summaryrefslogtreecommitdiff
path: root/drivers/net/ethernet/broadcom/tg3.c
AgeCommit message (Collapse)AuthorFilesLines
2012-03-22tg3: Fix RSS ring refill race conditionMichael Chan1-3/+22
The RSS feature in tg3 hardware has only one rx producer ring for all RSS rings. NAPI vector 1 is special and handles the refilling of the rx producer ring on behalf of all RSS rings. There is a race condition between these RSS NAPIs and the NAPI[1]. If NAPI[1] finishes checking for refill and then another RSS ring empties the rx producer ring before NAPI[1] exits NAPI, the chip will be completely out of SKBs in the rx producer ring. We fix this by adding a flag tp->rx_refill and rely on napi_schedule()/ napi_complete() to help synchronize it to close the race condition. Update driver version to 3.123. Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-10Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/netDavid S. Miller1-3/+3
2012-03-06tg3: Fix to use multi queue BQL interfacesTom Herbert1-3/+3
Fix tg3 to use BQL multi queue related netdev interfaces since the device supports multi queue. Signed-off-by: Tom Herbert <therbert@google.com> Reported-by: Christoph Lameter <cl@gentwo.org> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-05tg3: Recode PCI MRRS adjustment as a PCI quirkMatt Carlson1-6/+0
This patch recodes the MRRS cap for 5719 A0 devices as a PCI quirk. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-05tg3: Fix poor tx performance on 57766 after MTU changeMichael Chan1-2/+8
GRC reset causes the read DMA engine to go into a mode that breaks up requests into 256 bytes. A PHY reset is required to bring it back to the normal mode. Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-05tg3: Add memory barriers to sync BD dataMichael Chan1-0/+9
for weak memory model architectures to ensure that the chip will DMA valid BD data. Signed-off-by: Michael Chan <mchan@broadcom.com> Reviewed-by: Benjamin Li <benli@broadcom.com> Reviewed-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-05tg3: Fix jumbo loopback test on 5719Michael Chan1-3/+11
Loopback on 9K packet fails because the chip has a DMA limit of 4K. The loopback test logic uses a single BD for simplicity. Fix it by reducing the jumbo packet size to the DMA limit. Signed-off-by: Michael Chan <mchan@broadcom.com> Reviewed-by: Benjamin Li <benli@broadcom.com> Reviewed-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-02Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/netDavid S. Miller1-21/+22
Conflicts: drivers/net/ethernet/broadcom/tg3.c Conflicts in the statistics regression bug fix from 'net', but happily Matt Carlson originally posted the fix against 'net-next' so I used that to resolve this. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-29tg3: Fix tg3_get_stats64 for 5700 / 5701 devsMatt Carlson1-22/+23
tg3_get_stats64() takes tp->lock when dealing with non-serdes bcm5700 and bcm5701 devices. However, functions that call tg3_halt() have already acquired tp->lock. When tg3_get_stats64() is called in tg3_halt(), deadlock will occur. This patch fixes the problem by separating the stat gathering code into a new tg3_get_nstats() function. tg3_get_stats64() is recoded to call this function and take tp->lock. The code that takes tp->lock in tg3_calc_crc_errors() has been removed. Function signatures have been cleaned up too. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-24ethernet: unify return value of .ndo_set_mac_address if address is invalidDanny Kukawka1-1/+1
Unify return value of .ndo_set_mac_address if the given address isn't valid. Return -EADDRNOTAVAIL as eth_mac_addr() already does if is_valid_ether_addr() fails. Signed-off-by: Danny Kukawka <danny.kukawka@bisect.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-23tg3: Create timer helper functionsMatt Carlson1-96/+112
This patch seeks to clean up the timer related code. It begins by moving one-time timer setup code from tg3_open() to tg3_init_one(). It then creates a function that encapsulates the code needed to start the timer. A tg3_timer_stop() function was added for parity. Finally, this patch moves all the timer functions to a more suitable location. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-23tg3: Clear RECOVERY_PENDING with reset_task_cancelMatt Carlson1-1/+1
If an error happens in the tx completion thread, tg3_reset_task will be scheduled and TX_RECOVERY_PENDING will be set. The TX_RECOVERY_PENDING flag causes tg3_poll[_msix] to return early before doing much of its work. Tg3_reset_task() gets canceled when the configuration of the device is changing, which always results in a chip reset. When this happens, the TX_RECOVERY_PENDING flag may be left set, which would unnecessarily hinder tg3_poll from doing work. This patch fixes the problem. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-23tg3: Remove SPEED_UNKNOWN checksMatt Carlson1-44/+23
tg3_phy_copper_begin() has code that configures the link advertisements through the use of the link_config.speed and link_config.duplex members. The driver does not internally use these members in this way, nor is it (currently) permitted via the ethtool interface. This patch removes the dead code. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-23tg3: Fix link check in tg3_adjust_linkMatt Carlson1-2/+4
The tg3 driver tried to detect link changes by comparing the tg3 local active_speed member with SPEED_UNKNOWN (or formerly SPEED_INVALID). This check is not correct, since phylib will never set its speed member to either of these two values. The code only appeared to work because tg3 initializes active_speed to SPEED_INVALID during tg3_init_one. This patch introduces a new "old_link" tg3 member and then compares the phy_device's link member against it to detect link state changes. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-22tg3: remove IRQF_SAMPLE_RANDOM flagDavidlohr Bueso1-1/+1
