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

Cross-merge networking fixes after downstream PR (net-6.13-rc4).

No conflicts.

Adjacent changes:

drivers/net/ethernet/renesas/rswitch.h
  32fd46f5b6 ("net: renesas: rswitch: remove speed from gwca structure")
  922b4b955a ("net: renesas: rswitch: rework ts tags management")

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

Documentation/devicetree/bindings/phy/fsl,imx8mq-usb-phy.yaml

@@ -113,11 +113,8 @@ allOf:
           maxItems: 1
 
   - if:
-      properties:
+      required:
-        compatible:
+        - orientation-switch
-          contains:
-            enum:
-              - fsl,imx95-usb-phy
     then:
       $ref: /schemas/usb/usb-switch.yaml#
 

Documentation/devicetree/bindings/regulator/qcom,qca6390-pmu.yaml

@@ -18,6 +18,7 @@ properties:
   compatible:
     enum:
       - qcom,qca6390-pmu
+      - qcom,wcn6750-pmu
       - qcom,wcn6855-pmu
       - qcom,wcn7850-pmu
 
@@ -27,6 +28,9 @@ properties:
   vddaon-supply:
     description: VDD_AON supply regulator handle
 
+  vddasd-supply:
+    description: VDD_ASD supply regulator handle
+
   vdddig-supply:
     description: VDD_DIG supply regulator handle
 
@@ -42,6 +46,9 @@ properties:
   vddio1p2-supply:
     description: VDD_IO_1P2 supply regulator handle
 
+  vddrfa0p8-supply:
+    description: VDD_RFA_0P8 supply regulator handle
+
   vddrfa0p95-supply:
     description: VDD_RFA_0P95 supply regulator handle
 
@@ -51,12 +58,18 @@ properties:
   vddrfa1p3-supply:
     description: VDD_RFA_1P3 supply regulator handle
 
+  vddrfa1p7-supply:
+    description: VDD_RFA_1P7 supply regulator handle
+
   vddrfa1p8-supply:
     description: VDD_RFA_1P8 supply regulator handle
 
   vddrfa1p9-supply:
     description: VDD_RFA_1P9 supply regulator handle
 
+  vddrfa2p2-supply:
+    description: VDD_RFA_2P2 supply regulator handle
+
   vddpcie1p3-supply:
     description: VDD_PCIE_1P3 supply regulator handle
 
@@ -119,6 +132,20 @@ allOf:
         - vddpcie1p3-supply
         - vddpcie1p9-supply
         - vddio-supply
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: qcom,wcn6750-pmu
+    then:
+      required:
+        - vddaon-supply
+        - vddasd-supply
+        - vddpmu-supply
+        - vddrfa0p8-supply
+        - vddrfa1p2-supply
+        - vddrfa1p7-supply
+        - vddrfa2p2-supply
   - if:
       properties:
         compatible:

Documentation/power/runtime_pm.rst

@@ -347,7 +347,9 @@ drivers/base/power/runtime.c and include/linux/pm_runtime.h:
 
   `int pm_runtime_resume_and_get(struct device *dev);`
     - run pm_runtime_resume(dev) and if successful, increment the device's
-      usage counter; return the result of pm_runtime_resume
+      usage counter; returns 0 on success (whether or not the device's
+      runtime PM status was already 'active') or the error code from
+      pm_runtime_resume() on failure.
 
   `int pm_request_idle(struct device *dev);`
     - submit a request to execute the subsystem-level idle callback for the

MAINTAINERS

@@ -3900,7 +3900,7 @@ W:	http://www.baycom.org/~tom/ham/ham.html
 F:	drivers/net/hamradio/baycom*
 
 BCACHE (BLOCK LAYER CACHE)
-M:	Coly Li <colyli@suse.de>
+M:	Coly Li <colyli@kernel.org>
 M:	Kent Overstreet <kent.overstreet@linux.dev>
 L:	linux-bcache@vger.kernel.org
 S:	Maintained
@@ -8460,7 +8460,7 @@ F:	include/video/s1d13xxxfb.h
 EROFS FILE SYSTEM
 M:	Gao Xiang <xiang@kernel.org>
 M:	Chao Yu <chao@kernel.org>
-R:	Yue Hu <huyue2@coolpad.com>
+R:	Yue Hu <zbestahu@gmail.com>
 R:	Jeffle Xu <jefflexu@linux.alibaba.com>
 R:	Sandeep Dhavale <dhavale@google.com>
 L:	linux-erofs@lists.ozlabs.org

Makefile

@@ -2,7 +2,7 @@
 VERSION = 6
 PATCHLEVEL = 13
 SUBLEVEL = 0
-EXTRAVERSION = -rc2
+EXTRAVERSION = -rc3
 NAME = Baby Opossum Posse
 
 # *DOCUMENTATION*

arch/arc/Kconfig

@@ -297,7 +297,6 @@ config ARC_PAGE_SIZE_16K
 config ARC_PAGE_SIZE_4K
 	bool "4KB"
 	select HAVE_PAGE_SIZE_4KB
-	depends on ARC_MMU_V3 || ARC_MMU_V4
 
 endchoice
 
@@ -474,7 +473,8 @@ config HIGHMEM
 
 config ARC_HAS_PAE40
 	bool "Support for the 40-bit Physical Address Extension"
-	depends on ISA_ARCV2
+	depends on ARC_MMU_V4
+	depends on !ARC_PAGE_SIZE_4K
 	select HIGHMEM
 	select PHYS_ADDR_T_64BIT
 	help

arch/arc/Makefile

@@ -6,7 +6,7 @@
 KBUILD_DEFCONFIG := haps_hs_smp_defconfig
 
 ifeq ($(CROSS_COMPILE),)
-CROSS_COMPILE := $(call cc-cross-prefix, arc-linux- arceb-linux-)
+CROSS_COMPILE := $(call cc-cross-prefix, arc-linux- arceb-linux- arc-linux-gnu-)
 endif
 
 cflags-y	+= -fno-common -pipe -fno-builtin -mmedium-calls -D__linux__

arch/arc/boot/dts/axc001.dtsi

@@ -54,7 +54,7 @@ ictl_intc: gpio-controller@0 {
 				compatible = "snps,dw-apb-gpio-port";
 				gpio-controller;
 				#gpio-cells = <2>;
-				snps,nr-gpios = <30>;
+				ngpios = <30>;
 				reg = <0>;
 				interrupt-controller;
 				#interrupt-cells = <2>;

arch/arc/boot/dts/axc003.dtsi

@@ -62,7 +62,7 @@ ictl_intc: gpio-controller@0 {
 				compatible = "snps,dw-apb-gpio-port";
 				gpio-controller;
 				#gpio-cells = <2>;
-				snps,nr-gpios = <30>;
+				ngpios = <30>;
 				reg = <0>;
 				interrupt-controller;
 				#interrupt-cells = <2>;

arch/arc/boot/dts/axc003_idu.dtsi

@@ -69,7 +69,7 @@ ictl_intc: gpio-controller@0 {
 				compatible = "snps,dw-apb-gpio-port";
 				gpio-controller;
 				#gpio-cells = <2>;
-				snps,nr-gpios = <30>;
+				ngpios = <30>;
 				reg = <0>;
 				interrupt-controller;
 				#interrupt-cells = <2>;

arch/arc/boot/dts/axs10x_mb.dtsi

@@ -250,7 +250,7 @@ gpio0_banka: gpio-controller@0 {
 				compatible = "snps,dw-apb-gpio-port";
 				gpio-controller;
 				#gpio-cells = <2>;
-				snps,nr-gpios = <32>;
+				ngpios = <32>;
 				reg = <0>;
 			};
 
@@ -258,7 +258,7 @@ gpio0_bankb: gpio-controller@1 {
 				compatible = "snps,dw-apb-gpio-port";
 				gpio-controller;
 				#gpio-cells = <2>;
-				snps,nr-gpios = <8>;
+				ngpios = <8>;
 				reg = <1>;
 			};
 
@@ -266,7 +266,7 @@ gpio0_bankc: gpio-controller@2 {
 				compatible = "snps,dw-apb-gpio-port";
 				gpio-controller;
 				#gpio-cells = <2>;
-				snps,nr-gpios = <8>;
+				ngpios = <8>;
 				reg = <2>;
 			};
 		};
@@ -281,7 +281,7 @@ gpio1_banka: gpio-controller@0 {
 				compatible = "snps,dw-apb-gpio-port";
 				gpio-controller;
 				#gpio-cells = <2>;
-				snps,nr-gpios = <30>;
+				ngpios = <30>;
 				reg = <0>;
 			};
 
@@ -289,7 +289,7 @@ gpio1_bankb: gpio-controller@1 {
 				compatible = "snps,dw-apb-gpio-port";
 				gpio-controller;
 				#gpio-cells = <2>;
-				snps,nr-gpios = <10>;
+				ngpios = <10>;
 				reg = <1>;
 			};
 
@@ -297,7 +297,7 @@ gpio1_bankc: gpio-controller@2 {
 				compatible = "snps,dw-apb-gpio-port";
 				gpio-controller;
 				#gpio-cells = <2>;
-				snps,nr-gpios = <8>;
+				ngpios = <8>;
 				reg = <2>;
 			};
 		};

arch/arc/boot/dts/hsdk.dts

@@ -308,7 +308,7 @@ gpio_port_a: gpio-controller@0 {
 				compatible = "snps,dw-apb-gpio-port";
 				gpio-controller;
 				#gpio-cells = <2>;
-				snps,nr-gpios = <24>;
+				ngpios = <24>;
 				reg = <0>;
 			};
 		};

arch/arc/include/asm/arcregs.h

@@ -146,7 +146,7 @@
 
 #ifndef __ASSEMBLY__
 
-#include <soc/arc/aux.h>
+#include <soc/arc/arc_aux.h>
 
 /* Helpers */
 #define TO_KB(bytes)		((bytes) >> 10)

arch/arc/include/asm/cmpxchg.h

@@ -48,7 +48,7 @@
 									\
 	switch(sizeof((_p_))) {						\
 	case 1:								\
-		_prev_ = (__typeof__(*(ptr)))cmpxchg_emu_u8((volatile u8 *)_p_, (uintptr_t)_o_, (uintptr_t)_n_);	\
+		_prev_ = (__typeof__(*(ptr)))cmpxchg_emu_u8((volatile u8 *__force)_p_, (uintptr_t)_o_, (uintptr_t)_n_);	\
 		break;							\
 	case 4:								\
 		_prev_ = __cmpxchg(_p_, _o_, _n_);			\

arch/arc/include/asm/mmu-arcv2.h

@@ -9,7 +9,7 @@
 #ifndef _ASM_ARC_MMU_ARCV2_H
 #define _ASM_ARC_MMU_ARCV2_H
 
-#include <soc/arc/aux.h>
+#include <soc/arc/arc_aux.h>
 
 /*
  * TLB Management regs

arch/arc/net/bpf_jit_arcv2.c

@@ -2916,7 +2916,7 @@ bool check_jmp_32(u32 curr_off, u32 targ_off, u8 cond)
 	addendum = (cond == ARC_CC_AL) ? 0 : INSN_len_normal;
 	disp = get_displacement(curr_off + addendum, targ_off);
 
-	if (ARC_CC_AL)
+	if (cond == ARC_CC_AL)
 		return is_valid_far_disp(disp);
 	else
 		return is_valid_near_disp(disp);

arch/arm64/boot/dts/arm/fvp-base-revc.dts

@@ -233,7 +233,7 @@ pci: pci@40000000 {
 		#interrupt-cells = <0x1>;
 		compatible = "pci-host-ecam-generic";
 		device_type = "pci";
-		bus-range = <0x0 0x1>;
+		bus-range = <0x0 0xff>;
 		reg = <0x0 0x40000000 0x0 0x10000000>;
 		ranges = <0x2000000 0x0 0x50000000 0x0 0x50000000 0x0 0x10000000>;
 		interrupt-map = <0 0 0 1 &gic 0 0 GIC_SPI 168 IRQ_TYPE_LEVEL_HIGH>,

arch/arm64/include/asm/el2_setup.h

@@ -87,7 +87,7 @@
 		      1 << PMSCR_EL2_PA_SHIFT)
 	msr_s	SYS_PMSCR_EL2, x0		// addresses and physical counter
 .Lskip_spe_el2_\@:
-	mov	x0, #(MDCR_EL2_E2PB_MASK << MDCR_EL2_E2PB_SHIFT)
+	mov	x0, #MDCR_EL2_E2PB_MASK
 	orr	x2, x2, x0			// If we don't have VHE, then
 						// use EL1&0 translation.
 
@@ -100,7 +100,7 @@
 	and	x0, x0, TRBIDR_EL1_P
 	cbnz	x0, .Lskip_trace_\@		// If TRBE is available at EL2
 
-	mov	x0, #(MDCR_EL2_E2TB_MASK << MDCR_EL2_E2TB_SHIFT)
+	mov	x0, #MDCR_EL2_E2TB_MASK
 	orr	x2, x2, x0			// allow the EL1&0 translation
 						// to own it.
 

arch/arm64/kernel/hyp-stub.S

@@ -114,8 +114,8 @@ SYM_CODE_START_LOCAL(__finalise_el2)
 
 	// Use EL2 translations for SPE & TRBE and disable access from EL1
 	mrs	x0, mdcr_el2
-	bic	x0, x0, #(MDCR_EL2_E2PB_MASK << MDCR_EL2_E2PB_SHIFT)
+	bic	x0, x0, #MDCR_EL2_E2PB_MASK
-	bic	x0, x0, #(MDCR_EL2_E2TB_MASK << MDCR_EL2_E2TB_SHIFT)
+	bic	x0, x0, #MDCR_EL2_E2TB_MASK
 	msr	mdcr_el2, x0
 
 	// Transfer the MM state from EL1 to EL2

arch/arm64/kernel/signal.c

@@ -1462,10 +1462,33 @@ static int setup_return(struct pt_regs *regs, struct ksignal *ksig,
 			 struct rt_sigframe_user_layout *user, int usig)
 {
 	__sigrestore_t sigtramp;
+	int err;
+
+	if (ksig->ka.sa.sa_flags & SA_RESTORER)
+		sigtramp = ksig->ka.sa.sa_restorer;
+	else
+		sigtramp = VDSO_SYMBOL(current->mm->context.vdso, sigtramp);
+
+	err = gcs_signal_entry(sigtramp, ksig);
+	if (err)
+		return err;
+
+	/*
+	 * We must not fail from this point onwards. We are going to update
+	 * registers, including SP, in order to invoke the signal handler. If
+	 * we failed and attempted to deliver a nested SIGSEGV to a handler
+	 * after that point, the subsequent sigreturn would end up restoring
+	 * the (partial) state for the original signal handler.
+	 */
 
 	regs->regs[0] = usig;
+	if (ksig->ka.sa.sa_flags & SA_SIGINFO) {
+		regs->regs[1] = (unsigned long)&user->sigframe->info;
+		regs->regs[2] = (unsigned long)&user->sigframe->uc;
+	}
 	regs->sp = (unsigned long)user->sigframe;
 	regs->regs[29] = (unsigned long)&user->next_frame->fp;
+	regs->regs[30] = (unsigned long)sigtramp;
 	regs->pc = (unsigned long)ksig->ka.sa.sa_handler;
 
 	/*
@@ -1506,14 +1529,7 @@ static int setup_return(struct pt_regs *regs, struct ksignal *ksig,
 		sme_smstop();
 	}
 
-	if (ksig->ka.sa.sa_flags & SA_RESTORER)
+	return 0;
-		sigtramp = ksig->ka.sa.sa_restorer;
-	else
-		sigtramp = VDSO_SYMBOL(current->mm->context.vdso, sigtramp);
-
-	regs->regs[30] = (unsigned long)sigtramp;
-
-	return gcs_signal_entry(sigtramp, ksig);
 }
 
 static int setup_rt_frame(int usig, struct ksignal *ksig, sigset_t *set,
@@ -1537,14 +1553,16 @@ static int setup_rt_frame(int usig, struct ksignal *ksig, sigset_t *set,
 
 	err |= __save_altstack(&frame->uc.uc_stack, regs->sp);
 	err |= setup_sigframe(&user, regs, set, &ua_state);
-	if (err == 0) {
+	if (ksig->ka.sa.sa_flags & SA_SIGINFO)
+		err |= copy_siginfo_to_user(&frame->info, &ksig->info);
+
+	if (err == 0)
 		err = setup_return(regs, ksig, &user, usig);
-		if (ksig->ka.sa.sa_flags & SA_SIGINFO) {
+
-			err |= copy_siginfo_to_user(&frame->info, &ksig->info);
+	/*
-			regs->regs[1] = (unsigned long)&frame->info;
+	 * We must not fail if setup_return() succeeded - see comment at the
-			regs->regs[2] = (unsigned long)&frame->uc;
+	 * beginning of setup_return().
-		}
+	 */
-	}
 
 	if (err == 0)
 		set_handler_user_access_state();

arch/arm64/kernel/stacktrace.c

@@ -26,7 +26,6 @@ enum kunwind_source {
 	KUNWIND_SOURCE_CALLER,
 	KUNWIND_SOURCE_TASK,
 	KUNWIND_SOURCE_REGS_PC,
-	KUNWIND_SOURCE_REGS_LR,
 };
 
 union unwind_flags {
@@ -138,8 +137,10 @@ kunwind_recover_return_address(struct kunwind_state *state)
 		orig_pc = ftrace_graph_ret_addr(state->task, &state->graph_idx,
 						state->common.pc,
 						(void *)state->common.fp);
-		if (WARN_ON_ONCE(state->common.pc == orig_pc))
+		if (state->common.pc == orig_pc) {
+			WARN_ON_ONCE(state->task == current);
 			return -EINVAL;
+		}
 		state->common.pc = orig_pc;
 		state->flags.fgraph = 1;
 	}
@@ -178,23 +179,8 @@ int kunwind_next_regs_pc(struct kunwind_state *state)
 	state->regs = regs;
 	state->common.pc = regs->pc;
 	state->common.fp = regs->regs[29];
-	state->source = KUNWIND_SOURCE_REGS_PC;
-	return 0;
-}
-
-static __always_inline int
-kunwind_next_regs_lr(struct kunwind_state *state)
-{
-	/*
-	 * The stack for the regs was consumed by kunwind_next_regs_pc(), so we
-	 * cannot consume that again here, but we know the regs are safe to
-	 * access.
-	 */
-	state->common.pc = state->regs->regs[30];
-	state->common.fp = state->regs->regs[29];
 	state->regs = NULL;
-	state->source = KUNWIND_SOURCE_REGS_LR;
+	state->source = KUNWIND_SOURCE_REGS_PC;
-
 	return 0;
 }
 
@@ -215,12 +201,12 @@ kunwind_next_frame_record_meta(struct kunwind_state *state)
 	case FRAME_META_TYPE_FINAL:
 		if (meta == &task_pt_regs(tsk)->stackframe)
 			return -ENOENT;
-		WARN_ON_ONCE(1);
+		WARN_ON_ONCE(tsk == current);
 		return -EINVAL;
 	case FRAME_META_TYPE_PT_REGS:
 		return kunwind_next_regs_pc(state);
 	default:
-		WARN_ON_ONCE(1);
+		WARN_ON_ONCE(tsk == current);
 		return -EINVAL;
 	}
 }
@@ -274,11 +260,8 @@ kunwind_next(struct kunwind_state *state)
 	case KUNWIND_SOURCE_FRAME:
 	case KUNWIND_SOURCE_CALLER:
 	case KUNWIND_SOURCE_TASK:
-	case KUNWIND_SOURCE_REGS_LR:
-		err = kunwind_next_frame_record(state);
-		break;
 	case KUNWIND_SOURCE_REGS_PC:
-		err = kunwind_next_regs_lr(state);
+		err = kunwind_next_frame_record(state);
 		break;
 	default:
 		err = -EINVAL;
@@ -436,7 +419,6 @@ static const char *state_source_string(const struct kunwind_state *state)
 	case KUNWIND_SOURCE_CALLER:	return "C";
 	case KUNWIND_SOURCE_TASK:	return "T";
 	case KUNWIND_SOURCE_REGS_PC:	return "P";
-	case KUNWIND_SOURCE_REGS_LR:	return "L";
 	default:			return "U";
 	}
 }

arch/arm64/kvm/at.c

@@ -739,8 +739,15 @@ static u64 compute_par_s12(struct kvm_vcpu *vcpu, u64 s1_par,
 			final_attr = s1_parattr;
 			break;
 		default:
-			/* MemAttr[2]=0, Device from S2 */
+			/*
-			final_attr = s2_memattr & GENMASK(1,0) << 2;
+			 * MemAttr[2]=0, Device from S2.
+			 *
+			 * FWB does not influence the way that stage 1
+			 * memory types and attributes are combined
+			 * with stage 2 Device type and attributes.
+			 */
+			final_attr = min(s2_memattr_to_attr(s2_memattr),
+					 s1_parattr);
 		}
 	} else {
 		/* Combination of R_HMNDG, R_TNHFM and R_GQFSF */

