diff options
Diffstat (limited to 'sys')
96 files changed, 1555 insertions, 794 deletions
diff --git a/sys/amd64/sgx/sgx_linux.c b/sys/amd64/sgx/sgx_linux.c index 6ecef9207a38..d389edc1b2b0 100644 --- a/sys/amd64/sgx/sgx_linux.c +++ b/sys/amd64/sgx/sgx_linux.c @@ -92,16 +92,7 @@ out: return (error); } -static struct linux_ioctl_handler sgx_linux_handler = { - sgx_linux_ioctl, - SGX_LINUX_IOCTL_MIN, - SGX_LINUX_IOCTL_MAX, -}; - -SYSINIT(sgx_linux_register, SI_SUB_KLD, SI_ORDER_MIDDLE, - linux_ioctl_register_handler, &sgx_linux_handler); -SYSUNINIT(sgx_linux_unregister, SI_SUB_KLD, SI_ORDER_MIDDLE, - linux_ioctl_unregister_handler, &sgx_linux_handler); +LINUX_IOCTL_SET(sgx, SGX_LINUX_IOCTL_MIN, SGX_LINUX_IOCTL_MAX); static int sgx_linux_modevent(module_t mod, int type, void *data) diff --git a/sys/amd64/vmm/vmm.c b/sys/amd64/vmm/vmm.c index f7c59847140b..473887240b9b 100644 --- a/sys/amd64/vmm/vmm.c +++ b/sys/amd64/vmm/vmm.c @@ -870,7 +870,7 @@ vm_assign_pptdev(struct vm *vm, int bus, int slot, int func) int vm_get_register(struct vcpu *vcpu, int reg, uint64_t *retval) { - + /* Negative values represent VM control structure fields. */ if (reg >= VM_REG_LAST) return (EINVAL); @@ -882,6 +882,7 @@ vm_set_register(struct vcpu *vcpu, int reg, uint64_t val) { int error; + /* Negative values represent VM control structure fields. */ if (reg >= VM_REG_LAST) return (EINVAL); diff --git a/sys/amd64/vmm/vmm_dev_machdep.c b/sys/amd64/vmm/vmm_dev_machdep.c index dfebc9dcadbf..b84be809ea24 100644 --- a/sys/amd64/vmm/vmm_dev_machdep.c +++ b/sys/amd64/vmm/vmm_dev_machdep.c @@ -124,12 +124,16 @@ const struct vmmdev_ioctl vmmdev_machdep_ioctls[] = { VMMDEV_IOCTL(VM_SET_KERNEMU_DEV, VMMDEV_IOCTL_LOCK_ONE_VCPU), VMMDEV_IOCTL(VM_BIND_PPTDEV, - VMMDEV_IOCTL_XLOCK_MEMSEGS | VMMDEV_IOCTL_LOCK_ALL_VCPUS), + VMMDEV_IOCTL_XLOCK_MEMSEGS | VMMDEV_IOCTL_LOCK_ALL_VCPUS | + VMMDEV_IOCTL_PRIV_CHECK_DRIVER), VMMDEV_IOCTL(VM_UNBIND_PPTDEV, - VMMDEV_IOCTL_XLOCK_MEMSEGS | VMMDEV_IOCTL_LOCK_ALL_VCPUS), + VMMDEV_IOCTL_XLOCK_MEMSEGS | VMMDEV_IOCTL_LOCK_ALL_VCPUS | + VMMDEV_IOCTL_PRIV_CHECK_DRIVER), - VMMDEV_IOCTL(VM_MAP_PPTDEV_MMIO, VMMDEV_IOCTL_LOCK_ALL_VCPUS), - VMMDEV_IOCTL(VM_UNMAP_PPTDEV_MMIO, VMMDEV_IOCTL_LOCK_ALL_VCPUS), + VMMDEV_IOCTL(VM_MAP_PPTDEV_MMIO, VMMDEV_IOCTL_LOCK_ALL_VCPUS | + VMMDEV_IOCTL_PRIV_CHECK_DRIVER), + VMMDEV_IOCTL(VM_UNMAP_PPTDEV_MMIO, VMMDEV_IOCTL_LOCK_ALL_VCPUS | + VMMDEV_IOCTL_PRIV_CHECK_DRIVER), #ifdef BHYVE_SNAPSHOT #ifdef COMPAT_FREEBSD13 VMMDEV_IOCTL(VM_SNAPSHOT_REQ_13, VMMDEV_IOCTL_LOCK_ALL_VCPUS), @@ -147,9 +151,9 @@ const struct vmmdev_ioctl vmmdev_machdep_ioctls[] = { VMMDEV_IOCTL(VM_LAPIC_LOCAL_IRQ, VMMDEV_IOCTL_MAYBE_ALLOC_VCPU), - VMMDEV_IOCTL(VM_PPTDEV_MSI, 0), - VMMDEV_IOCTL(VM_PPTDEV_MSIX, 0), - VMMDEV_IOCTL(VM_PPTDEV_DISABLE_MSIX, 0), + VMMDEV_IOCTL(VM_PPTDEV_MSI, VMMDEV_IOCTL_PRIV_CHECK_DRIVER), + VMMDEV_IOCTL(VM_PPTDEV_MSIX, VMMDEV_IOCTL_PRIV_CHECK_DRIVER), + VMMDEV_IOCTL(VM_PPTDEV_DISABLE_MSIX, VMMDEV_IOCTL_PRIV_CHECK_DRIVER), VMMDEV_IOCTL(VM_LAPIC_MSI, 0), VMMDEV_IOCTL(VM_IOAPIC_ASSERT_IRQ, 0), VMMDEV_IOCTL(VM_IOAPIC_DEASSERT_IRQ, 0), @@ -172,40 +176,13 @@ int vmmdev_machdep_ioctl(struct vm *vm, struct vcpu *vcpu, u_long cmd, caddr_t data, int fflag, struct thread *td) { - struct vm_seg_desc *vmsegdesc; - struct vm_run *vmrun; -#ifdef COMPAT_FREEBSD13 - struct vm_run_13 *vmrun_13; -#endif - struct vm_exception *vmexc; - struct vm_lapic_irq *vmirq; - struct vm_lapic_msi *vmmsi; - struct vm_ioapic_irq *ioapic_irq; - struct vm_isa_irq *isa_irq; - struct vm_isa_irq_trigger *isa_irq_trigger; - struct vm_pptdev *pptdev; - struct vm_pptdev_mmio *pptmmio; - struct vm_pptdev_msi *pptmsi; - struct vm_pptdev_msix *pptmsix; - struct vm_x2apic *x2apic; - struct vm_gpa_pte *gpapte; - struct vm_gla2gpa *gg; - struct vm_intinfo *vmii; - struct vm_rtc_time *rtctime; - struct vm_rtc_data *rtcdata; - struct vm_readwrite_kernemu_device *kernemu; -#ifdef BHYVE_SNAPSHOT - struct vm_snapshot_meta *snapshot_meta; -#ifdef COMPAT_FREEBSD13 - struct vm_snapshot_meta_13 *snapshot_13; -#endif -#endif int error; error = 0; switch (cmd) { case VM_RUN: { struct vm_exit *vme; + struct vm_run *vmrun; vmrun = (struct vm_run *)data; vme = vm_exitinfo(vcpu); @@ -243,6 +220,7 @@ vmmdev_machdep_ioctl(struct vm *vm, struct vcpu *vcpu, u_long cmd, caddr_t data, case VM_RUN_13: { struct vm_exit *vme; struct vm_exit_13 *vme_13; + struct vm_run_13 *vmrun_13; vmrun_13 = (struct vm_run_13 *)data; vme_13 = &vmrun_13->vm_exit; @@ -281,85 +259,123 @@ vmmdev_machdep_ioctl(struct vm *vm, struct vcpu *vcpu, u_long cmd, caddr_t data, break; } #endif - case VM_PPTDEV_MSI: + case VM_PPTDEV_MSI: { + struct vm_pptdev_msi *pptmsi; + pptmsi = (struct vm_pptdev_msi *)data; - error = ppt_setup_msi(vm, - pptmsi->bus, pptmsi->slot, pptmsi->func, - pptmsi->addr, pptmsi->msg, - pptmsi->numvec); + error = ppt_setup_msi(vm, pptmsi->bus, pptmsi->slot, + pptmsi->func, pptmsi->addr, pptmsi->msg, pptmsi->numvec); break; - case VM_PPTDEV_MSIX: + } + case VM_PPTDEV_MSIX: { + struct vm_pptdev_msix *pptmsix; + pptmsix = (struct vm_pptdev_msix *)data; - error = ppt_setup_msix(vm, - pptmsix->bus, pptmsix->slot, - pptmsix->func, pptmsix->idx, - pptmsix->addr, pptmsix->msg, - pptmsix->vector_control); + error = ppt_setup_msix(vm, pptmsix->bus, pptmsix->slot, + pptmsix->func, pptmsix->idx, pptmsix->addr, pptmsix->msg, + pptmsix->vector_control); break; - case VM_PPTDEV_DISABLE_MSIX: + } + case VM_PPTDEV_DISABLE_MSIX: { + struct vm_pptdev *pptdev; + pptdev = (struct vm_pptdev *)data; error = ppt_disable_msix(vm, pptdev->bus, pptdev->slot, - pptdev->func); + pptdev->func); break; - case VM_MAP_PPTDEV_MMIO: + } + case VM_MAP_PPTDEV_MMIO: { + struct vm_pptdev_mmio *pptmmio; + pptmmio = (struct vm_pptdev_mmio *)data; error = ppt_map_mmio(vm, pptmmio->bus, pptmmio->slot, - pptmmio->func, pptmmio->gpa, pptmmio->len, - pptmmio->hpa); + pptmmio->func, pptmmio->gpa, pptmmio->len, pptmmio->hpa); break; - case VM_UNMAP_PPTDEV_MMIO: + } + case VM_UNMAP_PPTDEV_MMIO: { + struct vm_pptdev_mmio *pptmmio; + pptmmio = (struct vm_pptdev_mmio *)data; error = ppt_unmap_mmio(vm, pptmmio->bus, pptmmio->slot, - pptmmio->func, pptmmio->gpa, pptmmio->len); + pptmmio->func, pptmmio->gpa, pptmmio->len); break; - case VM_BIND_PPTDEV: + } + case VM_BIND_PPTDEV: { + struct vm_pptdev *pptdev; + pptdev = (struct vm_pptdev *)data; error = vm_assign_pptdev(vm, pptdev->bus, pptdev->slot, - pptdev->func); + pptdev->func); break; - case VM_UNBIND_PPTDEV: + } + case VM_UNBIND_PPTDEV: { + struct vm_pptdev *pptdev; + pptdev = (struct vm_pptdev *)data; error = vm_unassign_pptdev(vm, pptdev->bus, pptdev->slot, - pptdev->func); + pptdev->func); break; - case VM_INJECT_EXCEPTION: + } + case VM_INJECT_EXCEPTION: { + struct vm_exception *vmexc; + vmexc = (struct vm_exception *)data; error = vm_inject_exception(vcpu, vmexc->vector, vmexc->error_code_valid, vmexc->error_code, vmexc->restart_instruction); break; + } case VM_INJECT_NMI: error = vm_inject_nmi(vcpu); break; - case VM_LAPIC_IRQ: + case VM_LAPIC_IRQ: { + struct vm_lapic_irq *vmirq; + vmirq = (struct vm_lapic_irq *)data; error = lapic_intr_edge(vcpu, vmirq->vector); break; - case VM_LAPIC_LOCAL_IRQ: + } + case VM_LAPIC_LOCAL_IRQ: { + struct vm_lapic_irq *vmirq; + vmirq = (struct vm_lapic_irq *)data; error = lapic_set_local_intr(vm, vcpu, vmirq->vector); break; - case VM_LAPIC_MSI: + } + case VM_LAPIC_MSI: { + struct vm_lapic_msi *vmmsi; + vmmsi = (struct vm_lapic_msi *)data; error = lapic_intr_msi(vm, vmmsi->addr, vmmsi->msg); break; - case VM_IOAPIC_ASSERT_IRQ: + } + case VM_IOAPIC_ASSERT_IRQ: { + struct vm_ioapic_irq *ioapic_irq; + ioapic_irq = (struct vm_ioapic_irq *)data; error = vioapic_assert_irq(vm, ioapic_irq->irq); break; - case VM_IOAPIC_DEASSERT_IRQ: + } + case VM_IOAPIC_DEASSERT_IRQ: { + struct vm_ioapic_irq *ioapic_irq; + ioapic_irq = (struct vm_ioapic_irq *)data; error = vioapic_deassert_irq(vm, ioapic_irq->irq); break; - case VM_IOAPIC_PULSE_IRQ: + } + case VM_IOAPIC_PULSE_IRQ: { + struct vm_ioapic_irq *ioapic_irq; + ioapic_irq = (struct vm_ioapic_irq *)data; error = vioapic_pulse_irq(vm, ioapic_irq->irq); break; + } case VM_IOAPIC_PINCOUNT: *(int *)data = vioapic_pincount(vm); break; case VM_SET_KERNEMU_DEV: case VM_GET_KERNEMU_DEV: { + struct vm_readwrite_kernemu_device *kernemu; mem_region_write_t mwrite; mem_region_read_t mread; int size; @@ -396,60 +412,86 @@ vmmdev_machdep_ioctl(struct vm *vm, struct vcpu *vcpu, u_long cmd, caddr_t data, error = mread(vcpu, kernemu->gpa, &kernemu->value, size, &arg); break; - } - case VM_ISA_ASSERT_IRQ: + } + case VM_ISA_ASSERT_IRQ: { + struct vm_isa_irq *isa_irq; + isa_irq = (struct vm_isa_irq *)data; error = vatpic_assert_irq(vm, isa_irq->atpic_irq); if (error == 0 && isa_irq->ioapic_irq != -1) error = vioapic_assert_irq(vm, isa_irq->ioapic_irq); break; - case VM_ISA_DEASSERT_IRQ: + } + case VM_ISA_DEASSERT_IRQ: { + struct vm_isa_irq *isa_irq; + isa_irq = (struct vm_isa_irq *)data; error = vatpic_deassert_irq(vm, isa_irq->atpic_irq); if (error == 0 && isa_irq->ioapic_irq != -1) error = vioapic_deassert_irq(vm, isa_irq->ioapic_irq); break; - case VM_ISA_PULSE_IRQ: + } + case VM_ISA_PULSE_IRQ: { + struct vm_isa_irq *isa_irq; + isa_irq = (struct vm_isa_irq *)data; error = vatpic_pulse_irq(vm, isa_irq->atpic_irq); if (error == 0 && isa_irq->ioapic_irq != -1) error = vioapic_pulse_irq(vm, isa_irq->ioapic_irq); break; - case VM_ISA_SET_IRQ_TRIGGER: + } + case VM_ISA_SET_IRQ_TRIGGER: { + struct vm_isa_irq_trigger *isa_irq_trigger; + isa_irq_trigger = (struct vm_isa_irq_trigger *)data; error = vatpic_set_irq_trigger(vm, isa_irq_trigger->atpic_irq, isa_irq_trigger->trigger); break; - case VM_SET_SEGMENT_DESCRIPTOR: + } + case VM_SET_SEGMENT_DESCRIPTOR: { + struct vm_seg_desc *vmsegdesc; + vmsegdesc = (struct vm_seg_desc *)data; - error = vm_set_seg_desc(vcpu, - vmsegdesc->regnum, - &vmsegdesc->desc); + error = vm_set_seg_desc(vcpu, vmsegdesc->regnum, + &vmsegdesc->desc); break; - case VM_GET_SEGMENT_DESCRIPTOR: + } + case VM_GET_SEGMENT_DESCRIPTOR: { + struct vm_seg_desc *vmsegdesc; + vmsegdesc = (struct vm_seg_desc *)data; - error = vm_get_seg_desc(vcpu, - vmsegdesc->regnum, - &vmsegdesc->desc); + error = vm_get_seg_desc(vcpu, vmsegdesc->regnum, + &vmsegdesc->desc); break; - case VM_SET_X2APIC_STATE: + } + case VM_SET_X2APIC_STATE: { + struct vm_x2apic *x2apic; + x2apic = (struct vm_x2apic *)data; error = vm_set_x2apic_state(vcpu, x2apic->state); break; - case VM_GET_X2APIC_STATE: + } + case VM_GET_X2APIC_STATE: { + struct vm_x2apic *x2apic; + x2apic = (struct vm_x2apic *)data; error = vm_get_x2apic_state(vcpu, &x2apic->state); break; - case VM_GET_GPA_PMAP: + } + case VM_GET_GPA_PMAP: { + struct vm_gpa_pte *gpapte; + gpapte = (struct vm_gpa_pte *)data; - pmap_get_mapping(vmspace_pmap(vm_vmspace(vm)), - gpapte->gpa, gpapte->pte, &gpapte->ptenum); - error = 0; + pmap_get_mapping(vmspace_pmap(vm_vmspace(vm)), gpapte->gpa, + gpapte->pte, &gpapte->ptenum); break; + } case VM_GET_HPET_CAPABILITIES: error = vhpet_getcap((struct vm_hpet_cap *)data); break; case VM_GLA2GPA: { + struct vm_gla2gpa *gg; + CTASSERT(PROT_READ == VM_PROT_READ); CTASSERT(PROT_WRITE == VM_PROT_WRITE); CTASSERT(PROT_EXEC == VM_PROT_EXECUTE); @@ -460,50 +502,76 @@ vmmdev_machdep_ioctl(struct vm *vm, struct vcpu *vcpu, u_long cmd, caddr_t data, ("%s: vm_gla2gpa unknown error %d", __func__, error)); break; } - case VM_GLA2GPA_NOFAULT: + case VM_GLA2GPA_NOFAULT: { + struct vm_gla2gpa *gg; + gg = (struct vm_gla2gpa *)data; error = vm_gla2gpa_nofault(vcpu, &gg->paging, gg->gla, gg->prot, &gg->gpa, &gg->fault); KASSERT(error == 0 || error == EFAULT, ("%s: vm_gla2gpa unknown error %d", __func__, error)); break; - case VM_SET_INTINFO: + } + case VM_SET_INTINFO: { + struct vm_intinfo *vmii; + vmii = (struct vm_intinfo *)data; error = vm_exit_intinfo(vcpu, vmii->info1); break; - case VM_GET_INTINFO: + } + case VM_GET_INTINFO: { + struct vm_intinfo *vmii; + vmii = (struct vm_intinfo *)data; error = vm_get_intinfo(vcpu, &vmii->info1, &vmii->info2); break; - case VM_RTC_WRITE: + } + case VM_RTC_WRITE: { + struct vm_rtc_data *rtcdata; + rtcdata = (struct vm_rtc_data *)data; error = vrtc_nvram_write(vm, rtcdata->offset, rtcdata->value); break; - case VM_RTC_READ: + } + case VM_RTC_READ: { + struct vm_rtc_data *rtcdata; + rtcdata = (struct vm_rtc_data *)data; error = vrtc_nvram_read(vm, rtcdata->offset, &rtcdata->value); break; - case VM_RTC_SETTIME: + } + case VM_RTC_SETTIME: { + struct vm_rtc_time *rtctime; + rtctime = (struct vm_rtc_time *)data; error = vrtc_set_time(vm, rtctime->secs); break; - case VM_RTC_GETTIME: - error = 0; + } + case VM_RTC_GETTIME: { + struct vm_rtc_time *rtctime; + rtctime = (struct vm_rtc_time *)data; rtctime->secs = vrtc_get_time(vm); break; + } case VM_RESTART_INSTRUCTION: error = vm_restart_instruction(vcpu); break; #ifdef BHYVE_SNAPSHOT - case VM_SNAPSHOT_REQ: + case VM_SNAPSHOT_REQ: { + struct vm_snapshot_meta *snapshot_meta; + snapshot_meta = (struct vm_snapshot_meta *)data; error = vm_snapshot_req(vm, snapshot_meta); break; + } #ifdef COMPAT_FREEBSD13 - case VM_SNAPSHOT_REQ_13: + case VM_SNAPSHOT_REQ_13: { + struct vm_snapshot_meta *snapshot_meta; + struct vm_snapshot_meta_13 *snapshot_13; + /* * The old structure just has an additional pointer at * the start that is ignored. @@ -513,6 +581,7 @@ vmmdev_machdep_ioctl(struct vm *vm, struct vcpu *vcpu, u_long cmd, caddr_t data, (struct vm_snapshot_meta *)&snapshot_13->dev_data; error = vm_snapshot_req(vm, snapshot_meta); break; + } #endif case VM_RESTORE_TIME: error = vm_restore_time(vm); diff --git a/sys/arm/conf/GENERIC b/sys/arm/conf/GENERIC index 7394f3842d43..22bb75993834 100644 --- a/sys/arm/conf/GENERIC +++ b/sys/arm/conf/GENERIC @@ -270,7 +270,3 @@ makeoptions MODULES_EXTRA+="dtb/nvidia" makeoptions MODULES_EXTRA+="dtb/rockchip" makeoptions MODULES_EXTRA+="dtb/rpi" makeoptions MODULES_EXTRA+="dtb/zynq" - -# SOC-specific modules -makeoptions MODULES_EXTRA+="allwinner" -makeoptions MODULES_EXTRA+="imx" diff --git a/sys/arm/conf/NOTES b/sys/arm/conf/NOTES index 920d721dc3ba..2bd41d911124 100644 --- a/sys/arm/conf/NOTES +++ b/sys/arm/conf/NOTES @@ -92,11 +92,6 @@ nodevice mps nodevice bnxt -# Build SOC-specific modules... - -makeoptions MODULES_EXTRA+="allwinner" -makeoptions MODULES_EXTRA+="imx" - # Build dtb files... makeoptions MODULES_EXTRA+="dtb/allwinner" diff --git a/sys/arm/include/ieeefp.h b/sys/arm/include/ieeefp.h index 57dd058b8a95..57719b883d58 100644 --- a/sys/arm/include/ieeefp.h +++ b/sys/arm/include/ieeefp.h @@ -49,4 +49,14 @@ typedef enum { #define fp_except_t int +/* Augment the userland declarations. */ +__BEGIN_DECLS +extern fp_rnd_t fpgetround(void); +extern fp_rnd_t fpsetround(fp_rnd_t); +extern fp_except_t fpgetmask(void); +extern fp_except_t fpsetmask(fp_except_t); +extern fp_except_t fpgetsticky(void); +extern fp_except_t fpsetsticky(fp_except_t); +__END_DECLS + #endif /* _MACHINE_IEEEFP_H_ */ diff --git a/sys/arm/ti/clk/ti_clkctrl.c b/sys/arm/ti/clk/ti_clkctrl.c index 72fa8548d4f8..06e558d140f2 100644 --- a/sys/arm/ti/clk/ti_clkctrl.c +++ b/sys/arm/ti/clk/ti_clkctrl.c @@ -284,9 +284,9 @@ create_clkctrl(struct ti_clkctrl_softc *sc, cell_t *reg, uint32_t index, uint32_ /* * Check out XX_CLKCTRL-INDEX(offset)-macro dance in - * sys/gnu/dts/dts/include/dt-bindings/clock/am3.h - * sys/gnu/dts/dts/include/dt-bindings/clock/am4.h - * sys/gnu/dts/dts/include/dt-bindings/clock/dra7.h + * sys/contrib/device-tree/include/dt-bindings/clock/am3.h + * sys/contrib/device-tree/include/dt-bindings/clock/am4.h + * sys/contrib/device-tree/include/dt-bindings/clock/dra7.h * reg[0] are in practice the same as the offset described in the dts. */ /* special_gdbclk_reg are 0 or 1 */ diff --git a/sys/arm64/arm64/elf32_machdep.c b/sys/arm64/arm64/elf32_machdep.c index 8f8a934ad520..4cb8ee5f57ef 100644 --- a/sys/arm64/arm64/elf32_machdep.c +++ b/sys/arm64/arm64/elf32_machdep.c @@ -210,7 +210,7 @@ freebsd32_fetch_syscall_args(struct thread *td) sa->code = *ap++; nap--; } else if (sa->code == SYS___syscall) { - sa->code = ap[1]; + sa->code = ap[_QUAD_LOWWORD]; nap -= 2; ap += 2; } diff --git a/sys/arm64/vmm/vmm.c b/sys/arm64/vmm/vmm.c index bf52dc0fe916..14ea26c3668c 100644 --- a/sys/arm64/vmm/vmm.c +++ b/sys/arm64/vmm/vmm.c @@ -1279,8 +1279,7 @@ vcpu_get_state(struct vcpu *vcpu, int *hostcpu) int vm_get_register(struct vcpu *vcpu, int reg, uint64_t *retval) { - - if (reg >= VM_REG_LAST) + if (reg < 0 || reg >= VM_REG_LAST) return (EINVAL); return (vmmops_getreg(vcpu->cookie, reg, retval)); @@ -1291,7 +1290,7 @@ vm_set_register(struct vcpu *vcpu, int reg, uint64_t val) { int error; - if (reg >= VM_REG_LAST) + if (reg < 0 || reg >= VM_REG_LAST) return (EINVAL); error = vmmops_setreg(vcpu->cookie, reg, val); if (error || reg != VM_REG_GUEST_PC) diff --git a/sys/arm64/vmm/vmm_dev_machdep.c b/sys/arm64/vmm/vmm_dev_machdep.c index 926a74fa528b..29d14e1ba952 100644 --- a/sys/arm64/vmm/vmm_dev_machdep.c +++ b/sys/arm64/vmm/vmm_dev_machdep.c @@ -68,19 +68,13 @@ int vmmdev_machdep_ioctl(struct vm *vm, struct vcpu *vcpu, u_long cmd, caddr_t data, int fflag, struct thread *td) { - struct vm_run *vmrun; - struct vm_vgic_version *vgv; - struct vm_vgic_descr *vgic; - struct vm_irq *vi; - struct vm_exception *vmexc; - struct vm_gla2gpa *gg; - struct vm_msi *vmsi; int error; error = 0; switch (cmd) { case VM_RUN: { struct vm_exit *vme; + struct vm_run *vmrun; vmrun = (struct vm_run *)data; vme = vm_exitinfo(vcpu); @@ -94,41 +88,62 @@ vmmdev_machdep_ioctl(struct vm *vm, struct vcpu *vcpu, u_long cmd, caddr_t data, break; break; } - case VM_INJECT_EXCEPTION: + case VM_INJECT_EXCEPTION: { + struct vm_exception *vmexc; + vmexc = (struct vm_exception *)data; error = vm_inject_exception(vcpu, vmexc->esr, vmexc->far); break; - case VM_GLA2GPA_NOFAULT: + } + case VM_GLA2GPA_NOFAULT: { + struct vm_gla2gpa *gg; + gg = (struct vm_gla2gpa *)data; error = vm_gla2gpa_nofault(vcpu, &gg->paging, gg->gla, gg->prot, &gg->gpa, &gg->fault); KASSERT(error == 0 || error == EFAULT, ("%s: vm_gla2gpa unknown error %d", __func__, error)); break; - case VM_GET_VGIC_VERSION: + } + case VM_GET_VGIC_VERSION: { + struct vm_vgic_version *vgv; + vgv = (struct vm_vgic_version *)data; /* TODO: Query the vgic driver for this */ vgv->version = 3; vgv->flags = 0; error = 0; break; - case VM_ATTACH_VGIC: + } + case VM_ATTACH_VGIC: { + struct vm_vgic_descr *vgic; + vgic = (struct vm_vgic_descr *)data; error = vm_attach_vgic(vm, vgic); break; - case VM_RAISE_MSI: + } + case VM_RAISE_MSI: { + struct vm_msi *vmsi; + vmsi = (struct vm_msi *)data; error = vm_raise_msi(vm, vmsi->msg, vmsi->addr, vmsi->bus, vmsi->slot, vmsi->func); break; - case VM_ASSERT_IRQ: + } + case VM_ASSERT_IRQ: { + struct vm_irq *vi; + vi = (struct vm_irq *)data; error = vm_assert_irq(vm, vi->irq); break; - case VM_DEASSERT_IRQ: + } + case VM_DEASSERT_IRQ: { + struct vm_irq *vi; + vi = (struct vm_irq *)data; error = vm_deassert_irq(vm, vi->irq); break; + } default: error = ENOTTY; break; diff --git a/sys/compat/linux/linux_ioctl.h b/sys/compat/linux/linux_ioctl.h index ccc25bc919ab..8345b7e4b719 100644 --- a/sys/compat/linux/linux_ioctl.h +++ b/sys/compat/linux/linux_ioctl.h @@ -823,4 +823,16 @@ int linux32_ioctl_register_handler(struct linux_ioctl_handler *h); int linux32_ioctl_unregister_handler(struct linux_ioctl_handler *h); #endif +#define LINUX_IOCTL_SET(n, low, high) \ +static linux_ioctl_function_t n##_linux_ioctl; \ +static struct linux_ioctl_handler n##_linux_handler = { \ + n##_linux_ioctl, \ + low, \ + high \ +}; \ +SYSINIT(n##_ioctl_register, SI_SUB_KLD, SI_ORDER_MIDDLE, \ + linux_ioctl_register_handler, &n##_linux_handler); \ +SYSUNINIT(n##_ioctl_unregister, SI_SUB_KLD, SI_ORDER_MIDDLE, \ + linux_ioctl_unregister_handler, &n##_linux_handler) + #endif /* !