From 6e5f52674ff0756e61a8879f6232b9ac33735cba Mon Sep 17 00:00:00 2001 From: Baruch Siach Date: Sat, 25 Jan 2014 22:36:13 +0200 Subject: spi: spi.h: clarify the documentation of transfer_one Explicitly note the transfer_one and transfer_one_message are mutually exclusive, to make the text a little more newcomers friendly. Signed-off-by: Baruch Siach Signed-off-by: Mark Brown --- include/linux/spi/spi.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h index a1d4ca290862..ad050f0dd39d 100644 --- a/include/linux/spi/spi.h +++ b/include/linux/spi/spi.h @@ -287,7 +287,10 @@ static inline void spi_unregister_driver(struct spi_driver *sdrv) * - return 1 if the transfer is still in progress. When * the driver is finished with this transfer it must * call spi_finalize_current_transfer() so the subsystem - * can issue the next transfer + * can issue the next transfer. Note: transfer_one and + * transfer_one_message are mutually exclusive; when both + * are set, the generic subsystem does not call your + * transfer_one callback. * @unprepare_message: undo any work done by prepare_message(). * @cs_gpios: Array of GPIOs to use as chip select lines; one per CS * number. Any individual value may be -ENOENT for CS lines that -- cgit v1.2.3 From e9305331f64ab9e5210baa4307355635c5e849f4 Mon Sep 17 00:00:00 2001 From: Baruch Siach Date: Sat, 25 Jan 2014 22:36:15 +0200 Subject: spi: correct the transfer_one_message documentation wording The transfer_one_message callback handles messages, not transfers. Signed-off-by: Baruch Siach Signed-off-by: Mark Brown --- Documentation/spi/spi-summary | 2 +- include/linux/spi/spi.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/Documentation/spi/spi-summary b/Documentation/spi/spi-summary index dcaad476789c..7982bcc4d151 100644 --- a/Documentation/spi/spi-summary +++ b/Documentation/spi/spi-summary @@ -543,7 +543,7 @@ SPI MASTER METHODS queuing transfers that arrive in the meantime. When the driver is finished with this message, it must call spi_finalize_current_message() so the subsystem can issue the next - transfer. This may sleep. + message. This may sleep. master->transfer_one(struct spi_master *master, struct spi_device *spi, struct spi_transfer *transfer) diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h index ad050f0dd39d..4203c66d8803 100644 --- a/include/linux/spi/spi.h +++ b/include/linux/spi/spi.h @@ -273,7 +273,7 @@ static inline void spi_unregister_driver(struct spi_driver *sdrv) * message while queuing transfers that arrive in the meantime. When the * driver is finished with this message, it must call * spi_finalize_current_message() so the subsystem can issue the next - * transfer + * message * @unprepare_transfer_hardware: there are currently no more messages on the * queue so the subsystem notifies the driver that it may relax the * hardware by issuing this call -- cgit v1.2.3 From f0276924fa35a3607920a58cf5d878212824b951 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Tue, 31 Dec 2013 11:38:50 +0800 Subject: blk-mq: Don't reserve a tag for flush request Reserving a tag (request) for flush to avoid dead lock is a overkill. A tag is valuable resource. We can track the number of flush requests and disallow having too many pending flush requests allocated. With this patch, blk_mq_alloc_request_pinned() could do a busy nop (but not a dead loop) if too many pending requests are allocated and new flush request is allocated. But this should not be a problem, too many pending flush requests are very rare case. I verified this can fix the deadlock caused by too many pending flush requests. Signed-off-by: Shaohua Li Signed-off-by: Jens Axboe --- block/blk-flush.c | 8 +++++--- block/blk-mq.c | 46 ++++++++++++++++++++++++++++++---------------- include/linux/blk-mq.h | 3 +++ 3 files changed, 38 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/block/blk-flush.c b/block/blk-flush.c index 9288aaf35c21..9143e85226c7 100644 --- a/block/blk-flush.c +++ b/block/blk-flush.c @@ -284,9 +284,8 @@ static void mq_flush_work(struct work_struct *work) q = container_of(work, struct request_queue, mq_flush_work); - /* We don't need set REQ_FLUSH_SEQ, it's for consistency */ rq = blk_mq_alloc_request(q, WRITE_FLUSH|REQ_FLUSH_SEQ, - __GFP_WAIT|GFP_ATOMIC, true); + __GFP_WAIT|GFP_ATOMIC, false); rq->cmd_type = REQ_TYPE_FS; rq->end_io = flush_end_io; @@ -408,8 +407,11 @@ void blk_insert_flush(struct request *rq) /* * @policy now records what operations need to be done. Adjust * REQ_FLUSH and FUA for the driver. + * We keep REQ_FLUSH for mq to track flush requests. For !FUA, + * we never dispatch the request directly. */ - rq->cmd_flags &= ~REQ_FLUSH; + if (rq->cmd_flags & REQ_FUA) + rq->cmd_flags &= ~REQ_FLUSH; if (!(fflags & REQ_FUA)) rq->cmd_flags &= ~REQ_FUA; diff --git a/block/blk-mq.c b/block/blk-mq.c index 57039fcd9c93..9072d0ab184f 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -194,9 +194,27 @@ static void blk_mq_rq_ctx_init(struct request_queue *q, struct blk_mq_ctx *ctx, } static struct request *__blk_mq_alloc_request(struct blk_mq_hw_ctx *hctx, - gfp_t gfp, bool reserved) + gfp_t gfp, bool reserved, + int rw) { - return blk_mq_alloc_rq(hctx, gfp, reserved); + struct request *req; + bool is_flush = false; + /* + * flush need allocate a request, leave at least one request for + * non-flush IO to avoid deadlock + */ + if ((rw & REQ_FLUSH) && !(rw & REQ_FLUSH_SEQ)) { + if (atomic_inc_return(&hctx->pending_flush) >= + hctx->queue_depth - hctx->reserved_tags - 1) { + atomic_dec(&hctx->pending_flush); + return NULL; + } + is_flush = true; + } + req = blk_mq_alloc_rq(hctx, gfp, reserved); + if (!req && is_flush) + atomic_dec(&hctx->pending_flush); + return req; } static struct request *blk_mq_alloc_request_pinned(struct request_queue *q, @@ -209,7 +227,7 @@ static struct request *blk_mq_alloc_request_pinned(struct request_queue *q, struct blk_mq_ctx *ctx = blk_mq_get_ctx(q); struct blk_mq_hw_ctx *hctx = q->mq_ops->map_queue(q, ctx->cpu); - rq = __blk_mq_alloc_request(hctx, gfp & ~__GFP_WAIT, reserved); + rq = __blk_mq_alloc_request(hctx, gfp & ~__GFP_WAIT, reserved, rw); if (rq) { blk_mq_rq_ctx_init(q, ctx, rq, rw); break; @@ -272,6 +290,9 @@ static void __blk_mq_free_request(struct blk_mq_hw_ctx *hctx, const int tag = rq->tag; struct request_queue *q = rq->q; + if ((rq->cmd_flags & REQ_FLUSH) && !(rq->cmd_flags & REQ_FLUSH_SEQ)) + atomic_dec(&hctx->pending_flush); + blk_mq_rq_init(hctx, rq); blk_mq_put_tag(hctx->tags, tag); @@ -900,14 +921,14 @@ static void blk_mq_make_request(struct request_queue *q, struct bio *bio) hctx = q->mq_ops->map_queue(q, ctx->cpu); trace_block_getrq(q, bio, rw); - rq = __blk_mq_alloc_request(hctx, GFP_ATOMIC, false); + rq = __blk_mq_alloc_request(hctx, GFP_ATOMIC, false, bio->bi_rw); if (likely(rq)) - blk_mq_rq_ctx_init(q, ctx, rq, rw); + blk_mq_rq_ctx_init(q, ctx, rq, bio->bi_rw); else { blk_mq_put_ctx(ctx); trace_block_sleeprq(q, bio, rw); - rq = blk_mq_alloc_request_pinned(q, rw, __GFP_WAIT|GFP_ATOMIC, - false); + rq = blk_mq_alloc_request_pinned(q, bio->bi_rw, + __GFP_WAIT|GFP_ATOMIC, false); ctx = rq->mq_ctx; hctx = q->mq_ops->map_queue(q, ctx->cpu); } @@ -1184,7 +1205,9 @@ static int blk_mq_init_hw_queues(struct request_queue *q, hctx->queue_num = i; hctx->flags = reg->flags; hctx->queue_depth = reg->queue_depth; + hctx->reserved_tags = reg->reserved_tags; hctx->cmd_size = reg->cmd_size; + atomic_set(&hctx->pending_flush, 0); blk_mq_init_cpu_notifier(&hctx->cpu_notifier, blk_mq_hctx_notify, hctx); @@ -1309,15 +1332,6 @@ struct request_queue *blk_mq_init_queue(struct blk_mq_reg *reg, reg->queue_depth = BLK_MQ_MAX_DEPTH; } - /* - * Set aside a tag for flush requests. It will only be used while - * another flush request is in progress but outside the driver. - * - * TODO: only allocate if flushes are supported - */ - reg->queue_depth++; - reg->reserved_tags++; - if (reg->queue_depth < (reg->reserved_tags + BLK_MQ_TAG_MIN)) return ERR_PTR(-EINVAL); diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 161b23105b1e..1e8f16f65af4 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -36,12 +36,15 @@ struct blk_mq_hw_ctx { struct list_head page_list; struct blk_mq_tags *tags; + atomic_t pending_flush; + unsigned long queued; unsigned long run; #define BLK_MQ_MAX_DISPATCH_ORDER 10 unsigned long dispatched[BLK_MQ_MAX_DISPATCH_ORDER]; unsigned int queue_depth; + unsigned int reserved_tags; unsigned int numa_node; unsigned int cmd_size; /* per-request extra data */ -- cgit v1.2.3 From 48b8f6315a6a230f859fcae8ef4a0ccf4f48e802 Mon Sep 17 00:00:00 2001 From: Jani Nikula Date: Fri, 31 Jan 2014 15:49:07 +0200 Subject: drm: add DRM_INFO_ONCE() to print a one-time DRM_INFO() message Just like DRM_INFO(), but only do it once. Signed-off-by: Jani Nikula Reviewed-by: Chris Wilson Acked-by: Dave Airlie Signed-off-by: Daniel Vetter --- include/drm/drmP.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include') diff --git a/include/drm/drmP.h b/include/drm/drmP.h index 04086c5be930..04a7f31301f8 100644 --- a/include/drm/drmP.h +++ b/include/drm/drmP.h @@ -199,6 +199,9 @@ int drm_err(const char *func, const char *format, ...); #define DRM_INFO(fmt, ...) \ printk(KERN_INFO "[" DRM_NAME "] " fmt, ##__VA_ARGS__) +#define DRM_INFO_ONCE(fmt, ...) \ + printk_once(KERN_INFO "[" DRM_NAME "] " fmt, ##__VA_ARGS__) + /** * Debug output. * -- cgit v1.2.3 From 788a4d56ff378bff0b8e685d03a962b36903a149 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Tue, 4 Feb 2014 18:33:12 +0100 Subject: drivers: phy: Add support for optional phys Add devm_phy_optional_get and phy_optional_get, which should be used when the phy is optional. They does not return an error when the phy does not exist, rather they returns NULL, which is considered as a valid phy, but results in NOPs when used with the consumer API. Signed-off-by: Andrew Lunn Tested-by: Gregory CLEMENT Acked-by: Kishon Vijay Abraham I Signed-off-by: Jason Cooper --- Documentation/phy.txt | 20 +++++++++++++------- drivers/phy/phy-core.c | 45 +++++++++++++++++++++++++++++++++++++++++++++ include/linux/phy/phy.h | 14 ++++++++++++++ 3 files changed, 72 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/Documentation/phy.txt b/Documentation/phy.txt index 2e24b993e95f..ebff6ee52441 100644 --- a/Documentation/phy.txt +++ b/Documentation/phy.txt @@ -75,14 +75,20 @@ Before the controller can make use of the PHY, it has to get a reference to it. This framework provides the following APIs to get a reference to the PHY. struct phy *phy_get(struct device *dev, const char *string); +struct phy *phy_optional_get(struct device *dev, const char *string); struct phy *devm_phy_get(struct device *dev, const char *string); - -phy_get and devm_phy_get can be used to get the PHY. In the case of dt boot, -the string arguments should contain the phy name as given in the dt data and -in the case of non-dt boot, it should contain the label of the PHY. -The only difference between the two APIs is that devm_phy_get associates the -device with the PHY using devres on successful PHY get. On driver detach, -release function is invoked on the the devres data and devres data is freed. +struct phy *devm_phy_optional_get(struct device *dev, const char *string); + +phy_get, phy_optional_get, devm_phy_get and devm_phy_optional_get can +be used to get the PHY. In the case of dt boot, the string arguments +should contain the phy name as given in the dt data and in the case of +non-dt boot, it should contain the label of the PHY. The two +devm_phy_get associates the device with the PHY using devres on +successful PHY get. On driver detach, release function is invoked on +the the devres data and devres data is freed. phy_optional_get and +devm_phy_optional_get should be used when the phy is optional. These +two functions will never return -ENODEV, but instead returns NULL when +the phy cannot be found. It should be noted that NULL is a valid phy reference. All phy consumer calls on the NULL phy become NOPs. That is the release calls, diff --git a/drivers/phy/phy-core.c b/drivers/phy/phy-core.c index a9cdeee20d91..5f5b0f4be5be 100644 --- a/drivers/phy/phy-core.c +++ b/drivers/phy/phy-core.c @@ -425,6 +425,27 @@ struct phy *phy_get(struct device *dev, const char *string) } EXPORT_SYMBOL_GPL(phy_get); +/** + * phy_optional_get() - lookup and obtain a reference to an optional phy. + * @dev: device that requests this phy + * @string: the phy name as given in the dt data or the name of the controller + * port for non-dt case + * + * Returns the phy driver, after getting a refcount to it; or + * NULL if there is no such phy. The caller is responsible for + * calling phy_put() to release that count. + */ +struct phy *phy_optional_get(struct device *dev, const char *string) +{ + struct phy *phy = phy_get(dev, string); + + if (PTR_ERR(phy) == -ENODEV) + phy = NULL; + + return phy; +} +EXPORT_SYMBOL_GPL(phy_optional_get); + /** * devm_phy_get() - lookup and obtain a reference to a phy. * @dev: device that requests this phy @@ -455,6 +476,30 @@ struct phy *devm_phy_get(struct device *dev, const char *string) } EXPORT_SYMBOL_GPL(devm_phy_get); +/** + * devm_phy_optional_get() - lookup and obtain a reference to an optional phy. + * @dev: device that requests this phy + * @string: the phy name as given in the dt data or phy device name + * for non-dt case + * + * Gets the phy using phy_get(), and associates a device with it using + * devres. On driver detach, release function is invoked on the devres + * data, then, devres data is freed. This differs to devm_phy_get() in + * that if the phy does not exist, it is not considered an error and + * -ENODEV will not be returned. Instead the NULL phy is returned, + * which can be passed to all other phy consumer calls. + */ +struct phy *devm_phy_optional_get(struct device *dev, const char *string) +{ + struct phy *phy = devm_phy_get(dev, string); + + if (PTR_ERR(phy) == -ENODEV) + phy = NULL; + + return phy; +} +EXPORT_SYMBOL_GPL(devm_phy_optional_get); + /** * phy_create() - create a new phy * @dev: device that is creating the new phy diff --git a/include/linux/phy/phy.h b/include/linux/phy/phy.h index e273e5ac19c9..3f83459dbb20 100644 --- a/include/linux/phy/phy.h +++ b/include/linux/phy/phy.h @@ -146,7 +146,9 @@ static inline void phy_set_bus_width(struct phy *phy, int bus_width) phy->attrs.bus_width = bus_width; } struct phy *phy_get(struct device *dev, const char *string); +struct phy *phy_optional_get(struct device *dev, const char *string); struct phy *devm_phy_get(struct device *dev, const char *string); +struct phy *devm_phy_optional_get(struct device *dev, const char *string); void phy_put(struct phy *phy); void devm_phy_put(struct device *dev, struct phy *phy); struct phy *of_phy_simple_xlate(struct device *dev, @@ -232,11 +234,23 @@ static inline struct phy *phy_get(struct device *dev, const char *string) return ERR_PTR(-ENOSYS); } +static inline struct phy *phy_optional_get(struct device *dev, + const char *string) +{ + return ERR_PTR(-ENOSYS); +} + static inline struct phy *devm_phy_get(struct device *dev, const char *string) { return ERR_PTR(-ENOSYS); } +static inline struct phy *devm_phy_optional_get(struct device *dev, + const char *string) +{ + return ERR_PTR(-ENOSYS); +} + static inline void phy_put(struct phy *phy) { } -- cgit v1.2.3 From 662372e42e46d9bbfcb83e1cce81f6b33cebaddd Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 3 Feb 2014 08:53:44 -0600 Subject: of: restructure for_each macros to fix compile warnings Commit 00b2c76a6a "include/linux/of.h: make for_each_child_of_node() reference its args when CONFIG_OF=n" fixed warnings for unused variables, but introduced variable "used uninitialized" warnings. Simply initializing the variables would result in "set but not used" warnings with W=1. Fix both types of warnings by making all the for_each macros unconditional and rely on the dummy static inline functions to initialize and reference any variables. Reported-by: Geert Uytterhoeven Cc: David Howells Signed-off-by: Rob Herring Acked-by: Grant Likely --- include/linux/of.h | 153 +++++++++++++++++++++++++++++------------------------ 1 file changed, 84 insertions(+), 69 deletions(-) (limited to 'include') diff --git a/include/linux/of.h b/include/linux/of.h index 70c64ba17fa5..435cb995904d 100644 --- a/include/linux/of.h +++ b/include/linux/of.h @@ -169,35 +169,15 @@ static inline const char *of_node_full_name(const struct device_node *np) extern struct device_node *of_find_node_by_name(struct device_node *from, const char *name); -#define for_each_node_by_name(dn, name) \ - for (dn = of_find_node_by_name(NULL, name); dn; \ - dn = of_find_node_by_name(dn, name)) extern struct device_node *of_find_node_by_type(struct device_node *from, const char *type); -#define for_each_node_by_type(dn, type) \ - for (dn = of_find_node_by_type(NULL, type); dn; \ - dn = of_find_node_by_type(dn, type)) extern struct device_node *of_find_compatible_node(struct device_node *from, const char *type, const char *compat); -#define for_each_compatible_node(dn, type, compatible) \ - for (dn = of_find_compatible_node(NULL, type, compatible); dn; \ - dn = of_find_compatible_node(dn, type, compatible)) extern struct device_node *of_find_matching_node_and_match( struct device_node *from, const struct of_device_id *matches, const struct of_device_id **match); -static inline struct device_node *of_find_matching_node( - struct device_node *from, - const struct of_device_id *matches) -{ - return of_find_matching_node_and_match(from, matches, NULL); -} -#define for_each_matching_node(dn, matches) \ - for (dn = of_find_matching_node(NULL, matches); dn; \ - dn = of_find_matching_node(dn, matches)) -#define for_each_matching_node_and_match(dn, matches, match) \ - for (dn = of_find_matching_node_and_match(NULL, matches, match); \ - dn; dn = of_find_matching_node_and_match(dn, matches, match)) + extern struct device_node *of_find_node_by_path(const char *path); extern struct device_node *of_find_node_by_phandle(phandle handle); extern struct device_node *of_get_parent(const struct device_node *node); @@ -209,43 +189,11 @@ extern struct device_node *of_get_next_available_child( extern struct device_node *of_get_child_by_name(const struct device_node *node, const char *name); -#define for_each_child_of_node(parent, child) \ - for (child = of_get_next_child(parent, NULL); child != NULL; \ - child = of_get_next_child(parent, child)) - -#define for_each_available_child_of_node(parent, child) \ - for (child = of_get_next_available_child(parent, NULL); child != NULL; \ - child = of_get_next_available_child(parent, child)) - -static inline int of_get_child_count(const struct device_node *np) -{ - struct device_node *child; - int num = 0; - - for_each_child_of_node(np, child) - num++; - - return num; -} - -static inline int of_get_available_child_count(const struct device_node *np) -{ - struct device_node *child; - int num = 0; - - for_each_available_child_of_node(np, child) - num++; - - return num; -} /* cache lookup */ extern struct device_node *of_find_next_cache_node(const struct device_node *); extern struct device_node *of_find_node_with_property( struct device_node *from, const char *prop_name); -#define for_each_node_with_property(dn, prop_name) \ - for (dn = of_find_node_with_property(NULL, prop_name); dn; \ - dn = of_find_node_with_property(dn, prop_name)) extern struct property *of_find_property(const struct device_node *np, const char *name, @@ -367,42 +315,53 @@ static inline struct device_node *of_find_node_by_name(struct device_node *from, return NULL; } -static inline struct device_node *of_get_parent(const struct device_node *node) +static inline struct device_node *of_find_node_by_type(struct device_node *from, + const char *type) { return NULL; } -static inline bool of_have_populated_dt(void) +static inline struct device_node *of_find_matching_node_and_match( + struct device_node *from, + const struct of_device_id *matches, + const struct of_device_id **match) { - return false; + return NULL; } -/* Kill an unused variable warning on a device_node pointer */ -static inline void __of_use_dn(const struct device_node *np) +static inline struct device_node *of_get_parent(const struct device_node *node) { + return NULL; } -#define for_each_child_of_node(parent, child) \ - while (__of_use_dn(parent), __of_use_dn(child), 0) +static inline struct device_node *of_get_next_child( + const struct device_node *node, struct device_node *prev) +{ + return NULL; +} -#define for_each_available_child_of_node(parent, child) \ - while (0) +static inline struct device_node *of_get_next_available_child( + const struct device_node *node, struct device_node *prev) +{ + return NULL; +} -static inline struct device_node *of_get_child_by_name( - const struct device_node *node, - const char *name) +static inline struct device_node *of_find_node_with_property( + struct device_node *from, const char *prop_name) { return NULL; } -static inline int of_get_child_count(const struct device_node *np) +static inline bool of_have_populated_dt(void) { - return 0; + return false; } -static inline int of_get_available_child_count(const struct device_node *np) +static inline struct device_node *of_get_child_by_name( + const struct device_node *node, + const char *name) { - return 0; + return NULL; } static inline int of_device_is_compatible(const struct device_node *device, @@ -569,6 +528,13 @@ extern int of_node_to_nid(struct device_node *np); static inline int of_node_to_nid(struct device_node *device) { return 0; } #endif +static inline struct device_node *of_find_matching_node( + struct device_node *from, + const struct of_device_id *matches) +{ + return of_find_matching_node_and_match(from, matches, NULL); +} + /** * of_property_read_bool - Findfrom a property * @np: device node from which the property value is to be read. @@ -618,6 +584,55 @@ static inline int of_property_read_u32(const struct device_node *np, s; \ s = of_prop_next_string(prop, s)) +#define for_each_node_by_name(dn, name) \ + for (dn = of_find_node_by_name(NULL, name); dn; \ + dn = of_find_node_by_name(dn, name)) +#define for_each_node_by_type(dn, type) \ + for (dn = of_find_node_by_type(NULL, type); dn; \ + dn = of_find_node_by_type(dn, type)) +#define for_each_compatible_node(dn, type, compatible) \ + for (dn = of_find_compatible_node(NULL, type, compatible); dn; \ + dn = of_find_compatible_node(dn, type, compatible)) +#define for_each_matching_node(dn, matches) \ + for (dn = of_find_matching_node(NULL, matches); dn; \ + dn = of_find_matching_node(dn, matches)) +#define for_each_matching_node_and_match(dn, matches, match) \ + for (dn = of_find_matching_node_and_match(NULL, matches, match); \ + dn; dn = of_find_matching_node_and_match(dn, matches, match)) + +#define for_each_child_of_node(parent, child) \ + for (child = of_get_next_child(parent, NULL); child != NULL; \ + child = of_get_next_child(parent, child)) +#define for_each_available_child_of_node(parent, child) \ + for (child = of_get_next_available_child(parent, NULL); child != NULL; \ + child = of_get_next_available_child(parent, child)) + +#define for_each_node_with_property(dn, prop_name) \ + for (dn = of_find_node_with_property(NULL, prop_name); dn; \ + dn = of_find_node_with_property(dn, prop_name)) + +static inline int of_get_child_count(const struct device_node *np) +{ + struct device_node *child; + int num = 0; + + for_each_child_of_node(np, child) + num++; + + return num; +} + +static inline int of_get_available_child_count(const struct device_node *np) +{ + struct device_node *child; + int num = 0; + + for_each_available_child_of_node(np, child) + num++; + + return num; +} + #if defined(CONFIG_PROC_FS) && defined(CONFIG_PROC_DEVICETREE) extern void proc_device_tree_add_node(struct device_node *, struct proc_dir_entry *); extern void proc_device_tree_add_prop(struct proc_dir_entry *pde, struct property *prop); -- cgit v1.2.3 From 1db73ae39a97797b7d929eca2af469f82fbb337a Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 3 Feb 2014 09:27:40 +0100 Subject: of/device: Nullify match table in of_match_device() for CONFIG_OF=n If the of_device_id table inside a device driver is protected by #ifdef CONFIG_OF, the driver still has to provide a dummy declaration of the table, or wrap it inside of_match_ptr(), when calling of_match_device() in the CONFIG_OF=n case, else the driver fails to compile with e.g. drivers/spi/spi-rspi.c: In function 'rspi_probe': drivers/spi/spi-rspi.c:1203:26: error: 'rspi_of_match' undeclared (first use in this function) drivers/spi/spi-rspi.c:1203:26: note: each undeclared identifier is reported only once for each function it appears in Make of_match_device() nullify the table pointer if CONFIG_OF=n to fix this. Reported-by: Yoshihiro Shimoda Signed-off-by: Geert Uytterhoeven Signed-off-by: Rob Herring --- include/linux/of_device.