6.12.28-rc1 review

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

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

From: Cong Wang <[email protected]>

commit a7a15f39c682ac4268624da2abdb9114bdde96d5 upstream.

est_qlen_notify() deletes its class from its active list with list_del() when qlen is 0, therefore, it is not idempotent and not friendly to its callers, like fq_codel_dequeue().

Let's make it idempotent to ease qdisc_tree_reduce_backlog() callers' life. Also change other list_del()'s to list_del_init() just to be extra safe.

---
 net/sched/sch_ets.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

@@ -298,7 +298,7 @@ static void ets_class_qlen_notify(struct
 	 * to remove them.
 	 */
 	if (!ets_class_is_strict(q, cl) && sch->q.qlen)
-		list_del(&cl->alist);
+		list_del_init(&cl->alist);
 }
 
 static int ets_class_dump(struct Qdisc *sch, unsigned long arg,

@@ -491,7 +491,7 @@ static struct sk_buff *ets_qdisc_dequeue
 			if (unlikely(!skb))
 				goto out;
 			if (cl->qdisc->q.qlen == 0)
-				list_del(&cl->alist);
+				list_del_init(&cl->alist);
 			return ets_qdisc_dequeue_skb(sch, skb);
 		}
 

@@ -660,7 +660,7 @@ static int ets_qdisc_change(struct Qdisc
 	}
 	for (i = q->nbands; i < oldbands; i++) {
 		if (i >= q->nstrict && q->classes[i].qdisc->q.qlen)
-			list_del(&q->classes[i].alist);
+			list_del_init(&q->classes[i].alist);
 		qdisc_tree_flush_backlog(q->classes[i].qdisc);
 	}
 	WRITE_ONCE(q->nstrict, nstrict);

@@ -716,7 +716,7 @@ static void ets_qdisc_reset(struct Qdisc
 
 	for (band = q->nstrict; band < q->nbands; band++) {
 		if (q->classes[band].qdisc->q.qlen)
-			list_del(&q->classes[band].alist);
+			list_del_init(&q->classes[band].alist);
 	}
 	for (band = 0; band < q->nbands; band++)
 		qdisc_reset(q->classes[band].qdisc);

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

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

From: Niranjana Vishwanathapura <[email protected]>

commit 262de94a3a7ef23c326534b3d9483602b7af841e upstream.

The RCU_MODE_FIXED_SLICE_CCS_MODE setting is not getting invoked in the gt reset path after the ccs_mode setting by the user. Add it to engine register update list (in hw_engine_setup_default_state()) which ensures it gets set in the gt reset and engine reset paths.

v2: Add register update to engine list to ensure it gets updated after engine reset also.

---
 drivers/gpu/drm/xe/xe_hw_engine.c |   12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

@@ -381,12 +381,6 @@ xe_hw_engine_setup_default_lrc_state(str
 				 blit_cctl_val,
 				 XE_RTP_ACTION_FLAG(ENGINE_BASE)))
 		},
-		/* Use Fixed slice CCS mode */
-		{ XE_RTP_NAME("RCU_MODE_FIXED_SLICE_CCS_MODE"),
-		  XE_RTP_RULES(FUNC(xe_hw_engine_match_fixed_cslice_mode)),
-		  XE_RTP_ACTIONS(FIELD_SET(RCU_MODE, RCU_MODE_FIXED_SLICE_CCS_MODE,
-					   RCU_MODE_FIXED_SLICE_CCS_MODE))
-		},
 		/* Disable WMTP if HW doesn't support it */
 		{ XE_RTP_NAME("DISABLE_WMTP_ON_UNSUPPORTED_HW"),
 		  XE_RTP_RULES(FUNC(xe_rtp_cfeg_wmtp_disabled)),

@@ -454,6 +448,12 @@ hw_engine_setup_default_state(struct xe_
 		  XE_RTP_ACTIONS(SET(CSFE_CHICKEN1(0), CS_PRIORITY_MEM_READ,
 				     XE_RTP_ACTION_FLAG(ENGINE_BASE)))
 		},
+		/* Use Fixed slice CCS mode */
+		{ XE_RTP_NAME("RCU_MODE_FIXED_SLICE_CCS_MODE"),
+		  XE_RTP_RULES(FUNC(xe_hw_engine_match_fixed_cslice_mode)),
+		  XE_RTP_ACTIONS(FIELD_SET(RCU_MODE, RCU_MODE_FIXED_SLICE_CCS_MODE,
+					   RCU_MODE_FIXED_SLICE_CCS_MODE))
+		},
 		{}
 	};
 

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

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

From: Cristian Marussi <[email protected]>

[ Upstream commit 9ca67840c0ddf3f39407339624cef824a4f27599 ]

Using device_find_child() to lookup the proper SCMI device to destroy causes an unbalance in device refcount, since device_find_child() calls an implicit get_device(): this, in turns, inhibits the call of the provided release methods upon devices destruction.

As a consequence, one of the structures that is not freed properly upon destruction is the internal struct device_private dev->p populated by the drivers subsystem core.