_LINUX_IOCTL_H_ */ diff --git a/sys/compat/linuxkpi/common/include/linux/bitops.h b/sys/compat/linuxkpi/common/include/linux/bitops.h index 00dd1f9a1ec0..a5a7abd55287 100644 --- a/sys/compat/linuxkpi/common/include/linux/bitops.h +++ b/sys/compat/linuxkpi/common/include/linux/bitops.h @@ -37,13 +37,8 @@ #define BIT(nr) (1UL << (nr)) #define BIT_ULL(nr) (1ULL << (nr)) -#ifdef __LP64__ -#define BITS_PER_LONG 64 -#else -#define BITS_PER_LONG 32 -#endif - -#define BITS_PER_LONG_LONG 64 +#define BITS_PER_LONG (__SIZEOF_LONG__ * __CHAR_BIT__) +#define BITS_PER_LONG_LONG (__SIZEOF_LONG_LONG__ * __CHAR_BIT__) #define BITMAP_FIRST_WORD_MASK(start) (~0UL << ((start) % BITS_PER_LONG)) #define BITMAP_LAST_WORD_MASK(n) (~0UL >> (BITS_PER_LONG - (n))) diff --git a/sys/compat/linuxkpi/common/include/linux/compiler.h b/sys/compat/linuxkpi/common/include/linux/compiler.h index 948396144ad6..4146c829b936 100644 --- a/sys/compat/linuxkpi/common/include/linux/compiler.h +++ b/sys/compat/linuxkpi/common/include/linux/compiler.h @@ -31,6 +31,7 @@ #define _LINUXKPI_LINUX_COMPILER_H_ #include <sys/cdefs.h> +#include <sys/endian.h> #define __user #define __kernel @@ -79,6 +80,13 @@ #else #define __counted_by(_x) #endif +#if BYTE_ORDER == LITTLE_ENDIAN +#define __counted_by_le(_x) __counted_by(_x) +#define __counted_by_be(_x) +#else +#define __counted_by_le(_x) +#define __counted_by_be(_x) __counted_by(_x) +#endif #define likely(x) __builtin_expect(!!(x), 1) #define unlikely(x) __builtin_expect(!!(x), 0) diff --git a/sys/compat/linuxkpi/common/include/linux/device.h b/sys/compat/linuxkpi/common/include/linux/device.h index 7dd6340746d2..c291133e2e0b 100644 --- a/sys/compat/linuxkpi/common/include/linux/device.h +++ b/sys/compat/linuxkpi/common/include/linux/device.h @@ -92,6 +92,7 @@ struct device_driver { const struct dev_pm_ops *pm; void (*shutdown) (struct device *); + void (*coredump) (struct device *); }; struct device_type { diff --git a/sys/compat/linuxkpi/common/include/linux/etherdevice.h b/sys/compat/linuxkpi/common/include/linux/etherdevice.h index 1f2d6cf22d7e..b9a4951de8ac 100644 --- a/sys/compat/linuxkpi/common/include/linux/etherdevice.h +++ b/sys/compat/linuxkpi/common/include/linux/etherdevice.h @@ -27,6 +27,8 @@ #include <linux/types.h> #include <linux/device.h> +#include <linux/skbuff.h> +#include <linux/netdevice.h> #include <sys/random.h> #include <sys/libkern.h> @@ -137,4 +139,25 @@ device_get_mac_address(struct device *dev, char *dst) return (-ENOENT); } +/* Returns network byte order. */ +static inline uint16_t +eth_type_trans(struct sk_buff *skb, struct net_device *dev) +{ + pr_debug("%s: TODO\n", __func__); + return (htons(ETHERTYPE_8023)); +} + +static inline void +eth_hw_addr_set(struct net_device *dev, const u8 *addr) +{ + pr_debug("%s: TODO (if we want to)\n", __func__); +} + +static inline int +eth_platform_get_mac_address(struct device *dev __unused, u8 *addr __unused) +{ + pr_debug("%s: TODO\n", __func__); + return (-ENODEV); +} + #endif /* _LINUXKPI_LINUX_ETHERDEVICE_H_ */ diff --git a/sys/compat/linuxkpi/common/include/linux/fips.h b/sys/compat/linuxkpi/common/include/linux/fips.h new file mode 100644 index 000000000000..25c0c1fc1fa0 --- /dev/null +++ b/sys/compat/linuxkpi/common/include/linux/fips.h @@ -0,0 +1,12 @@ +/* + * Copyright (c) 2025 Bjoern A. Zeeb + * + * SPDX-License-Identifier: BSD-2-Clause + */ + +#ifndef _LINUXKPI_LINUX_FIPS_H +#define _LINUXKPI_LINUX_FIPS_H + +#define fips_enabled 0 + +#endif /* _LINUXKPI_LINUX_FIPS_H */ diff --git a/sys/compat/linuxkpi/common/include/linux/ieee80211.h b/sys/compat/linuxkpi/common/include/linux/ieee80211.h index 17041bb03ce8..ea8c0fc8ef5e 100644 --- a/sys/compat/linuxkpi/common/include/linux/ieee80211.h +++ b/sys/compat/linuxkpi/common/include/linux/ieee80211.h @@ -312,6 +312,7 @@ enum ieee80211_ac_numbers { #define IEEE80211_MLD_CAP_OP_MAX_SIMUL_LINKS 0xf #define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_SUPP 0x0060 #define IEEE80211_MLD_CAP_OP_TID_TO_LINK_MAP_NEG_SUPP_SAME 1 +#define IEEE80211_MLD_CAP_OP_LINK_RECONF_SUPPORT 0x2000 struct ieee80211_mcs_info { uint8_t rx_mask[IEEE80211_HT_MCS_MASK_LEN]; @@ -365,6 +366,7 @@ enum ieee80211_chanctx_change_flags { IEEE80211_CHANCTX_CHANGE_CHANNEL = BIT(4), IEEE80211_CHANCTX_CHANGE_PUNCTURING = BIT(5), IEEE80211_CHANCTX_CHANGE_MIN_DEF = BIT(6), + IEEE80211_CHANCTX_CHANGE_AP = BIT(7), }; enum ieee80211_frame_release_type { diff --git a/sys/compat/linuxkpi/common/include/linux/netdevice.h b/sys/compat/linuxkpi/common/include/linux/netdevice.h index 3b808a4a1749..cf27753bcb80 100644 --- a/sys/compat/linuxkpi/common/include/linux/netdevice.h +++ b/sys/compat/linuxkpi/common/include/linux/netdevice.h @@ -486,6 +486,21 @@ netdev_priv(const struct net_device *ndev) } /* -------------------------------------------------------------------------- */ + +static __inline void +netif_device_attach(struct net_device *ndev) +{ + pr_debug("%s: TODO\n", __func__); +} + +static __inline void +netif_device_detach(struct net_device *ndev) +{ + pr_debug("%s: TODO\n", __func__); +} + + +/* -------------------------------------------------------------------------- */ /* This is really rtnetlink and probably belongs elsewhere. */ #define rtnl_lock() do { } while(0) diff --git a/sys/compat/linuxkpi/common/include/linux/nl80211.h b/sys/compat/linuxkpi/common/include/linux/nl80211.h index f3979d3a2abc..845ffec4bcba 100644 --- a/sys/compat/linuxkpi/common/include/linux/nl80211.h +++ b/sys/compat/linuxkpi/common/include/linux/nl80211.h @@ -50,6 +50,7 @@ enum nl80211_feature_flags { NL80211_FEATURE_WFA_TPC_IE_IN_PROBES = BIT(15), NL80211_FEATURE_AP_SCAN = BIT(16), NL80211_FEATURE_ACTIVE_MONITOR = BIT(17), + NL80211_FEATURE_SAE = BIT(18), }; enum nl80211_pmsr_ftm_failure_flags { @@ -85,6 +86,7 @@ enum nl80211_reg_rule_flags { NL80211_RRF_NO_6GHZ_AFC_CLIENT = BIT(15), NL80211_RRF_PSD = BIT(16), NL80211_RRF_ALLOW_6GHZ_VLP_AP = BIT(17), + NL80211_RRF_ALLOW_20MHZ_ACTIVITY = BIT(18), }; #define NL80211_RRF_NO_HT40 (NL80211_RRF_NO_HT40MINUS|NL80211_RRF_NO_HT40PLUS) @@ -434,6 +436,14 @@ enum nl80211_hidden_ssid { NL80211_HIDDEN_SSID_NOT_IN_USE, }; +enum nl80211_external_auth_action { + NL80211_EXTERNAL_AUTH_START, +}; + +enum nl80211_rxmgmt_flags { + NL80211_RXMGMT_FLAG_EXTERNAL_AUTH = BIT(1), +}; + #define NL80211_TKIP_DATA_OFFSET_TX_MIC_KEY 16 #define NL80211_TKIP_DATA_OFFSET_RX_MIC_KEY 24 diff --git a/sys/compat/linuxkpi/common/include/linux/pci.h b/sys/compat/linuxkpi/common/include/linux/pci.h index ffc2be600c22..06336bf963d6 100644 --- a/sys/compat/linuxkpi/common/include/linux/pci.h +++ b/sys/compat/linuxkpi/common/include/linux/pci.h @@ -832,6 +832,19 @@ lkpi_pci_restore_state(struct pci_dev *pdev) #define pci_restore_state(dev) lkpi_pci_restore_state(dev) static inline int +linuxkpi_pci_enable_wake(struct pci_dev *pdev, pci_power_t state, bool ena) +{ + /* + * We do not currently support this in device.h either to + * check if the device is allowed to wake up in first place. + */ + pr_debug("%s: TODO\n", __func__); + return (0); +} +#define pci_enable_wake(dev, state, ena) \ + linuxkpi_pci_enable_wake(dev, state, ena) + +static inline int pci_reset_function(struct pci_dev *pdev) { diff --git a/sys/compat/linuxkpi/common/include/linux/platform_device.h b/sys/compat/linuxkpi/common/include/linux/platform_device.h index 6853e709cb70..dba79f5936cc 100644 --- a/sys/compat/linuxkpi/common/include/linux/platform_device.h +++ b/sys/compat/linuxkpi/common/include/linux/platform_device.h @@ -39,7 +39,7 @@ struct platform_device { }; struct platform_driver { - int (*remove)(struct platform_device *); + void (*remove)(struct platform_device *); struct device_driver driver; }; diff --git a/sys/compat/linuxkpi/common/include/linux/skbuff.h b/sys/compat/linuxkpi/common/include/linux/skbuff.h index 6e41c368a8b8..2e560a120e41 100644 --- a/sys/compat/linuxkpi/common/include/linux/skbuff.h +++ b/sys/compat/linuxkpi/common/include/linux/skbuff.h @@ -1159,6 +1159,9 @@ skb_cow_head(struct sk_buff *skb, unsigned int headroom) return (-1); } +/* Misplaced here really but sock comes from skbuff. */ +#define sk_pacing_shift_update(sock, n) + #define SKB_WITH_OVERHEAD(_s) \ (_s) - ALIGN(sizeof(struct skb_shared_info), CACHE_LINE_SIZE) diff --git a/sys/compat/linuxkpi/common/include/linux/soc/mediatek/mtk_wed.h b/sys/compat/linuxkpi/common/include/linux/soc/mediatek/mtk_wed.h index 903053e7f6e8..9f3a1ee4c139 100644 --- a/sys/compat/linuxkpi/common/include/linux/soc/mediatek/mtk_wed.h +++ b/sys/compat/linuxkpi/common/include/linux/soc/mediatek/mtk_wed.h @@ -1,54 +1,36 @@ /*- - * SPDX-License-Identifier: BSD-2-Clause - * - * Copyright (c) 2022-2023 Bjoern A. Zeeb - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. + * Copyright (c) 2022-2025 Bjoern A. Zeeb * - * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND - * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE - * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL - * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS - * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) - * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT - * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY - * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF - * SUCH DAMAGE. + * SPDX-License-Identifier: BSD-2-Clause */ #ifndef _LINUXKPI_LINUX_SOC_MEDIATEK_MTK_WED_H #define _LINUXKPI_LINUX_SOC_MEDIATEK_MTK_WED_H +#include <linux/kernel.h> /* pr_debug */ + struct mtk_wed_device { }; #define WED_WO_STA_REC 0x6 -#define mtk_wed_device_start(_dev, _mask) do { } while(0) -#define mtk_wed_device_detach(_dev) do { } while(0) +#define mtk_wed_device_start(_dev, _mask) do { pr_debug("%s: TODO\n", __func__); } while(0) +#define mtk_wed_device_detach(_dev) do { pr_debug("%s: TODO\n", __func__); } while(0) #define mtk_wed_device_irq_get(_dev, _mask) 0 -#define mtk_wed_device_irq_set_mask(_dev, _mask) do { } while(0) -#define mtk_wed_device_update_msg(_dev, _id, _msg, _len) (-ENODEV) -#define mtk_wed_device_dma_reset(_dev) do {} while (0) +#define mtk_wed_device_irq_set_mask(_dev, _mask) do { pr_debug("%s: TODO\n", __func__); } while(0) +#define mtk_wed_device_update_msg(_dev, _id, _msg, _len) ({ pr_debug("%s: TODO\n", __func__); -ENODEV; }) +#define mtk_wed_device_dma_reset(_dev) do { pr_debug("%s: TODO\n", __func__); } while (0) #define mtk_wed_device_ppe_check(_dev, _skb, _reason, _entry) \ - do {} while (0) -#define mtk_wed_device_stop(_dev) do { } while(0) -#define mtk_wed_device_start_hw_rro(_dev, _mask, _b) do { } while(0) -#define mtk_wed_device_setup_tc(_dev, _ndev, _type, _tdata) (-EOPNOTSUPP) + do { pr_debug("%s: TODO\n", __func__); } while (0) +#define mtk_wed_device_stop(_dev) do { pr_debug("%s: TODO\n", __func__); } while(0) +#define mtk_wed_device_start_hw_rro(_dev, _mask, _b) do { pr_debug("%s: TODO\n", __func__); } while(0) +#define mtk_wed_device_setup_tc(_dev, _ndev, _type, _tdata) ({ pr_debug("%s: TODO\n", __func__); -EOPNOTSUPP; }) static inline bool mtk_wed_device_active(struct mtk_wed_device *dev __unused) { + pr_debug("%s: TODO\n", __func__); return (false); } @@ -56,6 +38,7 @@ static inline bool mtk_wed_get_rx_capa(struct mtk_wed_device *dev __unused) { + pr_debug("%s: TODO\n", __func__); return (false); } diff --git a/sys/compat/linuxkpi/common/include/net/cfg80211.h b/sys/compat/linuxkpi/common/include/net/cfg80211.h index 239b4a5ae7b8..f769cfdd4075 100644 --- a/sys/compat/linuxkpi/common/include/net/cfg80211.h +++ b/sys/compat/linuxkpi/common/include/net/cfg80211.h @@ -56,7 +56,7 @@ extern int linuxkpi_debug_80211; #define D80211_IMPROVE 0x2 #endif #define TODO(fmt, ...) if (linuxkpi_debug_80211 & D80211_TODO) \ - printf("%s:%d: XXX LKPI80211 TODO " fmt "\n", __func__, __LINE__, ##__VA_ARGS__) + printf("%s:%d: XXX LKPI80211 TODO " fmt "\n", __func__, __LINE__, ##__VA_ARGS__) #define IMPROVE(fmt, ...) if (linuxkpi_debug_80211 & D80211_IMPROVE) \ printf("%s:%d: XXX LKPI80211 IMPROVE " fmt "\n", __func__, __LINE__, ##__VA_ARGS__) @@ -260,6 +260,19 @@ enum ieee80211_vht_opmode { IEEE80211_OPMODE_NOTIF_RX_NSS_SHIFT = 4, }; +struct cfg80211_bss_ies { + uint8_t *data; + size_t len; +}; + +struct cfg80211_bss { + /* XXX TODO */ + struct cfg80211_bss_ies *ies; + struct cfg80211_bss_ies *beacon_ies; + uint64_t ts_boottime; + int32_t signal; +}; + struct cfg80211_connect_resp_params { /* XXX TODO */ uint8_t *bssid; @@ -267,7 +280,13 @@ struct cfg80211_connect_resp_params { const uint8_t *resp_ie; uint32_t req_ie_len; uint32_t resp_ie_len; - int status; + int status; + struct { + const uint8_t *addr; + const uint8_t *bssid; + struct cfg80211_bss *bss; + uint16_t status; + } links[IEEE80211_MLD_MAX_NUM_LINKS]; }; struct cfg80211_inform_bss { @@ -284,19 +303,12 @@ struct cfg80211_roam_info { uint32_t req_ie_len; uint32_t resp_ie_len; struct linuxkpi_ieee80211_channel *channel; -}; - -struct cfg80211_bss_ies { - uint8_t *data; - size_t len; -}; - -struct cfg80211_bss { - /* XXX TODO */ - struct cfg80211_bss_ies *ies; - struct cfg80211_bss_ies *beacon_ies; - - int32_t signal; + struct { + const uint8_t *addr; + const uint8_t *bssid; + struct cfg80211_bss *bss; + struct linuxkpi_ieee80211_channel *channel; + } links[IEEE80211_MLD_MAX_NUM_LINKS]; }; struct cfg80211_chan_def { @@ -404,6 +416,7 @@ struct cfg80211_scan_request { bool no_cck; bool scan_6ghz; bool duration_mandatory; + bool first_part; int8_t tsf_report_link_id; uint16_t duration; uint32_t flags; @@ -463,6 +476,24 @@ struct cfg80211_beacon_data { uint32_t assocresp_ies_len; }; +struct cfg80211_ap_update { + /* XXX TODO */ + struct cfg80211_beacon_data beacon; +}; + +struct cfg80211_crypto_settings { + /* XXX TODO */ + enum nl80211_wpa_versions wpa_versions; + uint32_t cipher_group; /* WLAN_CIPHER_SUITE_* */ + uint32_t *akm_suites; + uint32_t *ciphers_pairwise; + const uint8_t *sae_pwd; + const uint8_t *psk; + int n_akm_suites; + int n_ciphers_pairwise; + int sae_pwd_len; +}; + struct cfg80211_ap_settings { /* XXX TODO */ int auth_type, beacon_interval, dtim_period, hidden_ssid, inactivity_timeout; @@ -470,6 +501,7 @@ struct cfg80211_ap_settings { size_t ssid_len; struct cfg80211_beacon_data beacon; struct cfg80211_chan_def chandef; + struct cfg80211_crypto_settings crypto; }; struct cfg80211_bss_selection { @@ -484,23 +516,12 @@ struct cfg80211_bss_selection { } param; }; -struct cfg80211_crypto { /* XXX made up name */ - /* XXX TODO */ - enum nl80211_wpa_versions wpa_versions; - uint32_t cipher_group; /* WLAN_CIPHER_SUITE_* */ - uint32_t *akm_suites; - uint32_t *ciphers_pairwise; - const uint8_t *sae_pwd; - const uint8_t *psk; - int n_akm_suites; - int n_ciphers_pairwise; - int sae_pwd_len; -}; - struct cfg80211_connect_params { /* XXX TODO */ struct linuxkpi_ieee80211_channel *channel; + struct linuxkpi_ieee80211_channel *channel_hint; uint8_t *bssid; + uint8_t *bssid_hint; const uint8_t *ie; const uint8_t *ssid; uint32_t ie_len; @@ -509,7 +530,7 @@ struct cfg80211_connect_params { uint32_t key_len; int auth_type, key_idx, privacy, want_1x; struct cfg80211_bss_selection bss_select; - struct cfg80211_crypto crypto; + struct cfg80211_crypto_settings crypto; }; enum bss_param_flags { /* Used as bitflags. XXX FIXME values? */ @@ -538,6 +559,14 @@ struct cfg80211_mgmt_tx_params { int wait; }; +struct cfg80211_external_auth_params { + uint8_t bssid[ETH_ALEN]; + uint16_t status; + enum nl80211_external_auth_action action; + unsigned int key_mgmt_suite; + struct cfg80211_ssid ssid; +}; + struct