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/of_device.h b/include/linux/of_device.h index 8d7dd6768cb7..ef370210ffb2 100644 --- a/include/linux/of_device.h +++ b/include/linux/of_device.h @@ -78,11 +78,13 @@ static inline int of_device_uevent_modalias(struct device *dev, static inline void of_device_node_put(struct device *dev) { } -static inline const struct of_device_id *of_match_device( +static inline const struct of_device_id *__of_match_device( const struct of_device_id *matches, const struct device *dev) { return NULL; } +#define of_match_device(matches, dev) \ + __of_match_device(of_match_ptr(matches), (dev)) static inline struct device_node *of_cpu_device_node_get(int cpu) { -- cgit v1.2.3 From a3485d088516be4c2ae2890e8ad64321481f2857 Mon Sep 17 00:00:00 2001 From: Lars-Peter Clausen Date: Tue, 4 Feb 2014 13:42:10 +0100 Subject: gpio: consumer.h: Move forward declarations outside #ifdef Make sure that the forward declared structs in gpio/consumer.h are also visible on the else branch of the CONFIG_GPIOLIB #ifdef. Fixes the following warnings and their associated errors when CONFIG_GPIOLIB is not selected: include/linux/gpio/consumer.h:67:14: warning: 'struct device' declared inside parameter list include/linux/gpio/consumer.h:67:14: warning: its scope is only this definition or declaration, which is probably not what you want [...] Signed-off-by: Lars-Peter Clausen Reviewed-by: Alexandre Courbot Signed-off-by: Linus Walleij --- include/linux/gpio/consumer.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/gpio/consumer.h b/include/linux/gpio/consumer.h index 4d34dbbbad4d..7a8144fef406 100644 --- a/include/linux/gpio/consumer.h +++ b/include/linux/gpio/consumer.h @@ -4,8 +4,6 @@ #include #include -#ifdef CONFIG_GPIOLIB - struct device; struct gpio_chip; @@ -18,6 +16,8 @@ struct gpio_chip; */ struct gpio_desc; +#ifdef CONFIG_GPIOLIB + /* Acquire and dispose GPIOs */ struct gpio_desc *__must_check gpiod_get(struct device *dev, const char *con_id); -- cgit v1.2.3 From 78c0f98cc9dd46824fa66f35f14ea24ba733d145 Mon Sep 17 00:00:00 2001 From: Eli Cohen Date: Thu, 30 Jan 2014 13:49:48 +0200 Subject: IB/mlx5: Fix binary compatibility with libmlx5 Commit c1be5232d21d ("Fix micro UAR allocator") broke binary compatibility between libmlx5 and mlx5_ib since it defines a different value to the number of micro UARs per page, leading to wrong calculation in libmlx5. This patch defines struct mlx5_ib_alloc_ucontext_req_v2 as an extension to struct mlx5_ib_alloc_ucontext_req. The extended size is determined in mlx5_ib_alloc_ucontext() and in case of old library we use uuarn 0 which works fine -- this is acheived due to create_user_qp() falling back from high to medium then to low class where low class will return 0. For new libraries we use the more sophisticated allocation algorithm. Signed-off-by: Eli Cohen Reviewed-by: Yann Droneaud Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx5/main.c | 19 +++++++++++++++++-- drivers/infiniband/hw/mlx5/qp.c | 10 ++++++++-- drivers/infiniband/hw/mlx5/user.h | 7 +++++++ include/linux/mlx5/driver.h | 1 + 4 files changed, 33 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 9660d093f8cf..f4ef4a24d410 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -536,24 +536,38 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev, struct ib_udata *udata) { struct mlx5_ib_dev *dev = to_mdev(ibdev); - struct mlx5_ib_alloc_ucontext_req req; + struct mlx5_ib_alloc_ucontext_req_v2 req; struct mlx5_ib_alloc_ucontext_resp resp; struct mlx5_ib_ucontext *context; struct mlx5_uuar_info *uuari; struct mlx5_uar *uars; int gross_uuars; int num_uars; + int ver; int uuarn; int err; int i; + int reqlen; if (!dev->ib_active) return ERR_PTR(-EAGAIN); - err = ib_copy_from_udata(&req, udata, sizeof(req)); + memset(&req, 0, sizeof(req)); + reqlen = udata->inlen - sizeof(struct ib_uverbs_cmd_hdr); + if (reqlen == sizeof(struct mlx5_ib_alloc_ucontext_req)) + ver = 0; + else if (reqlen == sizeof(struct mlx5_ib_alloc_ucontext_req_v2)) + ver = 2; + else + return ERR_PTR(-EINVAL); + + err = ib_copy_from_udata(&req, udata, reqlen); if (err) return ERR_PTR(err); + if (req.flags || req.reserved) + return ERR_PTR(-EINVAL); + if (req.total_num_uuars > MLX5_MAX_UUARS) return ERR_PTR(-ENOMEM); @@ -626,6 +640,7 @@ static struct ib_ucontext *mlx5_ib_alloc_ucontext(struct ib_device *ibdev, if (err) goto out_uars; + uuari->ver = ver; uuari->num_low_latency_uuars = req.num_low_latency_uuars; uuari->uars = uars; uuari->num_uars = num_uars; diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index 492dc330e907..091576a777e9 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -430,11 +430,17 @@ static int alloc_uuar(struct mlx5_uuar_info *uuari, break; case MLX5_IB_LATENCY_CLASS_MEDIUM: - uuarn = alloc_med_class_uuar(uuari); + if (uuari->ver < 2) + uuarn = -ENOMEM; + else + uuarn = alloc_med_class_uuar(uuari); break; case MLX5_IB_LATENCY_CLASS_HIGH: - uuarn = alloc_high_class_uuar(uuari); + if (uuari->ver < 2) + uuarn = -ENOMEM; + else + uuarn = alloc_high_class_uuar(uuari); break; case MLX5_IB_LATENCY_CLASS_FAST_PATH: diff --git a/drivers/infiniband/hw/mlx5/user.h b/drivers/infiniband/hw/mlx5/user.h index 32a2a5dfc523..0f4f8e42a17f 100644 --- a/drivers/infiniband/hw/mlx5/user.h +++ b/drivers/infiniband/hw/mlx5/user.h @@ -62,6 +62,13 @@ struct mlx5_ib_alloc_ucontext_req { __u32 num_low_latency_uuars; }; +struct mlx5_ib_alloc_ucontext_req_v2 { + __u32 total_num_uuars; + __u32 num_low_latency_uuars; + __u32 flags; + __u32 reserved; +}; + struct mlx5_ib_alloc_ucontext_resp { __u32 qp_tab_size; __u32 bf_reg_size; diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 554548cd3dd4..32cb18c399c2 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -227,6 +227,7 @@ struct mlx5_uuar_info { * protect uuar allocation data structs */ struct mutex lock; + u32 ver; }; struct mlx5_bf { -- cgit v1.2.3 From e28bab4828354583bb66ac09021ca69b341a7db4 Mon Sep 17 00:00:00 2001 From: "K. Y. Srinivasan" Date: Wed, 15 Jan 2014 17:12:58 -0800 Subject: Drivers: hv: vmbus: Specify the target CPU that should receive notification During the initial VMBUS connect phase, starting with WS2012 R2, we should specify the VPCU in the guest that should receive the notification. Fix this issue. This fix is required to properly connect to the host in the kexeced kernel. Signed-off-by: K. Y. Srinivasan Cc: [3.9+] Signed-off-by: Greg Kroah-Hartman --- drivers/hv/connection.c | 2 ++ include/linux/hyperv.h | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/drivers/hv/connection.c b/drivers/hv/connection.c index af6edf9b1936..855bbda9964d 100644 --- a/drivers/hv/connection.c +++ b/drivers/hv/connection.c @@ -78,6 +78,8 @@ static int vmbus_negotiate_version(struct vmbus_channel_msginfo *msginfo, msg->interrupt_page = virt_to_phys(vmbus_connection.int_page); msg->monitor_page1 = virt_to_phys(vmbus_connection.monitor_pages[0]); msg->monitor_page2 = virt_to_phys(vmbus_connection.monitor_pages[1]); + if (version == VERSION_WIN8) + msg->target_vcpu = hv_context.vp_index[smp_processor_id()]; /* * Add to list before we send the request since we may diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h index 15da677478dd..344883dce584 100644 --- a/include/linux/hyperv.h +++ b/include/linux/hyperv.h @@ -875,7 +875,7 @@ struct vmbus_channel_relid_released { struct vmbus_channel_initiate_contact { struct vmbus_channel_message_header header; u32 vmbus_version_requested; - u32 padding2; + u32 target_vcpu; /* The VCPU the host should respond to */ u64 interrupt_page; u64 monitor_page1; u64 monitor_page2; -- cgit v1.2.3 From 80bfa2f6e2e81049fc6cd3bfaeedcb64db3a9ba6 Mon Sep 17 00:00:00 2001 From: Roger Pau Monne Date: Tue, 4 Feb 2014 11:26:15 +0100 Subject: xen-blkif: drop struct blkif_request_segment_aligned MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This was wrongly introduced in commit 402b27f9, the only difference between blkif_request_segment_aligned and blkif_request_segment is that the former has a named padding, while both share the same memory layout. Also correct a few minor glitches in the description, including for it to no longer assume PAGE_SIZE == 4096. Signed-off-by: Roger Pau Monné [Description fix by Jan Beulich] Signed-off-by: Jan Beulich Reported-by: Jan Beulich Cc: Konrad Rzeszutek Wilk Cc: David Vrabel Cc: Boris Ostrovsky Tested-by: Matt Rushton Cc: Matt Wilson Signed-off-by: Konrad Rzeszutek Wilk --- drivers/block/xen-blkback/blkback.c | 2 +- drivers/block/xen-blkback/common.h | 2 +- drivers/block/xen-blkfront.c | 6 +++--- include/xen/interface/io/blkif.h | 34 ++++++++++++++-------------------- 4 files changed, 19 insertions(+), 25 deletions(-) (limited to 'include') diff --git a/drivers/block/xen-blkback/blkback.c b/drivers/block/xen-blkback/blkback.c index 394fa2eabf87..e612627ae981 100644 --- a/drivers/block/xen-blkback/blkback.c +++ b/drivers/block/xen-blkback/blkback.c @@ -847,7 +847,7 @@ static int xen_blkbk_parse_indirect(struct blkif_request *req, struct grant_page **pages = pending_req->indirect_pages; struct xen_blkif *blkif = pending_req->blkif; int indirect_grefs, rc, n, nseg, i; - struct blkif_request_segment_aligned *segments = NULL; + struct blkif_request_segment *segments = NULL; nseg = pending_req->nr_pages; indirect_grefs = INDIRECT_PAGES(nseg); diff --git a/drivers/block/xen-blkback/common.h b/drivers/block/xen-blkback/common.h index e40326a7f707..9eb34e24b4fe 100644 --- a/drivers/block/xen-blkback/common.h +++ b/drivers/block/xen-blkback/common.h @@ -57,7 +57,7 @@ #define MAX_INDIRECT_SEGMENTS 256 #define SEGS_PER_INDIRECT_FRAME \ - (PAGE_SIZE/sizeof(struct blkif_request_segment_aligned)) + (PAGE_SIZE/sizeof(struct blkif_request_segment)) #define MAX_INDIRECT_PAGES \ ((MAX_INDIRECT_SEGMENTS + SEGS_PER_INDIRECT_FRAME - 1)/SEGS_PER_INDIRECT_FRAME) #define INDIRECT_PAGES(_segs) \ diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index c4a4c9006288..7d09dfc9735c 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -162,7 +162,7 @@ static DEFINE_SPINLOCK(minor_lock); #define DEV_NAME "xvd" /* name in /dev */ #define SEGS_PER_INDIRECT_FRAME \ - (PAGE_SIZE/sizeof(struct blkif_request_segment_aligned)) + (PAGE_SIZE/sizeof(struct blkif_request_segment)) #define INDIRECT_GREFS(_segs) \ ((_segs + SEGS_PER_INDIRECT_FRAME - 1)/SEGS_PER_INDIRECT_FRAME) @@ -393,7 +393,7 @@ static int blkif_queue_request(struct request *req) unsigned long id; unsigned int fsect, lsect; int i, ref, n; - struct blkif_request_segment_aligned *segments = NULL; + struct blkif_request_segment *segments = NULL; /* * Used to store if we are able to queue the request by just using @@ -550,7 +550,7 @@ static int blkif_queue_request(struct request *req) } else { n = i % SEGS_PER_INDIRECT_FRAME; segments[n] = - (struct blkif_request_segment_aligned) { + (struct blkif_request_segment) { .gref = ref, .first_sect = fsect, .last_sect = lsect }; diff --git a/include/xen/interface/io/blkif.h b/include/xen/interface/io/blkif.h index ae665ac59c36..32ec05a6572f 100644 --- a/include/xen/interface/io/blkif.h +++ b/include/xen/interface/io/blkif.h @@ -113,13 +113,13 @@ typedef uint64_t blkif_sector_t; * it's less than the number provided by the backend. The indirect_grefs field * in blkif_request_indirect should be filled by the frontend with the * grant references of the pages that are holding the indirect segments. - * This pages are filled with an array of blkif_request_segment_aligned - * that hold the information about the segments. The number of indirect - * pages to use is determined by the maximum number of segments - * a indirect request contains. Every indirect page can contain a maximum - * of 512 segments (PAGE_SIZE/sizeof(blkif_request_segment_aligned)), - * so to calculate the number of indirect pages to use we have to do - * ceil(indirect_segments/512). + * These pages are filled with an array of blkif_request_segment that hold the + * information about the segments. The number of indirect pages to use is + * determined by the number of segments an indirect request contains. Every + * indirect page can contain a maximum of + * (PAGE_SIZE / sizeof(struct blkif_request_segment)) segments, so to + * calculate the number of indirect pages to use we have to do + * ceil(indirect_segments / (PAGE_SIZE / sizeof(struct blkif_request_segment))). * * If a backend does not recognize BLKIF_OP_INDIRECT, it should *not* * create the "feature-max-indirect-segments" node! @@ -135,13 +135,12 @@ typedef uint64_t blkif_sector_t; #define BLKIF_MAX_INDIRECT_PAGES_PER_REQUEST 8 -struct blkif_request_segment_aligned { - grant_ref_t gref; /* reference to I/O buffer frame */ - /* @first_sect: first sector in frame to transfer (inclusive). */ - /* @last_sect: last sector in frame to transfer (inclusive). */ - uint8_t first_sect, last_sect; - uint16_t _pad; /* padding to make it 8 bytes, so it's cache-aligned */ -} __attribute__((__packed__)); +struct blkif_request_segment { + grant_ref_t gref; /* reference to I/O buffer frame */ + /* @first_sect: first sector in frame to transfer (inclusive). */ + /* @last_sect: last sector in frame to transfer (inclusive). */ + uint8_t first_sect, last_sect; +}; struct blkif_request_rw { uint8_t nr_segments; /* number of segments */ @@ -151,12 +150,7 @@ struct blkif_request_rw { #endif uint64_t id; /* private guest value, echoed in resp */ blkif_sector_t sector_number;/* start sector idx on disk (r/w only) */ - struct blkif_request_segment { - grant_ref_t gref; /* reference to I/O buffer frame */ - /* @first_sect: first sector in frame to transfer (inclusive). */ - /* @last_sect: last sector in frame to transfer (inclusive). */ - uint8_t first_sect, last_sect; - } seg[BLKIF_MAX_SEGMENTS_PER_REQUEST]; + struct blkif_request_segment seg[BLKIF_MAX_SEGMENTS_PER_REQUEST]; } __attribute__((__packed__)); struct blkif_request_discard { -- cgit v1.2.3 From 72a0a36e2854a6eadb4cf2561858f613f9cd4639 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 7 Feb 2014 10:22:36 -0800 Subject: blk-mq: support at_head inserations for blk_execute_rq This is neede for proper SG_IO operation as well as various uses of blk_execute_rq from the SCSI midlayer. Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-exec.c | 2 +- block/blk-mq.c | 17 ++++++++++------- include/linux/blk-mq.h | 3 ++- 3 files changed, 13 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/block/blk-exec.c b/block/blk-exec.c index bbfc072a79c2..c68613bb4c79 100644 --- a/block/blk-exec.c +++ b/block/blk-exec.c @@ -65,7 +65,7 @@ void blk_execute_rq_nowait(struct request_queue *q, struct gendisk *bd_disk, * be resued after dying flag is set */ if (q->mq_ops) { - blk_mq_insert_request(q, rq, true); + blk_mq_insert_request(q, rq, at_head, true); return; } diff --git a/block/blk-mq.c b/block/blk-mq.c index 9072d0ab184f..c9306e3403fe 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -714,13 +714,16 @@ static void blk_mq_work_fn(struct work_struct *work) } static void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx, - struct request *rq) + struct request *rq, bool at_head) { struct blk_mq_ctx *ctx = rq->mq_ctx; trace_block_rq_insert(hctx->queue, rq); - list_add_tail(&rq->queuelist, &ctx->rq_list); + if (at_head) + list_add(&rq->queuelist, &ctx->rq_list); + else + list_add_tail(&rq->queuelist, &ctx->rq_list); blk_mq_hctx_mark_pending(hctx, ctx); /* @@ -730,7 +733,7 @@ static void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx, } void blk_mq_insert_request(struct request_queue *q, struct request *rq, - bool run_queue) + bool at_head, bool run_queue) { struct blk_mq_hw_ctx *hctx; struct blk_mq_ctx *ctx, *current_ctx; @@ -749,7 +752,7 @@ void blk_mq_insert_request(struct request_queue *q, struct request *rq, rq->mq_ctx = ctx; } spin_lock(&ctx->lock); - __blk_mq_insert_request(hctx, rq); + __blk_mq_insert_request(hctx, rq, at_head); spin_unlock(&ctx->lock); blk_mq_put_ctx(current_ctx); @@ -781,7 +784,7 @@ void blk_mq_run_request(struct request *rq, bool run_queue, bool async) /* ctx->cpu might be offline */ spin_lock(&ctx->lock); - __blk_mq_insert_request(hctx, rq); + __blk_mq_insert_request(hctx, rq, false); spin_unlock(&ctx->lock); blk_mq_put_ctx(current_ctx); @@ -819,7 +822,7 @@ static void blk_mq_insert_requests(struct request_queue *q, rq = list_first_entry(list, struct request, queuelist); list_del_init(&rq->queuelist); rq->mq_ctx = ctx; - __blk_mq_insert_request(hctx, rq); + __blk_mq_insert_request(hctx, rq, false); } spin_unlock(&ctx->lock); @@ -971,7 +974,7 @@ static void blk_mq_make_request(struct request_queue *q, struct bio *bio) __blk_mq_free_request(hctx, ctx, rq); else { blk_mq_bio_to_request(rq, bio); - __blk_mq_insert_request(hctx, rq); + __blk_mq_insert_request(hctx, rq, false); } spin_unlock(&ctx->lock); diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 1e8f16f65af4..b7638be58599 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -122,7 +122,8 @@ void blk_mq_init_commands(struct request_queue *, void (*init)(void *data, struc void blk_mq_flush_plug_list(struct blk_plug *plug, bool from_schedule); -void blk_mq_insert_request(struct request_queue *, struct request *, bool); +void blk_mq_insert_request(struct request_queue *, struct request *, + bool, bool); void blk_mq_run_queues(struct request_queue *q, bool async); void blk_mq_free_request(struct request *rq); bool blk_mq_can_queue(struct blk_mq_hw_ctx *); -- cgit v1.2.3 From 3d4b81eda2211f32886e2978daf6f39885042fc4 Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Fri, 31 Jan 2014 11:52:57 -0800 Subject: Revert "usb: xhci: Link TRB must not occur within a USB payload burst" This reverts commit 35773dac5f862cb1c82ea151eba3e2f6de51ec3e. It's a hack that caused regressions in the usb-storage and userspace USB drivers that use usbfs and libusb. Commit 70cabb7d992f "xhci 1.0: Limit arbitrarily-aligned scatter gather." should fix the issues seen with the ax88179_178a driver on xHCI 1.0 hosts, without causing regressions. Signed-off-by: Sarah Sharp Cc: stable@vger.kernel.org # 3.12 --- drivers/usb/host/xhci-ring.c | 54 ++------------------------------------------ include/linux/usb.h | 2 -- 2 files changed, 2 insertions(+), 54 deletions(-) (limited to 'include') diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index 0272450a1667..0ed64eb68e48 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -2967,58 +2967,8 @@ static int prepare_ring(struct xhci_hcd *xhci, struct xhci_ring *ep_ring, } while (1) { - if (room_on_ring(xhci, ep_ring, num_trbs)) { - union xhci_trb *trb = ep_ring->enqueue; - unsigned int usable = ep_ring->enq_seg->trbs + - TRBS_PER_SEGMENT - 1 - trb; - u32 nop_cmd; - - /* - * Section 4.11.7.1 TD Fragments states that a link - * TRB must only occur at the boundary between - * data bursts (eg 512 bytes for 480M). - * While it is possible to split a large fragment - * we don't know the size yet. - * Simplest solution is to fill the trb before the - * LINK with nop commands. - */ - if (num_trbs == 1 || num_trbs <= usable || usable == 0) - break; - - if (ep_ring->type != TYPE_BULK) - /* - * While isoc transfers might have a buffer that - * crosses a 64k boundary it is unlikely. - * Since we can't add NOPs without generating - * gaps in the traffic just hope it never - * happens at the end of the ring. - * This could be fixed by writing a LINK TRB - * instead of