diff options
169 files changed, 1619 insertions, 1103 deletions
diff --git a/Documentation/devicetree/bindings/arm/gic-v3.txt b/Documentation/devicetree/bindings/arm/gic-v3.txt index ddfade40ac59..7803e77d85cb 100644 --- a/Documentation/devicetree/bindings/arm/gic-v3.txt +++ b/Documentation/devicetree/bindings/arm/gic-v3.txt @@ -57,6 +57,8 @@ used to route Message Signalled Interrupts (MSI) to the CPUs. These nodes must have the following properties: - compatible : Should at least contain "arm,gic-v3-its". - msi-controller : Boolean property. Identifies the node as an MSI controller +- #msi-cells: Must be <1>. The single msi-cell is the DeviceID of the device + which will generate the MSI. - reg: Specifies the base physical address and size of the ITS registers. @@ -83,6 +85,7 @@ Examples: gic-its@2c200000 { compatible = "arm,gic-v3-its"; msi-controller; + #msi-cells = <1>; reg = <0x0 0x2c200000 0 0x200000>; }; }; @@ -107,12 +110,14 @@ Examples: gic-its@2c200000 { compatible = "arm,gic-v3-its"; msi-controller; + #msi-cells = <1>; reg = <0x0 0x2c200000 0 0x200000>; }; gic-its@2c400000 { compatible = "arm,gic-v3-its"; msi-controller; + #msi-cells = <1>; reg = <0x0 0x2c400000 0 0x200000>; }; }; diff --git a/Documentation/devicetree/bindings/arm/idle-states.txt b/Documentation/devicetree/bindings/arm/idle-states.txt index a8274eabae2e..b8e41c148a3c 100644 --- a/Documentation/devicetree/bindings/arm/idle-states.txt +++ b/Documentation/devicetree/bindings/arm/idle-states.txt @@ -497,7 +497,7 @@ cpus { }; idle-states { - entry-method = "arm,psci"; + entry-method = "psci"; CPU_RETENTION_0_0: cpu-retention-0-0 { compatible = "arm,idle-state"; diff --git a/Documentation/devicetree/bindings/gpio/gpio.txt b/Documentation/devicetree/bindings/gpio/gpio.txt index 5788d5cf1252..82d40e2505f6 100644 --- a/Documentation/devicetree/bindings/gpio/gpio.txt +++ b/Documentation/devicetree/bindings/gpio/gpio.txt @@ -16,7 +16,9 @@ properties, each containing a 'gpio-list': GPIO properties should be named "[<name>-]gpios", with <name> being the purpose of this GPIO for the device. While a non-existent <name> is considered valid for compatibility reasons (resolving to the "gpios" property), it is not allowed -for new bindings. +for new bindings. Also, GPIO properties named "[<name>-]gpio" are valid and old +bindings use it, but are only supported for compatibility reasons and should not +be used for newer bindings since it has been deprecated. GPIO properties can contain one or more GPIO phandles, but only in exceptional cases should they contain more than one. If your device uses several GPIOs with diff --git a/Documentation/devicetree/bindings/iio/accel/bma180.txt b/Documentation/devicetree/bindings/iio/accel/bma180.txt index c5933573e0f6..4a3679d54457 100644 --- a/Documentation/devicetree/bindings/iio/accel/bma180.txt +++ b/Documentation/devicetree/bindings/iio/accel/bma180.txt @@ -1,10 +1,11 @@ -* Bosch BMA180 triaxial acceleration sensor +* Bosch BMA180 / BMA250 triaxial acceleration sensor http://omapworld.com/BMA180_111_1002839.pdf +http://ae-bst.resource.bosch.com/media/products/dokumente/bma250/bst-bma250-ds002-05.pdf Required properties: - - compatible : should be "bosch,bma180" + - compatible : should be "bosch,bma180" or "bosch,bma250" - reg : the I2C address of the sensor Optional properties: @@ -13,6 +14,9 @@ Optional properties: - interrupts : interrupt mapping for GPIO IRQ, it should by configured with flags IRQ_TYPE_LEVEL_HIGH | IRQ_TYPE_EDGE_RISING + For the bma250 the first interrupt listed must be the one + connected to the INT1 pin, the second (optional) interrupt + listed must be the one connected to the INT2 pin. Example: diff --git a/Documentation/devicetree/bindings/pci/pci-rcar-gen2.txt b/Documentation/devicetree/bindings/pci/pci-rcar-gen2.txt index d8ef5bf50f11..7fab84b33531 100644 --- a/Documentation/devicetree/bindings/pci/pci-rcar-gen2.txt +++ b/Documentation/devicetree/bindings/pci/pci-rcar-gen2.txt @@ -7,7 +7,8 @@ OHCI and EHCI controllers. Required properties: - compatible: "renesas,pci-r8a7790" for the R8A7790 SoC; - "renesas,pci-r8a7791" for the R8A7791 SoC. + "renesas,pci-r8a7791" for the R8A7791 SoC; + "renesas,pci-r8a7794" for the R8A7794 SoC. - reg: A list of physical regions to access the device: the first is the operational registers for the OHCI/EHCI controllers and the second is for the bridge configuration and control registers. diff --git a/Documentation/devicetree/bindings/regulator/pbias-regulator.txt b/Documentation/devicetree/bindings/regulator/pbias-regulator.txt index 32aa26f1e434..acbcb452a69a 100644 --- a/Documentation/devicetree/bindings/regulator/pbias-regulator.txt +++ b/Documentation/devicetree/bindings/regulator/pbias-regulator.txt @@ -2,7 +2,12 @@ PBIAS internal regulator for SD card dual voltage i/o pads on OMAP SoCs. Required properties: - compatible: - - "ti,pbias-omap" for OMAP2, OMAP3, OMAP4, OMAP5, DRA7. + - should be "ti,pbias-dra7" for DRA7 + - should be "ti,pbias-omap2" for OMAP2 + - should be "ti,pbias-omap3" for OMAP3 + - should be "ti,pbias-omap4" for OMAP4 + - should be "ti,pbias-omap5" for OMAP5 + - "ti,pbias-omap" is deprecated - reg: pbias register offset from syscon base and size of pbias register. - syscon : phandle of the system control module - regulator-name : should be diff --git a/Documentation/devicetree/bindings/spi/spi-mt65xx.txt b/Documentation/devicetree/bindings/spi/spi-mt65xx.txt index dcefc438272f..6160ffbcb3d3 100644 --- a/Documentation/devicetree/bindings/spi/spi-mt65xx.txt +++ b/Documentation/devicetree/bindings/spi/spi-mt65xx.txt @@ -15,17 +15,18 @@ Required properties: - interrupts: Should contain spi interrupt - clocks: phandles to input clocks. - The first should be <&topckgen CLK_TOP_SPI_SEL>. - The second should be one of the following. + The first should be one of the following. It's PLL. - <&clk26m>: specify parent clock 26MHZ. - <&topckgen CLK_TOP_SYSPLL3_D2>: specify parent clock 109MHZ. It's the default one. - <&topckgen CLK_TOP_SYSPLL4_D2>: specify parent clock 78MHZ. - <&topckgen CLK_TOP_UNIVPLL2_D4>: specify parent clock 104MHZ. - <&topckgen CLK_TOP_UNIVPLL1_D8>: specify parent clock 78MHZ. + The second should be <&topckgen CLK_TOP_SPI_SEL>. It's clock mux. + The third is <&pericfg CLK_PERI_SPI0>. It's clock gate. -- clock-names: shall be "spi-clk" for the controller clock, and - "parent-clk" for the parent clock. +- clock-names: shall be "parent-clk" for the parent clock, "sel-clk" for the + muxes clock, and "spi-clk" for the clock gate. Optional properties: - mediatek,pad-select: specify which pins group(ck/mi/mo/cs) spi @@ -44,8 +45,11 @@ spi: spi@1100a000 { #size-cells = <0>; reg = <0 0x1100a000 0 0x1000>; interrupts = <GIC_SPI 110 IRQ_TYPE_LEVEL_LOW>; - clocks = <&topckgen CLK_TOP_SPI_SEL>, <&topckgen CLK_TOP_SYSPLL3_D2>; - clock-names = "spi-clk", "parent-clk"; + clocks = <&topckgen CLK_TOP_SYSPLL3_D2>, + <&topckgen CLK_TOP_SPI_SEL>, + <&pericfg CLK_PERI_SPI0>; + clock-names = "parent-clk", "sel-clk", "spi-clk"; + mediatek,pad-select = <0>; status = "disabled"; }; diff --git a/Documentation/devicetree/bindings/thermal/thermal.txt b/Documentation/devicetree/bindings/thermal/thermal.txt index 8a49362dea6e..41b817f7b670 100644 --- a/Documentation/devicetree/bindings/thermal/thermal.txt +++ b/Documentation/devicetree/bindings/thermal/thermal.txt @@ -55,19 +55,11 @@ of heat dissipation). For example a fan's cooling states correspond to the different fan speeds possible. Cooling states are referred to by single unsigned integers, where larger numbers mean greater heat dissipation. The precise set of cooling states associated with a device -(as referred to be the cooling-min-state and cooling-max-state +(as referred to by the cooling-min-level and cooling-max-level properties) should be defined in a particular device's binding. For more examples of cooling devices, refer to the example sections below. Required properties: -- cooling-min-state: An integer indicating the smallest - Type: unsigned cooling state accepted. Typically 0. - Size: one cell - -- cooling-max-state: An integer indicating the largest - Type: unsigned cooling state accepted. - Size: one cell - - #cooling-cells: Used to provide cooling device specific information Type: unsigned while referring to it. Must be at least 2, in order Size: one cell to specify minimum and maximum cooling state used @@ -77,6 +69,15 @@ Required properties: See Cooling device maps section below for more details on how consumers refer to cooling devices. +Optional properties: +- cooling-min-level: An integer indicating the smallest + Type: unsigned cooling state accepted. Typically 0. + Size: one cell + +- cooling-max-level: An integer indicating the largest + Type: unsigned cooling state accepted. + Size: one cell + * Trip points The trip node is a node to describe a point in the temperature domain @@ -225,8 +226,8 @@ cpus { 396000 950000 198000 850000 >; - cooling-min-state = <0>; - cooling-max-state = <3>; + cooling-min-level = <0>; + cooling-max-level = <3>; #cooling-cells = <2>; /* min followed by max */ }; ... @@ -240,8 +241,8 @@ cpus { */ fan0: fan@0x48 { ... - cooling-min-state = <0>; - cooling-max-state = <9>; + cooling-min-level = <0>; + cooling-max-level = <9>; #cooling-cells = <2>; /* min followed by max */ }; }; diff --git a/Documentation/devicetree/bindings/vendor-prefixes.txt b/Documentation/devicetree/bindings/vendor-prefixes.txt index ac5f0c34ae00..82d2ac97af74 100644 --- a/Documentation/devicetree/bindings/vendor-prefixes.txt +++ b/Documentation/devicetree/bindings/vendor-prefixes.txt @@ -203,6 +203,7 @@ sitronix Sitronix Technology Corporation skyworks Skyworks Solutions, Inc. smsc Standard Microsystems Corporation snps Synopsys, Inc. +socionext Socionext Inc. solidrun SolidRun solomon Solomon Systech Limited sony Sony Corporation diff --git a/Documentation/thermal/power_allocator.txt b/Documentation/thermal/power_allocator.txt index c3797b529991..a1ce2235f121 100644 --- a/Documentation/thermal/power_allocator.txt +++ b/Documentation/thermal/power_allocator.txt @@ -4,7 +4,7 @@ Power allocator governor tunables Trip points ----------- -The governor requires the following two passive trip points: +The governor works optimally with the following two passive trip points: 1. "switch on" trip point: temperature above which the governor control loop starts operating. This is the first passive trip diff --git a/MAINTAINERS b/MAINTAINERS index 0581f47e959b..45b06ab43ec0 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -10338,6 +10338,16 @@ F: include/uapi/linux/thermal.h F: include/linux/cpu_cooling.h F: Documentation/devicetree/bindings/thermal/ +THERMAL/CPU_COOLING +M: Amit Daniel Kachhap <amit.kachhap@gmail.com> +M: Viresh Kumar <viresh.kumar@linaro.org> +M: Javi Merino <javi.merino@arm.com> +L: linux-pm@vger.kernel.org +S: Supported +F: Documentation/thermal/cpu-cooling-api.txt +F: drivers/thermal/cpu_cooling.c +F: include/linux/cpu_cooling.h + THINGM BLINK(1) USB RGB LED DRIVER M: Vivien Didelot <vivien.didelot@savoirfairelinux.com> S: Maintained diff --git a/arch/alpha/kernel/pci.c b/arch/alpha/kernel/pci.c index cded02c890aa..5f387ee5b5c5 100644 --- a/arch/alpha/kernel/pci.c +++ b/arch/alpha/kernel/pci.c @@ -242,7 +242,12 @@ pci_restore_srm_config(void) void pcibios_fixup_bus(struct pci_bus *bus) { - struct pci_dev *dev; + struct pci_dev *dev = bus->self; + + if (pci_has_flag(PCI_PROBE_ONLY) && dev && + (dev->class >> 8) == PCI_CLASS_BRIDGE_PCI) { + pci_read_bridge_bases(bus); + } list_for_each_entry(dev, &bus->devices, bus_list) { pdev_save_srm_config(dev); diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h index 3df1e975f72a..c4072d9f32c7 100644 --- a/arch/arm/include/asm/kvm_host.h +++ b/arch/arm/include/asm/kvm_host.h @@ -33,6 +33,7 @@ #define KVM_PRIVATE_MEM_SLOTS 4 #define KVM_COALESCED_MMIO_PAGE_OFFSET 1 #define KVM_HAVE_ONE_REG +#define KVM_HALT_POLL_NS_DEFAULT 500000 #define KVM_VCPU_MAX_FEATURES 2 diff --git a/arch/arm64/boot/dts/mediatek/mt8173.dtsi b/arch/arm64/boot/dts/mediatek/mt8173.dtsi index d18ee4259ee5..06a15644be38 100644 --- a/arch/arm64/boot/dts/mediatek/mt8173.dtsi +++ b/arch/arm64/boot/dts/mediatek/mt8173.dtsi @@ -81,7 +81,7 @@ }; idle-states { - entry-method = "arm,psci"; + entry-method = "psci"; CPU_SLEEP_0: cpu-sleep-0 { compatible = "arm,idle-state"; diff --git a/arch/arm64/boot/dts/rockchip/rk3368.dtsi b/arch/arm64/boot/dts/rockchip/rk3368.dtsi index a712bea3bf2c..cc093a482aa4 100644 --- a/arch/arm64/boot/dts/rockchip/rk3368.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3368.dtsi @@ -106,7 +106,7 @@ }; idle-states { - entry-method = "arm,psci"; + entry-method = "psci"; cpu_sleep: cpu-sleep-0 { compatible = "arm,idle-state"; diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h index 4562459456a6..ed039688c221 100644 --- a/arch/arm64/include/asm/kvm_host.h +++ b/arch/arm64/include/asm/kvm_host.h @@ -33,6 +33,7 @@ #define KVM_USER_MEM_SLOTS 32 #define KVM_PRIVATE_MEM_SLOTS 4 #define KVM_COALESCED_MMIO_PAGE_OFFSET 1 +#define KVM_HALT_POLL_NS_DEFAULT 500000 #include <kvm/arm_vgic.h> #include <kvm/arm_arch_timer.h> diff --git a/arch/frv/mb93090-mb00/pci-vdk.c b/arch/frv/mb93090-mb00/pci-vdk.c index f9c86c475bbd..f211839e2cae 100644 --- a/arch/frv/mb93090-mb00/pci-vdk.c +++ b/arch/frv/mb93090-mb00/pci-vdk.c @@ -294,6 +294,8 @@ void pcibios_fixup_bus(struct pci_bus *bus) printk("### PCIBIOS_FIXUP_BUS(%d)\n",bus->number); #endif + pci_read_bridge_bases(bus); + if (bus->number == 0) { struct pci_dev *dev; list_for_each_entry(dev, &bus->devices, bus_list) { diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c index d89b6013c941..7cc3be9fa7c6 100644 --- a/arch/ia64/pci/pci.c +++ b/arch/ia64/pci/pci.c @@ -533,9 +533,10 @@ void pcibios_fixup_bus(struct pci_bus *b) { struct pci_dev *dev; - if (b->self) + if (b->self) { + pci_read_bridge_bases(b); pcibios_fixup_bridge_resources(b->self); - + } list_for_each_entry(dev, &b->devices, bus_list) pcibios_fixup_device_resources(dev); platform_pci_fixup_bus(b); diff --git a/arch/microblaze/pci/pci-common.c b/arch/microblaze/pci/pci-common.c index 6b8b75266801..ae838ed5fcf2 100644 --- a/arch/microblaze/pci/pci-common.c +++ b/arch/microblaze/pci/pci-common.c @@ -863,7 +863,14 @@ void pcibios_setup_bus_devices(struct pci_bus *bus) void pcibios_fixup_bus(struct pci_bus *bus) { - /* Fixup the bus */ + /* When called from the generic PCI probe, read PCI<->PCI bridge + * bases. This is -not- called when generating the PCI tree from + * the OF device-tree. + */ + if (bus->self != NULL) + pci_read_bridge_bases(bus); + + /* Now fixup the bus bus */ pcibios_setup_bus_self(bus); /* Now fixup devices on that bus */ diff --git a/arch/mips/include/asm/kvm_host.h b/arch/mips/include/asm/kvm_host.h index 3a54dbca9f7e..5a1a882e0a75 100644 --- a/arch/mips/include/asm/kvm_host.h +++ b/arch/mips/include/asm/kvm_host.h @@ -61,6 +61,7 @@ #define KVM_PRIVATE_MEM_SLOTS 0 #define KVM_COALESCED_MMIO_PAGE_OFFSET 1 +#define KVM_HALT_POLL_NS_DEFAULT 500000 diff --git a/arch/mips/pci/pci.c b/arch/mips/pci/pci.c index c6996cf67a5c..b8a0bf5766f2 100644 --- a/arch/mips/pci/pci.c +++ b/arch/mips/pci/pci.c @@ -311,6 +311,12 @@ int pcibios_enable_device(struct pci_dev *dev, int mask) void pcibios_fixup_bus(struct pci_bus *bus) { + struct pci_dev *dev = bus->self; + + if (pci_has_flag(PCI_PROBE_ONLY) && dev && + (dev->class >> 8) == PCI_CLASS_BRIDGE_PCI) { + pci_read_bridge_bases(bus); + } } EXPORT_SYMBOL(PCIBIOS_MIN_IO); diff --git a/arch/mn10300/unit-asb2305/pci.c b/arch/mn10300/unit-asb2305/pci.c index deaa893efba5..3dfe2d31c67b 100644 --- a/arch/mn10300/unit-asb2305/pci.c +++ b/arch/mn10300/unit-asb2305/pci.c @@ -324,6 +324,7 @@ void pcibios_fixup_bus(struct pci_bus *bus) struct pci_dev *dev; if (bus->self) { + pci_read_bridge_bases(bus); pcibios_fixup_bridge_resources(bus->self); } diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h index 195886a583ba..827a38d7a9db 100644 --- a/arch/powerpc/include/asm/kvm_host.h +++ b/arch/powerpc/include/asm/kvm_host.h @@ -44,6 +44,7 @@ #ifdef CONFIG_KVM_MMIO #define KVM_COALESCED_MMIO_PAGE_OFFSET 1 #endif +#define KVM_HALT_POLL_NS_DEFAULT 500000 /* These values are internal and can be increased later */ #define KVM_NR_IRQCHIPS 1 diff --git a/arch/powerpc/include/asm/systbl.h b/arch/powerpc/include/asm/systbl.h index 4d65499ee1c1..126d0c4f9b7d 100644 --- a/arch/powerpc/include/asm/systbl.h +++ b/arch/powerpc/include/asm/systbl.h @@ -369,3 +369,4 @@ SYSCALL_SPU(bpf) COMPAT_SYS(execveat) PPC64ONLY(switch_endian) SYSCALL_SPU(userfaultfd) +SYSCALL_SPU(membarrier) diff --git a/arch/powerpc/include/asm/unistd.h b/arch/powerpc/include/asm/unistd.h index 4a055b6c2a64..13411be86041 100644 --- a/arch/powerpc/include/asm/unistd.h +++ b/arch/powerpc/include/asm/unistd.h @@ -12,7 +12,7 @@ #include <uapi/asm/unistd.h> -#define __NR_syscalls 365 +#define __NR_syscalls 366 #define __NR__exit __NR_exit #define NR_syscalls __NR_syscalls diff --git a/arch/powerpc/include/uapi/asm/unistd.h b/arch/powerpc/include/uapi/asm/unistd.h index 6ad58d4c879b..6337738018aa 100644 --- a/arch/powerpc/include/uapi/asm/unistd.h +++ b/arch/powerpc/include/uapi/asm/unistd.h @@ -387,5 +387,6 @@ #define __NR_execveat 362 #define __NR_switch_endian 363 #define __NR_userfaultfd 364 +#define __NR_membarrier 365 #endif /* _UAPI_ASM_POWERPC_UNISTD_H_ */ diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c index a1d0632d97c6..7587b2ae5f77 100644 --- a/arch/powerpc/kernel/pci-common.c +++ b/arch/powerpc/kernel/pci-common.c @@ -1032,7 +1032,13 @@ void pcibios_set_master(struct pci_dev *dev) void pcibios_fixup_bus(struct pci_bus *bus) { - /* Fixup the bus */ + /* When called from the generic PCI probe, read PCI<->PCI bridge + * bases. This is -not- called when generating the PCI tree from + * the OF device-tree. + */ + pci_read_bridge_bases(bus); + + /* Now fixup the bus bus */ pcibios_setup_bus_self(bus); /* Now fixup devices on that bus */ diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c index cf009167d208..099c79d8c160 100644 --- a/arch/powerpc/kvm/book3s.c +++ b/arch/powerpc/kvm/book3s.c @@ -829,12 +829,15 @@ int kvmppc_h_logical_ci_load(struct kvm_vcpu *vcpu) unsigned long size = kvmppc_get_gpr(vcpu, 4); unsigned long addr = kvmppc_get_gpr(vcpu, 5); u64 buf; + int srcu_idx; int ret; if (!is_power_of_2(size) || (size > sizeof(buf))) return H_TOO_HARD; + srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); ret = kvm_io_bus_read(vcpu, KVM_MMIO_BUS, addr, size, &buf); + srcu_read_unlock(&vcpu->kvm->srcu, srcu_idx); if (ret != 0) return H_TOO_HARD; @@ -869,6 +872,7 @@ int kvmppc_h_logical_ci_store(struct kvm_vcpu *vcpu) unsigned long addr = kvmppc_get_gpr(vcpu, 5); unsigned long val = kvmppc_get_gpr(vcpu, 6); u64 buf; + int srcu_idx; int ret; switch (size) { @@ -892,7 +896,9 @@ int kvmppc_h_logical_ci_store(struct kvm_vcpu *vcpu) return H_TOO_HARD; } + srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); ret = kvm_io_bus_write(vcpu, KVM_MMIO_BUS, addr, size, &buf); + srcu_read_unlock(&vcpu->kvm->srcu, srcu_idx); if (ret != 0) return H_TOO_HARD; diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c index 9754e6815e52..228049786888 100644 --- a/arch/powerpc/kvm/book3s_hv.c +++ b/arch/powerpc/kvm/book3s_hv.c @@ -2692,9 +2692,13 @@ static int kvmppc_run_vcpu(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu) while (vcpu->arch.state == KVMPPC_VCPU_RUNNABLE && (vc->vcore_state == VCORE_RUNNING || - vc->vcore_state == VCORE_EXITING)) + vc->vcore_state == VCORE_EXITING || + vc->vcore_state == VCORE_PIGGYBACK)) kvmppc_wait_for_exec(vc, vcpu, TASK_UNINTERRUPTIBLE); + if (vc->vcore_state == VCORE_PREEMPT && vc->runner == NULL) + kvmppc_vcore_end_preempt(vc); + if (vcpu->arch.state == KVMPPC_VCPU_RUNNABLE) { kvmppc_remove_runnable(vc, vcpu); vcpu->stat.signal_exits++; diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S index 2273dcacef39..b98889e9851d 100644 --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S @@ -1257,6 +1257,7 @@ mc_cont: bl kvmhv_accumulate_time #endif + mr r3, r12 /* Increment exit count, poke other threads to exit */ bl kvmhv_commence_exit nop diff --git a/arch/s390/configs/zfcpdump_defconfig b/arch/s390/configs/zfcpdump_defconfig index 1b0184a0f7f2..92805d604173 100644 --- a/arch/s390/configs/zfcpdump_defconfig +++ b/arch/s390/configs/zfcpdump_defconfig @@ -1,7 +1,6 @@ # CONFIG_SWAP is not set CONFIG_NO_HZ=y CONFIG_HIGH_RES_TIMERS=y -CONFIG_RCU_FAST_NO_HZ=y CONFIG_BLK_DEV_INITRD=y CONFIG_CC_OPTIMIZE_FOR_SIZE=y # CONFIG_COMPAT_BRK is not set @@ -54,10 +53,6 @@ CONFIG_RAW_DRIVER=y # CONFIG_MONWRITER is not set # CONFIG_S390_VMUR is not set # CONFIG_HID is not set -CONFIG_MEMSTICK=y -CONFIG_MEMSTICK_DEBUG=y -CONFIG_MEMSTICK_UNSAFE_RESUME=y -CONFIG_MSPRO_BLOCK=y # CONFIG_IOMMU_SUPPORT is not set CONFIG_EXT2_FS=y CONFIG_EXT3_FS=y diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h index 6ce4a0b7e8da..8ced426091e1 100644 --- a/arch/s390/include/asm/kvm_host.h +++ b/arch/s390/include/asm/kvm_host.h @@ -35,6 +35,7 @@ */ #define KVM_NR_IRQCHIPS 1 #define KVM_IRQCHIP_NUM_PINS 4096 +#define KVM_HALT_POLL_NS_DEFAULT 0 #define SIGP_CTRL_C 0x80 #define SIGP_CTRL_SCN_MASK 0x3f diff --git a/arch/s390/include/asm/unistd.h b/arch/s390/include/asm/unistd.h index 525cef73b085..02613bad8bbb 100644 --- a/arch/s390/include/asm/unistd.h +++ b/arch/s390/include/asm/unistd.h @@ -8,28 +8,8 @@ #include <uapi/asm/unistd.h> - #define __IGNORE_time -/* Ignore system calls that are also reachable via sys_socketcall */ -#define __IGNORE_recvmmsg -#define __IGNORE_sendmmsg -#define __IGNORE_socket -#define __IGNORE_socketpair -#define __IGNORE_bind -#define __IGNORE_connect -#define __IGNORE_listen -#define __IGNORE_accept4 -#define __IGNORE_getsockopt -#define __IGNORE_setsockopt -#define __IGNORE_getsockname -#define __IGNORE_getpeername -#define __IGNORE_sendto -#define __IGNORE_sendmsg -#define __IGNORE_recvfrom -#define __IGNORE_recvmsg -#define __IGNORE_shutdown - #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_SYS_ALARM #define __ARCH_WANT_SYS_GETHOSTNAME diff --git a/arch/s390/include/uapi/asm/unistd.h b/arch/s390/include/uapi/asm/unistd.h index 59d2bb4e2d0c..a848adba1504 100644 --- a/arch/s390/include/uapi/asm/unistd.h +++ b/arch/s390/include/uapi/asm/unistd.h @@ -290,7 +290,26 @@ #define __NR_s390_pci_mmio_write 352 #define __NR_s390_pci_mmio_read 353 #define __NR_execveat 354 -#define NR_syscalls 355 +#define __NR_userfaultfd 355 +#define __NR_membarrier 356 +#define __NR_recvmmsg 357 +#define __NR_sendmmsg 358 +#define __NR_socket 359 +#define __NR_socketpair 360 +#define __NR_bind 361 +#define __NR_connect 362 +#define __NR_listen 363 +#define __NR_accept4 364 +#define __NR_getsockopt 365 +#define __NR_setsockopt 366 +#define __NR_getsockname 367 +#define __NR_getpeername 368 +#define __NR_sendto 369 +#define __NR_sendmsg 370 +#define __NR_recvfrom 371 +#define __NR_recvmsg 372 +#define __NR_shutdown 373 +#define NR_syscalls 374 /* * There are some system calls that are not present on 64 bit, some diff --git a/arch/s390/kernel/compat_signal.c b/arch/s390/kernel/compat_signal.c index eb4664238613..e0f9d270b30f 100644 --- a/arch/s390/kernel/compat_signal.c +++ b/arch/s390/kernel/compat_signal.c @@ -48,6 +48,19 @@ typedef struct struct ucontext32 uc; } rt_sigframe32; +static inline void sigset_to_sigset32(unsigned long *set64, + compat_sigset_word *set32) +{ + set32[0] = (compat_sigset_word) set64[0]; + set32[1] = (compat_sigset_word)(set64[0] >> 32); +} + +static inline void sigset32_to_sigset(compat_sigset_word *set32, + unsigned long *set64) +{ + set64[0] = (unsigned long) set32[0] | ((unsigned long) set32[1] << 32); +} + int copy_siginfo_to_user32(compat_siginfo_t __user *to, const siginfo_t *from) { int err; @@ -281,10 +294,12 @@ COMPAT_SYSCALL_DEFINE0(sigreturn) { struct pt_regs *regs = task_pt_regs(current); sigframe32 __user *frame = (sigframe32 __user *)regs->gprs[15]; + compat_sigset_t cset; sigset_t set; - if (__copy_from_user(&set.sig, &frame->sc.oldmask, _SIGMASK_COPY_SIZE32)) + if (__copy_from_user(&cset.sig, &frame->sc.oldmask, _SIGMASK_COPY_SIZE32)) goto badframe; + sigset32_to_sigset(cset.sig, set.sig); set_current_blocked(&set); save_fpu_regs(); if (restore_sigregs32(regs, &frame->sregs)) @@ -302,10 +317,12 @@ COMPAT_SYSCALL_DEFINE0(rt_sigreturn) { struct pt_regs *regs = task_pt_regs(current); rt_sigframe32 __user *frame = (rt_sigframe32 __user *)regs->gprs[15]; + compat_sigset_t cset; sigset_t set; - if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set))) + if (__copy_from_user(&cset, &frame->uc.uc_sigmask, sizeof(cset))) goto badframe; + sigset32_to_sigset(cset.sig, set.sig); set_current_blocked(&set); if (compat_restore_altstack(&frame->uc.uc_stack)) goto badframe; @@ -377,7 +394,7 @@ static int setup_frame32(struct ksignal *ksig, sigset_t *set, return -EFAULT; /* Create struct sigcontext32 on the signal stack */ - memcpy(&sc.oldmask, &set->sig, _SIGMASK_COPY_SIZE32); + sigset_to_sigset32(set->sig, sc.oldmask); sc.sregs = (__u32)(unsigned long __force) &frame->sregs; if (__copy_to_user(&frame->sc, &sc, sizeof(frame->sc))) return -EFAULT; @@ -438,6 +455,7 @@ static int setup_frame32(struct ksignal *ksig, sigset_t *set, static int setup_rt_frame32(struct ksignal *ksig, sigset_t *set, struct pt_regs *regs) { + compat_sigset_t cset; rt_sigframe32 __user *frame; unsigned long restorer; size_t frame_size; @@ -485,11 +503,12 @@ static int setup_rt_frame32(struct ksignal *ksig, sigset_t *set, store_sigregs(); /* Create ucontext on the signal stack. */ + sigset_to_sigset32(set->sig, cset.sig); if (__put_user(uc_flags, &frame->uc.uc_flags) || __put_user(0, &frame->uc.uc_link) || __compat_save_altstack(&frame->uc.uc_stack, regs->gprs[15]) || save_sigregs32(regs, &frame->uc.uc_mcontext) || - __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)) || + __copy_to_user(&frame->uc.uc_sigmask, &cset, sizeof(cset)) || save_sigregs_ext32(regs, &frame->uc.uc_mcontext_ext)) return -EFAULT; diff --git a/arch/s390/kernel/compat_wrapper.c b/arch/s390/kernel/compat_wrapper.c index f8498dde67b1..09f194052df3 100644 --- a/arch/s390/kernel/compat_wrapper.c +++ b/arch/s390/kernel/compat_wrapper.c @@ -52,15 +52,13 @@ * the regular system call wrappers. */ #define COMPAT_SYSCALL_WRAPx(x, name, ...) \ - asmlinkage long sys##name(__MAP(x,__SC_DECL,__VA_ARGS__)); \ - asmlinkage long compat_sys##name(__MAP(x,__SC_COMPAT_TYPE,__VA_ARGS__));\ - asmlinkage long compat_sys##name(__MAP(x,__SC_COMPAT_TYPE,__VA_ARGS__)) \ - { \ - return sys##name(__MAP(x,__SC_COMPAT_CAST,__VA_ARGS__)); \ - } +asmlinkage long sys##name(__MAP(x,__SC_DECL,__VA_ARGS__)); \ +asmlinkage long notrace compat_sys##name(__MAP(x,__SC_COMPAT_TYPE,__VA_ARGS__));\ +asmlinkage long notrace compat_sys##name(__MAP(x,__SC_COMPAT_TYPE,__VA_ARGS__)) \ +{ \ + return sys##name(__MAP(x,__SC_COMPAT_CAST,__VA_ARGS__)); \ +} -COMPAT_SYSCALL_WRAP1(exit, int, error_code); -COMPAT_SYSCALL_WRAP1(close, unsigned int, fd); COMPAT_SYSCALL_WRAP2(creat, const char __user *, pathname, umode_t, mode); COMPAT_SYSCALL_WRAP2(link, const char __user *, oldname, const char __user *, newname); COMPAT_SYSCALL_WRAP1(unlink, const char __user *, pathname); @@ -68,23 +66,16 @@ COMPAT_SYSCALL_WRAP1(chdir, const char __user *, filename); COMPAT_SYSCALL_WRAP3(mknod, const char __user *, filename, umode_t, mode, unsigned, dev); COMPAT_SYSCALL_WRAP2(chmod, const char __user *, filename, umode_t, mode); COMPAT_SYSCALL_WRAP1(oldumount, char __user *, name); -COMPAT_SYSCALL_WRAP1(alarm, unsigned int, seconds); COMPAT_SYSCALL_WRAP2(access, const char __user *, filename, int, mode); -COMPAT_SYSCALL_WRAP1(nice, int, increment); -COMPAT_SYSCALL_WRAP2(kill, int, pid, int, sig); COMPAT_SYSCALL_WRAP2(rename, const char __user *, oldname, const char __user *, newname); COMPAT_SYSCALL_WRAP2(mkdir, const char __user *, pathname, umode_t, mode); COMPAT_SYSCALL_WRAP1(rmdir, const char __user *, pathname); -COMPAT_SYSCALL_WRAP1(dup, unsigned int, fildes); COMPAT_SYSCALL_WRAP1(pipe, int __user *, fildes); COMPAT_SYSCALL_WRAP1(brk, unsigned long, brk); COMPAT_SYSCALL_WRAP2(signal, int, sig, __sighandler_t, handler); COMPAT_SYSCALL_WRAP1(acct, const char __user *, name); COMPAT_SYSCALL_WRAP2(umount, char __user *, name, int, flags); -COMPAT_SYSCALL_WRAP2(setpgid, pid_t, pid, pid_t, pgid); -COMPAT_SYSCALL_WRAP1(umask, int, mask); COMPAT_SYSCALL_WRAP1(chroot, const char __user *, filename); -COMPAT_SYSCALL_WRAP2(dup2, unsigned int, oldfd, unsigned int, newfd); COMPAT_SYSCALL_WRAP3(sigsuspend, int, unused1, int, unused2, old_sigset_t, mask); COMPAT_SYSCALL_WRAP2(sethostname, char __user *, name, int, len); COMPAT_SYSCALL_WRAP2(symlink, const char __user *, old, const char __user *, new); @@ -93,37 +84,23 @@ COMPAT_SYSCALL_WRAP1(uselib, const char __user *, library); COMPAT_SYSCALL_WRAP2(swapon, const char __user *, specialfile, int, swap_flags); COMPAT_SYSCALL_WRAP4(reboot, int, magic1, int, magic2, unsigned int, cmd, void __user *, arg); COMPAT_SYSCALL_WRAP2(munmap, unsigned long, addr, size_t, len); -COMPAT_SYSCALL_WRAP2(fchmod, unsigned int, fd, umode_t, mode); -COMPAT_SYSCALL_WRAP2(getpriority, int, which, int, who); -COMPAT_SYSCALL_WRAP3(setpriority, int, which, int, who, int, niceval); COMPAT_SYSCALL_WRAP3(syslog, int, type, char __user *, buf, int, len); COMPAT_SYSCALL_WRAP1(swapoff, const char __user *, specialfile); -COMPAT_SYSCALL_WRAP1(fsync, unsigned int, fd); COMPAT_SYSCALL_WRAP2(setdomainname, char __user *, name, int, len); COMPAT_SYSCALL_WRAP1(newuname, struct new_utsname __user *, name); COMPAT_SYSCALL_WRAP3(mprotect, unsigned long, start, size_t, len, unsigned long, prot); COMPAT_SYSCALL_WRAP3(init_module, void __user *, umod, unsigned long, len, const char __user *, uargs); COMPAT_SYSCALL_WRAP2(delete_module, const char __user *, name_user, unsigned int, flags); COMPAT_SYSCALL_WRAP4(quotactl, unsigned int, cmd, const char __user *, special, qid_t, id, void __user *, addr); -COMPAT_SYSCALL_WRAP1(getpgid, pid_t, pid); -COMPAT_SYSCALL_WRAP1(fchdir, unsigned int, fd); COMPAT_SYSCALL_WRAP2(bdflush, int, func, long, data); COMPAT_SYSCALL_WRAP3(sysfs, int, option, unsigned long, arg1, unsigned long, arg2); -COMPAT_SYSCALL_WRAP1(s390_personality, unsigned int, personality); COMPAT_SYSCALL_WRAP5(llseek, unsigned int, fd, unsigned long, high, unsigned long, low, loff_t __user *, result, unsigned int, whence); -COMPAT_SYSCALL_WRAP2(flock, unsigned int, fd, unsigned int, cmd); COMPAT_SYSCALL_WRAP3(msync, unsigned long, start, size_t, len, int, flags); -COMPAT_SYSCALL_WRAP1(getsid, pid_t, pid); -COMPAT_SYSCALL_WRAP1(fdatasync, unsigned int, fd); COMPAT_SYSCALL_WRAP2(mlock, unsigned long, start, size_t, len); COMPAT_SYSCALL_WRAP2(munlock, unsigned long, start, size_t, len); -COMPAT_SYSCALL_WRAP1(mlockall, int, flags); COMPAT_SYSCALL_WRAP2(sched_setparam, pid_t, pid, struct sched_param __user *, param); COMPAT_SYSCALL_WRAP2(sched_getparam, pid_t, pid, struct sched_param __user *, param); COMPAT_SYSCALL_WRAP3(sched_setscheduler, pid_t, pid, int, policy, struct sched_param __user *, param); -COMPAT_SYSCALL_WRAP1(sched_getscheduler, pid_t, pid); -COMPAT_SYSCALL_WRAP1(sched_get_priority_max, int, policy); -COMPAT_SYSCALL_WRAP1(sched_get_priority_min, int, policy); COMPAT_SYSCALL_WRAP5(mremap, unsigned long, addr, unsigned long, old_len, unsigned long, new_len, unsigned long, flags, unsigned long, new_addr); COMPAT_SYSCALL_WRAP3(poll, struct pollfd __user *, ufds, unsigned int, nfds, int, timeout); COMPAT_SYSCALL_WRAP5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, unsigned long, arg4, unsigned long, arg5); @@ -131,20 +108,11 @@ COMPAT_SYSCALL_WRAP2(getcwd, char __user *, buf, unsigned long, size); COMPAT_SYSCALL_WRAP2(capget, cap_user_header_t, header, cap_user_data_t, dataptr); COMPAT_SYSCALL_WRAP2(capset, cap_user_header_t, header, const cap_user_data_t, data); COMPAT_SYSCALL_WRAP3(lchown, const char __user *, filename, uid_t, user, gid_t, group); -COMPAT_SYSCALL_WRAP2(setreuid, uid_t, ruid, uid_t, euid); -COMPAT_SYSCALL_WRAP2(setregid, gid_t, rgid, gid_t, egid); COMPAT_SYSCALL_WRAP2(getgroups, int, gidsetsize, gid_t __user *, grouplist); COMPAT_SYSCALL_WRAP2(setgroups, int, gidsetsize, gid_t __user *, grouplist); -COMPAT_SYSCALL_WRAP3(fchown, unsigned int, fd, uid_t, user, gid_t, group); -COMPAT_SYSCALL_WRAP3(setresuid, uid_t, ruid, uid_t, euid, uid_t, suid); COMPAT_SYSCALL_WRAP3(getresuid, uid_t __user *, ruid, uid_t __user *, euid, uid_t __user *, suid); -COMPAT_SYSCALL_WRAP3(setresgid, gid_t, rgid, gid_t, egid, gid_t, sgid); COMPAT_SYSCALL_WRAP3(getresgid, gid_t __user *, rgid, gid_t __user *, egid, gid_t __user *, sgid); COMPAT_SYSCALL_WRAP3(chown, const char __user *, filename, uid_t, user, gid_t, group); -COMPAT_SYSCALL_WRAP1(setuid, uid_t, uid); -COMPAT_SYSCALL_WRAP1(setgid, gid_t, gid); -COMPAT_SYSCALL_WRAP1(setfsuid, uid_t, uid); -COMPAT_SYSCALL_WRAP1(setfsgid, gid_t, gid); COMPAT_SYSCALL_WRAP2(pivot_root, const char __user *, new_root, const char __user *, put_old); COMPAT_SYSCALL_WRAP3(mincore, unsigned long, start, size_t, len, unsigned char __user *, vec); COMPAT_SYSCALL_WRAP3(madvise, unsigned long, start, size_t, len, int, behavior); @@ -161,23 +129,16 @@ COMPAT_SYSCALL_WRAP3(flistxattr, int, fd, char __user *, list, size_t, size); COMPAT_SYSCALL_WRAP2(removexattr, const char __user *, path, const char __user *, name); COMPAT_SYSCALL_WRAP2(lremovexattr, const char __user *, path, const char __user *, name); COMPAT_SYSCALL_WRAP2(fremovexattr, int, fd, const char __user *, name); -COMPAT_SYSCALL_WRAP1(exit_group, int, error_code); COMPAT_SYSCALL_WRAP1(set_tid_address, int __user *, tidptr); -COMPAT_SYSCALL_WRAP1(epoll_create, int, size); COMPAT_SYSCALL_WRAP4(epoll_ctl, int, epfd, int, op, int, fd, struct epoll_event __user *, event); COMPAT_SYSCALL_WRAP4(epoll_wait, int, epfd, struct epoll_event __user *, events, int, maxevents, int, timeout); -COMPAT_SYSCALL_WRAP1(timer_getoverrun, timer_t, timer_id); -COMPAT_SYSCALL_WRAP1(timer_delete, compat_timer_t, compat_timer_id); COMPAT_SYSCALL_WRAP1(io_destroy, aio_context_t, ctx); COMPAT_SYSCALL_WRAP3(io_cancel, aio_context_t, ctx_id, struct iocb __user *, iocb, struct io_event __user *, result); COMPAT_SYSCALL_WRAP1(mq_unlink, const char __user *, name); COMPAT_SYSCALL_WRAP5(add_key, const char __user *, tp, const char __user *, dsc, const void __user *, pld, size_t, len, key_serial_t, id); COMPAT_SYSCALL_WRAP4(request_key, const char __user *, tp, const char __user *, dsc, const char __user *, info, key_serial_t, id); COMPAT_SYSCALL_WRAP5(remap_file_pages, unsigned long, start, unsigned long, size, unsigned long, prot, unsigned long, pgoff, unsigned long, flags); -COMPAT_SYSCALL_WRAP3(ioprio_set, int, which, int, who, int, ioprio); -COMPAT_SYSCALL_WRAP2(ioprio_get, int, which, int, who); COMPAT_SYSCALL_WRAP3(inotify_add_watch, int, fd, const char __user *, path, u32, mask); -COMPAT_SYSCALL_WRAP2(inotify_rm_watch, int, fd, __s32, wd); COMPAT_SYSCALL_WRAP3(mkdirat, int, dfd, const char __user *, pathname, umode_t, mode); COMPAT_SYSCALL_WRAP4(mknodat, int, dfd, const char __user *, filename, umode_t, mode, unsigned, dev); COMPAT_SYSCALL_WRAP5(fchownat, int, dfd, const char __user *, filename, uid_t, user, gid_t, group, int, flag); @@ -192,23 +153,11 @@ COMPAT_SYSCALL_WRAP1(unshare, unsigned long, unshare_flags); COMPAT_SYSCALL_WRAP6(splice, int, fd_in, loff_t __user *, off_in, int, fd_out, loff_t __user *, off_out, size_t, len, unsigned int, flags); COMPAT_SYSCALL_WRAP4(tee, int, fdin, int, fdout, size_t, len, unsigned int, flags); COMPAT_SYSCALL_WRAP3(getcpu, unsigned __user *, cpu, unsigned __user *, node, struct getcpu_cache __user *, cache); -COMPAT_SYSCALL_WRAP1(eventfd, unsigned int, count); -COMPAT_SYSCALL_WRAP2(timerfd_create, int, clockid, int, flags); -COMPAT_SYSCALL_WRAP2(eventfd2, unsigned int, count, int, flags); -COMPAT_SYSCALL_WRAP1(inotify_init1, int, flags); COMPAT_SYSCALL_WRAP2(pipe2, int __user *, fildes, int, flags); -COMPAT_SYSCALL_WRAP3(dup3, unsigned int, oldfd, unsigned int, newfd, int, flags); -COMPAT_SYSCALL_WRAP1(epoll_create1, int, flags); -COMPAT_SYSCALL_WRAP2(tkill, int, pid, int, sig); -COMPAT_SYSCALL_WRAP3(tgkill, int, tgid, int, pid, int, sig); COMPAT_SYSCALL_WRAP5(perf_event_open, struct perf_event_attr __user *, attr_uptr, pid_t, pid, int, cpu, int, group_fd, unsigned long, flags); COMPAT_SYSCALL_WRAP5(clone, unsigned long, newsp, unsigned long, clone_flags, int __user *, parent_tidptr, int __user *, child_tidptr, unsigned long, tls); -COMPAT_SYSCALL_WRAP2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags); COMPAT_SYSCALL_WRAP4(prlimit64, pid_t, pid, unsigned int, resource, const struct rlimit64 __user *, new_rlim, struct rlimit64 __user *, old_rlim); COMPAT_SYSCALL_WRAP5(name_to_handle_at, int, dfd, const char __user *, name, struct file_handle __user *, handle, int __user *, mnt_id, int, flag); -COMPAT_SYSCALL_WRAP1(syncfs, int, fd); -COMPAT_SYSCALL_WRAP2(setns, int, fd, int, nstype); -COMPAT_SYSCALL_WRAP2(s390_runtime_instr, int, command, int, signum); COMPAT_SYSCALL_WRAP5(kcmp, pid_t, pid1, pid_t, pid2, int, type, unsigned long, idx1, unsigned long, idx2); COMPAT_SYSCALL_WRAP3(finit_module, int, fd, const char __user *, uargs, int, flags); COMPAT_SYSCALL_WRAP3(sched_setattr, pid_t, pid, struct sched_attr __user *, attr, unsigned int, flags); @@ -220,3 +169,10 @@ COMPAT_SYSCALL_WRAP2(memfd_create, const char __user *, uname, unsigned int, fla COMPAT_SYSCALL_WRAP3(bpf, int, cmd, union bpf_attr *, attr, unsigned int, size); COMPAT_SYSCALL_WRAP3(s390_pci_mmio_write, const unsigned long, mmio_addr, const void __user *, user_buffer, const size_t, length); COMPAT_SYSCALL_WRAP3(s390_pci_mmio_read, const unsigned long, mmio_addr, void __user *, user_buffer, const size_t, length); +COMPAT_SYSCALL_WRAP4(socketpair, int, family, int, type, int, protocol, int __user *, usockvec); +COMPAT_SYSCALL_WRAP3(bind, int, fd, struct sockaddr __user *, umyaddr, int, addrlen); +COMPAT_SYSCALL_WRAP3(connect, int, fd, struct sockaddr __user *, uservaddr, int, addrlen); +COMPAT_SYSCALL_WRAP4(accept4, int, fd, struct sockaddr __user *, upeer_sockaddr, int __user *, upeer_addrlen, int, flags); +COMPAT_SYSCALL_WRAP3(getsockname, int, fd, struct sockaddr __user *, usockaddr, int __user *, usockaddr_len); +COMPAT_SYSCALL_WRAP3(getpeername, int, fd, struct sockaddr __user *, usockaddr, int __user *, usockaddr_len); +COMPAT_SYSCALL_WRAP6(sendto, int, fd, void __user *, buff, size_t, len, unsigned int, flags, struct sockaddr __user *, addr, int, addr_len); diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 247b7aae4c6d..09b039d7983d 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -1191,6 +1191,7 @@ cleanup_critical: clg %r9,BASED(.Lcleanup_save_fpu_fpc_end) jhe 1f lg %r2,__LC_CURRENT + aghi %r2,__TASK_thread 0: # Store floating-point controls stfpc __THREAD_FPU_fpc(%r2) 1: # Load register save area and check if VX is active @@ -1252,6 +1253,7 @@ cleanup_critical: clg %r9,BASED(.Lcleanup_load_fpu_regs_vx_ctl) jhe 6f lg %r4,__LC_CURRENT + aghi %r4,__TASK_thread lfpc __THREAD_FPU_fpc(%r4) tm __THREAD_FPU_flags+3(%r4),FPU_USE_VX # VX-enabled task ? lg %r4,__THREAD_FPU_regs(%r4) # %r4 <- reg save area diff --git a/arch/s390/kernel/perf_cpum_cf.c b/arch/s390/kernel/perf_cpum_cf.c index 56fdad479115..a9563409c36e 100644 --- a/arch/s390/kernel/perf_cpum_cf.c +++ b/arch/s390/kernel/perf_cpum_cf.c @@ -157,10 +157,14 @@ static int validate_ctr_auth(const struct hw_perf_event *hwc) cpuhw = &get_cpu_var(cpu_hw_events); - /* check authorization for cpu counter sets */ + /* Check authorization for cpu counter sets. + * If the particular CPU counter set is not authorized, + * return with -ENOENT in order to fall back to other + * PMUs that might suffice the event request. + */ ctrs_state = cpumf_state_ctl[hwc->config_base]; if (!(ctrs_state & cpuhw->info.auth_ctl)) - err = -EPERM; + err = -ENOENT; put_cpu_var(cpu_hw_events); return err; @@ -536,7 +540,7 @@ static int cpumf_pmu_add(struct perf_event *event, int flags) */ if (!(cpuhw->flags & PERF_EVENT_TXN)) if (validate_ctr_auth(&event->hw)) - return -EPERM; + return -ENOENT; ctr_set_enable(&cpuhw->state, event->hw.config_base); event->hw.state = PERF_HES_UPTODATE | PERF_HES_STOPPED; @@ -611,7 +615,7 @@ static int cpumf_pmu_commit_txn(struct pmu *pmu) state = cpuhw->state & ~((1 << CPUMF_LCCTL_ENABLE_SHIFT) - 1); state >>= CPUMF_LCCTL_ENABLE_SHIFT; if ((state & cpuhw->info.auth_ctl) != state) - return -EPERM; + return -ENOENT; cpuhw->flags &= ~PERF_EVENT_TXN; perf_pmu_enable(pmu); diff --git a/arch/s390/kernel/swsusp.S b/arch/s390/kernel/swsusp.S index ca6294645dd3..2d6b6e81f812 100644 --- a/arch/s390/kernel/swsusp.S +++ b/arch/s390/kernel/swsusp.S @@ -30,6 +30,9 @@ ENTRY(swsusp_arch_suspend) aghi %r15,-STACK_FRAME_OVERHEAD stg %r1,__SF_BACKCHAIN(%r15) + /* Store FPU registers */ + brasl %r14,save_fpu_regs + /* Deactivate DAT */ stnsm __SF_EMPTY(%r15),0xfb @@ -47,23 +50,6 @@ ENTRY(swsusp_arch_suspend) /* Store registers */ mvc 0x318(4,%r1),__SF_EMPTY(%r15) /* move prefix to lowcore */ - stfpc 0x31c(%r1) /* store fpu control */ - std 0,0x200(%r1) /* store f0 */ - std 1,0x208(%r1) /* store f1 */ - std 2,0x210(%r1) /* store f2 */ - std 3,0x218(%r1) /* store f3 */ - std 4,0x220(%r1) /* store f4 */ - std 5,0x228(%r1) /* store f5 */ - std 6,0x230(%r1) /* store f6 */ - std 7,0x238(%r1) /* store f7 */ - std 8,0x240(%r1) /* store f8 */ - std 9,0x248(%r1) /* store f9 */ - std 10,0x250(%r1) /* store f10 */ - std 11,0x258(%r1) /* store f11 */ - std 12,0x260(%r1) /* store f12 */ - std 13,0x268(%r1) /* store f13 */ - std 14,0x270(%r1) /* store f14 */ - std 15,0x278(%r1) /* store f15 */ stam %a0,%a15,0x340(%r1) /* store access registers */ stctg %c0,%c15,0x380(%r1) /* store control registers */ stmg %r0,%r15,0x280(%r1) /* store general registers */ @@ -249,24 +235,6 @@ restore_registers: lctlg %c0,%c15,0x380(%r13) /* load control registers */ lam %a0,%a15,0x340(%r13) /* load access registers */ - lfpc 0x31c(%r13) /* load fpu control */ - ld 0,0x200(%r13) /* load f0 */ - ld 1,0x208(%r13) /* load f1 */ - ld 2,0x210(%r13) /* load f2 */ - ld 3,0x218(%r13) /* load f3 */ - ld 4,0x220(%r13) /* load f4 */ - ld 5,0x228(%r13) /* load f5 */ - ld 6,0x230(%r13) /* load f6 */ - ld 7,0x238(%r13) /* load f7 */ - ld 8,0x240(%r13) /* load f8 */ - ld 9,0x248(%r13) /* load f9 */ - ld 10,0x250(%r13) /* load f10 */ - ld 11,0x258(%r13) /* load f11 */ - ld 12,0x260(%r13) /* load f12 */ - ld 13,0x268(%r13) /* load f13 */ - ld 14,0x270(%r13) /* load f14 */ - ld 15,0x278(%r13) /* load f15 */ - /* Load old stack */ lg %r15,0x2f8(%r13) diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S index f3f4a137aef6..8c56929c8d82 100644 --- a/arch/s390/kernel/syscalls.S +++ b/arch/s390/kernel/syscalls.S @@ -9,12 +9,12 @@ #define NI_SYSCALL SYSCALL(sys_ni_syscall,sys_ni_syscall) NI_SYSCALL /* 0 */ -SYSCALL(sys_exit,compat_sys_exit) +SYSCALL(sys_exit,sys_exit) SYSCALL(sys_fork,sys_fork) SYSCALL(sys_read,compat_sys_s390_read) SYSCALL(sys_write,compat_sys_s390_write) SYSCALL(sys_open,compat_sys_open) /* 5 */ -SYSCALL(sys_close,compat_sys_close) +SYSCALL(sys_close,sys_close) SYSCALL(sys_restart_syscall,sys_restart_syscall) SYSCALL(sys_creat,compat_sys_creat) SYSCALL(sys_link,compat_sys_link) @@ -35,21 +35,21 @@ SYSCALL(sys_ni_syscall,compat_sys_s390_setuid16) /* old setuid16 syscall*/ SYSCALL(sys_ni_syscall,compat_sys_s390_getuid16) /* old getuid16 syscall*/ SYSCALL(sys_ni_syscall,compat_sys_stime) /* 25 old stime syscall */ SYSCALL(sys_ptrace,compat_sys_ptrace) -SYSCALL(sys_alarm,compat_sys_alarm) +SYSCALL(sys_alarm,sys_alarm) NI_SYSCALL /* old fstat syscall */ SYSCALL(sys_pause,sys_pause) SYSCALL(sys_utime,compat_sys_utime) /* 30 */ NI_SYSCALL /* old stty syscall */ NI_SYSCALL /* old gtty syscall */ SYSCALL(sys_access,compat_sys_access) -SYSCALL(sys_nice,compat_sys_nice) +SYSCALL(sys_nice,sys_nice) NI_SYSCALL /* 35 old ftime syscall */ SYSCALL(sys_sync,sys_sync) -SYSCALL(sys_kill,compat_sys_kill) +SYSCALL(sys_kill,sys_kill) SYSCALL(sys_rename,compat_sys_rename) SYSCALL(sys_mkdir,compat_sys_mkdir) SYSCALL(sys_rmdir,compat_sys_rmdir) /* 40 */ -SYSCALL(sys_dup,compat_sys_dup) +SYSCALL(sys_dup,sys_dup) SYSCALL(sys_pipe,compat_sys_pipe) SYSCALL(sys_times,compat_sys_times) NI_SYSCALL /* old prof syscall */ @@ -65,13 +65,13 @@ NI_SYSCALL /* old lock syscall */ SYSCALL(sys_ioctl,compat_sys_ioctl) SYSCALL(sys_fcntl,compat_sys_fcntl) /* 55 */ NI_SYSCALL /* intel mpx syscall */ -SYSCALL(sys_setpgid,compat_sys_setpgid) +SYSCALL(sys_setpgid,sys_setpgid) NI_SYSCALL /* old ulimit syscall */ NI_SYSCALL /* old uname syscall */ -SYSCALL(sys_umask,compat_sys_umask) /* 60 */ +SYSCALL(sys_umask,sys_umask) /* 60 */ SYSCALL(sys_chroot,compat_sys_chroot) SYSCALL(sys_ustat,compat_sys_ustat) -SYSCALL(sys_dup2,compat_sys_dup2) +SYSCALL(sys_dup2,sys_dup2) SYSCALL(sys_getppid,sys_getppid) SYSCALL(sys_getpgrp,sys_getpgrp) /* 65 */ SYSCALL(sys_setsid,sys_setsid) @@ -102,10 +102,10 @@ SYSCALL(sys_old_mmap,compat_sys_s390_old_mmap) /* 90 */ SYSCALL(sys_munmap,compat_sys_munmap) SYSCALL(sys_truncate,compat_sys_truncate) SYSCALL(sys_ftruncate,compat_sys_ftruncate) -SYSCALL(sys_fchmod,compat_sys_fchmod) +SYSCALL(sys_fchmod,sys_fchmod) SYSCALL(sys_ni_syscall,compat_sys_s390_fchown16) /* 95 old fchown16 syscall*/ -SYSCALL(sys_getpriority,compat_sys_getpriority) -SYSCALL(sys_setpriority,compat_sys_setpriority) +SYSCALL(sys_getpriority,sys_getpriority) +SYSCALL(sys_setpriority,sys_setpriority) NI_SYSCALL /* old profil syscall */ SYSCALL(sys_statfs,compat_sys_statfs) SYSCALL(sys_fstatfs,compat_sys_fstatfs) /* 100 */ @@ -126,7 +126,7 @@ SYSCALL(sys_wait4,compat_sys_wait4) SYSCALL(sys_swapoff,compat_sys_swapoff) /* 115 */ SYSCALL(sys_sysinfo,compat_sys_sysinfo) SYSCALL(sys_s390_ipc,compat_sys_s390_ipc) -SYSCALL(sys_fsync,compat_sys_fsync) +SYSCALL(sys_fsync,sys_fsync) SYSCALL(sys_sigreturn,compat_sys_sigreturn) SYSCALL(sys_clone,compat_sys_clone) /* 120 */ SYSCALL(sys_setdomainname,compat_sys_setdomainname) @@ -140,35 +140,35 @@ SYSCALL(sys_init_module,compat_sys_init_module) SYSCALL(sys_delete_module,compat_sys_delete_module) NI_SYSCALL /* 130: old get_kernel_syms */ SYSCALL(sys_quotactl,compat_sys_quotactl) -SYSCALL(sys_getpgid,compat_sys_getpgid) -SYSCALL(sys_fchdir,compat_sys_fchdir) +SYSCALL(sys_getpgid,sys_getpgid) +SYSCALL(sys_fchdir,sys_fchdir) SYSCALL(sys_bdflush,compat_sys_bdflush) SYSCALL(sys_sysfs,compat_sys_sysfs) /* 135 */ -SYSCALL(sys_s390_personality,compat_sys_s390_personality) +SYSCALL(sys_s390_personality,sys_s390_personality) NI_SYSCALL /* for afs_syscall */ SYSCALL(sys_ni_syscall,compat_sys_s390_setfsuid16) /* old setfsuid16 syscall */ SYSCALL(sys_ni_syscall,compat_sys_s390_setfsgid16) /* old setfsgid16 syscall */ SYSCALL(sys_llseek,compat_sys_llseek) /* 140 */ SYSCALL(sys_getdents,compat_sys_getdents) SYSCALL(sys_select,compat_sys_select) -SYSCALL(sys_flock,compat_sys_flock) +SYSCALL(sys_flock,sys_flock) SYSCALL(sys_msync,compat_sys_msync) SYSCALL(sys_readv,compat_sys_readv) /* 145 */ SYSCALL(sys_writev,compat_sys_writev) -SYSCALL(sys_getsid,compat_sys_getsid) -SYSCALL(sys_fdatasync,compat_sys_fdatasync) +SYSCALL(sys_getsid,sys_getsid) +SYSCALL(sys_fdatasync,sys_fdatasync) SYSCALL(sys_sysctl,compat_sys_sysctl) SYSCALL(sys_mlock,compat_sys_mlock) /* 150 */ SYSCALL(sys_munlock,compat_sys_munlock) -SYSCALL(sys_mlockall,compat_sys_mlockall) +SYSCALL(sys_mlockall,sys_mlockall) SYSCALL(sys_munlockall,sys_munlockall) SYSCALL(sys_sched_setparam,compat_sys_sched_setparam) SYSCALL(sys_sched_getparam,compat_sys_sched_getparam) /* 155 */ SYSCALL(sys_sched_setscheduler,compat_sys_sched_setscheduler) -SYSCALL(sys_sched_getscheduler,compat_sys_sched_getscheduler) +SYSCALL(sys_sched_getscheduler,sys_sched_getscheduler) SYSCALL(sys_sched_yield,sys_sched_yield) -SYSCALL(sys_sched_get_priority_max,compat_sys_sched_get_priority_max) -SYSCALL(sys_sched_get_priority_min,compat_sys_sched_get_priority_min) /* 160 */ +SYSCALL(sys_sched_get_priority_max,sys_sched_get_priority_max) +SYSCALL(sys_sched_get_priority_min,sys_sched_get_priority_min) /* 160 */ SYSCALL(sys_sched_rr_get_interval,compat_sys_sched_rr_get_interval) SYSCALL(sys_nanosleep,compat_sys_nanosleep) SYSCALL(sys_mremap,compat_sys_mremap) @@ -211,20 +211,20 @@ SYSCALL(sys_getuid,sys_getuid) SYSCALL(sys_getgid,sys_getgid) /* 200 */ SYSCALL(sys_geteuid,sys_geteuid) SYSCALL(sys_getegid,sys_getegid) -SYSCALL(sys_setreuid,compat_sys_setreuid) -SYSCALL(sys_setregid,compat_sys_setregid) +SYSCALL(sys_setreuid,sys_setreuid) +SYSCALL(sys_setregid,sys_setregid) SYSCALL(sys_getgroups,compat_sys_getgroups) /* 205 */ SYSCALL(sys_setgroups,compat_sys_setgroups) -SYSCALL(sys_fchown,compat_sys_fchown) -SYSCALL(sys_setresuid,compat_sys_setresuid) +SYSCALL(sys_fchown,sys_fchown) +SYSCALL(sys_setresuid,sys_setresuid) SYSCALL(sys_getresuid,compat_sys_getresuid) -SYSCALL(sys_setresgid,compat_sys_setresgid) /* 210 */ +SYSCALL(sys_setresgid,sys_setresgid) /* 210 */ SYSCALL(sys_getresgid,compat_sys_getresgid) SYSCALL(sys_chown,compat_sys_chown) -SYSCALL(sys_setuid,compat_sys_setuid) -SYSCALL(sys_setgid,compat_sys_setgid) -SYSCALL(sys_setfsuid,compat_sys_setfsuid) /* 215 */ -SYSCALL(sys_setfsgid,compat_sys_setfsgid) +SYSCALL(sys_setuid,sys_setuid) +SYSCALL(sys_setgid,sys_setgid) +SYSCALL(sys_setfsuid,sys_setfsuid) /* 215 */ +SYSCALL(sys_setfsgid,sys_setfsgid) SYSCALL(sys_pivot_root,compat_sys_pivot_root) SYSCALL(sys_mincore,compat_sys_mincore) SYSCALL(sys_madvise,compat_sys_madvise) @@ -245,19 +245,19 @@ SYSCALL(sys_removexattr,compat_sys_removexattr) SYSCALL(sys_lremovexattr,compat_sys_lremovexattr) SYSCALL(sys_fremovexattr,compat_sys_fremovexattr) /* 235 */ SYSCALL(sys_gettid,sys_gettid) -SYSCALL(sys_tkill,compat_sys_tkill) +SYSCALL(sys_tkill,sys_tkill) SYSCALL(sys_futex,compat_sys_futex) SYSCALL(sys_sched_setaffinity,compat_sys_sched_setaffinity) SYSCALL(sys_sched_getaffinity,compat_sys_sched_getaffinity) /* 240 */ -SYSCALL(sys_tgkill,compat_sys_tgkill) +SYSCALL(sys_tgkill,sys_tgkill) NI_SYSCALL /* reserved for TUX */ SYSCALL(sys_io_setup,compat_sys_io_setup) SYSCALL(sys_io_destroy,compat_sys_io_destroy) SYSCALL(sys_io_getevents,compat_sys_io_getevents) /* 245 */ SYSCALL(sys_io_submit,compat_sys_io_submit) SYSCALL(sys_io_cancel,compat_sys_io_cancel) -SYSCALL(sys_exit_group,compat_sys_exit_group) -SYSCALL(sys_epoll_create,compat_sys_epoll_create) +SYSCALL(sys_exit_group,sys_exit_group) +SYSCALL(sys_epoll_create,sys_epoll_create) SYSCALL(sys_epoll_ctl,compat_sys_epoll_ctl) /* 250 */ SYSCALL(sys_epoll_wait,compat_sys_epoll_wait) SYSCALL(sys_set_tid_address,compat_sys_set_tid_address) @@ -265,8 +265,8 @@ SYSCALL(sys_fadvise64_64,compat_sys_s390_fadvise64) SYSCALL(sys_timer_create,compat_sys_timer_create) SYSCALL(sys_timer_settime,compat_sys_timer_settime) /* 255 */ SYSCALL(sys_timer_gettime,compat_sys_timer_gettime) -SYSCALL(sys_timer_getoverrun,compat_sys_timer_getoverrun) -SYSCALL(sys_timer_delete,compat_sys_timer_delete) +SYSCALL(sys_timer_getoverrun,sys_timer_getoverrun) +SYSCALL(sys_timer_delete,sys_timer_delete) SYSCALL(sys_clock_settime,compat_sys_clock_settime) SYSCALL(sys_clock_gettime,compat_sys_clock_gettime) /* 260 */ SYSCALL(sys_clock_getres,compat_sys_clock_getres) @@ -290,11 +290,11 @@ SYSCALL(sys_add_key,compat_sys_add_key) SYSCALL(sys_request_key,compat_sys_request_key) SYSCALL(sys_keyctl,compat_sys_keyctl) /* 280 */ SYSCALL(sys_waitid,compat_sys_waitid) -SYSCALL(sys_ioprio_set,compat_sys_ioprio_set) -SYSCALL(sys_ioprio_get,compat_sys_ioprio_get) +SYSCALL(sys_ioprio_set,sys_ioprio_set) +SYSCALL(sys_ioprio_get,sys_ioprio_get) SYSCALL(sys_inotify_init,sys_inotify_init) SYSCALL(sys_inotify_add_watch,compat_sys_inotify_add_watch) /* 285 */ -SYSCALL(sys_inotify_rm_watch,compat_sys_inotify_rm_watch) +SYSCALL(sys_inotify_rm_watch,sys_inotify_rm_watch) SYSCALL(sys_migrate_pages,compat_sys_migrate_pages) SYSCALL(sys_openat,compat_sys_openat) SYSCALL(sys_mkdirat,compat_sys_mkdirat) @@ -326,31 +326,31 @@ SYSCALL(sys_fallocate,compat_sys_s390_fallocate) SYSCALL(sys_utimensat,compat_sys_utimensat) /* 315 */ SYSCALL(sys_signalfd,compat_sys_signalfd) NI_SYSCALL /* 317 old sys_timer_fd */ -SYSCALL(sys_eventfd,compat_sys_eventfd) -SYSCALL(sys_timerfd_create,compat_sys_timerfd_create) +SYSCALL(sys_eventfd,sys_eventfd) +SYSCALL(sys_timerfd_create,sys_timerfd_create) SYSCALL(sys_timerfd_settime,compat_sys_timerfd_settime) /* 320 */ SYSCALL(sys_timerfd_gettime,compat_sys_timerfd_gettime) SYSCALL(sys_signalfd4,compat_sys_signalfd4) -SYSCALL(sys_eventfd2,compat_sys_eventfd2) -SYSCALL(sys_inotify_init1,compat_sys_inotify_init1) +SYSCALL(sys_eventfd2,sys_eventfd2) +SYSCALL(sys_inotify_init1,sys_inotify_init1) SYSCALL(sys_pipe2,compat_sys_pipe2) /* 325 */ -SYSCALL(sys_dup3,compat_sys_dup3) -SYSCALL(sys_epoll_create1,compat_sys_epoll_create1) +SYSCALL(sys_dup3,sys_dup3) +SYSCALL(sys_epoll_create1,sys_epoll_create1) SYSCALL(sys_preadv,compat_sys_preadv) SYSCALL(sys_pwritev,compat_sys_pwritev) SYSCALL(sys_rt_tgsigqueueinfo,compat_sys_rt_tgsigqueueinfo) /* 330 */ SYSCALL(sys_perf_event_open,compat_sys_perf_event_open) -SYSCALL(sys_fanotify_init,compat_sys_fanotify_init) +SYSCALL(sys_fanotify_init,sys_fanotify_init) SYSCALL(sys_fanotify_mark,compat_sys_fanotify_mark) SYSCALL(sys_prlimit64,compat_sys_prlimit64) SYSCALL(sys_name_to_handle_at,compat_sys_name_to_handle_at) /* 335 */ SYSCALL(sys_open_by_handle_at,compat_sys_open_by_handle_at) SYSCALL(sys_clock_adjtime,compat_sys_clock_adjtime) -SYSCALL(sys_syncfs,compat_sys_syncfs) -SYSCALL(sys_setns,compat_sys_setns) +SYSCALL(sys_syncfs,sys_syncfs) +SYSCALL(sys_setns,sys_setns) SYSCALL(sys_process_vm_readv,compat_sys_process_vm_readv) /* 340 */ SYSCALL(sys_process_vm_writev,compat_sys_process_vm_writev) -SYSCALL(sys_s390_runtime_instr,compat_sys_s390_runtime_instr) +SYSCALL(sys_s390_runtime_instr,sys_s390_runtime_instr) SYSCALL(sys_kcmp,compat_sys_kcmp) SYSCALL(sys_finit_module,compat_sys_finit_module) SYSCALL(sys_sched_setattr,compat_sys_sched_setattr) /* 345 */ @@ -363,3 +363,22 @@ SYSCALL(sys_bpf,compat_sys_bpf) SYSCALL(sys_s390_pci_mmio_write,compat_sys_s390_pci_mmio_write) SYSCALL(sys_s390_pci_mmio_read,compat_sys_s390_pci_mmio_read) SYSCALL(sys_execveat,compat_sys_execveat) +SYSCALL(sys_userfaultfd,sys_userfaultfd) /* 355 */ +SYSCALL(sys_membarrier,sys_membarrier) +SYSCALL(sys_recvmmsg,compat_sys_recvmmsg) +SYSCALL(sys_sendmmsg,compat_sys_sendmmsg) +SYSCALL(sys_socket,sys_socket) +SYSCALL(sys_socketpair,compat_sys_socketpair) /* 360 */ +SYSCALL(sys_bind,sys_bind) +SYSCALL(sys_connect,sys_connect) +SYSCALL(sys_listen,sys_listen) +SYSCALL(sys_accept4,sys_accept4) +SYSCALL(sys_getsockopt,compat_sys_getsockopt) /* 365 */ +SYSCALL(sys_setsockopt,compat_sys_setsockopt) +SYSCALL(sys_getsockname,compat_sys_getsockname) +SYSCALL(sys_getpeername,compat_sys_getpeername) +SYSCALL(sys_sendto,compat_sys_sendto) +SYSCALL(sys_sendmsg,compat_sys_sendmsg) /* 370 */ +SYSCALL(sys_recvfrom,compat_sys_recvfrom) +SYSCALL(sys_recvmsg,compat_sys_recvmsg) +SYSCALL(sys_shutdown,sys_shutdown) diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c index b9ce650e9e99..c8653435c70d 100644 --- a/arch/s390/kernel/vtime.c +++ b/arch/s390/kernel/vtime.c @@ -89,17 +89,21 @@ static int do_account_vtime(struct task_struct *tsk, int hardirq_offset) if (smp_cpu_mtid && time_after64(jiffies_64, __this_cpu_read(mt_scaling_jiffies))) { u64 cycles_new[32], *cycles_old; - u64 delta, mult, div; + u64 delta, fac, mult, div; cycles_old = this_cpu_ptr(mt_cycles); if (stcctm5(smp_cpu_mtid + 1, cycles_new) < 2) { + fac = 1; mult = div = 0; for (i = 0; i <= smp_cpu_mtid; i++) { delta = cycles_new[i] - cycles_old[i]; - mult += delta; - div += (i + 1) * delta; + div += delta; + mult *= i + 1; + mult += delta * fac; + fac *= i + 1; } - if (mult > 0) { + div *= fac; + if (div > 0) { /* Update scaling factor */ __this_cpu_write(mt_scaling_mult, mult); __this_cpu_write(mt_scaling_div, div); diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index 155162ea0e00..ab5f1d447ef9 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -86,6 +86,16 @@ extern u64 asmlinkage efi_call(void *fp, ...); extern void __iomem *__init efi_ioremap(unsigned long addr, unsigned long size, u32 type, u64 attribute); +/* + * CONFIG_KASAN may redefine memset to __memset. __memset function is present + * only in kernel binary. Since the EFI stub linked into a separate binary it + * doesn't have __memset(). So we should use standard memset from + * arch/x86/boot/compressed/string.c. The same applies to memcpy and memmove. + */ +#undef memcpy +#undef memset +#undef memmove + #endif /* CONFIG_X86_32 */ extern struct efi_scratch efi_scratch; diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 349f80a82b82..2beee0382088 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -40,6 +40,7 @@ #define KVM_PIO_PAGE_OFFSET 1 #define KVM_COALESCED_MMIO_PAGE_OFFSET 2 +#define KVM_HALT_POLL_NS_DEFAULT 500000 #define KVM_IRQCHIP_NUM_PINS KVM_IOAPIC_NUM_PINS diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index c1c0a1c14344..b98b471a3b7e 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -331,6 +331,7 @@ /* C1E active bits in int pending message */ #define K8_INTP_C1E_ACTIVE_MASK 0x18000000 #define MSR_K8_TSEG_ADDR 0xc0010112 +#define