5.15.195-rc1 review

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vineeth Vijayan <[email protected]>

[ Upstream commit 9daa5a8795865f9a3c93d8d1066785b07ded6073 ]

Starting with 'commit 2297791c92d0 ("s390/cio: dont unregister subchannel from child-drivers")', cio no longer unregisters subchannels when the attached device is invalid or unavailable.

As an unintended side-effect, the cio_ignore purge function no longer removes subchannels for devices on the cio_ignore list if no CCW device is attached. This situation occurs when a CCW device is non-operational or unavailable

To ensure the same outcome of the purge function as when the current cio_ignore list had been active during boot, update the purge function to remove I/O subchannels without working CCW devices if the associated device number is found on the cio_ignore list.

---
 drivers/s390/cio/device.c | 37 ++++++++++++++++++++++++-------------
 1 file changed, 24 insertions(+), 13 deletions(-)

index 70c5b85d2dfc9..22fa1296a5168 100644
@@ -1323,23 +1323,34 @@ void ccw_device_schedule_recovery(void)
 	spin_unlock_irqrestore(&recovery_lock, flags);
 }
 
-static int purge_fn(struct device *dev, void *data)
+static int purge_fn(struct subchannel *sch, void *data)
 {
-	struct ccw_device *cdev = to_ccwdev(dev);
-	struct ccw_dev_id *id = &cdev->private->dev_id;
-	struct subchannel *sch = to_subchannel(cdev->dev.parent);
+	struct ccw_device *cdev;
 
-	spin_lock_irq(cdev->ccwlock);
-	if (is_blacklisted(id->ssid, id->devno) &&
-	    (cdev->private->state == DEV_STATE_OFFLINE) &&
-	    (atomic_cmpxchg(&cdev->private->onoff, 0, 1) == 0)) {
-		CIO_MSG_EVENT(3, "ccw: purging 0.%x.%04x\n", id->ssid,
-			      id->devno);
+	spin_lock_irq(&sch->lock);
+	if (sch->st != SUBCHANNEL_TYPE_IO || !sch->schib.pmcw.dnv)
+		goto unlock;
+
+	if (!is_blacklisted(sch->schid.ssid, sch->schib.pmcw.dev))
+		goto unlock;
+
+	cdev = sch_get_cdev(sch);
+	if (cdev) {
+		if (cdev->private->state != DEV_STATE_OFFLINE)
+			goto unlock;
+
+		if (atomic_cmpxchg(&cdev->private->onoff, 0, 1) != 0)
+			goto unlock;
 		ccw_device_sched_todo(cdev, CDEV_TODO_UNREG);
-		css_sched_sch_todo(sch, SCH_TODO_UNREG);
 		atomic_set(&cdev->private->onoff, 0);
 	}
-	spin_unlock_irq(cdev->ccwlock);
+
+	css_sched_sch_todo(sch, SCH_TODO_UNREG);
+	CIO_MSG_EVENT(3, "ccw: purging 0.%x.%04x%s\n", sch->schid.ssid,
+		      sch->schib.pmcw.dev, cdev ? "" : " (no cdev)");
+
+unlock:
+	spin_unlock_irq(&sch->lock);
 	/* Abort loop in case of pending signal. */
 	if (signal_pending(current))
 		return -EINTR;

@@ -1355,7 +1366,7 @@ static int purge_fn(struct device *dev, void *data)
 int ccw_purge_blacklisted(void)
 {
 	CIO_MSG_EVENT(2, "ccw: purging blacklisted devices\n");
-	bus_for_each_dev(&ccw_bus_type, NULL, NULL, purge_fn);
+	for_each_subchannel_staged(purge_fn, NULL, NULL);
 	return 0;
 }
 

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Thomas Zimmermann <[email protected]>

[ Upstream commit 2985c96485b7ef4e015d13dc3081fb0479260951 ]

Besides some legacy code, vmwgfx is the only user of DRM's hash- table implementation. Copy the code into the driver, so that the core code can be retired.

No functional changes. However, the real solution for vmwgfx is to use Linux' generic hash-table functions.

v2:
* add TODO item for updating vmwgfx (Sam)

Stable-dep-of: dfe1323ab3c8 ("drm/vmwgfx: Fix Use-after-free in validation")

---
 Documentation/gpu/todo.rst                 |  11 ++
 drivers/gpu/drm/vmwgfx/Makefile            |   2 +-
 drivers/gpu/drm/vmwgfx/ttm_object.c        |  52 +++---
 drivers/gpu/drm/vmwgfx/ttm_object.h        |   3 +-
 drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf_res.c |  24 +--
 drivers/gpu/drm/vmwgfx/vmwgfx_drv.c        |   2 +-
 drivers/gpu/drm/vmwgfx/vmwgfx_drv.h        |   6 +-
 drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c    |   2 +-
 drivers/gpu/drm/vmwgfx/vmwgfx_hashtab.c    | 199 +++++++++++++++++++++
 drivers/gpu/drm/vmwgfx/vmwgfx_hashtab.h    |  83 +++++++++
 drivers/gpu/drm/vmwgfx/vmwgfx_validation.c |  22 +--
 drivers/gpu/drm/vmwgfx/vmwgfx_validation.h |   7 +-
 12 files changed, 353 insertions(+), 60 deletions(-)
 create mode 100644 drivers/gpu/drm/vmwgfx/vmwgfx_hashtab.c
 create mode 100644 drivers/gpu/drm/vmwgfx/vmwgfx_hashtab.h

index 67de1e94fdf76..f31a838d09fbb 100644
@@ -635,6 +635,17 @@ See drivers/gpu/drm/amd/display/TODO for tasks.
 
 Contact: Harry Wentland, Alex Deucher
 
+vmwgfx: Replace hashtable with Linux' implementation
+----------------------------------------------------
+
+The vmwgfx driver uses its own hashtable implementation. Replace the
+code with Linux' implementation and update the callers. It's mostly a
+refactoring task, but the interfaces are different.
+
+Contact: Zack Rusin, Thomas Zimmermann <[email protected]>
+
+Level: Intermediate
+
 Bootsplash
 ==========
 

index 18edc7ca5b454..59b0b77456dd2 100644
@@ -1,5 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
-vmwgfx-y := vmwgfx_execbuf.o vmwgfx_gmr.o vmwgfx_kms.o vmwgfx_drv.o \
+vmwgfx-y := vmwgfx_execbuf.o vmwgfx_gmr.o vmwgfx_hashtab.o vmwgfx_kms.o vmwgfx_drv.o \
 	    vmwgfx_ioctl.o vmwgfx_resource.o vmwgfx_ttm_buffer.o \
 	    vmwgfx_cmd.o vmwgfx_irq.o vmwgfx_ldu.o vmwgfx_ttm_glue.o \
 	    vmwgfx_overlay.o vmwgfx_gmrid_manager.o vmwgfx_fence.o \

index 49b064f0cb19c..6885ccbeec7a1 100644
@@ -42,9 +42,8 @@
 #include <linux/list.h>
 #include <linux/rcupdate.h>
 
-#include <drm/drm_hashtab.h>
-
 #include "ttm_memory.h"
+#include "vmwgfx_hashtab.h"
 
 /**
  * enum ttm_ref_type

index 8381750db81b6..494cb98061f22 100644
@@ -42,7 +42,7 @@
  */
 struct vmw_cmdbuf_res {
 	struct vmw_resource *res;
-	struct drm_hash_item hash;
+	struct vmwgfx_hash_item hash;
 	struct list_head head;
 	enum vmw_cmdbuf_res_state state;
 	struct vmw_cmdbuf_res_manager *man;

@@ -59,7 +59,7 @@ struct vmw_cmdbuf_res {
  * @resources and @list are protected by the cmdbuf mutex for now.
  */
 struct vmw_cmdbuf_res_manager {
-	struct drm_open_hash resources;
+	struct vmwgfx_open_hash resources;
 	struct list_head list;
 	struct vmw_private *dev_priv;
 };

@@ -81,11 +81,11 @@ vmw_cmdbuf_res_lookup(struct vmw_cmdbuf_res_manager *man,
 		      enum vmw_cmdbuf_res_type res_type,
 		      u32 user_key)
 {
-	struct drm_hash_item *hash;
+	struct vmwgfx_hash_item *hash;
 	int ret;
 	unsigned long key = user_key | (res_type << 24);
 
-	ret = drm_ht_find_item(&man->resources, key, &hash);
+	ret = vmwgfx_ht_find_item(&man->resources, key, &hash);
 	if (unlikely(ret != 0))
 		return ERR_PTR(ret);
 

@@ -105,7 +105,7 @@ static void vmw_cmdbuf_res_free(struct vmw_cmdbuf_res_manager *man,
 				struct vmw_cmdbuf_res *entry)
 {
 	list_del(&entry->head);
-	WARN_ON(drm_ht_remove_item(&man->resources, &entry->hash));
+	WARN_ON(vmwgfx_ht_remove_item(&man->resources, &entry->hash));
 	vmw_resource_unreference(&entry->res);
 	kfree(entry);
 }

@@ -167,7 +167,7 @@ void vmw_cmdbuf_res_revert(struct list_head *list)
 			vmw_cmdbuf_res_free(entry->man, entry);
 			break;
 		case VMW_CMDBUF_RES_DEL:
-			ret = drm_ht_insert_item(&entry->man->resources, &entry->hash);
+			ret = vmwgfx_ht_insert_item(&entry->man->resources, &entry->hash);
 			BUG_ON(ret);
 			list_move_tail(&entry->head, &entry->man->list);
 			entry->state = VMW_CMDBUF_RES_COMMITTED;

@@ -206,7 +206,7 @@ int vmw_cmdbuf_res_add(struct vmw_cmdbuf_res_manager *man,
 		return -ENOMEM;
 
 	cres->hash.key = user_key | (res_type << 24);
-	ret = drm_ht_insert_item(&man->resources, &cres->hash);
+	ret = vmwgfx_ht_insert_item(&man->resources, &cres->hash);
 	if (unlikely(ret != 0)) {
 		kfree(cres);
 		goto out_invalid_key;

@@ -244,10 +244,10 @@ int vmw_cmdbuf_res_remove(struct vmw_cmdbuf_res_manager *man,
 			  struct vmw_resource **res_p)
 {
 	struct vmw_cmdbuf_res *entry;
-	struct drm_hash_item *hash;
+	struct vmwgfx_hash_item *hash;
 	int ret;
 
-	ret = drm_ht_find_item(&man->resources, user_key | (res_type << 24),
+	ret = vmwgfx_ht_find_item(&man->resources, user_key | (res_type << 24),
 			       &hash);
 	if (likely(ret != 0))
 		return -EINVAL;

@@ -260,7 +260,7 @@ int vmw_cmdbuf_res_remove(struct vmw_cmdbuf_res_manager *man,
 		*res_p = NULL;
 		break;
 	case VMW_CMDBUF_RES_COMMITTED:
-		(void) drm_ht_remove_item(&man->resources, &entry->hash);
+		(void) vmwgfx_ht_remove_item(&man->resources, &entry->hash);
 		list_del(&entry->head);
 		entry->state = VMW_CMDBUF_RES_DEL;
 		list_add_tail(&entry->head, list);

@@ -295,7 +295,7 @@ vmw_cmdbuf_res_man_create(struct vmw_private *dev_priv)
 
 	man->dev_priv = dev_priv;
 	INIT_LIST_HEAD(&man->list);
-	ret = drm_ht_create(&man->resources, VMW_CMDBUF_RES_MAN_HT_ORDER);
+	ret = vmwgfx_ht_create(&man->resources, VMW_CMDBUF_RES_MAN_HT_ORDER);
 	if (ret == 0)
 		return man;
 

@@ -320,7 +320,7 @@ void vmw_cmdbuf_res_man_destroy(struct vmw_cmdbuf_res_manager *man)
 	list_for_each_entry_safe(entry, next, &man->list, head)
 		vmw_cmdbuf_res_free(man, entry);
 
-	drm_ht_remove(&man->resources);
+	vmwgfx_ht_remove(&man->resources);
 	kfree(man);
 }
 

index 0f09a9116b054..4fea95a650418 100644
@@ -1149,7 +1149,7 @@ static void vmw_driver_unload(struct drm_device *dev)
 	unregister_pm_notifier(&dev_priv->pm_nb);
 
 	if (dev_priv->ctx.res_ht_initialized)
-		drm_ht_remove(&dev_priv->ctx.res_ht);
+		vmwgfx_ht_remove(&dev_priv->ctx.res_ht);
 	vfree(dev_priv->ctx.cmd_bounce);
 	if (dev_priv->enable_fb) {
 		vmw_fb_off(dev_priv);

index 9c60bb2aefe1f..1099cb5e25006 100644
@@ -34,7 +34,6 @@
 #include <drm/drm_auth.h>
 #include <drm/drm_device.h>
 #include <drm/drm_file.h>
-#include <drm/drm_hashtab.h>
 #include <drm/drm_rect.h>
 
 #include <drm/ttm/ttm_bo_driver.h>

@@ -43,6 +42,7 @@
 #include "ttm_object.h"
 
 #include "vmwgfx_fence.h"
+#include "vmwgfx_hashtab.h"
 #include "vmwgfx_reg.h"
 #include "vmwgfx_validation.h"
 

@@ -131,7 +131,7 @@ struct vmw_buffer_object {
  */
 struct vmw_validate_buffer {
 	struct ttm_validate_buffer base;
-	struct drm_hash_item hash;
+	struct vmwgfx_hash_item hash;
 	bool validate_as_mob;
 };
 

@@ -404,7 +404,7 @@ struct vmw_ctx_validation_info;
  * @ctx: The validation context
  */
 struct vmw_sw_context{
-	struct drm_open_hash res_ht;
+	struct vmwgfx_open_hash res_ht;
 	bool res_ht_initialized;
 	bool kernel;
 	struct vmw_fpriv *fp;

index 21134c7f18382..7dd42c5a7fd62 100644
@@ -4112,7 +4112,7 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 		vmw_binding_state_reset(sw_context->staged_bindings);
 
 	if (!sw_context->res_ht_initialized) {
-		ret = drm_ht_create(&sw_context->res_ht, VMW_RES_HT_ORDER);
+		ret = vmwgfx_ht_create(&sw_context->res_ht, VMW_RES_HT_ORDER);
 		if (unlikely(ret != 0))
 			goto out_unlock;
 

index b09094b50c5d0..41b7417cb5d3d 100644
@@ -43,7 +43,7 @@
  */
 struct vmw_validation_bo_node {
 	struct ttm_validate_buffer base;
-	struct drm_hash_item hash;
+	struct vmwgfx_hash_item hash;
 	unsigned int coherent_count;
 	u32 as_mob : 1;
 	u32 cpu_blit : 1;

