From 02669b17a433c242a40f01f14b691c9c9d1f8a13 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Wed, 5 Dec 2018 16:37:03 -0500 Subject: XArray: Turn xa_init_flags into a static inline A regular xa_init_flags() put all dynamically-initialised XArrays into the same locking class. That leads to lockdep believing that taking one XArray lock while holding another is a deadlock. It's possible to work around some of these situations with separate locking classes for irq/bh/regular XArrays, and SINGLE_DEPTH_NESTING, but that's ugly, and it doesn't work for all situations (where we have completely unrelated XArrays). Signed-off-by: Matthew Wilcox --- include/linux/xarray.h | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/xarray.h b/include/linux/xarray.h index f492e21c4aa2..4cf3cd128689 100644 --- a/include/linux/xarray.h +++ b/include/linux/xarray.h @@ -286,7 +286,6 @@ struct xarray { */ #define DEFINE_XARRAY_ALLOC(name) DEFINE_XARRAY_FLAGS(name, XA_FLAGS_ALLOC) -void xa_init_flags(struct xarray *, gfp_t flags); void *xa_load(struct xarray *, unsigned long index); void *xa_store(struct xarray *, unsigned long index, void *entry, gfp_t); void *xa_erase(struct xarray *, unsigned long index); @@ -303,6 +302,24 @@ unsigned int xa_extract(struct xarray *, void **dst, unsigned long start, unsigned long max, unsigned int n, xa_mark_t); void xa_destroy(struct xarray *); +/** + * xa_init_flags() - Initialise an empty XArray with flags. + * @xa: XArray. + * @flags: XA_FLAG values. + * + * If you need to initialise an XArray with special flags (eg you need + * to take the lock from interrupt context), use this function instead + * of xa_init(). + * + * Context: Any context. + */ +static inline void xa_init_flags(struct xarray *xa, gfp_t flags) +{ + spin_lock_init(&xa->xa_lock); + xa->xa_flags = flags; + xa->xa_head = NULL; +} + /** * xa_init() - Initialise an empty XArray. * @xa: XArray. -- cgit v1.2.3 From 4a31896c5b5a2715ecf4033426aa0a35066d92d6 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Mon, 17 Dec 2018 14:45:36 -0500 Subject: XArray: Change xa_for_each iterator There were three problems with this API: 1. It took too many arguments; almost all users wanted to iterate over every element in the array rather than a subset. 2. It required that 'index' be initialised before use, and there's no realistic way to make GCC catch that. 3. 'index' and 'entry' were the opposite way round from every other member of the XArray APIs. So split it into three different APIs: xa_for_each(xa, index, entry) xa_for_each_start(xa, index, entry, start) xa_for_each_marked(xa, index, entry, filter) Signed-off-by: Matthew Wilcox --- include/linux/xarray.h | 78 +++++++++++++++++++++++++++++++++++++++++--------- lib/test_xarray.c | 11 ++++--- 2 files changed, 70 insertions(+), 19 deletions(-) (limited to 'include/linux') diff --git a/include/linux/xarray.h b/include/linux/xarray.h index 4cf3cd128689..3d0ce8b267e3 100644 --- a/include/linux/xarray.h +++ b/include/linux/xarray.h @@ -359,20 +359,45 @@ static inline bool xa_marked(const struct xarray *xa, xa_mark_t mark) } /** - * xa_for_each() - Iterate over a portion of an XArray. + * xa_for_each_start() - Iterate over a portion of an XArray. * @xa: XArray. + * @index: Index of @entry. * @entry: Entry retrieved from array. + * @start: First index to retrieve from array. + * + * During the iteration, @entry will have the value of the entry stored + * in @xa at @index. You may modify @index during the iteration if you + * want to skip or reprocess indices. It is safe to modify the array + * during the iteration. At the end of the iteration, @entry will be set + * to NULL and @index will have a value less than or equal to max. + * + * xa_for_each_start() is O(n.log(n)) while xas_for_each() is O(n). You have + * to handle your own locking with xas_for_each(), and if you have to unlock + * after each iteration, it will also end up being O(n.log(n)). + * xa_for_each_start() will spin if it hits a retry entry; if you intend to + * see retry entries, you should use the xas_for_each() iterator instead. + * The xas_for_each() iterator will expand into more inline code than + * xa_for_each_start(). + * + * Context: Any context. Takes and releases the RCU lock. + */ +#define xa_for_each_start(xa, index, entry, start) \ + for (index = start, \ + entry = xa_find(xa, &index, ULONG_MAX, XA_PRESENT); \ + entry; \ + entry = xa_find_after(xa, &index, ULONG_MAX, XA_PRESENT)) + +/** + * xa_for_each() - Iterate over present entries in an XArray. + * @xa: XArray. * @index: Index of @entry. - * @max: Maximum index to retrieve from array. - * @filter: Selection criterion. + * @entry: Entry retrieved from array. * - * Initialise @index to the lowest index you want to retrieve from the - * array. During the iteration, @entry will have the value of the entry - * stored in @xa at @index. The iteration will skip all entries in the - * array which do not match @filter. You may modify @index during the - * iteration if you want to skip or reprocess indices. It is safe to modify - * the array during the iteration. At the end of the iteration, @entry will - * be set to NULL and @index will have a value less than or equal to max. + * During the iteration, @entry will have the value of the entry stored + * in @xa at @index. You may modify @index during the iteration if you want + * to skip or reprocess indices. It is safe to modify the array during the + * iteration. At the end of the iteration, @entry will be set to NULL and + * @index will have a value less than or equal to max. * * xa_for_each() is O(n.log(n)) while xas_for_each() is O(n). You have * to handle your own locking with xas_for_each(), and if you have to unlock @@ -383,9 +408,36 @@ static inline bool xa_marked(const struct xarray *xa, xa_mark_t mark) * * Context: Any context. Takes and releases the RCU lock. */ -#define xa_for_each(xa, entry, index, max, filter) \ - for (entry = xa_find(xa, &index, max, filter); entry; \ - entry = xa_find_after(xa, &index, max, filter)) +#define xa_for_each(xa, index, entry) \ + xa_for_each_start(xa, index, entry, 0) + +/** + * xa_for_each_marked() - Iterate over marked entries in an XArray. + * @xa: XArray. + * @index: Index of @entry. + * @entry: Entry retrieved from array. + * @filter: Selection criterion. + * + * During the iteration, @entry will have the value of the entry stored + * in @xa at @index. The iteration will skip all entries in the array + * which do not match @filter. You may modify @index during the iteration + * if you want to skip or reprocess indices. It is safe to modify the array + * during the iteration. At the end of the iteration, @entry will be set to + * NULL and @index will have a value less than or equal to max. + * + * xa_for_each_marked() is O(n.log(n)) while xas_for_each_marked() is O(n). + * You have to handle your own locking with xas_for_each(), and if you have + * to unlock after each iteration, it will also end up being O(n.log(n)). + * xa_for_each_marked() will spin if it hits a retry entry; if you intend to + * see retry entries, you should use the xas_for_each_marked() iterator + * instead. The xas_for_each_marked() iterator will expand into more inline + * code than xa_for_each_marked(). + * + * Context: Any context. Takes and releases the RCU lock. + */ +#define xa_for_each_marked(xa, index, entry, filter) \ + for (index = 0, entry = xa_find(xa, &index, ULONG_MAX, filter); \ + entry; entry = xa_find_after(xa, &index, ULONG_MAX, filter)) #define xa_trylock(xa) spin_trylock(&(xa)->xa_lock) #define xa_lock(xa) spin_lock(&(xa)->xa_lock) diff --git a/lib/test_xarray.c b/lib/test_xarray.c index a885afde0aef..dc02eff562b8 100644 --- a/lib/test_xarray.c +++ b/lib/test_xarray.c @@ -357,7 +357,7 @@ static noinline void check_cmpxchg(struct xarray *xa) static noinline void check_reserve(struct xarray *xa) { void *entry; - unsigned long index = 0; + unsigned long index; /* An array with a reserved entry is not empty */ XA_BUG_ON(xa, !xa_empty(xa)); @@ -393,7 +393,7 @@ static noinline void check_reserve(struct xarray *xa) xa_reserve(xa, 6, GFP_KERNEL); xa_store_index(xa, 7, GFP_KERNEL); - xa_for_each(xa, entry, index, ULONG_MAX, XA_PRESENT) { + xa_for_each(xa, index, entry) { XA_BUG_ON(xa, index != 5 && index != 7); } xa_destroy(xa); @@ -812,17 +812,16 @@ static noinline void check_find_1(struct xarray *xa) static noinline void check_find_2(struct xarray *xa) { void *entry; - unsigned long i, j, index = 0; + unsigned long i, j, index; - xa_for_each(xa, entry, index, ULONG_MAX, XA_PRESENT) { + xa_for_each(xa, index, entry) { XA_BUG_ON(xa, true); } for (i = 0; i < 1024; i++) { xa_store_index(xa, index, GFP_KERNEL); j = 0; - index = 0; - xa_for_each(xa, entry, index, ULONG_MAX, XA_PRESENT) { + xa_for_each(xa, index, entry) { XA_BUG_ON(xa, xa_mk_index(index) != entry); XA_BUG_ON(xa, index != j++); } -- cgit v1.2.3 From 76b4e52995654af260f14558e0e07b5b039ae202 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Fri, 28 Dec 2018 23:20:44 -0500 Subject: XArray: Permit storing 2-byte-aligned pointers On m68k, statically allocated pointers may only be two-byte aligned. This clashes with the XArray's method for tagging internal pointers. Permit storing these pointers in single slots (ie not in multislots). Signed-off-by: Matthew Wilcox --- include/linux/xarray.h | 18 +++++++++++++++--- lib/test_xarray.c | 30 ++++++++++++++++++++++++++++++ lib/xarray.c | 22 +++++++++++++--------- 3 files changed, 58 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/xarray.h b/include/linux/xarray.h index 3d0ce8b267e3..435c25b29079 100644 --- a/include/linux/xarray.h +++ b/include/linux/xarray.h @@ -176,7 +176,8 @@ static inline bool xa_is_internal(const void *entry) */ static inline bool xa_is_err(const void *entry) { - return unlikely(xa_is_internal(entry)); + return unlikely(xa_is_internal(entry) && + (unsigned long)entry >= -((MAX_ERRNO << 2) + 2)); } /** @@ -1039,8 +1040,8 @@ static inline bool xa_is_sibling(const void *entry) (entry < xa_mk_sibling(XA_CHUNK_SIZE - 1)); } -#define XA_ZERO_ENTRY xa_mk_internal(256) -#define XA_RETRY_ENTRY xa_mk_internal(257) +#define XA_RETRY_ENTRY xa_mk_internal(256) +#define XA_ZERO_ENTRY xa_mk_internal(257) /** * xa_is_zero() - Is the entry a zero entry? @@ -1064,6 +1065,17 @@ static inline bool xa_is_retry(const void *entry) return unlikely(entry == XA_RETRY_ENTRY); } +/** + * xa_is_advanced() - Is the entry only permitted for the advanced API? + * @entry: Entry to be stored in the XArray. + * + * Return: %true if the entry cannot be stored by the normal API. + */ +static inline bool xa_is_advanced(const void *entry) +{ + return xa_is_internal(entry) && (entry <= XA_RETRY_ENTRY); +} + /** * typedef xa_update_node_t - A callback function from the XArray. * @node: The node which is being processed diff --git a/lib/test_xarray.c b/lib/test_xarray.c index dc02eff562b8..6e0212a60b08 100644 --- a/lib/test_xarray.c +++ b/lib/test_xarray.c @@ -1184,6 +1184,35 @@ static noinline void check_store_range(struct xarray *xa) } } +static void check_align_1(struct xarray *xa, char *name) +{ + int i; + unsigned int id; + unsigned long index; + void *entry; + + for (i = 0; i < 8; i++) { + id = 0; + XA_BUG_ON(xa, xa_alloc(xa, &id, UINT_MAX, name + i, GFP_KERNEL) + != 0); + XA_BUG_ON(xa, id != i); + } + xa_for_each(xa, index, entry) + XA_BUG_ON(xa, xa_is_err(entry)); + xa_destroy(xa); +} + +static noinline void check_align(struct xarray *xa) +{ + char name[] = "Motorola 68000"; + + check_align_1(xa, name); + check_align_1(xa, name + 1); + check_align_1(xa, name + 2); + check_align_1(xa, name + 3); +// check_align_2(xa, name); +} + static LIST_HEAD(shadow_nodes); static void test_update_node(struct xa_node *node) @@ -1333,6 +1362,7 @@ static int xarray_checks(void) check_create_range(&array); check_store_range(&array); check_store_iter(&array); + check_align(&xa0); check_workingset(&array, 0); check_workingset(&array, 64); diff --git a/lib/xarray.c b/lib/xarray.c index dda6026d202e..bffa26b1f0d6 100644 --- a/lib/xarray.c +++ b/lib/xarray.c @@ -232,6 +232,8 @@ void *xas_load(struct xa_state *xas) if (xas->xa_shift > node->shift) break; entry = xas_descend(xas, node); + if (node->shift == 0) + break; } return entry; } @@ -506,7 +508,7 @@ static void xas_free_nodes(struct xa_state *xas, struct xa_node *top) for (;;) { void *entry = xa_entry_locked(xas->xa, node, offset); - if (xa_is_node(entry)) { + if (node->shift && xa_is_node(entry)) { node = xa_to_node(entry); offset = 0; continue; @@ -604,6 +606,7 @@ static int xas_expand(struct xa_state *xas, void *head) /* * xas_create() - Create a slot to store an entry in. * @xas: XArray operation state. + * @allow_root: %true if we can store the entry in the root directly * * Most users will not need to call this function directly, as it is called * by xas_store(). It is useful for doing conditional store operations @@ -613,7 +616,7 @@ static int xas_expand(struct xa_state *xas, void *head) * If the slot was newly created, returns %NULL. If it failed to create the * slot, returns %NULL and indicates the error in @xas. */ -static void *xas_create(struct xa_state *xas) +static void *xas_create(struct xa_state *xas, bool allow_root) { struct xarray *xa = xas->xa; void *entry; @@ -628,6 +631,8 @@ static void *xas_create(struct xa_state *xas) shift = xas_expand(xas, entry); if (shift < 0) return NULL; + if (!shift && !allow_root) + shift = XA_CHUNK_SHIFT; entry = xa_head_locked(xa); slot = &xa->xa_head; } else if (xas_error(xas)) { @@ -687,7 +692,7 @@ void xas_create_range(struct xa_state *xas) xas->xa_sibs = 0; for (;;) { - xas_create(xas); + xas_create(xas, true); if (xas_error(xas)) goto restore; if (xas->xa_index <= (index | XA_CHUNK_MASK)) @@ -754,7 +759,7 @@ void *xas_store(struct xa_state *xas, void *entry) bool value = xa_is_value(entry); if (entry) - first = xas_create(xas); + first = xas_create(xas, !xa_is_node(entry)); else first = xas_load(xas); @@ -1279,7 +1284,6 @@ static void *xas_result(struct xa_state *xas, void *curr) { if (xa_is_zero(curr)) return NULL; - XA_NODE_BUG_ON(xas->xa_node, xa_is_internal(curr)); if (xas_error(xas)) curr = xas->xa_node; return curr; @@ -1349,7 +1353,7 @@ void *__xa_store(struct xarray *xa, unsigned long index, void *entry, gfp_t gfp) XA_STATE(xas, xa, index); void *curr; - if (WARN_ON_ONCE(xa_is_internal(entry))) + if (WARN_ON_ONCE(xa_is_advanced(entry))) return XA_ERROR(-EINVAL); if (xa_track_free(xa) && !entry) entry = XA_ZERO_ENTRY; @@ -1415,7 +1419,7 @@ void *__xa_cmpxchg(struct xarray *xa, unsigned long index, XA_STATE(xas, xa, index); void *curr; - if (WARN_ON_ONCE(xa_is_internal(entry))) + if (WARN_ON_ONCE(xa_is_advanced(entry))) return XA_ERROR(-EINVAL); if (xa_track_free(xa) && !entry) entry = XA_ZERO_ENTRY; @@ -1538,7 +1542,7 @@ void *xa_store_range(struct xarray *xa, unsigned long first, if (last + 1) order = __ffs(last + 1); xas_set_order(&xas, last, order); - xas_create(&xas); + xas_create(&xas, true); if (xas_error(&xas)) goto unlock; } @@ -1580,7 +1584,7 @@ int __xa_alloc(struct xarray *xa, u32 *id, u32 max, void *entry, gfp_t gfp) XA_STATE(xas, xa, 0); int err; - if (WARN_ON_ONCE(xa_is_internal(entry))) + if (WARN_ON_ONCE(xa_is_advanced(entry))) return -EINVAL; if (WARN_ON_ONCE(!xa_track_free(xa))) return -EINVAL; -- cgit v1.2.3 From b0606fed6eece16a421034eca0bbea9a08b90e91 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Wed, 2 Jan 2019 13:57:03 -0500 Subject: XArray: Honour reserved entries in xa_insert xa_insert() should treat reserved entries as occupied, not as available. Also, it should treat requests to insert a NULL pointer as a request to reserve the slot. Add xa_insert_bh() and xa_insert_irq() for completeness. Signed-off-by: Matthew Wilcox --- Documentation/core-api/xarray.rst | 15 +++--- include/linux/xarray.h | 110 ++++++++++++++++++++++++-------------- lib/test_xarray.c | 8 +-- lib/xarray.c | 41 ++++++++++++++ 4 files changed, 126 insertions(+), 48 deletions(-) (limited to 'include/linux') diff --git a/Documentation/core-api/xarray.rst b/Documentation/core-api/xarray.rst index 6a6d67acaf69..5d54b27c6eba 100644 --- a/Documentation/core-api/xarray.rst +++ b/Documentation/core-api/xarray.rst @@ -108,12 +108,13 @@ some, but not all of the other indices changing. Sometimes you need to ensure that a subsequent call to :c:func:`xa_store` will not need to allocate memory. The :c:func:`xa_reserve` function -will store a reserved entry at the indicated index. Users of the normal -API will see this entry as containing ``NULL``. If you do not need to -use the reserved entry, you can call :c:func:`xa_release` to remove the -unused entry. If another user has stored to the entry in the meantime, -:c:func:`xa_release` will do nothing; if instead you want the entry to -become ``NULL``, you should use :c:func:`xa_erase`. +will store a reserved entry at the indicated index. Users of the +normal API will see this entry as containing ``NULL``. If you do +not need to use the reserved entry, you can call :c:func:`xa_release` +to remove the unused entry. If another user has stored to the entry +in the meantime, :c:func:`xa_release` will do nothing; if instead you +want the entry to become ``NULL``, you should use :c:func:`xa_erase`. +Using :c:func:`xa_insert` on a reserved entry will fail. If all entries in the array are ``NULL``, the :c:func:`xa_empty` function will return ``true``. @@ -183,6 +184,8 @@ Takes xa_lock internally: * :c:func:`xa_store_bh` * :c:func:`xa_store_irq` * :c:func:`xa_insert` + * :c:func:`xa_insert_bh` + * :c:func:`xa_insert_irq` * :c:func:`xa_erase` * :c:func:`xa_erase_bh` * :c:func:`xa_erase_irq` diff --git a/include/linux/xarray.h b/include/linux/xarray.h index 435c25b29079..12244aa98a69 100644 --- a/include/linux/xarray.h +++ b/include/linux/xarray.h @@ -463,39 +463,12 @@ void *__xa_erase(struct xarray *, unsigned long index); void *__xa_store(struct xarray *, unsigned long index, void *entry, gfp_t); void *__xa_cmpxchg(struct xarray *, unsigned long index, void *old, void *entry, gfp_t); +int __xa_insert(struct xarray *, unsigned long index, void *entry, gfp_t); int __xa_alloc(struct xarray *, u32 *id, u32 max, void *entry, gfp_t); int __xa_reserve(struct xarray *, unsigned long index, gfp_t); void __xa_set_mark(struct xarray *, unsigned long index, xa_mark_t); void __xa_clear_mark(struct xarray *, unsigned long index, xa_mark_t); -/** - * __xa_insert() - Store this entry in the XArray unless another entry is - * already present. - * @xa: XArray. - * @index: Index into array. - * @entry: New entry. - * @gfp: Memory allocation flags. - * - * If you would rather see the existing entry in the array, use __xa_cmpxchg(). - * This function is for users who don't care what the entry is, only that - * one is present. - * - * Context: Any context. Expects xa_lock to be held on entry. May - * release and reacquire xa_lock if the @gfp flags permit. - * Return: 0 if the store succeeded. -EEXIST if another entry was present. - * -ENOMEM if memory could not be allocated. - */ -static inline int __xa_insert(struct xarray *xa, unsigned long index, - void *entry, gfp_t gfp) -{ - void *curr = __xa_cmpxchg(xa, index, NULL, entry, gfp); - if (!curr) - return 0; - if (xa_is_err(curr)) - return xa_err(curr); - return -EEXIST; -} - /** * xa_store_bh() - Store this entry in the XArray. * @xa: XArray. @@ -685,24 +658,83 @@ static inline void *xa_cmpxchg_irq(struct xarray *xa, unsigned long index, * @entry: New entry. * @gfp: Memory allocation flags. * - * If you would rather see the existing entry in the array, use xa_cmpxchg(). - * This function is for users who don't care what the entry is, only that - * one is present. + * Inserting a NULL entry will store a reserved entry (like xa_reserve()) + * if no entry is present. Inserting will fail if a reserved entry is + * present, even though loading from this index will return NULL. * - * Context: Process context. Takes and releases the xa_lock. - * May sleep if the @gfp flags permit. + * Context: Any context. Takes and releases the xa_lock. May sleep if + * the @gfp flags permit. * Return: 0 if the store succeeded. -EEXIST if another entry was present. * -ENOMEM if memory could not be allocated. */ static inline int xa_insert(struct xarray *xa, unsigned long index, void *entry, gfp_t gfp) { - void *curr = xa_cmpxchg(xa, index, NULL, entry, gfp); - if (!curr) - return 0; - if (xa_is_err(curr)) - return xa_err(curr); - return -EEXIST; + int err; + + xa_lock(xa); + err = __xa_insert(xa, index, entry, gfp); + xa_unlock(xa); + + return err; +} + +/** + * xa_insert_bh() - Store this entry in the XArray unless another entry is + * already present. + * @xa: XArray. + * @index: Index into array. + * @entry: New entry. + * @gfp: Memory allocation flags. + * + * Inserting a NULL entry will store a reserved entry (like xa_reserve()) + * if no entry is present. Inserting will fail if a reserved entry is + * present, even though loading from this index will return NULL. + * + * Context: Any context. Takes and releases the xa_lock while + * disabling softirqs. May sleep if the @gfp flags permit. + * Return: 0 if the store succeeded. -EEXIST if another entry was present. + * -ENOMEM if memory could not be allocated. + */ +static inline int xa_insert_bh(struct xarray *xa, unsigned long index, + void *entry, gfp_t gfp) +{ + int err; + + xa_lock_bh(xa); + err = __xa_insert(xa, index, entry, gfp); + xa_unlock_bh(xa); + + return err; +} + +/** + * xa_insert_irq() - Store this entry in the XArray unless another entry is + * already present. + * @xa: XArray. + * @index: Index into array. + * @entry: New entry. + * @gfp: Memory allocation flags. + * + * Inserting a NULL entry will store a reserved entry (like xa_reserve()) + * if no entry is present. Inserting will fail if a reserved entry is + * present, even though loading from this index will return NULL. + * + * Context: Process context. Takes and releases the xa_lock while + * disabling interrupts. May sleep if the @gfp flags permit. + * Return: 0 if the store succeeded. -EEXIST if another entry was present. + * -ENOMEM if memory could not be allocated. + */ +static inline int xa_insert_irq(struct xarray *xa, unsigned long index, + void *entry, gfp_t gfp) +{ + int err; + + xa_lock_irq(xa); + err = __xa_insert(xa, index, entry, gfp); + xa_unlock_irq(xa); + + return err; } /** diff --git a/lib/test_xarray.c b/lib/test_xarray.c index 6e0212a60b08..3cf17338b0a4 100644 --- a/lib/test_xarray.c +++ b/lib/test_xarray.c @@ -382,10 +382,12 @@ static noinline void check_reserve(struct xarray *xa) xa_erase_index(xa, 12345678); XA_BUG_ON(xa, !xa_empty(xa)); - /* And so does xa_insert */ + /* But xa_insert does not */ xa_reserve(xa, 12345678, GFP_KERNEL); - XA_BUG_ON(xa, xa_insert(xa, 12345678, xa_mk_value(12345678), 0) != 0); - xa_erase_index(xa, 12345678); + XA_BUG_ON(xa, xa_insert(xa, 12345678, xa_mk_value(12345678), 0) != + -EEXIST); + XA_BUG_ON(xa, xa_empty(xa)); + XA_BUG_ON(xa, xa_erase(xa, 12345678) != NULL); XA_BUG_ON(xa, !xa_empty(xa)); /* Can iterate through a reserved entry */ diff --git a/lib/xarray.c b/lib/xarray.c index bffa26b1f0d6..81c3171ddde9 100644 --- a/lib/xarray.c +++ b/lib/xarray.c @@ -1439,6 +1439,47 @@ void *__xa_cmpxchg(struct xarray *xa, unsigned long index, } EXPORT_SYMBOL(__xa_cmpxchg); +/** + * __xa_insert() - Store this entry in the XArray if no entry is present. + * @xa: XArray. + * @index: Index into array. + * @entry: New entry. + * @gfp: Memory allocation flags. + * + * Inserting a NULL entry will store a reserved entry (like xa_reserve()) + * if no entry is present. Inserting will fail if a reserved entry is + * present, even though loading from this index will return NULL. + * + * Context: Any context. Expects xa_lock to be held on entry. May + * release and reacquire xa_lock if @gfp flags permit. + * Return: 0 if the store succeeded. -EEXIST if another entry was present. + * -ENOMEM if memory could not be allocated. + */ +int __xa_insert(struct xarray *xa, unsigned long index, void *entry, gfp_t gfp) +{ + XA_STATE(xas, xa, index); + void *curr; + + if (WARN_ON_ONCE(xa_is_advanced(entry))) + return -EINVAL; + if (!entry) + entry = XA_ZERO_ENTRY; + + do { + curr = xas_load(&xas); + if (!curr) { + xas_store(&xas, entry); + if (xa_track_free(xa)) + xas_clear_mark(&xas, XA_FREE_MARK); + } else { + xas_set_err(&xas, -EEXIST); + } + } while (__xas_nomem(&xas, gfp)); + + return xas_error(&xas); +} +EXPORT_SYMBOL(__xa_insert); + /** * __xa_reserve() - Reserve this index in the XArray. * @xa: XArray. -- cgit v1.2.3 From 1cb95e072ede5e3d6a54eefd520db21b45985896 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Tue, 8 Jan 2019 15:34:52 -0800 Subject: libnvdimm/dimm: Fix security capability detection for non-Intel NVDIMMs Kees reports a crash with the following signature... RIP: 0010:nvdimm_visible+0x79/0x80 [..] Call Trace: internal_create_group+0xf4/0x380 sysfs_create_groups+0x46/0xb0 device_add+0x331/0x680 nd_async_device_register+0x15/0x60 async_run_entry_fn+0x38/0x100 ...when starting a QEMU environment with "label-less" DIMM. Without labels QEMU does not publish any DSM methods. Without defined methods the NVDIMM_FAMILY type is not established and the nfit driver will skip registering security operations. In that case the security state should be initialized to a negative value in __nvdimm_create() and nvdimm_visible() should skip interrogating the specific ops. However, since 'enum nvdimm_security_state' was only defined to contain positive values the "if (nvdimm->sec.state < 0)" check always fails. Define a negative error state to allow negative state values to be handled as expected. Fixes: f2989396553a ("acpi/nfit, libnvdimm: Introduce nvdimm_security_ops") Reviewed-by: Dave Jiang Reported-by: Kees Cook Tested-by: Kees Cook Signed-off-by: Dan Williams --- include/linux/libnvdimm.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h index 5440f11b0907..7315977b64da 100644 --- a/include/linux/libnvdimm.h +++ b/include/linux/libnvdimm.h @@ -160,6 +160,7 @@ static inline struct nd_blk_region_desc *to_blk_region_desc( } enum nvdimm_security_state { + NVDIMM_SECURITY_ERROR = -1, NVDIMM_SECURITY_DISABLED, NVDIMM_SECURITY_UNLOCKED, NVDIMM_SECURITY_LOCKED, -- cgit v1.2.3 From 3e2ffd655cc6a694608d997738989ff5572a8266 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Wed, 2 Jan 2019 15:57:49 -0500 Subject: include/linux/compiler*.h: fix OPTIMIZER_HIDE_VAR Since commit 815f0ddb346c ("include/linux/compiler*.h: make compiler-*.h mutually exclusive") clang no longer reuses the OPTIMIZER_HIDE_VAR macro from compiler-gcc - instead it gets the version in include/linux/compiler.h. Unfortunately that version doesn't actually prevent compiler from optimizing out the variable. Fix up by moving the macro out from compiler-gcc.h to compiler.h. Compilers without incline asm support will keep working since it's protected by an ifdef. Also fix up comments to match reality since we are no longer overriding any macros. Build-tested with gcc and clang. Fixes: 815f0ddb346c ("include/linux/compiler*.h: make compiler-*.h mutually exclusive") Cc: Eli Friedman Cc: Joe Perches Cc: Linus Torvalds Reviewed-by: Nick Desaulniers Signed-off-by: Michael S. Tsirkin Signed-off-by: Miguel Ojeda --- include/linux/compiler-clang.h | 5 ++--- include/linux/compiler-gcc.h | 4 ---- include/linux/compiler-intel.h | 4 +--- include/linux/compiler.h | 4 +++- 4 files changed, 6 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compiler-clang.h b/include/linux/compiler-clang.h index 39f668d5066b..333a6695a918 100644 --- a/include/linux/compiler-clang.h +++ b/include/linux/compiler-clang.h @@ -3,9 +3,8 @@ #error "Please don't include directly, include instead." #endif -/* Some compiler specific definitions are overwritten here - * for Clang compiler - */ +/* Compiler specific definitions for Clang compiler */ + #define uninitialized_var(x) x = *(&(x)) /* same as gcc, this was present in clang-2.6 so we can assume it works diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h index 5776da43da97..7b834e37d0c0 100644 --- a/include/linux/compiler-gcc.h +++ b/include/linux/compiler-gcc.h @@ -58,10 +58,6 @@ (typeof(ptr)) (__ptr + (off)); \ }) -/* Make the optimizer believe the variable can be manipulated arbitrarily. */ -#define OPTIMIZER_HIDE_VAR(var) \ - __asm__ ("" : "=r" (var) : "0" (var)) - /* * A trick to suppress uninitialized variable warning without generating any * code diff --git a/include/linux/compiler-intel.h b/include/linux/compiler-intel.h index 517bd14e1222..b17f3cd18334 100644 --- a/include/linux/compiler-intel.h +++ b/include/linux/compiler-intel.h @@ -5,9 +5,7 @@ #ifdef __ECC -/* Some compiler specific definitions are overwritten here - * for Intel ECC compiler - */ +/* Compiler specific definitions for Intel ECC compiler */ #include diff --git a/include/linux/compiler.h b/include/linux/compiler.h index fc5004a4b07d..445348facea9 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -161,7 +161,9 @@ void ftrace_likely_update(struct ftrace_likely_data *f, int val, #endif #ifndef OPTIMIZER_HIDE_VAR -#define OPTIMIZER_HIDE_VAR(var) barrier() +/* Make the optimizer believe the variable can be manipulated arbitrarily. */ +#define OPTIMIZER_HIDE_VAR(var) \ + __asm__ ("" : "=r" (var) : "0" (var)) #endif /* Not-quite-unique ID. */ -- cgit v1.2.3 From ba50bf1ce9a51fc97db58b96d01306aa70bc3979 Mon Sep 17 00:00:00 2001 From: Dexuan Cui Date: Mon, 17 Dec 2018 20:16:09 +0000 Subject: Drivers: hv: vmbus: Check for ring when getting debug info fc96df16a1ce is good and can already fix the "return stack garbage" issue, but let's also improve hv_ringbuffer_get_debuginfo(), which would silently return stack garbage, if people forget to check channel->state or ring_info->ring_buffer, when using the function in the future. Having an error check in the function would eliminate the potential risk. Add a Fixes tag to indicate the patch depdendency. Fixes: fc96df16a1ce ("Drivers: hv: vmbus: Return -EINVAL for the sys files for unopened channels") Cc: stable@vger.kernel.org Cc: K. Y. Srinivasan Cc: Haiyang Zhang Signed-off-by: Stephen Hemminger Signed-off-by: Dexuan Cui Signed-off-by: Sasha Levin --- drivers/hv/ring_buffer.c | 31 ++++++++--------- drivers/hv/vmbus_drv.c | 91 ++++++++++++++++++++++++++++++++---------------- include/linux/hyperv.h | 5 +-- 3 files changed, 79 insertions(+), 48 deletions(-) (limited to 'include/linux') diff --git a/drivers/hv/ring_buffer.c b/drivers/hv/ring_buffer.c index 64d0c85d5161..1f1a55e07733 100644 --- a/drivers/hv/ring_buffer.c +++ b/drivers/hv/ring_buffer.c @@ -164,26 +164,25 @@ hv_get_ringbuffer_availbytes(const struct hv_ring_buffer_info *rbi, } /* Get various debug metrics for the specified ring buffer. */ -void hv_ringbuffer_get_debuginfo(const struct hv_ring_buffer_info *ring_info, - struct hv_ring_buffer_debug_info *debug_info) +int hv_ringbuffer_get_debuginfo(const struct hv_ring_buffer_info *ring_info, + struct hv_ring_buffer_debug_info *debug_info) { u32 bytes_avail_towrite; u32 bytes_avail_toread; - if (ring_info->ring_buffer) { - hv_get_ringbuffer_availbytes(ring_info, - &bytes_avail_toread, - &bytes_avail_towrite); - - debug_info->bytes_avail_toread = bytes_avail_toread; - debug_info->bytes_avail_towrite = bytes_avail_towrite; - debug_info->current_read_index = - ring_info->ring_buffer->read_index; - debug_info->current_write_index = - ring_info->ring_buffer->write_index; - debug_info->current_interrupt_mask = - ring_info->ring_buffer->interrupt_mask; - } + if (!ring_info->ring_buffer) + return -EINVAL; + + hv_get_ringbuffer_availbytes(ring_info, + &bytes_avail_toread, + &bytes_avail_towrite); + debug_info->bytes_avail_toread = bytes_avail_toread; + debug_info->bytes_avail_towrite = bytes_avail_towrite; + debug_info->current_read_index = ring_info->ring_buffer->read_index; + debug_info->current_write_index = ring_info->ring_buffer->write_index; + debug_info->current_interrupt_mask + = ring_info->ring_buffer->interrupt_mask; + return 0; } EXPORT_SYMBOL_GPL(hv_ringbuffer_get_debuginfo); diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c index d0ff65675292..403fee01572c 100644 --- a/drivers/hv/vmbus_drv.c +++ b/drivers/hv/vmbus_drv.c @@ -313,12 +313,16 @@ static ssize_t out_intr_mask_show(struct device *dev, { struct hv_device *hv_dev = device_to_hv_device(dev); struct hv_ring_buffer_debug_info outbound; + int ret; if (!hv_dev->channel) return -ENODEV; - if (hv_dev->channel->state != CHANNEL_OPENED_STATE) - return -EINVAL; - hv_ringbuffer_get_debuginfo(&hv_dev->channel->outbound, &outbound); + + ret = hv_ringbuffer_get_debuginfo(&hv_dev->channel->outbound, + &outbound); + if (ret < 0) + return ret; + return sprintf(buf, "%d\n", outbound.current_interrupt_mask); } static DEVICE_ATTR_RO(out_intr_mask); @@ -328,12 +332,15 @@ static ssize_t out_read_index_show(struct device *dev, { struct hv_device *hv_dev = device_to_hv_device(dev); struct hv_ring_buffer_debug_info outbound; + int ret; if (!hv_dev->channel) return -ENODEV; - if (hv_dev->channel->state != CHANNEL_OPENED_STATE) - return -EINVAL; - hv_ringbuffer_get_debuginfo(&hv_dev->channel->outbound, &outbound); + + ret = hv_ringbuffer_get_debuginfo(&hv_dev->channel->outbound, + &outbound); + if (ret < 0) + return ret; return sprintf(buf, "%d\n", outbound.current_read_index); } static DEVICE_ATTR_RO(out_read_index); @@ -344,12 +351,15 @@ static ssize_t out_write_index_show(struct device *dev, { struct hv_device *hv_dev = device_to_hv_device(dev); struct hv_ring_buffer_debug_info outbound; + int ret; if (!hv_dev->channel) return -ENODEV; - if (hv_dev->channel->state != CHANNEL_OPENED_STATE) - return -EINVAL; - hv_ringbuffer_get_debuginfo(&hv_dev->channel->outbound, &outbound); + + ret = hv_ringbuffer_get_debuginfo(&hv_dev->channel->outbound, + &outbound); + if (ret < 0) + return ret; return sprintf(buf, "%d\n", outbound.current_write_index); } static DEVICE_ATTR_RO(out_write_index); @@ -360,12 +370,15 @@ static ssize_t out_read_bytes_avail_show(struct device *dev, { struct hv_device *hv_dev = device_to_hv_device(dev); struct hv_ring_buffer_debug_info outbound; + int ret; if (!hv_dev->channel) return -ENODEV; - if (hv_dev->channel->state != CHANNEL_OPENED_STATE) - return -EINVAL; - hv_ringbuffer_get_debuginfo(&hv_dev->channel->outbound, &outbound); + + ret = hv_ringbuffer_get_debuginfo(&hv_dev->channel->outbound, + &outbound); + if (ret < 0) + return ret; return sprintf(buf, "%d\n", outbound.bytes_avail_toread); } static DEVICE_ATTR_RO(out_read_bytes_avail); @@ -376,12 +389,15 @@ static ssize_t out_write_bytes_avail_show(struct device *dev, { struct hv_device *hv_dev = device_to_hv_device(dev); struct hv_ring_buffer_debug_info outbound; + int ret; if (!hv_dev->channel) return -ENODEV; - if (hv_dev->channel->state != CHANNEL_OPENED_STATE) - return -EINVAL; - hv_ringbuffer_get_debuginfo(&hv_dev->channel->outbound, &outbound); + + ret = hv_ringbuffer_get_debuginfo(&hv_dev->channel->outbound, + &outbound); + if (ret < 0) + return ret; return sprintf(buf, "%d\n", outbound.bytes_avail_towrite); } static DEVICE_ATTR_RO(out_write_bytes_avail); @@ -391,12 +407,15 @@ static ssize_t in_intr_mask_show(struct device *dev, { struct hv_device *hv_dev = device_to_hv_device(dev); struct hv_ring_buffer_debug_info inbound; + int ret; if (!hv_dev->channel) return -ENODEV; - if (hv_dev->channel->state != CHANNEL_OPENED_STATE) - return -EINVAL; - hv_ringbuffer_get_debuginfo(&hv_dev->channel->inbound, &inbound); + + ret = hv_ringbuffer_get_debuginfo(&hv_dev->channel->inbound, &inbound); + if (ret < 0) + return ret; + return sprintf(buf, "%d\n", inbound.current_interrupt_mask); } static DEVICE_ATTR_RO(in_intr_mask); @@ -406,12 +425,15 @@ static ssize_t in_read_index_show(struct device *dev, { struct hv_device *hv_dev = device_to_hv_device(dev); struct hv_ring_buffer_debug_info inbound; + int ret; if (!hv_dev->channel) return -ENODEV; - if (hv_dev->channel->state != CHANNEL_OPENED_STATE) - return -EINVAL; - hv_ringbuffer_get_debuginfo(&hv_dev->channel->inbound, &inbound); + + ret = hv_ringbuffer_get_debuginfo(&hv_dev->channel->inbound, &inbound); + if (ret < 0) + return ret; + return sprintf(buf, "%d\n", inbound.current_read_index); } static DEVICE_ATTR_RO(in_read_index); @@ -421,12 +443,15 @@ static ssize_t in_write_index_show(struct device *dev, { struct hv_device *hv_dev = device_to_hv_device(dev); struct hv_ring_buffer_debug_info inbound; + int ret; if (!hv_dev->channel) return -ENODEV; - if (hv_dev->channel->state != CHANNEL_OPENED_STATE) - return -EINVAL; - hv_ringbuffer_get_debuginfo(&hv_dev->channel->inbound, &inbound); + + ret = hv_ringbuffer_get_debuginfo(&hv_dev->channel->inbound, &inbound); + if (ret < 0) + return ret; + return sprintf(buf, "%d\n", inbound.current_write_index); } static DEVICE_ATTR_RO(in_write_index); @@ -437,12 +462,15 @@ static ssize_t in_read_bytes_avail_show(struct device *dev, { struct hv_device *hv_dev = device_to_hv_device(dev); struct hv_ring_buffer_debug_info inbound; + int ret; if (!hv_dev->channel) return -ENODEV; - if (hv_dev->channel->state != CHANNEL_OPENED_STATE) - return -EINVAL; - hv_ringbuffer_get_debuginfo(&hv_dev->channel->inbound, &inbound); + + ret = hv_ringbuffer_get_debuginfo(&hv_dev->channel->inbound, &inbound); + if (ret < 0) + return ret; + return sprintf(buf, "%d\n", inbound.bytes_avail_toread); } static DEVICE_ATTR_RO(in_read_bytes_avail); @@ -453,12 +481,15 @@ static ssize_t in_write_bytes_avail_show(struct device *dev, { struct hv_device *hv_dev = device_to_hv_device(dev); struct hv_ring_buffer_debug_info inbound; + int ret; if (!hv_dev->channel) return -ENODEV; - if (hv_dev->channel->state != CHANNEL_OPENED_STATE) - return -EINVAL; - hv_ringbuffer_get_debuginfo(&hv_dev->channel->inbound, &inbound); + + ret = hv_ringbuffer_get_debuginfo(&hv_dev->channel->inbound, &inbound); + if (ret < 0) + return ret; + return sprintf(buf, "%d\n", inbound.bytes_avail_towrite); } static DEVICE_ATTR_RO(in_write_bytes_avail); diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h index f0885cc01db6..dcb6977afce9 100644 --- a/include/linux/hyperv.h +++ b/include/linux/hyperv.h @@ -1159,8 +1159,9 @@ struct hv_ring_buffer_debug_info { u32 bytes_avail_towrite; }; -void hv_ringbuffer_get_debuginfo(const struct hv_ring_buffer_info *ring_info, - struct hv_ring_buffer_debug_info *debug_info); + +int hv_ringbuffer_get_debuginfo(const struct hv_ring_buffer_info *ring_info, + struct hv_ring_buffer_debug_info *debug_info); /* Vmbus interface */ #define vmbus_driver_register(driver) \ -- cgit v1.2.3 From 321c46b91550adc03054125fa7a1639390608e1a Mon Sep 17 00:00:00 2001 From: Rafał Miłecki Date: Thu, 3 Jan 2019 08:34:17 +0100 Subject: MIPS: BCM47XX: Setup struct device for the SoC MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit So far we never had any device registered for the SoC. This resulted in some small issues that we kept ignoring like: 1) Not working GPIOLIB_IRQCHIP (gpiochip_irqchip_add_key() failing) 2) Lack of proper tree in the /sys/devices/ 3) mips_dma_alloc_coherent() silently handling empty coherent_dma_mask Kernel 4.19 came with a lot of DMA changes and caused a regression on bcm47xx. Starting with the commit f8c55dc6e828 ("MIPS: use generic dma noncoherent ops for simple noncoherent platforms") DMA coherent allocations just fail. Example: [ 1.114914] bgmac_bcma bcma0:2: Allocation of TX ring 0x200 failed [ 1.121215] bgmac_bcma bcma0:2: Unable to alloc memory for DMA [ 1.127626] bgmac_bcma: probe of bcma0:2 failed with error -12 [ 1.133838] bgmac_bcma: Broadcom 47xx GBit MAC driver loaded The bgmac driver also triggers a WARNING: [ 0.959486] ------------[ cut here ]------------ [ 0.964387] WARNING: CPU: 0 PID: 1 at ./include/linux/dma-mapping.h:516 bgmac_enet_probe+0x1b4/0x5c4 [ 0.973751] Modules linked in: [ 0.976913] CPU: 0 PID: 1 Comm: swapper Not tainted 4.19.9 #0 [ 0.982750] Stack : 804a0000 804597c4 00000000 00000000 80458fd8 8381bc2c 838282d4 80481a47 [ 0.991367] 8042e3ec 00000001 804d38f0 00000204 83980000 00000065 8381bbe0 6f55b24f [ 0.999975] 00000000 00000000 80520000 00002018 00000000 00000075 00000007 00000000 [ 1.008583] 00000000 80480000 000ee811 00000000 00000000 00000000 80432c00 80248db8 [ 1.017196] 00000009 00000204 83980000 803ad7b0 00000000 801feeec 00000000 804d0000 [ 1.025804] ... [ 1.028325] Call Trace: [ 1.030875] [<8000aef8>] show_stack+0x58/0x100 [ 1.035513] [<8001f8b4>] __warn+0xe4/0x118 [ 1.039708] [<8001f9a4>] warn_slowpath_null+0x48/0x64 [ 1.044935] [<80248db8>] bgmac_enet_probe+0x1b4/0x5c4 [ 1.050101] [<802498e0>] bgmac_probe+0x558/0x590 [ 1.054906] [<80252fd0>] bcma_device_probe+0x38/0x70 [ 1.060017] [<8020e1e8>] really_probe+0x170/0x2e8 [ 1.064891] [<8020e714>] __driver_attach+0xa4/0xec [ 1.069784] [<8020c1e0>] bus_for_each_dev+0x58/0xb0 [ 1.074833] [<8020d590>] bus_add_driver+0xf8/0x218 [ 1.079731] [<8020ef24>] driver_register+0xcc/0x11c [ 1.084804] [<804b54cc>] bgmac_init+0x1c/0x44 [ 1.089258] [<8000121c>] do_one_initcall+0x7c/0x1a0 [ 1.094343] [<804a1d34>] kernel_init_freeable+0x150/0x218 [ 1.099886] [<803a082c>] kernel_init+0x10/0x104 [ 1.104583] [<80005878>] ret_from_kernel_thread+0x14/0x1c [ 1.110107] ---[ end trace f441c0d873d1fb5b ]--- This patch setups a "struct device" (and passes it to the bcma) which allows fixing all the mentioned problems. It'll also require a tiny bcma patch which will follow through the wireless tree & its maintainer. Fixes: f8c55dc6e828 ("MIPS: use generic dma noncoherent ops for simple noncoherent platforms") Signed-off-by: Rafał Miłecki Signed-off-by: Paul Burton Acked-by: Hauke Mehrtens Cc: Christoph Hellwig Cc: Linus Walleij Cc: linux-wireless@vger.kernel.org Cc: Ralf Baechle Cc: James Hogan Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Cc: stable@vger.kernel.org # v4.19+ --- arch/mips/bcm47xx/setup.c | 31 +++++++++++++++++++++++++++++++ include/linux/bcma/bcma_soc.h | 1 + 2 files changed, 32 insertions(+) (limited to 'include/linux') diff --git a/arch/mips/bcm47xx/setup.c b/arch/mips/bcm47xx/setup.c index 6054d49e608e..fe3773539eff 100644 --- a/arch/mips/bcm47xx/setup.c +++ b/arch/mips/bcm47xx/setup.c @@ -173,6 +173,31 @@ void __init plat_mem_setup(void) pm_power_off = bcm47xx_machine_halt; } +#ifdef CONFIG_BCM47XX_BCMA +static struct device * __init bcm47xx_setup_device(void) +{ + struct device *dev; + int err; + + dev = kzalloc(sizeof(*dev), GFP_KERNEL); + if (!dev) + return NULL; + + err = dev_set_name(dev, "bcm47xx_soc"); + if (err) { + pr_err("Failed to set SoC device name: %d\n", err); + kfree(dev); + return NULL; + } + + err = dma_coerce_mask_and_coherent(dev, DMA_BIT_MASK(32)); + if (err) + pr_err("Failed to set SoC DMA mask: %d\n", err); + + return dev; +} +#endif + /* * This finishes bus initialization doing things that were not possible without * kmalloc. Make sure to call it late enough (after mm_init). @@ -183,6 +208,10 @@ void __init bcm47xx_bus_setup(void) if (bcm47xx_bus_type == BCM47XX_BUS_TYPE_BCMA) { int err; + bcm47xx_bus.bcma.dev = bcm47xx_setup_device(); + if (!bcm47xx_bus.bcma.dev) + panic("Failed to setup SoC device\n"); + err = bcma_host_soc_init(&bcm47xx_bus.bcma); if (err) panic("Failed to initialize BCMA bus (err %d)", err); @@ -235,6 +264,8 @@ static int __init bcm47xx_register_bus_complete(void) #endif #ifdef CONFIG_BCM47XX_BCMA case BCM47XX_BUS_TYPE_BCMA: + if (device_register(bcm47xx_bus.bcma.dev)) + pr_err("Failed to register SoC device\n"); bcma_bus_register(&bcm47xx_bus.bcma.bus); break; #endif diff --git a/include/linux/bcma/bcma_soc.h b/include/linux/bcma/bcma_soc.h index 7cca5f859a90..f3c43519baa7 100644 --- a/include/linux/bcma/bcma_soc.h +++ b/include/linux/bcma/bcma_soc.h @@ -6,6 +6,7 @@ struct bcma_soc { struct bcma_bus bus; + struct device *dev; }; int __init bcma_host_soc_register(struct bcma_soc *soc); -- cgit v1.2.3 From ee46967fc6e74d412fe1ec15f77fdb8624bde2b0 Mon Sep 17 00:00:00 2001 From: Peter Hutterer Date: Wed, 9 Jan 2019 13:50:18 +1000 Subject: HID: core: replace the collection tree pointers with indices Previously, the pointer to the parent collection was stored. If a device exceeds 16 collections (HID_DEFAULT_NUM_COLLECTIONS), the array to store the collections is reallocated, the pointer to the parent collection becomes invalid. Replace the pointers with an index-based lookup into the collections array. Fixes: c53431eb696f3c ("HID: core: store the collections as a basic tree") Reported-by: Pandruvada, Srinivas Signed-off-by: Peter Hutterer Tested-by: Kyle Pelton Signed-off-by: Jiri Kosina --- drivers/hid/hid-core.c | 32 +++++++++++++++++++++----------- include/linux/hid.h | 4 ++-- 2 files changed, 23 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c index f41d5fe51abe..f9093dedf647 100644 --- a/drivers/hid/hid-core.c +++ b/drivers/hid/hid-core.c @@ -125,6 +125,7 @@ static int open_collection(struct hid_parser *parser, unsigned type) { struct hid_collection *collection; unsigned usage; + int collection_index; usage = parser->local.usage[0]; @@ -167,13 +168,13 @@ static int open_collection(struct hid_parser *parser, unsigned type) parser->collection_stack[parser->collection_stack_ptr++] = parser->device->maxcollection; - collection = parser->device->collection + - parser->device->maxcollection++; + collection_index = parser->device->maxcollection++; + collection = parser->device->collection + collection_index; collection->type = type; collection->usage = usage; collection->level = parser->collection_stack_ptr - 1; - collection->parent = parser->active_collection; - parser->active_collection = collection; + collection->parent_idx = parser->active_collection_idx; + parser->active_collection_idx = collection_index; if (type == HID_COLLECTION_APPLICATION) parser->device->maxapplication++; @@ -192,8 +193,13 @@ static int close_collection(struct hid_parser *parser) return -EINVAL; } parser->collection_stack_ptr--; - if (parser->active_collection) - parser->active_collection = parser->active_collection->parent; + if (parser->active_collection_idx != -1) { + struct hid_device *device = parser->device; + struct hid_collection *c; + + c = &device->collection[parser->active_collection_idx]; + parser->active_collection_idx = c->parent_idx; + } return 0; } @@ -819,6 +825,7 @@ static int hid_scan_report(struct hid_device *hid) return -ENOMEM; parser->device = hid; + parser->active_collection_idx = -1; hid->group = HID_GROUP_GENERIC; /* @@ -1006,10 +1013,12 @@ static void hid_apply_multiplier_to_field(struct hid_device *hid, usage = &field->usage[i]; collection = &hid->collection[usage->collection_index]; - while (collection && collection != multiplier_collection) - collection = collection->parent; + while (collection->parent_idx != -1 && + collection != multiplier_collection) + collection = &hid->collection[collection->parent_idx]; - if (collection || multiplier_collection == NULL) + if (collection->parent_idx != -1 || + multiplier_collection == NULL) usage->resolution_multiplier = effective_multiplier; } @@ -1044,9 +1053,9 @@ static void hid_apply_multiplier(struct hid_device *hid, * applicable fields later. */ multiplier_collection = &hid->collection[multiplier->usage->collection_index]; - while (multiplier_collection && + while (multiplier_collection->parent_idx != -1 && multiplier_collection->type != HID_COLLECTION_LOGICAL) - multiplier_collection = multiplier_collection->parent; + multiplier_collection = &hid->collection[multiplier_collection->parent_idx]; effective_multiplier = hid_calculate_multiplier(hid, multiplier); @@ -1170,6 +1179,7 @@ int hid_open_report(struct hid_device *device) } parser->device = device; + parser->active_collection_idx = -1; end = start + size; diff --git a/include/linux/hid.h b/include/linux/hid.h index d99287327ef2..992bbb7196df 100644 --- a/include/linux/hid.h +++ b/include/linux/hid.h @@ -430,7 +430,7 @@ struct hid_local { */ struct hid_collection { - struct hid_collection *parent; + int parent_idx; /* device->collection */ unsigned type; unsigned usage; unsigned level; @@ -658,7 +658,7 @@ struct hid_parser { unsigned int *collection_stack; unsigned int collection_stack_ptr; unsigned int collection_stack_size; - struct hid_collection *active_collection; + int active_collection_idx; /* device->collection */ struct hid_device *device; unsigned int scan_flags; }; -- cgit v1.2.3 From 8ce5f84157530ffa64b3e0acf00b9261f41c8da8 Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Tue, 11 Dec 2018 14:31:05 -0600 Subject: of: Remove struct device_node.type pointer Now that all users of device_node.type pointer have been removed in favor of accessor functions, we can remove it. Cc: Frank Rowand Cc: devicetree@vger.kernel.org Signed-off-by: Rob Herring --- drivers/of/dynamic.c | 3 --- drivers/of/fdt.c | 4 ---- drivers/of/overlay.c | 3 --- drivers/of/pdt.c | 1 - include/linux/of.h | 1 - 5 files changed, 12 deletions(-) (limited to 'include/linux') diff --git a/drivers/of/dynamic.c b/drivers/of/dynamic.c index a09c1c3cf831..49b16f76d78e 100644 --- a/drivers/of/dynamic.c +++ b/drivers/of/dynamic.c @@ -207,11 +207,8 @@ static void __of_attach_node(struct device_node *np) if (!of_node_check_flag(np, OF_OVERLAY)) { np->name = __of_get_property(np, "name", NULL); - np->type = __of_get_property(np, "device_type", NULL); if (!np->name) np->name = ""; - if (!np->type) - np->type = ""; phandle = __of_get_property(np, "phandle", &sz); if (!phandle) diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c index 7099c652c6a5..9cc1461aac7d 100644 --- a/drivers/of/fdt.c +++ b/drivers/of/fdt.c @@ -314,12 +314,8 @@ static bool populate_node(const void *blob, populate_properties(blob, offset, mem, np, pathp, dryrun); if (!dryrun) { np->name = of_get_property(np, "name", NULL); - np->type = of_get_property(np, "device_type", NULL); - if (!np->name) np->name = ""; - if (!np->type) - np->type = ""; } *pnp = np; diff --git a/drivers/of/overlay.c b/drivers/of/overlay.c index 2b5ac43a5690..c423e94baf0f 100644 --- a/drivers/of/overlay.c +++ b/drivers/of/overlay.c @@ -423,12 +423,9 @@ static int add_changeset_node(struct overlay_changeset *ovcs, tchild->parent = target->np; tchild->name = __of_get_property(node, "name", NULL); - tchild->type = __of_get_property(node, "device_type", NULL); if (!tchild->name) tchild->name = ""; - if (!tchild->type) - tchild->type = ""; /* ignore obsolete "linux,phandle" */ phandle = __of_get_property(node, "phandle", &size); diff --git a/drivers/of/pdt.c b/drivers/of/pdt.c index d3185063d369..7eda43c66c91 100644 --- a/drivers/of/pdt.c +++ b/drivers/of/pdt.c @@ -155,7 +155,6 @@ static struct device_node * __init of_pdt_create_node(phandle node, dp->parent = parent; dp->name = of_pdt_get_one_property(node, "name"); - dp->type = of_pdt_get_one_property(node, "device_type"); dp->phandle = node; dp->properties = of_pdt_build_prop_list(node); diff --git a/include/linux/of.h b/include/linux/of.h index fe472e5195a9..e240992e5cb6 100644 --- a/include/linux/of.h +++ b/include/linux/of.h @@ -50,7 +50,6 @@ struct of_irq_controller; struct device_node { const char *name; - const char *type; phandle phandle; const char *full_name; struct fwnode_handle fwnode; -- cgit v1.2.3 From 19ba9ecf24189bd74d070aa1b1c4bcb9fe4ae849 Mon Sep 17 00:00:00 2001 From: Cyrill Gorcunov Date: Mon, 14 Jan 2019 11:40:47 +0300 Subject: XArray: Fix typo in comment Seems copy and paste typo, not a big deal but still for consistency sake better to fix. Signed-off-by: Cyrill Gorcunov Signed-off-by: Matthew Wilcox --- include/linux/xarray.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/xarray.h b/include/linux/xarray.h index 12244aa98a69..7da665f5cb20 100644 --- a/include/linux/xarray.h +++ b/include/linux/xarray.h @@ -496,7 +496,7 @@ static inline void *xa_store_bh(struct xarray *xa, unsigned long index, } /** - * xa_store_irq() - Erase this entry from the XArray. + * xa_store_irq() - Store this entry in the XArray. * @xa: XArray. * @index: Index into array. * @entry: New entry. -- cgit v1.2.3 From b89a07c4373b27321b1f6d4b4fdc369fd45ef79d Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Thu, 3 Jan 2019 17:08:03 +0100 Subject: virtio: fix virtio_config_ops description - get_features has returned 64 bits since commit d025477368792 ("virtio: add support for 64 bit features.") - properly mark all optional callbacks Signed-off-by: Cornelia Huck Signed-off-by: Michael S. Tsirkin Reviewed-by: Halil Pasic --- include/linux/virtio_config.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/virtio_config.h b/include/linux/virtio_config.h index 32baf8e26735..7087ef946ba7 100644 --- a/include/linux/virtio_config.h +++ b/include/linux/virtio_config.h @@ -22,7 +22,7 @@ struct irq_affinity; * offset: the offset of the configuration field * buf: the buffer to read the field value from. * len: the length of the buffer - * @generation: config generation counter + * @generation: config generation counter (optional) * vdev: the virtio_device * Returns the config generation counter * @get_status: read the status byte @@ -48,17 +48,17 @@ struct irq_affinity; * @del_vqs: free virtqueues found by find_vqs(). * @get_features: get the array of feature bits for this device. * vdev: the virtio_device - * Returns the first 32 feature bits (all we currently need). + * Returns the first 64 feature bits (all we currently need). * @finalize_features: confirm what device features we'll be using. * vdev: the virtio_device * This gives the final feature bits for the device: it can change * the dev->feature bits if it wants. * Returns 0 on success or error status - * @bus_name: return the bus name associated with the device + * @bus_name: return the bus name associated with the device (optional) * vdev: the virtio_device * This returns a pointer to the bus name a la pci_name from which * the caller can then copy. - * @set_vq_affinity: set the affinity for a virtqueue. + * @set_vq_affinity: set the affinity for a virtqueue (optional). * @get_vq_affinity: get the affinity for a virtqueue (optional). */ typedef void vq_callback_t(struct virtqueue *); -- cgit v1.2.3 From d1c1dad89e7a8be2cfdc7b92deca2c8048f0d263 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Thu, 3 Jan 2019 17:08:04 +0100 Subject: virtio: document virtio_config_ops restrictions Some transports (e.g. virtio-ccw) implement virtio operations that seem to be a simple read/write as something more involved that cannot be done from an atomic context. Give at least a hint about that. Signed-off-by: Cornelia Huck Signed-off-by: Michael S. Tsirkin --- include/linux/virtio_config.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/virtio_config.h b/include/linux/virtio_config.h index 7087ef946ba7..987b6491b946 100644 --- a/include/linux/virtio_config.h +++ b/include/linux/virtio_config.h @@ -12,6 +12,11 @@ struct irq_affinity; /** * virtio_config_ops - operations for configuring a virtio device + * Note: Do not assume that a transport implements all of the operations + * getting/setting a value as a simple read/write! Generally speaking, + * any of @get/@set, @get_status/@set_status, or @get_features/ + * @finalize_features are NOT safe to be called from an atomic + * context. * @get: read the value of a configuration field * vdev: the virtio_device * offset: the offset of the configuration field -- cgit v1.2.3 From 1950f462916edc9581168ca8d5882a8101e8bbcf Mon Sep 17 00:00:00 2001 From: Philipp Zabel Date: Mon, 14 Jan 2019 08:19:22 +0100 Subject: HID: core: simplify active collection tracking Manually tracking an active collection to set collection parents is not necessary, we just have to look one step back into the collection stack to find the correct parent. Signed-off-by: Philipp Zabel Reviewed-by: Peter Hutterer Signed-off-by: Benjamin Tissoires --- drivers/hid/hid-core.c | 13 ++----------- include/linux/hid.h | 1 - 2 files changed, 2 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c index f9093dedf647..9993b692598f 100644 --- a/drivers/hid/hid-core.c +++ b/drivers/hid/hid-core.c @@ -173,8 +173,8 @@ static int open_collection(struct hid_parser *parser, unsigned type) collection->type = type; collection->usage = usage; collection->level = parser->collection_stack_ptr - 1; - collection->parent_idx = parser->active_collection_idx; - parser->active_collection_idx = collection_index; + collection->parent_idx = (collection->level == 0) ? -1 : + parser->collection_stack[collection->level - 1]; if (type == HID_COLLECTION_APPLICATION) parser->device->maxapplication++; @@ -193,13 +193,6 @@ static int close_collection(struct hid_parser *parser) return -EINVAL; } parser->collection_stack_ptr--; - if (parser->active_collection_idx != -1) { - struct hid_device *device = parser->device; - struct hid_collection *c; - - c = &device->collection[parser->active_collection_idx]; - parser->active_collection_idx = c->parent_idx; - } return 0; } @@ -825,7 +818,6 @@ static int hid_scan_report(struct hid_device *hid) return -ENOMEM; parser->device = hid; - parser->active_collection_idx = -1; hid->group = HID_GROUP_GENERIC; /* @@ -1179,7 +1171,6 @@ int hid_open_report(struct hid_device *device) } parser->device = device; - parser->active_collection_idx = -1; end = start + size; diff --git a/include/linux/hid.h b/include/linux/hid.h index 992bbb7196df..f9707d1dcb58 100644 --- a/include/linux/hid.h +++ b/include/linux/hid.h @@ -658,7 +658,6 @@ struct hid_parser { unsigned int *collection_stack; unsigned int collection_stack_ptr; unsigned int collection_stack_size; - int active_collection_idx; /* device->collection */ struct hid_device *device; unsigned int scan_flags; }; -- cgit v1.2.3 From 890d14d2d4b57ff5a149309da3ed36c8a529987f Mon Sep 17 00:00:00 2001 From: Peter Rosin Date: Wed, 16 Jan 2019 17:42:35 +0100 Subject: fbdev: fbmem: convert CONFIG_FB_LOGO_CENTER into a cmd line option A command line option is much more flexible than a config option and the supporting code is small. Gets rid of #ifdefs in the code too... Suggested-by: Geert Uytterhoeven Cc: Jonathan Corbet Signed-off-by: Peter Rosin Signed-off-by: Bartlomiej Zolnierkiewicz --- Documentation/fb/fbcon.txt | 8 ++++++++ drivers/video/fbdev/core/fbcon.c | 7 +++++++ drivers/video/fbdev/core/fbmem.c | 19 ++++++++++--------- drivers/video/logo/Kconfig | 9 --------- include/linux/fb.h | 1 + 5 files changed, 26 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/Documentation/fb/fbcon.txt b/Documentation/fb/fbcon.txt index 62af30511a95..60a5ec04e8f0 100644 --- a/Documentation/fb/fbcon.txt +++ b/Documentation/fb/fbcon.txt @@ -163,6 +163,14 @@ C. Boot options be preserved until there actually is some text is output to the console. This option causes fbcon to bind immediately to the fbdev device. +7. fbcon=logo-pos: + + The only possible 'location' is 'center' (without quotes), and when + given, the bootup logo is moved from the default top-left corner + location to the center of the framebuffer. If more than one logo is + displayed due to multiple CPUs, the collected line of logos is moved + as a whole. + C. Attaching, Detaching and Unloading Before going on to how to attach, detach and unload the framebuffer console, an diff --git a/drivers/video/fbdev/core/fbcon.c b/drivers/video/fbdev/core/fbcon.c index 8976190b6c1f..bfa1360ec750 100644 --- a/drivers/video/fbdev/core/fbcon.c +++ b/drivers/video/fbdev/core/fbcon.c @@ -510,6 +510,13 @@ static int __init fb_console_setup(char *this_opt) continue; } #endif + + if (!strncmp(options, "logo-pos:", 9)) { + options += 9; + if (!strcmp(options, "center")) + fb_center_logo = true; + continue; + } } return 1; } diff --git a/drivers/video/fbdev/core/fbmem.c b/drivers/video/fbdev/core/fbmem.c index 558ed2ed3124..cb43a2258c51 100644 --- a/drivers/video/fbdev/core/fbmem.c +++ b/drivers/video/fbdev/core/fbmem.c @@ -53,6 +53,9 @@ EXPORT_SYMBOL(registered_fb); int num_registered_fb __read_mostly; EXPORT_SYMBOL(num_registered_fb); +bool fb_center_logo __read_mostly; +EXPORT_SYMBOL(fb_center_logo); + static struct fb_info *get_fb_info(unsigned int idx) { struct fb_info *fb_info; @@ -506,8 +509,7 @@ static int fb_show_logo_line(struct fb_info *info, int rotate, fb_set_logo(info, logo, logo_new, fb_logo.depth); } -#ifdef CONFIG_FB_LOGO_CENTER - { + if (fb_center_logo) { int xres = info->var.xres; int yres = info->var.yres; @@ -520,11 +522,11 @@ static int fb_show_logo_line(struct fb_info *info, int rotate, --n; image.dx = (xres - n * (logo->width + 8) - 8) / 2; image.dy = y ?: (yres - logo->height) / 2; + } else { + image.dx = 0; + image.dy = y; } -#else - image.dx = 0; - image.dy = y; -#endif + image.width = logo->width; image.height = logo->height; @@ -684,9 +686,8 @@ int fb_prepare_logo(struct fb_info *info, int rotate) } height = fb_logo.logo->height; -#ifdef CONFIG_FB_LOGO_CENTER - height += (yres - fb_logo.logo->height) / 2; -#endif + if (fb_center_logo) + height += (yres - fb_logo.logo->height) / 2; return fb_prepare_extra_logos(info, height, yres); } diff --git a/drivers/video/logo/Kconfig b/drivers/video/logo/Kconfig index 1e972c4e88b1..d1f6196c8b9a 100644 --- a/drivers/video/logo/Kconfig +++ b/drivers/video/logo/Kconfig @@ -10,15 +10,6 @@ menuconfig LOGO if LOGO -config FB_LOGO_CENTER - bool "Center the logo" - depends on FB=y - help - When this option is selected, the bootup logo is centered both - horizontally and vertically. If more than one logo is displayed - due to multiple CPUs, the collected line of logos is centered - as a whole. - config FB_LOGO_EXTRA bool depends on FB=y diff --git a/include/linux/fb.h b/include/linux/fb.h index 7cdd31a69719..f52ef0ad6781 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -653,6 +653,7 @@ extern int fb_new_modelist(struct fb_info *info); extern struct fb_info *registered_fb[FB_MAX]; extern int num_registered_fb; +extern bool fb_center_logo; extern struct class *fb_class; #define for_each_registered_fb(i) \ -- cgit v1.2.3 From 9e857a40dc4eba15a739b4194d7db873d82c28a0 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Tue, 15 Jan 2019 16:55:30 +0100 Subject: net: phy: Add missing features to PHY drivers The bcm87xx and micrel driver has PHYs which are missing the .features value. Add them. The bcm87xx is a 10G FEC only PHY. Add the needed features definition of this PHY. Fixes: 719655a14971 ("net: phy: Replace phy driver features u32 with link_mode bitmap") Reported-by: Scott Wood Reported-by: Camelia Groza Signed-off-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/phy/bcm87xx.c | 2 ++ drivers/net/phy/micrel.c | 1 + drivers/net/phy/phy_device.c | 12 ++++++++++++ include/linux/phy.h | 2 ++ 4 files changed, 17 insertions(+) (limited to 'include/linux') diff --git a/drivers/net/phy/bcm87xx.c b/drivers/net/phy/bcm87xx.c index 1b350183bffb..a271239748f2 100644 --- a/drivers/net/phy/bcm87xx.c +++ b/drivers/net/phy/bcm87xx.c @@ -197,6 +197,7 @@ static struct phy_driver bcm87xx_driver[] = { .phy_id = PHY_ID_BCM8706, .phy_id_mask = 0xffffffff, .name = "Broadcom BCM8706", + .features = PHY_10GBIT_FEC_FEATURES, .config_init = bcm87xx_config_init, .config_aneg = bcm87xx_config_aneg, .read_status = bcm87xx_read_status, @@ -208,6 +209,7 @@ static struct phy_driver bcm87xx_driver[] = { .phy_id = PHY_ID_BCM8727, .phy_id_mask = 0xffffffff, .name = "Broadcom BCM8727", + .features = PHY_10GBIT_FEC_FEATURES, .config_init = bcm87xx_config_init, .config_aneg = bcm87xx_config_aneg, .read_status = bcm87xx_read_status, diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 7828d17f0662..b1f959935f50 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -1099,6 +1099,7 @@ static struct phy_driver ksphy_driver[] = { .phy_id = PHY_ID_KSZ8873MLL, .phy_id_mask = MICREL_PHY_ID_MASK, .name = "Micrel KSZ8873MLL Switch", + .features = PHY_BASIC_FEATURES, .config_init = kszphy_config_init, .config_aneg = ksz8873mll_config_aneg, .read_status = ksz8873mll_read_status, diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 51990002d495..bf3ce48a1e5d 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -61,6 +61,9 @@ EXPORT_SYMBOL_GPL(phy_gbit_all_ports_features); __ETHTOOL_DECLARE_LINK_MODE_MASK(phy_10gbit_features) __ro_after_init; EXPORT_SYMBOL_GPL(phy_10gbit_features); +__ETHTOOL_DECLARE_LINK_MODE_MASK(phy_10gbit_fec_features) __ro_after_init; +EXPORT_SYMBOL_GPL(phy_10gbit_fec_features); + static const int phy_basic_ports_array[] = { ETHTOOL_LINK_MODE_Autoneg_BIT, ETHTOOL_LINK_MODE_TP_BIT, @@ -109,6 +112,11 @@ const int phy_10gbit_features_array[1] = { }; EXPORT_SYMBOL_GPL(phy_10gbit_features_array); +const int phy_10gbit_fec_features_array[1] = { + ETHTOOL_LINK_MODE_10000baseR_FEC_BIT, +}; +EXPORT_SYMBOL_GPL(phy_10gbit_fec_features_array); + __ETHTOOL_DECLARE_LINK_MODE_MASK(phy_10gbit_full_features) __ro_after_init; EXPORT_SYMBOL_GPL(phy_10gbit_full_features); @@ -191,6 +199,10 @@ static void features_init(void) linkmode_set_bit_array(phy_10gbit_full_features_array, ARRAY_SIZE(phy_10gbit_full_features_array), phy_10gbit_full_features); + /* 10G FEC only */ + linkmode_set_bit_array(phy_10gbit_fec_features_array, + ARRAY_SIZE(phy_10gbit_fec_features_array), + phy_10gbit_fec_features); } void phy_device_free(struct phy_device *phydev) diff --git a/include/linux/phy.h b/include/linux/phy.h index 3b051f761450..55114657a577 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -48,6 +48,7 @@ extern __ETHTOOL_DECLARE_LINK_MODE_MASK(phy_gbit_features) __ro_after_init; extern __ETHTOOL_DECLARE_LINK_MODE_MASK(phy_gbit_fibre_features) __ro_after_init; extern __ETHTOOL_DECLARE_LINK_MODE_MASK(phy_gbit_all_ports_features) __ro_after_init; extern __ETHTOOL_DECLARE_LINK_MODE_MASK(phy_10gbit_features) __ro_after_init; +extern __ETHTOOL_DECLARE_LINK_MODE_MASK(phy_10gbit_fec_features) __ro_after_init; extern __ETHTOOL_DECLARE_LINK_MODE_MASK(phy_10gbit_full_features) __ro_after_init; #define PHY_BASIC_FEATURES ((unsigned long *)&phy_basic_features) @@ -56,6 +57,7 @@ extern __ETHTOOL_DECLARE_LINK_MODE_MASK(phy_10gbit_full_features) __ro_after_ini #define PHY_GBIT_FIBRE_FEATURES ((unsigned long *)&phy_gbit_fibre_features) #define PHY_GBIT_ALL_PORTS_FEATURES ((unsigned long *)&phy_gbit_all_ports_features) #define PHY_10GBIT_FEATURES ((unsigned long *)&phy_10gbit_features) +#define PHY_10GBIT_FEC_FEATURES ((unsigned long *)&phy_10gbit_fec_features) #define PHY_10GBIT_FULL_FEATURES ((unsigned long *)&phy_10gbit_full_features) extern const int phy_10_100_features_array[4]; -- cgit v1.2.3 From edcddd4c879af48ec922d680b2d56834c085683b Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 17 Jan 2019 07:15:35 -0500 Subject: XArray: Fix an arithmetic error in xa_is_err There is a math problem here which leads to a lot of static checker warnings for me: net/sunrpc/clnt.c:451 rpc_new_client() error: (-4096) too low for ERR_PTR Error values are from -1 to -4095 or from 0xffffffff to 0xfffff001 in hexadecimal. (I am assuming a 32 bit system for simplicity). We are using the lowest two bits to hold some internal XArray data so the error is shifted two spaces to the left. 0xfffff001 << 2 is 0xffffc004. And finally we want to check that BIT(1) is set so we add 2 which gives us 0xffffc006. In other words, we should be checking that "entry >= 0xffffc006", but the check is actually testing if "entry >= 0xffffc002". Fixes: 76b4e5299565 ("XArray: Permit storing 2-byte-aligned pointers") Signed-off-by: Dan Carpenter [Use xa_mk_internal() instead of changing the bracketing] Signed-off-by: Matthew Wilcox --- include/linux/xarray.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/xarray.h b/include/linux/xarray.h index 7da665f5cb20..5d9d318bcf7a 100644 --- a/include/linux/xarray.h +++ b/include/linux/xarray.h @@ -177,7 +177,7 @@ static inline bool xa_is_internal(const void *entry) static inline bool xa_is_err(const void *entry) { return unlikely(xa_is_internal(entry) && - (unsigned long)entry >= -((MAX_ERRNO << 2) + 2)); + entry >= xa_mk_internal(-MAX_ERRNO)); } /** -- cgit v1.2.3 From 70921ae25f944423f0abf096f73455c586da0652 Mon Sep 17 00:00:00 2001 From: Jonathan Corbet Date: Tue, 8 Jan 2019 17:04:32 -0700 Subject: genirq: Fix the kerneldoc comment for struct irq_affinity_desc A recent commit added a new field but did not update the kerneldoc comment, leading to this build warning: ./include/linux/interrupt.h:268: warning: Function parameter or member 'is_managed' not described in 'irq_affinity_desc' Add the missing information, making the docs build 0.001% quieter. Fixes: c410abbbacb9 ("genirq/affinity: Add is_managed to struct irq_affinity_desc") Signed-off-by: Jonathan Corbet Signed-off-by: Thomas Gleixner Cc: Dou Liyang Link: https://lkml.kernel.org/r/20190108170432.59bae8a6@lwn.net --- include/linux/interrupt.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index c672f34235e7..4a728dba02e2 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -260,6 +260,7 @@ struct irq_affinity { /** * struct irq_affinity_desc - Interrupt affinity descriptor * @mask: cpumask to hold the affinity assignment + * @is_managed: 1 if the interrupt is managed internally */ struct irq_affinity_desc { struct cpumask mask; -- cgit v1.2.3 From 202e651cd43c69a43f75b445e90f55b59f9af0ad Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 15 Jan 2019 22:03:34 +0100 Subject: netfilter: conntrack: gre: convert rwlock to rcu We can use gre. Lock is only needed when a new expectation is added. In case a single spinlock proves to be problematic we can either add one per netns or use an array of locks combined with net_hash_mix() or similar to pick the 'correct' one. But given this is only needed for an expectation rather than per packet a single one should be ok. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/nf_conntrack_proto_gre.h | 1 + net/netfilter/nf_conntrack_proto_gre.c | 37 ++++++++++-------------- 2 files changed, 16 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter/nf_conntrack_proto_gre.h b/include/linux/netfilter/nf_conntrack_proto_gre.h index 6989e2e4eabf..222c9d3d453f 100644 --- a/include/linux/netfilter/nf_conntrack_proto_gre.h +++ b/include/linux/netfilter/nf_conntrack_proto_gre.h @@ -19,6 +19,7 @@ struct nf_conn; struct nf_ct_gre_keymap { struct list_head list; struct nf_conntrack_tuple tuple; + struct rcu_head rcu; }; enum grep_conntrack { diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c index 8899b51aad44..34dd89485be2 100644 --- a/net/netfilter/nf_conntrack_proto_gre.c +++ b/net/netfilter/nf_conntrack_proto_gre.c @@ -49,6 +49,8 @@ static const unsigned int gre_timeouts[GRE_CT_MAX] = { }; static unsigned int proto_gre_net_id __read_mostly; +/* used when expectation is added */ +static DEFINE_SPINLOCK(keymap_lock); static inline struct netns_proto_gre *gre_pernet(struct net *net) { @@ -60,12 +62,12 @@ static void nf_ct_gre_keymap_flush(struct net *net) struct netns_proto_gre *net_gre = gre_pernet(net); struct nf_ct_gre_keymap *km, *tmp; - write_lock_bh(&net_gre->keymap_lock); + spin_lock_bh(&keymap_lock); list_for_each_entry_safe(km, tmp, &net_gre->keymap_list, list) { - list_del(&km->list); - kfree(km); + list_del_rcu(&km->list); + kfree_rcu(km, rcu); } - write_unlock_bh(&net_gre->keymap_lock); + spin_unlock_bh(&keymap_lock); } static inline int gre_key_cmpfn(const struct nf_ct_gre_keymap *km, @@ -85,14 +87,12 @@ static __be16 gre_keymap_lookup(struct net *net, struct nf_conntrack_tuple *t) struct nf_ct_gre_keymap *km; __be16 key = 0; - read_lock_bh(&net_gre->keymap_lock); - list_for_each_entry(km, &net_gre->keymap_list, list) { + list_for_each_entry_rcu(km, &net_gre->keymap_list, list) { if (gre_key_cmpfn(km, t)) { key = km->tuple.src.u.gre.key; break; } } - read_unlock_bh(&net_gre->keymap_lock); pr_debug("lookup src key 0x%x for ", key); nf_ct_dump_tuple(t); @@ -112,14 +112,10 @@ int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir, kmp = &ct_pptp_info->keymap[dir]; if (*kmp) { /* check whether it's a retransmission */ - read_lock_bh(&net_gre->keymap_lock); - list_for_each_entry(km, &net_gre->keymap_list, list) { - if (gre_key_cmpfn(km, t) && km == *kmp) { - read_unlock_bh(&net_gre->keymap_lock); + list_for_each_entry_rcu(km, &net_gre->keymap_list, list) { + if (gre_key_cmpfn(km, t) && km == *kmp) return 0; - } } - read_unlock_bh(&net_gre->keymap_lock); pr_debug("trying to override keymap_%s for ct %p\n", dir == IP_CT_DIR_REPLY ? "reply" : "orig", ct); return -EEXIST; @@ -134,9 +130,9 @@ int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir, pr_debug("adding new entry %p: ", km); nf_ct_dump_tuple(&km->tuple); - write_lock_bh(&net_gre->keymap_lock); + spin_lock_bh(&keymap_lock); list_add_tail(&km->list, &net_gre->keymap_list); - write_unlock_bh(&net_gre->keymap_lock); + spin_unlock_bh(&keymap_lock); return 0; } @@ -145,24 +141,22 @@ EXPORT_SYMBOL_GPL(nf_ct_gre_keymap_add); /* destroy the keymap entries associated with specified master ct */ void nf_ct_gre_keymap_destroy(struct nf_conn *ct) { - struct net *net = nf_ct_net(ct); - struct netns_proto_gre *net_gre = gre_pernet(net); struct nf_ct_pptp_master *ct_pptp_info = nfct_help_data(ct); enum ip_conntrack_dir dir; pr_debug("entering for ct %p\n", ct); - write_lock_bh(&net_gre->keymap_lock); + spin_lock_bh(&keymap_lock); for (dir = IP_CT_DIR_ORIGINAL; dir < IP_CT_DIR_MAX; dir++) { if (ct_pptp_info->keymap[dir]) { pr_debug("removing %p from list\n", ct_pptp_info->keymap[dir]); - list_del(&ct_pptp_info->keymap[dir]->list); - kfree(ct_pptp_info->keymap[dir]); + list_del_rcu(&ct_pptp_info->keymap[dir]->list); + kfree_rcu(ct_pptp_info->keymap[dir], rcu); ct_pptp_info->keymap[dir] = NULL; } } - write_unlock_bh(&net_gre->keymap_lock); + spin_unlock_bh(&keymap_lock); } EXPORT_SYMBOL_GPL(nf_ct_gre_keymap_destroy); @@ -365,7 +359,6 @@ static int gre_init_net(struct net *net) struct nf_proto_net *nf = &net_gre->nf; int i; - rwlock_init(&net_gre->keymap_lock); INIT_LIST_HEAD(&net_gre->keymap_list); for (i = 0; i < GRE_CT_MAX; i++) net_gre->gre_timeouts[i] = gre_timeouts[i]; -- cgit v1.2.3 From 22fc4c4c9fd60427bcda00878cee94e7622cfa7a Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 15 Jan 2019 22:03:35 +0100 Subject: netfilter: conntrack: gre: switch module to be built-in This makes the last of the modular l4 trackers 'bool'. After this, all infrastructure to handle dynamic l4 protocol registration becomes obsolete and can be removed in followup patches. Old: 302824 net/netfilter/nf_conntrack.ko 21504 net/netfilter/nf_conntrack_proto_gre.ko New: 313728 net/netfilter/nf_conntrack.ko Old: text data bss dec hex filename 6281 1732 4 8017 1f51 nf_conntrack_proto_gre.ko 108356 20613 236 129205 1f8b5 nf_conntrack.ko New: 112095 21381 240 133716 20a54 nf_conntrack.ko The size increase is only temporary. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/nf_conntrack_proto_gre.h | 14 +--- include/net/netfilter/ipv4/nf_conntrack_ipv4.h | 3 + include/net/netfilter/nf_conntrack_l4proto.h | 7 ++ include/net/netns/conntrack.h | 17 +++++ net/netfilter/Kconfig | 2 +- net/netfilter/Makefile | 3 +- net/netfilter/nf_conntrack_proto.c | 7 +- net/netfilter/nf_conntrack_proto_gre.c | 93 +++++------------------- net/netfilter/nfnetlink_cttimeout.c | 7 +- 9 files changed, 55 insertions(+), 98 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter/nf_conntrack_proto_gre.h b/include/linux/netfilter/nf_conntrack_proto_gre.h index 222c9d3d453f..59714e9ee4ef 100644 --- a/include/linux/netfilter/nf_conntrack_proto_gre.h +++ b/include/linux/netfilter/nf_conntrack_proto_gre.h @@ -22,23 +22,11 @@ struct nf_ct_gre_keymap { struct rcu_head rcu; }; -enum grep_conntrack { - GRE_CT_UNREPLIED, - GRE_CT_REPLIED, - GRE_CT_MAX -}; - -struct netns_proto_gre { - struct nf_proto_net nf; - rwlock_t keymap_lock; - struct list_head keymap_list; - unsigned int gre_timeouts[GRE_CT_MAX]; -}; - /* add new tuple->key_reply pair to keymap */ int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir, struct nf_conntrack_tuple *t); +void nf_ct_gre_keymap_flush(struct net *net); /* delete keymap entries */ void nf_ct_gre_keymap_destroy(struct nf_conn *ct); diff --git a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h index 135ee702c7b0..2c8c2b023848 100644 --- a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h +++ b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h @@ -22,5 +22,8 @@ extern const struct nf_conntrack_l4proto nf_conntrack_l4proto_sctp; #ifdef CONFIG_NF_CT_PROTO_UDPLITE extern const struct nf_conntrack_l4proto nf_conntrack_l4proto_udplite; #endif +#ifdef CONFIG_NF_CT_PROTO_GRE +extern const struct nf_conntrack_l4proto nf_conntrack_l4proto_gre; +#endif #endif /*_NF_CONNTRACK_IPV4_H*/ diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index 46d554806eb3..fded3f164dcc 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -239,4 +239,11 @@ static inline struct nf_sctp_net *nf_sctp_pernet(struct net *net) } #endif +#ifdef CONFIG_NF_CT_PROTO_GRE +static inline struct nf_gre_net *nf_gre_pernet(struct net *net) +{ + return &net->ct.nf_ct_proto.gre; +} +#endif + #endif /*_NF_CONNTRACK_PROTOCOL_H*/ diff --git a/include/net/netns/conntrack.h b/include/net/netns/conntrack.h index 51cba0b8adf5..c72f413a2d4d 100644 --- a/include/net/netns/conntrack.h +++ b/include/net/netns/conntrack.h @@ -70,6 +70,20 @@ struct nf_sctp_net { }; #endif +#ifdef CONFIG_NF_CT_PROTO_GRE +enum gre_conntrack { + GRE_CT_UNREPLIED, + GRE_CT_REPLIED, + GRE_CT_MAX +}; + +struct nf_gre_net { + struct nf_proto_net nf; + struct list_head keymap_list; + unsigned int timeouts[GRE_CT_MAX]; +}; +#endif + struct nf_ip_net { struct nf_generic_net generic; struct nf_tcp_net tcp; @@ -82,6 +96,9 @@ struct nf_ip_net { #ifdef CONFIG_NF_CT_PROTO_SCTP struct nf_sctp_net sctp; #endif +#ifdef CONFIG_NF_CT_PROTO_GRE + struct nf_gre_net gre; +#endif }; struct ct_pcpu { diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index beb3a69ce1d4..fefd63a243f2 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -174,7 +174,7 @@ config NF_CT_PROTO_DCCP If unsure, say Y. config NF_CT_PROTO_GRE - tristate + bool config NF_CT_PROTO_SCTP bool 'SCTP protocol connection tracking support' diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 1ae65a314d7a..e66067befa42 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -13,6 +13,7 @@ nf_conntrack-$(CONFIG_NF_CONNTRACK_EVENTS) += nf_conntrack_ecache.o nf_conntrack-$(CONFIG_NF_CONNTRACK_LABELS) += nf_conntrack_labels.o nf_conntrack-$(CONFIG_NF_CT_PROTO_DCCP) += nf_conntrack_proto_dccp.o nf_conntrack-$(CONFIG_NF_CT_PROTO_SCTP) += nf_conntrack_proto_sctp.o +nf_conntrack-$(CONFIG_NF_CT_PROTO_GRE) += nf_conntrack_proto_gre.o obj-$(CONFIG_NETFILTER) = netfilter.o @@ -25,8 +26,6 @@ obj-$(CONFIG_NETFILTER_NETLINK_OSF) += nfnetlink_osf.o # connection tracking obj-$(CONFIG_NF_CONNTRACK) += nf_conntrack.o -obj-$(CONFIG_NF_CT_PROTO_GRE) += nf_conntrack_proto_gre.o - # netlink interface for nf_conntrack obj-$(CONFIG_NF_CT_NETLINK) += nf_conntrack_netlink.o obj-$(CONFIG_NF_CT_NETLINK_TIMEOUT) += nfnetlink_cttimeout.o diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index 2bbc32d939e4..e113bb2dc88d 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -817,6 +817,9 @@ static const struct nf_conntrack_l4proto * const builtin_l4proto[] = { #ifdef CONFIG_NF_CT_PROTO_UDPLITE &nf_conntrack_l4proto_udplite, #endif +#ifdef CONFIG_NF_CT_PROTO_GRE + &nf_conntrack_l4proto_gre, +#endif #if IS_ENABLED(CONFIG_IPV6) &nf_conntrack_l4proto_icmpv6, #endif /* CONFIG_IPV6 */ @@ -897,9 +900,11 @@ void nf_conntrack_proto_pernet_fini(struct net *net) ARRAY_SIZE(builtin_l4proto)); pn->users--; nf_ct_l4proto_unregister_sysctl(pn); +#ifdef CONFIG_NF_CT_PROTO_GRE + nf_ct_gre_keymap_flush(net); +#endif } - module_param_call(hashsize, nf_conntrack_set_hashsize, param_get_uint, &nf_conntrack_htable_size, 0600); diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c index 34dd89485be2..68f9bfb79c4e 100644 --- a/net/netfilter/nf_conntrack_proto_gre.c +++ b/net/netfilter/nf_conntrack_proto_gre.c @@ -48,18 +48,17 @@ static const unsigned int gre_timeouts[GRE_CT_MAX] = { [GRE_CT_REPLIED] = 180*HZ, }; -static unsigned int proto_gre_net_id __read_mostly; /* used when expectation is added */ static DEFINE_SPINLOCK(keymap_lock); -static inline struct netns_proto_gre *gre_pernet(struct net *net) +static inline struct nf_gre_net *gre_pernet(struct net *net) { - return net_generic(net, proto_gre_net_id); + return &net->ct.nf_ct_proto.gre; } -static void nf_ct_gre_keymap_flush(struct net *net) +void nf_ct_gre_keymap_flush(struct net *net) { - struct netns_proto_gre *net_gre = gre_pernet(net); + struct nf_gre_net *net_gre = gre_pernet(net); struct nf_ct_gre_keymap *km, *tmp; spin_lock_bh(&keymap_lock); @@ -83,7 +82,7 @@ static inline int gre_key_cmpfn(const struct nf_ct_gre_keymap *km, /* look up the source key for a given tuple */ static __be16 gre_keymap_lookup(struct net *net, struct nf_conntrack_tuple *t) { - struct netns_proto_gre *net_gre = gre_pernet(net); + struct nf_gre_net *net_gre = gre_pernet(net); struct nf_ct_gre_keymap *km; __be16 key = 0; @@ -105,7 +104,7 @@ int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir, struct nf_conntrack_tuple *t) { struct net *net = nf_ct_net(ct); - struct netns_proto_gre *net_gre = gre_pernet(net); + struct nf_gre_net *net_gre = gre_pernet(net); struct nf_ct_pptp_master *ct_pptp_info = nfct_help_data(ct); struct nf_ct_gre_keymap **kmp, *km; @@ -210,7 +209,7 @@ static void gre_print_conntrack(struct seq_file *s, struct nf_conn *ct) static unsigned int *gre_get_timeouts(struct net *net) { - return gre_pernet(net)->gre_timeouts; + return gre_pernet(net)->timeouts; } /* Returns verdict for packet, and may modify conntrack */ @@ -272,13 +271,13 @@ static int gre_timeout_nlattr_to_obj(struct nlattr *tb[], struct net *net, void *data) { unsigned int *timeouts = data; - struct netns_proto_gre *net_gre = gre_pernet(net); + struct nf_gre_net *net_gre = gre_pernet(net); if (!timeouts) timeouts = gre_get_timeouts(net); /* set default timeouts for GRE. */ - timeouts[GRE_CT_UNREPLIED] = net_gre->gre_timeouts[GRE_CT_UNREPLIED]; - timeouts[GRE_CT_REPLIED] = net_gre->gre_timeouts[GRE_CT_REPLIED]; + timeouts[GRE_CT_UNREPLIED] = net_gre->timeouts[GRE_CT_UNREPLIED]; + timeouts[GRE_CT_REPLIED] = net_gre->timeouts[GRE_CT_REPLIED]; if (tb[CTA_TIMEOUT_GRE_UNREPLIED]) { timeouts[GRE_CT_UNREPLIED] = @@ -332,10 +331,11 @@ static struct ctl_table gre_sysctl_table[] = { }; #endif -static int gre_kmemdup_sysctl_table(struct net *net, struct nf_proto_net *nf, - struct netns_proto_gre *net_gre) +static int gre_kmemdup_sysctl_table(struct net *net) { #ifdef CONFIG_SYSCTL + struct nf_gre_net *net_gre = gre_pernet(net); + struct nf_proto_net *nf = &net_gre->nf; int i; if (nf->ctl_table) @@ -348,26 +348,25 @@ static int gre_kmemdup_sysctl_table(struct net *net, struct nf_proto_net *nf, return -ENOMEM; for (i = 0; i < GRE_CT_MAX; i++) - nf->ctl_table[i].data = &net_gre->gre_timeouts[i]; + nf->ctl_table[i].data = &net_gre->timeouts[i]; #endif return 0; } static int gre_init_net(struct net *net) { - struct netns_proto_gre *net_gre = gre_pernet(net); - struct nf_proto_net *nf = &net_gre->nf; + struct nf_gre_net *net_gre = gre_pernet(net); int i; INIT_LIST_HEAD(&net_gre->keymap_list); for (i = 0; i < GRE_CT_MAX; i++) - net_gre->gre_timeouts[i] = gre_timeouts[i]; + net_gre->timeouts[i] = gre_timeouts[i]; - return gre_kmemdup_sysctl_table(net, nf, net_gre); + return gre_kmemdup_sysctl_table(net); } /* protocol helper struct */ -static const struct nf_conntrack_l4proto nf_conntrack_l4proto_gre4 = { +const struct nf_conntrack_l4proto nf_conntrack_l4proto_gre = { .l4proto = IPPROTO_GRE, .pkt_to_tuple = gre_pkt_to_tuple, #ifdef CONFIG_NF_CONNTRACK_PROCFS @@ -391,61 +390,5 @@ static const struct nf_conntrack_l4proto nf_conntrack_l4proto_gre4 = { .nla_policy = gre_timeout_nla_policy, }, #endif /* CONFIG_NF_CONNTRACK_TIMEOUT */ - .net_id = &proto_gre_net_id, .init_net = gre_init_net, }; - -static int proto_gre_net_init(struct net *net) -{ - int ret = 0; - - ret = nf_ct_l4proto_pernet_register_one(net, - &nf_conntrack_l4proto_gre4); - if (ret < 0) - pr_err("nf_conntrack_gre4: pernet registration failed.\n"); - return ret; -} - -static void proto_gre_net_exit(struct net *net) -{ - nf_ct_l4proto_pernet_unregister_one(net, &nf_conntrack_l4proto_gre4); - nf_ct_gre_keymap_flush(net); -} - -static struct pernet_operations proto_gre_net_ops = { - .init = proto_gre_net_init, - .exit = proto_gre_net_exit, - .id = &proto_gre_net_id, - .size = sizeof(struct netns_proto_gre), -}; - -static int __init nf_ct_proto_gre_init(void) -{ - int ret; - - BUILD_BUG_ON(offsetof(struct netns_proto_gre, nf) != 0); - - ret = register_pernet_subsys(&proto_gre_net_ops); - if (ret < 0) - goto out_pernet; - ret = nf_ct_l4proto_register_one(&nf_conntrack_l4proto_gre4); - if (ret < 0) - goto out_gre4; - - return 0; -out_gre4: - unregister_pernet_subsys(&proto_gre_net_ops); -out_pernet: - return ret; -} - -static void __exit nf_ct_proto_gre_fini(void) -{ - nf_ct_l4proto_unregister_one(&nf_conntrack_l4proto_gre4); - unregister_pernet_subsys(&proto_gre_net_ops); -} - -module_init(nf_ct_proto_gre_init); -module_exit(nf_ct_proto_gre_fini); - -MODULE_LICENSE("GPL"); diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c index 109b0d27345a..0e3e1a018206 100644 --- a/net/netfilter/nfnetlink_cttimeout.c +++ b/net/netfilter/nfnetlink_cttimeout.c @@ -474,12 +474,7 @@ static int cttimeout_default_get(struct net *net, struct sock *ctnl, break; case IPPROTO_GRE: #ifdef CONFIG_NF_CT_PROTO_GRE - if (l4proto->net_id) { - struct netns_proto_gre *net_gre; - - net_gre = net_generic(net, *l4proto->net_id); - timeouts = net_gre->gre_timeouts; - } + timeouts = nf_gre_pernet(net)->timeouts; #endif break; case 255: -- cgit v1.2.3 From df5e1629087a45ca915fa0f69ea662175261855e Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 15 Jan 2019 22:03:37 +0100 Subject: netfilter: conntrack: remove pkt_to_tuple callback GRE is now builtin, so we can handle it via direct call and remove the callback. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/nf_conntrack_proto_gre.h | 2 ++ include/net/netfilter/nf_conntrack_l4proto.h | 5 ----- net/netfilter/nf_conntrack_core.c | 6 ++++-- net/netfilter/nf_conntrack_proto_generic.c | 11 ----------- net/netfilter/nf_conntrack_proto_gre.c | 5 ++--- 5 files changed, 8 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter/nf_conntrack_proto_gre.h b/include/linux/netfilter/nf_conntrack_proto_gre.h index 59714e9ee4ef..25f9a770fb84 100644 --- a/include/linux/netfilter/nf_conntrack_proto_gre.h +++ b/include/linux/netfilter/nf_conntrack_proto_gre.h @@ -30,5 +30,7 @@ void nf_ct_gre_keymap_flush(struct net *net); /* delete keymap entries */ void nf_ct_gre_keymap_destroy(struct nf_conn *ct); +bool gre_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, + struct net *net, struct nf_conntrack_tuple *tuple); #endif /* __KERNEL__ */ #endif /* _CONNTRACK_PROTO_GRE_H */ diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index 3585f8666fc0..0d4b0398aeb9 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -27,11 +27,6 @@ struct nf_conntrack_l4proto { /* protoinfo nlattr size, closes a hole */ u16 nlattr_size; - /* Try to fill in the third arg: dataoff is offset past network protocol - hdr. Return true if possible. */ - bool (*pkt_to_tuple)(const struct sk_buff *skb, unsigned int dataoff, - struct net *net, struct nf_conntrack_tuple *tuple); - /* Invert the per-proto part of the tuple: ie. turn xmit into reply. * Only used by icmp, most protocols use a generic version. */ diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index b3840d36c3a6..b71e271f2b44 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -279,9 +279,11 @@ nf_ct_get_tuple(const struct sk_buff *skb, return icmpv6_pkt_to_tuple(skb, dataoff, net, tuple); case IPPROTO_ICMP: return icmp_pkt_to_tuple(skb, dataoff, net, tuple); +#ifdef CONFIG_NF_CT_PROTO_GRE + case IPPROTO_GRE: + return gre_pkt_to_tuple(skb, dataoff, net, tuple); +#endif } - if (unlikely(l4proto->pkt_to_tuple)) - return l4proto->pkt_to_tuple(skb, dataoff, net, tuple); /* Actually only need first 4 bytes to get ports. */ inet_hdr = skb_header_pointer(skb, dataoff, sizeof(_inet_hdr), &_inet_hdr); diff --git a/net/netfilter/nf_conntrack_proto_generic.c b/net/netfilter/nf_conntrack_proto_generic.c index 5da19d5fbc76..5a5bf7cb6508 100644 --- a/net/netfilter/nf_conntrack_proto_generic.c +++ b/net/netfilter/nf_conntrack_proto_generic.c @@ -27,16 +27,6 @@ static bool nf_generic_should_process(u8 proto) } } -static bool generic_pkt_to_tuple(const struct sk_buff *skb, - unsigned int dataoff, - struct net *net, struct nf_conntrack_tuple *tuple) -{ - tuple->src.u.all = 0; - tuple->dst.u.all = 0; - - return true; -} - /* Returns verdict for packet, or -1 for invalid. */ static int generic_packet(struct nf_conn *ct, struct sk_buff *skb, @@ -149,7 +139,6 @@ static struct nf_proto_net *generic_get_net_proto(struct net *net) const struct nf_conntrack_l4proto nf_conntrack_l4proto_generic = { .l4proto = 255, - .pkt_to_tuple = generic_pkt_to_tuple, .packet = generic_packet, #ifdef CONFIG_NF_CONNTRACK_TIMEOUT .ctnl_timeout = { diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c index 68f9bfb79c4e..04bc982b274d 100644 --- a/net/netfilter/nf_conntrack_proto_gre.c +++ b/net/netfilter/nf_conntrack_proto_gre.c @@ -162,8 +162,8 @@ EXPORT_SYMBOL_GPL(nf_ct_gre_keymap_destroy); /* PUBLIC CONNTRACK PROTO HELPER FUNCTIONS */ /* gre hdr info to tuple */ -static bool gre_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, - struct net *net, struct nf_conntrack_tuple *tuple) +bool gre_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, + struct net *net, struct nf_conntrack_tuple *tuple) { const struct pptp_gre_header *pgrehdr; struct pptp_gre_header _pgrehdr; @@ -368,7 +368,6 @@ static int gre_init_net(struct net *net) /* protocol helper struct */ const struct nf_conntrack_l4proto nf_conntrack_l4proto_gre = { .l4proto = IPPROTO_GRE, - .pkt_to_tuple = gre_pkt_to_tuple, #ifdef CONFIG_NF_CONNTRACK_PROCFS .print_conntrack = gre_print_conntrack, #endif -- cgit v1.2.3 From 6c57f0458022298e4da1729c67bd33ce41c14e7a Mon Sep 17 00:00:00 2001 From: Ross Lagerwall Date: Thu, 17 Jan 2019 15:34:38 +0000 Subject: net: Fix usage of pskb_trim_rcsum In certain cases, pskb_trim_rcsum() may change skb pointers. Reinitialize header pointers afterwards to avoid potential use-after-frees. Add a note in the documentation of pskb_trim_rcsum(). Found by KASAN. Signed-off-by: Ross Lagerwall Signed-off-by: David S. Miller --- drivers/net/ppp/pppoe.c | 1 + include/linux/skbuff.h | 1 + net/bridge/br_netfilter_ipv6.c | 1 + net/bridge/netfilter/nft_reject_bridge.c | 1 + net/ipv4/ip_input.c | 1 + 5 files changed, 5 insertions(+) (limited to 'include/linux') diff --git a/drivers/net/ppp/pppoe.c b/drivers/net/ppp/pppoe.c index 62dc564b251d..f22639f0116a 100644 --- a/drivers/net/ppp/pppoe.c +++ b/drivers/net/ppp/pppoe.c @@ -445,6 +445,7 @@ static int pppoe_rcv(struct sk_buff *skb, struct net_device *dev, if (pskb_trim_rcsum(skb, len)) goto drop; + ph = pppoe_hdr(skb); pn = pppoe_pernet(dev_net(dev)); /* Note that get_item does a sock_hold(), so sk_pppox(po) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 93f56fddd92a..95d25b010a25 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3218,6 +3218,7 @@ int pskb_trim_rcsum_slow(struct sk_buff *skb, unsigned int len); * * This is exactly the same as pskb_trim except that it ensures the * checksum of received packets are still valid after the operation. + * It can change skb pointers. */ static inline int pskb_trim_rcsum(struct sk_buff *skb, unsigned int len) diff --git a/net/bridge/br_netfilter_ipv6.c b/net/bridge/br_netfilter_ipv6.c index 94039f588f1d..564710f88f93 100644 --- a/net/bridge/br_netfilter_ipv6.c +++ b/net/bridge/br_netfilter_ipv6.c @@ -131,6 +131,7 @@ int br_validate_ipv6(struct net *net, struct sk_buff *skb) IPSTATS_MIB_INDISCARDS); goto drop; } + hdr = ipv6_hdr(skb); } if (hdr->nexthdr == NEXTHDR_HOP && br_nf_check_hbh_len(skb)) goto drop; diff --git a/net/bridge/netfilter/nft_reject_bridge.c b/net/bridge/netfilter/nft_reject_bridge.c index 08cbed7d940e..419e8edf23ba 100644 --- a/net/bridge/netfilter/nft_reject_bridge.c +++ b/net/bridge/netfilter/nft_reject_bridge.c @@ -229,6 +229,7 @@ static bool reject6_br_csum_ok(struct sk_buff *skb, int hook) pskb_trim_rcsum(skb, ntohs(ip6h->payload_len) + sizeof(*ip6h))) return false; + ip6h = ipv6_hdr(skb); thoff = ipv6_skip_exthdr(skb, ((u8*)(ip6h+1) - skb->data), &proto, &fo); if (thoff < 0 || thoff >= skb->len || (fo & htons(~0x7)) != 0) return false; diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index 26921f6b3b92..51d8efba6de2 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -488,6 +488,7 @@ static struct sk_buff *ip_rcv_core(struct sk_buff *skb, struct net *net) goto drop; } + iph = ip_hdr(skb); skb->transport_header = skb->network_header + iph->ihl*4; /* Remove any debris in the socket control block */ -- cgit v1.2.3 From 3e64cf7a435ed0500e3adaa8aada2272d3ae8abc Mon Sep 17 00:00:00 2001 From: Camelia Groza Date: Thu, 17 Jan 2019 14:22:36 +0200 Subject: net: phy: phy driver features are mandatory Since phy driver features became a link_mode bitmap, phy drivers that don't have a list of features configured will cause the kernel to crash when probed. Prevent the phy driver from registering if the features field is missing. Fixes: 719655a14971 ("net: phy: Replace phy driver features u32 with link_mode bitmap") Reported-by: Scott Wood Signed-off-by: Camelia Groza Signed-off-by: David S. Miller --- drivers/net/phy/phy_device.c | 5 +++++ include/linux/phy.h | 4 ++-- 2 files changed, 7 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index bf3ce48a1e5d..46c86725a693 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -2255,6 +2255,11 @@ int phy_driver_register(struct phy_driver *new_driver, struct module *owner) { int retval; + if (WARN_ON(!new_driver->features)) { + pr_err("%s: Driver features are missing\n", new_driver->name); + return -EINVAL; + } + new_driver->mdiodrv.flags |= MDIO_DEVICE_IS_PHY; new_driver->mdiodrv.driver.name = new_driver->name; new_driver->mdiodrv.driver.bus = &mdio_bus_type; diff --git a/include/linux/phy.h b/include/linux/phy.h index 55114657a577..ef20aeea10cc 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -469,8 +469,8 @@ struct phy_device { * only works for PHYs with IDs which match this field * name: The friendly name of this PHY type * phy_id_mask: Defines the important bits of the phy_id - * features: A list of features (speed, duplex, etc) supported - * by this PHY + * features: A mandatory list of features (speed, duplex, etc) + * supported by this PHY * flags: A bitfield defining certain other features this PHY * supports (like interrupts) * -- cgit v1.2.3 From 59c28058fa7bb1cc7ab8b2c5607093cbbefafeb4 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 18 Jan 2019 10:46:13 -0800 Subject: net: netlink: add helper to retrieve NETLINK_F_STRICT_CHK Dumps can read state of the NETLINK_F_STRICT_CHK flag from a field in the callback structure. For non-dump GET requests we need a way to access the state of that flag from a socket. Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- include/linux/netlink.h | 1 + net/netlink/af_netlink.c | 8 ++++++++ 2 files changed, 9 insertions(+) (limited to 'include/linux') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index 4e8add270200..593d1b9c33a8 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -126,6 +126,7 @@ void __netlink_clear_multicast_users(struct sock *sk, unsigned int group); void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err, const struct netlink_ext_ack *extack); int netlink_has_listeners(struct sock *sk, unsigned int group); +bool netlink_strict_get_check(struct sk_buff *skb); int netlink_unicast(struct sock *ssk, struct sk_buff *skb, __u32 portid, int nonblock); int netlink_broadcast(struct sock *ssk, struct sk_buff *skb, __u32 portid, diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 3c023d6120f6..8fa35df94c07 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1371,6 +1371,14 @@ int netlink_has_listeners(struct sock *sk, unsigned int group) } EXPORT_SYMBOL_GPL(netlink_has_listeners); +bool netlink_strict_get_check(struct sk_buff *skb) +{ + const struct netlink_sock *nlk = nlk_sk(NETLINK_CB(skb).sk); + + return nlk->flags & NETLINK_F_STRICT_CHK; +} +EXPORT_SYMBOL_GPL(netlink_strict_get_check); + static int netlink_broadcast_deliver(struct sock *sk, struct sk_buff *skb) { struct netlink_sock *nlk = nlk_sk(sk); -- cgit v1.2.3 From e6018c0f5c996e61639adce6a0697391a2861916 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 17 Dec 2018 10:14:53 +0100 Subject: sched/wake_q: Document wake_q_add() The only guarantee provided by wake_q_add() is that a wakeup will happen after it, it does _NOT_ guarantee the wakeup will be delayed until the matching wake_up_q(). If wake_q_add() fails the cmpxchg() a concurrent wakeup is pending and that can happen at any time after the cmpxchg(). This means we should not rely on the wakeup happening at wake_q_up(), but should be ready for wake_q_add() to issue the wakeup. The delay; if provided (most likely); should only result in more efficient behaviour. Reported-by: Yongji Xie Signed-off-by: Peter Zijlstra (Intel) Cc: Davidlohr Bueso Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Waiman Long Cc: Will Deacon Signed-off-by: Ingo Molnar --- include/linux/sched/wake_q.h | 6 +++++- kernel/sched/core.c | 12 ++++++++++++ 2 files changed, 17 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sched/wake_q.h b/include/linux/sched/wake_q.h index 10b19a192b2d..545f37138057 100644 --- a/include/linux/sched/wake_q.h +++ b/include/linux/sched/wake_q.h @@ -24,9 +24,13 @@ * called near the end of a function. Otherwise, the list can be * re-initialized for later re-use by wake_q_init(). * - * Note that this can cause spurious wakeups. schedule() callers + * NOTE that this can cause spurious wakeups. schedule() callers * must ensure the call is done inside a loop, confirming that the * wakeup condition has in fact occurred. + * + * NOTE that there is no guarantee the wakeup will happen any later than the + * wake_q_add() location. Therefore task must be ready to be woken at the + * location of the wake_q_add(). */ #include diff --git a/kernel/sched/core.c b/kernel/sched/core.c index a674c7db2f29..cc814933f7d6 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -396,6 +396,18 @@ static bool set_nr_if_polling(struct task_struct *p) #endif #endif +/** + * wake_q_add() - queue a wakeup for 'later' waking. + * @head: the wake_q_head to add @task to + * @task: the task to queue for 'later' wakeup + * + * Queue a task for later wakeup, most likely by the wake_up_q() call in the + * same context, _HOWEVER_ this is not guaranteed, the wakeup can come + * instantly. + * + * This function must be used as-if it were wake_up_process(); IOW the task + * must be ready to be woken at this location. + */ void wake_q_add(struct wake_q_head *head, struct task_struct *task) { struct wake_q_node *node = &task->wake_q; -- cgit v1.2.3 From 1cd7386549f9b6f2f230da54aa9e7fe2d6c216d2 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Sat, 19 Jan 2019 08:45:56 -0800 Subject: libnvdimm/security: Require nvdimm_security_setup_events() to succeed The following warning: ACPI0012:00: security event setup failed: -19 ...is meant to capture exceptional failures of sysfs_get_dirent(), however it will also fail in the common case when security support is disabled. A few issues: 1/ A dev_warn() report for a common case is too chatty 2/ The setup of this notifier is generic, no need for it to be driven from the nfit driver, it can exist completely in the core. 3/ If it fails for any reason besides security support being disabled, that's fatal and should abort DIMM activation. Userspace may hang if it never gets overwrite notifications. 4/ The dirent needs to be released. Move the call to the core 'dimm' driver, make it conditional on security support being active, make it fatal for the exceptional case, add the missing sysfs_put() at device disable time. Fixes: 7d988097c546 ("...Add security DSM overwrite support") Reviewed-by: Dave Jiang Signed-off-by: Dan Williams --- drivers/acpi/nfit/core.c | 5 ----- drivers/nvdimm/dimm.c | 6 ++++++ drivers/nvdimm/dimm_devs.c | 22 +++++++++++++++++----- drivers/nvdimm/nd.h | 1 + include/linux/libnvdimm.h | 1 - 5 files changed, 24 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c index 5143e11e3b0f..c1fb06654749 100644 --- a/drivers/acpi/nfit/core.c +++ b/drivers/acpi/nfit/core.c @@ -2042,11 +2042,6 @@ static int acpi_nfit_register_dimms(struct acpi_nfit_desc *acpi_desc) if (!nvdimm) continue; - rc = nvdimm_security_setup_events(nvdimm); - if (rc < 0) - dev_warn(acpi_desc->dev, - "security event setup failed: %d\n", rc); - nfit_kernfs = sysfs_get_dirent(nvdimm_kobj(nvdimm)->sd, "nfit"); if (nfit_kernfs) nfit_mem->flags_attr = sysfs_get_dirent(nfit_kernfs, diff --git a/drivers/nvdimm/dimm.c b/drivers/nvdimm/dimm.c index 0cf58cabc9ed..3cf50274fadb 100644 --- a/drivers/nvdimm/dimm.c +++ b/drivers/nvdimm/dimm.c @@ -26,6 +26,12 @@ static int nvdimm_probe(struct device *dev) struct nvdimm_drvdata *ndd; int rc; + rc = nvdimm_security_setup_events(dev); + if (rc < 0) { + dev_err(dev, "security event setup failed: %d\n", rc); + return rc; + } + rc = nvdimm_check_config_data(dev); if (rc) { /* not required for non-aliased nvdimm, ex. NVDIMM-N */ diff --git a/drivers/nvdimm/dimm_devs.c b/drivers/nvdimm/dimm_devs.c index 4890310df874..efe412a6b5b9 100644 --- a/drivers/nvdimm/dimm_devs.c +++ b/drivers/nvdimm/dimm_devs.c @@ -578,13 +578,25 @@ struct nvdimm *__nvdimm_create(struct nvdimm_bus *nvdimm_bus, } EXPORT_SYMBOL_GPL(__nvdimm_create); -int nvdimm_security_setup_events(struct nvdimm *nvdimm) +static void shutdown_security_notify(void *data) { - nvdimm->sec.overwrite_state = sysfs_get_dirent(nvdimm->dev.kobj.sd, - "security"); + struct nvdimm *nvdimm = data; + + sysfs_put(nvdimm->sec.overwrite_state); +} + +int nvdimm_security_setup_events(struct device *dev) +{ + struct nvdimm *nvdimm = to_nvdimm(dev); + + if (nvdimm->sec.state < 0 || !nvdimm->sec.ops + || !nvdimm->sec.ops->overwrite) + return 0; + nvdimm->sec.overwrite_state = sysfs_get_dirent(dev->kobj.sd, "security"); if (!nvdimm->sec.overwrite_state) - return -ENODEV; - return 0; + return -ENOMEM; + + return devm_add_action_or_reset(dev, shutdown_security_notify, nvdimm); } EXPORT_SYMBOL_GPL(nvdimm_security_setup_events); diff --git a/drivers/nvdimm/nd.h b/drivers/nvdimm/nd.h index cfde992684e7..379bf4305e61 100644 --- a/drivers/nvdimm/nd.h +++ b/drivers/nvdimm/nd.h @@ -250,6 +250,7 @@ long nvdimm_clear_poison(struct device *dev, phys_addr_t phys, void nvdimm_set_aliasing(struct device *dev); void nvdimm_set_locked(struct device *dev); void nvdimm_clear_locked(struct device *dev); +int nvdimm_security_setup_events(struct device *dev); #if IS_ENABLED(CONFIG_NVDIMM_KEYS) int nvdimm_security_unlock(struct device *dev); #else diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h index 7315977b64da..ad609617aeb8 100644 --- a/include/linux/libnvdimm.h +++ b/include/linux/libnvdimm.h @@ -235,7 +235,6 @@ static inline struct nvdimm *nvdimm_create(struct nvdimm_bus *nvdimm_bus, cmd_mask, num_flush, flush_wpq, NULL, NULL); } -int nvdimm_security_setup_events(struct nvdimm *nvdimm); const struct nd_cmd_desc *nd_cmd_dimm_desc(int cmd); const struct nd_cmd_desc *nd_cmd_bus_desc(int cmd); u32 nd_cmd_in_size(struct nvdimm *nvdimm, int cmd, -- cgit v1.2.3 From 856c395cfa63b94a1d8215182f0243c222f6f927 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Thu, 17 Jan 2019 23:27:11 -0800 Subject: net: introduce a knob to control whether to inherit devconf config There have been many people complaining about the inconsistent behaviors of IPv4 and IPv6 devconf when creating new network namespaces. Currently, for IPv4, we inherit all current settings from init_net, but for IPv6 we reset all setting to default. This patch introduces a new /proc file /proc/sys/net/core/devconf_inherit_init_net to control the behavior of whether to inhert sysctl current settings from init_net. This file itself is only available in init_net. As demonstrated below: Initial setup in init_net: # cat /proc/sys/net/ipv4/conf/all/rp_filter 2 # cat /proc/sys/net/ipv6/conf/all/accept_dad 1 Default value 0 (current behavior): # ip netns del test # ip netns add test # ip netns exec test cat /proc/sys/net/ipv4/conf/all/rp_filter 2 # ip netns exec test cat /proc/sys/net/ipv6/conf/all/accept_dad 0 Set to 1 (inherit from init_net): # echo 1 > /proc/sys/net/core/devconf_inherit_init_net # ip netns del test # ip netns add test # ip netns exec test cat /proc/sys/net/ipv4/conf/all/rp_filter 2 # ip netns exec test cat /proc/sys/net/ipv6/conf/all/accept_dad 1 Set to 2 (reset to default): # echo 2 > /proc/sys/net/core/devconf_inherit_init_net # ip netns del test # ip netns add test # ip netns exec test cat /proc/sys/net/ipv4/conf/all/rp_filter 0 # ip netns exec test cat /proc/sys/net/ipv6/conf/all/accept_dad 0 Set to a value out of range (invalid): # echo 3 > /proc/sys/net/core/devconf_inherit_init_net -bash: echo: write error: Invalid argument # echo -1 > /proc/sys/net/core/devconf_inherit_init_net -bash: echo: write error: Invalid argument Reported-by: Zhu Yanjun Reported-by: Tonghao Zhang Cc: Nicolas Dichtel Signed-off-by: Cong Wang Acked-by: Nicolas Dichtel Acked-by: Tonghao Zhang Signed-off-by: David S. Miller --- Documentation/sysctl/net.txt | 14 ++++++++++++++ include/linux/netdevice.h | 1 + net/core/sysctl_net_core.c | 18 ++++++++++++++++++ net/ipv4/devinet.c | 43 ++++++++++++++++++++----------------------- net/ipv6/addrconf.c | 5 +++++ 5 files changed, 58 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/Documentation/sysctl/net.txt b/Documentation/sysctl/net.txt index 2793d4eac55f..bc0680706870 100644 --- a/Documentation/sysctl/net.txt +++ b/Documentation/sysctl/net.txt @@ -291,6 +291,20 @@ user space is responsible for creating them if needed. Default : 0 (for compatibility reasons) +devconf_inherit_init_net +---------------------------- + +Controls if a new network namespace should inherit all current +settings under /proc/sys/net/{ipv4,ipv6}/conf/{all,default}/. By +default, we keep the current behavior: for IPv4 we inherit all current +settings from init_net and for IPv6 we reset all settings to default. + +If set to 1, both IPv4 and IPv6 settings are forced to inherit from +current ones in init_net. If set to 2, both IPv4 and IPv6 settings are +forced to reset to their default values. + +Default : 0 (for compatibility reasons) + 2. /proc/sys/net/unix - Parameters for Unix domain sockets ------------------------------------------------------- diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index a57b9a853aab..e675ef97a426 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -630,6 +630,7 @@ struct netdev_queue { } ____cacheline_aligned_in_smp; extern int sysctl_fb_tunnels_only_for_init_net; +extern int sysctl_devconf_inherit_init_net; static inline bool net_has_fallback_tunnels(const struct net *net) { diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index d67ec17f2cc8..84bf2861f45f 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -36,6 +36,15 @@ static int net_msg_warn; /* Unused, but still a sysctl */ int sysctl_fb_tunnels_only_for_init_net __read_mostly = 0; EXPORT_SYMBOL(sysctl_fb_tunnels_only_for_init_net); +/* 0 - Keep current behavior: + * IPv4: inherit all current settings from init_net + * IPv6: reset all settings to default + * 1 - Both inherit all current settings from init_net + * 2 - Both reset all settings to default + */ +int sysctl_devconf_inherit_init_net __read_mostly; +EXPORT_SYMBOL(sysctl_devconf_inherit_init_net); + #ifdef CONFIG_RPS static int rps_sock_flow_sysctl(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) @@ -544,6 +553,15 @@ static struct ctl_table net_core_table[] = { .extra1 = &zero, .extra2 = &one, }, + { + .procname = "devconf_inherit_init_net", + .data = &sysctl_devconf_inherit_init_net, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &two, + }, { } }; diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index cd027639df2f..cd9033245b98 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -2591,32 +2591,32 @@ static __net_init int devinet_init_net(struct net *net) int err; struct ipv4_devconf *all, *dflt; #ifdef CONFIG_SYSCTL - struct ctl_table *tbl = ctl_forward_entry; + struct ctl_table *tbl; struct ctl_table_header *forw_hdr; #endif err = -ENOMEM; - all = &ipv4_devconf; - dflt = &ipv4_devconf_dflt; - - if (!net_eq(net, &init_net)) { - all = kmemdup(all, sizeof(ipv4_devconf), GFP_KERNEL); - if (!all) - goto err_alloc_all; + all = kmemdup(&ipv4_devconf, sizeof(ipv4_devconf), GFP_KERNEL); + if (!all) + goto err_alloc_all; - dflt = kmemdup(dflt, sizeof(ipv4_devconf_dflt), GFP_KERNEL); - if (!dflt) - goto err_alloc_dflt; + dflt = kmemdup(&ipv4_devconf_dflt, sizeof(ipv4_devconf_dflt), GFP_KERNEL); + if (!dflt) + goto err_alloc_dflt; #ifdef CONFIG_SYSCTL - tbl = kmemdup(tbl, sizeof(ctl_forward_entry), GFP_KERNEL); - if (!tbl) - goto err_alloc_ctl; + tbl = kmemdup(ctl_forward_entry, sizeof(ctl_forward_entry), GFP_KERNEL); + if (!tbl) + goto err_alloc_ctl; - tbl[0].data = &all->data[IPV4_DEVCONF_FORWARDING - 1]; - tbl[0].extra1 = all; - tbl[0].extra2 = net; + tbl[0].data = &all->data[IPV4_DEVCONF_FORWARDING - 1]; + tbl[0].extra1 = all; + tbl[0].extra2 = net; #endif + + if (sysctl_devconf_inherit_init_net != 2 && !net_eq(net, &init_net)) { + memcpy(all, init_net.ipv4.devconf_all, sizeof(ipv4_devconf)); + memcpy(dflt, init_net.ipv4.devconf_dflt, sizeof(ipv4_devconf_dflt)); } #ifdef CONFIG_SYSCTL @@ -2646,15 +2646,12 @@ err_reg_ctl: err_reg_dflt: __devinet_sysctl_unregister(net, all, NETCONFA_IFINDEX_ALL); err_reg_all: - if (tbl != ctl_forward_entry) - kfree(tbl); + kfree(tbl); err_alloc_ctl: #endif - if (dflt != &ipv4_devconf_dflt) - kfree(dflt); + kfree(dflt); err_alloc_dflt: - if (all != &ipv4_devconf) - kfree(all); + kfree(all); err_alloc_all: return err; } diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 57198b3c86da..48cd36311901 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -6902,6 +6902,11 @@ static int __net_init addrconf_init_net(struct net *net) if (!dflt) goto err_alloc_dflt; + if (sysctl_devconf_inherit_init_net == 1 && !net_eq(net, &init_net)) { + memcpy(all, init_net.ipv6.devconf_all, sizeof(ipv6_devconf)); + memcpy(dflt, init_net.ipv6.devconf_dflt, sizeof(ipv6_devconf_dflt)); + } + /* these will be inherited by all namespaces */ dflt->autoconf = ipv6_defaults.autoconf; dflt->disable_ipv6 = ipv6_defaults.disable_ipv6; -- cgit v1.2.3 From 7fc5854f8c6efae9e7624970ab49a1eac2faefb1 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Dec 2017 08:38:30 -0800 Subject: writeback: synchronize sync(2) against cgroup writeback membership switches sync_inodes_sb() can race against cgwb (cgroup writeback) membership switches and fail to writeback some inodes. For example, if an inode switches to another wb while sync_inodes_sb() is in progress, the new wb might not be visible to bdi_split_work_to_wbs() at all or the inode might jump from a wb which hasn't issued writebacks yet to one which already has. This patch adds backing_dev_info->wb_switch_rwsem to synchronize cgwb switch path against sync_inodes_sb() so that sync_inodes_sb() is guaranteed to see all the target wbs and inodes can't jump wbs to escape syncing. v2: Fixed misplaced rwsem init. Spotted by Jiufei. Signed-off-by: Tejun Heo Reported-by: Jiufei Xue Link: http://lkml.kernel.org/r/dc694ae2-f07f-61e1-7097-7c8411cee12d@gmail.com Acked-by: Jan Kara Signed-off-by: Jens Axboe --- fs/fs-writeback.c | 40 ++++++++++++++++++++++++++++++++++++++-- include/linux/backing-dev-defs.h | 1 + mm/backing-dev.c | 1 + 3 files changed, 40 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index b40168fcc94a..36855c1f8daf 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -331,11 +331,22 @@ struct inode_switch_wbs_context { struct work_struct work; }; +static void bdi_down_write_wb_switch_rwsem(struct backing_dev_info *bdi) +{ + down_write(&bdi->wb_switch_rwsem); +} + +static void bdi_up_write_wb_switch_rwsem(struct backing_dev_info *bdi) +{ + up_write(&bdi->wb_switch_rwsem); +} + static void inode_switch_wbs_work_fn(struct work_struct *work) { struct inode_switch_wbs_context *isw = container_of(work, struct inode_switch_wbs_context, work); struct inode *inode = isw->inode; + struct backing_dev_info *bdi = inode_to_bdi(inode); struct address_space *mapping = inode->i_mapping; struct bdi_writeback *old_wb = inode->i_wb; struct bdi_writeback *new_wb = isw->new_wb; @@ -343,6 +354,12 @@ static void inode_switch_wbs_work_fn(struct work_struct *work) struct page *page; bool switched = false; + /* + * If @inode switches cgwb membership while sync_inodes_sb() is + * being issued, sync_inodes_sb() might miss it. Synchronize. + */ + down_read(&bdi->wb_switch_rwsem); + /* * By the time control reaches here, RCU grace period has passed * since I_WB_SWITCH assertion and all wb stat update transactions @@ -428,6 +445,8 @@ skip_switch: spin_unlock(&new_wb->list_lock); spin_unlock(&old_wb->list_lock); + up_read(&bdi->wb_switch_rwsem); + if (switched) { wb_wakeup(new_wb); wb_put(old_wb); @@ -468,9 +487,18 @@ static void inode_switch_wbs(struct inode *inode, int new_wb_id) if (inode->i_state & I_WB_SWITCH) return; + /* + * Avoid starting new switches while sync_inodes_sb() is in + * progress. Otherwise, if the down_write protected issue path + * blocks heavily, we might end up starting a large number of + * switches which will block on the rwsem. + */ + if (!down_read_trylock(&bdi->wb_switch_rwsem)) + return; + isw = kzalloc(sizeof(*isw), GFP_ATOMIC); if (!isw) - return; + goto out_unlock; /* find and pin the new wb */ rcu_read_lock(); @@ -504,12 +532,14 @@ static void inode_switch_wbs(struct inode *inode, int new_wb_id) * Let's continue after I_WB_SWITCH is guaranteed to be visible. */ call_rcu(&isw->rcu_head, inode_switch_wbs_rcu_fn); - return; + goto out_unlock; out_free: if (isw->new_wb) wb_put(isw->new_wb); kfree(isw); +out_unlock: + up_read(&bdi->wb_switch_rwsem); } /** @@ -887,6 +917,9 @@ fs_initcall(cgroup_writeback_init); #else /* CONFIG_CGROUP_WRITEBACK */ +static void bdi_down_write_wb_switch_rwsem(struct backing_dev_info *bdi) { } +static void bdi_up_write_wb_switch_rwsem(struct backing_dev_info *bdi) { } + static struct bdi_writeback * locked_inode_to_wb_and_lock_list(struct inode *inode) __releases(&inode->i_lock) @@ -2413,8 +2446,11 @@ void sync_inodes_sb(struct super_block *sb) return; WARN_ON(!rwsem_is_locked(&sb->s_umount)); + /* protect against inode wb switch, see inode_switch_wbs_work_fn() */ + bdi_down_write_wb_switch_rwsem(bdi); bdi_split_work_to_wbs(bdi, &work, false); wb_wait_for_completion(bdi, &done); + bdi_up_write_wb_switch_rwsem(bdi); wait_sb_inodes(sb); } diff --git a/include/linux/backing-dev-defs.h b/include/linux/backing-dev-defs.h index c31157135598..07e02d6df5ad 100644 --- a/include/linux/backing-dev-defs.h +++ b/include/linux/backing-dev-defs.h @@ -190,6 +190,7 @@ struct backing_dev_info { struct radix_tree_root cgwb_tree; /* radix tree of active cgroup wbs */ struct rb_root cgwb_congested_tree; /* their congested states */ struct mutex cgwb_release_mutex; /* protect shutdown of wb structs */ + struct rw_semaphore wb_switch_rwsem; /* no cgwb switch while syncing */ #else struct bdi_writeback_congested *wb_congested; #endif diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 8a8bb8796c6c..72e6d0c55cfa 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -689,6 +689,7 @@ static int cgwb_bdi_init(struct backing_dev_info *bdi) INIT_RADIX_TREE(&bdi->cgwb_tree, GFP_ATOMIC); bdi->cgwb_congested_tree = RB_ROOT; mutex_init(&bdi->cgwb_release_mutex); + init_rwsem(&bdi->wb_switch_rwsem); ret = wb_init(&bdi->wb, bdi, 1, GFP_KERNEL); if (!ret) { -- cgit v1.2.3 From ba5ea614622dca6d675b4cc8a97270569ae13a23 Mon Sep 17 00:00:00 2001 From: Linus Lüssing Date: Mon, 21 Jan 2019 07:26:25 +0100 Subject: bridge: simplify ip_mc_check_igmp() and ipv6_mc_check_mld() calls MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch refactors ip_mc_check_igmp(), ipv6_mc_check_mld() and their callers (more precisely, the Linux bridge) to not rely on the skb_trimmed parameter anymore. An skb with its tail trimmed to the IP packet length was initially introduced for the following three reasons: 1) To be able to verify the ICMPv6 checksum. 