Skip to contentSkip to conversation

6.1.30-rc1 review

PATCH In review Greg Kroah-Hartman 15 participants 307 messages May 22, 2023 – May 23, 2023 Page 2 of 3

Subsystems (6)
lore
Share & tools
lore
author maintainer
#130 lore raw

PATCH 130/292 ASoC: amd: yc: Add ThinkBook 14 G5+ ARP to quirks list for acp6x

In reply to Greg Kroah-Hartman

From: Baishan Jiang <[email protected]>

[ Upstream commit a8f5da0bf4d85a6ad03810d902aba61c572102a6 ]

ThinkBook 14 G5+ ARP uses Ryzen 7735H processor, and has the same microphone problem as ThinkBook 14 G4+ ARA.

Adding 21HY to acp6x quirks table enables microphone for ThinkBook 14 G5+ ARP.

---
 sound/soc/amd/yc/acp6x-mach.c | 7 +++++++
 1 file changed, 7 insertions(+)
sound/soc/amd/yc/acp6x-mach.c +7 -0
index 1d59163a882ca..b9958e5553674 100644
@@ -185,6 +185,13 @@ static const struct dmi_system_id yc_acp_quirk_table[] = {
 			DMI_MATCH(DMI_PRODUCT_NAME, "21EN"),
 		}
 	},
+	{
+		.driver_data = &acp6x_card,
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "21HY"),
+		}
+	},
 	{
 		.driver_data = &acp6x_card,
 		.matches = {
Signature · 1 line
2.39.2
author maintainer
#131 lore raw

PATCH 131/292 HID: apple: Set the tilde quirk flag on the Geyser 3

In reply to Greg Kroah-Hartman

From: Alex Henrie <[email protected]>

[ Upstream commit 29e1ecc197d410ee59c8877098d54cf417075f7d ]

I was finally able to obtain a MacBook1,1 to test and I've now confirmed that it has the tilde key quirk as well:

Product Model Year System CPU Shape Labels Country Quirky
============================================================================
05ac:0218 A1181 2006 MacBook1,1 T2500 ISO British 13 Yes

---
 drivers/hid/hid-apple.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/hid/hid-apple.c +2 -1
index f21b1c4ca8254..37b2ce9b50fe8 100644
@@ -854,7 +854,8 @@ static const struct hid_device_id apple_devices[] = {
 	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER3_ANSI),
 		.driver_data = APPLE_NUMLOCK_EMULATION | APPLE_HAS_FN },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER3_ISO),
-		.driver_data = APPLE_NUMLOCK_EMULATION | APPLE_HAS_FN },
+		.driver_data = APPLE_NUMLOCK_EMULATION | APPLE_HAS_FN |
+			APPLE_ISO_TILDE_QUIRK },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER3_JIS),
 		.driver_data = APPLE_NUMLOCK_EMULATION | APPLE_HAS_FN |
 			APPLE_RDESC_JIS },
Signature · 1 line
2.39.2
author maintainer
#132 lore raw

PATCH 132/292 HID: Ignore battery for ELAN touchscreen on ROG Flow X13 GV301RA

In reply to Greg Kroah-Hartman

From: weiliang1503 <[email protected]>

[ Upstream commit 35903009dbde804a1565dc89e431c0f15179f054 ]

Ignore the reported battery level of the built-in touchscreen to suppress battery warnings when a stylus is used. The device ID was added and the battery ignore quirk was enabled.

---
 drivers/hid/hid-ids.h   | 1 +
 drivers/hid/hid-input.c | 2 ++
 2 files changed, 3 insertions(+)
drivers/hid/hid-ids.h +1 -0
index 2235d78784b1b..53c6692d77714 100644
@@ -415,6 +415,7 @@
 #define I2C_DEVICE_ID_HP_SPECTRE_X360_15	0x2817
 #define I2C_DEVICE_ID_HP_SPECTRE_X360_13_AW0020NG  0x29DF
 #define I2C_DEVICE_ID_ASUS_TP420IA_TOUCHSCREEN 0x2BC8
+#define I2C_DEVICE_ID_ASUS_GV301RA_TOUCHSCREEN 0x2C82
 #define USB_DEVICE_ID_ASUS_UX550VE_TOUCHSCREEN	0x2544
 #define USB_DEVICE_ID_ASUS_UX550_TOUCHSCREEN	0x2706
 #define I2C_DEVICE_ID_SURFACE_GO_TOUCHSCREEN	0x261A
drivers/hid/hid-input.c +2 -0
index c3f80b516f398..3acaaca888acd 100644
@@ -372,6 +372,8 @@ static const struct hid_device_id hid_battery_quirks[] = {
 	  HID_BATTERY_QUIRK_IGNORE },
 	{ HID_I2C_DEVICE(USB_VENDOR_ID_ELAN, I2C_DEVICE_ID_ASUS_TP420IA_TOUCHSCREEN),
 	  HID_BATTERY_QUIRK_IGNORE },
+	{ HID_I2C_DEVICE(USB_VENDOR_ID_ELAN, I2C_DEVICE_ID_ASUS_GV301RA_TOUCHSCREEN),
+	  HID_BATTERY_QUIRK_IGNORE },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ELAN, USB_DEVICE_ID_ASUS_UX550_TOUCHSCREEN),
 	  HID_BATTERY_QUIRK_IGNORE },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_ELAN, USB_DEVICE_ID_ASUS_UX550VE_TOUCHSCREEN),
Signature · 1 line
2.39.2
author maintainer
#133 lore raw

PATCH 133/292 HID: wacom: generic: Set battery quirk only when we see battery data

In reply to Greg Kroah-Hartman

From: Jason Gerecke <[email protected]>

[ Upstream commit bea407a427baa019758f29f4d31b26f008bb8cc6 ]

Some devices will include battery status usages in the HID descriptor but we won't see that battery data for one reason or another. For example, AES sensors won't send battery data unless an AES pen is in proximity. If a user does not have an AES pen but instead only interacts with the AES touchscreen with their fingers then there is no need for us to create a battery object. Similarly, if a family of peripherals shares the same HID descriptor between wired-only and wireless-capable SKUs, users of the former may never see a battery event and will not want a power_supply object created.

---
 drivers/hid/wacom_wac.c | 33 +++++++++++----------------------
 1 file changed, 11 insertions(+), 22 deletions(-)
drivers/hid/wacom_wac.c +11 -22
index 0c6a82c665c1d..d2f500242ed40 100644
@@ -1963,18 +1963,7 @@ static void wacom_map_usage(struct input_dev *input, struct hid_usage *usage,
 static void wacom_wac_battery_usage_mapping(struct hid_device *hdev,
 		struct hid_field *field, struct hid_usage *usage)
 {
-	struct wacom *wacom = hid_get_drvdata(hdev);
-	struct wacom_wac *wacom_wac = &wacom->wacom_wac;
-	struct wacom_features *features = &wacom_wac->features;
-	unsigned equivalent_usage = wacom_equivalent_usage(usage->hid);
-
-	switch (equivalent_usage) {
-	case HID_DG_BATTERYSTRENGTH:
-	case WACOM_HID_WD_BATTERY_LEVEL:
-	case WACOM_HID_WD_BATTERY_CHARGING:
-		features->quirks |= WACOM_QUIRK_BATTERY;
-		break;
-	}
+	return;
 }
 
 static void wacom_wac_battery_event(struct hid_device *hdev, struct hid_field *field,
@@ -1995,18 +1984,21 @@ static void wacom_wac_battery_event(struct hid_device *hdev, struct hid_field *f
 			wacom_wac->hid_data.bat_connected = 1;
 			wacom_wac->hid_data.bat_status = WACOM_POWER_SUPPLY_STATUS_AUTO;
 		}
+		wacom_wac->features.quirks |= WACOM_QUIRK_BATTERY;
 		break;
 	case WACOM_HID_WD_BATTERY_LEVEL:
 		value = value * 100 / (field->logical_maximum - field->logical_minimum);
 		wacom_wac->hid_data.battery_capacity = value;
 		wacom_wac->hid_data.bat_connected = 1;
 		wacom_wac->hid_data.bat_status = WACOM_POWER_SUPPLY_STATUS_AUTO;
+		wacom_wac->features.quirks |= WACOM_QUIRK_BATTERY;
 		break;
 	case WACOM_HID_WD_BATTERY_CHARGING:
 		wacom_wac->hid_data.bat_charging = value;
 		wacom_wac->hid_data.ps_connected = value;
 		wacom_wac->hid_data.bat_connected = 1;
 		wacom_wac->hid_data.bat_status = WACOM_POWER_SUPPLY_STATUS_AUTO;
+		wacom_wac->features.quirks |= WACOM_QUIRK_BATTERY;
 		break;
 	}
 }
@@ -2022,18 +2014,15 @@ static void wacom_wac_battery_report(struct hid_device *hdev,
 {
 	struct wacom *wacom = hid_get_drvdata(hdev);
 	struct wacom_wac *wacom_wac = &wacom->wacom_wac;
-	struct wacom_features *features = &wacom_wac->features;
 
-	if (features->quirks & WACOM_QUIRK_BATTERY) {
-		int status = wacom_wac->hid_data.bat_status;
-		int capacity = wacom_wac->hid_data.battery_capacity;
-		bool charging = wacom_wac->hid_data.bat_charging;
-		bool connected = wacom_wac->hid_data.bat_connected;
-		bool powered = wacom_wac->hid_data.ps_connected;
+	int status = wacom_wac->hid_data.bat_status;
+	int capacity = wacom_wac->hid_data.battery_capacity;
+	bool charging = wacom_wac->hid_data.bat_charging;
+	bool connected = wacom_wac->hid_data.bat_connected;
+	bool powered = wacom_wac->hid_data.ps_connected;
 
-		wacom_notify_battery(wacom_wac, status, capacity, charging,
-				     connected, powered);
-	}
+	wacom_notify_battery(wacom_wac, status, capacity, charging,
+			     connected, powered);
 }
 
 static void wacom_wac_pad_usage_mapping(struct hid_device *hdev,
Signature · 1 line
2.39.2
author maintainer
#134 lore raw

PATCH 134/292 usb: typec: tcpm: fix multiple times discover svids error

In reply to Greg Kroah-Hartman

From: Frank Wang <[email protected]>

[ Upstream commit dac3b192107b978198e89ec0f77375738352e0c8 ]

PD3.0 Spec 6.4.4.3.2 say that only Responder supports 12 or more SVIDs, the Discover SVIDs Command Shall be executed multiple times until a Discover SVIDs VDO is returned ending either with a SVID value of 0x0000 in the last part of the last VDO or with a VDO containing two SVIDs with values of 0x0000.

In the current implementation, if the last VDO does not find that the Discover SVIDs Command would be executed multiple times even if the Responder SVIDs are less than 12, and we found some odd dockers just meet this case. So fix it.

---
 drivers/usb/typec/tcpm/tcpm.c | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)
drivers/usb/typec/tcpm/tcpm.c +15 -1
index 032d21a967799..524099634a1d4 100644
@@ -1514,7 +1514,21 @@ static bool svdm_consume_svids(struct tcpm_port *port, const u32 *p, int cnt)
 		pmdata->svids[pmdata->nsvids++] = svid;
 		tcpm_log(port, "SVID %d: 0x%x", pmdata->nsvids, svid);
 	}
-	return true;
+
+	/*
+	 * PD3.0 Spec 6.4.4.3.2: The SVIDs are returned 2 per VDO (see Table
+	 * 6-43), and can be returned maximum 6 VDOs per response (see Figure
+	 * 6-19). If the Respondersupports 12 or more SVID then the Discover
+	 * SVIDs Command Shall be executed multiple times until a Discover
+	 * SVIDs VDO is returned ending either with a SVID value of 0x0000 in
+	 * the last part of the last VDO or with a VDO containing two SVIDs
+	 * with values of 0x0000.
+	 *
+	 * However, some odd dockers support SVIDs less than 12 but without
+	 * 0x0000 in the last VDO, so we need to break the Discover SVIDs
+	 * request and return false here.
+	 */
+	return cnt == 7;
 abort:
 	tcpm_log(port, "SVID_DISCOVERY_MAX(%d) too low!", SVID_DISCOVERY_MAX);
 	return false;
Signature · 1 line
2.39.2
author maintainer
#135 lore raw

PATCH 135/292 serial: 8250: Reinit port->pm on port specific driver unbind

In reply to Greg Kroah-Hartman

From: Tony Lindgren <[email protected]>

[ Upstream commit 04e82793f068d2f0ffe62fcea03d007a8cdc16a7 ]

When we unbind a serial port hardware specific 8250 driver, the generic serial8250 driver takes over the port. After that we see an oops about 10 seconds later. This can produce the following at least on some TI SoCs:

Unhandled fault: imprecise external abort (0x1406)

Internal error: : 1406 [#1] SMP ARM

Turns out that we may still have the serial port hardware specific driver port->pm in use, and serial8250_pm() tries to call it after the port specific driver is gone:

serial8250_pm [8250_base] from uart_change_pm+0x54/0x8c [serial_base] uart_change_pm [serial_base] from uart_hangup+0x154/0x198 [serial_base] uart_hangup [serial_base] from __tty_hangup.part.0+0x328/0x37c __tty_hangup.part.0 from disassociate_ctty+0x154/0x20c disassociate_ctty from do_exit+0x744/0xaac
do_exit from do_group_exit+0x40/0x8c
do_group_exit from __wake_up_parent+0x0/0x1c

Let's fix the issue by calling serial8250_set_defaults() in serial8250_unregister_port(). This will set the port back to using the serial8250 default functions, and sets the port->pm to point to serial8250_pm.

---
 drivers/tty/serial/8250/8250_core.c | 1 +
 1 file changed, 1 insertion(+)
drivers/tty/serial/8250/8250_core.c +1 -0
index 94fbf0add2ce2..81a5dab1a8286 100644
@@ -1157,6 +1157,7 @@ void serial8250_unregister_port(int line)
 		uart->port.type = PORT_UNKNOWN;
 		uart->port.dev = &serial8250_isa_devs->dev;
 		uart->capabilities = 0;
+		serial8250_init_port(uart);
 		serial8250_apply_quirks(uart);
 		uart_add_one_port(&serial8250_reg, &uart->port);
 	} else {
Signature · 1 line
2.39.2
author maintainer
#136 lore raw

PATCH 136/292 mcb-pci: Reallocate memory region to avoid memory overlapping

In reply to Greg Kroah-Hartman

From: Rodríguez Barbarin, José Javier <[email protected]>

[ Upstream commit 9be24faadd085c284890c3afcec7a0184642315a ]

mcb-pci requests a fixed-size memory region to parse the chameleon table, however, if the chameleon table is smaller that the allocated region, it could overlap with the IP Cores' memory regions.

After parsing the chameleon table, drop/reallocate the memory region with the actual chameleon table size.

---
 drivers/mcb/mcb-pci.c | 27 +++++++++++++++++++++++++--
 1 file changed, 25 insertions(+), 2 deletions(-)
drivers/mcb/mcb-pci.c +25 -2
index dc88232d9af83..53d9202ff9a7c 100644
@@ -31,7 +31,7 @@ static int mcb_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	struct resource *res;
 	struct priv *priv;
-	int ret;
+	int ret, table_size;
 	unsigned long flags;
 
 	priv = devm_kzalloc(&pdev->dev, sizeof(struct priv), GFP_KERNEL);
@@ -90,7 +90,30 @@ static int mcb_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (ret < 0)
 		goto out_mcb_bus;
 
-	dev_dbg(&pdev->dev, "Found %d cells\n", ret);
+	table_size = ret;
+
+	if (table_size < CHAM_HEADER_SIZE) {
+		/* Release the previous resources */
+		devm_iounmap(&pdev->dev, priv->base);
+		devm_release_mem_region(&pdev->dev, priv->mapbase, CHAM_HEADER_SIZE);
+
+		/* Then, allocate it again with the actual chameleon table size */
+		res = devm_request_mem_region(&pdev->dev, priv->mapbase,
+						table_size,
+						KBUILD_MODNAME);
+		if (!res) {
+			dev_err(&pdev->dev, "Failed to request PCI memory\n");
+			ret = -EBUSY;
+			goto out_mcb_bus;
+		}
+
+		priv->base = devm_ioremap(&pdev->dev, priv->mapbase, table_size);
+		if (!priv->base) {
+			dev_err(&pdev->dev, "Cannot ioremap\n");
+			ret = -ENOMEM;
+			goto out_mcb_bus;
+		}
+	}
 
 	mcb_bus_add_devices(priv->bus);
 
Signature · 1 line
2.39.2
author maintainer
#137 lore raw

PATCH 137/292 sched: Fix KCSAN noinstr violation

In reply to Greg Kroah-Hartman

From: Josh Poimboeuf <[email protected]>

[ Upstream commit e0b081d17a9f4e5c0cbb0e5fbeb1abe3de0f7e4e ]

With KCSAN enabled, end_of_stack() can get out-of-lined. Force it inline.

Fixes the following warnings:

vmlinux.o: warning: objtool: check_stackleak_irqoff+0x2b: call to end_of_stack() leaves .noinstr.text section

---
 include/linux/sched/task_stack.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/sched/task_stack.h +1 -1
index 5e799a47431e8..f158b025c1750 100644
@@ -23,7 +23,7 @@ static __always_inline void *task_stack_page(const struct task_struct *task)
 
 #define setup_thread_stack(new,old)	do { } while(0)
 
-static inline unsigned long *end_of_stack(const struct task_struct *task)
+static __always_inline unsigned long *end_of_stack(const struct task_struct *task)
 {
 #ifdef CONFIG_STACK_GROWSUP
 	return (unsigned long *)((unsigned long)task->stack + THREAD_SIZE) - 1;
Signature · 1 line
2.39.2
author maintainer
#138 lore raw

PATCH 138/292 lkdtm/stackleak: Fix noinstr violation

In reply to Greg Kroah-Hartman

From: Josh Poimboeuf <[email protected]>

[ Upstream commit f571da059f86fd9d432aea32c9c7e5aaa53245d8 ]

Fixes the following warning:

vmlinux.o: warning: objtool: check_stackleak_irqoff+0x2b6: call to _printk() leaves .noinstr.text section

---
 drivers/misc/lkdtm/stackleak.c | 6 ++++++
 1 file changed, 6 insertions(+)
drivers/misc/lkdtm/stackleak.c +6 -0
index 025b133297a6b..f1d0221609138 100644
@@ -43,12 +43,14 @@ static void noinstr check_stackleak_irqoff(void)
 	 * STACK_END_MAGIC, and in either casee something is seriously wrong.
 	 */
 	if (current_sp < task_stack_low || current_sp >= task_stack_high) {
+		instrumentation_begin();
 		pr_err("FAIL: current_stack_pointer (0x%lx) outside of task stack bounds [0x%lx..0x%lx]\n",
 		       current_sp, task_stack_low, task_stack_high - 1);
 		test_failed = true;
 		goto out;
 	}
 	if (lowest_sp < task_stack_low || lowest_sp >= task_stack_high) {
+		instrumentation_begin();
 		pr_err("FAIL: current->lowest_stack (0x%lx) outside of task stack bounds [0x%lx..0x%lx]\n",
 		       lowest_sp, task_stack_low, task_stack_high - 1);
 		test_failed = true;
@@ -86,11 +88,14 @@ static void noinstr check_stackleak_irqoff(void)
 		if (*(unsigned long *)poison_low == STACKLEAK_POISON)
 			continue;
 
+		instrumentation_begin();
 		pr_err("FAIL: non-poison value %lu bytes below poison boundary: 0x%lx\n",
 		       poison_high - poison_low, *(unsigned long *)poison_low);
 		test_failed = true;
+		goto out;
 	}
 
+	instrumentation_begin();
 	pr_info("stackleak stack usage:\n"
 		"  high offset: %lu bytes\n"
 		"  current:     %lu bytes\n"
@@ -113,6 +118,7 @@ static void noinstr check_stackleak_irqoff(void)
 	} else {
 		pr_info("OK: the rest of the thread stack is properly erased\n");
 	}
+	instrumentation_end();
 }
 
 static void lkdtm_STACKLEAK_ERASING(void)
Signature · 1 line
2.39.2
author maintainer
#139 lore raw

PATCH 139/292 recordmcount: Fix memory leaks in the uwrite function

In reply to Greg Kroah-Hartman

From: Hao Zeng <[email protected]>

[ Upstream commit fa359d068574d29e7d2f0fdd0ebe4c6a12b5cfb9 ]

Common realloc mistake: 'file_append' nulled but not freed upon failure

---
 scripts/recordmcount.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
scripts/recordmcount.c +5 -1
index cce12e1971d85..ec692af8ce9eb 100644
@@ -102,6 +102,7 @@ static ssize_t uwrite(void const *const buf, size_t const count)
 {
 	size_t cnt = count;
 	off_t idx = 0;
+	void *p = NULL;
 
 	file_updated = 1;
 
@@ -109,7 +110,10 @@ static ssize_t uwrite(void const *const buf, size_t const count)
 		off_t aoffset = (file_ptr + count) - file_end;
 
 		if (aoffset > file_append_size) {
-			file_append = realloc(file_append, aoffset);
+			p = realloc(file_append, aoffset);
+			if (!p)
+				free(file_append);
+			file_append = p;
 			file_append_size = aoffset;
 		}
 		if (!file_append) {
Signature · 1 line
2.39.2
author maintainer
#140 lore raw

PATCH 140/292 soundwire: dmi-quirks: add remapping for Intel Rooks County NUC M15

In reply to Greg Kroah-Hartman

From: Eugene Huang <[email protected]>

[ Upstream commit 01b33e284ca28cc977bdcfb23be2c719f2139175 ]

Same DSDT problem as the HP Omen 16-k0005TX, except rt1316 amp is on link2.

---
 drivers/soundwire/dmi-quirks.c | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)
drivers/soundwire/dmi-quirks.c +25 -0
index 7969881f126dc..58ea013fa918a 100644
@@ -73,6 +73,23 @@ static const struct adr_remap hp_omen_16[] = {
 	{}
 };
 
+/*
+ * Intel NUC M15 LAPRC510 and LAPRC710
+ */
+static const struct adr_remap intel_rooks_county[] = {
+	/* rt711-sdca on link0 */
+	{
+		0x000020025d071100ull,
+		0x000030025d071101ull
+	},
+	/* rt1316-sdca on link2 */
+	{
+		0x000120025d071100ull,
+		0x000230025d131601ull
+	},
+	{}
+};
+
 static const struct dmi_system_id adr_remap_quirk_table[] = {
 	/* TGL devices */
 	{
@@ -98,6 +115,14 @@ static const struct dmi_system_id adr_remap_quirk_table[] = {
 		},
 		.driver_data = (void *)intel_tgl_bios,
 	},
+	{
+		/* quirk used for NUC15 'Rooks County' LAPRC510 and LAPRC710 skews */
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Intel(R) Client Systems"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "LAPRC"),
+		},
+		.driver_data = (void *)intel_rooks_county,
+	},
 	{
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc"),
Signature · 1 line
2.39.2
author maintainer
#141 lore raw

PATCH 141/292 phy: st: miphy28lp: use _poll_timeout functions for waits

In reply to Greg Kroah-Hartman

From: Alain Volmat <[email protected]>

[ Upstream commit e3be4dd2c8d8aabfd2c3127d0e2e5754d3ae82d6 ]

This commit introduces _poll_timeout functions usage instead of wait loops waiting for a status bit.

---
 drivers/phy/st/phy-miphy28lp.c | 42 ++++++++--------------------------
 1 file changed, 10 insertions(+), 32 deletions(-)
drivers/phy/st/phy-miphy28lp.c +10 -32
index 068160a34f5cc..e30305b77f0d1 100644
@@ -9,6 +9,7 @@
 
 #include <linux/platform_device.h>
 #include <linux/io.h>
+#include <linux/iopoll.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/of.h>
@@ -484,19 +485,11 @@ static inline void miphy28lp_pcie_config_gen(struct miphy28lp_phy *miphy_phy)
 
 static inline int miphy28lp_wait_compensation(struct miphy28lp_phy *miphy_phy)
 {
-	unsigned long finish = jiffies + 5 * HZ;
 	u8 val;
 
 	/* Waiting for Compensation to complete */
-	do {
-		val = readb_relaxed(miphy_phy->base + MIPHY_COMP_FSM_6);
-
-		if (time_after_eq(jiffies, finish))
-			return -EBUSY;
-		cpu_relax();
-	} while (!(val & COMP_DONE));
-
-	return 0;
+	return readb_relaxed_poll_timeout(miphy_phy->base + MIPHY_COMP_FSM_6,
+					  val, val & COMP_DONE, 1, 5 * USEC_PER_SEC);
 }
 
 
@@ -805,7 +798,6 @@ static inline void miphy28lp_configure_usb3(struct miphy28lp_phy *miphy_phy)
 
 static inline int miphy_is_ready(struct miphy28lp_phy *miphy_phy)
 {
-	unsigned long finish = jiffies + 5 * HZ;
 	u8 mask = HFC_PLL | HFC_RDY;
 	u8 val;
 
@@ -816,21 +808,14 @@ static inline int miphy_is_ready(struct miphy28lp_phy *miphy_phy)
 	if (miphy_phy->type == PHY_TYPE_SATA)
 		mask |= PHY_RDY;
 
-	do {
-		val = readb_relaxed(miphy_phy->base + MIPHY_STATUS_1);
-		if ((val & mask) != mask)
-			cpu_relax();
-		else
-			return 0;
-	} while (!time_after_eq(jiffies, finish));
-
-	return -EBUSY;
+	return readb_relaxed_poll_timeout(miphy_phy->base + MIPHY_STATUS_1,
+					  val, (val & mask) == mask, 1,
+					  5 * USEC_PER_SEC);
 }
 
 static int miphy_osc_is_ready(struct miphy28lp_phy *miphy_phy)
 {
 	struct miphy28lp_dev *miphy_dev = miphy_phy->phydev;
-	unsigned long finish = jiffies + 5 * HZ;
 	u32 val;
 
 	if (!miphy_phy->osc_rdy)
@@ -839,17 +824,10 @@ static int miphy_osc_is_ready(struct miphy28lp_phy *miphy_phy)
 	if (!miphy_phy->syscfg_reg[SYSCFG_STATUS])
 		return -EINVAL;
 
-	do {
-		regmap_read(miphy_dev->regmap,
-				miphy_phy->syscfg_reg[SYSCFG_STATUS], &val);
-
-		if ((val & MIPHY_OSC_RDY) != MIPHY_OSC_RDY)
-			cpu_relax();
-		else
-			return 0;
-	} while (!time_after_eq(jiffies, finish));
-
-	return -EBUSY;
+	return regmap_read_poll_timeout(miphy_dev->regmap,
+					miphy_phy->syscfg_reg[SYSCFG_STATUS],
+					val, val & MIPHY_OSC_RDY, 1,
+					5 * USEC_PER_SEC);
 }
 
 static int miphy28lp_get_resource_byname(struct device_node *child,
Signature · 1 line
2.39.2
author maintainer
#142 lore raw

PATCH 142/292 soundwire: qcom: gracefully handle too many ports in DT

In reply to Greg Kroah-Hartman

From: Krzysztof Kozlowski <[email protected]>

[ Upstream commit 2367e0ecb498764e95cfda691ff0828f7d25f9a4 ]

There are two issues related to the number of ports coming from Devicetree when exceeding in total QCOM_SDW_MAX_PORTS. Both lead to incorrect memory accesses:
1. With DTS having too big value of input or output ports, the driver, when copying port parameters from local/stack arrays into 'pconfig' array in 'struct qcom_swrm_ctrl', will iterate over their sizes.

2. If DTS also has too many parameters for these ports (e.g. qcom,ports-sinterval-low), the driver will overflow buffers on the stack when reading these properties from DTS.

Add a sanity check so incorrect DTS will not cause kernel memory corruption.

---
 drivers/soundwire/qcom.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/soundwire/qcom.c +3 -0
index 866026185c669..21c50972047f5 100644
@@ -1209,6 +1209,9 @@ static int qcom_swrm_get_port_config(struct qcom_swrm_ctrl *ctrl)
 	ctrl->num_dout_ports = val;
 
 	nports = ctrl->num_dout_ports + ctrl->num_din_ports;
+	if (nports > QCOM_SDW_MAX_PORTS)
+		return -EINVAL;
+
 	/* Valid port numbers are from 1-14, so mask out port 0 explicitly */
 	set_bit(0, &ctrl->dout_port_mask);
 	set_bit(0, &ctrl->din_port_mask);
Signature · 1 line
2.39.2
author maintainer
#143 lore raw

PATCH 143/292 soundwire: bus: Fix unbalanced pm_runtime_put() causing usage count underflow

In reply to Greg Kroah-Hartman

From: Richard Fitzgerald <[email protected]>

[ Upstream commit e9537962519e88969f5f69cd0571eb4f6984403c ]

This reverts commit
443a98e649b4 ("soundwire: bus: use pm_runtime_resume_and_get()")

Change calls to pm_runtime_resume_and_get() back to pm_runtime_get_sync(). This fixes a usage count underrun caused by doing a pm_runtime_put() even though pm_runtime_resume_and_get() returned an error.

The three affected functions ignore -EACCES error from trying to get pm_runtime, and carry on, including a put at the end of the function. But pm_runtime_resume_and_get() does not increment the usage count if it returns an error. So in the -EACCES case you must not call pm_runtime_put().

The documentation for pm_runtime_get_sync() says:
"Consider using pm_runtime_resume_and_get() ... as this is likely to result in cleaner code."

In this case I don't think it results in cleaner code because the pm_runtime_put() at the end of the function would have to be conditional on the return value from pm_runtime_resume_and_get() at the top of the function.

pm_runtime_get_sync() doesn't have this problem because it always increments the count, so always needs a put. The code can just flow through and do the pm_runtime_put() unconditionally.

---
 drivers/soundwire/bus.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)
drivers/soundwire/bus.c +10 -5
index 76515c33e639e..4fd221d0cc818 100644
@@ -571,9 +571,11 @@ int sdw_nread(struct sdw_slave *slave, u32 addr, size_t count, u8 *val)
 {
 	int ret;
 
-	ret = pm_runtime_resume_and_get(&slave->dev);
-	if (ret < 0 && ret != -EACCES)
+	ret = pm_runtime_get_sync(&slave->dev);
+	if (ret < 0 && ret != -EACCES) {
+		pm_runtime_put_noidle(&slave->dev);
 		return ret;
+	}
 
 	ret = sdw_nread_no_pm(slave, addr, count, val);
 
@@ -595,9 +597,11 @@ int sdw_nwrite(struct sdw_slave *slave, u32 addr, size_t count, const u8 *val)
 {
 	int ret;
 
-	ret = pm_runtime_resume_and_get(&slave->dev);
-	if (ret < 0 && ret != -EACCES)
+	ret = pm_runtime_get_sync(&slave->dev);
+	if (ret < 0 && ret != -EACCES) {
+		pm_runtime_put_noidle(&slave->dev);
 		return ret;
+	}
 
 	ret = sdw_nwrite_no_pm(slave, addr, count, val);
 
@@ -1565,9 +1569,10 @@ static int sdw_handle_slave_alerts(struct sdw_slave *slave)
 
 	sdw_modify_slave_status(slave, SDW_SLAVE_ALERT);
 
-	ret = pm_runtime_resume_and_get(&slave->dev);
+	ret = pm_runtime_get_sync(&slave->dev);
 	if (ret < 0 && ret != -EACCES) {
 		dev_err(&slave->dev, "Failed to resume device: %d\n", ret);
+		pm_runtime_put_noidle(&slave->dev);
 		return ret;
 	}
 
Signature · 1 line
2.39.2
author maintainer
#144 lore raw

PATCH 144/292 mfd: intel_soc_pmic_chtwc: Add Lenovo Yoga Book X90F to intel_cht_wc_models

In reply to Greg Kroah-Hartman

From: Hans de Goede <[email protected]>

[ Upstream commit ded99b89d25fd73a1d7bd910378e0339fd9d4c4a ]

The Android Lenovo Yoga Book X90F / X90L uses the same charger / fuelgauge setup as the already supported Windows Lenovo Yoga Book X91F/L, add a DMI match for this to intel_cht_wc_models with driver_data set to INTEL_CHT_WC_LENOVO_YOGABOOK1.

When the quirk for the X91F/L was initially added it was written to also apply to the X90F/L but this does not work because the Android version of the Yoga Book uses completely different DMI strings. Also adjust the X91F/L quirk to reflect that it only applies to the X91F/L models.

---
 drivers/mfd/intel_soc_pmic_chtwc.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)
drivers/mfd/intel_soc_pmic_chtwc.c +11 -3
index 9216f0d34206c..a82b7cb661b7b 100644
@@ -159,11 +159,19 @@ static const struct dmi_system_id cht_wc_model_dmi_ids[] = {
 			DMI_MATCH(DMI_PRODUCT_NAME, "Mipad2"),
 		},
 	}, {
-		/* Lenovo Yoga Book X90F / X91F / X91L */
+		/* Lenovo Yoga Book X90F / X90L */
 		.driver_data = (void *)(long)INTEL_CHT_WC_LENOVO_YOGABOOK1,
 		.matches = {
-			/* Non exact match to match all versions */
-			DMI_MATCH(DMI_PRODUCT_NAME, "Lenovo YB1-X9"),
+			DMI_EXACT_MATCH(DMI_SYS_VENDOR, "Intel Corporation"),
+			DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "CHERRYVIEW D1 PLATFORM"),
+			DMI_EXACT_MATCH(DMI_PRODUCT_VERSION, "YETI-11"),
+		},
+	}, {
+		/* Lenovo Yoga Book X91F / X91L */
+		.driver_data = (void *)(long)INTEL_CHT_WC_LENOVO_YOGABOOK1,
+		.matches = {
+			/* Non exact match to match F + L versions */
+			DMI_MATCH(DMI_PRODUCT_NAME, "Lenovo YB1-X91"),
 		},
 	},
 	{ }
Signature · 1 line
2.39.2
author maintainer
#145 lore raw

PATCH 145/292 mfd: dln2: Fix memory leak in dln2_probe()

In reply to Greg Kroah-Hartman

From: Qiang Ning <[email protected]>

[ Upstream commit 96da8f148396329ba769246cb8ceaa35f1ddfc48 ]

When dln2_setup_rx_urbs() in dln2_probe() fails, error out_free forgets to call usb_put_dev() to decrease the refcount of dln2->usb_dev.

Fix this by adding usb_put_dev() in the error handling code of dln2_probe().

---
 drivers/mfd/dln2.c | 1 +
 1 file changed, 1 insertion(+)
drivers/mfd/dln2.c +1 -0
index 6cd0b0c752d6e..c3149729cec2e 100644
@@ -827,6 +827,7 @@ static int dln2_probe(struct usb_interface *interface,
 	dln2_stop_rx_urbs(dln2);
 
 out_free:
+	usb_put_dev(dln2->usb_dev);
 	dln2_free(dln2);
 
 	return ret;
Signature · 1 line
2.39.2
author maintainer
#146 lore raw

PATCH 146/292 mfd: intel-lpss: Add Intel Meteor Lake PCH-S LPSS PCI IDs

In reply to Greg Kroah-Hartman

From: Jarkko Nikula <[email protected]>

[ Upstream commit 72d4a1683741ee578da0e265886e6a7f3d42266c ]

Add Intel Meteor Lake PCH-S also called as Meteor Point-S LPSS PCI IDs.

---
 drivers/mfd/intel-lpss-pci.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
drivers/mfd/intel-lpss-pci.c +15 -0
index dde31c50a6320..699f44ffff0e4 100644
@@ -447,6 +447,21 @@ static const struct pci_device_id intel_lpss_pci_ids[] = {
 	{ PCI_VDEVICE(INTEL, 0x7e79), (kernel_ulong_t)&bxt_i2c_info },
 	{ PCI_VDEVICE(INTEL, 0x7e7a), (kernel_ulong_t)&bxt_i2c_info },
 	{ PCI_VDEVICE(INTEL, 0x7e7b), (kernel_ulong_t)&bxt_i2c_info },
+	/* MTP-S */
+	{ PCI_VDEVICE(INTEL, 0x7f28), (kernel_ulong_t)&bxt_uart_info },
+	{ PCI_VDEVICE(INTEL, 0x7f29), (kernel_ulong_t)&bxt_uart_info },
+	{ PCI_VDEVICE(INTEL, 0x7f2a), (kernel_ulong_t)&tgl_info },
+	{ PCI_VDEVICE(INTEL, 0x7f2b), (kernel_ulong_t)&tgl_info },
+	{ PCI_VDEVICE(INTEL, 0x7f4c), (kernel_ulong_t)&bxt_i2c_info },
+	{ PCI_VDEVICE(INTEL, 0x7f4d), (kernel_ulong_t)&bxt_i2c_info },
+	{ PCI_VDEVICE(INTEL, 0x7f4e), (kernel_ulong_t)&bxt_i2c_info },
+	{ PCI_VDEVICE(INTEL, 0x7f4f), (kernel_ulong_t)&bxt_i2c_info },
+	{ PCI_VDEVICE(INTEL, 0x7f5c), (kernel_ulong_t)&bxt_uart_info },
+	{ PCI_VDEVICE(INTEL, 0x7f5d), (kernel_ulong_t)&bxt_uart_info },
+	{ PCI_VDEVICE(INTEL, 0x7f5e), (kernel_ulong_t)&tgl_info },
+	{ PCI_VDEVICE(INTEL, 0x7f5f), (kernel_ulong_t)&tgl_info },
+	{ PCI_VDEVICE(INTEL, 0x7f7a), (kernel_ulong_t)&bxt_i2c_info },
+	{ PCI_VDEVICE(INTEL, 0x7f7b), (kernel_ulong_t)&bxt_i2c_info },
 	/* LKF */
 	{ PCI_VDEVICE(INTEL, 0x98a8), (kernel_ulong_t)&bxt_uart_info },
 	{ PCI_VDEVICE(INTEL, 0x98a9), (kernel_ulong_t)&bxt_uart_info },
Signature · 1 line
2.39.2
author maintainer
#147 lore raw

PATCH 147/292 parisc: Replace regular spinlock with spin_trylock on panic path

In reply to Greg Kroah-Hartman

From: Guilherme G. Piccoli <[email protected]>

[ Upstream commit 829632dae8321787525ee37dc4828bbe6edafdae ]

The panic notifiers' callbacks execute in an atomic context, with interrupts/preemption disabled, and all CPUs not running the panic function are off, so it's very dangerous to wait on a regular spinlock, there's a risk of deadlock.

Refactor the panic notifier of parisc/power driver to make use of spin_trylock - for that, we've added a second version of the soft-power function. Also, some comments were reorganized and trailing white spaces, useless header inclusion and blank lines were removed.

---
 arch/parisc/include/asm/pdc.h |  1 +
 arch/parisc/kernel/firmware.c | 27 +++++++++++++++++++++++----
 drivers/parisc/power.c        | 16 ++++++++++------
 3 files changed, 34 insertions(+), 10 deletions(-)
arch/parisc/include/asm/pdc.h +1 -0
index fcbcf9a96c111..77622558bf651 100644
@@ -80,6 +80,7 @@ int pdc_do_firm_test_reset(unsigned long ftc_bitmap);
 int pdc_do_reset(void);
 int pdc_soft_power_info(unsigned long *power_reg);
 int pdc_soft_power_button(int sw_control);
+int pdc_soft_power_button_panic(int sw_control);
 void pdc_io_reset(void);
 void pdc_io_reset_devices(void);
 int pdc_iodc_getc(void);
arch/parisc/kernel/firmware.c +23 -4
index bd325f2b5349e..3e051a973e9b2 100644
@@ -1232,15 +1232,18 @@ int __init pdc_soft_power_info(unsigned long *power_reg)
 }
 
 /*
- * pdc_soft_power_button - Control the soft power button behaviour
- * @sw_control: 0 for hardware control, 1 for software control 
+ * pdc_soft_power_button{_panic} - Control the soft power button behaviour
+ * @sw_control: 0 for hardware control, 1 for software control
  *
  *
  * This PDC function places the soft power button under software or
  * hardware control.
- * Under software control the OS may control to when to allow to shut 
- * down the system. Under hardware control pressing the power button 
+ * Under software control the OS may control to when to allow to shut
+ * down the system. Under hardware control pressing the power button
  * powers off the system immediately.
+ *
+ * The _panic version relies on spin_trylock to prevent deadlock
+ * on panic path.
  */
 int pdc_soft_power_button(int sw_control)
 {
@@ -1254,6 +1257,22 @@ int pdc_soft_power_button(int sw_control)
 	return retval;
 }
 
+int pdc_soft_power_button_panic(int sw_control)
+{
+	int retval;
+	unsigned long flags;
+
+	if (!spin_trylock_irqsave(&pdc_lock, flags)) {
+		pr_emerg("Couldn't enable soft power button\n");
+		return -EBUSY; /* ignored by the panic notifier */
+	}
+
+	retval = mem_pdc_call(PDC_SOFT_POWER, PDC_SOFT_POWER_ENABLE, __pa(pdc_result), sw_control);
+	spin_unlock_irqrestore(&pdc_lock, flags);
+
+	return retval;
+}
+
 /*
  * pdc_io_reset - Hack to avoid overlapping range registers of Bridges devices.
  * Primarily a problem on T600 (which parisc-linux doesn't support) but
drivers/parisc/power.c +10 -6
index 456776bd8ee66..6f5e5f0230d39 100644
@@ -37,7 +37,6 @@
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
-#include <linux/notifier.h>
 #include <linux/panic_notifier.h>
 #include <linux/reboot.h>
 #include <linux/sched/signal.h>
@@ -175,16 +174,21 @@ static void powerfail_interrupt(int code, void *x)
 
 
 
-/* parisc_panic_event() is called by the panic handler.
- * As soon as a panic occurs, our tasklets above will not be
- * executed any longer. This function then re-enables the 
- * soft-power switch and allows the user to switch off the system
+/*
+ * parisc_panic_event() is called by the panic handler.
+ *
+ * As soon as a panic occurs, our tasklets above will not
+ * be executed any longer. This function then re-enables
+ * the soft-power switch and allows the user to switch off
+ * the system. We rely in pdc_soft_power_button_panic()
+ * since this version spin_trylocks (instead of regular
+ * spinlock), preventing deadlocks on panic path.
  */
 static int parisc_panic_event(struct notifier_block *this,
 		unsigned long event, void *ptr)
 {
 	/* re-enable the soft-power switch */
-	pdc_soft_power_button(0);
+	pdc_soft_power_button_panic(0);
 	return NOTIFY_DONE;
 }
 
Signature · 1 line
2.39.2
author maintainer
#148 lore raw

PATCH 148/292 platform/x86: Move existing HP drivers to a new hp subdir

In reply to Greg Kroah-Hartman

From: Jorge Lopez <[email protected]>

[ Upstream commit 6e9b8992b122cb12688bd259fc99e67d1be234eb ]

The purpose of this patch is to provide a central location where all HP related drivers are found. HP drivers will recide under drivers/platform/x86/hp directory.

Introduce changes to Kconfig file to list all HP driver under "HP X86 Platform Specific Device Drivers" menu option. Additional changes include update MAINTAINERS file to indicate hp related drivers new path.

Stable-dep-of: decab2825c3e ("platform/x86: hp-wmi: add micmute to hp_wmi_keymap struct")

---
 MAINTAINERS                                |  4 +-
 drivers/platform/x86/Kconfig               | 43 +--------------
 drivers/platform/x86/Makefile              |  4 +-
 drivers/platform/x86/hp/Kconfig            | 63 ++++++++++++++++++++++
 drivers/platform/x86/hp/Makefile           | 10 ++++
 drivers/platform/x86/{ => hp}/hp-wmi.c     |  0
 drivers/platform/x86/{ => hp}/hp_accel.c   |  2 +-
 drivers/platform/x86/{ => hp}/tc1100-wmi.c |  0
 8 files changed, 78 insertions(+), 48 deletions(-)
 create mode 100644 drivers/platform/x86/hp/Kconfig
 create mode 100644 drivers/platform/x86/hp/Makefile
 rename drivers/platform/x86/{ => hp}/hp-wmi.c (100%)
 rename drivers/platform/x86/{ => hp}/hp_accel.c (99%)
 rename drivers/platform/x86/{ => hp}/tc1100-wmi.c (100%)
MAINTAINERS +2 -2
index 350d7e3ba94f9..301b9ba6af79f 100644
@@ -9362,7 +9362,7 @@ F:	drivers/net/wireless/intersil/hostap/
 HP COMPAQ TC1100 TABLET WMI EXTRAS DRIVER
 L:	[email protected]
 S:	Orphan
-F:	drivers/platform/x86/tc1100-wmi.c
+F:	drivers/platform/x86/hp/tc1100-wmi.c
 
 HPET:	High Precision Event Timers driver
 M:	Clemens Ladisch <[email protected]>
@@ -11858,7 +11858,7 @@ M:	Eric Piel <[email protected]>
 S:	Maintained
 F:	Documentation/misc-devices/lis3lv02d.rst
 F:	drivers/misc/lis3lv02d/
-F:	drivers/platform/x86/hp_accel.c
+F:	drivers/platform/x86/hp/hp_accel.c
 
 LIST KUNIT TEST
 M:	David Gow <[email protected]>
drivers/platform/x86/Kconfig +1 -42

55 lines not on this page · raw message

drivers/platform/x86/Makefile +1 -3
index 5a428caa654a7..1d3d1b02541b9 100644
@@ -55,9 +55,7 @@ obj-$(CONFIG_FUJITSU_TABLET)	+= fujitsu-tablet.o
 obj-$(CONFIG_GPD_POCKET_FAN)	+= gpd-pocket-fan.o
 
 # Hewlett Packard
-obj-$(CONFIG_HP_ACCEL)		+= hp_accel.o
-obj-$(CONFIG_HP_WMI)		+= hp-wmi.o
-obj-$(CONFIG_TC1100_WMI)	+= tc1100-wmi.o
+obj-$(CONFIG_X86_PLATFORM_DRIVERS_HP)	+= hp/
 
 # Hewlett Packard Enterprise
 obj-$(CONFIG_UV_SYSFS)       += uv_sysfs.o
drivers/platform/x86/hp/Kconfig new file +63 -0

63 lines not on this page · raw message

drivers/platform/x86/hp/Makefile new file +10 -0
new file mode 100644
index 0000000000000..db1eed4cd7c7d
@@ -0,0 +1,10 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Makefile for linux/drivers/platform/x86/hp
+# HP x86 Platform-Specific Drivers
+#
+
+# Hewlett Packard
+obj-$(CONFIG_HP_ACCEL)		+= hp_accel.o
+obj-$(CONFIG_HP_WMI)		+= hp-wmi.o
+obj-$(CONFIG_TC1100_WMI)	+= tc1100-wmi.o
drivers/platform/x86/hp/hp-wmi.c renamed from drivers/platform/x86/hp-wmi.c +0 -0
similarity index 100%
rename from drivers/platform/x86/hp-wmi.c
rename to drivers/platform/x86/hp/hp-wmi.c
drivers/platform/x86/hp/hp_accel.c renamed from drivers/platform/x86/hp_accel.c +1 -1
similarity index 99%
rename from drivers/platform/x86/hp_accel.c
rename to drivers/platform/x86/hp/hp_accel.c
index e9f852f7c27ff..6477591747cfd 100644
@@ -26,7 +26,7 @@
 #include <linux/acpi.h>
 #include <linux/i8042.h>
 #include <linux/serio.h>
-#include "../../misc/lis3lv02d/lis3lv02d.h"
+#include "../../../misc/lis3lv02d/lis3lv02d.h"
 
 /* Delayed LEDs infrastructure ------------------------------------ */
 
drivers/platform/x86/hp/tc1100-wmi.c renamed from drivers/platform/x86/tc1100-wmi.c +0 -0
similarity index 100%
rename from drivers/platform/x86/tc1100-wmi.c
rename to drivers/platform/x86/hp/tc1100-wmi.c
Signature · 1 line
2.39.2
author maintainer
#149 lore raw

PATCH 149/292 platform/x86: hp-wmi: add micmute to hp_wmi_keymap struct

In reply to Greg Kroah-Hartman

From: Fae <[email protected]>

[ Upstream commit decab2825c3ef9b154c6f76bce40872ffb41c36f ]

Fixes micmute key of HP Envy X360 ey0xxx.

---
 drivers/platform/x86/hp/hp-wmi.c | 1 +
 1 file changed, 1 insertion(+)
drivers/platform/x86/hp/hp-wmi.c +1 -0
index 4a3851332ef2c..94af7d398a1bf 100644
@@ -211,6 +211,7 @@ struct bios_rfkill2_state {
 static const struct key_entry hp_wmi_keymap[] = {
 	{ KE_KEY, 0x02,    { KEY_BRIGHTNESSUP } },
 	{ KE_KEY, 0x03,    { KEY_BRIGHTNESSDOWN } },
+	{ KE_KEY, 0x270,   { KEY_MICMUTE } },
 	{ KE_KEY, 0x20e6,  { KEY_PROG1 } },
 	{ KE_KEY, 0x20e8,  { KEY_MEDIA } },
 	{ KE_KEY, 0x2142,  { KEY_MEDIA } },
Signature · 1 line
2.39.2
author maintainer
#150 lore raw

PATCH 150/292 drm/amdgpu: drop gfx_v11_0_cp_ecc_error_irq_funcs

In reply to Greg Kroah-Hartman

From: Horatio Zhang <[email protected]>

[ Upstream commit 720b47229a5b24061d1c2e29ddb6043a59178d79 ]

The gfx.cp_ecc_error_irq is retired in gfx11. In gfx_v11_0_hw_fini still use amdgpu_irq_put to disable this interrupt, which caused the call trace in this function.

[  102.873958] Call Trace:
[  102.873959]  <TASK>
[  102.873961]  gfx_v11_0_hw_fini+0x23/0x1e0 [amdgpu]
[  102.874019]  gfx_v11_0_suspend+0xe/0x20 [amdgpu]
[  102.874072]  amdgpu_device_ip_suspend_phase2+0x240/0x460 [amdgpu]
[  102.874122]  amdgpu_device_ip_suspend+0x3d/0x80 [amdgpu]
[  102.874172]  amdgpu_device_pre_asic_reset+0xd9/0x490 [amdgpu]
[  102.874223]  amdgpu_device_gpu_recover.cold+0x548/0xce6 [amdgpu]
[  102.874321]  amdgpu_debugfs_reset_work+0x4c/0x70 [amdgpu]
[  102.874375]  process_one_work+0x21f/0x3f0
[  102.874377]  worker_thread+0x200/0x3e0
[  102.874378]  ? process_one_work+0x3f0/0x3f0
[  102.874379]  kthread+0xfd/0x130
[  102.874380]  ? kthread_complete_and_exit+0x20/0x20
[  102.874381]  ret_from_fork+0x22/0x30

v2:
- Handle umc and gfx ras cases in separated patch
- Retired the gfx_v11_0_cp_ecc_error_irq_funcs in gfx11

v3:
- Improve the subject and code comments
- Add judgment on gfx11 in the function of amdgpu_gfx_ras_late_init

v4:
- Drop the define of CP_ME1_PIPE_INST_ADDR_INTERVAL and SET_ECC_ME_PIPE_STATE which using in gfx_v11_0_set_cp_ecc_error_state
- Check cp_ecc_error_irq.funcs rather than ip version for a more sustainable life

v5:
- Simplify judgment conditions

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +5 -3
index 3380daf42da8a..b803e785d3aff 100644
@@ -683,9 +683,11 @@ int amdgpu_gfx_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *r
 		if (r)
 			return r;
 
-		r = amdgpu_irq_get(adev, &adev->gfx.cp_ecc_error_irq, 0);
-		if (r)
-			goto late_fini;
+		if (adev->gfx.cp_ecc_error_irq.funcs) {
+			r = amdgpu_irq_get(adev, &adev->gfx.cp_ecc_error_irq, 0);
+			if (r)
+				goto late_fini;
+		}
 	} else {
 		amdgpu_ras_feature_enable_on_boot(adev, ras_block, 0);
 	}
Signature · 1 line
2.39.2
author maintainer
#151 lore raw

PATCH 151/292 xfrm: dont check the default policy if the policy allows the packet

In reply to Greg Kroah-Hartman

From: Sabrina Dubroca <[email protected]>

[ Upstream commit 430cac487400494c19a8b85299e979bb07b4671f ]

The current code doesn't let a simple "allow" policy counteract a default policy blocking all incoming packets:

ip x p setdefault in block
ip x p a src 192.168.2.1/32 dst 192.168.2.2/32 dir in action allow

At this stage, we have an allow policy (with or without transforms) for this packet. It doesn't matter what the default policy says, since the policy we looked up lets the packet through. The case of a blocking policy is already handled separately, so we can remove this check.

Fixes 2d151d39073a ("xfrm: Add possibility to set the default to block if we have no policy")
---
 net/xfrm/xfrm_policy.c | 6 ------
 1 file changed, 6 deletions(-)
net/xfrm/xfrm_policy.c +0 -6
index 7f49dab3b6b59..bea48a73a7313 100644
@@ -3637,12 +3637,6 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 		}
 		xfrm_nr = ti;
 
-		if (net->xfrm.policy_default[dir] == XFRM_USERPOLICY_BLOCK &&
-		    !xfrm_nr) {
-			XFRM_INC_STATS(net, LINUX_MIB_XFRMINNOSTATES);
-			goto reject;
-		}
-
 		if (npols > 1) {
 			xfrm_tmpl_sort(stp, tpp, xfrm_nr, family);
 			tpp = stp;
Signature · 1 line
2.39.2
author maintainer
#152 lore raw

PATCH 152/292 Revert "Fix XFRM-I support for nested ESP tunnels"

In reply to Greg Kroah-Hartman

From: Martin Willi <[email protected]>

[ Upstream commit 5fc46f94219d1d103ffb5f0832be9da674d85a73 ]

This reverts commit b0355dbbf13c0052931dd14c38c789efed64d3de.