MSR_K8_TSEG_MASK 0xc0010113 #define K8_MTRRFIXRANGE_DRAM_ENABLE 0x00040000 /* MtrrFixDramEn bit */ #define K8_MTRRFIXRANGE_DRAM_MODIFY 0x00080000 /* MtrrFixDramModEn bit */ #define K8_MTRR_RDMEM_WRMEM_MASK 0x18181818 /* Mask: RdMem|WrMem */ diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 69088a1ba509..ff606f507913 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -3322,7 +3322,7 @@ walk_shadow_page_get_mmio_spte(struct kvm_vcpu *vcpu, u64 addr, u64 *sptep) break; reserved |= is_shadow_zero_bits_set(&vcpu->arch.mmu, spte, - leaf); + iterator.level); } walk_shadow_page_lockless_end(vcpu); @@ -3614,7 +3614,7 @@ static void __reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, struct rsvd_bits_validate *rsvd_check, int maxphyaddr, int level, bool nx, bool gbpages, - bool pse) + bool pse, bool amd) { u64 exb_bit_rsvd = 0; u64 gbpages_bit_rsvd = 0; @@ -3631,7 +3631,7 @@ __reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, * Non-leaf PML4Es and PDPEs reserve bit 8 (which would be the G bit for * leaf entries) on AMD CPUs only. */ - if (guest_cpuid_is_amd(vcpu)) + if (amd) nonleaf_bit8_rsvd = rsvd_bits(8, 8); switch (level) { @@ -3699,7 +3699,7 @@ static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, __reset_rsvds_bits_mask(vcpu, &context->guest_rsvd_check, cpuid_maxphyaddr(vcpu), context->root_level, context->nx, guest_cpuid_has_gbpages(vcpu), - is_pse(vcpu)); + is_pse(vcpu), guest_cpuid_is_amd(vcpu)); } static void @@ -3749,13 +3749,24 @@ static void reset_rsvds_bits_mask_ept(struct kvm_vcpu *vcpu, void reset_shadow_zero_bits_mask(struct kvm_vcpu *vcpu, struct kvm_mmu *context) { + /* + * Passing "true" to the last argument is okay; it adds a check + * on bit 8 of the SPTEs which KVM doesn't use anyway. + */ __reset_rsvds_bits_mask(vcpu, &context->shadow_zero_check, boot_cpu_data.x86_phys_bits, context->shadow_root_level, context->nx, - guest_cpuid_has_gbpages(vcpu), is_pse(vcpu)); + guest_cpuid_has_gbpages(vcpu), is_pse(vcpu), + true); } EXPORT_SYMBOL_GPL(reset_shadow_zero_bits_mask); +static inline bool boot_cpu_is_amd(void) +{ + WARN_ON_ONCE(!tdp_enabled); + return shadow_x_mask == 0; +} + /* * the direct page table on host, use as much mmu features as * possible, however, kvm currently does not do execution-protection. @@ -3764,11 +3775,11 @@ static void reset_tdp_shadow_zero_bits_mask(struct kvm_vcpu *vcpu, struct kvm_mmu *context) { - if (guest_cpuid_is_amd(vcpu)) + if (boot_cpu_is_amd()) __reset_rsvds_bits_mask(vcpu, &context->shadow_zero_check, boot_cpu_data.x86_phys_bits, context->shadow_root_level, false, - cpu_has_gbpages, true); + cpu_has_gbpages, true, true); else __reset_rsvds_bits_mask_ept(&context->shadow_zero_check, boot_cpu_data.x86_phys_bits, diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index fdb8cb63a6c0..94b7d15db3fc 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -202,6 +202,7 @@ module_param(npt, int, S_IRUGO); static int nested = true; module_param(nested, int, S_IRUGO); +static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0); static void svm_flush_tlb(struct kvm_vcpu *vcpu); static void svm_complete_interrupts(struct vcpu_svm *svm); @@ -1263,7 +1264,8 @@ static void init_vmcb(struct vcpu_svm *svm, bool init_event) * svm_set_cr0() sets PG and WP and clears NW and CD on save->cr0. * It also updates the guest-visible cr0 value. */ - (void)kvm_set_cr0(&svm->vcpu, X86_CR0_NW | X86_CR0_CD | X86_CR0_ET); + svm_set_cr0(&svm->vcpu, X86_CR0_NW | X86_CR0_CD | X86_CR0_ET); + kvm_mmu_reset_context(&svm->vcpu); save->cr4 = X86_CR4_PAE; /* rdx = ?? */ diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 6bbb0dfb99d0..991466bf8dee 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -2190,6 +2190,8 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info) case MSR_IA32_LASTINTFROMIP: case MSR_IA32_LASTINTTOIP: case MSR_K8_SYSCFG: + case MSR_K8_TSEG_ADDR: + case MSR_K8_TSEG_MASK: case MSR_K7_HWCR: case MSR_VM_HSAVE_PA: case MSR_K8_INT_PENDING_MSG: diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c index 09d3afc0a181..dc78a4a9a466 100644 --- a/arch/x86/pci/common.c +++ b/arch/x86/pci/common.c @@ -166,6 +166,7 @@ void pcibios_fixup_bus(struct pci_bus *b) { struct pci_dev *dev; + pci_read_bridge_bases(b); list_for_each_entry(dev, &b->devices, bus_list) pcibios_fixup_device_resources(dev); } diff --git a/arch/xtensa/kernel/pci.c b/arch/xtensa/kernel/pci.c index d27b4dcf221f..b848cc3dc913 100644 --- a/arch/xtensa/kernel/pci.c +++ b/arch/xtensa/kernel/pci.c @@ -210,6 +210,10 @@ subsys_initcall(pcibios_init); void pcibios_fixup_bus(struct pci_bus *bus) { + if (bus->parent) { + /* This is a subordinate bridge */ + pci_read_bridge_bases(bus); + } } void pcibios_set_master(struct pci_dev *dev) diff --git a/drivers/firmware/efi/libstub/efistub.h b/drivers/firmware/efi/libstub/efistub.h index e334a01cf92f..6b6548fda089 100644 --- a/drivers/firmware/efi/libstub/efistub.h +++ b/drivers/firmware/efi/libstub/efistub.h @@ -5,10 +5,6 @@ /* error code which can't be mistaken for valid address */ #define EFI_ERROR (~0UL) -#undef memcpy -#undef memset -#undef memmove - void efi_char16_printk(efi_system_table_t *, efi_char16_t *); efi_status_t efi_open_volume(efi_system_table_t *sys_table_arg, void *__image, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 668939a14206..6647fb26ef25 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -82,6 +82,7 @@ extern int amdgpu_vm_block_size; extern int amdgpu_enable_scheduler; extern int amdgpu_sched_jobs; extern int amdgpu_sched_hw_submission; +extern int amdgpu_enable_semaphores; #define AMDGPU_WAIT_IDLE_TIMEOUT_IN_MS 3000 #define AMDGPU_MAX_USEC_TIMEOUT 100000 /* 100 ms */ @@ -432,7 +433,7 @@ int amdgpu_fence_driver_init(struct amdgpu_device *adev); void amdgpu_fence_driver_fini(struct amdgpu_device *adev); void amdgpu_fence_driver_force_completion(struct amdgpu_device *adev); -void amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring); +int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring); int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring, struct amdgpu_irq_src *irq_src, unsigned irq_type); @@ -890,7 +891,7 @@ struct amdgpu_ring { struct amdgpu_device *adev; const struct amdgpu_ring_funcs *funcs; struct amdgpu_fence_driver fence_drv; - struct amd_gpu_scheduler *scheduler; + struct amd_gpu_scheduler sched; spinlock_t fence_lock; struct mutex *ring_lock; @@ -1201,8 +1202,6 @@ struct amdgpu_gfx { struct amdgpu_irq_src priv_inst_irq; /* gfx status */ uint32_t gfx_current_status; - /* sync signal for const engine */ - unsigned ce_sync_offs; /* ce ram size*/ unsigned ce_ram_size; }; @@ -1274,8 +1273,10 @@ struct amdgpu_job { uint32_t num_ibs; struct mutex job_lock; struct amdgpu_user_fence uf; - int (*free_job)(struct amdgpu_job *sched_job); + int (*free_job)(struct amdgpu_job *job); }; +#define to_amdgpu_job(sched_job) \ + container_of((sched_job), struct amdgpu_job, base) static inline u32 amdgpu_get_ib_value(struct amdgpu_cs_parser *p, uint32_t ib_idx, int idx) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index 496ed2192eba..84d68d658f8a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -183,7 +183,7 @@ int alloc_gtt_mem(struct kgd_dev *kgd, size_t size, return -ENOMEM; r = amdgpu_bo_create(rdev, size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_GTT, - AMDGPU_GEM_CREATE_CPU_GTT_USWC, NULL, &(*mem)->bo); + AMDGPU_GEM_CREATE_CPU_GTT_USWC, NULL, NULL, &(*mem)->bo); if (r) { dev_err(rdev->dev, "failed to allocate BO for amdkfd (%d)\n", r); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c index 98d59ee640ce..cd639c362df3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_benchmark.c @@ -79,7 +79,8 @@ static void amdgpu_benchmark_move(struct amdgpu_device *adev, unsigned size, int time; n = AMDGPU_BENCHMARK_ITERATIONS; - r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, sdomain, 0, NULL, &sobj); + r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, sdomain, 0, NULL, + NULL, &sobj); if (r) { goto out_cleanup; } @@ -91,7 +92,8 @@ static void amdgpu_benchmark_move(struct amdgpu_device *adev, unsigned size, if (r) { goto out_cleanup; } - r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, ddomain, 0, NULL, &dobj); + r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, ddomain, 0, NULL, + NULL, &dobj); if (r) { goto out_cleanup; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c index 6b1243f9f86d..1c3fc99c5465 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c @@ -86,7 +86,7 @@ static int amdgpu_cgs_gmap_kmem(void *cgs_device, void *kmem, struct sg_table *sg = drm_prime_pages_to_sg(&kmem_page, npages); ret = amdgpu_bo_create(adev, size, PAGE_SIZE, false, - AMDGPU_GEM_DOMAIN_GTT, 0, sg, &bo); + AMDGPU_GEM_DOMAIN_GTT, 0, sg, NULL, &bo); if (ret) return ret; ret = amdgpu_bo_reserve(bo, false); @@ -197,7 +197,8 @@ static int amdgpu_cgs_alloc_gpu_mem(void *cgs_device, ret = amdgpu_bo_create_restricted(adev, size, PAGE_SIZE, true, domain, flags, - NULL, &placement, &obj); + NULL, &placement, NULL, + &obj); if (ret) { DRM_ERROR("(%d) bo create failed\n", ret); return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 3b355aeb62fd..749420f1ea6f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -154,42 +154,41 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data) { union drm_amdgpu_cs *cs = data; uint64_t *chunk_array_user; - uint64_t *chunk_array = NULL; + uint64_t *chunk_array; struct amdgpu_fpriv *fpriv = p->filp->driver_priv; unsigned size, i; - int r = 0; + int ret; - if (!cs->in.num_chunks) - goto out; + if (cs->in.num_chunks == 0) + return 0; + + chunk_array = kmalloc_array(cs->in.num_chunks, sizeof(uint64_t), GFP_KERNEL); + if (!chunk_array) + return -ENOMEM; p->ctx = amdgpu_ctx_get(fpriv, cs->in.ctx_id); if (!p->ctx) { - r = -EINVAL; - goto out; + ret = -EINVAL; + goto free_chunk; } + p->bo_list = amdgpu_bo_list_get(fpriv, cs->in.bo_list_handle); /* get chunks */ INIT_LIST_HEAD(&p->validated); - chunk_array = kmalloc_array(cs->in.num_chunks, sizeof(uint64_t), GFP_KERNEL); - if (chunk_array == NULL) { - r = -ENOMEM; - goto out; - } - chunk_array_user = (uint64_t __user *)(cs->in.chunks); if (copy_from_user(chunk_array, chunk_array_user, sizeof(uint64_t)*cs->in.num_chunks)) { - r = -EFAULT; - goto out; + ret = -EFAULT; + goto put_bo_list; } p->nchunks = cs->in.num_chunks; p->chunks = kmalloc_array(p->nchunks, sizeof(struct amdgpu_cs_chunk), GFP_KERNEL); - if (p->chunks == NULL) { - r = -ENOMEM; - goto out; + if (!p->chunks) { + ret = -ENOMEM; + goto put_bo_list; } for (i = 0; i < p->nchunks; i++) { @@ -200,8 +199,9 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data) chunk_ptr = (void __user *)chunk_array[i]; if (copy_from_user(&user_chunk, chunk_ptr, sizeof(struct drm_amdgpu_cs_chunk))) { - r = -EFAULT; - goto out; + ret = -EFAULT; + i--; + goto free_partial_kdata; } p->chunks[i].chunk_id = user_chunk.chunk_id; p->chunks[i].length_dw = user_chunk.length_dw; @@ -212,13 +212,14 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data) p->chunks[i].kdata = drm_malloc_ab(size, sizeof(uint32_t)); if (p->chunks[i].kdata == NULL) { - r = -ENOMEM; - goto out; + ret = -ENOMEM; + i--; + goto free_partial_kdata; } size *= sizeof(uint32_t); if (copy_from_user(p->chunks[i].kdata, cdata, size)) { - r = -EFAULT; - goto out; + ret = -EFAULT; + goto free_partial_kdata; } switch (p->chunks[i].chunk_id) { @@ -238,15 +239,15 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data) gobj = drm_gem_object_lookup(p->adev->ddev, p->filp, handle); if (gobj == NULL) { - r = -EINVAL; - goto out; + ret = -EINVAL; + goto free_partial_kdata; } p->uf.bo = gem_to_amdgpu_bo(gobj); p->uf.offset = fence_data->offset; } else { - r = -EINVAL; - goto out; + ret = -EINVAL; + goto free_partial_kdata; } break; @@ -254,19 +255,35 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data) break; default: - r = -EINVAL; - goto out; + ret = -EINVAL; + goto free_partial_kdata; } } p->ibs = kcalloc(p->num_ibs, sizeof(struct amdgpu_ib), GFP_KERNEL); - if (!p->ibs) - r = -ENOMEM; + if (!p->ibs) { + ret = -ENOMEM; + goto free_all_kdata; + } -out: kfree(chunk_array); - return r; + return 0; + +free_all_kdata: + i = p->nchunks - 1; +free_partial_kdata: + for (; i >= 0; i--) + drm_free_large(p->chunks[i].kdata); + kfree(p->chunks); +put_bo_list: + if (p->bo_list) + amdgpu_bo_list_put(p->bo_list); + amdgpu_ctx_put(p->ctx); +free_chunk: + kfree(chunk_array); + + return ret; } /* Returns how many bytes TTM can move per IB. @@ -321,25 +338,17 @@ static u64 amdgpu_cs_get_threshold_for_moves(struct amdgpu_device *adev) return max(bytes_moved_threshold, 1024*1024ull); } -int amdgpu_cs_list_validate(struct amdgpu_cs_parser *p) +int amdgpu_cs_list_validate(struct amdgpu_device *adev, + struct amdgpu_vm *vm, + struct list_head *validated) { - struct amdgpu_fpriv *fpriv = p->filp->driver_priv; - struct amdgpu_vm *vm = &fpriv->vm; - struct amdgpu_device *adev = p->adev; struct amdgpu_bo_list_entry *lobj; - struct list_head duplicates; struct amdgpu_bo *bo; u64 bytes_moved = 0, initial_bytes_moved; u64 bytes_moved_threshold = amdgpu_cs_get_threshold_for_moves(adev); int r; - INIT_LIST_HEAD(&duplicates); - r = ttm_eu_reserve_buffers(&p->ticket, &p->validated, true, &duplicates); - if (unlikely(r != 0)) { - return r; - } - - list_for_each_entry(lobj, &p->validated, tv.head) { + list_for_each_entry(lobj, validated, tv.head) { bo = lobj->robj; if (!bo->pin_count) { u32 domain = lobj->prefered_domains; @@ -373,7 +382,6 @@ int amdgpu_cs_list_validate(struct amdgpu_cs_parser *p) domain = lobj->allowed_domains; goto retry; } - ttm_eu_backoff_reservation(&p->ticket, &p->validated); return r; } } @@ -386,6 +394,7 @@ static int amdgpu_cs_parser_relocs(struct amdgpu_cs_parser *p) { struct amdgpu_fpriv *fpriv = p->filp->driver_priv; struct amdgpu_cs_buckets buckets; + struct list_head duplicates; bool need_mmap_lock = false; int i, r; @@ -405,8 +414,22 @@ static int amdgpu_cs_parser_relocs(struct amdgpu_cs_parser *p) if (need_mmap_lock) down_read(¤t->mm->mmap_sem); - r = amdgpu_cs_list_validate(p); + INIT_LIST_HEAD(&duplicates); + r = ttm_eu_reserve_buffers(&p->ticket, &p->validated, true, &duplicates); + if (unlikely(r != 0)) + goto error_reserve; + + r = amdgpu_cs_list_validate(p->adev, &fpriv->vm, &p->validated); + if (r) + goto error_validate; + + r = amdgpu_cs_list_validate(p->adev, &fpriv->vm, &duplicates); + +error_validate: + if (r) + ttm_eu_backoff_reservation(&p->ticket, &p->validated); +error_reserve: if (need_mmap_lock) up_read(¤t->mm->mmap_sem); @@ -772,15 +795,15 @@ static int amdgpu_cs_dependencies(struct amdgpu_device *adev, return 0; } -static int amdgpu_cs_free_job(struct amdgpu_job *sched_job) +static int amdgpu_cs_free_job(struct amdgpu_job *job) { int i; - if (sched_job->ibs) - for (i = 0; i < sched_job->num_ibs; i++) - amdgpu_ib_free(sched_job->adev, &sched_job->ibs[i]); - kfree(sched_job->ibs); - if (sched_job->uf.bo) - drm_gem_object_unreference_unlocked(&sched_job->uf.bo->gem_base); + if (job->ibs) + for (i = 0; i < job->num_ibs; i++) + amdgpu_ib_free(job->adev, &job->ibs[i]); + kfree(job->ibs); + if (job->uf.bo) + drm_gem_object_unreference_unlocked(&job->uf.bo->gem_base); return 0; } @@ -804,7 +827,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) r = amdgpu_cs_parser_init(parser, data); if (r) { DRM_ERROR("Failed to initialize parser !\n"); - amdgpu_cs_parser_fini(parser, r, false); + kfree(parser); up_read(&adev->exclusive_lock); r = amdgpu_cs_handle_lockup(adev, r); return r; @@ -842,7 +865,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) job = kzalloc(sizeof(struct amdgpu_job), GFP_KERNEL); if (!job) return -ENOMEM; - job->base.sched = ring->scheduler; + job->base.sched = &ring->sched; job->base.s_entity = &parser->ctx->rings[ring->idx].entity; job->adev = parser->adev; job->ibs = parser->ibs; @@ -857,7 +880,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) job->free_job = amdgpu_cs_free_job; mutex_lock(&job->job_lock); - r = amd_sched_entity_push_job((struct amd_sched_job *)job); + r = amd_sched_entity_push_job(&job->base); if (r) { mutex_unlock(&job->job_lock); amdgpu_cs_free_job(job); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index 20cbc4eb5a6f..e0b80ccdfe8a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -43,10 +43,10 @@ int amdgpu_ctx_init(struct amdgpu_device *adev, bool kernel, for (i = 0; i < adev->num_rings; i++) { struct amd_sched_rq *rq; if (kernel) - rq = &adev->rings[i]->scheduler->kernel_rq; + rq = &adev->rings[i]->sched.kernel_rq; else - rq = &adev->rings[i]->scheduler->sched_rq; - r = amd_sched_entity_init(adev->rings[i]->scheduler, + rq = &adev->rings[i]->sched.sched_rq; + r = amd_sched_entity_init(&adev->rings[i]->sched, &ctx->rings[i].entity, rq, amdgpu_sched_jobs); if (r) @@ -55,7 +55,7 @@ int amdgpu_ctx_init(struct amdgpu_device *adev, bool kernel, if (i < adev->num_rings) { for (j = 0; j < i; j++) - amd_sched_entity_fini(adev->rings[j]->scheduler, + amd_sched_entity_fini(&adev->rings[j]->sched, &ctx->rings[j].entity); kfree(ctx); return r; @@ -75,7 +75,7 @@ void amdgpu_ctx_fini(struct amdgpu_ctx *ctx) if (amdgpu_enable_scheduler) { for (i = 0; i < adev->num_rings; i++) - amd_sched_entity_fini(adev->rings[i]->scheduler, + amd_sched_entity_fini(&adev->rings[i]->sched, &ctx->rings[i].entity); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 6ff6ae945794..6068d8207d10 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -246,7 +246,7 @@ static int amdgpu_vram_scratch_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, AMDGPU_GPU_PAGE_SIZE, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->vram_scratch.robj); + NULL, NULL, &adev->vram_scratch.robj); if (r) { return r; } @@ -449,7 +449,8 @@ static int amdgpu_wb_init(struct amdgpu_device *adev) if (adev->wb.wb_obj == NULL) { r = amdgpu_bo_create(adev, AMDGPU_MAX_WB * 4, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GTT, 0, NULL, &adev->wb.wb_obj); + AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL, + &adev->wb.wb_obj); if (r) { dev_warn(adev->dev, "(%d) create WB bo failed\n", r); return r; @@ -1650,9 +1651,11 @@ int amdgpu_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon) drm_kms_helper_poll_disable(dev); /* turn off display hw */ + drm_modeset_lock_all(dev); list_for_each_entry(connector, &dev->mode_config.connector_list, head) { drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF); } + drm_modeset_unlock_all(dev); /* unpin the front buffers */ list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { @@ -1747,9 +1750,11 @@ int amdgpu_resume_kms(struct drm_device *dev, bool resume, bool fbcon) if (fbcon) { drm_helper_resume_force_mode(dev); /* turn on display hw */ + drm_modeset_lock_all(dev); list_for_each_entry(connector, &dev->mode_config.connector_list, head) { drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON); } + drm_modeset_unlock_all(dev); } drm_kms_helper_poll_enable(dev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 0fcc0bd1622c..adb48353f2e1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -79,6 +79,7 @@ int amdgpu_exp_hw_support = 0; int amdgpu_enable_scheduler = 0; int amdgpu_sched_jobs = 16; int amdgpu_sched_hw_submission = 2; +int amdgpu_enable_semaphores = 1; MODULE_PARM_DESC(vramlimit, "Restrict VRAM for testing, in megabytes"); module_param_named(vramlimit, amdgpu_vram_limit, int, 0600); @@ -152,6 +153,9 @@ module_param_named(sched_jobs, amdgpu_sched_jobs, int, 0444); MODULE_PARM_DESC(sched_hw_submission, "the max number of HW submissions (default 2)"); module_param_named(sched_hw_submission, amdgpu_sched_hw_submission, int, 0444); +MODULE_PARM_DESC(enable_semaphores, "Enable semaphores (1 = enable (default), 0 = disable)"); +module_param_named(enable_semaphores, amdgpu_enable_semaphores, int, 0644); + static struct pci_device_id pciidlist[] = { #ifdef CONFIG_DRM_AMDGPU_CIK /* Kaveri */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index 1be2bd6d07ea..b3fc26c59787 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -609,9 +609,9 @@ int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring, * Init the fence driver for the requested ring (all asics). * Helper function for amdgpu_fence_driver_init(). */ -void amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring) +int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring) { - int i; + int i, r; ring->fence_drv.cpu_addr = NULL; ring->fence_drv.gpu_addr = 0; @@ -625,15 +625,19 @@ void amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring) amdgpu_fence_check_lockup); ring->fence_drv.ring = ring; + init_waitqueue_head(&ring->fence_drv.fence_queue); + if (amdgpu_enable_scheduler) { - ring->scheduler = amd_sched_create(&amdgpu_sched_ops, - ring->idx, - amdgpu_sched_hw_submission, - (void *)ring->adev); - if (!ring->scheduler) - DRM_ERROR("Failed to create scheduler on ring %d.\n", - ring->idx); + r = amd_sched_init(&ring->sched, &amdgpu_sched_ops, + amdgpu_sched_hw_submission, ring->name); + if (r) { + DRM_ERROR("Failed to create scheduler on ring %s.\n", + ring->name); + return r; + } } + + return 0; } /** @@ -681,8 +685,7 @@ void amdgpu_fence_driver_fini(struct amdgpu_device *adev) wake_up_all(&ring->fence_drv.fence_queue); amdgpu_irq_put(adev, ring->fence_drv.irq_src, ring->fence_drv.irq_type); - if (ring->scheduler) - amd_sched_destroy(ring->scheduler); + amd_sched_fini(&ring->sched); ring->fence_drv.initialized = false; } mutex_unlock(&adev->ring_lock); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c index cbd3a486c5c2..7312d729d300 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c @@ -127,7 +127,7 @@ int amdgpu_gart_table_vram_alloc(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, adev->gart.table_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->gart.robj); + NULL, NULL, &adev->gart.robj); if (r) { return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 5839fab374bf..7297ca3a0ba7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -69,7 +69,8 @@ int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size, } } retry: - r = amdgpu_bo_create(adev, size, alignment, kernel, initial_domain, flags, NULL, &robj); + r = amdgpu_bo_create(adev, size, alignment, kernel, initial_domain, + flags, NULL, NULL, &robj); if (r) { if (r != -ERESTARTSYS) { if (initial_domain == AMDGPU_GEM_DOMAIN_VRAM) { @@ -426,6 +427,10 @@ int amdgpu_gem_metadata_ioctl(struct drm_device *dev, void *data, &args->data.data_size_bytes, &args->data.flags); } else if (args->op == AMDGPU_GEM_METADATA_OP_SET_METADATA) { + if (args->data.data_size_bytes > sizeof(args->data.data)) { + r = -EINVAL; + goto unreserve; + } r = amdgpu_bo_set_tiling_flags(robj, args->data.tiling_info); if (!r) r = amdgpu_bo_set_metadata(robj, args->data.data, @@ -433,6 +438,7 @@ int amdgpu_gem_metadata_ioctl(struct drm_device *dev, void *data, args->data.flags); } +unreserve: amdgpu_bo_unreserve(robj); out: drm_gem_object_unreference_unlocked(gobj); @@ -454,11 +460,12 @@ static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev, struct ttm_validate_buffer tv, *entry; struct amdgpu_bo_list_entry *vm_bos; struct ww_acquire_ctx ticket; - struct list_head list; + struct list_head list, duplicates; unsigned domain; int r; INIT_LIST_HEAD(&list); + INIT_LIST_HEAD(&duplicates); tv.bo = &bo_va->bo->tbo; tv.shared = true; @@ -468,7 +475,8 @@ static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev, if (!vm_bos) return; - r = ttm_eu_reserve_buffers(&ticket, &list, true, NULL); + /* Provide duplicates to avoid -EALREADY */ + r = ttm_eu_reserve_buffers(&ticket, &list, true, &duplicates); if (r) goto error_free; @@ -651,7 +659,7 @@ int amdgpu_mode_dumb_create(struct drm_file *file_priv, int r; args->pitch = amdgpu_align_pitch(adev, args->width, args->bpp, 0) * ((args->bpp + 1) / 8); - args->size = args->pitch * args->height; + args->size = (u64)args->pitch * args->height; args->size = ALIGN(args->size, PAGE_SIZE); r = amdgpu_gem_object_create(adev, args->size, 0, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c index 5c8a803acedc..534fc04e80fd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c @@ -43,7 +43,7 @@ static int amdgpu_ih_ring_alloc(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, adev->irq.ih.ring_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_GTT, 0, - NULL, &adev->irq.ih.ring_obj); + NULL, NULL, &adev->irq.ih.ring_obj); if (r) { DRM_ERROR("amdgpu: failed to create ih ring buffer (%d).