cfg80211_pmk_conf { /* XXX TODO */ const uint8_t *pmk; @@ -548,6 +577,8 @@ struct cfg80211_pmksa { /* XXX TODO */ const uint8_t *bssid; const uint8_t *pmkid; + const uint8_t *ssid; + size_t ssid_len; }; struct station_del_parameters { @@ -961,6 +992,27 @@ struct cfg80211_set_hw_timestamp { bool enable; }; +struct survey_info { /* net80211::struct ieee80211_channel_survey */ + /* TODO FIXME */ + uint32_t filled; +#define SURVEY_INFO_TIME 0x0001 +#define SURVEY_INFO_TIME_RX 0x0002 +#define SURVEY_INFO_TIME_SCAN 0x0004 +#define SURVEY_INFO_TIME_TX 0x0008 +#define SURVEY_INFO_TIME_BSS_RX 0x0010 +#define SURVEY_INFO_TIME_BUSY 0x0020 +#define SURVEY_INFO_IN_USE 0x0040 +#define SURVEY_INFO_NOISE_DBM 0x0080 + uint32_t noise; + uint64_t time; + uint64_t time_bss_rx; + uint64_t time_busy; + uint64_t time_rx; + uint64_t time_scan; + uint64_t time_tx; + struct linuxkpi_ieee80211_channel *channel; +}; + enum wiphy_vendor_cmd_need_flags { WIPHY_VENDOR_CMD_NEED_NETDEV = 0x01, WIPHY_VENDOR_CMD_NEED_RUNNING = 0x02, @@ -1118,49 +1170,53 @@ struct wireless_dev { struct cfg80211_ops { /* XXX TODO */ struct wireless_dev *(*add_virtual_intf)(struct wiphy *, const char *, unsigned char, enum nl80211_iftype, struct vif_params *); - int (*del_virtual_intf)(struct wiphy *, struct wireless_dev *); - s32 (*change_virtual_intf)(struct wiphy *, struct net_device *, enum nl80211_iftype, struct vif_params *); - s32 (*scan)(struct wiphy *, struct cfg80211_scan_request *); - s32 (*set_wiphy_params)(struct wiphy *, u32); - s32 (*join_ibss)(struct wiphy *, struct net_device *, struct cfg80211_ibss_params *); - s32 (*leave_ibss)(struct wiphy *, struct net_device *); - s32 (*get_station)(struct wiphy *, struct net_device *, const u8 *, struct station_info *); - int (*dump_station)(struct wiphy *, struct net_device *, int, u8 *, struct station_info *); - s32 (*set_tx_power)(struct wiphy *, struct wireless_dev *, enum nl80211_tx_power_setting, s32); - s32 (*get_tx_power)(struct wiphy *, struct wireless_dev *, s32 *); - s32 (*add_key)(struct wiphy *, struct net_device *, u8, bool, const u8 *, struct key_params *); - s32 (*del_key)(struct wiphy *, struct net_device *, u8, bool, const u8 *); - s32 (*get_key)(struct wiphy *, struct net_device *, u8, bool, const u8 *, void *, void(*)(void *, struct key_params *)); - s32 (*set_default_key)(struct wiphy *, struct net_device *, u8, bool, bool); - s32 (*set_default_mgmt_key)(struct wiphy *, struct net_device *, u8); - s32 (*set_power_mgmt)(struct wiphy *, struct net_device *, bool, s32); - s32 (*connect)(struct wiphy *, struct net_device *, struct cfg80211_connect_params *); - s32 (*disconnect)(struct wiphy *, struct net_device *, u16); - s32 (*suspend)(struct wiphy *, struct cfg80211_wowlan *); - s32 (*resume)(struct wiphy *); - s32 (*set_pmksa)(struct wiphy *, struct net_device *, struct cfg80211_pmksa *); - s32 (*del_pmksa)(struct wiphy *, struct net_device *, struct cfg80211_pmksa *); - s32 (*flush_pmksa)(struct wiphy *, struct net_device *); - s32 (*start_ap)(struct wiphy *, struct net_device *, struct cfg80211_ap_settings *); - int (*stop_ap)(struct wiphy *, struct net_device *); - s32 (*change_beacon)(struct wiphy *, struct net_device *, struct cfg80211_beacon_data *); - int (*del_station)(struct wiphy *, struct net_device *, struct station_del_parameters *); - int (*change_station)(struct wiphy *, struct net_device *, const u8 *, struct station_parameters *); + int (*del_virtual_intf)(struct wiphy *, struct wireless_dev *); + int (*change_virtual_intf)(struct wiphy *, struct net_device *, enum nl80211_iftype, struct vif_params *); + int (*scan)(struct wiphy *, struct cfg80211_scan_request *); + int (*set_wiphy_params)(struct wiphy *, int, uint32_t); + int (*join_ibss)(struct wiphy *, struct net_device *, struct cfg80211_ibss_params *); + int (*leave_ibss)(struct wiphy *, struct net_device *); + int (*get_station)(struct wiphy *, struct net_device *, const uint8_t *, struct station_info *); + int (*dump_station)(struct wiphy *, struct net_device *, int, uint8_t *, struct station_info *); + int (*set_tx_power)(struct wiphy *, struct wireless_dev *, int, enum nl80211_tx_power_setting, int); + int (*get_tx_power)(struct wiphy *, struct wireless_dev *, int, unsigned int, int *); + int (*add_key)(struct wiphy *, struct net_device *, int, uint8_t, bool, const uint8_t *, struct key_params *); + int (*del_key)(struct wiphy *, struct net_device *, int, uint8_t, bool, const uint8_t *); + int (*get_key)(struct wiphy *, struct net_device *, int, uint8_t, bool, const uint8_t *, void *, void(*)(void *, struct key_params *)); + int (*set_default_key)(struct wiphy *, struct net_device *, int, uint8_t, bool, bool); + int (*set_default_mgmt_key)(struct wiphy *, struct net_device *, int, uint8_t); + int (*set_power_mgmt)(struct wiphy *, struct net_device *, bool, int); + int (*connect)(struct wiphy *, struct net_device *, struct cfg80211_connect_params *); + int (*disconnect)(struct wiphy *, struct net_device *, uint16_t); + int (*suspend)(struct wiphy *, struct cfg80211_wowlan *); + int (*resume)(struct wiphy *); + int (*set_pmksa)(struct wiphy *, struct net_device *, struct cfg80211_pmksa *); + int (*del_pmksa)(struct wiphy *, struct net_device *, struct cfg80211_pmksa *); + int (*flush_pmksa)(struct wiphy *, struct net_device *); + int (*start_ap)(struct wiphy *, struct net_device *, struct cfg80211_ap_settings *); + int (*stop_ap)(struct wiphy *, struct net_device *, unsigned int); + int (*change_beacon)(struct wiphy *, struct net_device *, struct cfg80211_ap_update *); + int (*del_station)(struct wiphy *, struct net_device *, struct station_del_parameters *); + int (*change_station)(struct wiphy *, struct net_device *, const uint8_t *, struct station_parameters *); int (*sched_scan_start)(struct wiphy *, struct net_device *, struct cfg80211_sched_scan_request *); - int (*sched_scan_stop)(struct wiphy *, struct net_device *, u64); + int (*sched_scan_stop)(struct wiphy *, struct net_device *, uint64_t); void (*update_mgmt_frame_registrations)(struct wiphy *, struct wireless_dev *, struct mgmt_frame_regs *); - int (*mgmt_tx)(struct wiphy *, struct wireless_dev *, struct cfg80211_mgmt_tx_params *, u64 *); - int (*cancel_remain_on_channel)(struct wiphy *, struct wireless_dev *, u64); - int (*get_channel)(struct wiphy *, struct wireless_dev *, struct cfg80211_chan_def *); - int (*crit_proto_start)(struct wiphy *, struct wireless_dev *, enum nl80211_crit_proto_id, u16); + int (*mgmt_tx)(struct wiphy *, struct wireless_dev *, struct cfg80211_mgmt_tx_params *, uint64_t *); + int (*cancel_remain_on_channel)(struct wiphy *, struct wireless_dev *, uint64_t); + int (*get_channel)(struct wiphy *, struct wireless_dev *, unsigned int, struct cfg80211_chan_def *); + int (*crit_proto_start)(struct wiphy *, struct wireless_dev *, enum nl80211_crit_proto_id, uint16_t); void (*crit_proto_stop)(struct wiphy *, struct wireless_dev *); - int (*tdls_oper)(struct wiphy *, struct net_device *, const u8 *, enum nl80211_tdls_operation); - int (*update_connect_params)(struct wiphy *, struct net_device *, struct cfg80211_connect_params *, u32); - int (*set_pmk)(struct wiphy *, struct net_device *, const struct cfg80211_pmk_conf *); - int (*del_pmk)(struct wiphy *, struct net_device *, const u8 *); - int (*remain_on_channel)(struct wiphy *, struct wireless_dev *, struct linuxkpi_ieee80211_channel *, unsigned int, u64 *); - int (*start_p2p_device)(struct wiphy *, struct wireless_dev *); - void (*stop_p2p_device)(struct wiphy *, struct wireless_dev *); + int (*tdls_oper)(struct wiphy *, struct net_device *, const uint8_t *, enum nl80211_tdls_operation); + int (*update_connect_params)(struct wiphy *, struct net_device *, struct cfg80211_connect_params *, uint32_t); + int (*set_pmk)(struct wiphy *, struct net_device *, const struct cfg80211_pmk_conf *); + int (*del_pmk)(struct wiphy *, struct net_device *, const uint8_t *); + int (*remain_on_channel)(struct wiphy *, struct wireless_dev *, struct linuxkpi_ieee80211_channel *, unsigned int, uint64_t *); + int (*start_p2p_device)(struct wiphy *, struct wireless_dev *); + void (*stop_p2p_device)(struct wiphy *, struct wireless_dev *); + int (*dump_survey)(struct wiphy *, struct net_device *, int, struct survey_info *); + int (*external_auth)(struct wiphy *, struct net_device *, struct cfg80211_external_auth_params *); + int (*set_cqm_rssi_range_config)(struct wiphy *, struct net_device *, int, int); + }; @@ -1179,6 +1235,8 @@ void linuxkpi_wiphy_delayed_work_queue(struct wiphy *, struct wiphy_delayed_work *, unsigned long); void linuxkpi_wiphy_delayed_work_cancel(struct wiphy *, struct wiphy_delayed_work *); +void linuxkpi_wiphy_delayed_work_flush(struct wiphy *, + struct wiphy_delayed_work *); int linuxkpi_regulatory_set_wiphy_regd_sync(struct wiphy *wiphy, struct linuxkpi_ieee80211_regdomain *regd); @@ -1247,6 +1305,21 @@ wiphy_rfkill_set_hw_state_reason(struct wiphy *wiphy, bool blocked, /* -------------------------------------------------------------------------- */ +static inline int +cfg80211_register_netdevice(struct net_device *ndev) +{ + TODO(); + return (-ENXIO); +} + +static inline void +cfg80211_unregister_netdevice(struct net_device *ndev) +{ + TODO(); +} + +/* -------------------------------------------------------------------------- */ + static inline struct cfg80211_bss * cfg80211_get_bss(struct wiphy *wiphy, struct linuxkpi_ieee80211_channel *chan, const uint8_t *bssid, const uint8_t *ssid, size_t ssid_len, @@ -1309,15 +1382,15 @@ reg_query_regdb_wmm(uint8_t *alpha2, uint32_t center_freq, return (-ENODATA); } -static __inline const u8 * -cfg80211_find_ie_match(uint32_t f, const u8 *ies, size_t ies_len, - const u8 *match, int x, int y) +static __inline const uint8_t * +cfg80211_find_ie_match(uint32_t f, const uint8_t *ies, size_t ies_len, + const uint8_t *match, int x, int y) { TODO(); return (NULL); } -static __inline const u8 * +static __inline const uint8_t * cfg80211_find_ie(uint8_t eid, const uint8_t *ie, uint32_t ielen) { TODO(); @@ -1339,6 +1412,36 @@ cfg80211_pmsr_report(struct wireless_dev *wdev, TODO(); } +static inline int +nl80211_chan_width_to_mhz(enum nl80211_chan_width width) +{ + switch (width) { + case NL80211_CHAN_WIDTH_5: + return (5); + break; + case NL80211_CHAN_WIDTH_10: + return (10); + break; + case NL80211_CHAN_WIDTH_20_NOHT: + case NL80211_CHAN_WIDTH_20: + return (20); + break; + case NL80211_CHAN_WIDTH_40: + return (40); + break; + case NL80211_CHAN_WIDTH_80: + case NL80211_CHAN_WIDTH_80P80: + return (80); + break; + case NL80211_CHAN_WIDTH_160: + return (160); + break; + case NL80211_CHAN_WIDTH_320: + return (320); + break; + } +} + static inline void cfg80211_chandef_create(struct cfg80211_chan_def *chandef, struct linuxkpi_ieee80211_channel *chan, enum nl80211_channel_type chan_type) @@ -1377,6 +1480,12 @@ cfg80211_chandef_valid(const struct cfg80211_chan_def *chandef) return (false); } +static inline int +cfg80211_chandef_get_width(const struct cfg80211_chan_def *chandef) +{ + return (nl80211_chan_width_to_mhz(chandef->width)); +} + static __inline bool cfg80211_chandef_dfs_usable(struct wiphy *wiphy, const struct cfg80211_chan_def *chandef) { @@ -1688,8 +1797,8 @@ cfg80211_disconnected(struct net_device *ndev, uint16_t reason, } static __inline int -cfg80211_get_p2p_attr(const u8 *ie, u32 ie_len, - enum ieee80211_p2p_attr_ids attr, u8 *p, size_t p_len) +cfg80211_get_p2p_attr(const uint8_t *ie, uint32_t ie_len, + enum ieee80211_p2p_attr_ids attr, uint8_t *p, size_t p_len) { TODO(); return (-1); @@ -1725,13 +1834,13 @@ cfg80211_inform_bss_data(struct wiphy *wiphy, static __inline void cfg80211_mgmt_tx_status(struct wireless_dev *wdev, uint64_t cookie, - const u8 *buf, size_t len, bool ack, gfp_t gfp) + const uint8_t *buf, size_t len, bool ack, gfp_t gfp) { TODO(); } static __inline void -cfg80211_michael_mic_failure(struct net_device *ndev, const uint8_t *addr, +cfg80211_michael_mic_failure(struct net_device *ndev, const uint8_t addr[ETH_ALEN], enum nl80211_key_type key_type, int _x, void *p, gfp_t gfp) { TODO(); @@ -1751,8 +1860,8 @@ cfg80211_del_sta(struct net_device *ndev, const uint8_t *addr, gfp_t gfp) } static __inline void -cfg80211_port_authorized(struct net_device *ndev, const uint8_t *bssid, - gfp_t gfp) +cfg80211_port_authorized(struct net_device *ndev, const uint8_t *addr, + const uint8_t *bitmap, uint8_t len, gfp_t gfp) { TODO(); } @@ -1935,7 +2044,7 @@ cfg80211_background_radar_event(struct wiphy *wiphy, TODO(); } -static __inline const u8 * +static __inline const uint8_t * cfg80211_find_ext_ie(uint8_t eid, const uint8_t *p, size_t len) { TODO(); @@ -2033,6 +2142,14 @@ cfg80211_get_iftype_ext_capa(struct wiphy *wiphy, enum nl80211_iftype iftype) return (NULL); } +static inline int +cfg80211_external_auth_request(struct net_device *ndev, + struct cfg80211_external_auth_params *params, gfp_t gfp) +{ + TODO(); + return (-ENXIO); +} + static inline uint16_t ieee80211_get_he_6ghz_capa(const struct ieee80211_supported_band *sband, enum nl80211_iftype iftype) @@ -2041,36 +2158,6 @@ ieee80211_get_he_6ghz_capa(const struct ieee80211_supported_band *sband, return (0); } -static inline int -nl80211_chan_width_to_mhz(enum nl80211_chan_width width) -{ - switch (width) { - case NL80211_CHAN_WIDTH_5: - return (5); - break; - case NL80211_CHAN_WIDTH_10: - return (10); - break; - case NL80211_CHAN_WIDTH_20_NOHT: - case NL80211_CHAN_WIDTH_20: - return (20); - break; - case NL80211_CHAN_WIDTH_40: - return (40); - break; - case NL80211_CHAN_WIDTH_80: - case NL80211_CHAN_WIDTH_80P80: - return (80); - break; - case NL80211_CHAN_WIDTH_160: - return (160); - break; - case NL80211_CHAN_WIDTH_320: - return (320); - break; - } -} - static __inline ssize_t wiphy_locked_debugfs_read(struct wiphy *wiphy, struct file *file, char *buf, size_t bufsize, const char __user *userbuf, size_t count, @@ -2093,6 +2180,13 @@ wiphy_locked_debugfs_write(struct wiphy *wiphy, struct file *file, return (-ENXIO); } +static inline void +cfg80211_cqm_rssi_notify(struct net_device *dev, + enum nl80211_cqm_rssi_threshold_event rssi_te, int32_t rssi, gfp_t gfp) +{ + TODO(); +} + /* -------------------------------------------------------------------------- */ static inline void @@ -2140,6 +2234,12 @@ wiphy_delayed_work_cancel(struct wiphy *wiphy, struct wiphy_delayed_work *wdwk) linuxkpi_wiphy_delayed_work_cancel(wiphy, wdwk); } +static inline void +wiphy_delayed_work_flush(struct wiphy *wiphy, struct wiphy_delayed_work *wdwk) +{ + linuxkpi_wiphy_delayed_work_flush(wiphy, wdwk); +} + /* -------------------------------------------------------------------------- */ #define wiphy_err(_wiphy, _fmt, ...) \ diff --git a/sys/compat/linuxkpi/common/include/net/mac80211.h b/sys/compat/linuxkpi/common/include/net/mac80211.h index 8de03410c6b6..523836b52a40 100644 --- a/sys/compat/linuxkpi/common/include/net/mac80211.h +++ b/sys/compat/linuxkpi/common/include/net/mac80211.h @@ -166,7 +166,7 @@ enum ieee80211_bss_changed { #define WLAN_AKM_SUITE_PSK_SHA256 WLAN_AKM_SUITE(6) /* TDLS 7 */ #define WLAN_AKM_SUITE_SAE WLAN_AKM_SUITE(8) -/* FToSAE 9 */ +#define WLAN_AKM_SUITE_FT_OVER_SAE WLAN_AKM_SUITE(9) /* AP peer key 10 */ /* 802.1x suite B 11 */ /* 802.1x suite B 384 12 */ @@ -857,7 +857,8 @@ struct ieee80211_vif_chanctx_switch { }; struct ieee80211_prep_tx_info { - u16 duration; + uint16_t duration; + uint16_t subtype; bool success; bool was_assoc; int link_id; @@ -904,27 +905,6 @@ struct linuxkpi_ieee80211_tim_ie { }; #define ieee80211_tim_ie linuxkpi_ieee80211_tim_ie -struct survey_info { /* net80211::struct ieee80211_channel_survey */ - /* TODO FIXME */ - uint32_t filled; -#define SURVEY_INFO_TIME 0x0001 -#define SURVEY_INFO_TIME_RX 0x0002 -#define SURVEY_INFO_TIME_SCAN 0x0004 -#define SURVEY_INFO_TIME_TX 0x0008 -#define SURVEY_INFO_TIME_BSS_RX 0x0010 -#define SURVEY_INFO_TIME_BUSY 0x0020 -#define SURVEY_INFO_IN_USE 0x0040 -#define SURVEY_INFO_NOISE_DBM 0x0080 - uint32_t noise; - uint64_t time; - uint64_t time_bss_rx; - uint64_t time_busy; - uint64_t time_rx; - uint64_t time_scan; - uint64_t time_tx; - struct ieee80211_channel *channel; -}; - enum ieee80211_iface_iter { IEEE80211_IFACE_ITER_NORMAL = BIT(0), IEEE80211_IFACE_ITER_RESUME_ALL = BIT(1), @@ -1553,6 +1533,15 @@ ieee80211_iter_chan_contexts_atomic(struct ieee80211_hw *hw, } static __inline void +ieee80211_iter_chan_contexts_mtx(struct ieee80211_hw *hw, + void (*iterfunc)(struct ieee80211_hw *, struct ieee80211_chanctx_conf *, void *), + void *arg) +{ + IMPROVE("XXX LKPI80211 TODO MTX\n"); + linuxkpi_ieee80211_iterate_chan_contexts(hw, iterfunc, arg); +} + +static __inline void ieee80211_iterate_stations_atomic(struct ieee80211_hw *hw, void (*iterfunc)(void *, struct ieee80211_sta *), void *arg) { @@ -2063,7 +2052,7 @@ ieee80211_tx_dequeue_ni(struct ieee80211_hw *hw, struct ieee80211_txq *txq) static __inline void ieee80211_update_mu_groups(struct ieee80211_vif *vif, - u_int _i, uint8_t *ms, uint8_t *up) + u_int link_id, const uint8_t *ms, const uint8_t *up) { TODO(); } diff --git a/sys/compat/linuxkpi/common/include/net/netmem.h b/sys/compat/linuxkpi/common/include/net/netmem.h new file mode 100644 index 000000000000..c8de09a2e8c2 --- /dev/null +++ b/sys/compat/linuxkpi/common/include/net/netmem.h @@ -0,0 +1,21 @@ +/*- + * Copyright (c) 2023-2025 Bjoern A. Zeeb + * + * SPDX-License-Identifier: BSD-2-Clause + */ + +#ifndef _LINUXKPI_NET_NETMEM_H +#define _LINUXKPI_NET_NETMEM_H + +struct page_pool; + +struct netmem_desc { + struct page_pool *pp; +}; + +#define pp_page_to_nmdesc(page) \ + (_Generic((page), \ + const struct page *: (const struct netmem_desc *)(page), \ + struct page *: (struct netmem_desc *)(page))) + +#endif /* _LINUXKPI_NET_NETMEM_H */ diff --git a/sys/compat/linuxkpi/common/include/net/page_pool.h b/sys/compat/linuxkpi/common/include/net/page_pool.h deleted file mode 100644 index 2dc8f74b31f3..000000000000 --- a/sys/compat/linuxkpi/common/include/net/page_pool.h +++ /dev/null @@ -1,119 +0,0 @@ -/*- - * Copyright (c) 2023 Bjoern A. Zeeb - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * - * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND - * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE - * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL - * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS - * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) - * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT - * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY - * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF - * SUCH DAMAGE. - */ - -#ifndef _LINUXKPI_NET_PAGE_POOL_H -#define _LINUXKPI_NET_PAGE_POOL_H - -#include <linux/kernel.h> /* pr_debug */ -#include <linux/types.h> -#include <linux/dma-mapping.h> -#include <linux/netdevice.h> - -struct device; - -struct page_pool_params { - struct device *dev; - uint32_t flags; - uint32_t order; - uint32_t pool_size; - uint32_t max_len; - uint32_t offset; - int nid; /* NUMA */ - enum dma_data_direction dma_dir; - struct napi_struct *napi; -}; - -struct page_pool { -}; - -#define PP_FLAG_DMA_MAP BIT(0) -#define PP_FLAG_DMA_SYNC_DEV BIT(1) -#define PP_FLAG_PAGE_FRAG BIT(2) - -static inline struct page_pool * -page_pool_create(const struct page_pool_params *ppparams) -{ - - pr_debug("%s: TODO\n", __func__); - return (NULL); -} - -static inline void -page_pool_destroy(struct page_pool *ppool) -{ - - pr_debug("%s: TODO\n", __func__); -} - -static inline struct page * -page_pool_dev_alloc_frag(struct page_pool *ppool, uint32_t *offset, - size_t size) -{ - - pr_debug("%s: TODO\n", __func__); - return (NULL); -} - -static inline dma_addr_t -page_pool_get_dma_addr(struct page *page) -{ - - pr_debug("%s: TODO\n", __func__); - return (0); -} - -static inline enum dma_data_direction -page_pool_get_dma_dir(const struct page_pool *ppool) -{ - - pr_debug("%s: TODO\n", __func__); - return (DMA_BIDIRECTIONAL); -} - -static inline void -page_pool_put_full_page(struct page_pool *ppool, struct page *page, - bool allow_direct) -{ - - pr_debug("%s: TODO\n", __func__); -} - -static inline int -page_pool_ethtool_stats_get_count(void) -{ - - pr_debug("%s: TODO\n", __func__); - return (0); -} - -static inline uint8_t * -page_pool_ethtool_stats_get_strings(uint8_t *x) -{ - - pr_debug("%s: TODO\n", __func__); - return (x); -} - -#endif /* _LINUXKPI_NET_PAGE_POOL_H */ diff --git a/sys/compat/linuxkpi/common/include/net/page_pool/helpers.h b/sys/compat/linuxkpi/common/include/net/page_pool/helpers.h new file mode 100644 index 000000000000..3469c39c7757 --- /dev/null +++ b/sys/compat/linuxkpi/common/include/net/page_pool/helpers.h @@ -0,0 +1,79 @@ +/*- + * Copyright (c) 2023-2025 Bjoern A. Zeeb + * + * SPDX-License-Identifier: BSD-2-Clause + */ + +#ifndef _LINUXKPI_NET_PAGE_POOL_HELPERS_H +#define _LINUXKPI_NET_PAGE_POOL_HELPERS_H + +#include <linux/kernel.h> /* pr_debug */ +#include <linux/types.h> +#include <linux/dma-mapping.h> +#include <net/page_pool/types.h> + +static inline struct page_pool * +page_pool_create(const struct page_pool_params *ppparams) +{ + + pr_debug("%s: TODO\n", __func__); + return (NULL); +} + +static inline void +page_pool_destroy(struct page_pool *ppool) +{ + + pr_debug("%s: TODO\n", __func__); +} + +static inline struct page * +page_pool_dev_alloc_frag(struct page_pool *ppool, uint32_t *offset, + size_t size) +{ + + pr_debug("%s: TODO\n", __func__); + return (NULL); +} + +static inline dma_addr_t +page_pool_get_dma_addr(struct page *page) +{ + + pr_debug("%s: TODO\n", __func__); + return (0); +} + +static inline enum dma_data_direction +page_pool_get_dma_dir(const struct page_pool *ppool) +{ + + pr_debug("%s: TODO\n", __func__); + return (DMA_BIDIRECTIONAL); +} + +static inline void +page_pool_put_full_page(struct page_pool *ppool, struct page *page, + bool allow_direct) +{ + + pr_debug("%s: TODO\n", __func__); +} + +static inline int +page_pool_ethtool_stats_get_count(void) +{ + + pr_debug("%s: TODO\n", __func__); + return (0); +} + +static inline uint8_t * +page_pool_ethtool_stats_get_strings(uint8_t *x) +{ + + pr_debug("%s: TODO\n", __func__); + return (x); +} + +#endif /* _LINUXKPI_NET_PAGE_POOL_HELPERS_H */ diff --git a/sys/compat/linuxkpi/common/include/net/page_pool/types.h b/sys/compat/linuxkpi/common/include/net/page_pool/types.h new file mode 100644 index 000000000000..6747be50b9b2 --- /dev/null +++ b/sys/compat/linuxkpi/common/include/net/page_pool/types.h @@ -0,0 +1,36 @@ +/*- + * Copyright (c) 2023-2025 Bjoern A. Zeeb + * + * SPDX-License-Identifier: BSD-2-Clause + */ + +#ifndef _LINUXKPI_NET_PAGE_POOL_TYPES_H +#define _LINUXKPI_NET_PAGE_POOL_TYPES_H + +#include <linux/types.h> +#include <linux/dma-mapping.h> +#include <net/netmem.h> + +struct device; +struct napi_struct; + +struct page_pool_params { + struct device *dev; + uint32_t flags; + uint32_t order; + uint32_t pool_size; + uint32_t max_len; + uint32_t offset; + int nid; /* NUMA */ + enum dma_data_direction dma_dir; + struct napi_struct *napi; +}; + +struct page_pool { +}; + +#define PP_FLAG_DMA_MAP BIT(0) +#define PP_FLAG_DMA_SYNC_DEV BIT(1) +#define PP_FLAG_PAGE_FRAG BIT(2) + +#endif /* _LINUXKPI_NET_PAGE_POOL_TYPES_H */ diff --git a/sys/compat/linuxkpi/common/src/linux_80211.c b/sys/compat/linuxkpi/common/src/linux_80211.c index bc4b334de28e..0dc3b2631804 100644 --- a/sys/compat/linuxkpi/common/src/linux_80211.c +++ b/sys/compat/linuxkpi/common/src/linux_80211.c @@ -7833,7 +7833,7 @@ lkpi_wiphy_delayed_work_timer(struct timer_list *tl) struct wiphy_delayed_work *wdwk; wdwk = timer_container_of(wdwk, tl, timer); - wiphy_work_queue(wdwk->wiphy, &wdwk->work); + wiphy_work_queue(wdwk->wiphy, &wdwk->work); } void @@ -7858,6 +7858,16 @@ linuxkpi_wiphy_delayed_work_cancel(struct wiphy *wiphy, wiphy_work_cancel(wiphy, &wdwk->work); } +void +linuxkpi_wiphy_delayed_work_flush(struct wiphy *wiphy, + struct wiphy_delayed_work *wdwk) +{ + lockdep_assert_held(&wiphy->mtx); + + del_timer_sync(&wdwk->timer); + wiphy_work_flush(wiphy, &wdwk->work); +} + /* -------------------------------------------------------------------------- */ struct wiphy * diff --git a/sys/compat/linuxkpi/common/src/linux_80211_macops.c b/sys/compat/linuxkpi/common/src/linux_80211_macops.c index 1046b753574f..04f9f6d7e7fc 100644 --- a/sys/compat/linuxkpi/common/src/linux_80211_macops.c +++ b/sys/compat/linuxkpi/common/src/linux_80211_macops.c @@ -42,7 +42,7 @@ if (linuxkpi_debug_80211 & D80211_TRACE_MO) \ printf("LKPI_80211_TRACE_MO %s:%d: %d %d %lu: " fmt "\n", \ __func__, __LINE__, curcpu, curthread->td_tid, \ - jiffies, __VA_ARGS__) + jiffies, ##__VA_ARGS__) #else #define LKPI_80211_TRACE_MO(...) do { } while(0) #endif diff --git a/sys/conf/dtb.build.mk b/sys/conf/dtb.build.mk index 7eb0db5e8b80..f0cd8f8d515c 100644 --- a/sys/conf/dtb.build.mk +++ b/sys/conf/dtb.build.mk @@ -15,9 +15,10 @@ SYSDIR= ${S} .endif .for _dts in ${DTS} -# DTB for aarch64 needs to preserve the immediate parent of the .dts, because -# these DTS are vendored and should be installed into their vendored directory. -.if ${MACHINE_CPUARCH} == "aarch64" +# DTBs for aarch64 and riscv need to preserve the immediate parent of the .dts, +# because these DTS are vendored and should be installed into their vendored +# directories. +.if ${MACHINE_CPUARCH} == "aarch64" || ${MACHINE_CPUARCH} == "riscv" DTB+= ${_dts:R:S/$/.dtb/} .else DTB+= ${_dts:T:R:S/$/.dtb/} @@ -54,7 +55,7 @@ _dtbinstall: # entries in the NO_ROOT case. test -d ${DESTDIR}${DTBDIR} || ${INSTALL} -d -o ${DTBOWN} -g ${DTBGRP} ${DESTDIR}${DTBDIR} .for _dtb in ${DTB} -.if ${MACHINE_CPUARCH} == "aarch64" +.if ${MACHINE_CPUARCH} == "aarch64" || ${MACHINE_CPUARCH} == "riscv" # :H:T here to grab the vendor component of the DTB path in a way that # allows out-of-tree DTS builds, too. We make the assumption that # out-of-tree DTS will have a similar directory structure to in-tree, diff --git a/sys/conf/files.x86 b/sys/conf/files.x86 index 953da7dd1284..21a1b8046f12 100644 --- a/sys/conf/files.x86 +++ b/sys/conf/files.x86 @@ -311,6 +311,7 @@ dev/ntb/test/ntb_tool.c optional ntb_tool dev/nvram/nvram.c optional nvram isa dev/random/ivy.c optional rdrand_rng !random_loadable dev/random/nehemiah.c optional padlock_rng !random_loadable +dev/random/rdseed.c optional rdrand_rng !random_loadable dev/qat_c2xxx/qat.c optional qat_c2xxx dev/qat_c2xxx/qat_ae.c optional qat_c2xxx dev/qat_c2xxx/qat_c2xxx.c optional qat_c2xxx diff --git a/sys/contrib/openzfs/.github/workflows/scripts/qemu-2-start.sh b/sys/contrib/openzfs/.github/workflows/scripts/qemu-2-start.sh index 1c608348ffcd..422b3e9df388 100755 --- a/sys/contrib/openzfs/.github/workflows/scripts/qemu-2-start.sh +++ b/sys/contrib/openzfs/.github/workflows/scripts/qemu-2-start.sh @@ -121,7 +121,14 @@ case "$OS" in KSRC="$FREEBSD_SNAP/../amd64/$FreeBSD/src.txz" ;; freebsd15-0c) - FreeBSD="15.0-ALPHA3" + FreeBSD="15.0-ALPHA4" + OSNAME="FreeBSD $FreeBSD" + OSv="freebsd14.0" + URLxz="$FREEBSD_SNAP/$FreeBSD/amd64/Latest/FreeBSD-$FreeBSD-amd64-BASIC-CI-ufs.raw.xz" + KSRC="$FREEBSD_SNAP/../amd64/$FreeBSD/src.txz" + ;; + freebsd16-0c) + FreeBSD="16.0-CURRENT" OSNAME="FreeBSD $FreeBSD" OSv="freebsd14.0" URLxz="$FREEBSD_SNAP/$FreeBSD/amd64/Latest/FreeBSD-$FreeBSD-amd64-BASIC-CI-ufs.raw.xz" @@ -287,7 +294,7 @@ else while pidof /usr/bin/qemu-system-x86_64 >/dev/null; do ssh 2>/dev/null root@vm0 "uname -a" && break done - ssh root@vm0 "pkg install -y bash ca_root_nss git qemu-guest-agent python3 py311-cloud-init" + ssh root@vm0 "env IGNORE_OSVERSION=yes pkg install -y bash ca_root_nss git qemu-guest-agent python3 py311-cloud-init" ssh root@vm0 "chsh -s $BASH root" ssh root@vm0 'sysrc qemu_guest_agent_enable="YES"' ssh root@vm0 'sysrc cloudinit_enable="YES"' diff --git a/sys/contrib/openzfs/.github/workflows/zfs-qemu.yml b/sys/contrib/openzfs/.github/workflows/zfs-qemu.yml index 69349678d84c..3b164548f9be 100644 --- a/sys/contrib/openzfs/.github/workflows/zfs-qemu.yml +++ b/sys/contrib/openzfs/.github/workflows/zfs-qemu.yml @@ -68,7 +68,7 @@ jobs: # FreeBSD variants of 2025-06: # FreeBSD Release: freebsd13-5r, freebsd14-2r, freebsd14-3r # FreeBSD Stable: freebsd13-5s, freebsd14-3s - # FreeBSD Current: freebsd15-0c + # FreeBSD Current: freebsd15-0c, freebsd16-0c os: ${{ fromJson(needs.test-config.outputs.test_os) }} runs-on: ubuntu-24.04 steps: diff --git a/sys/contrib/openzfs/cmd/zdb/zdb.c b/sys/contrib/openzfs/cmd/zdb/zdb.c index 70a4ed46f263..1ffb8ebc70f2 100644 --- a/sys/contrib/openzfs/cmd/zdb/zdb.c +++ b/sys/contrib/openzfs/cmd/zdb/zdb.c @@ -106,11 +106,15 @@ extern boolean_t spa_mode_readable_spacemaps; extern uint_t zfs_reconstruct_indirect_combinations_max; extern uint_t zfs_btree_verify_intensity; +enum { + ARG_ALLOCATED = 256, + ARG_BLOCK_BIN_MODE, + ARG_BLOCK_CLASSES, +}; + static const char cmdname[] = "zdb"; uint8_t dump_opt[512]; -#define ALLOCATED_OPT 256 - typedef void object_viewer_t(objset_t *, uint64_t, void *data, size_t size); static uint64_t *zopt_metaslab = NULL; @@ -131,6 +135,20 @@ static objset_t *os; static boolean_t kernel_init_done; static boolean_t corruption_found = B_FALSE; +static enum { + BIN_AUTO = 0, + BIN_PSIZE, + BIN_LSIZE, + BIN_ASIZE, +} block_bin_mode = BIN_AUTO; + +static enum { + CLASS_NORMAL = 1 << 1, + CLASS_SPECIAL = 1 << 2, + CLASS_DEDUP = 1 << 3, + CLASS_OTHER = 1 << 4, +} block_classes = 0; + static void snprintf_blkptr_compact(char *, size_t, const blkptr_t *, boolean_t); static void mos_obj_refd(uint64_t); @@ -749,6 +767,12 @@ usage(void) (void) fprintf(stderr, " Options to control amount of output:\n"); (void) fprintf(stderr, " -b --block-stats " "block statistics\n"); + (void) fprintf(stderr, " --bin=(lsize|psize|asize) " + "bin blocks based on this size in all three columns\n"); + (void) fprintf(stderr, + " --class=(normal|special|dedup|other)[,...]\n" + " only consider blocks from " + "these allocation classes\n"); (void) fprintf(stderr, " -B --backup " "backup stream\n"); (void) fprintf(stderr, " -c --checksum " @@ -1694,7 +1718,7 @@ dump_metaslab(metaslab_t *msp) (u_longlong_t)msp->ms_id, (u_longlong_t)msp->ms_start, (u_longlong_t)space_map_object(sm), freebuf); - if (dump_opt[ALLOCATED_OPT] || + if (dump_opt[ARG_ALLOCATED] || (dump_opt['m'] > 2 && !dump_opt['L'])) { mutex_enter(&msp->ms_lock); VERIFY0(metaslab_load(msp)); @@ -1705,7 +1729,7 @@ dump_metaslab(metaslab_t *msp) dump_metaslab_stats(msp); } - if (dump_opt[ALLOCATED_OPT]) { + if (dump_opt[ARG_ALLOCATED]) { uint64_t off = msp->ms_start; zfs_range_tree_walk(msp->ms_allocatable, dump_allocated, &off); @@ -1726,7 +1750,7 @@ dump_metaslab(metaslab_t *msp) SPACE_MAP_HISTOGRAM_SIZE, sm->sm_shift); } - if (dump_opt[ALLOCATED_OPT] || + if (dump_opt[ARG_ALLOCATED] || (dump_opt['m'] > 2 && !dump_opt['L'])) { metaslab_unload(msp); mutex_exit(&msp->ms_lock); @@ -5814,6 +5838,34 @@ dump_size_histograms(zdb_cb_t *zcb) (void) printf("\nBlock Size Histogram\n"); + switch (block_bin_mode) { + case BIN_PSIZE: + printf("(note: all categories are binned by %s)\n", "psize"); + break; + case BIN_LSIZE: + printf("(note: all categories are binned by %s)\n", "lsize"); + break; + case BIN_ASIZE: + printf("(note: all categories are binned by %s)\n", "asize"); + break; + default: + printf("(note: all categories are binned separately)\n"); + break; + } + if (block_classes != 0) { + char buf[256] = ""; + if (block_classes & CLASS_NORMAL) + strlcat(buf, "\"normal\", ", sizeof (buf)); + if (block_classes & CLASS_SPECIAL) + strlcat(buf, "\"special\", ", sizeof (buf)); + if (block_classes & CLASS_DEDUP) + strlcat(buf, "\"dedup\", ", sizeof (buf)); + if (block_classes & CLASS_OTHER) + strlcat(buf, "\"other\", ", sizeof (buf)); + buf[strlen(buf)-2] = '\0'; + printf("(note: only blocks in these classes are counted: %s)\n", + buf); + } /* * Print the first line titles */ @@ -6162,29 +6214,85 @@ skipped: [BPE_GET_PSIZE(bp)]++; return; } + + if (block_classes != 0) { + spa_config_enter(zcb->zcb_spa, SCL_CONFIG, FTAG, RW_READER); + + uint64_t vdev = DVA_GET_VDEV(&bp->blk_dva[0]); + uint64_t offset = DVA_GET_OFFSET(&bp->blk_dva[0]); + vdev_t *vd = vdev_lookup_top(zcb->zcb_spa, vdev); + ASSERT(vd != NULL); + metaslab_t *ms = vd->vdev_ms[offset >> vd->vdev_ms_shift]; + ASSERT(ms != NULL); + metaslab_group_t *mg = ms->ms_group; + ASSERT(mg != NULL); + metaslab_class_t *mc = mg->mg_class; + ASSERT(mc != NULL); + + spa_config_exit(zcb->zcb_spa, SCL_CONFIG, FTAG); + + int class; + if (mc == spa_normal_class(zcb->zcb_spa)) { + class = CLASS_NORMAL; + } else if (mc == spa_special_class(zcb->zcb_spa)) { + class = CLASS_SPECIAL; + } else if (mc == spa_dedup_class(zcb->zcb_spa)) { + class = CLASS_DEDUP; + } else { + class = CLASS_OTHER; + } + + if (!(block_classes & class)) { + goto hist_skipped; + } + } + /* * The binning histogram bins by powers of two up to * SPA_MAXBLOCKSIZE rather than creating bins for * every possible blocksize found in the pool. */ - int bin = highbit64(BP_GET_PSIZE(bp)) - 1; + int bin; + + /* + * Binning strategy: each bin includes blocks up to and including + * the given size (excluding blocks that fit into the previous bin). + * This way, the "4K" bin includes blocks within the (2K; 4K] range. + */ +#define BIN(size) (highbit64((size) - 1)) + + switch (block_bin_mode) { + case BIN_PSIZE: bin = BIN(BP_GET_PSIZE(bp)); break; + case BIN_LSIZE: bin = BIN(BP_GET_LSIZE(bp)); break; + case BIN_ASIZE: bin = BIN(BP_GET_ASIZE(bp)); break; + case BIN_AUTO: break; + default: PANIC("bad block_bin_mode"); abort(); + } + + if (block_bin_mode == BIN_AUTO) + bin = BIN(BP_GET_PSIZE(bp)); zcb->zcb_psize_count[bin]++; zcb->zcb_psize_len[bin] += BP_GET_PSIZE(bp); zcb->zcb_psize_total += BP_GET_PSIZE(bp); - bin = highbit64(BP_GET_LSIZE(bp)) - 1; + if (block_bin_mode == BIN_AUTO) + bin = BIN(BP_GET_LSIZE(bp)); zcb->zcb_lsize_count[bin]++; zcb->zcb_lsize_len[bin] += BP_GET_LSIZE(bp); zcb->zcb_lsize_total += BP_GET_LSIZE(bp); - bin = highbit64(BP_GET_ASIZE(bp)) - 1; + if (block_bin_mode == BIN_AUTO) + bin = BIN(BP_GET_ASIZE(bp)); zcb->zcb_asize_count[bin]++; zcb->zcb_asize_len[bin] += BP_GET_ASIZE(bp); zcb->zcb_asize_total += BP_GET_ASIZE(bp); +#undef BIN + +hist_skipped: if (!do_claim) return; @@ -9426,7 +9534,11 @@ main(int argc, char **argv) {"livelist", no_argument, NULL, 'y'}, {"zstd-headers", no_argument, NULL, 'Z'}, {"allocated-map", no_argument, NULL, - ALLOCATED_OPT}, + ARG_ALLOCATED}, + {"bin", required_argument, NULL, + ARG_BLOCK_BIN_MODE}, + {"class", required_argument, NULL, + ARG_BLOCK_CLASSES}, {0, 0, 0, 0} }; @@ -9457,7 +9569,7 @@ main(int argc, char **argv) case 'u': case 'y': case 'Z': - case ALLOCATED_OPT: + case ARG_ALLOCATED: dump_opt[c]++; dump_all = 0; break; @@ -9540,6 +9652,59 @@ main(int argc, char **argv) case 'x': vn_dumpdir = optarg; break; + case ARG_BLOCK_BIN_MODE: + if (strcmp(optarg, "lsize") == 0) { + block_bin_mode = BIN_LSIZE; + } else if (strcmp(optarg, "psize") == 0) { + block_bin_mode = BIN_PSIZE; + } else if (strcmp(optarg, "asize") == 0) { + block_bin_mode = BIN_ASIZE; + } else { + (void) fprintf(stderr, + "--bin=\"%s\" must be one of \"lsize\", " + "\"psize\" or \"asize\"\n", optarg); + usage(); + } + break; + + case ARG_BLOCK_CLASSES: { + char *buf = strdup(optarg), *tok = buf, *next, + *save = NULL; + + while ((next = strtok_r(tok, ",", &save)) != NULL) { + tok = NULL; + + if (strcmp(next, "normal") == 0) { + block_classes |= CLASS_NORMAL; + } else if (strcmp(next, "special") == 0) { + block_classes |= CLASS_SPECIAL; + } else if (strcmp(next, "dedup") == 0) { + block_classes |= CLASS_DEDUP; + } else if (strcmp(next, "other") == 0) { + block_classes |= CLASS_OTHER; + } else { + (void) fprintf(stderr, + "--class=\"%s\" must be a " + "comma-separated list of either " + "\"normal\", \"special\", " + "\"asize\" or \"other\"; " + "got \"%s\"\n", + optarg, next); + usage(); + } + } + + if (block_classes == 0) { + (void) fprintf(stderr, + "--class= must be a comma-separated " + "list of either \"normal\", \"special\", " + "\"asize\" or \"other\"; got empty\n"); + usage(); + } + + free(buf); + break; + } default: usage(); break; diff --git a/sys/contrib/openzfs/cmd/zpool/zpool_vdev.c b/sys/contrib/openzfs/cmd/zpool/zpool_vdev.c index 088c0108e911..222b5524669e 100644 --- a/sys/contrib/openzfs/cmd/zpool/zpool_vdev.c +++ b/sys/contrib/openzfs/cmd/zpool/zpool_vdev.c @@ -270,14 +270,13 @@ is_spare(nvlist_t *config, const char *path) * draid* Virtual dRAID spare */ static nvlist_t * -make_leaf_vdev(nvlist_t *props, const char *arg, boolean_t is_primary) +make_leaf_vdev(const char *arg, boolean_t is_primary, uint64_t ashift) { char path[MAXPATHLEN]; struct stat64 statbuf; nvlist_t *vdev = NULL; const char *type = NULL; boolean_t wholedisk = B_FALSE; - uint64_t ashift = 0; int err; /* @@ -382,31 +381,6 @@ make_leaf_vdev(nvlist_t *props, const char *arg, boolean_t is_primary) (uint64_t)wholedisk) == 0); /* - * Override defaults if custom properties are provided. - */ - if (props != NULL) { - const char *value = NULL; - - if (nvlist_lookup_string(props, - zpool_prop_to_name(ZPOOL_PROP_ASHIFT), &value) == 0) { - if (zfs_nicestrtonum(NULL, value, &ashift) != 0) { - (void) fprintf(stderr, - gettext("ashift must be a number.