the first NOP - however the - * TRB_TYPE_LINK_LE32() calls would all need - * changing to check the ring length. - */ - break; - - if (num_trbs >= TRBS_PER_SEGMENT) { - xhci_err(xhci, "Too many fragments %d, max %d\n", - num_trbs, TRBS_PER_SEGMENT - 1); - return -ENOMEM; - } - - nop_cmd = cpu_to_le32(TRB_TYPE(TRB_TR_NOOP) | - ep_ring->cycle_state); - ep_ring->num_trbs_free -= usable; - do { - trb->generic.field[0] = 0; - trb->generic.field[1] = 0; - trb->generic.field[2] = 0; - trb->generic.field[3] = nop_cmd; - trb++; - } while (--usable); - ep_ring->enqueue = trb; - if (room_on_ring(xhci, ep_ring, num_trbs)) - break; - } + if (room_on_ring(xhci, ep_ring, num_trbs)) + break; if (ep_ring == xhci->cmd_ring) { xhci_err(xhci, "Do not support expand command ring\n"); diff --git a/include/linux/usb.h b/include/linux/usb.h index c716da18c668..7f6eb859873e 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -1265,8 +1265,6 @@ typedef void (*usb_complete_t)(struct urb *); * @sg: scatter gather buffer list, the buffer size of each element in * the list (except the last) must be divisible by the endpoint's * max packet size if no_sg_constraint isn't set in 'struct usb_bus' - * (FIXME: scatter-gather under xHCI is broken for periodic transfers. - * Do not use urb->sg for interrupt endpoints for now, only bulk.) * @num_mapped_sgs: (internal) number of mapped sg entries * @num_sgs: number of entries in the sg list * @transfer_buffer_length: How big is transfer_buffer. The transfer may -- cgit v1.2.3 From 3b1cc9b9622a022208ec95b1259b05bbdf712eb7 Mon Sep 17 00:00:00 2001 From: Sudeep Dutt Date: Mon, 3 Feb 2014 14:53:19 -0800 Subject: misc: mic: fix possible signed underflow (undefined behavior) in userspace API iovcnt is declared as a signed integer in both the userspace API and as a local variable in mic_virtio.c. The while() loop in mic_virtio.c iterates until the local variable iovcnt reaches the value 0. If userspace passes e.g. INT_MIN as iovcnt field, this loop then appears to depend on an undefined behavior (signed underflow) to complete. The fix is to use unsigned integers in both the userspace API and the local variable. This issue was reported @ https://lkml.org/lkml/2014/1/10/10 Reported-by: Mathieu Desnoyers Reviewed-by: Ashutosh Dixit Signed-off-by: Sudeep Dutt Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/misc/mic/host/mic_virtio.c | 3 ++- include/uapi/linux/mic_ioctl.h | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/misc/mic/host/mic_virtio.c b/drivers/misc/mic/host/mic_virtio.c index 752ff873f891..7e1ef0ebbb80 100644 --- a/drivers/misc/mic/host/mic_virtio.c +++ b/drivers/misc/mic/host/mic_virtio.c @@ -156,7 +156,8 @@ static int mic_vringh_copy(struct mic_vdev *mvdev, struct vringh_kiov *iov, static int _mic_virtio_copy(struct mic_vdev *mvdev, struct mic_copy_desc *copy) { - int ret = 0, iovcnt = copy->iovcnt; + int ret = 0; + u32 iovcnt = copy->iovcnt; struct iovec iov; struct iovec __user *u_iov = copy->iov; void __user *ubuf = NULL; diff --git a/include/uapi/linux/mic_ioctl.h b/include/uapi/linux/mic_ioctl.h index 7fabba5059cf..feb0b4c0814c 100644 --- a/include/uapi/linux/mic_ioctl.h +++ b/include/uapi/linux/mic_ioctl.h @@ -39,7 +39,7 @@ struct mic_copy_desc { #else struct iovec *iov; #endif - int iovcnt; + __u32 iovcnt; __u8 vr_idx; __u8 update_used; __u32 out_len; -- cgit v1.2.3 From 0668d3065128d39449c097e62dbdb5707820137d Mon Sep 17 00:00:00 2001 From: Stephen Boyd Date: Thu, 2 Jan 2014 16:37:32 -0800 Subject: genirq: Add devm_request_any_context_irq() Some drivers use request_any_context_irq() but there isn't a devm_* function for it. Add one so that these drivers don't need to explicitly free the irq on driver detach. Signed-off-by: Stephen Boyd Cc: linux-arm-kernel@lists.infradead.org Cc: Dmitry Torokhov Link: http://lkml.kernel.org/r/1388709460-19222-3-git-send-email-sboyd@codeaurora.org Signed-off-by: Thomas Gleixner --- include/linux/interrupt.h | 5 +++++ kernel/irq/devres.c | 45 +++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 50 insertions(+) (limited to 'include') diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index 0053adde0ed9..a2678d35b5a2 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -158,6 +158,11 @@ devm_request_irq(struct device *dev, unsigned int irq, irq_handler_t handler, devname, dev_id); } +extern int __must_check +devm_request_any_context_irq(struct device *dev, unsigned int irq, + irq_handler_t handler, unsigned long irqflags, + const char *devname, void *dev_id); + extern void devm_free_irq(struct device *dev, unsigned int irq, void *dev_id); /* diff --git a/kernel/irq/devres.c b/kernel/irq/devres.c index bd8e788d71e0..1ef0606797c9 100644 --- a/kernel/irq/devres.c +++ b/kernel/irq/devres.c @@ -72,6 +72,51 @@ int devm_request_threaded_irq(struct device *dev, unsigned int irq, } EXPORT_SYMBOL(devm_request_threaded_irq); +/** + * devm_request_any_context_irq - allocate an interrupt line for a managed device + * @dev: device to request interrupt for + * @irq: Interrupt line to allocate + * @handler: Function to be called when the IRQ occurs + * @thread_fn: function to be called in a threaded interrupt context. NULL + * for devices which handle everything in @handler + * @irqflags: Interrupt type flags + * @devname: An ascii name for the claiming device + * @dev_id: A cookie passed back to the handler function + * + * Except for the extra @dev argument, this function takes the + * same arguments and performs the same function as + * request_any_context_irq(). IRQs requested with this function will be + * automatically freed on driver detach. + * + * If an IRQ allocated with this function needs to be freed + * separately, devm_free_irq() must be used. + */ +int devm_request_any_context_irq(struct device *dev, unsigned int irq, + irq_handler_t handler, unsigned long irqflags, + const char *devname, void *dev_id) +{ + struct irq_devres *dr; + int rc; + + dr = devres_alloc(devm_irq_release, sizeof(struct irq_devres), + GFP_KERNEL); + if (!dr) + return -ENOMEM; + + rc = request_any_context_irq(irq, handler, irqflags, devname, dev_id); + if (rc) { + devres_free(dr); + return rc; + } + + dr->irq = irq; + dr->dev_id = dev_id; + devres_add(dev, dr); + + return 0; +} +EXPORT_SYMBOL(devm_request_any_context_irq); + /** * devm_free_irq - free an interrupt * @dev: device to free interrupt for -- cgit v1.2.3 From c4540a7d8c1e595560e53acedf88901daf15a2b5 Mon Sep 17 00:00:00 2001 From: Rashika Kheria Date: Sun, 9 Feb 2014 18:30:39 +0530 Subject: fs: Add prototype declaration to appropriate header file include/linux/bio.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add prototype declaration to header file include/linux/bio.h because it is used by more than one file. This eliminates the following warning in bio-integrity.c: fs/bio-integrity.c:214:14: warning: no previous prototype for ‘bio_integrity_tag_size’ [-Wmissing-prototypes] Signed-off-by: Rashika Kheria Reviewed-by: Josh Triplett Signed-off-by: Jens Axboe --- include/linux/bio.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/bio.h b/include/linux/bio.h index 70654521dab6..d6791bba8264 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -332,6 +332,7 @@ extern struct bio *bio_clone_fast(struct bio *, gfp_t, struct bio_set *); extern struct bio *bio_clone_bioset(struct bio *, gfp_t, struct bio_set *bs); extern struct bio_set *fs_bio_set; +unsigned int bio_integrity_tag_size(struct bio *bio); static inline struct bio *bio_alloc(gfp_t gfp_mask, unsigned int nr_iovecs) { -- cgit v1.2.3 From 30a91cb4ef385fe1b260df204ef314d86fff2850 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 10 Feb 2014 03:24:38 -0800 Subject: blk-mq: rework I/O completions Rework I/O completions to work more like the old code path. blk_mq_end_io now stays out of the business of deferring completions to others CPUs and calling blk_mark_rq_complete. The latter is very important to allow completing requests that have timed out and thus are already marked completed, the former allows using the IPI callout even for driver specific completions instead of having to reimplement them. Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-mq.c | 52 ++++++++++++++++++++++++++++++-------------------- block/blk-mq.h | 3 +-- block/blk-timeout.c | 2 +- include/linux/blk-mq.h | 4 ++++ 4 files changed, 37 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/block/blk-mq.c b/block/blk-mq.c index cee96234bf58..14c8f35946e1 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -326,7 +326,7 @@ static void blk_mq_bio_endio(struct request *rq, struct bio *bio, int error) bio_endio(bio, error); } -void blk_mq_complete_request(struct request *rq, int error) +void blk_mq_end_io(struct request *rq, int error) { struct bio *bio = rq->bio; unsigned int bytes = 0; @@ -351,46 +351,53 @@ void blk_mq_complete_request(struct request *rq, int error) else blk_mq_free_request(rq); } +EXPORT_SYMBOL(blk_mq_end_io); -void __blk_mq_end_io(struct request *rq, int error) -{ - if (!blk_mark_rq_complete(rq)) - blk_mq_complete_request(rq, error); -} - -static void blk_mq_end_io_remote(void *data) +static void __blk_mq_complete_request_remote(void *data) { struct request *rq = data; - __blk_mq_end_io(rq, rq->errors); + rq->q->softirq_done_fn(rq); } -/* - * End IO on this request on a multiqueue enabled driver. We'll either do - * it directly inline, or punt to a local IPI handler on the matching - * remote CPU. - */ -void blk_mq_end_io(struct request *rq, int error) +void __blk_mq_complete_request(struct request *rq) { struct blk_mq_ctx *ctx = rq->mq_ctx; int cpu; - if (!ctx->ipi_redirect) - return __blk_mq_end_io(rq, error); + if (!ctx->ipi_redirect) { + rq->q->softirq_done_fn(rq); + return; + } cpu = get_cpu(); if (cpu != ctx->cpu && cpu_online(ctx->cpu)) { - rq->errors = error; - rq->csd.func = blk_mq_end_io_remote; + rq->csd.func = __blk_mq_complete_request_remote; rq->csd.info = rq; rq->csd.flags = 0; __smp_call_function_single(ctx->cpu, &rq->csd, 0); } else { - __blk_mq_end_io(rq, error); + rq->q->softirq_done_fn(rq); } put_cpu(); } -EXPORT_SYMBOL(blk_mq_end_io); + +/** + * blk_mq_complete_request - end I/O on a request + * @rq: the request being processed + * + * Description: + * Ends all I/O on a request. It does not handle partial completions. + * The actual completion happens out-of-order, through a IPI handler. + **/ +void blk_mq_complete_request(struct request *rq) +{ + if (unlikely(blk_should_fake_timeout(rq->q))) + return; + if (!blk_mark_rq_complete(rq)) + __blk_mq_complete_request(rq); +} +EXPORT_SYMBOL(blk_mq_complete_request); static void blk_mq_start_request(struct request *rq) { @@ -1399,6 +1406,9 @@ struct request_queue *blk_mq_init_queue(struct blk_mq_reg *reg, if (reg->timeout) blk_queue_rq_timeout(q, reg->timeout); + if (reg->ops->complete) + blk_queue_softirq_done(q, reg->ops->complete); + blk_mq_init_flush(q); blk_mq_init_cpu_queues(q, reg->nr_hw_queues); diff --git a/block/blk-mq.h b/block/blk-mq.h index 5c3917984b00..f29b645f0e1c 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -22,8 +22,7 @@ struct blk_mq_ctx { struct kobject kobj; }; -void __blk_mq_end_io(struct request *rq, int error); -void blk_mq_complete_request(struct request *rq, int error); +void __blk_mq_complete_request(struct request *rq); void blk_mq_run_request(struct request *rq, bool run_queue, bool async); void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async); void blk_mq_init_flush(struct request_queue *q); diff --git a/block/blk-timeout.c b/block/blk-timeout.c index bba81c9348e1..d96f7061c6fd 100644 --- a/block/blk-timeout.c +++ b/block/blk-timeout.c @@ -91,7 +91,7 @@ static void blk_rq_timed_out(struct request *req) case BLK_EH_HANDLED: /* Can we use req->errors here? */ if (q->mq_ops) - blk_mq_complete_request(req, req->errors); + __blk_mq_complete_request(req); else __blk_complete_request(req); break; diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index b7638be58599..468be242db90 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -86,6 +86,8 @@ struct blk_mq_ops { */ rq_timed_out_fn *timeout; + softirq_done_fn *complete; + /* * Override for hctx allocations (should probably go) */ @@ -137,6 +139,8 @@ void blk_mq_free_single_hw_queue(struct blk_mq_hw_ctx *, unsigned int); void blk_mq_end_io(struct request *rq, int error); +void blk_mq_complete_request(struct request *rq); + void blk_mq_stop_hw_queue(struct blk_mq_hw_ctx *hctx); void blk_mq_start_hw_queue(struct blk_mq_hw_ctx *hctx); void blk_mq_stop_hw_queues(struct request_queue *q); -- cgit v1.2.3 From 18741986a4b1dc4b1f171634c4191abc3b0fa023 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 10 Feb 2014 09:29:00 -0700 Subject: blk-mq: rework flush sequencing logic Witch to using a preallocated flush_rq for blk-mq similar to what's done with the old request path. This allows us to set up the request properly with a tag from the actually allowed range and ->rq_disk as needed by some drivers. To make life easier we also switch to dynamic allocation of ->flush_rq for the old path. This effectively reverts most of "blk-mq: fix for flush deadlock" and "blk-mq: Don't reserve a tag for flush request" Signed-off-by: Christoph Hellwig Signed-off-by: Jens Axboe --- block/blk-core.c | 15 +++++-- block/blk-flush.c | 105 ++++++++++++++++++------------------------------- block/blk-mq.c | 54 +++++++++---------------- block/blk-mq.h | 1 + block/blk-sysfs.c | 2 + include/linux/blk-mq.h | 5 +-- include/linux/blkdev.h | 11 ++---- 7 files changed, 76 insertions(+), 117 deletions(-) (limited to 'include') diff --git a/block/blk-core.c b/block/blk-core.c index 06636f3ad424..853f92749202 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -693,11 +693,20 @@ blk_init_queue_node(request_fn_proc *rfn, spinlock_t *lock, int node_id) if (!uninit_q) return NULL; + uninit_q->flush_rq = kzalloc(sizeof(struct request), GFP_KERNEL); + if (!uninit_q->flush_rq) + goto out_cleanup_queue; + q = blk_init_allocated_queue(uninit_q, rfn, lock); if (!q) - blk_cleanup_queue(uninit_q); - + goto out_free_flush_rq; return q; + +out_free_flush_rq: + kfree(uninit_q->flush_rq); +out_cleanup_queue: + blk_cleanup_queue(uninit_q); + return NULL; } EXPORT_SYMBOL(blk_init_queue_node); @@ -1127,7 +1136,7 @@ static struct request *blk_old_get_request(struct request_queue *q, int rw, struct request *blk_get_request(struct request_queue *q, int rw, gfp_t gfp_mask) { if (q->mq_ops) - return blk_mq_alloc_request(q, rw, gfp_mask, false); + return blk_mq_alloc_request(q, rw, gfp_mask); else return blk_old_get_request(q, rw, gfp_mask); } diff --git a/block/blk-flush.c b/block/blk-flush.c index 9143e85226c7..66e2b697f5db 100644 --- a/block/blk-flush.c +++ b/block/blk-flush.c @@ -130,20 +130,26 @@ static void blk_flush_restore_request(struct request *rq) blk_clear_rq_complete(rq); } -static void mq_flush_data_run(struct work_struct *work) +static void mq_flush_run(struct work_struct *work) { struct request *rq; - rq = container_of(work, struct request, mq_flush_data); + rq = container_of(work, struct request, mq_flush_work); memset(&rq->csd, 0, sizeof(rq->csd)); blk_mq_run_request(rq, true, false); } -static void blk_mq_flush_data_insert(struct request *rq) +static bool blk_flush_queue_rq(struct request *rq) { - INIT_WORK(&rq->mq_flush_data, mq_flush_data_run); - kblockd_schedule_work(rq->q, &rq->mq_flush_data); + if (rq->q->mq_ops) { + INIT_WORK(&rq->mq_flush_work, mq_flush_run); + kblockd_schedule_work(rq->q, &rq->mq_flush_work); + return false; + } else { + list_add_tail(&rq->queuelist, &rq->q->queue_head); + return true; + } } /** @@ -187,12 +193,7 @@ static bool blk_flush_complete_seq(struct request *rq, unsigned int seq, case REQ_FSEQ_DATA: list_move_tail(&rq->flush.list, &q->flush_data_in_flight); - if (q->mq_ops) - blk_mq_flush_data_insert(rq); - else { - list_add(&rq->queuelist, &q->queue_head); - queued = true; - } + queued = blk_flush_queue_rq(rq); break; case REQ_FSEQ_DONE: @@ -216,9 +217,6 @@ static bool blk_flush_complete_seq(struct request *rq, unsigned int seq, } kicked = blk_kick_flush(q); - /* blk_mq_run_flush will run queue */ - if (q->mq_ops) - return queued; return kicked | queued; } @@ -230,10 +228,9 @@ static void flush_end_io(struct request *flush_rq, int error) struct request *rq, *n; unsigned long flags = 0; - if (q->mq_ops) { - blk_mq_free_request(flush_rq); + if (q->mq_ops) spin_lock_irqsave(&q->mq_flush_lock, flags); - } + running = &q->flush_queue[q->flush_running_idx]; BUG_ON(q->flush_pending_idx == q->flush_running_idx); @@ -263,48 +260,14 @@ static void flush_end_io(struct request *flush_rq, int error) * kblockd. */ if (queued || q->flush_queue_delayed) { - if (!q->mq_ops) - blk_run_queue_async(q); - else - /* - * This can be optimized to only run queues with requests - * queued if necessary. - */ - blk_mq_run_queues(q, true); + WARN_ON(q->mq_ops); + blk_run_queue_async(q); } q->flush_queue_delayed = 0; if (q->mq_ops) spin_unlock_irqrestore(&q->mq_flush_lock, flags); } -static void mq_flush_work(struct work_struct *work) -{ - struct request_queue *q; - struct request *rq; - - q = container_of(work, struct request_queue, mq_flush_work); - - rq = blk_mq_alloc_request(q, WRITE_FLUSH|REQ_FLUSH_SEQ, - __GFP_WAIT|GFP_ATOMIC, false); - rq->cmd_type = REQ_TYPE_FS; - rq->end_io = flush_end_io; - - blk_mq_run_request(rq, true, false); -} - -/* - * We can't directly use q->flush_rq, because it doesn't have tag and is not in - * hctx->rqs[]. so we must allocate a new request, since we can't sleep here, - * so offload the work to workqueue. - * - * Note: we assume a flush request finished in any hardware queue will flush - * the whole disk cache. - */ -static void mq_run_flush(struct request_queue *q) -{ - kblockd_schedule_work(q, &q->mq_flush_work); -} - /** * blk_kick_flush - consider issuing flush request * @q: request_queue being kicked @@ -339,19 +302,31 @@ static bool blk_kick_flush(struct request_queue *q) * different from running_idx, which means flush is in flight. */ q->flush_pending_idx ^= 1; + if (q->mq_ops) { - mq_run_flush(q); - return true; + struct blk_mq_ctx *ctx = first_rq->mq_ctx; + struct blk_mq_hw_ctx *hctx = q->mq_ops->map_queue(q, ctx->cpu); + + blk_mq_rq_init(hctx, q->flush_rq); + q->flush_rq->mq_ctx = ctx; + + /* + * Reuse the tag value from the fist waiting request, + * with blk-mq the tag is generated during request + * allocation and drivers can rely on it being inside + * the range they asked for. + */ + q->flush_rq->tag = first_rq->tag; + } else { + blk_rq_init(q, q->flush_rq); } - blk_rq_init(q, &q->flush_rq); - q->flush_rq.cmd_type = REQ_TYPE_FS; - q->flush_rq.cmd_flags = WRITE_FLUSH | REQ_FLUSH_SEQ; - q->flush_rq.rq_disk = first_rq->rq_disk; - q->flush_rq.end_io = flush_end_io; + q->flush_rq->cmd_type = REQ_TYPE_FS; + q->flush_rq->cmd_flags = WRITE_FLUSH | REQ_FLUSH_SEQ; + q->flush_rq->rq_disk = first_rq->rq_disk; + q->flush_rq->end_io = flush_end_io; - list_add_tail(&q->flush_rq.queuelist, &q->queue_head); - return true; + return blk_flush_queue_rq(q->flush_rq); } static void flush_data_end_io(struct request *rq, int error) @@ -407,11 +382,8 @@ void blk_insert_flush(struct request *rq) /* * @policy now records what operations need to be done. Adjust * REQ_FLUSH and FUA for the driver. - * We keep REQ_FLUSH for mq to track flush requests. For !FUA, - * we never dispatch the request directly. */ - if (rq->cmd_flags & REQ_FUA) - rq->cmd_flags &= ~REQ_FLUSH; + rq->cmd_flags &= ~REQ_FLUSH; if (!(fflags & REQ_FUA)) rq->cmd_flags &= ~REQ_FUA; @@ -560,5 +532,4 @@ EXPORT_SYMBOL(blkdev_issue_flush); void blk_mq_init_flush(struct request_queue *q) { spin_lock_init(&q->mq_flush_lock); - INIT_WORK(&q->mq_flush_work, mq_flush_work); } diff --git a/block/blk-mq.c b/block/blk-mq.c index 14c8f35946e1..a59b0565e940 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -194,27 +194,9 @@ static void blk_mq_rq_ctx_init(struct request_queue *q, struct blk_mq_ctx *ctx, } static struct request *__blk_mq_alloc_request(struct blk_mq_hw_ctx *hctx, - gfp_t gfp, bool reserved, - int rw) + gfp_t gfp, bool reserved) { - struct request *req; - bool is_flush = false; - /* - * flush need allocate a request, leave at least one request for - * non-flush IO to avoid deadlock - */ - if ((rw & REQ_FLUSH) && !(rw & REQ_FLUSH_SEQ)) { - if (atomic_inc_return(&hctx->pending_flush) >= - hctx->queue_depth - hctx->reserved_tags - 1) { - atomic_dec(&hctx->pending_flush); - return NULL; - } - is_flush = true; - } - req = blk_mq_alloc_rq(hctx, gfp, reserved); - if (!req && is_flush) - atomic_dec(&hctx->pending_flush); - return req; + return blk_mq_alloc_rq(hctx, gfp, reserved); } static struct request *blk_mq_alloc_request_pinned(struct request_queue *q, @@ -227,7 +209,7 @@ static struct request *blk_mq_alloc_request_pinned(struct request_queue *q, struct blk_mq_ctx *ctx = blk_mq_get_ctx(q); struct blk_mq_hw_ctx *hctx = q->mq_ops->map_queue(q, ctx->cpu); - rq = __blk_mq_alloc_request(hctx, gfp & ~__GFP_WAIT, reserved, rw); + rq = __blk_mq_alloc_request(hctx, gfp & ~__GFP_WAIT, reserved); if (rq) { blk_mq_rq_ctx_init(q, ctx, rq, rw); break; @@ -244,15 +226,14 @@ static struct request *blk_mq_alloc_request_pinned(struct request_queue *q, return rq; } -struct request *blk_mq_alloc_request(struct request_queue *q, int rw, - gfp_t gfp, bool reserved) +struct request *blk_mq_alloc_request(struct request_queue *q, int rw, gfp_t gfp) { struct request *rq; if (blk_mq_queue_enter(q)) return NULL; - rq = blk_mq_alloc_request_pinned(q, rw, gfp, reserved); + rq = blk_mq_alloc_request_pinned(q, rw, gfp, false); if (rq) blk_mq_put_ctx(rq->mq_ctx); return rq; @@ -276,7 +257,7 @@ EXPORT_SYMBOL(blk_mq_alloc_reserved_request); /* * Re-init and set pdu, if we have it */ -static void blk_mq_rq_init(struct blk_mq_hw_ctx *hctx, struct request *rq) +void blk_mq_rq_init(struct blk_mq_hw_ctx *hctx, struct request *rq) { blk_rq_init(hctx->queue, rq); @@ -290,9 +271,6 @@ static void __blk_mq_free_request(struct blk_mq_hw_ctx *hctx, const int tag = rq->tag; struct request_queue *q = rq->q; - if ((rq->cmd_flags & REQ_FLUSH) && !