Skip to contentSkip to conversation

s390/cio: unregister device when the only path is gone

PATCH New Sasha Levin 1 participant 6 messages Jun 16, 2023

Subsystems (3)
lore
Share & tools
lore

Sasha Levin

author maintainer
#0 lore raw

PATCH 1/6 s390/cio: unregister device when the only path is gone

From: Vineeth Vijayan <[email protected]>

[ Upstream commit 89c0c62e947a01e7a36b54582fd9c9e346170255 ]

Currently, if the device is offline and all the channel paths are either configured or varied offline, the associated subchannel gets unregistered. Don't unregister the subchannel, instead unregister offline device.

---
 drivers/s390/cio/device.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/s390/cio/device.c +4 -1
index c9bc9a6bd73b7..ee4338158ae2e 100644
@@ -1353,6 +1353,7 @@ void ccw_device_set_notoper(struct ccw_device *cdev)
 enum io_sch_action {
 	IO_SCH_UNREG,
 	IO_SCH_ORPH_UNREG,
+	IO_SCH_UNREG_CDEV,
 	IO_SCH_ATTACH,
 	IO_SCH_UNREG_ATTACH,
 	IO_SCH_ORPH_ATTACH,
@@ -1385,7 +1386,7 @@ static enum io_sch_action sch_get_action(struct subchannel *sch)
 	}
 	if ((sch->schib.pmcw.pam & sch->opm) == 0) {
 		if (ccw_device_notify(cdev, CIO_NO_PATH) != NOTIFY_OK)
-			return IO_SCH_UNREG;
+			return IO_SCH_UNREG_CDEV;
 		return IO_SCH_DISC;
 	}
 	if (device_is_disconnected(cdev))
@@ -1447,6 +1448,7 @@ static int io_subchannel_sch_event(struct subchannel *sch, int process)
 	case IO_SCH_ORPH_ATTACH:
 		ccw_device_set_disconnected(cdev);
 		break;
+	case IO_SCH_UNREG_CDEV:
 	case IO_SCH_UNREG_ATTACH:
 	case IO_SCH_UNREG:
 		if (!cdev)
@@ -1480,6 +1482,7 @@ static int io_subchannel_sch_event(struct subchannel *sch, int process)
 		if (rc)
 			goto out;
 		break;
+	case IO_SCH_UNREG_CDEV:
 	case IO_SCH_UNREG_ATTACH:
 		spin_lock_irqsave(sch->lock, flags);
 		if (cdev->private->flags.resuming) {
Signature · 1 line
2.39.2
author maintainer
#1 lore raw

PATCH 2/6 ASoC: nau8824: Add quirk to active-high jack-detect

From: Edson Juliano Drosdeck <[email protected]>

[ Upstream commit e384dba03e3294ce7ea69e4da558e9bf8f0e8946 ]

Add entries for Positivo laptops: CW14Q01P, K1424G, N14ZP74G to the DMI table, so that active-high jack-detect will work properly on these laptops.

---
 sound/soc/codecs/nau8824.c | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)
sound/soc/codecs/nau8824.c +24 -0
index 4f18bb272e929..0ecea65a80b46 100644
@@ -1899,6 +1899,30 @@ static const struct dmi_system_id nau8824_quirk_table[] = {
 		},
 		.driver_data = (void *)(NAU8824_JD_ACTIVE_HIGH),
 	},
+	{
+		/* Positivo CW14Q01P */
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Positivo Tecnologia SA"),
+			DMI_MATCH(DMI_BOARD_NAME, "CW14Q01P"),
+		},
+		.driver_data = (void *)(NAU8824_JD_ACTIVE_HIGH),
+	},
+	{
+		/* Positivo K1424G */
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Positivo Tecnologia SA"),
+			DMI_MATCH(DMI_BOARD_NAME, "K1424G"),
+		},
+		.driver_data = (void *)(NAU8824_JD_ACTIVE_HIGH),
+	},
+	{
+		/* Positivo N14ZP74G */
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Positivo Tecnologia SA"),
+			DMI_MATCH(DMI_BOARD_NAME, "N14ZP74G"),
+		},
+		.driver_data = (void *)(NAU8824_JD_ACTIVE_HIGH),
+	},
 	{}
 };
 
Signature · 1 line
2.39.2
author maintainer
#2 lore raw

PATCH 3/6 drm/exynos: vidi: fix a wrong error return

In reply to Sasha Levin

From: Inki Dae <[email protected]>

[ Upstream commit 4a059559809fd1ddbf16f847c4d2237309c08edf ]

Fix a wrong error return by dropping an error return.

When vidi driver is remvoed, if ctx->raw_edid isn't same as fake_edid_info then only what we have to is to free ctx->raw_edid so that driver removing can work correctly - it's not an error case.

