summaryrefslogtreecommitdiffstats
path: root/drivers/net
Commit message (Collapse)AuthorAgeFilesLines
* e1000: Add enabled Jumbo frame support for 82573LJeff Kirsher2006-03-022-1/+20
| | | | | | Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
* e1000: Add 82573 controller support to TSO fixJeff Kirsher2006-03-021-11/+19
| | | | | | Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
* e1000: Fix filling skb descriptors while using packet splitJeff Kirsher2006-03-021-5/+2
| | | | | | | | - Simplified by calling skb_fill_page_desc(), which is more efficient. Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
* e1000: Fix network problems when forced at 100Mb/s and to fix TSO when ↵Jeff Kirsher2006-03-024-11/+66
| | | | | | | | forced at 100Mb/s Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
* e1000: Fix Quadport Wake on LANJeff Kirsher2006-03-023-1/+41
| | | | | | Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
* e1000: Fix RSS if enabled in mid-connectionJeff Kirsher2006-03-022-1/+30
| | | | | | Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
* e1000: Fix 82543 issue when reading eepromJeff Kirsher2006-03-021-1/+7
| | | | | | Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
* e1000: Fix dhcp issue when the skb structure fields are not filled properlyJeff Kirsher2006-03-021-1/+1
| | | | | | Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
* e1000: Fix AMT losing connectivity when switching VLAN in passive modeJeff Kirsher2006-03-021-1/+2
| | | | | | Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
* e1000: Fix unecessary delay for 82573 controllersJeff Kirsher2006-03-021-6/+0
| | | | | | Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
* e1000: Fix lock up while setting ring parametersJeff Kirsher2006-03-021-3/+3
| | | | | | Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
* e1000: Fix dead countersJeff Kirsher2006-03-021-1/+0
| | | | | | Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
* e1000: Remove Multiqueue code until we have support for MSI-X in our hardwareJeff Kirsher2006-03-023-292/+4
| | | | | | Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
* Merge branch 'lro'Jeff Garzik2006-03-022-62/+565
|\
| * [PATCH] s2io c99 warning fixAndrew Morton2006-02-071-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | drivers/net/s2io.c: In function `verify_l3_l4_lro_capable': drivers/net/s2io.c:6638: warning: ISO C90 forbids mixed declarations and code Cc: Ravinandan Arakali <Ravinandan.Arakali@neterion.com> Cc: Ananda Raju <Ananda.Raju@neterion.com> Cc: Adrian Bunk <bunk@stusta.de> Cc: Jeff Garzik <jgarzik@pobox.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * [PATCH] git-netdev-all: s2io fixesAndrew Morton2006-02-071-6/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix this: drivers/built-in.o(.text+0x89469): In function `s2io_get_ethtool_stats': drivers/net/s2io.c:5209: undefined reference to `__udivdi3' Cc: Ravinandan Arakali <Ravinandan.Arakali@neterion.com> Cc: Ananda Raju <Ananda.Raju@neterion.com> Cc: Adrian Bunk <bunk@stusta.de> Cc: Jeff Garzik <jgarzik@pobox.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
| * [PATCH] S2io: Large Receive Offload (LRO) feature(v2) for Neterion (s2io) ↵Ravinandan Arakali2006-01-272-62/+562
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 10GbE Xframe PCI-X and PCI-E NICs Hi, Below is a patch for the Large Receive Offload feature. Please review and let us know your comments. LRO algorithm was described in an OLS 2005 presentation, located at ftp.s2io.com user: linuxdocs password: HALdocs The same ftp site has Programming Manual for Xframe-I ASIC. LRO feature is supported on Neterion Xframe-I, Xframe-II and Xframe-Express 10GbE NICs. Brief description: The Large Receive Offload(LRO) feature is a stateless offload that is complementary to TSO feature but on the receive path. The idea is to combine and collapse(upto 64K maximum) in the driver, in-sequence TCP packets belonging to the same session. It is mainly designed to improve 1500 mtu receive performance, since Jumbo frame performance is already close to 10GbE line rate. Some performance numbers are attached below. Implementation details: 1. Handle packet chains from multiple sessions(current default MAX_LRO_SESSSIONS=32). 