| 1 |
1629 |
jcastillo |
|
| 2 |
|
|
Ongoing things.
|
| 3 |
|
|
|
| 4 |
|
|
0.0
|
| 5 |
|
|
---
|
| 6 |
|
|
Initial patches to catch up with things we want to add.
|
| 7 |
|
|
|
| 8 |
|
|
o Merged in the Jorge Cwik fast checksum. [TESTED]
|
| 9 |
|
|
o Added Arnt Gulbrandsen's fast UDP build. [TESTED]
|
| 10 |
|
|
o Pauline Middelinks masquerade patch [TESTED]
|
| 11 |
|
|
|
| 12 |
|
|
|
| 13 |
|
|
0.1
|
| 14 |
|
|
---
|
| 15 |
|
|
|
| 16 |
|
|
o Remove excess transmit request calls. [TESTED]
|
| 17 |
|
|
o Set type before calling netif_rx() [TESTED]
|
| 18 |
|
|
o Inline sock_rcv_skb [TESTED]
|
| 19 |
|
|
o Cache last socket for UDP [TESTED]
|
| 20 |
|
|
o Cache last socket for TCP [TESTED]
|
| 21 |
|
|
o Routing cache (only in ip_build_header so far) [TESTED]
|
| 22 |
|
|
------->>>>> ALPHA 001 <<<<<----------
|
| 23 |
|
|
o eql load balancing driver. [TESTED]
|
| 24 |
|
|
o Token ring drivers. [TESTED]
|
| 25 |
|
|
o IPIP and tunnels [TESTED]
|
| 26 |
|
|
o Fix Ethernet/token ring promisc broadcast error [TESTED]
|
| 27 |
|
|
(pkt_type set to OTHERHOST in error).
|
| 28 |
|
|
o Fixed bug in the routing caches [TESTED]
|
| 29 |
|
|
o Protocol header cache support [TESTED]
|
| 30 |
|
|
o Fixed ip_build_xmit loopback bugs [TESTED]
|
| 31 |
|
|
o Fixes for SIOCGSTAMP on SOCK_PACKET [TESTED]
|
| 32 |
|
|
o Perfect hash on net_bh(). [TESTED]
|
| 33 |
|
|
o Sonix ISDN driver. [SEPARATED/SENT]
|
| 34 |
|
|
o Use ip_build_xmit for raw sockets [TESTED]
|
| 35 |
|
|
o 3c501 fixed for speed [TESTED]
|
| 36 |
|
|
------->>>>> ALPHA 002 <<<<<--------
|
| 37 |
|
|
o PLIP, PPP and de bugs fixed [TESTED]
|
| 38 |
|
|
o Merged in G4KLX AX.25 state machine, with KA9Q
|
| 39 |
|
|
donated headers to get BSD free AX.25 done. [TESTED]
|
| 40 |
|
|
o Massive loopback device bug fixed [TESTED]
|
| 41 |
|
|
------->>>>> ALPHA 003 <<<<<----------
|
| 42 |
|
|
o Revised code layout [TESTED]
|
| 43 |
|
|
o More bug fixes (traceroute etc) [TESTED]
|
| 44 |
|
|
------->>>>> ALPHA 004 <<<<<----------
|
| 45 |
|
|
o IP build xmit fragment fixes [TESTED]
|
| 46 |
|
|
o TCP SWS probe fix [TESTED]
|
| 47 |
|
|
o Appletalk DDP [TESTED]
|
| 48 |
|
|
o IP firewall bug fixed [TESTED]
|
| 49 |
|
|
o IP masquerade ftp port spoof [TESTED]
|
| 50 |
|
|
o gcc 2.6.3 -O3 fix for checksum assembler [TESTED]
|
| 51 |
|
|
o /proc support shows both timers [TESTED]
|
| 52 |
|
|
o TCP irtt support [TESTED]
|
| 53 |
|
|
o RTF_REJECT routing support [TESTED]
|
| 54 |
|
|
o Fixed 0 length fragment bug [TESTED]
|
| 55 |
|
|
o Fixed overlapping reasm bug [TESTED]
|
| 56 |
|
|
o Newest AX.25 code from John Naylor [TESTED]
|
| 57 |
|
|
o NetROM from John Naylor [TESTED]
|
| 58 |
|
|
o Routerless DDP fixes from Wesley [TESTED]
|
| 59 |
|
|
|
| 60 |
