aboutsummaryrefslogtreecommitdiff
path: root/drivers/net/ixgb
AgeCommit message (Collapse)Author
2006-10-05IRQ: Maintain regs pointer globally rather than passing to IRQ handlersDavid Howells
Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-09-27e100, e1000, ixgb: increment version numbersAuke Kok
e100-3.5.17-k2 e1000-7.2.9-k2 ixgb-1.0.117-k2 Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-09-27ixgb: convert to netdev_priv(netdev)Auke Kok
PCI error recovery code recently merged needs to use netdev_priv Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-09-27ixgb: combine more rx descriptors to improve performanceJesse Brandeburg
We experimented with more descriptor buffer writebacks and found that values larger than 8 give HW problems, but 8 is safe and gives us some improved performance. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-09-27e100, e1000, ixgb: Fix an impossible memory overwrite bugAuke Kok
We keep getting requests from people that think that this might be an exploitable hole where we would overwrite 4 bytes in the netdev struct if the pci name would exceed 15 characters. In reality this will never happen but we fix it anyway. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-09-27e100, e1000, ixgb: update copyright header and remove LICENSEAuke Kok
This update to the copyright header adds the mailinglist, and aligns it with the kernel licensing as well as remove the offending 'all rights reserved'. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-09-24Merge branch 'upstream-linus' of ↵Linus Torvalds
master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6 * 'upstream-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6: (217 commits) net/ieee80211: fix more crypto-related build breakage [PATCH] Spidernet: add ethtool -S (show statistics) [NET] GT96100: Delete bitrotting ethernet driver [PATCH] mv643xx_eth: restrict to 32-bit PPC_MULTIPLATFORM [PATCH] Cirrus Logic ep93xx ethernet driver r8169: the MMIO region of the 8167 stands behin BAR#1 e1000, ixgb: Remove pointless wrappers [PATCH] Remove powerpc specific parts of 3c509 driver [PATCH] s2io: Switch to pci_get_device [PATCH] gt96100: move to pci_get_device API [PATCH] ehea: bugfix for register access functions [PATCH] e1000 disable device on PCI error drivers/net/phy/fixed: #if 0 some incomplete code drivers/net: const-ify ethtool_ops declarations [PATCH] ethtool: allow const ethtool_ops [PATCH] sky2: big endian [PATCH] sky2: fiber support [PATCH] sky2: tx pause bug fix drivers/net: Trim trailing whitespace [PATCH] ehea: IBM eHEA Ethernet Device Driver ... Manually resolved conflicts in drivers/net/ixgb/ixgb_main.c and drivers/net/sky2.c related to CHECKSUM_HW/CHECKSUM_PARTIAL changes by commit 84fa7933a33f806bbbaae6775e87459b1ec584c0 that just happened to be next to unrelated changes in this update.
2006-09-22[NET]: Replace CHECKSUM_HW by CHECKSUM_PARTIAL/CHECKSUM_COMPLETEPatrick McHardy
Replace CHECKSUM_HW by CHECKSUM_PARTIAL (for outgoing packets, whose checksum still needs to be completed) and CHECKSUM_COMPLETE (for incoming packets, device supplied full checksum). Patch originally from Herbert Xu, updated by myself for 2.6.18-rc3. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-09-19e1000, ixgb: Remove pointless wrappersJeff Garzik
Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-09-13drivers/net: const-ify ethtool_ops declarationsJeff Garzik
Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-08-31ixgb: Increment version to 1.0.112-k2Auke Kok
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-08-31ixgb: remove skb->dev assignmentAuke Kok
Same change as e1000: remove skb->dev assignment, it's now done by netdev_alloc_skb. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-08-31ixgb: Add PCI Error recovery callbacksLinas Vepstas
Adds PCI Error recovery callbacks to the Intel 10-gigabit ethernet ixgb device driver. Lightly tested, works. "Zhang, Yanmin" <yanmin_zhang@linux.intel.com> wrote: Both pci_disable_device and ixgb_down would access the device. It doesn't follow Documentation/pci-error-recovery.txt that error_detected shouldn't do any access to the device. Signed-off-by: Linas Vepstas <linas@austin.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-08-31ixgb: Add buffer_info and test like e1000 has.Jesse Brandeburg
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-08-31ixgb: Cache-align all TX components of the adapter struct.Jesse Brandeburg
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-08-31ixgb: recalculate after how many descriptors to wake the queueAuke Kok
Recalculate when to wake the queue using DESC_NEEDED instead of a static hardcoded number. Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-08-31ixgb: Set a constant blink rate for ixgb adapter identify (1sec on, 1sec off)Jesse Brandeburg
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-08-31ixgb: convert dev->priv to netdev_priv(dev).Auke Kok
