September 2024 - Kernel - mailweb.openeuler.org

[PATCH OLK-6.6 0/4] Fix CVE-2024-45025
by Long Li 13 Sep '24

13 Sep '24

This patch set fix CVE-2024-45025. Al Viro (1): fix bitmap corruption on close_range() with CLOSE_RANGE_UNSHARE Alexander Lobakin (3): s390/cio: rename bitmap_size() -> idset_bitmap_size() fs/ntfs3: add prefix to bitmap_size() and use BITS_TO_U64() bitmap: introduce generic optimized bitmap_size() drivers/md/dm-clone-metadata.c | 5 --- drivers/s390/cio/idset.c | 12 ++++--- fs/file.c | 30 +++++++--------- fs/ntfs3/bitmap.c | 4 +-- fs/ntfs3/fsntfs.c | 2 +- fs/ntfs3/index.c | 11 +++--- fs/ntfs3/ntfs_fs.h | 4 +-- fs/ntfs3/super.c | 2 +- include/linux/bitmap.h | 20 +++++++++-- include/linux/cpumask.h | 2 +- lib/math/prime_numbers.c | 2 -- tools/include/linux/bitmap.h | 7 ++-- .../testing/selftests/core/close_range_test.c | 35 +++++++++++++++++++ 13 files changed, 89 insertions(+), 47 deletions(-) -- 2.39.2

2 5

[PATCH OLK-5.10 0/4] Fix CVE-2024-45025
by Long Li 13 Sep '24

13 Sep '24

This patch set fix CVE-2024-45025. Al Viro (1): fix bitmap corruption on close_range() with CLOSE_RANGE_UNSHARE Alexander Lobakin (3): s390/cio: rename bitmap_size() -> idset_bitmap_size() fs/ntfs3: add prefix to bitmap_size() and use BITS_TO_U64() bitmap: introduce generic optimized bitmap_size() drivers/md/dm-clone-metadata.c | 5 --- drivers/s390/cio/idset.c | 12 ++++--- fs/file.c | 30 +++++++--------- fs/ntfs3/bitmap.c | 4 +-- fs/ntfs3/fsntfs.c | 2 +- fs/ntfs3/index.c | 11 +++--- fs/ntfs3/ntfs_fs.h | 4 +-- fs/ntfs3/super.c | 2 +- include/linux/bitmap.h | 20 +++++++++-- include/linux/cpumask.h | 2 +- lib/math/prime_numbers.c | 2 -- tools/include/linux/bitmap.h | 7 ++-- .../testing/selftests/core/close_range_test.c | 35 +++++++++++++++++++ 13 files changed, 89 insertions(+), 47 deletions(-) -- 2.39.2

2 5

[PATCH v2 OLK-6.6] KVM:arm64:Add a kvm parameter to control guest wfi trapping
by Zhiqiang Ni 13 Sep '24

