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

Cross-merge networking fixes after downstream PR (net-6.16-rc6).

No conflicts.

Adjacent changes:

Documentation/devicetree/bindings/net/allwinner,sun8i-a83t-emac.yaml
  0a12c435a1 ("dt-bindings: net: sun8i-emac: Add A100 EMAC compatible")
  b3603c0466 ("dt-bindings: net: sun8i-emac: Rename A523 EMAC0 to GMAC0")

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Documentation/ABI/testing/sysfs-devices-power

@@ -56,7 +56,7 @@ Date:		January 2009
 Contact:	Rafael J. Wysocki <rjw@rjwysocki.net>
 Description:
 		The /sys/devices/.../async attribute allows the user space to
-		enable or diasble the device's suspend and resume callbacks to
+		enable or disable the device's suspend and resume callbacks to
 		be executed asynchronously (ie. in separate threads, in parallel
 		with the main suspend/resume thread) during system-wide power
 		transitions (eg. suspend to RAM, hibernation).

Documentation/ABI/testing/sysfs-devices-system-cpu

@@ -584,6 +584,7 @@ What:		/sys/devices/system/cpu/vulnerabilities
 		/sys/devices/system/cpu/vulnerabilities/spectre_v1
 		/sys/devices/system/cpu/vulnerabilities/spectre_v2
 		/sys/devices/system/cpu/vulnerabilities/srbds
+		/sys/devices/system/cpu/vulnerabilities/tsa
 		/sys/devices/system/cpu/vulnerabilities/tsx_async_abort
 Date:		January 2018
 Contact:	Linux kernel mailing list <linux-kernel@vger.kernel.org>

Documentation/ABI/testing/sysfs-driver-ufs

@@ -711,7 +711,7 @@ Description:	This file shows the thin provisioning type. This is one of
 
 		The file is read only.
 
-What:		/sys/class/scsi_device/*/device/unit_descriptor/physical_memory_resourse_count
+What:		/sys/class/scsi_device/*/device/unit_descriptor/physical_memory_resource_count
 Date:		February 2018
 Contact:	Stanislav Nijnikov <stanislav.nijnikov@wdc.com>
 Description:	This file shows the total physical memory resources. This is

Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst

@@ -157,9 +157,7 @@ This is achieved by using the otherwise unused and obsolete VERW instruction in
 combination with a microcode update. The microcode clears the affected CPU
 buffers when the VERW instruction is executed.
 
-Kernel reuses the MDS function to invoke the buffer clearing:
+Kernel does the buffer clearing with x86_clear_cpu_buffers().
-
-	mds_clear_cpu_buffers()
 
 On MDS affected CPUs, the kernel already invokes CPU buffer clear on
 kernel/userspace, hypervisor/guest and C-state (idle) transitions. No

Documentation/admin-guide/kernel-parameters.txt

@@ -7488,6 +7488,19 @@
 			having this key zero'ed is acceptable. E.g. in testing
 			scenarios.
 
+	tsa=		[X86] Control mitigation for Transient Scheduler
+			Attacks on AMD CPUs. Search the following in your
+			favourite search engine for more details:
+
+			"Technical guidance for mitigating transient scheduler
+			attacks".
+
+			off		- disable the mitigation
+			on		- enable the mitigation (default)
+			user		- mitigate only user/kernel transitions
+			vm		- mitigate only guest/host transitions
+
+
 	tsc=		Disable clocksource stability checks for TSC.
 			Format: <string>
 			[x86] reliable: mark tsc clocksource as reliable, this

Documentation/arch/x86/mds.rst

@@ -93,7 +93,7 @@ enters a C-state.
 
 The kernel provides a function to invoke the buffer clearing:
 
-    mds_clear_cpu_buffers()
+    x86_clear_cpu_buffers()
 
 Also macro CLEAR_CPU_BUFFERS can be used in ASM late in exit-to-user path.
 Other than CFLAGS.ZF, this macro doesn't clobber any registers.
@@ -185,9 +185,9 @@ Mitigation points
    idle clearing would be a window dressing exercise and is therefore not
    activated.
 
-   The invocation is controlled by the static key mds_idle_clear which is
+   The invocation is controlled by the static key cpu_buf_idle_clear which is
-   switched depending on the chosen mitigation mode and the SMT state of
+   switched depending on the chosen mitigation mode and the SMT state of the
-   the system.
+   system.
 
    The buffer clear is only invoked before entering the C-State to prevent
    that stale data from the idling CPU from spilling to the Hyper-Thread

Documentation/devicetree/bindings/i2c/realtek,rtl9301-i2c.yaml

@@ -26,7 +26,8 @@ properties:
       - const: realtek,rtl9301-i2c
 
   reg:
-    description: Register offset and size this I2C controller.
+    items:
+      - description: Register offset and size this I2C controller.
 
   "#address-cells":
     const: 1

Documentation/devicetree/bindings/input/elan,ekth6915.yaml

@@ -4,14 +4,14 @@
 $id: http://devicetree.org/schemas/input/elan,ekth6915.yaml#
 $schema: http://devicetree.org/meta-schemas/core.yaml#
 
-title: Elan eKTH6915 touchscreen controller
+title: Elan I2C-HID touchscreen controllers
 
 maintainers:
   - Douglas Anderson <dianders@chromium.org>
 
 description:
-  Supports the Elan eKTH6915 touchscreen controller.
+  Supports the Elan eKTH6915 and other I2C-HID touchscreen controllers.
-  This touchscreen controller uses the i2c-hid protocol with a reset GPIO.
+  These touchscreen controller use the i2c-hid protocol with a reset GPIO.
 
 allOf:
   - $ref: /schemas/input/touchscreen/touchscreen.yaml#
@@ -23,12 +23,14 @@ properties:
           - enum:
               - elan,ekth5015m
           - const: elan,ekth6915
+      - items:
+          - const: elan,ekth8d18
+          - const: elan,ekth6a12nay
       - enum:
           - elan,ekth6915
           - elan,ekth6a12nay
 
-  reg:
+  reg: true
-    const: 0x10
 
   interrupts:
     maxItems: 1

Documentation/devicetree/bindings/net/allwinner,sun8i-a83t-emac.yaml

@@ -24,7 +24,7 @@ properties:
               - allwinner,sun50i-a100-emac
               - allwinner,sun50i-h6-emac
               - allwinner,sun50i-h616-emac0
-              - allwinner,sun55i-a523-emac0
+              - allwinner,sun55i-a523-gmac0
           - const: allwinner,sun50i-a64-emac
 
   reg:

Documentation/virt/kvm/api.rst

@@ -7196,6 +7196,10 @@ The valid value for 'flags' is:
 					u64 leaf;
 					u64 r11, r12, r13, r14;
 				} get_tdvmcall_info;
+				struct {
+					u64 ret;
+					u64 vector;
+				} setup_event_notify;
 			};
 		} tdx;
 
@@ -7210,21 +7214,24 @@ number from register R11.  The remaining field of the union provide the
 inputs and outputs of the TDVMCALL.  Currently the following values of
 ``nr`` are defined:
 
-* ``TDVMCALL_GET_QUOTE``: the guest has requested to generate a TD-Quote
+ * ``TDVMCALL_GET_QUOTE``: the guest has requested to generate a TD-Quote
-signed by a service hosting TD-Quoting Enclave operating on the host.
+   signed by a service hosting TD-Quoting Enclave operating on the host.
-Parameters and return value are in the ``get_quote`` field of the union.
+   Parameters and return value are in the ``get_quote`` field of the union.
-The ``gpa`` field and ``size`` specify the guest physical address
+   The ``gpa`` field and ``size`` specify the guest physical address
-(without the shared bit set) and the size of a shared-memory buffer, in
+   (without the shared bit set) and the size of a shared-memory buffer, in
-which the TDX guest passes a TD Report.  The ``ret`` field represents
+   which the TDX guest passes a TD Report.  The ``ret`` field represents
-the return value of the GetQuote request.  When the request has been
+   the return value of the GetQuote request.  When the request has been
-queued successfully, the TDX guest can poll the status field in the
+   queued successfully, the TDX guest can poll the status field in the
-shared-memory area to check whether the Quote generation is completed or
+   shared-memory area to check whether the Quote generation is completed or
-not. When completed, the generated Quote is returned via the same buffer.
+   not. When completed, the generated Quote is returned via the same buffer.
 
-* ``TDVMCALL_GET_TD_VM_CALL_INFO``: the guest has requested the support
+ * ``TDVMCALL_GET_TD_VM_CALL_INFO``: the guest has requested the support
-status of TDVMCALLs.  The output values for the given leaf should be
+   status of TDVMCALLs.  The output values for the given leaf should be
-placed in fields from ``r11`` to ``r14`` of the ``get_tdvmcall_info``
+   placed in fields from ``r11`` to ``r14`` of the ``get_tdvmcall_info``
-field of the union.
+   field of the union.
+
+* ``TDVMCALL_SETUP_EVENT_NOTIFY_INTERRUPT``: the guest has requested to
+set up a notification interrupt for vector ``vector``.
 
 KVM may add support for more values in the future that may cause a userspace
 exit, even without calls to ``KVM_ENABLE_CAP`` or similar.  In this case,

Documentation/virt/kvm/x86/intel-tdx.rst

@@ -79,7 +79,20 @@ to be configured to the TDX guest.
   struct kvm_tdx_capabilities {
         __u64 supported_attrs;
         __u64 supported_xfam;
-        __u64 reserved[254];
+
+        /* TDG.VP.VMCALL hypercalls executed in kernel and forwarded to
+         * userspace, respectively
+         */
+        __u64 kernel_tdvmcallinfo_1_r11;
+        __u64 user_tdvmcallinfo_1_r11;
+
+        /* TDG.VP.VMCALL instruction executions subfunctions executed in kernel
+         * and forwarded to userspace, respectively
+         */
+        __u64 kernel_tdvmcallinfo_1_r12;
+        __u64 user_tdvmcallinfo_1_r12;
+
+        __u64 reserved[250];
 
         /* Configurable CPUID bits for userspace */
         struct kvm_cpuid2 cpuid;

Documentation/wmi/acpi-interface.rst

@@ -36,7 +36,7 @@ Offset  Size (in bytes) Content
 
 The WMI object flags control whether the method or notification ID is used:
 
-- 0x1: Data block usage is expensive and must be explicitly enabled/disabled.
+- 0x1: Data block is expensive to collect.
 - 0x2: Data block contains WMI methods.
 - 0x4: Data block contains ASCIZ string.
 - 0x8: Data block describes a WMI event, use notification ID instead
@@ -83,14 +83,18 @@ event as hexadecimal value. Their first parameter is an integer with a value
 of 0 if the WMI event should be disabled, other values will enable
 the WMI event.
 
+Those ACPI methods are always called even for WMI events not registered as
+being expensive to collect to match the behavior of the Windows driver.
+
 WCxx ACPI methods
 -----------------
-Similar to the ``WExx`` ACPI methods, except that it controls data collection
+Similar to the ``WExx`` ACPI methods, except that instead of WMI events it controls
-instead of events and thus the last two characters of the ACPI method name are
+data collection of data blocks registered as being expensive to collect. Thus the
-the method ID of the data block to enable/disable.
+last two characters of the ACPI method name are the method ID of the data block
+to enable/disable.
 
 Those ACPI methods are also called before setting data blocks to match the
-behaviour of the Windows driver.
+behavior of the Windows driver.
 
 _WED ACPI method
 ----------------

MAINTAINERS

@@ -4186,6 +4186,7 @@ F:	include/linux/cpumask_types.h
 F:	include/linux/find.h
 F:	include/linux/nodemask.h
 F:	include/linux/nodemask_types.h
+F:	include/uapi/linux/bits.h
 F:	include/vdso/bits.h
 F:	lib/bitmap-str.c
 F:	lib/bitmap.c
@@ -4198,6 +4199,7 @@ F:	tools/include/linux/bitfield.h
 F:	tools/include/linux/bitmap.h
 F:	tools/include/linux/bits.h
 F:	tools/include/linux/find.h
+F:	tools/include/uapi/linux/bits.h
 F:	tools/include/vdso/bits.h
 F:	tools/lib/bitmap.c
 F:	tools/lib/find_bit.c
@@ -16843,8 +16845,8 @@ F:	include/dt-bindings/clock/mobileye,eyeq5-clk.h
 MODULE SUPPORT
 M:	Luis Chamberlain <mcgrof@kernel.org>
 M:	Petr Pavlu <petr.pavlu@suse.com>
+M:	Daniel Gomez <da.gomez@kernel.org>
 R:	Sami Tolvanen <samitolvanen@google.com>
-R:	Daniel Gomez <da.gomez@samsung.com>
 L:	linux-modules@vger.kernel.org
 L:	linux-kernel@vger.kernel.org
 S:	Maintained
@@ -17243,10 +17245,10 @@ F:	drivers/rtc/rtc-ntxec.c
 F:	include/linux/mfd/ntxec.h
 
 NETRONOME ETHERNET DRIVERS
-M:	Louis Peens <louis.peens@corigine.com>
 R:	Jakub Kicinski <kuba@kernel.org>
+R:	Simon Horman <horms@kernel.org>
 L:	oss-drivers@corigine.com
-S:	Maintained
+S:	Odd Fixes
 F:	drivers/net/ethernet/netronome/
 
 NETWORK BLOCK DEVICE (NBD)
@@ -19622,8 +19624,7 @@ T:	git git://git.kernel.org/pub/scm/linux/kernel/git/pinctrl/intel.git
 F:	drivers/pinctrl/intel/
 
 PIN CONTROLLER - KEEMBAY
-M:	Lakshmi Sowjanya D <lakshmi.sowjanya.d@intel.com>
+S:	Orphan
-S:	Supported
 F:	drivers/pinctrl/pinctrl-keembay*
 
 PIN CONTROLLER - MEDIATEK
@@ -20176,21 +20177,15 @@ S:	Supported
 F:	Documentation/devicetree/bindings/soc/qcom/qcom,apr*
 F:	Documentation/devicetree/bindings/sound/qcom,*
 F:	drivers/soc/qcom/apr.c
-F:	include/dt-bindings/sound/qcom,wcd9335.h
+F:	drivers/soundwire/qcom.c
-F:	include/dt-bindings/sound/qcom,wcd934x.h
+F:	include/dt-bindings/sound/qcom,wcd93*
-F:	sound/soc/codecs/lpass-rx-macro.*
+F:	sound/soc/codecs/lpass-*.*
-F:	sound/soc/codecs/lpass-tx-macro.*
-F:	sound/soc/codecs/lpass-va-macro.c
-F:	sound/soc/codecs/lpass-wsa-macro.*
 F:	sound/soc/codecs/msm8916-wcd-analog.c
 F:	sound/soc/codecs/msm8916-wcd-digital.c
 F:	sound/soc/codecs/wcd-clsh-v2.*
 F:	sound/soc/codecs/wcd-mbhc-v2.*
-F:	sound/soc/codecs/wcd9335.*
+F:	sound/soc/codecs/wcd93*.*
-F:	sound/soc/codecs/wcd934x.c
+F:	sound/soc/codecs/wsa88*.*
-F:	sound/soc/codecs/wsa881x.c
-F:	sound/soc/codecs/wsa883x.c
-F:	sound/soc/codecs/wsa884x.c
 F:	sound/soc/qcom/
 
 QCOM EMBEDDED USB DEBUGGER (EUD)

Makefile

@@ -2,7 +2,7 @@
 VERSION = 6
 PATCHLEVEL = 16
 SUBLEVEL = 0
-EXTRAVERSION = -rc4
+EXTRAVERSION = -rc5
 NAME = Baby Opossum Posse
 
 # *DOCUMENTATION*

arch/arm64/Kconfig

@@ -256,6 +256,7 @@ config ARM64
 	select HOTPLUG_SMT if HOTPLUG_CPU
 	select IRQ_DOMAIN
 	select IRQ_FORCED_THREADING
+	select JUMP_LABEL
 	select KASAN_VMALLOC if KASAN
 	select LOCK_MM_AND_FIND_VMA
 	select MODULES_USE_ELF_RELA

arch/arm64/boot/dts/apple/spi1-nvram.dtsi

@@ -20,8 +20,6 @@
 		compatible = "jedec,spi-nor";
 		reg = <0x0>;
 		spi-max-frequency = <25000000>;
-		#address-cells = <1>;
-		#size-cells = <1>;
 
 		partitions {
 			compatible = "fixed-partitions";

arch/arm64/boot/dts/apple/t8103-j293.dts

@@ -100,6 +100,8 @@
 
 &displaydfr_mipi {
 	status = "okay";
+	#address-cells = <1>;
+	#size-cells = <0>;
 
 	dfr_panel: panel@0 {
 		compatible = "apple,j293-summit", "apple,summit";

arch/arm64/boot/dts/apple/t8103-jxxx.dtsi

@@ -71,7 +71,7 @@
  */
 &port00 {
 	bus-range = <1 1>;
-	wifi0: network@0,0 {
+	wifi0: wifi@0,0 {
 		compatible = "pci14e4,4425";
 		reg = <0x10000 0x0 0x0 0x0 0x0>;
 		/* To be filled by the loader */

arch/arm64/boot/dts/apple/t8103.dtsi

@@ -405,8 +405,6 @@
 			compatible = "apple,t8103-display-pipe-mipi", "apple,h7-display-pipe-mipi";
 			reg = <0x2 0x28600000 0x0 0x100000>;
 			power-domains = <&ps_mipi_dsi>;
-			#address-cells = <1>;
-			#size-cells = <0>;
 			status = "disabled";
 
 			ports {

arch/arm64/boot/dts/apple/t8112-j493.dts

@@ -63,6 +63,8 @@
 
 &displaydfr_mipi {
 	status = "okay";
+	#address-cells = <1>;
+	#size-cells = <0>;
 
 	dfr_panel: panel@0 {
 		compatible = "apple,j493-summit", "apple,summit";

arch/arm64/boot/dts/apple/t8112.dtsi

@@ -420,8 +420,6 @@
 			compatible = "apple,t8112-display-pipe-mipi", "apple,h7-display-pipe-mipi";
 			reg = <0x2 0x28600000 0x0 0x100000>;
 			power-domains = <&ps_mipi_dsi>;
-			#address-cells = <1>;
-			#size-cells = <0>;
 			status = "disabled";
 
 			ports {

arch/arm64/configs/defconfig

@@ -1573,6 +1573,7 @@ CONFIG_RESET_QCOM_AOSS=y
 CONFIG_RESET_QCOM_PDC=m
 CONFIG_RESET_RZG2L_USBPHY_CTRL=y
 CONFIG_RESET_TI_SCI=y
+CONFIG_PHY_SNPS_EUSB2=m
 CONFIG_PHY_XGENE=y
 CONFIG_PHY_CAN_TRANSCEIVER=m
 CONFIG_PHY_NXP_PTN3222=m
@@ -1597,7 +1598,6 @@ CONFIG_PHY_QCOM_EDP=m
 CONFIG_PHY_QCOM_PCIE2=m
 CONFIG_PHY_QCOM_QMP=m
 CONFIG_PHY_QCOM_QUSB2=m
-CONFIG_PHY_QCOM_SNPS_EUSB2=m
 CONFIG_PHY_QCOM_EUSB2_REPEATER=m
 CONFIG_PHY_QCOM_M31_USB=m
 CONFIG_PHY_QCOM_USB_HS=m

arch/arm64/include/asm/el2_setup.h

@@ -287,17 +287,6 @@
 .Lskip_fgt2_\@:
 .endm
 
-.macro __init_el2_gcs
-	mrs_s	x1, SYS_ID_AA64PFR1_EL1
-	ubfx	x1, x1, #ID_AA64PFR1_EL1_GCS_SHIFT, #4
-	cbz	x1, .Lskip_gcs_\@
-
-	/* Ensure GCS is not enabled when we start trying to do BLs */
-	msr_s	SYS_GCSCR_EL1, xzr
-	msr_s	SYS_GCSCRE0_EL1, xzr
-.Lskip_gcs_\@:
-.endm
-
 /**
  * Initialize EL2 registers to sane values. This should be called early on all
  * cores that were booted in EL2. Note that everything gets initialised as
@@ -319,7 +308,6 @@
 	__init_el2_cptr
 	__init_el2_fgt
 	__init_el2_fgt2
-        __init_el2_gcs
 .endm
 