|
|
------->>>>> ALPHA 005 <<<<<----------
|
| 61 |
|
|
|
| 62 |
|
|
o Several compile and bugfixes from Jakko [TESTED]
|
| 63 |
|
|
o Connect fix from Matt Day (+ fix to fix) [TESTED]
|
| 64 |
|
|
o RTT, memory leak and other netrom/ax.25 cures
|
| 65 |
|
|
-- John Naylor [TESTED]
|
| 66 |
|
|
o IP source route via broadcast now illegal [TESTED]
|
| 67 |
|
|
|
| 68 |
|
|
------->>>>> ALPHA 006 <<<<<----------
|
| 69 |
|
|
|
| 70 |
|
|
o Yet more NetROM/AX.25 improvements [TESTED]
|
| 71 |
|
|
-- John Naylor
|
| 72 |
|
|
o Fixed a _stupid_ appletalk bug [TESTED]
|
| 73 |
|
|
o Missing include [TESTED]
|
| 74 |
|
|
-- Lots of people
|
| 75 |
|
|
o Can drop all source routes [TESTED]
|
| 76 |
|
|
o Printing fixes for ip_fw [TESTED]
|
| 77 |
|
|
o UDP checksum fix (Gerhard) [TESTED]
|
| 78 |
|
|
o Newer 3c505 driver from Juha Laiho [IN]
|
| 79 |
|
|
o Security fix to axassociate [TESTED]
|
| 80 |
|
|
o Loopback driver debugged (fixes named) [TESTED]
|
| 81 |
|
|
o SCC driver from Joerg Reuter [TESTED]
|
| 82 |
|
|
o IP Firewall accounting zero bug [TESTED]
|
| 83 |
|
|
|
| 84 |
|
|
////////////////////////////1.3.0///////////////////////////
|
| 85 |
|
|
|
| 86 |
|
|
|
| 87 |
|
|
o Merged loadable firewall code [TESTED]
|
| 88 |
|
|
o New buffers used totally non optimally [TESTED]
|
| 89 |
|
|
o Fast ip_forwarding (needs changing) [NOW INCLUDED IN 1.3.15]
|
| 90 |
|
|
o Fixed connection hang bug in new SWS code [TESTED]
|
| 91 |
|
|
o Buffer management hack putting skbuff control
|
| 92 |
|
|
after data in the frame because kmalloc is
|
| 93 |
|
|
totally cache non-optimal [TESTED]
|
| 94 |
|
|
o Faster checksum [Tom May] [TESTED]
|
| 95 |
|
|
o Appletalk router fixes [Michael Callahan] [TESTED]
|
| 96 |
|
|
o TCP state error fixes [Mark Tamsky] [TESTED]
|
| 97 |
|
|
o Verify area fixes [Heiko Eissfeldt] [TESTED]
|
| 98 |
|
|
o Routes use metric field [John Naylor] [TESTED]
|
| 99 |
|
|
o Major AX.25/NetROM fixes [John Nalor] [TESTED]
|
| 100 |
|
|
|
| 101 |
|
|
------->>>>> NET3 030 <<<<<----------
|
| 102 |
|
|
|
| 103 |
|
|
o Long word align Ethernet IP headers (64byte align for Pentium) [TESTED]
|
| 104 |
|
|
(less helpful than I'd have liked)
|
| 105 |
|
|
o Fixed variable length header support to really work [TESTED]
|
| 106 |
|
|
o Mend appletalk/ipx partially [TESTED]
|
| 107 |
|
|
o Start playing with input checksum & copy [TESTED]
|
| 108 |
|
|
o Fixed PPP and other oddments [TESTED]
|
| 109 |
|
|
o Mended IPIP [TESTED]
|
| 110 |
|
|
|
| 111 |
|
|
------->>>>> 1.3.7 <<<<<----------
|
| 112 |
|
|
|
| 113 |
|
|
o Checksum bug fixed [TESTED]
|
| 114 |
|
|
o Lance driver panic cured [TESTED]
|
| 115 |
|
|
o DEC ALPHA stuff (Linus) [ASK HIM NOT ME]
|
| 116 |
|
|
o Always try to keep output packet order
