source: trunk/kernel/devices/dev_nic.h @ 657

Last change on this file since 657 was 657, checked in by alain, 22 months ago

Introduce remote_buf.c/.h & socket.c/.h files.
Update dev_nic.c/.h files.

File size: 32.3 KB
Line 
1/*
2 * dev_nic.h - NIC (Network Controler) generic device API definition.
3 *
4 * Author  Alain Greiner    (2016,2017,2018,2019,2020)
5 *
6 * Copyright (c) UPMC Sorbonne Universites
7 *
8 * This file is part of ALMOS-MKH
9 *
10 * ALMOS-MKH is free software; you can redistribute it and/or modify it
11 * under the terms of the GNU General Public License as published by
12 * the Free Software Foundation; version 2.0 of the License.
13 *
14 * ALMOS-MKH is distributed in the hope that it will be useful, but
15 * WITHOUT ANY WARRANTY; without even the implied warranty of
16 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
17 * General Public License for more details.
18 *
19 * You should have received a copy of the GNU General Public License
20 * along with ALMOS-kernel; if not, write to the Free Software Foundation,
21 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
22 */
23
24#ifndef _DEV_NIC_H
25#define _DEV_NIC_H
26
27#include <kernel_config.h>
28#include <hal_kernel_types.h>
29#include <remote_busylock.h>
30#include <remote_buf.h>
31#include <xlist.h>
32
33/****  Forward declarations  ****/
34
35struct chdev_s;
36
37/*****************************************************************************************
38 *     Generic Network Interface Controler definition
39 *
40 * This device provides access to a generic Gigabit Ethernet network controler.
41 * It assumes that the NIC hardware peripheral handles two packets queues for sent (TX)
42 * and received (RX) packets.
43 *
44 * The supported protocols stack is : Ethernet / IPV4 / TCP or UDP
45 *
46 * 1) hardware assumptions
47 *
48 * The NIC device is handling two (infinite) streams of packets to or from the network.
49 * It is the driver responsibility to move the RX packets from the NIC to the RX queue,
50 * and the TX packets from the TX queue to the NIC.
51 *
52 * AS the RX and TX queues are independant, there is one NIC_RX device descriptor
53 * to handle RX packets, and another NIC_TX device descriptor to handle TX packets.
54 *
55 * In order to improve throughput, the NIC controller can implement multiple (N) channels.
56 * In this case, the channel index is defined by an hash function computed from the remote
57 * IP address and port. This index is computed by the hardware for an RX packet, and is
58 * computed by the kernel for a TX packet, using a specific driver function. TODO ...
59 * The 2*N chdevs, and the associated server threads implementing the protocols stack,
60 * are distributed in 2*N different clusters.
61 *
62 * 2) User API
63 *
64 * On the user side, ALMOS-MKH implements the POSIX socket API.
65 * The kernel functions implementing the socket related syscalls are :
66 * - dev_nic_socket()    : create a local socket registered in process fd_array[].
67 * - dev_nic_bind()      : attach a local IP address and port to a local socket.
68 * - dev_nic_listen()    : local server makes a passive open.
69 * - dev_nic_connect()   : local client makes an active open to a remote server.
70 * - dev_nic_accept()    : local server accept a new remote client.
71 * - dev_nic_send()      : send data on a connected socket.
72 * - dev_nic_recv()      : receive data on a connected socket.
73 * - dev_nic_sendto()    : send a packet to a remote (IP address/port).
74 * - dev_nic_recvfrom()  : receive a paket from a remote (IP address/port).
75 * - dev_nic_close()     : close a socket
76 *
77 * 3) TX stream
78 *
79 * The internal API between the client threads and the TX server thread defines
80 * the 3 following commands:
81 *   . SOCKET_TX_CONNECT : request to execute the 3 steps TCP connection handshake.
82 *   . SOCKET_TX_SEND    : send data to a remote socket (UDP or TCP).
83 *   . SOCKET_TX_CLOSE   : request to execute the 3 steps TCP close handshake.
84 *
85 * - These 3 commands are blocking for the client thread that registers the command in the
86 *   socket descriptor, blocks on the BLOCKED_IO condition, and deschedules.
