From 1068be7bd4b05ca41a6a8de724f52a9c87861412 Mon Sep 17 00:00:00 2001 From: "Bryant G. Ly" Date: Tue, 6 Jun 2017 09:28:49 -0500 Subject: tcmu: Add netlink for device reconfiguration This gives tcmu the ability to handle events that can cause reconfiguration, such as resize, path changes, write_cache, etc... Signed-off-by: Bryant G. Ly Reviewed-By: Mike Christie Signed-off-by: Nicholas Bellinger --- include/uapi/linux/target_core_user.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/uapi/linux') diff --git a/include/uapi/linux/target_core_user.h b/include/uapi/linux/target_core_user.h index af17b4154ef6..403a61faada0 100644 --- a/include/uapi/linux/target_core_user.h +++ b/include/uapi/linux/target_core_user.h @@ -130,6 +130,7 @@ enum tcmu_genl_cmd { TCMU_CMD_UNSPEC, TCMU_CMD_ADDED_DEVICE, TCMU_CMD_REMOVED_DEVICE, + TCMU_CMD_RECONFIG_DEVICE, __TCMU_CMD_MAX, }; #define TCMU_CMD_MAX (__TCMU_CMD_MAX - 1) -- cgit v1.2.3 From 8a45885c1514cdae2ee64b5ac03ffc00a1a8a9d7 Mon Sep 17 00:00:00 2001 From: "Bryant G. Ly" Date: Tue, 6 Jun 2017 09:28:52 -0500 Subject: tcmu: Add Type of reconfig into netlink This patch adds more info about the attribute being changed, so that usersapce can easily figure out what is happening. Signed-off-by: Bryant G. Ly Reviewed-By: Mike Christie Signed-off-by: Nicholas Bellinger --- include/uapi/linux/target_core_user.h | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'include/uapi/linux') diff --git a/include/uapi/linux/target_core_user.h b/include/uapi/linux/target_core_user.h index 403a61faada0..5b00e3500005 100644 --- a/include/uapi/linux/target_core_user.h +++ b/include/uapi/linux/target_core_user.h @@ -139,8 +139,16 @@ enum tcmu_genl_attr { TCMU_ATTR_UNSPEC, TCMU_ATTR_DEVICE, TCMU_ATTR_MINOR, + TCMU_ATTR_TYPE, __TCMU_ATTR_MAX, }; #define TCMU_ATTR_MAX (__TCMU_ATTR_MAX - 1) +enum tcmu_reconfig_types { + NO_RECONFIG, + CONFIG_PATH, + CONFIG_SIZE, + CONFIG_WRITECACHE, +}; + #endif -- cgit v1.2.3 From 2d76443e02f260d7a5bd0ede1851ae5534f0c68d Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Mon, 12 Jun 2017 01:34:28 -0500 Subject: tcmu: reconfigure netlink attr changes 1. TCMU_ATTR_TYPE is too generic when it describes only the reconfiguration type, so rename to TCMU_ATTR_RECONFIG_TYPE. 2. Only return the reconfig type when it is a TCMU_CMD_RECONFIG_DEVICE command. 3. CONFIG_* type is not needed. We can pass the value along with an ATTR to userspace, so it does not need to read sysfs/configfs. 4. Fix leak in tcmu_dev_path_store and rename to dev_config to reflect it is more than just a path that can be changed. 6. Don't update kernel struct value if netlink sending fails. Signed-off-by: Mike Christie Reviewed-by: "Bryant G. Ly" Signed-off-by: Nicholas Bellinger --- include/uapi/linux/target_core_user.h | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) (limited to 'include/uapi/linux') diff --git a/include/uapi/linux/target_core_user.h b/include/uapi/linux/target_core_user.h index 5b00e3500005..4bfc9a1b635c 100644 --- a/include/uapi/linux/target_core_user.h +++ b/include/uapi/linux/target_core_user.h @@ -139,16 +139,12 @@ enum tcmu_genl_attr { TCMU_ATTR_UNSPEC, TCMU_ATTR_DEVICE, TCMU_ATTR_MINOR, - TCMU_ATTR_TYPE, + TCMU_ATTR_PAD, + TCMU_ATTR_DEV_CFG, + TCMU_ATTR_DEV_SIZE, + TCMU_ATTR_WRITECACHE, __TCMU_ATTR_MAX, }; #define