The reverted commit clears the secpath on packets received via xfrm interfaces to support nested IPsec tunnels. This breaks Netfilter policy matching using xt_policy in the FORWARD chain, as the secpath is missing during forwarding. Additionally, Benedict Wong reports that it breaks Transport-in-Tunnel mode.

Fix this regression by reverting the commit until we have a better approach for nested IPsec tunnels.

Fixes b0355dbbf13c ("Fix XFRM-I support for nested ESP tunnels")
---
 net/xfrm/xfrm_interface.c | 54 +++------------------------------------
 net/xfrm/xfrm_policy.c    |  3 ---
 2 files changed, 4 insertions(+), 53 deletions(-)
net/xfrm/xfrm_interface.c +4 -50
index 94a3609548b11..5a67b120c4dbd 100644
@@ -310,52 +310,6 @@ static void xfrmi_scrub_packet(struct sk_buff *skb, bool xnet)
 	skb->mark = 0;
 }
 
-static int xfrmi_input(struct sk_buff *skb, int nexthdr, __be32 spi,
-		       int encap_type, unsigned short family)
-{
-	struct sec_path *sp;
-
-	sp = skb_sec_path(skb);
-	if (sp && (sp->len || sp->olen) &&
-	    !xfrm_policy_check(NULL, XFRM_POLICY_IN, skb, family))
-		goto discard;
-
-	XFRM_SPI_SKB_CB(skb)->family = family;
-	if (family == AF_INET) {
-		XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct iphdr, daddr);
-		XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4 = NULL;
-	} else {
-		XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct ipv6hdr, daddr);
-		XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip6 = NULL;
-	}
-
-	return xfrm_input(skb, nexthdr, spi, encap_type);
-discard:
-	kfree_skb(skb);
-	return 0;
-}
-
-static int xfrmi4_rcv(struct sk_buff *skb)
-{
-	return xfrmi_input(skb, ip_hdr(skb)->protocol, 0, 0, AF_INET);
-}
-
-static int xfrmi6_rcv(struct sk_buff *skb)
-{
-	return xfrmi_input(skb, skb_network_header(skb)[IP6CB(skb)->nhoff],
-			   0, 0, AF_INET6);
-}
-
-static int xfrmi4_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
-{
-	return xfrmi_input(skb, nexthdr, spi, encap_type, AF_INET);
-}
-
-static int xfrmi6_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
-{
-	return xfrmi_input(skb, nexthdr, spi, encap_type, AF_INET6);
-}
-
 static int xfrmi_rcv_cb(struct sk_buff *skb, int err)
 {
 	const struct xfrm_mode *inner_mode;
@@ -983,8 +937,8 @@ static struct pernet_operations xfrmi_net_ops = {
 };
 
 static struct xfrm6_protocol xfrmi_esp6_protocol __read_mostly = {
-	.handler	=	xfrmi6_rcv,
-	.input_handler	=	xfrmi6_input,
+	.handler	=	xfrm6_rcv,
+	.input_handler	=	xfrm_input,
 	.cb_handler	=	xfrmi_rcv_cb,
 	.err_handler	=	xfrmi6_err,
 	.priority	=	10,
@@ -1034,8 +988,8 @@ static struct xfrm6_tunnel xfrmi_ip6ip_handler __read_mostly = {
 #endif
 
 static struct xfrm4_protocol xfrmi_esp4_protocol __read_mostly = {
-	.handler	=	xfrmi4_rcv,
-	.input_handler	=	xfrmi4_input,
+	.handler	=	xfrm4_rcv,
+	.input_handler	=	xfrm_input,
 	.cb_handler	=	xfrmi_rcv_cb,
 	.err_handler	=	xfrmi4_err,
 	.priority	=	10,
net/xfrm/xfrm_policy.c +0 -3
index bea48a73a7313..bc04cb83215f9 100644
@@ -3664,9 +3664,6 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 			goto reject;
 		}
 
-		if (if_id)
-			secpath_reset(skb);
-
 		xfrm_pols_put(pols, npols);
 		return 1;
 	}
Signature · 1 line
2.39.2
author maintainer
#153 lore raw

PATCH 153/292 drm/msm/dp: unregister audio driver during unbind

In reply to Greg Kroah-Hartman

From: Srinivas Kandagatla <[email protected]>

[ Upstream commit 85c636284cb63b7740b4ae98881ace92158068d3 ]

while binding the code always registers a audio driver, however there is no corresponding unregistration done in unbind. This leads to multiple redundant audio platform devices if dp_display_bind and dp_display_unbind happens multiple times during startup. On X13s platform this resulted in 6 to 9 audio codec device instead of just 3 codec devices for 3 dp ports.

Fix this by unregistering codecs on unbind.

Fixes d13e36d7d222 ("drm/msm/dp: add audio support for Display Port on MSM")

Patchwork: https://patchwork.freedesktop.org/patch/533324/

---
 drivers/gpu/drm/msm/dp/dp_audio.c   | 12 ++++++++++++
 drivers/gpu/drm/msm/dp/dp_audio.h   |  2 ++
 drivers/gpu/drm/msm/dp/dp_display.c |  1 +
 3 files changed, 15 insertions(+)
drivers/gpu/drm/msm/dp/dp_audio.c +12 -0
index 6666783e1468e..1245c7aa49df8 100644
@@ -593,6 +593,18 @@ static struct hdmi_codec_pdata codec_data = {
 	.i2s = 1,
 };
 
+void dp_unregister_audio_driver(struct device *dev, struct dp_audio *dp_audio)
+{
+	struct dp_audio_private *audio_priv;
+
+	audio_priv = container_of(dp_audio, struct dp_audio_private, dp_audio);
+
+	if (audio_priv->audio_pdev) {
+		platform_device_unregister(audio_priv->audio_pdev);
+		audio_priv->audio_pdev = NULL;
+	}
+}
+
 int dp_register_audio_driver(struct device *dev,
 		struct dp_audio *dp_audio)
 {
drivers/gpu/drm/msm/dp/dp_audio.h +2 -0
index 84e5f4a5d26ba..4ab78880af829 100644
@@ -53,6 +53,8 @@ struct dp_audio *dp_audio_get(struct platform_device *pdev,
 int dp_register_audio_driver(struct device *dev,
 		struct dp_audio *dp_audio);
 
+void dp_unregister_audio_driver(struct device *dev, struct dp_audio *dp_audio);
+
 /**
  * dp_audio_put()
  *
drivers/gpu/drm/msm/dp/dp_display.c +1 -0
index c9d9b384ddd03..57b82e5d0ab12 100644
@@ -323,6 +323,7 @@ static void dp_display_unbind(struct device *dev, struct device *master,
 	kthread_stop(dp->ev_tsk);
 
 	dp_power_client_deinit(dp->power);
+	dp_unregister_audio_driver(dev, dp->audio);
 	dp_aux_unregister(dp->aux);
 	dp->drm_dev = NULL;
 	dp->aux->drm_dev = NULL;
Signature · 1 line
2.39.2
author maintainer
#154 lore raw

PATCH 154/292 drm/msm/dpu: Assign missing writeback log_mask

In reply to Greg Kroah-Hartman

From: Marijn Suijten <[email protected]>

[ Upstream commit a432fc31f03db2546a48bcf5dd69ca28ceb732bf ]

The WB debug log mask ended up never being assigned, leading to writes to this block to never be logged even if the mask is enabled in dpu_hw_util_log_mask via debugfs.

Fixes 84a33d0fd921 ("drm/msm/dpu: add dpu_hw_wb abstraction for writeback blocks")

Patchwork: https://patchwork.freedesktop.org/patch/533860/

---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c | 1 +
 1 file changed, 1 insertion(+)
drivers/gpu/drm/msm/disp/dpu1/dpu_hw_wb.c +1 -0
index 2d28afdf860ef..a3e413d277175 100644
@@ -61,6 +61,7 @@ static const struct dpu_wb_cfg *_wb_offset(enum dpu_wb wb,
 	for (i = 0; i < m->wb_count; i++) {
 		if (wb == m->wb[i].id) {
 			b->blk_addr = addr + m->wb[i].base;
+			b->log_mask = DPU_DBG_MASK_WB;
 			return &m->wb[i];
 		}
 	}
Signature · 1 line
2.39.2
author maintainer
#155 lore raw

PATCH 155/292 drm/msm/dpu: Move non-MDP_TOP INTF_INTR offsets out of hwio header

In reply to Greg Kroah-Hartman

From: Marijn Suijten <[email protected]>

[ Upstream commit e9d9ce5462fecdeefec87953de71df4d025cbc72 ]

These offsets do not fall under the MDP TOP block and do not fit the comment right above. Move them to dpu_hw_interrupts.c next to the repsective MDP_INTF_x_OFF interrupt block offsets.

Fixes 25fdd5933e4c ("drm/msm: Add SDM845 DPU support")

Patchwork: https://patchwork.freedesktop.org/patch/534203/

---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c | 5 ++++-
 drivers/gpu/drm/msm/disp/dpu1/dpu_hwio.h          | 3 ---
 2 files changed, 4 insertions(+), 4 deletions(-)
drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c +4 -1
index cf1b6d84c18a3..75e1b89c9eacf 100644
@@ -15,7 +15,7 @@
 
 /*
  * Register offsets in MDSS register file for the interrupt registers
- * w.r.t. to the MDP base
+ * w.r.t. the MDP base
  */
 #define MDP_SSPP_TOP0_OFF		0x0
 #define MDP_INTF_0_OFF			0x6A000
@@ -24,6 +24,9 @@
 #define MDP_INTF_3_OFF			0x6B800
 #define MDP_INTF_4_OFF			0x6C000
 #define MDP_INTF_5_OFF			0x6C800
+#define INTF_INTR_EN			0x1c0
+#define INTF_INTR_STATUS		0x1c4
+#define INTF_INTR_CLEAR			0x1c8
 #define MDP_AD4_0_OFF			0x7C000
 #define MDP_AD4_1_OFF			0x7D000
 #define MDP_AD4_INTR_EN_OFF		0x41c
drivers/gpu/drm/msm/disp/dpu1/dpu_hwio.h +0 -3
index c8156ed4b7fb8..93081e82c6d74 100644
@@ -20,9 +20,6 @@
 #define HIST_INTR_EN                    0x01c
 #define HIST_INTR_STATUS                0x020
 #define HIST_INTR_CLEAR                 0x024
-#define INTF_INTR_EN                    0x1C0
-#define INTF_INTR_STATUS                0x1C4
-#define INTF_INTR_CLEAR                 0x1C8
 #define SPLIT_DISPLAY_EN                0x2F4
 #define SPLIT_DISPLAY_UPPER_PIPE_CTRL   0x2F8
 #define DSPP_IGC_COLOR0_RAM_LUTN        0x300
Signature · 1 line
2.39.2
author maintainer
#156 lore raw

PATCH 156/292 drm/msm/dpu: Remove duplicate register defines from INTF

In reply to Greg Kroah-Hartman

From: Marijn Suijten <[email protected]>

[ Upstream commit 202c044203ac5860e3025169105368d99f9bc6a2 ]

The INTF_FRAME_LINE_COUNT_EN, INTF_FRAME_COUNT and INTF_LINE_COUNT registers are already defined higher up, in the right place when sorted numerically.

Fixes 25fdd5933e4c ("drm/msm: Add SDM845 DPU support")

Patchwork: https://patchwork.freedesktop.org/patch/534231/

---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c | 5 -----
 1 file changed, 5 deletions(-)
drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c +0 -5
index 7ce66bf3f4c8d..b2a94b9a3e987 100644
@@ -56,11 +56,6 @@
 #define   INTF_TPG_RGB_MAPPING          0x11C
 #define   INTF_PROG_FETCH_START         0x170
 #define   INTF_PROG_ROT_START           0x174
-
-#define   INTF_FRAME_LINE_COUNT_EN      0x0A8
-#define   INTF_FRAME_COUNT              0x0AC
-#define   INTF_LINE_COUNT               0x0B0
-
 #define   INTF_MUX                      0x25C
 
 #define INTF_CFG_ACTIVE_H_EN	BIT(29)
Signature · 1 line
2.39.2
author maintainer
#157 lore raw

PATCH 157/292 dt-bindings: display/msm: dsi-controller-main: Document qcom, master-dsi and qcom, sync-dual-dsi

In reply to Greg Kroah-Hartman

From: Jianhua Lu <[email protected]>

[ Upstream commit ca29699a57ecee6084a4056f5bfd6f11dd359a71 ]

This fixes warning:
sm8250-xiaomi-elish-csot.dtb: dsi@ae94000: Unevaluated properties are not allowed ('qcom,master-dsi', 'qcom,sync-dual-dsi' were unexpected)

Fixes 4dbe55c97741 ("dt-bindings: msm: dsi: add yaml schemas for DSI bindings")

Patchwork: https://patchwork.freedesktop.org/patch/534306/

---
 .../bindings/display/msm/dsi-controller-main.yaml    | 12 ++++++++++++
 1 file changed, 12 insertions(+)
Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml +12 -0
index 6c5b4783812ae..2fa1759e74d95 100644
@@ -65,6 +65,18 @@ properties:
       Indicates if the DSI controller is driving a panel which needs
       2 DSI links.
 
+  qcom,master-dsi:
+    type: boolean
+    description: |
+      Indicates if the DSI controller is the master DSI controller when
+      qcom,dual-dsi-mode enabled.
+
+  qcom,sync-dual-dsi:
+    type: boolean
+    description: |
+      Indicates if the DSI controller needs to sync the other DSI controller
+      with MIPI DCS commands when qcom,dual-dsi-mode enabled.
+
   assigned-clocks:
     maxItems: 2
     description: |
Signature · 1 line
2.39.2
author maintainer
#158 lore raw

PATCH 158/292 platform: Provide a remove callback that returns no value

In reply to Greg Kroah-Hartman

From: Uwe Kleine-König <[email protected]>

[ Upstream commit 5c5a7680e67ba6fbbb5f4d79fa41485450c1985c ]

struct platform_driver::remove returning an integer made driver authors expect that returning an error code was proper error handling. However the driver core ignores the error and continues to remove the device because there is nothing the core could do anyhow and reentering the remove callback again is only calling for trouble.

So this is an source for errors typically yielding resource leaks in the error path.

As there are too many platform drivers to neatly convert them all to return void in a single go, do it in several steps after this patch:

a) Convert all drivers to implement .remove_new() returning void instead

    of .remove() returning int;
 b) Change struct platform_driver::remove() to return void and so make
    it identical to .remove_new();

c) Change all drivers back to .remove() now with the better prototype;
d) drop struct platform_driver::remove_new().

While this touches all drivers eventually twice, steps a) and c) can be done one driver after another and so reduces coordination efforts immensely and simplifies review.

Stable-dep-of: 17955aba7877 ("ASoC: fsl_micfil: Fix error handler with pm_runtime_enable")

---
 drivers/base/platform.c         |  4 +++-
 include/linux/platform_device.h | 11 +++++++++++
 2 files changed, 14 insertions(+), 1 deletion(-)
drivers/base/platform.c +3 -1
index 51bb2289865c7..3a06c214ca1c6 100644
@@ -1416,7 +1416,9 @@ static void platform_remove(struct device *_dev)
 	struct platform_driver *drv = to_platform_driver(_dev->driver);
 	struct platform_device *dev = to_platform_device(_dev);
 
-	if (drv->remove) {
+	if (drv->remove_new) {
+		drv->remove_new(dev);
+	} else if (drv->remove) {
 		int ret = drv->remove(dev);
 
 		if (ret)
include/linux/platform_device.h +11 -0
index b0d5a253156ec..b845fd83f429b 100644
@@ -207,7 +207,18 @@ extern void platform_device_put(struct platform_device *pdev);
 
 struct platform_driver {
 	int (*probe)(struct platform_device *);
+
+	/*
+	 * Traditionally the remove callback returned an int which however is
+	 * ignored by the driver core. This led to wrong expectations by driver
+	 * authors who thought returning an error code was a valid error
+	 * handling strategy. To convert to a callback returning void, new
+	 * drivers should implement .remove_new() until the conversion it done
+	 * that eventually makes .remove() return void.
+	 */
 	int (*remove)(struct platform_device *);
+	void (*remove_new)(struct platform_device *);
+
 	void (*shutdown)(struct platform_device *);
 	int (*suspend)(struct platform_device *, pm_message_t state);
 	int (*resume)(struct platform_device *);
Signature · 1 line
2.39.2
author maintainer
#159 lore raw

PATCH 159/292 ASoC: fsl_micfil: Fix error handler with pm_runtime_enable

In reply to Greg Kroah-Hartman

From: Shengjiu Wang <[email protected]>

[ Upstream commit 17955aba7877a4494d8093ae5498e19469b01d57 ]

There is error message when defer probe happens:

fsl-micfil-dai 30ca0000.micfil: Unbalanced pm_runtime_enable!

Fix the error handler with pm_runtime_enable and add fsl_micfil_remove() for pm_runtime_disable.

Fixes 47a70e6fc9a8 ("ASoC: Add MICFIL SoC Digital Audio Interface driver.")
---
 sound/soc/fsl/fsl_micfil.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)
sound/soc/fsl/fsl_micfil.c +13 -1
index 4b8fe9b8be407..3a03f49452fa3 100644
@@ -712,7 +712,7 @@ static int fsl_micfil_probe(struct platform_device *pdev)
 	ret = devm_snd_dmaengine_pcm_register(&pdev->dev, NULL, 0);
 	if (ret) {
 		dev_err(&pdev->dev, "failed to pcm register\n");
-		return ret;
+		goto err_pm_disable;
 	}
 
 	fsl_micfil_dai.capture.formats = micfil->soc->formats;
@@ -722,9 +722,20 @@ static int fsl_micfil_probe(struct platform_device *pdev)
 	if (ret) {
 		dev_err(&pdev->dev, "failed to register component %s\n",
 			fsl_micfil_component.name);
+		goto err_pm_disable;
 	}
 
 	return ret;
+
+err_pm_disable:
+	pm_runtime_disable(&pdev->dev);
+
+	return ret;
+}
+
+static void fsl_micfil_remove(struct platform_device *pdev)
+{
+	pm_runtime_disable(&pdev->dev);
 }
 
 static int __maybe_unused fsl_micfil_runtime_suspend(struct device *dev)
@@ -785,6 +796,7 @@ static const struct dev_pm_ops fsl_micfil_pm_ops = {
 
 static struct platform_driver fsl_micfil_driver = {
 	.probe = fsl_micfil_probe,
+	.remove_new = fsl_micfil_remove,
 	.driver = {
 		.name = "fsl-micfil-dai",
 		.pm = &fsl_micfil_pm_ops,
Signature · 1 line
2.39.2
author maintainer
#160 lore raw

PATCH 160/292 cpupower: Make TSC read per CPU for Mperf monitor

In reply to Greg Kroah-Hartman

From: Wyes Karny <[email protected]>

[ Upstream commit c2adb1877b76fc81ae041e1db1a6ed2078c6746b ]

System-wide TSC read could cause a drift in C0 percentage calculation. Because if first TSC is read and then one by one mperf is read for all cpus, this introduces drift between mperf reading of later CPUs and TSC reading. To lower this drift read TSC per CPU and also just after mperf read. This technique improves C0 percentage calculation in Mperf monitor.

Before fix: (System 100% busy)

              | Mperf              || RAPL        || Idle_Stats
 PKG|CORE| CPU| C0   | Cx   | Freq  || pack | core  || POLL | C1   | C2
   0|   0|   0| 87.15| 12.85|  2695||168659003|3970468||  0.00|  0.00| 0.00
   0|   0| 256| 84.62| 15.38|  2695||168659003|3970468||  0.00|  0.00| 0.00
   0|   1|   1| 87.15| 12.85|  2695||168659003|3970468||  0.00|  0.00| 0.00
   0|   1| 257| 84.08| 15.92|  2695||168659003|3970468||  0.00|  0.00| 0.00
   0|   2|   2| 86.61| 13.39|  2695||168659003|3970468||  0.00|  0.00| 0.00
   0|   2| 258| 83.26| 16.74|  2695||168659003|3970468||  0.00|  0.00| 0.00
   0|   3|   3| 86.61| 13.39|  2695||168659003|3970468||  0.00|  0.00| 0.00
   0|   3| 259| 83.60| 16.40|  2695||168659003|3970468||  0.00|  0.00| 0.00
   0|   4|   4| 86.33| 13.67|  2695||168659003|3970468||  0.00|  0.00| 0.00
   0|   4| 260| 83.33| 16.67|  2695||168659003|3970468||  0.00|  0.00| 0.00
   0|   5|   5| 86.06| 13.94|  2695||168659003|3970468||  0.00|  0.00| 0.00
   0|   5| 261| 83.05| 16.95|  2695||168659003|3970468||  0.00|  0.00| 0.00
   0|   6|   6| 85.51| 14.49|  2695||168659003|3970468||  0.00|  0.00| 0.00

After fix: (System 100% busy)

             | Mperf              || RAPL        || Idle_Stats
 PKG|CORE| CPU| C0   | Cx   | Freq  || pack | core  || POLL | C1   | C2
   0|   0|   0| 98.03|  1.97|  2415||163295480|3811189||  0.00|  0.00| 0.00
   0|   0| 256| 98.50|  1.50|  2394||163295480|3811189||  0.00|  0.00| 0.00
   0|   1|   1| 99.99|  0.01|  2401||163295480|3811189||  0.00|  0.00| 0.00
   0|   1| 257| 99.99|  0.01|  2375||163295480|3811189||  0.00|  0.00| 0.00
   0|   2|   2| 99.99|  0.01|  2401||163295480|3811189||  0.00|  0.00| 0.00
   0|   2| 258|100.00|  0.00|  2401||163295480|3811189||  0.00|  0.00| 0.00
   0|   3|   3|100.00|  0.00|  2401||163295480|3811189||  0.00|  0.00| 0.00
   0|   3| 259| 99.99|  0.01|  2435||163295480|3811189||  0.00|  0.00| 0.00
   0|   4|   4|100.00|  0.00|  2401||163295480|3811189||  0.00|  0.00| 0.00
   0|   4| 260|100.00|  0.00|  2435||163295480|3811189||  0.00|  0.00| 0.00
   0|   5|   5| 99.99|  0.01|  2401||163295480|3811189||  0.00|  0.00| 0.00
   0|   5| 261|100.00|  0.00|  2435||163295480|3811189||  0.00|  0.00| 0.00
   0|   6|   6|100.00|  0.00|  2401||163295480|3811189||  0.00|  0.00| 0.00
   0|   6| 262|100.00|  0.00|  2435||163295480|3811189||  0.00|  0.00| 0.00
Cc Thomas Renninger <[email protected]>
Cc Shuah Khan <[email protected]>
Cc Dominik Brodowski <[email protected]>
Fixes 7fe2f6399a84 ("cpupowerutils - cpufrequtils extended with quite some features")
---
 .../utils/idle_monitor/mperf_monitor.c        | 31 +++++++++----------
 1 file changed, 14 insertions(+), 17 deletions(-)
tools/power/cpupower/utils/idle_monitor/mperf_monitor.c +14 -17
index e7d48cb563c0e..ae6af354a81db 100644
@@ -70,8 +70,8 @@ static int max_freq_mode;
  */
 static unsigned long max_frequency;
 
-static unsigned long long tsc_at_measure_start;
-static unsigned long long tsc_at_measure_end;
+static unsigned long long *tsc_at_measure_start;
+static unsigned long long *tsc_at_measure_end;
 static unsigned long long *mperf_previous_count;
 static unsigned long long *aperf_previous_count;
 static unsigned long long *mperf_current_count;
@@ -169,7 +169,7 @@ static int mperf_get_count_percent(unsigned int id, double *percent,
 	aperf_diff = aperf_current_count[cpu] - aperf_previous_count[cpu];
 
 	if (max_freq_mode == MAX_FREQ_TSC_REF) {
-		tsc_diff = tsc_at_measure_end - tsc_at_measure_start;
+		tsc_diff = tsc_at_measure_end[cpu] - tsc_at_measure_start[cpu];
 		*percent = 100.0 * mperf_diff / tsc_diff;
 		dprint("%s: TSC Ref - mperf_diff: %llu, tsc_diff: %llu\n",
 		       mperf_cstates[id].name, mperf_diff, tsc_diff);
@@ -206,7 +206,7 @@ static int mperf_get_count_freq(unsigned int id, unsigned long long *count,
 
 	if (max_freq_mode == MAX_FREQ_TSC_REF) {
 		/* Calculate max_freq from TSC count */
-		tsc_diff = tsc_at_measure_end - tsc_at_measure_start;
+		tsc_diff = tsc_at_measure_end[cpu] - tsc_at_measure_start[cpu];
 		time_diff = timespec_diff_us(time_start, time_end);
 		max_frequency = tsc_diff / time_diff;
 	}
@@ -225,33 +225,27 @@ static int mperf_get_count_freq(unsigned int id, unsigned long long *count,
 static int mperf_start(void)
 {
 	int cpu;
-	unsigned long long dbg;
 
 	clock_gettime(CLOCK_REALTIME, &time_start);
-	mperf_get_tsc(&tsc_at_measure_start);
 
-	for (cpu = 0; cpu < cpu_count; cpu++)
+	for (cpu = 0; cpu < cpu_count; cpu++) {
+		mperf_get_tsc(&tsc_at_measure_start[cpu]);
 		mperf_init_stats(cpu);
+	}
 
-	mperf_get_tsc(&dbg);
-	dprint("TSC diff: %llu\n", dbg - tsc_at_measure_start);
 	return 0;
 }
 
 static int mperf_stop(void)
 {
-	unsigned long long dbg;
 	int cpu;
 
-	for (cpu = 0; cpu < cpu_count; cpu++)
+	for (cpu = 0; cpu < cpu_count; cpu++) {
 		mperf_measure_stats(cpu);
+		mperf_get_tsc(&tsc_at_measure_end[cpu]);
+	}
 
-	mperf_get_tsc(&tsc_at_measure_end);
 	clock_gettime(CLOCK_REALTIME, &time_end);
-
-	mperf_get_tsc(&dbg);
-	dprint("TSC diff: %llu\n", dbg - tsc_at_measure_end);
-
 	return 0;
 }
 
@@ -353,7 +347,8 @@ struct cpuidle_monitor *mperf_register(void)
 	aperf_previous_count = calloc(cpu_count, sizeof(unsigned long long));
 	mperf_current_count = calloc(cpu_count, sizeof(unsigned long long));
 	aperf_current_count = calloc(cpu_count, sizeof(unsigned long long));
-
+	tsc_at_measure_start = calloc(cpu_count, sizeof(unsigned long long));
+	tsc_at_measure_end = calloc(cpu_count, sizeof(unsigned long long));
 	mperf_monitor.name_len = strlen(mperf_monitor.name);
 	return &mperf_monitor;
 }
@@ -364,6 +359,8 @@ void mperf_unregister(void)
 	free(aperf_previous_count);
 	free(mperf_current_count);
 	free(aperf_current_count);
+	free(tsc_at_measure_start);
+	free(tsc_at_measure_end);
 	free(is_valid);
 }
 
Signature · 1 line
2.39.2
author maintainer
#161 lore raw

PATCH 161/292 xfrm: Reject optional tunnel/BEET mode templates in outbound policies

In reply to Greg Kroah-Hartman

From: Tobias Brunner <[email protected]>

[ Upstream commit 3d776e31c841ba2f69895d2255a49320bec7cea6 ]

xfrm_state_find() uses encap_family of the current template with the passed local and remote addresses to find a matching state. If an optional tunnel or BEET mode template is skipped in a mixed-family scenario, there could be a mismatch causing an out-of-bounds read as the addresses were not replaced to match the family of the next template.

While there are theoretical use cases for optional templates in outbound policies, the only practical one is to skip IPComp states in inbound policies if uncompressed packets are received that are handled by an implicitly created IPIP state instead.

Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 net/xfrm/xfrm_user.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)
net/xfrm/xfrm_user.c +9 -5
index 83f35ecacf24f..2d68a173b2273 100644
@@ -1743,7 +1743,7 @@ static void copy_templates(struct xfrm_policy *xp, struct xfrm_user_tmpl *ut,
 }
 
 static int validate_tmpl(int nr, struct xfrm_user_tmpl *ut, u16 family,
-			 struct netlink_ext_ack *extack)
+			 int dir, struct netlink_ext_ack *extack)
 {
 	u16 prev_family;
 	int i;
@@ -1769,6 +1769,10 @@ static int validate_tmpl(int nr, struct xfrm_user_tmpl *ut, u16 family,
 		switch (ut[i].mode) {
 		case XFRM_MODE_TUNNEL:
 		case XFRM_MODE_BEET:
+			if (ut[i].optional && dir == XFRM_POLICY_OUT) {
+				NL_SET_ERR_MSG(extack, "Mode in optional template not allowed in outbound policy");
+				return -EINVAL;
+			}
 			break;
 		default:
 			if (ut[i].family != prev_family) {
@@ -1806,7 +1810,7 @@ static int validate_tmpl(int nr, struct xfrm_user_tmpl *ut, u16 family,
 }
 
 static int copy_from_user_tmpl(struct xfrm_policy *pol, struct nlattr **attrs,
-			       struct netlink_ext_ack *extack)
+			       int dir, struct netlink_ext_ack *extack)
 {
 	struct nlattr *rt = attrs[XFRMA_TMPL];
 
@@ -1817,7 +1821,7 @@ static int copy_from_user_tmpl(struct xfrm_policy *pol, struct nlattr **attrs,
 		int nr = nla_len(rt) / sizeof(*utmpl);
 		int err;
 
-		err = validate_tmpl(nr, utmpl, pol->family, extack);
+		err = validate_tmpl(nr, utmpl, pol->family, dir, extack);
 		if (err)
 			return err;
 
@@ -1894,7 +1898,7 @@ static struct xfrm_policy *xfrm_policy_construct(struct net *net,
 	if (err)
 		goto error;
 
-	if (!(err = copy_from_user_tmpl(xp, attrs, extack)))
+	if (!(err = copy_from_user_tmpl(xp, attrs, p->dir, extack)))
 		err = copy_from_user_sec_ctx(xp, attrs);
 	if (err)
 		goto error;
@@ -3443,7 +3447,7 @@ static struct xfrm_policy *xfrm_compile_policy(struct sock *sk, int opt,
 		return NULL;
 
 	nr = ((len - sizeof(*p)) / sizeof(*ut));
-	if (validate_tmpl(nr, ut, p->sel.family, NULL))
+	if (validate_tmpl(nr, ut, p->sel.family, p->dir, NULL))
 		return NULL;
 
 	if (p->dir > XFRM_POLICY_OUT)
Signature · 1 line
2.39.2
author maintainer
#162 lore raw

PATCH 162/292 af_key: Reject optional tunnel/BEET mode templates in outbound policies

In reply to Greg Kroah-Hartman

From: Tobias Brunner <[email protected]>

[ Upstream commit cf3128a7aca55b2eefb68281d44749c683bdc96f ]

xfrm_state_find() uses encap_family of the current template with the passed local and remote addresses to find a matching state. If an optional tunnel or BEET mode template is skipped in a mixed-family scenario, there could be a mismatch causing an out-of-bounds read as the addresses were not replaced to match the family of the next template.

While there are theoretical use cases for optional templates in outbound policies, the only practical one is to skip IPComp states in inbound policies if uncompressed packets are received that are handled by an implicitly created IPIP state instead.

Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 net/key/af_key.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
net/key/af_key.c +8 -4
index 95edcbedf6ef2..8c21de50eadf8 100644
@@ -1940,7 +1940,8 @@ static u32 gen_reqid(struct net *net)
 }
 
 static int
-parse_ipsecrequest(struct xfrm_policy *xp, struct sadb_x_ipsecrequest *rq)
+parse_ipsecrequest(struct xfrm_policy *xp, struct sadb_x_policy *pol,
+		   struct sadb_x_ipsecrequest *rq)
 {
 	struct net *net = xp_net(xp);
 	struct xfrm_tmpl *t = xp->xfrm_vec + xp->xfrm_nr;
@@ -1958,9 +1959,12 @@ parse_ipsecrequest(struct xfrm_policy *xp, struct sadb_x_ipsecrequest *rq)
 	if ((mode = pfkey_mode_to_xfrm(rq->sadb_x_ipsecrequest_mode)) < 0)
 		return -EINVAL;
 	t->mode = mode;
-	if (rq->sadb_x_ipsecrequest_level == IPSEC_LEVEL_USE)
+	if (rq->sadb_x_ipsecrequest_level == IPSEC_LEVEL_USE) {
+		if ((mode == XFRM_MODE_TUNNEL || mode == XFRM_MODE_BEET) &&
+		    pol->sadb_x_policy_dir == IPSEC_DIR_OUTBOUND)
+			return -EINVAL;
 		t->optional = 1;
-	else if (rq->sadb_x_ipsecrequest_level == IPSEC_LEVEL_UNIQUE) {
+	} else if (rq->sadb_x_ipsecrequest_level == IPSEC_LEVEL_UNIQUE) {
 		t->reqid = rq->sadb_x_ipsecrequest_reqid;
 		if (t->reqid > IPSEC_MANUAL_REQID_MAX)
 			t->reqid = 0;
@@ -2002,7 +2006,7 @@ parse_ipsecrequests(struct xfrm_policy *xp, struct sadb_x_policy *pol)
 		    rq->sadb_x_ipsecrequest_len < sizeof(*rq))
 			return -EINVAL;
 
-		if ((err = parse_ipsecrequest(xp, rq)) < 0)
+		if ((err = parse_ipsecrequest(xp, pol, rq)) < 0)
 			return err;
 		len -= rq->sadb_x_ipsecrequest_len;
 		rq = (void*)((u8*)rq + rq->sadb_x_ipsecrequest_len);
Signature · 1 line
2.39.2
author maintainer
#163 lore raw

PATCH 163/292 drm/msm: Fix submit error-path leaks

In reply to Greg Kroah-Hartman

From: Rob Clark <[email protected]>

[ Upstream commit 68dc6c2d5eec45515855cce99256162f45651a0b ]

For errors after msm_submitqueue_get(), we need to drop the submitqueue reference. Additionally after get_unused_fd() we need to drop the fd. The ordering for dropping the queue lock and put_unused_fd() is not important, so just move this all into out_post_unlock.

v2: Only drop queue ref if submit doesn't take it
v3: Fix unitialized submit ref in error path
v4: IS_ERR_OR_NULL()

Reported-by [email protected]
Fixes f0de40a131d9 drm/msm: ("Reorder lock vs submit alloc")

Patchwork: https://patchwork.freedesktop.org/patch/536073/

---
 drivers/gpu/drm/msm/msm_gem_submit.c | 25 ++++++++++++++++++-------
 1 file changed, 18 insertions(+), 7 deletions(-)
drivers/gpu/drm/msm/msm_gem_submit.c +18 -7
index d8c9d184190bb..d6162561141c5 100644
@@ -709,7 +709,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	struct msm_drm_private *priv = dev->dev_private;
 	struct drm_msm_gem_submit *args = data;
 	struct msm_file_private *ctx = file->driver_priv;
-	struct msm_gem_submit *submit;
+	struct msm_gem_submit *submit = NULL;
 	struct msm_gpu *gpu = priv->gpu;
 	struct msm_gpu_submitqueue *queue;
 	struct msm_ringbuffer *ring;
@@ -756,13 +756,15 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 		out_fence_fd = get_unused_fd_flags(O_CLOEXEC);
 		if (out_fence_fd < 0) {
 			ret = out_fence_fd;
-			return ret;
+			goto out_post_unlock;
 		}
 	}
 
 	submit = submit_create(dev, gpu, queue, args->nr_bos, args->nr_cmds);
-	if (IS_ERR(submit))
-		return PTR_ERR(submit);
+	if (IS_ERR(submit)) {
+		ret = PTR_ERR(submit);
+		goto out_post_unlock;
+	}
 
 	trace_msm_gpu_submit(pid_nr(submit->pid), ring->id, submit->ident,
 		args->nr_bos, args->nr_cmds);
@@ -945,11 +947,20 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	if (has_ww_ticket)
 		ww_acquire_fini(&submit->ticket);
 out_unlock:
-	if (ret && (out_fence_fd >= 0))
-		put_unused_fd(out_fence_fd);
 	mutex_unlock(&queue->lock);
 out_post_unlock:
-	msm_gem_submit_put(submit);
+	if (ret && (out_fence_fd >= 0))
+		put_unused_fd(out_fence_fd);
+
+	if (!IS_ERR_OR_NULL(submit)) {
+		msm_gem_submit_put(submit);
+	} else {
+		/*
+		 * If the submit hasn't yet taken ownership of the queue
+		 * then we need to drop the reference ourself:
+		 */
+		msm_submitqueue_put(queue);
+	}
 	if (!IS_ERR_OR_NULL(post_deps)) {
 		for (i = 0; i < args->nr_out_syncobjs; ++i) {
 			kfree(post_deps[i].chain);
Signature · 1 line
2.39.2
author maintainer
#164 lore raw

PATCH 164/292 selftests: seg6: disable DAD on IPv6 router cfg for srv6_end_dt4_l3vpn_test

In reply to Greg Kroah-Hartman

From: Andrea Mayer <[email protected]>

[ Upstream commit 21a933c79a33add3612808f3be4ad65dd4dc026b ]

The srv6_end_dt4_l3vpn_test instantiates a virtual network consisting of several routers (rt-1, rt-2) and hosts.
When the IPv6 addresses of rt-{1,2} routers are configured, the Deduplicate Address Detection (DAD) kicks in when enabled in the Linux distros running the selftests. DAD is used to check whether an IPv6 address is already assigned in a network. Such a mechanism consists of sending an ICMPv6 Echo Request and waiting for a reply.
As the DAD process could take too long to complete, it may cause the failing of some tests carried out by the srv6_end_dt4_l3vpn_test script.

To make the srv6_end_dt4_l3vpn_test more robust, we disable DAD on routers since we configure the virtual network manually and do not need any address deduplication mechanism at all.

Fixes 2195444e09b4 ("selftests: add selftest for the SRv6 End.DT4 behavior")
---
 tools/testing/selftests/net/srv6_end_dt4_l3vpn_test.sh | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
tools/testing/selftests/net/srv6_end_dt4_l3vpn_test.sh +5 -1
index 1003119773e5d..37f08d582d2fe 100755
@@ -232,10 +232,14 @@ setup_rt_networking()
 	local nsname=rt-${rt}
 
 	ip netns add ${nsname}
+
+	ip netns exec ${nsname} sysctl -wq net.ipv6.conf.all.accept_dad=0
+	ip netns exec ${nsname} sysctl -wq net.ipv6.conf.default.accept_dad=0
+
 	ip link set veth-rt-${rt} netns ${nsname}
 	ip -netns ${nsname} link set veth-rt-${rt} name veth0
 
-	ip -netns ${nsname} addr add ${IPv6_RT_NETWORK}::${rt}/64 dev veth0
+	ip -netns ${nsname} addr add ${IPv6_RT_NETWORK}::${rt}/64 dev veth0 nodad
 	ip -netns ${nsname} link set veth0 up
 	ip -netns ${nsname} link set lo up
 
Signature · 1 line
2.39.2
author maintainer
#165 lore raw

PATCH 165/292 selftets: seg6: disable rp_filter by default in srv6_end_dt4_l3vpn_test

In reply to Greg Kroah-Hartman

From: Andrea Mayer <[email protected]>

[ Upstream commit f97b8401e0deb46ad1e4245c21f651f64f55aaa6 ]

On some distributions, the rp_filter is automatically set (=1) by default on a netdev basis (also on VRFs).
In an SRv6 End.DT4 behavior, decapsulated IPv4 packets are routed using the table associated with the VRF bound to that tunnel. During lookup operations, the rp_filter can lead to packet loss when activated on the VRF.
Therefore, we chose to make this selftest more robust by explicitly disabling the rp_filter during tests (as it is automatically set by some Linux distributions).

Fixes 2195444e09b4 ("selftests: add selftest for the SRv6 End.DT4 behavior")
---
 .../testing/selftests/net/srv6_end_dt4_l3vpn_test.sh  | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)
tools/testing/selftests/net/srv6_end_dt4_l3vpn_test.sh +6 -5
index 37f08d582d2fe..f962823628119 100755
@@ -258,6 +258,12 @@ setup_hs()
 
 	# set the networking for the host
 	ip netns add ${hsname}
+
+	# disable the rp_filter otherwise the kernel gets confused about how
+	# to route decap ipv4 packets.
+	ip netns exec ${rtname} sysctl -wq net.ipv4.conf.all.rp_filter=0
+	ip netns exec ${rtname} sysctl -wq net.ipv4.conf.default.rp_filter=0
+
 	ip -netns ${hsname} link add veth0 type veth peer name ${rtveth}
 	ip -netns ${hsname} link set ${rtveth} netns ${rtname}
 	ip -netns ${hsname} addr add ${IPv4_HS_NETWORK}.${hs}/24 dev veth0
@@ -276,11 +282,6 @@ setup_hs()
 
 	ip netns exec ${rtname} sysctl -wq net.ipv4.conf.${rtveth}.proxy_arp=1
 
-	# disable the rp_filter otherwise the kernel gets confused about how
-	# to route decap ipv4 packets.
-	ip netns exec ${rtname} sysctl -wq net.ipv4.conf.all.rp_filter=0
-	ip netns exec ${rtname} sysctl -wq net.ipv4.conf.${rtveth}.rp_filter=0
-
 	ip netns exec ${rtname} sh -c "echo 1 > /proc/sys/net/vrf/strict_mode"
 }
 
Signature · 1 line
2.39.2
author maintainer
#166 lore raw

PATCH 166/292 net: fec: Better handle pm_runtime_get() failing in .remove()

In reply to Greg Kroah-Hartman

From: Uwe Kleine-König <[email protected]>

[ Upstream commit f816b9829b19394d318e01953aa3b2721bca040d ]

In the (unlikely) event that pm_runtime_get() (disguised as pm_runtime_resume_and_get()) fails, the remove callback returned an error early. The problem with this is that the driver core ignores the error value and continues removing the device. This results in a resource leak. Worse the devm allocated resources are freed and so if a callback of the driver is called later the register mapping is already gone which probably results in a crash.

---
 drivers/net/ethernet/freescale/fec_main.c | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)
drivers/net/ethernet/freescale/fec_main.c +11 -4
index 6f914180f4797..33226a22d8a4a 100644
@@ -4168,9 +4168,11 @@ fec_drv_remove(struct platform_device *pdev)
 	struct device_node *np = pdev->dev.of_node;
 	int ret;
 
-	ret = pm_runtime_resume_and_get(&pdev->dev);
+	ret = pm_runtime_get_sync(&pdev->dev);
 	if (ret < 0)
-		return ret;
+		dev_err(&pdev->dev,
+			"Failed to resume device in remove callback (%pe)\n",
+			ERR_PTR(ret));
 
 	cancel_work_sync(&fep->tx_timeout_work);
 	fec_ptp_stop(pdev);
@@ -4183,8 +4185,13 @@ fec_drv_remove(struct platform_device *pdev)
 		of_phy_deregister_fixed_link(np);
 	of_node_put(fep->phy_node);
 
-	clk_disable_unprepare(fep->clk_ahb);
-	clk_disable_unprepare(fep->clk_ipg);
+	/* After pm_runtime_get_sync() failed, the clks are still off, so skip
+	 * disabling them again.
+	 */
+	if (ret >= 0) {
+		clk_disable_unprepare(fep->clk_ahb);
+		clk_disable_unprepare(fep->clk_ipg);
+	}
 	pm_runtime_put_noidle(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
 
Signature · 1 line
2.39.2
author maintainer
#167 lore raw

PATCH 167/292 net: phy: dp83867: add w/a for packet errors seen with short cables

In reply to Greg Kroah-Hartman

From: Grygorii Strashko <[email protected]>

[ Upstream commit 0b01db274028f5acd207332686ffc92ac77491ac ]

Introduce the W/A for packet errors seen with short cables (<1m) between two DP83867 PHYs.

The W/A recommended by DM requires FFE Equalizer Configuration tuning by writing value 0x0E81 to DSP_FFE_CFG register (0x012C), surrounded by hard and soft resets as follows:

write_reg(0x001F, 0x8000); //hard reset
write_reg(DSP_FFE_CFG, 0x0E81);
write_reg(0x001F, 0x4000); //soft reset

Since DP83867 PHY DM says "Changing this register to 0x0E81, will not affect Long Cable performance.", enable the W/A by default.

Fixes 2a10154abcb7 ("net: phy: dp83867: Add TI dp83867 phy")
---
 drivers/net/phy/dp83867.c | 22 +++++++++++++++++++++-
 1 file changed, 21 insertions(+), 1 deletion(-)
drivers/net/phy/dp83867.c +21 -1
index 7446d5c6c7146..14990f8462ae3 100644
@@ -42,6 +42,7 @@
 #define DP83867_STRAP_STS1	0x006E
 #define DP83867_STRAP_STS2	0x006f
 #define DP83867_RGMIIDCTL	0x0086
+#define DP83867_DSP_FFE_CFG	0x012c
 #define DP83867_RXFCFG		0x0134
 #define DP83867_RXFPMD1	0x0136
 #define DP83867_RXFPMD2	0x0137
@@ -910,8 +911,27 @@ static int dp83867_phy_reset(struct phy_device *phydev)
 
 	usleep_range(10, 20);
 
-	return phy_modify(phydev, MII_DP83867_PHYCTRL,
+	err = phy_modify(phydev, MII_DP83867_PHYCTRL,
 			 DP83867_PHYCR_FORCE_LINK_GOOD, 0);
+	if (err < 0)
+		return err;
+
+	/* Configure the DSP Feedforward Equalizer Configuration register to
+	 * improve short cable (< 1 meter) performance. This will not affect
+	 * long cable performance.
+	 */
+	err = phy_write_mmd(phydev, DP83867_DEVADDR, DP83867_DSP_FFE_CFG,
+			    0x0e81);
+	if (err < 0)
+		return err;
+
+	err = phy_write(phydev, DP83867_CTRL, DP83867_SW_RESTART);
+	if (err < 0)
+		return err;
+
+	usleep_range(10, 20);
+
+	return 0;
 }
 
 static void dp83867_link_change_notify(struct phy_device *phydev)
Signature · 1 line
2.39.2
author maintainer
#168 lore raw

PATCH 168/292 ALSA: firewire-digi00x: prevent potential use after free

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit c0e72058d5e21982e61a29de6b098f7c1f0db498 ]

This code was supposed to return an error code if init_stream() failed, but it instead freed dg00x->rx_stream and returned success. This potentially leads to a use after free.

Fixes 9a08067ec318 ("ALSA: firewire-digi00x: support AMDTP domain")
---
 sound/firewire/digi00x/digi00x-stream.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
sound/firewire/digi00x/digi00x-stream.c +3 -1
index a15f55b0dce37..295163bb8abb6 100644
@@ -259,8 +259,10 @@ int snd_dg00x_stream_init_duplex(struct snd_dg00x *dg00x)
 		return err;
 
 	err = init_stream(dg00x, &dg00x->tx_stream);
-	if (err < 0)
+	if (err < 0) {
 		destroy_stream(dg00x, &dg00x->rx_stream);
+		return err;
+	}
 
 	err = amdtp_domain_init(&dg00x->domain);
 	if (err < 0) {
Signature · 1 line
2.39.2
author maintainer
#169 lore raw

PATCH 169/292 wifi: mt76: connac: fix stats->tx_bytes calculation

In reply to Greg Kroah-Hartman

From: Ryder Lee <[email protected]>

[ Upstream commit c7ab7a29ef5c0779574120d922256ce4651555d3 ]

The stats->tx_bytes shall subtract retry byte from tx byte.

---
 drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h | 2 +-
 drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c  | 3 ++-
 2 files changed, 3 insertions(+), 2 deletions(-)
drivers/net/wireless/mediatek/mt76/mt76_connac2_mac.h +1 -1
index f33171bcd3432..c3b692eac6f65 100644
@@ -163,7 +163,7 @@ enum {
 #define MT_TXS5_MPDU_TX_CNT		GENMASK(31, 23)
 
 #define MT_TXS6_MPDU_FAIL_CNT		GENMASK(31, 23)
-
+#define MT_TXS7_MPDU_RETRY_BYTE		GENMASK(22, 0)
 #define MT_TXS7_MPDU_RETRY_CNT		GENMASK(31, 23)
 
 /* RXD DW1 */
drivers/net/wireless/mediatek/mt76/mt76_connac_mac.c +2 -1
index 19f02b632a204..68511597599e3 100644
@@ -570,7 +570,8 @@ bool mt76_connac2_mac_fill_txs(struct mt76_dev *dev, struct mt76_wcid *wcid,
 	/* PPDU based reporting */
 	if (FIELD_GET(MT_TXS0_TXS_FORMAT, txs) > 1) {
 		stats->tx_bytes +=
-			le32_get_bits(txs_data[5], MT_TXS5_MPDU_TX_BYTE);
+			le32_get_bits(txs_data[5], MT_TXS5_MPDU_TX_BYTE) -
+			le32_get_bits(txs_data[7], MT_TXS7_MPDU_RETRY_BYTE);
 		stats->tx_packets +=
 			le32_get_bits(txs_data[5], MT_TXS5_MPDU_TX_CNT);
 		stats->tx_failed +=
Signature · 1 line
2.39.2
author maintainer
#170 lore raw

PATCH 170/292 ALSA: hda/realtek: Apply HP B&O top speaker profile to Pavilion 15

In reply to Greg Kroah-Hartman

From: Ryan C. Underwood <[email protected]>

[ Upstream commit 92553ee03166ef8fa978e7683f9f4af30c9c4e6b ]

The Pavilion 15 line has B&O top speakers similar to the x360 and applying the same profile produces good sound. Without this, the sound would be tinny and underpowered without either applying model=alc295-hp-x360 or booting another OS first.

Fixes 563785edfcef ("ALSA: hda/realtek - Add quirk entry for HP Pavilion 15")
---
 sound/pci/hda/patch_realtek.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/pci/hda/patch_realtek.c +1 -1
index 172ffc2c332b7..5d78d4ba1c959 100644
@@ -9363,7 +9363,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x103c, 0x802f, "HP Z240", ALC221_FIXUP_HP_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x103c, 0x8077, "HP", ALC256_FIXUP_HP_HEADSET_MIC),
 	SND_PCI_QUIRK(0x103c, 0x8158, "HP", ALC256_FIXUP_HP_HEADSET_MIC),
-	SND_PCI_QUIRK(0x103c, 0x820d, "HP Pavilion 15", ALC269_FIXUP_HP_MUTE_LED_MIC3),
+	SND_PCI_QUIRK(0x103c, 0x820d, "HP Pavilion 15", ALC295_FIXUP_HP_X360),
 	SND_PCI_QUIRK(0x103c, 0x8256, "HP", ALC221_FIXUP_HP_FRONT_MIC),
 	SND_PCI_QUIRK(0x103c, 0x827e, "HP x360", ALC295_FIXUP_HP_X360),
 	SND_PCI_QUIRK(0x103c, 0x827f, "HP x360", ALC269_FIXUP_HP_MUTE_LED_MIC3),
Signature · 1 line
2.39.2
author maintainer
#171 lore raw

PATCH 171/292 sfc: disable RXFCS and RXALL features by default

In reply to Greg Kroah-Hartman

From: Pieter Jansen van Vuuren <[email protected]>

[ Upstream commit 134120b066044399ef59564ff3ba66ab344cfc5b ]

By default we would not want RXFCS and RXALL features enabled as they are mainly intended for debugging purposes. This does not stop users from enabling them later on as needed.

---
 drivers/net/ethernet/sfc/ef100_netdev.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/ethernet/sfc/ef100_netdev.c +3 -1
index ddcc325ed5701..c6b9ba6803c8d 100644
@@ -372,7 +372,9 @@ int ef100_probe_netdev(struct efx_probe_data *probe_data)
 	efx->net_dev = net_dev;
 	SET_NETDEV_DEV(net_dev, &efx->pci_dev->dev);
 
-	net_dev->features |= efx->type->offload_features;
+	/* enable all supported features except rx-fcs and rx-all */
+	net_dev->features |= efx->type->offload_features &
+			     ~(NETIF_F_RXFCS | NETIF_F_RXALL);
 	net_dev->hw_features |= efx->type->offload_features;
 	net_dev->hw_enc_features |= efx->type->offload_features;
 	net_dev->vlan_features |= NETIF_F_HW_CSUM | NETIF_F_SG |
Signature · 1 line
2.39.2
author maintainer
#172 lore raw

PATCH 172/292 vsock: avoid to close connected socket after the timeout

In reply to Greg Kroah-Hartman

From: Zhuang Shengen <[email protected]>

[ Upstream commit 6d4486efe9c69626cab423456169e250a5cd3af5 ]

When client and server establish a connection through vsock, the client send a request to the server to initiate the connection, then start a timer to wait for the server's response. When the server's RESPONSE message arrives, the timer also times out and exits. The server's RESPONSE message is processed first, and the connection is established. However, the client's timer also times out, the original processing logic of the client is to directly set the state of this vsock to CLOSE and return ETIMEDOUT. It will not notify the server when the port is released, causing the server port remain.
when client's vsock_connect timeout,it should check sk state is ESTABLISHED or not. if sk state is ESTABLISHED, it means the connection is established, the client should not set the sk state to CLOSE

Note: I encountered this issue on kernel-4.18, which can be fixed by this patch. Then I checked the latest code in the community and found similar issue.