|
| 117 |
|
|
(eg for vat and BSD fast path tcp) [TESTED]
|
| 118 |
|
|
o Copy the mac pointer in skb_clone [TESTED]
|
| 119 |
|
|
o Fix tcpdump panic [TESTED]
|
| 120 |
|
|
o Fix dev_alloc_skb NULL deref bug [TESTED]
|
| 121 |
|
|
o Fix Security error in SIGURG stuff [TESTED]
|
| 122 |
|
|
o Missing 15 byte slack on ip_loopback [TESTED]
|
| 123 |
|
|
|
| 124 |
|
|
------->>>>> 1.3.8 <<<<<----------
|
| 125 |
|
|
|
| 126 |
|
|
o UDP snmp count fixed [TESTED]
|
| 127 |
|
|
o IP snmp out count fixed [TESTED]
|
| 128 |
|
|
o First bit of Dave Bonn's fast forwarding [TESTED/NOW WORKS]
|
| 129 |
|
|
o Fix leaks and double free in firewalling [TESTED]
|
| 130 |
|
|
o Fix memory scribble in ip_build_xmit [TESTED]
|
| 131 |
|
|
o Do fast cases of ip_build_xmit first
|
| 132 |
|
|
slows fragmented I/O down, speeds up smaller
|
| 133 |
|
|
packets. UDP send ttcp can now touch 7.5Mbyte/sec
|
| 134 |
|
|
with nothing else going on. UDP recv is slower 8( [TESTED]
|
| 135 |
|
|
o Fixed and enabled Ethernet header caches [TESTED]
|
| 136 |
|
|
o Removed junk from igmp [TESTED]
|
| 137 |
|
|
o Obscure UDP/copy&sum bug fix [TESTED]
|
| 138 |
|
|
o Fixed multicast [TESTED]
|
| 139 |
|
|
o TCP does rerouting for most cases [TESTED]
|
| 140 |
|
|
|
| 141 |
|
|
------->>>>> 1.3.14 <<<<<----------
|
| 142 |
|
|
|
| 143 |
|
|
o IPX works [TESTED]
|
| 144 |
|
|
o NetROM works [TESTED]
|
| 145 |
|
|
o AX.25 works [TESTED]
|
| 146 |
|
|
o Most modules need recompiling even though they
|
| 147 |
|
|
load OK [BLAME LINUS]
|
| 148 |
|
|
o Appletalk works nicely [CHECKED]
|
| 149 |
|
|
o Fast IP forwarding part 1 works [CHECKED]
|
| 150 |
|
|
|
| 151 |
|
|
------->>>>> 1.3.15 <<<<<---------
|
| 152 |
|
|
o Mike Shaver has started RFC1122 verification [DONE]
|
| 153 |
|
|
o Minor bug fixes [TESTED]
|
| 154 |
|
|
|
| 155 |
|
|
------->>>> 1.3.16 <<<--------
|
| 156 |
|
|
|
| 157 |
|
|
o Missing patches for device change in TCP [TESTED]
|
| 158 |
|
|
o Device locking [TESTED]
|
| 159 |
|
|
o Infinite slip devices [TESTED]
|
| 160 |
|
|
o New AF_UNIX sockets [TESTED]
|
| 161 |
|
|
o Sendmsg/recvmsg (for some stuff only) [TESTED]
|
| 162 |
|
|
o Device unload loopholes fixed [TESTED]
|
| 163 |
|
|
o Extra firewall abilities [TESTED]
|
| 164 |
|
|
o Appletalk node probe bug fix [TESTED]
|
| 165 |
|
|
|
| 166 |
|
|
------->>>> 1.3.18 <<<<---------
|
| 167 |
|
|
|
| 168 |
|
|
o AF_UNIX debugging [TESTED]
|
| 169 |
|
|
o TCP explode on SWS bug fix [TESTED]
|
| 170 |
|
|
o John Naylor's ARP hwtype fix [TESTED]
|
| 171 |
|
|
o Datagram select matches BSD semantics [TESTED]
|
| 172 |
|
|
|
| 173 |
|
|
-------->>>>> 1.3.21 <<<<<---------
|
| 174 |
|
|
|
| 175 |
|
|
o AF_UNIX fixes and 4K limiter [TESTED]
|
| 176 |
|
|
