aboutsummaryrefslogtreecommitdiff
path: root/drivers/net/sfc
AgeCommit message (Collapse)AuthorFilesLines
2009-10-24sfc: Move efx_xmit_done() declaration into correct stanzaBen Hutchings1-1/+1
Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Remove declarations of nonexistent functionsBen Hutchings1-2/+0
Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Change order of device removal to reverse of probe orderBen Hutchings1-1/+1
This makes efx_pci_remove_main() more obviously the inverse of efx_pci_probe_main(), and matches our out-of-tree driver. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Merge struct efx_blinker into struct efx_boardBen Hutchings2-29/+21
Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Move all TX DMA length limiting into tx.cBen Hutchings4-25/+25
Replace the duplicated logic in efx_enqueue_skb() and efx_tx_queue_insert() with an inline function, efx_max_tx_len(). Remove the failed attempt at abstracting hardware-specifics and put all the magic numbers in efx_max_tx_len(). Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Define DMA address mask explicitly in terms of descriptor field widthBen Hutchings1-5/+2
Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Eliminate indirect lookups of queue size constantsBen Hutchings7-98/+62
Move size and mask definitions into efx.h; calculate page orders in falcon.c. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Rename register I/O header and functions used by both Falcon and SienaBen Hutchings7-421/+426
While we're at it, use type suffixes of 'd', 'q' and 'o', consistent with register type names. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Update hardware definitions for SienaBen Hutchings7-1887/+3756
Siena is still based on the Falcon hardware architecture and will share many of these definitions, so replace falcon_hwdefs.h with regs.h. The new definitions have been generated according to a naming convention which incorporates the type and revision information. Update the code accordingly. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Move RX data FIFO thresholds out of struct efx_nic_typeBen Hutchings2-15/+12
Since there are now separate blocks of code to set the thresholds for each NIC type, it is no longer useful to include them in the NIC type description. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Remove versioned bitfield macrosBen Hutchings3-40/+47
These macros are not extensible to more than two NIC types without repetition of register definitions, and they are only used to deal with a few fields in RX_CFG_REG and global events which moved between Falcon rev A1 and B0. Therefore: - Move RX_CFG_REG initialisation into its own function which tests the NIC revision just once - Explicitly test the NIC revision when checking the RX_RECOVERY flag in global events - Merge definitions of RX_XOFF_MAC_EN flag, which did not move - Remove the macro definitions Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Remove boards.h, moving last remaining declaration to falcon.hBen Hutchings9-22/+2
Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Merge sfe4001.c into falcon_boards.cBen Hutchings4-443/+420
Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Rename Falcon-specific board code and typesBen Hutchings5-26/+25
Siena will require entirely different board code. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Remove redundant hardware initialisationBen Hutchings1-4/+0
Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-24sfc: Remove redundant header gmii.hBen Hutchings2-61/+0
Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-22sfc: 10Xpress: Report support for pause framesBen Hutchings1-0/+1
Commits 27fbc7d 'mdio: Expose pause frame advertising flags to ethtool' and c634263 'sfc: 10Xpress: Initialise pause advertising flags' added to our reported advertising flags. efx_mdio_set_settings() requires that all advertising flags are also present in the supported flags, so make sure that is true. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-10-12sfc: 10Xpress: Initialise pause advertising flagsBen Hutchings1-0/+10
The mdio module now handles reconfiguration of pause advertising through ethtool, but not initialisation. Add the necessary initialisation to tenxpress_phy_init(). Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-09-24cpumask: use zalloc_cpumask_var() where possibleLi Zefan1-2/+1
Remove open-coded zalloc_cpumask_var() and zalloc_cpumask_var_node(). Signed-off-by: Li Zefan <[email protected]> Signed-off-by: Rusty Russell <[email protected]>
2009-09-02netdev: drivers should make ethtool_ops constStephen Hemminger2-2/+2
No need to put ethtool_ops in data, they should be const. Signed-off-by: Stephen Hemminger <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-09-01netdev: convert bulk of drivers to netdev_tx_tStephen Hemminger4-14/+15