---
 drivers/gpu/drm/exynos/exynos_drm_vidi.c | 2 --
 1 file changed, 2 deletions(-)
drivers/gpu/drm/exynos/exynos_drm_vidi.c +0 -2
index 19697c1362d8f..947c9627c565a 100644
@@ -480,8 +480,6 @@ static int vidi_remove(struct platform_device *pdev)
 	if (ctx->raw_edid != (struct edid *)fake_edid_info) {
 		kfree(ctx->raw_edid);
 		ctx->raw_edid = NULL;
-
-		return -EINVAL;
 	}
 
 	component_del(&pdev->dev, &vidi_component_ops);
Signature · 1 line
2.39.2
author maintainer
#3 lore raw

PATCH 4/6 drm/exynos: fix race condition UAF in exynos_g2d_exec_ioctl

In reply to Sasha Levin

From: Min Li <[email protected]>

[ Upstream commit 48bfd02569f5db49cc033f259e66d57aa6efc9a3 ]

If it is async, runqueue_node is freed in g2d_runqueue_worker on another worker thread. So in extreme cases, if g2d_runqueue_worker runs first, and then executes the following if statement, there will be use-after-free.

---
 drivers/gpu/drm/exynos/exynos_drm_g2d.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/exynos/exynos_drm_g2d.c +1 -1
index f2481a2014bb3..2b7ecc02b2774 100644
@@ -1327,7 +1327,7 @@ int exynos_g2d_exec_ioctl(struct drm_device *drm_dev, void *data,
 	/* Let the runqueue know that there is work to do. */
 	queue_work(g2d->g2d_workq, &g2d->runqueue_work);
 
-	if (runqueue_node->async)
+	if (req->async)
 		goto out;
 
 	wait_for_completion(&runqueue_node->complete);
Signature · 1 line
2.39.2
author maintainer
#4 lore raw

PATCH 5/6 drm/radeon: fix race condition UAF in radeon_gem_set_domain_ioctl

In reply to Sasha Levin

From: Min Li <[email protected]>

[ Upstream commit 982b173a6c6d9472730c3116051977e05d17c8c5 ]

Userspace can race to free the gobj(robj converted from), robj should not be accessed again after drm_gem_object_put, otherwith it will result in use-after-free.

---
 drivers/gpu/drm/radeon/radeon_gem.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)
drivers/gpu/drm/radeon/radeon_gem.c +1 -3
index 27d8e7dd2d067..46f7789693ea0 100644
@@ -377,7 +377,6 @@ int radeon_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 	struct radeon_device *rdev = dev->dev_private;
 	struct drm_radeon_gem_set_domain *args = data;
 	struct drm_gem_object *gobj;
-	struct radeon_bo *robj;
 	int r;
 
 	/* for now if someone requests domain CPU -
@@ -390,13 +389,12 @@ int radeon_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 		up_read(&rdev->exclusive_lock);
 		return -ENOENT;
 	}
-	robj = gem_to_radeon_bo(gobj);
 
 	r = radeon_gem_set_domain(gobj, args->read_domains, args->write_domain);
 
 	drm_gem_object_put_unlocked(gobj);
 	up_read(&rdev->exclusive_lock);
-	r = radeon_gem_handle_lockup(robj->rdev, r);
+	r = radeon_gem_handle_lockup(rdev, r);
 	return r;
 }
 
Signature · 1 line
2.39.2
author maintainer
#5 lore raw

PATCH 6/6 ext4: only check dquot_initialize_needed() when debugging

In reply to Sasha Levin

From: Theodore Ts'o <[email protected]>

[ Upstream commit dea9d8f7643fab07bf89a1155f1f94f37d096a5e ]

ext4_xattr_block_set() relies on its caller to call dquot_initialize() on the inode. To assure that this has happened there are WARN_ON checks. Unfortunately, this is subject to false positives if there is an antagonist thread which is flipping the file system at high rates between r/o and rw. So only do the check if EXT4_XATTR_DEBUG is enabled.

---
 fs/ext4/xattr.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
fs/ext4/xattr.c +4 -2
index c2786bee4cb6e..f8a8807c20977 100644
@@ -2041,8 +2041,9 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode,
 			else {
 				u32 ref;
 
+#ifdef EXT4_XATTR_DEBUG
 				WARN_ON_ONCE(dquot_initialize_needed(inode));
-
+#endif
 				/* The old block is released after updating
 				   the inode. */
 				error = dquot_alloc_block(inode,
@@ -2104,8 +2105,9 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode,
 			/* We need to allocate a new block */
 			ext4_fsblk_t goal, block;
 
+#ifdef EXT4_XATTR_DEBUG
 			WARN_ON_ONCE(dquot_initialize_needed(inode));
-
+#endif
 			goal = ext4_group_first_block_no(sb,
 						EXT4_I(inode)->i_block_group);
 			block = ext4_new_meta_blocks(handle, inode, goal, 0,
Signature · 1 line
2.39.2