aboutsummaryrefslogtreecommitdiff
path: root/drivers
AgeCommit message (Collapse)AuthorFilesLines
2011-02-09iwl3945: remove plcp checkStanislaw Gruszka1-1/+0
Patch fixes: https://bugzilla.redhat.com/show_bug.cgi?id=654599 Many users report very low speed problem on 3945 devices, this patch fixes problem, but only for some of them. For unknown reason, sometimes after hw scanning, device is not able to receive frames at high rate. Since plcp health check may request hw scan to "reset radio", performance problem start to be observable after update kernel to .35, where plcp check was introduced. Bug reporter confirmed that removing plcp check fixed problem for him. Reported-and-tested-by: SilvioTO <[email protected]> Cc: [email protected] # 2.6.35+ Signed-off-by: Stanislaw Gruszka <[email protected]> Acked-by: Wey-Yi Guy <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-02-09Merge branch 'master' of ↵John W. Linville2-0/+5
git://git.kernel.org/pub/scm/linux/kernel/git/padovan/bluetooth-2.6
2011-02-07iwlagn: Re-enable RF_KILL interrupt when downDon Fry1-0/+6
With commit 554d1d027b19265c4aa3f718b3126d2b86e09a08 only one RF_KILL interrupt will be seen by the driver when the interface is down. Re-enable the interrupt when it occurs to see all transitions. Signed-off-by: Don Fry <[email protected]> Signed-off-by: Wey-Yi Guy <[email protected]> Cc: [email protected] Signed-off-by: John W. Linville <[email protected]>
2011-02-07ssb-pcmcia: Fix parsing of invariants tuplesMichael Büsch1-1/+1
This fixes parsing of the device invariants (MAC address) for PCMCIA SSB devices. ssb_pcmcia_do_get_invariants expects an iv pointer as data argument. Tested-by: dylan cristiani <[email protected]> Signed-off-by: Michael Buesch <[email protected]> Cc: [email protected] Signed-off-by: John W. Linville <[email protected]>
2011-02-04ath9k: Fix possible double free of PAPRD skb'sMohammed Shafi Shajakhan3-4/+7
This patch reverts the following commit ath9k: remove bfs_paprd_timestamp from struct ath_buf_state Under high interference/noisy environment conditions where PAPRD frames fails heavily introduces a possibility of double freeing skb's and causes kernel panic after some time.This patch reverts back to the original approach of using paprd_timestamp before freeing the PAPRD frame skb's [ 194.193705] Pid: 0, comm: swapper Tainted: G D WC 2.6.35-22-generic #33-Ubuntu [ 194.193712] Call Trace: [ 194.193722] [<c05c6468>] ? printk+0x2d/0x35 [ 194.193732] [<c05c63c3>] panic+0x5a/0xd2 [ 194.193741] [<c05ca3ed>] oops_end+0xcd/0xd0 [ 194.193750] [<c0105f74>] die+0x54/0x80 [ 194.193758] [<c05c9a16>] do_trap+0x96/0xc0 [ 194.193837] [<c0103fb0>] ? do_invalid_op+0x0/0xa0 [ 194.193846] [<c010403b>] do_invalid_op+0x8b/0xa0 [ 194.193856] [<c020bd4c>] ? kfree+0xec/0xf0 [ 194.193866] [<c012ce18>] ? default_spin_lock_flags+0x8/0x10 [ 194.193877] [<c01de47a>] ? free_one_page+0x12a/0x2d0 [ 194.193888] [<c01e04dc>] ? __free_pages+0x1c/0x40 [ 194.193897] [<c05c97a7>] error_code+0x73/0x78 [ 194.193906] [<c020bd4c>] ? kfree+0xec/0xf0 [ 194.193915] [<c04ecdd0>] ? skb_release_data+0x70/0xa0 [ 194.193924] [<c04ecdd0>] skb_release_data+0x70/0xa0 [ 194.193933] [<c04ec997>] __kfree_skb+0x17/0x90 [ 194.193941] [<c04eca31>] consume_skb+0x21/0x40 [ 194.193964] [<f85e0b70>] ieee80211_tx_status+0x760/0x860 [mac80211] [ 194.193979] [<f85caddf>] ath_tx_complete_buf+0x1bf/0x2c0 [ath9k] [ 194.193988] [<c05c8b9f>] ? _raw_spin_lock_irqsave+0x2f/0x50 [ 194.193997] [<c04ec40e>] ? skb_queue_tail+0x3e/0x50 [ 194.194010] [<f85cc803>] ath_tx_complete_aggr+0x823/0x940 [ath9k] [ 194.194021] [<c0108a28>] ? sched_clock+0x8/0x10 [ 194.194030] [<c016bf14>] ? sched_clock_local+0xa4/0x180 [ 194.194040] [<c0139f57>] ? enqueue_sleeper+0x1e7/0x2b0 [ 194.194051] [<c013a194>] ? enqueue_entity+0x174/0x200 [ 194.194064] [<f85ce83d>] ath_tx_edma_tasklet+0x2bd/0x3b0 [ath9k] [ 194.194074] [<c05c8b9f>] ? _raw_spin_lock_irqsave+0x2f/0x50 [ 194.194088] [<f85c7b9f>] ath9k_tasklet+0x9f/0x190 [ath9k] [ 194.194097] [<c01505d7>] tasklet_action+0xa7/0xb0 [ 194.194107] [<c015127c>] __do_softirq+0x9c/0x1b0 [ 194.194117] [<c01a7f64>] ? irq_to_desc+0x14/0x20 [ 194.194126] [<c0124fc4>] ? ack_apic_level+0x64/0x1f0 [ 194.194136] [<c01513d5>] do_softirq+0x45/0x50 [ 194.194145] [<c0151545>] irq_exit+0x65/0x70 [ 194.194153] [<c05cf665>] do_IRQ+0x55/0xc0 [ 194.194162] [<c016a6c7>] ? hrtimer_start+0x27/0x30 [ 194.194171] [<c0103630>] common_interrupt+0x30/0x38 [ 194.194181] [<c012c21a>] ? native_safe_halt+0xa/0x10 [ 194.194268] [<c010a2f9>] default_idle+0x49/0xb0 [ 194.194277] [<c0101fcc>] cpu_idle+0x8c/0xd0 [ 194.194286] [<c05b2431>] rest_init+0x71/0x80 [ 194.194295] [<c081981a>] start_kernel+0x36e/0x374 [ 194.194305] [<c08199dd>] ? pass_all_bootoptions+0x0/0xa [ 194.194314] [<c08190d7>] i386_start_kernel+0xd7/0xdf [ 194.194364] panic occurred, switching back to text console Signed-off-by: Mohammed Shafi Shajakhan <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-02-04carl9170: fix typo in PS codeChristian Lamparter1-1/+1