2) To be able to distinguish the version of an IGMP or MLD query. They are distinguishable only by their size. 3) To avoid parsing data for an IGMPv3 or MLDv2 report that is beyond the IP packet but still within the skb. The first case still uses a cloned and potentially trimmed skb to verfiy. However, there is no need to propagate it to the caller. For the second and third case explicit IP packet length checks were added. This hopefully makes ip_mc_check_igmp() and ipv6_mc_check_mld() easier to read and verfiy, as well as easier to use. Signed-off-by: Linus Lüssing Signed-off-by: David S. Miller --- include/linux/igmp.h | 11 ++++++++- include/linux/ip.h | 5 ++++ include/linux/ipv6.h | 6 +++++ include/net/addrconf.h | 12 +++++++++- net/batman-adv/multicast.c | 4 ++-- net/bridge/br_multicast.c | 57 +++++++++++++++++++++++----------------------- net/ipv4/igmp.c | 23 ++++--------------- net/ipv6/mcast_snoop.c | 24 ++++--------------- 8 files changed, 70 insertions(+), 72 deletions(-) (limited to 'include/linux') diff --git a/include/linux/igmp.h b/include/linux/igmp.h index 119f53941c12..8b4348f69bc5 100644 --- a/include/linux/igmp.h +++ b/include/linux/igmp.h @@ -18,6 +18,7 @@ #include #include #include +#include #include #include @@ -106,6 +107,14 @@ struct ip_mc_list { #define IGMPV3_QQIC(value) IGMPV3_EXP(0x80, 4, 3, value) #define IGMPV3_MRC(value) IGMPV3_EXP(0x80, 4, 3, value) +static inline int ip_mc_may_pull(struct sk_buff *skb, unsigned int len) +{ + if (skb_transport_offset(skb) + ip_transport_len(skb) < len) + return -EINVAL; + + return pskb_may_pull(skb, len); +} + extern int ip_check_mc_rcu(struct in_device *dev, __be32 mc_addr, __be32 src_addr, u8 proto); extern int igmp_rcv(struct sk_buff *); extern int ip_mc_join_group(struct sock *sk, struct ip_mreqn *imr); @@ -130,6 +139,6 @@ extern void ip_mc_unmap(struct in_device *); extern void ip_mc_remap(struct in_device *); extern void ip_mc_dec_group(struct in_device *in_dev, __be32 addr); extern void ip_mc_inc_group(struct in_device *in_dev, __be32 addr); -int ip_mc_check_igmp(struct sk_buff *skb, struct sk_buff **skb_trimmed); +int ip_mc_check_igmp(struct sk_buff *skb); #endif diff --git a/include/linux/ip.h b/include/linux/ip.h index 492bc6513533..482b7b7c9f30 100644 --- a/include/linux/ip.h +++ b/include/linux/ip.h @@ -34,4 +34,9 @@ static inline struct iphdr *ipip_hdr(const struct sk_buff *skb) { return (struct iphdr *)skb_transport_header(skb); } + +static inline unsigned int ip_transport_len(const struct sk_buff *skb) +{ + return ntohs(ip_hdr(skb)->tot_len) - skb_network_header_len(skb); +} #endif /* _LINUX_IP_H */ diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 495e834c1367..6d45ce784bea 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -104,6 +104,12 @@ static inline struct ipv6hdr *ipipv6_hdr(const struct sk_buff *skb) return (struct ipv6hdr *)skb_transport_header(skb); } +static inline unsigned int ipv6_transport_len(const struct sk_buff *skb) +{ + return ntohs(ipv6_hdr(skb)->payload_len) + sizeof(struct ipv6hdr) - + skb_network_header_len(skb); +} + /* This structure contains results of exthdrs parsing as offsets from skb->nh. diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 1656c5978498..daf11dcb0f70 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -49,6 +49,7 @@ struct prefix_info { struct in6_addr prefix; }; +#include #include #include #include @@ -201,6 +202,15 @@ u32 ipv6_addr_label(struct net *net, const struct in6_addr *addr, /* * multicast prototypes (mcast.c) */ +static inline int ipv6_mc_may_pull(struct sk_buff *skb, + unsigned int len) +{ + if (skb_transport_offset(skb) + ipv6_transport_len(skb) < len) + return -EINVAL; + + return pskb_may_pull(skb, len); +} + int ipv6_sock_mc_join(struct sock *sk, int ifindex, const struct in6_addr *addr); int ipv6_sock_mc_drop(struct sock *sk, int ifindex, @@ -219,7 +229,7 @@ void ipv6_mc_unmap(struct inet6_dev *idev); void ipv6_mc_remap(struct inet6_dev *idev); void ipv6_mc_init_dev(struct inet6_dev *idev); void ipv6_mc_destroy_dev(struct inet6_dev *idev); -int ipv6_mc_check_mld(struct sk_buff *skb, struct sk_buff **skb_trimmed); +int ipv6_mc_check_mld(struct sk_buff *skb); void addrconf_dad_failure(struct sk_buff *skb, struct inet6_ifaddr *ifp); bool ipv6_chk_mcast_addr(struct net_device *dev, const struct in6_addr *group, diff --git a/net/batman-adv/multicast.c b/net/batman-adv/multicast.c index 69244e4598f5..1dd70f048e7b 100644 --- a/net/batman-adv/multicast.c +++ b/net/batman-adv/multicast.c @@ -674,7 +674,7 @@ static void batadv_mcast_mla_update(struct work_struct *work) */ static bool batadv_mcast_is_report_ipv4(struct sk_buff *skb) { - if (ip_mc_check_igmp(skb, NULL) < 0) + if (ip_mc_check_igmp(skb) < 0) return false; switch (igmp_hdr(skb)->type) { @@ -741,7 +741,7 @@ static int batadv_mcast_forw_mode_check_ipv4(struct batadv_priv *bat_priv, */ static bool batadv_mcast_is_report_ipv6(struct sk_buff *skb) { - if (ipv6_mc_check_mld(skb, NULL) < 0) + if (ipv6_mc_check_mld(skb) < 0) return false; switch (icmp6_hdr(skb)->icmp6_type) { diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 3aeff0895669..156c4905639e 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -938,7 +938,7 @@ static int br_ip4_multicast_igmp3_report(struct net_bridge *br, for (i = 0; i < num; i++) { len += sizeof(*grec); - if (!pskb_may_pull(skb, len)) + if (!ip_mc_may_pull(skb, len)) return -EINVAL; grec = (void *)(skb->data + len - sizeof(*grec)); @@ -946,7 +946,7 @@ static int br_ip4_multicast_igmp3_report(struct net_bridge *br, type = grec->grec_type; len += ntohs(grec->grec_nsrcs) * 4; - if (!pskb_may_pull(skb, len)) + if (!ip_mc_may_pull(skb, len)) return -EINVAL; /* We treat this as an IGMPv2 report for now. */ @@ -985,15 +985,17 @@ static int br_ip6_multicast_mld2_report(struct net_bridge *br, struct sk_buff *skb, u16 vid) { + unsigned int nsrcs_offset; const unsigned char *src; struct icmp6hdr *icmp6h; struct mld2_grec *grec; + unsigned int grec_len; int i; int len; int num; int err = 0; - if (!pskb_may_pull(skb, sizeof(*icmp6h))) + if (!ipv6_mc_may_pull(skb, sizeof(*icmp6h))) return -EINVAL; icmp6h = icmp6_hdr(skb); @@ -1003,21 +1005,25 @@ static int br_ip6_multicast_mld2_report(struct net_bridge *br, for (i = 0; i < num; i++) { __be16 *nsrcs, _nsrcs; - nsrcs = skb_header_pointer(skb, - len + offsetof(struct mld2_grec, - grec_nsrcs), + nsrcs_offset = len + offsetof(struct mld2_grec, grec_nsrcs); + + if (skb_transport_offset(skb) + ipv6_transport_len(skb) < + nsrcs_offset + sizeof(_nsrcs)) + return -EINVAL; + + nsrcs = skb_header_pointer(skb, nsrcs_offset, sizeof(_nsrcs), &_nsrcs); if (!nsrcs) return -EINVAL; - if (!pskb_may_pull(skb, - len + sizeof(*grec) + - sizeof(struct in6_addr) * ntohs(*nsrcs))) + grec_len = sizeof(*grec) + + sizeof(struct in6_addr) * ntohs(*nsrcs); + + if (!ipv6_mc_may_pull(skb, len + grec_len)) return -EINVAL; grec = (struct mld2_grec *)(skb->data + len); - len += sizeof(*grec) + - sizeof(struct in6_addr) * ntohs(*nsrcs); + len += grec_len; /* We treat these as MLDv1 reports for now. */ switch (grec->grec_type) { @@ -1219,6 +1225,7 @@ static void br_ip4_multicast_query(struct net_bridge *br, struct sk_buff *skb, u16 vid) { + unsigned int transport_len = ip_transport_len(skb); const struct iphdr *iph = ip_hdr(skb); struct igmphdr *ih = igmp_hdr(skb); struct net_bridge_mdb_entry *mp; @@ -1228,7 +1235,6 @@ static void br_ip4_multicast_query(struct net_bridge *br, struct br_ip saddr; unsigned long max_delay; unsigned long now = jiffies; - unsigned int offset = skb_transport_offset(skb); __be32 group; spin_lock(&br->multicast_lock); @@ -1238,14 +1244,14 @@ static void br_ip4_multicast_query(struct net_bridge *br, group = ih->group; - if (skb->len == offset + sizeof(*ih)) { + if (transport_len == sizeof(*ih)) { max_delay = ih->code * (HZ / IGMP_TIMER_SCALE); if (!max_delay) { max_delay = 10 * HZ; group = 0; } - } else if (skb->len >= offset + sizeof(*ih3)) { + } else if (transport_len >= sizeof(*ih3)) { ih3 = igmpv3_query_hdr(skb); if (ih3->nsrcs) goto out; @@ -1296,6 +1302,7 @@ static int br_ip6_multicast_query(struct net_bridge *br, struct sk_buff *skb, u16 vid) { + unsigned int transport_len = ipv6_transport_len(skb); const struct ipv6hdr *ip6h = ipv6_hdr(skb); struct mld_msg *mld; struct net_bridge_mdb_entry *mp; @@ -1315,7 +1322,7 @@ static int br_ip6_multicast_query(struct net_bridge *br, (port && port->state == BR_STATE_DISABLED)) goto out; - if (skb->len == offset + sizeof(*mld)) { + if (transport_len == sizeof(*mld)) { if (!pskb_may_pull(skb, offset + sizeof(*mld))) { err = -EINVAL; goto out; @@ -1581,12 +1588,11 @@ static int br_multicast_ipv4_rcv(struct net_bridge *br, struct sk_buff *skb, u16 vid) { - struct sk_buff *skb_trimmed = NULL; const unsigned char *src; struct igmphdr *ih; int err; - err = ip_mc_check_igmp(skb, &skb_trimmed); + err = ip_mc_check_igmp(skb); if (err == -ENOMSG) { if (!ipv4_is_local_multicast(ip_hdr(skb)->daddr)) { @@ -1612,19 +1618,16 @@ static int br_multicast_ipv4_rcv(struct net_bridge *br, err = br_ip4_multicast_add_group(br, port, ih->group, vid, src); break; case IGMPV3_HOST_MEMBERSHIP_REPORT: - err = br_ip4_multicast_igmp3_report(br, port, skb_trimmed, vid); + err = br_ip4_multicast_igmp3_report(br, port, skb, vid); break; case IGMP_HOST_MEMBERSHIP_QUERY: - br_ip4_multicast_query(br, port, skb_trimmed, vid); + br_ip4_multicast_query(br, port, skb, vid); break; case IGMP_HOST_LEAVE_MESSAGE: br_ip4_multicast_leave_group(br, port, ih->group, vid, src); break; } - if (skb_trimmed && skb_trimmed != skb) - kfree_skb(skb_trimmed); - br_multicast_count(br, port, skb, BR_INPUT_SKB_CB(skb)->igmp, BR_MCAST_DIR_RX); @@ -1637,12 +1640,11 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, struct sk_buff *skb, u16 vid) { - struct sk_buff *skb_trimmed = NULL; const unsigned char *src; struct mld_msg *mld; int err; - err = ipv6_mc_check_mld(skb, &skb_trimmed); + err = ipv6_mc_check_mld(skb); if (err == -ENOMSG) { if (!ipv6_addr_is_ll_all_nodes(&ipv6_hdr(skb)->daddr)) @@ -1664,10 +1666,10 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, src); break; case ICMPV6_MLD2_REPORT: - err = br_ip6_multicast_mld2_report(br, port, skb_trimmed, vid); + err = br_ip6_multicast_mld2_report(br, port, skb, vid); break; case ICMPV6_MGM_QUERY: - err = br_ip6_multicast_query(br, port, skb_trimmed, vid); + err = br_ip6_multicast_query(br, port, skb, vid); break; case ICMPV6_MGM_REDUCTION: src = eth_hdr(skb)->h_source; @@ -1675,9 +1677,6 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, break; } - if (skb_trimmed && skb_trimmed != skb) - kfree_skb(skb_trimmed); - br_multicast_count(br, port, skb, BR_INPUT_SKB_CB(skb)->igmp, BR_MCAST_DIR_RX); diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 765b2b32c4a4..b1f6d93282d7 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -1544,7 +1544,7 @@ static inline __sum16 ip_mc_validate_checksum(struct sk_buff *skb) return skb_checksum_simple_validate(skb); } -static int __ip_mc_check_igmp(struct sk_buff *skb, struct sk_buff **skb_trimmed) +static int __ip_mc_check_igmp(struct sk_buff *skb) { struct sk_buff *skb_chk; @@ -1566,16 +1566,10 @@ static int __ip_mc_check_igmp(struct sk_buff *skb, struct sk_buff **skb_trimmed) if (ret) goto err; - if (skb_trimmed) - *skb_trimmed = skb_chk; - /* free now unneeded clone */ - else if (skb_chk != skb) - kfree_skb(skb_chk); - ret = 0; err: - if (ret && skb_chk && skb_chk != skb) + if (skb_chk && skb_chk != skb) kfree_skb(skb_chk); return ret; @@ -1584,7 +1578,6 @@ err: /** * ip_mc_check_igmp - checks whether this is a sane IGMP packet * @skb: the skb to validate - * @skb_trimmed: to store an skb pointer trimmed to IPv4 packet tail (optional) * * Checks whether an IPv4 packet is a valid IGMP packet. If so sets * skb transport header accordingly and returns zero. @@ -1594,18 +1587,10 @@ err: * -ENOMSG: IP header validation succeeded but it is not an IGMP packet. * -ENOMEM: A memory allocation failure happened. * - * Optionally, an skb pointer might be provided via skb_trimmed (or set it - * to NULL): After parsing an IGMP packet successfully it will point to - * an skb which has its tail aligned to the IP packet end. This might - * either be the originally provided skb or a trimmed, cloned version if - * the skb frame had data beyond the IP packet. A cloned skb allows us - * to leave the original skb and its full frame unchanged (which might be - * desirable for layer 2 frame jugglers). - * * Caller needs to set the skb network header and free any returned skb if it * differs from the provided skb. */ -int ip_mc_check_igmp(struct sk_buff *skb, struct sk_buff **skb_trimmed) +int ip_mc_check_igmp(struct sk_buff *skb) { int ret = ip_mc_check_iphdr(skb); @@ -1615,7 +1600,7 @@ int ip_mc_check_igmp(struct sk_buff *skb, struct sk_buff **skb_trimmed) if (ip_hdr(skb)->protocol != IPPROTO_IGMP) return -ENOMSG; - return __ip_mc_check_igmp(skb, skb_trimmed); + return __ip_mc_check_igmp(skb); } EXPORT_SYMBOL(ip_mc_check_igmp); diff --git a/net/ipv6/mcast_snoop.c b/net/ipv6/mcast_snoop.c index 9405b04eecc6..1a917dc80d5e 100644 --- a/net/ipv6/mcast_snoop.c +++ b/net/ipv6/mcast_snoop.c @@ -136,8 +136,7 @@ static inline __sum16 ipv6_mc_validate_checksum(struct sk_buff *skb) return skb_checksum_validate(skb, IPPROTO_ICMPV6, ip6_compute_pseudo); } -static int __ipv6_mc_check_mld(struct sk_buff *skb, - struct sk_buff **skb_trimmed) +static int __ipv6_mc_check_mld(struct sk_buff *skb) { struct sk_buff *skb_chk = NULL; @@ -160,16 +159,10 @@ static int __ipv6_mc_check_mld(struct sk_buff *skb, if (ret) goto err; - if (skb_trimmed) - *skb_trimmed = skb_chk; - /* free now unneeded clone */ - else if (skb_chk != skb) - kfree_skb(skb_chk); - ret = 0; err: - if (ret && skb_chk && skb_chk != skb) + if (skb_chk && skb_chk != skb) kfree_skb(skb_chk); return ret; @@ -178,7 +171,6 @@ err: /** * ipv6_mc_check_mld - checks whether this is a sane MLD packet * @skb: the skb to validate - * @skb_trimmed: to store an skb pointer trimmed to IPv6 packet tail (optional) * * Checks whether an IPv6 packet is a valid MLD packet. If so sets * skb transport header accordingly and returns zero. @@ -188,18 +180,10 @@ err: * -ENOMSG: IP header validation succeeded but it is not an MLD packet. * -ENOMEM: A memory allocation failure happened. * - * Optionally, an skb pointer might be provided via skb_trimmed (or set it - * to NULL): After parsing an MLD packet successfully it will point to - * an skb which has its tail aligned to the IP packet end. This might - * either be the originally provided skb or a trimmed, cloned version if - * the skb frame had data beyond the IP packet. A cloned skb allows us - * to leave the original skb and its full frame unchanged (which might be - * desirable for layer 2 frame jugglers). - * * Caller needs to set the skb network header and free any returned skb if it * differs from the provided skb. */ -int ipv6_mc_check_mld(struct sk_buff *skb, struct sk_buff **skb_trimmed) +int ipv6_mc_check_mld(struct sk_buff *skb) { int ret; @@ -211,6 +195,6 @@ int ipv6_mc_check_mld(struct sk_buff *skb, struct sk_buff **skb_trimmed) if (ret < 0) return ret; - return __ipv6_mc_check_mld(skb, skb_trimmed); + return __ipv6_mc_check_mld(skb); } EXPORT_SYMBOL(ipv6_mc_check_mld); -- cgit v1.2.3 From 4b3087c7e37f9e499127201849e33960dc81da11 Mon Sep 17 00:00:00 2001 From: Linus Lüssing Date: Mon, 21 Jan 2019 07:26:28 +0100 Subject: bridge: Snoop Multicast Router Advertisements MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When multiple multicast routers are present in a broadcast domain then only one of them will be detectable via IGMP/MLD query snooping. The multicast router with the lowest IP address will become the selected and active querier while all other multicast routers will then refrain from sending queries. To detect such rather silent multicast routers, too, RFC4286 ("Multicast Router Discovery") provides a standardized protocol to detect multicast routers for multicast snooping switches. This patch implements the necessary MRD Advertisement message parsing and after successful processing adds such routers to the internal multicast router list. Signed-off-by: Linus Lüssing Signed-off-by: David S. Miller --- include/linux/in.h | 5 +++++ include/net/addrconf.h | 15 +++++++++++++ include/uapi/linux/icmpv6.h | 2 ++ include/uapi/linux/igmp.h | 1 + net/bridge/br_multicast.c | 55 +++++++++++++++++++++++++++++++++++++++++++++ net/ipv6/mcast_snoop.c | 5 ++++- 6 files changed, 82 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/in.h b/include/linux/in.h index 31b493734763..435e7f2a513a 100644 --- a/include/linux/in.h +++ b/include/linux/in.h @@ -60,6 +60,11 @@ static inline bool ipv4_is_lbcast(__be32 addr) return addr == htonl(INADDR_BROADCAST); } +static inline bool ipv4_is_all_snoopers(__be32 addr) +{ + return addr == htonl(INADDR_ALLSNOOPERS_GROUP); +} + static inline bool ipv4_is_zeronet(__be32 addr) { return (addr & htonl(0xff000000)) == htonl(0x00000000); diff --git a/include/net/addrconf.h b/include/net/addrconf.h index daf11dcb0f70..20d523ee2fec 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -229,6 +229,7 @@ void ipv6_mc_unmap(struct inet6_dev *idev); void ipv6_mc_remap(struct inet6_dev *idev); void ipv6_mc_init_dev(struct inet6_dev *idev); void ipv6_mc_destroy_dev(struct inet6_dev *idev); +int ipv6_mc_check_icmpv6(struct sk_buff *skb); int ipv6_mc_check_mld(struct sk_buff *skb); void addrconf_dad_failure(struct sk_buff *skb, struct inet6_ifaddr *ifp); @@ -499,6 +500,20 @@ static inline bool ipv6_addr_is_solict_mult(const struct in6_addr *addr) #endif } +static inline bool ipv6_addr_is_all_snoopers(const struct in6_addr *addr) +{ +#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) && BITS_PER_LONG == 64 + __be64 *p = (__be64 *)addr; + + return ((p[0] ^ cpu_to_be64(0xff02000000000000UL)) | + (p[1] ^ cpu_to_be64(0x6a))) == 0UL; +#else + return ((addr->s6_addr32[0] ^ htonl(0xff020000)) | + addr->s6_addr32[1] | addr->s6_addr32[2] | + (addr->s6_addr32[3] ^ htonl(0x0000006a))) == 0; +#endif +} + #ifdef CONFIG_PROC_FS int if6_proc_init(void); void if6_proc_exit(void); diff --git a/include/uapi/linux/icmpv6.h b/include/uapi/linux/icmpv6.h index caf8dc019250..325395f56bfa 100644 --- a/include/uapi/linux/icmpv6.h +++ b/include/uapi/linux/icmpv6.h @@ -108,6 +108,8 @@ struct icmp6hdr { #define ICMPV6_MOBILE_PREFIX_SOL 146 #define ICMPV6_MOBILE_PREFIX_ADV 147 +#define ICMPV6_MRDISC_ADV 151 + /* * Codes for Destination Unreachable */ diff --git a/include/uapi/linux/igmp.h b/include/uapi/linux/igmp.h index 7e44ac02ca18..90c28bc466c6 100644 --- a/include/uapi/linux/igmp.h +++ b/include/uapi/linux/igmp.h @@ -93,6 +93,7 @@ struct igmpv3_query { #define IGMP_MTRACE_RESP 0x1e #define IGMP_MTRACE 0x1f +#define IGMP_MRDISC_ADV 0x30 /* From RFC4286 */ /* * Use the BSD names for these for compatibility diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 2366f4a2780e..2c46c7aca571 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -29,10 +30,12 @@ #include #include #if IS_ENABLED(CONFIG_IPV6) +#include #include #include #include #include +#include #endif #include "br_private.h" @@ -1583,6 +1586,19 @@ static void br_multicast_pim(struct net_bridge *br, br_multicast_mark_router(br, port); } +static int br_ip4_multicast_mrd_rcv(struct net_bridge *br, + struct net_bridge_port *port, + struct sk_buff *skb) +{ + if (ip_hdr(skb)->protocol != IPPROTO_IGMP || + igmp_hdr(skb)->type != IGMP_MRDISC_ADV) + return -ENOMSG; + + br_multicast_mark_router(br, port); + + return 0; +} + static int br_multicast_ipv4_rcv(struct net_bridge *br, struct net_bridge_port *port, struct sk_buff *skb, @@ -1600,7 +1616,15 @@ static int br_multicast_ipv4_rcv(struct net_bridge *br, } else if (pim_ipv4_all_pim_routers(ip_hdr(skb)->daddr)) { if (ip_hdr(skb)->protocol == IPPROTO_PIM) br_multicast_pim(br, port, skb); + } else if (ipv4_is_all_snoopers(ip_hdr(skb)->daddr)) { + err = br_ip4_multicast_mrd_rcv(br, port, skb); + + if (err < 0 && err != -ENOMSG) { + br_multicast_err_count(br, port, skb->protocol); + return err; + } } + return 0; } else if (err < 0) { br_multicast_err_count(br, port, skb->protocol); @@ -1635,6 +1659,27 @@ static int br_multicast_ipv4_rcv(struct net_bridge *br, } #if IS_ENABLED(CONFIG_IPV6) +static int br_ip6_multicast_mrd_rcv(struct net_bridge *br, + struct net_bridge_port *port, + struct sk_buff *skb) +{ + int ret; + + if (ipv6_hdr(skb)->nexthdr != IPPROTO_ICMPV6) + return -ENOMSG; + + ret = ipv6_mc_check_icmpv6(skb); + if (ret < 0) + return ret; + + if (icmp6_hdr(skb)->icmp6_type != ICMPV6_MRDISC_ADV) + return -ENOMSG; + + br_multicast_mark_router(br, port); + + return 0; +} + static int br_multicast_ipv6_rcv(struct net_bridge *br, struct net_bridge_port *port, struct sk_buff *skb, @@ -1649,6 +1694,16 @@ static int br_multicast_ipv6_rcv(struct net_bridge *br, if (err == -ENOMSG) { if (!ipv6_addr_is_ll_all_nodes(&ipv6_hdr(skb)->daddr)) BR_INPUT_SKB_CB(skb)->mrouters_only = 1; + + if (ipv6_addr_is_all_snoopers(&ipv6_hdr(skb)->daddr)) { + err = br_ip6_multicast_mrd_rcv(br, port, skb); + + if (err < 0 && err != -ENOMSG) { + br_multicast_err_count(br, port, skb->protocol); + return err; + } + } + return 0; } else if (err < 0) { br_multicast_err_count(br, port, skb->protocol); diff --git a/net/ipv6/mcast_snoop.c b/net/ipv6/mcast_snoop.c index a72ddfc40eb3..55e2ac179f28 100644 --- a/net/ipv6/mcast_snoop.c +++ b/net/ipv6/mcast_snoop.c @@ -41,6 +41,8 @@ static int ipv6_mc_check_ip6hdr(struct sk_buff *skb) if (skb->len < len || len <= offset) return -EINVAL; + skb_set_transport_header(skb, offset); + return 0; } @@ -142,7 +144,7 @@ static inline __sum16 ipv6_mc_validate_checksum(struct sk_buff *skb) return skb_checksum_validate(skb, IPPROTO_ICMPV6, ip6_compute_pseudo); } -static int ipv6_mc_check_icmpv6(struct sk_buff *skb) +int ipv6_mc_check_icmpv6(struct sk_buff *skb) { unsigned int len = skb_transport_offset(skb) + sizeof(struct icmp6hdr); unsigned int transport_len = ipv6_transport_len(skb); @@ -161,6 +163,7 @@ static int ipv6_mc_check_icmpv6(struct sk_buff *skb) return 0; } +EXPORT_SYMBOL(ipv6_mc_check_icmpv6); /** * ipv6_mc_check_mld - checks whether this is a sane MLD packet -- cgit v1.2.3 From c75860e48a7634ff8dc050842211f79a0e4e6c46 Mon Sep 17 00:00:00 2001 From: Tomer Tayar Date: Sun, 20 Jan 2019 11:36:38 +0200 Subject: qed: Add infrastructure for error detection and recovery This patch adds the detection and handling of a parity error ("process kill event"), including the update of the protocol drivers, and the prevention of any HW access that will lead to device access towards the host while recovery is in progress. It also provides the means for the protocol drivers to trigger a recovery process on their decision. Signed-off-by: Tomer Tayar Signed-off-by: Ariel Elior Signed-off-by: Michal Kalderon Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed.h | 4 ++ drivers/net/ethernet/qlogic/qed/qed_dev.c | 41 +++++++---- drivers/net/ethernet/qlogic/qed/qed_hsi.h | 2 +- drivers/net/ethernet/qlogic/qed/qed_hw.c | 11 +++ drivers/net/ethernet/qlogic/qed/qed_main.c | 30 ++++++++ drivers/net/ethernet/qlogic/qed/qed_mcp.c | 94 ++++++++++++++++++++++++++ drivers/net/ethernet/qlogic/qed/qed_mcp.h | 32 +++++++++ drivers/net/ethernet/qlogic/qed/qed_reg_addr.h | 2 + drivers/net/ethernet/qlogic/qed/qed_spq.c | 22 ++++++ drivers/net/ethernet/qlogic/qed/qed_sriov.c | 9 ++- include/linux/qed/qed_if.h | 20 ++++++ 11 files changed, 251 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h index b352e313e1f6..3b0955d34716 100644 --- a/drivers/net/ethernet/qlogic/qed/qed.h +++ b/drivers/net/ethernet/qlogic/qed/qed.h @@ -804,6 +804,9 @@ struct qed_dev { u32 mcp_nvm_resp; + /* Recovery */ + bool recov_in_prog; + /* Linux specific here */ struct qede_dev *edev; struct pci_dev *pdev; @@ -943,6 +946,7 @@ void qed_link_update(struct qed_hwfn *hwfn, struct qed_ptt *ptt); u32 qed_unzip_data(struct qed_hwfn *p_hwfn, u32 input_len, u8 *input_buf, u32 max_size, u8 *unzip_buf); +void qed_schedule_recovery_handler(struct qed_hwfn *p_hwfn); void qed_get_protocol_stats(struct qed_dev *cdev, enum qed_mcp_protocol_type type, union qed_mcp_protocol_stats *stats); diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index fa5f07e65672..b17003d9066c 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -2140,6 +2140,11 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params) "Load request was sent. Load code: 0x%x\n", load_code); + /* Only relevant for recovery: + * Clear the indication after LOAD_REQ is responded by the MFW. + */ + cdev->recov_in_prog = false; + qed_mcp_set_capabilities(p_hwfn, p_hwfn->p_main_ptt); qed_reset_mb_shadow(p_hwfn, p_hwfn->p_main_ptt); @@ -2291,6 +2296,9 @@ static void qed_hw_timers_stop(struct qed_dev *cdev, qed_wr(p_hwfn, p_ptt, TM_REG_PF_ENABLE_CONN, 0x0); qed_wr(p_hwfn, p_ptt, TM_REG_PF_ENABLE_TASK, 0x0); + if (cdev->recov_in_prog) + return; + for (i = 0; i < QED_HW_STOP_RETRY_LIMIT; i++) { if ((!qed_rd(p_hwfn, p_ptt, TM_REG_PF_SCAN_ACTIVE_CONN)) && @@ -2353,12 +2361,14 @@ int qed_hw_stop(struct qed_dev *cdev) p_hwfn->hw_init_done = false; /* Send unload command to MCP */ - rc = qed_mcp_unload_req(p_hwfn, p_ptt); - if (rc) { - DP_NOTICE(p_hwfn, - "Failed sending a UNLOAD_REQ command. rc = %d.\n", - rc); - rc2 = -EINVAL; + if (!cdev->recov_in_prog) { + rc = qed_mcp_unload_req(p_hwfn, p_ptt); + if (rc) { + DP_NOTICE(p_hwfn, + "Failed sending a UNLOAD_REQ command. rc = %d.\n", + rc); + rc2 = -EINVAL; + } } qed_slowpath_irq_sync(p_hwfn); @@ -2400,16 +2410,18 @@ int qed_hw_stop(struct qed_dev *cdev) qed_wr(p_hwfn, p_ptt, DORQ_REG_PF_DB_ENABLE, 0); qed_wr(p_hwfn, p_ptt, QM_REG_PF_EN, 0); - qed_mcp_unload_done(p_hwfn, p_ptt); - if (rc) { - DP_NOTICE(p_hwfn, - "Failed sending a UNLOAD_DONE command. rc = %d.\n", - rc); - rc2 = -EINVAL; + if (!cdev->recov_in_prog) { + rc = qed_mcp_unload_done(p_hwfn, p_ptt); + if (rc) { + DP_NOTICE(p_hwfn, + "Failed sending a UNLOAD_DONE command. rc = %d.\n", + rc); + rc2 = -EINVAL; + } } } - if (IS_PF(cdev)) { + if (IS_PF(cdev) && !cdev->recov_in_prog) { p_hwfn = QED_LEADING_HWFN(cdev); p_ptt = QED_LEADING_HWFN(cdev)->p_main_ptt; @@ -3459,6 +3471,7 @@ static int qed_hw_prepare_single(struct qed_hwfn *p_hwfn, void __iomem *p_doorbells, enum qed_pci_personality personality) { + struct qed_dev *cdev = p_hwfn->cdev; int rc = 0; /* Split PCI bars evenly between hwfns */ @@ -3511,7 +3524,7 @@ static int qed_hw_prepare_single(struct qed_hwfn *p_hwfn, /* Sending a mailbox to the MFW should be done after qed_get_hw_info() * is called as it sets the ports number in an engine. */ - if (IS_LEAD_HWFN(p_hwfn)) { + if (IS_LEAD_HWFN(p_hwfn) && !cdev->recov_in_prog) { rc = qed_mcp_initiate_pf_flr(p_hwfn, p_hwfn->p_main_ptt); if (rc) DP_NOTICE(p_hwfn, "Failed to initiate PF FLR\n"); diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h index b13cfb449d8f..417121e74ee9 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h @@ -12827,7 +12827,7 @@ enum MFW_DRV_MSG_TYPE { MFW_DRV_MSG_LLDP_DATA_UPDATED, MFW_DRV_MSG_DCBX_REMOTE_MIB_UPDATED, MFW_DRV_MSG_DCBX_OPERATIONAL_MIB_UPDATED, - MFW_DRV_MSG_RESERVED4, + MFW_DRV_MSG_ERROR_RECOVERY, MFW_DRV_MSG_BW_UPDATE, MFW_DRV_MSG_S_TAG_UPDATE, MFW_DRV_MSG_GET_LAN_STATS, diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.c b/drivers/net/ethernet/qlogic/qed/qed_hw.c index 70504dcf4087..72ec1c6bdf70 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hw.c +++ b/drivers/net/ethernet/qlogic/qed/qed_hw.c @@ -703,6 +703,17 @@ static int qed_dmae_execute_command(struct qed_hwfn *p_hwfn, int qed_status = 0; u32 offset = 0; + if (p_hwfn->cdev->recov_in_prog) { + DP_VERBOSE(p_hwfn, + NETIF_MSG_HW, + "Recovery is in progress. Avoid DMAE transaction [{src: addr 0x%llx, type %d}, {dst: addr 0x%llx, type %d}, size %d].\n", + src_addr, src_type, dst_addr, dst_type, + size_in_dwords); + + /* Let the flow complete w/o any error handling */ + return 0; + } + qed_dmae_opcode(p_hwfn, (src_type == QED_DMAE_ADDRESS_GRC), (dst_type == QED_DMAE_ADDRESS_GRC), diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index 6adf5bda9811..b47352643fb5 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -359,6 +359,8 @@ static struct qed_dev *qed_probe(struct pci_dev *pdev, qed_init_dp(cdev, params->dp_module, params->dp_level); + cdev->recov_in_prog = params->recov_in_prog; + rc = qed_init_pci(cdev, pdev); if (rc) { DP_ERR(cdev, "init pci failed\n"); @@ -2203,6 +2205,15 @@ static int qed_nvm_get_image(struct qed_dev *cdev, enum qed_nvm_images type, return qed_mcp_get_nvm_image(hwfn, type, buf, len); } +void qed_schedule_recovery_handler(struct qed_hwfn *p_hwfn) +{ + struct qed_common_cb_ops *ops = p_hwfn->cdev->protocol_ops.common; + void *cookie = p_hwfn->cdev->ops_cookie; + + if (ops && ops->schedule_recovery_handler) + ops->schedule_recovery_handler(cookie); +} + static int qed_set_coalesce(struct qed_dev *cdev, u16 rx_coal, u16 tx_coal, void *handle) { @@ -2226,6 +2237,23 @@ static int qed_set_led(struct qed_dev *cdev, enum qed_led_mode mode) return status; } +static int qed_recovery_process(struct qed_dev *cdev) +{ + struct qed_hwfn *p_hwfn = QED_LEADING_HWFN(cdev); + struct qed_ptt *p_ptt; + int rc = 0; + + p_ptt = qed_ptt_acquire(p_hwfn); + if (!p_ptt) + return -EAGAIN; + + rc = qed_start_recovery_process(p_hwfn, p_ptt); + + qed_ptt_release(p_hwfn, p_ptt); + + return rc; +} + static int qed_update_wol(struct qed_dev *cdev, bool enabled) { struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); @@ -2380,6 +2408,8 @@ const struct qed_common_ops qed_common_ops_pass = { .nvm_get_image = &qed_nvm_get_image, .set_coalesce = &qed_set_coalesce, .set_led = &qed_set_led, + .recovery_process = &qed_recovery_process, + .recovery_prolog = &qed_recovery_prolog, .update_drv_state = &qed_update_drv_state, .update_mac = &qed_update_mac, .update_mtu = &qed_update_mtu, diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c index 1024484d7dd8..bb8541847aa5 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c @@ -1549,6 +1549,60 @@ int qed_mcp_set_link(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, bool b_up) return 0; } +u32 qed_get_process_kill_counter(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt) +{ + u32 path_offsize_addr, path_offsize, path_addr, proc_kill_cnt; + + if (IS_VF(p_hwfn->cdev)) + return -EINVAL; + + path_offsize_addr = SECTION_OFFSIZE_ADDR(p_hwfn->mcp_info->public_base, + PUBLIC_PATH); + path_offsize = qed_rd(p_hwfn, p_ptt, path_offsize_addr); + path_addr = SECTION_ADDR(path_offsize, QED_PATH_ID(p_hwfn)); + + proc_kill_cnt = qed_rd(p_hwfn, p_ptt, + path_addr + + offsetof(struct public_path, process_kill)) & + PROCESS_KILL_COUNTER_MASK; + + return proc_kill_cnt; +} + +static void qed_mcp_handle_process_kill(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt) +{ + struct qed_dev *cdev = p_hwfn->cdev; + u32 proc_kill_cnt; + + /* Prevent possible attentions/interrupts during the recovery handling + * and till its load phase, during which they will be re-enabled. + */ + qed_int_igu_disable_int(p_hwfn, p_ptt); + + DP_NOTICE(p_hwfn, "Received a process kill indication\n"); + + /* The following operations should be done once, and thus in CMT mode + * are carried out by only the first HW function. + */ + if (p_hwfn != QED_LEADING_HWFN(cdev)) + return; + + if (cdev->recov_in_prog) { + DP_NOTICE(p_hwfn, + "Ignoring the indication since a recovery process is already in progress\n"); + return; + } + + cdev->recov_in_prog = true; + + proc_kill_cnt = qed_get_process_kill_counter(p_hwfn, p_ptt); + DP_NOTICE(p_hwfn, "Process kill counter: %d\n", proc_kill_cnt); + + qed_schedule_recovery_handler(p_hwfn); +} + static void qed_mcp_send_protocol_stats(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, enum MFW_DRV_MSG_TYPE type) @@ -1779,6 +1833,9 @@ int qed_mcp_handle_events(struct qed_hwfn *p_hwfn, case MFW_DRV_MSG_TRANSCEIVER_STATE_CHANGE: qed_mcp_handle_transceiver_change(p_hwfn, p_ptt); break; + case MFW_DRV_MSG_ERROR_RECOVERY: + qed_mcp_handle_process_kill(p_hwfn, p_ptt); + break; case MFW_DRV_MSG_GET_LAN_STATS: case MFW_DRV_MSG_GET_FCOE_STATS: case MFW_DRV_MSG_GET_ISCSI_STATS: @@ -2324,6 +2381,43 @@ int qed_mcp_get_flash_size(struct qed_hwfn *p_hwfn, return 0; } +int qed_start_recovery_process(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) +{ + struct qed_dev *cdev = p_hwfn->cdev; + + if (cdev->recov_in_prog) { + DP_NOTICE(p_hwfn, + "Avoid triggering a recovery since such a process is already in progress\n"); + return -EAGAIN; + } + + DP_NOTICE(p_hwfn, "Triggering a recovery process\n"); + qed_wr(p_hwfn, p_ptt, MISC_REG_AEU_GENERAL_ATTN_35, 0x1); + + return 0; +} + +#define QED_RECOVERY_PROLOG_SLEEP_MS 100 + +int qed_recovery_prolog(struct qed_dev *cdev) +{ + struct qed_hwfn *p_hwfn = QED_LEADING_HWFN(cdev); + struct qed_ptt *p_ptt = p_hwfn->p_main_ptt; + int rc; + + /* Allow ongoing PCIe transactions to complete */ + msleep(QED_RECOVERY_PROLOG_SLEEP_MS); + + /* Clear the PF's internal FID_enable in the PXP */ + rc = qed_pglueb_set_pfid_enable(p_hwfn, p_ptt, false); + if (rc) + DP_NOTICE(p_hwfn, + "qed_pglueb_set_pfid_enable() failed. rc = %d.