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

@@ -126,7 +126,7 @@ static void pvm_init_traps_aa64dfr0(struct kvm_vcpu *vcpu)
 	/* Trap SPE */
 	if (!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_PMSVer), feature_ids)) {
 		mdcr_set |= MDCR_EL2_TPMS;
-		mdcr_clear |= MDCR_EL2_E2PB_MASK << MDCR_EL2_E2PB_SHIFT;
+		mdcr_clear |= MDCR_EL2_E2PB_MASK;
 	}
 
 	/* Trap Trace Filter */
@@ -143,7 +143,7 @@ static void pvm_init_traps_aa64dfr0(struct kvm_vcpu *vcpu)
 
 	/* Trap External Trace */
 	if (!FIELD_GET(ARM64_FEATURE_MASK(ID_AA64DFR0_EL1_ExtTrcBuff), feature_ids))
-		mdcr_clear |= MDCR_EL2_E2TB_MASK << MDCR_EL2_E2TB_SHIFT;
+		mdcr_clear |= MDCR_EL2_E2TB_MASK;
 
 	vcpu->arch.mdcr_el2 |= mdcr_set;
 	vcpu->arch.mdcr_el2 &= ~mdcr_clear;

arch/arm64/kvm/sys_regs.c

@@ -2618,7 +2618,8 @@ static const struct sys_reg_desc sys_reg_descs[] = {
 	ID_WRITABLE(ID_AA64MMFR0_EL1, ~(ID_AA64MMFR0_EL1_RES0 |
 					ID_AA64MMFR0_EL1_TGRAN4_2 |
 					ID_AA64MMFR0_EL1_TGRAN64_2 |
-					ID_AA64MMFR0_EL1_TGRAN16_2)),
+					ID_AA64MMFR0_EL1_TGRAN16_2 |
+					ID_AA64MMFR0_EL1_ASIDBITS)),
 	ID_WRITABLE(ID_AA64MMFR1_EL1, ~(ID_AA64MMFR1_EL1_RES0 |
 					ID_AA64MMFR1_EL1_HCX |
 					ID_AA64MMFR1_EL1_TWED |

arch/arm64/kvm/vgic/vgic-its.c

@@ -608,12 +608,22 @@ static void vgic_its_cache_translation(struct kvm *kvm, struct vgic_its *its,
 	lockdep_assert_held(&its->its_lock);
 	vgic_get_irq_kref(irq);
 
+	old = xa_store(&its->translation_cache, cache_key, irq, GFP_KERNEL_ACCOUNT);
+
+	/*
+	 * Put the reference taken on @irq if the store fails. Intentionally do
+	 * not return the error as the translation cache is best effort.
+	 */
+	if (xa_is_err(old)) {
+		vgic_put_irq(kvm, irq);
+		return;
+	}
+
 	/*
 	 * We could have raced with another CPU caching the same
 	 * translation behind our back, ensure we don't leak a
 	 * reference if that is the case.
 	 */
-	old = xa_store(&its->translation_cache, cache_key, irq, GFP_KERNEL_ACCOUNT);
 	if (old)
 		vgic_put_irq(kvm, old);
 }

arch/hexagon/Makefile

@@ -32,3 +32,9 @@ KBUILD_LDFLAGS += $(ldflags-y)
 TIR_NAME := r19
 KBUILD_CFLAGS += -ffixed-$(TIR_NAME) -DTHREADINFO_REG=$(TIR_NAME) -D__linux__
 KBUILD_AFLAGS += -DTHREADINFO_REG=$(TIR_NAME)
+
+# Disable HexagonConstExtenders pass for LLVM versions prior to 19.1.0
+# https://github.com/llvm/llvm-project/issues/99714
+ifneq ($(call clang-min-version, 190100),y)
+KBUILD_CFLAGS += -mllvm -hexagon-cext=false
+endif

arch/openrisc/kernel/entry.S

@@ -239,6 +239,8 @@ handler:							;\
 
 /* =====================================================[ exceptions] === */
 
+	__REF
+
 /* ---[ 0x100: RESET exception ]----------------------------------------- */
 
 EXCEPTION_ENTRY(_tng_kernel_start)

arch/openrisc/kernel/head.S

@@ -26,15 +26,15 @@
 #include <asm/asm-offsets.h>
 #include <linux/of_fdt.h>
 
-#define tophys(rd,rs)				\
+#define tophys(rd,rs)						\
-	l.movhi	rd,hi(-KERNELBASE)		;\
+	l.movhi	rd,hi(-KERNELBASE)				;\
 	l.add	rd,rd,rs
 
-#define CLEAR_GPR(gpr)				\
+#define CLEAR_GPR(gpr)						\
 	l.movhi	gpr,0x0
 
-#define LOAD_SYMBOL_2_GPR(gpr,symbol)		\
+#define LOAD_SYMBOL_2_GPR(gpr,symbol)				\
-	l.movhi gpr,hi(symbol)			;\
+	l.movhi gpr,hi(symbol)					;\
 	l.ori   gpr,gpr,lo(symbol)
 
 
@@ -326,21 +326,21 @@
 	l.addi  r1,r1,-(INT_FRAME_SIZE)				;\
 	/* r1 is KSP, r30 is __pa(KSP) */			;\
 	tophys  (r30,r1)					;\
-	l.sw    PT_GPR12(r30),r12					;\
+	l.sw    PT_GPR12(r30),r12				;\
 	l.mfspr r12,r0,SPR_EPCR_BASE				;\
 	l.sw    PT_PC(r30),r12					;\
 	l.mfspr r12,r0,SPR_ESR_BASE				;\
 	l.sw    PT_SR(r30),r12					;\
 	/* save r31 */						;\
 	EXCEPTION_T_LOAD_GPR30(r12)				;\
-	l.sw	PT_GPR30(r30),r12					;\
+	l.sw	PT_GPR30(r30),r12				;\
 	/* save r10 as was prior to exception */		;\
 	EXCEPTION_T_LOAD_GPR10(r12)				;\
-	l.sw	PT_GPR10(r30),r12					;\
+	l.sw	PT_GPR10(r30),r12				;\
-	/* save PT_SP as was prior to exception */			;\
+	/* save PT_SP as was prior to exception */		;\
 	EXCEPTION_T_LOAD_SP(r12)				;\
 	l.sw	PT_SP(r30),r12					;\
-	l.sw    PT_GPR13(r30),r13					;\
+	l.sw    PT_GPR13(r30),r13				;\
 	/* --> */						;\
 	/* save exception r4, set r4 = EA */			;\
 	l.sw	PT_GPR4(r30),r4					;\
@@ -357,6 +357,8 @@
 
 /* =====================================================[ exceptions] === */
 
+	__HEAD
+
 /* ---[ 0x100: RESET exception ]----------------------------------------- */
     .org 0x100
 	/* Jump to .init code at _start which lives in the .head section
@@ -394,7 +396,7 @@ _dispatch_do_ipage_fault:
     .org 0x500
 	EXCEPTION_HANDLE(_timer_handler)
 
-/* ---[ 0x600: Alignment exception ]-------------------------------------- */
+/* ---[ 0x600: Alignment exception ]------------------------------------- */
     .org 0x600
 	EXCEPTION_HANDLE(_alignment_handler)
 
@@ -424,7 +426,7 @@ _dispatch_do_ipage_fault:
     .org 0xc00
 	EXCEPTION_HANDLE(_sys_call_handler)
 
-/* ---[ 0xd00: Floating point exception ]--------------------------------- */
+/* ---[ 0xd00: Floating point exception ]-------------------------------- */
     .org 0xd00
 	EXCEPTION_HANDLE(_fpe_trap_handler)
 
@@ -506,10 +508,10 @@ _dispatch_do_ipage_fault:
 
 /*    .text*/
 
-/* This early stuff belongs in HEAD, but some of the functions below definitely
+/* This early stuff belongs in the .init.text section, but some of the functions below definitely
  * don't... */
 
-	__HEAD
+	__INIT
 	.global _start
 _start:
 	/* Init r0 to zero as per spec */
@@ -816,7 +818,7 @@ secondary_start:
 
 #endif
 
-/* ========================================[ cache ]=== */
+/* ==========================================================[ cache ]=== */
 
 	/* alignment here so we don't change memory offsets with
 	 * memory controller defined

arch/openrisc/kernel/vmlinux.lds.S

@@ -50,6 +50,7 @@ SECTIONS
         .text                   : AT(ADDR(.text) - LOAD_OFFSET)
 	{
           _stext = .;
+	  HEAD_TEXT
 	  TEXT_TEXT
 	  SCHED_TEXT
 	  LOCK_TEXT
@@ -83,8 +84,6 @@ SECTIONS
 	. = ALIGN(PAGE_SIZE);
 	__init_begin = .;
 
-	HEAD_TEXT_SECTION
-
 	/* Page aligned */
 	INIT_TEXT_SECTION(PAGE_SIZE)
 

arch/riscv/include/asm/kfence.h

@@ -22,7 +22,9 @@ static inline bool kfence_protect_page(unsigned long addr, bool protect)
 	else
 		set_pte(pte, __pte(pte_val(ptep_get(pte)) | _PAGE_PRESENT));
 
-	flush_tlb_kernel_range(addr, addr + PAGE_SIZE);
+	preempt_disable();
+	local_flush_tlb_kernel_range(addr, addr + PAGE_SIZE);
+	preempt_enable();
 
 	return true;
 }

arch/riscv/kernel/jump_label.c

@@ -36,9 +36,15 @@ bool arch_jump_label_transform_queue(struct jump_entry *entry,
 		insn = RISCV_INSN_NOP;
 	}
 
-	mutex_lock(&text_mutex);
+	if (early_boot_irqs_disabled) {
-	patch_insn_write(addr, &insn, sizeof(insn));
+		riscv_patch_in_stop_machine = 1;
-	mutex_unlock(&text_mutex);
+		patch_insn_write(addr, &insn, sizeof(insn));
+		riscv_patch_in_stop_machine = 0;
+	} else {
+		mutex_lock(&text_mutex);
+		patch_insn_write(addr, &insn, sizeof(insn));
+		mutex_unlock(&text_mutex);
+	}
 
 	return true;
 }

arch/riscv/kernel/setup.c

@@ -227,7 +227,7 @@ static void __init init_resources(void)
 static void __init parse_dtb(void)
 {
 	/* Early scan of device tree from init memory */
-	if (early_init_dt_scan(dtb_early_va, __pa(dtb_early_va))) {
+	if (early_init_dt_scan(dtb_early_va, dtb_early_pa)) {
 		const char *name = of_flat_dt_get_machine_name();
 
 		if (name) {

arch/riscv/kvm/aia.c

@@ -590,7 +590,7 @@ void kvm_riscv_aia_enable(void)
 	csr_set(CSR_HIE, BIT(IRQ_S_GEXT));
 	/* Enable IRQ filtering for overflow interrupt only if sscofpmf is present */
 	if (__riscv_isa_extension_available(NULL, RISCV_ISA_EXT_SSCOFPMF))
-		csr_write(CSR_HVIEN, BIT(IRQ_PMU_OVF));
+		csr_set(CSR_HVIEN, BIT(IRQ_PMU_OVF));
 }
 
 void kvm_riscv_aia_disable(void)

arch/riscv/mm/init.c

@@ -1566,7 +1566,7 @@ static void __meminit free_pte_table(pte_t *pte_start, pmd_t *pmd)
 	pmd_clear(pmd);
 }
 
-static void __meminit free_pmd_table(pmd_t *pmd_start, pud_t *pud)
+static void __meminit free_pmd_table(pmd_t *pmd_start, pud_t *pud, bool is_vmemmap)
 {
 	struct page *page = pud_page(*pud);
 	struct ptdesc *ptdesc = page_ptdesc(page);
@@ -1579,7 +1579,8 @@ static void __meminit free_pmd_table(pmd_t *pmd_start, pud_t *pud)
 			return;
 	}
 
-	pagetable_pmd_dtor(ptdesc);
+	if (!is_vmemmap)
+		pagetable_pmd_dtor(ptdesc);
 	if (PageReserved(page))
 		free_reserved_page(page);
 	else
@@ -1703,7 +1704,7 @@ static void __meminit remove_pud_mapping(pud_t *pud_base, unsigned long addr, un
 		remove_pmd_mapping(pmd_base, addr, next, is_vmemmap, altmap);
 
 		if (pgtable_l4_enabled)
-			free_pmd_table(pmd_base, pudp);
+			free_pmd_table(pmd_base, pudp, is_vmemmap);
 	}
 }
 

arch/s390/boot/startup.c

@@ -234,6 +234,8 @@ static unsigned long get_vmem_size(unsigned long identity_size,
 	vsize = round_up(SZ_2G + max_mappable, rte_size) +
 		round_up(vmemmap_size, rte_size) +
 		FIXMAP_SIZE + MODULES_LEN + KASLR_LEN;
+	if (IS_ENABLED(CONFIG_KMSAN))
+		vsize += MODULES_LEN * 2;
 	return size_add(vsize, vmalloc_size);
 }
 

arch/s390/boot/vmem.c

@@ -306,7 +306,7 @@ static void pgtable_pte_populate(pmd_t *pmd, unsigned long addr, unsigned long e
 			pages++;
 		}
 	}
-	if (mode == POPULATE_DIRECT)
+	if (mode == POPULATE_IDENTITY)
 		update_page_count(PG_DIRECT_MAP_4K, pages);
 }
 
@@ -339,7 +339,7 @@ static void pgtable_pmd_populate(pud_t *pud, unsigned long addr, unsigned long e
 		}
 		pgtable_pte_populate(pmd, addr, next, mode);
 	}
-	if (mode == POPULATE_DIRECT)
+	if (mode == POPULATE_IDENTITY)
 		update_page_count(PG_DIRECT_MAP_1M, pages);
 }
 
@@ -372,7 +372,7 @@ static void pgtable_pud_populate(p4d_t *p4d, unsigned long addr, unsigned long e
 		}
 		pgtable_pmd_populate(pud, addr, next, mode);
 	}
-	if (mode == POPULATE_DIRECT)
+	if (mode == POPULATE_IDENTITY)
 		update_page_count(PG_DIRECT_MAP_2G, pages);
 }
 

arch/s390/kernel/ipl.c

@@ -270,7 +270,7 @@ static ssize_t sys_##_prefix##_##_name##_store(struct kobject *kobj,	\
 	if (len >= sizeof(_value))					\
 		return -E2BIG;						\
 	len = strscpy(_value, buf, sizeof(_value));			\
-	if (len < 0)							\
+	if ((ssize_t)len < 0)						\
 		return len;						\
 	strim(_value);							\
 	return len;							\

arch/x86/include/asm/processor.h

@@ -230,6 +230,8 @@ static inline unsigned long long l1tf_pfn_limit(void)
 	return BIT_ULL(boot_cpu_data.x86_cache_bits - 1 - PAGE_SHIFT);
 }
 
+void init_cpu_devs(void);
+void get_cpu_vendor(struct cpuinfo_x86 *c);
 extern void early_cpu_init(void);
 extern void identify_secondary_cpu(struct cpuinfo_x86 *);
 extern void print_cpu_info(struct cpuinfo_x86 *);

arch/x86/include/asm/static_call.h

@@ -65,4 +65,19 @@
 
 extern bool __static_call_fixup(void *tramp, u8 op, void *dest);
 
+extern void __static_call_update_early(void *tramp, void *func);
+
+#define static_call_update_early(name, _func)				\
+({									\
+	typeof(&STATIC_CALL_TRAMP(name)) __F = (_func);			\
+	if (static_call_initialized) {					\
+		__static_call_update(&STATIC_CALL_KEY(name),		\
+				     STATIC_CALL_TRAMP_ADDR(name), __F);\
+	} else {							\
+		WRITE_ONCE(STATIC_CALL_KEY(name).func, _func);		\
+		__static_call_update_early(STATIC_CALL_TRAMP_ADDR(name),\
+					   __F);			\
+	}								\
+})
+
 #endif /* _ASM_STATIC_CALL_H */

arch/x86/include/asm/sync_core.h

@@ -8,7 +8,7 @@
 #include <asm/special_insns.h>
 
 #ifdef CONFIG_X86_32
-static inline void iret_to_self(void)
+static __always_inline void iret_to_self(void)
 {
 	asm volatile (
 		"pushfl\n\t"
@@ -19,7 +19,7 @@ static inline void iret_to_self(void)
 		: ASM_CALL_CONSTRAINT : : "memory");
 }
 #else
-static inline void iret_to_self(void)
+static __always_inline void iret_to_self(void)
 {
 	unsigned int tmp;
 
@@ -55,7 +55,7 @@ static inline void iret_to_self(void)
  * Like all of Linux's memory ordering operations, this is a
  * compiler barrier as well.
  */
-static inline void sync_core(void)
+static __always_inline void sync_core(void)
 {
 	/*
 	 * The SERIALIZE instruction is the most straightforward way to

arch/x86/include/asm/xen/hypercall.h

@@ -39,9 +39,11 @@
 #include <linux/string.h>
 #include <linux/types.h>
 #include <linux/pgtable.h>
+#include <linux/instrumentation.h>
 
 #include <trace/events/xen.h>
 
+#include <asm/alternative.h>
 #include <asm/page.h>
 #include <asm/smap.h>
 #include <asm/nospec-branch.h>
@@ -86,11 +88,20 @@ struct xen_dm_op_buf;
  * there aren't more than 5 arguments...)
  */
 
-extern struct { char _entry[32]; } hypercall_page[];
+void xen_hypercall_func(void);
+DECLARE_STATIC_CALL(xen_hypercall, xen_hypercall_func);
 
-#define __HYPERCALL		"call hypercall_page+%c[offset]"
+#ifdef MODULE
-#define __HYPERCALL_ENTRY(x)						\
+#define __ADDRESSABLE_xen_hypercall
-	[offset] "i" (__HYPERVISOR_##x * sizeof(hypercall_page[0]))
+#else
+#define __ADDRESSABLE_xen_hypercall __ADDRESSABLE_ASM_STR(__SCK__xen_hypercall)
+#endif
+
+#define __HYPERCALL					\
+	__ADDRESSABLE_xen_hypercall			\
+	"call __SCT__xen_hypercall"
+
+#define __HYPERCALL_ENTRY(x)	"a" (x)
 
