This documentation 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 For more details see the file COPYING in the source distribution of Linux. Kernel Hackers Manual July 2017 enum sock_type 9 4.1.27 enum sock_type Socket types enum sock_type { SOCK_STREAM, SOCK_DGRAM, SOCK_RAW, SOCK_RDM, SOCK_SEQPACKET, SOCK_DCCP, SOCK_PACKET }; SOCK_STREAM stream (connection) socket SOCK_DGRAM datagram (conn.less) socket SOCK_RAW raw socket SOCK_RDM reliably-delivered message SOCK_SEQPACKET sequential packet socket SOCK_DCCP Datagram Congestion Control Protocol socket SOCK_PACKET linux specific way of getting packets at the dev level. For writing rarp and other similar things on the user level. When adding some new socket type please grep ARCH_HAS_SOCKET_TYPE include/asm-* /socket.h, at least MIPS overrides this enum for binary compat reasons. Kernel Hackers Manual July 2017 struct socket 9 4.1.27 struct socket general BSD socket struct socket { socket_state state; short type; unsigned long flags; struct socket_wq __rcu * wq; struct file * file; struct sock * sk; const struct proto_ops * ops; }; state socket state (SS_CONNECTED, etc) type socket type (SOCK_STREAM, etc) flags socket flags (SOCK_ASYNC_NOSPACE, etc) wq wait queue for several uses file File back pointer for gc sk internal networking protocol agnostic socket representation ops protocol specific socket operations Kernel Hackers Manual July 2017 struct skb_shared_hwtstamps 9 4.1.27 struct skb_shared_hwtstamps hardware time stamps struct skb_shared_hwtstamps { ktime_t hwtstamp; }; hwtstamp hardware time stamp transformed into duration since arbitrary point in time Software time stamps generated by ktime_get_real are stored in skb->tstamp. hwtstamps can only be compared against other hwtstamps from the same device. This structure is attached to packets as part of the skb_shared_info. Use skb_hwtstamps to get a pointer. Kernel Hackers Manual July 2017 struct skb_mstamp 9 4.1.27 struct skb_mstamp multi resolution time stamps struct skb_mstamp { union {unnamed_union}; }; {unnamed_union} anonymous Kernel Hackers Manual July 2017 skb_mstamp_get 9 4.1.27 skb_mstamp_get get current timestamp void skb_mstamp_get struct skb_mstamp * cl cl place to store timestamps Kernel Hackers Manual July 2017 skb_mstamp_us_delta 9 4.1.27 skb_mstamp_us_delta compute the difference in usec between two skb_mstamp u32 skb_mstamp_us_delta const struct skb_mstamp * t1 const struct skb_mstamp * t0 t1 pointer to newest sample t0 pointer to oldest sample Kernel Hackers Manual July 2017 struct sk_buff 9 4.1.27 struct sk_buff socket buffer struct sk_buff { union {unnamed_union}; __u16 inner_transport_header; __u16 inner_network_header; __u16 inner_mac_header; __be16 protocol; __u16 transport_header; __u16 network_header; __u16 mac_header; sk_buff_data_t tail; sk_buff_data_t end; unsigned char * head; unsigned char * data; unsigned int truesize; atomic_t users; }; {unnamed_union} anonymous inner_transport_header Inner transport layer header (encapsulation) inner_network_header Network layer header (encapsulation) inner_mac_header Link layer header (encapsulation) protocol Packet protocol from driver transport_header Transport layer header network_header Network layer header mac_header Link layer header tail Tail pointer end End pointer head Head of buffer data Data head pointer truesize Buffer size users User count - see {datagram,tcp}.c Kernel Hackers Manual July 2017 skb_dst 9 4.1.27 skb_dst returns skb dst_entry struct dst_entry * skb_dst const struct sk_buff * skb skb buffer Returns skb dst_entry, regardless of reference taken or not. Kernel Hackers Manual July 2017 skb_dst_set 9 4.1.27 skb_dst_set sets skb dst void skb_dst_set struct sk_buff * skb struct dst_entry * dst skb buffer dst dst entry Sets skb dst, assuming a reference was taken on dst and should be released by skb_dst_drop Kernel Hackers Manual July 2017 skb_dst_set_noref 9 4.1.27 skb_dst_set_noref sets skb dst, hopefully, without taking reference void skb_dst_set_noref struct sk_buff * skb struct dst_entry * dst skb buffer dst dst entry Sets skb dst, assuming a reference was not taken on dst. If dst entry is cached, we do not take reference and dst_release will be avoided by refdst_drop. If dst entry is not cached, we take reference, so that last dst_release can destroy the dst immediately. Kernel Hackers Manual July 2017 skb_dst_is_noref 9 4.1.27 skb_dst_is_noref Test if skb dst isn't refcounted bool skb_dst_is_noref const struct sk_buff * skb skb buffer Kernel Hackers Manual July 2017 skb_fclone_busy 9 4.1.27 skb_fclone_busy check if fclone is busy bool skb_fclone_busy const struct sock * sk const struct sk_buff * skb sk -- undescribed -- skb buffer Returns true is skb is a fast clone, and its clone is not freed. Some drivers call skb_orphan in their ndo_start_xmit, so we also check that this didnt happen. Kernel Hackers Manual July 2017 skb_queue_empty 9 4.1.27 skb_queue_empty check if a queue is empty int skb_queue_empty const struct sk_buff_head * list list queue head Returns true if the queue is empty, false otherwise. Kernel Hackers Manual July 2017 skb_queue_is_last 9 4.1.27 skb_queue_is_last check if skb is the last entry in the queue bool skb_queue_is_last const struct sk_buff_head * list const struct sk_buff * skb list queue head skb buffer Returns true if skb is the last buffer on the list. Kernel Hackers Manual July 2017 skb_queue_is_first 9 4.1.27 skb_queue_is_first check if skb is the first entry in the queue bool skb_queue_is_first const struct sk_buff_head * list const struct sk_buff * skb list queue head skb buffer Returns true if skb is the first buffer on the list. Kernel Hackers Manual July 2017 skb_queue_next 9 4.1.27 skb_queue_next return the next packet in the queue struct sk_buff * skb_queue_next const struct sk_buff_head * list const struct sk_buff * skb list queue head skb current buffer Return the next packet in list after skb. It is only valid to call this if skb_queue_is_last evaluates to false. Kernel Hackers Manual July 2017 skb_queue_prev 9 4.1.27 skb_queue_prev return the prev packet in the queue struct sk_buff * skb_queue_prev const struct sk_buff_head * list const struct sk_buff * skb list queue head skb current buffer Return the prev packet in list before skb. It is only valid to call this if skb_queue_is_first evaluates to false. Kernel Hackers Manual July 2017 skb_get 9 4.1.27 skb_get reference buffer struct sk_buff * skb_get struct sk_buff * skb skb buffer to reference Makes another reference to a socket buffer and returns a pointer to the buffer. Kernel Hackers Manual July 2017 skb_cloned 9 4.1.27 skb_cloned is the buffer a clone int skb_cloned const struct sk_buff * skb skb buffer to check Returns true if the buffer was generated with skb_clone and is one of multiple shared copies of the buffer. Cloned buffers are shared data so must not be written to under normal circumstances. Kernel Hackers Manual July 2017 skb_header_cloned 9 4.1.27 skb_header_cloned is the header a clone int skb_header_cloned const struct sk_buff * skb skb buffer to check Returns true if modifying the header part of the buffer requires the data to be copied. Kernel Hackers Manual July 2017 skb_header_release 9 4.1.27 skb_header_release release reference to header void skb_header_release struct sk_buff * skb skb buffer to operate on Drop a reference to the header part of the buffer. This is done by acquiring a payload reference. You must not read from the header part of skb->data after this. Check if you can use __skb_header_release instead. Kernel Hackers Manual July 2017 __skb_header_release 9 4.1.27 __skb_header_release release reference to header void __skb_header_release struct sk_buff * skb skb buffer to operate on Variant of skb_header_release assuming skb is private to caller. We can avoid one atomic operation. Kernel Hackers Manual July 2017 skb_shared 9 4.1.27 skb_shared is the buffer shared int skb_shared const struct sk_buff * skb skb buffer to check Returns true if more than one person has a reference to this buffer. Kernel Hackers Manual July 2017 skb_share_check 9 4.1.27 skb_share_check check if buffer is shared and if so clone it struct sk_buff * skb_share_check struct sk_buff * skb gfp_t pri skb buffer to check pri priority for memory allocation If the buffer is shared the buffer is cloned and the old copy drops a reference. A new clone with a single reference is returned. If the buffer is not shared the original buffer is returned. When being called from interrupt status or with spinlocks held pri must be GFP_ATOMIC. NULL is returned on a memory allocation failure. Kernel Hackers Manual July 2017 skb_unshare 9 4.1.27 skb_unshare make a copy of a shared buffer struct sk_buff * skb_unshare struct sk_buff * skb gfp_t pri skb buffer to check pri priority for memory allocation If the socket buffer is a clone then this function creates a new copy of the data, drops a reference count on the old copy and returns the new copy with the reference count at 1. If the buffer is not a clone the original buffer is returned. When called with a spinlock held or from interrupt state pri must be GFP_ATOMIC NULL is returned on a memory allocation failure. Kernel Hackers Manual July 2017 skb_peek 9 4.1.27 skb_peek peek at the head of an sk_buff_head struct sk_buff * skb_peek const struct sk_buff_head * list_ list_ list to peek at Peek an sk_buff. Unlike most other operations you _MUST_ be careful with this one. A peek leaves the buffer on the list and someone else may run off with it. You must hold the appropriate locks or have a private queue to do this. Returns NULL for an empty list or a pointer to the head element. The reference count is not incremented and the reference is therefore volatile. Use with caution. Kernel Hackers Manual July 2017 skb_peek_next 9 4.1.27 skb_peek_next peek skb following the given one from a queue struct sk_buff * skb_peek_next struct sk_buff * skb const struct sk_buff_head * list_ skb skb to start from list_ list to peek at Returns NULL when the end of the list is met or a pointer to the next element. The reference count is not incremented and the reference is therefore volatile. Use with caution. Kernel Hackers Manual July 2017 skb_peek_tail 9 4.1.27 skb_peek_tail peek at the tail of an sk_buff_head struct sk_buff * skb_peek_tail const struct sk_buff_head * list_ list_ list to peek at Peek an sk_buff. Unlike most other operations you _MUST_ be careful with this one. A peek leaves the buffer on the list and someone else may run off with it. You must hold the appropriate locks or have a private queue to do this. Returns NULL for an empty list or a pointer to the tail element. The reference count is not incremented and the reference is therefore volatile. Use with caution. Kernel Hackers Manual July 2017 skb_queue_len 9 4.1.27 skb_queue_len get queue length __u32 skb_queue_len const struct sk_buff_head * list_ list_ list to measure Return the length of an sk_buff queue. Kernel Hackers Manual July 2017 __skb_queue_head_init 9 4.1.27 __skb_queue_head_init initialize non-spinlock portions of sk_buff_head void __skb_queue_head_init struct sk_buff_head * list list queue to initialize This initializes only the list and queue length aspects of an sk_buff_head object. This allows to initialize the list aspects of an sk_buff_head without reinitializing things like the spinlock. It can also be used for on-stack sk_buff_head objects where the spinlock is known to not be used. Kernel Hackers Manual July 2017 skb_queue_splice 9 4.1.27 skb_queue_splice join two skb lists, this is designed for stacks void skb_queue_splice const struct sk_buff_head * list struct sk_buff_head * head list the new list to add head the place to add it in the first list Kernel Hackers Manual July 2017 skb_queue_splice_init 9 4.1.27 skb_queue_splice_init join two skb lists and reinitialise the emptied list void skb_queue_splice_init struct sk_buff_head * list struct sk_buff_head * head list the new list to add head the place to add it in the first list The list at list is reinitialised Kernel Hackers Manual July 2017 skb_queue_splice_tail 9 4.1.27 skb_queue_splice_tail join two skb lists, each list being a queue void skb_queue_splice_tail const struct sk_buff_head * list struct sk_buff_head * head list the new list to add head the place to add it in the first list Kernel Hackers Manual July 2017 skb_queue_splice_tail_init 9 4.1.27 skb_queue_splice_tail_init join two skb lists and reinitialise the emptied list void skb_queue_splice_tail_init struct sk_buff_head * list struct sk_buff_head * head list the new list to add head the place to add it in the first list Each of the lists is a queue. The list at list is reinitialised Kernel Hackers Manual July 2017 __skb_queue_after 9 4.1.27 __skb_queue_after queue a buffer at the list head void __skb_queue_after struct sk_buff_head * list struct sk_buff * prev struct sk_buff * newsk list list to use prev place after this buffer newsk buffer to queue Queue a buffer int the middle of a list. This function takes no locks and you must therefore hold required locks before calling it. A buffer cannot be placed on two lists at the same time. Kernel Hackers Manual July 2017 __skb_fill_page_desc 9 4.1.27 __skb_fill_page_desc initialise a paged fragment in an skb void __skb_fill_page_desc struct sk_buff * skb int i struct page * page int off int size skb buffer containing fragment to be initialised i paged fragment index to initialise page the page to use for this fragment off the offset to the data with page size the length of the data Initialises the i'th fragment of skb to point to size bytes at offset off within page. Does not take any additional reference on the fragment. Kernel Hackers Manual July 2017 skb_fill_page_desc 9 4.1.27 skb_fill_page_desc initialise a paged fragment in an skb void skb_fill_page_desc struct sk_buff * skb int i struct page * page int off int size skb buffer containing fragment to be initialised i paged fragment index to initialise page the page to use for this fragment off the offset to the data with page size the length of the data As per __skb_fill_page_desc -- initialises the i'th fragment of skb to point to size bytes at offset off within page. In addition updates skb such that i is the last fragment. Does not take any additional reference on the fragment. Kernel Hackers Manual July 2017 skb_headroom 9 4.1.27 skb_headroom bytes at buffer head unsigned int skb_headroom const struct sk_buff * skb skb buffer to check Return the number of bytes of free space at the head of an sk_buff. Kernel Hackers Manual July 2017 skb_tailroom 9 4.1.27 skb_tailroom bytes at buffer end int skb_tailroom const struct sk_buff * skb skb buffer to check Return the number of bytes of free space at the tail of an sk_buff Kernel Hackers Manual July 2017 skb_availroom 9 4.1.27 skb_availroom bytes at buffer end int skb_availroom const struct sk_buff * skb skb buffer to check Return the number of bytes of free space at the tail of an sk_buff allocated by sk_stream_alloc Kernel Hackers Manual July 2017 skb_reserve 9 4.1.27 skb_reserve adjust headroom void skb_reserve struct sk_buff * skb int len skb buffer to alter len bytes to move Increase the headroom of an empty sk_buff by reducing the tail room. This is only allowed for an empty buffer. Kernel Hackers Manual July 2017 pskb_trim_unique 9 4.1.27 pskb_trim_unique remove end from a paged unique (not cloned) buffer void pskb_trim_unique struct sk_buff * skb unsigned int len skb buffer to alter len new length This is identical to pskb_trim except that the caller knows that the skb is not cloned so we should never get an error due to out- of-memory. Kernel Hackers Manual July 2017 skb_orphan 9 4.1.27 skb_orphan orphan a buffer void skb_orphan struct sk_buff * skb skb buffer to orphan If a buffer currently has an owner then we call the owner's destructor function and make the skb unowned. The buffer continues to exist but is no longer charged to its former owner. Kernel Hackers Manual July 2017 skb_orphan_frags 9 4.1.27 skb_orphan_frags orphan the frags contained in a buffer int skb_orphan_frags struct sk_buff * skb gfp_t gfp_mask skb buffer to orphan frags from gfp_mask allocation mask for replacement pages For each frag in the SKB which needs a destructor (i.e. has an owner) create a copy of that frag and release the original page by calling the destructor. Kernel Hackers Manual July 2017 netdev_alloc_skb 9 4.1.27 netdev_alloc_skb allocate an skbuff for rx on a specific device struct sk_buff * netdev_alloc_skb struct net_device * dev unsigned int length dev network device to receive on length length to allocate Allocate a new sk_buff and assign it a usage count of one. The buffer has unspecified headroom built in. Users should allocate the headroom they think they need without accounting for the built in space. The built in space is used for optimisations. NULL is returned if there is no free memory. Although this function allocates memory it can be called from an interrupt. Kernel Hackers Manual July 2017 __dev_alloc_pages 9 4.1.27 __dev_alloc_pages allocate page for network Rx struct page * __dev_alloc_pages gfp_t gfp_mask unsigned int order gfp_mask allocation priority. Set __GFP_NOMEMALLOC if not for network Rx order size of the allocation Allocate a new page. NULL is returned if there is no free memory. Kernel Hackers Manual July 2017 __dev_alloc_page 9 4.1.27 __dev_alloc_page allocate a page for network Rx struct page * __dev_alloc_page gfp_t gfp_mask gfp_mask allocation priority. Set __GFP_NOMEMALLOC if not for network Rx Allocate a new page. NULL is returned if there is no free memory. Kernel Hackers Manual July 2017 skb_propagate_pfmemalloc 9 4.1.27 skb_propagate_pfmemalloc Propagate pfmemalloc if skb is allocated after RX page void skb_propagate_pfmemalloc struct page * page struct sk_buff * skb page The page that was allocated from skb_alloc_page skb The skb that may need pfmemalloc set Kernel Hackers Manual July 2017 skb_frag_page 9 4.1.27 skb_frag_page retrieve the page referred to by a paged fragment struct page * skb_frag_page const skb_frag_t * frag frag the paged fragment Returns the struct page associated with frag. Kernel Hackers Manual July 2017 __skb_frag_ref 9 4.1.27 __skb_frag_ref take an addition reference on a paged fragment. void __skb_frag_ref skb_frag_t * frag frag the paged fragment Takes an additional reference on the paged fragment frag. Kernel Hackers Manual July 2017 skb_frag_ref 9 4.1.27 skb_frag_ref take an addition reference on a paged fragment of an skb. void skb_frag_ref struct sk_buff * skb int f skb the buffer f the fragment offset. Takes an additional reference on the f'th paged fragment of skb. Kernel Hackers Manual July 2017 __skb_frag_unref 9 4.1.27 __skb_frag_unref release a reference on a paged fragment. void __skb_frag_unref skb_frag_t * frag frag the paged fragment Releases a reference on the paged fragment frag. Kernel Hackers Manual July 2017 skb_frag_unref 9 4.1.27 skb_frag_unref release a reference on a paged fragment of an skb. void skb_frag_unref struct sk_buff * skb int f skb the buffer f the fragment offset Releases a reference on the f'th paged fragment of skb. Kernel Hackers Manual July 2017 skb_frag_address 9 4.1.27 skb_frag_address gets the address of the data contained in a paged fragment void * skb_frag_address const skb_frag_t * frag frag the paged fragment buffer Returns the address of the data within frag. The page must already be mapped. Kernel Hackers Manual July 2017 skb_frag_address_safe 9 4.1.27 skb_frag_address_safe gets the address of the data contained in a paged fragment void * skb_frag_address_safe const skb_frag_t * frag frag the paged fragment buffer Returns the address of the data within frag. Checks that the page is mapped and returns NULL otherwise. Kernel Hackers Manual July 2017 __skb_frag_set_page 9 4.1.27 __skb_frag_set_page sets the page contained in a paged fragment void __skb_frag_set_page skb_frag_t * frag struct page * page frag the paged fragment page the page to set Sets the fragment frag to contain page. Kernel Hackers Manual July 2017 skb_frag_set_page 9 4.1.27 skb_frag_set_page sets the page contained in a paged fragment of an skb void skb_frag_set_page struct sk_buff * skb int f struct page * page skb the buffer f the fragment offset page the page to set Sets the f'th fragment of skb to contain page. Kernel Hackers Manual July 2017 skb_frag_dma_map 9 4.1.27 skb_frag_dma_map maps a paged fragment via the DMA API dma_addr_t skb_frag_dma_map struct device * dev const skb_frag_t * frag size_t offset size_t size enum dma_data_direction dir dev the device to map the fragment to frag the paged fragment to map offset the offset within the fragment (starting at the fragment's own offset) size the number of bytes to map dir the direction of the mapping (PCI_DMA_*) Maps the page associated with frag to device. Kernel Hackers Manual July 2017 skb_clone_writable 9 4.1.27 skb_clone_writable is the header of a clone writable int skb_clone_writable const struct sk_buff * skb unsigned int len skb buffer to check len length up to which to write Returns true if modifying the header part of the cloned buffer does not requires the data to be copied. Kernel Hackers Manual July 2017 skb_cow 9 4.1.27 skb_cow copy header of skb when it is required int skb_cow struct sk_buff * skb unsigned int headroom skb buffer to cow headroom needed headroom If the skb passed lacks sufficient headroom or its data part is shared, data is reallocated. If reallocation fails, an error is returned and original skb is not changed. The result is skb with writable area skb->head...skb->tail and at least headroom of space at head. Kernel Hackers Manual July 2017 skb_cow_head 9 4.1.27 skb_cow_head skb_cow but only making the head writable int skb_cow_head struct sk_buff * skb unsigned int headroom skb buffer to cow headroom needed headroom This function is identical to skb_cow except that we replace the skb_cloned check by skb_header_cloned. It should be used when you only need to push on some header and do not need to modify the data. Kernel Hackers Manual July 2017 skb_padto 9 4.1.27 skb_padto pad an skbuff up to a minimal size int skb_padto struct sk_buff * skb unsigned int len skb buffer to pad len minimal length Pads up a buffer to ensure the trailing bytes exist and are blanked. If the buffer already contains sufficient data it is untouched. Otherwise it is extended. Returns zero on success. The skb is freed on error. Kernel Hackers Manual July 2017 skb_put_padto 9 4.1.27 skb_put_padto increase size and pad an skbuff up to a minimal size int skb_put_padto struct sk_buff * skb unsigned int len skb buffer to pad len minimal length Pads up a buffer to ensure the trailing bytes exist and are blanked. If the buffer already contains sufficient data it is untouched. Otherwise it is extended. Returns zero on success. The skb is freed on error. Kernel Hackers Manual July 2017 skb_linearize 9 4.1.27 skb_linearize convert paged skb to linear one int skb_linearize struct sk_buff * skb skb buffer to linarize If there is no free memory -ENOMEM is returned, otherwise zero is returned and the old skb data released. Kernel Hackers Manual July 2017 skb_has_shared_frag 9 4.1.27 skb_has_shared_frag can any frag be overwritten bool skb_has_shared_frag const struct sk_buff * skb skb buffer to test Return true if the skb has at least one frag that might be modified by an external entity (as in vmsplice/sendfile) Kernel Hackers Manual July 2017 skb_linearize_cow 9 4.1.27 skb_linearize_cow make sure skb is linear and writable int skb_linearize_cow struct sk_buff * skb skb buffer to process If there is no free memory -ENOMEM is returned, otherwise zero is returned and the old skb data released. Kernel Hackers Manual July 2017 skb_postpull_rcsum 9 4.1.27 skb_postpull_rcsum update checksum for received skb after pull void skb_postpull_rcsum struct sk_buff * skb const void * start unsigned int len skb buffer to update start start of data before pull len length of data pulled After doing a pull on a received packet, you need to call this to update the CHECKSUM_COMPLETE checksum, or set ip_summed to CHECKSUM_NONE so that it can be recomputed from scratch. Kernel Hackers Manual July 2017 pskb_trim_rcsum 9 4.1.27 pskb_trim_rcsum trim received skb and update checksum int pskb_trim_rcsum struct sk_buff * skb unsigned int len skb buffer to trim len new length This is exactly the same as pskb_trim except that it ensures the checksum of received packets are still valid after the operation. Kernel Hackers Manual July 2017 skb_needs_linearize 9 4.1.27 skb_needs_linearize check if we need to linearize a given skb depending on the given device features. bool skb_needs_linearize struct sk_buff * skb netdev_features_t features skb socket buffer to check features net device features 1. skb has frag_list and the device doesn't support FRAGLIST, or 2. skb is fragmented and the device does not support SG. Kernel Hackers Manual July 2017 skb_get_timestamp 9 4.1.27 skb_get_timestamp get timestamp from a skb void skb_get_timestamp const struct sk_buff * skb struct timeval * stamp skb skb to get stamp from stamp pointer to struct timeval to store stamp in Timestamps are stored in the skb as offsets to a base timestamp. This function converts the offset back to a struct timeval and stores it in stamp. Kernel Hackers Manual July 2017 skb_tx_timestamp 9 4.1.27 skb_tx_timestamp Driver hook for transmit timestamping void skb_tx_timestamp struct sk_buff * skb skb A socket buffer. Ethernet MAC Drivers should call this function in their hard_xmit function immediately before giving the sk_buff to the MAC hardware. Specifically, one should make absolutely sure that this function is called before TX completion of this packet can trigger. Otherwise the packet could potentially already be freed. Kernel Hackers Manual July 2017 skb_checksum_complete 9 4.1.27 skb_checksum_complete Calculate checksum of an entire packet __sum16 skb_checksum_complete struct sk_buff * skb skb packet to process This function calculates the checksum over the entire packet plus the value of skb->csum. The latter can be used to supply the checksum of a pseudo header as used by TCP/UDP. It returns the checksum. For protocols that contain complete checksums such as ICMP/TCP/UDP, this function can be used to verify that checksum on received packets. In that case the function should return zero if the checksum is correct. In particular, this function will return zero if skb->ip_summed is CHECKSUM_UNNECESSARY which indicates that the hardware has already verified the correctness of the checksum. Kernel Hackers Manual July 2017 skb_checksum_none_assert 9 4.1.27 skb_checksum_none_assert make sure skb ip_summed is CHECKSUM_NONE void skb_checksum_none_assert const struct sk_buff * skb skb skb to check fresh skbs have their ip_summed set