\n", + rc); + + return rc; +} + static int qed_mcp_config_vf_msix_bb(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u8 vf_id, u8 num) diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.h b/drivers/net/ethernet/qlogic/qed/qed_mcp.h index 387c5e649136..6e1d72a669ae 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.h +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.h @@ -440,6 +440,38 @@ qed_mcp_send_drv_version(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, struct qed_mcp_drv_version *p_ver); +/** + * @brief Read the MFW process kill counter + * + * @param p_hwfn + * @param p_ptt + * + * @return u32 + */ +u32 qed_get_process_kill_counter(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt); + +/** + * @brief Trigger a recovery process + * + * @param p_hwfn + * @param p_ptt + * + * @return int + */ +int qed_start_recovery_process(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt); + +/** + * @brief A recovery handler must call this function as its first step. + * It is assumed that the handler is not run from an interrupt context. + * + * @param cdev + * @param p_ptt + * + * @return int + */ +int qed_recovery_prolog(struct qed_dev *cdev); + /** * @brief Notify MFW about the change in base device properties * diff --git a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h index 8939ed6e08b7..5ce825ca5f24 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h +++ b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h @@ -518,6 +518,8 @@ 0x180824UL #define MISC_REG_AEU_GENERAL_ATTN_0 \ 0x008400UL +#define MISC_REG_AEU_GENERAL_ATTN_35 \ + 0x00848cUL #define CAU_REG_SB_ADDR_MEMORY \ 0x1c8000UL #define CAU_REG_SB_VAR_MEMORY \ diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c index eb88bbc6b193..3e0f7c46bb1b 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_spq.c +++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c @@ -790,6 +790,17 @@ static int qed_spq_pend_post(struct qed_hwfn *p_hwfn) SPQ_HIGH_PRI_RESERVE_DEFAULT); } +static void qed_spq_recov_set_ret_code(struct qed_spq_entry *p_ent, + u8 *fw_return_code) +{ + if (!fw_return_code) + return; + + if (p_ent->elem.hdr.protocol_id == PROTOCOLID_ROCE || + p_ent->elem.hdr.protocol_id == PROTOCOLID_IWARP) + *fw_return_code = RDMA_RETURN_OK; +} + /* Avoid overriding of SPQ entries when getting out-of-order completions, by * marking the completions in a bitmap and increasing the chain consumer only * for the first successive completed entries. @@ -825,6 +836,17 @@ int qed_spq_post(struct qed_hwfn *p_hwfn, return -EINVAL; } + if (p_hwfn->cdev->recov_in_prog) { + DP_VERBOSE(p_hwfn, + QED_MSG_SPQ, + "Recovery is in progress. Skip spq post [cmd %02x protocol %02x]\n", + p_ent->elem.hdr.cmd_id, p_ent->elem.hdr.protocol_id); + + /* Let the flow complete w/o any error handling */ + qed_spq_recov_set_ret_code(p_ent, fw_return_code); + return 0; + } + /* Complete the entry */ rc = qed_spq_fill_entry(p_hwfn, p_ent); diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index ca6290fa0f30..71e28be58102 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -4447,6 +4447,13 @@ int qed_sriov_disable(struct qed_dev *cdev, bool pci_enabled) if (cdev->p_iov_info && cdev->p_iov_info->num_vfs && pci_enabled) pci_disable_sriov(cdev->pdev); + if (cdev->recov_in_prog) { + DP_VERBOSE(cdev, + QED_MSG_IOV, + "Skip SRIOV disable operations in the device since a recovery is in progress\n"); + goto out; + } + for_each_hwfn(cdev, i) { struct qed_hwfn *hwfn = &cdev->hwfns[i]; struct qed_ptt *ptt = qed_ptt_acquire(hwfn); @@ -4486,7 +4493,7 @@ int qed_sriov_disable(struct qed_dev *cdev, bool pci_enabled) qed_ptt_release(hwfn, ptt); } - +out: qed_iov_set_vfs_to_disable(cdev, false); return 0; diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h index 91c536a01b56..c2a1b7dbe4eb 100644 --- a/include/linux/qed/qed_if.h +++ b/include/linux/qed/qed_if.h @@ -764,6 +764,7 @@ struct qed_probe_params { u32 dp_module; u8 dp_level; bool is_vf; + bool recov_in_prog; }; #define QED_DRV_VER_STR_SIZE 12 @@ -810,6 +811,7 @@ struct qed_common_cb_ops { void (*arfs_filter_op)(void *dev, void *fltr, u8 fw_rc); void (*link_update)(void *dev, struct qed_link_output *link); + void (*schedule_recovery_handler)(void *dev); void (*dcbx_aen)(void *dev, struct qed_dcbx_get *get, u32 mib_type); void (*get_generic_tlv_data)(void *dev, struct qed_generic_tlvs *data); void (*get_protocol_tlv_data)(void *dev, void *data); @@ -1057,6 +1059,24 @@ struct qed_common_ops { int (*db_recovery_del)(struct qed_dev *cdev, void __iomem *db_addr, void *db_data); +/** + * @brief recovery_process - Trigger a recovery process + * + * @param cdev + * + * @return 0 on success, error otherwise. + */ + int (*recovery_process)(struct qed_dev *cdev); + +/** + * @brief recovery_prolog - Execute the prolog operations of a recovery process + * + * @param cdev + * + * @return 0 on success, error otherwise. + */ + int (*recovery_prolog)(struct qed_dev *cdev); + /** * @brief update_drv_state - API to inform the change in the driver state. * -- cgit v1.2.3 From 278396de78a9b59a692bc140233bde3a9d8a8a31 Mon Sep 17 00:00:00 2001 From: Tomer Tayar Date: Sun, 20 Jan 2019 11:36:39 +0200 Subject: qede: Error recovery process This patch adds the error recovery process in the qede driver. The process includes a partial/customized driver unload and load, which allows it to look like a short suspend period to the kernel while preserving the net devices' state. Signed-off-by: Tomer Tayar Signed-off-by: Ariel Elior Signed-off-by: Michal Kalderon Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qede/qede.h | 3 + drivers/net/ethernet/qlogic/qede/qede_main.c | 300 ++++++++++++++++++++++----- drivers/net/ethernet/qlogic/qede/qede_rdma.c | 64 ++++-- include/linux/qed/qede_rdma.h | 21 +- 4 files changed, 314 insertions(+), 74 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h index 613249d1e967..843416404aeb 100644 --- a/drivers/net/ethernet/qlogic/qede/qede.h +++ b/drivers/net/ethernet/qlogic/qede/qede.h @@ -162,6 +162,7 @@ struct qede_rdma_dev { struct list_head entry; struct list_head rdma_event_list; struct workqueue_struct *rdma_wq; + bool exp_recovery; }; struct qede_ptp; @@ -264,6 +265,7 @@ struct qede_dev { enum QEDE_STATE { QEDE_STATE_CLOSED, QEDE_STATE_OPEN, + QEDE_STATE_RECOVERY, }; #define HILO_U64(hi, lo) ((((u64)(hi)) << 32) + (lo)) @@ -462,6 +464,7 @@ struct qede_fastpath { #define QEDE_CSUM_UNNECESSARY BIT(1) #define QEDE_TUNN_CSUM_UNNECESSARY BIT(2) +#define QEDE_SP_RECOVERY 0 #define QEDE_SP_RX_MODE 1 #ifdef CONFIG_RFS_ACCEL diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index 5a74fcbdbc2b..de955f2b2980 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -133,23 +133,12 @@ static int qede_probe(struct pci_dev *pdev, const struct pci_device_id *id); static void qede_remove(struct pci_dev *pdev); static void qede_shutdown(struct pci_dev *pdev); static void qede_link_update(void *dev, struct qed_link_output *link); +static void qede_schedule_recovery_handler(void *dev); +static void qede_recovery_handler(struct qede_dev *edev); static void qede_get_eth_tlv_data(void *edev, void *data); static void qede_get_generic_tlv_data(void *edev, struct qed_generic_tlvs *data); -/* The qede lock is used to protect driver state change and driver flows that - * are not reentrant. - */ -void __qede_lock(struct qede_dev *edev) -{ - mutex_lock(&edev->qede_lock); -} - -void __qede_unlock(struct qede_dev *edev) -{ - mutex_unlock(&edev->qede_lock); -} - #ifdef CONFIG_QED_SRIOV static int qede_set_vf_vlan(struct net_device *ndev, int vf, u16 vlan, u8 qos, __be16 vlan_proto) @@ -231,6 +220,7 @@ static struct qed_eth_cb_ops qede_ll_ops = { .arfs_filter_op = qede_arfs_filter_op, #endif .link_update = qede_link_update, + .schedule_recovery_handler = qede_schedule_recovery_handler, .get_generic_tlv_data = qede_get_generic_tlv_data, .get_protocol_tlv_data = qede_get_eth_tlv_data, }, @@ -950,11 +940,57 @@ err: return -ENOMEM; } +/* The qede lock is used to protect driver state change and driver flows that + * are not reentrant. + */ +void __qede_lock(struct qede_dev *edev) +{ + mutex_lock(&edev->qede_lock); +} + +void __qede_unlock(struct qede_dev *edev) +{ + mutex_unlock(&edev->qede_lock); +} + +/* This version of the lock should be used when acquiring the RTNL lock is also + * needed in addition to the internal qede lock. + */ +void qede_lock(struct qede_dev *edev) +{ + rtnl_lock(); + __qede_lock(edev); +} + +void qede_unlock(struct qede_dev *edev) +{ + __qede_unlock(edev); + rtnl_unlock(); +} + static void qede_sp_task(struct work_struct *work) { struct qede_dev *edev = container_of(work, struct qede_dev, sp_task.work); + /* The locking scheme depends on the specific flag: + * In case of QEDE_SP_RECOVERY, acquiring the RTNL lock is required to + * ensure that ongoing flows are ended and new ones are not started. + * In other cases - only the internal qede lock should be acquired. + */ + + if (test_and_clear_bit(QEDE_SP_RECOVERY, &edev->sp_flags)) { +#ifdef CONFIG_QED_SRIOV + /* SRIOV must be disabled outside the lock to avoid a deadlock. + * The recovery of the active VFs is currently not supported. + */ + qede_sriov_configure(edev->pdev, 0); +#endif + qede_lock(edev); + qede_recovery_handler(edev); + qede_unlock(edev); + } + __qede_lock(edev); if (test_and_clear_bit(QEDE_SP_RX_MODE, &edev->sp_flags)) @@ -1031,8 +1067,13 @@ static void qede_log_probe(struct qede_dev *edev) enum qede_probe_mode { QEDE_PROBE_NORMAL, + QEDE_PROBE_RECOVERY, }; +#define QEDE_RDMA_PROBE_MODE(mode) \ + ((mode) == QEDE_PROBE_NORMAL ? QEDE_RDMA_PROBE_NORMAL \ + : QEDE_RDMA_PROBE_RECOVERY) + static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, bool is_vf, enum qede_probe_mode mode) { @@ -1051,6 +1092,7 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, probe_params.dp_module = dp_module; probe_params.dp_level = dp_level; probe_params.is_vf = is_vf; + probe_params.recov_in_prog = (mode == QEDE_PROBE_RECOVERY); cdev = qed_ops->common->probe(pdev, &probe_params); if (!cdev) { rc = -ENODEV; @@ -1078,11 +1120,20 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, if (rc) goto err2; - edev = qede_alloc_etherdev(cdev, pdev, &dev_info, dp_module, - dp_level); - if (!edev) { - rc = -ENOMEM; - goto err2; + if (mode != QEDE_PROBE_RECOVERY) { + edev = qede_alloc_etherdev(cdev, pdev, &dev_info, dp_module, + dp_level); + if (!edev) { + rc = -ENOMEM; + goto err2; + } + } else { + struct net_device *ndev = pci_get_drvdata(pdev); + + edev = netdev_priv(ndev); + edev->cdev = cdev; + memset(&edev->stats, 0, sizeof(edev->stats)); + memcpy(&edev->dev_info, &dev_info, sizeof(dev_info)); } if (is_vf) @@ -1090,28 +1141,31 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, qede_init_ndev(edev); - rc = qede_rdma_dev_add(edev); + rc = qede_rdma_dev_add(edev, QEDE_RDMA_PROBE_MODE(mode)); if (rc) goto err3; - /* Prepare the lock prior to the registration of the netdev, - * as once it's registered we might reach flows requiring it - * [it's even possible to reach a flow needing it directly - * from there, although it's unlikely]. - */ - INIT_DELAYED_WORK(&edev->sp_task, qede_sp_task); - mutex_init(&edev->qede_lock); - rc = register_netdev(edev->ndev); - if (rc) { - DP_NOTICE(edev, "Cannot register net-device\n"); - goto err4; + if (mode != QEDE_PROBE_RECOVERY) { + /* Prepare the lock prior to the registration of the netdev, + * as once it's registered we might reach flows requiring it + * [it's even possible to reach a flow needing it directly + * from there, although it's unlikely]. + */ + INIT_DELAYED_WORK(&edev->sp_task, qede_sp_task); + mutex_init(&edev->qede_lock); + + rc = register_netdev(edev->ndev); + if (rc) { + DP_NOTICE(edev, "Cannot register net-device\n"); + goto err4; + } } edev->ops->common->set_name(cdev, edev->ndev->name); /* PTP not supported on VFs */ if (!is_vf) - qede_ptp_enable(edev, true); + qede_ptp_enable(edev, (mode == QEDE_PROBE_NORMAL)); edev->ops->register_ops(cdev, &qede_ll_ops, edev); @@ -1126,7 +1180,7 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, return 0; err4: - qede_rdma_dev_remove(edev); + qede_rdma_dev_remove(edev, QEDE_RDMA_PROBE_MODE(mode)); err3: free_netdev(edev->ndev); err2: @@ -1162,8 +1216,13 @@ static int qede_probe(struct pci_dev *pdev, const struct pci_device_id *id) enum qede_remove_mode { QEDE_REMOVE_NORMAL, + QEDE_REMOVE_RECOVERY, }; +#define QEDE_RDMA_REMOVE_MODE(mode) \ + ((mode) == QEDE_REMOVE_NORMAL ? QEDE_RDMA_REMOVE_NORMAL \ + : QEDE_RDMA_REMOVE_RECOVERY) + static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode) { struct net_device *ndev = pci_get_drvdata(pdev); @@ -1172,15 +1231,19 @@ static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode) DP_INFO(edev, "Starting qede_remove\n"); - qede_rdma_dev_remove(edev); - unregister_netdev(ndev); - cancel_delayed_work_sync(&edev->sp_task); + qede_rdma_dev_remove(edev, QEDE_RDMA_REMOVE_MODE(mode)); - qede_ptp_disable(edev); + if (mode != QEDE_REMOVE_RECOVERY) { + unregister_netdev(ndev); - edev->ops->common->set_power_state(cdev, PCI_D0); + cancel_delayed_work_sync(&edev->sp_task); - pci_set_drvdata(pdev, NULL); + edev->ops->common->set_power_state(cdev, PCI_D0); + + pci_set_drvdata(pdev, NULL); + } + + qede_ptp_disable(edev); /* Use global ops since we've freed edev */ qed_ops->common->slowpath_stop(cdev); @@ -1194,7 +1257,8 @@ static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode) * [e.g., QED register callbacks] won't break anything when * accessing the netdevice. */ - free_netdev(ndev); + if (mode != QEDE_REMOVE_RECOVERY) + free_netdev(ndev); dev_info(&pdev->dev, "Ending qede_remove successfully\n"); } @@ -1539,6 +1603,58 @@ static int qede_alloc_mem_load(struct qede_dev *edev) return 0; } +static void qede_empty_tx_queue(struct qede_dev *edev, + struct qede_tx_queue *txq) +{ + unsigned int pkts_compl = 0, bytes_compl = 0; + struct netdev_queue *netdev_txq; + int rc, len = 0; + + netdev_txq = netdev_get_tx_queue(edev->ndev, txq->ndev_txq_id); + + while (qed_chain_get_cons_idx(&txq->tx_pbl) != + qed_chain_get_prod_idx(&txq->tx_pbl)) { + DP_VERBOSE(edev, NETIF_MSG_IFDOWN, + "Freeing a packet on tx queue[%d]: chain_cons 0x%x, chain_prod 0x%x\n", + txq->index, qed_chain_get_cons_idx(&txq->tx_pbl), + qed_chain_get_prod_idx(&txq->tx_pbl)); + + rc = qede_free_tx_pkt(edev, txq, &len); + if (rc) { + DP_NOTICE(edev, + "Failed to free a packet on tx queue[%d]: chain_cons 0x%x, chain_prod 0x%x\n", + txq->index, + qed_chain_get_cons_idx(&txq->tx_pbl), + qed_chain_get_prod_idx(&txq->tx_pbl)); + break; + } + + bytes_compl += len; + pkts_compl++; + txq->sw_tx_cons++; + } + + netdev_tx_completed_queue(netdev_txq, pkts_compl, bytes_compl); +} + +static void qede_empty_tx_queues(struct qede_dev *edev) +{ + int i; + + for_each_queue(i) + if (edev->fp_array[i].type & QEDE_FASTPATH_TX) { + int cos; + + for_each_cos_in_txq(edev, cos) { + struct qede_fastpath *fp; + + fp = &edev->fp_array[i]; + qede_empty_tx_queue(edev, + &fp->txq[cos]); + } + } +} + /* This function inits fp content and resets the SB, RXQ and TXQ structures */ static void qede_init_fp(struct qede_dev *edev) { @@ -2053,6 +2169,7 @@ out: enum qede_unload_mode { QEDE_UNLOAD_NORMAL, + QEDE_UNLOAD_RECOVERY, }; static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode, @@ -2068,7 +2185,8 @@ static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode, clear_bit(QEDE_FLAGS_LINK_REQUESTED, &edev->flags); - edev->state = QEDE_STATE_CLOSED; + if (mode != QEDE_UNLOAD_RECOVERY) + edev->state = QEDE_STATE_CLOSED; qede_rdma_dev_event_close(edev); @@ -2076,17 +2194,20 @@ static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode, netif_tx_disable(edev->ndev); netif_carrier_off(edev->ndev); - /* Reset the link */ - memset(&link_params, 0, sizeof(link_params)); - link_params.link_up = false; - edev->ops->common->set_link(edev->cdev, &link_params); - rc = qede_stop_queues(edev); - if (rc) { - qede_sync_free_irqs(edev); - goto out; - } + if (mode != QEDE_UNLOAD_RECOVERY) { + /* Reset the link */ + memset(&link_params, 0, sizeof(link_params)); + link_params.link_up = false; + edev->ops->common->set_link(edev->cdev, &link_params); - DP_INFO(edev, "Stopped Queues\n"); + rc = qede_stop_queues(edev); + if (rc) { + qede_sync_free_irqs(edev); + goto out; + } + + DP_INFO(edev, "Stopped Queues\n"); + } qede_vlan_mark_nonconfigured(edev); edev->ops->fastpath_stop(edev->cdev); @@ -2102,18 +2223,26 @@ static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode, qede_napi_disable_remove(edev); + if (mode == QEDE_UNLOAD_RECOVERY) + qede_empty_tx_queues(edev); + qede_free_mem_load(edev); qede_free_fp_array(edev); out: if (!is_locked) __qede_unlock(edev); + + if (mode != QEDE_UNLOAD_RECOVERY) + DP_NOTICE(edev, "Link is down\n"); + DP_INFO(edev, "Ending qede unload\n"); } enum qede_load_mode { QEDE_LOAD_NORMAL, QEDE_LOAD_RELOAD, + QEDE_LOAD_RECOVERY, }; static int qede_load(struct qede_dev *edev, enum qede_load_mode mode, @@ -2293,6 +2422,77 @@ static void qede_link_update(void *dev, struct qed_link_output *link) } } +static void qede_schedule_recovery_handler(void *dev) +{ + struct qede_dev *edev = dev; + + if (edev->state == QEDE_STATE_RECOVERY) { + DP_NOTICE(edev, + "Avoid scheduling a recovery handling since already in recovery state\n"); + return; + } + + set_bit(QEDE_SP_RECOVERY, &edev->sp_flags); + schedule_delayed_work(&edev->sp_task, 0); + + DP_INFO(edev, "Scheduled a recovery handler\n"); +} + +static void qede_recovery_failed(struct qede_dev *edev) +{ + netdev_err(edev->ndev, "Recovery handling has failed. Power cycle is needed.\n"); + + netif_device_detach(edev->ndev); + + if (edev->cdev) + edev->ops->common->set_power_state(edev->cdev, PCI_D3hot); +} + +static void qede_recovery_handler(struct qede_dev *edev) +{ + u32 curr_state = edev->state; + int rc; + + DP_NOTICE(edev, "Starting a recovery process\n"); + + /* No need to acquire first the qede_lock since is done by qede_sp_task + * before calling this function. + */ + edev->state = QEDE_STATE_RECOVERY; + + edev->ops->common->recovery_prolog(edev->cdev); + + if (curr_state == QEDE_STATE_OPEN) + qede_unload(edev, QEDE_UNLOAD_RECOVERY, true); + + __qede_remove(edev->pdev, QEDE_REMOVE_RECOVERY); + + rc = __qede_probe(edev->pdev, edev->dp_module, edev->dp_level, + IS_VF(edev), QEDE_PROBE_RECOVERY); + if (rc) { + edev->cdev = NULL; + goto err; + } + + if (curr_state == QEDE_STATE_OPEN) { + rc = qede_load(edev, QEDE_LOAD_RECOVERY, true); + if (rc) + goto err; + + qede_config_rx_mode(edev->ndev); + udp_tunnel_get_rx_info(edev->ndev); + } + + edev->state = curr_state; + + DP_NOTICE(edev, "Recovery handling is done\n"); + + return; + +err: + qede_recovery_failed(edev); +} + static bool qede_is_txq_full(struct qede_dev *edev, struct qede_tx_queue *txq) { struct netdev_queue *netdev_txq; diff --git a/drivers/net/ethernet/qlogic/qede/qede_rdma.c b/drivers/net/ethernet/qlogic/qede/qede_rdma.c index 1900bf7e67d1..9668e5e47d5f 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_rdma.c +++ b/drivers/net/ethernet/qlogic/qede/qede_rdma.c @@ -50,6 +50,8 @@ static void _qede_rdma_dev_add(struct qede_dev *edev) if (!qedr_drv) return; + /* Leftovers from previous error recovery */ + edev->rdma_info.exp_recovery = false; edev->rdma_info.qedr_dev = qedr_drv->add(edev->cdev, edev->pdev, edev->ndev); } @@ -87,21 +89,26 @@ static void qede_rdma_destroy_wq(struct qede_dev *edev) destroy_workqueue(edev->rdma_info.rdma_wq); } -int qede_rdma_dev_add(struct qede_dev *edev) +int qede_rdma_dev_add(struct qede_dev *edev, enum qede_rdma_probe_mode mode) { - int rc = 0; + int rc; - if (qede_rdma_supported(edev)) { - rc = qede_rdma_create_wq(edev); - if (rc) - return rc; + if (!qede_rdma_supported(edev)) + return 0; - INIT_LIST_HEAD(&edev->rdma_info.entry); - mutex_lock(&qedr_dev_list_lock); - list_add_tail(&edev->rdma_info.entry, &qedr_dev_list); - _qede_rdma_dev_add(edev); - mutex_unlock(&qedr_dev_list_lock); - } + /* Cannot start qedr while recovering since it wasn't fully stopped */ + if (mode == QEDE_RDMA_PROBE_RECOVERY) + return 0; + + rc = qede_rdma_create_wq(edev); + if (rc) + return rc; + + INIT_LIST_HEAD(&edev->rdma_info.entry); + mutex_lock(&qedr_dev_list_lock); + list_add_tail(&edev->rdma_info.entry, &qedr_dev_list); + _qede_rdma_dev_add(edev); + mutex_unlock(&qedr_dev_list_lock); return rc; } @@ -110,19 +117,31 @@ static void _qede_rdma_dev_remove(struct qede_dev *edev) { if (qedr_drv && qedr_drv->remove && edev->rdma_info.qedr_dev) qedr_drv->remove(edev->rdma_info.qedr_dev); - edev->rdma_info.qedr_dev = NULL; } -void qede_rdma_dev_remove(struct qede_dev *edev) +void qede_rdma_dev_remove(struct qede_dev *edev, + enum qede_rdma_remove_mode mode) { if (!qede_rdma_supported(edev)) return; - qede_rdma_destroy_wq(edev); - mutex_lock(&qedr_dev_list_lock); - _qede_rdma_dev_remove(edev); - list_del(&edev->rdma_info.entry); - mutex_unlock(&qedr_dev_list_lock); + /* Cannot remove qedr while recovering since it wasn't fully stopped */ + if (mode == QEDE_RDMA_REMOVE_NORMAL) { + qede_rdma_destroy_wq(edev); + mutex_lock(&qedr_dev_list_lock); + if (!edev->rdma_info.exp_recovery) + _qede_rdma_dev_remove(edev); + edev->rdma_info.qedr_dev = NULL; + list_del(&edev->rdma_info.entry); + mutex_unlock(&qedr_dev_list_lock); + } else { + if (!edev->rdma_info.exp_recovery) { + mutex_lock(&qedr_dev_list_lock); + _qede_rdma_dev_remove(edev); + mutex_unlock(&qedr_dev_list_lock); + } + edev->rdma_info.exp_recovery = true; + } } static void _qede_rdma_dev_open(struct qede_dev *edev) @@ -204,7 +223,8 @@ void qede_rdma_unregister_driver(struct qedr_driver *drv) mutex_lock(&qedr_dev_list_lock); list_for_each_entry(edev, &qedr_dev_list, rdma_info.entry) { - if (edev->rdma_info.qedr_dev) + /* If device has experienced recovery it was already removed */ + if (edev->rdma_info.qedr_dev && !edev->rdma_info.exp_recovery) _qede_rdma_dev_remove(edev); } qedr_drv = NULL; @@ -284,6 +304,10 @@ static void qede_rdma_add_event(struct qede_dev *edev, { struct qede_rdma_event_work *event_node; + /* If a recovery was experienced avoid adding the event */ + if (edev->rdma_info.exp_recovery) + return; + if (!edev->rdma_info.qedr_dev) return; diff --git a/include/linux/qed/qede_rdma.h b/include/linux/qed/qede_rdma.h index 9904617a9730..e29d7199c10e 100644 --- a/include/linux/qed/qede_rdma.h +++ b/include/linux/qed/qede_rdma.h @@ -55,6 +55,16 @@ struct qede_rdma_event_work { enum qede_rdma_event event; }; +enum qede_rdma_probe_mode { + QEDE_RDMA_PROBE_NORMAL, + QEDE_RDMA_PROBE_RECOVERY, +}; + +enum qede_rdma_remove_mode { + QEDE_RDMA_REMOVE_NORMAL, + QEDE_RDMA_REMOVE_RECOVERY, +}; + struct qedr_driver { unsigned char name[32]; @@ -74,21 +84,24 @@ void qede_rdma_unregister_driver(struct qedr_driver *drv); bool qede_rdma_supported(struct qede_dev *dev); #if IS_ENABLED(CONFIG_QED_RDMA) -int qede_rdma_dev_add(struct qede_dev *dev); +int qede_rdma_dev_add(struct qede_dev *dev, enum qede_rdma_probe_mode mode); void qede_rdma_dev_event_open(struct qede_dev *dev); void qede_rdma_dev_event_close(struct qede_dev *dev); -void qede_rdma_dev_remove(struct qede_dev *dev); +void qede_rdma_dev_remove(struct qede_dev *dev, + enum qede_rdma_remove_mode mode); void qede_rdma_event_changeaddr(struct qede_dev *edr); #else -static inline int qede_rdma_dev_add(struct qede_dev *dev) +static inline int qede_rdma_dev_add(struct qede_dev *dev, + enum qede_rdma_probe_mode mode) { return 0; } static inline void qede_rdma_dev_event_open(struct qede_dev *dev) {} static inline void qede_rdma_dev_event_close(struct qede_dev *dev) {} -static inline void qede_rdma_dev_remove(struct qede_dev *dev) {} +static inline void qede_rdma_dev_remove(struct qede_dev *dev, + enum qede_rdma_remove_mode mode) {} static inline void qede_rdma_event_changeaddr(struct qede_dev *edr) {} #endif #endif -- cgit v1.2.3 From 6815d8b09282c1df8e016bd2fabf25ada6d4462b Mon Sep 17 00:00:00 2001 From: Yangbo Lu Date: Mon, 21 Jan 2019 18:41:39 +0800 Subject: ptp_qoriq: support external trigger stamp FIFO The external trigger stamp FIFO was introduced as a new feature for QorIQ 1588 timer IP block. This patch is to support it by adding a new dts property "fsl,extts-fifo". Any QorIQ 1588 timer supporting this feature is required to add this property in its dts node. In addition, the FIFO should be cleaned up before enabling external trigger interrupts. Otherwise, there will be interrupts immediately just after enabling external trigger interrupts. Signed-off-by: Yangbo Lu Signed-off-by: Vladimir Oltean Signed-off-by: David S. Miller --- drivers/ptp/ptp_qoriq.c | 68 ++++++++++++++++++++++++++++++++++--------- include/linux/fsl/ptp_qoriq.h | 3 ++ 2 files changed, 57 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/drivers/ptp/ptp_qoriq.c b/drivers/ptp/ptp_qoriq.c index 274321471d50..a2e7702db3a4 100644 --- a/drivers/ptp/ptp_qoriq.c +++ b/drivers/ptp/ptp_qoriq.c @@ -88,6 +88,49 @@ static void set_fipers(struct qoriq_ptp *qoriq_ptp) qoriq_write(®s->fiper_regs->tmr_fiper2, qoriq_ptp->tmr_fiper2); } +static int extts_clean_up(struct qoriq_ptp *qoriq_ptp, int index, + bool update_event) +{ + struct qoriq_ptp_registers *regs = &qoriq_ptp->regs; + struct ptp_clock_event event; + void __iomem *reg_etts_l; + void __iomem *reg_etts_h; + u32 valid, stat, lo, hi; + + switch (index) { + case 0: + valid = ETS1_VLD; + reg_etts_l = ®s->etts_regs->tmr_etts1_l; + reg_etts_h = ®s->etts_regs->tmr_etts1_h; + break; + case 1: + valid = ETS2_VLD; + reg_etts_l = ®s->etts_regs->tmr_etts2_l; + reg_etts_h = ®s->etts_regs->tmr_etts2_h; + break; + default: + return -EINVAL; + } + + event.type = PTP_CLOCK_EXTTS; + event.index = index; + + do { + lo = qoriq_read(reg_etts_l); + hi = qoriq_read(reg_etts_h); + + if (update_event) { + event.timestamp = ((u64) hi) << 32; + event.timestamp |= lo; + ptp_clock_event(qoriq_ptp->clock, &event); + } + + stat = qoriq_read(®s->ctrl_regs->tmr_stat); + } while (qoriq_ptp->extts_fifo_support && (stat & valid)); + + return 0; +} + /* * Interrupt service routine */ @@ -111,24 +154,12 @@ static irqreturn_t isr(int irq, void *priv) if (irqs & ETS1) { ack |= ETS1; - hi = qoriq_read(®s->etts_regs->tmr_etts1_h); - lo = qoriq_read(®s->etts_regs->tmr_etts1_l); - event.type = PTP_CLOCK_EXTTS; - event.index = 0; - event.timestamp = ((u64) hi) << 32; - event.timestamp |= lo; - ptp_clock_event(qoriq_ptp->clock, &event); + extts_clean_up(qoriq_ptp, 0, true); } if (irqs & ETS2) { ack |= ETS2; - hi = qoriq_read(®s->etts_regs->tmr_etts2_h); - lo = qoriq_read(®s->etts_regs->tmr_etts2_l); - event.type = PTP_CLOCK_EXTTS; - event.index = 1; - event.timestamp = ((u64) hi) << 32; - event.timestamp |= lo; - ptp_clock_event(qoriq_ptp->clock, &event); + extts_clean_up(qoriq_ptp, 1, true); } if (irqs & ALM2) { @@ -278,6 +309,10 @@ static int ptp_qoriq_enable(struct ptp_clock_info *ptp, default: return -EINVAL; } + + if (on) + extts_clean_up(qoriq_ptp, rq->extts.index, false); + break; case PTP_CLK_REQ_PPS: bit = PP1EN; @@ -441,6 +476,11 @@ static int qoriq_ptp_probe(struct platform_device *dev) if (of_property_read_u32(node, "fsl,cksel", &qoriq_ptp->cksel)) qoriq_ptp->cksel = DEFAULT_CKSEL; + if (of_property_read_bool(node, "fsl,extts-fifo")) + qoriq_ptp->extts_fifo_support = true; + else + qoriq_ptp->extts_fifo_support = false; + if (of_property_read_u32(node, "fsl,tclk-period", &qoriq_ptp->tclk_period) || of_property_read_u32(node, diff --git a/include/linux/fsl/ptp_qoriq.h b/include/linux/fsl/ptp_qoriq.h index c1f003aadcce..43b4b442f6a4 100644 --- a/include/linux/fsl/ptp_qoriq.h +++ b/include/linux/fsl/ptp_qoriq.h @@ -120,6 +120,8 @@ struct qoriq_ptp_registers { /* Bit definitions for the TMR_STAT register */ #define STAT_VEC_SHIFT (0) /* Timer general purpose status vector */ #define STAT_VEC_MASK (0x3f) +#define ETS1_VLD (1<<24) +#define ETS2_VLD (1<<25) /* Bit definitions for the TMR_PRSC register */ #define PRSC_OCK_SHIFT (0) /* Output clock division/prescale factor. */ @@ -141,6 +143,7 @@ struct qoriq_ptp { struct ptp_clock *clock; struct ptp_clock_info caps; struct resource *rsrc; + bool extts_fifo_support; int irq; int phc_index; u64 alarm_interval; /* for periodic alarm */ -- cgit v1.2.3 From 19df7510d5cf077c2e88a7690fb7617e6d341beb Mon Sep 17 00:00:00 2001 From: Yangbo Lu Date: Mon, 21 Jan 2019 18:41:42 +0800 Subject: ptp: add debugfs support for ptp_qoriq This patch is to add debugfs support for ptp_qoriq. Current debugfs supports to control fiper1/fiper2 loopback mode. If the loopback mode is enabled, the fiper1/fiper2 pulse is looped back into trigger1/ trigger2 input. This is very useful for validating hardware and driver without external hardware. Below is an example to enable fiper1 loopback. echo 1 > /sys/kernel/debug/2d10e00.ptp_clock/fiper1-loopback Signed-off-by: Yangbo Lu Signed-off-by: David S. Miller --- drivers/ptp/Kconfig | 2 +- drivers/ptp/Makefile | 4 +- drivers/ptp/ptp_qoriq.c | 3 ++ drivers/ptp/ptp_qoriq_debugfs.c | 101 ++++++++++++++++++++++++++++++++++++++++ include/linux/fsl/ptp_qoriq.h | 12 +++++ 5 files changed, 120 insertions(+), 2 deletions(-) create mode 100644 drivers/ptp/ptp_qoriq_debugfs.c (limited to 'include/linux') diff --git a/drivers/ptp/Kconfig b/drivers/ptp/Kconfig index d137c480db46..aeb4a8b2e0af 100644 --- a/drivers/ptp/Kconfig +++ b/drivers/ptp/Kconfig @@ -53,7 +53,7 @@ config PTP_1588_CLOCK_QORIQ packets using the SO_TIMESTAMPING API. To compile this driver as a module, choose M here: the module - will be called ptp_qoriq. + will be called ptp-qoriq. config PTP_1588_CLOCK_IXP46X tristate "Intel IXP46x as PTP clock" diff --git a/drivers/ptp/Makefile b/drivers/ptp/Makefile index 19efa9cfa950..677d1d178a3e 100644 --- a/drivers/ptp/Makefile +++ b/drivers/ptp/Makefile @@ -9,4 +9,6 @@ obj-$(CONFIG_PTP_1588_CLOCK_DTE) += ptp_dte.o obj-$(CONFIG_PTP_1588_CLOCK_IXP46X) += ptp_ixp46x.o obj-$(CONFIG_PTP_1588_CLOCK_PCH) += ptp_pch.o obj-$(CONFIG_PTP_1588_CLOCK_KVM) += ptp_kvm.o -obj-$(CONFIG_PTP_1588_CLOCK_QORIQ) += ptp_qoriq.o +obj-$(CONFIG_PTP_1588_CLOCK_QORIQ) += ptp-qoriq.o +ptp-qoriq-y += ptp_qoriq.o +ptp-qoriq-$(CONFIG_DEBUG_FS) += ptp_qoriq_debugfs.o diff --git a/drivers/ptp/ptp_qoriq.c b/drivers/ptp/ptp_qoriq.c index a2e7702db3a4..43416b2e8a13 100644 --- a/drivers/ptp/ptp_qoriq.c +++ b/drivers/ptp/ptp_qoriq.c @@ -471,6 +471,7 @@ static int qoriq_ptp_probe(struct platform_device *dev) err = -EINVAL; + qoriq_ptp->dev = &dev->dev; qoriq_ptp->caps = ptp_qoriq_caps; if (of_property_read_u32(node, "fsl,cksel", &qoriq_ptp->cksel)) @@ -572,6 +573,7 @@ static int qoriq_ptp_probe(struct platform_device *dev) } qoriq_ptp->phc_index = ptp_clock_index(qoriq_ptp->clock); + ptp_qoriq_create_debugfs(qoriq_ptp); platform_set_drvdata(dev, qoriq_ptp); return 0; @@ -597,6 +599,7 @@ static int qoriq_ptp_remove(struct platform_device *dev) qoriq_write(®s->ctrl_regs->tmr_temask, 0); qoriq_write(®s->ctrl_regs->tmr_ctrl, 0); + ptp_qoriq_remove_debugfs(qoriq_ptp); ptp_clock_unregister(qoriq_ptp->clock); iounmap(qoriq_ptp->base); release_resource(qoriq_ptp->rsrc); diff --git a/drivers/ptp/ptp_qoriq_debugfs.c b/drivers/ptp/ptp_qoriq_debugfs.c new file mode 100644 index 000000000000..d904332b240d --- /dev/null +++ b/drivers/ptp/ptp_qoriq_debugfs.c @@ -0,0 +1,101 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* Copyright 2019 NXP + */ +#include +#include +#include + +static int ptp_qoriq_fiper1_lpbk_get(void *data, u64 *val) +{ + struct qoriq_ptp *qoriq_ptp = data; + struct qoriq_ptp_registers *regs = &qoriq_ptp->regs; + u32 ctrl; + + ctrl = qoriq_read(®s->ctrl_regs->tmr_ctrl); + *val = ctrl & PP1L ? 1 : 0; + + return 0; +} + +static int ptp_qoriq_fiper1_lpbk_set(void *data, u64 val) +{ + struct qoriq_ptp *qoriq_ptp = data; + struct qoriq_ptp_registers *regs = &qoriq_ptp->regs; + u32 ctrl; + + ctrl = qoriq_read(®s->ctrl_regs->tmr_ctrl); + if (val == 0) + ctrl &= ~PP1L; + else + ctrl |= PP1L; + + qoriq_write(®s->ctrl_regs->tmr_ctrl, ctrl); + return 0; +} + +DEFINE_SIMPLE_ATTRIBUTE(ptp_qoriq_fiper1_fops, ptp_qoriq_fiper1_lpbk_get, + ptp_qoriq_fiper1_lpbk_set, "%llu\n"); + +static int ptp_qoriq_fiper2_lpbk_get(void *data, u64 *val) +{ + struct qoriq_ptp *qoriq_ptp = data; + struct qoriq_ptp_registers *regs = &qoriq_ptp->regs; + u32 ctrl; + + ctrl = qoriq_read(®s->ctrl_regs->tmr_ctrl); + *val = ctrl & PP2L ? 1 : 0; + + return 0; +} + +static int ptp_qoriq_fiper2_lpbk_set(void *data, u64 val) +{ + struct qoriq_ptp *qoriq_ptp = data; + struct qoriq_ptp_registers *regs = &qoriq_ptp->regs; + u32 ctrl; + + ctrl = qoriq_read(®s->ctrl_regs->tmr_ctrl); + if (val == 0) + ctrl &= ~PP2L; + else + ctrl |= PP2L; + + qoriq_write(®s->ctrl_regs->tmr_ctrl, ctrl); + return 0; +} + +DEFINE_SIMPLE_ATTRIBUTE(ptp_qoriq_fiper2_fops, ptp_qoriq_fiper2_lpbk_get, + ptp_qoriq_fiper2_lpbk_set, "%llu\n"); + +void ptp_qoriq_create_debugfs(struct qoriq_ptp *qoriq_ptp) +{ + struct dentry *root; + + root = debugfs_create_dir(dev_name(qoriq_ptp->dev), NULL); + if (IS_ERR(root)) + return; + if (!root) + goto err_root; + + qoriq_ptp->debugfs_root = root; + + if (!debugfs_create_file("fiper1-loopback", 0600, root, qoriq_ptp, + &ptp_qoriq_fiper1_fops)) + goto err_node; + if (!debugfs_create_file("fiper2-loopback", 0600, root, qoriq_ptp, + &ptp_qoriq_fiper2_fops)) + goto err_node; + return; + +err_node: + debugfs_remove_recursive(root); + qoriq_ptp->debugfs_root = NULL; +err_root: + dev_err(qoriq_ptp->dev, "failed to initialize debugfs\n"); +} + +void ptp_qoriq_remove_debugfs(struct qoriq_ptp *qoriq_ptp) +{ + debugfs_remove_recursive(qoriq_ptp->debugfs_root); + qoriq_ptp->debugfs_root = NULL; +} diff --git a/include/linux/fsl/ptp_qoriq.h b/include/linux/fsl/ptp_qoriq.h index 43b4b442f6a4..94e9797e434c 100644 --- a/include/linux/fsl/ptp_qoriq.h +++ b/include/linux/fsl/ptp_qoriq.h @@ -143,6 +143,8 @@ struct qoriq_ptp { struct ptp_clock *clock; struct ptp_clock_info caps; struct resource *rsrc; + struct dentry *debugfs_root; + struct device *dev; bool extts_fifo_support; int irq; int phc_index; @@ -169,4 +171,14 @@ static inline void qoriq_write(unsigned __iomem *addr, u32 val) iowrite32be(val, addr); } +#ifdef CONFIG_DEBUG_FS +void ptp_qoriq_create_debugfs(struct qoriq_ptp *qoriq_ptp); +void ptp_qoriq_remove_debugfs(struct qoriq_ptp *qoriq_ptp); +#else +static inline void ptp_qoriq_create_debugfs(struct qoriq_ptp *qoriq_ptp) +{ } +static inline void ptp_qoriq_remove_debugfs(struct qoriq_ptp *qoriq_ptp) +{ } +#endif + #endif -- cgit v1.2.3 From 52875a04f4b26e7ef30a288ea096f7cfec0e93cd Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 22 Jan 2019 22:45:20 -0800 Subject: bpf: verifier: remove dead code Instead of overwriting dead code with jmp -1 instructions remove it completely for root. Adjust verifier state and line info appropriately. v2: - adjust func_info (Alexei); - make sure first instruction retains line info (Alexei). v4: (Yonghong) - remove unnecessary if (!insn to remove) checks; - always keep last line info if first live instruction lacks one. v5: (Martin Lau) - improve and clarify comments. Signed-off-by: Jakub Kicinski Acked-by: Yonghong Song Signed-off-by: Alexei Starovoitov --- include/linux/filter.h | 1 + kernel/bpf/core.c | 12 ++++ kernel/bpf/verifier.c | 176 ++++++++++++++++++++++++++++++++++++++++++++++++- 3 files changed, 186 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/filter.h b/include/linux/filter.h index ad106d845b22..be9af6b4a9e4 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -778,6 +778,7 @@ static inline bool bpf_dump_raw_ok(void) struct bpf_prog *bpf_patch_insn_single(struct bpf_prog *prog, u32 off, const struct bpf_insn *patch, u32 len); +int bpf_remove_insns(struct bpf_prog *prog, u32 off, u32 cnt); void bpf_clear_redirect_map(struct bpf_map *map); diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index ad08ba341197..2a81b8af3748 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -462,6 +462,18 @@ struct bpf_prog *bpf_patch_insn_single(struct bpf_prog *prog, u32 off, return prog_adj; } +int bpf_remove_insns(struct bpf_prog *prog, u32 off, u32 cnt) +{ + /* Branch offsets can't overflow when program is shrinking, no need + * to call bpf_adj_branches(..., true) here + */ + memmove(prog->insnsi + off, prog->insnsi + off + cnt, + sizeof(struct bpf_insn) * (prog->len - off - cnt)); + prog->len -= cnt; + + return WARN_ON_ONCE(bpf_adj_branches(prog, off, off + cnt, off, false)); +} + void bpf_prog_kallsyms_del_subprogs(struct bpf_prog *fp) { int i; diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index bf1f98e8beb6..099b2541f87f 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -6432,6 +6432,150 @@ static struct bpf_prog *bpf_patch_insn_data(struct bpf_verifier_env *env, u32 of return new_prog; } +static int adjust_subprog_starts_after_remove(struct bpf_verifier_env *env, + u32 off, u32 cnt) +{ + int i, j; + + /* find first prog starting at or after off (first to remove) */ + for (i = 0; i < env->subprog_cnt; i++) + if (env->subprog_info[i].start >= off) + break; + /* find first prog starting at or after off + cnt (first to stay) */ + for (j = i; j < env->subprog_cnt; j++) + if (env->subprog_info[j].start >= off + cnt) + break; + /* if j doesn't start exactly at off + cnt, we are just removing + * the front of previous prog + */ + if (env->subprog_info[j].start != off + cnt) + j--; + + if (j > i) { + struct bpf_prog_aux *aux = env->prog->aux; + int move; + + /* move fake 'exit' subprog as well */ + move = env->subprog_cnt + 1 - j; + + memmove(env->subprog_info + i, + env->subprog_info + j, + sizeof(*env->subprog_info) * move); + env->subprog_cnt -= j - i; + + /* remove func_info */ + if (aux->func_info) { + move = aux->func_info_cnt - j; + + memmove(aux->func_info + i, + aux->func_info + j, + sizeof(*aux->func_info) * move); + aux->func_info_cnt -= j - i; + /* func_info->insn_off is set after all code rewrites, + * in adjust_btf_func() - no need to adjust + */ + } + } else { + /* convert i from "first prog to remove" to "first to adjust" */ + if (env->subprog_info[i].start == off) + i++; + } + + /* update fake 'exit' subprog as well */ + for (; i <= env->subprog_cnt; i++) + env->subprog_info[i].start -= cnt; + + return 0; +} + +static int bpf_adj_linfo_after_remove(struct bpf_verifier_env *env, u32 off, + u32 cnt) +{ + struct bpf_prog *prog = env->prog; + u32 i, l_off, l_cnt, nr_linfo; + struct bpf_line_info *linfo; + + nr_linfo = prog->aux->nr_linfo; + if (!nr_linfo) + return 0; + + linfo = prog->aux->linfo; + + /* find first line info to remove, count lines to be removed */ + for (i = 0; i < nr_linfo; i++) + if (linfo[i].insn_off >= off) + break; + + l_off = i; + l_cnt = 0; + for (; i < nr_linfo; i++) + if (linfo[i].insn_off < off + cnt) + l_cnt++; + else + break; + + /* First live insn doesn't match first live linfo, it needs to "inherit" + * last removed linfo. prog is already modified, so prog->len == off + * means no live instructions after (tail of the program was removed). + */ + if (prog->len != off && l_cnt && + (i == nr_linfo || linfo[i].insn_off != off + cnt)) { + l_cnt--; + linfo[--i].insn_off = off + cnt; + } + + /* remove the line info which refer to the removed instructions */ + if (l_cnt) { + memmove(linfo + l_off, linfo + i, + sizeof(*linfo) * (nr_linfo - i)); + + prog->aux->nr_linfo -= l_cnt; + nr_linfo = prog->aux->nr_linfo; + } + + /* pull all linfo[i].insn_off >= off + cnt in by cnt */ + for (i = l_off; i < nr_linfo; i++) + linfo[i].insn_off -= cnt; + + /* fix up all subprogs (incl. 