13 Sep '24

virt inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/IAQV1W CVE: NA -------------------------------- Currently, on GICv4+ system if a vCPU is only runnable process on a pCPU and we are using direct injection of interrupts, the guest WFI trapping will be turn off. So in host view, the vcpu process's CPU util is almost 100%, which may confuse someone. Let's add a kvm parameter to control guest wfi trapping. Signed-off-by: Zhiqiang Ni <nizhiqiang1(a)huawei.com> --- arch/arm64/include/asm/kvm_emulate.h | 3 ++ arch/arm64/include/asm/kvm_host.h | 2 ++ arch/arm64/kvm/arm.c | 49 ++++++++++++++++++++++++++++ 3 files changed, 54 insertions(+) diff --git a/arch/arm64/include/asm/kvm_emulate.h b/arch/arm64/include/asm/kvm_emulate.h index 2293caab9..6992d010b 100644 --- a/arch/arm64/include/asm/kvm_emulate.h +++ b/arch/arm64/include/asm/kvm_emulate.h @@ -119,6 +119,9 @@ static inline void vcpu_clear_wfx_traps(struct kvm_vcpu *vcpu) vcpu->arch.hcr_el2 &= ~HCR_TWI; else vcpu->arch.hcr_el2 |= HCR_TWI; + + if (force_wfi_trap) + vcpu->arch.hcr_el2 |= HCR_TWI; } static inline void vcpu_set_wfx_traps(struct kvm_vcpu *vcpu) diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h index ac8115098..5ba983f71 100644 --- a/arch/arm64/include/asm/kvm_host.h +++ b/arch/arm64/include/asm/kvm_host.h @@ -54,6 +54,7 @@ #define KVM_REQ_SUSPEND KVM_ARCH_REQ(6) #define KVM_REQ_RESYNC_PMU_EL0 KVM_ARCH_REQ(7) #define KVM_REQ_RELOAD_TLBI_DVMBM KVM_ARCH_REQ(8) +#define KVM_REQ_RELOAD_WFI_TRAPS KVM_ARCH_REQ(9) #define KVM_DIRTY_LOG_MANUAL_CAPS (KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE | \ KVM_DIRTY_LOG_INITIALLY_SET) @@ -1244,6 +1245,7 @@ extern unsigned int twedel; void kvm_arm_vcpu_power_off(struct kvm_vcpu *vcpu); bool kvm_arm_vcpu_stopped(struct kvm_vcpu *vcpu); +extern bool force_wfi_trap; extern bool kvm_ncsnp_support; extern bool kvm_dvmbm_support; diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c index fe3add87e..3e0e84ce9 100644 --- a/arch/arm64/kvm/arm.c +++ b/arch/arm64/kvm/arm.c @@ -82,6 +82,48 @@ unsigned int twedel; module_param(twedel, uint, 0644); #endif +static int vcpu_req_reload_wfi_traps(const char *val, const struct kernel_param *kp); + +static const struct kernel_param_ops force_wfi_trap_ops = { + .set = vcpu_req_reload_wfi_traps, + .get = param_get_bool, +}; + +bool force_wfi_trap; +module_param_cb(force_wfi_trap, &force_wfi_trap_ops, &force_wfi_trap, 0644); + +static int vcpu_req_reload_wfi_traps(const char *val, const struct kernel_param *kp) +{ + struct kvm *kvm; + bool oldvalue; + int err; + + oldvalue = force_wfi_trap; + err = param_set_bool(val, kp); + if (err) + return err; + + if (oldvalue == force_wfi_trap) + return err; + + /* + * If set the force_wfi_trap from 1 to 0, no need to kick vcpus here. + * The HCR_TWI flag will be cleared in kvm_arch_vcpu_load(). + */ + if (force_wfi_trap == 0) + return 0; + + /* + * We need to kick vcpus out of guest mode here to reload + * wfx trapping config when re-enter guest mode. + */ + mutex_lock(&kvm_lock); + list_for_each_entry(kvm, &vm_list, vm_list) + kvm_make_all_cpus_request(kvm, KVM_REQ_RELOAD_WFI_TRAPS); + mutex_unlock(&kvm_lock); + return err; +} + int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu) { return kvm_vcpu_exiting_guest_mode(vcpu) == IN_GUEST_MODE; @@ -958,6 +1000,13 @@ static int check_vcpu_requests(struct kvm_vcpu *vcpu) if (kvm_check_request(KVM_REQ_RELOAD_TLBI_DVMBM, vcpu)) kvm_hisi_reload_lsudvmbm(vcpu->kvm); + + if (kvm_check_request(KVM_REQ_RELOAD_WFI_TRAPS, vcpu)) { + if (single_task_running()) + vcpu_clear_wfx_traps(vcpu); + else + vcpu_set_wfx_traps(vcpu); + } } return 1; -- 2.43.0

2 1

[openeuler:OLK-5.10 19395/30000] drivers/ub/urma/uburma/uburma_cmd.c:156:6: warning: no previous prototype for 'uburma_jfr_event_cb'
by kernel test robot 13 Sep '24