to CHECKSUM_NONE. Instead of forcing ip_summed to CHECKSUM_NONE, we can use this helper, to document places where we make this assertion. Kernel Hackers Manual July 2017 skb_head_is_locked 9 4.1.27 skb_head_is_locked Determine if the skb->head is locked down bool skb_head_is_locked const struct sk_buff * skb skb skb to check The head on skbs build around a head frag can be removed if they are not cloned. This function returns true if the skb head is locked down due to either being allocated via kmalloc, or by being a clone with multiple references to the head. Kernel Hackers Manual July 2017 skb_gso_network_seglen 9 4.1.27 skb_gso_network_seglen Return length of individual segments of a gso packet unsigned int skb_gso_network_seglen const struct sk_buff * skb 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. Kernel Hackers Manual July 2017 struct sock_common 9 4.1.27 struct sock_common minimal network layer representation of sockets struct sock_common { union {unnamed_union}; int skc_tx_queue_mapping; atomic_t skc_refcnt; }; {unnamed_union} anonymous skc_tx_queue_mapping tx queue number for this connection skc_refcnt reference count This is the minimal network layer representation of sockets, the header for struct sock and struct inet_timewait_sock. Kernel Hackers Manual July 2017 struct sock 9 4.1.27 struct sock network layer representation of sockets struct sock { struct sock_common __sk_common; #define sk_node __sk_common.skc_node #define sk_nulls_node __sk_common.skc_nulls_node #define sk_refcnt __sk_common.skc_refcnt #define sk_tx_queue_mapping __sk_common.skc_tx_queue_mapping #define sk_dontcopy_begin __sk_common.skc_dontcopy_begin #define sk_dontcopy_end __sk_common.skc_dontcopy_end #define sk_hash __sk_common.skc_hash #define sk_portpair __sk_common.skc_portpair #define sk_num __sk_common.skc_num #define sk_dport __sk_common.skc_dport #define sk_addrpair __sk_common.skc_addrpair #define sk_daddr __sk_common.skc_daddr #define sk_rcv_saddr __sk_common.skc_rcv_saddr #define sk_family __sk_common.skc_family #define sk_state __sk_common.skc_state #define sk_reuse __sk_common.skc_reuse #define sk_reuseport __sk_common.skc_reuseport #define sk_ipv6only __sk_common.skc_ipv6only #define sk_bound_dev_if __sk_common.skc_bound_dev_if #define sk_bind_node __sk_common.skc_bind_node #define sk_prot __sk_common.skc_prot #define sk_net __sk_common.skc_net #define sk_v6_daddr __sk_common.skc_v6_daddr #define sk_v6_rcv_saddr __sk_common.skc_v6_rcv_saddr #define sk_cookie __sk_common.skc_cookie socket_lock_t sk_lock; struct sk_buff_head sk_receive_queue; struct sk_backlog; #define sk_rmem_alloc sk_backlog.rmem_alloc int sk_forward_alloc; #ifdef CONFIG_RPS __u32 sk_rxhash; #endif u16 sk_incoming_cpu; __u32 sk_txhash; #ifdef CONFIG_NET_RX_BUSY_POLL unsigned int sk_napi_id; unsigned int sk_ll_usec; #endif atomic_t sk_drops; int sk_rcvbuf; struct sk_filter __rcu * sk_filter; struct socket_wq __rcu * sk_wq; #ifdef CONFIG_XFRM struct xfrm_policy * sk_policy[2]; #endif unsigned long sk_flags; struct dst_entry * sk_rx_dst; struct dst_entry __rcu * sk_dst_cache; spinlock_t sk_dst_lock; atomic_t sk_wmem_alloc; atomic_t sk_omem_alloc; int sk_sndbuf; struct sk_buff_head sk_write_queue; unsigned int sk_shutdown:2; unsigned int sk_no_check_tx:1; unsigned int sk_no_check_rx:1; unsigned int sk_userlocks:4; unsigned int sk_protocol:8; #define SK_PROTOCOL_MAX U8_MAX int sk_wmem_queued; gfp_t sk_allocation; u32 sk_pacing_rate; u32 sk_max_pacing_rate; netdev_features_t sk_route_caps; netdev_features_t sk_route_nocaps; int sk_gso_type; unsigned int sk_gso_max_size; u16 sk_gso_max_segs; int sk_rcvlowat; unsigned long sk_lingertime; struct sk_buff_head sk_error_queue; struct proto * sk_prot_creator; rwlock_t sk_callback_lock; int sk_err; int sk_err_soft; u32 sk_ack_backlog; u32 sk_max_ack_backlog; __u32 sk_priority; #if IS_ENABLED(CONFIG_CGROUP_NET_PRIO) __u32 sk_cgrp_prioidx; #endif struct pid * sk_peer_pid; const struct cred * sk_peer_cred; long sk_rcvtimeo; long sk_sndtimeo; void * sk_protinfo; struct timer_list sk_timer; ktime_t sk_stamp; u16 sk_tsflags; u32 sk_tskey; struct socket * sk_socket; void * sk_user_data; struct page_frag sk_frag; struct sk_buff * sk_send_head; __s32 sk_peek_off; int sk_write_pending; #ifdef CONFIG_SECURITY void * sk_security; #endif __u32 sk_mark; u32 sk_classid; struct cg_proto * sk_cgrp; void (* sk_state_change) (struct sock *sk); void (* sk_data_ready) (struct sock *sk); void (* sk_write_space) (struct sock *sk); void (* sk_error_report) (struct sock *sk); int (* sk_backlog_rcv) (struct sock *sk,struct sk_buff *skb); void (* sk_destruct) (struct sock *sk); }; __sk_common shared layout with inet_timewait_sock sk_lock synchronizer sk_receive_queue incoming packets sk_backlog always used with the per-socket spinlock held sk_forward_alloc space allocated forward sk_rxhash flow hash received from netif layer sk_incoming_cpu record cpu processing incoming packets sk_txhash computed flow hash for use on transmit sk_napi_id id of the last napi context to receive data for sk sk_ll_usec usecs to busypoll when there is no data sk_drops raw/udp drops counter sk_rcvbuf size of receive buffer in bytes sk_filter socket filtering instructions sk_wq sock wait queue and async head sk_policy[2] flow policy sk_flags SO_LINGER (l_onoff), SO_BROADCAST, SO_KEEPALIVE, SO_OOBINLINE settings, SO_TIMESTAMPING settings sk_rx_dst receive input route used by early demux sk_dst_cache destination cache sk_dst_lock destination cache lock sk_wmem_alloc transmit queue bytes committed sk_omem_alloc "o is option or other" sk_sndbuf size of send buffer in bytes sk_write_queue Packet sending queue sk_shutdown mask of SEND_SHUTDOWN and/or RCV_SHUTDOWN sk_no_check_tx SO_NO_CHECK setting, set checksum in TX packets sk_no_check_rx allow zero checksum in RX packets sk_userlocks SO_SNDBUF and SO_RCVBUF settings sk_protocol which protocol this socket belongs in this network family sk_wmem_queued persistent queue size sk_allocation allocation mode sk_pacing_rate Pacing rate (if supported by transport/packet scheduler) sk_max_pacing_rate Maximum pacing rate (SO_MAX_PACING_RATE) sk_route_caps route capabilities (e.g. NETIF_F_TSO) sk_route_nocaps forbidden route capabilities (e.g NETIF_F_GSO_MASK) sk_gso_type GSO type (e.g. SKB_GSO_TCPV4) sk_gso_max_size Maximum GSO segment size to build sk_gso_max_segs Maximum number of GSO segments sk_rcvlowat SO_RCVLOWAT setting sk_lingertime SO_LINGER l_linger setting sk_error_queue rarely used sk_prot_creator sk_prot of original sock creator (see ipv6_setsockopt, IPV6_ADDRFORM for instance) sk_callback_lock used with the callbacks in the end of this struct sk_err last error sk_err_soft errors that don't cause failure but are the cause of a persistent failure not just 'timed out' sk_ack_backlog current listen backlog sk_max_ack_backlog listen backlog set in listen sk_priority SO_PRIORITY setting sk_cgrp_prioidx socket group's priority map index sk_peer_pid struct pid for this socket's peer sk_peer_cred SO_PEERCRED setting sk_rcvtimeo SO_RCVTIMEO setting sk_sndtimeo SO_SNDTIMEO setting sk_protinfo private area, net family specific, when not using slab sk_timer sock cleanup timer sk_stamp time stamp of last packet received sk_tsflags SO_TIMESTAMPING socket options sk_tskey counter to disambiguate concurrent tstamp requests sk_socket Identd and reporting IO signals sk_user_data RPC layer private data sk_frag cached page frag sk_send_head front of stuff to transmit sk_peek_off current peek_offset value sk_write_pending a write to stream socket waits to start sk_security used by security modules sk_mark generic packet mark sk_classid this socket's cgroup classid sk_cgrp this socket's cgroup-specific proto data sk_state_change callback to indicate change in the state of the sock sk_data_ready callback to indicate there is data to be processed sk_write_space callback to indicate there is bf sending space available sk_error_report callback to indicate errors (e.g. MSG_ERRQUEUE) sk_backlog_rcv callback to process the backlog sk_destruct called at sock freeing time, i.e. when all refcnt == 0 Kernel Hackers Manual July 2017 sk_nulls_for_each_entry_offset 9 4.1.27 sk_nulls_for_each_entry_offset iterate over a list at a given struct offset sk_nulls_for_each_entry_offset tpos pos head offset tpos the type * to use as a loop cursor. pos the struct hlist_node to use as a loop cursor. head the head for your list. offset offset of hlist_node within the struct. Kernel Hackers Manual July 2017 unlock_sock_fast 9 4.1.27 unlock_sock_fast complement of lock_sock_fast void unlock_sock_fast struct sock * sk bool slow sk socket slow slow mode fast unlock socket for user context. If slow mode is on, we call regular release_sock Kernel Hackers Manual July 2017 sk_wmem_alloc_get 9 4.1.27 sk_wmem_alloc_get returns write allocations int sk_wmem_alloc_get const struct sock * sk sk socket Returns sk_wmem_alloc minus initial offset of one Kernel Hackers Manual July 2017 sk_rmem_alloc_get 9 4.1.27 sk_rmem_alloc_get returns read allocations int sk_rmem_alloc_get const struct sock * sk sk socket Returns sk_rmem_alloc Kernel Hackers Manual July 2017 sk_has_allocations 9 4.1.27 sk_has_allocations check if allocations are outstanding bool sk_has_allocations const struct sock * sk sk socket Returns true if socket has write or read allocations Kernel Hackers Manual July 2017 wq_has_sleeper 9 4.1.27 wq_has_sleeper check if there are any waiting processes bool wq_has_sleeper struct socket_wq * wq wq struct socket_wq Returns true if socket_wq has waiting processes The purpose of the wq_has_sleeper and sock_poll_wait is to wrap the memory barrier call. They were added due to the race found within the tcp code. CPU1 CPU2 sys_select receive packet ... ... __add_wait_queue update tp->rcv_nxt ... ... tp->rcv_nxt check sock_def_readable ... { schedule rcu_read_lock; wq = rcu_dereference(sk->sk_wq); if (wq && waitqueue_active(wq->wait)) wake_up_interruptible(wq->wait) ... } The race for tcp fires when the __add_wait_queue changes done by CPU1 stay in its cache, and so does the tp->rcv_nxt update on CPU2 side. The CPU1 could then endup calling schedule and sleep forever if there are no more data on the socket. Kernel Hackers Manual July 2017 sock_poll_wait 9 4.1.27 sock_poll_wait place memory barrier behind the poll_wait call. void sock_poll_wait struct file * filp wait_queue_head_t * wait_address poll_table * p filp file wait_address socket wait queue p poll_table See the comments in the wq_has_sleeper function. Kernel Hackers Manual July 2017 sk_page_frag 9 4.1.27 sk_page_frag return an appropriate page_frag struct page_frag * sk_page_frag struct sock * sk sk socket If socket allocation mode allows current thread to sleep, it means its safe to use the per task page_frag instead of the per socket one. Kernel Hackers Manual July 2017 sock_tx_timestamp 9 4.1.27 sock_tx_timestamp checks whether the outgoing packet is to be time stamped void sock_tx_timestamp const struct sock * sk __u8 * tx_flags sk socket sending this packet tx_flags completed with instructions for time stamping callers should take care of initial *tx_flags value (usually 0) Kernel Hackers Manual July 2017 sk_eat_skb 9 4.1.27 sk_eat_skb Release a skb if it is no longer needed void sk_eat_skb struct sock * sk struct sk_buff * skb sk socket to eat this skb from skb socket buffer to eat This routine must be called with interrupts disabled or with the socket locked so that the sk_buff queue operation is ok. Kernel Hackers Manual July 2017 sockfd_lookup 9 4.1.27 sockfd_lookup Go from a file number to its socket slot struct socket * sockfd_lookup int fd int * err fd file handle err pointer to an error code return The file handle passed in is locked and the socket it is bound too is returned. If an error occurs the err pointer is overwritten with a negative errno code and NULL is returned. The function checks for both invalid handles and passing a handle which is not a socket. On a success the socket object pointer is returned. Kernel Hackers Manual July 2017 sock_release 9 4.1.27 sock_release close a socket void sock_release struct socket * sock sock socket to close The socket is released from the protocol stack if it has a release callback, and the inode is then released if the socket is bound to an inode not a file. Kernel Hackers Manual July 2017 kernel_recvmsg 9 4.1.27 kernel_recvmsg Receive a message from a socket (kernel space) int kernel_recvmsg struct socket * sock struct msghdr * msg struct kvec * vec size_t num size_t size int flags sock The socket to receive the message from msg Received message vec Input s/g array for message data num Size of input s/g array size Number of bytes to read flags Message flags (MSG_DONTWAIT, etc...) On return the msg structure contains the scatter/gather array passed in the vec argument. The array is modified so that it consists of the unfilled portion of the original array. The returned value is the total number of bytes received, or an error. Kernel Hackers Manual July 2017 sock_register 9 4.1.27 sock_register add a socket protocol handler int sock_register const struct net_proto_family * ops ops description of protocol This function is called by a protocol handler that wants to advertise its address family, and have it linked into the socket interface. The value ops->family corresponds to the socket system call protocol family. Kernel Hackers Manual July 2017 sock_unregister 9 4.1.27 sock_unregister remove a protocol handler void sock_unregister int family family protocol family to remove This function is called by a protocol handler that wants to remove its address family, and have it unlinked from the new socket creation. If protocol handler is a module, then it can use module reference counts to protect against new references. If protocol handler is not a module then it needs to provide its own protection in the ops->create routine. Kernel Hackers Manual July 2017 __alloc_skb 9 4.1.27 __alloc_skb allocate a network buffer struct sk_buff * __alloc_skb unsigned int size gfp_t gfp_mask int flags int node size size to allocate gfp_mask allocation mask flags If SKB_ALLOC_FCLONE is set, allocate from fclone cache instead of head cache and allocate a cloned (child) skb. If SKB_ALLOC_RX is set, __GFP_MEMALLOC will be used for allocations in case the data is required for writeback node numa node to allocate memory on Allocate a new sk_buff. The returned buffer has no headroom and a tail room of at least size bytes. The object has a reference count of one. The return is the buffer. On a failure the return is NULL. Buffers may only be allocated from interrupts using a gfp_mask of GFP_ATOMIC. Kernel Hackers Manual July 2017 netdev_alloc_frag 9 4.1.27 netdev_alloc_frag allocate a page fragment void * netdev_alloc_frag unsigned int fragsz fragsz fragment size Allocates a frag from a page for receive buffer. Uses GFP_ATOMIC allocations. Kernel Hackers Manual July 2017 __netdev_alloc_skb 9 4.1.27 __netdev_alloc_skb allocate an skbuff for rx on a specific device struct sk_buff * __netdev_alloc_skb struct net_device * dev unsigned int length gfp_t gfp_mask dev network device to receive on length length to allocate gfp_mask get_free_pages mask, passed to alloc_skb Allocate a new sk_buff and assign it a usage count of one. The buffer has NET_SKB_PAD headroom built in. Users should allocate the headroom they think they need without accounting for the built in space. The built in space is used for optimisations. NULL is returned if there is no free memory. Kernel Hackers Manual July 2017 __napi_alloc_skb 9 4.1.27 __napi_alloc_skb allocate skbuff for rx in a specific NAPI instance struct sk_buff * __napi_alloc_skb struct napi_struct * napi unsigned int length gfp_t gfp_mask napi napi instance this buffer was allocated for length length to allocate gfp_mask get_free_pages mask, passed to alloc_skb and alloc_pages Allocate a new sk_buff for use in NAPI receive. This buffer will attempt to allocate the head from a special reserved region used only for NAPI Rx allocation. By doing this we can save several CPU cycles by avoiding having to disable and re-enable IRQs. NULL is returned if there is no free memory. Kernel Hackers Manual July 2017 __kfree_skb 9 4.1.27 __kfree_skb private function void __kfree_skb struct sk_buff * skb skb buffer Free an sk_buff. Release anything attached to the buffer. Clean the state. This is an internal helper function. Users should always call kfree_skb Kernel Hackers Manual July 2017 kfree_skb 9 4.1.27 kfree_skb free an sk_buff void kfree_skb struct sk_buff * skb skb buffer to free Drop a reference to the buffer and free it if the usage count has hit zero. Kernel Hackers Manual July 2017 skb_tx_error 9 4.1.27 skb_tx_error report an sk_buff xmit error void skb_tx_error struct sk_buff * skb skb buffer that triggered an error Report xmit error if a device callback is tracking this skb. skb must be freed afterwards. Kernel Hackers Manual July 2017 consume_skb 9 4.1.27 consume_skb free an skbuff void consume_skb struct sk_buff * skb skb buffer to free Drop a ref to the buffer and free it if the usage count has hit zero Functions identically to kfree_skb, but kfree_skb assumes that the frame is being dropped after a failure and notes that Kernel Hackers Manual July 2017 skb_morph 9 4.1.27 skb_morph morph one skb into another struct sk_buff * skb_morph struct sk_buff * dst struct sk_buff * src dst the skb to receive the contents src the skb to supply the contents This is identical to skb_clone except that the target skb is supplied by the user. The target skb is returned upon exit. Kernel Hackers Manual July 2017 skb_copy_ubufs 9 4.1.27 skb_copy_ubufs copy userspace skb frags buffers to kernel int skb_copy_ubufs struct sk_buff * skb gfp_t gfp_mask skb the skb to modify gfp_mask allocation priority This must be called on SKBTX_DEV_ZEROCOPY skb. It will copy all frags into kernel and drop the reference to userspace pages. If this function is called from an interrupt gfp_mask must be GFP_ATOMIC. Returns 0 on success or a negative error code on failure to allocate kernel memory to copy to. Kernel Hackers Manual July 2017 skb_clone 9 4.1.27 skb_clone duplicate an sk_buff struct sk_buff * skb_clone struct sk_buff * skb gfp_t gfp_mask skb buffer to clone gfp_mask allocation priority Duplicate an sk_buff. The new one is not owned by a socket. Both copies share the same packet data but not structure. The new buffer has a reference count of 1. If the allocation fails the function returns NULL otherwise the new buffer is returned. If this function is called from an interrupt gfp_mask must be GFP_ATOMIC. Kernel Hackers Manual July 2017 skb_copy 9 4.1.27 skb_copy create private copy of an sk_buff struct sk_buff * skb_copy const struct sk_buff * skb gfp_t gfp_mask skb buffer to copy gfp_mask allocation priority Make a copy of both an sk_buff and its data. This is used when the caller wishes to modify the data and needs a private copy of the data to alter. Returns NULL on failure or the pointer to the buffer on success. The returned buffer has a reference count of 1. As by-product this function converts non-linear sk_buff to linear one, so that sk_buff becomes completely private and caller is allowed to modify all the data of returned buffer. This means that this function is not recommended for use in circumstances when only header is going to be modified. Use pskb_copy instead. Kernel Hackers Manual July 2017 __pskb_copy_fclone 9 4.1.27 __pskb_copy_fclone create copy of an sk_buff with private head. struct sk_buff * __pskb_copy_fclone struct sk_buff * skb int headroom gfp_t gfp_mask bool fclone skb buffer to copy headroom headroom of new skb gfp_mask allocation priority fclone if true allocate the copy of the skb from the fclone cache instead of the head cache; it is recommended to set this to true for the cases where the copy will likely be cloned Make a copy of both an sk_buff and part of its data, located in header. Fragmented data remain shared. This is used when the caller wishes to modify only header of sk_buff and needs private copy of the header to alter. Returns NULL on failure or the pointer to the buffer on success. The returned buffer has a reference count of 1. Kernel Hackers Manual July 2017 pskb_expand_head 9 4.1.27 pskb_expand_head reallocate header of sk_buff int pskb_expand_head struct sk_buff * skb int nhead int ntail gfp_t gfp_mask skb buffer to reallocate nhead room to add at head ntail room to add at tail gfp_mask allocation priority Expands (or creates identical copy, if nhead and ntail are zero) header of skb. sk_buff itself is not changed. sk_buff MUST have reference count of 1. Returns zero in the case of success or error, if expansion failed. In the last case, sk_buff is not changed. All the pointers pointing into skb header may change and must be reloaded after call to this function. Kernel Hackers Manual July 2017 skb_copy_expand 9 4.1.27 skb_copy_expand copy and expand sk_buff struct sk_buff * skb_copy_expand const struct sk_buff * skb int newheadroom int newtailroom gfp_t gfp_mask skb buffer to copy newheadroom new free bytes at head newtailroom new free bytes at tail gfp_mask allocation priority Make a copy of both an sk_buff and its data and while doing so allocate additional space. This is used when the caller wishes to modify the data and needs a private copy of the data to alter as well as more space for new fields. Returns NULL on failure or the pointer to the buffer on success. The returned buffer has a reference count of 1. You must pass GFP_ATOMIC as the allocation priority if this function is called from an interrupt. Kernel Hackers Manual July 2017 skb_pad 9 4.1.27 skb_pad zero pad the tail of an skb int skb_pad struct sk_buff * skb int pad skb buffer to pad pad space to pad Ensure that a buffer is followed by a padding area that is zero filled. Used by network drivers which may DMA or transfer data beyond the buffer end onto the wire. May return error in out of memory cases. The skb is freed on error. Kernel Hackers Manual July 2017 pskb_put 9 4.1.27 pskb_put add data to the tail of a potentially fragmented buffer unsigned char * pskb_put struct sk_buff * skb struct sk_buff * tail int len skb start of the buffer to use tail tail fragment of the buffer to use len amount of data to add This function extends the used data area of the potentially fragmented buffer. tail must be the last fragment of skb -- or skb itself. If this would exceed the total buffer size the kernel will panic. A pointer to the first byte of the extra data is returned. Kernel Hackers Manual July 2017 skb_put 9 4.1.27 skb_put add data to a buffer unsigned char * skb_put struct sk_buff * skb unsigned int len skb buffer to use len amount of data to add This function extends the used data area of the buffer. If this would exceed the total buffer size the kernel will panic. A pointer to the first byte of the extra data is returned. Kernel Hackers Manual July 2017 skb_push 9 4.1.27 skb_push add data to the start of a buffer unsigned char * skb_push struct sk_buff * skb unsigned int len skb buffer to use len amount of data to add This function extends the used data area of the buffer at the buffer start. If this would exceed the total buffer headroom the kernel will panic. A pointer to the first byte of the extra data is returned. Kernel Hackers Manual July 2017 skb_pull 9 4.1.27 skb_pull remove data from the start of a buffer unsigned char * skb_pull struct sk_buff * skb unsigned int len skb buffer to use len amount of data to remove This function removes data from the start of a buffer, returning the memory to the headroom. A pointer to the next data in the buffer is returned. Once the data has been pulled future pushes will overwrite the old data. Kernel Hackers Manual July 2017 skb_trim 9 4.1.27 skb_trim remove end from a buffer void skb_trim struct sk_buff * skb unsigned int len skb buffer to alter len new length Cut the length of a buffer down by removing data from the tail. If the buffer is already under the length specified it is not modified. The skb must be linear. Kernel Hackers Manual July 2017 __pskb_pull_tail 9 4.1.27 __pskb_pull_tail advance tail of skb header unsigned char * __pskb_pull_tail struct sk_buff * skb int delta skb buffer to reallocate delta number of bytes to advance tail The function makes a sense only on a fragmented sk_buff, it expands header moving its tail forward and copying necessary data from fragmented part. sk_buff MUST have reference count of 1. Returns NULL (and sk_buff does not change) if pull failed or value of new tail of skb in the case of success. All the pointers pointing into skb header may change and must be reloaded after call to this function. Kernel Hackers Manual July 2017 skb_copy_bits 9 4.1.27 skb_copy_bits copy bits from skb to kernel buffer int skb_copy_bits const struct sk_buff * skb int offset void * to int len skb source skb offset offset in source to destination buffer len number of bytes to copy Copy the specified number of bytes from the source skb to the destination buffer. CAUTION ! : If its prototype is ever changed, check arch/{*}/net/{*}.S files, since it is called from BPF assembly code. Kernel Hackers Manual July 2017 skb_store_bits 9 4.1.27 skb_store_bits store bits from kernel buffer to skb int skb_store_bits struct sk_buff * skb int offset const void * from int len skb destination buffer offset offset in destination from source buffer len number of bytes to copy Copy the specified number of bytes from the source buffer to the destination skb. This function handles all the messy bits of traversing fragment lists and such. Kernel Hackers Manual July 2017 skb_zerocopy 9 4.1.27 skb_zerocopy Zero copy skb to skb int skb_zerocopy struct sk_buff * to struct sk_buff * from int len int hlen to destination buffer from source buffer len number of bytes to copy from source buffer hlen size of linear headroom in destination buffer Copies up to `len` bytes from `from` to `to` by creating references to the frags in the source buffer. The `hlen` as calculated by skb_zerocopy_headlen specifies the headroom in the `to` buffer. everything is OK -ENOMEM: couldn't orphan frags of from due to lack of memory -EFAULT: skb_copy_bits found some problem with skb geometry Kernel Hackers Manual July 2017 skb_dequeue 9 4.1.27 skb_dequeue remove from the head of the queue struct sk_buff * skb_dequeue struct sk_buff_head * list list list to dequeue from Remove the head of the list. The list lock is taken so the function may be used safely with other locking list functions. The head item is returned or NULL if the list is empty. Kernel Hackers Manual July 2017 skb_dequeue_tail 9 4.1.27 skb_dequeue_tail remove from the tail of the queue struct sk_buff * skb_dequeue_tail struct sk_buff_head * list list list to dequeue from Remove the tail of the list. The list lock is taken so