summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2006-08-30Horrible code to work around stupid VXT2000 rom getchar() behaviour, butMiod Vallat
at least this allows our boot blocks to work with glass console on these machines now. Caution: reading this code will hurt your eyes. ok deraadt@
2006-08-30Enable VXT2000 bits now that they are worth using; ok deraadt@Miod Vallat
2006-08-30Teach the SGEC driver about the Real World:Miod Vallat
- check for error bits in descriptors, and count and report errors accordingly; make sure we do not pass up known-to-be-bad packets up the receive path. - check for various error conditions which would cause the transmit process to go to suspended state, and recover from them, before the timeout expires. - finally, on the VXT2000, the transmit process for some reason will reset its ring position to the beginning of the ring every time it completes a transmission burst, while we don't, so play with the head of ring register behind its back. This makes my VXT2000+ (net)boot multiuser. Tested on VXT2000 (ze@vxtbus), KA49 (ze@vsbus) and KA53 (ze@ibus); ok deraadt@
2006-08-30Better diagnostic messages.Miod Vallat
2006-08-30allow DIOCNATLOOK to look up NAT states for protocols without portDamien Miller
numbers, reported by Raja Subramanian; ok henning@
2006-08-30Return from channel tuning code if we get IEEE80211_CHAN_ANYJonathan Gray
as per other drivers instead of doing panics. Should workaround PR 5206 which I could never manage to reproduce, though we really shouldn't be getting IEEE80211_CHAN_ANY here. ok deraadt@
2006-08-30when determining the size of the volumes and disks we want how manyDavid Gwynne
bytes each of them has, not the number of kilobytes. this solves the "why am i dividing by 2" question. the answer is "you dont, you multiply by 512".
2006-08-29enable arc, ok deraadt@Jolan Luff
2006-08-29#undef I2C_VERBOSEMark Kettenis
ok deraadt@
2006-08-29Fix handling of beacon frames by stations while they're associated.Damien Bergamini
Changes in the ERP information element and in the capinfo field were not detected properly, so the drivers were not able to adjust their settings (like short slot time, short preamble, protection mode etc...) dynamically. Should fix BSS mode with both 802.11b and 802.11g stations. ok reyk@
2006-08-29make ic_stats.is_rx_tooshort statistic more accurate.Damien Bergamini
cosmetic tweaks. ok reyk@
2006-08-29Change the way ieee80211_fix_rate() handles the IEEE80211_F_DOFRATE flag.Damien Bergamini
Instead of removing the rates from the node's rate set, just check that the fixed rate is present in the rate set. Otherwise, the node will be marked non-ERP which will force the use of protection mode (either CTS-to-self or RTS/CTS) in an 802.11g network and degrade performance. This fixes HostAP mode for 11g with a fixed rate. Problem originally reported by Sam Fourman Jr. Code from FreeBSD. ok reyk@, "no objections here" jsg@
2006-08-29Add support for onboard bge(4)'s on Sun UltraSPARC hardware that apparentlyMark Kettenis
come without an attached SEEPROM. ok deraadt@
2006-08-29Don't panic when watchdog timeout is called.Marcus Glocker
formal ok deraadt@
2006-08-29allow the carp demotion counter to be changed by arbitary values as longHenning Brauer
as the resulting demotion counter value is in range. previously, we only allowed +/- 1. ok mpf mcbride deraadt
2006-08-29Fix a possible kernel panic in ucom(4) that is caused by ttyclose() using dataMarc Balmer
already freed by ucom_close(). Problem reported and initial patch provided by Marc Winiger <marc@winiger.ch> ok miod, deraadt
2006-08-29zyd is not making this release; ok jsgTheo de Raadt
2006-08-29XXX when we find vgapci devices on a USIII machine, fail to matchTheo de Raadt
(for now, until the bugs are found. USIII machines remain serial-only for now); ok miod
2006-08-28move to official 4.0Theo de Raadt
2006-08-28Allow building procfs on systems that don't #define PT_STEP (such as sparc64).Marc Aurele La France
ok pedro@, miod@
2006-08-28Completely rework the allocation/management of Rx buffers.Damien Bergamini
The firmware expects 3KB Rx buffers so we can't just map mbuf clusters which are only 2KB. Also it seems that Rx buffers must be aligned on a 1KB boundary which is not necessary the case with mbuf clusters. We want to make damn sure the firmware will not DMA anything outside the mapped Rx buffers. The size of the Rx buffers is probably configurable but Intel will refuse to answer to any question regarding the programmation of the chipset, and a dump of the device registers led to nowhere.