Fixes d021c344051a ("VSOCK: Introduce VM Sockets")
---
 net/vmw_vsock/af_vsock.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/vmw_vsock/af_vsock.c +1 -1
index 884eca7f6743a..8360c790a8a01 100644
@@ -1427,7 +1427,7 @@ static int vsock_connect(struct socket *sock, struct sockaddr *addr,
 			vsock_transport_cancel_pkt(vsk);
 			vsock_remove_connected(vsk);
 			goto out_wait;
-		} else if (timeout == 0) {
+		} else if ((sk->sk_state != TCP_ESTABLISHED) && (timeout == 0)) {
 			err = -ETIMEDOUT;
 			sk->sk_state = TCP_CLOSE;
 			sock->state = SS_UNCONNECTED;
Signature · 1 line
2.39.2
author maintainer
#173 lore raw

PATCH 173/292 tcp: fix possible sk_priority leak in tcp_v4_send_reset()

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit 1e306ec49a1f206fd2cc89a42fac6e6f592a8cc1 ]

When tcp_v4_send_reset() is called with @sk == NULL, we do not change ctl_sk->sk_priority, which could have been set from a prior invocation.

Change tcp_v4_send_reset() to set sk_priority and sk_mark fields before calling ip_send_unicast_reply().

This means tcp_v4_send_reset() and tcp_v4_send_ack() no longer have to clear ctl_sk->sk_mark after their call to ip_send_unicast_reply().

Fixes f6c0f5d209fa ("tcp: honor SO_PRIORITY in TIME_WAIT state")
Cc Antoine Tenart <[email protected]>
---
 net/ipv4/tcp_ipv4.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
net/ipv4/tcp_ipv4.c +3 -2
index ad0a5f185a694..b37c1bcb15097 100644
@@ -829,6 +829,9 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 				   inet_twsk(sk)->tw_priority : sk->sk_priority;
 		transmit_time = tcp_transmit_time(sk);
 		xfrm_sk_clone_policy(ctl_sk, sk);
+	} else {
+		ctl_sk->sk_mark = 0;
+		ctl_sk->sk_priority = 0;
 	}
 	ip_send_unicast_reply(ctl_sk,
 			      skb, &TCP_SKB_CB(skb)->header.h4.opt,
@@ -836,7 +839,6 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 			      &arg, arg.iov[0].iov_len,
 			      transmit_time);
 
-	ctl_sk->sk_mark = 0;
 	xfrm_sk_free_policy(ctl_sk);
 	sock_net_set(ctl_sk, &init_net);
 	__TCP_INC_STATS(net, TCP_MIB_OUTSEGS);
@@ -935,7 +937,6 @@ static void tcp_v4_send_ack(const struct sock *sk,
 			      &arg, arg.iov[0].iov_len,
 			      transmit_time);
 
-	ctl_sk->sk_mark = 0;
 	sock_net_set(ctl_sk, &init_net);
 	__TCP_INC_STATS(net, TCP_MIB_OUTSEGS);
 	local_bh_enable();
Signature · 1 line
2.39.2
author maintainer
#174 lore raw

PATCH 174/292 serial: arc_uart: fix of_iomap leak in `arc_serial_probe`

In reply to Greg Kroah-Hartman

From: Ke Zhang <[email protected]>

[ Upstream commit 8ab5fc55d7f65d58a3c3aeadf11bdf60267cd2bd ]

Smatch reports:

drivers/tty/serial/arc_uart.c:631 arc_serial_probe() warn:
'port->membase' from of_iomap() not released on lines: 631.

In arc_serial_probe(), if uart_add_one_port() fails, port->membase is not released, which would cause a resource leak.

To fix this, I replace of_iomap with devm_platform_ioremap_resource.

---
 drivers/tty/serial/arc_uart.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/tty/serial/arc_uart.c +4 -3
index 2a65ea2660e10..f3ccc59d8c1f3 100644
@@ -607,10 +607,11 @@ static int arc_serial_probe(struct platform_device *pdev)
 	}
 	uart->baud = val;
 
-	port->membase = of_iomap(np, 0);
-	if (!port->membase)
+	port->membase = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(port->membase)) {
 		/* No point of dev_err since UART itself is hosed here */
-		return -ENXIO;
+		return PTR_ERR(port->membase);
+	}
 
 	port->irq = irq_of_parse_and_map(np, 0);
 
Signature · 1 line
2.39.2
author maintainer
#175 lore raw

PATCH 175/292 serial: 8250_bcm7271: balance clk_enable calls

In reply to Greg Kroah-Hartman

From: Doug Berger <[email protected]>

[ Upstream commit 8a3b5477256a54ae4a470dcebbcf8cdc18e4696d ]

The sw_baud clock must be disabled when the device driver is not connected to the device. This now occurs when probe fails and upon remove.

---
 drivers/tty/serial/8250/8250_bcm7271.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/tty/serial/8250/8250_bcm7271.c +4 -1
index 36e31b96ef4a5..1f0095cf57a7e 100644
@@ -1034,7 +1034,7 @@ static int brcmuart_probe(struct platform_device *pdev)
 	if (clk_rate == 0) {
 		dev_err(dev, "clock-frequency or clk not defined\n");
 		ret = -EINVAL;
-		goto release_dma;
+		goto err_clk_disable;
 	}
 
 	dev_dbg(dev, "DMA is %senabled\n", priv->dma_enabled ? "" : "not ");
@@ -1121,6 +1121,8 @@ static int brcmuart_probe(struct platform_device *pdev)
 	serial8250_unregister_port(priv->line);
 err:
 	brcmuart_free_bufs(dev, priv);
+err_clk_disable:
+	clk_disable_unprepare(baud_mux_clk);
 release_dma:
 	if (priv->dma_enabled)
 		brcmuart_arbitration(priv, 0);
@@ -1135,6 +1137,7 @@ static int brcmuart_remove(struct platform_device *pdev)
 	hrtimer_cancel(&priv->hrt);
 	serial8250_unregister_port(priv->line);
 	brcmuart_free_bufs(&pdev->dev, priv);
+	clk_disable_unprepare(priv->baud_mux_clk);
 	if (priv->dma_enabled)
 		brcmuart_arbitration(priv, 0);
 	return 0;
Signature · 1 line
2.39.2
author maintainer
#176 lore raw

PATCH 176/292 serial: 8250_bcm7271: fix leak in `brcmuart_probe`

In reply to Greg Kroah-Hartman

From: Doug Berger <[email protected]>

[ Upstream commit f264f2f6f4788dc031cef60a0cf2881902736709 ]

Smatch reports:
drivers/tty/serial/8250/8250_bcm7271.c:1120 brcmuart_probe() warn:
'baud_mux_clk' from clk_prepare_enable() not released on lines: 1032.

The issue is fixed by using a managed clock.

---
 drivers/tty/serial/8250/8250_bcm7271.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/tty/serial/8250/8250_bcm7271.c +1 -1
index 1f0095cf57a7e..ffc7f67e27e35 100644
@@ -1014,7 +1014,7 @@ static int brcmuart_probe(struct platform_device *pdev)
 	of_property_read_u32(np, "clock-frequency", &clk_rate);
 
 	/* See if a Baud clock has been specified */
-	baud_mux_clk = of_clk_get_by_name(np, "sw_baud");
+	baud_mux_clk = devm_clk_get(dev, "sw_baud");
 	if (IS_ERR(baud_mux_clk)) {
 		if (PTR_ERR(baud_mux_clk) == -EPROBE_DEFER) {
 			ret = -EPROBE_DEFER;
Signature · 1 line
2.39.2
author maintainer
#177 lore raw

PATCH 177/292 erspan: get the proto with the md version for collect_md

In reply to Greg Kroah-Hartman

From: Xin Long <[email protected]>

[ Upstream commit d80fc101d2eb9b3188c228d61223890aeea480a4 ]

In commit 20704bd1633d ("erspan: build the header with the right proto according to erspan_ver"), it gets the proto with t->parms.erspan_ver, but t->parms.erspan_ver is not used by collect_md branch, and instead it should get the proto with md->version for collect_md.

Thanks to Kevin for pointing this out.

Fixes 20704bd1633d ("erspan: build the header with the right proto according to erspan_ver")
Fixes 94d7d8f29287 ("ip6_gre: add erspan v2 support")
---
 net/ipv6/ip6_gre.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)
net/ipv6/ip6_gre.c +8 -5
index 4d5937af08ee9..216b40ccadae0 100644
@@ -1037,12 +1037,14 @@ static netdev_tx_t ip6erspan_tunnel_xmit(struct sk_buff *skb,
 					    ntohl(tun_id),
 					    ntohl(md->u.index), truncate,
 					    false);
+			proto = htons(ETH_P_ERSPAN);
 		} else if (md->version == 2) {
 			erspan_build_header_v2(skb,
 					       ntohl(tun_id),
 					       md->u.md2.dir,
 					       get_hwid(&md->u.md2),
 					       truncate, false);
+			proto = htons(ETH_P_ERSPAN2);
 		} else {
 			goto tx_err;
 		}
@@ -1065,24 +1067,25 @@ static netdev_tx_t ip6erspan_tunnel_xmit(struct sk_buff *skb,
 			break;
 		}
 
-		if (t->parms.erspan_ver == 1)
+		if (t->parms.erspan_ver == 1) {
 			erspan_build_header(skb, ntohl(t->parms.o_key),
 					    t->parms.index,
 					    truncate, false);
-		else if (t->parms.erspan_ver == 2)
+			proto = htons(ETH_P_ERSPAN);
+		} else if (t->parms.erspan_ver == 2) {
 			erspan_build_header_v2(skb, ntohl(t->parms.o_key),
 					       t->parms.dir,
 					       t->parms.hwid,
 					       truncate, false);
-		else
+			proto = htons(ETH_P_ERSPAN2);
+		} else {
 			goto tx_err;
+		}
 
 		fl6.daddr = t->parms.raddr;
 	}
 
 	/* Push GRE header. */
-	proto = (t->parms.erspan_ver == 1) ? htons(ETH_P_ERSPAN)
-					   : htons(ETH_P_ERSPAN2);
 	gre_build_header(skb, 8, TUNNEL_SEQ, proto, 0, htonl(atomic_fetch_inc(&t->o_seqno)));
 
 	/* TooBig packet may have updated dst->dev's mtu */
Signature · 1 line
2.39.2
author maintainer
#178 lore raw

PATCH 178/292 net: dsa: rzn1-a5psw: enable management frames for CPU port

In reply to Greg Kroah-Hartman

From: Clément Léger <[email protected]>

[ Upstream commit 9e4b45f20c5aac786c728619e5ee746bffce1798 ]

Currently, management frame were discarded before reaching the CPU port due to a misconfiguration of the MGMT_CONFIG register. Enable them by setting the correct value in this register in order to correctly receive management frame and handle STP.

Fixes 888cdb892b61 ("net: dsa: rzn1-a5psw: add Renesas RZ/N1 advanced 5 port switch driver")
---
 drivers/net/dsa/rzn1_a5psw.c | 2 +-
 drivers/net/dsa/rzn1_a5psw.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/net/dsa/rzn1_a5psw.c +1 -1
index ed413d555beca..92a3ac78ab1e5 100644
@@ -673,7 +673,7 @@ static int a5psw_setup(struct dsa_switch *ds)
 	}
 
 	/* Configure management port */
-	reg = A5PSW_CPU_PORT | A5PSW_MGMT_CFG_DISCARD;
+	reg = A5PSW_CPU_PORT | A5PSW_MGMT_CFG_ENABLE;
 	a5psw_reg_writel(a5psw, A5PSW_MGMT_CFG, reg);
 
 	/* Set pattern 0 to forward all frame to mgmt port */
drivers/net/dsa/rzn1_a5psw.h +1 -1
index c67abd49c013d..b4fbf453ff741 100644
@@ -36,7 +36,7 @@
 #define A5PSW_INPUT_LEARN_BLOCK(p)	BIT(p)
 
 #define A5PSW_MGMT_CFG			0x20
-#define A5PSW_MGMT_CFG_DISCARD		BIT(7)
+#define A5PSW_MGMT_CFG_ENABLE		BIT(6)
 
 #define A5PSW_MODE_CFG			0x24
 #define A5PSW_MODE_STATS_RESET		BIT(31)
Signature · 1 line
2.39.2
author maintainer
#179 lore raw

PATCH 179/292 net: dsa: rzn1-a5psw: fix STP states handling

In reply to Greg Kroah-Hartman

From: Alexis Lothoré <[email protected]>

[ Upstream commit ebe9bc50952757b4b25eaf514da7c464196c9606 ]

stp_set_state() should actually allow receiving BPDU while in LEARNING mode which is not the case. Additionally, the BLOCKEN bit does not actually forbid sending forwarded frames from that port. To fix this, add a5psw_port_tx_enable() function which allows to disable TX. However, while its name suggest that TX is totally disabled, it is not and can still allow to send BPDUs even if disabled. This can be done by using forced forwarding with the switch tagging mechanism but keeping "filtering" disabled (which is already the case in the rzn1-a5sw tag driver). With these fixes, STP support is now functional.

Fixes 888cdb892b61 ("net: dsa: rzn1-a5psw: add Renesas RZ/N1 advanced 5 port switch driver")
---
 drivers/net/dsa/rzn1_a5psw.c | 57 ++++++++++++++++++++++++++++++------
 drivers/net/dsa/rzn1_a5psw.h |  1 +
 2 files changed, 49 insertions(+), 9 deletions(-)
drivers/net/dsa/rzn1_a5psw.c +48 -9
index 92a3ac78ab1e5..2b0463263767c 100644
@@ -120,6 +120,22 @@ static void a5psw_port_mgmtfwd_set(struct a5psw *a5psw, int port, bool enable)
 	a5psw_port_pattern_set(a5psw, port, A5PSW_PATTERN_MGMTFWD, enable);
 }
 
+static void a5psw_port_tx_enable(struct a5psw *a5psw, int port, bool enable)
+{
+	u32 mask = A5PSW_PORT_ENA_TX(port);
+	u32 reg = enable ? mask : 0;
+
+	/* Even though the port TX is disabled through TXENA bit in the
+	 * PORT_ENA register, it can still send BPDUs. This depends on the tag
+	 * configuration added when sending packets from the CPU port to the
+	 * switch port. Indeed, when using forced forwarding without filtering,
+	 * even disabled ports will be able to send packets that are tagged.
+	 * This allows to implement STP support when ports are in a state where
+	 * forwarding traffic should be stopped but BPDUs should still be sent.
+	 */
+	a5psw_reg_rmw(a5psw, A5PSW_PORT_ENA, mask, reg);
+}
+
 static void a5psw_port_enable_set(struct a5psw *a5psw, int port, bool enable)
 {
 	u32 port_ena = 0;
@@ -292,6 +308,22 @@ static int a5psw_set_ageing_time(struct dsa_switch *ds, unsigned int msecs)
 	return 0;
 }
 
+static void a5psw_port_learning_set(struct a5psw *a5psw, int port, bool learn)
+{
+	u32 mask = A5PSW_INPUT_LEARN_DIS(port);
+	u32 reg = !learn ? mask : 0;
+
+	a5psw_reg_rmw(a5psw, A5PSW_INPUT_LEARN, mask, reg);
+}
+
+static void a5psw_port_rx_block_set(struct a5psw *a5psw, int port, bool block)
+{
+	u32 mask = A5PSW_INPUT_LEARN_BLOCK(port);
+	u32 reg = block ? mask : 0;
+
+	a5psw_reg_rmw(a5psw, A5PSW_INPUT_LEARN, mask, reg);
+}
+
 static void a5psw_flooding_set_resolution(struct a5psw *a5psw, int port,
 					  bool set)
 {
@@ -344,28 +376,35 @@ static void a5psw_port_bridge_leave(struct dsa_switch *ds, int port,
 
 static void a5psw_port_stp_state_set(struct dsa_switch *ds, int port, u8 state)
 {
-	u32 mask = A5PSW_INPUT_LEARN_DIS(port) | A5PSW_INPUT_LEARN_BLOCK(port);
+	bool learning_enabled, rx_enabled, tx_enabled;
 	struct a5psw *a5psw = ds->priv;
-	u32 reg = 0;
 
 	switch (state) {
 	case BR_STATE_DISABLED:
 	case BR_STATE_BLOCKING:
-		reg |= A5PSW_INPUT_LEARN_DIS(port);
-		reg |= A5PSW_INPUT_LEARN_BLOCK(port);
-		break;
 	case BR_STATE_LISTENING:
-		reg |= A5PSW_INPUT_LEARN_DIS(port);
+		rx_enabled = false;
+		tx_enabled = false;
+		learning_enabled = false;
 		break;
 	case BR_STATE_LEARNING:
-		reg |= A5PSW_INPUT_LEARN_BLOCK(port);
+		rx_enabled = false;
+		tx_enabled = false;
+		learning_enabled = true;
 		break;
 	case BR_STATE_FORWARDING:
-	default:
+		rx_enabled = true;
+		tx_enabled = true;
+		learning_enabled = true;
 		break;
+	default:
+		dev_err(ds->dev, "invalid STP state: %d\n", state);
+		return;
 	}
 
-	a5psw_reg_rmw(a5psw, A5PSW_INPUT_LEARN, mask, reg);
+	a5psw_port_learning_set(a5psw, port, learning_enabled);
+	a5psw_port_rx_block_set(a5psw, port, !rx_enabled);
+	a5psw_port_tx_enable(a5psw, port, tx_enabled);
 }
 
 static void a5psw_port_fast_age(struct dsa_switch *ds, int port)
drivers/net/dsa/rzn1_a5psw.h +1 -0
index b4fbf453ff741..b869192eef3f7 100644
@@ -19,6 +19,7 @@
 #define A5PSW_PORT_OFFSET(port)		(0x400 * (port))
 
 #define A5PSW_PORT_ENA			0x8
+#define A5PSW_PORT_ENA_TX(port)		BIT(port)
 #define A5PSW_PORT_ENA_RX_SHIFT		16
 #define A5PSW_PORT_ENA_TX_RX(port)	(BIT((port) + A5PSW_PORT_ENA_RX_SHIFT) | \
 					 BIT(port))
Signature · 1 line
2.39.2
author maintainer
#180 lore raw

PATCH 180/292 net: dsa: rzn1-a5psw: disable learning for standalone ports

In reply to Greg Kroah-Hartman

From: Clément Léger <[email protected]>

[ Upstream commit ec52b69c046a6219011af780aca155a96719637b ]

When ports are in standalone mode, they should have learning disabled to avoid adding new entries in the MAC lookup table which might be used by other bridge ports to forward packets. While adding that, also make sure learning is enabled for CPU port.

Fixes 888cdb892b61 ("net: dsa: rzn1-a5psw: add Renesas RZ/N1 advanced 5 port switch driver")
---
 drivers/net/dsa/rzn1_a5psw.c | 24 ++++++++++++++++--------
 1 file changed, 16 insertions(+), 8 deletions(-)
drivers/net/dsa/rzn1_a5psw.c +16 -8
index 2b0463263767c..790e177e2aef6 100644
@@ -340,6 +340,14 @@ static void a5psw_flooding_set_resolution(struct a5psw *a5psw, int port,
 		a5psw_reg_writel(a5psw, offsets[i], a5psw->bridged_ports);
 }
 
+static void a5psw_port_set_standalone(struct a5psw *a5psw, int port,
+				      bool standalone)
+{
+	a5psw_port_learning_set(a5psw, port, !standalone);
+	a5psw_flooding_set_resolution(a5psw, port, !standalone);
+	a5psw_port_mgmtfwd_set(a5psw, port, standalone);
+}
+
 static int a5psw_port_bridge_join(struct dsa_switch *ds, int port,
 				  struct dsa_bridge bridge,
 				  bool *tx_fwd_offload,
@@ -355,8 +363,7 @@ static int a5psw_port_bridge_join(struct dsa_switch *ds, int port,
 	}
 
 	a5psw->br_dev = bridge.dev;
-	a5psw_flooding_set_resolution(a5psw, port, true);
-	a5psw_port_mgmtfwd_set(a5psw, port, false);
+	a5psw_port_set_standalone(a5psw, port, false);
 
 	return 0;
 }
@@ -366,8 +373,7 @@ static void a5psw_port_bridge_leave(struct dsa_switch *ds, int port,
 {
 	struct a5psw *a5psw = ds->priv;
 
-	a5psw_flooding_set_resolution(a5psw, port, false);
-	a5psw_port_mgmtfwd_set(a5psw, port, true);
+	a5psw_port_set_standalone(a5psw, port, true);
 
 	/* No more ports bridged */
 	if (a5psw->bridged_ports == BIT(A5PSW_CPU_PORT))
@@ -761,13 +767,15 @@ static int a5psw_setup(struct dsa_switch *ds)
 		if (dsa_port_is_unused(dp))
 			continue;
 
-		/* Enable egress flooding for CPU port */
-		if (dsa_port_is_cpu(dp))
+		/* Enable egress flooding and learning for CPU port */
+		if (dsa_port_is_cpu(dp)) {
 			a5psw_flooding_set_resolution(a5psw, port, true);
+			a5psw_port_learning_set(a5psw, port, true);
+		}
 
-		/* Enable management forward only for user ports */
+		/* Enable standalone mode for user ports */
 		if (dsa_port_is_user(dp))
-			a5psw_port_mgmtfwd_set(a5psw, port, true);
+			a5psw_port_set_standalone(a5psw, port, true);
 	}
 
 	return 0;
Signature · 1 line
2.39.2
author maintainer
#181 lore raw

PATCH 181/292 net: hns3: fix output information incomplete for dumping tx queue info with debugfs

In reply to Greg Kroah-Hartman

From: Jie Wang <[email protected]>

[ Upstream commit 89f6bfb071182f05d7188c255b0e7251c3806f16 ]

In function hns3_dump_tx_queue_info, The print buffer is not enough when the tx BD number is configured to 32760. As a result several BD information wouldn't be displayed.

So fix it by increasing the tx queue print buffer length.

Fixes 630a6738da82 ("net: hns3: adjust string spaces of some parameters of tx bd info in debugfs")
---
 drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c | 2 +-
 drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.h | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)
drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c +1 -1
index 66feb23f7b7b6..bcccd82a2620f 100644
@@ -130,7 +130,7 @@ static struct hns3_dbg_cmd_info hns3_dbg_cmd[] = {
 		.name = "tx_bd_queue",
 		.cmd = HNAE3_DBG_CMD_TX_BD,
 		.dentry = HNS3_DBG_DENTRY_TX_BD,
-		.buf_len = HNS3_DBG_READ_LEN_4MB,
+		.buf_len = HNS3_DBG_READ_LEN_5MB,
 		.init = hns3_dbg_bd_file_init,
 	},
 	{
drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.h +1 -0
index 97578eabb7d8b..4a5ef8a90a104 100644
@@ -10,6 +10,7 @@
 #define HNS3_DBG_READ_LEN_128KB	0x20000
 #define HNS3_DBG_READ_LEN_1MB	0x100000
 #define HNS3_DBG_READ_LEN_4MB	0x400000
+#define HNS3_DBG_READ_LEN_5MB	0x500000
 #define HNS3_DBG_WRITE_LEN	1024
 
 #define HNS3_DBG_DATA_STR_LEN	32
Signature · 1 line
2.39.2
author maintainer
#182 lore raw

PATCH 182/292 net: hns3: fix sending pfc frames after reset issue

In reply to Greg Kroah-Hartman

From: Jijie Shao <[email protected]>

[ Upstream commit f14db07064727dd3bc0906c77a6d2759c1bbb395 ]

To prevent the system from abnormally sending PFC frames after an abnormal reset. The hns3 driver notifies the firmware to disable pfc before reset.

Fixes 35d93a30040c ("net: hns3: adjust the process of PF reset")
---
 .../ethernet/hisilicon/hns3/hns3pf/hclge_main.c   | 15 +++++++++------
 .../net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c |  4 ++--
 .../net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.h |  5 +++++
 3 files changed, 16 insertions(+), 8 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +9 -6
index 07ad5f35219e2..50e956d6c3b25 100644
@@ -8053,12 +8053,15 @@ static void hclge_ae_stop(struct hnae3_handle *handle)
 	/* If it is not PF reset or FLR, the firmware will disable the MAC,
 	 * so it only need to stop phy here.
 	 */
-	if (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state) &&
-	    hdev->reset_type != HNAE3_FUNC_RESET &&
-	    hdev->reset_type != HNAE3_FLR_RESET) {
-		hclge_mac_stop_phy(hdev);
-		hclge_update_link_status(hdev);
-		return;
+	if (test_bit(HCLGE_STATE_RST_HANDLING, &hdev->state)) {
+		hclge_pfc_pause_en_cfg(hdev, HCLGE_PFC_TX_RX_DISABLE,
+				       HCLGE_PFC_DISABLE);
+		if (hdev->reset_type != HNAE3_FUNC_RESET &&
+		    hdev->reset_type != HNAE3_FLR_RESET) {
+			hclge_mac_stop_phy(hdev);
+			hclge_update_link_status(hdev);
+			return;
+		}
 	}
 
 	hclge_reset_tqp(handle);
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c +2 -2
index 4a33f65190e2b..922c0da3660c7 100644
@@ -171,8 +171,8 @@ int hclge_mac_pause_en_cfg(struct hclge_dev *hdev, bool tx, bool rx)
 	return hclge_cmd_send(&hdev->hw, &desc, 1);
 }
 
-static int hclge_pfc_pause_en_cfg(struct hclge_dev *hdev, u8 tx_rx_bitmap,
-				  u8 pfc_bitmap)
+int hclge_pfc_pause_en_cfg(struct hclge_dev *hdev, u8 tx_rx_bitmap,
+			   u8 pfc_bitmap)
 {
 	struct hclge_desc desc;
 	struct hclge_pfc_en_cmd *pfc = (struct hclge_pfc_en_cmd *)desc.data;
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.h +5 -0
index 68f28a98e380b..dd6f1fd486cf2 100644
@@ -164,6 +164,9 @@ struct hclge_bp_to_qs_map_cmd {
 	u32 rsvd1;
 };
 
+#define HCLGE_PFC_DISABLE	0
+#define HCLGE_PFC_TX_RX_DISABLE	0
+
 struct hclge_pfc_en_cmd {
 	u8 tx_rx_en_bitmap;
 	u8 pri_en_bitmap;
@@ -235,6 +238,8 @@ void hclge_tm_schd_info_update(struct hclge_dev *hdev, u8 num_tc);
 void hclge_tm_pfc_info_update(struct hclge_dev *hdev);
 int hclge_tm_dwrr_cfg(struct hclge_dev *hdev);
 int hclge_tm_init_hw(struct hclge_dev *hdev, bool init);
+int hclge_pfc_pause_en_cfg(struct hclge_dev *hdev, u8 tx_rx_bitmap,
+			   u8 pfc_bitmap);
 int hclge_mac_pause_en_cfg(struct hclge_dev *hdev, bool tx, bool rx);
 int hclge_pause_addr_cfg(struct hclge_dev *hdev, const u8 *mac_addr);
 void hclge_pfc_rx_stats_get(struct hclge_dev *hdev, u64 *stats);
Signature · 1 line
2.39.2
author maintainer
#183 lore raw

PATCH 183/292 net: hns3: fix reset delay time to avoid configuration timeout

In reply to Greg Kroah-Hartman

From: Jie Wang <[email protected]>

[ Upstream commit 814d0c786068e858d889ada3153bff82f64223ad ]

Currently the hns3 vf function reset delays 5000ms before vf rebuild process. In product applications, this delay is too long for application configurations and causes configuration timeout.

According to the tests, 500ms delay is enough for reset process except PF FLR. So this patch modifies delay to 500ms in these scenarios.

Fixes 6988eb2a9b77 ("net: hns3: Add support to reset the enet/ring mgmt layer")
---
 drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +4 -1
index e84e5be8e59ed..b1b14850e958f 100644
@@ -1436,7 +1436,10 @@ static int hclgevf_reset_wait(struct hclgevf_dev *hdev)
 	 * might happen in case reset assertion was made by PF. Yes, this also
 	 * means we might end up waiting bit more even for VF reset.
 	 */
-	msleep(5000);
+	if (hdev->reset_type == HNAE3_VF_FULL_RESET)
+		msleep(5000);
+	else
+		msleep(500);
 
 	return 0;
 }
Signature · 1 line
2.39.2
author maintainer
#184 lore raw

PATCH 184/292 net: hns3: fix reset timeout when enable full VF

In reply to Greg Kroah-Hartman

From: Jijie Shao <[email protected]>

[ Upstream commit 6b45d5ff8c2c61baddd67d7510075ae121c5e704 ]

The timeout of the cmdq reset command has been increased to resolve the reset timeout issue in the full VF scenario. The timeout of other cmdq commands remains unchanged.

Fixes 8d307f8e8cf1 ("net: hns3: create new set of unified hclge_comm_cmd_send APIs")
---
 .../hns3/hns3_common/hclge_comm_cmd.c         | 25 ++++++++++++++++---
 .../hns3/hns3_common/hclge_comm_cmd.h         |  8 +++++-
 2 files changed, 28 insertions(+), 5 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.c +21 -4
index f671a63cecde4..c797d54f98caa 100644
@@ -330,9 +330,25 @@ static int hclge_comm_cmd_csq_done(struct hclge_comm_hw *hw)
 	return head == hw->cmq.csq.next_to_use;
 }
 
-static void hclge_comm_wait_for_resp(struct hclge_comm_hw *hw,
+static u32 hclge_get_cmdq_tx_timeout(u16 opcode, u32 tx_timeout)
+{
+	static const struct hclge_cmdq_tx_timeout_map cmdq_tx_timeout_map[] = {
+		{HCLGE_OPC_CFG_RST_TRIGGER, HCLGE_COMM_CMDQ_TX_TIMEOUT_500MS},
+	};
+	u32 i;
+
+	for (i = 0; i < ARRAY_SIZE(cmdq_tx_timeout_map); i++)
+		if (cmdq_tx_timeout_map[i].opcode == opcode)
+			return cmdq_tx_timeout_map[i].tx_timeout;
+
+	return tx_timeout;
+}
+
+static void hclge_comm_wait_for_resp(struct hclge_comm_hw *hw, u16 opcode,
 				     bool *is_completed)
 {
+	u32 cmdq_tx_timeout = hclge_get_cmdq_tx_timeout(opcode,
+							hw->cmq.tx_timeout);
 	u32 timeout = 0;
 
 	do {
@@ -342,7 +358,7 @@ static void hclge_comm_wait_for_resp(struct hclge_comm_hw *hw,
 		}
 		udelay(1);
 		timeout++;
-	} while (timeout < hw->cmq.tx_timeout);
+	} while (timeout < cmdq_tx_timeout);
 }
 
 static int hclge_comm_cmd_convert_err_code(u16 desc_ret)
@@ -406,7 +422,8 @@ static int hclge_comm_cmd_check_result(struct hclge_comm_hw *hw,
 	 * if multi descriptors to be sent, use the first one to check
 	 */
 	if (HCLGE_COMM_SEND_SYNC(le16_to_cpu(desc->flag)))
-		hclge_comm_wait_for_resp(hw, &is_completed);
+		hclge_comm_wait_for_resp(hw, le16_to_cpu(desc->opcode),
+					 &is_completed);
 
 	if (!is_completed)
 		ret = -EBADE;
@@ -528,7 +545,7 @@ int hclge_comm_cmd_queue_init(struct pci_dev *pdev, struct hclge_comm_hw *hw)
 	cmdq->crq.desc_num = HCLGE_COMM_NIC_CMQ_DESC_NUM;
 
 	/* Setup Tx write back timeout */
-	cmdq->tx_timeout = HCLGE_COMM_CMDQ_TX_TIMEOUT;
+	cmdq->tx_timeout = HCLGE_COMM_CMDQ_TX_TIMEOUT_DEFAULT;
 
 	/* Setup queue rings */
 	ret = hclge_comm_alloc_cmd_queue(hw, HCLGE_COMM_TYPE_CSQ);
drivers/net/ethernet/hisilicon/hns3/hns3_common/hclge_comm_cmd.h +7 -1
index b1f9383b418f4..2b2928c6dccfc 100644
@@ -54,7 +54,8 @@
 #define HCLGE_COMM_NIC_SW_RST_RDY		BIT(HCLGE_COMM_NIC_SW_RST_RDY_B)
 #define HCLGE_COMM_NIC_CMQ_DESC_NUM_S		3
 #define HCLGE_COMM_NIC_CMQ_DESC_NUM		1024
-#define HCLGE_COMM_CMDQ_TX_TIMEOUT		30000
+#define HCLGE_COMM_CMDQ_TX_TIMEOUT_DEFAULT	30000
+#define HCLGE_COMM_CMDQ_TX_TIMEOUT_500MS	500000
 
 enum hclge_opcode_type {
 	/* Generic commands */
@@ -357,6 +358,11 @@ struct hclge_comm_caps_bit_map {
 	u16 local_bit;
 };
 
+struct hclge_cmdq_tx_timeout_map {
+	u32 opcode;
+	u32 tx_timeout;
+};
+
 struct hclge_comm_firmware_compat_cmd {
 	__le32 compat;
 	u8 rsv[20];
Signature · 1 line
2.39.2
author maintainer
#185 lore raw

PATCH 185/292 media: netup_unidvb: fix use-after-free at del_timer()

In reply to Greg Kroah-Hartman

From: Duoming Zhou <[email protected]>

[ Upstream commit 0f5bb36bf9b39a2a96e730bf4455095b50713f63 ]

When Universal DVB card is detaching, netup_unidvb_dma_fini() uses del_timer() to stop dma->timeout timer. But when timer handler netup_unidvb_dma_timeout() is running, del_timer() could not stop it. As a result, the use-after-free bug could happen. The process is shown below:

    (cleanup routine)          |        (timer routine)
                               | mod_timer(&dev->tx_sim_timer, ..)
netup_unidvb_finidev()         | (wait a time)
  netup_unidvb_dma_fini()      | netup_unidvb_dma_timeout()
    del_timer(&dma->timeout);  |
                               |   ndev->pci_dev->dev //USE

Fix by changing del_timer() to del_timer_sync().

Fixes 52b1eaf4c59a ("[media] netup_unidvb: NetUP Universal DVB-S/S2/T/T2/C PCI-E card driver")
---
 drivers/media/pci/netup_unidvb/netup_unidvb_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/media/pci/netup_unidvb/netup_unidvb_core.c +1 -1
index 8287851b5ffdc..aaa1d2dedebdd 100644
@@ -697,7 +697,7 @@ static void netup_unidvb_dma_fini(struct netup_unidvb_dev *ndev, int num)
 	netup_unidvb_dma_enable(dma, 0);
 	msleep(50);
 	cancel_work_sync(&dma->work);
-	del_timer(&dma->timeout);
+	del_timer_sync(&dma->timeout);
 }
 
 static int netup_unidvb_dma_setup(struct netup_unidvb_dev *ndev)
Signature · 1 line
2.39.2
author maintainer
#186 lore raw

PATCH 186/292 SUNRPC: double free xprt_ctxt while still in use

In reply to Greg Kroah-Hartman

From: NeilBrown <[email protected]>

[ Upstream commit eb8d3a2c809abd73ab0a060fe971d6b9019aa3c1 ]

When an RPC request is deferred, the rq_xprt_ctxt pointer is moved out of the svc_rqst into the svc_deferred_req.
When the deferred request is revisited, the pointer is copied into the new svc_rqst - and also remains in the svc_deferred_req.

In the (rare?) case that the request is deferred a second time, the old svc_deferred_req is reused - it still has all the correct content. However in that case the rq_xprt_ctxt pointer is NOT cleared so that when xpo_release_xprt is called, the ctxt is freed (UDP) or possible added to a free list (RDMA).
When the deferred request is revisited for a second time, it will reference this ctxt which may be invalid, and the free the object a second time which is likely to oops.

So change svc_defer() to *always* clear rq_xprt_ctxt, and assert that the value is now stored in the svc_deferred_req.

Fixes 773f91b2cf3f ("SUNRPC: Fix NFSD's request deferral on RDMA transports")
---
 net/sunrpc/svc_xprt.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
net/sunrpc/svc_xprt.c +2 -1
index c2ce125380080..b4306cf1b458c 100644
@@ -1228,13 +1228,14 @@ static struct cache_deferred_req *svc_defer(struct cache_req *req)
 		dr->daddr = rqstp->rq_daddr;
 		dr->argslen = rqstp->rq_arg.len >> 2;
 		dr->xprt_ctxt = rqstp->rq_xprt_ctxt;
-		rqstp->rq_xprt_ctxt = NULL;
 
 		/* back up head to the start of the buffer and copy */
 		skip = rqstp->rq_arg.len - rqstp->rq_arg.head[0].iov_len;
 		memcpy(dr->args, rqstp->rq_arg.head[0].iov_base - skip,
 		       dr->argslen << 2);
 	}
+	WARN_ON_ONCE(rqstp->rq_xprt_ctxt != dr->xprt_ctxt);
+	rqstp->rq_xprt_ctxt = NULL;
 	trace_svc_defer(rqstp);
 	svc_xprt_get(rqstp->rq_xprt);
 	dr->xprt = rqstp->rq_xprt;
Signature · 1 line
2.39.2
author maintainer
#187 lore raw

PATCH 187/292 SUNRPC: always free ctxt when freeing deferred request

In reply to Greg Kroah-Hartman

From: NeilBrown <[email protected]>

[ Upstream commit 948f072ada23e0a504c5e4d7d71d4c83bd0785ec ]

Since the ->xprt_ctxt pointer was added to svc_deferred_req, it has not been sufficient to use kfree() to free a deferred request. We may need to free the ctxt as well.

As freeing the ctxt is all that ->xpo_release_rqst() does, we repurpose it to explicit do that even when the ctxt is not stored in an rqst. So we now have ->xpo_release_ctxt() which is given an xprt and a ctxt, which may have been taken either from an rqst or from a dreq. The caller is now responsible for clearing that pointer after the call to ->xpo_release_ctxt.

We also clear dr->xprt_ctxt when the ctxt is moved into a new rqst when revisiting a deferred request. This ensures there is only one pointer to the ctxt, so the risk of double freeing in future is reduced. The new code in svc_xprt_release which releases both the ctxt and any rq_deferred depends on this.

Fixes 773f91b2cf3f ("SUNRPC: Fix NFSD's request deferral on RDMA transports")
---
 include/linux/sunrpc/svc_rdma.h          |  2 +-
 include/linux/sunrpc/svc_xprt.h          |  2 +-
 net/sunrpc/svc_xprt.c                    | 23 +++++++++++++-----
 net/sunrpc/svcsock.c                     | 30 +++++++++++++-----------
 net/sunrpc/xprtrdma/svc_rdma_recvfrom.c  | 11 ++++-----
 net/sunrpc/xprtrdma/svc_rdma_transport.c |  2 +-
 6 files changed, 41 insertions(+), 29 deletions(-)
include/linux/sunrpc/svc_rdma.h +1 -1
index 24aa159d29a7f..fbc4bd423b355 100644
@@ -176,7 +176,7 @@ extern struct svc_rdma_recv_ctxt *
 extern void svc_rdma_recv_ctxt_put(struct svcxprt_rdma *rdma,
 				   struct svc_rdma_recv_ctxt *ctxt);
 extern void svc_rdma_flush_recv_queues(struct svcxprt_rdma *rdma);
-extern void svc_rdma_release_rqst(struct svc_rqst *rqstp);
+extern void svc_rdma_release_ctxt(struct svc_xprt *xprt, void *ctxt);
 extern int svc_rdma_recvfrom(struct svc_rqst *);
 
 /* svc_rdma_rw.c */
include/linux/sunrpc/svc_xprt.h +1 -1
index d42a75b3be102..e882fe16a5008 100644
@@ -23,7 +23,7 @@ struct svc_xprt_ops {
 	int		(*xpo_sendto)(struct svc_rqst *);
 	int		(*xpo_result_payload)(struct svc_rqst *, unsigned int,
 					      unsigned int);
-	void		(*xpo_release_rqst)(struct svc_rqst *);
+	void		(*xpo_release_ctxt)(struct svc_xprt *xprt, void *ctxt);
 	void		(*xpo_detach)(struct svc_xprt *);
 	void		(*xpo_free)(struct svc_xprt *);
 	void		(*xpo_secure_port)(struct svc_rqst *rqstp);
net/sunrpc/svc_xprt.c +17 -6
index b4306cf1b458c..8117d0e08d5a2 100644
@@ -534,13 +534,23 @@ void svc_reserve(struct svc_rqst *rqstp, int space)
 }
 EXPORT_SYMBOL_GPL(svc_reserve);
 
+static void free_deferred(struct svc_xprt *xprt, struct svc_deferred_req *dr)
+{
+	if (!dr)
+		return;
+
+	xprt->xpt_ops->xpo_release_ctxt(xprt, dr->xprt_ctxt);
+	kfree(dr);
+}
+
 static void svc_xprt_release(struct svc_rqst *rqstp)
 {
 	struct svc_xprt	*xprt = rqstp->rq_xprt;
 
-	xprt->xpt_ops->xpo_release_rqst(rqstp);
+	xprt->xpt_ops->xpo_release_ctxt(xprt, rqstp->rq_xprt_ctxt);
+	rqstp->rq_xprt_ctxt = NULL;
 
-	kfree(rqstp->rq_deferred);
+	free_deferred(xprt, rqstp->rq_deferred);
 	rqstp->rq_deferred = NULL;
 
 	pagevec_release(&rqstp->rq_pvec);
@@ -1059,7 +1069,7 @@ static void svc_delete_xprt(struct svc_xprt *xprt)
 	spin_unlock_bh(&serv->sv_lock);
 
 	while ((dr = svc_deferred_dequeue(xprt)) != NULL)
-		kfree(dr);
+		free_deferred(xprt, dr);
 
 	call_xpt_users(xprt);
 	svc_xprt_put(xprt);
@@ -1181,8 +1191,8 @@ static void svc_revisit(struct cache_deferred_req *dreq, int too_many)
 	if (too_many || test_bit(XPT_DEAD, &xprt->xpt_flags)) {
 		spin_unlock(&xprt->xpt_lock);
 		trace_svc_defer_drop(dr);
+		free_deferred(xprt, dr);
 		svc_xprt_put(xprt);
-		kfree(dr);
 		return;
 	}
 	dr->xprt = NULL;
@@ -1227,14 +1237,13 @@ static struct cache_deferred_req *svc_defer(struct cache_req *req)
 		dr->addrlen = rqstp->rq_addrlen;
 		dr->daddr = rqstp->rq_daddr;
 		dr->argslen = rqstp->rq_arg.len >> 2;
-		dr->xprt_ctxt = rqstp->rq_xprt_ctxt;
 
 		/* back up head to the start of the buffer and copy */
 		skip = rqstp->rq_arg.len - rqstp->rq_arg.head[0].iov_len;
 		memcpy(dr->args, rqstp->rq_arg.head[0].iov_base - skip,
 		       dr->argslen << 2);
 	}
-	WARN_ON_ONCE(rqstp->rq_xprt_ctxt != dr->xprt_ctxt);
+	dr->xprt_ctxt = rqstp->rq_xprt_ctxt;
 	rqstp->rq_xprt_ctxt = NULL;
 	trace_svc_defer(rqstp);
 	svc_xprt_get(rqstp->rq_xprt);
@@ -1268,6 +1277,8 @@ static noinline int svc_deferred_recv(struct svc_rqst *rqstp)
 	rqstp->rq_daddr       = dr->daddr;
 	rqstp->rq_respages    = rqstp->rq_pages;
 	rqstp->rq_xprt_ctxt   = dr->xprt_ctxt;
+
+	dr->xprt_ctxt = NULL;
 	svc_xprt_received(rqstp->rq_xprt);
 	return dr->argslen << 2;
 }
net/sunrpc/svcsock.c +16 -14
index 815baf308236a..7107fbcbff343 100644
@@ -111,27 +111,27 @@ static void svc_reclassify_socket(struct socket *sock)
 #endif
 
 /**
- * svc_tcp_release_rqst - Release transport-related resources
- * @rqstp: request structure with resources to be released
+ * svc_tcp_release_ctxt - Release transport-related resources
+ * @xprt: the transport which owned the context
+ * @ctxt: the context from rqstp->rq_xprt_ctxt or dr->xprt_ctxt
  *
  */
-static void svc_tcp_release_rqst(struct svc_rqst *rqstp)
+static void svc_tcp_release_ctxt(struct svc_xprt *xprt, void *ctxt)
 {
 }
 
 /**
- * svc_udp_release_rqst - Release transport-related resources
- * @rqstp: request structure with resources to be released
+ * svc_udp_release_ctxt - Release transport-related resources
+ * @xprt: the transport which owned the context
+ * @ctxt: the context from rqstp->rq_xprt_ctxt or dr->xprt_ctxt
  *
  */
-static void svc_udp_release_rqst(struct svc_rqst *rqstp)
+static void svc_udp_release_ctxt(struct svc_xprt *xprt, void *ctxt)
 {
-	struct sk_buff *skb = rqstp->rq_xprt_ctxt;
+	struct sk_buff *skb = ctxt;
 
-	if (skb) {
-		rqstp->rq_xprt_ctxt = NULL;
+	if (skb)
 		consume_skb(skb);
-	}
 }
 
 union svc_pktinfo_u {
@@ -559,7 +559,8 @@ static int svc_udp_sendto(struct svc_rqst *rqstp)
 	unsigned int sent;
 	int err;
 
-	svc_udp_release_rqst(rqstp);
+	svc_udp_release_ctxt(xprt, rqstp->rq_xprt_ctxt);
+	rqstp->rq_xprt_ctxt = NULL;
 
 	svc_set_cmsg_data(rqstp, cmh);
 
@@ -631,7 +632,7 @@ static const struct svc_xprt_ops svc_udp_ops = {
 	.xpo_recvfrom = svc_udp_recvfrom,
 	.xpo_sendto = svc_udp_sendto,
 	.xpo_result_payload = svc_sock_result_payload,
-	.xpo_release_rqst = svc_udp_release_rqst,
+	.xpo_release_ctxt = svc_udp_release_ctxt,
 	.xpo_detach = svc_sock_detach,
 	.xpo_free = svc_sock_free,
 	.xpo_has_wspace = svc_udp_has_wspace,
@@ -1159,7 +1160,8 @@ static int svc_tcp_sendto(struct svc_rqst *rqstp)
 	unsigned int sent;
 	int err;
 
-	svc_tcp_release_rqst(rqstp);
+	svc_tcp_release_ctxt(xprt, rqstp->rq_xprt_ctxt);
+	rqstp->rq_xprt_ctxt = NULL;
 
 	atomic_inc(&svsk->sk_sendqlen);
 	mutex_lock(&xprt->xpt_mutex);
@@ -1204,7 +1206,7 @@ static const struct svc_xprt_ops svc_tcp_ops = {
 	.xpo_recvfrom = svc_tcp_recvfrom,
 	.xpo_sendto = svc_tcp_sendto,
 	.xpo_result_payload = svc_sock_result_payload,
-	.xpo_release_rqst = svc_tcp_release_rqst,
+	.xpo_release_ctxt = svc_tcp_release_ctxt,
 	.xpo_detach = svc_tcp_sock_detach,
 	.xpo_free = svc_sock_free,
 	.xpo_has_wspace = svc_tcp_has_wspace,
net/sunrpc/xprtrdma/svc_rdma_recvfrom.c +5 -6
index 5242ad121450b..53a7cb2f6c07d 100644
@@ -239,21 +239,20 @@ void svc_rdma_recv_ctxt_put(struct svcxprt_rdma *rdma,
 }
 
 /**
- * svc_rdma_release_rqst - Release transport-specific per-rqst resources
- * @rqstp: svc_rqst being released
+ * svc_rdma_release_ctxt - Release transport-specific per-rqst resources
+ * @xprt: the transport which owned the context
+ * @vctxt: the context from rqstp->rq_xprt_ctxt or dr->xprt_ctxt
  *
  * Ensure that the recv_ctxt is released whether or not a Reply
  * was sent. For example, the client could close the connection,
  * or svc_process could drop an RPC, before the Reply is sent.
  */
-void svc_rdma_release_rqst(struct svc_rqst *rqstp)
+void svc_rdma_release_ctxt(struct svc_xprt *xprt, void *vctxt)
 {
-	struct svc_rdma_recv_ctxt *ctxt = rqstp->rq_xprt_ctxt;
-	struct svc_xprt *xprt = rqstp->rq_xprt;
+	struct svc_rdma_recv_ctxt *ctxt = vctxt;
 	struct svcxprt_rdma *rdma =
 		container_of(xprt, struct svcxprt_rdma, sc_xprt);
 
-	rqstp->rq_xprt_ctxt = NULL;
 	if (ctxt)
 		svc_rdma_recv_ctxt_put(rdma, ctxt);
 }
net/sunrpc/xprtrdma/svc_rdma_transport.c +1 -1
index 94b20fb471356..f776f0cb471f0 100644
@@ -81,7 +81,7 @@ static const struct svc_xprt_ops svc_rdma_ops = {
 	.xpo_recvfrom = svc_rdma_recvfrom,
 	.xpo_sendto = svc_rdma_sendto,
 	.xpo_result_payload = svc_rdma_result_payload,
-	.xpo_release_rqst = svc_rdma_release_rqst,
+	.xpo_release_ctxt = svc_rdma_release_ctxt,
 	.xpo_detach = svc_rdma_detach,
 	.xpo_free = svc_rdma_free,
 	.xpo_has_wspace = svc_rdma_has_wspace,
Signature · 1 line
2.39.2
author maintainer
#188 lore raw

PATCH 188/292 SUNRPC: Fix trace_svc_register() call site

In reply to Greg Kroah-Hartman

From: Chuck Lever <[email protected]>

[ Upstream commit 07a27305938559fb35f7a46fb90a5e37728bdee6 ]

The trace event recorded incorrect values for the registered family, protocol, and port because the arguments are in the wrong order.

Fixes b4af59328c25 ("SUNRPC: Trace server-side rpcbind registration events")
---
 net/sunrpc/svc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/sunrpc/svc.c +1 -1
index 9ee32e06f877e..9b0b21cccca9a 100644
@@ -1007,7 +1007,7 @@ static int __svc_register(struct net *net, const char *progname,
 #endif
 	}
 
-	trace_svc_register(progname, version, protocol, port, family, error);
+	trace_svc_register(progname, version, family, protocol, port, error);
 	return error;
 }
 
Signature · 1 line
2.39.2
author maintainer
#189 lore raw

PATCH 189/292 ASoC: mediatek: mt8186: Fix use-after-free in driver remove path

In reply to Greg Kroah-Hartman

From: Douglas Anderson <[email protected]>

[ Upstream commit a93d2afd3f77a7331271a0f25c6a11003db69b3c ]

When devm runs function in the "remove" path for a device it runs them in the reverse order. That means that if you have parts of your driver that aren't using devm or are using "roll your own" devm w/ devm_add_action_or_reset() you need to keep that in mind.

The mt8186 audio driver didn't quite get this right. Specifically, in mt8186_init_clock() it called mt8186_audsys_clk_register() and then went on to call a bunch of other devm function. The caller of mt8186_init_clock() used devm_add_action_or_reset() to call mt8186_deinit_clock() but, because of the intervening devm functions, the order was wrong.

Specifically at probe time, the order was:
1. mt8186_audsys_clk_register()
2. afe_priv->clk = devm_kcalloc(...)
3. afe_priv->clk[i] = devm_clk_get(...)

At remove time, the order (which should have been 3, 2, 1) was:
1. mt8186_audsys_clk_unregister()
3. Free all of afe_priv->clk[i]
2. Free afe_priv->clk

The above seemed to be causing a use-after-free. Luckily, it's easy to fix this by simply using devm more correctly. Let's move the devm_add_action_or_reset() to the right place. In addition to fixing the use-after-free, code inspection shows that this fixes a leak (missing call to mt8186_audsys_clk_unregister()) that would have happened if any of the syscon_regmap_lookup_by_phandle() calls in mt8186_init_clock() had failed.

Fixes 55b423d5623c ("ASoC: mediatek: mt8186: support audio clock control in platform driver")
---
 sound/soc/mediatek/mt8186/mt8186-afe-clk.c    |  6 ---
 sound/soc/mediatek/mt8186/mt8186-afe-clk.h    |  1 -
 sound/soc/mediatek/mt8186/mt8186-afe-pcm.c    |  4 --
 sound/soc/mediatek/mt8186/mt8186-audsys-clk.c | 46 ++++++++++---------
 sound/soc/mediatek/mt8186/mt8186-audsys-clk.h |  1 -
 5 files changed, 24 insertions(+), 34 deletions(-)
sound/soc/mediatek/mt8186/mt8186-afe-clk.c +0 -6
index a6b4f29049bbc..539e3a023bc4e 100644
@@ -644,9 +644,3 @@ int mt8186_init_clock(struct mtk_base_afe *afe)
 
 	return 0;
 }
-
-void mt8186_deinit_clock(void *priv)
-{
-	struct mtk_base_afe *afe = priv;
-	mt8186_audsys_clk_unregister(afe);
-}
sound/soc/mediatek/mt8186/mt8186-afe-clk.h +0 -1
index d5988717d8f2d..a9d59e506d9af 100644
@@ -81,7 +81,6 @@ enum {
 struct mtk_base_afe;
 int mt8186_set_audio_int_bus_parent(struct mtk_base_afe *afe, int clk_id);
 int mt8186_init_clock(struct mtk_base_afe *afe);
-void mt8186_deinit_clock(void *priv);
 int mt8186_afe_enable_cgs(struct mtk_base_afe *afe);
 void mt8186_afe_disable_cgs(struct mtk_base_afe *afe);
 int mt8186_afe_enable_clock(struct mtk_base_afe *afe);
sound/soc/mediatek/mt8186/mt8186-afe-pcm.c +0 -4
index d7e94e6a19c70..0e3792ccd49f6 100644
@@ -2847,10 +2847,6 @@ static int mt8186_afe_pcm_dev_probe(struct platform_device *pdev)
 		return ret;
 	}
 
-	ret = devm_add_action_or_reset(dev, mt8186_deinit_clock, (void *)afe);
-	if (ret)
-		return ret;
-
 	/* init memif */
 	afe->memif_32bit_supported = 0;
 	afe->memif_size = MT8186_MEMIF_NUM;
sound/soc/mediatek/mt8186/mt8186-audsys-clk.c +24 -22

57 lines not on this page · raw message

sound/soc/mediatek/mt8186/mt8186-audsys-clk.h +0 -1
index b8d6a06e11e8d..897a2914dc191 100644
@@ -10,6 +10,5 @@
 #define _MT8186_AUDSYS_CLK_H_
 
 int mt8186_audsys_clk_register(struct mtk_base_afe *afe);
-void mt8186_audsys_clk_unregister(struct mtk_base_afe *afe);
 
 #endif
Signature · 1 line
2.39.2
author maintainer
#190 lore raw

PATCH 190/292 ASoC: SOF: topology: Fix logic for copying tuples

In reply to Greg Kroah-Hartman

From: Ranjani Sridharan <[email protected]>

[ Upstream commit 41c5305cc3d827d2ea686533777a285176ae01a0 ]

Topology could have more instances of the tokens being searched for than the number of sets that need to be copied. Stop copying token after the limit of number of token instances has been reached. This worked before only by chance as we had allocated more size for the tuples array than the number of actual tokens being parsed.

