUNCTION_RSVD",
-	"ERR 09: ILLEGAL_BUFFER_LENGTH",
-	"ERR 10: VLSB_FAULT",
-	"ERR 11: ILLEGAL_MEM_ADDR",
-	"ERR 12: ILLEGAL_MEM_SEL",
-	"ERR 13: ILLEGAL_CONTEXT_ID",
-	"ERR 14: ILLEGAL_KEY_ADDR",
-	"ERR 15: 0xF Reserved",
-	"ERR 16: Zlib_ILLEGAL_MULTI_QUEUE",
-	"ERR 17: Zlib_ILLEGAL_JOBID_CHANGE",
-	"ERR 18: CMD_TIMEOUT",
-	"ERR 19: IDMA0_AXI_SLVERR",
-	"ERR 20: IDMA0_AXI_DECERR",
-	"ERR 21: 0x15 Reserved",
-	"ERR 22: IDMA1_AXI_SLAVE_FAULT",
-	"ERR 23: IDMA1_AIXI_DECERR",
-	"ERR 24: 0x18 Reserved",
-	"ERR 25: ZLIBVHB_AXI_SLVERR",
-	"ERR 26: ZLIBVHB_AXI_DECERR",
-	"ERR 27: 0x1B Reserved",
-	"ERR 27: ZLIB_UNEXPECTED_EOM",
-	"ERR 27: ZLIB_EXTRA_DATA",
-	"ERR 30: ZLIB_BTYPE",
-	"ERR 31: ZLIB_UNDEFINED_SYMBOL",
-	"ERR 32: ZLIB_UNDEFINED_DISTANCE_S",
-	"ERR 33: ZLIB_CODE_LENGTH_SYMBOL",
-	"ERR 34: ZLIB _VHB_ILLEGAL_FETCH",
-	"ERR 35: ZLIB_UNCOMPRESSED_LEN",
-	"ERR 36: ZLIB_LIMIT_REACHED",
-	"ERR 37: ZLIB_CHECKSUM_MISMATCH0",
-	"ERR 38: ODMA0_AXI_SLVERR",
-	"ERR 39: ODMA0_AXI_DECERR",
-	"ERR 40: 0x28 Reserved",
-	"ERR 41: ODMA1_AXI_SLVERR",
-	"ERR 42: ODMA1_AXI_DECERR",
-	"ERR 43: LSB_PARITY_ERR",
+	"ILLEGAL_ENGINE",
+	"ILLEGAL_KEY_ID",
+	"ILLEGAL_FUNCTION_TYPE",
+	"ILLEGAL_FUNCTION_MODE",
+	"ILLEGAL_FUNCTION_ENCRYPT",
+	"ILLEGAL_FUNCTION_SIZE",
+	"Zlib_MISSING_INIT_EOM",
+	"ILLEGAL_FUNCTION_RSVD",
+	"ILLEGAL_BUFFER_LENGTH",
+	"VLSB_FAULT",
+	"ILLEGAL_MEM_ADDR",
+	"ILLEGAL_MEM_SEL",
+	"ILLEGAL_CONTEXT_ID",
+	"ILLEGAL_KEY_ADDR",
+	"0xF Reserved",
+	"Zlib_ILLEGAL_MULTI_QUEUE",
+	"Zlib_ILLEGAL_JOBID_CHANGE",
+	"CMD_TIMEOUT",
+	"IDMA0_AXI_SLVERR",
+	"IDMA0_AXI_DECERR",
+	"0x15 Reserved",
+	"IDMA1_AXI_SLAVE_FAULT",
+	"IDMA1_AIXI_DECERR",
+	"0x18 Reserved",
+	"ZLIBVHB_AXI_SLVERR",
+	"ZLIBVHB_AXI_DECERR",
+	"0x1B Reserved",
+	"ZLIB_UNEXPECTED_EOM",
+	"ZLIB_EXTRA_DATA",
+	"ZLIB_BTYPE",
+	"ZLIB_UNDEFINED_SYMBOL",
+	"ZLIB_UNDEFINED_DISTANCE_S",
+	"ZLIB_CODE_LENGTH_SYMBOL",
+	"ZLIB _VHB_ILLEGAL_FETCH",
+	"ZLIB_UNCOMPRESSED_LEN",
+	"ZLIB_LIMIT_REACHED",
+	"ZLIB_CHECKSUM_MISMATCH0",
+	"ODMA0_AXI_SLVERR",
+	"ODMA0_AXI_DECERR",
+	"0x28 Reserved",
+	"ODMA1_AXI_SLVERR",
+	"ODMA1_AXI_DECERR",
 };
 
-void ccp_log_error(struct ccp_device *d, int e)
+void ccp_log_error(struct ccp_device *d, unsigned int e)
 {
-	dev_err(d->dev, "CCP error: %s (0x%x)\n", ccp_error_codes[e], e);
+	if (WARN_ON(e >= CCP_MAX_ERROR_CODE))
+		return;
+
+	if (e < ARRAY_SIZE(ccp_error_codes))
+		dev_err(d->dev, "CCP error %d: %s\n", e, ccp_error_codes[e]);
+	else
+		dev_err(d->dev, "CCP error %d: Unknown Error\n", e);
 }
 
 /* List of CCPs, CCP count, read-write access lock, and access functions
drivers/crypto/ccp/ccp-dev.h +1 -1
@@ -632,7 +632,7 @@ struct ccp5_desc {
 void ccp_add_device(struct ccp_device *ccp);
 void ccp_del_device(struct ccp_device *ccp);
 
-extern void ccp_log_error(struct ccp_device *, int);
+extern void ccp_log_error(struct ccp_device *, unsigned int);
 
 struct ccp_device *ccp_alloc_struct(struct sp_device *sp);
 bool ccp_queues_suspended(struct ccp_device *ccp);
author maintainer
#243 lore raw

PATCH 242/371 crypto: arm64/sha1-ce - correct digest for empty data in finup

In reply to Greg Kroah-Hartman

From: Elena Petrova <[email protected]>

commit 1d4aaf16defa86d2665ae7db0259d6cb07e2091f upstream.

The sha1-ce finup implementation for ARM64 produces wrong digest for empty input (len=0). Expected: da39a3ee..., result: 67452301... (initial value of SHA internal state). The error is in sha1_ce_finup:
for empty data finalize will be 1, so the code is relying on sha1_ce_transform to make the final round. However, in sha1_base_do_update, the block function will not be called when len == 0.

Fix it by setting finalize to 0 if data is empty.

Fixes 07eb54d306f4 ("crypto: arm64/sha1-ce - move SHA-1 ARMv8 implementation to base layer")
---
 arch/arm64/crypto/sha1-ce-glue.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/crypto/sha1-ce-glue.c +1 -1
@@ -54,7 +54,7 @@ static int sha1_ce_finup(struct shash_de
 			 unsigned int len, u8 *out)
 {
 	struct sha1_ce_state *sctx = shash_desc_ctx(desc);
-	bool finalize = !sctx->sst.count && !(len % SHA1_BLOCK_SIZE);
+	bool finalize = !sctx->sst.count && !(len % SHA1_BLOCK_SIZE) && len;
 
 	if (!may_use_simd())
 		return crypto_sha1_finup(desc, data, len, out);
author maintainer
#244 lore raw

PATCH 243/371 crypto: arm64/sha2-ce - correct digest for empty data in finup

In reply to Greg Kroah-Hartman

From: Elena Petrova <[email protected]>

commit 6bd934de1e393466b319d29c4427598fda096c57 upstream.

The sha256-ce finup implementation for ARM64 produces wrong digest for empty input (len=0). Expected: the actual digest, result: initial value of SHA internal state. The error is in sha256_ce_finup:
for empty data finalize will be 1, so the code is relying on sha2_ce_transform to make the final round. However, in sha256_base_do_update, the block function will not be called when len == 0.

Fix it by setting finalize to 0 if data is empty.

Fixes 03802f6a80b3a ("crypto: arm64/sha2-ce - move SHA-224/256 ARMv8 implementation to base layer")
---
 arch/arm64/crypto/sha2-ce-glue.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/crypto/sha2-ce-glue.c +1 -1
@@ -59,7 +59,7 @@ static int sha256_ce_finup(struct shash_
 			   unsigned int len, u8 *out)
 {
 	struct sha256_ce_state *sctx = shash_desc_ctx(desc);
-	bool finalize = !sctx->sst.count && !(len % SHA256_BLOCK_SIZE);
+	bool finalize = !sctx->sst.count && !(len % SHA256_BLOCK_SIZE) && len;
 
 	if (!may_use_simd()) {
 		if (len)
author maintainer
#245 lore raw

PATCH 244/371 crypto: chacha20poly1305 - fix atomic sleep when using async algorithm

In reply to Greg Kroah-Hartman

From: Eric Biggers <[email protected]>

commit 7545b6c2087f4ef0287c8c9b7eba6a728c67ff8e upstream.

Clear the CRYPTO_TFM_REQ_MAY_SLEEP flag when the chacha20poly1305 operation is being continued from an async completion callback, since sleeping may not be allowed in that context.

This is basically the same bug that was recently fixed in the xts and lrw templates. But, it's always been broken in chacha20poly1305 too. This was found using syzkaller in combination with the updated crypto self-tests which actually test the MAY_SLEEP flag now.

Reproducer:

    python -c 'import socket; socket.socket(socket.AF_ALG, 5, 0).bind(
    	       ("aead", "rfc7539(cryptd(chacha20-generic),poly1305-generic)"))'

Kernel output:

    BUG: sleeping function called from invalid context at include/crypto/algapi.h:426
    in_atomic(): 1, irqs_disabled(): 0, pid: 1001, name: kworker/2:2
    [...]
    CPU: 2 PID: 1001 Comm: kworker/2:2 Not tainted 5.2.0-rc2 #5
    Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-20181126_142135-anatol 04/01/2014
    Workqueue: crypto cryptd_queue_worker
    Call Trace:
     __dump_stack lib/dump_stack.c:77 [inline]
     dump_stack+0x4d/0x6a lib/dump_stack.c:113
     ___might_sleep kernel/sched/core.c:6138 [inline]
     ___might_sleep.cold.19+0x8e/0x9f kernel/sched/core.c:6095
     crypto_yield include/crypto/algapi.h:426 [inline]
     crypto_hash_walk_done+0xd6/0x100 crypto/ahash.c:113
     shash_ahash_update+0x41/0x60 crypto/shash.c:251
     shash_async_update+0xd/0x10 crypto/shash.c:260
     crypto_ahash_update include/crypto/hash.h:539 [inline]
     poly_setkey+0xf6/0x130 crypto/chacha20poly1305.c:337
     poly_init+0x51/0x60 crypto/chacha20poly1305.c:364
     async_done_continue crypto/chacha20poly1305.c:78 [inline]
     poly_genkey_done+0x15/0x30 crypto/chacha20poly1305.c:369
     cryptd_skcipher_complete+0x29/0x70 crypto/cryptd.c:279
     cryptd_skcipher_decrypt+0xcd/0x110 crypto/cryptd.c:339
     cryptd_queue_worker+0x70/0xa0 crypto/cryptd.c:184
     process_one_work+0x1ed/0x420 kernel/workqueue.c:2269
     worker_thread+0x3e/0x3a0 kernel/workqueue.c:2415
     kthread+0x11f/0x140 kernel/kthread.c:255
     ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:352
Fixes 71ebc4d1b27d ("crypto: chacha20poly1305 - Add a ChaCha20-Poly1305 AEAD construction, RFC7539")
Cc <[email protected]> # v4.2+
Cc Martin Willi <[email protected]>
---
 crypto/chacha20poly1305.c |   30 +++++++++++++++++++-----------
 1 file changed, 19 insertions(+), 11 deletions(-)
crypto/chacha20poly1305.c +19 -11
@@ -65,6 +65,8 @@ struct chachapoly_req_ctx {
 	unsigned int cryptlen;
 	/* Actual AD, excluding IV */
 	unsigned int assoclen;
+	/* request flags, with MAY_SLEEP cleared if needed */
+	u32 flags;
 	union {
 		struct poly_req poly;
 		struct chacha_req chacha;
@@ -74,8 +76,12 @@ struct chachapoly_req_ctx {
 static inline void async_done_continue(struct aead_request *req, int err,
 				       int (*cont)(struct aead_request *))
 {
-	if (!err)
+	if (!err) {
+		struct chachapoly_req_ctx *rctx = aead_request_ctx(req);
+
+		rctx->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP;
 		err = cont(req);
+	}
 
 	if (err != -EINPROGRESS && err != -EBUSY)
 		aead_request_complete(req, err);
@@ -142,7 +148,7 @@ static int chacha_decrypt(struct aead_re
 		dst = scatterwalk_ffwd(rctx->dst, req->dst, req->assoclen);
 	}
 
-	skcipher_request_set_callback(&creq->req, aead_request_flags(req),
+	skcipher_request_set_callback(&creq->req, rctx->flags,
 				      chacha_decrypt_done, req);
 	skcipher_request_set_tfm(&creq->req, ctx->chacha);
 	skcipher_request_set_crypt(&creq->req, src, dst,
@@ -186,7 +192,7 @@ static int poly_tail(struct aead_request
 	memcpy(&preq->tail.cryptlen, &len, sizeof(len));
 	sg_set_buf(preq->src, &preq->tail, sizeof(preq->tail));
 
-	ahash_request_set_callback(&preq->req, aead_request_flags(req),
+	ahash_request_set_callback(&preq->req, rctx->flags,
 				   poly_tail_done, req);
 	ahash_request_set_tfm(&preq->req, ctx->poly);
 	ahash_request_set_crypt(&preq->req, preq->src,
@@ -217,7 +223,7 @@ static int poly_cipherpad(struct aead_re
 	sg_init_table(preq->src, 1);
 	sg_set_buf(preq->src, &preq->pad, padlen);
 
-	ahash_request_set_callback(&preq->req, aead_request_flags(req),
+	ahash_request_set_callback(&preq->req, rctx->flags,
 				   poly_cipherpad_done, req);
 	ahash_request_set_tfm(&preq->req, ctx->poly);
 	ahash_request_set_crypt(&preq->req, preq->src, NULL, padlen);
@@ -248,7 +254,7 @@ static int poly_cipher(struct aead_reque
 	sg_init_table(rctx->src, 2);
 	crypt = scatterwalk_ffwd(rctx->src, crypt, req->assoclen);
 
-	ahash_request_set_callback(&preq->req, aead_request_flags(req),
+	ahash_request_set_callback(&preq->req, rctx->flags,
 				   poly_cipher_done, req);
 	ahash_request_set_tfm(&preq->req, ctx->poly);
 	ahash_request_set_crypt(&preq->req, crypt, NULL, rctx->cryptlen);
@@ -278,7 +284,7 @@ static int poly_adpad(struct aead_reques
 	sg_init_table(preq->src, 1);
 	sg_set_buf(preq->src, preq->pad, padlen);
 
-	ahash_request_set_callback(&preq->req, aead_request_flags(req),
+	ahash_request_set_callback(&preq->req, rctx->flags,
 				   poly_adpad_done, req);
 	ahash_request_set_tfm(&preq->req, ctx->poly);
 	ahash_request_set_crypt(&preq->req, preq->src, NULL, padlen);
@@ -302,7 +308,7 @@ static int poly_ad(struct aead_request *
 	struct poly_req *preq = &rctx->u.poly;
 	int err;
 
-	ahash_request_set_callback(&preq->req, aead_request_flags(req),
+	ahash_request_set_callback(&preq->req, rctx->flags,
 				   poly_ad_done, req);
 	ahash_request_set_tfm(&preq->req, ctx->poly);
 	ahash_request_set_crypt(&preq->req, req->src, NULL, rctx->assoclen);
@@ -329,7 +335,7 @@ static int poly_setkey(struct aead_reque
 	sg_init_table(preq->src, 1);
 	sg_set_buf(preq->src, rctx->key, sizeof(rctx->key));
 
-	ahash_request_set_callback(&preq->req, aead_request_flags(req),
+	ahash_request_set_callback(&preq->req, rctx->flags,
 				   poly_setkey_done, req);
 	ahash_request_set_tfm(&preq->req, ctx->poly);
 	ahash_request_set_crypt(&preq->req, preq->src, NULL, sizeof(rctx->key));
@@ -353,7 +359,7 @@ static int poly_init(struct aead_request
 	struct poly_req *preq = &rctx->u.poly;
 	int err;
 
-	ahash_request_set_callback(&preq->req, aead_request_flags(req),
+	ahash_request_set_callback(&preq->req, rctx->flags,
 				   poly_init_done, req);
 	ahash_request_set_tfm(&preq->req, ctx->poly);
 
@@ -391,7 +397,7 @@ static int poly_genkey(struct aead_reque
 
 	chacha_iv(creq->iv, req, 0);
 
-	skcipher_request_set_callback(&creq->req, aead_request_flags(req),
+	skcipher_request_set_callback(&creq->req, rctx->flags,
 				      poly_genkey_done, req);
 	skcipher_request_set_tfm(&creq->req, ctx->chacha);
 	skcipher_request_set_crypt(&creq->req, creq->src, creq->src,
@@ -431,7 +437,7 @@ static int chacha_encrypt(struct aead_re
 		dst = scatterwalk_ffwd(rctx->dst, req->dst, req->assoclen);
 	}
 
-	skcipher_request_set_callback(&creq->req, aead_request_flags(req),
+	skcipher_request_set_callback(&creq->req, rctx->flags,
 				      chacha_encrypt_done, req);
 	skcipher_request_set_tfm(&creq->req, ctx->chacha);
 	skcipher_request_set_crypt(&creq->req, src, dst,
@@ -449,6 +455,7 @@ static int chachapoly_encrypt(struct aea
 	struct chachapoly_req_ctx *rctx = aead_request_ctx(req);
 
 	rctx->cryptlen = req->cryptlen;
+	rctx->flags = aead_request_flags(req);
 
 	/* encrypt call chain:
 	 * - chacha_encrypt/done()
@@ -470,6 +477,7 @@ static int chachapoly_decrypt(struct aea
 	struct chachapoly_req_ctx *rctx = aead_request_ctx(req);
 
 	rctx->cryptlen = req->cryptlen - POLY1305_DIGEST_SIZE;
+	rctx->flags = aead_request_flags(req);
 
 	/* decrypt call chain:
 	 * - poly_genkey/done()
author maintainer
#246 lore raw

PATCH 245/371 crypto: crypto4xx - fix AES CTR blocksize value

In reply to Greg Kroah-Hartman

From: Christian Lamparter <[email protected]>

commit bfa2ba7d9e6b20aca82b99e6842fe18842ae3a0f upstream.

This patch fixes a issue with crypto4xx's ctr(aes) that was discovered by libcapi's kcapi-enc-test.sh test.

The some of the ctr(aes) encryptions test were failing on the non-power-of-two test:

kcapi-enc - Error: encryption failed with error 0
kcapi-enc - Error: decryption failed with error 0
[FAILED: 32-bit - 5.1.0-rc1+] 15 bytes: STDIN / STDOUT enc test (128 bits):
original file (1d100e..cc96184c) and generated file (e3b0c442..1b7852b855) [FAILED: 32-bit - 5.1.0-rc1+] 15 bytes: STDIN / STDOUT enc test (128 bits) (openssl generated CT): original file (e3b0..5) and generated file (3..8e) [PASSED: 32-bit - 5.1.0-rc1+] 15 bytes: STDIN / STDOUT enc test (128 bits) (openssl generated PT)
[FAILED: 32-bit - 5.1.0-rc1+] 15 bytes: STDIN / STDOUT enc test (password):
original file (1d1..84c) and generated file (e3b..852b855)

But the 16, 32, 512, 65536 tests always worked.

Thankfully, this isn't a hidden hardware problem like previously, instead this turned out to be a copy and paste issue.

With this patch, all the tests are passing with and kcapi-enc-test.sh gives crypto4xx's a clean bill of health:
"Number of failures: 0" :).

Fixes 98e87e3d933b ("crypto: crypto4xx - add aes-ctr support")
Fixes f2a13e7cba9e ("crypto: crypto4xx - enable AES RFC3686, ECB, CFB and OFB offloads")
---
 drivers/crypto/amcc/crypto4xx_core.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/crypto/amcc/crypto4xx_core.c +2 -2
@@ -1259,7 +1259,7 @@ static struct crypto4xx_alg_common crypt
 			.cra_flags = CRYPTO_ALG_NEED_FALLBACK |
 				CRYPTO_ALG_ASYNC |
 				CRYPTO_ALG_KERN_DRIVER_ONLY,
-			.cra_blocksize = AES_BLOCK_SIZE,
+			.cra_blocksize = 1,
 			.cra_ctxsize = sizeof(struct crypto4xx_ctx),
 			.cra_module = THIS_MODULE,
 		},
@@ -1279,7 +1279,7 @@ static struct crypto4xx_alg_common crypt
 			.cra_priority = CRYPTO4XX_CRYPTO_PRIORITY,
 			.cra_flags = CRYPTO_ALG_ASYNC |
 				CRYPTO_ALG_KERN_DRIVER_ONLY,
-			.cra_blocksize = AES_BLOCK_SIZE,
+			.cra_blocksize = 1,
 			.cra_ctxsize = sizeof(struct crypto4xx_ctx),
 			.cra_module = THIS_MODULE,
 		},
author maintainer
#247 lore raw

PATCH 246/371 crypto: crypto4xx - fix blocksize for cfb and ofb

In reply to Greg Kroah-Hartman

From: Christian Lamparter <[email protected]>

commit 70c4997f34b6c6888b3ac157adec49e01d0df2d5 upstream.

While the hardware consider them to be blockciphers, the reference implementation defines them as streamciphers.

Do the right thing and set the blocksize to 1. This was found by CONFIG_CRYPTO_MANAGER_EXTRA_TESTS.

This fixes the following issues:
skcipher: blocksize for ofb-aes-ppc4xx (16) doesn't match generic impl (1)
skcipher: blocksize for cfb-aes-ppc4xx (16) doesn't match generic impl (1)

Cc Eric Biggers <[email protected]>
Fixes f2a13e7cba9e ("crypto: crypto4xx - enable AES RFC3686, ECB, CFB and OFB offloads")
---
 drivers/crypto/amcc/crypto4xx_core.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/crypto/amcc/crypto4xx_core.c +2 -2
@@ -1238,7 +1238,7 @@ static struct crypto4xx_alg_common crypt
 			.cra_priority = CRYPTO4XX_CRYPTO_PRIORITY,
 			.cra_flags = CRYPTO_ALG_ASYNC |
 				CRYPTO_ALG_KERN_DRIVER_ONLY,
-			.cra_blocksize = AES_BLOCK_SIZE,
+			.cra_blocksize = 1,
 			.cra_ctxsize = sizeof(struct crypto4xx_ctx),
 			.cra_module = THIS_MODULE,
 		},
@@ -1318,7 +1318,7 @@ static struct crypto4xx_alg_common crypt
 			.cra_priority = CRYPTO4XX_CRYPTO_PRIORITY,
 			.cra_flags = CRYPTO_ALG_ASYNC |
 				CRYPTO_ALG_KERN_DRIVER_ONLY,
-			.cra_blocksize = AES_BLOCK_SIZE,
+			.cra_blocksize = 1,
 			.cra_ctxsize = sizeof(struct crypto4xx_ctx),
 			.cra_module = THIS_MODULE,
 		},
author maintainer
#248 lore raw

PATCH 247/371 crypto: crypto4xx - block ciphers should only accept complete blocks

In reply to Greg Kroah-Hartman

From: Christian Lamparter <[email protected]>

commit 0f7a81374060828280fcfdfbaa162cb559017f9f upstream.

