aboutsummaryrefslogtreecommitdiffstats
path: root/include/net
Commit message (Collapse)AuthorAge
...
| * | | | [PATCH] ieee80211: Type-o, capbility definition for QoS, and ERP parsingJames Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | tree 3ac0dd07b9972dfd68fee47ec2152d3d378de000 parent 9ada1d971d9829c34a14d98840080b7e69fdff6b author Mohamed Abbad <mohamed.abbas@intel.com> 1126054379 -0500 committer James Ketrenos <jketreno@linux.intel.com> 1127314340 -0500 Type-o, capbility definition for QoS, and ERP parsing Added WLAN_CAPABILITY_QOS Fixed type-o WLAN_CAPABILITY_OSSS_OFDM -> WLAN_CAPABILITY_DSSS_OFDM Added ERP IE parsing to ieee80211_rx Added handle_probe_request callback. Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | [PATCH] ieee80211: "extern inline" to "static inline"James Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | tree bce04549ce0a8239d8083d8da5c3d12f7e1aecd9 parent b15a5153d5f1c75d9435d5ce19b52287059d5d54 author Adrian Bunk <bunk@stusta.de> 1125026386 -0500 committer James Ketrenos <jketreno@linux.intel.com> 1127313953 -0500 "extern inline" doesn't make much sense. Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | [PATCH] ieee80211: Additional fixes for endian-aware typesJames Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | tree 589bbb92ce7cdf7c2ae820b0ebd3f8fbf1baeee9 parent c6ce9081e79e8836a11e86e3d38297521a2420be author Jiri Benc <jbenc@suse.cz> 1125015310 -0400 committer James Ketrenos <jketreno@linux.intel.com> 1127313914 -0500 Additional fixes for endian-aware types Based on the application of __le16/__be16 changes already made w/ a prior patch by Michael Wu <flamingice@sourmilk.net> Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | [PATCH] ieee80211: Added ieee80211_radiotap.hJames Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | tree 383c59b2516a61f2683f02dfebbed0caf6ee5dc3 parent a04948f63fd96c4b875a43f78afad1a0874cc441 author Mike Kershaw <dragorn@kismetwireless.net> 1124447833 -0500 committer James Ketrenos <jketreno@linux.intel.com> 1127313883 -0500 Added ieee80211_radiotap.h to enhance statistic reporting to user space from wireless drivers. Signed-off-by: Mike Kershaw <dragorn@kismetwireless.net> Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | [PATCH] ieee80211: Added ieee80211_geo to provide helper functionsJames Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | tree 385b391fc0d7c124cd0547fdb6183e9a0c333391 parent 97d7a47f76e72bedde7f402785559ed4c7a8e8e8 author James Ketrenos <jketreno@linux.intel.com> 1124447590 -0500 committer James Ketrenos <jketreno@linux.intel.com> 1127313735 -0500 Added ieee80211_geo to provide helper functions to drivers for implementing supported channel maps. Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | [PATCH] ieee80211: Add QoS (WME) support to the ieee80211 subsystemJames Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | tree a3ad796273e98036eb0e9fc063225070fa24508a parent 1b9c0aeb377abf8e4a43a86cff42382f74ca0259 author Mohamed Abbas <mabbas@linux.intel.com> 1124447069 -0500 committer James Ketrenos <jketreno@linux.intel.com> 1127313435 -0500 Add QoS (WME) support to the ieee80211 subsystem. NOTE: This requires drivers that use the ieee80211 hard_start_xmit (ipw2100 and ipw2200) to add the priority parameter to their callback. Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | [PATCH] ieee80211: Fix TKIP, repeated fragmentation problem, and ↵James Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | payload_size reporting tree 8428e9f510e6ad6c77baec89cb57374842abf733 parent d78bfd3ddae9c422dd350159110f9c4d7cfc50de author Liu Hong <hong.liu@intel.com> 1124446520 -0500 committer James Ketrenos <jketreno@linux.intel.com> 1127313183 -0500 Fix TKIP, repeated fragmentation problem, and payload_size reporting 1. TKIP encryption Originally, TKIP encryption issues msdu + mpdu encryption on every fragment. Change the behavior to msdu encryption on the whole packet, then mpdu encryption on every fragment. 2. Avoid repeated fragmentation when !host_encrypt. We only need do fragmentation when using host encryption. Otherwise we only need pass the whole packet to driver, letting driver do the fragmentation. 