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-08-31ixgb: Convert dev_alloc_skb to netdev_alloc_skb.Auke Kok
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-08-24Merge branch 'upstream-fixes' of ↵Jeff Garzik
git://lost.foo-projects.org/~ahkok/git/netdev-2.6 into tmp
2006-08-19drivers/net: Remove deprecated use of pci_module_init()Jeff Garzik
From: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Michal Piotrowski <michal.k.k.piotrowski@gmail.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-08-16ixgb: Increment version to 1.0.109-k4Auke Kok
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-08-16ixgb: fix cache miss due to miscalculationJesse Brandeburg
Reduce writeback threshold by 1. We were instructing the hardware to wait until the 17th descriptor which went over the cache line limit. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke.jan.h.kok@intel.com>
2006-08-16ixgb: Add CX4 PHY type detection and subdevice ID.Manasi Deval
Signed-off-by: Manasi Deval <manasi.deval@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
2006-07-12[PATCH] ixgb: fix tx unit hang - properly calculate desciptor countAuke Kok
There were some tso bugs that only showed up with heavy load and 16kB pages that this patch fixes by making the driver's internal use count of descriptors match the count that it was estimating it needed using the DESC_NEEDED macro. This bug caused NETDEV_WATCHDOG resets aka tx timeouts. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
2006-07-08[NET] gso: Add skb_is_gsoHerbert Xu
This patch adds the wrapper function skb_is_gso which can be used instead of directly testing skb_shinfo(skb)->gso_size. This makes things a little nicer and allows us to change the primary key for indicating whether an skb is GSO (if we ever want to do that). Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-07-02[PATCH] irq-flags: drivers/net: Use the new IRQF_ constantsThomas Gleixner
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@elte.hu> Cc: "David S. Miller" <davem@davemloft.net> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Jeff Garzik <jeff@garzik.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-30Remove obsolete #include <linux/config.h>Jörn Engel
Signed-off-by: Jörn Engel <joern@wohnheim.fh-wedel.de> Signed-off-by: Adrian Bunk <bunk@stusta.de>
2006-06-23[NET]: Merge TSO/UFO fields in sk_buffHerbert Xu
Having separate fields in sk_buff for TSO/UFO (tso_size/ufo_size) is not going to scale if we add any more segmentation methods (e.g., DCCP). So let's merge them. They were used to tell the protocol of a packet. This function has been subsumed by the new gso_type field. This is essentially a set of netdev feature bits (shifted by 16 bits) that are required to process a specific skb. As such it's easy to tell whether a given device can process a GSO skb: you just have to and the gso_type field and the netdev's features field. I've made gso_type a conjunction. The idea is that you have a base type (e.g., SKB_GSO_TCPV4) that can be modified further to support new features. For example, if we add a hardware TSO type that supports ECN, they would declare NETIF_F_TSO | NETIF_F_TSO_ECN. All TSO packets with CWR set would have a gso_type of SKB_GSO_TCPV4 | SKB_GSO_TCPV4_ECN while all other TSO packets would be SKB_GSO_TCPV4. This means that only the CWR packets need to be emulated in software. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-05-26ixgb: update version, datesAuke Kok
increase the year dates to 2006 and bump the version to 1.0.109-k2 Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-26ixgb: remove changelogAuke Kok
same as e1000 - remove the changelog from the driver code itself. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-26ixgb: replace netdev->priv with netdev_priv()Auke Kok
fix netdev->priv ==> netdev_priv(netdev) Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-26ixgb: remove inlines, allow compiler to chooseAuke Kok
deinline a few large functions as to allow the compiler to pick. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-26ixgb: remove lock access in the fast pathAuke Kok
This mimics a change made in the e1000 driver that imitates a slick tg3 way of avoiding grabbing the lock around restarting the tx queue. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-26ixgb: allocate only buffersize neededAuke Kok