In a couple of cases collapse some extra code like: int retval = NETDEV_TX_OK; ... return retval; into return NETDEV_TX_OK; Signed-off-by: Stephen Hemminger <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-08-26sfc: Improve reliability of RX queue flushingBen Hutchings1-0/+2
Reconfiguring the port requires us to flush all DMA queues. In repeated testing we have found that RX flushes would sometimes fail because the RX DMA engine was not properly isolated from the MACs. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-08-26sfc: Work around XMAC bug causing packet loss with some peersBen Hutchings1-1/+6
Received frames must be re-clocked by the local XGXS to the 156.25 MHz (DDR) clock of the XGMII. If the remote clock is slightly faster this can reduce a minimum IPG of 64 bit-times (1 cycle) to 32 bit-times (half a cycle). If the XMAC detects that a frame has reached the maximum RX frame length in the same cycle that it receives one of these reduced IPGs, it may miss the IPG, causing two valid frames to be treated as a single invalid frame (over-length with bad CRC). We work around this by increasing the maximum RX frame length so that peers with matched MTU will not provoke this bug. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-08-26sfc: QT2025C: Use hard reset onlyBen Hutchings1-9/+10
At probe time, falcon_reset_hw() performs a hard reset of the PHY along with Falcon. There is no need to perform a soft reset later, and any access to standard MDIO registers before the PHY firmware has booted can interrupt the boot process, making the port unusable. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-08-26sfc: Do not reinitialise XAUI serdes before it has completed resetBen Hutchings2-3/+7
falcon_reset_xaui() waits for XGXS reset to complete, but the XAUI serdes reset may take longer. It needs to check both reset active bits. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-08-26sfc: Fix ordering of device registration and initial netif_carrier_off()Ben Hutchings1-9/+17
We must call netif_carrier_off() after the device is registered, not before, to set the operational state and user-space IFF_RUNNING flag correctly. Since we don't want observers to see an intermediate state, open-code register_netdev() and add efx_update_name() and netif_carrier_off() into the locked region. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-06-11sfc: Expose 100/1000BASE-T MDI-X status via ethtoolBen Hutchings1-1/+10
Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-05-29net: dont update dev->trans_start in 10GB driversEric Dumazet2-7/+1
Followup of commits 9d21493b4beb8f918ba248032fefa393074a5e2b and 08baf561083bc27a953aa087dd8a664bb2b88e8e (net: tx scalability works : trans_start) (net: txq_trans_update() helper) Now that core network takes care of trans_start updates, dont do it in drivers themselves, if possible. Multi queue drivers can avoid one cache miss (on dev->trans_start) in their start_xmit() handler. Exceptions are NETIF_F_LLTX drivers (vxge & tehuti) Signed-off-by: Eric Dumazet <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-05-19sfc: Remove lro module parameterBen Hutchings1-13/+2
GRO/LRO can be controlled through ethtool so this is unnecessary. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-05-17sfc: Use generic XENPAK register definitionsBen Hutchings3-73/+4
Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-05-17sfc: modify allocation error messageMike Travis1-2/+2
Change error message when alloc_cpumask_var fails. Repairs "cpumask: convert drivers/net/sfc". Signed-off-by: Mike Travis <[email protected]> Acked-by: Ben Hutchings <[email protected]> Acked-by: David S. Miller <[email protected]> Cc: Rusty Russell <[email protected]> Cc: Ingo Molnar <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-04-29sfc: Use generic MDIO flow control auto-negotiation functionsBen Hutchings5-48/+7
Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-04-29sfc: Use generic MDIO functions and definitionsBen Hutchings14-845/+311
Make use of the newly-added generic MDIO clause 45 support and remove redundant definitions. Add an 'efx_' prefix to the remaining driver-specific MDIO functions and remove arguments which are redundant with efx->mdio.prtad. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-04-16Merge branch 'master' of ↵David S. Miller2-5/+6
master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
2009-04-16gro: New frags interface to avoid copying shinfoHerbert Xu1-8/+18
It turns out that copying a 16-byte area at ~800k times a second can be really expensive :) This patch redesigns the frags GRO interface to avoid copying that area twice. The two disciples of the frags interface have been converted. Signed-off-by: Herbert Xu <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-04-14sfc: Use correct macro to set event bitfieldBen Hutchings1-2/+2