 #ifdef CONFIG_X86_32
 #define __HYPERCALL_RETREG	"eax"
@@ -148,7 +159,7 @@ extern struct { char _entry[32]; } hypercall_page[];
 	__HYPERCALL_0ARG();						\
 	asm volatile (__HYPERCALL					\
 		      : __HYPERCALL_0PARAM				\
-		      : __HYPERCALL_ENTRY(name)				\
+		      : __HYPERCALL_ENTRY(__HYPERVISOR_ ## name)	\
 		      : __HYPERCALL_CLOBBER0);				\
 	(type)__res;							\
 })
@@ -159,7 +170,7 @@ extern struct { char _entry[32]; } hypercall_page[];
 	__HYPERCALL_1ARG(a1);						\
 	asm volatile (__HYPERCALL					\
 		      : __HYPERCALL_1PARAM				\
-		      : __HYPERCALL_ENTRY(name)				\
+		      : __HYPERCALL_ENTRY(__HYPERVISOR_ ## name)	\
 		      : __HYPERCALL_CLOBBER1);				\
 	(type)__res;							\
 })
@@ -170,7 +181,7 @@ extern struct { char _entry[32]; } hypercall_page[];
 	__HYPERCALL_2ARG(a1, a2);					\
 	asm volatile (__HYPERCALL					\
 		      : __HYPERCALL_2PARAM				\
-		      : __HYPERCALL_ENTRY(name)				\
+		      : __HYPERCALL_ENTRY(__HYPERVISOR_ ## name)	\
 		      : __HYPERCALL_CLOBBER2);				\
 	(type)__res;							\
 })
@@ -181,7 +192,7 @@ extern struct { char _entry[32]; } hypercall_page[];
 	__HYPERCALL_3ARG(a1, a2, a3);					\
 	asm volatile (__HYPERCALL					\
 		      : __HYPERCALL_3PARAM				\
-		      : __HYPERCALL_ENTRY(name)				\
+		      : __HYPERCALL_ENTRY(__HYPERVISOR_ ## name)	\
 		      : __HYPERCALL_CLOBBER3);				\
 	(type)__res;							\
 })
@@ -192,7 +203,7 @@ extern struct { char _entry[32]; } hypercall_page[];
 	__HYPERCALL_4ARG(a1, a2, a3, a4);				\
 	asm volatile (__HYPERCALL					\
 		      : __HYPERCALL_4PARAM				\
-		      : __HYPERCALL_ENTRY(name)				\
+		      : __HYPERCALL_ENTRY(__HYPERVISOR_ ## name)	\
 		      : __HYPERCALL_CLOBBER4);				\
 	(type)__res;							\
 })
@@ -206,12 +217,9 @@ xen_single_call(unsigned int call,
 	__HYPERCALL_DECLS;
 	__HYPERCALL_5ARG(a1, a2, a3, a4, a5);
 
-	if (call >= PAGE_SIZE / sizeof(hypercall_page[0]))
+	asm volatile(__HYPERCALL
-		return -EINVAL;
-
-	asm volatile(CALL_NOSPEC
 		     : __HYPERCALL_5PARAM
-		     : [thunk_target] "a" (&hypercall_page[call])
+		     : __HYPERCALL_ENTRY(call)
 		     : __HYPERCALL_CLOBBER5);
 
 	return (long)__res;

arch/x86/kernel/callthunks.c

@@ -142,11 +142,6 @@ static bool skip_addr(void *dest)
 	if (dest >= (void *)relocate_kernel &&
 	    dest < (void*)relocate_kernel + KEXEC_CONTROL_CODE_MAX_SIZE)
 		return true;
-#endif
-#ifdef CONFIG_XEN
-	if (dest >= (void *)hypercall_page &&
-	    dest < (void*)hypercall_page + PAGE_SIZE)
-		return true;
 #endif
 	return false;
 }

arch/x86/kernel/cpu/common.c

@@ -867,7 +867,7 @@ static void cpu_detect_tlb(struct cpuinfo_x86 *c)
 		tlb_lld_4m[ENTRIES], tlb_lld_1g[ENTRIES]);
 }
 
-static void get_cpu_vendor(struct cpuinfo_x86 *c)
+void get_cpu_vendor(struct cpuinfo_x86 *c)
 {
 	char *v = c->x86_vendor_id;
 	int i;
@@ -1649,15 +1649,11 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c)
 	detect_nopl();
 }
 
-void __init early_cpu_init(void)
+void __init init_cpu_devs(void)
 {
 	const struct cpu_dev *const *cdev;
 	int count = 0;
 
-#ifdef CONFIG_PROCESSOR_SELECT
-	pr_info("KERNEL supported cpus:\n");
-#endif
-
 	for (cdev = __x86_cpu_dev_start; cdev < __x86_cpu_dev_end; cdev++) {
 		const struct cpu_dev *cpudev = *cdev;
 
@@ -1665,20 +1661,30 @@ void __init early_cpu_init(void)
 			break;
 		cpu_devs[count] = cpudev;
 		count++;
+	}
+}
+
+void __init early_cpu_init(void)
+{
+#ifdef CONFIG_PROCESSOR_SELECT
+	unsigned int i, j;
+
+	pr_info("KERNEL supported cpus:\n");
+#endif
+
+	init_cpu_devs();
 
 #ifdef CONFIG_PROCESSOR_SELECT
-		{
+	for (i = 0; i < X86_VENDOR_NUM && cpu_devs[i]; i++) {
-			unsigned int j;
+		for (j = 0; j < 2; j++) {
-
+			if (!cpu_devs[i]->c_ident[j])
-			for (j = 0; j < 2; j++) {
+				continue;
-				if (!cpudev->c_ident[j])
+			pr_info("  %s %s\n", cpu_devs[i]->c_vendor,
-					continue;
+				cpu_devs[i]->c_ident[j]);
-				pr_info("  %s %s\n", cpudev->c_vendor,
-					cpudev->c_ident[j]);
-			}
 		}
-#endif
 	}
+#endif
+
 	early_identify_cpu(&boot_cpu_data);
 }
 

arch/x86/kernel/cpu/mshyperv.c

@@ -223,6 +223,63 @@ static void hv_machine_crash_shutdown(struct pt_regs *regs)
 	hyperv_cleanup();
 }
 #endif /* CONFIG_CRASH_DUMP */
+
+static u64 hv_ref_counter_at_suspend;
+static void (*old_save_sched_clock_state)(void);
+static void (*old_restore_sched_clock_state)(void);
+
+/*
+ * Hyper-V clock counter resets during hibernation. Save and restore clock
+ * offset during suspend/resume, while also considering the time passed
+ * before suspend. This is to make sure that sched_clock using hv tsc page
+ * based clocksource, proceeds from where it left off during suspend and
+ * it shows correct time for the timestamps of kernel messages after resume.
+ */
+static void save_hv_clock_tsc_state(void)
+{
+	hv_ref_counter_at_suspend = hv_read_reference_counter();
+}
+
+static void restore_hv_clock_tsc_state(void)
+{
+	/*
+	 * Adjust the offsets used by hv tsc clocksource to
+	 * account for the time spent before hibernation.
+	 * adjusted value = reference counter (time) at suspend
+	 *                - reference counter (time) now.
+	 */
+	hv_adj_sched_clock_offset(hv_ref_counter_at_suspend - hv_read_reference_counter());
+}
+
+/*
+ * Functions to override save_sched_clock_state and restore_sched_clock_state
+ * functions of x86_platform. The Hyper-V clock counter is reset during
+ * suspend-resume and the offset used to measure time needs to be
+ * corrected, post resume.
+ */
+static void hv_save_sched_clock_state(void)
+{
+	old_save_sched_clock_state();
+	save_hv_clock_tsc_state();
+}
+
+static void hv_restore_sched_clock_state(void)
+{
+	restore_hv_clock_tsc_state();
+	old_restore_sched_clock_state();
+}
+
+static void __init x86_setup_ops_for_tsc_pg_clock(void)
+{
+	if (!(ms_hyperv.features & HV_MSR_REFERENCE_TSC_AVAILABLE))
+		return;
+
+	old_save_sched_clock_state = x86_platform.save_sched_clock_state;
+	x86_platform.save_sched_clock_state = hv_save_sched_clock_state;
+
+	old_restore_sched_clock_state = x86_platform.restore_sched_clock_state;
+	x86_platform.restore_sched_clock_state = hv_restore_sched_clock_state;
+}
 #endif /* CONFIG_HYPERV */
 
 static uint32_t  __init ms_hyperv_platform(void)
@@ -579,6 +636,7 @@ static void __init ms_hyperv_init_platform(void)
 
 	/* Register Hyper-V specific clocksource */
 	hv_init_clocksource();
+	x86_setup_ops_for_tsc_pg_clock();
 	hv_vtl_init_platform();
 #endif
 	/*

arch/x86/kernel/static_call.c

@@ -172,6 +172,15 @@ void arch_static_call_transform(void *site, void *tramp, void *func, bool tail)
 }
 EXPORT_SYMBOL_GPL(arch_static_call_transform);
 
+noinstr void __static_call_update_early(void *tramp, void *func)
+{
+	BUG_ON(system_state != SYSTEM_BOOTING);
+	BUG_ON(!early_boot_irqs_disabled);
+	BUG_ON(static_call_initialized);
+	__text_gen_insn(tramp, JMP32_INSN_OPCODE, tramp, func, JMP32_INSN_SIZE);
+	sync_core();
+}
+
 #ifdef CONFIG_MITIGATION_RETHUNK
 /*
  * This is called by apply_returns() to fix up static call trampolines,

arch/x86/kernel/vmlinux.lds.S

@@ -519,14 +519,10 @@ INIT_PER_CPU(irq_stack_backing_store);
  * linker will never mark as relocatable. (Using just ABSOLUTE() is not
  * sufficient for that).
  */
-#ifdef CONFIG_XEN
 #ifdef CONFIG_XEN_PV
 xen_elfnote_entry_value =
 	ABSOLUTE(xen_elfnote_entry) + ABSOLUTE(startup_xen);
 #endif
-xen_elfnote_hypercall_page_value =
-	ABSOLUTE(xen_elfnote_hypercall_page) + ABSOLUTE(hypercall_page);
-#endif
 #ifdef CONFIG_PVH
 xen_elfnote_phys32_entry_value =
 	ABSOLUTE(xen_elfnote_phys32_entry) + ABSOLUTE(pvh_start_xen - LOAD_OFFSET);

arch/x86/kvm/cpuid.c

@@ -36,6 +36,26 @@
 u32 kvm_cpu_caps[NR_KVM_CPU_CAPS] __read_mostly;
 EXPORT_SYMBOL_GPL(kvm_cpu_caps);
 
+struct cpuid_xstate_sizes {
+	u32 eax;
+	u32 ebx;
+	u32 ecx;
+};
+
+static struct cpuid_xstate_sizes xstate_sizes[XFEATURE_MAX] __ro_after_init;
+
+void __init kvm_init_xstate_sizes(void)
+{
+	u32 ign;
+	int i;
+
+	for (i = XFEATURE_YMM; i < ARRAY_SIZE(xstate_sizes); i++) {
+		struct cpuid_xstate_sizes *xs = &xstate_sizes[i];
+
+		cpuid_count(0xD, i, &xs->eax, &xs->ebx, &xs->ecx, &ign);
+	}
+}
+
 u32 xstate_required_size(u64 xstate_bv, bool compacted)
 {
 	int feature_bit = 0;
@@ -44,14 +64,15 @@ u32 xstate_required_size(u64 xstate_bv, bool compacted)
 	xstate_bv &= XFEATURE_MASK_EXTEND;
 	while (xstate_bv) {
 		if (xstate_bv & 0x1) {
-		        u32 eax, ebx, ecx, edx, offset;
+			struct cpuid_xstate_sizes *xs = &xstate_sizes[feature_bit];
-		        cpuid_count(0xD, feature_bit, &eax, &ebx, &ecx, &edx);
+			u32 offset;
+
 			/* ECX[1]: 64B alignment in compacted form */
 			if (compacted)
-				offset = (ecx & 0x2) ? ALIGN(ret, 64) : ret;
+				offset = (xs->ecx & 0x2) ? ALIGN(ret, 64) : ret;
 			else
-				offset = ebx;
+				offset = xs->ebx;
-			ret = max(ret, offset + eax);
+			ret = max(ret, offset + xs->eax);
 		}
 
 		xstate_bv >>= 1;

arch/x86/kvm/cpuid.h

@@ -31,6 +31,7 @@ int kvm_vcpu_ioctl_get_cpuid2(struct kvm_vcpu *vcpu,
 bool kvm_cpuid(struct kvm_vcpu *vcpu, u32 *eax, u32 *ebx,
 	       u32 *ecx, u32 *edx, bool exact_only);
 
+void __init kvm_init_xstate_sizes(void);
 u32 xstate_required_size(u64 xstate_bv, bool compacted);
 
 int cpuid_query_maxphyaddr(struct kvm_vcpu *vcpu);

arch/x86/kvm/x86.c

@@ -13997,6 +13997,8 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_rmp_fault);
 
 static int __init kvm_x86_init(void)
 {
+	kvm_init_xstate_sizes();
+
 	kvm_mmu_x86_module_init();
 	mitigate_smt_rsb &= boot_cpu_has_bug(X86_BUG_SMT_RSB) && cpu_smt_possible();
 	return 0;

arch/x86/xen/enlighten.c

@@ -2,6 +2,7 @@
 
 #include <linux/console.h>
 #include <linux/cpu.h>
+#include <linux/instrumentation.h>
 #include <linux/kexec.h>
 #include <linux/memblock.h>
 #include <linux/slab.h>
@@ -21,7 +22,8 @@
 
 #include "xen-ops.h"
 
-EXPORT_SYMBOL_GPL(hypercall_page);
+DEFINE_STATIC_CALL(xen_hypercall, xen_hypercall_hvm);
+EXPORT_STATIC_CALL_TRAMP(xen_hypercall);
 
 /*
  * Pointer to the xen_vcpu_info structure or
@@ -68,6 +70,67 @@ EXPORT_SYMBOL(xen_start_flags);
  */
 struct shared_info *HYPERVISOR_shared_info = &xen_dummy_shared_info;
 
+static __ref void xen_get_vendor(void)
+{
+	init_cpu_devs();
+	cpu_detect(&boot_cpu_data);
+	get_cpu_vendor(&boot_cpu_data);
+}
+
+void xen_hypercall_setfunc(void)
+{
+	if (static_call_query(xen_hypercall) != xen_hypercall_hvm)
+		return;
+
+	if ((boot_cpu_data.x86_vendor == X86_VENDOR_AMD ||
+	     boot_cpu_data.x86_vendor == X86_VENDOR_HYGON))
+		static_call_update(xen_hypercall, xen_hypercall_amd);
+	else
+		static_call_update(xen_hypercall, xen_hypercall_intel);
+}
+
+/*
+ * Evaluate processor vendor in order to select the correct hypercall
+ * function for HVM/PVH guests.
+ * Might be called very early in boot before vendor has been set by
+ * early_cpu_init().
+ */
+noinstr void *__xen_hypercall_setfunc(void)
+{
+	void (*func)(void);
+
+	/*
+	 * Xen is supported only on CPUs with CPUID, so testing for
+	 * X86_FEATURE_CPUID is a test for early_cpu_init() having been
+	 * run.
+	 *
+	 * Note that __xen_hypercall_setfunc() is noinstr only due to a nasty
+	 * dependency chain: it is being called via the xen_hypercall static
+	 * call when running as a PVH or HVM guest. Hypercalls need to be
+	 * noinstr due to PV guests using hypercalls in noinstr code. So we
+	 * can safely tag the function body as "instrumentation ok", since
+	 * the PV guest requirement is not of interest here (xen_get_vendor()
+	 * calls noinstr functions, and static_call_update_early() might do
+	 * so, too).
+	 */
+	instrumentation_begin();
+
+	if (!boot_cpu_has(X86_FEATURE_CPUID))
+		xen_get_vendor();
+
+	if ((boot_cpu_data.x86_vendor == X86_VENDOR_AMD ||
+	     boot_cpu_data.x86_vendor == X86_VENDOR_HYGON))
+		func = xen_hypercall_amd;
+	else
+		func = xen_hypercall_intel;
+
+	static_call_update_early(xen_hypercall, func);
+
+	instrumentation_end();
+
+	return func;
+}
+
 static int xen_cpu_up_online(unsigned int cpu)
 {
 	xen_init_lock_cpu(cpu);

arch/x86/xen/enlighten_hvm.c

@@ -106,15 +106,8 @@ static void __init init_hvm_pv_info(void)
 	/* PVH set up hypercall page in xen_prepare_pvh(). */
 	if (xen_pvh_domain())
 		pv_info.name = "Xen PVH";
-	else {
+	else
-		u64 pfn;
-		uint32_t msr;
-
 		pv_info.name = "Xen HVM";
-		msr = cpuid_ebx(base + 2);
-		pfn = __pa(hypercall_page);
-		wrmsr_safe(msr, (u32)pfn, (u32)(pfn >> 32));
-	}
 
 	xen_setup_features();
 
@@ -300,6 +293,10 @@ static uint32_t __init xen_platform_hvm(void)
 	if (xen_pv_domain())
 		return 0;
 
+	/* Set correct hypercall function. */
+	if (xen_domain)
+		xen_hypercall_setfunc();
+
 	if (xen_pvh_domain() && nopv) {
 		/* Guest booting via the Xen-PVH boot entry goes here */
 		pr_info("\"nopv\" parameter is ignored in PVH guest\n");

arch/x86/xen/enlighten_pv.c

@@ -1341,6 +1341,9 @@ asmlinkage __visible void __init xen_start_kernel(struct start_info *si)
 
 	xen_domain_type = XEN_PV_DOMAIN;
 	xen_start_flags = xen_start_info->flags;
+	/* Interrupts are guaranteed to be off initially. */
+	early_boot_irqs_disabled = true;
+	static_call_update_early(xen_hypercall, xen_hypercall_pv);
 
 	xen_setup_features();
 
@@ -1431,7 +1434,6 @@ asmlinkage __visible void __init xen_start_kernel(struct start_info *si)
 	WARN_ON(xen_cpuhp_setup(xen_cpu_up_prepare_pv, xen_cpu_dead_pv));
 
 	local_irq_disable();
-	early_boot_irqs_disabled = true;
 
 	xen_raw_console_write("mapping kernel into physical memory\n");
 	xen_setup_kernel_pagetable((pgd_t *)xen_start_info->pt_base,

arch/x86/xen/enlighten_pvh.c

@@ -129,17 +129,10 @@ static void __init pvh_arch_setup(void)
 
 void __init xen_pvh_init(struct boot_params *boot_params)
 {
-	u32 msr;
-	u64 pfn;
-
 	xen_pvh = 1;
 	xen_domain_type = XEN_HVM_DOMAIN;
 	xen_start_flags = pvh_start_info.flags;
 
-	msr = cpuid_ebx(xen_cpuid_base() + 2);
-	pfn = __pa(hypercall_page);
-	wrmsr_safe(msr, (u32)pfn, (u32)(pfn >> 32));
-
 	x86_init.oem.arch_setup = pvh_arch_setup;
 	x86_init.oem.banner = xen_banner;
 

arch/x86/xen/xen-asm.S

@@ -20,9 +20,32 @@
 
 #include <linux/init.h>
 #include <linux/linkage.h>
+#include <linux/objtool.h>
 #include <../entry/calling.h>
 
 .pushsection .noinstr.text, "ax"
+/*
+ * PV hypercall interface to the hypervisor.
+ *
+ * Called via inline asm(), so better preserve %rcx and %r11.
+ *
+ * Input:
+ *	%eax: hypercall number
+ *	%rdi, %rsi, %rdx, %r10, %r8: args 1..5 for the hypercall
+ * Output: %rax
+ */
+SYM_FUNC_START(xen_hypercall_pv)
+	ANNOTATE_NOENDBR
+	push %rcx
+	push %r11
+	UNWIND_HINT_SAVE
+	syscall
+	UNWIND_HINT_RESTORE
+	pop %r11
+	pop %rcx
+	RET
+SYM_FUNC_END(xen_hypercall_pv)
+
 /*
  * Disabling events is simply a matter of making the event mask
  * non-zero.
@@ -176,7 +199,6 @@ SYM_CODE_START(xen_early_idt_handler_array)
 SYM_CODE_END(xen_early_idt_handler_array)
 	__FINIT
 
-hypercall_iret = hypercall_page + __HYPERVISOR_iret * 32
 /*
  * Xen64 iret frame:
  *
@@ -186,17 +208,28 @@ hypercall_iret = hypercall_page + __HYPERVISOR_iret * 32
  *	cs
  *	rip		<-- standard iret frame
  *
- *	flags
+ *	flags		<-- xen_iret must push from here on
  *
- *	rcx		}
+ *	rcx
- *	r11		}<-- pushed by hypercall page
+ *	r11
- * rsp->rax		}
+ * rsp->rax
  */
+.macro xen_hypercall_iret
+	pushq $0	/* Flags */
+	push %rcx
+	push %r11
+	push %rax
+	mov  $__HYPERVISOR_iret, %eax
+	syscall		/* Do the IRET. */
+#ifdef CONFIG_MITIGATION_SLS
+	int3
+#endif
+.endm
+
 SYM_CODE_START(xen_iret)
 	UNWIND_HINT_UNDEFINED
 	ANNOTATE_NOENDBR
-	pushq $0
+	xen_hypercall_iret
-	jmp hypercall_iret
 SYM_CODE_END(xen_iret)
 
