Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

Cross-merge networking fixes after downstream PR (net-6.12-rc4).

Conflicts:

107a034d5c1e ("net/mlx5: qos: Store rate groups in a qos domain")
1da9cfd6c41c ("net/mlx5: Unregister notifier on eswitch init failure")

Signed-off-by: Paolo Abeni <[e-mail]>

.mailmap

···
 Fenglin Wu <[e-mail]> <[e-mail]>
 Filipe Lautert <[e-mail]>
 Finn Thain <[e-mail]> <[e-mail]>
+Fiona Behrens <[e-mail]>
+Fiona Behrens <[e-mail]> <[e-mail]>
+Fiona Behrens <[e-mail]> <[e-mail]>
 Franck Bui-Huu <[e-mail]>
 Frank Rowand <[e-mail]> <[e-mail]>
 Frank Rowand <[e-mail]> <[e-mail]>

Documentation/devicetree/bindings/display/elgin,jg10309-01.yaml

···
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/elgin,jg10309-01.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Elgin JG10309-01 SPI-controlled display
+
+maintainers:
+  - Fabio Estevam <[e-mail]>
+
+description: |
+  The Elgin JG10309-01 SPI-controlled display is used on the RV1108-Elgin-r1
+  board and is a custom display.
+
+allOf:
+  - $ref: /schemas/spi/spi-peripheral-props.yaml#
+
+properties:
+  compatible:
+    const: elgin,jg10309-01
+
+  reg:
+    maxItems: 1
+
+  spi-max-frequency:
+    maximum: 24000000
+
+  spi-cpha: true
+
+  spi-cpol: true
+
+required:
+  - compatible
+  - reg
+  - spi-cpha
+  - spi-cpol
+
+additionalProperties: false
+
+examples:
+  - |
+    spi {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        display@0 {
+            compatible = "elgin,jg10309-01";
+            reg = <0>;
+            spi-max-frequency = <24000000>;
+            spi-cpha;
+            spi-cpol;
+        };
+    };

Documentation/devicetree/bindings/interrupt-controller/fsl,ls-extirq.yaml

···
             enum:
               - fsl,ls1043a-extirq
               - fsl,ls1046a-extirq
-              - fsl,ls1088a-extirq
-              - fsl,ls2080a-extirq
-              - fsl,lx2160a-extirq
     then:
       properties:
         interrupt-map:
           minItems: 12
           maxItems: 12
+        interrupt-map-mask:
+          items:
+            - const: 0xf
+            - const: 0
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - fsl,ls1088a-extirq
+              - fsl,ls2080a-extirq
+              - fsl,lx2160a-extirq
+# The driver(drivers/irqchip/irq-ls-extirq.c) have not use standard DT
+# function to parser interrupt-map. So it doesn't consider '#address-size'
+# in parent interrupt controller, such as GIC.
+#
+# When dt-binding verify interrupt-map, item data matrix is spitted at
+# incorrect position. Remove interrupt-map restriction because it always
+# wrong.
+
+    then:
+      properties:
         interrupt-map-mask:
           items:
             - const: 0xf

Documentation/devicetree/bindings/misc/fsl,qoriq-mc.yaml

···
 
   msi-parent:
     deprecated: true
-    $ref: /schemas/types.yaml#/definitions/phandle
+    maxItems: 1
     description:
       Describes the MSI controller node handling message
       interrupts for the MC. When there is no translation

Documentation/devicetree/bindings/net/brcm,unimac-mdio.yaml

···
       - brcm,asp-v2.1-mdio
       - brcm,asp-v2.2-mdio
       - brcm,unimac-mdio
+      - brcm,bcm6846-mdio
 
   reg:
     minItems: 1

Documentation/devicetree/bindings/trivial-devices.yaml

···
           - domintech,dmard09
             # DMARD10: 3-axis Accelerometer
           - domintech,dmard10
-            # Elgin SPI-controlled LCD
-          - elgin,jg10309-01
             # MMA7660FC: 3-Axis Orientation/Motion Detection Sensor
           - fsl,mma7660
             # MMA8450Q: Xtrinsic Low-power, 3-axis Xtrinsic Accelerometer

MAINTAINERS

···
 M:	Viresh Kumar <[e-mail]>
 L:	[e-mail]
 S:	Maintained
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git
 F:	drivers/ata/pata_arasan_cf.c
 F:	include/linux/pata_arasan_cf_data.h
 
···
 M:	Linus Walleij <[e-mail]>
 L:	[e-mail]
 S:	Maintained
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git
 F:	drivers/ata/pata_ftide010.c
 F:	drivers/ata/sata_gemini.c
 F:	drivers/ata/sata_gemini.h
 
 LIBATA SATA AHCI PLATFORM devices support
 M:	Hans de Goede <[e-mail]>
-M:	Jens Axboe <[e-mail]>
 L:	[e-mail]
 S:	Maintained
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git
 F:	drivers/ata/ahci_platform.c
 F:	drivers/ata/libahci_platform.c
 F:	include/linux/ahci_platform.h
···
 M:	Serge Semin <[e-mail]>
 L:	[e-mail]
 S:	Maintained
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/libata.git
 F:	Documentation/devicetree/bindings/ata/baikal,bt1-ahci.yaml
 F:	Documentation/devicetree/bindings/ata/snps,dwc-ahci.yaml
 F:	drivers/ata/ahci_dwc.c
···
 M:	Mikael Pettersson <[e-mail]>
 L:	[e-mail]
 S:	Maintained
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git
 F:	drivers/ata/sata_promise.*
 
 LIBATA SUBSYSTEM (Serial and Parallel ATA drivers)
···
 F:	net/core/drop_monitor.c
 
 NETWORKING DRIVERS
+M:	Andrew Lunn <[e-mail]>
 M:	"David S. Miller" <[e-mail]>
 M:	Eric Dumazet <[e-mail]>
 M:	Jakub Kicinski <[e-mail]>

Makefile

···
 VERSION = 6
 PATCHLEVEL = 12
 SUBLEVEL = 0
-EXTRAVERSION = -rc2
+EXTRAVERSION = -rc3
 NAME = Baby Opossum Posse
 
 # *DOCUMENTATION*

arch/powerpc/kernel/head_8xx.S

···
 	bctr				/* jump into table */
 152:
 	mfdar	r11
+	mtdar	r10
 	mtctr	r11			/* restore ctr reg from DAR */
 	mfspr	r11, SPRN_SPRG_THREAD
 	stw	r10, DAR(r11)

arch/x86/xen/enlighten_pv.c

···
 	switch (msr) {
 	case MSR_IA32_APICBASE:
 		val &= ~X2APIC_ENABLE;
+		if (smp_processor_id() == 0)
+			val |= MSR_IA32_APICBASE_BSP;
+		else
+			val &= ~MSR_IA32_APICBASE_BSP;
 		break;
 	}
 	return val;

crypto/algapi.c

···
 	q->cra_flags |= CRYPTO_ALG_DEAD;
 	alg = test->adult;
 
-	if (list_empty(&alg->cra_list))
+	if (crypto_is_dead(alg))
 		goto complete;
 
 	if (err == -ECANCELED)

crypto/testmgr.c

···
 	atfm = crypto_alloc_ahash(driver, type, mask);
 	if (IS_ERR(atfm)) {
 		if (PTR_ERR(atfm) == -ENOENT)
-			return -ENOENT;
+			return 0;
 		pr_err("alg: hash: failed to allocate transform for %s: %ld\n",
 		       driver, PTR_ERR(atfm));
 		return PTR_ERR(atfm);
···
 	tfm = crypto_alloc_aead(driver, type, mask);
 	if (IS_ERR(tfm)) {
 		if (PTR_ERR(tfm) == -ENOENT)
-			return -ENOENT;
+			return 0;
 		pr_err("alg: aead: failed to allocate transform for %s: %ld\n",
 		       driver, PTR_ERR(tfm));
 		return PTR_ERR(tfm);
···
 	tfm = crypto_alloc_skcipher(driver, type, mask);
 	if (IS_ERR(tfm)) {
 		if (PTR_ERR(tfm) == -ENOENT)
-			return -ENOENT;
+			return 0;
 		pr_err("alg: skcipher: failed to allocate transform for %s: %ld\n",
 		       driver, PTR_ERR(tfm));
 		return PTR_ERR(tfm);
···
 	tfm = crypto_alloc_cipher(driver, type, mask);
 	if (IS_ERR(tfm)) {
 		if (PTR_ERR(tfm) == -ENOENT)
-			return -ENOENT;
+			return 0;
 		printk(KERN_ERR "alg: cipher: Failed to load transform for "
 		       "%s: %ld\n", driver, PTR_ERR(tfm));
 		return PTR_ERR(tfm);
···
 		acomp = crypto_alloc_acomp(driver, type, mask);
 		if (IS_ERR(acomp)) {
 			if (PTR_ERR(acomp) == -ENOENT)
-				return -ENOENT;
+				return 0;
 			pr_err("alg: acomp: Failed to load transform for %s: %ld\n",
 			       driver, PTR_ERR(acomp));
 			return PTR_ERR(acomp);
···
 		comp = crypto_alloc_comp(driver, type, mask);
 		if (IS_ERR(comp)) {
 			if (PTR_ERR(comp) == -ENOENT)
-				return -ENOENT;
+				return 0;
 			pr_err("alg: comp: Failed to load transform for %s: %ld\n",
 			       driver, PTR_ERR(comp));
 			return PTR_ERR(comp);
···
 	rng = crypto_alloc_rng(driver, type, mask);
 	if (IS_ERR(rng)) {
 		if (PTR_ERR(rng) == -ENOENT)
-			return -ENOENT;
+			return 0;
 		printk(KERN_ERR "alg: cprng: Failed to load transform for %s: "
 		       "%ld\n", driver, PTR_ERR(rng));
 		return PTR_ERR(rng);
···
 
 	drng = crypto_alloc_rng(driver, type, mask);
 	if (IS_ERR(drng)) {
+		kfree_sensitive(buf);
 		if (PTR_ERR(drng) == -ENOENT)
-			goto out_no_rng;
+			return 0;
 		printk(KERN_ERR "alg: drbg: could not allocate DRNG handle for "
 		       "%s\n", driver);
-out_no_rng:
-		kfree_sensitive(buf);
 		return PTR_ERR(drng);
 	}
 
···
 	tfm = crypto_alloc_kpp(driver, type, mask);
 	if (IS_ERR(tfm)) {
 		if (PTR_ERR(tfm) == -ENOENT)
-			return -ENOENT;
+			return 0;
 		pr_err("alg: kpp: Failed to load tfm for %s: %ld\n",
 		       driver, PTR_ERR(tfm));
 		return PTR_ERR(tfm);
···
 	tfm = crypto_alloc_akcipher(driver, type, mask);
 	if (IS_ERR(tfm)) {
 		if (PTR_ERR(tfm) == -ENOENT)
-			return -ENOENT;
+			return 0;
 		pr_err("alg: akcipher: Failed to load tfm for %s: %ld\n",
 		       driver, PTR_ERR(tfm));
 		return PTR_ERR(tfm);

drivers/acpi/resource.c

···
 		},
 	},
 	{
-		/* Asus ExpertBook B1402CBA */
+		/* Asus ExpertBook B1402C* */
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-			DMI_MATCH(DMI_BOARD_NAME, "B1402CBA"),
+			DMI_MATCH(DMI_BOARD_NAME, "B1402C"),
 		},
 	},
 	{
-		/* Asus ExpertBook B1402CVA */
+		/* Asus ExpertBook B1502C* */
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-			DMI_MATCH(DMI_BOARD_NAME, "B1402CVA"),
+			DMI_MATCH(DMI_BOARD_NAME, "B1502C"),
 		},
 	},
 	{
-		/* Asus ExpertBook B1502CBA */
+		/* Asus ExpertBook B2402 (B2402CBA / B2402FBA / B2402CVA / B2402FVA) */
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-			DMI_MATCH(DMI_BOARD_NAME, "B1502CBA"),
+			DMI_MATCH(DMI_BOARD_NAME, "B2402"),
 		},
 	},
 	{
-		/* Asus ExpertBook B1502CGA */
+		/* Asus ExpertBook B2502 (B2502CBA / B2502FBA / B2502CVA / B2502FVA) */
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-			DMI_MATCH(DMI_BOARD_NAME, "B1502CGA"),
-		},
-	},
-        {
-                /* Asus ExpertBook B1502CVA */
-                .matches = {
-                        DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-                        DMI_MATCH(DMI_BOARD_NAME, "B1502CVA"),
-                },
-        },
-	{
-		/* Asus ExpertBook B2402CBA */
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-			DMI_MATCH(DMI_BOARD_NAME, "B2402CBA"),
-		},
-	},
-	{
-		/* Asus ExpertBook B2402FBA */
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-			DMI_MATCH(DMI_BOARD_NAME, "B2402FBA"),
-		},
-	},
-	{
-		/* Asus ExpertBook B2502 */
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-			DMI_MATCH(DMI_BOARD_NAME, "B2502CBA"),
-		},
-	},
-	{
-		/* Asus ExpertBook B2502FBA */
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-			DMI_MATCH(DMI_BOARD_NAME, "B2502FBA"),
-		},
-	},
-	{
-		/* Asus ExpertBook B2502CVA */
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-			DMI_MATCH(DMI_BOARD_NAME, "B2502CVA"),
+			DMI_MATCH(DMI_BOARD_NAME, "B2502"),
 		},
 	},
 	{
···
 		},
 	},
 	{
-		/* Asus Vivobook Pro N6506MV */
+		/* Asus Vivobook Pro N6506M* */
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-			DMI_MATCH(DMI_BOARD_NAME, "N6506MV"),
-		},
-	},
-	{
-		/* Asus Vivobook Pro N6506MU */
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-			DMI_MATCH(DMI_BOARD_NAME, "N6506MU"),
-		},
-	},
-	{
-		/* Asus Vivobook Pro N6506MJ */
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-			DMI_MATCH(DMI_BOARD_NAME, "N6506MJ"),
+			DMI_MATCH(DMI_BOARD_NAME, "N6506M"),
 		},
 	},
 	{

drivers/ata/libata-eh.c

···
 
 	WARN_ON(ap->pflags & ATA_PFLAG_SUSPENDED);
 
-	/* Set all devices attached to the port in standby mode */
-	ata_for_each_link(link, ap, HOST_FIRST) {
-		ata_for_each_dev(dev, link, ENABLED)
-			ata_dev_power_set_standby(dev);
+	/*
+	 * We will reach this point for all of the PM events:
+	 * PM_EVENT_SUSPEND (if runtime pm, PM_EVENT_AUTO will also be set)
+	 * PM_EVENT_FREEZE, and PM_EVENT_HIBERNATE.
+	 *
+	 * We do not want to perform disk spin down for PM_EVENT_FREEZE.
+	 * (Spin down will be performed by the subsequent PM_EVENT_HIBERNATE.)
+	 */
+	if (!(ap->pm_mesg.event & PM_EVENT_FREEZE)) {
+		/* Set all devices attached to the port in standby mode */
+		ata_for_each_link(link, ap, HOST_FIRST) {
+			ata_for_each_dev(dev, link, ENABLED)
+				ata_dev_power_set_standby(dev);
+		}
 	}
 
 	/*

drivers/base/power/common.c

···
 	struct device *pd_dev = NULL;
 	int ret, i, num_pds = 0;
 	bool by_id = true;
+	size_t size;
 	u32 pd_flags = data ? data->pd_flags : 0;
 	u32 link_flags = pd_flags & PD_FLAG_NO_DEV_LINK ? 0 :
 			DL_FLAG_STATELESS | DL_FLAG_PM_RUNTIME;
···
 	if (num_pds <= 0)
 		return 0;
 
-	pds = devm_kzalloc(dev, sizeof(*pds), GFP_KERNEL);
+	pds = kzalloc(sizeof(*pds), GFP_KERNEL);
 	if (!pds)
 		return -ENOMEM;
 
-	pds->pd_devs = devm_kcalloc(dev, num_pds, sizeof(*pds->pd_devs),
-				    GFP_KERNEL);
-	if (!pds->pd_devs)
-		return -ENOMEM;
-
-	pds->pd_links = devm_kcalloc(dev, num_pds, sizeof(*pds->pd_links),
-				     GFP_KERNEL);
-	if (!pds->pd_links)
-		return -ENOMEM;
+	size = sizeof(*pds->pd_devs) + sizeof(*pds->pd_links);
+	pds->pd_devs = kcalloc(num_pds, size, GFP_KERNEL);
+	if (!pds->pd_devs) {
+		ret = -ENOMEM;
+		goto free_pds;
+	}
+	pds->pd_links = (void *)(pds->pd_devs + num_pds);
 
 	if (link_flags && pd_flags & PD_FLAG_DEV_LINK_ON)
 		link_flags |= DL_FLAG_RPM_ACTIVE;
···
 			device_link_del(pds->pd_links[i]);
 		dev_pm_domain_detach(pds->pd_devs[i], true);
 	}
+	kfree(pds->pd_devs);
+free_pds:
+	kfree(pds);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(dev_pm_domain_attach_list);
···
 			device_link_del(list->pd_links[i]);
 		dev_pm_domain_detach(list->pd_devs[i], true);
 	}
+
+	kfree(list->pd_devs);
+	kfree(list);
 }
 EXPORT_SYMBOL_GPL(dev_pm_domain_detach_list);
 

drivers/crypto/marvell/cesa/hash.c

···
 		.base = {
 			.cra_name = "md5",
 			.cra_driver_name = "mv-md5",
-			.cra_priority = 300,
+			.cra_priority = 0,
 			.cra_flags = CRYPTO_ALG_ASYNC |
 				     CRYPTO_ALG_ALLOCATES_MEMORY |
 				     CRYPTO_ALG_KERN_DRIVER_ONLY,
···
 		.base = {
 			.cra_name = "sha1",
 			.cra_driver_name = "mv-sha1",
-			.cra_priority = 300,
+			.cra_priority = 0,
 			.cra_flags = CRYPTO_ALG_ASYNC |
 				     CRYPTO_ALG_ALLOCATES_MEMORY |
 				     CRYPTO_ALG_KERN_DRIVER_ONLY,
···
 		.base = {
 			.cra_name = "sha256",
 			.cra_driver_name = "mv-sha256",
-			.cra_priority = 300,
+			.cra_priority = 0,
 			.cra_flags = CRYPTO_ALG_ASYNC |
 				     CRYPTO_ALG_ALLOCATES_MEMORY |
 				     CRYPTO_ALG_KERN_DRIVER_ONLY,
···
 		.base = {
 			.cra_name = "hmac(md5)",
 			.cra_driver_name = "mv-hmac-md5",
-			.cra_priority = 300,
+			.cra_priority = 0,
 			.cra_flags = CRYPTO_ALG_ASYNC |
 				     CRYPTO_ALG_ALLOCATES_MEMORY |
 				     CRYPTO_ALG_KERN_DRIVER_ONLY,
···
 		.base = {
 			.cra_name = "hmac(sha1)",
 			.cra_driver_name = "mv-hmac-sha1",
-			.cra_priority = 300,
+			.cra_priority = 0,
 			.cra_flags = CRYPTO_ALG_ASYNC |
 				     CRYPTO_ALG_ALLOCATES_MEMORY |
 				     CRYPTO_ALG_KERN_DRIVER_ONLY,
···
 		.base = {
 			.cra_name = "hmac(sha256)",
 			.cra_driver_name = "mv-hmac-sha256",
-			.cra_priority = 300,
+			.cra_priority = 0,
 			.cra_flags = CRYPTO_ALG_ASYNC |
 				     CRYPTO_ALG_ALLOCATES_MEMORY |
 				     CRYPTO_ALG_KERN_DRIVER_ONLY,

drivers/gpio/gpio-aspeed.c

···
 	gpio->dcache[GPIO_BANK(offset)] = reg;
 
 	iowrite32(reg, addr);
+	/* Flush write */
+	ioread32(addr);
 }
 
 static void aspeed_gpio_set(struct gpio_chip *gc, unsigned int offset,
···
 	if (!gpio_id)
 		return -EINVAL;
 
-	gpio->clk = of_clk_get(pdev->dev.of_node, 0);
+	gpio->clk = devm_clk_get_enabled(&pdev->dev, NULL);
 	if (IS_ERR(gpio->clk)) {
 		dev_warn(&pdev->dev,
 				"Failed to get clock from devicetree, debouncing disabled\n");

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c

···
 	list_add_tail(&vm->vm_list_node,
 			&(vm->process_info->vm_list_head));
 	vm->process_info->n_vms++;
-
-	*ef = dma_fence_get(&vm->process_info->eviction_fence->base);
+	if (ef)
+		*ef = dma_fence_get(&vm->process_info->eviction_fence->base);
 	mutex_unlock(&vm->process_info->lock);
 
 	return 0;

drivers/gpu/drm/amd/amdkfd/kfd_process.c

···
 
 	ret = amdgpu_amdkfd_gpuvm_acquire_process_vm(dev->adev, avm,
 						     &p->kgd_process_info,
-						     &ef);
+						     p->ef ? NULL : &ef);
 	if (ret) {
 		dev_err(dev->adev->dev, "Failed to create process VM object\n");
 		return ret;
 	}
-	RCU_INIT_POINTER(p->ef, ef);
+
+	if (!p->ef)
+		RCU_INIT_POINTER(p->ef, ef);
+
 	pdd->drm_priv = drm_file->private_data;
 
 	ret = kfd_process_device_reserve_ib_mem(pdd);

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c

···
 
 	hpd_rx_irq_work_suspend(dm);
 
-	if (adev->dm.dc->caps.ips_support)
-		dc_allow_idle_optimizations(adev->dm.dc, true);
-
 	dc_set_power_state(dm->dc, DC_ACPI_CM_POWER_STATE_D3);
+
+	if (dm->dc->caps.ips_support && adev->in_s0ix)
+		dc_allow_idle_optimizations(dm->dc, true);
+
 	dc_dmub_srv_set_power_state(dm->dc->ctx->dmub_srv, DC_ACPI_CM_POWER_STATE_D3);
 
 	return 0;

drivers/gpu/drm/amd/display/dc/core/dc.c

···
 	return true;
 }
 
+static void clear_update_flags(struct dc_surface_update *srf_updates,
+	int surface_count, struct dc_stream_state *stream)
+{
+	int i;
+
+	if (stream)
+		stream->update_flags.raw = 0;
+
+	for (i = 0; i < surface_count; i++)
+		if (srf_updates[i].surface)
+			srf_updates[i].surface->update_flags.raw = 0;
+}
+
 bool dc_update_planes_and_stream(struct dc *dc,
 		struct dc_surface_update *srf_updates, int surface_count,
 		struct dc_stream_state *stream,
 		struct dc_stream_update *stream_update)
 {
+	bool ret = false;
+
 	dc_exit_ips_for_hw_access(dc);
 	/*
 	 * update planes and stream version 3 separates FULL and FAST updates
···
 	 * features as they are now transparent to the new sequence.
 	 */
 	if (dc->ctx->dce_version >= DCN_VERSION_4_01)
-		return update_planes_and_stream_v3(dc, srf_updates,
+		ret = update_planes_and_stream_v3(dc, srf_updates,
 				surface_count, stream, stream_update);
-	return update_planes_and_stream_v2(dc, srf_updates,
+	else
+		ret = update_planes_and_stream_v2(dc, srf_updates,
 			surface_count, stream, stream_update);
+
+	if (ret)
+		clear_update_flags(srf_updates, surface_count, stream);
+
+	return ret;
 }
 
 void dc_commit_updates_for_stream(struct dc *dc,
···
 		struct dc_stream_update *stream_update,
 		struct dc_state *state)
 {
+	bool ret = false;
+
 	dc_exit_ips_for_hw_access(dc);
 	/* TODO: Since change commit sequence can have a huge impact,
 	 * we decided to only enable it for DCN3x. However, as soon as
···
 	 * the new sequence for all ASICs.
 	 */
 	if (dc->ctx->dce_version >= DCN_VERSION_4_01) {
-		update_planes_and_stream_v3(dc, srf_updates, surface_count,
+		ret = update_planes_and_stream_v3(dc, srf_updates, surface_count,
 				stream, stream_update);
-		return;
-	}
-	if (dc->ctx->dce_version >= DCN_VERSION_3_2) {
-		update_planes_and_stream_v2(dc, srf_updates, surface_count,
+	} else if (dc->ctx->dce_version >= DCN_VERSION_3_2) {
+		ret = update_planes_and_stream_v2(dc, srf_updates, surface_count,
 				stream, stream_update);
-		return;
-	}
-	update_planes_and_stream_v1(dc, srf_updates, surface_count, stream,
-			stream_update, state);
+	} else
+		ret = update_planes_and_stream_v1(dc, srf_updates, surface_count, stream,
+				stream_update, state);
+
+	if (ret)
+		clear_update_flags(srf_updates, surface_count, stream);
 }
 
 uint8_t dc_get_current_stream_count(struct dc *dc)

drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h

···
 
 struct vi_dpm_table {
 	uint32_t count;
-	struct vi_dpm_level dpm_level[] __counted_by(count);
+	struct vi_dpm_level dpm_level[];
 };
 
 #define PCIE_PERF_REQ_REMOVE_REGISTRY   0
···
 
 struct phm_clock_array {
 	uint32_t count;
-	uint32_t values[] __counted_by(count);
+	uint32_t values[];
 };
 
 struct phm_clock_voltage_dependency_record {
···
 
 struct phm_clock_voltage_dependency_table {
 	uint32_t count;
-	struct phm_clock_voltage_dependency_record entries[] __counted_by(count);
+	struct phm_clock_voltage_dependency_record entries[];
 };
 
 struct phm_phase_shedding_limits_record {
···
 
 struct phm_uvd_clock_voltage_dependency_table {
 	uint8_t count;
-	struct phm_uvd_clock_voltage_dependency_record entries[] __counted_by(count);
+	struct phm_uvd_clock_voltage_dependency_record entries[];
 };
 
 struct phm_acp_clock_voltage_dependency_record {
···
 
 struct phm_acp_clock_voltage_dependency_table {
 	uint32_t count;
-	struct phm_acp_clock_voltage_dependency_record entries[] __counted_by(count);
+	struct phm_acp_clock_voltage_dependency_record entries[];
 };
 
 struct phm_vce_clock_voltage_dependency_record {
···
 
 struct phm_phase_shedding_limits_table {
 	uint32_t count;
-	struct phm_phase_shedding_limits_record  entries[] __counted_by(count);
+	struct phm_phase_shedding_limits_record  entries[];
 };
 
 struct phm_vceclock_voltage_dependency_table {
 	uint8_t count;
-	struct phm_vceclock_voltage_dependency_record entries[] __counted_by(count);
+	struct phm_vceclock_voltage_dependency_record entries[];
 };
 
 struct phm_uvdclock_voltage_dependency_table {
 	uint8_t count;
-	struct phm_uvdclock_voltage_dependency_record entries[] __counted_by(count);
+	struct phm_uvdclock_voltage_dependency_record entries[];
 };
 
 struct phm_samuclock_voltage_dependency_table {
 	uint8_t count;
-	struct phm_samuclock_voltage_dependency_record entries[] __counted_by(count);
+	struct phm_samuclock_voltage_dependency_record entries[];
 };
 
 struct phm_acpclock_voltage_dependency_table {
 	uint32_t count;
-	struct phm_acpclock_voltage_dependency_record entries[] __counted_by(count);
+	struct phm_acpclock_voltage_dependency_record entries[];
 };
 
 struct phm_vce_clock_voltage_dependency_table {
 	uint8_t count;
-	struct phm_vce_clock_voltage_dependency_record entries[] __counted_by(count);
+	struct phm_vce_clock_voltage_dependency_record entries[];
 };
 
 
···
 
 struct phm_cac_leakage_table {
 	uint32_t count;
-	union phm_cac_leakage_record entries[] __counted_by(count);
+	union phm_cac_leakage_record entries[];
 };
 
 struct phm_samu_clock_voltage_dependency_record {
···
 
 struct phm_samu_clock_voltage_dependency_table {
 	uint8_t count;
-	struct phm_samu_clock_voltage_dependency_record entries[] __counted_by(count);
+	struct phm_samu_clock_voltage_dependency_record entries[];
 };
 
 struct phm_cac_tdp_table {

drivers/gpu/drm/drm_fbdev_dma.c

···
 	if (!fb_helper->dev)
 		return;
 
-	fb_deferred_io_cleanup(info);
+	if (info->fbdefio)
+		fb_deferred_io_cleanup(info);
 	drm_fb_helper_fini(fb_helper);
 
 	drm_client_buffer_vunmap(fb_helper->buffer);

drivers/gpu/drm/i915/display/intel_hdcp.c

···
 	hdcp->value = value;
 	if (update_property) {
 		drm_connector_get(&connector->base);
-		queue_work(i915->unordered_wq, &hdcp->prop_work);
+		if (!queue_work(i915->unordered_wq, &hdcp->prop_work))
+			drm_connector_put(&connector->base);
 	}
 }
 
···
 		mutex_lock(&hdcp->mutex);
 		hdcp->value = DRM_MODE_CONTENT_PROTECTION_DESIRED;
 		drm_connector_get(&connector->base);
-		queue_work(i915->unordered_wq, &hdcp->prop_work);
+		if (!queue_work(i915->unordered_wq, &hdcp->prop_work))
+			drm_connector_put(&connector->base);
 		mutex_unlock(&hdcp->mutex);
 	}
 
···
 		 */
 		if (!desired_and_not_enabled && !content_protection_type_changed) {
 			drm_connector_get(&connector->base);
-			queue_work(i915->unordered_wq, &hdcp->prop_work);
+			if (!queue_work(i915->unordered_wq, &hdcp->prop_work))
+				drm_connector_put(&connector->base);
+
 		}
 	}
 

drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h

···
 	} *rm;
 
 	struct {
-		struct mutex mutex;;
+		struct mutex mutex;
 		struct idr idr;
 	} client_id;
 

drivers/gpu/drm/nouveau/nouveau_dmem.c

···
 	if (!spage || !(src & MIGRATE_PFN_MIGRATE))
 		goto done;
 
-	dpage = alloc_page_vma(GFP_HIGHUSER, vmf->vma, vmf->address);
+	dpage = alloc_page_vma(GFP_HIGHUSER | __GFP_ZERO, vmf->vma, vmf->address);
 	if (!dpage)
 		goto done;
 

drivers/gpu/drm/nouveau/nouveau_drm.c

···
 		return;
 	}
 
-	ret = nouveau_channel_new(&drm->client, false, runm, NvDmaFB, NvDmaTT, &drm->cechan);
+	ret = nouveau_channel_new(&drm->client, true, runm, NvDmaFB, NvDmaTT, &drm->cechan);
 	if (ret)
 		NV_ERROR(drm, "failed to create ce channel, %d\n", ret);
 }

drivers/gpu/drm/radeon/atombios_dp.c

···
 {
 	struct drm_device *dev = radeon_connector->base.dev;
 	struct radeon_device *rdev = dev->dev_private;
-	int ret;
 
 	radeon_connector->ddc_bus->rec.hpd = radeon_connector->hpd.hpd;
-	radeon_connector->ddc_bus->aux.dev = radeon_connector->base.kdev;
 	radeon_connector->ddc_bus->aux.drm_dev = radeon_connector->base.dev;
 	if (ASIC_IS_DCE5(rdev)) {
 		if (radeon_auxch)
···
 		radeon_connector->ddc_bus->aux.transfer = radeon_dp_aux_transfer_atom;
 	}
 
-	ret = drm_dp_aux_register(&radeon_connector->ddc_bus->aux);
-	if (!ret)
-		radeon_connector->ddc_bus->has_aux = true;
-
-	WARN(ret, "drm_dp_aux_register() failed with error %d\n", ret);
+	drm_dp_aux_init(&radeon_connector->ddc_bus->aux);
+	radeon_connector->ddc_bus->has_aux = true;
 }
 
 /***** general DP utility functions *****/

drivers/gpu/drm/radeon/radeon_connectors.c

···
 	return MODE_OK;
 }
 
+static int
+radeon_connector_late_register(struct drm_connector *connector)
+{
+	struct radeon_connector *radeon_connector = to_radeon_connector(connector);
+	int r = 0;
+
+	if (radeon_connector->ddc_bus->has_aux) {
+		radeon_connector->ddc_bus->aux.dev = radeon_connector->base.kdev;
+		r = drm_dp_aux_register(&radeon_connector->ddc_bus->aux);
+	}
+
+	return r;
+}
+
 static const struct drm_connector_helper_funcs radeon_dp_connector_helper_funcs = {
 	.get_modes = radeon_dp_get_modes,
 	.mode_valid = radeon_dp_mode_valid,
···
 	.early_unregister = radeon_connector_unregister,
 	.destroy = radeon_connector_destroy,
 	.force = radeon_dvi_force,
+	.late_register = radeon_connector_late_register,
 };
 
 static const struct drm_connector_funcs radeon_edp_connector_funcs = {
···
 	.early_unregister = radeon_connector_unregister,
 	.destroy = radeon_connector_destroy,
 	.force = radeon_dvi_force,
+	.late_register = radeon_connector_late_register,
 };
 
 static const struct drm_connector_funcs radeon_lvds_bridge_connector_funcs = {
···
 	.early_unregister = radeon_connector_unregister,
 	.destroy = radeon_connector_destroy,
 	.force = radeon_dvi_force,
+	.late_register = radeon_connector_late_register,
 };
 
 void

drivers/gpu/drm/radeon/radeon_gem.c

···
 int radeon_gem_prime_pin(struct drm_gem_object *obj);
 void radeon_gem_prime_unpin(struct drm_gem_object *obj);
 
-const struct drm_gem_object_funcs radeon_gem_object_funcs;
-
 static vm_fault_t radeon_gem_fault(struct vm_fault *vmf)
 {
 	struct ttm_buffer_object *bo = vmf->vma->vm_private_data;
···
 		return r;
 	}
 	*obj = &robj->tbo.base;
-	(*obj)->funcs = &radeon_gem_object_funcs;
 	robj->pid = task_pid_nr(current);
 
 	mutex_lock(&rdev->gem.mutex);

drivers/gpu/drm/radeon/radeon_object.c

···
 	if (bo == NULL)
 		return -ENOMEM;
 	drm_gem_private_object_init(rdev_to_drm(rdev), &bo->tbo.base, size);
+	bo->tbo.base.funcs = &radeon_gem_object_funcs;
 	bo->rdev = rdev;
 	bo->surface_reg = -1;
 	INIT_LIST_HEAD(&bo->list);

drivers/gpu/drm/scheduler/sched_main.c

···
 #define CREATE_TRACE_POINTS
 #include "gpu_scheduler_trace.h"
 
+#ifdef CONFIG_LOCKDEP
+static struct lockdep_map drm_sched_lockdep_map = {
+	.name = "drm_sched_lockdep_map"
+};
+#endif
+
 #define to_drm_sched_job(sched_job)		\
 		container_of((sched_job), struct drm_sched_job, queue_node)
 
···
 		sched->submit_wq = submit_wq;
 		sched->own_submit_wq = false;
 	} else {
+#ifdef CONFIG_LOCKDEP
+		sched->submit_wq = alloc_ordered_workqueue_lockdep_map(name, 0,
+								       &drm_sched_lockdep_map);
+#else
 		sched->submit_wq = alloc_ordered_workqueue(name, 0);
+#endif
 		if (!sched->submit_wq)
 			return -ENOMEM;
 

drivers/gpu/drm/tegra/gr3d.c

···
 	unsigned int nclocks;
 	struct reset_control_bulk_data resets[RST_GR3D_MAX];
 	unsigned int nresets;
-	struct dev_pm_domain_list *pd_list;
 
 	DECLARE_BITMAP(addr_regs, GR3D_NUM_REGS);
 };
···
 	return 0;
 }
 
+static void gr3d_del_link(void *link)
+{
+	device_link_del(link);
+}
+
 static int gr3d_init_power(struct device *dev, struct gr3d *gr3d)
 {
-	struct dev_pm_domain_attach_data pd_data = {
-		.pd_names = (const char *[]) { "3d0", "3d1" },
-		.num_pd_names = 2,
-	};
+	static const char * const opp_genpd_names[] = { "3d0", "3d1", NULL };
+	const u32 link_flags = DL_FLAG_STATELESS | DL_FLAG_PM_RUNTIME;
+	struct device **opp_virt_devs, *pd_dev;
+	struct device_link *link;
+	unsigned int i;
 	int err;
 
 	err = of_count_phandle_with_args(dev->of_node, "power-domains",
···
 	if (dev->pm_domain)
 		return 0;
 
-	err = dev_pm_domain_attach_list(dev, &pd_data, &gr3d->pd_list);
-	if (err < 0)
+	err = devm_pm_opp_attach_genpd(dev, opp_genpd_names, &opp_virt_devs);
+	if (err)
 		return err;
+
+	for (i = 0; opp_genpd_names[i]; i++) {
+		pd_dev = opp_virt_devs[i];
+		if (!pd_dev) {
+			dev_err(dev, "failed to get %s power domain\n",
+				opp_genpd_names[i]);
+			return -EINVAL;
+		}
+
+		link = device_link_add(dev, pd_dev, link_flags);
+		if (!link) {
+			dev_err(dev, "failed to link to %s\n", dev_name(pd_dev));
+			return -EINVAL;
+		}
+
+		err = devm_add_action_or_reset(dev, gr3d_del_link, link);
+		if (err)
+			return err;
+	}
 
 	return 0;
 }
···
 
 	err = devm_tegra_core_dev_init_opp_table_common(&pdev->dev);
 	if (err)
-		goto err;
+		return err;
 
 	err = host1x_client_register(&gr3d->client.base);
 	if (err < 0) {
 		dev_err(&pdev->dev, "failed to register host1x client: %d\n",
 			err);
-		goto err;
+		return err;
 	}
 
 	/* initialize address register map */
···
 		set_bit(gr3d_addr_regs[i], gr3d->addr_regs);
 
 	return 0;
-err:
-	dev_pm_domain_detach_list(gr3d->pd_list);
-	return err;
 }
 
 static void gr3d_remove(struct platform_device *pdev)
···
 
 	pm_runtime_disable(&pdev->dev);
 	host1x_client_unregister(&gr3d->client.base);
-	dev_pm_domain_detach_list(gr3d->pd_list);
 }
 
 static int __maybe_unused gr3d_runtime_suspend(struct device *dev)

drivers/gpu/drm/v3d/v3d_perfmon.c

···
 static int v3d_perfmon_idr_del(int id, void *elem, void *data)
 {
 	struct v3d_perfmon *perfmon = elem;
+	struct v3d_dev *v3d = (struct v3d_dev *)data;
+
+	/* If the active perfmon is being destroyed, stop it first */
+	if (perfmon == v3d->active_perfmon)
+		v3d_perfmon_stop(v3d, perfmon, false);
 
 	v3d_perfmon_put(perfmon);
 
···
 
 void v3d_perfmon_close_file(struct v3d_file_priv *v3d_priv)
 {
+	struct v3d_dev *v3d = v3d_priv->v3d;
+
 	mutex_lock(&v3d_priv->perfmon.lock);
-	idr_for_each(&v3d_priv->perfmon.idr, v3d_perfmon_idr_del, NULL);
+	idr_for_each(&v3d_priv->perfmon.idr, v3d_perfmon_idr_del, v3d);
 	idr_destroy(&v3d_priv->perfmon.idr);
 	mutex_unlock(&v3d_priv->perfmon.lock);
 	mutex_destroy(&v3d_priv->perfmon.lock);

drivers/gpu/drm/vc4/vc4_perfmon.c

···
 static int vc4_perfmon_idr_del(int id, void *elem, void *data)
 {
 	struct vc4_perfmon *perfmon = elem;
+	struct vc4_dev *vc4 = (struct vc4_dev *)data;
+
+	/* If the active perfmon is being destroyed, stop it first */
+	if (perfmon == vc4->active_perfmon)
+		vc4_perfmon_stop(vc4, perfmon, false);
 
 	vc4_perfmon_put(perfmon);
 
···
 		return;
 
 	mutex_lock(&vc4file->perfmon.lock);
-	idr_for_each(&vc4file->perfmon.idr, vc4_perfmon_idr_del, NULL);
+	idr_for_each(&vc4file->perfmon.idr, vc4_perfmon_idr_del, vc4);
 	idr_destroy(&vc4file->perfmon.idr);
 	mutex_unlock(&vc4file->perfmon.lock);
 	mutex_destroy(&vc4file->perfmon.lock);

drivers/gpu/drm/xe/xe_debugfs.c

···
 	debugfs_create_file("forcewake_all", 0400, root, xe,
 			    &forcewake_all_fops);
 
-	debugfs_create_file("wedged_mode", 0400, root, xe,
+	debugfs_create_file("wedged_mode", 0600, root, xe,
 			    &wedged_mode_fops);
 
 	for (mem_type = XE_PL_VRAM0; mem_type <= XE_PL_VRAM1; ++mem_type) {

drivers/gpu/drm/xe/xe_gt.c

···
 	int ret = 0;
 
 	if ((!xe_uc_fw_is_available(&gt->uc.gsc.fw) ||
-	     xe_uc_fw_is_loaded(&gt->uc.gsc.fw)) && XE_WA(gt, 22019338487))
+	     xe_uc_fw_is_loaded(&gt->uc.gsc.fw) ||
+	     xe_uc_fw_is_in_error_state(&gt->uc.gsc.fw)) &&
+	    XE_WA(gt, 22019338487))
 		ret = xe_guc_pc_restore_stashed_freq(&gt->uc.guc.pc);
 
 	return ret;

drivers/gpu/drm/xe/xe_guc_ct.c

···
 		num_g2h = 1;
 
 		if (g2h_fence_needs_alloc(g2h_fence)) {
-			void *ptr;
-
 			g2h_fence->seqno = next_ct_seqno(ct, true);
-			ptr = xa_store(&ct->fence_lookup,
-				       g2h_fence->seqno,
-				       g2h_fence, GFP_ATOMIC);
-			if (IS_ERR(ptr)) {
-				ret = PTR_ERR(ptr);
+			ret = xa_err(xa_store(&ct->fence_lookup,
+					      g2h_fence->seqno, g2h_fence,
+					      GFP_ATOMIC));
+			if (ret)
 				goto out;
-			}
 		}
 
 		seqno = g2h_fence->seqno;
···
 retry_same_fence:
 	ret = guc_ct_send(ct, action, len, 0, 0, &g2h_fence);
 	if (unlikely(ret == -ENOMEM)) {
-		void *ptr;
-
 		/* Retry allocation /w GFP_KERNEL */
-		ptr = xa_store(&ct->fence_lookup,
-			       g2h_fence.seqno,
-			       &g2h_fence, GFP_KERNEL);
-		if (IS_ERR(ptr))
-			return PTR_ERR(ptr);
+		ret = xa_err(xa_store(&ct->fence_lookup, g2h_fence.seqno,
+				      &g2h_fence, GFP_KERNEL));
+		if (ret)
+			return ret;
 
 		goto retry_same_fence;
 	} else if (unlikely(ret)) {
···
 	}
 
 	ret = wait_event_timeout(ct->g2h_fence_wq, g2h_fence.done, HZ);
+
+	/*
+	 * Ensure we serialize with completion side to prevent UAF with fence going out of scope on
+	 * the stack, since we have no clue if it will fire after the timeout before we can erase
+	 * from the xa. Also we have some dependent loads and stores below for which we need the
+	 * correct ordering, and we lack the needed barriers.
+	 */
+	mutex_lock(&ct->lock);
 	if (!ret) {
-		xe_gt_err(gt, "Timed out wait for G2H, fence %u, action %04x",
-			  g2h_fence.seqno, action[0]);
+		xe_gt_err(gt, "Timed out wait for G2H, fence %u, action %04x, done %s",
+			  g2h_fence.seqno, action[0], str_yes_no(g2h_fence.done));
 		xa_erase_irq(&ct->fence_lookup, g2h_fence.seqno);
+		mutex_unlock(&ct->lock);
 		return -ETIME;
 	}
 
 	if (g2h_fence.retry) {
 		xe_gt_dbg(gt, "H2G action %#x retrying: reason %#x\n",
 			  action[0], g2h_fence.reason);
+		mutex_unlock(&ct->lock);
 		goto retry;
 	}
 	if (g2h_fence.fail) {
···
 		ret = -EIO;
 	}
 
-	return ret > 0 ? response_buffer ? g2h_fence.response_len : g2h_fence.response_data : ret;
+	if (ret > 0)
+		ret = response_buffer ? g2h_fence.response_len : g2h_fence.response_data;
+
+	mutex_unlock(&ct->lock);
+
+	return ret;
 }
 
 /**

drivers/gpu/drm/xe/xe_guc_submit.c

···
 		 EXEC_QUEUE_STATE_BANNED));
 }
 
-#ifdef CONFIG_PROVE_LOCKING
-static int alloc_submit_wq(struct xe_guc *guc)
-{
-	int i;
-
-	for (i = 0; i < NUM_SUBMIT_WQ; ++i) {
-		guc->submission_state.submit_wq_pool[i] =
-			alloc_ordered_workqueue("submit_wq", 0);
-		if (!guc->submission_state.submit_wq_pool[i])
-			goto err_free;
-	}
-
-	return 0;
-
-err_free:
-	while (i)
-		destroy_workqueue(guc->submission_state.submit_wq_pool[--i]);
-
-	return -ENOMEM;
-}
-
-static void free_submit_wq(struct xe_guc *guc)
-{
-	int i;
-
-	for (i = 0; i < NUM_SUBMIT_WQ; ++i)
-		destroy_workqueue(guc->submission_state.submit_wq_pool[i]);
-}
-
-static struct workqueue_struct *get_submit_wq(struct xe_guc *guc)
-{
-	int idx = guc->submission_state.submit_wq_idx++ % NUM_SUBMIT_WQ;
-
-	return guc->submission_state.submit_wq_pool[idx];
-}
-#else
-static int alloc_submit_wq(struct xe_guc *guc)
-{
-	return 0;
-}
-
-static void free_submit_wq(struct xe_guc *guc)
-{
-
-}
-
-static struct workqueue_struct *get_submit_wq(struct xe_guc *guc)
-{
-	return NULL;
-}
-#endif
-
-static void xe_guc_submit_fini(struct xe_guc *guc)
-{
-	struct xe_device *xe = guc_to_xe(guc);
-	struct xe_gt *gt = guc_to_gt(guc);
-	int ret;
-
-	ret = wait_event_timeout(guc->submission_state.fini_wq,
-				 xa_empty(&guc->submission_state.exec_queue_lookup),
-				 HZ * 5);
-
-	drain_workqueue(xe->destroy_wq);
-
-	xe_gt_assert(gt, ret);
-}
-
 static void guc_submit_fini(struct drm_device *drm, void *arg)
 {
 	struct xe_guc *guc = arg;
 