This patch fixes a off-by-one bug which bugged the driver's PS-POLL capability. Cc: <[email protected]> Signed-off-by: Christian Lamparter <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-02-03wireless, wl1251: Fix potential NULL pointer dereference in ↵Jesper Juhl1-0/+3
wl1251_op_bss_info_changed() In drivers/net/wireless/wl1251/main.c:wl1251_op_bss_info_changed() we make a call to ieee80211_beacon_get() which may return NULL, but we do not check the return value before dereferencing the pointer. Signed-off-by: Jesper Juhl <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-02-02iwlagn: overwrite EEPROM chain setting for 6250 devicesWey-Yi Guy1-0/+2
6250 2x2 devices have 2 tx chain and 2 rx chain. For some reason, the EEPROM contain incorrect information and indicate it only has single tx chain. overwrite it with .cfg parameter to make sure both chain 'A' and chain 'B' can be used for transmit and receive Signed-off-by: Wey-Yi Guy <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-02-01ath9k: Fix memory leak due to failed PAPRD framesMohammed Shafi Shajakhan1-1/+7
free the skb's when the Tx of PAPRD frames fails and also add a debug message indicating that. Signed-off-by: Mohammed Shafi Shajakhan <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-31wl12xx: fix use after freeMathias Krause1-2/+1
When DEBUG_SPI is included in the debug log level wl1271_spi_reset() will dump the already freed memory instead of the SPI buffer. This bug was spotted by the semantic patch tool coccinelle using the script found at scripts/coccinelle/free/kfree.cocci. More information about semantic patching is available at http://coccinelle.lip6.fr/ Signed-off-by: Mathias Krause <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-28ath9k: Fix power save usage count imbalance on deinitRajkumar Manoharan2-2/+2
While unloading the driver, the ps_usecount is incremented before configuring gpio registers in deinit_device. But it is failed to restore the ps_usecount after that. The problem is that the chip is forcibly moved to FULL SLEEP by radio_disable when mac80211 is reporting as idle though ps_usecount is not zero. This patch retores ps_usecount properly and ensures that the chip is always moved to full sleep only if ps usage count is zero which also helps in debugging deadbeef on multivif case. And also fixes the following warning. ath: DMA failed to stop in 10 ms AR_CR=0xdeadbeef AR_DIAG_SW=0xdeadbeef ath: Could not stop RX, we could be confusing the DMA engine when we start RX up ------------[ cut here ]------------ WARNING: at drivers/net/wireless/ath/ath9k/recv.c:536 ath_stoprecv+0xf4/0x100 [ath9k]() Cc: [email protected] Cc: Paul Stewart <[email protected]> Signed-off-by: Rajkumar Manoharan <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-28ath9k_hw: Fix system hang when resuming from S3/S4Rajkumar Manoharan1-2/+1
The bit 6 & 7 of AR_WA (0x4004) should be enabled only for the chips that are supporting L0s functionality while resuming back from S3/S4. Enabling these bits for AR9280 is causing system hang within a few S3/S4-resume cycles. Cc: [email protected] Cc: Jack Lee <[email protected]> Signed-off-by: Rajkumar Manoharan <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-27ath9k_htc: fix race conditions when stop deviceStanislaw Gruszka2-9/+15
We do not kill any scheduled tasklets when stopping device, that may cause usage of resources after free. Disable interrupts, kill tasklets and then works in correct order. Cc: [email protected] Tested-by: Sujith <[email protected]> Signed-off-by: Stanislaw Gruszka <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-27ath9k: fix race conditions when stop deviceStanislaw Gruszka2-5/+9