o Sendmsg/recvmsg for AX.25/Appletalk [TESTED]
|
| 177 |
|
|
o Datagram generic iovec support [TESTED]
|
| 178 |
|
|
o Misc minor bug fixes [TESTED]
|
| 179 |
|
|
|
| 180 |
|
|
-------->>>>> 1.3.22 <<<<<-------
|
| 181 |
|
|
|
| 182 |
|
|
o Device lock against page fault [TESTED]
|
| 183 |
|
|
o IP_HDRINCL [TESTED]
|
| 184 |
|
|
o IP firewalling spoofing protection [TESTED]
|
| 185 |
|
|
o IGMP bug fixes and workarounds [TESTED]
|
| 186 |
|
|
o IFF_ALLMULTI protocol layer support [TESTED]
|
| 187 |
|
|
o First parts of IP multicast routing code [TESTED]
|
| 188 |
|
|
o Generate BSD ENETDOWN errors [TESTED]
|
| 189 |
|
|
o Clean device unload bug [TESTED]
|
| 190 |
|
|
|
| 191 |
|
|
-------->>>>> 1.3.24 <<<<<-------
|
| 192 |
|
|
|
| 193 |
|
|
o Missing IGMP includes fixes [TESTED]
|
| 194 |
|
|
o Smarter buffer use options for sockets [TESTED]
|
| 195 |
|
|
o AF_UNIX smarter buffer driving [TESTED]
|
| 196 |
|
|
o AF_UNIX full BSD semantics on STREAM writes [TESTED]
|
| 197 |
|
|
o IOVEC's support repeated calls to copy more [TESTED]
|
| 198 |
|
|
o Zero fragment 'Solaris NFS' bug fixed [TESTED]
|
| 199 |
|
|
o NetROM supports sendmsg/recvmsg [TESTED]
|
| 200 |
|
|
o Sendmsg verify_iovec bugfix [TESTED]
|
| 201 |
|
|
o ARP PERM is really permanent now [TESTED]
|
| 202 |
|
|
o IPIP tunnels work fully we hope [UMM...]
|
| 203 |
|
|
o Closing socket change (Marc Tamsky) [TESTED]
|
| 204 |
|
|
o RFC1122 verification of tcp.c [DONE]
|
| 205 |
|
|
|
| 206 |
|
|
-------->>>>> 1.3.26 <<<<<-------
|
| 207 |
|
|
|
| 208 |
|
|
o Rewrote ICMP completely [TESTED]
|
| 209 |
|
|
o Moved most IP addresses to __u32 [TESTED]
|
| 210 |
|
|
o Cleaned up ICMP reporting [TESTED]
|
| 211 |
|
|
o Tidied remove_sock [TESTED]
|
| 212 |
|
|
o Added memory allocation type to ip_build_xmit [TESTED]
|
| 213 |
|
|
o Cleaned up af_inet to use inet_error [TESTED]
|
| 214 |
|
|
o Named firewall returns [TESTED]
|
| 215 |
|
|
o Added firewall output checks to ip_build_xmit [TESTED]
|
| 216 |
|
|
o Multicast router downs VIF's when their
|
| 217 |
|
|
physical interface is dropped [TESTED]
|
| 218 |
|
|
o Reformatted ipv4/protocol.c, dropped frag field [TESTED]
|
| 219 |
|
|
o Fixed MSS for TCP [TESTED]
|
| 220 |
|
|
o Dropped sock_awaitconn [TESTED]
|
| 221 |
|
|
o Added ip_forward to ksyms for IPIP etc [TESTED]
|
| 222 |
|
|
o Appletalk TIOCINQ/TIOCOUTQ bug fix [TESTED]
|
| 223 |
|
|
o Rewrote the IFF_UP/IFF_DOWN handling code [TESTED]
|
| 224 |
|
|
|
| 225 |
|
|
-------->>>>> 1.3.29 <<<<<-------
|
| 226 |
|
|
|
| 227 |
|
|
o Major AX.25/NetROM fixes [John Naylor] [TESTED]
|
| 228 |
|
|
o Error in ip_mr ioctls fixed [Michael Chastain] [TESTED]
|
| 229 |
|
|
o TCP cache zap bugs hopefully fixed [CLOSE BUT NO COOKIE]
|
| 230 |
|
|