-	xe_guc_submit_fini(guc);
 	xa_destroy(&guc->submission_state.exec_queue_lookup);
-	free_submit_wq(guc);
 }
 
 static void guc_submit_wedged_fini(void *arg)
···
 	if (err)
 		return err;
 
-	err = alloc_submit_wq(guc);
-	if (err)
-		return err;
-
 	gt->exec_queue_ops = &guc_exec_queue_ops;
 
 	xa_init(&guc->submission_state.exec_queue_lookup);
···
 static int alloc_guc_id(struct xe_guc *guc, struct xe_exec_queue *q)
 {
 	int ret;
-	void *ptr;
 	int i;
 
 	/*
···
 	q->guc->id = ret;
 
 	for (i = 0; i < q->width; ++i) {
-		ptr = xa_store(&guc->submission_state.exec_queue_lookup,
-			       q->guc->id + i, q, GFP_NOWAIT);
-		if (IS_ERR(ptr)) {
-			ret = PTR_ERR(ptr);
+		ret = xa_err(xa_store(&guc->submission_state.exec_queue_lookup,
+				      q->guc->id + i, q, GFP_NOWAIT));
+		if (ret)
 			goto err_release;
-		}
 	}
 
 	return 0;
···
 	timeout = (q->vm && xe_vm_in_lr_mode(q->vm)) ? MAX_SCHEDULE_TIMEOUT :
 		  msecs_to_jiffies(q->sched_props.job_timeout_ms);
 	err = xe_sched_init(&ge->sched, &drm_sched_ops, &xe_sched_ops,
-			    get_submit_wq(guc),
-			    q->lrc[0]->ring.size / MAX_JOB_SIZE_BYTES, 64,
+			    NULL, q->lrc[0]->ring.size / MAX_JOB_SIZE_BYTES, 64,
 			    timeout, guc_to_gt(guc)->ordered_wq, NULL,
 			    q->name, gt_to_xe(q->gt)->drm.dev);
 	if (err)

drivers/gpu/drm/xe/xe_guc_types.h

···
 		atomic_t stopped;
 		/** @submission_state.lock: protects submission state */
 		struct mutex lock;
-#ifdef CONFIG_PROVE_LOCKING
-#define NUM_SUBMIT_WQ	256
-		/** @submission_state.submit_wq_pool: submission ordered workqueues pool */
-		struct workqueue_struct *submit_wq_pool[NUM_SUBMIT_WQ];
-		/** @submission_state.submit_wq_idx: submission ordered workqueue index */
-		int submit_wq_idx;
-#endif
 		/** @submission_state.enabled: submission is enabled */
 		bool enabled;
 		/** @submission_state.fini_wq: submit fini wait queue */

drivers/hid/amd-sfh-hid/amd_sfh_client.c

···
 	cl_data->in_data = in_data;
 
 	for (i = 0; i < cl_data->num_hid_devices; i++) {
-		in_data->sensor_virt_addr[i] = dma_alloc_coherent(dev, sizeof(int) * 8,
-								  &cl_data->sensor_dma_addr[i],
-								  GFP_KERNEL);
+		in_data->sensor_virt_addr[i] = dmam_alloc_coherent(dev, sizeof(int) * 8,
+								   &cl_data->sensor_dma_addr[i],
+								   GFP_KERNEL);
 		if (!in_data->sensor_virt_addr[i]) {
 			rc = -ENOMEM;
 			goto cleanup;
···
 int amd_sfh_hid_client_deinit(struct amd_mp2_dev *privdata)
 {
 	struct amdtp_cl_data *cl_data = privdata->cl_data;
-	struct amd_input_data *in_data = cl_data->in_data;
 	int i, status;
 
 	for (i = 0; i < cl_data->num_hid_devices; i++) {
···
 	cancel_delayed_work_sync(&cl_data->work_buffer);
 	amdtp_hid_remove(cl_data);
 
-	for (i = 0; i < cl_data->num_hid_devices; i++) {
-		if (in_data->sensor_virt_addr[i]) {
-			dma_free_coherent(&privdata->pdev->dev, 8 * sizeof(int),
-					  in_data->sensor_virt_addr[i],
-					  cl_data->sensor_dma_addr[i]);
-		}
-	}
 	return 0;
 }

drivers/hid/hid-ids.h

···
 #define USB_DEVICE_ID_PLANTRONICS_BLACKWIRE_3220_SERIES	0xc056
 #define USB_DEVICE_ID_PLANTRONICS_BLACKWIRE_3215_SERIES	0xc057
 #define USB_DEVICE_ID_PLANTRONICS_BLACKWIRE_3225_SERIES	0xc058
+#define USB_DEVICE_ID_PLANTRONICS_BLACKWIRE_3325_SERIES	0x430c
+#define USB_DEVICE_ID_PLANTRONICS_ENCOREPRO_500_SERIES		0x431e
 
 #define USB_VENDOR_ID_PANASONIC		0x04da
 #define USB_DEVICE_ID_PANABOARD_UBT780	0x1044

drivers/hid/hid-multitouch.c

···
 		HID_DEVICE(BUS_I2C, HID_GROUP_MULTITOUCH_WIN_8,
 			USB_VENDOR_ID_ELAN, 0x3148) },
 
+	{ .driver_data = MT_CLS_WIN_8_FORCE_MULTI_INPUT_NSMU,
+		HID_DEVICE(BUS_I2C, HID_GROUP_MULTITOUCH_WIN_8,
+			USB_VENDOR_ID_ELAN, 0x32ae) },
+
 	/* Elitegroup panel */
 	{ .driver_data = MT_CLS_SERIAL,
 		MT_USB_DEVICE(USB_VENDOR_ID_ELITEGROUP,
···
 	{ .driver_data = MT_CLS_VTL,
 		HID_DEVICE(BUS_I2C, HID_GROUP_MULTITOUCH_WIN_8,
 			0x347d, 0x7853) },
+
+	/* HONOR MagicBook Art 14 touchpad */
+	{ .driver_data = MT_CLS_VTL,
+		HID_DEVICE(BUS_I2C, HID_GROUP_MULTITOUCH_WIN_8,
+			0x35cc, 0x0104) },
 
 	/* Ilitek dual touch panel */
 	{  .driver_data = MT_CLS_NSMU,

drivers/hid/hid-plantronics.c

···
 			    (usage->hid & HID_USAGE_PAGE) == HID_UP_CONSUMER)
 
 #define PLT_QUIRK_DOUBLE_VOLUME_KEYS BIT(0)
+#define PLT_QUIRK_FOLLOWED_OPPOSITE_VOLUME_KEYS BIT(1)
 
 #define PLT_DOUBLE_KEY_TIMEOUT 5 /* ms */
+#define PLT_FOLLOWED_OPPOSITE_KEY_TIMEOUT 220 /* ms */
 
 struct plt_drv_data {
 	unsigned long device_type;
···
 
 		drv_data->last_volume_key_ts = cur_ts;
 	}
+	if (drv_data->quirks & PLT_QUIRK_FOLLOWED_OPPOSITE_VOLUME_KEYS) {
+		unsigned long prev_ts, cur_ts;
+
+		/* Usages are filtered in plantronics_usages. */
+
+		if (!value) /* Handle key presses only. */
+			return 0;
+
+		prev_ts = drv_data->last_volume_key_ts;
+		cur_ts = jiffies;
+		if (jiffies_to_msecs(cur_ts - prev_ts) <= PLT_FOLLOWED_OPPOSITE_KEY_TIMEOUT)
+			return 1; /* Ignore the followed opposite volume key. */
+
+		drv_data->last_volume_key_ts = cur_ts;
+	}
 
 	return 0;
 }
···
 	{ HID_USB_DEVICE(USB_VENDOR_ID_PLANTRONICS,
 					 USB_DEVICE_ID_PLANTRONICS_BLACKWIRE_3225_SERIES),
 		.driver_data = PLT_QUIRK_DOUBLE_VOLUME_KEYS },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_PLANTRONICS,
+					 USB_DEVICE_ID_PLANTRONICS_BLACKWIRE_3325_SERIES),
+		.driver_data = PLT_QUIRK_FOLLOWED_OPPOSITE_VOLUME_KEYS },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_PLANTRONICS,
+					 USB_DEVICE_ID_PLANTRONICS_ENCOREPRO_500_SERIES),
+		.driver_data = PLT_QUIRK_FOLLOWED_OPPOSITE_VOLUME_KEYS },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_PLANTRONICS, HID_ANY_ID) },
 	{ }
 };

drivers/hid/intel-ish-hid/ishtp-fw-loader.c

···
 				  const struct firmware *fw,
 				  const struct shim_fw_info fw_info)
 {
-	int rv;
+	int rv = 0;
 	void *dma_buf;
 	dma_addr_t dma_buf_phy;
 	u32 fragment_offset, fragment_size, payload_max_size;

drivers/hid/wacom_wac.c

···
 		/* Going into range select tool */
 		if (wacom_wac->hid_data.invert_state)
 			wacom_wac->tool[0] = BTN_TOOL_RUBBER;
+		else if (wacom_wac->features.quirks & WACOM_QUIRK_AESPEN)
+			wacom_wac->tool[0] = BTN_TOOL_PEN;
 		else if (wacom_wac->id[0])
 			wacom_wac->tool[0] = wacom_intuos_get_tool_type(wacom_wac->id[0]);
 		else

drivers/hwmon/Kconfig

···
 	tristate "Analog Devices ADM9240 and compatibles"
 	depends on I2C
 	select HWMON_VID
+	select REGMAP_I2C
 	help
 	  If you say yes here you get support for Analog Devices ADM9240,
 	  Dallas DS1780, National Semiconductor LM81 sensor chips.
···
 config SENSORS_ADT7470
 	tristate "Analog Devices ADT7470"
 	depends on I2C
+	select REGMAP_I2C
 	help
 	  If you say yes here you get support for the Analog Devices
 	  ADT7470 temperature monitoring chips.
···
 config SENSORS_LTC2991
 	tristate "Analog Devices LTC2991"
 	depends on I2C
+	select REGMAP_I2C
 	help
 	  If you say yes here you get support for Analog Devices LTC2991
 	  Octal I2C Voltage, Current, and Temperature Monitor. The LTC2991
···
 config SENSORS_MAX1668
 	tristate "Maxim MAX1668 and compatibles"
 	depends on I2C
+	select REGMAP_I2C
 	help
 	  If you say yes here you get support for MAX1668, MAX1989 and
 	  MAX1805 chips.
···
 config SENSORS_MC34VR500
 	tristate "NXP MC34VR500 hardware monitoring driver"
 	depends on I2C
+	select REGMAP_I2C
 	help
 	  If you say yes here you get support for the temperature and input
 	  voltage sensors of the NXP MC34VR500.
···
 config SENSORS_TMP513
 	tristate "Texas Instruments TMP513 and compatibles"
 	depends on I2C
+	select REGMAP_I2C
 	help
 	  If you say yes here you get support for Texas Instruments TMP512,
 	  and TMP513 temperature and power supply sensor chips.

drivers/hwmon/adt7475.c

···
 static int adt7475_fan_pwm_config(struct i2c_client *client)
 {
 	struct adt7475_data *data = i2c_get_clientdata(client);
-	struct fwnode_handle *child;
 	struct adt7475_pwm_config cfg = {};
 	int ret;
 
-	device_for_each_child_node(&client->dev, child) {
+	device_for_each_child_node_scoped(&client->dev, child) {
 		if (!fwnode_property_present(child, "pwms"))
 			continue;
 

drivers/hwmon/intel-m10-bmc-hwmon.c

···
 	{ 0x4f0, 0x4f4, 0x4f8, 0x52c, 0x0, 500, "Board Top Near FPGA Temperature" },
 	{ 0x4fc, 0x500, 0x504, 0x52c, 0x0, 500, "Board Bottom Near CVL Temperature" },
 	{ 0x508, 0x50c, 0x510, 0x52c, 0x0, 500, "Board Top East Near VRs Temperature" },
-	{ 0x514, 0x518, 0x51c, 0x52c, 0x0, 500, "Columbiaville Die Temperature" },
+	{ 0x514, 0x518, 0x51c, 0x52c, 0x0, 500, "CVL Die Temperature" },
 	{ 0x520, 0x524, 0x528, 0x52c, 0x0, 500, "Board Rear Side Temperature" },
 	{ 0x530, 0x534, 0x538, 0x52c, 0x0, 500, "Board Front Side Temperature" },
 	{ 0x53c, 0x540, 0x544, 0x0, 0x0, 500, "QSFP1 Case Temperature" },

drivers/infiniband/core/addr.c

···
 		break;
 #endif
 	}
+	if (!ret && dev && is_vlan_dev(dev))
+		dev = vlan_dev_real_dev(dev);
 	return ret ? ERR_PTR(ret) : dev;
 }
 

drivers/infiniband/core/nldev.c

···
 	nlh = nlmsg_put(skb, 0, 0,
 			RDMA_NL_GET_TYPE(RDMA_NL_NLDEV, RDMA_NLDEV_CMD_MONITOR),
 			0, 0);
+	if (!nlh)
+		goto err_free;
 
 	switch (type) {
 	case RDMA_REGISTER_EVENT:

drivers/infiniband/hw/bnxt_re/hw_counters.c

···
 				goto done;
 			}
 		}
-		if (rdev->pacing.dbr_pacing)
+		if (rdev->pacing.dbr_pacing && bnxt_qplib_is_chip_gen_p5_p7(rdev->chip_ctx))
 			bnxt_re_copy_db_pacing_stats(rdev, stats);
 	}
 

drivers/infiniband/hw/bnxt_re/ib_verbs.c

···
 			0 : BNXT_QPLIB_RESERVED_QP_WRS;
 		entries = bnxt_re_init_depth(entries + diff + 1, uctx);
 		sq->max_wqe = min_t(u32, entries, dev_attr->max_qp_wqes + diff + 1);
-		sq->max_sw_wqe = bnxt_qplib_get_depth(sq, qplqp->wqe_mode, true);
+		if (qplqp->wqe_mode == BNXT_QPLIB_WQE_MODE_VARIABLE)
+			sq->max_sw_wqe = bnxt_qplib_get_depth(sq, qplqp->wqe_mode, true);
+		else
+			sq->max_sw_wqe = sq->max_wqe;
+
 	}
 	sq->q_full_delta = diff + 1;
 	/*

drivers/infiniband/hw/bnxt_re/main.c

···
 
 	bnxt_re_set_db_offset(rdev);
 	rc = bnxt_qplib_map_db_bar(&rdev->qplib_res);
-	if (rc)
+	if (rc) {
+		kfree(rdev->chip_ctx);
+		rdev->chip_ctx = NULL;
 		return rc;
+	}
 
 	if (bnxt_qplib_determine_atomics(en_dev->pdev))
 		ibdev_info(&rdev->ibdev,
···
 static void __wait_for_fifo_occupancy_below_th(struct bnxt_re_dev *rdev)
 {
 	struct bnxt_qplib_db_pacing_data *pacing_data = rdev->qplib_res.pacing_data;
+	u32 retry_fifo_check = 1000;
 	u32 fifo_occup;
 
 	/* loop shouldn't run infintely as the occupancy usually goes
···
 
 		if (fifo_occup < pacing_data->pacing_th)
 			break;
+		if (!retry_fifo_check--) {
+			dev_info_once(rdev_to_dev(rdev),
+				      "%s: fifo_occup = 0x%xfifo_max_depth = 0x%x pacing_th = 0x%x\n",
+				      __func__, fifo_occup, pacing_data->fifo_max_depth,
+					pacing_data->pacing_th);
+			break;
+		}
+
 	}
 }
 
···
 	return ib_register_device(ibdev, "bnxt_re%d", &rdev->en_dev->pdev->dev);
 }
 
-static struct bnxt_re_dev *bnxt_re_dev_add(struct bnxt_aux_priv *aux_priv,
+static struct bnxt_re_dev *bnxt_re_dev_add(struct auxiliary_device *adev,
 					   struct bnxt_en_dev *en_dev)
 {
 	struct bnxt_re_dev *rdev;
···
 	rdev->nb.notifier_call = NULL;
 	rdev->netdev = en_dev->net;
 	rdev->en_dev = en_dev;
+	rdev->adev = adev;
 	rdev->id = rdev->en_dev->pdev->devfn;
 	INIT_LIST_HEAD(&rdev->qp_list);
 	mutex_init(&rdev->qp_lock);
···
 static int bnxt_re_handle_qp_async_event(struct creq_qp_event *qp_event,
 					 struct bnxt_re_qp *qp)
 {
-	struct bnxt_re_srq *srq = container_of(qp->qplib_qp.srq, struct bnxt_re_srq,
-					       qplib_srq);
 	struct creq_qp_error_notification *err_event;
+	struct bnxt_re_srq *srq = NULL;
 	struct ib_event event = {};
 	unsigned int flags;
+
+	if (qp->qplib_qp.srq)
+		srq =  container_of(qp->qplib_qp.srq, struct bnxt_re_srq,
+				    qplib_srq);
 
 	if (qp->qplib_qp.state == CMDQ_MODIFY_QP_NEW_STATE_ERR &&
 	    rdma_is_kernel_res(&qp->ib_qp.res)) {
···
 {
 	struct bnxt_re_cq *cq = container_of(handle, struct bnxt_re_cq,
 					     qplib_cq);
-	u32 *cq_ptr;
 
-	if (cq->ib_cq.comp_handler) {
-		if (cq->uctx_cq_page) {
-			cq_ptr = (u32 *)cq->uctx_cq_page;
-			*cq_ptr = cq->qplib_cq.toggle;
-		}
+	if (cq->ib_cq.comp_handler)
 		(*cq->ib_cq.comp_handler)(&cq->ib_cq, cq->ib_cq.cq_context);
-	}
 
 	return 0;
 }
···
 	 */
 	rtnl_lock();
 	en_info->rdev = rdev;
-	rdev->adev = adev;
 	rtnl_unlock();
 }
 
···
 	en_dev = en_info->en_dev;
 
 
-	rdev = bnxt_re_dev_add(aux_priv, en_dev);
+	rdev = bnxt_re_dev_add(adev, en_dev);
 	if (!rdev || !rdev_to_dev(rdev)) {
 		rc = -ENOMEM;
 		goto exit;
···
 		rdev->nb.notifier_call = NULL;
 		pr_err("%s: Cannot register to netdevice_notifier",
 		       ROCE_DRV_MODULE_NAME);
-		return rc;
+		goto re_dev_unreg;
 	}
 	bnxt_re_setup_cc(rdev, true);
 
 	return 0;
 
+re_dev_unreg:
+	ib_unregister_device(&rdev->ibdev);
 re_dev_uninit:
 	bnxt_re_update_en_info_rdev(NULL, en_info, adev);
 	bnxt_re_dev_uninit(rdev, BNXT_RE_COMPLETE_REMOVE);
···
 	auxiliary_set_drvdata(adev, en_info);
 
 	rc = bnxt_re_add_device(adev, BNXT_RE_COMPLETE_INIT);
-	if (rc)
-		goto err;
 	mutex_unlock(&bnxt_re_mutex);
-	return 0;
-
-err:
-	mutex_unlock(&bnxt_re_mutex);
-	bnxt_re_remove(adev);
-
 	return rc;
 }
 

drivers/infiniband/hw/bnxt_re/qplib_fp.c

···
 		case NQ_BASE_TYPE_CQ_NOTIFICATION:
 		{
 			struct nq_cn *nqcne = (struct nq_cn *)nqe;
+			struct bnxt_re_cq *cq_p;
 
 			q_handle = le32_to_cpu(nqcne->cq_handle_low);
 			q_handle |= (u64)le32_to_cpu(nqcne->cq_handle_high)
···
 			cq->toggle = (le16_to_cpu(nqe->info10_type) &
 					NQ_CN_TOGGLE_MASK) >> NQ_CN_TOGGLE_SFT;
 			cq->dbinfo.toggle = cq->toggle;
+			cq_p = container_of(cq, struct bnxt_re_cq, qplib_cq);
+			if (cq_p->uctx_cq_page)
+				*((u32 *)cq_p->uctx_cq_page) = cq->toggle;
+
 			bnxt_qplib_armen_db(&cq->dbinfo,
 					    DBC_DBC_TYPE_CQ_ARMENA);
 			spin_lock_bh(&cq->compl_lock);

drivers/infiniband/hw/bnxt_re/qplib_fp.h

···
 			};
 			u32		q_key;
 			u32		dst_qp;
-			u16		avid;
+			u32		avid;
 		} send;
 
 		/* Send Raw Ethernet and QP1 */

drivers/infiniband/hw/bnxt_re/qplib_rcfw.c

···
 		/* failed with status */
 		dev_err(&rcfw->pdev->dev, "cmdq[%#x]=%#x status %#x\n",
 			cookie, opcode, evnt->status);
-		rc = -EFAULT;
+		rc = -EIO;
 	}
 
 	return rc;

drivers/infiniband/hw/bnxt_re/qplib_res.c

···
 			sginfo.pgsize = npde * pg_size;
 			sginfo.npages = 1;
 			rc = __alloc_pbl(res, &hwq->pbl[PBL_LVL_0], &sginfo);
+			if (rc)
+				goto fail;
 
 			/* Alloc PBL pages */
 			sginfo.npages = npbl;
···
 			dst_virt_ptr =
 				(dma_addr_t **)hwq->pbl[PBL_LVL_0].pg_arr;
 			src_phys_ptr = hwq->pbl[PBL_LVL_1].pg_map_arr;
-			if (hwq_attr->type == HWQ_TYPE_MR) {
-			/* For MR it is expected that we supply only 1 contigous
-			 * page i.e only 1 entry in the PDL that will contain
-			 * all the PBLs for the user supplied memory region
-			 */
-				for (i = 0; i < hwq->pbl[PBL_LVL_1].pg_count;
-				     i++)
-					dst_virt_ptr[0][i] = src_phys_ptr[i] |
-						flag;
-			} else {
-				for (i = 0; i < hwq->pbl[PBL_LVL_1].pg_count;
-				     i++)
-					dst_virt_ptr[PTR_PG(i)][PTR_IDX(i)] =
-						src_phys_ptr[i] |
-						PTU_PDE_VALID;
-			}
+			for (i = 0; i < hwq->pbl[PBL_LVL_1].pg_count; i++)
+				dst_virt_ptr[0][i] = src_phys_ptr[i] | flag;
+
 			/* Alloc or init PTEs */
 			rc = __alloc_pbl(res, &hwq->pbl[PBL_LVL_2],
 					 hwq_attr->sginfo);

drivers/infiniband/hw/bnxt_re/qplib_sp.c

···
 			    min_t(u32, sb->max_sge_var_wqe, BNXT_VAR_MAX_SGE) : 6;
 	attr->max_cq = le32_to_cpu(sb->max_cq);
 	attr->max_cq_wqes = le32_to_cpu(sb->max_cqe);
+	if (!bnxt_qplib_is_chip_gen_p7(rcfw->res->cctx))
+		attr->max_cq_wqes = min_t(u32, BNXT_QPLIB_MAX_CQ_WQES, attr->max_cq_wqes);
 	attr->max_cq_sges = attr->max_qp_sges;
 	attr->max_mr = le32_to_cpu(sb->max_mr);
 	attr->max_mw = le32_to_cpu(sb->max_mw);
···
 	if (!bnxt_qplib_is_chip_gen_p7(rcfw->res->cctx))
 		attr->l2_db_size = (sb->l2_db_space_size + 1) *
 				    (0x01 << RCFW_DBR_BASE_PAGE_SHIFT);
-	attr->max_sgid = BNXT_QPLIB_NUM_GIDS_SUPPORTED;
+	/*
+	 * Read the max gid supported by HW.
+	 * For each entry in HW  GID in HW table, we consume 2
+	 * GID entries in the kernel GID table.  So max_gid reported
+	 * to stack can be up to twice the value reported by the HW, up to 256 gids.
+	 */
+	attr->max_sgid = le32_to_cpu(sb->max_gid);
+	attr->max_sgid = min_t(u32, BNXT_QPLIB_NUM_GIDS_SUPPORTED, 2 * attr->max_sgid);
 	attr->dev_cap_flags = le16_to_cpu(sb->dev_cap_flags);
 	attr->dev_cap_flags2 = le16_to_cpu(sb->dev_cap_ext_flags_2);
 

drivers/infiniband/hw/bnxt_re/qplib_sp.h

···
 	u32				max_qp_wqes;
 	u32				max_qp_sges;
 	u32				max_cq;
+#define BNXT_QPLIB_MAX_CQ_WQES          0xfffff
 	u32				max_cq_wqes;
 	u32				max_cq_sges;
 	u32				max_mr;

drivers/infiniband/hw/cxgb4/cm.c

···
 	err = -ENOMEM;
 	if (n->dev->flags & IFF_LOOPBACK) {
 		if (iptype == 4)
-			pdev = ip_dev_find(&init_net, *(__be32 *)peer_ip);
+			pdev = __ip_dev_find(&init_net, *(__be32 *)peer_ip, false);
 		else if (IS_ENABLED(CONFIG_IPV6))
 			for_each_netdev(&init_net, pdev) {
 				if (ipv6_chk_addr(&init_net,
···
 			err = -ENODEV;
 			goto out;
 		}
+		if (is_vlan_dev(pdev))
+			pdev = vlan_dev_real_dev(pdev);
 		ep->l2t = cxgb4_l2t_get(cdev->rdev.lldi.l2t,
 					n, pdev, rt_tos2priority(tos));
-		if (!ep->l2t) {
-			dev_put(pdev);
+		if (!ep->l2t)
 			goto out;
-		}
 		ep->mtu = pdev->mtu;
 		ep->tx_chan = cxgb4_port_chan(pdev);
 		ep->smac_idx = ((struct port_info *)netdev_priv(pdev))->smt_idx;
···
 		ep->rss_qid = cdev->rdev.lldi.rxq_ids[
 			cxgb4_port_idx(pdev) * step];
 		set_tcp_window(ep, (struct port_info *)netdev_priv(pdev));
-		dev_put(pdev);
 	} else {
 		pdev = get_real_dev(n->dev);
 		ep->l2t = cxgb4_l2t_get(cdev->rdev.lldi.l2t,

drivers/infiniband/hw/irdma/cm.c

···
 /**
  * irdma_accept - registered call for connection to be accepted
  * @cm_id: cm information for passive connection
- * @conn_param: accpet parameters
+ * @conn_param: accept parameters
  */
 int irdma_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 {

drivers/infiniband/sw/siw/siw_qp_tx.c

···
 			msg.msg_flags &= ~MSG_MORE;
 
 		tcp_rate_check_app_limited(sk);
+		if (!sendpage_ok(page[i]))
+			msg.msg_flags &= ~MSG_SPLICE_PAGES;
 		bvec_set_page(&bvec, page[i], bytes, offset);
 		iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, &bvec, 1, size);
 

drivers/infiniband/ulp/srpt/ib_srpt.c

···
 static u64 srpt_service_guid;
 static DEFINE_SPINLOCK(srpt_dev_lock);	/* Protects srpt_dev_list. */
 static LIST_HEAD(srpt_dev_list);	/* List of srpt_device structures. */
+static DEFINE_MUTEX(srpt_mc_mutex);	/* Protects srpt_memory_caches. */
+static DEFINE_XARRAY(srpt_memory_caches); /* See also srpt_memory_cache_entry */
 
 static unsigned srp_max_req_size = DEFAULT_MAX_REQ_SIZE;
 module_param(srp_max_req_size, int, 0444);
···
 static void srpt_recv_done(struct ib_cq *cq, struct ib_wc *wc);
 static void srpt_send_done(struct ib_cq *cq, struct ib_wc *wc);
 static void srpt_process_wait_list(struct srpt_rdma_ch *ch);
+
+/* Type of the entries in srpt_memory_caches. */
+struct srpt_memory_cache_entry {
+	refcount_t ref;
+	struct kmem_cache *c;
+};
+
+static struct kmem_cache *srpt_cache_get(unsigned int object_size)
+{
+	struct srpt_memory_cache_entry *e;
+	char name[32];
+	void *res;
+
+	guard(mutex)(&srpt_mc_mutex);
+	e = xa_load(&srpt_memory_caches, object_size);
+	if (e) {
+		refcount_inc(&e->ref);
+		return e->c;
+	}
+	snprintf(name, sizeof(name), "srpt-%u", object_size);
+	e = kmalloc(sizeof(*e), GFP_KERNEL);
+	if (!e)
+		return NULL;
+	refcount_set(&e->ref, 1);
+	e->c = kmem_cache_create(name, object_size, /*align=*/512, 0, NULL);
+	if (!e->c)
+		goto free_entry;
+	res = xa_store(&srpt_memory_caches, object_size, e, GFP_KERNEL);
+	if (xa_is_err(res))
+		goto destroy_cache;
+	return e->c;
+
+destroy_cache:
+	kmem_cache_destroy(e->c);
+
+free_entry:
+	kfree(e);
+	return NULL;
+}
+
+static void srpt_cache_put(struct kmem_cache *c)
+{
+	struct srpt_memory_cache_entry *e = NULL;
+	unsigned long object_size;
+
+	guard(mutex)(&srpt_mc_mutex);
+	xa_for_each(&srpt_memory_caches, object_size, e)
+		if (e->c == c)
+			break;
+	if (WARN_ON_ONCE(!e))
+		return;
+	if (!refcount_dec_and_test(&e->ref))
+		return;
+	WARN_ON_ONCE(xa_erase(&srpt_memory_caches, object_size) != e);
+	kmem_cache_destroy(e->c);
+	kfree(e);
+}
 
 /*
  * The only allowed channel state changes are those that change the channel
···
 			     ch->sport->sdev, ch->rq_size,
 			     ch->rsp_buf_cache, DMA_TO_DEVICE);
 
-	kmem_cache_destroy(ch->rsp_buf_cache);
+	srpt_cache_put(ch->rsp_buf_cache);
 
 	srpt_free_ioctx_ring((struct srpt_ioctx **)ch->ioctx_recv_ring,
 			     sdev, ch->rq_size,
 			     ch->req_buf_cache, DMA_FROM_DEVICE);
 
-	kmem_cache_destroy(ch->req_buf_cache);
+	srpt_cache_put(ch->req_buf_cache);
 
 	kref_put(&ch->kref, srpt_free_ch);
 }
···
 	INIT_LIST_HEAD(&ch->cmd_wait_list);
 	ch->max_rsp_size = ch->sport->port_attrib.srp_max_rsp_size;
 
-	ch->rsp_buf_cache = kmem_cache_create("srpt-rsp-buf", ch->max_rsp_size,
-					      512, 0, NULL);
+	ch->rsp_buf_cache = srpt_cache_get(ch->max_rsp_size);
 	if (!ch->rsp_buf_cache)
 		goto free_ch;
 
···
 		alignment_offset = round_up(imm_data_offset, 512) -
 			imm_data_offset;
 		req_sz = alignment_offset + imm_data_offset + srp_max_req_size;
-		ch->req_buf_cache = kmem_cache_create("srpt-req-buf", req_sz,
-						      512, 0, NULL);
+		ch->req_buf_cache = srpt_cache_get(req_sz);
 		if (!ch->req_buf_cache)
 			goto free_rsp_ring;
 
···
 			     ch->req_buf_cache, DMA_FROM_DEVICE);
 
 free_recv_cache:
-	kmem_cache_destroy(ch->req_buf_cache);
+	srpt_cache_put(ch->req_buf_cache);
 
 free_rsp_ring:
 	srpt_free_ioctx_ring((struct srpt_ioctx **)ch->ioctx_ring,
···
 			     ch->rsp_buf_cache, DMA_TO_DEVICE);
 
 free_rsp_cache:
-	kmem_cache_destroy(ch->rsp_buf_cache);
+	srpt_cache_put(ch->rsp_buf_cache);
 
 free_ch:
 	if (rdma_cm_id)
···
 	srpt_free_ioctx_ring((struct srpt_ioctx **)sdev->ioctx_ring, sdev,
 			     sdev->srq_size, sdev->req_buf_cache,
 			     DMA_FROM_DEVICE);
-	kmem_cache_destroy(sdev->req_buf_cache);
+	srpt_cache_put(sdev->req_buf_cache);
 	sdev->srq = NULL;
 }
 
···
 	pr_debug("create SRQ #wr= %d max_allow=%d dev= %s\n", sdev->srq_size,
 		 sdev->device->attrs.max_srq_wr, dev_name(&device->dev));
 
-	sdev->req_buf_cache = kmem_cache_create("srpt-srq-req-buf",
-						srp_max_req_size, 0, 0, NULL);
+	sdev->req_buf_cache = srpt_cache_get(srp_max_req_size);
 	if (!sdev->req_buf_cache)
 		goto free_srq;
 
···
 	return 0;
 
 free_cache:
-	kmem_cache_destroy(sdev->req_buf_cache);
+	srpt_cache_put(sdev->req_buf_cache);
 
 free_srq:
 	ib_destroy_srq(srq);

drivers/mmc/core/queue.c

···
 
 	blk_queue_rq_timeout(mq->queue, 60 * HZ);
 
-	dma_set_max_seg_size(mmc_dev(host), queue_max_segment_size(mq->queue));
+	if (mmc_dev(host)->dma_parms)
+		dma_set_max_seg_size(mmc_dev(host), queue_max_segment_size(mq->queue));
 
 	INIT_WORK(&mq->recovery_work, mmc_mq_recovery_handler);
 	INIT_WORK(&mq->complete_work, mmc_blk_mq_complete_work);

drivers/mmc/host/mvsdio.c

···
 	unsigned int xfer_mode;
 	unsigned int intr_en;
 	unsigned int ctrl;
-	bool use_pio;
-	struct sg_mapping_iter sg_miter;
 	unsigned int pio_size;
+	void *pio_ptr;
 	unsigned int sg_frags;
 	unsigned int ns_per_clk;
 	unsigned int clock;
···
 		 * data when the buffer is not aligned on a 64 byte
 		 * boundary.
 		 */
-		unsigned int miter_flags = SG_MITER_ATOMIC; /* Used from IRQ */
-
-		if (data->flags & MMC_DATA_READ)
-			miter_flags |= SG_MITER_TO_SG;
-		else
-			miter_flags |= SG_MITER_FROM_SG;
-
 		host->pio_size = data->blocks * data->blksz;
-		sg_miter_start(&host->sg_miter, data->sg, data->sg_len, miter_flags);
+		host->pio_ptr = sg_virt(data->sg);
 		if (!nodma)
-			dev_dbg(host->dev, "fallback to PIO for data\n");
-		host->use_pio = true;
+			dev_dbg(host->dev, "fallback to PIO for data at 0x%p size %d\n",
+				host->pio_ptr, host->pio_size);
 		return 1;
 	} else {
 		dma_addr_t phys_addr;
···
 		phys_addr = sg_dma_address(data->sg);
 		mvsd_write(MVSD_SYS_ADDR_LOW, (u32)phys_addr & 0xffff);
 		mvsd_write(MVSD_SYS_ADDR_HI,  (u32)phys_addr >> 16);
-		host->use_pio = false;
 		return 0;
 	}
 }
···
 {
 	void __iomem *iobase = host->base;
 
-	if (host->use_pio) {
-		sg_miter_stop(&host->sg_miter);
+	if (host->pio_ptr) {
+		host->pio_ptr = NULL;
 		host->pio_size = 0;
 	} else {
 		dma_unmap_sg(mmc_dev(host->mmc), data->sg, host->sg_frags,
···
 static irqreturn_t mvsd_irq(int irq, void *dev)
 {
 	struct mvsd_host *host = dev;
-	struct sg_mapping_iter *sgm = &host->sg_miter;
 	void __iomem *iobase = host->base;
 	u32 intr_status, intr_done_mask;
 	int irq_handled = 0;
-	u16 *p;
-	int s;
 
 	intr_status = mvsd_read(MVSD_NOR_INTR_STATUS);
 	dev_dbg(host->dev, "intr 0x%04x intr_en 0x%04x hw_state 0x%04x\n",
···
 	spin_lock(&host->lock);
 
 	/* PIO handling, if needed. Messy business... */
-	if (host->use_pio) {
-		/*
-		 * As we set sgm->consumed this always gives a valid buffer
-		 * position.
-		 */
-		if (!sg_miter_next(sgm)) {
-			/* This should not happen */
-			dev_err(host->dev, "ran out of scatter segments\n");
-			spin_unlock(&host->lock);
-			host->intr_en &=
-				~(MVSD_NOR_RX_READY | MVSD_NOR_RX_FIFO_8W |
-				  MVSD_NOR_TX_AVAIL | MVSD_NOR_TX_FIFO_8W);
-			mvsd_write(MVSD_NOR_INTR_EN, host->intr_en);
-			return IRQ_HANDLED;
-		}
-		p = sgm->addr;
-		s = sgm->length;
-		if (s > host->pio_size)
-			s = host->pio_size;
-	}
-
-	if (host->use_pio &&
+	if (host->pio_size &&
 	    (intr_status & host->intr_en &
 	     (MVSD_NOR_RX_READY | MVSD_NOR_RX_FIFO_8W))) {
-
+		u16 *p = host->pio_ptr;
+		int s = host->pio_size;
 		while (s >= 32 && (intr_status & MVSD_NOR_RX_FIFO_8W)) {
 			readsw(iobase + MVSD_FIFO, p, 16);
 			p += 16;
 			s -= 32;
-			sgm->consumed += 32;
 			intr_status = mvsd_read(MVSD_NOR_INTR_STATUS);
 		}
 		/*
···
 				put_unaligned(mvsd_read(MVSD_FIFO), p++);
 				put_unaligned(mvsd_read(MVSD_FIFO), p++);
 				s -= 4;
-				sgm->consumed += 4;
 				intr_status = mvsd_read(MVSD_NOR_INTR_STATUS);
 			}
 			if (s && s < 4 && (intr_status & MVSD_NOR_RX_READY)) {
···
 				val[0] = mvsd_read(MVSD_FIFO);
 				val[1] = mvsd_read(MVSD_FIFO);
 				memcpy(p, ((void *)&val) + 4 - s, s);
-				sgm->consumed += s;
 				s = 0;
 				intr_status = mvsd_read(MVSD_NOR_INTR_STATUS);
 			}
-			/* PIO transfer done */
-			host->pio_size -= sgm->consumed;
-			if (host->pio_size == 0) {
+			if (s == 0) {
 				host->intr_en &=
 				     ~(MVSD_NOR_RX_READY | MVSD_NOR_RX_FIFO_8W);
 				mvsd_write(MVSD_NOR_INTR_EN, host->intr_en);
···
 		}
 		dev_dbg(host->dev, "pio %d intr 0x%04x hw_state 0x%04x\n",
 			s, intr_status, mvsd_read(MVSD_HW_STATE));
+		host->pio_ptr = p;
+		host->pio_size = s;
 		irq_handled = 1;
-	} else if (host->use_pio &&
+	} else if (host->pio_size &&
 		   (intr_status & host->intr_en &
 		    (MVSD_NOR_TX_AVAIL | MVSD_NOR_TX_FIFO_8W))) {
+		u16 *p = host->pio_ptr;
+		int s = host->pio_size;
 		/*
 		 * The TX_FIFO_8W bit is unreliable. When set, bursting
 		 * 16 halfwords all at once in the FIFO drops data. Actually
···
 			mvsd_write(MVSD_FIFO, get_unaligned(p++));
 			mvsd_write(MVSD_FIFO, get_unaligned(p++));
 			s -= 4;
-			sgm->consumed += 4;
 			intr_status = mvsd_read(MVSD_NOR_INTR_STATUS);
 		}
 		if (s < 4) {
···
 				memcpy(((void *)&val) + 4 - s, p, s);
 				mvsd_write(MVSD_FIFO, val[0]);
 				mvsd_write(MVSD_FIFO, val[1]);
-				sgm->consumed += s;
 				s = 0;
 				intr_status = mvsd_read(MVSD_NOR_INTR_STATUS);
 			}
-			/* PIO transfer done */
-			host->pio_size -= sgm->consumed;
-			if (host->pio_size == 0) {
+			if (s == 0) {
 				host->intr_en &=
 				     ~(MVSD_NOR_TX_AVAIL | MVSD_NOR_TX_FIFO_8W);
 				mvsd_write(MVSD_NOR_INTR_EN, host->intr_en);
···
 		}
 		dev_dbg(host->dev, "pio %d intr 0x%04x hw_state 0x%04x\n",
 			s, intr_status, mvsd_read(MVSD_HW_STATE));
+		host->pio_ptr = p;
+		host->pio_size = s;
 		irq_handled = 1;
 	}
 

drivers/mmc/host/sdhci-of-dwcmshc.c

···
 
 	sdhci_reset(host, mask);
 
+	/* The T-Head 1520 SoC does not comply with the SDHCI specification
+	 * regarding the "Software Reset for CMD line should clear 'Command
+	 * Complete' in the Normal Interrupt Status Register." Clear the bit
+	 * here to compensate for this quirk.
+	 */
+	if (mask & SDHCI_RESET_CMD)
+		sdhci_writel(host, SDHCI_INT_RESPONSE, SDHCI_INT_STATUS);
+
 	if (priv->flags & FLAG_IO_FIXED_1V8) {
 		ctrl_2 = sdhci_readw(host, SDHCI_HOST_CONTROL2);
 		if (!(ctrl_2 & SDHCI_CTRL_VDD_180)) {

drivers/net/dsa/mv88e6xxx/chip.c

···
 		.invalid_port_mask = BIT(1) | BIT(2) | BIT(8),
 		.num_internal_phys = 5,
 		.internal_phys_offset = 3,
-		.max_vid = 4095,
+		.max_vid = 8191,
 		.max_sid = 63,
 		.port_base_addr = 0x0,
 		.phy_base_addr = 0x0,

drivers/net/dsa/vitesse-vsc73xx-core.c

···
 
 	dev_info(vsc->dev, "set up the switch\n");
 
-	ds->untag_bridge_pvid = true;
 	ds->max_num_bridges = DSA_TAG_8021Q_MAX_NUM_BRIDGES;
 	ds->fdb_isolation = true;
 

drivers/net/ethernet/aeroflex/greth.c

···
 
 	if (unlikely(skb->len > MAX_FRAME_SIZE)) {
 		dev->stats.tx_errors++;
-		goto out;
+		goto len_error;
 	}
 
 	/* Save skb pointer. */
···
 map_error:
 	if (net_ratelimit())
 		dev_warn(greth->dev, "Could not create TX DMA mapping\n");
+len_error:
 	dev_kfree_skb(skb);
 out:
 	return err;

drivers/net/ethernet/broadcom/asp2/bcmasp_intf.c

···
 			}
 			/* Rewind so we do not have a hole */
 			spb_index = intf->tx_spb_index;
+			dev_kfree_skb(skb);
 			return NETDEV_TX_OK;
 		}
 

drivers/net/ethernet/broadcom/bcmsysport.c

···
 		netif_err(priv, tx_err, dev, "DMA map failed at %p (len=%d)\n",
 			  skb->data, skb_len);
 		ret = NETDEV_TX_OK;
+		dev_kfree_skb_any(skb);
 		goto out;
 	}
 

drivers/net/ethernet/cadence/macb_main.c

···
 		return ret;
 	}
 
-	if (of_phy_is_fixed_link(np))
-		return mdiobus_register(bp->mii_bus);
-
 	/* Only create the PHY from the device tree if at least one PHY is
 	 * described. Otherwise scan the entire MDIO bus. We do this to support
 	 * old device tree that did not follow the best practices and did not
···
 
 static int macb_mii_init(struct macb *bp)
 {
+	struct device_node *child, *np = bp->pdev->dev.of_node;
 	int err = -ENXIO;
+
+	/* With fixed-link, we don't need to register the MDIO bus,
+	 * except if we have a child named "mdio" in the device tree.
+	 * In that case, some devices may be attached to the MACB's MDIO bus.
+	 */
+	child = of_get_child_by_name(np, "mdio");
+	if (child)
+		of_node_put(child);
+	else if (of_phy_is_fixed_link(np))
+		return macb_mii_probe(bp->dev);
 
 	/* Enable management port */
 	macb_writel(bp, NCR, MACB_BIT(MPE));

drivers/net/ethernet/freescale/enetc/enetc.c

···
 
 	if (unlikely(tx_frm_cnt && netif_carrier_ok(ndev) &&
 		     __netif_subqueue_stopped(ndev, tx_ring->index) &&
+		     !test_bit(ENETC_TX_DOWN, &priv->flags) &&
 		     (enetc_bd_unused(tx_ring) >= ENETC_TXBDS_MAX_NEEDED))) {
 		netif_wake_subqueue(ndev, tx_ring->index);
 	}
···
 	int xdp_tx_bd_cnt, i, k;
 	int xdp_tx_frm_cnt = 0;
 
+	if (unlikely(test_bit(ENETC_TX_DOWN, &priv->flags)))
+		return -ENETDOWN;
+
 	enetc_lock_mdio();
 
 	tx_ring = priv->xdp_tx_ring[smp_processor_id()];
···
 				  &rx_ring->rx_swbd[rx_ring_first]);
 		enetc_bdr_idx_inc(rx_ring, &rx_ring_first);
 	}
-	rx_ring->stats.xdp_drops++;
 }
 
 static int enetc_clean_rx_ring_xdp(struct enetc_bdr *rx_ring,
···
 			fallthrough;
 		case XDP_DROP:
 			enetc_xdp_drop(rx_ring, orig_i, i);
+			rx_ring->stats.xdp_drops++;
 			break;
 		case XDP_PASS:
 			rxbd = orig_rxbd;
···
 			break;
 		case XDP_TX:
 			tx_ring = priv->xdp_tx_ring[rx_ring->index];
+			if (unlikely(test_bit(ENETC_TX_DOWN, &priv->flags))) {
+				enetc_xdp_drop(rx_ring, orig_i, i);
+				tx_ring->stats.xdp_tx_drops++;
+				break;
+			}
+
 			xdp_tx_bd_cnt = enetc_rx_swbd_to_xdp_tx_swbd(xdp_tx_arr,
 								     rx_ring,
 								     orig_i, i);
···
 	enetc_rxbdr_wr(hw, idx, ENETC_RBMR, rbmr);
 }
 
-static void enetc_enable_bdrs(struct enetc_ndev_priv *priv)
+static void enetc_enable_rx_bdrs(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	int i;
+
+	for (i = 0; i < priv->num_rx_rings; i++)
+		enetc_enable_rxbdr(hw, priv->rx_ring[i]);
+}
+
+static void enetc_enable_tx_bdrs(struct enetc_ndev_priv *priv)
 {
 	struct enetc_hw *hw = &priv->si->hw;
 	int i;
 
 	for (i = 0; i < priv->num_tx_rings; i++)
 		enetc_enable_txbdr(hw, priv->tx_ring[i]);
-
-	for (i = 0; i < priv->num_rx_rings; i++)
-		enetc_enable_rxbdr(hw, priv->rx_ring[i]);
 }
 
 static void enetc_disable_rxbdr(struct enetc_hw *hw, struct enetc_bdr *rx_ring)
···
 	enetc_txbdr_wr(hw, idx, ENETC_TBMR, 0);
 }
 