The hardware automatically zero pads incomplete block ciphers blocks without raising any errors. This is a screw-up. This was noticed by CONFIG_CRYPTO_MANAGER_EXTRA_TESTS tests that sent a incomplete blocks and expect them to fail.

This fixes:
cbc-aes-ppc4xx encryption unexpectedly succeeded on test vector "random: len=2409 klen=32"; expected_error=-22, cfg="random:
may_sleep use_digest src_divs=[96.90%@+2295, 2.34%@+4066, 0.32%@alignmask+12, 0.34%@+4087, 0.9%@alignmask+1787, 0.1%@+3767] iv_offset=6"

ecb-aes-ppc4xx encryption unexpectedly succeeded on test vector "random: len=1011 klen=32"; expected_error=-22, cfg="random:
may_sleep use_digest src_divs=[100.0%@alignmask+20] dst_divs=[3.12%@+3001, 96.88%@+4070]"

---
 drivers/crypto/amcc/crypto4xx_alg.c  |   36 ++++++++++++++++++++++++-----------
 drivers/crypto/amcc/crypto4xx_core.c |   16 +++++++--------
 drivers/crypto/amcc/crypto4xx_core.h |   10 +++++----
 3 files changed, 39 insertions(+), 23 deletions(-)
drivers/crypto/amcc/crypto4xx_alg.c +25 -11
@@ -76,12 +76,16 @@ static void set_dynamic_sa_command_1(str
 }
 
 static inline int crypto4xx_crypt(struct skcipher_request *req,
-				  const unsigned int ivlen, bool decrypt)
+				  const unsigned int ivlen, bool decrypt,
+				  bool check_blocksize)
 {
 	struct crypto_skcipher *cipher = crypto_skcipher_reqtfm(req);
 	struct crypto4xx_ctx *ctx = crypto_skcipher_ctx(cipher);
 	__le32 iv[AES_IV_SIZE];
 
+	if (check_blocksize && !IS_ALIGNED(req->cryptlen, AES_BLOCK_SIZE))
+		return -EINVAL;
+
 	if (ivlen)
 		crypto4xx_memcpy_to_le32(iv, req->iv, ivlen);
 
@@ -90,24 +94,34 @@ static inline int crypto4xx_crypt(struct
 		ctx->sa_len, 0, NULL);
 }
 
-int crypto4xx_encrypt_noiv(struct skcipher_request *req)
+int crypto4xx_encrypt_noiv_block(struct skcipher_request *req)
+{
+	return crypto4xx_crypt(req, 0, false, true);
+}
+
+int crypto4xx_encrypt_iv_stream(struct skcipher_request *req)
+{
+	return crypto4xx_crypt(req, AES_IV_SIZE, false, false);
+}
+
+int crypto4xx_decrypt_noiv_block(struct skcipher_request *req)
 {
-	return crypto4xx_crypt(req, 0, false);
+	return crypto4xx_crypt(req, 0, true, true);
 }
 
-int crypto4xx_encrypt_iv(struct skcipher_request *req)
+int crypto4xx_decrypt_iv_stream(struct skcipher_request *req)
 {
-	return crypto4xx_crypt(req, AES_IV_SIZE, false);
+	return crypto4xx_crypt(req, AES_IV_SIZE, true, false);
 }
 
-int crypto4xx_decrypt_noiv(struct skcipher_request *req)
+int crypto4xx_encrypt_iv_block(struct skcipher_request *req)
 {
-	return crypto4xx_crypt(req, 0, true);
+	return crypto4xx_crypt(req, AES_IV_SIZE, false, true);
 }
 
-int crypto4xx_decrypt_iv(struct skcipher_request *req)
+int crypto4xx_decrypt_iv_block(struct skcipher_request *req)
 {
-	return crypto4xx_crypt(req, AES_IV_SIZE, true);
+	return crypto4xx_crypt(req, AES_IV_SIZE, true, true);
 }
 
 /**
@@ -278,8 +292,8 @@ crypto4xx_ctr_crypt(struct skcipher_requ
 		return ret;
 	}
 
-	return encrypt ? crypto4xx_encrypt_iv(req)
-		       : crypto4xx_decrypt_iv(req);
+	return encrypt ? crypto4xx_encrypt_iv_stream(req)
+		       : crypto4xx_decrypt_iv_stream(req);
 }
 
 static int crypto4xx_sk_setup_fallback(struct crypto4xx_ctx *ctx,
drivers/crypto/amcc/crypto4xx_core.c +8 -8
@@ -1226,8 +1226,8 @@ static struct crypto4xx_alg_common crypt
 		.max_keysize = AES_MAX_KEY_SIZE,
 		.ivsize	= AES_IV_SIZE,
 		.setkey = crypto4xx_setkey_aes_cbc,
-		.encrypt = crypto4xx_encrypt_iv,
-		.decrypt = crypto4xx_decrypt_iv,
+		.encrypt = crypto4xx_encrypt_iv_block,
+		.decrypt = crypto4xx_decrypt_iv_block,
 		.init = crypto4xx_sk_init,
 		.exit = crypto4xx_sk_exit,
 	} },
@@ -1246,8 +1246,8 @@ static struct crypto4xx_alg_common crypt
 		.max_keysize = AES_MAX_KEY_SIZE,
 		.ivsize	= AES_IV_SIZE,
 		.setkey	= crypto4xx_setkey_aes_cfb,
-		.encrypt = crypto4xx_encrypt_iv,
-		.decrypt = crypto4xx_decrypt_iv,
+		.encrypt = crypto4xx_encrypt_iv_stream,
+		.decrypt = crypto4xx_decrypt_iv_stream,
 		.init = crypto4xx_sk_init,
 		.exit = crypto4xx_sk_exit,
 	} },
@@ -1306,8 +1306,8 @@ static struct crypto4xx_alg_common crypt
 		.min_keysize = AES_MIN_KEY_SIZE,
 		.max_keysize = AES_MAX_KEY_SIZE,
 		.setkey	= crypto4xx_setkey_aes_ecb,
-		.encrypt = crypto4xx_encrypt_noiv,
-		.decrypt = crypto4xx_decrypt_noiv,
+		.encrypt = crypto4xx_encrypt_noiv_block,
+		.decrypt = crypto4xx_decrypt_noiv_block,
 		.init = crypto4xx_sk_init,
 		.exit = crypto4xx_sk_exit,
 	} },
@@ -1326,8 +1326,8 @@ static struct crypto4xx_alg_common crypt
 		.max_keysize = AES_MAX_KEY_SIZE,
 		.ivsize	= AES_IV_SIZE,
 		.setkey	= crypto4xx_setkey_aes_ofb,
-		.encrypt = crypto4xx_encrypt_iv,
-		.decrypt = crypto4xx_decrypt_iv,
+		.encrypt = crypto4xx_encrypt_iv_stream,
+		.decrypt = crypto4xx_decrypt_iv_stream,
 		.init = crypto4xx_sk_init,
 		.exit = crypto4xx_sk_exit,
 	} },
drivers/crypto/amcc/crypto4xx_core.h +6 -4
@@ -183,10 +183,12 @@ int crypto4xx_setkey_rfc3686(struct cryp
 			     const u8 *key, unsigned int keylen);
 int crypto4xx_encrypt_ctr(struct skcipher_request *req);
 int crypto4xx_decrypt_ctr(struct skcipher_request *req);
-int crypto4xx_encrypt_iv(struct skcipher_request *req);
-int crypto4xx_decrypt_iv(struct skcipher_request *req);
-int crypto4xx_encrypt_noiv(struct skcipher_request *req);
-int crypto4xx_decrypt_noiv(struct skcipher_request *req);
+int crypto4xx_encrypt_iv_stream(struct skcipher_request *req);
+int crypto4xx_decrypt_iv_stream(struct skcipher_request *req);
+int crypto4xx_encrypt_iv_block(struct skcipher_request *req);
+int crypto4xx_decrypt_iv_block(struct skcipher_request *req);
+int crypto4xx_encrypt_noiv_block(struct skcipher_request *req);
+int crypto4xx_decrypt_noiv_block(struct skcipher_request *req);
 int crypto4xx_rfc3686_encrypt(struct skcipher_request *req);
 int crypto4xx_rfc3686_decrypt(struct skcipher_request *req);
 int crypto4xx_sha1_alg_init(struct crypto_tfm *tfm);
author maintainer
#249 lore raw

PATCH 248/371 crypto: ccp - memset structure fields to zero before reuse

In reply to Greg Kroah-Hartman

From: Hook, Gary <[email protected]>

commit 20e833dc36355ed642d00067641a679c618303fa upstream.

The AES GCM function reuses an 'op' data structure, which members contain values that must be cleared for each (re)use.

This fix resolves a crypto self-test failure:
alg: aead: gcm-aes-ccp encryption test failed (wrong result) on test vector 2, cfg="two even aligned splits"

Fixes 36cf515b9bbe ("crypto: ccp - Enable support for AES GCM on v5 CCPs")
---
 drivers/crypto/ccp/ccp-ops.c |   12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)
drivers/crypto/ccp/ccp-ops.c +11 -1
@@ -625,6 +625,7 @@ static int ccp_run_aes_gcm_cmd(struct cc
 
 	unsigned long long *final;
 	unsigned int dm_offset;
+	unsigned int jobid;
 	unsigned int ilen;
 	bool in_place = true; /* Default value */
 	int ret;
@@ -663,9 +664,11 @@ static int ccp_run_aes_gcm_cmd(struct cc
 		p_tag = scatterwalk_ffwd(sg_tag, p_inp, ilen);
 	}
 
+	jobid = CCP_NEW_JOBID(cmd_q->ccp);
+
 	memset(&op, 0, sizeof(op));
 	op.cmd_q = cmd_q;
-	op.jobid = CCP_NEW_JOBID(cmd_q->ccp);
+	op.jobid = jobid;
 	op.sb_key = cmd_q->sb_key; /* Pre-allocated */
 	op.sb_ctx = cmd_q->sb_ctx; /* Pre-allocated */
 	op.init = 1;
@@ -816,6 +819,13 @@ static int ccp_run_aes_gcm_cmd(struct cc
 	final[0] = cpu_to_be64(aes->aad_len * 8);
 	final[1] = cpu_to_be64(ilen * 8);
 
+	memset(&op, 0, sizeof(op));
+	op.cmd_q = cmd_q;
+	op.jobid = jobid;
+	op.sb_key = cmd_q->sb_key; /* Pre-allocated */
+	op.sb_ctx = cmd_q->sb_ctx; /* Pre-allocated */
+	op.init = 1;
+	op.u.aes.type = aes->type;
 	op.u.aes.mode = CCP_AES_MODE_GHASH;
 	op.u.aes.action = CCP_AES_GHASHFINAL;
 	op.src.type = CCP_MEMTYPE_SYSTEM;
author maintainer
#250 lore raw

PATCH 249/371 crypto: ccp/gcm - use const time tag comparison.

In reply to Greg Kroah-Hartman

From: Cfir Cohen <[email protected]>

commit 538a5a072e6ef04377b180ee9b3ce5bae0a85da4 upstream.

Avoid leaking GCM tag through timing side channel.

Fixes 36cf515b9bbe ("crypto: ccp - Enable support for AES GCM on v5 CCPs")
Cc <[email protected]> # v4.12+
---
 drivers/crypto/ccp/ccp-ops.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/crypto/ccp/ccp-ops.c +2 -1
@@ -853,7 +853,8 @@ static int ccp_run_aes_gcm_cmd(struct cc
 		if (ret)
 			goto e_tag;
 
-		ret = memcmp(tag.address, final_wa.address, AES_BLOCK_SIZE);
+		ret = crypto_memneq(tag.address, final_wa.address,
+				    AES_BLOCK_SIZE) ? -EBADMSG : 0;
 		ccp_dm_free(&tag);
 	}
 
author maintainer
#251 lore raw

PATCH 250/371 crypto: crypto4xx - fix a potential double free in ppc4xx_trng_probe

In reply to Greg Kroah-Hartman

From: Wen Yang <[email protected]>

commit 95566aa75cd6b3b404502c06f66956b5481194b3 upstream.

There is a possible double free issue in ppc4xx_trng_probe():

85: dev->trng_base = of_iomap(trng, 0);
86: of_node_put(trng); ---> released here
87: if (!dev->trng_base)
88: goto err_out;
...
110: ierr_out:
111: of_node_put(trng); ---> double released here
...

This issue was detected by using the Coccinelle software. We fix it by removing the unnecessary of_node_put().

Fixes 5343e674f32f ("crypto4xx: integrate ppc4xx-rng into crypto4xx")
Cc "David S. Miller" <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc Greg Kroah-Hartman <[email protected]>
Cc Allison Randal <[email protected]>
Cc Armijn Hemel <[email protected]>
Cc Julia Lawall <[email protected]>
---
 drivers/crypto/amcc/crypto4xx_trng.c |    1 -
 1 file changed, 1 deletion(-)
drivers/crypto/amcc/crypto4xx_trng.c +0 -1
@@ -111,7 +111,6 @@ void ppc4xx_trng_probe(struct crypto4xx_
 	return;
 
 err_out:
-	of_node_put(trng);
 	iounmap(dev->trng_base);
 	kfree(rng);
 	dev->trng_base = NULL;
author maintainer
#252 lore raw

PATCH 251/371 cifs: always add credits back for unsolicited PDUs

In reply to Greg Kroah-Hartman

From: Ronnie Sahlberg <[email protected]>

commit 3e2725796cbdfe4efc7eb7b27cacaeac2ddad1a5 upstream.

not just if CONFIG_CIFS_DEBUG2 is enabled.

---
 fs/cifs/connect.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/cifs/connect.c +1 -1
@@ -1221,11 +1221,11 @@ next_pdu:
 					 atomic_read(&midCount));
 				cifs_dump_mem("Received Data is: ", bufs[i],
 					      HEADER_SIZE(server));
+				smb2_add_credits_from_hdr(bufs[i], server);
 #ifdef CONFIG_CIFS_DEBUG2
 				if (server->ops->dump_detail)
 					server->ops->dump_detail(bufs[i],
 								 server);
-				smb2_add_credits_from_hdr(bufs[i], server);
 				cifs_dump_mids(server);
 #endif /* CIFS_DEBUG2 */
 			}
author maintainer
#253 lore raw

PATCH 252/371 cifs: fix crash in smb2_compound_op()/smb2_set_next_command()

In reply to Greg Kroah-Hartman

From: Ronnie Sahlberg <[email protected]>

commit 88a92c913cef09e70b1744a8877d177aa6cb2189 upstream.

RHBZ: 1722704

In low memory situations the various SMB2_*_init() functions can fail to allocate a request PDU and thus leave the request iovector as NULL.

If we don't check the return code for failure we end up calling smb2_set_next_command() with a NULL iovector causing a crash when it tries to dereference it.

---
 fs/cifs/smb2inode.c |   12 ++++++++++++
 fs/cifs/smb2ops.c   |   11 ++++++++++-
 2 files changed, 22 insertions(+), 1 deletion(-)
fs/cifs/smb2inode.c +12 -0
@@ -120,6 +120,8 @@ smb2_compound_op(const unsigned int xid,
 				SMB2_O_INFO_FILE, 0,
 				sizeof(struct smb2_file_all_info) +
 					  PATH_MAX * 2, 0, NULL);
+		if (rc)
+			goto finished;
 		smb2_set_next_command(tcon, &rqst[num_rqst]);
 		smb2_set_related(&rqst[num_rqst++]);
 		trace_smb3_query_info_compound_enter(xid, ses->Suid, tcon->tid,
@@ -147,6 +149,8 @@ smb2_compound_op(const unsigned int xid,
 					COMPOUND_FID, current->tgid,
 					FILE_DISPOSITION_INFORMATION,
 					SMB2_O_INFO_FILE, 0, data, size);
+		if (rc)
+			goto finished;
 		smb2_set_next_command(tcon, &rqst[num_rqst]);
 		smb2_set_related(&rqst[num_rqst++]);
 		trace_smb3_rmdir_enter(xid, ses->Suid, tcon->tid, full_path);
@@ -163,6 +167,8 @@ smb2_compound_op(const unsigned int xid,
 					COMPOUND_FID, current->tgid,
 					FILE_END_OF_FILE_INFORMATION,
 					SMB2_O_INFO_FILE, 0, data, size);
+		if (rc)
+			goto finished;
 		smb2_set_next_command(tcon, &rqst[num_rqst]);
 		smb2_set_related(&rqst[num_rqst++]);
 		trace_smb3_set_eof_enter(xid, ses->Suid, tcon->tid, full_path);
@@ -180,6 +186,8 @@ smb2_compound_op(const unsigned int xid,
 					COMPOUND_FID, current->tgid,
 					FILE_BASIC_INFORMATION,
 					SMB2_O_INFO_FILE, 0, data, size);
+		if (rc)
+			goto finished;
 		smb2_set_next_command(tcon, &rqst[num_rqst]);
 		smb2_set_related(&rqst[num_rqst++]);
 		trace_smb3_set_info_compound_enter(xid, ses->Suid, tcon->tid,
@@ -206,6 +214,8 @@ smb2_compound_op(const unsigned int xid,
 					COMPOUND_FID, current->tgid,
 					FILE_RENAME_INFORMATION,
 					SMB2_O_INFO_FILE, 0, data, size);
+		if (rc)
+			goto finished;
 		smb2_set_next_command(tcon, &rqst[num_rqst]);
 		smb2_set_related(&rqst[num_rqst++]);
 		trace_smb3_rename_enter(xid, ses->Suid, tcon->tid, full_path);
@@ -231,6 +241,8 @@ smb2_compound_op(const unsigned int xid,
 					COMPOUND_FID, current->tgid,
 					FILE_LINK_INFORMATION,
 					SMB2_O_INFO_FILE, 0, data, size);
+		if (rc)
+			goto finished;
 		smb2_set_next_command(tcon, &rqst[num_rqst]);
 		smb2_set_related(&rqst[num_rqst++]);
 		trace_smb3_hardlink_enter(xid, ses->Suid, tcon->tid, full_path);
fs/cifs/smb2ops.c +10 -1
@@ -2004,6 +2004,10 @@ smb2_set_related(struct smb_rqst *rqst)
 	struct smb2_sync_hdr *shdr;
 
 	shdr = (struct smb2_sync_hdr *)(rqst->rq_iov[0].iov_base);
+	if (shdr == NULL) {
+		cifs_dbg(FYI, "shdr NULL in smb2_set_related\n");
+		return;
+	}
 	shdr->Flags |= SMB2_FLAGS_RELATED_OPERATIONS;
 }
 
@@ -2018,6 +2022,12 @@ smb2_set_next_command(struct cifs_tcon *
 	unsigned long len = smb_rqst_len(server, rqst);
 	int i, num_padding;
 
+	shdr = (struct smb2_sync_hdr *)(rqst->rq_iov[0].iov_base);
+	if (shdr == NULL) {
+		cifs_dbg(FYI, "shdr NULL in smb2_set_next_command\n");
+		return;
+	}
+
 	/* SMB headers in a compound are 8 byte aligned. */
 
 	/* No padding needed */
@@ -2057,7 +2067,6 @@ smb2_set_next_command(struct cifs_tcon *
 	}
 
  finished:
-	shdr = (struct smb2_sync_hdr *)(rqst->rq_iov[0].iov_base);
 	shdr->NextCommand = cpu_to_le32(len);
 }
 
author maintainer
#254 lore raw

PATCH 253/371 cifs: Properly handle auto disabling of serverino option

In reply to Greg Kroah-Hartman

From: Paulo Alcantara (SUSE) <[email protected]>

commit 29fbeb7a908a60a5ae8c50fbe171cb8fdcef1980 upstream.

Fix mount options comparison when serverino option is turned off later in cifs_autodisable_serverino() and thus avoiding mismatch of new cifs mounts.

---
 fs/cifs/cifs_fs_sb.h |    5 +++++
 fs/cifs/connect.c    |    8 ++++++--
 fs/cifs/misc.c       |    1 +
 3 files changed, 12 insertions(+), 2 deletions(-)
fs/cifs/cifs_fs_sb.h +5 -0
@@ -83,5 +83,10 @@ struct cifs_sb_info {
 	 * failover properly.
 	 */
 	char *origin_fullpath; /* \\HOST\SHARE\[OPTIONAL PATH] */
+	/*
+	 * Indicate whether serverino option was turned off later
+	 * (cifs_autodisable_serverino) in order to match new mounts.
+	 */
+	bool mnt_cifs_serverino_autodisabled;
 };
 #endif				/* _CIFS_FS_SB_H */
fs/cifs/connect.c +6 -2
@@ -3455,12 +3455,16 @@ compare_mount_options(struct super_block
 {
 	struct cifs_sb_info *old = CIFS_SB(sb);
 	struct cifs_sb_info *new = mnt_data->cifs_sb;
+	unsigned int oldflags = old->mnt_cifs_flags & CIFS_MOUNT_MASK;
+	unsigned int newflags = new->mnt_cifs_flags & CIFS_MOUNT_MASK;
 
 	if ((sb->s_flags & CIFS_MS_MASK) != (mnt_data->flags & CIFS_MS_MASK))
 		return 0;
 
-	if ((old->mnt_cifs_flags & CIFS_MOUNT_MASK) !=
-	    (new->mnt_cifs_flags & CIFS_MOUNT_MASK))
+	if (old->mnt_cifs_serverino_autodisabled)
+		newflags &= ~CIFS_MOUNT_SERVER_INUM;
+
+	if (oldflags != newflags)
 		return 0;
 
 	/*
fs/cifs/misc.c +1 -0
@@ -539,6 +539,7 @@ cifs_autodisable_serverino(struct cifs_s
 			tcon = cifs_sb_master_tcon(cifs_sb);
 
 		cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_SERVER_INUM;
+		cifs_sb->mnt_cifs_serverino_autodisabled = true;
 		cifs_dbg(VFS, "Autodisabling the use of server inode numbers on %s.\n",
 			 tcon ? tcon->treeName : "new server");
 		cifs_dbg(VFS, "The server doesn't seem to support them properly or the files might be on different servers (DFS).\n");
author maintainer
#255 lore raw

PATCH 254/371 cifs: flush before set-info if we have writeable handles

In reply to Greg Kroah-Hartman

From: Ronnie Sahlberg <[email protected]>

commit aa081859b10c5d8b19f5c525c78883a59d73c2b8 upstream.

Servers can defer destaging any data and updating the mtime until close(). This means that if we do a setinfo to modify the mtime while other handles are open for write the server may overwrite our setinfo timestamps when if flushes the file on close() of the writeable handle.

To solve this we add an explicit flush when the mtime is about to be updated.

This fixes "cp -p" to preserve mtime when copying a file onto an SMB2 share.