3. change the txb->payload_size to correct value FW will use this value to determine whether to do fragmentation. If we pass the wrong value, fw may cut on the wrong bound which will make decryption fail when we do host encryption. NOTE: This requires changing drivers (hostap) that have extra_prefix_len used within them (structure member name change). Signed-off-by: Hong Liu <liu.hong@intel.com> Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | [PATCH] ieee82011: Added ieee80211_tx_frame to convert generic 802.11 data ↵James Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | frames, and callbacks tree 40adc78b623ae70d56074934ec6334eb4f0ae6a5 parent db43d847bcebaa3df6414e26d0008eb21690e8cf author James Ketrenos <jketreno@linux.intel.com> 1124445938 -0500 committer James Ketrenos <jketreno@linux.intel.com> 1127313102 -0500 Added ieee80211_tx_frame to convert generic 802.11 data frames into txbs for transmission. Added several purpose specific callbacks (handle_assoc, handle_auth, etc.) which the driver can register with for being notified on reception of variouf frame elements. Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | [PATCH] ieee80211: adds support for the creation of RTS packetsJames Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | tree b45c9c1017fd23216bfbe71e441aed9aa297fc84 parent 04aacdd71e904656a304d923bdcf57ad3bd2b254 author Ivo van Doorn <IvDoorn@gmail.com> 1124445405 -0500 committer James Ketrenos <jketreno@linux.intel.com> 1127313029 -0500 This patch adds support for the creation of RTS packets when the config flag CFG_IEEE80211_RTS has been set. Signed-Off-By: Ivo van Doorn <IvDoorn@gmail.com> Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | [PATCH] ieee80211: Renamed ieee80211_hdr to ieee80211_hdr_3addrJames Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | tree e9c18b2c8e5ad446a4d213243c2dcf9fd1652a7b parent 4e97ad6ae7084a4f741e94e76c41c68bc7c5a76a author James Ketrenos <jketreno@linux.intel.com> 1124444315 -0500 committer James Ketrenos <jketreno@linux.intel.com> 1127312922 -0500 Renamed ieee80211_hdr to ieee80211_hdr_3addr and modified ieee80211_hdr to just contain the frame_ctrl and duration_id. Changed uses of ieee80211_hdr to ieee80211_hdr_4addr or ieee80211_hdr_3addr based on what was expected for that portion of code. NOTE: This requires changes to ipw2100, ipw2200, hostap, and atmel drivers. Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | [PATCH] ieee82011: Added WE-18 support to default wireless extension handlerJames Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | tree 1536f39c18756698d033da72c49300a561be1289 parent 07172d7c9f10ee3d05d6f6489ba6d6ee2628da06 author Liu Hong <hong.liu@intel.com> 1124436225 -0500 committer James Ketrenos <jketreno@linux.intel.com> 1127312664 -0500 Added WE-18 support to default wireless extension handler in ieee80211 subsystem. Updated patch since last send to account for ieee80211_device parameter being added to the crypto init method. Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | [PATCH] ieee80211: Allow drivers to fix an issue when using wpa_supplicant ↵James Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | with WEP tree 898fedef6ca1b5b58b8bdf7e6d8894a78bbde4cd parent 8720fff53090ae428d2159332b6f4b2749dea10f author Zhu Yi <jketreno@io.(none)> 1124435746 -0500 committer James Ketrenos <jketreno@linux.intel.com> 1127312509 -0500 Allow drivers to fix an issue when using wpa_supplicant with WEP. The problem is introduced by the hwcrypto patch. We changed indicator of the encryption request from the upper layer (i.e. wpa_supplicant): In the original host based crypto the driver could use: crypt && crypt->ops. In the new hardware based crypto, the driver should use the flags specified in ieee->sec.encrypt. Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | [PATCH] ieee80211: Fix kernel Oops when module unloadJames Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | tree b69e983266840983183a00f5ac02c66d5270ca47 parent cdd6372949b76694622ed74fe36e1dd17a92eb71 author Zhu Yi <jketreno@io.(none)> 1124435425 -0500 committer James Ketrenos <jketreno@linux.intel.com> 1127312421 -0500 Fix kernel Oops when module unload. Export a new function ieee80211_crypt_quiescing from ieee80211. Device drivers call it to make the host crypto stack enter the quiescence state, which means "process existing requests, but don't accept new ones". This is usually called during a driver's host crypto data structure free (module unload) path. Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | [PATCH] ieee80211: Hardware crypto and fragmentation offload supportJames Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | tree 5322d496af90d03ffbec27292dc1a6268a746ede parent 6c9364386ccb786e4a84427ab3ad712f0b7b8904 author James Ketrenos <jketreno@linux.intel.com> 1124432367 -0500 committer James Ketrenos <jketreno@linux.intel.com> 1127311810 -0500 Hardware crypto and fragmentation offload support added (Zhu Yi) Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | [PATCH] ieee80211: Fixed a kernel oops on module unloadJames Ketrenos2005-09-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | tree 367069f24fc38b4aa910e86ff40094d2078d8aa7 parent a33a1982012e9070736e3717231714dc9892303b author James Ketrenos <jketreno@linux.intel.com> 1124430800 -0500 committer James Ketrenos <jketreno@linux.intel.com> 1127310571 -0500 Fixed a kernel oops on module unload by adding spin lock protection to ieee80211's crypt handlers (thanks to Zhu Yi) Modified scan result logic to report WPA and RSN IEs if set (vs.being based on wpa_enabled) Added ieee80211_device as the first parameter to the crypt init() method. TKIP modified to use that structure for determining whether to countermeasures are active. Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | Merge /spare/repo/linux-2.6/Jeff Garzik2005-09-21
| |\ \ \ \
| * | | | | [PATCH] ieee80211 Fixed type-o of abg_ture -> abg_trueJames Ketrenos2005-09-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | [PATCH 14/29] Fixed type-o of abg_ture -> abg_true. Signed-off-by: James Ketrenos <jketreno@linux.intel.com> NOTE: This patch requires drivers using abg_ture to be updated. Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | | [PATCH] ieee80211 Removed ieee80211_info_element_hdrJames Ketrenos2005-09-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Removed ieee80211_info_element_hdr structure as ieee80211_info_element provides the same use. Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | | [PATCH] Changed 802.11 headers to use ieee80211_info_element[0]James Ketrenos2005-09-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Changed 802.11 headers to use ieee80211_info_element as zero sized array so that sizeof calculations do not account for IE sizes. Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | | [PATCH] ieee80211 Added wireless spy supportJames Ketrenos2005-09-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Added wireless spy support to Rx code path. Signed-off-by: James Ketrenos <jketreno@linux.intel.com> NOTE: Looks like scripts/Lindent generated output different than the Lindented version already in-kernel, hence all the whitespace deltas... *sigh* Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | | [PATCH] ieee80211 quality scaling algorithm extension handlerJames Ketrenos2005-09-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Incorporated Bill Moss' quality scaling algorithm into default wireless extension handler. Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * | | | | [PATCH] drivers/net/wan/: possible cleanupsAdrian Bunk2005-09-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch contains possible cleanups including the following: - make needlessly global code static - #if 0 the following unused global function: - sdladrv.c: sdla_intde - remove the following unused global variable: - lmc_media.c: lmc_t1_cables - remove the following unneeded EXPORT_SYMBOL's: - cycx_drv.c: cycx_inten - sdladrv.c: sdla_inten - sdladrv.c: sdla_intde - sdladrv.c: sdla_intack - sdladrv.c: sdla_intr - syncppp.c: sppp_input - syncppp.c: sppp_change_mtu Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
* | | | | | Merge master.kernel.org:/pub/scm/linux/kernel/git/acme/net-2.6.15Linus Torvalds2005-10-28
|\ \ \ \ \ \
| * | | | | | [IPSEC]: Kill obsolete get_mss functionHerbert Xu2005-10-25
| | |_|_|_|/ | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Now that we've switched over to storing MTUs in the xfrm_dst entries, we no longer need the dst's get_mss methods. This patch gets rid of them. It also documents the fact that our MTU calculation is not optimal for ESP. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