87 * - The TX server thread is acting as a multiplexer. It scans the list of attached sockets,
88 *   to handle all valid commands: one UDP packet or TCP segment per iteration.
89 *   It uses the user buffer defined by the client thread, and attached to socket descriptor,
90 *   as a retransmission buffer. It blocks and deschedules on the BLOCKED_CLIENT condition,
91 *   when there is no more active TX command registered in any socket. It is re-activated
92 *   by the first client thread registering a new TX command in the socket descriptor.
93 *   It unblocks a client thread only when a command is fully completed. It signals errors
94 *   to the client thread using the tx_error field in socket descriptor.
95 *
96 * 4) RX stream
97 *
98 * The communication between the RX server thread and the client threads expecting data
99 * is done through receive buffers (one private buffer per socket) that are handled
100 * as single-writer / single reader-FIFOs, called rx_buf.
101 * - The RX server thread is acting as a demultiplexor: it handle one TCP segment or UDP
102 *   packet per iteration, and register the data in the rx_buf of the socket matching
103 *   the packet. It simply discard all packets that does not match a registered socket.
104 *   When a client thread is registered in the socket descriptor, the RX server thread
105 *   unblocks this client thread as soon as there is data available in rx_buf.
106 *   It blocks and deschedules on the BLOCKED_ISR condition when there is no more packets
107 *   in the NIC_RX queue. It is unblocked by the hardware ISR.
108 * - The client thread simply access the rx_buf attached to socket descriptor, and consumes
109 *   the available data when the rx_buf is non empty. It blocks on the BLOCKED_IO condition,
110 *   and deschedules when the rx_buf is empty.
111 *
112 * 5) R2T queue
113 *
114 * To implement the TCP "3 steps handshake" protocol, the RX server thread can directly
115 * request the associated TX server thread to send control packets in  the TX stream,
116 * using a dedicate R2T (RX to TX) FIFO stored in the socket descriptor.
117 *
118 * 6) NIC driver API
119 *
120 * The generic NIC device "driver" API defines the following commands to the NIC driver:
121 * - READABLE : returns true if at least one RX paquet is available in RX queue.
122 * - WRITABLE : returns true if at least one empty slot is available in TX queue.
123 * - READ     : consume one packet from the RX queue.
124 * - WRITE    : produce one packet to the TX queue.
125 * All RX or TX paquets are sent or received in standard 2 Kbytes kernel buffers,
126 * that are dynamically allocated by the protocols stack.
127 *
128 * The actual TX an RX queues structures depends on the hardware NIC implementation,
129 * and are defined in the HAL specific driver code.
130 *
131 * WARNING: the WTI mailboxes used by the driver ro receive events from the hardware
132 * (available RX packet, or available free TX slot, for a given channel), must be
133 * statically allocated during the kernel initialisation phase, and must be
134 * routed to the cluster containing the associated TX/RX chdev and server thread.
135 *
136 *****************************************************************************************/
137
138/****  Forward declarations  ****/
139
140struct chdev_s;
141
142/******************************************************************************************
143 *   Various constants used by the Protocols stack
144 *****************************************************************************************/
145
146#define SRC_MAC_54             0x54
147#define SRC_MAC_32             0x32
148#define SRC_MAC_10             0x10
149#define DST_MAC_54             0x54
150#define DST_MAC_32             0x32
151#define DST_MAC_10             0x10
152
153#define TCP_HEAD_LEN           20
154#define UDP_HEAD_LEN           8
155#define IP_HEAD_LEN            20
156#define ETH_HEAD_LEN           14
157
158#define PROTOCOL_UDP           0x11
159#define PROTOCOL_TCP           0x06
160
161#define TCP_ISS                0x10000
162
163#define PAYLOAD_MAX_LEN        1500         // max payload for and UDP packet or a TCP segment
164
165#define TCP_FLAG_FIN           0x01
166#define TCP_FLAG_SYN           0x02
167#define TCP_FLAG_RST           0x04
168#define TCP_FLAG_PSH           0x08
169#define TCP_FLAG_ACK           0x10
170#define TCP_FLAG_URG           0x20
171
172#define NIC_RX_BUF_SIZE        0x100000     // 1 Mbytes
173#define NIC_R2T_QUEUE_SIZE     0x64         // smallest KCM size
174#define NIC_CRQ_QUEUE_SIZE     0x8          // 8 * sizeof(sockaddr_t) = smallest KCM size
175#define NIC_PKT_MAX_SIZE       1500         // for Ethernet
176#define NIC_KERNEL_BUF_SIZE    2000         // for on ETH/IP/TCP packet
177
178/*****************************************************************************************
179 * This defines the extension for the generic NIC device.