---
 fs/cifs/inode.c |   16 ++++++++++++++++
 1 file changed, 16 insertions(+)
fs/cifs/inode.c +16 -0
@@ -2371,6 +2371,8 @@ cifs_setattr_nounix(struct dentry *diren
 	struct inode *inode = d_inode(direntry);
 	struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
 	struct cifsInodeInfo *cifsInode = CIFS_I(inode);
+	struct cifsFileInfo *wfile;
+	struct cifs_tcon *tcon;
 	char *full_path = NULL;
 	int rc = -EACCES;
 	__u32 dosattr = 0;
@@ -2417,6 +2419,20 @@ cifs_setattr_nounix(struct dentry *diren
 	mapping_set_error(inode->i_mapping, rc);
 	rc = 0;
 
+	if (attrs->ia_valid & ATTR_MTIME) {
+		rc = cifs_get_writable_file(cifsInode, false, &wfile);
+		if (!rc) {
+			tcon = tlink_tcon(wfile->tlink);
+			rc = tcon->ses->server->ops->flush(xid, tcon, &wfile->fid);
+			cifsFileInfo_put(wfile);
+			if (rc)
+				return rc;
+		} else if (rc != -EBADF)
+			return rc;
+		else
+			rc = 0;
+	}
+
 	if (attrs->ia_valid & ATTR_SIZE) {
 		rc = cifs_set_file_size(inode, attrs, xid, full_path);
 		if (rc != 0)
author maintainer
#256 lore raw

PATCH 255/371 CIFS: fix deadlock in cached root handling

In reply to Greg Kroah-Hartman

From: Aurelien Aptel <[email protected]>

commit 7e5a70ad88b1e6f6d9b934b2efb41afff496820f upstream.

Prevent deadlock between open_shroot() and cifs_mark_open_files_invalid() by releasing the lock before entering SMB2_open, taking it again after and checking if we still need to use the result.

---
 fs/cifs/smb2ops.c |   46 +++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 45 insertions(+), 1 deletion(-)
fs/cifs/smb2ops.c +45 -1
@@ -705,8 +705,51 @@ int open_shroot(unsigned int xid, struct
 
 	smb2_set_related(&rqst[1]);
 
+	/*
+	 * We do not hold the lock for the open because in case
+	 * SMB2_open needs to reconnect, it will end up calling
+	 * cifs_mark_open_files_invalid() which takes the lock again
+	 * thus causing a deadlock
+	 */
+
+	mutex_unlock(&tcon->crfid.fid_mutex);
 	rc = compound_send_recv(xid, ses, flags, 2, rqst,
 				resp_buftype, rsp_iov);
+	mutex_lock(&tcon->crfid.fid_mutex);
+
+	/*
+	 * Now we need to check again as the cached root might have
+	 * been successfully re-opened from a concurrent process
+	 */
+
+	if (tcon->crfid.is_valid) {
+		/* work was already done */
+
+		/* stash fids for close() later */
+		struct cifs_fid fid = {
+			.persistent_fid = pfid->persistent_fid,
+			.volatile_fid = pfid->volatile_fid,
+		};
+
+		/*
+		 * caller expects this func to set pfid to a valid
+		 * cached root, so we copy the existing one and get a
+		 * reference.
+		 */
+		memcpy(pfid, tcon->crfid.fid, sizeof(*pfid));
+		kref_get(&tcon->crfid.refcount);
+
+		mutex_unlock(&tcon->crfid.fid_mutex);
+
+		if (rc == 0) {
+			/* close extra handle outside of crit sec */
+			SMB2_close(xid, tcon, fid.persistent_fid, fid.volatile_fid);
+		}
+		goto oshr_free;
+	}
+
+	/* Cached root is still invalid, continue normaly */
+
 	if (rc)
 		goto oshr_exit;
 
@@ -740,8 +783,9 @@ int open_shroot(unsigned int xid, struct
 				(char *)&tcon->crfid.file_all_info))
 		tcon->crfid.file_all_info_is_valid = 1;
 
- oshr_exit:
+oshr_exit:
 	mutex_unlock(&tcon->crfid.fid_mutex);
+oshr_free:
 	SMB2_open_free(&rqst[0]);
 	SMB2_query_info_free(&rqst[1]);
 	free_rsp_buf(resp_buftype[0], rsp_iov[0].iov_base);
author maintainer
#257 lore raw

PATCH 256/371 Revert "bcache: set CACHE_SET_IO_DISABLE in bch_cached_dev_error()"

In reply to Greg Kroah-Hartman

From: Coly Li <[email protected]>

commit 695277f16b3a102fcc22c97fdf2de77c7b19f0b3 upstream.

This reverts commit 6147305c73e4511ca1a975b766b97a779d442567.

Although this patch helps the failed bcache device to stop faster when too many I/O errors detected on corresponding cached device, setting CACHE_SET_IO_DISABLE bit to cache set c->flags was not a good idea. This operation will disable all I/Os on cache set, which means other attached bcache devices won't work neither.

Without this patch, the failed bcache device can also be stopped eventually if internal I/O accomplished (e.g. writeback). Therefore here I revert it.

Fixes 6147305c73e4 ("bcache: set CACHE_SET_IO_DISABLE in bch_cached_dev_error()")
---
 drivers/md/bcache/super.c |   17 -----------------
 1 file changed, 17 deletions(-)
drivers/md/bcache/super.c +0 -17
@@ -1429,8 +1429,6 @@ int bch_flash_dev_create(struct cache_se
 
 bool bch_cached_dev_error(struct cached_dev *dc)
 {
-	struct cache_set *c;
-
 	if (!dc || test_bit(BCACHE_DEV_CLOSING, &dc->disk.flags))
 		return false;
 
@@ -1441,21 +1439,6 @@ bool bch_cached_dev_error(struct cached_
 	pr_err("stop %s: too many IO errors on backing device %s\n",
 		dc->disk.disk->disk_name, dc->backing_dev_name);
 
-	/*
-	 * If the cached device is still attached to a cache set,
-	 * even dc->io_disable is true and no more I/O requests
-	 * accepted, cache device internal I/O (writeback scan or
-	 * garbage collection) may still prevent bcache device from
-	 * being stopped. So here CACHE_SET_IO_DISABLE should be
-	 * set to c->flags too, to make the internal I/O to cache
-	 * device rejected and stopped immediately.
-	 * If c is NULL, that means the bcache device is not attached
-	 * to any cache set, then no CACHE_SET_IO_DISABLE bit to set.
-	 */
-	c = dc->disk.c;
-	if (c && test_and_set_bit(CACHE_SET_IO_DISABLE, &c->flags))
-		pr_info("CACHE_SET_IO_DISABLE already set");
-
 	bcache_device_stop(&dc->disk);
 	return true;
 }
author maintainer
#258 lore raw

PATCH 257/371 bcache: Revert "bcache: fix high CPU occupancy during journal"

In reply to Greg Kroah-Hartman

From: Coly Li <[email protected]>

commit 249a5f6da57c28a903c75d81505d58ec8c10030d upstream.

This reverts commit c4dc2497d50d9c6fb16aa0d07b6a14f3b2adb1e0.

This patch enlarges a race between normal btree flush code path and flush_btree_write(), which causes deadlock when journal space is exhausted. Reverts this patch makes the race window from 128 btree nodes to only 1 btree nodes.

Fixes c4dc2497d50d ("bcache: fix high CPU occupancy during journal")
Cc Tang Junhui <[email protected]>
---
 drivers/md/bcache/bcache.h  |    2 -
 drivers/md/bcache/journal.c |   47 ++++++++++++++------------------------------
 drivers/md/bcache/util.h    |    2 -
 3 files changed, 15 insertions(+), 36 deletions(-)
drivers/md/bcache/bcache.h +0 -2
@@ -726,8 +726,6 @@ struct cache_set {
 
 #define BUCKET_HASH_BITS	12
 	struct hlist_head	bucket_hash[1 << BUCKET_HASH_BITS];
-
-	DECLARE_HEAP(struct btree *, flush_btree);
 };
 
 struct bbio {
drivers/md/bcache/journal.c +15 -32

70 lines not on this page · raw message

drivers/md/bcache/util.h +0 -2
@@ -113,8 +113,6 @@ do {									\
 
 #define heap_full(h)	((h)->used == (h)->size)
 
-#define heap_empty(h)	((h)->used == 0)
-
 #define DECLARE_FIFO(type, name)					\
 	struct {							\
 		size_t front, back, size, mask;				\
author maintainer
#259 lore raw

PATCH 258/371 bcache: Revert "bcache: free heap cache_set->flush_btree in bch_journal_free"

In reply to Greg Kroah-Hartman

From: Coly Li <[email protected]>

commit ba82c1ac1667d6efb91a268edb13fc9cdaecec9b upstream.

This reverts commit 6268dc2c4703aabfb0b35681be709acf4c2826c6.

This patch depends on commit c4dc2497d50d ("bcache: fix high CPU occupancy during journal") which is reverted in previous patch. So revert this one too.

Fixes 6268dc2c4703 ("bcache: free heap cache_set->flush_btree in bch_journal_free")
Cc Shenghui Wang <[email protected]>
---
 drivers/md/bcache/journal.c |    1 -
 1 file changed, 1 deletion(-)
drivers/md/bcache/journal.c +0 -1
@@ -842,7 +842,6 @@ void bch_journal_free(struct cache_set *
 	free_pages((unsigned long) c->journal.w[1].data, JSET_BITS);
 	free_pages((unsigned long) c->journal.w[0].data, JSET_BITS);
 	free_fifo(&c->journal.pin);
-	free_heap(&c->flush_btree);
 }
 
 int bch_journal_alloc(struct cache_set *c)
author maintainer
#260 lore raw

PATCH 259/371 bcache: ignore read-ahead request failure on backing device

In reply to Greg Kroah-Hartman

From: Coly Li <[email protected]>

commit 578df99b1b0531d19af956530fe4da63d01a1604 upstream.

When md raid device (e.g. raid456) is used as backing device, read-ahead requests on a degrading and recovering md raid device might be failured immediately by md raid code, but indeed this md raid array can still be read or write for normal I/O requests. Therefore such failed read-ahead request are not real hardware failure. Further more, after degrading and recovering accomplished, read-ahead requests will be handled by md raid array again.

For such condition, I/O failures of read-ahead requests don't indicate real health status (because normal I/O still be served), they should not be counted into I/O error counter dc->io_errors.

Since there is no simple way to detect whether the backing divice is a md raid device, this patch simply ignores I/O failures for read-ahead bios on backing device, to avoid bogus backing device failure on a degrading md raid array.

---
 drivers/md/bcache/io.c |   12 ++++++++++++
 1 file changed, 12 insertions(+)
drivers/md/bcache/io.c +12 -0
@@ -58,6 +58,18 @@ void bch_count_backing_io_errors(struct
 
 	WARN_ONCE(!dc, "NULL pointer of struct cached_dev");
 
+	/*
+	 * Read-ahead requests on a degrading and recovering md raid
+	 * (e.g. raid6) device might be failured immediately by md
+	 * raid code, which is not a real hardware media failure. So
+	 * we shouldn't count failed REQ_RAHEAD bio to dc->io_errors.
+	 */
+	if (bio->bi_opf & REQ_RAHEAD) {
+		pr_warn_ratelimited("%s: Read-ahead I/O failed on backing device, ignore",
+				    dc->backing_dev_name);
+		return;
+	}
+
 	errors = atomic_add_return(1, &dc->io_errors);
 	if (errors < dc->error_limit)
 		pr_err("%s: IO error on backing device, unrecoverable",
author maintainer
#261 lore raw

PATCH 260/371 bcache: fix mistaken sysfs entry for io_error counter

In reply to Greg Kroah-Hartman

From: Coly Li <[email protected]>

commit 5461999848e0462c14f306a62923d22de820a59c upstream.

In bch_cached_dev_files[] from driver/md/bcache/sysfs.c, sysfs_errors is incorrectly inserted in. The correct entry should be sysfs_io_errors.

This patch fixes the problem and now I/O errors of cached device can be read from /sys/block/bcache<N>/bcache/io_errors.

Fixes c7b7bd07404c5 ("bcache: add io_disable to struct cached_dev")
---
 drivers/md/bcache/sysfs.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/md/bcache/sysfs.c +2 -2
@@ -180,7 +180,7 @@ SHOW(__bch_cached_dev)
 	var_print(writeback_percent);
 	sysfs_hprint(writeback_rate,
 		     wb ? atomic_long_read(&dc->writeback_rate.rate) << 9 : 0);
-	sysfs_hprint(io_errors,		atomic_read(&dc->io_errors));
+	sysfs_printf(io_errors,		"%i", atomic_read(&dc->io_errors));
 	sysfs_printf(io_error_limit,	"%i", dc->error_limit);
 	sysfs_printf(io_disable,	"%i", dc->io_disable);
 	var_print(writeback_rate_update_seconds);
@@ -464,7 +464,7 @@ static struct attribute *bch_cached_dev_
 	&sysfs_writeback_rate_p_term_inverse,
 	&sysfs_writeback_rate_minimum,
 	&sysfs_writeback_rate_debug,
-	&sysfs_errors,
+	&sysfs_io_errors,
 	&sysfs_io_error_limit,
 	&sysfs_io_disable,
 	&sysfs_dirty_data,
author maintainer
#262 lore raw

PATCH 261/371 bcache: destroy dc->writeback_write_wq if failed to create dc->writeback_thread

In reply to Greg Kroah-Hartman

From: Coly Li <[email protected]>

commit f54d801dda14942dbefa00541d10603015b7859c upstream.

Commit 9baf30972b55 ("bcache: fix for gc and write-back race") added a new work queue dc->writeback_write_wq, but forgot to destroy it in the error condition when creating dc->writeback_thread failed.

This patch destroys dc->writeback_write_wq if kthread_create() returns error pointer to dc->writeback_thread, then a memory leak is avoided.

Fixes 9baf30972b55 ("bcache: fix for gc and write-back race")
---
 drivers/md/bcache/writeback.c |    1 +
 1 file changed, 1 insertion(+)
drivers/md/bcache/writeback.c +1 -0
@@ -834,6 +834,7 @@ int bch_cached_dev_writeback_start(struc
 					      "bcache_writeback");
 	if (IS_ERR(dc->writeback_thread)) {
 		cached_dev_put(dc);
+		destroy_workqueue(dc->writeback_write_wq);
 		return PTR_ERR(dc->writeback_thread);
 	}
 	dc->writeback_running = true;
author maintainer
#263 lore raw

PATCH 262/371 Input: gtco - bounds check collection indent level

In reply to Greg Kroah-Hartman

From: Grant Hernandez <[email protected]>

commit 2a017fd82c5402b3c8df5e3d6e5165d9e6147dc1 upstream.

The GTCO tablet input driver configures itself from an HID report sent via USB during the initial enumeration process. Some debugging messages are generated during the parsing. A debugging message indentation counter is not bounds checked, leading to the ability for a specially crafted HID report to cause '-' and null bytes be written past the end of the indentation array. As long as the kernel has CONFIG_DYNAMIC_DEBUG enabled, this code will not be optimized out. This was discovered during code review after a previous syzkaller bug was found in this driver.

---
 drivers/input/tablet/gtco.c |   20 +++++++++++++++++---
 1 file changed, 17 insertions(+), 3 deletions(-)
drivers/input/tablet/gtco.c +17 -3
@@ -78,6 +78,7 @@ Scott Hill [email protected]
 
 /* Max size of a single report */
 #define REPORT_MAX_SIZE       10
+#define MAX_COLLECTION_LEVELS  10
 
 
 /* Bitmask whether pen is in range */
@@ -223,8 +224,7 @@ static void parse_hid_report_descriptor(
 	char  maintype = 'x';
 	char  globtype[12];
 	int   indent = 0;
-	char  indentstr[10] = "";
-
+	char  indentstr[MAX_COLLECTION_LEVELS + 1] = { 0 };
 
 	dev_dbg(ddev, "======>>>>>>PARSE<<<<<<======\n");
 
@@ -350,6 +350,13 @@ static void parse_hid_report_descriptor(
 			case TAG_MAIN_COL_START:
 				maintype = 'S';
 
+				if (indent == MAX_COLLECTION_LEVELS) {
+					dev_err(ddev, "Collection level %d would exceed limit of %d\n",
+						indent + 1,
+						MAX_COLLECTION_LEVELS);
+					break;
+				}
+
 				if (data == 0) {
 					dev_dbg(ddev, "======>>>>>> Physical\n");
 					strcpy(globtype, "Physical");
@@ -369,8 +376,15 @@ static void parse_hid_report_descriptor(
 				break;
 
 			case TAG_MAIN_COL_END:
-				dev_dbg(ddev, "<<<<<<======\n");
 				maintype = 'E';
+
+				if (indent == 0) {
+					dev_err(ddev, "Collection level already at zero\n");
+					break;
+				}
+
+				dev_dbg(ddev, "<<<<<<======\n");
+
 				indent--;
 				for (x = 0; x < indent; x++)
 					indentstr[x] = '-';
author maintainer
#264 lore raw

PATCH 263/371 Input: alps - dont handle ALPS cs19 trackpoint-only device

In reply to Greg Kroah-Hartman

From: Hui Wang <[email protected]>

commit 7e4935ccc3236751e5fe4bd6846f86e46bb2e427 upstream.

On a latest Lenovo laptop, the trackpoint and 3 buttons below it don't work at all, when we move the trackpoint or press those 3 buttons, the kernel will print out:
"Rejected trackstick packet from non DualPoint device"

This device is identified as an alps touchpad but the packet has trackpoint format, so the alps.c drops the packet and prints out the message above.

According to XiaoXiao's explanation, this device is named cs19 and is trackpoint-only device, its firmware is only for trackpoint, it is independent of touchpad and is a device completely different from DualPoint ones.

To drive this device with mininal changes to the existing driver, we just let the alps driver not handle this device, then the trackpoint.c will be the driver of this device if the trackpoint driver is enabled. (if not, this device will fallback to a bare PS/2 device)

With the trackpoint.c, this trackpoint and 3 buttons all work well, they have all features that the trackpoint should have, like scrolling-screen, drag-and-drop and frame-selection.

---
 drivers/input/mouse/alps.c |   32 ++++++++++++++++++++++++++++++++
 1 file changed, 32 insertions(+)
drivers/input/mouse/alps.c +32 -0
@@ -24,6 +24,7 @@
 
 #include "psmouse.h"
 #include "alps.h"
+#include "trackpoint.h"
 
 /*
  * Definitions for ALPS version 3 and 4 command mode protocol
@@ -2864,6 +2865,23 @@ static const struct alps_protocol_info *
 	return NULL;
 }
 
+static bool alps_is_cs19_trackpoint(struct psmouse *psmouse)
+{
+	u8 param[2] = { 0 };
+
+	if (ps2_command(&psmouse->ps2dev,
+			param, MAKE_PS2_CMD(0, 2, TP_READ_ID)))
+		return false;
+
+	/*
+	 * param[0] contains the trackpoint device variant_id while
+	 * param[1] contains the firmware_id. So far all alps
+	 * trackpoint-only devices have their variant_ids equal
+	 * TP_VARIANT_ALPS and their firmware_ids are in 0x20~0x2f range.
+	 */
+	return param[0] == TP_VARIANT_ALPS && (param[1] & 0x20);
+}
+
 static int alps_identify(struct psmouse *psmouse, struct alps_data *priv)
 {
 	const struct alps_protocol_info *protocol;
@@ -3165,6 +3183,20 @@ int alps_detect(struct psmouse *psmouse,
 		return error;
 
 	/*
+	 * ALPS cs19 is a trackpoint-only device, and uses different
+	 * protocol than DualPoint ones, so we return -EINVAL here and let
+	 * trackpoint.c drive this device. If the trackpoint driver is not
+	 * enabled, the device will fall back to a bare PS/2 mouse.
+	 * If ps2_command() fails here, we depend on the immediately
+	 * followed psmouse_reset() to reset the device to normal state.
+	 */
+	if (alps_is_cs19_trackpoint(psmouse)) {
+		psmouse_dbg(psmouse,
+			    "ALPS CS19 trackpoint-only device detected, ignoring\n");
+		return -EINVAL;
+	}
+
+	/*
 	 * Reset the device to make sure it is fully operational:
 	 * on some laptops, like certain Dell Latitudes, we may
 	 * fail to properly detect presence of trackstick if device
author maintainer
#265 lore raw

PATCH 264/371 Input: synaptics - whitelist Lenovo T580 SMBus intertouch

In reply to Greg Kroah-Hartman

From: Nick Black <[email protected]>

commit 1976d7d200c5a32e72293a2ada36b7b7c9d6dd6e upstream.

Adds the Lenovo T580 to the SMBus intertouch list for Synaptics touchpads. I've tested with this for a week now, and it seems a great improvement. It's also nice to have the complaint gone from dmesg.

---
 drivers/input/mouse/synaptics.c |    1 +
 1 file changed, 1 insertion(+)
drivers/input/mouse/synaptics.c +1 -0
@@ -179,6 +179,7 @@ static const char * const smbus_pnp_ids[
 	"LEN0093", /* T480 */
 	"LEN0096", /* X280 */
 	"LEN0097", /* X280 -> ALPS trackpoint */
+	"LEN009b", /* T580 */
 	"LEN200f", /* T450s */
 	"LEN2054", /* E480 */
 	"LEN2055", /* E580 */
author maintainer
#266 lore raw

PATCH 265/371 Input: alps - fix a mismatch between a condition check and its comment

In reply to Greg Kroah-Hartman

From: Hui Wang <[email protected]>

commit 771a081e44a9baa1991ef011cc453ef425591740 upstream.

In the function alps_is_cs19_trackpoint(), we check if the param[1] is in the 0x20~0x2f range, but the code we wrote for this checking is not correct:
(param[1] & 0x20) does not mean param[1] is in the range of 0x20~0x2f, it also means the param[1] is in the range of 0x30~0x3f, 0x60~0x6f...

Now fix it with a new condition checking ((param[1] & 0xf0) == 0x20).

Fixes 7e4935ccc323 ("Input: alps - don't handle ALPS cs19 trackpoint-only device")
---
 drivers/input/mouse/alps.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/input/mouse/alps.c +1 -1
@@ -2879,7 +2879,7 @@ static bool alps_is_cs19_trackpoint(stru
 	 * trackpoint-only devices have their variant_ids equal
 	 * TP_VARIANT_ALPS and their firmware_ids are in 0x20~0x2f range.
 	 */
-	return param[0] == TP_VARIANT_ALPS && (param[1] & 0x20);
+	return param[0] == TP_VARIANT_ALPS && ((param[1] & 0xf0) == 0x20);
 }
 
 static int alps_identify(struct psmouse *psmouse, struct alps_data *priv)
author maintainer
#267 lore raw

PATCH 266/371 regulator: s2mps11: Fix ERR_PTR dereference on GPIO lookup failure

In reply to Greg Kroah-Hartman

From: Krzysztof Kozlowski <[email protected]>

commit 70ca117b02f3b1c8830fe95e4e3dea2937038e11 upstream.

If devm_gpiod_get_from_of_node() call returns ERR_PTR, it is assigned into an array of GPIO descriptors and used later because such error is not treated as critical thus it is not propagated back to the probe function.

All code later expects that such GPIO descriptor is either a NULL or proper value. This later might lead to dereference of ERR_PTR.

Only devices with S2MPS14 flavor are affected (other do not control regulators with GPIOs).

Fixes 1c984942f0a4 ("regulator: s2mps11: Pass descriptor instead of GPIO number")
---
 drivers/regulator/s2mps11.c |    1 +
 1 file changed, 1 insertion(+)
drivers/regulator/s2mps11.c +1 -0
@@ -826,6 +826,7 @@ static void s2mps14_pmic_dt_parse_ext_co
 		else if (IS_ERR(gpio[reg])) {
 			dev_err(&pdev->dev, "Failed to get control GPIO for %d/%s\n",
 				reg, rdata[reg].name);
+			gpio[reg] = NULL;
 			continue;
 		}
 		if (gpio[reg])
author maintainer
#268 lore raw

PATCH 267/371 regulator: s2mps11: Fix buck7 and buck8 wrong voltages

In reply to Greg Kroah-Hartman

From: Krzysztof Kozlowski <[email protected]>

commit 16da0eb5ab6ef2dd1d33431199126e63db9997cc upstream.