'exit') which start >= off */ + for (i = 0; i <= env->subprog_cnt; i++) + if (env->subprog_info[i].linfo_idx > l_off) { + /* program may have started in the removed region but + * may not be fully removed + */ + if (env->subprog_info[i].linfo_idx >= l_off + l_cnt) + env->subprog_info[i].linfo_idx -= l_cnt; + else + env->subprog_info[i].linfo_idx = l_off; + } + + return 0; +} + +static int verifier_remove_insns(struct bpf_verifier_env *env, u32 off, u32 cnt) +{ + struct bpf_insn_aux_data *aux_data = env->insn_aux_data; + unsigned int orig_prog_len = env->prog->len; + int err; + + err = bpf_remove_insns(env->prog, off, cnt); + if (err) + return err; + + err = adjust_subprog_starts_after_remove(env, off, cnt); + if (err) + return err; + + err = bpf_adj_linfo_after_remove(env, off, cnt); + if (err) + return err; + + memmove(aux_data + off, aux_data + off + cnt, + sizeof(*aux_data) * (orig_prog_len - off - cnt)); + + return 0; +} + /* The verifier does more data flow analysis than llvm and will not * explore branches that are dead at run time. Malicious programs can * have dead code too. Therefore replace all dead at-run-time code @@ -6492,6 +6636,30 @@ static void opt_hard_wire_dead_code_branches(struct bpf_verifier_env *env) } } +static int opt_remove_dead_code(struct bpf_verifier_env *env) +{ + struct bpf_insn_aux_data *aux_data = env->insn_aux_data; + int insn_cnt = env->prog->len; + int i, err; + + for (i = 0; i < insn_cnt; i++) { + int j; + + j = 0; + while (i + j < insn_cnt && !aux_data[i + j].seen) + j++; + if (!j) + continue; + + err = verifier_remove_insns(env, i, j); + if (err) + return err; + insn_cnt = env->prog->len; + } + + return 0; +} + /* convert load instructions that access fields of a context type into a * sequence of instructions that access fields of the underlying structure: * struct __sk_buff -> struct sk_buff @@ -7282,11 +7450,13 @@ skip_full_check: if (is_priv) { if (ret == 0) opt_hard_wire_dead_code_branches(env); + if (ret == 0) + ret = opt_remove_dead_code(env); + } else { + if (ret == 0) + sanitize_dead_code(env); } - if (ret == 0) - sanitize_dead_code(env); - if (ret == 0) /* program is valid, convert *(u32*)(ctx + off) accesses */ ret = convert_ctx_accesses(env); -- cgit v1.2.3 From 9e4c24e7ee7dfd3898269519103e823892b730d8 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 22 Jan 2019 22:45:23 -0800 Subject: bpf: verifier: record original instruction index The communication between the verifier and advanced JITs is based on instruction indexes. We have to keep them stable throughout the optimizations otherwise referring to a particular instruction gets messy quickly. Signed-off-by: Jakub Kicinski Reviewed-by: Quentin Monnet Signed-off-by: Alexei Starovoitov --- include/linux/bpf_verifier.h | 1 + kernel/bpf/verifier.c | 8 +++++--- 2 files changed, 6 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h index 573cca00a0e6..f3ae00ee5516 100644 --- a/include/linux/bpf_verifier.h +++ b/include/linux/bpf_verifier.h @@ -187,6 +187,7 @@ struct bpf_insn_aux_data { int sanitize_stack_off; /* stack slot to be cleared */ bool seen; /* this insn was processed by the verifier */ u8 alu_state; /* used in combination with alu_limit */ + unsigned int orig_idx; /* original instruction index */ }; #define MAX_USED_MAPS 64 /* max number of maps accessed by one eBPF program */ diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index f39bca188a5c..f2c49b4235df 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -7371,7 +7371,7 @@ int bpf_check(struct bpf_prog **prog, union bpf_attr *attr, { struct bpf_verifier_env *env; struct bpf_verifier_log *log; - int ret = -EINVAL; + int i, len, ret = -EINVAL; bool is_priv; /* no program is valid */ @@ -7386,12 +7386,14 @@ int bpf_check(struct bpf_prog **prog, union bpf_attr *attr, return -ENOMEM; log = &env->log; + len = (*prog)->len; env->insn_aux_data = - vzalloc(array_size(sizeof(struct bpf_insn_aux_data), - (*prog)->len)); + vzalloc(array_size(sizeof(struct bpf_insn_aux_data), len)); ret = -ENOMEM; if (!env->insn_aux_data) goto err_free_env; + for (i = 0; i < len; i++) + env->insn_aux_data[i].orig_idx = i; env->prog = *prog; env->ops = bpf_verifier_ops[env->prog->type]; -- cgit v1.2.3 From 08ca90afba255d05dc3253caa44056e7aecbe8c5 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 22 Jan 2019 22:45:24 -0800 Subject: bpf: notify offload JITs about optimizations Let offload JITs know when instructions are replaced and optimized out, so they can update their state appropriately. The optimizations are best effort, if JIT returns an error from any callback verifier will stop notifying it as state may now be out of sync, but the verifier continues making progress. Signed-off-by: Jakub Kicinski Reviewed-by: Quentin Monnet Signed-off-by: Alexei Starovoitov --- include/linux/bpf.h | 7 +++++++ include/linux/bpf_verifier.h | 5 +++++ kernel/bpf/offload.c | 35 +++++++++++++++++++++++++++++++++++ kernel/bpf/verifier.c | 6 ++++++ 4 files changed, 53 insertions(+) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index e734f163bd0b..3851529062ec 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -268,9 +268,15 @@ struct bpf_verifier_ops { }; struct bpf_prog_offload_ops { + /* verifier basic callbacks */ int (*insn_hook)(struct bpf_verifier_env *env, int insn_idx, int prev_insn_idx); int (*finalize)(struct bpf_verifier_env *env); + /* verifier optimization callbacks (called after .finalize) */ + int (*replace_insn)(struct bpf_verifier_env *env, u32 off, + struct bpf_insn *insn); + int (*remove_insns)(struct bpf_verifier_env *env, u32 off, u32 cnt); + /* program management callbacks */ int (*prepare)(struct bpf_prog *prog); int (*translate)(struct bpf_prog *prog); void (*destroy)(struct bpf_prog *prog); @@ -283,6 +289,7 @@ struct bpf_prog_offload { void *dev_priv; struct list_head offloads; bool dev_state; + bool opt_failed; void *jited_image; u32 jited_len; }; diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h index f3ae00ee5516..0620e418dde5 100644 --- a/include/linux/bpf_verifier.h +++ b/include/linux/bpf_verifier.h @@ -266,5 +266,10 @@ int bpf_prog_offload_verifier_prep(struct bpf_prog *prog); int bpf_prog_offload_verify_insn(struct bpf_verifier_env *env, int insn_idx, int prev_insn_idx); int bpf_prog_offload_finalize(struct bpf_verifier_env *env); +void +bpf_prog_offload_replace_insn(struct bpf_verifier_env *env, u32 off, + struct bpf_insn *insn); +void +bpf_prog_offload_remove_insns(struct bpf_verifier_env *env, u32 off, u32 cnt); #endif /* _LINUX_BPF_VERIFIER_H */ diff --git a/kernel/bpf/offload.c b/kernel/bpf/offload.c index 54cf2b9c44a4..39dba8c90331 100644 --- a/kernel/bpf/offload.c +++ b/kernel/bpf/offload.c @@ -173,6 +173,41 @@ int bpf_prog_offload_finalize(struct bpf_verifier_env *env) return ret; } +void +bpf_prog_offload_replace_insn(struct bpf_verifier_env *env, u32 off, + struct bpf_insn *insn) +{ + const struct bpf_prog_offload_ops *ops; + struct bpf_prog_offload *offload; + int ret = -EOPNOTSUPP; + + down_read(&bpf_devs_lock); + offload = env->prog->aux->offload; + if (offload) { + ops = offload->offdev->ops; + if (!offload->opt_failed && ops->replace_insn) + ret = ops->replace_insn(env, off, insn); + offload->opt_failed |= ret; + } + up_read(&bpf_devs_lock); +} + +void +bpf_prog_offload_remove_insns(struct bpf_verifier_env *env, u32 off, u32 cnt) +{ + struct bpf_prog_offload *offload; + int ret = -EOPNOTSUPP; + + down_read(&bpf_devs_lock); + offload = env->prog->aux->offload; + if (offload) { + if (!offload->opt_failed && offload->offdev->ops->remove_insns) + ret = offload->offdev->ops->remove_insns(env, off, cnt); + offload->opt_failed |= ret; + } + up_read(&bpf_devs_lock); +} + static void __bpf_prog_offload_destroy(struct bpf_prog *prog) { struct bpf_prog_offload *offload = prog->aux->offload; diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index f2c49b4235df..8cfe39ef770f 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -6558,6 +6558,9 @@ static int verifier_remove_insns(struct bpf_verifier_env *env, u32 off, u32 cnt) unsigned int orig_prog_len = env->prog->len; int err; + if (bpf_prog_is_dev_bound(env->prog->aux)) + bpf_prog_offload_remove_insns(env, off, cnt); + err = bpf_remove_insns(env->prog, off, cnt); if (err) return err; @@ -6632,6 +6635,9 @@ static void opt_hard_wire_dead_code_branches(struct bpf_verifier_env *env) else continue; + if (bpf_prog_is_dev_bound(env->prog->aux)) + bpf_prog_offload_replace_insn(env, i, &ja); + memcpy(insn, &ja, sizeof(ja)); } } -- cgit v1.2.3 From 8367de2c99a13d35960a51d6084631c883e93a4d Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Thu, 24 Jan 2019 18:20:14 +0900 Subject: block: Fix comment typo Fix typo in REQ_OP_ZONE_RESET description. Signed-off-by: Damien Le Moal Signed-off-by: Jens Axboe --- include/linux/blk_types.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index 5c7e7f859a24..d66bf5f32610 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -287,7 +287,7 @@ enum req_opf { REQ_OP_DISCARD = 3, /* securely erase sectors */ REQ_OP_SECURE_ERASE = 5, - /* seset a zone write pointer */ + /* reset a zone write pointer */ REQ_OP_ZONE_RESET = 6, /* write the same sector many times */ REQ_OP_WRITE_SAME = 7, -- cgit v1.2.3 From 434a4315b9617bf1742bc64712bf44a208502f7f Mon Sep 17 00:00:00 2001 From: Heiner Kallweit Date: Wed, 23 Jan 2019 07:31:58 +0100 Subject: net: phy: change phy_start_interrupts to phy_request_interrupt Now that we enable the interrupts in phy_start() we don't have to do it before. Therefore remove enabling interrupts from phy_start_interrupts() and rename this function to reflect the changed functionality. v2: - improve warning to clearly state that we fall back to polling Signed-off-by: Heiner Kallweit Signed-off-by: David S. Miller --- drivers/net/phy/phy.c | 23 +++++++++++------------ drivers/net/phy/phy_device.c | 4 ++-- drivers/net/phy/phylink.c | 4 ++-- include/linux/phy.h | 2 +- 4 files changed, 16 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index 079b6a617fc8..d12aa512b7f5 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -785,28 +785,27 @@ static int phy_enable_interrupts(struct phy_device *phydev) } /** - * phy_start_interrupts - request and enable interrupts for a PHY device + * phy_request_interrupt - request interrupt for a PHY device * @phydev: target phy_device struct * * Description: Request the interrupt for the given PHY. * If this fails, then we set irq to PHY_POLL. - * Otherwise, we enable the interrupts in the PHY. * This should only be called with a valid IRQ number. - * Returns 0 on success or < 0 on error. */ -int phy_start_interrupts(struct phy_device *phydev) +void phy_request_interrupt(struct phy_device *phydev) { - if (request_threaded_irq(phydev->irq, NULL, phy_interrupt, - IRQF_ONESHOT | IRQF_SHARED, - phydev_name(phydev), phydev) < 0) { - phydev_warn(phydev, "Can't get IRQ %d\n", phydev->irq); + int err; + + err = request_threaded_irq(phydev->irq, NULL, phy_interrupt, + IRQF_ONESHOT | IRQF_SHARED, + phydev_name(phydev), phydev); + if (err) { + phydev_warn(phydev, "Error %d requesting IRQ %d, falling back to polling\n", + err, phydev->irq); phydev->irq = PHY_POLL; - return 0; } - - return phy_enable_interrupts(phydev); } -EXPORT_SYMBOL(phy_start_interrupts); +EXPORT_SYMBOL(phy_request_interrupt); /** * phy_stop - Bring down the PHY link, and stop checking the status diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 64c25a0684ac..891e0178b97f 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -951,8 +951,8 @@ int phy_connect_direct(struct net_device *dev, struct phy_device *phydev, return rc; phy_prepare_link(phydev, handler); - if (phydev->irq > 0) - phy_start_interrupts(phydev); + if (phy_interrupt_is_valid(phydev)) + phy_request_interrupt(phydev); return 0; } diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c index c1b6e05ba60c..2e21ce42e388 100644 --- a/drivers/net/phy/phylink.c +++ b/drivers/net/phy/phylink.c @@ -676,8 +676,8 @@ static int phylink_bringup_phy(struct phylink *pl, struct phy_device *phy) __ETHTOOL_LINK_MODE_MASK_NBITS, pl->supported, __ETHTOOL_LINK_MODE_MASK_NBITS, phy->advertising); - if (phy->irq > 0) - phy_start_interrupts(phy); + if (phy_interrupt_is_valid(phy)) + phy_request_interrupt(phy); return 0; } diff --git a/include/linux/phy.h b/include/linux/phy.h index 1f3873a2ff29..70f83d0d7469 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -1047,7 +1047,7 @@ void phy_ethtool_ksettings_get(struct phy_device *phydev, int phy_ethtool_ksettings_set(struct phy_device *phydev, const struct ethtool_link_ksettings *cmd); int phy_mii_ioctl(struct phy_device *phydev, struct ifreq *ifr, int cmd); -int phy_start_interrupts(struct phy_device *phydev); +void phy_request_interrupt(struct phy_device *phydev); void phy_print_status(struct phy_device *phydev); int phy_set_max_speed(struct phy_device *phydev, u32 max_speed); void phy_remove_link_mode(struct phy_device *phydev, u32 link_mode); -- cgit v1.2.3 From 3b707c3008cad04604c1f50e39f456621821c414 Mon Sep 17 00:00:00 2001 From: Maciej Żenczykowski Date: Thu, 24 Jan 2019 03:07:02 -0800 Subject: net: dev_is_mac_header_xmit() true for ARPHRD_RAWIP MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit __bpf_redirect() and act_mirred checks this boolean to determine whether to prefix an ethernet header. Signed-off-by: Maciej Żenczykowski Acked-by: Daniel Borkmann Signed-off-by: David S. Miller --- include/linux/if_arp.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/if_arp.h b/include/linux/if_arp.h index 6756fea18b69..e44746de95cd 100644 --- a/include/linux/if_arp.h +++ b/include/linux/if_arp.h @@ -54,6 +54,7 @@ static inline bool dev_is_mac_header_xmit(const struct net_device *dev) case ARPHRD_IPGRE: case ARPHRD_VOID: case ARPHRD_NONE: + case ARPHRD_RAWIP: return false; default: return true; -- cgit v1.2.3 From abfd04f738c2625f63e04c8fc7cadb3b7a70d580 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 25 Jan 2019 15:32:28 -0800 Subject: qed: Revert error handling changes. This is new code and not bug fixes. This reverts all changes added by merge commit 8fb18be93efd7292d6ee403b9f61af1008239639 Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed.h | 5 +- drivers/net/ethernet/qlogic/qed/qed_dev.c | 158 ++++++------- drivers/net/ethernet/qlogic/qed/qed_dev_api.h | 12 - drivers/net/ethernet/qlogic/qed/qed_hsi.h | 2 +- drivers/net/ethernet/qlogic/qed/qed_hw.c | 11 - drivers/net/ethernet/qlogic/qed/qed_int.c | 126 +++++------ drivers/net/ethernet/qlogic/qed/qed_int.h | 3 - drivers/net/ethernet/qlogic/qed/qed_main.c | 30 --- drivers/net/ethernet/qlogic/qed/qed_mcp.c | 115 ---------- drivers/net/ethernet/qlogic/qed/qed_mcp.h | 42 ---- drivers/net/ethernet/qlogic/qed/qed_reg_addr.h | 2 - drivers/net/ethernet/qlogic/qed/qed_spq.c | 22 -- drivers/net/ethernet/qlogic/qed/qed_sriov.c | 9 +- drivers/net/ethernet/qlogic/qede/qede.h | 3 - drivers/net/ethernet/qlogic/qede/qede_main.c | 300 +++++-------------------- drivers/net/ethernet/qlogic/qede/qede_rdma.c | 64 ++---- include/linux/qed/qed_if.h | 20 -- include/linux/qed/qede_rdma.h | 21 +- 18 files changed, 202 insertions(+), 743 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h index 3b0955d34716..24a90163775e 100644 --- a/drivers/net/ethernet/qlogic/qed/qed.h +++ b/drivers/net/ethernet/qlogic/qed/qed.h @@ -554,6 +554,7 @@ struct qed_hwfn { u8 dp_level; char name[NAME_SIZE]; + bool first_on_engine; bool hw_init_done; u8 num_funcs_on_engine; @@ -804,9 +805,6 @@ struct qed_dev { u32 mcp_nvm_resp; - /* Recovery */ - bool recov_in_prog; - /* Linux specific here */ struct qede_dev *edev; struct pci_dev *pdev; @@ -946,7 +944,6 @@ void qed_link_update(struct qed_hwfn *hwfn, struct qed_ptt *ptt); u32 qed_unzip_data(struct qed_hwfn *p_hwfn, u32 input_len, u8 *input_buf, u32 max_size, u8 *unzip_buf); -void qed_schedule_recovery_handler(struct qed_hwfn *p_hwfn); void qed_get_protocol_stats(struct qed_dev *cdev, enum qed_mcp_protocol_type type, union qed_mcp_protocol_stats *stats); diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index b17003d9066c..8f6551421945 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -1959,6 +1959,11 @@ static int qed_hw_init_pf(struct qed_hwfn *p_hwfn, (p_hwfn->hw_info.personality == QED_PCI_FCOE) ? 1 : 0); STORE_RT_REG(p_hwfn, PRS_REG_SEARCH_ROCE_RT_OFFSET, 0); + /* Cleanup chip from previous driver if such remains exist */ + rc = qed_final_cleanup(p_hwfn, p_ptt, rel_pf_id, false); + if (rc) + return rc; + /* Sanity check before the PF init sequence that uses DMAE */ rc = qed_dmae_sanity(p_hwfn, p_ptt, "pf_phase"); if (rc) @@ -2002,15 +2007,17 @@ static int qed_hw_init_pf(struct qed_hwfn *p_hwfn, return rc; } -int qed_pglueb_set_pfid_enable(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, bool b_enable) +static int qed_change_pci_hwfn(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + u8 enable) { - u32 delay_idx = 0, val, set_val = b_enable ? 1 : 0; + u32 delay_idx = 0, val, set_val = enable ? 1 : 0; - /* Configure the PF's internal FID_enable for master transactions */ - qed_wr(p_hwfn, p_ptt, PGLUE_B_REG_INTERNAL_PFID_ENABLE_MASTER, set_val); + /* Change PF in PXP */ + qed_wr(p_hwfn, p_ptt, + PGLUE_B_REG_INTERNAL_PFID_ENABLE_MASTER, set_val); - /* Wait until value is set - try for 1 second every 50us */ + /* wait until value is set - try for 1 second every 50us */ for (delay_idx = 0; delay_idx < 20000; delay_idx++) { val = qed_rd(p_hwfn, p_ptt, PGLUE_B_REG_INTERNAL_PFID_ENABLE_MASTER); @@ -2064,19 +2071,13 @@ static int qed_vf_start(struct qed_hwfn *p_hwfn, return 0; } -static void qed_pglueb_clear_err(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) -{ - qed_wr(p_hwfn, p_ptt, PGLUE_B_REG_WAS_ERROR_PF_31_0_CLR, - BIT(p_hwfn->abs_pf_id)); -} - int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params) { struct qed_load_req_params load_req_params; u32 load_code, resp, param, drv_mb_param; bool b_default_mtu = true; struct qed_hwfn *p_hwfn; - int rc = 0, i; + int rc = 0, mfw_rc, i; u16 ether_type; if ((p_params->int_mode == QED_INT_MODE_MSI) && (cdev->num_hwfns > 1)) { @@ -2091,7 +2092,7 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params) } for_each_hwfn(cdev, i) { - p_hwfn = &cdev->hwfns[i]; + struct qed_hwfn *p_hwfn = &cdev->hwfns[i]; /* If management didn't provide a default, set one of our own */ if (!p_hwfn->hw_info.mtu) { @@ -2104,6 +2105,9 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params) continue; } + /* Enable DMAE in PXP */ + rc = qed_change_pci_hwfn(p_hwfn, p_hwfn->p_main_ptt, true); + rc = qed_calc_hw_mode(p_hwfn); if (rc) return rc; @@ -2140,43 +2144,12 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params) "Load request was sent. Load code: 0x%x\n", load_code); - /* Only relevant for recovery: - * Clear the indication after LOAD_REQ is responded by the MFW. - */ - cdev->recov_in_prog = false; - qed_mcp_set_capabilities(p_hwfn, p_hwfn->p_main_ptt); qed_reset_mb_shadow(p_hwfn, p_hwfn->p_main_ptt); - /* Clean up chip from previous driver if such remains exist. - * This is not needed when the PF is the first one on the - * engine, since afterwards we are going to init the FW. - */ - if (load_code != FW_MSG_CODE_DRV_LOAD_ENGINE) { - rc = qed_final_cleanup(p_hwfn, p_hwfn->p_main_ptt, - p_hwfn->rel_pf_id, false); - if (rc) { - DP_NOTICE(p_hwfn, "Final cleanup failed\n"); - goto load_err; - } - } - - /* Log and clear previous pglue_b errors if such exist */ - qed_pglueb_rbc_attn_handler(p_hwfn, p_hwfn->p_main_ptt); - - /* Enable the PF's internal FID_enable in the PXP */ - rc = qed_pglueb_set_pfid_enable(p_hwfn, p_hwfn->p_main_ptt, - true); - if (rc) - goto load_err; - - /* Clear the pglue_b was_error indication. - * In E4 it must be done after the BME and the internal - * FID_enable for the PF are set, since VDMs may cause the - * indication to be set again. - */ - qed_pglueb_clear_err(p_hwfn, p_hwfn->p_main_ptt); + p_hwfn->first_on_engine = (load_code == + FW_MSG_CODE_DRV_LOAD_ENGINE); switch (load_code) { case FW_MSG_CODE_DRV_LOAD_ENGINE: @@ -2207,29 +2180,39 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params) break; } - if (rc) { + if (rc) DP_NOTICE(p_hwfn, "init phase failed for loadcode 0x%x (rc %d)\n", - load_code, rc); - goto load_err; - } + load_code, rc); - rc = qed_mcp_load_done(p_hwfn, p_hwfn->p_main_ptt); + /* ACK mfw regardless of success or failure of initialization */ + mfw_rc = qed_mcp_cmd(p_hwfn, p_hwfn->p_main_ptt, + DRV_MSG_CODE_LOAD_DONE, + 0, &load_code, ¶m); if (rc) return rc; + if (mfw_rc) { + DP_NOTICE(p_hwfn, "Failed sending LOAD_DONE command\n"); + return mfw_rc; + } + + /* Check if there is a DID mismatch between nvm-cfg/efuse */ + if (param & FW_MB_PARAM_LOAD_DONE_DID_EFUSE_ERROR) + DP_NOTICE(p_hwfn, + "warning: device configuration is not supported on this board type. The device may not function as expected.\n"); /* send DCBX attention request command */ DP_VERBOSE(p_hwfn, QED_MSG_DCB, "sending phony dcbx set command to trigger DCBx attention handling\n"); - rc = qed_mcp_cmd(p_hwfn, p_hwfn->p_main_ptt, - DRV_MSG_CODE_SET_DCBX, - 1 << DRV_MB_PARAM_DCBX_NOTIFY_SHIFT, - &resp, ¶m); - if (rc) { + mfw_rc = qed_mcp_cmd(p_hwfn, p_hwfn->p_main_ptt, + DRV_MSG_CODE_SET_DCBX, + 1 << DRV_MB_PARAM_DCBX_NOTIFY_SHIFT, + &load_code, ¶m); + if (mfw_rc) { DP_NOTICE(p_hwfn, "Failed to send DCBX attention request\n"); - return rc; + return mfw_rc; } p_hwfn->hw_init_done = true; @@ -2278,12 +2261,6 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params) } return 0; - -load_err: - /* The MFW load lock should be released also when initialization fails. - */ - qed_mcp_load_done(p_hwfn, p_hwfn->p_main_ptt); - return rc; } #define QED_HW_STOP_RETRY_LIMIT (10) @@ -2296,9 +2273,6 @@ static void qed_hw_timers_stop(struct qed_dev *cdev, qed_wr(p_hwfn, p_ptt, TM_REG_PF_ENABLE_CONN, 0x0); qed_wr(p_hwfn, p_ptt, TM_REG_PF_ENABLE_TASK, 0x0); - if (cdev->recov_in_prog) - return; - for (i = 0; i < QED_HW_STOP_RETRY_LIMIT; i++) { if ((!qed_rd(p_hwfn, p_ptt, TM_REG_PF_SCAN_ACTIVE_CONN)) && @@ -2361,14 +2335,12 @@ int qed_hw_stop(struct qed_dev *cdev) p_hwfn->hw_init_done = false; /* Send unload command to MCP */ - if (!cdev->recov_in_prog) { - rc = qed_mcp_unload_req(p_hwfn, p_ptt); - if (rc) { - DP_NOTICE(p_hwfn, - "Failed sending a UNLOAD_REQ command. rc = %d.\n", - rc); - rc2 = -EINVAL; - } + rc = qed_mcp_unload_req(p_hwfn, p_ptt); + if (rc) { + DP_NOTICE(p_hwfn, + "Failed sending a UNLOAD_REQ command. rc = %d.\n", + rc); + rc2 = -EINVAL; } qed_slowpath_irq_sync(p_hwfn); @@ -2410,31 +2382,27 @@ int qed_hw_stop(struct qed_dev *cdev) qed_wr(p_hwfn, p_ptt, DORQ_REG_PF_DB_ENABLE, 0); qed_wr(p_hwfn, p_ptt, QM_REG_PF_EN, 0); - if (!cdev->recov_in_prog) { - rc = qed_mcp_unload_done(p_hwfn, p_ptt); - if (rc) { - DP_NOTICE(p_hwfn, - "Failed sending a UNLOAD_DONE command. rc = %d.\n", - rc); - rc2 = -EINVAL; - } + qed_mcp_unload_done(p_hwfn, p_ptt); + if (rc) { + DP_NOTICE(p_hwfn, + "Failed sending a UNLOAD_DONE command. rc = %d.\n", + rc); + rc2 = -EINVAL; } } - if (IS_PF(cdev) && !cdev->recov_in_prog) { + if (IS_PF(cdev)) { p_hwfn = QED_LEADING_HWFN(cdev); p_ptt = QED_LEADING_HWFN(cdev)->p_main_ptt; - /* Clear the PF's internal FID_enable in the PXP. - * In CMT this should only be done for first hw-function, and - * only after all transactions have stopped for all active - * hw-functions. + /* Disable DMAE in PXP - in CMT, this should only be done for + * first hw-function, and only after all transactions have + * stopped for all active hw-functions. */ - rc = qed_pglueb_set_pfid_enable(p_hwfn, p_ptt, false); + rc = qed_change_pci_hwfn(p_hwfn, p_ptt, false); if (rc) { DP_NOTICE(p_hwfn, - "qed_pglueb_set_pfid_enable() failed. rc = %d.\n", - rc); + "qed_change_pci_hwfn failed. rc = %d.\n", rc); rc2 = -EINVAL; } } @@ -2534,8 +2502,9 @@ static void qed_hw_hwfn_prepare(struct qed_hwfn *p_hwfn) PGLUE_B_REG_PGL_ADDR_94_F0_BB, 0); } - /* Clean previous pglue_b errors if such exist */ - qed_pglueb_clear_err(p_hwfn, p_hwfn->p_main_ptt); + /* Clean Previous errors if such exist */ + qed_wr(p_hwfn, p_hwfn->p_main_ptt, + PGLUE_B_REG_WAS_ERROR_PF_31_0_CLR, 1 << p_hwfn->abs_pf_id); /* enable internal target-read */ qed_wr(p_hwfn, p_hwfn->p_main_ptt, @@ -3471,7 +3440,6 @@ static int qed_hw_prepare_single(struct qed_hwfn *p_hwfn, void __iomem *p_doorbells, enum qed_pci_personality personality) { - struct qed_dev *cdev = p_hwfn->cdev; int rc = 0; /* Split PCI bars evenly between hwfns */ @@ -3524,7 +3492,7 @@ static int qed_hw_prepare_single(struct qed_hwfn *p_hwfn, /* Sending a mailbox to the MFW should be done after qed_get_hw_info() * is called as it sets the ports number in an engine. */ - if (IS_LEAD_HWFN(p_hwfn) && !cdev->recov_in_prog) { + if (IS_LEAD_HWFN(p_hwfn)) { rc = qed_mcp_initiate_pf_flr(p_hwfn, p_hwfn->p_main_ptt); if (rc) DP_NOTICE(p_hwfn, "Failed to initiate PF FLR\n"); diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h index e4b4e3b78e8a..acccd85170aa 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h +++ b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h @@ -472,18 +472,6 @@ int qed_get_queue_coalesce(struct qed_hwfn *p_hwfn, u16 *coal, void *handle); int qed_set_queue_coalesce(u16 rx_coal, u16 tx_coal, void *p_handle); -/** - * @brief qed_pglueb_set_pfid_enable - Enable or disable PCI BUS MASTER - * - * @param p_hwfn - * @param p_ptt - * @param b_enable - true/false - * - * @return int - */ -int qed_pglueb_set_pfid_enable(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt, bool b_enable); - /** * @brief db_recovery_add - add doorbell information to the doorbell * recovery mechanism. diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h index 417121e74ee9..b13cfb449d8f 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h @@ -12827,7 +12827,7 @@ enum MFW_DRV_MSG_TYPE { MFW_DRV_MSG_LLDP_DATA_UPDATED, MFW_DRV_MSG_DCBX_REMOTE_MIB_UPDATED, MFW_DRV_MSG_DCBX_OPERATIONAL_MIB_UPDATED, - MFW_DRV_MSG_ERROR_RECOVERY, + MFW_DRV_MSG_RESERVED4, MFW_DRV_MSG_BW_UPDATE, MFW_DRV_MSG_S_TAG_UPDATE, MFW_DRV_MSG_GET_LAN_STATS, diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.c b/drivers/net/ethernet/qlogic/qed/qed_hw.c index 72ec1c6bdf70..70504dcf4087 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hw.c +++ b/drivers/net/ethernet/qlogic/qed/qed_hw.c @@ -703,17 +703,6 @@ static int qed_dmae_execute_command(struct qed_hwfn *p_hwfn, int qed_status = 0; u32 offset = 0; - if (p_hwfn->cdev->recov_in_prog) { - DP_VERBOSE(p_hwfn, - NETIF_MSG_HW, - "Recovery is in progress. Avoid DMAE transaction [{src: addr 0x%llx, type %d}, {dst: addr 0x%llx, type %d}, size %d].\n", - src_addr, src_type, dst_addr, dst_type, - size_in_dwords); - - /* Let the flow complete w/o any error handling */ - return 0; - } - qed_dmae_opcode(p_hwfn, (src_type == QED_DMAE_ADDRESS_GRC), (dst_type == QED_DMAE_ADDRESS_GRC), diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c index e23980e301b6..92340919d852 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_int.c +++ b/drivers/net/ethernet/qlogic/qed/qed_int.c @@ -255,114 +255,112 @@ out: #define PGLUE_ATTENTION_ICPL_VALID (1 << 23) #define PGLUE_ATTENTION_ZLR_VALID (1 << 25) #define PGLUE_ATTENTION_ILT_VALID (1 << 23) - -int qed_pglueb_rbc_attn_handler(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) +static int qed_pglub_rbc_attn_cb(struct qed_hwfn *p_hwfn) { u32 tmp; - tmp = qed_rd(p_hwfn, p_ptt, PGLUE_B_REG_TX_ERR_WR_DETAILS2); + tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, + PGLUE_B_REG_TX_ERR_WR_DETAILS2); if (tmp & PGLUE_ATTENTION_VALID) { u32 addr_lo, addr_hi, details; - addr_lo = qed_rd(p_hwfn, p_ptt, + addr_lo = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, PGLUE_B_REG_TX_ERR_WR_ADD_31_0); - addr_hi = qed_rd(p_hwfn, p_ptt, + addr_hi = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, PGLUE_B_REG_TX_ERR_WR_ADD_63_32); - details = qed_rd(p_hwfn, p_ptt, + details = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, PGLUE_B_REG_TX_ERR_WR_DETAILS); - DP_NOTICE(p_hwfn, - "Illegal write by chip to [%08x:%08x] blocked.\n" - "Details: %08x [PFID %02x, VFID %02x, VF_VALID %02x]\n" - "Details2 %08x [Was_error %02x BME deassert %02x FID_enable deassert %02x]\n", - addr_hi, addr_lo, details, - (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_PFID), - (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_VFID), - GET_FIELD(details, - PGLUE_ATTENTION_DETAILS_VF_VALID) ? 1 : 0, - tmp, - GET_FIELD(tmp, - PGLUE_ATTENTION_DETAILS2_WAS_ERR) ? 1 : 0, - GET_FIELD(tmp, - PGLUE_ATTENTION_DETAILS2_BME) ? 1 : 0, - GET_FIELD(tmp, - PGLUE_ATTENTION_DETAILS2_FID_EN) ? 1 : 0); + DP_INFO(p_hwfn, + "Illegal write by chip to [%08x:%08x] blocked.\n" + "Details: %08x [PFID %02x, VFID %02x, VF_VALID %02x]\n" + "Details2 %08x [Was_error %02x BME deassert %02x FID_enable deassert %02x]\n", + addr_hi, addr_lo, details, + (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_PFID), + (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_VFID), + GET_FIELD(details, + PGLUE_ATTENTION_DETAILS_VF_VALID) ? 1 : 0, + tmp, + GET_FIELD(tmp, + PGLUE_ATTENTION_DETAILS2_WAS_ERR) ? 1 : 0, + GET_FIELD(tmp, + PGLUE_ATTENTION_DETAILS2_BME) ? 1 : 0, + GET_FIELD(tmp, + PGLUE_ATTENTION_DETAILS2_FID_EN) ? 1 : 0); } - tmp = qed_rd(p_hwfn, p_ptt, PGLUE_B_REG_TX_ERR_RD_DETAILS2); + tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, + PGLUE_B_REG_TX_ERR_RD_DETAILS2); if (tmp & PGLUE_ATTENTION_RD_VALID) { u32 addr_lo, addr_hi, details; - addr_lo = qed_rd(p_hwfn, p_ptt, + addr_lo = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, PGLUE_B_REG_TX_ERR_RD_ADD_31_0); - addr_hi = qed_rd(p_hwfn, p_ptt, + addr_hi = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, PGLUE_B_REG_TX_ERR_RD_ADD_63_32); - details = qed_rd(p_hwfn, p_ptt, + details = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, PGLUE_B_REG_TX_ERR_RD_DETAILS); - DP_NOTICE(p_hwfn, - "Illegal read by chip from [%08x:%08x] blocked.\n" - "Details: %08x [PFID %02x, VFID %02x, VF_VALID %02x]\n" - "Details2 %08x [Was_error %02x BME deassert %02x FID_enable deassert %02x]\n", - addr_hi, addr_lo, details, - (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_PFID), - (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_VFID), - GET_FIELD(details, - PGLUE_ATTENTION_DETAILS_VF_VALID) ? 1 : 0, - tmp, - GET_FIELD(tmp, - PGLUE_ATTENTION_DETAILS2_WAS_ERR) ? 1 : 0, - GET_FIELD(tmp, - PGLUE_ATTENTION_DETAILS2_BME) ? 1 : 0, - GET_FIELD(tmp, - PGLUE_ATTENTION_DETAILS2_FID_EN) ? 1 : 0); + DP_INFO(p_hwfn, + "Illegal read by chip from [%08x:%08x] blocked.\n" + " Details: %08x [PFID %02x, VFID %02x, VF_VALID %02x]\n" + " Details2 %08x [Was_error %02x BME deassert %02x FID_enable deassert %02x]\n", + addr_hi, addr_lo, details, + (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_PFID), + (u8)GET_FIELD(details, PGLUE_ATTENTION_DETAILS_VFID), + GET_FIELD(details, + PGLUE_ATTENTION_DETAILS_VF_VALID) ? 1 : 0, + tmp, + GET_FIELD(tmp, PGLUE_ATTENTION_DETAILS2_WAS_ERR) ? 1 + : 0, + GET_FIELD(tmp, PGLUE_ATTENTION_DETAILS2_BME) ? 1 : 0, + GET_FIELD(tmp, PGLUE_ATTENTION_DETAILS2_FID_EN) ? 1 + : 0); } - tmp = qed_rd(p_hwfn, p_ptt, PGLUE_B_REG_TX_ERR_WR_DETAILS_ICPL); + tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, + PGLUE_B_REG_TX_ERR_WR_DETAILS_ICPL); if (tmp & PGLUE_ATTENTION_ICPL_VALID) - DP_NOTICE(p_hwfn, "ICPL error - %08x\n", tmp); + DP_INFO(p_hwfn, "ICPL error - %08x\n", tmp); - tmp = qed_rd(p_hwfn, p_ptt, PGLUE_B_REG_MASTER_ZLR_ERR_DETAILS); + tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, + PGLUE_B_REG_MASTER_ZLR_ERR_DETAILS); if (tmp & PGLUE_ATTENTION_ZLR_VALID) { u32 addr_hi, addr_lo; - addr_lo = qed_rd(p_hwfn, p_ptt, + addr_lo = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, PGLUE_B_REG_MASTER_ZLR_ERR_ADD_31_0); - addr_hi = qed_rd(p_hwfn, p_ptt, + addr_hi = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, PGLUE_B_REG_MASTER_ZLR_ERR_ADD_63_32); - DP_NOTICE(p_hwfn, "ZLR error - %08x [Address %08x:%08x]\n", - tmp, addr_hi, addr_lo); + DP_INFO(p_hwfn, "ZLR eror - %08x [Address %08x:%08x]\n", + tmp, addr_hi, addr_lo); } - tmp = qed_rd(p_hwfn, p_ptt, PGLUE_B_REG_VF_ILT_ERR_DETAILS2); + tmp = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, + PGLUE_B_REG_VF_ILT_ERR_DETAILS2); if (tmp & PGLUE_ATTENTION_ILT_VALID) { u32 addr_hi, addr_lo, details; - addr_lo = qed_rd(p_hwfn, p_ptt, + addr_lo = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, PGLUE_B_REG_VF_ILT_ERR_ADD_31_0); - addr_hi = qed_rd(p_hwfn, p_ptt, + addr_hi = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, PGLUE_B_REG_VF_ILT_ERR_ADD_63_32); - details = qed_rd(p_hwfn, p_ptt, + details = qed_rd(p_hwfn, p_hwfn->p_dpc_ptt, PGLUE_B_REG_VF_ILT_ERR_DETAILS); - DP_NOTICE(p_hwfn, - "ILT error - Details %08x Details2 %08x [Address %08x:%08x]\n", - details, tmp, addr_hi, addr_lo); + DP_INFO(p_hwfn, + "ILT error - Details %08x Details2 %08x [Address %08x:%08x]\n", + details, tmp, addr_hi, addr_lo); } /* Clear the indications */ - qed_wr(p_hwfn, p_ptt, PGLUE_B_REG_LATCHED_ERRORS_CLR, BIT(2)); + qed_wr(p_hwfn, p_hwfn->p_dpc_ptt, + PGLUE_B_REG_LATCHED_ERRORS_CLR, (1 << 2)); return 0; } -static int qed_pglueb_rbc_attn_cb(struct qed_hwfn *p_hwfn) -{ - return qed_pglueb_rbc_attn_handler(p_hwfn, p_hwfn->p_dpc_ptt); -} - #define QED_DORQ_ATTENTION_REASON_MASK (0xfffff) #define QED_DORQ_ATTENTION_OPAQUE_MASK (0xffff) #define QED_DORQ_ATTENTION_OPAQUE_SHIFT (0x0) @@ -542,7 +540,7 @@ static struct aeu_invert_reg aeu_descs[NUM_ATTN_REGS] = { {"PGLUE misc_flr", ATTENTION_SINGLE, NULL, MAX_BLOCK_ID}, {"PGLUE B RBC", ATTENTION_PAR_INT, - qed_pglueb_rbc_attn_cb, BLOCK_PGLUE_B}, + qed_pglub_rbc_attn_cb, BLOCK_PGLUE_B}, {"PGLUE misc_mctp", ATTENTION_SINGLE, NULL, MAX_BLOCK_ID}, {"Flash event", ATTENTION_SINGLE, NULL, MAX_BLOCK_ID}, diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.h b/drivers/net/ethernet/qlogic/qed/qed_int.h index 1f356ed4f761..d81a62ebd524 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_int.h +++ b/drivers/net/ethernet/qlogic/qed/qed_int.h @@ -431,7 +431,4 @@ int qed_int_set_timer_res(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, #define QED_MAPPING_MEMORY_SIZE(dev) (NUM_OF_SBS(dev)) -int qed_pglueb_rbc_attn_handler(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt); - #endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index b47352643fb5..6adf5bda9811 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -359,8 +359,6 @@ static struct qed_dev *qed_probe(struct pci_dev *pdev, qed_init_dp(cdev, params->dp_module, params->dp_level); - cdev->recov_in_prog = params->recov_in_prog; - rc = qed_init_pci(cdev, pdev); if (rc) { DP_ERR(cdev, "init pci failed\n"); @@ -2205,15 +2203,6 @@ static int qed_nvm_get_image(struct qed_dev *cdev, enum qed_nvm_images type, return qed_mcp_get_nvm_image(hwfn, type, buf, len); } -void qed_schedule_recovery_handler(struct qed_hwfn *p_hwfn) -{ - struct qed_common_cb_ops *ops = p_hwfn->cdev->protocol_ops.common; - void *cookie = p_hwfn->cdev->ops_cookie; - - if (ops && ops->schedule_recovery_handler) - ops->schedule_recovery_handler(cookie); -} - static int qed_set_coalesce(struct qed_dev *cdev, u16 rx_coal, u16 tx_coal, void *handle) { @@ -2237,23 +2226,6 @@ static int qed_set_led(struct qed_dev *cdev, enum qed_led_mode mode) return status; } -static int qed_recovery_process(struct qed_dev *cdev) -{ - struct qed_hwfn *p_hwfn = QED_LEADING_HWFN(cdev); - struct qed_ptt *p_ptt; - int rc = 0; - - p_ptt = qed_ptt_acquire(p_hwfn); - if (!p_ptt) - return -EAGAIN; - - rc = qed_start_recovery_process(p_hwfn, p_ptt); - - qed_ptt_release(p_hwfn, p_ptt); - - return rc; -} - static int qed_update_wol(struct qed_dev *cdev, bool enabled) { struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); @@ -2408,8 +2380,6 @@ const struct qed_common_ops qed_common_ops_pass = { .nvm_get_image = &qed_nvm_get_image, .set_coalesce = &qed_set_coalesce, .set_led = &qed_set_led, - .recovery_process = &qed_recovery_process, - .recovery_prolog = &qed_recovery_prolog, .update_drv_state = &qed_update_drv_state, .update_mac = &qed_update_mac, .update_mtu = &qed_update_mtu, diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c index bb8541847aa5..e7f18e34ff0d 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c @@ -1070,27 +1070,6 @@ int qed_mcp_load_req(struct qed_hwfn *p_hwfn, return 0; } -int qed_mcp_load_done(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) -{ - u32 resp = 0, param = 0; - int rc; - - rc = qed_mcp_cmd(p_hwfn, p_ptt, DRV_MSG_CODE_LOAD_DONE, 0, &resp, - ¶m); - if (rc) { - DP_NOTICE(p_hwfn, - "Failed to send a LOAD_DONE command, rc = %d\n", rc); - return rc; - } - - /* Check if there is a DID mismatch between nvm-cfg/efuse */ - if (param & FW_MB_PARAM_LOAD_DONE_DID_EFUSE_ERROR) - DP_NOTICE(p_hwfn, - "warning: device configuration is not supported on this board type. The device may not function as expected.