(rq->cmd_flags & REQ_FLUSH_SEQ)) - atomic_dec(&hctx->pending_flush); - blk_mq_rq_init(hctx, rq); blk_mq_put_tag(hctx->tags, tag); @@ -946,14 +924,14 @@ static void blk_mq_make_request(struct request_queue *q, struct bio *bio) hctx = q->mq_ops->map_queue(q, ctx->cpu); trace_block_getrq(q, bio, rw); - rq = __blk_mq_alloc_request(hctx, GFP_ATOMIC, false, bio->bi_rw); + rq = __blk_mq_alloc_request(hctx, GFP_ATOMIC, false); if (likely(rq)) - blk_mq_rq_ctx_init(q, ctx, rq, bio->bi_rw); + blk_mq_rq_ctx_init(q, ctx, rq, rw); else { blk_mq_put_ctx(ctx); trace_block_sleeprq(q, bio, rw); - rq = blk_mq_alloc_request_pinned(q, bio->bi_rw, - __GFP_WAIT|GFP_ATOMIC, false); + rq = blk_mq_alloc_request_pinned(q, rw, __GFP_WAIT|GFP_ATOMIC, + false); ctx = rq->mq_ctx; hctx = q->mq_ops->map_queue(q, ctx->cpu); } @@ -1230,9 +1208,7 @@ static int blk_mq_init_hw_queues(struct request_queue *q, hctx->queue_num = i; hctx->flags = reg->flags; hctx->queue_depth = reg->queue_depth; - hctx->reserved_tags = reg->reserved_tags; hctx->cmd_size = reg->cmd_size; - atomic_set(&hctx->pending_flush, 0); blk_mq_init_cpu_notifier(&hctx->cpu_notifier, blk_mq_hctx_notify, hctx); @@ -1412,9 +1388,14 @@ struct request_queue *blk_mq_init_queue(struct blk_mq_reg *reg, blk_mq_init_flush(q); blk_mq_init_cpu_queues(q, reg->nr_hw_queues); - if (blk_mq_init_hw_queues(q, reg, driver_data)) + q->flush_rq = kzalloc(round_up(sizeof(struct request) + reg->cmd_size, + cache_line_size()), GFP_KERNEL); + if (!q->flush_rq) goto err_hw; + if (blk_mq_init_hw_queues(q, reg, driver_data)) + goto err_flush_rq; + blk_mq_map_swqueue(q); mutex_lock(&all_q_mutex); @@ -1422,6 +1403,9 @@ struct request_queue *blk_mq_init_queue(struct blk_mq_reg *reg, mutex_unlock(&all_q_mutex); return q; + +err_flush_rq: + kfree(q->flush_rq); err_hw: kfree(q->mq_map); err_map: diff --git a/block/blk-mq.h b/block/blk-mq.h index f29b645f0e1c..ed0035cd458e 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -28,6 +28,7 @@ void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async); void blk_mq_init_flush(struct request_queue *q); void blk_mq_drain_queue(struct request_queue *q); void blk_mq_free_queue(struct request_queue *q); +void blk_mq_rq_init(struct blk_mq_hw_ctx *hctx, struct request *rq); /* * CPU hotplug helpers diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index 8095c4a21fc0..7500f876dae4 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -549,6 +549,8 @@ static void blk_release_queue(struct kobject *kobj) if (q->mq_ops) blk_mq_free_queue(q); + kfree(q->flush_rq); + blk_trace_shutdown(q); bdi_destroy(&q->backing_dev_info); diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 468be242db90..18ba8a627f46 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -36,15 +36,12 @@ struct blk_mq_hw_ctx { struct list_head page_list; struct blk_mq_tags *tags; - atomic_t pending_flush; - unsigned long queued; unsigned long run; #define BLK_MQ_MAX_DISPATCH_ORDER 10 unsigned long dispatched[BLK_MQ_MAX_DISPATCH_ORDER]; unsigned int queue_depth; - unsigned int reserved_tags; unsigned int numa_node; unsigned int cmd_size; /* per-request extra data */ @@ -129,7 +126,7 @@ void blk_mq_insert_request(struct request_queue *, struct request *, void blk_mq_run_queues(struct request_queue *q, bool async); void blk_mq_free_request(struct request *rq); bool blk_mq_can_queue(struct blk_mq_hw_ctx *); -struct request *blk_mq_alloc_request(struct request_queue *q, int rw, gfp_t gfp, bool reserved); +struct request *blk_mq_alloc_request(struct request_queue *q, int rw, gfp_t gfp); struct request *blk_mq_alloc_reserved_request(struct request_queue *q, int rw, gfp_t gfp); struct request *blk_mq_rq_from_tag(struct request_queue *q, unsigned int tag); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 0375654adb28..b2d25ecbcbc1 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -101,7 +101,7 @@ struct request { }; union { struct call_single_data csd; - struct work_struct mq_flush_data; + struct work_struct mq_flush_work; }; struct request_queue *q; @@ -451,13 +451,8 @@ struct request_queue { unsigned long flush_pending_since; struct list_head flush_queue[2]; struct list_head flush_data_in_flight; - union { - struct request flush_rq; - struct { - spinlock_t mq_flush_lock; - struct work_struct mq_flush_work; - }; - }; + struct request *flush_rq; + spinlock_t mq_flush_lock; struct mutex sysfs_lock; -- cgit v1.2.3 From 564eb714f5f09ac733c26860d5f0831f213fbdf1 Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Mon, 10 Feb 2014 13:54:02 +0000 Subject: xen: install xen/gntdev.h and xen/gntalloc.h xen/gntdev.h and xen/gntalloc.h both provide userspace ABIs so they should be installed. CC: stable@vger.kernel.org Signed-off-by: David Vrabel Signed-off-by: Konrad Rzeszutek Wilk --- include/uapi/xen/Kbuild | 2 + include/uapi/xen/gntalloc.h | 82 ++++++++++++++++++++++++ include/uapi/xen/gntdev.h | 150 ++++++++++++++++++++++++++++++++++++++++++++ include/xen/gntalloc.h | 82 ------------------------ include/xen/gntdev.h | 150 -------------------------------------------- 5 files changed, 234 insertions(+), 232 deletions(-) create mode 100644 include/uapi/xen/gntalloc.h create mode 100644 include/uapi/xen/gntdev.h delete mode 100644 include/xen/gntalloc.h delete mode 100644 include/xen/gntdev.h (limited to 'include') diff --git a/include/uapi/xen/Kbuild b/include/uapi/xen/Kbuild index 61257cb14653..5c459628e8c7 100644 --- a/include/uapi/xen/Kbuild +++ b/include/uapi/xen/Kbuild @@ -1,3 +1,5 @@ # UAPI Header export list header-y += evtchn.h +header-y += gntalloc.h +header-y += gntdev.h header-y += privcmd.h diff --git a/include/uapi/xen/gntalloc.h b/include/uapi/xen/gntalloc.h new file mode 100644 index 000000000000..76bd58065f4f --- /dev/null +++ b/include/uapi/xen/gntalloc.h @@ -0,0 +1,82 @@ +/****************************************************************************** + * gntalloc.h + * + * Interface to /dev/xen/gntalloc. + * + * Author: Daniel De Graaf + * + * This file is in the public domain. + */ + +#ifndef __LINUX_PUBLIC_GNTALLOC_H__ +#define __LINUX_PUBLIC_GNTALLOC_H__ + +/* + * Allocates a new page and creates a new grant reference. + */ +#define IOCTL_GNTALLOC_ALLOC_GREF \ +_IOC(_IOC_NONE, 'G', 5, sizeof(struct ioctl_gntalloc_alloc_gref)) +struct ioctl_gntalloc_alloc_gref { + /* IN parameters */ + /* The ID of the domain to be given access to the grants. */ + uint16_t domid; + /* Flags for this mapping */ + uint16_t flags; + /* Number of pages to map */ + uint32_t count; + /* OUT parameters */ + /* The offset to be used on a subsequent call to mmap(). */ + uint64_t index; + /* The grant references of the newly created grant, one per page */ + /* Variable size, depending on count */ + uint32_t gref_ids[1]; +}; + +#define GNTALLOC_FLAG_WRITABLE 1 + +/* + * Deallocates the grant reference, allowing the associated page to be freed if + * no other domains are using it. + */ +#define IOCTL_GNTALLOC_DEALLOC_GREF \ +_IOC(_IOC_NONE, 'G', 6, sizeof(struct ioctl_gntalloc_dealloc_gref)) +struct ioctl_gntalloc_dealloc_gref { + /* IN parameters */ + /* The offset returned in the map operation */ + uint64_t index; + /* Number of references to unmap */ + uint32_t count; +}; + +/* + * Sets up an unmap notification within the page, so that the other side can do + * cleanup if this side crashes. Required to implement cross-domain robust + * mutexes or close notification on communication channels. + * + * Each mapped page only supports one notification; multiple calls referring to + * the same page overwrite the previous notification. You must clear the + * notification prior to the IOCTL_GNTALLOC_DEALLOC_GREF if you do not want it + * to occur. + */ +#define IOCTL_GNTALLOC_SET_UNMAP_NOTIFY \ +_IOC(_IOC_NONE, 'G', 7, sizeof(struct ioctl_gntalloc_unmap_notify)) +struct ioctl_gntalloc_unmap_notify { + /* IN parameters */ + /* Offset in the file descriptor for a byte within the page (same as + * used in mmap). If using UNMAP_NOTIFY_CLEAR_BYTE, this is the byte to + * be cleared. Otherwise, it can be any byte in the page whose + * notification we are adjusting. + */ + uint64_t index; + /* Action(s) to take on unmap */ + uint32_t action; + /* Event channel to notify */ + uint32_t event_channel_port; +}; + +/* Clear (set to zero) the byte specified by index */ +#define UNMAP_NOTIFY_CLEAR_BYTE 0x1 +/* Send an interrupt on the indicated event channel */ +#define UNMAP_NOTIFY_SEND_EVENT 0x2 + +#endif /* __LINUX_PUBLIC_GNTALLOC_H__ */ diff --git a/include/uapi/xen/gntdev.h b/include/uapi/xen/gntdev.h new file mode 100644 index 000000000000..5304bd3c84c5 --- /dev/null +++ b/include/uapi/xen/gntdev.h @@ -0,0 +1,150 @@ +/****************************************************************************** + * gntdev.h + * + * Interface to /dev/xen/gntdev. + * + * Copyright (c) 2007, D G Murray + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#ifndef __LINUX_PUBLIC_GNTDEV_H__ +#define __LINUX_PUBLIC_GNTDEV_H__ + +struct ioctl_gntdev_grant_ref { + /* The domain ID of the grant to be mapped. */ + uint32_t domid; + /* The grant reference of the grant to be mapped. */ + uint32_t ref; +}; + +/* + * Inserts the grant references into the mapping table of an instance + * of gntdev. N.B. This does not perform the mapping, which is deferred + * until mmap() is called with @index as the offset. + */ +#define IOCTL_GNTDEV_MAP_GRANT_REF \ +_IOC(_IOC_NONE, 'G', 0, sizeof(struct ioctl_gntdev_map_grant_ref)) +struct ioctl_gntdev_map_grant_ref { + /* IN parameters */ + /* The number of grants to be mapped. */ + uint32_t count; + uint32_t pad; + /* OUT parameters */ + /* The offset to be used on a subsequent call to mmap(). */ + uint64_t index; + /* Variable IN parameter. */ + /* Array of grant references, of size @count. */ + struct ioctl_gntdev_grant_ref refs[1]; +}; + +/* + * Removes the grant references from the mapping table of an instance of + * of gntdev. N.B. munmap() must be called on the relevant virtual address(es) + * before this ioctl is called, or an error will result. + */ +#define IOCTL_GNTDEV_UNMAP_GRANT_REF \ +_IOC(_IOC_NONE, 'G', 1, sizeof(struct ioctl_gntdev_unmap_grant_ref)) +struct ioctl_gntdev_unmap_grant_ref { + /* IN parameters */ + /* The offset was returned by the corresponding map operation. */ + uint64_t index; + /* The number of pages to be unmapped. */ + uint32_t count; + uint32_t pad; +}; + +/* + * Returns the offset in the driver's address space that corresponds + * to @vaddr. This can be used to perform a munmap(), followed by an + * UNMAP_GRANT_REF ioctl, where no state about the offset is retained by + * the caller. The number of pages that were allocated at the same time as + * @vaddr is returned in @count. + * + * N.B. Where more than one page has been mapped into a contiguous range, the + * supplied @vaddr must correspond to the start of the range; otherwise + * an error will result. It is only possible to munmap() the entire + * contiguously-allocated range at once, and not any subrange thereof. + */ +#define IOCTL_GNTDEV_GET_OFFSET_FOR_VADDR \ +_IOC(_IOC_NONE, 'G', 2, sizeof(struct ioctl_gntdev_get_offset_for_vaddr)) +struct ioctl_gntdev_get_offset_for_vaddr { + /* IN parameters */ + /* The virtual address of the first mapped page in a range. */ + uint64_t vaddr; + /* OUT parameters */ + /* The offset that was used in the initial mmap() operation. */ + uint64_t offset; + /* The number of pages mapped in the VM area that begins at @vaddr. */ + uint32_t count; + uint32_t pad; +}; + +/* + * Sets the maximum number of grants that may mapped at once by this gntdev + * instance. + * + * N.B. This must be called before any other ioctl is performed on the device. + */ +#define IOCTL_GNTDEV_SET_MAX_GRANTS \ +_IOC(_IOC_NONE, 'G', 3, sizeof(struct ioctl_gntdev_set_max_grants)) +struct ioctl_gntdev_set_max_grants { + /* IN parameter */ + /* The maximum number of grants that may be mapped at once. */ + uint32_t count; +}; + +/* + * Sets up an unmap notification within the page, so that the other side can do + * cleanup if this side crashes. Required to implement cross-domain robust + * mutexes or close notification on communication channels. + * + * Each mapped page only supports one notification; multiple calls referring to + * the same page overwrite the previous notification. You must clear the + * notification prior to the IOCTL_GNTALLOC_DEALLOC_GREF if you do not want it + * to occur. + */ +#define IOCTL_GNTDEV_SET_UNMAP_NOTIFY \ +_IOC(_IOC_NONE, 'G', 7, sizeof(struct ioctl_gntdev_unmap_notify)) +struct ioctl_gntdev_unmap_notify { + /* IN parameters */ + /* Offset in the file descriptor for a byte within the page (same as + * used in mmap). If using UNMAP_NOTIFY_CLEAR_BYTE, this is the byte to + * be cleared. Otherwise, it can be any byte in the page whose + * notification we are adjusting. + */ + uint64_t index; + /* Action(s) to take on unmap */ + uint32_t action; + /* Event channel to notify */ + uint32_t event_channel_port; +}; + +/* Clear (set to zero) the byte specified by index */ +#define UNMAP_NOTIFY_CLEAR_BYTE 0x1 +/* Send an interrupt on the indicated event channel */ +#define UNMAP_NOTIFY_SEND_EVENT 0x2 + +#endif /* __LINUX_PUBLIC_GNTDEV_H__ */ diff --git a/include/xen/gntalloc.h b/include/xen/gntalloc.h deleted file mode 100644 index 76bd58065f4f..000000000000 --- a/include/xen/gntalloc.h +++ /dev/null @@ -1,82 +0,0 @@ -/****************************************************************************** - * gntalloc.h - * - * Interface to /dev/xen/gntalloc. - * - * Author: Daniel De Graaf - * - * This file is in the public domain. - */ - -#ifndef __LINUX_PUBLIC_GNTALLOC_H__ -#define __LINUX_PUBLIC_GNTALLOC_H__ - -/* - * Allocates a new page and creates a new grant reference. - */ -#define IOCTL_GNTALLOC_ALLOC_GREF \ -_IOC(_IOC_NONE, 'G', 5, sizeof(struct ioctl_gntalloc_alloc_gref)) -struct ioctl_gntalloc_alloc_gref { - /* IN parameters */ - /* The ID of the domain to be given access to the grants. */ - uint16_t domid; - /* Flags for this mapping */ - uint16_t flags; - /* Number of pages to map */ - uint32_t count; - /* OUT parameters */ - /* The offset to be used on a subsequent call to mmap(). */ - uint64_t index; - /* The grant references of the newly created grant, one per page */ - /* Variable size, depending on count */ - uint32_t gref_ids[1]; -}; - -#define GNTALLOC_FLAG_WRITABLE 1 - -/* - * Deallocates the grant reference, allowing the associated page to be freed if - * no other domains are using it. - */ -#define IOCTL_GNTALLOC_DEALLOC_GREF \ -_IOC(_IOC_NONE, 'G', 6, sizeof(struct ioctl_gntalloc_dealloc_gref)) -struct ioctl_gntalloc_dealloc_gref { - /* IN parameters */ - /* The offset returned in the map operation */ - uint64_t index; - /* Number of references to unmap */ - uint32_t count; -}; - -/* - * Sets up an unmap notification within the page, so that the other side can do - * cleanup if this side crashes. Required to implement cross-domain robust - * mutexes or close notification on communication channels. - * - * Each mapped page only supports one notification; multiple calls referring to - * the same page overwrite the previous notification. You must clear the - * notification prior to the IOCTL_GNTALLOC_DEALLOC_GREF if you do not want it - * to occur. - */ -#define IOCTL_GNTALLOC_SET_UNMAP_NOTIFY \ -_IOC(_IOC_NONE, 'G', 7, sizeof(struct ioctl_gntalloc_unmap_notify)) -struct ioctl_gntalloc_unmap_notify { - /* IN parameters */ - /* Offset in the file descriptor for a byte within the page (same as - * used in mmap). If using UNMAP_NOTIFY_CLEAR_BYTE, this is the byte to - * be cleared. Otherwise, it can be any byte in the page whose - * notification we are adjusting. - */ - uint64_t index; - /* Action(s) to take on unmap */ - uint32_t action; - /* Event channel to notify */ - uint32_t event_channel_port; -}; - -/* Clear (set to zero) the byte specified by index */ -#define UNMAP_NOTIFY_CLEAR_BYTE 0x1 -/* Send an interrupt on the indicated event channel */ -#define UNMAP_NOTIFY_SEND_EVENT 0x2 - -#endif /* __LINUX_PUBLIC_GNTALLOC_H__ */ diff --git a/include/xen/gntdev.h b/include/xen/gntdev.h deleted file mode 100644 index 5304bd3c84c5..000000000000 --- a/include/xen/gntdev.h +++ /dev/null @@ -1,150 +0,0 @@ -/****************************************************************************** - * gntdev.h - * - * Interface to /dev/xen/gntdev. - * - * Copyright (c) 2007, D G Murray - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License version 2 - * as published by the Free Software Foundation; or, when distributed - * separately from the Linux kernel or incorporated into other - * software packages, subject to the following license: - * - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this source file (the "Software"), to deal in the Software without - * restriction, including without limitation the rights to use, copy, modify, - * merge, publish, distribute, sublicense, and/or sell copies of the Software, - * and to permit persons to whom the Software is furnished to do so, subject to - * the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING - * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS - * IN THE SOFTWARE. - */ - -#ifndef __LINUX_PUBLIC_GNTDEV_H__ -#define __LINUX_PUBLIC_GNTDEV_H__ - -struct ioctl_gntdev_grant_ref { - /* The domain ID of the grant to be mapped. */ - uint32_t domid; - /* The grant reference of the grant to be mapped. */ - uint32_t ref; -}; - -/* - * Inserts the grant references into the mapping table of an instance - * of gntdev. N.B. This does not perform the mapping, which is deferred - * until mmap() is called with @index as the offset. - */ -#define IOCTL_GNTDEV_MAP_GRANT_REF \ -_IOC(_IOC_NONE, 'G', 0, sizeof(struct ioctl_gntdev_map_grant_ref)) -struct ioctl_gntdev_map_grant_ref { - /* IN parameters */ - /* The number of grants to be mapped. */ - uint32_t count; - uint32_t pad; - /* OUT parameters */ - /* The offset to be used on a subsequent call to mmap(). */ - uint64_t index; - /* Variable IN parameter. */ - /* Array of grant references, of size @count. */ - struct ioctl_gntdev_grant_ref refs[1]; -}; - -/* - * Removes the grant references from the mapping table of an instance of - * of gntdev. N.B. munmap() must be called on the relevant virtual address(es) - * before this ioctl is called, or an error will result. - */ -#define IOCTL_GNTDEV_UNMAP_GRANT_REF \ -_IOC(_IOC_NONE, 'G', 1, sizeof(struct ioctl_gntdev_unmap_grant_ref)) -struct ioctl_gntdev_unmap_grant_ref { - /* IN parameters */ - /* The offset was returned by the corresponding map operation. */ - uint64_t index; - /* The number of pages to be unmapped. */ - uint32_t count; - uint32_t pad; -}; - -/* - * Returns the offset in the driver's address space that corresponds - * to @vaddr. This can be used to perform a munmap(), followed by an - * UNMAP_GRANT_REF ioctl, where no state about the offset is retained by - * the caller. The number of pages that were allocated at the same time as - * @vaddr is returned in @count. - * - * N.B. Where more than one page has been mapped into a contiguous range, the - * supplied @vaddr must correspond to the start of the range; otherwise - * an error will result. It is only possible to munmap() the entire - * contiguously-allocated range at once, and not any subrange thereof. - */ -#define IOCTL_GNTDEV_GET_OFFSET_FOR_VADDR \ -_IOC(_IOC_NONE, 'G', 2, sizeof(struct ioctl_gntdev_get_offset_for_vaddr)) -struct ioctl_gntdev_get_offset_for_vaddr { - /* IN parameters */ - /* The virtual address of the first mapped page in a range. */ - uint64_t vaddr; - /* OUT parameters */ - /* The offset that was used in the initial mmap() operation. */ - uint64_t offset; - /* The number of pages mapped in the VM area that begins at @vaddr. */ - uint32_t count; - uint32_t pad; -}; - -/* - * Sets the maximum number of grants that may mapped at once by this gntdev - * instance. - * - * N.B. This must be called before any other ioctl is performed on the device. - */ -#define IOCTL_GNTDEV_SET_MAX_GRANTS \ -_IOC(_IOC_NONE, 'G', 3, sizeof(struct ioctl_gntdev_set_max_grants)) -struct ioctl_gntdev_set_max_grants { - /* IN parameter */ - /* The maximum number of grants that may be mapped at once. */ - uint32_t count; -}; - -/* - * Sets up an unmap notification within the page, so that the other side can do - * cleanup if this side crashes. Required to implement cross-domain robust - * mutexes or close notification on communication channels. - * - * Each mapped page only supports one notification; multiple calls referring to - * the same page overwrite the previous notification. You must clear the - * notification prior to the IOCTL_GNTALLOC_DEALLOC_GREF if you do not want it - * to occur. - */ -#define IOCTL_GNTDEV_SET_UNMAP_NOTIFY \ -_IOC(_IOC_NONE, 'G', 7, sizeof(struct ioctl_gntdev_unmap_notify)) -struct ioctl_gntdev_unmap_notify { - /* IN parameters */ - /* Offset in the file descriptor for a byte within the page (same as - * used in mmap). If using UNMAP_NOTIFY_CLEAR_BYTE, this is the byte to - * be cleared. Otherwise, it can be any byte in the page whose - * notification we are adjusting. - */ - uint64_t index; - /* Action(s) to take on unmap */ - uint32_t action; - /* Event channel to notify */ - uint32_t event_channel_port; -}; - -/* Clear (set to zero) the byte specified by index */ -#define UNMAP_NOTIFY_CLEAR_BYTE 0x1 -/* Send an interrupt on the indicated event channel */ -#define UNMAP_NOTIFY_SEND_EVENT 0x2 - -#endif /* __LINUX_PUBLIC_GNTDEV_H__ */ -- cgit v1.2.3 From d8320b2d2e7d1cda8216d496c7c685c5fdf74ff0 Mon Sep 17 00:00:00 2001 From: Paul Bolle Date: Sat, 8 Feb 2014 23:35:43 +0100 Subject: ia64/xen: Remove Xen support for ia64 even more Commit d52eefb47d4e ("ia64/xen: Remove Xen support for ia64") removed the Kconfig symbol XEN_XENCOMM. But it didn't remove the code depending on that symbol. Remove that code now. Signed-off-by: Paul Bolle Acked-by: David Vrabel Signed-off-by: Konrad Rzeszutek Wilk --- drivers/xen/Makefile | 1 - drivers/xen/xencomm.c | 219 ---------------------------------------- include/xen/interface/xencomm.h | 41 -------- include/xen/xencomm.h | 77 -------------- 4 files changed, 338 deletions(-) delete mode 100644 drivers/xen/xencomm.c delete mode 100644 include/xen/interface/xencomm.h delete mode 100644 include/xen/xencomm.h (limited to 'include') diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile index d75c811bfa56..45e00afa7f2d 100644 --- a/drivers/xen/Makefile +++ b/drivers/xen/Makefile @@ -16,7 +16,6 @@ xen-pad-$(CONFIG_X86) += xen-acpi-pad.o dom0-$(CONFIG_X86) += pcpu.o obj-$(CONFIG_XEN_DOM0) += $(dom0-y) obj-$(CONFIG_BLOCK) += biomerge.o -obj-$(CONFIG_XEN_XENCOMM) += xencomm.o obj-$(CONFIG_XEN_BALLOON) += xen-balloon.o obj-$(CONFIG_XEN_SELFBALLOONING) += xen-selfballoon.o obj-$(CONFIG_XEN_DEV_EVTCHN) += xen-evtchn.o diff --git a/drivers/xen/xencomm.c b/drivers/xen/xencomm.c deleted file mode 100644 index 4793fc594549..000000000000 --- a/drivers/xen/xencomm.c +++ /dev/null @@ -1,219 +0,0 @@ -/* - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - * - * Copyright (C) IBM Corp. 2006 - * - * Authors: Hollis Blanchard - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include -#include -#include -#include -#include -#include /* for xencomm_is_phys_contiguous() */ - -static int xencomm_init(struct xencomm_desc *desc, - void *buffer, unsigned long bytes) -{ - unsigned long recorded = 0; - int i = 0; - - while ((recorded < bytes) && (i < desc->nr_addrs)) { - unsigned long vaddr = (unsigned long)buffer + recorded; - unsigned long paddr; - int offset; - int chunksz; - - offset = vaddr % PAGE_SIZE; /* handle partial pages */ - chunksz = min(PAGE_SIZE - offset, bytes - recorded); - - paddr = xencomm_vtop(vaddr); - if (paddr == ~0UL) { - printk(KERN_DEBUG "%s: couldn't translate vaddr %lx\n", - __func__, vaddr); - return -EINVAL; - } - - desc->address[i++] = paddr; - recorded += chunksz; - } - - if (recorded < bytes) { - printk(KERN_DEBUG - "%s: could only translate %ld of %ld bytes\n", - __func__, recorded, bytes); - return -ENOSPC; - } - - /* mark remaining addresses invalid (just for safety) */ - while (i < desc->nr_addrs) - desc->address[i++] = XENCOMM_INVALID; - - desc->magic = XENCOMM_MAGIC; - - return 0; -} - -static struct xencomm_desc *xencomm_alloc(gfp_t gfp_mask, - void *buffer, unsigned long bytes) -{ - struct xencomm_desc *desc; - unsigned long buffer_ulong = (unsigned long)buffer; - unsigned long start = buffer_ulong & PAGE_MASK; - unsigned long end = (buffer_ulong + bytes) | ~PAGE_MASK; - unsigned long nr_addrs = (end - start + 1) >> PAGE_SHIFT; - unsigned long size = sizeof(*desc) + - sizeof(desc->address[0]) * nr_addrs; - - /* - * slab allocator returns at least sizeof(void*) aligned pointer. - * When sizeof(*desc) > sizeof(void*), struct xencomm_desc might - * cross page boundary. - */ - if (sizeof(*desc) > sizeof(void *)) { - unsigned long order = get_order(size); - desc = (struct xencomm_desc *)__get_free_pages(gfp_mask, - order); - if (desc == NULL) - return NULL; - - desc->nr_addrs = - ((PAGE_SIZE << order) - sizeof(struct xencomm_desc)) / - sizeof(*desc->address); - } else { - desc = kmalloc(size, gfp_mask); - if (desc == NULL) - return NULL; - - desc->nr_addrs = nr_addrs; - } - return desc; -} - -void xencomm_free(struct xencomm_handle *desc) -{ - if (desc && !