13 Sep '24

Hi Yizhen, FYI, the error/warning still remains. tree: https://gitee.com/openeuler/kernel.git OLK-5.10 head: 5a1d9701155c6908c76c68951170f10279685143 commit: 79b60243a9ff771b76add8e16cfe55cc4f5edef2 [19395/30000] ub: uburma add cmd create jfr implementation. config: arm64-randconfig-004-20240912 (https://download.01.org/0day-ci/archive/20240913/202409131601.MyaN9hL3-lkp@…) compiler: aarch64-linux-gcc (GCC) 14.1.0 reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240913/202409131601.MyaN9hL3-lkp@…) If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202409131601.MyaN9hL3-lkp@intel.com/ All warnings (new ones prefixed by >>): drivers/ub/urma/uburma/uburma_cmd.c:144:6: warning: no previous prototype for 'uburma_jfs_event_cb' [-Wmissing-prototypes] 144 | void uburma_jfs_event_cb(struct ubcore_event *event, struct ubcore_ucontext *ctx) | ^~~~~~~~~~~~~~~~~~~ >> drivers/ub/urma/uburma/uburma_cmd.c:156:6: warning: no previous prototype for 'uburma_jfr_event_cb' [-Wmissing-prototypes] 156 | void uburma_jfr_event_cb(struct ubcore_event *event, struct ubcore_ucontext *ctx) | ^~~~~~~~~~~~~~~~~~~ Kconfig warnings: (for reference only) WARNING: unmet direct dependencies detected for PGP_KEY_PARSER Depends on [n]: CRYPTO [=y] && ASYMMETRIC_KEY_TYPE [=y] && ASYMMETRIC_PUBLIC_KEY_SUBTYPE [=n] Selected by [y]: - PGP_PRELOAD [=y] && CRYPTO [=y] && ASYMMETRIC_KEY_TYPE [=y] vim +/uburma_jfr_event_cb +156 drivers/ub/urma/uburma/uburma_cmd.c 155 > 156 void uburma_jfr_event_cb(struct ubcore_event *event, struct ubcore_ucontext *ctx) 157 { 158 struct uburma_jfr_uobj *jfr_uobj; 159 160 if (event->element.jfr == NULL) 161 return; 162 163 jfr_uobj = (struct uburma_jfr_uobj *)event->element.jfr->jfr_cfg.jfr_context; 164 uburma_write_async_event(ctx, event->element.jfr->urma_jfr, event->event_type, 165 &jfr_uobj->async_event_list, &jfr_uobj->async_events_reported); 166 } 167 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[PATCH v2 OLK-6.6] KVM:arm64:Add a kvm module param to get current enabled GIC version
by Zhiqiang Ni 13 Sep '24

13 Sep '24

From: Yanan Wang <wangyanan55(a)huawei.com> virt inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/IAR5EZ CVE: NA Using GICv4.1 is not a 100% win in all scenarios, sometimes we need to use different GIC version in different usecases. Given that there is not a straight forward way to check the GIC version that system is using (e.g., kernel messages may be overided with time elapses), let's add a module param in KVM such that userspace can check the GIC version simply by reading /sys/module/kvm/parameters/gic_version. Signed-off-by: Yanan Wang <wangyanan55(a)huawei.com> Signed-off-by: Zhiqiang Ni <nizhiqiang1(a)huawei.com> --- arch/arm64/kvm/vgic/vgic-v3.c | 37 +++++++++++++++++++++++++++++++++++ arch/arm64/kvm/vgic/vgic.h | 8 ++++++++ 2 files changed, 45 insertions(+) diff --git a/arch/arm64/kvm/vgic/vgic-v3.c b/arch/arm64/kvm/vgic/vgic-v3.c index dab599e857b5..227fa1874c3b 100644 --- a/arch/arm64/kvm/vgic/vgic-v3.c +++ b/arch/arm64/kvm/vgic/vgic-v3.c @@ -20,6 +20,37 @@ static bool common_trap; static bool dir_trap; static bool gicv4_enable; +enum gic_version global_gic_version = UNKNOWN_GIC_VERSION; +static const char * const gic_version_str[] = { + "GICv2", + "GICv3", + "GICv4", + "GICv4.1", + "unknown" +}; + +static int get_gic_version(char *buffer, const struct kernel_param *kp); +static const struct kernel_param_ops gic_version_ops = { + .get = get_gic_version, +}; + +static const char *gic_version; +module_param_cb(gic_version, &gic_version_ops, &gic_version, 0444); +MODULE_PARM_DESC(gic_version, "Version of current enabled GIC"); + +static int get_gic_version(char *buffer, const struct kernel_param *kp) +{ + int index = global_gic_version; + + if (index < 0 || index >= ARRAY_SIZE(gic_version_str)) { + kvm_err("Invalid GIC version enum %d\n", index); + return -EINVAL; + } + gic_version = gic_version_str[index]; + + return param_get_charp(buffer, kp); +} + void vgic_v3_set_underflow(struct kvm_vcpu *vcpu) { struct vgic_v3_cpu_if *cpuif = &vcpu->arch.vgic_cpu.vgic_v3; @@ -654,6 +685,7 @@ int vgic_v3_probe(const struct gic_kvm_info *info) kvm_vgic_global_state.nr_lr = (ich_vtr_el2 & 0xf) + 1; kvm_vgic_global_state.can_emulate_gicv2 = false; kvm_vgic_global_state.ich_vtr_el2 = ich_vtr_el2; + global_gic_version = GICV3; /* GICv4 support? */ if (info->has_v4) { @@ -663,6 +695,11 @@ int vgic_v3_probe(const struct gic_kvm_info *info) kvm_vgic_global_state.has_gicv4_1 ? ".1" : "", gicv4_enable ? "en" : "dis"); + if (kvm_vgic_global_state.has_gicv4_1) + global_gic_version = GICV4_1; + else if (kvm_vgic_global_state.has_gicv4) + global_gic_version = GICV4; + #ifdef CONFIG_VIRT_VTIMER_IRQ_BYPASS kvm_vgic_global_state.has_direct_vtimer = info->has_vtimer && gicv4_enable; if (kvm_vgic_global_state.has_direct_vtimer) diff --git a/arch/arm64/kvm/vgic/vgic.h b/arch/arm64/kvm/vgic/vgic.h index 5135c690839e..539909a681d7 100644 --- a/arch/arm64/kvm/vgic/vgic.h +++ b/arch/arm64/kvm/vgic/vgic.h @@ -157,6 +157,14 @@ static inline bool vgic_direct_sgi_or_ppi(struct vgic_irq *irq) #endif } +enum gic_version { + GICV2, + GICV3, + GICV4, + GICV4_1, + UNKNOWN_GIC_VERSION +}; + /* * This struct provides an intermediate representation of the fields contained * in the GICH_VMCR and ICH_VMCR registers, such that code exporting the GIC -- 2.43.0