On S2MPS11 device, the buck7 and buck8 regulator voltages start at 750 mV, not 600 mV. Using wrong minimal value caused shifting of these regulator values by 150 mV (e.g. buck7 usually configured to v1.35 V was reported as 1.2 V).

On most of the boards these regulators are left in default state so this was only affecting reported voltage. However if any driver wanted to change them, then effectively it would set voltage 150 mV higher than intended.

Fixes cb74685ecb39 ("regulator: s2mps11: Add samsung s2mps11 regulator driver")
---
 drivers/regulator/s2mps11.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/regulator/s2mps11.c +2 -2
@@ -372,8 +372,8 @@ static const struct regulator_desc s2mps
 	regulator_desc_s2mps11_buck1_4(4),
 	regulator_desc_s2mps11_buck5,
 	regulator_desc_s2mps11_buck67810(6, MIN_600_MV, STEP_6_25_MV),
-	regulator_desc_s2mps11_buck67810(7, MIN_600_MV, STEP_12_5_MV),
-	regulator_desc_s2mps11_buck67810(8, MIN_600_MV, STEP_12_5_MV),
+	regulator_desc_s2mps11_buck67810(7, MIN_750_MV, STEP_12_5_MV),
+	regulator_desc_s2mps11_buck67810(8, MIN_750_MV, STEP_12_5_MV),
 	regulator_desc_s2mps11_buck9,
 	regulator_desc_s2mps11_buck67810(10, MIN_750_MV, STEP_12_5_MV),
 };
author maintainer
#269 lore raw

PATCH 268/371 arm64: tegra: Update Jetson TX1 GPU regulator timings

In reply to Greg Kroah-Hartman

From: Jon Hunter <[email protected]>

commit ece6031ece2dd64d63708cfe1088016cee5b10c0 upstream.

The GPU regulator enable ramp delay for Jetson TX1 is set to 1ms which not sufficient because the enable ramp delay has been measured to be greater than 1ms. Furthermore, the downstream kernels released by NVIDIA for Jetson TX1 are using a enable ramp delay 2ms and a settling delay of 160us. Update the GPU regulator enable ramp delay for Jetson TX1 to be 2ms and add a settling delay of 160us.

Fixes 5e6b9a89afce ("arm64: tegra: Add VDD_GPU regulator to Jetson TX1")
---
 arch/arm64/boot/dts/nvidia/tegra210-p2180.dtsi |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
arch/arm64/boot/dts/nvidia/tegra210-p2180.dtsi +2 -1
@@ -322,7 +322,8 @@
 			regulator-max-microvolt = <1320000>;
 			enable-gpios = <&pmic 6 GPIO_ACTIVE_HIGH>;
 			regulator-ramp-delay = <80>;
-			regulator-enable-ramp-delay = <1000>;
+			regulator-enable-ramp-delay = <2000>;
+			regulator-settling-time-us = <160>;
 		};
 	};
 };
author maintainer
#270 lore raw

PATCH 269/371 iwlwifi: add support for hr1 RF ID

In reply to Greg Kroah-Hartman

From: Oren Givon <[email protected]>

commit 498d3eb5bfbb2e05e40005152976a7b9eadfb59c upstream.

The 22000 series FW that was meant to be used with hr is also the FW that is used for hr1 and has a different RF ID. Add support to load the hr FW when hr1 RF ID is detected.

---
 drivers/net/wireless/intel/iwlwifi/iwl-csr.h    |    1 +
 drivers/net/wireless/intel/iwlwifi/pcie/trans.c |    8 +++++---
 2 files changed, 6 insertions(+), 3 deletions(-)
drivers/net/wireless/intel/iwlwifi/iwl-csr.h +1 -0
@@ -335,6 +335,7 @@ enum {
 /* RF_ID value */
 #define CSR_HW_RF_ID_TYPE_JF		(0x00105100)
 #define CSR_HW_RF_ID_TYPE_HR		(0x0010A000)
+#define CSR_HW_RF_ID_TYPE_HR1		(0x0010c100)
 #define CSR_HW_RF_ID_TYPE_HRCDB		(0x00109F00)
 #define CSR_HW_RF_ID_TYPE_GF		(0x0010D000)
 
drivers/net/wireless/intel/iwlwifi/pcie/trans.c +5 -3
@@ -3562,9 +3562,11 @@ struct iwl_trans *iwl_trans_pcie_alloc(s
 			trans->cfg = &iwlax210_2ax_cfg_so_gf_a0;
 		}
 	} else if (cfg == &iwl_ax101_cfg_qu_hr) {
-		if (CSR_HW_RF_ID_TYPE_CHIP_ID(trans->hw_rf_id) ==
-		    CSR_HW_RF_ID_TYPE_CHIP_ID(CSR_HW_RF_ID_TYPE_HR) &&
-		    trans->hw_rev == CSR_HW_REV_TYPE_QNJ_B0) {
+		if ((CSR_HW_RF_ID_TYPE_CHIP_ID(trans->hw_rf_id) ==
+		     CSR_HW_RF_ID_TYPE_CHIP_ID(CSR_HW_RF_ID_TYPE_HR) &&
+		     trans->hw_rev == CSR_HW_REV_TYPE_QNJ_B0) ||
+		    (CSR_HW_RF_ID_TYPE_CHIP_ID(trans->hw_rf_id) ==
+		     CSR_HW_RF_ID_TYPE_CHIP_ID(CSR_HW_RF_ID_TYPE_HR1))) {
 			trans->cfg = &iwl22000_2ax_cfg_qnj_hr_b0;
 		} else if (CSR_HW_RF_ID_TYPE_CHIP_ID(trans->hw_rf_id) ==
 		    CSR_HW_RF_ID_TYPE_CHIP_ID(CSR_HW_RF_ID_TYPE_HR)) {
author maintainer
#271 lore raw

PATCH 270/371 iwlwifi: pcie: dont service an interrupt that was masked

In reply to Greg Kroah-Hartman

From: Emmanuel Grumbach <[email protected]>

commit 3b57a10ca14c619707398dc58fe5ece18c95b20b upstream.

Sometimes the register status can include interrupts that were masked. We can, for example, get the RF-Kill bit set in the interrupt status register although this interrupt was masked. Then if we get the ALIVE interrupt (for example) that was not masked, we need to *not* service the RF-Kill interrupt.
Fix this in the MSI-X interrupt handler.

---
 drivers/net/wireless/intel/iwlwifi/pcie/rx.c |   27 +++++++++++++++++++++------
 1 file changed, 21 insertions(+), 6 deletions(-)
drivers/net/wireless/intel/iwlwifi/pcie/rx.c +21 -6
@@ -2113,10 +2113,18 @@ irqreturn_t iwl_pcie_irq_msix_handler(in
 		return IRQ_NONE;
 	}
 
-	if (iwl_have_debug_level(IWL_DL_ISR))
-		IWL_DEBUG_ISR(trans, "ISR inta_fh 0x%08x, enabled 0x%08x\n",
-			      inta_fh,
+	if (iwl_have_debug_level(IWL_DL_ISR)) {
+		IWL_DEBUG_ISR(trans,
+			      "ISR inta_fh 0x%08x, enabled (sw) 0x%08x (hw) 0x%08x\n",
+			      inta_fh, trans_pcie->fh_mask,
 			      iwl_read32(trans, CSR_MSIX_FH_INT_MASK_AD));
+		if (inta_fh & ~trans_pcie->fh_mask)
+			IWL_DEBUG_ISR(trans,
+				      "We got a masked interrupt (0x%08x)\n",
+				      inta_fh & ~trans_pcie->fh_mask);
+	}
+
+	inta_fh &= trans_pcie->fh_mask;
 
 	if ((trans_pcie->shared_vec_mask & IWL_SHARED_IRQ_NON_RX) &&
 	    inta_fh & MSIX_FH_INT_CAUSES_Q0) {
@@ -2156,11 +2164,18 @@ irqreturn_t iwl_pcie_irq_msix_handler(in
 	}
 
 	/* After checking FH register check HW register */
-	if (iwl_have_debug_level(IWL_DL_ISR))
+	if (iwl_have_debug_level(IWL_DL_ISR)) {
 		IWL_DEBUG_ISR(trans,
-			      "ISR inta_hw 0x%08x, enabled 0x%08x\n",
-			      inta_hw,
+			      "ISR inta_hw 0x%08x, enabled (sw) 0x%08x (hw) 0x%08x\n",
+			      inta_hw, trans_pcie->hw_mask,
 			      iwl_read32(trans, CSR_MSIX_HW_INT_MASK_AD));
+		if (inta_hw & ~trans_pcie->hw_mask)
+			IWL_DEBUG_ISR(trans,
+				      "We got a masked interrupt 0x%08x\n",
+				      inta_hw & ~trans_pcie->hw_mask);
+	}
+
+	inta_hw &= trans_pcie->hw_mask;
 
 	/* Alive notification via Rx interrupt will do the real work */
 	if (inta_hw & MSIX_HW_INT_CAUSES_REG_ALIVE) {
author maintainer
#272 lore raw

PATCH 271/371 iwlwifi: pcie: fix ALIVE interrupt handling for gen2 devices w/o MSI-X

In reply to Greg Kroah-Hartman

From: Emmanuel Grumbach <[email protected]>

commit ec46ae30245ecb41d73f8254613db07c653fb498 upstream.

We added code to restock the buffer upon ALIVE interrupt when MSI-X is disabled. This was added as part of the context info code. This code was added only if the ISR debug level is set which is very unlikely to be related.
Move this code to run even when the ISR debug level is not set.

Note that gen2 devices work with MSI-X in most cases so that this path is seldom used.

---
 drivers/net/wireless/intel/iwlwifi/pcie/rx.c |   34 ++++++++++++---------------
 1 file changed, 16 insertions(+), 18 deletions(-)
drivers/net/wireless/intel/iwlwifi/pcie/rx.c +16 -18
@@ -1832,25 +1832,23 @@ irqreturn_t iwl_pcie_irq_handler(int irq
 		goto out;
 	}
 
-	if (iwl_have_debug_level(IWL_DL_ISR)) {
-		/* NIC fires this, but we don't use it, redundant with WAKEUP */
-		if (inta & CSR_INT_BIT_SCD) {
-			IWL_DEBUG_ISR(trans,
-				      "Scheduler finished to transmit the frame/frames.\n");
-			isr_stats->sch++;
-		}
+	/* NIC fires this, but we don't use it, redundant with WAKEUP */
+	if (inta & CSR_INT_BIT_SCD) {
+		IWL_DEBUG_ISR(trans,
+			      "Scheduler finished to transmit the frame/frames.\n");
+		isr_stats->sch++;
+	}
 
-		/* Alive notification via Rx interrupt will do the real work */
-		if (inta & CSR_INT_BIT_ALIVE) {
-			IWL_DEBUG_ISR(trans, "Alive interrupt\n");
-			isr_stats->alive++;
-			if (trans->cfg->gen2) {
-				/*
-				 * We can restock, since firmware configured
-				 * the RFH
-				 */
-				iwl_pcie_rxmq_restock(trans, trans_pcie->rxq);
-			}
+	/* Alive notification via Rx interrupt will do the real work */
+	if (inta & CSR_INT_BIT_ALIVE) {
+		IWL_DEBUG_ISR(trans, "Alive interrupt\n");
+		isr_stats->alive++;
+		if (trans->cfg->gen2) {
+			/*
+			 * We can restock, since firmware configured
+			 * the RFH
+			 */
+			iwl_pcie_rxmq_restock(trans, trans_pcie->rxq);
 		}
 	}
 
author maintainer
#273 lore raw

PATCH 272/371 iwlwifi: dont WARN when calling iwl_get_shared_mem_conf with RF-Kill

In reply to Greg Kroah-Hartman

From: Emmanuel Grumbach <[email protected]>

commit 0d53cfd0cca3c729a089c39eef0e7d8ae7662974 upstream.

iwl_mvm_send_cmd returns 0 when the command won't be sent because RF-Kill is asserted. Do the same when we call iwl_get_shared_mem_conf since it is not sent through iwl_mvm_send_cmd but directly calls the transport layer.

---
 drivers/net/wireless/intel/iwlwifi/fw/smem.c |   12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)
drivers/net/wireless/intel/iwlwifi/fw/smem.c +9 -3
@@ -8,7 +8,7 @@
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
  * Copyright(c) 2016 - 2017 Intel Deutschland GmbH
- * Copyright(c) 2018 Intel Corporation
+ * Copyright(c) 2018 - 2019 Intel Corporation
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -31,7 +31,7 @@
  * Copyright(c) 2012 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2013 - 2015 Intel Mobile Communications GmbH
  * Copyright(c) 2016 - 2017 Intel Deutschland GmbH
- * Copyright(c) 2018 Intel Corporation
+ * Copyright(c) 2018 - 2019 Intel Corporation
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -134,6 +134,7 @@ void iwl_get_shared_mem_conf(struct iwl_
 		.len = { 0, },
 	};
 	struct iwl_rx_packet *pkt;
+	int ret;
 
 	if (fw_has_capa(&fwrt->fw->ucode_capa,
 			IWL_UCODE_TLV_CAPA_EXTEND_SHARED_MEM_CFG))
@@ -141,8 +142,13 @@ void iwl_get_shared_mem_conf(struct iwl_
 	else
 		cmd.id = SHARED_MEM_CFG;
 
-	if (WARN_ON(iwl_trans_send_cmd(fwrt->trans, &cmd)))
+	ret = iwl_trans_send_cmd(fwrt->trans, &cmd);
+
+	if (ret) {
+		WARN(ret != -ERFKILL,
+		     "Could not send the SMEM command: %d\n", ret);
 		return;
+	}
 
 	pkt = cmd.resp_pkt;
 	if (fwrt->trans->cfg->device_family >= IWL_DEVICE_FAMILY_22000)
author maintainer
#274 lore raw

PATCH 273/371 iwlwifi: fix RF-Kill interrupt while FW load for gen2 devices

In reply to Greg Kroah-Hartman

From: Emmanuel Grumbach <[email protected]>

commit ed3e4c6d3cd8f093a3636cb05492429fe2af228d upstream.

Newest devices have a new firmware load mechanism. This mechanism is called the context info. It means that the driver doesn't need to load the sections of the firmware. The driver rather prepares a place in DRAM, with pointers to the relevant sections of the firmware, and the firmware loads itself.
At the end of the process, the firmware sends the ALIVE interrupt. This is different from the previous scheme in which the driver expected the FH_TX interrupt after each section being transferred over the DMA.

In order to support this new flow, we enabled all the interrupts. This broke the assumption that we have in the code that the RF-Kill interrupt can't interrupt the firmware load flow.

Change the context info flow to enable only the ALIVE interrupt, and re-enable all the other interrupts only after the firmware is alive. Then, we won't see the RF-Kill interrupt until then. Getting the RF-Kill interrupt while loading the firmware made us kill the firmware while it is loading and we ended up dumping garbage instead of the firmware state.

Re-enable the ALIVE | RX interrupts from the ISR when we get the ALIVE interrupt to be able to get the RX interrupt that comes immediately afterwards for the ALIVE notification. This is needed for non MSI-X only.

---
 drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c |    2 -
 drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info.c      |    2 -
 drivers/net/wireless/intel/iwlwifi/pcie/internal.h       |   27 +++++++++++++++
 drivers/net/wireless/intel/iwlwifi/pcie/rx.c             |    5 ++
 drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c     |    9 +++++
 5 files changed, 43 insertions(+), 2 deletions(-)
drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info-gen3.c +1 -1
@@ -168,7 +168,7 @@ int iwl_pcie_ctxt_info_gen3_init(struct
 
 	memcpy(iml_img, trans->iml, trans->iml_len);
 
-	iwl_enable_interrupts(trans);
+	iwl_enable_fw_load_int_ctx_info(trans);
 
 	/* kick FW self load */
 	iwl_write64(trans, CSR_CTXT_INFO_ADDR,
drivers/net/wireless/intel/iwlwifi/pcie/ctxt-info.c +1 -1
@@ -222,7 +222,7 @@ int iwl_pcie_ctxt_info_init(struct iwl_t
 
 	trans_pcie->ctxt_info = ctxt_info;
 
-	iwl_enable_interrupts(trans);
+	iwl_enable_fw_load_int_ctx_info(trans);
 
 	/* Configure debug, if exists */
 	if (iwl_pcie_dbg_on(trans))
drivers/net/wireless/intel/iwlwifi/pcie/internal.h +27 -0
@@ -894,6 +894,33 @@ static inline void iwl_enable_fw_load_in
 	}
 }
 
+static inline void iwl_enable_fw_load_int_ctx_info(struct iwl_trans *trans)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+
+	IWL_DEBUG_ISR(trans, "Enabling ALIVE interrupt only\n");
+
+	if (!trans_pcie->msix_enabled) {
+		/*
+		 * When we'll receive the ALIVE interrupt, the ISR will call
+		 * iwl_enable_fw_load_int_ctx_info again to set the ALIVE
+		 * interrupt (which is not really needed anymore) but also the
+		 * RX interrupt which will allow us to receive the ALIVE
+		 * notification (which is Rx) and continue the flow.
+		 */
+		trans_pcie->inta_mask =  CSR_INT_BIT_ALIVE | CSR_INT_BIT_FH_RX;
+		iwl_write32(trans, CSR_INT_MASK, trans_pcie->inta_mask);
+	} else {
+		iwl_enable_hw_int_msk_msix(trans,
+					   MSIX_HW_INT_CAUSES_REG_ALIVE);
+		/*
+		 * Leave all the FH causes enabled to get the ALIVE
+		 * notification.
+		 */
+		iwl_enable_fh_int_msk_msix(trans, trans_pcie->fh_init_mask);
+	}
+}
+
 static inline u16 iwl_pcie_get_cmd_index(const struct iwl_txq *q, u32 index)
 {
 	return index & (q->n_window - 1);
drivers/net/wireless/intel/iwlwifi/pcie/rx.c +5 -0
@@ -1850,6 +1850,8 @@ irqreturn_t iwl_pcie_irq_handler(int irq
 			 */
 			iwl_pcie_rxmq_restock(trans, trans_pcie->rxq);
 		}
+
+		handled |= CSR_INT_BIT_ALIVE;
 	}
 
 	/* Safely ignore these bits for debug checks below */
@@ -1968,6 +1970,9 @@ irqreturn_t iwl_pcie_irq_handler(int irq
 	/* Re-enable RF_KILL if it occurred */
 	else if (handled & CSR_INT_BIT_RF_KILL)
 		iwl_enable_rfkill_int(trans);
+	/* Re-enable the ALIVE / Rx interrupt if it occurred */
+	else if (handled & (CSR_INT_BIT_ALIVE | CSR_INT_BIT_FH_RX))
+		iwl_enable_fw_load_int_ctx_info(trans);
 	spin_unlock(&trans_pcie->irq_lock);
 
 out:
drivers/net/wireless/intel/iwlwifi/pcie/trans-gen2.c +9 -0
@@ -272,6 +272,15 @@ void iwl_trans_pcie_gen2_fw_alive(struct
 	 * paging memory cannot be freed included since FW will still use it
 	 */
 	iwl_pcie_ctxt_info_free(trans);
+
+	/*
+	 * Re-enable all the interrupts, including the RF-Kill one, now that
+	 * the firmware is alive.
+	 */
+	iwl_enable_interrupts(trans);
+	mutex_lock(&trans_pcie->mutex);
+	iwl_pcie_check_hw_rf_kill(trans);
+	mutex_unlock(&trans_pcie->mutex);
 }
 
 int iwl_trans_pcie_gen2_start_fw(struct iwl_trans *trans,
author maintainer
#275 lore raw

PATCH 274/371 iwlwifi: mvm: delay GTK setting in FW in AP mode

In reply to Greg Kroah-Hartman

From: Johannes Berg <[email protected]>

commit c56e00a3feaee2b46b7d33875fb7f52efd30241f upstream.

In AP (and IBSS) mode, we can only set GTKs to firmware after we have sent down the multicast station, but this we can only do after we've enabled beaconing, etc.

However, during rfkill exit, hostapd will configure the keys before starting the AP, and cfg80211/mac80211 accept it happily.

On earlier devices, this didn't bother us as GTK TX wasn't really handled in firmware, we just put the key material into the TX cmd and thus it only mattered when we actually transmitted a frame.

On newer devices, however, the firmware needs to track all of this and that doesn't work if we add the key before the (multicast) sta it belongs to.

To fix this, keep a list of keys to add during AP enable, and call the function there.

---
 drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c |   53 +++++++++++++++++++++-
 drivers/net/wireless/intel/iwlwifi/mvm/mvm.h      |    3 +
 2 files changed, 54 insertions(+), 2 deletions(-)
drivers/net/wireless/intel/iwlwifi/mvm/mac80211.c +51 -2
@@ -207,6 +207,12 @@ static const struct cfg80211_pmsr_capabi
 	},
 };
 
+static int iwl_mvm_mac_set_key(struct ieee80211_hw *hw,
+			       enum set_key_cmd cmd,
+			       struct ieee80211_vif *vif,
+			       struct ieee80211_sta *sta,
+			       struct ieee80211_key_conf *key);
+
 void iwl_mvm_ref(struct iwl_mvm *mvm, enum iwl_mvm_ref_type ref_type)
 {
 	if (!iwl_mvm_is_d0i3_supported(mvm))
@@ -2535,7 +2541,7 @@ static int iwl_mvm_start_ap_ibss(struct
 {
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
-	int ret;
+	int ret, i;
 
 	/*
 	 * iwl_mvm_mac_ctxt_add() might read directly from the device
@@ -2609,6 +2615,20 @@ static int iwl_mvm_start_ap_ibss(struct
 	/* must be set before quota calculations */
 	mvmvif->ap_ibss_active = true;
 
+	/* send all the early keys to the device now */
+	for (i = 0; i < ARRAY_SIZE(mvmvif->ap_early_keys); i++) {
+		struct ieee80211_key_conf *key = mvmvif->ap_early_keys[i];
+
+		if (!key)
+			continue;
+
+		mvmvif->ap_early_keys[i] = NULL;
+
+		ret = iwl_mvm_mac_set_key(hw, SET_KEY, vif, NULL, key);
+		if (ret)
+			goto out_quota_failed;
+	}
+
 	if (vif->type == NL80211_IFTYPE_AP && !vif->p2p) {
 		iwl_mvm_vif_set_low_latency(mvmvif, true,
 					    LOW_LATENCY_VIF_TYPE);
@@ -3378,11 +3398,12 @@ static int iwl_mvm_mac_set_key(struct ie
 			       struct ieee80211_sta *sta,
 			       struct ieee80211_key_conf *key)
 {
+	struct iwl_mvm_vif *mvmvif = iwl_mvm_vif_from_mac80211(vif);
 	struct iwl_mvm *mvm = IWL_MAC80211_GET_MVM(hw);
 	struct iwl_mvm_sta *mvmsta;
 	struct iwl_mvm_key_pn *ptk_pn;
 	int keyidx = key->keyidx;
-	int ret;
+	int ret, i;
 	u8 key_offset;
 
 	if (iwlwifi_mod_params.swcrypto) {
@@ -3455,6 +3476,22 @@ static int iwl_mvm_mac_set_key(struct ie
 				key->hw_key_idx = STA_KEY_IDX_INVALID;
 				break;
 			}
+
+			if (!mvmvif->ap_ibss_active) {
+				for (i = 0;
+				     i < ARRAY_SIZE(mvmvif->ap_early_keys);
+				     i++) {
+					if (!mvmvif->ap_early_keys[i]) {
+						mvmvif->ap_early_keys[i] = key;
+						break;
+					}
+				}
+
+				if (i >= ARRAY_SIZE(mvmvif->ap_early_keys))
+					ret = -ENOSPC;
+
+				break;
+			}
 		}
 
 		/* During FW restart, in order to restore the state as it was,
@@ -3523,6 +3560,18 @@ static int iwl_mvm_mac_set_key(struct ie
 
 		break;
 	case DISABLE_KEY:
+		ret = -ENOENT;
+		for (i = 0; i < ARRAY_SIZE(mvmvif->ap_early_keys); i++) {
+			if (mvmvif->ap_early_keys[i] == key) {
+				mvmvif->ap_early_keys[i] = NULL;
+				ret = 0;
+			}
+		}
+
+		/* found in pending list - don't do anything else */
+		if (ret == 0)
+			break;
+
 		if (key->hw_key_idx == STA_KEY_IDX_INVALID) {
 			ret = 0;
 			break;
drivers/net/wireless/intel/iwlwifi/mvm/mvm.h +3 -0
@@ -498,6 +498,9 @@ struct iwl_mvm_vif {
 	netdev_features_t features;
 
 	struct iwl_probe_resp_data __rcu *probe_resp_data;
+
+	/* we can only have 2 GTK + 2 IGTK active at a time */
+	struct ieee80211_key_conf *ap_early_keys[4];
 };
 
 static inline struct iwl_mvm_vif *
author maintainer
#276 lore raw

PATCH 275/371 iwlwifi: mvm: clear rfkill_safe_init_done when we start the firmware

In reply to Greg Kroah-Hartman

From: Emmanuel Grumbach <[email protected]>

commit 940225628652b340b2bfe99f42f3d2db9fd9ce6c upstream.

