From 5d5de3a431d87ac51d43da8d796891d014975ab7 Mon Sep 17 00:00:00 2001 From: Hou Tao Date: Thu, 16 Feb 2023 10:48:21 +0800 Subject: bpf: Only allocate one bpf_mem_cache for bpf_cpumask_ma The size of bpf_cpumask is fixed, so there is no need to allocate many bpf_mem_caches for bpf_cpumask_ma, just one bpf_mem_cache is enough. Also add comments for bpf_mem_alloc_init() in bpf_mem_alloc.h to prevent future miuse. Signed-off-by: Hou Tao Acked-by: Jiri Olsa Link: https://lore.kernel.org/r/20230216024821.2202916-1-houtao@huaweicloud.com Signed-off-by: Alexei Starovoitov --- include/linux/bpf_mem_alloc.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/linux') diff --git a/include/linux/bpf_mem_alloc.h b/include/linux/bpf_mem_alloc.h index 3e164b8efaa9..a7104af61ab4 100644 --- a/include/linux/bpf_mem_alloc.h +++ b/include/linux/bpf_mem_alloc.h @@ -14,6 +14,13 @@ struct bpf_mem_alloc { struct work_struct work; }; +/* 'size != 0' is for bpf_mem_alloc which manages fixed-size objects. + * Alloc and free are done with bpf_mem_cache_{alloc,free}(). + * + * 'size = 0' is for bpf_mem_alloc which manages many fixed-size objects. + * Alloc and free are done with bpf_mem_{alloc,free}() and the size of + * the returned object is given by the size argument of bpf_mem_alloc(). + */ int bpf_mem_alloc_init(struct bpf_mem_alloc *ma, int size, bool percpu); void bpf_mem_alloc_destroy(struct bpf_mem_alloc *ma); -- cgit v1.2.3 From d5f7638eb5fed0eb12e45a127764c4111b11c50e Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 24 Oct 2022 05:34:14 -0700 Subject: Input: matrix_keypad - replace header inclusions by forward declarations When the data structure is only referred by pointer, compiler may not need to see the contents of the data type. Thus, we may replace header inclusions by respective forward declarations. Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/r/20220923184632.2157-2-andriy.shevchenko@linux.intel.com Signed-off-by: Dmitry Torokhov --- include/linux/input/matrix_keypad.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/input/matrix_keypad.h b/include/linux/input/matrix_keypad.h index 9476768c3b90..b8d8d69eba29 100644 --- a/include/linux/input/matrix_keypad.h +++ b/include/linux/input/matrix_keypad.h @@ -3,8 +3,9 @@ #define _MATRIX_KEYPAD_H #include -#include -#include + +struct device; +struct input_dev; #define MATRIX_MAX_ROWS 32 #define MATRIX_MAX_COLS 32 -- cgit v1.2.3 From 7e0dac2807e6c4ae8c56941d74971fdb0763b4f9 Mon Sep 17 00:00:00 2001 From: Joanne Koong Date: Wed, 1 Mar 2023 07:49:45 -0800 Subject: bpf: Refactor process_dynptr_func This change cleans up process_dynptr_func's flow to be more intuitive and updates some comments with more context. Signed-off-by: Joanne Koong Link: https://lore.kernel.org/r/20230301154953.641654-3-joannelkoong@gmail.com Signed-off-by: Alexei Starovoitov --- include/linux/bpf_verifier.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h index cf1bb1cf4a7b..b26ff2a8f63b 100644 --- a/include/linux/bpf_verifier.h +++ b/include/linux/bpf_verifier.h @@ -616,9 +616,6 @@ int check_func_arg_reg_off(struct bpf_verifier_env *env, enum bpf_arg_type arg_type); int check_mem_reg(struct bpf_verifier_env *env, struct bpf_reg_state *reg, u32 regno, u32 mem_size); -struct bpf_call_arg_meta; -int process_dynptr_func(struct bpf_verifier_env *env, int regno, - enum bpf_arg_type arg_type, struct bpf_call_arg_meta *meta); /* this lives here instead of in bpf.h because it needs to dereference tgt_prog */ static inline u64 bpf_trampoline_compute_key(const struct bpf_prog *tgt_prog, -- cgit v1.2.3 From 8357b366cbb09b17c90e2cd758360a6bd2ea7507 Mon Sep 17 00:00:00 2001 From: Joanne Koong Date: Wed, 1 Mar 2023 07:49:47 -0800 Subject: bpf: Define no-ops for externally called bpf dynptr functions Some bpf dynptr functions will be called from places where if CONFIG_BPF_SYSCALL is not set, then the dynptr function is undefined. For example, when skb type dynptrs are added in the next commit, dynptr functions are called from net/core/filter.c This patch defines no-op implementations of these dynptr functions so that they do not break compilation by being an undefined reference. Signed-off-by: Joanne Koong Link: https://lore.kernel.org/r/20230301154953.641654-5-joannelkoong@gmail.com Signed-off-by: Alexei Starovoitov --- include/linux/bpf.h | 75 ++++++++++++++++++++++++++++++++--------------------- 1 file changed, 45 insertions(+), 30 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 520b238abd5a..296841a31749 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -1124,6 +1124,33 @@ static __always_inline __nocfi unsigned int bpf_dispatcher_nop_func( return bpf_func(ctx, insnsi); } +/* the implementation of the opaque uapi struct bpf_dynptr */ +struct bpf_dynptr_kern { + void *data; + /* Size represents the number of usable bytes of dynptr data. + * If for example the offset is at 4 for a local dynptr whose data is + * of type u64, the number of usable bytes is 4. + * + * The upper 8 bits are reserved. It is as follows: + * Bits 0 - 23 = size + * Bits 24 - 30 = dynptr type + * Bit 31 = whether dynptr is read-only + */ + u32 size; + u32 offset; +} __aligned(8); + +enum bpf_dynptr_type { + BPF_DYNPTR_TYPE_INVALID, + /* Points to memory that is local to the bpf program */ + BPF_DYNPTR_TYPE_LOCAL, + /* Underlying data is a ringbuf record */ + BPF_DYNPTR_TYPE_RINGBUF, +}; + +int bpf_dynptr_check_size(u32 size); +u32 bpf_dynptr_get_size(const struct bpf_dynptr_kern *ptr); + #ifdef CONFIG_BPF_JIT int bpf_trampoline_link_prog(struct bpf_tramp_link *link, struct bpf_trampoline *tr); int bpf_trampoline_unlink_prog(struct bpf_tramp_link *link, struct bpf_trampoline *tr); @@ -2266,6 +2293,11 @@ static inline bool has_current_bpf_ctx(void) } void notrace bpf_prog_inc_misses_counter(struct bpf_prog *prog); + +void bpf_dynptr_init(struct bpf_dynptr_kern *ptr, void *data, + enum bpf_dynptr_type type, u32 offset, u32 size); +void bpf_dynptr_set_null(struct bpf_dynptr_kern *ptr); +void bpf_dynptr_set_rdonly(struct bpf_dynptr_kern *ptr); #else /* !CONFIG_BPF_SYSCALL */ static inline struct bpf_prog *bpf_prog_get(u32 ufd) { @@ -2495,6 +2527,19 @@ static inline void bpf_prog_inc_misses_counter(struct bpf_prog *prog) static inline void bpf_cgrp_storage_free(struct cgroup *cgroup) { } + +static inline void bpf_dynptr_init(struct bpf_dynptr_kern *ptr, void *data, + enum bpf_dynptr_type type, u32 offset, u32 size) +{ +} + +static inline void bpf_dynptr_set_null(struct bpf_dynptr_kern *ptr) +{ +} + +static inline void bpf_dynptr_set_rdonly(struct bpf_dynptr_kern *ptr) +{ +} #endif /* CONFIG_BPF_SYSCALL */ void __bpf_free_used_btfs(struct bpf_prog_aux *aux, @@ -2913,36 +2958,6 @@ int bpf_bprintf_prepare(char *fmt, u32 fmt_size, const u64 *raw_args, u32 num_args, struct bpf_bprintf_data *data); void bpf_bprintf_cleanup(struct bpf_bprintf_data *data); -/* the implementation of the opaque uapi struct bpf_dynptr */ -struct bpf_dynptr_kern { - void *data; - /* Size represents the number of usable bytes of dynptr data. - * If for example the offset is at 4 for a local dynptr whose data is - * of type u64, the number of usable bytes is 4. - * - * The upper 8 bits are reserved. It is as follows: - * Bits 0 - 23 = size - * Bits 24 - 30 = dynptr type - * Bit 31 = whether dynptr is read-only - */ - u32 size; - u32 offset; -} __aligned(8); - -enum bpf_dynptr_type { - BPF_DYNPTR_TYPE_INVALID, - /* Points to memory that is local to the bpf program */ - BPF_DYNPTR_TYPE_LOCAL, - /* Underlying data is a kernel-produced ringbuf record */ - BPF_DYNPTR_TYPE_RINGBUF, -}; - -void bpf_dynptr_init(struct bpf_dynptr_kern *ptr, void *data, - enum bpf_dynptr_type type, u32 offset, u32 size); -void bpf_dynptr_set_null(struct bpf_dynptr_kern *ptr); -int bpf_dynptr_check_size(u32 size); -u32 bpf_dynptr_get_size(const struct bpf_dynptr_kern *ptr); - #ifdef CONFIG_BPF_LSM void bpf_cgroup_atype_get(u32 attach_btf_id, int cgroup_atype); void bpf_cgroup_atype_put(int cgroup_atype); -- cgit v1.2.3 From b5964b968ac64c2ec2debee7518499113b27c34e Mon Sep 17 00:00:00 2001 From: Joanne Koong Date: Wed, 1 Mar 2023 07:49:50 -0800 Subject: bpf: Add skb dynptrs Add skb dynptrs, which are dynptrs whose underlying pointer points to a skb. The dynptr acts on skb data. skb dynptrs have two main benefits. One is that they allow operations on sizes that are not statically known at compile-time (eg variable-sized accesses). Another is that parsing the packet data through dynptrs (instead of through direct access of skb->data and skb->data_end) can be more ergonomic and less brittle (eg does not need manual if checking for being within bounds of data_end). For bpf prog types that don't support writes on skb data, the dynptr is read-only (bpf_dynptr_write() will return an error) For reads and writes through the bpf_dynptr_read() and bpf_dynptr_write() interfaces, reading and writing from/to data in the head as well as from/to non-linear paged buffers is supported. Data slices through the bpf_dynptr_data API are not supported; instead bpf_dynptr_slice() and bpf_dynptr_slice_rdwr() (added in subsequent commit) should be used. For examples of how skb dynptrs can be used, please see the attached selftests. Signed-off-by: Joanne Koong Link: https://lore.kernel.org/r/20230301154953.641654-8-joannelkoong@gmail.com Signed-off-by: Alexei Starovoitov --- include/linux/bpf.h | 14 +++++++++++++- include/linux/filter.h | 18 ++++++++++++++++++ 2 files changed, 31 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 296841a31749..e7436d7615b0 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -607,11 +607,14 @@ enum bpf_type_flag { */ NON_OWN_REF = BIT(14 + BPF_BASE_TYPE_BITS), + /* DYNPTR points to sk_buff */ + DYNPTR_TYPE_SKB = BIT(15 + BPF_BASE_TYPE_BITS), + __BPF_TYPE_FLAG_MAX, __BPF_TYPE_LAST_FLAG = __BPF_TYPE_FLAG_MAX - 1, }; -#define DYNPTR_TYPE_FLAG_MASK (DYNPTR_TYPE_LOCAL | DYNPTR_TYPE_RINGBUF) +#define DYNPTR_TYPE_FLAG_MASK (DYNPTR_TYPE_LOCAL | DYNPTR_TYPE_RINGBUF | DYNPTR_TYPE_SKB) /* Max number of base types. */ #define BPF_BASE_TYPE_LIMIT (1UL << BPF_BASE_TYPE_BITS) @@ -1146,6 +1149,8 @@ enum bpf_dynptr_type { BPF_DYNPTR_TYPE_LOCAL, /* Underlying data is a ringbuf record */ BPF_DYNPTR_TYPE_RINGBUF, + /* Underlying data is a sk_buff */ + BPF_DYNPTR_TYPE_SKB, }; int bpf_dynptr_check_size(u32 size); @@ -2846,6 +2851,8 @@ u32 bpf_sock_convert_ctx_access(enum bpf_access_type type, struct bpf_insn *insn_buf, struct bpf_prog *prog, u32 *target_size); +int bpf_dynptr_from_skb_rdonly(struct sk_buff *skb, u64 flags, + struct bpf_dynptr_kern *ptr); #else static inline bool bpf_sock_common_is_valid_access(int off, int size, enum bpf_access_type type, @@ -2867,6 +2874,11 @@ static inline u32 bpf_sock_convert_ctx_access(enum bpf_access_type type, { return 0; } +static inline int bpf_dynptr_from_skb_rdonly(struct sk_buff *skb, u64 flags, + struct bpf_dynptr_kern *ptr) +{ + return -EOPNOTSUPP; +} #endif #ifdef CONFIG_INET diff --git a/include/linux/filter.h b/include/linux/filter.h index 1727898f1641..de18e844d15e 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -1542,4 +1542,22 @@ static __always_inline int __bpf_xdp_redirect_map(struct bpf_map *map, u64 index return XDP_REDIRECT; } +#ifdef CONFIG_NET +int __bpf_skb_load_bytes(const struct sk_buff *skb, u32 offset, void *to, u32 len); +int __bpf_skb_store_bytes(struct sk_buff *skb, u32 offset, const void *from, + u32 len, u64 flags); +#else /* CONFIG_NET */ +static inline int __bpf_skb_load_bytes(const struct sk_buff *skb, u32 offset, + void *to, u32 len) +{ + return -EOPNOTSUPP; +} + +static inline int __bpf_skb_store_bytes(struct sk_buff *skb, u32 offset, + const void *from, u32 len, u64 flags) +{ + return -EOPNOTSUPP; +} +#endif /* CONFIG_NET */ + #endif /* __LINUX_FILTER_H__ */ -- cgit v1.2.3 From 05421aecd4ed65da0dc17b0c3c13779ef334e9e5 Mon Sep 17 00:00:00 2001 From: Joanne Koong Date: Wed, 1 Mar 2023 07:49:51 -0800 Subject: bpf: Add xdp dynptrs Add xdp dynptrs, which are dynptrs whose underlying pointer points to a xdp_buff. The dynptr acts on xdp data. xdp dynptrs have two main benefits. One is that they allow operations on sizes that are not statically known at compile-time (eg variable-sized accesses). Another is that parsing the packet data through dynptrs (instead of through direct access of xdp->data and xdp->data_end) can be more ergonomic and less brittle (eg does not need manual if checking for being within bounds of data_end). For reads and writes on the dynptr, this includes reading/writing from/to and across fragments. Data slices through the bpf_dynptr_data API are not supported; instead bpf_dynptr_slice() and bpf_dynptr_slice_rdwr() should be used. For examples of how xdp dynptrs can be used, please see the attached selftests. Signed-off-by: Joanne Koong Link: https://lore.kernel.org/r/20230301154953.641654-9-joannelkoong@gmail.com Signed-off-by: Alexei Starovoitov --- include/linux/bpf.h | 8 +++++++- include/linux/filter.h | 14 ++++++++++++++ 2 files changed, 21 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index e7436d7615b0..23ec684e660d 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -610,11 +610,15 @@ enum bpf_type_flag { /* DYNPTR points to sk_buff */ DYNPTR_TYPE_SKB = BIT(15 + BPF_BASE_TYPE_BITS), + /* DYNPTR points to xdp_buff */ + DYNPTR_TYPE_XDP = BIT(16 + BPF_BASE_TYPE_BITS), + __BPF_TYPE_FLAG_MAX, __BPF_TYPE_LAST_FLAG = __BPF_TYPE_FLAG_MAX - 1, }; -#define DYNPTR_TYPE_FLAG_MASK (DYNPTR_TYPE_LOCAL | DYNPTR_TYPE_RINGBUF | DYNPTR_TYPE_SKB) +#define DYNPTR_TYPE_FLAG_MASK (DYNPTR_TYPE_LOCAL | DYNPTR_TYPE_RINGBUF | DYNPTR_TYPE_SKB \ + | DYNPTR_TYPE_XDP) /* Max number of base types. */ #define BPF_BASE_TYPE_LIMIT (1UL << BPF_BASE_TYPE_BITS) @@ -1151,6 +1155,8 @@ enum bpf_dynptr_type { BPF_DYNPTR_TYPE_RINGBUF, /* Underlying data is a sk_buff */ BPF_DYNPTR_TYPE_SKB, + /* Underlying data is a xdp_buff */ + BPF_DYNPTR_TYPE_XDP, }; int bpf_dynptr_check_size(u32 size); diff --git a/include/linux/filter.h b/include/linux/filter.h index de18e844d15e..3f6992261ec5 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -1546,6 +1546,8 @@ static __always_inline int __bpf_xdp_redirect_map(struct bpf_map *map, u64 index int __bpf_skb_load_bytes(const struct sk_buff *skb, u32 offset, void *to, u32 len); int __bpf_skb_store_bytes(struct sk_buff *skb, u32 offset, const void *from, u32 len, u64 flags); +int __bpf_xdp_load_bytes(struct xdp_buff *xdp, u32 offset, void *buf, u32 len); +int __bpf_xdp_store_bytes(struct xdp_buff *xdp, u32 offset, void *buf, u32 len); #else /* CONFIG_NET */ static inline int __bpf_skb_load_bytes(const struct sk_buff *skb, u32 offset, void *to, u32 len) @@ -1558,6 +1560,18 @@ static inline int __bpf_skb_store_bytes(struct sk_buff *skb, u32 offset, { return -EOPNOTSUPP; } + +static inline int __bpf_xdp_load_bytes(struct xdp_buff *xdp, u32 offset, + void *buf, u32 len) +{ + return -EOPNOTSUPP; +} + +static inline int __bpf_xdp_store_bytes(struct xdp_buff *xdp, u32 offset, + void *buf, u32 len) +{ + return -EOPNOTSUPP; +} #endif /* CONFIG_NET */ #endif /* __LINUX_FILTER_H__ */ -- cgit v1.2.3 From 66e3a13e7c2c44d0c9dd6bb244680ca7529a8845 Mon Sep 17 00:00:00 2001 From: Joanne Koong Date: Wed, 1 Mar 2023 07:49:52 -0800 Subject: bpf: Add bpf_dynptr_slice and bpf_dynptr_slice_rdwr Two new kfuncs are added, bpf_dynptr_slice and bpf_dynptr_slice_rdwr. The user must pass in a buffer to store the contents of the data slice if a direct pointer to the data cannot be obtained. For skb and xdp type dynptrs, these two APIs are the only way to obtain a data slice. However, for other types of dynptrs, there is no difference between bpf_dynptr_slice(_rdwr) and bpf_dynptr_data. For skb type dynptrs, the data is copied into the user provided buffer if any of the data is not in the linear portion of the skb. For xdp type dynptrs, the data is copied into the user provided buffer if the data is between xdp frags. If the skb is cloned and a call to bpf_dynptr_data_rdwr is made, then the skb will be uncloned (see bpf_unclone_prologue()). Please note that any bpf_dynptr_write() automatically invalidates any prior data slices of the skb dynptr. This is because the skb may be cloned or may need to pull its paged buffer into the head. As such, any bpf_dynptr_write() will automatically have its prior data slices invalidated, even if the write is to data in the skb head of an uncloned skb. Please note as well that any other helper calls that change the underlying packet buffer (eg bpf_skb_pull_data()) invalidates any data slices of the skb dynptr as well, for the same reasons. Signed-off-by: Joanne Koong Link: https://lore.kernel.org/r/20230301154953.641654-10-joannelkoong@gmail.com Signed-off-by: Alexei Starovoitov --- include/linux/filter.h | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'include/linux') diff --git a/include/linux/filter.h b/include/linux/filter.h index 3f6992261ec5..efa5d4a1677e 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -1548,6 +1548,9 @@ int __bpf_skb_store_bytes(struct sk_buff *skb, u32 offset, const void *from, u32 len, u64 flags); int __bpf_xdp_load_bytes(struct xdp_buff *xdp, u32 offset, void *buf, u32 len); int __bpf_xdp_store_bytes(struct xdp_buff *xdp, u32 offset, void *buf, u32 len); +void *bpf_xdp_pointer(struct xdp_buff *xdp, u32 offset, u32 len); +void bpf_xdp_copy_buf(struct xdp_buff *xdp, unsigned long off, + void *buf, unsigned long len, bool flush); #else /* CONFIG_NET */ static inline int __bpf_skb_load_bytes(const struct sk_buff *skb, u32 offset, void *to, u32 len) @@ -1572,6 +1575,17 @@ static inline int __bpf_xdp_store_bytes(struct xdp_buff *xdp, u32 offset, { return -EOPNOTSUPP; } + +static inline void *bpf_xdp_pointer(struct xdp_buff *xdp, u32 offset, u32 len) +{ + return NULL; +} + +static inline void *bpf_xdp_copy_buf(struct xdp_buff *xdp, unsigned long off, void *buf, + unsigned long len, bool flush) +{ + return NULL; +} #endif /* CONFIG_NET */ #endif /* __LINUX_FILTER_H__ */ -- cgit v1.2.3 From 9db44fdd8105da00669d425acab887c668df75f6 Mon Sep 17 00:00:00 2001 From: Kumar Kartikeya Dwivedi Date: Sat, 25 Feb 2023 16:40:09 +0100 Subject: bpf: Support kptrs in local storage maps Enable support for kptrs in local storage maps by wiring up the freeing of these kptrs from map value. Freeing of bpf_local_storage_map is only delayed in case there are special fields, therefore bpf_selem_free_* path can also only dereference smap safely in that case. This is recorded using a bool utilizing a hole in bpF_local_storage_elem. It could have been tagged in the pointer value smap using the lowest bit (since alignment > 1), but since there was already a hole I went with the simpler option. Only the map structure freeing is delayed using RCU barriers, as the buckets aren't used when selem is being freed, so they can be freed once all readers of the bucket lists can no longer access it. Cc: Martin KaFai Lau Cc: KP Singh Cc: Paul E. McKenney Signed-off-by: Kumar Kartikeya Dwivedi Link: https://lore.kernel.org/r/20230225154010.391965-3-memxor@gmail.com Signed-off-by: Alexei Starovoitov --- include/linux/bpf_local_storage.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include/linux') diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h index 6d37a40cd90e..0fe92986412b 100644 --- a/include/linux/bpf_local_storage.h +++ b/include/linux/bpf_local_storage.h @@ -74,6 +74,12 @@ struct bpf_local_storage_elem { struct hlist_node snode; /* Linked to bpf_local_storage */ struct bpf_local_storage __rcu *local_storage; struct rcu_head rcu; + bool can_use_smap; /* Is it safe to access smap in bpf_selem_free_* RCU + * callbacks? bpf_local_storage_map_free only + * executes rcu_barrier when there are special + * fields, this field remembers that to ensure we + * don't access already freed smap in sdata. + */ /* 8 bytes hole */ /* The data is stored in another cacheline to minimize * the number of cachelines access during a cache hit. -- cgit v1.2.3 From 20c09d92faeefb8536f705d3a4629e0dc314c8a1 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Thu, 2 Mar 2023 20:14:43 -0800 Subject: bpf: Introduce kptr_rcu. The life time of certain kernel structures like 'struct cgroup' is protected by RCU. Hence it's safe to dereference them directly from __kptr tagged pointers in bpf maps. The resulting pointer is MEM_RCU and can be passed to kfuncs that expect KF_RCU. Derefrence of other kptr-s returns PTR_UNTRUSTED. For example: struct map_value { struct cgroup __kptr *cgrp; }; SEC("tp_btf/cgroup_mkdir") int BPF_PROG(test_cgrp_get_ancestors, struct cgroup *cgrp_arg, const char *path) { struct cgroup *cg, *cg2; cg = bpf_cgroup_acquire(cgrp_arg); // cg is PTR_TRUSTED and ref_obj_id > 0 bpf_kptr_xchg(&v->cgrp, cg); cg2 = v->cgrp; // This is new feature introduced by this patch. // cg2 is PTR_MAYBE_NULL | MEM_RCU. // When cg2 != NULL, it's a valid cgroup, but its percpu_ref could be zero if (cg2) bpf_cgroup_ancestor(cg2, level); // safe to do. } Signed-off-by: Alexei Starovoitov Signed-off-by: Daniel Borkmann Acked-by: Tejun Heo Acked-by: David Vernet Link: https://lore.kernel.org/bpf/20230303041446.3630-4-alexei.starovoitov@gmail.com --- include/linux/btf.