the function may be used safely with other locking list functions. The tail item is returned or NULL if the list is empty. Kernel Hackers Manual July 2017 skb_queue_purge 9 4.1.27 skb_queue_purge empty a list void skb_queue_purge struct sk_buff_head * list list list to empty Delete all buffers on an sk_buff list. Each buffer is removed from the list and one reference dropped. This function takes the list lock and is atomic with respect to other list locking functions. Kernel Hackers Manual July 2017 skb_queue_head 9 4.1.27 skb_queue_head queue a buffer at the list head void skb_queue_head struct sk_buff_head * list struct sk_buff * newsk list list to use newsk buffer to queue Queue a buffer at the start of the list. This function takes the list lock and can be used safely with other locking sk_buff functions safely. A buffer cannot be placed on two lists at the same time. Kernel Hackers Manual July 2017 skb_queue_tail 9 4.1.27 skb_queue_tail queue a buffer at the list tail void skb_queue_tail struct sk_buff_head * list struct sk_buff * newsk list list to use newsk buffer to queue Queue a buffer at the tail of the list. This function takes the list lock and can be used safely with other locking sk_buff functions safely. A buffer cannot be placed on two lists at the same time. Kernel Hackers Manual July 2017 skb_unlink 9 4.1.27 skb_unlink remove a buffer from a list void skb_unlink struct sk_buff * skb struct sk_buff_head * list skb buffer to remove list list to use Remove a packet from a list. The list locks are taken and this function is atomic with respect to other list locked calls You must know what list the SKB is on. Kernel Hackers Manual July 2017 skb_append 9 4.1.27 skb_append append a buffer void skb_append struct sk_buff * old struct sk_buff * newsk struct sk_buff_head * list old buffer to insert after newsk buffer to insert list list to use Place a packet after a given packet in a list. The list locks are taken and this function is atomic with respect to other list locked calls. A buffer cannot be placed on two lists at the same time. Kernel Hackers Manual July 2017 skb_insert 9 4.1.27 skb_insert insert a buffer void skb_insert struct sk_buff * old struct sk_buff * newsk struct sk_buff_head * list old buffer to insert before newsk buffer to insert list list to use Place a packet before a given packet in a list. The list locks are taken and this function is atomic with respect to other list locked calls. A buffer cannot be placed on two lists at the same time. Kernel Hackers Manual July 2017 skb_split 9 4.1.27 skb_split Split fragmented skb to two parts at length len. void skb_split struct sk_buff * skb struct sk_buff * skb1 const u32 len skb the buffer to split skb1 the buffer to receive the second part len new length for skb Kernel Hackers Manual July 2017 skb_prepare_seq_read 9 4.1.27 skb_prepare_seq_read Prepare a sequential read of skb data void skb_prepare_seq_read struct sk_buff * skb unsigned int from unsigned int to struct skb_seq_state * st skb the buffer to read from lower offset of data to be read to upper offset of data to be read st state variable Initializes the specified state variable. Must be called before invoking skb_seq_read for the first time. Kernel Hackers Manual July 2017 skb_seq_read 9 4.1.27 skb_seq_read Sequentially read skb data unsigned int skb_seq_read unsigned int consumed const u8 ** data struct skb_seq_state * st consumed number of bytes consumed by the caller so far data destination pointer for data to be returned st state variable Reads a block of skb data at consumed relative to the lower offset specified to skb_prepare_seq_read. Assigns the head of the data block to data and returns the length of the block or 0 if the end of the skb data or the upper offset has been reached. The caller is not required to consume all of the data returned, i.e. consumed is typically set to the number of bytes already consumed and the next call to skb_seq_read will return the remaining part of the block. The size of each block of data returned can be arbitrary, this limitation is the cost for zerocopy sequential reads of potentially non linear data. Fragment lists within fragments are not implemented at the moment, state->root_skb could be replaced with a stack for this purpose. Kernel Hackers Manual July 2017 skb_abort_seq_read 9 4.1.27 skb_abort_seq_read Abort a sequential read of skb data void skb_abort_seq_read struct skb_seq_state * st st state variable Must be called if skb_seq_read was not called until it returned 0. Kernel Hackers Manual July 2017 skb_find_text 9 4.1.27 skb_find_text Find a text pattern in skb data unsigned int skb_find_text struct sk_buff * skb unsigned int from unsigned int to struct ts_config * config skb the buffer to look in from search offset to search limit config textsearch configuration Finds a pattern in the skb data according to the specified textsearch configuration. Use textsearch_next to retrieve subsequent occurrences of the pattern. Returns the offset to the first occurrence or UINT_MAX if no match was found. Kernel Hackers Manual July 2017 skb_append_datato_frags 9 4.1.27 skb_append_datato_frags append the user data to a skb int skb_append_datato_frags struct sock * sk struct sk_buff * skb int (*getfrag) void *from, char *to, int offset, int len, int odd, struct sk_buff *skb void * from int length sk sock structure skb skb structure to be appended with user data. getfrag call back function to be used for getting the user data from pointer to user message iov length length of the iov message This procedure append the user data in the fragment part of the skb if any page alloc fails user this procedure returns -ENOMEM Kernel Hackers Manual July 2017 skb_pull_rcsum 9 4.1.27 skb_pull_rcsum pull skb and update receive checksum unsigned char * skb_pull_rcsum struct sk_buff * skb unsigned int len skb buffer to update len length of data pulled This function performs an skb_pull on the packet and updates the CHECKSUM_COMPLETE checksum. It should be used on receive path processing instead of skb_pull unless you know that the checksum difference is zero (e.g., a valid IP header) or you are setting ip_summed to CHECKSUM_NONE. Kernel Hackers Manual July 2017 skb_segment 9 4.1.27 skb_segment Perform protocol segmentation on skb. struct sk_buff * skb_segment struct sk_buff * head_skb netdev_features_t features head_skb buffer to segment features features for the output path (see dev->features) This function performs segmentation on the given skb. It returns a pointer to the first in a list of new skbs for the segments. In case of error it returns ERR_PTR(err). Kernel Hackers Manual July 2017 skb_cow_data 9 4.1.27 skb_cow_data Check that a socket buffer's data buffers are writable int skb_cow_data struct sk_buff * skb int tailbits struct sk_buff ** trailer skb The socket buffer to check. tailbits Amount of trailing space to be added trailer Returned pointer to the skb where the tailbits space begins Make sure that the data buffers attached to a socket buffer are writable. If they are not, private copies are made of the data buffers and the socket buffer is set to use these instead. If tailbits is given, make sure that there is space to write tailbits bytes of data beyond current end of socket buffer. trailer will be set to point to the skb in which this space begins. The number of scatterlist elements required to completely map the COW'd and extended socket buffer will be returned. Kernel Hackers Manual July 2017 skb_clone_sk 9 4.1.27 skb_clone_sk create clone of skb, and take reference to socket struct sk_buff * skb_clone_sk struct sk_buff * skb skb the skb to clone This function creates a clone of a buffer that holds a reference on sk_refcnt. Buffers created via this function are meant to be returned using sock_queue_err_skb, or free via kfree_skb. When passing buffers allocated with this function to sock_queue_err_skb it is necessary to wrap the call with sock_hold/sock_put in order to prevent the socket from being released prior to being enqueued on the sk_error_queue. Kernel Hackers Manual July 2017 skb_partial_csum_set 9 4.1.27 skb_partial_csum_set set up and verify partial csum values for packet bool skb_partial_csum_set struct sk_buff * skb u16 start u16 off skb the skb to set start the number of bytes after skb->data to start checksumming. off the offset from start to place the checksum. For untrusted partially-checksummed packets, we need to make sure the values for skb->csum_start and skb->csum_offset are valid so we don't oops. This function checks and sets those values and skb->ip_summed: if this returns false you should drop the packet. Kernel Hackers Manual July 2017 skb_checksum_setup 9 4.1.27 skb_checksum_setup set up partial checksum offset int skb_checksum_setup struct sk_buff * skb bool recalculate skb the skb to set up recalculate if true the pseudo-header checksum will be recalculated Kernel Hackers Manual July 2017 skb_try_coalesce 9 4.1.27 skb_try_coalesce try to merge skb to prior one bool skb_try_coalesce struct sk_buff * to struct sk_buff * from bool * fragstolen int * delta_truesize to prior buffer from buffer to add fragstolen pointer to boolean delta_truesize how much more was allocated than was requested Kernel Hackers Manual July 2017 skb_scrub_packet 9 4.1.27 skb_scrub_packet scrub an skb void skb_scrub_packet struct sk_buff * skb bool xnet skb buffer to clean xnet packet is crossing netns skb_scrub_packet can be used after encapsulating or decapsulting a packet into/from a tunnel. Some information have to be cleared during these operations. skb_scrub_packet can also be used to clean a skb before injecting it in another namespace (xnet == true). We have to clear all information in the skb that could impact namespace isolation. Kernel Hackers Manual July 2017 skb_gso_transport_seglen 9 4.1.27 skb_gso_transport_seglen Return length of individual segments of a gso packet unsigned int skb_gso_transport_seglen const struct sk_buff * skb skb GSO skb skb_gso_transport_seglen is used to determine the real size of the individual segments, including Layer4 headers (TCP/UDP). The MAC/L2 or network (IP, IPv6) headers are not accounted for. Kernel Hackers Manual July 2017 alloc_skb_with_frags 9 4.1.27 alloc_skb_with_frags allocate skb with page frags struct sk_buff * alloc_skb_with_frags unsigned long header_len unsigned long data_len int max_page_order int * errcode gfp_t gfp_mask header_len size of linear part data_len needed length in frags max_page_order max page order desired. errcode pointer to error code if any gfp_mask allocation mask This can be used to allocate a paged skb, given a maximal order for frags. Kernel Hackers Manual July 2017 sk_ns_capable 9 4.1.27 sk_ns_capable General socket capability test bool sk_ns_capable const struct sock * sk struct user_namespace * user_ns int cap sk Socket to use a capability on or through user_ns The user namespace of the capability to use cap The capability to use Test to see if the opener of the socket had when the socket was created and the current process has the capability cap in the user namespace user_ns. Kernel Hackers Manual July 2017 sk_capable 9 4.1.27 sk_capable Socket global capability test bool sk_capable const struct sock * sk int cap sk Socket to use a capability on or through cap The global capability to use Test to see if the opener of the socket had when the socket was created and the current process has the capability cap in all user namespaces. Kernel Hackers Manual July 2017 sk_net_capable 9 4.1.27 sk_net_capable Network namespace socket capability test bool sk_net_capable const struct sock * sk int cap sk Socket to use a capability on or through cap The capability to use Test to see if the opener of the socket had when the socket was created and the current process has the capability cap over the network namespace the socket is a member of. Kernel Hackers Manual July 2017 sk_set_memalloc 9 4.1.27 sk_set_memalloc sets SOCK_MEMALLOC void sk_set_memalloc struct sock * sk sk socket to set it on Set SOCK_MEMALLOC on a socket for access to emergency reserves. It's the responsibility of the admin to adjust min_free_kbytes to meet the requirements Kernel Hackers Manual July 2017 sk_alloc 9 4.1.27 sk_alloc All socket objects are allocated here struct sock * sk_alloc struct net * net int family gfp_t priority struct proto * prot net the applicable net namespace family protocol family priority for allocation (GFP_KERNEL, GFP_ATOMIC, etc) prot struct proto associated with this new sock instance Kernel Hackers Manual July 2017 sk_clone_lock 9 4.1.27 sk_clone_lock clone a socket, and lock its clone struct sock * sk_clone_lock const struct sock * sk const gfp_t priority sk the socket to clone priority for allocation (GFP_KERNEL, GFP_ATOMIC, etc) Caller must unlock socket even in error path (bh_unlock_sock(newsk)) Kernel Hackers Manual July 2017 skb_page_frag_refill 9 4.1.27 skb_page_frag_refill check that a page_frag contains enough room bool skb_page_frag_refill unsigned int sz struct page_frag * pfrag gfp_t gfp sz minimum size of the fragment we want to get pfrag pointer to page_frag gfp priority for memory allocation While this allocator tries to use high order pages, there is no guarantee that allocations succeed. Therefore, sz MUST be less or equal than PAGE_SIZE. Kernel Hackers Manual July 2017 sk_wait_data 9 4.1.27 sk_wait_data wait for data to arrive at sk_receive_queue int sk_wait_data struct sock * sk long * timeo sk sock to wait on timeo for how long Now socket state including sk->sk_err is changed only under lock, hence we may omit checks after joining wait queue. We check receive queue before schedule only as optimization; it is very likely that release_sock added new data. Kernel Hackers Manual July 2017 __sk_mem_schedule 9 4.1.27 __sk_mem_schedule increase sk_forward_alloc and memory_allocated int __sk_mem_schedule struct sock * sk int size int kind sk socket size memory size to allocate kind allocation type If kind is SK_MEM_SEND, it means wmem allocation. Otherwise it means rmem allocation. This function assumes that protocols which have memory_pressure use sk_wmem_queued as write buffer accounting. Kernel Hackers Manual July 2017 __sk_mem_reclaim 9 4.1.27 __sk_mem_reclaim reclaim memory_allocated void __sk_mem_reclaim struct sock * sk sk socket Kernel Hackers Manual July 2017 lock_sock_fast 9 4.1.27 lock_sock_fast fast version of lock_sock bool lock_sock_fast struct sock * sk sk socket This version should be used for very small section, where process wont block return false if fast path is taken sk_lock.slock locked, owned = 0, BH disabled return true if slow path is taken sk_lock.slock unlocked, owned = 1, BH enabled Kernel Hackers Manual July 2017 __skb_recv_datagram 9 4.1.27 __skb_recv_datagram Receive a datagram skbuff struct sk_buff * __skb_recv_datagram struct sock * sk unsigned int flags int * peeked int * off int * err sk socket flags MSG_ flags peeked returns non-zero if this packet has been seen before off an offset in bytes to peek skb from. Returns an offset within an skb where data actually starts err error code returned Get a datagram skbuff, understands the peeking, nonblocking wakeups and possible races. This replaces identical code in packet, raw and udp, as well as the IPX AX.25 and Appletalk. It also finally fixes the long standing peek and read race for datagram sockets. If you alter this routine remember it must be re-entrant. This function will lock the socket if a skb is returned, so the caller needs to unlock the socket in that case (usually by calling skb_free_datagram) * It does not lock socket since today. This function is * free of race conditions. This measure should/can improve * significantly datagram socket latencies at high loads, * when data copying to user space takes lots of time. * (BTW I've just killed the last cli in IP/IPv6/core/netlink/packet * 8) Great win.) * --ANK (980729) The order of the tests when we find no data waiting are specified quite explicitly by POSIX 1003.1g, don't change them without having the standard around please. Kernel Hackers Manual July 2017 skb_kill_datagram 9 4.1.27 skb_kill_datagram Free a datagram skbuff forcibly int skb_kill_datagram struct sock * sk struct sk_buff * skb unsigned int flags sk socket skb datagram skbuff flags MSG_ flags This function frees a datagram skbuff that was received by skb_recv_datagram. The flags argument must match the one used for skb_recv_datagram. If the MSG_PEEK flag is set, and the packet is still on the receive queue of the socket, it will be taken off the queue before it is freed. This function currently only disables BH when acquiring the sk_receive_queue lock. Therefore it must not be used in a context where that lock is acquired in an IRQ context. It returns 0 if the packet was removed by us. Kernel Hackers Manual July 2017 skb_copy_datagram_iter 9 4.1.27 skb_copy_datagram_iter Copy a datagram to an iovec iterator. int skb_copy_datagram_iter const struct sk_buff * skb int offset struct iov_iter * to int len skb buffer to copy offset offset in the buffer to start copying from to iovec iterator to copy to len amount of data to copy from buffer to iovec Kernel Hackers Manual July 2017 skb_copy_datagram_from_iter 9 4.1.27 skb_copy_datagram_from_iter Copy a datagram from an iov_iter. int skb_copy_datagram_from_iter struct sk_buff * skb int offset struct iov_iter * from int len skb buffer to copy offset offset in the buffer to start copying to from the copy source len amount of data to copy to buffer from iovec Returns 0 or -EFAULT. Kernel Hackers Manual July 2017 zerocopy_sg_from_iter 9 4.1.27 zerocopy_sg_from_iter Build a zerocopy datagram from an iov_iter int zerocopy_sg_from_iter struct sk_buff * skb struct iov_iter * from skb buffer to copy from the source to copy from The function will first copy up to headlen, and then pin the userspace pages and build frags through them. Returns 0, -EFAULT or -EMSGSIZE. Kernel Hackers Manual July 2017 skb_copy_and_csum_datagram_msg 9 4.1.27 skb_copy_and_csum_datagram_msg Copy and checksum skb to user iovec. int skb_copy_and_csum_datagram_msg struct sk_buff * skb int hlen struct msghdr * msg skb skbuff hlen hardware length msg destination Caller _must_ check that skb will fit to this iovec. 0 - success. -EINVAL - checksum failure. -EFAULT - fault during copy. Kernel Hackers Manual July 2017 datagram_poll 9 4.1.27 datagram_poll generic datagram poll unsigned int datagram_poll struct file * file struct socket * sock poll_table * wait file file struct sock socket wait poll table Again totally generic. This also handles sequenced packet sockets providing the socket receive queue is only ever holding data ready to receive. when you _don't_ use this routine for this protocol, and you use a different write policy from sock_writeable then please supply your own write_space callback. Kernel Hackers Manual July 2017 sk_stream_write_space 9 4.1.27 sk_stream_write_space stream socket write_space callback. void sk_stream_write_space struct sock * sk sk socket write proper description Kernel Hackers Manual July 2017 sk_stream_wait_connect 9 4.1.27 sk_stream_wait_connect Wait for a socket to get into the connected state int sk_stream_wait_connect struct sock * sk long * timeo_p sk sock to wait on timeo_p for how long to wait Must be called with the socket locked. Kernel Hackers Manual July 2017 sk_stream_wait_memory 9 4.1.27 sk_stream_wait_memory Wait for more memory for a socket int sk_stream_wait_memory struct sock * sk long * timeo_p sk socket to wait for memory timeo_p for how long Kernel Hackers Manual July 2017 sk_filter 9 4.1.27 sk_filter run a packet through a socket filter int sk_filter struct sock * sk struct sk_buff * skb sk sock associated with sk_buff skb buffer to filter Run the filter code and then cut skb->data to correct size returned by SK_RUN_FILTER. If pkt_len is 0 we toss packet. If skb->len is smaller than pkt_len we keep whole skb->data. This is the socket level wrapper to SK_RUN_FILTER. It returns 0 if the packet should be accepted or -EPERM if the packet should be tossed. Kernel Hackers Manual July 2017 bpf_check_classic 9 4.1.27 bpf_check_classic verify socket filter code int bpf_check_classic const struct sock_filter * filter unsigned int flen filter filter to verify flen length of filter Check the user's filter code. If we let some ugly filter code slip through kaboom! The filter must contain no references or jumps that are out of range, no illegal instructions, and must end with a RET instruction. All jumps are forward as they are not signed. Returns 0 if the rule set is legal or -EINVAL if not. Kernel Hackers Manual July 2017 bpf_prog_create 9 4.1.27 bpf_prog_create create an unattached filter int bpf_prog_create struct bpf_prog ** pfp struct sock_fprog_kern * fprog pfp the unattached filter that is created fprog the filter program Create a filter independent of any socket. We first run some sanity checks on it to make sure it does not explode on us later. If an error occurs or there is insufficient memory for the filter a negative errno code is returned. On success the return is zero. Kernel Hackers Manual July 2017 sk_attach_filter 9 4.1.27 sk_attach_filter attach a socket filter int sk_attach_filter struct sock_fprog * fprog struct sock * sk fprog the filter program sk the socket to use Attach the user's filter code. We first run some sanity checks on it to make sure it does not explode on us later. If an error occurs or there is insufficient memory for the filter a negative errno code is returned. On success the return is zero. Kernel Hackers Manual July 2017 struct gnet_stats_basic 9 4.1.27 struct gnet_stats_basic byte/packet throughput statistics struct gnet_stats_basic { __u64 bytes; __u32 packets; }; bytes number of seen bytes packets number of seen packets Kernel Hackers Manual July 2017 struct gnet_stats_rate_est 9 4.1.27 struct gnet_stats_rate_est rate estimator struct gnet_stats_rate_est { __u32 bps; __u32 pps; }; bps current byte rate pps current packet rate Kernel Hackers Manual July 2017 struct gnet_stats_rate_est64 9 4.1.27 struct gnet_stats_rate_est64 rate estimator struct gnet_stats_rate_est64 { __u64 bps; __u64 pps; }; bps current byte rate pps current packet rate Kernel Hackers Manual July 2017 struct gnet_stats_queue 9 4.1.27 struct gnet_stats_queue queuing statistics struct gnet_stats_queue { __u32 qlen; __u32 backlog; __u32 drops; __u32 requeues; __u32 overlimits; }; qlen queue length backlog backlog size of queue drops number of dropped packets requeues number of requeues overlimits number of enqueues over the limit Kernel Hackers Manual July 2017 struct gnet_estimator 9 4.1.27 struct gnet_estimator rate estimator configuration struct gnet_estimator { signed char interval; unsigned char ewma_log; }; interval sampling period ewma_log the log of measurement window weight Kernel Hackers Manual July 2017 gnet_stats_start_copy_compat 9 4.1.27 gnet_stats_start_copy_compat start dumping procedure in compatibility mode int gnet_stats_start_copy_compat struct sk_buff * skb int type int tc_stats_type int xstats_type spinlock_t * lock struct gnet_dump * d skb socket buffer to put statistics TLVs into type TLV type for top level statistic TLV tc_stats_type TLV type for backward compatibility struct tc_stats TLV xstats_type TLV type for backward compatibility xstats TLV lock statistics lock d dumping handle Initializes the dumping handle, grabs the statistic lock and appends an empty TLV header to the socket buffer for use a container for all other statistic TLVS. The dumping handle is marked to be in backward compatibility mode telling all gnet_stats_copy_XXX functions to fill a local copy of struct tc_stats. Returns 0 on success or -1 if the room in the socket buffer was not sufficient. Kernel Hackers Manual July 2017 gnet_stats_start_copy 9 4.1.27 gnet_stats_start_copy start dumping procedure in compatibility mode int gnet_stats_start_copy struct sk_buff * skb int type spinlock_t * lock struct gnet_dump * d skb socket buffer to put statistics TLVs into type TLV type for top level statistic TLV lock statistics lock d dumping handle Initializes the dumping handle, grabs the statistic lock and appends an empty TLV header to the socket buffer for use a container for all other statistic TLVS. Returns 0 on success or -1 if the room in the socket buffer was not sufficient. Kernel Hackers Manual July 2017 gnet_stats_copy_basic 9 4.1.27 gnet_stats_copy_basic copy basic statistics into statistic TLV int gnet_stats_copy_basic struct gnet_dump * d struct gnet_stats_basic_cpu __percpu * cpu struct gnet_stats_basic_packed * b d dumping handle cpu -- undescribed -- b basic statistics Appends the basic statistics to the top level TLV created by gnet_stats_start_copy. Returns 0 on success or -1 with the statistic lock released if the room in the socket buffer was not sufficient. Kernel Hackers Manual July 2017 gnet_stats_copy_rate_est 9 4.1.27 gnet_stats_copy_rate_est copy rate estimator statistics into statistics TLV int gnet_stats_copy_rate_est struct gnet_dump * d const struct gnet_stats_basic_packed * b struct gnet_stats_rate_est64 * r d dumping handle b basic statistics r rate estimator statistics Appends the rate estimator statistics to the top level TLV created by gnet_stats_start_copy. Returns 0 on success or -1 with the statistic lock released if the room in the socket buffer was not sufficient. Kernel Hackers Manual July 2017 gnet_stats_copy_queue 9 4.1.27 gnet_stats_copy_queue copy queue statistics into statistics TLV int gnet_stats_copy_queue struct gnet_dump * d struct gnet_stats_queue __percpu * cpu_q struct gnet_stats_queue * q __u32 qlen d dumping handle cpu_q per cpu queue statistics q queue statistics qlen queue length statistics Appends the queue statistics to the top level TLV created by gnet_stats_start_copy. Using per cpu queue statistics if they are available. Returns 0 on success or -1 with the statistic lock released if the room in the socket buffer was not sufficient. Kernel Hackers Manual July 2017 gnet_stats_copy_app 9 4.1.27 gnet_stats_copy_app copy application specific statistics into statistics TLV int gnet_stats_copy_app struct gnet_dump * d void * st int len d dumping handle st application specific statistics data len length of data Appends the application specific statistics to the top level TLV created by gnet_stats_start_copy and remembers the data for XSTATS if the dumping handle is in backward compatibility mode. Returns 0 on success or -1 with the statistic lock released if the room in the socket buffer was not sufficient. Kernel Hackers Manual July 2017 gnet_stats_finish_copy 9 4.1.27 gnet_stats_finish_copy finish dumping procedure int gnet_stats_finish_copy struct gnet_dump * d d dumping handle Corrects the length of the top level TLV to include all TLVs added by gnet_stats_copy_XXX calls. Adds the backward