180 * The actual queue descriptor depends on the implementation.
181 *
182 * WARNING : for all NIC_TX and NIC_RX chdevs, the xlist rooted in in the chdev
183 *           ("wait_root" and "wait_lock" fields) is actually a list of sockets.
184 ****************************************************************************************/
185
186typedef struct nic_extend_s
187{
188    void         * queue;       /*! local pointer on NIC queue descriptor (RX or TX)    */
189}
190nic_extend_t;
191
192/*****************************************************************************************
193 * This enum defines the various implementations of the generic NIC peripheral.
194 * This array must be kept consistent with the define in the arch_info.h file.
195 ****************************************************************************************/
196
197typedef enum nic_impl_e
198{
199    IMPL_NIC_CBF =   0,     
200    IMPL_NIC_I86 =   1,
201}
202nic_impl_t;
203
204/****************************************************************************************
205 * This defines the (implementation independant) commands to access the NIC hardware.
206 * These commands are registered by the NIC_TX and NIC_RX server threads in the
207 * server thread descriptor, to be used by the NIC driver.
208 * The buffer is always a 2K bytes kernel buffer, containing an Ethernet packet.
209 ****************************************************************************************/
210
211typedef enum nic_cmd_e
212{
213    NIC_CMD_WRITABLE  = 10,   /*! test TX queue not full (for a given packet length)    */
214    NIC_CMD_WRITE     = 11,   /*! put one (given length) packet to TX queue             */
215    NIC_CMD_READABLE  = 12,   /*! test RX queue not empty (for any packet length)       */
216    NIC_CMD_READ      = 13,   /*! get one (any length) packet from RX queue             */
217}
218nic_cmd_t;
219
220typedef struct nic_command_s
221{
222    xptr_t      dev_xp;       /*! extended pointer on NIC chdev descriptor              */
223    nic_cmd_t   type;         /*! command type                                          */
224    uint8_t   * buffer;       /*! local pointer on buffer (kernel or user space)        */ 
225    uint32_t    length;       /*! number of bytes in buffer                             */
226    uint32_t    status;       /*! return value (depends on command type)                */
227    uint32_t    error;        /*! return an error from the hardware (0 if no error)     */
228}
229nic_command_t;
230
231/*****************************************************************************************
232 * This structure defines a socket descriptor. In order to parallelize the transfers,
233 * the set of all registered sockets is split in several subsets.
234 * The number of subsets is the number of  NIC channels.
235 * The distribution key is computed from the (remote_addr/remote_port) couple.
236 * This computation is done by the NIC hardware for RX packets,
237 * and by the dev_nic_connect() function for the TX packets.
238 *
239 * A socket is attached to the NIC_TX[channel] & NIC_RX[channel] chdevs.
240 * Each socket descriptor allows the TX and TX server threads to access various buffers:
241 * - the user "send" buffer contains the data to be send by the TX server thread.
242 * - the kernel "receive" buffer contains the data received by the RX server thread.
243 * - the kernel "r2t" buffer allows the RX server thread to make direct requests
244 *   to the associated TX server (to implement the TCP 3 steps handshake).
245 *
246 * The synchronisation mechanism between the clients threads and the servers threads
247 * is different for TX and RX transfers:
248 *
249 * 1) For a TX transfer, it can exist only one client thread for a given socket,
250 *    the transfer is always initiated by the local process, and all TX commands
251 *    (CONNECT/SEND/CLOSE) are blocking for the client thread. The user buffer is
252 *    used by TCP to handle retransmissions when required.in case of re
253 *    The client thread registers the command in the thread descriptor, registers itself
254 *    in the socket descriptor, unblocks the TX server thread from the BLOCKED_CLIENT
255 *    condition, blocks itself on the BLOCKED_IO condition, and deschedules.
256 *    When the command is completed, the TX server thread unblocks the client thread.