2. Examine each packet for eligiblity to aggregate. A packet is considered eligible if it meets all the below criteria. a. It is a TCP/IP packet and L2 type is not LLC or SNAP. b. The packet has no checksum errors(L3 and L4). c. There are no IP options. The only TCP option supported is timestamps. d. Search and locate the LRO object corresponding to this socket and ensure packet is in TCP sequence. e. It's not a special packet(SYN, FIN, RST, URG, PSH etc. flags are not set). f. TCP payload is non-zero(It's not a pure ACK). g. It's not an IP-fragmented packet. 3. If a packet is found eligible, the LRO object is updated with information such as next sequence number expected, current length of aggregated packet and so on. If not eligible or max packets reached, update IP and TCP headers of first packet in the chain and pass it up to stack. 4. The frag_list in skb structure is used to chain packets into one large packet. Kernel changes required: None Performance results: Main focus of the initial testing was on 1500 mtu receiver, since this is a bottleneck not covered by the existing stateless offloads. There are couple disclaimers about the performance results below: 1. Your mileage will vary!!!! We initially concentrated on couple pci-x 2.0 platforms that are powerful enough to push 10 GbE NIC and do not have bottlenecks other than cpu%; testing on other platforms is still in progress. On some lower end systems we are seeing lower gains. 2. Current LRO implementation is still (for the most part) software based, and therefore performance potential of the feature is far from being realized. Full hw implementation of LRO is expected in the next version of Xframe ASIC. Performance delta(with MTU=1500) going from LRO disabled to enabled: IBM 2-way Xeon (x366) : 3.5 to 7.1 Gbps 2-way Opteron : 4.5 to 6.1 Gbps Signed-off-by: Ravinandan Arakali <ravinandan.arakali@neterion.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
* | Merge branch 'upstream' of ↵Jeff Garzik2006-03-018-102/+81
|\ \ | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-2.6
| * | [PATCH] wireless/airo: Remove 'Setting transmit key' info messagesDan Williams2006-02-271-2/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | These messages end up polluting logs when things like NetworkManager or wpa_supplicant are controlling the driver. They aren't really that useful, and no other drivers output messages like this when the user fiddles with encryption keys. Users can use iwconfig and other wireless-tools methods to determine and change the current transmit key if they wish to do so manually. Therefore, remove the messages. Signed-off-by: Dan Williams <dcbw@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * | Merge branch 'from-jgarzik-upstream'John W. Linville2006-02-271-48/+545
| |\ \
| * \ \ Merge branch 'from-linus'John W. Linville2006-02-2717-197/+513
| |\ \ \
| * | | | [PATCH] wireless/airo: fix setting TX key index plus key in ENCODEEXTDan Williams2006-02-171-3/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The previous patch that added ENCODEEXT and AUTH support to the airo driver contained a slight error which would cause setting the TX key index ignore a valid key-set request at the same time. This patch allows any combination of setting the TX key index and setting an encryption key. Signed-off-by: Dan Williams <dcbw@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * | | | [PATCH] ipw2200: Fix software crypto shared WEP authentication problemZhu Yi2006-02-171-2/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We didn't set the WEP key to hardware when we are using software based crypto. Hardware needs the key to do WEP authentication even for software based encryption. Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * | | | [PATCH] trivial: fix spelling errors in KconfigsJon Mason2006-02-175-17/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch corrects a few spelling and grammar errors found in drivers/net Signed-off-by: Jon Mason <jdmason@us.ibm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * | | | [PATCH] ipw2200: Fix WPA network selection problemStefan Rompf2006-02-171-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Do not avoid APs with wpa_ie or rsn_ie if !ieee->wpa_enabled There are broken APs out there that fill these elements even though encryption is disnabled. Also, this breaks legit WEP to WPA migration scenarious. We add a checking