2 1

[PATCH v2 OLK-6.6] KVM:arm64:Add a kvm module param to get current enabled GIC version
by Zhiqiang Ni 13 Sep '24

13 Sep '24

From: Yanan Wang <wangyanan55(a)huawei.com> virt inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/IAR5EZ CVE: NA Using GICv4.1 is not a 100% win in all scenarios, sometimes we need to use different GIC version in different usecases. Given that there is not a straight forward way to check the GIC version that system is using (e.g., kernel messages may be overided with time elapses), let's add a module param in KVM such that userspace can check the GIC version simply by reading /sys/module/kvm/parameters/gic_version. Signed-off-by: Yanan Wang <wangyanan55(a)huawei.com> Signed-off-by: Zhiqiang Ni <nizhiqiang1(a)huawei.com> --- arch/arm64/kvm/vgic/vgic-v3.c | 37 +++++++++++++++++++++++++++++++++++ arch/arm64/kvm/vgic/vgic.h | 8 ++++++++ 2 files changed, 45 insertions(+) diff --git a/arch/arm64/kvm/vgic/vgic-v3.c b/arch/arm64/kvm/vgic/vgic-v3.c index dab599e857b5..227fa1874c3b 100644 --- a/arch/arm64/kvm/vgic/vgic-v3.c +++ b/arch/arm64/kvm/vgic/vgic-v3.c @@ -20,6 +20,37 @@ static bool common_trap; static bool dir_trap; static bool gicv4_enable; +enum gic_version global_gic_version = UNKNOWN_GIC_VERSION; +static const char * const gic_version_str[] = { + "GICv2", + "GICv3", + "GICv4", + "GICv4.1", + "unknown" +}; + +static int get_gic_version(char *buffer, const struct kernel_param *kp); +static const struct kernel_param_ops gic_version_ops = { + .get = get_gic_version, +}; + +static const char *gic_version; +module_param_cb(gic_version, &gic_version_ops, &gic_version, 0444); +MODULE_PARM_DESC(gic_version, "Version of current enabled GIC"); + +static int get_gic_version(char *buffer, const struct kernel_param *kp) +{ + int index = global_gic_version; + + if (index < 0 || index >= ARRAY_SIZE(gic_version_str)) { + kvm_err("Invalid GIC version enum %d\n", index); + return -EINVAL; + } + gic_version = gic_version_str[index]; + + return param_get_charp(buffer, kp); +} + void vgic_v3_set_underflow(struct kvm_vcpu *vcpu) { struct vgic_v3_cpu_if *cpuif = &vcpu->arch.vgic_cpu.vgic_v3; @@ -654,6 +685,7 @@ int vgic_v3_probe(const struct gic_kvm_info *info) kvm_vgic_global_state.nr_lr = (ich_vtr_el2 & 0xf) + 1; kvm_vgic_global_state.can_emulate_gicv2 = false; kvm_vgic_global_state.ich_vtr_el2 = ich_vtr_el2; + global_gic_version = GICV3; /* GICv4 support? */ if (info->has_v4) { @@ -663,6 +695,11 @@ int vgic_v3_probe(const struct gic_kvm_info *info) kvm_vgic_global_state.has_gicv4_1 ? ".1" : "", gicv4_enable ? "en" : "dis"); + if (kvm_vgic_global_state.has_gicv4_1) + global_gic_version = GICV4_1; + else if (kvm_vgic_global_state.has_gicv4) + global_gic_version = GICV4; + #ifdef CONFIG_VIRT_VTIMER_IRQ_BYPASS kvm_vgic_global_state.has_direct_vtimer = info->has_vtimer && gicv4_enable; if (kvm_vgic_global_state.has_direct_vtimer) diff --git a/arch/arm64/kvm/vgic/vgic.h b/arch/arm64/kvm/vgic/vgic.h index 5135c690839e..539909a681d7 100644 --- a/arch/arm64/kvm/vgic/vgic.h +++ b/arch/arm64/kvm/vgic/vgic.h @@ -157,6 +157,14 @@ static inline bool vgic_direct_sgi_or_ppi(struct vgic_irq *irq) #endif } +enum gic_version { + GICV2, + GICV3, + GICV4, + GICV4_1, + UNKNOWN_GIC_VERSION +}; + /* * This struct provides an intermediate representation of the fields contained * in the GICH_VMCR and ICH_VMCR registers, such that code exporting the GIC -- 2.43.0