((ulong)desc & XENCOMM_INLINE_FLAG)) { - struct xencomm_desc *desc__ = (struct xencomm_desc *)desc; - if (sizeof(*desc__) > sizeof(void *)) { - unsigned long size = sizeof(*desc__) + - sizeof(desc__->address[0]) * desc__->nr_addrs; - unsigned long order = get_order(size); - free_pages((unsigned long)__va(desc), order); - } else - kfree(__va(desc)); - } -} - -static int xencomm_create(void *buffer, unsigned long bytes, - struct xencomm_desc **ret, gfp_t gfp_mask) -{ - struct xencomm_desc *desc; - int rc; - - pr_debug("%s: %p[%ld]\n", __func__, buffer, bytes); - - if (bytes == 0) { - /* don't create a descriptor; Xen recognizes NULL. */ - BUG_ON(buffer != NULL); - *ret = NULL; - return 0; - } - - BUG_ON(buffer == NULL); /* 'bytes' is non-zero */ - - desc = xencomm_alloc(gfp_mask, buffer, bytes); - if (!desc) { - printk(KERN_DEBUG "%s failure\n", "xencomm_alloc"); - return -ENOMEM; - } - - rc = xencomm_init(desc, buffer, bytes); - if (rc) { - printk(KERN_DEBUG "%s failure: %d\n", "xencomm_init", rc); - xencomm_free((struct xencomm_handle *)__pa(desc)); - return rc; - } - - *ret = desc; - return 0; -} - -static struct xencomm_handle *xencomm_create_inline(void *ptr) -{ - unsigned long paddr; - - BUG_ON(!xencomm_is_phys_contiguous((unsigned long)ptr)); - - paddr = (unsigned long)xencomm_pa(ptr); - BUG_ON(paddr & XENCOMM_INLINE_FLAG); - return (struct xencomm_handle *)(paddr | XENCOMM_INLINE_FLAG); -} - -/* "mini" routine, for stack-based communications: */ -static int xencomm_create_mini(void *buffer, - unsigned long bytes, struct xencomm_mini *xc_desc, - struct xencomm_desc **ret) -{ - int rc = 0; - struct xencomm_desc *desc; - BUG_ON(((unsigned long)xc_desc) % sizeof(*xc_desc) != 0); - - desc = (void *)xc_desc; - - desc->nr_addrs = XENCOMM_MINI_ADDRS; - - rc = xencomm_init(desc, buffer, bytes); - if (!rc) - *ret = desc; - - return rc; -} - -struct xencomm_handle *xencomm_map(void *ptr, unsigned long bytes) -{ - int rc; - struct xencomm_desc *desc; - - if (xencomm_is_phys_contiguous((unsigned long)ptr)) - return xencomm_create_inline(ptr); - - rc = xencomm_create(ptr, bytes, &desc, GFP_KERNEL); - - if (rc || desc == NULL) - return NULL; - - return xencomm_pa(desc); -} - -struct xencomm_handle *__xencomm_map_no_alloc(void *ptr, unsigned long bytes, - struct xencomm_mini *xc_desc) -{ - int rc; - struct xencomm_desc *desc = NULL; - - if (xencomm_is_phys_contiguous((unsigned long)ptr)) - return xencomm_create_inline(ptr); - - rc = xencomm_create_mini(ptr, bytes, xc_desc, - &desc); - - if (rc) - return NULL; - - return xencomm_pa(desc); -} diff --git a/include/xen/interface/xencomm.h b/include/xen/interface/xencomm.h deleted file mode 100644 index ac45e0712afa..000000000000 --- a/include/xen/interface/xencomm.h +++ /dev/null @@ -1,41 +0,0 @@ -/* - * Permission is hereby granted, free of charge, to any person obtaining a copy - * of this software and associated documentation files (the "Software"), to - * deal in the Software without restriction, including without limitation the - * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or - * sell copies of the Software, and to permit persons to whom the Software is - * furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE - * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER - * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING - * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER - * DEALINGS IN THE SOFTWARE. - * - * Copyright (C) IBM Corp. 2006 - */ - -#ifndef _XEN_XENCOMM_H_ -#define _XEN_XENCOMM_H_ - -/* A xencomm descriptor is a scatter/gather list containing physical - * addresses corresponding to a virtually contiguous memory area. The - * hypervisor translates these physical addresses to machine addresses to copy - * to and from the virtually contiguous area. - */ - -#define XENCOMM_MAGIC 0x58434F4D /* 'XCOM' */ -#define XENCOMM_INVALID (~0UL) - -struct xencomm_desc { - uint32_t magic; - uint32_t nr_addrs; /* the number of entries in address[] */ - uint64_t address[0]; -}; - -#endif /* _XEN_XENCOMM_H_ */ diff --git a/include/xen/xencomm.h b/include/xen/xencomm.h deleted file mode 100644 index e43b039be112..000000000000 --- a/include/xen/xencomm.h +++ /dev/null @@ -1,77 +0,0 @@ -/* - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA - * - * Copyright (C) IBM Corp. 2006 - * - * Authors: Hollis Blanchard - * Jerone Young - */ - -#ifndef _LINUX_XENCOMM_H_ -#define _LINUX_XENCOMM_H_ - -#include - -#define XENCOMM_MINI_ADDRS 3 -struct xencomm_mini { - struct xencomm_desc _desc; - uint64_t address[XENCOMM_MINI_ADDRS]; -}; - -/* To avoid additionnal virt to phys conversion, an opaque structure is - presented. */ -struct xencomm_handle; - -extern void xencomm_free(struct xencomm_handle *desc); -extern struct xencomm_handle *xencomm_map(void *ptr, unsigned long bytes); -extern struct xencomm_handle *__xencomm_map_no_alloc(void *ptr, - unsigned long bytes, struct xencomm_mini *xc_area); - -#if 0 -#define XENCOMM_MINI_ALIGNED(xc_desc, n) \ - struct xencomm_mini xc_desc ## _base[(n)] \ - __attribute__((__aligned__(sizeof(struct xencomm_mini)))); \ - struct xencomm_mini *xc_desc = &xc_desc ## _base[0]; -#else -/* - * gcc bug workaround: - * http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16660 - * gcc doesn't handle properly stack variable with - * __attribute__((__align__(sizeof(struct xencomm_mini)))) - */ -#define XENCOMM_MINI_ALIGNED(xc_desc, n) \ - unsigned char xc_desc ## _base[((n) + 1 ) * \ - sizeof(struct xencomm_mini)]; \ - struct xencomm_mini *xc_desc = (struct xencomm_mini *) \ - ((unsigned long)xc_desc ## _base + \ - (sizeof(struct xencomm_mini) - \ - ((unsigned long)xc_desc ## _base) % \ - sizeof(struct xencomm_mini))); -#endif -#define xencomm_map_no_alloc(ptr, bytes) \ - ({ XENCOMM_MINI_ALIGNED(xc_desc, 1); \ - __xencomm_map_no_alloc(ptr, bytes, xc_desc); }) - -/* provided by architecture code: */ -extern unsigned long xencomm_vtop(unsigned long vaddr); - -static inline void *xencomm_pa(void *ptr) -{ - return (void *)xencomm_vtop((unsigned long)ptr); -} - -#define xen_guest_handle(hnd) ((hnd).p) - -#endif /* _LINUX_XENCOMM_H_ */ -- cgit v1.2.3 From 8423ae3d7a3cfe084865262cfaeba1359d405182 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Mon, 10 Feb 2014 17:45:50 -0800 Subject: block: Fix cloning of discard/write same bios Immutable biovecs changed the way bio segments are treated in such a way that bio_for_each_segment() cannot now do what we want for discard/write same bios, since bi_size means something completely different for them. Fortunately discard and write same bios never have more than a single biovec, so bio_for_each_segment() is unnecessary and not terribly meaningful for them, but we still have to special case them in a few places. Signed-off-by: Kent Overstreet Tested-by: Richard W.M. Jones Signed-off-by: Jens Axboe --- fs/bio.c | 15 ++++++++++----- include/linux/bio.h | 11 +++++++++++ 2 files changed, 21 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/fs/bio.c b/fs/bio.c index 75c49a382239..8754e7b6eb49 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -611,7 +611,6 @@ EXPORT_SYMBOL(bio_clone_fast); struct bio *bio_clone_bioset(struct bio *bio_src, gfp_t gfp_mask, struct bio_set *bs) { - unsigned nr_iovecs = 0; struct bvec_iter iter; struct bio_vec bv; struct bio *bio; @@ -638,10 +637,7 @@ struct bio *bio_clone_bioset(struct bio *bio_src, gfp_t gfp_mask, * __bio_clone_fast() anyways. */ - bio_for_each_segment(bv, bio_src, iter) - nr_iovecs++; - - bio = bio_alloc_bioset(gfp_mask, nr_iovecs, bs); + bio = bio_alloc_bioset(gfp_mask, bio_segments(bio_src), bs); if (!bio) return NULL; @@ -650,9 +646,18 @@ struct bio *bio_clone_bioset(struct bio *bio_src, gfp_t gfp_mask, bio->bi_iter.bi_sector = bio_src->bi_iter.bi_sector; bio->bi_iter.bi_size = bio_src->bi_iter.bi_size; + if (bio->bi_rw & REQ_DISCARD) + goto integrity_clone; + + if (bio->bi_rw & REQ_WRITE_SAME) { + bio->bi_io_vec[bio->bi_vcnt++] = bio_src->bi_io_vec[0]; + goto integrity_clone; + } + bio_for_each_segment(bv, bio_src, iter) bio->bi_io_vec[bio->bi_vcnt++] = bv; +integrity_clone: if (bio_integrity(bio_src)) { int ret; diff --git a/include/linux/bio.h b/include/linux/bio.h index d6791bba8264..5a4d39b4686b 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -250,6 +250,17 @@ static inline unsigned bio_segments(struct bio *bio) struct bio_vec bv; struct bvec_iter iter; + /* + * We special case discard/write same, because they interpret bi_size + * differently: + */ + + if (bio->bi_rw & REQ_DISCARD) + return 1; + + if (bio->bi_rw & REQ_WRITE_SAME) + return 1; + bio_for_each_segment(bv, bio, iter) segs++; -- cgit v1.2.3 From 857aea1c5769386b1905bb38fbb4464c800b2df6 Mon Sep 17 00:00:00 2001 From: Charmaine Lee Date: Wed, 12 Feb 2014 12:07:38 +0100 Subject: drm/vmwgfx: Get maximum mob size from register SVGA_REG_MOB_MAX_SIZE This patch queries the register SVGA_REG_MOB_MAX_SIZE for the maximum size of a single mob. Signed-off-by: Charmaine Lee Reviewed-by: Thomas Hellstrom --- drivers/gpu/drm/vmwgfx/svga_reg.h | 9 ++++++++- drivers/gpu/drm/vmwgfx/vmwgfx_drv.c | 3 +++ drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 1 + drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c | 3 +++ include/uapi/drm/vmwgfx_drm.h | 1 + 5 files changed, 16 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/drivers/gpu/drm/vmwgfx/svga_reg.h b/drivers/gpu/drm/vmwgfx/svga_reg.h index 71defa4d2d75..11323dd5196f 100644 --- a/drivers/gpu/drm/vmwgfx/svga_reg.h +++ b/drivers/gpu/drm/vmwgfx/svga_reg.h @@ -169,10 +169,17 @@ enum { SVGA_REG_TRACES = 45, /* Enable trace-based updates even when FIFO is on */ SVGA_REG_GMRS_MAX_PAGES = 46, /* Maximum number of 4KB pages for all GMRs */ SVGA_REG_MEMORY_SIZE = 47, /* Total dedicated device memory excluding FIFO */ + SVGA_REG_COMMAND_LOW = 48, /* Lower 32 bits and submits commands */ + SVGA_REG_COMMAND_HIGH = 49, /* Upper 32 bits of command buffer PA */ SVGA_REG_MAX_PRIMARY_BOUNDING_BOX_MEM = 50, /* Max primary memory */ SVGA_REG_SUGGESTED_GBOBJECT_MEM_SIZE_KB = 51, /* Suggested limit on mob mem */ SVGA_REG_DEV_CAP = 52, /* Write dev cap index, read value */ - SVGA_REG_TOP = 53, /* Must be 1 more than the last register */ + SVGA_REG_CMD_PREPEND_LOW = 53, + SVGA_REG_CMD_PREPEND_HIGH = 54, + SVGA_REG_SCREENTARGET_MAX_WIDTH = 55, + SVGA_REG_SCREENTARGET_MAX_HEIGHT = 56, + SVGA_REG_MOB_MAX_SIZE = 57, + SVGA_REG_TOP = 58, /* Must be 1 more than the last register */ SVGA_PALETTE_BASE = 1024, /* Base of SVGA color map */ /* Next 768 (== 256*3) registers exist for colormap */ diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index 3bdc0adc656d..0083cbf99edf 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -667,6 +667,7 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) dev_priv->memory_size = 512*1024*1024; } dev_priv->max_mob_pages = 0; + dev_priv->max_mob_size = 0; if (dev_priv->capabilities & SVGA_CAP_GBOBJECTS) { uint64_t mem_size = vmw_read(dev_priv, @@ -676,6 +677,8 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) dev_priv->prim_bb_mem = vmw_read(dev_priv, SVGA_REG_MAX_PRIMARY_BOUNDING_BOX_MEM); + dev_priv->max_mob_size = + vmw_read(dev_priv, SVGA_REG_MOB_MAX_SIZE); } else dev_priv->prim_bb_mem = dev_priv->vram_size; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index ecaa302a6154..9e4be1725985 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -386,6 +386,7 @@ struct vmw_private { uint32_t max_gmr_ids; uint32_t max_gmr_pages; uint32_t max_mob_pages; + uint32_t max_mob_size; uint32_t memory_size; bool has_gmr; bool has_mob; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c index f9881f9e62bd..47b70949bf3a 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c @@ -102,6 +102,9 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data, vmw_fp->gb_aware = true; param->value = dev_priv->max_mob_pages * PAGE_SIZE; break; + case DRM_VMW_PARAM_MAX_MOB_SIZE: + param->value = dev_priv->max_mob_size; + break; default: DRM_ERROR("Illegal vmwgfx get param request: %d\n", param->param); diff --git a/include/uapi/drm/vmwgfx_drm.h b/include/uapi/drm/vmwgfx_drm.h index 9971c560ed9a..87792a5fee3b 100644 --- a/include/uapi/drm/vmwgfx_drm.h +++ b/include/uapi/drm/vmwgfx_drm.h @@ -87,6 +87,7 @@ #define DRM_VMW_PARAM_MAX_SURF_MEMORY 7 #define DRM_VMW_PARAM_3D_CAPS_SIZE 8 #define DRM_VMW_PARAM_MAX_MOB_MEMORY 9 +#define DRM_VMW_PARAM_MAX_MOB_SIZE 10 /** * struct drm_vmw_getparam_arg -- cgit v1.2.3 From 3dca1471993f9b89f3184468f8bbab2b1e024451 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Mon, 3 Feb 2014 14:08:26 -0800 Subject: target: Simplify command completion by removing CMD_T_FAILED flag The CMD_T_FAILED flag is set used in one place to record the result of a trivial test, and it is only tested once, few lines later. We might as well make the code simpler and easier to read by directly doing the test of "success" where we want to use it. Signed-off-by: Roland Dreier Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_transport.c | 5 +---- include/target/target_core_base.h | 1 - 2 files changed, 1 insertion(+), 5 deletions(-) (limited to 'include') diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c index c50fd9f11aab..24b4f65d8777 100644 --- a/drivers/target/target_core_transport.c +++ b/drivers/target/target_core_transport.c @@ -669,9 +669,6 @@ void target_complete_cmd(struct se_cmd *cmd, u8 scsi_status) return; } - if (!success) - cmd->transport_state |= CMD_T_FAILED; - /* * Check for case where an explicit ABORT_TASK has been received * and transport_wait_for_tasks() will be waiting for completion.. @@ -681,7 +678,7 @@ void target_complete_cmd(struct se_cmd *cmd, u8 scsi_status) spin_unlock_irqrestore(&cmd->t_state_lock, flags); complete(&cmd->t_transport_stop_comp); return; - } else if (cmd->transport_state & CMD_T_FAILED) { + } else if (!success) { INIT_WORK(&cmd->work, target_complete_failure_work); } else { INIT_WORK(&cmd->work, target_complete_ok_work); diff --git a/include/target/target_core_base.h b/include/target/target_core_base.h index c9c791209cd1..1772fadcff62 100644 --- a/include/target/target_core_base.h +++ b/include/target/target_core_base.h @@ -525,7 +525,6 @@ struct se_cmd { #define CMD_T_COMPLETE (1 << 2) #define CMD_T_SENT (1 << 4) #define CMD_T_STOP (1 << 5) -#define CMD_T_FAILED (1 << 6) #define CMD_T_DEV_ACTIVE (1 << 7) #define CMD_T_REQUEST_STOP (1 << 8) #define CMD_T_BUSY (1 << 9) -- cgit v1.2.3 From 709c078e176bd47227e89bb34de7c64b57aaaeab Mon Sep 17 00:00:00 2001 From: Dirk Brandewie Date: Wed, 12 Feb 2014 10:01:03 -0800 Subject: intel_pstate: Remove energy reporting from pstate_sample tracepoint Remove the reporting of energy since it does not provide any useful information about the state of the driver and will be a maintainance headache going forward since the RAPL energy units register is not architectural and subject to change between micro-architectures References: https://bugzilla.kernel.org/show_bug.cgi?id=69831 Fixes: b69880f9ccf7 (intel_pstate: Add trace point to report internal state.) Signed-off-by: Dirk Brandewie Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/intel_pstate.c | 9 --------- include/trace/events/power.h | 7 +------ 2 files changed, 1 insertion(+), 15 deletions(-) (limited to 'include') diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index 79606f473f48..c788abf1c457 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -51,8 +51,6 @@ static inline int32_t div_fp(int32_t x, int32_t y) return div_s64((int64_t)x << FRAC_BITS, (int64_t)y); } -static u64 energy_divisor; - struct sample { int32_t core_pct_busy; u64 aperf; @@ -630,12 +628,10 @@ static void intel_pstate_timer_func(unsigned long __data) { struct cpudata *cpu = (struct cpudata *) __data; struct sample *sample; - u64 energy; intel_pstate_sample(cpu); sample = &cpu->samples[cpu->sample_ptr]; - rdmsrl(MSR_PKG_ENERGY_STATUS, energy); intel_pstate_adjust_busy_pstate(cpu); @@ -644,7 +640,6 @@ static void intel_pstate_timer_func(unsigned long __data) cpu->pstate.current_pstate, sample->mperf, sample->aperf, - div64_u64(energy, energy_divisor), sample->freq); intel_pstate_set_sample_time(cpu); @@ -926,7 +921,6 @@ static int __init intel_pstate_init(void) int cpu, rc = 0; const struct x86_cpu_id *id; struct cpu_defaults *cpu_info; - u64 units; if (no_load) return -ENODEV; @@ -960,9 +954,6 @@ static int __init intel_pstate_init(void) if (rc) goto out; - rdmsrl(MSR_RAPL_POWER_UNIT, units); - energy_divisor = 1 << ((units >> 8) & 0x1f); /* bits{12:8} */ - intel_pstate_debug_expose_params(); intel_pstate_sysfs_expose_params(); diff --git a/include/trace/events/power.h b/include/trace/events/power.h index 9e9475c85de5..e5bf9a76f169 100644 --- a/include/trace/events/power.h +++ b/include/trace/events/power.h @@ -42,7 +42,6 @@ TRACE_EVENT(pstate_sample, u32 state, u64 mperf, u64 aperf, - u32 energy, u32 freq ), @@ -51,7 +50,6 @@ TRACE_EVENT(pstate_sample, state, mperf, aperf, - energy, freq ), @@ -61,7 +59,6 @@ TRACE_EVENT(pstate_sample, __field(u32, state) __field(u64, mperf) __field(u64, aperf) - __field(u32, energy) __field(u32, freq) ), @@ -72,17 +69,15 @@ TRACE_EVENT(pstate_sample, __entry->state = state; __entry->mperf = mperf; __entry->aperf = aperf; - __entry->energy = energy; __entry->freq = freq; ), - TP_printk("core_busy=%lu scaled=%lu state=%lu mperf=%llu aperf=%llu energy=%lu freq=%lu ", + TP_printk("core_busy=%lu scaled=%lu state=%lu mperf=%llu aperf=%llu freq=%lu ", (unsigned long)__entry->core_busy, (unsigned long)__entry->scaled_busy, (unsigned long)__entry->state, (unsigned long long)__entry->mperf, (unsigned long long)__entry->aperf, - (unsigned long)__entry->energy, (unsigned long)__entry->freq ) -- cgit v1.2.3 From c0b00a525c127d0055c1df6283300e17f601a1a1 Mon Sep 17 00:00:00 2001 From: Sumit Semwal Date: Mon, 3 Feb 2014 15:09:12 +0530 Subject: dma-buf: update debugfs output Russell King observed 'wierd' looking output from debugfs, and also suggested better ways of getting device names (use KBUILD_MODNAME, dev_name()) This patch addresses these issues to make the debugfs output correct and better looking. While at it, replace seq_printf with seq_puts to remove the checkpatch.pl warnings. Reported-by: Russell King - ARM Linux Signed-off-by: Sumit Semwal --- drivers/base/dma-buf.c | 25 ++++++++++++------------- include/linux/dma-buf.h | 2 +- 2 files changed, 13 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/drivers/base/dma-buf.c b/drivers/base/dma-buf.c index 1e16cbd61da2..61d6d62cc0d3 100644 --- a/drivers/base/dma-buf.c +++ b/drivers/base/dma-buf.c @@ -616,36 +616,35 @@ static int dma_buf_describe(struct seq_file *s) if (ret) return ret; - seq_printf(s, "\nDma-buf Objects:\n"); - seq_printf(s, "\texp_name\tsize\tflags\tmode\tcount\n"); + seq_puts(s, "\nDma-buf Objects:\n"); + seq_puts(s, "size\tflags\tmode\tcount\texp_name\n"); list_for_each_entry(buf_obj, &db_list.head, list_node) { ret = mutex_lock_interruptible(&buf_obj->lock); if (ret) { - seq_printf(s, - "\tERROR locking