compatibility TLVs if gnet_stats_start_copy_compat was used and releases the statistics lock. Returns 0 on success or -1 with the statistic lock released if the room in the socket buffer was not sufficient. Kernel Hackers Manual July 2017 gen_new_estimator 9 4.1.27 gen_new_estimator create a new rate estimator int gen_new_estimator struct gnet_stats_basic_packed * bstats struct gnet_stats_basic_cpu __percpu * cpu_bstats struct gnet_stats_rate_est64 * rate_est spinlock_t * stats_lock struct nlattr * opt bstats basic statistics cpu_bstats -- undescribed -- rate_est rate estimator statistics stats_lock statistics lock opt rate estimator configuration TLV Creates a new rate estimator with bstats as source and rate_est as destination. A new timer with the interval specified in the configuration TLV is created. Upon each interval, the latest statistics will be read from bstats and the estimated rate will be stored in rate_est with the statistics lock grabbed during this period. Returns 0 on success or a negative error code. Kernel Hackers Manual July 2017 gen_kill_estimator 9 4.1.27 gen_kill_estimator remove a rate estimator void gen_kill_estimator struct gnet_stats_basic_packed * bstats struct gnet_stats_rate_est64 * rate_est bstats basic statistics rate_est rate estimator statistics Removes the rate estimator specified by bstats and rate_est. Caller should respect an RCU grace period before freeing stats_lock Kernel Hackers Manual July 2017 gen_replace_estimator 9 4.1.27 gen_replace_estimator replace rate estimator configuration int gen_replace_estimator struct gnet_stats_basic_packed * bstats struct gnet_stats_basic_cpu __percpu * cpu_bstats struct gnet_stats_rate_est64 * rate_est spinlock_t * stats_lock struct nlattr * opt bstats basic statistics cpu_bstats -- undescribed -- rate_est rate estimator statistics stats_lock statistics lock opt rate estimator configuration TLV Replaces the configuration of a rate estimator by calling gen_kill_estimator and gen_new_estimator. Returns 0 on success or a negative error code. Kernel Hackers Manual July 2017 gen_estimator_active 9 4.1.27 gen_estimator_active test if estimator is currently in use bool gen_estimator_active const struct gnet_stats_basic_packed * bstats const struct gnet_stats_rate_est64 * rate_est bstats basic statistics rate_est rate estimator statistics Returns true if estimator is active, and false if not. Kernel Hackers Manual July 2017 xdr_encode_opaque_fixed 9 4.1.27 xdr_encode_opaque_fixed Encode fixed length opaque data __be32 * xdr_encode_opaque_fixed __be32 * p const void * ptr unsigned int nbytes p pointer to current position in XDR buffer. ptr pointer to data to encode (or NULL) nbytes size of data. Copy the array of data of length nbytes at ptr to the XDR buffer at position p, then align to the next 32-bit boundary by padding with zero bytes (see RFC1832). if ptr is NULL, only the padding is performed. Returns the updated current XDR buffer position Kernel Hackers Manual July 2017 xdr_encode_opaque 9 4.1.27 xdr_encode_opaque Encode variable length opaque data __be32 * xdr_encode_opaque __be32 * p const void * ptr unsigned int nbytes p pointer to current position in XDR buffer. ptr pointer to data to encode (or NULL) nbytes size of data. Returns the updated current XDR buffer position Kernel Hackers Manual July 2017 xdr_terminate_string 9 4.1.27 xdr_terminate_string '\0'-terminate a string residing in an xdr_buf void xdr_terminate_string struct xdr_buf * buf const u32 len buf XDR buffer where string resides len length of string, in bytes Kernel Hackers Manual July 2017 _copy_from_pages 9 4.1.27 _copy_from_pages void _copy_from_pages char * p struct page ** pages size_t pgbase size_t len p pointer to destination pages array of pages pgbase offset of source data len length Copies data into an arbitrary memory location from an array of pages The copy is assumed to be non-overlapping. Kernel Hackers Manual July 2017 xdr_stream_pos 9 4.1.27 xdr_stream_pos Return the current offset from the start of the xdr_stream unsigned int xdr_stream_pos const struct xdr_stream * xdr xdr pointer to struct xdr_stream Kernel Hackers Manual July 2017 xdr_init_encode 9 4.1.27 xdr_init_encode Initialize a struct xdr_stream for sending data. void xdr_init_encode struct xdr_stream * xdr struct xdr_buf * buf __be32 * p xdr pointer to xdr_stream struct buf pointer to XDR buffer in which to encode data p current pointer inside XDR buffer at the moment the RPC client only passes the length of our scratch buffer in the xdr_buf's header kvec. Previously this meant we needed to call xdr_adjust_iovec after encoding the data. With the new scheme, the xdr_stream manages the details of the buffer length, and takes care of adjusting the kvec length for us. Kernel Hackers Manual July 2017 xdr_commit_encode 9 4.1.27 xdr_commit_encode Ensure all data is written to buffer void xdr_commit_encode struct xdr_stream * xdr xdr pointer to xdr_stream We handle encoding across page boundaries by giving the caller a temporary location to write to, then later copying the data into place; xdr_commit_encode does that copying. Normally the caller doesn't need to call this directly, as the following xdr_reserve_space will do it. But an explicit call may be required at the end of encoding, or any other time when the xdr_buf data might be read. Kernel Hackers Manual July 2017 xdr_reserve_space 9 4.1.27 xdr_reserve_space Reserve buffer space for sending __be32 * xdr_reserve_space struct xdr_stream * xdr size_t nbytes xdr pointer to xdr_stream nbytes number of bytes to reserve Checks that we have enough buffer space to encode 'nbytes' more bytes of data. If so, update the total xdr_buf length, and adjust the length of the current kvec. Kernel Hackers Manual July 2017 xdr_truncate_encode 9 4.1.27 xdr_truncate_encode truncate an encode buffer void xdr_truncate_encode struct xdr_stream * xdr size_t len xdr pointer to xdr_stream len new length of buffer Truncates the xdr stream, so that xdr->buf->len == len, and xdr->p points at offset len from the start of the buffer, and head, tail, and page lengths are adjusted to correspond. If this means moving xdr->p to a different buffer, we assume that that the end pointer should be set to the end of the current page, except in the case of the head buffer when we assume the head buffer's current length represents the end of the available buffer. This is *not* safe to use on a buffer that already has inlined page cache pages (as in a zero-copy server read reply), except for the simple case of truncating from one position in the tail to another. Kernel Hackers Manual July 2017 xdr_restrict_buflen 9 4.1.27 xdr_restrict_buflen decrease available buffer space int xdr_restrict_buflen struct xdr_stream * xdr int newbuflen xdr pointer to xdr_stream newbuflen new maximum number of bytes available Adjust our idea of how much space is available in the buffer. If we've already used too much space in the buffer, returns -1. If the available space is already smaller than newbuflen, returns 0 and does nothing. Otherwise, adjusts xdr->buf->buflen to newbuflen and ensures xdr->end is set at most offset newbuflen from the start of the buffer. Kernel Hackers Manual July 2017 xdr_write_pages 9 4.1.27 xdr_write_pages Insert a list of pages into an XDR buffer for sending void xdr_write_pages struct xdr_stream * xdr struct page ** pages unsigned int base unsigned int len xdr pointer to xdr_stream pages list of pages base offset of first byte len length of data in bytes Kernel Hackers Manual July 2017 xdr_init_decode 9 4.1.27 xdr_init_decode Initialize an xdr_stream for decoding data. void xdr_init_decode struct xdr_stream * xdr struct xdr_buf * buf __be32 * p xdr pointer to xdr_stream struct buf pointer to XDR buffer from which to decode data p current pointer inside XDR buffer Kernel Hackers Manual July 2017 xdr_init_decode_pages 9 4.1.27 xdr_init_decode_pages Initialize an xdr_stream for decoding data. void xdr_init_decode_pages struct xdr_stream * xdr struct xdr_buf * buf struct page ** pages unsigned int len xdr pointer to xdr_stream struct buf pointer to XDR buffer from which to decode data pages list of pages to decode into len length in bytes of buffer in pages Kernel Hackers Manual July 2017 xdr_set_scratch_buffer 9 4.1.27 xdr_set_scratch_buffer Attach a scratch buffer for decoding data. void xdr_set_scratch_buffer struct xdr_stream * xdr void * buf size_t buflen xdr pointer to xdr_stream struct buf pointer to an empty buffer buflen size of 'buf' The scratch buffer is used when decoding from an array of pages. If an xdr_inline_decode call spans across page boundaries, then we copy the data into the scratch buffer in order to allow linear access. Kernel Hackers Manual July 2017 xdr_inline_decode 9 4.1.27 xdr_inline_decode Retrieve XDR data to decode __be32 * xdr_inline_decode struct xdr_stream * xdr size_t nbytes xdr pointer to xdr_stream struct nbytes number of bytes of data to decode Check if the input buffer is long enough to enable us to decode 'nbytes' more bytes of data starting at the current position. If so return the current pointer, then update the current pointer position. Kernel Hackers Manual July 2017 xdr_read_pages 9 4.1.27 xdr_read_pages Ensure page-based XDR data to decode is aligned at current pointer position unsigned int xdr_read_pages struct xdr_stream * xdr unsigned int len xdr pointer to xdr_stream struct len number of bytes of page data Moves data beyond the current pointer position from the XDR head[] buffer into the page list. Any data that lies beyond current position + len bytes is moved into the XDR tail[]. Returns the number of XDR encoded bytes now contained in the pages Kernel Hackers Manual July 2017 xdr_enter_page 9 4.1.27 xdr_enter_page decode data from the XDR page void xdr_enter_page struct xdr_stream * xdr unsigned int len xdr pointer to xdr_stream struct len number of bytes of page data Moves data beyond the current pointer position from the XDR head[] buffer into the page list. Any data that lies beyond current position + len bytes is moved into the XDR tail[]. The current pointer is then repositioned at the beginning of the first XDR page. Kernel Hackers Manual July 2017 xdr_buf_subsegment 9 4.1.27 xdr_buf_subsegment set subbuf to a portion of buf int xdr_buf_subsegment struct xdr_buf * buf struct xdr_buf * subbuf unsigned int base unsigned int len buf an xdr buffer subbuf the result buffer base beginning of range in bytes len length of range in bytes sets subbuf to an xdr buffer representing the portion of buf of length len starting at offset base. buf and subbuf may be pointers to the same struct xdr_buf. Returns -1 if base of length are out of bounds. Kernel Hackers Manual July 2017 xdr_buf_trim 9 4.1.27 xdr_buf_trim lop at most len bytes off the end of buf void xdr_buf_trim struct xdr_buf * buf unsigned int len buf buf to be trimmed len number of bytes to reduce buf by Trim an xdr_buf by the given number of bytes by fixing up the lengths. Note that it's possible that we'll trim less than that amount if the xdr_buf is too small, or if (for instance) it's all in the head and the parser has already read too far into it. Kernel Hackers Manual July 2017 svc_print_addr 9 4.1.27 svc_print_addr Format rq_addr field for printing char * svc_print_addr struct svc_rqst * rqstp char * buf size_t len rqstp svc_rqst struct containing address to print buf target buffer for formatted address len length of target buffer Kernel Hackers Manual July 2017 svc_reserve 9 4.1.27 svc_reserve change the space reserved for the reply to a request. void svc_reserve struct svc_rqst * rqstp int space rqstp The request in question space new max space to reserve Each request reserves some space on the output queue of the transport to make sure the reply fits. This function reduces that reserved space to be the amount of space used already, plus space. Kernel Hackers Manual July 2017 svc_find_xprt 9 4.1.27 svc_find_xprt find an RPC transport instance struct svc_xprt * svc_find_xprt struct svc_serv * serv const char * xcl_name struct net * net const sa_family_t af const unsigned short port serv pointer to svc_serv to search xcl_name C string containing transport's class name net owner net pointer af Address family of transport's local address port transport's IP port number Return the transport instance pointer for the endpoint accepting connections/peer traffic from the specified transport class, address family and port. Specifying 0 for the address family or port is effectively a wild-card, and will result in matching the first transport in the service's list that has a matching class name. Kernel Hackers Manual July 2017 svc_xprt_names 9 4.1.27 svc_xprt_names format a buffer with a list of transport names int svc_xprt_names struct svc_serv * serv char * buf const int buflen serv pointer to an RPC service buf pointer to a buffer to be filled in buflen length of buffer to be filled in Fills in buf with a string containing a list of transport names, each name terminated with '\n'. Returns positive length of the filled-in string on success; otherwise a negative errno value is returned if an error occurs. Kernel Hackers Manual July 2017 xprt_register_transport 9 4.1.27 xprt_register_transport register a transport implementation int xprt_register_transport struct xprt_class * transport transport transport to register If a transport implementation is loaded as a kernel module, it can call this interface to make itself known to the RPC client. transport successfully registered -EEXIST: transport already registered -EINVAL: transport module being unloaded Kernel Hackers Manual July 2017 xprt_unregister_transport 9 4.1.27 xprt_unregister_transport unregister a transport implementation int xprt_unregister_transport struct xprt_class * transport transport transport to unregister transport successfully unregistered -ENOENT: transport never registered Kernel Hackers Manual July 2017 xprt_load_transport 9 4.1.27 xprt_load_transport load a transport implementation int xprt_load_transport const char * transport_name transport_name transport to load transport successfully loaded -ENOENT: transport module not available Kernel Hackers Manual July 2017 xprt_reserve_xprt 9 4.1.27 xprt_reserve_xprt serialize write access to transports int xprt_reserve_xprt struct rpc_xprt * xprt struct rpc_task * task xprt pointer to the target transport task task that is requesting access to the transport This prevents mixing the payload of separate requests, and prevents transport connects from colliding with writes. No congestion control is provided. Kernel Hackers Manual July 2017 xprt_release_xprt 9 4.1.27 xprt_release_xprt allow other requests to use a transport void xprt_release_xprt struct rpc_xprt * xprt struct rpc_task * task xprt transport with other tasks potentially waiting task task that is releasing access to the transport Note that task can be NULL. No congestion control is provided. Kernel Hackers Manual July 2017 xprt_release_xprt_cong 9 4.1.27 xprt_release_xprt_cong allow other requests to use a transport void xprt_release_xprt_cong struct rpc_xprt * xprt struct rpc_task * task xprt transport with other tasks potentially waiting task task that is releasing access to the transport Note that task can be NULL. Another task is awoken to use the transport if the transport's congestion window allows it. Kernel Hackers Manual July 2017 xprt_release_rqst_cong 9 4.1.27 xprt_release_rqst_cong housekeeping when request is complete void xprt_release_rqst_cong struct rpc_task * task task RPC request that recently completed Useful for transports that require congestion control. Kernel Hackers Manual July 2017 xprt_adjust_cwnd 9 4.1.27 xprt_adjust_cwnd adjust transport congestion window void xprt_adjust_cwnd struct rpc_xprt * xprt struct rpc_task * task int result xprt pointer to xprt task recently completed RPC request used to adjust window result result code of completed RPC request The transport code maintains an estimate on the maximum number of out- standing RPC requests, using a smoothed version of the congestion avoidance implemented in 44BSD. This is basically the Van Jacobson If a retransmit occurs, the congestion window is halved; otherwise, it is incremented by 1/cwnd when - a reply is received and - a full number of requests are outstanding and - the congestion window hasn't been updated recently. Kernel Hackers Manual July 2017 xprt_wake_pending_tasks 9 4.1.27 xprt_wake_pending_tasks wake all tasks on a transport's pending queue void xprt_wake_pending_tasks struct rpc_xprt * xprt int status xprt transport with waiting tasks status result code to plant in each task before waking it Kernel Hackers Manual July 2017 xprt_wait_for_buffer_space 9 4.1.27 xprt_wait_for_buffer_space wait for transport output buffer to clear void xprt_wait_for_buffer_space struct rpc_task * task rpc_action action task task to be put to sleep action function pointer to be executed after wait Note that we only set the timer for the case of RPC_IS_SOFT, since we don't in general want to force a socket disconnection due to an incomplete RPC call transmission. Kernel Hackers Manual July 2017 xprt_write_space 9 4.1.27 xprt_write_space wake the task waiting for transport output buffer space void xprt_write_space struct rpc_xprt * xprt xprt transport with waiting tasks Can be called in a soft IRQ context, so xprt_write_space never sleeps. Kernel Hackers Manual July 2017 xprt_set_retrans_timeout_def 9 4.1.27 xprt_set_retrans_timeout_def set a request's retransmit timeout void xprt_set_retrans_timeout_def struct rpc_task * task task task whose timeout is to be set Set a request's retransmit timeout based on the transport's default timeout parameters. Used by transports that don't adjust the retransmit timeout based on round-trip time estimation. Kernel Hackers Manual July 2017 xprt_set_retrans_timeout_rtt 9 4.1.27 xprt_set_retrans_timeout_rtt set a request's retransmit timeout void xprt_set_retrans_timeout_rtt struct rpc_task * task task task whose timeout is to be set Set a request's retransmit timeout using the RTT estimator. Kernel Hackers Manual July 2017 xprt_disconnect_done 9 4.1.27 xprt_disconnect_done mark a transport as disconnected void xprt_disconnect_done struct rpc_xprt * xprt xprt transport to flag for disconnect Kernel Hackers Manual July 2017 xprt_lookup_rqst 9 4.1.27 xprt_lookup_rqst find an RPC request corresponding to an XID struct rpc_rqst * xprt_lookup_rqst struct rpc_xprt * xprt __be32 xid xprt transport on which the original request was transmitted xid RPC XID of incoming reply Kernel Hackers Manual July 2017 xprt_complete_rqst 9 4.1.27 xprt_complete_rqst called when reply processing is complete void xprt_complete_rqst struct rpc_task * task int copied task RPC request that recently completed copied actual number of bytes received from the transport Caller holds transport lock. Kernel Hackers Manual July 2017 rpc_wake_up 9 4.1.27 rpc_wake_up wake up all rpc_tasks void rpc_wake_up struct rpc_wait_queue * queue queue rpc_wait_queue on which the tasks are sleeping Grabs queue->lock Kernel Hackers Manual July 2017 rpc_wake_up_status 9 4.1.27 rpc_wake_up_status wake up all rpc_tasks and set their status value. void rpc_wake_up_status struct rpc_wait_queue * queue int status queue rpc_wait_queue on which the tasks are sleeping status status value to set Grabs queue->lock Kernel Hackers Manual July 2017 rpc_malloc 9 4.1.27 rpc_malloc allocate an RPC buffer void * rpc_malloc struct rpc_task * task size_t size task RPC task that will use this buffer size requested byte size To prevent rpciod from hanging, this allocator never sleeps, returning NULL and suppressing warning if the request cannot be serviced immediately. The caller can arrange to sleep in a way that is safe for rpciod. Most requests are 'small' (under 2KiB) and can be serviced from a mempool, ensuring that NFS reads and writes can always proceed, and that there is good locality of reference for these buffers. In order to avoid memory starvation triggering more writebacks of NFS requests, we avoid using GFP_KERNEL. Kernel Hackers Manual July 2017 rpc_free 9 4.1.27 rpc_free free buffer allocated via rpc_malloc void rpc_free void * buffer buffer buffer to free Kernel Hackers Manual July 2017 xdr_skb_read_bits 9 4.1.27 xdr_skb_read_bits copy some data bits from skb to internal buffer size_t xdr_skb_read_bits struct xdr_skb_reader * desc void * to size_t len desc sk_buff copy helper to copy destination len number of bytes to copy Possibly called several times to iterate over an sk_buff and copy data out of it. Kernel Hackers Manual July 2017 xdr_partial_copy_from_skb 9 4.1.27 xdr_partial_copy_from_skb copy data out of an skb ssize_t xdr_partial_copy_from_skb struct xdr_buf * xdr unsigned int base struct xdr_skb_reader * desc xdr_skb_read_actor copy_actor xdr target XDR buffer base starting offset desc sk_buff copy helper copy_actor virtual method for copying data Kernel Hackers Manual July 2017 csum_partial_copy_to_xdr 9 4.1.27 csum_partial_copy_to_xdr checksum and copy data int csum_partial_copy_to_xdr struct xdr_buf * xdr struct sk_buff * skb xdr target XDR buffer skb source skb We have set things up such that we perform the checksum of the UDP packet in parallel with the copies into the RPC client iovec. -DaveM Kernel Hackers Manual July 2017 rpc_alloc_iostats 9 4.1.27 rpc_alloc_iostats allocate an rpc_iostats structure struct rpc_iostats * rpc_alloc_iostats struct rpc_clnt * clnt clnt RPC program, version, and xprt Kernel Hackers Manual July 2017 rpc_free_iostats 9 4.1.27 rpc_free_iostats release an rpc_iostats structure void rpc_free_iostats struct rpc_iostats * stats stats doomed rpc_iostats structure Kernel Hackers Manual July 2017 rpc_count_iostats_metrics 9 4.1.27 rpc_count_iostats_metrics tally up per-task stats void rpc_count_iostats_metrics const struct rpc_task * task struct rpc_iostats * op_metrics task completed rpc_task op_metrics stat structure for OP that will accumulate stats from task Kernel Hackers Manual July 2017 rpc_count_iostats 9 4.1.27 rpc_count_iostats tally up per-task stats void rpc_count_iostats const struct rpc_task * task struct rpc_iostats * stats task completed rpc_task stats array of stat structures Uses the statidx from task Kernel Hackers Manual July 2017 rpc_queue_upcall 9 4.1.27 rpc_queue_upcall queue an upcall message to userspace int rpc_queue_upcall struct rpc_pipe * pipe struct rpc_pipe_msg * msg pipe upcall pipe on which to queue given message msg message to queue Call with an inode created by rpc_mkpipe to queue an upcall. A userspace process may then later read the upcall by performing a read on an open file for this inode. It is up to the caller to initialize the fields of msg (other than msg->list) appropriately. Kernel Hackers Manual July 2017 rpc_mkpipe_dentry 9 4.1.27 rpc_mkpipe_dentry make an rpc_pipefs file for kernel<->userspace communication struct dentry * rpc_mkpipe_dentry struct dentry * parent const char * name void * private struct rpc_pipe * pipe parent dentry of directory to create new pipe in name name of pipe private private data to associate with the pipe, for the caller's use pipe rpc_pipe containing input parameters Data is made available for userspace to read by calls to rpc_queue_upcall. The actual reads will result in calls to ops->upcall, which will be called with the file pointer, message, and userspace buffer to copy to. Writes can come at any time, and do not necessarily have to be responses to upcalls. They will result in calls to msg->downcall. The private argument passed here will be available to all these methods from the file pointer, via RPC_I(file_inode(file))->private. Kernel Hackers Manual July 2017 rpc_unlink 9 4.1.27 rpc_unlink remove a pipe int rpc_unlink struct dentry * dentry dentry dentry for the pipe, as returned from rpc_mkpipe After this call, lookups will no longer find the pipe, and any attempts to read or write using preexisting opens of the pipe will return -EPIPE. Kernel Hackers Manual July 2017 rpc_init_pipe_dir_head 9 4.1.27 rpc_init_pipe_dir_head initialise a struct rpc_pipe_dir_head void rpc_init_pipe_dir_head struct rpc_pipe_dir_head * pdh pdh pointer to struct rpc_pipe_dir_head Kernel Hackers Manual July 2017 rpc_init_pipe_dir_object 9 4.1.27 rpc_init_pipe_dir_object initialise a struct rpc_pipe_dir_object void rpc_init_pipe_dir_object struct rpc_pipe_dir_object * pdo const struct rpc_pipe_dir_object_ops * pdo_ops void * pdo_data pdo pointer to struct rpc_pipe_dir_object pdo_ops pointer to const struct rpc_pipe_dir_object_ops pdo_data pointer to caller-defined data Kernel Hackers Manual July 2017 rpc_add_pipe_dir_object 9 4.1.27 rpc_add_pipe_dir_object associate a rpc_pipe_dir_object to a directory int rpc_add_pipe_dir_object struct net * net struct rpc_pipe_dir_head * pdh struct rpc_pipe_dir_object * pdo net pointer to struct net pdh pointer to struct rpc_pipe_dir_head pdo pointer to struct rpc_pipe_dir_object Kernel Hackers Manual July 2017 rpc_remove_pipe_dir_object 9 4.1.27 rpc_remove_pipe_dir_object remove a rpc_pipe_dir_object from a directory void rpc_remove_pipe_dir_object struct net * net struct rpc_pipe_dir_head * pdh struct rpc_pipe_dir_object * pdo net pointer to struct net pdh pointer to struct rpc_pipe_dir_head pdo pointer to struct rpc_pipe_dir_object Kernel Hackers Manual July 2017 rpc_find_or_alloc_pipe_dir_object 9 4.1.27 rpc_find_or_alloc_pipe_dir_object struct rpc_pipe_dir_object * rpc_find_or_alloc_pipe_dir_object struct net * net struct rpc_pipe_dir_head * pdh int (*match) struct rpc_pipe_dir_object *, void * struct rpc_pipe_dir_object *(*alloc) void * void * data net pointer to struct net pdh pointer to struct rpc_pipe_dir_head match match struct rpc_pipe_dir_object to data alloc allocate a new struct rpc_pipe_dir_object data user defined data for match and alloc Kernel Hackers Manual July 2017 rpcb_getport_async 9 4.1.27 rpcb_getport_async obtain the port for a given RPC service on a given host void rpcb_getport_async struct rpc_task * task task task that is waiting for portmapper request This one can be called for an ongoing RPC request, and can be used in an async (rpciod) context. Kernel Hackers Manual July 2017 rpc_create 9 4.1.27 rpc_create create an RPC client and transport with one call struct rpc_clnt * rpc_create struct rpc_create_args * args args rpc_clnt create argument structure Creates and initializes an RPC transport and an RPC client. It can ping the server in order to determine if it is up, and to see if it supports this program and version. RPC_CLNT_CREATE_NOPING disables this behavior so asynchronous tasks can also use rpc_create. Kernel Hackers Manual July 2017 rpc_clone_client 9 4.1.27 rpc_clone_client Clone an RPC client structure struct rpc_clnt * rpc_clone_client struct rpc_clnt * clnt clnt RPC client whose parameters are copied Returns a fresh RPC client