 #ifndef __KVM_NVHE_HYPERVISOR__
@@ -371,6 +359,13 @@
 	msr_s   SYS_MPAMHCR_EL2, xzr		// clear TRAP_MPAMIDR_EL1 -> EL2
 
 .Lskip_mpam_\@:
+	check_override id_aa64pfr1, ID_AA64PFR1_EL1_GCS_SHIFT, .Linit_gcs_\@, .Lskip_gcs_\@, x1, x2
+
+.Linit_gcs_\@:
+	msr_s	SYS_GCSCR_EL1, xzr
+	msr_s	SYS_GCSCRE0_EL1, xzr
+
+.Lskip_gcs_\@:
 	check_override id_aa64pfr0, ID_AA64PFR0_EL1_SVE_SHIFT, .Linit_sve_\@, .Lskip_sve_\@, x1, x2
 
 .Linit_sve_\@:	/* SVE register access */

arch/arm64/include/asm/kvm_host.h

@@ -1480,7 +1480,6 @@ int kvm_vm_ioctl_get_reg_writable_masks(struct kvm *kvm,
 					struct reg_mask_range *range);
 
 /* Guest/host FPSIMD coordination helpers */
-int kvm_arch_vcpu_run_map_fp(struct kvm_vcpu *vcpu);
 void kvm_arch_vcpu_load_fp(struct kvm_vcpu *vcpu);
 void kvm_arch_vcpu_ctxflush_fp(struct kvm_vcpu *vcpu);
 void kvm_arch_vcpu_ctxsync_fp(struct kvm_vcpu *vcpu);

arch/arm64/kernel/Makefile

@@ -34,7 +34,7 @@ obj-y			:= debug-monitors.o entry.o irq.o fpsimd.o		\
 			   cpufeature.o alternative.o cacheinfo.o		\
 			   smp.o smp_spin_table.o topology.o smccc-call.o	\
 			   syscall.o proton-pack.o idle.o patching.o pi/	\
-			   rsi.o
+			   rsi.o jump_label.o
 
 obj-$(CONFIG_COMPAT)			+= sys32.o signal32.o			\
 					   sys_compat.o
@@ -47,7 +47,6 @@ obj-$(CONFIG_PERF_EVENTS)		+= perf_regs.o perf_callchain.o
 obj-$(CONFIG_HARDLOCKUP_DETECTOR_PERF)	+= watchdog_hld.o
 obj-$(CONFIG_HAVE_HW_BREAKPOINT)	+= hw_breakpoint.o
 obj-$(CONFIG_CPU_PM)			+= sleep.o suspend.o
-obj-$(CONFIG_JUMP_LABEL)		+= jump_label.o
 obj-$(CONFIG_KGDB)			+= kgdb.o
 obj-$(CONFIG_EFI)			+= efi.o efi-rt-wrapper.o
 obj-$(CONFIG_PCI)			+= pci.o

arch/arm64/kernel/cpufeature.c

@@ -3135,6 +3135,13 @@ static bool has_sve_feature(const struct arm64_cpu_capabilities *cap, int scope)
 }
 #endif
 
+#ifdef CONFIG_ARM64_SME
+static bool has_sme_feature(const struct arm64_cpu_capabilities *cap, int scope)
+{
+	return system_supports_sme() && has_user_cpuid_feature(cap, scope);
+}
+#endif
+
 static const struct arm64_cpu_capabilities arm64_elf_hwcaps[] = {
 	HWCAP_CAP(ID_AA64ISAR0_EL1, AES, PMULL, CAP_HWCAP, KERNEL_HWCAP_PMULL),
 	HWCAP_CAP(ID_AA64ISAR0_EL1, AES, AES, CAP_HWCAP, KERNEL_HWCAP_AES),
@@ -3223,31 +3230,31 @@ static const struct arm64_cpu_capabilities arm64_elf_hwcaps[] = {
 	HWCAP_CAP(ID_AA64ISAR2_EL1, BC, IMP, CAP_HWCAP, KERNEL_HWCAP_HBC),
 #ifdef CONFIG_ARM64_SME
 	HWCAP_CAP(ID_AA64PFR1_EL1, SME, IMP, CAP_HWCAP, KERNEL_HWCAP_SME),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, FA64, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_FA64),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, FA64, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_FA64),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, LUTv2, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_LUTV2),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, LUTv2, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_LUTV2),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, SMEver, SME2p2, CAP_HWCAP, KERNEL_HWCAP_SME2P2),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, SMEver, SME2p2, CAP_HWCAP, KERNEL_HWCAP_SME2P2),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, SMEver, SME2p1, CAP_HWCAP, KERNEL_HWCAP_SME2P1),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, SMEver, SME2p1, CAP_HWCAP, KERNEL_HWCAP_SME2P1),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, SMEver, SME2, CAP_HWCAP, KERNEL_HWCAP_SME2),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, SMEver, SME2, CAP_HWCAP, KERNEL_HWCAP_SME2),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, I16I64, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_I16I64),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, I16I64, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_I16I64),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, F64F64, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_F64F64),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, F64F64, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_F64F64),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, I16I32, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_I16I32),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, I16I32, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_I16I32),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, B16B16, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_B16B16),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, B16B16, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_B16B16),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, F16F16, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_F16F16),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, F16F16, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_F16F16),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, F8F16, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_F8F16),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, F8F16, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_F8F16),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, F8F32, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_F8F32),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, F8F32, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_F8F32),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, I8I32, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_I8I32),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, I8I32, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_I8I32),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, F16F32, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_F16F32),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, F16F32, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_F16F32),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, B16F32, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_B16F32),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, B16F32, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_B16F32),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, BI32I32, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_BI32I32),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, BI32I32, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_BI32I32),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, F32F32, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_F32F32),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, F32F32, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_F32F32),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, SF8FMA, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SF8FMA),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, SF8FMA, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SF8FMA),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, SF8DP4, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SF8DP4),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, SF8DP4, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SF8DP4),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, SF8DP2, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SF8DP2),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, SF8DP2, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SF8DP2),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, SBitPerm, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SBITPERM),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, SBitPerm, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SBITPERM),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, AES, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_AES),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, AES, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_AES),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, SFEXPA, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SFEXPA),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, SFEXPA, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SFEXPA),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, STMOP, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_STMOP),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, STMOP, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_STMOP),
-	HWCAP_CAP(ID_AA64SMFR0_EL1, SMOP4, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SMOP4),
+	HWCAP_CAP_MATCH_ID(has_sme_feature, ID_AA64SMFR0_EL1, SMOP4, IMP, CAP_HWCAP, KERNEL_HWCAP_SME_SMOP4),
 #endif /* CONFIG_ARM64_SME */
 	HWCAP_CAP(ID_AA64FPFR0_EL1, F8CVT, IMP, CAP_HWCAP, KERNEL_HWCAP_F8CVT),
 	HWCAP_CAP(ID_AA64FPFR0_EL1, F8FMA, IMP, CAP_HWCAP, KERNEL_HWCAP_F8FMA),

arch/arm64/kernel/efi.c

@@ -15,6 +15,7 @@
 
 #include <asm/efi.h>
 #include <asm/stacktrace.h>
+#include <asm/vmap_stack.h>
 
 static bool region_is_misaligned(const efi_memory_desc_t *md)
 {
@@ -214,9 +215,13 @@ static int __init arm64_efi_rt_init(void)
 	if (!efi_enabled(EFI_RUNTIME_SERVICES))
 		return 0;
 
-	p = __vmalloc_node(THREAD_SIZE, THREAD_ALIGN, GFP_KERNEL,
+	if (!IS_ENABLED(CONFIG_VMAP_STACK)) {
-			   NUMA_NO_NODE, &&l);
+		clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
-l:	if (!p) {
+		return -ENOMEM;
+	}
+
+	p = arch_alloc_vmap_stack(THREAD_SIZE, NUMA_NO_NODE);
+	if (!p) {
 		pr_warn("Failed to allocate EFI runtime stack\n");
 		clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
 		return -ENOMEM;

arch/arm64/kernel/process.c

@@ -673,6 +673,11 @@ static void permission_overlay_switch(struct task_struct *next)
 	current->thread.por_el0 = read_sysreg_s(SYS_POR_EL0);
 	if (current->thread.por_el0 != next->thread.por_el0) {
 		write_sysreg_s(next->thread.por_el0, SYS_POR_EL0);
+		/*
+		 * No ISB required as we can tolerate spurious Overlay faults -
+		 * the fault handler will check again based on the new value
+		 * of POR_EL0.
+		 */
 	}
 }
 

arch/arm64/kernel/smp.c

@@ -1143,7 +1143,7 @@ static inline unsigned int num_other_online_cpus(void)
 void smp_send_stop(void)
 {
 	static unsigned long stop_in_progress;
-	cpumask_t mask;
+	static cpumask_t mask;
 	unsigned long timeout;
 
 	/*

arch/arm64/kvm/arm.c

@@ -825,10 +825,6 @@ int kvm_arch_vcpu_run_pid_change(struct kvm_vcpu *vcpu)
 	if (!kvm_arm_vcpu_is_finalized(vcpu))
 		return -EPERM;
 
-	ret = kvm_arch_vcpu_run_map_fp(vcpu);
-	if (ret)
-		return ret;
-
 	if (likely(vcpu_has_run_once(vcpu)))
 		return 0;
 
@@ -2129,7 +2125,7 @@ static void cpu_hyp_init(void *discard)
 
 static void cpu_hyp_uninit(void *discard)
 {
-	if (__this_cpu_read(kvm_hyp_initialized)) {
+	if (!is_protected_kvm_enabled() && __this_cpu_read(kvm_hyp_initialized)) {
 		cpu_hyp_reset();
 		__this_cpu_write(kvm_hyp_initialized, 0);
 	}
@@ -2345,8 +2341,13 @@ static void __init teardown_hyp_mode(void)
 
 	free_hyp_pgds();
 	for_each_possible_cpu(cpu) {
+		if (per_cpu(kvm_hyp_initialized, cpu))
+			continue;
+
 		free_pages(per_cpu(kvm_arm_hyp_stack_base, cpu), NVHE_STACK_SHIFT - PAGE_SHIFT);
-		free_pages(kvm_nvhe_sym(kvm_arm_hyp_percpu_base)[cpu], nvhe_percpu_order());
+
+		if (!kvm_nvhe_sym(kvm_arm_hyp_percpu_base)[cpu])
+			continue;
 
 		if (free_sve) {
 			struct cpu_sve_state *sve_state;
@@ -2354,6 +2355,9 @@ static void __init teardown_hyp_mode(void)
 			sve_state = per_cpu_ptr_nvhe_sym(kvm_host_data, cpu)->sve_state;
 			free_pages((unsigned long) sve_state, pkvm_host_sve_state_order());
 		}
+
+		free_pages(kvm_nvhe_sym(kvm_arm_hyp_percpu_base)[cpu], nvhe_percpu_order());
+
 	}
 }
 

arch/arm64/kvm/fpsimd.c

@@ -14,32 +14,6 @@
 #include <asm/kvm_mmu.h>
 #include <asm/sysreg.h>
 
-/*
- * Called on entry to KVM_RUN unless this vcpu previously ran at least
- * once and the most recent prior KVM_RUN for this vcpu was called from
- * the same task as current (highly likely).
- *
- * This is guaranteed to execute before kvm_arch_vcpu_load_fp(vcpu),
- * such that on entering hyp the relevant parts of current are already
- * mapped.
- */
-int kvm_arch_vcpu_run_map_fp(struct kvm_vcpu *vcpu)
-{
-	struct user_fpsimd_state *fpsimd = &current->thread.uw.fpsimd_state;
-	int ret;
-
-	/* pKVM has its own tracking of the host fpsimd state. */
-	if (is_protected_kvm_enabled())
-		return 0;
-
-	/* Make sure the host task fpsimd state is visible to hyp: */
-	ret = kvm_share_hyp(fpsimd, fpsimd + 1);
-	if (ret)
-		return ret;
-
-	return 0;
-}
-
 /*
  * Prepare vcpu for saving the host's FPSIMD state and loading the guest's.
  * The actual loading is done by the FPSIMD access trap taken to hyp.

arch/arm64/kvm/hyp/nvhe/mem_protect.c

@@ -479,6 +479,7 @@ static int host_stage2_adjust_range(u64 addr, struct kvm_mem_range *range)
 {
 	struct kvm_mem_range cur;
 	kvm_pte_t pte;
+	u64 granule;
 	s8 level;
 	int ret;
 
@@ -496,18 +497,21 @@ static int host_stage2_adjust_range(u64 addr, struct kvm_mem_range *range)
 		return -EPERM;
 	}
 
-	do {
+	for (; level <= KVM_PGTABLE_LAST_LEVEL; level++) {
-		u64 granule = kvm_granule_size(level);
+		if (!kvm_level_supports_block_mapping(level))
+			continue;
+		granule = kvm_granule_size(level);
 		cur.start = ALIGN_DOWN(addr, granule);
 		cur.end = cur.start + granule;
-		level++;
+		if (!range_included(&cur, range))
-	} while ((level <= KVM_PGTABLE_LAST_LEVEL) &&
+			continue;
-			!(kvm_level_supports_block_mapping(level) &&
+		*range = cur;
-			  range_included(&cur, range)));
+		return 0;
+	}
 
-	*range = cur;
+	WARN_ON(1);
 
-	return 0;
+	return -EINVAL;
 }
 
 int host_stage2_idmap_locked(phys_addr_t addr, u64 size,

arch/arm64/kvm/nested.c

@@ -1402,6 +1402,21 @@ static void kvm_map_l1_vncr(struct kvm_vcpu *vcpu)
 	}
 }
 
+#define has_tgran_2(__r, __sz)						\
+	({								\
+		u64 _s1, _s2, _mmfr0 = __r;				\
+									\
+		_s2 = SYS_FIELD_GET(ID_AA64MMFR0_EL1,			\
+				    TGRAN##__sz##_2, _mmfr0);		\
+									\
+		_s1 = SYS_FIELD_GET(ID_AA64MMFR0_EL1,			\
+				    TGRAN##__sz, _mmfr0);		\
+									\
+		((_s2 != ID_AA64MMFR0_EL1_TGRAN##__sz##_2_NI &&		\
+		  _s2 != ID_AA64MMFR0_EL1_TGRAN##__sz##_2_TGRAN##__sz) || \
+		 (_s2 == ID_AA64MMFR0_EL1_TGRAN##__sz##_2_TGRAN##__sz && \
+		  _s1 != ID_AA64MMFR0_EL1_TGRAN##__sz##_NI));		\
+	})
 /*
  * Our emulated CPU doesn't support all the possible features. For the
  * sake of simplicity (and probably mental sanity), wipe out a number
@@ -1411,6 +1426,8 @@ static void kvm_map_l1_vncr(struct kvm_vcpu *vcpu)
  */
 u64 limit_nv_id_reg(struct kvm *kvm, u32 reg, u64 val)
 {
+	u64 orig_val = val;
+
 	switch (reg) {
 	case SYS_ID_AA64ISAR0_EL1:
 		/* Support everything but TME */
@@ -1480,13 +1497,16 @@ u64 limit_nv_id_reg(struct kvm *kvm, u32 reg, u64 val)
 		 */
 		switch (PAGE_SIZE) {
 		case SZ_4K:
-			val |= SYS_FIELD_PREP_ENUM(ID_AA64MMFR0_EL1, TGRAN4_2, IMP);
+			if (has_tgran_2(orig_val, 4))
+				val |= SYS_FIELD_PREP_ENUM(ID_AA64MMFR0_EL1, TGRAN4_2, IMP);
 			fallthrough;
 		case SZ_16K:
-			val |= SYS_FIELD_PREP_ENUM(ID_AA64MMFR0_EL1, TGRAN16_2, IMP);
+			if (has_tgran_2(orig_val, 16))
+				val |= SYS_FIELD_PREP_ENUM(ID_AA64MMFR0_EL1, TGRAN16_2, IMP);
 			fallthrough;
 		case SZ_64K:
-			val |= SYS_FIELD_PREP_ENUM(ID_AA64MMFR0_EL1, TGRAN64_2, IMP);
+			if (has_tgran_2(orig_val, 64))
+				val |= SYS_FIELD_PREP_ENUM(ID_AA64MMFR0_EL1, TGRAN64_2, IMP);
 			break;
 		}
 

arch/arm64/kvm/vgic/vgic-v3-nested.c

@@ -401,9 +401,7 @@ void vgic_v3_nested_update_mi(struct kvm_vcpu *vcpu)
 {
 	bool level;
 