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/btf.h b/include/linux/btf.h index 49e0fe6d8274..556b3e2e7471 100644 --- a/include/linux/btf.h +++ b/include/linux/btf.h @@ -70,7 +70,7 @@ #define KF_TRUSTED_ARGS (1 << 4) /* kfunc only takes trusted pointer arguments */ #define KF_SLEEPABLE (1 << 5) /* kfunc may sleep */ #define KF_DESTRUCTIVE (1 << 6) /* kfunc performs destructive actions */ -#define KF_RCU (1 << 7) /* kfunc only takes rcu pointer arguments */ +#define KF_RCU (1 << 7) /* kfunc takes either rcu or trusted pointer arguments */ /* * Tag marking a kernel function as a kfunc. This is meant to minimize the -- cgit v1.2.3 From 6fcd486b3a0a628c41f12b3a7329a18a2c74b351 Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Thu, 2 Mar 2023 20:14:46 -0800 Subject: bpf: Refactor RCU enforcement in the verifier. bpf_rcu_read_lock/unlock() are only available in clang compiled kernels. Lack of such key mechanism makes it impossible for sleepable bpf programs to use RCU pointers. Allow bpf_rcu_read_lock/unlock() in GCC compiled kernels (though GCC doesn't support btf_type_tag yet) and allowlist certain field dereferences in important data structures like tast_struct, cgroup, socket that are used by sleepable programs either as RCU pointer or full trusted pointer (which is valid outside of RCU CS). Use BTF_TYPE_SAFE_RCU and BTF_TYPE_SAFE_TRUSTED macros for such tagging. They will be removed once GCC supports btf_type_tag. With that refactor check_ptr_to_btf_access(). Make it strict in enforcing PTR_TRUSTED and PTR_UNTRUSTED while deprecating old PTR_TO_BTF_ID without modifier flags. There is a chance that this strict enforcement might break existing programs (especially on GCC compiled kernels), but this cleanup has to start sooner than later. Note PTR_TO_CTX access still yields old deprecated PTR_TO_BTF_ID. Once it's converted to strict PTR_TRUSTED or PTR_UNTRUSTED the kfuncs and helpers will be able to default to KF_TRUSTED_ARGS. KF_RCU will remain as a weaker version of KF_TRUSTED_ARGS where obj refcnt could be 0. Adjust rcu_read_lock selftest to run on gcc and clang compiled kernels. Signed-off-by: Alexei Starovoitov Signed-off-by: Daniel Borkmann Acked-by: David Vernet Link: https://lore.kernel.org/bpf/20230303041446.3630-7-alexei.starovoitov@gmail.com --- include/linux/bpf.h | 2 +- include/linux/bpf_verifier.h | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 23ec684e660d..d3456804f7aa 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -2279,7 +2279,7 @@ struct bpf_core_ctx { bool btf_nested_type_is_trusted(struct bpf_verifier_log *log, const struct bpf_reg_state *reg, - int off); + int off, const char *suffix); bool btf_type_ids_nocast_alias(struct bpf_verifier_log *log, const struct btf *reg_btf, u32 reg_id, diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h index b26ff2a8f63b..18538bad2b8c 100644 --- a/include/linux/bpf_verifier.h +++ b/include/linux/bpf_verifier.h @@ -537,7 +537,6 @@ struct bpf_verifier_env { bool bypass_spec_v1; bool bypass_spec_v4; bool seen_direct_write; - bool rcu_tag_supported; struct bpf_insn_aux_data *insn_aux_data; /* array of per-insn state */ const struct bpf_line_info *prev_linfo; struct bpf_verifier_log log; -- cgit v1.2.3 From e768e3c5aab44ee63f58649d4c8cbbb3270e5c06 Mon Sep 17 00:00:00 2001 From: Kumar Kartikeya Dwivedi Date: Fri, 3 Mar 2023 15:15:42 +0100 Subject: bpf: Use separate RCU callbacks for freeing selem Martin suggested that instead of using a byte in the hole (which he has a use for in his future patch) in bpf_local_storage_elem, we can dispatch a different call_rcu callback based on whether we need to free special fields in bpf_local_storage_elem data. The free path, described in commit 9db44fdd8105 ("bpf: Support kptrs in local storage maps"), only waits for call_rcu callbacks when there are special (kptrs, etc.) fields in the map value, hence it is necessary that we only access smap in this case. Therefore, dispatch different RCU callbacks based on the BPF map has a valid btf_record, which dereference and use smap's btf_record only when it is valid. Signed-off-by: Kumar Kartikeya Dwivedi Link: https://lore.kernel.org/r/20230303141542.300068-1-memxor@gmail.com Signed-off-by: Martin KaFai Lau --- include/linux/bpf_local_storage.h | 6 ------ 1 file changed, 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf_local_storage.h b/include/linux/bpf_local_storage.h index 0fe92986412b..6d37a40cd90e 100644 --- a/include/linux/bpf_local_storage.h +++ b/include/linux/bpf_local_storage.h @@ -74,12 +74,6 @@ struct bpf_local_storage_elem { struct hlist_node snode; /* Linked to bpf_local_storage */ struct bpf_local_storage __rcu *local_storage; struct rcu_head rcu; - bool can_use_smap; /* Is it safe to access smap in bpf_selem_free_* RCU - * callbacks? bpf_local_storage_map_free only - * executes rcu_barrier when there are special - * fields, this field remembers that to ensure we - * don't access already freed smap in sdata. - */ /* 8 bytes hole */ /* The data is stored in another cacheline to minimize * the number of cachelines access during a cache hit. -- cgit v1.2.3 From a6ff3c0021468721b96e84892a8cae24bde8d65f Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Wed, 1 Mar 2023 21:14:29 +0100 Subject: thermal/core: Add a thermal zone 'devdata' accessor The thermal zone device structure is exposed to the different drivers and obviously they access the internals while that should be restricted to the core thermal code. In order to self-encapsulate the thermal core code, we need to prevent the drivers accessing directly the thermal zone structure and provide accessor functions to deal with. Provide an accessor to the 'devdata' structure and make use of it in the different drivers. No functional changes intended. Signed-off-by: Daniel Lezcano Acked-by: Rafael J. Wysocki Acked-by: Mark Brown Signed-off-by: Rafael J. Wysocki --- include/linux/thermal.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/linux') diff --git a/include/linux/thermal.h b/include/linux/thermal.h index 2bb4bf33f4f3..7dbb5712434c 100644 --- a/include/linux/thermal.h +++ b/include/linux/thermal.h @@ -365,6 +365,8 @@ thermal_zone_device_register_with_trips(const char *, struct thermal_trip *, int void *, struct thermal_zone_device_ops *, struct thermal_zone_params *, int, int); +void *thermal_zone_device_priv(struct thermal_zone_device *tzd); + int thermal_zone_bind_cooling_device(struct thermal_zone_device *, int, struct thermal_cooling_device *, unsigned long, unsigned long, @@ -436,6 +438,11 @@ static inline int thermal_zone_get_offset( struct thermal_zone_device *tz) { return -ENODEV; } +static inline void *thermal_zone_device_priv(struct thermal_zone_device *tz) +{ + return NULL; +} + static inline int thermal_zone_device_enable(struct thermal_zone_device *tz) { return -ENODEV; } -- cgit v1.2.3 From 072e35c98806100182c0a7263cf4cba09ce43463 Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Wed, 1 Mar 2023 21:14:38 +0100 Subject: thermal/core: Add thermal_zone_device structure 'type' accessor The thermal zone device structure is exposed via the exported thermal.h header. This structure should stay private the thermal core code. In order to encapsulate the structure, let's add an accessor to get the 'type' of the thermal zone. Signed-off-by: Daniel Lezcano Signed-off-by: Rafael J. Wysocki --- include/linux/thermal.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include/linux') diff --git a/include/linux/thermal.h b/include/linux/thermal.h index 7dbb5712434c..21686e676b3d 100644 --- a/include/linux/thermal.h +++ b/include/linux/thermal.h @@ -366,6 +366,7 @@ thermal_zone_device_register_with_trips(const char *, struct thermal_trip *, int struct thermal_zone_params *, int, int); void *thermal_zone_device_priv(struct thermal_zone_device *tzd); +const char *thermal_zone_device_type(struct thermal_zone_device *tzd); int thermal_zone_bind_cooling_device(struct thermal_zone_device *, int, struct thermal_cooling_device *, @@ -443,6 +444,11 @@ static inline void *thermal_zone_device_priv(struct thermal_zone_device *tz) return NULL; } +static inline const char *thermal_zone_device_type(struct thermal_zone_device *tzd) +{ + return NULL; +} + static inline int thermal_zone_device_enable(struct thermal_zone_device *tz) { return -ENODEV; } -- cgit v1.2.3 From 3034f859b90d49ee661fad4d23593589d0d37779 Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Wed, 1 Mar 2023 21:14:40 +0100 Subject: thermal: Add a thermal zone id accessor In order to get the thermal zone id but without directly accessing the thermal zone device structure, add an accessor. Use the accessor in the hwmon_scmi and acpi_thermal. No functional change intented. Signed-off-by: Daniel Lezcano Acked-by: Guenter Roeck Acked-by: Rafael J. Wysocki Signed-off-by: Rafael J. Wysocki --- include/linux/thermal.