---
 sound/soc/sof/topology.c | 4 ++++
 1 file changed, 4 insertions(+)
sound/soc/sof/topology.c +4 -0
index 6a0e7f3b50234..872e44408298f 100644
@@ -545,6 +545,10 @@ static int sof_copy_tuples(struct snd_sof_dev *sdev, struct snd_soc_tplg_vendor_
 				if (*num_copied_tuples == tuples_size)
 					return 0;
 			}
+
+			/* stop when we've found the required token instances */
+			if (found == num_tokens * token_instance_num)
+				return 0;
 		}
 
 		/* next array */
Signature · 1 line
2.39.2
author maintainer
#191 lore raw

PATCH 191/292 drm/exynos: fix g2d_open/close helper function definitions

In reply to Greg Kroah-Hartman

From: Arnd Bergmann <[email protected]>

[ Upstream commit 2ef0785b30bd6549ddbc124979f1b6596e065ae2 ]

The empty stub functions are defined as global functions, which causes a warning because of missing prototypes:

drivers/gpu/drm/exynos/exynos_drm_g2d.h:37:5: error: no previous prototype for 'g2d_open' drivers/gpu/drm/exynos/exynos_drm_g2d.h:42:5: error: no previous prototype for 'g2d_close'

Mark them as 'static inline' to avoid the warning and to make them behave as intended.

Fixes eb4d9796fa34 ("drm/exynos: g2d: Convert to driver component API")
---
 drivers/gpu/drm/exynos/exynos_drm_g2d.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/exynos/exynos_drm_g2d.h +2 -2
index 74ea3c26deadc..1a5ae781b56c6 100644
@@ -34,11 +34,11 @@ static inline int exynos_g2d_exec_ioctl(struct drm_device *dev, void *data,
 	return -ENODEV;
 }
 
-int g2d_open(struct drm_device *drm_dev, struct drm_file *file)
+static inline int g2d_open(struct drm_device *drm_dev, struct drm_file *file)
 {
 	return 0;
 }
 
-void g2d_close(struct drm_device *drm_dev, struct drm_file *file)
+static inline void g2d_close(struct drm_device *drm_dev, struct drm_file *file)
 { }
 #endif
Signature · 1 line
2.39.2
author maintainer
#192 lore raw

PATCH 192/292 net: nsh: Use correct mac_offset to unwind gso skb in nsh_gso_segment()

In reply to Greg Kroah-Hartman

From: Dong Chenchen <[email protected]>

[ Upstream commit c83b49383b595be50647f0c764a48c78b5f3c4f8 ]

As the call trace shows, skb_panic was caused by wrong skb->mac_header in nsh_gso_segment():

invalid opcode: 0000 [#1] PREEMPT SMP KASAN PTI

CPU: 3 PID: 2737 Comm: syz Not tainted 6.3.0-next-20230505 #1
RIP: 0010:skb_panic+0xda/0xe0
call Trace:
 skb_push+0x91/0xa0
 nsh_gso_segment+0x4f3/0x570
 skb_mac_gso_segment+0x19e/0x270
 __skb_gso_segment+0x1e8/0x3c0
 validate_xmit_skb+0x452/0x890
 validate_xmit_skb_list+0x99/0xd0
 sch_direct_xmit+0x294/0x7c0
 __dev_queue_xmit+0x16f0/0x1d70
 packet_xmit+0x185/0x210
 packet_snd+0xc15/0x1170
 packet_sendmsg+0x7b/0xa0
 sock_sendmsg+0x14f/0x160

The root cause is:
nsh_gso_segment() use skb->network_header - nhoff to reset mac_header in skb_gso_error_unwind() if inner-layer protocol gso fails. However, skb->network_header may be reset by inner-layer protocol gso function e.g. mpls_gso_segment. skb->mac_header reset by the inaccurate network_header will be larger than skb headroom.

nsh_gso_segment

    nhoff = skb->network_header - skb->mac_header;
    __skb_pull(skb,nsh_len)
    skb_mac_gso_segment
        mpls_gso_segment
            skb_reset_network_header(skb);//skb->network_header+=nsh_len
            return -EINVAL;
    skb_gso_error_unwind
        skb_push(skb, nsh_len);
        skb->mac_header = skb->network_header - nhoff;
        // skb->mac_header > skb->headroom, cause skb_push panic

Use correct mac_offset to restore mac_header and get rid of nhoff.

Fixes c411ed854584 ("nsh: add GSO support")
Reported-by [email protected]
---
 net/nsh/nsh.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)
net/nsh/nsh.c +3 -5
index e9ca007718b7e..0f23e5e8e03eb 100644
@@ -77,13 +77,12 @@ static struct sk_buff *nsh_gso_segment(struct sk_buff *skb,
 				       netdev_features_t features)
 {
 	struct sk_buff *segs = ERR_PTR(-EINVAL);
+	u16 mac_offset = skb->mac_header;
 	unsigned int nsh_len, mac_len;
 	__be16 proto;
-	int nhoff;
 
 	skb_reset_network_header(skb);
 
-	nhoff = skb->network_header - skb->mac_header;
 	mac_len = skb->mac_len;
 
 	if (unlikely(!pskb_may_pull(skb, NSH_BASE_HDR_LEN)))
@@ -108,15 +107,14 @@ static struct sk_buff *nsh_gso_segment(struct sk_buff *skb,
 	segs = skb_mac_gso_segment(skb, features);
 	if (IS_ERR_OR_NULL(segs)) {
 		skb_gso_error_unwind(skb, htons(ETH_P_NSH), nsh_len,
-				     skb->network_header - nhoff,
-				     mac_len);
+				     mac_offset, mac_len);
 		goto out;
 	}
 
 	for (skb = segs; skb; skb = skb->next) {
 		skb->protocol = htons(ETH_P_NSH);
 		__skb_push(skb, nsh_len);
-		skb_set_mac_header(skb, -nhoff);
+		skb->mac_header = mac_offset;
 		skb->network_header = skb->mac_header + mac_len;
 		skb->mac_len = mac_len;
 	}
Signature · 1 line
2.39.2
author maintainer
#193 lore raw

PATCH 193/292 virtio-net: Maintain reverse cleanup order

In reply to Greg Kroah-Hartman

From: Parav Pandit <[email protected]>

[ Upstream commit 27369c9c2b722617063d6b80c758ab153f1d95d4 ]

To easily audit the code, better to keep the device stop() sequence to be mirror of the device open() sequence.

Stable-dep-of: 5306623a9826 ("virtio_net: Fix error unwinding of XDP initialization")

---
 drivers/net/virtio_net.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/virtio_net.c +1 -1
index 9a612b13b4e46..08a23ba3d68a2 100644
@@ -2158,9 +2158,9 @@ static int virtnet_close(struct net_device *dev)
 	cancel_delayed_work_sync(&vi->refill);
 
 	for (i = 0; i < vi->max_queue_pairs; i++) {
+		virtnet_napi_tx_disable(&vi->sq[i].napi);
 		napi_disable(&vi->rq[i].napi);
 		xdp_rxq_info_unreg(&vi->rq[i].xdp_rxq);
-		virtnet_napi_tx_disable(&vi->sq[i].napi);
 	}
 
 	return 0;
Signature · 1 line
2.39.2
author maintainer
#194 lore raw

PATCH 194/292 virtio_net: Fix error unwinding of XDP initialization

In reply to Greg Kroah-Hartman

From: Feng Liu <[email protected]>

[ Upstream commit 5306623a9826aa7d63b32c6a3803c798a765474d ]

When initializing XDP in virtnet_open(), some rq xdp initialization may hit an error causing net device open failed. However, previous rqs have already initialized XDP and enabled NAPI, which is not the expected behavior. Need to roll back the previous rq initialization to avoid leaks in error unwinding of init code.

Also extract helper functions of disable and enable queue pairs. Use newly introduced disable helper function in error unwinding and virtnet_close. Use enable helper function in virtnet_open.

---
 drivers/net/virtio_net.c | 61 +++++++++++++++++++++++++++++-----------
 1 file changed, 44 insertions(+), 17 deletions(-)
drivers/net/virtio_net.c +44 -17
index 08a23ba3d68a2..47788f0935514 100644
@@ -1697,6 +1697,38 @@ static int virtnet_poll(struct napi_struct *napi, int budget)
 	return received;
 }
 
+static void virtnet_disable_queue_pair(struct virtnet_info *vi, int qp_index)
+{
+	virtnet_napi_tx_disable(&vi->sq[qp_index].napi);
+	napi_disable(&vi->rq[qp_index].napi);
+	xdp_rxq_info_unreg(&vi->rq[qp_index].xdp_rxq);
+}
+
+static int virtnet_enable_queue_pair(struct virtnet_info *vi, int qp_index)
+{
+	struct net_device *dev = vi->dev;
+	int err;
+
+	err = xdp_rxq_info_reg(&vi->rq[qp_index].xdp_rxq, dev, qp_index,
+			       vi->rq[qp_index].napi.napi_id);
+	if (err < 0)
+		return err;
+
+	err = xdp_rxq_info_reg_mem_model(&vi->rq[qp_index].xdp_rxq,
+					 MEM_TYPE_PAGE_SHARED, NULL);
+	if (err < 0)
+		goto err_xdp_reg_mem_model;
+
+	virtnet_napi_enable(vi->rq[qp_index].vq, &vi->rq[qp_index].napi);
+	virtnet_napi_tx_enable(vi, vi->sq[qp_index].vq, &vi->sq[qp_index].napi);
+
+	return 0;
+
+err_xdp_reg_mem_model:
+	xdp_rxq_info_unreg(&vi->rq[qp_index].xdp_rxq);
+	return err;
+}
+
 static int virtnet_open(struct net_device *dev)
 {
 	struct virtnet_info *vi = netdev_priv(dev);
@@ -1710,22 +1742,20 @@ static int virtnet_open(struct net_device *dev)
 			if (!try_fill_recv(vi, &vi->rq[i], GFP_KERNEL))
 				schedule_delayed_work(&vi->refill, 0);
 
-		err = xdp_rxq_info_reg(&vi->rq[i].xdp_rxq, dev, i, vi->rq[i].napi.napi_id);
+		err = virtnet_enable_queue_pair(vi, i);
 		if (err < 0)
-			return err;
-
-		err = xdp_rxq_info_reg_mem_model(&vi->rq[i].xdp_rxq,
-						 MEM_TYPE_PAGE_SHARED, NULL);
-		if (err < 0) {
-			xdp_rxq_info_unreg(&vi->rq[i].xdp_rxq);
-			return err;
-		}
-
-		virtnet_napi_enable(vi->rq[i].vq, &vi->rq[i].napi);
-		virtnet_napi_tx_enable(vi, vi->sq[i].vq, &vi->sq[i].napi);
+			goto err_enable_qp;
 	}
 
 	return 0;
+
+err_enable_qp:
+	disable_delayed_refill(vi);
+	cancel_delayed_work_sync(&vi->refill);
+
+	for (i--; i >= 0; i--)
+		virtnet_disable_queue_pair(vi, i);
+	return err;
 }
 
 static int virtnet_poll_tx(struct napi_struct *napi, int budget)
@@ -2157,11 +2187,8 @@ static int virtnet_close(struct net_device *dev)
 	/* Make sure refill_work doesn't re-enable napi! */
 	cancel_delayed_work_sync(&vi->refill);
 
-	for (i = 0; i < vi->max_queue_pairs; i++) {
-		virtnet_napi_tx_disable(&vi->sq[i].napi);
-		napi_disable(&vi->rq[i].napi);
-		xdp_rxq_info_unreg(&vi->rq[i].xdp_rxq);
-	}
+	for (i = 0; i < vi->max_queue_pairs; i++)
+		virtnet_disable_queue_pair(vi, i);
 
 	return 0;
 }
Signature · 1 line
2.39.2
author maintainer
#195 lore raw

PATCH 195/292 tipc: add tipc_bearer_min_mtu to calculate min mtu

In reply to Greg Kroah-Hartman

From: Xin Long <[email protected]>

[ Upstream commit 3ae6d66b605be604644d4bb5708a7ffd9cf1abe8 ]

As different media may requires different min mtu, and even the same media with different net family requires different min mtu, add tipc_bearer_min_mtu() to calculate min mtu accordingly.

This API will be used to check the new mtu when doing the link mtu negotiation in the next patch.

Stable-dep-of: 56077b56cd3f ("tipc: do not update mtu if msg_max is too small in mtu negotiation")

---
 net/tipc/bearer.c    | 13 +++++++++++++
 net/tipc/bearer.h    |  3 +++
 net/tipc/udp_media.c |  5 +++--
 3 files changed, 19 insertions(+), 2 deletions(-)
net/tipc/bearer.c +13 -0
index 35cac7733fd3a..0e9a29e1536b7 100644
@@ -541,6 +541,19 @@ int tipc_bearer_mtu(struct net *net, u32 bearer_id)
 	return mtu;
 }
 
+int tipc_bearer_min_mtu(struct net *net, u32 bearer_id)
+{
+	int mtu = TIPC_MIN_BEARER_MTU;
+	struct tipc_bearer *b;
+
+	rcu_read_lock();
+	b = bearer_get(net, bearer_id);
+	if (b)
+		mtu += b->encap_hlen;
+	rcu_read_unlock();
+	return mtu;
+}
+
 /* tipc_bearer_xmit_skb - sends buffer to destination over bearer
  */
 void tipc_bearer_xmit_skb(struct net *net, u32 bearer_id,
net/tipc/bearer.h +3 -0
index 490ad6e5f7a3c..bd0cc5c287ef8 100644
@@ -146,6 +146,7 @@ struct tipc_media {
  * @identity: array index of this bearer within TIPC bearer array
  * @disc: ptr to link setup request
  * @net_plane: network plane ('A' through 'H') currently associated with bearer
+ * @encap_hlen: encap headers length
  * @up: bearer up flag (bit 0)
  * @refcnt: tipc_bearer reference counter
  *
@@ -170,6 +171,7 @@ struct tipc_bearer {
 	u32 identity;
 	struct tipc_discoverer *disc;
 	char net_plane;
+	u16 encap_hlen;
 	unsigned long up;
 	refcount_t refcnt;
 };
@@ -232,6 +234,7 @@ int tipc_bearer_setup(void);
 void tipc_bearer_cleanup(void);
 void tipc_bearer_stop(struct net *net);
 int tipc_bearer_mtu(struct net *net, u32 bearer_id);
+int tipc_bearer_min_mtu(struct net *net, u32 bearer_id);
 bool tipc_bearer_bcast_support(struct net *net, u32 bearer_id);
 void tipc_bearer_xmit_skb(struct net *net, u32 bearer_id,
 			  struct sk_buff *skb,
net/tipc/udp_media.c +3 -2
index c2bb818704c8f..0a85244fd6188 100644
@@ -738,8 +738,8 @@ static int tipc_udp_enable(struct net *net, struct tipc_bearer *b,
 			udp_conf.local_ip.s_addr = local.ipv4.s_addr;
 		udp_conf.use_udp_checksums = false;
 		ub->ifindex = dev->ifindex;
-		if (tipc_mtu_bad(dev, sizeof(struct iphdr) +
-				      sizeof(struct udphdr))) {
+		b->encap_hlen = sizeof(struct iphdr) + sizeof(struct udphdr);
+		if (tipc_mtu_bad(dev, b->encap_hlen)) {
 			err = -EINVAL;
 			goto err;
 		}
@@ -760,6 +760,7 @@ static int tipc_udp_enable(struct net *net, struct tipc_bearer *b,
 		else
 			udp_conf.local_ip6 = local.ipv6;
 		ub->ifindex = dev->ifindex;
+		b->encap_hlen = sizeof(struct ipv6hdr) + sizeof(struct udphdr);
 		b->mtu = 1280;
 #endif
 	} else {
Signature · 1 line
2.39.2
author maintainer
#196 lore raw

PATCH 196/292 tipc: do not update mtu if msg_max is too small in mtu negotiation

In reply to Greg Kroah-Hartman

From: Xin Long <[email protected]>

[ Upstream commit 56077b56cd3fb78e1c8619e29581ba25a5c55e86 ]

When doing link mtu negotiation, a malicious peer may send Activate msg with a very small mtu, e.g. 4 in Shuang's testing, without checking for the minimum mtu, l->mtu will be set to 4 in tipc_link_proto_rcv(), then n->links[bearer_id].mtu is set to 4294967228, which is a overflow of '4 - INT_H_SIZE - EMSG_OVERHEAD' in tipc_link_mss().

With tipc_link.mtu = 4, tipc_link_xmit() kept printing the warning:

tipc: Too large msg, purging xmit list 1 5 0 40 4!
tipc: Too large msg, purging xmit list 1 15 0 60 4!

And with tipc_link_entry.mtu 4294967228, a huge skb was allocated in named_distribute(), and when purging it in tipc_link_xmit(), a crash was even caused:

  general protection fault, probably for non-canonical address 0x2100001011000dd: 0000 [#1] PREEMPT SMP PTI
  CPU: 0 PID: 0 Comm: swapper/0 Kdump: loaded Not tainted 6.3.0.neta #19
  RIP: 0010:kfree_skb_list_reason+0x7e/0x1f0
  Call Trace:
   <IRQ>
   skb_release_data+0xf9/0x1d0
   kfree_skb_reason+0x40/0x100
   tipc_link_xmit+0x57a/0x740 [tipc]
   tipc_node_xmit+0x16c/0x5c0 [tipc]
   tipc_named_node_up+0x27f/0x2c0 [tipc]
   tipc_node_write_unlock+0x149/0x170 [tipc]
   tipc_rcv+0x608/0x740 [tipc]
   tipc_udp_recv+0xdc/0x1f0 [tipc]
   udp_queue_rcv_one_skb+0x33e/0x620
   udp_unicast_rcv_skb.isra.72+0x75/0x90
   __udp4_lib_rcv+0x56d/0xc20
   ip_protocol_deliver_rcu+0x100/0x2d0

This patch fixes it by checking the new mtu against tipc_bearer_min_mtu(), and not updating mtu if it is too small.

Fixes ed193ece2649 ("tipc: simplify link mtu negotiation")
---
 net/tipc/link.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
net/tipc/link.c +6 -3
index b3ce24823f503..2eff1c7949cbc 100644
@@ -2200,7 +2200,7 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
 	struct tipc_msg *hdr = buf_msg(skb);
 	struct tipc_gap_ack_blks *ga = NULL;
 	bool reply = msg_probe(hdr), retransmitted = false;
-	u32 dlen = msg_data_sz(hdr), glen = 0;
+	u32 dlen = msg_data_sz(hdr), glen = 0, msg_max;
 	u16 peers_snd_nxt =  msg_next_sent(hdr);
 	u16 peers_tol = msg_link_tolerance(hdr);
 	u16 peers_prio = msg_linkprio(hdr);
@@ -2239,6 +2239,9 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
 	switch (mtyp) {
 	case RESET_MSG:
 	case ACTIVATE_MSG:
+		msg_max = msg_max_pkt(hdr);
+		if (msg_max < tipc_bearer_min_mtu(l->net, l->bearer_id))
+			break;
 		/* Complete own link name with peer's interface name */
 		if_name =  strrchr(l->name, ':') + 1;
 		if (sizeof(l->name) - (if_name - l->name) <= TIPC_MAX_IF_NAME)
@@ -2283,8 +2286,8 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
 		l->peer_session = msg_session(hdr);
 		l->in_session = true;
 		l->peer_bearer_id = msg_bearer_id(hdr);
-		if (l->mtu > msg_max_pkt(hdr))
-			l->mtu = msg_max_pkt(hdr);
+		if (l->mtu > msg_max)
+			l->mtu = msg_max;
 		break;
 
 	case STATE_MSG:
Signature · 1 line
2.39.2
author maintainer
#197 lore raw

PATCH 197/292 tipc: check the bearer min mtu properly when setting it by netlink

In reply to Greg Kroah-Hartman

From: Xin Long <[email protected]>

[ Upstream commit 35a089b5d793d2bfd2cc7cfa6104545184de2ce7 ]

Checking the bearer min mtu with tipc_udp_mtu_bad() only works for IPv4 UDP bearer, and IPv6 UDP bearer has a different value for the min mtu. This patch checks with encap_hlen + TIPC_MIN_BEARER_MTU for min mtu, which works for both IPv4 and IPv6 UDP bearer.

Note that tipc_udp_mtu_bad() is still used to check media min mtu in __tipc_nl_media_set(), as m->mtu currently is only used by the IPv4 UDP bearer as its default mtu value.

Fixes 682cd3cf946b ("tipc: confgiure and apply UDP bearer MTU on running links")
---
 net/tipc/bearer.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
net/tipc/bearer.c +2 -2
index 0e9a29e1536b7..53881406e2006 100644
@@ -1151,8 +1151,8 @@ int __tipc_nl_bearer_set(struct sk_buff *skb, struct genl_info *info)
 				return -EINVAL;
 			}
 #ifdef CONFIG_TIPC_MEDIA_UDP
-			if (tipc_udp_mtu_bad(nla_get_u32
-					     (props[TIPC_NLA_PROP_MTU]))) {
+			if (nla_get_u32(props[TIPC_NLA_PROP_MTU]) <
+			    b->encap_hlen + TIPC_MIN_BEARER_MTU) {
 				NL_SET_ERR_MSG(info->extack,
 					       "MTU value is out-of-range");
 				return -EINVAL;
Signature · 1 line
2.39.2
author maintainer
#198 lore raw

PATCH 198/292 s390/cio: include subchannels without devices also for evaluation

In reply to Greg Kroah-Hartman

From: Vineeth Vijayan <[email protected]>

[ Upstream commit b1b0d5aec1cf9f9a900a14964f869c68688d923e ]

Currently when the new channel-path is enabled, we do evaluation only on the subchannels with a device connected on it. This is because, in the past, if the device in the subchannel is not working or not available, we used to unregister the subchannels. But, from the 'commit 2297791c92d0 ("s390/cio: dont unregister subchannel from child-drivers")' we allow subchannels with or without an active device connected on it. So, when we do the io_subchannel_verify, make sure that, we are evaluating the subchannels without any device too.

Fixes 2297791c92d0 ("s390/cio: dont unregister subchannel from child-drivers")
---
 drivers/s390/cio/device.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/s390/cio/device.c +2 -0
index 3b1cd0c96a74b..ba4c69226c337 100644
@@ -1102,6 +1102,8 @@ static void io_subchannel_verify(struct subchannel *sch)
 	cdev = sch_get_cdev(sch);
 	if (cdev)
 		dev_fsm_event(cdev, DEV_EVENT_VERIFY);
+	else
+		css_schedule_eval(sch->schid);
 }
 
 static void io_subchannel_terminate_path(struct subchannel *sch, u8 mask)
Signature · 1 line
2.39.2
author maintainer
#199 lore raw

PATCH 199/292 can: dev: fix missing CAN XL support in can_put_echo_skb()

In reply to Greg Kroah-Hartman

From: Oliver Hartkopp <[email protected]>

[ Upstream commit 6bffdc38f9935bae49f980448f3f6be2dada0564 ]

can_put_echo_skb() checks for the enabled IFF_ECHO flag and the correct ETH_P type of the given skbuff. When implementing the CAN XL support the new check for ETH_P_CANXL has been forgotten.

Fixes fb08cba12b52 ("can: canxl: update CAN infrastructure for CAN XL frames")
---
 drivers/net/can/dev/skb.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/can/dev/skb.c +2 -1
index 241ec636e91fd..f6d05b3ef59ab 100644
@@ -54,7 +54,8 @@ int can_put_echo_skb(struct sk_buff *skb, struct net_device *dev,
 	/* check flag whether this packet has to be looped back */
 	if (!(dev->flags & IFF_ECHO) ||
 	    (skb->protocol != htons(ETH_P_CAN) &&
-	     skb->protocol != htons(ETH_P_CANFD))) {
+	     skb->protocol != htons(ETH_P_CANFD) &&
+	     skb->protocol != htons(ETH_P_CANXL))) {
 		kfree_skb(skb);
 		return 0;
 	}
Signature · 1 line
2.39.2
author maintainer
#200 lore raw

PATCH 200/292 net: bcmgenet: Remove phy_stop() from bcmgenet_netif_stop()

In reply to Greg Kroah-Hartman

From: Florian Fainelli <[email protected]>

[ Upstream commit 93e0401e0fc0c54b0ac05b687cd135c2ac38187c ]

The call to phy_stop() races with the later call to phy_disconnect(), resulting in concurrent phy_suspend() calls being run from different CPUs. The final call to phy_disconnect() ensures that the PHY is stopped and suspended, too.

Fixes c96e731c93ff ("net: bcmgenet: connect and disconnect from the PHY state machine")
---
 drivers/net/ethernet/broadcom/genet/bcmgenet.c | 1 -
 1 file changed, 1 deletion(-)
drivers/net/ethernet/broadcom/genet/bcmgenet.c +0 -1
index f679ed54b3ef2..9860fd66f3bca 100644
@@ -3460,7 +3460,6 @@ static void bcmgenet_netif_stop(struct net_device *dev)
 	/* Disable MAC transmit. TX DMA disabled must be done before this */
 	umac_enable_set(priv, CMD_TX_EN, false);
 
-	phy_stop(dev->phydev);
 	bcmgenet_disable_rx_napi(priv);
 	bcmgenet_intr_disable(priv);
 
Signature · 1 line
2.39.2
author maintainer
#201 lore raw

PATCH 201/292 net: bcmgenet: Restore phy_stop() depending upon suspend/close

In reply to Greg Kroah-Hartman

From: Florian Fainelli <[email protected]>

[ Upstream commit 225c657945c4a6307741cb3cc89467eadcc26e9b ]

Removing the phy_stop() from bcmgenet_netif_stop() ended up causing warnings from the PHY library that phy_start() is called from the RUNNING state since we are no longer stopping the PHY state machine during bcmgenet_suspend().

Restore the call to phy_stop() but make it conditional on being called from the close or suspend path.

Fixes c96e731c93ff ("net: bcmgenet: connect and disconnect from the PHY state machine")
Fixes 93e0401e0fc0 ("net: bcmgenet: Remove phy_stop() from bcmgenet_netif_stop()")
---
 drivers/net/ethernet/broadcom/genet/bcmgenet.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)
drivers/net/ethernet/broadcom/genet/bcmgenet.c +5 -3
index 9860fd66f3bca..4da2becfa950c 100644
@@ -3445,7 +3445,7 @@ static int bcmgenet_open(struct net_device *dev)
 	return ret;
 }
 
-static void bcmgenet_netif_stop(struct net_device *dev)
+static void bcmgenet_netif_stop(struct net_device *dev, bool stop_phy)
 {
 	struct bcmgenet_priv *priv = netdev_priv(dev);
 
@@ -3460,6 +3460,8 @@ static void bcmgenet_netif_stop(struct net_device *dev)
 	/* Disable MAC transmit. TX DMA disabled must be done before this */
 	umac_enable_set(priv, CMD_TX_EN, false);
 
+	if (stop_phy)
+		phy_stop(dev->phydev);
 	bcmgenet_disable_rx_napi(priv);
 	bcmgenet_intr_disable(priv);
 
@@ -3480,7 +3482,7 @@ static int bcmgenet_close(struct net_device *dev)
 
 	netif_dbg(priv, ifdown, dev, "bcmgenet_close\n");
 
-	bcmgenet_netif_stop(dev);
+	bcmgenet_netif_stop(dev, false);
 
 	/* Really kill the PHY state machine and disconnect from it */
 	phy_disconnect(dev->phydev);
@@ -4298,7 +4300,7 @@ static int bcmgenet_suspend(struct device *d)
 
 	netif_device_detach(dev);
 
-	bcmgenet_netif_stop(dev);
+	bcmgenet_netif_stop(dev, true);
 
 	if (!device_may_wakeup(d))
 		phy_suspend(dev->phydev);
Signature · 1 line
2.39.2
author maintainer
#202 lore raw

PATCH 202/292 ice: introduce clear_reset_state operation

In reply to Greg Kroah-Hartman

From: Jacob Keller <[email protected]>

[ Upstream commit fa4a15c85c849e92257da6dbffeb1e3a6399fd7b ]

When hardware is reset, the VF relies on the VFGEN_RSTAT register to detect when the VF is finished resetting. This is a tri-state register where 0 indicates a reset is in progress, 1 indicates the hardware is done resetting, and 2 indicates that the software is done resetting.

Currently the PF driver relies on the device hardware resetting VFGEN_RSTAT when a global reset occurs. This works ok, but it does mean that the VF might not immediately notice a reset when the driver first detects that the global reset is occurring.

This is also problematic for Scalable IOV, because there is no read/write equivalent VFGEN_RSTAT register for the Scalable VSI type. Instead, the Scalable IOV VFs will need to emulate this register.

To support this, introduce a new VF operation, clear_reset_state, which is called when the PF driver first detects a global reset. The Single Root IOV implementation can just write to VFGEN_RSTAT to ensure it's cleared immediately, without waiting for the actual hardware reset to begin. The Scalable IOV implementation will use this as part of its tracking of the reset status to allow properly reporting the emulated VFGEN_RSTAT to the VF driver.

Stable-dep-of: 7255355a0636 ("ice: Fix ice VF reset during iavf initialization")

---
 drivers/net/ethernet/intel/ice/ice_main.c   |  2 +-
 drivers/net/ethernet/intel/ice/ice_sriov.c  | 16 ++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_vf_lib.c | 12 +++++++++++-
 drivers/net/ethernet/intel/ice/ice_vf_lib.h |  5 +++--
 4 files changed, 31 insertions(+), 4 deletions(-)
drivers/net/ethernet/intel/ice/ice_main.c +1 -1
index cfc57cfc46e42..6a50f8ba3940c 100644
@@ -573,7 +573,7 @@ ice_prepare_for_reset(struct ice_pf *pf, enum ice_reset_req reset_type)
 	/* Disable VFs until reset is completed */
 	mutex_lock(&pf->vfs.table_lock);
 	ice_for_each_vf(pf, bkt, vf)
-		ice_set_vf_state_qs_dis(vf);
+		ice_set_vf_state_dis(vf);
 	mutex_unlock(&pf->vfs.table_lock);
 
 	if (ice_is_eswitch_mode_switchdev(pf)) {
drivers/net/ethernet/intel/ice/ice_sriov.c +16 -0
index b3849bc3d4fc6..b719e9a771e36 100644
@@ -696,6 +696,21 @@ static void ice_sriov_free_vf(struct ice_vf *vf)
 	kfree_rcu(vf, rcu);
 }
 
+/**
+ * ice_sriov_clear_reset_state - clears VF Reset status register
+ * @vf: the vf to configure
+ */
+static void ice_sriov_clear_reset_state(struct ice_vf *vf)
+{
+	struct ice_hw *hw = &vf->pf->hw;
+
+	/* Clear the reset status register so that VF immediately sees that
+	 * the device is resetting, even if hardware hasn't yet gotten around
+	 * to clearing VFGEN_RSTAT for us.
+	 */
+	wr32(hw, VFGEN_RSTAT(vf->vf_id), VIRTCHNL_VFR_INPROGRESS);
+}
+
 /**
  * ice_sriov_clear_mbx_register - clears SRIOV VF's mailbox registers
  * @vf: the vf to configure
@@ -835,6 +850,7 @@ static void ice_sriov_post_vsi_rebuild(struct ice_vf *vf)
 static const struct ice_vf_ops ice_sriov_vf_ops = {
 	.reset_type = ICE_VF_RESET,
 	.free = ice_sriov_free_vf,
+	.clear_reset_state = ice_sriov_clear_reset_state,
 	.clear_mbx_register = ice_sriov_clear_mbx_register,
 	.trigger_reset_register = ice_sriov_trigger_reset_register,
 	.poll_reset_status = ice_sriov_poll_reset_status,
drivers/net/ethernet/intel/ice/ice_vf_lib.c +11 -1
index 1c51778db951b..86abbcb480d9d 100644
@@ -673,7 +673,7 @@ int ice_reset_vf(struct ice_vf *vf, u32 flags)
  * ice_set_vf_state_qs_dis - Set VF queues state to disabled
  * @vf: pointer to the VF structure
  */
-void ice_set_vf_state_qs_dis(struct ice_vf *vf)
+static void ice_set_vf_state_qs_dis(struct ice_vf *vf)
 {
 	/* Clear Rx/Tx enabled queues flag */
 	bitmap_zero(vf->txq_ena, ICE_MAX_RSS_QS_PER_VF);
@@ -681,6 +681,16 @@ void ice_set_vf_state_qs_dis(struct ice_vf *vf)
 	clear_bit(ICE_VF_STATE_QS_ENA, vf->vf_states);
 }
 
+/**
+ * ice_set_vf_state_dis - Set VF state to disabled
+ * @vf: pointer to the VF structure
+ */
+void ice_set_vf_state_dis(struct ice_vf *vf)
+{
+	ice_set_vf_state_qs_dis(vf);
+	vf->vf_ops->clear_reset_state(vf);
+}
+
 /* Private functions only accessed from other virtualization files */
 
 /**
drivers/net/ethernet/intel/ice/ice_vf_lib.h +3 -2
index 52bd9a3816bf2..9f7fcd8e5714b 100644
@@ -56,6 +56,7 @@ struct ice_mdd_vf_events {
 struct ice_vf_ops {
 	enum ice_disq_rst_src reset_type;
 	void (*free)(struct ice_vf *vf);
+	void (*clear_reset_state)(struct ice_vf *vf);
 	void (*clear_mbx_register)(struct ice_vf *vf);
 	void (*trigger_reset_register)(struct ice_vf *vf, bool is_vflr);
 	bool (*poll_reset_status)(struct ice_vf *vf);
@@ -213,7 +214,7 @@ u16 ice_get_num_vfs(struct ice_pf *pf);
 struct ice_vsi *ice_get_vf_vsi(struct ice_vf *vf);
 bool ice_is_vf_disabled(struct ice_vf *vf);
 int ice_check_vf_ready_for_cfg(struct ice_vf *vf);
-void ice_set_vf_state_qs_dis(struct ice_vf *vf);
+void ice_set_vf_state_dis(struct ice_vf *vf);
 bool ice_is_any_vf_in_unicast_promisc(struct ice_pf *pf);
 void
 ice_vf_get_promisc_masks(struct ice_vf *vf, struct ice_vsi *vsi,
@@ -259,7 +260,7 @@ static inline int ice_check_vf_ready_for_cfg(struct ice_vf *vf)
 	return -EOPNOTSUPP;
 }
 
-static inline void ice_set_vf_state_qs_dis(struct ice_vf *vf)
+static inline void ice_set_vf_state_dis(struct ice_vf *vf)
 {
 }
 
Signature · 1 line
2.39.2
author maintainer
#203 lore raw

PATCH 203/292 ice: Fix ice VF reset during iavf initialization

In reply to Greg Kroah-Hartman

From: Dawid Wesierski <[email protected]>

[ Upstream commit 7255355a0636b4eff08d5e8139c77d98f151c4fc ]

Fix the current implementation that causes ice_trigger_vf_reset() to start resetting the VF even when the VF-NIC is still initializing.

When we reset NIC with ice driver it can interfere with iavf-vf initialization e.g. during consecutive resets induced by ice

iavf ice

  |                  |
  |<-----------------|
  |            ice resets vf
 iavf                |
 reset               |
 start               |
  |<-----------------|
  |             ice resets vf
  |             causing iavf
  |             initialization
  |             error
  |                  |

iavf
reset
end

This leads to a series of -53 errors
(failed to init adminq) from the IAVF.

Change the state of the vf_state field to be not active when the IAVF is still initializing. Make sure to wait until receiving the message on the message box to ensure that the vf is ready and initializded.

In simple terms we use the ACTIVE flag to make sure that the ice driver knows if the iavf is ready for another reset

iavf ice

    |                    |
    |                    |
    |<------------- ice resets vf
  iavf           vf_state != ACTIVE
  reset                  |
  start                  |
    |                    |
    |                    |
  iavf                   |
  reset-------> vf_state == ACTIVE
  end              ice resets vf
    |                    |
    |                    |
Fixes c54d209c78b8 ("ice: Wait for VF to be reset/ready before configuration")
---
 drivers/net/ethernet/intel/ice/ice_sriov.c    |  8 ++++----
 drivers/net/ethernet/intel/ice/ice_vf_lib.c   | 19 +++++++++++++++++++
 drivers/net/ethernet/intel/ice/ice_vf_lib.h   |  1 +
 drivers/net/ethernet/intel/ice/ice_virtchnl.c |  1 +
 4 files changed, 25 insertions(+), 4 deletions(-)
drivers/net/ethernet/intel/ice/ice_sriov.c +4 -4
index b719e9a771e36..b8c31bf721ad1 100644
@@ -1240,7 +1240,7 @@ int ice_set_vf_spoofchk(struct net_device *netdev, int vf_id, bool ena)
 	if (!vf)
 		return -EINVAL;
 
-	ret = ice_check_vf_ready_for_cfg(vf);
+	ret = ice_check_vf_ready_for_reset(vf);
 	if (ret)
 		goto out_put_vf;
 
@@ -1355,7 +1355,7 @@ int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac)
 		goto out_put_vf;
 	}
 
-	ret = ice_check_vf_ready_for_cfg(vf);
+	ret = ice_check_vf_ready_for_reset(vf);
 	if (ret)
 		goto out_put_vf;
 
@@ -1409,7 +1409,7 @@ int ice_set_vf_trust(struct net_device *netdev, int vf_id, bool trusted)
 		return -EOPNOTSUPP;
 	}
 
-	ret = ice_check_vf_ready_for_cfg(vf);
+	ret = ice_check_vf_ready_for_reset(vf);
 	if (ret)
 		goto out_put_vf;
 
@@ -1722,7 +1722,7 @@ ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos,
 	if (!vf)
 		return -EINVAL;
 
-	ret = ice_check_vf_ready_for_cfg(vf);
+	ret = ice_check_vf_ready_for_reset(vf);
 	if (ret)
 		goto out_put_vf;
 
drivers/net/ethernet/intel/ice/ice_vf_lib.c +19 -0
index 86abbcb480d9d..71047fc341392 100644
@@ -185,6 +185,25 @@ int ice_check_vf_ready_for_cfg(struct ice_vf *vf)
 	return 0;
 }
 
+/**
+ * ice_check_vf_ready_for_reset - check if VF is ready to be reset
+ * @vf: VF to check if it's ready to be reset
+ *
+ * The purpose of this function is to ensure that the VF is not in reset,
+ * disabled, and is both initialized and active, thus enabling us to safely
+ * initialize another reset.
+ */
+int ice_check_vf_ready_for_reset(struct ice_vf *vf)
+{
+	int ret;
+
+	ret = ice_check_vf_ready_for_cfg(vf);
+	if (!ret && !test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states))
+		ret = -EAGAIN;
+
+	return ret;
+}
+
 /**
  * ice_trigger_vf_reset - Reset a VF on HW
  * @vf: pointer to the VF structure
drivers/net/ethernet/intel/ice/ice_vf_lib.h +1 -0
index 9f7fcd8e5714b..e5bed85724622 100644
@@ -214,6 +214,7 @@ u16 ice_get_num_vfs(struct ice_pf *pf);
 struct ice_vsi *ice_get_vf_vsi(struct ice_vf *vf);
 bool ice_is_vf_disabled(struct ice_vf *vf);
 int ice_check_vf_ready_for_cfg(struct ice_vf *vf);
+int ice_check_vf_ready_for_reset(struct ice_vf *vf);
 void ice_set_vf_state_dis(struct ice_vf *vf);
 bool ice_is_any_vf_in_unicast_promisc(struct ice_pf *pf);
 void
drivers/net/ethernet/intel/ice/ice_virtchnl.c +1 -0
index 2b4c791b6cbad..ef3c709d6a750 100644
@@ -3722,6 +3722,7 @@ void ice_vc_process_vf_msg(struct ice_pf *pf, struct ice_rq_event_info *event)
 		ice_vc_notify_vf_link_state(vf);
 		break;
 	case VIRTCHNL_OP_RESET_VF:
+		clear_bit(ICE_VF_STATE_ACTIVE, vf->vf_states);
 		ops->reset_vf(vf);
 		break;
 	case VIRTCHNL_OP_ADD_ETH_ADDR:
Signature · 1 line
2.39.2
author maintainer
#204 lore raw

PATCH 204/292 wifi: cfg80211: Drop entries with invalid BSSIDs in RNR

In reply to Greg Kroah-Hartman

From: Ilan Peer <[email protected]>

[ Upstream commit 1b6b4ed01493b7ea2205ab83c49198f7d13ca9d2 ]

Ignore AP information for entries that include an invalid BSSID in the TBTT information field, e.g., all zeros BSSIDs.

---
 net/wireless/scan.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
net/wireless/scan.c +5 -1
index 3d86482e83f51..6c2b73c0d36e8 100644
@@ -5,7 +5,7 @@
  * Copyright 2008 Johannes Berg <[email protected]>
  * Copyright 2013-2014  Intel Mobile Communications GmbH
  * Copyright 2016	Intel Deutschland GmbH
- * Copyright (C) 2018-2022 Intel Corporation
+ * Copyright (C) 2018-2023 Intel Corporation
  */
 #include <linux/kernel.h>
 #include <linux/slab.h>
@@ -543,6 +543,10 @@ static int cfg80211_parse_ap_info(struct cfg80211_colocated_ap *entry,
 	/* skip the TBTT offset */
 	pos++;
 
+	/* ignore entries with invalid BSSID */
+	if (!is_valid_ether_addr(pos))
+		return -EINVAL;
+
 	memcpy(entry->bssid, pos, ETH_ALEN);
 	pos += ETH_ALEN;
 
Signature · 1 line
2.39.2
author maintainer
#205 lore raw

PATCH 205/292 wifi: mac80211: fortify the spinlock against deadlock by interrupt

In reply to Greg Kroah-Hartman

From: Mirsad Goran Todorovac <[email protected]>

[ Upstream commit ef6e1997da63ad0ac3fe33153fec9524c9ae56c9 ]

In the function ieee80211_tx_dequeue() there is a particular locking sequence:

begin:

	spin_lock(&local->queue_stop_reason_lock);
	q_stopped = local->queue_stop_reasons[q];
	spin_unlock(&local->queue_stop_reason_lock);

However small the chance (increased by ftracetest), an asynchronous interrupt can occur in between of spin_lock() and spin_unlock(), and the interrupt routine will attempt to lock the same &local->queue_stop_reason_lock again.

This will cause a costly reset of the CPU and the wifi device or an altogether hang in the single CPU and single core scenario.

The only remaining spin_lock(&local->queue_stop_reason_lock) that did not disable interrupts was patched, which should prevent any deadlocks on the same CPU/core and the same wifi device.

This is the probable trace of the deadlock:

kernel: ================================
kernel: WARNING: inconsistent lock state
kernel: 6.3.0-rc6-mt-20230401-00001-gf86822a1170f #4 Tainted: G        W
kernel: --------------------------------
kernel: inconsistent {IN-SOFTIRQ-W} -> {SOFTIRQ-ON-W} usage.
kernel: kworker/5:0/25656 [HC0[0]:SC0[0]:HE1:SE1] takes:
kernel: ffff9d6190779478 (&local->queue_stop_reason_lock){+.?.}-{2:2}, at: return_to_handler+0x0/0x40
kernel: {IN-SOFTIRQ-W} state was registered at:
kernel:   lock_acquire+0xc7/0x2d0
kernel:   _raw_spin_lock+0x36/0x50
kernel:   ieee80211_tx_dequeue+0xb4/0x1330 [mac80211]
kernel:   iwl_mvm_mac_itxq_xmit+0xae/0x210 [iwlmvm]
kernel:   iwl_mvm_mac_wake_tx_queue+0x2d/0xd0 [iwlmvm]
kernel:   ieee80211_queue_skb+0x450/0x730 [mac80211]
kernel:   __ieee80211_xmit_fast.constprop.66+0x834/0xa50 [mac80211]
kernel:   __ieee80211_subif_start_xmit+0x217/0x530 [mac80211]
kernel:   ieee80211_subif_start_xmit+0x60/0x580 [mac80211]
kernel:   dev_hard_start_xmit+0xb5/0x260
kernel:   __dev_queue_xmit+0xdbe/0x1200
kernel:   neigh_resolve_output+0x166/0x260
kernel:   ip_finish_output2+0x216/0xb80
kernel:   __ip_finish_output+0x2a4/0x4d0
kernel:   ip_finish_output+0x2d/0xd0
kernel:   ip_output+0x82/0x2b0
kernel:   ip_local_out+0xec/0x110
kernel:   igmpv3_sendpack+0x5c/0x90
kernel:   igmp_ifc_timer_expire+0x26e/0x4e0
kernel:   call_timer_fn+0xa5/0x230
kernel:   run_timer_softirq+0x27f/0x550
kernel:   __do_softirq+0xb4/0x3a4
kernel:   irq_exit_rcu+0x9b/0xc0
kernel:   sysvec_apic_timer_interrupt+0x80/0xa0
kernel:   asm_sysvec_apic_timer_interrupt+0x1f/0x30
kernel:   _raw_spin_unlock_irqrestore+0x3f/0x70
kernel:   free_to_partial_list+0x3d6/0x590
kernel:   __slab_free+0x1b7/0x310
kernel:   kmem_cache_free+0x52d/0x550
kernel:   putname+0x5d/0x70
kernel:   do_sys_openat2+0x1d7/0x310
kernel:   do_sys_open+0x51/0x80
kernel:   __x64_sys_openat+0x24/0x30
kernel:   do_syscall_64+0x5c/0x90
kernel:   entry_SYSCALL_64_after_hwframe+0x72/0xdc
kernel: irq event stamp: 5120729
kernel: hardirqs last  enabled at (5120729): [<ffffffff9d149936>] trace_graph_return+0xd6/0x120
kernel: hardirqs last disabled at (5120728): [<ffffffff9d149950>] trace_graph_return+0xf0/0x120
kernel: softirqs last  enabled at (5069900): [<ffffffff9cf65b60>] return_to_handler+0x0/0x40
kernel: softirqs last disabled at (5067555): [<ffffffff9cf65b60>] return_to_handler+0x0/0x40
kernel:
        other info that might help us debug this:
kernel:  Possible unsafe locking scenario:
kernel:        CPU0
kernel:        ----
kernel:   lock(&local->queue_stop_reason_lock);
kernel:   <Interrupt>
kernel:     lock(&local->queue_stop_reason_lock);
kernel:
         *** DEADLOCK ***
kernel: 8 locks held by kworker/5:0/25656:
kernel:  #0: ffff9d618009d138 ((wq_completion)events_freezable){+.+.}-{0:0}, at: process_one_work+0x1ca/0x530
kernel:  #1: ffffb1ef4637fe68 ((work_completion)(&local->restart_work)){+.+.}-{0:0}, at: process_one_work+0x1ce/0x530
kernel:  #2: ffffffff9f166548 (rtnl_mutex){+.+.}-{3:3}, at: return_to_handler+0x0/0x40
kernel:  #3: ffff9d6190778728 (&rdev->wiphy.mtx){+.+.}-{3:3}, at: return_to_handler+0x0/0x40
kernel:  #4: ffff9d619077b480 (&mvm->mutex){+.+.}-{3:3}, at: return_to_handler+0x0/0x40
kernel:  #5: ffff9d61907bacd8 (&trans_pcie->mutex){+.+.}-{3:3}, at: return_to_handler+0x0/0x40
kernel:  #6: ffffffff9ef9cda0 (rcu_read_lock){....}-{1:2}, at: iwl_mvm_queue_state_change+0x59/0x3a0 [iwlmvm]
kernel:  #7: ffffffff9ef9cda0 (rcu_read_lock){....}-{1:2}, at: iwl_mvm_mac_itxq_xmit+0x42/0x210 [iwlmvm]
kernel:
        stack backtrace:
kernel: CPU: 5 PID: 25656 Comm: kworker/5:0 Tainted: G        W          6.3.0-rc6-mt-20230401-00001-gf86822a1170f #4
kernel: Hardware name: LENOVO 82H8/LNVNB161216, BIOS GGCN51WW 11/16/2022
kernel: Workqueue: events_freezable ieee80211_restart_work [mac80211]
kernel: Call Trace:
kernel:  <TASK>
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  dump_stack_lvl+0x5f/0xa0
kernel:  dump_stack+0x14/0x20
kernel:  print_usage_bug.part.46+0x208/0x2a0
kernel:  mark_lock.part.47+0x605/0x630
kernel:  ? sched_clock+0xd/0x20
kernel:  ? trace_clock_local+0x14/0x30
kernel:  ? __rb_reserve_next+0x5f/0x490
kernel:  ? _raw_spin_lock+0x1b/0x50
kernel:  __lock_acquire+0x464/0x1990
kernel:  ? mark_held_locks+0x4e/0x80
kernel:  lock_acquire+0xc7/0x2d0
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  ? ftrace_return_to_handler+0x8b/0x100
kernel:  ? preempt_count_add+0x4/0x70
kernel:  _raw_spin_lock+0x36/0x50
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  ieee80211_tx_dequeue+0xb4/0x1330 [mac80211]
kernel:  ? prepare_ftrace_return+0xc5/0x190
kernel:  ? ftrace_graph_func+0x16/0x20
kernel:  ? 0xffffffffc02ab0b1
kernel:  ? lock_acquire+0xc7/0x2d0
kernel:  ? iwl_mvm_mac_itxq_xmit+0x42/0x210 [iwlmvm]
kernel:  ? ieee80211_tx_dequeue+0x9/0x1330 [mac80211]
kernel:  ? __rcu_read_lock+0x4/0x40
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  iwl_mvm_mac_itxq_xmit+0xae/0x210 [iwlmvm]
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  iwl_mvm_queue_state_change+0x311/0x3a0 [iwlmvm]
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  iwl_mvm_wake_sw_queue+0x17/0x20 [iwlmvm]
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  iwl_txq_gen2_unmap+0x1c9/0x1f0 [iwlwifi]
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  iwl_txq_gen2_free+0x55/0x130 [iwlwifi]
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  iwl_txq_gen2_tx_free+0x63/0x80 [iwlwifi]
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  _iwl_trans_pcie_gen2_stop_device+0x3f3/0x5b0 [iwlwifi]
kernel:  ? _iwl_trans_pcie_gen2_stop_device+0x9/0x5b0 [iwlwifi]
kernel:  ? mutex_lock_nested+0x4/0x30
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  iwl_trans_pcie_gen2_stop_device+0x5f/0x90 [iwlwifi]
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  iwl_mvm_stop_device+0x78/0xd0 [iwlmvm]
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  __iwl_mvm_mac_start+0x114/0x210 [iwlmvm]
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  iwl_mvm_mac_start+0x76/0x150 [iwlmvm]
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  drv_start+0x79/0x180 [mac80211]
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  ieee80211_reconfig+0x1523/0x1ce0 [mac80211]
kernel:  ? synchronize_net+0x4/0x50
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  ieee80211_restart_work+0x108/0x170 [mac80211]
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  process_one_work+0x250/0x530
kernel:  ? ftrace_regs_caller_end+0x66/0x66
kernel:  worker_thread+0x48/0x3a0
kernel:  ? __pfx_worker_thread+0x10/0x10
kernel:  kthread+0x10f/0x140
kernel:  ? __pfx_kthread+0x10/0x10
kernel:  ret_from_fork+0x29/0x50
kernel:  </TASK>
Fixes 4444bc2116ae ("wifi: mac80211: Proper mark iTXQs for resumption")
Cc Gregory Greenman <[email protected]>
Cc Johannes Berg <[email protected]>
Cc David S. Miller <[email protected]>
Cc Eric Dumazet <[email protected]>
Cc Jakub Kicinski <[email protected]>
Cc Paolo Abeni <[email protected]>
Cc Leon Romanovsky <[email protected]>
Cc Alexander Wetzel <[email protected]>
Reviewed-by tag, or it goes automatically?
---
 net/mac80211/tx.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
net/mac80211/tx.c +3 -2
index 6a1708db652f2..763cefd0cc268 100644
@@ -3718,6 +3718,7 @@ struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw,
 	ieee80211_tx_result r;
 	struct ieee80211_vif *vif = txq->vif;
 	int q = vif->hw_queue[txq->ac];
+	unsigned long flags;
 	bool q_stopped;
 
 	WARN_ON_ONCE(softirq_count() == 0);
@@ -3726,9 +3727,9 @@ struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw,
 		return NULL;
 
 begin:
-	spin_lock(&local->queue_stop_reason_lock);
+	spin_lock_irqsave(&local->queue_stop_reason_lock, flags);
 	q_stopped = local->queue_stop_reasons[q];
-	spin_unlock(&local->queue_stop_reason_lock);
+	spin_unlock_irqrestore(&local->queue_stop_reason_lock, flags);
 
 	if (unlikely(q_stopped)) {
 		/* mark for waking later */
Signature · 1 line
2.39.2
author maintainer
#206 lore raw

PATCH 206/292 wifi: mac80211: fix min center freq offset tracing

In reply to Greg Kroah-Hartman

From: Johannes Berg <[email protected]>

[ Upstream commit 248e4776514bf70236e6b1a54c65aa5324c8b1eb ]

We need to set the correct trace variable, otherwise we're overwriting something else instead and the right one that we print later is not initialized.