\n", r); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c index 0aba8e9bc8a0..7c42ff670080 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c @@ -140,7 +140,7 @@ void amdgpu_irq_preinstall(struct drm_device *dev) */ int amdgpu_irq_postinstall(struct drm_device *dev) { - dev->max_vblank_count = 0x001fffff; + dev->max_vblank_count = 0x00ffffff; return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 22367939ebf1..8c735f544b66 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -390,7 +390,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file min((size_t)size, sizeof(vram_gtt))) ? -EFAULT : 0; } case AMDGPU_INFO_READ_MMR_REG: { - unsigned n, alloc_size = info->read_mmr_reg.count * 4; + unsigned n, alloc_size; uint32_t *regs; unsigned se_num = (info->read_mmr_reg.instance >> AMDGPU_INFO_MMR_SE_INDEX_SHIFT) & @@ -406,9 +406,10 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file if (sh_num == AMDGPU_INFO_MMR_SH_INDEX_MASK) sh_num = 0xffffffff; - regs = kmalloc(alloc_size, GFP_KERNEL); + regs = kmalloc_array(info->read_mmr_reg.count, sizeof(*regs), GFP_KERNEL); if (!regs) return -ENOMEM; + alloc_size = info->read_mmr_reg.count * sizeof(*regs); for (i = 0; i < info->read_mmr_reg.count; i++) if (amdgpu_asic_read_register(adev, se_num, sh_num, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index 08b09d55b96f..1a7708f365f3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -215,6 +215,7 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev, bool kernel, u32 domain, u64 flags, struct sg_table *sg, struct ttm_placement *placement, + struct reservation_object *resv, struct amdgpu_bo **bo_ptr) { struct amdgpu_bo *bo; @@ -261,7 +262,7 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev, /* Kernel allocation are uninterruptible */ r = ttm_bo_init(&adev->mman.bdev, &bo->tbo, size, type, &bo->placement, page_align, !kernel, NULL, - acc_size, sg, NULL, &amdgpu_ttm_bo_destroy); + acc_size, sg, resv, &amdgpu_ttm_bo_destroy); if (unlikely(r != 0)) { return r; } @@ -275,7 +276,9 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev, int amdgpu_bo_create(struct amdgpu_device *adev, unsigned long size, int byte_align, bool kernel, u32 domain, u64 flags, - struct sg_table *sg, struct amdgpu_bo **bo_ptr) + struct sg_table *sg, + struct reservation_object *resv, + struct amdgpu_bo **bo_ptr) { struct ttm_placement placement = {0}; struct ttm_place placements[AMDGPU_GEM_DOMAIN_MAX + 1]; @@ -286,11 +289,9 @@ int amdgpu_bo_create(struct amdgpu_device *adev, amdgpu_ttm_placement_init(adev, &placement, placements, domain, flags); - return amdgpu_bo_create_restricted(adev, size, byte_align, - kernel, domain, flags, - sg, - &placement, - bo_ptr); + return amdgpu_bo_create_restricted(adev, size, byte_align, kernel, + domain, flags, sg, &placement, + resv, bo_ptr); } int amdgpu_bo_kmap(struct amdgpu_bo *bo, void **ptr) @@ -535,12 +536,10 @@ int amdgpu_bo_set_metadata (struct amdgpu_bo *bo, void *metadata, if (metadata == NULL) return -EINVAL; - buffer = kzalloc(metadata_size, GFP_KERNEL); + buffer = kmemdup(metadata, metadata_size, GFP_KERNEL); if (buffer == NULL) return -ENOMEM; - memcpy(buffer, metadata, metadata_size); - kfree(bo->metadata); bo->metadata_flags = flags; bo->metadata = buffer; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h index 6ea18dcec561..3c2ff4567798 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h @@ -129,12 +129,14 @@ int amdgpu_bo_create(struct amdgpu_device *adev, unsigned long size, int byte_align, bool kernel, u32 domain, u64 flags, struct sg_table *sg, + struct reservation_object *resv, struct amdgpu_bo **bo_ptr); int amdgpu_bo_create_restricted(struct amdgpu_device *adev, unsigned long size, int byte_align, bool kernel, u32 domain, u64 flags, struct sg_table *sg, struct ttm_placement *placement, + struct reservation_object *resv, struct amdgpu_bo **bo_ptr); int amdgpu_bo_kmap(struct amdgpu_bo *bo, void **ptr); void amdgpu_bo_kunmap(struct amdgpu_bo *bo); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c index d9652fe32d6a..59f735a933a9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c @@ -61,12 +61,15 @@ struct drm_gem_object *amdgpu_gem_prime_import_sg_table(struct drm_device *dev, struct dma_buf_attachment *attach, struct sg_table *sg) { + struct reservation_object *resv = attach->dmabuf->resv; struct amdgpu_device *adev = dev->dev_private; struct amdgpu_bo *bo; int ret; + ww_mutex_lock(&resv->lock, NULL); ret = amdgpu_bo_create(adev, attach->dmabuf->size, PAGE_SIZE, false, - AMDGPU_GEM_DOMAIN_GTT, 0, sg, &bo); + AMDGPU_GEM_DOMAIN_GTT, 0, sg, resv, &bo); + ww_mutex_unlock(&resv->lock); if (ret) return ERR_PTR(ret); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index 9bec91484c24..30dce235ddeb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -357,11 +357,11 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, ring->adev = adev; ring->idx = adev->num_rings++; adev->rings[ring->idx] = ring; - amdgpu_fence_driver_init_ring(ring); + r = amdgpu_fence_driver_init_ring(ring); + if (r) + return r; } - init_waitqueue_head(&ring->fence_drv.fence_queue); - r = amdgpu_wb_get(adev, &ring->rptr_offs); if (r) { dev_err(adev->dev, "(%d) ring rptr_offs wb alloc failed\n", r); @@ -407,7 +407,7 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, if (ring->ring_obj == NULL) { r = amdgpu_bo_create(adev, ring->ring_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_GTT, 0, - NULL, &ring->ring_obj); + NULL, NULL, &ring->ring_obj); if (r) { dev_err(adev->dev, "(%d) ring create failed\n", r); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c index 74dad270362c..e90712443fe9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c @@ -64,8 +64,8 @@ int amdgpu_sa_bo_manager_init(struct amdgpu_device *adev, INIT_LIST_HEAD(&sa_manager->flist[i]); } - r = amdgpu_bo_create(adev, size, align, true, - domain, 0, NULL, &sa_manager->bo); + r = amdgpu_bo_create(adev, size, align, true, domain, + 0, NULL, NULL, &sa_manager->bo); if (r) { dev_err(adev->dev, "(%d) failed to allocate bo for manager\n", r); return r; @@ -145,8 +145,13 @@ static uint32_t amdgpu_sa_get_ring_from_fence(struct fence *f) struct amd_sched_fence *s_fence; s_fence = to_amd_sched_fence(f); - if (s_fence) - return s_fence->scheduler->ring_id; + if (s_fence) { + struct amdgpu_ring *ring; + + ring = container_of(s_fence->sched, struct amdgpu_ring, sched); + return ring->idx; + } + a_fence = to_amdgpu_fence(f); if (a_fence) return a_fence->ring->idx; @@ -412,6 +417,26 @@ void amdgpu_sa_bo_free(struct amdgpu_device *adev, struct amdgpu_sa_bo **sa_bo, } #if defined(CONFIG_DEBUG_FS) + +static void amdgpu_sa_bo_dump_fence(struct fence *fence, struct seq_file *m) +{ + struct amdgpu_fence *a_fence = to_amdgpu_fence(fence); + struct amd_sched_fence *s_fence = to_amd_sched_fence(fence); + + if (a_fence) + seq_printf(m, " protected by 0x%016llx on ring %d", + a_fence->seq, a_fence->ring->idx); + + if (s_fence) { + struct amdgpu_ring *ring; + + + ring = container_of(s_fence->sched, struct amdgpu_ring, sched); + seq_printf(m, " protected by 0x%016x on ring %d", + s_fence->base.seqno, ring->idx); + } +} + void amdgpu_sa_bo_dump_debug_info(struct amdgpu_sa_manager *sa_manager, struct seq_file *m) { @@ -428,18 +453,8 @@ void amdgpu_sa_bo_dump_debug_info(struct amdgpu_sa_manager *sa_manager, } seq_printf(m, "[0x%010llx 0x%010llx] size %8lld", soffset, eoffset, eoffset - soffset); - if (i->fence) { - struct amdgpu_fence *a_fence = to_amdgpu_fence(i->fence); - struct amd_sched_fence *s_fence = to_amd_sched_fence(i->fence); - if (a_fence) - seq_printf(m, " protected by 0x%016llx on ring %d", - a_fence->seq, a_fence->ring->idx); - if (s_fence) - seq_printf(m, " protected by 0x%016x on ring %d", - s_fence->base.seqno, - s_fence->scheduler->ring_id); - - } + if (i->fence) + amdgpu_sa_bo_dump_fence(i->fence, m); seq_printf(m, "\n"); } spin_unlock(&sa_manager->wq.lock); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c index de98fbd2971e..2e946b2cad88 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c @@ -27,63 +27,48 @@ #include <drm/drmP.h> #include "amdgpu.h" -static struct fence *amdgpu_sched_dependency(struct amd_sched_job *job) +static struct fence *amdgpu_sched_dependency(struct amd_sched_job *sched_job) { - struct amdgpu_job *sched_job = (struct amdgpu_job *)job; - return amdgpu_sync_get_fence(&sched_job->ibs->sync); + struct amdgpu_job *job = to_amdgpu_job(sched_job); + return amdgpu_sync_get_fence(&job->ibs->sync); } -static struct fence *amdgpu_sched_run_job(struct amd_sched_job *job) +static struct fence *amdgpu_sched_run_job(struct amd_sched_job *sched_job) { - struct amdgpu_job *sched_job; - struct amdgpu_fence *fence; + struct amdgpu_fence *fence = NULL; + struct amdgpu_job *job; int r; - if (!job) { + if (!sched_job) { DRM_ERROR("job is null\n"); return NULL; } - sched_job = (struct amdgpu_job *)job; - mutex_lock(&sched_job->job_lock); - r = amdgpu_ib_schedule(sched_job->adev, - sched_job->num_ibs, - sched_job->ibs, - sched_job->base.owner); - if (r) + job = to_amdgpu_job(sched_job); + mutex_lock(&job->job_lock); + r = amdgpu_ib_schedule(job->adev, + job->num_ibs, + job->ibs, + job->base.owner); + if (r) { + DRM_ERROR("Error scheduling IBs (%d)\n", r); goto err; - fence = amdgpu_fence_ref(sched_job->ibs[sched_job->num_ibs - 1].fence); - - if (sched_job->free_job) - sched_job->free_job(sched_job); + } - mutex_unlock(&sched_job->job_lock); - return &fence->base; + fence = amdgpu_fence_ref(job->ibs[job->num_ibs - 1].fence); err: - DRM_ERROR("Run job error\n"); - mutex_unlock(&sched_job->job_lock); - job->sched->ops->process_job(job); - return NULL; -} + if (job->free_job) + job->free_job(job); -static void amdgpu_sched_process_job(struct amd_sched_job *job) -{ - struct amdgpu_job *sched_job; - - if (!job) { - DRM_ERROR("job is null\n"); - return; - } - sched_job = (struct amdgpu_job *)job; - /* after processing job, free memory */ - fence_put(&sched_job->base.s_fence->base); - kfree(sched_job); + mutex_unlock(&job->job_lock); + fence_put(&job->base.s_fence->base); + kfree(job); + return fence ? &fence->base : NULL; } struct amd_sched_backend_ops amdgpu_sched_ops = { .dependency = amdgpu_sched_dependency, .run_job = amdgpu_sched_run_job, - .process_job = amdgpu_sched_process_job }; int amdgpu_sched_ib_submit_kernel_helper(struct amdgpu_device *adev, @@ -100,7 +85,7 @@ int amdgpu_sched_ib_submit_kernel_helper(struct amdgpu_device *adev, kzalloc(sizeof(struct amdgpu_job), GFP_KERNEL); if (!job) return -ENOMEM; - job->base.sched = ring->scheduler; + job->base.sched = &ring->sched; job->base.s_entity = &adev->kernel_ctx.rings[ring->idx].entity; job->adev = adev; job->ibs = ibs; @@ -109,7 +94,7 @@ int amdgpu_sched_ib_submit_kernel_helper(struct amdgpu_device *adev, mutex_init(&job->job_lock); job->free_job = free_job; mutex_lock(&job->job_lock); - r = amd_sched_entity_push_job((struct amd_sched_job *)job); + r = amd_sched_entity_push_job(&job->base); if (r) { mutex_unlock(&job->job_lock); kfree(job); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c index 068aeaff7183..4921de15b451 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c @@ -65,8 +65,14 @@ static bool amdgpu_sync_same_dev(struct amdgpu_device *adev, struct fence *f) if (a_fence) return a_fence->ring->adev == adev; - if (s_fence) - return (struct amdgpu_device *)s_fence->scheduler->priv == adev; + + if (s_fence) { + struct amdgpu_ring *ring; + + ring = container_of(s_fence->sched, struct amdgpu_ring, sched); + return ring->adev == adev; + } + return false; } @@ -251,6 +257,20 @@ int amdgpu_sync_wait(struct amdgpu_sync *sync) fence_put(e->fence); kfree(e); } + + if (amdgpu_enable_semaphores) + return 0; + + for (i = 0; i < AMDGPU_MAX_RINGS; ++i) { + struct amdgpu_fence *fence = sync->sync_to[i]; + if (!fence) + continue; + + r = fence_wait(&fence->base, false); + if (r) + return r; + } + return 0; } @@ -285,7 +305,8 @@ int amdgpu_sync_rings(struct amdgpu_sync *sync, return -EINVAL; } - if (amdgpu_enable_scheduler || (count >= AMDGPU_NUM_SYNCS)) { + if (amdgpu_enable_scheduler || !amdgpu_enable_semaphores || + (count >= AMDGPU_NUM_SYNCS)) { /* not enough room, wait manually */ r = fence_wait(&fence->base, false); if (r) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c index f80b1a43be8a..4865615e9c06 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_test.c @@ -59,8 +59,9 @@ static void amdgpu_do_test_moves(struct amdgpu_device *adev) goto out_cleanup; } - r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, 0, - NULL, &vram_obj); + r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, + AMDGPU_GEM_DOMAIN_VRAM, 0, + NULL, NULL, &vram_obj); if (r) { DRM_ERROR("Failed to create VRAM object\n"); goto out_cleanup; @@ -80,7 +81,8 @@ static void amdgpu_do_test_moves(struct amdgpu_device *adev) struct fence *fence = NULL; r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GTT, 0, NULL, gtt_obj + i); + AMDGPU_GEM_DOMAIN_GTT, 0, NULL, + NULL, gtt_obj + i); if (r) { DRM_ERROR("Failed to create GTT object %d\n", i); goto out_lclean; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index b5abd5cde413..364cbe975332 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -861,7 +861,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, 256 * 1024, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->stollen_vga_memory); + NULL, NULL, &adev->stollen_vga_memory); if (r) { return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c index 482e66797ae6..5cc95f1a7dab 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c @@ -247,7 +247,7 @@ int amdgpu_ucode_init_bo(struct amdgpu_device *adev) const struct common_firmware_header *header = NULL; err = amdgpu_bo_create(adev, adev->firmware.fw_size, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GTT, 0, NULL, bo); + AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL, bo); if (err) { dev_err(adev->dev, "(%d) Firmware buffer allocate failed\n", err); err = -ENOMEM; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c index 2cf6c6b06e3b..d0312364d950 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c @@ -156,7 +156,7 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, bo_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->uvd.vcpu_bo); + NULL, NULL, &adev->uvd.vcpu_bo); if (r) { dev_err(adev->dev, "(%d) failed to allocate UVD bo\n", r); return r; @@ -543,46 +543,60 @@ static int amdgpu_uvd_cs_msg(struct amdgpu_uvd_cs_ctx *ctx, return -EINVAL; } - if (msg_type == 1) { + switch (msg_type) { + case 0: + /* it's a create msg, calc image size (width * height) */ + amdgpu_bo_kunmap(bo); + + /* try to alloc a new handle */ + for (i = 0; i < AMDGPU_MAX_UVD_HANDLES; ++i) { + if (atomic_read(&adev->uvd.handles[i]) == handle) { + DRM_ERROR("Handle 0x%x already in use!\n", handle); + return -EINVAL; + } + + if (!atomic_cmpxchg(&adev->uvd.handles[i], 0, handle)) { + adev->uvd.filp[i] = ctx->parser->filp; + return 0; + } + } + + DRM_ERROR("No more free UVD handles!\n"); + return -EINVAL; + + case 1: /* it's a decode msg, calc buffer sizes */ r = amdgpu_uvd_cs_msg_decode(msg, ctx->buf_sizes); amdgpu_bo_kunmap(bo); if (r) return r; - } else if (msg_type == 2) { + /* validate the handle */ + for (i = 0; i < AMDGPU_MAX_UVD_HANDLES; ++i) { + if (atomic_read(&adev->uvd.handles[i]) == handle) { + if (adev->uvd.filp[i] != ctx->parser->filp) { + DRM_ERROR("UVD handle collision detected!\n"); + return -EINVAL; + } + return 0; + } + } + + DRM_ERROR("Invalid UVD handle 0x%x!\n", handle); + return -ENOENT; + + case 2: /* it's a destroy msg, free the handle */ for (i = 0; i < AMDGPU_MAX_UVD_HANDLES; ++i) atomic_cmpxchg(&adev->uvd.handles[i], handle, 0); amdgpu_bo_kunmap(bo); return 0; - } else { - /* it's a create msg */ - amdgpu_bo_kunmap(bo); - - if (msg_type != 0) { - DRM_ERROR("Illegal UVD message type (%d)!\n", msg_type); - return -EINVAL; - } - - /* it's a create msg, no special handling needed */ - } - - /* create or decode, validate the handle */ - for (i = 0; i < AMDGPU_MAX_UVD_HANDLES; ++i) { - if (atomic_read(&adev->uvd.handles[i]) == handle) - return 0; - } - /* handle not found try to alloc a new one */ - for (i = 0; i < AMDGPU_MAX_UVD_HANDLES; ++i) { - if (!atomic_cmpxchg(&adev->uvd.handles[i], 0, handle)) { - adev->uvd.filp[i] = ctx->parser->filp; - return 0; - } + default: + DRM_ERROR("Illegal UVD message type (%d)!\n", msg_type); + return -EINVAL; } - - DRM_ERROR("No more free UVD handles!\n"); + BUG(); return -EINVAL; } @@ -805,10 +819,10 @@ int amdgpu_uvd_ring_parse_cs(struct amdgpu_cs_parser *parser, uint32_t ib_idx) } static int amdgpu_uvd_free_job( - struct amdgpu_job *sched_job) + struct amdgpu_job *job) { - amdgpu_ib_free(sched_job->adev, sched_job->ibs); - kfree(sched_job->ibs); + amdgpu_ib_free(job->adev, job->ibs); + kfree(job->ibs); return 0; } @@ -905,7 +919,7 @@ int amdgpu_uvd_get_create_msg(struct amdgpu_ring *ring, uint32_t handle, r = amdgpu_bo_create(adev, 1024, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &bo); + NULL, NULL, &bo); if (r) return r; @@ -954,7 +968,7 @@ int amdgpu_uvd_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle, r = amdgpu_bo_create(adev, 1024, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &bo); + NULL, NULL, &bo); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c index 3cab96c42aa8..74f2038ac747 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c @@ -143,7 +143,7 @@ int amdgpu_vce_sw_init(struct amdgpu_device *adev, unsigned long size) r = amdgpu_bo_create(adev, size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->vce.vcpu_bo); + NULL, NULL, &adev->vce.vcpu_bo); if (r) { dev_err(adev->dev, "(%d) failed to allocate VCE bo\n", r); return r; @@ -342,10 +342,10 @@ void amdgpu_vce_free_handles(struct amdgpu_device *adev, struct drm_file *filp) } static int amdgpu_vce_free_job( - struct amdgpu_job *sched_job) + struct amdgpu_job *job) { - amdgpu_ib_free(sched_job->adev, sched_job->ibs); - kfree(sched_job->ibs); + amdgpu_ib_free(job->adev, job->ibs); + kfree(job->ibs); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index f68b7cdc370a..1e14531353e0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -316,12 +316,12 @@ static void amdgpu_vm_update_pages(struct amdgpu_device *adev, } } -int amdgpu_vm_free_job(struct amdgpu_job *sched_job) +int amdgpu_vm_free_job(struct amdgpu_job *job) { int i; - for (i = 0; i < sched_job->num_ibs; i++) - amdgpu_ib_free(sched_job->adev, &sched_job->ibs[i]); - kfree(sched_job->ibs); + for (i = 0; i < job->num_ibs; i++) + amdgpu_ib_free(job->adev, &job->ibs[i]); + kfree(job->ibs); return 0; } @@ -686,31 +686,6 @@ static int amdgpu_vm_update_ptes(struct amdgpu_device *adev, } /** - * amdgpu_vm_fence_pts - fence page tables after an update - * - * @vm: requested vm - * @start: start of GPU address range - * @end: end of GPU address range - * @fence: fence to use - * - * Fence the page tables in the range @start - @end (cayman+). - * - * Global and local mutex must be locked! - */ -static void amdgpu_vm_fence_pts(struct amdgpu_vm *vm, - uint64_t start, uint64_t end, - struct fence *fence) -{ - unsigned i; - - start >>= amdgpu_vm_block_size; - end >>= amdgpu_vm_block_size; - - for (i = start; i <= end; ++i) - amdgpu_bo_fence(vm->page_tables[i].bo, fence, true); -} - -/** * amdgpu_vm_bo_update_mapping - update a mapping in the vm page table * * @adev: amdgpu_device pointer @@ -813,8 +788,7 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev, if (r) goto error_free; - amdgpu_vm_fence_pts(vm, mapping->it.start, - mapping->it.last + 1, f); + amdgpu_bo_fence(vm->page_directory, f, true); if (fence) { fence_put(*fence); *fence = fence_get(f); @@ -855,7 +829,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, int r; if (mem) { - addr = mem->start << PAGE_SHIFT; + addr = (u64)mem->start << PAGE_SHIFT; if (mem->mem_type != TTM_PL_TT) addr += adev->vm_manager.vram_base_offset; } else { @@ -1089,6 +1063,7 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev, /* walk over the address space and allocate the page tables */ for (pt_idx = saddr; pt_idx <= eaddr; ++pt_idx) { + struct reservation_object *resv = vm->page_directory->tbo.resv; struct amdgpu_bo *pt; if (vm->page_tables[pt_idx].bo) @@ -1097,11 +1072,13 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev, /* drop mutex to allocate and clear page table */ mutex_unlock(&vm->mutex); + ww_mutex_lock(&resv->lock, NULL); r = amdgpu_bo_create(adev, AMDGPU_VM_PTE_COUNT * 8, AMDGPU_GPU_PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_NO_CPU_ACCESS, - NULL, &pt); + NULL, resv, &pt); + ww_mutex_unlock(&resv->lock); if (r) goto error_free; @@ -1303,7 +1280,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm) r = amdgpu_bo_create(adev, pd_size, align, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_NO_CPU_ACCESS, - NULL, &vm->page_directory); + NULL, NULL, &vm->page_directory); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/cz_smc.c b/drivers/gpu/drm/amd/amdgpu/cz_smc.c index a72ffc7d6c26..e33180d3314a 100644 --- a/drivers/gpu/drm/amd/amdgpu/cz_smc.c +++ b/drivers/gpu/drm/amd/amdgpu/cz_smc.c @@ -814,7 +814,8 @@ int cz_smu_init(struct amdgpu_device *adev) * 3. map kernel virtual address */ ret = amdgpu_bo_create(adev, priv->toc_buffer.data_size, PAGE_SIZE, - true, AMDGPU_GEM_DOMAIN_GTT, 0, NULL, toc_buf); + true, AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL, + toc_buf); if (ret) { dev_err(adev->dev, "(%d) SMC TOC buffer allocation failed\n", ret); @@ -822,7 +823,8 @@ int cz_smu_init(struct amdgpu_device *adev) } ret = amdgpu_bo_create(adev, priv->smu_buffer.data_size, PAGE_SIZE, - true, AMDGPU_GEM_DOMAIN_GTT, 0, NULL, smu_buf); + true, AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL, + smu_buf); if (ret) { dev_err(adev->dev, "(%d) SMC Internal buffer allocation failed\n", ret); diff --git a/drivers/gpu/drm/amd/amdgpu/fiji_smc.c b/drivers/gpu/drm/amd/amdgpu/fiji_smc.c index 322edea65857..bda1249eb871 100644 --- a/drivers/gpu/drm/amd/amdgpu/fiji_smc.c +++ b/drivers/gpu/drm/amd/amdgpu/fiji_smc.c @@ -764,7 +764,7 @@ int fiji_smu_init(struct amdgpu_device *adev) ret = amdgpu_bo_create(adev, image_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, toc_buf); + NULL, NULL, toc_buf); if (ret) { DRM_ERROR("Failed to allocate memory for TOC buffer\n"); return -ENOMEM; @@ -774,7 +774,7 @@ int fiji_smu_init(struct amdgpu_device *adev) ret = amdgpu_bo_create(adev, smu_internal_buffer_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, smu_buf); + NULL, NULL, smu_buf); if (ret) { DRM_ERROR("Failed to allocate memory for SMU internal buffer\n"); return -ENOMEM; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index 4bd1e5cf65ca..e992bf2ff66c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -3206,7 +3206,7 @@ static int gfx_v7_0_mec_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, adev->gfx.mec.num_mec *adev->gfx.mec.num_pipe * MEC_HPD_SIZE * 2, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GTT, 0, NULL, + AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL, &adev->gfx.mec.hpd_eop_obj); if (r) { dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r); @@ -3373,7 +3373,7 @@ static int gfx_v7_0_cp_compute_resume(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, sizeof(struct bonaire_mqd), PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GTT, 0, NULL, + AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL, &ring->mqd_obj); if (r) { dev_warn(adev->dev, "(%d) create MQD bo failed\n", r); @@ -3610,41 +3610,6 @@ static int gfx_v7_0_cp_resume(struct amdgpu_device *adev) return 0; } -static void gfx_v7_0_ce_sync_me(struct amdgpu_ring *ring) -{ - struct amdgpu_device *adev = ring->adev; - u64 gpu_addr = adev->wb.gpu_addr + adev->gfx.ce_sync_offs * 4; - - /* instruct DE to set a magic number */ - amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); - amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) | - WRITE_DATA_DST_SEL(5))); - amdgpu_ring_write(ring, gpu_addr & 0xfffffffc); - amdgpu_ring_write(ring, upper_32_bits(gpu_addr) & 0xffffffff); - amdgpu_ring_write(ring, 1); - - /* let CE wait till condition satisfied */ - amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5)); - amdgpu_ring_write(ring, (WAIT_REG_MEM_OPERATION(0) | /* wait */ - WAIT_REG_MEM_MEM_SPACE(1) | /* memory */ - WAIT_REG_MEM_FUNCTION(3) | /* == */ - WAIT_REG_MEM_ENGINE(2))); /* ce */ - amdgpu_ring_write(ring, gpu_addr & 0xfffffffc); - amdgpu_ring_write(ring, upper_32_bits(gpu_addr) & 0xffffffff); - amdgpu_ring_write(ring, 1); - amdgpu_ring_write(ring, 0xffffffff); - amdgpu_ring_write(ring, 4); /* poll interval */ - - /* instruct CE to reset wb of ce_sync to zero */ - amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); - amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(2) | - WRITE_DATA_DST_SEL(5) | - WR_CONFIRM)); - amdgpu_ring_write(ring, gpu_addr & 0xfffffffc); - amdgpu_ring_write(ring, upper_32_bits(gpu_addr) & 0xffffffff); - amdgpu_ring_write(ring, 0); -} - /* * vm * VMID 0 is the physical GPU addresses as used by the kernel. @@ -3663,6 +3628,13 @@ static void gfx_v7_0_ring_emit_vm_flush(struct amdgpu_ring *ring, unsigned vm_id, uint64_t pd_addr) { int usepfp = (ring->type == AMDGPU_RING_TYPE_GFX); + if (usepfp) { + /* synce CE with ME to prevent CE fetch CEIB before context switch done */ + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); + } amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(usepfp) | @@ -3703,7 +3675,10 @@ static void gfx_v7_0_ring_emit_vm_flush(struct amdgpu_ring *ring, amdgpu_ring_write(ring, 0x0); /* synce CE with ME to prevent CE fetch CEIB before context switch done */ - gfx_v7_0_ce_sync_me(ring); + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); } } @@ -3788,7 +3763,8 @@ static int gfx_v7_0_rlc_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, dws * 4, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->gfx.rlc.save_restore_obj); + NULL, NULL, + &adev->gfx.rlc.save_restore_obj); if (r) { dev_warn(adev->dev, "(%d) create RLC sr bo failed\n", r); return r; @@ -3831,7 +3807,8 @@ static int gfx_v7_0_rlc_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, dws * 4, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->gfx.rlc.clear_state_obj); + NULL, NULL, + &adev->gfx.rlc.clear_state_obj); if (r) { dev_warn(adev->dev, "(%d) create RLC c bo failed\n", r); gfx_v7_0_rlc_fini(adev); @@ -3870,7 +3847,8 @@ static int gfx_v7_0_rlc_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, adev->gfx.rlc.cp_table_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, &adev->gfx.rlc.cp_table_obj); + NULL, NULL, + &adev->gfx.rlc.cp_table_obj); if (r) { dev_warn(adev->dev, "(%d) create RLC cp table bo failed\n", r); gfx_v7_0_rlc_fini(adev); @@ -4802,12 +4780,6 @@ static int gfx_v7_0_sw_init(void *handle) return r; } - r = amdgpu_wb_get(adev, &adev->gfx.ce_sync_offs); - if (r) { - DRM_ERROR("(%d) gfx.ce_sync_offs wb alloc failed\n", r); - return r; - } - for (i = 0; i < adev->gfx.num_gfx_rings; i++) { ring = &adev->gfx.gfx_ring[i]; ring->ring_obj = NULL; @@ -4851,21 +4823,21 @@ static int gfx_v7_0_sw_init(void *handle) r = amdgpu_bo_create(adev, adev->gds.mem.gfx_partition_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_GDS, 0, - NULL, &adev->gds.gds_gfx_bo); + NULL, NULL, &adev->gds.gds_gfx_bo); if (r) return r; r = amdgpu_bo_create(adev, adev->gds.gws.gfx_partition_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_GWS, 0, - NULL, &adev->gds.gws_gfx_bo); + NULL, NULL, &adev->gds.gws_gfx_bo); if (r) return r; r = amdgpu_bo_create(adev, adev->gds.oa.gfx_partition_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_OA, 0, - NULL, &adev->gds.oa_gfx_bo); + NULL, NULL, &adev->gds.oa_gfx_bo); if (r) return r; @@ -4886,8 +4858,6 @@ static int gfx_v7_0_sw_fini(void *handle) for (i = 0; i < adev->gfx.num_compute_rings; i++) amdgpu_ring_fini(&adev->gfx.compute_ring[i]); - amdgpu_wb_free(adev, adev->gfx.ce_sync_offs); - gfx_v7_0_cp_compute_fini(adev); gfx_v7_0_rlc_fini(adev); gfx_v7_0_mec_fini(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 53f07439a512..cb4f68f53f24 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -868,7 +868,7 @@ static int gfx_v8_0_mec_init(struct amdgpu_device *adev) r = amdgpu_bo_create(adev, adev->gfx.mec.num_mec *adev->gfx.mec.num_pipe * MEC_HPD_SIZE * 2, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GTT, 0, NULL, + AMDGPU_GEM_DOMAIN_GTT, 0, NULL, NULL, &adev->gfx.mec.hpd_eop_obj); if (r) { dev_warn(adev->dev, "(%d) create HDP EOP bo failed\n", r); @@ -940,12 +940,6 @@ static int gfx_v8_0_sw_init(void *handle) return r; } - r = amdgpu_wb_get(adev, &adev->gfx.ce_sync_offs); - if (r) { - DRM_ERROR("(%d) gfx.ce_sync_offs wb alloc failed\n", r); - return r; - } - /* set up the gfx ring */ for (i = 0; i < adev->gfx.num_gfx_rings; i++) { ring = &adev->gfx.gfx_ring[i]; @@ -995,21 +989,21 @@ static int gfx_v8_0_sw_init(void *handle) /* reserve GDS, GWS and OA resource for gfx */ r = amdgpu_bo_create(adev, adev->gds.mem.gfx_partition_size, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GDS, 0, + AMDGPU_GEM_DOMAIN_GDS, 0, NULL, NULL, &adev->gds.gds_gfx_bo); if (r) return r; r = amdgpu_bo_create(adev, adev->gds.gws.gfx_partition_size, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_GWS, 0, + AMDGPU_GEM_DOMAIN_GWS, 0, NULL, NULL, &adev->gds.gws_gfx_bo); if (r) return r; r = amdgpu_bo_create(adev, adev->gds.oa.gfx_partition_size, PAGE_SIZE, true, - AMDGPU_GEM_DOMAIN_OA, 0, + AMDGPU_GEM_DOMAIN_OA, 0, NULL, NULL, &adev->gds.oa_gfx_bo); if (r) return r; @@ -1033,8 +1027,6 @@ static int gfx_v8_0_sw_fini(void *handle) for (i = 0; i < adev->gfx.num_compute_rings; i++) amdgpu_ring_fini(&adev->gfx.compute_ring[i]); - amdgpu_wb_free(adev, adev->gfx.ce_sync_offs); - gfx_v8_0_mec_fini(adev); return 0; @@ -3106,7 +3098,7 @@ static int gfx_v8_0_cp_compute_resume(struct amdgpu_device *adev) sizeof(struct vi_mqd), PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_GTT, 0, NULL, - &ring->mqd_obj); + NULL, &ring->mqd_obj); if (r) { dev_warn(adev->dev, "(%d) create MQD bo failed\n", r); return r; @@ -3965,6 +3957,7 @@ static void gfx_v8_0_ring_emit_fence_gfx(struct amdgpu_ring *ring, u64 addr, DATA_SEL(write64bit ? 2 : 1) | INT_SEL(int_sel ? 