 /*
@@ -301,8 +334,7 @@ SYM_CODE_START(xen_entry_SYSENTER_compat)
 	ENDBR
 	lea 16(%rsp), %rsp	/* strip %rcx, %r11 */
 	mov $-ENOSYS, %rax
-	pushq $0
+	xen_hypercall_iret
-	jmp hypercall_iret
 SYM_CODE_END(xen_entry_SYSENTER_compat)
 SYM_CODE_END(xen_entry_SYSCALL_compat)
 

arch/x86/xen/xen-head.S

@@ -6,9 +6,11 @@
 
 #include <linux/elfnote.h>
 #include <linux/init.h>
+#include <linux/instrumentation.h>
 
 #include <asm/boot.h>
 #include <asm/asm.h>
+#include <asm/frame.h>
 #include <asm/msr.h>
 #include <asm/page_types.h>
 #include <asm/percpu.h>
@@ -20,28 +22,6 @@
 #include <xen/interface/xen-mca.h>
 #include <asm/xen/interface.h>
 
-.pushsection .noinstr.text, "ax"
-	.balign PAGE_SIZE
-SYM_CODE_START(hypercall_page)
-	.rept (PAGE_SIZE / 32)
-		UNWIND_HINT_FUNC
-		ANNOTATE_NOENDBR
-		ANNOTATE_UNRET_SAFE
-		ret
-		/*
-		 * Xen will write the hypercall page, and sort out ENDBR.
-		 */
-		.skip 31, 0xcc
-	.endr
-
-#define HYPERCALL(n) \
-	.equ xen_hypercall_##n, hypercall_page + __HYPERVISOR_##n * 32; \
-	.type xen_hypercall_##n, @function; .size xen_hypercall_##n, 32
-#include <asm/xen-hypercalls.h>
-#undef HYPERCALL
-SYM_CODE_END(hypercall_page)
-.popsection
-
 #ifdef CONFIG_XEN_PV
 	__INIT
 SYM_CODE_START(startup_xen)
@@ -87,6 +67,87 @@ SYM_CODE_END(xen_cpu_bringup_again)
 #endif
 #endif
 
+	.pushsection .noinstr.text, "ax"
+/*
+ * Xen hypercall interface to the hypervisor.
+ *
+ * Input:
+ *     %eax: hypercall number
+ *   32-bit:
+ *     %ebx, %ecx, %edx, %esi, %edi: args 1..5 for the hypercall
+ *   64-bit:
+ *     %rdi, %rsi, %rdx, %r10, %r8: args 1..5 for the hypercall
+ * Output: %[er]ax
+ */
+SYM_FUNC_START(xen_hypercall_hvm)
+	ENDBR
+	FRAME_BEGIN
+	/* Save all relevant registers (caller save and arguments). */
+#ifdef CONFIG_X86_32
+	push %eax
+	push %ebx
+	push %ecx
+	push %edx
+	push %esi
+	push %edi
+#else
+	push %rax
+	push %rcx
+	push %rdx
+	push %rdi
+	push %rsi
+	push %r11
+	push %r10
+	push %r9
+	push %r8
+#ifdef CONFIG_FRAME_POINTER
+	pushq $0	/* Dummy push for stack alignment. */
+#endif
+#endif
+	/* Set the vendor specific function. */
+	call __xen_hypercall_setfunc
+	/* Set ZF = 1 if AMD, Restore saved registers. */
+#ifdef CONFIG_X86_32
+	lea xen_hypercall_amd, %ebx
+	cmp %eax, %ebx
+	pop %edi
+	pop %esi
+	pop %edx
+	pop %ecx
+	pop %ebx
+	pop %eax
+#else
+	lea xen_hypercall_amd(%rip), %rbx
+	cmp %rax, %rbx
+#ifdef CONFIG_FRAME_POINTER
+	pop %rax	/* Dummy pop. */
+#endif
+	pop %r8
+	pop %r9
+	pop %r10
+	pop %r11
+	pop %rsi
+	pop %rdi
+	pop %rdx
+	pop %rcx
+	pop %rax
+#endif
+	/* Use correct hypercall function. */
+	jz xen_hypercall_amd
+	jmp xen_hypercall_intel
+SYM_FUNC_END(xen_hypercall_hvm)
+
+SYM_FUNC_START(xen_hypercall_amd)
+	vmmcall
+	RET
+SYM_FUNC_END(xen_hypercall_amd)
+
+SYM_FUNC_START(xen_hypercall_intel)
+	vmcall
+	RET
+SYM_FUNC_END(xen_hypercall_intel)
+	.popsection
+
 	ELFNOTE(Xen, XEN_ELFNOTE_GUEST_OS,       .asciz "linux")
 	ELFNOTE(Xen, XEN_ELFNOTE_GUEST_VERSION,  .asciz "2.6")
 	ELFNOTE(Xen, XEN_ELFNOTE_XEN_VERSION,    .asciz "xen-3.0")
@@ -116,8 +177,6 @@ SYM_CODE_END(xen_cpu_bringup_again)
 #else
 # define FEATURES_DOM0 0
 #endif
-	ELFNOTE(Xen, XEN_ELFNOTE_HYPERCALL_PAGE, .globl xen_elfnote_hypercall_page;
-		xen_elfnote_hypercall_page: _ASM_PTR xen_elfnote_hypercall_page_value - .)
 	ELFNOTE(Xen, XEN_ELFNOTE_SUPPORTED_FEATURES,
 		.long FEATURES_PV | FEATURES_PVH | FEATURES_DOM0)
 	ELFNOTE(Xen, XEN_ELFNOTE_LOADER,         .asciz "generic")

arch/x86/xen/xen-ops.h

@@ -326,4 +326,13 @@ static inline void xen_smp_intr_free_pv(unsigned int cpu) {}
 static inline void xen_smp_count_cpus(void) { }
 #endif /* CONFIG_SMP */
 
+#ifdef CONFIG_XEN_PV
+void xen_hypercall_pv(void);
+#endif
+void xen_hypercall_hvm(void);
+void xen_hypercall_amd(void);
+void xen_hypercall_intel(void);
+void xen_hypercall_setfunc(void);
+void *__xen_hypercall_setfunc(void);
+
 #endif /* XEN_OPS_H */

block/bio.c

@@ -1171,7 +1171,7 @@ void __bio_release_pages(struct bio *bio, bool mark_dirty)
 }
 EXPORT_SYMBOL_GPL(__bio_release_pages);
 
-void bio_iov_bvec_set(struct bio *bio, struct iov_iter *iter)
+void bio_iov_bvec_set(struct bio *bio, const struct iov_iter *iter)
 {
 	WARN_ON_ONCE(bio->bi_max_vecs);
 

block/blk-cgroup.c

@@ -1324,10 +1324,14 @@ void blkcg_unpin_online(struct cgroup_subsys_state *blkcg_css)
 	struct blkcg *blkcg = css_to_blkcg(blkcg_css);
 
 	do {
+		struct blkcg *parent;
+
 		if (!refcount_dec_and_test(&blkcg->online_pin))
 			break;
+
+		parent = blkcg_parent(blkcg);
 		blkcg_destroy_blkgs(blkcg);
-		blkcg = blkcg_parent(blkcg);
+		blkcg = parent;
 	} while (blkcg);
 }
 

block/blk-iocost.c

@@ -1098,7 +1098,14 @@ static void __propagate_weights(struct ioc_gq *iocg, u32 active, u32 inuse,
 		inuse = DIV64_U64_ROUND_UP(active * iocg->child_inuse_sum,
 					   iocg->child_active_sum);
 	} else {
-		inuse = clamp_t(u32, inuse, 1, active);
+		/*
+		 * It may be tempting to turn this into a clamp expression with
+		 * a lower limit of 1 but active may be 0, which cannot be used
+		 * as an upper limit in that situation. This expression allows
+		 * active to clamp inuse unless it is 0, in which case inuse
+		 * becomes 1.
+		 */
+		inuse = min(inuse, active) ?: 1;
 	}
 
 	iocg->last_inuse = iocg->inuse;

block/blk-map.c

@@ -574,7 +574,7 @@ static int blk_rq_map_user_bvec(struct request *rq, const struct iov_iter *iter)
 	bio = blk_rq_map_bio_alloc(rq, 0, GFP_KERNEL);
 	if (!bio)
 		return -ENOMEM;
-	bio_iov_bvec_set(bio, (struct iov_iter *)iter);
+	bio_iov_bvec_set(bio, iter);
 
 	/* check that the data layout matches the hardware restrictions */
 	ret = bio_split_rw_at(bio, lim, &nsegs, max_bytes);

block/blk-mq-sysfs.c

@@ -275,15 +275,13 @@ void blk_mq_sysfs_unregister_hctxs(struct request_queue *q)
 	struct blk_mq_hw_ctx *hctx;
 	unsigned long i;
 
-	mutex_lock(&q->sysfs_dir_lock);
+	lockdep_assert_held(&q->sysfs_dir_lock);
+
 	if (!q->mq_sysfs_init_done)
-		goto unlock;
+		return;
 
 	queue_for_each_hw_ctx(q, hctx, i)
 		blk_mq_unregister_hctx(hctx);
-
-unlock:
-	mutex_unlock(&q->sysfs_dir_lock);
 }
 
 int blk_mq_sysfs_register_hctxs(struct request_queue *q)
@@ -292,9 +290,10 @@ int blk_mq_sysfs_register_hctxs(struct request_queue *q)
 	unsigned long i;
 	int ret = 0;
 
-	mutex_lock(&q->sysfs_dir_lock);
+	lockdep_assert_held(&q->sysfs_dir_lock);
+
 	if (!q->mq_sysfs_init_done)
-		goto unlock;
+		return ret;
 
 	queue_for_each_hw_ctx(q, hctx, i) {
 		ret = blk_mq_register_hctx(hctx);
@@ -302,8 +301,5 @@ int blk_mq_sysfs_register_hctxs(struct request_queue *q)
 			break;
 	}
 
-unlock:
-	mutex_unlock(&q->sysfs_dir_lock);
-
 	return ret;
 }

block/blk-mq.c

@@ -1544,19 +1544,17 @@ static void blk_mq_requeue_work(struct work_struct *work)
 
 	while (!list_empty(&rq_list)) {
 		rq = list_entry(rq_list.next, struct request, queuelist);
+		list_del_init(&rq->queuelist);
 		/*
-		 * If RQF_DONTPREP ist set, the request has been started by the
+		 * If RQF_DONTPREP is set, the request has been started by the
 		 * driver already and might have driver-specific data allocated
 		 * already.  Insert it into the hctx dispatch list to avoid
 		 * block layer merges for the request.
 		 */
-		if (rq->rq_flags & RQF_DONTPREP) {
+		if (rq->rq_flags & RQF_DONTPREP)
-			list_del_init(&rq->queuelist);
 			blk_mq_request_bypass_insert(rq, 0);
-		} else {
+		else
-			list_del_init(&rq->queuelist);
 			blk_mq_insert_request(rq, BLK_MQ_INSERT_AT_HEAD);
-		}
 	}
 
 	while (!list_empty(&flush_list)) {
@@ -4455,7 +4453,8 @@ static void blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set,
 	unsigned long i, j;
 
 	/* protect against switching io scheduler  */
-	mutex_lock(&q->sysfs_lock);
+	lockdep_assert_held(&q->sysfs_lock);
+
 	for (i = 0; i < set->nr_hw_queues; i++) {
 		int old_node;
 		int node = blk_mq_get_hctx_node(set, i);
@@ -4488,7 +4487,6 @@ static void blk_mq_realloc_hw_ctxs(struct blk_mq_tag_set *set,
 
 	xa_for_each_start(&q->hctx_table, j, hctx, j)
 		blk_mq_exit_hctx(q, set, hctx, j);
-	mutex_unlock(&q->sysfs_lock);
 
 	/* unregister cpuhp callbacks for exited hctxs */
 	blk_mq_remove_hw_queues_cpuhp(q);
@@ -4520,10 +4518,14 @@ int blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
 
 	xa_init(&q->hctx_table);
 
+	mutex_lock(&q->sysfs_lock);
+
 	blk_mq_realloc_hw_ctxs(set, q);
 	if (!q->nr_hw_queues)
 		goto err_hctxs;
 
+	mutex_unlock(&q->sysfs_lock);
+
 	INIT_WORK(&q->timeout_work, blk_mq_timeout_work);
 	blk_queue_rq_timeout(q, set->timeout ? set->timeout : 30 * HZ);
 
@@ -4542,6 +4544,7 @@ int blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
 	return 0;
 
 err_hctxs:
+	mutex_unlock(&q->sysfs_lock);
 	blk_mq_release(q);
 err_exit:
 	q->mq_ops = NULL;
@@ -4922,12 +4925,12 @@ static bool blk_mq_elv_switch_none(struct list_head *head,
 		return false;
 
 	/* q->elevator needs protection from ->sysfs_lock */
-	mutex_lock(&q->sysfs_lock);
+	lockdep_assert_held(&q->sysfs_lock);
 
 	/* the check has to be done with holding sysfs_lock */
 	if (!q->elevator) {
 		kfree(qe);
-		goto unlock;
+		goto out;
 	}
 
 	INIT_LIST_HEAD(&qe->node);
@@ -4937,9 +4940,7 @@ static bool blk_mq_elv_switch_none(struct list_head *head,
 	__elevator_get(qe->type);
 	list_add(&qe->node, head);
 	elevator_disable(q);
-unlock:
+out:
-	mutex_unlock(&q->sysfs_lock);
-
 	return true;
 }
 
@@ -4968,11 +4969,9 @@ static void blk_mq_elv_switch_back(struct list_head *head,
 	list_del(&qe->node);
 	kfree(qe);
 
-	mutex_lock(&q->sysfs_lock);
 	elevator_switch(q, t);
 	/* drop the reference acquired in blk_mq_elv_switch_none */
 	elevator_put(t);
-	mutex_unlock(&q->sysfs_lock);
 }
 
 static void __blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set,
@@ -4992,8 +4991,11 @@ static void __blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set,
 	if (set->nr_maps == 1 && nr_hw_queues == set->nr_hw_queues)
 		return;
 
-	list_for_each_entry(q, &set->tag_list, tag_set_list)
+	list_for_each_entry(q, &set->tag_list, tag_set_list) {
+		mutex_lock(&q->sysfs_dir_lock);
+		mutex_lock(&q->sysfs_lock);
 		blk_mq_freeze_queue(q);
+	}
 	/*
 	 * Switch IO scheduler to 'none', cleaning up the data associated
 	 * with the previous scheduler. We will switch back once we are done
@@ -5049,8 +5051,11 @@ static void __blk_mq_update_nr_hw_queues(struct blk_mq_tag_set *set,
 	list_for_each_entry(q, &set->tag_list, tag_set_list)
 		blk_mq_elv_switch_back(&head, q);
 
-	list_for_each_entry(q, &set->tag_list, tag_set_list)
+	list_for_each_entry(q, &set->tag_list, tag_set_list) {
 		blk_mq_unfreeze_queue(q);
+		mutex_unlock(&q->sysfs_lock);
+		mutex_unlock(&q->sysfs_dir_lock);
+	}
 
 	/* Free the excess tags when nr_hw_queues shrink. */
 	for (i = set->nr_hw_queues; i < prev_nr_hw_queues; i++)

block/blk-sysfs.c

@@ -263,7 +263,7 @@ static ssize_t queue_nr_zones_show(struct gendisk *disk, char *page)
 
 static ssize_t queue_iostats_passthrough_show(struct gendisk *disk, char *page)
 {
-	return queue_var_show(blk_queue_passthrough_stat(disk->queue), page);
+	return queue_var_show(!!blk_queue_passthrough_stat(disk->queue), page);
 }
 
 static ssize_t queue_iostats_passthrough_store(struct gendisk *disk,
@@ -706,11 +706,11 @@ queue_attr_store(struct kobject *kobj, struct attribute *attr,
 	if (entry->load_module)
 		entry->load_module(disk, page, length);
 
-	blk_mq_freeze_queue(q);
 	mutex_lock(&q->sysfs_lock);
+	blk_mq_freeze_queue(q);
 	res = entry->store(disk, page, length);
-	mutex_unlock(&q->sysfs_lock);
 	blk_mq_unfreeze_queue(q);
+	mutex_unlock(&q->sysfs_lock);
 	return res;
 }
 

block/blk-zoned.c

@@ -41,7 +41,6 @@ static const char *const zone_cond_name[] = {
 /*
  * Per-zone write plug.
  * @node: hlist_node structure for managing the plug using a hash table.
- * @link: To list the plug in the zone write plug error list of the disk.
  * @ref: Zone write plug reference counter. A zone write plug reference is
  *       always at least 1 when the plug is hashed in the disk plug hash table.
  *       The reference is incremented whenever a new BIO needing plugging is
@@ -63,7 +62,6 @@ static const char *const zone_cond_name[] = {
  */
 struct blk_zone_wplug {
 	struct hlist_node	node;
-	struct list_head	link;
 	refcount_t		ref;
 	spinlock_t		lock;
 	unsigned int		flags;
@@ -80,8 +78,8 @@ struct blk_zone_wplug {
  *  - BLK_ZONE_WPLUG_PLUGGED: Indicates that the zone write plug is plugged,
  *    that is, that write BIOs are being throttled due to a write BIO already
  *    being executed or the zone write plug bio list is not empty.
- *  - BLK_ZONE_WPLUG_ERROR: Indicates that a write error happened which will be
+ *  - BLK_ZONE_WPLUG_NEED_WP_UPDATE: Indicates that we lost track of a zone
- *    recovered with a report zone to update the zone write pointer offset.
+ *    write pointer offset and need to update it.
  *  - BLK_ZONE_WPLUG_UNHASHED: Indicates that the zone write plug was removed
  *    from the disk hash table and that the initial reference to the zone
  *    write plug set when the plug was first added to the hash table has been
@@ -91,11 +89,9 @@ struct blk_zone_wplug {
  *    freed once all remaining references from BIOs or functions are dropped.
  */
 #define BLK_ZONE_WPLUG_PLUGGED		(1U << 0)
-#define BLK_ZONE_WPLUG_ERROR		(1U << 1)
+#define BLK_ZONE_WPLUG_NEED_WP_UPDATE	(1U << 1)
 #define BLK_ZONE_WPLUG_UNHASHED		(1U << 2)
 
-#define BLK_ZONE_WPLUG_BUSY	(BLK_ZONE_WPLUG_PLUGGED | BLK_ZONE_WPLUG_ERROR)
-
 /**
  * blk_zone_cond_str - Return string XXX in BLK_ZONE_COND_XXX.
  * @zone_cond: BLK_ZONE_COND_XXX.
@@ -115,6 +111,30 @@ const char *blk_zone_cond_str(enum blk_zone_cond zone_cond)
 }
 EXPORT_SYMBOL_GPL(blk_zone_cond_str);
 
+struct disk_report_zones_cb_args {
+	struct gendisk	*disk;
+	report_zones_cb	user_cb;
+	void		*user_data;
+};
+
+static void disk_zone_wplug_sync_wp_offset(struct gendisk *disk,
+					   struct blk_zone *zone);
+
+static int disk_report_zones_cb(struct blk_zone *zone, unsigned int idx,
+				void *data)
+{
+	struct disk_report_zones_cb_args *args = data;
+	struct gendisk *disk = args->disk;
+
+	if (disk->zone_wplugs_hash)
+		disk_zone_wplug_sync_wp_offset(disk, zone);
+
+	if (!args->user_cb)
+		return 0;
+
+	return args->user_cb(zone, idx, args->user_data);
+}
+
 /**
  * blkdev_report_zones - Get zones information
  * @bdev:	Target block device
@@ -139,6 +159,11 @@ int blkdev_report_zones(struct block_device *bdev, sector_t sector,
 {
 	struct gendisk *disk = bdev->bd_disk;
 	sector_t capacity = get_capacity(disk);
+	struct disk_report_zones_cb_args args = {
+		.disk = disk,
+		.user_cb = cb,
+		.user_data = data,
+	};
 
 	if (!bdev_is_zoned(bdev) || WARN_ON_ONCE(!disk->fops->report_zones))
 		return -EOPNOTSUPP;
@@ -146,7 +171,8 @@ int blkdev_report_zones(struct block_device *bdev, sector_t sector,
 	if (!nr_zones || sector >= capacity)
 		return 0;
 