KMemleak detects this situation since loading/unloding some SCMI driver causes related devices to be created/destroyed without calling any device_release method.

unreferenced object 0xffff00000f583800 (size 512):
comm "insmod", pid 227, jiffies 4294912190 hex dump (first 32 bytes):

    00 00 00 00 ad 4e ad de ff ff ff ff 00 00 00 00  .....N..........
    ff ff ff ff ff ff ff ff 60 36 1d 8a 00 80 ff ff  ........`6......
  backtrace (crc 114e2eed):
    kmemleak_alloc+0xbc/0xd8
    __kmalloc_cache_noprof+0x2dc/0x398
    device_add+0x954/0x12d0
    device_register+0x28/0x40
    __scmi_device_create.part.0+0x1bc/0x380
    scmi_device_create+0x2d0/0x390
    scmi_create_protocol_devices+0x74/0xf8
    scmi_device_request_notifier+0x1f8/0x2a8
    notifier_call_chain+0x110/0x3b0
    blocking_notifier_call_chain+0x70/0xb0
    scmi_driver_register+0x350/0x7f0
    0xffff80000a3b3038
    do_one_initcall+0x12c/0x730
    do_init_module+0x1dc/0x640
    load_module+0x4b20/0x5b70
    init_module_from_file+0xec/0x158

$ ./scripts/faddr2line ./vmlinux device_add+0x954/0x12d0
device_add+0x954/0x12d0:
kmalloc_noprof at include/linux/slab.h:901
(inlined by) kzalloc_noprof at include/linux/slab.h:1037
(inlined by) device_private_init at drivers/base/core.c:3510
(inlined by) device_add at drivers/base/core.c:3561

Balance device refcount by issuing a put_device() on devices found via device_find_child().

Message-Id: <[email protected]>

---
 drivers/firmware/arm_scmi/bus.c | 3 +++
 1 file changed, 3 insertions(+)

index 157172a5f2b57..782c9bec8361c 100644
@@ -260,6 +260,9 @@ static struct scmi_device *scmi_child_dev_find(struct device *parent,
 	if (!dev)
 		return NULL;
 
+	/* Drop the refcnt bumped implicitly by device_find_child */
+	put_device(dev);
+
 	return to_scmi_dev(dev);
 }
 

2.39.5

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

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

From: Sudeep Holla <[email protected]>

[ Upstream commit 4567bdaaaaa1744da3d7da07d9aca2f941f5b4e5 ]

Completion of the FFA_PARTITION_INFO_GET ABI transfers the ownership of the caller’s Rx buffer from the producer(typically partition mnager) to the consumer(this driver/OS). FFA_RX_RELEASE transfers the ownership from the consumer back to the producer.

However, when we set the flag to just return the count of partitions deployed in the system corresponding to the specified UUID while invoking FFA_PARTITION_INFO_GET, the Rx buffer ownership shouldn't be transferred to this driver. We must be able to skip transferring back the ownership to the partition manager when we request just to get the count of the partitions as the buffers are not acquired in this case.

Firmware may return FFA_RET_DENIED or other error for the ffa_rx_release() in such cases.

Message-Id: <[email protected]>

---
 drivers/firmware/arm_ffa/driver.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

index 2c2ec3c35f156..dce448687e28e 100644
@@ -280,7 +280,8 @@ __ffa_partition_info_get(u32 uuid0, u32 uuid1, u32 uuid2, u32 uuid3,
 			memcpy(buffer + idx, drv_info->rx_buffer + idx * sz,
 			       buf_sz);
 
-	ffa_rx_release();
+	if (!(flags & PARTITION_INFO_GET_RETURN_COUNT_ONLY))
+		ffa_rx_release();
 
 	mutex_unlock(&drv_info->rx_lock);
 

2.39.5

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

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

From: Richard Zhu <[email protected]>

[ Upstream commit 02e4232998db357bb8199778722d81ffcff0cb98 ]

Correct the range of PCIe app-reg region from 0x2000 to 0x4000 refer to SerDes_SS memory map of i.MX95 Rerference Manual.

---
 arch/arm64/boot/dts/freescale/imx95.dtsi | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

index 40cbb071f265c..f904d6b1c84bf 100644
@@ -1478,7 +1478,7 @@
 			reg = <0 0x4c300000 0 0x10000>,
 			      <0 0x60100000 0 0xfe00000>,
 			      <0 0x4c360000 0 0x10000>,
-			      <0 0x4c340000 0 0x2000>;
+			      <0 0x4c340000 0 0x4000>;
 			reg-names = "dbi", "config", "atu", "app";
 			ranges = <0x81000000 0x0 0x00000000 0x0 0x6ff00000 0 0x00100000>,
 				 <0x82000000 0x0 0x10000000 0x9 0x10000000 0 0x10000000>;

@@ -1518,7 +1518,7 @@
 			reg = <0 0x4c300000 0 0x10000>,
 			      <0 0x4c360000 0 0x1000>,
 			      <0 0x4c320000 0 0x1000>,
-			      <0 0x4c340000 0 0x2000>,
+			      <0 0x4c340000 0 0x4000>,
 			      <0 0x4c370000 0 0x10000>,
 			      <0x9 0 1 0>;
 			reg-names = "dbi","atu", "dbi2", "app", "dma", "addr_space";

@@ -1545,7 +1545,7 @@
 			reg = <0 0x4c380000 0 0x10000>,
 			      <8 0x80100000 0 0xfe00000>,
 			      <0 0x4c3e0000 0 0x10000>,
-			      <0 0x4c3c0000 0 0x2000>;
+			      <0 0x4c3c0000 0 0x4000>;
 			reg-names = "dbi", "config", "atu", "app";
 			ranges = <0x81000000 0 0x00000000 0x8 0x8ff00000 0 0x00100000>,
 				 <0x82000000 0 0x10000000 0xa 0x10000000 0 0x10000000>;

@@ -1585,7 +1585,7 @@
 			reg = <0 0x4c380000 0 0x10000>,
 			      <0 0x4c3e0000 0 0x1000>,
 			      <0 0x4c3a0000 0 0x1000>,
-			      <0 0x4c3c0000 0 0x2000>,
+			      <0 0x4c3c0000 0 0x4000>,
 			      <0 0x4c3f0000 0 0x10000>,
 			      <0xa 0 1 0>;
 			reg-names = "dbi", "atu", "dbi2", "app", "dma", "addr_space";

2.39.5

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

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

From: Sébastien Szymanski <[email protected]>

[ Upstream commit 6e1a7bc8382b0d4208258f7d2a4474fae788dd90 ]

Commit c7e73b5051d6 ("ARM: imx: mach-imx6ul: remove 14x14 EVK specific PHY fixup") removed a PHY fixup that setted the clock mode and the LED mode.
Make the Ethernet interface work again by doing as advised in the commit's log, set clock mode and the LED mode in the device tree.