to prohibite WPA configured STA trying to associate with non-WPA supported APs. Signed-off-by: Stefan Rompf <stefan@loplof.de> Signed-off-by: James Ketrenos <jketreno@linux.intel.com> Signed-off-by: Zhu Yi <yi.zhu@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * | | | [PATCH] ipw2200: add monitor and qos entries to KconfigAndreas Happe2006-02-171-0/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I have made a stupid copy&paste error: QoS option is named IPW_QOS not IPW2200_MONITOR. Spotted by Daniel Paschka, thanks. Add the following config entries for the ipw2200 driver to drivers/net/wireless/Kconfig * IPW2200_MONITOR enables Monitor mode * IPW_QOS enables QoS feature - this is under development right now, so it depends upon EXPERIMENTAL Signed-off-by: Andreas Happe <andreashappe@snikt.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * | | | [PATCH] wireless/atmel: convert constants to ieee80211 layer equivalentsDan Williams2006-02-171-74/+36
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch converts the Atmel driver-only IEEE 802.11 constants to their equivalents from the kernel's ieee80211 layer headers. Signed-off-by: Dan Williams <dcbw@redhat.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * | | | [PATCH] let IPW2{1,2}00 select IEEE80211Jan Niehusmann2006-02-171-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch makes the IPW2100 and IPW2200 options available in the configuration menu even if IEEE80211 has not been selected before. This behaviour is more intuitive for people which are not familiar with the driver internals. The suggestion for this change was made by Alejandro Bonilla Beeche. Signed-off-by: Jan Niehusmann <jan@gondor.com> Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: John W. Linville <linville@tuxdriver.com>
| * | | | Merge branch 'upstream-fixes'John W. Linville2006-02-171-39/+59
| |\ \ \ \
| * \ \ \ \ Merge branch 'upstream' of ↵John W. Linville2006-02-1773-2208/+1548
| |\ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/netdev-2.6
* | \ \ \ \ \ Merge branch 'upstream-fixes'Jeff Garzik2006-03-017-170/+79
|\ \ \ \ \ \ \ | |_|_|_|_|/ / |/| | | | | |
| * | | | | | Merge branch 'e1000' of ssh://198.78.49.142/srv/git/intel/linux-2.6Jeff Garzik2006-03-012-76/+46
| |\ \ \ \ \ \
| | * | | | | | [PATCH] e1000: revert to single descriptor for legacy receive pathJeff Kirsher2006-02-282-76/+46
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | A recent patch attempted to enable more efficient memory usage by using only 2kB descriptors for jumbo frames. The method used to implement this has since been commented upon as "illegal" and in recent kernels even causes a BUG when receiving ip fragments while using jumbo frames. This patch simply goes back to the way things were. We expect some complaints due to order 3 allocations failing to come back due to this change. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
| * | | | | | | [PATCH] sky2: remove MSI supportStephen Hemminger2006-02-282-78/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Remove Message Signaled Interrupt support (for 2.6.16). MSI is inherently edge-triggered and that is incompatiable (without more work) with NAPI. In future, will replace with smarter lockless-IRQ handling like tg3.c Signed-off-by: Stephen Hemminger <shemminger@osdl.org> Signed-off-by: Jeff Garzik <jeff@garzik.org>
| * | | | | | | Merge branch 'for-jeff' of ↵Jeff Garzik2006-02-282-16/+24
| |\ \ \ \ \ \ \ | | |_|_|_|_|/ / | |/| | | | | | | | | | | | | | git://electric-eye.fr.zoreil.com/home/romieu/linux-2.6
| | * | | | | | 8139cp: fix broken suspend/resumeFrancois Romieu2006-02-271-16/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - check that the device is up before it is enabled again; - the descriptor ring indexes must be set to zero before cp_init_hw() is issued. Add a nice comment to remember that skb allocation failure is still not handled. Fixes http://bugzilla.kernel.org/show_bug.cgi?id=5681 Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
| | * | | | | | via-velocity: fix memory corruption when changing the mtuFrancois Romieu2006-02-271-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | velocity_rx_refill() only replenishes the descriptor entries which belong to the CPU. It works great in the Rx path but the driver must ensure that all the descriptors are freed before velocity_rx_refill() is used in velocity_change_mtu(). The patch resets the Rx descriptors in velocity_free_rd_ring(). Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