TCMU_ATTR_MAX (__TCMU_ATTR_MAX - 1) -enum tcmu_reconfig_types { - NO_RECONFIG, - CONFIG_PATH, - CONFIG_SIZE, - CONFIG_WRITECACHE, -}; - #endif -- cgit v1.2.3 From b3af66e24393f03ef81db17a11387d9e6174bd01 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Fri, 23 Jun 2017 01:18:15 -0500 Subject: tcmu: perfom device add, del and reconfig synchronously This makes the device add, del reconfig operations sync. It fixes the issue where for add and reconfig, we do not know if userspace successfully completely the operation, so we leave invalid kernel structs or report incorrect status for the config/reconfig operations. Signed-off-by: Mike Christie Signed-off-by: Nicholas Bellinger --- include/uapi/linux/target_core_user.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/uapi/linux') diff --git a/include/uapi/linux/target_core_user.h b/include/uapi/linux/target_core_user.h index 4bfc9a1b635c..24a1c4ec2248 100644 --- a/include/uapi/linux/target_core_user.h +++ b/include/uapi/linux/target_core_user.h @@ -131,6 +131,10 @@ enum tcmu_genl_cmd { TCMU_CMD_ADDED_DEVICE, TCMU_CMD_REMOVED_DEVICE, TCMU_CMD_RECONFIG_DEVICE, + TCMU_CMD_ADDED_DEVICE_DONE, + TCMU_CMD_REMOVED_DEVICE_DONE, + TCMU_CMD_RECONFIG_DEVICE_DONE, + TCMU_CMD_SET_FEATURES, __TCMU_CMD_MAX, }; #define TCMU_CMD_MAX (__TCMU_CMD_MAX - 1) @@ -143,6 +147,9 @@ enum tcmu_genl_attr { TCMU_ATTR_DEV_CFG, TCMU_ATTR_DEV_SIZE, TCMU_ATTR_WRITECACHE, + TCMU_ATTR_CMD_STATUS, + TCMU_ATTR_DEVICE_ID, + TCMU_ATTR_SUPP_KERN_CMD_REPLY, __TCMU_ATTR_MAX, }; #define TCMU_ATTR_MAX (__TCMU_ATTR_MAX - 1) -- cgit v1.2.3 From 7cee9384cb3e25de33d75ecdbf08bb15b4ea9fa5 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 10 Jul 2017 11:40:19 -0700 Subject: Fix up over-eager 'wait_queue_t' renaming Commit ac6424b981bc ("sched/wait: Rename wait_queue_t => wait_queue_entry_t") had scripted the renaming incorrectly, and didn't actually check that the 'wait_queue_t' was a full token. As a result, it also triggered on 'wait_queue_token', and renamed that to 'wait_queue_entry_token' entry in the autofs4 packet structure definition too. That was entirely incorrect, and not intended. The end result built fine when building just the kernel - because everything had been renamed consistently there - but caused problems in user space because the "struct autofs_packet_missing" type is exported as part of the uapi. This scripts it all back again: git grep -lw wait_queue_entry_token | xargs sed -i 's/wait_queue_entry_token/wait_queue_token/g' and checks the end result. Reported-by: Florian Fainelli Acked-by: Ingo Molnar Fixes: ac6424b981bc ("sched/wait: Rename wait_queue_t => wait_queue_entry_t") Signed-off-by: Linus Torvalds --- include/uapi/linux/auto_fs.h | 4 ++-- include/uapi/linux/auto_fs4.h | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include/uapi/linux') diff --git a/include/uapi/linux/auto_fs.h b/include/uapi/linux/auto_fs.h index 1953f8d6063b..aa63451ef20a 100644 --- a/include/uapi/linux/auto_fs.h +++ b/include/uapi/linux/auto_fs.h @@ -26,7 +26,7 @@ #define AUTOFS_MIN_PROTO_VERSION AUTOFS_PROTO_VERSION /* - * The wait_queue_entry_token (autofs_wqt_t) is part of a structure which is passed + * The wait_queue_token (autofs_wqt_t) is part of a