o Length checks in udp/raw sending [Craig Metz] [TESTED]
|
| 231 |
|
|
|
| 232 |
|
|
-------->>>>> 1.3.31 <<<<<<-------
|
| 233 |
|
|
|
| 234 |
|
|
o IP_OPTIONS [A.N.Kuznetsov] [TESTED]
|
| 235 |
|
|
o TCP cache zap more fixes [TESTED]
|
| 236 |
|
|
o Most of the IP multicast routing cache added [TESTED - WORK NEEDED]
|
| 237 |
|
|
o Kernel/user communication module (not used yet) [TESTED]
|
| 238 |
|
|
|
| 239 |
|
|
-------->>>>> 1.3.31 <<<<<<-------
|
| 240 |
|
|
|
| 241 |
|
|
o IFF_ALLMULTI support for 3c501,3c509,8390 and
|
| 242 |
|
|
tulip(SMC etherpower) boards [TESTED]
|
| 243 |
|
|
|
| 244 |
|
|
-------->>>>> 1.3.33 <<<<<<--------
|
| 245 |
|
|
|
| 246 |
|
|
o IFF_ALLMULTI causes an address check on ether [TESTED]
|
| 247 |
|
|
o Added multicast ability readme file [TESTED]
|
| 248 |
|
|
o Assorted driver/multicast fixes [TESTED]
|
| 249 |
|
|
o IP routing change errors resemble BSD more [TESTED/MORE TO COME]
|
| 250 |
|
|
o IP port masquerading fixes [TESTED]
|
| 251 |
|
|
|
| 252 |
|
|
-------->>>>> 1.3.35 <<<<<<--------
|
| 253 |
|
|
|
| 254 |
|
|
o Appletalk data now in the protinfo union [TESTED]
|
| 255 |
|
|
o BSD style bind to broadcast address supported [TESTED]
|
| 256 |
|
|
o Standard loadable firewall chains [TESTED]
|
| 257 |
|
|
o IPFW uses the firewall chains for firewall but
|
| 258 |
|
|
not yet acct/masquerade [TESTED]
|
| 259 |
|
|
o Firewall chain hooks in all other protocols [TESTED]
|
| 260 |
|
|
o Sendmsg/recvmsg for IPX. [TESTED]
|
| 261 |
|
|
o IPX uses sock_alloc_send_skb [TESTED]
|
| 262 |
|
|
o Recvmsg for all IP, sendmsg for TCP [TESTED]
|
| 263 |
|
|
(nearly ready to go all *msg())
|
| 264 |
|
|
|
| 265 |
|
|
-------->>>>> 1.3.42 <<<<<<--------
|
| 266 |
|
|
|
| 267 |
|
|
o ip udp/raw nonblock bug fixed [TESTED]
|
| 268 |
|
|
o ICMP lockup fix [TESTED]
|
| 269 |
|
|
o Fundamental operations now only sendmsg/recvmsg [TESTED]
|
| 270 |
|
|
o bind() for SOCK_PACKET [IN]
|
| 271 |
|
|
o set_mac_addr fixed up [TESTED]
|
| 272 |
|
|
o BSD SIOCSIFADDR, AF_UNSPEC behaviour [TESTED]
|
| 273 |
|
|
o Updated this list [OK]
|
| 274 |
|
|
o Massive ARP/cache/routing rewrite [ANK] [TESTED]
|
| 275 |
|
|
o AX.25 connect return fixed in using sock_error [TESTED]
|
| 276 |
|
|
o Proper netlink device major(36) [TESTED]
|
| 277 |
|
|
o First parts of the SKIP support [IN, not useful]
|
| 278 |
|
|
o TCP ICMP (SOSS should work again) [TESTED]
|
| 279 |
|
|
o IPFW support for TOS changing (Al Longyear) [TESTED]
|
| 280 |
|
|
o DECNET PPP test code [Steve] [IN]
|
| 281 |
|
|
o NFS root [Miguel/Gero] [TESTED]
|
| 282 |
|
|
o Path MTU discovery [ANK] [TESTED]
|
| 283 |
|
|
|
| 284 |
|
|
-------->>>>> 1.3.44 <<<<<<--------
|
| 285 |
|
|
|
| 286 |
|
|
o NFS root/ FPU clash fixed [TESTED]
|