buffer object: skipping\n"); + seq_puts(s, + "\tERROR locking buffer object: skipping\n"); continue; } - seq_printf(s, "\t"); - - seq_printf(s, "\t%s\t%08zu\t%08x\t%08x\t%08ld\n", - buf_obj->exp_name, buf_obj->size, + seq_printf(s, "%08zu\t%08x\t%08x\t%08ld\t%s\n", + buf_obj->size, buf_obj->file->f_flags, buf_obj->file->f_mode, - (long)(buf_obj->file->f_count.counter)); + (long)(buf_obj->file->f_count.counter), + buf_obj->exp_name); - seq_printf(s, "\t\tAttached Devices:\n"); + seq_puts(s, "\tAttached Devices:\n"); attach_count = 0; list_for_each_entry(attach_obj, &buf_obj->attachments, node) { - seq_printf(s, "\t\t"); + seq_puts(s, "\t"); - seq_printf(s, "%s\n", attach_obj->dev->init_name); + seq_printf(s, "%s\n", dev_name(attach_obj->dev)); attach_count++; } - seq_printf(s, "\n\t\tTotal %d devices attached\n", + seq_printf(s, "Total %d devices attached\n\n", attach_count); count++; diff --git a/include/linux/dma-buf.h b/include/linux/dma-buf.h index dfac5ed31120..f886985a28b2 100644 --- a/include/linux/dma-buf.h +++ b/include/linux/dma-buf.h @@ -171,7 +171,7 @@ struct dma_buf *dma_buf_export_named(void *priv, const struct dma_buf_ops *ops, size_t size, int flags, const char *); #define dma_buf_export(priv, ops, size, flags) \ - dma_buf_export_named(priv, ops, size, flags, __FILE__) + dma_buf_export_named(priv, ops, size, flags, KBUILD_MODNAME) int dma_buf_fd(struct dma_buf *dmabuf, int flags); struct dma_buf *dma_buf_get(int fd); -- cgit v1.2.3 From a9f180345f5378ac87d80ed0bea55ba421d83859 Mon Sep 17 00:00:00 2001 From: Steven Noonan Date: Wed, 12 Feb 2014 23:01:07 -0800 Subject: compiler/gcc4: Make quirk for asm_volatile_goto() unconditional I started noticing problems with KVM guest destruction on Linux 3.12+, where guest memory wasn't being cleaned up. I bisected it down to the commit introducing the new 'asm goto'-based atomics, and found this quirk was later applied to those. Unfortunately, even with GCC 4.8.2 (which ostensibly fixed the known 'asm goto' bug) I am still getting some kind of miscompilation. If I enable the asm_volatile_goto quirk for my compiler, KVM guests are destroyed correctly and the memory is cleaned up. So make the quirk unconditional for now, until bug is found and fixed. Suggested-by: Linus Torvalds Signed-off-by: Steven Noonan Cc: Peter Zijlstra Cc: Steven Rostedt Cc: Jakub Jelinek Cc: Richard Henderson Cc: Andrew Morton Cc: Oleg Nesterov Cc: Link: http://lkml.kernel.org/r/1392274867-15236-1-git-send-email-steven@uplinklabs.net Link: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58670 Signed-off-by: Ingo Molnar --- include/linux/compiler-gcc4.h | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/compiler-gcc4.h b/include/linux/compiler-gcc4.h index ded429966c1f..2507fd2a1eb4 100644 --- a/include/linux/compiler-gcc4.h +++ b/include/linux/compiler-gcc4.h @@ -75,11 +75,7 @@ * * (asm goto is automatically volatile - the naming reflects this.) */ -#if GCC_VERSION <= 40801 -# define asm_volatile_goto(x...) do { asm goto(x); asm (""); } while (0) -#else -# define asm_volatile_goto(x...) do { asm goto(x); } while (0) -#endif +#define asm_volatile_goto(x...) do { asm goto(x); asm (""); } while (0) #ifdef CONFIG_ARCH_USE_BUILTIN_BSWAP #if GCC_VERSION >= 40400 -- cgit v1.2.3 From d206940319c41df4299db75ed56142177bb2e5f6 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 13 Feb 2014 23:09:11 +0100 Subject: net: core: introduce netif_skb_dev_features Will be used by upcoming ipv4 forward path change that needs to determine feature mask using skb->dst->dev instead of skb->dev. Signed-off-by: Florian Westphal Signed-off-by: David S. Miller --- include/linux/netdevice.h | 7 ++++++- net/core/dev.c | 22 ++++++++++++---------- 2 files changed, 18 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 440a02ee6f92..21d4e6be8949 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3068,7 +3068,12 @@ void netdev_change_features(struct net_device *dev); void netif_stacked_transfer_operstate(const struct net_device *rootdev, struct net_device *dev); -netdev_features_t netif_skb_features(struct sk_buff *skb); +netdev_features_t netif_skb_dev_features(struct sk_buff *skb, + const struct net_device *dev); +static inline netdev_features_t netif_skb_features(struct sk_buff *skb) +{ + return netif_skb_dev_features(skb, skb->dev); +} static inline bool net_gso_ok(netdev_features_t features, int gso_type) { diff --git a/net/core/dev.c b/net/core/dev.c index 4ad1b78c9c77..b1b0c8d4d7df 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2420,7 +2420,7 @@ EXPORT_SYMBOL(netdev_rx_csum_fault); * 2. No high memory really exists on this machine. */ -static int illegal_highdma(struct net_device *dev, struct sk_buff *skb) +static int illegal_highdma(const struct net_device *dev, struct sk_buff *skb) { #ifdef CONFIG_HIGHMEM int i; @@ -2495,34 +2495,36 @@ static int dev_gso_segment(struct sk_buff *skb, netdev_features_t features) } static netdev_features_t harmonize_features(struct sk_buff *skb, - netdev_features_t features) + const struct net_device *dev, + netdev_features_t features) { if (skb->ip_summed != CHECKSUM_NONE && !can_checksum_protocol(features, skb_network_protocol(skb))) { features &= ~NETIF_F_ALL_CSUM; - } else if (illegal_highdma(skb->dev, skb)) { + } else if (illegal_highdma(dev, skb)) { features &= ~NETIF_F_SG; } return features; } -netdev_features_t netif_skb_features(struct sk_buff *skb) +netdev_features_t netif_skb_dev_features(struct sk_buff *skb, + const struct net_device *dev) { __be16 protocol = skb->protocol; - netdev_features_t features = skb->dev->features; + netdev_features_t features = dev->features; - if (skb_shinfo(skb)->gso_segs > skb->dev->gso_max_segs) + if (skb_shinfo(skb)->gso_segs > dev->gso_max_segs) features &= ~NETIF_F_GSO_MASK; if (protocol == htons(ETH_P_8021Q) || protocol == htons(ETH_P_8021AD)) { struct vlan_ethhdr *veh = (struct vlan_ethhdr *)skb->data; protocol = veh->h_vlan_encapsulated_proto; } else if (!vlan_tx_tag_present(skb)) { - return harmonize_features(skb, features); + return harmonize_features(skb, dev, features); } - features &= (skb->dev->vlan_features | NETIF_F_HW_VLAN_CTAG_TX | + features &= (dev->vlan_features | NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX); if (protocol == htons(ETH_P_8021Q) || protocol == htons(ETH_P_8021AD)) @@ -2530,9 +2532,9 @@ netdev_features_t netif_skb_features(struct sk_buff *skb) NETIF_F_GEN_CSUM | NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX; - return harmonize_features(skb, features); + return harmonize_features(skb, dev, features); } -EXPORT_SYMBOL(netif_skb_features); +EXPORT_SYMBOL(netif_skb_dev_features); int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, struct netdev_queue *txq) -- cgit v1.2.3 From fe6cc55f3a9a053482a76f5a6b2257cee51b4663 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 13 Feb 2014 23:09:12 +0100 Subject: net: ip, ipv6: handle gso skbs in forwarding path Marcelo Ricardo Leitner reported problems when the forwarding link path has a lower mtu than the incoming one if the inbound interface supports GRO. Given: Host R1 R2 Host sends tcp stream which is routed via R1 and R2. R1 performs GRO. In this case, the kernel will fail to send ICMP fragmentation needed messages (or pkt too big for ipv6), as GSO packets currently bypass dstmtu checks in forward path. Instead, Linux tries to send out packets exceeding the mtu. When locking route MTU on Host (i.e., no ipv4 DF bit set), R1 does not fragment the packets when forwarding, and again tries to send out packets exceeding R1-R2 link mtu. This alters the forwarding dstmtu checks to take the individual gso segment lengths into account. For ipv6, we send out pkt too big error for gso if the individual segments are too big. For ipv4, we either send icmp fragmentation needed, or, if the DF bit is not set, perform software segmentation and let the output path create fragments when the packet is leaving the machine. It is not 100% correct as the error message will contain the headers of the GRO skb instead of the original/segmented one, but it seems to work fine in my (limited) tests. Eric Dumazet suggested to simply shrink mss via ->gso_size to avoid sofware segmentation. However it turns out that skb_segment() assumes skb nr_frags is related to mss size so we would BUG there. I don't want to mess with it considering Herbert and Eric disagree on what the correct behavior should be. Hannes Frederic Sowa notes that when we would shrink gso_size skb_segment would then also need to deal with the case where SKB_MAX_FRAGS would be exceeded. This uses sofware segmentation in the forward path when we hit ipv4 non-DF packets and the outgoing link mtu is too small. Its not perfect, but given the lack of bug reports wrt. GRO fwd being broken this is a rare case anyway. Also its not like this could not be improved later once the dust settles. Acked-by: Herbert Xu Reported-by: Marcelo Ricardo Leitner Signed-off-by: Florian Westphal Signed-off-by: David S. Miller --- include/linux/skbuff.h | 17 ++++++++++++ net/ipv4/ip_forward.c | 71 ++++++++++++++++++++++++++++++++++++++++++++++++-- net/ipv6/ip6_output.c | 17 ++++++++++-- 3 files changed, 101 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index f589c9af8cbf..3ebbbe7b6d05 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -2916,5 +2916,22 @@ static inline bool skb_head_is_locked(const struct sk_buff *skb) { return !skb->head_frag || skb_cloned(skb); } + +/** + * skb_gso_network_seglen - Return length of individual segments of a gso packet + * + * @skb: GSO skb + * + * skb_gso_network_seglen is used to determine the real size of the + * individual segments, including Layer3 (IP, IPv6) and L4 headers (TCP/UDP). + * + * The MAC/L2 header is not accounted for. + */ +static inline unsigned int skb_gso_network_seglen(const struct sk_buff *skb) +{ + unsigned int hdr_len = skb_transport_header(skb) - + skb_network_header(skb); + return hdr_len + skb_gso_transport_seglen(skb); +} #endif /* __KERNEL__ */ #endif /* _LINUX_SKBUFF_H */ diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c index e9f1217a8afd..f3869c186d97 100644 --- a/net/ipv4/ip_forward.c +++ b/net/ipv4/ip_forward.c @@ -39,6 +39,71 @@ #include #include +static bool ip_may_fragment(const struct sk_buff *skb) +{ + return unlikely((ip_hdr(skb)->frag_off & htons(IP_DF)) == 0) || + !skb->local_df; +} + +static bool ip_exceeds_mtu(const struct sk_buff *skb, unsigned int mtu) +{ + if (skb->len <= mtu || skb->local_df) + return false; + + if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu) + return false; + + return true; +} + +static bool ip_gso_exceeds_dst_mtu(const struct sk_buff *skb) +{ + unsigned int mtu; + + if (skb->local_df || !skb_is_gso(skb)) + return false; + + mtu = ip_dst_mtu_maybe_forward(skb_dst(skb), true); + + /* if seglen > mtu, do software segmentation for IP fragmentation on + * output. DF bit cannot be set since ip_forward would have sent + * icmp error. + */ + return skb_gso_network_seglen(skb) > mtu; +} + +/* called if GSO skb needs to be fragmented on forward */ +static int ip_forward_finish_gso(struct sk_buff *skb) +{ + struct dst_entry *dst = skb_dst(skb); + netdev_features_t features; + struct sk_buff *segs; + int ret = 0; + + features = netif_skb_dev_features(skb, dst->dev); + segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK); + if (IS_ERR(segs)) { + kfree_skb(skb); + return -ENOMEM; + } + + consume_skb(skb); + + do { + struct sk_buff *nskb = segs->next; + int err; + + segs->next = NULL; + err = dst_output(segs); + + if (err && ret == 0) + ret = err; + segs = nskb; + } while (segs); + + return ret; +} + static int ip_forward_finish(struct sk_buff *skb) { struct ip_options *opt = &(IPCB(skb)->opt); @@ -49,6 +114,9 @@ static int ip_forward_finish(struct sk_buff *skb) if (unlikely(opt->optlen)) ip_forward_options(skb); + if (ip_gso_exceeds_dst_mtu(skb)) + return ip_forward_finish_gso(skb); + return dst_output(skb); } @@ -91,8 +159,7 @@ int ip_forward(struct sk_buff *skb) IPCB(skb)->flags |= IPSKB_FORWARDED; mtu = ip_dst_mtu_maybe_forward(&rt->dst, true); - if (unlikely(skb->len > mtu && !skb_is_gso(skb) && - (ip_hdr(skb)->frag_off & htons(IP_DF))) && !skb->local_df) { + if (!ip_may_fragment(skb) && ip_exceeds_mtu(skb, mtu)) { IP_INC_STATS(dev_net(rt->dst.dev), IPSTATS_MIB_FRAGFAILS); icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu)); diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index ef02b26ccf81..070a2fae2375 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -342,6 +342,20 @@ static unsigned int ip6_dst_mtu_forward(const struct dst_entry *dst) return mtu; } +static bool ip6_pkt_too_big(const struct sk_buff *skb, unsigned int mtu) +{ + if (skb->len <= mtu || skb->local_df) + return false; + + if (IP6CB(skb)->frag_max_size && IP6CB(skb)->frag_max_size > mtu) + return true; + + if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu) + return false; + + return true; +} + int ip6_forward(struct sk_buff *skb) { struct dst_entry *dst = skb_dst(skb); @@ -466,8 +480,7 @@ int ip6_forward(struct sk_buff *skb) if (mtu < IPV6_MIN_MTU) mtu = IPV6_MIN_MTU; - if ((!skb->local_df && skb->len > mtu && !skb_is_gso(skb)) || - (IP6CB(skb)->frag_max_size && IP6CB(skb)->frag_max_size > mtu)) { + if (ip6_pkt_too_big(skb, mtu)) { /* Again, force OUTPUT device used as source address */ skb->dev = dst->dev; icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu); -- cgit v1.2.3 From b4a26a27287a7f81933ba016aeed6c69dd155323 Mon Sep 17 00:00:00 2001 From: Moni Shoua Date: Sun, 9 Feb 2014 11:54:34 +0200 Subject: IB: Report using RoCE IP based gids in port caps For userspace RoCE UD QPs we need to know the GID format that the kernel uses, e.g when working over older kernels. For that end, add a new port capability IB_PORT_IP_BASED_GIDS and report it when query port is issued. Signed-off-by: Moni Shoua Signed-off-by: Matan Barak Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx4/main.c | 2 +- drivers/infiniband/hw/ocrdma/ocrdma_verbs.c | 2 +- include/rdma/ib_verbs.h | 3 ++- 3 files changed, 4 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index c2702f549f10..5d4088ad6bd5 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -347,7 +347,7 @@ static int eth_link_query_port(struct ib_device *ibdev, u8 port, props->active_width = (((u8 *)mailbox->buf)[5] == 0x40) ? IB_WIDTH_4X : IB_WIDTH_1X; props->active_speed = IB_SPEED_QDR; - props->port_cap_flags = IB_PORT_CM_SUP; + props->port_cap_flags = IB_PORT_CM_SUP | IB_PORT_IP_BASED_GIDS; props->gid_tbl_len = mdev->dev->caps.gid_table_len[port]; props->max_msg_sz = mdev->dev->caps.max_msg_sz; props->pkey_tbl_len = 1; diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c index aa92f40c9d50..7890e7b14d81 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c @@ -176,7 +176,7 @@ int ocrdma_query_port(struct ib_device *ibdev, props->port_cap_flags = IB_PORT_CM_SUP | IB_PORT_REINIT_SUP | - IB_PORT_DEVICE_MGMT_SUP | IB_PORT_VENDOR_CLASS_SUP; + IB_PORT_DEVICE_MGMT_SUP | IB_PORT_VENDOR_CLASS_SUP | IB_PORT_IP_BASED_GIDS; props->gid_tbl_len = OCRDMA_MAX_SGID; props->pkey_tbl_len = 1; props->bad_pkey_cntr = 0; diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 8d4a1c06f7e4..6793f32ccb58 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -226,7 +226,8 @@ enum ib_port_cap_flags { IB_PORT_CAP_MASK_NOTICE_SUP = 1 << 22, IB_PORT_BOOT_MGMT_SUP = 1 << 23, IB_PORT_LINK_LATENCY_SUP = 1 << 24, - IB_PORT_CLIENT_REG_SUP = 1 << 25 + IB_PORT_CLIENT_REG_SUP = 1 << 25, + IB_PORT_IP_BASED_GIDS = 1 << 26 }; enum ib_port_width { -- cgit v1.2.3 From 6ecde51dd7894ffe2f959cca1fea3ea2b9ee2394 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Thu, 13 Feb 2014 20:45:17 -0800 Subject: mlx5: Add include of because of kzalloc()/kfree() use On some architectures (for example, arm), we don't end up indirectly pulling in the declaration of kzalloc() and kfree(), and so building anything that includes breaks. Fix this by adding an explicit include to get the declaration. Reported-by: kbuild test robot Signed-off-by: Roland Dreier --- include/linux/mlx5/driver.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 32cb18c399c2..130bc8d77fa5 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -38,8 +38,10 @@ #include #include #include +#include #include #include + #include #include -- cgit v1.2.3 From 11bcac89c0d73dea42f1cb8646b532035796a5d6 Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Fri, 14 Feb 2014 13:42:13 -0800 Subject: Revert "btrfs: add ioctl to export size of global metadata reservation" This reverts commit 01e219e8069516cdb98594d417b8bb8d906ed30d. David Sterba found a different way to provide these features without adding a new ioctl. We haven't released any progs with this ioctl yet, so I'm taking this out for now until we finalize things. Signed-off-by: Chris Mason Signed-off-by: David Sterba CC: Jeff Mahoney --- fs/btrfs/ioctl.c | 16 ---------------- include/uapi/linux/btrfs.h | 1 - 2 files changed, 17 deletions(-) (limited to 'include') diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index ebdd866d4cfd..9a9044585da7 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -3559,20 +3559,6 @@ out: return ret; } -static long btrfs_ioctl_global_rsv(struct btrfs_root *root, void __user *arg) -{ - struct btrfs_block_rsv *block_rsv = &root->fs_info->global_block_rsv; - u64 reserved; - - spin_lock(&block_rsv->lock); - reserved = block_rsv->reserved; - spin_unlock(&block_rsv->lock); - - if (arg && copy_to_user(arg, &reserved, sizeof(reserved))) - return -EFAULT; - return 0; -} - /* * there are many ways the trans_start and trans_end ioctls can lead * to deadlocks. They should only be used by applications that @@ -4779,8 +4765,6 @@ long btrfs_ioctl(struct file *file, unsigned int return btrfs_ioctl_logical_to_ino(root, argp); case BTRFS_IOC_SPACE_INFO: return btrfs_ioctl_space_info(root, argp); - case BTRFS_IOC_GLOBAL_RSV: - return btrfs_ioctl_global_rsv(root, argp); case BTRFS_IOC_SYNC: { int ret; diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h index 1b8a0f4c9590..b4d69092fbdb 100644 --- a/include/uapi/linux/btrfs.h +++ b/include/uapi/linux/btrfs.h @@ -558,7 +558,6 @@ static inline char *btrfs_err_str(enum btrfs_err_code err_code) #define BTRFS_IOC_DEFAULT_SUBVOL _IOW(BTRFS_IOCTL_MAGIC, 19, __u64) #define BTRFS_IOC_SPACE_INFO _IOWR(BTRFS_IOCTL_MAGIC, 20, \ struct btrfs_ioctl_space_args) -#define BTRFS_IOC_GLOBAL_RSV _IOR(BTRFS_IOCTL_MAGIC, 20, __u64) #define BTRFS_IOC_START_SYNC _IOR(BTRFS_IOCTL_MAGIC, 24, __u64) #define BTRFS_IOC_WAIT_SYNC _IOW(BTRFS_IOCTL_MAGIC, 22, __u64) #define BTRFS_IOC_SNAP_CREATE_V2 _IOW(BTRFS_IOCTL_MAGIC, 23, \ -- cgit v1.2.3 From 56eecdb912b536a4fa97fb5bfe5a940a54d79be6 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Wed, 12 Feb 2014 09:13:38 +0530 Subject: mm: Use ptep/pmdp_set_numa() for updating _PAGE_NUMA bit Archs like ppc64 doesn't do tlb flush in set_pte/pmd functions when using a hash table MMU for various reasons (the flush is handled as part of the PTE modification when necessary). ppc64 thus doesn't implement flush_tlb_range for hash based MMUs. Additionally ppc64 require the tlb flushing to be batched within ptl locks. The reason to do that is to ensure that the hash page table is in sync with linux page table. We track the hpte index in linux pte and if we clear them without flushing hash and drop the ptl lock, we can have another cpu update the pte and can end up with duplicate entry in the hash table, which is fatal. We also want to keep set_pte_at simpler by not requiring them to do hash flush for performance reason. We do that by assuming that set_pte_at() is never *ever* called on a PTE that is already valid. This was the case until the NUMA code went in which broke that assumption. Fix that by introducing a new pair of helpers to set _PAGE_NUMA in a way similar to ptep/pmdp_set_wrprotect(), with a generic implementation using set_pte_at() and a powerpc specific one using the appropriate mechanism needed to keep the hash table in sync. Acked-by: Mel Gorman Reviewed-by: Rik van Riel Signed-off-by: Aneesh Kumar K.V Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/pgtable.h | 22 +++++++++++++++++++++ include/asm-generic/pgtable.h | 39 ++++++++++++++++++++++++++++++++++++++ mm/huge_memory.c | 9 ++------- mm/mprotect.c | 4 +--- 4 files changed, 64 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/arch/powerpc/include/asm/pgtable.h b/arch/powerpc/include/asm/pgtable.h index f83b6f3e1b39..3ebb188c3ff5 100644 --- a/arch/powerpc/include/asm/pgtable.h +++ b/arch/powerpc/include/asm/pgtable.h @@ -75,12 +75,34 @@ static inline pte_t pte_mknuma(pte_t pte) return pte; } +#define ptep_set_numa ptep_set_numa +static inline void ptep_set_numa(struct mm_struct *mm, unsigned long addr, + pte_t *ptep) +{ + if ((pte_val(*ptep) & _PAGE_PRESENT) == 0) + VM_BUG_ON(1); + + pte_update(mm, addr, ptep, _PAGE_PRESENT, _PAGE_NUMA, 0); + return; +} + #define pmd_numa pmd_numa static inline int pmd_numa(pmd_t pmd) { return pte_numa(pmd_pte(pmd)); } +#define pmdp_set_numa pmdp_set_numa +static inline void pmdp_set_numa(struct mm_struct *mm, unsigned long addr, + pmd_t *pmdp) +{ + if ((pmd_val(*pmdp) & _PAGE_PRESENT) == 0) + VM_BUG_ON(1); + + pmd_hugepage_update(mm, addr, pmdp, _PAGE_PRESENT, _PAGE_NUMA); + return; +} + #define pmd_mknonnuma pmd_mknonnuma static inline pmd_t pmd_mknonnuma(pmd_t pmd) { diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h index 8e4f41d9af4d..34c7bdc06014 100644 --- a/include/asm-generic/pgtable.h +++ b/include/asm-generic/pgtable.h @@ -701,6 +701,18 @@ static inline pte_t pte_mknuma(pte_t pte) } #endif +#ifndef ptep_set_numa +static inline void ptep_set_numa(struct mm_struct *mm, unsigned long addr, + pte_t *ptep) +{ + pte_t ptent = *ptep; + + ptent = pte_mknuma(ptent); + set_pte_at(mm, addr, ptep, ptent); + return; +} +#endif + #ifndef pmd_mknuma static inline pmd_t pmd_mknuma(pmd_t pmd) { @@ -708,6 +720,18 @@ static inline pmd_t pmd_mknuma(pmd_t pmd) return pmd_clear_flags(pmd, _PAGE_PRESENT); } #endif + +#ifndef pmdp_set_numa +static inline void pmdp_set_numa(struct mm_struct *mm, unsigned long addr, + pmd_t *pmdp) +{ + pmd_t pmd = *pmdp; + + pmd = pmd_mknuma(pmd); + set_pmd_at(mm, addr, pmdp, pmd); + return; +} +#endif #else extern int pte_numa(pte_t pte); extern int pmd_numa(pmd_t pmd); @@ -715,6 +739,8 @@ extern pte_t pte_mknonnuma(pte_t pte); extern pmd_t pmd_mknonnuma(pmd_t pmd); extern pte_t pte_mknuma(pte_t pte); extern pmd_t pmd_mknuma(pmd_t pmd); +extern void ptep_set_numa(struct mm_struct *mm, unsigned long addr, pte_t *ptep); +extern void pmdp_set_numa(struct mm_struct *mm, unsigned long addr, pmd_t *pmdp); #endif /* CONFIG_ARCH_USES_NUMA_PROT_NONE */ #else static inline int pmd_numa(pmd_t pmd) @@ -742,10 +768,23 @@ static inline pte_t pte_mknuma(pte_t pte) return pte; } +static inline void ptep_set_numa(struct mm_struct *mm, unsigned long addr, + pte_t *ptep) +{ + return; +} + + static inline pmd_t pmd_mknuma(pmd_t pmd) { return pmd; } + +static inline void pmdp_set_numa(struct mm_struct *mm, unsigned long addr, + pmd_t *pmdp) +{ + return ; +} #endif /* CONFIG_NUMA_BALANCING */ #endif /* CONFIG_MMU */ diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 82166bf974e1..da23eb96779f 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1545,6 +1545,7 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd, entry = pmd_mknonnuma(entry); entry = pmd_modify(entry, newprot); ret = HPAGE_PMD_NR; + set_pmd_at(mm, addr, pmd, entry); BUG_ON(pmd_write(entry)); } else { struct page *page = pmd_page(*pmd); @@ -1557,16 +1558,10 @@ int change_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd, */ if (!is_huge_zero_page(page) && !pmd_numa(*pmd)) { - entry = *pmd; - entry = pmd_mknuma(entry); + pmdp_set_numa(mm, addr, pmd); ret = HPAGE_PMD_NR; } } - - /* Set PMD if cleared earlier */ - if (ret == HPAGE_PMD_NR) - set_pmd_at(mm, addr, pmd, entry); - spin_unlock(ptl); } diff --git a/mm/mprotect.c b/mm/mprotect.c index 33eab902f10e..769a67a15803 100644 --- a/mm/mprotect.c +++ b/mm/mprotect.c @@ -69,12 +69,10 @@ static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd, } else { struct page *page; - ptent = *pte; page = vm_normal_page(vma, addr, oldpte); if (page && !PageKsm(page)) { if (!pte_numa(oldpte)) { - ptent = pte_mknuma(ptent); - set_pte_at(mm, addr, pte, ptent); + ptep_set_numa(mm, addr, pte); updated = true; } } -- cgit v1.2.3 From ef2820a735f74ea60335f8ba3801b844f0cb184d Mon Sep 17 00:00:00 2001 From: Matija Glavinic Pecotic Date: Fri, 14 Feb 2014 14:51:18 +0100 Subject: net: sctp: Fix a_rwnd/rwnd management to reflect real state of the receiver's buffer Implementation of (a)rwnd calculation might lead to severe performance issues and associations completely stalling. These problems are described and solution is proposed which improves lksctp's robustness in congestion state. 