---
 arch/arm/boot/dts/nxp/imx/imx6ul-imx6ull-opos6ul.dtsi | 3 +++
 1 file changed, 3 insertions(+)

index f2386dcb9ff2c..dda4fa91b2f2c 100644
@@ -40,6 +40,9 @@
 			reg = <1>;
 			interrupt-parent = <&gpio4>;
 			interrupts = <16 IRQ_TYPE_LEVEL_LOW>;
+			micrel,led-mode = <1>;
+			clocks = <&clks IMX6UL_CLK_ENET_REF>;
+			clock-names = "rmii-ref";
 			status = "okay";
 		};
 	};

2.39.5

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

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

From: Christian Bruel <[email protected]>

[ Upstream commit de2b2107d5a41a91ab603e135fb6e408abbee28e ]

Use gic-400 compatible and remove address-cells = <1> on aarch64

---
 arch/arm64/boot/dts/st/stm32mp251.dtsi | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

index 1167cf63d7e87..896d92032284b 100644
@@ -114,9 +114,8 @@
 	};
 
 	intc: interrupt-controller@4ac00000 {
-		compatible = "arm,cortex-a7-gic";
+		compatible = "arm,gic-400";
 		#interrupt-cells = <3>;
-		#address-cells = <1>;
 		interrupt-controller;
 		reg = <0x0 0x4ac10000 0x0 0x1000>,
 		      <0x0 0x4ac20000 0x0 0x2000>,

2.39.5

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

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

From: Christian Bruel <[email protected]>

[ Upstream commit 06c231fe953a26f4bc9d7a37ba1b9b288a59c7c2 ]

Adjust the size of 8kB GIC regions to 128kB so that each 4kB is mapped 16 times over a 64kB region.
The offset is then adjusted in the irq-gic driver.

see commit 12e14066f4835 ("irqchip/GIC: Add workaround for aliased GIC400")

---
 arch/arm64/boot/dts/st/stm32mp251.dtsi | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

index 896d92032284b..cd9b92144a42c 100644
@@ -118,9 +118,9 @@
 		#interrupt-cells = <3>;
 		interrupt-controller;
 		reg = <0x0 0x4ac10000 0x0 0x1000>,
-		      <0x0 0x4ac20000 0x0 0x2000>,
-		      <0x0 0x4ac40000 0x0 0x2000>,
-		      <0x0 0x4ac60000 0x0 0x2000>;
+		      <0x0 0x4ac20000 0x0 0x20000>,
+		      <0x0 0x4ac40000 0x0 0x20000>,
+		      <0x0 0x4ac60000 0x0 0x20000>;
 	};
 
 	psci {

2.39.5

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

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

From: Christian Hewitt <[email protected]>

[ Upstream commit f37bb5486ea536c1d61df89feeaeff3f84f0b560 ]

This reverts commit bfbc68e.

The patch does permit the offending YUV420 @ 59.94 phy_freq and vclk_freq mode to match in calculations. It also results in all fractional rates being unavailable for use. This was unintended and requires the patch to be reverted.

---
 drivers/gpu/drm/meson/meson_vclk.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

index 2a942dc6a6dc2..2a82119eb58ed 100644
@@ -790,13 +790,13 @@ meson_vclk_vic_supported_freq(struct meson_drm *priv, unsigned int phy_freq,
 				 FREQ_1000_1001(params[i].pixel_freq));
 		DRM_DEBUG_DRIVER("i = %d phy_freq = %d alt = %d\n",
 				 i, params[i].phy_freq,
-				 FREQ_1000_1001(params[i].phy_freq/1000)*1000);
+				 FREQ_1000_1001(params[i].phy_freq/10)*10);
 		/* Match strict frequency */
 		if (phy_freq == params[i].phy_freq &&
 		    vclk_freq == params[i].vclk_freq)
 			return MODE_OK;
 		/* Match 1000/1001 variant */
-		if (phy_freq == (FREQ_1000_1001(params[i].phy_freq/1000)*1000) &&
+		if (phy_freq == (FREQ_1000_1001(params[i].phy_freq/10)*10) &&
 		    vclk_freq == FREQ_1000_1001(params[i].vclk_freq))
 			return MODE_OK;
 	}

@@ -1070,7 +1070,7 @@ void meson_vclk_setup(struct meson_drm *priv, unsigned int target,
 
 	for (freq = 0 ; params[freq].pixel_freq ; ++freq) {
 		if ((phy_freq == params[freq].phy_freq ||
-		     phy_freq == FREQ_1000_1001(params[freq].phy_freq/1000)*1000) &&
+		     phy_freq == FREQ_1000_1001(params[freq].phy_freq/10)*10) &&
 		    (vclk_freq == params[freq].vclk_freq ||
 		     vclk_freq == FREQ_1000_1001(params[freq].vclk_freq))) {
 			if (vclk_freq != params[freq].vclk_freq)

2.39.5

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

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

From: Shyam Saini <[email protected]>

[ Upstream commit bbc9462f0cb0c8917a4908e856731708f0cee910 ]

The locate_module_kobject() function looks up an existing module_kobject for a given module name. If it cannot find the corresponding module_kobject, it creates one for the given name.