@@ -72,7 +72,7 @@ struct vmw_validation_bo_node {
  */
 struct vmw_validation_res_node {
 	struct list_head head;
-	struct drm_hash_item hash;
+	struct vmwgfx_hash_item hash;
 	struct vmw_resource *res;
 	struct vmw_buffer_object *new_backup;
 	unsigned long new_backup_offset;

@@ -184,9 +184,9 @@ vmw_validation_find_bo_dup(struct vmw_validation_context *ctx,
 		return NULL;
 
 	if (ctx->ht) {
-		struct drm_hash_item *hash;
+		struct vmwgfx_hash_item *hash;
 
-		if (!drm_ht_find_item(ctx->ht, (unsigned long) vbo, &hash))
+		if (!vmwgfx_ht_find_item(ctx->ht, (unsigned long) vbo, &hash))
 			bo_node = container_of(hash, typeof(*bo_node), hash);
 	} else {
 		struct  vmw_validation_bo_node *entry;

@@ -221,9 +221,9 @@ vmw_validation_find_res_dup(struct vmw_validation_context *ctx,
 		return NULL;
 
 	if (ctx->ht) {
-		struct drm_hash_item *hash;
+		struct vmwgfx_hash_item *hash;
 
-		if (!drm_ht_find_item(ctx->ht, (unsigned long) res, &hash))
+		if (!vmwgfx_ht_find_item(ctx->ht, (unsigned long) res, &hash))
 			res_node = container_of(hash, typeof(*res_node), hash);
 	} else {
 		struct  vmw_validation_res_node *entry;

@@ -280,7 +280,7 @@ int vmw_validation_add_bo(struct vmw_validation_context *ctx,
 
 		if (ctx->ht) {
 			bo_node->hash.key = (unsigned long) vbo;
-			ret = drm_ht_insert_item(ctx->ht, &bo_node->hash);
+			ret = vmwgfx_ht_insert_item(ctx->ht, &bo_node->hash);
 			if (ret) {
 				DRM_ERROR("Failed to initialize a buffer "
 					  "validation entry.\n");

@@ -335,7 +335,7 @@ int vmw_validation_add_resource(struct vmw_validation_context *ctx,
 
 	if (ctx->ht) {
 		node->hash.key = (unsigned long) res;
-		ret = drm_ht_insert_item(ctx->ht, &node->hash);
+		ret = vmwgfx_ht_insert_item(ctx->ht, &node->hash);
 		if (ret) {
 			DRM_ERROR("Failed to initialize a resource validation "
 				  "entry.\n");

@@ -688,13 +688,13 @@ void vmw_validation_drop_ht(struct vmw_validation_context *ctx)
 		return;
 
 	list_for_each_entry(entry, &ctx->bo_list, base.head)
-		(void) drm_ht_remove_item(ctx->ht, &entry->hash);
+		(void) vmwgfx_ht_remove_item(ctx->ht, &entry->hash);
 
 	list_for_each_entry(val, &ctx->resource_list, head)
-		(void) drm_ht_remove_item(ctx->ht, &val->hash);
+		(void) vmwgfx_ht_remove_item(ctx->ht, &val->hash);
 
 	list_for_each_entry(val, &ctx->resource_ctx_list, head)
-		(void) drm_ht_remove_item(ctx->ht, &val->hash);
+		(void) vmwgfx_ht_remove_item(ctx->ht, &val->hash);
 
 	ctx->ht = NULL;
 }

index 739906d1b3ebb..495fd504b8c62 100644
@@ -31,9 +31,10 @@
 #include <linux/list.h>
 #include <linux/ww_mutex.h>
 
-#include <drm/drm_hashtab.h>
 #include <drm/ttm/ttm_execbuf_util.h>
 
+#include "vmwgfx_hashtab.h"
+
 #define VMW_RES_DIRTY_NONE 0
 #define VMW_RES_DIRTY_SET BIT(0)
 #define VMW_RES_DIRTY_CLEAR BIT(1)

@@ -73,7 +74,7 @@ struct vmw_validation_mem {
  * @total_mem: Amount of reserved memory.
  */
 struct vmw_validation_context {
-	struct drm_open_hash *ht;
+	struct vmwgfx_open_hash *ht;
 	struct list_head resource_list;
 	struct list_head resource_ctx_list;
 	struct list_head bo_list;

@@ -151,7 +152,7 @@ vmw_validation_set_val_mem(struct vmw_validation_context *ctx,
  * available at validation context declaration time
  */
 static inline void vmw_validation_set_ht(struct vmw_validation_context *ctx,
-					 struct drm_open_hash *ht)
+					 struct vmwgfx_open_hash *ht)
 {
 	ctx->ht = ht;
 }

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ian Forbes <[email protected]>

[ Upstream commit dfe1323ab3c8a4dd5625ebfdba44dc47df84512a ]

Nodes stored in the validation duplicates hashtable come from an arena allocator that is cleared at the end of vmw_execbuf_process. All nodes are expected to be cleared in vmw_validation_drop_ht but this node escaped because its resource was destroyed prematurely.

---
 drivers/gpu/drm/vmwgfx/vmwgfx_validation.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

index 41b7417cb5d3d..4633bd3081852 100644
@@ -343,8 +343,10 @@ int vmw_validation_add_resource(struct vmw_validation_context *ctx,
 		}
 	}
 	node->res = vmw_resource_reference_unless_doomed(res);
-	if (!node->res)
+	if (!node->res) {
+		hash_del_rcu(&node->hash.head);
 		return -ESRCH;
+	}
 
 	node->first_usage = 1;
 	if (!res->dev_priv->has_mob) {

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alexandr Sapozhnikov <[email protected]>

[ Upstream commit 2f3119686ef50319490ccaec81a575973da98815 ]

If new_asoc->peer.adaptation_ind=0 and sctp_ulpevent_make_authkey=0 and sctp_ulpevent_make_authkey() returns 0, then the variable ai_ev remains zero and the zero will be dereferenced
in the sctp_ulpevent_free() function.

---
 net/sctp/sm_statefuns.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

index b5f5ee233b59d..5a883bd722f5d 100644
@@ -880,7 +880,8 @@ enum sctp_disposition sctp_sf_do_5_1D_ce(struct net *net,
 	return SCTP_DISPOSITION_CONSUME;
 
 nomem_authev:
-	sctp_ulpevent_free(ai_ev);
+	if (ai_ev)
+		sctp_ulpevent_free(ai_ev);
 nomem_aiev:
 	sctp_ulpevent_free(ev);
 nomem_ev:

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit 2e7cbbbe3d61c63606994b7ff73c72537afe2e1c ]

syzbot reported the splat below in tcp_conn_request(). [0]

If a listener is close()d while a TFO socket is being processed in tcp_conn_request(), inet_csk_reqsk_queue_add() does not set reqsk->sk and calls inet_child_forget(), which calls tcp_disconnect() for the TFO socket.

After the cited commit, tcp_disconnect() calls reqsk_fastopen_remove(), where reqsk_put() is called due to !reqsk->sk.

Then, reqsk_fastopen_remove() in tcp_conn_request() decrements the last req->rsk_refcnt and frees reqsk, and __reqsk_free() at the drop_and_free label causes the refcount underflow for the listener and double-free of the reqsk.

Let's remove reqsk_fastopen_remove() in tcp_conn_request().

Note that other callers make sure tp->fastopen_rsk is not NULL.

[0]:
refcount_t: underflow; use-after-free.

WARNING: CPU: 12 PID: 5563 at lib/refcount.c:28 refcount_warn_saturate (lib/refcount.c:28)
Modules linked in:
CPU: 12 UID: 0 PID: 5563 Comm: syz-executor Not tainted syzkaller #0 PREEMPT(full)
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/12/2025
RIP: 0010:refcount_warn_saturate (lib/refcount.c:28)
Code: ab e8 8e b4 98 ff 0f 0b c3 cc cc cc cc cc 80 3d a4 e4 d6 01 00 75 9c c6 05 9b e4 d6 01 01 48 c7 c7 e8 df fb ab e8 6a b4 98 ff <0f> 0b e9 03 5b 76 00 cc 80 3d 7d e4 d6 01 00 0f 85 74 ff ff ff c6
RSP: 0018:ffffa79fc0304a98 EFLAGS: 00010246
RAX: d83af4db1c6b3900 RBX: ffff9f65c7a69020 RCX: d83af4db1c6b3900
RDX: 0000000000000000 RSI: 00000000ffff7fff RDI: ffffffffac78a280
RBP: 000000009d781b60 R08: 0000000000007fff R09: ffffffffac6ca280
R10: 0000000000017ffd R11: 0000000000000004 R12: ffff9f65c7b4f100
R13: ffff9f65c7d23c00 R14: ffff9f65c7d26000 R15: ffff9f65c7a64ef8
FS:  00007f9f962176c0(0000) GS:ffff9f65fcf00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000200000000180 CR3: 000000000dbbe006 CR4: 0000000000372ef0
Call Trace:
 <IRQ>
 tcp_conn_request (./include/linux/refcount.h:400 ./include/linux/refcount.h:432 ./include/linux/refcount.h:450 ./include/net/sock.h:1965 ./include/net/request_sock.h:131 net/ipv4/tcp_input.c:7301)
 tcp_rcv_state_process (net/ipv4/tcp_input.c:6708)
 tcp_v6_do_rcv (net/ipv6/tcp_ipv6.c:1670)
 tcp_v6_rcv (net/ipv6/tcp_ipv6.c:1906)
 ip6_protocol_deliver_rcu (net/ipv6/ip6_input.c:438)
 ip6_input (net/ipv6/ip6_input.c:500)
 ipv6_rcv (net/ipv6/ip6_input.c:311)
 __netif_receive_skb (net/core/dev.c:6104)
 process_backlog (net/core/dev.c:6456)
 __napi_poll (net/core/dev.c:7506)
 net_rx_action (net/core/dev.c:7569 net/core/dev.c:7696)
 handle_softirqs (kernel/softirq.c:579)
 do_softirq (kernel/softirq.c:480)
 </IRQ>

---
 net/ipv4/tcp_input.c | 1 -
 1 file changed, 1 deletion(-)

index fea019cc92d3c..15548dc3cc5c5 100644
@@ -7060,7 +7060,6 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
 				    &foc, TCP_SYNACK_FASTOPEN, skb);
 		/* Add the child socket directly into the accept queue */
 		if (!inet_csk_reqsk_queue_add(sk, req, fastopen_sk)) {
-			reqsk_fastopen_remove(fastopen_sk, req, false);
 			bh_unlock_sock(fastopen_sk);
 			sock_put(fastopen_sk);
 			goto drop_and_free;

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Erick Karanja <[email protected]>

[ Upstream commit 521405cb54cd2812bbb6dedd5afc14bca1e7e98a ]

Add missing of_node_put call to release device node tbi obtained via for_each_child_of_node.

---
 drivers/net/ethernet/freescale/fsl_pq_mdio.c | 2 ++
 1 file changed, 2 insertions(+)

index 9d58d83344670..ea49b0df397e5 100644
@@ -482,10 +482,12 @@ static int fsl_pq_mdio_probe(struct platform_device *pdev)
 					"missing 'reg' property in node %pOF\n",
 					tbi);
 				err = -EBUSY;
+				of_node_put(tbi);
 				goto error;
 			}
 			set_tbipa(*prop, pdev,
 				  data->get_tbipa, priv->map, &res);
+			of_node_put(tbi);
 		}
 	}
 

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Leo Yan <[email protected]>

[ Upstream commit 53d067feb8c4f16d1f24ce3f4df4450bb18c555f ]

The feature test programs are built without enabling '-Wall -Werror' options. As a result, a feature may appear to be available, but later building in perf can fail with stricter checks.

Make the feature test program use the same warning options as perf.

---
 tools/build/feature/Makefile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

index aa3b0d75e44b7..37aa85a81e0a0 100644
@@ -268,10 +268,10 @@ $(OUTPUT)test-libbabeltrace.bin:
 	$(BUILD) # -lbabeltrace provided by $(FEATURE_CHECK_LDFLAGS-libbabeltrace)
 
 $(OUTPUT)test-compile-32.bin:
-	$(CC) -m32 -o $@ test-compile.c
+	$(CC) -m32 -Wall -Werror -o $@ test-compile.c
 
 $(OUTPUT)test-compile-x32.bin:
-	$(CC) -mx32 -o $@ test-compile.c
+	$(CC) -mx32 -Wall -Werror -o $@ test-compile.c
 
 $(OUTPUT)test-zlib.bin:
 	$(BUILD) -lz

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Harini T <[email protected]>

[ Upstream commit 341867f730d3d3bb54491ee64e8b1a0c446656e7 ]

The controller is registered using the device-managed function 'devm_mbox_controller_register()'. As documented in mailbox.c, this ensures the devres framework automatically calls mbox_controller_unregister() when device_unregister() is invoked, making the explicit call unnecessary.

Remove redundant mbox_controller_unregister() call as device_unregister() handles controller cleanup.

---
 drivers/mailbox/zynqmp-ipi-mailbox.c | 1 -
 1 file changed, 1 deletion(-)

index be06de791c544..136c1f67dd223 100644
@@ -619,7 +619,6 @@ static void zynqmp_ipi_free_mboxes(struct zynqmp_ipi_pdata *pdata)
 	for (; i >= 0; i--) {
 		ipi_mbox = &pdata->ipi_mboxes[i];
 		if (ipi_mbox->dev.parent) {
-			mbox_controller_unregister(&ipi_mbox->mbox);
 			if (device_is_registered(&ipi_mbox->dev))
 				device_unregister(&ipi_mbox->dev);
 		}

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Harini T <[email protected]>

[ Upstream commit 019e3f4550fc7d319a7fd03eff487255f8e8aecd ]

The ipi_mbox->dev.parent check is unreliable proxy for registration status as it fails to protect against probe failures that occur after the parent is assigned but before device_register() completes.

device_is_registered() is the canonical and robust method to verify the registration status.