or an ERR_PTR. Kernel Hackers Manual July 2017 rpc_clone_client_set_auth 9 4.1.27 rpc_clone_client_set_auth Clone an RPC client structure and set its auth struct rpc_clnt * rpc_clone_client_set_auth struct rpc_clnt * clnt rpc_authflavor_t flavor clnt RPC client whose parameters are copied flavor security flavor for new client Returns a fresh RPC client or an ERR_PTR. Kernel Hackers Manual July 2017 rpc_switch_client_transport 9 4.1.27 rpc_switch_client_transport int rpc_switch_client_transport struct rpc_clnt * clnt struct xprt_create * args const struct rpc_timeout * timeout clnt pointer to a struct rpc_clnt args pointer to the new transport arguments timeout pointer to the new timeout parameters This function allows the caller to switch the RPC transport for the rpc_clnt structure 'clnt' to allow it to connect to a mirrored NFS server, for instance. It assumes that the caller has ensured that there are no active RPC tasks by using some form of locking. Returns zero if clnt is now using the new xprt. Otherwise a negative errno is returned, and clnt continues to use the old xprt. Kernel Hackers Manual July 2017 rpc_bind_new_program 9 4.1.27 rpc_bind_new_program bind a new RPC program to an existing client struct rpc_clnt * rpc_bind_new_program struct rpc_clnt * old const struct rpc_program * program u32 vers old old rpc_client program rpc program to set vers rpc program version Clones the rpc client and sets up a new RPC program. This is mainly of use for enabling different RPC programs to share the same transport. The Sun NFSv2/v3 ACL protocol can do this. Kernel Hackers Manual July 2017 rpc_run_task 9 4.1.27 rpc_run_task Allocate a new RPC task, then run rpc_execute against it struct rpc_task * rpc_run_task const struct rpc_task_setup * task_setup_data task_setup_data pointer to task initialisation data Kernel Hackers Manual July 2017 rpc_call_sync 9 4.1.27 rpc_call_sync Perform a synchronous RPC call int rpc_call_sync struct rpc_clnt * clnt const struct rpc_message * msg int flags clnt pointer to RPC client msg RPC call parameters flags RPC call flags Kernel Hackers Manual July 2017 rpc_call_async 9 4.1.27 rpc_call_async Perform an asynchronous RPC call int rpc_call_async struct rpc_clnt * clnt const struct rpc_message * msg int flags const struct rpc_call_ops * tk_ops void * data clnt pointer to RPC client msg RPC call parameters flags RPC call flags tk_ops RPC call ops data user call data Kernel Hackers Manual July 2017 rpc_peeraddr 9 4.1.27 rpc_peeraddr extract remote peer address from clnt's xprt size_t rpc_peeraddr struct rpc_clnt * clnt struct sockaddr * buf size_t bufsize clnt RPC client structure buf target buffer bufsize length of target buffer Returns the number of bytes that are actually in the stored address. Kernel Hackers Manual July 2017 rpc_peeraddr2str 9 4.1.27 rpc_peeraddr2str return remote peer address in printable format const char * rpc_peeraddr2str struct rpc_clnt * clnt enum rpc_display_format_t format clnt RPC client structure format address format the lifetime of the memory referenced by the returned pointer is the same as the rpc_xprt itself. As long as the caller uses this pointer, it must hold the RCU read lock. Kernel Hackers Manual July 2017 rpc_localaddr 9 4.1.27 rpc_localaddr discover local endpoint address for an RPC client int rpc_localaddr struct rpc_clnt * clnt struct sockaddr * buf size_t buflen clnt RPC client structure buf target buffer buflen size of target buffer, in bytes Returns zero and fills in buf and buflen if successful; otherwise, a negative errno is returned. This works even if the underlying transport is not currently connected, or if the upper layer never previously provided a source address. multiple calls in succession may give different results, depending on how local networking configuration changes over time. Kernel Hackers Manual July 2017 rpc_protocol 9 4.1.27 rpc_protocol Get transport protocol number for an RPC client int rpc_protocol struct rpc_clnt * clnt clnt RPC client to query Kernel Hackers Manual July 2017 rpc_net_ns 9 4.1.27 rpc_net_ns Get the network namespace for this RPC client struct net * rpc_net_ns struct rpc_clnt * clnt clnt RPC client to query Kernel Hackers Manual July 2017 rpc_max_payload 9 4.1.27 rpc_max_payload Get maximum payload size for a transport, in bytes size_t rpc_max_payload struct rpc_clnt * clnt clnt RPC client to query For stream transports, this is one RPC record fragment (see RFC 1831), as we don't support multi-record requests yet. For datagram transports, this is the size of an IP packet minus the IP, UDP, and RPC header sizes. Kernel Hackers Manual July 2017 rpc_get_timeout 9 4.1.27 rpc_get_timeout Get timeout for transport in units of HZ unsigned long rpc_get_timeout struct rpc_clnt * clnt clnt RPC client to query Kernel Hackers Manual July 2017 rpc_force_rebind 9 4.1.27 rpc_force_rebind force transport to check that remote port is unchanged void rpc_force_rebind struct rpc_clnt * clnt clnt client to rebind Kernel Hackers Manual July 2017 wimax_msg_alloc 9 4.1.27 wimax_msg_alloc Create a new skb for sending a message to userspace struct sk_buff * wimax_msg_alloc struct wimax_dev * wimax_dev const char * pipe_name const void * msg size_t size gfp_t gfp_flags wimax_dev WiMAX device descriptor pipe_name "named pipe" the message will be sent to msg pointer to the message data to send size size of the message to send (in bytes), including the header. gfp_flags flags for memory allocation. 0 if ok, negative errno code on error Allocates an skb that will contain the message to send to user space over the messaging pipe and initializes it, copying the payload. Once this call is done, you can deliver it with wimax_msg_send. Don't use skb_push/skb_pull/skb_reserve on the skb, as wimax_msg_send depends on skb->data being placed at the beginning of the user message. Unlike other WiMAX stack calls, this call can be used way early, even before wimax_dev_add is called, as long as the wimax_dev->net_dev pointer is set to point to a proper net_dev. This is so that drivers can use it early in case they need to send stuff around or communicate with user space. Kernel Hackers Manual July 2017 wimax_msg_data_len 9 4.1.27 wimax_msg_data_len Return a pointer and size of a message's payload const void * wimax_msg_data_len struct sk_buff * msg size_t * size msg Pointer to a message created with wimax_msg_alloc size Pointer to where to store the message's size Returns the pointer to the message data. Kernel Hackers Manual July 2017 wimax_msg_data 9 4.1.27 wimax_msg_data Return a pointer to a message's payload const void * wimax_msg_data struct sk_buff * msg msg Pointer to a message created with wimax_msg_alloc Kernel Hackers Manual July 2017 wimax_msg_len 9 4.1.27 wimax_msg_len Return a message's payload length ssize_t wimax_msg_len struct sk_buff * msg msg Pointer to a message created with wimax_msg_alloc Kernel Hackers Manual July 2017 wimax_msg_send 9 4.1.27 wimax_msg_send Send a pre-allocated message to user space int wimax_msg_send struct wimax_dev * wimax_dev struct sk_buff * skb wimax_dev WiMAX device descriptor skb struct sk_buff returned by wimax_msg_alloc. Note the ownership of skb is transferred to this function. 0 if ok, < 0 errno code on error Sends a free-form message that was preallocated with wimax_msg_alloc and filled up. Assumes that once you pass an skb to this function for sending, it owns it and will release it when done (on success). Don't use skb_push/skb_pull/skb_reserve on the skb, as wimax_msg_send depends on skb->data being placed at the beginning of the user message. Unlike other WiMAX stack calls, this call can be used way early, even before wimax_dev_add is called, as long as the wimax_dev->net_dev pointer is set to point to a proper net_dev. This is so that drivers can use it early in case they need to send stuff around or communicate with user space. Kernel Hackers Manual July 2017 wimax_msg 9 4.1.27 wimax_msg Send a message to user space int wimax_msg struct wimax_dev * wimax_dev const char * pipe_name const void * buf size_t size gfp_t gfp_flags wimax_dev WiMAX device descriptor (properly referenced) pipe_name "named pipe" the message will be sent to buf pointer to the message to send. size size of the buffer pointed to by buf (in bytes). gfp_flags flags for memory allocation. 0 if ok, negative errno code on error. Sends a free-form message to user space on the device wimax_dev. Once the skb is given to this function, who will own it and will release it when done (unless it returns error). Kernel Hackers Manual July 2017 wimax_reset 9 4.1.27 wimax_reset Reset a WiMAX device int wimax_reset struct wimax_dev * wimax_dev wimax_dev WiMAX device descriptor 0 if ok and a warm reset was done (the device still exists in the system). -ENODEV if a cold/bus reset had to be done (device has disconnected and reconnected, so current handle is not valid any more). -EINVAL if the device is not even registered. Any other negative error code shall be considered as non-recoverable. Called when wanting to reset the device for any reason. Device is taken back to power on status. This call blocks; on successful return, the device has completed the reset process and is ready to operate. Kernel Hackers Manual July 2017 wimax_report_rfkill_hw 9 4.1.27 wimax_report_rfkill_hw Reports changes in the hardware RF switch void wimax_report_rfkill_hw struct wimax_dev * wimax_dev enum wimax_rf_state state wimax_dev WiMAX device descriptor state New state of the RF Kill switch. WIMAX_RF_ON radio on, WIMAX_RF_OFF radio off. When the device detects a change in the state of thehardware RF switch, it must call this function to let the WiMAX kernel stack know that the state has changed so it can be properly propagated. The WiMAX stack caches the state (the driver doesn't need to). As well, as the change is propagated it will come back as a request to change the software state to mirror the hardware state. If the device doesn't have a hardware kill switch, just report it on initialization as always on (WIMAX_RF_ON, radio on). Kernel Hackers Manual July 2017 wimax_report_rfkill_sw 9 4.1.27 wimax_report_rfkill_sw Reports changes in the software RF switch void wimax_report_rfkill_sw struct wimax_dev * wimax_dev enum wimax_rf_state state wimax_dev WiMAX device descriptor state New state of the RF kill switch. WIMAX_RF_ON radio on, WIMAX_RF_OFF radio off. Reports changes in the software RF switch state to the the WiMAX stack. The main use is during initialization, so the driver can query the device for its current software radio kill switch state and feed it to the system. On the side, the device does not change the software state by itself. In practice, this can happen, as the device might decide to switch (in software) the radio off for different reasons. Kernel Hackers Manual July 2017 wimax_rfkill 9 4.1.27 wimax_rfkill Set the software RF switch state for a WiMAX device int wimax_rfkill struct wimax_dev * wimax_dev enum wimax_rf_state state wimax_dev WiMAX device descriptor state New RF state. >= 0 toggle state if ok, < 0 errno code on error. The toggle state is returned as a bitmap, bit 0 being the hardware RF state, bit 1 the software RF state. 0 means disabled (WIMAX_RF_ON, radio on), 1 means enabled radio off (WIMAX_RF_OFF). Called by the user when he wants to request the WiMAX radio to be switched on (WIMAX_RF_ON) or off (WIMAX_RF_OFF). With WIMAX_RF_QUERY, just the current state is returned. This call will block until the operation is complete. Kernel Hackers Manual July 2017 wimax_state_change 9 4.1.27 wimax_state_change Set the current state of a WiMAX device void wimax_state_change struct wimax_dev * wimax_dev enum wimax_st new_state wimax_dev WiMAX device descriptor (properly referenced) new_state New state to switch to This implements the state changes for the wimax devices. It will - verify that the state transition is legal (for now it'll just print a warning if not) according to the table in linux/wimax.h's documentation for 'enum wimax_st'. - perform the actions needed for leaving the current state and whichever are needed for entering the new state. - issue a report to user space indicating the new state (and an optional payload with information about the new state). wimax_dev must be locked Kernel Hackers Manual July 2017 wimax_state_get 9 4.1.27 wimax_state_get Return the current state of a WiMAX device enum wimax_st wimax_state_get struct wimax_dev * wimax_dev wimax_dev WiMAX device descriptor Current state of the device according to its driver. Kernel Hackers Manual July 2017 wimax_dev_init 9 4.1.27 wimax_dev_init initialize a newly allocated instance void wimax_dev_init struct wimax_dev * wimax_dev wimax_dev WiMAX device descriptor to initialize. Initializes fields of a freshly allocated wimax_dev instance. This function assumes that after allocation, the memory occupied by wimax_dev was zeroed. Kernel Hackers Manual July 2017 wimax_dev_add 9 4.1.27 wimax_dev_add Register a new WiMAX device int wimax_dev_add struct wimax_dev * wimax_dev struct net_device * net_dev wimax_dev WiMAX device descriptor (as embedded in your net_dev's priv data). You must have called wimax_dev_init on it before. net_dev net device the wimax_dev is associated with. The function expects SET_NETDEV_DEV and register_netdev were already called on it. Registers the new WiMAX device, sets up the user-kernel control interface (generic netlink) and common WiMAX infrastructure. Note that the parts that will allow interaction with user space are setup at the very end, when the rest is in place, as once that happens, the driver might get user space control requests via netlink or from debugfs that might translate into calls into wimax_dev->op_*(). Kernel Hackers Manual July 2017 wimax_dev_rm 9 4.1.27 wimax_dev_rm Unregister an existing WiMAX device void wimax_dev_rm struct wimax_dev * wimax_dev wimax_dev WiMAX device descriptor Unregisters a WiMAX device previously registered for use with wimax_add_rm. IMPORTANT! Must call before calling unregister_netdev. After this function returns, you will not get any more user space control requests (via netlink or debugfs) and thus to wimax_dev->ops. Reentrancy control is ensured by setting the state to __WIMAX_ST_QUIESCING. rfkill operations coming through wimax_*rfkill*() will be stopped by the quiescing state; ops coming from the rfkill subsystem will be stopped by the support being removed by wimax_rfkill_rm. Kernel Hackers Manual July 2017 struct wimax_dev 9 4.1.27 struct wimax_dev Generic WiMAX device struct wimax_dev { struct net_device * net_dev; struct list_head id_table_node; struct mutex mutex; struct mutex mutex_reset; enum wimax_st state; int (* op_msg_from_user) (struct wimax_dev *wimax_dev,const char *,const void *, size_t,const struct genl_info *info); int (* op_rfkill_sw_toggle) (struct wimax_dev *wimax_dev,enum wimax_rf_state); int (* op_reset) (struct wimax_dev *wimax_dev); struct rfkill * rfkill; unsigned int rf_hw; unsigned int rf_sw; char name[32]; struct dentry * debugfs_dentry; }; net_dev [fill] Pointer to the struct net_device this WiMAX device implements. id_table_node [private] link to the list of wimax devices kept by id-table.c. Protected by it's own spinlock. mutex [private] Serializes all concurrent access and execution of operations. mutex_reset [private] Serializes reset operations. Needs to be a different mutex because as part of the reset operation, the driver has to call back into the stack to do things such as state change, that require wimax_dev->mutex. state [private] Current state of the WiMAX device. op_msg_from_user [fill] Driver-specific operation to handle a raw message from user space to the driver. The driver can send messages to user space using with wimax_msg_to_user. op_rfkill_sw_toggle [fill] Driver-specific operation to act on userspace (or any other agent) requesting the WiMAX device to change the RF Kill software switch (WIMAX_RF_ON or WIMAX_RF_OFF). If such hardware support is not present, it is assumed the radio cannot be switched off and it is always on (and the stack will error out when trying to switch it off). In such case, this function pointer can be left as NULL. op_reset [fill] Driver specific operation to reset the device. This operation should always attempt first a warm reset that does not disconnect the device from the bus and return 0. If that fails, it should resort to some sort of cold or bus reset (even if it implies a bus disconnection and device disappearance). In that case, -ENODEV should be returned to indicate the device is gone. This operation has to be synchronous, and return only when the reset is complete. In case of having had to resort to bus/cold reset implying a device disconnection, the call is allowed to return immediately. rfkill [private] integration into the RF-Kill infrastructure. rf_hw [private] State of the hardware radio switch (OFF/ON) rf_sw [private] State of the software radio switch (OFF/ON) name[32] [fill] A way to identify this device. We need to register a name with many subsystems (rfkill, workqueue creation, etc). We can't use the network device name as that might change and in some instances we don't know it yet (until we don't call register_netdev). So we generate an unique one using the driver name and device bus id, place it here and use it across the board. Recommended naming: DRIVERNAME-BUSNAME:BUSID (dev->bus->name, dev->bus_id). debugfs_dentry [private] Used to hook up a debugfs entry. This shows up in the debugfs root as wimax\:DEVICENAME. wimax_dev->mutex is NOT locked when this op is being called; however, wimax_dev->mutex_reset IS locked to ensure serialization of calls to wimax_reset. See wimax_reset's documentation. This structure defines a common interface to access all WiMAX devices from different vendors and provides a common API as well as a free-form device-specific messaging channel. 1. Embed a struct wimax_dev at *the beginning* the network device structure so that netdev_priv points to it. 2. memset it to zero 3. Initialize with wimax_dev_init. This will leave the WiMAX device in the __WIMAX_ST_NULL state. 4. Fill all the fields marked with [fill]; once called wimax_dev_add, those fields CANNOT be modified. 5. Call wimax_dev_add *after* registering the network device. This will leave the WiMAX device in the WIMAX_ST_DOWN state. Protect the driver's net_device->open against succeeding if the wimax device state is lower than WIMAX_ST_DOWN. 6. Select when the device is going to be turned on/initialized; for example, it could be initialized on 'ifconfig up' (when the netdev op 'open' is called on the driver). When the device is initialized (at `ifconfig up` time, or right after calling wimax_dev_add from _probe, make sure the following steps are taken a. Move the device to WIMAX_ST_UNINITIALIZED. This is needed so some API calls that shouldn't work until the device is ready can be blocked. b. Initialize the device. Make sure to turn the SW radio switch off and move the device to state WIMAX_ST_RADIO_OFF when done. When just initialized, a device should be left in RADIO OFF state until user space devices to turn it on. c. Query the device for the state of the hardware rfkill switch and call wimax_rfkill_report_hw and wimax_rfkill_report_sw as needed. See below. wimax_dev_rm undoes before unregistering the network device. Once wimax_dev_add is called, the driver can get called on the wimax_dev->op_* function pointers The stack provides a mutex for each device that will disallow API calls happening concurrently; thus, op calls into the driver through the wimax_dev->op*() function pointers will always be serialized and *never* concurrent. For locking, take wimax_dev->mutex is taken; (most) operations in the API have to check for wimax_dev_is_ready to return 0 before continuing (this is done internally). The WiMAX device is reference counted by the associated network device. The only operation that can be used to reference the device is wimax_dev_get_by_genl_info, and the reference it acquires has to be released with dev_put(wimax_dev->net_dev). At startup, both HW and SW radio switchess are assumed to be off. At initialization time [after calling wimax_dev_add], have the driver query the device for the status of the software and hardware RF kill switches and call wimax_report_rfkill_hw and wimax_rfkill_report_sw to indicate their state. If any is missing, just call it to indicate it is ON (radio always on). Whenever the driver detects a change in the state of the RF kill switches, it should call wimax_report_rfkill_hw or wimax_report_rfkill_sw to report it to the stack. Kernel Hackers Manual July 2017 enum wimax_st 9 4.1.27 enum wimax_st The different states of a WiMAX device enum wimax_st { __WIMAX_ST_NULL, WIMAX_ST_DOWN, __WIMAX_ST_QUIESCING, WIMAX_ST_UNINITIALIZED, WIMAX_ST_RADIO_OFF, WIMAX_ST_READY, WIMAX_ST_SCANNING, WIMAX_ST_CONNECTING, WIMAX_ST_CONNECTED, __WIMAX_ST_INVALID }; __WIMAX_ST_NULL The device structure has been allocated and zeroed, but still wimax_dev_add hasn't been called. There is no state. WIMAX_ST_DOWN The device has been registered with the WiMAX and networking stacks, but it is not initialized (normally that is done with 'ifconfig DEV up' [or equivalent], which can upload firmware and enable communications with the device). In this state, the device is powered down and using as less power as possible. This state is the default after a call to wimax_dev_add. It is ok to have drivers move directly to WIMAX_ST_UNINITIALIZED or WIMAX_ST_RADIO_OFF in _probe after the call to wimax_dev_add. It is recommended that the driver leaves this state when calling 'ifconfig DEV up' and enters it back on 'ifconfig DEV down'. __WIMAX_ST_QUIESCING The device is being torn down, so no API operations are allowed to proceed except the ones needed to complete the device clean up process. WIMAX_ST_UNINITIALIZED [optional] Communication with the device is setup, but the device still requires some configuration before being operational. Some WiMAX API calls might work. WIMAX_ST_RADIO_OFF The device is fully up; radio is off (wether by hardware or software switches). It is recommended to always leave the device in this state after initialization. WIMAX_ST_READY The device is fully up and radio is on. WIMAX_ST_SCANNING [optional] The device has been instructed to scan. In this state, the device cannot be actively connected to a network. WIMAX_ST_CONNECTING The device is connecting to a network. This state exists because in some devices, the connect process can include a number of negotiations between user space, kernel space and the device. User space needs to know what the device is doing. If the connect sequence in a device is atomic and fast, the device can transition directly to CONNECTED WIMAX_ST_CONNECTED The device is connected to a network. __WIMAX_ST_INVALID This is an invalid state used to mark the maximum numeric value of states. Transitions from one state to another one are atomic and can only be caused in kernel space with wimax_state_change. To read the state, use wimax_state_get. States starting with __ are internal and shall not be used or referred to by drivers or userspace. They look ugly, but that's the point -- if any use is made non-internal to the stack, it is easier to catch on review. All API operations [with well defined exceptions] will take the device mutex before starting and then check the state. If the state is __WIMAX_ST_NULL, WIMAX_ST_DOWN, WIMAX_ST_UNINITIALIZED or __WIMAX_ST_QUIESCING, it will drop the lock and quit with -EINVAL, -ENOMEDIUM, -ENOTCONN or -ESHUTDOWN. The order of the definitions is important, so we can do numerical comparisons (eg: < WIMAX_ST_RADIO_OFF means the device is not ready to operate). Kernel Hackers Manual July 2017 dev_add_pack 9 4.1.27 dev_add_pack add packet handler void dev_add_pack struct packet_type * pt pt packet type declaration Add a protocol handler to the networking stack. The passed packet_type is linked into kernel lists and may not be freed until it has been removed from the kernel lists. This call does not sleep therefore it can not guarantee all CPU's that are in middle of receiving packets will see the new packet type (until the next received packet). Kernel Hackers Manual July 2017 __dev_remove_pack 9 4.1.27 __dev_remove_pack remove packet handler void __dev_remove_pack struct packet_type * pt pt packet type declaration Remove a protocol handler that was previously added to the kernel protocol handlers by dev_add_pack. The passed packet_type is removed from the kernel lists and can be freed or reused once this function returns. The packet type might still be in use by receivers and must not be freed until after all the CPU's have gone through a quiescent state. Kernel Hackers Manual July 2017 dev_remove_pack 9 4.1.27 dev_remove_pack remove packet handler void dev_remove_pack struct packet_type * pt pt packet type declaration Remove a protocol handler that was previously added to the kernel protocol handlers by dev_add_pack. The passed packet_type is removed from the kernel lists and can be freed or reused once this function returns. This call sleeps to guarantee that no CPU is looking at the packet type after return. Kernel Hackers Manual July 2017 dev_add_offload 9 4.1.27 dev_add_offload register offload handlers void dev_add_offload struct packet_offload * po po protocol offload declaration Add protocol offload handlers to the networking stack. The passed proto_offload is linked into kernel lists and may not be freed until it has been removed from the kernel lists. This call does not sleep therefore it can not guarantee all CPU's that are in middle of receiving packets will see the new offload handlers (until the next received packet). Kernel Hackers Manual July 2017 dev_remove_offload 9 4.1.27 dev_remove_offload remove packet offload handler void dev_remove_offload struct packet_offload * po po packet offload declaration Remove a packet offload handler that was previously added to the kernel offload handlers by dev_add_offload. The passed offload_type is removed from the kernel lists and can be freed or reused once this function returns. This call sleeps to guarantee that no CPU is looking at the packet type after return. Kernel Hackers Manual July 2017 netdev_boot_setup_check 9 4.1.27 netdev_boot_setup_check check boot time settings int netdev_boot_setup_check struct net_device * dev dev the netdevice Check boot time settings for the device. The found settings are set for the device to be used later in the device probing. Returns 0 if no settings found, 1 if they are. Kernel Hackers Manual July 2017 dev_get_iflink 9 4.1.27 dev_get_iflink get 'iflink' value of a interface int dev_get_iflink const struct net_device * dev dev targeted interface Indicates the ifindex the interface is linked to. Physical interfaces have the same 'ifindex' and 'iflink' values. Kernel Hackers