This commit renames locate_module_kobject() to lookup_or_create_module_kobject() to better describe its operations.

This doesn't change anything functionality wise.

Stable-dep-of: f95bbfe18512 ("drivers: base: handle module_kobject creation")

---
 kernel/params.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

index 2e447f8ae183e..d62a116918cd0 100644
@@ -763,7 +763,7 @@ void destroy_params(const struct kernel_param *params, unsigned num)
 			params[i].ops->free(params[i].arg);
 }
 
-static struct module_kobject * __init locate_module_kobject(const char *name)
+static struct module_kobject * __init lookup_or_create_module_kobject(const char *name)
 {
 	struct module_kobject *mk;
 	struct kobject *kobj;

@@ -805,7 +805,7 @@ static void __init kernel_add_sysfs_param(const char *name,
 	struct module_kobject *mk;
 	int err;
 
-	mk = locate_module_kobject(name);
+	mk = lookup_or_create_module_kobject(name);
 	if (!mk)
 		return;
 

@@ -876,7 +876,7 @@ static void __init version_sysfs_builtin(void)
 	int err;
 
 	for (vattr = __start___modver; vattr < __stop___modver; vattr++) {
-		mk = locate_module_kobject(vattr->module_name);
+		mk = lookup_or_create_module_kobject(vattr->module_name);
 		if (mk) {
 			err = sysfs_create_file(&mk->kobj, &vattr->mattr.attr);
 			WARN_ON_ONCE(err);

2.39.5

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

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

From: Shyam Saini <[email protected]>

[ Upstream commit 7c76c813cfc42a7376378a0c4b7250db2eebab81 ]

lookup_or_create_module_kobject() is marked as static and __init, to make it global drop static keyword.
Since this function can be called from non-init code, use __modinit instead of __init, __modinit marker will make it __init if CONFIG_MODULES is not defined.

Stable-dep-of: f95bbfe18512 ("drivers: base: handle module_kobject creation")

---
 include/linux/module.h | 2 ++
 kernel/params.c        | 2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

index 88ecc5e9f5230..82a9527d43c76 100644
@@ -162,6 +162,8 @@ extern void cleanup_module(void);
 #define __INITRODATA_OR_MODULE __INITRODATA
 #endif /*CONFIG_MODULES*/
 
+struct module_kobject *lookup_or_create_module_kobject(const char *name);
+
 /* Generic info of form tag = "info" */
 #define MODULE_INFO(tag, info) __MODULE_INFO(tag, tag, info)
 

index d62a116918cd0..33b2985b31c7f 100644
@@ -763,7 +763,7 @@ void destroy_params(const struct kernel_param *params, unsigned num)
 			params[i].ops->free(params[i].arg);
 }
 
-static struct module_kobject * __init lookup_or_create_module_kobject(const char *name)
+struct module_kobject __modinit * lookup_or_create_module_kobject(const char *name)
 {
 	struct module_kobject *mk;
 	struct kobject *kobj;

2.39.5

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

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

From: Shyam Saini <[email protected]>

[ Upstream commit f95bbfe18512c5c018720468959edac056a17196 ]

module_add_driver() relies on module_kset list for /sys/module/<built-in-module>/drivers directory creation.

Since,
commit 96a1a2412acba ("kernel/params.c: defer most of param_sysfs_init() to late_initcall time") drivers which are initialized from subsys_initcall() or any other
higher precedence initcall couldn't find the related kobject entry
in the module_kset list because module_kset is not fully populated
by the time module_add_driver() refers it. As a consequence,
module_add_driver() returns early without calling make_driver_name().
Therefore, /sys/module/<built-in-module>/drivers is never created.

Fix this issue by letting module_add_driver() handle module_kobject creation itself.

---
 drivers/base/module.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

index 5bc71bea883a0..218aaa0964552 100644
@@ -42,16 +42,13 @@ int module_add_driver(struct module *mod, const struct device_driver *drv)
 	if (mod)
 		mk = &mod->mkobj;
 	else if (drv->mod_name) {
-		struct kobject *mkobj;
-
-		/* Lookup built-in module entry in /sys/modules */
-		mkobj = kset_find_obj(module_kset, drv->mod_name);
-		if (mkobj) {
-			mk = container_of(mkobj, struct module_kobject, kobj);
+		/* Lookup or create built-in module entry in /sys/modules */
+		mk = lookup_or_create_module_kobject(drv->mod_name);
+		if (mk) {
 			/* remember our module structure */
 			drv->p->mkobj = mk;
-			/* kset_find_obj took a reference */
-			kobject_put(mkobj);
+			/* lookup_or_create_module_kobject took a reference */
+			kobject_put(&mk->kobj);
 		}
 	}
 

2.39.5

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

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

From: Mario Limonciello <[email protected]>

[ Upstream commit 6b675ab8efbf2bcee25be29e865455c56e246401 ]

[Why]
Guards automatically release mutex when it goes out of scope making code easier to follow.

[How]
Replace all use of mutex_lock()/mutex_unlock() with guard(mutex).