1) Sudden drop of a_rwnd and incomplete window recovery afterwards Data accounted in sctp_assoc_rwnd_decrease takes only payload size (sctp data), but size of sk_buff, which is blamed against receiver buffer, is not accounted in rwnd. Theoretically, this should not be the problem as actual size of buffer is double the amount requested on the socket (SO_RECVBUF). Problem here is that this will have bad scaling for data which is less then sizeof sk_buff. E.g. in 4G (LTE) networks, link interfacing radio side will have a large portion of traffic of this size (less then 100B). An example of sudden drop and incomplete window recovery is given below. Node B exhibits problematic behavior. Node A initiates association and B is configured to advertise rwnd of 10000. A sends messages of size 43B (size of typical sctp message in 4G (LTE) network). On B data is left in buffer by not reading socket in userspace. Lets examine when we will hit pressure state and declare rwnd to be 0 for scenario with above stated parameters (rwnd == 10000, chunk size == 43, each chunk is sent in separate sctp packet) Logic is implemented in sctp_assoc_rwnd_decrease: socket_buffer (see below) is maximum size which can be held in socket buffer (sk_rcvbuf). current_alloced is amount of data currently allocated (rx_count) A simple expression is given for which it will be examined after how many packets for above stated parameters we enter pressure state: We start by condition which has to be met in order to enter pressure state: socket_buffer < currently_alloced; currently_alloced is represented as size of sctp packets received so far and not yet delivered to userspace. x is the number of chunks/packets (since there is no bundling, and each chunk is delivered in separate packet, we can observe each chunk also as sctp packet, and what is important here, having its own sk_buff): socket_buffer < x*each_sctp_packet; each_sctp_packet is sctp chunk size + sizeof(struct sk_buff). socket_buffer is twice the amount of initially requested size of socket buffer, which is in case of sctp, twice the a_rwnd requested: 2*rwnd < x*(payload+sizeof(struc sk_buff)); sizeof(struct sk_buff) is 190 (3.13.0-rc4+). Above is stated that rwnd is 10000 and each payload size is 43 20000 < x(43+190); x > 20000/233; x ~> 84; After ~84 messages, pressure state is entered and 0 rwnd is advertised while received 84*43B ~= 3612B sctp data. This is why external observer notices sudden drop from 6474 to 0, as it will be now shown in example: IP A.34340 > B.12345: sctp (1) [INIT] [init tag: 1875509148] [rwnd: 81920] [OS: 10] [MIS: 65535] [init TSN: 1096057017] IP B.12345 > A.34340: sctp (1) [INIT ACK] [init tag: 3198966556] [rwnd: 10000] [OS: 10] [MIS: 10] [init TSN: 902132839] IP A.34340 > B.12345: sctp (1) [COOKIE ECHO] IP B.12345 > A.34340: sctp (1) [COOKIE ACK] IP A.34340 > B.12345: sctp (1) [DATA] (B)(E) [TSN: 1096057017] [SID: 0] [SSEQ 0] [PPID 0x18] IP B.12345 > A.34340: sctp (1) [SACK] [cum ack 1096057017] [a_rwnd 9957] [#gap acks 0] [#dup tsns 0] IP A.34340 > B.12345: sctp (1) [DATA] (B)(E) [TSN: 1096057018] [SID: 0] [SSEQ 1] [PPID 0x18] IP B.12345 > A.34340: sctp (1) [SACK] [cum ack 1096057018] [a_rwnd 9957] [#gap acks 0] [#dup tsns 0] IP A.34340 > B.12345: sctp (1) [DATA] (B)(E) [TSN: 1096057019] [SID: 0] [SSEQ 2] [PPID 0x18] IP B.12345 > A.34340: sctp (1) [SACK] [cum ack 1096057019] [a_rwnd 9914] [#gap acks 0] [#dup tsns 0] <...> IP A.34340 > B.12345: sctp (1) [DATA] (B)(E) [TSN: 1096057098] [SID: 0] [SSEQ 81] [PPID 0x18] IP B.12345 > A.34340: sctp (1) [SACK] [cum ack 1096057098] [a_rwnd 6517] [#gap acks 0] [#dup tsns 0] IP A.34340 > B.12345: sctp (1) [DATA] (B)(E) [TSN: 1096057099] [SID: 0] [SSEQ 82] [PPID 0x18] IP B.12345 > A.34340: sctp (1) [SACK] [cum ack 1096057099] [a_rwnd 6474] [#gap acks 0] [#dup tsns 0] IP A.34340 > B.12345: sctp (1) [DATA] (B)(E) [TSN: 1096057100] [SID: 0] [SSEQ 83] [PPID 0x18] --> Sudden drop IP B.12345 > A.34340: sctp (1) [SACK] [cum ack 1096057100] [a_rwnd 0] [#gap acks 0] [#dup tsns 0] At this point, rwnd_press stores current rwnd value so it can be later restored in sctp_assoc_rwnd_increase. This however doesn't happen as condition to start slowly increasing rwnd until rwnd_press is returned to rwnd is never met. This condition is not met since rwnd, after it hit 0, must first reach rwnd_press by adding amount which is read from userspace. Let us observe values in above example. Initial a_rwnd is 10000, pressure was hit when rwnd was ~6500 and the amount of actual sctp data currently waiting to be delivered to userspace is ~3500. When userspace starts to read, sctp_assoc_rwnd_increase will be blamed only for sctp data, which is ~3500. Condition is never met, and when userspace reads all data, rwnd stays on 3569. IP B.12345 > A.34340: sctp (1) [SACK] [cum ack 1096057100] [a_rwnd 1505] [#gap acks 0] [#dup tsns 0] IP B.12345 > A.34340: sctp (1) [SACK] [cum ack 1096057100] [a_rwnd 3010] [#gap acks 0] [#dup tsns 0] IP A.34340 > B.12345: sctp (1) [DATA] (B)(E) [TSN: 1096057101] [SID: 0] [SSEQ 84] [PPID 0x18] IP B.12345 > A.34340: sctp (1) [SACK] [cum ack 1096057101] [a_rwnd 3569] [#gap acks 0] [#dup tsns 0] --> At this point userspace read everything, rwnd recovered only to 3569 IP A.34340 > B.12345: sctp (1) [DATA] (B)(E) [TSN: 1096057102] [SID: 0] [SSEQ 85] [PPID 0x18] IP B.12345 > A.34340: sctp (1) [SACK] [cum ack 1096057102] [a_rwnd 3569] [#gap acks 0] [#dup tsns 0] Reproduction is straight forward, it is enough for sender to send packets of size less then sizeof(struct sk_buff) and receiver keeping them in its buffers. 2) Minute size window for associations sharing the same socket buffer In case multiple associations share the same socket, and same socket buffer (sctp.rcvbuf_policy == 0), different scenarios exist in which congestion on one of the associations can permanently drop rwnd of other association(s). Situation will be typically observed as one association suddenly having rwnd dropped to size of last packet received and never recovering beyond that point. Different scenarios will lead to it, but all have in common that one of the associations (let it be association from 1)) nearly depleted socket buffer, and the other association blames socket buffer just for the amount enough to start the pressure. This association will enter pressure state, set rwnd_press and announce 0 rwnd. When data is read by userspace, similar situation as in 1) will occur, rwnd will increase just for the size read by userspace but rwnd_press will be high enough so that association doesn't have enough credit to reach rwnd_press and restore to previous state. This case is special case of 1), being worse as there is, in the worst case, only one packet in buffer for which size rwnd will be increased. Consequence is association which has very low maximum rwnd ('minute size', in our case down to 43B - size of packet which caused pressure) and as such unusable. Scenario happened in the field and labs frequently after congestion state (link breaks, different probabilities of packet drop, packet reordering) and with scenario 1) preceding. Here is given a deterministic scenario for reproduction: >From node A establish two associations on the same socket, with rcvbuf_policy being set to share one common buffer (sctp.rcvbuf_policy == 0). On association 1 repeat scenario from 1), that is, bring it down to 0 and restore up. Observe scenario 1). Use small payload size (here we use 43). Once rwnd is 'recovered', bring it down close to 0, as in just one more packet would close it. This has as a consequence that association number 2 is able to receive (at least) one more packet which will bring it in pressure state. E.g. if association 2 had rwnd of 10000, packet received was 43, and we enter at this point into pressure, rwnd_press will have 9957. Once payload is delivered to userspace, rwnd will increase for 43, but conditions to restore rwnd to original state, just as in 1), will never be satisfied. --> Association 1, between A.y and B.12345 IP A.55915 > B.12345: sctp (1) [INIT] [init tag: 836880897] [rwnd: 10000] [OS: 10] [MIS: 65535] [init TSN: 4032536569] IP B.12345 > A.55915: sctp (1) [INIT ACK] [init tag: 2873310749] [rwnd: 81920] [OS: 10] [MIS: 10] [init TSN: 3799315613] IP A.55915 > B.12345: sctp (1) [COOKIE ECHO] IP B.12345 > A.55915: sctp (1) [COOKIE ACK] --> Association 2, between A.z and B.12346 IP A.55915 > B.12346: sctp (1) [INIT] [init tag: 534798321] [rwnd: 10000] [OS: 10] [MIS: 65535] [init TSN: 2099285173] IP B.12346 > A.55915: sctp (1) [INIT ACK] [init tag: 516668823] [rwnd: 81920] [OS: 10] [MIS: 10] [init TSN: 3676403240] IP A.55915 > B.12346: sctp (1) [COOKIE ECHO] IP B.12346 > A.55915: sctp (1) [COOKIE ACK] --> Deplete socket buffer by sending messages of size 43B over association 1 IP B.12345 > A.55915: sctp (1) [DATA] (B)(E) [TSN: 3799315613] [SID: 0] [SSEQ 0] [PPID 0x18] IP A.55915 > B.12345: sctp (1) [SACK] [cum ack 3799315613] [a_rwnd 9957] [#gap acks 0] [#dup tsns 0] <...> IP A.55915 > B.12345: sctp (1) [SACK] [cum ack 3799315696] [a_rwnd 6388] [#gap acks 0] [#dup tsns 0] IP B.12345 > A.55915: sctp (1) [DATA] (B)(E) [TSN: 3799315697] [SID: 0] [SSEQ 84] [PPID 0x18] IP A.55915 > B.12345: sctp (1) [SACK] [cum ack 3799315697] [a_rwnd 6345] [#gap acks 0] [#dup tsns 0] --> Sudden drop on 1 IP B.12345 > A.55915: sctp (1) [DATA] (B)(E) [TSN: 3799315698] [SID: 0] [SSEQ 85] [PPID 0x18] IP A.55915 > B.12345: sctp (1) [SACK] [cum ack 3799315698] [a_rwnd 0] [#gap acks 0] [#dup tsns 0] --> Here userspace read, rwnd 'recovered' to 3698, now deplete again using association 1 so there is place in buffer for only one more packet IP B.12345 > A.55915: sctp (1) [DATA] (B)(E) [TSN: 3799315799] [SID: 0] [SSEQ 186] [PPID 0x18] IP A.55915 > B.12345: sctp (1) [SACK] [cum ack 3799315799] [a_rwnd 86] [#gap acks 0] [#dup tsns 0] IP B.12345 > A.55915: sctp (1) [DATA] (B)(E) [TSN: 3799315800] [SID: 0] [SSEQ 187] [PPID 0x18] IP A.55915 > B.12345: sctp (1) [SACK] [cum ack 3799315800] [a_rwnd 43] [#gap acks 0] [#dup tsns 0] --> Socket buffer is almost depleted, but there is space for one more packet, send them over association 2, size 43B IP B.12346 > A.55915: sctp (1) [DATA] (B)(E) [TSN: 3676403240] [SID: 0] [SSEQ 0] [PPID 0x18] IP A.55915 > B.12346: sctp (1) [SACK] [cum ack 3676403240] [a_rwnd 0] [#gap acks 0] [#dup tsns 0] --> Immediate drop IP A.60995 > B.12346: sctp (1) [SACK] [cum ack 387491510] [a_rwnd 0] [#gap acks 0] [#dup tsns 0] --> Read everything from the socket, both association recover up to maximum rwnd they are capable of reaching, note that association 1 recovered up to 3698, and association 2 recovered only to 43 IP A.55915 > B.12345: sctp (1) [SACK] [cum ack 3799315800] [a_rwnd 1548] [#gap acks 0] [#dup tsns 0] IP A.55915 > B.12345: sctp (1) [SACK] [cum ack 3799315800] [a_rwnd 3053] [#gap acks 0] [#dup tsns 0] IP B.12345 > A.55915: sctp (1) [DATA] (B)(E) [TSN: 3799315801] [SID: 0] [SSEQ 188] [PPID 0x18] IP A.55915 > B.12345: sctp (1) [SACK] [cum ack 3799315801] [a_rwnd 3698] [#gap acks 0] [#dup tsns 0] IP B.12346 > A.55915: sctp (1) [DATA] (B)(E) [TSN: 3676403241] [SID: 0] [SSEQ 1] [PPID 0x18] IP A.55915 > B.12346: sctp (1) [SACK] [cum ack 3676403241] [a_rwnd 43] [#gap acks 0] [#dup tsns 0] A careful reader might wonder why it is necessary to reproduce 1) prior reproduction of 2). It is simply easier to observe when to send packet over association 2 which will push association into the pressure state. Proposed solution: Both problems share the same root cause, and that is improper scaling of socket buffer with rwnd. Solution in which sizeof(sk_buff) is taken into concern while calculating rwnd is not possible due to fact that there is no linear relationship between amount of data blamed in increase/decrease with IP packet in which payload arrived. Even in case such solution would be followed, complexity of the code would increase. Due to nature of current rwnd handling, slow increase (in sctp_assoc_rwnd_increase) of rwnd after pressure state is entered is rationale, but it gives false representation to the sender of current buffer space. Furthermore, it implements additional congestion control mechanism which is defined on implementation, and not on standard basis. Proposed solution simplifies whole algorithm having on mind definition from rfc: o Receiver Window (rwnd): This gives the sender an indication of the space available in the receiver's inbound buffer. Core of the proposed solution is given with these lines: sctp_assoc_rwnd_update: if ((asoc->base.sk->sk_rcvbuf - rx_count) > 0) asoc->rwnd = (asoc->base.sk->sk_rcvbuf - rx_count) >> 1; else asoc->rwnd = 0; We advertise to sender (half of) actual space we have. Half is in the braces depending whether you would like to observe size of socket buffer as SO_RECVBUF or twice the amount, i.e. size is the one visible from userspace, that is, from kernelspace. In this way sender is given with good approximation of our buffer space, regardless of the buffer policy - we always advertise what we have. Proposed solution fixes described problems and removes necessity for rwnd restoration algorithm. Finally, as proposed solution is simplification, some lines of code, along with some bytes in struct sctp_association are saved. Version 2 of the patch addressed comments from Vlad. Name of the function is set to be more descriptive, and two parts of code are changed, in one removing the superfluous call to sctp_assoc_rwnd_update since call would not result in update of rwnd, and the other being reordering of the code in a way that call to sctp_assoc_rwnd_update updates rwnd. Version 3 corrected change introduced in v2 in a way that existing function is not reordered/copied in line, but it is correctly called. Thanks Vlad for suggesting. Signed-off-by: Matija Glavinic Pecotic Reviewed-by: Alexander Sverdlin Acked-by: Vlad Yasevich Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 14 +------- net/sctp/associola.c | 82 ++++++++++------------------------------------ net/sctp/sm_statefuns.c | 2 +- net/sctp/socket.c | 6 ---- net/sctp/ulpevent.c | 8 +++-- 5 files changed, 25 insertions(+), 87 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index d992ca3145fe..6ee76c804893 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1653,17 +1653,6 @@ struct sctp_association { /* This is the last advertised value of rwnd over a SACK chunk. */ __u32 a_rwnd; - /* Number of bytes by which the rwnd has slopped. The rwnd is allowed - * to slop over a maximum of the association's frag_point. - */ - __u32 rwnd_over; - - /* Keeps treack of rwnd pressure. This happens when we have - * a window, but not recevie buffer (i.e small packets). This one - * is releases slowly (1 PMTU at a time ). - */ - __u32 rwnd_press; - /* This is the sndbuf size in use for the association. * This corresponds to the sndbuf size for the association, * as specified in the sk->sndbuf. @@ -1892,8 +1881,7 @@ void sctp_assoc_update(struct sctp_association *old, __u32 sctp_association_get_next_tsn(struct sctp_association *); void sctp_assoc_sync_pmtu(struct sock *, struct sctp_association *); -void sctp_assoc_rwnd_increase(struct sctp_association *, unsigned int); -void sctp_assoc_rwnd_decrease(struct sctp_association *, unsigned int); +void sctp_assoc_rwnd_update(struct sctp_association *, bool); void sctp_assoc_set_primary(struct sctp_association *, struct sctp_transport *); void sctp_assoc_del_nonprimary_peers(struct sctp_association *, diff --git a/net/sctp/associola.c b/net/sctp/associola.c index 5ae609200674..f558433537b8 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -1367,44 +1367,35 @@ static inline bool sctp_peer_needs_update(struct sctp_association *asoc) return false; } -/* Increase asoc's rwnd by len and send any window update SACK if needed. */ -void sctp_assoc_rwnd_increase(struct sctp_association *asoc, unsigned int len) +/* Update asoc's rwnd for the approximated state in the buffer, + * and check whether SACK needs to be sent. + */ +void sctp_assoc_rwnd_update(struct sctp_association *asoc, bool update_peer) { + int rx_count; struct sctp_chunk *sack; struct timer_list *timer; - if (asoc->rwnd_over) { - if (asoc->rwnd_over >= len) { - asoc->rwnd_over -= len; - } else { - asoc->rwnd += (len - asoc->rwnd_over); - asoc->rwnd_over = 0; - } - } else { - asoc->rwnd += len; - } + if (asoc->ep->rcvbuf_policy) + rx_count = atomic_read(&asoc->rmem_alloc); + else + rx_count = atomic_read(&asoc->base.sk->sk_rmem_alloc); - /* If we had window pressure, start recovering it - * once our rwnd had reached the accumulated pressure - * threshold. The idea is to recover slowly, but up - * to the initial advertised window. - */ - if (asoc->rwnd_press && asoc->rwnd >= asoc->rwnd_press) { - int change = min(asoc->pathmtu, asoc->rwnd_press); - asoc->rwnd += change; - asoc->rwnd_press -= change; - } + if ((asoc->base.sk->sk_rcvbuf - rx_count) > 0) + asoc->rwnd = (asoc->base.sk->sk_rcvbuf - rx_count) >> 1; + else + asoc->rwnd = 0; - pr_debug("%s: asoc:%p rwnd increased by %d to (%u, %u) - %u\n", - __func__, asoc, len, asoc->rwnd, asoc->rwnd_over, - asoc->a_rwnd); + pr_debug("%s: asoc:%p rwnd=%u, rx_count=%d, sk_rcvbuf=%d\n", + __func__, asoc, asoc->rwnd, rx_count, + asoc->base.sk->sk_rcvbuf); /* Send a window update SACK if the rwnd has increased by at least the * minimum of the association's PMTU and half of the receive buffer. * The algorithm used is similar to the one described in * Section 4.2.3.3 of RFC 1122. */ - if (sctp_peer_needs_update(asoc)) { + if (update_peer && sctp_peer_needs_update(asoc)) { asoc->a_rwnd = asoc->rwnd; pr_debug("%s: sending window update SACK- asoc:%p rwnd:%u " @@ -1426,45 +1417,6 @@ void sctp_assoc_rwnd_increase(struct sctp_association *asoc, unsigned int len) } } -/* Decrease asoc's rwnd by len. */ -void sctp_assoc_rwnd_decrease(struct sctp_association *asoc, unsigned int len) -{ - int rx_count; - int over = 0; - - if (unlikely(!asoc->rwnd || asoc->rwnd_over)) - pr_debug("%s: association:%p has asoc->rwnd:%u, " - "asoc->rwnd_over:%u!