-	return disk->fops->report_zones(disk, sector, nr_zones, cb, data);
+	return disk->fops->report_zones(disk, sector, nr_zones,
+					disk_report_zones_cb, &args);
 }
 EXPORT_SYMBOL_GPL(blkdev_report_zones);
 
@@ -427,7 +453,7 @@ static inline void disk_put_zone_wplug(struct blk_zone_wplug *zwplug)
 {
 	if (refcount_dec_and_test(&zwplug->ref)) {
 		WARN_ON_ONCE(!bio_list_empty(&zwplug->bio_list));
-		WARN_ON_ONCE(!list_empty(&zwplug->link));
+		WARN_ON_ONCE(zwplug->flags & BLK_ZONE_WPLUG_PLUGGED);
 		WARN_ON_ONCE(!(zwplug->flags & BLK_ZONE_WPLUG_UNHASHED));
 
 		call_rcu(&zwplug->rcu_head, disk_free_zone_wplug_rcu);
@@ -441,8 +467,8 @@ static inline bool disk_should_remove_zone_wplug(struct gendisk *disk,
 	if (zwplug->flags & BLK_ZONE_WPLUG_UNHASHED)
 		return false;
 
-	/* If the zone write plug is still busy, it cannot be removed. */
+	/* If the zone write plug is still plugged, it cannot be removed. */
-	if (zwplug->flags & BLK_ZONE_WPLUG_BUSY)
+	if (zwplug->flags & BLK_ZONE_WPLUG_PLUGGED)
 		return false;
 
 	/*
@@ -525,12 +551,11 @@ static struct blk_zone_wplug *disk_get_and_lock_zone_wplug(struct gendisk *disk,
 		return NULL;
 
 	INIT_HLIST_NODE(&zwplug->node);
-	INIT_LIST_HEAD(&zwplug->link);
 	refcount_set(&zwplug->ref, 2);
 	spin_lock_init(&zwplug->lock);
 	zwplug->flags = 0;
 	zwplug->zone_no = zno;
-	zwplug->wp_offset = sector & (disk->queue->limits.chunk_sectors - 1);
+	zwplug->wp_offset = bdev_offset_from_zone_start(disk->part0, sector);
 	bio_list_init(&zwplug->bio_list);
 	INIT_WORK(&zwplug->bio_work, blk_zone_wplug_bio_work);
 	zwplug->disk = disk;
@@ -574,115 +599,22 @@ static void disk_zone_wplug_abort(struct blk_zone_wplug *zwplug)
 }
 
 /*
- * Abort (fail) all plugged BIOs of a zone write plug that are not aligned
+ * Set a zone write plug write pointer offset to the specified value.
- * with the assumed write pointer location of the zone when the BIO will
+ * This aborts all plugged BIOs, which is fine as this function is called for
- * be unplugged.
+ * a zone reset operation, a zone finish operation or if the zone needs a wp
- */
+ * update from a report zone after a write error.
-static void disk_zone_wplug_abort_unaligned(struct gendisk *disk,
-					    struct blk_zone_wplug *zwplug)
-{
-	unsigned int wp_offset = zwplug->wp_offset;
-	struct bio_list bl = BIO_EMPTY_LIST;
-	struct bio *bio;
-
-	while ((bio = bio_list_pop(&zwplug->bio_list))) {
-		if (disk_zone_is_full(disk, zwplug->zone_no, wp_offset) ||
-		    (bio_op(bio) != REQ_OP_ZONE_APPEND &&
-		     bio_offset_from_zone_start(bio) != wp_offset)) {
-			blk_zone_wplug_bio_io_error(zwplug, bio);
-			continue;
-		}
-
-		wp_offset += bio_sectors(bio);
-		bio_list_add(&bl, bio);
-	}
-
-	bio_list_merge(&zwplug->bio_list, &bl);
-}
-
-static inline void disk_zone_wplug_set_error(struct gendisk *disk,
-					     struct blk_zone_wplug *zwplug)
-{
-	unsigned long flags;
-
-	if (zwplug->flags & BLK_ZONE_WPLUG_ERROR)
-		return;
-
-	/*
-	 * At this point, we already have a reference on the zone write plug.
-	 * However, since we are going to add the plug to the disk zone write
-	 * plugs work list, increase its reference count. This reference will
-	 * be dropped in disk_zone_wplugs_work() once the error state is
-	 * handled, or in disk_zone_wplug_clear_error() if the zone is reset or
-	 * finished.
-	 */
-	zwplug->flags |= BLK_ZONE_WPLUG_ERROR;
-	refcount_inc(&zwplug->ref);
-
-	spin_lock_irqsave(&disk->zone_wplugs_lock, flags);
-	list_add_tail(&zwplug->link, &disk->zone_wplugs_err_list);
-	spin_unlock_irqrestore(&disk->zone_wplugs_lock, flags);
-}
-
-static inline void disk_zone_wplug_clear_error(struct gendisk *disk,
-					       struct blk_zone_wplug *zwplug)
-{
-	unsigned long flags;
-
-	if (!(zwplug->flags & BLK_ZONE_WPLUG_ERROR))
-		return;
-
-	/*
-	 * We are racing with the error handling work which drops the reference
-	 * on the zone write plug after handling the error state. So remove the
-	 * plug from the error list and drop its reference count only if the
-	 * error handling has not yet started, that is, if the zone write plug
-	 * is still listed.
-	 */
-	spin_lock_irqsave(&disk->zone_wplugs_lock, flags);
-	if (!list_empty(&zwplug->link)) {
-		list_del_init(&zwplug->link);
-		zwplug->flags &= ~BLK_ZONE_WPLUG_ERROR;
-		disk_put_zone_wplug(zwplug);
-	}
-	spin_unlock_irqrestore(&disk->zone_wplugs_lock, flags);
-}
-
-/*
- * Set a zone write plug write pointer offset to either 0 (zone reset case)
- * or to the zone size (zone finish case). This aborts all plugged BIOs, which
- * is fine to do as doing a zone reset or zone finish while writes are in-flight
- * is a mistake from the user which will most likely cause all plugged BIOs to
- * fail anyway.
  */
 static void disk_zone_wplug_set_wp_offset(struct gendisk *disk,
 					  struct blk_zone_wplug *zwplug,
 					  unsigned int wp_offset)
 {
-	unsigned long flags;
+	lockdep_assert_held(&zwplug->lock);
-
-	spin_lock_irqsave(&zwplug->lock, flags);
-
-	/*
-	 * Make sure that a BIO completion or another zone reset or finish
-	 * operation has not already removed the plug from the hash table.
-	 */
-	if (zwplug->flags & BLK_ZONE_WPLUG_UNHASHED) {
-		spin_unlock_irqrestore(&zwplug->lock, flags);
-		return;
-	}
 
 	/* Update the zone write pointer and abort all plugged BIOs. */
+	zwplug->flags &= ~BLK_ZONE_WPLUG_NEED_WP_UPDATE;
 	zwplug->wp_offset = wp_offset;
 	disk_zone_wplug_abort(zwplug);
 
-	/*
-	 * Updating the write pointer offset puts back the zone
-	 * in a good state. So clear the error flag and decrement the
-	 * error count if we were in error state.
-	 */
-	disk_zone_wplug_clear_error(disk, zwplug);
-
 	/*
 	 * The zone write plug now has no BIO plugged: remove it from the
 	 * hash table so that it cannot be seen. The plug will be freed
@@ -690,8 +622,58 @@ static void disk_zone_wplug_set_wp_offset(struct gendisk *disk,
 	 */
 	if (disk_should_remove_zone_wplug(disk, zwplug))
 		disk_remove_zone_wplug(disk, zwplug);
+}
 
+static unsigned int blk_zone_wp_offset(struct blk_zone *zone)
+{
+	switch (zone->cond) {
+	case BLK_ZONE_COND_IMP_OPEN:
+	case BLK_ZONE_COND_EXP_OPEN:
+	case BLK_ZONE_COND_CLOSED:
+		return zone->wp - zone->start;
+	case BLK_ZONE_COND_FULL:
+		return zone->len;
+	case BLK_ZONE_COND_EMPTY:
+		return 0;
+	case BLK_ZONE_COND_NOT_WP:
+	case BLK_ZONE_COND_OFFLINE:
+	case BLK_ZONE_COND_READONLY:
+	default:
+		/*
+		 * Conventional, offline and read-only zones do not have a valid
+		 * write pointer.
+		 */
+		return UINT_MAX;
+	}
+}
+
+static void disk_zone_wplug_sync_wp_offset(struct gendisk *disk,
+					   struct blk_zone *zone)
+{
+	struct blk_zone_wplug *zwplug;
+	unsigned long flags;
+
+	zwplug = disk_get_zone_wplug(disk, zone->start);
+	if (!zwplug)
+		return;
+
+	spin_lock_irqsave(&zwplug->lock, flags);
+	if (zwplug->flags & BLK_ZONE_WPLUG_NEED_WP_UPDATE)
+		disk_zone_wplug_set_wp_offset(disk, zwplug,
+					      blk_zone_wp_offset(zone));
 	spin_unlock_irqrestore(&zwplug->lock, flags);
+
+	disk_put_zone_wplug(zwplug);
+}
+
+static int disk_zone_sync_wp_offset(struct gendisk *disk, sector_t sector)
+{
+	struct disk_report_zones_cb_args args = {
+		.disk = disk,
+	};
+
+	return disk->fops->report_zones(disk, sector, 1,
+					disk_report_zones_cb, &args);
 }
 
 static bool blk_zone_wplug_handle_reset_or_finish(struct bio *bio,
@@ -700,6 +682,7 @@ static bool blk_zone_wplug_handle_reset_or_finish(struct bio *bio,
 	struct gendisk *disk = bio->bi_bdev->bd_disk;
 	sector_t sector = bio->bi_iter.bi_sector;
 	struct blk_zone_wplug *zwplug;
+	unsigned long flags;
 
 	/* Conventional zones cannot be reset nor finished. */
 	if (!bdev_zone_is_seq(bio->bi_bdev, sector)) {
@@ -707,6 +690,15 @@ static bool blk_zone_wplug_handle_reset_or_finish(struct bio *bio,
 		return true;
 	}
 
+	/*
+	 * No-wait reset or finish BIOs do not make much sense as the callers
+	 * issue these as blocking operations in most cases. To avoid issues
+	 * the BIO execution potentially failing with BLK_STS_AGAIN, warn about
+	 * REQ_NOWAIT being set and ignore that flag.
+	 */
+	if (WARN_ON_ONCE(bio->bi_opf & REQ_NOWAIT))
+		bio->bi_opf &= ~REQ_NOWAIT;
+
 	/*
 	 * If we have a zone write plug, set its write pointer offset to 0
 	 * (reset case) or to the zone size (finish case). This will abort all
@@ -716,7 +708,9 @@ static bool blk_zone_wplug_handle_reset_or_finish(struct bio *bio,
 	 */
 	zwplug = disk_get_zone_wplug(disk, sector);
 	if (zwplug) {
+		spin_lock_irqsave(&zwplug->lock, flags);
 		disk_zone_wplug_set_wp_offset(disk, zwplug, wp_offset);
+		spin_unlock_irqrestore(&zwplug->lock, flags);
 		disk_put_zone_wplug(zwplug);
 	}
 
@@ -727,6 +721,7 @@ static bool blk_zone_wplug_handle_reset_all(struct bio *bio)
 {
 	struct gendisk *disk = bio->bi_bdev->bd_disk;
 	struct blk_zone_wplug *zwplug;
+	unsigned long flags;
 	sector_t sector;
 
 	/*
@@ -738,7 +733,9 @@ static bool blk_zone_wplug_handle_reset_all(struct bio *bio)
 	     sector += disk->queue->limits.chunk_sectors) {
 		zwplug = disk_get_zone_wplug(disk, sector);
 		if (zwplug) {
+			spin_lock_irqsave(&zwplug->lock, flags);
 			disk_zone_wplug_set_wp_offset(disk, zwplug, 0);
+			spin_unlock_irqrestore(&zwplug->lock, flags);
 			disk_put_zone_wplug(zwplug);
 		}
 	}
@@ -746,9 +743,25 @@ static bool blk_zone_wplug_handle_reset_all(struct bio *bio)
 	return false;
 }
 
-static inline void blk_zone_wplug_add_bio(struct blk_zone_wplug *zwplug,
+static void disk_zone_wplug_schedule_bio_work(struct gendisk *disk,
-					  struct bio *bio, unsigned int nr_segs)
+					      struct blk_zone_wplug *zwplug)
 {
+	/*
+	 * Take a reference on the zone write plug and schedule the submission
+	 * of the next plugged BIO. blk_zone_wplug_bio_work() will release the
+	 * reference we take here.
+	 */
+	WARN_ON_ONCE(!(zwplug->flags & BLK_ZONE_WPLUG_PLUGGED));
+	refcount_inc(&zwplug->ref);
+	queue_work(disk->zone_wplugs_wq, &zwplug->bio_work);
+}
+
+static inline void disk_zone_wplug_add_bio(struct gendisk *disk,
+				struct blk_zone_wplug *zwplug,
+				struct bio *bio, unsigned int nr_segs)
+{
+	bool schedule_bio_work = false;
+
 	/*
 	 * Grab an extra reference on the BIO request queue usage counter.
 	 * This reference will be reused to submit a request for the BIO for
@@ -764,6 +777,16 @@ static inline void blk_zone_wplug_add_bio(struct blk_zone_wplug *zwplug,
 	 */
 	bio_clear_polled(bio);
 
+	/*
+	 * REQ_NOWAIT BIOs are always handled using the zone write plug BIO
+	 * work, which can block. So clear the REQ_NOWAIT flag and schedule the
+	 * work if this is the first BIO we are plugging.
+	 */
+	if (bio->bi_opf & REQ_NOWAIT) {
+		schedule_bio_work = !(zwplug->flags & BLK_ZONE_WPLUG_PLUGGED);
+		bio->bi_opf &= ~REQ_NOWAIT;
+	}
+
 	/*
 	 * Reuse the poll cookie field to store the number of segments when
 	 * split to the hardware limits.
@@ -777,6 +800,11 @@ static inline void blk_zone_wplug_add_bio(struct blk_zone_wplug *zwplug,
 	 * at the tail of the list to preserve the sequential write order.
 	 */
 	bio_list_add(&zwplug->bio_list, bio);
+
+	zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;
+
+	if (schedule_bio_work)
+		disk_zone_wplug_schedule_bio_work(disk, zwplug);
 }
 
 /*
@@ -889,13 +917,23 @@ static bool blk_zone_wplug_prepare_bio(struct blk_zone_wplug *zwplug,
 {
 	struct gendisk *disk = bio->bi_bdev->bd_disk;
 
+	/*
+	 * If we lost track of the zone write pointer due to a write error,
+	 * the user must either execute a report zones, reset the zone or finish
+	 * the to recover a reliable write pointer position. Fail BIOs if the
+	 * user did not do that as we cannot handle emulated zone append
+	 * otherwise.
+	 */
+	if (zwplug->flags & BLK_ZONE_WPLUG_NEED_WP_UPDATE)
+		return false;
+
 	/*
 	 * Check that the user is not attempting to write to a full zone.
 	 * We know such BIO will fail, and that would potentially overflow our
 	 * write pointer offset beyond the end of the zone.
 	 */
 	if (disk_zone_wplug_is_full(disk, zwplug))
-		goto err;
+		return false;
 
 	if (bio_op(bio) == REQ_OP_ZONE_APPEND) {
 		/*
@@ -914,24 +952,18 @@ static bool blk_zone_wplug_prepare_bio(struct blk_zone_wplug *zwplug,
 		bio_set_flag(bio, BIO_EMULATES_ZONE_APPEND);
 	} else {
 		/*
-		 * Check for non-sequential writes early because we avoid a
+		 * Check for non-sequential writes early as we know that BIOs
-		 * whole lot of error handling trouble if we don't send it off
+		 * with a start sector not unaligned to the zone write pointer
-		 * to the driver.
+		 * will fail.
 		 */
 		if (bio_offset_from_zone_start(bio) != zwplug->wp_offset)
-			goto err;
+			return false;
 	}
 
 	/* Advance the zone write pointer offset. */
 	zwplug->wp_offset += bio_sectors(bio);
 
 	return true;
-
-err:
-	/* We detected an invalid write BIO: schedule error recovery. */
-	disk_zone_wplug_set_error(disk, zwplug);
-	kblockd_schedule_work(&disk->zone_wplugs_work);
-	return false;
 }
 
 static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
@@ -970,7 +1002,10 @@ static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
 
 	zwplug = disk_get_and_lock_zone_wplug(disk, sector, gfp_mask, &flags);
 	if (!zwplug) {
-		bio_io_error(bio);
+		if (bio->bi_opf & REQ_NOWAIT)
+			bio_wouldblock_error(bio);
+		else
+			bio_io_error(bio);
 		return true;
 	}
 
@@ -978,18 +1013,20 @@ static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
 	bio_set_flag(bio, BIO_ZONE_WRITE_PLUGGING);
 
 	/*
-	 * If the zone is already plugged or has a pending error, add the BIO
+	 * If the zone is already plugged, add the BIO to the plug BIO list.
-	 * to the plug BIO list. Otherwise, plug and let the BIO execute.
+	 * Do the same for REQ_NOWAIT BIOs to ensure that we will not see a
+	 * BLK_STS_AGAIN failure if we let the BIO execute.
+	 * Otherwise, plug and let the BIO execute.
 	 */
-	if (zwplug->flags & BLK_ZONE_WPLUG_BUSY)
+	if ((zwplug->flags & BLK_ZONE_WPLUG_PLUGGED) ||
+	    (bio->bi_opf & REQ_NOWAIT))
 		goto plug;
 
-	/*
+	if (!blk_zone_wplug_prepare_bio(zwplug, bio)) {
-	 * If an error is detected when preparing the BIO, add it to the BIO
+		spin_unlock_irqrestore(&zwplug->lock, flags);
-	 * list so that error recovery can deal with it.
+		bio_io_error(bio);
-	 */
+		return true;
-	if (!blk_zone_wplug_prepare_bio(zwplug, bio))
+	}
-		goto plug;
 
 	zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;
 
@@ -998,8 +1035,7 @@ static bool blk_zone_wplug_handle_write(struct bio *bio, unsigned int nr_segs)
 	return false;
 
 plug:
-	zwplug->flags |= BLK_ZONE_WPLUG_PLUGGED;
+	disk_zone_wplug_add_bio(disk, zwplug, bio, nr_segs);
-	blk_zone_wplug_add_bio(zwplug, bio, nr_segs);
 
 	spin_unlock_irqrestore(&zwplug->lock, flags);
 
@@ -1083,19 +1119,6 @@ bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs)
 }
 EXPORT_SYMBOL_GPL(blk_zone_plug_bio);
 
-static void disk_zone_wplug_schedule_bio_work(struct gendisk *disk,
-					      struct blk_zone_wplug *zwplug)
-{
-	/*
-	 * Take a reference on the zone write plug and schedule the submission
-	 * of the next plugged BIO. blk_zone_wplug_bio_work() will release the
-	 * reference we take here.
-	 */
-	WARN_ON_ONCE(!(zwplug->flags & BLK_ZONE_WPLUG_PLUGGED));
-	refcount_inc(&zwplug->ref);
-	queue_work(disk->zone_wplugs_wq, &zwplug->bio_work);
-}
-
 static void disk_zone_wplug_unplug_bio(struct gendisk *disk,
 				       struct blk_zone_wplug *zwplug)
 {
@@ -1103,16 +1126,6 @@ static void disk_zone_wplug_unplug_bio(struct gendisk *disk,
 
 	spin_lock_irqsave(&zwplug->lock, flags);
 