* / | | | | [PATCH] gfp_t: net/*Al Viro2005-10-28
|/ / / / / | | | | | | | | | | | | | | | | | | | | Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* | | | | [LLC]: Strip RIF flag from source MAC addressJochen Friedrich2005-10-25
| | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Jochen Friedrich <jochen@scram.de> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
* | | | | [AX.25]: Fix signed char bugRalf Baechle2005-10-22
| |_|_|/ |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On architectures where the char type defaults to unsigned some of the arithmetic in the AX.25 stack to fail, resulting in some packets being dropped on receive. Credits for tracking this down and the original patch to Bob Brose N0QBJ <linuxhams@n0qbj-11.ampr.org>. Signed-off-by: Ralf Baechle DL5RB <ralf@linux-mips.org> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
* | | | [TWSK]: Grab the module refcount for timewait socketsArnaldo Carvalho de Melo2005-10-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is required to avoid unloading a module that has active timewait sockets, such as DCCP. Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | | [PATCH] gfp flags annotations - part 1Al Viro2005-10-08
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - added typedef unsigned int __nocast gfp_t; - replaced __nocast uses for gfp flags with gfp_t - it gives exactly the same warnings as far as sparse is concerned, doesn't change generated code (from gcc point of view we replaced unsigned int with typedef) and documents what's going on far better. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* | | | [SCTP] Fix SCTP socket options to work with 32-bit apps on 64-bit kernels.Sridhar Samudrala2005-10-07
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Adds alignment attribute to a few structures used with SCTP socket options so that the sizes and offsets remain the same when built using either 32 or 64 bit tools. Signed-off-by: Sridhar Samudrala <sri@us.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | | [SCTP] Fix sctp_get{pl}addrs() API to work with 32-bit apps on 64-bit kernels.Ivan Skytte Jørgensen2005-10-07
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The old socket options are marked with a _OLD suffix so that the existing 32-bit apps on 32-bit kernels do not break. Signed-off-by: Ivan Skytte Jørgensen <isj-sctp@i1.dk> Signed-off-by: Sridhar Samudrala <sri@us.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | | [IPSEC]: Document that policy direction is derived from the index.Herbert Xu2005-10-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Here is a patch that adds a helper called xfrm_policy_id2dir to document the fact that the policy direction can be and is derived from the index. This is based on a patch by YOSHIFUJI Hideaki and 210313105@suda.edu.cn. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | | [XFRM]: fix sparse gfp nocast warningsRandy Dunlap2005-10-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix implicit nocast warnings in xfrm code: net/xfrm/xfrm_policy.c:232:47: warning: implicit cast to nocast type Signed-off-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | | [IPVS]: fix sparse gfp nocast warningsRandy Dunlap2005-10-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | From: Randy Dunlap <rdunlap@xenotime.net> Fix implicit nocast warnings in ip_vs code: net/ipv4/ipvs/ip_vs_app.c:631:54: warning: implicit cast to nocast type Signed-off-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | | [DECNET]: fix sparse gfp nocast warningsRandy Dunlap2005-10-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix implicit nocast warnings in decnet code: net/decnet/af_decnet.c:458:40: warning: implicit cast to nocast type net/decnet/dn_nsp_out.c:125:35: warning: implicit cast to nocast type net/decnet/dn_nsp_out.c:219:29: warning: implicit cast to nocast type Signed-off-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | | [INET]: Shrink struct inet_ehash_bucket on 32 bits UPEric Dumazet2005-10-04