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include/linux') diff --git a/include/linux/thermal.h b/include/linux/thermal.h index 21686e676b3d..eb80cee4f64f 100644 --- a/include/linux/thermal.h +++ b/include/linux/thermal.h @@ -367,6 +367,7 @@ thermal_zone_device_register_with_trips(const char *, struct thermal_trip *, int void *thermal_zone_device_priv(struct thermal_zone_device *tzd); const char *thermal_zone_device_type(struct thermal_zone_device *tzd); +int thermal_zone_device_id(struct thermal_zone_device *tzd); int thermal_zone_bind_cooling_device(struct thermal_zone_device *, int, struct thermal_cooling_device *, @@ -449,6 +450,11 @@ static inline const char *thermal_zone_device_type(struct thermal_zone_device *t return NULL; } +static inline int thermal_zone_device_id(struct thermal_zone_device *tzd) +{ + return -ENODEV; +} + static inline int thermal_zone_device_enable(struct thermal_zone_device *tz) { return -ENODEV; } -- cgit v1.2.3 From 4d60cac951fd2dfbf261b83abb805748abc8b995 Mon Sep 17 00:00:00 2001 From: William Breathitt Gray Date: Mon, 27 Feb 2023 11:45:22 -0500 Subject: regmap-irq: Add no_status support Some devices lack status registers, yet expect to handle interrupts. Introduce a no_status flag to indicate such a configuration, where rather than read a status register to verify, all interrupts received are assumed to be active. Cc: Mark Brown Signed-off-by: William Breathitt Gray Link: https://lore.kernel.org/r/bd501b4b5ff88da24d467f75e8c71b4e0e6f21e2.1677515341.git.william.gray@linaro.org Signed-off-by: Mark Brown --- include/linux/regmap.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 4d10790adeb0..301286149643 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -1567,6 +1567,7 @@ struct regmap_irq_chip_data; * the need for a @sub_reg_offsets table. * @status_invert: Inverted status register: cleared bits are active interrupts. * @runtime_pm: Hold a runtime PM lock on the device when accessing it. + * @no_status: No status register: all interrupts assumed generated by device. * * @num_regs: Number of registers in each control bank. * @irqs: Descriptors for individual IRQs. Interrupt numbers are @@ -1631,6 +1632,7 @@ struct regmap_irq_chip { unsigned int clear_on_unmask:1; unsigned int not_fixed_stride:1; unsigned int status_invert:1; + unsigned int no_status:1; int num_regs; -- cgit v1.2.3 From 9b400171a69d2487c3196cc3b6de60de3b08e1ee Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 20 Feb 2023 17:33:34 +0200 Subject: regmap-irq: Place kernel doc of struct regmap_irq_chip in order It seems that a couple of members got lost theirorder, put them back. Besides that, split field descriptions into groups in the same way as it's done in the structure definition. Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/r/20230220153334.87049-1-andriy.shevchenko@linux.intel.com Signed-off-by: Mark Brown --- include/linux/regmap.h | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 4d10790adeb0..1c777566fb7d 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -1551,6 +1551,7 @@ struct regmap_irq_chip_data; * @use_ack: Use @ack register even if it is zero. * @ack_invert: Inverted ack register: cleared bits for ack. * @clear_ack: Use this to set 1 and 0 or vice-versa to clear interrupts. + * @status_invert: Inverted status register: cleared bits are active interrupts. * @wake_invert: Inverted wake register: cleared bits are wake enabled. * @type_in_mask: Use the mask registers for controlling irq type. Use this if * the hardware provides separate bits for rising/falling edge @@ -1560,18 +1561,19 @@ struct regmap_irq_chip_data; * @clear_on_unmask: For chips with interrupts cleared on read: read the status * registers before unmasking interrupts to clear any bits * set when they were masked. + * @runtime_pm: Hold a runtime PM lock on the device when accessing it. * @not_fixed_stride: Used when chip peripherals are not laid out with fixed * stride. Must be used with sub_reg_offsets containing the * offsets to each peripheral. Deprecated; the same thing * can be accomplished with a @get_irq_reg callback, without * the need for a @sub_reg_offsets table. - * @status_invert: Inverted status register: cleared bits are active interrupts. - * @runtime_pm: Hold a runtime PM lock on the device when accessing it. * * @num_regs: Number of registers in each control bank. + * * @irqs: Descriptors for individual IRQs. Interrupt numbers are * assigned based on the index in the array of the interrupt. * @num_irqs: Number of descriptors. + * * @num_type_reg: Number of type registers. Deprecated, use config registers * instead. * @num_virt_regs: Number of non-standard irq configuration registers. @@ -1579,6 +1581,7 @@ struct regmap_irq_chip_data; * instead. * @num_config_bases: Number of config base registers. * @num_config_regs: Number of config registers for each config base register. + * * @handle_pre_irq: Driver specific callback to handle interrupt from device * before regmap_irq_handler process the interrupts. * @handle_post_irq: Driver specific callback to handle interrupt from device @@ -1625,12 +1628,12 @@ struct regmap_irq_chip { unsigned int use_ack:1; unsigned int ack_invert:1; unsigned int clear_ack:1; + unsigned int status_invert:1; unsigned int wake_invert:1; - unsigned int runtime_pm:1; unsigned int type_in_mask:1; unsigned int clear_on_unmask:1; + unsigned int runtime_pm:1; unsigned int not_fixed_stride:1; - unsigned int status_invert:1; int num_regs; -- cgit v1.2.3 From ae2ade4ba58167f165fbf3db19380f9b72c56db8 Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Wed, 1 Mar 2023 21:58:52 +0100 Subject: spi: Reorder fields in 'struct spi_message' Group some variables based on their sizes to reduce hole and avoid padding. On x86_64, this shrinks the size from 112 to 96 bytes. This should have no real impact on memory allocation because 'struct spi_message' is mostly used on stack, but it can save a few cycles when the structure is initialized with spi_message_init() and co. Signed-off-by: Christophe JAILLET Tested-by: Muhammad Usama Anjum Reviewed-by: Muhammad Usama Anjum Link: https://lore.kernel.org/r/c112aad16eb47808e1ec10abd87b3d273c969a68.1677704283.git.christophe.jaillet@wanadoo.fr Signed-off-by: Mark Brown --- include/linux/spi/spi.