| 287 |
|
|
o ARP lock bug fixed [TESTED]
|
| 288 |
|
|
o SO_BSDCOMPAT option(libbsd/ibcs2 ought to set) [SEMIDONE]
|
| 289 |
|
|
o Changed to new set_multicast_list() [TESTED]
|
| 290 |
|
|
o ARP ioctl() call fixes [Bernd] [TESTED]
|
| 291 |
|
|
o Fixes to the name set functions (maybe fixes
|
| 292 |
|
|
netrom) [Steve] [TESTED]
|
| 293 |
|
|
o Packet protocol labelling (not IPX yet) [TESTED]
|
| 294 |
|
|
o Faster buffer copy/clone [Linus] [TESTED]
|
| 295 |
|
|
|
| 296 |
|
|
-------->>>>> 1.3.46 <<<<<<--------
|
| 297 |
|
|
|
| 298 |
|
|
o AX.25/NetROM fixes/changes [John Naylor] [TESTED]
|
| 299 |
|
|
o Further attempts to fix the IPX memory bug [NOW FIXED]
|
| 300 |
|
|
o ARP fixes (Assorted) [TESTED]
|
| 301 |
|
|
o Driver fixes for multicast lists [TESTED]
|
| 302 |
|
|
|
| 303 |
|
|
-------->>>>> 1.3.48 <<<<<<--------
|
| 304 |
|
|
|
| 305 |
|
|
o IPalias [TESTED]
|
| 306 |
|
|
|
| 307 |
|
|
-------->>>>> 1.3.50 <<<<<<--------
|
| 308 |
|
|
|
| 309 |
|
|
o TCP soft error support [TESTED]
|
| 310 |
|
|
o Further 3c501 tweaking [TESTED]
|
| 311 |
|
|
o Still trying to make IPX work right [TESTED]
|
| 312 |
|
|
o Trap faulty boxes sending IGMP using 0.0.0.0 [TESTED]
|
| 313 |
|
|
o Only allow SMBFS selection with IP configured [TESTED]
|
| 314 |
|
|
o Packetwin driver [Craig] [IN]
|
| 315 |
|
|
o Net alias changes [Juan] [TESTED]
|
| 316 |
|
|
|
| 317 |
|
|
-------->>>>> 1.3.53 <<<<<<--------
|
| 318 |
|
|
|
| 319 |
|
|
o Missing htons() in ip_build_xmit [Mike Kilburn] [TESTED]
|
| 320 |
|
|
o Missing protocol type set in appletalk [TESTED]
|
| 321 |
|
|
o Net alias changes/documentation [Juan Ciarlante][TESTED]
|
| 322 |
|
|
o Set protocol type on IPX sends [Various] [TESTED]
|
| 323 |
|
|
o Lance driver packet length sanity check [TESTED]
|
| 324 |
|
|
|
| 325 |
|
|
-------->>>>> 1.3.60 <<<<<<--------
|
| 326 |
|
|
|
| 327 |
|
|
o Fixed NFS notice message [IN]
|
| 328 |
|
|
o Extra ETH_P_xx types [IN]
|
| 329 |
|
|
o Added skb_copy [IN]
|
| 330 |
|
|
o AX.25 unlock bug fix [Joerg] [IN]
|
| 331 |
|
|
o Disabled buggy kerneld support [IN]
|
| 332 |
|
|
o Removed dev_rint [IN]
|
| 333 |
|
|
o Fixed bind checking [IN]
|
| 334 |
|
|
o ARP changes [Bernd] [IN]
|
| 335 |
|
|
o IPX memory leak fix [Martin] [IN]
|
| 336 |
|
|
o Net builds without /proc [Paul] [IN]
|
| 337 |
|
|
o IP multicast races removed [IN]
|
| 338 |
|
|
o Device queue length limits and packet discarder [IN]
|
| 339 |
|
|
|
| 340 |
|
|
---------- Things I thought Linus had for a while and not merged ----------------
|
| 341 |
|
|
|
| 342 |
|
|
|
| 343 |
|
|
---------- Things pending from other people -------------
|
| 344 |
|
|
|
| 345 |
|
|
o Improved IPX support for lwared.
|
| 346 |
|
|
o Decnet pre pre pre pre pre Alpha 0.0.