Otherwise it'll stay set forever which is clearly buggy.

---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/net/wireless/intel/iwlwifi/mvm/fw.c +6 -2
@@ -419,6 +419,8 @@ static int iwl_run_unified_mvm_ucode(str
 
 	lockdep_assert_held(&mvm->mutex);
 
+	mvm->rfkill_safe_init_done = false;
+
 	iwl_init_notification_wait(&mvm->notif_wait,
 				   &init_wait,
 				   init_complete,
@@ -537,8 +539,7 @@ int iwl_run_init_mvm_ucode(struct iwl_mv
 
 	lockdep_assert_held(&mvm->mutex);
 
-	if (WARN_ON_ONCE(mvm->rfkill_safe_init_done))
-		return 0;
+	mvm->rfkill_safe_init_done = false;
 
 	iwl_init_notification_wait(&mvm->notif_wait,
 				   &calib_wait,
@@ -1108,10 +1109,13 @@ static int iwl_mvm_load_rt_fw(struct iwl
 
 	iwl_fw_dbg_apply_point(&mvm->fwrt, IWL_FW_INI_APPLY_EARLY);
 
+	mvm->rfkill_safe_init_done = false;
 	ret = iwl_mvm_load_ucode_wait_alive(mvm, IWL_UCODE_REGULAR);
 	if (ret)
 		return ret;
 
+	mvm->rfkill_safe_init_done = true;
+
 	iwl_fw_dbg_apply_point(&mvm->fwrt, IWL_FW_INI_APPLY_AFTER_ALIVE);
 
 	return iwl_init_paging(&mvm->fwrt, mvm->fwrt.cur_fw_img);
author maintainer
#277 lore raw

PATCH 276/371 opp: Dont use IS_ERR on invalid supplies

In reply to Greg Kroah-Hartman

From: Dmitry Osipenko <[email protected]>

commit 560d1bcad715c215e7ffe5d7cffe045974b623d0 upstream.

_set_opp_custom() receives a set of OPP supplies as its arguments and the caller of it passes NULL when the supplies are not valid. But _set_opp_custom(), by mistake, checks for error by performing IS_ERR(old_supply) on it which will always evaluate to false.

The problem was spotted during of testing of upcoming update for the NVIDIA Tegra CPUFreq driver.

Cc stable <[email protected]>
Fixes 7e535993fa4f ("OPP: Separate out custom OPP handler specific code")
[ Viresh: Massaged changelog ]
---
 drivers/opp/core.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/opp/core.c +1 -1
@@ -631,7 +631,7 @@ static int _set_opp_custom(const struct
 
 	data->old_opp.rate = old_freq;
 	size = sizeof(*old_supply) * opp_table->regulator_count;
-	if (IS_ERR(old_supply))
+	if (!old_supply)
 		memset(data->old_opp.supplies, 0, size);
 	else
 		memcpy(data->old_opp.supplies, old_supply, size);
author maintainer
#278 lore raw

PATCH 277/371 arm64: Fix interrupt tracing in the presence of NMIs

In reply to Greg Kroah-Hartman

From: Julien Thierry <[email protected]>

commit 17ce302f3117e9518395847a3120c8a108b587b8 upstream.

In the presence of any form of instrumentation, nmi_enter() should be done before calling any traceable code and any instrumentation code.

Currently, nmi_enter() is done in handle_domain_nmi(), which is much too late as instrumentation code might get called before. Move the nmi_enter/exit() calls to the arch IRQ vector handler.

On arm64, it is not possible to know if the IRQ vector handler was called because of an NMI before acknowledging the interrupt. However, It is possible to know whether normal interrupts could be taken in the interrupted context (i.e. if taking an NMI in that context could introduce a potential race condition).

When interrupting a context with IRQs disabled, call nmi_enter() as soon as possible. In contexts with IRQs enabled, defer this to the interrupt controller, which is in a better position to know if an interrupt taken is an NMI.

Fixes bc3c03ccb464 ("arm64: Enable the support of pseudo-NMIs")
Cc <[email protected]> # 5.1.x-
Cc Will Deacon <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc Jason Cooper <[email protected]>
Cc Mark Rutland <[email protected]>
---
 arch/arm64/kernel/entry.S    |   44 ++++++++++++++++++++++++++++++++-----------
 arch/arm64/kernel/irq.c      |   17 ++++++++++++++++
 drivers/irqchip/irq-gic-v3.c |    7 ++++++
 kernel/irq/irqdesc.c         |    8 +++++--
 4 files changed, 63 insertions(+), 13 deletions(-)
arch/arm64/kernel/entry.S +33 -11
@@ -420,6 +420,20 @@ tsk	.req	x28		// current thread_info
 	irq_stack_exit
 	.endm
 
+#ifdef CONFIG_ARM64_PSEUDO_NMI
+	/*
+	 * Set res to 0 if irqs were unmasked in interrupted context.
+	 * Otherwise set res to non-0 value.
+	 */
+	.macro	test_irqs_unmasked res:req, pmr:req
+alternative_if ARM64_HAS_IRQ_PRIO_MASKING
+	sub	\res, \pmr, #GIC_PRIO_IRQON
+alternative_else
+	mov	\res, xzr
+alternative_endif
+	.endm
+#endif
+
 	.text
 
 /*
@@ -616,19 +630,19 @@ ENDPROC(el1_sync)
 el1_irq:
 	kernel_entry 1
 	enable_da_f
-#ifdef CONFIG_TRACE_IRQFLAGS
+
 #ifdef CONFIG_ARM64_PSEUDO_NMI
 alternative_if ARM64_HAS_IRQ_PRIO_MASKING
 	ldr	x20, [sp, #S_PMR_SAVE]
-alternative_else
-	mov	x20, #GIC_PRIO_IRQON
-alternative_endif
-	cmp	x20, #GIC_PRIO_IRQOFF
-	/* Irqs were disabled, don't trace */
-	b.ls	1f
+alternative_else_nop_endif
+	test_irqs_unmasked	res=x0, pmr=x20
+	cbz	x0, 1f
+	bl	asm_nmi_enter
+1:
 #endif
+
+#ifdef CONFIG_TRACE_IRQFLAGS
 	bl	trace_hardirqs_off
-1:
 #endif
 
 	irq_handler
@@ -647,14 +661,22 @@ alternative_else_nop_endif
 	bl	preempt_schedule_irq		// irq en/disable is done inside
 1:
 #endif
-#ifdef CONFIG_TRACE_IRQFLAGS
+
 #ifdef CONFIG_ARM64_PSEUDO_NMI
 	/*
 	 * if IRQs were disabled when we received the interrupt, we have an NMI
 	 * and we are not re-enabling interrupt upon eret. Skip tracing.
 	 */
-	cmp	x20, #GIC_PRIO_IRQOFF
-	b.ls	1f
+	test_irqs_unmasked	res=x0, pmr=x20
+	cbz	x0, 1f
+	bl	asm_nmi_exit
+1:
+#endif
+
+#ifdef CONFIG_TRACE_IRQFLAGS
+#ifdef CONFIG_ARM64_PSEUDO_NMI
+	test_irqs_unmasked	res=x0, pmr=x20
+	cbnz	x0, 1f
 #endif
 	bl	trace_hardirqs_on
 1:
arch/arm64/kernel/irq.c +17 -0
@@ -27,8 +27,10 @@
 #include <linux/smp.h>
 #include <linux/init.h>
 #include <linux/irqchip.h>
+#include <linux/kprobes.h>
 #include <linux/seq_file.h>
 #include <linux/vmalloc.h>
+#include <asm/daifflags.h>
 #include <asm/vmap_stack.h>
 
 unsigned long irq_err_count;
@@ -76,3 +78,18 @@ void __init init_IRQ(void)
 	if (!handle_arch_irq)
 		panic("No interrupt controller found.");
 }
+
+/*
+ * Stubs to make nmi_enter/exit() code callable from ASM
+ */
+asmlinkage void notrace asm_nmi_enter(void)
+{
+	nmi_enter();
+}
+NOKPROBE_SYMBOL(asm_nmi_enter);
+
+asmlinkage void notrace asm_nmi_exit(void)
+{
+	nmi_exit();
+}
+NOKPROBE_SYMBOL(asm_nmi_exit);
drivers/irqchip/irq-gic-v3.c +7 -0
@@ -472,8 +472,12 @@ static void gic_deactivate_unhandled(u32
 
 static inline void gic_handle_nmi(u32 irqnr, struct pt_regs *regs)
 {
+	bool irqs_enabled = interrupts_enabled(regs);
 	int err;
 
+	if (irqs_enabled)
+		nmi_enter();
+
 	if (static_branch_likely(&supports_deactivate_key))
 		gic_write_eoir(irqnr);
 	/*
@@ -485,6 +489,9 @@ static inline void gic_handle_nmi(u32 ir
 	err = handle_domain_nmi(gic_data.domain, irqnr, regs);
 	if (err)
 		gic_deactivate_unhandled(irqnr);
+
+	if (irqs_enabled)
+		nmi_exit();
 }
 
 static asmlinkage void __exception_irq_entry gic_handle_irq(struct pt_regs *regs)
kernel/irq/irqdesc.c +6 -2
@@ -679,6 +679,8 @@ int __handle_domain_irq(struct irq_domai
  * @hwirq:	The HW irq number to convert to a logical one
  * @regs:	Register file coming from the low-level handling code
  *
+ *		This function must be called from an NMI context.
+ *
  * Returns:	0 on success, or -EINVAL if conversion has failed
  */
 int handle_domain_nmi(struct irq_domain *domain, unsigned int hwirq,
@@ -688,7 +690,10 @@ int handle_domain_nmi(struct irq_domain
 	unsigned int irq;
 	int ret = 0;
 
-	nmi_enter();
+	/*
+	 * NMI context needs to be setup earlier in order to deal with tracing.
+	 */
+	WARN_ON(!in_nmi());
 
 	irq = irq_find_mapping(domain, hwirq);
 
@@ -701,7 +706,6 @@ int handle_domain_nmi(struct irq_domain
 	else
 		ret = -EINVAL;
 
-	nmi_exit();
 	set_irq_regs(old_regs);
 	return ret;
 }
author maintainer
#279 lore raw

PATCH 278/371 NFSv4: Handle the special Linux file open access mode

In reply to Greg Kroah-Hartman

From: Trond Myklebust <[email protected]>

commit 44942b4e457beda00981f616402a1a791e8c616e upstream.

According to the open() manpage, Linux reserves the access mode 3 to mean "check for read and write permission on the file and return a file descriptor that can't be used for reading or writing."

Currently, the NFSv4 code will ask the server to open the file, and will use an incorrect share access mode of 0. Since it has an incorrect share access mode, the client later forgets to send a corresponding close, meaning it can leak stateids on the server.

Fixes ce4ef7c0a8a05 ("NFS: Split out NFS v4 file operations")
---
 fs/nfs/inode.c    |    1 +
 fs/nfs/nfs4file.c |    2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)
fs/nfs/inode.c +1 -0
@@ -1094,6 +1094,7 @@ int nfs_open(struct inode *inode, struct
 	nfs_fscache_open_file(inode, filp);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(nfs_open);
 
 /*
  * This function is called whenever some part of NFS notices that
fs/nfs/nfs4file.c +1 -1
@@ -49,7 +49,7 @@ nfs4_file_open(struct inode *inode, stru
 		return err;
 
 	if ((openflags & O_ACCMODE) == 3)
-		openflags--;
+		return nfs_open(inode, filp);
 
 	/* We can't create new files here */
 	openflags &= ~(O_CREAT|O_EXCL);
author maintainer
#280 lore raw

PATCH 279/371 Revert "NFS: readdirplus optimization by cache mechanism" (memleak)

In reply to Greg Kroah-Hartman

From: Max Kellermann <[email protected]>

commit db531db951f950b86d274cc8ed7b21b9e2240036 upstream.

This reverts commit be4c2d4723a4a637f0d1b4f7c66447141a4b3564.

That commit caused a severe memory leak in nfs_readdir_make_qstr().

When listing a directory with more than 100 files (this is how many struct nfs_cache_array_entry elements fit in one 4kB page), all allocated file name strings past those 100 leak.

The root of the leakage is that those string pointers are managed in pages which are never linked into the page cache.

fs/nfs/dir.c puts pages into the page cache by calling read_cache_page(); the callback function nfs_readdir_filler() will then fill the given page struct which was passed to it, which is already linked in the page cache (by do_read_cache_page() calling add_to_page_cache_lru()).

Commit be4c2d4723a4 added another (local) array of allocated pages, to be filled with more data, instead of discarding excess items received from the NFS server. Those additional pages can be used by the next nfs_readdir_filler() call (from within the same nfs_readdir() call).

The leak happens when some of those additional pages are never used (copied to the page cache using copy_highpage()). The pages will be freed by nfs_readdir_free_pages(), but their contents will not. The commit did not invoke nfs_readdir_clear_array() (and doing so would have been dangerous, because it did not track which of those pages were already copied to the page cache, risking double free bugs).

How to reproduce the leak:

- Use a kernel with CONFIG_SLUB_DEBUG_ON.

- Create a directory on a NFS mount with more than 100 files with names long enough to use the "kmalloc-32" slab (so we can easily look up the allocation counts):

  for i in `seq 110`; do touch ${i}_0123456789abcdef; done

- Drop all caches:

echo 3 >/proc/sys/vm/drop_caches

- Check the allocation counter:

grep nfs_readdir /sys/kernel/slab/kmalloc-32/alloc_calls
30564391 nfs_readdir_add_to_array+0x73/0xd0 age=534558/4791307/6540952 pid=370-1048386 cpus=0-47 nodes=0-1

- Request a directory listing and check the allocation counters again:

ls
[...]
grep nfs_readdir /sys/kernel/slab/kmalloc-32/alloc_calls
30564511 nfs_readdir_add_to_array+0x73/0xd0 age=207/4792999/6542663 pid=370-1048386 cpus=0-47 nodes=0-1

There are now 120 new allocations.

- Drop all caches and check the counters again:

echo 3 >/proc/sys/vm/drop_caches
grep nfs_readdir /sys/kernel/slab/kmalloc-32/alloc_calls
30564401 nfs_readdir_add_to_array+0x73/0xd0 age=735/4793524/6543176 pid=370-1048386 cpus=0-47 nodes=0-1

110 allocations are gone, but 10 have leaked and will never be freed.

Unhelpfully, those allocations are explicitly excluded from KMEMLEAK, that's why my initial attempts with KMEMLEAK were not successful:

	/*
	 * Avoid a kmemleak false positive. The pointer to the name is stored
	 * in a page cache page which kmemleak does not scan.
	 */
	kmemleak_not_leak(string->name);

It would be possible to solve this bug without reverting the whole commit:

- keep track of which pages were not used, and call nfs_readdir_clear_array() on them, or
- manually link those pages into the page cache

But for now I have decided to just revert the commit, because the real fix would require complex considerations, risking more dangerous (crash) bugs, which may seem unsuitable for the stable branches.

---
 fs/nfs/dir.c      |   90 +++---------------------------------------------------
 fs/nfs/internal.h |    3 -
 2 files changed, 7 insertions(+), 86 deletions(-)
fs/nfs/dir.c +6 -84
@@ -139,19 +139,12 @@ struct nfs_cache_array {
 	struct nfs_cache_array_entry array[0];
 };
 
-struct readdirvec {
-	unsigned long nr;
-	unsigned long index;
-	struct page *pages[NFS_MAX_READDIR_RAPAGES];
-};
-
 typedef int (*decode_dirent_t)(struct xdr_stream *, struct nfs_entry *, bool);
 typedef struct {
 	struct file	*file;
 	struct page	*page;
 	struct dir_context *ctx;
 	unsigned long	page_index;
-	struct readdirvec pvec;
 	u64		*dir_cookie;
 	u64		last_cookie;
 	loff_t		current_index;
@@ -531,10 +524,6 @@ int nfs_readdir_page_filler(nfs_readdir_
 	struct nfs_cache_array *array;
 	unsigned int count = 0;
 	int status;
-	int max_rapages = NFS_MAX_READDIR_RAPAGES;
-
-	desc->pvec.index = desc->page_index;
-	desc->pvec.nr = 0;
 
 	scratch = alloc_page(GFP_KERNEL);
 	if (scratch == NULL)
@@ -559,40 +548,20 @@ int nfs_readdir_page_filler(nfs_readdir_
 		if (desc->plus)
 			nfs_prime_dcache(file_dentry(desc->file), entry);
 
-		status = nfs_readdir_add_to_array(entry, desc->pvec.pages[desc->pvec.nr]);
-		if (status == -ENOSPC) {
-			desc->pvec.nr++;
-			if (desc->pvec.nr == max_rapages)
-				break;
-			status = nfs_readdir_add_to_array(entry, desc->pvec.pages[desc->pvec.nr]);
-		}
+		status = nfs_readdir_add_to_array(entry, page);
 		if (status != 0)
 			break;
 	} while (!entry->eof);
 
-	/*
-	 * page and desc->pvec.pages[0] are valid, don't need to check
-	 * whether or not to be NULL.
-	 */
-	copy_highpage(page, desc->pvec.pages[0]);
-
 out_nopages:
 	if (count == 0 || (status == -EBADCOOKIE && entry->eof != 0)) {
-		array = kmap_atomic(desc->pvec.pages[desc->pvec.nr]);
+		array = kmap(page);
 		array->eof_index = array->size;
 		status = 0;
-		kunmap_atomic(array);
+		kunmap(page);
 	}
 
 	put_page(scratch);
-
-	/*
-	 * desc->pvec.nr > 0 means at least one page was completely filled,
-	 * we should return -ENOSPC. Otherwise function
-	 * nfs_readdir_xdr_to_array will enter infinite loop.
-	 */
-	if (desc->pvec.nr > 0)
-		return -ENOSPC;
 	return status;
 }
 
@@ -626,24 +595,6 @@ out_freepages:
 	return -ENOMEM;
 }
 
-/*
- * nfs_readdir_rapages_init initialize rapages by nfs_cache_array structure.
- */
-static
-void nfs_readdir_rapages_init(nfs_readdir_descriptor_t *desc)
-{
-	struct nfs_cache_array *array;
-	int max_rapages = NFS_MAX_READDIR_RAPAGES;
-	int index;
-
-	for (index = 0; index < max_rapages; index++) {
-		array = kmap_atomic(desc->pvec.pages[index]);
-		memset(array, 0, sizeof(struct nfs_cache_array));
-		array->eof_index = -1;
-		kunmap_atomic(array);
-	}
-}
-
 static
 int nfs_readdir_xdr_to_array(nfs_readdir_descriptor_t *desc, struct page *page, struct inode *inode)
 {
@@ -654,12 +605,6 @@ int nfs_readdir_xdr_to_array(nfs_readdir
 	int status = -ENOMEM;
 	unsigned int array_size = ARRAY_SIZE(pages);
 
-	/*
-	 * This means we hit readdir rdpages miss, the preallocated rdpages
-	 * are useless, the preallocate rdpages should be reinitialized.
-	 */
-	nfs_readdir_rapages_init(desc);
-
 	entry.prev_cookie = 0;
 	entry.cookie = desc->last_cookie;
 	entry.eof = 0;
@@ -719,24 +664,9 @@ int nfs_readdir_filler(nfs_readdir_descr
 	struct inode	*inode = file_inode(desc->file);
 	int ret;
 
-	/*
-	 * If desc->page_index in range desc->pvec.index and
-	 * desc->pvec.index + desc->pvec.nr, we get readdir cache hit.
-	 */
-	if (desc->page_index >= desc->pvec.index &&
-		desc->page_index < (desc->pvec.index + desc->pvec.nr)) {
-		/*
-		 * page and desc->pvec.pages[x] are valid, don't need to check
-		 * whether or not to be NULL.
-		 */
-		copy_highpage(page, desc->pvec.pages[desc->page_index - desc->pvec.index]);
-		ret = 0;
-	} else {
-		ret = nfs_readdir_xdr_to_array(desc, page, inode);
-		if (ret < 0)
-			goto error;
-	}
-
+	ret = nfs_readdir_xdr_to_array(desc, page, inode);
+	if (ret < 0)
+		goto error;
 	SetPageUptodate(page);
 
 	if (invalidate_inode_pages2_range(inode->i_mapping, page->index + 1, -1) < 0) {
@@ -901,7 +831,6 @@ static int nfs_readdir(struct file *file
 			*desc = &my_desc;
 	struct nfs_open_dir_context *dir_ctx = file->private_data;
 	int res = 0;
-	int max_rapages = NFS_MAX_READDIR_RAPAGES;
 
 	dfprintk(FILE, "NFS: readdir(%pD2) starting at cookie %llu\n",
 			file, (long long)ctx->pos);
@@ -921,12 +850,6 @@ static int nfs_readdir(struct file *file
 	desc->decode = NFS_PROTO(inode)->decode_dirent;
 	desc->plus = nfs_use_readdirplus(inode, ctx);
 
-	res = nfs_readdir_alloc_pages(desc->pvec.pages, max_rapages);
-	if (res < 0)
-		return -ENOMEM;
-
-	nfs_readdir_rapages_init(desc);
-
 	if (ctx->pos == 0 || nfs_attribute_cache_expired(inode))
 		res = nfs_revalidate_mapping(inode, file->f_mapping);
 	if (res < 0)
@@ -962,7 +885,6 @@ static int nfs_readdir(struct file *file
 			break;
 	} while (!desc->eof);
 out:
-	nfs_readdir_free_pages(desc->pvec.pages, max_rapages);
 	if (res > 0)
 		res = 0;
 	dfprintk(FILE, "NFS: readdir(%pD2) returns %d\n", file, res);
fs/nfs/internal.h +1 -2
@@ -69,8 +69,7 @@ struct nfs_clone_mount {
  * Maximum number of pages that readdir can use for creating
  * a vmapped array of pages.
  */
-#define NFS_MAX_READDIR_PAGES 64
-#define NFS_MAX_READDIR_RAPAGES 8
+#define NFS_MAX_READDIR_PAGES 8
 
 struct nfs_client_initdata {
 	unsigned long init_flags;
author maintainer
#281 lore raw

PATCH 280/371 pnfs/flexfiles: Fix PTR_ERR() dereferences in ff_layout_track_ds_error

In reply to Greg Kroah-Hartman

From: Trond Myklebust <[email protected]>

commit 8e04fdfadda75a849c649f7e50fe7d97772e1fcb upstream.

mirror->mirror_ds can be NULL if uninitialised, but can contain a PTR_ERR() if call to GETDEVICEINFO failed.