Stable-dep-of: be593d9d91c5 ("drm/amd/display: Fix slab-use-after-free in hdcp")

---
 .../amd/display/amdgpu_dm/amdgpu_dm_hdcp.c    | 37 +++++--------------
 1 file changed, 10 insertions(+), 27 deletions(-)

index c0dc232440490..53796a74b7a65 100644
@@ -172,7 +172,7 @@ void hdcp_update_display(struct hdcp_workqueue *hdcp_work,
 	struct mod_hdcp_display_adjustment display_adjust;
 	unsigned int conn_index = aconnector->base.index;
 
-	mutex_lock(&hdcp_w->mutex);
+	guard(mutex)(&hdcp_w->mutex);
 	hdcp_w->aconnector[conn_index] = aconnector;
 
 	memset(&link_adjust, 0, sizeof(link_adjust));

@@ -209,7 +209,6 @@ void hdcp_update_display(struct hdcp_workqueue *hdcp_work,
 	mod_hdcp_update_display(&hdcp_w->hdcp, conn_index, &link_adjust, &display_adjust, &hdcp_w->output);
 
 	process_output(hdcp_w);
-	mutex_unlock(&hdcp_w->mutex);
 }
 
 static void hdcp_remove_display(struct hdcp_workqueue *hdcp_work,

@@ -220,7 +219,7 @@ static void hdcp_remove_display(struct hdcp_workqueue *hdcp_work,
 	struct drm_connector_state *conn_state = aconnector->base.state;
 	unsigned int conn_index = aconnector->base.index;
 
-	mutex_lock(&hdcp_w->mutex);
+	guard(mutex)(&hdcp_w->mutex);
 	hdcp_w->aconnector[conn_index] = aconnector;
 
 	/* the removal of display will invoke auth reset -> hdcp destroy and

@@ -239,7 +238,6 @@ static void hdcp_remove_display(struct hdcp_workqueue *hdcp_work,
 	mod_hdcp_remove_display(&hdcp_w->hdcp, aconnector->base.index, &hdcp_w->output);
 
 	process_output(hdcp_w);
-	mutex_unlock(&hdcp_w->mutex);
 }
 
 void hdcp_reset_display(struct hdcp_workqueue *hdcp_work, unsigned int link_index)

@@ -247,7 +245,7 @@ void hdcp_reset_display(struct hdcp_workqueue *hdcp_work, unsigned int link_inde
 	struct hdcp_workqueue *hdcp_w = &hdcp_work[link_index];
 	unsigned int conn_index;
 
-	mutex_lock(&hdcp_w->mutex);
+	guard(mutex)(&hdcp_w->mutex);
 
 	mod_hdcp_reset_connection(&hdcp_w->hdcp,  &hdcp_w->output);
 

@@ -259,8 +257,6 @@ void hdcp_reset_display(struct hdcp_workqueue *hdcp_work, unsigned int link_inde
 	}
 
 	process_output(hdcp_w);
-
-	mutex_unlock(&hdcp_w->mutex);
 }
 
 void hdcp_handle_cpirq(struct hdcp_workqueue *hdcp_work, unsigned int link_index)

@@ -277,7 +273,7 @@ static void event_callback(struct work_struct *work)
 	hdcp_work = container_of(to_delayed_work(work), struct hdcp_workqueue,
 				 callback_dwork);
 
-	mutex_lock(&hdcp_work->mutex);
+	guard(mutex)(&hdcp_work->mutex);
 
 	cancel_delayed_work(&hdcp_work->callback_dwork);
 

@@ -285,8 +281,6 @@ static void event_callback(struct work_struct *work)
 			       &hdcp_work->output);
 
 	process_output(hdcp_work);
-
-	mutex_unlock(&hdcp_work->mutex);
 }
 
 static void event_property_update(struct work_struct *work)

@@ -323,7 +317,7 @@ static void event_property_update(struct work_struct *work)
 			continue;
 
 		drm_modeset_lock(&dev->mode_config.connection_mutex, NULL);
-		mutex_lock(&hdcp_work->mutex);
+		guard(mutex)(&hdcp_work->mutex);
 
 		if (conn_state->commit) {
 			ret = wait_for_completion_interruptible_timeout(&conn_state->commit->hw_done,

@@ -355,7 +349,6 @@ static void event_property_update(struct work_struct *work)
 			drm_hdcp_update_content_protection(connector,
 							   DRM_MODE_CONTENT_PROTECTION_DESIRED);
 		}
-		mutex_unlock(&hdcp_work->mutex);
 		drm_modeset_unlock(&dev->mode_config.connection_mutex);
 	}
 }

@@ -368,7 +361,7 @@ static void event_property_validate(struct work_struct *work)
 	struct amdgpu_dm_connector *aconnector;
 	unsigned int conn_index;
 
-	mutex_lock(&hdcp_work->mutex);
+	guard(mutex)(&hdcp_work->mutex);
 
 	for (conn_index = 0; conn_index < AMDGPU_DM_MAX_DISPLAY_INDEX;
 	     conn_index++) {

@@ -408,8 +401,6 @@ static void event_property_validate(struct work_struct *work)
 			schedule_work(&hdcp_work->property_update_work);
 		}
 	}
-
-	mutex_unlock(&hdcp_work->mutex);
 }
 
 static void event_watchdog_timer(struct work_struct *work)

@@ -420,7 +411,7 @@ static void event_watchdog_timer(struct work_struct *work)
 				 struct hdcp_workqueue,
 				      watchdog_timer_dwork);
 
-	mutex_lock(&hdcp_work->mutex);
+	guard(mutex)(&hdcp_work->mutex);
 
 	cancel_delayed_work(&hdcp_work->watchdog_timer_dwork);
 

@@ -429,8 +420,6 @@ static void event_watchdog_timer(struct work_struct *work)
 			       &hdcp_work->output);
 
 	process_output(hdcp_work);
-
-	mutex_unlock(&hdcp_work->mutex);
 }
 
 static void event_cpirq(struct work_struct *work)

@@ -439,13 +428,11 @@ static void event_cpirq(struct work_struct *work)
 
 	hdcp_work = container_of(work, struct hdcp_workqueue, cpirq_work);
 