We do not kill any scheduled tasklets when stopping device, that may cause usage of resources after free. Moreover we enable interrupts in tasklet function, so we could potentially end with interrupts enabled when driver is not ready to receive them. I think patch should fix Ben's kernel crash from: http://marc.info/?l=linux-wireless&m=129438358921501&w=2 Cc: [email protected] Signed-off-by: Stanislaw Gruszka <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-27ath5k: correct endianness of frame durationBob Copeland1-3/+1
The ath5k version of ieee80211_generic_frame_duration() returns an __le16 for standard modes but a cpu-endian int for turbo/half/ quarter rates. Make it always return cpu-endian values. Signed-off-by: Bob Copeland <[email protected]> Acked-by: Bruno Randolf <[email protected]> Acked-by: Nick Kossifidis <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-27ath5k: fix error handling in ath5k_hw_dma_stopBob Copeland1-2/+2
Review spotted a problem with the error handling in ath5k_hw_dma_stop: a successful return from ath5k_hw_stop_tx_dma will be treated as an error, so we always bail out of the loop after processing a single active queue. As a result, we may not actually stop some queues during reset. Signed-off-by: Bob Copeland <[email protected]> Acked-by: Bruno Randolf <[email protected]> Acked-by: Nick Kossifidis <[email protected]> Reviewed-by: Stanislaw Gruszka <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-27rtlwifi: Fix firmware upload errorsChaoming Li1-20/+20
When the source code from Realtek was prepared for kernel inclusion, some routines were refactored to reduce the level of indentation. This patch repairs errors introduced in that process. Signed-off-by: Chaoming Li <[email protected]> Signed-off-by: Larry Finger <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-26Bluetooth: add Atheros BT AR9285 fw supportedCho, Yu-Chen2-0/+5
Add the btusb.c blacklist [03f0:311d] for Atheros AR9285 Malbec BT and add to ath3k.c ath3-1.fw (md5:1211fa34c09e10ba48381586b7c3883d) supported this device. Signed-off-by: Cho, Yu-Chen <[email protected]> Signed-off-by: Gustavo F. Padovan <[email protected]>
2011-01-25rt2x00: add device id for windy31 usb deviceGreg Kroah-Hartman1-0/+1
This patch adds the device id for the windy31 USB device to the rt73usb driver. Thanks to Ralf Flaxa for reporting this and providing testing and a sample device. Reported-by: Ralf Flaxa <[email protected]> Tested-by: Ralf Flaxa <[email protected]> Cc: stable <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]> Acked-by: Ivo van Doorn <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-21rtlwifi: Fix possible NULL dereferenceJesper Juhl1-2/+9
In drivers/net/wireless/rtlwifi/pci.c::_rtl_pci_rx_interrupt() we call dev_alloc_skb(), which may fail and return NULL, but we do not check the returned value against NULL before dereferencing the returned pointer. This may lead to a NULL pointer dereference which means we'll crash - not good. In a separate call to dev_alloc_skb(), the debug level is changed so that the failure message will always be logged. Signed-off-by: Jesper Juhl <[email protected]> Signed-off-by: Larry Finger <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-21ath9k: add missing ps wakeup/restore callsFelix Fietkau2-5/+5
There are several places where ath_reset() was called without proper calls to ath9k_ps_wakeup/ath9k_ps_restore. To fix this, add those calls directly to ath_reset and drop them from callers where it makes sense. Also add them to the config callback around ath_update_txpow to fix a crash that happens when the tx power changed before any vif is brought up. Signed-off-by: Felix Fietkau <[email protected]> Cc: [email protected] Signed-off-by: John W. Linville <[email protected]>
2011-01-21ath9k_hw: disabled PAPRD for AR9003Luis R. Rodriguez2-1/+6
AR9003's PAPRD was enabled prematurely, and is causing some large discrepancies on throughput and network connectivity. For example downlink (RX) throughput against an AR9280 AP can vary widlely from 43-73 Mbit/s while disabling this gets AR9382 (2x2) up to around 93 Mbit/s in a 2.4 GHz HT20 setup. Cc: [email protected] Cc: Paul Shaw <[email protected]> Signed-off-by: Luis R. Rodriguez <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-21iwlwifi: don't read sku information from EEPROM for 4965Wey-Yi Guy2-4/+8