-	level  = __vcpu_sys_reg(vcpu, ICH_HCR_EL2) & ICH_HCR_EL2_En;
+	level = (__vcpu_sys_reg(vcpu, ICH_HCR_EL2) & ICH_HCR_EL2_En) && vgic_v3_get_misr(vcpu);
-	if (level)
-		level &= vgic_v3_get_misr(vcpu);
 	kvm_vgic_inject_irq(vcpu->kvm, vcpu,
 			    vcpu->kvm->arch.vgic.mi_intid, level, vcpu);
 }

arch/arm64/mm/fault.c

@@ -487,17 +487,29 @@ static void do_bad_area(unsigned long far, unsigned long esr,
 	}
 }
 
-static bool fault_from_pkey(unsigned long esr, struct vm_area_struct *vma,
+static bool fault_from_pkey(struct vm_area_struct *vma, unsigned int mm_flags)
-			unsigned int mm_flags)
 {
-	unsigned long iss2 = ESR_ELx_ISS2(esr);
-
 	if (!system_supports_poe())
 		return false;
 
-	if (esr_fsc_is_permission_fault(esr) && (iss2 & ESR_ELx_Overlay))
+	/*
-		return true;
+	 * We do not check whether an Overlay fault has occurred because we
-
+	 * cannot make a decision based solely on its value:
+	 *
+	 * - If Overlay is set, a fault did occur due to POE, but it may be
+	 *   spurious in those cases where we update POR_EL0 without ISB (e.g.
+	 *   on context-switch). We would then need to manually check POR_EL0
+	 *   against vma_pkey(vma), which is exactly what
+	 *   arch_vma_access_permitted() does.
+	 *
+	 * - If Overlay is not set, we may still need to report a pkey fault.
+	 *   This is the case if an access was made within a mapping but with no
+	 *   page mapped, and POR_EL0 forbids the access (according to
+	 *   vma_pkey()). Such access will result in a SIGSEGV regardless
+	 *   because core code checks arch_vma_access_permitted(), but in order
+	 *   to report the correct error code - SEGV_PKUERR - we must handle
+	 *   that case here.
+	 */
 	return !arch_vma_access_permitted(vma,
 			mm_flags & FAULT_FLAG_WRITE,
 			mm_flags & FAULT_FLAG_INSTRUCTION,
@@ -635,7 +647,7 @@ static int __kprobes do_page_fault(unsigned long far, unsigned long esr,
 		goto bad_area;
 	}
 
-	if (fault_from_pkey(esr, vma, mm_flags)) {
+	if (fault_from_pkey(vma, mm_flags)) {
 		pkey = vma_pkey(vma);
 		vma_end_read(vma);
 		fault = 0;
@@ -679,7 +691,7 @@ retry:
 		goto bad_area;
 	}
 
-	if (fault_from_pkey(esr, vma, mm_flags)) {
+	if (fault_from_pkey(vma, mm_flags)) {
 		pkey = vma_pkey(vma);
 		mmap_read_unlock(mm);
 		fault = 0;

arch/arm64/mm/proc.S

@@ -518,7 +518,6 @@ alternative_else_nop_endif
 	msr	REG_PIR_EL1, x0
 
 	orr	tcr2, tcr2, TCR2_EL1_PIE
-	msr	REG_TCR2_EL1, x0
 
 .Lskip_indirection:
 

arch/riscv/Kconfig

@@ -63,7 +63,8 @@ config RISCV
 	select ARCH_OPTIONAL_KERNEL_RWX_DEFAULT
 	select ARCH_STACKWALK
 	select ARCH_SUPPORTS_ATOMIC_RMW
-	select ARCH_SUPPORTS_CFI_CLANG
+	# clang >= 17: https://github.com/llvm/llvm-project/commit/62fa708ceb027713b386c7e0efda994f8bdc27e2
+	select ARCH_SUPPORTS_CFI_CLANG if CLANG_VERSION >= 170000
 	select ARCH_SUPPORTS_DEBUG_PAGEALLOC if MMU
 	select ARCH_SUPPORTS_HUGE_PFNMAP if TRANSPARENT_HUGEPAGE
 	select ARCH_SUPPORTS_HUGETLBFS if MMU

arch/riscv/kernel/cpu_ops_sbi.c

@@ -18,10 +18,10 @@ const struct cpu_operations cpu_ops_sbi;
 
 /*
  * Ordered booting via HSM brings one cpu at a time. However, cpu hotplug can
- * be invoked from multiple threads in parallel. Define a per cpu data
+ * be invoked from multiple threads in parallel. Define an array of boot data
  * to handle that.
  */
-static DEFINE_PER_CPU(struct sbi_hart_boot_data, boot_data);
+static struct sbi_hart_boot_data boot_data[NR_CPUS];
 
 static int sbi_hsm_hart_start(unsigned long hartid, unsigned long saddr,
 			      unsigned long priv)
@@ -67,7 +67,7 @@ static int sbi_cpu_start(unsigned int cpuid, struct task_struct *tidle)
 	unsigned long boot_addr = __pa_symbol(secondary_start_sbi);
 	unsigned long hartid = cpuid_to_hartid_map(cpuid);
 	unsigned long hsm_data;
-	struct sbi_hart_boot_data *bdata = &per_cpu(boot_data, cpuid);
+	struct sbi_hart_boot_data *bdata = &boot_data[cpuid];
 
 	/* Make sure tidle is updated */
 	smp_mb();

arch/s390/crypto/sha1_s390.c

@@ -38,6 +38,7 @@ static int s390_sha1_init(struct shash_desc *desc)
 	sctx->state[4] = SHA1_H4;
 	sctx->count = 0;
 	sctx->func = CPACF_KIMD_SHA_1;
+	sctx->first_message_part = 0;
 
 	return 0;
 }
@@ -60,6 +61,7 @@ static int s390_sha1_import(struct shash_desc *desc, const void *in)
 	sctx->count = ictx->count;
 	memcpy(sctx->state, ictx->state, sizeof(ictx->state));
 	sctx->func = CPACF_KIMD_SHA_1;
+	sctx->first_message_part = 0;
 	return 0;
 }
 

arch/s390/crypto/sha512_s390.c

@@ -32,6 +32,7 @@ static int sha512_init(struct shash_desc *desc)
 	ctx->count = 0;
 	ctx->sha512.count_hi = 0;
 	ctx->func = CPACF_KIMD_SHA_512;
+	ctx->first_message_part = 0;
 
 	return 0;
 }
@@ -57,6 +58,7 @@ static int sha512_import(struct shash_desc *desc, const void *in)
 
 	memcpy(sctx->state, ictx->state, sizeof(ictx->state));
 	sctx->func = CPACF_KIMD_SHA_512;
+	sctx->first_message_part = 0;
 	return 0;
 }
 
@@ -97,6 +99,7 @@ static int sha384_init(struct shash_desc *desc)
 	ctx->count = 0;
 	ctx->sha512.count_hi = 0;
 	ctx->func = CPACF_KIMD_SHA_512;
+	ctx->first_message_part = 0;
 
 	return 0;
 }

arch/x86/Kconfig

@@ -2695,6 +2695,15 @@ config MITIGATION_ITS
 	  disabled, mitigation cannot be enabled via cmdline.
 	  See <file:Documentation/admin-guide/hw-vuln/indirect-target-selection.rst>
 
+config MITIGATION_TSA
+	bool "Mitigate Transient Scheduler Attacks"
+	depends on CPU_SUP_AMD
+	default y
+	help
+	  Enable mitigation for Transient Scheduler Attacks. TSA is a hardware
+	  security vulnerability on AMD CPUs which can lead to forwarding of
+	  invalid info to subsequent instructions and thus can affect their
+	  timing and thereby cause a leakage.
 endif
 
 config ARCH_HAS_ADD_PAGES

arch/x86/coco/sev/core.c

@@ -88,7 +88,7 @@ static const char * const sev_status_feat_names[] = {
  */
 static u64 snp_tsc_scale __ro_after_init;
 static u64 snp_tsc_offset __ro_after_init;
-static u64 snp_tsc_freq_khz __ro_after_init;
+static unsigned long snp_tsc_freq_khz __ro_after_init;
 
 DEFINE_PER_CPU(struct sev_es_runtime_data*, runtime_data);
 DEFINE_PER_CPU(struct sev_es_save_area *, sev_vmsa);
@@ -2167,15 +2167,31 @@ static unsigned long securetsc_get_tsc_khz(void)
 
 void __init snp_secure_tsc_init(void)
 {
-	unsigned long long tsc_freq_mhz;
+	struct snp_secrets_page *secrets;
+	unsigned long tsc_freq_mhz;
+	void *mem;
 
 	if (!cc_platform_has(CC_ATTR_GUEST_SNP_SECURE_TSC))
 		return;
 
+	mem = early_memremap_encrypted(sev_secrets_pa, PAGE_SIZE);
+	if (!mem) {
+		pr_err("Unable to get TSC_FACTOR: failed to map the SNP secrets page.\n");
+		sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_SECURE_TSC);
+	}
+
+	secrets = (__force struct snp_secrets_page *)mem;
+
 	setup_force_cpu_cap(X86_FEATURE_TSC_KNOWN_FREQ);
 	rdmsrq(MSR_AMD64_GUEST_TSC_FREQ, tsc_freq_mhz);
-	snp_tsc_freq_khz = (unsigned long)(tsc_freq_mhz * 1000);
+
+	/* Extract the GUEST TSC MHZ from BIT[17:0], rest is reserved space */
+	tsc_freq_mhz &= GENMASK_ULL(17, 0);
+
+	snp_tsc_freq_khz = SNP_SCALE_TSC_FREQ(tsc_freq_mhz * 1000, secrets->tsc_factor);
 
 	x86_platform.calibrate_cpu = securetsc_get_tsc_khz;
 	x86_platform.calibrate_tsc = securetsc_get_tsc_khz;
+
+	early_memunmap(mem, PAGE_SIZE);
 }

arch/x86/entry/entry.S

@@ -36,20 +36,20 @@ EXPORT_SYMBOL_GPL(write_ibpb);
 
 /*
  * Define the VERW operand that is disguised as entry code so that
- * it can be referenced with KPTI enabled. This ensure VERW can be
+ * it can be referenced with KPTI enabled. This ensures VERW can be
  * used late in exit-to-user path after page tables are switched.
  */
 .pushsection .entry.text, "ax"
 
 .align L1_CACHE_BYTES, 0xcc
-SYM_CODE_START_NOALIGN(mds_verw_sel)
+SYM_CODE_START_NOALIGN(x86_verw_sel)
 	UNWIND_HINT_UNDEFINED
 	ANNOTATE_NOENDBR
 	.word __KERNEL_DS
 .align L1_CACHE_BYTES, 0xcc
-SYM_CODE_END(mds_verw_sel);
+SYM_CODE_END(x86_verw_sel);
 /* For KVM */
-EXPORT_SYMBOL_GPL(mds_verw_sel);
+EXPORT_SYMBOL_GPL(x86_verw_sel);
 
 .popsection
 

arch/x86/include/asm/cpufeatures.h

@@ -456,6 +456,7 @@
 #define X86_FEATURE_NO_NESTED_DATA_BP	(20*32+ 0) /* No Nested Data Breakpoints */
 #define X86_FEATURE_WRMSR_XX_BASE_NS	(20*32+ 1) /* WRMSR to {FS,GS,KERNEL_GS}_BASE is non-serializing */
 #define X86_FEATURE_LFENCE_RDTSC	(20*32+ 2) /* LFENCE always serializing / synchronizes RDTSC */
+#define X86_FEATURE_VERW_CLEAR		(20*32+ 5) /* The memory form of VERW mitigates TSA */
 #define X86_FEATURE_NULL_SEL_CLR_BASE	(20*32+ 6) /* Null Selector Clears Base */
 #define X86_FEATURE_AUTOIBRS		(20*32+ 8) /* Automatic IBRS */
 #define X86_FEATURE_NO_SMM_CTL_MSR	(20*32+ 9) /* SMM_CTL MSR is not present */
@@ -487,6 +488,9 @@
 #define X86_FEATURE_PREFER_YMM		(21*32+ 8) /* Avoid ZMM registers due to downclocking */
 #define X86_FEATURE_APX			(21*32+ 9) /* Advanced Performance Extensions */
 #define X86_FEATURE_INDIRECT_THUNK_ITS	(21*32+10) /* Use thunk for indirect branches in lower half of cacheline */
+#define X86_FEATURE_TSA_SQ_NO		(21*32+11) /* AMD CPU not vulnerable to TSA-SQ */
+#define X86_FEATURE_TSA_L1_NO		(21*32+12) /* AMD CPU not vulnerable to TSA-L1 */
+#define X86_FEATURE_CLEAR_CPU_BUF_VM	(21*32+13) /* Clear CPU buffers using VERW before VMRUN */
 
 /*
  * BUG word(s)
@@ -542,5 +546,5 @@
 #define X86_BUG_OLD_MICROCODE		X86_BUG( 1*32+ 6) /* "old_microcode" CPU has old microcode, it is surely vulnerable to something */
 #define X86_BUG_ITS			X86_BUG( 1*32+ 7) /* "its" CPU is affected by Indirect Target Selection */
 #define X86_BUG_ITS_NATIVE_ONLY		X86_BUG( 1*32+ 8) /* "its_native_only" CPU is affected by ITS, VMX is not affected */
-
+#define X86_BUG_TSA			X86_BUG( 1*32+ 9) /* "tsa" CPU is affected by Transient Scheduler Attacks */
 #endif /* _ASM_X86_CPUFEATURES_H */

arch/x86/include/asm/irqflags.h

@@ -44,13 +44,13 @@ static __always_inline void native_irq_enable(void)
 
 static __always_inline void native_safe_halt(void)
 {
-	mds_idle_clear_cpu_buffers();
+	x86_idle_clear_cpu_buffers();
 	asm volatile("sti; hlt": : :"memory");
 }
 
 static __always_inline void native_halt(void)
 {
-	mds_idle_clear_cpu_buffers();
+	x86_idle_clear_cpu_buffers();
 	asm volatile("hlt": : :"memory");
 }
 

arch/x86/include/asm/kvm_host.h

@@ -700,8 +700,13 @@ struct kvm_vcpu_hv {
 
 	struct kvm_vcpu_hv_tlb_flush_fifo tlb_flush_fifo[HV_NR_TLB_FLUSH_FIFOS];
 
-	/* Preallocated buffer for handling hypercalls passing sparse vCPU set */
+	/*
+	 * Preallocated buffers for handling hypercalls that pass sparse vCPU
+	 * sets (for high vCPU counts, they're too large to comfortably fit on
+	 * the stack).
+	 */
 	u64 sparse_banks[HV_MAX_SPARSE_VCPU_BANKS];
+	DECLARE_BITMAP(vcpu_mask, KVM_MAX_VCPUS);
 
 	struct hv_vp_assist_page vp_assist_page;
 
@@ -764,6 +769,7 @@ enum kvm_only_cpuid_leafs {
 	CPUID_8000_0022_EAX,
 	CPUID_7_2_EDX,
 	CPUID_24_0_EBX,
+	CPUID_8000_0021_ECX,
 	NR_KVM_CPU_CAPS,
 
 	NKVMCAPINTS = NR_KVM_CPU_CAPS - NCAPINTS,

arch/x86/include/asm/mwait.h

@@ -43,8 +43,6 @@ static __always_inline void __monitorx(const void *eax, u32 ecx, u32 edx)
 
 static __always_inline void __mwait(u32 eax, u32 ecx)
 {
-	mds_idle_clear_cpu_buffers();
-
 	/*
 	 * Use the instruction mnemonic with implicit operands, as the LLVM
 	 * assembler fails to assemble the mnemonic with explicit operands:
@@ -80,7 +78,7 @@ static __always_inline void __mwait(u32 eax, u32 ecx)
  */
 static __always_inline void __mwaitx(u32 eax, u32 ebx, u32 ecx)
 {
-	/* No MDS buffer clear as this is AMD/HYGON only */
+	/* No need for TSA buffer clearing on AMD */
 
 	/* "mwaitx %eax, %ebx, %ecx" */
 	asm volatile(".byte 0x0f, 0x01, 0xfb"
@@ -98,7 +96,6 @@ static __always_inline void __mwaitx(u32 eax, u32 ebx, u32 ecx)
  */
 static __always_inline void __sti_mwait(u32 eax, u32 ecx)
 {
-	mds_idle_clear_cpu_buffers();
 
 	asm volatile("sti; mwait" :: "a" (eax), "c" (ecx));
 }
@@ -115,21 +112,29 @@ static __always_inline void __sti_mwait(u32 eax, u32 ecx)
  */
 static __always_inline void mwait_idle_with_hints(u32 eax, u32 ecx)
 {
+	if (need_resched())
+		return;
+
+	x86_idle_clear_cpu_buffers();
+
 	if (static_cpu_has_bug(X86_BUG_MONITOR) || !current_set_polling_and_test()) {
 		const void *addr = &current_thread_info()->flags;
 
 		alternative_input("", "clflush (%[addr])", X86_BUG_CLFLUSH_MONITOR, [addr] "a" (addr));
 		__monitor(addr, 0, 0);
 
-		if (!need_resched()) {
+		if (need_resched())
-			if (ecx & 1) {
+			goto out;
-				__mwait(eax, ecx);
+
-			} else {
+		if (ecx & 1) {
-				__sti_mwait(eax, ecx);
+			__mwait(eax, ecx);
-				raw_local_irq_disable();
+		} else {
-			}
+			__sti_mwait(eax, ecx);
+			raw_local_irq_disable();
 		}
 	}
+
+out:
 	current_clr_polling();
 }
 

arch/x86/include/asm/nospec-branch.h

@@ -302,25 +302,31 @@
 .endm
 
 /*
- * Macro to execute VERW instruction that mitigate transient data sampling
+ * Macro to execute VERW insns that mitigate transient data sampling
- * attacks such as MDS. On affected systems a microcode update overloaded VERW
+ * attacks such as MDS or TSA. On affected systems a microcode update
- * instruction to also clear the CPU buffers. VERW clobbers CFLAGS.ZF.
+ * overloaded VERW insns to also clear the CPU buffers. VERW clobbers
- *
+ * CFLAGS.ZF.
  * Note: Only the memory operand variant of VERW clears the CPU buffers.
  */
-.macro CLEAR_CPU_BUFFERS
+.macro __CLEAR_CPU_BUFFERS feature
 #ifdef CONFIG_X86_64
-	ALTERNATIVE "", "verw mds_verw_sel(%rip)", X86_FEATURE_CLEAR_CPU_BUF
+	ALTERNATIVE "", "verw x86_verw_sel(%rip)", \feature
 #else
 	/*
 	 * In 32bit mode, the memory operand must be a %cs reference. The data
 	 * segments may not be usable (vm86 mode), and the stack segment may not
 	 * be flat (ESPFIX32).
 	 */
-	ALTERNATIVE "", "verw %cs:mds_verw_sel", X86_FEATURE_CLEAR_CPU_BUF
+	ALTERNATIVE "", "verw %cs:x86_verw_sel", \feature
 #endif
 .endm
 