-	/*
-	 * If we had an error, schedule error recovery. The recovery work
-	 * will restart submission of plugged BIOs.
-	 */
-	if (zwplug->flags & BLK_ZONE_WPLUG_ERROR) {
-		spin_unlock_irqrestore(&zwplug->lock, flags);
-		kblockd_schedule_work(&disk->zone_wplugs_work);
-		return;
-	}
-
 	/* Schedule submission of the next plugged BIO if we have one. */
 	if (!bio_list_empty(&zwplug->bio_list)) {
 		disk_zone_wplug_schedule_bio_work(disk, zwplug);
@@ -1155,12 +1168,13 @@ void blk_zone_write_plug_bio_endio(struct bio *bio)
 	}
 
 	/*
-	 * If the BIO failed, mark the plug as having an error to trigger
+	 * If the BIO failed, abort all plugged BIOs and mark the plug as
-	 * recovery.
+	 * needing a write pointer update.
 	 */
 	if (bio->bi_status != BLK_STS_OK) {
 		spin_lock_irqsave(&zwplug->lock, flags);
-		disk_zone_wplug_set_error(disk, zwplug);
+		disk_zone_wplug_abort(zwplug);
+		zwplug->flags |= BLK_ZONE_WPLUG_NEED_WP_UPDATE;
 		spin_unlock_irqrestore(&zwplug->lock, flags);
 	}
 
@@ -1216,6 +1230,7 @@ static void blk_zone_wplug_bio_work(struct work_struct *work)
 	 */
 	spin_lock_irqsave(&zwplug->lock, flags);
 
+again:
 	bio = bio_list_pop(&zwplug->bio_list);
 	if (!bio) {
 		zwplug->flags &= ~BLK_ZONE_WPLUG_PLUGGED;
@@ -1224,10 +1239,8 @@ static void blk_zone_wplug_bio_work(struct work_struct *work)
 	}
 
 	if (!blk_zone_wplug_prepare_bio(zwplug, bio)) {
-		/* Error recovery will decide what to do with the BIO. */
+		blk_zone_wplug_bio_io_error(zwplug, bio);
-		bio_list_add_head(&zwplug->bio_list, bio);
+		goto again;
-		spin_unlock_irqrestore(&zwplug->lock, flags);
-		goto put_zwplug;
 	}
 
 	spin_unlock_irqrestore(&zwplug->lock, flags);
@@ -1249,120 +1262,6 @@ static void blk_zone_wplug_bio_work(struct work_struct *work)
 	disk_put_zone_wplug(zwplug);
 }
 
-static unsigned int blk_zone_wp_offset(struct blk_zone *zone)
-{
-	switch (zone->cond) {
-	case BLK_ZONE_COND_IMP_OPEN:
-	case BLK_ZONE_COND_EXP_OPEN:
-	case BLK_ZONE_COND_CLOSED:
-		return zone->wp - zone->start;
-	case BLK_ZONE_COND_FULL:
-		return zone->len;
-	case BLK_ZONE_COND_EMPTY:
-		return 0;
-	case BLK_ZONE_COND_NOT_WP:
-	case BLK_ZONE_COND_OFFLINE:
-	case BLK_ZONE_COND_READONLY:
-	default:
-		/*
-		 * Conventional, offline and read-only zones do not have a valid
-		 * write pointer.
-		 */
-		return UINT_MAX;
-	}
-}
-
-static int blk_zone_wplug_report_zone_cb(struct blk_zone *zone,
-					 unsigned int idx, void *data)
-{
-	struct blk_zone *zonep = data;
-
-	*zonep = *zone;
-	return 0;
-}
-
-static void disk_zone_wplug_handle_error(struct gendisk *disk,
-					 struct blk_zone_wplug *zwplug)
-{
-	sector_t zone_start_sector =
-		bdev_zone_sectors(disk->part0) * zwplug->zone_no;
-	unsigned int noio_flag;
-	struct blk_zone zone;
-	unsigned long flags;
-	int ret;
-
-	/* Get the current zone information from the device. */
-	noio_flag = memalloc_noio_save();
-	ret = disk->fops->report_zones(disk, zone_start_sector, 1,
-				       blk_zone_wplug_report_zone_cb, &zone);
-	memalloc_noio_restore(noio_flag);
-
-	spin_lock_irqsave(&zwplug->lock, flags);
-
-	/*
-	 * A zone reset or finish may have cleared the error already. In such
-	 * case, do nothing as the report zones may have seen the "old" write
-	 * pointer value before the reset/finish operation completed.
-	 */
-	if (!(zwplug->flags & BLK_ZONE_WPLUG_ERROR))
-		goto unlock;
-
-	zwplug->flags &= ~BLK_ZONE_WPLUG_ERROR;
-
-	if (ret != 1) {
-		/*
-		 * We failed to get the zone information, meaning that something
-		 * is likely really wrong with the device. Abort all remaining
-		 * plugged BIOs as otherwise we could endup waiting forever on
-		 * plugged BIOs to complete if there is a queue freeze on-going.
-		 */
-		disk_zone_wplug_abort(zwplug);
-		goto unplug;
-	}
-
-	/* Update the zone write pointer offset. */
-	zwplug->wp_offset = blk_zone_wp_offset(&zone);
-	disk_zone_wplug_abort_unaligned(disk, zwplug);
-
-	/* Restart BIO submission if we still have any BIO left. */
-	if (!bio_list_empty(&zwplug->bio_list)) {
-		disk_zone_wplug_schedule_bio_work(disk, zwplug);
-		goto unlock;
-	}
-
-unplug:
-	zwplug->flags &= ~BLK_ZONE_WPLUG_PLUGGED;
-	if (disk_should_remove_zone_wplug(disk, zwplug))
-		disk_remove_zone_wplug(disk, zwplug);
-
-unlock:
-	spin_unlock_irqrestore(&zwplug->lock, flags);
-}
-
-static void disk_zone_wplugs_work(struct work_struct *work)
-{
-	struct gendisk *disk =
-		container_of(work, struct gendisk, zone_wplugs_work);
-	struct blk_zone_wplug *zwplug;
-	unsigned long flags;
-
-	spin_lock_irqsave(&disk->zone_wplugs_lock, flags);
-
-	while (!list_empty(&disk->zone_wplugs_err_list)) {
-		zwplug = list_first_entry(&disk->zone_wplugs_err_list,
-					  struct blk_zone_wplug, link);
-		list_del_init(&zwplug->link);
-		spin_unlock_irqrestore(&disk->zone_wplugs_lock, flags);
-
-		disk_zone_wplug_handle_error(disk, zwplug);
-		disk_put_zone_wplug(zwplug);
-
-		spin_lock_irqsave(&disk->zone_wplugs_lock, flags);
-	}
-
-	spin_unlock_irqrestore(&disk->zone_wplugs_lock, flags);
-}
-
 static inline unsigned int disk_zone_wplugs_hash_size(struct gendisk *disk)
 {
 	return 1U << disk->zone_wplugs_hash_bits;
@@ -1371,8 +1270,6 @@ static inline unsigned int disk_zone_wplugs_hash_size(struct gendisk *disk)
 void disk_init_zone_resources(struct gendisk *disk)
 {
 	spin_lock_init(&disk->zone_wplugs_lock);
-	INIT_LIST_HEAD(&disk->zone_wplugs_err_list);
-	INIT_WORK(&disk->zone_wplugs_work, disk_zone_wplugs_work);
 }
 
 /*
@@ -1471,8 +1368,6 @@ void disk_free_zone_resources(struct gendisk *disk)
 	if (!disk->zone_wplugs_pool)
 		return;
 
-	cancel_work_sync(&disk->zone_wplugs_work);
-
 	if (disk->zone_wplugs_wq) {
 		destroy_workqueue(disk->zone_wplugs_wq);
 		disk->zone_wplugs_wq = NULL;
@@ -1669,6 +1564,8 @@ static int blk_revalidate_seq_zone(struct blk_zone *zone, unsigned int idx,
 	if (!disk->zone_wplugs_hash)
 		return 0;
 
+	disk_zone_wplug_sync_wp_offset(disk, zone);
+
 	wp_offset = blk_zone_wp_offset(zone);
 	if (!wp_offset || wp_offset >= zone->capacity)
 		return 0;
@@ -1799,6 +1696,7 @@ int blk_revalidate_disk_zones(struct gendisk *disk)
 		memalloc_noio_restore(noio_flag);
 		return ret;
 	}
+
 	ret = disk->fops->report_zones(disk, 0, UINT_MAX,
 				       blk_revalidate_zone_cb, &args);
 	if (!ret) {
@@ -1835,6 +1733,48 @@ int blk_revalidate_disk_zones(struct gendisk *disk)
 }
 EXPORT_SYMBOL_GPL(blk_revalidate_disk_zones);
 
+/**
+ * blk_zone_issue_zeroout - zero-fill a block range in a zone
+ * @bdev:	blockdev to write
+ * @sector:	start sector
+ * @nr_sects:	number of sectors to write
+ * @gfp_mask:	memory allocation flags (for bio_alloc)
+ *
+ * Description:
+ *  Zero-fill a block range in a zone (@sector must be equal to the zone write
+ *  pointer), handling potential errors due to the (initially unknown) lack of
+ *  hardware offload (See blkdev_issue_zeroout()).
+ */
+int blk_zone_issue_zeroout(struct block_device *bdev, sector_t sector,
+			   sector_t nr_sects, gfp_t gfp_mask)
+{
+	int ret;
+
+	if (WARN_ON_ONCE(!bdev_is_zoned(bdev)))
+		return -EIO;
+
+	ret = blkdev_issue_zeroout(bdev, sector, nr_sects, gfp_mask,
+				   BLKDEV_ZERO_NOFALLBACK);
+	if (ret != -EOPNOTSUPP)
+		return ret;
+
+	/*
+	 * The failed call to blkdev_issue_zeroout() advanced the zone write
+	 * pointer. Undo this using a report zone to update the zone write
+	 * pointer to the correct current value.
+	 */
+	ret = disk_zone_sync_wp_offset(bdev->bd_disk, sector);
+	if (ret != 1)
+		return ret < 0 ? ret : -EIO;
+
+	/*
+	 * Retry without BLKDEV_ZERO_NOFALLBACK to force the fallback to a
+	 * regular write with zero-pages.
+	 */
+	return blkdev_issue_zeroout(bdev, sector, nr_sects, gfp_mask, 0);
+}
+EXPORT_SYMBOL_GPL(blk_zone_issue_zeroout);
+
 #ifdef CONFIG_BLK_DEBUG_FS
 
 int queue_zone_wplugs_show(void *data, struct seq_file *m)

block/mq-deadline.c

@@ -698,8 +698,6 @@ static void dd_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq,
 		list_add(&rq->queuelist, &per_prio->dispatch);
 		rq->fifo_time = jiffies;
 	} else {
-		struct list_head *insert_before;
-
 		deadline_add_rq_rb(per_prio, rq);
 
 		if (rq_mergeable(rq)) {
@@ -712,8 +710,7 @@ static void dd_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq,
 		 * set expire time and add to fifo list
 		 */
 		rq->fifo_time = jiffies + dd->fifo_expire[data_dir];
-		insert_before = &per_prio->fifo_list[data_dir];
+		list_add_tail(&rq->queuelist, &per_prio->fifo_list[data_dir]);
-		list_add_tail(&rq->queuelist, insert_before);
 	}
 }
 

crypto/rsassa-pkcs1.c

@@ -163,10 +163,6 @@ static int rsassa_pkcs1_sign(struct crypto_sig *tfm,
 	struct rsassa_pkcs1_inst_ctx *ictx = sig_instance_ctx(inst);
 	const struct hash_prefix *hash_prefix = ictx->hash_prefix;
 	struct rsassa_pkcs1_ctx *ctx = crypto_sig_ctx(tfm);
-	unsigned int child_reqsize = crypto_akcipher_reqsize(ctx->child);
-	struct akcipher_request *child_req __free(kfree_sensitive) = NULL;
-	struct scatterlist in_sg[3], out_sg;
-	struct crypto_wait cwait;
 	unsigned int pad_len;
 	unsigned int ps_end;
 	unsigned int len;
@@ -187,37 +183,25 @@ static int rsassa_pkcs1_sign(struct crypto_sig *tfm,
 
 	pad_len = ctx->key_size - slen - hash_prefix->size - 1;
 
-	child_req = kmalloc(sizeof(*child_req) + child_reqsize + pad_len,
-			    GFP_KERNEL);
-	if (!child_req)
-		return -ENOMEM;
-
 	/* RFC 8017 sec 8.2.1 step 1 - EMSA-PKCS1-v1_5 encoding generation */
-	in_buf = (u8 *)(child_req + 1) + child_reqsize;
+	in_buf = dst;
+	memmove(in_buf + pad_len + hash_prefix->size, src, slen);
+	memcpy(in_buf + pad_len, hash_prefix->data, hash_prefix->size);
+
 	ps_end = pad_len - 1;
 	in_buf[0] = 0x01;
 	memset(in_buf + 1, 0xff, ps_end - 1);
 	in_buf[ps_end] = 0x00;
 
-	/* RFC 8017 sec 8.2.1 step 2 - RSA signature */
-	crypto_init_wait(&cwait);
-	sg_init_table(in_sg, 3);
-	sg_set_buf(&in_sg[0], in_buf, pad_len);
-	sg_set_buf(&in_sg[1], hash_prefix->data, hash_prefix->size);
-	sg_set_buf(&in_sg[2], src, slen);
-	sg_init_one(&out_sg, dst, dlen);
-	akcipher_request_set_tfm(child_req, ctx->child);
-	akcipher_request_set_crypt(child_req, in_sg, &out_sg,
-				   ctx->key_size - 1, dlen);
-	akcipher_request_set_callback(child_req, CRYPTO_TFM_REQ_MAY_SLEEP,
-				      crypto_req_done, &cwait);
 
-	err = crypto_akcipher_decrypt(child_req);
+	/* RFC 8017 sec 8.2.1 step 2 - RSA signature */
-	err = crypto_wait_req(err, &cwait);
+	err = crypto_akcipher_sync_decrypt(ctx->child, in_buf,
-	if (err)
+					   ctx->key_size - 1, in_buf,
+					   ctx->key_size);
+	if (err < 0)
 		return err;
 
-	len = child_req->dst_len;
+	len = err;
 	pad_len = ctx->key_size - len;
 
 	/* Four billion to one */
@@ -239,8 +223,8 @@ static int rsassa_pkcs1_verify(struct crypto_sig *tfm,
 	struct rsassa_pkcs1_ctx *ctx = crypto_sig_ctx(tfm);
 	unsigned int child_reqsize = crypto_akcipher_reqsize(ctx->child);
 	struct akcipher_request *child_req __free(kfree_sensitive) = NULL;
-	struct scatterlist in_sg, out_sg;
 	struct crypto_wait cwait;
+	struct scatterlist sg;
 	unsigned int dst_len;
 	unsigned int pos;
 	u8 *out_buf;
@@ -259,13 +243,12 @@ static int rsassa_pkcs1_verify(struct crypto_sig *tfm,
 		return -ENOMEM;
 
 	out_buf = (u8 *)(child_req + 1) + child_reqsize;
+	memcpy(out_buf, src, slen);
 
 	crypto_init_wait(&cwait);
-	sg_init_one(&in_sg, src, slen);
+	sg_init_one(&sg, out_buf, slen);
-	sg_init_one(&out_sg, out_buf, ctx->key_size);
 	akcipher_request_set_tfm(child_req, ctx->child);
-	akcipher_request_set_crypt(child_req, &in_sg, &out_sg,
+	akcipher_request_set_crypt(child_req, &sg, &sg, slen, slen);
-				   slen, ctx->key_size);
 	akcipher_request_set_callback(child_req, CRYPTO_TFM_REQ_MAY_SLEEP,
 				      crypto_req_done, &cwait);
 

drivers/acpi/acpica/evxfregn.c

@@ -232,8 +232,6 @@ acpi_remove_address_space_handler(acpi_handle device,
 
 			/* Now we can delete the handler object */
 
-			acpi_os_release_mutex(handler_obj->address_space.
-					      context_mutex);
 			acpi_ut_remove_reference(handler_obj);
 			goto unlock_and_exit;
 		}

drivers/acpi/nfit/core.c

@@ -454,8 +454,13 @@ int acpi_nfit_ctl(struct nvdimm_bus_descriptor *nd_desc, struct nvdimm *nvdimm,
 	if (cmd_rc)
 		*cmd_rc = -EINVAL;
 
-	if (cmd == ND_CMD_CALL)
+	if (cmd == ND_CMD_CALL) {
+		if (!buf || buf_len < sizeof(*call_pkg))
+			return -EINVAL;
+
 		call_pkg = buf;
+	}
+
 	func = cmd_to_func(nfit_mem, cmd, call_pkg, &family);
 	if (func < 0)
 		return func;

drivers/acpi/resource.c

@@ -250,6 +250,9 @@ static bool acpi_decode_space(struct resource_win *win,
 	switch (addr->resource_type) {
 	case ACPI_MEMORY_RANGE:
 		acpi_dev_memresource_flags(res, len, wp);
+
+		if (addr->info.mem.caching == ACPI_PREFETCHABLE_MEMORY)
+			res->flags |= IORESOURCE_PREFETCH;
 		break;
 	case ACPI_IO_RANGE:
 		acpi_dev_ioresource_flags(res, len, iodec,
@@ -265,9 +268,6 @@ static bool acpi_decode_space(struct resource_win *win,
 	if (addr->producer_consumer == ACPI_PRODUCER)
 		res->flags |= IORESOURCE_WINDOW;
 
-	if (addr->info.mem.caching == ACPI_PREFETCHABLE_MEMORY)
-		res->flags |= IORESOURCE_PREFETCH;
-
 	return !(res->flags & IORESOURCE_DISABLED);
 }
 

drivers/ata/sata_highbank.c

@@ -348,6 +348,7 @@ static int highbank_initialize_phys(struct device *dev, void __iomem *addr)
 			phy_nodes[phy] = phy_data.np;
 			cphy_base[phy] = of_iomap(phy_nodes[phy], 0);
 			if (cphy_base[phy] == NULL) {
+				of_node_put(phy_data.np);
 				return 0;
 			}
 			phy_count += 1;

drivers/clocksource/hyperv_timer.c

@@ -27,7 +27,8 @@
 #include <asm/mshyperv.h>
 
 static struct clock_event_device __percpu *hv_clock_event;
-static u64 hv_sched_clock_offset __ro_after_init;
+/* Note: offset can hold negative values after hibernation. */
+static u64 hv_sched_clock_offset __read_mostly;
 
 /*
  * If false, we're using the old mechanism for stimer0 interrupts
@@ -470,6 +471,17 @@ static void resume_hv_clock_tsc(struct clocksource *arg)
 	hv_set_msr(HV_MSR_REFERENCE_TSC, tsc_msr.as_uint64);
 }
 