2 1

[PATCH openEuler-22.03-LTS-SP1] [Backport] usb: gadget: f_ncm: fix potential NULL ptr deref in ncm_bitrate()
by Wang Hai 13 Sep '24

13 Sep '24

From: Maciej Żenczykowski <maze(a)google.com> stable inclusion from stable-v5.10.165 commit e92c70059178da751e5af7de02384b7dfadb5ec7 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IALIDP CVE: CVE-2023-52894 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit c6ec929595c7443250b2a4faea988c62019d5cd2 upstream. In Google internal bug 265639009 we've received an (as yet) unreproducible crash report from an aarch64 GKI 5.10.149-android13 running device. AFAICT the source code is at: https://android.googlesource.com/kernel/common/+/refs/tags/ASB-2022-12-05_1… The call stack is: ncm_close() -> ncm_notify() -> ncm_do_notify() with the crash at: ncm_do_notify+0x98/0x270 Code: 79000d0b b9000a6c f940012a f9400269 (b9405d4b) Which I believe disassembles to (I don't know ARM assembly, but it looks sane enough to me...): // halfword (16-bit) store presumably to event->wLength (at offset 6 of struct usb_cdc_notification) 0B 0D 00 79 strh w11, [x8, #6] // word (32-bit) store presumably to req->Length (at offset 8 of struct usb_request) 6C 0A 00 B9 str w12, [x19, #8] // x10 (NULL) was read here from offset 0 of valid pointer x9 // IMHO we're reading 'cdev->gadget' and getting NULL // gadget is indeed at offset 0 of struct usb_composite_dev 2A 01 40 F9 ldr x10, [x9] // loading req->buf pointer, which is at offset 0 of struct usb_request 69 02 40 F9 ldr x9, [x19] // x10 is null, crash, appears to be attempt to read cdev->gadget->max_speed 4B 5D 40 B9 ldr w11, [x10, #0x5c] which seems to line up with ncm_do_notify() case NCM_NOTIFY_SPEED code fragment: event->wLength = cpu_to_le16(8); req->length = NCM_STATUS_BYTECOUNT; /* SPEED_CHANGE data is up/down speeds in bits/sec */ data = req->buf + sizeof *event; data[0] = cpu_to_le32(ncm_bitrate(cdev->gadget)); My analysis of registers and NULL ptr deref crash offset (Unable to handle kernel NULL pointer dereference at virtual address 000000000000005c) heavily suggests that the crash is due to 'cdev->gadget' being NULL when executing: data[0] = cpu_to_le32(ncm_bitrate(cdev->gadget)); which calls: ncm_bitrate(NULL) which then calls: gadget_is_superspeed(NULL) which reads ((struct usb_gadget *)NULL)->max_speed and hits a panic. AFAICT, if I'm counting right, the offset of max_speed is indeed 0x5C. (remember there's a GKI KABI reservation of 16 bytes in struct work_struct) It's not at all clear to me how this is all supposed to work... but returning 0 seems much better than panic-ing... Cc: Felipe Balbi <balbi(a)kernel.org> Cc: Lorenzo Colitti <lorenzo(a)google.com> Cc: Carlos Llamas <cmllamas(a)google.com> Cc: stable(a)vger.kernel.org Signed-off-by: Maciej Żenczykowski <maze(a)google.com> Cc: stable <stable(a)kernel.org> Link: https://lore.kernel.org/r/20230117131839.1138208-1-maze@google.com Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Wang Hai <wanghai38(a)huawei.com> --- drivers/usb/gadget/function/f_ncm.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/usb/gadget/function/f_ncm.c b/drivers/usb/gadget/function/f_ncm.c index 855127249f24..f56147489835 100644 --- a/drivers/usb/gadget/function/f_ncm.c +++ b/drivers/usb/gadget/function/f_ncm.c @@ -85,7 +85,9 @@ static inline struct f_ncm *func_to_ncm(struct usb_function *f) /* peak (theoretical) bulk transfer rate in bits-per-second */ static inline unsigned ncm_bitrate(struct usb_gadget *g) { - if (gadget_is_superspeed(g) && g->speed >= USB_SPEED_SUPER_PLUS) + if (!g) + return 0; + else if (gadget_is_superspeed(g) && g->speed >= USB_SPEED_SUPER_PLUS) return 4250000000U; else if (gadget_is_superspeed(g) && g->speed == USB_SPEED_SUPER) return 3750000000U; -- 2.17.1