|
| 347 |
|
|
o Chase Donald for new drivers, get people to sort out what net
|
| 348 |
|
|
drivers should cease to be 'Alpha'.
|
| 349 |
|
|
o IPX PPP support
|
| 350 |
|
|
o IP multicast bug fixes
|
| 351 |
|
|
|
| 352 |
|
|
---------- Things pending for me to merge --------------
|
| 353 |
|
|
|
| 354 |
|
|
o AF_UNIX garbage collect code
|
| 355 |
|
|
o Faster closedown option for heavy use sites (me)
|
| 356 |
|
|
o Tom May's insw_and_checksum()
|
| 357 |
|
|
o SPARC patches [Dave] [partly in]
|
| 358 |
|
|
|
| 359 |
|
|
--------------- Things That Need Doing Before 1.4 ------------------
|
| 360 |
|
|
|
| 361 |
|
|
o Clean up RAW AX.25 sockets. [Sorted by skb_clone change]
|
| 362 |
|
|
o Finish IPIP bug fixes [Done hopefully]
|
| 363 |
|
|
o Multicast routing [Nearly right]
|
| 364 |
|
|
o PPP/IPX
|
| 365 |
|
|
o IPX for Lwared
|
| 366 |
|
|
o SKIP [Available in user mode]
|
| 367 |
|
|
o AX.25/NetROM locking changes
|
| 368 |
|
|
o insw_and_csum
|
| 369 |
|
|
o AF_UNIX fd passing
|
| 370 |
|
|
|
| 371 |
|
|
-------------------------- Bugs to fix ------------------------------
|
| 372 |
|
|
|
| 373 |
|
|
o signal interrupting a unix domain connect can occasionally hang the
|
| 374 |
|
|
machine ??
|
| 375 |
|
|
o TCP socket cache gets things wrong very very occasionally under high
|
| 376 |
|
|
load. [TRYING THINGS]
|
| 377 |
|
|
o AX.25/NetROM needs more locking.
|
| 378 |
|
|
o NFS flow control is needed with the new multirequest NFS support.
|
| 379 |
|
|
o Need to be able to turn off the intelligent arp refreshing as it's not so
|
| 380 |
|
|
hot over AX.25 and upsets some people with very dumb ISDN bridges.
|
| 381 |
|
|
o Matti Arnio's TCP problem.
|
| 382 |
|
|
o Should unix domain connect never block ?
|
| 383 |
|
|
o Sort out kerneld getting things right.
|
| 384 |
|
|
|
| 385 |
|
|
0.2
|
| 386 |
|
|
---
|
| 387 |
|
|
o Fast checksum/copy on outgoing TCP
|
| 388 |
|
|
o Add tty support to sonix driver.
|
| 389 |
|
|
o PPP for Sonix ISDN.
|
| 390 |
|
|
o Screend loadable firewall module
|
| 391 |
|
|
o AXIP [AVAILABLE IN USER MODE]
|
| 392 |
|
|
o Finish merging the bridge code [LEAVE POST 1.4]
|
| 393 |
|
|
o Finish 802.2 Class I code to be compliant to the oddities of 802.2
|
| 394 |
|
|
o Tidy BPQ support to use a bpqip tunnel device
|
| 395 |
|
|
o Kill off old ip_queue_xmit/ip_send stuff.
|
| 396 |
|
|
o Remove kernel RARP and replace with user mode daemon.
|
| 397 |
|
|
o Throw out existing firewall ioctl()'s and use a single table load.
|
| 398 |
|
|
o SPARC merge
|
| 399 |
|
|
|
| 400 |
|
|
0.3
|
| 401 |
|
|
---
|
| 402 |
|
|
o 802.2 Class 2 services (eg netbios).
|
| 403 |
|
|
|
| 404 |
|
|
Possible projects for victim^H^H^H^H^Holunteers
|
| 405 |
|
|
|
| 406 |
|
|
|
| 407 |
|
|
2. Verifying all the error returns match the BSD ones (grotty job I
|
| 408 |
|
|
wouldn't wish on anyone).