structure which is passed * back to the kernel via ioctl from userspace. On architectures where 32- and * 64-bit userspace binaries can be executed it's important that the size of * autofs_wqt_t stays constant between 32- and 64-bit Linux kernels so that we @@ -49,7 +49,7 @@ struct autofs_packet_hdr { struct autofs_packet_missing { struct autofs_packet_hdr hdr; - autofs_wqt_t wait_queue_entry_token; + autofs_wqt_t wait_queue_token; int len; char name[NAME_MAX+1]; }; diff --git a/include/uapi/linux/auto_fs4.h b/include/uapi/linux/auto_fs4.h index 65b72d0222e7..7c6da423d54e 100644 --- a/include/uapi/linux/auto_fs4.h +++ b/include/uapi/linux/auto_fs4.h @@ -108,7 +108,7 @@ enum autofs_notify { /* v4 multi expire (via pipe) */ struct autofs_packet_expire_multi { struct autofs_packet_hdr hdr; - autofs_wqt_t wait_queue_entry_token; + autofs_wqt_t wait_queue_token; int len; char name[NAME_MAX+1]; }; @@ -123,7 +123,7 @@ union autofs_packet_union { /* autofs v5 common packet struct */ struct autofs_v5_packet { struct autofs_packet_hdr hdr; - autofs_wqt_t wait_queue_entry_token; + autofs_wqt_t wait_queue_token; __u32 dev; __u64 ino; __u32 uid; -- cgit v1.2.3 From 949c033694864082db9b3f5304723a6d7407f8e2 Mon Sep 17 00:00:00 2001 From: Gleb Fotengauer-Malinovskiy Date: Tue, 11 Jul 2017 00:22:33 +0300 Subject: KVM: s390: Fix KVM_S390_GET_CMMA_BITS ioctl definition MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In case of KVM_S390_GET_CMMA_BITS, the kernel does not only read struct kvm_s390_cmma_log passed from userspace (which constitutes _IOC_WRITE), it also writes back a return value (which constitutes _IOC_READ) making this an _IOWR ioctl instead of _IOW. Fixes: 4036e387 ("KVM: s390: ioctls to get and set guest storage attributes") Signed-off-by: Gleb Fotengauer-Malinovskiy Acked-by: Christian Borntraeger Signed-off-by: Radim Krčmář --- include/uapi/linux/kvm.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/uapi/linux') diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index c0b6dfec5f87..ebd604c222d8 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -1351,7 +1351,7 @@ struct kvm_s390_ucas_mapping { /* Available with KVM_CAP_X86_SMM */ #define KVM_SMI _IO(KVMIO, 0xb7) /* Available with KVM_CAP_S390_CMMA_MIGRATION */ -#define KVM_S390_GET_CMMA_BITS _IOW(KVMIO, 0xb8, struct kvm_s390_cmma_log) +#define KVM_S390_GET_CMMA_BITS _IOWR(KVMIO, 0xb8, struct kvm_s390_cmma_log) #define KVM_S390_SET_CMMA_BITS _IOW(KVMIO, 0xb9, struct kvm_s390_cmma_log) #define KVM_DEV_ASSIGN_ENABLE_IOMMU (1 << 0) -- cgit v1.2.3 From 0791e3644e5ef21646fe565b9061788d05ec71d4 Mon Sep 17 00:00:00 2001 From: Cyrill Gorcunov Date: Wed, 12 Jul 2017 14:34:28 -0700 Subject: kcmp: add KCMP_EPOLL_TFD mode to compare epoll target files With current epoll architecture target files are addressed with file_struct and file descriptor number, where the last is not unique. Moreover files can be transferred from another process via unix socket, added into queue and closed then so we won't find this descriptor in the task fdinfo list. Thus to checkpoint and restore such processes CRIU needs to find out where exactly the target file is present to add it into epoll queue. For this sake one can use kcmp call where some particular target file from the queue is