\n")); - return (NULL); - } - if (ashift != 0 && - (ashift < ASHIFT_MIN || ashift > ASHIFT_MAX)) { - (void) fprintf(stderr, - gettext("invalid 'ashift=%" PRIu64 "' " - "property: only values between %" PRId32 " " - "and %" PRId32 " are allowed.\n"), - ashift, ASHIFT_MIN, ASHIFT_MAX); - return (NULL); - } - } - } - - /* * If the device is known to incorrectly report its physical sector * size explicitly provide the known correct value. */ @@ -1513,6 +1487,29 @@ construct_spec(nvlist_t *props, int argc, char **argv) const char *type, *fulltype; boolean_t is_log, is_special, is_dedup, is_spare; boolean_t seen_logs; + uint64_t ashift = 0; + + if (props != NULL) { + const char *value = NULL; + + if (nvlist_lookup_string(props, + zpool_prop_to_name(ZPOOL_PROP_ASHIFT), &value) == 0) { + if (zfs_nicestrtonum(NULL, value, &ashift) != 0) { + (void) fprintf(stderr, + gettext("ashift must be a number.\n")); + return (NULL); + } + if (ashift != 0 && + (ashift < ASHIFT_MIN || ashift > ASHIFT_MAX)) { + (void) fprintf(stderr, + gettext("invalid 'ashift=%" PRIu64 "' " + "property: only values between %" PRId32 " " + "and %" PRId32 " are allowed.\n"), + ashift, ASHIFT_MIN, ASHIFT_MAX); + return (NULL); + } + } + } top = NULL; toplevels = 0; @@ -1618,9 +1615,9 @@ construct_spec(nvlist_t *props, int argc, char **argv) children * sizeof (nvlist_t *)); if (child == NULL) zpool_no_memory(); - if ((nv = make_leaf_vdev(props, argv[c], + if ((nv = make_leaf_vdev(argv[c], !(is_log || is_special || is_dedup || - is_spare))) == NULL) { + is_spare), ashift)) == NULL) { for (c = 0; c < children - 1; c++) nvlist_free(child[c]); free(child); @@ -1684,6 +1681,10 @@ construct_spec(nvlist_t *props, int argc, char **argv) ZPOOL_CONFIG_ALLOCATION_BIAS, VDEV_ALLOC_BIAS_DEDUP) == 0); } + if (ashift > 0) { + fnvlist_add_uint64(nv, + ZPOOL_CONFIG_ASHIFT, ashift); + } if (strcmp(type, VDEV_TYPE_RAIDZ) == 0) { verify(nvlist_add_uint64(nv, ZPOOL_CONFIG_NPARITY, @@ -1711,8 +1712,9 @@ construct_spec(nvlist_t *props, int argc, char **argv) * We have a device. Pass off to make_leaf_vdev() to * construct the appropriate nvlist describing the vdev. */ - if ((nv = make_leaf_vdev(props, argv[0], !(is_log || - is_special || is_dedup || is_spare))) == NULL) + if ((nv = make_leaf_vdev(argv[0], !(is_log || + is_special || is_dedup || is_spare), + ashift)) == NULL) goto spec_out; verify(nvlist_add_uint64(nv, diff --git a/sys/contrib/openzfs/lib/libspl/include/sys/uio.h b/sys/contrib/openzfs/lib/libspl/include/sys/uio.h index 93aa4984d734..9ada482be000 100644 --- a/sys/contrib/openzfs/lib/libspl/include/sys/uio.h +++ b/sys/contrib/openzfs/lib/libspl/include/sys/uio.h @@ -41,6 +41,7 @@ #ifndef _LIBSPL_SYS_UIO_H #define _LIBSPL_SYS_UIO_H +#include <sys/sysmacros.h> #include <sys/types.h> #include_next <sys/uio.h> diff --git a/sys/contrib/openzfs/lib/libzfs/libzfs_status.c b/sys/contrib/openzfs/lib/libzfs/libzfs_status.c index bdddefb92165..a589ca6896f0 100644 --- a/sys/contrib/openzfs/lib/libzfs/libzfs_status.c +++ b/sys/contrib/openzfs/lib/libzfs/libzfs_status.c @@ -98,57 +98,57 @@ static const char *const zfs_msgid_table[] = { #define NMSGID (sizeof (zfs_msgid_table) / sizeof (zfs_msgid_table[0])) static int -vdev_missing(vdev_stat_t *vs, uint_t vsc) +vdev_missing(vdev_stat_t *vs, uint_t vsc, void *arg) { - (void) vsc; + (void) vsc, (void) arg; return (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_OPEN_FAILED); } static int -vdev_faulted(vdev_stat_t *vs, uint_t vsc) +vdev_faulted(vdev_stat_t *vs, uint_t vsc, void *arg) { - (void) vsc; + (void) vsc, (void) arg; return (vs->vs_state == VDEV_STATE_FAULTED); } static int -vdev_errors(vdev_stat_t *vs, uint_t vsc) +vdev_errors(vdev_stat_t *vs, uint_t vsc, void *arg) { - (void) vsc; + (void) vsc, (void) arg; return (vs->vs_state == VDEV_STATE_DEGRADED || vs->vs_read_errors != 0 || vs->vs_write_errors != 0 || vs->vs_checksum_errors != 0); } static int -vdev_broken(vdev_stat_t *vs, uint_t vsc) +vdev_broken(vdev_stat_t *vs, uint_t vsc, void *arg) { - (void) vsc; + (void) vsc, (void) arg; return (vs->vs_state == VDEV_STATE_CANT_OPEN); } static int -vdev_offlined(vdev_stat_t *vs, uint_t vsc) +vdev_offlined(vdev_stat_t *vs, uint_t vsc, void *arg) { - (void) vsc; + (void) vsc, (void) arg; return (vs->vs_state == VDEV_STATE_OFFLINE); } static int -vdev_removed(vdev_stat_t *vs, uint_t vsc) +vdev_removed(vdev_stat_t *vs, uint_t vsc, void *arg) { - (void) vsc; + (void) vsc, (void) arg; return (vs->vs_state == VDEV_STATE_REMOVED); } static int -vdev_non_native_ashift(vdev_stat_t *vs, uint_t vsc) +vdev_non_native_ashift(vdev_stat_t *vs, uint_t vsc, void *arg) { - if (getenv("ZPOOL_STATUS_NON_NATIVE_ASHIFT_IGNORE") != NULL) - return (0); + uint64_t ashift = *(uint64_t *)arg; return (VDEV_STAT_VALID(vs_physical_ashift, vsc) && + (ashift == 0 || vs->vs_configured_ashift < ashift) && vs->vs_configured_ashift < vs->vs_physical_ashift); } @@ -156,8 +156,8 @@ vdev_non_native_ashift(vdev_stat_t *vs, uint_t vsc) * Detect if any leaf devices that have seen errors or could not be opened. */ static boolean_t -find_vdev_problem(nvlist_t *vdev, int (*func)(vdev_stat_t *, uint_t), - boolean_t ignore_replacing) +find_vdev_problem(nvlist_t *vdev, int (*func)(vdev_stat_t *, uint_t, void *), + void *arg, boolean_t ignore_replacing) { nvlist_t **child; uint_t c, children; @@ -177,14 +177,16 @@ find_vdev_problem(nvlist_t *vdev, int (*func)(vdev_stat_t *, uint_t), if (nvlist_lookup_nvlist_array(vdev, ZPOOL_CONFIG_CHILDREN, &child, &children) == 0) { - for (c = 0; c < children; c++) - if (find_vdev_problem(child[c], func, ignore_replacing)) + for (c = 0; c < children; c++) { + if (find_vdev_problem(child[c], func, arg, + ignore_replacing)) return (B_TRUE); + } } else { uint_t vsc; vdev_stat_t *vs = (vdev_stat_t *)fnvlist_lookup_uint64_array( vdev, ZPOOL_CONFIG_VDEV_STATS, &vsc); - if (func(vs, vsc) != 0) + if (func(vs, vsc, arg) != 0) return (B_TRUE); } @@ -193,9 +195,11 @@ find_vdev_problem(nvlist_t *vdev, int (*func)(vdev_stat_t *, uint_t), */ if (nvlist_lookup_nvlist_array(vdev, ZPOOL_CONFIG_L2CACHE, &child, &children) == 0) { - for (c = 0; c < children; c++) - if (find_vdev_problem(child[c], func, ignore_replacing)) + for (c = 0; c < children; c++) { + if (find_vdev_problem(child[c], func, arg, + ignore_replacing)) return (B_TRUE); + } } return (B_FALSE); @@ -220,7 +224,7 @@ find_vdev_problem(nvlist_t *vdev, int (*func)(vdev_stat_t *, uint_t), */ static zpool_status_t check_status(nvlist_t *config, boolean_t isimport, - zpool_errata_t *erratap, const char *compat) + zpool_errata_t *erratap, const char *compat, uint64_t ashift) { pool_scan_stat_t *ps = NULL; uint_t vsc, psc; @@ -371,15 +375,15 @@ check_status(nvlist_t *config, boolean_t isimport, * Bad devices in non-replicated config. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && - find_vdev_problem(nvroot, vdev_faulted, B_TRUE)) + find_vdev_problem(nvroot, vdev_faulted, NULL, B_TRUE)) return (ZPOOL_STATUS_FAULTED_DEV_NR); if (vs->vs_state == VDEV_STATE_CANT_OPEN && - find_vdev_problem(nvroot, vdev_missing, B_TRUE)) + find_vdev_problem(nvroot, vdev_missing, NULL, B_TRUE)) return (ZPOOL_STATUS_MISSING_DEV_NR); if (vs->vs_state == VDEV_STATE_CANT_OPEN && - find_vdev_problem(nvroot, vdev_broken, B_TRUE)) + find_vdev_problem(nvroot, vdev_broken, NULL, B_TRUE)) return (ZPOOL_STATUS_CORRUPT_LABEL_NR); /* @@ -402,35 +406,37 @@ check_status(nvlist_t *config, boolean_t isimport, /* * Missing devices in a replicated config. */ - if (find_vdev_problem(nvroot, vdev_faulted, B_TRUE)) + if (find_vdev_problem(nvroot, vdev_faulted, NULL, B_TRUE)) return (ZPOOL_STATUS_FAULTED_DEV_R); - if (find_vdev_problem(nvroot, vdev_missing, B_TRUE)) + if (find_vdev_problem(nvroot, vdev_missing, NULL, B_TRUE)) return (ZPOOL_STATUS_MISSING_DEV_R); - if (find_vdev_problem(nvroot, vdev_broken, B_TRUE)) + if (find_vdev_problem(nvroot, vdev_broken, NULL, B_TRUE)) return (ZPOOL_STATUS_CORRUPT_LABEL_R); /* * Devices with errors */ - if (!isimport && find_vdev_problem(nvroot, vdev_errors, B_TRUE)) + if (!isimport && find_vdev_problem(nvroot, vdev_errors, NULL, B_TRUE)) return (ZPOOL_STATUS_FAILING_DEV); /* * Offlined devices */ - if (find_vdev_problem(nvroot, vdev_offlined, B_TRUE)) + if (find_vdev_problem(nvroot, vdev_offlined, NULL, B_TRUE)) return (ZPOOL_STATUS_OFFLINE_DEV); /* * Removed device */ - if (find_vdev_problem(nvroot, vdev_removed, B_TRUE)) + if (find_vdev_problem(nvroot, vdev_removed, NULL, B_TRUE)) return (ZPOOL_STATUS_REMOVED_DEV); /* * Suboptimal, but usable, ashift configuration. */ - if (find_vdev_problem(nvroot, vdev_non_native_ashift, B_FALSE)) + if (!isimport && + getenv("ZPOOL_STATUS_NON_NATIVE_ASHIFT_IGNORE") == NULL && + find_vdev_problem(nvroot, vdev_non_native_ashift, &ashift, B_FALSE)) return (ZPOOL_STATUS_NON_NATIVE_ASHIFT); /* @@ -510,8 +516,10 @@ zpool_get_status(zpool_handle_t *zhp, const char **msgid, ZFS_MAXPROPLEN, NULL, B_FALSE) != 0) compatibility[0] = '\0'; + uint64_t ashift = zpool_get_prop_int(zhp, ZPOOL_PROP_ASHIFT, NULL); + zpool_status_t ret = check_status(zhp->zpool_config, B_FALSE, errata, - compatibility); + compatibility, ashift); if (msgid != NULL) { if (ret >= NMSGID) @@ -526,7 +534,7 @@ zpool_status_t zpool_import_status(nvlist_t *config, const char **msgid, zpool_errata_t *errata) { - zpool_status_t ret = check_status(config, B_TRUE, errata, NULL); + zpool_status_t ret = check_status(config, B_TRUE, errata, NULL, 0); if (ret >= NMSGID) *msgid = NULL; diff --git a/sys/contrib/openzfs/lib/libzpool/kernel.c b/sys/contrib/openzfs/lib/libzpool/kernel.c index 8ed374627264..70eba5099119 100644 --- a/sys/contrib/openzfs/lib/libzpool/kernel.c +++ b/sys/contrib/openzfs/lib/libzpool/kernel.c @@ -23,6 +23,7 @@ * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. * Copyright (c) 2012, 2018 by Delphix. All rights reserved. * Copyright (c) 2016 Actifio, Inc. All rights reserved. + * Copyright (c) 2025, Klara, Inc. */ #include <assert.h> @@ -645,39 +646,60 @@ __dprintf(boolean_t dprint, const char *file, const char *func, * cmn_err() and panic() * ========================================================================= */ -static char ce_prefix[CE_IGNORE][10] = { "", "NOTICE: ", "WARNING: ", "" }; -static char ce_suffix[CE_IGNORE][2] = { "", "\n", "\n", "" }; -__attribute__((noreturn)) void -vpanic(const char *fmt, va_list adx) +static __attribute__((noreturn)) void +panic_stop_or_abort(void) { - (void) fprintf(stderr, "error: "); - (void) vfprintf(stderr, fmt, adx); - (void) fprintf(stderr, "\n"); + const char *stopenv = getenv("LIBZPOOL_PANIC_STOP"); + if (stopenv != NULL && atoi(stopenv)) { + fputs("libzpool: LIBZPOOL_PANIC_STOP is set, sending " + "SIGSTOP to process group\n", stderr); + fflush(stderr); + + kill(0, SIGSTOP); + + fputs("libzpool: continued after panic stop, " + "aborting\n", stderr); + } abort(); /* think of it as a "user-level crash dump" */ } -__attribute__((noreturn)) void -panic(const char *fmt, ...) +static void +vcmn_msg(int ce, const char *fmt, va_list adx) { - va_list adx; + switch (ce) { + case CE_IGNORE: + return; + case CE_CONT: + break; + case CE_NOTE: + fputs("libzpool: NOTICE: ", stderr); + break; + case CE_WARN: + fputs("libzpool: WARNING: ", stderr); + break; + case CE_PANIC: + fputs("libzpool: PANIC: ", stderr); + break; + default: + fputs("libzpool: [unknown severity %d]: ", stderr); + break; + } - va_start(adx, fmt); - vpanic(fmt, adx); - va_end(adx); + vfprintf(stderr, fmt, adx); + if (ce != CE_CONT) + fputc('\n', stderr); + fflush(stderr); } void vcmn_err(int ce, const char *fmt, va_list adx) { + vcmn_msg(ce, fmt, adx); + if (ce == CE_PANIC) - vpanic(fmt, adx); - if (ce != CE_NOTE) { /* suppress noise in userland stress testing */ - (void) fprintf(stderr, "%s", ce_prefix[ce]); - (void) vfprintf(stderr, fmt, adx); - (void) fprintf(stderr, "%s", ce_suffix[ce]); - } + panic_stop_or_abort(); } void @@ -690,6 +712,25 @@ cmn_err(int ce, const char *fmt, ...) va_end(adx); } +__attribute__((noreturn)) void +panic(const char *fmt, ...) +{ + va_list adx; + + va_start(adx, fmt); + vcmn_msg(CE_PANIC, fmt, adx); + va_end(adx); + + panic_stop_or_abort(); +} + +__attribute__((noreturn)) void +vpanic(const char *fmt, va_list adx) +{ + vcmn_msg(CE_PANIC, fmt, adx); + panic_stop_or_abort(); +} + /* * ========================================================================= * misc routines diff --git a/sys/contrib/openzfs/man/man8/zdb.8 b/sys/contrib/openzfs/man/man8/zdb.8 index c3290ea14769..f51e24fa849c 100644 --- a/sys/contrib/openzfs/man/man8/zdb.8 +++ b/sys/contrib/openzfs/man/man8/zdb.8 @@ -144,6 +144,20 @@ subcommand. Display statistics regarding the number, size .Pq logical, physical and allocated and deduplication of blocks. +.It Fl -bin Ns = Ns ( Li lsize Ns | Ns Li psize Ns | Ns Li asize ) +When used with +.Fl bb , +sort blocks into all three bins according to the given size (instead of binning +a block for each size separately). +.Pp +For instance, with +.Fl -bin Ns = Ns Li lsize , +a block with lsize of 16K and psize of 4K will be added to the 16K bin +in all three columns. +.It Fl -class Ns = Ns ( Li normal Ns | Ns Li special Ns | Ns Li dedup Ns | Ns Li other ) Ns Op , Ns … +When used with +.Fl bb , +only consider blocks from these allocation classes. .It Fl B , -backup Generate a backup stream, similar to .Nm zfs Cm send , diff --git a/sys/contrib/openzfs/module/os/freebsd/zfs/zfs_ctldir.c b/sys/contrib/openzfs/module/os/freebsd/zfs/zfs_ctldir.c index 4de48e013ec4..d0a9c662e6f0 100644 --- a/sys/contrib/openzfs/module/os/freebsd/zfs/zfs_ctldir.c +++ b/sys/contrib/openzfs/module/os/freebsd/zfs/zfs_ctldir.c @@ -762,8 +762,7 @@ zfsctl_common_pathconf(struct vop_pathconf_args *ap) return (0); case _PC_MIN_HOLE_SIZE: - *ap->a_retval = (int)SPA_MINBLOCKSIZE; - return (0); + return (EINVAL); case _PC_ACL_EXTENDED: *ap->a_retval = 0; diff --git a/sys/contrib/openzfs/module/os/freebsd/zfs/zfs_vnops_os.c b/sys/contrib/openzfs/module/os/freebsd/zfs/zfs_vnops_os.c index 411225786089..f34a2fd37a77 100644 --- a/sys/contrib/openzfs/module/os/freebsd/zfs/zfs_vnops_os.c +++ b/sys/contrib/openzfs/module/os/freebsd/zfs/zfs_vnops_os.c @@ -4116,6 +4116,7 @@ zfs_pathconf(vnode_t *vp, int cmd, ulong_t *valp, cred_t *cr, { znode_t *zp; zfsvfs_t *zfsvfs; + uint_t blksize, iosize; int error; switch (cmd) { @@ -4127,8 +4128,20 @@ zfs_pathconf(vnode_t *vp, int cmd, ulong_t *valp, cred_t *cr, *valp = 64; return (0); case _PC_MIN_HOLE_SIZE: - *valp = (int)SPA_MINBLOCKSIZE; - return (0); + iosize = vp->v_mount->mnt_stat.f_iosize; + if (vp->v_type == VREG) { + zp = VTOZ(vp); + blksize = zp->z_blksz; + if (zp->z_size <= blksize) + blksize = MAX(blksize, iosize); + *valp = (int)blksize; + return (0); + } + if (vp->v_type == VDIR) { + *valp = (int)iosize; + return (0); + } + return (EINVAL); case _PC_ACL_EXTENDED: #if 0 /* POSIX ACLs are not implemented for ZFS on FreeBSD yet. */ zp = VTOZ(vp); @@ -4210,8 +4223,20 @@ zfs_getpages(struct vnode *vp, vm_page_t *ma, int count, int *rbehind, zfs_vmobject_wlock(object); (void) vm_page_grab_pages(object, OFF_TO_IDX(start), - VM_ALLOC_NORMAL | VM_ALLOC_WAITOK | VM_ALLOC_ZERO, + VM_ALLOC_NORMAL | VM_ALLOC_WAITOK, ma, count); + if (!vm_page_all_valid(ma[count - 1])) { + /* + * Later in this function, we copy DMU data to + * invalid pages only. The last page may not be + * entirely filled though, if the file does not + * end on a page boundary. Therefore, we zero + * that last page here to make sure it does not + * contain garbage after the end of file. + */ + ASSERT(vm_page_none_valid(ma[count - 1])); + vm_page_zero_invalid(ma[count - 1], FALSE); + } zfs_vmobject_wunlock(object); } if (blksz == zp->z_blksz) diff --git