-static void enetc_disable_bdrs(struct enetc_ndev_priv *priv)
+static void enetc_disable_rx_bdrs(struct enetc_ndev_priv *priv)
+{
+	struct enetc_hw *hw = &priv->si->hw;
+	int i;
+
+	for (i = 0; i < priv->num_rx_rings; i++)
+		enetc_disable_rxbdr(hw, priv->rx_ring[i]);
+}
+
+static void enetc_disable_tx_bdrs(struct enetc_ndev_priv *priv)
 {
 	struct enetc_hw *hw = &priv->si->hw;
 	int i;
 
 	for (i = 0; i < priv->num_tx_rings; i++)
 		enetc_disable_txbdr(hw, priv->tx_ring[i]);
-
-	for (i = 0; i < priv->num_rx_rings; i++)
-		enetc_disable_rxbdr(hw, priv->rx_ring[i]);
 }
 
 static void enetc_wait_txbdr(struct enetc_hw *hw, struct enetc_bdr *tx_ring)
···
 		enable_irq(irq);
 	}
 
-	enetc_enable_bdrs(priv);
+	enetc_enable_tx_bdrs(priv);
+
+	enetc_enable_rx_bdrs(priv);
 
 	netif_tx_start_all_queues(ndev);
+
+	clear_bit(ENETC_TX_DOWN, &priv->flags);
 }
 EXPORT_SYMBOL_GPL(enetc_start);
 
···
 	struct enetc_ndev_priv *priv = netdev_priv(ndev);
 	int i;
 
+	set_bit(ENETC_TX_DOWN, &priv->flags);
+
 	netif_tx_stop_all_queues(ndev);
 
-	enetc_disable_bdrs(priv);
+	enetc_disable_rx_bdrs(priv);
+
+	enetc_wait_bdrs(priv);
+
+	enetc_disable_tx_bdrs(priv);
 
 	for (i = 0; i < priv->bdr_int_num; i++) {
 		int irq = pci_irq_vector(priv->si->pdev,
···
 		napi_synchronize(&priv->int_vector[i]->napi);
 		napi_disable(&priv->int_vector[i]->napi);
 	}
-
-	enetc_wait_bdrs(priv);
 
 	enetc_clear_interrupts(priv);
 }

drivers/net/ethernet/freescale/enetc/enetc.h

···
 
 enum enetc_flags_bit {
 	ENETC_TX_ONESTEP_TSTAMP_IN_PROGRESS = 0,
+	ENETC_TX_DOWN,
 };
 
 /* interrupt coalescing modes */

drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c

···
 				 NIX_AF_TL3_TL2X_LINKX_CFG(tl2_tl3_link_schq, link));
 		if (!(cfg & BIT_ULL(12)))
 			continue;
-		bmap |= (1 << i);
+		bmap |= BIT_ULL(i);
 		cfg &= ~BIT_ULL(12);
 		rvu_write64(rvu, blkaddr,
 			    NIX_AF_TL3_TL2X_LINKX_CFG(tl2_tl3_link_schq, link), cfg);
···
 
 	/* Set NIX_AF_TL3_TL2_LINKX_CFG[ENA] for the TL3/TL2 queue */
 	for (i = 0; i < (rvu->hw->cgx_links + rvu->hw->lbk_links); i++) {
-		if (!(bmap & (1 << i)))
+		if (!(bmap & BIT_ULL(i)))
 			continue;
 		cfg = rvu_read64(rvu, blkaddr,
 				 NIX_AF_TL3_TL2X_LINKX_CFG(tl2_tl3_link_schq, link));

drivers/net/ethernet/mediatek/mtk_eth_soc.c

···
 		if (unlikely(dma_mapping_error(eth->dma_dev, dma_addr)))
 			return -ENOMEM;
 
-		for (i = 0; i < cnt; i++) {
+		for (i = 0; i < len; i++) {
 			struct mtk_tx_dma_v2 *txd;
 
 			txd = eth->scratch_ring + (j * MTK_FQ_DMA_LENGTH + i) * soc->tx.desc_size;

drivers/net/ethernet/mellanox/mlx5/core/cmd.c

···
 	}
 }
 
+#define MLX5_MAX_MANAGE_PAGES_CMD_ENT 1
+#define MLX5_CMD_MASK ((1UL << (cmd->vars.max_reg_cmds + \
+			   MLX5_MAX_MANAGE_PAGES_CMD_ENT)) - 1)
+
 static void mlx5_cmd_trigger_completions(struct mlx5_core_dev *dev)
 {
 	struct mlx5_cmd *cmd = &dev->cmd;
···
 	/* wait for pending handlers to complete */
 	mlx5_eq_synchronize_cmd_irq(dev);
 	spin_lock_irqsave(&dev->cmd.alloc_lock, flags);
-	vector = ~dev->cmd.vars.bitmask & ((1ul << (1 << dev->cmd.vars.log_sz)) - 1);
+	vector = ~dev->cmd.vars.bitmask & MLX5_CMD_MASK;
 	if (!vector)
 		goto no_trig;
 
···
 
 	cmd->state = MLX5_CMDIF_STATE_DOWN;
 	cmd->vars.max_reg_cmds = (1 << cmd->vars.log_sz) - 1;
-	cmd->vars.bitmask = (1UL << cmd->vars.max_reg_cmds) - 1;
+	cmd->vars.bitmask = MLX5_CMD_MASK;
 
 	sema_init(&cmd->vars.sem, cmd->vars.max_reg_cmds);
 	sema_init(&cmd->vars.pages_sem, 1);

drivers/net/ethernet/mellanox/mlx5/core/en_main.c

···
 	mlx5e_dcbnl_delete_app(priv);
 	unregister_netdev(priv->netdev);
 	_mlx5e_suspend(adev, false);
-	priv->profile->cleanup(priv);
+	/* Avoid cleanup if profile rollback failed. */
+	if (priv->profile)
+		priv->profile->cleanup(priv);
 	mlx5e_destroy_netdev(priv);
 	mlx5e_devlink_port_unregister(mlx5e_dev);
 	mlx5e_destroy_devlink(mlx5e_dev);

drivers/net/ethernet/mellanox/mlx5/core/eq.c

···
 	struct mlx5_eq_comp *eq;
 	int ret = 0;
 
+	if (vecidx >= table->max_comp_eqs) {
+		mlx5_core_dbg(dev, "Requested vector index %u should be less than %u",
+			      vecidx, table->max_comp_eqs);
+		return -EINVAL;
+	}
+
 	mutex_lock(&table->comp_lock);
 	eq = xa_load(&table->comp_eqs, vecidx);
 	if (eq) {

drivers/net/ethernet/mellanox/mlx5/core/steering/hws/mlx5hws_bwc.c

···
 static int
 hws_bwc_matcher_rehash_size(struct mlx5hws_bwc_matcher *bwc_matcher)
 {
-	u32 num_of_rules;
 	int ret;
 
 	/* If the current matcher size is already at its max size, we can't
···
 	 * Need to check again if we really need rehash.
 	 * If the reason for rehash was size, but not any more - skip rehash.
 	 */
-	num_of_rules = __atomic_load_n(&bwc_matcher->num_of_rules, __ATOMIC_RELAXED);
-	if (!hws_bwc_matcher_rehash_size_needed(bwc_matcher, num_of_rules))
+	if (!hws_bwc_matcher_rehash_size_needed(bwc_matcher, bwc_matcher->num_of_rules))
 		return 0;
 
 	/* Now we're done all the checking - do the rehash:

drivers/net/ethernet/mellanox/mlx5/core/steering/hws/mlx5hws_context.h

···
 	struct mlx5hws_send_engine *send_queue;
 	size_t queues;
 	struct mutex *bwc_send_queue_locks; /* protect BWC queues */
+	struct lock_class_key *bwc_lock_class_keys;
 	struct list_head tbl_list;
 	struct mlx5hws_context_debug_info debug_info;
 	struct xarray peer_ctx_xa;

drivers/net/ethernet/mellanox/mlx5/core/steering/hws/mlx5hws_definer.c

···
 	ret = hws_definer_conv_match_params_to_hl(ctx, mt, match_hl);
 	if (ret) {
 		mlx5hws_err(ctx, "Failed to convert items to header layout\n");
-		goto free_fc;
+		goto free_match_hl;
 	}
 
 	/* Find the match definer layout for header layout match union */
···
 
 free_fc:
 	kfree(mt->fc);
-
+free_match_hl:
 	kfree(match_hl);
 	return ret;
 }

drivers/net/ethernet/mellanox/mlx5/core/steering/hws/mlx5hws_send.c

···
 
 static void hws_send_queues_bwc_locks_destroy(struct mlx5hws_context *ctx)
 {
-	int bwc_queues = ctx->queues - 1;
+	int bwc_queues = mlx5hws_bwc_queues(ctx);
 	int i;
 
 	if (!mlx5hws_context_bwc_supported(ctx))
 		return;
 
-	for (i = 0; i < bwc_queues; i++)
+	for (i = 0; i < bwc_queues; i++) {
 		mutex_destroy(&ctx->bwc_send_queue_locks[i]);
+		lockdep_unregister_key(ctx->bwc_lock_class_keys + i);
+	}
+
+	kfree(ctx->bwc_lock_class_keys);
 	kfree(ctx->bwc_send_queue_locks);
 }
 
···
 	if (!ctx->bwc_send_queue_locks)
 		return -ENOMEM;
 
-	for (i = 0; i < bwc_queues; i++)
+	ctx->bwc_lock_class_keys = kcalloc(bwc_queues,
+					   sizeof(*ctx->bwc_lock_class_keys),
+					   GFP_KERNEL);
+	if (!ctx->bwc_lock_class_keys)
+		goto err_lock_class_keys;
+
+	for (i = 0; i < bwc_queues; i++) {
 		mutex_init(&ctx->bwc_send_queue_locks[i]);
+		lockdep_register_key(ctx->bwc_lock_class_keys + i);
+	}
 
 	return 0;
+
+err_lock_class_keys:
+	kfree(ctx->bwc_send_queue_locks);
+	return -ENOMEM;
 }
 
 int mlx5hws_send_queues_open(struct mlx5hws_context *ctx,

drivers/net/ethernet/microchip/lan743x_ptp.c

···
 	u32 nano_seconds = 0;
 	u32 seconds = 0;
 
-	if (ts) {
-		if (ts->tv_sec > 0xFFFFFFFFLL ||
-		    ts->tv_sec < 0) {
-			netif_warn(adapter, drv, adapter->netdev,
-				   "ts->tv_sec out of range, %lld\n",
-				   ts->tv_sec);
-			return -ERANGE;
-		}
-		if (ts->tv_nsec >= 1000000000L ||
-		    ts->tv_nsec < 0) {
-			netif_warn(adapter, drv, adapter->netdev,
-				   "ts->tv_nsec out of range, %ld\n",
-				   ts->tv_nsec);
-			return -ERANGE;
-		}
-		seconds = ts->tv_sec;
-		nano_seconds = ts->tv_nsec;
-		lan743x_ptp_clock_set(adapter, seconds, nano_seconds, 0);
-	} else {
-		netif_warn(adapter, drv, adapter->netdev, "ts == NULL\n");
-		return -EINVAL;
+	if (ts->tv_sec > 0xFFFFFFFFLL) {
+		netif_warn(adapter, drv, adapter->netdev,
+			   "ts->tv_sec out of range, %lld\n",
+			   ts->tv_sec);
+		return -ERANGE;
 	}
+	if (ts->tv_nsec < 0) {
+		netif_warn(adapter, drv, adapter->netdev,
+			   "ts->tv_nsec out of range, %ld\n",
+			   ts->tv_nsec);
+		return -ERANGE;
+	}
+	seconds = ts->tv_sec;
+	nano_seconds = ts->tv_nsec;
+	lan743x_ptp_clock_set(adapter, seconds, nano_seconds, 0);
 
 	return 0;
 }

drivers/net/ethernet/microchip/sparx5/sparx5_mirror.c

···
 /* Add port to mirror (only front ports) */
 static void sparx5_mirror_port_add(struct sparx5 *sparx5, u32 idx, u32 portno)
 {
-	u32 val, reg = portno;
+	u64 reg = portno;
+	u32 val;
 
-	reg = portno / BITS_PER_BYTE;
-	val = BIT(portno % BITS_PER_BYTE);
+	val = BIT(do_div(reg, 32));
 
 	if (reg == 0)
 		return spx5_rmw(val, val, sparx5, ANA_AC_PROBE_PORT_CFG(idx));
···
 /* Delete port from mirror (only front ports) */
 static void sparx5_mirror_port_del(struct sparx5 *sparx5, u32 idx, u32 portno)
 {
-	u32 val, reg = portno;
+	u64 reg = portno;
+	u32 val;
 
-	reg = portno / BITS_PER_BYTE;
-	val = BIT(portno % BITS_PER_BYTE);
+	val = BIT(do_div(reg, 32));
 
 	if (reg == 0)
 		return spx5_rmw(0, val, sparx5, ANA_AC_PROBE_PORT_CFG(idx));

drivers/net/ethernet/microchip/vcap/vcap_api_kunit.c

···
 
 	ret = vcap_del_rule(&test_vctrl, &test_netdev, id);
 	KUNIT_EXPECT_EQ(test, 0, ret);
+
+	vcap_free_rule(rule);
 }
 
 static void vcap_api_set_rule_counter_test(struct kunit *test)

drivers/net/ethernet/renesas/ravb_main.c

···
 	struct ravb_private *priv = netdev_priv(ndev);
 	const struct ravb_hw_info *hw_info = priv->info;
 
-	info->so_timestamping =
-		SOF_TIMESTAMPING_TX_SOFTWARE |
-		SOF_TIMESTAMPING_TX_HARDWARE |
-		SOF_TIMESTAMPING_RX_HARDWARE |
-		SOF_TIMESTAMPING_RAW_HARDWARE;
-	info->tx_types = (1 << HWTSTAMP_TX_OFF) | (1 << HWTSTAMP_TX_ON);
-	info->rx_filters =
-		(1 << HWTSTAMP_FILTER_NONE) |
-		(1 << HWTSTAMP_FILTER_PTP_V2_L2_EVENT) |
-		(1 << HWTSTAMP_FILTER_ALL);
-	if (hw_info->gptp || hw_info->ccc_gac)
+	if (hw_info->gptp || hw_info->ccc_gac) {
+		info->so_timestamping =
+			SOF_TIMESTAMPING_TX_SOFTWARE |
+			SOF_TIMESTAMPING_TX_HARDWARE |
+			SOF_TIMESTAMPING_RX_HARDWARE |
+			SOF_TIMESTAMPING_RAW_HARDWARE;
+		info->tx_types = (1 << HWTSTAMP_TX_OFF) | (1 << HWTSTAMP_TX_ON);
+		info->rx_filters =
+			(1 << HWTSTAMP_FILTER_NONE) |
+			(1 << HWTSTAMP_FILTER_PTP_V2_L2_EVENT) |
+			(1 << HWTSTAMP_FILTER_ALL);
 		info->phc_index = ptp_clock_index(priv->ptp.clock);
-	else
-		info->phc_index = 0;
+	}
 
 	return 0;
 }

drivers/net/ethernet/renesas/rtsn.c

···
 	if (skb->len >= TX_DS) {
 		priv->stats.tx_dropped++;
 		priv->stats.tx_errors++;
+		dev_kfree_skb_any(skb);
 		goto out;
 	}
 

drivers/net/ethernet/stmicro/stmmac/dwmac-tegra.c

···
 	value &= ~XPCS_WRAP_UPHY_RX_CONTROL_AUX_RX_IDDQ;
 	writel(value, mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
 
+	usleep_range(10, 20);  /* 50ns min delay needed as per HW design */
 	value = readl(mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
 	value &= ~XPCS_WRAP_UPHY_RX_CONTROL_RX_SLEEP;
 	writel(value, mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
 
+	usleep_range(10, 20);  /* 500ns min delay needed as per HW design */
 	value = readl(mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
 	value |= XPCS_WRAP_UPHY_RX_CONTROL_RX_CAL_EN;
 	writel(value, mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
···
 		return err;
 	}
 
+	usleep_range(10, 20);  /* 50ns min delay needed as per HW design */
 	value = readl(mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
 	value |= XPCS_WRAP_UPHY_RX_CONTROL_RX_DATA_EN;
 	writel(value, mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
 
 	value = readl(mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
+	value &= ~XPCS_WRAP_UPHY_RX_CONTROL_RX_PCS_PHY_RDY;
+	writel(value, mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
+
+	usleep_range(10, 20);  /* 50ns min delay needed as per HW design */
+	value = readl(mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
 	value |= XPCS_WRAP_UPHY_RX_CONTROL_RX_CDR_RESET;
 	writel(value, mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
 
-	value = readl(mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
-	value &= ~XPCS_WRAP_UPHY_RX_CONTROL_RX_CDR_RESET;
-	writel(value, mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
-
+	usleep_range(10, 20);  /* 50ns min delay needed as per HW design */
 	value = readl(mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
 	value |= XPCS_WRAP_UPHY_RX_CONTROL_RX_PCS_PHY_RDY;
+	writel(value, mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
+
+	msleep(30);  /* 30ms delay needed as per HW design */
+	value = readl(mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
+	value &= ~XPCS_WRAP_UPHY_RX_CONTROL_RX_CDR_RESET;
 	writel(value, mgbe->xpcs + XPCS_WRAP_UPHY_RX_CONTROL);
 
 	err = readl_poll_timeout(mgbe->xpcs + XPCS_WRAP_IRQ_STATUS, value,

drivers/net/ethernet/xilinx/xilinx_axienet_main.c

···
 		if (net_ratelimit())
 			netdev_err(ndev, "TX DMA mapping error\n");
 		ndev->stats.tx_dropped++;
+		dev_kfree_skb_any(skb);
 		return NETDEV_TX_OK;
 	}
 	desc_set_phys_addr(lp, phys, cur_p);
···
 			ndev->stats.tx_dropped++;
 			axienet_free_tx_chain(lp, orig_tail_ptr, ii + 1,
 					      true, NULL, 0);
+			dev_kfree_skb_any(skb);
 			return NETDEV_TX_OK;
 		}
 		desc_set_phys_addr(lp, phys, cur_p);

drivers/net/macsec.c

···
 	return sa;
 }
 
-static struct macsec_rx_sa *macsec_active_rxsa_get(struct macsec_rx_sc *rx_sc)
-{
-	struct macsec_rx_sa *sa = NULL;
-	int an;
-
-	for (an = 0; an < MACSEC_NUM_AN; an++)	{
-		sa = macsec_rxsa_get(rx_sc->sa[an]);
-		if (sa)
-			break;
-	}
-	return sa;
-}
-
 static void free_rx_sc_rcu(struct rcu_head *head)
 {
 	struct macsec_rx_sc *rx_sc = container_of(head, struct macsec_rx_sc, rcu_head);
···
 		/* If validateFrames is Strict or the C bit in the
 		 * SecTAG is set, discard
 		 */
-		struct macsec_rx_sa *active_rx_sa = macsec_active_rxsa_get(rx_sc);
 		if (hdr->tci_an & MACSEC_TCI_C ||
 		    secy->validate_frames == MACSEC_VALIDATE_STRICT) {
 			u64_stats_update_begin(&rxsc_stats->syncp);
 			rxsc_stats->stats.InPktsNotUsingSA++;
 			u64_stats_update_end(&rxsc_stats->syncp);
 			DEV_STATS_INC(secy->netdev, rx_errors);
-			if (active_rx_sa)
-				this_cpu_inc(active_rx_sa->stats->InPktsNotUsingSA);
 			goto drop_nosa;
 		}
 
···
 		u64_stats_update_begin(&rxsc_stats->syncp);
 		rxsc_stats->stats.InPktsUnusedSA++;
 		u64_stats_update_end(&rxsc_stats->syncp);
-		if (active_rx_sa)
-			this_cpu_inc(active_rx_sa->stats->InPktsUnusedSA);
 		goto deliver;
 	}
 

drivers/net/mdio/mdio-bcm-unimac.c

···
 	{ .compatible = "brcm,asp-v2.2-mdio", },
 	{ .compatible = "brcm,asp-v2.1-mdio", },
 	{ .compatible = "brcm,asp-v2.0-mdio", },
+	{ .compatible = "brcm,bcm6846-mdio", },
 	{ .compatible = "brcm,genet-mdio-v5", },
 	{ .compatible = "brcm,genet-mdio-v4", },
 	{ .compatible = "brcm,genet-mdio-v3", },

drivers/net/netdevsim/dev.c

···
 	nsim_dev = nsim_trap_data->nsim_dev;
 
 	if (!devl_trylock(priv_to_devlink(nsim_dev))) {
-		schedule_delayed_work(&nsim_dev->trap_data->trap_report_dw, 1);
+		queue_delayed_work(system_unbound_wq,
+				   &nsim_dev->trap_data->trap_report_dw, 1);
 		return;
 	}
 
···
 			continue;
 
 		nsim_dev_trap_report(nsim_dev_port);
+		cond_resched();
 	}
 	devl_unlock(priv_to_devlink(nsim_dev));
-
-	schedule_delayed_work(&nsim_dev->trap_data->trap_report_dw,
-			      msecs_to_jiffies(NSIM_TRAP_REPORT_INTERVAL_MS));
+	queue_delayed_work(system_unbound_wq,
+			   &nsim_dev->trap_data->trap_report_dw,
+			   msecs_to_jiffies(NSIM_TRAP_REPORT_INTERVAL_MS));
 }
 
 static int nsim_dev_traps_init(struct devlink *devlink)
···
 
 	INIT_DELAYED_WORK(&nsim_dev->trap_data->trap_report_dw,
 			  nsim_dev_trap_report_work);
-	schedule_delayed_work(&nsim_dev->trap_data->trap_report_dw,
-			      msecs_to_jiffies(NSIM_TRAP_REPORT_INTERVAL_MS));
+	queue_delayed_work(system_unbound_wq,
+			   &nsim_dev->trap_data->trap_report_dw,
+			   msecs_to_jiffies(NSIM_TRAP_REPORT_INTERVAL_MS));
 
 	return 0;
 

drivers/net/usb/usbnet.c

···
 	 * may trigger an error resubmitting itself and, worse,
 	 * schedule a timer. So we kill it all just in case.
 	 */
+	usbnet_mark_going_away(dev);
 	cancel_work_sync(&dev->kevent);
 	del_timer_sync(&dev->delay);
 	free_netdev(net);

drivers/net/vmxnet3/vmxnet3_xdp.c

···
 	} else { /* XDP buffer from page pool */
 		page = virt_to_page(xdpf->data);
 		tbi->dma_addr = page_pool_get_dma_addr(page) +
-				VMXNET3_XDP_HEADROOM;
+				(xdpf->data - (void *)xdpf);
 		dma_sync_single_for_device(&adapter->pdev->dev,
 					   tbi->dma_addr, buf_size,
 					   DMA_TO_DEVICE);

drivers/of/of_kunit_helpers.c

···
 #include <kunit/test.h>
 #include <kunit/resource.h>
 
+#include "of_private.h"
+
+/**
+ * of_root_kunit_skip() - Skip test if the root node isn't populated
+ * @test: test to skip if the root node isn't populated
+ */
+void of_root_kunit_skip(struct kunit *test)
+{
+	if (IS_ENABLED(CONFIG_ARM64) && IS_ENABLED(CONFIG_ACPI) && !of_root)
+		kunit_skip(test, "arm64+acpi doesn't populate a root node");
+}
+EXPORT_SYMBOL_GPL(of_root_kunit_skip);
+
 #if defined(CONFIG_OF_OVERLAY) && defined(CONFIG_OF_EARLY_FLATTREE)
 
 static void of_overlay_fdt_apply_kunit_exit(void *ovcs_id)
···
 {
 	int ret;
 	int *copy_id;
+
+	of_root_kunit_skip(test);
 
 	copy_id = kunit_kmalloc(test, sizeof(*copy_id), GFP_KERNEL);
 	if (!copy_id)

drivers/of/of_private.h

···
 extern struct list_head aliases_lookup;
 extern struct kset *of_kset;
 
+struct kunit;
+extern void of_root_kunit_skip(struct kunit *test);
+
 #if defined(CONFIG_OF_DYNAMIC)
 extern int of_property_notify(int action, struct device_node *np,
 			      struct property *prop, struct property *old_prop);

drivers/of/of_test.c

···
 
 #include <kunit/test.h>
 
+#include "of_private.h"
+
 /*
  * Test that the root node "/" can be found by path.
  */
···
 
 static int of_dtb_test_init(struct kunit *test)
 {
+	of_root_kunit_skip(test);
 	if (!IS_ENABLED(CONFIG_OF_EARLY_FLATTREE))
 		kunit_skip(test, "requires CONFIG_OF_EARLY_FLATTREE");
 

drivers/of/overlay_test.c

···
 #include <kunit/of.h>
 #include <kunit/test.h>
 
+#include "of_private.h"
+
 static const char * const kunit_node_name = "kunit-test";
 static const char * const kunit_compatible = "test,empty";
 
···
 	struct device *dev;
 	struct device_node *np;
 
+	of_root_kunit_skip(test);
 	if (!IS_ENABLED(CONFIG_OF_EARLY_FLATTREE))
 		kunit_skip(test, "requires CONFIG_OF_EARLY_FLATTREE for root node");
 
···
 
 	np = of_find_node_by_name(NULL, kunit_node_name);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, np);
-	of_node_put_kunit(test, np);
+	of_node_put_kunit(&fake, np);
 
 	pdev = of_find_device_by_node(np);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, pdev);

drivers/opp/core.c

···
 
 	/* Attach genpds */
 	if (config->genpd_names) {
-		if (config->required_devs)
+		if (config->required_devs) {
+			ret = -EINVAL;
 			goto err;
+		}
 
 		ret = _opp_attach_genpd(opp_table, dev, config->genpd_names,
 					config->virt_devs);

drivers/pmdomain/qcom/cpr.c

···
 			of_parse_phandle(child_np, "required-opps", 0);
 
 		if (child_req_np == ref_np) {
-			u64 rate;
+			u64 rate = 0;
 
 			of_property_read_u64(child_np, "opp-hz", &rate);
 			return (unsigned long) rate;

drivers/powercap/intel_rapl_msr.c

···
 	X86_MATCH_VFM(INTEL_RAPTORLAKE_P, NULL),
 	X86_MATCH_VFM(INTEL_METEORLAKE, NULL),
 	X86_MATCH_VFM(INTEL_METEORLAKE_L, NULL),
+	X86_MATCH_VFM(INTEL_ARROWLAKE_U, NULL),
 	{}
 };
 

drivers/powercap/intel_rapl_tpmi.c

···
 #include <linux/module.h>
 #include <linux/slab.h>
 
-#define TPMI_RAPL_VERSION 1
+#define TPMI_RAPL_MAJOR_VERSION 0
+#define TPMI_RAPL_MINOR_VERSION 1
 
 /* 1 header + 10 registers + 5 reserved. 8 bytes for each. */
 #define TPMI_RAPL_DOMAIN_SIZE 128
···
 	tpmi_domain_size = tpmi_domain_header >> 16 & 0xff;
 	tpmi_domain_flags = tpmi_domain_header >> 32 & 0xffff;
 
-	if (tpmi_domain_version != TPMI_RAPL_VERSION) {
-		pr_warn(FW_BUG "Unsupported version:%d\n", tpmi_domain_version);
+	if (tpmi_domain_version == TPMI_VERSION_INVALID) {
+		pr_warn(FW_BUG "Invalid version\n");
 		return -ENODEV;
 	}
+
+	if (TPMI_MAJOR_VERSION(tpmi_domain_version) != TPMI_RAPL_MAJOR_VERSION) {
+		pr_warn(FW_BUG "Unsupported major version:%ld\n",
+			TPMI_MAJOR_VERSION(tpmi_domain_version));
+		return -ENODEV;
+	}
+
+	if (TPMI_MINOR_VERSION(tpmi_domain_version) > TPMI_RAPL_MINOR_VERSION)
+		pr_info("Ignore: Unsupported minor version:%ld\n",
+			TPMI_MINOR_VERSION(tpmi_domain_version));
 
 	/* Domain size: in unit of 128 Bytes */
 	if (tpmi_domain_size != 1) {
···
 			pr_warn(FW_BUG "System domain must support Domain Info register\n");
 			return -ENODEV;
 		}
-		tpmi_domain_info = readq(trp->base + offset + TPMI_RAPL_REG_DOMAIN_INFO);
+		tpmi_domain_info = readq(trp->base + offset + TPMI_RAPL_REG_DOMAIN_INFO * 8);
 		if (!(tpmi_domain_info & TPMI_RAPL_DOMAIN_ROOT))
 			return 0;
 		domain_type = RAPL_DOMAIN_PLATFORM;

drivers/scsi/fnic/fnic_main.c

···
 		spin_lock_init(&fnic->vlans_lock);
 		INIT_WORK(&fnic->fip_frame_work, fnic_handle_fip_frame);
 		INIT_WORK(&fnic->event_work, fnic_handle_event);
-		INIT_WORK(&fnic->flush_work, fnic_flush_tx);
 		skb_queue_head_init(&fnic->fip_frame_queue);
 		INIT_LIST_HEAD(&fnic->evlist);
 		INIT_LIST_HEAD(&fnic->vlans);
···
 
 	INIT_WORK(&fnic->link_work, fnic_handle_link);
 	INIT_WORK(&fnic->frame_work, fnic_handle_frame);
+	INIT_WORK(&fnic->flush_work, fnic_flush_tx);
 	skb_queue_head_init(&fnic->frame_queue);
 	skb_queue_head_init(&fnic->tx_queue);
 

drivers/scsi/scsi_transport_fc.c

···
 		 */
 		if (rport->port_state == FC_PORTSTATE_ONLINE)
 			rport->port_state = port_state;
-		else
+		else if (port_state != rport->port_state)
 			return -EINVAL;
 	} else if (port_state == FC_PORTSTATE_ONLINE) {
 		/*
···
 		 */
 		if (rport->port_state == FC_PORTSTATE_MARGINAL)
 			rport->port_state = port_state;
-		else
+		else if (port_state != rport->port_state)
 			return -EINVAL;
 	} else
 		return -EINVAL;

drivers/scsi/wd33c93.c

···
 		/* construct an IDENTIFY message with correct disconnect bit */
 
 		hostdata->outgoing_msg[0] = IDENTIFY(0, cmd->device->lun);
-		if (scsi_pointer->phase)
+		if (WD33C93_scsi_pointer(cmd)->phase)
 			hostdata->outgoing_msg[0] |= 0x40;
 
 		if (hostdata->sync_stat[cmd->device->id] == SS_FIRST) {

drivers/target/target_core_user.c

···
 	}
 
 	ret = genlmsg_multicast_allns(&tcmu_genl_family, skb, 0,
-				      TCMU_MCGRP_CONFIG, GFP_KERNEL);
+				      TCMU_MCGRP_CONFIG);
 
 	/* Wait during an add as the listener may not be up yet */
 	if (ret == 0 ||

drivers/thermal/intel/int340x_thermal/processor_thermal_device_pci.c

···
 	if (!pci_info->no_legacy)
 		proc_thermal_remove(proc_priv);
 	proc_thermal_mmio_remove(pdev, proc_priv);
-	pci_disable_device(pdev);
 
 	return ret;
 }
···
 	proc_thermal_mmio_remove(pdev, pci_info->proc_priv);
 	if (!pci_info->no_legacy)
 		proc_thermal_remove(proc_priv);
-	pci_disable_device(pdev);
 }
 
 #ifdef CONFIG_PM_SLEEP

drivers/thermal/intel/int340x_thermal/processor_thermal_rapl.c

···
 
 static const struct rapl_mmio_regs rapl_mmio_default = {
 	.reg_unit = 0x5938,
-	.regs[RAPL_DOMAIN_PACKAGE] = { 0x59a0, 0x593c, 0x58f0, 0, 0x5930},
+	.regs[RAPL_DOMAIN_PACKAGE] = { 0x59a0, 0x593c, 0x58f0, 0, 0x5930, 0x59b0},
 	.regs[RAPL_DOMAIN_DRAM] = { 0x58e0, 0x58e8, 0x58ec, 0, 0},
-	.limits[RAPL_DOMAIN_PACKAGE] = BIT(POWER_LIMIT2),
+	.limits[RAPL_DOMAIN_PACKAGE] = BIT(POWER_LIMIT2) | BIT(POWER_LIMIT4),
 	.limits[RAPL_DOMAIN_DRAM] = BIT(POWER_LIMIT2),
 };
-
-static int rapl_mmio_cpu_online(unsigned int cpu)
-{
-	struct rapl_package *rp;
-
-	/* mmio rapl supports package 0 only for now */
-	if (topology_physical_package_id(cpu))
-		return 0;
-
-	rp = rapl_find_package_domain_cpuslocked(cpu, &rapl_mmio_priv, true);
-	if (!rp) {
-		rp = rapl_add_package_cpuslocked(cpu, &rapl_mmio_priv, true);
-		if (IS_ERR(rp))
-			return PTR_ERR(rp);
-	}
-	cpumask_set_cpu(cpu, &rp->cpumask);
-	return 0;
-}
-
-static int rapl_mmio_cpu_down_prep(unsigned int cpu)
-{
-	struct rapl_package *rp;
-	int lead_cpu;
-
-	rp = rapl_find_package_domain_cpuslocked(cpu, &rapl_mmio_priv, true);
-	if (!rp)
-		return 0;
-
-	cpumask_clear_cpu(cpu, &rp->cpumask);
-	lead_cpu = cpumask_first(&rp->cpumask);
-	if (lead_cpu >= nr_cpu_ids)
-		rapl_remove_package_cpuslocked(rp);
-	else if (rp->lead_cpu == cpu)
-		rp->lead_cpu = lead_cpu;
-	return 0;
-}
 
 static int rapl_mmio_read_raw(int cpu, struct reg_action *ra)
 {
···
 int proc_thermal_rapl_add(struct pci_dev *pdev, struct proc_thermal_device *proc_priv)
 {
 	const struct rapl_mmio_regs *rapl_regs = &rapl_mmio_default;
+	struct rapl_package *rp;
 	enum rapl_domain_reg_id reg;
 	enum rapl_domain_type domain;
 	int ret;
···
 		return PTR_ERR(rapl_mmio_priv.control_type);
 	}
 
-	ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "powercap/rapl:online",
-				rapl_mmio_cpu_online, rapl_mmio_cpu_down_prep);
-	if (ret < 0) {
-		powercap_unregister_control_type(rapl_mmio_priv.control_type);
-		rapl_mmio_priv.control_type = NULL;
-		return ret;
+	/* Register a RAPL package device for package 0 which is always online */
+	rp = rapl_find_package_domain(0, &rapl_mmio_priv, false);
+	if (rp) {
+		ret = -EEXIST;
+		goto err;
 	}
-	rapl_mmio_priv.pcap_rapl_online = ret;
+
+	rp = rapl_add_package(0, &rapl_mmio_priv, false);
+	if (IS_ERR(rp)) {
+		ret = PTR_ERR(rp);
+		goto err;
+	}
 
 	return 0;
+
+err:
+	powercap_unregister_control_type(rapl_mmio_priv.control_type);
+	rapl_mmio_priv.control_type = NULL;
+	return ret;
 }
 EXPORT_SYMBOL_GPL(proc_thermal_rapl_add);
 
 void proc_thermal_rapl_remove(void)
 {
+	struct rapl_package *rp;
+
 	if (IS_ERR_OR_NULL(rapl_mmio_priv.control_type))
 		return;
 
-	cpuhp_remove_state(rapl_mmio_priv.pcap_rapl_online);
+	rp = rapl_find_package_domain(0, &rapl_mmio_priv, false);
+	if (rp)
+		rapl_remove_package(rp);
 	powercap_unregister_control_type(rapl_mmio_priv.control_type);
 }
 EXPORT_SYMBOL_GPL(proc_thermal_rapl_remove);

drivers/thermal/thermal_core.c

···
 	mutex_lock(&thermal_list_lock);
 	list_for_each_entry(tz, &thermal_tz_list, node) {
 		if (tz->id == id) {
+			get_device(&tz->device);
 			match = tz;
 			break;
 		}
···
 	ida_destroy(&tz->ida);
 
 	device_del(&tz->device);
-
-	kfree(tz->tzp);
-
 	put_device(&tz->device);
 
 	thermal_notify_tz_delete(tz);
 
 	wait_for_completion(&tz->removal);
+	kfree(tz->tzp);
 	kfree(tz);
 }
 EXPORT_SYMBOL_GPL(thermal_zone_device_unregister);

drivers/thermal/thermal_core.h

···
 
 struct thermal_zone_device *thermal_zone_get_by_id(int id);
 
+DEFINE_CLASS(thermal_zone_get_by_id, struct thermal_zone_device *,
+	     if (_T) put_device(&_T->device), thermal_zone_get_by_id(id), int id)
+
 static inline bool cdev_is_power_actor(struct thermal_cooling_device *cdev)
 {
 	return cdev->ops->get_requested_power && cdev->ops->state2power &&

drivers/thermal/thermal_netlink.c

···
 {
 	struct sk_buff *msg = p->msg;
 	const struct thermal_trip_desc *td;
-	struct thermal_zone_device *tz;
 	struct nlattr *start_trip;
 	int id;
 
···
 
 	id = nla_get_u32(p->attrs[THERMAL_GENL_ATTR_TZ_ID]);
 
-	tz = thermal_zone_get_by_id(id);
+	CLASS(thermal_zone_get_by_id, tz)(id);
 	if (!tz)
 		return -EINVAL;
 
···
 static int thermal_genl_cmd_tz_get_temp(struct param *p)
 {
 	struct sk_buff *msg = p->msg;
-	struct thermal_zone_device *tz;
 	int temp, ret, id;
 
 	if (!p->attrs[THERMAL_GENL_ATTR_TZ_ID])
···
 
 	id = nla_get_u32(p->attrs[THERMAL_GENL_ATTR_TZ_ID]);
 
-	tz = thermal_zone_get_by_id(id);
+	CLASS(thermal_zone_get_by_id, tz)(id);
 	if (!tz)
 		return -EINVAL;
 
···
 static int thermal_genl_cmd_tz_get_gov(struct param *p)
 {
 	struct sk_buff *msg = p->msg;
-	struct thermal_zone_device *tz;
 	int id, ret = 0;
 
 	if (!p->attrs[THERMAL_GENL_ATTR_TZ_ID])
···
 
 	id = nla_get_u32(p->attrs[THERMAL_GENL_ATTR_TZ_ID]);
 
-	tz = thermal_zone_get_by_id(id);
+	CLASS(thermal_zone_get_by_id, tz)(id);
 	if (!tz)
 		return -EINVAL;
 

drivers/ufs/core/ufshcd.c

···
 	struct utp_transfer_req_desc *utrdlp = hba->utrdl_base_addr;
 	dma_addr_t cmd_desc_element_addr = hba->ucdl_dma_addr +
 		i * ufshcd_get_ucd_size(hba);
-	u16 response_offset = offsetof(struct utp_transfer_cmd_desc,
-				       response_upiu);
-	u16 prdt_offset = offsetof(struct utp_transfer_cmd_desc, prd_table);
+	u16 response_offset = le16_to_cpu(utrdlp[i].response_upiu_offset);
+	u16 prdt_offset = le16_to_cpu(utrdlp[i].prd_table_offset);
 
 	lrb->utr_descriptor_ptr = utrdlp + i;
 	lrb->utrd_dma_addr = hba->utrdl_dma_addr +

drivers/usb/dwc3/core.c

···
 int dwc3_event_buffers_setup(struct dwc3 *dwc)
 {
 	struct dwc3_event_buffer	*evt;
+	u32				reg;
 
 	if (!dwc->ev_buf)
 		return 0;
···
 			upper_32_bits(evt->dma));
 	dwc3_writel(dwc->regs, DWC3_GEVNTSIZ(0),
 			DWC3_GEVNTSIZ_SIZE(evt->length));
-	dwc3_writel(dwc->regs, DWC3_GEVNTCOUNT(0), 0);
 
+	/* Clear any stale event */
+	reg = dwc3_readl(dwc->regs, DWC3_GEVNTCOUNT(0));
+	dwc3_writel(dwc->regs, DWC3_GEVNTCOUNT(0), reg);
 	return 0;
 }
 
···
 	dwc3_writel(dwc->regs, DWC3_GEVNTADRHI(0), 0);
 	dwc3_writel(dwc->regs, DWC3_GEVNTSIZ(0), DWC3_GEVNTSIZ_INTMASK
 			| DWC3_GEVNTSIZ_SIZE(0));
-	dwc3_writel(dwc->regs, DWC3_GEVNTCOUNT(0), 0);
+
+	/* Clear any stale event */
+	reg = dwc3_readl(dwc->regs, DWC3_GEVNTCOUNT(0));
+	dwc3_writel(dwc->regs, DWC3_GEVNTCOUNT(0), reg);
 }
 
 static void dwc3_core_num_eps(struct dwc3 *dwc)
···
 
 	switch (dwc->current_dr_role) {
 	case DWC3_GCTL_PRTCAP_DEVICE:
-		dwc3_gadget_process_pending_events(dwc);
+		if (dwc->pending_events) {
+			pm_runtime_put(dwc->dev);
+			dwc->pending_events = false;
+			enable_irq(dwc->irq_gadget);
+		}
 		break;
 	case DWC3_GCTL_PRTCAP_HOST:
 	default:
···
 static int dwc3_resume(struct device *dev)
 {
 	struct dwc3	*dwc = dev_get_drvdata(dev);
-	int		ret;
+	int		ret = 0;
 
 	pinctrl_pm_select_default_state(dev);
 
···
 	pm_runtime_set_active(dev);
 
 	ret = dwc3_resume_common(dwc, PMSG_RESUME);
-	if (ret) {
+	if (ret)
 		pm_runtime_set_suspended(dev);
-		return ret;
-	}
 
 	pm_runtime_enable(dev);
 
-	return 0;
+	return ret;
 }
 
 static void dwc3_complete(struct device *dev)
···
 static const struct dev_pm_ops dwc3_dev_pm_ops = {
 	SET_SYSTEM_SLEEP_PM_OPS(dwc3_suspend, dwc3_resume)
 	.complete = dwc3_complete,
+
+	/*
+	 * Runtime suspend halts the controller on disconnection. It relies on
+	 * platforms with custom connection notification to start the controller
+	 * again.
+	 */
 	SET_RUNTIME_PM_OPS(dwc3_runtime_suspend, dwc3_runtime_resume,
 			dwc3_runtime_idle)
 };

drivers/usb/dwc3/core.h

···
 #if !IS_ENABLED(CONFIG_USB_DWC3_HOST)
 int dwc3_gadget_suspend(struct dwc3 *dwc);
 int dwc3_gadget_resume(struct dwc3 *dwc);
-void dwc3_gadget_process_pending_events(struct dwc3 *dwc);
 #else
 static inline int dwc3_gadget_suspend(struct dwc3 *dwc)
 {
···
 	return 0;
 }
 
-static inline void dwc3_gadget_process_pending_events(struct dwc3 *dwc)
-{
-}
 #endif /* !IS_ENABLED(CONFIG_USB_DWC3_HOST) */
 
 #if IS_ENABLED(CONFIG_USB_DWC3_ULPI)

drivers/usb/dwc3/gadget.c

···
 
 	return dwc3_gadget_soft_connect(dwc);
 }
-
-void dwc3_gadget_process_pending_events(struct dwc3 *dwc)
-{
-	if (dwc->pending_events) {
-		dwc3_interrupt(dwc->irq_gadget, dwc->ev_buf);
-		dwc3_thread_interrupt(dwc->irq_gadget, dwc->ev_buf);
-		pm_runtime_put(dwc->dev);
-		dwc->pending_events = false;
-		enable_irq(dwc->irq_gadget);
-	}
-}

drivers/usb/gadget/udc/core.c

···
 	driver->driver.bus = &gadget_bus_type;
 	driver->driver.owner = owner;
 	driver->driver.mod_name = mod_name;
+	driver->driver.probe_type = PROBE_FORCE_SYNCHRONOUS;
 	ret = driver_register(&driver->driver);
 	if (ret) {
 		pr_warn("%s: driver registration failed: %d\n",

drivers/usb/host/xhci-pci.c

···
 #define PCI_DEVICE_ID_ASMEDIA_1042A_XHCI		0x1142
 #define PCI_DEVICE_ID_ASMEDIA_1142_XHCI			0x1242
 #define PCI_DEVICE_ID_ASMEDIA_2142_XHCI			0x2142
+#define PCI_DEVICE_ID_ASMEDIA_3042_XHCI			0x3042
 #define PCI_DEVICE_ID_ASMEDIA_3242_XHCI			0x3242
 
 #define PCI_DEVICE_ID_CADENCE				0x17CD
···
 	if (pdev->vendor == PCI_VENDOR_ID_ASMEDIA &&
 		pdev->device == PCI_DEVICE_ID_ASMEDIA_1042A_XHCI)
 		xhci->quirks |= XHCI_ASMEDIA_MODIFY_FLOWCONTROL;
+
+	if (pdev->vendor == PCI_VENDOR_ID_ASMEDIA &&
+	    pdev->device == PCI_DEVICE_ID_ASMEDIA_3042_XHCI)
+		xhci->quirks |= XHCI_RESET_ON_RESUME;
 
 	if (pdev->vendor == PCI_VENDOR_ID_TI && pdev->device == 0x8241)
 		xhci->quirks |= XHCI_LIMIT_ENDPOINT_INTERVAL_7;

drivers/usb/misc/Kconfig

···
 	  this config will enable the driver and it will automatically
 	  match the state of the USB subsystem. If this driver is a
 	  module it will be called onboard_usb_dev.
+
+config USB_ONBOARD_DEV_USB5744
+	bool "Onboard USB Microchip usb5744 hub with SMBus support"
+	depends on (USB_ONBOARD_DEV && I2C=y) || (USB_ONBOARD_DEV=m && I2C=m)
+	help
+	  Say Y here if you want to support onboard USB Microchip usb5744
+	  hub that requires SMBus initialization.
+
+	  This options enables usb5744 i2c default initialization sequence
+	  during hub start-up configuration stage. It is must to enable this
+	  option on AMD Kria KR260 Robotics Starter Kit as this hub is
+	  connected to USB-SD converter which mounts the root filesystem.

drivers/usb/misc/onboard_usb_dev.c

···
 
 static int onboard_dev_5744_i2c_init(struct i2c_client *client)
 {
-#if IS_ENABLED(CONFIG_I2C)
+#if IS_ENABLED(CONFIG_USB_ONBOARD_DEV_USB5744)
 	struct device *dev = &client->dev;
 	int ret;
 