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h index 4fa26b9a3572..bdb35a91b4bf 100644 --- a/include/linux/spi/spi.h +++ b/include/linux/spi/spi.h @@ -1093,6 +1093,9 @@ struct spi_message { unsigned is_dma_mapped:1; + /* spi_prepare_message() was called for this message */ + bool prepared; + /* REVISIT: we might want a flag affecting the behavior of the * last transfer ... allowing things like "read 16 bit length L" * immediately followed by "read L bytes". Basically imposing @@ -1105,11 +1108,11 @@ struct spi_message { */ /* Completion is reported through a callback */ + int status; void (*complete)(void *context); void *context; unsigned frame_length; unsigned actual_length; - int status; /* For optional use by whatever driver currently owns the * spi_message ... between calls to spi_async and then later @@ -1120,9 +1123,6 @@ struct spi_message { /* List of spi_res reources when the spi message is processed */ struct list_head resources; - - /* spi_prepare_message() was called for this message */ - bool prepared; }; static inline void spi_message_init_no_memset(struct spi_message *m) -- cgit v1.2.3 From 0aaf08de8426f823bd0e36797445222e6392e374 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 19 Jan 2023 19:22:22 -0500 Subject: __blockdev_direct_IO(): get rid of submit_io callback always NULL... Signed-off-by: Al Viro --- include/linux/fs.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index c85916e9f7db..b07d13e546d3 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2778,7 +2778,7 @@ enum { ssize_t __blockdev_direct_IO(struct kiocb *iocb, struct inode *inode, struct block_device *bdev, struct iov_iter *iter, get_block_t get_block, - dio_iodone_t end_io, dio_submit_t submit_io, + dio_iodone_t end_io, int flags); static inline ssize_t blockdev_direct_IO(struct kiocb *iocb, @@ -2787,7 +2787,7 @@ static inline ssize_t blockdev_direct_IO(struct kiocb *iocb, get_block_t get_block) { return __blockdev_direct_IO(iocb, inode, inode->i_sb->s_bdev, iter, - get_block, NULL, NULL, DIO_LOCKING | DIO_SKIP_HOLES); + get_block, NULL, DIO_LOCKING | DIO_SKIP_HOLES); } #endif -- cgit v1.2.3 From 721da5cee9d43901105f5b8bd33fcb9101b12fc3 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 23 Feb 2023 08:33:26 +0100 Subject: driver core: remove CONFIG_SYSFS_DEPRECATED and CONFIG_SYSFS_DEPRECATED_V2 CONFIG_SYSFS_DEPRECATED was added in commit 88a22c985e35 ("CONFIG_SYSFS_DEPRECATED") in 2006 to allow systems with older versions of some tools (i.e. Fedora 3's version of udev) to boot properly. Four years later, in 2010, the option was attempted to be removed as most of userspace should have been fixed up properly by then, but some kernel developers clung to those old systems and refused to update, so we added CONFIG_SYSFS_DEPRECATED_V2 in commit e52eec13cd6b ("SYSFS: Allow boot time switching between deprecated and modern sysfs layout") to allow them to continue to boot properly, and we allowed a boot time parameter to be used to switch back to the old format if needed. Over time, the logic that was covered under these config options was slowly removed from individual driver subsystems successfully, removed, and the only thing that is now left in the kernel are some changes in the block layer's representation in sysfs where real directories are used instead of symlinks like normal. Because the original changes were done to userspace tools in 2006, and all distros that use those tools are long end-of-life, and older non-udev-based systems do not care about the block layer's sysfs representation, it is time to finally remove this old logic and the config entries from the kernel. Cc: Jonathan Corbet Cc: "Rafael J. Wysocki" Cc: linux-block@vger.kernel.org Cc: linux-doc@vger.kernel.org Acked-by: Jens Axboe Link: https://lore.kernel.org/r/20230223073326.2073220-1-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman --- include/linux/device.h | 6 ------ 1 file changed, 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/device.h b/include/linux/device.h index 1508e637bb26..19b6ba478fbf 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -1092,10 +1092,4 @@ int dev_err_probe(const struct device *dev, int err, const char *fmt, ...); #define MODULE_ALIAS_CHARDEV_MAJOR(major) \ MODULE_ALIAS("char-major-" __stringify(major) "-*") -#ifdef CONFIG_SYSFS_DEPRECATED -extern long sysfs_deprecated; -#else -#define sysfs_deprecated 0 -#endif - #endif /* _DEVICE_H_ */ -- cgit v1.2.3 From f2620f166e2a4db08f016b7b30b904ab28c265e4 Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Wed, 1 Feb 2023 14:14:52 +0100 Subject: xattr: simplify listxattr helpers The generic_listxattr() and simple_xattr_list() helpers list xattrs and contain duplicated code. Add two helpers that both generic_listxattr() and simple_xattr_list() can use. Reviewed-by: Christoph Hellwig Signed-off-by: Christian Brauner (Microsoft) --- include/linux/posix_acl.h | 7 +++++++ include/linux/xattr.h | 1 + 2 files changed, 8 insertions(+) (limited to 'include/linux') diff --git a/include/linux/posix_acl.h b/include/linux/posix_acl.h index 21cc29b8a9e8..0e65b3d634d9 100644 --- a/include/linux/posix_acl.h +++ b/include/linux/posix_acl.h @@ -106,6 +106,8 @@ struct posix_acl *vfs_get_acl(struct mnt_idmap *idmap, struct dentry *dentry, const char *acl_name); int vfs_remove_acl(struct mnt_idmap *idmap, struct dentry *dentry, const char *acl_name); +int posix_acl_listxattr(struct inode *inode, char **buffer, + ssize_t *remaining_size); #else static inline int posix_acl_chmod(struct mnt_idmap *idmap, struct dentry *dentry, umode_t mode) @@ -153,6 +155,11 @@ static inline int vfs_remove_acl(struct mnt_idmap *idmap, { return -EOPNOTSUPP; } +static inline int posix_acl_listxattr(struct inode *inode, char **buffer, + ssize_t *remaining_size) +{ + return 0; +} #endif /* CONFIG_FS_POSIX_ACL */ struct posix_acl *get_inode_acl(struct inode *inode, int type); diff --git a/include/linux/xattr.h b/include/linux/xattr.h index 6af72461397d..ce8fb0bd56c6 100644 --- a/include/linux/xattr.h +++ b/include/linux/xattr.h @@ -109,5 +109,6 @@ ssize_t simple_xattr_list(struct inode *inode, struct simple_xattrs *xattrs, char *buffer, size_t size); void simple_xattr_add(struct simple_xattrs *xattrs, struct simple_xattr *new_xattr); +int xattr_list_one(char **buffer, ssize_t *remaining_size, const char *name); #endif /* _LINUX_XATTR_H */ -- cgit v1.2.3 From 2db8a948046cab3a2f707561592906a3d096972f Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Wed, 1 Feb 2023 14:14:53 +0100 Subject: xattr: add listxattr helper Add a tiny helper to determine whether an xattr handler given a specific dentry supports listing the requested xattr. We will use this helper in various filesystems in later commits. Reviewed-by: Christoph Hellwig Signed-off-by: Christian Brauner (Microsoft) --- include/linux/xattr.h | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) (limited to 'include/linux') diff --git a/include/linux/xattr.h b/include/linux/xattr.h index ce8fb0bd56c6..7adf768d4396 100644 --- a/include/linux/xattr.h +++ b/include/linux/xattr.h @@ -47,6 +47,22 @@ struct xattr_handler { size_t size, int flags); }; +/** + * xattr_handler_can_list - check whether xattr can be listed + * @handler: handler for this type of xattr + * @dentry: dentry whose inode xattr to list + * + * Determine whether the xattr associated with @dentry can be listed given + * @handler. + * + * Return: true if xattr can be listed, false if not. + */ +static inline bool xattr_handler_can_list(const struct xattr_handler *handler, + struct dentry *dentry) +{ + return handler && (!handler->list || handler->list(dentry)); +} + const char *xattr_full_name(const struct xattr_handler *, const char *); struct xattr { -- cgit v1.2.3 From 831be973aa21d1cf8948bf4b1d4e73e6d5d028c0 Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Wed, 1 Feb 2023 14:14:54 +0100 Subject: xattr: remove unused argument his helpers is really just used to check for user.