+#define CLEAR_CPU_BUFFERS \
+	__CLEAR_CPU_BUFFERS X86_FEATURE_CLEAR_CPU_BUF
+
+#define VM_CLEAR_CPU_BUFFERS \
+	__CLEAR_CPU_BUFFERS X86_FEATURE_CLEAR_CPU_BUF_VM
+
 #ifdef CONFIG_X86_64
 .macro CLEAR_BRANCH_HISTORY
 	ALTERNATIVE "", "call clear_bhb_loop", X86_FEATURE_CLEAR_BHB_LOOP
@@ -567,24 +573,24 @@ DECLARE_STATIC_KEY_FALSE(switch_mm_always_ibpb);
 
 DECLARE_STATIC_KEY_FALSE(switch_vcpu_ibpb);
 
-DECLARE_STATIC_KEY_FALSE(mds_idle_clear);
+DECLARE_STATIC_KEY_FALSE(cpu_buf_idle_clear);
 
 DECLARE_STATIC_KEY_FALSE(switch_mm_cond_l1d_flush);
 
 DECLARE_STATIC_KEY_FALSE(cpu_buf_vm_clear);
 
-extern u16 mds_verw_sel;
+extern u16 x86_verw_sel;
 
 #include <asm/segment.h>
 
 /**
- * mds_clear_cpu_buffers - Mitigation for MDS and TAA vulnerability
+ * x86_clear_cpu_buffers - Buffer clearing support for different x86 CPU vulns
  *
  * This uses the otherwise unused and obsolete VERW instruction in
  * combination with microcode which triggers a CPU buffer flush when the
  * instruction is executed.
  */
-static __always_inline void mds_clear_cpu_buffers(void)
+static __always_inline void x86_clear_cpu_buffers(void)
 {
 	static const u16 ds = __KERNEL_DS;
 
@@ -601,14 +607,15 @@ static __always_inline void mds_clear_cpu_buffers(void)
 }
 
 /**
- * mds_idle_clear_cpu_buffers - Mitigation for MDS vulnerability
+ * x86_idle_clear_cpu_buffers - Buffer clearing support in idle for the MDS
+ * and TSA vulnerabilities.
  *
  * Clear CPU buffers if the corresponding static key is enabled
  */
-static __always_inline void mds_idle_clear_cpu_buffers(void)
+static __always_inline void x86_idle_clear_cpu_buffers(void)
 {
-	if (static_branch_likely(&mds_idle_clear))
+	if (static_branch_likely(&cpu_buf_idle_clear))
-		mds_clear_cpu_buffers();
+		x86_clear_cpu_buffers();
 }
 
 #endif /* __ASSEMBLER__ */

arch/x86/include/asm/sev.h

@@ -223,6 +223,18 @@ struct snp_tsc_info_resp {
 	u8 rsvd2[100];
 } __packed;
 
+/*
+ * Obtain the mean TSC frequency by decreasing the nominal TSC frequency with
+ * TSC_FACTOR as documented in the SNP Firmware ABI specification:
+ *
+ * GUEST_TSC_FREQ * (1 - (TSC_FACTOR * 0.00001))
+ *
+ * which is equivalent to:
+ *
+ * GUEST_TSC_FREQ -= (GUEST_TSC_FREQ * TSC_FACTOR) / 100000;
+ */
+#define SNP_SCALE_TSC_FREQ(freq, factor) ((freq) - (freq) * (factor) / 100000)
+
 struct snp_guest_req {
 	void *req_buf;
 	size_t req_sz;
@@ -282,8 +294,11 @@ struct snp_secrets_page {
 	u8 svsm_guest_vmpl;
 	u8 rsvd3[3];
 
+	/* The percentage decrease from nominal to mean TSC frequency. */
+	u32 tsc_factor;
+
 	/* Remainder of page */
-	u8 rsvd4[3744];
+	u8 rsvd4[3740];
 } __packed;
 
 struct snp_msg_desc {

arch/x86/include/asm/shared/tdx.h

@@ -72,6 +72,7 @@
 #define TDVMCALL_MAP_GPA		0x10001
 #define TDVMCALL_GET_QUOTE		0x10002
 #define TDVMCALL_REPORT_FATAL_ERROR	0x10003
+#define TDVMCALL_SETUP_EVENT_NOTIFY_INTERRUPT 0x10004ULL
 
 /*
  * TDG.VP.VMCALL Status Codes (returned in R10)

arch/x86/include/uapi/asm/kvm.h

@@ -965,7 +965,13 @@ struct kvm_tdx_cmd {
 struct kvm_tdx_capabilities {
 	__u64 supported_attrs;
 	__u64 supported_xfam;
-	__u64 reserved[254];
+
+	__u64 kernel_tdvmcallinfo_1_r11;
+	__u64 user_tdvmcallinfo_1_r11;
+	__u64 kernel_tdvmcallinfo_1_r12;
+	__u64 user_tdvmcallinfo_1_r12;
+
+	__u64 reserved[250];
 
 	/* Configurable CPUID bits for userspace */
 	struct kvm_cpuid2 cpuid;

arch/x86/kernel/cpu/amd.c

@@ -9,7 +9,7 @@
 #include <linux/sched/clock.h>
 #include <linux/random.h>
 #include <linux/topology.h>
-#include <asm/amd/fch.h>
+#include <linux/platform_data/x86/amd-fch.h>
 #include <asm/processor.h>
 #include <asm/apic.h>
 #include <asm/cacheinfo.h>
@@ -377,6 +377,47 @@ static void bsp_determine_snp(struct cpuinfo_x86 *c)
 #endif
 }
 
+#define ZEN_MODEL_STEP_UCODE(fam, model, step, ucode) \
+	X86_MATCH_VFM_STEPS(VFM_MAKE(X86_VENDOR_AMD, fam, model), \
+			    step, step, ucode)
+
+static const struct x86_cpu_id amd_tsa_microcode[] = {
+	ZEN_MODEL_STEP_UCODE(0x19, 0x01, 0x1, 0x0a0011d7),
+	ZEN_MODEL_STEP_UCODE(0x19, 0x01, 0x2, 0x0a00123b),
+	ZEN_MODEL_STEP_UCODE(0x19, 0x08, 0x2, 0x0a00820d),
+	ZEN_MODEL_STEP_UCODE(0x19, 0x11, 0x1, 0x0a10114c),
+	ZEN_MODEL_STEP_UCODE(0x19, 0x11, 0x2, 0x0a10124c),
+	ZEN_MODEL_STEP_UCODE(0x19, 0x18, 0x1, 0x0a108109),
+	ZEN_MODEL_STEP_UCODE(0x19, 0x21, 0x0, 0x0a20102e),
+	ZEN_MODEL_STEP_UCODE(0x19, 0x21, 0x2, 0x0a201211),
+	ZEN_MODEL_STEP_UCODE(0x19, 0x44, 0x1, 0x0a404108),
+	ZEN_MODEL_STEP_UCODE(0x19, 0x50, 0x0, 0x0a500012),
+	ZEN_MODEL_STEP_UCODE(0x19, 0x61, 0x2, 0x0a60120a),
+	ZEN_MODEL_STEP_UCODE(0x19, 0x74, 0x1, 0x0a704108),
+	ZEN_MODEL_STEP_UCODE(0x19, 0x75, 0x2, 0x0a705208),
+	ZEN_MODEL_STEP_UCODE(0x19, 0x78, 0x0, 0x0a708008),
+	ZEN_MODEL_STEP_UCODE(0x19, 0x7c, 0x0, 0x0a70c008),
+	ZEN_MODEL_STEP_UCODE(0x19, 0xa0, 0x2, 0x0aa00216),
+	{},
+};
+
+static void tsa_init(struct cpuinfo_x86 *c)
+{
+	if (cpu_has(c, X86_FEATURE_HYPERVISOR))
+		return;
+
+	if (cpu_has(c, X86_FEATURE_ZEN3) ||
+	    cpu_has(c, X86_FEATURE_ZEN4)) {
+		if (x86_match_min_microcode_rev(amd_tsa_microcode))
+			setup_force_cpu_cap(X86_FEATURE_VERW_CLEAR);
+		else
+			pr_debug("%s: current revision: 0x%x\n", __func__, c->microcode);
+	} else {
+		setup_force_cpu_cap(X86_FEATURE_TSA_SQ_NO);
+		setup_force_cpu_cap(X86_FEATURE_TSA_L1_NO);
+	}
+}
+
 static void bsp_init_amd(struct cpuinfo_x86 *c)
 {
 	if (cpu_has(c, X86_FEATURE_CONSTANT_TSC)) {
@@ -489,6 +530,9 @@ static void bsp_init_amd(struct cpuinfo_x86 *c)
 	}
 
 	bsp_determine_snp(c);
+
+	tsa_init(c);
+
 	return;
 
 warn:

arch/x86/kernel/cpu/bugs.c

@@ -94,6 +94,8 @@ static void __init bhi_apply_mitigation(void);
 static void __init its_select_mitigation(void);
 static void __init its_update_mitigation(void);
 static void __init its_apply_mitigation(void);
+static void __init tsa_select_mitigation(void);
+static void __init tsa_apply_mitigation(void);
 
 /* The base value of the SPEC_CTRL MSR without task-specific bits set */
 u64 x86_spec_ctrl_base;
@@ -169,9 +171,9 @@ DEFINE_STATIC_KEY_FALSE(switch_mm_always_ibpb);
 DEFINE_STATIC_KEY_FALSE(switch_vcpu_ibpb);
 EXPORT_SYMBOL_GPL(switch_vcpu_ibpb);
 
-/* Control MDS CPU buffer clear before idling (halt, mwait) */
+/* Control CPU buffer clear before idling (halt, mwait) */
-DEFINE_STATIC_KEY_FALSE(mds_idle_clear);
+DEFINE_STATIC_KEY_FALSE(cpu_buf_idle_clear);
-EXPORT_SYMBOL_GPL(mds_idle_clear);
+EXPORT_SYMBOL_GPL(cpu_buf_idle_clear);
 
 /*
  * Controls whether l1d flush based mitigations are enabled,
@@ -225,6 +227,7 @@ void __init cpu_select_mitigations(void)
 	gds_select_mitigation();
 	its_select_mitigation();
 	bhi_select_mitigation();
+	tsa_select_mitigation();
 
 	/*
 	 * After mitigations are selected, some may need to update their
@@ -272,6 +275,7 @@ void __init cpu_select_mitigations(void)
 	gds_apply_mitigation();
 	its_apply_mitigation();
 	bhi_apply_mitigation();
+	tsa_apply_mitigation();
 }
 
 /*
@@ -637,7 +641,7 @@ static void __init mmio_apply_mitigation(void)
 	 * is required irrespective of SMT state.
 	 */
 	if (!(x86_arch_cap_msr & ARCH_CAP_FBSDP_NO))
-		static_branch_enable(&mds_idle_clear);
+		static_branch_enable(&cpu_buf_idle_clear);
 
 	if (mmio_nosmt || cpu_mitigations_auto_nosmt())
 		cpu_smt_disable(false);
@@ -1487,6 +1491,94 @@ static void __init its_apply_mitigation(void)
 	set_return_thunk(its_return_thunk);
 }
 
+#undef pr_fmt
+#define pr_fmt(fmt)	"Transient Scheduler Attacks: " fmt
+
+enum tsa_mitigations {
+	TSA_MITIGATION_NONE,
+	TSA_MITIGATION_AUTO,
+	TSA_MITIGATION_UCODE_NEEDED,
+	TSA_MITIGATION_USER_KERNEL,
+	TSA_MITIGATION_VM,
+	TSA_MITIGATION_FULL,
+};
+
+static const char * const tsa_strings[] = {
+	[TSA_MITIGATION_NONE]		= "Vulnerable",
+	[TSA_MITIGATION_UCODE_NEEDED]	= "Vulnerable: No microcode",
+	[TSA_MITIGATION_USER_KERNEL]	= "Mitigation: Clear CPU buffers: user/kernel boundary",
+	[TSA_MITIGATION_VM]		= "Mitigation: Clear CPU buffers: VM",
+	[TSA_MITIGATION_FULL]		= "Mitigation: Clear CPU buffers",
+};
+
+static enum tsa_mitigations tsa_mitigation __ro_after_init =
+	IS_ENABLED(CONFIG_MITIGATION_TSA) ? TSA_MITIGATION_AUTO : TSA_MITIGATION_NONE;
+
+static int __init tsa_parse_cmdline(char *str)
+{
+	if (!str)
+		return -EINVAL;
+
+	if (!strcmp(str, "off"))
+		tsa_mitigation = TSA_MITIGATION_NONE;
+	else if (!strcmp(str, "on"))
+		tsa_mitigation = TSA_MITIGATION_FULL;
+	else if (!strcmp(str, "user"))
+		tsa_mitigation = TSA_MITIGATION_USER_KERNEL;
+	else if (!strcmp(str, "vm"))
+		tsa_mitigation = TSA_MITIGATION_VM;
+	else
+		pr_err("Ignoring unknown tsa=%s option.\n", str);
+
+	return 0;
+}
+early_param("tsa", tsa_parse_cmdline);
+
+static void __init tsa_select_mitigation(void)
+{
+	if (cpu_mitigations_off() || !boot_cpu_has_bug(X86_BUG_TSA)) {
+		tsa_mitigation = TSA_MITIGATION_NONE;
+		return;
+	}
+
+	if (tsa_mitigation == TSA_MITIGATION_NONE)
+		return;
+
+	if (!boot_cpu_has(X86_FEATURE_VERW_CLEAR)) {
+		tsa_mitigation = TSA_MITIGATION_UCODE_NEEDED;
+		goto out;
+	}
+
+	if (tsa_mitigation == TSA_MITIGATION_AUTO)
+		tsa_mitigation = TSA_MITIGATION_FULL;
+
+	/*
+	 * No need to set verw_clear_cpu_buf_mitigation_selected - it
+	 * doesn't fit all cases here and it is not needed because this
+	 * is the only VERW-based mitigation on AMD.
+	 */
+out:
+	pr_info("%s\n", tsa_strings[tsa_mitigation]);
+}
+
+static void __init tsa_apply_mitigation(void)
+{
+	switch (tsa_mitigation) {
+	case TSA_MITIGATION_USER_KERNEL:
+		setup_force_cpu_cap(X86_FEATURE_CLEAR_CPU_BUF);
+		break;
+	case TSA_MITIGATION_VM:
+		setup_force_cpu_cap(X86_FEATURE_CLEAR_CPU_BUF_VM);
+		break;
+	case TSA_MITIGATION_FULL:
+		setup_force_cpu_cap(X86_FEATURE_CLEAR_CPU_BUF);
+		setup_force_cpu_cap(X86_FEATURE_CLEAR_CPU_BUF_VM);
+		break;
+	default:
+		break;
+	}
+}
+
 #undef pr_fmt
 #define pr_fmt(fmt)     "Spectre V2 : " fmt
 
@@ -2249,10 +2341,10 @@ static void update_mds_branch_idle(void)
 		return;
 
 	if (sched_smt_active()) {
-		static_branch_enable(&mds_idle_clear);
+		static_branch_enable(&cpu_buf_idle_clear);
 	} else if (mmio_mitigation == MMIO_MITIGATION_OFF ||
 		   (x86_arch_cap_msr & ARCH_CAP_FBSDP_NO)) {
-		static_branch_disable(&mds_idle_clear);
+		static_branch_disable(&cpu_buf_idle_clear);
 	}
 }
 
@@ -2316,6 +2408,25 @@ void cpu_bugs_smt_update(void)
 		break;
 	}
 
+	switch (tsa_mitigation) {
+	case TSA_MITIGATION_USER_KERNEL:
+	case TSA_MITIGATION_VM:
+	case TSA_MITIGATION_AUTO:
+	case TSA_MITIGATION_FULL:
+		/*
+		 * TSA-SQ can potentially lead to info leakage between
+		 * SMT threads.
+		 */
+		if (sched_smt_active())
+			static_branch_enable(&cpu_buf_idle_clear);
+		else
+			static_branch_disable(&cpu_buf_idle_clear);
+		break;
+	case TSA_MITIGATION_NONE:
+	case TSA_MITIGATION_UCODE_NEEDED:
+		break;
+	}
+
 	mutex_unlock(&spec_ctrl_mutex);
 }
 
@@ -3265,6 +3376,11 @@ static ssize_t gds_show_state(char *buf)
 	return sysfs_emit(buf, "%s\n", gds_strings[gds_mitigation]);
 }
 
+static ssize_t tsa_show_state(char *buf)
+{
+	return sysfs_emit(buf, "%s\n", tsa_strings[tsa_mitigation]);
+}
+
 static ssize_t cpu_show_common(struct device *dev, struct device_attribute *attr,
 			       char *buf, unsigned int bug)
 {
@@ -3328,6 +3444,9 @@ static ssize_t cpu_show_common(struct device *dev, struct device_attribute *attr
 	case X86_BUG_ITS:
 		return its_show_state(buf);
 