-	mutex_lock(&hdcp_work->mutex);
+	guard(mutex)(&hdcp_work->mutex);
 
 	mod_hdcp_process_event(&hdcp_work->hdcp, MOD_HDCP_EVENT_CPIRQ, &hdcp_work->output);
 
 	process_output(hdcp_work);
-
-	mutex_unlock(&hdcp_work->mutex);
 }
 
 void hdcp_destroy(struct kobject *kobj, struct hdcp_workqueue *hdcp_work)

@@ -479,7 +466,7 @@ static bool enable_assr(void *handle, struct dc_link *link)
 
 	dtm_cmd = (struct ta_dtm_shared_memory *)psp->dtm_context.context.mem_context.shared_buf;
 
-	mutex_lock(&psp->dtm_context.mutex);
+	guard(mutex)(&psp->dtm_context.mutex);
 	memset(dtm_cmd, 0, sizeof(struct ta_dtm_shared_memory));
 
 	dtm_cmd->cmd_id = TA_DTM_COMMAND__TOPOLOGY_ASSR_ENABLE;

@@ -494,8 +481,6 @@ static bool enable_assr(void *handle, struct dc_link *link)
 		res = false;
 	}
 
-	mutex_unlock(&psp->dtm_context.mutex);
-
 	return res;
 }
 

@@ -557,13 +542,11 @@ static void update_config(void *handle, struct cp_psp_stream_config *config)
 			 (!!aconnector->base.state) ?
 			 aconnector->base.state->hdcp_content_type : -1);
 
-	mutex_lock(&hdcp_w->mutex);
+	guard(mutex)(&hdcp_w->mutex);
 
 	mod_hdcp_add_display(&hdcp_w->hdcp, link, display, &hdcp_w->output);
 
 	process_output(hdcp_w);
-	mutex_unlock(&hdcp_w->mutex);
-
 }
 
 /**

2.39.5

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

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

From: Chris Bainbridge <[email protected]>

[ Upstream commit be593d9d91c5a3a363d456b9aceb71029aeb3f1d ]

The HDCP code in amdgpu_dm_hdcp.c copies pointers to amdgpu_dm_connector objects without incrementing the kref reference counts. When using a USB-C dock, and the dock is unplugged, the corresponding amdgpu_dm_connector objects are freed, creating dangling pointers in the HDCP code. When the dock is plugged back, the dangling pointers are dereferenced, resulting in a slab-use-after-free:

[   66.775837] BUG: KASAN: slab-use-after-free in event_property_validate+0x42f/0x6c0 [amdgpu]
[   66.776171] Read of size 4 at addr ffff888127804120 by task kworker/0:1/10

[   66.776179] CPU: 0 UID: 0 PID: 10 Comm: kworker/0:1 Not tainted 6.14.0-rc7-00180-g54505f727a38-dirty #233
[   66.776183] Hardware name: HP HP Pavilion Aero Laptop 13-be0xxx/8916, BIOS F.17 12/18/2024
[   66.776186] Workqueue: events event_property_validate [amdgpu]
[   66.776494] Call Trace:
[   66.776496]  <TASK>
[   66.776497]  dump_stack_lvl+0x70/0xa0
[   66.776504]  print_report+0x175/0x555
[   66.776507]  ? __virt_addr_valid+0x243/0x450
[   66.776510]  ? kasan_complete_mode_report_info+0x66/0x1c0
[   66.776515]  kasan_report+0xeb/0x1c0
[   66.776518]  ? event_property_validate+0x42f/0x6c0 [amdgpu]
[   66.776819]  ? event_property_validate+0x42f/0x6c0 [amdgpu]
[   66.777121]  __asan_report_load4_noabort+0x14/0x20
[   66.777124]  event_property_validate+0x42f/0x6c0 [amdgpu]
[   66.777342]  ? __lock_acquire+0x6b40/0x6b40
[   66.777347]  ? enable_assr+0x250/0x250 [amdgpu]
[   66.777571]  process_one_work+0x86b/0x1510
[   66.777575]  ? pwq_dec_nr_in_flight+0xcf0/0xcf0
[   66.777578]  ? assign_work+0x16b/0x280
[   66.777580]  ? lock_is_held_type+0xa3/0x130
[   66.777583]  worker_thread+0x5c0/0xfa0
[   66.777587]  ? process_one_work+0x1510/0x1510
[   66.777588]  kthread+0x3a2/0x840
[   66.777591]  ? kthread_is_per_cpu+0xd0/0xd0
[   66.777594]  ? trace_hardirqs_on+0x4f/0x60
[   66.777597]  ? _raw_spin_unlock_irq+0x27/0x60
[   66.777599]  ? calculate_sigpending+0x77/0xa0
[   66.777602]  ? kthread_is_per_cpu+0xd0/0xd0
[   66.777605]  ret_from_fork+0x40/0x90
[   66.777607]  ? kthread_is_per_cpu+0xd0/0xd0
[   66.777609]  ret_from_fork_asm+0x11/0x20
[   66.777614]  </TASK>