For all the new devices, the sku information should read from EEPROM but for legacy devices such as 4965, appearly the EEPROM does not contain the necessary information. so skip the read from EEPROM and go back to use software configuration. Reported-by: Helmut Schaa <[email protected]> Signed-off-by: Wey-Yi Guy <[email protected]> Tested-by: Helmut Schaa <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-19Bluetooth: ath3k: reduce memory usageAlexander Holler1-55/+20
There is no need to hold the firmware in memory. Signed-off-by: Alexander Holler <[email protected]> Signed-off-by: Gustavo F. Padovan <[email protected]>
2011-01-18iwlwifi: fix valid chain reading from EEPROMWey-Yi Guy1-1/+1
When read valid tx/rx chains from EEPROM, there is a bug to use the tx chain value for both tx and rx, the result of this cause low receive throughput on 1x2 devices becuase rx will only utilize single chain instead of two chains Signed-off-by: Wey-Yi Guy <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-18ath5k: fix locking in tx_complete_poll_workBob Copeland1-0/+4
ath5k_reset must be called with sc->lock. Since the tx queue watchdog runs in a workqueue and accesses sc, it's appropriate to just take the lock over the whole function. Signed-off-by: Bob Copeland <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-17ath9k_hw: do PA offset calibration only on longcal intervalRajkumar Manoharan1-5/+5
The power detector adc offset calibration has to be done on 4 minutes interval (longcal * pa_skip_count). But the commit "ath9k_hw: fix a noise floor calibration related race condition" makes the PA calibration executed more frequently beased on nfcal_pending value. Running PAOffset calibration lesser than longcal interval doesn't help anything and the worse part is that it causes NF load timeouts and RX deaf conditions. In a very noisy environment, where the distance b/w AP & station is ~10 meter and running a downlink udp traffic with frequent background scan causes "Timeout while waiting for nf to load: AR_PHY_AGC_CONTROL=0x40d1a" and moves the chip into deaf state. This issue was originaly reported in Android platform where the network-manager application does bgscan more frequently on AR9271 chips. (AR9285 family usb device). Cc: [email protected] Signed-off-by: Vasanthakumar Thiagarajan <[email protected]> Signed-off-by: Rajkumar Manoharan <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-17ath9k_htc: Fix endian issue in tx headerRajkumar Manoharan2-4/+6
Signed-off-by: Rajkumar Manoharan <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-17ath9k_hw: ASPM interoperability fix for AR9380/AR9382Luis R. Rodriguez2-3/+3
There is an interoperability with AR9382/AR9380 in L1 state with a few root complexes which can cause a hang. This is fixed by setting some work around bits on the PCIE PHY. We fix by using a new ini array to modify these bits when the radio is idle. Cc: [email protected] Cc: Jack Lee <[email protected]> Cc: Carl Huang <[email protected]> Cc: David Quan <[email protected]> Cc: Nael Atallah <[email protected]> Cc: Sarvesh Shrivastava <[email protected]> Signed-off-by: Luis R. Rodriguez <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-13rt2x00: Don't leak mem in error path of rt2x00lib_request_firmware()Jesper Juhl1-0/+1
We need to release_firmware() in order not to leak memory. Signed-off-by: Jesper Juhl <[email protected]> Acked-by: Ivo van Doorn <[email protected]> Acked-by: Pekka Enberg <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-13iwmc3200wifi: Return proper error for iwm_if_allocAxel Lin1-0/+2
In the case of alloc_netdev_mq failure and kmalloc failure, current implementation returns ERR_PTR(0). As a result, the caller of iwm_if_alloc does not catch the error by IS_ERR macro. Fix it by setting proper error code for ret variable in the failure cases. Signed-off-by: Axel Lin <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-10hostap_cs: fix sleeping function called from invalid contextStanislaw Gruszka1-9/+6
pcmcia_request_irq() and pcmcia_enable_device() are intended to be called from process context (first function allocate memory with GFP_KERNEL, second take a mutex). We can not take spin lock and call them. It's safe to move spin lock after pcmcia_enable_device() as we still hold off IRQ until dev->base_addr is 0 and driver will not proceed with interrupts when is not ready. Patch resolves: https://bugzilla.redhat.com/show_bug.cgi?id=643758 Reported-and-tested-by: [email protected] Cc: [email protected] # 2.6.34+ Signed-off-by: Stanislaw Gruszka <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-10ipw2200: Check for -1 INTA in tasklet too.Indan Zupancic1-0/+7