257 *    The TX server blocks itself on the BLOCKED_CLIENT condition, when there is no
258 *    pending commands and the R2T queue is empty. It is unblocked when a client
259 *    register a new command, or when the TX server thread register a mew request
260 *    in the R2T queue.
261 *    The tx_valid flip-flop is SET by the client thread to signal a valid command.
262 *    It is RESET by the server thread when the command is completed: For a SEND,
263 *    all bytes have been sent (UDP) or acknowledged (TCP).
264 *
265 * 2) For an RX transfer, it can exist only one client thread for a given socket,
266 *    but the transfer is initiated by the remote process, and the  RECV command
267 *    is not really blocking: the data can arrive before the local RECV command is
268 *    executed, and the server thread does not wait to receive all requested data
269 *    to deliver data to client thread. Therefore each socket contains a receive
270 *    buffer (rx_buf) handled as a single-writer/single-reader fifo.
271 *    The client thread consumes data from the rx_buf when possible. It blocks on the
272 *    BLOCKED_IO condition and deschedules when the rx_buf is empty.
273 *    It is unblocked by the RX server thread when new data is available in the rx_buf.
274 *    The RX server blocks itself on the BLOCKED_ISR condition When the NIC_RX packets
275 *    queue is empty. It is unblocked by the hardware when new packets are available.
276 *
277 * Note : the socket domains and types are defined in the "shared_socket.h" file.
278 ****************************************************************************************/
279
280/******************************************************************************************
281 * This function returns a printable string for a given NIC command <type>.
282 ******************************************************************************************
283 * @ type    : NIC command type
284 *****************************************************************************************/
285char * nic_cmd_str( uint32_t type );
286   
287/******************************************************************************************
288 * This function returns a printable string for a given socket <state>.
289 ******************************************************************************************
290 * @ state    : socket state
291 *****************************************************************************************/
292char * socket_state_str( uint32_t state );
293
294/******************************************************************************************
295 * This function completes the NIC-RX and NIC-TX chdev descriptors initialisation.
296 * namely the link with the implementation specific driver.
297 * The func, impl, channel, is_rx, base fields have been previously initialised.
298 * It calls the specific driver initialisation function, to initialise the hardware
299 * device and the specific data structures when required.
300 * It creates the associated server thread and allocates a WTI from local ICU.
301 * For a TX_NIC chedv, it allocates and initializes the R2T waiting queue used by the
302 * NIC_RX[channel] server to send direct requests to the NIC_TX[channel] server.
303 * It must de executed by a local thread.
304 ******************************************************************************************
305 * @ chdev    : local pointer on NIC chdev descriptor.
306 *****************************************************************************************/
307void dev_nic_init( struct chdev_s * chdev );
308
309
310/* functions implementing the socket API */
311
312/****************************************************************************************
313 * This function implements the socket() syscall.
314 * This function allocates and intializes in the calling thread cluster:
315 * - a new socket descriptor, defined by the <domain> and <type> arguments,
316 * - a new file descriptor, associated to this socket,
317 * It registers the file descriptor in the reference process fd_array[], set
318 * the socket state to IDLE, and returns the <fdid> value.
319 ****************************************************************************************
320 * @ domain  : [in] socket protocol family (AF_UNIX / AF_INET)
321 * @ type    : [in] socket type (SOCK_DGRAM / SOCK_STREAM).
322 * @ return a file descriptor <fdid> if success / return -1 if failure.
323***************************************************************************************/
324int dev_nic_socket( uint32_t   domain,
325                    uint32_t   type );
326
327/****************************************************************************************
328 * This function implements the bind() syscall.
329 * It initializes the  "local_addr" and "local_port" fields in the socket
330 * descriptor identified by the <fdid> argument and set the socket state to BOUND.
331 * It can be called by a thread running in any cluster.
332 ****************************************************************************************
333 * @ fdid      : [in] file descriptor identifying the socket.
334 * @ addr      : [in] local IP address.
335 * @ port      : [in] local port.
336 * @ return 0 if success / return -1 if failure.
337 ***************************************************************************************/
338int dev_nic_bind( uint32_t fdid,
339                  uint32_t addr,
340                  uint16_t port );
341
342/****************************************************************************************
343 * This function implements the listen() syscall().