+	case X86_BUG_TSA:
+		return tsa_show_state(buf);
+
 	default:
 		break;
 	}
@@ -3414,6 +3533,11 @@ ssize_t cpu_show_indirect_target_selection(struct device *dev, struct device_att
 {
 	return cpu_show_common(dev, attr, buf, X86_BUG_ITS);
 }
+
+ssize_t cpu_show_tsa(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	return cpu_show_common(dev, attr, buf, X86_BUG_TSA);
+}
 #endif
 
 void __warn_thunk(void)

arch/x86/kernel/cpu/common.c

@@ -1233,6 +1233,8 @@ static const __initconst struct x86_cpu_id cpu_vuln_whitelist[] = {
 #define ITS		BIT(8)
 /* CPU is affected by Indirect Target Selection, but guest-host isolation is not affected */
 #define ITS_NATIVE_ONLY	BIT(9)
+/* CPU is affected by Transient Scheduler Attacks */
+#define TSA		BIT(10)
 
 static const struct x86_cpu_id cpu_vuln_blacklist[] __initconst = {
 	VULNBL_INTEL_STEPS(INTEL_IVYBRIDGE,	     X86_STEP_MAX,	SRBDS),
@@ -1280,7 +1282,7 @@ static const struct x86_cpu_id cpu_vuln_blacklist[] __initconst = {
 	VULNBL_AMD(0x16, RETBLEED),
 	VULNBL_AMD(0x17, RETBLEED | SMT_RSB | SRSO),
 	VULNBL_HYGON(0x18, RETBLEED | SMT_RSB | SRSO),
-	VULNBL_AMD(0x19, SRSO),
+	VULNBL_AMD(0x19, SRSO | TSA),
 	VULNBL_AMD(0x1a, SRSO),
 	{}
 };
@@ -1530,6 +1532,16 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c)
 			setup_force_cpu_bug(X86_BUG_ITS_NATIVE_ONLY);
 	}
 
+	if (c->x86_vendor == X86_VENDOR_AMD) {
+		if (!cpu_has(c, X86_FEATURE_TSA_SQ_NO) ||
+		    !cpu_has(c, X86_FEATURE_TSA_L1_NO)) {
+			if (cpu_matches(cpu_vuln_blacklist, TSA) ||
+			    /* Enable bug on Zen guests to allow for live migration. */
+			    (cpu_has(c, X86_FEATURE_HYPERVISOR) && cpu_has(c, X86_FEATURE_ZEN)))
+				setup_force_cpu_bug(X86_BUG_TSA);
+		}
+	}
+
 	if (cpu_matches(cpu_vuln_whitelist, NO_MELTDOWN))
 		return;
 

arch/x86/kernel/cpu/mce/amd.c

@@ -350,7 +350,6 @@ static void smca_configure(unsigned int bank, unsigned int cpu)
 
 struct thresh_restart {
 	struct threshold_block	*b;
-	int			reset;
 	int			set_lvt_off;
 	int			lvt_off;
 	u16			old_limit;
@@ -432,13 +431,13 @@ static void threshold_restart_bank(void *_tr)
 
 	rdmsr(tr->b->address, lo, hi);
 
-	if (tr->b->threshold_limit < (hi & THRESHOLD_MAX))
+	/*
-		tr->reset = 1;	/* limit cannot be lower than err count */
+	 * Reset error count and overflow bit.
-
+	 * This is done during init or after handling an interrupt.
-	if (tr->reset) {		/* reset err count and overflow bit */
+	 */
-		hi =
+	if (hi & MASK_OVERFLOW_HI || tr->set_lvt_off) {
-		    (hi & ~(MASK_ERR_COUNT_HI | MASK_OVERFLOW_HI)) |
+		hi &= ~(MASK_ERR_COUNT_HI | MASK_OVERFLOW_HI);
-		    (THRESHOLD_MAX - tr->b->threshold_limit);
+		hi |= THRESHOLD_MAX - tr->b->threshold_limit;
 	} else if (tr->old_limit) {	/* change limit w/o reset */
 		int new_count = (hi & THRESHOLD_MAX) +
 		    (tr->old_limit - tr->b->threshold_limit);
@@ -1113,13 +1112,20 @@ static const char *get_name(unsigned int cpu, unsigned int bank, struct threshol
 	}
 
 	bank_type = smca_get_bank_type(cpu, bank);
-	if (bank_type >= N_SMCA_BANK_TYPES)
-		return NULL;
 
 	if (b && (bank_type == SMCA_UMC || bank_type == SMCA_UMC_V2)) {
 		if (b->block < ARRAY_SIZE(smca_umc_block_names))
 			return smca_umc_block_names[b->block];
-		return NULL;
+	}
+
+	if (b && b->block) {
+		snprintf(buf_mcatype, MAX_MCATYPE_NAME_LEN, "th_block_%u", b->block);
+		return buf_mcatype;
+	}
+
+	if (bank_type >= N_SMCA_BANK_TYPES) {
+		snprintf(buf_mcatype, MAX_MCATYPE_NAME_LEN, "th_bank_%u", bank);
+		return buf_mcatype;
 	}
 
 	if (per_cpu(smca_bank_counts, cpu)[bank_type] == 1)

arch/x86/kernel/cpu/mce/core.c

@@ -1740,6 +1740,11 @@ static void mc_poll_banks_default(void)
 
 void (*mc_poll_banks)(void) = mc_poll_banks_default;
 
+static bool should_enable_timer(unsigned long iv)
+{
+	return !mca_cfg.ignore_ce && iv;
+}
+
 static void mce_timer_fn(struct timer_list *t)
 {
 	struct timer_list *cpu_t = this_cpu_ptr(&mce_timer);
@@ -1763,7 +1768,7 @@ static void mce_timer_fn(struct timer_list *t)
 
 	if (mce_get_storm_mode()) {
 		__start_timer(t, HZ);
-	} else {
+	} else if (should_enable_timer(iv)) {
 		__this_cpu_write(mce_next_interval, iv);
 		__start_timer(t, iv);
 	}
@@ -2156,11 +2161,10 @@ static void mce_start_timer(struct timer_list *t)
 {
 	unsigned long iv = check_interval * HZ;
 
-	if (mca_cfg.ignore_ce || !iv)
+	if (should_enable_timer(iv)) {
-		return;
+		this_cpu_write(mce_next_interval, iv);
-
+		__start_timer(t, iv);
-	this_cpu_write(mce_next_interval, iv);
+	}
-	__start_timer(t, iv);
 }
 
 static void __mcheck_cpu_setup_timer(void)
@@ -2801,15 +2805,9 @@ static int mce_cpu_dead(unsigned int cpu)
 static int mce_cpu_online(unsigned int cpu)
 {
 	struct timer_list *t = this_cpu_ptr(&mce_timer);
-	int ret;
 
 	mce_device_create(cpu);
-
+	mce_threshold_create_device(cpu);
-	ret = mce_threshold_create_device(cpu);
-	if (ret) {
-		mce_device_remove(cpu);
-		return ret;
-	}
 	mce_reenable_cpu();
 	mce_start_timer(t);
 	return 0;

arch/x86/kernel/cpu/mce/intel.c

@@ -478,6 +478,7 @@ void mce_intel_feature_init(struct cpuinfo_x86 *c)
 void mce_intel_feature_clear(struct cpuinfo_x86 *c)
 {
 	intel_clear_lmce();
+	cmci_clear();
 }
 
 bool intel_filter_mce(struct mce *m)

arch/x86/kernel/cpu/microcode/amd_shas.c

@@ -231,6 +231,13 @@ static const struct patch_digest phashes[] = {
 		0x0d,0x5b,0x65,0x34,0x69,0xb2,0x62,0x21,
 	}
  },
+ { 0xa0011d7, {
+                0x35,0x07,0xcd,0x40,0x94,0xbc,0x81,0x6b,
+                0xfc,0x61,0x56,0x1a,0xe2,0xdb,0x96,0x12,
+                0x1c,0x1c,0x31,0xb1,0x02,0x6f,0xe5,0xd2,
+                0xfe,0x1b,0x04,0x03,0x2c,0x8f,0x4c,0x36,
+        }
+ },
  { 0xa001223, {
 		0xfb,0x32,0x5f,0xc6,0x83,0x4f,0x8c,0xb8,
 		0xa4,0x05,0xf9,0x71,0x53,0x01,0x16,0xc4,
@@ -294,6 +301,13 @@ static const struct patch_digest phashes[] = {
 		0xc0,0xcd,0x33,0xf2,0x8d,0xf9,0xef,0x59,
 	}
  },
+ { 0xa00123b, {
+		0xef,0xa1,0x1e,0x71,0xf1,0xc3,0x2c,0xe2,
+		0xc3,0xef,0x69,0x41,0x7a,0x54,0xca,0xc3,
+		0x8f,0x62,0x84,0xee,0xc2,0x39,0xd9,0x28,
+		0x95,0xa7,0x12,0x49,0x1e,0x30,0x71,0x72,
+	}
+ },
  { 0xa00820c, {
 		0xa8,0x0c,0x81,0xc0,0xa6,0x00,0xe7,0xf3,
 		0x5f,0x65,0xd3,0xb9,0x6f,0xea,0x93,0x63,
@@ -301,6 +315,13 @@ static const struct patch_digest phashes[] = {
 		0xe1,0x3b,0x8d,0xb2,0xf8,0x22,0x03,0xe2,
 	}
  },
+ { 0xa00820d, {
+		0xf9,0x2a,0xc0,0xf4,0x9e,0xa4,0x87,0xa4,
+		0x7d,0x87,0x00,0xfd,0xab,0xda,0x19,0xca,
+		0x26,0x51,0x32,0xc1,0x57,0x91,0xdf,0xc1,
+		0x05,0xeb,0x01,0x7c,0x5a,0x95,0x21,0xb7,
+	}
+ },
  { 0xa10113e, {
 		0x05,0x3c,0x66,0xd7,0xa9,0x5a,0x33,0x10,
 		0x1b,0xf8,0x9c,0x8f,0xed,0xfc,0xa7,0xa0,
@@ -322,6 +343,13 @@ static const struct patch_digest phashes[] = {
 		0xf1,0x5e,0xb0,0xde,0xb4,0x98,0xae,0xc4,
 	}
  },
+ { 0xa10114c, {
+		0x9e,0xb6,0xa2,0xd9,0x87,0x38,0xc5,0x64,
+		0xd8,0x88,0xfa,0x78,0x98,0xf9,0x6f,0x74,
+		0x39,0x90,0x1b,0xa5,0xcf,0x5e,0xb4,0x2a,
+		0x02,0xff,0xd4,0x8c,0x71,0x8b,0xe2,0xc0,
+	}
+ },
  { 0xa10123e, {
 		0x03,0xb9,0x2c,0x76,0x48,0x93,0xc9,0x18,
 		0xfb,0x56,0xfd,0xf7,0xe2,0x1d,0xca,0x4d,
@@ -343,6 +371,13 @@ static const struct patch_digest phashes[] = {
 		0x1b,0x7d,0x64,0x9d,0x4b,0x53,0x13,0x75,
 	}
  },
+ { 0xa10124c, {
+		0x29,0xea,0xf1,0x2c,0xb2,0xe4,0xef,0x90,
+		0xa4,0xcd,0x1d,0x86,0x97,0x17,0x61,0x46,
+		0xfc,0x22,0xcb,0x57,0x75,0x19,0xc8,0xcc,
+		0x0c,0xf5,0xbc,0xac,0x81,0x9d,0x9a,0xd2,
+	}
+ },
  { 0xa108108, {
 		0xed,0xc2,0xec,0xa1,0x15,0xc6,0x65,0xe9,
 		0xd0,0xef,0x39,0xaa,0x7f,0x55,0x06,0xc6,
@@ -350,6 +385,13 @@ static const struct patch_digest phashes[] = {
 		0x28,0x1e,0x9c,0x59,0x69,0x99,0x4d,0x16,
 	}
  },
+ { 0xa108109, {
+		0x85,0xb4,0xbd,0x7c,0x49,0xa7,0xbd,0xfa,
+		0x49,0x36,0x80,0x81,0xc5,0xb7,0x39,0x1b,
+		0x9a,0xaa,0x50,0xde,0x9b,0xe9,0x32,0x35,
+		0x42,0x7e,0x51,0x4f,0x52,0x2c,0x28,0x59,
+	}
+ },
  { 0xa20102d, {
 		0xf9,0x6e,0xf2,0x32,0xd3,0x0f,0x5f,0x11,
 		0x59,0xa1,0xfe,0xcc,0xcd,0x9b,0x42,0x89,
@@ -357,6 +399,13 @@ static const struct patch_digest phashes[] = {
 		0x8c,0xe9,0x19,0x3e,0xcc,0x3f,0x7b,0xb4,
 	}
  },
+ { 0xa20102e, {
+		0xbe,0x1f,0x32,0x04,0x0d,0x3c,0x9c,0xdd,
+		0xe1,0xa4,0xbf,0x76,0x3a,0xec,0xc2,0xf6,
+		0x11,0x00,0xa7,0xaf,0x0f,0xe5,0x02,0xc5,
+		0x54,0x3a,0x1f,0x8c,0x16,0xb5,0xff,0xbe,
+	}
+ },
  { 0xa201210, {
 		0xe8,0x6d,0x51,0x6a,0x8e,0x72,0xf3,0xfe,
 		0x6e,0x16,0xbc,0x62,0x59,0x40,0x17,0xe9,
@@ -364,6 +413,13 @@ static const struct patch_digest phashes[] = {
 		0xf7,0x55,0xf0,0x13,0xbb,0x22,0xf6,0x41,
 	}
  },
+ { 0xa201211, {
+		0x69,0xa1,0x17,0xec,0xd0,0xf6,0x6c,0x95,
+		0xe2,0x1e,0xc5,0x59,0x1a,0x52,0x0a,0x27,
+		0xc4,0xed,0xd5,0x59,0x1f,0xbf,0x00,0xff,
+		0x08,0x88,0xb5,0xe1,0x12,0xb6,0xcc,0x27,
+	}
+ },
  { 0xa404107, {
 		0xbb,0x04,0x4e,0x47,0xdd,0x5e,0x26,0x45,
 		0x1a,0xc9,0x56,0x24,0xa4,0x4c,0x82,0xb0,
@@ -371,6 +427,13 @@ static const struct patch_digest phashes[] = {
 		0x13,0xbc,0xc5,0x25,0xe4,0xc5,0xc3,0x99,
 	}
  },
+ { 0xa404108, {
+		0x69,0x67,0x43,0x06,0xf8,0x0c,0x62,0xdc,
+		0xa4,0x21,0x30,0x4f,0x0f,0x21,0x2c,0xcb,
+		0xcc,0x37,0xf1,0x1c,0xc3,0xf8,0x2f,0x19,
+		0xdf,0x53,0x53,0x46,0xb1,0x15,0xea,0x00,
+	}
+ },
  { 0xa500011, {
 		0x23,0x3d,0x70,0x7d,0x03,0xc3,0xc4,0xf4,
 		0x2b,0x82,0xc6,0x05,0xda,0x80,0x0a,0xf1,
@@ -378,6 +441,13 @@ static const struct patch_digest phashes[] = {
 		0x11,0x5e,0x96,0x7e,0x71,0xe9,0xfc,0x74,
 	}
  },
+ { 0xa500012, {
+		0xeb,0x74,0x0d,0x47,0xa1,0x8e,0x09,0xe4,
+		0x93,0x4c,0xad,0x03,0x32,0x4c,0x38,0x16,
+		0x10,0x39,0xdd,0x06,0xaa,0xce,0xd6,0x0f,
+		0x62,0x83,0x9d,0x8e,0x64,0x55,0xbe,0x63,
+	}
+ },
  { 0xa601209, {
 		0x66,0x48,0xd4,0x09,0x05,0xcb,0x29,0x32,
 		0x66,0xb7,0x9a,0x76,0xcd,0x11,0xf3,0x30,
@@ -385,6 +455,13 @@ static const struct patch_digest phashes[] = {
 		0xe8,0x73,0xe2,0xd6,0xdb,0xd2,0x77,0x1d,
 	}
  },
+ { 0xa60120a, {
+		0x0c,0x8b,0x3d,0xfd,0x52,0x52,0x85,0x7d,
+		0x20,0x3a,0xe1,0x7e,0xa4,0x21,0x3b,0x7b,
+		0x17,0x86,0xae,0xac,0x13,0xb8,0x63,0x9d,
+		0x06,0x01,0xd0,0xa0,0x51,0x9a,0x91,0x2c,
+	}
+ },
  { 0xa704107, {
 		0xf3,0xc6,0x58,0x26,0xee,0xac,0x3f,0xd6,
 		0xce,0xa1,0x72,0x47,0x3b,0xba,0x2b,0x93,
@@ -392,6 +469,13 @@ static const struct patch_digest phashes[] = {
 		0x64,0x39,0x71,0x8c,0xce,0xe7,0x41,0x39,
 	}
  },
+ { 0xa704108, {
+		0xd7,0x55,0x15,0x2b,0xfe,0xc4,0xbc,0x93,
+		0xec,0x91,0xa0,0xae,0x45,0xb7,0xc3,0x98,
+		0x4e,0xff,0x61,0x77,0x88,0xc2,0x70,0x49,
+		0xe0,0x3a,0x1d,0x84,0x38,0x52,0xbf,0x5a,
+	}
+ },
  { 0xa705206, {
 		0x8d,0xc0,0x76,0xbd,0x58,0x9f,0x8f,0xa4,
 		0x12,0x9d,0x21,0xfb,0x48,0x21,0xbc,0xe7,
@@ -399,6 +483,13 @@ static const struct patch_digest phashes[] = {
 		0x03,0x35,0xe9,0xbe,0xfb,0x06,0xdf,0xfc,
 	}
  },
+ { 0xa705208, {
+		0x30,0x1d,0x55,0x24,0xbc,0x6b,0x5a,0x19,
+		0x0c,0x7d,0x1d,0x74,0xaa,0xd1,0xeb,0xd2,
+		0x16,0x62,0xf7,0x5b,0xe1,0x1f,0x18,0x11,
+		0x5c,0xf0,0x94,0x90,0x26,0xec,0x69,0xff,
+	}
+ },
  { 0xa708007, {
 		0x6b,0x76,0xcc,0x78,0xc5,0x8a,0xa3,0xe3,
 		0x32,0x2d,0x79,0xe4,0xc3,0x80,0xdb,0xb2,
@@ -406,6 +497,13 @@ static const struct patch_digest phashes[] = {
 		0xdf,0x92,0x73,0x84,0x87,0x3c,0x73,0x93,
 	}
  },
+ { 0xa708008, {
+		0x08,0x6e,0xf0,0x22,0x4b,0x8e,0xc4,0x46,
+		0x58,0x34,0xe6,0x47,0xa2,0x28,0xfd,0xab,
+		0x22,0x3d,0xdd,0xd8,0x52,0x9e,0x1d,0x16,
+		0xfa,0x01,0x68,0x14,0x79,0x3e,0xe8,0x6b,
+	}
+ },
  { 0xa70c005, {
 		0x88,0x5d,0xfb,0x79,0x64,0xd8,0x46,0x3b,
 		0x4a,0x83,0x8e,0x77,0x7e,0xcf,0xb3,0x0f,
@@ -413,6 +511,13 @@ static const struct patch_digest phashes[] = {
 		0xee,0x49,0xac,0xe1,0x8b,0x13,0xc5,0x13,
 	}
  },
+ { 0xa70c008, {
+		0x0f,0xdb,0x37,0xa1,0x10,0xaf,0xd4,0x21,
+		0x94,0x0d,0xa4,0xa2,0xe9,0x86,0x6c,0x0e,
+		0x85,0x7c,0x36,0x30,0xa3,0x3a,0x78,0x66,
+		0x18,0x10,0x60,0x0d,0x78,0x3d,0x44,0xd0,
+	}
+ },
  { 0xaa00116, {
 		0xe8,0x4c,0x2c,0x88,0xa1,0xac,0x24,0x63,
 		0x65,0xe5,0xaa,0x2d,0x16,0xa9,0xc3,0xf5,
@@ -441,4 +546,11 @@ static const struct patch_digest phashes[] = {
 		0x68,0x2f,0x46,0xee,0xfe,0xc6,0x6d,0xef,
 	}
  },
+ { 0xaa00216, {
+		0x79,0xfb,0x5b,0x9f,0xb6,0xe6,0xa8,0xf5,
+		0x4e,0x7c,0x4f,0x8e,0x1d,0xad,0xd0,0x08,
+		0xc2,0x43,0x7c,0x8b,0xe6,0xdb,0xd0,0xd2,
+		0xe8,0x39,0x26,0xc1,0xe5,0x5a,0x48,0xf1,
+	}
+ },
 };