compared with arbitrary file passed as an argument. Because epoll target files can have same file descriptor number but different file_struct a caller should explicitly specify the offset within. To test if some particular file is matching entry inside epoll one have to - fill kcmp_epoll_slot structure with epoll file descriptor, target file number and target file offset (in case if only one target is present then it should be 0) - call kcmp as kcmp(pid1, pid2, KCMP_EPOLL_TFD, fd, &kcmp_epoll_slot) - the kernel fetch file pointer matching file descriptor @fd of pid1 - lookups for file struct in epoll queue of pid2 and returns traditional 0,1,2 result for sorting purpose Link: http://lkml.kernel.org/r/20170424154423.511592110@gmail.com Signed-off-by: Cyrill Gorcunov Acked-by: Andrey Vagin Cc: Al Viro Cc: Pavel Emelyanov Cc: Michael Kerrisk Cc: Jason Baron Cc: Andy Lutomirski Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/uapi/linux/kcmp.h | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'include/uapi/linux') diff --git a/include/uapi/linux/kcmp.h b/include/uapi/linux/kcmp.h index 84df14b37360..481e103da78e 100644 --- a/include/uapi/linux/kcmp.h +++ b/include/uapi/linux/kcmp.h @@ -1,6 +1,8 @@ #ifndef _UAPI_LINUX_KCMP_H #define _UAPI_LINUX_KCMP_H +#include + /* Comparison type */ enum kcmp_type { KCMP_FILE, @@ -10,8 +12,16 @@ enum kcmp_type { KCMP_SIGHAND, KCMP_IO, KCMP_SYSVSEM, + KCMP_EPOLL_TFD, KCMP_TYPES, }; +/* Slot for KCMP_EPOLL_TFD */ +struct kcmp_epoll_slot { + __u32 efd; /* epoll file descriptor */ + __u32 tfd; /* target file number */ + __u32 toff; /* target offset within same numbered sequence */ +}; + #endif /* _UAPI_LINUX_KCMP_H */ -- cgit v1.2.3 From 2cd648c110b5570c3280bd645797658cabbe5f5c Mon Sep 17 00:00:00 2001 From: Manfred Spraul Date: Wed, 12 Jul 2017 14:34:44 -0700 Subject: include/linux/sem.h: correctly document sem_ctime sem_ctime is initialized to the semget() time and then updated at every semctl() that changes the array. Thus it does not represent the time of the last change. Especially, semop() calls are only stored in sem_otime, not in sem_ctime. This is already described in ipc/sem.c, I just overlooked that there is a comment in include/linux/sem.h and man semctl(2) as well. So: Correct wrong comments. Link: http://lkml.kernel.org/r/20170515171912.6298-4-manfred@colorfullife.com Signed-off-by: Manfred Spraul Cc: Kees Cook Cc: <1vier1@web.de> Cc: Davidlohr Bueso Cc: Ingo Molnar Cc: Peter Zijlstra Cc: Fabian Frederick Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/uapi/linux/sem.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/uapi/linux') diff --git a/include/uapi/linux/sem.h b/include/uapi/linux/sem.h index dd73b908b2f3..67eb90361692 100644 --- a/include/uapi/linux/sem.h +++ b/include/uapi/linux/sem.h @@ -23,7 +23,7 @@ struct semid_ds { struct ipc_perm sem_perm; /* permissions .. see ipc.h */ __kernel_time_t sem_otime; /* last semop time */ - __kernel_time_t sem_ctime; /* last change time */ + __kernel_time_t sem_ctime; /* create/last semctl() time */ struct sem *sem_base; /* ptr to first semaphore in array */ struct sem_queue *sem_pending; /* pending operations to be processed */ struct sem_queue **sem_pending_last; /* last pending operation */ -- cgit v1.2.3 From