\n"); - - return 0; -} - int qed_mcp_unload_req(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { struct qed_mcp_mb_params mb_params; @@ -1549,60 +1528,6 @@ int qed_mcp_set_link(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, bool b_up) return 0; } -u32 qed_get_process_kill_counter(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) -{ - u32 path_offsize_addr, path_offsize, path_addr, proc_kill_cnt; - - if (IS_VF(p_hwfn->cdev)) - return -EINVAL; - - path_offsize_addr = SECTION_OFFSIZE_ADDR(p_hwfn->mcp_info->public_base, - PUBLIC_PATH); - path_offsize = qed_rd(p_hwfn, p_ptt, path_offsize_addr); - path_addr = SECTION_ADDR(path_offsize, QED_PATH_ID(p_hwfn)); - - proc_kill_cnt = qed_rd(p_hwfn, p_ptt, - path_addr + - offsetof(struct public_path, process_kill)) & - PROCESS_KILL_COUNTER_MASK; - - return proc_kill_cnt; -} - -static void qed_mcp_handle_process_kill(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt) -{ - struct qed_dev *cdev = p_hwfn->cdev; - u32 proc_kill_cnt; - - /* Prevent possible attentions/interrupts during the recovery handling - * and till its load phase, during which they will be re-enabled. - */ - qed_int_igu_disable_int(p_hwfn, p_ptt); - - DP_NOTICE(p_hwfn, "Received a process kill indication\n"); - - /* The following operations should be done once, and thus in CMT mode - * are carried out by only the first HW function. - */ - if (p_hwfn != QED_LEADING_HWFN(cdev)) - return; - - if (cdev->recov_in_prog) { - DP_NOTICE(p_hwfn, - "Ignoring the indication since a recovery process is already in progress\n"); - return; - } - - cdev->recov_in_prog = true; - - proc_kill_cnt = qed_get_process_kill_counter(p_hwfn, p_ptt); - DP_NOTICE(p_hwfn, "Process kill counter: %d\n", proc_kill_cnt); - - qed_schedule_recovery_handler(p_hwfn); -} - static void qed_mcp_send_protocol_stats(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, enum MFW_DRV_MSG_TYPE type) @@ -1833,9 +1758,6 @@ int qed_mcp_handle_events(struct qed_hwfn *p_hwfn, case MFW_DRV_MSG_TRANSCEIVER_STATE_CHANGE: qed_mcp_handle_transceiver_change(p_hwfn, p_ptt); break; - case MFW_DRV_MSG_ERROR_RECOVERY: - qed_mcp_handle_process_kill(p_hwfn, p_ptt); - break; case MFW_DRV_MSG_GET_LAN_STATS: case MFW_DRV_MSG_GET_FCOE_STATS: case MFW_DRV_MSG_GET_ISCSI_STATS: @@ -2381,43 +2303,6 @@ int qed_mcp_get_flash_size(struct qed_hwfn *p_hwfn, return 0; } -int qed_start_recovery_process(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) -{ - struct qed_dev *cdev = p_hwfn->cdev; - - if (cdev->recov_in_prog) { - DP_NOTICE(p_hwfn, - "Avoid triggering a recovery since such a process is already in progress\n"); - return -EAGAIN; - } - - DP_NOTICE(p_hwfn, "Triggering a recovery process\n"); - qed_wr(p_hwfn, p_ptt, MISC_REG_AEU_GENERAL_ATTN_35, 0x1); - - return 0; -} - -#define QED_RECOVERY_PROLOG_SLEEP_MS 100 - -int qed_recovery_prolog(struct qed_dev *cdev) -{ - struct qed_hwfn *p_hwfn = QED_LEADING_HWFN(cdev); - struct qed_ptt *p_ptt = p_hwfn->p_main_ptt; - int rc; - - /* Allow ongoing PCIe transactions to complete */ - msleep(QED_RECOVERY_PROLOG_SLEEP_MS); - - /* Clear the PF's internal FID_enable in the PXP */ - rc = qed_pglueb_set_pfid_enable(p_hwfn, p_ptt, false); - if (rc) - DP_NOTICE(p_hwfn, - "qed_pglueb_set_pfid_enable() failed. rc = %d.\n", - rc); - - return rc; -} - static int qed_mcp_config_vf_msix_bb(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u8 vf_id, u8 num) diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.h b/drivers/net/ethernet/qlogic/qed/qed_mcp.h index 6e1d72a669ae..eddf67798d6f 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.h +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.h @@ -440,38 +440,6 @@ qed_mcp_send_drv_version(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, struct qed_mcp_drv_version *p_ver); -/** - * @brief Read the MFW process kill counter - * - * @param p_hwfn - * @param p_ptt - * - * @return u32 - */ -u32 qed_get_process_kill_counter(struct qed_hwfn *p_hwfn, - struct qed_ptt *p_ptt); - -/** - * @brief Trigger a recovery process - * - * @param p_hwfn - * @param p_ptt - * - * @return int - */ -int qed_start_recovery_process(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt); - -/** - * @brief A recovery handler must call this function as its first step. - * It is assumed that the handler is not run from an interrupt context. - * - * @param cdev - * @param p_ptt - * - * @return int - */ -int qed_recovery_prolog(struct qed_dev *cdev); - /** * @brief Notify MFW about the change in base device properties * @@ -832,16 +800,6 @@ int qed_mcp_load_req(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, struct qed_load_req_params *p_params); -/** - * @brief Sends a LOAD_DONE message to the MFW - * - * @param p_hwfn - * @param p_ptt - * - * @return int - 0 - Operation was successful. - */ -int qed_mcp_load_done(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt); - /** * @brief Sends a UNLOAD_REQ message to the MFW * diff --git a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h index 5ce825ca5f24..8939ed6e08b7 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h +++ b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h @@ -518,8 +518,6 @@ 0x180824UL #define MISC_REG_AEU_GENERAL_ATTN_0 \ 0x008400UL -#define MISC_REG_AEU_GENERAL_ATTN_35 \ - 0x00848cUL #define CAU_REG_SB_ADDR_MEMORY \ 0x1c8000UL #define CAU_REG_SB_VAR_MEMORY \ diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c index 3e0f7c46bb1b..eb88bbc6b193 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_spq.c +++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c @@ -790,17 +790,6 @@ static int qed_spq_pend_post(struct qed_hwfn *p_hwfn) SPQ_HIGH_PRI_RESERVE_DEFAULT); } -static void qed_spq_recov_set_ret_code(struct qed_spq_entry *p_ent, - u8 *fw_return_code) -{ - if (!fw_return_code) - return; - - if (p_ent->elem.hdr.protocol_id == PROTOCOLID_ROCE || - p_ent->elem.hdr.protocol_id == PROTOCOLID_IWARP) - *fw_return_code = RDMA_RETURN_OK; -} - /* Avoid overriding of SPQ entries when getting out-of-order completions, by * marking the completions in a bitmap and increasing the chain consumer only * for the first successive completed entries. @@ -836,17 +825,6 @@ int qed_spq_post(struct qed_hwfn *p_hwfn, return -EINVAL; } - if (p_hwfn->cdev->recov_in_prog) { - DP_VERBOSE(p_hwfn, - QED_MSG_SPQ, - "Recovery is in progress. Skip spq post [cmd %02x protocol %02x]\n", - p_ent->elem.hdr.cmd_id, p_ent->elem.hdr.protocol_id); - - /* Let the flow complete w/o any error handling */ - qed_spq_recov_set_ret_code(p_ent, fw_return_code); - return 0; - } - /* Complete the entry */ rc = qed_spq_fill_entry(p_hwfn, p_ent); diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index 71e28be58102..ca6290fa0f30 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -4447,13 +4447,6 @@ int qed_sriov_disable(struct qed_dev *cdev, bool pci_enabled) if (cdev->p_iov_info && cdev->p_iov_info->num_vfs && pci_enabled) pci_disable_sriov(cdev->pdev); - if (cdev->recov_in_prog) { - DP_VERBOSE(cdev, - QED_MSG_IOV, - "Skip SRIOV disable operations in the device since a recovery is in progress\n"); - goto out; - } - for_each_hwfn(cdev, i) { struct qed_hwfn *hwfn = &cdev->hwfns[i]; struct qed_ptt *ptt = qed_ptt_acquire(hwfn); @@ -4493,7 +4486,7 @@ int qed_sriov_disable(struct qed_dev *cdev, bool pci_enabled) qed_ptt_release(hwfn, ptt); } -out: + qed_iov_set_vfs_to_disable(cdev, false); return 0; diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h index 843416404aeb..613249d1e967 100644 --- a/drivers/net/ethernet/qlogic/qede/qede.h +++ b/drivers/net/ethernet/qlogic/qede/qede.h @@ -162,7 +162,6 @@ struct qede_rdma_dev { struct list_head entry; struct list_head rdma_event_list; struct workqueue_struct *rdma_wq; - bool exp_recovery; }; struct qede_ptp; @@ -265,7 +264,6 @@ struct qede_dev { enum QEDE_STATE { QEDE_STATE_CLOSED, QEDE_STATE_OPEN, - QEDE_STATE_RECOVERY, }; #define HILO_U64(hi, lo) ((((u64)(hi)) << 32) + (lo)) @@ -464,7 +462,6 @@ struct qede_fastpath { #define QEDE_CSUM_UNNECESSARY BIT(1) #define QEDE_TUNN_CSUM_UNNECESSARY BIT(2) -#define QEDE_SP_RECOVERY 0 #define QEDE_SP_RX_MODE 1 #ifdef CONFIG_RFS_ACCEL diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index de955f2b2980..5a74fcbdbc2b 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -133,12 +133,23 @@ static int qede_probe(struct pci_dev *pdev, const struct pci_device_id *id); static void qede_remove(struct pci_dev *pdev); static void qede_shutdown(struct pci_dev *pdev); static void qede_link_update(void *dev, struct qed_link_output *link); -static void qede_schedule_recovery_handler(void *dev); -static void qede_recovery_handler(struct qede_dev *edev); static void qede_get_eth_tlv_data(void *edev, void *data); static void qede_get_generic_tlv_data(void *edev, struct qed_generic_tlvs *data); +/* The qede lock is used to protect driver state change and driver flows that + * are not reentrant. + */ +void __qede_lock(struct qede_dev *edev) +{ + mutex_lock(&edev->qede_lock); +} + +void __qede_unlock(struct qede_dev *edev) +{ + mutex_unlock(&edev->qede_lock); +} + #ifdef CONFIG_QED_SRIOV static int qede_set_vf_vlan(struct net_device *ndev, int vf, u16 vlan, u8 qos, __be16 vlan_proto) @@ -220,7 +231,6 @@ static struct qed_eth_cb_ops qede_ll_ops = { .arfs_filter_op = qede_arfs_filter_op, #endif .link_update = qede_link_update, - .schedule_recovery_handler = qede_schedule_recovery_handler, .get_generic_tlv_data = qede_get_generic_tlv_data, .get_protocol_tlv_data = qede_get_eth_tlv_data, }, @@ -940,57 +950,11 @@ err: return -ENOMEM; } -/* The qede lock is used to protect driver state change and driver flows that - * are not reentrant. - */ -void __qede_lock(struct qede_dev *edev) -{ - mutex_lock(&edev->qede_lock); -} - -void __qede_unlock(struct qede_dev *edev) -{ - mutex_unlock(&edev->qede_lock); -} - -/* This version of the lock should be used when acquiring the RTNL lock is also - * needed in addition to the internal qede lock. - */ -void qede_lock(struct qede_dev *edev) -{ - rtnl_lock(); - __qede_lock(edev); -} - -void qede_unlock(struct qede_dev *edev) -{ - __qede_unlock(edev); - rtnl_unlock(); -} - static void qede_sp_task(struct work_struct *work) { struct qede_dev *edev = container_of(work, struct qede_dev, sp_task.work); - /* The locking scheme depends on the specific flag: - * In case of QEDE_SP_RECOVERY, acquiring the RTNL lock is required to - * ensure that ongoing flows are ended and new ones are not started. - * In other cases - only the internal qede lock should be acquired. - */ - - if (test_and_clear_bit(QEDE_SP_RECOVERY, &edev->sp_flags)) { -#ifdef CONFIG_QED_SRIOV - /* SRIOV must be disabled outside the lock to avoid a deadlock. - * The recovery of the active VFs is currently not supported. - */ - qede_sriov_configure(edev->pdev, 0); -#endif - qede_lock(edev); - qede_recovery_handler(edev); - qede_unlock(edev); - } - __qede_lock(edev); if (test_and_clear_bit(QEDE_SP_RX_MODE, &edev->sp_flags)) @@ -1067,13 +1031,8 @@ static void qede_log_probe(struct qede_dev *edev) enum qede_probe_mode { QEDE_PROBE_NORMAL, - QEDE_PROBE_RECOVERY, }; -#define QEDE_RDMA_PROBE_MODE(mode) \ - ((mode) == QEDE_PROBE_NORMAL ? QEDE_RDMA_PROBE_NORMAL \ - : QEDE_RDMA_PROBE_RECOVERY) - static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, bool is_vf, enum qede_probe_mode mode) { @@ -1092,7 +1051,6 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, probe_params.dp_module = dp_module; probe_params.dp_level = dp_level; probe_params.is_vf = is_vf; - probe_params.recov_in_prog = (mode == QEDE_PROBE_RECOVERY); cdev = qed_ops->common->probe(pdev, &probe_params); if (!cdev) { rc = -ENODEV; @@ -1120,20 +1078,11 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, if (rc) goto err2; - if (mode != QEDE_PROBE_RECOVERY) { - edev = qede_alloc_etherdev(cdev, pdev, &dev_info, dp_module, - dp_level); - if (!edev) { - rc = -ENOMEM; - goto err2; - } - } else { - struct net_device *ndev = pci_get_drvdata(pdev); - - edev = netdev_priv(ndev); - edev->cdev = cdev; - memset(&edev->stats, 0, sizeof(edev->stats)); - memcpy(&edev->dev_info, &dev_info, sizeof(dev_info)); + edev = qede_alloc_etherdev(cdev, pdev, &dev_info, dp_module, + dp_level); + if (!edev) { + rc = -ENOMEM; + goto err2; } if (is_vf) @@ -1141,31 +1090,28 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, qede_init_ndev(edev); - rc = qede_rdma_dev_add(edev, QEDE_RDMA_PROBE_MODE(mode)); + rc = qede_rdma_dev_add(edev); if (rc) goto err3; - if (mode != QEDE_PROBE_RECOVERY) { - /* Prepare the lock prior to the registration of the netdev, - * as once it's registered we might reach flows requiring it - * [it's even possible to reach a flow needing it directly - * from there, although it's unlikely]. - */ - INIT_DELAYED_WORK(&edev->sp_task, qede_sp_task); - mutex_init(&edev->qede_lock); - - rc = register_netdev(edev->ndev); - if (rc) { - DP_NOTICE(edev, "Cannot register net-device\n"); - goto err4; - } + /* Prepare the lock prior to the registration of the netdev, + * as once it's registered we might reach flows requiring it + * [it's even possible to reach a flow needing it directly + * from there, although it's unlikely]. + */ + INIT_DELAYED_WORK(&edev->sp_task, qede_sp_task); + mutex_init(&edev->qede_lock); + rc = register_netdev(edev->ndev); + if (rc) { + DP_NOTICE(edev, "Cannot register net-device\n"); + goto err4; } edev->ops->common->set_name(cdev, edev->ndev->name); /* PTP not supported on VFs */ if (!is_vf) - qede_ptp_enable(edev, (mode == QEDE_PROBE_NORMAL)); + qede_ptp_enable(edev, true); edev->ops->register_ops(cdev, &qede_ll_ops, edev); @@ -1180,7 +1126,7 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, return 0; err4: - qede_rdma_dev_remove(edev, QEDE_RDMA_PROBE_MODE(mode)); + qede_rdma_dev_remove(edev); err3: free_netdev(edev->ndev); err2: @@ -1216,13 +1162,8 @@ static int qede_probe(struct pci_dev *pdev, const struct pci_device_id *id) enum qede_remove_mode { QEDE_REMOVE_NORMAL, - QEDE_REMOVE_RECOVERY, }; -#define QEDE_RDMA_REMOVE_MODE(mode) \ - ((mode) == QEDE_REMOVE_NORMAL ? QEDE_RDMA_REMOVE_NORMAL \ - : QEDE_RDMA_REMOVE_RECOVERY) - static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode) { struct net_device *ndev = pci_get_drvdata(pdev); @@ -1231,19 +1172,15 @@ static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode) DP_INFO(edev, "Starting qede_remove\n"); - qede_rdma_dev_remove(edev, QEDE_RDMA_REMOVE_MODE(mode)); - - if (mode != QEDE_REMOVE_RECOVERY) { - unregister_netdev(ndev); + qede_rdma_dev_remove(edev); + unregister_netdev(ndev); + cancel_delayed_work_sync(&edev->sp_task); - cancel_delayed_work_sync(&edev->sp_task); + qede_ptp_disable(edev); - edev->ops->common->set_power_state(cdev, PCI_D0); + edev->ops->common->set_power_state(cdev, PCI_D0); - pci_set_drvdata(pdev, NULL); - } - - qede_ptp_disable(edev); + pci_set_drvdata(pdev, NULL); /* Use global ops since we've freed edev */ qed_ops->common->slowpath_stop(cdev); @@ -1257,8 +1194,7 @@ static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode) * [e.g., QED register callbacks] won't break anything when * accessing the netdevice. */ - if (mode != QEDE_REMOVE_RECOVERY) - free_netdev(ndev); + free_netdev(ndev); dev_info(&pdev->dev, "Ending qede_remove successfully\n"); } @@ -1603,58 +1539,6 @@ static int qede_alloc_mem_load(struct qede_dev *edev) return 0; } -static void qede_empty_tx_queue(struct qede_dev *edev, - struct qede_tx_queue *txq) -{ - unsigned int pkts_compl = 0, bytes_compl = 0; - struct netdev_queue *netdev_txq; - int rc, len = 0; - - netdev_txq = netdev_get_tx_queue(edev->ndev, txq->ndev_txq_id); - - while (qed_chain_get_cons_idx(&txq->tx_pbl) != - qed_chain_get_prod_idx(&txq->tx_pbl)) { - DP_VERBOSE(edev, NETIF_MSG_IFDOWN, - "Freeing a packet on tx queue[%d]: chain_cons 0x%x, chain_prod 0x%x\n", - txq->index, qed_chain_get_cons_idx(&txq->tx_pbl), - qed_chain_get_prod_idx(&txq->tx_pbl)); - - rc = qede_free_tx_pkt(edev, txq, &len); - if (rc) { - DP_NOTICE(edev, - "Failed to free a packet on tx queue[%d]: chain_cons 0x%x, chain_prod 0x%x\n", - txq->index, - qed_chain_get_cons_idx(&txq->tx_pbl), - qed_chain_get_prod_idx(&txq->tx_pbl)); - break; - } - - bytes_compl += len; - pkts_compl++; - txq->sw_tx_cons++; - } - - netdev_tx_completed_queue(netdev_txq, pkts_compl, bytes_compl); -} - -static void qede_empty_tx_queues(struct qede_dev *edev) -{ - int i; - - for_each_queue(i) - if (edev->fp_array[i].type & QEDE_FASTPATH_TX) { - int cos; - - for_each_cos_in_txq(edev, cos) { - struct qede_fastpath *fp; - - fp = &edev->fp_array[i]; - qede_empty_tx_queue(edev, - &fp->txq[cos]); - } - } -} - /* This function inits fp content and resets the SB, RXQ and TXQ structures */ static void qede_init_fp(struct qede_dev *edev) { @@ -2169,7 +2053,6 @@ out: enum qede_unload_mode { QEDE_UNLOAD_NORMAL, - QEDE_UNLOAD_RECOVERY, }; static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode, @@ -2185,8 +2068,7 @@ static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode, clear_bit(QEDE_FLAGS_LINK_REQUESTED, &edev->flags); - if (mode != QEDE_UNLOAD_RECOVERY) - edev->state = QEDE_STATE_CLOSED; + edev->state = QEDE_STATE_CLOSED; qede_rdma_dev_event_close(edev); @@ -2194,21 +2076,18 @@ static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode, netif_tx_disable(edev->ndev); netif_carrier_off(edev->ndev); - if (mode != QEDE_UNLOAD_RECOVERY) { - /* Reset the link */ - memset(&link_params, 0, sizeof(link_params)); - link_params.link_up = false; - edev->ops->common->set_link(edev->cdev, &link_params); - - rc = qede_stop_queues(edev); - if (rc) { - qede_sync_free_irqs(edev); - goto out; - } - - DP_INFO(edev, "Stopped Queues\n"); + /* Reset the link */ + memset(&link_params, 0, sizeof(link_params)); + link_params.link_up = false; + edev->ops->common->set_link(edev->cdev, &link_params); + rc = qede_stop_queues(edev); + if (rc) { + qede_sync_free_irqs(edev); + goto out; } + DP_INFO(edev, "Stopped Queues\n"); + qede_vlan_mark_nonconfigured(edev); edev->ops->fastpath_stop(edev->cdev); @@ -2223,26 +2102,18 @@ static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode, qede_napi_disable_remove(edev); - if (mode == QEDE_UNLOAD_RECOVERY) - qede_empty_tx_queues(edev); - qede_free_mem_load(edev); qede_free_fp_array(edev); out: if (!is_locked) __qede_unlock(edev); - - if (mode != QEDE_UNLOAD_RECOVERY) - DP_NOTICE(edev, "Link is down\n"); - DP_INFO(edev, "Ending qede unload\n"); } enum qede_load_mode { QEDE_LOAD_NORMAL, QEDE_LOAD_RELOAD, - QEDE_LOAD_RECOVERY, }; static int qede_load(struct qede_dev *edev, enum qede_load_mode mode, @@ -2422,77 +2293,6 @@ static void qede_link_update(void *dev, struct qed_link_output *link) } } -static void qede_schedule_recovery_handler(void *dev) -{ - struct qede_dev *edev = dev; - - if (edev->state == QEDE_STATE_RECOVERY) { - DP_NOTICE(edev, - "Avoid scheduling a recovery handling since already in recovery state\n"); - return; - } - - set_bit(QEDE_SP_RECOVERY, &edev->sp_flags); - schedule_delayed_work(&edev->sp_task, 0); - - DP_INFO(edev, "Scheduled a recovery handler\n"); -} - -static void qede_recovery_failed(struct qede_dev *edev) -{ - netdev_err(edev->ndev, "Recovery handling has failed. Power cycle is needed.\n"); - - netif_device_detach(edev->ndev); - - if (edev->cdev) - edev->ops->common->set_power_state(edev->cdev, PCI_D3hot); -} - -static void qede_recovery_handler(struct qede_dev *edev) -{ - u32 curr_state = edev->state; - int rc; - - DP_NOTICE(edev, "Starting a recovery process\n"); - - /* No need to acquire first the qede_lock since is done by qede_sp_task - * before calling this function. - */ - edev->state = QEDE_STATE_RECOVERY; - - edev->ops->common->recovery_prolog(edev->cdev); - - if (curr_state == QEDE_STATE_OPEN) - qede_unload(edev, QEDE_UNLOAD_RECOVERY, true); - - __qede_remove(edev->pdev, QEDE_REMOVE_RECOVERY); - - rc = __qede_probe(edev->pdev, edev->dp_module, edev->dp_level, - IS_VF(edev), QEDE_PROBE_RECOVERY); - if (rc) { - edev->cdev = NULL; - goto err; - } - - if (curr_state == QEDE_STATE_OPEN) { - rc = qede_load(edev, QEDE_LOAD_RECOVERY, true); - if (rc) - goto err; - - qede_config_rx_mode(edev->ndev); - udp_tunnel_get_rx_info(edev->ndev); - } - - edev->state = curr_state; - - DP_NOTICE(edev, "Recovery handling is done\n"); - - return; - -err: - qede_recovery_failed(edev); -} - static bool qede_is_txq_full(struct qede_dev *edev, struct qede_tx_queue *txq) { struct netdev_queue *netdev_txq; diff --git a/drivers/net/ethernet/qlogic/qede/qede_rdma.c b/drivers/net/ethernet/qlogic/qede/qede_rdma.c index 9668e5e47d5f..1900bf7e67d1 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_rdma.c +++ b/drivers/net/ethernet/qlogic/qede/qede_rdma.c @@ -50,8 +50,6 @@ static void _qede_rdma_dev_add(struct qede_dev *edev) if (!qedr_drv) return; - /* Leftovers from previous error recovery */ - edev->rdma_info.exp_recovery = false; edev->rdma_info.qedr_dev = qedr_drv->add(edev->cdev, edev->pdev, edev->ndev); } @@ -89,26 +87,21 @@ static void qede_rdma_destroy_wq(struct qede_dev *edev) destroy_workqueue(edev->rdma_info.rdma_wq); } -int qede_rdma_dev_add(struct qede_dev *edev, enum qede_rdma_probe_mode mode) +int qede_rdma_dev_add(struct qede_dev *edev) { - int rc; + int rc = 0; - if (!qede_rdma_supported(edev)) - return 0; - - /* Cannot start qedr while recovering since it wasn't fully stopped */ - if (mode == QEDE_RDMA_PROBE_RECOVERY) - return 0; - - rc = qede_rdma_create_wq(edev); - if (rc) - return rc; + if (qede_rdma_supported(edev)) { + rc = qede_rdma_create_wq(edev); + if (rc) + return rc; - INIT_LIST_HEAD(&edev->rdma_info.entry); - mutex_lock(&qedr_dev_list_lock); - list_add_tail(&edev->rdma_info.entry, &qedr_dev_list); - _qede_rdma_dev_add(edev); - mutex_unlock(&qedr_dev_list_lock); + INIT_LIST_HEAD(&edev->rdma_info.entry); + mutex_lock(&qedr_dev_list_lock); + list_add_tail(&edev->rdma_info.entry, &qedr_dev_list); + _qede_rdma_dev_add(edev); + mutex_unlock(&qedr_dev_list_lock); + } return rc; } @@ -117,31 +110,19 @@ static void _qede_rdma_dev_remove(struct qede_dev *edev) { if (qedr_drv && qedr_drv->remove && edev->rdma_info.qedr_dev) qedr_drv->remove(edev->rdma_info.qedr_dev); + edev->rdma_info.qedr_dev = NULL; } -void qede_rdma_dev_remove(struct qede_dev *edev, - enum qede_rdma_remove_mode mode) +void qede_rdma_dev_remove(struct qede_dev *edev) { if (!qede_rdma_supported(edev)) return; - /* Cannot remove qedr while recovering since it wasn't fully stopped */ - if (mode == QEDE_RDMA_REMOVE_NORMAL) { - qede_rdma_destroy_wq(edev); - mutex_lock(&qedr_dev_list_lock); - if (!edev->rdma_info.exp_recovery) - _qede_rdma_dev_remove(edev); - edev->rdma_info.qedr_dev = NULL; - list_del(&edev->rdma_info.entry); - mutex_unlock(&qedr_dev_list_lock); - } else { - if (!edev->rdma_info.exp_recovery) { - mutex_lock(&qedr_dev_list_lock); - _qede_rdma_dev_remove(edev); - mutex_unlock(&qedr_dev_list_lock); - } - edev->rdma_info.exp_recovery = true; - } + qede_rdma_destroy_wq(edev); + mutex_lock(&qedr_dev_list_lock); + _qede_rdma_dev_remove(edev); + list_del(&edev->rdma_info.entry); + mutex_unlock(&qedr_dev_list_lock); } static void _qede_rdma_dev_open(struct qede_dev *edev) @@ -223,8 +204,7 @@ void qede_rdma_unregister_driver(struct qedr_driver *drv) mutex_lock(&qedr_dev_list_lock); list_for_each_entry(edev, &qedr_dev_list, rdma_info.entry) { - /* If device has experienced recovery it was already removed */ - if (edev->rdma_info.qedr_dev && !edev->rdma_info.exp_recovery) + if (edev->rdma_info.qedr_dev) _qede_rdma_dev_remove(edev); } qedr_drv = NULL; @@ -304,10 +284,6 @@ static void qede_rdma_add_event(struct qede_dev *edev, { struct qede_rdma_event_work *event_node; - /* If a recovery was experienced avoid adding the event */ - if (edev->rdma_info.exp_recovery) - return; - if (!edev->rdma_info.qedr_dev) return; diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h index c2a1b7dbe4eb..91c536a01b56 100644 --- a/include/linux/qed/qed_if.h +++ b/include/linux/qed/qed_if.h @@ -764,7 +764,6 @@ struct qed_probe_params { u32 dp_module; u8 dp_level; bool is_vf; - bool recov_in_prog; }; #define QED_DRV_VER_STR_SIZE 12 @@ -811,7 +810,6 @@ struct qed_common_cb_ops { void (*arfs_filter_op)(void *dev, void *fltr, u8 fw_rc); void (*link_update)(void *dev, struct qed_link_output *link); - void (*schedule_recovery_handler)(void *dev); void (*dcbx_aen)(void *dev, struct qed_dcbx_get *get, u32 mib_type); void (*get_generic_tlv_data)(void *dev, struct qed_generic_tlvs *data); void (*get_protocol_tlv_data)(void *dev, void *data); @@ -1059,24 +1057,6 @@ struct qed_common_ops { int (*db_recovery_del)(struct qed_dev *cdev, void __iomem *db_addr, void *db_data); -/** - * @brief recovery_process - Trigger a recovery process - * - * @param cdev - * - * @return 0 on success, error otherwise. - */ - int (*recovery_process)(struct qed_dev *cdev); - -/** - * @brief recovery_prolog - Execute the prolog operations of a recovery process - * - * @param cdev - * - * @return 0 on success, error otherwise. - */ - int (*recovery_prolog)(struct qed_dev *cdev); - /** * @brief update_drv_state - API to inform the change in the driver state. * diff --git a/include/linux/qed/qede_rdma.h b/include/linux/qed/qede_rdma.h index e29d7199c10e..9904617a9730 100644 --- a/include/linux/qed/qede_rdma.h +++ b/include/linux/qed/qede_rdma.h @@ -55,16 +55,6 @@ struct qede_rdma_event_work { enum qede_rdma_event event; }; -enum qede_rdma_probe_mode { - QEDE_RDMA_PROBE_NORMAL, - QEDE_RDMA_PROBE_RECOVERY, -}; - -enum qede_rdma_remove_mode { - QEDE_RDMA_REMOVE_NORMAL, - QEDE_RDMA_REMOVE_RECOVERY, -}; - struct qedr_driver { unsigned char name[32]; @@ -84,24 +74,21 @@ void qede_rdma_unregister_driver(struct qedr_driver *drv); bool qede_rdma_supported(struct qede_dev *dev); #if IS_ENABLED(CONFIG_QED_RDMA) -int qede_rdma_dev_add(struct qede_dev *dev, enum qede_rdma_probe_mode mode); +int qede_rdma_dev_add(struct qede_dev *dev); void qede_rdma_dev_event_open(struct qede_dev *dev); void qede_rdma_dev_event_close(struct qede_dev *dev); -void qede_rdma_dev_remove(struct qede_dev *dev, - enum qede_rdma_remove_mode mode); +void qede_rdma_dev_remove(struct qede_dev *dev); void qede_rdma_event_changeaddr(struct qede_dev *edr); #else -static inline int qede_rdma_dev_add(struct qede_dev *dev, - enum qede_rdma_probe_mode mode) +static inline int qede_rdma_dev_add(struct qede_dev *dev) { return 0; } static inline void qede_rdma_dev_event_open(struct qede_dev *dev) {} static inline void qede_rdma_dev_event_close(struct qede_dev *dev) {} -static inline void qede_rdma_dev_remove(struct qede_dev *dev, - enum qede_rdma_remove_mode mode) {} +static inline void qede_rdma_dev_remove(struct qede_dev *dev) {} static inline void qede_rdma_event_changeaddr(struct qede_dev *edr) {} #endif #endif -- cgit v1.2.3 From a7b76c8857692b0fce063b94ed83da11c396d341 Mon Sep 17 00:00:00 2001 From: Jiong Wang Date: Sat, 26 Jan 2019 12:26:05 -0500 Subject: bpf: JIT blinds support JMP32 This patch adds JIT blinds support for JMP32. Like BPF_JMP_REG/IMM, JMP32 version are needed for building raw bpf insn. They are added to both include/linux/filter.h and tools/include/linux/filter.h. Reviewed-by: Jakub Kicinski Signed-off-by: Jiong Wang Signed-off-by: Alexei Starovoitov --- include/linux/filter.h | 20 ++++++++++++++++++++ kernel/bpf/core.c | 21 +++++++++++++++++++++ tools/include/linux/filter.h | 20 ++++++++++++++++++++ 3 files changed, 61 insertions(+) (limited to 'include/linux') diff --git a/include/linux/filter.h b/include/linux/filter.h index be9af6b4a9e4..e4b473f85b46 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -277,6 +277,26 @@ struct sock_reuseport; .off = OFF, \ .imm = IMM }) +/* Like BPF_JMP_REG, but with 32-bit wide operands for comparison. */ + +#define BPF_JMP32_REG(OP, DST, SRC, OFF) \ + ((struct bpf_insn) { \ + .code = BPF_JMP32 | BPF_OP(OP) | BPF_X, \ + .dst_reg = DST, \ + .src_reg = SRC, \ + .off = OFF, \ + .imm = 0 }) + +/* Like BPF_JMP_IMM, but with 32-bit wide operands for comparison. */ + +#define BPF_JMP32_IMM(OP, DST, IMM, OFF) \ + ((struct bpf_insn) { \ + .code = BPF_JMP32 | BPF_OP(OP) | BPF_K, \ + .dst_reg = DST, \ + .src_reg = 0, \ + .off = OFF, \ + .imm = IMM }) + /* Unconditional jumps, goto pc + off16 */ #define BPF_JMP_A(OFF) \ diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index bba11c2565ee..a7bcb23bee84 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -949,6 +949,27 @@ static int bpf_jit_blind_insn(const struct bpf_insn *from, *to++ = BPF_JMP_REG(from->code, from->dst_reg, BPF_REG_AX, off); break; + case BPF_JMP32 | BPF_JEQ | BPF_K: + case BPF_JMP32 | BPF_JNE | BPF_K: + case BPF_JMP32 | BPF_JGT | BPF_K: + case BPF_JMP32 | BPF_JLT | BPF_K: + case BPF_JMP32 | BPF_JGE | BPF_K: + case BPF_JMP32 | BPF_JLE | BPF_K: + case BPF_JMP32 | BPF_JSGT | BPF_K: + case BPF_JMP32 | BPF_JSLT | BPF_K: + case BPF_JMP32 | BPF_JSGE | BPF_K: + case BPF_JMP32 | BPF_JSLE | BPF_K: + case BPF_JMP32 | BPF_JSET | BPF_K: + /* Accommodate for extra offset in case of a backjump. */ + off = from->off; + if (off < 0) + off -= 2; + *to++ = BPF_ALU32_IMM(BPF_MOV, BPF_REG_AX, imm_rnd ^ from->imm); + *to++ = BPF_ALU32_IMM(BPF_XOR, BPF_REG_AX, imm_rnd); + *to++ = BPF_JMP32_REG(from->code, from->dst_reg, BPF_REG_AX, + off); + break; + case BPF_LD | BPF_IMM | BPF_DW: *to++ = BPF_ALU64_IMM(BPF_MOV, BPF_REG_AX, imm_rnd ^ aux[1].imm); *to++ = BPF_ALU64_IMM(BPF_XOR, BPF_REG_AX, imm_rnd); diff --git a/tools/include/linux/filter.h b/tools/include/linux/filter.h index af55acf73e75..cce0b02c0e28 100644 --- a/tools/include/linux/filter.h +++ b/tools/include/linux/filter.h @@ -199,6 +199,16 @@ .off = OFF, \ .imm = 0 }) +/* Like BPF_JMP_REG, but with 32-bit wide operands for comparison. */ + +#define BPF_JMP32_REG(OP, DST, SRC, OFF) \ + ((struct bpf_insn) { \ + .code = BPF_JMP32 | BPF_OP(OP) | BPF_X, \ + .dst_reg = DST, \ + .src_reg = SRC, \ + .off = OFF, \ + .imm = 0 }) + /* Conditional jumps against immediates, if (dst_reg 'op' imm32) goto pc + off16 */ #define BPF_JMP_IMM(OP, DST, IMM, OFF) \ @@ -209,6 +219,16 @@ .off = OFF, \ .imm = IMM }) +/* Like BPF_JMP_IMM, but with 32-bit wide operands for comparison. */ + +#define BPF_JMP32_IMM(OP, DST, IMM, OFF) \ + ((struct bpf_insn) { \ + .code = BPF_JMP32 | BPF_OP(OP) | BPF_K, \ + .dst_reg = DST, \ + .src_reg = 0, \ + .off = OFF, \ + .imm = IMM }) + /* Unconditional jumps, goto pc + off16 */ #define BPF_JMP_A(OFF) \ -- cgit v1.2.3 From 83f529281d7aa42b10c2c5cb64fcbd2c7cab4409 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sun, 27 Jan 2019 19:18:57 +0100 Subject: netfilter: ipv4: remove useless export_symbol Only one caller; place it where needed and get rid of the EXPORT_SYMBOL. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter_ipv4.h | 6 ------ net/ipv4/netfilter.c | 18 ------------------ net/netfilter/utils.c | 19 +++++++++++++++++++ 3 files changed, 19 insertions(+), 24 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv4.h b/include/linux/netfilter_ipv4.h index 95ab5cc64422..082e2c41b7ff 100644 --- a/include/linux/netfilter_ipv4.h +++ b/include/linux/netfilter_ipv4.h @@ -25,7 +25,6 @@ __sum16 nf_ip_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol); int nf_ip_route(struct net *net, struct dst_entry **dst, struct flowi *fl, bool strict); -int nf_ip_reroute(struct sk_buff *skb, const struct nf_queue_entry *entry); #else static inline __sum16 nf_ip_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol) @@ -37,11 +36,6 @@ static inline int nf_ip_route(struct net *net, struct dst_entry **dst, { return -EOPNOTSUPP; } -static inline int nf_ip_reroute(struct sk_buff *skb, - const struct nf_queue_entry *entry) -{ - return -EOPNOTSUPP; -} #endif /* CONFIG_INET */ #endif /*__LINUX_IP_NETFILTER_H*/ diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c index 8d2e5dc9a827..a058213b77a7 100644 --- a/net/ipv4/netfilter.c +++ b/net/ipv4/netfilter.c @@ -80,24 +80,6 @@ int ip_route_me_harder(struct net *net, struct sk_buff *skb, unsigned int addr_t } EXPORT_SYMBOL(ip_route_me_harder); -int nf_ip_reroute(struct sk_buff *skb, const struct nf_queue_entry *entry) -{ - const struct ip_rt_info *rt_info = nf_queue_entry_reroute(entry); - - if (entry->state.hook == NF_INET_LOCAL_OUT) { - const struct iphdr *iph = ip_hdr(skb); - - if (!(iph->tos == rt_info->tos && - skb->mark == rt_info->mark && - iph->daddr == rt_info->daddr && - iph->saddr == rt_info->saddr)) - return ip_route_me_harder(entry->state.net, skb, - RTN_UNSPEC); - } - return 0; -} -EXPORT_SYMBOL_GPL(nf_ip_reroute); - int nf_ip_route(struct net *net, struct dst_entry **dst, struct flowi *fl, bool strict __always_unused) { diff --git a/net/netfilter/utils.c b/net/netfilter/utils.c index e8da9a9bba73..55af9f247993 100644 --- a/net/netfilter/utils.c +++ b/net/netfilter/utils.c @@ -180,6 +180,25 @@ int nf_route(struct net *net, struct dst_entry **dst, struct flowi *fl, } EXPORT_SYMBOL_GPL(nf_route); +static int nf_ip_reroute(struct sk_buff *skb, const struct nf_queue_entry *entry) +{ +#ifdef CONFIG_INET + const struct ip_rt_info *rt_info = nf_queue_entry_reroute(entry); + + if (entry->state.hook == NF_INET_LOCAL_OUT) { + const struct iphdr *iph = ip_hdr(skb); + + if (!(iph->tos == rt_info->tos && + skb->mark == rt_info->mark && + iph->daddr == rt_info->daddr && + iph->saddr == rt_info->saddr)) + return ip_route_me_harder(entry->state.net, skb, + RTN_UNSPEC); + } +#endif + return 0; +} + int nf_reroute(struct sk_buff *skb, struct nf_queue_entry *entry) { const struct nf_ipv6_ops *v6ops; -- cgit v1.2.3 From 64515dc899df898991b2b7e56f69f56f014ea888 Mon Sep 17 00:00:00 2001 From: Tomer Tayar Date: Mon, 28 Jan 2019 19:27:55 +0200 Subject: qed: Add infrastructure for error detection and recovery This patch adds the detection and handling of a parity error ("process kill event"), including the update of the protocol drivers, and the prevention of any HW access that will lead to device access towards the host while recovery is in progress. It also provides the means for the protocol drivers to trigger a recovery process on their decision. Signed-off-by: Tomer Tayar Signed-off-by: Ariel Elior Signed-off-by: Michal Kalderon Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed.h | 4 ++ drivers/net/ethernet/qlogic/qed/qed_dev.c | 41 +++++++---- drivers/net/ethernet/qlogic/qed/qed_hsi.h | 2 +- drivers/net/ethernet/qlogic/qed/qed_hw.c | 11 +++ drivers/net/ethernet/qlogic/qed/qed_main.c | 30 ++++++++ drivers/net/ethernet/qlogic/qed/qed_mcp.c | 94 ++++++++++++++++++++++++++ drivers/net/ethernet/qlogic/qed/qed_mcp.h | 32 +++++++++ drivers/net/ethernet/qlogic/qed/qed_reg_addr.h | 2 + drivers/net/ethernet/qlogic/qed/qed_spq.c | 22 ++++++ drivers/net/ethernet/qlogic/qed/qed_sriov.c | 9 ++- include/linux/qed/qed_if.h | 20 ++++++ 11 files changed, 251 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h index b352e313e1f6..3b0955d34716 100644 --- a/drivers/net/ethernet/qlogic/qed/qed.h +++ b/drivers/net/ethernet/qlogic/qed/qed.h @@ -804,6 +804,9 @@ struct qed_dev { u32 mcp_nvm_resp; + /* Recovery */ + bool recov_in_prog; + /* Linux specific here */ struct qede_dev *edev; struct pci_dev *pdev; @@ -943,6 +946,7 @@ void qed_link_update(struct qed_hwfn *hwfn, struct qed_ptt *ptt); u32 qed_unzip_data(struct qed_hwfn *p_hwfn, u32 input_len, u8 *input_buf, u32 max_size, u8 *unzip_buf); +void qed_schedule_recovery_handler(struct qed_hwfn *p_hwfn); void qed_get_protocol_stats(struct qed_dev *cdev, enum qed_mcp_protocol_type type, union qed_mcp_protocol_stats *stats); diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index fa5f07e65672..b17003d9066c 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -2140,6 +2140,11 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params) "Load request was sent. Load code: 0x%x\n", load_code); + /* Only relevant for recovery: + * Clear the indication after LOAD_REQ is responded by the MFW. + */ + cdev->recov_in_prog = false; + qed_mcp_set_capabilities(p_hwfn, p_hwfn->p_main_ptt); qed_reset_mb_shadow(p_hwfn, p_hwfn->p_main_ptt); @@ -2291,6 +2296,9 @@ static void qed_hw_timers_stop(struct qed_dev *cdev, qed_wr(p_hwfn, p_ptt, TM_REG_PF_ENABLE_CONN, 0x0); qed_wr(p_hwfn, p_ptt, TM_REG_PF_ENABLE_TASK, 0x0); + if (cdev->recov_in_prog) + return; + for (i = 0; i < QED_HW_STOP_RETRY_LIMIT; i++) { if ((!qed_rd(p_hwfn, p_ptt, TM_REG_PF_SCAN_ACTIVE_CONN)) && @@ -2353,12 +2361,14 @@ int qed_hw_stop(struct qed_dev *cdev) p_hwfn->hw_init_done = false; /* Send unload command to MCP */ - rc = qed_mcp_unload_req(p_hwfn, p_ptt); - if (rc) { - DP_NOTICE(p_hwfn, - "Failed sending a UNLOAD_REQ command. rc = %d.\n", - rc); - rc2 = -EINVAL; + if (!cdev->recov_in_prog) { + rc = qed_mcp_unload_req(p_hwfn, p_ptt); + if (rc) { + DP_NOTICE(p_hwfn, + "Failed sending a UNLOAD_REQ command. rc = %d.\n", + rc); + rc2 = -EINVAL; + } } qed_slowpath_irq_sync(p_hwfn); @@ -2400,16 +2410,18 @@ int qed_hw_stop(struct qed_dev *cdev) qed_wr(p_hwfn, p_ptt, DORQ_REG_PF_DB_ENABLE, 0); qed_wr(p_hwfn, p_ptt, QM_REG_PF_EN, 0); - qed_mcp_unload_done(p_hwfn, p_ptt); - if (rc) { - DP_NOTICE(p_hwfn, - "Failed sending a UNLOAD_DONE command. rc = %d.\n", - rc); - rc2 = -EINVAL; + if (!cdev->recov_in_prog) { + rc = qed_mcp_unload_done(p_hwfn, p_ptt); + if (rc) { + DP_NOTICE(p_hwfn, + "Failed sending a UNLOAD_DONE command. rc = %d.\n", + rc); + rc2 = -EINVAL; + } } } - if (IS_PF(cdev)) { + if (IS_PF(cdev) && !cdev->recov_in_prog) { p_hwfn = QED_LEADING_HWFN(cdev); p_ptt = QED_LEADING_HWFN(cdev)->p_main_ptt; @@ -3459,6 +3471,7 @@ static int qed_hw_prepare_single(struct qed_hwfn *p_hwfn, void __iomem *p_doorbells, enum qed_pci_personality personality) { + struct qed_dev *cdev = p_hwfn->cdev; int rc = 0; /* Split PCI bars evenly between hwfns */ @@ -3511,7 +3524,7 @@ static int qed_hw_prepare_single(struct qed_hwfn *p_hwfn, /* Sending a mailbox to the MFW should be done after qed_get_hw_info() * is called as it sets the ports number in an engine. */ - if (IS_LEAD_HWFN(p_hwfn)) { + if (IS_LEAD_HWFN(p_hwfn) && !cdev->recov_in_prog) { rc = qed_mcp_initiate_pf_flr(p_hwfn, p_hwfn->p_main_ptt); if (rc) DP_NOTICE(p_hwfn, "Failed to initiate PF FLR\n"); diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h index b13cfb449d8f..417121e74ee9 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h @@ -12827,7 +12827,7 @@ enum MFW_DRV_MSG_TYPE { MFW_DRV_MSG_LLDP_DATA_UPDATED, MFW_DRV_MSG_DCBX_REMOTE_MIB_UPDATED, MFW_DRV_MSG_DCBX_OPERATIONAL_MIB_UPDATED, - MFW_DRV_MSG_RESERVED4, + MFW_DRV_MSG_ERROR_RECOVERY, MFW_DRV_MSG_BW_UPDATE, MFW_DRV_MSG_S_TAG_UPDATE, MFW_DRV_MSG_GET_LAN_STATS, diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.c b/drivers/net/ethernet/qlogic/qed/qed_hw.c index 70504dcf4087..72ec1c6bdf70 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hw.c +++ b/drivers/net/ethernet/qlogic/qed/qed_hw.c @@ -703,6 +703,17 @@ static int qed_dmae_execute_command(struct qed_hwfn *p_hwfn, int qed_status = 0; u32 offset = 0; + if (p_hwfn->cdev->recov_in_prog) { + DP_VERBOSE(p_hwfn, + NETIF_MSG_HW, + "Recovery is in progress. Avoid DMAE transaction [{src: addr 0x%llx, type %d}, {dst: addr 0x%llx, type %d}, size %d].