···
 
 	i2c_node = of_parse_phandle(pdev->dev.of_node, "i2c-bus", 0);
 	if (i2c_node) {
-		struct i2c_client *client;
+		struct i2c_client *client = NULL;
 
+#if IS_ENABLED(CONFIG_USB_ONBOARD_DEV_USB5744)
 		client = of_find_i2c_device_by_node(i2c_node);
+#endif
 		of_node_put(i2c_node);
 
 		if (!client) {

drivers/usb/misc/yurex.c

···
 #define YUREX_BUF_SIZE		8
 #define YUREX_WRITE_TIMEOUT	(HZ*2)
 
-#define MAX_S64_STRLEN 20 /* {-}922337203685477580{7,8} */
-
 /* table of devices that work with this driver */
 static struct usb_device_id yurex_table[] = {
 	{ USB_DEVICE(YUREX_VENDOR_ID, YUREX_PRODUCT_ID) },
···
 			  loff_t *ppos)
 {
 	struct usb_yurex *dev;
-	int len = 0;
-	char in_buffer[MAX_S64_STRLEN];
+	int len;
+	char in_buffer[20];
+	unsigned long flags;
 
 	dev = file->private_data;
 
···
 		return -ENODEV;
 	}
 
-	if (WARN_ON_ONCE(dev->bbu > S64_MAX || dev->bbu < S64_MIN)) {
-		mutex_unlock(&dev->io_mutex);
-		return -EIO;
-	}
-
-	spin_lock_irq(&dev->lock);
-	scnprintf(in_buffer, MAX_S64_STRLEN, "%lld\n", dev->bbu);
-	spin_unlock_irq(&dev->lock);
+	spin_lock_irqsave(&dev->lock, flags);
+	len = snprintf(in_buffer, 20, "%lld\n", dev->bbu);
+	spin_unlock_irqrestore(&dev->lock, flags);
 	mutex_unlock(&dev->io_mutex);
+
+	if (WARN_ON_ONCE(len >= sizeof(in_buffer)))
+		return -EIO;
 
 	return simple_read_from_buffer(buffer, count, ppos, in_buffer, len);
 }

drivers/usb/storage/unusual_devs.h

···
 		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
 		US_FL_NOT_LOCKABLE),
 
+/*
+ * Reported by Icenowy Zheng <[e-mail]>
+ * This is an interface for vendor-specific cryptic commands instead
+ * of real USB storage device.
+ */
+UNUSUAL_DEV(  0xe5b7, 0x0811, 0x0100, 0x0100,
+		"ZhuHai JieLi Technology",
+		"JieLi BR21",
+		USB_SC_DEVICE, USB_PR_DEVICE, NULL,
+		US_FL_IGNORE_DEVICE),
+
 /* Reported by Andrew Simmons <[e-mail]> */
 UNUSUAL_DEV(  0xed06, 0x4500, 0x0001, 0x0001,
 		"DataStor",

drivers/video/fbdev/amifb.c

···
  * triggers a section mismatch warning.
  */
 static struct platform_driver amifb_driver __refdata = {
-	.remove_new = __exit_p(amifb_remove),
-	.driver   = {
+	.remove = __exit_p(amifb_remove),
+	.driver = {
 		.name	= "amiga-video",
 	},
 };

drivers/video/fbdev/arcfb.c

···
 
 static struct platform_driver arcfb_driver = {
 	.probe	= arcfb_probe,
-	.remove_new = arcfb_remove,
+	.remove	= arcfb_remove,
 	.driver	= {
 		.name	= "arcfb",
 	},

drivers/video/fbdev/atmel_lcdfb.c

···
 
 static struct platform_driver atmel_lcdfb_driver = {
 	.probe		= atmel_lcdfb_probe,
-	.remove_new	= atmel_lcdfb_remove,
+	.remove		= atmel_lcdfb_remove,
 	.suspend	= atmel_lcdfb_suspend,
 	.resume		= atmel_lcdfb_resume,
 	.driver		= {

drivers/video/fbdev/au1100fb.c

···
 		.name		= "au1100-lcd",
 	},
 	.probe		= au1100fb_drv_probe,
-	.remove_new	= au1100fb_drv_remove,
+	.remove		= au1100fb_drv_remove,
 	.suspend	= au1100fb_drv_suspend,
 	.resume		= au1100fb_drv_resume,
 };

drivers/video/fbdev/au1200fb.c

···
 		.pm	= AU1200FB_PMOPS,
 	},
 	.probe		= au1200fb_drv_probe,
-	.remove_new	= au1200fb_drv_remove,
+	.remove		= au1200fb_drv_remove,
 };
 module_platform_driver(au1200fb_driver);
 

drivers/video/fbdev/broadsheetfb.c

···
 
 static struct platform_driver broadsheetfb_driver = {
 	.probe	= broadsheetfb_probe,
-	.remove_new = broadsheetfb_remove,
+	.remove	= broadsheetfb_remove,
 	.driver	= {
 		.name	= "broadsheetfb",
 	},

drivers/video/fbdev/bw2.c

···
 		.of_match_table = bw2_match,
 	},
 	.probe		= bw2_probe,
-	.remove_new	= bw2_remove,
+	.remove		= bw2_remove,
 };
 
 static int __init bw2_init(void)

drivers/video/fbdev/cg14.c

···
 		.of_match_table = cg14_match,
 	},
 	.probe		= cg14_probe,
-	.remove_new	= cg14_remove,
+	.remove		= cg14_remove,
 };
 
 static int __init cg14_init(void)

drivers/video/fbdev/cg3.c

···
 		.of_match_table = cg3_match,
 	},
 	.probe		= cg3_probe,
-	.remove_new	= cg3_remove,
+	.remove		= cg3_remove,
 };
 
 static int __init cg3_init(void)

drivers/video/fbdev/cg6.c

···
 		.of_match_table = cg6_match,
 	},
 	.probe		= cg6_probe,
-	.remove_new	= cg6_remove,
+	.remove		= cg6_remove,
 };
 
 static int __init cg6_init(void)

drivers/video/fbdev/clps711x-fb.c

···
 		.of_match_table	= clps711x_fb_dt_ids,
 	},
 	.probe	= clps711x_fb_probe,
-	.remove_new = clps711x_fb_remove,
+	.remove	= clps711x_fb_remove,
 };
 module_platform_driver(clps711x_fb_driver);
 

drivers/video/fbdev/cobalt_lcdfb.c

···
 
 static struct platform_driver cobalt_lcdfb_driver = {
 	.probe	= cobalt_lcdfb_probe,
-	.remove_new = cobalt_lcdfb_remove,
+	.remove	= cobalt_lcdfb_remove,
 	.driver	= {
 		.name	= "cobalt-lcd",
 	},

drivers/video/fbdev/da8xx-fb.c

···
 
 static struct platform_driver da8xx_fb_driver = {
 	.probe = fb_probe,
-	.remove_new = fb_remove,
+	.remove = fb_remove,
 	.driver = {
 		   .name = DRIVER_NAME,
 		   .pm	= &fb_pm_ops,

drivers/video/fbdev/ep93xx-fb.c

···
 
 static struct platform_driver ep93xxfb_driver = {
 	.probe		= ep93xxfb_probe,
-	.remove_new	= ep93xxfb_remove,
+	.remove		= ep93xxfb_remove,
 	.driver = {
 		.name	= "ep93xx-fb",
 	},

drivers/video/fbdev/ffb.c

···
 		.of_match_table = ffb_match,
 	},
 	.probe		= ffb_probe,
-	.remove_new	= ffb_remove,
+	.remove		= ffb_remove,
 };
 
 static int __init ffb_init(void)

drivers/video/fbdev/fsl-diu-fb.c

···
 MODULE_DEVICE_TABLE(of, fsl_diu_match);
 
 static struct platform_driver fsl_diu_driver = {
-	.driver = {
+	.driver		= {
 		.name = "fsl-diu-fb",
 		.of_match_table = fsl_diu_match,
 	},
-	.probe  	= fsl_diu_probe,
-	.remove_new 	= fsl_diu_remove,
+	.probe		= fsl_diu_probe,
+	.remove		= fsl_diu_remove,
 	.suspend	= fsl_diu_suspend,
 	.resume		= fsl_diu_resume,
 };

drivers/video/fbdev/gbefb.c

···
 
 static struct platform_driver gbefb_driver = {
 	.probe = gbefb_probe,
-	.remove_new = gbefb_remove,
-	.driver	= {
+	.remove = gbefb_remove,
+	.driver = {
 		.name = "gbefb",
-		.dev_groups	= gbefb_groups,
+		.dev_groups = gbefb_groups,
 	},
 };
 

drivers/video/fbdev/goldfishfb.c

···
 
 static struct platform_driver goldfish_fb_driver = {
 	.probe		= goldfish_fb_probe,
-	.remove_new	= goldfish_fb_remove,
+	.remove		= goldfish_fb_remove,
 	.driver = {
 		.name = "goldfish_fb",
 		.of_match_table = goldfish_fb_of_match,

drivers/video/fbdev/grvga.c

···
 		.of_match_table = svgactrl_of_match,
 	},
 	.probe		= grvga_probe,
-	.remove_new	= grvga_remove,
+	.remove		= grvga_remove,
 };
 
 module_platform_driver(grvga_driver);

drivers/video/fbdev/hecubafb.c

···
 
 static struct platform_driver hecubafb_driver = {
 	.probe	= hecubafb_probe,
-	.remove_new = hecubafb_remove,
+	.remove	= hecubafb_remove,
 	.driver	= {
 		.name	= "hecubafb",
 	},

drivers/video/fbdev/hgafb.c

···
 
 static struct platform_driver hgafb_driver = {
 	.probe = hgafb_probe,
-	.remove_new = hgafb_remove,
+	.remove = hgafb_remove,
 	.driver = {
 		.name = "hgafb",
 	},

drivers/video/fbdev/hitfb.c

···
 
 static struct platform_driver hitfb_driver = {
 	.probe		= hitfb_probe,
-	.remove_new	= hitfb_remove,
+	.remove		= hitfb_remove,
 	.driver		= {
 		.name	= "hitfb",
 		.pm	= &hitfb_dev_pm_ops,

drivers/video/fbdev/imxfb.c

···
 		.pm	= pm_sleep_ptr(&imxfb_pm_ops),
 	},
 	.probe		= imxfb_probe,
-	.remove_new	= imxfb_remove,
+	.remove		= imxfb_remove,
 	.id_table	= imxfb_devtype,
 };
 module_platform_driver(imxfb_driver);

drivers/video/fbdev/leo.c

···
 		.of_match_table = leo_match,
 	},
 	.probe		= leo_probe,
-	.remove_new	= leo_remove,
+	.remove		= leo_remove,
 };
 
 static int __init leo_init(void)

drivers/video/fbdev/mb862xx/mb862xxfbdrv.c

···
 		.of_match_table = of_platform_mb862xx_tbl,
 	},
 	.probe		= of_platform_mb862xx_probe,
-	.remove_new	= of_platform_mb862xx_remove,
+	.remove		= of_platform_mb862xx_remove,
 };
 #endif
 

drivers/video/fbdev/metronomefb.c

···
 
 static struct platform_driver metronomefb_driver = {
 	.probe	= metronomefb_probe,
-	.remove_new = metronomefb_remove,
+	.remove	= metronomefb_remove,
 	.driver	= {
 		.name	= "metronomefb",
 	},

drivers/video/fbdev/ocfb.c

···
 
 static struct platform_driver ocfb_driver = {
 	.probe  = ocfb_probe,
-	.remove_new = ocfb_remove,
+	.remove = ocfb_remove,
 	.driver = {
 		.name = "ocfb_fb",
 		.of_match_table = ocfb_match,

drivers/video/fbdev/offb.c

···
 		.name = "bootx-noscreen",
 	},
 	.probe = offb_probe_bootx_noscreen,
-	.remove_new = offb_remove,
+	.remove = offb_remove,
 };
 
 static int offb_probe_display(struct platform_device *pdev)
···
 		.of_match_table = offb_of_match_display,
 	},
 	.probe = offb_probe_display,
-	.remove_new = offb_remove,
+	.remove = offb_remove,
 };
 
 static int __init offb_init(void)

drivers/video/fbdev/omap/omapfb_main.c

···
 
 static struct platform_driver omapfb_driver = {
 	.probe		= omapfb_probe,
-	.remove_new	= omapfb_remove,
+	.remove		= omapfb_remove,
 	.suspend	= omapfb_suspend,
 	.resume		= omapfb_resume,
 	.driver		= {

drivers/video/fbdev/omap2/omapfb/displays/connector-analog-tv.c

···
 
 static struct platform_driver tvc_connector_driver = {
 	.probe	= tvc_probe,
-	.remove_new = tvc_remove,
+	.remove	= tvc_remove,
 	.driver	= {
 		.name	= "connector-analog-tv",
 		.of_match_table = tvc_of_match,

drivers/video/fbdev/omap2/omapfb/displays/connector-dvi.c

···
 
 static struct platform_driver dvi_connector_driver = {
 	.probe	= dvic_probe,
-	.remove_new = dvic_remove,
+	.remove	= dvic_remove,
 	.driver	= {
 		.name	= "connector-dvi",
 		.of_match_table = dvic_of_match,

drivers/video/fbdev/omap2/omapfb/displays/connector-hdmi.c

···
 
 static struct platform_driver hdmi_connector_driver = {
 	.probe	= hdmic_probe,
-	.remove_new = hdmic_remove,
+	.remove	= hdmic_remove,
 	.driver	= {
 		.name	= "connector-hdmi",
 		.of_match_table = hdmic_of_match,

drivers/video/fbdev/omap2/omapfb/displays/encoder-opa362.c

···
 
 static struct platform_driver opa362_driver = {
 	.probe	= opa362_probe,
-	.remove_new = opa362_remove,
+	.remove	= opa362_remove,
 	.driver	= {
 		.name	= "amplifier-opa362",
 		.of_match_table = opa362_of_match,

drivers/video/fbdev/omap2/omapfb/displays/encoder-tfp410.c

···
 
 static struct platform_driver tfp410_driver = {
 	.probe	= tfp410_probe,
-	.remove_new = tfp410_remove,
+	.remove	= tfp410_remove,
 	.driver	= {
 		.name	= "tfp410",
 		.of_match_table = tfp410_of_match,

drivers/video/fbdev/omap2/omapfb/displays/encoder-tpd12s015.c

···
 
 static struct platform_driver tpd_driver = {
 	.probe	= tpd_probe,
-	.remove_new = tpd_remove,
+	.remove	= tpd_remove,
 	.driver	= {
 		.name	= "tpd12s015",
 		.of_match_table = tpd_of_match,

drivers/video/fbdev/omap2/omapfb/displays/panel-dpi.c

···
 
 static struct platform_driver panel_dpi_driver = {
 	.probe = panel_dpi_probe,
-	.remove_new = panel_dpi_remove,
+	.remove = panel_dpi_remove,
 	.driver = {
 		.name = "panel-dpi",
 		.of_match_table = panel_dpi_of_match,

drivers/video/fbdev/omap2/omapfb/displays/panel-dsi-cm.c

···
 
 static struct platform_driver dsicm_driver = {
 	.probe = dsicm_probe,
-	.remove_new = dsicm_remove,
+	.remove = dsicm_remove,
 	.driver = {
 		.name = "panel-dsi-cm",
 		.of_match_table = dsicm_of_match,

drivers/video/fbdev/omap2/omapfb/displays/panel-sharp-ls037v7dw01.c

···
 
 static struct platform_driver sharp_ls_driver = {
 	.probe = sharp_ls_probe,
-	.remove_new = sharp_ls_remove,
+	.remove = sharp_ls_remove,
 	.driver = {
 		.name = "panel-sharp-ls037v7dw01",
 		.of_match_table = sharp_ls_of_match,

drivers/video/fbdev/omap2/omapfb/dss/core.c

···
 }
 
 static struct platform_driver omap_dss_driver = {
-	.remove_new     = omap_dss_remove,
+	.remove		= omap_dss_remove,
 	.shutdown	= omap_dss_shutdown,
-	.driver         = {
-		.name   = "omapdss",
+	.driver		= {
+		.name	= "omapdss",
 	},
 };
 

drivers/video/fbdev/omap2/omapfb/dss/dispc.c

···
 
 static struct platform_driver omap_dispchw_driver = {
 	.probe		= dispc_probe,
-	.remove_new     = dispc_remove,
+	.remove         = dispc_remove,
 	.driver         = {
 		.name   = "omapdss_dispc",
 		.pm	= &dispc_pm_ops,

drivers/video/fbdev/omap2/omapfb/dss/dpi.c

···
 
 static struct platform_driver omap_dpi_driver = {
 	.probe		= dpi_probe,
-	.remove_new	= dpi_remove,
-	.driver         = {
+	.remove		= dpi_remove,
+	.driver		= {
 		.name   = "omapdss_dpi",
 		.suppress_bind_attrs = true,
 	},

drivers/video/fbdev/omap2/omapfb/dss/dsi.c

···
 
 static struct platform_driver omap_dsihw_driver = {
 	.probe		= dsi_probe,
-	.remove_new	= dsi_remove,
+	.remove		= dsi_remove,
 	.driver         = {
 		.name   = "omapdss_dsi",
 		.pm	= &dsi_pm_ops,

drivers/video/fbdev/omap2/omapfb/dss/dss.c

···
 
 static struct platform_driver omap_dsshw_driver = {
 	.probe		= dss_probe,
-	.remove_new	= dss_remove,
+	.remove		= dss_remove,
 	.driver         = {
 		.name   = "omapdss_dss",
 		.pm	= &dss_pm_ops,

drivers/video/fbdev/omap2/omapfb/dss/hdmi4.c

···
 
 static struct platform_driver omapdss_hdmihw_driver = {
 	.probe		= hdmi4_probe,
-	.remove_new	= hdmi4_remove,
-	.driver         = {
-		.name   = "omapdss_hdmi",
+	.remove		= hdmi4_remove,
+	.driver		= {
+		.name	= "omapdss_hdmi",
 		.pm	= &hdmi_pm_ops,
 		.of_match_table = hdmi_of_match,
 		.suppress_bind_attrs = true,

drivers/video/fbdev/omap2/omapfb/dss/hdmi5.c

···
 
 static struct platform_driver omapdss_hdmihw_driver = {
 	.probe		= hdmi5_probe,
-	.remove_new	= hdmi5_remove,
+	.remove		= hdmi5_remove,
 	.driver         = {
 		.name   = "omapdss_hdmi5",
 		.pm	= &hdmi_pm_ops,

drivers/video/fbdev/omap2/omapfb/dss/sdi.c

···
 
 static struct platform_driver omap_sdi_driver = {
 	.probe		= sdi_probe,
-	.remove_new     = sdi_remove,
-	.driver         = {
-		.name   = "omapdss_sdi",
+	.remove		= sdi_remove,
+	.driver		= {
+		.name	= "omapdss_sdi",
 		.suppress_bind_attrs = true,
 	},
 };

drivers/video/fbdev/omap2/omapfb/dss/venc.c

···
 
 static struct platform_driver omap_venchw_driver = {
 	.probe		= venc_probe,
-	.remove_new	= venc_remove,
-	.driver         = {
-		.name   = "omapdss_venc",
+	.remove		= venc_remove,
+	.driver		= {
+		.name	= "omapdss_venc",
 		.pm	= &venc_pm_ops,
 		.of_match_table = venc_of_match,
 		.suppress_bind_attrs = true,

drivers/video/fbdev/omap2/omapfb/omapfb-main.c

···
 
 static struct platform_driver omapfb_driver = {
 	.probe		= omapfb_probe,
-	.remove_new     = omapfb_remove,
+	.remove         = omapfb_remove,
 	.driver         = {
 		.name   = "omapfb",
 	},

drivers/video/fbdev/p9100.c

···
 		.of_match_table = p9100_match,
 	},
 	.probe		= p9100_probe,
-	.remove_new	= p9100_remove,
+	.remove		= p9100_remove,
 };
 
 static int __init p9100_init(void)

drivers/video/fbdev/platinumfb.c

···
 		.of_match_table = platinumfb_match,
 	},
 	.probe		= platinumfb_probe,
-	.remove_new	= platinumfb_remove,
+	.remove		= platinumfb_remove,
 };
 
 static int __init platinumfb_init(void)

drivers/video/fbdev/pxa168fb.c

···
 		.name	= "pxa168-fb",
 	},
 	.probe		= pxa168fb_probe,
-	.remove_new	= pxa168fb_remove,
+	.remove		= pxa168fb_remove,
 };
 
 module_platform_driver(pxa168fb_driver);

drivers/video/fbdev/pxa3xx-gcu.c

···
 #endif
 
 static struct platform_driver pxa3xx_gcu_driver = {
-	.probe	  = pxa3xx_gcu_probe,
-	.remove_new	 = pxa3xx_gcu_remove,
-	.driver	 = {
-		.name   = DRV_NAME,
+	.probe = pxa3xx_gcu_probe,
+	.remove = pxa3xx_gcu_remove,
+	.driver = {
+		.name = DRV_NAME,
 		.of_match_table = of_match_ptr(pxa3xx_gcu_of_match),
 	},
 };

drivers/video/fbdev/pxafb.c

···
 
 static struct platform_driver pxafb_driver = {
 	.probe		= pxafb_probe,
-	.remove_new 	= pxafb_remove,
+	.remove		= pxafb_remove,
 	.driver		= {
 		.name	= "pxa2xx-fb",
 		.of_match_table = pxafb_of_dev_id,

drivers/video/fbdev/s1d13xxxfb.c

···
 
 static struct platform_driver s1d13xxxfb_driver = {
 	.probe		= s1d13xxxfb_probe,
-	.remove_new	= s1d13xxxfb_remove,
+	.remove		= s1d13xxxfb_remove,
 #ifdef CONFIG_PM
 	.suspend	= s1d13xxxfb_suspend,
 	.resume		= s1d13xxxfb_resume,

drivers/video/fbdev/s3c-fb.c

···
 
 static struct platform_driver s3c_fb_driver = {
 	.probe		= s3c_fb_probe,
-	.remove_new	= s3c_fb_remove,
+	.remove		= s3c_fb_remove,
 	.id_table	= s3c_fb_driver_ids,
 	.driver		= {
 		.name	= "s3c-fb",

drivers/video/fbdev/sh7760fb.c

···
 		   .name = "sh7760-lcdc",
 		   },
 	.probe = sh7760fb_probe,
-	.remove_new = sh7760fb_remove,
+	.remove = sh7760fb_remove,
 };
 
 module_platform_driver(sh7760_lcdc_driver);

drivers/video/fbdev/sh_mobile_lcdcfb.c

···
 		.pm		= &sh_mobile_lcdc_dev_pm_ops,
 	},
 	.probe		= sh_mobile_lcdc_probe,
-	.remove_new	= sh_mobile_lcdc_remove,
+	.remove		= sh_mobile_lcdc_remove,
 };
 
 module_platform_driver(sh_mobile_lcdc_driver);

drivers/video/fbdev/simplefb.c

···
 		.of_match_table = simplefb_of_match,
 	},
 	.probe = simplefb_probe,
-	.remove_new = simplefb_remove,
+	.remove = simplefb_remove,
 };
 
 module_platform_driver(simplefb_driver);

drivers/video/fbdev/sm501fb.c

···
 
 static struct platform_driver sm501fb_driver = {
 	.probe		= sm501fb_probe,
-	.remove_new	= sm501fb_remove,
+	.remove		= sm501fb_remove,
 	.suspend	= sm501fb_suspend,
 	.resume		= sm501fb_resume,
 	.driver		= {

drivers/video/fbdev/tcx.c

···
 		.of_match_table = tcx_match,
 	},
 	.probe		= tcx_probe,
-	.remove_new	= tcx_remove,
+	.remove		= tcx_remove,
 };
 
 static int __init tcx_init(void)

drivers/video/fbdev/uvesafb.c

···
 
 static struct platform_driver uvesafb_driver = {
 	.probe  = uvesafb_probe,
-	.remove_new = uvesafb_remove,
+	.remove = uvesafb_remove,
 	.driver = {
 		.name = "uvesafb",
 	},

drivers/video/fbdev/vesafb.c

···
 		.name = "vesa-framebuffer",
 	},
 	.probe = vesafb_probe,
-	.remove_new = vesafb_remove,
+	.remove = vesafb_remove,
 };
 
 module_platform_driver(vesafb_driver);

drivers/video/fbdev/vfb.c

···
 
 static struct platform_driver vfb_driver = {
 	.probe	= vfb_probe,
-	.remove_new = vfb_remove,
+	.remove = vfb_remove,
 	.driver = {
 		.name	= "vfb",
 	},

drivers/video/fbdev/vga16fb.c

···
 
 static struct platform_driver vga16fb_driver = {
 	.probe = vga16fb_probe,
-	.remove_new = vga16fb_remove,
+	.remove = vga16fb_remove,
 	.driver = {
 		.name = "vga16fb",
 	},

drivers/video/fbdev/via/via-gpio.c

···
 		.name = "viafb-gpio",
 	},
 	.probe = viafb_gpio_probe,
-	.remove_new = viafb_gpio_remove,
+	.remove = viafb_gpio_remove,
 };
 
 int viafb_gpio_init(void)

drivers/video/fbdev/via/via_i2c.c

···
 		.name = "viafb-i2c",
 	},
 	.probe = viafb_i2c_probe,
-	.remove_new = viafb_i2c_remove,
+	.remove = viafb_i2c_remove,
 };
 
 int viafb_i2c_init(void)

drivers/video/fbdev/vt8500lcdfb.c

···
 
 static struct platform_driver vt8500lcd_driver = {
 	.probe		= vt8500lcd_probe,
-	.remove_new	= vt8500lcd_remove,
+	.remove		= vt8500lcd_remove,
 	.driver		= {
 		.name	= "vt8500-lcd",
 		.of_match_table = of_match_ptr(via_dt_ids),

drivers/video/fbdev/wm8505fb.c

···
 
 static struct platform_driver wm8505fb_driver = {
 	.probe		= wm8505fb_probe,
-	.remove_new	= wm8505fb_remove,
+	.remove		= wm8505fb_remove,
 	.driver		= {
 		.name	= DRIVER_NAME,
 		.of_match_table = wmt_dt_ids,

drivers/video/fbdev/wmt_ge_rops.c

···
 
 static struct platform_driver wmt_ge_rops_driver = {
 	.probe		= wmt_ge_rops_probe,
-	.remove_new	= wmt_ge_rops_remove,
+	.remove		= wmt_ge_rops_remove,
 	.driver		= {
 		.name	= "wmt_ge_rops",
 		.of_match_table = wmt_dt_ids,

drivers/video/fbdev/xilinxfb.c

···
 
 static struct platform_driver xilinxfb_of_driver = {
 	.probe = xilinxfb_of_probe,
-	.remove_new = xilinxfb_of_remove,
+	.remove = xilinxfb_of_remove,
 	.driver = {
 		.name = DRIVER_NAME,
 		.of_match_table = xilinxfb_of_match,

fs/Kconfig

···
 
 config NFS_COMMON_LOCALIO_SUPPORT
 	tristate
-	default n
+	depends on NFS_LOCALIO
 	default y if NFSD=y || NFS_FS=y
 	default m if NFSD=m && NFS_FS=m
 	select SUNRPC

fs/bcachefs/alloc_background.c

···
 				continue;
 			}
 
+			if (k.k->p.offset < ca->mi.first_bucket) {
+				bch2_btree_iter_set_pos(&iter, POS(k.k->p.inode, ca->mi.first_bucket));
+				continue;
+			}
+
+			if (k.k->p.offset >= ca->mi.nbuckets) {
+				bch2_btree_iter_set_pos(&iter, POS(k.k->p.inode + 1, 0));
+				continue;
+			}
+
 			struct bch_alloc_v4 a;
 			*bucket_gen(ca, k.k->p.offset) = bch2_alloc_to_v4(k, &a)->gen;
 			0;

fs/bcachefs/bcachefs_format.h

···
 	x(disk_accounting_v2,		BCH_VERSION(1,  9))		\
 	x(disk_accounting_v3,		BCH_VERSION(1, 10))		\
 	x(disk_accounting_inum,		BCH_VERSION(1, 11))		\
-	x(rebalance_work_acct_fix,	BCH_VERSION(1, 12))
+	x(rebalance_work_acct_fix,	BCH_VERSION(1, 12))		\
+	x(inode_has_child_snapshots,	BCH_VERSION(1, 13))
 
 enum bcachefs_metadata_version {
 	bcachefs_metadata_version_min = 9,

fs/bcachefs/btree_gc.c

···
 	u64 b, start_time = local_clock();
 	int ret;
 
-	/*
-	 * Ideally we would be using state_lock and not gc_gens_lock here, but that
-	 * introduces a deadlock in the RO path - we currently take the state
-	 * lock at the start of going RO, thus the gc thread may get stuck:
-	 */
 	if (!mutex_trylock(&c->gc_gens_lock))
 		return 0;
 
 	trace_and_count(c, gc_gens_start, c);
 
-	down_read(&c->state_lock);
+	/*
+	 * We have to use trylock here. Otherwise, we would
+	 * introduce a deadlock in the RO path - we take the
+	 * state lock at the start of going RO.
+	 */
+	if (!down_read_trylock(&c->state_lock)) {
+		mutex_unlock(&c->gc_gens_lock);
+		return 0;
+	}
 
 	for_each_member_device(c, ca) {
 		struct bucket_gens *gens = bucket_gens(ca);

fs/bcachefs/btree_io.c

···
 	struct btree_trans *trans = bch2_trans_get(c);
 
 	btree_node_lock_nopath_nofail(trans, &b->c, SIX_LOCK_read);
+
+	/* we don't need transaction context anymore after we got the lock. */
+	bch2_trans_put(trans);
 	__btree_node_write_done(c, b);
 	six_unlock_read(&b->c.lock);
-
-	bch2_trans_put(trans);
 }
 
 static void btree_node_write_work(struct work_struct *work)

fs/bcachefs/btree_iter.c

···
 		else
 			iter_pos = bkey_max(iter->pos, bkey_start_pos(k.k));
 
-		if (unlikely(!(iter->flags & BTREE_ITER_is_extents)
-			     ? bkey_gt(iter_pos, end)
-			     : bkey_ge(iter_pos, end)))
+		if (unlikely(iter->flags & BTREE_ITER_all_snapshots	? bpos_gt(iter_pos, end) :
+			     iter->flags & BTREE_ITER_is_extents	? bkey_ge(iter_pos, end) :
+									  bkey_gt(iter_pos, end)))
 			goto end;
 
 		break;

fs/bcachefs/btree_iter.h

···
 	for_each_btree_key_upto_norestart(_trans, _iter, _btree_id, _start,\
 					  SPOS_MAX, _flags, _k, _ret)
 
+#define for_each_btree_key_reverse_norestart(_trans, _iter, _btree_id,	\
+					     _start, _flags, _k, _ret)	\
+	for (bch2_trans_iter_init((_trans), &(_iter), (_btree_id),	\
+				  (_start), (_flags));			\
+	     (_k) = bch2_btree_iter_peek_prev_type(&(_iter), _flags),	\
+	     !((_ret) = bkey_err(_k)) && (_k).k;			\
+	     bch2_btree_iter_rewind(&(_iter)))
+
 #define for_each_btree_key_continue_norestart(_iter, _flags, _k, _ret)	\
 	for_each_btree_key_upto_continue_norestart(_iter, SPOS_MAX, _flags, _k, _ret)
 

fs/bcachefs/btree_node_scan.c

···
 	if (BTREE_NODE_LEVEL(bn) >= BTREE_MAX_DEPTH)
 		return;
 
+	if (BTREE_NODE_ID(bn) >= BTREE_ID_NR_MAX)
+		return;
+
 	rcu_read_lock();
 	struct found_btree_node n = {
 		.btree_id	= BTREE_NODE_ID(bn),

fs/bcachefs/data_update.c

···
 					if (ptr2 == ptr)
 						break;
 
+					ca = bch2_dev_have_ref(c, ptr2->dev);
 					bucket = PTR_BUCKET_POS(ca, ptr2);
 					bch2_bucket_nocow_unlock(&c->nocow_locks, bucket, 0);
 				}

fs/bcachefs/disk_accounting.c

···
 		*p = swab64(*p);
 }
 
+static inline void __accounting_to_replicas(struct bch_replicas_entry_v1 *r,
+					    struct disk_accounting_pos acc)
+{
+	unsafe_memcpy(r, &acc.replicas,
+		      replicas_entry_bytes(&acc.replicas),
+		      "variable length struct");
+}
+
 static inline bool accounting_to_replicas(struct bch_replicas_entry_v1 *r, struct bpos p)
 {
 	struct disk_accounting_pos acc_k;
···
 
 	switch (acc_k.type) {
 	case BCH_DISK_ACCOUNTING_replicas:
-		unsafe_memcpy(r, &acc_k.replicas,
-			      replicas_entry_bytes(&acc_k.replicas),
-			      "variable length struct");
+		__accounting_to_replicas(r, acc_k);
 		return true;
 	default:
 		return false;
···
 	return ret;
 }
 
+static int bch2_disk_accounting_validate_late(struct btree_trans *trans,
+					      struct disk_accounting_pos acc,
+					      u64 *v, unsigned nr)
+{
+	struct bch_fs *c = trans->c;
+	struct printbuf buf = PRINTBUF;
+	int ret = 0, invalid_dev = -1;
+
+	switch (acc.type) {
+	case BCH_DISK_ACCOUNTING_replicas: {
+		struct bch_replicas_padded r;
+		__accounting_to_replicas(&r.e, acc);
+
+		for (unsigned i = 0; i < r.e.nr_devs; i++)
+			if (r.e.devs[i] != BCH_SB_MEMBER_INVALID &&
+			    !bch2_dev_exists(c, r.e.devs[i])) {
+				invalid_dev = r.e.devs[i];
+				goto invalid_device;
+			}
+
+		/*
+		 * All replicas entry checks except for invalid device are done
+		 * in bch2_accounting_validate
+		 */
+		BUG_ON(bch2_replicas_entry_validate(&r.e, c, &buf));
+
+		if (fsck_err_on(!bch2_replicas_marked_locked(c, &r.e),
+				trans, accounting_replicas_not_marked,
+				"accounting not marked in superblock replicas\n  %s",
+				(printbuf_reset(&buf),
+				 bch2_accounting_key_to_text(&buf, &acc),
+				 buf.buf))) {
+			/*
+			 * We're not RW yet and still single threaded, dropping
+			 * and retaking lock is ok:
+			 */
+			percpu_up_write(&c->mark_lock);
+			ret = bch2_mark_replicas(c, &r.e);
+			if (ret)
+				goto fsck_err;
+			percpu_down_write(&c->mark_lock);
+		}
+		break;
+	}
+
+	case BCH_DISK_ACCOUNTING_dev_data_type:
+		if (!bch2_dev_exists(c, acc.dev_data_type.dev)) {
+			invalid_dev = acc.dev_data_type.dev;
+			goto invalid_device;
+		}
+		break;
+	}
+
+fsck_err:
+	printbuf_exit(&buf);
+	return ret;
+invalid_device:
+	if (fsck_err(trans, accounting_to_invalid_device,
+		     "accounting entry points to invalid device %i\n  %s",
+		     invalid_dev,
+		     (printbuf_reset(&buf),
+		      bch2_accounting_key_to_text(&buf, &acc),
+		      buf.buf))) {
+		for (unsigned i = 0; i < nr; i++)
+			v[i] = -v[i];
+
+		ret = commit_do(trans, NULL, NULL, 0,
+				bch2_disk_accounting_mod(trans, &acc, v, nr, false)) ?:
+			-BCH_ERR_remove_disk_accounting_entry;
+	} else {
+		ret = -BCH_ERR_remove_disk_accounting_entry;
+	}
+	goto fsck_err;
+}
+
 /*
  * At startup time, initialize the in memory accounting from the btree (and
  * journal)
···
 	}
 	keys->gap = keys->nr = dst - keys->data;
 
-	percpu_down_read(&c->mark_lock);
-	for (unsigned i = 0; i < acc->k.nr; i++) {
+	percpu_down_write(&c->mark_lock);
+	unsigned i = 0;
+	while (i < acc->k.nr) {
+		unsigned idx = inorder_to_eytzinger0(i, acc->k.nr);
+
+		struct disk_accounting_pos acc_k;
+		bpos_to_disk_accounting_pos(&acc_k, acc->k.data[idx].pos);
+
 		u64 v[BCH_ACCOUNTING_MAX_COUNTERS];
-		bch2_accounting_mem_read_counters(acc, i, v, ARRAY_SIZE(v), false);
-
-		if (bch2_is_zero(v, sizeof(v[0]) * acc->k.data[i].nr_counters))
-			continue;
-
-		struct bch_replicas_padded r;
-		if (!accounting_to_replicas(&r.e, acc->k.data[i].pos))
-			continue;
+		bch2_accounting_mem_read_counters(acc, idx, v, ARRAY_SIZE(v), false);
 
 		/*
-		 * If the replicas entry is invalid it'll get cleaned up by
-		 * check_allocations:
+		 * If the entry counters are zeroed, it should be treated as
+		 * nonexistent - it might point to an invalid device.
+		 *
+		 * Remove it, so that if it's re-added it gets re-marked in the
+		 * superblock:
 		 */
-		if (bch2_replicas_entry_validate(&r.e, c, &buf))
+		ret = bch2_is_zero(v, sizeof(v[0]) * acc->k.data[idx].nr_counters)
+			? -BCH_ERR_remove_disk_accounting_entry
+			: bch2_disk_accounting_validate_late(trans, acc_k,
+							v, acc->k.data[idx].nr_counters);
+
+		if (ret == -BCH_ERR_remove_disk_accounting_entry) {
+			free_percpu(acc->k.data[idx].v[0]);
+			free_percpu(acc->k.data[idx].v[1]);
+			darray_remove_item(&acc->k, &acc->k.data[idx]);
+			eytzinger0_sort(acc->k.data, acc->k.nr, sizeof(acc->k.data[0]),
+					accounting_pos_cmp, NULL);
+			ret = 0;
 			continue;
-
-		struct disk_accounting_pos k;
-		bpos_to_disk_accounting_pos(&k, acc->k.data[i].pos);
-
-		if (fsck_err_on(!bch2_replicas_marked_locked(c, &r.e),
-				trans, accounting_replicas_not_marked,
-				"accounting not marked in superblock replicas\n  %s",
-				(printbuf_reset(&buf),
-				 bch2_accounting_key_to_text(&buf, &k),
-				 buf.buf))) {
-			/*
-			 * We're not RW yet and still single threaded, dropping
-			 * and retaking lock is ok:
-			 */
-			percpu_up_read(&c->mark_lock);
-			ret = bch2_mark_replicas(c, &r.e);
-			if (ret)
-				goto fsck_err;
-			percpu_down_read(&c->mark_lock);
 		}
+
+		if (ret)
+			goto fsck_err;
+		i++;
 	}
 
 	preempt_disable();
···
 	}
 	preempt_enable();
 fsck_err:
-	percpu_up_read(&c->mark_lock);
+	percpu_up_write(&c->mark_lock);
 err:
 	printbuf_exit(&buf);
 	bch2_trans_put(trans);

fs/bcachefs/ec.c

···
 			 "incorrect value size (%zu < %u)",
 			 bkey_val_u64s(k.k), stripe_val_u64s(s));
 
+	bkey_fsck_err_on(s->csum_granularity_bits >= 64,
+			 c, stripe_csum_granularity_bad,
+			 "invalid csum granularity (%u >= 64)",
+			 s->csum_granularity_bits);
+
 	ret = bch2_bkey_ptrs_validate(c, k, flags);
 fsck_err:
 	return ret;
···
 		   nr_data,
 		   s.nr_redundant);
 	bch2_prt_csum_type(out, s.csum_type);
-	prt_printf(out, " gran %u", 1U << s.csum_granularity_bits);
+	prt_str(out, " gran ");
+	if (s.csum_granularity_bits < 64)
+		prt_printf(out, "%llu", 1ULL << s.csum_granularity_bits);
+	else
+		prt_printf(out, "(invalid shift %u)", s.csum_granularity_bits);
 
 	if (s.disk_label) {
 		prt_str(out, " label");
···
 /* stripe creation: */
 
 static int ec_stripe_key_update(struct btree_trans *trans,
-				struct bkey_i_stripe *new,
-				bool create)
+				struct bkey_i_stripe *old,
+				struct bkey_i_stripe *new)
 {
 	struct bch_fs *c = trans->c;
-	struct btree_iter iter;
-	struct bkey_s_c k;
-	int ret;
+	bool create = !old;
 
-	k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_stripes,
-			       new->k.p, BTREE_ITER_intent);
-	ret = bkey_err(k);
+	struct btree_iter iter;
+	struct bkey_s_c k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_stripes,
+					       new->k.p, BTREE_ITER_intent);
+	int ret = bkey_err(k);
 	if (ret)
 		goto err;
 
-	if (k.k->type != (create ? KEY_TYPE_deleted : KEY_TYPE_stripe)) {
-		bch2_fs_inconsistent(c, "error %s stripe: got existing key type %s",
-				     create ? "creating" : "updating",
-				     bch2_bkey_types[k.k->type]);
+	if (bch2_fs_inconsistent_on(k.k->type != (create ? KEY_TYPE_deleted : KEY_TYPE_stripe),
+				    c, "error %s stripe: got existing key type %s",
+				    create ? "creating" : "updating",
+				    bch2_bkey_types[k.k->type])) {
 		ret = -EINVAL;
 		goto err;
 	}
 
 	if (k.k->type == KEY_TYPE_stripe) {
-		const struct bch_stripe *old = bkey_s_c_to_stripe(k).v;
-		unsigned i;
+		const struct bch_stripe *v = bkey_s_c_to_stripe(k).v;
 
-		if (old->nr_blocks != new->v.nr_blocks) {
-			bch_err(c, "error updating stripe: nr_blocks does not match");
-			ret = -EINVAL;
-			goto err;
-		}
+		BUG_ON(old->v.nr_blocks != new->v.nr_blocks);
+		BUG_ON(old->v.nr_blocks != v->nr_blocks);
 
-		for (i = 0; i < new->v.nr_blocks; i++) {
-			unsigned v = stripe_blockcount_get(old, i);
+		for (unsigned i = 0; i < new->v.nr_blocks; i++) {
+			unsigned sectors = stripe_blockcount_get(v, i);
 
-			BUG_ON(v &&
-			       (old->ptrs[i].dev != new->v.ptrs[i].dev ||
-				old->ptrs[i].gen != new->v.ptrs[i].gen ||
-				old->ptrs[i].offset != new->v.ptrs[i].offset));
+			if (!bch2_extent_ptr_eq(old->v.ptrs[i], new->v.ptrs[i]) && sectors) {
+				struct printbuf buf = PRINTBUF;
 
-			stripe_blockcount_set(&new->v, i, v);
+				prt_printf(&buf, "stripe changed nonempty block %u", i);
+				prt_str(&buf, "\nold: ");
+				bch2_bkey_val_to_text(&buf, c, k);
+				prt_str(&buf, "\nnew: ");
+				bch2_bkey_val_to_text(&buf, c, bkey_i_to_s_c(&new->k_i));
+				bch2_fs_inconsistent(c, "%s", buf.buf);
+				printbuf_exit(&buf);
+				ret = -EINVAL;
+				goto err;
+			}
+
+			/*
+			 * If the stripe ptr changed underneath us, it must have
+			 * been dev_remove_stripes() -> * invalidate_stripe_to_dev()
+			 */
+			if (!bch2_extent_ptr_eq(old->v.ptrs[i], v->ptrs[i])) {
+				BUG_ON(v->ptrs[i].dev != BCH_SB_MEMBER_INVALID);
+
+				if (bch2_extent_ptr_eq(old->v.ptrs[i], new->v.ptrs[i]))
+					new->v.ptrs[i].dev = BCH_SB_MEMBER_INVALID;
+			}
+
+			stripe_blockcount_set(&new->v, i, sectors);
 		}
 	}
 
···
 			    BCH_TRANS_COMMIT_no_check_rw|
 			    BCH_TRANS_COMMIT_no_enospc,
 			    ec_stripe_key_update(trans,
-					bkey_i_to_stripe(&s->new_stripe.key),
-					!s->have_existing_stripe));
+					s->have_existing_stripe
+					? bkey_i_to_stripe(&s->existing_stripe.key)
+					: NULL,
+					bkey_i_to_stripe(&s->new_stripe.key)));
 	bch_err_msg(c, ret, "creating stripe key");
 	if (ret) {
 		goto err;
···
 	bitmap_and(devs.d, devs.d, c->rw_devs[BCH_DATA_user].d, BCH_SB_MEMBERS_MAX);
 
 	for_each_set_bit(i, h->s->blocks_gotten, v->nr_blocks) {
-		__clear_bit(v->ptrs[i].dev, devs.d);
+		/*
+		 * Note: we don't yet repair invalid blocks (failed/removed
+		 * devices) when reusing stripes - we still need a codepath to
+		 * walk backpointers and update all extents that point to that
+		 * block when updating the stripe
+		 */
+		if (v->ptrs[i].dev != BCH_SB_MEMBER_INVALID)
+			__clear_bit(v->ptrs[i].dev, devs.d);
+
 		if (i < h->s->nr_data)
 			nr_have_data++;
 		else

fs/bcachefs/errcode.h

···
 	x(BCH_ERR_nopromote,		nopromote_no_writes)			\
 	x(BCH_ERR_nopromote,		nopromote_enomem)			\
 	x(0,				invalid_snapshot_node)			\
-	x(0,				option_needs_open_fs)
+	x(0,				option_needs_open_fs)			\
+	x(0,				remove_disk_accounting_entry)
 
 enum bch_errcode {
 	BCH_ERR_START		= 2048,

fs/bcachefs/extents.h

···
 int bch2_bkey_ptrs_validate(struct bch_fs *, struct bkey_s_c,
 			    enum bch_validate_flags);
 
+static inline bool bch2_extent_ptr_eq(struct bch_extent_ptr ptr1,
+				      struct bch_extent_ptr ptr2)
+{
+	return (ptr1.cached	== ptr2.cached &&
+		ptr1.unwritten	== ptr2.unwritten &&
+		ptr1.offset	== ptr2.offset &&
+		ptr1.dev	== ptr2.dev &&
+		ptr1.dev	== ptr2.dev);
+}
+
 void bch2_ptr_swab(struct bkey_s);
 
 const struct bch_extent_rebalance *bch2_bkey_rebalance_opts(struct bkey_s_c);

fs/bcachefs/fs-io-direct.c

···
 
 static __always_inline long bch2_dio_write_done(struct dio_write *dio)
 {
+	struct bch_fs *c = dio->op.c;
 	struct kiocb *req = dio->req;
 	struct bch_inode_info *inode = dio->inode;
 	bool sync = dio->sync;
···
 	ret = dio->op.error ?: ((long) dio->written << 9);
 	bio_put(&dio->op.wbio.bio);
 