Fixes 65990d1afbd2 ("pNFS/flexfiles: Fix a deadlock on LAYOUTGET")
---
 fs/nfs/flexfilelayout/flexfilelayoutdev.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/nfs/flexfilelayout/flexfilelayoutdev.c +1 -1
@@ -257,7 +257,7 @@ int ff_layout_track_ds_error(struct nfs4
 	if (status == 0)
 		return 0;
 
-	if (mirror->mirror_ds == NULL)
+	if (IS_ERR_OR_NULL(mirror->mirror_ds))
 		return -EINVAL;
 
 	dserr = kmalloc(sizeof(*dserr), gfp_flags);
author maintainer
#282 lore raw

PATCH 281/371 pnfs: Fix a problem where we gratuitously start doing I/O through the MDS

In reply to Greg Kroah-Hartman

From: Trond Myklebust <[email protected]>

commit 58bbeab425c6c5e318f5b6ae31d351331ddfb34b upstream.

If the client has to stop in pnfs_update_layout() to wait for another layoutget to complete, it currently exits and defaults to I/O through the MDS if the layoutget was successful.

Fixes d03360aaf5cc ("pNFS: Ensure we return the error if someone kills...")
Cc [email protected] # v4.20+
---
 fs/nfs/pnfs.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/nfs/pnfs.c +1 -1
@@ -1890,7 +1890,7 @@ lookup_again:
 		spin_unlock(&ino->i_lock);
 		lseg = ERR_PTR(wait_var_event_killable(&lo->plh_outstanding,
 					!atomic_read(&lo->plh_outstanding)));
-		if (IS_ERR(lseg) || !list_empty(&lo->plh_segs))
+		if (IS_ERR(lseg))
 			goto out_put_layout_hdr;
 		pnfs_put_layout_hdr(lo);
 		goto lookup_again;
author maintainer
#283 lore raw

PATCH 282/371 SUNRPC: Ensure the bvecs are reset when we re-encode the RPC request

In reply to Greg Kroah-Hartman

From: Trond Myklebust <[email protected]>

commit 75369089820473eac45e9ddd970081901a373c08 upstream.

The bvec tracks the list of pages, so if the number of pages changes due to a re-encode, we need to reset the bvec as well.

Fixes 277e4ab7d530 ("SUNRPC: Simplify TCP receive code by switching...")
Cc [email protected] # v4.20+
---
 net/sunrpc/clnt.c     |    3 +--
 net/sunrpc/xprt.c     |    2 ++
 net/sunrpc/xprtsock.c |    1 +
 3 files changed, 4 insertions(+), 2 deletions(-)
net/sunrpc/clnt.c +1 -2
@@ -1767,6 +1767,7 @@ rpc_xdr_encode(struct rpc_task *task)
 	req->rq_snd_buf.head[0].iov_len = 0;
 	xdr_init_encode(&xdr, &req->rq_snd_buf,
 			req->rq_snd_buf.head[0].iov_base, req);
+	xdr_free_bvec(&req->rq_snd_buf);
 	if (rpc_encode_header(task, &xdr))
 		return;
 
@@ -1799,8 +1800,6 @@ call_encode(struct rpc_task *task)
 			rpc_exit(task, task->tk_status);
 		}
 		return;
-	} else {
-		xprt_request_prepare(task->tk_rqstp);
 	}
 
 	/* Add task to reply queue before transmission to avoid races */
net/sunrpc/xprt.c +2 -0
@@ -1006,6 +1006,8 @@ xprt_request_enqueue_receive(struct rpc_
 
 	if (!xprt_request_need_enqueue_receive(task, req))
 		return;
+
+	xprt_request_prepare(task->tk_rqstp);
 	spin_lock(&xprt->queue_lock);
 
 	/* Update the softirq receive buffer */
net/sunrpc/xprtsock.c +1 -0
@@ -909,6 +909,7 @@ static int xs_nospace(struct rpc_rqst *r
 static void
 xs_stream_prepare_request(struct rpc_rqst *req)
 {
+	xdr_free_bvec(&req->rq_rcv_buf);
 	req->rq_task->tk_status = xdr_alloc_bvec(&req->rq_rcv_buf, GFP_KERNEL);
 }
 
author maintainer
#284 lore raw

PATCH 283/371 lib/scatterlist: Fix mapping iterator when sg->offset is greater than PAGE_SIZE

In reply to Greg Kroah-Hartman

From: Christophe Leroy <[email protected]>

commit aeb87246537a83c2aff482f3f34a2e0991e02cbc upstream.

All mapping iterator logic is based on the assumption that sg->offset is always lower than PAGE_SIZE.

But there are situations where sg->offset is such that the SG item is on the second page. In that case sg_copy_to_buffer() fails properly copying the data into the buffer. One of the reason is that the data will be outside the kmapped area used to access that data.

This patch fixes the issue by adjusting the mapping iterator offset and pgoffset fields such that offset is always lower than PAGE_SIZE.

Fixes 4225fc8555a9 ("lib/scatterlist: use page iterator in the mapping iterator")
---
 lib/scatterlist.c |    9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)
lib/scatterlist.c +5 -4
@@ -678,17 +678,18 @@ static bool sg_miter_get_next_page(struc
 {
 	if (!miter->__remaining) {
 		struct scatterlist *sg;
-		unsigned long pgoffset;
 
 		if (!__sg_page_iter_next(&miter->piter))
 			return false;
 
 		sg = miter->piter.sg;
-		pgoffset = miter->piter.sg_pgoffset;
 
-		miter->__offset = pgoffset ? 0 : sg->offset;
+		miter->__offset = miter->piter.sg_pgoffset ? 0 : sg->offset;
+		miter->piter.sg_pgoffset += miter->__offset >> PAGE_SHIFT;
+		miter->__offset &= PAGE_SIZE - 1;
 		miter->__remaining = sg->offset + sg->length -
-				(pgoffset << PAGE_SHIFT) - miter->__offset;
+				     (miter->piter.sg_pgoffset << PAGE_SHIFT) -
+				     miter->__offset;
 		miter->__remaining = min_t(unsigned long, miter->__remaining,
 					   PAGE_SIZE - miter->__offset);
 	}
author maintainer
#285 lore raw

PATCH 284/371 ASoC: dapm: Adapt for debugfs API change

In reply to Greg Kroah-Hartman

From: Mark Brown <[email protected]>

commit ceaea851b9ea75f9ea2bbefb53ff0d4b27cd5a6e upstream.

Back in ff9fb72bc07705c (debugfs: return error values, not NULL) the debugfs APIs were changed to return error pointers rather than NULL pointers on error, breaking the error checking in ASoC. Update the code to use IS_ERR() and log the codes that are returned as part of the error messages.

Fixes ff9fb72bc07705c (debugfs: return error values, not NULL)
---
 sound/soc/soc-dapm.c |   18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)
sound/soc/soc-dapm.c +10 -8
@@ -2154,23 +2154,25 @@ void snd_soc_dapm_debugfs_init(struct sn
 {
 	struct dentry *d;
 
-	if (!parent)
+	if (!parent || IS_ERR(parent))
 		return;
 
 	dapm->debugfs_dapm = debugfs_create_dir("dapm", parent);
 
-	if (!dapm->debugfs_dapm) {
+	if (IS_ERR(dapm->debugfs_dapm)) {
 		dev_warn(dapm->dev,
-		       "ASoC: Failed to create DAPM debugfs directory\n");
+			 "ASoC: Failed to create DAPM debugfs directory %ld\n",
+			 PTR_ERR(dapm->debugfs_dapm));
 		return;
 	}
 
 	d = debugfs_create_file("bias_level", 0444,
 				dapm->debugfs_dapm, dapm,
 				&dapm_bias_fops);
-	if (!d)
+	if (IS_ERR(d))
 		dev_warn(dapm->dev,
-			 "ASoC: Failed to create bias level debugfs file\n");
+			 "ASoC: Failed to create bias level debugfs file: %ld\n",
+			 PTR_ERR(d));
 }
 
 static void dapm_debugfs_add_widget(struct snd_soc_dapm_widget *w)
@@ -2184,10 +2186,10 @@ static void dapm_debugfs_add_widget(stru
 	d = debugfs_create_file(w->name, 0444,
 				dapm->debugfs_dapm, w,
 				&dapm_widget_power_fops);
-	if (!d)
+	if (IS_ERR(d))
 		dev_warn(w->dapm->dev,
-			"ASoC: Failed to create %s debugfs file\n",
-			w->name);
+			 "ASoC: Failed to create %s debugfs file: %ld\n",
+			 w->name, PTR_ERR(d));
 }
 
 static void dapm_debugfs_cleanup(struct snd_soc_dapm_context *dapm)
author maintainer
#286 lore raw

PATCH 285/371 ASoC: core: Adapt for debugfs API change

In reply to Greg Kroah-Hartman

From: Mark Brown <[email protected]>

commit c2c928c93173f220955030e8440517b87ec7df92 upstream.

Back in ff9fb72bc07705c (debugfs: return error values, not NULL) the debugfs APIs were changed to return error pointers rather than NULL pointers on error, breaking the error checking in ASoC. Update the code to use IS_ERR() and log the codes that are returned as part of the error messages.

Fixes ff9fb72bc07705c (debugfs: return error values, not NULL)
---
 sound/soc/soc-core.c |   16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)
sound/soc/soc-core.c +10 -6
@@ -158,9 +158,10 @@ static void soc_init_component_debugfs(s
 				component->card->debugfs_card_root);
 	}
 
-	if (!component->debugfs_root) {
+	if (IS_ERR(component->debugfs_root)) {
 		dev_warn(component->dev,
-			"ASoC: Failed to create component debugfs directory\n");
+			"ASoC: Failed to create component debugfs directory: %ld\n",
+			PTR_ERR(component->debugfs_root));
 		return;
 	}
 
@@ -212,18 +213,21 @@ static void soc_init_card_debugfs(struct
 
 	card->debugfs_card_root = debugfs_create_dir(card->name,
 						     snd_soc_debugfs_root);
-	if (!card->debugfs_card_root) {
+	if (IS_ERR(card->debugfs_card_root)) {
 		dev_warn(card->dev,
-			 "ASoC: Failed to create card debugfs directory\n");
+			 "ASoC: Failed to create card debugfs directory: %ld\n",
+			 PTR_ERR(card->debugfs_card_root));
+		card->debugfs_card_root = NULL;
 		return;
 	}
 
 	card->debugfs_pop_time = debugfs_create_u32("dapm_pop_time", 0644,
 						    card->debugfs_card_root,
 						    &card->pop_time);
-	if (!card->debugfs_pop_time)
+	if (IS_ERR(card->debugfs_pop_time))
 		dev_warn(card->dev,
-			 "ASoC: Failed to create pop time debugfs file\n");
+			 "ASoC: Failed to create pop time debugfs file: %ld\n",
+			 PTR_ERR(card->debugfs_pop_time));
 }
 
 static void soc_cleanup_card_debugfs(struct snd_soc_card *card)
author maintainer
#287 lore raw

PATCH 286/371 raid5-cache: Need to do start() part job after adding journal device

In reply to Greg Kroah-Hartman

From: Xiao Ni <[email protected]>

commit d9771f5ec46c282d518b453c793635dbdc3a2a94 upstream.

commit d5d885fd514f ("md: introduce new personality funciton start()") splits the init job to two parts. The first part run() does the jobs that do not require the md threads. The second part start() does the jobs that require the md threads.

Now it just does run() in adding new journal device. It needs to do the second part start() too.

Fixes d5d885fd514f ("md: introduce new personality funciton start()")
---
 drivers/md/raid5.c |   11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)
drivers/md/raid5.c +9 -2
@@ -7674,7 +7674,7 @@ abort:
 static int raid5_add_disk(struct mddev *mddev, struct md_rdev *rdev)
 {
 	struct r5conf *conf = mddev->private;
-	int err = -EEXIST;
+	int ret, err = -EEXIST;
 	int disk;
 	struct disk_info *p;
 	int first = 0;
@@ -7689,7 +7689,14 @@ static int raid5_add_disk(struct mddev *
 		 * The array is in readonly mode if journal is missing, so no
 		 * write requests running. We should be safe
 		 */
-		log_init(conf, rdev, false);
+		ret = log_init(conf, rdev, false);
+		if (ret)
+			return ret;
+
+		ret = r5l_start(conf->log);
+		if (ret)
+			return ret;
+
 		return 0;
 	}
 	if (mddev->recovery_disabled == conf->recovery_disabled)
author maintainer
#288 lore raw

PATCH 287/371 ALSA: seq: Break too long mutex context in the write loop

In reply to Greg Kroah-Hartman

From: Takashi Iwai <[email protected]>

commit ede34f397ddb063b145b9e7d79c6026f819ded13 upstream.

The fix for the racy writes and ioctls to sequencer widened the application of client->ioctl_mutex to the whole write loop. Although it does unlock/relock for the lengthy operation like the event dup, the loop keeps the ioctl_mutex for the whole time in other situations. This may take quite long time if the user-space would give a huge buffer, and this is a likely cause of some weird behavior spotted by syzcaller fuzzer.

This patch puts a simple workaround, just adding a mutex break in the loop when a large number of events have been processed. This shouldn't hit any performance drop because the threshold is set high enough for usual operations.

Fixes 7bd800915677 ("ALSA: seq: More protection for concurrent write and ioctl races")
Reported-by [email protected]
Reported-by [email protected]
---
 sound/core/seq/seq_clientmgr.c |   11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)
sound/core/seq/seq_clientmgr.c +10 -1
@@ -1004,7 +1004,7 @@ static ssize_t snd_seq_write(struct file
 {
 	struct snd_seq_client *client = file->private_data;
 	int written = 0, len;
-	int err;
+	int err, handled;
 	struct snd_seq_event event;
 
 	if (!(snd_seq_file_flags(file) & SNDRV_SEQ_LFLG_OUTPUT))
@@ -1017,6 +1017,8 @@ static ssize_t snd_seq_write(struct file
 	if (!client->accept_output || client->pool == NULL)
 		return -ENXIO;
 
+ repeat:
+	handled = 0;
 	/* allocate the pool now if the pool is not allocated yet */ 
 	mutex_lock(&client->ioctl_mutex);
 	if (client->pool->size > 0 && !snd_seq_write_pool_allocated(client)) {
@@ -1076,12 +1078,19 @@ static ssize_t snd_seq_write(struct file
 						   0, 0, &client->ioctl_mutex);
 		if (err < 0)
 			break;
+		handled++;
 
 	__skip_event:
 		/* Update pointers and counts */
 		count -= len;
 		buf += len;
 		written += len;
+
+		/* let's have a coffee break if too many events are queued */
+		if (++handled >= 200) {
+			mutex_unlock(&client->ioctl_mutex);
+			goto repeat;
+		}
 	}
 
  out:
author maintainer
#289 lore raw

PATCH 288/371 ALSA: hda - Dont resume forcibly i915 HDMI/DP codec

In reply to Greg Kroah-Hartman

From: Takashi Iwai <[email protected]>

commit 4914da2fb0c89205790503f20dfdde854f3afdd8 upstream.

We apply the codec resume forcibly at system resume callback for updating and syncing the jack detection state that may have changed during sleeping. This is, however, superfluous for the codec like Intel HDMI/DP, where the jack detection is managed via the audio component notification; i.e. the jack state change shall be reported sooner or later from the graphics side at mode change.

This patch changes the codec resume callback to avoid the forcible resume conditionally with a new flag, codec->relaxed_resume, for reducing the resume time. The flag is set in the codec probe.

Although this doesn't fix the entire bug mentioned in the bugzilla entry below, it's still a good optimization and some improvements are seen.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=201901

---
 include/sound/hda_codec.h  |    2 ++
 sound/pci/hda/hda_codec.c  |    8 ++++++--
 sound/pci/hda/patch_hdmi.c |    6 +++++-
 3 files changed, 13 insertions(+), 3 deletions(-)
include/sound/hda_codec.h +2 -0
@@ -262,6 +262,8 @@ struct hda_codec {
 	unsigned int auto_runtime_pm:1; /* enable automatic codec runtime pm */
 	unsigned int force_pin_prefix:1; /* Add location prefix */
 	unsigned int link_down_at_suspend:1; /* link down at runtime suspend */
+	unsigned int relaxed_resume:1;	/* don't resume forcibly for jack */
+
 #ifdef CONFIG_PM
 	unsigned long power_on_acct;
 	unsigned long power_off_acct;
sound/pci/hda/hda_codec.c +6 -2
@@ -2955,15 +2955,19 @@ static int hda_codec_runtime_resume(stru
 #ifdef CONFIG_PM_SLEEP
 static int hda_codec_force_resume(struct device *dev)
 {
+	struct hda_codec *codec = dev_to_hda_codec(dev);
+	bool forced_resume = !codec->relaxed_resume;
 	int ret;
 
 	/* The get/put pair below enforces the runtime resume even if the
 	 * device hasn't been used at suspend time.  This trick is needed to
 	 * update the jack state change during the sleep.
 	 */
-	pm_runtime_get_noresume(dev);
+	if (forced_resume)
+		pm_runtime_get_noresume(dev);
 	ret = pm_runtime_force_resume(dev);
-	pm_runtime_put(dev);
+	if (forced_resume)
+		pm_runtime_put(dev);
 	return ret;
 }
 
sound/pci/hda/patch_hdmi.c +5 -1
@@ -2304,8 +2304,10 @@ static void generic_hdmi_free(struct hda
 	struct hdmi_spec *spec = codec->spec;
 	int pin_idx, pcm_idx;
 
-	if (codec_has_acomp(codec))
+	if (codec_has_acomp(codec)) {
 		snd_hdac_acomp_register_notifier(&codec->bus->core, NULL);
+		codec->relaxed_resume = 0;
+	}
 
 	for (pin_idx = 0; pin_idx < spec->num_pins; pin_idx++) {
 		struct hdmi_spec_per_pin *per_pin = get_pin(spec, pin_idx);
@@ -2578,6 +2580,8 @@ static void register_i915_notifier(struc
 	spec->drm_audio_ops.pin_eld_notify = intel_pin_eld_notify;
 	snd_hdac_acomp_register_notifier(&codec->bus->core,
 					&spec->drm_audio_ops);
+	/* no need for forcible resume for jack check thanks to notifier */
+	codec->relaxed_resume = 1;
 }
 
 /* setup_stream ops override for HSW+ */
author maintainer
#290 lore raw

PATCH 289/371 ALSA: hda/realtek - Fixed Headphone Mic cant record on Dell platform

In reply to Greg Kroah-Hartman

From: Kailang Yang <[email protected]>

commit fbc571290d9f7bfe089c50f4ac4028dd98ebfe98 upstream.

It assigned to wrong model. So, The headphone Mic can't work.

Fixes 3f640970a414 ("ALSA: hda - Fix headset mic detection problem for several Dell laptops")
---
 sound/pci/hda/patch_realtek.c |    5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
sound/pci/hda/patch_realtek.c +4 -1
@@ -7614,9 +7614,12 @@ static const struct snd_hda_pin_quirk al
 		{0x12, 0x90a60130},
 		{0x17, 0x90170110},
 		{0x21, 0x03211020}),
-	SND_HDA_PIN_QUIRK(0x10ec0295, 0x1028, "Dell", ALC269_FIXUP_DELL1_MIC_NO_PRESENCE,
+	SND_HDA_PIN_QUIRK(0x10ec0295, 0x1028, "Dell", ALC269_FIXUP_DELL4_MIC_NO_PRESENCE,
 		{0x14, 0x90170110},
 		{0x21, 0x04211020}),
+	SND_HDA_PIN_QUIRK(0x10ec0295, 0x1028, "Dell", ALC269_FIXUP_DELL4_MIC_NO_PRESENCE,
+		{0x14, 0x90170110},
+		{0x21, 0x04211030}),
 	SND_HDA_PIN_QUIRK(0x10ec0295, 0x1028, "Dell", ALC269_FIXUP_DELL1_MIC_NO_PRESENCE,
 		ALC295_STANDARD_PINS,
 		{0x17, 0x21014020},
author maintainer
#291 lore raw

PATCH 290/371 ALSA: hda/realtek: apply ALC891 headset fixup to one Dell machine

In reply to Greg Kroah-Hartman

From: Hui Wang <[email protected]>

commit 4b4e0e32e4b09274dbc9d173016c1a026f44608c upstream.

Without this patch, the headset-mic and headphone-mic don't work.

---
 sound/pci/hda/patch_realtek.c |    5 +++++
 1 file changed, 5 insertions(+)
sound/pci/hda/patch_realtek.c +5 -0
@@ -8754,6 +8754,11 @@ static const struct snd_hda_pin_quirk al
 		{0x18, 0x01a19030},
 		{0x1a, 0x01813040},
 		{0x21, 0x01014020}),
+	SND_HDA_PIN_QUIRK(0x10ec0867, 0x1028, "Dell", ALC891_FIXUP_DELL_MIC_NO_PRESENCE,
+		{0x16, 0x01813030},
+		{0x17, 0x02211010},
+		{0x18, 0x01a19040},
+		{0x21, 0x01014020}),
 	SND_HDA_PIN_QUIRK(0x10ec0662, 0x1028, "Dell", ALC662_FIXUP_DELL_MIC_NO_PRESENCE,
 		{0x14, 0x01014010},
 		{0x18, 0x01a19020},
author maintainer
#292 lore raw

PATCH 291/371 ALSA: hda/hdmi - Remove duplicated define

In reply to Greg Kroah-Hartman

From: Takashi Iwai <[email protected]>

commit eb4177116bf568a413c544eca3f4446cb4064be9 upstream.

INTEL_GET_VENDOR_VERB is defined twice identically. Let's remove a superfluous line.

Fixes b0d8bc50b9f2 ("ALSA: hda: hdmi - add Icelake support")
---
 sound/pci/hda/patch_hdmi.c |    1 -
 1 file changed, 1 deletion(-)
sound/pci/hda/patch_hdmi.c +0 -1
@@ -2431,7 +2431,6 @@ static void intel_haswell_fixup_connect_
 }
 
 #define INTEL_GET_VENDOR_VERB	0xf81
-#define INTEL_GET_VENDOR_VERB	0xf81
 #define INTEL_SET_VENDOR_VERB	0x781
 #define INTEL_EN_DP12		0x02	/* enable DP 1.2 features */
 #define INTEL_EN_ALL_PIN_CVTS	0x01	/* enable 2nd & 3rd pins and convertors */
author maintainer
#293 lore raw

PATCH 292/371 ALSA: hda/hdmi - Fix i915 reverse port/pin mapping

In reply to Greg Kroah-Hartman

From: Takashi Iwai <[email protected]>

commit 3140aafb22edeab0cc41f15f53b12a118c0ac215 upstream.