\n", + src_addr, src_type, dst_addr, dst_type, + size_in_dwords); + + /* Let the flow complete w/o any error handling */ + return 0; + } + qed_dmae_opcode(p_hwfn, (src_type == QED_DMAE_ADDRESS_GRC), (dst_type == QED_DMAE_ADDRESS_GRC), diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index 6adf5bda9811..b47352643fb5 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -359,6 +359,8 @@ static struct qed_dev *qed_probe(struct pci_dev *pdev, qed_init_dp(cdev, params->dp_module, params->dp_level); + cdev->recov_in_prog = params->recov_in_prog; + rc = qed_init_pci(cdev, pdev); if (rc) { DP_ERR(cdev, "init pci failed\n"); @@ -2203,6 +2205,15 @@ static int qed_nvm_get_image(struct qed_dev *cdev, enum qed_nvm_images type, return qed_mcp_get_nvm_image(hwfn, type, buf, len); } +void qed_schedule_recovery_handler(struct qed_hwfn *p_hwfn) +{ + struct qed_common_cb_ops *ops = p_hwfn->cdev->protocol_ops.common; + void *cookie = p_hwfn->cdev->ops_cookie; + + if (ops && ops->schedule_recovery_handler) + ops->schedule_recovery_handler(cookie); +} + static int qed_set_coalesce(struct qed_dev *cdev, u16 rx_coal, u16 tx_coal, void *handle) { @@ -2226,6 +2237,23 @@ static int qed_set_led(struct qed_dev *cdev, enum qed_led_mode mode) return status; } +static int qed_recovery_process(struct qed_dev *cdev) +{ + struct qed_hwfn *p_hwfn = QED_LEADING_HWFN(cdev); + struct qed_ptt *p_ptt; + int rc = 0; + + p_ptt = qed_ptt_acquire(p_hwfn); + if (!p_ptt) + return -EAGAIN; + + rc = qed_start_recovery_process(p_hwfn, p_ptt); + + qed_ptt_release(p_hwfn, p_ptt); + + return rc; +} + static int qed_update_wol(struct qed_dev *cdev, bool enabled) { struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); @@ -2380,6 +2408,8 @@ const struct qed_common_ops qed_common_ops_pass = { .nvm_get_image = &qed_nvm_get_image, .set_coalesce = &qed_set_coalesce, .set_led = &qed_set_led, + .recovery_process = &qed_recovery_process, + .recovery_prolog = &qed_recovery_prolog, .update_drv_state = &qed_update_drv_state, .update_mac = &qed_update_mac, .update_mtu = &qed_update_mtu, diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c index 1024484d7dd8..bb8541847aa5 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c @@ -1549,6 +1549,60 @@ int qed_mcp_set_link(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, bool b_up) return 0; } +u32 qed_get_process_kill_counter(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt) +{ + u32 path_offsize_addr, path_offsize, path_addr, proc_kill_cnt; + + if (IS_VF(p_hwfn->cdev)) + return -EINVAL; + + path_offsize_addr = SECTION_OFFSIZE_ADDR(p_hwfn->mcp_info->public_base, + PUBLIC_PATH); + path_offsize = qed_rd(p_hwfn, p_ptt, path_offsize_addr); + path_addr = SECTION_ADDR(path_offsize, QED_PATH_ID(p_hwfn)); + + proc_kill_cnt = qed_rd(p_hwfn, p_ptt, + path_addr + + offsetof(struct public_path, process_kill)) & + PROCESS_KILL_COUNTER_MASK; + + return proc_kill_cnt; +} + +static void qed_mcp_handle_process_kill(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt) +{ + struct qed_dev *cdev = p_hwfn->cdev; + u32 proc_kill_cnt; + + /* Prevent possible attentions/interrupts during the recovery handling + * and till its load phase, during which they will be re-enabled. + */ + qed_int_igu_disable_int(p_hwfn, p_ptt); + + DP_NOTICE(p_hwfn, "Received a process kill indication\n"); + + /* The following operations should be done once, and thus in CMT mode + * are carried out by only the first HW function. + */ + if (p_hwfn != QED_LEADING_HWFN(cdev)) + return; + + if (cdev->recov_in_prog) { + DP_NOTICE(p_hwfn, + "Ignoring the indication since a recovery process is already in progress\n"); + return; + } + + cdev->recov_in_prog = true; + + proc_kill_cnt = qed_get_process_kill_counter(p_hwfn, p_ptt); + DP_NOTICE(p_hwfn, "Process kill counter: %d\n", proc_kill_cnt); + + qed_schedule_recovery_handler(p_hwfn); +} + static void qed_mcp_send_protocol_stats(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, enum MFW_DRV_MSG_TYPE type) @@ -1779,6 +1833,9 @@ int qed_mcp_handle_events(struct qed_hwfn *p_hwfn, case MFW_DRV_MSG_TRANSCEIVER_STATE_CHANGE: qed_mcp_handle_transceiver_change(p_hwfn, p_ptt); break; + case MFW_DRV_MSG_ERROR_RECOVERY: + qed_mcp_handle_process_kill(p_hwfn, p_ptt); + break; case MFW_DRV_MSG_GET_LAN_STATS: case MFW_DRV_MSG_GET_FCOE_STATS: case MFW_DRV_MSG_GET_ISCSI_STATS: @@ -2324,6 +2381,43 @@ int qed_mcp_get_flash_size(struct qed_hwfn *p_hwfn, return 0; } +int qed_start_recovery_process(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) +{ + struct qed_dev *cdev = p_hwfn->cdev; + + if (cdev->recov_in_prog) { + DP_NOTICE(p_hwfn, + "Avoid triggering a recovery since such a process is already in progress\n"); + return -EAGAIN; + } + + DP_NOTICE(p_hwfn, "Triggering a recovery process\n"); + qed_wr(p_hwfn, p_ptt, MISC_REG_AEU_GENERAL_ATTN_35, 0x1); + + return 0; +} + +#define QED_RECOVERY_PROLOG_SLEEP_MS 100 + +int qed_recovery_prolog(struct qed_dev *cdev) +{ + struct qed_hwfn *p_hwfn = QED_LEADING_HWFN(cdev); + struct qed_ptt *p_ptt = p_hwfn->p_main_ptt; + int rc; + + /* Allow ongoing PCIe transactions to complete */ + msleep(QED_RECOVERY_PROLOG_SLEEP_MS); + + /* Clear the PF's internal FID_enable in the PXP */ + rc = qed_pglueb_set_pfid_enable(p_hwfn, p_ptt, false); + if (rc) + DP_NOTICE(p_hwfn, + "qed_pglueb_set_pfid_enable() failed. rc = %d.\n", + rc); + + return rc; +} + static int qed_mcp_config_vf_msix_bb(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u8 vf_id, u8 num) diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.h b/drivers/net/ethernet/qlogic/qed/qed_mcp.h index 387c5e649136..6e1d72a669ae 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.h +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.h @@ -440,6 +440,38 @@ qed_mcp_send_drv_version(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, struct qed_mcp_drv_version *p_ver); +/** + * @brief Read the MFW process kill counter + * + * @param p_hwfn + * @param p_ptt + * + * @return u32 + */ +u32 qed_get_process_kill_counter(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt); + +/** + * @brief Trigger a recovery process + * + * @param p_hwfn + * @param p_ptt + * + * @return int + */ +int qed_start_recovery_process(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt); + +/** + * @brief A recovery handler must call this function as its first step. + * It is assumed that the handler is not run from an interrupt context. + * + * @param cdev + * @param p_ptt + * + * @return int + */ +int qed_recovery_prolog(struct qed_dev *cdev); + /** * @brief Notify MFW about the change in base device properties * diff --git a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h index 8939ed6e08b7..5ce825ca5f24 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h +++ b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h @@ -518,6 +518,8 @@ 0x180824UL #define MISC_REG_AEU_GENERAL_ATTN_0 \ 0x008400UL +#define MISC_REG_AEU_GENERAL_ATTN_35 \ + 0x00848cUL #define CAU_REG_SB_ADDR_MEMORY \ 0x1c8000UL #define CAU_REG_SB_VAR_MEMORY \ diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c index eb88bbc6b193..3e0f7c46bb1b 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_spq.c +++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c @@ -790,6 +790,17 @@ static int qed_spq_pend_post(struct qed_hwfn *p_hwfn) SPQ_HIGH_PRI_RESERVE_DEFAULT); } +static void qed_spq_recov_set_ret_code(struct qed_spq_entry *p_ent, + u8 *fw_return_code) +{ + if (!fw_return_code) + return; + + if (p_ent->elem.hdr.protocol_id == PROTOCOLID_ROCE || + p_ent->elem.hdr.protocol_id == PROTOCOLID_IWARP) + *fw_return_code = RDMA_RETURN_OK; +} + /* Avoid overriding of SPQ entries when getting out-of-order completions, by * marking the completions in a bitmap and increasing the chain consumer only * for the first successive completed entries. @@ -825,6 +836,17 @@ int qed_spq_post(struct qed_hwfn *p_hwfn, return -EINVAL; } + if (p_hwfn->cdev->recov_in_prog) { + DP_VERBOSE(p_hwfn, + QED_MSG_SPQ, + "Recovery is in progress. Skip spq post [cmd %02x protocol %02x]\n", + p_ent->elem.hdr.cmd_id, p_ent->elem.hdr.protocol_id); + + /* Let the flow complete w/o any error handling */ + qed_spq_recov_set_ret_code(p_ent, fw_return_code); + return 0; + } + /* Complete the entry */ rc = qed_spq_fill_entry(p_hwfn, p_ent); diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index ca6290fa0f30..71e28be58102 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -4447,6 +4447,13 @@ int qed_sriov_disable(struct qed_dev *cdev, bool pci_enabled) if (cdev->p_iov_info && cdev->p_iov_info->num_vfs && pci_enabled) pci_disable_sriov(cdev->pdev); + if (cdev->recov_in_prog) { + DP_VERBOSE(cdev, + QED_MSG_IOV, + "Skip SRIOV disable operations in the device since a recovery is in progress\n"); + goto out; + } + for_each_hwfn(cdev, i) { struct qed_hwfn *hwfn = &cdev->hwfns[i]; struct qed_ptt *ptt = qed_ptt_acquire(hwfn); @@ -4486,7 +4493,7 @@ int qed_sriov_disable(struct qed_dev *cdev, bool pci_enabled) qed_ptt_release(hwfn, ptt); } - +out: qed_iov_set_vfs_to_disable(cdev, false); return 0; diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h index 5f818fda96bd..35170f74ed80 100644 --- a/include/linux/qed/qed_if.h +++ b/include/linux/qed/qed_if.h @@ -763,6 +763,7 @@ struct qed_probe_params { u32 dp_module; u8 dp_level; bool is_vf; + bool recov_in_prog; }; #define QED_DRV_VER_STR_SIZE 12 @@ -809,6 +810,7 @@ struct qed_common_cb_ops { void (*arfs_filter_op)(void *dev, void *fltr, u8 fw_rc); void (*link_update)(void *dev, struct qed_link_output *link); + void (*schedule_recovery_handler)(void *dev); void (*dcbx_aen)(void *dev, struct qed_dcbx_get *get, u32 mib_type); void (*get_generic_tlv_data)(void *dev, struct qed_generic_tlvs *data); void (*get_protocol_tlv_data)(void *dev, void *data); @@ -1056,6 +1058,24 @@ struct qed_common_ops { int (*db_recovery_del)(struct qed_dev *cdev, void __iomem *db_addr, void *db_data); +/** + * @brief recovery_process - Trigger a recovery process + * + * @param cdev + * + * @return 0 on success, error otherwise. + */ + int (*recovery_process)(struct qed_dev *cdev); + +/** + * @brief recovery_prolog - Execute the prolog operations of a recovery process + * + * @param cdev + * + * @return 0 on success, error otherwise. + */ + int (*recovery_prolog)(struct qed_dev *cdev); + /** * @brief update_drv_state - API to inform the change in the driver state. * -- cgit v1.2.3 From ccc67ef50b9085b895738d7720840eb6fe98745e Mon Sep 17 00:00:00 2001 From: Tomer Tayar Date: Mon, 28 Jan 2019 19:27:56 +0200 Subject: qede: Error recovery process This patch adds the error recovery process in the qede driver. The process includes a partial/customized driver unload and load, which allows it to look like a short suspend period to the kernel while preserving the net devices' state. Signed-off-by: Tomer Tayar Signed-off-by: Ariel Elior Signed-off-by: Michal Kalderon Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qede/qede.h | 3 + drivers/net/ethernet/qlogic/qede/qede_main.c | 292 ++++++++++++++++++++++----- drivers/net/ethernet/qlogic/qede/qede_rdma.c | 63 ++++-- include/linux/qed/qede_rdma.h | 10 +- 4 files changed, 294 insertions(+), 74 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h index 613249d1e967..843416404aeb 100644 --- a/drivers/net/ethernet/qlogic/qede/qede.h +++ b/drivers/net/ethernet/qlogic/qede/qede.h @@ -162,6 +162,7 @@ struct qede_rdma_dev { struct list_head entry; struct list_head rdma_event_list; struct workqueue_struct *rdma_wq; + bool exp_recovery; }; struct qede_ptp; @@ -264,6 +265,7 @@ struct qede_dev { enum QEDE_STATE { QEDE_STATE_CLOSED, QEDE_STATE_OPEN, + QEDE_STATE_RECOVERY, }; #define HILO_U64(hi, lo) ((((u64)(hi)) << 32) + (lo)) @@ -462,6 +464,7 @@ struct qede_fastpath { #define QEDE_CSUM_UNNECESSARY BIT(1) #define QEDE_TUNN_CSUM_UNNECESSARY BIT(2) +#define QEDE_SP_RECOVERY 0 #define QEDE_SP_RX_MODE 1 #ifdef CONFIG_RFS_ACCEL diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index 5a74fcbdbc2b..6b4d96635238 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -133,23 +133,12 @@ static int qede_probe(struct pci_dev *pdev, const struct pci_device_id *id); static void qede_remove(struct pci_dev *pdev); static void qede_shutdown(struct pci_dev *pdev); static void qede_link_update(void *dev, struct qed_link_output *link); +static void qede_schedule_recovery_handler(void *dev); +static void qede_recovery_handler(struct qede_dev *edev); static void qede_get_eth_tlv_data(void *edev, void *data); static void qede_get_generic_tlv_data(void *edev, struct qed_generic_tlvs *data); -/* The qede lock is used to protect driver state change and driver flows that - * are not reentrant. - */ -void __qede_lock(struct qede_dev *edev) -{ - mutex_lock(&edev->qede_lock); -} - -void __qede_unlock(struct qede_dev *edev) -{ - mutex_unlock(&edev->qede_lock); -} - #ifdef CONFIG_QED_SRIOV static int qede_set_vf_vlan(struct net_device *ndev, int vf, u16 vlan, u8 qos, __be16 vlan_proto) @@ -231,6 +220,7 @@ static struct qed_eth_cb_ops qede_ll_ops = { .arfs_filter_op = qede_arfs_filter_op, #endif .link_update = qede_link_update, + .schedule_recovery_handler = qede_schedule_recovery_handler, .get_generic_tlv_data = qede_get_generic_tlv_data, .get_protocol_tlv_data = qede_get_eth_tlv_data, }, @@ -950,11 +940,57 @@ err: return -ENOMEM; } +/* The qede lock is used to protect driver state change and driver flows that + * are not reentrant. + */ +void __qede_lock(struct qede_dev *edev) +{ + mutex_lock(&edev->qede_lock); +} + +void __qede_unlock(struct qede_dev *edev) +{ + mutex_unlock(&edev->qede_lock); +} + +/* This version of the lock should be used when acquiring the RTNL lock is also + * needed in addition to the internal qede lock. + */ +void qede_lock(struct qede_dev *edev) +{ + rtnl_lock(); + __qede_lock(edev); +} + +void qede_unlock(struct qede_dev *edev) +{ + __qede_unlock(edev); + rtnl_unlock(); +} + static void qede_sp_task(struct work_struct *work) { struct qede_dev *edev = container_of(work, struct qede_dev, sp_task.work); + /* The locking scheme depends on the specific flag: + * In case of QEDE_SP_RECOVERY, acquiring the RTNL lock is required to + * ensure that ongoing flows are ended and new ones are not started. + * In other cases - only the internal qede lock should be acquired. + */ + + if (test_and_clear_bit(QEDE_SP_RECOVERY, &edev->sp_flags)) { +#ifdef CONFIG_QED_SRIOV + /* SRIOV must be disabled outside the lock to avoid a deadlock. + * The recovery of the active VFs is currently not supported. + */ + qede_sriov_configure(edev->pdev, 0); +#endif + qede_lock(edev); + qede_recovery_handler(edev); + qede_unlock(edev); + } + __qede_lock(edev); if (test_and_clear_bit(QEDE_SP_RX_MODE, &edev->sp_flags)) @@ -1031,6 +1067,7 @@ static void qede_log_probe(struct qede_dev *edev) enum qede_probe_mode { QEDE_PROBE_NORMAL, + QEDE_PROBE_RECOVERY, }; static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, @@ -1051,6 +1088,7 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, probe_params.dp_module = dp_module; probe_params.dp_level = dp_level; probe_params.is_vf = is_vf; + probe_params.recov_in_prog = (mode == QEDE_PROBE_RECOVERY); cdev = qed_ops->common->probe(pdev, &probe_params); if (!cdev) { rc = -ENODEV; @@ -1078,11 +1116,20 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, if (rc) goto err2; - edev = qede_alloc_etherdev(cdev, pdev, &dev_info, dp_module, - dp_level); - if (!edev) { - rc = -ENOMEM; - goto err2; + if (mode != QEDE_PROBE_RECOVERY) { + edev = qede_alloc_etherdev(cdev, pdev, &dev_info, dp_module, + dp_level); + if (!edev) { + rc = -ENOMEM; + goto err2; + } + } else { + struct net_device *ndev = pci_get_drvdata(pdev); + + edev = netdev_priv(ndev); + edev->cdev = cdev; + memset(&edev->stats, 0, sizeof(edev->stats)); + memcpy(&edev->dev_info, &dev_info, sizeof(dev_info)); } if (is_vf) @@ -1090,28 +1137,31 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, qede_init_ndev(edev); - rc = qede_rdma_dev_add(edev); + rc = qede_rdma_dev_add(edev, (mode == QEDE_PROBE_RECOVERY)); if (rc) goto err3; - /* Prepare the lock prior to the registration of the netdev, - * as once it's registered we might reach flows requiring it - * [it's even possible to reach a flow needing it directly - * from there, although it's unlikely]. - */ - INIT_DELAYED_WORK(&edev->sp_task, qede_sp_task); - mutex_init(&edev->qede_lock); - rc = register_netdev(edev->ndev); - if (rc) { - DP_NOTICE(edev, "Cannot register net-device\n"); - goto err4; + if (mode != QEDE_PROBE_RECOVERY) { + /* Prepare the lock prior to the registration of the netdev, + * as once it's registered we might reach flows requiring it + * [it's even possible to reach a flow needing it directly + * from there, although it's unlikely]. + */ + INIT_DELAYED_WORK(&edev->sp_task, qede_sp_task); + mutex_init(&edev->qede_lock); + + rc = register_netdev(edev->ndev); + if (rc) { + DP_NOTICE(edev, "Cannot register net-device\n"); + goto err4; + } } edev->ops->common->set_name(cdev, edev->ndev->name); /* PTP not supported on VFs */ if (!is_vf) - qede_ptp_enable(edev, true); + qede_ptp_enable(edev, (mode == QEDE_PROBE_NORMAL)); edev->ops->register_ops(cdev, &qede_ll_ops, edev); @@ -1126,7 +1176,7 @@ static int __qede_probe(struct pci_dev *pdev, u32 dp_module, u8 dp_level, return 0; err4: - qede_rdma_dev_remove(edev); + qede_rdma_dev_remove(edev, (mode == QEDE_PROBE_RECOVERY)); err3: free_netdev(edev->ndev); err2: @@ -1162,6 +1212,7 @@ static int qede_probe(struct pci_dev *pdev, const struct pci_device_id *id) enum qede_remove_mode { QEDE_REMOVE_NORMAL, + QEDE_REMOVE_RECOVERY, }; static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode) @@ -1172,15 +1223,19 @@ static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode) DP_INFO(edev, "Starting qede_remove\n"); - qede_rdma_dev_remove(edev); - unregister_netdev(ndev); - cancel_delayed_work_sync(&edev->sp_task); + qede_rdma_dev_remove(edev, (mode == QEDE_REMOVE_RECOVERY)); - qede_ptp_disable(edev); + if (mode != QEDE_REMOVE_RECOVERY) { + unregister_netdev(ndev); - edev->ops->common->set_power_state(cdev, PCI_D0); + cancel_delayed_work_sync(&edev->sp_task); - pci_set_drvdata(pdev, NULL); + edev->ops->common->set_power_state(cdev, PCI_D0); + + pci_set_drvdata(pdev, NULL); + } + + qede_ptp_disable(edev); /* Use global ops since we've freed edev */ qed_ops->common->slowpath_stop(cdev); @@ -1194,7 +1249,8 @@ static void __qede_remove(struct pci_dev *pdev, enum qede_remove_mode mode) * [e.g., QED register callbacks] won't break anything when * accessing the netdevice. */ - free_netdev(ndev); + if (mode != QEDE_REMOVE_RECOVERY) + free_netdev(ndev); dev_info(&pdev->dev, "Ending qede_remove successfully\n"); } @@ -1539,6 +1595,58 @@ static int qede_alloc_mem_load(struct qede_dev *edev) return 0; } +static void qede_empty_tx_queue(struct qede_dev *edev, + struct qede_tx_queue *txq) +{ + unsigned int pkts_compl = 0, bytes_compl = 0; + struct netdev_queue *netdev_txq; + int rc, len = 0; + + netdev_txq = netdev_get_tx_queue(edev->ndev, txq->ndev_txq_id); + + while (qed_chain_get_cons_idx(&txq->tx_pbl) != + qed_chain_get_prod_idx(&txq->tx_pbl)) { + DP_VERBOSE(edev, NETIF_MSG_IFDOWN, + "Freeing a packet on tx queue[%d]: chain_cons 0x%x, chain_prod 0x%x\n", + txq->index, qed_chain_get_cons_idx(&txq->tx_pbl), + qed_chain_get_prod_idx(&txq->tx_pbl)); + + rc = qede_free_tx_pkt(edev, txq, &len); + if (rc) { + DP_NOTICE(edev, + "Failed to free a packet on tx queue[%d]: chain_cons 0x%x, chain_prod 0x%x\n", + txq->index, + qed_chain_get_cons_idx(&txq->tx_pbl), + qed_chain_get_prod_idx(&txq->tx_pbl)); + break; + } + + bytes_compl += len; + pkts_compl++; + txq->sw_tx_cons++; + } + + netdev_tx_completed_queue(netdev_txq, pkts_compl, bytes_compl); +} + +static void qede_empty_tx_queues(struct qede_dev *edev) +{ + int i; + + for_each_queue(i) + if (edev->fp_array[i].type & QEDE_FASTPATH_TX) { + int cos; + + for_each_cos_in_txq(edev, cos) { + struct qede_fastpath *fp; + + fp = &edev->fp_array[i]; + qede_empty_tx_queue(edev, + &fp->txq[cos]); + } + } +} + /* This function inits fp content and resets the SB, RXQ and TXQ structures */ static void qede_init_fp(struct qede_dev *edev) { @@ -2053,6 +2161,7 @@ out: enum qede_unload_mode { QEDE_UNLOAD_NORMAL, + QEDE_UNLOAD_RECOVERY, }; static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode, @@ -2068,7 +2177,8 @@ static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode, clear_bit(QEDE_FLAGS_LINK_REQUESTED, &edev->flags); - edev->state = QEDE_STATE_CLOSED; + if (mode != QEDE_UNLOAD_RECOVERY) + edev->state = QEDE_STATE_CLOSED; qede_rdma_dev_event_close(edev); @@ -2076,17 +2186,20 @@ static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode, netif_tx_disable(edev->ndev); netif_carrier_off(edev->ndev); - /* Reset the link */ - memset(&link_params, 0, sizeof(link_params)); - link_params.link_up = false; - edev->ops->common->set_link(edev->cdev, &link_params); - rc = qede_stop_queues(edev); - if (rc) { - qede_sync_free_irqs(edev); - goto out; - } + if (mode != QEDE_UNLOAD_RECOVERY) { + /* Reset the link */ + memset(&link_params, 0, sizeof(link_params)); + link_params.link_up = false; + edev->ops->common->set_link(edev->cdev, &link_params); - DP_INFO(edev, "Stopped Queues\n"); + rc = qede_stop_queues(edev); + if (rc) { + qede_sync_free_irqs(edev); + goto out; + } + + DP_INFO(edev, "Stopped Queues\n"); + } qede_vlan_mark_nonconfigured(edev); edev->ops->fastpath_stop(edev->cdev); @@ -2102,18 +2215,26 @@ static void qede_unload(struct qede_dev *edev, enum qede_unload_mode mode, qede_napi_disable_remove(edev); + if (mode == QEDE_UNLOAD_RECOVERY) + qede_empty_tx_queues(edev); + qede_free_mem_load(edev); qede_free_fp_array(edev); out: if (!is_locked) __qede_unlock(edev); + + if (mode != QEDE_UNLOAD_RECOVERY) + DP_NOTICE(edev, "Link is down\n"); + DP_INFO(edev, "Ending qede unload\n"); } enum qede_load_mode { QEDE_LOAD_NORMAL, QEDE_LOAD_RELOAD, + QEDE_LOAD_RECOVERY, }; static int qede_load(struct qede_dev *edev, enum qede_load_mode mode, @@ -2293,6 +2414,77 @@ static void qede_link_update(void *dev, struct qed_link_output *link) } } +static void qede_schedule_recovery_handler(void *dev) +{ + struct qede_dev *edev = dev; + + if (edev->state == QEDE_STATE_RECOVERY) { + DP_NOTICE(edev, + "Avoid scheduling a recovery handling since already in recovery state\n"); + return; + } + + set_bit(QEDE_SP_RECOVERY, &edev->sp_flags); + schedule_delayed_work(&edev->sp_task, 0); + + DP_INFO(edev, "Scheduled a recovery handler\n"); +} + +static void qede_recovery_failed(struct qede_dev *edev) +{ + netdev_err(edev->ndev, "Recovery handling has failed. Power cycle is needed.\n"); + + netif_device_detach(edev->ndev); + + if (edev->cdev) + edev->ops->common->set_power_state(edev->cdev, PCI_D3hot); +} + +static void qede_recovery_handler(struct qede_dev *edev) +{ + u32 curr_state = edev->state; + int rc; + + DP_NOTICE(edev, "Starting a recovery process\n"); + + /* No need to acquire first the qede_lock since is done by qede_sp_task + * before calling this function. + */ + edev->state = QEDE_STATE_RECOVERY; + + edev->ops->common->recovery_prolog(edev->cdev); + + if (curr_state == QEDE_STATE_OPEN) + qede_unload(edev, QEDE_UNLOAD_RECOVERY, true); + + __qede_remove(edev->pdev, QEDE_REMOVE_RECOVERY); + + rc = __qede_probe(edev->pdev, edev->dp_module, edev->dp_level, + IS_VF(edev), QEDE_PROBE_RECOVERY); + if (rc) { + edev->cdev = NULL; + goto err; + } + + if (curr_state == QEDE_STATE_OPEN) { + rc = qede_load(edev, QEDE_LOAD_RECOVERY, true); + if (rc) + goto err; + + qede_config_rx_mode(edev->ndev); + udp_tunnel_get_rx_info(edev->ndev); + } + + edev->state = curr_state; + + DP_NOTICE(edev, "Recovery handling is done\n"); + + return; + +err: + qede_recovery_failed(edev); +} + static bool qede_is_txq_full(struct qede_dev *edev, struct qede_tx_queue *txq) { struct netdev_queue *netdev_txq; diff --git a/drivers/net/ethernet/qlogic/qede/qede_rdma.c b/drivers/net/ethernet/qlogic/qede/qede_rdma.c index 1900bf7e67d1..ffabc2d2f082 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_rdma.c +++ b/drivers/net/ethernet/qlogic/qede/qede_rdma.c @@ -50,6 +50,8 @@ static void _qede_rdma_dev_add(struct qede_dev *edev) if (!qedr_drv) return; + /* Leftovers from previous error recovery */ + edev->rdma_info.exp_recovery = false; edev->rdma_info.qedr_dev = qedr_drv->add(edev->cdev, edev->pdev, edev->ndev); } @@ -87,21 +89,26 @@ static void qede_rdma_destroy_wq(struct qede_dev *edev) destroy_workqueue(edev->rdma_info.rdma_wq); } -int qede_rdma_dev_add(struct qede_dev *edev) +int qede_rdma_dev_add(struct qede_dev *edev, bool recovery) { - int rc = 0; + int rc; - if (qede_rdma_supported(edev)) { - rc = qede_rdma_create_wq(edev); - if (rc) - return rc; + if (!qede_rdma_supported(edev)) + return 0; - INIT_LIST_HEAD(&edev->rdma_info.entry); - mutex_lock(&qedr_dev_list_lock); - list_add_tail(&edev->rdma_info.entry, &qedr_dev_list); - _qede_rdma_dev_add(edev); - mutex_unlock(&qedr_dev_list_lock); - } + /* Cannot start qedr while recovering since it wasn't fully stopped */ + if (recovery) + return 0; + + rc = qede_rdma_create_wq(edev); + if (rc) + return rc; + + INIT_LIST_HEAD(&edev->rdma_info.entry); + mutex_lock(&qedr_dev_list_lock); + list_add_tail(&edev->rdma_info.entry, &qedr_dev_list); + _qede_rdma_dev_add(edev); + mutex_unlock(&qedr_dev_list_lock); return rc; } @@ -110,19 +117,30 @@ static void _qede_rdma_dev_remove(struct qede_dev *edev) { if (qedr_drv && qedr_drv->remove && edev->rdma_info.qedr_dev) qedr_drv->remove(edev->rdma_info.qedr_dev); - edev->rdma_info.qedr_dev = NULL; } -void qede_rdma_dev_remove(struct qede_dev *edev) +void qede_rdma_dev_remove(struct qede_dev *edev, bool recovery) { if (!qede_rdma_supported(edev)) return; - qede_rdma_destroy_wq(edev); - mutex_lock(&qedr_dev_list_lock); - _qede_rdma_dev_remove(edev); - list_del(&edev->rdma_info.entry); - mutex_unlock(&qedr_dev_list_lock); + /* Cannot remove qedr while recovering since it wasn't fully stopped */ + if (!recovery) { + qede_rdma_destroy_wq(edev); + mutex_lock(&qedr_dev_list_lock); + if (!edev->rdma_info.exp_recovery) + _qede_rdma_dev_remove(edev); + edev->rdma_info.qedr_dev = NULL; + list_del(&edev->rdma_info.entry); + mutex_unlock(&qedr_dev_list_lock); + } else { + if (!edev->rdma_info.exp_recovery) { + mutex_lock(&qedr_dev_list_lock); + _qede_rdma_dev_remove(edev); + mutex_unlock(&qedr_dev_list_lock); + } + edev->rdma_info.exp_recovery = true; + } } static void _qede_rdma_dev_open(struct qede_dev *edev) @@ -204,7 +222,8 @@ void qede_rdma_unregister_driver(struct qedr_driver *drv) mutex_lock(&qedr_dev_list_lock); list_for_each_entry(edev, &qedr_dev_list, rdma_info.entry) { - if (edev->rdma_info.qedr_dev) + /* If device has experienced recovery it was already removed */ + if (edev->rdma_info.qedr_dev && !edev->rdma_info.exp_recovery) _qede_rdma_dev_remove(edev); } qedr_drv = NULL; @@ -284,6 +303,10 @@ static void qede_rdma_add_event(struct qede_dev *edev, { struct qede_rdma_event_work *event_node; + /* If a recovery was experienced avoid adding the event */ + if (edev->rdma_info.exp_recovery) + return; + if (!edev->rdma_info.qedr_dev) return; diff --git a/include/linux/qed/qede_rdma.h b/include/linux/qed/qede_rdma.h index 9904617a9730..5a00c7a473bf 100644 --- a/include/linux/qed/qede_rdma.h +++ b/include/linux/qed/qede_rdma.h @@ -74,21 +74,23 @@ void qede_rdma_unregister_driver(struct qedr_driver *drv); bool qede_rdma_supported(struct qede_dev *dev); #if IS_ENABLED(CONFIG_QED_RDMA) -int qede_rdma_dev_add(struct qede_dev *dev); +int qede_rdma_dev_add(struct qede_dev *dev, bool recovery); void qede_rdma_dev_event_open(struct qede_dev *dev); void qede_rdma_dev_event_close(struct qede_dev *dev); -void qede_rdma_dev_remove(struct qede_dev *dev); +void qede_rdma_dev_remove(struct qede_dev *dev, bool recovery); void qede_rdma_event_changeaddr(struct qede_dev *edr); #else -static inline int qede_rdma_dev_add(struct qede_dev *dev) +static inline int qede_rdma_dev_add(struct qede_dev *dev, + bool recovery) { return 0; } static inline void qede_rdma_dev_event_open(struct qede_dev *dev) {} static inline void qede_rdma_dev_event_close(struct qede_dev *dev) {} -static inline void qede_rdma_dev_remove(struct qede_dev *dev) {} +static inline void qede_rdma_dev_remove(struct qede_dev *dev, + bool recovery) {} static inline void qede_rdma_event_changeaddr(struct qede_dev *edr) {} #endif #endif -- cgit v1.2.3 From c8aa703822bf811269975cf7251b5eaad4c38e9c Mon Sep 17 00:00:00 2001 From: Stanislav Fomichev Date: Mon, 28 Jan 2019 08:53:53 -0800 Subject: net/flow_dissector: move bpf case into __skb_flow_bpf_dissect This way, we can reuse it for flow dissector in BPF_PROG_TEST_RUN. No functional changes. Signed-off-by: Stanislav Fomichev Acked-by: Song Liu Signed-off-by: Daniel Borkmann --- include/linux/skbuff.h | 5 +++ net/core/flow_dissector.c | 92 +++++++++++++++++++++++++++-------------------- 2 files changed, 59 insertions(+), 38 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 93f56fddd92a..be762fc34ff3 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1221,6 +1221,11 @@ static inline int skb_flow_dissector_bpf_prog_detach(const union bpf_attr *attr) } #endif +struct bpf_flow_keys; +bool __skb_flow_bpf_dissect(struct bpf_prog *prog, + const struct sk_buff *skb, + struct flow_dissector *flow_dissector, + struct bpf_flow_keys *flow_keys); bool __skb_flow_dissect(const struct sk_buff *skb, struct flow_dissector *flow_dissector, void *target_container, diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c index 9f2840510e63..bb1a54747d64 100644 --- a/net/core/flow_dissector.c +++ b/net/core/flow_dissector.c @@ -683,6 +683,46 @@ static void __skb_flow_bpf_to_target(const struct bpf_flow_keys *flow_keys, } } +bool __skb_flow_bpf_dissect(struct bpf_prog *prog, + const struct sk_buff *skb, + struct flow_dissector *flow_dissector, + struct bpf_flow_keys *flow_keys) +{ + struct bpf_skb_data_end cb_saved; + struct bpf_skb_data_end *cb; + u32 result; + + /* Note that even though the const qualifier is discarded + * throughout the execution of the BPF program, all changes(the + * control block) are reverted after the BPF program returns. + * Therefore, __skb_flow_dissect does not alter the skb. + */ + + cb = (struct bpf_skb_data_end *)skb->cb; + + /* Save Control Block */ + memcpy(&cb_saved, cb, sizeof(cb_saved)); + memset(cb, 0, sizeof(*cb)); + + /* Pass parameters to the BPF program */ + memset(flow_keys, 0, sizeof(*flow_keys)); + cb->qdisc_cb.flow_keys = flow_keys; + flow_keys->nhoff = skb_network_offset(skb); + flow_keys->thoff = flow_keys->nhoff; + + bpf_compute_data_pointers((struct sk_buff *)skb); + result = BPF_PROG_RUN(prog, skb); + + /* Restore state */ + memcpy(cb, &cb_saved, sizeof(cb_saved)); + + flow_keys->nhoff = clamp_t(u16, flow_keys->nhoff, 0, skb->len); + flow_keys->thoff = clamp_t(u16, flow_keys->thoff, + flow_keys->nhoff, skb->len); + + return result == BPF_OK; +} + /** * __skb_flow_dissect - extract the flow_keys struct and return it * @skb: sk_buff to extract the flow from, can be NULL if the rest are specified @@ -714,7 +754,6 @@ bool __skb_flow_dissect(const struct sk_buff *skb, struct flow_dissector_key_vlan *key_vlan; enum flow_dissect_ret fdret; enum flow_dissector_key_id dissector_vlan = FLOW_DISSECTOR_KEY_MAX; - struct bpf_prog *attached = NULL; int num_hdrs = 0; u8 ip_proto = 0; bool ret; @@ -754,53 +793,30 @@ bool __skb_flow_dissect(const struct sk_buff *skb, FLOW_DISSECTOR_KEY_BASIC, target_container); - rcu_read_lock(); if (skb) { + struct bpf_flow_keys flow_keys; + struct bpf_prog *attached = NULL; + + rcu_read_lock(); + if (skb->dev) attached = rcu_dereference(dev_net(skb->dev)->flow_dissector_prog); else if (skb->sk) attached = rcu_dereference(sock_net(skb->sk)->flow_dissector_prog); else WARN_ON_ONCE(1); - } - if (attached) { - /* Note that even though the const qualifier is discarded - * throughout the execution of the BPF program, all changes(the - * control block) are reverted after the BPF program returns. - * Therefore, __skb_flow_dissect does not alter the skb. - */ - struct bpf_flow_keys flow_keys = {}; - struct bpf_skb_data_end cb_saved; - struct bpf_skb_data_end *cb; - u32 result; - - cb = (struct bpf_skb_data_end *)skb->cb; - - /* Save Control Block */ - memcpy(&cb_saved, cb, sizeof(cb_saved)); - memset(cb, 0, sizeof(cb_saved)); - /* Pass parameters to the BPF program */ - cb->qdisc_cb.flow_keys = &flow_keys; - flow_keys.nhoff = nhoff; - flow_keys.thoff = nhoff; - - bpf_compute_data_pointers((struct sk_buff *)skb); - result = BPF_PROG_RUN(attached, skb); - - /* Restore state */ - memcpy(cb, &cb_saved, sizeof(cb_saved)); - - flow_keys.nhoff = clamp_t(u16, flow_keys.nhoff, 0, skb->len); - flow_keys.thoff = clamp_t(u16, flow_keys.thoff, - flow_keys.nhoff, skb->len); - - __skb_flow_bpf_to_target(&flow_keys, flow_dissector, - target_container); + if (attached) { + ret = __skb_flow_bpf_dissect(attached, skb, + flow_dissector, + &flow_keys); + __skb_flow_bpf_to_target(&flow_keys, flow_dissector, + target_container); + rcu_read_unlock(); + return ret; + } rcu_read_unlock(); - return result == BPF_OK; } - rcu_read_unlock(); if (dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_ETH_ADDRS)) { -- cgit v1.2.3 From b7a1848e8398b8396c990279e6a10272d818577e Mon Sep 17 00:00:00 2001 From: Stanislav Fomichev Date: Mon, 28 Jan 2019 08:53:54 -0800 Subject: bpf: add BPF_PROG_TEST_RUN support for flow dissector The input is packet data, the output is struct bpf_flow_key. This should make it easy to test flow dissector programs without elaborate setup. Signed-off-by: Stanislav Fomichev Acked-by: Song Liu Signed-off-by: Daniel Borkmann --- include/linux/bpf.h | 3 ++ net/bpf/test_run.c | 82 +++++++++++++++++++++++++++++++++++++++++++++++++++++ net/core/filter.c | 1 + 3 files changed, 86 insertions(+) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 3851529062ec..0394f1f9213b 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -404,6 +404,9 @@ int bpf_prog_test_run_xdp(struct bpf_prog *prog, const union bpf_attr *kattr, union bpf_attr __user *uattr); int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr, union bpf_attr __user *uattr); +int bpf_prog_test_run_flow_dissector(struct bpf_prog *prog, + const union bpf_attr *kattr, + union bpf_attr __user *uattr); /* an array of programs to be executed under rcu_lock. * diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c index fa2644d276ef..2c5172b33209 100644 --- a/net/bpf/test_run.c +++ b/net/bpf/test_run.c @@ -240,3 +240,85 @@ out: kfree(data); return ret; } + +int bpf_prog_test_run_flow_dissector(struct bpf_prog *prog, + const union bpf_attr *kattr, + union bpf_attr __user *uattr) +{ + u32 size = kattr->test.data_size_in; + u32 repeat = kattr->test.repeat; + struct bpf_flow_keys flow_keys; + u64 time_start, time_spent = 0; + struct bpf_skb_data_end *cb; + u32 retval, duration; + struct sk_buff *skb; + struct sock *sk; + void *data; + int ret; + u32 i; + + if (prog->type != BPF_PROG_TYPE_FLOW_DISSECTOR) + return -EINVAL; + + data = bpf_test_init(kattr, size, NET_SKB_PAD + NET_IP_ALIGN, + SKB_DATA_ALIGN(sizeof(struct skb_shared_info))); + if (IS_ERR(data)) + return PTR_ERR(data); + + sk = kzalloc(sizeof(*sk), GFP_USER); + if (!sk) { + kfree(data); + return -ENOMEM; + } + sock_net_set(sk, current->nsproxy->net_ns); + sock_init_data(NULL, sk); + + skb = build_skb(data, 0); + if (!skb) { + kfree(data); + kfree(sk); + return -ENOMEM; + } + skb->sk = sk; + + skb_reserve(skb, NET_SKB_PAD + NET_IP_ALIGN); + __skb_put(skb, size); + skb->protocol = eth_type_trans(skb, + current->nsproxy->net_ns->loopback_dev); + skb_reset_network_header(skb); + + cb = (struct bpf_skb_data_end *)skb->cb; + cb->qdisc_cb.flow_keys = &flow_keys; + + if (!repeat) + repeat = 1; + + time_start = ktime_get_ns(); + for (i = 0; i < repeat; i++) { + preempt_disable(); + rcu_read_lock(); + retval = __skb_flow_bpf_dissect(prog, skb, + &flow_keys_dissector, + &flow_keys); + rcu_read_unlock(); + preempt_enable(); + + if (need_resched()) { + if (signal_pending(current)) + break; + time_spent += ktime_get_ns() - time_start; + cond_resched(); + time_start = ktime_get_ns(); + } + } + time_spent += ktime_get_ns() - time_start; + do_div(time_spent, repeat); + duration = time_spent > U32_MAX ? U32_MAX : (u32)time_spent; + + ret = bpf_test_finish(kattr, uattr, &flow_keys, sizeof(flow_keys), + retval, duration); + + kfree_skb(skb); + kfree(sk); + return ret; +} diff --git a/net/core/filter.c b/net/core/filter.c index 8e587dd1da20..8ce421796ac6 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -7711,6 +7711,7 @@ const struct bpf_verifier_ops flow_dissector_verifier_ops = { }; const struct bpf_prog_ops flow_dissector_prog_ops = { + .test_run = bpf_prog_test_run_flow_dissector, }; int sk_detach_filter(struct sock *sk) -- cgit v1.2.3