2 : 0)); amdgpu_ring_write(ring, lower_32_bits(seq)); amdgpu_ring_write(ring, upper_32_bits(seq)); + } /** @@ -4005,49 +3998,34 @@ static bool gfx_v8_0_ring_emit_semaphore(struct amdgpu_ring *ring, return true; } -static void gfx_v8_0_ce_sync_me(struct amdgpu_ring *ring) +static void gfx_v8_0_ring_emit_vm_flush(struct amdgpu_ring *ring, + unsigned vm_id, uint64_t pd_addr) { - struct amdgpu_device *adev = ring->adev; - u64 gpu_addr = adev->wb.gpu_addr + adev->gfx.ce_sync_offs * 4; - - /* instruct DE to set a magic number */ - amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); - amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) | - WRITE_DATA_DST_SEL(5))); - amdgpu_ring_write(ring, gpu_addr & 0xfffffffc); - amdgpu_ring_write(ring, upper_32_bits(gpu_addr) & 0xffffffff); - amdgpu_ring_write(ring, 1); + int usepfp = (ring->type == AMDGPU_RING_TYPE_GFX); + uint32_t seq = ring->fence_drv.sync_seq[ring->idx]; + uint64_t addr = ring->fence_drv.gpu_addr; - /* let CE wait till condition satisfied */ amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5)); - amdgpu_ring_write(ring, (WAIT_REG_MEM_OPERATION(0) | /* wait */ - WAIT_REG_MEM_MEM_SPACE(1) | /* memory */ - WAIT_REG_MEM_FUNCTION(3) | /* == */ - WAIT_REG_MEM_ENGINE(2))); /* ce */ - amdgpu_ring_write(ring, gpu_addr & 0xfffffffc); - amdgpu_ring_write(ring, upper_32_bits(gpu_addr) & 0xffffffff); - amdgpu_ring_write(ring, 1); + amdgpu_ring_write(ring, (WAIT_REG_MEM_MEM_SPACE(1) | /* memory */ + WAIT_REG_MEM_FUNCTION(3))); /* equal */ + amdgpu_ring_write(ring, addr & 0xfffffffc); + amdgpu_ring_write(ring, upper_32_bits(addr) & 0xffffffff); + amdgpu_ring_write(ring, seq); amdgpu_ring_write(ring, 0xffffffff); amdgpu_ring_write(ring, 4); /* poll interval */ - /* instruct CE to reset wb of ce_sync to zero */ - amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); - amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(2) | - WRITE_DATA_DST_SEL(5) | - WR_CONFIRM)); - amdgpu_ring_write(ring, gpu_addr & 0xfffffffc); - amdgpu_ring_write(ring, upper_32_bits(gpu_addr) & 0xffffffff); - amdgpu_ring_write(ring, 0); -} - -static void gfx_v8_0_ring_emit_vm_flush(struct amdgpu_ring *ring, - unsigned vm_id, uint64_t pd_addr) -{ - int usepfp = (ring->type == AMDGPU_RING_TYPE_GFX); + if (usepfp) { + /* synce CE with ME to prevent CE fetch CEIB before context switch done */ + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); + } amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(usepfp) | - WRITE_DATA_DST_SEL(0))); + WRITE_DATA_DST_SEL(0)) | + WR_CONFIRM); if (vm_id < 8) { amdgpu_ring_write(ring, (mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR + vm_id)); @@ -4083,9 +4061,10 @@ static void gfx_v8_0_ring_emit_vm_flush(struct amdgpu_ring *ring, /* sync PFP to ME, otherwise we might get invalid PFP reads */ amdgpu_ring_write(ring, PACKET3(PACKET3_PFP_SYNC_ME, 0)); amdgpu_ring_write(ring, 0x0); - - /* synce CE with ME to prevent CE fetch CEIB before context switch done */ - gfx_v8_0_ce_sync_me(ring); + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); + amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0)); + amdgpu_ring_write(ring, 0); } } diff --git a/drivers/gpu/drm/amd/amdgpu/iceland_smc.c b/drivers/gpu/drm/amd/amdgpu/iceland_smc.c index c900aa942ade..966d4b2ed9da 100644 --- a/drivers/gpu/drm/amd/amdgpu/iceland_smc.c +++ b/drivers/gpu/drm/amd/amdgpu/iceland_smc.c @@ -625,7 +625,7 @@ int iceland_smu_init(struct amdgpu_device *adev) ret = amdgpu_bo_create(adev, image_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, toc_buf); + NULL, NULL, toc_buf); if (ret) { DRM_ERROR("Failed to allocate memory for TOC buffer\n"); return -ENOMEM; diff --git a/drivers/gpu/drm/amd/amdgpu/tonga_smc.c b/drivers/gpu/drm/amd/amdgpu/tonga_smc.c index 1f5ac941a610..5421309c1862 100644 --- a/drivers/gpu/drm/amd/amdgpu/tonga_smc.c +++ b/drivers/gpu/drm/amd/amdgpu/tonga_smc.c @@ -763,7 +763,7 @@ int tonga_smu_init(struct amdgpu_device *adev) ret = amdgpu_bo_create(adev, image_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, toc_buf); + NULL, NULL, toc_buf); if (ret) { DRM_ERROR("Failed to allocate memory for TOC buffer\n"); return -ENOMEM; @@ -773,7 +773,7 @@ int tonga_smu_init(struct amdgpu_device *adev) ret = amdgpu_bo_create(adev, smu_internal_buffer_size, PAGE_SIZE, true, AMDGPU_GEM_DOMAIN_VRAM, AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED, - NULL, smu_buf); + NULL, NULL, smu_buf); if (ret) { DRM_ERROR("Failed to allocate memory for SMU internal buffer\n"); return -ENOMEM; diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c index 5fac5da694f0..ed50dd725788 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c @@ -224,11 +224,11 @@ static int uvd_v4_2_suspend(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = uvd_v4_2_hw_fini(adev); + r = amdgpu_uvd_suspend(adev); if (r) return r; - r = amdgpu_uvd_suspend(adev); + r = uvd_v4_2_hw_fini(adev); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c index 2d5c59c318af..9ad8b9906c0b 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c @@ -220,11 +220,11 @@ static int uvd_v5_0_suspend(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = uvd_v5_0_hw_fini(adev); + r = amdgpu_uvd_suspend(adev); if (r) return r; - r = amdgpu_uvd_suspend(adev); + r = uvd_v5_0_hw_fini(adev); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c index d9f553fce531..7e9934fa4193 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c @@ -214,14 +214,16 @@ static int uvd_v6_0_suspend(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + /* Skip this for APU for now */ + if (!(adev->flags & AMD_IS_APU)) { + r = amdgpu_uvd_suspend(adev); + if (r) + return r; + } r = uvd_v6_0_hw_fini(adev); if (r) return r; - r = amdgpu_uvd_suspend(adev); - if (r) - return r; - return r; } @@ -230,10 +232,12 @@ static int uvd_v6_0_resume(void *handle) int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_uvd_resume(adev); - if (r) - return r; - + /* Skip this for APU for now */ + if (!(adev->flags & AMD_IS_APU)) { + r = amdgpu_uvd_resume(adev); + if (r) + return r; + } r = uvd_v6_0_hw_init(adev); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c index 552d9e75ad1b..b55ceb14fdcd 100644 --- a/drivers/gpu/drm/amd/amdgpu/vi.c +++ b/drivers/gpu/drm/amd/amdgpu/vi.c @@ -1400,7 +1400,8 @@ static int vi_common_early_init(void *handle) case CHIP_CARRIZO: adev->has_uvd = true; adev->cg_flags = 0; - adev->pg_flags = AMDGPU_PG_SUPPORT_UVD | AMDGPU_PG_SUPPORT_VCE; + /* Disable UVD pg */ + adev->pg_flags = /* AMDGPU_PG_SUPPORT_UVD | */AMDGPU_PG_SUPPORT_VCE; adev->external_rev_id = adev->rev_id + 0x1; if (amdgpu_smc_load_fw && smc_enabled) adev->firmware.smu_load = true; diff --git a/drivers/gpu/drm/amd/scheduler/gpu_sched_trace.h b/drivers/gpu/drm/amd/scheduler/gpu_sched_trace.h new file mode 100644 index 000000000000..144f50acc971 --- /dev/null +++ b/drivers/gpu/drm/amd/scheduler/gpu_sched_trace.h @@ -0,0 +1,41 @@ +#if !defined(_GPU_SCHED_TRACE_H) || defined(TRACE_HEADER_MULTI_READ) +#define _GPU_SCHED_TRACE_H_ + +#include <linux/stringify.h> +#include <linux/types.h> +#include <linux/tracepoint.h> + +#include <drm/drmP.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM gpu_sched +#define TRACE_INCLUDE_FILE gpu_sched_trace + +TRACE_EVENT(amd_sched_job, + TP_PROTO(struct amd_sched_job *sched_job), + TP_ARGS(sched_job), + TP_STRUCT__entry( + __field(struct amd_sched_entity *, entity) + __field(const char *, name) + __field(u32, job_count) + __field(int, hw_job_count) + ), + + TP_fast_assign( + __entry->entity = sched_job->s_entity; + __entry->name = sched_job->sched->name; + __entry->job_count = kfifo_len( + &sched_job->s_entity->job_queue) / sizeof(sched_job); + __entry->hw_job_count = atomic_read( + &sched_job->sched->hw_rq_count); + ), + TP_printk("entity=%p, ring=%s, job count:%u, hw job count:%d", + __entry->entity, __entry->name, __entry->job_count, + __entry->hw_job_count) +); +#endif + +/* This part must be outside protection */ +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#include <trace/define_trace.h> diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c index 9259f1b6664c..3697eeeecf82 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c @@ -27,6 +27,9 @@ #include <drm/drmP.h> #include "gpu_scheduler.h" +#define CREATE_TRACE_POINTS +#include "gpu_sched_trace.h" + static struct amd_sched_job * amd_sched_entity_pop_job(struct amd_sched_entity *entity); static void amd_sched_wakeup(struct amd_gpu_scheduler *sched); @@ -65,29 +68,29 @@ static struct amd_sched_job * amd_sched_rq_select_job(struct amd_sched_rq *rq) { struct amd_sched_entity *entity; - struct amd_sched_job *job; + struct amd_sched_job *sched_job; spin_lock(&rq->lock); entity = rq->current_entity; if (entity) { list_for_each_entry_continue(entity, &rq->entities, list) { - job = amd_sched_entity_pop_job(entity); - if (job) { + sched_job = amd_sched_entity_pop_job(entity); + if (sched_job) { rq->current_entity = entity; spin_unlock(&rq->lock); - return job; + return sched_job; } } } list_for_each_entry(entity, &rq->entities, list) { - job = amd_sched_entity_pop_job(entity); - if (job) { + sched_job = amd_sched_entity_pop_job(entity); + if (sched_job) { rq->current_entity = entity; spin_unlock(&rq->lock); - return job; + return sched_job; } if (entity == rq->current_entity) @@ -115,23 +118,27 @@ int amd_sched_entity_init(struct amd_gpu_scheduler *sched, struct amd_sched_rq *rq, uint32_t jobs) { + int r; + if (!(sched && entity && rq)) return -EINVAL; memset(entity, 0, sizeof(struct amd_sched_entity)); - entity->belongto_rq = rq; - entity->scheduler = sched; - entity->fence_context = fence_context_alloc(1); - if(kfifo_alloc(&entity->job_queue, - jobs * sizeof(void *), - GFP_KERNEL)) - return -EINVAL; + INIT_LIST_HEAD(&entity->list); + entity->rq = rq; + entity->sched = sched; spin_lock_init(&entity->queue_lock); + r = kfifo_alloc(&entity->job_queue, jobs * sizeof(void *), GFP_KERNEL); + if (r) + return r; + atomic_set(&entity->fence_seq, 0); + entity->fence_context = fence_context_alloc(1); /* Add the entity to the run queue */ amd_sched_rq_add_entity(rq, entity); + return 0; } @@ -146,8 +153,8 @@ int amd_sched_entity_init(struct amd_gpu_scheduler *sched, static bool amd_sched_entity_is_initialized(struct amd_gpu_scheduler *sched, struct amd_sched_entity *entity) { - return entity->scheduler == sched && - entity->belongto_rq != NULL; + return entity->sched == sched && + entity->rq != NULL; } /** @@ -177,7 +184,7 @@ static bool amd_sched_entity_is_idle(struct amd_sched_entity *entity) void amd_sched_entity_fini(struct amd_gpu_scheduler *sched, struct amd_sched_entity *entity) { - struct amd_sched_rq *rq = entity->belongto_rq; + struct amd_sched_rq *rq = entity->rq; if (!amd_sched_entity_is_initialized(sched, entity)) return; @@ -198,22 +205,22 @@ static void amd_sched_entity_wakeup(struct fence *f, struct fence_cb *cb) container_of(cb, struct amd_sched_entity, cb); entity->dependency = NULL; fence_put(f); - amd_sched_wakeup(entity->scheduler); + amd_sched_wakeup(entity->sched); } static struct amd_sched_job * amd_sched_entity_pop_job(struct amd_sched_entity *entity) { - struct amd_gpu_scheduler *sched = entity->scheduler; - struct amd_sched_job *job; + struct amd_gpu_scheduler *sched = entity->sched; + struct amd_sched_job *sched_job; if (ACCESS_ONCE(entity->dependency)) return NULL; - if (!kfifo_out_peek(&entity->job_queue, &job, sizeof(job))) + if (!kfifo_out_peek(&entity->job_queue, &sched_job, sizeof(sched_job))) return NULL; - while ((entity->dependency = sched->ops->dependency(job))) { + while ((entity->dependency = sched->ops->dependency(sched_job))) { if (fence_add_callback(entity->dependency, &entity->cb, amd_sched_entity_wakeup)) @@ -222,32 +229,33 @@ amd_sched_entity_pop_job(struct amd_sched_entity *entity) return NULL; } - return job; + return sched_job; } /** * Helper to submit a job to the job queue * - * @job The pointer to job required to submit + * @sched_job The pointer to job required to submit * * Returns true if we could submit the job. */ -static bool amd_sched_entity_in(struct amd_sched_job *job) +static bool amd_sched_entity_in(struct amd_sched_job *sched_job) { - struct amd_sched_entity *entity = job->s_entity; + struct amd_sched_entity *entity = sched_job->s_entity; bool added, first = false; spin_lock(&entity->queue_lock); - added = kfifo_in(&entity->job_queue, &job, sizeof(job)) == sizeof(job); + added = kfifo_in(&entity->job_queue, &sched_job, + sizeof(sched_job)) == sizeof(sched_job); - if (added && kfifo_len(&entity->job_queue) == sizeof(job)) + if (added && kfifo_len(&entity->job_queue) == sizeof(sched_job)) first = true; spin_unlock(&entity->queue_lock); /* first job wakes up scheduler */ if (first) - amd_sched_wakeup(job->sched); + amd_sched_wakeup(sched_job->sched); return added; } @@ -255,7 +263,7 @@ static bool amd_sched_entity_in(struct amd_sched_job *job) /** * Submit a job to the job queue * - * @job The pointer to job required to submit + * @sched_job The pointer to job required to submit * * Returns 0 for success, negative error code otherwise. */ @@ -271,9 +279,9 @@ int amd_sched_entity_push_job(struct amd_sched_job *sched_job) fence_get(&fence->base); sched_job->s_fence = fence; - wait_event(entity->scheduler->job_scheduled, + wait_event(entity->sched->job_scheduled, amd_sched_entity_in(sched_job)); - + trace_amd_sched_job(sched_job); return 0; } @@ -301,30 +309,28 @@ static void amd_sched_wakeup(struct amd_gpu_scheduler *sched) static struct amd_sched_job * amd_sched_select_job(struct amd_gpu_scheduler *sched) { - struct amd_sched_job *job; + struct amd_sched_job *sched_job; if (!amd_sched_ready(sched)) return NULL; /* Kernel run queue has higher priority than normal run queue*/ - job = amd_sched_rq_select_job(&sched->kernel_rq); - if (job == NULL) - job = amd_sched_rq_select_job(&sched->sched_rq); + sched_job = amd_sched_rq_select_job(&sched->kernel_rq); + if (sched_job == NULL) + sched_job = amd_sched_rq_select_job(&sched->sched_rq); - return job; + return sched_job; } static void amd_sched_process_job(struct fence *f, struct fence_cb *cb) { - struct amd_sched_job *sched_job = - container_of(cb, struct amd_sched_job, cb); - struct amd_gpu_scheduler *sched; + struct amd_sched_fence *s_fence = + container_of(cb, struct amd_sched_fence, cb); + struct amd_gpu_scheduler *sched = s_fence->sched; - sched = sched_job->sched; - amd_sched_fence_signal(sched_job->s_fence); atomic_dec(&sched->hw_rq_count); - fence_put(&sched_job->s_fence->base); - sched->ops->process_job(sched_job); + amd_sched_fence_signal(s_fence); + fence_put(&s_fence->base); wake_up_interruptible(&sched->wake_up_worker); } @@ -338,87 +344,82 @@ static int amd_sched_main(void *param) while (!kthread_should_stop()) { struct amd_sched_entity *entity; - struct amd_sched_job *job; + struct amd_sched_fence *s_fence; + struct amd_sched_job *sched_job; struct fence *fence; wait_event_interruptible(sched->wake_up_worker, kthread_should_stop() || - (job = amd_sched_select_job(sched))); + (sched_job = amd_sched_select_job(sched))); - if (!job) + if (!sched_job) continue; - entity = job->s_entity; + entity = sched_job->s_entity; + s_fence = sched_job->s_fence; atomic_inc(&sched->hw_rq_count); - fence = sched->ops->run_job(job); + fence = sched->ops->run_job(sched_job); if (fence) { - r = fence_add_callback(fence, &job->cb, + r = fence_add_callback(fence, &s_fence->cb, amd_sched_process_job); if (r == -ENOENT) - amd_sched_process_job(fence, &job->cb); + amd_sched_process_job(fence, &s_fence->cb); else if (r) DRM_ERROR("fence add callback failed (%d)\n", r); fence_put(fence); + } else { + DRM_ERROR("Failed to run job!\n"); + amd_sched_process_job(NULL, &s_fence->cb); } - count = kfifo_out(&entity->job_queue, &job, sizeof(job)); - WARN_ON(count != sizeof(job)); + count = kfifo_out(&entity->job_queue, &sched_job, + sizeof(sched_job)); + WARN_ON(count != sizeof(sched_job)); wake_up(&sched->job_scheduled); } return 0; } /** - * Create a gpu scheduler + * Init a gpu scheduler instance * + * @sched The pointer to the scheduler * @ops The backend operations for this scheduler. - * @ring The the ring id for the scheduler. * @hw_submissions Number of hw submissions to do. + * @name Name used for debugging * - * Return the pointer to scheduler for success, otherwise return NULL + * Return 0 on success, otherwise error code. */ -struct amd_gpu_scheduler *amd_sched_create(struct amd_sched_backend_ops *ops, - unsigned ring, unsigned hw_submission, - void *priv) +int amd_sched_init(struct amd_gpu_scheduler *sched, + struct amd_sched_backend_ops *ops, + unsigned hw_submission, const char *name) { - struct amd_gpu_scheduler *sched; - - sched = kzalloc(sizeof(struct amd_gpu_scheduler), GFP_KERNEL); - if (!sched) - return NULL; - sched->ops = ops; - sched->ring_id = ring; sched->hw_submission_limit = hw_submission; - sched->priv = priv; - snprintf(sched->name, sizeof(sched->name), "amdgpu[%d]", ring); + sched->name = name; amd_sched_rq_init(&sched->sched_rq); amd_sched_rq_init(&sched->kernel_rq); init_waitqueue_head(&sched->wake_up_worker); init_waitqueue_head(&sched->job_scheduled); atomic_set(&sched->hw_rq_count, 0); + /* Each scheduler will run on a seperate kernel thread */ sched->thread = kthread_run(amd_sched_main, sched, sched->name); if (IS_ERR(sched->thread)) { - DRM_ERROR("Failed to create scheduler for id %d.\n", ring); - kfree(sched); - return NULL; + DRM_ERROR("Failed to create scheduler for %s.\n", name); + return PTR_ERR(sched->thread); } - return sched; + return 0; } /** * Destroy a gpu scheduler * * @sched The pointer to the scheduler - * - * return 0 if succeed. -1 if failed. */ -int amd_sched_destroy(struct amd_gpu_scheduler *sched) +void amd_sched_fini(struct amd_gpu_scheduler *sched) { kthread_stop(sched->thread); - kfree(sched); - return 0; } diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h index 2af0e4d4d817..80b64dc22214 100644 --- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h +++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h @@ -38,13 +38,15 @@ struct amd_sched_rq; */ struct amd_sched_entity { struct list_head list; - struct amd_sched_rq *belongto_rq; - atomic_t fence_seq; - /* the job_queue maintains the jobs submitted by clients */ - struct kfifo job_queue; + struct amd_sched_rq *rq; + struct amd_gpu_scheduler *sched; + spinlock_t queue_lock; - struct amd_gpu_scheduler *scheduler; + struct kfifo job_queue; + + atomic_t fence_seq; uint64_t fence_context; + struct fence *dependency; struct fence_cb cb; }; @@ -62,13 +64,13 @@ struct amd_sched_rq { struct amd_sched_fence { struct fence base; - struct amd_gpu_scheduler *scheduler; + struct fence_cb cb; + struct amd_gpu_scheduler *sched; spinlock_t lock; void *owner; }; struct amd_sched_job { - struct fence_cb cb; struct amd_gpu_scheduler *sched; struct amd_sched_entity *s_entity; struct amd_sched_fence *s_fence; @@ -91,32 +93,29 @@ static inline struct amd_sched_fence *to_amd_sched_fence(struct fence *f) * these functions should be implemented in driver side */ struct amd_sched_backend_ops { - struct fence *(*dependency)(struct amd_sched_job *job); - struct fence *(*run_job)(struct amd_sched_job *job); - void (*process_job)(struct amd_sched_job *job); + struct fence *(*dependency)(struct amd_sched_job *sched_job); + struct fence *(*run_job)(struct amd_sched_job *sched_job); }; /** * One scheduler is implemented for each hardware ring */ struct amd_gpu_scheduler { - struct task_struct *thread; + struct amd_sched_backend_ops *ops; + uint32_t hw_submission_limit; + const char *name; struct amd_sched_rq sched_rq; struct amd_sched_rq kernel_rq; - atomic_t hw_rq_count; - struct amd_sched_backend_ops *ops; - uint32_t ring_id; wait_queue_head_t wake_up_worker; wait_queue_head_t job_scheduled; - uint32_t hw_submission_limit; - char name[20]; - void *priv; + atomic_t hw_rq_count; + struct task_struct *thread; }; -struct amd_gpu_scheduler * -amd_sched_create(struct amd_sched_backend_ops *ops, - uint32_t ring, uint32_t hw_submission, void *priv); -int amd_sched_destroy(struct amd_gpu_scheduler *sched); +int amd_sched_init(struct amd_gpu_scheduler *sched, + struct amd_sched_backend_ops *ops, + uint32_t hw_submission, const char *name); +void amd_sched_fini(struct amd_gpu_scheduler *sched); int amd_sched_entity_init(struct amd_gpu_scheduler *sched, struct amd_sched_entity *entity, diff --git a/drivers/gpu/drm/amd/scheduler/sched_fence.c b/drivers/gpu/drm/amd/scheduler/sched_fence.c index e62c37920e11..d802638094f4 100644 --- a/drivers/gpu/drm/amd/scheduler/sched_fence.c +++ b/drivers/gpu/drm/amd/scheduler/sched_fence.c @@ -36,7 +36,7 @@ struct amd_sched_fence *amd_sched_fence_create(struct amd_sched_entity *s_entity if (fence == NULL) return NULL; fence->owner = owner; - fence->scheduler = s_entity->scheduler; + fence->sched = s_entity->sched; spin_lock_init(&fence->lock); seq = atomic_inc_return(&s_entity->fence_seq); @@ -63,7 +63,7 @@ static const char *amd_sched_fence_get_driver_name(struct fence *fence) static const char *amd_sched_fence_get_timeline_name(struct fence *f) { struct amd_sched_fence *fence = to_amd_sched_fence(f); - return (const char *)fence->scheduler->name; + return (const char *)fence->sched->name; } static bool amd_sched_fence_enable_signaling(struct fence *f) diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c index 9a860ca1e9d7..d93e7378c077 100644 --- a/drivers/gpu/drm/drm_ioctl.c +++ b/drivers/gpu/drm/drm_ioctl.c @@ -520,7 +520,8 @@ EXPORT_SYMBOL(drm_ioctl_permit); /** Ioctl table */ static const struct drm_ioctl_desc drm_ioctls[] = { - DRM_IOCTL_DEF(DRM_IOCTL_VERSION, drm_version, DRM_UNLOCKED|DRM_RENDER_ALLOW), + DRM_IOCTL_DEF(DRM_IOCTL_VERSION, drm_version, + DRM_UNLOCKED|DRM_RENDER_ALLOW|DRM_CONTROL_ALLOW), DRM_IOCTL_DEF(DRM_IOCTL_GET_UNIQUE, drm_getunique, 0), DRM_IOCTL_DEF(DRM_IOCTL_GET_MAGIC, drm_getmagic, 0), DRM_IOCTL_DEF(DRM_IOCTL_IRQ_BUSID, drm_irq_by_busid, DRM_MASTER|DRM_ROOT_ONLY), diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c index 82be6b86a168..d1e300dcd544 100644 --- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c +++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_plane.c @@ -58,7 +58,8 @@ static void fsl_dcu_drm_plane_atomic_disable(struct drm_plane *plane, struct drm_plane_state *old_state) { struct fsl_dcu_drm_device *fsl_dev = plane->dev->dev_private; - unsigned int index, value, ret; + unsigned int value; + int index, ret; index = fsl_dcu_drm_plane_index(plane); if (index < 0) diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 5a244ab9395b..39d73dbc1c47 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -640,6 +640,32 @@ static int __intel_get_crtc_scanline(struct intel_crtc *crtc) position = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN3; /* + * On HSW, the DSL reg (0x70000) appears to return 0 if we + * read it just before the start of vblank. So try it again + * so we don't accidentally end up spanning a vblank frame + * increment, causing the pipe_update_end() code to squak at us. + * + * The nature of this problem means we can't simply check the ISR + * bit and return the vblank start value; nor can we use the scanline + * debug register in the transcoder as it appears to have the same + * problem. We may need to extend this to include other platforms, + * but so far testing only shows the problem on HSW. + */ + if (IS_HASWELL(dev) && !position) { + int i, temp; + + for (i = 0; i < 100; i++) { + udelay(1); + temp = __raw_i915_read32(dev_priv, PIPEDSL(pipe)) & + DSL_LINEMASK_GEN3; + if (temp != position) { + position = temp; + break; + } + } + } + + /* * See update_scanline_offset() for the details on the * scanline_offset adjustment. */ diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c index 89c1a8ce1f98..2a5c76faf9f8 100644 --- a/drivers/gpu/drm/i915/intel_audio.c +++ b/drivers/gpu/drm/i915/intel_audio.c @@ -430,7 +430,7 @@ void intel_audio_codec_enable(struct intel_encoder *intel_encoder) /** * intel_audio_codec_disable - Disable the audio codec for HD audio - * @encoder: encoder on which to disable audio + * @intel_encoder: encoder on which to disable audio * * The disable sequences must be performed before disabling the transcoder or * port. diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c index b3e437b3bb54..c19e669ffe50 100644 --- a/drivers/gpu/drm/i915/intel_bios.c +++ b/drivers/gpu/drm/i915/intel_bios.c @@ -42,7 +42,7 @@ find_section(const void *_bdb, int section_id) const struct bdb_header *bdb = _bdb; const u8 *base = _bdb; int index = 0; - u16 total, current_size; + u32 total, current_size; u8 current_id; /* skip to first section */ @@ -57,6 +57,10 @@ find_section(const void *_bdb, int section_id) current_size = *((const u16 *)(base + index)); index += 2; + /* The MIPI Sequence Block v3+ has a separate size field. */ + if (current_id == BDB_MIPI_SEQUENCE && *(base + index) >= 3) + current_size = *((const u32 *)(base + index + 1)); + if (index + current_size > total) return NULL; @@ -799,6 +803,12 @@ parse_mipi(struct drm_i915_private *dev_priv, const struct bdb_header *bdb) return; } + /* Fail gracefully for forward incompatible sequence block. */ + if (sequence->version >= 3) { + DRM_ERROR("Unable to parse MIPI Sequence Block v3+\n"); + return; + } + DRM_DEBUG_DRIVER("Found MIPI sequence block\n"); block_size = get_blocksize(sequence); diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 8cc9264f7809..cf418be7d30a 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -15087,9 +15087,12 @@ static void readout_plane_state(struct intel_crtc *crtc, plane_state = to_intel_plane_state(p->base.state); - if (p->base.type == DRM_PLANE_TYPE_PRIMARY) + if (p->base.type == DRM_PLANE_TYPE_PRIMARY) { plane_state->visible = primary_get_hw_state(crtc); - else { + if (plane_state->visible) + crtc->base.state->plane_mask |= + 1 << drm_plane_index(&p->base); + } else { if (active) p->disable_plane(&p->base, &crtc->base); diff --git a/drivers/gpu/drm/mgag200/mgag200_fb.c b/drivers/gpu/drm/mgag200/mgag200_fb.c index 87de15ea1f93..b35b5b2db4ec 100644 --- a/drivers/gpu/drm/mgag200/mgag200_fb.c +++ b/drivers/gpu/drm/mgag200/mgag200_fb.c @@ -186,17 +186,19 @@ static int mgag200fb_create(struct drm_fb_helper *helper, sysram = vmalloc(size); if (!sysram) - return -ENOMEM; + goto err_sysram; info = drm_fb_helper_alloc_fbi(helper); - if (IS_ERR(info)) - return PTR_ERR(info); + if (IS_ERR(info)) { + ret = PTR_ERR(info); + goto err_alloc_fbi; + } info->par = mfbdev; ret = mgag200_framebuffer_init(dev, &mfbdev->mfb, &mode_cmd, gobj); if (ret) - return ret; + goto err_framebuffer_init; mfbdev->sysram = sysram; mfbdev->size = size; @@ -225,7 +227,17 @@ static int mgag200fb_create(struct drm_fb_helper *helper, DRM_DEBUG_KMS("allocated %dx%d\n", fb->width, fb->height); + return 0; + +err_framebuffer_init: + drm_fb_helper_release_fbi(helper); +err_alloc_fbi: + vfree(sysram); +err_sysram: + drm_gem_object_unreference_unlocked(gobj); + + return ret; } static int mga_fbdev_destroy(struct drm_device *dev, @@ -276,23 +288,26 @@ int mgag200_fbdev_init(struct mga_device *mdev) ret = drm_fb_helper_init(mdev->dev, &mfbdev->helper, mdev->num_crtc, MGAG200FB_CONN_LIMIT); if (ret) - return ret; + goto err_fb_helper; ret = drm_fb_helper_single_add_all_connectors(&mfbdev->helper); if (ret) - goto fini; + goto err_fb_setup; /* disable all the possible outputs/crtcs before entering KMS mode */ drm_helper_disable_unused_functions(mdev->dev); ret = drm_fb_helper_initial_config(&mfbdev->helper, bpp_sel); if (ret) - goto fini; + goto err_fb_setup; return 0; -fini: +err_fb_setup: drm_fb_helper_fini(&mfbdev->helper); +err_fb_helper: + mdev->mfbdev = NULL; + return ret; } diff --git a/drivers/gpu/drm/mgag200/mgag200_main.c b/drivers/gpu/drm/mgag200/mgag200_main.c index de06388069e7..b1a0f5656175 100644 --- a/drivers/gpu/drm/mgag200/mgag200_main.c +++ b/drivers/gpu/drm/mgag200/mgag200_main.c @@ -220,7 +220,7 @@ int mgag200_driver_load(struct drm_device *dev, unsigned long flags) } r = mgag200_mm_init(mdev); if (r) - goto out; + goto err_mm; drm_mode_config_init(dev); dev->mode_config.funcs = (void *)&mga_mode_funcs; @@ -233,7 +233,7 @@ int mgag200_driver_load(struct drm_device *dev, unsigned long flags) r = mgag200_modeset_init(mdev); if (r) { dev_err(&dev->pdev->dev, "Fatal error during modeset init: %d\n", r); - goto out; + goto err_modeset; } /* Make small buffers to store a hardware cursor (double buffered icon updates) */ @@ -241,20 +241,24 @@ int mgag200_driver_load(struct drm_device *dev, unsigned long flags) &mdev->cursor.pixels_1); mgag200_bo_create(dev, roundup(48*64, PAGE_SIZE), 0, 0, &mdev->cursor.pixels_2); - if (!mdev->cursor.pixels_2 || !mdev->cursor.pixels_1) - goto cursor_nospace; - mdev->cursor.pixels_current = mdev->cursor.pixels_1; - mdev->cursor.pixels_prev = mdev->cursor.pixels_2; - goto cursor_done; - cursor_nospace: - mdev->cursor.pixels_1 = NULL; - mdev->cursor.pixels_2 = NULL; - dev_warn(&dev->pdev->dev, "Could not allocate space for cursors. Not doing hardware cursors.\n"); - cursor_done: - -out: - if (r) - mgag200_driver_unload(dev); + if (!mdev->cursor.pixels_2 || !mdev->cursor.pixels_1) { + mdev->cursor.pixels_1 = NULL; + mdev->cursor.pixels_2 = NULL; + dev_warn(&dev->pdev->dev, + "Could not allocate space for cursors. Not doing hardware cursors.