\n", __func__, asoc, - asoc->rwnd, asoc->rwnd_over); - - if (asoc->ep->rcvbuf_policy) - rx_count = atomic_read(&asoc->rmem_alloc); - else - rx_count = atomic_read(&asoc->base.sk->sk_rmem_alloc); - - /* If we've reached or overflowed our receive buffer, announce - * a 0 rwnd if rwnd would still be positive. Store the - * the potential pressure overflow so that the window can be restored - * back to original value. - */ - if (rx_count >= asoc->base.sk->sk_rcvbuf) - over = 1; - - if (asoc->rwnd >= len) { - asoc->rwnd -= len; - if (over) { - asoc->rwnd_press += asoc->rwnd; - asoc->rwnd = 0; - } - } else { - asoc->rwnd_over = len - asoc->rwnd; - asoc->rwnd = 0; - } - - pr_debug("%s: asoc:%p rwnd decreased by %d to (%u, %u, %u)\n", - __func__, asoc, len, asoc->rwnd, asoc->rwnd_over, - asoc->rwnd_press); -} /* Build the bind address list for the association based on info from the * local endpoint and the remote peer. diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 483dcd71b3c5..591b44d3b7de 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -6176,7 +6176,7 @@ static int sctp_eat_data(const struct sctp_association *asoc, * PMTU. In cases, such as loopback, this might be a rather * large spill over. */ - if ((!chunk->data_accepted) && (!asoc->rwnd || asoc->rwnd_over || + if ((!chunk->data_accepted) && (!asoc->rwnd || (datalen > asoc->rwnd + asoc->frag_point))) { /* If this is the next TSN, consider reneging to make diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 9e91d6e5df63..7075ac847fde 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -2092,12 +2092,6 @@ static int sctp_recvmsg(struct kiocb *iocb, struct sock *sk, sctp_skb_pull(skb, copied); skb_queue_head(&sk->sk_receive_queue, skb); - /* When only partial message is copied to the user, increase - * rwnd by that amount. If all the data in the skb is read, - * rwnd is updated when the event is freed. - */ - if (!sctp_ulpevent_is_notification(event)) - sctp_assoc_rwnd_increase(event->asoc, copied); goto out; } else if ((event->msg_flags & MSG_NOTIFICATION) || (event->msg_flags & MSG_EOR)) diff --git a/net/sctp/ulpevent.c b/net/sctp/ulpevent.c index 85c64658bd0b..8d198ae03606 100644 --- a/net/sctp/ulpevent.c +++ b/net/sctp/ulpevent.c @@ -989,7 +989,7 @@ static void sctp_ulpevent_receive_data(struct sctp_ulpevent *event, skb = sctp_event2skb(event); /* Set the owner and charge rwnd for bytes received. */ sctp_ulpevent_set_owner(event, asoc); - sctp_assoc_rwnd_decrease(asoc, skb_headlen(skb)); + sctp_assoc_rwnd_update(asoc, false); if (!skb->data_len) return; @@ -1011,6 +1011,7 @@ static void sctp_ulpevent_release_data(struct sctp_ulpevent *event) { struct sk_buff *skb, *frag; unsigned int len; + struct sctp_association *asoc; /* Current stack structures assume that the rcv buffer is * per socket. For UDP style sockets this is not true as @@ -1035,8 +1036,11 @@ static void sctp_ulpevent_release_data(struct sctp_ulpevent *event) } done: - sctp_assoc_rwnd_increase(event->asoc, len); + asoc = event->asoc; + sctp_association_hold(asoc); sctp_ulpevent_release_owner(event); + sctp_assoc_rwnd_update(asoc, true); + sctp_association_put(asoc); } static void sctp_ulpevent_release_frag_data(struct sctp_ulpevent *event) -- cgit v1.2.3 From 99932d4fc03a13bb3e94938fe25458fabc8f2fc3 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Sun, 16 Feb 2014 15:55:20 +0100 Subject: netdevice: add queue selection fallback handler for ndo_select_queue Add a new argument for ndo_select_queue() callback that passes a fallback handler. This gets invoked through netdev_pick_tx(); fallback handler is currently __netdev_pick_tx() as most drivers invoke this function within their customized implementation in case for skbs that don't need any special handling. This fallback handler can then be replaced on other call-sites with different queue selection methods (e.g. in packet sockets, pktgen etc). This also has the nice side-effect that __netdev_pick_tx() is then only invoked from netdev_pick_tx() and export of that function to modules can be undone. Suggested-by: David S. Miller Signed-off-by: Daniel Borkmann Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 2 +- drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 4 ++-- drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h | 2 +- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 6 +++--- drivers/net/ethernet/lantiq_etop.c | 2 +- drivers/net/ethernet/mellanox/mlx4/en_tx.c | 4 ++-- drivers/net/ethernet/mellanox/mlx4/mlx4_en.h | 2 +- drivers/net/ethernet/tile/tilegx.c | 2 +- drivers/net/team/team.c | 2 +- drivers/net/tun.c | 2 +- drivers/net/wireless/mwifiex/main.c | 2 +- drivers/staging/bcm/Bcmnet.c | 2 +- drivers/staging/netlogic/xlr_net.c | 2 +- drivers/staging/rtl8188eu/os_dep/os_intfs.c | 2 +- include/linux/netdevice.h | 9 ++++++--- net/core/flow_dissector.c | 7 +++---- net/mac80211/iface.c | 6 ++++-- 17 files changed, 31 insertions(+), 27 deletions(-) (limited to 'include') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 867664918715..1c6104d3501d 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -3707,7 +3707,7 @@ static inline int bond_slave_override(struct bonding *bond, static u16 bond_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, select_queue_fallback_t fallback) { /* * This helper function exists to help dev_pick_tx get the correct diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index 9d7419e0390b..66c0df78c3ff 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -1873,7 +1873,7 @@ void bnx2x_netif_stop(struct bnx2x *bp, int disable_hw) } u16 bnx2x_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, select_queue_fallback_t fallback) { struct bnx2x *bp = netdev_priv(dev); @@ -1895,7 +1895,7 @@ u16 bnx2x_select_queue(struct net_device *dev, struct sk_buff *skb, } /* select a non-FCoE queue */ - return __netdev_pick_tx(dev, skb) % BNX2X_NUM_ETH_QUEUES(bp); + return fallback(dev, skb) % BNX2X_NUM_ETH_QUEUES(bp); } void bnx2x_set_num_queues(struct bnx2x *bp) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h index bfc58d488bb5..a89a40f88c25 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h @@ -496,7 +496,7 @@ int bnx2x_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos); /* select_queue callback */ u16 bnx2x_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv); + void *accel_priv, select_queue_fallback_t fallback); static inline void bnx2x_update_rx_prod(struct bnx2x *bp, struct bnx2x_fastpath *fp, diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 6d4ada72dfd0..18076c4178b4 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -6881,7 +6881,7 @@ static inline int ixgbe_maybe_stop_tx(struct ixgbe_ring *tx_ring, u16 size) } static u16 ixgbe_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, select_queue_fallback_t fallback) { struct ixgbe_fwd_adapter *fwd_adapter = accel_priv; #ifdef IXGBE_FCOE @@ -6907,7 +6907,7 @@ static u16 ixgbe_select_queue(struct net_device *dev, struct sk_buff *skb, if (adapter->flags & IXGBE_FLAG_FCOE_ENABLED) break; default: - return __netdev_pick_tx(dev, skb); + return fallback(dev, skb); } f = &adapter->ring_feature[RING_F_FCOE]; @@ -6920,7 +6920,7 @@ static u16 ixgbe_select_queue(struct net_device *dev, struct sk_buff *skb, return txq + f->offset; #else - return __netdev_pick_tx(dev, skb); + return fallback(dev, skb); #endif } diff --git a/drivers/net/ethernet/lantiq_etop.c b/drivers/net/ethernet/lantiq_etop.c index 8f9266c64c75..fd4b6aecf6ee 100644 --- a/drivers/net/ethernet/lantiq_etop.c +++ b/drivers/net/ethernet/lantiq_etop.c @@ -619,7 +619,7 @@ ltq_etop_set_multicast_list(struct net_device *dev) static u16 ltq_etop_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, select_queue_fallback_t fallback) { /* we are currently only using the first queue */ return 0; diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c index 8e8a7eb43a2c..13457032d15f 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c @@ -629,7 +629,7 @@ static void build_inline_wqe(struct mlx4_en_tx_desc *tx_desc, struct sk_buff *sk } u16 mlx4_en_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, select_queue_fallback_t fallback) { struct mlx4_en_priv *priv = netdev_priv(dev); u16 rings_p_up = priv->num_tx_rings_p_up; @@ -641,7 +641,7 @@ u16 mlx4_en_select_queue(struct net_device *dev, struct sk_buff *skb, if (vlan_tx_tag_present(skb)) up = vlan_tx_tag_get(skb) >> VLAN_PRIO_SHIFT; - return __netdev_pick_tx(dev, skb) % rings_p_up + up * rings_p_up; + return fallback(dev, skb) % rings_p_up + up * rings_p_up; } static void mlx4_bf_copy(void __iomem *dst, unsigned long *src, unsigned bytecnt) diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index 3af04c3f42ea..9ca223bc90fc 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -723,7 +723,7 @@ int mlx4_en_arm_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq); void mlx4_en_tx_irq(struct mlx4_cq *mcq); u16 mlx4_en_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv); + void *accel_priv, select_queue_fallback_t fallback); netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev); int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv, diff --git a/drivers/net/ethernet/tile/tilegx.c b/drivers/net/ethernet/tile/tilegx.c index 023237a65720..17503da9f7a5 100644 --- a/drivers/net/ethernet/tile/tilegx.c +++ b/drivers/net/ethernet/tile/tilegx.c @@ -2071,7 +2071,7 @@ static int tile_net_tx(struct sk_buff *skb, struct net_device *dev) /* Return subqueue id on this core (one per core). */ static u16 tile_net_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, select_queue_fallback_t fallback) { return smp_processor_id(); } diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index 28407426fd6f..c8624a8235ab 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -1648,7 +1648,7 @@ static netdev_tx_t team_xmit(struct sk_buff *skb, struct net_device *dev) } static u16 team_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, select_queue_fallback_t fallback) { /* * This helper function exists to help dev_pick_tx get the correct diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 44c4db8450f0..8fe9cb7d0f72 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -366,7 +366,7 @@ static inline void tun_flow_save_rps_rxhash(struct tun_flow_entry *e, u32 hash) * hope the rxq no. may help here. */ static u16 tun_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, select_queue_fallback_t fallback) { struct tun_struct *tun = netdev_priv(dev); struct tun_flow_entry *e; diff --git a/drivers/net/wireless/mwifiex/main.c b/drivers/net/wireless/mwifiex/main.c index 4d79761b9c87..9d3d2758ec35 100644 --- a/drivers/net/wireless/mwifiex/main.c +++ b/drivers/net/wireless/mwifiex/main.c @@ -748,7 +748,7 @@ static struct net_device_stats *mwifiex_get_stats(struct net_device *dev) static u16 mwifiex_netdev_select_wmm_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, select_queue_fallback_t fallback) { skb->priority = cfg80211_classify8021d(skb, NULL); return mwifiex_1d_to_wmm_queue[skb->priority]; diff --git a/drivers/staging/bcm/Bcmnet.c b/drivers/staging/bcm/Bcmnet.c index 8dfdd2732bdc..95a2358267ba 100644 --- a/drivers/staging/bcm/Bcmnet.c +++ b/drivers/staging/bcm/Bcmnet.c @@ -40,7 +40,7 @@ static INT bcm_close(struct net_device *dev) } static u16 bcm_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, select_queue_fallback_t fallback) { return ClassifyPacket(netdev_priv(dev), skb); } diff --git a/drivers/staging/netlogic/xlr_net.c b/drivers/staging/netlogic/xlr_net.c index eedffed17e39..6f9ac27730af 100644 --- a/drivers/staging/netlogic/xlr_net.c +++ b/drivers/staging/netlogic/xlr_net.c @@ -307,7 +307,7 @@ static netdev_tx_t xlr_net_start_xmit(struct sk_buff *skb, } static u16 xlr_net_select_queue(struct net_device *ndev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, select_queue_fallback_t fallback) { return (u16)smp_processor_id(); } diff --git a/drivers/staging/rtl8188eu/os_dep/os_intfs.c b/drivers/staging/rtl8188eu/os_dep/os_intfs.c index 68f98fa114d2..7c9ee58f47bb 100644 --- a/drivers/staging/rtl8188eu/os_dep/os_intfs.c +++ b/drivers/staging/rtl8188eu/os_dep/os_intfs.c @@ -653,7 +653,7 @@ static unsigned int rtw_classify8021d(struct sk_buff *skb) } static u16 rtw_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, select_queue_fallback_t fallback) { struct adapter *padapter = rtw_netdev_priv(dev); struct mlme_priv *pmlmepriv = &padapter->mlmepriv; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 21d4e6be8949..1de9c136b066 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -752,6 +752,9 @@ struct netdev_phys_port_id { unsigned char id_len; }; +typedef u16 (*select_queue_fallback_t)(struct net_device *dev, + struct sk_buff *skb); + /* * This structure defines the management hooks for network devices. * The following hooks can be defined; unless noted otherwise, they are @@ -783,7 +786,7 @@ struct netdev_phys_port_id { * Required can not be NULL. * * u16 (*ndo_select_queue)(struct net_device *dev, struct sk_buff *skb, - * void *accel_priv); + * void *accel_priv, select_queue_fallback_t fallback); * Called to decide which queue to when device supports multiple * transmit queues. * @@ -1005,7 +1008,8 @@ struct net_device_ops { struct net_device *dev); u16 (*ndo_select_queue)(struct net_device *dev, struct sk_buff *skb, - void *accel_priv); + void *accel_priv, + select_queue_fallback_t fallback); void (*ndo_change_rx_flags)(struct net_device *dev, int flags); void (*ndo_set_rx_mode)(struct net_device *dev); @@ -1551,7 +1555,6 @@ static inline void netdev_for_each_tx_queue(struct net_device *dev, struct netdev_queue *netdev_pick_tx(struct net_device *dev, struct sk_buff *skb, void *accel_priv); -u16 __netdev_pick_tx(struct net_device *dev, struct sk_buff *skb); /* * Net namespace inlines diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c index 87577d447554..75fe83f590ea 100644 --- a/net/core/flow_dissector.c +++ b/net/core/flow_dissector.c @@ -372,7 +372,7 @@ static inline int get_xps_queue(struct net_device *dev, struct sk_buff *skb) #endif } -u16 __netdev_pick_tx(struct net_device *dev, struct sk_buff *skb) +static u16 __netdev_pick_tx(struct net_device *dev, struct sk_buff *skb) { struct sock *sk = skb->sk; int queue_index = sk_tx_queue_get(sk); @@ -392,7 +392,6 @@ u16 __netdev_pick_tx(struct net_device *dev, struct sk_buff *skb) return queue_index; } -EXPORT_SYMBOL(__netdev_pick_tx); struct netdev_queue *netdev_pick_tx(struct net_device *dev, struct sk_buff *skb, @@ -403,8 +402,8 @@ struct netdev_queue *netdev_pick_tx(struct net_device *dev, if (dev->real_num_tx_queues != 1) { const struct net_device_ops *ops = dev->netdev_ops; if (ops->ndo_select_queue) - queue_index = ops->ndo_select_queue(dev, skb, - accel_priv); + queue_index = ops->ndo_select_queue(dev, skb, accel_priv, + __netdev_pick_tx); else queue_index = __netdev_pick_tx(dev, skb); diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c index d6d1f1df9119..ce1c44370610 100644 --- a/net/mac80211/iface.c +++ b/net/mac80211/iface.c @@ -1057,7 +1057,8 @@ static void ieee80211_uninit(struct net_device *dev) static u16 ieee80211_netdev_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, + select_queue_fallback_t fallback) { return ieee80211_select_queue(IEEE80211_DEV_TO_SUB_IF(dev), skb); } @@ -1075,7 +1076,8 @@ static const struct net_device_ops ieee80211_dataif_ops = { static u16 ieee80211_monitor_select_queue(struct net_device *dev, struct sk_buff *skb, - void *accel_priv) + void *accel_priv, + select_queue_fallback_t fallback) { struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); struct ieee80211_local *local = sdata->local; -- cgit v1.2.3 From b9507bdaf40e91fea2b1c0c1ee7dc627c8ee6fd6 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Sun, 16 Feb 2014 15:55:21 +0100 Subject: netdevice: move netdev_cap_txqueue for shared usage to header In order to allow users to invoke netdev_cap_txqueue, it needs to be moved into netdevice.h header file. While at it, also add kernel doc header to document the API. Signed-off-by: Daniel Borkmann Signed-off-by: David S. Miller --- include/linux/netdevice.h | 20 ++++++++++++++++++++ net/core/flow_dissector.c | 13 +------------ 2 files changed, 21 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 1de9c136b066..e8eeebd49a98 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2278,6 +2278,26 @@ static inline void netdev_reset_queue(struct net_device *dev_queue) netdev_tx_reset_queue(netdev_get_tx_queue(dev_queue, 0)); } +/** + * netdev_cap_txqueue - check if selected tx queue exceeds device queues + * @dev: network device + * @queue_index: given tx queue index + * + * Returns 0 if given tx queue index >= number of device tx queues, + * otherwise returns the originally passed tx queue index. + */ +static inline u16 netdev_cap_txqueue(struct net_device *dev, u16 queue_index) +{ + if (unlikely(queue_index >= dev->real_num_tx_queues)) { + net_warn_ratelimited("%s selects TX queue %d, but real number of TX queues is %d\n", + dev->name, queue_index, + dev->real_num_tx_queues); + return 0; + } + + return queue_index; +} + /** * netif_running - test if up * @dev: network device diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c index 75fe83f590ea..e29e810663d7 100644 --- a/net/core/flow_dissector.c +++ b/net/core/flow_dissector.c @@ -323,17 +323,6 @@ u32 __skb_get_poff(const struct sk_buff *skb) return poff; } -static inline u16 dev_cap_txqueue(struct net_device *dev, u16 queue_index) -{ - if (unlikely(queue_index >= dev->real_num_tx_queues)) { - net_warn_ratelimited("%s selects TX queue %d, but real number of TX queues is %d\n", - dev->name, queue_index, - dev->real_num_tx_queues); - return 0; - } - return queue_index; -} - static inline int get_xps_queue(struct net_device *dev, struct sk_buff *skb) { #ifdef CONFIG_XPS @@ -408,7 +397,7 @@ struct netdev_queue *netdev_pick_tx(struct net_device *dev, queue_index = __netdev_pick_tx(dev, skb); if (!accel_priv) - queue_index = dev_cap_txqueue(dev, queue_index); + queue_index = netdev_cap_txqueue(dev, queue_index); } skb_set_queue_mapping(skb, queue_index); -- cgit v1.2.3 From bcdfeb2eb4e42b811950b9cd226109291051732a Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Tue, 11 Feb 2014 13:04:19 +0800 Subject: ceph: remove xattr when null value is given to setxattr() For the setxattr request, introduce a new flag CEPH_XATTR_REMOVE to distinguish null value case from the zero-length value case. Signed-off-by: Yan, Zheng --- fs/ceph/xattr.c | 16 ++++++++++++++-- include/linux/ceph/ceph_fs.h | 5 +++-- 2 files changed, 17 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c index 28f9793b9167..231c02b16c0c 100644 --- a/fs/ceph/xattr.c +++ b/fs/ceph/xattr.c @@ -12,6 +12,9 @@ #define XATTR_CEPH_PREFIX "ceph." #define XATTR_CEPH_PREFIX_LEN (sizeof (XATTR_CEPH_PREFIX) - 1) +static int __remove_xattr(struct ceph_inode_info *ci, + struct ceph_inode_xattr *xattr); + /* * List of handlers for synthetic system.* attributes. Other * attributes are handled directly. @@ -359,6 +362,12 @@ static int __set_xattr(struct ceph_inode_info *ci, kfree(val); return err; } + if (update_xattr < 0) { + if (xattr) + __remove_xattr(ci, xattr); + kfree(name); + return 0; + } } if (!xattr) { @@ -862,6 +871,9 @@ static int ceph_sync_setxattr(struct dentry *dentry, const char *name, dout("setxattr value=%.*s\n", (int)size, value); + if (!value) + flags |= CEPH_XATTR_REMOVE; + /* do request */ req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_SETXATTR, USE_AUTH_MDS); @@ -965,8 +977,8 @@ retry: goto retry; } - err = __set_xattr(ci, newname, name_len, newval, - val_len, flags, 1, &xattr); + err = __set_xattr(ci, newname, name_len, newval, val_len, + flags, value ? 1 : -1, &xattr); if (!err) { dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL); diff --git a/include/linux/ceph/ceph_fs.h b/include/linux/ceph/ceph_fs.h index 2623cffc73a1..25bfb0eff772 100644 --- a/include/linux/ceph/ceph_fs.h +++ b/include/linux/ceph/ceph_fs.h @@ -373,8 +373,9 @@ extern const char *ceph_mds_op_name(int op); /* * Ceph setxattr request flags. */ -#define CEPH_XATTR_CREATE 1 -#define CEPH_XATTR_REPLACE 2 +#define CEPH_XATTR_CREATE (1 << 0) +#define CEPH_XATTR_REPLACE (1 << 1) +#define CEPH_XATTR_REMOVE (1 << 31) union ceph_mds_request_args { struct { -- cgit v1.2.3 From 728a0cdf065b9f42d1ff25d94858f824bb0989ba Mon Sep 17 00:00:00 2001 From: Alexandre Courbot Date: Sun, 9 Feb 2014 18:43:18 +0900 Subject: drm/ttm: declare 'struct device' in ttm_page_alloc.h Declare 'struct device' explicitly in ttm_page_alloc.h as this file does not include any file declaring it. This removes the following warning: warning: 'struct device' declared inside parameter list Signed-off-by: Alexandre Courbot Reviewed-by: Thierry Reding --- include/drm/ttm/ttm_page_alloc.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/drm/ttm/ttm_page_alloc.h b/include/drm/ttm/ttm_page_alloc.h index d1f61bfe0ebe..49a828425fa2 100644 --- a/include/drm/ttm/ttm_page_alloc.h +++ b/include/drm/ttm/ttm_page_alloc.h @@ -29,6 +29,8 @@ #include #include +struct device; + /** * Initialize pool allocator. */ -- cgit v1.2.3 From 8716ed4e7bed4e4c7e3f37940e950ddc0362f450 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 12 Feb 2014 12:48:23 -0500 Subject: drm: add DRM_CAPs for cursor size Some hardware may not support standard 64x64 cursors. Add a drm cap to query the cursor size from the kernel. Some examples include radeon CIK parts (128x128 cursors) and armada (32x64 or 64x32). This allows things like device specific ddxes to remove asics specific logic and also allows xf86-video-modesetting to work properly with hw cursors on this hardware. Default to 64 if the driver doesn't specify a size. Signed-off-by: Alex Deucher Reviewed-by: Rob Clark --- drivers/gpu/drm/drm_ioctl.c | 12 ++++++++++++ include/drm/drm_crtc.h | 3 +++ include/uapi/drm/drm.h | 2 ++ 3 files changed, 17 insertions(+) (limited to 'include') diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c index dffc836144cc..f4dc9b7a3831 100644 --- a/drivers/gpu/drm/drm_ioctl.c +++ b/drivers/gpu/drm/drm_ioctl.c @@ -296,6 +296,18 @@ int drm_getcap(struct drm_device *dev, void *data, struct drm_file *file_priv) case DRM_CAP_ASYNC_PAGE_FLIP: req->value = dev->mode_config.async_page_flip; break; + case DRM_CAP_CURSOR_WIDTH: + if (dev->mode_config.cursor_width) + req->value = dev->mode_config.cursor_width; + else + req->value = 64; + break; + case DRM_CAP_CURSOR_HEIGHT: + if (dev->mode_config.cursor_height) + req->value = dev->mode_config.cursor_height; + else + req->value = 64; + break; default: return -EINVAL; } diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h index 71727b6210ae..8f3dee097579 100644 --- a/include/drm/drm_crtc.h +++ b/include/drm/drm_crtc.h @@ -907,6 +907,9 @@ struct drm_mode_config { /* whether async page flip is supported or not */ bool async_page_flip; + + /* cursor size */ + uint32_t cursor_width, cursor_height; }; #define obj_to_crtc(x) container_of(x, struct drm_crtc, base) diff --git a/include/uapi/drm/drm.h b/include/uapi/drm/drm.h index 3c9a833992e8..b06c8ed68707 100644 --- a/include/uapi/drm/drm.h +++ b/include/uapi/drm/drm.h @@ -619,6 +619,8 @@ struct drm_gem_open { #define DRM_PRIME_CAP_EXPORT 0x2 #define DRM_CAP_TIMESTAMP_MONOTONIC 0x6 #define DRM_CAP_ASYNC_PAGE_FLIP 0x7 +#define DRM_CAP_CURSOR_WIDTH 0x8 +#define DRM_CAP_CURSOR_HEIGHT 0x9 /** DRM_IOCTL_GET_CAP ioctl argument type */ struct drm_get_cap { -- cgit v1.2.3