arch/x86/kernel/cpu/scattered.c

@@ -50,6 +50,8 @@ static const struct cpuid_bit cpuid_bits[] = {
 	{ X86_FEATURE_MBA,			CPUID_EBX,  6, 0x80000008, 0 },
 	{ X86_FEATURE_SMBA,			CPUID_EBX,  2, 0x80000020, 0 },
 	{ X86_FEATURE_BMEC,			CPUID_EBX,  3, 0x80000020, 0 },
+	{ X86_FEATURE_TSA_SQ_NO,		CPUID_ECX,  1, 0x80000021, 0 },
+	{ X86_FEATURE_TSA_L1_NO,		CPUID_ECX,  2, 0x80000021, 0 },
 	{ X86_FEATURE_AMD_WORKLOAD_CLASS,	CPUID_EAX, 22, 0x80000021, 0 },
 	{ X86_FEATURE_PERFMON_V2,		CPUID_EAX,  0, 0x80000022, 0 },
 	{ X86_FEATURE_AMD_LBR_V2,		CPUID_EAX,  1, 0x80000022, 0 },

arch/x86/kernel/process.c

@@ -907,16 +907,24 @@ static __init bool prefer_mwait_c1_over_halt(void)
  */
 static __cpuidle void mwait_idle(void)
 {
+	if (need_resched())
+		return;
+
+	x86_idle_clear_cpu_buffers();
+
 	if (!current_set_polling_and_test()) {
 		const void *addr = &current_thread_info()->flags;
 
 		alternative_input("", "clflush (%[addr])", X86_BUG_CLFLUSH_MONITOR, [addr] "a" (addr));
 		__monitor(addr, 0, 0);
-		if (!need_resched()) {
+		if (need_resched())
-			__sti_mwait(0, 0);
+			goto out;
-			raw_local_irq_disable();
+
-		}
+		__sti_mwait(0, 0);
+		raw_local_irq_disable();
 	}
+
+out:
 	__current_clr_polling();
 }
 

arch/x86/kvm/cpuid.c

@@ -1165,6 +1165,8 @@ void kvm_set_cpu_caps(void)
 		 */
 		SYNTHESIZED_F(LFENCE_RDTSC),
 		/* SmmPgCfgLock */
+		/* 4: Resv */
+		SYNTHESIZED_F(VERW_CLEAR),
 		F(NULL_SEL_CLR_BASE),
 		/* UpperAddressIgnore */
 		F(AUTOIBRS),
@@ -1179,6 +1181,11 @@ void kvm_set_cpu_caps(void)
 		F(SRSO_USER_KERNEL_NO),
 	);
 
+	kvm_cpu_cap_init(CPUID_8000_0021_ECX,
+		SYNTHESIZED_F(TSA_SQ_NO),
+		SYNTHESIZED_F(TSA_L1_NO),
+	);
+
 	kvm_cpu_cap_init(CPUID_8000_0022_EAX,
 		F(PERFMON_V2),
 	);
@@ -1748,8 +1755,9 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
 		entry->eax = entry->ebx = entry->ecx = entry->edx = 0;
 		break;
 	case 0x80000021:
-		entry->ebx = entry->ecx = entry->edx = 0;
+		entry->ebx = entry->edx = 0;
 		cpuid_entry_override(entry, CPUID_8000_0021_EAX);
+		cpuid_entry_override(entry, CPUID_8000_0021_ECX);
 		break;
 	/* AMD Extended Performance Monitoring and Debug */
 	case 0x80000022: {

arch/x86/kvm/hyperv.c

@@ -1979,6 +1979,9 @@ int kvm_hv_vcpu_flush_tlb(struct kvm_vcpu *vcpu)
 		if (entries[i] == KVM_HV_TLB_FLUSHALL_ENTRY)
 			goto out_flush_all;
 
+		if (is_noncanonical_invlpg_address(entries[i], vcpu))
+			continue;
+
 		/*
 		 * Lower 12 bits of 'address' encode the number of additional
 		 * pages to flush.
@@ -2001,11 +2004,11 @@ out_flush_all:
 static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc)
 {
 	struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu);
+	unsigned long *vcpu_mask = hv_vcpu->vcpu_mask;
 	u64 *sparse_banks = hv_vcpu->sparse_banks;
 	struct kvm *kvm = vcpu->kvm;
 	struct hv_tlb_flush_ex flush_ex;
 	struct hv_tlb_flush flush;
-	DECLARE_BITMAP(vcpu_mask, KVM_MAX_VCPUS);
 	struct kvm_vcpu_hv_tlb_flush_fifo *tlb_flush_fifo;
 	/*
 	 * Normally, there can be no more than 'KVM_HV_TLB_FLUSH_FIFO_SIZE'

arch/x86/kvm/reverse_cpuid.h

@@ -52,6 +52,10 @@
 /* CPUID level 0x80000022 (EAX) */
 #define KVM_X86_FEATURE_PERFMON_V2	KVM_X86_FEATURE(CPUID_8000_0022_EAX, 0)
 
+/* CPUID level 0x80000021 (ECX) */
+#define KVM_X86_FEATURE_TSA_SQ_NO	KVM_X86_FEATURE(CPUID_8000_0021_ECX, 1)
+#define KVM_X86_FEATURE_TSA_L1_NO	KVM_X86_FEATURE(CPUID_8000_0021_ECX, 2)
+
 struct cpuid_reg {
 	u32 function;
 	u32 index;
@@ -82,6 +86,7 @@ static const struct cpuid_reg reverse_cpuid[] = {
 	[CPUID_8000_0022_EAX] = {0x80000022, 0, CPUID_EAX},
 	[CPUID_7_2_EDX]       = {         7, 2, CPUID_EDX},
 	[CPUID_24_0_EBX]      = {      0x24, 0, CPUID_EBX},
+	[CPUID_8000_0021_ECX] = {0x80000021, 0, CPUID_ECX},
 };
 
 /*
@@ -121,6 +126,8 @@ static __always_inline u32 __feature_translate(int x86_feature)
 	KVM_X86_TRANSLATE_FEATURE(PERFMON_V2);
 	KVM_X86_TRANSLATE_FEATURE(RRSBA_CTRL);
 	KVM_X86_TRANSLATE_FEATURE(BHI_CTRL);
+	KVM_X86_TRANSLATE_FEATURE(TSA_SQ_NO);
+	KVM_X86_TRANSLATE_FEATURE(TSA_L1_NO);
 	default:
 		return x86_feature;
 	}

arch/x86/kvm/svm/sev.c

@@ -1971,6 +1971,10 @@ static int sev_check_source_vcpus(struct kvm *dst, struct kvm *src)
 	struct kvm_vcpu *src_vcpu;
 	unsigned long i;
 
+	if (src->created_vcpus != atomic_read(&src->online_vcpus) ||
+	    dst->created_vcpus != atomic_read(&dst->online_vcpus))
+		return -EBUSY;
+
 	if (!sev_es_guest(src))
 		return 0;
 
@@ -4445,8 +4449,12 @@ static void sev_es_init_vmcb(struct vcpu_svm *svm)
 	 * the VMSA will be NULL if this vCPU is the destination for intrahost
 	 * migration, and will be copied later.
 	 */
-	if (svm->sev_es.vmsa && !svm->sev_es.snp_has_guest_vmsa)
+	if (!svm->sev_es.snp_has_guest_vmsa) {
-		svm->vmcb->control.vmsa_pa = __pa(svm->sev_es.vmsa);
+		if (svm->sev_es.vmsa)
+			svm->vmcb->control.vmsa_pa = __pa(svm->sev_es.vmsa);
+		else
+			svm->vmcb->control.vmsa_pa = INVALID_PAGE;
+	}
 
 	if (cpu_feature_enabled(X86_FEATURE_ALLOWED_SEV_FEATURES))
 		svm->vmcb->control.allowed_sev_features = sev->vmsa_features |

arch/x86/kvm/svm/vmenter.S

@@ -169,6 +169,9 @@ SYM_FUNC_START(__svm_vcpu_run)
 #endif
 	mov VCPU_RDI(%_ASM_DI), %_ASM_DI
 
+	/* Clobbers EFLAGS.ZF */
+	VM_CLEAR_CPU_BUFFERS
+
 	/* Enter guest mode */
 3:	vmrun %_ASM_AX
 4:
@@ -335,6 +338,9 @@ SYM_FUNC_START(__svm_sev_es_vcpu_run)
 	mov SVM_current_vmcb(%rdi), %rax
 	mov KVM_VMCB_pa(%rax), %rax
 
+	/* Clobbers EFLAGS.ZF */
+	VM_CLEAR_CPU_BUFFERS
+
 	/* Enter guest mode */
 1:	vmrun %rax
 2:

arch/x86/kvm/vmx/tdx.c

@@ -173,6 +173,9 @@ static void td_init_cpuid_entry2(struct kvm_cpuid_entry2 *entry, unsigned char i
 	tdx_clear_unsupported_cpuid(entry);
 }
 
+#define TDVMCALLINFO_GET_QUOTE				BIT(0)
+#define TDVMCALLINFO_SETUP_EVENT_NOTIFY_INTERRUPT	BIT(1)
+
 static int init_kvm_tdx_caps(const struct tdx_sys_info_td_conf *td_conf,
 			     struct kvm_tdx_capabilities *caps)
 {
@@ -188,6 +191,10 @@ static int init_kvm_tdx_caps(const struct tdx_sys_info_td_conf *td_conf,
 
 	caps->cpuid.nent = td_conf->num_cpuid_config;
 
+	caps->user_tdvmcallinfo_1_r11 =
+		TDVMCALLINFO_GET_QUOTE |
+		TDVMCALLINFO_SETUP_EVENT_NOTIFY_INTERRUPT;
+
 	for (i = 0; i < td_conf->num_cpuid_config; i++)
 		td_init_cpuid_entry2(&caps->cpuid.entries[i], i);
 
@@ -1530,6 +1537,27 @@ static int tdx_get_quote(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
+static int tdx_setup_event_notify_interrupt(struct kvm_vcpu *vcpu)
+{
+	struct vcpu_tdx *tdx = to_tdx(vcpu);
+	u64 vector = tdx->vp_enter_args.r12;
+
+	if (vector < 32 || vector > 255) {
+		tdvmcall_set_return_code(vcpu, TDVMCALL_STATUS_INVALID_OPERAND);
+		return 1;
+	}
+
+	vcpu->run->exit_reason = KVM_EXIT_TDX;
+	vcpu->run->tdx.flags = 0;
+	vcpu->run->tdx.nr = TDVMCALL_SETUP_EVENT_NOTIFY_INTERRUPT;
+	vcpu->run->tdx.setup_event_notify.ret = TDVMCALL_STATUS_SUBFUNC_UNSUPPORTED;
+	vcpu->run->tdx.setup_event_notify.vector = vector;
+
+	vcpu->arch.complete_userspace_io = tdx_complete_simple;
+
+	return 0;
+}
+
 static int handle_tdvmcall(struct kvm_vcpu *vcpu)
 {
 	switch (tdvmcall_leaf(vcpu)) {
@@ -1541,6 +1569,8 @@ static int handle_tdvmcall(struct kvm_vcpu *vcpu)
 		return tdx_get_td_vm_call_info(vcpu);
 	case TDVMCALL_GET_QUOTE:
 		return tdx_get_quote(vcpu);
+	case TDVMCALL_SETUP_EVENT_NOTIFY_INTERRUPT:
+		return tdx_setup_event_notify_interrupt(vcpu);
 	default:
 		break;
 	}

arch/x86/kvm/vmx/vmx.c

@@ -7291,7 +7291,7 @@ static noinstr void vmx_vcpu_enter_exit(struct kvm_vcpu *vcpu,
 		vmx_l1d_flush(vcpu);
 	else if (static_branch_unlikely(&cpu_buf_vm_clear) &&
 		 kvm_arch_has_assigned_device(vcpu->kvm))
-		mds_clear_cpu_buffers();
+		x86_clear_cpu_buffers();
 
 	vmx_disable_fb_clear(vmx);
 

arch/x86/kvm/x86.c

@@ -3258,9 +3258,11 @@ int kvm_guest_time_update(struct kvm_vcpu *v)
 
 	/* With all the info we got, fill in the values */
 
-	if (kvm_caps.has_tsc_control)
+	if (kvm_caps.has_tsc_control) {
 		tgt_tsc_khz = kvm_scale_tsc(tgt_tsc_khz,
 					    v->arch.l1_tsc_scaling_ratio);
+		tgt_tsc_khz = tgt_tsc_khz ? : 1;
+	}
 
 	if (unlikely(vcpu->hw_tsc_khz != tgt_tsc_khz)) {
 		kvm_get_time_scale(NSEC_PER_SEC, tgt_tsc_khz * 1000LL,

arch/x86/kvm/xen.c

@@ -1971,8 +1971,19 @@ int kvm_xen_setup_evtchn(struct kvm *kvm,
 {
 	struct kvm_vcpu *vcpu;
 
-	if (ue->u.xen_evtchn.port >= max_evtchn_port(kvm))
+	/*
-		return -EINVAL;
+	 * Don't check for the port being within range of max_evtchn_port().
+	 * Userspace can configure what ever targets it likes; events just won't
+	 * be delivered if/while the target is invalid, just like userspace can
+	 * configure MSIs which target non-existent APICs.
+	 *
+	 * This allow on Live Migration and Live Update, the IRQ routing table
+	 * can be restored *independently* of other things like creating vCPUs,
+	 * without imposing an ordering dependency on userspace.  In this
+	 * particular case, the problematic ordering would be with setting the
+	 * Xen 'long mode' flag, which changes max_evtchn_port() to allow 4096
+	 * instead of 1024 event channels.
+	 */
 
 	/* We only support 2 level event channels for now */
 	if (ue->u.xen_evtchn.priority != KVM_IRQ_ROUTING_XEN_EVTCHN_PRIO_2LEVEL)

drivers/acpi/battery.c

@@ -243,23 +243,10 @@ static int acpi_battery_get_property(struct power_supply *psy,
 		break;
 	case POWER_SUPPLY_PROP_CURRENT_NOW:
 	case POWER_SUPPLY_PROP_POWER_NOW:
-		if (battery->rate_now == ACPI_BATTERY_VALUE_UNKNOWN) {
+		if (battery->rate_now == ACPI_BATTERY_VALUE_UNKNOWN)
 			ret = -ENODEV;
-			break;
+		else
-		}
+			val->intval = battery->rate_now * 1000;
-
-		val->intval = battery->rate_now * 1000;
-		/*
-		 * When discharging, the current should be reported as a
-		 * negative number as per the power supply class interface
-		 * definition.
-		 */
-		if (psp == POWER_SUPPLY_PROP_CURRENT_NOW &&
-		    (battery->state & ACPI_BATTERY_STATE_DISCHARGING) &&
-		    acpi_battery_handle_discharging(battery)
-				== POWER_SUPPLY_STATUS_DISCHARGING)
-			val->intval = -val->intval;
-
 		break;
 	case POWER_SUPPLY_PROP_CHARGE_FULL_DESIGN:
 	case POWER_SUPPLY_PROP_ENERGY_FULL_DESIGN:

drivers/base/cpu.c

@@ -602,6 +602,7 @@ CPU_SHOW_VULN_FALLBACK(reg_file_data_sampling);
 CPU_SHOW_VULN_FALLBACK(ghostwrite);
 CPU_SHOW_VULN_FALLBACK(old_microcode);
 CPU_SHOW_VULN_FALLBACK(indirect_target_selection);
+CPU_SHOW_VULN_FALLBACK(tsa);
 
 static DEVICE_ATTR(meltdown, 0444, cpu_show_meltdown, NULL);
 static DEVICE_ATTR(spectre_v1, 0444, cpu_show_spectre_v1, NULL);
@@ -620,6 +621,7 @@ static DEVICE_ATTR(reg_file_data_sampling, 0444, cpu_show_reg_file_data_sampling
 static DEVICE_ATTR(ghostwrite, 0444, cpu_show_ghostwrite, NULL);
 static DEVICE_ATTR(old_microcode, 0444, cpu_show_old_microcode, NULL);
 static DEVICE_ATTR(indirect_target_selection, 0444, cpu_show_indirect_target_selection, NULL);
+static DEVICE_ATTR(tsa, 0444, cpu_show_tsa, NULL);
 
 static struct attribute *cpu_root_vulnerabilities_attrs[] = {
 	&dev_attr_meltdown.attr,
@@ -639,6 +641,7 @@ static struct attribute *cpu_root_vulnerabilities_attrs[] = {
 	&dev_attr_ghostwrite.attr,
 	&dev_attr_old_microcode.attr,
 	&dev_attr_indirect_target_selection.attr,
+	&dev_attr_tsa.attr,
 	NULL
 };
 

drivers/base/power/main.c

@@ -1236,6 +1236,7 @@ void dpm_complete(pm_message_t state)
  */
 void dpm_resume_end(pm_message_t state)
 {
+	pm_restore_gfp_mask();
 	dpm_resume(state);
 	dpm_complete(state);
 }
@@ -2176,8 +2177,10 @@ int dpm_suspend_start(pm_message_t state)
 	error = dpm_prepare(state);
 	if (error)
 		dpm_save_failed_step(SUSPEND_PREPARE);
-	else
+	else {
+		pm_restrict_gfp_mask();
 		error = dpm_suspend(state);
+	}
 
 	dpm_show_time(starttime, state, error, "start");
 	return error;

drivers/block/brd.c

@@ -64,13 +64,15 @@ static struct page *brd_insert_page(struct brd_device *brd, sector_t sector,
 
 	rcu_read_unlock();
 	page = alloc_page(gfp | __GFP_ZERO | __GFP_HIGHMEM);
-	rcu_read_lock();
+	if (!page) {
-	if (!page)
+		rcu_read_lock();
 		return ERR_PTR(-ENOMEM);
+	}
 
 	xa_lock(&brd->brd_pages);
 	ret = __xa_cmpxchg(&brd->brd_pages, sector >> PAGE_SECTORS_SHIFT, NULL,
 			page, gfp);
+	rcu_read_lock();
 	if (ret) {
 		xa_unlock(&brd->brd_pages);
 		__free_page(page);

drivers/block/ublk_drv.c

@@ -1442,15 +1442,16 @@ static void ublk_queue_rqs(struct rq_list *rqlist)
 		struct ublk_queue *this_q = req->mq_hctx->driver_data;
 		struct ublk_io *this_io = &this_q->ios[req->tag];
 