\n"); + } else { + mdev->cursor.pixels_current = mdev->cursor.pixels_1; + mdev->cursor.pixels_prev = mdev->cursor.pixels_2; + } + + return 0; + +err_modeset: + drm_mode_config_cleanup(dev); + mgag200_mm_fini(mdev); +err_mm: + dev->dev_private = NULL; + return r; } diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c index 7c6225c84ba6..dd845f82cc24 100644 --- a/drivers/gpu/drm/qxl/qxl_display.c +++ b/drivers/gpu/drm/qxl/qxl_display.c @@ -886,13 +886,15 @@ static enum drm_connector_status qxl_conn_detect( drm_connector_to_qxl_output(connector); struct drm_device *ddev = connector->dev; struct qxl_device *qdev = ddev->dev_private; - int connected; + bool connected = false; /* The first monitor is always connected */ - connected = (output->index == 0) || - (qdev->client_monitors_config && - qdev->client_monitors_config->count > output->index && - qxl_head_enabled(&qdev->client_monitors_config->heads[output->index])); + if (!qdev->client_monitors_config) { + if (output->index == 0) + connected = true; + } else + connected = qdev->client_monitors_config->count > output->index && + qxl_head_enabled(&qdev->client_monitors_config->heads[output->index]); DRM_DEBUG("#%d connected: %d\n", output->index, connected); if (!connected) diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index d8319dae8358..f3f562f6d848 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -1573,10 +1573,12 @@ int radeon_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon) drm_kms_helper_poll_disable(dev); + drm_modeset_lock_all(dev); /* turn off display hw */ list_for_each_entry(connector, &dev->mode_config.connector_list, head) { drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF); } + drm_modeset_unlock_all(dev); /* unpin the front buffers and cursors */ list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { @@ -1734,9 +1736,11 @@ int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon) if (fbcon) { drm_helper_resume_force_mode(dev); /* turn on display hw */ + drm_modeset_lock_all(dev); list_for_each_entry(connector, &dev->mode_config.connector_list, head) { drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON); } + drm_modeset_unlock_all(dev); } drm_kms_helper_poll_enable(dev); diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index 787cd8fd897f..e9115d3f67b0 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -2927,6 +2927,7 @@ static struct si_dpm_quirk si_dpm_quirk_list[] = { { PCI_VENDOR_ID_ATI, 0x6810, 0x1462, 0x3036, 0, 120000 }, { PCI_VENDOR_ID_ATI, 0x6811, 0x174b, 0xe271, 0, 120000 }, { PCI_VENDOR_ID_ATI, 0x6810, 0x174b, 0xe271, 85000, 90000 }, + { PCI_VENDOR_ID_ATI, 0x6811, 0x1762, 0x2015, 0, 120000 }, { 0, 0, 0, 0 }, }; diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index 8d9b7de25613..745e996d2dbc 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -882,6 +882,8 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo, if (ret) return ret; man = &bdev->man[mem_type]; + if (!man->has_type || !man->use_type) + continue; type_ok = ttm_bo_mt_compatible(man, mem_type, place, &cur_flags); @@ -889,6 +891,7 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo, if (!type_ok) continue; + type_found = true; cur_flags = ttm_bo_select_caching(man, bo->mem.placement, cur_flags); /* @@ -901,12 +904,10 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo, if (mem_type == TTM_PL_SYSTEM) break; - if (man->has_type && man->use_type) { - type_found = true; - ret = (*man->func->get_node)(man, bo, place, mem); - if (unlikely(ret)) - return ret; - } + ret = (*man->func->get_node)(man, bo, place, mem); + if (unlikely(ret)) + return ret; + if (mem->mm_node) break; } @@ -917,9 +918,6 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo, return 0; } - if (!type_found) - return -EINVAL; - for (i = 0; i < placement->num_busy_placement; ++i) { const struct ttm_place *place = &placement->busy_placement[i]; @@ -927,11 +925,12 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo, if (ret) return ret; man = &bdev->man[mem_type]; - if (!man->has_type) + if (!man->has_type || !man->use_type) continue; if (!ttm_bo_mt_compatible(man, mem_type, place, &cur_flags)) continue; + type_found = true; cur_flags = ttm_bo_select_caching(man, bo->mem.placement, cur_flags); /* @@ -957,8 +956,13 @@ int ttm_bo_mem_space(struct ttm_buffer_object *bo, if (ret == -ERESTARTSYS) has_erestartsys = true; } - ret = (has_erestartsys) ? -ERESTARTSYS : -ENOMEM; - return ret; + + if (!type_found) { + printk(KERN_ERR TTM_PFX "No compatible memory type found.\n"); + return -EINVAL; + } + + return (has_erestartsys) ? -ERESTARTSYS : -ENOMEM; } EXPORT_SYMBOL(ttm_bo_mem_space); diff --git a/drivers/gpu/drm/vmwgfx/Kconfig b/drivers/gpu/drm/vmwgfx/Kconfig index 67720f70fe29..b49445df8a7e 100644 --- a/drivers/gpu/drm/vmwgfx/Kconfig +++ b/drivers/gpu/drm/vmwgfx/Kconfig @@ -1,6 +1,6 @@ config DRM_VMWGFX tristate "DRM driver for VMware Virtual GPU" - depends on DRM && PCI + depends on DRM && PCI && X86 select FB_DEFERRED_IO select FB_CFB_FILLRECT select FB_CFB_COPYAREA diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c index ce659a125f2b..092ea81eeff7 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cotable.c @@ -311,7 +311,6 @@ static int vmw_cotable_unbind(struct vmw_resource *res, struct vmw_private *dev_priv = res->dev_priv; struct ttm_buffer_object *bo = val_buf->bo; struct vmw_fence_obj *fence; - int ret; if (list_empty(&res->mob_head)) return 0; @@ -328,7 +327,7 @@ static int vmw_cotable_unbind(struct vmw_resource *res, if (likely(fence != NULL)) vmw_fence_obj_unreference(&fence); - return ret; + return 0; } /** diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index e13b20bd9908..2c7a25c71af2 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -752,12 +752,8 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) ttm_lock_set_kill(&dev_priv->fbdev_master.lock, false, SIGTERM); dev_priv->active_master = &dev_priv->fbdev_master; - - dev_priv->mmio_mtrr = arch_phys_wc_add(dev_priv->mmio_start, - dev_priv->mmio_size); - - dev_priv->mmio_virt = ioremap_wc(dev_priv->mmio_start, - dev_priv->mmio_size); + dev_priv->mmio_virt = ioremap_cache(dev_priv->mmio_start, + dev_priv->mmio_size); if (unlikely(dev_priv->mmio_virt == NULL)) { ret = -ENOMEM; @@ -913,7 +909,6 @@ out_no_device: out_err4: iounmap(dev_priv->mmio_virt); out_err3: - arch_phys_wc_del(dev_priv->mmio_mtrr); vmw_ttm_global_release(dev_priv); out_err0: for (i = vmw_res_context; i < vmw_res_max; ++i) @@ -964,7 +959,6 @@ static int vmw_driver_unload(struct drm_device *dev) ttm_object_device_release(&dev_priv->tdev); iounmap(dev_priv->mmio_virt); - arch_phys_wc_del(dev_priv->mmio_mtrr); if (dev_priv->ctx.staged_bindings) vmw_binding_state_free(dev_priv->ctx.staged_bindings); vmw_ttm_global_release(dev_priv); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index 6d02de6dc36c..f19fd39b43e1 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -376,7 +376,6 @@ struct vmw_private { uint32_t initial_width; uint32_t initial_height; u32 __iomem *mmio_virt; - int mmio_mtrr; uint32_t capabilities; uint32_t max_gmr_ids; uint32_t max_gmr_pages; @@ -631,7 +630,8 @@ extern int vmw_user_dmabuf_alloc(struct vmw_private *dev_priv, uint32_t size, bool shareable, uint32_t *handle, - struct vmw_dma_buffer **p_dma_buf); + struct vmw_dma_buffer **p_dma_buf, + struct ttm_base_object **p_base); extern int vmw_user_dmabuf_reference(struct ttm_object_file *tfile, struct vmw_dma_buffer *dma_buf, uint32_t *handle); @@ -645,7 +645,8 @@ extern uint32_t vmw_dmabuf_validate_node(struct ttm_buffer_object *bo, uint32_t cur_validate_node); extern void vmw_dmabuf_validate_clear(struct ttm_buffer_object *bo); extern int vmw_user_dmabuf_lookup(struct ttm_object_file *tfile, - uint32_t id, struct vmw_dma_buffer **out); + uint32_t id, struct vmw_dma_buffer **out, + struct ttm_base_object **base); extern int vmw_stream_claim_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); extern int vmw_stream_unref_ioctl(struct drm_device *dev, void *data, diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index b56565457c96..5da5de0cb522 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -1236,7 +1236,8 @@ static int vmw_translate_mob_ptr(struct vmw_private *dev_priv, struct vmw_relocation *reloc; int ret; - ret = vmw_user_dmabuf_lookup(sw_context->fp->tfile, handle, &vmw_bo); + ret = vmw_user_dmabuf_lookup(sw_context->fp->tfile, handle, &vmw_bo, + NULL); if (unlikely(ret != 0)) { DRM_ERROR("Could not find or use MOB buffer.\n"); ret = -EINVAL; @@ -1296,7 +1297,8 @@ static int vmw_translate_guest_ptr(struct vmw_private *dev_priv, struct vmw_relocation *reloc; int ret; - ret = vmw_user_dmabuf_lookup(sw_context->fp->tfile, handle, &vmw_bo); + ret = vmw_user_dmabuf_lookup(sw_context->fp->tfile, handle, &vmw_bo, + NULL); if (unlikely(ret != 0)) { DRM_ERROR("Could not find or use GMR region.\n"); ret = -EINVAL; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index 61fb7f3de311..15a6c01cd016 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -1685,7 +1685,6 @@ int vmw_kms_helper_dirty(struct vmw_private *dev_priv, struct drm_crtc *crtc; u32 num_units = 0; u32 i, k; - int ret; dirty->dev_priv = dev_priv; @@ -1711,7 +1710,7 @@ int vmw_kms_helper_dirty(struct vmw_private *dev_priv, if (!dirty->cmd) { DRM_ERROR("Couldn't reserve fifo space " "for dirty blits.\n"); - return ret; + return -ENOMEM; } memset(dirty->cmd, 0, dirty->fifo_reserve_size); } diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c b/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c index 76069f093ccf..222c9c2123a1 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c @@ -484,7 +484,7 @@ int vmw_overlay_ioctl(struct drm_device *dev, void *data, goto out_unlock; } - ret = vmw_user_dmabuf_lookup(tfile, arg->handle, &buf); + ret = vmw_user_dmabuf_lookup(tfile, arg->handle, &buf, NULL); if (ret) goto out_unlock; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c index c1912f852b42..e57667ca7557 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c @@ -354,7 +354,7 @@ int vmw_user_lookup_handle(struct vmw_private *dev_priv, } *out_surf = NULL; - ret = vmw_user_dmabuf_lookup(tfile, handle, out_buf); + ret = vmw_user_dmabuf_lookup(tfile, handle, out_buf, NULL); return ret; } @@ -481,7 +481,8 @@ int vmw_user_dmabuf_alloc(struct vmw_private *dev_priv, uint32_t size, bool shareable, uint32_t *handle, - struct vmw_dma_buffer **p_dma_buf) + struct vmw_dma_buffer **p_dma_buf, + struct ttm_base_object **p_base) { struct vmw_user_dma_buffer *user_bo; struct ttm_buffer_object *tmp; @@ -515,6 +516,10 @@ int vmw_user_dmabuf_alloc(struct vmw_private *dev_priv, } *p_dma_buf = &user_bo->dma; + if (p_base) { + *p_base = &user_bo->prime.base; + kref_get(&(*p_base)->refcount); + } *handle = user_bo->prime.base.hash.key; out_no_base_object: @@ -631,6 +636,7 @@ int vmw_user_dmabuf_synccpu_ioctl(struct drm_device *dev, void *data, struct vmw_dma_buffer *dma_buf; struct vmw_user_dma_buffer *user_bo; struct ttm_object_file *tfile = vmw_fpriv(file_priv)->tfile; + struct ttm_base_object *buffer_base; int ret; if ((arg->flags & (drm_vmw_synccpu_read | drm_vmw_synccpu_write)) == 0 @@ -643,7 +649,8 @@ int vmw_user_dmabuf_synccpu_ioctl(struct drm_device *dev, void *data, switch (arg->op) { case drm_vmw_synccpu_grab: - ret = vmw_user_dmabuf_lookup(tfile, arg->handle, &dma_buf); + ret = vmw_user_dmabuf_lookup(tfile, arg->handle, &dma_buf, + &buffer_base); if (unlikely(ret != 0)) return ret; @@ -651,6 +658,7 @@ int vmw_user_dmabuf_synccpu_ioctl(struct drm_device *dev, void *data, dma); ret = vmw_user_dmabuf_synccpu_grab(user_bo, tfile, arg->flags); vmw_dmabuf_unreference(&dma_buf); + ttm_base_object_unref(&buffer_base); if (unlikely(ret != 0 && ret != -ERESTARTSYS && ret != -EBUSY)) { DRM_ERROR("Failed synccpu grab on handle 0x%08x.\n", @@ -692,7 +700,8 @@ int vmw_dmabuf_alloc_ioctl(struct drm_device *dev, void *data, return ret; ret = vmw_user_dmabuf_alloc(dev_priv, vmw_fpriv(file_priv)->tfile, - req->size, false, &handle, &dma_buf); + req->size, false, &handle, &dma_buf, + NULL); if (unlikely(ret != 0)) goto out_no_dmabuf; @@ -721,7 +730,8 @@ int vmw_dmabuf_unref_ioctl(struct drm_device *dev, void *data, } int vmw_user_dmabuf_lookup(struct ttm_object_file *tfile, - uint32_t handle, struct vmw_dma_buffer **out) + uint32_t handle, struct vmw_dma_buffer **out, + struct ttm_base_object **p_base) { struct vmw_user_dma_buffer *vmw_user_bo; struct ttm_base_object *base; @@ -743,7 +753,10 @@ int vmw_user_dmabuf_lookup(struct ttm_object_file *tfile, vmw_user_bo = container_of(base, struct vmw_user_dma_buffer, prime.base); (void)ttm_bo_reference(&vmw_user_bo->dma.base); - ttm_base_object_unref(&base); + if (p_base) + *p_base = base; + else + ttm_base_object_unref(&base); *out = &vmw_user_bo->dma; return 0; @@ -1004,7 +1017,7 @@ int vmw_dumb_create(struct drm_file *file_priv, ret = vmw_user_dmabuf_alloc(dev_priv, vmw_fpriv(file_priv)->tfile, args->size, false, &args->handle, - &dma_buf); + &dma_buf, NULL); if (unlikely(ret != 0)) goto out_no_dmabuf; @@ -1032,7 +1045,7 @@ int vmw_dumb_map_offset(struct drm_file *file_priv, struct vmw_dma_buffer *out_buf; int ret; - ret = vmw_user_dmabuf_lookup(tfile, handle, &out_buf); + ret = vmw_user_dmabuf_lookup(tfile, handle, &out_buf, NULL); if (ret != 0) return -EINVAL; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c index bba1ee395478..fd47547b0234 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c @@ -855,7 +855,7 @@ static int vmw_shader_define(struct drm_device *dev, struct drm_file *file_priv, if (buffer_handle != SVGA3D_INVALID_ID) { ret = vmw_user_dmabuf_lookup(tfile, buffer_handle, - &buffer); + &buffer, NULL); if (unlikely(ret != 0)) { DRM_ERROR("Could not find buffer for shader " "creation.\n"); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c index 3361769842f4..64b50409fa07 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c @@ -46,6 +46,7 @@ struct vmw_user_surface { struct vmw_surface srf; uint32_t size; struct drm_master *master; + struct ttm_base_object *backup_base; }; /** @@ -656,6 +657,7 @@ static void vmw_user_surface_base_release(struct ttm_base_object **p_base) struct vmw_resource *res = &user_srf->srf.res; *p_base = NULL; + ttm_base_object_unref(&user_srf->backup_base); vmw_resource_unreference(&res); } @@ -851,7 +853,8 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, res->backup_size, true, &backup_handle, - &res->backup); + &res->backup, + &user_srf->backup_base); if (unlikely(ret != 0)) { vmw_resource_unreference(&res); goto out_unlock; @@ -1321,7 +1324,8 @@ int vmw_gb_surface_define_ioctl(struct drm_device *dev, void *data, if (req->buffer_handle != SVGA3D_INVALID_ID) { ret = vmw_user_dmabuf_lookup(tfile, req->buffer_handle, - &res->backup); + &res->backup, + &user_srf->backup_base); if (ret == 0 && res->backup->base.num_pages * PAGE_SIZE < res->backup_size) { DRM_ERROR("Surface backup buffer is too small.\n"); @@ -1335,7 +1339,8 @@ int vmw_gb_surface_define_ioctl(struct drm_device *dev, void *data, req->drm_surface_flags & drm_vmw_surface_flag_shareable, &backup_handle, - &res->backup); + &res->backup, + &user_srf->backup_base); if (unlikely(ret != 0)) { vmw_resource_unreference(&res); diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index d60c88df5234..4b3b6f8aff0c 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -968,7 +968,8 @@ static void crypt_free_buffer_pages(struct crypt_config *cc, struct bio *clone); /* * Generate a new unfragmented bio with the given size - * This should never violate the device limitations + * This should never violate the device limitations (but only because + * max_segment_size is being constrained to PAGE_SIZE). * * This function may be called concurrently. If we allocate from the mempool * concurrently, there is a possibility of deadlock. For example, if we have @@ -2045,9 +2046,20 @@ static int crypt_iterate_devices(struct dm_target *ti, return fn(ti, cc->dev, cc->start, ti->len, data); } +static void crypt_io_hints(struct dm_target *ti, struct queue_limits *limits) +{ + /* + * Unfortunate constraint that is required to avoid the potential + * for exceeding underlying device's max_segments limits -- due to + * crypt_alloc_buffer() possibly allocating pages for the encryption + * bio that are not as physically contiguous as the original bio. + */ + limits->max_segment_size = PAGE_SIZE; +} + static struct target_type crypt_target = { .name = "crypt", - .version = {1, 14, 0}, + .version = {1, 14, 1}, .module = THIS_MODULE, .ctr = crypt_ctr, .dtr = crypt_dtr, @@ -2058,6 +2070,7 @@ static struct target_type crypt_target = { .resume = crypt_resume, .message = crypt_message, .iterate_devices = crypt_iterate_devices, + .io_hints = crypt_io_hints, }; static int __init dm_crypt_init(void) diff --git a/drivers/md/dm-thin.c b/drivers/md/dm-thin.c index 6578b7bc1fbb..6fcbfb063366 100644 --- a/drivers/md/dm-thin.c +++ b/drivers/md/dm-thin.c @@ -4249,6 +4249,10 @@ static void thin_io_hints(struct dm_target *ti, struct queue_limits *limits) { struct thin_c *tc = ti->private; struct pool *pool = tc->pool; + struct queue_limits *pool_limits = dm_get_queue_limits(pool->pool_md); + + if (!pool_limits->discard_granularity) + return; /* pool's discard support is disabled */ limits->discard_granularity = pool->sectors_per_block << SECTOR_SHIFT; limits->max_discard_sectors = 2048 * 1024 * 16; /* 16G */ diff --git a/drivers/misc/cxl/sysfs.c b/drivers/misc/cxl/sysfs.c index 25868c2ec03e..02006f7109a8 100644 --- a/drivers/misc/cxl/sysfs.c +++ b/drivers/misc/cxl/sysfs.c @@ -592,6 +592,8 @@ int cxl_sysfs_afu_add(struct cxl_afu *afu) /* conditionally create the add the binary file for error info buffer */ if (afu->eb_len) { + sysfs_attr_init(&afu->attr_eb.attr); + afu->attr_eb.attr.name = "afu_err_buff"; afu->attr_eb.attr.mode = S_IRUGO; afu->attr_eb.size = afu->eb_len; diff --git a/drivers/of/of_pci_irq.c b/drivers/of/of_pci_irq.c index 1710d9dc7fc2..2306313c0029 100644 --- a/drivers/of/of_pci_irq.c +++ b/drivers/of/of_pci_irq.c @@ -38,8 +38,8 @@ int of_irq_parse_pci(const struct pci_dev *pdev, struct of_phandle_args *out_irq */ rc = pci_read_config_byte(pdev, PCI_INTERRUPT_PIN, &pin); if (rc != 0) - return rc; - /* No pin, exit */ + goto err; + /* No pin, exit with no error message. */ if (pin == 0) return -ENODEV; @@ -53,8 +53,10 @@ int of_irq_parse_pci(const struct pci_dev *pdev, struct of_phandle_args *out_irq ppnode = pci_bus_to_OF_node(pdev->bus); /* No node for host bridge ? give up */ - if (ppnode == NULL) - return -EINVAL; + if (ppnode == NULL) { + rc = -EINVAL; + goto err; + } } else { /* We found a P2P bridge, check if it has a node */ ppnode = pci_device_to_OF_node(ppdev); @@ -86,7 +88,13 @@ int of_irq_parse_pci(const struct pci_dev *pdev, struct of_phandle_args *out_irq out_irq->args[0] = pin; laddr[0] = cpu_to_be32((pdev->bus->number << 16) | (pdev->devfn << 8)); laddr[1] = laddr[2] = cpu_to_be32(0); - return of_irq_parse_raw(laddr, out_irq); + rc = of_irq_parse_raw(laddr, out_irq); + if (rc) + goto err; + return 0; +err: + dev_err(&pdev->dev, "of_irq_parse_pci() failed with rc=%d\n", rc); + return rc; } EXPORT_SYMBOL_GPL(of_irq_parse_pci); @@ -105,10 +113,8 @@ int of_irq_parse_and_map_pci(const struct pci_dev *dev, u8 slot, u8 pin) int ret; ret = of_irq_parse_pci(dev, &oirq); - if (ret) { - dev_err(&dev->dev, "of_irq_parse_pci() failed with rc=%d\n", ret); + if (ret) return 0; /* Proper return code 0 == NO_IRQ */ - } return irq_create_of_mapping(&oirq); } diff --git a/drivers/parisc/dino.c b/drivers/parisc/dino.c index baec33c4e698..a0580afe1713 100644 --- a/drivers/parisc/dino.c +++ b/drivers/parisc/dino.c @@ -560,6 +560,9 @@ dino_fixup_bus(struct pci_bus *bus) } else if (bus->parent) { int i; + pci_read_bridge_bases(bus); + + for(i = PCI_BRIDGE_RESOURCES; i < PCI_NUM_RESOURCES; i++) { if((bus->self->resource[i].flags & (IORESOURCE_IO | IORESOURCE_MEM)) == 0) diff --git a/drivers/parisc/lba_pci.c b/drivers/parisc/lba_pci.c index 7b9e89ba0465..a32c1f6c252c 100644 --- a/drivers/parisc/lba_pci.c +++ b/drivers/parisc/lba_pci.c @@ -693,6 +693,7 @@ lba_fixup_bus(struct pci_bus *bus) if (bus->parent) { int i; /* PCI-PCI Bridge */ + pci_read_bridge_bases(bus); for (i = PCI_BRIDGE_RESOURCES; i < PCI_NUM_RESOURCES; i++) pci_claim_bridge_resource(bus->self, i); } else { diff --git a/drivers/pci/access.c b/drivers/pci/access.c index 769f7e35f1a2..59ac36fe7c42 100644 --- a/drivers/pci/access.c +++ b/drivers/pci/access.c @@ -442,7 +442,8 @@ static const struct pci_vpd_ops pci_vpd_pci22_ops = { static ssize_t pci_vpd_f0_read(struct pci_dev *dev, loff_t pos, size_t count, void *arg) { - struct pci_dev *tdev = pci_get_slot(dev->bus, PCI_SLOT(dev->devfn)); + struct pci_dev *tdev = pci_get_slot(dev->bus, + PCI_DEVFN(PCI_SLOT(dev->devfn), 0)); ssize_t ret; if (!tdev) @@ -456,7 +457,8 @@ static ssize_t pci_vpd_f0_read(struct pci_dev *dev, loff_t pos, size_t count, static ssize_t pci_vpd_f0_write(struct pci_dev *dev, loff_t pos, size_t count, const void *arg) { - struct pci_dev *tdev = pci_get_slot(dev->bus, PCI_SLOT(dev->devfn)); + struct pci_dev *tdev = pci_get_slot(dev->bus, + PCI_DEVFN(PCI_SLOT(dev->devfn), 0)); ssize_t ret; if (!tdev) @@ -473,22 +475,6 @@ static const struct pci_vpd_ops pci_vpd_f0_ops = { .release = pci_vpd_pci22_release, }; -static int pci_vpd_f0_dev_check(struct pci_dev *dev) -{ - struct pci_dev *tdev = pci_get_slot(dev->bus, PCI_SLOT(dev->devfn)); - int ret = 0; - - if (!tdev) - return -ENODEV; - if (!tdev->vpd || !tdev->multifunction || - dev->class != tdev->class || dev->vendor != tdev->vendor || - dev->device != tdev->device) - ret = -ENODEV; - - pci_dev_put(tdev); - return ret; -} - int pci_vpd_pci22_init(struct pci_dev *dev) { struct pci_vpd_pci22 *vpd; @@ -497,12 +483,7 @@ int pci_vpd_pci22_init(struct pci_dev *dev) cap = pci_find_capability(dev, PCI_CAP_ID_VPD); if (!cap) return -ENODEV; - if (dev->dev_flags & PCI_DEV_FLAGS_VPD_REF_F0) { - int ret = pci_vpd_f0_dev_check(dev); - if (ret) - return ret; - } vpd = kzalloc(sizeof(*vpd), GFP_ATOMIC); if (!vpd) return -ENOMEM; diff --git a/drivers/pci/bus.c b/drivers/pci/bus.c index 6fbd3f2b5992..d3346d23963b 100644 --- a/drivers/pci/bus.c +++ b/drivers/pci/bus.c @@ -256,6 +256,8 @@ bool pci_bus_clip_resource(struct pci_dev *dev, int idx) res->start = start; res->end = end; + res->flags &= ~IORESOURCE_UNSET; + orig_res.flags &= ~IORESOURCE_UNSET; dev_printk(KERN_DEBUG, &dev->dev, "%pR clipped to %pR\n", &orig_res, res); diff --git a/drivers/pci/host/pci-rcar-gen2.c b/drivers/pci/host/pci-rcar-gen2.c index 367e28fa7564..c4f64bfee551 100644 --- a/drivers/pci/host/pci-rcar-gen2.c +++ b/drivers/pci/host/pci-rcar-gen2.c @@ -362,6 +362,7 @@ static int rcar_pci_probe(struct platform_device *pdev) static struct of_device_id rcar_pci_of_match[] = { { .compatible = "renesas,pci-r8a7790", }, { .compatible = "renesas,pci-r8a7791", }, + { .compatible = "renesas,pci-r8a7794", }, { }, }; diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 0b2be174d981..8361d27e5eca 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -676,15 +676,20 @@ static struct irq_domain *pci_host_bridge_msi_domain(struct pci_bus *bus) static void pci_set_bus_msi_domain(struct pci_bus *bus) { struct irq_domain *d; + struct pci_bus *b; /* - * Either bus is the root, and we must obtain it from the - * firmware, or we inherit it from the bridge device. + * The bus can be a root bus, a subordinate bus, or a virtual bus + * created by an SR-IOV device. Walk up to the first bridge device + * found or derive the domain from the host bridge. */ - if (pci_is_root_bus(bus)) - d = pci_host_bridge_msi_domain(bus); - else - d = dev_get_msi_domain(&bus->self->dev); + for (b = bus, d = NULL; !d && !pci_is_root_bus(b); b = b->parent) { + if (b->self) + d = dev_get_msi_domain(&b->self->dev); + } + + if (!d) + d = pci_host_bridge_msi_domain(b); dev_set_msi_domain(&bus->dev, d); } @@ -855,9 +860,6 @@ int pci_scan_bridge(struct pci_bus *bus, struct pci_dev *dev, int max, int pass) child->bridge_ctl = bctl; } - /* Read and initialize bridge resources */ - pci_read_bridge_bases(child); - cmax = pci_scan_child_bus(child); if (cmax > subordinate) dev_warn(&dev->dev, "bridge has subordinate %02x but max busn %02x\n", @@ -918,9 +920,6 @@ int pci_scan_bridge(struct pci_bus *bus, struct pci_dev *dev, int max, int pass) if (!is_cardbus) { child->bridge_ctl = bctl; - - /* Read and initialize bridge resources */ - pci_read_bridge_bases(child); max = pci_scan_child_bus(child); } else { /* diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 6a30252cd79f..b03373fd05ca 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -1907,11 +1907,27 @@ static void quirk_netmos(struct pci_dev *dev) DECLARE_PCI_FIXUP_CLASS_HEADER(PCI_VENDOR_ID_NETMOS, PCI_ANY_ID, PCI_CLASS_COMMUNICATION_SERIAL, 8, quirk_netmos); +/* + * Quirk non-zero PCI functions to route VPD access through function 0 for + * devices that share VPD resources between functions. The functions are + * expected to be identical devices. + */ static void quirk_f0_vpd_link(struct pci_dev *dev) { - if (!dev->multifunction || !PCI_FUNC(dev->devfn)) + struct pci_dev *f0; + + if (!PCI_FUNC(dev->devfn)) return; - dev->dev_flags |= PCI_DEV_FLAGS_VPD_REF_F0; + + f0 = pci_get_slot(dev->bus, PCI_DEVFN(PCI_SLOT(dev->devfn), 0)); + if (!f0) + return; + + if (f0->vpd && dev->class == f0->class && + dev->vendor == f0->vendor && dev->device == f0->device) + dev->dev_flags |= PCI_DEV_FLAGS_VPD_REF_F0; + + pci_dev_put(f0); } DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_INTEL, PCI_ANY_ID, PCI_CLASS_NETWORK_ETHERNET, 8, quirk_f0_vpd_link); diff --git a/drivers/regulator/anatop-regulator.c b/drivers/regulator/anatop-regulator.c index 738adfa5332b..52ea605f8130 100644 --- a/drivers/regulator/anatop-regulator.c +++ b/drivers/regulator/anatop-regulator.c @@ -318,6 +318,7 @@ static const struct of_device_id of_anatop_regulator_match_tbl[] = { { .compatible = "fsl,anatop-regulator", }, { /* end */ } }; +MODULE_DEVICE_TABLE(of, of_anatop_regulator_match_tbl); static struct platform_driver anatop_regulator_driver = { .driver = { diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c index 7a85ac9e32c5..7849187d91ae 100644 --- a/drivers/regulator/core.c +++ b/drivers/regulator/core.c @@ -1394,15 +1394,15 @@ static int regulator_resolve_supply(struct regulator_dev *rdev) return 0; r = regulator_dev_lookup(dev, rdev->supply_name, &ret); - if (ret == -ENODEV) { - /* - * No supply was specified for this regulator and - * there will never be one. - */ - return 0; - } - if (!r) { + if (ret == -ENODEV) { + /* + * No supply was specified for this regulator and + * there will never be one. + */ + return 0; + } + if (have_full_constraints()) { r = dummy_regulator_rdev; } else { @@ -1422,11 +1422,10 @@ static int regulator_resolve_supply(struct regulator_dev *rdev) return ret; /* Cascade always-on state to supply */ - if (_regulator_is_enabled(rdev)) { + if (_regulator_is_enabled(rdev) && rdev->supply) { ret = regulator_enable(rdev->supply); if (ret < 0) { - if (rdev->supply) - _regulator_put(rdev->supply); + _regulator_put(rdev->supply); return ret; } } diff --git a/drivers/regulator/gpio-regulator.c b/drivers/regulator/gpio-regulator.c index 464018de7e97..7bba8b747f30 100644 --- a/drivers/regulator/gpio-regulator.c +++ b/drivers/regulator/gpio-regulator.c @@ -394,6 +394,7 @@ static const struct of_device_id regulator_gpio_of_match[] = { { .compatible = "regulator-gpio", }, {}, }; +MODULE_DEVICE_TABLE(of, regulator_gpio_of_match); #endif static struct platform_driver gpio_regulator_driver = { diff --git a/drivers/regulator/pbias-regulator.c b/drivers/regulator/pbias-regulator.c index 4fa7bcaf454e..f9d74d63be7c 100644 --- a/drivers/regulator/pbias-regulator.c +++ b/drivers/regulator/pbias-regulator.c @@ -45,6 +45,10 @@ struct pbias_regulator_data { int voltage; }; +struct pbias_of_data { + unsigned int offset; +}; + static const unsigned int pbias_volt_table[] = { 1800000, 3000000 @@ -102,8 +106,35 @@ static struct of_regulator_match pbias_matches[] = { }; #define PBIAS_NUM_REGS ARRAY_SIZE(pbias_matches) +/* Offset from SCM general area (and syscon) base */ + +static const struct pbias_of_data pbias_of_data_omap2 = { + .offset = 0x230, +}; + +static const struct pbias_of_data pbias_of_data_omap3 = { + .offset = 0x2b0, +}; + +static const struct pbias_of_data pbias_of_data_omap4 = { + .offset = 0x60, +}; + +static const struct pbias_of_data pbias_of_data_omap5 = { + .offset = 0x60, +}; + +static const struct pbias_of_data pbias_of_data_dra7 = { + .offset = 0xe00, +}; + static const struct of_device_id pbias_of_match[] = { { .compatible = "ti,pbias-omap", }, + { .compatible = "ti,pbias-omap2", .data = &pbias_of_data_omap2, }, + { .compatible = "ti,pbias-omap3", .data = &pbias_of_data_omap3, }, + { .compatible = "ti,pbias-omap4", .data = &pbias_of_data_omap4, }, + { .compatible = "ti,pbias-omap5", .data = &pbias_of_data_omap5, }, + { .compatible = "ti,pbias-dra7", .data = &pbias_of_data_dra7, }, {}, }; MODULE_DEVICE_TABLE(of, pbias_of_match); @@ -118,6 +149,9 @@ static int pbias_regulator_probe(struct platform_device *pdev) const struct pbias_reg_info *info; int ret = 0; int count, idx, data_idx = 0; + const struct of_device_id *match; + const struct pbias_of_data *data; + unsigned int offset; count = of_regulator_match(&pdev->dev, np, pbias_matches, PBIAS_NUM_REGS); @@ -133,6 +167,20 @@ static int pbias_regulator_probe(struct platform_device *pdev) if (IS_ERR(syscon)) return PTR_ERR(syscon); + match = of_match_device(of_match_ptr(pbias_of_match), &pdev->dev); + if (match && match->data) { + data = match->data; + offset = data->offset; + } else { + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!res) + return -EINVAL; + + offset = res->start; + dev_WARN(&pdev->dev, + "using legacy dt data for pbias offset\n"); + } + cfg.regmap = syscon; cfg.dev = &pdev->dev; @@ -145,10 +193,6 @@ static int pbias_regulator_probe(struct platform_device *pdev) if (!info) return -ENODEV; - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (!res) - return -EINVAL; - drvdata[data_idx].syscon = syscon; drvdata[data_idx].info = info; drvdata[data_idx].desc.name = info->name; @@ -158,9 +202,9 @@ static int pbias_regulator_probe(struct platform_device *pdev) drvdata[data_idx].desc.volt_table = pbias_volt_table; drvdata[data_idx].desc.n_voltages = 2; drvdata[data_idx].desc.enable_time = info->enable_time; - drvdata[data_idx].desc.vsel_reg = res->start; + drvdata[data_idx].desc.vsel_reg = offset; drvdata[data_idx].desc.vsel_mask = info->vmode; - drvdata[data_idx].desc.enable_reg = res->start; + drvdata[data_idx].desc.enable_reg = offset; drvdata[data_idx].desc.enable_mask = info->enable_mask; drvdata[data_idx].desc.enable_val = info->enable; drvdata[data_idx].desc.disable_val = info->disable_val; diff --git a/drivers/regulator/tps65218-regulator.c b/drivers/regulator/tps65218-regulator.c index 7f97223f95c5..a02c1b961039 100644 --- a/drivers/regulator/tps65218-regulator.c +++ b/drivers/regulator/tps65218-regulator.c @@ -73,7 +73,7 @@ static const struct regulator_linear_range dcdc4_ranges[] = { }; static struct tps_info tps65218_pmic_regs[] = { - TPS65218_INFO(DCDC1, "DCDC1", 850000, 167500), + TPS65218_INFO(DCDC1, "DCDC1", 850000, 1675000), TPS65218_INFO(DCDC2, "DCDC2", 850000, 1675000), TPS65218_INFO(DCDC3, "DCDC3", 900000, 3400000), TPS65218_INFO(DCDC4, "DCDC4", 1175000, 3400000), diff --git a/drivers/regulator/vexpress.c b/drivers/regulator/vexpress.c index bed9d3ee4198..c810cbbd463f 100644 --- a/drivers/regulator/vexpress.c +++ b/drivers/regulator/vexpress.c @@ -103,6 +103,7 @@ static const struct of_device_id vexpress_regulator_of_match[] = { { .compatible = "arm,vexpress-volt", }, { } }; +MODULE_DEVICE_TABLE(of, vexpress_regulator_of_match); static struct platform_driver vexpress_regulator_driver = { .probe = vexpress_regulator_probe, diff --git a/drivers/sh/pm_runtime.c b/drivers/sh/pm_runtime.c index d3d1891cda3c..25abd4eb7d10 100644 --- a/drivers/sh/pm_runtime.c +++ b/drivers/sh/pm_runtime.c @@ -35,20 +35,11 @@ static struct pm_clk_notifier_block platform_bus_notifier = { static int __init sh_pm_runtime_init(void) { if (IS_ENABLED(CONFIG_ARCH_SHMOBILE_MULTI)) { - if (!of_machine_is_compatible("renesas,emev2") && - !of_machine_is_compatible("renesas,r7s72100") && -#ifndef CONFIG_PM_GENERIC_DOMAINS_OF - !of_machine_is_compatible("renesas,r8a73a4") && - !of_machine_is_compatible("renesas,r8a7740") && - !of_machine_is_compatible("renesas,sh73a0") && -#endif - !of_machine_is_compatible("renesas,r8a7778") && - !of_machine_is_compatible("renesas,r8a7779") && - !of_machine_is_compatible("renesas,r8a7790") && - !of_machine_is_compatible("renesas,r8a7791") && - !of_machine_is_compatible("renesas,r8a7792") && - !of_machine_is_compatible("renesas,r8a7793") && - !of_machine_is_compatible("renesas,r8a7794")) + if (!of_find_compatible_node(NULL, NULL, + "renesas,cpg-mstp-clocks")) + return 0; + if (IS_ENABLED(CONFIG_PM_GENERIC_DOMAINS_OF) && + of_find_node_with_property(NULL, "#power-domain-cells")) return 0; } diff --git a/drivers/spi/spi-atmel.c b/drivers/spi/spi-atmel.c index bf9ed380bb1c..63318e2afba1 100644 --- a/drivers/spi/spi-atmel.c +++ b/drivers/spi/spi-atmel.c @@ -1720,6 +1720,7 @@ static int atmel_spi_runtime_resume(struct device *dev) return clk_prepare_enable(as->clk); } +#ifdef CONFIG_PM_SLEEP static int atmel_spi_suspend(struct device *dev) { struct spi_master *master = dev_get_drvdata(dev); @@ -1756,6 +1757,7 @@ static int atmel_spi_resume(struct device *dev) return ret; } +#endif static const struct dev_pm_ops atmel_spi_pm_ops = { SET_SYSTEM_SLEEP_PM_OPS(atmel_spi_suspend, atmel_spi_resume) diff --git a/drivers/spi/spi-bcm2835.c b/drivers/spi/spi-bcm2835.c index e7874a6171ec..3e8eeb23d4e9 100644 --- a/drivers/spi/spi-bcm2835.c +++ b/drivers/spi/spi-bcm2835.c @@ -386,14 +386,14 @@ static bool bcm2835_spi_can_dma(struct spi_master *master, /* otherwise we only allow transfers within the same page * to avoid wasting time on dma_mapping when it is not practical */ - if (((size_t)tfr->tx_buf & PAGE_MASK) + tfr->len > PAGE_SIZE) { + if (((size_t)tfr->tx_buf & (PAGE_SIZE - 1)) + tfr->len > PAGE_SIZE) { dev_warn_once(&spi->dev, "Unaligned spi tx-transfer bridging page\n"); return false; } - if (((size_t)tfr->rx_buf & PAGE_MASK) + tfr->len > PAGE_SIZE) { + if (((size_t)tfr->rx_buf & (PAGE_SIZE - 1)) + tfr->len > PAGE_SIZE) { dev_warn_once(&spi->dev, - "Unaligned spi tx-transfer bridging page\n"); + "Unaligned spi rx-transfer bridging page\n"); return false; } diff --git a/drivers/spi/spi-meson-spifc.c b/drivers/spi/spi-meson-spifc.c index 5468fc70dbf8..2465259f6241 100644 --- a/drivers/spi/spi-meson-spifc.c +++ b/drivers/spi/spi-meson-spifc.c @@ -444,6 +444,7 @@ static const struct of_device_id meson_spifc_dt_match[] = { { .compatible = "amlogic,meson6-spifc", }, { }, }; +MODULE_DEVICE_TABLE(of, meson_spifc_dt_match); static struct platform_driver meson_spifc_driver = { .probe = meson_spifc_probe, diff --git a/drivers/spi/spi-mt65xx.c b/drivers/spi/spi-mt65xx.c index 5f6315c47920..ecb6c58238c4 100644 --- a/drivers/spi/spi-mt65xx.c +++ b/drivers/spi/spi-mt65xx.c @@ -85,7 +85,7 @@ struct mtk_spi { void __iomem *base; u32 state; u32 pad_sel; - struct clk *spi_clk, *parent_clk; + struct clk *parent_clk, *sel_clk, *spi_clk; struct spi_transfer *cur_transfer; u32 xfer_len; struct scatterlist *tx_sgl, *rx_sgl; @@ -173,22 +173,6 @@ static void mtk_spi_config(struct mtk_spi *mdata, writel(mdata->pad_sel, mdata->base + SPI_PAD_SEL_REG); } -static int mtk_spi_prepare_hardware(struct spi_master *master) -{ - struct spi_transfer *trans; - struct mtk_spi *mdata = spi_master_get_devdata(master); - struct spi_message *msg = master->cur_msg; - - trans = list_first_entry(&msg->transfers, struct spi_transfer, - transfer_list); - if (!trans->cs_change) { - mdata->state = MTK_SPI_IDLE; - mtk_spi_reset(mdata); - } - - return 0; -} - static int mtk_spi_prepare_message(struct spi_master *master, struct spi_message *msg) { @@ -228,11 +212,15 @@ static void mtk_spi_set_cs(struct spi_device *spi, bool enable) struct mtk_spi *mdata = spi_master_get_devdata(spi->master); reg_val = readl(mdata->base + SPI_CMD_REG); - if (!enable) + if (!enable) { reg_val |= SPI_CMD_PAUSE_EN; - else + writel(reg_val, mdata->base + SPI_CMD_REG); + } else { reg_val &= ~SPI_CMD_PAUSE_EN; - writel(reg_val, mdata->base + SPI_CMD_REG); + writel(reg_val, mdata->base + SPI_CMD_REG); + mdata->state = MTK_SPI_IDLE; + mtk_spi_reset(mdata); + } } static void mtk_spi_prepare_transfer(struct spi_master *master, @@ -509,7 +497,6 @@ static int mtk_spi_probe(struct platform_device *pdev) master->mode_bits = SPI_CPOL | SPI_CPHA; master->set_cs = mtk_spi_set_cs; - master->prepare_transfer_hardware = mtk_spi_prepare_hardware; master->prepare_message = mtk_spi_prepare_message; master->transfer_one = mtk_spi_transfer_one; master->can_dma = mtk_spi_can_dma; @@ -576,13 +563,6 @@ static int mtk_spi_probe(struct platform_device *pdev) goto err_put_master; } - mdata->spi_clk = devm_clk_get(&pdev->dev, "spi-clk"); - if (IS_ERR(mdata->spi_clk)) { - ret = PTR_ERR(mdata->spi_clk); - dev_err(&pdev->dev, "failed to get spi-clk: %d\n", ret); - goto err_put_master; - } - mdata->parent_clk = devm_clk_get(&pdev->dev, "parent-clk"); if (IS_ERR(mdata->parent_clk)) { ret = PTR_ERR(mdata->parent_clk); @@ -590,13 +570,27 @@ static int mtk_spi_probe(struct platform_device *pdev) goto err_put_master; } + mdata->sel_clk = devm_clk_get(&pdev->dev, "sel-clk"); + if (IS_ERR(mdata->sel_clk)) { + ret = PTR_ERR(mdata->sel_clk); + dev_err(&pdev->dev, "failed to get sel-clk: %d\n", ret); + goto err_put_master; + } + + mdata->spi_clk = devm_clk_get(&pdev->dev, "spi-clk"); + if (IS_ERR(mdata->spi_clk)) { + ret = PTR_ERR(mdata->spi_clk); + dev_err(&pdev->dev, "failed to get spi-clk: %d\n", ret); + goto err_put_master; + } + ret = clk_prepare_enable(mdata->spi_clk); if (ret < 0) { dev_err(&pdev->dev, "failed to enable spi_clk (%d)\n", ret); goto err_put_master; } - ret = clk_set_parent(mdata->spi_clk, mdata->parent_clk); + ret = clk_set_parent(mdata->sel_clk, mdata->parent_clk); if (ret < 0) { dev_err(&pdev->dev, "failed to clk_set_parent (%d)\n", ret); goto err_disable_clk; @@ -630,7 +624,6 @@ static int mtk_spi_remove(struct platform_device *pdev) pm_runtime_disable(&pdev->dev); mtk_spi_reset(mdata); - clk_disable_unprepare(mdata->spi_clk); spi_master_put(master); return 0; diff --git a/drivers/spi/spi-pxa2xx.c b/drivers/spi/spi-pxa2xx.c index fdd791977041..a8ef38ebb9c9 100644 --- a/drivers/spi/spi-pxa2xx.c +++ b/drivers/spi/spi-pxa2xx.c @@ -654,6 +654,10 @@ static irqreturn_t ssp_int(int irq, void *dev_id) if (!(sccr1_reg & SSCR1_TIE)) mask &= ~SSSR_TFS; + /* Ignore RX timeout interrupt if it is disabled */ + if (!(sccr1_reg & SSCR1_TINTE)) + mask &= ~SSSR_TINT; + if (!(status & mask)) return IRQ_NONE; diff --git a/drivers/spi/spi-xtensa-xtfpga.c b/drivers/spi/spi-xtensa-xtfpga.c index 2e32ea2f194f..be6155cba9de 100644 --- a/drivers/spi/spi-xtensa-xtfpga.c +++ b/drivers/spi/spi-xtensa-xtfpga.c @@ -34,13 +34,13 @@ struct xtfpga_spi { static inline void xtfpga_spi_write32(const struct xtfpga_spi *spi, unsigned addr, u32 val) { - iowrite32(val, spi->regs + addr); + __raw_writel(val, spi->regs + addr); } static inline unsigned int xtfpga_spi_read32(const struct xtfpga_spi *spi, unsigned addr) { - return ioread32(spi->regs + addr); + return __raw_readl(spi->regs + addr); } static inline void xtfpga_spi_wait_busy(struct xtfpga_spi *xspi) diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index 3abb3903f2ad..a5f53de813d3 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -1610,8 +1610,7 @@ static struct class spi_master_class = { * * The caller is responsible for assigning the bus number and initializing * the master's methods before calling spi_register_master(); and (after errors - * adding the device) calling spi_master_put() and kfree() to prevent a memory - * leak. + * adding the device) calling spi_master_put() to prevent a memory leak. */ struct spi_master *spi_alloc_master(struct device *dev, unsigned size) { diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c index fba92a526531..ef008e52f953 100644 --- a/drivers/spi/spidev.c +++ b/drivers/spi/spidev.c @@ -651,7 +651,8 @@ static int spidev_release(struct inode *inode, struct file *filp) kfree(spidev->rx_buffer); spidev->rx_buffer = NULL; - spidev->speed_hz = spidev->spi->max_speed_hz; + if (spidev->spi) + spidev->speed_hz = spidev->spi->max_speed_hz; /* ... after we unbound from the underlying device? */ spin_lock_irq(&spidev->spi_lock); diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig index 039004400987..5aabc4bc0d75 100644 --- a/drivers/thermal/Kconfig +++ b/drivers/thermal/Kconfig @@ -163,7 +163,7 @@ config THERMAL_EMULATION config HISI_THERMAL tristate "Hisilicon thermal driver" - depends on ARCH_HISI && CPU_THERMAL && OF + depends on (ARCH_HISI && CPU_THERMAL && OF) || COMPILE_TEST help Enable this to plug hisilicon's thermal sensor driver into the Linux thermal framework. cpufreq is used as the cooling device to throttle @@ -182,7 +182,7 @@ config IMX_THERMAL config SPEAR_THERMAL bool "SPEAr thermal sensor driver" - depends on PLAT_SPEAR + depends on PLAT_SPEAR || COMPILE_TEST depends on OF help Enable this to plug the SPEAr thermal sensor driver into the Linux @@ -190,7 +190,7 @@ config SPEAR_THERMAL config ROCKCHIP_THERMAL tristate "Rockchip thermal driver" - depends on ARCH_ROCKCHIP + depends on ARCH_ROCKCHIP || COMPILE_TEST depends on RESET_CONTROLLER help Rockchip thermal driver provides support for Temperature sensor @@ -208,7 +208,7 @@ config RCAR_THERMAL config KIRKWOOD_THERMAL tristate "Temperature sensor on Marvell Kirkwood SoCs" - depends on MACH_KIRKWOOD + depends on MACH_KIRKWOOD || COMPILE_TEST depends on OF help Support for the Kirkwood thermal sensor driver into the Linux thermal @@ -216,7 +216,7 @@ config KIRKWOOD_THERMAL config DOVE_THERMAL tristate "Temperature sensor on Marvell Dove SoCs" - depends on ARCH_DOVE || MACH_DOVE + depends on ARCH_DOVE || MACH_DOVE || COMPILE_TEST depends on OF help Support for the Dove thermal sensor driver in the Linux thermal @@ -234,7 +234,7 @@ config DB8500_THERMAL config ARMADA_THERMAL tristate "Armada 370/XP thermal management" - depends on ARCH_MVEBU + depends on ARCH_MVEBU || COMPILE_TEST depends on OF help Enable this option if you want to have support for thermal management @@ -349,11 +349,12 @@ config INTEL_PCH_THERMAL programmable trip points and other information. menu "Texas Instruments thermal drivers" +depends on ARCH_HAS_BANDGAP || COMPILE_TEST source "drivers/thermal/ti-soc-thermal/Kconfig" endmenu menu "Samsung thermal drivers" -depends on ARCH_EXYNOS +depends on ARCH_EXYNOS || COMPILE_TEST source "drivers/thermal/samsung/Kconfig" endmenu @@ -364,7 +365,7 @@ endmenu config QCOM_SPMI_TEMP_ALARM tristate "Qualcomm SPMI PMIC Temperature Alarm" - depends on OF && SPMI && IIO + depends on OF && (SPMI || COMPILE_TEST) && IIO select REGMAP_SPMI help This enables a thermal sysfs driver for Qualcomm plug-and-play (QPNP) diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c index 620dcd405ff6..42c6f71bdcc1 100644 --- a/drivers/thermal/cpu_cooling.c +++ b/drivers/thermal/cpu_cooling.c @@ -262,7 +262,9 @@ static int cpufreq_thermal_notifier(struct notifier_block *nb, * efficiently. Power is stored in mW, frequency in KHz. The * resulting table is in ascending order. * - * Return: 0 on success, -E* on error. + * Return: 0 on success, -EINVAL if there are no OPPs for any CPUs, + * -ENOMEM if we run out of memory or -EAGAIN if an OPP was + * added/enabled while the function was executing. */ static int build_dyn_power_table(struct cpufreq_cooling_device *cpufreq_device, u32 capacitance) @@ -273,8 +275,6 @@ static int build_dyn_power_table(struct cpufreq_cooling_device *cpufreq_device, int num_opps = 0, cpu, i, ret = 0; unsigned long freq; - rcu_read_lock(); - for_each_cpu(cpu, &cpufreq_device->allowed_cpus) { dev = get_cpu_device(cpu); if (!dev) { @@ -284,24 +284,20 @@ static int build_dyn_power_table(struct cpufreq_cooling_device *cpufreq_device, } num_opps = dev_pm_opp_get_opp_count(dev); - if (num_opps > 0) { + if (num_opps > 0) break; - } else if (num_opps < 0) { - ret = num_opps; - goto unlock; - } + else if (num_opps < 0) + return num_opps; } - if (num_opps == 0) { - ret = -EINVAL; - goto unlock; - } + if (num_opps == 0) + return -EINVAL; power_table = kcalloc(num_opps, sizeof(*power_table), GFP_KERNEL); - if (!power_table) { - ret = -ENOMEM; - goto unlock; - } + if (!power_table) + return -ENOMEM; + + rcu_read_lock(); for (freq = 0, i = 0; opp = dev_pm_opp_find_freq_ceil(dev, &freq), !IS_ERR(opp); @@ -309,6 +305,12 @@ static int build_dyn_power_table(struct cpufreq_cooling_device *cpufreq_device, u32 freq_mhz, voltage_mv; u64 power; + if (i >= num_opps) { + rcu_read_unlock(); + ret = -EAGAIN; + goto free_power_table; + } + freq_mhz = freq / 1000000; voltage_mv = dev_pm_opp_get_voltage(opp) / 1000; @@ -326,17 +328,22 @@ static int build_dyn_power_table(struct cpufreq_cooling_device *cpufreq_device, power_table[i].power = power; } - if (i == 0) { + rcu_read_unlock(); + + if (i != num_opps) { ret = PTR_ERR(opp); - goto unlock; + goto free_power_table; } cpufreq_device->cpu_dev = dev; cpufreq_device->dyn_power_table = power_table; cpufreq_device->dyn_power_table_entries = i; -unlock: - rcu_read_unlock(); + return 0; + +free_power_table: + kfree(power_table); + return ret; } @@ -847,7 +854,7 @@ __cpufreq_cooling_register(struct device_node *np, ret = get_idr(&cpufreq_idr, &cpufreq_dev->id); if (ret) { cool_dev = ERR_PTR(ret); - goto free_table; + goto free_power_table; } snprintf(dev_name, sizeof(dev_name), "thermal-cpufreq-%d", @@ -889,6 +896,8 @@ __cpufreq_cooling_register(struct device_node *np, remove_idr: release_idr(&cpufreq_idr, cpufreq_dev->id); +free_power_table: + kfree(cpufreq_dev->dyn_power_table); free_table: kfree(cpufreq_dev->freq_table); free_time_in_idle_timestamp: @@ -1039,6 +1048,7 @@ void cpufreq_cooling_unregister(struct thermal_cooling_device *cdev) thermal_cooling_device_unregister(cpufreq_dev->cool_dev); release_idr(&cpufreq_idr, cpufreq_dev->id); + kfree(cpufreq_dev->dyn_power_table); kfree(cpufreq_dev->time_in_idle_timestamp); kfree(cpufreq_dev->time_in_idle); kfree(cpufreq_dev->freq_table); diff --git a/drivers/thermal/db8500_cpufreq_cooling.c b/drivers/thermal/db8500_cpufreq_cooling.c index 607b62c7e611..e58bd0b658b5 100644 --- a/drivers/thermal/db8500_cpufreq_cooling.c +++ b/drivers/thermal/db8500_cpufreq_cooling.c @@ -72,6 +72,7 @@ static const struct of_device_id db8500_cpufreq_cooling_match[] = { { .compatible = "stericsson,db8500-cpufreq-cooling" }, {}, }; +MODULE_DEVICE_TABLE(of, db8500_cpufreq_cooling_match); #endif static struct platform_driver db8500_cpufreq_cooling_driver = { diff --git a/drivers/thermal/power_allocator.c b/drivers/thermal/power_allocator.c index 9c8a7aad0252..7ff96270c933 100644 --- a/drivers/thermal/power_allocator.c +++ b/drivers/thermal/power_allocator.c @@ -24,6 +24,8 @@ #include "thermal_core.h" +#define INVALID_TRIP -1 + #define FRAC_BITS 10 #define int_to_frac(x) ((x) << FRAC_BITS) #define frac_to_int(x) ((x) >> FRAC_BITS) @@ -56,16 +58,21 @@ static inline s64 div_frac(s64 x, s64 y) /** * struct power_allocator_params - parameters for the power allocator governor + * @allocated_tzp: whether we have allocated tzp for this thermal zone and + * it needs to be freed on unbind * @err_integral: accumulated error in the PID controller. * @prev_err: error in the previous iteration of the PID controller. * Used to calculate the derivative term. * @trip_switch_on: first passive trip point of the thermal zone. The * governor switches on when this trip point is crossed. + * If the thermal zone only has one passive trip point, + * @trip_switch_on should be INVALID_TRIP. * @trip_max_desired_temperature: last passive trip point of the thermal * zone. The temperature we are * controlling for. */ struct power_allocator_params { + bool allocated_tzp; s64 err_integral; s32 prev_err; int trip_switch_on; @@ -73,6 +80,88 @@ struct power_allocator_params { }; /** + * estimate_sustainable_power() - Estimate the sustainable power of a thermal zone + * @tz: thermal zone we are operating in + * + * For thermal zones that don't provide a sustainable_power in their + * thermal_zone_params, estimate one. Calculate it using the minimum + * power of all the cooling devices as that gives a valid value that + * can give some degree of functionality. For optimal performance of + * this governor, provide a sustainable_power in the thermal zone's + * thermal_zone_params. + */ +static u32 estimate_sustainable_power(struct thermal_zone_device *tz) +{ + u32 sustainable_power = 0; + struct thermal_instance *instance; + struct power_allocator_params *params = tz->governor_data; + + list_for_each_entry(instance, &tz->thermal_instances, tz_node) { + struct thermal_cooling_device *cdev = instance->cdev; + u32 min_power; + + if (instance->trip != params->trip_max_desired_temperature) + continue; + + if (power_actor_get_min_power(cdev, tz, &min_power)) + continue; + + sustainable_power += min_power; + } + + return sustainable_power; +} + +/** + * estimate_pid_constants() - Estimate the constants for the PID controller + * @tz: thermal zone for which to estimate the constants + * @sustainable_power: sustainable power for the thermal zone + * @trip_switch_on: trip point number for the switch on temperature + * @control_temp: target temperature for the power allocator governor + * @force: whether to force the update of the constants + * + * This function is used to update the estimation of the PID + * controller constants in struct thermal_zone_parameters. + * Sustainable power is provided in case it was estimated. The + * estimated sustainable_power should not be stored in the + * thermal_zone_parameters so it has to be passed explicitly to this + * function. + * + * If @force is not set, the values in the thermal zone's parameters + * are preserved if they are not zero. If @force is set, the values + * in thermal zone's parameters are overwritten. + */ +static void estimate_pid_constants(struct thermal_zone_device *tz, + u32 sustainable_power, int trip_switch_on, + int control_temp, bool force) +{ + int ret; + int switch_on_temp; + u32 temperature_threshold; + + ret = tz->ops->get_trip_temp(tz, trip_switch_on, &switch_on_temp); + if (ret) + switch_on_temp = 0; + + temperature_threshold = control_temp - switch_on_temp; + + if (!tz->tzp->k_po || force) + tz->tzp->k_po = int_to_frac(sustainable_power) / + temperature_threshold; + + if (!tz->tzp->k_pu || force) + tz->tzp->k_pu = int_to_frac(2 * sustainable_power) / + temperature_threshold; + + if (!tz->tzp->k_i || force) + tz->tzp->k_i = int_to_frac(10) / 1000; + /* + * The default for k_d and integral_cutoff is 0, so we can + * leave them as they are. + */ +} + +/** * pid_controller() - PID controller * @tz: thermal zone we are operating in * @current_temp: the current temperature in millicelsius @@ -98,10 +187,20 @@ static u32 pid_controller(struct thermal_zone_device *tz, { s64 p, i, d, power_range; s32 err, max_power_frac; + u32 sustainable_power; struct power_allocator_params *params = tz->governor_data; max_power_frac = int_to_frac(max_allocatable_power); + if (tz->tzp->sustainable_power) { + sustainable_power = tz->tzp->sustainable_power; + } else { + sustainable_power = estimate_sustainable_power(tz); + estimate_pid_constants(tz, sustainable_power, + params->trip_switch_on, control_temp, + true); + } + err = control_temp - current_temp; err = int_to_frac(err); @@ -139,7 +238,7 @@ static u32 pid_controller(struct thermal_zone_device *tz, power_range = p + i + d; /* feed-forward the known sustainable dissipatable power */ - power_range = tz->tzp->sustainable_power + frac_to_int(power_range); + power_range = sustainable_power + frac_to_int(power_range); power_range = clamp(power_range, (s64)0, (s64)max_allocatable_power); @@ -247,6 +346,11 @@ static int allocate_power(struct thermal_zone_device *tz, } } + if (!num_actors) { + ret = -ENODEV; + goto unlock; + } + /* * We need to allocate five arrays of the same size: * req_power, max_power, granted_power, extra_actor_power and @@ -340,43 +444,66 @@ unlock: return ret; } -static int get_governor_trips(struct thermal_zone_device *tz, - struct power_allocator_params *params) +/** + * get_governor_trips() - get the number of the two trip points that are key for this governor + * @tz: thermal zone to operate on + * @params: pointer to private data for this governor + * + * The power allocator governor works optimally with two trips points: + * a "switch on" trip point and a "maximum desired temperature". These + * are defined as the first and last passive trip points. + * + * If there is only one trip point, then that's considered to be the + * "maximum desired temperature" trip point and the governor is always + * on. If there are no passive or active trip points, then the + * governor won't do anything. In fact, its throttle function + * won't be called at all. + */ +static void get_governor_trips(struct thermal_zone_device *tz, + struct power_allocator_params *params) { - int i, ret, last_passive; + int i, last_active, last_passive; bool found_first_passive; found_first_passive = false; - last_passive = -1; - ret = -EINVAL; + last_active = INVALID_TRIP; + last_passive = INVALID_TRIP; for (i = 0; i < tz->trips; i++) { enum thermal_trip_type type; + int ret; ret = tz->ops->get_trip_type(tz, i, &type); - if (ret) - return ret; + if (ret) { + dev_warn(&tz->device, + "Failed to get trip point %d type: %d\n", i, + ret); + continue; + } - if (!found_first_passive) { - if (type == THERMAL_TRIP_PASSIVE) { + if (type == THERMAL_TRIP_PASSIVE) { + if (!found_first_passive) { params->trip_switch_on = i; found_first_passive = true; + } else { + last_passive = i; } - } else if (type == THERMAL_TRIP_PASSIVE) { - last_passive = i; + } else if (type == THERMAL_TRIP_ACTIVE) { + last_active = i; } else { break; } } - if (last_passive != -1) { + if (last_passive != INVALID_TRIP) { params->trip_max_desired_temperature = last_passive; - ret = 0; + } else if (found_first_passive) { + params->trip_max_desired_temperature = params->trip_switch_on; + params->trip_switch_on = INVALID_TRIP; } else { - ret = -EINVAL; + params->trip_switch_on = INVALID_TRIP; + params->trip_max_desired_temperature = last_active; } - - return ret; } static void reset_pid_controller(struct power_allocator_params *params) @@ -405,60 +532,45 @@ static void allow_maximum_power(struct thermal_zone_device *tz) * power_allocator_bind() - bind the power_allocator governor to a thermal zone * @tz: thermal zone to bind it to * - * Check that the thermal zone is valid for this governor, that is, it - * has two thermal trips. If so, initialize the PID controller - * parameters and bind it to the thermal zone. + * Initialize the PID controller parameters and bind it to the thermal + * zone. * - * Return: 0 on success, -EINVAL if the trips were invalid or -ENOMEM - * if we ran out of memory. + * Return: 0 on success, or -ENOMEM if we ran out of memory. */ static int power_allocator_bind(struct thermal_zone_device *tz) { int ret; struct power_allocator_params *params; - int switch_on_temp, control_temp; - u32 temperature_threshold; - - if (!tz->tzp || !tz->tzp->sustainable_power) { - dev_err(&tz->device, - "power_allocator: missing sustainable_power\n"); - return -EINVAL; - } + int control_temp; params = kzalloc(sizeof(*params), GFP_KERNEL); if (!params) return -ENOMEM; - ret = get_governor_trips(tz, params); - if (ret) { - dev_err(&tz->device, - "thermal zone %s has wrong trip setup for power allocator\n", - tz->type); - goto free; - } + if (!tz->tzp) { + tz->tzp = kzalloc(sizeof(*tz->tzp), GFP_KERNEL); + if (!tz->tzp) { + ret = -ENOMEM; + goto free_params; + } - ret = tz->ops->get_trip_temp(tz, params->trip_switch_on, - &switch_on_temp); - if (ret) - goto free; + params->allocated_tzp = true; + } - ret = tz->ops->get_trip_temp(tz, params->trip_max_desired_temperature, - &control_temp); - if (ret) - goto free; + if (!tz->tzp->sustainable_power) + dev_warn(&tz->device, "power_allocator: sustainable_power will be estimated\n"); - temperature_threshold = control_temp - switch_on_temp; + get_governor_trips(tz, params); - tz->tzp->k_po = tz->tzp->k_po ?: - int_to_frac(tz->tzp->sustainable_power) / temperature_threshold; - tz->tzp->k_pu = tz->tzp->k_pu ?: - int_to_frac(2 * tz->tzp->sustainable_power) / - temperature_threshold; - tz->tzp->k_i = tz->tzp->k_i ?: int_to_frac(10) / 1000; - /* - * The default for k_d and integral_cutoff is 0, so we can - * leave them as they are. - */ + if (tz->trips > 0) { + ret = tz->ops->get_trip_temp(tz, + params->trip_max_desired_temperature, + &control_temp); + if (!ret) + estimate_pid_constants(tz, tz->tzp->sustainable_power, + params->trip_switch_on, + control_temp, false); + } reset_pid_controller(params); @@ -466,14 +578,23 @@ static int power_allocator_bind(struct thermal_zone_device *tz) return 0; -free: +free_params: kfree(params); + return ret; } static void power_allocator_unbind(struct thermal_zone_device *tz) { + struct power_allocator_params *params = tz->governor_data; + dev_dbg(&tz->device, "Unbinding from thermal zone %d\n", tz->id); + + if (params->allocated_tzp) { + kfree(tz->tzp); + tz->tzp = NULL; + } + kfree(tz->governor_data); tz->governor_data = NULL; } @@ -499,13 +620,7 @@ static int power_allocator_throttle(struct thermal_zone_device *tz, int trip) ret = tz->ops->get_trip_temp(tz, params->trip_switch_on, &switch_on_temp); - if (ret) { - dev_warn(&tz->device, - "Failed to get switch on temperature: %d\n", ret); - return ret; - } - - if (current_temp < switch_on_temp) { + if (!ret && (current_temp < switch_on_temp)) { tz->passive = 0; reset_pid_controller(params); allow_maximum_power(tz); diff --git a/drivers/thermal/thermal_core.c b/drivers/thermal/thermal_core.c index 5e5fc7015c7f..d9e525cc9c1c 100644 --- a/drivers/thermal/thermal_core.c +++ b/drivers/thermal/thermal_core.c @@ -1013,6 +1013,34 @@ int power_actor_get_max_power(struct thermal_cooling_device *cdev, } /** + * power_actor_get_min_power() - get the mainimum power that a cdev can consume + * @cdev: pointer to &thermal_cooling_device + * @tz: a valid thermal zone device pointer + * @min_power: pointer in which to store the minimum power + * + * Calculate the minimum power consumption in milliwatts that the + * cooling device can currently consume and store it in @min_power. + * + * Return: 0 on success, -EINVAL if @cdev doesn't support the + * power_actor API or -E* on other error. + */ +int power_actor_get_min_power(struct thermal_cooling_device *cdev, + struct thermal_zone_device *tz, u32 *min_power) +{ + unsigned long max_state; + int ret; + + if (!cdev_is_power_actor(cdev)) + return -EINVAL; + + ret = cdev->ops->get_max_state(cdev, &max_state); + if (ret) + return ret; + + return cdev->ops->state2power(cdev, tz, max_state, min_power); +} + +/** * power_actor_set_power() - limit the maximum power that a cooling device can consume * @cdev: pointer to &thermal_cooling_device * @instance: thermal