a/sys/contrib/openzfs/module/os/linux/zfs/zvol_os.c b/sys/contrib/openzfs/module/os/linux/zfs/zvol_os.c index 967a018640e1..4e66bee7744d 100644 --- a/sys/contrib/openzfs/module/os/linux/zfs/zvol_os.c +++ b/sys/contrib/openzfs/module/os/linux/zfs/zvol_os.c @@ -337,16 +337,14 @@ zvol_discard(zv_request_t *zvr) } /* - * Align the request to volume block boundaries when a secure erase is - * not required. This will prevent dnode_free_range() from zeroing out - * the unaligned parts which is slow (read-modify-write) and useless - * since we are not freeing any space by doing so. + * Align the request to volume block boundaries. This will prevent + * dnode_free_range() from zeroing out the unaligned parts which is + * slow (read-modify-write) and useless since we are not freeing any + * space by doing so. */ - if (!io_is_secure_erase(bio, rq)) { - start = P2ROUNDUP(start, zv->zv_volblocksize); - end = P2ALIGN_TYPED(end, zv->zv_volblocksize, uint64_t); - size = end - start; - } + start = P2ROUNDUP(start, zv->zv_volblocksize); + end = P2ALIGN_TYPED(end, zv->zv_volblocksize, uint64_t); + size = end - start; if (start >= end) goto unlock; @@ -467,6 +465,24 @@ zvol_read_task(void *arg) zv_request_task_free(task); } +/* + * Note: + * + * The kernel uses different enum names for the IO opcode, depending on the + * kernel version ('req_opf', 'req_op'). To sidestep this, use macros rather + * than inline functions for these checks. + */ +/* Should this IO go down the zvol write path? */ +#define ZVOL_OP_IS_WRITE(op) \ + (op == REQ_OP_WRITE || \ + op == REQ_OP_FLUSH || \ + op == REQ_OP_DISCARD) + +/* Is this IO type supported by zvols? */ +#define ZVOL_OP_IS_SUPPORTED(op) (op == REQ_OP_READ || ZVOL_OP_IS_WRITE(op)) + +/* Get the IO opcode */ +#define ZVOL_OP(bio, rq) (bio != NULL ? bio_op(bio) : req_op(rq)) /* * Process a BIO or request @@ -486,27 +502,32 @@ zvol_request_impl(zvol_state_t *zv, struct bio *bio, struct request *rq, uint64_t size = io_size(bio, rq); int rw; - if (rq != NULL) { - /* - * Flush & trim requests go down the zvol_write codepath. Or - * more specifically: - * - * If request is a write, or if it's op_is_sync() and not a - * read, or if it's a flush, or if it's a discard, then send the - * request down the write path. - */ - if (op_is_write(rq->cmd_flags) || - (op_is_sync(rq->cmd_flags) && req_op(rq) != REQ_OP_READ) || - req_op(rq) == REQ_OP_FLUSH || - op_is_discard(rq->cmd_flags)) { - rw = WRITE; - } else { - rw = READ; - } + if (unlikely(!ZVOL_OP_IS_SUPPORTED(ZVOL_OP(bio, rq)))) { + zfs_dbgmsg("Unsupported zvol %s, op=%d, flags=0x%x", + rq != NULL ? "request" : "BIO", + ZVOL_OP(bio, rq), + rq != NULL ? rq->cmd_flags : bio->bi_opf); + ASSERT(ZVOL_OP_IS_SUPPORTED(ZVOL_OP(bio, rq))); + zvol_end_io(bio, rq, SET_ERROR(ENOTSUPP)); + goto out; + } + + if (ZVOL_OP_IS_WRITE(ZVOL_OP(bio, rq))) { + rw = WRITE; } else { - rw = bio_data_dir(bio); + rw = READ; } + /* + * Sanity check + * + * If we're a BIO, check our rw matches the kernel's + * bio_data_dir(bio) rw. We need to check because we support fewer + * IO operations, and want to verify that what we think are reads and + * writes from those operations match what the kernel thinks. + */ + ASSERT(rq != NULL || rw == bio_data_dir(bio)); + if (unlikely(zv->zv_flags & ZVOL_REMOVING)) { zvol_end_io(bio, rq, SET_ERROR(ENXIO)); goto out; @@ -610,7 +631,7 @@ zvol_request_impl(zvol_state_t *zv, struct bio *bio, struct request *rq, * interfaces lack this functionality (they block waiting for * the i/o to complete). */ - if (io_is_discard(bio, rq) || io_is_secure_erase(bio, rq)) { + if (io_is_discard(bio, rq)) { if (force_sync) { zvol_discard(&zvr); } else { diff --git a/sys/contrib/openzfs/module/zfs/arc.c b/sys/contrib/openzfs/module/zfs/arc.c index b677f90280d7..dbb5e942e2e6 100644 --- a/sys/contrib/openzfs/module/zfs/arc.c +++ b/sys/contrib/openzfs/module/zfs/arc.c @@ -1157,7 +1157,7 @@ buf_fini(void) #if defined(_KERNEL) /* * Large allocations which do not require contiguous pages - * should be using vmem_free() in the linux kernel\ + * should be using vmem_free() in the linux kernel. */ vmem_free(buf_hash_table.ht_table, (buf_hash_table.ht_mask + 1) * sizeof (void *)); @@ -4651,10 +4651,10 @@ arc_flush_task(void *arg) arc_flush_impl(spa_guid, B_FALSE); arc_async_flush_remove(spa_guid, af->af_cache_level); - uint64_t elaspsed = NSEC2MSEC(gethrtime() - start_time); - if (elaspsed > 0) { + uint64_t elapsed = NSEC2MSEC(gethrtime() - start_time); + if (elapsed > 0) { zfs_dbgmsg("spa %llu arc flushed in %llu ms", - (u_longlong_t)spa_guid, (u_longlong_t)elaspsed); + (u_longlong_t)spa_guid, (u_longlong_t)elapsed); } } @@ -9152,7 +9152,7 @@ top: if (dev->l2ad_first) { /* * This is the first sweep through the device. There is - * nothing to evict. We have already trimmmed the + * nothing to evict. We have already trimmed the * whole device. */ goto out; @@ -10086,12 +10086,12 @@ l2arc_device_teardown(void *arg) kmem_free(remdev->l2ad_dev_hdr, remdev->l2ad_dev_hdr_asize); vmem_free(remdev, sizeof (l2arc_dev_t)); - uint64_t elaspsed = NSEC2MSEC(gethrtime() - start_time); - if (elaspsed > 0) { + uint64_t elapsed = NSEC2MSEC(gethrtime() - start_time); + if (elapsed > 0) { zfs_dbgmsg("spa %llu, vdev %llu removed in %llu ms", (u_longlong_t)rva->rva_spa_gid, (u_longlong_t)rva->rva_vdev_gid, - (u_longlong_t)elaspsed); + (u_longlong_t)elapsed); } if (rva->rva_async) diff --git a/sys/contrib/openzfs/tests/zfs-tests/cmd/mmap_seek.c b/sys/contrib/openzfs/tests/zfs-tests/cmd/mmap_seek.c index 45ba17e36c35..f46980cad111 100644 --- a/sys/contrib/openzfs/tests/zfs-tests/cmd/mmap_seek.c +++ b/sys/contrib/openzfs/tests/zfs-tests/cmd/mmap_seek.c @@ -47,25 +47,34 @@ #endif static void +seek_expect(int fd, off_t offset, int whence, off_t expect_offset) +{ + errno = 0; + off_t seek_offset = lseek(fd, offset, whence); + if (seek_offset == expect_offset) + return; + + int err = errno; + fprintf(stderr, "lseek(fd, %ld, SEEK_%s) = %ld (expected %ld)", + offset, (whence == SEEK_DATA ? "DATA" : "HOLE"), + seek_offset, expect_offset); + if (err != 0) + fprintf(stderr, " (errno %d [%s])\n", err, strerror(err)); + else + fputc('\n', stderr); + exit(2); +} + +static inline void seek_data(int fd, off_t offset, off_t expected) { - off_t data_offset = lseek(fd, offset, SEEK_DATA); - if (data_offset != expected) { - fprintf(stderr, "lseek(fd, %d, SEEK_DATA) = %d (expected %d)\n", - (int)offset, (int)data_offset, (int)expected); - exit(2); - } + seek_expect(fd, offset, SEEK_DATA, expected); } -static void +static inline void seek_hole(int fd, off_t offset, off_t expected) { - off_t hole_offset = lseek(fd, offset, SEEK_HOLE); - if (hole_offset != expected) { - fprintf(stderr, "lseek(fd, %d, SEEK_HOLE) = %d (expected %d)\n", - (int)offset, (int)hole_offset, (int)expected); - exit(2); - } + seek_expect(fd, offset, SEEK_HOLE, expected); } int diff --git a/sys/contrib/openzfs/tests/zfs-tests/tests/functional/cli_root/zpool_reopen/zpool_reopen_004_pos.ksh b/sys/contrib/openzfs/tests/zfs-tests/tests/functional/cli_root/zpool_reopen/zpool_reopen_004_pos.ksh index 2bd9f616170b..1a9774331ba1 100755 --- a/sys/contrib/openzfs/tests/zfs-tests/tests/functional/cli_root/zpool_reopen/zpool_reopen_004_pos.ksh +++ b/sys/contrib/openzfs/tests/zfs-tests/tests/functional/cli_root/zpool_reopen/zpool_reopen_004_pos.ksh @@ -47,6 +47,8 @@ function cleanup # bring back removed disk online for further tests insert_disk $REMOVED_DISK $scsi_host poolexists $TESTPOOL && destroy_pool $TESTPOOL + # Since the disk was offline during destroy, remove the label + zpool labelclear $DISK2 -f } log_assert "Testing zpool reopen with pool name as argument" diff --git a/sys/contrib/openzfs/tests/zfs-tests/tests/functional/zvol/zvol_misc/zvol_misc_trim.ksh b/sys/contrib/openzfs/tests/zfs-tests/tests/functional/zvol/zvol_misc/zvol_misc_trim.ksh index bb697b76a9f3..9de308c4a11a 100755 --- a/sys/contrib/openzfs/tests/zfs-tests/tests/functional/zvol/zvol_misc/zvol_misc_trim.ksh +++ b/sys/contrib/openzfs/tests/zfs-tests/tests/functional/zvol/zvol_misc/zvol_misc_trim.ksh @@ -41,6 +41,7 @@ # 5. TRIM the first 1MB and last 2MB of the 5MB block of data. # 6. Observe 2MB of used space on the zvol # 7. Verify the trimmed regions are zero'd on the zvol +# 8. Verify Secure Erase does not work on zvols (Linux only) verify_runnable "global" @@ -56,6 +57,7 @@ if is_linux ; then else trimcmd='blkdiscard' fi + secure_trimcmd="$trimcmd --secure" else # By default, FreeBSD 'trim' always does a dry-run. '-f' makes # it perform the actual operation. @@ -114,6 +116,11 @@ function do_test { log_must diff $datafile1 $datafile2 log_must rm $datafile1 $datafile2 + + # Secure erase should not work (Linux check only). + if [ -n "$secure_trimcmd" ] ; then + log_mustnot $secure_trimcmd $zvolpath + fi } log_assert "Verify that a ZFS volume can be TRIMed" diff --git a/sys/dev/aac/aac_linux.c b/sys/dev/aac/aac_linux.c index 609315f50939..65008c562342 100644 --- a/sys/dev/aac/aac_linux.c +++ b/sys/dev/aac/aac_linux.c @@ -52,15 +52,7 @@ #define AAC_LINUX_IOCTL_MIN 0x0000 #define AAC_LINUX_IOCTL_MAX 0x21ff -static linux_ioctl_function_t aac_linux_ioctl; -static struct linux_ioctl_handler aac_linux_handler = {aac_linux_ioctl, - AAC_LINUX_IOCTL_MIN, - AAC_LINUX_IOCTL_MAX}; - -SYSINIT (aac_linux_register, SI_SUB_KLD, SI_ORDER_MIDDLE, - linux_ioctl_register_handler, &aac_linux_handler); -SYSUNINIT(aac_linux_unregister, SI_SUB_KLD, SI_ORDER_MIDDLE, - linux_ioctl_unregister_handler, &aac_linux_handler); +LINUX_IOCTL_SET(aac, AAC_LINUX_IOCTL_MIN, AAC_LINUX_IOCTL_MAX); static int aac_linux_modevent(module_t mod, int type, void *data) diff --git a/sys/dev/aacraid/aacraid_linux.c b/sys/dev/aacraid/aacraid_linux.c index 267dd84c65d5..6b6259ed7059 100644 --- a/sys/dev/aacraid/aacraid_linux.c +++ b/sys/dev/aacraid/aacraid_linux.c @@ -54,15 +54,7 @@ #define AAC_LINUX_IOCTL_MIN 0x0000 #define AAC_LINUX_IOCTL_MAX 0x21ff -static linux_ioctl_function_t aacraid_linux_ioctl; -static struct linux_ioctl_handler aacraid_linux_handler = {aacraid_linux_ioctl, - AAC_LINUX_IOCTL_MIN, - AAC_LINUX_IOCTL_MAX}; - -SYSINIT (aacraid_linux_register, SI_SUB_KLD, SI_ORDER_MIDDLE, - linux_ioctl_register_handler, &aacraid_linux_handler); -SYSUNINIT(aacraid_linux_unregister, SI_SUB_KLD, SI_ORDER_MIDDLE, - linux_ioctl_unregister_handler, &aacraid_linux_handler); +LINUX_IOCTL_SET(aacraid, AAC_LINUX_IOCTL_MIN, AAC_LINUX_IOCTL_MAX); static int aacraid_linux_modevent(module_t mod, int type, void *data) diff --git a/sys/dev/iommu/busdma_iommu.c b/sys/dev/iommu/busdma_iommu.c index 668ccf056463..82f73d469585 100644 --- a/sys/dev/iommu/busdma_iommu.c +++ b/sys/dev/iommu/busdma_iommu.c @@ -295,7 +295,6 @@ iommu_instantiate_ctx(struct iommu_unit *unit, device_t dev, bool rmrr) } else { iommu_free_ctx_locked(unit, ctx); } - ctx = NULL; } return (ctx); } @@ -303,6 +302,7 @@ iommu_instantiate_ctx(struct iommu_unit *unit, device_t dev, bool rmrr) struct iommu_ctx * iommu_get_dev_ctx(device_t dev) { + struct iommu_ctx *ctx; struct iommu_unit *unit; unit = iommu_find(dev, bootverbose); @@ -313,7 +313,10 @@ iommu_get_dev_ctx(device_t dev) return (NULL); iommu_unit_pre_instantiate_ctx(unit); - return (iommu_instantiate_ctx(unit, dev, false)); + ctx = iommu_instantiate_ctx(unit, dev, false); + if (ctx != NULL && (ctx->flags & IOMMU_CTX_DISABLED) != 0) + ctx = NULL; + return (ctx); } bus_dma_tag_t diff --git a/sys/dev/ipmi/ipmi_linux.c b/sys/dev/ipmi/ipmi_linux.c index 05eb30a0aa77..58872de12003 100644 --- a/sys/dev/ipmi/ipmi_linux.c +++ b/sys/dev/ipmi/ipmi_linux.c @@ -66,15 +66,7 @@ #define L_IPMICTL_SET_MY_LUN_CMD _IOW(IPMI_IOC_MAGIC, 19, unsigned int) #define L_IPMICTL_GET_MY_LUN_CMD _IOW(IPMI_IOC_MAGIC, 20, unsigned int) -static linux_ioctl_function_t ipmi_linux_ioctl; -static struct linux_ioctl_handler ipmi_linux_handler = {ipmi_linux_ioctl, - IPMI_LINUX_IOCTL_MIN, - IPMI_LINUX_IOCTL_MAX}; - -SYSINIT (ipmi_linux_register, SI_SUB_KLD, SI_ORDER_MIDDLE, - linux_ioctl_register_handler, &ipmi_linux_handler); -SYSUNINIT(ipmi_linux_unregister, SI_SUB_KLD, SI_ORDER_MIDDLE, - linux_ioctl_unregister_handler, &ipmi_linux_handler); +LINUX_IOCTL_SET(ipmi, IPMI_LINUX_IOCTL_MIN, IPMI_LINUX_IOCTL_MAX); static int ipmi_linux_modevent(module_t mod, int type, void *data) diff --git a/sys/dev/mfi/mfi_linux.c b/sys/dev/mfi/mfi_linux.c index 8ed8baa3858a..9541ff37336a 100644 --- a/sys/dev/mfi/mfi_linux.c +++ b/sys/dev/mfi/mfi_linux.c @@ -53,15 +53,7 @@ #define MFI_LINUX_IOCTL_MIN 0x4d00 #define MFI_LINUX_IOCTL_MAX 0x4d04 -static linux_ioctl_function_t mfi_linux_ioctl; -static struct linux_ioctl_handler mfi_linux_handler = {mfi_linux_ioctl, - MFI_LINUX_IOCTL_MIN, - MFI_LINUX_IOCTL_MAX}; - -SYSINIT (mfi_register, SI_SUB_KLD, SI_ORDER_MIDDLE, - linux_ioctl_register_handler, &mfi_linux_handler); -SYSUNINIT(mfi_unregister, SI_SUB_KLD, SI_ORDER_MIDDLE, - linux_ioctl_unregister_handler, &mfi_linux_handler); +LINUX_IOCTL_SET(mfi, MFI_LINUX_IOCTL_MIN, MFI_LINUX_IOCTL_MAX); static struct linux_device_handler mfi_device_handler = { "mfi", "megaraid_sas", "mfi0", "megaraid_sas_ioctl_node", -1, 0, 1}; diff --git a/sys/dev/mrsas/mrsas_linux.c b/sys/dev/mrsas/mrsas_linux.c index d7d48740a204..b06788fffc82 100644 --- a/sys/dev/mrsas/mrsas_linux.c +++ b/sys/dev/mrsas/mrsas_linux.c @@ -67,15 +67,7 @@ #define MRSAS_LINUX_IOCTL_MIN 0x4d00 #define MRSAS_LINUX_IOCTL_MAX 0x4d01 -static linux_ioctl_function_t mrsas_linux_ioctl; -static struct linux_ioctl_handler mrsas_linux_handler = {mrsas_linux_ioctl, - MRSAS_LINUX_IOCTL_MIN, -MRSAS_LINUX_IOCTL_MAX}; - -SYSINIT(mrsas_register, SI_SUB_KLD, SI_ORDER_MIDDLE, - linux_ioctl_register_handler, &mrsas_linux_handler); -SYSUNINIT(mrsas_unregister, SI_SUB_KLD, SI_ORDER_MIDDLE, - linux_ioctl_unregister_handler, &mrsas_linux_handler); +LINUX_IOCTL_SET(mrsas, MRSAS_LINUX_IOCTL_MIN, MRSAS_LINUX_IOCTL_MAX); static struct linux_device_handler mrsas_device_handler = {"mrsas", "megaraid_sas", "mrsas0", "megaraid_sas_ioctl_node", -1, 0, 1}; diff --git a/sys/dev/nvme/nvme_ctrlr.c b/sys/dev/nvme/nvme_ctrlr.c index 3a1894bf754d..f212759a5500 100644 --- a/sys/dev/nvme/nvme_ctrlr.c +++ b/sys/dev/nvme/nvme_ctrlr.c @@ -33,6 +33,7 @@ #include <sys/buf.h> #include <sys/bus.h> #include <sys/conf.h> +#include <sys/disk.h> #include <sys/ioccom.h> #include <sys/proc.h> #include <sys/smp.h> @@ -1254,6 +1255,24 @@ nvme_ctrlr_poll(struct nvme_controller *ctrlr) } /* + * Copy the NVME device's serial number to the provided buffer, which must be + * at least DISK_IDENT_SIZE bytes large. + */ +void +nvme_ctrlr_get_ident(const struct nvme_controller *ctrlr, uint8_t *sn) +{ + _Static_assert(NVME_SERIAL_NUMBER_LENGTH < DISK_IDENT_SIZE, + "NVME serial number too big for disk ident"); + + memmove(sn, ctrlr->cdata.sn, NVME_SERIAL_NUMBER_LENGTH); + sn[NVME_SERIAL_NUMBER_LENGTH] = '\0'; + for (int i = 0; sn[i] != '\0'; i++) { + if (sn[i] < 0x20 || sn[i] >= 0x80) + sn[i] = ' '; + } +} + +/* * Poll the single-vector interrupt case: num_io_queues will be 1 and * there's only a single vector. While we're polling, we mask further * interrupts in the controller. @@ -1495,6 +1514,11 @@ nvme_ctrlr_ioctl(struct cdev *cdev, u_long cmd, caddr_t arg, int flag, case NVME_GET_CONTROLLER_DATA: memcpy(arg, &ctrlr->cdata, sizeof(ctrlr->cdata)); break; + case DIOCGIDENT: { + uint8_t *sn = arg; + nvme_ctrlr_get_ident(ctrlr, sn); + break; + } /* Linux Compatible (see nvme_linux.h) */ case NVME_IOCTL_ID: td->td_retval[0] = 0xfffffffful; diff --git a/sys/dev/nvme/nvme_ns.c b/sys/dev/nvme/nvme_ns.c index e84d2066930e..a759181a8c16 100644 --- a/sys/dev/nvme/nvme_ns.c +++ b/sys/dev/nvme/nvme_ns.c @@ -88,6 +88,11 @@ nvme_ns_ioctl(struct cdev *cdev, u_long cmd, caddr_t arg, int flag, gnsid->nsid = ns->id; break; } + case DIOCGIDENT: { + uint8_t *sn = arg; + nvme_ctrlr_get_ident(ctrlr, sn); + break; + } case DIOCGMEDIASIZE: *(off_t *)arg = (off_t)nvme_ns_get_size(ns); break; diff --git a/sys/dev/nvme/nvme_private.h b/sys/dev/nvme/nvme_private.h index 52e9fcbbebcd..04a47d799350 100644 --- a/sys/dev/nvme/nvme_private.h +++ b/sys/dev/nvme/nvme_private.h @@ -563,6 +563,7 @@ void nvme_notify_new_controller(struct nvme_controller *ctrlr); void nvme_notify_ns(struct nvme_controller *ctrlr, int nsid); void nvme_ctrlr_shared_handler(void *arg); +void nvme_ctrlr_get_ident(const struct nvme_controller *ctrlr, uint8_t *sn); void nvme_ctrlr_poll(struct nvme_controller *ctrlr); int nvme_ctrlr_suspend(struct nvme_controller *ctrlr); diff --git a/sys/dev/random/fenestrasX/fx_pool.c b/sys/dev/random/fenestrasX/fx_pool.c index f4ad1e295d54..ec59b97a2070 100644 --- a/sys/dev/random/fenestrasX/fx_pool.c +++ b/sys/dev/random/fenestrasX/fx_pool.c @@ -167,9 +167,6 @@ static const struct fxrng_ent_char { [RANDOM_RANDOMDEV] = { .entc_cls = &fxrng_lo_push, }, - [RANDOM_PURE_OCTEON] = { - .entc_cls = &fxrng_hi_push, /* Could be made pull. */ - }, [RANDOM_PURE_SAFE] = { .entc_cls = &fxrng_hi_push, }, diff --git a/sys/dev/random/ivy.c b/sys/dev/random/ivy.c index fa1e4831f1b9..3eb0f261e6dc 100644 --- a/sys/dev/random/ivy.c +++ b/sys/dev/random/ivy.c @@ -1,6 +1,6 @@ /*- + * Copyright (c) 2013, 2025, David E. O'Brien <deo@NUXI.org> * Copyright (c) 2013 The FreeBSD Foundation - * Copyright (c) 2013 David E. O'Brien <obrien@NUXI.org> * Copyright (c) 2012 Konstantin Belousov <kib@FreeBSD.org> * All rights reserved. * @@ -48,7 +48,6 @@ #define RETRY_COUNT 10 -static bool has_rdrand, has_rdseed; static u_int random_ivy_read(void *, u_int); static const struct random_source random_ivy = { @@ -57,13 +56,7 @@ static const struct random_source random_ivy = { .rs_read = random_ivy_read }; -SYSCTL_NODE(_kern_random, OID_AUTO, rdrand, CTLFLAG_RW | CTLFLAG_MPSAFE, 0, - "rdrand (ivy) entropy source"); static bool acquire_independent_seed_samples = false; -SYSCTL_BOOL(_kern_random_rdrand, OID_AUTO, rdrand_independent_seed, - CTLFLAG_RWTUN, &acquire_independent_seed_samples, 0, - "If non-zero, use more expensive and slow, but safer, seeded samples " - "where RDSEED is not present."); static bool x86_rdrand_store(u_long *buf) @@ -99,45 +92,6 @@ x86_rdrand_store(u_long *buf) return (true); } -static bool -x86_rdseed_store(u_long *buf) -{ - u_long rndval; - int retry; - - retry = RETRY_COUNT; - __asm __volatile( - "1:\n\t" - "rdseed %1\n\t" /* read randomness into rndval */ - "jc 2f\n\t" /* CF is set on success, exit retry loop */ - "dec %0\n\t" /* otherwise, retry-- */ - "jne 1b\n\t" /* and loop if retries are not exhausted */ - "2:" - : "+r" (retry), "=r" (rndval) : : "cc"); - *buf = rndval; - return (retry != 0); -} - -static bool -x86_unimpl_store(u_long *buf __unused) -{ - - panic("%s called", __func__); -} - -DEFINE_IFUNC(static, bool, x86_rng_store, (u_long *buf)) -{ - has_rdrand = (cpu_feature2 & CPUID2_RDRAND); - has_rdseed = (cpu_stdext_feature & CPUID_STDEXT_RDSEED); - - if (has_rdseed) - return (x86_rdseed_store); - else if (has_rdrand) - return (x86_rdrand_store); - else - return (x86_unimpl_store); -} - /* It is required that buf length is a multiple of sizeof(u_long). */ static u_int random_ivy_read(void *buf, u_int c) @@ -148,7 +102,7 @@ random_ivy_read(void *buf, u_int c) KASSERT(c % sizeof(*b) == 0, ("partial read %d", c)); b = buf; for (count = c; count > 0; count -= sizeof(*b)) { - if (!x86_rng_store(&rndval)) + if (!x86_rdrand_store(&rndval)) break; *b++ = rndval; } @@ -158,18 +112,33 @@ random_ivy_read(void *buf, u_int c) static int rdrand_modevent(module_t mod, int type, void *unused) { + struct sysctl_ctx_list ctx; + struct sysctl_oid *o; + bool has_rdrand, has_rdseed; int error = 0; + has_rdrand = (cpu_feature2 & CPUID2_RDRAND); + has_rdseed = (cpu_stdext_feature & CPUID_STDEXT_RDSEED); + switch (type) { case MOD_LOAD: - if (has_rdrand || has_rdseed) { + if (has_rdrand && !has_rdseed) { + sysctl_ctx_init(&ctx); + o = SYSCTL_ADD_NODE(&ctx, SYSCTL_STATIC_CHILDREN(_kern_random), + OID_AUTO, "rdrand", CTLFLAG_RW | CTLFLAG_MPSAFE, 0, + "rdrand (ivy) entropy source"); + SYSCTL_ADD_BOOL(&ctx, SYSCTL_CHILDREN(o), OID_AUTO, + "rdrand_independent_seed", CTLFLAG_RDTUN, + &acquire_independent_seed_samples, 0, + "If non-zero, use more expensive and slow, but safer, seeded samples " + "where RDSEED is not present."); random_source_register(&random_ivy); printf("random: fast provider: \"%s\"\n", random_ivy.rs_ident); } break; case MOD_UNLOAD: - if (has_rdrand || has_rdseed) + if (has_rdrand && !has_rdseed) random_source_deregister(&random_ivy); break; diff --git a/sys/dev/random/random_harvestq.c b/sys/dev/random/random_harvestq.c index 2d7af254c52c..e38fd38c310b 100644 --- a/sys/dev/random/random_harvestq.c +++ b/sys/dev/random/random_harvestq.c @@ -661,11 +661,11 @@ static const char *random_source_descr[ENTROPYSOURCE] = { [RANDOM_UMA] = "UMA", [RANDOM_CALLOUT] = "CALLOUT", [RANDOM_RANDOMDEV] = "RANDOMDEV", /* ENVIRONMENTAL_END */ - [RANDOM_PURE_OCTEON] = "PURE_OCTEON", /* PURE_START */ - [RANDOM_PURE_SAFE] = "PURE_SAFE", + [RANDOM_PURE_SAFE] = "PURE_SAFE", /* PURE_START */ [RANDOM_PURE_GLXSB] = "PURE_GLXSB", [RANDOM_PURE_HIFN] = "PURE_HIFN", [RANDOM_PURE_RDRAND] = "PURE_RDRAND", + [RANDOM_PURE_RDSEED] = "PURE_RDSEED", [RANDOM_PURE_NEHEMIAH] = "PURE_NEHEMIAH", [RANDOM_PURE_RNDTEST] = "PURE_RNDTEST", [RANDOM_PURE_VIRTIO] = "PURE_VIRTIO", diff --git a/sys/dev/random/rdseed.c b/sys/dev/random/rdseed.c new file mode 100644 index 000000000000..af084aab4ed9 --- /dev/null +++ b/sys/dev/random/rdseed.c @@ -0,0 +1,169 @@ +/*- + * Copyright (c) 2013, 2025, David E. O'Brien <deo@NUXI.org> + * Copyright (c) 2013 The FreeBSD Foundation + * Copyright (c) 2012 Konstantin Belousov <kib@FreeBSD.org> + * All rights reserved. + * + * Portions of this software were developed by Konstantin Belousov + * under sponsorship from the FreeBSD Foundation. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer + * in this position and unchanged. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR + * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES + * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. + * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, + * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT + * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF + * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + */ + +#include <sys/param.h> +#include <sys/kernel.h> +#include <sys/conf.h> +#include <sys/lock.h> +#include <sys/malloc.h> +#include <sys/module.h> +#include <sys/random.h> +#include <sys/sysctl.h> +#include <sys/systm.h> + +#include <machine/md_var.h> +#include <machine/specialreg.h> +#include <x86/ifunc.h> + +#include <dev/random/randomdev.h> + +#define RETRY_COUNT 10 + +static u_int random_rdseed_read(void *, u_int); + +static struct random_source random_rdseed = { + .rs_ident = "Intel Secure Key Seed", + .rs_source = RANDOM_PURE_RDSEED, + .rs_read = random_rdseed_read +}; + +SYSCTL_NODE(_kern_random, OID_AUTO, rdseed, CTLFLAG_RW, 0, + "rdseed (x86) entropy source"); +/* XXX: kern.random.rdseed.enabled=0 also disables RDRAND */ +static bool enabled = true; +SYSCTL_BOOL(_kern_random_rdseed, OID_AUTO, enabled, CTLFLAG_RDTUN, &enabled, 0, + "If zero, disable the use of RDSEED."); + +static bool +x86_rdseed_store(u_long *buf) +{ + u_long rndval; + int retry; + + retry = RETRY_COUNT; + __asm __volatile( + "1:\n\t" + "rdseed %1\n\t" /* read randomness into rndval */ + "jc 2f\n\t" /* CF is set on success, exit retry loop */ + "dec %0\n\t" /* otherwise, retry-- */ + "jne 1b\n\t" /* and loop if retries are not exhausted */ + "2:" + : "+r" (retry), "=r" (rndval) : : "cc"); + *buf = rndval; + return (retry != 0); +} + +/* It is required that buf length is a multiple of sizeof(u_long). */ +static u_int +random_rdseed_read(void *buf, u_int c) +{ + u_long *b, rndval; + u_int count; + + KASSERT(c % sizeof(*b) == 0, ("partial read %d", c)); + b = buf; + for (count = c; count > 0; count -= sizeof(*b)) { + if (!x86_rdseed_store(&rndval)) + break; + *b++ = rndval; + } + return (c - count); +} + +static int +rdseed_modevent(module_t mod, int type, void *unused) +{ + bool has_rdseed; + int error = 0; + + has_rdseed = (cpu_stdext_feature & CPUID_STDEXT_RDSEED); + + switch (type) { + case MOD_LOAD: + if (has_rdseed && enabled) { + random_source_register(&random_rdseed); + printf("random: fast provider: \"%s\"\n", random_rdseed.rs_ident); + } + break; + + case MOD_UNLOAD: + if (has_rdseed) + random_source_deregister(&random_rdseed); + break; + + case MOD_SHUTDOWN: + break; + + default: + error = EOPNOTSUPP; + break; + + } + + return (error); +} + +static moduledata_t rdseed_mod = { + "rdseed", + rdseed_modevent, + 0 +}; + +DECLARE_MODULE(rdseed, rdseed_mod, SI_SUB_RANDOM, SI_ORDER_FOURTH); +MODULE_VERSION(rdseed, 1); +MODULE_DEPEND(rdseed, random_harvestq, 1, 1, 1); + +/* + * Intel's RDSEED Entropy Assessment Report min-entropy claim is 0.6 Shannons + * per bit of data output. Rrefer to the following Entropy Source Validation + * (ESV) certificates: + * + * E#87: Junos OS Physical Entropy Source - Broadwell EP 10-Core Die + * Broadwell-EP-10 FCLGA2011 Intel(R) Xeon(R) E5-2620 V4 Processor + * https://csrc.nist.gov/projects/cryptographic-module-validation-program/entropy-validations/certificate/87 + * (URLs below omitted for brevity but follow same format.) + * + * E#121: Junos OS Physical Entropy Source - Intel Atom C3000 Series + * (Denverton) 16 Core Die with FCBGA1310 Package + * + * E#122: Junos OS Physical Entropy Source - Intel Xeon D-1500 Family + * (Broadwell) 8 Core Die with FCBGA1667 Package + * + * E#123: Junos OS Physical Entropy Source - Intel Xeon D-2100 Series + * (Skylake) 18 Core Die with FCBGA2518 Package + * + * E#141: Junos OS Physical Entropy Source - Intel Xeon D-10 Series + * (Ice Lake-D-10) Die with FCBGA2227 Package + * + * E#169: Junos OS Physical Entropy Source - Intel Xeon AWS-1000 v4 and + * E5 v4 (Broadwell EP) 15 Core Die with FCLGA2011 Package + */ diff --git a/sys/dev/sound/dummy.c b/sys/dev/sound/dummy.c index 4df5b112d3f4..1f2d69708eec 100644 --- a/sys/dev/sound/dummy.c +++ b/sys/dev/sound/dummy.c @@ -346,6 +346,12 @@ dummy_attach(device_t dev) return (ENXIO); mixer_init(dev, &dummy_mixer_class, sc); + /* + * Create an alias so that tests do not need to guess which one is the + * dummy device if there are more devices present in the system. + */ + make_dev_alias(sc->info.dsp_dev, "dsp.dummy"); + return (0); } diff --git a/sys/dev/sound/pci/hda/hdac.c b/sys/dev/sound/pci/hda/hdac.c index 80028063bb0d..8a325c538b9b 100644 --- a/sys/dev/sound/pci/hda/hdac.c +++ b/sys/dev/sound/pci/hda/hdac.c @@ -170,6 +170,7 @@ static const struct { { HDA_NVIDIA_GF119, "NVIDIA GF119", 0, 0 }, { HDA_NVIDIA_GF110_1, "NVIDIA GF110", 0, HDAC_QUIRK_MSI }, { HDA_NVIDIA_GF110_2, "NVIDIA GF110", 0, HDAC_QUIRK_MSI }, + { HDA_ATI_RAVEN, "ATI Raven", 0, 0 }, { HDA_ATI_SB450, "ATI SB450", 0, 0 }, { HDA_ATI_SB600, "ATI SB600", 0, 0 }, { HDA_ATI_RS600, "ATI RS600", 0, 0 }, diff --git a/sys/dev/sound/pci/hda/hdac.h b/sys/dev/sound/pci/hda/hdac.h index c11e6b2d6810..8fb54108a833 100644 --- a/sys/dev/sound/pci/hda/hdac.h +++ b/sys/dev/sound/pci/hda/hdac.h @@ -154,6 +154,7 @@ /* ATI */ #define ATI_VENDORID 0x1002 +#define HDA_ATI_RAVEN HDA_MODEL_CONSTRUCT(ATI, 0x15de) #define HDA_ATI_SB450 HDA_MODEL_CONSTRUCT(ATI, 0x437b) #define HDA_ATI_SB600 HDA_MODEL_CONSTRUCT(ATI, 0x4383) #define HDA_ATI_RS600 HDA_MODEL_CONSTRUCT(ATI, 0x793b) diff --git a/sys/dev/sound/pcm/dsp.c b/sys/dev/sound/pcm/dsp.c index fe5576baf017..27d5b740b90b 100644 --- a/sys/dev/sound/pcm/dsp.c +++ b/sys/dev/sound/pcm/dsp.c @@ -83,15 +83,15 @@ static d_mmap_t dsp_mmap; static d_mmap_single_t dsp_mmap_single; struct cdevsw dsp_cdevsw = { - .d_version = D_VERSION, - .d_open = dsp_open, - .d_read = dsp_read, - .d_write = dsp_write, - .d_ioctl = dsp_ioctl, - .d_poll = dsp_poll, - .d_mmap = dsp_mmap, - .d_mmap_single = dsp_mmap_single, - .d_name = "dsp", + .d_version = D_VERSION, + .d_open = dsp_open, + .d_read = dsp_read, + .d_write = dsp_write, + .d_ioctl = dsp_ioctl, + .d_poll = dsp_poll, + .d_mmap = dsp_mmap, + .d_mmap_single = dsp_mmap_single, + .d_name = "dsp", }; static eventhandler_tag dsp_ehtag = NULL; diff --git a/sys/dev/tdfx/tdfx_linux.c b/sys/dev/tdfx/tdfx_linux.c index f3410106bad2..777144d21bb6 100644 --- a/sys/dev/tdfx/tdfx_linux.c +++ b/sys/dev/tdfx/tdfx_linux.c @@ -42,7 +42,7 @@ LINUX_IOCTL_SET(tdfx, LINUX_IOCTL_TDFX_MIN, LINUX_IOCTL_TDFX_MAX); * Linux emulation IOCTL for /dev/tdfx */ static int -linux_ioctl_tdfx(struct thread *td, struct linux_ioctl_args* args) +tdfx_linux_ioctl(struct thread *td, struct linux_ioctl_args* args) { cap_rights_t rights; int error = 0; diff --git a/sys/dev/tdfx/tdfx_linux.h b/sys/dev/tdfx/tdfx_linux.h index b87cb41f38fe..9d012c12274b 100644 --- a/sys/dev/tdfx/tdfx_linux.h +++ b/sys/dev/tdfx/tdfx_linux.h @@ -35,18 +35,6 @@ #include <machine/../linux/linux_proto.h> #include <compat/linux/linux_ioctl.h> -/* - * This code was donated by Vladimir N. Silynaev to allow for defining - * ioctls within modules - */ -#define LINUX_IOCTL_SET(n,low,high) \ -static linux_ioctl_function_t linux_ioctl_##n; \ -static struct linux_ioctl_handler n##_handler = {linux_ioctl_##n, low, high}; \ -SYSINIT(n##register, SI_SUB_KLD, SI_ORDER_MIDDLE,\ -linux_ioctl_register_handler, &n##_handler); \ -SYSUNINIT(n##unregister, SI_SUB_KLD, SI_ORDER_MIDDLE,\ -linux_ioctl_unregister_handler, &n##_handler); - /* Values for /dev/3dfx */ /* Query IOCTLs */ #define LINUX_IOCTL_TDFX_QUERY_BOARDS 0x3302 diff --git a/sys/dev/usb/controller/ehci_pci.c b/sys/dev/usb/controller/ehci_pci.c index d7298ab89df7..9550002e3b70 100644 --- a/sys/dev/usb/controller/ehci_pci.c +++ b/sys/dev/usb/controller/ehci_pci.c @@ -88,6 +88,7 @@ #define PCI_EHCI_VENDORID_NEC 0x1033 #define PCI_EHCI_VENDORID_OPTI 0x1045 #define PCI_EHCI_VENDORID_PHILIPS 0x1131 +#define PCI_EHCI_VENDORID_REALTEK 0x10ec #define PCI_EHCI_VENDORID_SIS 0x1039 #define PCI_EHCI_VENDORID_NVIDIA 0x12D2 #define PCI_EHCI_VENDORID_NVIDIA2 0x10DE @@ -218,6 +219,9 @@ ehci_pci_match(device_t self) case 0x15621131: return "Philips ISP156x USB 2.0 controller"; + case 0x816d10ec: + return ("Realtek RTL811x USB 2.0 controller"); + case 0x70021039: return "SiS 968 USB 2.0 controller"; @@ -402,6 +406,9 @@ ehci_pci_attach(device_t self) case PCI_EHCI_VENDORID_PHILIPS: sprintf(sc->sc_vendor, "Philips"); break; + case PCI_EHCI_VENDORID_REALTEK: + sprintf(sc->sc_vendor, "Realtek"); + break; case PCI_EHCI_VENDORID_SIS: sprintf(sc->sc_vendor, "SiS"); break; diff --git a/sys/dev/vmm/vmm_dev.c b/sys/dev/vmm/vmm_dev.c index 4961b21180e1..ebbceb25b69e 100644 --- a/sys/dev/vmm/vmm_dev.c +++ b/sys/dev/vmm/vmm_dev.c @@ -14,6 +14,7 @@ #include <sys/kernel.h> #include <sys/malloc.h> #include <sys/mman.h> +#include <sys/priv.h> #include <sys/proc.h> #include <sys/queue.h> #include <sys/sx.h> @@ -470,6 +471,12 @@ vmmdev_ioctl(struct cdev *cdev, u_long cmd, caddr_t data, int fflag, if (ioctl == NULL) return (ENOTTY); + if ((ioctl->flags & VMMDEV_IOCTL_PRIV_CHECK_DRIVER) != 0) { + error = priv_check(td, PRIV_DRIVER); + if (error != 0) + return (error); + } + if ((ioctl->flags & VMMDEV_IOCTL_XLOCK_MEMSEGS) != 0) vm_xlock_memsegs(sc->vm); else if ((ioctl->flags & VMMDEV_IOCTL_SLOCK_MEMSEGS) != 0) @@ -656,10 +663,10 @@ vmmdev_ioctl(struct cdev *cdev, u_long cmd, caddr_t data, int fflag, error = EINVAL; break; } - regvals = malloc(sizeof(regvals[0]) * vmregset->count, M_VMMDEV, - M_WAITOK); - regnums = malloc(sizeof(regnums[0]) * vmregset->count, M_VMMDEV, - M_WAITOK); + regvals = mallocarray(vmregset->count, sizeof(regvals[0]), + M_VMMDEV, M_WAITOK); + regnums = mallocarray(vmregset->count, sizeof(regnums[0]), + M_VMMDEV, M_WAITOK); error = copyin(vmregset->regnums, regnums, sizeof(regnums[0]) * vmregset->count); if (error == 0) @@ -682,10 +689,10 @@ vmmdev_ioctl(struct cdev *cdev, u_long cmd, caddr_t data, int fflag, error = EINVAL; break; } - regvals = malloc(sizeof(regvals[0]) * vmregset->count, M_VMMDEV, - M_WAITOK); - regnums = malloc(sizeof(regnums[0]) * vmregset->count, M_VMMDEV, - M_WAITOK); + regvals = mallocarray(vmregset->count, sizeof(regvals[0]), + M_VMMDEV, M_WAITOK); + regnums = mallocarray(vmregset->count, sizeof(regnums[0]), + M_VMMDEV, M_WAITOK); error = copyin(vmregset->regnums, regnums, sizeof(regnums[0]) * vmregset->count); if (error == 0) diff --git a/sys/dev/vmm/vmm_dev.h b/sys/dev/vmm/vmm_dev.h index 410066c49cf2..2881a7063565 100644 --- a/sys/dev/vmm/vmm_dev.h +++ b/sys/dev/vmm/vmm_dev.h @@ -44,6 +44,7 @@ struct vmmdev_ioctl { #define VMMDEV_IOCTL_LOCK_ALL_VCPUS 0x08 #define VMMDEV_IOCTL_ALLOC_VCPU 0x10 #define VMMDEV_IOCTL_MAYBE_ALLOC_VCPU 0x20 +#define VMMDEV_IOCTL_PRIV_CHECK_DRIVER 0x40 int flags; }; diff --git a/sys/kern/kern_jail.c b/sys/kern/kern_jail.c index 3697d95fe0e5..267b60ffb5bc 100644 --- a/sys/kern/kern_jail.c +++ b/sys/kern/kern_jail.c @@ -1088,6 +1088,7 @@ kern_jail_set(struct thread *td, struct uio *optuio, int flags) else { if (!(flags & (JAIL_USE_DESC | JAIL_AT_DESC | JAIL_GET_DESC | JAIL_OWN_DESC))) { + error = EINVAL; vfs_opterror(opts, "unexpected desc"); goto done_errmsg; } @@ -2518,6 +2519,7 @@ kern_jail_get(struct thread *td, struct uio *optuio, int flags) } else if (error == 0) { if (!(flags & (JAIL_USE_DESC | JAIL_AT_DESC | JAIL_GET_DESC | JAIL_OWN_DESC))) { + error = EINVAL; vfs_opterror(opts, "unexpected desc"); goto done; } @@ -2909,12 +2911,6 @@ prison_remove(struct prison *pr) { sx_assert(&allprison_lock, SA_XLOCKED); mtx_assert(&pr->pr_mtx, MA_OWNED); - if (!prison_isalive(pr)) { - /* Silently ignore already-dying prisons. */ - mtx_unlock(&pr->pr_mtx); - sx_xunlock(&allprison_lock); - return; - } prison_deref(pr, PD_KILL | PD_DEREF | PD_LOCKED | PD_LIST_XLOCKED); } @@ -3461,12 +3457,17 @@ prison_deref(struct prison *pr, int flags) /* Kill the prison and its descendents. */ KASSERT(pr != &prison0, ("prison_deref trying to kill prison0")); - if (!(flags & PD_DEREF)) { - prison_hold(pr); - flags |= PD_DEREF; + if (!prison_isalive(pr)) { + /* Silently ignore already-dying prisons. */ + flags &= ~PD_KILL; + } else { + if (!