---
 net/mac80211/trace.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/mac80211/trace.h +1 -1
index 9f4377566c425..c85367a4757a9 100644
@@ -67,7 +67,7 @@
 			__entry->min_freq_offset = (c)->chan ? (c)->chan->freq_offset : 0;	\
 			__entry->min_chan_width = (c)->width;				\
 			__entry->min_center_freq1 = (c)->center_freq1;			\
-			__entry->freq1_offset = (c)->freq1_offset;			\
+			__entry->min_freq1_offset = (c)->freq1_offset;			\
 			__entry->min_center_freq2 = (c)->center_freq2;
 #define MIN_CHANDEF_PR_FMT	" min_control:%d.%03d MHz min_width:%d min_center: %d.%03d/%d MHz"
 #define MIN_CHANDEF_PR_ARG	__entry->min_control_freq, __entry->min_freq_offset,	\
Signature · 1 line
2.39.2
author maintainer
#207 lore raw

PATCH 207/292 wifi: mac80211: Abort running color change when stopping the AP

In reply to Greg Kroah-Hartman

From: Michael Lee <[email protected]>

[ Upstream commit a23d7f5b2fbda114de60c4b53311e052281d7533 ]

When stopping the AP, there might be a color change in progress. It should be deactivated here, or the driver might later finalize a color change on a stopped AP.

Fixes 5f9404abdf2a (mac80211: add support for BSS color change)
---
 net/mac80211/cfg.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
net/mac80211/cfg.c +2 -1
index e8beec0a0ae1c..06b9df2fbcd77 100644
@@ -1477,9 +1477,10 @@ static int ieee80211_stop_ap(struct wiphy *wiphy, struct net_device *dev,
 		sdata_dereference(link->u.ap.unsol_bcast_probe_resp,
 				  sdata);
 
-	/* abort any running channel switch */
+	/* abort any running channel switch or color change */
 	mutex_lock(&local->mtx);
 	link_conf->csa_active = false;
+	link_conf->color_change_active = false;
 	if (link->csa_block_tx) {
 		ieee80211_wake_vif_queues(local, sdata,
 					  IEEE80211_QUEUE_STOP_REASON_CSA);
Signature · 1 line
2.39.2
author maintainer
#208 lore raw

PATCH 208/292 wifi: iwlwifi: mvm: fix cancel_delayed_work_sync() deadlock

In reply to Greg Kroah-Hartman

From: Johannes Berg <[email protected]>

[ Upstream commit c2d8b7f257b2398f2d866205365895e038beca12 ]

Lockdep points out that we can deadlock here by calling cancel_delayed_work_sync() because that might be already running and gotten interrupted by the NAPI soft-IRQ. Even just calling something that can sleep is wrong in this context though.

Luckily, it doesn't even really matter since the things we need to do are idempotent, so just drop the _sync().

---
 drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/intel/iwlwifi/mvm/rxmq.c +1 -1
index 091225894037c..02c2a06301076 100644
@@ -1975,7 +1975,7 @@ void iwl_mvm_rx_mpdu_mq(struct iwl_mvm *mvm, struct napi_struct *napi,
 				RCU_INIT_POINTER(mvm->csa_tx_blocked_vif, NULL);
 				/* Unblock BCAST / MCAST station */
 				iwl_mvm_modify_all_sta_disable_tx(mvm, mvmvif, false);
-				cancel_delayed_work_sync(&mvm->cs_tx_unblock_dwork);
+				cancel_delayed_work(&mvm->cs_tx_unblock_dwork);
 			}
 		}
 
Signature · 1 line
2.39.2
author maintainer
#209 lore raw

PATCH 209/292 wifi: iwlwifi: fw: fix DBGI dump

In reply to Greg Kroah-Hartman

From: Johannes Berg <[email protected]>

[ Upstream commit d3ae69180bbd74bcbc03a2b6d10ed7eccbe98c23 ]

The DBGI dump is (unsurprisingly) of type DBGI, not SRAM. This leads to bad register accesses because the union is built differently, there's no allocation ID, and thus the allocation ID ends up being 0x8000.

Note that this was already wrong for DRAM vs. SMEM since they use different parts of the union, but the allocation ID is at the same place, so it worked.

Fix all of this but set the allocation ID in a way that the offset calculation ends up without any offset.

---
 drivers/net/wireless/intel/iwlwifi/fw/dbg.c | 19 +++++++++++--------
 1 file changed, 11 insertions(+), 8 deletions(-)
drivers/net/wireless/intel/iwlwifi/fw/dbg.c +11 -8
index 027360e63b926..3ef0b776b7727 100644
@@ -1664,14 +1664,10 @@ static __le32 iwl_get_mon_reg(struct iwl_fw_runtime *fwrt, u32 alloc_id,
 }
 
 static void *
-iwl_dump_ini_mon_fill_header(struct iwl_fw_runtime *fwrt,
-			     struct iwl_dump_ini_region_data *reg_data,
+iwl_dump_ini_mon_fill_header(struct iwl_fw_runtime *fwrt, u32 alloc_id,
 			     struct iwl_fw_ini_monitor_dump *data,
 			     const struct iwl_fw_mon_regs *addrs)
 {
-	struct iwl_fw_ini_region_tlv *reg = (void *)reg_data->reg_tlv->data;
-	u32 alloc_id = le32_to_cpu(reg->dram_alloc_id);
-
 	if (!iwl_trans_grab_nic_access(fwrt->trans)) {
 		IWL_ERR(fwrt, "Failed to get monitor header\n");
 		return NULL;
@@ -1702,8 +1698,10 @@ iwl_dump_ini_mon_dram_fill_header(struct iwl_fw_runtime *fwrt,
 				  void *data, u32 data_len)
 {
 	struct iwl_fw_ini_monitor_dump *mon_dump = (void *)data;
+	struct iwl_fw_ini_region_tlv *reg = (void *)reg_data->reg_tlv->data;
+	u32 alloc_id = le32_to_cpu(reg->dram_alloc_id);
 
-	return iwl_dump_ini_mon_fill_header(fwrt, reg_data, mon_dump,
+	return iwl_dump_ini_mon_fill_header(fwrt, alloc_id, mon_dump,
 					    &fwrt->trans->cfg->mon_dram_regs);
 }
 
@@ -1713,8 +1711,10 @@ iwl_dump_ini_mon_smem_fill_header(struct iwl_fw_runtime *fwrt,
 				  void *data, u32 data_len)
 {
 	struct iwl_fw_ini_monitor_dump *mon_dump = (void *)data;
+	struct iwl_fw_ini_region_tlv *reg = (void *)reg_data->reg_tlv->data;
+	u32 alloc_id = le32_to_cpu(reg->internal_buffer.alloc_id);
 
-	return iwl_dump_ini_mon_fill_header(fwrt, reg_data, mon_dump,
+	return iwl_dump_ini_mon_fill_header(fwrt, alloc_id, mon_dump,
 					    &fwrt->trans->cfg->mon_smem_regs);
 }
 
@@ -1725,7 +1725,10 @@ iwl_dump_ini_mon_dbgi_fill_header(struct iwl_fw_runtime *fwrt,
 {
 	struct iwl_fw_ini_monitor_dump *mon_dump = (void *)data;
 
-	return iwl_dump_ini_mon_fill_header(fwrt, reg_data, mon_dump,
+	return iwl_dump_ini_mon_fill_header(fwrt,
+					    /* no offset calculation later */
+					    IWL_FW_INI_ALLOCATION_ID_DBGC1,
+					    mon_dump,
 					    &fwrt->trans->cfg->mon_dbgi_regs);
 }
 
Signature · 1 line
2.39.2
author maintainer
#210 lore raw

PATCH 210/292 wifi: iwlwifi: fix OEMs name in the ppag approved list

In reply to Greg Kroah-Hartman

From: Alon Giladi <[email protected]>

[ Upstream commit eca7296d9a671e9961834d2ace9cc0ce21fc15b3 ]

Fix a spelling mistake.

---
 drivers/net/wireless/intel/iwlwifi/fw/acpi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/intel/iwlwifi/fw/acpi.c +1 -1
index a02e5a67b7066..585e8cd2d332d 100644
@@ -38,7 +38,7 @@ static const struct dmi_system_id dmi_ppag_approved_list[] = {
 	},
 	{ .ident = "ASUS",
 	  .matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTek COMPUTER INC."),
+			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
 		},
 	},
 	{}
Signature · 1 line
2.39.2
author maintainer
#211 lore raw

PATCH 211/292 wifi: iwlwifi: mvm: fix OEMs name in the tas approved list

In reply to Greg Kroah-Hartman

From: Alon Giladi <[email protected]>

[ Upstream commit d0246a0e49efee0f8649d0e4f2350614cdfe6565 ]

Fix a spelling mistake.

---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/intel/iwlwifi/mvm/fw.c +1 -1
index 5de34edc51fe9..887d0789c96c3 100644
@@ -1055,7 +1055,7 @@ static const struct dmi_system_id dmi_tas_approved_list[] = {
 	},
 		{ .ident = "LENOVO",
 	  .matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Lenovo"),
+			DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
 		},
 	},
 	{ .ident = "DELL",
Signature · 1 line
2.39.2
author maintainer
#212 lore raw

PATCH 212/292 wifi: iwlwifi: mvm: dont trust firmware n_channels

In reply to Greg Kroah-Hartman

From: Johannes Berg <[email protected]>

[ Upstream commit 682b6dc29d98e857e6ca4bbc077c7dc2899b7473 ]

If the firmware sends us a corrupted MCC response with n_channels much larger than the command response can be, we might copy far too much (uninitialized) memory and even crash if the n_channels is large enough to make it run out of the one page allocated for the FW response.

Fix that by checking the lengths. Doing a < comparison would be sufficient, but the firmware should be doing it correctly, so check more strictly.

---
 drivers/net/wireless/intel/iwlwifi/mvm/nvm.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
drivers/net/wireless/intel/iwlwifi/mvm/nvm.c +10 -0
index 6d18a1fd649b9..fdf60afb0f3f2 100644
@@ -445,6 +445,11 @@ iwl_mvm_update_mcc(struct iwl_mvm *mvm, const char *alpha2,
 		struct iwl_mcc_update_resp *mcc_resp = (void *)pkt->data;
 
 		n_channels =  __le32_to_cpu(mcc_resp->n_channels);
+		if (iwl_rx_packet_payload_len(pkt) !=
+		    struct_size(mcc_resp, channels, n_channels)) {
+			resp_cp = ERR_PTR(-EINVAL);
+			goto exit;
+		}
 		resp_len = sizeof(struct iwl_mcc_update_resp) +
 			   n_channels * sizeof(__le32);
 		resp_cp = kmemdup(mcc_resp, resp_len, GFP_KERNEL);
@@ -456,6 +461,11 @@ iwl_mvm_update_mcc(struct iwl_mvm *mvm, const char *alpha2,
 		struct iwl_mcc_update_resp_v3 *mcc_resp_v3 = (void *)pkt->data;
 
 		n_channels =  __le32_to_cpu(mcc_resp_v3->n_channels);
+		if (iwl_rx_packet_payload_len(pkt) !=
+		    struct_size(mcc_resp_v3, channels, n_channels)) {
+			resp_cp = ERR_PTR(-EINVAL);
+			goto exit;
+		}
 		resp_len = sizeof(struct iwl_mcc_update_resp) +
 			   n_channels * sizeof(__le32);
 		resp_cp = kzalloc(resp_len, GFP_KERNEL);
Signature · 1 line
2.39.2
author maintainer
#213 lore raw

PATCH 213/292 scsi: storvsc: Dont pass unused PFNs to Hyper-V host

In reply to Greg Kroah-Hartman

From: Michael Kelley <[email protected]>

[ Upstream commit 4e81a6cba517cb33584308a331f14f5e3fec369b ]

In a SCSI request, storvsc pre-allocates space for up to MAX_PAGE_BUFFER_COUNT physical frame numbers to be passed to Hyper-V. If the size of the I/O request requires more PFNs, a separate memory area of exactly the correct size is dynamically allocated.

But when the pre-allocated area is used, current code always passes MAX_PAGE_BUFFER_COUNT PFNs to Hyper-V, even if fewer are needed. While this doesn't break anything because the additional PFNs are always zero, more bytes than necessary are copied into the VMBus channel ring buffer. This takes CPU cycles and wastes space in the ring buffer. For a typical 4 Kbyte I/O that requires only a single PFN, 248 unnecessary bytes are copied.

Fix this by setting the payload_sz based on the actual number of PFNs required, not the size of the pre-allocated space.

Fixes 8f43710543ef ("scsi: storvsc: Support PAGE_SIZE larger than 4K")
---
 drivers/scsi/storvsc_drv.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/scsi/storvsc_drv.c +4 -4
index a0665bca54b99..5284f9a0b826e 100644
@@ -1780,7 +1780,7 @@ static int storvsc_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *scmnd)
 
 	length = scsi_bufflen(scmnd);
 	payload = (struct vmbus_packet_mpb_array *)&cmd_request->mpb;
-	payload_sz = sizeof(cmd_request->mpb);
+	payload_sz = 0;
 
 	if (scsi_sg_count(scmnd)) {
 		unsigned long offset_in_hvpg = offset_in_hvpage(sgl->offset);
@@ -1789,10 +1789,10 @@ static int storvsc_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *scmnd)
 		unsigned long hvpfn, hvpfns_to_add;
 		int j, i = 0, sg_count;
 
-		if (hvpg_count > MAX_PAGE_BUFFER_COUNT) {
+		payload_sz = (hvpg_count * sizeof(u64) +
+			      sizeof(struct vmbus_packet_mpb_array));
 
-			payload_sz = (hvpg_count * sizeof(u64) +
-				      sizeof(struct vmbus_packet_mpb_array));
+		if (hvpg_count > MAX_PAGE_BUFFER_COUNT) {
 			payload = kzalloc(payload_sz, GFP_ATOMIC);
 			if (!payload)
 				return SCSI_MLQUEUE_DEVICE_BUSY;
Signature · 1 line
2.39.2
author maintainer
#214 lore raw

PATCH 214/292 net: tun: rebuild error handling in tun_get_user

In reply to Greg Kroah-Hartman

From: Chuang Wang <[email protected]>

[ Upstream commit ab00af85d2f886a8e4ace1342d9cc2b232eab6a8 ]

The error handling in tun_get_user is very scattered. This patch unifies error handling, reduces duplication of code, and makes the logic clearer.

Stable-dep-of: 82b2bc279467 ("tun: Fix memory leak for detached NAPI queue.")

---
 drivers/net/tun.c | 65 +++++++++++++++++++++--------------------------
 1 file changed, 29 insertions(+), 36 deletions(-)
drivers/net/tun.c +29 -36
index 91d198aff2f9a..65706824eb828 100644
@@ -1748,7 +1748,7 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
 	u32 rxhash = 0;
 	int skb_xdp = 1;
 	bool frags = tun_napi_frags_enabled(tfile);
-	enum skb_drop_reason drop_reason;
+	enum skb_drop_reason drop_reason = SKB_DROP_REASON_NOT_SPECIFIED;
 
 	if (!(tun->flags & IFF_NO_PI)) {
 		if (len < sizeof(pi))
@@ -1809,10 +1809,9 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
 		 * skb was created with generic XDP routine.
 		 */
 		skb = tun_build_skb(tun, tfile, from, &gso, len, &skb_xdp);
-		if (IS_ERR(skb)) {
-			dev_core_stats_rx_dropped_inc(tun->dev);
-			return PTR_ERR(skb);
-		}
+		err = PTR_ERR_OR_ZERO(skb);
+		if (err)
+			goto drop;
 		if (!skb)
 			return total_len;
 	} else {
@@ -1837,13 +1836,9 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
 					    noblock);
 		}
 
-		if (IS_ERR(skb)) {
-			if (PTR_ERR(skb) != -EAGAIN)
-				dev_core_stats_rx_dropped_inc(tun->dev);
-			if (frags)
-				mutex_unlock(&tfile->napi_mutex);
-			return PTR_ERR(skb);
-		}
+		err = PTR_ERR_OR_ZERO(skb);
+		if (err)
+			goto drop;
 
 		if (zerocopy)
 			err = zerocopy_sg_from_iter(skb, from);
@@ -1853,27 +1848,14 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
 		if (err) {
 			err = -EFAULT;
 			drop_reason = SKB_DROP_REASON_SKB_UCOPY_FAULT;
-drop:
-			dev_core_stats_rx_dropped_inc(tun->dev);
-			kfree_skb_reason(skb, drop_reason);
-			if (frags) {
-				tfile->napi.skb = NULL;
-				mutex_unlock(&tfile->napi_mutex);
-			}
-
-			return err;
+			goto drop;
 		}
 	}
 
 	if (virtio_net_hdr_to_skb(skb, &gso, tun_is_little_endian(tun))) {
 		atomic_long_inc(&tun->rx_frame_errors);
-		kfree_skb(skb);
-		if (frags) {
-			tfile->napi.skb = NULL;
-			mutex_unlock(&tfile->napi_mutex);
-		}
-
-		return -EINVAL;
+		err = -EINVAL;
+		goto free_skb;
 	}
 
 	switch (tun->flags & TUN_TYPE_MASK) {
@@ -1889,9 +1871,8 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
 				pi.proto = htons(ETH_P_IPV6);
 				break;
 			default:
-				dev_core_stats_rx_dropped_inc(tun->dev);
-				kfree_skb(skb);
-				return -EINVAL;
+				err = -EINVAL;
+				goto drop;
 			}
 		}
 
@@ -1933,11 +1914,7 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
 			if (ret != XDP_PASS) {
 				rcu_read_unlock();
 				local_bh_enable();
-				if (frags) {
-					tfile->napi.skb = NULL;
-					mutex_unlock(&tfile->napi_mutex);
-				}
-				return total_len;
+				goto unlock_frags;
 			}
 		}
 		rcu_read_unlock();
@@ -2017,6 +1994,22 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
 		tun_flow_update(tun, rxhash, tfile);
 
 	return total_len;
+
+drop:
+	if (err != -EAGAIN)
+		dev_core_stats_rx_dropped_inc(tun->dev);
+
+free_skb:
+	if (!IS_ERR_OR_NULL(skb))
+		kfree_skb_reason(skb, drop_reason);
+
+unlock_frags:
+	if (frags) {
+		tfile->napi.skb = NULL;
+		mutex_unlock(&tfile->napi_mutex);
+	}
+
+	return err ?: total_len;
 }
 
 static ssize_t tun_chr_write_iter(struct kiocb *iocb, struct iov_iter *from)
Signature · 1 line
2.39.2
author maintainer
#215 lore raw

PATCH 215/292 tun: Fix memory leak for detached NAPI queue.

In reply to Greg Kroah-Hartman

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit 82b2bc279467c875ec36f8ef820f00997c2a4e8e ]

syzkaller reported [0] memory leaks of sk and skb related to the TUN device with no repro, but we can reproduce it easily with:

  struct ifreq ifr = {}
  int fd_tun, fd_tmp;
  char buf[4] = {};

  fd_tun = openat(AT_FDCWD, "/dev/net/tun", O_WRONLY, 0);
  ifr.ifr_flags = IFF_TUN | IFF_NAPI | IFF_MULTI_QUEUE;
  ioctl(fd_tun, TUNSETIFF, &ifr);

  ifr.ifr_flags = IFF_DETACH_QUEUE;
  ioctl(fd_tun, TUNSETQUEUE, &ifr);

  fd_tmp = socket(AF_PACKET, SOCK_PACKET, 0);
  ifr.ifr_flags = IFF_UP;
  ioctl(fd_tmp, SIOCSIFFLAGS, &ifr);

  write(fd_tun, buf, sizeof(buf));
  close(fd_tun);

If we enable NAPI and multi-queue on a TUN device, we can put skb into tfile->sk.sk_write_queue after the queue is detached. We should prevent it by checking tfile->detached before queuing skb.

Note this must be done under tfile->sk.sk_write_queue.lock because write() and ioctl(IFF_DETACH_QUEUE) can run concurrently. Otherwise, there would be a small race window:

write() ioctl(IFF_DETACH_QUEUE)
- tun_get_user - __tun_detach

     |- if (tfile->detached)             |- tun_disable_queue
     |  `-> false                        |  `- tfile->detached = tun
     |                                   `- tun_queue_purge
     |- spin_lock_bh(&queue->lock)
     `- __skb_queue_tail(queue, skb)

Another solution is to call tun_queue_purge() when closing and reattaching the detached queue, but it could paper over another problems. Also, we do the same kind of test for IFF_NAPI_FRAGS.

[0]:
unreferenced object 0xffff88801edbc800 (size 2048):
comm "syz-executor.1", pid 33269, jiffies 4295743834 (age 18.756s) hex dump (first 32 bytes):

    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
    00 00 07 40 00 00 00 00 00 00 00 00 00 00 00 00  ...@............
  backtrace:
    [<000000008c16ea3d>] __do_kmalloc_node mm/slab_common.c:965 [inline]
    [<000000008c16ea3d>] __kmalloc+0x4a/0x130 mm/slab_common.c:979
    [<000000003addde56>] kmalloc include/linux/slab.h:563 [inline]
    [<000000003addde56>] sk_prot_alloc+0xef/0x1b0 net/core/sock.c:2035
    [<000000003e20621f>] sk_alloc+0x36/0x2f0 net/core/sock.c:2088
    [<0000000028e43843>] tun_chr_open+0x3d/0x190 drivers/net/tun.c:3438
    [<000000001b0f1f28>] misc_open+0x1a6/0x1f0 drivers/char/misc.c:165
    [<000000004376f706>] chrdev_open+0x111/0x300 fs/char_dev.c:414
    [<00000000614d379f>] do_dentry_open+0x2f9/0x750 fs/open.c:920
    [<000000008eb24774>] do_open fs/namei.c:3636 [inline]
    [<000000008eb24774>] path_openat+0x143f/0x1a30 fs/namei.c:3791
    [<00000000955077b5>] do_filp_open+0xce/0x1c0 fs/namei.c:3818
    [<00000000b78973b0>] do_sys_openat2+0xf0/0x260 fs/open.c:1356
    [<00000000057be699>] do_sys_open fs/open.c:1372 [inline]
    [<00000000057be699>] __do_sys_openat fs/open.c:1388 [inline]
    [<00000000057be699>] __se_sys_openat fs/open.c:1383 [inline]
    [<00000000057be699>] __x64_sys_openat+0x83/0xf0 fs/open.c:1383
    [<00000000a7d2182d>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
    [<00000000a7d2182d>] do_syscall_64+0x3c/0x90 arch/x86/entry/common.c:80
    [<000000004cc4e8c4>] entry_SYSCALL_64_after_hwframe+0x72/0xdc

unreferenced object 0xffff88802f671700 (size 240):
comm "syz-executor.1", pid 33269, jiffies 4295743854 (age 18.736s) hex dump (first 32 bytes):

    68 c9 db 1e 80 88 ff ff 68 c9 db 1e 80 88 ff ff  h.......h.......
    00 c0 7b 2f 80 88 ff ff 00 c8 db 1e 80 88 ff ff  ..{/............
  backtrace:
    [<00000000e9d9fdb6>] __alloc_skb+0x223/0x250 net/core/skbuff.c:644
    [<000000002c3e4e0b>] alloc_skb include/linux/skbuff.h:1288 [inline]
    [<000000002c3e4e0b>] alloc_skb_with_frags+0x6f/0x350 net/core/skbuff.c:6378
    [<00000000825f98d7>] sock_alloc_send_pskb+0x3ac/0x3e0 net/core/sock.c:2729
    [<00000000e9eb3df3>] tun_alloc_skb drivers/net/tun.c:1529 [inline]
    [<00000000e9eb3df3>] tun_get_user+0x5e1/0x1f90 drivers/net/tun.c:1841
    [<0000000053096912>] tun_chr_write_iter+0xac/0x120 drivers/net/tun.c:2035
    [<00000000b9282ae0>] call_write_iter include/linux/fs.h:1868 [inline]
    [<00000000b9282ae0>] new_sync_write fs/read_write.c:491 [inline]
    [<00000000b9282ae0>] vfs_write+0x40f/0x530 fs/read_write.c:584
    [<00000000524566e4>] ksys_write+0xa1/0x170 fs/read_write.c:637
    [<00000000a7d2182d>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
    [<00000000a7d2182d>] do_syscall_64+0x3c/0x90 arch/x86/entry/common.c:80
    [<000000004cc4e8c4>] entry_SYSCALL_64_after_hwframe+0x72/0xdc
Fixes cde8b15f1aab ("tuntap: add ioctl to attach or detach a file form tuntap device")
---
 drivers/net/tun.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
drivers/net/tun.c +15 -0
index 65706824eb828..7c8db8f6f661e 100644
@@ -1971,6 +1971,14 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile,
 		int queue_len;
 
 		spin_lock_bh(&queue->lock);
+
+		if (unlikely(tfile->detached)) {
+			spin_unlock_bh(&queue->lock);
+			rcu_read_unlock();
+			err = -EBUSY;
+			goto free_skb;
+		}
+
 		__skb_queue_tail(queue, skb);
 		queue_len = skb_queue_len(queue);
 		spin_unlock(&queue->lock);
@@ -2506,6 +2514,13 @@ static int tun_xdp_one(struct tun_struct *tun,
 	if (tfile->napi_enabled) {
 		queue = &tfile->sk.sk_write_queue;
 		spin_lock(&queue->lock);
+
+		if (unlikely(tfile->detached)) {
+			spin_unlock(&queue->lock);
+			kfree_skb(skb);
+			return -EBUSY;
+		}
+
 		__skb_queue_tail(queue, skb);
 		spin_unlock(&queue->lock);
 		ret = 1;
Signature · 1 line
2.39.2
author maintainer
#216 lore raw

PATCH 216/292 cassini: Fix a memory leak in the error handling path of cas_init_one()

In reply to Greg Kroah-Hartman

From: Christophe JAILLET <[email protected]>

[ Upstream commit 412cd77a2c24b191c65ea53025222418db09817c ]

cas_saturn_firmware_init() allocates some memory using vmalloc(). This memory is freed in the .remove() function but not it the error handling path of the probe.

Add the missing vfree() to avoid a memory leak, should an error occur.

Fixes fcaa40669cd7 ("cassini: use request_firmware")
---
 drivers/net/ethernet/sun/cassini.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/net/ethernet/sun/cassini.c +2 -0
index 0aca193d9550d..800956d5464b4 100644
@@ -5095,6 +5095,8 @@ static int cas_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 		cas_shutdown(cp);
 	mutex_unlock(&cp->pm_mutex);
 
+	vfree(cp->fw_data);
+
 	pci_iounmap(pdev, cp->regs);
 
 
Signature · 1 line
2.39.2
author maintainer
#217 lore raw

PATCH 217/292 net: dsa: mv88e6xxx: Fix mv88e6393x EPC write command offset

In reply to Greg Kroah-Hartman

From: Marco Migliore <[email protected]>

[ Upstream commit 1323e0c6e1d7e103d59384c3ac50f72b17a6936c ]

According to datasheet, the command opcode must be specified into bits [14:12] of the Extended Port Control register (EPC).

Fixes de776d0d316f ("net: dsa: mv88e6xxx: add support for mv88e6393x family")
---
 drivers/net/dsa/mv88e6xxx/port.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/dsa/mv88e6xxx/port.h +1 -1
index cb04243f37c1e..a91e22d9a6cb3 100644
@@ -276,7 +276,7 @@
 /* Offset 0x10: Extended Port Control Command */
 #define MV88E6393X_PORT_EPC_CMD		0x10
 #define MV88E6393X_PORT_EPC_CMD_BUSY	0x8000
-#define MV88E6393X_PORT_EPC_CMD_WRITE	0x0300
+#define MV88E6393X_PORT_EPC_CMD_WRITE	0x3000
 #define MV88E6393X_PORT_EPC_INDEX_PORT_ETYPE	0x02
 
 /* Offset 0x11: Extended Port Control Data */
Signature · 1 line
2.39.2
author maintainer
#218 lore raw

PATCH 218/292 igb: fix bit_shift to be in [1..8] range

In reply to Greg Kroah-Hartman

From: Aleksandr Loktionov <[email protected]>

[ Upstream commit 60d758659f1fb49e0d5b6ac2691ede8c0958795b ]

In igb_hash_mc_addr() the expression:

        "mc_addr[4] >> 8 - bit_shift", right shifting "mc_addr[4]"

shift by more than 7 bits always yields zero, so hash becomes not so different. Add initialization with bit_shift = 1 and add a loop condition to ensure bit_shift will be always in [1..8] range.

---
 drivers/net/ethernet/intel/igb/e1000_mac.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/intel/igb/e1000_mac.c +2 -2
index 205d577bdbbaa..caf91c6f52b4d 100644
@@ -426,7 +426,7 @@ void igb_mta_set(struct e1000_hw *hw, u32 hash_value)
 static u32 igb_hash_mc_addr(struct e1000_hw *hw, u8 *mc_addr)
 {
 	u32 hash_value, hash_mask;
-	u8 bit_shift = 0;
+	u8 bit_shift = 1;
 
 	/* Register count multiplied by bits per register */
 	hash_mask = (hw->mac.mta_reg_count * 32) - 1;
@@ -434,7 +434,7 @@ static u32 igb_hash_mc_addr(struct e1000_hw *hw, u8 *mc_addr)
 	/* For a mc_filter_type of 0, bit_shift is the number of left-shifts
 	 * where 0xFF would still fall within the hash mask.
 	 */
-	while (hash_mask >> bit_shift != 0xFF)
+	while (hash_mask >> bit_shift != 0xFF && bit_shift < 4)
 		bit_shift++;
 
 	/* The portion of the address that is used for the hash table
Signature · 1 line
2.39.2
author maintainer
#219 lore raw

PATCH 219/292 vlan: fix a potential uninit-value in vlan_dev_hard_start_xmit()

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit dacab578c7c6cd06c50c89dfa36b0e0f10decd4e ]

syzbot triggered the following splat [1], sending an empty message through pppoe_sendmsg().

When VLAN_FLAG_REORDER_HDR flag is set, vlan_dev_hard_header() does not push extra bytes for the VLAN header, because vlan is offloaded.

Unfortunately vlan_dev_hard_start_xmit() first reads veth->h_vlan_proto before testing (vlan->flags & VLAN_FLAG_REORDER_HDR).

We need to swap the two conditions.

[1]

BUG: KMSAN: uninit-value in vlan_dev_hard_start_xmit+0x171/0x7f0 net/8021q/vlan_dev.c:111
vlan_dev_hard_start_xmit+0x171/0x7f0 net/8021q/vlan_dev.c:111
__netdev_start_xmit include/linux/netdevice.h:4883 [inline]
netdev_start_xmit include/linux/netdevice.h:4897 [inline]
xmit_one net/core/dev.c:3580 [inline]
dev_hard_start_xmit+0x253/0xa20 net/core/dev.c:3596
__dev_queue_xmit+0x3c7f/0x5ac0 net/core/dev.c:4246
dev_queue_xmit include/linux/netdevice.h:3053 [inline]
pppoe_sendmsg+0xa93/0xb80 drivers/net/ppp/pppoe.c:900
sock_sendmsg_nosec net/socket.c:724 [inline]
sock_sendmsg net/socket.c:747 [inline]
____sys_sendmsg+0xa24/0xe40 net/socket.c:2501
___sys_sendmsg+0x2a1/0x3f0 net/socket.c:2555
__sys_sendmmsg+0x411/0xa50 net/socket.c:2641
__do_sys_sendmmsg net/socket.c:2670 [inline]
__se_sys_sendmmsg net/socket.c:2667 [inline]
__x64_sys_sendmmsg+0xbc/0x120 net/socket.c:2667
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

Uninit was created at:

slab_post_alloc_hook+0x12d/0xb60 mm/slab.h:774
slab_alloc_node mm/slub.c:3452 [inline]
kmem_cache_alloc_node+0x543/0xab0 mm/slub.c:3497
kmalloc_reserve+0x148/0x470 net/core/skbuff.c:520
__alloc_skb+0x3a7/0x850 net/core/skbuff.c:606
alloc_skb include/linux/skbuff.h:1277 [inline]
sock_wmalloc+0xfe/0x1a0 net/core/sock.c:2583
pppoe_sendmsg+0x3af/0xb80 drivers/net/ppp/pppoe.c:867
sock_sendmsg_nosec net/socket.c:724 [inline]
sock_sendmsg net/socket.c:747 [inline]
____sys_sendmsg+0xa24/0xe40 net/socket.c:2501
___sys_sendmsg+0x2a1/0x3f0 net/socket.c:2555
__sys_sendmmsg+0x411/0xa50 net/socket.c:2641
__do_sys_sendmmsg net/socket.c:2670 [inline]
__se_sys_sendmmsg net/socket.c:2667 [inline]
__x64_sys_sendmmsg+0xbc/0x120 net/socket.c:2667
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

CPU: 0 PID: 29770 Comm: syz-executor.0 Not tainted 6.3.0-rc6-syzkaller-gc478e5b17829 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/30/2023
Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 net/8021q/vlan_dev.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
net/8021q/vlan_dev.c +2 -2
index 07e86d03d4bae..d3e511e1eba8a 100644
@@ -108,8 +108,8 @@ static netdev_tx_t vlan_dev_hard_start_xmit(struct sk_buff *skb,
 	 * NOTE: THIS ASSUMES DIX ETHERNET, SPECIFICALLY NOT SUPPORTING
 	 * OTHER THINGS LIKE FDDI/TokenRing/802.3 SNAPs...
 	 */
-	if (veth->h_vlan_proto != vlan->vlan_proto ||
-	    vlan->flags & VLAN_FLAG_REORDER_HDR) {
+	if (vlan->flags & VLAN_FLAG_REORDER_HDR ||
+	    veth->h_vlan_proto != vlan->vlan_proto) {
 		u16 vlan_tci;
 		vlan_tci = vlan->vlan_id;
 		vlan_tci |= vlan_dev_get_egress_qos_mask(dev, skb->priority);
Signature · 1 line
2.39.2
author maintainer
#220 lore raw

PATCH 220/292 net: wwan: iosm: fix NULL pointer dereference when removing device

In reply to Greg Kroah-Hartman

From: M Chetan Kumar <[email protected]>

[ Upstream commit 60829145f1e2650b31ebe6a0ec70a9725b38fa2c ]

In suspend and resume cycle, the removal and rescan of device ends up in NULL pointer dereference.

During driver initialization, if the ipc_imem_wwan_channel_init() fails to get the valid device capabilities it returns an error and further no resource (wwan struct) will be allocated. Now in this situation if driver removal procedure is initiated it would result in NULL pointer exception since unallocated wwan struct is dereferenced inside ipc_wwan_deinit().

ipc_imem_run_state_worker() to handle the called functions return value and to release the resource in failure case. It also reports the link down event in failure cases. The user space application can handle this event to do a device reset for restoring the device communication.

---
 drivers/net/wwan/iosm/iosm_ipc_imem.c     | 27 ++++++++++++++++++-----
 drivers/net/wwan/iosm/iosm_ipc_imem_ops.c | 12 ++++++----
 drivers/net/wwan/iosm/iosm_ipc_imem_ops.h |  6 +++--
 3 files changed, 33 insertions(+), 12 deletions(-)
drivers/net/wwan/iosm/iosm_ipc_imem.c +21 -6
index 1e6a479766429..8ccd4d26b9060 100644
@@ -565,24 +565,32 @@ static void ipc_imem_run_state_worker(struct work_struct *instance)
 	struct ipc_mux_config mux_cfg;
 	struct iosm_imem *ipc_imem;
 	u8 ctrl_chl_idx = 0;
+	int ret;
 
 	ipc_imem = container_of(instance, struct iosm_imem, run_state_worker);
 
 	if (ipc_imem->phase != IPC_P_RUN) {
 		dev_err(ipc_imem->dev,
 			"Modem link down. Exit run state worker.");
-		return;
+		goto err_out;
 	}
 
 	if (test_and_clear_bit(IOSM_DEVLINK_INIT, &ipc_imem->flag))
 		ipc_devlink_deinit(ipc_imem->ipc_devlink);
 
-	if (!ipc_imem_setup_cp_mux_cap_init(ipc_imem, &mux_cfg))
-		ipc_imem->mux = ipc_mux_init(&mux_cfg, ipc_imem);
+	ret = ipc_imem_setup_cp_mux_cap_init(ipc_imem, &mux_cfg);
+	if (ret < 0)
+		goto err_out;
+
+	ipc_imem->mux = ipc_mux_init(&mux_cfg, ipc_imem);
+	if (!ipc_imem->mux)
+		goto err_out;
+
+	ret = ipc_imem_wwan_channel_init(ipc_imem, mux_cfg.protocol);
+	if (ret < 0)
+		goto err_ipc_mux_deinit;
 
-	ipc_imem_wwan_channel_init(ipc_imem, mux_cfg.protocol);
-	if (ipc_imem->mux)
-		ipc_imem->mux->wwan = ipc_imem->wwan;
+	ipc_imem->mux->wwan = ipc_imem->wwan;
 
 	while (ctrl_chl_idx < IPC_MEM_MAX_CHANNELS) {
 		if (!ipc_chnl_cfg_get(&chnl_cfg_port, ctrl_chl_idx)) {
@@ -615,6 +623,13 @@ static void ipc_imem_run_state_worker(struct work_struct *instance)
 
 	/* Complete all memory stores after setting bit */
 	smp_mb__after_atomic();
+
+	return;
+
+err_ipc_mux_deinit:
+	ipc_mux_deinit(ipc_imem->mux);
+err_out:
+	ipc_uevent_send(ipc_imem->dev, UEVENT_CD_READY_LINK_DOWN);
 }
 
 static void ipc_imem_handle_irq(struct iosm_imem *ipc_imem, int irq)
drivers/net/wwan/iosm/iosm_ipc_imem_ops.c +8 -4
index 66b90cc4c3460..109cf89304888 100644
@@ -77,8 +77,8 @@ int ipc_imem_sys_wwan_transmit(struct iosm_imem *ipc_imem,
 }
 
 /* Initialize wwan channel */
-void ipc_imem_wwan_channel_init(struct iosm_imem *ipc_imem,
-				enum ipc_mux_protocol mux_type)
+int ipc_imem_wwan_channel_init(struct iosm_imem *ipc_imem,
+			       enum ipc_mux_protocol mux_type)
 {
 	struct ipc_chnl_cfg chnl_cfg = { 0 };
 
@@ -87,7 +87,7 @@ void ipc_imem_wwan_channel_init(struct iosm_imem *ipc_imem,
 	/* If modem version is invalid (0xffffffff), do not initialize WWAN. */
 	if (ipc_imem->cp_version == -1) {
 		dev_err(ipc_imem->dev, "invalid CP version");
-		return;
+		return -EIO;
 	}
 
 	ipc_chnl_cfg_get(&chnl_cfg, ipc_imem->nr_of_channels);
@@ -104,9 +104,13 @@ void ipc_imem_wwan_channel_init(struct iosm_imem *ipc_imem,
 
 	/* WWAN registration. */
 	ipc_imem->wwan = ipc_wwan_init(ipc_imem, ipc_imem->dev);
-	if (!ipc_imem->wwan)
+	if (!ipc_imem->wwan) {
 		dev_err(ipc_imem->dev,
 			"failed to register the ipc_wwan interfaces");
+		return -ENOMEM;
+	}
+
+	return 0;
 }
 
 /* Map SKB to DMA for transfer */
drivers/net/wwan/iosm/iosm_ipc_imem_ops.h +4 -2
index f8afb217d9e2f..026c5bd0f9992 100644
@@ -91,9 +91,11 @@ int ipc_imem_sys_wwan_transmit(struct iosm_imem *ipc_imem, int if_id,
  *				MUX.
  * @ipc_imem:		Pointer to iosm_imem struct.
  * @mux_type:		Type of mux protocol.
+ *
+ * Return: 0 on success and failure value on error
  */
-void ipc_imem_wwan_channel_init(struct iosm_imem *ipc_imem,
-				enum ipc_mux_protocol mux_type);
+int ipc_imem_wwan_channel_init(struct iosm_imem *ipc_imem,
+			       enum ipc_mux_protocol mux_type);
 
 /**
  * ipc_imem_sys_devlink_open - Open a Flash/CD Channel link to CP
Signature · 1 line
2.39.2
author maintainer
#221 lore raw

PATCH 221/292 net: pcs: xpcs: fix C73 AN not getting enabled

In reply to Greg Kroah-Hartman

From: Vladimir Oltean <[email protected]>

[ Upstream commit c46e78ba9a7a09da4f192dc8df15c4e8a07fb9e0 ]

The XPCS expects clause 73 (copper backplane) autoneg to follow the ethtool autoneg bit. It actually did that until the blamed commit inaptly replaced state->an_enabled (coming from ethtool) with phylink_autoneg_inband() (coming from the device tree or struct phylink_config), as part of an unrelated phylink_pcs API conversion.

Russell King suggests that state->an_enabled from the original code was just a proxy for the ethtool Autoneg bit, and that the correct way of restoring the functionality is to check for this bit in the advertising mask.

---
 drivers/net/pcs/pcs-xpcs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/pcs/pcs-xpcs.c +1 -1
index dd88624593c71..3f882bce37f42 100644
@@ -881,7 +881,7 @@ int xpcs_do_config(struct dw_xpcs *xpcs, phy_interface_t interface,
 
 	switch (compat->an_mode) {
 	case DW_AN_C73:
-		if (phylink_autoneg_inband(mode)) {
+		if (test_bit(ETHTOOL_LINK_MODE_Autoneg_BIT, advertising)) {
 			ret = xpcs_config_aneg_c73(xpcs, compat);
 			if (ret)
 				return ret;
Signature · 1 line
2.39.2
author maintainer
#222 lore raw

PATCH 222/292 net: selftests: Fix optstring

In reply to Greg Kroah-Hartman

From: Benjamin Poirier <[email protected]>

[ Upstream commit 9ba9485b87ac97fd159abdb4cbd53099bc9f01c6 ]

The cited commit added a stray colon to the 'v' option. That makes the option work incorrectly.

ex:
tools/testing/selftests/net# ./fib_nexthops.sh -v
(should enable verbose mode, instead it shows help text due to missing arg)

Fixes 5feba4727395 ("selftests: fib_nexthops: Make ping timeout configurable")
---
 tools/testing/selftests/net/fib_nexthops.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/testing/selftests/net/fib_nexthops.sh +1 -1
index a47b26ab48f23..0f5e88c8f4ffe 100755
@@ -2283,7 +2283,7 @@ EOF
 ################################################################################
 # main
 
-while getopts :t:pP46hv:w: o
+while getopts :t:pP46hvw: o
 do
 	case $o in
 		t) TESTS=$OPTARG;;
Signature · 1 line
2.39.2
author maintainer
#223 lore raw

PATCH 223/292 netfilter: nf_tables: fix nft_trans type confusion

In reply to Greg Kroah-Hartman

From: Florian Westphal <[email protected]>

[ Upstream commit e3c361b8acd636f5fe80c02849ca175201edf10c ]

nft_trans_FOO objects all share a common nft_trans base structure, but trailing fields depend on the real object size. Access is only safe after trans->msg_type check.

Check for rule type first. Found by code inspection.

Fixes 1a94e38d254b ("netfilter: nf_tables: add NFTA_RULE_ID attribute")
---
 net/netfilter/nf_tables_api.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)
net/netfilter/nf_tables_api.c +1 -3
index f663262df6987..31775d54f4b40 100644
@@ -3692,12 +3692,10 @@ static struct nft_rule *nft_rule_lookup_byid(const struct net *net,
 	struct nft_trans *trans;
 
 	list_for_each_entry(trans, &nft_net->commit_list, list) {
-		struct nft_rule *rule = nft_trans_rule(trans);
-
 		if (trans->msg_type == NFT_MSG_NEWRULE &&
 		    trans->ctx.chain == chain &&
 		    id == nft_trans_rule_id(trans))
-			return rule;
+			return nft_trans_rule(trans);
 	}
 	return ERR_PTR(-ENOENT);
 }
Signature · 1 line
2.39.2
author maintainer
#224 lore raw

PATCH 224/292 netfilter: nft_set_rbtree: fix null deref on element insertion

In reply to Greg Kroah-Hartman

From: Florian Westphal <[email protected]>

[ Upstream commit 61ae320a29b0540c16931816299eb86bf2b66c08 ]

There is no guarantee that rb_prev() will not return NULL in nft_rbtree_gc_elem():

general protection fault, probably for non-canonical address 0xdffffc0000000003: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000018-0x000000000000001f]
 nft_add_set_elem+0x14b0/0x2990
  nf_tables_newsetelem+0x528/0xb30

Furthermore, there is a possible use-after-free while iterating, 'node' can be free'd so we need to cache the next value to use.

Fixes c9e6978e2725 ("netfilter: nft_set_rbtree: Switch to node list walk for overlap detection")
---
 net/netfilter/nft_set_rbtree.c | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)
net/netfilter/nft_set_rbtree.c +13 -7
index 19ea4d3c35535..2f114aa10f1a7 100644
@@ -221,7 +221,7 @@ static int nft_rbtree_gc_elem(const struct nft_set *__set,
 {
 	struct nft_set *set = (struct nft_set *)__set;
 	struct rb_node *prev = rb_prev(&rbe->node);
-	struct nft_rbtree_elem *rbe_prev;
+	struct nft_rbtree_elem *rbe_prev = NULL;
 	struct nft_set_gc_batch *gcb;
 
 	gcb = nft_set_gc_batch_check(set, NULL, GFP_ATOMIC);
@@ -229,17 +229,21 @@ static int nft_rbtree_gc_elem(const struct nft_set *__set,
 		return -ENOMEM;
 
 	/* search for expired end interval coming before this element. */
-	do {
+	while (prev) {
 		rbe_prev = rb_entry(prev, struct nft_rbtree_elem, node);
 		if (nft_rbtree_interval_end(rbe_prev))
 			break;
 
 		prev = rb_prev(prev);
-	} while (prev != NULL);
+	}
+
+	if (rbe_prev) {
+		rb_erase(&rbe_prev->node, &priv->root);
+		atomic_dec(&set->nelems);
+	}
 
-	rb_erase(&rbe_prev->node, &priv->root);
 	rb_erase(&rbe->node, &priv->root);
-	atomic_sub(2, &set->nelems);
+	atomic_dec(&set->nelems);
 
 	nft_set_gc_batch_add(gcb, rbe);
 	nft_set_gc_batch_complete(gcb);
@@ -268,7 +272,7 @@ static int __nft_rbtree_insert(const struct net *net, const struct nft_set *set,
 			       struct nft_set_ext **ext)
 {
 	struct nft_rbtree_elem *rbe, *rbe_le = NULL, *rbe_ge = NULL;
-	struct rb_node *node, *parent, **p, *first = NULL;
+	struct rb_node *node, *next, *parent, **p, *first = NULL;
 	struct nft_rbtree *priv = nft_set_priv(set);
 	u8 genmask = nft_genmask_next(net);
 	int d, err;
@@ -307,7 +311,9 @@ static int __nft_rbtree_insert(const struct net *net, const struct nft_set *set,
 	 * Values stored in the tree are in reversed order, starting from
 	 * highest to lowest value.
 	 */
-	for (node = first; node != NULL; node = rb_next(node)) {
+	for (node = first; node != NULL; node = next) {
+		next = rb_next(node);
+
 		rbe = rb_entry(node, struct nft_rbtree_elem, node);
 
 		if (!nft_set_elem_active(&rbe->ext, genmask))
Signature · 1 line
2.39.2
author maintainer
#225 lore raw

PATCH 225/292 bridge: always declare tunnel functions

In reply to Greg Kroah-Hartman

From: Arnd Bergmann <[email protected]>

[ Upstream commit 89dcd87ce534a3a7f267cfd58505803006f51301 ]

When CONFIG_BRIDGE_VLAN_FILTERING is disabled, two functions are still defined but have no prototype or caller. This causes a W=1 warning for the missing prototypes:

net/bridge/br_netlink_tunnel.c:29:6: error: no previous prototype for 'vlan_tunid_inrange' [-Werror=missing-prototypes] net/bridge/br_netlink_tunnel.c:199:5: error: no previous prototype for 'br_vlan_tunnel_info' [-Werror=missing-prototypes]

The functions are already contitional on CONFIG_BRIDGE_VLAN_FILTERING, and I coulnd't easily figure out the right set of #ifdefs, so just move the declarations out of the #ifdef to avoid the warning, at a small cost in code size over a more elaborate fix.

Fixes 188c67dd1906 ("net: bridge: vlan options: add support for tunnel id dumping")
Fixes 569da0822808 ("net: bridge: vlan options: add support for tunnel mapping set/del")
---
 net/bridge/br_private_tunnel.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
net/bridge/br_private_tunnel.h +4 -4
index 2b053289f0166..efb096025151a 100644
@@ -27,6 +27,10 @@ int br_process_vlan_tunnel_info(const struct net_bridge *br,
 int br_get_vlan_tunnel_info_size(struct net_bridge_vlan_group *vg);
 int br_fill_vlan_tunnel_info(struct sk_buff *skb,
 			     struct net_bridge_vlan_group *vg);
+bool vlan_tunid_inrange(const struct net_bridge_vlan *v_curr,
+			const struct net_bridge_vlan *v_last);
+int br_vlan_tunnel_info(const struct net_bridge_port *p, int cmd,
+			u16 vid, u32 tun_id, bool *changed);
 
 #ifdef CONFIG_BRIDGE_VLAN_FILTERING
 /* br_vlan_tunnel.c */
@@ -43,10 +47,6 @@ void br_handle_ingress_vlan_tunnel(struct sk_buff *skb,
 				   struct net_bridge_vlan_group *vg);
 int br_handle_egress_vlan_tunnel(struct sk_buff *skb,
 				 struct net_bridge_vlan *vlan);
-bool vlan_tunid_inrange(const struct net_bridge_vlan *v_curr,
-			const struct net_bridge_vlan *v_last);
-int br_vlan_tunnel_info(const struct net_bridge_port *p, int cmd,
-			u16 vid, u32 tun_id, bool *changed);
 #else
 static inline int vlan_tunnel_init(struct net_bridge_vlan_group *vg)
 {
Signature · 1 line
2.39.2
author maintainer
#226 lore raw

PATCH 226/292 ALSA: usb-audio: Add a sample rate workaround for Line6 Pod Go

In reply to Greg Kroah-Hartman

From: Takashi Iwai <[email protected]>

commit 359b4315471181f108723c61612d96e383e56179 upstream.

Line6 Pod Go (0e41:424b) requires the similar workaround for the fixed 48k sample rate like other Line6 models. This patch adds the corresponding entry to line6_parse_audio_format_rate_quirk().

---
 sound/usb/format.c |    1 +
 1 file changed, 1 insertion(+)
sound/usb/format.c +1 -0
@@ -423,6 +423,7 @@ static int line6_parse_audio_format_rate
 	case USB_ID(0x0e41, 0x4248): /* Line6 Helix >= fw 2.82 */
 	case USB_ID(0x0e41, 0x4249): /* Line6 Helix Rack >= fw 2.82 */
 	case USB_ID(0x0e41, 0x424a): /* Line6 Helix LT >= fw 2.82 */
+	case USB_ID(0x0e41, 0x424b): /* Line6 Pod Go */
 	case USB_ID(0x19f7, 0x0011): /* Rode Rodecaster Pro */
 		return set_fixed_rate(fp, 48000, SNDRV_PCM_RATE_48000);
 	}
author maintainer
#227 lore raw

PATCH 227/292 USB: usbtmc: Fix direction for 0-length ioctl control messages

In reply to Greg Kroah-Hartman

From: Alan Stern <[email protected]>

commit 94d25e9128988c6a1fc9070f6e98215a95795bd8 upstream.

The syzbot fuzzer found a problem in the usbtmc driver: When a user submits an ioctl for a 0-length control transfer, the driver does not check that the direction is set to OUT:

------------[ cut here ]------------
usb 3-1: BOGUS control dir, pipe 80000b80 doesn't match bRequestType fd
WARNING: CPU: 0 PID: 5100 at drivers/usb/core/urb.c:411 usb_submit_urb+0x14a7/0x1880 drivers/usb/core/urb.c:411
Modules linked in:
CPU: 0 PID: 5100 Comm: syz-executor428 Not tainted 6.3.0-syzkaller-12049-g58390c8ce1bd #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 04/14/2023
RIP: 0010:usb_submit_urb+0x14a7/0x1880 drivers/usb/core/urb.c:411
Code: 7c 24 40 e8 1b 13 5c fb 48 8b 7c 24 40 e8 21 1d f0 fe 45 89 e8 44 89 f1 4c 89 e2 48 89 c6 48 c7 c7 e0 b5 fc 8a e8 19 c8 23 fb <0f> 0b e9 9f ee ff ff e8 ed 12 5c fb 0f b6 1d 12 8a 3c 08 31 ff 41
RSP: 0018:ffffc90003d2fb00 EFLAGS: 00010282
RAX: 0000000000000000 RBX: ffff8880789e9058 RCX: 0000000000000000
RDX: ffff888029593b80 RSI: ffffffff814c1447 RDI: 0000000000000001
RBP: ffff88801ea742f8 R08: 0000000000000001 R09: 0000000000000000
R10: 0000000000000001 R11: 0000000000000001 R12: ffff88802915e528
R13: 00000000000000fd R14: 0000000080000b80 R15: ffff8880222b3100
FS:  0000555556ca63c0(0000) GS:ffff8880b9800000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f9ef4d18150 CR3: 0000000073e5b000 CR4: 00000000003506f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
 usb_start_wait_urb+0x101/0x4b0 drivers/usb/core/message.c:58
 usb_internal_control_msg drivers/usb/core/message.c:102 [inline]
 usb_control_msg+0x320/0x4a0 drivers/usb/core/message.c:153
 usbtmc_ioctl_request drivers/usb/class/usbtmc.c:1954 [inline]
 usbtmc_ioctl+0x1b3d/0x2840 drivers/usb/class/usbtmc.c:2097

To fix this, we must override the direction in the bRequestType field of the control request structure when the length is 0.