+		if (ublk_prep_req(this_q, req, true) != BLK_STS_OK) {
+			rq_list_add_tail(&requeue_list, req);
+			continue;
+		}
+
 		if (io && !ublk_belong_to_same_batch(io, this_io) &&
 				!rq_list_empty(&submit_list))
 			ublk_queue_cmd_list(io, &submit_list);
 		io = this_io;
-
+		rq_list_add_tail(&submit_list, req);
-		if (ublk_prep_req(this_q, req, true) == BLK_STS_OK)
-			rq_list_add_tail(&submit_list, req);
-		else
-			rq_list_add_tail(&requeue_list, req);
 	}
 
 	if (!rq_list_empty(&submit_list))

drivers/dma-buf/dma-resv.c

@@ -685,11 +685,13 @@ long dma_resv_wait_timeout(struct dma_resv *obj, enum dma_resv_usage usage,
 	dma_resv_iter_begin(&cursor, obj, usage);
 	dma_resv_for_each_fence_unlocked(&cursor, fence) {
 
-		ret = dma_fence_wait_timeout(fence, intr, ret);
+		ret = dma_fence_wait_timeout(fence, intr, timeout);
-		if (ret <= 0) {
+		if (ret <= 0)
-			dma_resv_iter_end(&cursor);
+			break;
-			return ret;
+
-		}
+		/* Even for zero timeout the return value is 1 */
+		if (timeout)
+			timeout = ret;
 	}
 	dma_resv_iter_end(&cursor);
 

drivers/edac/ecs.c

@@ -170,8 +170,10 @@ static int ecs_create_desc(struct device *ecs_dev, const struct attribute_group
 		fru_ctx->dev_attr[ECS_RESET]		= EDAC_ECS_ATTR_WO(reset, fru);
 		fru_ctx->dev_attr[ECS_THRESHOLD]	= EDAC_ECS_ATTR_RW(threshold, fru);
 
-		for (i = 0; i < ECS_MAX_ATTRS; i++)
+		for (i = 0; i < ECS_MAX_ATTRS; i++) {
+			sysfs_attr_init(&fru_ctx->dev_attr[i].dev_attr.attr);
 			fru_ctx->ecs_attrs[i] = &fru_ctx->dev_attr[i].dev_attr.attr;
+		}
 
 		sprintf(fru_ctx->name, "%s%d", EDAC_ECS_FRU_NAME, fru);
 		group->name = fru_ctx->name;

drivers/edac/mem_repair.c

@@ -333,6 +333,7 @@ static int mem_repair_create_desc(struct device *dev,
 	for (i = 0; i < MR_MAX_ATTRS; i++) {
 		memcpy(&ctx->mem_repair_dev_attr[i],
 		       &dev_attr[i], sizeof(dev_attr[i]));
+		sysfs_attr_init(&ctx->mem_repair_dev_attr[i].dev_attr.attr);
 		ctx->mem_repair_attrs[i] =
 			&ctx->mem_repair_dev_attr[i].dev_attr.attr;
 	}

drivers/edac/scrub.c

@@ -176,6 +176,7 @@ static int scrub_create_desc(struct device *scrub_dev,
 	group = &scrub_ctx->group;
 	for (i = 0; i < SCRUB_MAX_ATTRS; i++) {
 		memcpy(&scrub_ctx->scrub_dev_attr[i], &dev_attr[i], sizeof(dev_attr[i]));
+		sysfs_attr_init(&scrub_ctx->scrub_dev_attr[i].dev_attr.attr);
 		scrub_ctx->scrub_attrs[i] = &scrub_ctx->scrub_dev_attr[i].dev_attr.attr;
 	}
 	sprintf(scrub_ctx->name, "%s%d", "scrub", instance);

drivers/firmware/arm_ffa/driver.c

@@ -110,7 +110,7 @@ struct ffa_drv_info {
 	struct work_struct sched_recv_irq_work;
 	struct xarray partition_info;
 	DECLARE_HASHTABLE(notifier_hash, ilog2(FFA_MAX_NOTIFICATIONS));
-	struct mutex notify_lock; /* lock to protect notifier hashtable  */
+	rwlock_t notify_lock; /* lock to protect notifier hashtable  */
 };
 
 static struct ffa_drv_info *drv_info;
@@ -1250,13 +1250,12 @@ notifier_hnode_get_by_type(u16 notify_id, enum notify_type type)
 	return NULL;
 }
 
-static int
+static int update_notifier_cb(struct ffa_device *dev, int notify_id,
-update_notifier_cb(struct ffa_device *dev, int notify_id, void *cb,
+			      struct notifier_cb_info *cb, bool is_framework)
-		   void *cb_data, bool is_registration, bool is_framework)
 {
 	struct notifier_cb_info *cb_info = NULL;
 	enum notify_type type = ffa_notify_type_get(dev->vm_id);
-	bool cb_found;
+	bool cb_found, is_registration = !!cb;
 
 	if (is_framework)
 		cb_info = notifier_hnode_get_by_vmid_uuid(notify_id, dev->vm_id,
@@ -1270,20 +1269,10 @@ update_notifier_cb(struct ffa_device *dev, int notify_id, void *cb,
 		return -EINVAL;
 
 	if (is_registration) {
-		cb_info = kzalloc(sizeof(*cb_info), GFP_KERNEL);
+		hash_add(drv_info->notifier_hash, &cb->hnode, notify_id);
-		if (!cb_info)
-			return -ENOMEM;
-
-		cb_info->dev = dev;
-		cb_info->cb_data = cb_data;
-		if (is_framework)
-			cb_info->fwk_cb = cb;
-		else
-			cb_info->cb = cb;
-
-		hash_add(drv_info->notifier_hash, &cb_info->hnode, notify_id);
 	} else {
 		hash_del(&cb_info->hnode);
+		kfree(cb_info);
 	}
 
 	return 0;
@@ -1300,20 +1289,19 @@ static int __ffa_notify_relinquish(struct ffa_device *dev, int notify_id,
 	if (notify_id >= FFA_MAX_NOTIFICATIONS)
 		return -EINVAL;
 
-	mutex_lock(&drv_info->notify_lock);
+	write_lock(&drv_info->notify_lock);
 
-	rc = update_notifier_cb(dev, notify_id, NULL, NULL, false,
+	rc = update_notifier_cb(dev, notify_id, NULL, is_framework);
-				is_framework);
 	if (rc) {
 		pr_err("Could not unregister notification callback\n");
-		mutex_unlock(&drv_info->notify_lock);
+		write_unlock(&drv_info->notify_lock);
 		return rc;
 	}
 
 	if (!is_framework)
 		rc = ffa_notification_unbind(dev->vm_id, BIT(notify_id));
 
-	mutex_unlock(&drv_info->notify_lock);
+	write_unlock(&drv_info->notify_lock);
 
 	return rc;
 }
@@ -1334,6 +1322,7 @@ static int __ffa_notify_request(struct ffa_device *dev, bool is_per_vcpu,
 {
 	int rc;
 	u32 flags = 0;
+	struct notifier_cb_info *cb_info = NULL;
 
 	if (ffa_notifications_disabled())
 		return -EOPNOTSUPP;
@@ -1341,28 +1330,40 @@ static int __ffa_notify_request(struct ffa_device *dev, bool is_per_vcpu,
 	if (notify_id >= FFA_MAX_NOTIFICATIONS)
 		return -EINVAL;
 
-	mutex_lock(&drv_info->notify_lock);
+	cb_info = kzalloc(sizeof(*cb_info), GFP_KERNEL);
+	if (!cb_info)
+		return -ENOMEM;
+
+	cb_info->dev = dev;
+	cb_info->cb_data = cb_data;
+	if (is_framework)
+		cb_info->fwk_cb = cb;
+	else
+		cb_info->cb = cb;
+
+	write_lock(&drv_info->notify_lock);
 
 	if (!is_framework) {
 		if (is_per_vcpu)
 			flags = PER_VCPU_NOTIFICATION_FLAG;
 
 		rc = ffa_notification_bind(dev->vm_id, BIT(notify_id), flags);
-		if (rc) {
+		if (rc)
-			mutex_unlock(&drv_info->notify_lock);
+			goto out_unlock_free;
-			return rc;
-		}
 	}
 
-	rc = update_notifier_cb(dev, notify_id, cb, cb_data, true,
+	rc = update_notifier_cb(dev, notify_id, cb_info, is_framework);
-				is_framework);
 	if (rc) {
 		pr_err("Failed to register callback for %d - %d\n",
 		       notify_id, rc);
 		if (!is_framework)
 			ffa_notification_unbind(dev->vm_id, BIT(notify_id));
 	}
-	mutex_unlock(&drv_info->notify_lock);
+
+out_unlock_free:
+	write_unlock(&drv_info->notify_lock);
+	if (rc)
+		kfree(cb_info);
 
 	return rc;
 }
@@ -1406,9 +1407,9 @@ static void handle_notif_callbacks(u64 bitmap, enum notify_type type)
 		if (!(bitmap & 1))
 			continue;
 
-		mutex_lock(&drv_info->notify_lock);
+		read_lock(&drv_info->notify_lock);
 		cb_info = notifier_hnode_get_by_type(notify_id, type);
-		mutex_unlock(&drv_info->notify_lock);
+		read_unlock(&drv_info->notify_lock);
 
 		if (cb_info && cb_info->cb)
 			cb_info->cb(notify_id, cb_info->cb_data);
@@ -1446,9 +1447,9 @@ static void handle_fwk_notif_callbacks(u32 bitmap)
 
 	ffa_rx_release();
 
-	mutex_lock(&drv_info->notify_lock);
+	read_lock(&drv_info->notify_lock);
 	cb_info = notifier_hnode_get_by_vmid_uuid(notify_id, target, &uuid);
-	mutex_unlock(&drv_info->notify_lock);
+	read_unlock(&drv_info->notify_lock);
 
 	if (cb_info && cb_info->fwk_cb)
 		cb_info->fwk_cb(notify_id, cb_info->cb_data, buf);
@@ -1973,7 +1974,7 @@ static void ffa_notifications_setup(void)
 		goto cleanup;
 
 	hash_init(drv_info->notifier_hash);
-	mutex_init(&drv_info->notify_lock);
+	rwlock_init(&drv_info->notify_lock);
 
 	drv_info->notif_enabled = true;
 	return;

drivers/firmware/efi/libstub/zboot.lds

@@ -29,14 +29,12 @@ SECTIONS
 		. = _etext;
 	}
 
-#ifdef CONFIG_EFI_SBAT
         .sbat : ALIGN(4096) {
 		_sbat = .;
 		*(.sbat)
 		_esbat = ALIGN(4096);
 		. = _esbat;
 	}
-#endif
 
 	.data : ALIGN(4096) {
 		_data = .;
@@ -60,6 +58,6 @@ SECTIONS
 PROVIDE(__efistub__gzdata_size =
 		ABSOLUTE(__efistub__gzdata_end - __efistub__gzdata_start));
 
-PROVIDE(__data_rawsize = ABSOLUTE(_edata - _etext));
+PROVIDE(__data_rawsize = ABSOLUTE(_edata - _data));
-PROVIDE(__data_size = ABSOLUTE(_end - _etext));
+PROVIDE(__data_size = ABSOLUTE(_end - _data));
 PROVIDE(__sbat_size = ABSOLUTE(_esbat - _sbat));

drivers/firmware/samsung/exynos-acpm.c

@@ -430,6 +430,9 @@ int acpm_do_xfer(const struct acpm_handle *handle, const struct acpm_xfer *xfer)
 		return -EOPNOTSUPP;
 	}
 
+	msg.chan_id = xfer->acpm_chan_id;
+	msg.chan_type = EXYNOS_MBOX_CHAN_TYPE_DOORBELL;
+
 	scoped_guard(mutex, &achan->tx_lock) {
 		tx_front = readl(achan->tx.front);
 		idx = (tx_front + 1) % achan->qlen;
@@ -446,25 +449,15 @@ int acpm_do_xfer(const struct acpm_handle *handle, const struct acpm_xfer *xfer)
 
 		/* Advance TX front. */
 		writel(idx, achan->tx.front);
+
+		ret = mbox_send_message(achan->chan, (void *)&msg);
+		if (ret < 0)
+			return ret;
+
+		mbox_client_txdone(achan->chan, 0);
 	}
 
-	msg.chan_id = xfer->acpm_chan_id;
+	return acpm_wait_for_message_response(achan, xfer);
-	msg.chan_type = EXYNOS_MBOX_CHAN_TYPE_DOORBELL;
-	ret = mbox_send_message(achan->chan, (void *)&msg);
-	if (ret < 0)
-		return ret;
-
-	ret = acpm_wait_for_message_response(achan, xfer);
-
-	/*
-	 * NOTE: we might prefer not to need the mailbox ticker to manage the
-	 * transfer queueing since the protocol layer queues things by itself.
-	 * Unfortunately, we have to kick the mailbox framework after we have
-	 * received our message.
-	 */
-	mbox_client_txdone(achan->chan, ret);
-
-	return ret;
 }
 
 /**

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c

@@ -561,6 +561,13 @@ static uint32_t read_vmid_from_vmfault_reg(struct amdgpu_device *adev)
 	return REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, VMID);
 }
 
+static uint32_t kgd_hqd_sdma_get_doorbell(struct amdgpu_device *adev,
+					  int engine, int queue)
+
+{
+	return 0;
+}
+
 const struct kfd2kgd_calls gfx_v7_kfd2kgd = {
 	.program_sh_mem_settings = kgd_program_sh_mem_settings,
 	.set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping,
@@ -578,4 +585,5 @@ const struct kfd2kgd_calls gfx_v7_kfd2kgd = {
 	.set_scratch_backing_va = set_scratch_backing_va,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
 	.read_vmid_from_vmfault_reg = read_vmid_from_vmfault_reg,
+	.hqd_sdma_get_doorbell = kgd_hqd_sdma_get_doorbell,
 };

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c

@@ -582,6 +582,13 @@ static void set_vm_context_page_table_base(struct amdgpu_device *adev,
 			lower_32_bits(page_table_base));
 }
 
+static uint32_t kgd_hqd_sdma_get_doorbell(struct amdgpu_device *adev,
+					  int engine, int queue)
+
+{
+	return 0;
+}
+
 const struct kfd2kgd_calls gfx_v8_kfd2kgd = {
 	.program_sh_mem_settings = kgd_program_sh_mem_settings,
 	.set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping,
@@ -599,4 +606,5 @@ const struct kfd2kgd_calls gfx_v8_kfd2kgd = {
 			get_atc_vmid_pasid_mapping_info,
 	.set_scratch_backing_va = set_scratch_backing_va,
 	.set_vm_context_page_table_base = set_vm_context_page_table_base,
+	.hqd_sdma_get_doorbell = kgd_hqd_sdma_get_doorbell,
 };

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

@@ -944,6 +944,7 @@ static void amdgpu_ctx_mgr_entity_fini(struct amdgpu_ctx_mgr *mgr)
 				drm_sched_entity_fini(entity);
 			}
 		}
+		kref_put(&ctx->refcount, amdgpu_ctx_fini);
 	}
 }
 

drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c

@@ -45,6 +45,7 @@
 #include "amdgpu_ras.h"
 
 MODULE_FIRMWARE("amdgpu/sdma_4_4_2.bin");
+MODULE_FIRMWARE("amdgpu/sdma_4_4_4.bin");
 MODULE_FIRMWARE("amdgpu/sdma_4_4_5.bin");
 
 static const struct amdgpu_hwip_reg_entry sdma_reg_list_4_4_2[] = {

drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c

@@ -1543,8 +1543,13 @@ static int sdma_v5_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid)
 {
 	struct amdgpu_device *adev = ring->adev;
 	u32 inst_id = ring->me;
+	int r;
 
-	return amdgpu_sdma_reset_engine(adev, inst_id);
+	amdgpu_amdkfd_suspend(adev, true);
+	r = amdgpu_sdma_reset_engine(adev, inst_id);
+	amdgpu_amdkfd_resume(adev, true);
+
+	return r;
 }
 
 static int sdma_v5_0_stop_queue(struct amdgpu_ring *ring)

drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c

@@ -1456,8 +1456,13 @@ static int sdma_v5_2_reset_queue(struct amdgpu_ring *ring, unsigned int vmid)
 {
 	struct amdgpu_device *adev = ring->adev;
 	u32 inst_id = ring->me;
+	int r;
 
-	return amdgpu_sdma_reset_engine(adev, inst_id);
+	amdgpu_amdkfd_suspend(adev, true);
+	r = amdgpu_sdma_reset_engine(adev, inst_id);
+	amdgpu_amdkfd_resume(adev, true);
+
+	return r;
 }
 
 static int sdma_v5_2_stop_queue(struct amdgpu_ring *ring)