Remove ipi_mbox->dev.parent check in zynqmp_ipi_free_mboxes().

---
 drivers/mailbox/zynqmp-ipi-mailbox.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

index 136c1f67dd223..e64f7157f065f 100644
@@ -618,10 +618,8 @@ static void zynqmp_ipi_free_mboxes(struct zynqmp_ipi_pdata *pdata)
 	i = pdata->num_mboxes;
 	for (; i >= 0; i--) {
 		ipi_mbox = &pdata->ipi_mboxes[i];
-		if (ipi_mbox->dev.parent) {
-			if (device_is_registered(&ipi_mbox->dev))
-				device_unregister(&ipi_mbox->dev);
-		}
+		if (device_is_registered(&ipi_mbox->dev))
+			device_unregister(&ipi_mbox->dev);
 	}
 }
 

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Daniel Borkmann <[email protected]>

[ Upstream commit 23f3770e1a53e6c7a553135011f547209e141e72 ]

Cilium has a BPF egress gateway feature which forces outgoing K8s Pod traffic to pass through dedicated egress gateways which then SNAT the traffic in order to interact with stable IPs outside the cluster.

The traffic is directed to the gateway via vxlan tunnel in collect md mode. A recent BPF change utilized the bpf_redirect_neigh() helper to forward packets after the arrival and decap on vxlan, which turned out over time that the kmalloc-256 slab usage in kernel was ever-increasing.

The issue was that vxlan allocates the metadata_dst object and attaches it through a fake dst entry to the skb. The latter was never released though given bpf_redirect_neigh() was merely setting the new dst entry via skb_dst_set() without dropping an existing one first.

---
 net/core/filter.c | 2 ++
 1 file changed, 2 insertions(+)

index b95af925b9c27..1403829b96db9 100644
@@ -2257,6 +2257,7 @@ static int __bpf_redirect_neigh_v6(struct sk_buff *skb, struct net_device *dev,
 		if (IS_ERR(dst))
 			goto out_drop;
 
+		skb_dst_drop(skb);
 		skb_dst_set(skb, dst);
 	} else if (nh->nh_family != AF_INET6) {
 		goto out_drop;

@@ -2364,6 +2365,7 @@ static int __bpf_redirect_neigh_v4(struct sk_buff *skb, struct net_device *dev,
 			goto out_drop;
 		}
 
+		skb_dst_drop(skb);
 		skb_dst_set(skb, &rt->dst);
 	}
 

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alex Deucher <[email protected]>

[ Upstream commit 507296328b36ffd00ec1f4fde5b8acafb7222ec7 ]

---
 drivers/gpu/drm/amd/include/asic_reg/dce/dce_6_0_d.h       | 7 +++++++
 drivers/gpu/drm/amd/include/asic_reg/dce/dce_6_0_sh_mask.h | 2 ++
 2 files changed, 9 insertions(+)

index 9de01ae574c03..067eddd9c62d8 100644
@@ -4115,6 +4115,7 @@
 #define mmSCL0_SCL_COEF_RAM_CONFLICT_STATUS 0x1B55
 #define mmSCL0_SCL_COEF_RAM_SELECT 0x1B40
 #define mmSCL0_SCL_COEF_RAM_TAP_DATA 0x1B41
+#define mmSCL0_SCL_SCALER_ENABLE 0x1B42
 #define mmSCL0_SCL_CONTROL 0x1B44
 #define mmSCL0_SCL_DEBUG 0x1B6A
 #define mmSCL0_SCL_DEBUG2 0x1B69

@@ -4144,6 +4145,7 @@
 #define mmSCL1_SCL_COEF_RAM_CONFLICT_STATUS 0x1E55
 #define mmSCL1_SCL_COEF_RAM_SELECT 0x1E40
 #define mmSCL1_SCL_COEF_RAM_TAP_DATA 0x1E41
+#define mmSCL1_SCL_SCALER_ENABLE 0x1E42
 #define mmSCL1_SCL_CONTROL 0x1E44
 #define mmSCL1_SCL_DEBUG 0x1E6A
 #define mmSCL1_SCL_DEBUG2 0x1E69

@@ -4173,6 +4175,7 @@
 #define mmSCL2_SCL_COEF_RAM_CONFLICT_STATUS 0x4155
 #define mmSCL2_SCL_COEF_RAM_SELECT 0x4140
 #define mmSCL2_SCL_COEF_RAM_TAP_DATA 0x4141
+#define mmSCL2_SCL_SCALER_ENABLE 0x4142
 #define mmSCL2_SCL_CONTROL 0x4144
 #define mmSCL2_SCL_DEBUG 0x416A
 #define mmSCL2_SCL_DEBUG2 0x4169

@@ -4202,6 +4205,7 @@
 #define mmSCL3_SCL_COEF_RAM_CONFLICT_STATUS 0x4455
 #define mmSCL3_SCL_COEF_RAM_SELECT 0x4440
 #define mmSCL3_SCL_COEF_RAM_TAP_DATA 0x4441
+#define mmSCL3_SCL_SCALER_ENABLE 0x4442
 #define mmSCL3_SCL_CONTROL 0x4444
 #define mmSCL3_SCL_DEBUG 0x446A
 #define mmSCL3_SCL_DEBUG2 0x4469

@@ -4231,6 +4235,7 @@
 #define mmSCL4_SCL_COEF_RAM_CONFLICT_STATUS 0x4755
 #define mmSCL4_SCL_COEF_RAM_SELECT 0x4740
 #define mmSCL4_SCL_COEF_RAM_TAP_DATA 0x4741
+#define mmSCL4_SCL_SCALER_ENABLE 0x4742
 #define mmSCL4_SCL_CONTROL 0x4744
 #define mmSCL4_SCL_DEBUG 0x476A
 #define mmSCL4_SCL_DEBUG2 0x4769

@@ -4260,6 +4265,7 @@
 #define mmSCL5_SCL_COEF_RAM_CONFLICT_STATUS 0x4A55
 #define mmSCL5_SCL_COEF_RAM_SELECT 0x4A40
 #define mmSCL5_SCL_COEF_RAM_TAP_DATA 0x4A41
+#define mmSCL5_SCL_SCALER_ENABLE 0x4A42
 #define mmSCL5_SCL_CONTROL 0x4A44
 #define mmSCL5_SCL_DEBUG 0x4A6A
 #define mmSCL5_SCL_DEBUG2 0x4A69

@@ -4287,6 +4293,7 @@
 #define mmSCL_COEF_RAM_CONFLICT_STATUS 0x1B55
 #define mmSCL_COEF_RAM_SELECT 0x1B40
 #define mmSCL_COEF_RAM_TAP_DATA 0x1B41
+#define mmSCL_SCALER_ENABLE 0x1B42
 #define mmSCL_CONTROL 0x1B44
 #define mmSCL_DEBUG 0x1B6A
 #define mmSCL_DEBUG2 0x1B69

index 41c4a46ce3572..afe7303802c61 100644
@@ -8646,6 +8646,8 @@
 #define REGAMMA_LUT_INDEX__REGAMMA_LUT_INDEX__SHIFT 0x00000000
 #define REGAMMA_LUT_WRITE_EN_MASK__REGAMMA_LUT_WRITE_EN_MASK_MASK 0x00000007L
 #define REGAMMA_LUT_WRITE_EN_MASK__REGAMMA_LUT_WRITE_EN_MASK__SHIFT 0x00000000
+#define SCL_SCALER_ENABLE__SCL_SCALE_EN_MASK 0x00000001L
+#define SCL_SCALER_ENABLE__SCL_SCALE_EN__SHIFT 0x00000000
 #define SCL_ALU_CONTROL__SCL_ALU_DISABLE_MASK 0x00000001L
 #define SCL_ALU_CONTROL__SCL_ALU_DISABLE__SHIFT 0x00000000
 #define SCL_BYPASS_CONTROL__SCL_BYPASS_MODE_MASK 0x00000003L

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Timur Kristóf <[email protected]>

[ Upstream commit d60f9c45d1bff7e20ecd57492ef7a5e33c94a37c ]

Without these, it's impossible to program these registers.

---
 drivers/gpu/drm/amd/display/dc/dce/dce_transform.h | 2 ++
 1 file changed, 2 insertions(+)

index cbce194ec7b82..ff746fba850bc 100644
@@ -155,6 +155,8 @@
 	SRI(SCL_COEF_RAM_TAP_DATA, SCL, id), \
 	SRI(VIEWPORT_START, SCL, id), \
 	SRI(VIEWPORT_SIZE, SCL, id), \
+	SRI(SCL_HORZ_FILTER_INIT_RGB_LUMA, SCL, id), \
+	SRI(SCL_HORZ_FILTER_INIT_CHROMA, SCL, id), \
 	SRI(SCL_HORZ_FILTER_SCALE_RATIO, SCL, id), \
 	SRI(SCL_VERT_FILTER_SCALE_RATIO, SCL, id), \
 	SRI(SCL_VERT_FILTER_INIT, SCL, id), \

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Timur Kristóf <[email protected]>

[ Upstream commit c0aa7cf49dd6cb302fe28e7183992b772cb7420c ]

Previously, the code would set a bit field which didn't exist on DCE6 so it would be effectively a no-op.

---
 drivers/gpu/drm/amd/display/dc/dce/dce_transform.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

index 670d5ab9d9984..b761dda491d54 100644
@@ -527,8 +527,7 @@ static void dce60_transform_set_scaler(
 		if (coeffs_v != xfm_dce->filter_v || coeffs_h != xfm_dce->filter_h) {
 			/* 4. Program vertical filters */
 			if (xfm_dce->filter_v == NULL)
-				REG_SET(SCL_VERT_FILTER_CONTROL, 0,
-						SCL_V_2TAP_HARDCODE_COEF_EN, 0);
+				REG_WRITE(SCL_VERT_FILTER_CONTROL, 0);
 			program_multi_taps_filter(
 					xfm_dce,
 					data->taps.v_taps,

@@ -542,8 +541,7 @@ static void dce60_transform_set_scaler(
 
 			/* 5. Program horizontal filters */
 			if (xfm_dce->filter_h == NULL)
-				REG_SET(SCL_HORZ_FILTER_CONTROL, 0,
-						SCL_H_2TAP_HARDCODE_COEF_EN, 0);
+				REG_WRITE(SCL_HORZ_FILTER_CONTROL, 0);
 			program_multi_taps_filter(
 					xfm_dce,
 					data->taps.h_taps,

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Timur Kristóf <[email protected]>

[ Upstream commit a7dc87f3448bea5ebe054f14e861074b9c289c65 ]

SCL_SCALER_ENABLE can be used to enable/disable the scaler on DCE6. Program it to 0 when scaling isn't used, 1 when used. Additionally, clear some other registers when scaling is disabled and program the SCL_UPDATE register as recommended.

This fixes visible glitches for users whose BIOS sets up a mode with scaling at boot, which DC was unable to clean up.

---
 .../gpu/drm/amd/display/dc/dce/dce_transform.c    | 15 +++++++++++----
 .../gpu/drm/amd/display/dc/dce/dce_transform.h    |  2 ++
 2 files changed, 13 insertions(+), 4 deletions(-)

index b761dda491d54..f97c182677082 100644
@@ -154,10 +154,13 @@ static bool dce60_setup_scaling_configuration(
 	REG_SET(SCL_BYPASS_CONTROL, 0, SCL_BYPASS_MODE, 0);
 
 	if (data->taps.h_taps + data->taps.v_taps <= 2) {
-		/* Set bypass */
-
-		/* DCE6 has no SCL_MODE register, skip scale mode programming */
+		/* Disable scaler functionality */
+		REG_WRITE(SCL_SCALER_ENABLE, 0);
 
+		/* Clear registers that can cause glitches even when the scaler is off */
+		REG_WRITE(SCL_TAP_CONTROL, 0);
+		REG_WRITE(SCL_AUTOMATIC_MODE_CONTROL, 0);
+		REG_WRITE(SCL_F_SHARP_CONTROL, 0);
 		return false;
 	}
 

@@ -165,7 +168,7 @@ static bool dce60_setup_scaling_configuration(
 			SCL_H_NUM_OF_TAPS, data->taps.h_taps - 1,
 			SCL_V_NUM_OF_TAPS, data->taps.v_taps - 1);
 
-	/* DCE6 has no SCL_MODE register, skip scale mode programming */
+	REG_WRITE(SCL_SCALER_ENABLE, 1);
 
 	/* DCE6 has no SCL_BOUNDARY_MODE bit, skip replace out of bound pixels */
 

@@ -502,6 +505,8 @@ static void dce60_transform_set_scaler(
 	REG_SET(DC_LB_MEM_SIZE, 0,
 		DC_LB_MEM_SIZE, xfm_dce->lb_memory_size);
 
+	REG_WRITE(SCL_UPDATE, 0x00010000);
+
 	/* Clear SCL_F_SHARP_CONTROL value to 0 */
 	REG_WRITE(SCL_F_SHARP_CONTROL, 0);
 

@@ -564,6 +569,8 @@ static void dce60_transform_set_scaler(
 	/* DCE6 has no SCL_COEF_UPDATE_COMPLETE bit to flip to new coefficient memory */
 
 	/* DCE6 DATA_FORMAT register does not support ALPHA_EN */
+
+	REG_WRITE(SCL_UPDATE, 0);
 }
 #endif
 

index ff746fba850bc..eb716e8337e23 100644
@@ -155,6 +155,7 @@
 	SRI(SCL_COEF_RAM_TAP_DATA, SCL, id), \
 	SRI(VIEWPORT_START, SCL, id), \
 	SRI(VIEWPORT_SIZE, SCL, id), \
+	SRI(SCL_SCALER_ENABLE, SCL, id), \
 	SRI(SCL_HORZ_FILTER_INIT_RGB_LUMA, SCL, id), \
 	SRI(SCL_HORZ_FILTER_INIT_CHROMA, SCL, id), \
 	SRI(SCL_HORZ_FILTER_SCALE_RATIO, SCL, id), \

@@ -592,6 +593,7 @@ struct dce_transform_registers {
 	uint32_t SCL_VERT_FILTER_SCALE_RATIO;
 	uint32_t SCL_HORZ_FILTER_INIT;
 #if defined(CONFIG_DRM_AMD_DC_SI)
+	uint32_t SCL_SCALER_ENABLE;
 	uint32_t SCL_HORZ_FILTER_INIT_RGB_LUMA;
 	uint32_t SCL_HORZ_FILTER_INIT_CHROMA;
 #endif

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Woudstra <[email protected]>

[ Upstream commit bbf0c98b3ad9edaea1f982de6c199cc11d3b7705 ]

net/bridge/br_private.h:1627 suspicious rcu_dereference_protected() usage! other info that might help us debug this:

rcu_scheduler_active = 2, debug_locks = 1
7 locks held by socat/410:
#0: ffff88800d7a9c90 (sk_lock-AF_INET){+.+.}-{0:0}, at: inet_stream_connect+0x43/0xa0 #1: ffffffff9a779900 (rcu_read_lock){....}-{1:3}, at: __ip_queue_xmit+0x62/0x1830 [..]
#6: ffffffff9a779900 (rcu_read_lock){....}-{1:3}, at: nf_hook.constprop.0+0x8a/0x440

Call Trace:
 lockdep_rcu_suspicious.cold+0x4f/0xb1
 br_vlan_fill_forward_path_pvid+0x32c/0x410 [bridge]
 br_fill_forward_path+0x7a/0x4d0 [bridge]

Use to correct helper, non _rcu variant requires RTNL mutex.