|
| 409 |
|
|
|
| 410 |
|
|
3. 'Fast Retransmit'. This is a TCP algorithm extension BSD uses. If
|
| 411 |
|
|
you see about 3 acks in a row that are for the same 'old' value. You resend
|
| 412 |
|
|
the frame following the ack. (The assumption being that the acks are
|
| 413 |
|
|
because a single frame in the data stream has been lost). Given a
|
| 414 |
|
|
mathematician with some queue theory you can show this allows you to
|
| 415 |
|
|
lose one frame per window full of data without measurable speed loss.
|
| 416 |
|
|
|
| 417 |
|
|
4. RFC1323. These are the extensions for very fast nets.
|
| 418 |
|
|
RFC1323 will be useful for Linux talking to systems over 100 Mb/sec
|
| 419 |
|
|
Ethernet and over ATM as it allows large windows and protects from some
|
| 420 |
|
|
potential high speed TCP problems.
|
| 421 |
|
|
|
| 422 |
|
|
6. Delayed ack. This is mostly supported but not actually set up and
|
| 423 |
|
|
used yet. Basically ack frames are held back 1/10th of a second in the hope
|
| 424 |
|
|
that two acks can be merged into one or for interactive use the ack can
|
| 425 |
|
|
piggyback on the next character typed (great improvement on 2400 baud
|
| 426 |
|
|
modems). Johannes Stille did some work on this about 0.99.13 but it never
|
| 427 |
|
|
got merged in. [Pedro Roque]
|
| 428 |
|
|
|
| 429 |
|
|
7. One on my tempting project list. Add an extra (unofficial - but so
|
| 430 |
|
|
is SLIP6) SLIP mode that does packet data compression [maybe use the code
|
| 431 |
|
|
from term].
|
| 432 |
|
|
|
| 433 |
|
|
9. Implementing streams. Not as a blind slow SYS5.4 style copy but actually
|
| 434 |
|
|
working out how to do it so it runs like greased lightning. Quite a big
|
| 435 |
|
|
problem. [See the LiS project]
|
| 436 |
|
|
|
| 437 |
|
|
10. Frame Relay/WAN/ISDN drivers [I'm working on the sonix EuroISDN board
|
| 438 |
|
|
driver but that's for an internal project and its general release is still
|
| 439 |
|
|
a maybe (so is finishing it ;))][Jim Freeman is working on Frame Relay as is
|
| 440 |
|
|
Mike McLagan][Fritz Elfert is doing the isdn4linux kit].
|
| 441 |
|
|
|
| 442 |
|
|
11. IP over SCSI.
|
| 443 |
|
|
|
| 444 |
|
|
14. Bidirectional PLIP. Also PLIP for the newer style parallel ports.
|
| 445 |
|
|
|
| 446 |
|
|
15. 802.2LLC and thus Netbeui sockets. Becoming less important since the
|
| 447 |
|
|
rumour is microsoft are phasing out netbeui for netbios/IP. Microsoft have
|
| 448 |
|
|
gone for netbios/funny-ipx-variant it seems in Win95, but TCP is selectable.
|
| 449 |
|
|
|
| 450 |
|
|
16. X.25. This is one for a real head case with far too much time on
|
| 451 |
|
|
their hands. [Provisionally taken]
|
| 452 |
|
|
|
| 453 |
|
|
17. PPP multilink. Another nasty job.
|
| 454 |
|
|
[In progress]
|
| 455 |
|
|
|
| 456 |
|
|
18. Implement swIPe under Linux.
|
| 457 |
|
|
[Reportedly in progress]
|
| 458 |
|
|
|
| 459 |
|
|
19. IPv4 IP-AH and IP-ESP.
|
| 460 |
|
|
[Taken]
|
| 461 |
|
|
|
| 462 |
|
|
20. SKIP IP security using ENskip-0.10 - started
|
| 463 |
|
|
[Abandoned]
|
| 464 |
|
|
|
| 465 |
|
|
21. T/TCP support.
|
| 466 |
|
|
|
| 467 |
|
|
BTW: Don't let the magic words 'kernel programming' worry you. Its like DOS
|
| 468 |
|
|
- you make a mistake you have to reboot. You do at least get dumps and a
|
| 469 |
|
|
kernel logger that is reliable. There is now a loadable module allowing
|
| 470 |
|
|
use of gdb on the kernel (no breakpoints though!). No magic involved.
|
| 471 |
|
|
|
| 472 |
|
|
Alan
|