---
 drivers/usb/class/usbtmc.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/usb/class/usbtmc.c +2 -0
@@ -1928,6 +1928,8 @@ static int usbtmc_ioctl_request(struct u
 
 	if (request.req.wLength > USBTMC_BUFSIZE)
 		return -EMSGSIZE;
+	if (request.req.wLength == 0)	/* Length-0 requests are never IN */
+		request.req.bRequestType &= ~USB_DIR_IN;
 
 	is_in = request.req.bRequestType & USB_DIR_IN;
 
author maintainer
#228 lore raw

PATCH 228/292 usb-storage: fix deadlock when a scsi command timeouts more than once

In reply to Greg Kroah-Hartman

From: Maxime Bizon <[email protected]>

commit a398d5eac6984316e71474e25b975688f282379b upstream.

With faulty usb-storage devices, read/write can timeout, in that case the SCSI layer will abort and re-issue the command. USB storage has no internal timeout, it relies on SCSI layer aborting commands via .eh_abort_handler() for non those responsive devices.

After two consecutive timeouts of the same command, SCSI layer calls .eh_device_reset_handler(), without calling .eh_abort_handler() first.

With usb-storage, this causes a deadlock:

  -> .eh_device_reset_handler
    -> device_reset
      -> mutex_lock(&(us->dev_mutex));

mutex already by usb_stor_control_thread(), which is waiting for command completion:

-> usb_stor_control_thread (mutex taken here)

    -> usb_stor_invoke_transport
      -> usb_stor_Bulk_transport
        -> usb_stor_bulk_srb
	  -> usb_stor_bulk_transfer_sglist
	    -> usb_sg_wait

Make sure we cancel any pending command in .eh_device_reset_handler() to avoid this.

---
 drivers/usb/storage/scsiglue.c |   28 +++++++++++++++++++++-------
 1 file changed, 21 insertions(+), 7 deletions(-)
drivers/usb/storage/scsiglue.c +21 -7
@@ -406,22 +406,25 @@ static DEF_SCSI_QCMD(queuecommand)
  ***********************************************************************/
 
 /* Command timeout and abort */
-static int command_abort(struct scsi_cmnd *srb)
+static int command_abort_matching(struct us_data *us, struct scsi_cmnd *srb_match)
 {
-	struct us_data *us = host_to_us(srb->device->host);
-
-	usb_stor_dbg(us, "%s called\n", __func__);
-
 	/*
 	 * us->srb together with the TIMED_OUT, RESETTING, and ABORTING
 	 * bits are protected by the host lock.
 	 */
 	scsi_lock(us_to_host(us));
 
-	/* Is this command still active? */
-	if (us->srb != srb) {
+	/* is there any active pending command to abort ? */
+	if (!us->srb) {
 		scsi_unlock(us_to_host(us));
 		usb_stor_dbg(us, "-- nothing to abort\n");
+		return SUCCESS;
+	}
+
+	/* Does the command match the passed srb if any ? */
+	if (srb_match && us->srb != srb_match) {
+		scsi_unlock(us_to_host(us));
+		usb_stor_dbg(us, "-- pending command mismatch\n");
 		return FAILED;
 	}
 
@@ -444,6 +447,14 @@ static int command_abort(struct scsi_cmn
 	return SUCCESS;
 }
 
+static int command_abort(struct scsi_cmnd *srb)
+{
+	struct us_data *us = host_to_us(srb->device->host);
+
+	usb_stor_dbg(us, "%s called\n", __func__);
+	return command_abort_matching(us, srb);
+}
+
 /*
  * This invokes the transport reset mechanism to reset the state of the
  * device
@@ -455,6 +466,9 @@ static int device_reset(struct scsi_cmnd
 
 	usb_stor_dbg(us, "%s called\n", __func__);
 
+	/* abort any pending command before reset */
+	command_abort_matching(us, NULL);
+
 	/* lock the device pointers and do the reset */
 	mutex_lock(&(us->dev_mutex));
 	result = us->transport_reset(us);
author maintainer
#229 lore raw

PATCH 229/292 USB: UHCI: adjust zhaoxin UHCI controllers OverCurrent bit value

In reply to Greg Kroah-Hartman

From: Weitao Wang <[email protected]>

commit dddb342b5b9e482bb213aecc08cbdb201ea4f8da upstream.

OverCurrent condition is not standardized in the UHCI spec. Zhaoxin UHCI controllers report OverCurrent bit active off. In order to handle OverCurrent condition correctly, the uhci-hcd driver needs to be told to expect the active-off behavior.

---
 drivers/usb/host/uhci-pci.c |   10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/usb/host/uhci-pci.c +6 -4
@@ -119,11 +119,13 @@ static int uhci_pci_init(struct usb_hcd
 
 	uhci->rh_numports = uhci_count_ports(hcd);
 
-	/* Intel controllers report the OverCurrent bit active on.
-	 * VIA controllers report it active off, so we'll adjust the
-	 * bit value.  (It's not standardized in the UHCI spec.)
+	/*
+	 * Intel controllers report the OverCurrent bit active on.  VIA
+	 * and ZHAOXIN controllers report it active off, so we'll adjust
+	 * the bit value.  (It's not standardized in the UHCI spec.)
 	 */
-	if (to_pci_dev(uhci_dev(uhci))->vendor == PCI_VENDOR_ID_VIA)
+	if (to_pci_dev(uhci_dev(uhci))->vendor == PCI_VENDOR_ID_VIA ||
+			to_pci_dev(uhci_dev(uhci))->vendor == PCI_VENDOR_ID_ZHAOXIN)
 		uhci->oc_low = 1;
 
 	/* HP's server management chip requires a longer port reset delay. */
author maintainer
#230 lore raw

PATCH 230/292 usb: dwc3: gadget: Improve dwc3_gadget_suspend() and dwc3_gadget_resume()

In reply to Greg Kroah-Hartman

From: Roger Quadros <[email protected]>

commit c8540870af4ce6ddeb27a7bb5498b75fb29b643c upstream.

Prevent -ETIMEDOUT error on .suspend().
e.g. If gadget driver is loaded and we are connected to a USB host, all transfers must be stopped before stopping the controller else we will not get a clean stop i.e. dwc3_gadget_run_stop() will take several seconds to complete and will return -ETIMEDOUT.

Handle error cases properly in dwc3_gadget_suspend(). Simplify dwc3_gadget_resume() by using the introduced helper function.

---
 drivers/usb/dwc3/gadget.c |   67 +++++++++++++++++++++++-----------------------
 1 file changed, 34 insertions(+), 33 deletions(-)
drivers/usb/dwc3/gadget.c +34 -33
@@ -2587,6 +2587,21 @@ static int dwc3_gadget_soft_disconnect(s
 	return ret;
 }
 
+static int dwc3_gadget_soft_connect(struct dwc3 *dwc)
+{
+	/*
+	 * In the Synopsys DWC_usb31 1.90a programming guide section
+	 * 4.1.9, it specifies that for a reconnect after a
+	 * device-initiated disconnect requires a core soft reset
+	 * (DCTL.CSftRst) before enabling the run/stop bit.
+	 */
+	dwc3_core_soft_reset(dwc);
+
+	dwc3_event_buffers_setup(dwc);
+	__dwc3_gadget_start(dwc);
+	return dwc3_gadget_run_stop(dwc, true);
+}
+
 static int dwc3_gadget_pullup(struct usb_gadget *g, int is_on)
 {
 	struct dwc3		*dwc = gadget_to_dwc(g);
@@ -2625,21 +2640,10 @@ static int dwc3_gadget_pullup(struct usb
 
 	synchronize_irq(dwc->irq_gadget);
 
-	if (!is_on) {
+	if (!is_on)
 		ret = dwc3_gadget_soft_disconnect(dwc);
-	} else {
-		/*
-		 * In the Synopsys DWC_usb31 1.90a programming guide section
-		 * 4.1.9, it specifies that for a reconnect after a
-		 * device-initiated disconnect requires a core soft reset
-		 * (DCTL.CSftRst) before enabling the run/stop bit.
-		 */
-		dwc3_core_soft_reset(dwc);
-
-		dwc3_event_buffers_setup(dwc);
-		__dwc3_gadget_start(dwc);
-		ret = dwc3_gadget_run_stop(dwc, true);
-	}
+	else
+		ret = dwc3_gadget_soft_connect(dwc);
 
 	pm_runtime_put(dwc->dev);
 
@@ -4555,42 +4559,39 @@ void dwc3_gadget_exit(struct dwc3 *dwc)
 int dwc3_gadget_suspend(struct dwc3 *dwc)
 {
 	unsigned long flags;
+	int ret;
 
 	if (!dwc->gadget_driver)
 		return 0;
 
-	dwc3_gadget_run_stop(dwc, false);
+	ret = dwc3_gadget_soft_disconnect(dwc);
+	if (ret)
+		goto err;
 
 	spin_lock_irqsave(&dwc->lock, flags);
 	dwc3_disconnect_gadget(dwc);
-	__dwc3_gadget_stop(dwc);
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
 	return 0;
+
+err:
+	/*
+	 * Attempt to reset the controller's state. Likely no
+	 * communication can be established until the host
+	 * performs a port reset.
+	 */
+	if (dwc->softconnect)
+		dwc3_gadget_soft_connect(dwc);
+
+	return ret;
 }
 
 int dwc3_gadget_resume(struct dwc3 *dwc)
 {
-	int			ret;
-
 	if (!dwc->gadget_driver || !dwc->softconnect)
 		return 0;
 
-	ret = __dwc3_gadget_start(dwc);
-	if (ret < 0)
-		goto err0;
-
-	ret = dwc3_gadget_run_stop(dwc, true);
-	if (ret < 0)
-		goto err1;
-
-	return 0;
-
-err1:
-	__dwc3_gadget_stop(dwc);
-
-err0:
-	return ret;
+	return dwc3_gadget_soft_connect(dwc);
 }
 
 void dwc3_gadget_process_pending_events(struct dwc3 *dwc)
author maintainer
#231 lore raw

PATCH 231/292 usb: dwc3: debugfs: Resume dwc3 before accessing registers

In reply to Greg Kroah-Hartman

From: Udipto Goswami <[email protected]>

commit 614ce6a2ea50068b45339257891e51e639ac9001 upstream.

When the dwc3 device is runtime suspended, various required clocks are in disabled state and it is not guaranteed that access to any registers would work. Depending on the SoC glue, a register read could be as benign as returning 0 or be fatal enough to hang the system.

In order to prevent such scenarios of fatal errors, make sure to resume dwc3 then allow the function to proceed.

Fixes 72246da40f37 ("usb: Introduce DesignWare USB3 DRD Driver")
Cc [email protected] #3.2: 30332eeefec8: debugfs: regset32: Add Runtime PM support
---
 drivers/usb/dwc3/debugfs.c |  109 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 109 insertions(+)
drivers/usb/dwc3/debugfs.c +109 -0
@@ -327,6 +327,11 @@ static int dwc3_lsp_show(struct seq_file
 	unsigned int		current_mode;
 	unsigned long		flags;
 	u32			reg;
+	int			ret;
+
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&dwc->lock, flags);
 	reg = dwc3_readl(dwc->regs, DWC3_GSTS);
@@ -345,6 +350,8 @@ static int dwc3_lsp_show(struct seq_file
 	}
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return 0;
 }
 
@@ -390,6 +397,11 @@ static int dwc3_mode_show(struct seq_fil
 	struct dwc3		*dwc = s->private;
 	unsigned long		flags;
 	u32			reg;
+	int			ret;
+
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&dwc->lock, flags);
 	reg = dwc3_readl(dwc->regs, DWC3_GCTL);
@@ -409,6 +421,8 @@ static int dwc3_mode_show(struct seq_fil
 		seq_printf(s, "UNKNOWN %08x\n", DWC3_GCTL_PRTCAP(reg));
 	}
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return 0;
 }
 
@@ -458,6 +472,11 @@ static int dwc3_testmode_show(struct seq
 	struct dwc3		*dwc = s->private;
 	unsigned long		flags;
 	u32			reg;
+	int			ret;
+
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&dwc->lock, flags);
 	reg = dwc3_readl(dwc->regs, DWC3_DCTL);
@@ -488,6 +507,8 @@ static int dwc3_testmode_show(struct seq
 		seq_printf(s, "UNKNOWN %d\n", reg);
 	}
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return 0;
 }
 
@@ -504,6 +525,7 @@ static ssize_t dwc3_testmode_write(struc
 	unsigned long		flags;
 	u32			testmode = 0;
 	char			buf[32];
+	int			ret;
 
 	if (copy_from_user(&buf, ubuf, min_t(size_t, sizeof(buf) - 1, count)))
 		return -EFAULT;
@@ -521,10 +543,16 @@ static ssize_t dwc3_testmode_write(struc
 	else
 		testmode = 0;
 
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
+
 	spin_lock_irqsave(&dwc->lock, flags);
 	dwc3_gadget_set_test_mode(dwc, testmode);
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return count;
 }
 
@@ -543,12 +571,18 @@ static int dwc3_link_state_show(struct s
 	enum dwc3_link_state	state;
 	u32			reg;
 	u8			speed;
+	int			ret;
+
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&dwc->lock, flags);
 	reg = dwc3_readl(dwc->regs, DWC3_GSTS);
 	if (DWC3_GSTS_CURMOD(reg) != DWC3_GSTS_CURMOD_DEVICE) {
 		seq_puts(s, "Not available\n");
 		spin_unlock_irqrestore(&dwc->lock, flags);
+		pm_runtime_put_sync(dwc->dev);
 		return 0;
 	}
 
@@ -561,6 +595,8 @@ static int dwc3_link_state_show(struct s
 		   dwc3_gadget_hs_link_string(state));
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return 0;
 }
 
@@ -579,6 +615,7 @@ static ssize_t dwc3_link_state_write(str
 	char			buf[32];
 	u32			reg;
 	u8			speed;
+	int			ret;
 
 	if (copy_from_user(&buf, ubuf, min_t(size_t, sizeof(buf) - 1, count)))
 		return -EFAULT;
@@ -598,10 +635,15 @@ static ssize_t dwc3_link_state_write(str
 	else
 		return -EINVAL;
 
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
+
 	spin_lock_irqsave(&dwc->lock, flags);
 	reg = dwc3_readl(dwc->regs, DWC3_GSTS);
 	if (DWC3_GSTS_CURMOD(reg) != DWC3_GSTS_CURMOD_DEVICE) {
 		spin_unlock_irqrestore(&dwc->lock, flags);
+		pm_runtime_put_sync(dwc->dev);
 		return -EINVAL;
 	}
 
@@ -611,12 +653,15 @@ static ssize_t dwc3_link_state_write(str
 	if (speed < DWC3_DSTS_SUPERSPEED &&
 	    state != DWC3_LINK_STATE_RECOV) {
 		spin_unlock_irqrestore(&dwc->lock, flags);
+		pm_runtime_put_sync(dwc->dev);
 		return -EINVAL;
 	}
 
 	dwc3_gadget_set_link_state(dwc, state);
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return count;
 }
 
@@ -640,6 +685,11 @@ static int dwc3_tx_fifo_size_show(struct
 	unsigned long		flags;
 	u32			mdwidth;
 	u32			val;
+	int			ret;
+
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&dwc->lock, flags);
 	val = dwc3_core_fifo_space(dep, DWC3_TXFIFO);
@@ -652,6 +702,8 @@ static int dwc3_tx_fifo_size_show(struct
 	seq_printf(s, "%u\n", val);
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return 0;
 }
 
@@ -662,6 +714,11 @@ static int dwc3_rx_fifo_size_show(struct
 	unsigned long		flags;
 	u32			mdwidth;
 	u32			val;
+	int			ret;
+
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&dwc->lock, flags);
 	val = dwc3_core_fifo_space(dep, DWC3_RXFIFO);
@@ -674,6 +731,8 @@ static int dwc3_rx_fifo_size_show(struct
 	seq_printf(s, "%u\n", val);
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return 0;
 }
 
@@ -683,12 +742,19 @@ static int dwc3_tx_request_queue_show(st
 	struct dwc3		*dwc = dep->dwc;
 	unsigned long		flags;
 	u32			val;
+	int			ret;
+
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&dwc->lock, flags);
 	val = dwc3_core_fifo_space(dep, DWC3_TXREQQ);
 	seq_printf(s, "%u\n", val);
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return 0;
 }
 
@@ -698,12 +764,19 @@ static int dwc3_rx_request_queue_show(st
 	struct dwc3		*dwc = dep->dwc;
 	unsigned long		flags;
 	u32			val;
+	int			ret;
+
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&dwc->lock, flags);
 	val = dwc3_core_fifo_space(dep, DWC3_RXREQQ);
 	seq_printf(s, "%u\n", val);
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return 0;
 }
 
@@ -713,12 +786,19 @@ static int dwc3_rx_info_queue_show(struc
 	struct dwc3		*dwc = dep->dwc;
 	unsigned long		flags;
 	u32			val;
+	int			ret;
+
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&dwc->lock, flags);
 	val = dwc3_core_fifo_space(dep, DWC3_RXINFOQ);
 	seq_printf(s, "%u\n", val);
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return 0;
 }
 
@@ -728,12 +808,19 @@ static int dwc3_descriptor_fetch_queue_s
 	struct dwc3		*dwc = dep->dwc;
 	unsigned long		flags;
 	u32			val;
+	int			ret;
+
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&dwc->lock, flags);
 	val = dwc3_core_fifo_space(dep, DWC3_DESCFETCHQ);
 	seq_printf(s, "%u\n", val);
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return 0;
 }
 
@@ -743,12 +830,19 @@ static int dwc3_event_queue_show(struct
 	struct dwc3		*dwc = dep->dwc;
 	unsigned long		flags;
 	u32			val;
+	int			ret;
+
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&dwc->lock, flags);
 	val = dwc3_core_fifo_space(dep, DWC3_EVENTQ);
 	seq_printf(s, "%u\n", val);
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return 0;
 }
 
@@ -793,6 +887,11 @@ static int dwc3_trb_ring_show(struct seq
 	struct dwc3		*dwc = dep->dwc;
 	unsigned long		flags;
 	int			i;
+	int			ret;
+
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&dwc->lock, flags);
 	if (dep->number <= 1) {
@@ -822,6 +921,8 @@ static int dwc3_trb_ring_show(struct seq
 out:
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return 0;
 }
 
@@ -834,6 +935,11 @@ static int dwc3_ep_info_register_show(st
 	u32			lower_32_bits;
 	u32			upper_32_bits;
 	u32			reg;
+	int			ret;
+
+	ret = pm_runtime_resume_and_get(dwc->dev);
+	if (ret < 0)
+		return ret;
 
 	spin_lock_irqsave(&dwc->lock, flags);
 	reg = DWC3_GDBGLSPMUX_EPSELECT(dep->number);
@@ -846,6 +952,8 @@ static int dwc3_ep_info_register_show(st
 	seq_printf(s, "0x%016llx\n", ep_info);
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
+	pm_runtime_put_sync(dwc->dev);
+
 	return 0;
 }
 
@@ -905,6 +1013,7 @@ void dwc3_debugfs_init(struct dwc3 *dwc)
 	dwc->regset->regs = dwc3_regs;
 	dwc->regset->nregs = ARRAY_SIZE(dwc3_regs);
 	dwc->regset->base = dwc->regs - DWC3_GLOBALS_REGS_START;
+	dwc->regset->dev = dwc->dev;
 
 	root = debugfs_create_dir(dev_name(dwc->dev), usb_debug_root);
 	dwc->debug_root = root;
author maintainer
#232 lore raw

PATCH 232/292 usb: gadget: u_ether: Fix host MAC address case

In reply to Greg Kroah-Hartman

From: Konrad Gräfe <[email protected]>

commit 3c0f4f09c063e143822393d99cb2b19a85451c07 upstream.

The CDC-ECM specification [1] requires to send the host MAC address as an uppercase hexadecimal string in chapter "5.4 Ethernet Networking Functional Descriptor":
The Unicode character is chosen from the set of values 30h through 39h and 41h through 46h (0-9 and A-F).

However, snprintf(.., "%pm", ..) generates a lowercase MAC address string. While most host drivers are tolerant to this, UsbNcm.sys on Windows 10 is not. Instead it uses a different MAC address with all bytes set to zero including and after the first byte containing a lowercase letter. On Windows 11 Microsoft fixed it, but apparently they did not backport the fix.

This change fixes the issue by upper-casing the MAC to comply with the specification.

[1]: https://www.usb.org/document-library/class-definitions-communication-devices-12, file ECM120.pdf

Fixes bcd4a1c40bee ("usb: gadget: u_ether: construct with default values and add setters/getters")
---
 drivers/usb/gadget/function/u_ether.c |    3 +++
 1 file changed, 3 insertions(+)
drivers/usb/gadget/function/u_ether.c +3 -0
@@ -17,6 +17,7 @@
 #include <linux/etherdevice.h>
 #include <linux/ethtool.h>
 #include <linux/if_vlan.h>
+#include <linux/string_helpers.h>
 
 #include "u_ether.h"
 
@@ -978,6 +979,8 @@ int gether_get_host_addr_cdc(struct net_
 	dev = netdev_priv(net);
 	snprintf(host_addr, len, "%pm", dev->host_mac);
 
+	string_upper(host_addr, host_addr);
+
 	return strlen(host_addr);
 }
 EXPORT_SYMBOL_GPL(gether_get_host_addr_cdc);
author maintainer
#233 lore raw

PATCH 233/292 usb: typec: altmodes/displayport: fix pin_assignment_show

In reply to Greg Kroah-Hartman

From: Badhri Jagan Sridharan <[email protected]>

commit d8f28269dd4bf9b55c3fb376ae31512730a96fce upstream.

This patch fixes negative indexing of buf array in pin_assignment_show when get_current_pin_assignments returns 0 i.e. no compatible pin assignments are found.

BUG: KASAN: use-after-free in pin_assignment_show+0x26c/0x33c
...
Call trace:
dump_backtrace+0x110/0x204
dump_stack_lvl+0x84/0xbc
print_report+0x358/0x974
kasan_report+0x9c/0xfc
__do_kernel_fault+0xd4/0x2d4
do_bad_area+0x48/0x168
do_tag_check_fault+0x24/0x38
do_mem_abort+0x6c/0x14c
el1_abort+0x44/0x68
el1h_64_sync_handler+0x64/0xa4
el1h_64_sync+0x78/0x7c
pin_assignment_show+0x26c/0x33c
dev_attr_show+0x50/0xc0
---
 drivers/usb/typec/altmodes/displayport.c |    4 ++++
 1 file changed, 4 insertions(+)
drivers/usb/typec/altmodes/displayport.c +4 -0
@@ -513,6 +513,10 @@ static ssize_t pin_assignment_show(struc
 
 	mutex_unlock(&dp->lock);
 
+	/* get_current_pin_assignments can return 0 when no matching pin assignments are found */
+	if (len == 0)
+		len++;
+
 	buf[len - 1] = '\n';
 	return len;
 }
author maintainer
#234 lore raw

PATCH 234/292 Revert "usb: gadget: udc: core: Prevent redundant calls to pullup"

In reply to Greg Kroah-Hartman

From: Francesco Dolcini <[email protected]>

commit 5e1617210aede9f1b91bb9819c93097b6da481f9 upstream.

This reverts commit a3afbf5cc887fc3401f012fe629810998ed61859.

This depends on commit 0db213ea8eed ("usb: gadget: udc: core: Invoke usb_gadget_connect only when started") that introduces a regression, revert it till the issue is fixed.

---
 drivers/usb/gadget/udc/core.c |    3 ---
 1 file changed, 3 deletions(-)
drivers/usb/gadget/udc/core.c +0 -3
@@ -676,9 +676,6 @@ static int usb_gadget_connect_locked(str
 		goto out;
 	}
 
-	if (gadget->connected)
-		goto out;
-
 	if (gadget->deactivated || !gadget->udc->started) {
 		/*
 		 * If gadget is deactivated we only save new state.
author maintainer
#235 lore raw

PATCH 235/292 Revert "usb: gadget: udc: core: Invoke usb_gadget_connect only when started"

In reply to Greg Kroah-Hartman

From: Francesco Dolcini <[email protected]>

commit f22e9b67f19ccc73de1ae04375d4b30684e261f8 upstream.

This reverts commit 0db213ea8eed5534a5169e807f28103cbc9d23df.

It introduces an issues with configuring the USB gadget hangs forever on multiple Qualcomm and NXP i.MX SoC at least.

---
 drivers/usb/gadget/udc/core.c |  148 ++++++++++++------------------------------
 1 file changed, 44 insertions(+), 104 deletions(-)
drivers/usb/gadget/udc/core.c +44 -104
@@ -37,10 +37,6 @@ static struct bus_type gadget_bus_type;
  * @vbus: for udcs who care about vbus status, this value is real vbus status;
  * for udcs who do not care about vbus status, this value is always true
  * @started: the UDC's started state. True if the UDC had started.
- * @connect_lock: protects udc->vbus, udc->started, gadget->connect, gadget->deactivate related
- * functions. usb_gadget_connect_locked, usb_gadget_disconnect_locked,
- * usb_udc_connect_control_locked, usb_gadget_udc_start_locked, usb_gadget_udc_stop_locked are
- * called with this lock held.
  *
  * This represents the internal data structure which is used by the UDC-class
  * to hold information about udc driver and gadget together.
@@ -52,7 +48,6 @@ struct usb_udc {
 	struct list_head		list;
 	bool				vbus;
 	bool				started;
-	struct mutex			connect_lock;
 };
 
 static struct class *udc_class;
@@ -665,9 +660,17 @@ out:
 }
 EXPORT_SYMBOL_GPL(usb_gadget_vbus_disconnect);
 
-/* Internal version of usb_gadget_connect needs to be called with connect_lock held. */
-static int usb_gadget_connect_locked(struct usb_gadget *gadget)
-	__must_hold(&gadget->udc->connect_lock)
+/**
+ * usb_gadget_connect - software-controlled connect to USB host
+ * @gadget:the peripheral being connected
+ *
+ * Enables the D+ (or potentially D-) pullup.  The host will start
+ * enumerating this gadget when the pullup is active and a VBUS session
+ * is active (the link is powered).
+ *
+ * Returns zero on success, else negative errno.
+ */
+int usb_gadget_connect(struct usb_gadget *gadget)
 {
 	int ret = 0;
 
@@ -676,12 +679,10 @@ static int usb_gadget_connect_locked(str
 		goto out;
 	}
 
-	if (gadget->deactivated || !gadget->udc->started) {
+	if (gadget->deactivated) {
 		/*
 		 * If gadget is deactivated we only save new state.
 		 * Gadget will be connected automatically after activation.
-		 *
-		 * udc first needs to be started before gadget can be pulled up.
 		 */
 		gadget->connected = true;
 		goto out;
@@ -696,32 +697,22 @@ out:
 
 	return ret;
 }
+EXPORT_SYMBOL_GPL(usb_gadget_connect);
 
 /**
- * usb_gadget_connect - software-controlled connect to USB host
- * @gadget:the peripheral being connected
+ * usb_gadget_disconnect - software-controlled disconnect from USB host
+ * @gadget:the peripheral being disconnected
  *
- * Enables the D+ (or potentially D-) pullup.  The host will start
- * enumerating this gadget when the pullup is active and a VBUS session
- * is active (the link is powered).
+ * Disables the D+ (or potentially D-) pullup, which the host may see
+ * as a disconnect (when a VBUS session is active).  Not all systems
+ * support software pullup controls.
+ *
+ * Following a successful disconnect, invoke the ->disconnect() callback
+ * for the current gadget driver so that UDC drivers don't need to.
  *
  * Returns zero on success, else negative errno.
  */
-int usb_gadget_connect(struct usb_gadget *gadget)
-{
-	int ret;
-
-	mutex_lock(&gadget->udc->connect_lock);
-	ret = usb_gadget_connect_locked(gadget);
-	mutex_unlock(&gadget->udc->connect_lock);
-
-	return ret;
-}
-EXPORT_SYMBOL_GPL(usb_gadget_connect);
-
-/* Internal version of usb_gadget_disconnect needs to be called with connect_lock held. */
-static int usb_gadget_disconnect_locked(struct usb_gadget *gadget)
-	__must_hold(&gadget->udc->connect_lock)
+int usb_gadget_disconnect(struct usb_gadget *gadget)
 {
 	int ret = 0;
 
@@ -733,12 +724,10 @@ static int usb_gadget_disconnect_locked(
 	if (!gadget->connected)
 		goto out;
 
-	if (gadget->deactivated || !gadget->udc->started) {
+	if (gadget->deactivated) {
 		/*
 		 * If gadget is deactivated we only save new state.
 		 * Gadget will stay disconnected after activation.
-		 *
-		 * udc should have been started before gadget being pulled down.
 		 */
 		gadget->connected = false;
 		goto out;
@@ -758,30 +747,6 @@ out:
 
 	return ret;
 }
-
-/**
- * usb_gadget_disconnect - software-controlled disconnect from USB host
- * @gadget:the peripheral being disconnected
- *
- * Disables the D+ (or potentially D-) pullup, which the host may see
- * as a disconnect (when a VBUS session is active).  Not all systems
- * support software pullup controls.
- *
- * Following a successful disconnect, invoke the ->disconnect() callback
- * for the current gadget driver so that UDC drivers don't need to.
- *
- * Returns zero on success, else negative errno.
- */
-int usb_gadget_disconnect(struct usb_gadget *gadget)
-{
-	int ret;
-
-	mutex_lock(&gadget->udc->connect_lock);
-	ret = usb_gadget_disconnect_locked(gadget);
-	mutex_unlock(&gadget->udc->connect_lock);
-
-	return ret;
-}
 EXPORT_SYMBOL_GPL(usb_gadget_disconnect);
 
 /**
@@ -802,11 +767,10 @@ int usb_gadget_deactivate(struct usb_gad
 	if (gadget->deactivated)
 		goto out;
 
-	mutex_lock(&gadget->udc->connect_lock);
 	if (gadget->connected) {
-		ret = usb_gadget_disconnect_locked(gadget);
+		ret = usb_gadget_disconnect(gadget);
 		if (ret)
-			goto unlock;
+			goto out;
 
 		/*
 		 * If gadget was being connected before deactivation, we want
@@ -816,8 +780,6 @@ int usb_gadget_deactivate(struct usb_gad
 	}
 	gadget->deactivated = true;
 
-unlock:
-	mutex_unlock(&gadget->udc->connect_lock);
 out:
 	trace_usb_gadget_deactivate(gadget, ret);
 
@@ -841,7 +803,6 @@ int usb_gadget_activate(struct usb_gadge
 	if (!gadget->deactivated)
 		goto out;
 
-	mutex_lock(&gadget->udc->connect_lock);
 	gadget->deactivated = false;
 
 	/*
@@ -849,8 +810,7 @@ int usb_gadget_activate(struct usb_gadge
 	 * while it was being deactivated, we call usb_gadget_connect().
 	 */
 	if (gadget->connected)
-		ret = usb_gadget_connect_locked(gadget);
-	mutex_unlock(&gadget->udc->connect_lock);
+		ret = usb_gadget_connect(gadget);
 
 out:
 	trace_usb_gadget_activate(gadget, ret);
@@ -1091,13 +1051,12 @@ EXPORT_SYMBOL_GPL(usb_gadget_set_state);
 
 /* ------------------------------------------------------------------------- */
 
-/* Acquire connect_lock before calling this function. */
-static void usb_udc_connect_control_locked(struct usb_udc *udc) __must_hold(&udc->connect_lock)
+static void usb_udc_connect_control(struct usb_udc *udc)
 {
-	if (udc->vbus && udc->started)
-		usb_gadget_connect_locked(udc->gadget);
+	if (udc->vbus)
+		usb_gadget_connect(udc->gadget);
 	else
-		usb_gadget_disconnect_locked(udc->gadget);
+		usb_gadget_disconnect(udc->gadget);
 }
 
 /**
@@ -1113,12 +1072,10 @@ void usb_udc_vbus_handler(struct usb_gad
 {
 	struct usb_udc *udc = gadget->udc;
 
-	mutex_lock(&udc->connect_lock);
 	if (udc) {
 		udc->vbus = status;
-		usb_udc_connect_control_locked(udc);
+		usb_udc_connect_control(udc);
 	}
-	mutex_unlock(&udc->connect_lock);
 }
 EXPORT_SYMBOL_GPL(usb_udc_vbus_handler);
 
@@ -1140,7 +1097,7 @@ void usb_gadget_udc_reset(struct usb_gad
 EXPORT_SYMBOL_GPL(usb_gadget_udc_reset);
 
 /**
- * usb_gadget_udc_start_locked - tells usb device controller to start up
+ * usb_gadget_udc_start - tells usb device controller to start up
  * @udc: The UDC to be started
  *
  * This call is issued by the UDC Class driver when it's about
@@ -1151,11 +1108,8 @@ EXPORT_SYMBOL_GPL(usb_gadget_udc_reset);
  * necessary to have it powered on.
  *
  * Returns zero on success, else negative errno.
- *
- * Caller should acquire connect_lock before invoking this function.
  */
-static inline int usb_gadget_udc_start_locked(struct usb_udc *udc)
-	__must_hold(&udc->connect_lock)
+static inline int usb_gadget_udc_start(struct usb_udc *udc)
 {
 	int ret;
 
@@ -1172,7 +1126,7 @@ static inline int usb_gadget_udc_start_l
 }
 
 /**
- * usb_gadget_udc_stop_locked - tells usb device controller we don't need it anymore
+ * usb_gadget_udc_stop - tells usb device controller we don't need it anymore
  * @udc: The UDC to be stopped
  *
  * This call is issued by the UDC Class driver after calling
@@ -1181,11 +1135,8 @@ static inline int usb_gadget_udc_start_l
  * The details are implementation specific, but it can go as
  * far as powering off UDC completely and disable its data
  * line pullups.
- *
- * Caller should acquire connect lock before invoking this function.
  */
-static inline void usb_gadget_udc_stop_locked(struct usb_udc *udc)
-	__must_hold(&udc->connect_lock)
+static inline void usb_gadget_udc_stop(struct usb_udc *udc)
 {
 	if (!udc->started) {
 		dev_err(&udc->dev, "UDC had already stopped\n");
@@ -1344,7 +1295,6 @@ int usb_add_gadget(struct usb_gadget *ga
 
 	udc->gadget = gadget;
 	gadget->udc = udc;
-	mutex_init(&udc->connect_lock);
 
 	udc->started = false;
 
@@ -1546,15 +1496,11 @@ static int gadget_bind_driver(struct dev
 	if (ret)
 		goto err_bind;
 
-	mutex_lock(&udc->connect_lock);
-	ret = usb_gadget_udc_start_locked(udc);
-	if (ret) {
-		mutex_unlock(&udc->connect_lock);
+	ret = usb_gadget_udc_start(udc);
+	if (ret)
 		goto err_start;
-	}
 	usb_gadget_enable_async_callbacks(udc);
-	usb_udc_connect_control_locked(udc);
-	mutex_unlock(&udc->connect_lock);
+	usb_udc_connect_control(udc);
 
 	kobject_uevent(&udc->dev.kobj, KOBJ_CHANGE);
 	return 0;
@@ -1585,14 +1531,12 @@ static void gadget_unbind_driver(struct
 
 	kobject_uevent(&udc->dev.kobj, KOBJ_CHANGE);
 
-	mutex_lock(&udc->connect_lock);
-	usb_gadget_disconnect_locked(gadget);
+	usb_gadget_disconnect(gadget);
 	usb_gadget_disable_async_callbacks(udc);
 	if (gadget->irq)
 		synchronize_irq(gadget->irq);
 	udc->driver->unbind(gadget);
-	usb_gadget_udc_stop_locked(udc);
-	mutex_unlock(&udc->connect_lock);
+	usb_gadget_udc_stop(udc);
 
 	mutex_lock(&udc_lock);
 	driver->is_bound = false;
@@ -1678,15 +1622,11 @@ static ssize_t soft_connect_store(struct
 	}
 
 	if (sysfs_streq(buf, "connect")) {
-		mutex_lock(&udc->connect_lock);
-		usb_gadget_udc_start_locked(udc);
-		usb_gadget_connect_locked(udc->gadget);
-		mutex_unlock(&udc->connect_lock);
+		usb_gadget_udc_start(udc);
+		usb_gadget_connect(udc->gadget);
 	} else if (sysfs_streq(buf, "disconnect")) {
-		mutex_lock(&udc->connect_lock);
-		usb_gadget_disconnect_locked(udc->gadget);
-		usb_gadget_udc_stop_locked(udc);
-		mutex_unlock(&udc->connect_lock);
+		usb_gadget_disconnect(udc->gadget);
+		usb_gadget_udc_stop(udc);
 	} else {
 		dev_err(dev, "unsupported command '%s'\n", buf);
 		ret = -EINVAL;
author maintainer
#236 lore raw

PATCH 236/292 xhci-pci: Only run d3cold avoidance quirk for s2idle

In reply to Greg Kroah-Hartman

From: Mario Limonciello <[email protected]>

commit 2a821fc3136d5d99dcb9de152be8a052ca27d870 upstream.

Donghun reports that a notebook that has an AMD Ryzen 5700U but supports S3 has problems with USB after resuming from suspend. The issue was bisected down to commit d1658268e439 ("usb: pci-quirks: disable D3cold on xhci suspend for s2idle on AMD Renoir").

As this issue only happens on S3, narrow the broken D3cold quirk to only run in s2idle.

Fixes d1658268e439 ("usb: pci-quirks: disable D3cold on xhci suspend for s2idle on AMD Renoir")
Reported-and-tested-by Donghun Yoon <[email protected]>
---
 drivers/usb/host/xhci-pci.c |   12 ++++++++++--
 drivers/usb/host/xhci.h     |    2 +-
 2 files changed, 11 insertions(+), 3 deletions(-)
drivers/usb/host/xhci-pci.c +10 -2
@@ -13,6 +13,7 @@
 #include <linux/module.h>
 #include <linux/acpi.h>
 #include <linux/reset.h>
+#include <linux/suspend.h>
 
 #include "xhci.h"
 #include "xhci-trace.h"
@@ -194,7 +195,7 @@ static void xhci_pci_quirks(struct devic
 
 	if (pdev->vendor == PCI_VENDOR_ID_AMD &&
 		pdev->device == PCI_DEVICE_ID_AMD_RENOIR_XHCI)
-		xhci->quirks |= XHCI_BROKEN_D3COLD;
+		xhci->quirks |= XHCI_BROKEN_D3COLD_S2I;
 
 	if (pdev->vendor == PCI_VENDOR_ID_INTEL) {
 		xhci->quirks |= XHCI_LPM_SUPPORT;
@@ -609,9 +610,16 @@ static int xhci_pci_suspend(struct usb_h
 	 * Systems with the TI redriver that loses port status change events
 	 * need to have the registers polled during D3, so avoid D3cold.
 	 */
-	if (xhci->quirks & (XHCI_COMP_MODE_QUIRK | XHCI_BROKEN_D3COLD))
+	if (xhci->quirks & XHCI_COMP_MODE_QUIRK)
 		pci_d3cold_disable(pdev);
 
+#ifdef CONFIG_SUSPEND
+	/* d3cold is broken, but only when s2idle is used */
+	if (pm_suspend_target_state == PM_SUSPEND_TO_IDLE &&
+	    xhci->quirks & (XHCI_BROKEN_D3COLD_S2I))
+		pci_d3cold_disable(pdev);
+#endif
+
 	if (xhci->quirks & XHCI_PME_STUCK_QUIRK)
 		xhci_pme_quirk(hcd);
 
drivers/usb/host/xhci.h +1 -1
@@ -1895,7 +1895,7 @@ struct xhci_hcd {
 #define XHCI_DISABLE_SPARSE	BIT_ULL(38)
 #define XHCI_SG_TRB_CACHE_SIZE_QUIRK	BIT_ULL(39)
 #define XHCI_NO_SOFT_RETRY	BIT_ULL(40)
-#define XHCI_BROKEN_D3COLD	BIT_ULL(41)
+#define XHCI_BROKEN_D3COLD_S2I	BIT_ULL(41)
 #define XHCI_EP_CTX_BROKEN_DCS	BIT_ULL(42)
 #define XHCI_SUSPEND_RESUME_CLKS	BIT_ULL(43)
 #define XHCI_RESET_TO_DEFAULT	BIT_ULL(44)
author maintainer
#237 lore raw

PATCH 237/292 xhci: Fix incorrect tracking of free space on transfer rings

In reply to Greg Kroah-Hartman

From: Mathias Nyman <[email protected]>

commit fe82f16aafdaf8002281d3b9524291d4a4a28460 upstream.

This incorrect tracking caused unnecessary ring expansion in some usecases which over days of use consume a lot of memory.

xhci driver tries to keep track of free transfer blocks (TRBs) on the ring buffer, but failed to add back some cancelled transfers that were turned into no-op operations instead of just moving past them.

This can happen if there are several queued pending transfers which then are cancelled in reverse order.

Solve this by counting the numer of steps we move the dequeue pointer once we complete a transfer, and add it to the number of free trbs instead of just adding the trb number of the current transfer. This way we ensure we count the no-op trbs on the way as well.

---
 drivers/usb/host/xhci-ring.c |   29 ++++++++++++++++++++++++++++-
 1 file changed, 28 insertions(+), 1 deletion(-)
drivers/usb/host/xhci-ring.c +28 -1
@@ -276,6 +276,26 @@ static void inc_enq(struct xhci_hcd *xhc
 	trace_xhci_inc_enq(ring);
 }
 
+static int xhci_num_trbs_to(struct xhci_segment *start_seg, union xhci_trb *start,
+			    struct xhci_segment *end_seg, union xhci_trb *end,
+			    unsigned int num_segs)
+{
+	union xhci_trb *last_on_seg;
+	int num = 0;
+	int i = 0;
+
+	do {
+		if (start_seg == end_seg && end >= start)
+			return num + (end - start);
+		last_on_seg = &start_seg->trbs[TRBS_PER_SEGMENT - 1];
+		num += last_on_seg - start;
+		start_seg = start_seg->next;
+		start = start_seg->trbs;
+	} while (i++ <= num_segs);
+
+	return -EINVAL;
+}
+
 /*
  * Check to see if there's room to enqueue num_trbs on the ring and make sure
  * enqueue pointer will not advance into dequeue segment. See rules above.
@@ -2141,6 +2161,7 @@ static int finish_td(struct xhci_hcd *xh
 		     u32 trb_comp_code)
 {
 	struct xhci_ep_ctx *ep_ctx;
+	int trbs_freed;
 
 	ep_ctx = xhci_get_ep_ctx(xhci, ep->vdev->out_ctx, ep->ep_index);
 
@@ -2212,9 +2233,15 @@ static int finish_td(struct xhci_hcd *xh
 	}
 
 	/* Update ring dequeue pointer */
+	trbs_freed = xhci_num_trbs_to(ep_ring->deq_seg, ep_ring->dequeue,
+				      td->last_trb_seg, td->last_trb,
+				      ep_ring->num_segs);
+	if (trbs_freed < 0)
+		xhci_dbg(xhci, "Failed to count freed trbs at TD finish\n");
+	else
+		ep_ring->num_trbs_free += trbs_freed;
 	ep_ring->dequeue = td->last_trb;
 	ep_ring->deq_seg = td->last_trb_seg;
-	ep_ring->num_trbs_free += td->num_trbs - 1;
 	inc_deq(xhci, ep_ring);
 
 	return xhci_td_cleanup(xhci, td, ep_ring, td->status);
author maintainer
#238 lore raw

PATCH 238/292 ALSA: hda: Fix Oops by 9.1 surround channel names

In reply to Greg Kroah-Hartman

From: Takashi Iwai <[email protected]>

commit 3b44ec8c5c44790a82f07e90db45643c762878c6 upstream.

get_line_out_pfx() may trigger an Oops by overflowing the static array with more than 8 channels. This was reported for MacBookPro 12,1 with Cirrus codec.

As a workaround, extend for the 9.1 channels and also fix the potential Oops by unifying the code paths accessing the same array with the proper size check.

---
 sound/pci/hda/hda_generic.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
sound/pci/hda/hda_generic.c +4 -3
@@ -1155,8 +1155,8 @@ static bool path_has_mixer(struct hda_co
 	return path && path->ctls[ctl_type];
 }
 
-static const char * const channel_name[4] = {
-	"Front", "Surround", "CLFE", "Side"
+static const char * const channel_name[] = {
+	"Front", "Surround", "CLFE", "Side", "Back",
 };
 
 /* give some appropriate ctl name prefix for the given line out channel */
@@ -1182,7 +1182,7 @@ static const char *get_line_out_pfx(stru
 
 	/* multi-io channels */
 	if (ch >= cfg->line_outs)
-		return channel_name[ch];
+		goto fixed_name;
 
 	switch (cfg->line_out_type) {
 	case AUTO_PIN_SPEAKER_OUT:
@@ -1234,6 +1234,7 @@ static const char *get_line_out_pfx(stru
 	if (cfg->line_outs == 1 && !spec->multi_ios)
 		return "Line Out";
 
+ fixed_name:
 	if (ch >= ARRAY_SIZE(channel_name)) {
 		snd_BUG();
 		return "PCM";
author maintainer
#239 lore raw

PATCH 239/292 ALSA: hda: Add NVIDIA codec IDs a3 through a7 to patch table

In reply to Greg Kroah-Hartman

From: Nikhil Mahale <[email protected]>

commit dc4f2ccaedddb489a83e7b12ebbdc347272aacc9 upstream.

These IDs are for AD102, AD103, AD104, AD106, and AD107 gpus with audio functions that are largely similar to the existing ones.

Tested audio using gnome-settings, over HDMI, DP-SST and DP-MST connections on AD106 gpu.

---
 sound/pci/hda/patch_hdmi.c |    5 +++++
 1 file changed, 5 insertions(+)
sound/pci/hda/patch_hdmi.c +5 -0
@@ -4577,6 +4577,11 @@ HDA_CODEC_ENTRY(0x10de009d, "GPU 9d HDMI
 HDA_CODEC_ENTRY(0x10de009e, "GPU 9e HDMI/DP",	patch_nvhdmi),
 HDA_CODEC_ENTRY(0x10de009f, "GPU 9f HDMI/DP",	patch_nvhdmi),
 HDA_CODEC_ENTRY(0x10de00a0, "GPU a0 HDMI/DP",	patch_nvhdmi),
+HDA_CODEC_ENTRY(0x10de00a3, "GPU a3 HDMI/DP",	patch_nvhdmi),
+HDA_CODEC_ENTRY(0x10de00a4, "GPU a4 HDMI/DP",	patch_nvhdmi),
+HDA_CODEC_ENTRY(0x10de00a5, "GPU a5 HDMI/DP",	patch_nvhdmi),
+HDA_CODEC_ENTRY(0x10de00a6, "GPU a6 HDMI/DP",	patch_nvhdmi),
+HDA_CODEC_ENTRY(0x10de00a7, "GPU a7 HDMI/DP",	patch_nvhdmi),
 HDA_CODEC_ENTRY(0x10de8001, "MCP73 HDMI",	patch_nvhdmi_2ch),
 HDA_CODEC_ENTRY(0x10de8067, "MCP67/68 HDMI",	patch_nvhdmi_2ch),
 HDA_CODEC_ENTRY(0x11069f80, "VX900 HDMI/DP",	patch_via_hdmi),
author maintainer
#240 lore raw

PATCH 240/292 ALSA: hda/realtek: Add quirk for Clevo L140AU

In reply to Greg Kroah-Hartman

From: Jeremy Soller <[email protected]>

commit 0a6b36c5dc3dda0196f4fb65bdb34c38b8d060c3 upstream.

Fixes headset detection on Clevo L140AU.

---
 sound/pci/hda/patch_realtek.c |    1 +
 1 file changed, 1 insertion(+)
sound/pci/hda/patch_realtek.c +1 -0
@@ -9618,6 +9618,7 @@ static const struct snd_pci_quirk alc269
 	SND_PCI_QUIRK(0x1558, 0x7716, "Clevo NS50PU", ALC256_FIXUP_SYSTEM76_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x1558, 0x7717, "Clevo NS70PU", ALC256_FIXUP_SYSTEM76_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x1558, 0x7718, "Clevo L140PU", ALC256_FIXUP_SYSTEM76_MIC_NO_PRESENCE),
+	SND_PCI_QUIRK(0x1558, 0x7724, "Clevo L140AU", ALC256_FIXUP_SYSTEM76_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x1558, 0x8228, "Clevo NR40BU", ALC293_FIXUP_SYSTEM76_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x1558, 0x8520, "Clevo NH50D[CD]", ALC293_FIXUP_SYSTEM76_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x1558, 0x8521, "Clevo NH77D[CD]", ALC293_FIXUP_SYSTEM76_MIC_NO_PRESENCE),
author maintainer
#241 lore raw

PATCH 241/292 ALSA: hda/realtek: Add a quirk for HP EliteDesk 805

In reply to Greg Kroah-Hartman

From: Ai Chao <[email protected]>

commit 90670ef774a8b6700c38ce1222e6aa263be54d5f upstream.

Add a quirk for HP EliteDesk 805 to fixup ALC3867 headset MIC no sound.

---
 sound/pci/hda/patch_realtek.c |    1 +
 1 file changed, 1 insertion(+)
sound/pci/hda/patch_realtek.c +1 -0
@@ -11664,6 +11664,7 @@ static const struct snd_pci_quirk alc662
 	SND_PCI_QUIRK(0x103c, 0x1632, "HP RP5800", ALC662_FIXUP_HP_RP5800),
 	SND_PCI_QUIRK(0x103c, 0x870c, "HP", ALC897_FIXUP_HP_HSMIC_VERB),
 	SND_PCI_QUIRK(0x103c, 0x8719, "HP", ALC897_FIXUP_HP_HSMIC_VERB),
+	SND_PCI_QUIRK(0x103c, 0x872b, "HP", ALC897_FIXUP_HP_HSMIC_VERB),
 	SND_PCI_QUIRK(0x103c, 0x873e, "HP", ALC671_FIXUP_HP_HEADSET_MIC2),
 	SND_PCI_QUIRK(0x103c, 0x877e, "HP 288 Pro G6", ALC671_FIXUP_HP_HEADSET_MIC2),
 	SND_PCI_QUIRK(0x103c, 0x885f, "HP 288 Pro G8", ALC671_FIXUP_HP_HEADSET_MIC2),
author maintainer
#242 lore raw

PATCH 242/292 ALSA: hda/realtek: Add quirk for 2nd ASUS GU603

In reply to Greg Kroah-Hartman

From: Luke D. Jones <[email protected]>

commit a4671b7fba59775845ee60cfbdfc4ba64300211b upstream.

Add quirk for GU603 with 0x1c62 variant of codec.

---
 sound/pci/hda/patch_realtek.c |    1 +
 1 file changed, 1 insertion(+)
sound/pci/hda/patch_realtek.c +1 -0
@@ -9522,6 +9522,7 @@ static const struct snd_pci_quirk alc269
 	SND_PCI_QUIRK(0x1043, 0x1b13, "Asus U41SV", ALC269_FIXUP_INV_DMIC),
 	SND_PCI_QUIRK(0x1043, 0x1bbd, "ASUS Z550MA", ALC255_FIXUP_ASUS_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x1043, 0x1c23, "Asus X55U", ALC269_FIXUP_LIMIT_INT_MIC_BOOST),
+	SND_PCI_QUIRK(0x1043, 0x1c62, "ASUS GU603", ALC289_FIXUP_ASUS_GA401),
 	SND_PCI_QUIRK(0x1043, 0x1c92, "ASUS ROG Strix G15", ALC285_FIXUP_ASUS_G533Z_PINS),
 	SND_PCI_QUIRK(0x1043, 0x1ccd, "ASUS X555UB", ALC256_FIXUP_ASUS_MIC),
 	SND_PCI_QUIRK(0x1043, 0x1d42, "ASUS Zephyrus G14 2022", ALC289_FIXUP_ASUS_GA401),
author maintainer
#243 lore raw

PATCH 243/292 ALSA: hda/realtek: Add quirk for HP EliteBook G10 laptops

In reply to Greg Kroah-Hartman

From: Vitaly Rodionov <[email protected]>

commit 3e10f6ca76c4d00019badebd235c9d7f0068261e upstream.

Add support for HP EliteBook 835/845/845W/865 G10 laptops with CS35L41 amplifiers on I2C/SPI bus connected to Realtek codec.