---
 net/bridge/br_vlan.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 86441ff78a0f8..055d988d280cd 100644
@@ -1391,7 +1391,7 @@ void br_vlan_fill_forward_path_pvid(struct net_bridge *br,
 	if (!br_opt_get(br, BROPT_VLAN_ENABLED))
 		return;
 
-	vg = br_vlan_group(br);
+	vg = br_vlan_group_rcu(br);
 
 	if (idx >= 0 &&
 	    ctx->vlan[idx].proto == br->vlan_proto) {

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Herbert Xu <[email protected]>

[ Upstream commit 6bb73db6948c2de23e407fe1b7ef94bf02b7529f ]

Move the ssize check to the start in essiv_aead_crypt so that it's also checked for decryption and in-place encryption.

---
 crypto/essiv.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

index 3505b071e6471..365f3082ea041 100644
@@ -186,9 +186,14 @@ static int essiv_aead_crypt(struct aead_request *req, bool enc)
 	const struct essiv_tfm_ctx *tctx = crypto_aead_ctx(tfm);
 	struct essiv_aead_request_ctx *rctx = aead_request_ctx(req);
 	struct aead_request *subreq = &rctx->aead_req;
+	int ivsize = crypto_aead_ivsize(tfm);
+	int ssize = req->assoclen - ivsize;
 	struct scatterlist *src = req->src;
 	int err;
 
+	if (ssize < 0)
+		return -EINVAL;
+
 	crypto_cipher_encrypt_one(tctx->essiv_cipher, req->iv, req->iv);
 
 	/*

@@ -198,19 +203,12 @@ static int essiv_aead_crypt(struct aead_request *req, bool enc)
 	 */
 	rctx->assoc = NULL;
 	if (req->src == req->dst || !enc) {
-		scatterwalk_map_and_copy(req->iv, req->dst,
-					 req->assoclen - crypto_aead_ivsize(tfm),
-					 crypto_aead_ivsize(tfm), 1);
+		scatterwalk_map_and_copy(req->iv, req->dst, ssize, ivsize, 1);
 	} else {
 		u8 *iv = (u8 *)aead_request_ctx(req) + tctx->ivoffset;
-		int ivsize = crypto_aead_ivsize(tfm);
-		int ssize = req->assoclen - ivsize;
 		struct scatterlist *sg;
 		int nents;
 
-		if (ssize < 0)
-			return -EINVAL;
-
 		nents = sg_nents_for_len(req->src, ssize);
 		if (nents < 0)
 			return -EINVAL;

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Gunnar Kudrjavets <[email protected]>

[ Upstream commit 8a81236f2cb0882c7ea6c621ce357f7f3f601fe5 ]

The tpm_tis_write8() call specifies arguments in wrong order. Should be (data, addr, value) not (data, value, addr). The initial correct order was changed during the major refactoring when the code was split.

---
 drivers/char/tpm/tpm_tis_core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

index b3452259d6e0b..c8c68301543b2 100644
@@ -831,8 +831,8 @@ static int tpm_tis_probe_irq_single(struct tpm_chip *chip, u32 intmask,
 	 * will call disable_irq which undoes all of the above.
 	 */
 	if (!(chip->flags & TPM_CHIP_FLAG_IRQ)) {
-		tpm_tis_write8(priv, original_int_vec,
-			       TPM_INT_VECTOR(priv->locality));
+		tpm_tis_write8(priv, TPM_INT_VECTOR(priv->locality),
+			       original_int_vec);
 		rc = -1;
 	}
 

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andy Shevchenko <[email protected]>

[ Upstream commit a060dc6620c13435b78e92cd2ebdbb6d11af237a ]

The of_gpio_n_cells default is 2 when ->of_xlate() callback is not defined. No need to assign it explicitly in the driver.

Stable-dep-of: b5f8aa8d4bde ("gpio: wcd934x: mark the GPIO controller as sleeping")

---
 drivers/gpio/gpio-wcd934x.c | 1 -
 1 file changed, 1 deletion(-)

index c00968ce7a569..cbbbd105a5a7b 100644
@@ -101,7 +101,6 @@ static int wcd_gpio_probe(struct platform_device *pdev)
 	chip->base = -1;
 	chip->ngpio = WCD934X_NPINS;
 	chip->label = dev_name(dev);
-	chip->of_gpio_n_cells = 2;
 	chip->can_sleep = false;
 
 	return devm_gpiochip_add_data(dev, chip, data);

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Bartosz Golaszewski <[email protected]>

[ Upstream commit b5f8aa8d4bde0cf3e4595af5a536da337e5f1c78 ]

The slimbus regmap passed to the GPIO driver down from MFD does not use fast_io. This means a mutex is used for locking and thus this GPIO chip must not be used in atomic context. Change the can_sleep switch in struct gpio_chip to true.

---
 drivers/gpio/gpio-wcd934x.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index cbbbd105a5a7b..26d70ac90933c 100644
@@ -101,7 +101,7 @@ static int wcd_gpio_probe(struct platform_device *pdev)
 	chip->base = -1;
 	chip->ngpio = WCD934X_NPINS;
 	chip->label = dev_name(dev);
-	chip->can_sleep = false;
+	chip->can_sleep = true;
 
 	return devm_gpiochip_add_data(dev, chip, data);
 }

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: KaFai Wan <[email protected]>

[ Upstream commit 4f375ade6aa9f37fd72d7a78682f639772089eed ]

When unpinning a BPF hash table (htab or htab_lru) that contains internal structures (timer, workqueue, or task_work) in its values, a BUG warning is triggered:

 BUG: sleeping function called from invalid context at kernel/bpf/hashtab.c:244
 in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 14, name: ksoftirqd/0
 ...

The issue arises from the interaction between BPF object unpinning and RCU callback mechanisms:
1. BPF object unpinning uses ->free_inode() which schedules cleanup via call_rcu(), deferring the actual freeing to an RCU callback that executes within the RCU_SOFTIRQ context.
2. During cleanup of hash tables containing internal structures, htab_map_free_internal_structs() is invoked, which includes cond_resched() or cond_resched_rcu() calls to yield the CPU during potentially long operations.

However, cond_resched() or cond_resched_rcu() cannot be safely called from atomic RCU softirq context, leading to the BUG warning when attempting to reschedule.

Fix this by changing from ->free_inode() to ->destroy_inode() and rename bpf_free_inode() to bpf_destroy_inode() for BPF objects (prog, map, link). This allows direct inode freeing without RCU callback scheduling, avoiding the invalid context warning.

---
 kernel/bpf/inode.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

index 5a8d9f7467bf4..849df8268af57 100644
@@ -610,7 +610,7 @@ static int bpf_show_options(struct seq_file *m, struct dentry *root)
 	return 0;
 }
 
-static void bpf_free_inode(struct inode *inode)
+static void bpf_destroy_inode(struct inode *inode)
 {
 	enum bpf_type type;
 

@@ -625,7 +625,7 @@ static const struct super_operations bpf_super_ops = {
 	.statfs		= simple_statfs,
 	.drop_inode	= generic_delete_inode,
 	.show_options	= bpf_show_options,
-	.free_inode	= bpf_free_inode,
+	.destroy_inode	= bpf_destroy_inode,
 };
 
 enum {

2.51.0

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Daniel Tang <[email protected]>

commit 4aac453deca0d9c61df18d968f8864c3ae7d3d8d upstream.

Previously, after rmmod acpi_tad, modprobe acpi_tad would fail with this dmesg:

sysfs: cannot create duplicate filename '/devices/platform/ACPI000E:00/time'

Call Trace:
 <TASK>
 dump_stack_lvl+0x6c/0x90
 dump_stack+0x10/0x20
 sysfs_warn_dup+0x8b/0xa0
 sysfs_add_file_mode_ns+0x122/0x130
 internal_create_group+0x1dd/0x4c0
 sysfs_create_group+0x13/0x20
 acpi_tad_probe+0x147/0x1f0 [acpi_tad]
 platform_probe+0x42/0xb0
 </TASK>

acpi-tad ACPI000E:00: probe with driver acpi-tad failed with error -17

---
 drivers/acpi/acpi_tad.c |    3 +++
 1 file changed, 3 insertions(+)

@@ -563,6 +563,9 @@ static int acpi_tad_remove(struct platfo
 
 	pm_runtime_get_sync(dev);
 
+	if (dd->capabilities & ACPI_TAD_RT)
+		sysfs_remove_group(&dev->kobj, &acpi_tad_time_attr_group);
+
 	if (dd->capabilities & ACPI_TAD_DC_WAKE)
 		sysfs_remove_group(&dev->kobj, &acpi_tad_dc_attr_group);
 

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Amir Mohammad Jahangirzad <[email protected]>

commit 496f9372eae14775e0524e83e952814691fe850a upstream.

In the ACPI debugger interface, the helper functions for read and write operations use "int" as the length parameter data type. When a large "size_t count" is passed from the file operations, this cast to "int" results in truncation and a negative value due to signed integer representation.

Logically, this negative number propagates to the min() calculation, where it is selected over the positive buffer space value, leading to unexpected behavior. Subsequently, when this negative value is used in copy_to_user() or copy_from_user(), it is interpreted as a large positive value due to the unsigned nature of the size parameter in these functions, causing the copy operations to attempt handling sizes far beyond the intended buffer limits.

Address the issue by:
- Changing the length parameters in acpi_aml_read_user() and acpi_aml_write_user() from "int" to "size_t", aligning with the expected unsigned size semantics.
- Updating return types and local variables in acpi_aml_read() and acpi_aml_write() to "ssize_t" for consistency with kernel file operation conventions.
- Using "size_t" for the "n" variable to ensure calculations remain unsigned.
- Using min_t() for circ_count_to_end() and circ_space_to_end() to ensure type-safe comparisons and prevent integer overflow.

---
 drivers/acpi/acpi_dbg.c |   26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

@@ -569,11 +569,11 @@ static int acpi_aml_release(struct inode
 	return 0;
 }
 
-static int acpi_aml_read_user(char __user *buf, int len)
+static ssize_t acpi_aml_read_user(char __user *buf, size_t len)
 {
-	int ret;
 	struct circ_buf *crc = &acpi_aml_io.out_crc;
-	int n;
+	ssize_t ret;
+	size_t n;
 	char *p;
 
 	ret = acpi_aml_lock_read(crc, ACPI_AML_OUT_USER);

@@ -582,7 +582,7 @@ static int acpi_aml_read_user(char __use
 	/* sync head before removing logs */
 	smp_rmb();
 	p = &crc->buf[crc->tail];
-	n = min(len, circ_count_to_end(crc));
+	n = min_t(size_t, len, circ_count_to_end(crc));
 	if (copy_to_user(buf, p, n)) {
 		ret = -EFAULT;
 		goto out;

@@ -599,8 +599,8 @@ out:
 static ssize_t acpi_aml_read(struct file *file, char __user *buf,
 			     size_t count, loff_t *ppos)
 {
-	int ret = 0;
-	int size = 0;
+	ssize_t ret = 0;
+	ssize_t size = 0;
 
 	if (!count)
 		return 0;

@@ -639,11 +639,11 @@ again:
 	return size > 0 ? size : ret;
 }
 
-static int acpi_aml_write_user(const char __user *buf, int len)
+static ssize_t acpi_aml_write_user(const char __user *buf, size_t len)
 {
-	int ret;
 	struct circ_buf *crc = &acpi_aml_io.in_crc;
-	int n;
+	ssize_t ret;
+	size_t n;
 	char *p;
 
 	ret = acpi_aml_lock_write(crc, ACPI_AML_IN_USER);

@@ -652,7 +652,7 @@ static int acpi_aml_write_user(const cha
 	/* sync tail before inserting cmds */
 	smp_mb();
 	p = &crc->buf[crc->head];
-	n = min(len, circ_space_to_end(crc));
+	n = min_t(size_t, len, circ_space_to_end(crc));
 	if (copy_from_user(p, buf, n)) {
 		ret = -EFAULT;
 		goto out;

@@ -663,14 +663,14 @@ static int acpi_aml_write_user(const cha
 	ret = n;
 out:
 	acpi_aml_unlock_fifo(ACPI_AML_IN_USER, ret >= 0);
-	return n;
+	return ret;
 }
 
 static ssize_t acpi_aml_write(struct file *file, const char __user *buf,
 			      size_t count, loff_t *ppos)
 {
-	int ret = 0;
-	int size = 0;
+	ssize_t ret = 0;
+	ssize_t size = 0;
 
 	if (!count)
 		return 0;

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Stephan Gerhold <[email protected]>

commit 99b78773c2ae55dcc01025f94eae8ce9700ae985 upstream.