[   66.777643] Allocated by task 10:
[   66.777646]  kasan_save_stack+0x39/0x60
[   66.777649]  kasan_save_track+0x14/0x40
[   66.777652]  kasan_save_alloc_info+0x37/0x50
[   66.777655]  __kasan_kmalloc+0xbb/0xc0
[   66.777658]  __kmalloc_cache_noprof+0x1c8/0x4b0
[   66.777661]  dm_dp_add_mst_connector+0xdd/0x5c0 [amdgpu]
[   66.777880]  drm_dp_mst_port_add_connector+0x47e/0x770 [drm_display_helper]
[   66.777892]  drm_dp_send_link_address+0x1554/0x2bf0 [drm_display_helper]
[   66.777901]  drm_dp_check_and_send_link_address+0x187/0x1f0 [drm_display_helper]
[   66.777909]  drm_dp_mst_link_probe_work+0x2b8/0x410 [drm_display_helper]
[   66.777917]  process_one_work+0x86b/0x1510
[   66.777919]  worker_thread+0x5c0/0xfa0
[   66.777922]  kthread+0x3a2/0x840
[   66.777925]  ret_from_fork+0x40/0x90
[   66.777927]  ret_from_fork_asm+0x11/0x20

[   66.777932] Freed by task 1713:
[   66.777935]  kasan_save_stack+0x39/0x60
[   66.777938]  kasan_save_track+0x14/0x40
[   66.777940]  kasan_save_free_info+0x3b/0x60
[   66.777944]  __kasan_slab_free+0x52/0x70
[   66.777946]  kfree+0x13f/0x4b0
[   66.777949]  dm_dp_mst_connector_destroy+0xfa/0x150 [amdgpu]
[   66.778179]  drm_connector_free+0x7d/0xb0
[   66.778184]  drm_mode_object_put.part.0+0xee/0x160
[   66.778188]  drm_mode_object_put+0x37/0x50
[   66.778191]  drm_atomic_state_default_clear+0x220/0xd60
[   66.778194]  __drm_atomic_state_free+0x16e/0x2a0
[   66.778197]  drm_mode_atomic_ioctl+0x15ed/0x2ba0
[   66.778200]  drm_ioctl_kernel+0x17a/0x310
[   66.778203]  drm_ioctl+0x584/0xd10
[   66.778206]  amdgpu_drm_ioctl+0xd2/0x1c0 [amdgpu]
[   66.778375]  __x64_sys_ioctl+0x139/0x1a0
[   66.778378]  x64_sys_call+0xee7/0xfb0
[   66.778381]  do_syscall_64+0x87/0x140
[   66.778385]  entry_SYSCALL_64_after_hwframe+0x4b/0x53

Fix this by properly incrementing and decrementing the reference counts when making and deleting copies of the amdgpu_dm_connector pointers.

(Mario: rebase on current code and update fixes tag)

---
 .../amd/display/amdgpu_dm/amdgpu_dm_hdcp.c    | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)

index 53796a74b7a65..10ba4d7bf6325 100644
@@ -173,6 +173,9 @@ void hdcp_update_display(struct hdcp_workqueue *hdcp_work,
 	unsigned int conn_index = aconnector->base.index;
 
 	guard(mutex)(&hdcp_w->mutex);
+	drm_connector_get(&aconnector->base);
+	if (hdcp_w->aconnector[conn_index])
+		drm_connector_put(&hdcp_w->aconnector[conn_index]->base);
 	hdcp_w->aconnector[conn_index] = aconnector;
 
 	memset(&link_adjust, 0, sizeof(link_adjust));

@@ -220,7 +223,6 @@ static void hdcp_remove_display(struct hdcp_workqueue *hdcp_work,
 	unsigned int conn_index = aconnector->base.index;
 
 	guard(mutex)(&hdcp_w->mutex);
-	hdcp_w->aconnector[conn_index] = aconnector;
 
 	/* the removal of display will invoke auth reset -> hdcp destroy and
 	 * we'd expect the Content Protection (CP) property changed back to

@@ -236,7 +238,10 @@ static void hdcp_remove_display(struct hdcp_workqueue *hdcp_work,
 	}
 
 	mod_hdcp_remove_display(&hdcp_w->hdcp, aconnector->base.index, &hdcp_w->output);
-
+	if (hdcp_w->aconnector[conn_index]) {
+		drm_connector_put(&hdcp_w->aconnector[conn_index]->base);
+		hdcp_w->aconnector[conn_index] = NULL;
+	}
 	process_output(hdcp_w);
 }
 

@@ -254,6 +259,10 @@ void hdcp_reset_display(struct hdcp_workqueue *hdcp_work, unsigned int link_inde
 	for (conn_index = 0; conn_index < AMDGPU_DM_MAX_DISPLAY_INDEX; conn_index++) {
 		hdcp_w->encryption_status[conn_index] =
 			MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF;
+		if (hdcp_w->aconnector[conn_index]) {
+			drm_connector_put(&hdcp_w->aconnector[conn_index]->base);
+			hdcp_w->aconnector[conn_index] = NULL;
+		}
 	}
 
 	process_output(hdcp_w);

@@ -489,6 +498,7 @@ static void update_config(void *handle, struct cp_psp_stream_config *config)
 	struct hdcp_workqueue *hdcp_work = handle;
 	struct amdgpu_dm_connector *aconnector = config->dm_stream_ctx;
 	int link_index = aconnector->dc_link->link_index;
+	unsigned int conn_index = aconnector->base.index;
 	struct mod_hdcp_display *display = &hdcp_work[link_index].display;
 	struct mod_hdcp_link *link = &hdcp_work[link_index].link;
 	struct hdcp_workqueue *hdcp_w = &hdcp_work[link_index];

@@ -545,7 +555,10 @@ static void update_config(void *handle, struct cp_psp_stream_config *config)
 	guard(mutex)(&hdcp_w->mutex);
 
 	mod_hdcp_add_display(&hdcp_w->hdcp, link, display, &hdcp_w->output);
-
+	drm_connector_get(&aconnector->base);
+	if (hdcp_w->aconnector[conn_index])
+		drm_connector_put(&hdcp_w->aconnector[conn_index]->base);
+	hdcp_w->aconnector[conn_index] = aconnector;
 	process_output(hdcp_w);
 }
 

2.39.5

Results from Linaro’s test farm.
No regressions on arm64, arm, x86_64, and i386.