2 1

[PATCH OLK-6.6] KVM:arm64:Add a kvm module param to get current enabled GIC version
by Zhiqiang Ni 13 Sep '24

13 Sep '24

From: Yanan Wang <wangyanan55(a)huawei.com> virt inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/IAR5EZ CVE: NA Using GICv4.1 is not a 100% win in all scenarios, sometimes we need to use different GIC version in different usecases. Given that there is not a straight forward way to check the GIC version that system is using (e.g., kernel messages may be overriden with time elapses), let's add a module param in KVM such that userspace can check the GIC version simply by reading /sys/module/kvm/parameters/gic_version. Signed-off-by: Yanan Wang <wangyanan55(a)huawei.com> Signed-off-by: Zhiqiang Ni <nizhiqiang1(a)huawei.com> --- arch/arm64/kvm/vgic/vgic-v3.c | 37 +++++++++++++++++++++++++++++++++++ arch/arm64/kvm/vgic/vgic.h | 8 ++++++++ 2 files changed, 45 insertions(+) diff --git a/arch/arm64/kvm/vgic/vgic-v3.c b/arch/arm64/kvm/vgic/vgic-v3.c index dab599e857b5..3af4046b0e6d 100644 --- a/arch/arm64/kvm/vgic/vgic-v3.c +++ b/arch/arm64/kvm/vgic/vgic-v3.c @@ -20,6 +20,37 @@ static bool common_trap; static bool dir_trap; static bool gicv4_enable; +enum gic_version global_gic_version = UNKNOWN_GIC_VERSION; +static const char * const gic_version_str[] = { + "GICv2", + "GICv3", + "GICv4", + "GICv4.1", + "unknown" +}; + +static int get_gic_version(char *buffer, const struct kernel_param *kp); +static const struct kernel_param_ops gic_version_ops = { + .get = get_gic_version, +}; + +static const char *gic_version; +module_param_cb(gic_version, &gic_version_ops, &gic_version, S_IRUGO); +MODULE_PARM_DESC(gic_version, "Version of current enabled GIC"); + +static int get_gic_version(char *buffer, const struct kernel_param *kp) +{ + int index = global_gic_version; + + if (index < 0 || index >= ARRAY_SIZE(gic_version_str)) { + kvm_err("Invalid GIC version enum %d\n", index); + return -EINVAL; + } + gic_version = gic_version_str[index]; + + return param_get_charp(buffer, kp); +} + void vgic_v3_set_underflow(struct kvm_vcpu *vcpu) { struct vgic_v3_cpu_if *cpuif = &vcpu->arch.vgic_cpu.vgic_v3; @@ -654,6 +685,7 @@ int vgic_v3_probe(const struct gic_kvm_info *info) kvm_vgic_global_state.nr_lr = (ich_vtr_el2 & 0xf) + 1; kvm_vgic_global_state.can_emulate_gicv2 = false; kvm_vgic_global_state.ich_vtr_el2 = ich_vtr_el2; + global_gic_version = GICV3; /* GICv4 support? */ if (info->has_v4) { @@ -663,6 +695,11 @@ int vgic_v3_probe(const struct gic_kvm_info *info) kvm_vgic_global_state.has_gicv4_1 ? ".1" : "", gicv4_enable ? "en" : "dis"); + if (kvm_vgic_global_state.has_gicv4_1) + global_gic_version = GICV4_1; + else if (kvm_vgic_global_state.has_gicv4) + global_gic_version = GICV4; + #ifdef CONFIG_VIRT_VTIMER_IRQ_BYPASS kvm_vgic_global_state.has_direct_vtimer = info->has_vtimer && gicv4_enable; if (kvm_vgic_global_state.has_direct_vtimer) diff --git a/arch/arm64/kvm/vgic/vgic.h b/arch/arm64/kvm/vgic/vgic.h index 5135c690839e..539909a681d7 100644 --- a/arch/arm64/kvm/vgic/vgic.h +++ b/arch/arm64/kvm/vgic/vgic.h @@ -157,6 +157,14 @@ static inline bool vgic_direct_sgi_or_ppi(struct vgic_irq *irq) #endif } +enum gic_version { + GICV2, + GICV3, + GICV4, + GICV4_1, + UNKNOWN_GIC_VERSION +}; + /* * This struct provides an intermediate representation of the fields contained * in the GICH_VMCR and ICH_VMCR registers, such that code exporting the GIC -- 2.43.0