This is an attempt to fix a long standing open bug: http://bugzilla.intellinuxwireless.org/show_bug.cgi?id=1334 The interrupt handler checks for INTA being -1, apparently that means that the hardware is gone. But the interrupt handler defers actual interrupt processing to a tasklet. By the time the tasklet is run and checks INTA again, the hardware might be gone and INTA be -1, which confuses the driver because all event bits are set. The patch applies to 2.6.37. Signed-off-by: Indan Zupancic <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-10ssb: Ignore dangling ethernet cores on wireless devicesMichael Büsch1-0/+10
Some Broadcom based wireless devices contain dangling ethernet cores. This triggers the ssb probing mechanism and tries to load the b44 driver on this core. Ignore the dangling core in the ssb core scanning code to avoid access to the core and failure of b44 probing. Signed-off-by: Michael Buesch <[email protected]> Tested-by: Larry Finger <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-10p54: fix sequence no. accounting off-by-one errorChristian Lamparter1-1/+1
P54_HDR_FLAG_DATA_OUT_SEQNR is meant to tell the firmware that "the frame's sequence number has already been set by the application." Whereas IEEE80211_TX_CTL_ASSIGN_SEQ is set for frames which lack a valid sequence number and either the driver or firmware has to assign one. Yup, it's the exact opposite! Cc: <[email protected]> Signed-off-by: Christian Lamparter <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-10ath9k_htc: Really fix packet injectionSujith Manoharan2-11/+27
The chainmask value along with other configuration has to be set on the target for packet injection. Fix this and also move the monitor interface addition before the channel set segment to ensure that the opmode is updated properly. Signed-off-by: Sujith Manoharan <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-10ath9k_hw: Fix RX handling for USB devicesSujith Manoharan1-1/+1
Commit "ath9k_hw: Abort rx if hw is not coming out of full sleep in reset" uncondionally added aborting RX DMA in a HW reset, though it is a bit unclear as to why this is needed. Anyway, RX DMA is handled in the target for USB devices, and this would interfere with normal operations (scanning etc.), so fix this. Signed-off-by: Sujith Manoharan <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-10ath9k_hw: Fix thermal issue with UB94Sujith Manoharan1-0/+4
Hardcode the output voltage of x-PA bias LDO to the lowest value for UB94. The card doesn't get too hot now. Signed-off-by: Sujith Manoharan <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-10ath9k_hw: Fix calibration for AR9287 devicesSujith Manoharan1-0/+3
AR9287 based devices have issues with ADC gain calibration which would cause uplink throughput drops in HT40 mode. Remove ADC gain from the supported calibration algorithms. Signed-off-by: Sujith Manoharan <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-10ath9k_hw: Fix chip testSujith Manoharan1-1/+2
USB devices do not require the chip test routine. Signed-off-by: Sujith Manoharan <[email protected]> Signed-off-by: John W. Linville <[email protected]>
2011-01-09ixgbe: update ntuple filter configurationAlexander Duyck5-371/+250
This change fixes several issues found in ntuple filtering while I was doing the ATR refactor. Specifically I updated the masks to work correctly with the latest version of ethtool, I cleaned up the exception handling and added detailed error output when a filter is rejected, and corrected several bits that were set incorrectly in ixgbe_type.h. The previous version of this patch included a printk that was left over from me fixing the filter setup. This patch does not include that printk. Signed-off-by: Alexander Duyck <[email protected]> Tested-by: Stephen Ko <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2011-01-09ixgbe: further flow director performance optimizationsAlexander Duyck4-44/+194
This change adds a compressed input type for atr signature hash computation. It also drops the use of the set functions when setting up the ATR input since we can then directly setup the hash input as two dwords that can be stored and passed as registers. With these changes the cost of computing the has is low enough that we can perform a hash computation on each TCP SYN flagged packet allowing us to drop the number of flow director misses considerably in tests such as netperf TCP_CRR. Signed-off-by: Alexander Duyck <[email protected]> Tested-by: Stephen Ko <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2011-01-09ixgbe: cleanup flow director hash computation to improve performanceAlexander Duyck5-253/+182