On most MSM8916 devices (aside from the DragonBoard 410c), the bootloader already initializes the display to show the boot splash screen. In this situation, MDSS is already configured and left running when starting Linux. To avoid side effects from the bootloader configuration, the MDSS reset can be specified in the device tree to start again with a clean hardware state.

The reset for MDSS is currently missing in msm8916.dtsi, which causes errors when the MDSS driver tries to re-initialize the registers:

dsi_err_worker: status=6
dsi_err_worker: status=6
dsi_err_worker: status=6
...

It turns out that we have always indirectly worked around this by building the MDSS driver as a module. Before v6.17, the power domain was temporarily turned off until the module was loaded, long enough to clear the register contents. In v6.17, power domains are not turned off during boot until sync_state() happens, so this is no longer working. Even before v6.17 this resulted in broken behavior, but notably only when the MDSS driver was built-in instead of a module.

---
 arch/arm64/boot/dts/qcom/msm8916.dtsi |    2 ++
 1 file changed, 2 insertions(+)

@@ -957,6 +957,8 @@
 
 			interrupts = <GIC_SPI 72 IRQ_TYPE_LEVEL_HIGH>;
 
+			resets = <&gcc GCC_MDSS_BCR>;
+
 			interrupt-controller;
 			#interrupt-cells = <1>;
 

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Miaoqian Lin <[email protected]>

commit 74139a64e8cedb6d971c78d5d17384efeced1725 upstream.

Add missing of_node_put() calls to release device node references obtained via of_parse_phandle().

---
 arch/arm/mach-omap2/pm33xx-core.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

@@ -393,12 +393,15 @@ static int __init amx3_idle_init(struct
 		if (!state_node)
 			break;
 
-		if (!of_device_is_available(state_node))
+		if (!of_device_is_available(state_node)) {
+			of_node_put(state_node);
 			continue;
+		}
 
 		if (i == CPUIDLE_STATE_MAX) {
 			pr_warn("%s: cpuidle states reached max possible\n",
 				__func__);
+			of_node_put(state_node);
 			break;
 		}
 

@@ -408,6 +411,7 @@ static int __init amx3_idle_init(struct
 			states[state_count].wfi_flags |= WFI_FLAG_WAKE_M3 |
 							 WFI_FLAG_FLUSH_CACHE;
 
+		of_node_put(state_node);
 		state_count++;
 	}
 

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jason Andryuk <[email protected]>

commit 08df2d7dd4ab2db8a172d824cda7872d5eca460a upstream.

rc is overwritten by the evtchn_status hypercall in each iteration, so the return value will be whatever the last iteration is. This could incorrectly return success even if the event channel was not found. Change to an explicit -ENOENT for an un-found virq and return 0 on a successful match.

Message-ID: <[email protected]>

---
 drivers/xen/events/events_base.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

@@ -1331,10 +1331,11 @@ static int find_virq(unsigned int virq,
 {
 	struct evtchn_status status;
 	evtchn_port_t port;
-	int rc = -ENOENT;
 
 	memset(&status, 0, sizeof(status));
 	for (port = 0; port < xen_evtchn_max_channels(); port++) {
+		int rc;
+
 		status.dom = DOMID_SELF;
 		status.port = port;
 		rc = HYPERVISOR_event_channel_op(EVTCHNOP_status, &status);

@@ -1344,10 +1345,10 @@ static int find_virq(unsigned int virq,
 			continue;
 		if (status.u.virq == virq && status.vcpu == xen_vcpu_nr(cpu)) {
 			*evtchn = port;
-			break;
+			return 0;
 		}
 	}
-	return rc;
+	return -ENOENT;
 }
 
 /**

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Lukas Wunner <[email protected]>

commit f770c3d858687252f1270265ba152d5c622e793f upstream.

The device power management API has the following asymmetry:
* dpm_suspend_start() does not clean up on failure (it requires a call to dpm_resume_end())
* dpm_suspend_end() does clean up on failure
(it does not require a call to dpm_resume_start())

The asymmetry was introduced by commit d8f3de0d2412 ("Suspend-related patches for 2.6.27") in June 2008: It removed a call to device_resume() from device_suspend() (which was later renamed to dpm_suspend_start()).

When Xen began using the device power management API in May 2008 with commit 0e91398f2a5d ("xen: implement save/restore"), the asymmetry did not yet exist. But since it was introduced, a call to dpm_resume_end() is missing in the error path of dpm_suspend_start(). Fix it.

Message-ID: <22453676d1ddcebbe81641bb68ddf587fee7e21e.1756990799.git.lukas@wunner.de>

---
 drivers/xen/manage.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -116,7 +116,7 @@ static void do_suspend(void)
 	err = dpm_suspend_start(PMSG_FREEZE);
 	if (err) {
 		pr_err("%s: dpm_suspend_start %d\n", __func__, err);
-		goto out_thaw;
+		goto out_resume_end;
 	}
 
 	printk(KERN_DEBUG "suspending xenstore...\n");

@@ -156,6 +156,7 @@ out_resume:
 	else
 		xs_suspend_cancel();
 
+out_resume_end:
 	dpm_resume_end(si.cancelled ? PMSG_THAW : PMSG_RESTORE);
 
 out_thaw:

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johan Hovold <[email protected]>

commit 8ece3173f87df03935906d0c612c2aeda9db92ca upstream.

Make sure to drop the reference to the secure monitor device taken by of_find_device_by_node() when looking up its driver data on behalf of other drivers (e.g. during probe).

Note that holding a reference to the platform device does not prevent its driver data from going away so there is no point in keeping the reference after the helper returns.

---
 drivers/firmware/meson/meson_sm.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

@@ -225,11 +225,16 @@ EXPORT_SYMBOL(meson_sm_call_write);
 struct meson_sm_firmware *meson_sm_get(struct device_node *sm_node)
 {
 	struct platform_device *pdev = of_find_device_by_node(sm_node);
+	struct meson_sm_firmware *fw;
 
 	if (!pdev)
 		return NULL;
 
-	return platform_get_drvdata(pdev);
+	fw = platform_get_drvdata(pdev);
+
+	put_device(&pdev->dev);
+
+	return fw;
 }
 EXPORT_SYMBOL_GPL(meson_sm_get);
 

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Qianfeng Rong <[email protected]>

commit bacd713145443dce7764bb2967d30832a95e5ec8 upstream.

Change "ret" from unsigned int to int type in mt9v111_calc_frame_rate() to store negative error codes or zero returned by __mt9v111_hw_reset() and other functions.

Storing the negative error codes in unsigned type, doesn't cause an issue at runtime but it's ugly as pants.

No effect on runtime.

---
 drivers/media/i2c/mt9v111.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -534,8 +534,8 @@ static int mt9v111_calc_frame_rate(struc
 static int mt9v111_hw_config(struct mt9v111_dev *mt9v111)
 {
 	struct i2c_client *c = mt9v111->client;
-	unsigned int ret;
 	u16 outfmtctrl2;
+	int ret;
 
 	/* Force device reset. */
 	ret = __mt9v111_hw_reset(mt9v111);

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Shuhao Fu <[email protected]>

commit e4bea919584ff292c9156cf7d641a2ab3cbe27b0 upstream.

In nouveau_bo_move_prep, if nouveau_mem_map fails, an error code should be returned. Currently, it returns zero even if vmm addr is not correctly mapped.

---
 drivers/gpu/drm/nouveau/nouveau_bo.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -791,7 +791,7 @@ done:
 		nvif_vmm_put(vmm, &old_mem->vma[1]);
 		nvif_vmm_put(vmm, &old_mem->vma[0]);
 	}
-	return 0;
+	return ret;
 }
 
 static int

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Anderson Nascimento <[email protected]>

commit dff4f9ff5d7f289e4545cc936362e01ed3252742 upstream.

The function btrfs_encode_fh() does not properly account for the three cases it handles.

Before writing to the file handle (fh), the function only returns to the user BTRFS_FID_SIZE_NON_CONNECTABLE (5 dwords, 20 bytes) or BTRFS_FID_SIZE_CONNECTABLE (8 dwords, 32 bytes).

However, when a parent exists and the root ID of the parent and the inode are different, the function writes BTRFS_FID_SIZE_CONNECTABLE_ROOT (10 dwords, 40 bytes).

If *max_len is not large enough, this write goes out of bounds because BTRFS_FID_SIZE_CONNECTABLE_ROOT is greater than BTRFS_FID_SIZE_CONNECTABLE originally returned.

This results in an 8-byte out-of-bounds write at
fid->parent_root_objectid = parent_root_id.

A previous attempt to fix this issue was made but was lost.

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

Although this issue does not seem to be easily triggerable, it is a potential memory corruption bug that should be fixed. This patch resolves the issue by ensuring the function returns the appropriate size for all three cases and validates that *max_len is large enough before writing any data.

---
 fs/btrfs/export.c |    8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

@@ -22,7 +22,11 @@ static int btrfs_encode_fh(struct inode
 	int type;
 
 	if (parent && (len < BTRFS_FID_SIZE_CONNECTABLE)) {
-		*max_len = BTRFS_FID_SIZE_CONNECTABLE;
+		if (btrfs_root_id(BTRFS_I(inode)->root) !=
+		    btrfs_root_id(BTRFS_I(parent)->root))
+			*max_len = BTRFS_FID_SIZE_CONNECTABLE_ROOT;
+		else
+			*max_len = BTRFS_FID_SIZE_CONNECTABLE;
 		return FILEID_INVALID;
 	} else if (len < BTRFS_FID_SIZE_NON_CONNECTABLE) {
 		*max_len = BTRFS_FID_SIZE_NON_CONNECTABLE;

@@ -44,6 +48,8 @@ static int btrfs_encode_fh(struct inode
 		parent_root_id = BTRFS_I(parent)->root->root_key.objectid;
 
 		if (parent_root_id != fid->root_objectid) {
+			if (*max_len < BTRFS_FID_SIZE_CONNECTABLE_ROOT)
+				return FILEID_INVALID;
 			fid->parent_root_objectid = parent_root_id;
 			len = BTRFS_FID_SIZE_CONNECTABLE_ROOT;
 			type = FILEID_BTRFS_WITH_PARENT_ROOT;

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Adam Xue <[email protected]>

commit d0856a6dff57f95cc5d2d74e50880f01697d0cc4 upstream.

In mhi_init_irq_setup, the device pointer used for dev_err() was not initialized. Use the pointer from mhi_cntrl instead.

---
 drivers/bus/mhi/host/init.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

@@ -161,7 +161,6 @@ void mhi_deinit_free_irq(struct mhi_cont
 int mhi_init_irq_setup(struct mhi_controller *mhi_cntrl)
 {
 	struct mhi_event *mhi_event = mhi_cntrl->mhi_event;
-	struct device *dev = &mhi_cntrl->mhi_dev->dev;
 	unsigned long irq_flags = IRQF_SHARED | IRQF_NO_SUSPEND;
 	int i, ret;
 

@@ -182,7 +181,7 @@ int mhi_init_irq_setup(struct mhi_contro
 			continue;
 
 		if (mhi_event->irq >= mhi_cntrl->nr_irqs) {
-			dev_err(dev, "irq %d not available for event ring\n",
+			dev_err(mhi_cntrl->cntrl_dev, "irq %d not available for event ring\n",
 				mhi_event->irq);
 			ret = -EINVAL;
 			goto error_request;

@@ -193,7 +192,7 @@ int mhi_init_irq_setup(struct mhi_contro
 				  irq_flags,
 				  "mhi", mhi_event);
 		if (ret) {
-			dev_err(dev, "Error requesting irq:%d for ev:%d\n",
+			dev_err(mhi_cntrl->cntrl_dev, "Error requesting irq:%d for ev:%d\n",
 				mhi_cntrl->irq[mhi_event->irq], i);
 			goto error_request;
 		}

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Simon Schuster <[email protected]>

commit 04ff48239f46e8b493571e260bd0e6c3a6400371 upstream.

With the introduction of clone3 in commit 7f192e3cd316 ("fork: add clone3") the effective bit width of clone_flags on all architectures was increased from 32-bit to 64-bit. However, the signature of the copy_* helper functions (e.g., copy_sighand) used by copy_process was not adapted.

As such, they truncate the flags on any 32-bit architectures that supports clone3 (arc, arm, csky, m68k, microblaze, mips32, openrisc, parisc32, powerpc32, riscv32, x86-32 and xtensa).

For copy_sighand with CLONE_CLEAR_SIGHAND being an actual u64 constant, this triggers an observable bug in kernel selftest clone3_clear_sighand:

if (clone_flags & CLONE_CLEAR_SIGHAND)

in function copy_sighand within fork.c will always fail given:

        unsigned long /* == uint32_t */ clone_flags
        #define CLONE_CLEAR_SIGHAND 0x100000000ULL

This commit fixes the bug by always passing clone_flags to copy_sighand via their declared u64 type, invariant of architecture-dependent integer sizes.

---
 kernel/fork.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -1595,7 +1595,7 @@ static int copy_io(unsigned long clone_f
 	return 0;
 }
 
-static int copy_sighand(unsigned long clone_flags, struct task_struct *tsk)
+static int copy_sighand(u64 clone_flags, struct task_struct *tsk)
 {
 	struct sighand_struct *sig;
 

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafael J. Wysocki <[email protected]>

commit 69e5d50fcf4093fb3f9f41c4f931f12c2ca8c467 upstream.

The cpufreq_cpu_put() call in update_qos_request() takes place too early because the latter subsequently calls freq_qos_update_request() that indirectly accesses the policy object in question through the QoS request object passed to it.

Fortunately, update_qos_request() is called under intel_pstate_driver_lock, so this issue does not matter for changing the intel_pstate operation mode, but it theoretically can cause a crash to occur on CPU device hot removal (which currently can only happen in virt, but it is formally supported nevertheless).

Address this issue by modifying update_qos_request() to drop the reference to the policy later.

---
 drivers/cpufreq/intel_pstate.c |    8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

@@ -1307,10 +1307,10 @@ static void update_qos_request(enum freq
 			continue;
 
 		req = policy->driver_data;
-		cpufreq_cpu_put(policy);
-
-		if (!req)
+		if (!req) {
+			cpufreq_cpu_put(policy);
 			continue;
+		}
 
 		if (hwp_active)
 			intel_pstate_get_hwp_cap(cpu);

@@ -1326,6 +1326,8 @@ static void update_qos_request(enum freq
 
 		if (freq_qos_update_request(req, freq) < 0)
 			pr_warn("Failed to update freq constraint: CPU%d\n", i);
+
+		cpufreq_cpu_put(policy);
 	}
 }
 

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Thomas Fourier <[email protected]>

commit f5d643156ef62216955c119216d2f3815bd51cb1 upstream.