Manual July 2017 __dev_get_by_name 9 4.1.27 __dev_get_by_name find a device by its name struct net_device * __dev_get_by_name struct net * net const char * name net the applicable net namespace name name to find Find an interface by name. Must be called under RTNL semaphore or dev_base_lock. If the name is found a pointer to the device is returned. If the name is not found then NULL is returned. The reference counters are not incremented so the caller must be careful with locks. Kernel Hackers Manual July 2017 dev_get_by_name_rcu 9 4.1.27 dev_get_by_name_rcu find a device by its name struct net_device * dev_get_by_name_rcu struct net * net const char * name net the applicable net namespace name name to find Find an interface by name. If the name is found a pointer to the device is returned. If the name is not found then NULL is returned. The reference counters are not incremented so the caller must be careful with locks. The caller must hold RCU lock. Kernel Hackers Manual July 2017 dev_get_by_name 9 4.1.27 dev_get_by_name find a device by its name struct net_device * dev_get_by_name struct net * net const char * name net the applicable net namespace name name to find Find an interface by name. This can be called from any context and does its own locking. The returned handle has the usage count incremented and the caller must use dev_put to release it when it is no longer needed. NULL is returned if no matching device is found. Kernel Hackers Manual July 2017 __dev_get_by_index 9 4.1.27 __dev_get_by_index find a device by its ifindex struct net_device * __dev_get_by_index struct net * net int ifindex net the applicable net namespace ifindex index of device Search for an interface by index. Returns NULL if the device is not found or a pointer to the device. The device has not had its reference counter increased so the caller must be careful about locking. The caller must hold either the RTNL semaphore or dev_base_lock. Kernel Hackers Manual July 2017 dev_get_by_index_rcu 9 4.1.27 dev_get_by_index_rcu find a device by its ifindex struct net_device * dev_get_by_index_rcu struct net * net int ifindex net the applicable net namespace ifindex index of device Search for an interface by index. Returns NULL if the device is not found or a pointer to the device. The device has not had its reference counter increased so the caller must be careful about locking. The caller must hold RCU lock. Kernel Hackers Manual July 2017 dev_get_by_index 9 4.1.27 dev_get_by_index find a device by its ifindex struct net_device * dev_get_by_index struct net * net int ifindex net the applicable net namespace ifindex index of device Search for an interface by index. Returns NULL if the device is not found or a pointer to the device. The device returned has had a reference added and the pointer is safe until the user calls dev_put to indicate they have finished with it. Kernel Hackers Manual July 2017 dev_getbyhwaddr_rcu 9 4.1.27 dev_getbyhwaddr_rcu find a device by its hardware address struct net_device * dev_getbyhwaddr_rcu struct net * net unsigned short type const char * ha net the applicable net namespace type media type of device ha hardware address Search for an interface by MAC address. Returns NULL if the device is not found or a pointer to the device. The caller must hold RCU or RTNL. The returned device has not had its ref count increased and the caller must therefore be careful about locking Kernel Hackers Manual July 2017 __dev_get_by_flags 9 4.1.27 __dev_get_by_flags find any device with given flags struct net_device * __dev_get_by_flags struct net * net unsigned short if_flags unsigned short mask net the applicable net namespace if_flags IFF_* values mask bitmask of bits in if_flags to check Search for any interface with the given flags. Returns NULL if a device is not found or a pointer to the device. Must be called inside rtnl_lock, and result refcount is unchanged. Kernel Hackers Manual July 2017 dev_valid_name 9 4.1.27 dev_valid_name check if name is okay for network device bool dev_valid_name const char * name name name string Network device names need to be valid file names to to allow sysfs to work. We also disallow any kind of whitespace. Kernel Hackers Manual July 2017 dev_alloc_name 9 4.1.27 dev_alloc_name allocate a name for a device int dev_alloc_name struct net_device * dev const char * name dev device name name format string Passed a format string - eg ltd it will try and find a suitable id. It scans list of devices to build up a free map, then chooses the first empty slot. The caller must hold the dev_base or rtnl lock while allocating the name and adding the device in order to avoid duplicates. Limited to bits_per_byte * page size devices (ie 32K on most platforms). Returns the number of the unit assigned or a negative errno code. Kernel Hackers Manual July 2017 netdev_features_change 9 4.1.27 netdev_features_change device changes features void netdev_features_change struct net_device * dev dev device to cause notification Called to indicate a device has changed features. Kernel Hackers Manual July 2017 netdev_state_change 9 4.1.27 netdev_state_change device changes state void netdev_state_change struct net_device * dev dev device to cause notification Called to indicate a device has changed state. This function calls the notifier chains for netdev_chain and sends a NEWLINK message to the routing socket. Kernel Hackers Manual July 2017 netdev_notify_peers 9 4.1.27 netdev_notify_peers notify network peers about existence of dev void netdev_notify_peers struct net_device * dev dev network device Generate traffic such that interested network peers are aware of dev, such as by generating a gratuitous ARP. This may be used when a device wants to inform the rest of the network about some sort of reconfiguration such as a failover event or virtual machine migration. Kernel Hackers Manual July 2017 dev_open 9 4.1.27 dev_open prepare an interface for use. int dev_open struct net_device * dev dev device to open Takes a device from down to up state. The device's private open function is invoked and then the multicast lists are loaded. Finally the device is moved into the up state and a NETDEV_UP message is sent to the netdev notifier chain. Calling this function on an active interface is a nop. On a failure a negative errno code is returned. Kernel Hackers Manual July 2017 dev_close 9 4.1.27 dev_close shutdown an interface. int dev_close struct net_device * dev dev device to shutdown This function moves an active device into down state. A NETDEV_GOING_DOWN is sent to the netdev notifier chain. The device is then deactivated and finally a NETDEV_DOWN is sent to the notifier chain. Kernel Hackers Manual July 2017 dev_disable_lro 9 4.1.27 dev_disable_lro disable Large Receive Offload on a device void dev_disable_lro struct net_device * dev dev device Disable Large Receive Offload (LRO) on a net device. Must be called under RTNL. This is needed if received packets may be forwarded to another interface. Kernel Hackers Manual July 2017 register_netdevice_notifier 9 4.1.27 register_netdevice_notifier register a network notifier block int register_netdevice_notifier struct notifier_block * nb nb notifier Register a notifier to be called when network device events occur. The notifier passed is linked into the kernel structures and must not be reused until it has been unregistered. A negative errno code is returned on a failure. When registered all registration and up events are replayed to the new notifier to allow device to have a race free view of the network device list. Kernel Hackers Manual July 2017 unregister_netdevice_notifier 9 4.1.27 unregister_netdevice_notifier unregister a network notifier block int unregister_netdevice_notifier struct notifier_block * nb nb notifier Unregister a notifier previously registered by register_netdevice_notifier. The notifier is unlinked into the kernel structures and may then be reused. A negative errno code is returned on a failure. After unregistering unregister and down device events are synthesized for all devices on the device list to the removed notifier to remove the need for special case cleanup code. Kernel Hackers Manual July 2017 call_netdevice_notifiers 9 4.1.27 call_netdevice_notifiers call all network notifier blocks int call_netdevice_notifiers unsigned long val struct net_device * dev val value passed unmodified to notifier function dev net_device pointer passed unmodified to notifier function Call all network notifier blocks. Parameters and return value are as for raw_notifier_call_chain. Kernel Hackers Manual July 2017 dev_forward_skb 9 4.1.27 dev_forward_skb loopback an skb to another netif int dev_forward_skb struct net_device * dev struct sk_buff * skb dev destination network device skb buffer to forward NET_RX_SUCCESS (no congestion) NET_RX_DROP (packet was dropped, but freed) dev_forward_skb can be used for injecting an skb from the start_xmit function of one device into the receive queue of another device. The receiving device may be in another namespace, so we have to clear all information in the skb that could impact namespace isolation. Kernel Hackers Manual July 2017 netif_set_real_num_rx_queues 9 4.1.27 netif_set_real_num_rx_queues set actual number of RX queues used int netif_set_real_num_rx_queues struct net_device * dev unsigned int rxq dev Network device rxq Actual number of RX queues This must be called either with the rtnl_lock held or before registration of the net device. Returns 0 on success, or a negative error code. If called before registration, it always succeeds. Kernel Hackers Manual July 2017 netif_get_num_default_rss_queues 9 4.1.27 netif_get_num_default_rss_queues default number of RSS queues int netif_get_num_default_rss_queues void void no arguments This routine should set an upper limit on the number of RSS queues used by default by multiqueue devices. Kernel Hackers Manual July 2017 netif_wake_subqueue 9 4.1.27 netif_wake_subqueue allow sending packets on subqueue void netif_wake_subqueue struct net_device * dev u16 queue_index dev network device queue_index sub queue index Resume individual transmit queue of a device with multiple transmit queues. Kernel Hackers Manual July 2017 netif_device_detach 9 4.1.27 netif_device_detach mark device as removed void netif_device_detach struct net_device * dev dev network device Mark device as removed from system and therefore no longer available. Kernel Hackers Manual July 2017 netif_device_attach 9 4.1.27 netif_device_attach mark device as attached void netif_device_attach struct net_device * dev dev network device Mark device as attached from system and restart if needed. Kernel Hackers Manual July 2017 skb_mac_gso_segment 9 4.1.27 skb_mac_gso_segment mac layer segmentation handler. struct sk_buff * skb_mac_gso_segment struct sk_buff * skb netdev_features_t features skb buffer to segment features features for the output path (see dev->features) Kernel Hackers Manual July 2017 __skb_gso_segment 9 4.1.27 __skb_gso_segment Perform segmentation on skb. struct sk_buff * __skb_gso_segment struct sk_buff * skb netdev_features_t features bool tx_path skb buffer to segment features features for the output path (see dev->features) tx_path whether it is called in TX path This function segments the given skb and returns a list of segments. It may return NULL if the skb requires no segmentation. This is only possible when GSO is used for verifying header integrity. Segmentation preserves SKB_SGO_CB_OFFSET bytes of previous skb cb. Kernel Hackers Manual July 2017 dev_loopback_xmit 9 4.1.27 dev_loopback_xmit loop back skb int dev_loopback_xmit struct sock * sk struct sk_buff * skb sk -- undescribed -- skb buffer to transmit Kernel Hackers Manual July 2017 rps_may_expire_flow 9 4.1.27 rps_may_expire_flow check whether an RFS hardware filter may be removed bool rps_may_expire_flow struct net_device * dev u16 rxq_index u32 flow_id u16 filter_id dev Device on which the filter was set rxq_index RX queue index flow_id Flow ID passed to ndo_rx_flow_steer filter_id Filter ID returned by ndo_rx_flow_steer Drivers that implement ndo_rx_flow_steer should periodically call this function for each installed filter and remove the filters for which it returns true. Kernel Hackers Manual July 2017 netif_rx 9 4.1.27 netif_rx post buffer to the network code int netif_rx struct sk_buff * skb skb buffer to post This function receives a packet from a device driver and queues it for the upper (protocol) levels to process. It always succeeds. The buffer may be dropped during processing for congestion control or by the protocol layers. NET_RX_SUCCESS (no congestion) NET_RX_DROP (packet was dropped) Kernel Hackers Manual July 2017 netdev_rx_handler_register 9 4.1.27 netdev_rx_handler_register register receive handler int netdev_rx_handler_register struct net_device * dev rx_handler_func_t * rx_handler void * rx_handler_data dev device to register a handler for rx_handler receive handler to register rx_handler_data data pointer that is used by rx handler Register a receive handler for a device. This handler will then be called from __netif_receive_skb. A negative errno code is returned on a failure. The caller must hold the rtnl_mutex. For a general description of rx_handler, see enum rx_handler_result. Kernel Hackers Manual July 2017 netdev_rx_handler_unregister 9 4.1.27 netdev_rx_handler_unregister unregister receive handler void netdev_rx_handler_unregister struct net_device * dev dev device to unregister a handler from Unregister a receive handler from a device. The caller must hold the rtnl_mutex. Kernel Hackers Manual July 2017 netif_receive_skb_sk 9 4.1.27 netif_receive_skb_sk process receive buffer from network int netif_receive_skb_sk struct sock * sk struct sk_buff * skb sk -- undescribed -- skb buffer to process netif_receive_skb is the main receive data processing function. It always succeeds. The buffer may be dropped during processing for congestion control or by the protocol layers. This function may only be called from softirq context and interrupts should be enabled. Return values (usually ignored): no congestion packet was dropped Kernel Hackers Manual July 2017 __napi_schedule 9 4.1.27 __napi_schedule schedule for receive void __napi_schedule struct napi_struct * n n entry to schedule The entry's receive function will be scheduled to run. Consider using __napi_schedule_irqoff if hard irqs are masked. Kernel Hackers Manual July 2017 __napi_schedule_irqoff 9 4.1.27 __napi_schedule_irqoff schedule for receive void __napi_schedule_irqoff struct napi_struct * n n entry to schedule Variant of __napi_schedule assuming hard irqs are masked Kernel Hackers Manual July 2017 netdev_has_upper_dev 9 4.1.27 netdev_has_upper_dev Check if device is linked to an upper device bool netdev_has_upper_dev struct net_device * dev struct net_device * upper_dev dev device upper_dev upper device to check Find out if a device is linked to specified upper device and return true in case it is. Note that this checks only immediate upper device, not through a complete stack of devices. The caller must hold the RTNL lock. Kernel Hackers Manual July 2017 netdev_master_upper_dev_get 9 4.1.27 netdev_master_upper_dev_get Get master upper device struct net_device * netdev_master_upper_dev_get struct net_device * dev dev device Find a master upper device and return pointer to it or NULL in case it's not there. The caller must hold the RTNL lock. Kernel Hackers Manual July 2017 netdev_upper_get_next_dev_rcu 9 4.1.27 netdev_upper_get_next_dev_rcu Get the next dev from upper list struct net_device * netdev_upper_get_next_dev_rcu struct net_device * dev struct list_head ** iter dev device iter list_head ** of the current position Gets the next device from the dev's upper list, starting from iter position. The caller must hold RCU read lock. Kernel Hackers Manual July 2017 netdev_all_upper_get_next_dev_rcu 9 4.1.27 netdev_all_upper_get_next_dev_rcu Get the next dev from upper list struct net_device * netdev_all_upper_get_next_dev_rcu struct net_device * dev struct list_head ** iter dev device iter list_head ** of the current position Gets the next device from the dev's upper list, starting from iter position. The caller must hold RCU read lock. Kernel Hackers Manual July 2017 netdev_lower_get_next_private 9 4.1.27 netdev_lower_get_next_private Get the next ->private from the lower neighbour list void * netdev_lower_get_next_private struct net_device * dev struct list_head ** iter dev device iter list_head ** of the current position Gets the next netdev_adjacent->private from the dev's lower neighbour list, starting from iter position. The caller must hold either hold the RTNL lock or its own locking that guarantees that the neighbour lower list will remain unchainged. Kernel Hackers Manual July 2017 netdev_lower_get_next_private_rcu 9 4.1.27 netdev_lower_get_next_private_rcu Get the next ->private from the lower neighbour list, RCU variant void * netdev_lower_get_next_private_rcu struct net_device * dev struct list_head ** iter dev device iter list_head ** of the current position Gets the next netdev_adjacent->private from the dev's lower neighbour list, starting from iter position. The caller must hold RCU read lock. Kernel Hackers Manual July 2017 netdev_lower_get_next 9 4.1.27 netdev_lower_get_next Get the next device from the lower neighbour list void * netdev_lower_get_next struct net_device * dev struct list_head ** iter dev device iter list_head ** of the current position Gets the next netdev_adjacent from the dev's lower neighbour list, starting from iter position. The caller must hold RTNL lock or its own locking that guarantees that the neighbour lower list will remain unchainged. Kernel Hackers Manual July 2017 netdev_lower_get_first_private_rcu 9 4.1.27 netdev_lower_get_first_private_rcu Get the first ->private from the lower neighbour list, RCU variant void * netdev_lower_get_first_private_rcu struct net_device * dev dev device Gets the first netdev_adjacent->private from the dev's lower neighbour list. The caller must hold RCU read lock. Kernel Hackers Manual July 2017 netdev_master_upper_dev_get_rcu 9 4.1.27 netdev_master_upper_dev_get_rcu Get master upper device struct net_device * netdev_master_upper_dev_get_rcu struct net_device * dev dev device Find a master upper device and return pointer to it or NULL in case it's not there. The caller must hold the RCU read lock. Kernel Hackers Manual July 2017 netdev_upper_dev_link 9 4.1.27 netdev_upper_dev_link Add a link to the upper device int netdev_upper_dev_link struct net_device * dev struct net_device * upper_dev dev device upper_dev new upper device Adds a link to device which is upper to this one. The caller must hold the RTNL lock. On a failure a negative errno code is returned. On success the reference counts are adjusted and the function returns zero. Kernel Hackers Manual July 2017 netdev_master_upper_dev_link 9 4.1.27 netdev_master_upper_dev_link Add a master link to the upper device int netdev_master_upper_dev_link struct net_device * dev struct net_device * upper_dev dev device upper_dev new upper device Adds a link to device which is upper to this one. In this case, only one master upper device can be linked, although other non-master devices might be linked as well. The caller must hold the RTNL lock. On a failure a negative errno code is returned. On success the reference counts are adjusted and the function returns zero. Kernel Hackers Manual July 2017 netdev_upper_dev_unlink 9 4.1.27 netdev_upper_dev_unlink Removes a link to upper device void netdev_upper_dev_unlink struct net_device * dev struct net_device * upper_dev dev device upper_dev new upper device Removes a link to device which is upper to this one. The caller must hold the RTNL lock. Kernel Hackers Manual July 2017 netdev_bonding_info_change 9 4.1.27 netdev_bonding_info_change Dispatch event about slave change void netdev_bonding_info_change struct net_device * dev struct netdev_bonding_info * bonding_info dev device bonding_info info to dispatch Send NETDEV_BONDING_INFO to netdev notifiers with info. The caller must hold the RTNL lock. Kernel Hackers Manual July 2017 dev_set_promiscuity 9 4.1.27 dev_set_promiscuity update promiscuity count on a device int dev_set_promiscuity struct net_device * dev int inc dev device inc modifier Add or remove promiscuity from a device. While the count in the device remains above zero the interface remains promiscuous. Once it hits zero the device reverts back to normal filtering operation. A negative inc value is used to drop promiscuity on the device. Return 0 if successful or a negative errno code on error. Kernel Hackers Manual July 2017 dev_set_allmulti 9 4.1.27 dev_set_allmulti update allmulti count on a device int dev_set_allmulti struct net_device * dev int inc dev device inc modifier Add or remove reception of all multicast frames to a device. While the count in the device remains above zero the interface remains listening to all interfaces. Once it hits zero the device reverts back to normal filtering operation. A negative inc value is used to drop the counter when releasing a resource needing all multicasts. Return 0 if successful or a negative errno code on error. Kernel Hackers Manual July 2017 dev_get_flags 9 4.1.27 dev_get_flags get flags reported to userspace unsigned int dev_get_flags const struct net_device * dev dev device Get the combination of flag bits exported through APIs to userspace. Kernel Hackers Manual July 2017 dev_change_flags 9 4.1.27 dev_change_flags change device settings int dev_change_flags struct net_device * dev unsigned int flags dev device flags device state flags Change settings on device based state flags. The flags are in the userspace exported format. Kernel Hackers Manual July 2017 dev_set_mtu 9 4.1.27 dev_set_mtu Change maximum transfer unit int dev_set_mtu struct net_device * dev int new_mtu dev device new_mtu new transfer unit Change the maximum transfer size of the network device. Kernel Hackers Manual July 2017 dev_set_group 9 4.1.27 dev_set_group Change group this device belongs to void dev_set_group struct net_device * dev int new_group dev device new_group group this device should belong to Kernel Hackers Manual July 2017 dev_set_mac_address 9 4.1.27 dev_set_mac_address Change Media Access Control Address int dev_set_mac_address struct net_device * dev struct sockaddr * sa dev device sa new address Change the hardware (MAC) address of the device Kernel Hackers Manual July 2017 dev_change_carrier 9 4.1.27 dev_change_carrier Change device carrier int dev_change_carrier struct net_device * dev bool new_carrier dev device new_carrier new value Change device carrier Kernel Hackers Manual July 2017 dev_get_phys_port_id 9 4.1.27 dev_get_phys_port_id Get device physical port ID int dev_get_phys_port_id struct net_device * dev struct netdev_phys_item_id * ppid dev device ppid port ID Get device physical port ID Kernel Hackers Manual July 2017 dev_get_phys_port_name 9 4.1.27 dev_get_phys_port_name Get device physical port name int dev_get_phys_port_name struct net_device * dev char * name size_t len dev device name port name len -- undescribed -- Get device physical port name Kernel Hackers Manual July 2017 netdev_update_features 9 4.1.27 netdev_update_features recalculate device features void netdev_update_features struct net_device * dev dev the device to check Recalculate dev->features set and send notifications if it has changed. Should be called after driver or hardware dependent conditions might have changed that influence the features. Kernel Hackers Manual July 2017 netdev_change_features 9 4.1.27 netdev_change_features recalculate device features void netdev_change_features struct net_device * dev dev the device to check Recalculate dev->features set and send notifications even if they have not changed. Should be called instead of netdev_update_features if also dev->vlan_features might have changed to allow the changes to be propagated to stacked VLAN devices. Kernel Hackers Manual July 2017 netif_stacked_transfer_operstate 9 4.1.27 netif_stacked_transfer_operstate transfer operstate void netif_stacked_transfer_operstate const struct net_device * rootdev struct net_device * dev rootdev the root or lower level device to transfer state from dev the device to transfer operstate to Transfer operational state from root to device. This is normally called when a stacking relationship exists between the root device and the device(a leaf device). Kernel Hackers Manual July 2017 register_netdevice 9 4.1.27 register_netdevice register a network device int register_netdevice struct net_device * dev dev device to register Take a completed network device structure and add it to the kernel interfaces. A NETDEV_REGISTER message is sent to the netdev notifier chain. 0 is returned on success. A negative errno code is returned on a failure to set up the device, or if the name is a duplicate. Callers must hold the rtnl semaphore. You may want register_netdev instead of this. The locking appears insufficient to guarantee two parallel registers will not get the same name. Kernel Hackers Manual July 2017 init_dummy_netdev 9 4.1.27 init_dummy_netdev init a dummy network device for NAPI int init_dummy_netdev struct net_device * dev dev device to init This takes a network device structure and initialize the minimum amount of fields so it can be used to schedule NAPI polls without registering a full blown interface. This is to be used by drivers that need to tie several hardware interfaces to a single NAPI poll scheduler due to HW limitations. Kernel Hackers Manual July 2017 register_netdev 9 4.1.27 register_netdev register a network device int register_netdev struct net_device * dev dev device to register Take a completed network device structure and add it to the kernel interfaces. A NETDEV_REGISTER message is sent to the netdev notifier chain. 