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

@@ -1171,13 +1171,12 @@ svm_range_split_head(struct svm_range *prange, uint64_t new_start,
 }
 
 static void
-svm_range_add_child(struct svm_range *prange, struct mm_struct *mm,
+svm_range_add_child(struct svm_range *prange, struct svm_range *pchild, enum svm_work_list_ops op)
-		    struct svm_range *pchild, enum svm_work_list_ops op)
 {
 	pr_debug("add child 0x%p [0x%lx 0x%lx] to prange 0x%p child list %d\n",
 		 pchild, pchild->start, pchild->last, prange, op);
 
-	pchild->work_item.mm = mm;
+	pchild->work_item.mm = NULL;
 	pchild->work_item.op = op;
 	list_add_tail(&pchild->child_list, &prange->child_list);
 }
@@ -1278,7 +1277,7 @@ svm_range_get_pte_flags(struct kfd_node *node,
 				mapping_flags |= ext_coherent ? AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC;
 		/* system memory accessed by the dGPU */
 		} else {
-			if (gc_ip_version < IP_VERSION(9, 5, 0))
+			if (gc_ip_version < IP_VERSION(9, 5, 0) || ext_coherent)
 				mapping_flags |= AMDGPU_VM_MTYPE_UC;
 			else
 				mapping_flags |= AMDGPU_VM_MTYPE_NC;
@@ -2394,15 +2393,17 @@ svm_range_add_list_work(struct svm_range_list *svms, struct svm_range *prange,
 		    prange->work_item.op != SVM_OP_UNMAP_RANGE)
 			prange->work_item.op = op;
 	} else {
-		prange->work_item.op = op;
+		/* Pairs with mmput in deferred_list_work.
-
+		 * If process is exiting and mm is gone, don't update mmu notifier.
-		/* Pairs with mmput in deferred_list_work */
+		 */
-		mmget(mm);
+		if (mmget_not_zero(mm)) {
-		prange->work_item.mm = mm;
+			prange->work_item.mm = mm;
-		list_add_tail(&prange->deferred_list,
+			prange->work_item.op = op;
-			      &prange->svms->deferred_range_list);
+			list_add_tail(&prange->deferred_list,
-		pr_debug("add prange 0x%p [0x%lx 0x%lx] to work list op %d\n",
+				      &prange->svms->deferred_range_list);
-			 prange, prange->start, prange->last, op);
+			pr_debug("add prange 0x%p [0x%lx 0x%lx] to work list op %d\n",
+				 prange, prange->start, prange->last, op);
+		}
 	}
 	spin_unlock(&svms->deferred_list_lock);
 }
@@ -2416,8 +2417,7 @@ void schedule_deferred_list_work(struct svm_range_list *svms)
 }
 
 static void
-svm_range_unmap_split(struct mm_struct *mm, struct svm_range *parent,
+svm_range_unmap_split(struct svm_range *parent, struct svm_range *prange, unsigned long start,
-		      struct svm_range *prange, unsigned long start,
 		      unsigned long last)
 {
 	struct svm_range *head;
@@ -2438,12 +2438,12 @@ svm_range_unmap_split(struct mm_struct *mm, struct svm_range *parent,
 		svm_range_split(tail, last + 1, tail->last, &head);
 
 	if (head != prange && tail != prange) {
-		svm_range_add_child(parent, mm, head, SVM_OP_UNMAP_RANGE);
+		svm_range_add_child(parent, head, SVM_OP_UNMAP_RANGE);
-		svm_range_add_child(parent, mm, tail, SVM_OP_ADD_RANGE);
+		svm_range_add_child(parent, tail, SVM_OP_ADD_RANGE);
 	} else if (tail != prange) {
-		svm_range_add_child(parent, mm, tail, SVM_OP_UNMAP_RANGE);
+		svm_range_add_child(parent, tail, SVM_OP_UNMAP_RANGE);
 	} else if (head != prange) {
-		svm_range_add_child(parent, mm, head, SVM_OP_UNMAP_RANGE);
+		svm_range_add_child(parent, head, SVM_OP_UNMAP_RANGE);
 	} else if (parent != prange) {
 		prange->work_item.op = SVM_OP_UNMAP_RANGE;
 	}
@@ -2520,14 +2520,14 @@ svm_range_unmap_from_cpu(struct mm_struct *mm, struct svm_range *prange,
 		l = min(last, pchild->last);
 		if (l >= s)
 			svm_range_unmap_from_gpus(pchild, s, l, trigger);
-		svm_range_unmap_split(mm, prange, pchild, start, last);
+		svm_range_unmap_split(prange, pchild, start, last);
 		mutex_unlock(&pchild->lock);
 	}
 	s = max(start, prange->start);
 	l = min(last, prange->last);
 	if (l >= s)
 		svm_range_unmap_from_gpus(prange, s, l, trigger);
-	svm_range_unmap_split(mm, prange, prange, start, last);
+	svm_range_unmap_split(prange, prange, start, last);
 
 	if (unmap_parent)
 		svm_range_add_list_work(svms, prange, mm, SVM_OP_UNMAP_RANGE);
@@ -2570,8 +2570,6 @@ svm_range_cpu_invalidate_pagetables(struct mmu_interval_notifier *mni,
 
 	if (range->event == MMU_NOTIFY_RELEASE)
 		return true;
-	if (!mmget_not_zero(mni->mm))
-		return true;
 
 	start = mni->interval_tree.start;
 	last = mni->interval_tree.last;
@@ -2598,7 +2596,6 @@ svm_range_cpu_invalidate_pagetables(struct mmu_interval_notifier *mni,
 	}
 
 	svm_range_unlock(prange);
-	mmput(mni->mm);
 
 	return true;
 }

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

@@ -3610,13 +3610,15 @@ static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector)
 
 	luminance_range = &conn_base->display_info.luminance_range;
 
-	if (luminance_range->max_luminance) {
+	if (luminance_range->max_luminance)
-		caps->aux_min_input_signal = luminance_range->min_luminance;
 		caps->aux_max_input_signal = luminance_range->max_luminance;
-	} else {
+	else
-		caps->aux_min_input_signal = 0;
 		caps->aux_max_input_signal = 512;
-	}
+
+	if (luminance_range->min_luminance)
+		caps->aux_min_input_signal = luminance_range->min_luminance;
+	else
+		caps->aux_min_input_signal = 1;
 
 	min_input_signal_override = drm_get_panel_min_brightness_quirk(aconnector->drm_edid);
 	if (min_input_signal_override >= 0)

drivers/gpu/drm/amd/display/dc/dc_hw_types.h

@@ -974,6 +974,7 @@ struct dc_crtc_timing {
 	uint32_t pix_clk_100hz;
 
 	uint32_t min_refresh_in_uhz;
+	uint32_t max_refresh_in_uhz;
 
 	uint32_t vic;
 	uint32_t hdmi_vic;

drivers/gpu/drm/amd/display/modules/freesync/freesync.c

@@ -155,6 +155,14 @@ unsigned int mod_freesync_calc_v_total_from_refresh(
 		v_total = div64_u64(div64_u64(((unsigned long long)(
 				frame_duration_in_ns) * (stream->timing.pix_clk_100hz / 10)),
 				stream->timing.h_total), 1000000);
+	} else if (refresh_in_uhz >= stream->timing.max_refresh_in_uhz) {
+		/* When the target refresh rate is the maximum panel refresh rate
+		 * round up the vtotal value to prevent off-by-one error causing
+		 * v_total_min to be below the panel's lower bound
+		 */
+		v_total = div64_u64(div64_u64(((unsigned long long)(
+				frame_duration_in_ns) * (stream->timing.pix_clk_100hz / 10)),
+				stream->timing.h_total) + (1000000 - 1), 1000000);
 	} else {
 		v_total = div64_u64(div64_u64(((unsigned long long)(
 				frame_duration_in_ns) * (stream->timing.pix_clk_100hz / 10)),

drivers/gpu/drm/bridge/aux-hpd-bridge.c

@@ -64,10 +64,11 @@ struct auxiliary_device *devm_drm_dp_hpd_bridge_alloc(struct device *parent, str
 	adev->id = ret;
 	adev->name = "dp_hpd_bridge";
 	adev->dev.parent = parent;
-	adev->dev.of_node = of_node_get(parent->of_node);
 	adev->dev.release = drm_aux_hpd_bridge_release;
 	adev->dev.platform_data = of_node_get(np);
 
+	device_set_of_node_from_dev(&adev->dev, parent);
+
 	ret = auxiliary_device_init(adev);
 	if (ret) {
 		of_node_put(adev->dev.platform_data);

drivers/gpu/drm/bridge/panel.c

@@ -299,6 +299,7 @@ struct drm_bridge *drm_panel_bridge_add_typed(struct drm_panel *panel,
 	panel_bridge->bridge.of_node = panel->dev->of_node;
 	panel_bridge->bridge.ops = DRM_BRIDGE_OP_MODES;
 	panel_bridge->bridge.type = connector_type;
+	panel_bridge->bridge.pre_enable_prev_first = panel->prepare_prev_first;
 
 	drm_bridge_add(&panel_bridge->bridge);
 
@@ -413,8 +414,6 @@ struct drm_bridge *devm_drm_panel_bridge_add_typed(struct device *dev,
 		return bridge;
 	}
 
-	bridge->pre_enable_prev_first = panel->prepare_prev_first;
-
 	*ptr = bridge;
 	devres_add(dev, ptr);
 
@@ -456,8 +455,6 @@ struct drm_bridge *drmm_panel_bridge_add(struct drm_device *drm,
 	if (ret)
 		return ERR_PTR(ret);
 
-	bridge->pre_enable_prev_first = panel->prepare_prev_first;
-
 	return bridge;
 }
 EXPORT_SYMBOL(drmm_panel_bridge_add);

drivers/gpu/drm/bridge/samsung-dsim.c

@@ -1095,7 +1095,7 @@ static void samsung_dsim_send_to_fifo(struct samsung_dsim *dsi,
 	bool first = !xfer->tx_done;
 	u32 reg;
 
-	dev_dbg(dev, "< xfer %pK: tx len %u, done %u, rx len %u, done %u\n",
+	dev_dbg(dev, "< xfer %p: tx len %u, done %u, rx len %u, done %u\n",
 		xfer, length, xfer->tx_done, xfer->rx_len, xfer->rx_done);
 
 	if (length > DSI_TX_FIFO_SIZE)
@@ -1293,7 +1293,7 @@ static bool samsung_dsim_transfer_finish(struct samsung_dsim *dsi)
 	spin_unlock_irqrestore(&dsi->transfer_lock, flags);
 
 	dev_dbg(dsi->dev,
-		"> xfer %pK, tx_len %zu, tx_done %u, rx_len %u, rx_done %u\n",
+		"> xfer %p, tx_len %zu, tx_done %u, rx_len %u, rx_done %u\n",
 		xfer, xfer->packet.payload_length, xfer->tx_done, xfer->rx_len,
 		xfer->rx_done);
 

drivers/gpu/drm/drm_gem.c

@@ -212,6 +212,35 @@ void drm_gem_private_object_fini(struct drm_gem_object *obj)
 }
 EXPORT_SYMBOL(drm_gem_private_object_fini);
 
+static void drm_gem_object_handle_get(struct drm_gem_object *obj)
+{
+	struct drm_device *dev = obj->dev;
+
+	drm_WARN_ON(dev, !mutex_is_locked(&dev->object_name_lock));
+
+	if (obj->handle_count++ == 0)
+		drm_gem_object_get(obj);
+}
+
+/**
+ * drm_gem_object_handle_get_unlocked - acquire reference on user-space handles
+ * @obj: GEM object
+ *
+ * Acquires a reference on the GEM buffer object's handle. Required
+ * to keep the GEM object alive. Call drm_gem_object_handle_put_unlocked()
+ * to release the reference.
+ */
+void drm_gem_object_handle_get_unlocked(struct drm_gem_object *obj)
+{
+	struct drm_device *dev = obj->dev;
+
+	guard(mutex)(&dev->object_name_lock);
+
+	drm_WARN_ON(dev, !obj->handle_count); /* first ref taken in create-tail helper */
+	drm_gem_object_handle_get(obj);
+}
+EXPORT_SYMBOL(drm_gem_object_handle_get_unlocked);
+
 /**
  * drm_gem_object_handle_free - release resources bound to userspace handles
  * @obj: GEM object to clean up.
@@ -242,8 +271,14 @@ static void drm_gem_object_exported_dma_buf_free(struct drm_gem_object *obj)
 	}
 }
 
-static void
+/**
-drm_gem_object_handle_put_unlocked(struct drm_gem_object *obj)
+ * drm_gem_object_handle_put_unlocked - releases reference on user-space handles
+ * @obj: GEM object
+ *
+ * Releases a reference on the GEM buffer object's handle. Possibly releases
+ * the GEM buffer object and associated dma-buf objects.
+ */
+void drm_gem_object_handle_put_unlocked(struct drm_gem_object *obj)
 {
 	struct drm_device *dev = obj->dev;
 	bool final = false;
@@ -268,6 +303,7 @@ drm_gem_object_handle_put_unlocked(struct drm_gem_object *obj)
 	if (final)
 		drm_gem_object_put(obj);
 }
+EXPORT_SYMBOL(drm_gem_object_handle_put_unlocked);
 
 /*
  * Called at device or object close to release the file's
@@ -389,8 +425,8 @@ drm_gem_handle_create_tail(struct drm_file *file_priv,
 	int ret;
 
 	WARN_ON(!mutex_is_locked(&dev->object_name_lock));
-	if (obj->handle_count++ == 0)
+
-		drm_gem_object_get(obj);
+	drm_gem_object_handle_get(obj);
 
 	/*
 	 * Get the user-visible handle using idr.  Preload and perform

drivers/gpu/drm/drm_gem_framebuffer_helper.c

@@ -99,7 +99,7 @@ void drm_gem_fb_destroy(struct drm_framebuffer *fb)
 	unsigned int i;
 
 	for (i = 0; i < fb->format->num_planes; i++)
-		drm_gem_object_put(fb->obj[i]);
+		drm_gem_object_handle_put_unlocked(fb->obj[i]);
 
 	drm_framebuffer_cleanup(fb);
 	kfree(fb);
@@ -182,8 +182,10 @@ int drm_gem_fb_init_with_funcs(struct drm_device *dev,
 		if (!objs[i]) {
 			drm_dbg_kms(dev, "Failed to lookup GEM object\n");
 			ret = -ENOENT;
-			goto err_gem_object_put;
+			goto err_gem_object_handle_put_unlocked;
 		}
+		drm_gem_object_handle_get_unlocked(objs[i]);
+		drm_gem_object_put(objs[i]);
 
 		min_size = (height - 1) * mode_cmd->pitches[i]
 			 + drm_format_info_min_pitch(info, i, width)
@@ -193,22 +195,22 @@ int drm_gem_fb_init_with_funcs(struct drm_device *dev,
 			drm_dbg_kms(dev,
 				    "GEM object size (%zu) smaller than minimum size (%u) for plane %d\n",
 				    objs[i]->size, min_size, i);
-			drm_gem_object_put(objs[i]);
+			drm_gem_object_handle_put_unlocked(objs[i]);
 			ret = -EINVAL;
-			goto err_gem_object_put;
+			goto err_gem_object_handle_put_unlocked;
 		}
 	}
 
 	ret = drm_gem_fb_init(dev, fb, mode_cmd, objs, i, funcs);
 	if (ret)
-		goto err_gem_object_put;
+		goto err_gem_object_handle_put_unlocked;
 
 	return 0;
 
-err_gem_object_put:
+err_gem_object_handle_put_unlocked:
 	while (i > 0) {
 		--i;
-		drm_gem_object_put(objs[i]);
+		drm_gem_object_handle_put_unlocked(objs[i]);
 	}
 	return ret;
 }

drivers/gpu/drm/drm_internal.h

@@ -161,6 +161,8 @@ void drm_sysfs_lease_event(struct drm_device *dev);
 
 /* drm_gem.c */
 int drm_gem_init(struct drm_device *dev);
+void drm_gem_object_handle_get_unlocked(struct drm_gem_object *obj);
+void drm_gem_object_handle_put_unlocked(struct drm_gem_object *obj);
 int drm_gem_handle_create_tail(struct drm_file *file_priv,
 			       struct drm_gem_object *obj,
 			       u32 *handlep);

drivers/gpu/drm/drm_mipi_dsi.c

@@ -91,12 +91,13 @@ static const struct dev_pm_ops mipi_dsi_device_pm_ops = {
 	.restore = pm_generic_restore,
 };
 
-static const struct bus_type mipi_dsi_bus_type = {
+const struct bus_type mipi_dsi_bus_type = {
 	.name = "mipi-dsi",
 	.match = mipi_dsi_device_match,
 	.uevent = mipi_dsi_uevent,
 	.pm = &mipi_dsi_device_pm_ops,
 };
+EXPORT_SYMBOL_GPL(mipi_dsi_bus_type);
 
 /**
  * of_find_mipi_dsi_device_by_node() - find the MIPI DSI device matching a

drivers/gpu/drm/exynos/exynos7_drm_decon.c

@@ -636,6 +636,10 @@ static irqreturn_t decon_irq_handler(int irq, void *dev_id)
 	if (!ctx->drm_dev)
 		goto out;
 
+	/* check if crtc and vblank have been initialized properly */
+	if (!drm_dev_has_vblank(ctx->drm_dev))
+		goto out;
+
 	if (!ctx->i80_if) {
 		drm_crtc_handle_vblank(&ctx->crtc->base);
 

drivers/gpu/drm/exynos/exynos_drm_fimd.c

@@ -187,6 +187,7 @@ struct fimd_context {
 	u32				i80ifcon;
 	bool				i80_if;
 	bool				suspended;
+	bool				dp_clk_enabled;
 	wait_queue_head_t		wait_vsync_queue;
 	atomic_t			wait_vsync_event;
 	atomic_t			win_updated;
@@ -1047,7 +1048,18 @@ static void fimd_dp_clock_enable(struct exynos_drm_clk *clk, bool enable)
 	struct fimd_context *ctx = container_of(clk, struct fimd_context,
 						dp_clk);
 	u32 val = enable ? DP_MIE_CLK_DP_ENABLE : DP_MIE_CLK_DISABLE;
+
+	if (enable == ctx->dp_clk_enabled)
+		return;
+
+	if (enable)
+		pm_runtime_resume_and_get(ctx->dev);
+
+	ctx->dp_clk_enabled = enable;
 	writel(val, ctx->regs + DP_MIE_CLKCON);
+
+	if (!enable)
+		pm_runtime_put(ctx->dev);
 }
 
 static const struct exynos_drm_crtc_ops fimd_crtc_ops = {