In order to help correct window size growth, use the MFS register to limit the packet sizes received and allocate only the buffer size necessary Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com> index 0905a82..84a8064 100644 --- a/drivers/net/ixgb/ixgb_main.c +++ b/drivers/net/ixgb/ixgb_main.c @@ -574,9 +574,8 @@ ixgb_sw_init(struct ixgb_adapter *adapte hw->subsystem_vendor_id = pdev->subsystem_vendor; hw->subsystem_id = pdev->subsystem_device; - adapter->rx_buffer_len = IXGB_RXBUFFER_2048; - hw->max_frame_size = netdev->mtu + ENET_HEADER_SIZE + ENET_FCS_LENGTH; + adapter->rx_buffer_len = hw->max_frame_size; if((hw->device_id == IXGB_DEVICE_ID_82597EX) || (hw->device_id == IXGB_DEVICE_ID_82597EX_CX4) @@ -820,21 +819,14 @@ ixgb_setup_rctl(struct ixgb_adapter *ada rctl |= IXGB_RCTL_SECRC; - switch (adapter->rx_buffer_len) { - case IXGB_RXBUFFER_2048: - default: + if (adapter->rx_buffer_len <= IXGB_RXBUFFER_2048) rctl |= IXGB_RCTL_BSIZE_2048; - break; - case IXGB_RXBUFFER_4096: + else if (adapter->rx_buffer_len <= IXGB_RXBUFFER_4096) rctl |= IXGB_RCTL_BSIZE_4096; - break; - case IXGB_RXBUFFER_8192: + else if (adapter->rx_buffer_len <= IXGB_RXBUFFER_8192) rctl |= IXGB_RCTL_BSIZE_8192; - break; - case IXGB_RXBUFFER_16384: + else if (adapter->rx_buffer_len <= IXGB_RXBUFFER_16384) rctl |= IXGB_RCTL_BSIZE_16384; - break; - } IXGB_WRITE_REG(&adapter->hw, RCTL, rctl); } @@ -1551,25 +1543,12 @@ ixgb_change_mtu(struct net_device *netde DPRINTK(PROBE, ERR, "Invalid MTU setting %d\n", new_mtu); return -EINVAL; } - - if((max_frame <= IXGB_MAX_ENET_FRAME_SIZE_WITHOUT_FCS + ENET_FCS_LENGTH) - || (max_frame <= IXGB_RXBUFFER_2048)) { - adapter->rx_buffer_len = IXGB_RXBUFFER_2048; - - } else if(max_frame <= IXGB_RXBUFFER_4096) { - adapter->rx_buffer_len = IXGB_RXBUFFER_4096; - } else if(max_frame <= IXGB_RXBUFFER_8192) { - adapter->rx_buffer_len = IXGB_RXBUFFER_8192; + adapter->rx_buffer_len = max_frame; - } else { - adapter->rx_buffer_len = IXGB_RXBUFFER_16384; - } - netdev->mtu = new_mtu; - - if(old_max_frame != max_frame && netif_running(netdev)) { + if ((old_max_frame != max_frame) && netif_running(netdev)) { ixgb_down(adapter, TRUE); ixgb_up(adapter); }
2006-05-26ixgb: revert an unwanted fix regarding tso/descriptorsAuke Kok
There seemed to be another bug introduced as well as a performance hit with the addtion of the sentinel descriptor workaround. Removal of this workaround appears to prevent the hang. We'll take a risk and remove it, as we had never seen the originally reported bug under linux. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-26ixgb: fix interface losing macaddr on ifdn/upAuke Kok
user contributed fix for LAA across down/up, from tonychung00@users.sf.net. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-26ixgb: fix smp polling race conditionAuke Kok
Moved interrupt masking to before requesting the interrupt from the OS. Moved interrupt enable to after netif_poll_enable. This fixes a racy BUG() where polling would be running on another CPU at the same time that netif_poll_enable would run. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-25ixgb: increment version to 1.0.104-k4Auke Kok
Increment the driver version to 1.0.104-k4 Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-25ixgb: use rx copybreak/skb recycleAuke Kok
o use rx copybreak/skb recycle Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-25ixgb: add performance enhancements to the buffer_info structAuke Kok
o modify the rx refill logic and tail bump o add counter for failures Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-23ixgb: increment version to 1.0.104-k2Auke Kok
Increment the driver version to 1.0.104-k2 Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-23ixgb: add tx timeout counterAuke Kok
This adds a TX timeout counter to the ethtool stats, a tx timeout debug message, and sets the timer to 5 seconds. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-23ixgb: use DPRINTK and msglvl, and ethtool to control itAuke Kok
Use DPRINTK and msglvl, and ethtool to control it. Add proper names to netdev structs and mappings. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-23ixgb: remove hardcoded numberAuke Kok
This adds a define for an awkward and uncommented value. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-23ixgb: add copper 10gig driver idAuke Kok
Add support for Copper 10GbE device ID 109E Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-23ixgb: add NETIF_F_LLTX analogous to e1000Auke Kok
add NETIF_F_LLTX code like e1000 has Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-23ixgb: fix flow controlAuke Kok
Make default flow control only have *sending* of flow control packets enabled, and fix to disable / enable flow control correctly. Set flow control defaults to disable receiving flow control from the link partner, to fix the transmit fifo overlow errata Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-23ixgb: remove duplicate code setting duplex and speedAuke Kok
Removed duplicate code. Create ixgb_set_speed_duplex function to contain duplicate code. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>
2006-05-23ixgb: fix rare early tso completionAuke Kok
Fix rare early completion when using TSO. This essentially is the e1000 fix, with code that was mostly already written. Another skb frag was also needed. Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com> Signed-off-by: John Ronciak <john.ronciak@intel.com>