-	bch2_write_ref_put(dio->op.c, BCH_WRITE_REF_dio_write);
+	bch2_write_ref_put(c, BCH_WRITE_REF_dio_write);
 
 	/* inode->i_dio_count is our ref on inode and thus bch_fs */
 	inode_dio_end(&inode->v);

fs/bcachefs/fs.c

···
 	return a.subvol == b.subvol && a.inum == b.inum;
 }
 
+static u32 bch2_vfs_inode_hash_fn(const void *data, u32 len, u32 seed)
+{
+	const subvol_inum *inum = data;
+
+	return jhash(&inum->inum, sizeof(inum->inum), seed);
+}
+
+static u32 bch2_vfs_inode_obj_hash_fn(const void *data, u32 len, u32 seed)
+{
+	const struct bch_inode_info *inode = data;
+
+	return bch2_vfs_inode_hash_fn(&inode->ei_inum, sizeof(inode->ei_inum), seed);
+}
+
 static int bch2_vfs_inode_cmp_fn(struct rhashtable_compare_arg *arg,
 				 const void *obj)
 {
···
 	.head_offset		= offsetof(struct bch_inode_info, hash),
 	.key_offset		= offsetof(struct bch_inode_info, ei_inum),
 	.key_len		= sizeof(subvol_inum),
+	.hashfn			= bch2_vfs_inode_hash_fn,
+	.obj_hashfn		= bch2_vfs_inode_obj_hash_fn,
 	.obj_cmpfn		= bch2_vfs_inode_cmp_fn,
 	.automatic_shrinking	= true,
 };
 
-struct bch_inode_info *__bch2_inode_hash_find(struct bch_fs *c, subvol_inum inum)
+int bch2_inode_or_descendents_is_open(struct btree_trans *trans, struct bpos p)
+{
+	struct bch_fs *c = trans->c;
+	struct rhashtable *ht = &c->vfs_inodes_table;
+	subvol_inum inum = (subvol_inum) { .inum = p.offset };
+	DARRAY(u32) subvols;
+	int ret = 0;
+
+	if (!test_bit(BCH_FS_started, &c->flags))
+		return false;
+
+	darray_init(&subvols);
+restart_from_top:
+
+	/*
+	 * Tweaked version of __rhashtable_lookup(); we need to get a list of
+	 * subvolumes in which the given inode number is open.
+	 *
+	 * For this to work, we don't include the subvolume ID in the key that
+	 * we hash - all inodes with the same inode number regardless of
+	 * subvolume will hash to the same slot.
+	 *
+	 * This will be less than ideal if the same file is ever open
+	 * simultaneously in many different snapshots:
+	 */
+	rcu_read_lock();
+	struct rhash_lock_head __rcu *const *bkt;
+	struct rhash_head *he;
+	unsigned int hash;
+	struct bucket_table *tbl = rht_dereference_rcu(ht->tbl, ht);
+restart:
+	hash = rht_key_hashfn(ht, tbl, &inum, bch2_vfs_inodes_params);
+	bkt = rht_bucket(tbl, hash);
+	do {
+		struct bch_inode_info *inode;
+
+		rht_for_each_entry_rcu_from(inode, he, rht_ptr_rcu(bkt), tbl, hash, hash) {
+			if (inode->ei_inum.inum == inum.inum) {
+				ret = darray_push_gfp(&subvols, inode->ei_inum.subvol,
+						      GFP_NOWAIT|__GFP_NOWARN);
+				if (ret) {
+					rcu_read_unlock();
+					ret = darray_make_room(&subvols, 1);
+					if (ret)
+						goto err;
+					subvols.nr = 0;
+					goto restart_from_top;
+				}
+			}
+		}
+		/* An object might have been moved to a different hash chain,
+		 * while we walk along it - better check and retry.
+		 */
+	} while (he != RHT_NULLS_MARKER(bkt));
+
+	/* Ensure we see any new tables. */
+	smp_rmb();
+
+	tbl = rht_dereference_rcu(tbl->future_tbl, ht);
+	if (unlikely(tbl))
+		goto restart;
+	rcu_read_unlock();
+
+	darray_for_each(subvols, i) {
+		u32 snap;
+		ret = bch2_subvolume_get_snapshot(trans, *i, &snap);
+		if (ret)
+			goto err;
+
+		ret = bch2_snapshot_is_ancestor(c, snap, p.snapshot);
+		if (ret)
+			break;
+	}
+err:
+	darray_exit(&subvols);
+	return ret;
+}
+
+static struct bch_inode_info *__bch2_inode_hash_find(struct bch_fs *c, subvol_inum inum)
 {
 	return rhashtable_lookup_fast(&c->vfs_inodes_table, &inum, bch2_vfs_inodes_params);
 }
···
 				    subvol_inum inum)
 {
 	wait_queue_head_t *wq;
-	DEFINE_WAIT_BIT(wait, &inode->v.i_state, __I_NEW);
+	struct wait_bit_queue_entry wait;
+
 	wq = inode_bit_waitqueue(&wait, &inode->v, __I_NEW);
 	prepare_to_wait(wq, &wait.wq_entry, TASK_UNINTERRUPTIBLE);
 	spin_unlock(&inode->v.i_lock);
···
 
 	set_bit(EI_INODE_HASHED, &inode->ei_flags);
 retry:
-	if (unlikely(rhashtable_lookup_insert_fast(&c->vfs_inodes_table,
+	if (unlikely(rhashtable_lookup_insert_key(&c->vfs_inodes_table,
+					&inode->ei_inum,
 					&inode->hash,
 					bch2_vfs_inodes_params))) {
 		old = bch2_inode_hash_find(c, trans, inode->ei_inum);

fs/bcachefs/fs.h

···
 	return inode->ei_inum;
 }
 
-struct bch_inode_info *__bch2_inode_hash_find(struct bch_fs *, subvol_inum);
-
 /*
  * Set if we've gotten a btree error for this inode, and thus the vfs inode and
  * btree inode may be inconsistent:
···
 __bch2_create(struct mnt_idmap *, struct bch_inode_info *,
 	      struct dentry *, umode_t, dev_t, subvol_inum, unsigned);
 
+int bch2_inode_or_descendents_is_open(struct btree_trans *trans, struct bpos p);
+
 int bch2_fs_quota_transfer(struct bch_fs *,
 			   struct bch_inode_info *,
 			   struct bch_qid,
···
 
 #define bch2_inode_update_after_write(_trans, _inode, _inode_u, _fields)	({ do {} while (0); })
 
-static inline struct bch_inode_info *__bch2_inode_hash_find(struct bch_fs *c, subvol_inum inum)
-{
-	return NULL;
-}
+static inline int bch2_inode_or_descendents_is_open(struct btree_trans *trans, struct bpos p) { return 0; }
 
 static inline void bch2_evict_subvolume_inodes(struct bch_fs *c,
 					       snapshot_id_list *s) {}

fs/bcachefs/fsck.c

···
 	return ret;
 }
 
+static inline bool inode_should_reattach(struct bch_inode_unpacked *inode)
+{
+	if (inode->bi_inum == BCACHEFS_ROOT_INO &&
+	    inode->bi_subvol == BCACHEFS_ROOT_SUBVOL)
+		return false;
+
+	return !inode->bi_dir && !(inode->bi_flags & BCH_INODE_unlinked);
+}
+
+static int maybe_delete_dirent(struct btree_trans *trans, struct bpos d_pos, u32 snapshot)
+{
+	struct btree_iter iter;
+	struct bkey_s_c k = bch2_bkey_get_iter(trans, &iter, BTREE_ID_dirents,
+					SPOS(d_pos.inode, d_pos.offset, snapshot),
+					BTREE_ITER_intent|
+					BTREE_ITER_with_updates);
+	int ret = bkey_err(k);
+	if (ret)
+		return ret;
+
+	if (bpos_eq(k.k->p, d_pos)) {
+		/*
+		 * delet_at() doesn't work because the update path doesn't
+		 * internally use BTREE_ITER_with_updates yet
+		 */
+		struct bkey_i *k = bch2_trans_kmalloc(trans, sizeof(*k));
+		ret = PTR_ERR_OR_ZERO(k);
+		if (ret)
+			goto err;
+
+		bkey_init(&k->k);
+		k->k.type = KEY_TYPE_whiteout;
+		k->k.p = iter.pos;
+		ret = bch2_trans_update(trans, &iter, k, BTREE_UPDATE_internal_snapshot_node);
+	}
+err:
+	bch2_trans_iter_exit(trans, &iter);
+	return ret;
+}
+
 static int reattach_inode(struct btree_trans *trans, struct bch_inode_unpacked *inode)
 {
 	struct bch_fs *c = trans->c;
-	struct bch_hash_info dir_hash;
 	struct bch_inode_unpacked lostfound;
 	char name_buf[20];
-	struct qstr name;
-	u64 dir_offset = 0;
-	u32 dirent_snapshot = inode->bi_snapshot;
 	int ret;
 
+	u32 dirent_snapshot = inode->bi_snapshot;
 	if (inode->bi_subvol) {
 		inode->bi_parent_subvol = BCACHEFS_ROOT_SUBVOL;
 
···
 	if (ret)
 		return ret;
 
-	dir_hash = bch2_hash_info_init(c, &lostfound);
+	struct bch_hash_info dir_hash = bch2_hash_info_init(c, &lostfound);
+	struct qstr name = (struct qstr) QSTR(name_buf);
 
-	name = (struct qstr) QSTR(name_buf);
+	inode->bi_dir = lostfound.bi_inum;
 
 	ret = bch2_dirent_create_snapshot(trans,
 				inode->bi_parent_subvol, lostfound.bi_inum,
···
 				inode_d_type(inode),
 				&name,
 				inode->bi_subvol ?: inode->bi_inum,
-				&dir_offset,
+				&inode->bi_dir_offset,
 				STR_HASH_must_create);
 	if (ret) {
 		bch_err_msg(c, ret, "error creating dirent");
 		return ret;
 	}
 
-	inode->bi_dir		= lostfound.bi_inum;
-	inode->bi_dir_offset	= dir_offset;
+	ret = __bch2_fsck_write_inode(trans, inode);
+	if (ret)
+		return ret;
 
-	return __bch2_fsck_write_inode(trans, inode);
+	/*
+	 * Fix up inodes in child snapshots: if they should also be reattached
+	 * update the backpointer field, if they should not be we need to emit
+	 * whiteouts for the dirent we just created.
+	 */
+	if (!inode->bi_subvol && bch2_snapshot_is_leaf(c, inode->bi_snapshot) <= 0) {
+		snapshot_id_list whiteouts_done;
+		struct btree_iter iter;
+		struct bkey_s_c k;
+
+		darray_init(&whiteouts_done);
+
+		for_each_btree_key_reverse_norestart(trans, iter,
+				BTREE_ID_inodes, SPOS(0, inode->bi_inum, inode->bi_snapshot - 1),
+				BTREE_ITER_all_snapshots|BTREE_ITER_intent, k, ret) {
+			if (k.k->p.offset != inode->bi_inum)
+				break;
+
+			if (!bkey_is_inode(k.k) ||
+			    !bch2_snapshot_is_ancestor(c, k.k->p.snapshot, inode->bi_snapshot) ||
+			    snapshot_list_has_ancestor(c, &whiteouts_done, k.k->p.snapshot))
+				continue;
+
+			struct bch_inode_unpacked child_inode;
+			bch2_inode_unpack(k, &child_inode);
+
+			if (!inode_should_reattach(&child_inode)) {
+				ret = maybe_delete_dirent(trans,
+							  SPOS(lostfound.bi_inum, inode->bi_dir_offset,
+							       dirent_snapshot),
+							  k.k->p.snapshot);
+				if (ret)
+					break;
+
+				ret = snapshot_list_add(c, &whiteouts_done, k.k->p.snapshot);
+				if (ret)
+					break;
+			} else {
+				iter.snapshot = k.k->p.snapshot;
+				child_inode.bi_dir = inode->bi_dir;
+				child_inode.bi_dir_offset = inode->bi_dir_offset;
+
+				ret = bch2_inode_write_flags(trans, &iter, &child_inode,
+							     BTREE_UPDATE_internal_snapshot_node);
+				if (ret)
+					break;
+			}
+		}
+		darray_exit(&whiteouts_done);
+		bch2_trans_iter_exit(trans, &iter);
+	}
+
+	return ret;
 }
 
 static int remove_backpointer(struct btree_trans *trans,
···
 		 */
 		inode->bi_dir = 0;
 		inode->bi_dir_offset = 0;
-		inode->bi_flags &= ~BCH_INODE_backptr_untrusted;
 		*write_inode = true;
 	}
 
···
 	return ret;
 }
 
-static bool bch2_inode_is_open(struct bch_fs *c, struct bpos p)
-{
-	subvol_inum inum = {
-		.subvol = snapshot_t(c, p.snapshot)->subvol,
-		.inum	= p.offset,
-	};
-
-	/* snapshot tree corruption, can't safely delete */
-	if (!inum.subvol) {
-		bch_warn_ratelimited(c, "%s(): snapshot %u has no subvol, unlinked but can't safely delete", __func__, p.snapshot);
-		return true;
-	}
-
-	return __bch2_inode_hash_find(c, inum) != NULL;
-}
-
 static int check_inode(struct btree_trans *trans,
 		       struct btree_iter *iter,
 		       struct bkey_s_c k,
 		       struct bch_inode_unpacked *prev,
-		       struct snapshots_seen *s,
-		       bool full)
+		       struct snapshots_seen *s)
 {
 	struct bch_fs *c = trans->c;
 	struct printbuf buf = PRINTBUF;
···
 		return 0;
 
 	BUG_ON(bch2_inode_unpack(k, &u));
-
-	if (!full &&
-	    !(u.bi_flags & (BCH_INODE_i_size_dirty|
-			    BCH_INODE_i_sectors_dirty|
-			    BCH_INODE_unlinked)))
-		return 0;
 
 	if (prev->bi_inum != u.bi_inum)
 		*prev = u;
···
 		ret = 0;
 	}
 
-	if ((u.bi_flags & (BCH_INODE_i_size_dirty|BCH_INODE_unlinked)) &&
-	    bch2_key_has_snapshot_overwrites(trans, BTREE_ID_inodes, k.k->p)) {
-		struct bpos new_min_pos;
+	ret = bch2_inode_has_child_snapshots(trans, k.k->p);
+	if (ret < 0)
+		goto err;
 
-		ret = bch2_propagate_key_to_snapshot_leaves(trans, iter->btree_id, k, &new_min_pos);
+	if (fsck_err_on(ret != !!(u.bi_flags & BCH_INODE_has_child_snapshot),
+			trans, inode_has_child_snapshots_wrong,
+			"inode has_child_snapshots flag wrong (should be %u)\n%s",
+			ret,
+			(printbuf_reset(&buf),
+			 bch2_inode_unpacked_to_text(&buf, &u),
+			 buf.buf))) {
 		if (ret)
-			goto err;
-
-		u.bi_flags &= ~BCH_INODE_i_size_dirty|BCH_INODE_unlinked;
-
-		ret = __bch2_fsck_write_inode(trans, &u);
-
-		bch_err_msg(c, ret, "in fsck updating inode");
-		if (ret)
-			goto err_noprint;
-
-		if (!bpos_eq(new_min_pos, POS_MIN))
-			bch2_btree_iter_set_pos(iter, bpos_predecessor(new_min_pos));
-		goto err_noprint;
+			u.bi_flags |= BCH_INODE_has_child_snapshot;
+		else
+			u.bi_flags &= ~BCH_INODE_has_child_snapshot;
+		do_update = true;
 	}
+	ret = 0;
 
-	if (u.bi_flags & BCH_INODE_unlinked) {
+	if ((u.bi_flags & BCH_INODE_unlinked) &&
+	    !(u.bi_flags & BCH_INODE_has_child_snapshot)) {
 		if (!test_bit(BCH_FS_started, &c->flags)) {
 			/*
 			 * If we're not in online fsck, don't delete unlinked
···
 			if (ret)
 				goto err;
 		} else {
-			if (fsck_err_on(!bch2_inode_is_open(c, k.k->p),
+			ret = bch2_inode_or_descendents_is_open(trans, k.k->p);
+			if (ret < 0)
+				goto err;
+
+			if (fsck_err_on(!ret,
 					trans, inode_unlinked_and_not_open,
 				      "inode %llu%u unlinked and not open",
 				      u.bi_inum, u.bi_snapshot)) {
···
 				bch_err_msg(c, ret, "in fsck deleting inode");
 				goto err_noprint;
 			}
+			ret = 0;
 		}
-	}
-
-	/* i_size_dirty is vestigal, since we now have logged ops for truncate * */
-	if (u.bi_flags & BCH_INODE_i_size_dirty &&
-	    (!test_bit(BCH_FS_clean_recovery, &c->flags) ||
-	     fsck_err(trans, inode_i_size_dirty_but_clean,
-		      "filesystem marked clean, but inode %llu has i_size dirty",
-		      u.bi_inum))) {
-		bch_verbose(c, "truncating inode %llu", u.bi_inum);
-
-		/*
-		 * XXX: need to truncate partial blocks too here - or ideally
-		 * just switch units to bytes and that issue goes away
-		 */
-		ret = bch2_btree_delete_range_trans(trans, BTREE_ID_extents,
-				SPOS(u.bi_inum, round_up(u.bi_size, block_bytes(c)) >> 9,
-				     iter->pos.snapshot),
-				POS(u.bi_inum, U64_MAX),
-				0, NULL);
-		bch_err_msg(c, ret, "in fsck truncating inode");
-		if (ret)
-			return ret;
-
-		/*
-		 * We truncated without our normal sector accounting hook, just
-		 * make sure we recalculate it:
-		 */
-		u.bi_flags |= BCH_INODE_i_sectors_dirty;
-
-		u.bi_flags &= ~BCH_INODE_i_size_dirty;
-		do_update = true;
-	}
-
-	/* i_sectors_dirty is vestigal, i_sectors is always updated transactionally */
-	if (u.bi_flags & BCH_INODE_i_sectors_dirty &&
-	    (!test_bit(BCH_FS_clean_recovery, &c->flags) ||
-	     fsck_err(trans, inode_i_sectors_dirty_but_clean,
-		      "filesystem marked clean, but inode %llu has i_sectors dirty",
-		      u.bi_inum))) {
-		s64 sectors;
-
-		bch_verbose(c, "recounting sectors for inode %llu",
-			    u.bi_inum);
-
-		sectors = bch2_count_inode_sectors(trans, u.bi_inum, iter->pos.snapshot);
-		if (sectors < 0) {
-			bch_err_msg(c, sectors, "in fsck recounting inode sectors");
-			return sectors;
-		}
-
-		u.bi_sectors = sectors;
-		u.bi_flags &= ~BCH_INODE_i_sectors_dirty;
-		do_update = true;
-	}
-
-	if (u.bi_flags & BCH_INODE_backptr_untrusted) {
-		u.bi_dir = 0;
-		u.bi_dir_offset = 0;
-		u.bi_flags &= ~BCH_INODE_backptr_untrusted;
-		do_update = true;
 	}
 
 	if (fsck_err_on(u.bi_parent_subvol &&
···
 
 int bch2_check_inodes(struct bch_fs *c)
 {
-	bool full = c->opts.fsck;
 	struct bch_inode_unpacked prev = { 0 };
 	struct snapshots_seen s;
 
···
 				POS_MIN,
 				BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k,
 				NULL, NULL, BCH_TRANS_COMMIT_no_enospc,
-			check_inode(trans, &iter, k, &prev, &s, full)));
+			check_inode(trans, &iter, k, &prev, &s)));
 
 	snapshots_seen_exit(&s);
+	bch_err_fn(c, ret);
+	return ret;
+}
+
+static int find_oldest_inode_needs_reattach(struct btree_trans *trans,
+					    struct bch_inode_unpacked *inode)
+{
+	struct bch_fs *c = trans->c;
+	struct btree_iter iter;
+	struct bkey_s_c k;
+	int ret = 0;
+
+	/*
+	 * We look for inodes to reattach in natural key order, leaves first,
+	 * but we should do the reattach at the oldest version that needs to be
+	 * reattached:
+	 */
+	for_each_btree_key_norestart(trans, iter,
+				     BTREE_ID_inodes,
+				     SPOS(0, inode->bi_inum, inode->bi_snapshot + 1),
+				     BTREE_ITER_all_snapshots, k, ret) {
+		if (k.k->p.offset != inode->bi_inum)
+			break;
+
+		if (!bch2_snapshot_is_ancestor(c, inode->bi_snapshot, k.k->p.snapshot))
+			continue;
+
+		if (!bkey_is_inode(k.k))
+			break;
+
+		struct bch_inode_unpacked parent_inode;
+		bch2_inode_unpack(k, &parent_inode);
+
+		if (!inode_should_reattach(&parent_inode))
+			break;
+
+		*inode = parent_inode;
+	}
+	bch2_trans_iter_exit(trans, &iter);
+
+	return ret;
+}
+
+static int check_unreachable_inode(struct btree_trans *trans,
+				   struct btree_iter *iter,
+				   struct bkey_s_c k)
+{
+	struct printbuf buf = PRINTBUF;
+	int ret = 0;
+
+	if (!bkey_is_inode(k.k))
+		return 0;
+
+	struct bch_inode_unpacked inode;
+	BUG_ON(bch2_inode_unpack(k, &inode));
+
+	if (!inode_should_reattach(&inode))
+		return 0;
+
+	ret = find_oldest_inode_needs_reattach(trans, &inode);
+	if (ret)
+		return ret;
+
+	if (fsck_err(trans, inode_unreachable,
+		     "unreachable inode:\n%s",
+		     (bch2_inode_unpacked_to_text(&buf, &inode),
+		      buf.buf)))
+		ret = reattach_inode(trans, &inode);
+fsck_err:
+	printbuf_exit(&buf);
+	return ret;
+}
+
+/*
+ * Reattach unreachable (but not unlinked) inodes
+ *
+ * Run after check_inodes() and check_dirents(), so we node that inode
+ * backpointer fields point to valid dirents, and every inode that has a dirent
+ * that points to it has its backpointer field set - so we're just looking for
+ * non-unlinked inodes without backpointers:
+ *
+ * XXX: this is racy w.r.t. hardlink removal in online fsck
+ */
+int bch2_check_unreachable_inodes(struct bch_fs *c)
+{
+	int ret = bch2_trans_run(c,
+		for_each_btree_key_commit(trans, iter, BTREE_ID_inodes,
+				POS_MIN,
+				BTREE_ITER_prefetch|BTREE_ITER_all_snapshots, k,
+				NULL, NULL, BCH_TRANS_COMMIT_no_enospc,
+			check_unreachable_inode(trans, &iter, k)));
 	bch_err_fn(c, ret);
 	return ret;
 }
···
 			    !key_visible_in_snapshot(c, s, i->snapshot, k.k->p.snapshot))
 				continue;
 
-			if (fsck_err_on(!(i->inode.bi_flags & BCH_INODE_i_size_dirty) &&
-					k.k->p.offset > round_up(i->inode.bi_size, block_bytes(c)) >> 9 &&
+			if (fsck_err_on(k.k->p.offset > round_up(i->inode.bi_size, block_bytes(c)) >> 9 &&
 					!bkey_extent_is_reservation(k),
 					trans, extent_past_end_of_inode,
 					"extent type past end of inode %llu:%u, i_size %llu\n  %s",
···
 		if (ret)
 			break;
 
-		/*
-		 * We've checked that inode backpointers point to valid dirents;
-		 * here, it's sufficient to check that the subvolume root has a
-		 * dirent:
-		 */
-		if (fsck_err_on(!subvol_root.bi_dir,
-				trans, subvol_unreachable,
-				"unreachable subvolume %s",
-				(bch2_bkey_val_to_text(&buf, c, s.s_c),
-				 prt_newline(&buf),
-				 bch2_inode_unpacked_to_text(&buf, &subvol_root),
-				 buf.buf))) {
-			ret = reattach_subvol(trans, s);
-			break;
-		}
-
 		u32 parent = le32_to_cpu(s.v->fs_path_parent);
 
 		if (darray_u32_has(&subvol_path, parent)) {
···
 	return false;
 }
 
-/*
- * Check that a given inode is reachable from its subvolume root - we already
- * verified subvolume connectivity:
- *
- * XXX: we should also be verifying that inodes are in the right subvolumes
- */
 static int check_path(struct btree_trans *trans, pathbuf *p, struct bkey_s_c inode_k)
 {
 	struct bch_fs *c = trans->c;
···
 	p->nr = 0;
 
 	BUG_ON(bch2_inode_unpack(inode_k, &inode));
+
+	if (!S_ISDIR(inode.bi_mode))
+		return 0;
 
 	while (!inode.bi_subvol) {
 		struct btree_iter dirent_iter;
···
 			bch2_trans_iter_exit(trans, &dirent_iter);
 
 		if (bch2_err_matches(ret, ENOENT)) {
-			ret = 0;
-			if (fsck_err(trans, inode_unreachable,
-				     "unreachable inode\n%s",
-				     (printbuf_reset(&buf),
-				      bch2_bkey_val_to_text(&buf, c, inode_k),
-				      buf.buf)))
-				ret = reattach_inode(trans, &inode);
+			printbuf_reset(&buf);
+			bch2_bkey_val_to_text(&buf, c, inode_k);
+			bch_err(c, "unreachable inode in check_directory_structure: %s\n%s",
+				bch2_err_str(ret), buf.buf);
 			goto out;
 		}
 
 		bch2_trans_iter_exit(trans, &dirent_iter);
-
-		if (!S_ISDIR(inode.bi_mode))
-			break;
 
 		ret = darray_push(p, ((struct pathbuf_entry) {
 			.inum		= inode.bi_inum,
···
 }
 
 /*
- * Check for unreachable inodes, as well as loops in the directory structure:
- * After bch2_check_dirents(), if an inode backpointer doesn't exist that means it's
- * unreachable:
+ * Check for loops in the directory structure: all other connectivity issues
+ * have been fixed by prior passes
  */
 int bch2_check_directory_structure(struct bch_fs *c)
 {
···
 			if (S_ISDIR(u.bi_mode))
 				continue;
 
+			/*
+			 * Previous passes ensured that bi_nlink is nonzero if
+			 * it had multiple hardlinks:
+			 */
 			if (!u.bi_nlink)
 				continue;
 

fs/bcachefs/fsck.h

···
 int bch2_check_xattrs(struct bch_fs *);
 int bch2_check_root(struct bch_fs *);
 int bch2_check_subvolume_structure(struct bch_fs *);
+int bch2_check_unreachable_inodes(struct bch_fs *);
 int bch2_check_directory_structure(struct bch_fs *);
 int bch2_check_nlinks(struct bch_fs *);
 int bch2_fix_reflink_p(struct bch_fs *);

fs/bcachefs/inode.c

···
 #include "error.h"
 #include "extents.h"
 #include "extent_update.h"
+#include "fs.h"
 #include "inode.h"
 #include "str_hash.h"
 #include "snapshot.h"
···
 	NULL
 };
 #undef  x
+
+static int delete_ancestor_snapshot_inodes(struct btree_trans *, struct bpos);
 
 static const u8 byte_table[8] = { 1, 2, 3, 4, 6, 8, 10, 13 };
 
···
 	}
 }
 
-static inline bool bkey_is_deleted_inode(struct bkey_s_c k)
+static inline void bkey_inode_flags_set(struct bkey_s k, u64 f)
 {
-	return bkey_inode_flags(k) & BCH_INODE_unlinked;
+	switch (k.k->type) {
+	case KEY_TYPE_inode:
+		bkey_s_to_inode(k).v->bi_flags = cpu_to_le32(f);
+		return;
+	case KEY_TYPE_inode_v2:
+		bkey_s_to_inode_v2(k).v->bi_flags = cpu_to_le64(f);
+		return;
+	case KEY_TYPE_inode_v3:
+		bkey_s_to_inode_v3(k).v->bi_flags = cpu_to_le64(f);
+		return;
+	default:
+		BUG();
+	}
+}
+
+static inline bool bkey_is_unlinked_inode(struct bkey_s_c k)
+{
+	unsigned f = bkey_inode_flags(k) & BCH_INODE_unlinked;
+
+	return (f & BCH_INODE_unlinked) && !(f & BCH_INODE_has_child_snapshot);
+}
+
+static struct bkey_s_c
+bch2_bkey_get_iter_snapshot_parent(struct btree_trans *trans, struct btree_iter *iter,
+				   enum btree_id btree, struct bpos pos,
+				   unsigned flags)
+{
+	struct bch_fs *c = trans->c;
+	struct bkey_s_c k;
+	int ret = 0;
+
+	for_each_btree_key_upto_norestart(trans, *iter, btree,
+					  bpos_successor(pos),
+					  SPOS(pos.inode, pos.offset, U32_MAX),
+					  flags|BTREE_ITER_all_snapshots, k, ret)
+		if (bch2_snapshot_is_ancestor(c, pos.snapshot, k.k->p.snapshot))
+			return k;
+
+	bch2_trans_iter_exit(trans, iter);
+	return ret ? bkey_s_c_err(ret) : bkey_s_c_null;
+}
+
+static struct bkey_s_c
+bch2_inode_get_iter_snapshot_parent(struct btree_trans *trans, struct btree_iter *iter,
+				    struct bpos pos, unsigned flags)
+{
+	struct bkey_s_c k;
+again:
+	k = bch2_bkey_get_iter_snapshot_parent(trans, iter, BTREE_ID_inodes, pos, flags);
+	if (!k.k ||
+	    bkey_err(k) ||
+	    bkey_is_inode(k.k))
+		return k;
+
+	bch2_trans_iter_exit(trans, iter);
+	pos = k.k->p;
+	goto again;
+}
+
+int __bch2_inode_has_child_snapshots(struct btree_trans *trans, struct bpos pos)
+{
+	struct bch_fs *c = trans->c;
+	struct btree_iter iter;
+	struct bkey_s_c k;
+	int ret = 0;
+
+	for_each_btree_key_upto_norestart(trans, iter,
+			BTREE_ID_inodes, POS(0, pos.offset), bpos_predecessor(pos),
+			BTREE_ITER_all_snapshots|
+			BTREE_ITER_with_updates, k, ret)
+		if (bch2_snapshot_is_ancestor(c, k.k->p.snapshot, pos.snapshot) &&
+		    bkey_is_inode(k.k)) {
+			ret = 1;
+			break;
+		}
+	bch2_trans_iter_exit(trans, &iter);
+	return ret;
+}
+
+static int update_inode_has_children(struct btree_trans *trans,
+				     struct bkey_s k,
+				     bool have_child)
+{
+	if (!have_child) {
+		int ret = bch2_inode_has_child_snapshots(trans, k.k->p);
+		if (ret)
+			return ret < 0 ? ret : 0;
+	}
+
+	u64 f = bkey_inode_flags(k.s_c);
+	if (have_child != !!(f & BCH_INODE_has_child_snapshot))
+		bkey_inode_flags_set(k, f ^ BCH_INODE_has_child_snapshot);
+
+	return 0;
+}
+
+static int update_parent_inode_has_children(struct btree_trans *trans, struct bpos pos,
+					    bool have_child)
+{
+	struct btree_iter iter;
+	struct bkey_s_c k = bch2_inode_get_iter_snapshot_parent(trans,
+						&iter, pos, BTREE_ITER_with_updates);
+	int ret = bkey_err(k);
+	if (ret)
+		return ret;
+	if (!k.k)
+		return 0;
+
+	if (!have_child) {
+		ret = bch2_inode_has_child_snapshots(trans, k.k->p);
+		if (ret) {
+			ret = ret < 0 ? ret : 0;
+			goto err;
+		}
+	}
+
+	u64 f = bkey_inode_flags(k);
+	if (have_child != !!(f & BCH_INODE_has_child_snapshot)) {
+		struct bkey_i *update = bch2_bkey_make_mut(trans, &iter, &k,
+					     BTREE_UPDATE_internal_snapshot_node);
+		ret = PTR_ERR_OR_ZERO(update);
+		if (ret)
+			goto err;
+
+		bkey_inode_flags_set(bkey_i_to_s(update), f ^ BCH_INODE_has_child_snapshot);
+	}
+err:
+	bch2_trans_iter_exit(trans, &iter);
+	return ret;
 }
 
 int bch2_trigger_inode(struct btree_trans *trans,
···
 		       struct bkey_s new,
 		       enum btree_iter_update_trigger_flags flags)
 {
+	struct bch_fs *c = trans->c;
+
 	if ((flags & BTREE_TRIGGER_atomic) && (flags & BTREE_TRIGGER_insert)) {
 		BUG_ON(!trans->journal_res.seq);
 		bkey_s_to_inode_v3(new).v->bi_journal_seq = cpu_to_le64(trans->journal_res.seq);
···
 			return ret;
 	}
 
-	int deleted_delta =	(int) bkey_is_deleted_inode(new.s_c) -
-				(int) bkey_is_deleted_inode(old);
-	if ((flags & BTREE_TRIGGER_transactional) && deleted_delta) {
-		int ret = bch2_btree_bit_mod_buffered(trans, BTREE_ID_deleted_inodes,
-						      new.k->p, deleted_delta > 0);
-		if (ret)
-			return ret;
+	if (flags & BTREE_TRIGGER_transactional) {
+		int unlinked_delta =	(int) bkey_is_unlinked_inode(new.s_c) -
+					(int) bkey_is_unlinked_inode(old);
+		if (unlinked_delta) {
+			int ret = bch2_btree_bit_mod_buffered(trans, BTREE_ID_deleted_inodes,
+							      new.k->p, unlinked_delta > 0);
+			if (ret)
+				return ret;
+		}
+
+		/*
+		 * If we're creating or deleting an inode at this snapshot ID,
+		 * and there might be an inode in a parent snapshot ID, we might
+		 * need to set or clear the has_child_snapshot flag on the
+		 * parent.
+		 */
+		int deleted_delta = (int) bkey_is_inode(new.k) -
+				    (int) bkey_is_inode(old.k);
+		if (deleted_delta &&
+		    bch2_snapshot_parent(c, new.k->p.snapshot)) {
+			int ret = update_parent_inode_has_children(trans, new.k->p,
+								   deleted_delta > 0);
+			if (ret)
+				return ret;
+		}
+
+		/*
+		 * When an inode is first updated in a new snapshot, we may need
+		 * to clear has_child_snapshot
+		 */
+		if (deleted_delta > 0) {
+			int ret = update_inode_has_children(trans, new, false);
+			if (ret)
+				return ret;
+		}
 	}
 
 	return 0;
···
 	if (bch2_err_matches(ret, BCH_ERR_transaction_restart))
 		goto retry;
 
+	if (ret)
+		goto err2;
+
+	ret = delete_ancestor_snapshot_inodes(trans, SPOS(0, inum.inum, snapshot));
+err2:
 	bch2_trans_put(trans);
 	return ret;
 }
···
 	return 0;
 }
 
-int bch2_inode_rm_snapshot(struct btree_trans *trans, u64 inum, u32 snapshot)
+static noinline int __bch2_inode_rm_snapshot(struct btree_trans *trans, u64 inum, u32 snapshot)
 {
 	struct bch_fs *c = trans->c;
 	struct btree_iter iter = { NULL };
···
 	return ret ?: -BCH_ERR_transaction_restart_nested;
 }
 
+/*
+ * After deleting an inode, there may be versions in older snapshots that should
+ * also be deleted - if they're not referenced by sibling snapshots and not open
+ * in other subvolumes:
+ */
+static int delete_ancestor_snapshot_inodes(struct btree_trans *trans, struct bpos pos)
+{
+	struct btree_iter iter;
+	struct bkey_s_c k;
+	int ret;
+next_parent:
+	ret = lockrestart_do(trans,
+		bkey_err(k = bch2_inode_get_iter_snapshot_parent(trans, &iter, pos, 0)));
+	if (ret || !k.k)
+		return ret;
+
+	bool unlinked = bkey_is_unlinked_inode(k);
+	pos = k.k->p;
+	bch2_trans_iter_exit(trans, &iter);
+
+	if (!unlinked)
+		return 0;
+
+	ret = lockrestart_do(trans, bch2_inode_or_descendents_is_open(trans, pos));
+	if (ret)
+		return ret < 0 ? ret : 0;
+
+	ret = __bch2_inode_rm_snapshot(trans, pos.offset, pos.snapshot);
+	if (ret)
+		return ret;
+	goto next_parent;
+}
+
+int bch2_inode_rm_snapshot(struct btree_trans *trans, u64 inum, u32 snapshot)
+{
+	return __bch2_inode_rm_snapshot(trans, inum, snapshot) ?:
+		delete_ancestor_snapshot_inodes(trans, SPOS(0, inum, snapshot));
+}
+
 static int may_delete_deleted_inode(struct btree_trans *trans,
 				    struct btree_iter *iter,
 				    struct bpos pos,
···
 	struct btree_iter inode_iter;
 	struct bkey_s_c k;
 	struct bch_inode_unpacked inode;
+	struct printbuf buf = PRINTBUF;
 	int ret;
 
 	k = bch2_bkey_get_iter(trans, &inode_iter, BTREE_ID_inodes, pos, BTREE_ITER_cached);
···
 			pos.offset, pos.snapshot))
 		goto delete;
 
+	if (fsck_err_on(inode.bi_flags & BCH_INODE_has_child_snapshot,
+			trans, deleted_inode_has_child_snapshots,
+			"inode with child snapshots %llu:%u in deleted_inodes btree",
+			pos.offset, pos.snapshot))
+		goto delete;
+
+	ret = bch2_inode_has_child_snapshots(trans, k.k->p);
+	if (ret < 0)
+		goto out;
+
+	if (ret) {
+		if (fsck_err(trans, inode_has_child_snapshots_wrong,
+			     "inode has_child_snapshots flag wrong (should be set)\n%s",
+			     (printbuf_reset(&buf),
+			      bch2_inode_unpacked_to_text(&buf, &inode),
+			      buf.buf))) {
+			inode.bi_flags |= BCH_INODE_has_child_snapshot;
+			ret = __bch2_fsck_write_inode(trans, &inode);
+			if (ret)
+				goto out;
+		}
+		goto delete;
+
+	}
+
 	if (test_bit(BCH_FS_clean_recovery, &c->flags) &&
 	    !fsck_err(trans, deleted_inode_but_clean,
 		      "filesystem marked as clean but have deleted inode %llu:%u",
···
 		goto out;
 	}
 
-	if (bch2_snapshot_is_internal_node(c, pos.snapshot)) {
-		struct bpos new_min_pos;
-
-		ret = bch2_propagate_key_to_snapshot_leaves(trans, inode_iter.btree_id, k, &new_min_pos);
-		if (ret)
-			goto out;
-
-		inode.bi_flags &= ~BCH_INODE_unlinked;
-
-		ret = bch2_inode_write_flags(trans, &inode_iter, &inode,
-					     BTREE_UPDATE_internal_snapshot_node);
-		bch_err_msg(c, ret, "clearing inode unlinked flag");
-		if (ret)
-			goto out;
-
-		/*
-		 * We'll need another write buffer flush to pick up the new
-		 * unlinked inodes in the snapshot leaves:
-		 */
-		*need_another_pass = true;
-		goto out;
-	}
-
 	ret = 1;
 out:
 fsck_err:
 	bch2_trans_iter_exit(trans, &inode_iter);
+	printbuf_exit(&buf);
 	return ret;
 delete:
 	ret = bch2_btree_bit_mod_buffered(trans, BTREE_ID_deleted_inodes, pos, false);

fs/bcachefs/inode.h

···
 #include "bkey.h"
 #include "bkey_methods.h"
 #include "opts.h"
+#include "snapshot.h"
 
 enum bch_validate_flags;
 extern const char * const bch2_inode_opts[];
···
 int bch2_inode_v3_validate(struct bch_fs *, struct bkey_s_c,
 			  enum bch_validate_flags);
 void bch2_inode_to_text(struct printbuf *, struct bch_fs *, struct bkey_s_c);
+
+int __bch2_inode_has_child_snapshots(struct btree_trans *, struct bpos);
+
+static inline int bch2_inode_has_child_snapshots(struct btree_trans *trans, struct bpos pos)
+{
+	return bch2_snapshot_is_leaf(trans->c, pos.snapshot) <= 0
+		? __bch2_inode_has_child_snapshots(trans, pos)
+		: 0;
+}
 
 int bch2_trigger_inode(struct btree_trans *, enum btree_id, unsigned,
 		       struct bkey_s_c, struct bkey_s,

fs/bcachefs/inode_format.h

···
 	x(i_size_dirty,			5)	\
 	x(i_sectors_dirty,		6)	\
 	x(unlinked,			7)	\
-	x(backptr_untrusted,		8)
+	x(backptr_untrusted,		8)	\
+	x(has_child_snapshot,		9)
 
 /* bits 20+ reserved for packed fields below: */
 

fs/bcachefs/journal.c

···
 {
 	int ret;
 
+	if (closure_wait_event_timeout(&j->async_wait,
+		   (ret = __journal_res_get(j, res, flags)) != -BCH_ERR_journal_res_get_blocked ||
+		   (flags & JOURNAL_RES_GET_NONBLOCK),
+		   HZ * 10))
+		return ret;
+
+	struct bch_fs *c = container_of(j, struct bch_fs, journal);
+	struct printbuf buf = PRINTBUF;
+	bch2_journal_debug_to_text(&buf, j);
+	bch_err(c, "Journal stuck? Waited for 10 seconds...\n%s",
+		buf.buf);
+	printbuf_exit(&buf);
+
 	closure_wait_event(&j->async_wait,
 		   (ret = __journal_res_get(j, res, flags)) != -BCH_ERR_journal_res_get_blocked ||
 		   (flags & JOURNAL_RES_GET_NONBLOCK));

fs/bcachefs/opts.c

···
 			prt_printf(out, "%lli", v);
 		break;
 	case BCH_OPT_STR:
-		if (flags & OPT_SHOW_FULL_LIST)
+		if (v < opt->min || v >= opt->max - 1)
+			prt_printf(out, "(invalid option %lli)", v);
+		else if (flags & OPT_SHOW_FULL_LIST)
 			prt_string_option(out, opt->choices, v);
 		else
 			prt_str(out, opt->choices[v]);

fs/bcachefs/recovery.c

···
 				BCH_TRANS_COMMIT_no_enospc|
 				BCH_TRANS_COMMIT_journal_reclaim|
 				BCH_TRANS_COMMIT_skip_accounting_apply|
-				BCH_TRANS_COMMIT_no_journal_res,
+				BCH_TRANS_COMMIT_no_journal_res|
+				BCH_WATERMARK_reclaim,
 			     bch2_journal_replay_accounting_key(trans, k));
 		if (bch2_fs_fatal_err_on(ret, c, "error replaying accounting; %s", bch2_err_str(ret)))
 			goto err;

fs/bcachefs/recovery_passes_types.h

···
 	x(check_dirents,			27, PASS_FSCK)			\
 	x(check_xattrs,				28, PASS_FSCK)			\
 	x(check_root,				29, PASS_ONLINE|PASS_FSCK)	\
+	x(check_unreachable_inodes,		40, PASS_ONLINE|PASS_FSCK)	\
 	x(check_subvolume_structure,		36, PASS_ONLINE|PASS_FSCK)	\
 	x(check_directory_structure,		30, PASS_ONLINE|PASS_FSCK)	\
 	x(check_nlinks,				31, PASS_FSCK)			\

fs/bcachefs/replicas.c

···
 	prt_printf(out, "]");
 }
 
-static int bch2_replicas_entry_validate_locked(struct bch_replicas_entry_v1 *r,
-					       struct bch_sb *sb,
-					       struct printbuf *err)
+static int bch2_replicas_entry_sb_validate(struct bch_replicas_entry_v1 *r,
+					   struct bch_sb *sb,
+					   struct printbuf *err)
 {
 	if (!r->nr_devs) {
 		prt_printf(err, "no devices in entry ");
···
 				 struct bch_fs *c,
 				 struct printbuf *err)
 {
-	mutex_lock(&c->sb_lock);
-	int ret = bch2_replicas_entry_validate_locked(r, c->disk_sb.sb, err);
-	mutex_unlock(&c->sb_lock);
-	return ret;
+	if (!r->nr_devs) {
+		prt_printf(err, "no devices in entry ");
+		goto bad;
+	}
+
+	if (r->nr_required > 1 &&
+	    r->nr_required >= r->nr_devs) {
+		prt_printf(err, "bad nr_required in entry ");
+		goto bad;
+	}
+
+	for (unsigned i = 0; i < r->nr_devs; i++)
+		if (r->devs[i] != BCH_SB_MEMBER_INVALID &&
+		    !bch2_dev_exists(c, r->devs[i])) {
+			prt_printf(err, "invalid device %u in entry ", r->devs[i]);
+			goto bad;
+		}
+
+	return 0;
+bad:
+	bch2_replicas_entry_to_text(err, r);
+	return -BCH_ERR_invalid_replicas_entry;
 }
 
 void bch2_cpu_replicas_to_text(struct printbuf *out,
···
 		struct bch_replicas_entry_v1 *e =
 			cpu_replicas_entry(cpu_r, i);
 
-		int ret = bch2_replicas_entry_validate_locked(e, sb, err);
+		int ret = bch2_replicas_entry_sb_validate(e, sb, err);
 		if (ret)
 			return ret;
 
···
 
 		rcu_read_lock();
 		for (unsigned i = 0; i < e->nr_devs; i++) {
+			if (e->devs[i] == BCH_SB_MEMBER_INVALID) {
+				nr_failed++;
+				continue;
+			}
+
 			nr_online += test_bit(e->devs[i], devs.d);
 
 			struct bch_dev *ca = bch2_dev_rcu_noerror(c, e->devs[i]);

fs/bcachefs/sb-downgrade.c

···
 	  BCH_FSCK_ERR_accounting_mismatch)			\
 	x(rebalance_work_acct_fix,				\
 	  BIT_ULL(BCH_RECOVERY_PASS_check_allocations),		\
-	  BCH_FSCK_ERR_accounting_mismatch)
+	  BCH_FSCK_ERR_accounting_mismatch)			\
+	x(inode_has_child_snapshots,				\
+	  BIT_ULL(BCH_RECOVERY_PASS_check_inodes),		\
+	  BCH_FSCK_ERR_inode_has_child_snapshots_wrong)
 