(flags & PD_DEREF)) { + prison_hold(pr); + flags |= PD_DEREF; + } + flags = prison_lock_xlock(pr, flags); + prison_deref_kill(pr, &freeprison); } - flags = prison_lock_xlock(pr, flags); - prison_deref_kill(pr, &freeprison); } if (flags & PD_DEUREF) { /* Drop a user reference. */ diff --git a/sys/kern/subr_kdb.c b/sys/kern/subr_kdb.c index 56264a96c9fa..909dd10a6e69 100644 --- a/sys/kern/subr_kdb.c +++ b/sys/kern/subr_kdb.c @@ -330,7 +330,7 @@ kdb_reboot(void) #define KEY_CRTLP 16 /* ^P */ #define KEY_CRTLR 18 /* ^R */ -/* States of th KDB "alternate break sequence" detecting state machine. */ +/* States of the KDB "alternate break sequence" detecting state machine. */ enum { KDB_ALT_BREAK_SEEN_NONE, KDB_ALT_BREAK_SEEN_CR, diff --git a/sys/kern/vfs_cluster.c b/sys/kern/vfs_cluster.c index 2e397b8e9e8f..b674313993c4 100644 --- a/sys/kern/vfs_cluster.c +++ b/sys/kern/vfs_cluster.c @@ -260,8 +260,10 @@ cluster_read(struct vnode *vp, u_quad_t filesize, daddr_t lblkno, long size, */ while (lblkno < (origblkno + maxra)) { error = VOP_BMAP(vp, lblkno, NULL, &blkno, &ncontig, NULL); - if (error) + if (error) { + error = 0; break; + } if (blkno == -1) break; diff --git a/sys/modules/Makefile b/sys/modules/Makefile index 63a0b3260e6d..9bc743c0c6d1 100644 --- a/sys/modules/Makefile +++ b/sys/modules/Makefile @@ -173,6 +173,7 @@ SUBDIR= \ iflib \ ${_igc} \ imgact_binmisc \ + ${_imx} \ ${_intelspi} \ ${_io} \ ${_ioat} \ @@ -346,6 +347,7 @@ SUBDIR= \ rc4 \ ${_rdma} \ ${_rdrand_rng} \ + ${_rdseed_rng} \ re \ rl \ ${_rockchip} \ @@ -680,7 +682,9 @@ _irdma= irdma .if ${MACHINE_CPUARCH} == "aarch64" || ${MACHINE_CPUARCH} == "arm" || \ ${MACHINE_CPUARCH} == "riscv" .if !empty(OPT_FDT) +_allwinner= allwinner _if_cgem= if_cgem +_sdhci_fdt= sdhci_fdt .endif .endif @@ -713,7 +717,6 @@ _hyperv= hyperv _vf_i2c= vf_i2c .if !empty(OPT_FDT) -_allwinner= allwinner _dwwdt= dwwdt _enetc= enetc _felix= felix @@ -721,12 +724,8 @@ _rockchip= rockchip .endif .endif -.if ${MACHINE_CPUARCH} == "aarch64" || ${MACHINE_CPUARCH} == "arm" || \ - ${MACHINE_CPUARCH} == "riscv" -.if !empty(OPT_FDT) -_sdhci_fdt= sdhci_fdt -.endif -_neta= neta +.if ${MACHINE_CPUARCH} == "arm" +_imx= imx .endif .if ${MACHINE_CPUARCH} == "i386" || ${MACHINE_CPUARCH} == "amd64" @@ -822,6 +821,7 @@ _nvram= nvram _padlock= padlock _padlock_rng= padlock_rng _rdrand_rng= rdrand_rng +_rdseed_rng= rdseed_rng .endif _pchtherm = pchtherm _s3= s3 @@ -923,7 +923,8 @@ _nvram+= powermac_nvram .if ${MACHINE_CPUARCH} == "arm" || ${MACHINE_CPUARCH} == "aarch64" _bcm283x_clkman= bcm283x_clkman -_bcm283x_pwm= bcm283x_pwm +_bcm283x_pwm= bcm283x_pwm +_neta= neta .endif .if ${MACHINE_CPUARCH} == "amd64" diff --git a/sys/modules/allwinner/aw_sid/Makefile b/sys/modules/allwinner/aw_sid/Makefile index 7d3dad68acb9..2679aa83d09d 100644 --- a/sys/modules/allwinner/aw_sid/Makefile +++ b/sys/modules/allwinner/aw_sid/Makefile @@ -7,6 +7,7 @@ SRCS+= \ bus_if.h \ clknode_if.h \ device_if.h \ - ofw_bus_if.h \ + nvmem_if.h \ + ofw_bus_if.h .include <bsd.kmod.mk> diff --git a/sys/modules/allwinner/aw_thermal/Makefile b/sys/modules/allwinner/aw_thermal/Makefile index 911e6406d947..924b09a3b3c8 100644 --- a/sys/modules/allwinner/aw_thermal/Makefile +++ b/sys/modules/allwinner/aw_thermal/Makefile @@ -7,6 +7,7 @@ SRCS+= \ bus_if.h \ clknode_if.h \ device_if.h \ - ofw_bus_if.h \ + nvmem_if.h \ + ofw_bus_if.h .include <bsd.kmod.mk> diff --git a/sys/modules/dtb/allwinner/Makefile b/sys/modules/dtb/allwinner/Makefile index 242ee5d974ad..2666e786a9df 100644 --- a/sys/modules/dtb/allwinner/Makefile +++ b/sys/modules/dtb/allwinner/Makefile @@ -65,7 +65,12 @@ DTSO= sun50i-a64-mmc0-disable.dtso \ sun50i-a64-timer.dtso \ sun50i-h5-opp.dtso \ sun50i-h5-nanopi-neo2-opp.dtso - +.elif ${MACHINE_CPUARCH} == "riscv" +DTS= \ + allwinner/sun20i-d1-dongshan-nezha-stu.dts \ + allwinner/sun20i-d1-lichee-rv.dts \ + allwinner/sun20i-d1-mangopi-mq-pro.dts \ + allwinner/sun20i-d1-nezha.dts .endif .include <bsd.dtb.mk> diff --git a/sys/modules/dtb/starfive/Makefile b/sys/modules/dtb/starfive/Makefile new file mode 100644 index 000000000000..2da30f0985c7 --- /dev/null +++ b/sys/modules/dtb/starfive/Makefile @@ -0,0 +1,7 @@ +DTS= \ + starfive/jh7110-pine64-star64.dts \ + starfive/jh7110-milkv-mars.dts \ + starfive/jh7110-starfive-visionfive-2-v1.3b.dts \ + starfive/jh7110-starfive-visionfive-2-v1.2a.dts + +.include <bsd.dtb.mk> diff --git a/sys/modules/rdrand_rng/Makefile b/sys/modules/rdrand_rng/Makefile index 7fa7a8bb8fb9..496fc863033f 100644 --- a/sys/modules/rdrand_rng/Makefile +++ b/sys/modules/rdrand_rng/Makefile @@ -6,9 +6,4 @@ SRCS+= bus_if.h device_if.h CFLAGS+= -I${SRCTOP}/sys -# ld.bfd doesn't support ifuncs invoked non-PIC -.if ${MACHINE_CPUARCH} == "i386" -CFLAGS.gcc= -fPIC -.endif - .include <bsd.kmod.mk> diff --git a/sys/modules/rdseed_rng/Makefile b/sys/modules/rdseed_rng/Makefile new file mode 100644 index 000000000000..6593505546dd --- /dev/null +++ b/sys/modules/rdseed_rng/Makefile @@ -0,0 +1,9 @@ +.PATH: ${SRCTOP}/sys/dev/random + +KMOD= rdseed_rng +SRCS= rdseed.c +SRCS+= bus_if.h device_if.h + +CFLAGS+= -I${SRCTOP}/sys + +.include <bsd.kmod.mk> diff --git a/sys/modules/zfs/zfs_config.h b/sys/modules/zfs/zfs_config.h index db1b6f33a8ef..3a17ed289235 100644 --- a/sys/modules/zfs/zfs_config.h +++ b/sys/modules/zfs/zfs_config.h @@ -843,7 +843,7 @@ /* #undef ZFS_DEVICE_MINOR */ /* Define the project alias string. */ -#define ZFS_META_ALIAS "zfs-2.4.99-95-FreeBSD_g5605a6d79" +#define ZFS_META_ALIAS "zfs-2.4.99-113-FreeBSD_g6ae99d269" /* Define the project author. */ #define ZFS_META_AUTHOR "OpenZFS" @@ -873,7 +873,7 @@ #define ZFS_META_NAME "zfs" /* Define the project release. */ -#define ZFS_META_RELEASE "95-FreeBSD_g5605a6d79" +#define ZFS_META_RELEASE "113-FreeBSD_g6ae99d269" /* Define the project version. */ #define ZFS_META_VERSION "2.4.99" diff --git a/sys/modules/zfs/zfs_gitrev.h b/sys/modules/zfs/zfs_gitrev.h index 8a1802f5480b..6f568754f61d 100644 --- a/sys/modules/zfs/zfs_gitrev.h +++ b/sys/modules/zfs/zfs_gitrev.h @@ -1 +1 @@ -#define ZFS_META_GITREV "zfs-2.4.99-95-g5605a6d79" +#define ZFS_META_GITREV "zfs-2.4.99-113-g6ae99d269" diff --git a/sys/net/if_vxlan.c b/sys/net/if_vxlan.c index 03184c1fb678..f3a8410a2258 100644 --- a/sys/net/if_vxlan.c +++ b/sys/net/if_vxlan.c @@ -2533,7 +2533,7 @@ vxlan_encap4(struct vxlan_softc *sc, const union vxlan_sockaddr *fvxlsa, ifp = sc->vxl_ifp; srcaddr = sc->vxl_src_addr.in4.sin_addr; - srcport = vxlan_pick_source_port(sc, m); + srcport = htons(vxlan_pick_source_port(sc, m)); dstaddr = fvxlsa->in4.sin_addr; dstport = fvxlsa->in4.sin_port; @@ -2644,7 +2644,7 @@ vxlan_encap6(struct vxlan_softc *sc, const union vxlan_sockaddr *fvxlsa, ifp = sc->vxl_ifp; srcaddr = &sc->vxl_src_addr.in6.sin6_addr; - srcport = vxlan_pick_source_port(sc, m); + srcport = htons(vxlan_pick_source_port(sc, m)); dstaddr = &fvxlsa->in6.sin6_addr; dstport = fvxlsa->in6.sin6_port; diff --git a/sys/netinet/in_pcb.c b/sys/netinet/in_pcb.c index dbe48242381d..712ff28768dc 100644 --- a/sys/netinet/in_pcb.c +++ b/sys/netinet/in_pcb.c @@ -2665,10 +2665,13 @@ in_pcbinshash(struct inpcb *inp) INP_PCBPORTHASH(inp->inp_lport, pcbinfo->ipi_porthashmask)]; /* - * Add entry to load balance group. - * Only do this if SO_REUSEPORT_LB is set. + * Ignore SO_REUSEPORT_LB if the socket is connected. Really this case + * should be an error, but for UDP sockets it is not, and some + * applications erroneously set it on connected UDP sockets, so we can't + * change this without breaking compatibility. */ - if ((inp->inp_socket->so_options & SO_REUSEPORT_LB) != 0) { + if (!connected && + (inp->inp_socket->so_options & SO_REUSEPORT_LB) != 0) { int error = in_pcbinslbgrouphash(inp, M_NODOM); if (error != 0) return (error); @@ -2770,6 +2773,10 @@ in_pcbrehash(struct inpcb *inp) connected = !in_nullhost(inp->inp_faddr); } + /* See the comment in in_pcbinshash(). */ + if (connected && (inp->inp_flags & INP_INLBGROUP) != 0) + in_pcbremlbgrouphash(inp); + /* * When rehashing, the caller must ensure that either the new or the old * foreign address was unspecified. diff --git a/sys/netinet/tcp_stacks/bbr.c b/sys/netinet/tcp_stacks/bbr.c index 66983edcdd73..10383bc0801e 100644 --- a/sys/netinet/tcp_stacks/bbr.c +++ b/sys/netinet/tcp_stacks/bbr.c @@ -477,7 +477,7 @@ bbr_log_rtt_shrinks(struct tcp_bbr *bbr, uint32_t cts, uint32_t applied, uint16_t set); static struct bbr_sendmap * bbr_find_lowest_rsm(struct tcp_bbr *bbr); -static __inline uint32_t +static inline uint32_t bbr_get_rtt(struct tcp_bbr *bbr, int32_t rtt_type); static void bbr_log_to_start(struct tcp_bbr *bbr, uint32_t cts, uint32_t to, int32_t pacing_delay, @@ -1841,7 +1841,7 @@ bbr_counter_destroy(void) } -static __inline void +static inline void bbr_fill_in_logging_data(struct tcp_bbr *bbr, struct tcp_log_bbr *l, uint32_t cts) { memset(l, 0, sizeof(union tcp_log_stackspecific)); @@ -4206,7 +4206,7 @@ bbr_calc_thresh_tlp(struct tcpcb *tp, struct tcp_bbr *bbr, /* * Return one of three RTTs to use (in microseconds). */ -static __inline uint32_t +static inline uint32_t bbr_get_rtt(struct tcp_bbr *bbr, int32_t rtt_type) { uint32_t f_rtt; @@ -4370,7 +4370,7 @@ bbr_timeout_rack(struct tcpcb *tp, struct tcp_bbr *bbr, uint32_t cts) return (0); } -static __inline void +static inline void bbr_clone_rsm(struct tcp_bbr *bbr, struct bbr_sendmap *nrsm, struct bbr_sendmap *rsm, uint32_t start) { int idx; diff --git a/sys/netinet/tcp_stacks/rack.c b/sys/netinet/tcp_stacks/rack.c index 0320479e882e..50077abdfd86 100644 --- a/sys/netinet/tcp_stacks/rack.c +++ b/sys/netinet/tcp_stacks/rack.c @@ -4730,7 +4730,7 @@ rack_make_timely_judgement(struct tcp_rack *rack, uint32_t rtt, int32_t rtt_diff return (timely_says); } -static __inline int +static inline int rack_in_gp_window(struct tcpcb *tp, struct rack_sendmap *rsm) { if (SEQ_GEQ(rsm->r_start, tp->gput_seq) && @@ -4767,7 +4767,7 @@ rack_in_gp_window(struct tcpcb *tp, struct rack_sendmap *rsm) return (0); } -static __inline void +static inline void rack_mark_in_gp_win(struct tcpcb *tp, struct rack_sendmap *rsm) { @@ -4784,7 +4784,7 @@ rack_mark_in_gp_win(struct tcpcb *tp, struct rack_sendmap *rsm) rsm->r_flags &= ~RACK_IN_GP_WIN; } -static __inline void +static inline void rack_clear_gp_marks(struct tcpcb *tp, struct tcp_rack *rack) { /* A GP measurement is ending, clear all marks on the send map*/ @@ -4802,7 +4802,7 @@ rack_clear_gp_marks(struct tcpcb *tp, struct tcp_rack *rack) } -static __inline void +static inline void rack_tend_gp_marks(struct tcpcb *tp, struct tcp_rack *rack) { struct rack_sendmap *rsm = NULL; @@ -7017,7 +7017,7 @@ rack_setup_offset_for_rsm(struct tcp_rack *rack, struct rack_sendmap *src_rsm, s rsm->orig_t_space = M_TRAILINGROOM(rsm->m); } -static __inline void +static inline void rack_clone_rsm(struct tcp_rack *rack, struct rack_sendmap *nrsm, struct rack_sendmap *rsm, uint32_t start) { diff --git a/sys/netinet/tcp_syncache.c b/sys/netinet/tcp_syncache.c index be20fb44a820..3cb538f7054d 100644 --- a/sys/netinet/tcp_syncache.c +++ b/sys/netinet/tcp_syncache.c @@ -1168,7 +1168,7 @@ syncache_expand(struct in_conninfo *inc, struct tcpopt *to, struct tcphdr *th, /* * If listening socket requested TCP digests, check that * received ACK has signature and it is correct. - * If not, drop the ACK and leave sc entry in th cache, + * If not, drop the ACK and leave sc entry in the cache, * because SYN was received with correct signature. */ if (sc->sc_flags & SCF_SIGNATURE) { diff --git a/sys/netinet/udp_usrreq.c b/sys/netinet/udp_usrreq.c index cea8a916679b..0a89d91dfc37 100644 --- a/sys/netinet/udp_usrreq.c +++ b/sys/netinet/udp_usrreq.c @@ -787,7 +787,8 @@ udplite_ctlinput(struct icmp *icmp) static int udp_pcblist(SYSCTL_HANDLER_ARGS) { - struct inpcb_iterator inpi = INP_ALL_ITERATOR(&V_udbinfo, + struct inpcbinfo *pcbinfo = udp_get_inpcbinfo(arg2); + struct inpcb_iterator inpi = INP_ALL_ITERATOR(pcbinfo, INPLOOKUP_RLOCKPCB); struct xinpgen xig; struct inpcb *inp; @@ -799,7 +800,7 @@ udp_pcblist(SYSCTL_HANDLER_ARGS) if (req->oldptr == 0) { int n; - n = V_udbinfo.ipi_count; + n = pcbinfo->ipi_count; n += imax(n / 8, 10); req->oldidx = 2 * (sizeof xig) + n * sizeof(struct xinpcb); return (0); @@ -810,8 +811,8 @@ udp_pcblist(SYSCTL_HANDLER_ARGS) bzero(&xig, sizeof(xig)); xig.xig_len = sizeof xig; - xig.xig_count = V_udbinfo.ipi_count; - xig.xig_gen = V_udbinfo.ipi_gencnt; + xig.xig_count = pcbinfo->ipi_count; + xig.xig_gen = pcbinfo->ipi_gencnt; xig.xig_sogen = so_gencnt; error = SYSCTL_OUT(req, &xig, sizeof xig); if (error) @@ -838,9 +839,9 @@ udp_pcblist(SYSCTL_HANDLER_ARGS) * that something happened while we were processing this * request, and it might be necessary to retry. */ - xig.xig_gen = V_udbinfo.ipi_gencnt; + xig.xig_gen = pcbinfo->ipi_gencnt; xig.xig_sogen = so_gencnt; - xig.xig_count = V_udbinfo.ipi_count; + xig.xig_count = pcbinfo->ipi_count; error = SYSCTL_OUT(req, &xig, sizeof xig); } @@ -848,7 +849,7 @@ udp_pcblist(SYSCTL_HANDLER_ARGS) } SYSCTL_PROC(_net_inet_udp, UDPCTL_PCBLIST, pcblist, - CTLTYPE_OPAQUE | CTLFLAG_RD | CTLFLAG_MPSAFE, NULL, 0, + CTLTYPE_OPAQUE | CTLFLAG_RD | CTLFLAG_MPSAFE, NULL, IPPROTO_UDP, udp_pcblist, "S,xinpcb", "List of active UDP sockets"); diff --git a/sys/riscv/conf/std.allwinner b/sys/riscv/conf/std.allwinner index 34fe195b01ba..ecd789f39963 100644 --- a/sys/riscv/conf/std.allwinner +++ b/sys/riscv/conf/std.allwinner @@ -17,4 +17,7 @@ device awg # Allwinner EMAC Gigabit Ethernet device musb # Mentor Graphics USB OTG controller +# DTBs +makeoptions MODULES_EXTRA+="dtb/allwinner" + files "../allwinner/files.allwinner" diff --git a/sys/riscv/conf/std.starfive b/sys/riscv/conf/std.starfive index 9bdb1af9e79c..6a0e56cc84bd 100644 --- a/sys/riscv/conf/std.starfive +++ b/sys/riscv/conf/std.starfive @@ -10,4 +10,7 @@ device eqos device dwmmc device dwmmc_starfive +# DTBs +makeoptions MODULES_EXTRA+="dtb/starfive" + files "../starfive/files.starfive" diff --git a/sys/riscv/vmm/vmm.c b/sys/riscv/vmm/vmm.c index 4c9b1fa53f7a..24b4be89af48 100644 --- a/sys/riscv/vmm/vmm.c +++ b/sys/riscv/vmm/vmm.c @@ -954,8 +954,7 @@ vcpu_get_state(struct vcpu *vcpu, int *hostcpu) int vm_get_register(struct vcpu *vcpu, int reg, uint64_t *retval) { - - if (reg >= VM_REG_LAST) + if (reg < 0 || reg >= VM_REG_LAST) return (EINVAL); return (vmmops_getreg(vcpu->cookie, reg, retval)); @@ -966,7 +965,7 @@ vm_set_register(struct vcpu *vcpu, int reg, uint64_t val) { int error; - if (reg >= VM_REG_LAST) + if (reg < 0 || reg >= VM_REG_LAST) return (EINVAL); error = vmmops_setreg(vcpu->cookie, reg, val); if (error || reg != VM_REG_GUEST_SEPC) diff --git a/sys/riscv/vmm/vmm_dev_machdep.c b/sys/riscv/vmm/vmm_dev_machdep.c index ba15d8dcd79e..c736b10dc9c0 100644 --- a/sys/riscv/vmm/vmm_dev_machdep.c +++ b/sys/riscv/vmm/vmm_dev_machdep.c @@ -67,18 +67,13 @@ int vmmdev_machdep_ioctl(struct vm *vm, struct vcpu *vcpu, u_long cmd, caddr_t data, int fflag, struct thread *td) { - struct vm_run *vmrun; - struct vm_aplic_descr *aplic; - struct vm_irq *vi; - struct vm_exception *vmexc; - struct vm_gla2gpa *gg; - struct vm_msi *vmsi; int error; error = 0; switch (cmd) { case VM_RUN: { struct vm_exit *vme; + struct vm_run *vmrun; vmrun = (struct vm_run *)data; vme = vm_exitinfo(vcpu); @@ -90,34 +85,52 @@ vmmdev_machdep_ioctl(struct vm *vm, struct vcpu *vcpu, u_long cmd, caddr_t data, error = copyout(vme, vmrun->vm_exit, sizeof(*vme)); break; } - case VM_INJECT_EXCEPTION: + case VM_INJECT_EXCEPTION: { + struct vm_exception *vmexc; + vmexc = (struct vm_exception *)data; error = vm_inject_exception(vcpu, vmexc->scause); break; - case VM_GLA2GPA_NOFAULT: + } + case VM_GLA2GPA_NOFAULT: { + struct vm_gla2gpa *gg; + gg = (struct vm_gla2gpa *)data; error = vm_gla2gpa_nofault(vcpu, &gg->paging, gg->gla, gg->prot, &gg->gpa, &gg->fault); KASSERT(error == 0 || error == EFAULT, ("%s: vm_gla2gpa unknown error %d", __func__, error)); break; - case VM_ATTACH_APLIC: + } + case VM_ATTACH_APLIC: { + struct vm_aplic_descr *aplic; + aplic = (struct vm_aplic_descr *)data; error = vm_attach_aplic(vm, aplic); break; - case VM_RAISE_MSI: + } + case VM_RAISE_MSI: { + struct vm_msi *vmsi; + vmsi = (struct vm_msi *)data; error = vm_raise_msi(vm, vmsi->msg, vmsi->addr, vmsi->bus, vmsi->slot, vmsi->func); break; - case VM_ASSERT_IRQ: + } + case VM_ASSERT_IRQ: { + struct vm_irq *vi; + vi = (struct vm_irq *)data; error = vm_assert_irq(vm, vi->irq); break; - case VM_DEASSERT_IRQ: + } + case VM_DEASSERT_IRQ: { + struct vm_irq *vi; + vi = (struct vm_irq *)data; error = vm_deassert_irq(vm, vi->irq); break; + } default: error = ENOTTY; break; diff --git a/sys/sys/ata.h b/sys/sys/ata.h index 5e9c2bcb9aa0..9ce822214f6d 100644 --- a/sys/sys/ata.h +++ b/sys/sys/ata.h @@ -347,8 +347,11 @@ struct ata_params { #define ATA_STATUS_BUSY 0x80 /* ATA Error Register */ +/* COMMAND TIMEOUT 0x01 */ #define ATA_ERROR_ABORT 0x04 #define ATA_ERROR_ID_NOT_FOUND 0x10 +/* UNCORRECTABLE ERROR 0x40 */ +/* INTERFACE CRC 0x80 */ /* ATA HPA Features */ #define ATA_HPA_FEAT_MAX_ADDR 0x00 diff --git a/sys/sys/param.h b/sys/sys/param.h index 8a71693cff3d..7cfa3c6aa4a8 100644 --- a/sys/sys/param.h +++ b/sys/sys/param.h @@ -74,7 +74,7 @@ * cannot include sys/param.h and should only be updated here. */ #undef __FreeBSD_version -#define __FreeBSD_version 1600001 +#define __FreeBSD_version 1600002 /* * __FreeBSD_kernel__ indicates that this system uses the kernel of FreeBSD, diff --git a/sys/sys/random.h b/sys/sys/random.h index 2a68f0c99b6d..af6b1e117423 100644 --- a/sys/sys/random.h +++ b/sys/sys/random.h @@ -89,11 +89,11 @@ enum random_entropy_source { RANDOM_ENVIRONMENTAL_END = RANDOM_RANDOMDEV, /* Fast hardware random-number sources from here on. */ RANDOM_PURE_START, - RANDOM_PURE_OCTEON = RANDOM_PURE_START, - RANDOM_PURE_SAFE, + RANDOM_PURE_SAFE = RANDOM_PURE_START, RANDOM_PURE_GLXSB, RANDOM_PURE_HIFN, RANDOM_PURE_RDRAND, + RANDOM_PURE_RDSEED, RANDOM_PURE_NEHEMIAH, RANDOM_PURE_RNDTEST, RANDOM_PURE_VIRTIO, diff --git a/sys/x86/iommu/amd_intrmap.c b/sys/x86/iommu/amd_intrmap.c index f8900fe0561f..cce4f57ca323 100644 --- a/sys/x86/iommu/amd_intrmap.c +++ b/sys/x86/iommu/amd_intrmap.c @@ -223,9 +223,9 @@ static struct amdiommu_ctx * amdiommu_ir_find(device_t src, uint16_t *ridp, bool *is_iommu) { devclass_t src_class; - device_t requester; struct amdiommu_unit *unit; struct amdiommu_ctx *ctx; + struct iommu_ctx *ioctx; uint32_t edte; uint16_t rid; uint8_t dte; @@ -255,10 +255,9 @@ amdiommu_ir_find(device_t src, uint16_t *ridp, bool *is_iommu) error = amdiommu_find_unit(src, &unit, &rid, &dte, &edte, bootverbose); if (error == 0) { - error = iommu_get_requester(src, &requester, &rid); - MPASS(error == 0); - ctx = amdiommu_get_ctx_for_dev(unit, src, - rid, 0, false /* XXXKIB */, false, dte, edte); + ioctx = iommu_instantiate_ctx(AMD2IOMMU(unit), src, false); + if (ioctx != NULL) + ctx = IOCTX2CTX(ioctx); } } if (ridp != NULL) |