falcon_sim_phy_event() used EFX_OWORD_FIELD, which operates on bitfields in 128-bit values, on an event, which is a 64-bit value. This should be harmless - these macros always use little-endian ordering, so it would read and write back the following 8 bytes unchanged - but it is obviously wrong. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-04-14sfc: Match calls to netif_napi_add() and netif_napi_del()Ben Hutchings1-3/+4
sfc could call netif_napi_add() multiple times for the same napi_struct, corrupting the list of napi_structs for the associated device and leading to a busy-loop on device removal. Move the call to netif_napi_add() and add a call to netif_napi_del() in the obvious places. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-04-08sfc: Don't specify unexistent IRQJean Delvare2-5/+0
Neither the lm90 driver nor the lm87 driver do support interrupts, so there is no point in specifying one when declaring the devices. Signed-off-by: Jean Delvare <[email protected]> Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-03-31sfc: Call netif_napi_add() before registering our interrupt handlerBen Hutchings1-4/+3
If we use a shared legacy IRQ then our interrupt handler may be called as soon as it is registered even though IRQs are disabled on the NIC. Now that the legacy interrupt handler also checks for event delivery, it may decide to schedule polling in this case. Ensure that the NAPI context is valid but disabled at this point. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-03-27Merge branch 'core/percpu' into percpu-cpumask-x86-for-linus-2Ingo Molnar2-17/+24
Conflicts: arch/parisc/kernel/irq.c arch/x86/include/asm/fixmap_64.h arch/x86/include/asm/setup.h kernel/irq/handle.c Semantic merge: arch/x86/include/asm/fixmap.h Signed-off-by: Ingo Molnar <[email protected]>
2009-03-21sfc: Implement adaptive IRQ moderationBen Hutchings6-21/+73
Calculate a score for each 1000 IRQs: - TX completions are worth 1 point - RX completions are worth 4 if merged using LRO or 2 otherwise Reduce moderation if the score is less than 10000, down to a minimum of 5 us. Increase moderation if the score is more than 20000, up to the specified maximum. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-03-21sfc: Optimise falcon_writel_page_locked() for page > 0Ben Hutchings1-5/+8
The bug this function works around only applies to the first set of page-mapped registers; other pages can be written without locking. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-03-21sfc: Remove unused private PCI register definitionsBen Hutchings1-7/+0
Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-03-21sfc: Work around unreliable legacy interrupt statusBen Hutchings1-12/+14
In rare cases, reading the legacy interrupt status register can acknowledge an event queue whose attention flag has not yet been set in the register. Until we service this event queue it will not generate any more interrupts. Therefore, as a secondary check, poll the next slot in each active event queue whose flag is not set. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-03-21sfc: Pad packets to 33 bytes to prevent TX packet parser lockupBen Hutchings2-0/+10
The packet parser used in the TX data path for locating checksum fields can lose synchronisation with the TX queue manager when handling packets that look like IPv4 but are too short (17-32 bytes). Work around this by padding to 33 bytes. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-03-04sfc: Improve NIC internal error recoveryBen Hutchings1-4/+19
Make the error count a per-NIC variable. Reset this the count after an hour if it has not reached the critical value. Set the critical value back to 5. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-03-04sfc: Fix search for flush completion eventsBen Hutchings1-12/+9
When flushing queues we disable normal interrupt and event handling and poll event queue 0 looking for flush completions. Unfortunately the flush event polling loop fails to move past any other type of event. This problem has not been observed in production hardware but appears to be a possibility. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-03-04sfc: Don't wake TX queues while they're being flushedBen Hutchings1-1/+1
Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-03-04sfc: Clear I2C adapter structure in falcon_remove_nic()Ben Hutchings1-0/+2
i2c_del_adapter() leaves dangling pointers in the structure. If we retry the NIC probe and pass the structure to i2c_add_adapter() again it will lead to an oops unless we clear it first. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2009-03-04sfc: Clean up properly on reset failure pathsBen Hutchings1-3/+6
If MAC switching fails, stop the port properly. If PHY reinitialisation fails, clear the port_initialized flag. Signed-off-by: Ben Hutchings <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.