| |_|/ |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | No need to align struct inet_ehash_bucket on a 8 bytes boundary. On 32 bits Uniprocessor, that's a waste of 4 bytes per struct (50 %) On other platforms, the attribute is useless, natual alignement is already 8. platform | Size before | Size after patch -------------+-------------+------------------ 32 bits, UP | 8 | 4 32 bits, SMP | 8 | 8 64 bits, UP | 8 | 8 64 bits, SMP | 16 | 16 Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | [INET]: speedup inet (tcp/dccp) lookupsEric Dumazet2005-10-03
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Arnaldo and I agreed it could be applied now, because I have other pending patches depending on this one (Thank you Arnaldo) (The other important patch moves skc_refcnt in a separate cache line, so that the SMP/NUMA performance doesnt suffer from cache line ping pongs) 1) First some performance data : -------------------------------- tcp_v4_rcv() wastes a *lot* of time in __inet_lookup_established() The most time critical code is : sk_for_each(sk, node, &head->chain) { if (INET_MATCH(sk, acookie, saddr, daddr, ports, dif)) goto hit; /* You sunk my battleship! */ } The sk_for_each() does use prefetch() hints but only the begining of "struct sock" is prefetched. As INET_MATCH first comparison uses inet_sk(__sk)->daddr, wich is far away from the begining of "struct sock", it has to bring into CPU cache cold cache line. Each iteration has to use at least 2 cache lines. This can be problematic if some chains are very long. 2) The goal ----------- The idea I had is to change things so that INET_MATCH() may return FALSE in 99% of cases only using the data already in the CPU cache, using one cache line per iteration. 3) Description of the patch --------------------------- Adds a new 'unsigned int skc_hash' field in 'struct sock_common', filling a 32 bits hole on 64 bits platform. struct sock_common { unsigned short skc_family; volatile unsigned char skc_state; unsigned char skc_reuse; int skc_bound_dev_if; struct hlist_node skc_node; struct hlist_node skc_bind_node; atomic_t skc_refcnt; + unsigned int skc_hash; struct proto *skc_prot; }; Store in this 32 bits field the full hash, not masked by (ehash_size - 1) Using this full hash as the first comparison done in INET_MATCH permits us immediatly skip the element without touching a second cache line in case of a miss. Suppress the sk_hashent/tw_hashent fields since skc_hash (aliased to sk_hash and tw_hash) already contains the slot number if we mask with (ehash_size - 1) File include/net/inet_hashtables.h 64 bits platforms : #define INET_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\ (((__sk)->sk_hash == (__hash)) ((*((__u64 *)&(inet_sk(__sk)->daddr)))== (__cookie)) && \ ((*((__u32 *)&(inet_sk(__sk)->dport))) == (__ports)) && \ (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) 32bits platforms: #define TCP_IPV4_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\ (((__sk)->sk_hash == (__hash)) && \ (inet_sk(__sk)->daddr == (__saddr)) && \ (inet_sk(__sk)->rcv_saddr == (__daddr)) && \ (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) - Adds a prefetch(head->chain.first) in __inet_lookup_established()/__tcp_v4_check_established() and __inet6_lookup_established()/__tcp_v6_check_established() and __dccp_v4_check_established() to bring into cache the first element of the list, before the {read|write}_lock(&head->lock); Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | [LLC]: fix llc_ui_recvmsg, making it behave like tcp_recvmsgArnaldo Carvalho de Melo2005-09-22
| | | | | | | | | | | | | | | | | | In fact it is an exact copy of the parts that makes sense to LLC :-) Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
* | | [LLC]: Fix the accept pathArnaldo Carvalho de Melo2005-09-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Borrowing the structure of TCP/IP for this. On the receive of new connections I was bh_lock_socking the _new_ sock, not the listening one, duh, now it survives the ssh connections storm I've been using to test this specific bug. Also fixes send side skb sock accounting. Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
* | | [LLC]: Fix sparse warningsArnaldo Carvalho de Melo2005-09-22
| | | | | | | | | | | | Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
* | | [LLC]: Use refcounting with struct llc_sapArnaldo Carvalho de Melo2005-09-22