+/*
+ * Called during resume from hibernation, from overridden
+ * x86_platform.restore_sched_clock_state routine. This is to adjust offsets
+ * used to calculate time for hv tsc page based sched_clock, to account for
+ * time spent before hibernation.
+ */
+void hv_adj_sched_clock_offset(u64 offset)
+{
+	hv_sched_clock_offset -= offset;
+}
+
 #ifdef HAVE_VDSO_CLOCKMODE_HVCLOCK
 static int hv_cs_enable(struct clocksource *cs)
 {

drivers/crypto/hisilicon/debugfs.c

@@ -192,7 +192,7 @@ static int qm_sqc_dump(struct hisi_qm *qm, char *s, char *name)
 
 	down_read(&qm->qps_lock);
 	if (qm->sqc) {
-		memcpy(&sqc, qm->sqc + qp_id * sizeof(struct qm_sqc), sizeof(struct qm_sqc));
+		memcpy(&sqc, qm->sqc + qp_id, sizeof(struct qm_sqc));
 		sqc.base_h = cpu_to_le32(QM_XQC_ADDR_MASK);
 		sqc.base_l = cpu_to_le32(QM_XQC_ADDR_MASK);
 		dump_show(qm, &sqc, sizeof(struct qm_sqc), "SOFT SQC");
@@ -229,7 +229,7 @@ static int qm_cqc_dump(struct hisi_qm *qm, char *s, char *name)
 
 	down_read(&qm->qps_lock);
 	if (qm->cqc) {
-		memcpy(&cqc, qm->cqc + qp_id * sizeof(struct qm_cqc), sizeof(struct qm_cqc));
+		memcpy(&cqc, qm->cqc + qp_id, sizeof(struct qm_cqc));
 		cqc.base_h = cpu_to_le32(QM_XQC_ADDR_MASK);
 		cqc.base_l = cpu_to_le32(QM_XQC_ADDR_MASK);
 		dump_show(qm, &cqc, sizeof(struct qm_cqc), "SOFT CQC");

drivers/cxl/core/region.c

@@ -1295,6 +1295,7 @@ static int cxl_port_setup_targets(struct cxl_port *port,
 	struct cxl_region_params *p = &cxlr->params;
 	struct cxl_decoder *cxld = cxl_rr->decoder;
 	struct cxl_switch_decoder *cxlsd;
+	struct cxl_port *iter = port;
 	u16 eig, peig;
 	u8 eiw, peiw;
 
@@ -1311,16 +1312,26 @@ static int cxl_port_setup_targets(struct cxl_port *port,
 
 	cxlsd = to_cxl_switch_decoder(&cxld->dev);
 	if (cxl_rr->nr_targets_set) {
-		int i, distance;
+		int i, distance = 1;
+		struct cxl_region_ref *cxl_rr_iter;
 
 		/*
-		 * Passthrough decoders impose no distance requirements between
+		 * The "distance" between peer downstream ports represents which
-		 * peers
+		 * endpoint positions in the region interleave a given port can
+		 * host.
+		 *
+		 * For example, at the root of a hierarchy the distance is
+		 * always 1 as every index targets a different host-bridge. At
+		 * each subsequent switch level those ports map every Nth region
+		 * position where N is the width of the switch == distance.
 		 */
-		if (cxl_rr->nr_targets == 1)
+		do {
-			distance = 0;
+			cxl_rr_iter = cxl_rr_load(iter, cxlr);
-		else
+			distance *= cxl_rr_iter->nr_targets;
-			distance = p->nr_targets / cxl_rr->nr_targets;
+			iter = to_cxl_port(iter->dev.parent);
+		} while (!is_cxl_root(iter));
+		distance *= cxlrd->cxlsd.cxld.interleave_ways;
+
 		for (i = 0; i < cxl_rr->nr_targets_set; i++)
 			if (ep->dport == cxlsd->target[i]) {
 				rc = check_last_peer(cxled, ep, cxl_rr,

drivers/cxl/pci.c

@@ -836,6 +836,9 @@ static ssize_t rcd_pcie_cap_emit(struct device *dev, u16 offset, char *buf, size
 	if (!root_dev)
 		return -ENXIO;
 
+	if (!dport->regs.rcd_pcie_cap)
+		return -ENXIO;
+
 	guard(device)(root_dev);
 	if (!root_dev->driver)
 		return -ENXIO;
@@ -1032,8 +1035,7 @@ static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (rc)
 		return rc;
 
-	rc = cxl_pci_ras_unmask(pdev);
+	if (cxl_pci_ras_unmask(pdev))
-	if (rc)
 		dev_dbg(&pdev->dev, "No RAS reporting unmasked\n");
 
 	pci_save_state(pdev);

drivers/edac/amd64_edac.c

@@ -3362,36 +3362,24 @@ static bool dct_ecc_enabled(struct amd64_pvt *pvt)
 
 static bool umc_ecc_enabled(struct amd64_pvt *pvt)
 {
-	u8 umc_en_mask = 0, ecc_en_mask = 0;
-	u16 nid = pvt->mc_node_id;
 	struct amd64_umc *umc;
-	u8 ecc_en = 0, i;
+	bool ecc_en = false;
+	int i;
 
+	/* Check whether at least one UMC is enabled: */
 	for_each_umc(i) {
 		umc = &pvt->umc[i];
 
-		/* Only check enabled UMCs. */
+		if (umc->sdp_ctrl & UMC_SDP_INIT &&
-		if (!(umc->sdp_ctrl & UMC_SDP_INIT))
+		    umc->umc_cap_hi & UMC_ECC_ENABLED) {
-			continue;
+			ecc_en = true;
-
+			break;
-		umc_en_mask |= BIT(i);
+		}
-
-		if (umc->umc_cap_hi & UMC_ECC_ENABLED)
-			ecc_en_mask |= BIT(i);
 	}
 
-	/* Check whether at least one UMC is enabled: */
+	edac_dbg(3, "Node %d: DRAM ECC %s.\n", pvt->mc_node_id, (ecc_en ? "enabled" : "disabled"));
-	if (umc_en_mask)
-		ecc_en = umc_en_mask == ecc_en_mask;
-	else
-		edac_dbg(0, "Node %d: No enabled UMCs.\n", nid);
 
-	edac_dbg(3, "Node %d: DRAM ECC %s.\n", nid, (ecc_en ? "enabled" : "disabled"));
+	return ecc_en;
-
-	if (!ecc_en)
-		return false;
-	else
-		return true;
 }
 
 static inline void

drivers/firmware/arm_ffa/bus.c

@@ -187,13 +187,18 @@ bool ffa_device_is_valid(struct ffa_device *ffa_dev)
 	return valid;
 }
 
-struct ffa_device *ffa_device_register(const uuid_t *uuid, int vm_id,
+struct ffa_device *
-				       const struct ffa_ops *ops)
+ffa_device_register(const struct ffa_partition_info *part_info,
+		    const struct ffa_ops *ops)
 {
 	int id, ret;
+	uuid_t uuid;
 	struct device *dev;
 	struct ffa_device *ffa_dev;
 
+	if (!part_info)
+		return NULL;
+
 	id = ida_alloc_min(&ffa_bus_id, 1, GFP_KERNEL);
 	if (id < 0)
 		return NULL;
@@ -210,9 +215,11 @@ struct ffa_device *ffa_device_register(const uuid_t *uuid, int vm_id,
 	dev_set_name(&ffa_dev->dev, "arm-ffa-%d", id);
 
 	ffa_dev->id = id;
-	ffa_dev->vm_id = vm_id;
+	ffa_dev->vm_id = part_info->id;
+	ffa_dev->properties = part_info->properties;
 	ffa_dev->ops = ops;
-	uuid_copy(&ffa_dev->uuid, uuid);
+	import_uuid(&uuid, (u8 *)part_info->uuid);
+	uuid_copy(&ffa_dev->uuid, &uuid);
 
 	ret = device_register(&ffa_dev->dev);
 	if (ret) {

drivers/firmware/arm_ffa/driver.c

@@ -1387,7 +1387,6 @@ static struct notifier_block ffa_bus_nb = {
 static int ffa_setup_partitions(void)
 {
 	int count, idx, ret;
-	uuid_t uuid;
 	struct ffa_device *ffa_dev;
 	struct ffa_dev_part_info *info;
 	struct ffa_partition_info *pbuf, *tpbuf;
@@ -1406,23 +1405,19 @@ static int ffa_setup_partitions(void)
 
 	xa_init(&drv_info->partition_info);
 	for (idx = 0, tpbuf = pbuf; idx < count; idx++, tpbuf++) {
-		import_uuid(&uuid, (u8 *)tpbuf->uuid);
-
 		/* Note that if the UUID will be uuid_null, that will require
 		 * ffa_bus_notifier() to find the UUID of this partition id
 		 * with help of ffa_device_match_uuid(). FF-A v1.1 and above
 		 * provides UUID here for each partition as part of the
 		 * discovery API and the same is passed.
 		 */
-		ffa_dev = ffa_device_register(&uuid, tpbuf->id, &ffa_drv_ops);
+		ffa_dev = ffa_device_register(tpbuf, &ffa_drv_ops);
 		if (!ffa_dev) {
 			pr_err("%s: failed to register partition ID 0x%x\n",
 			       __func__, tpbuf->id);
 			continue;
 		}
 
-		ffa_dev->properties = tpbuf->properties;
-
 		if (drv_info->version > FFA_VERSION_1_0 &&
 		    !(tpbuf->properties & FFA_PARTITION_AARCH64_EXEC))
 			ffa_mode_32bit_set(ffa_dev);

drivers/firmware/arm_scmi/vendors/imx/Kconfig

@@ -15,6 +15,7 @@ config IMX_SCMI_BBM_EXT
 config IMX_SCMI_MISC_EXT
 	tristate "i.MX SCMI MISC EXTENSION"
 	depends on ARM_SCMI_PROTOCOL || (COMPILE_TEST && OF)
+	depends on IMX_SCMI_MISC_DRV
 	default y if ARCH_MXC
 	help
 	  This enables i.MX System MISC control logic such as gpio expander

drivers/firmware/efi/Kconfig

@@ -76,10 +76,6 @@ config EFI_ZBOOT
 	bool "Enable the generic EFI decompressor"
 	depends on EFI_GENERIC_STUB && !ARM
 	select HAVE_KERNEL_GZIP
-	select HAVE_KERNEL_LZ4
-	select HAVE_KERNEL_LZMA
-	select HAVE_KERNEL_LZO
-	select HAVE_KERNEL_XZ
 	select HAVE_KERNEL_ZSTD
 	help
 	  Create the bootable image as an EFI application that carries the

drivers/firmware/efi/esrt.c

@@ -75,8 +75,6 @@ static LIST_HEAD(entry_list);
 struct esre_attribute {
 	struct attribute attr;
 	ssize_t (*show)(struct esre_entry *entry, char *buf);
-	ssize_t (*store)(struct esre_entry *entry,
-			 const char *buf, size_t count);
 };
 
 static struct esre_entry *to_entry(struct kobject *kobj)

drivers/firmware/efi/libstub/Makefile.zboot

@@ -12,22 +12,16 @@ quiet_cmd_copy_and_pad = PAD     $@
 $(obj)/vmlinux.bin: $(obj)/$(EFI_ZBOOT_PAYLOAD) FORCE
 	$(call if_changed,copy_and_pad)
 
-comp-type-$(CONFIG_KERNEL_GZIP)		:= gzip
-comp-type-$(CONFIG_KERNEL_LZ4)		:= lz4
-comp-type-$(CONFIG_KERNEL_LZMA)		:= lzma
-comp-type-$(CONFIG_KERNEL_LZO)		:= lzo
-comp-type-$(CONFIG_KERNEL_XZ)		:= xzkern
-comp-type-$(CONFIG_KERNEL_ZSTD)		:= zstd22
-
 # in GZIP, the appended le32 carrying the uncompressed size is part of the
 # format, but in other cases, we just append it at the end for convenience,
 # causing the original tools to complain when checking image integrity.
-# So disregard it when calculating the payload size in the zimage header.
+comp-type-y				:= gzip
-zboot-method-y                         := $(comp-type-y)_with_size
+zboot-method-y				:= gzip
-zboot-size-len-y                       := 4
+zboot-size-len-y			:= 0
 
-zboot-method-$(CONFIG_KERNEL_GZIP)     := gzip
+comp-type-$(CONFIG_KERNEL_ZSTD)		:= zstd
-zboot-size-len-$(CONFIG_KERNEL_GZIP)   := 0
+zboot-method-$(CONFIG_KERNEL_ZSTD)	:= zstd22_with_size
+zboot-size-len-$(CONFIG_KERNEL_ZSTD)	:= 4
 
 $(obj)/vmlinuz: $(obj)/vmlinux.bin FORCE
 	$(call if_changed,$(zboot-method-y))

drivers/firmware/imx/Kconfig

@@ -25,7 +25,6 @@ config IMX_SCU
 
 config IMX_SCMI_MISC_DRV
 	tristate "IMX SCMI MISC Protocol driver"
-	depends on IMX_SCMI_MISC_EXT || COMPILE_TEST
 	default y if ARCH_MXC
 	help
 	  The System Controller Management Interface firmware (SCMI FW) is

drivers/gpio/Kconfig

@@ -482,8 +482,9 @@ config GPIO_MT7621
 	  Say yes here to support the Mediatek MT7621 SoC GPIO device.
 
 config GPIO_MVEBU
-	def_bool y
+	bool "Marvell Orion and EBU GPIO support" if COMPILE_TEST
 	depends on PLAT_ORION || ARCH_MVEBU || COMPILE_TEST
+	default PLAT_ORION || ARCH_MVEBU
 	select GENERIC_IRQ_CHIP
 	select REGMAP_MMIO
 

drivers/gpio/gpio-graniterapids.c

@@ -32,12 +32,14 @@
 #define GNR_PINS_PER_REG 32
 #define GNR_NUM_REGS DIV_ROUND_UP(GNR_NUM_PINS, GNR_PINS_PER_REG)
 
-#define GNR_CFG_BAR		0x00
+#define GNR_CFG_PADBAR		0x00
 #define GNR_CFG_LOCK_OFFSET	0x04
-#define GNR_GPI_STATUS_OFFSET	0x20
+#define GNR_GPI_STATUS_OFFSET	0x14
 #define GNR_GPI_ENABLE_OFFSET	0x24
 
-#define GNR_CFG_DW_RX_MASK	GENMASK(25, 22)
+#define GNR_CFG_DW_HOSTSW_MODE	BIT(27)
+#define GNR_CFG_DW_RX_MASK	GENMASK(23, 22)
+#define GNR_CFG_DW_INTSEL_MASK	GENMASK(21, 14)
 #define GNR_CFG_DW_RX_DISABLE	FIELD_PREP(GNR_CFG_DW_RX_MASK, 2)
 #define GNR_CFG_DW_RX_EDGE	FIELD_PREP(GNR_CFG_DW_RX_MASK, 1)
 #define GNR_CFG_DW_RX_LEVEL	FIELD_PREP(GNR_CFG_DW_RX_MASK, 0)
@@ -50,6 +52,7 @@
  * struct gnr_gpio - Intel Granite Rapids-D vGPIO driver state
  * @gc: GPIO controller interface
  * @reg_base: base address of the GPIO registers
+ * @pad_base: base address of the vGPIO pad configuration registers
  * @ro_bitmap: bitmap of read-only pins
  * @lock: guard the registers
  * @pad_backup: backup of the register state for suspend
@@ -57,6 +60,7 @@
 struct gnr_gpio {
 	struct gpio_chip gc;
 	void __iomem *reg_base;
+	void __iomem *pad_base;
 	DECLARE_BITMAP(ro_bitmap, GNR_NUM_PINS);
 	raw_spinlock_t lock;
 	u32 pad_backup[];
@@ -65,7 +69,7 @@ struct gnr_gpio {
 static void __iomem *gnr_gpio_get_padcfg_addr(const struct gnr_gpio *priv,
 					      unsigned int gpio)
 {
-	return priv->reg_base + gpio * sizeof(u32);
+	return priv->pad_base + gpio * sizeof(u32);
 }
 
 static int gnr_gpio_configure_line(struct gpio_chip *gc, unsigned int gpio,
@@ -88,6 +92,20 @@ static int gnr_gpio_configure_line(struct gpio_chip *gc, unsigned int gpio,
 	return 0;
 }
 
+static int gnr_gpio_request(struct gpio_chip *gc, unsigned int gpio)
+{
+	struct gnr_gpio *priv = gpiochip_get_data(gc);
+	u32 dw;
+
+	dw = readl(gnr_gpio_get_padcfg_addr(priv, gpio));
+	if (!(dw & GNR_CFG_DW_HOSTSW_MODE)) {
+		dev_warn(gc->parent, "GPIO %u is not owned by host", gpio);
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
 static int gnr_gpio_get(struct gpio_chip *gc, unsigned int gpio)
 {
 	const struct gnr_gpio *priv = gpiochip_get_data(gc);
@@ -139,6 +157,7 @@ static int gnr_gpio_direction_output(struct gpio_chip *gc, unsigned int gpio, in
 
 static const struct gpio_chip gnr_gpio_chip = {
 	.owner		  = THIS_MODULE,
+	.request	  = gnr_gpio_request,
 	.get		  = gnr_gpio_get,
 	.set		  = gnr_gpio_set,
 	.get_direction    = gnr_gpio_get_direction,
@@ -166,7 +185,7 @@ static void gnr_gpio_irq_ack(struct irq_data *d)
 	guard(raw_spinlock_irqsave)(&priv->lock);
 
 	reg = readl(addr);
-	reg &= ~BIT(bit_idx);
+	reg |= BIT(bit_idx);
 	writel(reg, addr);
 }
 
@@ -209,10 +228,18 @@ static void gnr_gpio_irq_unmask(struct irq_data *d)
 static int gnr_gpio_irq_set_type(struct irq_data *d, unsigned int type)
 {
 	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
-	irq_hw_number_t pin = irqd_to_hwirq(d);
+	struct gnr_gpio *priv = gpiochip_get_data(gc);
-	u32 mask = GNR_CFG_DW_RX_MASK;
+	irq_hw_number_t hwirq = irqd_to_hwirq(d);
+	u32 reg;
 	u32 set;
 
+	/* Allow interrupts only if Interrupt Select field is non-zero */
+	reg = readl(gnr_gpio_get_padcfg_addr(priv, hwirq));
+	if (!(reg & GNR_CFG_DW_INTSEL_MASK)) {
+		dev_dbg(gc->parent, "GPIO %lu cannot be used as IRQ", hwirq);
+		return -EPERM;
+	}
+
 	/* Falling edge and level low triggers not supported by the GPIO controller */
 	switch (type) {
 	case IRQ_TYPE_NONE:
@@ -230,10 +257,11 @@ static int gnr_gpio_irq_set_type(struct irq_data *d, unsigned int type)
 		return -EINVAL;
 	}
 
-	return gnr_gpio_configure_line(gc, pin, mask, set);
+	return gnr_gpio_configure_line(gc, hwirq, GNR_CFG_DW_RX_MASK, set);
 }
 
 static const struct irq_chip gnr_gpio_irq_chip = {
+	.name		= "gpio-graniterapids",
 	.irq_ack	= gnr_gpio_irq_ack,
 	.irq_mask	= gnr_gpio_irq_mask,
 	.irq_unmask	= gnr_gpio_irq_unmask,
@@ -291,6 +319,7 @@ static int gnr_gpio_probe(struct platform_device *pdev)
 	struct gnr_gpio *priv;
 	void __iomem *regs;
 	int irq, ret;
+	u32 offset;
 
 	priv = devm_kzalloc(dev, struct_size(priv, pad_backup, num_backup_pins), GFP_KERNEL);
 	if (!priv)
@@ -302,6 +331,10 @@ static int gnr_gpio_probe(struct platform_device *pdev)
 	if (IS_ERR(regs))
 		return PTR_ERR(regs);
 
+	priv->reg_base = regs;
+	offset = readl(priv->reg_base + GNR_CFG_PADBAR);
+	priv->pad_base = priv->reg_base + offset;
+
 	irq = platform_get_irq(pdev, 0);
 	if (irq < 0)
 		return irq;
@@ -311,8 +344,6 @@ static int gnr_gpio_probe(struct platform_device *pdev)
 	if (ret)
 		return dev_err_probe(dev, ret, "failed to request interrupt\n");
 
-	priv->reg_base = regs + readl(regs + GNR_CFG_BAR);
-
 	gnr_gpio_init_pin_ro_bits(dev, priv->reg_base + GNR_CFG_LOCK_OFFSET,
 				  priv->ro_bitmap);
 
@@ -324,7 +355,6 @@ static int gnr_gpio_probe(struct platform_device *pdev)
 
 	girq = &priv->gc.irq;
 	gpio_irq_chip_set_chip(girq, &gnr_gpio_irq_chip);
-	girq->chip->name	= dev_name(dev);
 	girq->parent_handler	= NULL;
 	girq->num_parents	= 0;
 	girq->parents		= NULL;

drivers/gpio/gpio-idio-16.c

@@ -3,6 +3,9 @@
  * GPIO library for the ACCES IDIO-16 family
  * Copyright (C) 2022 William Breathitt Gray
  */
+
+#define DEFAULT_SYMBOL_NAMESPACE "GPIO_IDIO_16"
+
 #include <linux/bits.h>
 #include <linux/device.h>
 #include <linux/err.h>
@@ -14,8 +17,6 @@
 