 #define DOWNGRADE_TABLE()					\
 	x(bucket_stripe_sectors,				\

fs/bcachefs/sb-errors_format.h

···
 	x(reflink_p_to_missing_reflink_v,			166,	0)		\
 	x(stripe_pos_bad,					167,	0)		\
 	x(stripe_val_size_bad,					168,	0)		\
+	x(stripe_csum_granularity_bad,				290,	0)		\
 	x(stripe_sector_count_wrong,				169,	0)		\
 	x(snapshot_tree_pos_bad,				170,	0)		\
 	x(snapshot_tree_to_missing_snapshot,			171,	0)		\
···
 	x(inode_multiple_links_but_nlink_0,			207,	FSCK_AUTOFIX)	\
 	x(inode_wrong_backpointer,				208,	FSCK_AUTOFIX)	\
 	x(inode_wrong_nlink,					209,	FSCK_AUTOFIX)	\
+	x(inode_has_child_snapshots_wrong,			287,	0)		\
 	x(inode_unreachable,					210,	FSCK_AUTOFIX)	\
 	x(deleted_inode_but_clean,				211,	FSCK_AUTOFIX)	\
 	x(deleted_inode_missing,				212,	FSCK_AUTOFIX)	\
 	x(deleted_inode_is_dir,					213,	FSCK_AUTOFIX)	\
 	x(deleted_inode_not_unlinked,				214,	FSCK_AUTOFIX)	\
+	x(deleted_inode_has_child_snapshots,			288,	FSCK_AUTOFIX)	\
 	x(extent_overlapping,					215,	0)		\
 	x(key_in_missing_inode,					216,	0)		\
 	x(key_in_wrong_inode_type,				217,	0)		\
···
 	x(alloc_key_stripe_sectors_wrong,			271,	FSCK_AUTOFIX)	\
 	x(accounting_mismatch,					272,	FSCK_AUTOFIX)	\
 	x(accounting_replicas_not_marked,			273,	0)		\
+	x(accounting_to_invalid_device,				289,	0)		\
 	x(invalid_btree_id,					274,	0)		\
 	x(alloc_key_io_time_bad,				275,	0)		\
 	x(alloc_key_fragmentation_lru_wrong,			276,	FSCK_AUTOFIX)	\
···
 	x(accounting_key_replicas_devs_unsorted,		280,	FSCK_AUTOFIX)	\
 	x(accounting_key_version_0,				282,	FSCK_AUTOFIX)	\
 	x(logged_op_but_clean,					283,	FSCK_AUTOFIX)	\
-	x(MAX,							287,	0)
+	x(MAX,							291,	0)
 
 enum bch_sb_error_id {
 #define x(t, n, ...) BCH_FSCK_ERR_##t = n,

fs/bcachefs/sb-members.c

···
 		return -BCH_ERR_invalid_sb_members;
 	}
 
+	if (m.btree_bitmap_shift >= 64) {
+		prt_printf(err, "device %u: invalid btree_bitmap_shift %u", i, m.btree_bitmap_shift);
+		return -BCH_ERR_invalid_sb_members;
+	}
+
 	return 0;
 }
 
···
 	prt_newline(out);
 
 	prt_printf(out, "Btree allocated bitmap blocksize:\t");
-	prt_units_u64(out, 1ULL << m.btree_bitmap_shift);
+	if (m.btree_bitmap_shift < 64)
+		prt_units_u64(out, 1ULL << m.btree_bitmap_shift);
+	else
+		prt_printf(out, "(invalid shift %u)", m.btree_bitmap_shift);
 	prt_newline(out);
 
 	prt_printf(out, "Btree allocated bitmap:\t");

fs/bcachefs/snapshot.c

···
 	if (bch2_snapshot_equiv(c, id))
 		return 0;
 
-	/* 0 is an invalid tree ID */
+	/* Do we need to reconstruct the snapshot_tree entry as well? */
+	struct btree_iter iter;
+	struct bkey_s_c k;
+	int ret = 0;
 	u32 tree_id = 0;
-	int ret = bch2_snapshot_tree_create(trans, id, 0, &tree_id);
+
+	for_each_btree_key_norestart(trans, iter, BTREE_ID_snapshot_trees, POS_MIN,
+				     0, k, ret) {
+		if (le32_to_cpu(bkey_s_c_to_snapshot_tree(k).v->root_snapshot) == id) {
+			tree_id = k.k->p.offset;
+			break;
+		}
+	}
+	bch2_trans_iter_exit(trans, &iter);
+
 	if (ret)
 		return ret;
+
+	if (!tree_id) {
+		ret = bch2_snapshot_tree_create(trans, id, 0, &tree_id);
+		if (ret)
+			return ret;
+	}
 
 	struct bkey_i_snapshot *snapshot = bch2_trans_kmalloc(trans, sizeof(*snapshot));
 	ret = PTR_ERR_OR_ZERO(snapshot);
···
 	snapshot->k.p		= POS(0, id);
 	snapshot->v.tree	= cpu_to_le32(tree_id);
 	snapshot->v.btime.lo	= cpu_to_le64(bch2_current_time(c));
+
+	for_each_btree_key_norestart(trans, iter, BTREE_ID_subvolumes, POS_MIN,
+				     0, k, ret) {
+		if (le32_to_cpu(bkey_s_c_to_subvolume(k).v->snapshot) == id) {
+			snapshot->v.subvol = cpu_to_le32(k.k->p.offset);
+			SET_BCH_SNAPSHOT_SUBVOL(&snapshot->v, true);
+			break;
+		}
+	}
+	bch2_trans_iter_exit(trans, &iter);
 
 	return  bch2_btree_insert_trans(trans, BTREE_ID_snapshots, &snapshot->k_i, 0) ?:
 		bch2_mark_snapshot(trans, BTREE_ID_snapshots, 0,
···
 	bch2_trans_iter_exit(trans, &iter);
 
 	return ret;
-}
-
-static u32 bch2_snapshot_smallest_child(struct bch_fs *c, u32 id)
-{
-	const struct snapshot_t *s = snapshot_t(c, id);
-
-	return s->children[1] ?: s->children[0];
-}
-
-static u32 bch2_snapshot_smallest_descendent(struct bch_fs *c, u32 id)
-{
-	u32 child;
-
-	while ((child = bch2_snapshot_smallest_child(c, id)))
-		id = child;
-	return id;
-}
-
-static int bch2_propagate_key_to_snapshot_leaf(struct btree_trans *trans,
-					       enum btree_id btree,
-					       struct bkey_s_c interior_k,
-					       u32 leaf_id, struct bpos *new_min_pos)
-{
-	struct btree_iter iter;
-	struct bpos pos = interior_k.k->p;
-	struct bkey_s_c k;
-	struct bkey_i *new;
-	int ret;
-
-	pos.snapshot = leaf_id;
-
-	bch2_trans_iter_init(trans, &iter, btree, pos, BTREE_ITER_intent);
-	k = bch2_btree_iter_peek_slot(&iter);
-	ret = bkey_err(k);
-	if (ret)
-		goto out;
-
-	/* key already overwritten in this snapshot? */
-	if (k.k->p.snapshot != interior_k.k->p.snapshot)
-		goto out;
-
-	if (bpos_eq(*new_min_pos, POS_MIN)) {
-		*new_min_pos = k.k->p;
-		new_min_pos->snapshot = leaf_id;
-	}
-
-	new = bch2_bkey_make_mut_noupdate(trans, interior_k);
-	ret = PTR_ERR_OR_ZERO(new);
-	if (ret)
-		goto out;
-
-	new->k.p.snapshot = leaf_id;
-	ret = bch2_trans_update(trans, &iter, new, 0);
-out:
-	bch2_set_btree_iter_dontneed(&iter);
-	bch2_trans_iter_exit(trans, &iter);
-	return ret;
-}
-
-int bch2_propagate_key_to_snapshot_leaves(struct btree_trans *trans,
-					  enum btree_id btree,
-					  struct bkey_s_c k,
-					  struct bpos *new_min_pos)
-{
-	struct bch_fs *c = trans->c;
-	struct bkey_buf sk;
-	u32 restart_count = trans->restart_count;
-	int ret = 0;
-
-	bch2_bkey_buf_init(&sk);
-	bch2_bkey_buf_reassemble(&sk, c, k);
-	k = bkey_i_to_s_c(sk.k);
-
-	*new_min_pos = POS_MIN;
-
-	for (u32 id = bch2_snapshot_smallest_descendent(c, k.k->p.snapshot);
-	     id < k.k->p.snapshot;
-	     id++) {
-		if (!bch2_snapshot_is_ancestor(c, id, k.k->p.snapshot) ||
-		    !bch2_snapshot_is_leaf(c, id))
-			continue;
-again:
-		ret =   btree_trans_too_many_iters(trans) ?:
-			bch2_propagate_key_to_snapshot_leaf(trans, btree, k, id, new_min_pos) ?:
-			bch2_trans_commit(trans, NULL, NULL, 0);
-		if (ret && bch2_err_matches(ret, BCH_ERR_transaction_restart)) {
-			bch2_trans_begin(trans);
-			goto again;
-		}
-
-		if (ret)
-			break;
-	}
-
-	bch2_bkey_buf_exit(&sk, c);
-
-	return ret ?: trans_was_restarted(trans, restart_count);
 }
 
 static int bch2_check_snapshot_needs_deletion(struct btree_trans *trans, struct bkey_s_c k)

fs/bcachefs/snapshot.h

···
 	return __bch2_key_has_snapshot_overwrites(trans, id, pos);
 }
 
-int bch2_propagate_key_to_snapshot_leaves(struct btree_trans *, enum btree_id,
-					  struct bkey_s_c, struct bpos *);
-
 int bch2_snapshots_read(struct bch_fs *);
 void bch2_fs_snapshots_exit(struct bch_fs *);
 

fs/bcachefs/super.c

···
 
 DECLARE_WAIT_QUEUE_HEAD(bch2_read_only_wait);
 
+static void bch2_dev_unlink(struct bch_dev *);
 static void bch2_dev_free(struct bch_dev *);
 static int bch2_dev_alloc(struct bch_fs *, unsigned);
 static int bch2_dev_sysfs_online(struct bch_fs *, struct bch_dev *);
···
 	up_write(&c->state_lock);
 
 	for_each_member_device(c, ca)
-		if (ca->kobj.state_in_sysfs &&
-		    ca->disk_sb.bdev)
-			sysfs_remove_link(bdev_kobj(ca->disk_sb.bdev), "bcachefs");
+		bch2_dev_unlink(ca);
 
 	if (c->kobj.state_in_sysfs)
 		kobject_del(&c->kobj);
···
 {
 	cancel_work_sync(&ca->io_error_work);
 
-	if (ca->kobj.state_in_sysfs &&
-	    ca->disk_sb.bdev)
-		sysfs_remove_link(bdev_kobj(ca->disk_sb.bdev), "bcachefs");
+	bch2_dev_unlink(ca);
 
 	if (ca->kobj.state_in_sysfs)
 		kobject_del(&ca->kobj);
···
 	percpu_ref_kill(&ca->io_ref);
 	wait_for_completion(&ca->io_ref_completion);
 
-	if (ca->kobj.state_in_sysfs) {
-		sysfs_remove_link(bdev_kobj(ca->disk_sb.bdev), "bcachefs");
-		sysfs_remove_link(&ca->kobj, "block");
-	}
+	bch2_dev_unlink(ca);
 
 	bch2_free_super(&ca->disk_sb);
 	bch2_dev_journal_exit(ca);
···
 	struct bch_dev *ca = container_of(ref, struct bch_dev, io_ref);
 
 	complete(&ca->io_ref_completion);
+}
+
+static void bch2_dev_unlink(struct bch_dev *ca)
+{
+	struct kobject *b;
+
+	/*
+	 * This is racy w.r.t. the underlying block device being hot-removed,
+	 * which removes it from sysfs.
+	 *
+	 * It'd be lovely if we had a way to handle this race, but the sysfs
+	 * code doesn't appear to provide a good method and block/holder.c is
+	 * susceptible as well:
+	 */
+	if (ca->kobj.state_in_sysfs &&
+	    ca->disk_sb.bdev &&
+	    (b = bdev_kobj(ca->disk_sb.bdev))->state_in_sysfs) {
+		sysfs_remove_link(b, "bcachefs");
+		sysfs_remove_link(&ca->kobj, "block");
+	}
 }
 
 static int bch2_dev_sysfs_online(struct bch_fs *c, struct bch_dev *ca)

fs/btrfs/delayed-ref.c

···
 		     struct btrfs_qgroup_extent_record *qrecord,
 		     int action, bool *qrecord_inserted_ret)
 {
+	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_delayed_ref_head *existing;
 	struct btrfs_delayed_ref_root *delayed_refs;
 	bool qrecord_inserted = false;
···
 	if (qrecord) {
 		int ret;
 
-		ret = btrfs_qgroup_trace_extent_nolock(trans->fs_info,
-						       delayed_refs, qrecord);
+		ret = btrfs_qgroup_trace_extent_nolock(fs_info, delayed_refs, qrecord);
 		if (ret) {
 			/* Clean up if insertion fails or item exists. */
-			xa_release(&delayed_refs->dirty_extents, qrecord->bytenr);
+			xa_release(&delayed_refs->dirty_extents,
+				   qrecord->bytenr >> fs_info->sectorsize_bits);
 			/* Caller responsible for freeing qrecord on error. */
 			if (ret < 0)
 				return ERR_PTR(ret);
···
 		}
 	}
 
-	trace_add_delayed_ref_head(trans->fs_info, head_ref, action);
+	trace_add_delayed_ref_head(fs_info, head_ref, action);
 
 	existing = htree_insert(&delayed_refs->href_root,
 				&head_ref->href_node);
···
 		if (head_ref->is_data && head_ref->ref_mod < 0) {
 			delayed_refs->pending_csums += head_ref->num_bytes;
 			trans->delayed_ref_csum_deletions +=
-				btrfs_csum_bytes_to_leaves(trans->fs_info,
-							   head_ref->num_bytes);
+				btrfs_csum_bytes_to_leaves(fs_info, head_ref->num_bytes);
 		}
 		delayed_refs->num_heads++;
 		delayed_refs->num_heads_ready++;
···
 			goto free_head_ref;
 		}
 		if (xa_reserve(&trans->transaction->delayed_refs.dirty_extents,
-			       generic_ref->bytenr, GFP_NOFS)) {
+			       generic_ref->bytenr >> fs_info->sectorsize_bits,
+			       GFP_NOFS)) {
 			ret = -ENOMEM;
 			goto free_record;
 		}

fs/btrfs/delayed-ref.h

···
 	/* head ref rbtree */
 	struct rb_root_cached href_root;
 
-	/* Track dirty extent records. */
+	/*
+	 * Track dirty extent records.
+	 * The keys correspond to the logical address of the extent ("bytenr")
+	 * right shifted by fs_info->sectorsize_bits. This is both to get a more
+	 * dense index space (optimizes xarray structure) and because indexes in
+	 * xarrays are of "unsigned long" type, meaning they are 32 bits wide on
+	 * 32 bits platforms, limiting the extent range to 4G which is too low
+	 * and makes it unusable (truncated index values) on 32 bits platforms.
+	 */
 	struct xarray dirty_extents;
 
 	/* this spin lock protects the rbtree and the entries inside */

fs/btrfs/qgroup.c

···
 				struct btrfs_qgroup_extent_record *record)
 {
 	struct btrfs_qgroup_extent_record *existing, *ret;
-	unsigned long bytenr = record->bytenr;
+	const unsigned long index = (record->bytenr >> fs_info->sectorsize_bits);
 
 	if (!btrfs_qgroup_full_accounting(fs_info))
 		return 1;
+
+#if BITS_PER_LONG == 32
+	if (record->bytenr >= MAX_LFS_FILESIZE) {
+		btrfs_err_rl(fs_info,
+"qgroup record for extent at %llu is beyond 32bit page cache and xarray index limit",
+			     record->bytenr);
+		btrfs_err_32bit_limit(fs_info);
+		return -EOVERFLOW;
+	}
+#endif
 
 	lockdep_assert_held(&delayed_refs->lock);
 	trace_btrfs_qgroup_trace_extent(fs_info, record);
 
 	xa_lock(&delayed_refs->dirty_extents);
-	existing = xa_load(&delayed_refs->dirty_extents, bytenr);
+	existing = xa_load(&delayed_refs->dirty_extents, index);
 	if (existing) {
 		if (record->data_rsv && !existing->data_rsv) {
 			existing->data_rsv = record->data_rsv;
···
 		return 1;
 	}
 
-	ret = __xa_store(&delayed_refs->dirty_extents, record->bytenr, record, GFP_ATOMIC);
+	ret = __xa_store(&delayed_refs->dirty_extents, index, record, GFP_ATOMIC);
 	xa_unlock(&delayed_refs->dirty_extents);
 	if (xa_is_err(ret)) {
 		qgroup_mark_inconsistent(fs_info);
···
 	struct btrfs_fs_info *fs_info = trans->fs_info;
 	struct btrfs_qgroup_extent_record *record;
 	struct btrfs_delayed_ref_root *delayed_refs;
+	const unsigned long index = (bytenr >> fs_info->sectorsize_bits);
 	int ret;
 
 	if (!btrfs_qgroup_full_accounting(fs_info) || bytenr == 0 || num_bytes == 0)
···
 	if (!record)
 		return -ENOMEM;
 
-	if (xa_reserve(&trans->transaction->delayed_refs.dirty_extents, bytenr, GFP_NOFS)) {
+	if (xa_reserve(&trans->transaction->delayed_refs.dirty_extents, index, GFP_NOFS)) {
 		kfree(record);
 		return -ENOMEM;
 	}
···
 	spin_unlock(&delayed_refs->lock);
 	if (ret) {
 		/* Clean up if insertion fails or item exists. */
-		xa_release(&delayed_refs->dirty_extents, record->bytenr);
+		xa_release(&delayed_refs->dirty_extents, index);
 		kfree(record);
 		return 0;
 	}

fs/btrfs/send.c

···
 
 static int changed_verity(struct send_ctx *sctx, enum btrfs_compare_tree_result result)
 {
-	int ret = 0;
-
 	if (!sctx->cur_inode_new_gen && !sctx->cur_inode_deleted) {
 		if (result == BTRFS_COMPARE_TREE_NEW)
 			sctx->cur_inode_needs_verity = true;
 	}
-	return ret;
+	return 0;
 }
 
 static int dir_changed(struct send_ctx *sctx, u64 dir)

fs/btrfs/tree-log.c

···
 	struct inode *inode = NULL;
 	unsigned long ref_ptr;
 	unsigned long ref_end;
-	struct fscrypt_str name;
+	struct fscrypt_str name = { 0 };
 	int ret;
 	int log_ref_ver = 0;
 	u64 parent_objectid;
···
 				    struct btrfs_dir_item *di,
 				    struct btrfs_key *key)
 {
-	struct fscrypt_str name;
+	struct fscrypt_str name = { 0 };
 	struct btrfs_dir_item *dir_dst_di;
 	struct btrfs_dir_item *index_dst_di;
 	bool dir_dst_matches = false;
···
 	struct extent_buffer *eb;
 	int slot;
 	struct btrfs_dir_item *di;
-	struct fscrypt_str name;
+	struct fscrypt_str name = { 0 };
 	struct inode *inode = NULL;
 	struct btrfs_key location;
 

fs/erofs/super.c

···
 		if (IS_ERR(file))
 			return PTR_ERR(file);
 
-		dif->file = file;
-		if (!erofs_is_fileio_mode(sbi))
+		if (!erofs_is_fileio_mode(sbi)) {
 			dif->dax_dev = fs_dax_get_by_bdev(file_bdev(file),
 					&dif->dax_part_off, NULL, NULL);
+		} else if (!S_ISREG(file_inode(file)->i_mode)) {
+			fput(file);
+			return -EINVAL;
+		}
+		dif->file = file;
 	}
 
 	dif->blocks = le32_to_cpu(dis->blocks);
···
 		if (IS_ERR(sbi->fdev))
 			return PTR_ERR(sbi->fdev);
 
-		return get_tree_nodev(fc, erofs_fc_fill_super);
+		if (S_ISREG(file_inode(sbi->fdev)->i_mode) &&
+		    sbi->fdev->f_mapping->a_ops->read_folio)
+			return get_tree_nodev(fc, erofs_fc_fill_super);
+		fput(sbi->fdev);
 	}
 #endif
 	return ret;

fs/erofs/zdata.c

···
 	return ret;
 }
 
-static void z_erofs_try_to_claim_pcluster(struct z_erofs_decompress_frontend *f)
-{
-	struct z_erofs_pcluster *pcl = f->pcl;
-	z_erofs_next_pcluster_t *owned_head = &f->owned_head;
-
-	/* type 1, nil pcluster (this pcluster doesn't belong to any chain.) */
-	if (cmpxchg(&pcl->next, Z_EROFS_PCLUSTER_NIL,
-		    *owned_head) == Z_EROFS_PCLUSTER_NIL) {
-		*owned_head = &pcl->next;
-		/* so we can attach this pcluster to our submission chain. */
-		f->mode = Z_EROFS_PCLUSTER_FOLLOWED;
-		return;
-	}
-
-	/* type 2, it belongs to an ongoing chain */
-	f->mode = Z_EROFS_PCLUSTER_INFLIGHT;
-}
-
 static int z_erofs_register_pcluster(struct z_erofs_decompress_frontend *fe)
 {
 	struct erofs_map_blocks *map = &fe->map;
···
 	int ret;
 
 	DBG_BUGON(fe->pcl);
-
 	/* must be Z_EROFS_PCLUSTER_TAIL or pointed to previous pcluster */
 	DBG_BUGON(fe->owned_head == Z_EROFS_PCLUSTER_NIL);
 
···
 
 	if (ret == -EEXIST) {
 		mutex_lock(&fe->pcl->lock);
-		z_erofs_try_to_claim_pcluster(fe);
+		/* check if this pcluster hasn't been linked into any chain. */
+		if (cmpxchg(&fe->pcl->next, Z_EROFS_PCLUSTER_NIL,
+			    fe->owned_head) == Z_EROFS_PCLUSTER_NIL) {
+			/* .. so it can be attached to our submission chain */
+			fe->owned_head = &fe->pcl->next;
+			fe->mode = Z_EROFS_PCLUSTER_FOLLOWED;
+		} else {	/* otherwise, it belongs to an inflight chain */
+			fe->mode = Z_EROFS_PCLUSTER_INFLIGHT;
+		}
 	} else if (ret) {
 		return ret;
 	}

fs/erofs/zmap.c

···
 struct z_erofs_maprecorder {
 	struct inode *inode;
 	struct erofs_map_blocks *map;
-	void *kaddr;
-
 	unsigned long lcn;
 	/* compression extent information gathered */
 	u8  type, headtype;
···
 	struct z_erofs_lcluster_index *di;
 	unsigned int advise;
 
-	m->kaddr = erofs_read_metabuf(&m->map->buf, inode->i_sb,
-				      pos, EROFS_KMAP);
-	if (IS_ERR(m->kaddr))
-		return PTR_ERR(m->kaddr);
-
-	m->nextpackoff = pos + sizeof(struct z_erofs_lcluster_index);
+	di = erofs_read_metabuf(&m->map->buf, inode->i_sb, pos, EROFS_KMAP);
+	if (IS_ERR(di))
+		return PTR_ERR(di);
 	m->lcn = lcn;
-	di = m->kaddr;
+	m->nextpackoff = pos + sizeof(struct z_erofs_lcluster_index);
 
 	advise = le16_to_cpu(di->di_advise);
 	m->type = advise & Z_EROFS_LI_LCLUSTER_TYPE_MASK;
···
 				DBG_BUGON(1);
 				return -EFSCORRUPTED;
 			}
-			m->compressedblks = m->delta[0] &
-				~Z_EROFS_LI_D0_CBLKCNT;
+			m->compressedblks = m->delta[0] & ~Z_EROFS_LI_D0_CBLKCNT;
 			m->delta[0] = 1;
 		}
 		m->delta[1] = le16_to_cpu(di->di_u.delta[1]);
···
 	struct erofs_inode *const vi = EROFS_I(m->inode);
 	const unsigned int lclusterbits = vi->z_logical_clusterbits;
 	unsigned int vcnt, lo, lobits, encodebits, nblk, bytes;
-	int i;
-	u8 *in, type;
 	bool big_pcluster;
+	u8 *in, type;
+	int i;
 
 	if (1 << amortizedshift == 4 && lclusterbits <= 14)
 		vcnt = 2;
···
 	else
 		return -EOPNOTSUPP;
 
+	in = erofs_read_metabuf(&m->map->buf, m->inode->i_sb, pos, EROFS_KMAP);
+	if (IS_ERR(in))
+		return PTR_ERR(in);
+
 	/* it doesn't equal to round_up(..) */
 	m->nextpackoff = round_down(pos, vcnt << amortizedshift) +
 			 (vcnt << amortizedshift);
···
 	lobits = max(lclusterbits, ilog2(Z_EROFS_LI_D0_CBLKCNT) + 1U);
 	encodebits = ((vcnt << amortizedshift) - sizeof(__le32)) * 8 / vcnt;
 	bytes = pos & ((vcnt << amortizedshift) - 1);
-
-	in = m->kaddr - bytes;
-
+	in -= bytes;
 	i = bytes >> amortizedshift;
 
 	lo = decode_compactedbits(lobits, in, encodebits * i, &type);
···
 	amortizedshift = 2;
 out:
 	pos += lcn * (1 << amortizedshift);
-	m->kaddr = erofs_read_metabuf(&m->map->buf, inode->i_sb,
-				      pos, EROFS_KMAP);
-	if (IS_ERR(m->kaddr))
-		return PTR_ERR(m->kaddr);
 	return unpack_compacted_index(m, amortizedshift, pos, lookahead);
 }
 

fs/f2fs/file.c

···
 					iov_iter_count(to), READ);
 
 	/* In LFS mode, if there is inflight dio, wait for its completion */
-	if (f2fs_lfs_mode(F2FS_I_SB(inode)))
+	if (f2fs_lfs_mode(F2FS_I_SB(inode)) &&
+	    get_pages(F2FS_I_SB(inode), F2FS_DIO_WRITE))
 		inode_dio_wait(inode);
 
 	if (f2fs_should_use_dio(inode, iocb, to)) {

fs/nfs/callback_xdr.c

···
 
 	rc_list->rcl_nrefcalls = ntohl(*p++);
 	if (rc_list->rcl_nrefcalls) {
+		if (unlikely(rc_list->rcl_nrefcalls > xdr->buf->len))
+			goto out;
 		p = xdr_inline_decode(xdr,
 			     rc_list->rcl_nrefcalls * 2 * sizeof(uint32_t));
 		if (unlikely(p == NULL))

fs/nfs/client.c

···
 	INIT_LIST_HEAD(&server->layouts);
 	INIT_LIST_HEAD(&server->state_owners_lru);
 	INIT_LIST_HEAD(&server->ss_copies);
+	INIT_LIST_HEAD(&server->ss_src_copies);
 
 	atomic_set(&server->active, 0);
 

fs/nfs/delegation.c

···
 	}
 
 	nfs_mark_delegation_revoked(delegation);
+	clear_bit(NFS_DELEGATION_RETURNING, &delegation->flags);
+	spin_unlock(&delegation->lock);
+	if (nfs_detach_delegation(NFS_I(inode), delegation, NFS_SERVER(inode)))
+		nfs_put_delegation(delegation);
+	goto out_rcu_unlock;
 
 out_clear_returning:
 	clear_bit(NFS_DELEGATION_RETURNING, &delegation->flags);

fs/nfs/localio.c

···
 #include <net/addrconf.h>
 #include <linux/nfs_common.h>
 #include <linux/nfslocalio.h>
-#include <linux/module.h>
 #include <linux/bvec.h>
 
 #include <linux/nfs.h>
···
 {
 	struct nfs_pgio_header *hdr = iocb->hdr;
 
-	nfs_to->nfsd_file_put_local(iocb->localio);
+	nfs_to_nfsd_file_put_local(iocb->localio);
 	nfs_local_iocb_free(iocb);
 	nfs_local_hdr_release(hdr, hdr->task.tk_ops);
 }
···
 	}
 out:
 	if (status != 0) {
-		nfs_to->nfsd_file_put_local(localio);
+		nfs_to_nfsd_file_put_local(localio);
 		hdr->task.tk_status = status;
 		nfs_local_hdr_release(hdr, call_ops);
 	}
···
 		struct nfs_commit_data *data,
 		const struct rpc_call_ops *call_ops)
 {
-	nfs_to->nfsd_file_put_local(localio);
+	nfs_to_nfsd_file_put_local(localio);
 	call_ops->rpc_call_done(&data->task, data);
 	call_ops->rpc_release(data);
 }

fs/nfs/nfs42proc.c

···
 
 	if (dst_server != src_server) {
 		spin_lock(&src_server->nfs_client->cl_lock);
-		list_add_tail(&copy->src_copies, &src_server->ss_copies);
+		list_add_tail(&copy->src_copies, &src_server->ss_src_copies);
 		spin_unlock(&src_server->nfs_client->cl_lock);
 	}
 

fs/nfs/nfs4state.c

···
 			complete(&copy->completion);
 		}
 	}
-	list_for_each_entry(copy, &sp->so_server->ss_copies, src_copies) {
+	list_for_each_entry(copy, &sp->so_server->ss_src_copies, src_copies) {
 		if ((test_bit(NFS_CLNT_SRC_SSC_COPY_STATE, &state->flags) &&
 				!nfs4_stateid_match_other(&state->stateid,
 				&copy->parent_src_state->stateid)))

fs/nfs_common/nfslocalio.c

···
 	/* We have an implied reference to net thanks to nfsd_serv_try_get */
 	localio = nfs_to->nfsd_open_local_fh(net, uuid->dom, rpc_clnt,
 					     cred, nfs_fh, fmode);
-	if (IS_ERR(localio))
+	if (IS_ERR(localio)) {
+		rcu_read_lock();
 		nfs_to->nfsd_serv_put(net);
+		rcu_read_unlock();
+	}
 	return localio;
 }
 EXPORT_SYMBOL_GPL(nfs_open_local_fh);

fs/nfsd/filecache.c

···
  * reference to the associated nn->nfsd_serv.
  */
 void
-nfsd_file_put_local(struct nfsd_file *nf)
+nfsd_file_put_local(struct nfsd_file *nf) __must_hold(rcu)
 {
 	struct net *net = nf->nf_net;
 

fs/nfsd/localio.c

···
  *
  * On successful return, returned nfsd_file will have its nf_net member
  * set. Caller (NFS client) is responsible for calling nfsd_serv_put and
- * nfsd_file_put (via nfs_to->nfsd_file_put_local).
+ * nfsd_file_put (via nfs_to_nfsd_file_put_local).
  */
 struct nfsd_file *
 nfsd_open_local_fh(struct net *net, struct auth_domain *dom,

fs/nfsd/nfssvc.c

···
 	return 0;
 }
 
-bool nfsd_serv_try_get(struct net *net)
+bool nfsd_serv_try_get(struct net *net) __must_hold(rcu)
 {
 	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 
 	return (nn && percpu_ref_tryget_live(&nn->nfsd_serv_ref));
 }
 
-void nfsd_serv_put(struct net *net)
+void nfsd_serv_put(struct net *net) __must_hold(rcu)
 {
 	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 

fs/nfsd/trace.h

···
 	),
 
 	TP_fast_assign(
-		__entry->xid = be32_to_cpu(rqstp->rq_xid);
+		__entry->xid = rqstp ? be32_to_cpu(rqstp->rq_xid) : 0;
 		__entry->inode = inode;
 		__entry->may_flags = may_flags;
 		__entry->nf_ref = nf ? refcount_read(&nf->nf_ref) : 0;
···
 		__field(long, error)
 	),
 	TP_fast_assign(
-		__entry->xid = be32_to_cpu(rqstp->rq_xid);
+		__entry->xid = rqstp ? be32_to_cpu(rqstp->rq_xid) : 0;
 		__entry->inode = inode;
 		__entry->may_flags = may_flags;
 		__entry->error = error;
···
 		__field(const void *, nf_file)
 	),
 	TP_fast_assign(
-		__entry->xid = be32_to_cpu(rqstp->rq_xid);
+		__entry->xid = rqstp ? be32_to_cpu(rqstp->rq_xid) : 0;
 		__entry->inode = inode;
 		__entry->may_flags = may_flags;
 		__entry->nf_ref = refcount_read(&nf->nf_ref);

fs/smb/client/cifs_unicode.c

···
 			/**
 			 * Remap spaces and periods found at the end of every
 			 * component of the path. The special cases of '.' and
-			 * '..' do not need to be dealt with explicitly because
-			 * they are addressed in namei.c:link_path_walk().
+			 * '..' are need to be handled because of symlinks.
+			 * They are treated as non-end-of-string to avoid
+			 * remapping and breaking symlinks pointing to . or ..
 			 **/
-			if ((i == srclen - 1) || (source[i+1] == '\\'))
+			if ((i == 0 || source[i-1] == '\\') &&
+			    source[i] == '.' &&
+			    (i == srclen-1 || source[i+1] == '\\'))
+				end_of_string = false; /* "." case */
+			else if (i >= 1 &&
+				 (i == 1 || source[i-2] == '\\') &&
+				 source[i-1] == '.' &&
+				 source[i] == '.' &&
+				 (i == srclen-1 || source[i+1] == '\\'))
+				end_of_string = false; /* ".." case */
+			else if ((i == srclen - 1) || (source[i+1] == '\\'))
 				end_of_string = true;
 			else
 				end_of_string = false;

fs/smb/client/reparse.c

···
 #include "fs_context.h"
 #include "reparse.h"
 
+static int detect_directory_symlink_target(struct cifs_sb_info *cifs_sb,
+					   const unsigned int xid,
+					   const char *full_path,
+					   const char *symname,
+					   bool *directory);
+
 int smb2_create_reparse_symlink(const unsigned int xid, struct inode *inode,
 				struct dentry *dentry, struct cifs_tcon *tcon,
 				const char *full_path, const char *symname)
···
 	struct inode *new;
 	struct kvec iov;
 	__le16 *path;
+	bool directory;
 	char *sym, sep = CIFS_DIR_SEP(cifs_sb);
 	u16 len, plen;
 	int rc = 0;
···
 		rc = -ENOMEM;
 		goto out;
 	}
+
+	/*
+	 * SMB distinguish between symlink to directory and symlink to file.
+	 * They cannot be exchanged (symlink of file type which points to
+	 * directory cannot be resolved and vice-versa). Try to detect if
+	 * the symlink target could be a directory or not. When detection
+	 * fails then treat symlink as a file (non-directory) symlink.
+	 */
+	directory = false;
+	rc = detect_directory_symlink_target(cifs_sb, xid, full_path, symname, &directory);
+	if (rc < 0)
+		goto out;
 
 	plen = 2 * UniStrnlen((wchar_t *)path, PATH_MAX);
 	len = sizeof(*buf) + plen * 2;
···
 	iov.iov_base = buf;
 	iov.iov_len = len;
 	new = smb2_get_reparse_inode(&data, inode->i_sb, xid,
-				     tcon, full_path, &iov, NULL);
+				     tcon, full_path, directory,
+				     &iov, NULL);
 	if (!IS_ERR(new))
 		d_instantiate(dentry, new);
 	else
···
 	cifs_free_open_info(&data);
 	kfree(buf);
 	return rc;
+}
+
+static int detect_directory_symlink_target(struct cifs_sb_info *cifs_sb,
+					   const unsigned int xid,
+					   const char *full_path,
+					   const char *symname,
+					   bool *directory)
+{
+	char sep = CIFS_DIR_SEP(cifs_sb);
+	struct cifs_open_parms oparms;
+	struct tcon_link *tlink;
+	struct cifs_tcon *tcon;
+	const char *basename;
+	struct cifs_fid fid;
+	char *resolved_path;
+	int full_path_len;
+	int basename_len;
+	int symname_len;
+	char *path_sep;
+	__u32 oplock;
+	int open_rc;
+
+	/*
+	 * First do some simple check. If the original Linux symlink target ends
+	 * with slash, or last path component is dot or dot-dot then it is for
+	 * sure symlink to the directory.
+	 */
+	basename = kbasename(symname);
+	basename_len = strlen(basename);
+	if (basename_len == 0 || /* symname ends with slash */
+	    (basename_len == 1 && basename[0] == '.') || /* last component is "." */
+	    (basename_len == 2 && basename[0] == '.' && basename[1] == '.')) { /* or ".." */
+		*directory = true;
+		return 0;
+	}
+
+	/*
+	 * For absolute symlinks it is not possible to determinate
+	 * if it should point to directory or file.
+	 */
+	if (symname[0] == '/') {
+		cifs_dbg(FYI,
+			 "%s: cannot determinate if the symlink target path '%s' "
+			 "is directory or not, creating '%s' as file symlink\n",
+			 __func__, symname, full_path);
+		return 0;
+	}
+
+	/*
+	 * If it was not detected as directory yet and the symlink is relative
+	 * then try to resolve the path on the SMB server, check if the path
+	 * exists and determinate if it is a directory or not.
+	 */
+
+	full_path_len = strlen(full_path);
+	symname_len = strlen(symname);
+
+	tlink = cifs_sb_tlink(cifs_sb);
+	if (IS_ERR(tlink))
+		return PTR_ERR(tlink);
+
+	resolved_path = kzalloc(full_path_len + symname_len + 1, GFP_KERNEL);
+	if (!resolved_path) {
+		cifs_put_tlink(tlink);
+		return -ENOMEM;
+	}
+
+	/*
+	 * Compose the resolved SMB symlink path from the SMB full path
+	 * and Linux target symlink path.
+	 */
+	memcpy(resolved_path, full_path, full_path_len+1);
+	path_sep = strrchr(resolved_path, sep);
+	if (path_sep)
+		path_sep++;
+	else
+		path_sep = resolved_path;
+	memcpy(path_sep, symname, symname_len+1);
+	if (sep == '\\')
+		convert_delimiter(path_sep, sep);
+
+	tcon = tlink_tcon(tlink);
+	oparms = CIFS_OPARMS(cifs_sb, tcon, resolved_path,
+			     FILE_READ_ATTRIBUTES, FILE_OPEN, 0, ACL_NO_MODE);
+	oparms.fid = &fid;
+
+	/* Try to open as a directory (NOT_FILE) */
+	oplock = 0;
+	oparms.create_options = cifs_create_options(cifs_sb,
+						    CREATE_NOT_FILE | OPEN_REPARSE_POINT);
+	open_rc = tcon->ses->server->ops->open(xid, &oparms, &oplock, NULL);
+	if (open_rc == 0) {
+		/* Successful open means that the target path is definitely a directory. */
+		*directory = true;
+		tcon->ses->server->ops->close(xid, tcon, &fid);
+	} else if (open_rc == -ENOTDIR) {
+		/* -ENOTDIR means that the target path is definitely a file. */
+		*directory = false;
+	} else if (open_rc == -ENOENT) {
+		/* -ENOENT means that the target path does not exist. */
+		cifs_dbg(FYI,
+			 "%s: symlink target path '%s' does not exist, "
+			 "creating '%s' as file symlink\n",
+			 __func__, symname, full_path);
+	} else {
+		/* Try to open as a file (NOT_DIR) */
+		oplock = 0;
+		oparms.create_options = cifs_create_options(cifs_sb,
+							    CREATE_NOT_DIR | OPEN_REPARSE_POINT);
+		open_rc = tcon->ses->server->ops->open(xid, &oparms, &oplock, NULL);
+		if (open_rc == 0) {
+			/* Successful open means that the target path is definitely a file. */
+			*directory = false;
+			tcon->ses->server->ops->close(xid, tcon, &fid);
+		} else if (open_rc == -EISDIR) {
+			/* -EISDIR means that the target path is definitely a directory. */
+			*directory = true;
+		} else {
+			/*
+			 * This code branch is called when we do not have a permission to
+			 * open the resolved_path or some other client/process denied
+			 * opening the resolved_path.
+			 *
+			 * TODO: Try to use ops->query_dir_first on the parent directory
+			 * of resolved_path, search for basename of resolved_path and
+			 * check if the ATTR_DIRECTORY is set in fi.Attributes. In some
+			 * case this could work also when opening of the path is denied.
+			 */
+			cifs_dbg(FYI,
+				 "%s: cannot determinate if the symlink target path '%s' "
+				 "is directory or not, creating '%s' as file symlink\n",
+				 __func__, symname, full_path);
+		}
+	}
+
+	kfree(resolved_path);
+	cifs_put_tlink(tlink);
+	return 0;
 }
 
 static int nfs_set_reparse_buf(struct reparse_posix_data *buf,
···
 	};
 
 	new = smb2_get_reparse_inode(&data, inode->i_sb, xid,
-				     tcon, full_path, &iov, NULL);
+				     tcon, full_path, false, &iov, NULL);
 	if (!IS_ERR(new))
 		d_instantiate(dentry, new);
 	else
···
 	data.wsl.eas_len = len;
 
 	new = smb2_get_reparse_inode(&data, inode->i_sb,
-				     xid, tcon, full_path,
+				     xid, tcon, full_path, false,
 				     &reparse_iov, &xattr_iov);
 	if (!IS_ERR(new))
 		d_instantiate(dentry, new);

fs/smb/client/smb2inode.c

···
 				     const unsigned int xid,
 				     struct cifs_tcon *tcon,
 				     const char *full_path,
+				     bool directory,
 				     struct kvec *reparse_iov,
 				     struct kvec *xattr_iov)
 {
···
 			     FILE_READ_ATTRIBUTES |
 			     FILE_WRITE_ATTRIBUTES,
 			     FILE_CREATE,
-			     CREATE_NOT_DIR | OPEN_REPARSE_POINT,
+			     (directory ? CREATE_NOT_FILE : CREATE_NOT_DIR) | OPEN_REPARSE_POINT,
 			     ACL_NO_MODE);
 	if (xattr_iov)
 		oparms.ea_cctx = xattr_iov;

fs/smb/client/smb2proto.h

···
 				     const unsigned int xid,
 				     struct cifs_tcon *tcon,
 				     const char *full_path,
+				     bool directory,
 				     struct kvec *reparse_iov,
 				     struct kvec *xattr_iov);
 int smb2_query_reparse_point(const unsigned int xid,

fs/smb/server/auth.c

···
 			    int in_len, char *out_blob, int *out_len)
 {
 	struct ksmbd_spnego_authen_response *resp;
+	struct ksmbd_login_response_ext *resp_ext = NULL;
 	struct ksmbd_user *user = NULL;
 	int retval;
 
···
 		goto out;
 	}
 
-	user = ksmbd_alloc_user(&resp->login_response);
+	if (resp->login_response.status & KSMBD_USER_FLAG_EXTENSION)
+		resp_ext = ksmbd_ipc_login_request_ext(resp->login_response.account);
+
+	user = ksmbd_alloc_user(&resp->login_response, resp_ext);
 	if (!user) {
 		ksmbd_debug(AUTH, "login failure\n");
 		retval = -ENOMEM;

fs/smb/server/ksmbd_netlink.h

···
  *  - KSMBD_EVENT_SPNEGO_AUTHEN_REQUEST/RESPONSE(ksmbd_spnego_authen_request/response)
  *    This event is to make kerberos authentication to be processed in
  *    userspace.
+ *
+ *  - KSMBD_EVENT_LOGIN_REQUEST_EXT/RESPONSE_EXT(ksmbd_login_request_ext/response_ext)
+ *    This event is to get user account extension info to user IPC daemon.
  */
 
 #define KSMBD_GENL_NAME		"SMBD_GENL"
···
 	__u16	hash_sz;			/* hash size */
 	__s8	hash[KSMBD_REQ_MAX_HASH_SZ];	/* password hash */
 	__u32	reserved[16];			/* Reserved room */
+};
+
+/*
+ * IPC user login response extension.
+ */
+struct ksmbd_login_response_ext {
+	__u32	handle;
+	__s32	ngroups;			/* supplementary group count */
+	__s8	reserved[128];			/* Reserved room */
+	__s8	____payload[];
 };
 
 /*
···
 	KSMBD_EVENT_SPNEGO_AUTHEN_REQUEST,
 	KSMBD_EVENT_SPNEGO_AUTHEN_RESPONSE	= 15,
 
+	KSMBD_EVENT_LOGIN_REQUEST_EXT,
+	KSMBD_EVENT_LOGIN_RESPONSE_EXT,
+
 	__KSMBD_EVENT_MAX,
 	KSMBD_EVENT_MAX = __KSMBD_EVENT_MAX - 1
 };
···
 #define KSMBD_USER_FLAG_BAD_USER	BIT(3)
 #define KSMBD_USER_FLAG_GUEST_ACCOUNT	BIT(4)
 #define KSMBD_USER_FLAG_DELAY_SESSION	BIT(5)
+#define KSMBD_USER_FLAG_EXTENSION	BIT(6)
 
 /*
  * Share config flags.

fs/smb/server/mgmt/user_config.c

···
 struct ksmbd_user *ksmbd_login_user(const char *account)
 {
 	struct ksmbd_login_response *resp;
+	struct ksmbd_login_response_ext *resp_ext = NULL;
 	struct ksmbd_user *user = NULL;
 
 	resp = ksmbd_ipc_login_request(account);
···
 	if (!(resp->status & KSMBD_USER_FLAG_OK))
 		goto out;
 
-	user = ksmbd_alloc_user(resp);
+	if (resp->status & KSMBD_USER_FLAG_EXTENSION)
+		resp_ext = ksmbd_ipc_login_request_ext(account);
+
+	user = ksmbd_alloc_user(resp, resp_ext);
 out:
 	kvfree(resp);
 	return user;
 }
 
-struct ksmbd_user *ksmbd_alloc_user(struct ksmbd_login_response *resp)
+struct ksmbd_user *ksmbd_alloc_user(struct ksmbd_login_response *resp,
+		struct ksmbd_login_response_ext *resp_ext)
 {
-	struct ksmbd_user *user = NULL;
+	struct ksmbd_user *user;
 
 	user = kmalloc(sizeof(struct ksmbd_user), GFP_KERNEL);
 	if (!user)
···
 	if (user->passkey)
 		memcpy(user->passkey, resp->hash, resp->hash_sz);
 