| | | | | | | | | | | | | | | Signed-off-by: Jochen Friedrich <jochen@scram.de> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
* | | [LLC]: Do better struct sock accounting on skbsArnaldo Carvalho de Melo2005-09-22
| | | | | | | | | | | | | | | Signed-off-by: Jochen Friedrich <jochen@scram.de> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
* | | [LLC]: Add sysctl support for the LLC timeoutsArnaldo Carvalho de Melo2005-09-22
| | | | | | | | | | | | | | | Signed-off-by: Jochen Friedrich <jochen@scram.de> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
* | | [LLC]: Make llc_frame_alloc take a net_device as an argumentArnaldo Carvalho de Melo2005-09-22
| |/ |/| | | | | | | | | | | | | | | | | | | | | | | | | So as to set the newly created sk_buff ->dev member with it, that way we stop using dev_base->next, that is the wrong thing to do, as there may well be several interfaces being used with LLC. This was not such a big problem after all as most of the users of llc_alloc_frame were setting the correct dev, but this way code is reduced. This also fixes another bug in llc_station_ac_send_null_dsap_xid_c, that was not setting the skb->dev field. Signed-off-by: Jochen Friedrich <jochen@scram.de> Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
* | [IPVS]: ip_vs_ftp breaks connections using persistenceJulian Anastasov2005-09-15
|/ | | | | | | | | | | | | | | | | | | | | | | | ip_vs_ftp when loaded can create NAT connections with unknown client port for passive FTP. For such expectations we lookup with cport=0 on incoming packet but it matches the format of the persistence templates causing packets to other persistent virtual servers to be forwarded to real server without creating connection. Later the reply packets are treated as foreign and not SNAT-ed. This patch changes the connection lookup for packets from clients: * introduce IP_VS_CONN_F_TEMPLATE connection flag to mark the connection as template * create new connection lookup function just for templates - ip_vs_ct_in_get * make sure ip_vs_conn_in_get hits only connections with IP_VS_CONN_F_NO_CPORT flag set when s_port is 0. By this way we avoid returning template when looking for cport=0 (ftp) Signed-off-by: Julian Anastasov <ja@ssi.bg> Signed-off-by: David S. Miller <davem@davemloft.net>
* Merge master.kernel.org:/pub/scm/linux/kernel/git/holtmann/bluetooth-2.6 David S. Miller2005-09-13
|\
| * [Bluetooth] Add support for extended inquiry responsesMarcel Holtmann2005-09-12
| | | | | | | | | | | | | | This patch adds the handling of the extended inquiry responses and inserts them into the inquiry cache. Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
* | [NETROM]: Introduct stuct nr_privateRalf Baechle2005-09-12
| | | | | | | | | | | | | | | | NET/ROM's virtual interfaces don't have a proper private data structure yet. Create struct nr_private and put the statistics there. Signed-off-by: Ralf Baechle DL5RB <ralf@linux-mips.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* | [NETROM]: Implement G8PZT Circuit reset for NET/ROMRalf Baechle2005-09-12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | NET/ROM is lacking a connection reset like TCP's RST flag which at times may result in a connecting having to slowly timing out instead of just being reset. An earlier attempt to reset the connection by sending a NR_CONNACK | NR_CHOKE_FLAG transport was inacceptable as it did result in crashes of BPQ systems. An alternative approach of introducing a new transport type 7 (NR_RESET) has be implemented several years ago in Paula Jayne Dowie G8PZT's Xrouter. Implement NR_RESET for Linux's NET/ROM but like any messing with the state engine consider this experimental for now and thus control it by a sysctl (net.netrom.reset) which for the time being defaults to off. Signed-off-by: Ralf Baechle DL5RB <ralf@linux-mips.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* | [AX.25]: Add descriptions to constantsRalf Baechle2005-09-12
| | | | | | | | | | | | | | Comment the names used for the AX.25 state machine. Signed-off-by: Ralf Baechle DL5RB <ralf@linux-mips.org> Signed-off-by: David S. Miller <davem@davemloft.net>
generated by cgit v1.2.2 (git 2.25.0) at 2025-04-18 09:43:11 -0400