344 * It is called by a (local) server process to specify the max size of the queue
345 * registering the (remote) client process connections, and set the socket identified
346 * by the <fdid> argument to LISTEN state. It applies only to sockets of type TCP.
347 * It can be called by a thread running in any cluster.
348 * TODO handle the <max_pending> argument...
349 ****************************************************************************************
350 * @ fdid        : [in] file descriptor identifying the local server socket.
351 * @ max_pending : [in] max number of accepted remote client connections.
352 ***************************************************************************************/
353int dev_nic_listen( uint32_t fdid,
354                    uint32_t max_pending );
355
356/****************************************************************************************
357 * This function implements the connect() syscall.
358 * It is used by a (local) client process to connect a local socket identified by
359 * the <fdid> argument, to a remote socket identified by the <remote_addr> and
360 * <remote_port> arguments. It can be used for both  UDP and TCP sockets.
361 * It computes the nic_channel index from <remote_addr> and <remote_port> values,
362 * and initializes "remote_addr","remote_port", "nic_channel" in local socket.
363 * It registers the socket in the two lists of clients rooted in the NIC_RX[channel]
364 * and NIC_TX[channel] chdevs.  It can be called by a thread running in any cluster.
365 * WARNING : the clients are the socket descriptors, and NOT the threads descriptors.
366 ****************************************************************************************
367 * Implementation Note:
368 * - For a TCP socket, it updates the "remote_addr", "remote_port", "nic_channel" fields
369 *   in the socket descriptor defined by the <fdid> argument, and register this socket,
370 *   in the lists of sockets attached to the NIC_TX and NIC_RX chdevs.
371 *   Then, it registers a CONNECT command in the "nic_cmd" field ot the client thread
372 *   descriptor to request the NIC_TX server thread to execute the 3 steps handshake,
373 *   and updates the "tx_client" field in the socket descriptor. It unblocks the NIC_TX
374 *   server thread, blocks on the THREAD_BLOCKED_IO condition and deschedules.
375 * - For an UDP socket, it simply updates "remote_addr", "remote_port", "nic_channel"
376 *   in the socket descriptor defined by the <fdid> argument, and register this socket,
377 *   in the lists of sockets attached to the NIC_TX and NIC_RX chdevs.
378 *   Then, it set the socket state to CONNECT, without unblocking the NIC_TX server
379 *   thread, and without blocking itself.
380 * TODO : the nic_channel index computation must be done by a driver specific function.
381 ****************************************************************************************
382 * @ fdid          : [in] file descriptor identifying the socket.
383 * @ remote_addr   : [in] remote IP address.
384 * @ remote_port   : [in] remote port.
385 * @ return 0 if success / return -1 if failure.
386 ***************************************************************************************/
387int dev_nic_connect( uint32_t  fdid,
388                     uint32_t  remote_addr,
389                     uint16_t  remote_port );
390
391/****************************************************************************************
392 * This function implements the accept() syscall().
393 * It is executed by a server process, waiting for one (or several) client process(es)
394 * requesting a connection on a socket identified by the <fdid> argument.
395 * This socket was previouly created with socket(), bound to a local address with bind(),
396 * and is listening for connections after a listen().
397 * This function extracts the first connection request on the CRQQ queue of pending
398 * requests, creates a new socket with the same properties as the existing socket,
399 * and allocates a new file descriptor for this new socket.
400 * If no pending connections are present on the queue, it blocks the caller until a
401 * connection is present.
402 * The new socket cannot accept more connections, but the original socket remains open.
403 * It returns the new socket <fdid>, and register in the <address> an <port> arguments
404 * the remote client IP address & port. It applies only to sockets of type SOCK_STREAM.
405 ****************************************************************************************
406 * @ fdid         : [in] file descriptor identifying the listening socket.
407 * @ address      : [out] server IP address.
408 * @ port         : [out] server port address length in bytes.
409 * @ return the new socket <fdid> if success / return -1 if failure
410 ***************************************************************************************/
411int dev_nic_accept( uint32_t   fdid,
412                    uint32_t * address,
413                    uint16_t * port );
414
415/****************************************************************************************
416 * This blocking function implements the send() syscall.
417 * It is used to send data stored in the user buffer, identified the <u_buf> and <length>
418 * arguments, to a connected (TCP or UDP) socket, identified by the <fdid> argument.