| * | | | | | | [TG3]: Fix Sun tg3 variant detection.David S. Miller2006-02-271-0/+9
| |/ / / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Some Sun parts don't have PCI_VENDOR_ID_SUN in the subsystem vendor ID. So add another fallback test, which is the name of the OBP firmware device tree node. If it's a Sun part we'll get "network", else it will be named "ethernet". Signed-off-by: David S. Miller <davem@davemloft.net>
* | | | | | | Merge branch 'master'Jeff Garzik2006-02-273-3/+5
|\ \ \ \ \ \ \ | |/ / / / / /
| * | | | | | Fix Wake on LAN support in sis900Daniele Venzano2006-02-251-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix two bugs in the WoL implementation of sis900. The first causes hangs on some system on driver load, the second causes troubles when disabling WoL support. Both fixes are one liner and really simple. Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org> Signed-off-by: Daniele Venzano <venza@brownhat.org>
| * | | | | | Merge branch 'upstream-fixes' of ↵Linus Torvalds2006-02-236-153/+370
| |\ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6
| | * | | | | | [PATCH] drivers/net/tlan.c: #ifdef CONFIG_PCI the PCI specific codeAdrian Bunk2006-02-241-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | drivers/net/tlan.c compiles with CONFIG_PCI=n only with a warning and due to the dead code elimination of gcc. Additionally, this fixes the only compile error I found with CONFIG_PCI=n and the gcc -Werror-implicit-function-declaration flag on i386. Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Jeff Garzik <jeff@garzik.org>
| * | | | | | | [NET]: Increase default IFB device count.Richard Lucassen2006-02-231-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The most usable number of ifb devices is 2. Change the default to 2. Signed-off-by: Richard Lucassen <spamtrap@lucassen.org> Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | | | | | | Merge branch 'upstream-fixes'Jeff Garzik2006-02-238-174/+450
|\ \ \ \ \ \ \ \ | | |/ / / / / / | |/| | | | | |
| * | | | | | | skge: protect interrupt maskStephen Hemminger2006-02-232-7/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There is a race between updating the irq mask and setting it which can be triggered on SMP with a bad cable. Similar patch from Ingo Molnar and Thomas Gleixner Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
| * | | | | | | skge: genesis phy initialzationStephen Hemminger2006-02-231-11/+26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The SysKonnect Genesis based board would fail on initialization with phy_read errors caused by not waiting for last phy write. Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
| * | | | | | | skge: NAPI/irq race fixStephen Hemminger2006-02-231-12/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix a race in the receive NAPI, irq handling. The interrupt clear and the start need to be separated. Otherwise there is a window between the last frame received and the NAPI done level handling. Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
| * | | | | | | r8169: enable wake on lanFrancois Romieu2006-02-231-0/+87
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Similar to 8139cp code but more inspired/lucky. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
| * | | | | | | r8169: fix broken ring index handling in suspend/resumeFrancois Romieu2006-02-231-43/+59
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | rtl8169_hw_start() requires that the descriptor ring indexes be set to zero. Let a deferred invocation of rtl8169_reset_task() handle it. Enabling a few power management bits will not hurt either. suspend/resume is issued with irq on: the spinlock do not need to save the irq flag. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
| * | | | | | | [PATCH] sky2: close race on IRQ mask update.Stephen Hemminger2006-02-232-7/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Need to avoid race in updating IRQ mask. This can probably be replaced smarter use of the interrupt control registers (if/when chipset docs are available). Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
OpenPOWER on IntegriCloud