efc479e6900c22bad9a2b649d13405ed9cde2d53 Mon Sep 17 00:00:00 2001 From: Roman Kagan Date: Thu, 22 Jun 2017 16:51:01 +0300 Subject: kvm: x86: hyperv: add KVM_CAP_HYPERV_SYNIC2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is a flaw in the Hyper-V SynIC implementation in KVM: when message page or event flags page is enabled by setting the corresponding msr, KVM zeroes it out. This is problematic because on migration the corresponding MSRs are loaded on the destination, so the content of those pages is lost. This went unnoticed so far because the only user of those pages was in-KVM hyperv synic timers, which could continue working despite that zeroing. Newer QEMU uses those pages for Hyper-V VMBus implementation, and zeroing them breaks the migration. Besides, in newer QEMU the content of those pages is fully managed by QEMU, so zeroing them is undesirable even when writing the MSRs from the guest side. To support this new scheme, introduce a new capability, KVM_CAP_HYPERV_SYNIC2, which, when enabled, makes sure that the synic pages aren't zeroed out in KVM. Signed-off-by: Roman Kagan Signed-off-by: Radim Krčmář --- include/uapi/linux/kvm.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/uapi/linux') diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index ebd604c222d8..38b2cfbc8112 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -927,6 +927,7 @@ struct kvm_ppc_resize_hpt { #define KVM_CAP_S390_CMMA_MIGRATION 145 #define KVM_CAP_PPC_FWNMI 146 #define KVM_CAP_PPC_SMT_POSSIBLE 147 +#define KVM_CAP_HYPERV_SYNIC2 148 #ifdef KVM_CAP_IRQ_ROUTING -- cgit v1.2.3 From d3457c877b14aaee8c52923eedf05a3b78af0476 Mon Sep 17 00:00:00 2001 From: Roman Kagan Date: Fri, 14 Jul 2017 17:13:20 +0300 Subject: kvm: x86: hyperv: make VP_INDEX managed by userspace MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Hyper-V identifies vCPUs by Virtual Processor Index, which can be queried via HV_X64_MSR_VP_INDEX msr. It is defined by the spec as a sequential number which can't exceed the maximum number of vCPUs per VM. APIC ids can be sparse and thus aren't a valid replacement for VP indices. Current KVM uses its internal vcpu index as VP_INDEX. However, to make it predictable and persistent across VM migrations, the userspace has to control the value of VP_INDEX. This patch achieves that, by storing vp_index explicitly on vcpu, and allowing HV_X64_MSR_VP_INDEX to be set from the host side. For compatibility it's initialized to KVM vcpu index. Also a few variables are renamed to make clear distinction betweed this Hyper-V vp_index and KVM vcpu_id (== APIC id). Besides, a new capability, KVM_CAP_HYPERV_VP_INDEX, is added to allow the userspace to skip attempting msr writes where unsupported, to avoid spamming error logs. Signed-off-by: Roman Kagan Signed-off-by: Radim Krčmář --- include/uapi/linux/kvm.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/uapi/linux') diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index 38b2cfbc8112..6cd63c18708a 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -928,6 +928,7 @@ struct kvm_ppc_resize_hpt { #define KVM_CAP_PPC_FWNMI 146 #define KVM_CAP_PPC_SMT_POSSIBLE 147 #define KVM_CAP_HYPERV_SYNIC2 148 +#define KVM_CAP_HYPERV_VP_INDEX 149 #ifdef KVM_CAP_IRQ_ROUTING -- cgit v1.2.3