2006-08-28Make carp see all IPv6 address additions and remove dead code in carp_ioctl()Ryan Thomas McBride
ok mpf@ henning@
2006-08-28Another grammar nit; "please go ahead" jmc@.Tom Cosgrove
2006-08-28Port bus_dmamap_load(9) and bus_dmamap_sync(9).Marcus Glocker
2006-08-28typos; from tbertJason McIntyre
(one not taken)
2006-08-28avoid avail_end common; no binary change; found by grunk@Michael Shalayeff
2006-08-28enable arcDavid Gwynne
2006-08-28Check for illegal option lengths when parsing LCP packetsCan Erkin Acar
and drop such bad packets. Also remove some redundant mallocs. This fixes possible heap overflows when forming replys to such bad packets as discovered by Martin Husemann and Pavel Cahyna. reported by NetBSD, initial diff from markus@, additional comments by claudio@, ok markus@
2006-08-28Move DMA mapping code from if_pgt_pci.c to pgt.c and port it as is for now.Marcus Glocker
Remove dead code chunk from if_pgt_pci.c. For those who want to refer to the original code you can find it in ~mglocker/pgt/orig.
2006-08-28Get MAC address from the OpenFirmware local-mac-address property on sparc64.Mark Kettenis
Needed to support onboard bge(4)'s on Sun machines which don't seem to be fitted with the EEPROM that normally contains the MAC address. ok brad@
2006-08-28move the firmware messaging under NBIO since it's only used by bio.David Gwynne
2006-08-28async events defines and structsMichael Shalayeff
2006-08-28revert flags usage commits. too close to release. will go in post release ↵Brad Smith
instead. requested by deraadt@
2006-08-28make a define for dealing w/ the big bit (0x80); no func changeMichael Shalayeff
2006-08-28properly detect failed drives and do not crash bioctl; krw@ testingMichael Shalayeff
2006-08-28Fix manual media types by always leaving the PHY in autoneg mode but adjustingBrad Smith
the ANAR as appropriate. Derived from a diff for FreeBSD; same technique used by the Linux r8169 driver.
2006-08-28bzero() the sensor memory after allocating it.Kenneth R Westerback
ok marco@
2006-08-28set error and break -- not return an error before spl is set back in ioctl! ↵Michael Shalayeff
found by krw@
2006-08-28add a BGE_JUMBO flag.Brad Smith
2006-08-28turns out the volume info contains a bitmap of failed physdisks in it, soDavid Gwynne
i can just check if this is non zero to see if a volume is degraded. this means i dont have to pull the raid info page to find failed disks in the volume inquiry.
2006-08-28Dell PCI vendor subid.Brad Smith
2006-08-28Set the BGE_NO3LED flag to properly adjust the PHY LED mode if attachingBrad Smith
to either a BCM5700 or BCM5701 chip and have the Dell PCI vendor subid. The appropriate code has been in brgphy for 2 years, now the bge(4) driver will use it. Also found in the Broadcom bcm5700 and Linux tg3 drivers.
2006-08-27display proper volume device name in the volume line (; krw@ testingMichael Shalayeff
2006-08-27it works a little better like this.Brad Smith
2006-08-27Enable HORRID_III_HACK, but make sure we only disable the data cache onMark Kettenis
UltraSPARC III and up. That way it isn't really that horrid and doesn't really affect UltraSPARC I & II. This gives us a GENERIC that runs on UltraSPARC III. ok deraadt@
2006-08-27pack several boolean fields into the existing bge_flags field.Brad Smith
2006-08-27basic monitoring and management support. there are a few more things to fix ↵Michael Shalayeff
but fetching array status and rebuild/blinking works on compaq/hp boxen; testing and some fixes by krw@
2006-08-27Use the correct IGN for the B side of schizo. Makes interrupts actually workMark Kettenis
in the PCI slot of the v210. ok jason@
2006-08-27Work-in-progress VXT2000 support (commented out in GENERIC so far). GlassMiod Vallat
console works on color model, serial console works, ethernet attaches but does not work correctly yet.
2006-08-27Extract the device-independent parts of dzms, to let lkms@non-dz exist.Miod Vallat
Don't forget to rerun config(8)!