-	if (!user->name || !user->passkey) {
-		kfree(user->name);
-		kfree(user->passkey);
-		kfree(user);
-		user = NULL;
+	user->ngroups = 0;
+	user->sgid = NULL;
+
+	if (!user->name || !user->passkey)
+		goto err_free;
+
+	if (resp_ext) {
+		if (resp_ext->ngroups > NGROUPS_MAX) {
+			pr_err("ngroups(%u) from login response exceeds max groups(%d)\n",
+					resp_ext->ngroups, NGROUPS_MAX);
+			goto err_free;
+		}
+
+		user->sgid = kmemdup(resp_ext->____payload,
+				     resp_ext->ngroups * sizeof(gid_t),
+				     GFP_KERNEL);
+		if (!user->sgid)
+			goto err_free;
+
+		user->ngroups = resp_ext->ngroups;
+		ksmbd_debug(SMB, "supplementary groups : %d\n", user->ngroups);
 	}
+
 	return user;
+
+err_free:
+	kfree(user->name);
+	kfree(user->passkey);
+	kfree(user);
+	return NULL;
 }
 
 void ksmbd_free_user(struct ksmbd_user *user)
 {
 	ksmbd_ipc_logout_request(user->name, user->flags);
+	kfree(user->sgid);
 	kfree(user->name);
 	kfree(user->passkey);
 	kfree(user);

fs/smb/server/mgmt/user_config.h

···
 
 	size_t			passkey_sz;
 	char			*passkey;
+	int			ngroups;
+	gid_t			*sgid;
 };
 
 static inline bool user_guest(struct ksmbd_user *user)
···
 }
 
 struct ksmbd_user *ksmbd_login_user(const char *account);
-struct ksmbd_user *ksmbd_alloc_user(struct ksmbd_login_response *resp);
+struct ksmbd_user *ksmbd_alloc_user(struct ksmbd_login_response *resp,
+		struct ksmbd_login_response_ext *resp_ext);
 void ksmbd_free_user(struct ksmbd_user *user);
 int ksmbd_anonymous_user(struct ksmbd_user *user);
 bool ksmbd_compare_user(struct ksmbd_user *u1, struct ksmbd_user *u2);

fs/smb/server/mgmt/user_session.c

···
 
 	down_write(&conn->session_lock);
 	xa_for_each(&conn->sessions, id, sess) {
-		if (sess->state != SMB2_SESSION_VALID ||
-		    time_after(jiffies,
-			       sess->last_active + SMB2_SESSION_TIMEOUT)) {
+		if (atomic_read(&sess->refcnt) == 0 &&
+		    (sess->state != SMB2_SESSION_VALID ||
+		     time_after(jiffies,
+			       sess->last_active + SMB2_SESSION_TIMEOUT))) {
 			xa_erase(&conn->sessions, sess->id);
 			hash_del(&sess->hlist);
 			ksmbd_session_destroy(sess);
···
 
 	down_read(&sessions_table_lock);
 	sess = __session_lookup(id);
-	if (sess)
-		sess->last_active = jiffies;
 	up_read(&sessions_table_lock);
 
 	return sess;
···
 	if (sess && sess->state != SMB2_SESSION_VALID)
 		sess = NULL;
 	return sess;
+}
+
+void ksmbd_user_session_get(struct ksmbd_session *sess)
+{
+	atomic_inc(&sess->refcnt);
+}
+
+void ksmbd_user_session_put(struct ksmbd_session *sess)
+{
+	if (!sess)
+		return;
+
+	if (atomic_read(&sess->refcnt) <= 0)
+		WARN_ON(1);
+	else
+		atomic_dec(&sess->refcnt);
 }
 
 struct preauth_session *ksmbd_preauth_session_alloc(struct ksmbd_conn *conn,
···
 	xa_init(&sess->rpc_handle_list);
 	sess->sequence_number = 1;
 	rwlock_init(&sess->tree_conns_lock);
+	atomic_set(&sess->refcnt, 1);
 
 	ret = __init_smb2_session(sess);
 	if (ret)

fs/smb/server/mgmt/user_session.h

···
 	struct ksmbd_file_table		file_table;
 	unsigned long			last_active;
 	rwlock_t			tree_conns_lock;
+
+	atomic_t			refcnt;
 };
 
 static inline int test_session_flag(struct ksmbd_session *sess, int bit)
···
 int ksmbd_session_rpc_open(struct ksmbd_session *sess, char *rpc_name);
 void ksmbd_session_rpc_close(struct ksmbd_session *sess, int id);
 int ksmbd_session_rpc_method(struct ksmbd_session *sess, int id);
+void ksmbd_user_session_get(struct ksmbd_session *sess);
+void ksmbd_user_session_put(struct ksmbd_session *sess);
 #endif /* __USER_SESSION_MANAGEMENT_H__ */

fs/smb/server/server.c

···
 	} while (is_chained == true);
 
 send:
+	if (work->sess)
+		ksmbd_user_session_put(work->sess);
 	if (work->tcon)
 		ksmbd_tree_connect_put(work->tcon);
 	smb3_preauth_hash_rsp(work);

fs/smb/server/smb2pdu.c

···
 
 	/* Check for validity of user session */
 	work->sess = ksmbd_session_lookup_all(conn, sess_id);
-	if (work->sess)
+	if (work->sess) {
+		ksmbd_user_session_get(work->sess);
 		return 1;
+	}
 	ksmbd_debug(SMB, "Invalid user session, Uid %llu\n", sess_id);
 	return -ENOENT;
 }
···
 		}
 
 		conn->binding = true;
+		ksmbd_user_session_get(sess);
 	} else if ((conn->dialect < SMB30_PROT_ID ||
 		    server_conf.flags & KSMBD_GLOBAL_FLAG_SMB3_MULTICHANNEL) &&
 		   (req->Flags & SMB2_SESSION_REQ_FLAG_BINDING)) {
···
 		}
 
 		conn->binding = false;
+		ksmbd_user_session_get(sess);
 	}
 	work->sess = sess;
 
···
 	}
 
 	ksmbd_destroy_file_table(&sess->file_table);
+	down_write(&conn->session_lock);
 	sess->state = SMB2_SESSION_EXPIRED;
+	up_write(&conn->session_lock);
 
 	ksmbd_free_user(sess->user);
 	sess->user = NULL;

fs/smb/server/smb_common.c

···
 		struct ksmbd_share_config *share)
 {
 	struct ksmbd_session *sess = work->sess;
+	struct ksmbd_user *user = sess->user;
 	struct cred *cred;
 	struct group_info *gi;
 	unsigned int uid;
 	unsigned int gid;
+	int i;
 
-	uid = user_uid(sess->user);
-	gid = user_gid(sess->user);
+	uid = user_uid(user);
+	gid = user_gid(user);
 	if (share->force_uid != KSMBD_SHARE_INVALID_UID)
 		uid = share->force_uid;
 	if (share->force_gid != KSMBD_SHARE_INVALID_GID)
···
 	cred->fsuid = make_kuid(&init_user_ns, uid);
 	cred->fsgid = make_kgid(&init_user_ns, gid);
 
-	gi = groups_alloc(0);
+	gi = groups_alloc(user->ngroups);
 	if (!gi) {
 		abort_creds(cred);
 		return -ENOMEM;
 	}
+
+	for (i = 0; i < user->ngroups; i++)
+		gi->gid[i] = make_kgid(&init_user_ns, user->sgid[i]);
+
+	if (user->ngroups)
+		groups_sort(gi);
+
 	set_groups(cred, gi);
 	put_group_info(gi);
 

fs/smb/server/transport_ipc.c

···
 	},
 	[KSMBD_EVENT_SPNEGO_AUTHEN_RESPONSE] = {
 	},
+	[KSMBD_EVENT_LOGIN_REQUEST_EXT] = {
+		.len = sizeof(struct ksmbd_login_request),
+	},
+	[KSMBD_EVENT_LOGIN_RESPONSE_EXT] = {
+		.len = sizeof(struct ksmbd_login_response_ext),
+	},
 };
 
 static struct genl_ops ksmbd_genl_ops[] = {
···
 		.cmd	= KSMBD_EVENT_SPNEGO_AUTHEN_RESPONSE,
 		.doit	= handle_generic_event,
 	},
+	{
+		.cmd	= KSMBD_EVENT_LOGIN_REQUEST_EXT,
+		.doit	= handle_unsupported_event,
+	},
+	{
+		.cmd	= KSMBD_EVENT_LOGIN_RESPONSE_EXT,
+		.doit	= handle_generic_event,
+	},
 };
 
 static struct genl_family ksmbd_genl_family = {
···
 	.module		= THIS_MODULE,
 	.ops		= ksmbd_genl_ops,
 	.n_ops		= ARRAY_SIZE(ksmbd_genl_ops),
-	.resv_start_op	= KSMBD_EVENT_SPNEGO_AUTHEN_RESPONSE + 1,
+	.resv_start_op	= KSMBD_EVENT_LOGIN_RESPONSE_EXT + 1,
 };
 
 static void ksmbd_nl_init_fixup(void)
···
 {
 	unsigned int msg_sz = entry->msg_sz;
 
-	if (entry->type == KSMBD_EVENT_RPC_REQUEST) {
+	switch (entry->type) {
+	case KSMBD_EVENT_RPC_REQUEST:
+	{
 		struct ksmbd_rpc_command *resp = entry->response;
 
 		msg_sz = sizeof(struct ksmbd_rpc_command) + resp->payload_sz;
-	} else if (entry->type == KSMBD_EVENT_SPNEGO_AUTHEN_REQUEST) {
+		break;
+	}
+	case KSMBD_EVENT_SPNEGO_AUTHEN_REQUEST:
+	{
 		struct ksmbd_spnego_authen_response *resp = entry->response;
 
 		msg_sz = sizeof(struct ksmbd_spnego_authen_response) +
 				resp->session_key_len + resp->spnego_blob_len;
-	} else if (entry->type == KSMBD_EVENT_SHARE_CONFIG_REQUEST) {
+		break;
+	}
+	case KSMBD_EVENT_SHARE_CONFIG_REQUEST:
+	{
 		struct ksmbd_share_config_response *resp = entry->response;
 
 		if (resp->payload_sz) {
···
 			msg_sz = sizeof(struct ksmbd_share_config_response) +
 					resp->payload_sz;
 		}
+		break;
+	}
+	case KSMBD_EVENT_LOGIN_REQUEST_EXT:
+	{
+		struct ksmbd_login_response_ext *resp = entry->response;
+
+		if (resp->ngroups) {
+			msg_sz = sizeof(struct ksmbd_login_response_ext) +
+					resp->ngroups * sizeof(gid_t);
+		}
+	}
 	}
 
 	return entry->msg_sz != msg_sz ? -EINVAL : 0;
···
 	req->handle = ksmbd_acquire_id(&ipc_ida);
 	strscpy(req->account, account, KSMBD_REQ_MAX_ACCOUNT_NAME_SZ);
 
+	resp = ipc_msg_send_request(msg, req->handle);
+	ipc_msg_handle_free(req->handle);
+	ipc_msg_free(msg);
+	return resp;
+}
+
+struct ksmbd_login_response_ext *ksmbd_ipc_login_request_ext(const char *account)
+{
+	struct ksmbd_ipc_msg *msg;
+	struct ksmbd_login_request *req;
+	struct ksmbd_login_response_ext *resp;
+
+	if (strlen(account) >= KSMBD_REQ_MAX_ACCOUNT_NAME_SZ)
+		return NULL;
+
+	msg = ipc_msg_alloc(sizeof(struct ksmbd_login_request));
+	if (!msg)
+		return NULL;
+
+	msg->type = KSMBD_EVENT_LOGIN_REQUEST_EXT;
+	req = (struct ksmbd_login_request *)msg->payload;
+	req->handle = ksmbd_acquire_id(&ipc_ida);
+	strscpy(req->account, account, KSMBD_REQ_MAX_ACCOUNT_NAME_SZ);
 	resp = ipc_msg_send_request(msg, req->handle);
 	ipc_msg_handle_free(req->handle);
 	ipc_msg_free(msg);

fs/smb/server/transport_ipc.h

···
 
 struct ksmbd_login_response *
 ksmbd_ipc_login_request(const char *account);
+struct ksmbd_login_response_ext *
+ksmbd_ipc_login_request_ext(const char *account);
 
 struct ksmbd_session;
 struct ksmbd_share_config;

include/linux/closure.h

···
 		__closure_wait_event(waitlist, _cond);			\
 } while (0)
 
+#define __closure_wait_event_timeout(waitlist, _cond, _until)		\
+({									\
+	struct closure cl;						\
+	long _t;							\
+									\
+	closure_init_stack(&cl);					\
+									\
+	while (1) {							\
+		closure_wait(waitlist, &cl);				\
+		if (_cond) {						\
+			_t = max_t(long, 1L, _until - jiffies);		\
+			break;						\
+		}							\
+		_t = max_t(long, 0L, _until - jiffies);			\
+		if (!_t)						\
+			break;						\
+		closure_sync_timeout(&cl, _t);				\
+	}								\
+	closure_wake_up(waitlist);					\
+	closure_sync(&cl);						\
+	_t;								\
+})
+
+/*
+ * Returns 0 if timeout expired, remaining time in jiffies (at least 1) if
+ * condition became true
+ */
+#define closure_wait_event_timeout(waitlist, _cond, _timeout)		\
+({									\
+	unsigned long _until = jiffies + _timeout;			\
+	(_cond)								\
+		? max_t(long, 1L, _until - jiffies)			\
+		: __closure_wait_event_timeout(waitlist, _cond, _until);\
+})
+
 #endif /* _LINUX_CLOSURE_H */

include/linux/fsl/enetc_mdio.h

···
 static inline int enetc_mdio_write_c45(struct mii_bus *bus, int phy_id,
 				       int devad, int regnum, u16 value)
 { return -EINVAL; }
-struct enetc_hw *enetc_hw_alloc(struct device *dev, void __iomem *port_regs)
+static inline struct enetc_hw *enetc_hw_alloc(struct device *dev,
+					      void __iomem *port_regs)
 { return ERR_PTR(-EINVAL); }
 
 #endif

include/linux/nfs_fs_sb.h

···
 	struct list_head	layouts;
 	struct list_head	delegations;
 	struct list_head	ss_copies;
+	struct list_head	ss_src_copies;
 
 	unsigned long		delegation_gen;
 	unsigned long		mig_gen;

include/linux/nfslocalio.h

···
 		   struct rpc_clnt *, const struct cred *,
 		   const struct nfs_fh *, const fmode_t);
 
+static inline void nfs_to_nfsd_file_put_local(struct nfsd_file *localio)
+{
+	/*
+	 * Once reference to nfsd_serv is dropped, NFSD could be
+	 * unloaded, so ensure safe return from nfsd_file_put_local()
+	 * by always taking RCU.
+	 */
+	rcu_read_lock();
+	nfs_to->nfsd_file_put_local(localio);
+	rcu_read_unlock();
+}
+
 #else   /* CONFIG_NFS_LOCALIO */
 static inline void nfsd_localio_ops_init(void)
+{
+}
+static inline void nfs_to_nfsd_file_put_local(struct nfsd_file *localio)
 {
 }
 #endif  /* CONFIG_NFS_LOCALIO */

include/net/genetlink.h

···
  * @skb: netlink message as socket buffer
  * @portid: own netlink portid to avoid sending to yourself
  * @group: offset of multicast group in groups array
- * @flags: allocation flags
  *
  * This function must hold the RTNL or rcu_read_lock().
  */
 int genlmsg_multicast_allns(const struct genl_family *family,
 			    struct sk_buff *skb, u32 portid,
-			    unsigned int group, gfp_t flags);
+			    unsigned int group);
 
 /**
  * genlmsg_unicast - unicast a netlink message

io_uring/rw.c

···
 	rwf_t				flags;
 };
 
-static inline bool io_file_supports_nowait(struct io_kiocb *req)
+static bool io_file_supports_nowait(struct io_kiocb *req, __poll_t mask)
 {
-	return req->flags & REQ_F_SUPPORT_NOWAIT;
+	/* If FMODE_NOWAIT is set for a file, we're golden */
+	if (req->flags & REQ_F_SUPPORT_NOWAIT)
+		return true;
+	/* No FMODE_NOWAIT, if we can poll, check the status */
+	if (io_file_can_poll(req)) {
+		struct poll_table_struct pt = { ._key = mask };
+
+		return vfs_poll(req->file, &pt) & mask;
+	}
+	/* No FMODE_NOWAIT support, and file isn't pollable. Tough luck. */
+	return false;
 }
 
 #ifdef CONFIG_COMPAT
···
 	 * supports async. Otherwise it's impossible to use O_NONBLOCK files
 	 * reliably. If not, or it IOCB_NOWAIT is set, don't retry.
 	 */
-	if ((kiocb->ki_flags & IOCB_NOWAIT) ||
-	    ((file->f_flags & O_NONBLOCK) && !io_file_supports_nowait(req)))
+	if (kiocb->ki_flags & IOCB_NOWAIT ||
+	    ((file->f_flags & O_NONBLOCK && (req->flags & REQ_F_SUPPORT_NOWAIT))))
 		req->flags |= REQ_F_NOWAIT;
 
 	if (ctx->flags & IORING_SETUP_IOPOLL) {
···
 
 	if (force_nonblock) {
 		/* If the file doesn't support async, just async punt */
-		if (unlikely(!io_file_supports_nowait(req)))
+		if (unlikely(!io_file_supports_nowait(req, EPOLLIN)))
 			return -EAGAIN;
 		kiocb->ki_flags |= IOCB_NOWAIT;
 	} else {
···
 	ret = __io_read(req, issue_flags);
 
 	/*
-	 * If the file doesn't support proper NOWAIT, then disable multishot
-	 * and stay in single shot mode.
-	 */
-	if (!io_file_supports_nowait(req))
-		req->flags &= ~REQ_F_APOLL_MULTISHOT;
-
-	/*
 	 * If we get -EAGAIN, recycle our buffer and just let normal poll
 	 * handling arm it.
 	 */
···
 		if (issue_flags & IO_URING_F_MULTISHOT)
 			return IOU_ISSUE_SKIP_COMPLETE;
 		return -EAGAIN;
-	}
-
-	/*
-	 * Any successful return value will keep the multishot read armed.
-	 */
-	if (ret > 0 && req->flags & REQ_F_APOLL_MULTISHOT) {
+	} else if (ret <= 0) {
+		io_kbuf_recycle(req, issue_flags);
+		if (ret < 0)
+			req_set_fail(req);
+	} else {
 		/*
-		 * Put our buffer and post a CQE. If we fail to post a CQE, then
+		 * Any successful return value will keep the multishot read
+		 * armed, if it's still set. Put our buffer and post a CQE. If
+		 * we fail to post a CQE, or multishot is no longer set, then
 		 * jump to the termination path. This request is then done.
 		 */
 		cflags = io_put_kbuf(req, ret, issue_flags);
···
 
 	if (force_nonblock) {
 		/* If the file doesn't support async, just async punt */
-		if (unlikely(!io_file_supports_nowait(req)))
+		if (unlikely(!io_file_supports_nowait(req, EPOLLOUT)))
 			goto ret_eagain;
 
 		/* Check if we can support NOWAIT. */

kernel/rcu/tree_nocb.h

···
 			rcu_nocb_unlock(rdp);
 			wake_nocb_gp_defer(rdp, RCU_NOCB_WAKE_LAZY,
 					   TPS("WakeLazy"));
-		} else if (!irqs_disabled_flags(flags)) {
+		} else if (!irqs_disabled_flags(flags) && cpu_online(rdp->cpu)) {
 			/* ... if queue was empty ... */
 			rcu_nocb_unlock(rdp);
 			wake_nocb_gp(rdp, false);
 			trace_rcu_nocb_wake(rcu_state.name, rdp->cpu,
 					    TPS("WakeEmpty"));
 		} else {
+			/*
+			 * Don't do the wake-up upfront on fragile paths.
+			 * Also offline CPUs can't call swake_up_one_online() from
+			 * (soft-)IRQs. Rely on the final deferred wake-up from
+			 * rcutree_report_cpu_dead()
+			 */
 			rcu_nocb_unlock(rdp);
 			wake_nocb_gp_defer(rdp, RCU_NOCB_WAKE,
 					   TPS("WakeEmptyIsDeferred"));

kernel/sched/ext.c

···
 #define SCX_OP_IDX(op)		(offsetof(struct sched_ext_ops, op) / sizeof(void (*)(void)))
 
 enum scx_consts {
-	SCX_SLICE_BYPASS		= SCX_SLICE_DFL / 4,
 	SCX_DSP_DFL_MAX_BATCH		= 32,
 	SCX_DSP_MAX_LOOPS		= 32,
 	SCX_WATCHDOG_MAX_TIMEOUT	= 30 * HZ,
···
 	SCX_EXIT_DUMP_DFL_LEN		= 32768,
 
 	SCX_CPUPERF_ONE			= SCHED_CAPACITY_SCALE,
+
+	/*
+	 * Iterating all tasks may take a while. Periodically drop
+	 * scx_tasks_lock to avoid causing e.g. CSD and RCU stalls.
+	 */
+	SCX_OPS_TASK_ITER_BATCH		= 32,
 };
 
 enum scx_exit_kind {
···
 	struct task_struct		*locked;
 	struct rq			*rq;
 	struct rq_flags			rf;
+	u32				cnt;
 };
 
 /**
- * scx_task_iter_init - Initialize a task iterator
+ * scx_task_iter_start - Lock scx_tasks_lock and start a task iteration
  * @iter: iterator to init
  *
- * Initialize @iter. Must be called with scx_tasks_lock held. Once initialized,
- * @iter must eventually be exited with scx_task_iter_exit().
+ * Initialize @iter and return with scx_tasks_lock held. Once initialized, @iter
+ * must eventually be stopped with scx_task_iter_stop().
  *
- * scx_tasks_lock may be released between this and the first next() call or
- * between any two next() calls. If scx_tasks_lock is released between two
- * next() calls, the caller is responsible for ensuring that the task being
- * iterated remains accessible either through RCU read lock or obtaining a
- * reference count.
+ * scx_tasks_lock and the rq lock may be released using scx_task_iter_unlock()
+ * between this and the first next() call or between any two next() calls. If
+ * the locks are released between two next() calls, the caller is responsible
+ * for ensuring that the task being iterated remains accessible either through
+ * RCU read lock or obtaining a reference count.
  *
  * All tasks which existed when the iteration started are guaranteed to be
  * visited as long as they still exist.
  */
-static void scx_task_iter_init(struct scx_task_iter *iter)
+static void scx_task_iter_start(struct scx_task_iter *iter)
 {
-	lockdep_assert_held(&scx_tasks_lock);
-
 	BUILD_BUG_ON(__SCX_DSQ_ITER_ALL_FLAGS &
 		     ((1U << __SCX_DSQ_LNODE_PRIV_SHIFT) - 1));
+
+	spin_lock_irq(&scx_tasks_lock);
 
 	iter->cursor = (struct sched_ext_entity){ .flags = SCX_TASK_CURSOR };
 	list_add(&iter->cursor.tasks_node, &scx_tasks);
 	iter->locked = NULL;
+	iter->cnt = 0;
 }
 
-/**
- * scx_task_iter_rq_unlock - Unlock rq locked by a task iterator
- * @iter: iterator to unlock rq for
- *
- * If @iter is in the middle of a locked iteration, it may be locking the rq of
- * the task currently being visited. Unlock the rq if so. This function can be
- * safely called anytime during an iteration.
- *
- * Returns %true if the rq @iter was locking is unlocked. %false if @iter was
- * not locking an rq.
- */
-static bool scx_task_iter_rq_unlock(struct scx_task_iter *iter)
+static void __scx_task_iter_rq_unlock(struct scx_task_iter *iter)
 {
 	if (iter->locked) {
 		task_rq_unlock(iter->rq, iter->locked, &iter->rf);
 		iter->locked = NULL;
-		return true;
-	} else {
-		return false;
 	}
 }
 
 /**
- * scx_task_iter_exit - Exit a task iterator
+ * scx_task_iter_unlock - Unlock rq and scx_tasks_lock held by a task iterator
+ * @iter: iterator to unlock
+ *
+ * If @iter is in the middle of a locked iteration, it may be locking the rq of
+ * the task currently being visited in addition to scx_tasks_lock. Unlock both.
+ * This function can be safely called anytime during an iteration.
+ */
+static void scx_task_iter_unlock(struct scx_task_iter *iter)
+{
+	__scx_task_iter_rq_unlock(iter);
+	spin_unlock_irq(&scx_tasks_lock);
+}
+
+/**
+ * scx_task_iter_relock - Lock scx_tasks_lock released by scx_task_iter_unlock()
+ * @iter: iterator to re-lock
+ *
+ * Re-lock scx_tasks_lock unlocked by scx_task_iter_unlock(). Note that it
+ * doesn't re-lock the rq lock. Must be called before other iterator operations.
+ */
+static void scx_task_iter_relock(struct scx_task_iter *iter)
+{
+	spin_lock_irq(&scx_tasks_lock);
+}
+
+/**
+ * scx_task_iter_stop - Stop a task iteration and unlock scx_tasks_lock
  * @iter: iterator to exit
  *
- * Exit a previously initialized @iter. Must be called with scx_tasks_lock held.
- * If the iterator holds a task's rq lock, that rq lock is released. See
- * scx_task_iter_init() for details.
+ * Exit a previously initialized @iter. Must be called with scx_tasks_lock held
+ * which is released on return. If the iterator holds a task's rq lock, that rq
+ * lock is also released. See scx_task_iter_start() for details.
  */
-static void scx_task_iter_exit(struct scx_task_iter *iter)
+static void scx_task_iter_stop(struct scx_task_iter *iter)
 {
-	lockdep_assert_held(&scx_tasks_lock);
-
-	scx_task_iter_rq_unlock(iter);
 	list_del_init(&iter->cursor.tasks_node);
+	scx_task_iter_unlock(iter);
 }
 
 /**
  * scx_task_iter_next - Next task
  * @iter: iterator to walk
  *
- * Visit the next task. See scx_task_iter_init() for details.
+ * Visit the next task. See scx_task_iter_start() for details. Locks are dropped
+ * and re-acquired every %SCX_OPS_TASK_ITER_BATCH iterations to avoid causing
+ * stalls by holding scx_tasks_lock for too long.
  */
 static struct task_struct *scx_task_iter_next(struct scx_task_iter *iter)
 {
 	struct list_head *cursor = &iter->cursor.tasks_node;
 	struct sched_ext_entity *pos;
 
-	lockdep_assert_held(&scx_tasks_lock);
+	if (!(++iter->cnt % SCX_OPS_TASK_ITER_BATCH)) {
+		scx_task_iter_unlock(iter);
+		cond_resched();
+		scx_task_iter_relock(iter);
+	}
 
 	list_for_each_entry(pos, cursor, tasks_node) {
 		if (&pos->tasks_node == &scx_tasks)
···
  * @include_dead: Whether we should include dead tasks in the iteration
  *
  * Visit the non-idle task with its rq lock held. Allows callers to specify
- * whether they would like to filter out dead tasks. See scx_task_iter_init()
+ * whether they would like to filter out dead tasks. See scx_task_iter_start()
  * for details.
  */
 static struct task_struct *scx_task_iter_next_locked(struct scx_task_iter *iter)
 {
 	struct task_struct *p;
 
-	scx_task_iter_rq_unlock(iter);
+	__scx_task_iter_rq_unlock(iter);
 
 	while ((p = scx_task_iter_next(iter))) {
 		/*
···
 static void do_enqueue_task(struct rq *rq, struct task_struct *p, u64 enq_flags,
 			    int sticky_cpu)
 {
-	bool bypassing = scx_rq_bypassing(rq);
 	struct task_struct **ddsp_taskp;
 	unsigned long qseq;
 
···
 	if (!scx_rq_online(rq))
 		goto local;
 
-	if (bypassing)
+	if (scx_rq_bypassing(rq))
 		goto global;
 
 	if (p->scx.ddsp_dsq_id != SCX_DSQ_INVALID)
···
 
 global:
 	touch_core_sched(rq, p);	/* see the comment in local: */
-	p->scx.slice = bypassing ? SCX_SLICE_BYPASS : SCX_SLICE_DFL;
+	p->scx.slice = SCX_SLICE_DFL;
 	dispatch_enqueue(find_global_dsq(p), p, enq_flags);
 }
 
···
 
 		if (unlikely(!p->scx.slice)) {
 			if (!scx_rq_bypassing(rq) && !scx_warned_zero_slice) {
-				printk_deferred(KERN_WARNING "sched_ext: %s[%d] has zero slice in pick_next_task_scx()\n",
-						p->comm, p->pid);
+				printk_deferred(KERN_WARNING "sched_ext: %s[%d] has zero slice in %s()\n",
+						p->comm, p->pid, __func__);
 				scx_warned_zero_slice = true;
 			}
 			p->scx.slice = SCX_SLICE_DFL;
···
 
 	*found = false;
 
-	if (!static_branch_likely(&scx_builtin_idle_enabled)) {
-		scx_ops_error("built-in idle tracking is disabled");
-		return prev_cpu;
-	}
-
 	/*
 	 * If WAKE_SYNC, the waker's local DSQ is empty, and the system is
 	 * under utilized, wake up @p to the local DSQ of the waker. Checking
···
 	if (unlikely(wake_flags & WF_EXEC))
 		return prev_cpu;
 
-	if (SCX_HAS_OP(select_cpu)) {
+	if (SCX_HAS_OP(select_cpu) && !scx_rq_bypassing(task_rq(p))) {
 		s32 cpu;
 		struct task_struct **ddsp_taskp;
 
···
 {
 	int cpu = cpu_of(rq);
 
-	if (SCX_HAS_OP(update_idle)) {
+	if (SCX_HAS_OP(update_idle) && !scx_rq_bypassing(rq)) {
 		SCX_CALL_OP(SCX_KF_REST, update_idle, cpu_of(rq), idle);
 		if (!static_branch_unlikely(&scx_builtin_idle_enabled))
 			return;
···
  * the DISABLING state and then cycling the queued tasks through dequeue/enqueue
  * to force global FIFO scheduling.
  *
- * a. ops.enqueue() is ignored and tasks are queued in simple global FIFO order.
- *    %SCX_OPS_ENQ_LAST is also ignored.
+ * - ops.select_cpu() is ignored and the default select_cpu() is used.
  *
- * b. ops.dispatch() is ignored.
+ * - ops.enqueue() is ignored and tasks are queued in simple global FIFO order.
+ *   %SCX_OPS_ENQ_LAST is also ignored.
  *
- * c. balance_scx() does not set %SCX_RQ_BAL_KEEP on non-zero slice as slice
- *    can't be trusted. Whenever a tick triggers, the running task is rotated to
- *    the tail of the queue with core_sched_at touched.
+ * - ops.dispatch() is ignored.
  *
- * d. pick_next_task() suppresses zero slice warning.
+ * - balance_scx() does not set %SCX_RQ_BAL_KEEP on non-zero slice as slice
+ *   can't be trusted. Whenever a tick triggers, the running task is rotated to
+ *   the tail of the queue with core_sched_at touched.
  *
- * e. scx_bpf_kick_cpu() is disabled to avoid irq_work malfunction during PM
- *    operations.
+ * - pick_next_task() suppresses zero slice warning.
  *
- * f. scx_prio_less() reverts to the default core_sched_at order.
+ * - scx_bpf_kick_cpu() is disabled to avoid irq_work malfunction during PM
+ *   operations.
+ *
+ * - scx_prio_less() reverts to the default core_sched_at order.
  */
 static void scx_ops_bypass(bool bypass)
 {
···
 
 		rq_unlock_irqrestore(rq, &rf);
 
-		/* kick to restore ticks */
+		/* resched to restore ticks and idle state */
 		resched_cpu(cpu);
 	}
 }
···
 
 	scx_ops_init_task_enabled = false;
 
-	spin_lock_irq(&scx_tasks_lock);
-	scx_task_iter_init(&sti);
+	scx_task_iter_start(&sti);
 	while ((p = scx_task_iter_next_locked(&sti))) {
 		const struct sched_class *old_class = p->sched_class;
 		struct sched_enq_and_set_ctx ctx;
 
 		sched_deq_and_put_task(p, DEQUEUE_SAVE | DEQUEUE_MOVE, &ctx);
 
-		p->scx.slice = min_t(u64, p->scx.slice, SCX_SLICE_DFL);
 		__setscheduler_prio(p, p->prio);
 		check_class_changing(task_rq(p), p, old_class);
 
···
 		check_class_changed(task_rq(p), p, old_class, p->prio);
 		scx_ops_exit_task(p);
 	}
-	scx_task_iter_exit(&sti);
-	spin_unlock_irq(&scx_tasks_lock);
+	scx_task_iter_stop(&sti);
 	percpu_up_write(&scx_fork_rwsem);
 
 	/* no task is on scx, turn off all the switches and flush in-progress calls */
···
 	if (ret)
 		goto err_disable_unlock_all;
 
-	spin_lock_irq(&scx_tasks_lock);
-	scx_task_iter_init(&sti);
+	scx_task_iter_start(&sti);
 	while ((p = scx_task_iter_next_locked(&sti))) {
 		/*
 		 * @p may already be dead, have lost all its usages counts and
···
 		if (!tryget_task_struct(p))
 			continue;
 
-		scx_task_iter_rq_unlock(&sti);
-		spin_unlock_irq(&scx_tasks_lock);
+		scx_task_iter_unlock(&sti);
 
 		ret = scx_ops_init_task(p, task_group(p), false);
 		if (ret) {
 			put_task_struct(p);
-			spin_lock_irq(&scx_tasks_lock);
-			scx_task_iter_exit(&sti);
-			spin_unlock_irq(&scx_tasks_lock);
+			scx_task_iter_relock(&sti);
+			scx_task_iter_stop(&sti);
 			scx_ops_error("ops.init_task() failed (%d) for %s[%d]",
 				      ret, p->comm, p->pid);
 			goto err_disable_unlock_all;
···
 		scx_set_task_state(p, SCX_TASK_READY);
 
 		put_task_struct(p);
-		spin_lock_irq(&scx_tasks_lock);
+		scx_task_iter_relock(&sti);
 	}
-	scx_task_iter_exit(&sti);
-	spin_unlock_irq(&scx_tasks_lock);
+	scx_task_iter_stop(&sti);
 	scx_cgroup_unlock();
 	percpu_up_write(&scx_fork_rwsem);
 
···
 	 * scx_tasks_lock.
 	 */
 	percpu_down_write(&scx_fork_rwsem);
-	spin_lock_irq(&scx_tasks_lock);
-	scx_task_iter_init(&sti);
+	scx_task_iter_start(&sti);
 	while ((p = scx_task_iter_next_locked(&sti))) {
 		const struct sched_class *old_class = p->sched_class;
 		struct sched_enq_and_set_ctx ctx;
 
 		sched_deq_and_put_task(p, DEQUEUE_SAVE | DEQUEUE_MOVE, &ctx);
 
+		p->scx.slice = SCX_SLICE_DFL;
 		__setscheduler_prio(p, p->prio);
 		check_class_changing(task_rq(p), p, old_class);
 
···
 
 		check_class_changed(task_rq(p), p, old_class, p->prio);
 	}
-	scx_task_iter_exit(&sti);
-	spin_unlock_irq(&scx_tasks_lock);
+	scx_task_iter_stop(&sti);
 	percpu_up_write(&scx_fork_rwsem);
 
 	scx_ops_bypass(false);
···
 __bpf_kfunc s32 scx_bpf_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
 				       u64 wake_flags, bool *is_idle)
 {
-	if (!scx_kf_allowed(SCX_KF_SELECT_CPU)) {
-		*is_idle = false;
-		return prev_cpu;
+	if (!static_branch_likely(&scx_builtin_idle_enabled)) {
+		scx_ops_error("built-in idle tracking is disabled");
+		goto prev_cpu;
 	}
+
+	if (!scx_kf_allowed(SCX_KF_SELECT_CPU))
+		goto prev_cpu;
+
 #ifdef CONFIG_SMP
 	return scx_select_cpu_dfl(p, prev_cpu, wake_flags, is_idle);
-#else
+#endif
+
+prev_cpu:
 	*is_idle = false;
 	return prev_cpu;
-#endif
 }
 
 __bpf_kfunc_end_defs();

kernel/time/posix-clock.c

···
 		goto out;
 	}
 
+	if (!timespec64_valid_strict(ts))
+		return -EINVAL;
+
 	if (cd.clk->ops.clock_settime)
 		err = cd.clk->ops.clock_settime(cd.clk, ts);
 	else

kernel/trace/ring_buffer.c

···
 	}
 
 	for_each_buffer_cpu(buffer, cpu) {
+		struct buffer_data_page *old_free_data_page;
+		struct list_head old_pages;
+		unsigned long flags;
 
 		if (!cpumask_test_cpu(cpu, buffer->cpumask))
 			continue;
 
 		cpu_buffer = buffer->buffers[cpu];
 
+		raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags);
+
 		/* Clear the head bit to make the link list normal to read */
 		rb_head_page_deactivate(cpu_buffer);
 
-		/* Now walk the list and free all the old sub buffers */
-		list_for_each_entry_safe(bpage, tmp, cpu_buffer->pages, list) {
-			list_del_init(&bpage->list);
-			free_buffer_page(bpage);
-		}
-		/* The above loop stopped an the last page needing to be freed */
-		bpage = list_entry(cpu_buffer->pages, struct buffer_page, list);
-		free_buffer_page(bpage);
-
-		/* Free the current reader page */
-		free_buffer_page(cpu_buffer->reader_page);
+		/*
+		 * Collect buffers from the cpu_buffer pages list and the
+		 * reader_page on old_pages, so they can be freed later when not
+		 * under a spinlock. The pages list is a linked list with no
+		 * head, adding old_pages turns it into a regular list with
+		 * old_pages being the head.
+		 */
+		list_add(&old_pages, cpu_buffer->pages);
+		list_add(&cpu_buffer->reader_page->list, &old_pages);
 
 		/* One page was allocated for the reader page */
 		cpu_buffer->reader_page = list_entry(cpu_buffer->new_pages.next,
 						     struct buffer_page, list);
 		list_del_init(&cpu_buffer->reader_page->list);
 
-		/* The cpu_buffer pages are a link list with no head */
+		/* Install the new pages, remove the head from the list */
 		cpu_buffer->pages = cpu_buffer->new_pages.next;
-		cpu_buffer->new_pages.next->prev = cpu_buffer->new_pages.prev;
-		cpu_buffer->new_pages.prev->next = cpu_buffer->new_pages.next;
-
-		/* Clear the new_pages list */
-		INIT_LIST_HEAD(&cpu_buffer->new_pages);
+		list_del_init(&cpu_buffer->new_pages);
 
 		cpu_buffer->head_page
 			= list_entry(cpu_buffer->pages, struct buffer_page, list);
···
 		cpu_buffer->nr_pages = cpu_buffer->nr_pages_to_update;
 		cpu_buffer->nr_pages_to_update = 0;
 
-		free_pages((unsigned long)cpu_buffer->free_page, old_order);
+		old_free_data_page = cpu_buffer->free_page;
 		cpu_buffer->free_page = NULL;
 
 		rb_head_page_activate(cpu_buffer);
+
+		raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags);
+
+		/* Free old sub buffers */
+		list_for_each_entry_safe(bpage, tmp, &old_pages, list) {
+			list_del_init(&bpage->list);
+			free_buffer_page(bpage);
+		}
+		free_pages((unsigned long)old_free_data_page, old_order);
 
 		rb_check_pages(cpu_buffer);
 	}

kernel/trace/trace.c

···
 		 * cannot be deleted by user space, so keep the reference
 		 * to it.
 		 */
-		if (start)
+		if (start) {
 			tr->flags |= TRACE_ARRAY_FL_BOOT;
-		else
-			trace_array_put(tr);
+			tr->ref++;
+		}
 
 		while ((tok = strsep(&curr_str, ","))) {
 			early_enable_events(tr, tok, true);

net/9p/Kconfig

···
 config NET_9P_USBG
 	bool "9P USB Gadget Transport"
 	depends on USB_GADGET=y || USB_GADGET=NET_9P
+	select CONFIGFS_FS
+	select USB_LIBCOMPOSITE
 	help
 	  This builds support for a transport for 9pfs over
 	  usb gadget.

net/ipv4/devinet.c

···
 	/* Account for reference dev->ip_ptr (below) */
 	refcount_set(&in_dev->refcnt, 1);
 
-	err = devinet_sysctl_register(in_dev);
-	if (err) {
-		in_dev->dead = 1;
-		neigh_parms_release(&arp_tbl, in_dev->arp_parms);
-		in_dev_put(in_dev);
-		in_dev = NULL;
-		goto out;
+	if (dev != blackhole_netdev) {
+		err = devinet_sysctl_register(in_dev);
+		if (err) {
+			in_dev->dead = 1;
+			neigh_parms_release(&arp_tbl, in_dev->arp_parms);
+			in_dev_put(in_dev);
+			in_dev = NULL;
+			goto out;
+		}
+		ip_mc_init_dev(in_dev);
+		if (dev->flags & IFF_UP)
+			ip_mc_up(in_dev);
 	}
-	ip_mc_init_dev(in_dev);
-	if (dev->flags & IFF_UP)
-		ip_mc_up(in_dev);
 
 	/* we can receive as soon as ip_ptr is set -- do this last */
 	rcu_assign_pointer(dev->ip_ptr, in_dev);
···
 
 	in_dev_put(in_dev);
 }
+
+static int __init inet_blackhole_dev_init(void)
+{
+	int err = 0;
+
+	rtnl_lock();
+	if (!inetdev_init(blackhole_netdev))
+		err = -ENOMEM;
+	rtnl_unlock();
+
+	return err;
+}
+late_initcall(inet_blackhole_dev_init);
 
 int inet_addr_onlink(struct in_device *in_dev, __be32 a, __be32 b)
 {

net/ipv4/inet_connection_sock.c

···
 		found = __sk_nulls_del_node_init_rcu(sk);
 		spin_unlock(lock);
 	}
-	if (timer_pending(&req->rsk_timer) && del_timer_sync(&req->rsk_timer))
-		reqsk_put(req);
+
 	return found;
 }
 
-bool inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req)
+static bool __inet_csk_reqsk_queue_drop(struct sock *sk,
+					struct request_sock *req,
+					bool from_timer)
 {
 	bool unlinked = reqsk_queue_unlink(req);
+
+	if (!from_timer && timer_delete_sync(&req->rsk_timer))
+		reqsk_put(req);
 
 	if (unlinked) {
 		reqsk_queue_removed(&inet_csk(sk)->icsk_accept_queue, req);
 		reqsk_put(req);
 	}
+
 	return unlinked;
+}
+
+bool inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req)
+{
+	return __inet_csk_reqsk_queue_drop(sk, req, false);
 }
 EXPORT_SYMBOL(inet_csk_reqsk_queue_drop);
 
···
 
 		if (!inet_ehash_insert(req_to_sk(nreq), req_to_sk(oreq), NULL)) {
 			/* delete timer */
-			inet_csk_reqsk_queue_drop(sk_listener, nreq);
+			__inet_csk_reqsk_queue_drop(sk_listener, nreq, true);
 			goto no_ownership;
 		}
 
···
 	}
 
 drop:
-	inet_csk_reqsk_queue_drop_and_put(oreq->rsk_listener, oreq);
+	__inet_csk_reqsk_queue_drop(sk_listener, oreq, true);
+	reqsk_put(req);
 }
 
 static bool reqsk_queue_hash_req(struct request_sock *req,

net/ipv4/udp.c

···
 			skb_shinfo(skb)->gso_type = SKB_GSO_UDP_L4;
 			skb_shinfo(skb)->gso_segs = DIV_ROUND_UP(datalen,
 								 cork->gso_size);
+
+			/* Don't checksum the payload, skb will get segmented */
+			goto csum_partial;
 		}
-		goto csum_partial;
 	}
 
 	if (is_udplite)  				 /*     UDP-Lite      */

net/ipv6/udp.c

···
 			skb_shinfo(skb)->gso_type = SKB_GSO_UDP_L4;
 			skb_shinfo(skb)->gso_segs = DIV_ROUND_UP(datalen,
 								 cork->gso_size);
+
+			/* Don't checksum the payload, skb will get segmented */
+			goto csum_partial;
 		}
-		goto csum_partial;
 	}
 
 	if (is_udplite)

net/l2tp/l2tp_netlink.c

···
 				  NLM_F_ACK, tunnel, cmd);
 
 	if (ret >= 0) {
-		ret = genlmsg_multicast_allns(family, msg, 0, 0, GFP_ATOMIC);
+		ret = genlmsg_multicast_allns(family, msg, 0, 0);
 		/* We don't care if no one is listening */
 		if (ret == -ESRCH)
 			ret = 0;
···
 				   NLM_F_ACK, session, cmd);
 
 	if (ret >= 0) {
-		ret = genlmsg_multicast_allns(family, msg, 0, 0, GFP_ATOMIC);
+		ret = genlmsg_multicast_allns(family, msg, 0, 0);
 		/* We don't care if no one is listening */
 		if (ret == -ESRCH)
 			ret = 0;

net/mptcp/mib.c

···
 	SNMP_MIB_ITEM("MPCapableFallbackSYNACK", MPTCP_MIB_MPCAPABLEACTIVEFALLBACK),
 	SNMP_MIB_ITEM("MPCapableSYNTXDrop", MPTCP_MIB_MPCAPABLEACTIVEDROP),
 	SNMP_MIB_ITEM("MPCapableSYNTXDisabled", MPTCP_MIB_MPCAPABLEACTIVEDISABLED),
+	SNMP_MIB_ITEM("MPCapableEndpAttempt", MPTCP_MIB_MPCAPABLEENDPATTEMPT),
 	SNMP_MIB_ITEM("MPFallbackTokenInit", MPTCP_MIB_TOKENFALLBACKINIT),
 	SNMP_MIB_ITEM("MPTCPRetrans", MPTCP_MIB_RETRANSSEGS),
 	SNMP_MIB_ITEM("MPJoinNoTokenFound", MPTCP_MIB_JOINNOTOKEN),

net/mptcp/mib.h

···
 	MPTCP_MIB_MPCAPABLEACTIVEFALLBACK, /* Client-side fallback during 3-way handshake */
 	MPTCP_MIB_MPCAPABLEACTIVEDROP,	/* Client-side fallback due to a MPC drop */
 	MPTCP_MIB_MPCAPABLEACTIVEDISABLED, /* Client-side disabled due to past issues */
+	MPTCP_MIB_MPCAPABLEENDPATTEMPT,	/* Prohibited MPC to port-based endp */
 	MPTCP_MIB_TOKENFALLBACKINIT,	/* Could not init/allocate token */
 	MPTCP_MIB_RETRANSSEGS,		/* Segments retransmitted at the MPTCP-level */
 	MPTCP_MIB_JOINNOTOKEN,		/* Received MP_JOIN but the token was not found */

net/mptcp/pm_netlink.c

···
 				 i, rm_id, id, remote_id, msk->mpc_endpoint_id);
 			spin_unlock_bh(&msk->pm.lock);
 			mptcp_subflow_shutdown(sk, ssk, how);
+			removed |= subflow->request_join;
 
 			/* the following takes care of updating the subflows counter */
 			mptcp_close_ssk(sk, ssk, subflow);
 			spin_lock_bh(&msk->pm.lock);
 