The recent fix for Icelake HDMI codec introduced the mapping from pin NID to the i915 gfx port number. However, it forgot the reverse mapping from the port number to the pin NID that is used in the ELD notifier callback. As a result, it's processed to a wrong widget and gives a warning like
snd_hda_codec_hdmi hdaudioC0D2: HDMI: pin nid 5 not registered

This patch corrects it with a proper reverse mapping function.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=204133

Fixes b0d8bc50b9f2 ("ALSA: hda: hdmi - add Icelake support")
---
 sound/pci/hda/patch_hdmi.c |   24 +++++++++++++++++++-----
 1 file changed, 19 insertions(+), 5 deletions(-)
sound/pci/hda/patch_hdmi.c +19 -5
@@ -2538,18 +2538,32 @@ static int intel_pin2port(void *audio_pt
 	return -1;
 }
 
+static int intel_port2pin(struct hda_codec *codec, int port)
+{
+	struct hdmi_spec *spec = codec->spec;
+
+	if (!spec->port_num) {
+		/* we assume only from port-B to port-D */
+		if (port < 1 || port > 3)
+			return 0;
+		/* intel port is 1-based */
+		return port + intel_base_nid(codec) - 1;
+	}
+
+	if (port < 1 || port > spec->port_num)
+		return 0;
+	return spec->port_map[port - 1];
+}
+
 static void intel_pin_eld_notify(void *audio_ptr, int port, int pipe)
 {
 	struct hda_codec *codec = audio_ptr;
 	int pin_nid;
 	int dev_id = pipe;
 
-	/* we assume only from port-B to port-D */
-	if (port < 1 || port > 3)
+	pin_nid = intel_port2pin(codec, port);
+	if (!pin_nid)
 		return;
-
-	pin_nid = port + intel_base_nid(codec) - 1; /* intel port is 1-based */
-
 	/* skip notification during system suspend (but not in runtime PM);
 	 * the state will be updated at resume
 	 */
author maintainer
#294 lore raw

PATCH 293/371 ceph: fix end offset in truncate_inode_pages_range call

In reply to Greg Kroah-Hartman

From: Luis Henriques <[email protected]>

commit d31d07b97a5e76f41e00eb81dcca740e84aa7782 upstream.

Commit e450f4d1a5d6 ("ceph: pass inclusive lend parameter to filemap_write_and_wait_range()") fixed the end offset parameter used to call filemap_write_and_wait_range and invalidate_inode_pages2_range. Unfortunately it missed truncate_inode_pages_range, introducing a regression that is easily detected by xfstest generic/130.

The problem is that when doing direct IO it is possible that an extra page is truncated from the page cache when the end offset is page aligned. This can cause data loss if that page hasn't been sync'ed to the OSDs.

While there, change code to use PAGE_ALIGN macro instead.

Fixes e450f4d1a5d6 ("ceph: pass inclusive lend parameter to filemap_write_and_wait_range()")
---
 fs/ceph/file.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/ceph/file.c +1 -1
@@ -1006,7 +1006,7 @@ ceph_direct_read_write(struct kiocb *ioc
 			 * may block.
 			 */
 			truncate_inode_pages_range(inode->i_mapping, pos,
-					(pos+len) | (PAGE_SIZE - 1));
+						   PAGE_ALIGN(pos + len) - 1);
 
 			req->r_mtime = mtime;
 		}
author maintainer
#295 lore raw

PATCH 294/371 media: v4l2: Test type instead of cfg->type in v4l2_ctrl_new_custom()

In reply to Greg Kroah-Hartman

From: Boris Brezillon <[email protected]>

commit 07d89227a983df957a6a7c56f7c040cde9ac571f upstream.

cfg->type can be overridden by v4l2_ctrl_fill() and the new value is stored in the local type var. Fix the tests to use this local var.

Fixes 0996517cf8ea ("V4L/DVB: v4l2: Add new control handling framework")
[[email protected]: change to !qmenu and !qmenu_int (checkpatch)]
---
 drivers/media/v4l2-core/v4l2-ctrls.c |    9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)
drivers/media/v4l2-core/v4l2-ctrls.c +4 -5
@@ -2365,16 +2365,15 @@ struct v4l2_ctrl *v4l2_ctrl_new_custom(s
 		v4l2_ctrl_fill(cfg->id, &name, &type, &min, &max, &step,
 								&def, &flags);
 
-	is_menu = (cfg->type == V4L2_CTRL_TYPE_MENU ||
-		   cfg->type == V4L2_CTRL_TYPE_INTEGER_MENU);
+	is_menu = (type == V4L2_CTRL_TYPE_MENU ||
+		   type == V4L2_CTRL_TYPE_INTEGER_MENU);
 	if (is_menu)
 		WARN_ON(step);
 	else
 		WARN_ON(cfg->menu_skip_mask);
-	if (cfg->type == V4L2_CTRL_TYPE_MENU && qmenu == NULL)
+	if (type == V4L2_CTRL_TYPE_MENU && !qmenu) {
 		qmenu = v4l2_ctrl_get_menu(cfg->id);
-	else if (cfg->type == V4L2_CTRL_TYPE_INTEGER_MENU &&
-		 qmenu_int == NULL) {
+	} else if (type == V4L2_CTRL_TYPE_INTEGER_MENU && !qmenu_int) {
 		handler_set_err(hdl, -EINVAL);
 		return NULL;
 	}
author maintainer
#296 lore raw

PATCH 295/371 media: coda: Remove unbalanced and unneeded mutex unlock

In reply to Greg Kroah-Hartman

From: Ezequiel Garcia <[email protected]>

commit 766b9b168f6c75c350dd87c3e0bc6a9b322f0013 upstream.

The mutex unlock in the threaded interrupt handler is not paired with any mutex lock. Remove it.

This bug has been here for a really long time, so it applies to any stable repo.

---
 drivers/media/platform/coda/coda-bit.c |    1 -
 1 file changed, 1 deletion(-)
drivers/media/platform/coda/coda-bit.c +0 -1
@@ -2314,7 +2314,6 @@ irqreturn_t coda_irq_handler(int irq, vo
 	if (ctx == NULL) {
 		v4l2_err(&dev->v4l2_dev,
 			 "Instance released before the end of transaction\n");
-		mutex_unlock(&dev->coda_mutex);
 		return IRQ_HANDLED;
 	}
 
author maintainer
#297 lore raw

PATCH 296/371 media: videobuf2-core: Prevent size alignment wrapping buffer size to 0

In reply to Greg Kroah-Hartman

From: Sakari Ailus <[email protected]>

commit defcdc5d89ced780fb45196d539d6570ec5b1ba5 upstream.

PAGE_ALIGN() may wrap the buffer size around to 0. Prevent this by checking that the aligned value is not smaller than the unaligned one.

Note on backporting to stable: the file used to be under drivers/media/v4l2-core, it was moved to the current location after 4.14.

---
 drivers/media/common/videobuf2/videobuf2-core.c |    4 ++++
 1 file changed, 4 insertions(+)
drivers/media/common/videobuf2/videobuf2-core.c +4 -0
@@ -207,6 +207,10 @@ static int __vb2_buf_mem_alloc(struct vb
 	for (plane = 0; plane < vb->num_planes; ++plane) {
 		unsigned long size = PAGE_ALIGN(vb->planes[plane].length);
 
+		/* Did it wrap around? */
+		if (size < vb->planes[plane].length)
+			goto free;
+
 		mem_priv = call_ptr_memop(vb, alloc,
 				q->alloc_devs[plane] ? : q->dev,
 				q->dma_attrs, size, q->dma_dir, q->gfp_flags);
author maintainer
#298 lore raw

PATCH 297/371 media: videobuf2-dma-sg: Prevent size from overflowing

In reply to Greg Kroah-Hartman

From: Sakari Ailus <[email protected]>

commit 14f28f5cea9e3998442de87846d1907a531b6748 upstream.

buf->size is an unsigned long; casting that to int will lead to an overflow if buf->size exceeds INT_MAX.

Fix this by changing the type to unsigned long instead. This is possible as the buf->size is always aligned to PAGE_SIZE, and therefore the size will never have values lesser than 0.

Note on backporting to stable: the file used to be under drivers/media/v4l2-core, it was moved to the current location after 4.14.

---
 drivers/media/common/videobuf2/videobuf2-dma-sg.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/media/common/videobuf2/videobuf2-dma-sg.c +1 -1
@@ -59,7 +59,7 @@ static int vb2_dma_sg_alloc_compacted(st
 		gfp_t gfp_flags)
 {
 	unsigned int last_page = 0;
-	int size = buf->size;
+	unsigned long size = buf->size;
 
 	while (size > 0) {
 		struct page *pages;
author maintainer
#299 lore raw

PATCH 298/371 KVM: nVMX: Dont dump VMCS if virtual APIC page cant be mapped

In reply to Greg Kroah-Hartman

From: Sean Christopherson <[email protected]>

commit 73cb85568433feadb79e963bf2efba9b3e9ae3df upstream.

... as a malicious userspace can run a toy guest to generate invalid virtual-APIC page addresses in L1, i.e. flood the kernel log with error messages.

Fixes 690908104e39d ("KVM: nVMX: allow tests to use bad virtual-APIC page address")
Cc Paolo Bonzini <[email protected]>
---
 arch/x86/kvm/vmx/nested.c |    3 ---
 1 file changed, 3 deletions(-)
arch/x86/kvm/vmx/nested.c +0 -3
@@ -2891,9 +2891,6 @@ static void nested_get_vmcs12_pages(stru
 			 */
 			vmcs_clear_bits(CPU_BASED_VM_EXEC_CONTROL,
 					CPU_BASED_TPR_SHADOW);
-		} else {
-			printk("bad virtual-APIC page address\n");
-			dump_vmcs();
 		}
 	}
 
author maintainer
#300 lore raw

PATCH 299/371 KVM: nVMX: Always sync GUEST_BNDCFGS when it comes from vmcs01

In reply to Greg Kroah-Hartman

From: Sean Christopherson <[email protected]>

commit 3b013a2972d5bc344d6eaa8f24fdfe268211e45f upstream.

If L1 does not set VM_ENTRY_LOAD_BNDCFGS, then L1's BNDCFGS value must be propagated to vmcs02 since KVM always runs with VM_ENTRY_LOAD_BNDCFGS when MPX is supported. Because the value effectively comes from vmcs01, vmcs02 must be updated even if vmcs12 is clean.

Fixes 62cf9bd8118c4 ("KVM: nVMX: Fix emulation of VM_ENTRY_LOAD_BNDCFGS")
Cc Liran Alon <[email protected]>
---
 arch/x86/kvm/vmx/nested.c |   13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)
arch/x86/kvm/vmx/nested.c +6 -7
@@ -2243,13 +2243,9 @@ static void prepare_vmcs02_full(struct v
 
 	set_cr4_guest_host_mask(vmx);
 
-	if (kvm_mpx_supported()) {
-		if (vmx->nested.nested_run_pending &&
-			(vmcs12->vm_entry_controls & VM_ENTRY_LOAD_BNDCFGS))
-			vmcs_write64(GUEST_BNDCFGS, vmcs12->guest_bndcfgs);
-		else
-			vmcs_write64(GUEST_BNDCFGS, vmx->nested.vmcs01_guest_bndcfgs);
-	}
+	if (kvm_mpx_supported() && vmx->nested.nested_run_pending &&
+	    (vmcs12->vm_entry_controls & VM_ENTRY_LOAD_BNDCFGS))
+		vmcs_write64(GUEST_BNDCFGS, vmcs12->guest_bndcfgs);
 }
 
 /*
@@ -2292,6 +2288,9 @@ static int prepare_vmcs02(struct kvm_vcp
 		kvm_set_dr(vcpu, 7, vcpu->arch.dr7);
 		vmcs_write64(GUEST_IA32_DEBUGCTL, vmx->nested.vmcs01_debugctl);
 	}
+	if (kvm_mpx_supported() && (!vmx->nested.nested_run_pending ||
+	    !(vmcs12->vm_entry_controls & VM_ENTRY_LOAD_BNDCFGS)))
+		vmcs_write64(GUEST_BNDCFGS, vmx->nested.vmcs01_guest_bndcfgs);
 	vmx_set_rflags(vcpu, vmcs12->guest_rflags);
 
 	/* EXCEPTION_BITMAP and CR0_GUEST_HOST_MASK should basically be the
author maintainer
#301 lore raw

PATCH 300/371 KVM: VMX: Fix handling of #MC that occurs during VM-Entry

In reply to Greg Kroah-Hartman

From: Sean Christopherson <[email protected]>

commit beb8d93b3e423043e079ef3dda19dad7b28467a8 upstream.

A previous fix to prevent KVM from consuming stale VMCS state after a failed VM-Entry inadvertantly blocked KVM's handling of machine checks that occur during VM-Entry.

Per Intel's SDM, a #MC during VM-Entry is handled in one of three ways, depending on when the #MC is recognoized. As it pertains to this bug fix, the third case explicitly states EXIT_REASON_MCE_DURING_VMENTRY is handled like any other VM-Exit during VM-Entry, i.e. sets bit 31 to indicate the VM-Entry failed.

If a machine-check event occurs during a VM entry, one of the following occurs:
- The machine-check event is handled as if it occurred before the VM entry:

        ...
 - The machine-check event is handled after VM entry completes:
        ...

- A VM-entry failure occurs as described in Section 26.7. The basic exit reason is 41, for "VM-entry failure due to machine-check event".

Explicitly handle EXIT_REASON_MCE_DURING_VMENTRY as a one-off case in vmx_vcpu_run() instead of binning it into vmx_complete_atomic_exit(). Doing so allows vmx_vcpu_run() to handle VMX_EXIT_REASONS_FAILED_VMENTRY in a sane fashion and also simplifies vmx_complete_atomic_exit() since VMCS.VM_EXIT_INTR_INFO is guaranteed to be fresh.

Fixes b060ca3b2e9e7 ("kvm: vmx: Handle VMLAUNCH/VMRESUME failure properly")
---
 arch/x86/kvm/vmx/vmx.c |   20 ++++++++------------
 1 file changed, 8 insertions(+), 12 deletions(-)
arch/x86/kvm/vmx/vmx.c +8 -12
@@ -6102,28 +6102,21 @@ static void vmx_apicv_post_state_restore
 
 static void vmx_complete_atomic_exit(struct vcpu_vmx *vmx)
 {
-	u32 exit_intr_info = 0;
-	u16 basic_exit_reason = (u16)vmx->exit_reason;
-
-	if (!(basic_exit_reason == EXIT_REASON_MCE_DURING_VMENTRY
-	      || basic_exit_reason == EXIT_REASON_EXCEPTION_NMI))
+	if (vmx->exit_reason != EXIT_REASON_EXCEPTION_NMI)
 		return;
 
-	if (!(vmx->exit_reason & VMX_EXIT_REASONS_FAILED_VMENTRY))
-		exit_intr_info = vmcs_read32(VM_EXIT_INTR_INFO);
-	vmx->exit_intr_info = exit_intr_info;
+	vmx->exit_intr_info = vmcs_read32(VM_EXIT_INTR_INFO);
 
 	/* if exit due to PF check for async PF */
-	if (is_page_fault(exit_intr_info))
+	if (is_page_fault(vmx->exit_intr_info))
 		vmx->vcpu.arch.apf.host_apf_reason = kvm_read_and_reset_pf_reason();
 
 	/* Handle machine checks before interrupts are enabled */
-	if (basic_exit_reason == EXIT_REASON_MCE_DURING_VMENTRY ||
-	    is_machine_check(exit_intr_info))
+	if (is_machine_check(vmx->exit_intr_info))
 		kvm_machine_check();
 
 	/* We need to handle NMIs before interrupts are enabled */
-	if (is_nmi(exit_intr_info)) {
+	if (is_nmi(vmx->exit_intr_info)) {
 		kvm_before_interrupt(&vmx->vcpu);
 		asm("int $2");
 		kvm_after_interrupt(&vmx->vcpu);
@@ -6526,6 +6519,9 @@ static void vmx_vcpu_run(struct kvm_vcpu
 	vmx->idt_vectoring_info = 0;
 
 	vmx->exit_reason = vmx->fail ? 0xdead : vmcs_read32(VM_EXIT_REASON);
+	if ((u16)vmx->exit_reason == EXIT_REASON_MCE_DURING_VMENTRY)
+		kvm_machine_check();
+
 	if (vmx->fail || (vmx->exit_reason & VMX_EXIT_REASONS_FAILED_VMENTRY))
 		return;
 
author maintainer
#302 lore raw

PATCH 301/371 KVM: VMX: check CPUID before allowing read/write of IA32_XSS

In reply to Greg Kroah-Hartman

From: Wanpeng Li <[email protected]>

commit 4d763b168e9c5c366b05812c7bba7662e5ea3669 upstream.

Raise #GP when guest read/write IA32_XSS, but the CPUID bits say that it shouldn't exist.

Fixes 203000993de5 (kvm: vmx: add MSR logic for XSAVES)
Cc Paolo Bonzini <[email protected]>
Cc Radim Krčmář <[email protected]>
---
 arch/x86/kvm/vmx/vmx.c |   10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
arch/x86/kvm/vmx/vmx.c +8 -2
@@ -1718,7 +1718,10 @@ static int vmx_get_msr(struct kvm_vcpu *
 		return vmx_get_vmx_msr(&vmx->nested.msrs, msr_info->index,
 				       &msr_info->data);
 	case MSR_IA32_XSS:
-		if (!vmx_xsaves_supported())
+		if (!vmx_xsaves_supported() ||
+		    (!msr_info->host_initiated &&
+		     !(guest_cpuid_has(vcpu, X86_FEATURE_XSAVE) &&
+		       guest_cpuid_has(vcpu, X86_FEATURE_XSAVES))))
 			return 1;
 		msr_info->data = vcpu->arch.ia32_xss;
 		break;
@@ -1929,7 +1932,10 @@ static int vmx_set_msr(struct kvm_vcpu *
 			return 1;
 		return vmx_set_vmx_msr(vcpu, msr_index, data);
 	case MSR_IA32_XSS:
-		if (!vmx_xsaves_supported())
+		if (!vmx_xsaves_supported() ||
+		    (!msr_info->host_initiated &&
+		     !(guest_cpuid_has(vcpu, X86_FEATURE_XSAVE) &&
+		       guest_cpuid_has(vcpu, X86_FEATURE_XSAVES))))
 			return 1;
 		/*
 		 * The only supported bit as of Skylake is bit 8, but
author maintainer
#303 lore raw

PATCH 302/371 KVM: PPC: Book3S HV: Signed extend decrementer value if not using large decrementer

In reply to Greg Kroah-Hartman

From: Suraj Jitindar Singh <[email protected]>

commit 869537709ebf1dc865e75c3fc97b23f8acf37c16 upstream.

On POWER9 the decrementer can operate in large decrementer mode where the decrementer is 56 bits and signed extended to 64 bits. When not operating in this mode the decrementer behaves as a 32 bit decrementer which is NOT signed extended (as on POWER8).

Currently when reading a guest decrementer value we don't take into account whether the large decrementer is enabled or not, and this means the value will be incorrect when the guest is not using the large decrementer. Fix this by sign extending the value read when the guest isn't using the large decrementer.

Fixes 95a6432ce903 ("KVM: PPC: Book3S HV: Streamlined guest entry/exit path on P9 for radix guests")
Cc [email protected] # v4.20+
---
 arch/powerpc/kvm/book3s_hv.c |    2 ++
 1 file changed, 2 insertions(+)
arch/powerpc/kvm/book3s_hv.c +2 -0
@@ -3568,6 +3568,8 @@ int kvmhv_p9_guest_entry(struct kvm_vcpu
 
 	vcpu->arch.slb_max = 0;
 	dec = mfspr(SPRN_DEC);
+	if (!(lpcr & LPCR_LD)) /* Sign extend if not using large decrementer */
+		dec = (s32) dec;
 	tb = mftb();
 	vcpu->arch.dec_expires = dec + tb;
 	vcpu->cpu = -1;
author maintainer
#304 lore raw

PATCH 303/371 KVM: PPC: Book3S HV: Clear pending decrementer exceptions on nested guest entry

In reply to Greg Kroah-Hartman

From: Suraj Jitindar Singh <[email protected]>

commit 3c25ab35fbc8526ac0c9b298e8a78e7ad7a55479 upstream.

If we enter an L1 guest with a pending decrementer exception then this is cleared on guest exit if the guest has writtien a positive value into the decrementer (indicating that it handled the decrementer exception) since there is no other way to detect that the guest has handled the pending exception and that it should be dequeued. In the event that the L1 guest tries to run a nested (L2) guest immediately after this and the L2 guest decrementer is negative (which is loaded by L1 before making the H_ENTER_NESTED hcall), then the pending decrementer exception isn't cleared and the L2 entry is blocked since L1 has a pending exception, even though L1 may have already handled the exception and written a positive value for it's decrementer. This results in a loop of L1 trying to enter the L2 guest and L0 blocking the entry since L1 has an interrupt pending with the outcome being that L2 never gets to run and hangs.

Fix this by clearing any pending decrementer exceptions when L1 makes the H_ENTER_NESTED hcall since it won't do this if it's decrementer has gone negative, and anyway it's decrementer has been communicated to L0 in the hdec_expires field and L0 will return control to L1 when this goes negative by delivering an H_DECREMENTER exception.

Fixes 95a6432ce903 ("KVM: PPC: Book3S HV: Streamlined guest entry/exit path on P9 for radix guests")
Cc [email protected] # v4.20+
---
 arch/powerpc/kvm/book3s_hv.c |   11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)
arch/powerpc/kvm/book3s_hv.c +9 -2
@@ -4084,8 +4084,15 @@ int kvmhv_run_single_vcpu(struct kvm_run
 
 	preempt_enable();
 
-	/* cancel pending decrementer exception if DEC is now positive */
-	if (get_tb() < vcpu->arch.dec_expires && kvmppc_core_pending_dec(vcpu))
+	/*
+	 * cancel pending decrementer exception if DEC is now positive, or if
+	 * entering a nested guest in which case the decrementer is now owned
+	 * by L2 and the L1 decrementer is provided in hdec_expires
+	 */
+	if (kvmppc_core_pending_dec(vcpu) &&
+			((get_tb() < vcpu->arch.dec_expires) ||
+			 (trap == BOOK3S_INTERRUPT_SYSCALL &&
+			  kvmppc_get_gpr(vcpu, 3) == H_ENTER_NESTED)))
 		kvmppc_core_dequeue_dec(vcpu);
 
 	trace_kvm_guest_exit(vcpu);
author maintainer
#305 lore raw

PATCH 304/371 KVM: PPC: Book3S HV: Fix CR0 setting in TM emulation

In reply to Greg Kroah-Hartman

From: Michael Neuling <[email protected]>

commit 3fefd1cd95df04da67c83c1cb93b663f04b3324f upstream.

When emulating tsr, treclaim and trechkpt, we incorrectly set CR0. The code currently sets:

    CR0 <- 00 || MSR[TS]
but according to the ISA it should be:
    CR0 <-  0 || MSR[TS] || 0

This fixes the bit shift to put the bits in the correct location.

This is a data integrity issue as CR0 is corrupted.