* xattr support so don't make it pointlessly generic. Reviewed-by: Christoph Hellwig Signed-off-by: Christian Brauner (Microsoft) --- include/linux/xattr.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/xattr.h b/include/linux/xattr.h index 7adf768d4396..d591ef59aa98 100644 --- a/include/linux/xattr.h +++ b/include/linux/xattr.h @@ -94,7 +94,7 @@ int vfs_getxattr_alloc(struct mnt_idmap *idmap, struct dentry *dentry, const char *name, char **xattr_value, size_t size, gfp_t flags); -int xattr_supported_namespace(struct inode *inode, const char *prefix); +int xattr_supports_user_prefix(struct inode *inode); static inline const char *xattr_prefix(const struct xattr_handler *handler) { -- cgit v1.2.3 From d549b741740e63e87e661754e2d1b336fdc51d50 Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Wed, 1 Feb 2023 14:14:58 +0100 Subject: fs: rename generic posix acl handlers Reflect in their naming and document that they are kept around for legacy reasons and shouldn't be used anymore by new code. Reviewed-by: Christoph Hellwig Signed-off-by: Christian Brauner (Microsoft) --- include/linux/posix_acl_xattr.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/posix_acl_xattr.h b/include/linux/posix_acl_xattr.h index 54cd7a14330d..e86f3b731da2 100644 --- a/include/linux/posix_acl_xattr.h +++ b/include/linux/posix_acl_xattr.h @@ -68,7 +68,8 @@ static inline int posix_acl_type(const char *name) return -1; } -extern const struct xattr_handler posix_acl_access_xattr_handler; -extern const struct xattr_handler posix_acl_default_xattr_handler; +/* These are legacy handlers. Don't use them for new code. */ +extern const struct xattr_handler nop_posix_acl_access; +extern const struct xattr_handler nop_posix_acl_default; #endif /* _POSIX_ACL_XATTR_H */ -- cgit v1.2.3 From 94d20f7d674d589e94aea77e2cbe37fc58541d04 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 7 Feb 2023 16:29:42 +0200 Subject: gpiolib: coldfire: remove custom asm/gpio.h Now that coldfire is the only user of a custom asm/gpio.h, it seems better to remove this as well, and have the same interface everywhere. For the gpio_get_value()/gpio_set_value()/gpio_to_irq(), gpio_cansleep() functions, the custom version is only a micro-optimization to inline the function for constant GPIO numbers. However, in the coldfire defconfigs, I was unable to find a single instance where this micro-optimization was even used, and according to Geert the only user appears to be the QSPI chip that is disabled everywhere. The custom gpio_request_one() function is even less useful, as it is guarded by an #ifdef that is never true. Reviewed-by: Geert Uytterhoeven Reviewed-by: Andy Shevchenko Acked-by: Bartosz Golaszewski Signed-off-by: Arnd Bergmann Reviewed-by: Linus Walleij Signed-off-by: Andy Shevchenko --- include/linux/gpio.h | 7 ------- 1 file changed, 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/gpio.h b/include/linux/gpio.h index 85beb236c925..2b75017b3aad 100644 --- a/include/linux/gpio.h +++ b/include/linux/gpio.h @@ -54,11 +54,6 @@ struct gpio { }; #ifdef CONFIG_GPIOLIB - -#ifdef CONFIG_ARCH_HAVE_CUSTOM_GPIO_H -#include -#else - #include static inline int gpio_get_value(unsigned int gpio) @@ -81,8 +76,6 @@ static inline int gpio_to_irq(unsigned int gpio) return __gpio_to_irq(gpio); } -#endif /* ! CONFIG_ARCH_HAVE_CUSTOM_GPIO_H */ - /* CONFIG_GPIOLIB: bindings for managed devices that want to request gpios */ struct device; -- cgit v1.2.3 From eccb7a00613c804ec7244676090bf6ee43a23da2 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 8 Feb 2023 17:37:22 +0200 Subject: gpiolib: remove asm-generic/gpio.h The asm-generic/gpio.h file is now always included when using gpiolib, so just move its contents into linux/gpio.h with a few minor simplifications. Signed-off-by: Arnd Bergmann Reviewed-by: Linus Walleij Reviewed-by: Andy Shevchenko Signed-off-by: Andy Shevchenko --- include/linux/gpio.h | 94 ++++++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 84 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/gpio.h b/include/linux/gpio.h index 2b75017b3aad..d5ce78e2bdd9 100644 --- a/include/linux/gpio.h +++ b/include/linux/gpio.h @@ -13,6 +13,7 @@ #define __LINUX_GPIO_H #include +#include /* see Documentation/driver-api/gpio/legacy.rst */ @@ -54,26 +55,100 @@ struct gpio { }; #ifdef CONFIG_GPIOLIB -#include -static inline int gpio_get_value(unsigned int gpio) +#include + +/* + * "valid" GPIO numbers are nonnegative and may be passed to + * setup routines like gpio_request(). Only some valid numbers + * can successfully be requested and used. + * + * Invalid GPIO numbers are useful for indicating no-such-GPIO in + * platform data and other tables. + */ +static inline bool gpio_is_valid(int number) +{ + /* only non-negative numbers are valid */ + return number >= 0; +} + +/* + * Platforms may implement their GPIO interface with library code, + * at a small performance cost for non-inlined operations and some + * extra memory (for code and for per-GPIO table entries). + */ + +/* + * At the end we want all GPIOs to be dynamically allocated from 0. + * However, some legacy drivers still perform fixed allocation. + * Until they are all fixed, leave 0-512 space for them. + */ +#define GPIO_DYNAMIC_BASE 512 + +/* Always use the library code for GPIO management calls, + * or when sleeping may be involved. + */ +int gpio_request(unsigned gpio, const char *label); +void gpio_free(unsigned gpio); + +static inline int gpio_direction_input(unsigned gpio) +{ + return gpiod_direction_input(gpio_to_desc(gpio)); +} +static inline int gpio_direction_output(unsigned gpio, int value) +{ + return gpiod_direction_output_raw(gpio_to_desc(gpio), value); +} + +static inline int gpio_set_debounce(unsigned gpio, unsigned debounce) +{ + return gpiod_set_debounce(gpio_to_desc(gpio), debounce); +} + +static inline int gpio_get_value_cansleep(unsigned gpio) +{ + return gpiod_get_raw_value_cansleep(gpio_to_desc(gpio)); +} +static inline void gpio_set_value_cansleep(unsigned gpio, int value) { - return __gpio_get_value(gpio); + return gpiod_set_raw_value_cansleep(gpio_to_desc(gpio), value); } -static inline void gpio_set_value(unsigned int gpio, int value) +static inline int gpio_get_value(unsigned gpio) +{ + return gpiod_get_raw_value(gpio_to_desc(gpio)); +} +static inline void gpio_set_value(unsigned gpio, int value) { - __gpio_set_value(gpio, value); + return gpiod_set_raw_value(gpio_to_desc(gpio), value); } -static inline int gpio_cansleep(unsigned int gpio) +static inline int gpio_cansleep(unsigned gpio) { - return __gpio_cansleep(gpio); + return gpiod_cansleep(gpio_to_desc(gpio)); } -static inline int gpio_to_irq(unsigned int gpio) +static inline int gpio_to_irq(unsigned gpio) { - return __gpio_to_irq(gpio); + return gpiod_to_irq(gpio_to_desc(gpio)); +} + +int gpio_request_one(unsigned gpio, unsigned long flags, const char *label); +int gpio_request_array(const struct gpio *array, size_t num); +void gpio_free_array(const struct gpio *array, size_t num); + +/* + * A sysfs interface can be exported by individual drivers if they want, + * but more typically is configured entirely from userspace. + */ +static inline int gpio_export(unsigned gpio, bool direction_may_change) +{ + return gpiod_export(gpio_to_desc(gpio), direction_may_change); +} + +static inline void gpio_unexport(unsigned gpio) +{ + gpiod_unexport(gpio_to_desc(gpio)); } /* CONFIG_GPIOLIB: bindings for managed devices that want to request gpios */ @@ -88,7 +163,6 @@ int devm_gpio_request_one(struct device *dev, unsigned gpio, #include #include -#include struct device; struct gpio_chip; -- cgit v1.2.3 From 0e685c3e7158d35626d6d76b9f859eae806d87fa Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 7 Feb 2023 16:29:44 +0200 Subject: gpiolib: remove gpio_set_debounce() gpio_set_debounce() only has a single user, which is trivially converted to gpiod_set_debounce(). Signed-off-by: Arnd Bergmann Reviewed-by: Linus Walleij Reviewed-by: Andy Shevchenko Signed-off-by: Andy Shevchenko --- include/linux/gpio.h | 10 ---------- 1 file changed, 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/gpio.h b/include/linux/gpio.h index d5ce78e2bdd9..fc56733e8514 100644 --- a/include/linux/gpio.h +++ b/include/linux/gpio.h @@ -100,11 +100,6 @@ static inline int gpio_direction_output(unsigned gpio, int value) return gpiod_direction_output_raw(gpio_to_desc(gpio), value); } -static inline int gpio_set_debounce(unsigned gpio, unsigned debounce) -{ - return gpiod_set_debounce(gpio_to_desc(gpio), debounce); -} - static inline int gpio_get_value_cansleep(unsigned gpio) { return gpiod_get_raw_value_cansleep(gpio_to_desc(gpio)); @@ -214,11 +209,6 @@ static inline int gpio_direction_output(unsigned gpio, int value) return -ENOSYS; } -static inline int gpio_set_debounce(unsigned gpio, unsigned debounce) -{ - return -ENOSYS; -} - static inline int gpio_get_value(unsigned gpio) { /* GPIO can never have been requested or set as {in,out}put */ -- cgit v1.2.3 From d74e316633e49f44756c23997fa071979a939405 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 7 Feb 2023 16:29:45 +0200 Subject: gpiolib: remove legacy gpio_export() There are only a handful of users of gpio_export() and related functions. As these are just wrappers around the modern gpiod_export() helper, remove the wrappers and open-code the gpio_to_desc in all callers to shrink the legacy API. Reviewed-by: Andy Shevchenko Signed-off-by: Arnd Bergmann Reviewed-by: Linus Walleij Signed-off-by: Andy Shevchenko --- include/linux/gpio.h | 27 --------------------------- 1 file changed, 27 deletions(-) (limited to 'include/linux') diff --git a/include/linux/gpio.h b/include/linux/gpio.h index fc56733e8514..a86953696e47 100644 --- a/include/linux/gpio.h +++ b/include/linux/gpio.h @@ -132,20 +132,6 @@ int gpio_request_one(unsigned gpio, unsigned long flags, const char *label); int gpio_request_array(const struct gpio *array, size_t num); void gpio_free_array(const struct gpio *array, size_t num); -/* - * A sysfs interface can be exported by individual drivers if they want, - * but more typically is configured entirely from userspace. - */ -static inline int gpio_export(unsigned gpio, bool direction_may_change) -{ - return gpiod_export(gpio_to_desc(gpio), direction_may_change); -} - -static inline void gpio_unexport(unsigned gpio) -{ - gpiod_unexport(gpio_to_desc(gpio)); -} - /* CONFIG_GPIOLIB: bindings for managed devices that want to request gpios */ struct device; @@ -242,19 +228,6 @@ static inline void gpio_set_value_cansleep(unsigned gpio, int value) WARN_ON(1); } -static inline int gpio_export(unsigned gpio, bool direction_may_change) -{ - /* GPIO can never have been requested or set as {in,out}put */ - WARN_ON(1); - return -EINVAL; -} - -static inline void gpio_unexport(unsigned gpio) -{ - /* GPIO can never have been exported */ - WARN_ON(1); -} - static inline int gpio_to_irq(unsigned gpio) { /* GPIO can never have been requested or set as input */ -- cgit v1.2.3 From a8e59744e16b9ae18fab773a0fd3b23cdf21ad75 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 8 Feb 2023 17:41:47 +0200 Subject: gpiolib: split linux/gpio/driver.h out of linux/gpio.h Almost all gpio drivers include linux/gpio/driver.h, and other files should not rely on includes from this header. Remove the indirect include from here and include the correct headers directly from where they are used. Reviewed-by: Andy Shevchenko Signed-off-by: Arnd Bergmann Reviewed-by: Linus Walleij Signed-off-by: Andy Shevchenko Acked-by: Lee Jones --- include/linux/mfd/ucb1x00.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/mfd/ucb1x00.h b/include/linux/mfd/ucb1x00.h index 43bcf35afe27..ede237384723 100644 --- a/include/linux/mfd/ucb1x00.h +++ b/include/linux/mfd/ucb1x00.h @@ -10,6 +10,7 @@ #include #include #include +#include #include #define UCB_IO_DATA 0x00 -- cgit v1.2.3 From a99cc66807d6c854a7f65f962766c530c91be149 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 7 Feb 2023 16:29:47 +0200 Subject: gpiolib: split of_mm_gpio_chip out of linux/of_gpio.h This is a rarely used feature that has nothing to do with the client-side of_gpio.h. Split it out with a separate header file and Kconfig option so it can be removed on its own timeline aside from removing the of_gpio consumer interfaces. Reviewed-by: Andy Shevchenko Signed-off-by: Arnd Bergmann Reviewed-by: Linus Walleij Signed-off-by: Andy Shevchenko --- include/linux/gpio/legacy-of-mm-gpiochip.h | 36 ++++++++++++++++++++++++++++++ include/linux/of_gpio.h | 21 ----------------- 2 files changed, 36 insertions(+), 21 deletions(-) create mode 100644 include/linux/gpio/legacy-of-mm-gpiochip.h (limited to 'include/linux') diff --git a/include/linux/gpio/legacy-of-mm-gpiochip.h b/include/linux/gpio/legacy-of-mm-gpiochip.h new file mode 100644 index 000000000000..2e2bd3b19cc3 --- /dev/null +++ b/include/linux/gpio/legacy-of-mm-gpiochip.h @@ -0,0 +1,36 @@ +/* SPDX-License-Identifier: GPL-2.0+ */ +/* + * OF helpers for the old of_mm_gpio_chip, used on ppc32 and nios2, + * do not use in new code. + * + * Copyright (c) 2007-2008 MontaVista Software, Inc. + * + * Author: Anton Vorontsov + */ + +#ifndef __LINUX_GPIO_LEGACY_OF_MM_GPIO_CHIP_H +#define __LINUX_GPIO_LEGACY_OF_MM_GPIO_CHIP_H + +#include +#include + +/* + * OF GPIO chip for memory mapped banks + */ +struct of_mm_gpio_chip { + struct gpio_chip gc; + void (*save_regs)(struct of_mm_gpio_chip *mm_gc); + void __iomem *regs; +}; + +static inline struct of_mm_gpio_chip *to_of_mm_gpio_chip(struct gpio_chip *gc) +{ + return container_of(gc, struct of_mm_gpio_chip, gc); +} + +extern int of_mm_gpiochip_add_data(struct device_node *np, + struct of_mm_gpio_chip *mm_gc, + void *data); +extern void of_mm_gpiochip_remove(struct of_mm_gpio_chip *mm_gc); + +#endif /* __LINUX_GPIO_LEGACY_OF_MM_GPIO_CHIP_H */ diff --git a/include/linux/of_gpio.h b/include/linux/of_gpio.h index 5d58b3b0a97e..d0f66a5e1b2a 100644 --- a/include/linux/of_gpio.h +++ b/include/linux/of_gpio.h @@ -19,30 +19,9 @@ struct device_node; #ifdef CONFIG_OF_GPIO -#include - -/* - * OF GPIO chip for memory mapped banks - */ -struct of_mm_gpio_chip { - struct gpio_chip gc; - void (*save_regs)(struct of_mm_gpio_chip *mm_gc); - void __iomem *regs; -}; - -static inline struct of_mm_gpio_chip *to_of_mm_gpio_chip(struct gpio_chip *gc) -{ - return container_of(gc, struct of_mm_gpio_chip, gc); -} - extern int of_get_named_gpio(const struct device_node *np, const char *list_name, int index); -extern int of_mm_gpiochip_add_data(struct device_node *np, - struct of_mm_gpio_chip *mm_gc, - void *data); -extern void of_mm_gpiochip_remove(struct of_mm_gpio_chip *mm_gc); - #else /* CONFIG_OF_GPIO */ #include -- cgit v1.2.3 From 6cfd84c4f4f619