-			removed |= subflow->request_join;
 			if (rm_type == MPTCP_MIB_RMSUBFLOW)
 				__MPTCP_INC_STATS(sock_net(sk), rm_type);
 		}
···
 	 */
 	inet_sk_state_store(newsk, TCP_LISTEN);
 	lock_sock(ssk);
+	WRITE_ONCE(mptcp_subflow_ctx(ssk)->pm_listener, true);
 	err = __inet_listen_sk(ssk, backlog);
 	if (!err)
 		mptcp_event_pm_listener(ssk, MPTCP_EVENT_LISTENER_CREATED);

net/mptcp/protocol.h

···
 		__unused : 8;
 	bool	data_avail;
 	bool	scheduled;
+	bool	pm_listener;	    /* a listener managed by the kernel PM? */
 	u32	remote_nonce;
 	u64	thmac;
 	u32	local_nonce;

net/mptcp/subflow.c

···
 	}
 }
 
+static int subflow_reset_req_endp(struct request_sock *req, struct sk_buff *skb)
+{
+	SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_MPCAPABLEENDPATTEMPT);
+	subflow_add_reset_reason(skb, MPTCP_RST_EPROHIBIT);
+	return -EPERM;
+}
+
 /* Init mptcp request socket.
  *
  * Returns an error code if a JOIN has failed and a TCP reset
···
 	if (opt_mp_capable) {
 		SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_MPCAPABLEPASSIVE);
 
+		if (unlikely(listener->pm_listener))
+			return subflow_reset_req_endp(req, skb);
 		if (opt_mp_join)
 			return 0;
 	} else if (opt_mp_join) {
···
 
 		if (mp_opt.backup)
 			SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINSYNBACKUPRX);
+	} else if (unlikely(listener->pm_listener)) {
+		return subflow_reset_req_endp(req, skb);
 	}
 
 	if (opt_mp_capable && listener->request_mptcp) {

net/netlink/genetlink.c

···
 	if (IS_ERR(msg))
 		return PTR_ERR(msg);
 
-	if (!family->netnsok) {
+	if (!family->netnsok)
 		genlmsg_multicast_netns(&genl_ctrl, &init_net, msg, 0,
 					0, GFP_KERNEL);
-	} else {
-		rcu_read_lock();
-		genlmsg_multicast_allns(&genl_ctrl, msg, 0,
-					0, GFP_ATOMIC);
-		rcu_read_unlock();
-	}
+	else
+		genlmsg_multicast_allns(&genl_ctrl, msg, 0, 0);
 
 	return 0;
 }
···
 
 core_initcall(genl_init);
 
-static int genlmsg_mcast(struct sk_buff *skb, u32 portid, unsigned long group,
-			 gfp_t flags)
+static int genlmsg_mcast(struct sk_buff *skb, u32 portid, unsigned long group)
 {
 	struct sk_buff *tmp;
 	struct net *net, *prev = NULL;
 	bool delivered = false;
 	int err;
 
+	rcu_read_lock();
 	for_each_net_rcu(net) {
 		if (prev) {
-			tmp = skb_clone(skb, flags);
+			tmp = skb_clone(skb, GFP_ATOMIC);
 			if (!tmp) {
 				err = -ENOMEM;
 				goto error;
 			}
 			err = nlmsg_multicast(prev->genl_sock, tmp,
-					      portid, group, flags);
+					      portid, group, GFP_ATOMIC);
 			if (!err)
 				delivered = true;
 			else if (err != -ESRCH)
···
 
 		prev = net;
 	}
+	err = nlmsg_multicast(prev->genl_sock, skb, portid, group, GFP_ATOMIC);
 
-	err = nlmsg_multicast(prev->genl_sock, skb, portid, group, flags);
+	rcu_read_unlock();
+
 	if (!err)
 		delivered = true;
 	else if (err != -ESRCH)
 		return err;
 	return delivered ? 0 : -ESRCH;
  error:
+	rcu_read_unlock();
+
 	kfree_skb(skb);
 	return err;
 }
 
 int genlmsg_multicast_allns(const struct genl_family *family,
 			    struct sk_buff *skb, u32 portid,
-			    unsigned int group, gfp_t flags)
+			    unsigned int group)
 {
 	if (WARN_ON_ONCE(group >= family->n_mcgrps))
 		return -EINVAL;
 
 	group = family->mcgrp_offset + group;
-	return genlmsg_mcast(skb, portid, group, flags);
+	return genlmsg_mcast(skb, portid, group);
 }
 EXPORT_SYMBOL(genlmsg_multicast_allns);
 

net/smc/smc_pnet.c

···
 
 	write_lock(&sn->pnetids_ndev.lock);
 	list_for_each_entry(pi, &sn->pnetids_ndev.list, list) {
-		if (smc_pnet_match(pnetid, pe->pnetid)) {
+		if (smc_pnet_match(pnetid, pi->pnetid)) {
 			refcount_inc(&pi->refcnt);
 			kfree(pe);
 			goto unlock;

net/smc/smc_wr.c

···
 	smc_wr_tx_wait_no_pending_sends(lnk);
 	percpu_ref_kill(&lnk->wr_reg_refs);
 	wait_for_completion(&lnk->reg_ref_comp);
+	percpu_ref_exit(&lnk->wr_reg_refs);
 	percpu_ref_kill(&lnk->wr_tx_refs);
 	wait_for_completion(&lnk->tx_ref_comp);
+	percpu_ref_exit(&lnk->wr_tx_refs);
 
 	if (lnk->wr_rx_dma_addr) {
 		ib_dma_unmap_single(ibdev, lnk->wr_rx_dma_addr,
···
 	init_waitqueue_head(&lnk->wr_reg_wait);
 	rc = percpu_ref_init(&lnk->wr_reg_refs, smcr_wr_reg_refs_free, 0, GFP_KERNEL);
 	if (rc)
-		goto dma_unmap;
+		goto cancel_ref;
 	init_completion(&lnk->reg_ref_comp);
 	init_waitqueue_head(&lnk->wr_rx_empty_wait);
 	return rc;
 
+cancel_ref:
+	percpu_ref_exit(&lnk->wr_tx_refs);
 dma_unmap:
 	if (lnk->wr_rx_v2_dma_addr) {
 		ib_dma_unmap_single(ibdev, lnk->wr_rx_v2_dma_addr,

net/sunrpc/svc.c

···
 svc_process_common(struct svc_rqst *rqstp)
 {
 	struct xdr_stream	*xdr = &rqstp->rq_res_stream;
-	struct svc_program	*progp;
+	struct svc_program	*progp = NULL;
 	const struct svc_procedure *procp = NULL;
 	struct svc_serv		*serv = rqstp->rq_server;
 	struct svc_process_info process;
···
 	rqstp->rq_vers = be32_to_cpup(p++);
 	rqstp->rq_proc = be32_to_cpup(p);
 
-	for (pr = 0; pr < serv->sv_nprogs; pr++) {
-		progp = &serv->sv_programs[pr];
-
-		if (rqstp->rq_prog == progp->pg_prog)
-			break;
-	}
+	for (pr = 0; pr < serv->sv_nprogs; pr++)
+		if (rqstp->rq_prog == serv->sv_programs[pr].pg_prog)
+			progp = &serv->sv_programs[pr];
 
 	/*
 	 * Decode auth data, and add verifier to reply buffer.

net/wireless/nl80211.c

···
 
 	genlmsg_end(msg, hdr);
 
-	rcu_read_lock();
 	genlmsg_multicast_allns(&nl80211_fam, msg, 0,
-				NL80211_MCGRP_REGULATORY, GFP_ATOMIC);
-	rcu_read_unlock();
+				NL80211_MCGRP_REGULATORY);
 
 	return;
 
···
 
 	genlmsg_end(msg, hdr);
 
-	rcu_read_lock();
 	genlmsg_multicast_allns(&nl80211_fam, msg, 0,
-				NL80211_MCGRP_REGULATORY, GFP_ATOMIC);
-	rcu_read_unlock();
+				NL80211_MCGRP_REGULATORY);
 
 	return;
 

rust/kernel/device.rs

···
     ///
     /// It must also be ensured that `bindings::device::release` can be called from any thread.
     /// While not officially documented, this should be the case for any `struct device`.
-    pub unsafe fn from_raw(ptr: *mut bindings::device) -> ARef<Self> {
-        // SAFETY: By the safety requirements, ptr is valid.
-        // Initially increase the reference count by one to compensate for the final decrement once
-        // this newly created `ARef<Device>` instance is dropped.
-        unsafe { bindings::get_device(ptr) };
-
-        // CAST: `Self` is a `repr(transparent)` wrapper around `bindings::device`.
-        let ptr = ptr.cast::<Self>();
-
-        // SAFETY: `ptr` is valid by the safety requirements of this function. By the above call to
-        // `bindings::get_device` we also own a reference to the underlying `struct device`.
-        unsafe { ARef::from_raw(ptr::NonNull::new_unchecked(ptr)) }
+    pub unsafe fn get_device(ptr: *mut bindings::device) -> ARef<Self> {
+        // SAFETY: By the safety requirements ptr is valid
+        unsafe { Self::as_ref(ptr) }.into()
     }
 
     /// Obtain the raw `struct device *`.

rust/kernel/firmware.rs

···
 ///
 /// # fn no_run() -> Result<(), Error> {
 /// # // SAFETY: *NOT* safe, just for the example to get an `ARef<Device>` instance
-/// # let dev = unsafe { Device::from_raw(core::ptr::null_mut()) };
+/// # let dev = unsafe { Device::get_device(core::ptr::null_mut()) };
 ///
 /// let fw = Firmware::request(c_str!("path/to/firmware.bin"), &dev)?;
 /// let blob = fw.data();

tools/testing/selftests/Makefile

···
 TARGETS += rseq
 TARGETS += rtc
 TARGETS += rust
+TARGETS += sched_ext
 TARGETS += seccomp
 TARGETS += sgx
 TARGETS += sigaltstack
···
 endif
 endif
 
-# User can optionally provide a TARGETS skiplist.  By default we skip
-# BPF since it has cutting edge build time dependencies which require
-# more effort to install.
-SKIP_TARGETS ?= bpf
+# User can optionally provide a TARGETS skiplist. By default we skip
+# targets using BPF since it has cutting edge build time dependencies
+# which require more effort to install.
+SKIP_TARGETS ?= bpf sched_ext
 ifneq ($(SKIP_TARGETS),)
 	TMP := $(filter-out $(SKIP_TARGETS), $(TARGETS))
 	override TARGETS := $(TMP)

tools/testing/selftests/ftrace/test.d/ftrace/fgraph-profiler.tc

···
+#!/bin/sh
+# SPDX-License-Identifier: GPL-2.0
+# description: ftrace - function profiler with function graph tracing
+# requires: function_profile_enabled set_ftrace_filter function_graph:tracer
+
+# The function graph tracer can now be run along side of the function
+# profiler. But there was a bug that caused the combination of the two
+# to crash. It also required the function graph tracer to be started
+# first.
+#
+# This test triggers that bug
+#
+# We need both function_graph and profiling to run this test
+
+fail() { # mesg
+    echo $1
+    exit_fail
+}
+
+echo "Enabling function graph tracer:"
+echo function_graph > current_tracer
+echo "enable profiler"
+
+# Older kernels do not allow function_profile to be enabled with
+# function graph tracer. If the below fails, mark it as unsupported
+echo 1 > function_profile_enabled || exit_unsupported
+
+# Let it run for a bit to make sure nothing explodes
+sleep 1
+
+exit 0

tools/testing/selftests/net/lib/py/nsim.py

···
 # SPDX-License-Identifier: GPL-2.0
 
+import errno
 import json
 import os
 import random

tools/testing/selftests/net/mptcp/mptcp_join.sh

···
 cout=""
 err=""
 capout=""
+cappid=""
 ns1=""
 ns2=""
 iptables="iptables"
···
 	fi
 }
 
+cond_start_capture()
+{
+	local ns="$1"
+
+	:> "$capout"
+
+	if $capture; then
+		local capuser capfile
+		if [ -z $SUDO_USER ]; then
+			capuser=""
+		else
+			capuser="-Z $SUDO_USER"
+		fi
+
+		capfile=$(printf "mp_join-%02u-%s.pcap" "$MPTCP_LIB_TEST_COUNTER" "$ns")
+
+		echo "Capturing traffic for test $MPTCP_LIB_TEST_COUNTER into $capfile"
+		ip netns exec "$ns" tcpdump -i any -s 65535 -B 32768 $capuser -w "$capfile" > "$capout" 2>&1 &
+		cappid=$!
+
+		sleep 1
+	fi
+}
+
+cond_stop_capture()
+{
+	if $capture; then
+		sleep 1
+		kill $cappid
+		cat "$capout"
+	fi
+}
+
+get_port()
+{
+	echo "$((10000 + MPTCP_LIB_TEST_COUNTER - 1))"
+}
+
 do_transfer()
 {
 	local listener_ns="$1"
···
 	local cl_proto="$3"
 	local srv_proto="$4"
 	local connect_addr="$5"
+	local port
 
-	local port=$((10000 + MPTCP_LIB_TEST_COUNTER - 1))
-	local cappid
 	local FAILING_LINKS=${FAILING_LINKS:-""}
 	local fastclose=${fastclose:-""}
 	local speed=${speed:-"fast"}
+	port=$(get_port)
 
 	:> "$cout"
 	:> "$sout"
-	:> "$capout"
 
-	if $capture; then
-		local capuser
-		if [ -z $SUDO_USER ] ; then
-			capuser=""
-		else
-			capuser="-Z $SUDO_USER"
-		fi
-
-		capfile=$(printf "mp_join-%02u-%s.pcap" "$MPTCP_LIB_TEST_COUNTER" "${listener_ns}")
-
-		echo "Capturing traffic for test $MPTCP_LIB_TEST_COUNTER into $capfile"
-		ip netns exec ${listener_ns} tcpdump -i any -s 65535 -B 32768 $capuser -w $capfile > "$capout" 2>&1 &
-		cappid=$!
-
-		sleep 1
-	fi
+	cond_start_capture ${listener_ns}
 
 	NSTAT_HISTORY=/tmp/${listener_ns}.nstat ip netns exec ${listener_ns} \
 		nstat -n
···
 	wait $spid
 	local rets=$?
 
-	if $capture; then
-	    sleep 1
-	    kill $cappid
-	fi
+	cond_stop_capture
 
 	NSTAT_HISTORY=/tmp/${listener_ns}.nstat ip netns exec ${listener_ns} \
 		nstat | grep Tcp > /tmp/${listener_ns}.out
···
 		ip netns exec ${connector_ns} ss -Menita 1>&2 -o "dport = :$port"
 		cat /tmp/${connector_ns}.out
 
-		cat "$capout"
 		return 1
 	fi
 
···
 	fi
 	rets=$?
 
-	if [ $retc -eq 0 ] && [ $rets -eq 0 ];then
-		cat "$capout"
-		return 0
-	fi
-
-	cat "$capout"
-	return 1
+	[ $retc -eq 0 ] && [ $rets -eq 0 ]
 }
 
 make_file()
···
 	fail_test
 }
 
+chk_mpc_endp_attempt()
+{
+	local retl=$1
+	local attempts=$2
+
+	print_check "Connect"
+
+	if [ ${retl} = 124 ]; then
+		fail_test "timeout on connect"
+	elif [ ${retl} = 0 ]; then
+		fail_test "unexpected successful connect"
+	else
+		print_ok
+
+		print_check "Attempts"
+		count=$(mptcp_lib_get_counter ${ns1} "MPTcpExtMPCapableEndpAttempt")
+		if [ -z "$count" ]; then
+			print_skip
+		elif [ "$count" != "$attempts" ]; then
+			fail_test "got ${count} MPC attempt[s] on port-based endpoint, expected ${attempts}"
+		else
+			print_ok
+		fi
+	fi
+}
+
 add_addr_ports_tests()
 {
 	# signal address with port
···
 		run_tests $ns1 $ns2 10.0.1.1
 		chk_join_nr 2 2 2
 		chk_add_nr 2 2 2
+	fi
+
+	if reset "port-based signal endpoint must not accept mpc"; then
+		local port retl count
+		port=$(get_port)
+
+		cond_start_capture ${ns1}
+		pm_nl_add_endpoint ${ns1} 10.0.2.1 flags signal port ${port}
+		mptcp_lib_wait_local_port_listen ${ns1} ${port}
+
+		timeout 1 ip netns exec ${ns2} \
+			./mptcp_connect -t ${timeout_poll} -p $port -s MPTCP 10.0.2.1 >/dev/null 2>&1
+		retl=$?
+		cond_stop_capture
+
+		chk_mpc_endp_attempt ${retl} 1
 	fi
 }
 

tools/testing/selftests/net/rds/test.py

···
 import atexit
 from pwd import getpwuid
 from os import stat
-from lib.py import ip
 
+# Allow utils module to be imported from different directory
+this_dir = os.path.dirname(os.path.realpath(__file__))
+sys.path.append(os.path.join(this_dir, "../"))
+from lib.py.utils import ip
 
 libc = ctypes.cdll.LoadLibrary('libc.so.6')
 setns = libc.setns

tools/testing/selftests/rseq/rseq.c

···
 /* Flags used during rseq registration.  */
 unsigned int rseq_flags;
 
-/*
- * rseq feature size supported by the kernel. 0 if the registration was
- * unsuccessful.
- */
-unsigned int rseq_feature_size = -1U;
-
 static int rseq_ownership;
 static int rseq_reg_success;	/* At least one rseq registration has succeded. */
 
···
 	}
 }
 
+/* The rseq areas need to be at least 32 bytes. */
+static
+unsigned int get_rseq_min_alloc_size(void)
+{
+	unsigned int alloc_size = rseq_size;
+
+	if (alloc_size < ORIG_RSEQ_ALLOC_SIZE)
+		alloc_size = ORIG_RSEQ_ALLOC_SIZE;
+	return alloc_size;
+}
+
+/*
+ * Return the feature size supported by the kernel.
+ *
+ * Depending on the value returned by getauxval(AT_RSEQ_FEATURE_SIZE):
+ *
+ * 0:   Return ORIG_RSEQ_FEATURE_SIZE (20)
+ * > 0: Return the value from getauxval(AT_RSEQ_FEATURE_SIZE).
+ *
+ * It should never return a value below ORIG_RSEQ_FEATURE_SIZE.
+ */
+static
+unsigned int get_rseq_kernel_feature_size(void)
+{
+	unsigned long auxv_rseq_feature_size, auxv_rseq_align;
+
+	auxv_rseq_align = getauxval(AT_RSEQ_ALIGN);
+	assert(!auxv_rseq_align || auxv_rseq_align <= RSEQ_THREAD_AREA_ALLOC_SIZE);
+
+	auxv_rseq_feature_size = getauxval(AT_RSEQ_FEATURE_SIZE);
+	assert(!auxv_rseq_feature_size || auxv_rseq_feature_size <= RSEQ_THREAD_AREA_ALLOC_SIZE);
+	if (auxv_rseq_feature_size)
+		return auxv_rseq_feature_size;
+	else
+		return ORIG_RSEQ_FEATURE_SIZE;
+}
+
 int rseq_register_current_thread(void)
 {
 	int rc;
···
 		/* Treat libc's ownership as a successful registration. */
 		return 0;
 	}
-	rc = sys_rseq(&__rseq_abi, rseq_size, 0, RSEQ_SIG);
+	rc = sys_rseq(&__rseq_abi, get_rseq_min_alloc_size(), 0, RSEQ_SIG);
 	if (rc) {
 		if (RSEQ_READ_ONCE(rseq_reg_success)) {
 			/* Incoherent success/failure within process. */
···
 		/* Treat libc's ownership as a successful unregistration. */
 		return 0;
 	}
-	rc = sys_rseq(&__rseq_abi, rseq_size, RSEQ_ABI_FLAG_UNREGISTER, RSEQ_SIG);
+	rc = sys_rseq(&__rseq_abi, get_rseq_min_alloc_size(), RSEQ_ABI_FLAG_UNREGISTER, RSEQ_SIG);
 	if (rc)
 		return -1;
 	return 0;
-}
-
-static
-unsigned int get_rseq_feature_size(void)
-{
-	unsigned long auxv_rseq_feature_size, auxv_rseq_align;
-
-	auxv_rseq_align = getauxval(AT_RSEQ_ALIGN);
-	assert(!auxv_rseq_align || auxv_rseq_align <= RSEQ_THREAD_AREA_ALLOC_SIZE);
-
-	auxv_rseq_feature_size = getauxval(AT_RSEQ_FEATURE_SIZE);
-	assert(!auxv_rseq_feature_size || auxv_rseq_feature_size <= RSEQ_THREAD_AREA_ALLOC_SIZE);
-	if (auxv_rseq_feature_size)
-		return auxv_rseq_feature_size;
-	else
-		return ORIG_RSEQ_FEATURE_SIZE;
 }
 
 static __attribute__((constructor))
···
 	}
 	if (libc_rseq_size_p && libc_rseq_offset_p && libc_rseq_flags_p &&
 			*libc_rseq_size_p != 0) {
+		unsigned int libc_rseq_size;
+
 		/* rseq registration owned by glibc */
 		rseq_offset = *libc_rseq_offset_p;
-		rseq_size = *libc_rseq_size_p;
+		libc_rseq_size = *libc_rseq_size_p;
 		rseq_flags = *libc_rseq_flags_p;
-		rseq_feature_size = get_rseq_feature_size();
-		if (rseq_feature_size > rseq_size)
-			rseq_feature_size = rseq_size;
+
+		/*
+		 * Previous versions of glibc expose the value
+		 * 32 even though the kernel only supported 20
+		 * bytes initially. Therefore treat 32 as a
+		 * special-case. glibc 2.40 exposes a 20 bytes
+		 * __rseq_size without using getauxval(3) to
+		 * query the supported size, while still allocating a 32
+		 * bytes area. Also treat 20 as a special-case.
+		 *
+		 * Special-cases are handled by using the following
+		 * value as active feature set size:
+		 *
+		 *   rseq_size = min(32, get_rseq_kernel_feature_size())
+		 */
+		switch (libc_rseq_size) {
+		case ORIG_RSEQ_FEATURE_SIZE:
+			fallthrough;
+		case ORIG_RSEQ_ALLOC_SIZE:
+		{
+			unsigned int rseq_kernel_feature_size = get_rseq_kernel_feature_size();
+
+			if (rseq_kernel_feature_size < ORIG_RSEQ_ALLOC_SIZE)
+				rseq_size = rseq_kernel_feature_size;
+			else
+				rseq_size = ORIG_RSEQ_ALLOC_SIZE;
+			break;
+		}
+		default:
+			/* Otherwise just use the __rseq_size from libc as rseq_size. */
+			rseq_size = libc_rseq_size;
+			break;
+		}
 		return;
 	}
 	rseq_ownership = 1;
 	if (!rseq_available()) {
 		rseq_size = 0;
-		rseq_feature_size = 0;
 		return;
 	}
 	rseq_offset = (void *)&__rseq_abi - rseq_thread_pointer();
 	rseq_flags = 0;
-	rseq_feature_size = get_rseq_feature_size();
-	if (rseq_feature_size == ORIG_RSEQ_FEATURE_SIZE)
-		rseq_size = ORIG_RSEQ_ALLOC_SIZE;
-	else
-		rseq_size = RSEQ_THREAD_AREA_ALLOC_SIZE;
 }
 
 static __attribute__((destructor))
···
 		return;
 	rseq_offset = 0;
 	rseq_size = -1U;
-	rseq_feature_size = -1U;
 	rseq_ownership = 0;
 }
 

tools/testing/selftests/rseq/rseq.h

···
 /* Flags used during rseq registration. */
 extern unsigned int rseq_flags;
 
-/*
- * rseq feature size supported by the kernel. 0 if the registration was
- * unsuccessful.
- */
-extern unsigned int rseq_feature_size;
-
 enum rseq_mo {
 	RSEQ_MO_RELAXED = 0,
 	RSEQ_MO_CONSUME = 1,	/* Unused */
···
 
 static inline bool rseq_node_id_available(void)
 {
-	return (int) rseq_feature_size >= rseq_offsetofend(struct rseq_abi, node_id);
+	return (int) rseq_size >= rseq_offsetofend(struct rseq_abi, node_id);
 }
 
 /*
···
 
 static inline bool rseq_mm_cid_available(void)
 {
-	return (int) rseq_feature_size >= rseq_offsetofend(struct rseq_abi, mm_cid);
+	return (int) rseq_size >= rseq_offsetofend(struct rseq_abi, mm_cid);
 }
 
 static inline uint32_t rseq_current_mm_cid(void)

tools/testing/selftests/sched_ext/Makefile

···
 include ../../../build/Build.include
 include ../../../scripts/Makefile.arch
 include ../../../scripts/Makefile.include
+
+TEST_GEN_PROGS := runner
+
+# override lib.mk's default rules
+OVERRIDE_TARGETS := 1
 include ../lib.mk
-
-ifneq ($(LLVM),)
-ifneq ($(filter %/,$(LLVM)),)
-LLVM_PREFIX := $(LLVM)
-else ifneq ($(filter -%,$(LLVM)),)
-LLVM_SUFFIX := $(LLVM)
-endif
-
-CC := $(LLVM_PREFIX)clang$(LLVM_SUFFIX) $(CLANG_FLAGS) -fintegrated-as
-else
-CC := gcc
-endif # LLVM
-
-ifneq ($(CROSS_COMPILE),)
-$(error CROSS_COMPILE not supported for scx selftests)
-endif # CROSS_COMPILE
 
 CURDIR := $(abspath .)
 REPOROOT := $(abspath ../../../..)
···
 SCXTOOLSDIR := $(TOOLSDIR)/sched_ext
 SCXTOOLSINCDIR := $(TOOLSDIR)/sched_ext/include
 
-OUTPUT_DIR := $(CURDIR)/build
+OUTPUT_DIR := $(OUTPUT)/build
 OBJ_DIR := $(OUTPUT_DIR)/obj
 INCLUDE_DIR := $(OUTPUT_DIR)/include
 BPFOBJ_DIR := $(OBJ_DIR)/libbpf
 SCXOBJ_DIR := $(OBJ_DIR)/sched_ext
 BPFOBJ := $(BPFOBJ_DIR)/libbpf.a
 LIBBPF_OUTPUT := $(OBJ_DIR)/libbpf/libbpf.a
-DEFAULT_BPFTOOL := $(OUTPUT_DIR)/sbin/bpftool
-HOST_BUILD_DIR := $(OBJ_DIR)
-HOST_OUTPUT_DIR := $(OUTPUT_DIR)
 
-VMLINUX_BTF_PATHS ?= ../../../../vmlinux					\
+DEFAULT_BPFTOOL := $(OUTPUT_DIR)/host/sbin/bpftool
+HOST_OBJ_DIR := $(OBJ_DIR)/host/bpftool
+HOST_LIBBPF_OUTPUT := $(OBJ_DIR)/host/libbpf/
+HOST_LIBBPF_DESTDIR := $(OUTPUT_DIR)/host/
+HOST_DESTDIR := $(OUTPUT_DIR)/host/
+
+VMLINUX_BTF_PATHS ?= $(if $(O),$(O)/vmlinux)					\
+		     $(if $(KBUILD_OUTPUT),$(KBUILD_OUTPUT)/vmlinux)		\
+		     ../../../../vmlinux					\
 		     /sys/kernel/btf/vmlinux					\
 		     /boot/vmlinux-$(shell uname -r)
 VMLINUX_BTF ?= $(abspath $(firstword $(wildcard $(VMLINUX_BTF_PATHS))))
···
 # Use '-idirafter': Don't interfere with include mechanics except where the
 # build would have failed anyways.
 define get_sys_includes
-$(shell $(1) -v -E - </dev/null 2>&1 \
+$(shell $(1) $(2) -v -E - </dev/null 2>&1 \
 	| sed -n '/<...> search starts here:/,/End of search list./{ s| \(/.*\)|-idirafter \1|p }') \
-$(shell $(1) -dM -E - </dev/null | grep '__riscv_xlen ' | awk '{printf("-D__riscv_xlen=%d -D__BITS_PER_LONG=%d", $$3, $$3)}')
+$(shell $(1) $(2) -dM -E - </dev/null | grep '__riscv_xlen ' | awk '{printf("-D__riscv_xlen=%d -D__BITS_PER_LONG=%d", $$3, $$3)}')
 endef
+
+ifneq ($(CROSS_COMPILE),)
+CLANG_TARGET_ARCH = --target=$(notdir $(CROSS_COMPILE:%-=%))
+endif
+
+CLANG_SYS_INCLUDES = $(call get_sys_includes,$(CLANG),$(CLANG_TARGET_ARCH))
 
 BPF_CFLAGS = -g -D__TARGET_ARCH_$(SRCARCH)					\
 	     $(if $(IS_LITTLE_ENDIAN),-mlittle-endian,-mbig-endian)		\
 	     -I$(CURDIR)/include -I$(CURDIR)/include/bpf-compat			\
 	     -I$(INCLUDE_DIR) -I$(APIDIR) -I$(SCXTOOLSINCDIR)			\
 	     -I$(REPOROOT)/include						\
-	     $(call get_sys_includes,$(CLANG))					\
+	     $(CLANG_SYS_INCLUDES) 						\
 	     -Wall -Wno-compare-distinct-pointer-types				\
 	     -Wno-incompatible-function-pointer-types				\
 	     -O2 -mcpu=v3
···
 # sort removes libbpf duplicates when not cross-building
 MAKE_DIRS := $(sort $(OBJ_DIR)/libbpf $(OBJ_DIR)/libbpf				\
 	       $(OBJ_DIR)/bpftool $(OBJ_DIR)/resolve_btfids			\
-	       $(INCLUDE_DIR) $(SCXOBJ_DIR))
+	       $(HOST_OBJ_DIR) $(INCLUDE_DIR) $(SCXOBJ_DIR))
 
 $(MAKE_DIRS):
 	$(call msg,MKDIR,,$@)
···
 	   $(APIDIR)/linux/bpf.h						\
 	   | $(OBJ_DIR)/libbpf
 	$(Q)$(MAKE) $(submake_extras) -C $(BPFDIR) OUTPUT=$(OBJ_DIR)/libbpf/	\
+		    ARCH=$(ARCH) CC="$(CC)" CROSS_COMPILE=$(CROSS_COMPILE)	\
 		    EXTRA_CFLAGS='-g -O0 -fPIC'					\
 		    DESTDIR=$(OUTPUT_DIR) prefix= all install_headers
 
 $(DEFAULT_BPFTOOL): $(wildcard $(BPFTOOLDIR)/*.[ch] $(BPFTOOLDIR)/Makefile)	\
-		    $(LIBBPF_OUTPUT) | $(OBJ_DIR)/bpftool
+		    $(LIBBPF_OUTPUT) | $(HOST_OBJ_DIR)
 	$(Q)$(MAKE) $(submake_extras)  -C $(BPFTOOLDIR)				\
 		    ARCH= CROSS_COMPILE= CC=$(HOSTCC) LD=$(HOSTLD)		\
 		    EXTRA_CFLAGS='-g -O0'					\
-		    OUTPUT=$(OBJ_DIR)/bpftool/					\
-		    LIBBPF_OUTPUT=$(OBJ_DIR)/libbpf/				\
-		    LIBBPF_DESTDIR=$(OUTPUT_DIR)/				\
-		    prefix= DESTDIR=$(OUTPUT_DIR)/ install-bin
+		    OUTPUT=$(HOST_OBJ_DIR)/					\
+		    LIBBPF_OUTPUT=$(HOST_LIBBPF_OUTPUT)				\
+		    LIBBPF_DESTDIR=$(HOST_LIBBPF_DESTDIR)			\
+		    prefix= DESTDIR=$(HOST_DESTDIR) install-bin
 
 $(INCLUDE_DIR)/vmlinux.h: $(VMLINUX_BTF) $(BPFTOOL) | $(INCLUDE_DIR)
 ifeq ($(VMLINUX_H),)
···
 
 override define CLEAN
 	rm -rf $(OUTPUT_DIR)
-	rm -f *.o *.bpf.o *.bpf.skel.h *.bpf.subskel.h
 	rm -f $(TEST_GEN_PROGS)
-	rm -f runner
 endef
 
 # Every testcase takes all of the BPF progs are dependencies by default. This
···
 # function doesn't support using implicit rules otherwise.
 $(testcase-targets): $(SCXOBJ_DIR)/%.o: %.c $(SCXOBJ_DIR)/runner.o $(all_test_bpfprogs) | $(SCXOBJ_DIR)
 	$(eval test=$(patsubst %.o,%.c,$(notdir $@)))
-	$(CC) $(CFLAGS) -c $< -o $@ $(SCXOBJ_DIR)/runner.o
+	$(CC) $(CFLAGS) -c $< -o $@
 
 $(SCXOBJ_DIR)/util.o: util.c | $(SCXOBJ_DIR)
 	$(CC) $(CFLAGS) -c $< -o $@
 
-runner: $(SCXOBJ_DIR)/runner.o $(SCXOBJ_DIR)/util.o $(BPFOBJ) $(testcase-targets)
+$(OUTPUT)/runner: $(SCXOBJ_DIR)/runner.o $(SCXOBJ_DIR)/util.o $(BPFOBJ) $(testcase-targets)
 	@echo "$(testcase-targets)"
 	$(CC) $(CFLAGS) -o $@ $^ $(LDFLAGS)
-
-TEST_GEN_PROGS := runner
-
-all: runner
-
-.PHONY: all clean help
 
 .DEFAULT_GOAL := all
 

tools/testing/selftests/vDSO/Makefile

···
 TEST_GEN_PROGS += vdso_standalone_test_x86
 endif
 TEST_GEN_PROGS += vdso_test_correctness
-ifeq ($(ARCH)$(CONFIG_X86_32),$(filter $(ARCH)$(CONFIG_X86_32),x86 x86_64 loongarch arm64 powerpc s390))
 TEST_GEN_PROGS += vdso_test_getrandom
 TEST_GEN_PROGS += vdso_test_chacha
-endif
 
 CFLAGS := -std=gnu99 -O2
 
···
                                          $(KHDR_INCLUDES) \
                                          -isystem $(top_srcdir)/include/uapi
 
-$(OUTPUT)/vdso_test_chacha: $(top_srcdir)/tools/arch/$(SRCARCH)/vdso/vgetrandom-chacha.S
+$(OUTPUT)/vdso_test_chacha: vgetrandom-chacha.S
 $(OUTPUT)/vdso_test_chacha: CFLAGS += -idirafter $(top_srcdir)/tools/include \
                                       -idirafter $(top_srcdir)/tools/include/generated \
                                       -idirafter $(top_srcdir)/arch/$(SRCARCH)/include \
                                       -idirafter $(top_srcdir)/include \
-                                      -D__ASSEMBLY__ -Wa,--noexecstack
+                                      -Wa,--noexecstack

tools/testing/selftests/vDSO/vdso_test_chacha.c

···
  * Copyright (C) 2022-2024 Jason A. Donenfeld <[e-mail]>. All Rights Reserved.
  */
 
+#include <linux/compiler.h>
 #include <tools/le_byteshift.h>
 #include <sys/random.h>
 #include <sys/auxv.h>
···
 	counter[1] = s[13];
 }
 
-typedef uint8_t u8;
-typedef uint32_t u32;
-typedef uint64_t u64;
-#include <vdso/getrandom.h>
+void __weak __arch_chacha20_blocks_nostack(uint8_t *dst_bytes, const uint32_t *key, uint32_t *counter, size_t nblocks)
+{
+	ksft_exit_skip("Not implemented on architecture\n");
+}
 
 int main(int argc, char *argv[])
 {
···
 	ksft_set_plan(1);
 
 	for (unsigned int trial = 0; trial < TRIALS; ++trial) {
-		if (getrandom(key, sizeof(key), 0) != sizeof(key)) {
-			printf("getrandom() failed!\n");
-			return KSFT_SKIP;
-		}
+		if (getrandom(key, sizeof(key), 0) != sizeof(key))
+			ksft_exit_skip("getrandom() failed unexpectedly\n");
 		memset(counter1, 0, sizeof(counter1));
 		reference_chacha20_blocks(output1, key, counter1, BLOCKS);
 		for (unsigned int split = 0; split < BLOCKS; ++split) {
···
 			if (split)
 				__arch_chacha20_blocks_nostack(output2, key, counter2, split);
 			__arch_chacha20_blocks_nostack(output2 + split * BLOCK_SIZE, key, counter2, BLOCKS - split);
-			if (memcmp(output1, output2, sizeof(output1)) || memcmp(counter1, counter2, sizeof(counter1)))
-				return KSFT_FAIL;
+			if (memcmp(output1, output2, sizeof(output1)))
+				ksft_exit_fail_msg("Main loop outputs do not match on trial %u, split %u\n", trial, split);
+			if (memcmp(counter1, counter2, sizeof(counter1)))
+				ksft_exit_fail_msg("Main loop counters do not match on trial %u, split %u\n", trial, split);
 		}
 	}
 	memset(counter1, 0, sizeof(counter1));
···
 
 	reference_chacha20_blocks(output1, key, counter1, BLOCKS);
 	__arch_chacha20_blocks_nostack(output2, key, counter2, BLOCKS);
-	if (memcmp(output1, output2, sizeof(output1)) || memcmp(counter1, counter2, sizeof(counter1)))
-		return KSFT_FAIL;
+	if (memcmp(output1, output2, sizeof(output1)))
+		ksft_exit_fail_msg("Block limit outputs do not match after first round\n");
+	if (memcmp(counter1, counter2, sizeof(counter1)))
+		ksft_exit_fail_msg("Block limit counters do not match after first round\n");
 
 	reference_chacha20_blocks(output1, key, counter1, BLOCKS);
 	__arch_chacha20_blocks_nostack(output2, key, counter2, BLOCKS);
-	if (memcmp(output1, output2, sizeof(output1)) || memcmp(counter1, counter2, sizeof(counter1)))
-		return KSFT_FAIL;
+	if (memcmp(output1, output2, sizeof(output1)))
+		ksft_exit_fail_msg("Block limit outputs do not match after second round\n");
+	if (memcmp(counter1, counter2, sizeof(counter1)))
+		ksft_exit_fail_msg("Block limit counters do not match after second round\n");
 
 	ksft_test_result_pass("chacha: PASS\n");
-	return KSFT_PASS;
+	ksft_exit_pass();
+	return 0;
 }

tools/testing/selftests/vDSO/vdso_test_getrandom.c

···
 #include <string.h>
 #include <time.h>
 #include <unistd.h>
+#include <sched.h>
 #include <signal.h>
 #include <sys/auxv.h>
 #include <sys/mman.h>
···
 		}							\
 	} while (0)
 #endif
+
+#define ksft_assert(condition) \
+	do { if (!(condition)) ksft_exit_fail_msg("Assertion failed: %s\n", #condition); } while (0)
 
 static struct {
 	pthread_mutex_t lock;
···
 	const char *version = versions[VDSO_VERSION];
 	const char *name = names[VDSO_NAMES][6];
 	unsigned long sysinfo_ehdr = getauxval(AT_SYSINFO_EHDR);
-	size_t ret;
+	ssize_t ret;
 
-	if (!sysinfo_ehdr) {
-		printf("AT_SYSINFO_EHDR is not present!\n");
-		exit(KSFT_SKIP);
-	}
+	if (!sysinfo_ehdr)
+		ksft_exit_skip("AT_SYSINFO_EHDR is not present\n");
 	vdso_init_from_sysinfo_ehdr(sysinfo_ehdr);
 	vgrnd.fn = (__typeof__(vgrnd.fn))vdso_sym(version, name);
-	if (!vgrnd.fn) {
-		printf("%s is missing!\n", name);
-		exit(KSFT_FAIL);
-	}
+	if (!vgrnd.fn)
+		ksft_exit_skip("%s@%s symbol is missing from vDSO\n", name, version);
 	ret = VDSO_CALL(vgrnd.fn, 5, NULL, 0, 0, &vgrnd.params, ~0UL);
-	if (ret == -ENOSYS) {
-		printf("unsupported architecture\n");
-		exit(KSFT_SKIP);
-	} else if (ret) {
-		printf("failed to fetch vgetrandom params!\n");
-		exit(KSFT_FAIL);
-	}
+	if (ret == -ENOSYS)
+		ksft_exit_skip("CPU does not have runtime support\n");
+	else if (ret)
+		ksft_exit_fail_msg("Failed to fetch vgetrandom params: %zd\n", ret);
 }
 
 static ssize_t vgetrandom(void *buf, size_t len, unsigned long flags)
···
 
 	if (!state) {
 		state = vgetrandom_get_state();
-		if (!state) {
-			printf("vgetrandom_get_state failed!\n");
-			exit(KSFT_FAIL);
-		}
+		ksft_assert(state);
 	}
 	return VDSO_CALL(vgrnd.fn, 5, buf, len, flags, state, vgrnd.params.size_of_opaque_state);
 }
···
 	for (size_t i = 0; i < TRIALS; ++i) {
 		unsigned int val;
 		ssize_t ret = vgetrandom(&val, sizeof(val), 0);
-		assert(ret == sizeof(val));
+		ksft_assert(ret == sizeof(val));
 	}
 	return NULL;
 }
···
 	for (size_t i = 0; i < TRIALS; ++i) {
 		unsigned int val;
 		ssize_t ret = getrandom(&val, sizeof(val), 0);
-		assert(ret == sizeof(val));
+		ksft_assert(ret == sizeof(val));
 	}
 	return NULL;
 }
···
 	for (size_t i = 0; i < TRIALS; ++i) {
 		unsigned int val;
 		ssize_t ret = syscall(__NR_getrandom, &val, sizeof(val), 0);
-		assert(ret == sizeof(val));
+		ksft_assert(ret == sizeof(val));
 	}
 	return NULL;
 }
···
 
 	clock_gettime(CLOCK_MONOTONIC, &start);
 	for (size_t i = 0; i < THREADS; ++i)
-		assert(pthread_create(&threads[i], NULL, test_vdso_getrandom, NULL) == 0);
+		ksft_assert(pthread_create(&threads[i], NULL, test_vdso_getrandom, NULL) == 0);
 	for (size_t i = 0; i < THREADS; ++i)
 		pthread_join(threads[i], NULL);
 	clock_gettime(CLOCK_MONOTONIC, &end);
···
 
 	clock_gettime(CLOCK_MONOTONIC, &start);
 	for (size_t i = 0; i < THREADS; ++i)
-		assert(pthread_create(&threads[i], NULL, test_libc_getrandom, NULL) == 0);
+		ksft_assert(pthread_create(&threads[i], NULL, test_libc_getrandom, NULL) == 0);
 	for (size_t i = 0; i < THREADS; ++i)
 		pthread_join(threads[i], NULL);
 	clock_gettime(CLOCK_MONOTONIC, &end);
···
 
 	clock_gettime(CLOCK_MONOTONIC, &start);
 	for (size_t i = 0; i < THREADS; ++i)
-		assert(pthread_create(&threads[i], NULL, test_syscall_getrandom, NULL) == 0);
+		ksft_assert(pthread_create(&threads[i], NULL, test_syscall_getrandom, NULL) == 0);
 	for (size_t i = 0; i < THREADS; ++i)
 		pthread_join(threads[i], NULL);
 	clock_gettime(CLOCK_MONOTONIC, &end);
···
 
 	for (size_t i = 0; i < 1000; ++i) {
 		ssize_t ret = vgetrandom(weird_size, sizeof(weird_size), 0);
-		if (ret != sizeof(weird_size))
-			exit(KSFT_FAIL);
+		ksft_assert(ret == sizeof(weird_size));
 	}
 
 	ksft_test_result_pass("getrandom: PASS\n");
 
 	unshare(CLONE_NEWUSER);
-	assert(unshare(CLONE_NEWTIME) == 0);
+	ksft_assert(unshare(CLONE_NEWTIME) == 0);
 	child = fork();
-	assert(child >= 0);
+	ksft_assert(child >= 0);
 	if (!child) {
 		vgetrandom_init();
 		child = getpid();
-		assert(ptrace(PTRACE_TRACEME, 0, NULL, NULL) == 0);
-		assert(kill(child, SIGSTOP) == 0);
-		assert(vgetrandom(weird_size, sizeof(weird_size), 0) == sizeof(weird_size));
+		ksft_assert(ptrace(PTRACE_TRACEME, 0, NULL, NULL) == 0);
+		ksft_assert(kill(child, SIGSTOP) == 0);
+		ksft_assert(vgetrandom(weird_size, sizeof(weird_size), 0) == sizeof(weird_size));
 		_exit(0);
 	}
 	for (;;) {
 		struct ptrace_syscall_info info = { 0 };
 		int status, ret;
-		assert(waitpid(child, &status, 0) >= 0);
+		ksft_assert(waitpid(child, &status, 0) >= 0);
 		if (WIFEXITED(status)) {
-			if (WEXITSTATUS(status) != 0)
-				exit(KSFT_FAIL);
+			ksft_assert(WEXITSTATUS(status) == 0);
 			break;
 		}
-		assert(WIFSTOPPED(status));
+		ksft_assert(WIFSTOPPED(status));
 		if (WSTOPSIG(status) == SIGSTOP)
-			assert(ptrace(PTRACE_SETOPTIONS, child, 0, PTRACE_O_TRACESYSGOOD) == 0);
+			ksft_assert(ptrace(PTRACE_SETOPTIONS, child, 0, PTRACE_O_TRACESYSGOOD) == 0);
 		else if (WSTOPSIG(status) == (SIGTRAP | 0x80)) {
-			assert(ptrace(PTRACE_GET_SYSCALL_INFO, child, sizeof(info), &info) > 0);
+			ksft_assert(ptrace(PTRACE_GET_SYSCALL_INFO, child, sizeof(info), &info) > 0);
 			if (info.op == PTRACE_SYSCALL_INFO_ENTRY && info.entry.nr == __NR_getrandom &&
 			    info.entry.args[0] == (uintptr_t)weird_size && info.entry.args[1] == sizeof(weird_size))
-				exit(KSFT_FAIL);
+				ksft_exit_fail_msg("vgetrandom passed buffer to syscall getrandom unexpectedly\n");
 		}
-		assert(ptrace(PTRACE_SYSCALL, child, 0, 0) == 0);
+		ksft_assert(ptrace(PTRACE_SYSCALL, child, 0, 0) == 0);
 	}
 
 	ksft_test_result_pass("getrandom timens: PASS\n");
 
-	exit(KSFT_PASS);
+	ksft_exit_pass();
 }
 
 static void usage(const char *argv0)

tools/testing/selftests/vDSO/vgetrandom-chacha.S

···
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2024 Jason A. Donenfeld <[e-mail]>. All Rights Reserved.
+ */
+
+#define __ASSEMBLY__
+
+#if defined(__aarch64__)
+#include "../../../../arch/arm64/kernel/vdso/vgetrandom-chacha.S"
+#elif defined(__loongarch__)
+#include "../../../../arch/loongarch/vdso/vgetrandom-chacha.S"
+#elif defined(__powerpc__) || defined(__powerpc64__)
+#include "../../../../arch/powerpc/kernel/vdso/vgetrandom-chacha.S"
+#elif defined(__s390x__)
+#include "../../../../arch/s390/kernel/vdso64/vgetrandom-chacha.S"
+#elif defined(__x86_64__)
+#include "../../../../arch/x86/entry/vdso/vgetrandom-chacha.S"
+#endif