summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2004-07-06sent in by loki@animata.net: This includes all the changes from netbsdTheo de Raadt
related to working with the data toggle, but also includes some endian fixes. The major changes in the netbsd code were in revisions 1.55 and 1.64. their log messages are: 1.55: Set the data toggle correctly, and use EHCI_QTD_DTC. This fixes problems with my ALi-based drive enclosure (it works now, rather than failing to attach). Also seems to work with a GL811-based enclosure and an ASUS enclosure with a CD-RW, on both Intel and NEC controllers. Note: The ALi enclosure is currently very SLOW, due to some issue with taking too long to notice that the QTD is complete. This requires more investigation. 1.64: Further cleanup of toggle handling. Now that we use EHCI_QH_DTC, we don't need to fiddle with the TOGGLE bit in the overlay descriptor, so minimize how much we fuss with it.
2004-07-06from netbsd, via loki@animata.net, ehci.c revision 1.58 without someTheo de Raadt
printfs that are more related to the data toggle. log message: Fix a stupid bug in ehci_check_intr() that caused use to try to complete a transaction that was still running. Now ehci can handle multiple devices being active at once.
2004-07-06from netbsd, via loki@animata.net, ehci.c revision 1.56 sans some debuggingTheo de Raadt
printfs that crept in. log message: Failure to properly mask off UE_DIR_IN from the endpoint address was causing OHCI_ED_FORMAT_ISO and EHCI_QH_HRECL to get set spuriously, causing rather interesting lossage.
2004-07-06Replace magic number with a more descriptive #define. This magic number wasAaron Campbell
particularly vile because it looks like a typo (i.e., looks like one "f" is missing), but it's really not. I happened to notice this myself, but the fix is inspired by NetBSD in the interest of keeping the code similar. deraadt@ ok
2004-07-05Allocate buffers in randomread and randomwrite via malloc insteadTodd C. Miller
of from the stack. Keeps us from running out of kernel stack during stack-heavy networking load. The bug manifests as a panic during bootup with an fpu-related assertion failure on i386. Fix from Christopher Pascoe; commit requested by deraadt@
2004-07-05remove some stuff to make the blocks slightly smaller. we could go furtherTheo de Raadt
still. tested by matthieu
2004-07-05clean & knfTheo de Raadt
2004-07-05do not queue events when no daemon running. flush all events on last close.Theo de Raadt
ok tedu
2004-07-05sync various changes in old_identifycpu() into identifycpu()David Krause
first step toward merging these two functions ok deraadt@
2004-07-05Completes the structs for the isochronous and split transactionTheo de Raadt
isochronous transfer descriptor types. Also updates the doco link at the top of the file; from netbsd; via loki@animata.net
2004-07-05Adds missing port routing rules field, fixes an incorrect offset forTheo de Raadt
EHCI_QH_GET_CTL and adds a macro for setting the data toggle on a qtd token; from netbsd; via loki@animata.net
2004-07-05This fixes a couple of spelling errors (EHCI_HCS_P_INCICATOR toTheo de Raadt
EHCI_HCS_P_INDICATOR and EHCI_QG_MPLMASK to EHCI_QH_MPLMASK); from netbsd; via, loki@animata.net
2004-07-05KNFHenning Brauer
2004-07-04syncTheo de Raadt
2004-07-04shrink, shrink, shrinkTheo de Raadt
2004-07-04- two typos, stallen -> stolenBrad Smith
- cast base correctly for format string From NetBSD ok deraadt@
2004-07-04sis rev 0x91 seeprom support to get eaddr; from fgschTheo de Raadt
we supplied 10 soekris boxes to 10 openbsd developers, and these all have some variety or other of sis ethernet. not ONE of them commented about this diff when it was mailed out. how very sad.
2004-07-04remove the half-baked and bogus pfi_dynamic_drivers() which tries to guessHenning Brauer
which drivers are hotpluggable. since we removed the stupid check from pfctl a few days ago nothing relies on this any more. ok pb@ mcbride@
2004-07-04fix a NetBSDism and a pasto left over from the import; ok millert@Christian Weisgerber
2004-07-04re fits on this floppyTheo de Raadt
2004-07-04proc_stop needs sched_lock locked. From art, verified with NetBSD.Niklas Hallqvist
2004-07-04syncTheo de Raadt
2004-07-04two hp netserver LH 3 devicesTheo de Raadt
2004-07-04turn off _KERN_DO_ECOFF unless COMPAT_OSF1 is enabled. This shrinks theTheo de Raadt
ramdisks, too.
2004-07-03Check return value instead of uninitialized variable.Mark Kettenis
ok deraadt@
2004-07-03sigh, not yetTheo de Raadt
2004-07-03Cruft removal; ed(4) has been dead for ages. OK deraadt@ miod@Todd C. Miller
2004-07-03oops something went wrong in commitTheo de Raadt
2004-07-03{e,}intr{names,cnt} are extinctTheo de Raadt
2004-07-03if vinvalbuf() fails, unlock the vnode and release it. ok marius@Pedro Martelletto
2004-07-03Oops, one more mmap() bounds check.Miod Vallat
2004-07-03On 1[67]7 boards, always use the VMEChip2 timers for delay(), even if theMiod Vallat
device is not attached yet.
2004-07-03quick workaround until proper PF_FORWARD reass gets implemented.Jun-ichiro itojun Hagino
2004-07-02move into #ifdef; various peopleTheo de Raadt
2004-07-02During bootup, only print the VMEChip2 decoder settings for those which areMiod Vallat
enabled.
2004-07-02Remove unused code, per mvme88k.Miod Vallat
2004-07-02Stricter bounds checking for mmap() calls.Miod Vallat
2004-07-02Kill ca_master in autoconf structures. Drivers either only attach to oneMiod Vallat
specific bus (mc/pcc/pcctwo), or they know it from the ca_bustype value, and can use the globals sys_busname variables. This also allows for some code simplifications.
2004-07-02Merge *ramread() and *ramwrite(), as done on mvme88k.Miod Vallat
2004-07-02Remove unused getdevunit().Miod Vallat
2004-07-02Cope with SONPROC scheduler changes.Miod Vallat
2004-07-02Maintain %f and %gs over traps. Mostly from NetBSD. Preparation for SMPNiklas Hallqvist
speedups.
2004-07-02Make page-aligned variable contiguous. Saves almost one page of data.Miod Vallat
2004-07-02Switch mvme88k to evcount, but keep maintaining intrcnt[] too for now.Miod Vallat
2004-07-02Disable a -O2 specific optimization which causes kern_fork (and probably otherMiod Vallat
files) to be miscompiled, until the problem is fixed in the compiler.
2004-07-02printf inode numer as unsigned -- just the way it isMichael Shalayeff
2004-07-02Be sure to fill intrhand structures correctly.Miod Vallat
2004-07-02Remove unused extra intrhand.Miod Vallat
2004-07-02Since these vector tables are not final tables, they don't need to beMiod Vallat
page aligned.
2004-07-02fill out error info into ret* regs upon failed syscall returnMichael Shalayeff
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-30 16:23:19 +0000