This driver is the last user of the IRQF_SAMPLE_RANDOM flag for net drivers, and since add_*_randomness interfaces have now deprecated the flag as a source of external noise, we can remove it. Signed-off-by: Davidlohr Bueso <dave@gnu.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Update copyrightMatt Carlson1-1/+1
This patch updates the copyright dates in the tg3 driver. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Use *_UNKNOWN ethtool definitionsMatt Carlson1-20/+20
This patch replaces tg3's private SPEED_INVALID and DUPLEX_INVALID definitions with SPEED_UNKNOWN and DUPLEX_UNKNOWN respectively. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Remove unneeded link_config.orig_... membersMatt Carlson1-27/+12
This patch removes the unneeded link_config.orig_* members. When the phy is in a low power state, the TG3_PHYFLG_IS_LOW_POWER flag solely determines how the link is configured. When the phy is powered back up, it can resume using the original settings. For the phylib case, the link configuration still needs to be saved, but since the phylib maintains its own configuration, we can repurpose the (unused in this case) tg3 link configuration members. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Remove unused link config codeMatt Carlson1-5/+1
When tg3 devices are shutdown, the driver uses the TG3_PHYFLG_IS_LOW_POWER flag to enable a static phy configuration. Any attempt to use the link configuration variables is dead code. This patch removes such code. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Consolidate ASIC rev detection codeMatt Carlson1-80/+79
Detecting the ASIC revision of a device is getting to be an increasingly complex process. This patch consolidates all the ASIC rev detection code to a single routine for better maintainability. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Reduce UMP event collision windowMatt Carlson1-16/+26
The tg3 driver needs to submit a few phy register values to the UMP firmware each time the link state changes. Up until now, the driver would wait for the previous event to complete, then proceed to gather data through a series of phy accesses. Since phy accesses are relatively slow, it is possible for another thread to attempt to submit its own event while the UMP code is still construction its message. This patch seeks to minimize the collision window as much as possible by preloading the phy data. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Fix NVRAM page writes on newer devicesMatt Carlson1-2/+5
On newer devices, the hardware expects the NVRAM address register to be written only once per NVRAM page. To do otherwise causes NVRAM corruption. This patch fixes the problem. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Fix copper autoneg adv checksMatt Carlson1-1/+10
When checking the autoneg advertisements, the driver failed to include the master and master enable bits for the bcm5701. This patch fixes the problem. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Fix stats while interface is downMatt Carlson1-4/+4
If the tg3 interface is down, the driver will return ethtool stats uninitialized. This patch zeroes out the destination stat buffer in such a case. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Disable new DMA engine for 57766Matt Carlson1-5/+0
A bug was found in the new DMA engine for the 57766. This patch disables it, which causes the device to fallback to the old DMA engine. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Move transmit comment to a better locationMatt Carlson1-1/+3
This patch moves a comment in the transmit path to a better location. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Eliminate unneeded prototypeMatt Carlson1-3/+2
This patch eliminates the unneeded tg3_halt_cpu() prototype and moves the tg3_setup_phy() prototype closer to where it is needed. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Relocate tg3_find_peerMatt Carlson1-30/+28
This patch relocates tg3_find_peer to eliminate a prototype. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Move tg3_nvram_write_block functionsMatt Carlson1-250/+250
This patch moves the tg3_nvram_write_block functions higher in the file to eliminate a prototype. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Move tg3_set_rx_modeMatt Carlson1-100/+99
This patch moves __tg3_set_rx_mode above its first use and moves tg3_set_rx_mode down closer to where the netdev_ops functions should be. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Move tg3_change_mtu to a better locationMatt Carlson1-81/+78
This patch moves tg3_change_mtu to a better location. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Relocate tg3_reset_taskMatt Carlson1-47/+45
This patch moves tg3_reset_task further down in the file where it makes more sense to be. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14tg3: Move tg3_restart_hw to a better locationMatt Carlson1-27/+29
This patch relocates tg3_restart_hw() to a better location. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-01drivers/net: Remove alloc_etherdev error messagesJoe Perches1-1/+0
alloc_etherdev has a generic OOM/unable to alloc message. Remove the duplicative messages after alloc_etherdev calls. Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-23tg3: fix ipv6 header length computationEric Dumazet1-7/+2