## Build
* kernel: 6.12.28-rc1
* git: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git
* git commit: 483b39c5e6de6bcb0adeeab81c10cac4aa25f8ec
* git describe: v6.12.26-167-g483b39c5e6de
* test details:
https://qa-reports.linaro.org/lkft/linux-stable-rc-linux-6.12.y/build/v6.12.26-167-g483b39c5e6de

## Test Regressions (compared to v6.12.24-499-g990f4938689a)

## Metric Regressions (compared to v6.12.24-499-g990f4938689a)

## Test Fixes (compared to v6.12.24-499-g990f4938689a)

## Metric Fixes (compared to v6.12.24-499-g990f4938689a)

## Test result summary
total: 118401, pass: 96123, fail: 6121, skip: 15784, xfail: 373

## Build Summary
* arc: 5 total, 5 passed, 0 failed
* arm: 139 total, 137 passed, 2 failed
* arm64: 57 total, 56 passed, 1 failed
* i386: 18 total, 16 passed, 2 failed
* mips: 34 total, 33 passed, 1 failed
* parisc: 4 total, 4 passed, 0 failed
* powerpc: 40 total, 40 passed, 0 failed
* riscv: 25 total, 23 passed, 2 failed
* s390: 22 total, 21 passed, 1 failed
* sh: 5 total, 5 passed, 0 failed
* sparc: 4 total, 3 passed, 1 failed
* x86_64: 49 total, 42 passed, 7 failed

## Test suites summary
* boot
* commands
* kselftest-arm64
* kselftest-breakpoints
* kselftest-capabilities
* kselftest-cgroup
* kselftest-clone3
* kselftest-core
* kselftest-cpu-hotplug
* kselftest-cpufreq
* kselftest-efivarfs
* kselftest-exec
* kselftest-fpu
* kselftest-futex
* kselftest-gpio
* kselftest-intel_pstate
* kselftest-ipc
* kselftest-kcmp
* kselftest-kvm
* kselftest-livepatch
* kselftest-membarrier
* kselftest-memfd
* kselftest-mincore
* kselftest-mqueue
* kselftest-net
* kselftest-net-mptcp
* kselftest-openat2
* kselftest-ptrace
* kselftest-rseq
* kselftest-rtc
* kselftest-seccomp
* kselftest-sigaltstack
* kselftest-size
* kselftest-tc-testing
* kselftest-timers
* kselftest-tmpfs
* kselftest-tpm2
* kselftest-user_events
* kselftest-vDSO
* kselftest-x86
* kunit
* kvm-unit-tests
* lava
* libgpiod
* libhugetlbfs
* log-parser-boot
* log-parser-build-clang
* log-parser-build-gcc
* log-parser-test
* ltp-capability
* ltp-commands
* ltp-containers
* ltp-controllers
* ltp-cpuhotplug
* ltp-crypto
* ltp-cve
* ltp-dio
* ltp-fcntl-locktests
* ltp-fs
* ltp-fs_bind
* ltp-fs_perms_simple
* ltp-hugetlb
* ltp-ipc
* ltp-math
* ltp-mm
* ltp-nptl
* ltp-pty
* ltp-sched
* ltp-smoke
* ltp-syscalls
* ltp-tracing
* perf
* rcutorture
* rt-tests-cyclicdeadline
* rt-tests-pi-stress
* rt-tests-pmqtest
* rt-tests-rt-migrate-test
* rt-tests-signaltest

--
Linaro LKFT
https://lkft.linaro.org

All tests passing for Tegra ...

Test results for stable-v6.12:

    10 builds:	10 pass, 0 fail
    28 boots:	28 pass, 0 fail
    116 tests:	116 pass, 0 fail

Linux version: 6.12.28-rc1-g483b39c5e6de
Boards tested: tegra124-jetson-tk1, tegra186-p2771-0000,

                tegra186-p3509-0000+p3636-0001, tegra194-p2972-0000,
                tegra194-p3509-0000+p3668-0000, tegra20-ventana,
                tegra210-p2371-2180, tegra210-p3450-0000,
                tegra30-cardhu-a04

Jon

Boot-tested under QEMU for Rust x86_64, arm64 and riscv64; built-tested for loongarch64:

Thanks!

Cheers,
Miguel

Builds successfully. Boots and works on qemu and Dell XPS 15 9520 w/ Intel Core i7-12600H

Thanks,
Brett

Am 07.05.2025 um 20:38 schrieb Greg Kroah-Hartman:

Builds, boots and works on my 2-socket Ivy Bridge Xeon E5-2697 v2 server. No dmesg oddities or regressions found.

Beste Grüße,
Peter Schneider

Climb the mountain not to plant your flag, but to embrace the challenge,
enjoy the air and behold the view. Climb it so you can see the world,
not so the world can see you.                    -- David McCullough Jr.

OpenPGP:  0xA3828BD796CCE11A8CADE8866E3A92C92C3FF244
Download: https://www.peters-netzplatz.de/download/pschneider1968_pub.asc
https://keys.mailvelope.com/pks/lookup?op=get&[email protected]
https://keys.mailvelope.com/pks/lookup?op=get&[email protected]

Compiled and booted on my test system. No dmesg regressions.

thanks,
-- Shuah

On ARCH_BRCMSTB using 32-bit and 64-bit ARM kernels build tested on BMIPS_GENERIC:

Florian

Built and booted successfully on RISC-V RV64 (HiFive Unmatched).