 #include "gpio-idio-16.h"
 
-#define DEFAULT_SYMBOL_NAMESPACE "GPIO_IDIO_16"
-
 #define IDIO_16_DAT_BASE 0x0
 #define IDIO_16_OUT_BASE IDIO_16_DAT_BASE
 #define IDIO_16_IN_BASE (IDIO_16_DAT_BASE + 1)

drivers/gpio/gpio-ljca.c

@@ -82,9 +82,9 @@ static int ljca_gpio_config(struct ljca_gpio_dev *ljca_gpio, u8 gpio_id,
 	int ret;
 
 	mutex_lock(&ljca_gpio->trans_lock);
+	packet->num = 1;
 	packet->item[0].index = gpio_id;
 	packet->item[0].value = config | ljca_gpio->connect_mode[gpio_id];
-	packet->num = 1;
 
 	ret = ljca_transfer(ljca_gpio->ljca, LJCA_GPIO_CONFIG, (u8 *)packet,
 			    struct_size(packet, item, packet->num), NULL, 0);

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

@@ -1801,13 +1801,18 @@ int amdgpu_cs_find_mapping(struct amdgpu_cs_parser *parser,
 	if (dma_resv_locking_ctx((*bo)->tbo.base.resv) != &parser->exec.ticket)
 		return -EINVAL;
 
+	/* Make sure VRAM is allocated contigiously */
 	(*bo)->flags |= AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS;
-	amdgpu_bo_placement_from_domain(*bo, (*bo)->allowed_domains);
+	if ((*bo)->tbo.resource->mem_type == TTM_PL_VRAM &&
-	for (i = 0; i < (*bo)->placement.num_placement; i++)
+	    !((*bo)->tbo.resource->placement & TTM_PL_FLAG_CONTIGUOUS)) {
-		(*bo)->placements[i].flags |= TTM_PL_FLAG_CONTIGUOUS;
+
-	r = ttm_bo_validate(&(*bo)->tbo, &(*bo)->placement, &ctx);
+		amdgpu_bo_placement_from_domain(*bo, (*bo)->allowed_domains);
-	if (r)
+		for (i = 0; i < (*bo)->placement.num_placement; i++)
-		return r;
+			(*bo)->placements[i].flags |= TTM_PL_FLAG_CONTIGUOUS;
+		r = ttm_bo_validate(&(*bo)->tbo, &(*bo)->placement, &ctx);
+		if (r)
+			return r;
+	}
 
 	return amdgpu_ttm_alloc_gart(&(*bo)->tbo);
 }

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

@@ -145,7 +145,7 @@ const char *amdgpu_asic_name[] = {
 	"LAST",
 };
 
-#define AMDGPU_IP_BLK_MASK_ALL GENMASK(AMDGPU_MAX_IP_NUM, 0)
+#define AMDGPU_IP_BLK_MASK_ALL GENMASK(AMD_IP_BLOCK_TYPE_NUM  - 1, 0)
 /*
  * Default init level where all blocks are expected to be initialized. This is
  * the level of initialization expected by default and also after a full reset

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

@@ -551,6 +551,8 @@ static void amdgpu_uvd_force_into_uvd_segment(struct amdgpu_bo *abo)
 	for (i = 0; i < abo->placement.num_placement; ++i) {
 		abo->placements[i].fpfn = 0 >> PAGE_SHIFT;
 		abo->placements[i].lpfn = (256 * 1024 * 1024) >> PAGE_SHIFT;
+		if (abo->placements[i].mem_type == TTM_PL_VRAM)
+			abo->placements[i].flags |= TTM_PL_FLAG_CONTIGUOUS;
 	}
 }
 

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

@@ -674,12 +674,8 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job,
 	pasid_mapping_needed &= adev->gmc.gmc_funcs->emit_pasid_mapping &&
 		ring->funcs->emit_wreg;
 
-	if (adev->gfx.enable_cleaner_shader &&
+	if (!vm_flush_needed && !gds_switch_needed && !need_pipe_sync &&
-	    ring->funcs->emit_cleaner_shader &&
+	    !(job->enforce_isolation && !job->vmid))
-	    job->enforce_isolation)
-		ring->funcs->emit_cleaner_shader(ring);
-
-	if (!vm_flush_needed && !gds_switch_needed && !need_pipe_sync)
 		return 0;
 
 	amdgpu_ring_ib_begin(ring);
@@ -690,6 +686,11 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job,
 	if (need_pipe_sync)
 		amdgpu_ring_emit_pipeline_sync(ring);
 
+	if (adev->gfx.enable_cleaner_shader &&
+	    ring->funcs->emit_cleaner_shader &&
+	    job->enforce_isolation)
+		ring->funcs->emit_cleaner_shader(ring);
+
 	if (vm_flush_needed) {
 		trace_amdgpu_vm_flush(ring, job->vmid, job->vm_pd_addr);
 		amdgpu_ring_emit_vm_flush(ring, job->vmid, job->vm_pd_addr);

drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c

@@ -45,6 +45,8 @@ MODULE_FIRMWARE("amdgpu/gc_9_4_3_mec.bin");
 MODULE_FIRMWARE("amdgpu/gc_9_4_4_mec.bin");
 MODULE_FIRMWARE("amdgpu/gc_9_4_3_rlc.bin");
 MODULE_FIRMWARE("amdgpu/gc_9_4_4_rlc.bin");
+MODULE_FIRMWARE("amdgpu/gc_9_4_3_sjt_mec.bin");
+MODULE_FIRMWARE("amdgpu/gc_9_4_4_sjt_mec.bin");
 
 #define GFX9_MEC_HPD_SIZE 4096
 #define RLCG_UCODE_LOADING_START_ADDRESS 0x00002000L
@@ -574,8 +576,12 @@ static int gfx_v9_4_3_init_cp_compute_microcode(struct amdgpu_device *adev,
 {
 	int err;
 
-	err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw,
+	if (amdgpu_sriov_vf(adev))
-				   "amdgpu/%s_mec.bin", chip_name);
+		err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw,
+				"amdgpu/%s_sjt_mec.bin", chip_name);
+	else
+		err = amdgpu_ucode_request(adev, &adev->gfx.mec_fw,
+				"amdgpu/%s_mec.bin", chip_name);
 	if (err)
 		goto out;
 	amdgpu_gfx_cp_init_microcode(adev, AMDGPU_UCODE_ID_CP_MEC1);

drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c

@@ -1288,7 +1288,7 @@ static int uvd_v7_0_ring_patch_cs_in_place(struct amdgpu_cs_parser *p,
 					   struct amdgpu_job *job,
 					   struct amdgpu_ib *ib)
 {
-	struct amdgpu_ring *ring = to_amdgpu_ring(job->base.sched);
+	struct amdgpu_ring *ring = amdgpu_job_ring(job);
 	unsigned i;
 
 	/* No patching necessary for the first instance */

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

@@ -1423,6 +1423,7 @@ int kfd_parse_crat_table(void *crat_image, struct list_head *device_list,
 
 
 static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
+						   bool cache_line_size_missing,
 						   struct kfd_gpu_cache_info *pcache_info)
 {
 	struct amdgpu_device *adev = kdev->adev;
@@ -1437,6 +1438,8 @@ static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
 					CRAT_CACHE_FLAGS_SIMD_CACHE);
 		pcache_info[i].num_cu_shared = adev->gfx.config.gc_num_tcp_per_wpg / 2;
 		pcache_info[i].cache_line_size = adev->gfx.config.gc_tcp_cache_line_size;
+		if (cache_line_size_missing && !pcache_info[i].cache_line_size)
+			pcache_info[i].cache_line_size = 128;
 		i++;
 	}
 	/* Scalar L1 Instruction Cache per SQC */
@@ -1449,6 +1452,8 @@ static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
 					CRAT_CACHE_FLAGS_SIMD_CACHE);
 		pcache_info[i].num_cu_shared = adev->gfx.config.gc_num_sqc_per_wgp * 2;
 		pcache_info[i].cache_line_size = adev->gfx.config.gc_instruction_cache_line_size;
+		if (cache_line_size_missing && !pcache_info[i].cache_line_size)
+			pcache_info[i].cache_line_size = 128;
 		i++;
 	}
 	/* Scalar L1 Data Cache per SQC */
@@ -1460,6 +1465,8 @@ static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
 					CRAT_CACHE_FLAGS_SIMD_CACHE);
 		pcache_info[i].num_cu_shared = adev->gfx.config.gc_num_sqc_per_wgp * 2;
 		pcache_info[i].cache_line_size = adev->gfx.config.gc_scalar_data_cache_line_size;
+		if (cache_line_size_missing && !pcache_info[i].cache_line_size)
+			pcache_info[i].cache_line_size = 64;
 		i++;
 	}
 	/* GL1 Data Cache per SA */
@@ -1472,7 +1479,8 @@ static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
 					CRAT_CACHE_FLAGS_DATA_CACHE |
 					CRAT_CACHE_FLAGS_SIMD_CACHE);
 		pcache_info[i].num_cu_shared = adev->gfx.config.max_cu_per_sh;
-		pcache_info[i].cache_line_size = 0;
+		if (cache_line_size_missing)
+			pcache_info[i].cache_line_size = 128;
 		i++;
 	}
 	/* L2 Data Cache per GPU (Total Tex Cache) */
@@ -1484,6 +1492,8 @@ static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
 					CRAT_CACHE_FLAGS_SIMD_CACHE);
 		pcache_info[i].num_cu_shared = adev->gfx.config.max_cu_per_sh;
 		pcache_info[i].cache_line_size = adev->gfx.config.gc_tcc_cache_line_size;
+		if (cache_line_size_missing && !pcache_info[i].cache_line_size)
+			pcache_info[i].cache_line_size = 128;
 		i++;
 	}
 	/* L3 Data Cache per GPU */
@@ -1494,7 +1504,7 @@ static int kfd_fill_gpu_cache_info_from_gfx_config(struct kfd_dev *kdev,
 					CRAT_CACHE_FLAGS_DATA_CACHE |
 					CRAT_CACHE_FLAGS_SIMD_CACHE);
 		pcache_info[i].num_cu_shared = adev->gfx.config.max_cu_per_sh;
-		pcache_info[i].cache_line_size = 0;
+		pcache_info[i].cache_line_size = 64;
 		i++;
 	}
 	return i;
@@ -1569,6 +1579,7 @@ static int kfd_fill_gpu_cache_info_from_gfx_config_v2(struct kfd_dev *kdev,
 int kfd_get_gpu_cache_info(struct kfd_node *kdev, struct kfd_gpu_cache_info **pcache_info)
 {
 	int num_of_cache_types = 0;
+	bool cache_line_size_missing = false;
 
 	switch (kdev->adev->asic_type) {
 	case CHIP_KAVERI:
@@ -1692,10 +1703,17 @@ int kfd_get_gpu_cache_info(struct kfd_node *kdev, struct kfd_gpu_cache_info **pc
 		case IP_VERSION(11, 5, 0):
 		case IP_VERSION(11, 5, 1):
 		case IP_VERSION(11, 5, 2):
+			/* Cacheline size not available in IP discovery for gc11.
+			 * kfd_fill_gpu_cache_info_from_gfx_config to hard code it
+			 */
+			cache_line_size_missing = true;
+			fallthrough;
 		case IP_VERSION(12, 0, 0):
 		case IP_VERSION(12, 0, 1):
 			num_of_cache_types =
-				kfd_fill_gpu_cache_info_from_gfx_config(kdev->kfd, *pcache_info);
+				kfd_fill_gpu_cache_info_from_gfx_config(kdev->kfd,
+									cache_line_size_missing,
+									*pcache_info);
 			break;
 		default:
 			*pcache_info = dummy_cache_info;

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

@@ -207,6 +207,21 @@ static int add_queue_mes(struct device_queue_manager *dqm, struct queue *q,
 	if (!down_read_trylock(&adev->reset_domain->sem))
 		return -EIO;
 
+	if (!pdd->proc_ctx_cpu_ptr) {
+		r = amdgpu_amdkfd_alloc_gtt_mem(adev,
+				AMDGPU_MES_PROC_CTX_SIZE,
+				&pdd->proc_ctx_bo,
+				&pdd->proc_ctx_gpu_addr,
+				&pdd->proc_ctx_cpu_ptr,
+				false);
+		if (r) {
+			dev_err(adev->dev,
+				"failed to allocate process context bo\n");
+			return r;
+		}
+		memset(pdd->proc_ctx_cpu_ptr, 0, AMDGPU_MES_PROC_CTX_SIZE);
+	}
+
 	memset(&queue_input, 0x0, sizeof(struct mes_add_queue_input));
 	queue_input.process_id = qpd->pqm->process->pasid;
 	queue_input.page_table_base_addr =  qpd->page_table_base;

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

@@ -306,7 +306,7 @@ svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange,
 		spage = migrate_pfn_to_page(migrate->src[i]);
 		if (spage && !is_zone_device_page(spage)) {
 			src[i] = dma_map_page(dev, spage, 0, PAGE_SIZE,
-					      DMA_TO_DEVICE);
+					      DMA_BIDIRECTIONAL);
 			r = dma_mapping_error(dev, src[i]);
 			if (r) {
 				dev_err(dev, "%s: fail %d dma_map_page\n",
@@ -629,7 +629,7 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 			goto out_oom;
 		}
 
-		dst[i] = dma_map_page(dev, dpage, 0, PAGE_SIZE, DMA_FROM_DEVICE);
+		dst[i] = dma_map_page(dev, dpage, 0, PAGE_SIZE, DMA_BIDIRECTIONAL);
 		r = dma_mapping_error(dev, dst[i]);
 		if (r) {
 			dev_err(adev->dev, "%s: fail %d dma_map_page\n", __func__, r);

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

@@ -1076,7 +1076,8 @@ static void kfd_process_destroy_pdds(struct kfd_process *p)
 
 		kfd_free_process_doorbells(pdd->dev->kfd, pdd);
 
-		if (pdd->dev->kfd->shared_resources.enable_mes)
+		if (pdd->dev->kfd->shared_resources.enable_mes &&
+			pdd->proc_ctx_cpu_ptr)
 			amdgpu_amdkfd_free_gtt_mem(pdd->dev->adev,
 						   &pdd->proc_ctx_bo);
 		/*
@@ -1608,7 +1609,6 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev,
 							struct kfd_process *p)
 {
 	struct kfd_process_device *pdd = NULL;
-	int retval = 0;
 
 	if (WARN_ON_ONCE(p->n_pdds >= MAX_GPU_INSTANCE))
 		return NULL;
@@ -1632,21 +1632,6 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev,
 	pdd->user_gpu_id = dev->id;
 	atomic64_set(&pdd->evict_duration_counter, 0);
 
-	if (dev->kfd->shared_resources.enable_mes) {
-		retval = amdgpu_amdkfd_alloc_gtt_mem(dev->adev,
-						AMDGPU_MES_PROC_CTX_SIZE,
-						&pdd->proc_ctx_bo,
-						&pdd->proc_ctx_gpu_addr,
-						&pdd->proc_ctx_cpu_ptr,
-						false);
-		if (retval) {
-			dev_err(dev->adev->dev,
-				"failed to allocate process context bo\n");
-			goto err_free_pdd;
-		}
-		memset(pdd->proc_ctx_cpu_ptr, 0, AMDGPU_MES_PROC_CTX_SIZE);
-	}
-
 	p->pdds[p->n_pdds++] = pdd;
 	if (kfd_dbg_is_per_vmid_supported(pdd->dev))
 		pdd->spi_dbg_override = pdd->dev->kfd2kgd->disable_debug_trap(
@@ -1658,10 +1643,6 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev,
 	idr_init(&pdd->alloc_idr);
 
 	return pdd;
-
-err_free_pdd:
-	kfree(pdd);
-	return NULL;
 }
 
 /**

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

@@ -212,13 +212,17 @@ static void pqm_clean_queue_resource(struct process_queue_manager *pqm,
 void pqm_uninit(struct process_queue_manager *pqm)
 {
 	struct process_queue_node *pqn, *next;
-	struct kfd_process_device *pdd;
 
 	list_for_each_entry_safe(pqn, next, &pqm->queues, process_queue_list) {
 		if (pqn->q) {
-			pdd = kfd_get_process_device_data(pqn->q->device, pqm->process);
+			struct kfd_process_device *pdd = kfd_get_process_device_data(pqn->q->device,
-			kfd_queue_unref_bo_vas(pdd, &pqn->q->properties);
+										     pqm->process);
-			kfd_queue_release_buffers(pdd, &pqn->q->properties);
+			if (pdd) {
+				kfd_queue_unref_bo_vas(pdd, &pqn->q->properties);
+				kfd_queue_release_buffers(pdd, &pqn->q->properties);
+			} else {
+				WARN_ON(!pdd);
+			}
 			pqm_clean_queue_resource(pqm, pqn);
 		}
 

drivers/gpu/drm/amd/include/kgd_pp_interface.h

@@ -164,6 +164,7 @@ enum amd_pp_task {
 };
 
 enum PP_SMC_POWER_PROFILE {
+	PP_SMC_POWER_PROFILE_UNKNOWN = -1,
 	PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT = 0x0,
 	PP_SMC_POWER_PROFILE_FULLSCREEN3D = 0x1,
 	PP_SMC_POWER_PROFILE_POWERSAVING  = 0x2,

drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c

@@ -764,6 +764,7 @@ static int smu_early_init(struct amdgpu_ip_block *ip_block)
 	smu->smu_baco.platform_support = false;
 	smu->smu_baco.maco_support = false;
 	smu->user_dpm_profile.fan_mode = -1;
+	smu->power_profile_mode = PP_SMC_POWER_PROFILE_UNKNOWN;
 
 	mutex_init(&smu->message_lock);
 
@@ -1248,6 +1249,21 @@ static bool smu_is_workload_profile_available(struct smu_context *smu,
 	return smu->workload_map && smu->workload_map[profile].valid_mapping;
 }
 
+static void smu_init_power_profile(struct smu_context *smu)
+{
+	if (smu->power_profile_mode == PP_SMC_POWER_PROFILE_UNKNOWN) {
+		if (smu->is_apu ||
+		    !smu_is_workload_profile_available(
+			    smu, PP_SMC_POWER_PROFILE_FULLSCREEN3D))
+			smu->power_profile_mode =
+				PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
+		else
+			smu->power_profile_mode =
+				PP_SMC_POWER_PROFILE_FULLSCREEN3D;
+	}
+	smu_power_profile_mode_get(smu, smu->power_profile_mode);
+}
+
 static int smu_sw_init(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
@@ -1269,13 +1285,7 @@ static int smu_sw_init(struct amdgpu_ip_block *ip_block)
 	atomic_set(&smu->smu_power.power_gate.vpe_gated, 1);
 	atomic_set(&smu->smu_power.power_gate.umsch_mm_gated, 1);
 
-	if (smu->is_apu ||
+	smu_init_power_profile(smu);
-	    !smu_is_workload_profile_available(smu, PP_SMC_POWER_PROFILE_FULLSCREEN3D))
-		smu->power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
-	else
-		smu->power_profile_mode = PP_SMC_POWER_PROFILE_FULLSCREEN3D;
-	smu_power_profile_mode_get(smu, smu->power_profile_mode);
-
 	smu->display_config = &adev->pm.pm_display_cfg;
 
 	smu->smu_dpm.dpm_level = AMD_DPM_FORCED_LEVEL_AUTO;

drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c

@@ -2810,4 +2810,5 @@ void smu_v13_0_7_set_ppt_funcs(struct smu_context *smu)
 	smu->workload_map = smu_v13_0_7_workload_map;
 	smu->smc_driver_if_version = SMU13_0_7_DRIVER_IF_VERSION;
 	smu_v13_0_set_smu_mailbox_registers(smu);
+	smu->power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
 }