419 * The work is actually done by the NIC_TX server thread, and the synchronisation
420 * between the client and the server threads uses the "rx_valid" set/reset flip-flop:
421 * The client thread registers itself in the socket descriptor, registers in the queue
422 * rooted in the NIC_TX[index] chdev, set "rx_valid", unblocks the server thread, and
423 * finally blocks on THREAD_BLOCKED_IO, and deschedules.
424 * When the TX server thread completes the command (all data has been sent for an UDP
425 * socket, or acknowledeged for a TCP socket), the server thread reset "rx_valid" and
426 * unblocks the client thread.
427 * This function can be called by a thread running in any cluster.
428 * WARNING : This implementation does not support several concurent SEND/SENDTO commands
429 * on the same socket, as only one TX thread can register in a given socket.
430 ****************************************************************************************
431 * @ fdid      : [in] file descriptor identifying the socket.
432 * @ u_buf     : [in] pointer on buffer containing packet in user space.
433 * @ length    : [in] packet size in bytes.
434 * @ return number of sent bytes if success / return -1 if failure.
435 ***************************************************************************************/
436int dev_nic_send( uint32_t    fdid,
437                  uint8_t   * u_buf,
438                  uint32_t    length );
439
440/****************************************************************************************
441 * This blocking function implements the sendto() syscall.
442 * It registers the <remote_addr> and <remote_port> arguments in the local socket
443 * descriptor, and does the same thing as the dev_nic_send() function above,
444 * but can be called  on an unconnected UDP socket.
445 ****************************************************************************************
446 * @ fdid        : [in] file descriptor identifying the socket.
447 * @ u_buf       : [in] pointer on buffer containing packet in user space.
448 * @ length      : [in] packet size in bytes.
449 * @ remote_addr : [in] destination IP address.
450 * @ remote_port : [in] destination port.
451 * @ return number of sent bytes if success / return -1 if failure.
452 ***************************************************************************************/
453int dev_nic_sendto( uint32_t    fdid,
454                    uint8_t   * u_buf,
455                    uint32_t    length,
456                    uint32_t    remote_addr,
457                    uint32_t    remote_port );
458
459/****************************************************************************************
460 * This blocking function implements the recv() syscall.
461 * It is used to receive data that has been stored by the NIC_RX server thread in the
462 * rx_buf of a connected (TCP or UDP) socket, identified by the <fdid> argument.
463 * The synchronisation between the client and the server threads uses the "rx_valid"
464 * set/reset flip-flop: If "rx_valid" is set, the client simply moves the available
465 * data from the "rx_buf" to the user buffer identified by the <u_buf> and <length>
466 * arguments, and reset the "rx_valid" flip_flop. If "rx_valid" is not set, the client
467 * thread register itself in the socket descriptor, registers in the clients queue rooted
468 * in the NIC_RX[index] chdev, and finally blocks on THREAD_BLOCKED_IO, and deschedules.
469 * The client thread is re-activated by the RX server, that set the "rx_valid" flip-flop
470 * as soon as data is available in the "rcv_buf" (can be less than the user buffer size).
471 * This  function can be called by a thread running in any cluster.
472 * WARNING : This implementation does not support several concurent RECV/RECVFROM
473 * commands on the same socket, as only one RX thread can register in a given socket.
474 ****************************************************************************************
475 * @ fdid      : [in] file descriptor identifying the socket.
476 * @ u_buf     : [in] pointer on buffer in user space.
477 * @ length    : [in] buffer size in bytes.
478 * @ return number of received bytes if success / return -1 if failure.
479 ***************************************************************************************/
480int dev_nic_recv( uint32_t    fdid,
481                  uint8_t   * u_buf,
482                  uint32_t    length );
483
484/****************************************************************************************
485 * This blocking function implements the recvfrom() syscall.
486 * It registers the <remote_addr> and <remote_port> arguments in the local socket
487 * descriptor, and does the same thing as the dev_nic_recv() function above,
488 * but can be called  on an unconnected UDP socket.
489 ****************************************************************************************
490 * @ fdid        : [in] file descriptor identifying the socket.
491 * @ u_buf       : [in] pointer on buffer containing packet in user space.