It seems like everywhere in this file, dd->in_sg is mapped with DMA_TO_DEVICE and dd->out_sg is mapped with DMA_FROM_DEVICE.

---
 drivers/crypto/atmel-tdes.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -548,7 +548,7 @@ static int atmel_tdes_crypt_start(struct
 
 	if (err && (dd->flags & TDES_FLAGS_FAST)) {
 		dma_unmap_sg(dd->dev, dd->in_sg, 1, DMA_TO_DEVICE);
-		dma_unmap_sg(dd->dev, dd->out_sg, 1, DMA_TO_DEVICE);
+		dma_unmap_sg(dd->dev, dd->out_sg, 1, DMA_FROM_DEVICE);
 	}
 
 	return err;

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Haoxiang Li <[email protected]>

commit d68318471aa2e16222ebf492883e05a2d72b9b17 upstream.

Add put_bh() to decrease the refcount of 'bh' after the job is finished, preventing a resource leak.

---
 fs/ntfs3/bitmap.c |    1 +
 1 file changed, 1 insertion(+)

@@ -1381,6 +1381,7 @@ int wnd_extend(struct wnd_bitmap *wnd, s
 		mark_buffer_dirty(bh);
 		unlock_buffer(bh);
 		/* err = sync_dirty_buffer(bh); */
+		put_bh(bh);
 
 		b0 = 0;
 		bits -= op;

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Qianfeng Rong <[email protected]>

commit f9381ece76de999a2065d5b4fdd87fa17883978c upstream.

Change the 'ret' variable in ad5360_update_ctrl() from unsigned int to int, as it needs to store either negative error codes or zero returned by ad5360_write_unlocked().

---
 drivers/iio/dac/ad5360.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -262,7 +262,7 @@ static int ad5360_update_ctrl(struct iio
 	unsigned int clr)
 {
 	struct ad5360_state *st = iio_priv(indio_dev);
-	unsigned int ret;
+	int ret;
 
 	mutex_lock(&st->lock);
 

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Qianfeng Rong <[email protected]>

commit 3379c900320954d768ed9903691fb2520926bbe3 upstream.

Change the 'ret' variable in ad5421_update_ctrl() from unsigned int to int, as it needs to store either negative error codes or zero returned by ad5421_write_unlocked().

---
 drivers/iio/dac/ad5421.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -186,7 +186,7 @@ static int ad5421_update_ctrl(struct iio
 	unsigned int clr)
 {
 	struct ad5421_state *st = iio_priv(indio_dev);
-	unsigned int ret;
+	int ret;
 
 	mutex_lock(&st->lock);
 

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Michael Hennerich <[email protected]>

commit 33d7ecbf69aa7dd4145e3b77962bcb8759eede3d upstream.

The ADF4350/1 features a programmable dual-modulus prescaler of 4/5 or 8/9. When set to 4/5, the maximum RF frequency allowed is 3 GHz. Therefore, when operating the ADF4351 above 3 GHz, this must be set to 8/9. In this context not the RF output frequency is meant
- it's the VCO frequency.

Therefore move the prescaler selection after we derived the VCO frequency from the desired RF output frequency.

This BUG may have caused PLL lock instabilities when operating the VCO at the very high range close to 4.4 GHz.

---
 drivers/iio/frequency/adf4350.c |   20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)

@@ -143,6 +143,19 @@ static int adf4350_set_freq(struct adf43
 	if (freq > ADF4350_MAX_OUT_FREQ || freq < st->min_out_freq)
 		return -EINVAL;
 
+	st->r4_rf_div_sel = 0;
+
+	/*
+	 * !\TODO: The below computation is making sure we get a power of 2
+	 * shift (st->r4_rf_div_sel) so that freq becomes higher or equal to
+	 * ADF4350_MIN_VCO_FREQ. This might be simplified with fls()/fls_long()
+	 * and friends.
+	 */
+	while (freq < ADF4350_MIN_VCO_FREQ) {
+		freq <<= 1;
+		st->r4_rf_div_sel++;
+	}
+
 	if (freq > ADF4350_MAX_FREQ_45_PRESC) {
 		prescaler = ADF4350_REG1_PRESCALER;
 		mdiv = 75;

@@ -151,13 +164,6 @@ static int adf4350_set_freq(struct adf43
 		mdiv = 23;
 	}
 
-	st->r4_rf_div_sel = 0;
-
-	while (freq < ADF4350_MIN_VCO_FREQ) {
-		freq <<= 1;
-		st->r4_rf_div_sel++;
-	}
-
 	/*
 	 * Allow a predefined reference division factor
 	 * if not set, compute our own

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Huacai Chen <[email protected]>

commit e416f0ed3c500c05c55fb62ee62662717b1c7f71 upstream.

BootLoaders (Grub, LILO, etc) may pass an identifier such as "BOOT_IMAGE= /boot/vmlinuz-x.y.z" to kernel parameters. But these identifiers are not recognized by the kernel itself so will be passed to userspace. However user space init program also don't recognize it.

KEXEC/KDUMP (kexec-tools) may also pass an identifier such as "kexec" on some architectures.

We cannot change BootLoader's behavior, because this behavior exists for many years, and there are already user space programs search BOOT_IMAGE= in /proc/cmdline to obtain the kernel image locations:

https://github.com/linuxdeepin/deepin-ab-recovery/blob/master/util.go
(search getBootOptions)
https://github.com/linuxdeepin/deepin-ab-recovery/blob/master/main.go
(search getKernelReleaseWithBootOption) So the the best way is handle (ignore) it by the kernel itself, which can avoid such boot warnings (if we use something like init=/bin/bash, bootloader identifier can even cause a crash):

Kernel command line: BOOT_IMAGE=(hd0,1)/vmlinuz-6.x root=/dev/sda3 ro console=tty Unknown kernel command line parameters "BOOT_IMAGE=(hd0,1)/vmlinuz-6.x", will be passed to user space.

---
 init/main.c |   14 ++++++++++++++
 1 file changed, 14 insertions(+)

@@ -540,6 +540,14 @@ static int __init unknown_bootoption(cha
 				     const char *unused, void *arg)
 {
 	size_t len = strlen(param);
+	int i;
+
+	/*
+	 * Well-known bootloader identifiers:
+	 * 1. LILO/Grub pass "BOOT_IMAGE=...";
+	 * 2. kexec/kdump (kexec-tools) pass "kexec".
+	 */
+	const char *bootloader[] = { "BOOT_IMAGE=", "kexec", NULL };
 
 	/* Handle params aliased to sysctls */
 	if (sysctl_is_alias(param))

@@ -547,6 +555,12 @@ static int __init unknown_bootoption(cha
 
 	repair_env_string(param, val);
 
+	/* Handle bootloader identifier */
+	for (i = 0; bootloader[i]; i++) {
+		if (strstarts(param, bootloader[i]))
+			return 0;
+	}
+
 	/* Handle obsolete-style parameters */
 	if (obsolete_checksetup(param))
 		return 0;

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Nyekjaer <[email protected]>

commit a95a0b4e471a6d8860f40c6ac8f1cad9dde3189a upstream.

Remove unnecessary calls to pm_runtime_disable(), pm_runtime_set_active(), and pm_runtime_enable() from the resume path. These operations are not required here and can interfere with proper pm_runtime state handling, especially when resuming from a pm_runtime suspended state.

---
 drivers/iio/imu/inv_icm42600/inv_icm42600_core.c |    4 ----
 1 file changed, 4 deletions(-)

@@ -730,10 +730,6 @@ static int __maybe_unused inv_icm42600_r
 	if (ret)
 		goto out_unlock;
 
-	pm_runtime_disable(dev);
-	pm_runtime_set_active(dev);
-	pm_runtime_enable(dev);
-
 	/* restore sensors state */
 	ret = inv_icm42600_set_pwr_mgmt0(st, st->suspended.gyro,
 					 st->suspended.accel,

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Lu Baolu <[email protected]>

commit 5ef7e24c742038a5d8c626fdc0e3a21834358341 upstream.

The specification, Section 7.10, "Software Steps to Drain Page Requests & Responses," requires software to submit an Invalidation Wait Descriptor (inv_wait_dsc) with the Page-request Drain (PD=1) flag set, along with the Invalidation Wait Completion Status Write flag (SW=1). It then waits for the Invalidation Wait Descriptor's completion.

However, the PD field in the Invalidation Wait Descriptor is optional, as stated in Section 6.5.2.9, "Invalidation Wait Descriptor":

"Page-request Drain (PD): Remapping hardware implementations reporting
Page-request draining as not supported (PDS = 0 in ECAP_REG) treat this field as reserved."

This implies that if the IOMMU doesn't support the PDS capability, software can't drain page requests and group responses as expected.

Do not enable PCI/PRI if the IOMMU doesn't support PDS.

---
 drivers/iommu/intel/iommu.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -2659,7 +2659,7 @@ static struct dmar_domain *dmar_insert_o
 			}
 
 			if (info->ats_supported && ecap_prs(iommu->ecap) &&
-			    pci_pri_supported(pdev))
+			    ecap_pds(iommu->ecap) && pci_pri_supported(pdev))
 				info->pri_supported = 1;
 		}
 	}

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Biggers <[email protected]>

commit eed0e3d305530066b4fc5370107cff8ef1a0d229 upstream.

To prevent timing attacks, HMAC value comparison needs to be constant time. Replace the memcmp() with the correct function, crypto_memneq().

[For the Fixes commit I used the commit that introduced the memcmp(). It predates the introduction of crypto_memneq(), but it was still a bug at the time even though a helper function didn't exist yet.]

---
 security/keys/trusted-keys/trusted_tpm1.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

@@ -7,6 +7,7 @@
  */
 
 #include <crypto/hash_info.h>
+#include <crypto/algapi.h>
 #include <linux/init.h>
 #include <linux/slab.h>
 #include <linux/parser.h>

@@ -241,7 +242,7 @@ int TSS_checkhmac1(unsigned char *buffer
 	if (ret < 0)
 		goto out;
 
-	if (memcmp(testhmac, authdata, SHA1_DIGEST_SIZE))
+	if (crypto_memneq(testhmac, authdata, SHA1_DIGEST_SIZE))
 		ret = -EINVAL;
 out:
 	kfree_sensitive(sdesc);

@@ -334,7 +335,7 @@ static int TSS_checkhmac2(unsigned char
 			  TPM_NONCE_SIZE, ononce, 1, continueflag1, 0, 0);
 	if (ret < 0)
 		goto out;
-	if (memcmp(testhmac1, authdata1, SHA1_DIGEST_SIZE)) {
+	if (crypto_memneq(testhmac1, authdata1, SHA1_DIGEST_SIZE)) {
 		ret = -EINVAL;
 		goto out;
 	}

@@ -343,7 +344,7 @@ static int TSS_checkhmac2(unsigned char
 			  TPM_NONCE_SIZE, ononce, 1, continueflag2, 0, 0);
 	if (ret < 0)
 		goto out;
-	if (memcmp(testhmac2, authdata2, SHA1_DIGEST_SIZE))
+	if (crypto_memneq(testhmac2, authdata2, SHA1_DIGEST_SIZE))
 		ret = -EINVAL;
 out:
 	kfree_sensitive(sdesc);

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johan Hovold <[email protected]>

commit 1260cbcffa608219fc9188a6cbe9c45a300ef8b5 upstream.

Make sure to drop the reference taken when looking up the genpool platform device in of_gen_pool_get() before returning the pool.

Note that holding a reference to a device does typically not prevent its devres managed resources from being released so there is no point in keeping the reference.

---
 lib/genalloc.c |    5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

@@ -899,8 +899,11 @@ struct gen_pool *of_gen_pool_get(struct
 		if (!name)
 			name = np_pool->name;
 	}
-	if (pdev)
+	if (pdev) {
 		pool = gen_pool_get(&pdev->dev, name);
+		put_device(&pdev->dev);
+	}
+
 	of_node_put(np_pool);
 
 	return pool;

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Askar Safin <[email protected]>

commit 042a60680de43175eb4df0977ff04a4eba9da082 upstream.

openat2 had a bug: if we pass RESOLVE_NO_XDEV, then openat2 doesn't traverse through automounts, but may still trigger them. (See the link for full bug report with reproducer.)

This commit fixes this bug.

---
 fs/namei.c |    8 ++++++++
 1 file changed, 8 insertions(+)

@@ -1307,6 +1307,10 @@ static int follow_automount(struct path
 	    dentry->d_inode)
 		return -EISDIR;
 
+	/* No need to trigger automounts if mountpoint crossing is disabled. */
+	if (lookup_flags & LOOKUP_NO_XDEV)
+		return -EXDEV;
+
 	if (count && (*count)++ >= MAXSYMLINKS)
 		return -ELOOP;
 

@@ -1330,6 +1334,10 @@ static int __traverse_mounts(struct path
 		/* Allow the filesystem to manage the transit without i_mutex
 		 * being held. */
 		if (flags & DCACHE_MANAGE_TRANSIT) {
+			if (lookup_flags & LOOKUP_NO_XDEV) {
+				ret = -EXDEV;
+				break;
+			}
 			ret = path->dentry->d_op->d_manage(path, false);
 			flags = smp_load_acquire(&path->dentry->d_flags);
 			if (ret < 0)

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sam James <[email protected]>

commit 8ec5a066f88f89bd52094ba18792b34c49dcd55a upstream.

Similar in nature to ab107276607af90b13a5994997e19b7b9731e251. glibc-2.42 drops the legacy termio struct, but the ioctls.h header still defines some TC* constants in terms of termio (via sizeof). Hardcode the values instead.