tg3_start_xmit() makes the wrong assumption for TSOV6 that skb->head doesnt include any payload data. if (skb_is_gso_v6(skb)) hdr_len = skb_headlen(skb) - ETH_HLEN; This is not true anymore after commit f07d960df3 (tcp: avoid frag allocation for small frames) We should instead use : skb_transport_offset(skb) + tcp_hdrlen(skb) Its also true for IPv4 Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> CC: Matt Carlson <mcarlson@broadcom.com> CC: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-18tg3: Fix single-vector MSI-X codeMatt Carlson1-13/+14
Kdump kernels leave MSI-X interrupts (as setup by the crashed kernel) enabled. However, kdump only enables one CPU in the new environment, thus causing tg3 to abort MSI-X setup. When the driver attempts to enable INTA or MSI interrupt modes on a kdump kernel, interrupt delivery fails. This patch attempts to workaround the problem by forcing the driver to enable a single MSI-X interrupt. In such a configuration, the device's multivector interrupt mode must be disabled. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-20net: fix assignment of 0/1 to bool variables.Rusty Russell1-4/+4
DaveM said: Please, this kind of stuff rots forever and not using bool properly drives me crazy. Joe Perches <joe@perches.com> gave me the spatch script: @@ bool b; @@ -b = 0 +b = false @@ bool b; @@ -b = 1 +b = true I merely installed coccinelle, read the documentation and took credit. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-20tg3: Make the RSS indir tbl admin configurableMatt Carlson1-7/+101
This patch adds the ethtool callbacks necessary to change the rss indirection table from userspace. Should the number of interrupts change (e.g. across a close / open call, or through a reset) and any one of the indirection table values fall out-of-range, the driver will reset the indirection table to a default layout. [Integrated many suggestions made by Ben Hutchings.] Changes since v3 * Removed TG3_FLAG_SUPPORT_MSIX checks at the start of tg3_get_rxfh_indir() and tg3_set_rxfh_indir(). Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: Michael Chan <mchan@broadcom.com> Reviewed-by: Benjamin Li <benli@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-15tg3: Break out RSS indir table init and assignmentMatt Carlson1-23/+34
This patch creates a new device member to hold the RSS indirection table and separates out the code that initializes the table from the code that programs the table into device registers. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Reviewed-by: Benjamin Li <benli@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-15tg3: Use mii_advertise_flowctrlMatt Carlson1-19/+3
This patch replaces tg3's internal tg3_advert_flowctrl_1000T function with mii_advertise_flowctrl provided by the kernel headers. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Reviewed-by: Benjamin Li <benli@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-15tg3: Add 57766 ASIC rev supportMatt Carlson1-17/+41
This patch adds support for the 57766 ASIC revision. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Reviewed-by: Benjamin Li <benli@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-15tg3: Make the TX BD DMA limit configurableMatt Carlson1-8/+8
The 57766 ASIC rev will impose a new TX BD DMA limit on the driver. This patch prepares for 57766 support by making the tx BD DMA limit tunable. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Reviewed-by: Benjamin Li <benli@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-15tg3: Enable EEE support for capable 10/100 devsMatt Carlson1-10/+9
There are some devices in the 57765 ASIC rev that are EEE capable. Unfortunately the EEE setup code only gets executed if the device is gigabit capable. This patch fixes the problem. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Reviewed-by: Benjamin Li <benli@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-09tg3: Update version to 3.122Matt Carlson1-2/+2
This patch updates the tg3 version to 3.122. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Reviewed-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-09tg3: Return flowctrl config through ethtoolMatt Carlson1-2/+2
This patch changes the driver to return the flow control configuration rather than the flow control status through the ETHTOOL_GPAUSEPARAM ioctl. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Reviewed-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-09tg3: Track LP advertisingMatt Carlson1-3/+37
This patch adds code to track the autonegotiation advertisements of the link partner and report them through ethtool. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-09tg3: Integrate flowctrl check into AN adv checkMatt Carlson1-56/+25
This patch integrates tg3_adv_1000T_flowctrl_ok() into tg3_copper_is_advertising_all() and renames the function tg3_phy_copper_an_config_ok(). Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Reviewed-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-09tg3: Save stats across chip resetsMatt Carlson1-9/+17
Tg3 has a place to store stats, but doesn't really use it. This patch modifies the driver so that stats are saved across chip resets and gets cleared across close / open calls. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Reviewed-by: Michael Chan <mchan@broadcom.com> Reviewed-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-09tg3: Remove ethtool stats member from dev structMatt Carlson1-7/+7
This patch removes the ethtool stats member from the tg3 device structure. Signed-off-by: Matt Carlson <mcarlson@broadcom.com> Signed-off-by: Michael Chan <mchan@broadcom.com> Reviewed-by: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-11-29tg3: Support for byte queue limitsTom Herbert1-0/+8
Changes to tg3 to use byte queue limits. Signed-off-by: Tom Herbert <therbert@google.com> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
generated by cgit v1.2.3 (git 2.39.0) at 2025-09-26 18:15:47 +0300