summaryrefslogtreecommitdiff
path: root/sys/arch/arm
AgeCommit message (Collapse)Author
2006-04-09#ifdef lint wraps for va_startTheo de Raadt
2006-04-04Add input routines to the i2s layer to support future Zaurus audio input.Christopher Pascoe
ok robert@
2006-04-04Add pxa2x0_dma_from_fifo to support DMA from internal peripheral FIFOsChristopher Pascoe
to memory. ok robert@
2006-04-04Abort any DMA that is in progress before changing the DMA completion handler,Christopher Pascoe
not after.
2006-04-04Configure the I2S controller's GPIOs, not than the USB's.Christopher Pascoe
tested robert@
2006-03-19remove unused bdbtofsb(bn) macroMartin Reindl
found by drahn@
2006-03-15use sysctl_int_lower() for the aperture variable. This lets root closeTheo de Raadt
the aperture without having to reboot, but does not allow re-opening; ok matthieu
2006-03-15Nuke dk_establish(), no longer used.Miod Vallat
2006-03-13* Add "pcitag_t *pba_bridgetag" to pci_attach_args. This is set toBrad Smith
NULL for root PCI busses. For busses behind a bridge, it points to a persistent copy of the bridge's pcitag_t. This can be very useful for machine-dependent PCI bus enumeration code. From NetBSD ok grange@ kettenis@
2006-03-12remove splimp.Brad Smith
2006-03-07Fix collision between SI_SOFTSERIAL and OS timers in the interrupt mask.Uwe Stuehler
The new allocation of softintr bits complicates SI_TO_IRQBIT, but I see no safe way around it.
2006-03-07Add the proper constant to replace a magic interrupt number.Uwe Stuehler
2006-03-07Remove COMPAT_1x stuff inherited from NetBSD, which does not apply to us.Miod Vallat
ok uwe@
2006-03-04Typos grab bag of the month, eyeballed by jmc@Miod Vallat
2006-03-04Do not deliver SIGBUS twice.Miod Vallat
2006-02-14Kill unused __HAVE_MINIMAL_EMULMiod Vallat
2006-02-12Remove dead sysbeep driver and unused sysbeep() routine. No functional change.Miod Vallat
2006-01-17Upon halt, invoke resettodr() on arm platforms, unless we have been sittingMiod Vallat
in DDB; tested at least by aanriot@
2006-01-13Make size_t, ssize_t, intptr_t and uintptr_t consistent amongst ourTodd C. Miller
architectures. They are now defined as unsigned long, long, long and unsigned long respectively.
2006-01-08Expose struct sigcontext, struct sigstat and struct sigaltstack for XPG4.2Todd C. Miller
2006-01-06Don't redefine SIZE_MAXTodd C. Miller
2006-01-06Adapt things to use __type_t instead of _BSD_TYPE_T_Todd C. Miller
Add new sys/_types.h header Include machine/_types.h or sys/_types.h where applicable
2006-01-06Merge machine/ansi.h and machine/types.h into machine/_types.h andTodd C. Miller
rename the types using the __ protected namespace (e.g. __size_t). Idea from FreeBSD.
2005-12-31Search for the devices to attach to mainbus, not hardcoded. ok miod@Dale Rahn
2005-12-31whitespace cleanup.Dale Rahn
2005-12-31Split the d_type field of cdevsw entries into d_type and d_flags for clarity.Miod Vallat
Discussed with and ok deraadt@ millert@
2005-12-27Instead of using options to configure code, use machine header filesDale Rahn
<machine/intr.h>. On the road to cleanup.
2005-12-27print real memory in MB as well as KB, also remove unused ifdef code.Dale Rahn
2005-12-22Fully implemented microtime() operation.Dale Rahn
2005-12-22typoTheo de Raadt
2005-12-22do not do delay(HUGEVALUE) after enabling/disabling the lcd dma. yes, itTheo de Raadt
has a weird effect when you enable it and the backlight at the same time, but most people don't see it because they are flipping the lid open. and it is not actually weird, it is actually rather neat. ok miod
2005-12-22actual step X-# instead of rev # for pxa270Theo de Raadt
2005-12-22'spoofonly' does not mean what I thought it meant. The whole point isKenneth R Westerback
to read non-OpenBSD label information like MBRs and populate ('spoof') the OpenBSD disk label appropriately. So avoiding an I/O when doing 'spoofonly' is to miss the whole point. Backout all Sept. 26 and subsequent changes to disksubr.c files. The original CD problem with I/O errors on blank media will be addressed another way. Should fix the ext3 vs amd64 problems discussed on misc@. Thanks to steven@ for finally putting it terms that I understood. ok deraadt@
2005-12-22remove another debug messageTheo de Raadt
2005-12-22silence setperf codeTheo de Raadt
2005-12-21sysctl machdep.lidsuspend=1 makes screen closures to cause a suspend.Theo de Raadt
(opening the screen does not yet wake the machine up, not yet sure if that is even possible yet)
2005-12-20volatile variable for looping; ok drahnTheo de Raadt
2005-12-20Add machdep.maxspeed sysctl to change the meaning of hw.setperf=100 on zaurus,Dale Rahn
This will scale the speed for hw.setperf, this allows limiting apm -A to lower speed, or even overclocking. ok deraadt@
2005-12-20switch to using clock4 instead of clock0 so that we get clocks when runningDale Rahn
at 91MHz (clock4 is programmed to be based of the 32.768KHz clock.
2005-12-18Don't include cdefs.h if _LOCORE is defined and hide some more definesTodd C. Miller
unless __BSD_VISIBLE or _LOCORE are set. OK deraadt@
2005-12-17Get rid of deprecated vm_{offset,size}_t types for good, use {p,v}{addr,size}_tMiod Vallat
instead; looked at millert@
2005-12-16Return EINVAL in process_sstep() as this is not implemented on arm yet;Miod Vallat
ok kettenis@
2005-12-14convert _FOO_SOURCE -> __FOO_VISIBLE in machine. OK deraadt@Todd C. Miller
2005-12-14Make clock_t consistent across platforms as a 32 bit int. OK deraadt@Todd C. Miller
2005-12-14Don't hard-code card present and number of socket bits in the PCMCIA busUwe Stuehler
configuration register, but control these bits from PCMCIA code. ok drahn@
2005-12-13It would seem that some cards do not get reset correctly if RESET isUwe Stuehler
held only for 10us but the same cards work just fine on other platforms. A really long delay lets those cards reset correctly on zaurus, but it is only an ugly workaround until the socket setup is fixed. ok for now fgs@, drahn@
2005-12-13First step in include files overhaul. Use __FOO_VISIBLE (as definedTodd C. Miller
in sys/cdefs.h) instead of _FOO_SOURCE. Also fix several namespace pollution issues, including the byte order defines. OK deraadt@
2005-12-13Remove the advertising clause in the UCB license which BerkeleyJonathan Gray
rescinded 22 July 1999. Checked by ian@, deraadt@ and millert@, arm portion checked by drahn@
2005-12-05- Run only the necessary code to transition from one speed to another.Uwe Stuehler
- Don't clobber perflevel, instead leave it to the user requested value. precise description and ok drahn@ :)
2005-12-02add registers for 4 new clocksTheo de Raadt