instance to update diff --git a/drivers/thermal/ti-soc-thermal/Kconfig b/drivers/thermal/ti-soc-thermal/Kconfig index bd4c7beba679..cb6686ff09ae 100644 --- a/drivers/thermal/ti-soc-thermal/Kconfig +++ b/drivers/thermal/ti-soc-thermal/Kconfig @@ -1,7 +1,5 @@ config TI_SOC_THERMAL tristate "Texas Instruments SoCs temperature sensor driver" - depends on THERMAL - depends on ARCH_HAS_BANDGAP help If you say yes here you get support for the Texas Instruments OMAP4460+ on die bandgap temperature sensor support. The register @@ -24,7 +22,7 @@ config TI_THERMAL config OMAP4_THERMAL bool "Texas Instruments OMAP4 thermal support" depends on TI_SOC_THERMAL - depends on ARCH_OMAP4 + depends on ARCH_OMAP4 || COMPILE_TEST help If you say yes here you get thermal support for the Texas Instruments OMAP4 SoC family. The current chip supported are: @@ -38,7 +36,7 @@ config OMAP4_THERMAL config OMAP5_THERMAL bool "Texas Instruments OMAP5 thermal support" depends on TI_SOC_THERMAL - depends on SOC_OMAP5 + depends on SOC_OMAP5 || COMPILE_TEST help If you say yes here you get thermal support for the Texas Instruments OMAP5 SoC family. The current chip supported are: @@ -50,7 +48,7 @@ config OMAP5_THERMAL config DRA752_THERMAL bool "Texas Instruments DRA752 thermal support" depends on TI_SOC_THERMAL - depends on SOC_DRA7XX + depends on SOC_DRA7XX || COMPILE_TEST help If you say yes here you get thermal support for the Texas Instruments DRA752 SoC family. The current chip supported are: diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index 46b8b2bbc95a..ee5aa4daaea0 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -1439,6 +1439,7 @@ int dlm_master_request_handler(struct o2net_msg *msg, u32 len, void *data, int found, ret; int set_maybe; int dispatch_assert = 0; + int dispatched = 0; if (!dlm_grab(dlm)) return DLM_MASTER_RESP_NO; @@ -1658,15 +1659,18 @@ send_response: mlog(ML_ERROR, "failed to dispatch assert master work\n"); response = DLM_MASTER_RESP_ERROR; dlm_lockres_put(res); - } else + } else { + dispatched = 1; __dlm_lockres_grab_inflight_worker(dlm, res); + } spin_unlock(&res->spinlock); } else { if (res) dlm_lockres_put(res); } - dlm_put(dlm); + if (!dispatched) + dlm_put(dlm); return response; } @@ -2090,7 +2094,6 @@ int dlm_dispatch_assert_master(struct dlm_ctxt *dlm, /* queue up work for dlm_assert_master_worker */ - dlm_grab(dlm); /* get an extra ref for the work item */ dlm_init_work_item(dlm, item, dlm_assert_master_worker, NULL); item->u.am.lockres = res; /* already have a ref */ /* can optionally ignore node numbers higher than this node */ diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index ce12e0b1a31f..3d90ad7ff91f 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -1694,6 +1694,7 @@ int dlm_master_requery_handler(struct o2net_msg *msg, u32 len, void *data, unsigned int hash; int master = DLM_LOCK_RES_OWNER_UNKNOWN; u32 flags = DLM_ASSERT_MASTER_REQUERY; + int dispatched = 0; if (!dlm_grab(dlm)) { /* since the domain has gone away on this @@ -1719,8 +1720,10 @@ int dlm_master_requery_handler(struct o2net_msg *msg, u32 len, void *data, dlm_put(dlm); /* sender will take care of this and retry */ return ret; - } else + } else { + dispatched = 1; __dlm_lockres_grab_inflight_worker(dlm, res); + } spin_unlock(&res->spinlock); } else { /* put.. incase we are not the master */ @@ -1730,7 +1733,8 @@ int dlm_master_requery_handler(struct o2net_msg *msg, u32 len, void *data, } spin_unlock(&dlm->spinlock); - dlm_put(dlm); + if (!dispatched) + dlm_put(dlm); return master; } diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c index f9aeb40a7197..50311703135b 100644 --- a/fs/userfaultfd.c +++ b/fs/userfaultfd.c @@ -467,8 +467,8 @@ static int userfaultfd_release(struct inode *inode, struct file *file) * the fault_*wqh. */ spin_lock(&ctx->fault_pending_wqh.lock); - __wake_up_locked_key(&ctx->fault_pending_wqh, TASK_NORMAL, 0, &range); - __wake_up_locked_key(&ctx->fault_wqh, TASK_NORMAL, 0, &range); + __wake_up_locked_key(&ctx->fault_pending_wqh, TASK_NORMAL, &range); + __wake_up_locked_key(&ctx->fault_wqh, TASK_NORMAL, &range); spin_unlock(&ctx->fault_pending_wqh.lock); wake_up_poll(&ctx->fd_wqh, POLLHUP); @@ -650,10 +650,10 @@ static void __wake_userfault(struct userfaultfd_ctx *ctx, spin_lock(&ctx->fault_pending_wqh.lock); /* wake all in the range and autoremove */ if (waitqueue_active(&ctx->fault_pending_wqh)) - __wake_up_locked_key(&ctx->fault_pending_wqh, TASK_NORMAL, 0, + __wake_up_locked_key(&ctx->fault_pending_wqh, TASK_NORMAL, range); if (waitqueue_active(&ctx->fault_wqh)) - __wake_up_locked_key(&ctx->fault_wqh, TASK_NORMAL, 0, range); + __wake_up_locked_key(&ctx->fault_wqh, TASK_NORMAL, range); spin_unlock(&ctx->fault_pending_wqh.lock); } diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h index 4d8fcf2187dc..8492721b39be 100644 --- a/include/linux/cgroup-defs.h +++ b/include/linux/cgroup-defs.h @@ -473,31 +473,8 @@ struct cgroup_subsys { unsigned int depends_on; }; -extern struct percpu_rw_semaphore cgroup_threadgroup_rwsem; - -/** - * cgroup_threadgroup_change_begin - threadgroup exclusion for cgroups - * @tsk: target task - * - * Called from threadgroup_change_begin() and allows cgroup operations to - * synchronize against threadgroup changes using a percpu_rw_semaphore. - */ -static inline void cgroup_threadgroup_change_begin(struct task_struct *tsk) -{ - percpu_down_read(&cgroup_threadgroup_rwsem); -} - -/** - * cgroup_threadgroup_change_end - threadgroup exclusion for cgroups - * @tsk: target task - * - * Called from threadgroup_change_end(). Counterpart of - * cgroup_threadcgroup_change_begin(). - */ -static inline void cgroup_threadgroup_change_end(struct task_struct *tsk) -{ - percpu_up_read(&cgroup_threadgroup_rwsem); -} +void cgroup_threadgroup_change_begin(struct task_struct *tsk); +void cgroup_threadgroup_change_end(struct task_struct *tsk); #else /* CONFIG_CGROUPS */ diff --git a/include/linux/init_task.h b/include/linux/init_task.h index d0b380ee7d67..e38681f4912d 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -25,6 +25,13 @@ extern struct files_struct init_files; extern struct fs_struct init_fs; +#ifdef CONFIG_CGROUPS +#define INIT_GROUP_RWSEM(sig) \ + .group_rwsem = __RWSEM_INITIALIZER(sig.group_rwsem), +#else +#define INIT_GROUP_RWSEM(sig) +#endif + #ifdef CONFIG_CPUSETS #define INIT_CPUSET_SEQ(tsk) \ .mems_allowed_seq = SEQCNT_ZERO(tsk.mems_allowed_seq), @@ -57,6 +64,7 @@ extern struct fs_struct init_fs; INIT_PREV_CPUTIME(sig) \ .cred_guard_mutex = \ __MUTEX_INITIALIZER(sig.cred_guard_mutex), \ + INIT_GROUP_RWSEM(sig) \ } extern struct nsproxy init_nsproxy; diff --git a/include/linux/sched.h b/include/linux/sched.h index a4ab9daa387c..b7b9501b41af 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -762,6 +762,18 @@ struct signal_struct { unsigned audit_tty_log_passwd; struct tty_audit_buf *tty_audit_buf; #endif +#ifdef CONFIG_CGROUPS + /* + * group_rwsem prevents new tasks from entering the threadgroup and + * member tasks from exiting,a more specifically, setting of + * PF_EXITING. fork and exit paths are protected with this rwsem + * using threadgroup_change_begin/end(). Users which require + * threadgroup to remain stable should use threadgroup_[un]lock() + * which also takes care of exec path. Currently, cgroup is the + * only user. + */ + struct rw_semaphore group_rwsem; +#endif oom_flags_t oom_flags; short oom_score_adj; /* OOM kill score adjustment */ diff --git a/include/linux/security.h b/include/linux/security.h index 79d85ddf8093..2f4c1f7aa7db 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -946,7 +946,7 @@ static inline int security_task_prctl(int option, unsigned long arg2, unsigned long arg4, unsigned long arg5) { - return cap_task_prctl(option, arg2, arg3, arg3, arg5); + return cap_task_prctl(option, arg2, arg3, arg4, arg5); } static inline void security_task_to_inode(struct task_struct *p, struct inode *inode) diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h index 269e8afd3e2a..6b00f18f5e6b 100644 --- a/include/linux/spi/spi.h +++ b/include/linux/spi/spi.h @@ -34,7 +34,7 @@ extern struct bus_type spi_bus_type; /** * struct spi_statistics - statistics for spi transfers - * @clock: lock protecting this structure + * @lock: lock protecting this structure * * @messages: number of spi-messages handled * @transfers: number of spi_transfers handled diff --git a/include/linux/thermal.h b/include/linux/thermal.h index 17292fee8686..157d366e761b 100644 --- a/include/linux/thermal.h +++ b/include/linux/thermal.h @@ -360,7 +360,7 @@ static inline struct thermal_zone_device * thermal_zone_of_sensor_register(struct device *dev, int id, void *data, const struct thermal_zone_of_device_ops *ops) { - return NULL; + return ERR_PTR(-ENODEV); } static inline @@ -380,6 +380,8 @@ static inline bool cdev_is_power_actor(struct thermal_cooling_device *cdev) int power_actor_get_max_power(struct thermal_cooling_device *, struct thermal_zone_device *tz, u32 *max_power); +int power_actor_get_min_power(struct thermal_cooling_device *, + struct thermal_zone_device *tz, u32 *min_power); int power_actor_set_power(struct thermal_cooling_device *, struct thermal_instance *, u32); struct thermal_zone_device *thermal_zone_device_register(const char *, int, int, @@ -415,6 +417,10 @@ static inline bool cdev_is_power_actor(struct thermal_cooling_device *cdev) static inline int power_actor_get_max_power(struct thermal_cooling_device *cdev, struct thermal_zone_device *tz, u32 *max_power) { return 0; } +static inline int power_actor_get_min_power(struct thermal_cooling_device *cdev, + struct thermal_zone_device *tz, + u32 *min_power) +{ return -ENODEV; } static inline int power_actor_set_power(struct thermal_cooling_device *cdev, struct thermal_instance *tz, u32 power) { return 0; } diff --git a/include/linux/wait.h b/include/linux/wait.h index d3d077228d4c..1e1bf9f963a9 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -147,8 +147,7 @@ __remove_wait_queue(wait_queue_head_t *head, wait_queue_t *old) typedef int wait_bit_action_f(struct wait_bit_key *); void __wake_up(wait_queue_head_t *q, unsigned int mode, int nr, void *key); -void __wake_up_locked_key(wait_queue_head_t *q, unsigned int mode, int nr, - void *key); +void __wake_up_locked_key(wait_queue_head_t *q, unsigned int mode, void *key); void __wake_up_sync_key(wait_queue_head_t *q, unsigned int mode, int nr, void *key); void __wake_up_locked(wait_queue_head_t *q, unsigned int mode, int nr); void __wake_up_sync(wait_queue_head_t *q, unsigned int mode, int nr); @@ -180,7 +179,7 @@ wait_queue_head_t *bit_waitqueue(void *, int); #define wake_up_poll(x, m) \ __wake_up(x, TASK_NORMAL, 1, (void *) (m)) #define wake_up_locked_poll(x, m) \ - __wake_up_locked_key((x), TASK_NORMAL, 1, (void *) (m)) + __wake_up_locked_key((x), TASK_NORMAL, (void *) (m)) #define wake_up_interruptible_poll(x, m) \ __wake_up(x, TASK_INTERRUPTIBLE, 1, (void *) (m)) #define wake_up_interruptible_sync_poll(x, m) \ diff --git a/include/uapi/asm-generic/unistd.h b/include/uapi/asm-generic/unistd.h index 8da542a2874d..ee124009e12a 100644 --- a/include/uapi/asm-generic/unistd.h +++ b/include/uapi/asm-generic/unistd.h @@ -709,17 +709,19 @@ __SYSCALL(__NR_memfd_create, sys_memfd_create) __SYSCALL(__NR_bpf, sys_bpf) #define __NR_execveat 281 __SC_COMP(__NR_execveat, sys_execveat, compat_sys_execveat) -#define __NR_membarrier 282 +#define __NR_userfaultfd 282 +__SYSCALL(__NR_userfaultfd, sys_userfaultfd) +#define __NR_membarrier 283 __SYSCALL(__NR_membarrier, sys_membarrier) #undef __NR_syscalls -#define __NR_syscalls 283 +#define __NR_syscalls 284 /* * All syscalls below here should go away really, * these are provided for both review and as a porting * help for the C library version. -* + * * Last chance: are any of these important enough to * enable by default? */ diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 2cf0f79f1fc9..2c9eae6ad970 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -46,7 +46,6 @@ #include <linux/slab.h> #include <linux/spinlock.h> #include <linux/rwsem.h> -#include <linux/percpu-rwsem.h> #include <linux/string.h> #include <linux/sort.h> #include <linux/kmod.h> @@ -104,8 +103,6 @@ static DEFINE_SPINLOCK(cgroup_idr_lock); */ static DEFINE_SPINLOCK(release_agent_path_lock); -struct percpu_rw_semaphore cgroup_threadgroup_rwsem; - #define cgroup_assert_mutex_or_rcu_locked() \ RCU_LOCKDEP_WARN(!rcu_read_lock_held() && \ !lockdep_is_held(&cgroup_mutex), \ @@ -874,6 +871,48 @@ static struct css_set *find_css_set(struct css_set *old_cset, return cset; } +void cgroup_threadgroup_change_begin(struct task_struct *tsk) +{ + down_read(&tsk->signal->group_rwsem); +} + +void cgroup_threadgroup_change_end(struct task_struct *tsk) +{ + up_read(&tsk->signal->group_rwsem); +} + +/** + * threadgroup_lock - lock threadgroup + * @tsk: member task of the threadgroup to lock + * + * Lock the threadgroup @tsk belongs to. No new task is allowed to enter + * and member tasks aren't allowed to exit (as indicated by PF_EXITING) or + * change ->group_leader/pid. This is useful for cases where the threadgroup + * needs to stay stable across blockable operations. + * + * fork and exit explicitly call threadgroup_change_{begin|end}() for + * synchronization. While held, no new task will be added to threadgroup + * and no existing live task will have its PF_EXITING set. + * + * de_thread() does threadgroup_change_{begin|end}() when a non-leader + * sub-thread becomes a new leader. + */ +static void threadgroup_lock(struct task_struct *tsk) +{ + down_write(&tsk->signal->group_rwsem); +} + +/** + * threadgroup_unlock - unlock threadgroup + * @tsk: member task of the threadgroup to unlock + * + * Reverse threadgroup_lock(). + */ +static inline void threadgroup_unlock(struct task_struct *tsk) +{ + up_write(&tsk->signal->group_rwsem); +} + static struct cgroup_root *cgroup_root_from_kf(struct kernfs_root *kf_root) { struct cgroup *root_cgrp = kf_root->kn->priv; @@ -2074,9 +2113,9 @@ static void cgroup_task_migrate(struct cgroup *old_cgrp, lockdep_assert_held(&css_set_rwsem); /* - * We are synchronized through cgroup_threadgroup_rwsem against - * PF_EXITING setting such that we can't race against cgroup_exit() - * changing the css_set to init_css_set and dropping the old one. + * We are synchronized through threadgroup_lock() against PF_EXITING + * setting such that we can't race against cgroup_exit() changing the + * css_set to init_css_set and dropping the old one. */ WARN_ON_ONCE(tsk->flags & PF_EXITING); old_cset = task_css_set(tsk); @@ -2133,11 +2172,10 @@ static void cgroup_migrate_finish(struct list_head *preloaded_csets) * @src_cset and add it to @preloaded_csets, which should later be cleaned * up by cgroup_migrate_finish(). * - * This function may be called without holding cgroup_threadgroup_rwsem - * even if the target is a process. Threads may be created and destroyed - * but as long as cgroup_mutex is not dropped, no new css_set can be put - * into play and the preloaded css_sets are guaranteed to cover all - * migrations. + * This function may be called without holding threadgroup_lock even if the + * target is a process. Threads may be created and destroyed but as long + * as cgroup_mutex is not dropped, no new css_set can be put into play and + * the preloaded css_sets are guaranteed to cover all migrations. */ static void cgroup_migrate_add_src(struct css_set *src_cset, struct cgroup *dst_cgrp, @@ -2240,7 +2278,7 @@ err: * @threadgroup: whether @leader points to the whole process or a single task * * Migrate a process or task denoted by @leader to @cgrp. If migrating a - * process, the caller must be holding cgroup_threadgroup_rwsem. The + * process, the caller must be holding threadgroup_lock of @leader. The * caller is also responsible for invoking cgroup_migrate_add_src() and * cgroup_migrate_prepare_dst() on the targets before invoking this * function and following up with cgroup_migrate_finish(). @@ -2368,7 +2406,7 @@ out_release_tset: * @leader: the task or the leader of the threadgroup to be attached * @threadgroup: attach the whole threadgroup? * - * Call holding cgroup_mutex and cgroup_threadgroup_rwsem. + * Call holding cgroup_mutex and threadgroup_lock of @leader. */ static int cgroup_attach_task(struct cgroup *dst_cgrp, struct task_struct *leader, bool threadgroup) @@ -2460,13 +2498,14 @@ static ssize_t __cgroup_procs_write(struct kernfs_open_file *of, char *buf, if (!cgrp) return -ENODEV; - percpu_down_write(&cgroup_threadgroup_rwsem); +retry_find_task: rcu_read_lock(); if (pid) { tsk = find_task_by_vpid(pid); if (!tsk) { + rcu_read_unlock(); ret = -ESRCH; - goto out_unlock_rcu; + goto out_unlock_cgroup; } } else { tsk = current; @@ -2482,23 +2521,37 @@ static ssize_t __cgroup_procs_write(struct kernfs_open_file *of, char *buf, */ if (tsk == kthreadd_task || (tsk->flags & PF_NO_SETAFFINITY)) { ret = -EINVAL; - goto out_unlock_rcu; + rcu_read_unlock(); + goto out_unlock_cgroup; } get_task_struct(tsk); rcu_read_unlock(); + threadgroup_lock(tsk); + if (threadgroup) { + if (!thread_group_leader(tsk)) { + /* + * a race with de_thread from another thread's exec() + * may strip us of our leadership, if this happens, + * there is no choice but to throw this task away and + * try again; this is + * "double-double-toil-and-trouble-check locking". + */ + threadgroup_unlock(tsk); + put_task_struct(tsk); + goto retry_find_task; + } + } + ret = cgroup_procs_write_permission(tsk, cgrp, of); if (!ret) ret = cgroup_attach_task(cgrp, tsk, threadgroup); - put_task_struct(tsk); - goto out_unlock_threadgroup; + threadgroup_unlock(tsk); -out_unlock_rcu: - rcu_read_unlock(); -out_unlock_threadgroup: - percpu_up_write(&cgroup_threadgroup_rwsem); + put_task_struct(tsk); +out_unlock_cgroup: cgroup_kn_unlock(of->kn); return ret ?: nbytes; } @@ -2643,8 +2696,6 @@ static int cgroup_update_dfl_csses(struct cgroup *cgrp) lockdep_assert_held(&cgroup_mutex); - percpu_down_write(&cgroup_threadgroup_rwsem); - /* look up all csses currently attached to @cgrp's subtree */ down_read(&css_set_rwsem); css_for_each_descendant_pre(css, cgroup_css(cgrp, NULL)) { @@ -2700,8 +2751,17 @@ static int cgroup_update_dfl_csses(struct cgroup *cgrp) goto out_finish; last_task = task; + threadgroup_lock(task); + /* raced against de_thread() from another thread? */ + if (!thread_group_leader(task)) { + threadgroup_unlock(task); + put_task_struct(task); + continue; + } + ret = cgroup_migrate(src_cset->dfl_cgrp, task, true); + threadgroup_unlock(task); put_task_struct(task); if (WARN(ret, "cgroup: failed to update controllers for the default hierarchy (%d), further operations may crash or hang\n", ret)) @@ -2711,7 +2771,6 @@ static int cgroup_update_dfl_csses(struct cgroup *cgrp) out_finish: cgroup_migrate_finish(&preloaded_csets); - percpu_up_write(&cgroup_threadgroup_rwsem); return ret; } @@ -5024,7 +5083,6 @@ int __init cgroup_init(void) unsigned long key; int ssid, err; - BUG_ON(percpu_init_rwsem(&cgroup_threadgroup_rwsem)); BUG_ON(cgroup_init_cftypes(NULL, cgroup_dfl_base_files)); BUG_ON(cgroup_init_cftypes(NULL, cgroup_legacy_base_files)); diff --git a/kernel/fork.c b/kernel/fork.c index 7d5f0f118a63..2845623fb582 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1149,6 +1149,10 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) tty_audit_fork(sig); sched_autogroup_fork(sig); +#ifdef CONFIG_CGROUPS + init_rwsem(&sig->group_rwsem); +#endif + sig->oom_score_adj = current->signal->oom_score_adj; sig->oom_score_adj_min = current->signal->oom_score_adj_min; diff --git a/kernel/sched/wait.c b/kernel/sched/wait.c index 272d9322bc5d..052e02672d12 100644 --- a/kernel/sched/wait.c +++ b/kernel/sched/wait.c @@ -106,10 +106,9 @@ void __wake_up_locked(wait_queue_head_t *q, unsigned int mode, int nr) } EXPORT_SYMBOL_GPL(__wake_up_locked); -void __wake_up_locked_key(wait_queue_head_t *q, unsigned int mode, int nr, - void *key) +void __wake_up_locked_key(wait_queue_head_t *q, unsigned int mode, void *key) { - __wake_up_common(q, mode, nr, 0, key); + __wake_up_common(q, mode, 1, 0, key); } EXPORT_SYMBOL_GPL(__wake_up_locked_key); @@ -284,7 +283,7 @@ void abort_exclusive_wait(wait_queue_head_t *q, wait_queue_t *wait, if (!list_empty(&wait->task_list)) list_del_init(&wait->task_list); else if (waitqueue_active(q)) - __wake_up_locked_key(q, mode, 1, key); + __wake_up_locked_key(q, mode, key); spin_unlock_irqrestore(&q->lock, flags); } EXPORT_SYMBOL(abort_exclusive_wait); diff --git a/lib/iommu-common.c b/lib/iommu-common.c index ff19f66d3f7f..b1c93e94ca7a 100644 --- a/lib/iommu-common.c +++ b/lib/iommu-common.c @@ -21,8 +21,7 @@ static DEFINE_PER_CPU(unsigned int, iommu_hash_common); static inline bool need_flush(struct iommu_map_table *iommu) { - return (iommu->lazy_flush != NULL && - (iommu->flags & IOMMU_NEED_FLUSH) != 0); + return ((iommu->flags & IOMMU_NEED_FLUSH) != 0); } static inline void set_flush(struct iommu_map_table *iommu) @@ -211,7 +210,8 @@ unsigned long iommu_tbl_range_alloc(struct device *dev, goto bail; } } - if (n < pool->hint || need_flush(iommu)) { + if (iommu->lazy_flush && + (n < pool->hint || need_flush(iommu))) { clear_flush(iommu); iommu->lazy_flush(iommu); } diff --git a/mm/migrate.c b/mm/migrate.c index c3cb566af3e2..7452a00bbb50 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1075,7 +1075,7 @@ out: if (rc != MIGRATEPAGE_SUCCESS && put_new_page) put_new_page(new_hpage, private); else - put_page(new_hpage); + putback_active_hugepage(new_hpage); if (result) { if (rc) diff --git a/mm/mmap.c b/mm/mmap.c index c739d6db7193..79bcc9f92e48 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1490,13 +1490,14 @@ SYSCALL_DEFINE1(old_mmap, struct mmap_arg_struct __user *, arg) int vma_wants_writenotify(struct vm_area_struct *vma) { vm_flags_t vm_flags = vma->vm_flags; + const struct vm_operations_struct *vm_ops = vma->vm_ops; /* If it was private or non-writable, the write bit is already clear */ if ((vm_flags & (VM_WRITE|VM_SHARED)) != ((VM_WRITE|VM_SHARED))) return 0; /* The backer wishes to know when pages are first written to? */ - if (vma->vm_ops && vma->vm_ops->page_mkwrite) + if (vm_ops && (vm_ops->page_mkwrite || vm_ops->pfn_mkwrite)) return 1; /* The open routine did something to the protections that pgprot_modify diff --git a/mm/vmscan.c b/mm/vmscan.c index 2d978b28a410..7f63a9381f71 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -175,7 +175,7 @@ static bool sane_reclaim(struct scan_control *sc) if (!memcg) return true; #ifdef CONFIG_CGROUP_WRITEBACK - if (memcg->css.cgroup) + if (cgroup_on_dfl(memcg->css.cgroup)) return true; #endif return false; diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index b140c092d226..337ca851a350 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -297,7 +297,7 @@ static int rpc_complete_task(struct rpc_task *task) clear_bit(RPC_TASK_ACTIVE, &task->tk_runstate); ret = atomic_dec_and_test(&task->tk_count); if (waitqueue_active(wq)) - __wake_up_locked_key(wq, TASK_NORMAL, 1, &k); + __wake_up_locked_key(wq, TASK_NORMAL, &k); spin_unlock_irqrestore(&wq->lock, flags); return ret; } diff --git a/tools/testing/selftests/membarrier/Makefile b/tools/testing/selftests/membarrier/Makefile index 877a50355d7f..a1a97085847d 100644 --- a/tools/testing/selftests/membarrier/Makefile +++ b/tools/testing/selftests/membarrier/Makefile @@ -1,11 +1,10 @@ CFLAGS += -g -I../../../../usr/include/ -all: - $(CC) $(CFLAGS) membarrier_test.c -o membarrier_test - TEST_PROGS := membarrier_test +all: $(TEST_PROGS) + include ../lib.mk clean: - $(RM) membarrier_test + $(RM) $(TEST_PROGS) diff --git a/tools/testing/selftests/membarrier/membarrier_test.c b/tools/testing/selftests/membarrier/membarrier_test.c index dde312508007..535f0fef4d0b 100644 --- a/tools/testing/selftests/membarrier/membarrier_test.c +++ b/tools/testing/selftests/membarrier/membarrier_test.c @@ -1,9 +1,6 @@ #define _GNU_SOURCE -#define __EXPORTED_HEADERS__ - #include <linux/membarrier.h> -#include <asm-generic/unistd.h> -#include <sys/syscall.h> +#include <syscall.h> #include <stdio.h> #include <errno.h> #include <string.h> diff --git a/tools/testing/selftests/vm/Makefile b/tools/testing/selftests/vm/Makefile index d36fab7d8ebd..3c53cac15de1 100644 --- a/tools/testing/selftests/vm/Makefile +++ b/tools/testing/selftests/vm/Makefile @@ -1,6 +1,6 @@ # Makefile for vm selftests -CFLAGS = -Wall +CFLAGS = -Wall -I ../../../../usr/include $(EXTRA_CFLAGS) BINARIES = compaction_test BINARIES += hugepage-mmap BINARIES += hugepage-shm @@ -12,8 +12,11 @@ BINARIES += userfaultfd all: $(BINARIES) %: %.c $(CC) $(CFLAGS) -o $@ $^ -lrt -userfaultfd: userfaultfd.c - $(CC) $(CFLAGS) -O2 -o $@ $^ -lpthread +userfaultfd: userfaultfd.c ../../../../usr/include/linux/kernel.h + $(CC) $(CFLAGS) -O2 -o $@ $< -lpthread + +../../../../usr/include/linux/kernel.h: + make -C ../../../.. headers_install TEST_PROGS := run_vmtests TEST_FILES := $(BINARIES) diff --git a/tools/testing/selftests/vm/userfaultfd.c b/tools/testing/selftests/vm/userfaultfd.c index 2c7cca6f26a4..d77ed41b2094 100644 --- a/tools/testing/selftests/vm/userfaultfd.c +++ b/tools/testing/selftests/vm/userfaultfd.c @@ -64,17 +64,9 @@ #include <sys/syscall.h> #include <sys/ioctl.h> #include <pthread.h> -#include "../../../../include/uapi/linux/userfaultfd.h" - -#ifdef __x86_64__ -#define __NR_userfaultfd 323 -#elif defined(__i386__) -#define __NR_userfaultfd 374 -#elif defined(__powewrpc__) -#define __NR_userfaultfd 364 -#else -#error "missing __NR_userfaultfd definition" -#endif +#include <linux/userfaultfd.h> + +#ifdef __NR_userfaultfd static unsigned long nr_cpus, nr_pages, nr_pages_per_cpu, page_size; @@ -430,7 +422,7 @@ static int userfaultfd_stress(void) struct uffdio_register uffdio_register; struct uffdio_api uffdio_api; unsigned long cpu; - int uffd_flags; + int uffd_flags, err; unsigned long userfaults[nr_cpus]; if (posix_memalign(&area, page_size, nr_pages * page_size)) { @@ -473,6 +465,14 @@ static int userfaultfd_stress(void) *area_mutex(area_src, nr) = (pthread_mutex_t) PTHREAD_MUTEX_INITIALIZER; count_verify[nr] = *area_count(area_src, nr) = 1; + /* + * In the transition between 255 to 256, powerpc will + * read out of order in my_bcmp and see both bytes as + * zero, so leave a placeholder below always non-zero + * after the count, to avoid my_bcmp to trigger false + * positives. + */ + *(area_count(area_src, nr) + 1) = 1; } pipefd = malloc(sizeof(int) * nr_cpus * 2); @@ -499,6 +499,7 @@ static int userfaultfd_stress(void) pthread_attr_init(&attr); pthread_attr_setstacksize(&attr, 16*1024*1024); + err = 0; while (bounces--) { unsigned long expected_ioctls; @@ -579,20 +580,13 @@ static int userfaultfd_stress(void) /* verification */ if (bounces & BOUNCE_VERIFY) { for (nr = 0; nr < nr_pages; nr++) { - if (my_bcmp(area_dst, - area_dst + nr * page_size, - sizeof(pthread_mutex_t))) { - fprintf(stderr, - "error mutex 2 %lu\n", - nr); - bounces = 0; - } if (*area_count(area_dst, nr) != count_verify[nr]) { fprintf(stderr, "error area_count %Lu %Lu %lu\n", *area_count(area_src, nr), count_verify[nr], nr); + err = 1; bounces = 0; } } @@ -609,7 +603,7 @@ static int userfaultfd_stress(void) printf("\n"); } - return 0; + return err; } int main(int argc, char **argv) @@ -618,8 +612,8 @@ int main(int argc, char **argv) fprintf(stderr, "Usage: <MiB> <bounces>\n"), exit(1); nr_cpus = sysconf(_SC_NPROCESSORS_ONLN); page_size = sysconf(_SC_PAGE_SIZE); - if ((unsigned long) area_count(NULL, 0) + sizeof(unsigned long long) > - page_size) + if ((unsigned long) area_count(NULL, 0) + sizeof(unsigned long long) * 2 + > page_size) fprintf(stderr, "Impossible to run this test\n"), exit(2); nr_pages_per_cpu = atol(argv[1]) * 1024*1024 / page_size / nr_cpus; @@ -637,3 +631,15 @@ int main(int argc, char **argv) nr_pages, nr_pages_per_cpu); return userfaultfd_stress(); } + +#else /* __NR_userfaultfd */ + +#warning "missing __NR_userfaultfd definition" + +int main(void) +{ + printf("skip: Skipping userfaultfd test (missing __NR_userfaultfd)\n"); + return 0; +} + +#endif /* __NR_userfaultfd */ diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 04146a2e1d81..8db1d9361993 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -66,8 +66,8 @@ MODULE_AUTHOR("Qumranet"); MODULE_LICENSE("GPL"); -/* halt polling only reduces halt latency by 5-7 us, 500us is enough */ -static unsigned int halt_poll_ns = 500000; +/* Architectures should define their poll value according to the halt latency */ +static unsigned int halt_poll_ns = KVM_HALT_POLL_NS_DEFAULT; module_param(halt_poll_ns, uint, S_IRUGO | S_IWUSR); /* Default doubles per-vcpu halt_poll_ns. */ |