0 is returned on success. A negative errno code is returned on a failure to set up the device, or if the name is a duplicate. This is a wrapper around register_netdevice that takes the rtnl semaphore and expands the device name if you passed a format string to alloc_netdev. Kernel Hackers Manual July 2017 dev_get_stats 9 4.1.27 dev_get_stats get network device statistics struct rtnl_link_stats64 * dev_get_stats struct net_device * dev struct rtnl_link_stats64 * storage dev device to get statistics from storage place to store stats Get network statistics from device. Return storage. The device driver may provide its own method by setting dev->netdev_ops->get_stats64 or dev->netdev_ops->get_stats; otherwise the internal statistics structure is used. Kernel Hackers Manual July 2017 alloc_netdev_mqs 9 4.1.27 alloc_netdev_mqs allocate network device struct net_device * alloc_netdev_mqs int sizeof_priv const char * name unsigned char name_assign_type void (*setup) struct net_device * unsigned int txqs unsigned int rxqs sizeof_priv size of private data to allocate space for name device name format string name_assign_type origin of device name setup callback to initialize device txqs the number of TX subqueues to allocate rxqs the number of RX subqueues to allocate Allocates a struct net_device with private data area for driver use and performs basic initialization. Also allocates subqueue structs for each queue on the device. Kernel Hackers Manual July 2017 free_netdev 9 4.1.27 free_netdev free network device void free_netdev struct net_device * dev dev device This function does the last stage of destroying an allocated device interface. The reference to the device object is released. If this is the last reference then it will be freed. Kernel Hackers Manual July 2017 synchronize_net 9 4.1.27 synchronize_net Synchronize with packet receive processing void synchronize_net void void no arguments Wait for packets currently being received to be done. Does not block later packets from starting. Kernel Hackers Manual July 2017 unregister_netdevice_queue 9 4.1.27 unregister_netdevice_queue remove device from the kernel void unregister_netdevice_queue struct net_device * dev struct list_head * head dev device head list This function shuts down a device interface and removes it from the kernel tables. If head not NULL, device is queued to be unregistered later. Callers must hold the rtnl semaphore. You may want unregister_netdev instead of this. Kernel Hackers Manual July 2017 unregister_netdevice_many 9 4.1.27 unregister_netdevice_many unregister many devices void unregister_netdevice_many struct list_head * head head list of devices As most callers use a stack allocated list_head, we force a list_del to make sure stack wont be corrupted later. Kernel Hackers Manual July 2017 unregister_netdev 9 4.1.27 unregister_netdev remove device from the kernel void unregister_netdev struct net_device * dev dev device This function shuts down a device interface and removes it from the kernel tables. This is just a wrapper for unregister_netdevice that takes the rtnl semaphore. In general you want to use this and not unregister_netdevice. Kernel Hackers Manual July 2017 dev_change_net_namespace 9 4.1.27 dev_change_net_namespace move device to different nethost namespace int dev_change_net_namespace struct net_device * dev struct net * net const char * pat dev device net network namespace pat If not NULL name pattern to try if the current device name is already taken in the destination network namespace. This function shuts down a device interface and moves it to a new network namespace. On success 0 is returned, on a failure a netagive errno code is returned. Callers must hold the rtnl semaphore. Kernel Hackers Manual July 2017 netdev_increment_features 9 4.1.27 netdev_increment_features increment feature set by one netdev_features_t netdev_increment_features netdev_features_t all netdev_features_t one netdev_features_t mask all current feature set one new feature set mask mask feature set Computes a new feature set after adding a device with feature set one to the master device with current feature set all. Will not enable anything that is off in mask. Returns the new feature set. Kernel Hackers Manual July 2017 eth_header 9 4.1.27 eth_header create the Ethernet header int eth_header struct sk_buff * skb struct net_device * dev unsigned short type const void * daddr const void * saddr unsigned int len skb buffer to alter dev source device type Ethernet type field daddr destination address (NULL leave destination address) saddr source address (NULL use device source address) len packet length (<= skb->len) Set the protocol type. For a packet of type ETH_P_802_3/2 we put the length in here instead. Kernel Hackers Manual July 2017 eth_get_headlen 9 4.1.27 eth_get_headlen determine the the length of header for an ethernet frame u32 eth_get_headlen void * data unsigned int len data pointer to start of frame len total length of frame Make a best effort attempt to pull the length for all of the headers for a given frame in a linear buffer. Kernel Hackers Manual July 2017 eth_type_trans 9 4.1.27 eth_type_trans determine the packet's protocol ID. __be16 eth_type_trans struct sk_buff * skb struct net_device * dev skb received socket data dev receiving network device The rule here is that we assume 802.3 if the type field is short enough to be a length. This is normal practice and works for any 'now in use' protocol. Kernel Hackers Manual July 2017 eth_header_parse 9 4.1.27 eth_header_parse extract hardware address from packet int eth_header_parse const struct sk_buff * skb unsigned char * haddr skb packet to extract header from haddr destination buffer Kernel Hackers Manual July 2017 eth_header_cache 9 4.1.27 eth_header_cache fill cache entry from neighbour int eth_header_cache const struct neighbour * neigh struct hh_cache * hh __be16 type neigh source neighbour hh destination cache entry type Ethernet type field Create an Ethernet header template from the neighbour. Kernel Hackers Manual July 2017 eth_header_cache_update 9 4.1.27 eth_header_cache_update update cache entry void eth_header_cache_update struct hh_cache * hh const struct net_device * dev const unsigned char * haddr hh destination cache entry dev network device haddr new hardware address Called by Address Resolution module to notify changes in address. Kernel Hackers Manual July 2017 eth_prepare_mac_addr_change 9 4.1.27 eth_prepare_mac_addr_change prepare for mac change int eth_prepare_mac_addr_change struct net_device * dev void * p dev network device p socket address Kernel Hackers Manual July 2017 eth_commit_mac_addr_change 9 4.1.27 eth_commit_mac_addr_change commit mac change void eth_commit_mac_addr_change struct net_device * dev void * p dev network device p socket address Kernel Hackers Manual July 2017 eth_mac_addr 9 4.1.27 eth_mac_addr set new Ethernet hardware address int eth_mac_addr struct net_device * dev void * p dev network device p socket address Change hardware address of device. This doesn't change hardware matching, so needs to be overridden for most real devices. Kernel Hackers Manual July 2017 eth_change_mtu 9 4.1.27 eth_change_mtu set new MTU size int eth_change_mtu struct net_device * dev int new_mtu dev network device new_mtu new Maximum Transfer Unit Allow changing MTU size. Needs to be overridden for devices supporting jumbo frames. Kernel Hackers Manual July 2017 ether_setup 9 4.1.27 ether_setup setup Ethernet network device void ether_setup struct net_device * dev dev network device Fill in the fields of the device structure with Ethernet-generic values. Kernel Hackers Manual July 2017 alloc_etherdev_mqs 9 4.1.27 alloc_etherdev_mqs Allocates and sets up an Ethernet device struct net_device * alloc_etherdev_mqs int sizeof_priv unsigned int txqs unsigned int rxqs sizeof_priv Size of additional driver-private structure to be allocated for this Ethernet device txqs The number of TX queues this device has. rxqs The number of RX queues this device has. Fill in the fields of the device structure with Ethernet-generic values. Basically does everything except registering the device. Constructs a new net device, complete with a private data area of size (sizeof_priv). A 32-byte (not bit) alignment is enforced for this private data area. Kernel Hackers Manual July 2017 netif_carrier_on 9 4.1.27 netif_carrier_on set carrier void netif_carrier_on struct net_device * dev dev network device Device has detected that carrier. Kernel Hackers Manual July 2017 netif_carrier_off 9 4.1.27 netif_carrier_off clear carrier void netif_carrier_off struct net_device * dev dev network device Device has detected loss of carrier. Kernel Hackers Manual July 2017 is_link_local_ether_addr 9 4.1.27 is_link_local_ether_addr Determine if given Ethernet address is link-local bool is_link_local_ether_addr const u8 * addr addr Pointer to a six-byte array containing the Ethernet address Return true if address is link local reserved addr (01:80:c2:00:00:0X) per IEEE 802.1Q 8.6.3 Frame filtering. addr must be aligned to u16. Kernel Hackers Manual July 2017 is_zero_ether_addr 9 4.1.27 is_zero_ether_addr Determine if give Ethernet address is all zeros. bool is_zero_ether_addr const u8 * addr addr Pointer to a six-byte array containing the Ethernet address Return true if the address is all zeroes. addr must be aligned to u16. Kernel Hackers Manual July 2017 is_multicast_ether_addr 9 4.1.27 is_multicast_ether_addr Determine if the Ethernet address is a multicast. bool is_multicast_ether_addr const u8 * addr addr Pointer to a six-byte array containing the Ethernet address Return true if the address is a multicast address. By definition the broadcast address is also a multicast address. Kernel Hackers Manual July 2017 is_local_ether_addr 9 4.1.27 is_local_ether_addr Determine if the Ethernet address is locally-assigned one (IEEE 802). bool is_local_ether_addr const u8 * addr addr Pointer to a six-byte array containing the Ethernet address Return true if the address is a local address. Kernel Hackers Manual July 2017 is_broadcast_ether_addr 9 4.1.27 is_broadcast_ether_addr Determine if the Ethernet address is broadcast bool is_broadcast_ether_addr const u8 * addr addr Pointer to a six-byte array containing the Ethernet address Return true if the address is the broadcast address. addr must be aligned to u16. Kernel Hackers Manual July 2017 is_unicast_ether_addr 9 4.1.27 is_unicast_ether_addr Determine if the Ethernet address is unicast bool is_unicast_ether_addr const u8 * addr addr Pointer to a six-byte array containing the Ethernet address Return true if the address is a unicast address. Kernel Hackers Manual July 2017 is_valid_ether_addr 9 4.1.27 is_valid_ether_addr Determine if the given Ethernet address is valid bool is_valid_ether_addr const u8 * addr addr Pointer to a six-byte array containing the Ethernet address Check that the Ethernet address (MAC) is not 00:00:00:00:00:00, is not a multicast address, and is not FF:FF:FF:FF:FF:FF. Return true if the address is valid. addr must be aligned to u16. Kernel Hackers Manual July 2017 eth_random_addr 9 4.1.27 eth_random_addr Generate software assigned random Ethernet address void eth_random_addr u8 * addr addr Pointer to a six-byte array containing the Ethernet address Generate a random Ethernet address (MAC) that is not multicast and has the local assigned bit set. Kernel Hackers Manual July 2017 eth_broadcast_addr 9 4.1.27 eth_broadcast_addr Assign broadcast address void eth_broadcast_addr u8 * addr addr Pointer to a six-byte array containing the Ethernet address Assign the broadcast address to the given address array. Kernel Hackers Manual July 2017 eth_zero_addr 9 4.1.27 eth_zero_addr Assign zero address void eth_zero_addr u8 * addr addr Pointer to a six-byte array containing the Ethernet address Assign the zero address to the given address array. Kernel Hackers Manual July 2017 eth_hw_addr_random 9 4.1.27 eth_hw_addr_random Generate software assigned random Ethernet and set device flag void eth_hw_addr_random struct net_device * dev dev pointer to net_device structure Generate a random Ethernet address (MAC) to be used by a net device and set addr_assign_type so the state can be read by sysfs and be used by userspace. Kernel Hackers Manual July 2017 ether_addr_copy 9 4.1.27 ether_addr_copy Copy an Ethernet address void ether_addr_copy u8 * dst const u8 * src dst Pointer to a six-byte array Ethernet address destination src Pointer to a six-byte array Ethernet address source dst & src must both be aligned to u16. Kernel Hackers Manual July 2017 eth_hw_addr_inherit 9 4.1.27 eth_hw_addr_inherit Copy dev_addr from another net_device void eth_hw_addr_inherit struct net_device * dst struct net_device * src dst pointer to net_device to copy dev_addr to src pointer to net_device to copy dev_addr from Copy the Ethernet address from one net_device to another along with the address attributes (addr_assign_type). Kernel Hackers Manual July 2017 ether_addr_equal 9 4.1.27 ether_addr_equal Compare two Ethernet addresses bool ether_addr_equal const u8 * addr1 const u8 * addr2 addr1 Pointer to a six-byte array containing the Ethernet address addr2 Pointer other six-byte array containing the Ethernet address Compare two Ethernet addresses, returns true if equal addr1 & addr2 must both be aligned to u16. Kernel Hackers Manual July 2017 ether_addr_equal_64bits 9 4.1.27 ether_addr_equal_64bits Compare two Ethernet addresses bool ether_addr_equal_64bits const u8 addr1[6+2] const u8 addr2[6+2] addr1[6+2] Pointer to an array of 8 bytes addr2[6+2] Pointer to an other array of 8 bytes Compare two Ethernet addresses, returns true if equal, false otherwise. The function doesn't need any conditional branches and possibly uses word memory accesses on CPU allowing cheap unaligned memory reads. arrays = { byte1, byte2, byte3, byte4, byte5, byte6, pad1, pad2 } Please note that alignment of addr1 & addr2 are only guaranteed to be 16 bits. Kernel Hackers Manual July 2017 ether_addr_equal_unaligned 9 4.1.27 ether_addr_equal_unaligned Compare two not u16 aligned Ethernet addresses bool ether_addr_equal_unaligned const u8 * addr1 const u8 * addr2 addr1 Pointer to a six-byte array containing the Ethernet address addr2 Pointer other six-byte array containing the Ethernet address Compare two Ethernet addresses, returns true if equal Use only when any Ethernet address may not be u16 aligned. Kernel Hackers Manual July 2017 is_etherdev_addr 9 4.1.27 is_etherdev_addr Tell if given Ethernet address belongs to the device. bool is_etherdev_addr const struct net_device * dev const u8 addr[6 + 2] dev Pointer to a device structure addr[6 + 2] Pointer to a six-byte array containing the Ethernet address Compare passed address with all addresses of the device. Return true if the address if one of the device addresses. Note that this function calls ether_addr_equal_64bits so take care of the right padding. Kernel Hackers Manual July 2017 compare_ether_header 9 4.1.27 compare_ether_header Compare two Ethernet headers unsigned long compare_ether_header const void * a const void * b a Pointer to Ethernet header b Pointer to Ethernet header Compare two Ethernet headers, returns 0 if equal. This assumes that the network header (i.e., IP header) is 4-byte aligned OR the platform can handle unaligned access. This is the case for all packets coming into netif_receive_skb or similar entry points. Kernel Hackers Manual July 2017 eth_skb_pad 9 4.1.27 eth_skb_pad Pad buffer to mininum number of octets for Ethernet frame int eth_skb_pad struct sk_buff * skb skb Buffer to pad An Ethernet frame should have a minimum size of 60 bytes. This function takes short frames and pads them with zeros up to the 60 byte limit. Kernel Hackers Manual July 2017 napi_schedule_prep 9 4.1.27 napi_schedule_prep check if napi can be scheduled bool napi_schedule_prep struct napi_struct * n n napi context Test if NAPI routine is already running, and if not mark it as running. This is used as a condition variable insure only one NAPI poll instance runs. We also make sure there is no pending NAPI disable. Kernel Hackers Manual July 2017 napi_schedule 9 4.1.27 napi_schedule schedule NAPI poll void napi_schedule struct napi_struct * n n napi context Schedule NAPI poll routine to be called if it is not already running. Kernel Hackers Manual July 2017 napi_schedule_irqoff 9 4.1.27 napi_schedule_irqoff schedule NAPI poll void napi_schedule_irqoff struct napi_struct * n n napi context Variant of napi_schedule, assuming hard irqs are masked. Kernel Hackers Manual July 2017 napi_complete 9 4.1.27 napi_complete NAPI processing complete void napi_complete struct napi_struct * n n napi context Mark NAPI processing as complete. Consider using napi_complete_done instead. Kernel Hackers Manual July 2017 napi_enable 9 4.1.27 napi_enable enable NAPI scheduling void napi_enable struct napi_struct * n n napi context Resume NAPI from being scheduled on this context. Must be paired with napi_disable. Kernel Hackers Manual July 2017 napi_synchronize 9 4.1.27 napi_synchronize wait until NAPI is not running void napi_synchronize const struct napi_struct * n n napi context Wait until NAPI is done being scheduled on this context. Waits till any outstanding processing completes but does not disable future activations. Kernel Hackers Manual July 2017 enum netdev_priv_flags 9 4.1.27 enum netdev_priv_flags struct net_device priv_flags enum netdev_priv_flags { IFF_802_1Q_VLAN, IFF_EBRIDGE, IFF_SLAVE_INACTIVE, IFF_MASTER_8023AD, IFF_MASTER_ALB, IFF_BONDING, IFF_SLAVE_NEEDARP, IFF_ISATAP, IFF_MASTER_ARPMON, IFF_WAN_HDLC, IFF_XMIT_DST_RELEASE, IFF_DONT_BRIDGE, IFF_DISABLE_NETPOLL, IFF_MACVLAN_PORT, IFF_BRIDGE_PORT, IFF_OVS_DATAPATH, IFF_TX_SKB_SHARING, IFF_UNICAST_FLT, IFF_TEAM_PORT, IFF_SUPP_NOFCS, IFF_LIVE_ADDR_CHANGE, IFF_MACVLAN, IFF_XMIT_DST_RELEASE_PERM, IFF_IPVLAN_MASTER, IFF_IPVLAN_SLAVE }; IFF_802_1Q_VLAN 802.1Q VLAN device IFF_EBRIDGE Ethernet bridging device IFF_SLAVE_INACTIVE bonding slave not the curr. active IFF_MASTER_8023AD bonding master, 802.3ad IFF_MASTER_ALB bonding master, balance-alb IFF_BONDING bonding master or slave IFF_SLAVE_NEEDARP need ARPs for validation IFF_ISATAP ISATAP interface (RFC4214) IFF_MASTER_ARPMON bonding master, ARP mon in use IFF_WAN_HDLC WAN HDLC device IFF_XMIT_DST_RELEASE dev_hard_start_xmit is allowed to release skb->dst IFF_DONT_BRIDGE disallow bridging this ether dev IFF_DISABLE_NETPOLL disable netpoll at run-time IFF_MACVLAN_PORT device used as macvlan port IFF_BRIDGE_PORT device used as bridge port IFF_OVS_DATAPATH device used as Open vSwitch datapath port IFF_TX_SKB_SHARING The interface supports sharing skbs on transmit IFF_UNICAST_FLT Supports unicast filtering IFF_TEAM_PORT device used as team port IFF_SUPP_NOFCS device supports sending custom FCS IFF_LIVE_ADDR_CHANGE device supports hardware address change when it's running IFF_MACVLAN Macvlan device IFF_XMIT_DST_RELEASE_PERM -- undescribed -- IFF_IPVLAN_MASTER -- undescribed -- IFF_IPVLAN_SLAVE -- undescribed -- These are the struct net_device, they are only set internally by drivers and used in the kernel. These flags are invisible to userspace, this means that the order of these flags can change during any kernel release. You should have a pretty good reason to be extending these flags. Kernel Hackers Manual July 2017 struct net_device 9 4.1.27 struct net_device The DEVICE structure. Actually, this whole structure is a big mistake. It mixes I/O data with strictly high-level data, and it has to know about almost every data structure used in the INET module. struct net_device { char name[IFNAMSIZ]; struct hlist_node name_hlist; char * ifalias; unsigned long mem_end; unsigned long mem_start; unsigned long base_addr; int irq; atomic_t carrier_changes; unsigned long state; struct list_head dev_list; struct list_head napi_list; struct list_head unreg_list; struct list_head close_list; struct {unnamed_struct}; struct garp_port __rcu * garp_port; struct mrp_port __rcu * mrp_port; struct device dev; const struct attribute_group * sysfs_groups[4]; const struct attribute_group * sysfs_rx_queue_group; const struct rtnl_link_ops * rtnl_link_ops; #define GSO_MAX_SIZE 65536 unsigned int gso_max_size; #define GSO_MAX_SEGS 65535 u16 gso_max_segs; u16 gso_min_segs; #ifdef CONFIG_DCB const struct dcbnl_rtnl_ops * dcbnl_ops; #endif u8 num_tc; struct netdev_tc_txq tc_to_txq[TC_MAX_QUEUE]; u8 prio_tc_map[TC_BITMASK + 1]; #if IS_ENABLED(CONFIG_FCOE) unsigned int fcoe_ddp_xid; #endif #if IS_ENABLED(CONFIG_CGROUP_NET_PRIO) struct netprio_map __rcu * priomap; #endif struct phy_device * phydev; struct lock_class_key * qdisc_tx_busylock; }; name[IFNAMSIZ] This is the first field of the visible part of this structure (i.e. as seen by users in the Space.c file). It is the name of the interface. name_hlist Device name hash chain, please keep it close to name[] ifalias SNMP alias mem_end Shared memory end mem_start Shared memory start base_addr Device I/O address irq Device IRQ number carrier_changes Stats to monitor carrier on<->off transitions state Generic network queuing layer state, see netdev_state_t dev_list The global list of network devices napi_list List entry, that is used for polling napi devices unreg_list List entry, that is used, when we are unregistering the device, see the function unregister_netdev close_list List entry, that is used, when we are closing the device {unnamed_struct} anonymous garp_port GARP mrp_port MRP dev Class/net/name entry sysfs_groups[4] Space for optional device, statistics and wireless sysfs groups sysfs_rx_queue_group Space for optional per-rx queue attributes rtnl_link_ops Rtnl_link_ops gso_max_size Maximum size of generic segmentation offload gso_max_segs Maximum number of segments that can be passed to the NIC for GSO gso_min_segs Minimum number of segments that can be passed to the NIC for GSO dcbnl_ops Data Center Bridging netlink ops num_tc Number of traffic classes in the net device tc_to_txq[TC_MAX_QUEUE] XXX: need comments on this one prio_tc_map[TC_BITMASK + 1] need comments on this one fcoe_ddp_xid Max exchange id for FCoE LRO by ddp priomap XXX: need comments on this one phydev Physical device may attach itself for hardware timestamping qdisc_tx_busylock XXX: need comments on this one cleanup struct net_device such that network protocol info moves out. Kernel Hackers Manual July 2017 netdev_priv 9 4.1.27 netdev_priv access network device private data void * netdev_priv const struct net_device * dev dev network device Get network device private data Kernel Hackers Manual July 2017 netif_start_queue 9 4.1.27 netif_start_queue allow transmit void netif_start_queue struct net_device * dev dev network device Allow upper layers to call the device hard_start_xmit routine. Kernel Hackers Manual July 2017 netif_wake_queue 9 4.1.27 netif_wake_queue restart transmit void netif_wake_queue struct net_device * dev dev network device Allow upper layers to call the device hard_start_xmit routine. Used for flow control when transmit resources are available. Kernel Hackers Manual July 2017 netif_stop_queue 9 4.1.27 netif_stop_queue stop transmitted packets void netif_stop_queue struct net_device * dev dev network device Stop upper layers calling the device hard_start_xmit routine. Used for flow control when transmit resources are unavailable. Kernel Hackers Manual July 2017 netif_queue_stopped 9 4.1.27 netif_queue_stopped test if transmit queue is flowblocked bool netif_queue_stopped const struct net_device * dev dev network device Test if transmit queue on device is currently unable to send. Kernel Hackers Manual July 2017 netdev_txq_bql_enqueue_prefetchw 9 4.1.27 netdev_txq_bql_enqueue_prefetchw prefetch bql data for write void netdev_txq_bql_enqueue_prefetchw struct netdev_queue * dev_queue dev_queue pointer to transmit queue BQL enabled drivers might use this helper in their ndo_start_xmit, to give appropriate hint to the cpu. Kernel Hackers Manual July 2017 netdev_txq_bql_complete_prefetchw 9 4.1.27 netdev_txq_bql_complete_prefetchw prefetch bql data for write void netdev_txq_bql_complete_prefetchw struct netdev_queue * dev_queue dev_queue pointer to transmit queue BQL enabled drivers might use this helper in their TX completion path, to give appropriate hint to the cpu. Kernel Hackers Manual July 2017 netdev_sent_queue 9 4.1.27 netdev_sent_queue report the number of bytes queued to hardware void netdev_sent_queue struct net_device * dev unsigned int bytes dev network device bytes number of bytes queued to the hardware device queue Report the number of bytes queued for sending/completion to the network device hardware queue. bytes should be a good approximation and should exactly match netdev_completed_queue bytes Kernel Hackers Manual July 2017 netdev_completed_queue 9 4.1.27 netdev_completed_queue report bytes and packets completed by device void netdev_completed_queue struct net_device * dev unsigned int pkts unsigned int bytes dev network device pkts actual number of packets sent over the medium bytes actual number of bytes sent over the medium Report the number of bytes and packets transmitted by the network device hardware queue over the physical medium, bytes must exactly match the bytes amount passed to netdev_sent_queue Kernel Hackers Manual July 2017 netdev_reset_queue 9 4.1.27 netdev_reset_queue reset the packets and bytes count of a network device void netdev_reset_queue struct net_device * dev_queue dev_queue network device Reset the bytes and packet count of a network device and clear the software flow control OFF bit for this network device Kernel Hackers Manual July 2017 netdev_cap_txqueue 9 4.1.27 netdev_cap_txqueue check if selected tx queue exceeds device queues u16 netdev_cap_txqueue struct net_device * dev u16 queue_index 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. Kernel Hackers Manual July 2017 netif_running 9 4.1.27 netif_running test if up bool netif_running const struct net_device * dev dev network device Test if the device has been brought up. Kernel Hackers Manual July 2017 netif_start_subqueue 9 4.1.27 netif_start_subqueue allow sending packets on subqueue void netif_start_subqueue struct net_device * dev u16 queue_index dev network device queue_index sub queue index Start individual transmit queue of a device with multiple transmit queues. Kernel Hackers Manual July 2017 netif_stop_subqueue 9 4.1.27 netif_stop_subqueue stop sending packets on subqueue void netif_stop_subqueue struct net_device * dev u16 queue_index dev network device queue_index sub queue index Stop individual transmit queue of a device with multiple transmit queues. Kernel Hackers Manual July 2017 __netif_subqueue_stopped 9 4.1.27 __netif_subqueue_stopped test status of subqueue bool __netif_subqueue_stopped const struct net_device * dev u16 queue_index dev network device queue_index sub queue index Check individual transmit queue of a device with multiple transmit queues. Kernel Hackers Manual July 2017 netif_is_multiqueue 9 4.1.27 netif_is_multiqueue test if device has multiple transmit queues bool netif_is_multiqueue const struct net_device * dev dev network device Check if device has multiple transmit queues Kernel Hackers Manual July 2017 dev_put 9 4.1.27 dev_put release reference to device void dev_put struct net_device * dev dev network device Release reference to device to allow it to be freed. Kernel Hackers Manual July 2017 dev_hold 9 4.1.27 dev_hold get reference to device void dev_hold struct net_device * dev dev network device Hold reference to device to keep it from being freed. Kernel Hackers Manual July 2017 netif_carrier_ok 9 4.1.27 netif_carrier_ok test if carrier present bool netif_carrier_ok const struct net_device * dev dev network device Check if carrier is present on device Kernel Hackers Manual July 2017 netif_dormant_on 