---
 sound/pci/hda/patch_realtek.c |    8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
sound/pci/hda/patch_realtek.c +7 -1
@@ -9458,7 +9458,7 @@ static const struct snd_pci_quirk alc269
 	SND_PCI_QUIRK(0x103c, 0x8aa3, "HP ProBook 450 G9 (MB 8AA1)", ALC236_FIXUP_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8aa8, "HP EliteBook 640 G9 (MB 8AA6)", ALC236_FIXUP_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8aab, "HP EliteBook 650 G9 (MB 8AA9)", ALC236_FIXUP_HP_GPIO_LED),
-	 SND_PCI_QUIRK(0x103c, 0x8abb, "HP ZBook Firefly 14 G9", ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED),
+	SND_PCI_QUIRK(0x103c, 0x8abb, "HP ZBook Firefly 14 G9", ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8ad1, "HP EliteBook 840 14 inch G9 Notebook PC", ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8ad2, "HP EliteBook 860 16 inch G9 Notebook PC", ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8b42, "HP", ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED),
@@ -9469,8 +9469,13 @@ static const struct snd_pci_quirk alc269
 	SND_PCI_QUIRK(0x103c, 0x8b47, "HP", ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8b5d, "HP", ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF),
 	SND_PCI_QUIRK(0x103c, 0x8b5e, "HP", ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF),
+	SND_PCI_QUIRK(0x103c, 0x8b63, "HP Elite Dragonfly 13.5 inch G4", ALC245_FIXUP_CS35L41_SPI_4_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8b65, "HP ProBook 455 15.6 inch G10 Notebook PC", ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF),
 	SND_PCI_QUIRK(0x103c, 0x8b66, "HP", ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF),
+	SND_PCI_QUIRK(0x103c, 0x8b70, "HP EliteBook 835 G10", ALC287_FIXUP_CS35L41_I2C_2),
+	SND_PCI_QUIRK(0x103c, 0x8b72, "HP EliteBook 845 G10", ALC287_FIXUP_CS35L41_I2C_2),
+	SND_PCI_QUIRK(0x103c, 0x8b74, "HP EliteBook 845W G10", ALC287_FIXUP_CS35L41_I2C_2),
+	SND_PCI_QUIRK(0x103c, 0x8b77, "HP ElieBook 865 G10", ALC287_FIXUP_CS35L41_I2C_2),
 	SND_PCI_QUIRK(0x103c, 0x8b7a, "HP", ALC236_FIXUP_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8b7d, "HP", ALC236_FIXUP_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8b87, "HP", ALC236_FIXUP_HP_GPIO_LED),
@@ -9481,6 +9486,7 @@ static const struct snd_pci_quirk alc269
 	SND_PCI_QUIRK(0x103c, 0x8b92, "HP", ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8b96, "HP", ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF),
 	SND_PCI_QUIRK(0x103c, 0x8bf0, "HP", ALC236_FIXUP_HP_GPIO_LED),
+	SND_PCI_QUIRK(0x103c, 0x8c26, "HP HP EliteBook 800G11", ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x1043, 0x103e, "ASUS X540SA", ALC256_FIXUP_ASUS_MIC),
 	SND_PCI_QUIRK(0x1043, 0x103f, "ASUS TX300", ALC282_FIXUP_ASUS_TX300),
 	SND_PCI_QUIRK(0x1043, 0x106d, "Asus K53BE", ALC269_FIXUP_LIMIT_INT_MIC_BOOST),
author maintainer
#244 lore raw

PATCH 244/292 ALSA: hda/realtek: Fix mute and micmute LEDs for yet another HP laptop

In reply to Greg Kroah-Hartman

From: Kai-Heng Feng <[email protected]>

commit 9dc68a4fe70893b000fb3c92c68b9f72369cf448 upstream.

There's yet another laptop that needs the fixup to enable mute and micmute LEDs. So do it accordingly.

---
 sound/pci/hda/patch_realtek.c |    1 +
 1 file changed, 1 insertion(+)
sound/pci/hda/patch_realtek.c +1 -0
@@ -9485,6 +9485,7 @@ static const struct snd_pci_quirk alc269
 	SND_PCI_QUIRK(0x103c, 0x8b8f, "HP", ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8b92, "HP", ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8b96, "HP", ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF),
+	SND_PCI_QUIRK(0x103c, 0x8b97, "HP", ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF),
 	SND_PCI_QUIRK(0x103c, 0x8bf0, "HP", ALC236_FIXUP_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x103c, 0x8c26, "HP HP EliteBook 800G11", ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED),
 	SND_PCI_QUIRK(0x1043, 0x103e, "ASUS X540SA", ALC256_FIXUP_ASUS_MIC),
author maintainer
#245 lore raw

PATCH 245/292 can: j1939: recvmsg(): allow MSG_CMSG_COMPAT flag

In reply to Greg Kroah-Hartman

From: Oliver Hartkopp <[email protected]>

commit 1db080cbdbab28752bbb1c86d64daf96253a5da1 upstream.

The control message provided by J1939 support MSG_CMSG_COMPAT but blocked recvmsg() syscalls that have set this flag, i.e. on 32bit user space on 64 bit kernels.

---
 net/can/j1939/socket.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/can/j1939/socket.c +1 -1
@@ -798,7 +798,7 @@ static int j1939_sk_recvmsg(struct socke
 	struct j1939_sk_buff_cb *skcb;
 	int ret = 0;
 
-	if (flags & ~(MSG_DONTWAIT | MSG_ERRQUEUE))
+	if (flags & ~(MSG_DONTWAIT | MSG_ERRQUEUE | MSG_CMSG_COMPAT))
 		return -EINVAL;
 
 	if (flags & MSG_ERRQUEUE)
author maintainer
#246 lore raw

PATCH 246/292 can: isotp: recvmsg(): allow MSG_CMSG_COMPAT flag

In reply to Greg Kroah-Hartman

From: Oliver Hartkopp <[email protected]>

commit db2773d65b02aed319a93efdfb958087771d4e19 upstream.

The control message provided by isotp support MSG_CMSG_COMPAT but blocked recvmsg() syscalls that have set this flag, i.e. on 32bit user space on 64 bit kernels.

---
 net/can/isotp.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/can/isotp.c +1 -1
@@ -1106,7 +1106,7 @@ static int isotp_recvmsg(struct socket *
 	struct isotp_sock *so = isotp_sk(sk);
 	int ret = 0;
 
-	if (flags & ~(MSG_DONTWAIT | MSG_TRUNC | MSG_PEEK))
+	if (flags & ~(MSG_DONTWAIT | MSG_TRUNC | MSG_PEEK | MSG_CMSG_COMPAT))
 		return -EINVAL;
 
 	if (!so->bound)
author maintainer
#247 lore raw

PATCH 247/292 can: kvaser_pciefd: Set CAN_STATE_STOPPED in kvaser_pciefd_stop()

In reply to Greg Kroah-Hartman

From: Jimmy Assarsson <[email protected]>

commit aed0e6ca7dbb8fbea9bc69c9ac663d5533c8c5d8 upstream.

Set can.state to CAN_STATE_STOPPED in kvaser_pciefd_stop(). Without this fix, wrong CAN state was repported after the interface was brought down.

Fixes 26ad340e582d ("can: kvaser_pciefd: Add driver for Kvaser PCIEcan devices")
---
 drivers/net/can/kvaser_pciefd.c |    1 +
 1 file changed, 1 insertion(+)
drivers/net/can/kvaser_pciefd.c +1 -0
@@ -719,6 +719,7 @@ static int kvaser_pciefd_stop(struct net
 		iowrite32(0, can->reg_base + KVASER_PCIEFD_KCAN_IEN_REG);
 		del_timer(&can->bec_poll_timer);
 	}
+	can->can.state = CAN_STATE_STOPPED;
 	close_candev(netdev);
 
 	return ret;
author maintainer
#248 lore raw

PATCH 248/292 can: kvaser_pciefd: Call request_irq() before enabling interrupts

In reply to Greg Kroah-Hartman

From: Jimmy Assarsson <[email protected]>

commit 84762d8da89d29ba842317eb842973e628c27391 upstream.

Make sure the interrupt handler is registered before enabling interrupts.

Fixes 26ad340e582d ("can: kvaser_pciefd: Add driver for Kvaser PCIEcan devices")
---
 drivers/net/can/kvaser_pciefd.c |   10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
drivers/net/can/kvaser_pciefd.c +5 -5
@@ -1825,6 +1825,11 @@ static int kvaser_pciefd_probe(struct pc
 	if (err)
 		goto err_teardown_can_ctrls;
 
+	err = request_irq(pcie->pci->irq, kvaser_pciefd_irq_handler,
+			  IRQF_SHARED, KVASER_PCIEFD_DRV_NAME, pcie);
+	if (err)
+		goto err_teardown_can_ctrls;
+
 	iowrite32(KVASER_PCIEFD_SRB_IRQ_DPD0 | KVASER_PCIEFD_SRB_IRQ_DPD1,
 		  pcie->reg_base + KVASER_PCIEFD_SRB_IRQ_REG);
 
@@ -1845,11 +1850,6 @@ static int kvaser_pciefd_probe(struct pc
 	iowrite32(KVASER_PCIEFD_SRB_CMD_RDB1,
 		  pcie->reg_base + KVASER_PCIEFD_SRB_CMD_REG);
 
-	err = request_irq(pcie->pci->irq, kvaser_pciefd_irq_handler,
-			  IRQF_SHARED, KVASER_PCIEFD_DRV_NAME, pcie);
-	if (err)
-		goto err_teardown_can_ctrls;
-
 	err = kvaser_pciefd_reg_candev(pcie);
 	if (err)
 		goto err_free_irq;
author maintainer
#249 lore raw

PATCH 249/292 can: kvaser_pciefd: Empty SRB buffer in probe

In reply to Greg Kroah-Hartman

From: Jimmy Assarsson <[email protected]>

commit c589557dd1426f5adf90c7a919d4fde5a3e4ef64 upstream.

Empty the "Shared receive buffer" (SRB) in probe, to assure we start in a known state, and don't process any irrelevant packets.

Fixes 26ad340e582d ("can: kvaser_pciefd: Add driver for Kvaser PCIEcan devices")
---
 drivers/net/can/kvaser_pciefd.c |   15 +++++++++++++++
 1 file changed, 15 insertions(+)
drivers/net/can/kvaser_pciefd.c +15 -0
@@ -71,10 +71,12 @@ MODULE_DESCRIPTION("CAN driver for Kvase
 #define KVASER_PCIEFD_SYSID_BUILD_REG (KVASER_PCIEFD_SYSID_BASE + 0x14)
 /* Shared receive buffer registers */
 #define KVASER_PCIEFD_SRB_BASE 0x1f200
+#define KVASER_PCIEFD_SRB_FIFO_LAST_REG (KVASER_PCIEFD_SRB_BASE + 0x1f4)
 #define KVASER_PCIEFD_SRB_CMD_REG (KVASER_PCIEFD_SRB_BASE + 0x200)
 #define KVASER_PCIEFD_SRB_IEN_REG (KVASER_PCIEFD_SRB_BASE + 0x204)
 #define KVASER_PCIEFD_SRB_IRQ_REG (KVASER_PCIEFD_SRB_BASE + 0x20c)
 #define KVASER_PCIEFD_SRB_STAT_REG (KVASER_PCIEFD_SRB_BASE + 0x210)
+#define KVASER_PCIEFD_SRB_RX_NR_PACKETS_REG (KVASER_PCIEFD_SRB_BASE + 0x214)
 #define KVASER_PCIEFD_SRB_CTRL_REG (KVASER_PCIEFD_SRB_BASE + 0x218)
 /* EPCS flash controller registers */
 #define KVASER_PCIEFD_SPI_BASE 0x1fc00
@@ -111,6 +113,9 @@ MODULE_DESCRIPTION("CAN driver for Kvase
 /* DMA support */
 #define KVASER_PCIEFD_SRB_STAT_DMA BIT(24)
 
+/* SRB current packet level */
+#define KVASER_PCIEFD_SRB_RX_NR_PACKETS_MASK 0xff
+
 /* DMA Enable */
 #define KVASER_PCIEFD_SRB_CTRL_DMA_ENABLE BIT(0)
 
@@ -1059,6 +1064,7 @@ static int kvaser_pciefd_setup_dma(struc
 {
 	int i;
 	u32 srb_status;
+	u32 srb_packet_count;
 	dma_addr_t dma_addr[KVASER_PCIEFD_DMA_COUNT];
 
 	/* Disable the DMA */
@@ -1086,6 +1092,15 @@ static int kvaser_pciefd_setup_dma(struc
 		  KVASER_PCIEFD_SRB_CMD_RDB1,
 		  pcie->reg_base + KVASER_PCIEFD_SRB_CMD_REG);
 
+	/* Empty Rx FIFO */
+	srb_packet_count = ioread32(pcie->reg_base + KVASER_PCIEFD_SRB_RX_NR_PACKETS_REG) &
+			   KVASER_PCIEFD_SRB_RX_NR_PACKETS_MASK;
+	while (srb_packet_count) {
+		/* Drop current packet in FIFO */
+		ioread32(pcie->reg_base + KVASER_PCIEFD_SRB_FIFO_LAST_REG);
+		srb_packet_count--;
+	}
+
 	srb_status = ioread32(pcie->reg_base + KVASER_PCIEFD_SRB_STAT_REG);
 	if (!(srb_status & KVASER_PCIEFD_SRB_STAT_DI)) {
 		dev_err(&pcie->pci->dev, "DMA not idle before enabling\n");
author maintainer
#250 lore raw

PATCH 250/292 can: kvaser_pciefd: Clear listen-only bit if not explicitly requested

In reply to Greg Kroah-Hartman

From: Jimmy Assarsson <[email protected]>

commit bf7ac55e991ca177f1ac16be51152f1ef291a4df upstream.

The listen-only bit was never cleared, causing the controller to always use listen-only mode, if previously set.

Fixes 26ad340e582d ("can: kvaser_pciefd: Add driver for Kvaser PCIEcan devices")
---
 drivers/net/can/kvaser_pciefd.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/net/can/kvaser_pciefd.c +2 -0
@@ -559,6 +559,8 @@ static void kvaser_pciefd_setup_controll
 
 	if (can->can.ctrlmode & CAN_CTRLMODE_LISTENONLY)
 		mode |= KVASER_PCIEFD_KCAN_MODE_LOM;
+	else
+		mode &= ~KVASER_PCIEFD_KCAN_MODE_LOM;
 
 	mode |= KVASER_PCIEFD_KCAN_MODE_EEN;
 	mode |= KVASER_PCIEFD_KCAN_MODE_EPEN;
author maintainer
#251 lore raw

PATCH 251/292 can: kvaser_pciefd: Do not send EFLUSH command on TFD interrupt

In reply to Greg Kroah-Hartman

From: Jimmy Assarsson <[email protected]>

commit 262d7a52ba27525e3c1203230c9f0524e48bbb34 upstream.

Under certain circumstances we send two EFLUSH commands, resulting in two EFLUSH ack packets, while only expecting a single EFLUSH ack. This can cause the driver Tx flush completion to get out of sync.

To avoid this problem, don't enable the "Transmit buffer flush done" (TFD) interrupt and remove the code handling it.
Now we only send EFLUSH command after receiving status packet with "Init detected" (IDET) bit set.

Fixes 26ad340e582d ("can: kvaser_pciefd: Add driver for Kvaser PCIEcan devices")
---
 drivers/net/can/kvaser_pciefd.c |   21 ++++-----------------
 1 file changed, 4 insertions(+), 17 deletions(-)
drivers/net/can/kvaser_pciefd.c +4 -17
@@ -531,7 +531,7 @@ static int kvaser_pciefd_set_tx_irq(stru
 	      KVASER_PCIEFD_KCAN_IRQ_TOF | KVASER_PCIEFD_KCAN_IRQ_ABD |
 	      KVASER_PCIEFD_KCAN_IRQ_TAE | KVASER_PCIEFD_KCAN_IRQ_TAL |
 	      KVASER_PCIEFD_KCAN_IRQ_FDIC | KVASER_PCIEFD_KCAN_IRQ_BPP |
-	      KVASER_PCIEFD_KCAN_IRQ_TAR | KVASER_PCIEFD_KCAN_IRQ_TFD;
+	      KVASER_PCIEFD_KCAN_IRQ_TAR;
 
 	iowrite32(msk, can->reg_base + KVASER_PCIEFD_KCAN_IEN_REG);
 
@@ -579,7 +579,7 @@ static void kvaser_pciefd_start_controll
 
 	spin_lock_irqsave(&can->lock, irq);
 	iowrite32(-1, can->reg_base + KVASER_PCIEFD_KCAN_IRQ_REG);
-	iowrite32(KVASER_PCIEFD_KCAN_IRQ_ABD | KVASER_PCIEFD_KCAN_IRQ_TFD,
+	iowrite32(KVASER_PCIEFD_KCAN_IRQ_ABD,
 		  can->reg_base + KVASER_PCIEFD_KCAN_IEN_REG);
 
 	status = ioread32(can->reg_base + KVASER_PCIEFD_KCAN_STAT_REG);
@@ -622,7 +622,7 @@ static int kvaser_pciefd_bus_on(struct k
 	iowrite32(0, can->reg_base + KVASER_PCIEFD_KCAN_IEN_REG);
 	iowrite32(-1, can->reg_base + KVASER_PCIEFD_KCAN_IRQ_REG);
 
-	iowrite32(KVASER_PCIEFD_KCAN_IRQ_ABD | KVASER_PCIEFD_KCAN_IRQ_TFD,
+	iowrite32(KVASER_PCIEFD_KCAN_IRQ_ABD,
 		  can->reg_base + KVASER_PCIEFD_KCAN_IEN_REG);
 
 	mode = ioread32(can->reg_base + KVASER_PCIEFD_KCAN_MODE_REG);
@@ -1015,8 +1015,7 @@ static int kvaser_pciefd_setup_can_ctrls
 		SET_NETDEV_DEV(netdev, &pcie->pci->dev);
 
 		iowrite32(-1, can->reg_base + KVASER_PCIEFD_KCAN_IRQ_REG);
-		iowrite32(KVASER_PCIEFD_KCAN_IRQ_ABD |
-			  KVASER_PCIEFD_KCAN_IRQ_TFD,
+		iowrite32(KVASER_PCIEFD_KCAN_IRQ_ABD,
 			  can->reg_base + KVASER_PCIEFD_KCAN_IEN_REG);
 
 		pcie->can[i] = can;
@@ -1443,9 +1442,6 @@ static int kvaser_pciefd_handle_status_p
 		cmd = KVASER_PCIEFD_KCAN_CMD_AT;
 		cmd |= ++can->cmd_seq << KVASER_PCIEFD_KCAN_CMD_SEQ_SHIFT;
 		iowrite32(cmd, can->reg_base + KVASER_PCIEFD_KCAN_CMD_REG);
-
-		iowrite32(KVASER_PCIEFD_KCAN_IRQ_TFD,
-			  can->reg_base + KVASER_PCIEFD_KCAN_IEN_REG);
 	} else if (p->header[0] & KVASER_PCIEFD_SPACK_IDET &&
 		   p->header[0] & KVASER_PCIEFD_SPACK_IRM &&
 		   cmdseq == (p->header[1] & KVASER_PCIEFD_PACKET_SEQ_MSK) &&
@@ -1732,15 +1728,6 @@ static int kvaser_pciefd_transmit_irq(st
 	if (irq & KVASER_PCIEFD_KCAN_IRQ_TOF)
 		netdev_err(can->can.dev, "Tx FIFO overflow\n");
 
-	if (irq & KVASER_PCIEFD_KCAN_IRQ_TFD) {
-		u8 count = ioread32(can->reg_base +
-				    KVASER_PCIEFD_KCAN_TX_NPACKETS_REG) & 0xff;
-
-		if (count == 0)
-			iowrite32(KVASER_PCIEFD_KCAN_CTRL_EFLUSH,
-				  can->reg_base + KVASER_PCIEFD_KCAN_CTRL_REG);
-	}
-
 	if (irq & KVASER_PCIEFD_KCAN_IRQ_BPP)
 		netdev_err(can->can.dev,
 			   "Fail to change bittiming, when not in reset mode\n");
author maintainer
#252 lore raw

PATCH 252/292 can: kvaser_pciefd: Disable interrupts in probe error path

In reply to Greg Kroah-Hartman

From: Jimmy Assarsson <[email protected]>

commit 11164bc39459335ab93c6e99d53b7e4292fba38b upstream.

Disable interrupts in error path of probe function.

Fixes 26ad340e582d ("can: kvaser_pciefd: Add driver for Kvaser PCIEcan devices")
---
 drivers/net/can/kvaser_pciefd.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/net/can/kvaser_pciefd.c +2 -0
@@ -1861,6 +1861,8 @@ static int kvaser_pciefd_probe(struct pc
 	return 0;
 
 err_free_irq:
+	/* Disable PCI interrupts */
+	iowrite32(0, pcie->reg_base + KVASER_PCIEFD_IEN_REG);
 	free_irq(pcie->pci->irq, pcie);
 
 err_teardown_can_ctrls:
author maintainer
#253 lore raw

PATCH 253/292 wifi: rtw88: use work to update rate to avoid RCU warning

In reply to Greg Kroah-Hartman

From: Ping-Ke Shih <[email protected]>

commit bcafcb959a57a6890e900199690c5fc47da1a304 upstream.

The ieee80211_ops::sta_rc_update must be atomic, because ieee80211_chan_bw_change() holds rcu_read lock while calling drv_sta_rc_update(), so create a work to do original things.

Voluntary context switch within RCU read-side critical section!

 WARNING: CPU: 0 PID: 4621 at kernel/rcu/tree_plugin.h:318
 rcu_note_context_switch+0x571/0x5d0
 CPU: 0 PID: 4621 Comm: kworker/u16:2 Tainted: G        W  OE
 Workqueue: phy3 ieee80211_chswitch_work [mac80211]
 RIP: 0010:rcu_note_context_switch+0x571/0x5d0
 Call Trace:
  <TASK>
  __schedule+0xb0/0x1460
  ? __mod_timer+0x116/0x360
  schedule+0x5a/0xc0
  schedule_timeout+0x87/0x150
  ? trace_raw_output_tick_stop+0x60/0x60
  wait_for_completion_timeout+0x7b/0x140
  usb_start_wait_urb+0x82/0x160 [usbcore
  usb_control_msg+0xe3/0x140 [usbcore
  rtw_usb_read+0x88/0xe0 [rtw_usb
  rtw_usb_read8+0xf/0x10 [rtw_usb
  rtw_fw_send_h2c_command+0xa0/0x170 [rtw_core
  rtw_fw_send_ra_info+0xc9/0xf0 [rtw_core
  drv_sta_rc_update+0x7c/0x160 [mac80211
  ieee80211_chan_bw_change+0xfb/0x110 [mac80211
  ieee80211_change_chanctx+0x38/0x130 [mac80211
  ieee80211_vif_use_reserved_switch+0x34e/0x900 [mac80211
  ieee80211_link_use_reserved_context+0x88/0xe0 [mac80211
  ieee80211_chswitch_work+0x95/0x170 [mac80211
  process_one_work+0x201/0x410
  worker_thread+0x4a/0x3b0
  ? process_one_work+0x410/0x410
  kthread+0xe1/0x110
  ? kthread_complete_and_exit+0x20/0x20
  ret_from_fork+0x1f/0x30
  </TASK>
---
 drivers/net/wireless/realtek/rtw88/mac80211.c |    2 +-
 drivers/net/wireless/realtek/rtw88/main.c     |   15 +++++++++++++++
 drivers/net/wireless/realtek/rtw88/main.h     |    3 +++
 3 files changed, 19 insertions(+), 1 deletion(-)
drivers/net/wireless/realtek/rtw88/mac80211.c +1 -1
@@ -891,7 +891,7 @@ static void rtw_ops_sta_rc_update(struct
 	struct rtw_sta_info *si = (struct rtw_sta_info *)sta->drv_priv;
 
 	if (changed & IEEE80211_RC_BW_CHANGED)
-		rtw_update_sta_info(rtwdev, si, true);
+		ieee80211_queue_work(rtwdev->hw, &si->rc_work);
 }
 
 const struct ieee80211_ops rtw_ops = {
drivers/net/wireless/realtek/rtw88/main.c +15 -0
@@ -296,6 +296,17 @@ static u8 rtw_acquire_macid(struct rtw_d
 	return mac_id;
 }
 
+static void rtw_sta_rc_work(struct work_struct *work)
+{
+	struct rtw_sta_info *si = container_of(work, struct rtw_sta_info,
+					       rc_work);
+	struct rtw_dev *rtwdev = si->rtwdev;
+
+	mutex_lock(&rtwdev->mutex);
+	rtw_update_sta_info(rtwdev, si, true);
+	mutex_unlock(&rtwdev->mutex);
+}
+
 int rtw_sta_add(struct rtw_dev *rtwdev, struct ieee80211_sta *sta,
 		struct ieee80211_vif *vif)
 {
@@ -306,12 +317,14 @@ int rtw_sta_add(struct rtw_dev *rtwdev,
 	if (si->mac_id >= RTW_MAX_MAC_ID_NUM)
 		return -ENOSPC;
 
+	si->rtwdev = rtwdev;
 	si->sta = sta;
 	si->vif = vif;
 	si->init_ra_lv = 1;
 	ewma_rssi_init(&si->avg_rssi);
 	for (i = 0; i < ARRAY_SIZE(sta->txq); i++)
 		rtw_txq_init(rtwdev, sta->txq[i]);
+	INIT_WORK(&si->rc_work, rtw_sta_rc_work);
 
 	rtw_update_sta_info(rtwdev, si, true);
 	rtw_fw_media_status_report(rtwdev, si->mac_id, true);
@@ -330,6 +343,8 @@ void rtw_sta_remove(struct rtw_dev *rtwd
 	struct rtw_sta_info *si = (struct rtw_sta_info *)sta->drv_priv;
 	int i;
 
+	cancel_work_sync(&si->rc_work);
+
 	rtw_release_macid(rtwdev, si->mac_id);
 	if (fw_exist)
 		rtw_fw_media_status_report(rtwdev, si->mac_id, false);
drivers/net/wireless/realtek/rtw88/main.h +3 -0
@@ -734,6 +734,7 @@ struct rtw_txq {
 DECLARE_EWMA(rssi, 10, 16);
 
 struct rtw_sta_info {
+	struct rtw_dev *rtwdev;
 	struct ieee80211_sta *sta;
 	struct ieee80211_vif *vif;
 
@@ -758,6 +759,8 @@ struct rtw_sta_info {
 
 	bool use_cfg_mask;
 	struct cfg80211_bitrate_mask *mask;
+
+	struct work_struct rc_work;
 };
 
 enum rtw_bfee_role {
author maintainer
#254 lore raw

PATCH 254/292 SMB3: Close all deferred handles of inode in case of handle lease break

In reply to Greg Kroah-Hartman

From: Bharath SM <[email protected]>

commit 47592fa8eb03742048b096b4696ec133384c45eb upstream.

Oplock break may occur for different file handle than the deferred handle. Check for inode deferred closes list, if it's not empty then close all the deferred handles of inode because we should not cache handles if we dont have handle lease.

Eg: If openfilelist has one deferred file handle and another open file handle from app for a same file, then on a lease break we choose the first handle in openfile list. The first handle in list can be deferred handle or actual open file handle from app. In case if it is actual open handle then today, we don't close deferred handles if we lose handle lease on a file. Problem with this is, later if app decides to close the existing open handle then we still be caching deferred handles until deferred close timeout. Leaving open handle may result in sharing violation when windows client tries to open a file with limited file share access.

So we should check for deferred list of inode and walk through the list of deferred files in inode and close all deferred files.

Fixes 9e31678fb403 ("SMB3: fix lease break timeout when multiple deferred close handles for the same file.")
---
 fs/cifs/file.c |    9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)
fs/cifs/file.c +1 -8
@@ -5087,8 +5087,6 @@ void cifs_oplock_break(struct work_struc
 	struct TCP_Server_Info *server = tcon->ses->server;
 	int rc = 0;
 	bool purge_cache = false;
-	struct cifs_deferred_close *dclose;
-	bool is_deferred = false;
 
 	wait_on_bit(&cinode->flags, CIFS_INODE_PENDING_WRITERS,
 			TASK_UNINTERRUPTIBLE);
@@ -5129,14 +5127,9 @@ oplock_break_ack:
 	 * file handles but cached, then schedule deferred close immediately.
 	 * So, new open will not use cached handle.
 	 */
-	spin_lock(&CIFS_I(inode)->deferred_lock);
-	is_deferred = cifs_is_deferred_close(cfile, &dclose);
-	spin_unlock(&CIFS_I(inode)->deferred_lock);
 
-	if (!CIFS_CACHE_HANDLE(cinode) && is_deferred &&
-			cfile->deferred_close_scheduled && delayed_work_pending(&cfile->deferred)) {
+	if (!CIFS_CACHE_HANDLE(cinode) && !list_empty(&cinode->deferred_closes))
 		cifs_close_deferred_file(cinode);
-	}
 
 	/*
 	 * releasing stale oplock after recent reconnect of smb session using
author maintainer
#255 lore raw

PATCH 255/292 SMB3: drop reference to cfile before sending oplock break

In reply to Greg Kroah-Hartman

From: Bharath SM <[email protected]>

commit 59a556aebc43dded08535fe97d94ca3f657915e4 upstream.

In cifs_oplock_break function we drop reference to a cfile at the end of function, due to which close command goes on wire after lease break acknowledgment even if file is already closed by application but we had deferred the handle close.
If other client with limited file shareaccess waiting on lease break ack proceeds operation on that file as soon as first client sends ack, then we may encounter status sharing violation error because of open handle.
Solution is to put reference to cfile(send close on wire if last ref) and then send oplock acknowledgment to server.

Fixes 9e31678fb403 ("SMB3: fix lease break timeout when multiple deferred close handles for the same file.")
---
 fs/cifs/cifsglob.h |    4 ++--
 fs/cifs/file.c     |   17 ++++++++++++-----
 fs/cifs/smb1ops.c  |    9 ++++-----
 fs/cifs/smb2ops.c  |    7 +++----
 4 files changed, 21 insertions(+), 16 deletions(-)
fs/cifs/cifsglob.h +2 -2
@@ -428,8 +428,8 @@ struct smb_version_operations {
 	/* check for STATUS_NETWORK_SESSION_EXPIRED */
 	bool (*is_session_expired)(char *);
 	/* send oplock break response */
-	int (*oplock_response)(struct cifs_tcon *, struct cifs_fid *,
-			       struct cifsInodeInfo *);
+	int (*oplock_response)(struct cifs_tcon *tcon, __u64 persistent_fid, __u64 volatile_fid,
+			__u16 net_fid, struct cifsInodeInfo *cifs_inode);
 	/* query remote filesystem */
 	int (*queryfs)(const unsigned int, struct cifs_tcon *,
 		       struct cifs_sb_info *, struct kstatfs *);
fs/cifs/file.c +12 -5
@@ -5086,7 +5086,9 @@ void cifs_oplock_break(struct work_struc
 	struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
 	struct TCP_Server_Info *server = tcon->ses->server;
 	int rc = 0;
-	bool purge_cache = false;
+	bool purge_cache = false, oplock_break_cancelled;
+	__u64 persistent_fid, volatile_fid;
+	__u16 net_fid;
 
 	wait_on_bit(&cinode->flags, CIFS_INODE_PENDING_WRITERS,
 			TASK_UNINTERRUPTIBLE);
@@ -5131,19 +5133,24 @@ oplock_break_ack:
 	if (!CIFS_CACHE_HANDLE(cinode) && !list_empty(&cinode->deferred_closes))
 		cifs_close_deferred_file(cinode);
 
+	persistent_fid = cfile->fid.persistent_fid;
+	volatile_fid = cfile->fid.volatile_fid;
+	net_fid = cfile->fid.netfid;
+	oplock_break_cancelled = cfile->oplock_break_cancelled;
+
+	_cifsFileInfo_put(cfile, false /* do not wait for ourself */, false);
 	/*
 	 * releasing stale oplock after recent reconnect of smb session using
 	 * a now incorrect file handle is not a data integrity issue but do
 	 * not bother sending an oplock release if session to server still is
 	 * disconnected since oplock already released by the server
 	 */
-	if (!cfile->oplock_break_cancelled) {
-		rc = tcon->ses->server->ops->oplock_response(tcon, &cfile->fid,
-							     cinode);
+	if (!oplock_break_cancelled) {
+		rc = tcon->ses->server->ops->oplock_response(tcon, persistent_fid,
+				volatile_fid, net_fid, cinode);
 		cifs_dbg(FYI, "Oplock release rc = %d\n", rc);
 	}
 
-	_cifsFileInfo_put(cfile, false /* do not wait for ourself */, false);
 	cifs_done_oplock_break(cinode);
 }
 
fs/cifs/smb1ops.c +4 -5
@@ -897,12 +897,11 @@ cifs_close_dir(const unsigned int xid, s
 }
 
 static int
-cifs_oplock_response(struct cifs_tcon *tcon, struct cifs_fid *fid,
-		     struct cifsInodeInfo *cinode)
+cifs_oplock_response(struct cifs_tcon *tcon, __u64 persistent_fid,
+		__u64 volatile_fid, __u16 net_fid, struct cifsInodeInfo *cinode)
 {
-	return CIFSSMBLock(0, tcon, fid->netfid, current->tgid, 0, 0, 0, 0,
-			   LOCKING_ANDX_OPLOCK_RELEASE, false,
-			   CIFS_CACHE_READ(cinode) ? 1 : 0);
+	return CIFSSMBLock(0, tcon, net_fid, current->tgid, 0, 0, 0, 0,
+			   LOCKING_ANDX_OPLOCK_RELEASE, false, CIFS_CACHE_READ(cinode) ? 1 : 0);
 }
 
 static int
fs/cifs/smb2ops.c +3 -4
@@ -2383,15 +2383,14 @@ smb2_is_network_name_deleted(char *buf,
 }
 
 static int
-smb2_oplock_response(struct cifs_tcon *tcon, struct cifs_fid *fid,
-		     struct cifsInodeInfo *cinode)
+smb2_oplock_response(struct cifs_tcon *tcon, __u64 persistent_fid,
+		__u64 volatile_fid, __u16 net_fid, struct cifsInodeInfo *cinode)
 {
 	if (tcon->ses->server->capabilities & SMB2_GLOBAL_CAP_LEASING)
 		return SMB2_lease_break(0, tcon, cinode->lease_key,
 					smb2_get_lease_state(cinode));
 
-	return SMB2_oplock_break(0, tcon, fid->persistent_fid,
-				 fid->volatile_fid,
+	return SMB2_oplock_break(0, tcon, persistent_fid, volatile_fid,
 				 CIFS_CACHE_READ(cinode) ? 1 : 0);
 }
 
author maintainer
#256 lore raw

PATCH 256/292 ksmbd: smb2: Allow messages padded to 8byte boundary

In reply to Greg Kroah-Hartman

From: Gustav Johansson <[email protected]>

commit e7b8b8ed9960bf699bf4029f482d9e869c094ed6 upstream.

clc length is now accepted to <= 8 less than length, rather than < 8.

Solve issues on some of Axis's smb clients which send messages where clc length is 8 bytes less than length.

The specific client was running kernel 4.19.217 with smb dialect 3.0.2 on armv7l.

---
 fs/ksmbd/smb2misc.c |    5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
fs/ksmbd/smb2misc.c +4 -1
@@ -416,8 +416,11 @@ int ksmbd_smb2_check_message(struct ksmb
 
 		/*
 		 * Allow a message that padded to 8byte boundary.
+		 * Linux 4.19.217 with smb 3.0.2 are sometimes
+		 * sending messages where the cls_len is exactly
+		 * 8 bytes less than len.
 		 */
-		if (clc_len < len && (len - clc_len) < 8)
+		if (clc_len < len && (len - clc_len) <= 8)
 			goto validate_credit;
 
 		pr_err_ratelimited(
author maintainer
#257 lore raw

PATCH 257/292 ksmbd: allocate one more byte for implied bcc[0]

In reply to Greg Kroah-Hartman

From: Chih-Yen Chang <[email protected]>

commit 443d61d1fa9faa60ef925513d83742902390100f upstream.

ksmbd_smb2_check_message allows client to return one byte more, so we need to allocate additional memory in ksmbd_conn_handler_loop to avoid out-of-bound access.

---
 fs/ksmbd/connection.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
fs/ksmbd/connection.c +2 -1
@@ -353,7 +353,8 @@ int ksmbd_conn_handler_loop(void *p)
 			break;
 
 		/* 4 for rfc1002 length field */
-		size = pdu_size + 4;
+		/* 1 for implied bcc[0] */
+		size = pdu_size + 4 + 1;
 		conn->request_buf = kvmalloc(size, GFP_KERNEL);
 		if (!conn->request_buf)
 			break;
author maintainer
#258 lore raw

PATCH 258/292 ksmbd: fix wrong UserName check in session_user

In reply to Greg Kroah-Hartman

From: Chih-Yen Chang <[email protected]>

commit f0a96d1aafd8964e1f9955c830a3e5cb3c60a90f upstream.

The offset of UserName is related to the address of security buffer. To ensure the validaty of UserName, we need to compare name_off
+ name_len with secbuf_len instead of auth_msg_len.

[   27.096243] ==================================================================
[   27.096890] BUG: KASAN: slab-out-of-bounds in smb_strndup_from_utf16+0x188/0x350
[   27.097609] Read of size 2 at addr ffff888005e3b542 by task kworker/0:0/7
...
[   27.099950] Call Trace:
[   27.100194]  <TASK>
[   27.100397]  dump_stack_lvl+0x33/0x50
[   27.100752]  print_report+0xcc/0x620
[   27.102305]  kasan_report+0xae/0xe0
[   27.103072]  kasan_check_range+0x35/0x1b0
[   27.103757]  smb_strndup_from_utf16+0x188/0x350
[   27.105474]  smb2_sess_setup+0xaf8/0x19c0
[   27.107935]  handle_ksmbd_work+0x274/0x810
[   27.108315]  process_one_work+0x419/0x760
[   27.108689]  worker_thread+0x2a2/0x6f0
[   27.109385]  kthread+0x160/0x190
[   27.110129]  ret_from_fork+0x1f/0x30
[   27.110454]  </TASK>
---
 fs/ksmbd/smb2pdu.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
fs/ksmbd/smb2pdu.c +2 -3
@@ -1373,7 +1373,7 @@ static struct ksmbd_user *session_user(s
 	struct authenticate_message *authblob;
 	struct ksmbd_user *user;
 	char *name;
-	unsigned int auth_msg_len, name_off, name_len, secbuf_len;
+	unsigned int name_off, name_len, secbuf_len;
 
 	secbuf_len = le16_to_cpu(req->SecurityBufferLength);
 	if (secbuf_len < sizeof(struct authenticate_message)) {
@@ -1383,9 +1383,8 @@ static struct ksmbd_user *session_user(s
 	authblob = user_authblob(conn, req);
 	name_off = le32_to_cpu(authblob->UserName.BufferOffset);
 	name_len = le16_to_cpu(authblob->UserName.Length);
-	auth_msg_len = le16_to_cpu(req->SecurityBufferOffset) + secbuf_len;
 
-	if (auth_msg_len < (u64)name_off + name_len)
+	if (secbuf_len < (u64)name_off + name_len)
 		return NULL;
 
 	name = smb_strndup_from_utf16((const char *)authblob + name_off,
author maintainer
#259 lore raw

PATCH 259/292 ksmbd: fix global-out-of-bounds in smb2_find_context_vals

In reply to Greg Kroah-Hartman

From: Chih-Yen Chang <[email protected]>

commit 02f76c401d17e409ed45bf7887148fcc22c93c85 upstream.

Add tag_len argument in smb2_find_context_vals() to avoid out-of-bound read when create_context's name_len is larger than tag length.

[    7.995411] ==================================================================
[    7.995866] BUG: KASAN: global-out-of-bounds in memcmp+0x83/0xa0
[    7.996248] Read of size 8 at addr ffffffff8258d940 by task kworker/0:0/7
...
[    7.998191] Call Trace:
[    7.998358]  <TASK>
[    7.998503]  dump_stack_lvl+0x33/0x50
[    7.998743]  print_report+0xcc/0x620
[    7.999458]  kasan_report+0xae/0xe0
[    7.999895]  kasan_check_range+0x35/0x1b0
[    8.000152]  memcmp+0x83/0xa0
[    8.000347]  smb2_find_context_vals+0xf7/0x1e0
[    8.000635]  smb2_open+0x1df2/0x43a0
[    8.006398]  handle_ksmbd_work+0x274/0x810
[    8.006666]  process_one_work+0x419/0x760
[    8.006922]  worker_thread+0x2a2/0x6f0
[    8.007429]  kthread+0x160/0x190
[    8.007946]  ret_from_fork+0x1f/0x30
[    8.008181]  </TASK>
---
 fs/ksmbd/oplock.c  |    5 +++--
 fs/ksmbd/oplock.h  |    2 +-
 fs/ksmbd/smb2pdu.c |   14 +++++++-------
 3 files changed, 11 insertions(+), 10 deletions(-)
fs/ksmbd/oplock.c +3 -2
@@ -1449,11 +1449,12 @@ struct lease_ctx_info *parse_lease_state
  * smb2_find_context_vals() - find a particular context info in open request
  * @open_req:	buffer containing smb2 file open(create) request
  * @tag:	context name to search for
+ * @tag_len:	the length of tag
  *
  * Return:	pointer to requested context, NULL if @str context not found
  *		or error pointer if name length is invalid.
  */
-struct create_context *smb2_find_context_vals(void *open_req, const char *tag)
+struct create_context *smb2_find_context_vals(void *open_req, const char *tag, int tag_len)
 {
 	struct create_context *cc;
 	unsigned int next = 0;
@@ -1492,7 +1493,7 @@ struct create_context *smb2_find_context
 			return ERR_PTR(-EINVAL);
 
 		name = (char *)cc + name_off;
-		if (memcmp(name, tag, name_len) == 0)
+		if (name_len == tag_len && !memcmp(name, tag, name_len))
 			return cc;
 
 		remain_len -= next;
fs/ksmbd/oplock.h +1 -1
@@ -118,7 +118,7 @@ void create_durable_v2_rsp_buf(char *cc,
 void create_mxac_rsp_buf(char *cc, int maximal_access);
 void create_disk_id_rsp_buf(char *cc, __u64 file_id, __u64 vol_id);
 void create_posix_rsp_buf(char *cc, struct ksmbd_file *fp);
-struct create_context *smb2_find_context_vals(void *open_req, const char *str);
+struct create_context *smb2_find_context_vals(void *open_req, const char *tag, int tag_len);
 struct oplock_info *lookup_lease_in_table(struct ksmbd_conn *conn,
 					  char *lease_key);
 int find_same_lease_key(struct ksmbd_session *sess, struct ksmbd_inode *ci,
fs/ksmbd/smb2pdu.c +7 -7
@@ -2478,7 +2478,7 @@ static int smb2_create_sd_buffer(struct
 		return -ENOENT;
 
 	/* Parse SD BUFFER create contexts */
-	context = smb2_find_context_vals(req, SMB2_CREATE_SD_BUFFER);
+	context = smb2_find_context_vals(req, SMB2_CREATE_SD_BUFFER, 4);
 	if (!context)
 		return -ENOENT;
 	else if (IS_ERR(context))
@@ -2680,7 +2680,7 @@ int smb2_open(struct ksmbd_work *work)
 
 	if (req->CreateContextsOffset) {
 		/* Parse non-durable handle create contexts */
-		context = smb2_find_context_vals(req, SMB2_CREATE_EA_BUFFER);
+		context = smb2_find_context_vals(req, SMB2_CREATE_EA_BUFFER, 4);
 		if (IS_ERR(context)) {
 			rc = PTR_ERR(context);
 			goto err_out1;
@@ -2700,7 +2700,7 @@ int smb2_open(struct ksmbd_work *work)
 		}
 
 		context = smb2_find_context_vals(req,
-						 SMB2_CREATE_QUERY_MAXIMAL_ACCESS_REQUEST);
+						 SMB2_CREATE_QUERY_MAXIMAL_ACCESS_REQUEST, 4);
 		if (IS_ERR(context)) {
 			rc = PTR_ERR(context);
 			goto err_out1;
@@ -2711,7 +2711,7 @@ int smb2_open(struct ksmbd_work *work)
 		}
 
 		context = smb2_find_context_vals(req,
-						 SMB2_CREATE_TIMEWARP_REQUEST);
+						 SMB2_CREATE_TIMEWARP_REQUEST, 4);
 		if (IS_ERR(context)) {
 			rc = PTR_ERR(context);
 			goto err_out1;
@@ -2723,7 +2723,7 @@ int smb2_open(struct ksmbd_work *work)
 
 		if (tcon->posix_extensions) {
 			context = smb2_find_context_vals(req,
-							 SMB2_CREATE_TAG_POSIX);
+							 SMB2_CREATE_TAG_POSIX, 16);
 			if (IS_ERR(context)) {
 				rc = PTR_ERR(context);
 				goto err_out1;
@@ -3122,7 +3122,7 @@ int smb2_open(struct ksmbd_work *work)
 		struct create_alloc_size_req *az_req;
 
 		az_req = (struct create_alloc_size_req *)smb2_find_context_vals(req,
-					SMB2_CREATE_ALLOCATION_SIZE);
+					SMB2_CREATE_ALLOCATION_SIZE, 4);
 		if (IS_ERR(az_req)) {
 			rc = PTR_ERR(az_req);
 			goto err_out;
@@ -3149,7 +3149,7 @@ int smb2_open(struct ksmbd_work *work)
 					    err);
 		}
 
-		context = smb2_find_context_vals(req, SMB2_CREATE_QUERY_ON_DISK_ID);
+		context = smb2_find_context_vals(req, SMB2_CREATE_QUERY_ON_DISK_ID, 4);
 		if (IS_ERR(context)) {
 			rc = PTR_ERR(context);
 			goto err_out;
author maintainer
#260 lore raw

PATCH 260/292 KVM: Fix vcpu_array[0] races

In reply to Greg Kroah-Hartman

From: Michal Luczaj <[email protected]>

commit afb2acb2e3a32e4d56f7fbd819769b98ed1b7520 upstream.

In kvm_vm_ioctl_create_vcpu(), add vcpu to vcpu_array iff it's safe to access vcpu via kvm_get_vcpu() and kvm_for_each_vcpu(), i.e. when there's no failure path requiring vcpu removal and destruction. Such order is important because vcpu_array accessors may end up referencing vcpu at vcpu_array[0] even before online_vcpus is set to 1.

When online_vcpus=0, any call to kvm_get_vcpu() goes through array_index_nospec() and ends with an attempt to xa_load(vcpu_array, 0):

	int num_vcpus = atomic_read(&kvm->online_vcpus);
	i = array_index_nospec(i, num_vcpus);
	return xa_load(&kvm->vcpu_array, i);

Similarly, when online_vcpus=0, a kvm_for_each_vcpu() does not iterate over an "empty" range, but actually [0, ULONG_MAX]:

	xa_for_each_range(&kvm->vcpu_array, idx, vcpup, 0, \
			  (atomic_read(&kvm->online_vcpus) - 1))

In both cases, such online_vcpus=0 edge case, even if leading to unnecessary calls to XArray API, should not be an issue; requesting unpopulated indexes/ranges is handled by xa_load() and xa_for_each_range().

However, this means that when the first vCPU is created and inserted in vcpu_array *and* before online_vcpus is incremented, code calling kvm_get_vcpu()/kvm_for_each_vcpu() already has access to that first vCPU.

This should not pose a problem assuming that once a vcpu is stored in vcpu_array, it will remain there, but that's not the case:
kvm_vm_ioctl_create_vcpu() first inserts to vcpu_array, then requests a file descriptor. If create_vcpu_fd() fails, newly inserted vcpu is removed from the vcpu_array, then destroyed:

	vcpu->vcpu_idx = atomic_read(&kvm->online_vcpus);
	r = xa_insert(&kvm->vcpu_array, vcpu->vcpu_idx, vcpu, GFP_KERNEL_ACCOUNT);
	kvm_get_kvm(kvm);
	r = create_vcpu_fd(vcpu);
	if (r < 0) {
		xa_erase(&kvm->vcpu_array, vcpu->vcpu_idx);
		kvm_put_kvm_no_destroy(kvm);
		goto unlock_vcpu_destroy;
	}
	atomic_inc(&kvm->online_vcpus);

This results in a possible race condition when a reference to a vcpu is acquired (via kvm_get_vcpu() or kvm_for_each_vcpu()) moments before said vcpu is destroyed.

Message-Id: <[email protected]>

Fixes c5b077549136 ("KVM: Convert the kvm->vcpus array to a xarray", 2021-12-08)
---
 virt/kvm/kvm_main.c |   16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)
virt/kvm/kvm_main.c +10 -6
@@ -3947,18 +3947,19 @@ static int kvm_vm_ioctl_create_vcpu(stru
 	}
 
 	vcpu->vcpu_idx = atomic_read(&kvm->online_vcpus);
-	r = xa_insert(&kvm->vcpu_array, vcpu->vcpu_idx, vcpu, GFP_KERNEL_ACCOUNT);
-	BUG_ON(r == -EBUSY);
+	r = xa_reserve(&kvm->vcpu_array, vcpu->vcpu_idx, GFP_KERNEL_ACCOUNT);
 	if (r)
 		goto unlock_vcpu_destroy;
 
 	/* Now it's all set up, let userspace reach it */
 	kvm_get_kvm(kvm);
 	r = create_vcpu_fd(vcpu);
-	if (r < 0) {
-		xa_erase(&kvm->vcpu_array, vcpu->vcpu_idx);
-		kvm_put_kvm_no_destroy(kvm);
-		goto unlock_vcpu_destroy;
+	if (r < 0)
+		goto kvm_put_xa_release;
+
+	if (KVM_BUG_ON(!!xa_store(&kvm->vcpu_array, vcpu->vcpu_idx, vcpu, 0), kvm)) {
+		r = -EINVAL;
+		goto kvm_put_xa_release;
 	}
 
 	/*
@@ -3973,6 +3974,9 @@ static int kvm_vm_ioctl_create_vcpu(stru
 	kvm_create_vcpu_debugfs(vcpu);
 	return r;
 
+kvm_put_xa_release:
+	kvm_put_kvm_no_destroy(kvm);
+	xa_release(&kvm->vcpu_array, vcpu->vcpu_idx);
 unlock_vcpu_destroy:
 	mutex_unlock(&kvm->lock);
 	kvm_dirty_ring_free(&vcpu->dirty_ring);
author maintainer
#261 lore raw

PATCH 261/292 statfs: enforce statfs[64] structure initialization

In reply to Greg Kroah-Hartman

From: Ilya Leoshkevich <[email protected]>

commit ed40866ec7d328b3dfb70db7e2011640a16202c3 upstream.

s390's struct statfs and struct statfs64 contain padding, which field-by-field copying does not set. Initialize the respective structs with zeros before filling them and copying them to userspace, like it's already done for the compat versions of these structs.

Found by KMSAN.

---
 fs/statfs.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
fs/statfs.c +2 -2
@@ -130,6 +130,7 @@ static int do_statfs_native(struct kstat
 	if (sizeof(buf) == sizeof(*st))
 		memcpy(&buf, st, sizeof(*st));
 	else {
+		memset(&buf, 0, sizeof(buf));
 		if (sizeof buf.f_blocks == 4) {
 			if ((st->f_blocks | st->f_bfree | st->f_bavail |
 			     st->f_bsize | st->f_frsize) &
@@ -158,7 +159,6 @@ static int do_statfs_native(struct kstat
 		buf.f_namelen = st->f_namelen;
 		buf.f_frsize = st->f_frsize;
 		buf.f_flags = st->f_flags;
-		memset(buf.f_spare, 0, sizeof(buf.f_spare));
 	}
 	if (copy_to_user(p, &buf, sizeof(buf)))
 		return -EFAULT;
@@ -171,6 +171,7 @@ static int do_statfs64(struct kstatfs *s
 	if (sizeof(buf) == sizeof(*st))
 		memcpy(&buf, st, sizeof(*st));
 	else {
+		memset(&buf, 0, sizeof(buf));
 		buf.f_type = st->f_type;
 		buf.f_bsize = st->f_bsize;
 		buf.f_blocks = st->f_blocks;
@@ -182,7 +183,6 @@ static int do_statfs64(struct kstatfs *s
 		buf.f_namelen = st->f_namelen;
 		buf.f_frsize = st->f_frsize;
 		buf.f_flags = st->f_flags;
-		memset(buf.f_spare, 0, sizeof(buf.f_spare));
 	}
 	if (copy_to_user(p, &buf, sizeof(buf)))
 		return -EFAULT;
author maintainer
#262 lore raw

PATCH 262/292 maple_tree: make maple state reusable after mas_empty_area()

In reply to Greg Kroah-Hartman

From: Peng Zhang <[email protected]>

commit 0257d9908d38c0b1669af4bb1bc4dbca1f273fe6 upstream.

Make mas->min and mas->max point to a node range instead of a leaf entry range. This allows mas to still be usable after mas_empty_area() returns. Users would get unexpected results from other operations on the maple state after calling the affected function.

For example, x86 MAP_32BIT mmap() acts as if there is no suitable gap when there should be one.

---
 lib/maple_tree.c |   12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)
lib/maple_tree.c +3 -9
@@ -5338,15 +5338,9 @@ int mas_empty_area(struct ma_state *mas,
 
 	mt = mte_node_type(mas->node);
 	pivots = ma_pivots(mas_mn(mas), mt);
-	if (offset)
-		mas->min = pivots[offset - 1] + 1;
-
-	if (offset < mt_pivots[mt])
-		mas->max = pivots[offset];
-
-	if (mas->index < mas->min)
-		mas->index = mas->min;
-
+	min = mas_safe_min(mas, pivots, offset);
+	if (mas->index < min)
+		mas->index = min;
 	mas->last = mas->index + size - 1;
 	return 0;
 }
author maintainer
#263 lore raw

PATCH 263/292 mm: fix zswap writeback race condition

In reply to Greg Kroah-Hartman

From: Domenico Cerasuolo <[email protected]>

commit 04fc7816089c5a32c29a04ec94b998e219dfb946 upstream.

The zswap writeback mechanism can cause a race condition resulting in memory corruption, where a swapped out page gets swapped in with data that was written to a different page.