Fixes 4bb3c7a0208f ("KVM: PPC: Book3S HV: Work around transactional memory bugs in POWER9")
Cc [email protected] # v4.17+
---
 arch/powerpc/kvm/book3s_hv_tm.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
arch/powerpc/kvm/book3s_hv_tm.c +3 -3
@@ -131,7 +131,7 @@ int kvmhv_p9_tm_emulation(struct kvm_vcp
 		}
 		/* Set CR0 to indicate previous transactional state */
 		vcpu->arch.regs.ccr = (vcpu->arch.regs.ccr & 0x0fffffff) |
-			(((msr & MSR_TS_MASK) >> MSR_TS_S_LG) << 28);
+			(((msr & MSR_TS_MASK) >> MSR_TS_S_LG) << 29);
 		/* L=1 => tresume, L=0 => tsuspend */
 		if (instr & (1 << 21)) {
 			if (MSR_TM_SUSPENDED(msr))
@@ -175,7 +175,7 @@ int kvmhv_p9_tm_emulation(struct kvm_vcp
 
 		/* Set CR0 to indicate previous transactional state */
 		vcpu->arch.regs.ccr = (vcpu->arch.regs.ccr & 0x0fffffff) |
-			(((msr & MSR_TS_MASK) >> MSR_TS_S_LG) << 28);
+			(((msr & MSR_TS_MASK) >> MSR_TS_S_LG) << 29);
 		vcpu->arch.shregs.msr &= ~MSR_TS_MASK;
 		return RESUME_GUEST;
 
@@ -205,7 +205,7 @@ int kvmhv_p9_tm_emulation(struct kvm_vcp
 
 		/* Set CR0 to indicate previous transactional state */
 		vcpu->arch.regs.ccr = (vcpu->arch.regs.ccr & 0x0fffffff) |
-			(((msr & MSR_TS_MASK) >> MSR_TS_S_LG) << 28);
+			(((msr & MSR_TS_MASK) >> MSR_TS_S_LG) << 29);
 		vcpu->arch.shregs.msr = msr | MSR_TS_S;
 		return RESUME_GUEST;
 	}
author maintainer
#306 lore raw

PATCH 305/371 KVM: x86/vPMU: refine kvm_pmu err msg when event creation failed

In reply to Greg Kroah-Hartman

From: Like Xu <[email protected]>

commit 6fc3977ccc5d3c22e851f2dce2d3ce2a0a843842 upstream.

If a perf_event creation fails due to any reason of the host perf subsystem, it has no chance to log the corresponding event for guest which may cause abnormal sampling data in guest result. In debug mode, this message helps to understand the state of vPMC and we may not limit the number of occurrences but not in a spamming style.

---
 arch/x86/kvm/pmu.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/x86/kvm/pmu.c +2 -2
@@ -131,8 +131,8 @@ static void pmc_reprogram_counter(struct
 						 intr ? kvm_perf_overflow_intr :
 						 kvm_perf_overflow, pmc);
 	if (IS_ERR(event)) {
-		printk_once("kvm_pmu: event creation failed %ld\n",
-			    PTR_ERR(event));
+		pr_debug_ratelimited("kvm_pmu: event creation failed %ld for pmc->idx = %d\n",
+			    PTR_ERR(event), pmc->idx);
 		return;
 	}
 
author maintainer
#307 lore raw

PATCH 306/371 arm64: tegra: Fix AGIC register range

In reply to Greg Kroah-Hartman

From: Jon Hunter <[email protected]>

commit ba24eee6686f6ed3738602b54d959253316a9541 upstream.

The Tegra AGIC interrupt controller is an ARM GIC400 interrupt controller. Per the ARM GIC device-tree binding, the first address region is for the GIC distributor registers and the second address region is for the GIC CPU interface registers. The address space for the distributor registers is 4kB, but currently this is incorrectly defined as 8kB for the Tegra AGIC and overlaps with the CPU interface registers. Correct the address space for the distributor to be 4kB.

Fixes bcdbde433542 ("arm64: tegra: Add AGIC node for Tegra210")
---
 arch/arm64/boot/dts/nvidia/tegra210.dtsi |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/boot/dts/nvidia/tegra210.dtsi +1 -1
@@ -1250,7 +1250,7 @@
 			compatible = "nvidia,tegra210-agic";
 			#interrupt-cells = <3>;
 			interrupt-controller;
-			reg = <0x702f9000 0x2000>,
+			reg = <0x702f9000 0x1000>,
 			      <0x702fa000 0x2000>;
 			interrupts = <GIC_SPI 102 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_HIGH)>;
 			clocks = <&tegra_car TEGRA210_CLK_APE>;
author maintainer
#308 lore raw

PATCH 307/371 arm64: irqflags: Add condition flags to inline asm clobber list

In reply to Greg Kroah-Hartman

From: Julien Thierry <[email protected]>

commit f57065782f245ca96f1472209a485073bbc11247 upstream.

Some of the inline assembly instruction use the condition flags and need to include "cc" in the clobber list.

Fixes 4a503217ce37 ("arm64: irqflags: Use ICC_PMR_EL1 for interrupt masking")
Cc <[email protected]> # 5.1.x-
Cc Will Deacon <[email protected]>
---
 arch/arm64/include/asm/irqflags.h |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/include/asm/irqflags.h +2 -2
@@ -92,7 +92,7 @@ static inline unsigned long arch_local_s
 			ARM64_HAS_IRQ_PRIO_MASKING)
 		: "=&r" (flags), "+r" (daif_bits)
 		: "r" ((unsigned long) GIC_PRIO_IRQOFF)
-		: "memory");
+		: "cc", "memory");
 
 	return flags;
 }
@@ -136,7 +136,7 @@ static inline int arch_irqs_disabled_fla
 			ARM64_HAS_IRQ_PRIO_MASKING)
 		: "=&r" (res)
 		: "r" ((int) flags)
-		: "memory");
+		: "cc", "memory");
 
 	return res;
 }
author maintainer
#309 lore raw

PATCH 308/371 signal/usb: Replace kill_pid_info_as_cred with kill_pid_usb_asyncio

In reply to Greg Kroah-Hartman

From: Eric W. Biederman <[email protected]>

commit 70f1b0d34bdf03065fe869e93cc17cad1ea20c4a upstream.

The usb support for asyncio encoded one of it's values in the wrong field. It should have used si_value but instead used si_addr which is not present in the _rt union member of struct siginfo.

The practical result of this is that on a 64bit big endian kernel when delivering a signal to a 32bit process the si_addr field is set to NULL, instead of the expected pointer value.

This issue can not be fixed in copy_siginfo_to_user32 as the usb usage of the the _sigfault (aka si_addr) member of the siginfo union when SI_ASYNCIO is set is incompatible with the POSIX and glibc usage of the _rt member of the siginfo union.

Therefore replace kill_pid_info_as_cred with kill_pid_usb_asyncio a dedicated function for this one specific case. There are no other users of kill_pid_info_as_cred so this specialization should have no impact on the amount of code in the kernel. Have kill_pid_usb_asyncio take instead of a siginfo_t which is difficult and error prone, 3 arguments, a signal number, an errno value, and an address enconded as a sigval_t. The encoding of the address as a sigval_t allows the code that reads the userspace request for a signal to handle this compat issue along with all of the other compat issues.

Add BUILD_BUG_ONs in kernel/signal.c to ensure that we can now place the pointer value at the in si_pid (instead of si_addr). That is the code now verifies that si_pid and si_addr always occur at the same location. Further the code veries that for native structures a value placed in si_pid and spilling into si_uid will appear in userspace in si_addr (on a byte by byte copy of siginfo or a field by field copy of siginfo). The code also verifies that for a 64bit kernel and a 32bit userspace the 32bit pointer will fit in si_pid.

I have used the usbsig.c program below written by Alan Stern and slightly tweaked by me to run on a big endian machine to verify the issue exists (on sparc64) and to confirm the patch below fixes the issue.

/* usbsig.c -- test USB async signal delivery */

#define _GNU_SOURCE
#include <stdio.h>
#include <fcntl.h>
#include <signal.h>
#include <string.h>
#include <sys/ioctl.h>
#include <unistd.h>
#include <endian.h>
#include <linux/usb/ch9.h>
#include <linux/usbdevice_fs.h>

static struct usbdevfs_urb urb;
static struct usbdevfs_disconnectsignal ds;
static volatile sig_atomic_t done = 0;

void urb_handler(int sig, siginfo_t *info , void *ucontext)
{
printf("Got signal %d, signo %d errno %d code %d addr: %p urb: %p\n",
sig, info->si_signo, info->si_errno, info->si_code, info->si_addr, &urb);

printf("%s\n", (info->si_addr == &urb) ? "Good" : "Bad");
}

void ds_handler(int sig, siginfo_t *info , void *ucontext)
{
printf("Got signal %d, signo %d errno %d code %d addr: %p ds: %p\n",
sig, info->si_signo, info->si_errno, info->si_code, info->si_addr, &ds);

printf("%s\n", (info->si_addr == &ds) ? "Good" : "Bad"); done = 1;
}

int main(int argc, char **argv)
{
char *devfilename;
int fd;
int rc;
struct sigaction act;
struct usb_ctrlrequest *req;
void *ptr;
char buf[80];

if (argc != 2) {
fprintf(stderr, "Usage: usbsig device-file-name\n"); return 1;
}

devfilename = argv[1];
fd = open(devfilename, O_RDWR);
if (fd == -1) {
perror("Error opening device file"); return 1;
}

act.sa_sigaction = urb_handler;
sigemptyset(&act.sa_mask);
act.sa_flags = SA_SIGINFO;

rc = sigaction(SIGUSR1, &act, NULL);
if (rc == -1) {
perror("Error in sigaction");
return 1;
}

act.sa_sigaction = ds_handler;
sigemptyset(&act.sa_mask);
act.sa_flags = SA_SIGINFO;

rc = sigaction(SIGUSR2, &act, NULL);
if (rc == -1) {
perror("Error in sigaction");
return 1;
}

memset(&urb, 0, sizeof(urb));
urb.type = USBDEVFS_URB_TYPE_CONTROL;
urb.endpoint = USB_DIR_IN | 0;
urb.buffer = buf;
urb.buffer_length = sizeof(buf);
urb.signr = SIGUSR1;

req = (struct usb_ctrlrequest *) buf;
req->bRequestType = USB_DIR_IN | USB_TYPE_STANDARD | USB_RECIP_DEVICE; req->bRequest = USB_REQ_GET_DESCRIPTOR;
req->wValue = htole16(USB_DT_DEVICE << 8);
req->wIndex = htole16(0);
req->wLength = htole16(sizeof(buf) - sizeof(*req));

rc = ioctl(fd, USBDEVFS_SUBMITURB, &urb); if (rc == -1) {
perror("Error in SUBMITURB ioctl"); return 1;
}

rc = ioctl(fd, USBDEVFS_REAPURB, &ptr);
if (rc == -1) {
perror("Error in REAPURB ioctl");
return 1;
}

memset(&ds, 0, sizeof(ds));
ds.signr = SIGUSR2;
ds.context = &ds;
rc = ioctl(fd, USBDEVFS_DISCSIGNAL, &ds); if (rc == -1) {
perror("Error in DISCSIGNAL ioctl"); return 1;
}

printf("Waiting for usb disconnect\n");
while (!done) {
sleep(1);
}

close(fd);
return 0;
}

---
 drivers/usb/core/devio.c     |   48 ++++++++++++++---------------
 include/linux/sched/signal.h |    2 -
 kernel/signal.c              |   69 ++++++++++++++++++++++++++++++++++++++-----
 3 files changed, 86 insertions(+), 33 deletions(-)
drivers/usb/core/devio.c +24 -24
@@ -63,7 +63,7 @@ struct usb_dev_state {
 	unsigned int discsignr;
 	struct pid *disc_pid;
 	const struct cred *cred;
-	void __user *disccontext;
+	sigval_t disccontext;
 	unsigned long ifclaimed;
 	u32 disabled_bulk_eps;
 	bool privileges_dropped;
@@ -90,6 +90,7 @@ struct async {
 	unsigned int ifnum;
 	void __user *userbuffer;
 	void __user *userurb;
+	sigval_t userurb_sigval;
 	struct urb *urb;
 	struct usb_memory *usbm;
 	unsigned int mem_usage;
@@ -582,22 +583,19 @@ static void async_completed(struct urb *
 {
 	struct async *as = urb->context;
 	struct usb_dev_state *ps = as->ps;
-	struct kernel_siginfo sinfo;
 	struct pid *pid = NULL;
 	const struct cred *cred = NULL;
 	unsigned long flags;
-	int signr;
+	sigval_t addr;
+	int signr, errno;
 
 	spin_lock_irqsave(&ps->lock, flags);
 	list_move_tail(&as->asynclist, &ps->async_completed);
 	as->status = urb->status;
 	signr = as->signr;
 	if (signr) {
-		clear_siginfo(&sinfo);
-		sinfo.si_signo = as->signr;
-		sinfo.si_errno = as->status;
-		sinfo.si_code = SI_ASYNCIO;
-		sinfo.si_addr = as->userurb;
+		errno = as->status;
+		addr = as->userurb_sigval;
 		pid = get_pid(as->pid);
 		cred = get_cred(as->cred);
 	}
@@ -615,7 +613,7 @@ static void async_completed(struct urb *
 	spin_unlock_irqrestore(&ps->lock, flags);
 
 	if (signr) {
-		kill_pid_info_as_cred(sinfo.si_signo, &sinfo, pid, cred);
+		kill_pid_usb_asyncio(signr, errno, addr, pid, cred);
 		put_pid(pid);
 		put_cred(cred);
 	}
@@ -1427,7 +1425,7 @@ find_memory_area(struct usb_dev_state *p
 
 static int proc_do_submiturb(struct usb_dev_state *ps, struct usbdevfs_urb *uurb,
 			struct usbdevfs_iso_packet_desc __user *iso_frame_desc,
-			void __user *arg)
+			void __user *arg, sigval_t userurb_sigval)
 {
 	struct usbdevfs_iso_packet_desc *isopkt = NULL;
 	struct usb_host_endpoint *ep;
@@ -1727,6 +1725,7 @@ static int proc_do_submiturb(struct usb_
 	isopkt = NULL;
 	as->ps = ps;
 	as->userurb = arg;
+	as->userurb_sigval = userurb_sigval;
 	if (as->usbm) {
 		unsigned long uurb_start = (unsigned long)uurb->buffer;
 
@@ -1801,13 +1800,17 @@ static int proc_do_submiturb(struct usb_
 static int proc_submiturb(struct usb_dev_state *ps, void __user *arg)
 {
 	struct usbdevfs_urb uurb;
+	sigval_t userurb_sigval;
 
 	if (copy_from_user(&uurb, arg, sizeof(uurb)))
 		return -EFAULT;
 
+	memset(&userurb_sigval, 0, sizeof(userurb_sigval));
+	userurb_sigval.sival_ptr = arg;
+
 	return proc_do_submiturb(ps, &uurb,
 			(((struct usbdevfs_urb __user *)arg)->iso_frame_desc),
-			arg);
+			arg, userurb_sigval);
 }
 
 static int proc_unlinkurb(struct usb_dev_state *ps, void __user *arg)
@@ -1977,7 +1980,7 @@ static int proc_disconnectsignal_compat(
 	if (copy_from_user(&ds, arg, sizeof(ds)))
 		return -EFAULT;
 	ps->discsignr = ds.signr;
-	ps->disccontext = compat_ptr(ds.context);
+	ps->disccontext.sival_int = ds.context;
 	return 0;
 }
 
@@ -2005,13 +2008,17 @@ static int get_urb32(struct usbdevfs_urb
 static int proc_submiturb_compat(struct usb_dev_state *ps, void __user *arg)
 {
 	struct usbdevfs_urb uurb;
+	sigval_t userurb_sigval;
 
 	if (get_urb32(&uurb, (struct usbdevfs_urb32 __user *)arg))
 		return -EFAULT;
 
+	memset(&userurb_sigval, 0, sizeof(userurb_sigval));
+	userurb_sigval.sival_int = ptr_to_compat(arg);
+
 	return proc_do_submiturb(ps, &uurb,
 			((struct usbdevfs_urb32 __user *)arg)->iso_frame_desc,
-			arg);
+			arg, userurb_sigval);
 }
 
 static int processcompl_compat(struct async *as, void __user * __user *arg)
@@ -2092,7 +2099,7 @@ static int proc_disconnectsignal(struct
 	if (copy_from_user(&ds, arg, sizeof(ds)))
 		return -EFAULT;
 	ps->discsignr = ds.signr;
-	ps->disccontext = ds.context;
+	ps->disccontext.sival_ptr = ds.context;
 	return 0;
 }
 
@@ -2614,22 +2621,15 @@ const struct file_operations usbdev_file
 static void usbdev_remove(struct usb_device *udev)
 {
 	struct usb_dev_state *ps;
-	struct kernel_siginfo sinfo;
 
 	while (!list_empty(&udev->filelist)) {
 		ps = list_entry(udev->filelist.next, struct usb_dev_state, list);
 		destroy_all_async(ps);
 		wake_up_all(&ps->wait);
 		list_del_init(&ps->list);
-		if (ps->discsignr) {
-			clear_siginfo(&sinfo);
-			sinfo.si_signo = ps->discsignr;
-			sinfo.si_errno = EPIPE;
-			sinfo.si_code = SI_ASYNCIO;
-			sinfo.si_addr = ps->disccontext;
-			kill_pid_info_as_cred(ps->discsignr, &sinfo,
-					ps->disc_pid, ps->cred);
-		}
+		if (ps->discsignr)
+			kill_pid_usb_asyncio(ps->discsignr, EPIPE, ps->disccontext,
+					     ps->disc_pid, ps->cred);
 	}
 }
 
include/linux/sched/signal.h +1 -1
@@ -328,7 +328,7 @@ extern void force_sigsegv(int sig, struc
 extern int force_sig_info(int, struct kernel_siginfo *, struct task_struct *);
 extern int __kill_pgrp_info(int sig, struct kernel_siginfo *info, struct pid *pgrp);
 extern int kill_pid_info(int sig, struct kernel_siginfo *info, struct pid *pid);
-extern int kill_pid_info_as_cred(int, struct kernel_siginfo *, struct pid *,
+extern int kill_pid_usb_asyncio(int sig, int errno, sigval_t addr, struct pid *,
 				const struct cred *);
 extern int kill_pgrp(struct pid *pid, int sig, int priv);
 extern int kill_pid(struct pid *pid, int sig, int priv);
kernel/signal.c +61 -8

104 lines not on this page · raw message

author maintainer
#310 lore raw

PATCH 309/371 signal: Correct namespace fixups of si_pid and si_uid

In reply to Greg Kroah-Hartman

From: Eric W. Biederman <[email protected]>

commit 7a0cf094944e2540758b7f957eb6846d5126f535 upstream.

The function send_signal was split from __send_signal so that it would be possible to bypass the namespace logic based upon current[1]. As it turns out the si_pid and the si_uid fixup are both inappropriate in the case of kill_pid_usb_asyncio so move that logic into send_signal.

It is difficult to arrange but possible for a signal with an si_code of SI_TIMER or SI_SIGIO to be sent across namespace boundaries. In which case tests for when it is ok to change si_pid and si_uid based on SI_FROMUSER are incorrect. Replace the use of SI_FROMUSER with a new test has_si_pid_and_used based on siginfo_layout.

Now that the uid fixup is no longer present after expanding SEND_SIG_NOINFO properly calculate the si_uid that the target task needs to read.

[1] 7978b567d315 ("signals: add from_ancestor_ns parameter to send_signal()")

Fixes 6588c1e3ff01 ("signals: SI_USER: Masquerade si_pid when crossing pid ns boundary")
Fixes 6b550f949594 ("user namespace: make signal.c respect user namespaces")
---
 kernel/signal.c |   67 +++++++++++++++++++++++++++++++++-----------------------
 1 file changed, 40 insertions(+), 27 deletions(-)
kernel/signal.c +40 -27
@@ -1053,27 +1053,6 @@ static inline bool legacy_queue(struct s
 	return (sig < SIGRTMIN) && sigismember(&signals->signal, sig);
 }
 
-#ifdef CONFIG_USER_NS
-static inline void userns_fixup_signal_uid(struct kernel_siginfo *info, struct task_struct *t)
-{
-	if (current_user_ns() == task_cred_xxx(t, user_ns))
-		return;
-
-	if (SI_FROMKERNEL(info))
-		return;
-
-	rcu_read_lock();
-	info->si_uid = from_kuid_munged(task_cred_xxx(t, user_ns),
-					make_kuid(current_user_ns(), info->si_uid));
-	rcu_read_unlock();
-}
-#else
-static inline void userns_fixup_signal_uid(struct kernel_siginfo *info, struct task_struct *t)
-{
-	return;
-}
-#endif
-
 static int __send_signal(int sig, struct kernel_siginfo *info, struct task_struct *t,
 			enum pid_type type, int from_ancestor_ns)
 {
@@ -1131,7 +1110,11 @@ static int __send_signal(int sig, struct
 			q->info.si_code = SI_USER;
 			q->info.si_pid = task_tgid_nr_ns(current,
 							task_active_pid_ns(t));
-			q->info.si_uid = from_kuid_munged(current_user_ns(), current_uid());
+			rcu_read_lock();
+			q->info.si_uid =
+				from_kuid_munged(task_cred_xxx(t, user_ns),
+						 current_uid());
+			rcu_read_unlock();
 			break;
 		case (unsigned long) SEND_SIG_PRIV:
 			clear_siginfo(&q->info);
@@ -1143,13 +1126,8 @@ static int __send_signal(int sig, struct
 			break;
 		default:
 			copy_siginfo(&q->info, info);
-			if (from_ancestor_ns)
-				q->info.si_pid = 0;
 			break;
 		}
-
-		userns_fixup_signal_uid(&q->info, t);
-
 	} else if (!is_si_special(info)) {
 		if (sig >= SIGRTMIN && info->si_code != SI_USER) {
 			/*
@@ -1193,6 +1171,28 @@ ret:
 	return ret;
 }
 
+static inline bool has_si_pid_and_uid(struct kernel_siginfo *info)
+{
+	bool ret = false;
+	switch (siginfo_layout(info->si_signo, info->si_code)) {
+	case SIL_KILL:
+	case SIL_CHLD:
+	case SIL_RT:
+		ret = true;
+		break;
+	case SIL_TIMER:
+	case SIL_POLL:
+	case SIL_FAULT:
+	case SIL_FAULT_MCEERR:
+	case SIL_FAULT_BNDERR:
+	case SIL_FAULT_PKUERR:
+	case SIL_SYS:
+		ret = false;
+		break;
+	}
+	return ret;
+}
+
 static int send_signal(int sig, struct kernel_siginfo *info, struct task_struct *t,
 			enum pid_type type)
 {
@@ -1202,7 +1202,20 @@ static int send_signal(int sig, struct k
 	from_ancestor_ns = si_fromuser(info) &&
 			   !task_pid_nr_ns(current, task_active_pid_ns(t));
 #endif
+	if (!is_si_special(info) && has_si_pid_and_uid(info)) {
+		struct user_namespace *t_user_ns;
+
+		rcu_read_lock();
+		t_user_ns = task_cred_xxx(t, user_ns);
+		if (current_user_ns() != t_user_ns) {
+			kuid_t uid = make_kuid(current_user_ns(), info->si_uid);
+			info->si_uid = from_kuid_munged(t_user_ns, uid);
+		}
+		rcu_read_unlock();
 
+		if (!task_pid_nr_ns(current, task_active_pid_ns(t)))
+			info->si_pid = 0;
+	}
 	return __send_signal(sig, info, t, type, from_ancestor_ns);
 }