492 * @ length      : [in] packet size in bytes.
493 * @ remote_addr : [in] destination IP address.
494 * @ remote_port : [in] destination port.
495 * @ return number of received bytes if success / return -1 if failure.
496 ***************************************************************************************/
497int dev_nic_recvfrom( uint32_t    fdid,
498                      uint8_t   * u_buf,
499                      uint32_t    length,
500                      uint32_t    remote_addr,
501                      uint32_t    remote_port );
502
503
504/*  Instrumentation functions */
505
506
507/******************************************************************************************
508 * This instrumentation function displays on the TXT0 kernel terminal the content
509 * of the instrumentation registers contained in the NIC device.
510 *****************************************************************************************/
511void dev_nic_print_stats( void );
512
513/******************************************************************************************
514 * This instrumentation function reset all instrumentation registers contained
515 * in the NIC device.
516 *****************************************************************************************/
517void dev_nic_clear_stats( void );
518
519
520/* Functions executed by the TX and RX server threads */ 
521 
522/******************************************************************************************
523 * This function is executed by the server thread associated to a NIC_TX[channel] chdev.
524 * This TX server thread is created by the dev_nic_init() function.
525 * It build and send UDP packets or TCP segments for all clients threads registered in
526 * the NIC_TX[channel] chdev. The command types are (CONNECT / SEND / CLOSE), and the
527 * priority between clients is round-robin. It takes into account the request registered
528 * by the RX server thread in the R2T queue associated to the involved socket.
529 * When a command is completed, it unblocks the client thread. For a SEND command, the
530 * last byte must have been sent for an UDP socket, and it must have been acknowledged
531 * for a TCP socket.
532 * When the TX client threads queue is empty, it blocks on THREAD_BLOCKED_CLIENT
533 * condition and deschedules. It is re-activated by a client thread registering a command.
534 ******************************************************************************************
535 * Implementation note:
536 * It execute an infinite loop in which it takes the lock protecting the clients list
537 * to build a "kleenex" list of currently registered clients.
538 * For each client registered in this "kleenex" list, it takes the lock protecting the
539 * socket state, build one packet/segment in a local 2K bytes kernel buffer, calls the
540 * transport layer to add the UDP/TCP header, calls the IP layer to add the IP header,
541 * calls the ETH layer to add the ETH header, and moves the packet to the NIC_TX_QUEUE.
542 * Finally, it updates the socket state, and release the socket lock.
543 ******************************************************************************************
544 * @ chdev    : [in] local pointer on one local NIC_TX[channel] chdev descriptor.
545 *****************************************************************************************/
546void dev_nic_tx_server( struct chdev_s * chdev );
547
548
549/******************************************************************************************
550 * This function is executed by the server thread associated to a NIC_RX[channel] chdev.
551 * This RX server thread is created by the dev_nic_init() function.
552 * It handles all UDP packets or TCP segments received by the sockets attached to
553 * the NIC_RX[channel] chdev. It writes the received data in the socket rcv_buf, and
554 * unblocks the client thread waiting on a RECV command.
555 * To implement the three steps handshahke required by a TCP connection, it posts direct
556 * requests to the TX server, using the R2T queue attached to the involved socket.
557 * It blocks on the THREAD_BLOCKED_ISR condition and deschedules when the NIC_RX_QUEUE
558 * is empty. It is re-activated by the NIC_RX_ISR, when the queue becomes non empty.
559 ******************************************************************************************
560 * Implementation note:
561 * It executes an infinite loop in which it extracts one packet from the NIC_RX_QUEUE
562 * of received packets, copies this packet in a local 2 kbytes kernel buffer, checks
563 * the Ethernet header, checks the IP header, calls the relevant (TCP or UDP) transport
564 * protocol that search a matching socket for the received packet. It copies the payload
565 * to the relevant socket rcv_buf when the packet is acceptable, and unblocks the client
566 * thread. It discard the packet if no socket found.
567 ******************************************************************************************
568 * @ chdev    : [in] local pointer on one local NIC_RX[channel] chdev descriptor.
569 *****************************************************************************************/
570void dev_nic_rx_server( struct chdev_s * chdev );
571
572#endif  /* _DEV_NIC_H */
Note: See TracBrowser for help on using the repository browser.