This fixes building Python for example, which falls over like:
./Modules/termios.c:1119:16: error: invalid application of 'sizeof' to incomplete type 'struct termio'

---
 arch/parisc/include/uapi/asm/ioctls.h |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

@@ -10,10 +10,10 @@
 #define TCSETS		_IOW('T', 17, struct termios) /* TCSETATTR */
 #define TCSETSW		_IOW('T', 18, struct termios) /* TCSETATTRD */
 #define TCSETSF		_IOW('T', 19, struct termios) /* TCSETATTRF */
-#define TCGETA		_IOR('T', 1, struct termio)
-#define TCSETA		_IOW('T', 2, struct termio)
-#define TCSETAW		_IOW('T', 3, struct termio)
-#define TCSETAF		_IOW('T', 4, struct termio)
+#define TCGETA          0x40125401
+#define TCSETA          0x80125402
+#define TCSETAW         0x80125403
+#define TCSETAF         0x80125404
 #define TCSBRK		_IO('T', 5)
 #define TCXONC		_IO('T', 6)
 #define TCFLSH		_IO('T', 7)

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Georg Gottleuber <[email protected]>

commit eeaed48980a7aeb0d3d8b438185d4b5a66154ff9 upstream.

On the TUXEDO InfinityBook S Gen8, a Samsung 990 Evo NVMe leads to a high power consumption in s2idle sleep (3.5 watts).

This patch applies 'Force No Simple Suspend' quirk to achieve a sleep with a lower power consumption, typically around 1 watts.

---
 drivers/nvme/host/pci.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -2979,10 +2979,12 @@ static unsigned long check_vendor_combin
 		 * Exclude Samsung 990 Evo from NVME_QUIRK_SIMPLE_SUSPEND
 		 * because of high power consumption (> 2 Watt) in s2idle
 		 * sleep. Only some boards with Intel CPU are affected.
+		 * (Note for testing: Samsung 990 Evo Plus has same PCI ID)
 		 */
 		if (dmi_match(DMI_BOARD_NAME, "DN50Z-140HC-YD") ||
 		    dmi_match(DMI_BOARD_NAME, "GMxPXxx") ||
 		    dmi_match(DMI_BOARD_NAME, "GXxMRXx") ||
+		    dmi_match(DMI_BOARD_NAME, "NS5X_NS7XAU") ||
 		    dmi_match(DMI_BOARD_NAME, "PH4PG31") ||
 		    dmi_match(DMI_BOARD_NAME, "PH4PRX1_PH6PRX1") ||
 		    dmi_match(DMI_BOARD_NAME, "PH6PG01_PH6PG71"))

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nam Cao <[email protected]>

commit a39087905af9ffecaa237a918a2c03a04e479934 upstream.

pnv_irq_domain_alloc() allocates interrupts at parent's interrupt domain. If it fails in the progress, all allocated interrupts are freed.

The number of successfully allocated interrupts so far is stored "i". However, "i - 1" interrupts are freed. This is broken:

- One interrupt is not be freed

- If "i" is zero, "i - 1" wraps around

Correct the number of freed interrupts to "i".

---
 arch/powerpc/platforms/powernv/pci-ioda.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -2243,7 +2243,7 @@ static int pnv_irq_domain_alloc(struct i
 	return 0;
 
 out:
-	irq_domain_free_irqs_parent(domain, virq, i - 1);
+	irq_domain_free_irqs_parent(domain, virq, i);
 	msi_bitmap_free_hwirqs(&phb->msi_bmp, hwirq, nr_irqs);
 	return ret;
 }

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nam Cao <[email protected]>

commit 3443ff3be6e59b80d74036bb39f5b6409eb23cc9 upstream.

pseries_irq_domain_alloc() allocates interrupts at parent's interrupt domain. If it fails in the progress, all allocated interrupts are freed.

The number of successfully allocated interrupts so far is stored "i". However, "i - 1" interrupts are freed. This is broken:

- One interrupt is not be freed

- If "i" is zero, "i - 1" wraps around

Correct the number of freed interrupts to 'i'.

---
 arch/powerpc/platforms/pseries/msi.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -607,7 +607,7 @@ static int pseries_irq_domain_alloc(stru
 
 out:
 	/* TODO: handle RTAS cleanup in ->msi_finish() ? */
-	irq_domain_free_irqs_parent(domain, virq, i - 1);
+	irq_domain_free_irqs_parent(domain, virq, i);
 	return ret;
 }
 

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jisheng Zhang <[email protected]>

commit 3a4b9d027e4061766f618292df91760ea64a1fcc upstream.

The 'enable' register should be BERLIN_PWM_EN rather than BERLIN_PWM_ENABLE, otherwise, the driver accesses wrong address, there will be cpu exception then kernel panic during suspend/resume.

---
 drivers/pwm/pwm-berlin.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -274,7 +274,7 @@ static int berlin_pwm_suspend(struct dev
 		if (!channel)
 			continue;
 
-		channel->enable = berlin_pwm_readl(bpc, i, BERLIN_PWM_ENABLE);
+		channel->enable = berlin_pwm_readl(bpc, i, BERLIN_PWM_EN);
 		channel->ctrl = berlin_pwm_readl(bpc, i, BERLIN_PWM_CONTROL);
 		channel->duty = berlin_pwm_readl(bpc, i, BERLIN_PWM_DUTY);
 		channel->tcnt = berlin_pwm_readl(bpc, i, BERLIN_PWM_TCNT);

@@ -305,7 +305,7 @@ static int berlin_pwm_resume(struct devi
 		berlin_pwm_writel(bpc, i, channel->ctrl, BERLIN_PWM_CONTROL);
 		berlin_pwm_writel(bpc, i, channel->duty, BERLIN_PWM_DUTY);
 		berlin_pwm_writel(bpc, i, channel->tcnt, BERLIN_PWM_TCNT);
-		berlin_pwm_writel(bpc, i, channel->enable, BERLIN_PWM_ENABLE);
+		berlin_pwm_writel(bpc, i, channel->enable, BERLIN_PWM_EN);
 	}
 
 	return 0;

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Thorsten Blum <[email protected]>

commit b81296591c567b12d3873b05a37b975707959b94 upstream.

Replace kmalloc() followed by copy_from_user() with memdup_user() to fix a memory leak that occurs when copy_from_user(buff[sg_used],,) fails and the 'cleanup1:' path does not free the memory for 'buff[sg_used]'. Using memdup_user() avoids this by freeing the memory internally.

Since memdup_user() already allocates memory, use kzalloc() in the else branch instead of manually zeroing 'buff[sg_used]' using memset(0).

---
 drivers/scsi/hpsa.c |   21 ++++++++++++---------
 1 file changed, 12 insertions(+), 9 deletions(-)

@@ -6528,18 +6528,21 @@ static int hpsa_big_passthru_ioctl(struc
 	while (left) {
 		sz = (left > ioc->malloc_size) ? ioc->malloc_size : left;
 		buff_size[sg_used] = sz;
-		buff[sg_used] = kmalloc(sz, GFP_KERNEL);
-		if (buff[sg_used] == NULL) {
-			status = -ENOMEM;
-			goto cleanup1;
-		}
+
 		if (ioc->Request.Type.Direction & XFER_WRITE) {
-			if (copy_from_user(buff[sg_used], data_ptr, sz)) {
-				status = -EFAULT;
+			buff[sg_used] = memdup_user(data_ptr, sz);
+			if (IS_ERR(buff[sg_used])) {
+				status = PTR_ERR(buff[sg_used]);
+				goto cleanup1;
+			}
+		} else {
+			buff[sg_used] = kzalloc(sz, GFP_KERNEL);
+			if (!buff[sg_used]) {
+				status = -ENOMEM;
 				goto cleanup1;
 			}
-		} else
-			memset(buff[sg_used], 0, sz);
+		}
+
 		left -= sz;
 		data_ptr += sz;
 		sg_used++;

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Biggers <[email protected]>

commit dd91c79e4f58fbe2898dac84858033700e0e99fb upstream.

To prevent timing attacks, MACs need to be compared in constant time. Use the appropriate helper function for this.

---
 net/sctp/sm_make_chunk.c |    3 ++-
 net/sctp/sm_statefuns.c  |    3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

@@ -31,6 +31,7 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include <crypto/hash.h>
+#include <crypto/algapi.h>
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/ip.h>

@@ -1796,7 +1797,7 @@ struct sctp_association *sctp_unpack_coo
 		}
 	}
 
-	if (memcmp(digest, cookie->signature, SCTP_SIGNATURE_SIZE)) {
+	if (crypto_memneq(digest, cookie->signature, SCTP_SIGNATURE_SIZE)) {
 		*error = -SCTP_IERROR_BAD_SIG;
 		goto fail;
 	}

@@ -30,6 +30,7 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
+#include <crypto/algapi.h>
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/ip.h>

@@ -4402,7 +4403,7 @@ static enum sctp_ierror sctp_sf_authenti
 				 sh_key, GFP_ATOMIC);
 
 	/* Discard the packet if the digests do not match */
-	if (memcmp(save_digest, digest, sig_len)) {
+	if (crypto_memneq(save_digest, digest, sig_len)) {
 		kfree(save_digest);
 		return SCTP_IERROR_BAD_SIG;
 	}

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Anthony Yznaga <[email protected]>

commit 6fd44a481b3c6111e4801cec964627791d0f3ec5 upstream.

An attempt to exercise sparc hugetlb code in a sun4u-based guest running under qemu results in the guest hanging due to being stuck in a trap loop. This is due to invalid hugetlb TTEs being installed that do not have the expected _PAGE_PMD_HUGE and page size bits set. Although the breakage has gone apparently unnoticed for several years, fix it now so there is the option to exercise sparc hugetlb code under qemu. This can be useful because sun4v support in qemu does not support linux guests currently and sun4v-based hardware resources may not be readily available.

Fix tested with a 6.15.2 and 6.16-rc6 kernels by running libhugetlbfs tests on a qemu guest running Debian 13.

---
 arch/sparc/mm/hugetlbpage.c |   20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

@@ -133,6 +133,26 @@ hugetlb_get_unmapped_area(struct file *f
 
 static pte_t sun4u_hugepage_shift_to_tte(pte_t entry, unsigned int shift)
 {
+	unsigned long hugepage_size = _PAGE_SZ4MB_4U;
+
+	pte_val(entry) = pte_val(entry) & ~_PAGE_SZALL_4U;
+
+	switch (shift) {
+	case HPAGE_256MB_SHIFT:
+		hugepage_size = _PAGE_SZ256MB_4U;
+		pte_val(entry) |= _PAGE_PMD_HUGE;
+		break;
+	case HPAGE_SHIFT:
+		pte_val(entry) |= _PAGE_PMD_HUGE;
+		break;
+	case HPAGE_64K_SHIFT:
+		hugepage_size = _PAGE_SZ64K_4U;
+		break;
+	default:
+		WARN_ONCE(1, "unsupported hugepage shift=%u\n", shift);
+	}
+
+	pte_val(entry) = pte_val(entry) | hugepage_size;
 	return entry;
 }
 

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ma Ke <[email protected]>

commit 302c04110f0ce70d25add2496b521132548cd408 upstream.

Once of_device_register() failed, we should call put_device() to decrement reference count for cleanup. Or it could cause memory leak. So fix this by calling put_device(), then the name can be freed in kobject_cleanup().

Calling path: of_device_register() -> of_device_add() -> device_add(). As comment of device_add() says, 'if device_add() succeeds, you should call device_del() when you want to get rid of it. If device_add() has not succeeded, use only put_device() to drop the reference count'.

Found by code review.

---
 arch/sparc/kernel/of_device_32.c |    1 +
 arch/sparc/kernel/of_device_64.c |    1 +
 2 files changed, 2 insertions(+)

@@ -387,6 +387,7 @@ static struct platform_device * __init s
 
 	if (of_device_register(op)) {
 		printk("%pOF: Could not register of device.\n", dp);
+		put_device(&op->dev);
 		kfree(op);
 		op = NULL;
 	}

@@ -680,6 +680,7 @@ static struct platform_device * __init s
 
 	if (of_device_register(op)) {
 		printk("%pOF: Could not register of device.\n", dp);
+		put_device(&op->dev);
 		kfree(op);
 		op = NULL;
 	}

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Linus Walleij <[email protected]>

commit b8df622cf7f6808c85764e681847150ed6d85f3d upstream.

If you don't specify buswidth 2 (16 bits) in the device tree, FSMC doesn't even probe anymore:

fsmc-nand 10100000.flash: FSMC device partno 090,
manufacturer 80, revision 00, config 00
nand: device found, Manufacturer ID: 0x20, Chip ID: 0xb1
nand: ST Micro 10100000.flash
nand: bus width 8 instead of 16 bits
nand: No NAND device found
fsmc-nand 10100000.flash: probe with driver fsmc-nand failed
with error -22

With this patch to use autodetection unless buswidth is specified, the device is properly detected again:

fsmc-nand 10100000.flash: FSMC device partno 090,
manufacturer 80, revision 00, config 00
nand: device found, Manufacturer ID: 0x20, Chip ID: 0xb1
nand: ST Micro NAND 128MiB 1,8V 16-bit
nand: 128 MiB, SLC, erase size: 128 KiB, page size: 2048, OOB size: 64 fsmc-nand 10100000.flash: Using 1-bit HW ECC scheme Scanning device for bad blocks

I don't know where or how this happened, I think some change in the nand core.

---
 drivers/mtd/nand/raw/fsmc_nand.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

@@ -876,10 +876,14 @@ static int fsmc_nand_probe_config_dt(str
 	if (!of_property_read_u32(np, "bank-width", &val)) {
 		if (val == 2) {
 			nand->options |= NAND_BUSWIDTH_16;
-		} else if (val != 1) {
+		} else if (val == 1) {
+			nand->options |= NAND_BUSWIDTH_AUTO;
+		} else {
 			dev_err(&pdev->dev, "invalid bank-width %u\n", val);
 			return -EINVAL;
 		}
+	} else {
+		nand->options |= NAND_BUSWIDTH_AUTO;
 	}
 
 	if (of_get_property(np, "nand-skip-bbtscan", NULL))

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rex Chen <[email protected]>

commit fec40f44afdabcbc4a7748e4278f30737b54bb1a upstream.

SPI mode doesn't support cmd7, so remove it in mmc_sdio_alive() and confirm if sdio is active by checking CCCR register value is available or not.