This change cleans up the layout of the flow director data, and the algorithm used to calculate the hash resulting in a 35x / 3500% performance increase versus the old flow director hash computation. The overall effect is only a 1% increase in transactions per second though due to the fact that only 1 packet in 20 are actually hashed upon. TCP_RR before: Socket Size Request Resp. Elapsed Trans. Send Recv Size Size Time Rate bytes Bytes bytes bytes secs. per sec 16384 87380 1 1 60.00 23059.27 16384 87380 TCP_RR after: Socket Size Request Resp. Elapsed Trans. Send Recv Size Size Time Rate bytes Bytes bytes bytes secs. per sec 16384 87380 1 1 60.00 23239.98 16384 87380 Signed-off-by: Alexander Duyck <[email protected]> Tested-by: Stephen Ko <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2011-01-09ixgbe: make sure per Rx queue is disabled before unmapping the receive bufferYi Zou3-7/+39
When disable the Rx logic globally, we would also want to disable the per Rx queue receive logic by per queue Rx control register RXDCTL so no more DMA is happening from the packet buffer to the receive buffer associated with the Rx ring, before we start unmapping Rx ring receive buffer. The hardware may take max of 100us before the corresponding Rx queue is really disabled. Added ixgbe_disable_rx_queue() for this purpose. Signed-off-by: Yi Zou <[email protected]> Tested-by: Stephen Ko <[email protected]> Tested-by: Ross Brattain <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2011-01-09e1000: Add support for the CE4100 reference platformDirk Brandewie4-76/+365
This patch adds support for the gigabit phys present on the CE4100 reference platforms. Signed-off-by: Dirk Brandewie <[email protected]> Tested-by: Jeff Pieper <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2011-01-09e1000e: add custom set_d[0|3]_lplu_state function pointer for 82574Bruce Allan2-0/+57
82574 needs to configure Low Power Link Up (or LPLU) differently than the other parts in the 8257x family supported by the driver. Signed-off-by: Bruce Allan <[email protected]> Tested-by: Emil Tantilov <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2011-01-09e1000e: power off PHY after reset when interface is downBruce Allan3-35/+53
Some Phys supported by the driver do not remain powered off across a reset of the device when the interface is down, e.g. on 82571, but not on 82574. This patch powers down (only when WoL is disabled) the PHY after a reset if the interface is down and the ethtool diagnostics are not currently running. The ethtool diagnostic function required a minor re-factor as a result, and the e1000_[get|put]_hw_control() functions are renamed since they are no longer static to netdev.c as they are needed by the ethtool diagnostics. A couple minor whitespace issues were cleaned up, too. Reported-by: Arthur Jones <[email protected]> Signed-off-by: Bruce Allan <[email protected]> Tested-by: Jeff Pieper <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2011-01-09e1000e: use either_crc_le() rather than re-write itBruce Allan2-18/+2
For the 82579 jumbo frame workaround, there is no need to re-write the CRC calculation functionality already found in the kernel's ether_crc_le(). Signed-off-by: Bruce Allan <[email protected]> Tested-by: Jeff Pieper <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2011-01-09e1000e: properly bounds-check string functionsBruce Allan2-7/+11
Use string functions with bounds checking rather than their non-bounds checking counterparts, and do not hard code these boundaries. Signed-off-by: Bruce Allan <[email protected]> Tested-by: Emil Tantilov <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2011-01-09e1000e: convert calls of ops.[read|write]_reg to e1e_[r|w]phyBruce Allan3-53/+44
Cleans up the code a bit by using the driver-specific e1e_rphy and e1e_wphy macros instead of the full function pointer variants. Fix a couple whitespace issue with two already existing calls to e1e_wphy. Signed-off-by: Bruce Allan <[email protected]> Tested-by: Jeff Pieper <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.