summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2012-08-10Add O2 Micro OZ776 HubDavid Coppa
Ok jsg@
2012-08-10cleanup ixgbe_start routine; from brad, ok jsgMike Belopuhov
2012-08-09don't compile rss functions in unless OCE_RSS is specifiedMike Belopuhov
2012-08-09oops. missed these conflicts in the previous commitMike Belopuhov
2012-08-09try to make sense of the firmware statistics by counting rx and tx erorrsMike Belopuhov
2012-08-09schedule a rx refill if the ring is empty, sprinkle some dma syncsMike Belopuhov
2012-08-09more htole conversionsMike Belopuhov
2012-08-09remove internal queue stats, use if_* counters wherever possibleMike Belopuhov
2012-08-09respect the mtu value that ifconfig setsMike Belopuhov
2012-08-09set a timeout in case the chip goes out to lunchMike Belopuhov
2012-08-09remove oce_dmamap_sync wrapperMike Belopuhov
2012-08-09Add support for BCM5717/19/20C PHYs. From Brad.Mark Kettenis
2012-08-09regenMark Kettenis
2012-08-09Add BCM5717/19/20C PHYs. From Brad.Mark Kettenis
2012-08-08call m_cluncount to account correctly for cluster chains the driverMike Belopuhov
builds upon receiving a jumbo frame.
2012-08-08We're not going to loop in the rxeof refilling our ring as it wouldMike Belopuhov
work against the mclgeti algorithm and besides it doesn't make any difference if [repaired and] enabled since interrupt mitigation was fixed some time ago. So remove the leftovers altogether so that nobody would be tempted to use them. ok claudio, jsg
2012-08-08remove rx debugging codeMike Belopuhov
2012-08-08Do not depend upon an initialized variable in the common case; gives init(8)Miod Vallat
a chance to run...
2012-08-08Make this compile again.Miod Vallat
2012-08-08Convert from K&R to ISO function signatures.Matthew Dempsky
From Brad; no binary change on amd64.
2012-08-07Make all *_device_*_start() functions (e.g. ohci_device_intr_start())Kenneth R Westerback
wait for an interrupt when the bus is in polling mode. Otherwise some devices like my YE-Data USB Floppy take a short sharp trip to ddb> at 'halt -p'. Most of the functions were already doing the wait. ok miod@. With modification that printf's become DPRINTF's.
2012-08-07Fix compilation when !defined(KTRACE) by hiding the then-unused 'code'Philip Guenthe
variable Pointed out by Amit Kulkarni (amitkulz at gmail.com)
2012-08-07Handle __syscall as syscall, since these are identical on 64-bit platforms,Miod Vallat
fixing an old n32-inherited bug in the process. Spotted during guenther@'s syscall factorization work.
2012-08-07Store the data used to generate an ICMP error message on a stackMike Belopuhov
instead of allocating a new mbuf. This is a third or fourth attempt to incorporate a change like this meaning a handful of people have lost their hair trying to make it work, namely dlg@, henning@, deraadt@, and thib@. Unfortunately the fixed version was never put back which is exceptionally unfortunate since the impact on performance is huge: it nearly doubles the forwarding performance on selected hardware in simple setups. So after being beaten in test and production environments on several architectures it's ready to be put back again. We're doing it early in the release cycle so that it will receive a good test exposure. ok derradt, henning
2012-08-07<sys/ktrace.h> is now pulled in via <sys/syscall_mi.h>, so drop it from herePhilip Guenthe
2012-08-07a bit of cleanup and a tx delay tweakMike Belopuhov
2012-08-07Don't include <sys/ktrace.h>; makes this file compile again.Mark Kettenis
2012-08-07don't forget to set IFF_ALLMULTI; from bradMike Belopuhov
2012-08-07Move the common bits of syscall invocation and return handling intoPhilip Guenthe
an MI file, <sys/syscall_mi.h>, correcting inconsistencies and the handling when copyin() of arguments fails. Tested on i386, amd64, sparc64, and alpha (thanks naddy@) Any issues with other platforms will be fixed in tree. header name from millert@; ok miod@
2012-08-06kill sc->promiscMike Belopuhov
2012-08-06Add support for 10Gb ethernet cards based on the Intel X540 chipset.Mike Belopuhov
The code was obtained from FreeBSD and tested on the hardware kindly donated by Tony Sarendal <tony () polarcap ! org>. Thanks a lot! ok jsg
2012-08-05Remove a self-assignment, from BradMartin Pieuchot
2012-08-04PCI_PRODUCT_ATI_RADEON_X1250 -> PCI_PRODUCT_ATI_RADEON_X1250_{1,2}Antoine Jacoutot
prodded by matthieu@
2012-08-04Regen.Antoine Jacoutot
2012-08-040x7942 is also a Radeon Xpress 1500Antoine Jacoutot
ok matthieu@
2012-08-03enable oce(4)Mike Belopuhov
2012-08-03add HUAWEI E392, tested by Alexei MalininJonathan Gray
2012-08-03regenJonathan Gray
2012-08-03HUAWEI E392Jonathan Gray
2012-08-02get rid of some "capabilities" leftoversMike Belopuhov
2012-08-02we use a central pci database; no reason to keep duplicates aroundMike Belopuhov
2012-08-02Add a driver for Emulex OneConnect 10Gb Ethernet obtained from FreeBSDMike Belopuhov
but heavily massaged to look like other BSD network drivers. Support is provided for cards based on the following controllers: o ServerEngines BladeEngine 2 o ServerEngines BladeEngine 3 o Emulex Lancer
2012-08-02regenMike Belopuhov
2012-08-02Emulex Lancer (XE201)Mike Belopuhov
2012-08-02remove leftover NOLINT, WANTLINT, LINTFLAGS, LOBJ vars and lint targets.Okan Demirmen
ok guenther@
2012-08-02Apply profiling to all threads instead of just the thread that calledPhilip Guenthe
profil() by moving P_PROFIL from proc->p_flag to process->ps_flags with matching adjustment in fork1() and exit1() ok matthew@
2012-08-01Use mtrrmask instead of hardcoded values to accommodate newer CPUsMike Belopuhov
with a large physical address size (greater than 36 bit). Fixes a major performance hit seen on newer servers where an incorrectly programmed memory region length affects the PCI device mappings. While here, make sure to invalidate the TLB after programming MSRs and fix an incorrect behavior found by deraadt@ where MTRRdefType was updated outside of the protected region. The fix was partly obtained from FreeBSD, tested by many. With and OK deraadt
2012-07-30Revert previous change, and don't set IFM_AVALID | IFM_ACTIVE in ifm_statusMiod Vallat
to appease dhcpd, as dhcpd has now been fixed to not require this. repeated prodding and special ok deraadt@
2012-07-29whitespace cleanup, no binary changeMike Belopuhov
2012-07-28zap a misleading commentPhilip Guenthe
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-05 20:25:48 +0000