---
 drivers/mmc/core/sdio.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

@@ -929,7 +929,11 @@ static void mmc_sdio_remove(struct mmc_h
  */
 static int mmc_sdio_alive(struct mmc_host *host)
 {
-	return mmc_select_card(host->card);
+	if (!mmc_host_is_spi(host))
+		return mmc_select_card(host->card);
+	else
+		return mmc_io_rw_direct(host->card, 0, 0, SDIO_CCCR_CCCR, 0,
+					NULL);
 }
 
 /*

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhen Ni <[email protected]>

commit 6744085079e785dae5f7a2239456135407c58b25 upstream.

The of_platform_populate() call at the end of the function has a possible failure path, causing a resource leak.

Replace of_iomap() with devm_platform_ioremap_resource() to ensure automatic cleanup of srom->reg_base.

This issue was detected by smatch static analysis:
drivers/memory/samsung/exynos-srom.c:155 exynos_srom_probe()warn:
'srom->reg_base' from of_iomap() not released on lines: 155.

---
 drivers/memory/samsung/exynos-srom.c |   10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

@@ -121,20 +121,18 @@ static int exynos_srom_probe(struct plat
 		return -ENOMEM;
 
 	srom->dev = dev;
-	srom->reg_base = of_iomap(np, 0);
-	if (!srom->reg_base) {
+	srom->reg_base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(srom->reg_base)) {
 		dev_err(&pdev->dev, "iomap of exynos srom controller failed\n");
-		return -ENOMEM;
+		return PTR_ERR(srom->reg_base);
 	}
 
 	platform_set_drvdata(pdev, srom);
 
 	srom->reg_offset = exynos_srom_alloc_reg_dump(exynos_srom_offsets,
 						      ARRAY_SIZE(exynos_srom_offsets));
-	if (!srom->reg_offset) {
-		iounmap(srom->reg_base);
+	if (!srom->reg_offset)
 		return -ENOMEM;
-	}
 
 	for_each_child_of_node(np, child) {
 		if (exynos_srom_configure_bank(srom, child)) {

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Esben Haabendal <[email protected]>

commit 9db26d5855d0374d4652487bfb5aacf40821c469 upstream.

When setting a normal alarm, user-space is responsible for using RTC_AIE_ON/RTC_AIE_OFF to control if alarm irq should be enabled.

But when RTC_UIE_ON is used, interrupts must be enabled so that the requested irq events are generated.
When RTC_UIE_OFF is used, alarm irq is disabled if there are no other alarms queued, so this commit brings symmetry to that.

---
 drivers/rtc/interface.c |    4 ++++
 1 file changed, 4 insertions(+)

@@ -583,6 +583,10 @@ int rtc_update_irq_enable(struct rtc_dev
 		rtc->uie_rtctimer.node.expires = ktime_add(now, onesec);
 		rtc->uie_rtctimer.period = ktime_set(1, 0);
 		err = rtc_timer_enqueue(rtc, &rtc->uie_rtctimer);
+		if (!err && rtc->ops && rtc->ops->alarm_irq_enable)
+			err = rtc->ops->alarm_irq_enable(rtc->dev.parent, 1);
+		if (err)
+			goto out;
 	} else {
 		rtc_timer_remove(rtc, &rtc->uie_rtctimer);
 	}

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Esben Haabendal <[email protected]>

commit 795cda8338eab036013314dbc0b04aae728880ab upstream.

As described in the old comment dating back to
commit 6610e0893b8b ("RTC: Rework RTC code to use timerqueue for events") from 2010, we have been living with a race window when setting alarm with an expiry in the near future (i.e. next second).
With 1 second resolution, it can happen that the second ticks after the check for the timer having expired, but before the alarm is actually set. When this happen, no alarm IRQ is generated, at least not with some RTC chips (isl12022 is an example of this).

With UIE RTC timer being implemented on top of alarm irq, being re-armed every second, UIE will occasionally fail to work, as an alarm irq lost due to this race will stop the re-arming loop.

For now, I have limited the additional expiry check to only be done for alarms set to next seconds. I expect it should be good enough, although I don't know if we can now for sure that systems with loads could end up causing the same problems for alarms set 2 seconds or even longer in the future.

I haven't been able to reproduce the problem with this check in place.

---
 drivers/rtc/interface.c |   23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

@@ -442,6 +442,29 @@ static int __rtc_set_alarm(struct rtc_de
 	else
 		err = rtc->ops->set_alarm(rtc->dev.parent, alarm);
 
+	/*
+	 * Check for potential race described above. If the waiting for next
+	 * second, and the second just ticked since the check above, either
+	 *
+	 * 1) It ticked after the alarm was set, and an alarm irq should be
+	 *    generated.
+	 *
+	 * 2) It ticked before the alarm was set, and alarm irq most likely will
+	 * not be generated.
+	 *
+	 * While we cannot easily check for which of these two scenarios we
+	 * are in, we can return -ETIME to signal that the timer has already
+	 * expired, which is true in both cases.
+	 */
+	if ((scheduled - now) <= 1) {
+		err = __rtc_read_time(rtc, &tm);
+		if (err)
+			return err;
+		now = rtc_tm_to_time64(&tm);
+		if (scheduled <= now)
+			return -ETIME;
+	}
+
 	trace_rtc_set_alarm(rtc_tm_to_time64(&alarm->time), err);
 	return err;
 }

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Christopherson <[email protected]>

commit a001cd248ab244633c5fabe4f7c707e13fc1d1cc upstream.

Add "extern" to the glibc-defined weak rseq symbols to convert the rseq selftest's usage from weak symbol definitions to weak symbol _references_. Effectively re-defining the glibc symbols wreaks havoc when building with -fno-common, e.g. generates segfaults when running multi-threaded programs, as dynamically linked applications end up with multiple versions of the symbols.

Building with -fcommon, which until recently has the been the default for GCC and clang, papers over the bug by allowing the linker to resolve the weak/tentative definition to glibc's "real" definition.

Note, the symbol itself (or rather its address), not the value of the symbol, is set to 0/NULL for unresolved weak symbol references, as the symbol doesn't exist and thus can't have a value. Check for a NULL rseq size pointer to handle the scenario where the test is statically linked against a libc that doesn't support rseq in any capacity.

---
 tools/testing/selftests/rseq/rseq.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

@@ -38,9 +38,9 @@
  * Define weak versions to play nice with binaries that are statically linked
  * against a libc that doesn't support registering its own rseq.
  */
-__weak ptrdiff_t __rseq_offset;
-__weak unsigned int __rseq_size;
-__weak unsigned int __rseq_flags;
+extern __weak ptrdiff_t __rseq_offset;
+extern __weak unsigned int __rseq_size;
+extern __weak unsigned int __rseq_flags;
 
 static const ptrdiff_t *libc_rseq_offset_p = &__rseq_offset;
 static const unsigned int *libc_rseq_size_p = &__rseq_size;

@@ -124,7 +124,7 @@ void rseq_init(void)
 	 * libc not having registered a restartable sequence.  Try to find the
 	 * symbols if that's the case.
 	 */
-	if (!*libc_rseq_size_p) {
+	if (!libc_rseq_size_p || !*libc_rseq_size_p) {
 		libc_rseq_offset_p = dlsym(RTLD_NEXT, "__rseq_offset");
 		libc_rseq_size_p = dlsym(RTLD_NEXT, "__rseq_size");
 		libc_rseq_flags_p = dlsym(RTLD_NEXT, "__rseq_flags");

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Brian Norris <[email protected]>

commit 48991e4935078b05f80616c75d1ee2ea3ae18e58 upstream.

The "max_link_width", "current_link_speed", "current_link_width", "secondary_bus_number", and "subordinate_bus_number" sysfs files all access config registers, but they don't check the runtime PM state. If the device is in D3cold or a parent bridge is suspended, we may see -EINVAL, bogus values, or worse, depending on implementation details.

Wrap these access in pci_config_pm_runtime_{get,put}() like most of the rest of the similar sysfs attributes.

Notably, "max_link_speed" does not access config registers; it returns a cached value since d2bd39c0456b ("PCI: Store all PCIe Supported Link Speeds").

---
 drivers/pci/pci-sysfs.c |   20 +++++++++++++++++++-
 1 file changed, 19 insertions(+), 1 deletion(-)

@@ -174,8 +174,14 @@ static ssize_t max_link_width_show(struc
 				   struct device_attribute *attr, char *buf)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
+	ssize_t ret;
 
-	return sysfs_emit(buf, "%u\n", pcie_get_width_cap(pdev));
+	/* We read PCI_EXP_LNKCAP, so we need the device to be accessible. */
+	pci_config_pm_runtime_get(pdev);
+	ret = sysfs_emit(buf, "%u\n", pcie_get_width_cap(pdev));
+	pci_config_pm_runtime_put(pdev);
+
+	return ret;
 }
 static DEVICE_ATTR_RO(max_link_width);
 

@@ -187,7 +193,10 @@ static ssize_t current_link_speed_show(s
 	int err;
 	enum pci_bus_speed speed;
 
+	pci_config_pm_runtime_get(pci_dev);
 	err = pcie_capability_read_word(pci_dev, PCI_EXP_LNKSTA, &linkstat);
+	pci_config_pm_runtime_put(pci_dev);
+
 	if (err)
 		return -EINVAL;
 

@@ -204,7 +213,10 @@ static ssize_t current_link_width_show(s
 	u16 linkstat;
 	int err;
 
+	pci_config_pm_runtime_get(pci_dev);
 	err = pcie_capability_read_word(pci_dev, PCI_EXP_LNKSTA, &linkstat);
+	pci_config_pm_runtime_put(pci_dev);
+
 	if (err)
 		return -EINVAL;
 

@@ -220,7 +232,10 @@ static ssize_t secondary_bus_number_show
 	u8 sec_bus;
 	int err;
 
+	pci_config_pm_runtime_get(pci_dev);
 	err = pci_read_config_byte(pci_dev, PCI_SECONDARY_BUS, &sec_bus);
+	pci_config_pm_runtime_put(pci_dev);
+
 	if (err)
 		return -EINVAL;
 

@@ -236,7 +251,10 @@ static ssize_t subordinate_bus_number_sh
 	u8 sub_bus;
 	int err;
 
+	pci_config_pm_runtime_get(pci_dev);
 	err = pci_read_config_byte(pci_dev, PCI_SUBORDINATE_BUS, &sub_bus);
+	pci_config_pm_runtime_put(pci_dev);
+
 	if (err)
 		return -EINVAL;
 

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Niklas Schnelle <[email protected]>

commit 05703271c3cdcc0f2a8cf6ebdc45892b8ca83520 upstream.

Before disabling SR-IOV via config space accesses to the parent PF, sriov_disable() first removes the PCI devices representing the VFs.

Since commit 9d16947b7583 ("PCI: Add global pci_lock_rescan_remove()") such removal operations are serialized against concurrent remove and rescan using the pci_rescan_remove_lock. No such locking was ever added in sriov_disable() however. In particular when commit 18f9e9d150fc ("PCI/IOV: Factor out sriov_add_vfs()") factored out the PCI device removal into sriov_del_vfs() there was still no locking around the pci_iov_remove_virtfn() calls.

On s390 the lack of serialization in sriov_disable() may cause double remove and list corruption with the below (amended) trace being observed:

PSW: 0704c00180000000 0000000c914e4b38 (klist_put+56)
GPRS: 000003800313fb48 0000000000000000 0000000100000001 0000000000000001

	00000000f9b520a8 0000000000000000 0000000000002fbd 00000000f4cc9480
	0000000000000001 0000000000000000 0000000000000000 0000000180692828
	00000000818e8000 000003800313fe2c 000003800313fb20 000003800313fad8

#0 [3800313fb20] device_del at c9158ad5c
#1 [3800313fb88] pci_remove_bus_device at c915105ba #2 [3800313fbd0] pci_iov_remove_virtfn at c9152f198 #3 [3800313fc28] zpci_iov_remove_virtfn at c90fb67c0 #4 [3800313fc60] zpci_bus_remove_device at c90fb6104 #5 [3800313fca0] __zpci_event_availability at c90fb3dca #6 [3800313fd08] chsc_process_sei_nt0 at c918fe4a2 #7 [3800313fd60] crw_collect_info at c91905822 #8 [3800313fe10] kthread at c90feb390
#9 [3800313fe68] __ret_from_fork at c90f6aa64 #10 [3800313fe98] ret_from_fork at c9194f3f2.

This is because in addition to sriov_disable() removing the VFs, the platform also generates hot-unplug events for the VFs. This being the reverse operation to the hotplug events generated by sriov_enable() and handled via pdev->no_vf_scan. And while the event processing takes pci_rescan_remove_lock and checks whether the struct pci_dev still exists, the lack of synchronization makes this checking racy.

Other races may also be possible of course though given that this lack of locking persisted so long observable races seem very rare. Even on s390 the list corruption was only observed with certain devices since the platform events are only triggered by config accesses after the removal, so as long as the removal finished synchronously they would not race. Either way the locking is missing so fix this by adding it to the sriov_del_vfs() helper.

Just like PCI rescan-remove, locking is also missing in sriov_add_vfs() including for the error case where pci_stop_and_remove_bus_device() is called without the PCI rescan-remove lock being held. Even in the non-error case, adding new PCI devices and buses should be serialized via the PCI rescan-remove lock. Add the necessary locking.

---
 drivers/pci/iov.c |    5 +++++
 1 file changed, 5 insertions(+)

@@ -541,15 +541,18 @@ static int sriov_add_vfs(struct pci_dev
 	if (dev->no_vf_scan)
 		return 0;
 
+	pci_lock_rescan_remove();
 	for (i = 0; i < num_vfs; i++) {
 		rc = pci_iov_add_virtfn(dev, i);
 		if (rc)
 			goto failed;
 	}
+	pci_unlock_rescan_remove();
 	return 0;
 failed:
 	while (i--)
 		pci_iov_remove_virtfn(dev, i);
+	pci_unlock_rescan_remove();
 
 	return rc;
 }

@@ -669,8 +672,10 @@ static void sriov_del_vfs(struct pci_dev
 	struct pci_sriov *iov = dev->sriov;
 	int i;
 
+	pci_lock_rescan_remove();
 	for (i = 0; i < iov->num_VFs; i++)
 		pci_iov_remove_virtfn(dev, i);
+	pci_unlock_rescan_remove();
 }
 
 static void sriov_disable(struct pci_dev *dev)