2 1

[PATCH OLK-6.6] KVM:arm64:Add a kvm parameter to control guest wfi trapping
by Zhiqiang Ni 13 Sep '24

13 Sep '24

virt inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/IAQV1W CVE: NA -------------------------------- Currently, on GICv4+ system if a vCPU is only runnable process on a pCPU and we are using direct injection of interrupts, the guest WFI trapping will be turn off. So in host view, the vcpu process's CPU util is almost 100%, which may confuse someone. Let's add a kvm parameter to control guest wfi trapping. Signed-off-by: Zhiqiang Ni <nizhiqiang1(a)huawei.com> --- arch/arm64/include/asm/kvm_emulate.h | 3 ++ arch/arm64/include/asm/kvm_host.h | 2 ++ arch/arm64/kvm/arm.c | 49 ++++++++++++++++++++++++++++ 3 files changed, 54 insertions(+) diff --git a/arch/arm64/include/asm/kvm_emulate.h b/arch/arm64/include/asm/kvm_emulate.h index 2293caab9..6992d010b 100644 --- a/arch/arm64/include/asm/kvm_emulate.h +++ b/arch/arm64/include/asm/kvm_emulate.h @@ -119,6 +119,9 @@ static inline void vcpu_clear_wfx_traps(struct kvm_vcpu *vcpu) vcpu->arch.hcr_el2 &= ~HCR_TWI; else vcpu->arch.hcr_el2 |= HCR_TWI; + + if (force_wfi_trap) + vcpu->arch.hcr_el2 |= HCR_TWI; } static inline void vcpu_set_wfx_traps(struct kvm_vcpu *vcpu) diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h index ac8115098..5ba983f71 100644 --- a/arch/arm64/include/asm/kvm_host.h +++ b/arch/arm64/include/asm/kvm_host.h @@ -54,6 +54,7 @@ #define KVM_REQ_SUSPEND KVM_ARCH_REQ(6) #define KVM_REQ_RESYNC_PMU_EL0 KVM_ARCH_REQ(7) #define KVM_REQ_RELOAD_TLBI_DVMBM KVM_ARCH_REQ(8) +#define KVM_REQ_RELOAD_WFI_TRAPS KVM_ARCH_REQ(9) #define KVM_DIRTY_LOG_MANUAL_CAPS (KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE | \ KVM_DIRTY_LOG_INITIALLY_SET) @@ -1244,6 +1245,7 @@ extern unsigned int twedel; void kvm_arm_vcpu_power_off(struct kvm_vcpu *vcpu); bool kvm_arm_vcpu_stopped(struct kvm_vcpu *vcpu); +extern bool force_wfi_trap; extern bool kvm_ncsnp_support; extern bool kvm_dvmbm_support; diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c index fe3add87e..22560263f 100644 --- a/arch/arm64/kvm/arm.c +++ b/arch/arm64/kvm/arm.c @@ -82,6 +82,48 @@ unsigned int twedel; module_param(twedel, uint, 0644); #endif +static int vcpu_req_reload_wfi_traps(const char *val, const struct kernel_param *kp); + +static const struct kernel_param_ops force_wfi_trap_ops = { + .set = vcpu_req_reload_wfi_traps, + .get = param_get_bool, +}; + +bool force_wfi_trap = false; +module_param_cb(force_wfi_trap, &force_wfi_trap_ops, &force_wfi_trap, S_IRUGO | S_IWUSR); + +static int vcpu_req_reload_wfi_traps(const char *val, const struct kernel_param *kp) +{ + struct kvm *kvm; + bool oldvalue; + int err; + + oldvalue = force_wfi_trap; + err = param_set_bool(val, kp); + if (err) + return err; + + if (oldvalue == force_wfi_trap) + return err; + + /* + * If set the force_wfi_trap from 1 to 0, no need to kick vcpus here. + * The HCR_TWI flag will be cleared in kvm_arch_vcpu_load(). + */ + if (force_wfi_trap == 0) + return 0; + + /* + * We need to kick vcpus out of guest mode here to reload + * wfx trapping config when re-enter guest mode. + */ + mutex_lock(&kvm_lock); + list_for_each_entry(kvm, &vm_list, vm_list) + kvm_make_all_cpus_request(kvm, KVM_REQ_RELOAD_WFI_TRAPS); + mutex_unlock(&kvm_lock); + return err; +} + int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu) { return kvm_vcpu_exiting_guest_mode(vcpu) == IN_GUEST_MODE; @@ -958,6 +1000,13 @@ static int check_vcpu_requests(struct kvm_vcpu *vcpu) if (kvm_check_request(KVM_REQ_RELOAD_TLBI_DVMBM, vcpu)) kvm_hisi_reload_lsudvmbm(vcpu->kvm); + + if (kvm_check_request(KVM_REQ_RELOAD_WFI_TRAPS, vcpu)) { + if (single_task_running()) + vcpu_clear_wfx_traps(vcpu); + else + vcpu_set_wfx_traps(vcpu); + } } return 1; -- 2.43.0