9 4.1.27 netif_dormant_on mark device as dormant. void netif_dormant_on struct net_device * dev dev network device Mark device as dormant (as per RFC2863). The dormant state indicates that the relevant interface is not actually in a condition to pass packets (i.e., it is not 'up') but is in a pending state, waiting for some external event. For on- demand interfaces, this new state identifies the situation where the interface is waiting for events to place it in the up state. Kernel Hackers Manual July 2017 netif_dormant_off 9 4.1.27 netif_dormant_off set device as not dormant. void netif_dormant_off struct net_device * dev dev network device Device is not in dormant state. Kernel Hackers Manual July 2017 netif_dormant 9 4.1.27 netif_dormant test if carrier present bool netif_dormant const struct net_device * dev dev network device Check if carrier is present on device Kernel Hackers Manual July 2017 netif_oper_up 9 4.1.27 netif_oper_up test if device is operational bool netif_oper_up const struct net_device * dev dev network device Check if carrier is operational Kernel Hackers Manual July 2017 netif_device_present 9 4.1.27 netif_device_present is device available or removed bool netif_device_present struct net_device * dev dev network device Check if device has not been removed from system. Kernel Hackers Manual July 2017 netif_tx_lock 9 4.1.27 netif_tx_lock grab network device transmit lock void netif_tx_lock struct net_device * dev dev network device Get network device transmit lock Kernel Hackers Manual July 2017 __dev_uc_sync 9 4.1.27 __dev_uc_sync Synchonize device's unicast list int __dev_uc_sync struct net_device * dev int (*sync) struct net_device *, const unsigned char * int (*unsync) struct net_device *, const unsigned char * dev device to sync sync function to call if address should be added unsync function to call if address should be removed Add newly added addresses to the interface, and release addresses that have been deleted. Kernel Hackers Manual July 2017 __dev_uc_unsync 9 4.1.27 __dev_uc_unsync Remove synchronized addresses from device void __dev_uc_unsync struct net_device * dev int (*unsync) struct net_device *, const unsigned char * dev device to sync unsync function to call if address should be removed Remove all addresses that were added to the device by dev_uc_sync. Kernel Hackers Manual July 2017 __dev_mc_sync 9 4.1.27 __dev_mc_sync Synchonize device's multicast list int __dev_mc_sync struct net_device * dev int (*sync) struct net_device *, const unsigned char * int (*unsync) struct net_device *, const unsigned char * dev device to sync sync function to call if address should be added unsync function to call if address should be removed Add newly added addresses to the interface, and release addresses that have been deleted. Kernel Hackers Manual July 2017 __dev_mc_unsync 9 4.1.27 __dev_mc_unsync Remove synchronized addresses from device void __dev_mc_unsync struct net_device * dev int (*unsync) struct net_device *, const unsigned char * dev device to sync unsync function to call if address should be removed Remove all addresses that were added to the device by dev_mc_sync. Kernel Hackers Manual July 2017 phy_print_status 9 4.1.27 phy_print_status Convenience function to print out the current phy status void phy_print_status struct phy_device * phydev phydev the phy_device struct Kernel Hackers Manual July 2017 phy_ethtool_sset 9 4.1.27 phy_ethtool_sset generic ethtool sset function, handles all the details int phy_ethtool_sset struct phy_device * phydev struct ethtool_cmd * cmd phydev target phy_device struct cmd ethtool_cmd - We don't set port or transceiver, so we don't care what they were set to. - phy_start_aneg will make sure forced settings are sane, and choose the next best ones from the ones selected, so we don't care if ethtool tries to give us bad values. Kernel Hackers Manual July 2017 phy_mii_ioctl 9 4.1.27 phy_mii_ioctl generic PHY MII ioctl interface int phy_mii_ioctl struct phy_device * phydev struct ifreq * ifr int cmd phydev the phy_device struct ifr struct ifreq for socket ioctl's cmd ioctl cmd to execute Note that this function is currently incompatible with the PHYCONTROL layer. It changes registers without regard to current state. Use at own risk. Kernel Hackers Manual July 2017 phy_start_aneg 9 4.1.27 phy_start_aneg start auto-negotiation for this PHY device int phy_start_aneg struct phy_device * phydev phydev the phy_device struct Sanitizes the settings (if we're not autonegotiating them), and then calls the driver's config_aneg function. If the PHYCONTROL Layer is operating, we change the state to reflect the beginning of Auto-negotiation or forcing. Kernel Hackers Manual July 2017 phy_start_interrupts 9 4.1.27 phy_start_interrupts request and enable interrupts for a PHY device int phy_start_interrupts struct phy_device * phydev phydev target phy_device struct Request the interrupt for the given PHY. If this fails, then we set irq to PHY_POLL. Otherwise, we enable the interrupts in the PHY. This should only be called with a valid IRQ number. Returns 0 on success or < 0 on error. Kernel Hackers Manual July 2017 phy_stop_interrupts 9 4.1.27 phy_stop_interrupts disable interrupts from a PHY device int phy_stop_interrupts struct phy_device * phydev phydev target phy_device struct Kernel Hackers Manual July 2017 phy_stop 9 4.1.27 phy_stop Bring down the PHY link, and stop checking the status void phy_stop struct phy_device * phydev phydev target phy_device struct Kernel Hackers Manual July 2017 phy_start 9 4.1.27 phy_start start or restart a PHY device void phy_start struct phy_device * phydev phydev target phy_device struct Indicates the attached device's readiness to handle PHY-related work. Used during startup to start the PHY, and after a call to phy_stop to resume operation. Also used to indicate the MDIO bus has cleared an error condition. Kernel Hackers Manual July 2017 phy_read_mmd_indirect 9 4.1.27 phy_read_mmd_indirect reads data from the MMD registers int phy_read_mmd_indirect struct phy_device * phydev int prtad int devad int addr phydev The PHY device bus prtad MMD Address devad MMD DEVAD addr PHY address on the MII bus it reads data from the MMD registers (clause 22 to access to clause 45) of the specified phy address. 1) Write reg 13 // DEVAD 2) Write reg 14 // MMD Address 3) Write reg 13 // MMD Data Command for MMD DEVAD 3) Read reg 14 // Read MMD data Kernel Hackers Manual July 2017 phy_write_mmd_indirect 9 4.1.27 phy_write_mmd_indirect writes data to the MMD registers void phy_write_mmd_indirect struct phy_device * phydev int prtad int devad int addr u32 data phydev The PHY device prtad MMD Address devad MMD DEVAD addr PHY address on the MII bus data data to write in the MMD register Write data from the MMD registers of the specified phy address. 1) Write reg 13 // DEVAD 2) Write reg 14 // MMD Address 3) Write reg 13 // MMD Data Command for MMD DEVAD 3) Write reg 14 // Write MMD data Kernel Hackers Manual July 2017 phy_init_eee 9 4.1.27 phy_init_eee init and check the EEE feature int phy_init_eee struct phy_device * phydev bool clk_stop_enable phydev target phy_device struct clk_stop_enable PHY may stop the clock during LPI it checks if the Energy-Efficient Ethernet (EEE) is supported by looking at the MMD registers 3.20 and 7.60/61 and it programs the MMD register 3.0 setting the Clock stop enable bit if required. Kernel Hackers Manual July 2017 phy_get_eee_err 9 4.1.27 phy_get_eee_err report the EEE wake error count int phy_get_eee_err struct phy_device * phydev phydev target phy_device struct it is to report the number of time where the PHY failed to complete its normal wake sequence. Kernel Hackers Manual July 2017 phy_ethtool_get_eee 9 4.1.27 phy_ethtool_get_eee get EEE supported and status int phy_ethtool_get_eee struct phy_device * phydev struct ethtool_eee * data phydev target phy_device struct data ethtool_eee data it reportes the Supported/Advertisement/LP Advertisement capabilities. Kernel Hackers Manual July 2017 phy_ethtool_set_eee 9 4.1.27 phy_ethtool_set_eee set EEE supported and status int phy_ethtool_set_eee struct phy_device * phydev struct ethtool_eee * data phydev target phy_device struct data ethtool_eee data it is to program the Advertisement EEE register. Kernel Hackers Manual July 2017 phy_clear_interrupt 9 4.1.27 phy_clear_interrupt Ack the phy device's interrupt int phy_clear_interrupt struct phy_device * phydev phydev the phy_device struct If the phydev driver has an ack_interrupt function, call it to ack and clear the phy device's interrupt. Returns 0 on success or < 0 on error. Kernel Hackers Manual July 2017 phy_config_interrupt 9 4.1.27 phy_config_interrupt configure the PHY device for the requested interrupts int phy_config_interrupt struct phy_device * phydev u32 interrupts phydev the phy_device struct interrupts interrupt flags to configure for this phydev Returns 0 on success or < 0 on error. Kernel Hackers Manual July 2017 phy_aneg_done 9 4.1.27 phy_aneg_done return auto-negotiation status int phy_aneg_done struct phy_device * phydev phydev target phy_device struct Return the auto-negotiation status from this phydev Returns > 0 on success or < 0 on error. 0 means that auto-negotiation is still pending. Kernel Hackers Manual July 2017 phy_find_setting 9 4.1.27 phy_find_setting find a PHY settings array entry that matches speed & duplex unsigned int phy_find_setting int speed int duplex speed speed to match duplex duplex to match Searches the settings array for the setting which matches the desired speed and duplex, and returns the index of that setting. Returns the index of the last setting if none of the others match. Kernel Hackers Manual July 2017 phy_find_valid 9 4.1.27 phy_find_valid find a PHY setting that matches the requested features mask unsigned int phy_find_valid unsigned int idx u32 features idx The first index in settings[] to search features A mask of the valid settings Returns the index of the first valid setting less than or equal to the one pointed to by idx, as determined by the mask in features. Returns the index of the last setting if nothing else matches. Kernel Hackers Manual July 2017 phy_check_valid 9 4.1.27 phy_check_valid check if there is a valid PHY setting which matches speed, duplex, and feature mask bool phy_check_valid int speed int duplex u32 features speed speed to match duplex duplex to match features A mask of the valid settings Returns true if there is a valid setting, false otherwise. Kernel Hackers Manual July 2017 phy_sanitize_settings 9 4.1.27 phy_sanitize_settings make sure the PHY is set to supported speed and duplex void phy_sanitize_settings struct phy_device * phydev phydev the target phy_device struct Make sure the PHY is set to supported speeds and duplexes. Drop down by one in this order: 1000/FULL, 1000/HALF, 100/FULL, 100/HALF, 10/FULL, 10/HALF. Kernel Hackers Manual July 2017 phy_start_machine 9 4.1.27 phy_start_machine start PHY state machine tracking void phy_start_machine struct phy_device * phydev phydev the phy_device struct The PHY infrastructure can run a state machine which tracks whether the PHY is starting up, negotiating, etc. This function starts the timer which tracks the state of the PHY. If you want to maintain your own state machine, do not call this function. Kernel Hackers Manual July 2017 phy_stop_machine 9 4.1.27 phy_stop_machine stop the PHY state machine tracking void phy_stop_machine struct phy_device * phydev phydev target phy_device struct Stops the state machine timer, sets the state to UP (unless it wasn't up yet). This function must be called BEFORE phy_detach. Kernel Hackers Manual July 2017 phy_error 9 4.1.27 phy_error enter HALTED state for this PHY device void phy_error struct phy_device * phydev phydev target phy_device struct Moves the PHY to the HALTED state in response to a read or write error, and tells the controller the link is down. Must not be called from interrupt context, or while the phydev->lock is held. Kernel Hackers Manual July 2017 phy_interrupt 9 4.1.27 phy_interrupt PHY interrupt handler irqreturn_t phy_interrupt int irq void * phy_dat irq interrupt line phy_dat phy_device pointer When a PHY interrupt occurs, the handler disables interrupts, and schedules a work task to clear the interrupt. Kernel Hackers Manual July 2017 phy_enable_interrupts 9 4.1.27 phy_enable_interrupts Enable the interrupts from the PHY side int phy_enable_interrupts struct phy_device * phydev phydev target phy_device struct Kernel Hackers Manual July 2017 phy_disable_interrupts 9 4.1.27 phy_disable_interrupts Disable the PHY interrupts from the PHY side int phy_disable_interrupts struct phy_device * phydev phydev target phy_device struct Kernel Hackers Manual July 2017 phy_change 9 4.1.27 phy_change Scheduled by the phy_interrupt/timer to handle PHY changes void phy_change struct work_struct * work work work_struct that describes the work to be done Kernel Hackers Manual July 2017 phy_state_machine 9 4.1.27 phy_state_machine Handle the state machine void phy_state_machine struct work_struct * work work work_struct that describes the work to be done Kernel Hackers Manual July 2017 phy_register_fixup 9 4.1.27 phy_register_fixup creates a new phy_fixup and adds it to the list int phy_register_fixup const char * bus_id u32 phy_uid u32 phy_uid_mask int (*run) struct phy_device * bus_id A string which matches phydev->dev.bus_id (or PHY_ANY_ID) phy_uid Used to match against phydev->phy_id (the UID of the PHY) It can also be PHY_ANY_UID phy_uid_mask Applied to phydev->phy_id and fixup->phy_uid before comparison run The actual code to be run when a matching PHY is found Kernel Hackers Manual July 2017 get_phy_device 9 4.1.27 get_phy_device reads the specified PHY device and returns its phy_device struct struct phy_device * get_phy_device struct mii_bus * bus int addr bool is_c45 bus the target MII bus addr PHY address on the MII bus is_c45 If true the PHY uses the 802.3 clause 45 protocol Reads the ID registers of the PHY at addr on the bus, then allocates and returns the phy_device to represent it. Kernel Hackers Manual July 2017 phy_device_register 9 4.1.27 phy_device_register Register the phy device on the MDIO bus int phy_device_register struct phy_device * phydev phydev phy_device structure to be added to the MDIO bus Kernel Hackers Manual July 2017 phy_find_first 9 4.1.27 phy_find_first finds the first PHY device on the bus struct phy_device * phy_find_first struct mii_bus * bus bus the target MII bus Kernel Hackers Manual July 2017 phy_connect_direct 9 4.1.27 phy_connect_direct connect an ethernet device to a specific phy_device int phy_connect_direct struct net_device * dev struct phy_device * phydev void (*handler) struct net_device * phy_interface_t interface dev the network device to connect phydev the pointer to the phy device handler callback function for state change notifications interface PHY device's interface Kernel Hackers Manual July 2017 phy_connect 9 4.1.27 phy_connect connect an ethernet device to a PHY device struct phy_device * phy_connect struct net_device * dev const char * bus_id void (*handler) struct net_device * phy_interface_t interface dev the network device to connect bus_id the id string of the PHY device to connect handler callback function for state change notifications interface PHY device's interface Convenience function for connecting ethernet devices to PHY devices. The default behavior is for the PHY infrastructure to handle everything, and only notify the connected driver when the link status changes. If you don't want, or can't use the provided functionality, you may choose to call only the subset of functions which provide the desired functionality. Kernel Hackers Manual July 2017 phy_disconnect 9 4.1.27 phy_disconnect disable interrupts, stop state machine, and detach a PHY device void phy_disconnect struct phy_device * phydev phydev target phy_device struct Kernel Hackers Manual July 2017 phy_attach_direct 9 4.1.27 phy_attach_direct attach a network device to a given PHY device pointer int phy_attach_direct struct net_device * dev struct phy_device * phydev u32 flags phy_interface_t interface dev network device to attach phydev Pointer to phy_device to attach flags PHY device's dev_flags interface PHY device's interface Called by drivers to attach to a particular PHY device. The phy_device is found, and properly hooked up to the phy_driver. If no driver is attached, then a generic driver is used. The phy_device is given a ptr to the attaching device, and given a callback for link status change. The phy_device is returned to the attaching driver. Kernel Hackers Manual July 2017 phy_attach 9 4.1.27 phy_attach attach a network device to a particular PHY device struct phy_device * phy_attach struct net_device * dev const char * bus_id phy_interface_t interface dev network device to attach bus_id Bus ID of PHY device to attach interface PHY device's interface Same as phy_attach_direct except that a PHY bus_id string is passed instead of a pointer to a struct phy_device. Kernel Hackers Manual July 2017 phy_detach 9 4.1.27 phy_detach detach a PHY device from its network device void phy_detach struct phy_device * phydev phydev target phy_device struct Kernel Hackers Manual July 2017 genphy_setup_forced 9 4.1.27 genphy_setup_forced configures/forces speed/duplex from phydev int genphy_setup_forced struct phy_device * phydev phydev target phy_device struct Configures MII_BMCR to force speed/duplex to the values in phydev. Assumes that the values are valid. Please see phy_sanitize_settings. Kernel Hackers Manual July 2017 genphy_restart_aneg 9 4.1.27 genphy_restart_aneg Enable and Restart Autonegotiation int genphy_restart_aneg struct phy_device * phydev phydev target phy_device struct Kernel Hackers Manual July 2017 genphy_config_aneg 9 4.1.27 genphy_config_aneg restart auto-negotiation or write BMCR int genphy_config_aneg struct phy_device * phydev phydev target phy_device struct If auto-negotiation is enabled, we configure the advertising, and then restart auto-negotiation. If it is not enabled, then we write the BMCR. Kernel Hackers Manual July 2017 genphy_aneg_done 9 4.1.27 genphy_aneg_done return auto-negotiation status int genphy_aneg_done struct phy_device * phydev phydev target phy_device struct Reads the status register and returns 0 either if auto-negotiation is incomplete, or if there was an error. Returns BMSR_ANEGCOMPLETE if auto-negotiation is done. Kernel Hackers Manual July 2017 genphy_update_link 9 4.1.27 genphy_update_link update link status in phydev int genphy_update_link struct phy_device * phydev phydev target phy_device struct Update the value in phydev->link to reflect the current link value. In order to do this, we need to read the status register twice, keeping the second value. Kernel Hackers Manual July 2017 genphy_read_status 9 4.1.27 genphy_read_status check the link status and update current link state int genphy_read_status struct phy_device * phydev phydev target phy_device struct Check the link, then figure out the current state by comparing what we advertise with what the link partner advertises. Start by checking the gigabit possibilities, then move on to 10/100. Kernel Hackers Manual July 2017 genphy_soft_reset 9 4.1.27 genphy_soft_reset software reset the PHY via BMCR_RESET bit int genphy_soft_reset struct phy_device * phydev phydev target phy_device struct Perform a software PHY reset using the standard BMCR_RESET bit and poll for the reset bit to be cleared. 0 on success, < 0 on failure Kernel Hackers Manual July 2017 phy_driver_register 9 4.1.27 phy_driver_register register a phy_driver with the PHY layer int phy_driver_register struct phy_driver * new_driver new_driver new phy_driver to register Kernel Hackers Manual July 2017 get_phy_c45_ids 9 4.1.27 get_phy_c45_ids reads the specified addr for its 802.3-c45 IDs. int get_phy_c45_ids struct mii_bus * bus int addr u32 * phy_id struct phy_c45_device_ids * c45_ids bus the target MII bus addr PHY address on the MII bus phy_id where to store the ID retrieved. c45_ids where to store the c45 ID information. If the PHY devices-in-package appears to be valid, it and the corresponding identifiers are stored in c45_ids, zero is stored in phy_id. Otherwise 0xffffffff is stored in phy_id. Returns zero on success. Kernel Hackers Manual July 2017 get_phy_id 9 4.1.27 get_phy_id reads the specified addr for its ID. int get_phy_id struct mii_bus * bus int addr u32 * phy_id bool is_c45 struct phy_c45_device_ids * c45_ids bus the target MII bus addr PHY address on the MII bus phy_id where to store the ID retrieved. is_c45 If true the PHY uses the 802.3 clause 45 protocol c45_ids where to store the c45 ID information. In the case of a 802.3-c22 PHY, reads the ID registers of the PHY at addr on the bus, stores it in phy_id and returns zero on success. In the case of a 802.3-c45 PHY, get_phy_c45_ids is invoked, and its return value is in turn returned. Kernel Hackers Manual July 2017 phy_prepare_link 9 4.1.27 phy_prepare_link prepares the PHY layer to monitor link status void phy_prepare_link struct phy_device * phydev void (*handler) struct net_device * phydev target phy_device struct handler callback function for link status change notifications Tells the PHY infrastructure to handle the gory details on monitoring link status (whether through polling or an interrupt), and to call back to the connected device driver when the link status changes. If you want to monitor your own link state, don't call this function. Kernel Hackers Manual July 2017 phy_poll_reset 9 4.1.27 phy_poll_reset Safely wait until a PHY reset has properly completed int phy_poll_reset struct phy_device * phydev phydev The PHY device to poll According to IEEE 802.3, Section 2, Subsection 22.2.4.1.1, as published in 2008, a PHY reset may take up to 0.5 seconds. The MII BMCR register must be polled until the BMCR_RESET bit clears. Furthermore, any attempts to write to PHY registers may have no effect or even generate MDIO bus errors until this is complete. Some PHYs (such as the Marvell 88E1111) don't entirely conform to the standard and do not fully reset after the BMCR_RESET bit is set, and may even *REQUIRE* a soft-reset to properly restart autonegotiation. In an effort to support such broken PHYs, this function is separate from the standard phy_init_hw which will zero all the other bits in the BMCR and reapply all driver-specific and board-specific fixups. Kernel Hackers Manual July 2017 genphy_config_advert 9 4.1.27 genphy_config_advert sanitize and advertise auto-negotiation parameters int genphy_config_advert struct phy_device * phydev phydev target phy_device struct Writes MII_ADVERTISE with the appropriate values, after sanitizing the values to make sure we only advertise what is supported. Returns < 0 on error, 0 if the PHY's advertisement hasn't changed, and > 0 if it has changed. Kernel Hackers Manual July 2017 phy_probe 9 4.1.27 phy_probe probe and init a PHY device int phy_probe struct device * dev dev device to probe and init Take care of setting up the phy_device structure, set the state to READY (the driver's init function should set it to STARTING if needed). Kernel Hackers Manual July 2017 mdiobus_alloc_size 9 4.1.27 mdiobus_alloc_size allocate a mii_bus structure struct mii_bus * mdiobus_alloc_size size_t size size extra amount of memory to allocate for private storage. If non-zero, then bus->priv is points to that memory. called by a bus driver to allocate an mii_bus structure to fill in. Kernel Hackers Manual July 2017 devm_mdiobus_alloc_size 9 4.1.27 devm_mdiobus_alloc_size Resource-managed mdiobus_alloc_size struct mii_bus * devm_mdiobus_alloc_size struct device * dev int sizeof_priv dev Device to allocate mii_bus for sizeof_priv Space to allocate for private structure. Managed mdiobus_alloc_size. mii_bus allocated with this function is automatically freed on driver detach. If an mii_bus allocated with this function needs to be freed separately, devm_mdiobus_free must be used. Pointer to allocated mii_bus on success, NULL on failure. Kernel Hackers Manual July 2017 devm_mdiobus_free 9 4.1.27 devm_mdiobus_free Resource-managed mdiobus_free void devm_mdiobus_free struct device * dev struct mii_bus * bus dev Device this mii_bus belongs to bus the mii_bus associated with the device Free mii_bus allocated with devm_mdiobus_alloc_size. Kernel Hackers Manual July 2017 of_mdio_find_bus 9 4.1.27 of_mdio_find_bus Given an mii_bus node, find the mii_bus. struct mii_bus * of_mdio_find_bus struct device_node * mdio_bus_np mdio_bus_np Pointer to the mii_bus. Returns a pointer to the mii_bus, or NULL if none found. Because the association of a device_node and mii_bus is made via of_mdiobus_register, the mii_bus cannot be found before it is registered with of_mdiobus_register. Kernel Hackers Manual July 2017 mdiobus_register 9 4.1.27 mdiobus_register bring up all the PHYs on a given bus and attach them to bus int mdiobus_register struct mii_bus * bus bus target mii_bus Called by a bus driver to bring up all the PHYs on a given bus, and attach them to the bus. Returns 0 on success or < 0 on error. Kernel Hackers Manual July 2017 mdiobus_free 9 4.1.27 mdiobus_free free a struct mii_bus void mdiobus_free struct mii_bus * bus bus mii_bus to free This function releases the reference to the underlying device object in the mii_bus. If this is the last reference, the mii_bus will be freed. Kernel Hackers Manual July 2017 mdiobus_read 9 4.1.27 mdiobus_read Convenience function for reading a given MII mgmt register int mdiobus_read struct mii_bus * bus int addr u32 regnum bus the mii_bus struct addr the phy address regnum register number to read MUST NOT be called from interrupt context, because the bus read/write functions may wait for an interrupt to conclude the operation. Kernel Hackers Manual July 2017 mdiobus_write 9 4.1.27 mdiobus_write Convenience function for writing a given MII mgmt register int mdiobus_write struct mii_bus * bus int addr u32 regnum u16 val bus the mii_bus struct addr the phy address regnum register number to write val value to write to regnum MUST NOT be called from interrupt context, because the bus read/write functions may wait for an interrupt to conclude the operation. Kernel Hackers Manual July 2017 mdiobus_release 9 4.1.27 mdiobus_release mii_bus device release callback void mdiobus_release struct device * d d the target struct device that contains the mii_bus called when the last reference to an mii_bus is dropped, to free the underlying memory. Kernel Hackers Manual July 2017 mdio_bus_match 9 4.1.27 mdio_bus_match determine if given PHY driver supports the given PHY device int mdio_bus_match struct device * dev struct device_driver * drv dev target PHY device drv given PHY driver Given a PHY device, and a PHY driver, return 1 if the driver supports the device. Otherwise, return 0.