The race unfolds like this:
1. a page with data A and swap offset X is stored in zswap
2. page A is removed off the LRU by zpool driver for writeback in zswap-shrink work, data for A is mapped by zpool driver
3. user space program faults and invalidates page entry A, offset X is considered free
4. kswapd stores page B at offset X in zswap (zswap could also be full, if so, page B would then be IOed to X, then skip step 5.)
5. entry A is replaced by B in tree->rbroot, this doesn't affect the local reference held by zswap-shrink work
6. zswap-shrink work writes back A at X, and frees zswap entry A
7. swapin of slot X brings A in memory instead of B

The fix:
Once the swap page cache has been allocated (case ZSWAP_SWAPCACHE_NEW), zswap-shrink work just checks that the local zswap_entry reference is still the same as the one in the tree. If it's not the same it means that it's either been invalidated or replaced, in both cases the writeback is aborted because the local entry contains stale data.

Reproducer:
I originally found this by running stress overnight to validate my work on the zswap writeback mechanism, it manifested after hours on my test machine. The key to make it happen is having zswap writebacks, so whatever setup pumps /sys/kernel/debug/zswap/written_back_pages should do the trick.

In order to reproduce this faster on a vm, I setup a system with ~100M of available memory and a 500M swap file, then running stress --vm 1 --vm-bytes 300000000 --vm-stride 4000 makes it happen in matter of tens of minutes. One can speed things up even more by swinging /sys/module/zswap/parameters/max_pool_percent up and down between, say, 20 and 1; this makes it reproduce in tens of seconds. It's crucial to set --vm-stride to something other than 4096 otherwise stress won't realize that memory has been corrupted because all pages would have the same data.

---
 mm/zswap.c |   16 ++++++++++++++++
 1 file changed, 16 insertions(+)
mm/zswap.c +16 -0
@@ -1002,6 +1002,22 @@ static int zswap_writeback_entry(struct
 		goto fail;
 
 	case ZSWAP_SWAPCACHE_NEW: /* page is locked */
+		/*
+		 * Having a local reference to the zswap entry doesn't exclude
+		 * swapping from invalidating and recycling the swap slot. Once
+		 * the swapcache is secured against concurrent swapping to and
+		 * from the slot, recheck that the entry is still current before
+		 * writing.
+		 */
+		spin_lock(&tree->lock);
+		if (zswap_rb_search(&tree->rbroot, entry->offset) != entry) {
+			spin_unlock(&tree->lock);
+			delete_from_swap_cache(page_folio(page));
+			ret = -ENOMEM;
+			goto fail;
+		}
+		spin_unlock(&tree->lock);
+
 		/* decompress */
 		acomp_ctx = raw_cpu_ptr(entry->pool->acomp_ctx);
 		dlen = PAGE_SIZE;
author maintainer
#264 lore raw

PATCH 264/292 serial: Add support for Advantech PCI-1611U card

In reply to Greg Kroah-Hartman

From: Vitaliy Tomin <[email protected]>

commit d2b00516de0e1d696724247098f6733a6ea53908 upstream.

Add support for Advantech PCI-1611U card

Advantech provides opensource drivers for this and many others card based on legacy copy of 8250_pci driver called adv950

https://www.advantech.com/emt/support/details/driver?id=1-TDOIMJ

It is hard to maintain to run as out of tree module on newer kernels. Just adding PCI ID to kernel 8250_pci works perfect.

---
 drivers/tty/serial/8250/8250_pci.c |    5 +++++
 1 file changed, 5 insertions(+)
drivers/tty/serial/8250/8250_pci.c +5 -0
@@ -1940,6 +1940,8 @@ pci_moxa_setup(struct serial_private *pr
 #define PCI_SUBDEVICE_ID_SIIG_DUAL_30	0x2530
 #define PCI_VENDOR_ID_ADVANTECH		0x13fe
 #define PCI_DEVICE_ID_INTEL_CE4100_UART 0x2e66
+#define PCI_DEVICE_ID_ADVANTECH_PCI1600	0x1600
+#define PCI_DEVICE_ID_ADVANTECH_PCI1600_1611	0x1611
 #define PCI_DEVICE_ID_ADVANTECH_PCI3620	0x3620
 #define PCI_DEVICE_ID_ADVANTECH_PCI3618	0x3618
 #define PCI_DEVICE_ID_ADVANTECH_PCIf618	0xf618
@@ -4105,6 +4107,9 @@ static SIMPLE_DEV_PM_OPS(pciserial_pm_op
 			 pciserial_resume_one);
 
 static const struct pci_device_id serial_pci_tbl[] = {
+	{	PCI_VENDOR_ID_ADVANTECH, PCI_DEVICE_ID_ADVANTECH_PCI1600,
+		PCI_DEVICE_ID_ADVANTECH_PCI1600_1611, PCI_ANY_ID, 0, 0,
+		pbn_b0_4_921600 },
 	/* Advantech use PCI_DEVICE_ID_ADVANTECH_PCI3620 (0x3620) as 'PCI_SUBVENDOR_ID' */
 	{	PCI_VENDOR_ID_ADVANTECH, PCI_DEVICE_ID_ADVANTECH_PCI3620,
 		PCI_DEVICE_ID_ADVANTECH_PCI3620, 0x0001, 0, 0,
author maintainer
#265 lore raw

PATCH 265/292 serial: 8250_exar: Add support for USR298x PCI Modems

In reply to Greg Kroah-Hartman

From: Andrew Davis <[email protected]>

commit 95d698869b404772cc8b72560df71548491c10bc upstream.

Possibly the last PCI controller-based (i.e. not a soft/winmodem) dial-up modem one can still buy.

Looks to have a stock XR17C154 PCI UART chip for communication, but for some reason when provisioning the PCI IDs they swapped the vendor and subvendor IDs. Otherwise this card would have worked out of the box.

Searching online, some folks seem to not have this issue and others do, so it is possible only some batches of cards have this error.

Create a new macro to handle the switched IDs and add support here.

---
 drivers/tty/serial/8250/8250_exar.c |   17 +++++++++++++++++
 1 file changed, 17 insertions(+)
drivers/tty/serial/8250/8250_exar.c +17 -0
@@ -40,9 +40,13 @@
 #define PCI_DEVICE_ID_COMMTECH_4224PCIE		0x0020
 #define PCI_DEVICE_ID_COMMTECH_4228PCIE		0x0021
 #define PCI_DEVICE_ID_COMMTECH_4222PCIE		0x0022
+
 #define PCI_DEVICE_ID_EXAR_XR17V4358		0x4358
 #define PCI_DEVICE_ID_EXAR_XR17V8358		0x8358
 
+#define PCI_SUBDEVICE_ID_USR_2980		0x0128
+#define PCI_SUBDEVICE_ID_USR_2981		0x0129
+
 #define PCI_DEVICE_ID_SEALEVEL_710xC		0x1001
 #define PCI_DEVICE_ID_SEALEVEL_720xC		0x1002
 #define PCI_DEVICE_ID_SEALEVEL_740xC		0x1004
@@ -829,6 +833,15 @@ static const struct exar8250_board pbn_e
 		(kernel_ulong_t)&bd			\
 	}
 
+#define USR_DEVICE(devid, sdevid, bd) {			\
+	PCI_DEVICE_SUB(					\
+		PCI_VENDOR_ID_USR,			\
+		PCI_DEVICE_ID_EXAR_##devid,		\
+		PCI_VENDOR_ID_EXAR,			\
+		PCI_SUBDEVICE_ID_USR_##sdevid), 0, 0,	\
+		(kernel_ulong_t)&bd			\
+	}
+
 static const struct pci_device_id exar_pci_tbl[] = {
 	EXAR_DEVICE(ACCESSIO, COM_2S, pbn_exar_XR17C15x),
 	EXAR_DEVICE(ACCESSIO, COM_4S, pbn_exar_XR17C15x),
@@ -853,6 +866,10 @@ static const struct pci_device_id exar_p
 
 	IBM_DEVICE(XR17C152, SATURN_SERIAL_ONE_PORT, pbn_exar_ibm_saturn),
 
+	/* USRobotics USR298x-OEM PCI Modems */
+	USR_DEVICE(XR17C152, 2980, pbn_exar_XR17C15x),
+	USR_DEVICE(XR17C152, 2981, pbn_exar_XR17C15x),
+
 	/* Exar Corp. XR17C15[248] Dual/Quad/Octal UART */
 	EXAR_DEVICE(EXAR, XR17C152, pbn_exar_XR17C15x),
 	EXAR_DEVICE(EXAR, XR17C154, pbn_exar_XR17C15x),
author maintainer
#266 lore raw

PATCH 266/292 serial: qcom-geni: fix enabling deactivated interrupt

In reply to Greg Kroah-Hartman

From: Krzysztof Kozlowski <[email protected]>

commit 5f949f140f73696f64acb89a1f16ff9153d017e0 upstream.

The driver have a race, experienced only with PREEMPT_RT patchset:

CPU0 | CPU1
==================================================================

qcom_geni_serial_probe       |
  uart_add_one_port          |
                             | serdev_drv_probe
                             |   qca_serdev_probe
                             |     serdev_device_open
                             |       uart_open
                             |         uart_startup
                             |           qcom_geni_serial_startup
                             |             enable_irq
                             |               __irq_startup
                             |                 WARN_ON()
                             |                 IRQ not activated
  request_threaded_irq       |
    irq_domain_activate_irq  |

The warning:

894000.serial: ttyHS1 at MMIO 0x894000 (irq = 144, base_baud = 0) is a MSM serial serial0: tty port ttyHS1 registered

  WARNING: CPU: 7 PID: 107 at kernel/irq/chip.c:241 __irq_startup+0x78/0xd8
  ...
  qcom_geni_serial 894000.serial: serial engine reports 0 RX bytes in!

Adding UART port triggers probe of child serial devices - serdev and eventually Qualcomm Bluetooth hci_qca driver. This opens UART port which enables the interrupt before it got activated in request_threaded_irq(). The issue originates in commit f3974413cf02 ("tty: serial: qcom_geni_serial: Wakeup IRQ cleanup") and discussion on mailing list [1]. However the above commit does not explain why the uart_add_one_port() is moved above requesting interrupt.

[1] https://lore.kernel.org/all/[email protected]/

---
 drivers/tty/serial/qcom_geni_serial.c |    9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)
drivers/tty/serial/qcom_geni_serial.c +4 -5
@@ -1483,19 +1483,18 @@ static int qcom_geni_serial_probe(struct
 	platform_set_drvdata(pdev, port);
 	port->handle_rx = console ? handle_rx_console : handle_rx_uart;
 
-	ret = uart_add_one_port(drv, uport);
-	if (ret)
-		return ret;
-
 	irq_set_status_flags(uport->irq, IRQ_NOAUTOEN);
 	ret = devm_request_irq(uport->dev, uport->irq, qcom_geni_serial_isr,
 			IRQF_TRIGGER_HIGH, port->name, uport);
 	if (ret) {
 		dev_err(uport->dev, "Failed to get IRQ ret %d\n", ret);
-		uart_remove_one_port(drv, uport);
 		return ret;
 	}
 
+	ret = uart_add_one_port(drv, uport);
+	if (ret)
+		return ret;
+
 	/*
 	 * Set pm_runtime status as ACTIVE so that wakeup_irq gets
 	 * enabled/disabled from dev_pm_arm_wake_irq during system
author maintainer
#267 lore raw

PATCH 267/292 thunderbolt: Clear registers properly when auto clear isnt in use

In reply to Greg Kroah-Hartman

From: Mario Limonciello <[email protected]>

commit c4af8e3fecd03b0aedcd38145955605cfebe7e3a upstream.

When QUIRK_AUTO_CLEAR_INT isn't set, interrupt masking should be cleared by writing to Interrupt Mask Clear (IMR) and interrupt status should be cleared properly at shutdown/init.

This fixes an error where interrupts are left enabled during resume from hibernation with CONFIG_USB4=y.

Fixes 468c49f44759 ("thunderbolt: Disable interrupt auto clear for rings")
---
 drivers/thunderbolt/nhi.c      |   29 ++++++++++++++++++++++++-----
 drivers/thunderbolt/nhi_regs.h |    2 ++
 2 files changed, 26 insertions(+), 5 deletions(-)
drivers/thunderbolt/nhi.c +24 -5
@@ -54,6 +54,21 @@ static int ring_interrupt_index(const st
 	return bit;
 }
 
+static void nhi_mask_interrupt(struct tb_nhi *nhi, int mask, int ring)
+{
+	if (nhi->quirks & QUIRK_AUTO_CLEAR_INT)
+		return;
+	iowrite32(mask, nhi->iobase + REG_RING_INTERRUPT_MASK_CLEAR_BASE + ring);
+}
+
+static void nhi_clear_interrupt(struct tb_nhi *nhi, int ring)
+{
+	if (nhi->quirks & QUIRK_AUTO_CLEAR_INT)
+		ioread32(nhi->iobase + REG_RING_NOTIFY_BASE + ring);
+	else
+		iowrite32(~0, nhi->iobase + REG_RING_INT_CLEAR + ring);
+}
+
 /*
  * ring_interrupt_active() - activate/deactivate interrupts for a single ring
  *
@@ -61,8 +76,8 @@ static int ring_interrupt_index(const st
  */
 static void ring_interrupt_active(struct tb_ring *ring, bool active)
 {
-	int reg = REG_RING_INTERRUPT_BASE +
-		  ring_interrupt_index(ring) / 32 * 4;
+	int index = ring_interrupt_index(ring) / 32 * 4;
+	int reg = REG_RING_INTERRUPT_BASE + index;
 	int interrupt_bit = ring_interrupt_index(ring) & 31;
 	int mask = 1 << interrupt_bit;
 	u32 old, new;
@@ -123,7 +138,11 @@ static void ring_interrupt_active(struct
 					 "interrupt for %s %d is already %s\n",
 					 RING_TYPE(ring), ring->hop,
 					 active ? "enabled" : "disabled");
-	iowrite32(new, ring->nhi->iobase + reg);
+
+	if (active)
+		iowrite32(new, ring->nhi->iobase + reg);
+	else
+		nhi_mask_interrupt(ring->nhi, mask, index);
 }
 
 /*
@@ -136,11 +155,11 @@ static void nhi_disable_interrupts(struc
 	int i = 0;
 	/* disable interrupts */
 	for (i = 0; i < RING_INTERRUPT_REG_COUNT(nhi); i++)
-		iowrite32(0, nhi->iobase + REG_RING_INTERRUPT_BASE + 4 * i);
+		nhi_mask_interrupt(nhi, ~0, 4 * i);
 
 	/* clear interrupt status bits */
 	for (i = 0; i < RING_NOTIFY_REG_COUNT(nhi); i++)
-		ioread32(nhi->iobase + REG_RING_NOTIFY_BASE + 4 * i);
+		nhi_clear_interrupt(nhi, 4 * i);
 }
 
 /* ring helper methods */
drivers/thunderbolt/nhi_regs.h +2 -0
@@ -93,6 +93,8 @@ struct ring_desc {
 #define REG_RING_INTERRUPT_BASE	0x38200
 #define RING_INTERRUPT_REG_COUNT(nhi) ((31 + 2 * nhi->hop_count) / 32)
 
+#define REG_RING_INTERRUPT_MASK_CLEAR_BASE	0x38208
+
 #define REG_INT_THROTTLING_RATE	0x38c00
 
 /* Interrupt Vector Allocation */
author maintainer
#268 lore raw

PATCH 268/292 vc_screen: reload load of struct vc_data pointer in vcs_write() to avoid UAF

In reply to Greg Kroah-Hartman

From: George Kennedy <[email protected]>

commit 8fb9ea65c9d1338b0d2bb0a9122dc942cdd32357 upstream.

After a call to console_unlock() in vcs_write() the vc_data struct can be freed by vc_port_destruct(). Because of that, the struct vc_data pointer must be reloaded in the while loop in vcs_write() after console_lock() to avoid a UAF when vcs_size() is called.

Syzkaller reported a UAF in vcs_size().

BUG: KASAN: slab-use-after-free in vcs_size (drivers/tty/vt/vc_screen.c:215)
Read of size 4 at addr ffff8880beab89a8 by task repro_vcs_size/4119

Call Trace:
 <TASK>

__asan_report_load4_noabort (mm/kasan/report_generic.c:380) vcs_size (drivers/tty/vt/vc_screen.c:215) vcs_write (drivers/tty/vt/vc_screen.c:664) vfs_write (fs/read_write.c:582 fs/read_write.c:564)
...

 <TASK>

Allocated by task 1213:
kmalloc_trace (mm/slab_common.c:1064)
vc_allocate (./include/linux/slab.h:559 ./include/linux/slab.h:680
    drivers/tty/vt/vt.c:1078 drivers/tty/vt/vt.c:1058)
con_install (drivers/tty/vt/vt.c:3334)
tty_init_dev (drivers/tty/tty_io.c:1303 drivers/tty/tty_io.c:1415
    drivers/tty/tty_io.c:1392)
tty_open (drivers/tty/tty_io.c:2082 drivers/tty/tty_io.c:2128)

chrdev_open (fs/char_dev.c:415)
do_dentry_open (fs/open.c:921)
vfs_open (fs/open.c:1052)
...

Freed by task 4116:

kfree (mm/slab_common.c:1016)
vc_port_destruct (drivers/tty/vt/vt.c:1044) tty_port_destructor (drivers/tty/tty_port.c:296) tty_port_put (drivers/tty/tty_port.c:312) vt_disallocate_all (drivers/tty/vt/vt_ioctl.c:662 (discriminator 2)) vt_ioctl (drivers/tty/vt/vt_ioctl.c:903)
tty_ioctl (drivers/tty/tty_io.c:2778)
...

The buggy address belongs to the object at ffff8880beab8800
 which belongs to the cache kmalloc-1k of size 1024

The buggy address is located 424 bytes inside of
freed 1024-byte region [ffff8880beab8800, ffff8880beab8c00)

The buggy address belongs to the physical page:
page:00000000afc77580 refcount:1 mapcount:0 mapping:0000000000000000
    index:0x0 pfn:0xbeab8

head:00000000afc77580 order:3 entire_mapcount:0 nr_pages_mapped:0
pincount:0
flags: 0xfffffc0010200(slab|head|node=0|zone=1|lastcpupid=0x1fffff)
page_type: 0xffffffff()

raw: 000fffffc0010200 ffff888100042dc0 ffffea000426de00 dead000000000002
raw: 0000000000000000 0000000000100010 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected

Memory state around the buggy address:
 ffff8880beab8880: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff8880beab8900: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff8880beab8980: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                                  ^
 ffff8880beab8a00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff8880beab8a80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
==================================================================

Disabling lock debugging due to kernel taint

Fixes ac751efa6a0d ("console: rename acquire/release_console_sem() to console_lock/unlock()")
Cc stable <[email protected]>
---
 drivers/tty/vt/vc_screen.c |   11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)
drivers/tty/vt/vc_screen.c +9 -2
@@ -656,10 +656,17 @@ vcs_write(struct file *file, const char
 			}
 		}
 
-		/* The vcs_size might have changed while we slept to grab
-		 * the user buffer, so recheck.
+		/* The vc might have been freed or vcs_size might have changed
+		 * while we slept to grab the user buffer, so recheck.
 		 * Return data written up to now on failure.
 		 */
+		vc = vcs_vc(inode, &viewed);
+		if (!vc) {
+			if (written)
+				break;
+			ret = -ENXIO;
+			goto unlock_out;
+		}
 		size = vcs_size(vc, attr, false);
 		if (size < 0) {
 			if (written)
author maintainer
#269 lore raw

PATCH 269/292 ceph: force updating the msg pointer in non-split case

In reply to Greg Kroah-Hartman

From: Xiubo Li <[email protected]>

commit 4cafd0400bcb6187c0d4ab4d4b0229a89ac4f8c2 upstream.

When the MClientSnap reqeust's op is not CEPH_SNAP_OP_SPLIT the request may still contain a list of 'split_realms', and we need to skip it anyway. Or it will be parsed as a corrupt snaptrace.

---
 fs/ceph/snap.c |   13 +++++++++++++
 1 file changed, 13 insertions(+)
fs/ceph/snap.c +13 -0
@@ -1111,6 +1111,19 @@ skip_inode:
 				continue;
 			adjust_snap_realm_parent(mdsc, child, realm->ino);
 		}
+	} else {
+		/*
+		 * In the non-split case both 'num_split_inos' and
+		 * 'num_split_realms' should be 0, making this a no-op.
+		 * However the MDS happens to populate 'split_realms' list
+		 * in one of the UPDATE op cases by mistake.
+		 *
+		 * Skip both lists just in case to ensure that 'p' is
+		 * positioned at the start of realm info, as expected by
+		 * ceph_update_snap_trace().
+		 */
+		p += sizeof(u64) * num_split_inos;
+		p += sizeof(u64) * num_split_realms;
 	}
 
 	/*
author maintainer
#270 lore raw

PATCH 270/292 drm/amd/pm: fix possible power mode mismatch between driver and PMFW

In reply to Greg Kroah-Hartman

From: Evan Quan <[email protected]>

commit bf4823267a817f7c155876a125b94336d7113e77 upstream.

PMFW may boots the ASIC with a different power mode from the system's real one. Notify PMFW explicitly the power mode the system in. This is needed only when ACDC switch via gpio is not supported.

---
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c            |   18 +++++++++++++++++
 drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c      |   20 -------------------
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c |    1 
 3 files changed, 20 insertions(+), 19 deletions(-)
drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c +18 -0
@@ -723,6 +723,24 @@ static int smu_late_init(void *handle)
 		return ret;
 	}
 
+	/*
+	 * Explicitly notify PMFW the power mode the system in. Since
+	 * the PMFW may boot the ASIC with a different mode.
+	 * For those supporting ACDC switch via gpio, PMFW will
+	 * handle the switch automatically. Driver involvement
+	 * is unnecessary.
+	 */
+	if (!smu->dc_controlled_by_gpio) {
+		ret = smu_set_power_source(smu,
+					   adev->pm.ac_power ? SMU_POWER_SOURCE_AC :
+					   SMU_POWER_SOURCE_DC);
+		if (ret) {
+			dev_err(adev->dev, "Failed to switch to %s mode!\n",
+				adev->pm.ac_power ? "AC" : "DC");
+			return ret;
+		}
+	}
+
 	if ((adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 1)) ||
 	    (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 3)))
 		return 0;
drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c +1 -19
@@ -3406,26 +3406,8 @@ static int navi10_post_smu_init(struct s
 		return 0;
 
 	ret = navi10_run_umc_cdr_workaround(smu);
-	if (ret) {
+	if (ret)
 		dev_err(adev->dev, "Failed to apply umc cdr workaround!\n");
-		return ret;
-	}
-
-	if (!smu->dc_controlled_by_gpio) {
-		/*
-		 * For Navi1X, manually switch it to AC mode as PMFW
-		 * may boot it with DC mode.
-		 */
-		ret = smu_v11_0_set_power_source(smu,
-						 adev->pm.ac_power ?
-						 SMU_POWER_SOURCE_AC :
-						 SMU_POWER_SOURCE_DC);
-		if (ret) {
-			dev_err(adev->dev, "Failed to switch to %s mode!\n",
-					adev->pm.ac_power ? "AC" : "DC");
-			return ret;
-		}
-	}
 
 	return ret;
 }
drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +1 -0
@@ -1767,6 +1767,7 @@ static const struct pptable_funcs smu_v1
 	.enable_mgpu_fan_boost = smu_v13_0_7_enable_mgpu_fan_boost,
 	.get_power_limit = smu_v13_0_7_get_power_limit,
 	.set_power_limit = smu_v13_0_set_power_limit,
+	.set_power_source = smu_v13_0_set_power_source,
 	.get_power_profile_mode = smu_v13_0_7_get_power_profile_mode,
 	.set_power_profile_mode = smu_v13_0_7_set_power_profile_mode,
 	.set_tool_table_location = smu_v13_0_set_tool_table_location,
author maintainer
#271 lore raw

PATCH 271/292 drm/amdgpu/gmc11: implement get_vbios_fb_size()

In reply to Greg Kroah-Hartman

From: Alex Deucher <[email protected]>

commit 68518294d00da6a2433357af75a63abc6030676e upstream.

Implement get_vbios_fb_size() so we can properly reserve the vbios splash screen to avoid potential artifacts on the screen during the transition from the pre-OS console to the OS console.

---
 drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c |   21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c +20 -1
@@ -31,6 +31,8 @@
 #include "umc_v8_10.h"
 #include "athub/athub_3_0_0_sh_mask.h"
 #include "athub/athub_3_0_0_offset.h"
+#include "dcn/dcn_3_2_0_offset.h"
+#include "dcn/dcn_3_2_0_sh_mask.h"
 #include "oss/osssys_6_0_0_offset.h"
 #include "ivsrcid/vmc/irqsrcs_vmc_1_0.h"
 #include "navi10_enum.h"
@@ -523,7 +525,24 @@ static void gmc_v11_0_get_vm_pte(struct
 
 static unsigned gmc_v11_0_get_vbios_fb_size(struct amdgpu_device *adev)
 {
-	return 0;
+	u32 d1vga_control = RREG32_SOC15(DCE, 0, regD1VGA_CONTROL);
+	unsigned size;
+
+	if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) {
+		size = AMDGPU_VBIOS_VGA_ALLOCATION;
+	} else {
+		u32 viewport;
+		u32 pitch;
+
+		viewport = RREG32_SOC15(DCE, 0, regHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION);
+		pitch = RREG32_SOC15(DCE, 0, regHUBPREQ0_DCSURF_SURFACE_PITCH);
+		size = (REG_GET_FIELD(viewport,
+					HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_HEIGHT) *
+				REG_GET_FIELD(pitch, HUBPREQ0_DCSURF_SURFACE_PITCH, PITCH) *
+				4);
+	}
+
+	return size;
 }
 
 static const struct amdgpu_gmc_funcs gmc_v11_0_gmc_funcs = {
author maintainer
#272 lore raw

PATCH 272/292 drm/amdgpu/gfx10: Disable gfxoff before disabling powergating.

In reply to Greg Kroah-Hartman

From: Bas Nieuwenhuizen <[email protected]>

commit 8173cab3368a13cdc3cad0bd5cf14e9399b0f501 upstream.

Otherwise we get a full system lock (looks like a FW mess).

Copied the order from the GFX9 powergating code.

---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c |    8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +7 -1
@@ -8236,8 +8236,14 @@ static int gfx_v10_0_set_powergating_sta
 	case IP_VERSION(10, 3, 3):
 	case IP_VERSION(10, 3, 6):
 	case IP_VERSION(10, 3, 7):
+		if (!enable)
+			amdgpu_gfx_off_ctrl(adev, false);
+
 		gfx_v10_cntl_pg(adev, enable);
-		amdgpu_gfx_off_ctrl(adev, enable);
+
+		if (enable)
+			amdgpu_gfx_off_ctrl(adev, true);
+
 		break;
 	default:
 		break;
author maintainer
#273 lore raw

PATCH 273/292 drm/amdgpu/gfx11: Adjust gfxoff before powergating on gfx11 as well

In reply to Greg Kroah-Hartman

From: Guilherme G. Piccoli <[email protected]>

commit 11fbdda2ab6bf049e2869139c07016022b4e045b upstream.

(Bas: speculative change to mirror gfx10/gfx9)

---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c |    8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +7 -1
@@ -5085,8 +5085,14 @@ static int gfx_v11_0_set_powergating_sta
 		break;
 	case IP_VERSION(11, 0, 1):
 	case IP_VERSION(11, 0, 4):
+		if (!enable)
+			amdgpu_gfx_off_ctrl(adev, false);
+
 		gfx_v11_cntl_pg(adev, enable);
-		amdgpu_gfx_off_ctrl(adev, enable);
+
+		if (enable)
+			amdgpu_gfx_off_ctrl(adev, true);
+
 		break;
 	default:
 		break;
author maintainer
#274 lore raw

PATCH 274/292 drm/amdgpu: refine get gpu clock counter method

In reply to Greg Kroah-Hartman

From: Tong Liu01 <[email protected]>

commit 5591a051b86be170a84943698ab140342602ff7b upstream.

[why]
regGOLDEN_TSC_COUNT_LOWER/regGOLDEN_TSC_COUNT_UPPER are protected and unaccessible under sriov.
The clock counter high bit may update during reading process.

[How]
Replace regGOLDEN_TSC_COUNT_LOWER/regGOLDEN_TSC_COUNT_UPPER with regCP_MES_MTIME_LO/regCP_MES_MTIME_HI to get gpu clock under sriov. Refine get gpu clock counter method to make the result more precise.

---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c |   17 +++++++++++++++--
 1 file changed, 15 insertions(+), 2 deletions(-)
drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +15 -2
@@ -4641,11 +4641,24 @@ static int gfx_v11_0_post_soft_reset(voi
 static uint64_t gfx_v11_0_get_gpu_clock_counter(struct amdgpu_device *adev)
 {
 	uint64_t clock;
+	uint64_t clock_counter_lo, clock_counter_hi_pre, clock_counter_hi_after;
 
 	amdgpu_gfx_off_ctrl(adev, false);
 	mutex_lock(&adev->gfx.gpu_clock_mutex);
-	clock = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_LOWER) |
-		((uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_UPPER) << 32ULL);
+	if (amdgpu_sriov_vf(adev)) {
+		clock_counter_hi_pre = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_HI);
+		clock_counter_lo = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_LO);
+		clock_counter_hi_after = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_HI);
+		if (clock_counter_hi_pre != clock_counter_hi_after)
+			clock_counter_lo = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_LO);
+	} else {
+		clock_counter_hi_pre = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_UPPER);
+		clock_counter_lo = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_LOWER);
+		clock_counter_hi_after = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_UPPER);
+		if (clock_counter_hi_pre != clock_counter_hi_after)
+			clock_counter_lo = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_LOWER);
+	}
+	clock = clock_counter_lo | (clock_counter_hi_after << 32ULL);
 	mutex_unlock(&adev->gfx.gpu_clock_mutex);
 	amdgpu_gfx_off_ctrl(adev, true);
 	return clock;
author maintainer
#275 lore raw

PATCH 275/292 drm/amdgpu/gfx11: update gpu_clock_counter logic

In reply to Greg Kroah-Hartman

From: Alex Deucher <[email protected]>

commit d5aa417808cf14c052ca042920b3c6b9f1dc6aa4 upstream.

This code was written prior to previous updates to this logic for other chips. The RSC registers are part of SMUIO which is an always on block so there is no need to disable gfxoff. Additionally add the carryover and preemption checks.

v2: rebase

Cc [email protected] # 6.1.y: 5591a051b86b: drm/amdgpu: refine get gpu clock counter method
Cc [email protected] # 6.2.y: 5591a051b86b: drm/amdgpu: refine get gpu clock counter method
Cc [email protected] # 6.3.y: 5591a051b86b: drm/amdgpu: refine get gpu clock counter method
---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c |   11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)
drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +7 -4
@@ -4643,24 +4643,27 @@ static uint64_t gfx_v11_0_get_gpu_clock_
 	uint64_t clock;
 	uint64_t clock_counter_lo, clock_counter_hi_pre, clock_counter_hi_after;
 
-	amdgpu_gfx_off_ctrl(adev, false);
-	mutex_lock(&adev->gfx.gpu_clock_mutex);
 	if (amdgpu_sriov_vf(adev)) {
+		amdgpu_gfx_off_ctrl(adev, false);
+		mutex_lock(&adev->gfx.gpu_clock_mutex);
 		clock_counter_hi_pre = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_HI);
 		clock_counter_lo = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_LO);
 		clock_counter_hi_after = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_HI);
 		if (clock_counter_hi_pre != clock_counter_hi_after)
 			clock_counter_lo = (uint64_t)RREG32_SOC15(GC, 0, regCP_MES_MTIME_LO);
+		mutex_unlock(&adev->gfx.gpu_clock_mutex);
+		amdgpu_gfx_off_ctrl(adev, true);
 	} else {
+		preempt_disable();
 		clock_counter_hi_pre = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_UPPER);
 		clock_counter_lo = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_LOWER);
 		clock_counter_hi_after = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_UPPER);
 		if (clock_counter_hi_pre != clock_counter_hi_after)
 			clock_counter_lo = (uint64_t)RREG32_SOC15(SMUIO, 0, regGOLDEN_TSC_COUNT_LOWER);
+		preempt_enable();
 	}
 	clock = clock_counter_lo | (clock_counter_hi_after << 32ULL);
-	mutex_unlock(&adev->gfx.gpu_clock_mutex);
-	amdgpu_gfx_off_ctrl(adev, true);
+
 	return clock;
 }
 
author maintainer
#276 lore raw

PATCH 276/292 dt-bindings: ata: ahci-ceva: Cover all 4 iommus entries

In reply to Greg Kroah-Hartman

From: Michal Simek <[email protected]>

commit a7844528722619d2f97740ae5ec747afff18c4be upstream.

Current only one entry is enabled but IP itself is using 4 different IDs which are already listed in zynqmp.dtsi.

sata: ahci@fd0c0000 {
	compatible = "ceva,ahci-1v84";
	...
	iommus = <&smmu 0x4c0>, <&smmu 0x4c1>,
		 <&smmu 0x4c2>, <&smmu 0x4c3>;
};
Fixes 8ac47837f0e0 ("arm64: dts: zynqmp: Add missing iommu IDs")
Cc [email protected] # v5.12+
---
 Documentation/devicetree/bindings/ata/ceva,ahci-1v84.yaml |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Documentation/devicetree/bindings/ata/ceva,ahci-1v84.yaml +1 -1
@@ -32,7 +32,7 @@ properties:
     maxItems: 1
 
   iommus:
-    maxItems: 1
+    maxItems: 4
 
   power-domains:
     maxItems: 1
author maintainer
#277 lore raw

PATCH 277/292 powerpc/iommu: DMA address offset is incorrectly calculated with 2MB TCEs

In reply to Greg Kroah-Hartman

From: Gaurav Batra <[email protected]>

commit 096339ab84f36beae0b1db25e0ce63fb3873e8b2 upstream.

When DMA window is backed by 2MB TCEs, the DMA address for the mapped page should be the offset of the page relative to the 2MB TCE. The code was incorrectly setting the DMA address to the beginning of the TCE range.

Mellanox driver is reporting timeout trying to ENABLE_HCA for an SR-IOV ethernet port, when DMA window is backed by 2MB TCEs.

Fixes 387273118714 ("powerps/pseries/dma: Add support for 2M IOMMU page size")
Cc [email protected] # v5.16+
---
 arch/powerpc/kernel/iommu.c |   11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)
arch/powerpc/kernel/iommu.c +7 -4
@@ -517,7 +517,7 @@ int ppc_iommu_map_sg(struct device *dev,
 		/* Convert entry to a dma_addr_t */
 		entry += tbl->it_offset;
 		dma_addr = entry << tbl->it_page_shift;
-		dma_addr |= (s->offset & ~IOMMU_PAGE_MASK(tbl));
+		dma_addr |= (vaddr & ~IOMMU_PAGE_MASK(tbl));
 
 		DBG("  - %lu pages, entry: %lx, dma_addr: %lx\n",
 			    npages, entry, dma_addr);
@@ -904,6 +904,7 @@ void *iommu_alloc_coherent(struct device
 	unsigned int order;
 	unsigned int nio_pages, io_order;
 	struct page *page;
+	int tcesize = (1 << tbl->it_page_shift);
 
 	size = PAGE_ALIGN(size);
 	order = get_order(size);
@@ -930,7 +931,8 @@ void *iommu_alloc_coherent(struct device
 	memset(ret, 0, size);
 
 	/* Set up tces to cover the allocated range */
-	nio_pages = size >> tbl->it_page_shift;
+	nio_pages = IOMMU_PAGE_ALIGN(size, tbl) >> tbl->it_page_shift;
+
 	io_order = get_iommu_order(size, tbl);
 	mapping = iommu_alloc(dev, tbl, ret, nio_pages, DMA_BIDIRECTIONAL,
 			      mask >> tbl->it_page_shift, io_order, 0);
@@ -938,7 +940,8 @@ void *iommu_alloc_coherent(struct device
 		free_pages((unsigned long)ret, order);
 		return NULL;
 	}
-	*dma_handle = mapping;
+
+	*dma_handle = mapping | ((u64)ret & (tcesize - 1));
 	return ret;
 }
 
@@ -949,7 +952,7 @@ void iommu_free_coherent(struct iommu_ta
 		unsigned int nio_pages;
 
 		size = PAGE_ALIGN(size);
-		nio_pages = size >> tbl->it_page_shift;
+		nio_pages = IOMMU_PAGE_ALIGN(size, tbl) >> tbl->it_page_shift;
 		iommu_free(tbl, dma_handle, nio_pages);
 		size = PAGE_ALIGN(size);
 		free_pages((unsigned long)vaddr, get_order(size));
author maintainer
#278 lore raw

PATCH 278/292 powerpc/iommu: Incorrect DDW Table is referenced for SR-IOV device

In reply to Greg Kroah-Hartman

From: Gaurav Batra <[email protected]>

commit 1f7aacc5eb9ed2cc17be7a90da5cd559effb9d59 upstream.

For an SR-IOV device, while enabling DDW, a new table is created and added at index 1 in the group. In the below 2 scenarios, the table is incorrectly referenced at index 0 (which is where the table is for default DMA window).

1. When adding DDW

This issue is exposed with "slub_debug". Error thrown out from dma_iommu_dma_supported()

Warning: IOMMU offset too big for device mask
mask: 0xffffffff, table offset: 0x800000000000000

2. During Dynamic removal of the PCI device.

Error is from iommu_tce_table_put() since a NULL table pointer is passed in.

Fixes 381ceda88c4c ("powerpc/pseries/iommu: Make use of DDW for indirect mapping")
Cc [email protected] # v5.15+
---
 arch/powerpc/kernel/dma-iommu.c        |    4 +++-
 arch/powerpc/platforms/pseries/iommu.c |   13 +++++++++----
 2 files changed, 12 insertions(+), 5 deletions(-)
arch/powerpc/kernel/dma-iommu.c +3 -1
@@ -144,7 +144,7 @@ static bool dma_iommu_bypass_supported(s
 /* We support DMA to/from any memory page via the iommu */
 int dma_iommu_dma_supported(struct device *dev, u64 mask)
 {
-	struct iommu_table *tbl = get_iommu_table_base(dev);
+	struct iommu_table *tbl;
 
 	if (dev_is_pci(dev) && dma_iommu_bypass_supported(dev, mask)) {
 		/*
@@ -162,6 +162,8 @@ int dma_iommu_dma_supported(struct devic
 		return 1;
 	}
 
+	tbl = get_iommu_table_base(dev);
+
 	if (!tbl) {
 		dev_err(dev, "Warning: IOMMU dma not supported: mask 0x%08llx, table unavailable\n", mask);
 		return 0;
arch/powerpc/platforms/pseries/iommu.c +9 -4
@@ -85,19 +85,24 @@ static struct iommu_table_group *iommu_p
 static void iommu_pseries_free_group(struct iommu_table_group *table_group,
 		const char *node_name)
 {
-	struct iommu_table *tbl;
-
 	if (!table_group)
 		return;
 
-	tbl = table_group->tables[0];
 #ifdef CONFIG_IOMMU_API
 	if (table_group->group) {
 		iommu_group_put(table_group->group);
 		BUG_ON(table_group->group);
 	}
 #endif
-	iommu_tce_table_put(tbl);
+
+	/* Default DMA window table is at index 0, while DDW at 1. SR-IOV
+	 * adapters only have table on index 1.
+	 */
+	if (table_group->tables[0])
+		iommu_tce_table_put(table_group->tables[0]);
+
+	if (table_group->tables[1])
+		iommu_tce_table_put(table_group->tables[1]);
 
 	kfree(table_group);
 }
author maintainer
#279 lore raw

PATCH 279/292 tpm/tpm_tis: Disable interrupts for more Lenovo devices

In reply to Greg Kroah-Hartman

From: Jerry Snitselaar <[email protected]>

commit e7d3e5c4b1dd50a70b31524c3228c62bb41bbab2 upstream.

The P360 Tiny suffers from an irq storm issue like the T490s, so add an entry for it to tpm_tis_dmi_table, and force polling. There also previously was a report from the previous attempt to enable interrupts that involved a ThinkPad L490. So an entry is added for it as well.

---
 drivers/char/tpm/tpm_tis.c |   16 ++++++++++++++++
 1 file changed, 16 insertions(+)
drivers/char/tpm/tpm_tis.c +16 -0
@@ -83,6 +83,22 @@ static const struct dmi_system_id tpm_ti
 			DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkPad T490s"),
 		},
 	},
+	{
+		.callback = tpm_tis_disable_irq,
+		.ident = "ThinkStation P360 Tiny",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkStation P360 Tiny"),
+		},
+	},
+	{
+		.callback = tpm_tis_disable_irq,
+		.ident = "ThinkPad L490",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_VERSION, "ThinkPad L490"),
+		},
+	},
 	{}
 };
 
author maintainer
#280 lore raw

PATCH 280/292 powerpc/64s/radix: Fix soft dirty tracking

In reply to Greg Kroah-Hartman

From: Michael Ellerman <[email protected]>

commit 66b2ca086210732954a7790d63d35542936fc664 upstream.

It was reported that soft dirty tracking doesn't work when using the Radix MMU.

The tracking is supposed to work by clearing the soft dirty bit for a mapping and then write protecting the PTE. If/when the page is written to, a page fault occurs and the soft dirty bit is added back via pte_mkdirty(). For example in wp_page_reuse():

	entry = maybe_mkwrite(pte_mkdirty(entry), vma);
	if (ptep_set_access_flags(vma, vmf->address, vmf->pte, entry, 1))
		update_mmu_cache(vma, vmf->address, vmf->pte);

Unfortunately on radix _PAGE_SOFTDIRTY is being dropped by radix__ptep_set_access_flags(), called from ptep_set_access_flags(), meaning the soft dirty bit is not set even though the page has been written to.

Fix it by adding _PAGE_SOFTDIRTY to the set of bits that are able to be changed in radix__ptep_set_access_flags().

---
 arch/powerpc/mm/book3s64/radix_pgtable.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/powerpc/mm/book3s64/radix_pgtable.c +2 -2
@@ -1040,8 +1040,8 @@ void radix__ptep_set_access_flags(struct
 				  pte_t entry, unsigned long address, int psize)
 {
 	struct mm_struct *mm = vma->vm_mm;
-	unsigned long set = pte_val(entry) & (_PAGE_DIRTY | _PAGE_ACCESSED |
-					      _PAGE_RW | _PAGE_EXEC);
+	unsigned long set = pte_val(entry) & (_PAGE_DIRTY | _PAGE_SOFT_DIRTY |
+					      _PAGE_ACCESSED | _PAGE_RW | _PAGE_EXEC);
 
 	unsigned long change = pte_val(entry) ^ pte_val(*ptep);
 	/*
author maintainer
#281 lore raw

PATCH 281/292 nilfs2: fix use-after-free bug of nilfs_root in nilfs_evict_inode()

In reply to Greg Kroah-Hartman

From: Ryusuke Konishi <[email protected]>

commit 9b5a04ac3ad9898c4745cba46ea26de74ba56a8e upstream.

During unmount process of nilfs2, nothing holds nilfs_root structure after nilfs2 detaches its writer in nilfs_detach_log_writer(). However, since nilfs_evict_inode() uses nilfs_root for some cleanup operations, it may cause use-after-free read if inodes are left in "garbage_list" and released by nilfs_dispose_list() at the end of nilfs_detach_log_writer().

Fix this issue by modifying nilfs_evict_inode() to only clear inode without additional metadata changes that use nilfs_root if the file system is degraded to read-only or the writer is detached.

---
 fs/nilfs2/inode.c |   18 ++++++++++++++++++
 1 file changed, 18 insertions(+)
fs/nilfs2/inode.c +18 -0
@@ -917,6 +917,7 @@ void nilfs_evict_inode(struct inode *ino
 	struct nilfs_transaction_info ti;
 	struct super_block *sb = inode->i_sb;
 	struct nilfs_inode_info *ii = NILFS_I(inode);
+	struct the_nilfs *nilfs;
 	int ret;
 
 	if (inode->i_nlink || !ii->i_root || unlikely(is_bad_inode(inode))) {
@@ -929,6 +930,23 @@ void nilfs_evict_inode(struct inode *ino
 
 	truncate_inode_pages_final(&inode->i_data);
 
+	nilfs = sb->s_fs_info;
+	if (unlikely(sb_rdonly(sb) || !nilfs->ns_writer)) {
+		/*
+		 * If this inode is about to be disposed after the file system
+		 * has been degraded to read-only due to file system corruption
+		 * or after the writer has been detached, do not make any
+		 * changes that cause writes, just clear it.
+		 * Do this check after read-locking ns_segctor_sem by
+		 * nilfs_transaction_begin() in order to avoid a race with
+		 * the writer detach operation.
+		 */
+		clear_inode(inode);
+		nilfs_clear_inode(inode);
+		nilfs_transaction_abort(sb);
+		return;
+	}
+
 	/* TODO: some of the following operations may fail.  */
 	nilfs_truncate_bmap(ii, 0);
 	nilfs_mark_inode_dirty(inode);
author maintainer
#282 lore raw

PATCH 282/292 s390/dasd: fix command reject error on ESE devices

In reply to Greg Kroah-Hartman

From: Stefan Haberland <[email protected]>

commit c99bff34290f1b994073557b754aff86e4c7b22e upstream.

Formatting a thin-provisioned (ESE) device that is part of a PPRC copy relation might fail with the following error:

dasd-eckd 0.0.f500: An error occurred in the DASD device driver, reason=09 [...]
24 Byte: 0 MSG 4, no MSGb to SYSOP

During format of an ESE disk the Release Allocated Space command is used. A bit in the payload of the command is set that is not allowed to be set for devices in a copy relation. This bit is set to allow the partial release of an extent.

Check for the existence of a copy relation before setting the respective bit.

---
 drivers/s390/block/dasd_eckd.c |   33 +++++++++++++++++++++++++++++++--
 1 file changed, 31 insertions(+), 2 deletions(-)
drivers/s390/block/dasd_eckd.c +31 -2
@@ -127,6 +127,8 @@ static int prepare_itcw(struct itcw *, u
 			struct dasd_device *, struct dasd_device *,
 			unsigned int, int, unsigned int, unsigned int,
 			unsigned int, unsigned int);
+static int dasd_eckd_query_pprc_status(struct dasd_device *,
+				       struct dasd_pprc_data_sc4 *);
 
 /* initial attempt at a probe function. this can be simplified once
  * the other detection code is gone */
@@ -3732,6 +3734,26 @@ static int count_exts(unsigned int from,
 	return count;
 }
 
+static int dasd_in_copy_relation(struct dasd_device *device)
+{
+	struct dasd_pprc_data_sc4 *temp;
+	int rc;
+
+	if (!dasd_eckd_pprc_enabled(device))
+		return 0;
+
+	temp = kzalloc(sizeof(*temp), GFP_KERNEL);
+	if (!temp)
+		return -ENOMEM;
+
+	rc = dasd_eckd_query_pprc_status(device, temp);
+	if (!rc)
+		rc = temp->dev_info[0].state;
+
+	kfree(temp);
+	return rc;
+}
+
 /*
  * Release allocated space for a given range or an entire volume.
  */
@@ -3748,6 +3770,7 @@ dasd_eckd_dso_ras(struct dasd_device *de
 	int cur_to_trk, cur_from_trk;
 	struct dasd_ccw_req *cqr;
 	u32 beg_cyl, end_cyl;
+	int copy_relation;
 	struct ccw1 *ccw;
 	int trks_per_ext;
 	size_t ras_size;
@@ -3759,6 +3782,10 @@ dasd_eckd_dso_ras(struct dasd_device *de
 	if (dasd_eckd_ras_sanity_checks(device, first_trk, last_trk))
 		return ERR_PTR(-EINVAL);
 
+	copy_relation = dasd_in_copy_relation(device);
+	if (copy_relation < 0)
+		return ERR_PTR(copy_relation);
+
 	rq = req ? blk_mq_rq_to_pdu(req) : NULL;
 
 	features = &private->features;
@@ -3787,9 +3814,11 @@ dasd_eckd_dso_ras(struct dasd_device *de
 	/*
 	 * This bit guarantees initialisation of tracks within an extent that is
 	 * not fully specified, but is only supported with a certain feature
-	 * subset.
+	 * subset and for devices not in a copy relation.
 	 */
-	ras_data->op_flags.guarantee_init = !!(features->feature[56] & 0x01);
+	if (features->feature[56] & 0x01 && !copy_relation)
+		ras_data->op_flags.guarantee_init = 1;
+
 	ras_data->lss = private->conf.ned->ID;
 	ras_data->dev_addr = private->conf.ned->unit_addr;
 	ras_data->nr_exts = nr_exts;
author maintainer
#283 lore raw

PATCH 283/292 s390/crypto: use vector instructions only if available for ChaCha20

In reply to Greg Kroah-Hartman

From: Heiko Carstens <[email protected]>

commit 8703dd6b238da0ec6c276e53836f8200983d3d9b upstream.

Commit 349d03ffd5f6 ("crypto: s390 - add crypto library interface for ChaCha20") added a library interface to the s390 specific ChaCha20 implementation. However no check was added to verify if the required facilities are installed before branching into the assembler code.

If compiled into the kernel, this will lead to the following crash, if vector instructions are not available:

data exception: 0007 ilc:3 [#1] SMP

Modules linked in:
CPU: 0 PID: 1 Comm: swapper/0 Not tainted 6.3.0-rc7+ #11
Hardware name: IBM 3931 A01 704 (KVM/Linux)
Krnl PSW : 0704e00180000000 000000001857277a (chacha20_vx+0x32/0x818)
           R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:2 PM:0 RI:0 EA:3
Krnl GPRS: 0000037f0000000a ffffffffffffff60 000000008184b000 0000000019f5c8e6
           0000000000000109 0000037fffb13c58 0000037fffb13c78 0000000019bb1780
           0000037fffb13c58 0000000019f5c8e6 000000008184b000 0000000000000109
           00000000802d8000 0000000000000109 0000000018571ebc 0000037fffb13718
Krnl Code: 000000001857276a: c07000b1f80b        larl    %r7,0000000019bb1780
           0000000018572770: a708000a            lhi     %r0,10
          #0000000018572774: e78950000c36        vlm     %v24,%v25,0(%r5),0
          >000000001857277a: e7a060000806        vl      %v26,0(%r6),0
           0000000018572780: e7bf70004c36        vlm     %v27,%v31,0(%r7),4
           0000000018572786: e70b00000456        vlr     %v0,%v27
           000000001857278c: e71800000456        vlr     %v1,%v24
           0000000018572792: e74b00000456        vlr     %v4,%v27
Call Trace:
 [<000000001857277a>] chacha20_vx+0x32/0x818
Last Breaking-Event-Address:
 [<0000000018571eb6>] chacha20_crypt_s390.constprop.0+0x6e/0xd8
---[ end trace 0000000000000000 ]---
Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b

Fix this by adding a missing MACHINE_HAS_VX check.

Fixes 349d03ffd5f6 ("crypto: s390 - add crypto library interface for ChaCha20")
Cc <[email protected]> # 5.19+
[[email protected]: remove duplicates in commit message]
---
 arch/s390/crypto/chacha-glue.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/s390/crypto/chacha-glue.c +1 -1
@@ -82,7 +82,7 @@ void chacha_crypt_arch(u32 *state, u8 *d
 	 * it cannot handle a block of data or less, but otherwise
 	 * it can handle data of arbitrary size
 	 */
-	if (bytes <= CHACHA_BLOCK_SIZE || nrounds != 20)
+	if (bytes <= CHACHA_BLOCK_SIZE || nrounds != 20 || !MACHINE_HAS_VX)
 		chacha_crypt_generic(state, dst, src, bytes, nrounds);
 	else
 		chacha20_crypt_s390(state, dst, src, bytes,
author maintainer
#284 lore raw

PATCH 284/292 s390/qdio: fix do_sqbs() inline assembly constraint

In reply to Greg Kroah-Hartman

From: Heiko Carstens <[email protected]>

commit 2862a2fdfae875888e3c1c3634e3422e01d98147 upstream.

Use "a" constraint instead of "d" constraint to pass the state parameter to the do_sqbs() inline assembly. This prevents that general purpose register zero is used for the state parameter.

If the compiler would select general purpose register zero this would be problematic for the used instruction in rsy format: the register used for the state parameter is a base register. If the base register is general purpose register zero the contents of the register are unexpectedly ignored when the instruction is executed.

This only applies to z/VM guests using QIOASSIST with dedicated (pass through) QDIO-based devices such as FCP [zfcp driver] as well as real OSA or HiperSockets [qeth driver].

A possible symptom for this case using zfcp is the following repeating kernel message pattern:

zfcp <devbusid>: A QDIO problem occurred
zfcp <devbusid>: A QDIO problem occurred
zfcp <devbusid>: qdio: ZFCP on SC <sc> using AI:1 QEBSM:1 PRI:1 TDD:1 SIGA: W zfcp <devbusid>: A QDIO problem occurred
zfcp <devbusid>: A QDIO problem occurred

Each of the qdio problem message can be accompanied by the following entries for the affected subchannel <sc> in /sys/kernel/debug/s390dbf/qdio_error/hex_ascii for zfcp or qeth:

<sc> ccq: 69....
<sc> SQBS ERROR.

Cc Steffen Maier <[email protected]>
Fixes 8129ee164267 ("[PATCH] s390: qdio V=V pass-through")
---
 drivers/s390/cio/qdio.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/s390/cio/qdio.h +1 -1
@@ -95,7 +95,7 @@ static inline int do_sqbs(u64 token, uns
 		"	lgr	1,%[token]\n"
 		"	.insn	rsy,0xeb000000008a,%[qs],%[ccq],0(%[state])"
 		: [ccq] "+&d" (_ccq), [qs] "+&d" (_queuestart)
-		: [state] "d" ((unsigned long)state), [token] "d" (token)
+		: [state] "a" ((unsigned long)state), [token] "d" (token)
 		: "memory", "cc", "1");
 	*count = _ccq & 0xff;
 	*start = _queuestart & 0xff;