2 1

[openeuler:openEuler-1.0-LTS] BUILD REGRESSION c46f803e3465bd0ca66716804a4d3e20f586ac0d
by kernel test robot 13 Sep '24

13 Sep '24

tree/branch: https://gitee.com/openeuler/kernel.git openEuler-1.0-LTS branch HEAD: c46f803e3465bd0ca66716804a4d3e20f586ac0d !11525 v2 usb: gadget: core: Check for unset descriptor Error/Warning (recently discovered and may have been fixed): drivers/iommu/arm-smmu-v3.c:223:55: error: 'CONFIG_CMA_ALIGNMENT' undeclared (first use in this function); did you mean 'CONFIG_CMDLINE'? include/linux/kernel.h:875:9: error: first argument to '__builtin_choose_expr' not a constant kernel/sched/core.c:5976:9: error: implicit declaration of function 'init_auto_affinity'; did you mean 'irq_set_affinity'? [-Werror=implicit-function-declaration] Error/Warning ids grouped by kconfigs: recent_errors |-- arm64-allmodconfig | |-- block-blk-io-hierarchy-iodump.c:warning:no-previous-prototype-for-__bio_stage_hierarchy_start | `-- include-linux-thread_info.h:warning:b-may-be-used-uninitialized |-- arm64-randconfig-001-20240913 | `-- arch-arm64-include-asm-irqflags.h:warning:flags-may-be-used-uninitialized |-- arm64-randconfig-004-20240913 | |-- drivers-iommu-arm-smmu-v3.c:error:CONFIG_CMA_ALIGNMENT-undeclared-(first-use-in-this-function) | |-- drivers-tty-serial-sc16is7xx.c:warning:label-err_i2c-defined-but-not-used | |-- include-linux-kernel.h:error:first-argument-to-__builtin_choose_expr-not-a-constant | `-- kernel-sched-core.c:error:implicit-declaration-of-function-init_auto_affinity |-- x86_64-allyesconfig | `-- block-blk-io-hierarchy-iodump.c:warning:no-previous-prototype-for-function-__bio_stage_hierarchy_start `-- x86_64-randconfig-001-20240913 `-- kernel-sched-core.c:error:implicit-declaration-of-function-init_auto_affinity elapsed time: 995m configs tested: 20 configs skipped: 139 tested configs: arm64 allmodconfig gcc-14.1.0 arm64 allnoconfig gcc-14.1.0 arm64 randconfig-001-20240913 gcc-14.1.0 arm64 randconfig-002-20240913 gcc-14.1.0 arm64 randconfig-003-20240913 gcc-14.1.0 arm64 randconfig-004-20240913 gcc-14.1.0 x86_64 allnoconfig clang-18 x86_64 allyesconfig clang-18 x86_64 buildonly-randconfig-001-20240913 clang-18 x86_64 buildonly-randconfig-002-20240913 gcc-12 x86_64 buildonly-randconfig-003-20240913 clang-18 x86_64 buildonly-randconfig-004-20240913 clang-18 x86_64 buildonly-randconfig-005-20240913 gcc-12 x86_64 buildonly-randconfig-006-20240913 clang-18 x86_64 defconfig gcc-11 x86_64 randconfig-001-20240913 gcc-12 x86_64 randconfig-002-20240913 clang-18 x86_64 randconfig-003-20240913 gcc-12 x86_64 randconfig-004-20240913 clang-18 x86_64 rhel-8.3-rust clang-18 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0