| Age | Commit message (Collapse) | Author |
|---|
|
transfers. This fixes some cases where the software toggle tracking
was not doing the right thing. For example, a short transfer that
transferred 0 bytes of the requested qTD transfer size does cause
a toggle change, but the existing code was assuming it didn't.
Derived from work in FreeBSD.
Fixes at least one USB2 cdce device, which would otherwise drop
every second packet due to incorrect toggle state.
commit it, dlg@
|
|
UltraSPARC III. Hack around it by temporarily switching to nucleus context
when flushing TLBs. Put this hack under HORRID_III_HACK for now.
Compiling with HORRID_III_HACK results in a kernel that runs on the Blade
2000 "at the speed of a 140MHz UltraSPARC I".
ok jason@
|
|
in ether_input/ether_output.
- only count input packets when sure the packet can be received as opposed
to counting it unconditionally, then potentially running into an error
and then dropping the packet.
ok reyk@
|
|
is closed and then reopened. This may be necessary now that
we no longer clear endpoint stalls every time a pipe is opened.
Previously we could assume an initial toggle value of zero because
a clear-stall operation resets the device's toggle state.
Derived from work in FreeBSD.
This is most likely to affect devices like printers, which open
pipes for short periods of time and close them again.
ok dlg@
|
|
i got the page fetching right, its not really useful for anything in the
real world.
|
|
ok brad@ jason@
|
|
|
|
again, pointed out by miod@
|
|
|
|
Detect device type using some simple checks and use table driven
interrupt maps for known devices.
IO space appears to not want to be at 0, but at some bus offset (0x90000000)
so skew mappings into this io bus base.
Thecus N2100 and I/O Data Landisk (HDL-G) are now partially working.
|
|
|
|
|
|
to them by default, so disable mpt to get mpi to attach.
requested by deraadt@
|
|
|
|
length is smaller. Tested by joris@, ok gwk@
|
|
on sparc64 last night. macppc gets it in RAMDISK as well.
|
|
and fc controllers), so mpi will be slower than mpt for io on those hbas.
for now.
ifdef the scsi stuff out till we make it go fast, and so we can enable mpi
everywhere.
|
|
on a port. since 256 wont fit into 8 bits they say 0 means 256. this diff
does the appropriate interpretation. it also avoids a divide by zero when
we figure the openings out by dividing the number of commands the
controller can support by the number of devices it supports.
panic found while testing the fc controller at home.
|
|
and adapted from FreeBSD. This adds the ability for status requests to ask
for the volume tags (usually bar codes, on changers with a bar code reader)
on media inthe library.
ok deraadt@, krw@
|
|
From NetBSD
|
|
This fixes a few cases where the hardcoded constant indicated a reserved
operation instead of the intended secundary context flush.
ok jason@
|
|
|
|
|
|
copying the Ethernet header to the new mbuf. The new mbuf is then prepended
into the existing mbuf chain.
From FreeBSD
ok reyk@ pascoe@ jason@
|
|
|
|
ok brad@
|
|
|
|
more or less the same code. This means that all our watchdog drivers
use the framework, and that they all can be configured using sysctl
kern.watchdog.{period,auto}, and they can all use watchdogd.
At least for now we still support the /dev/wdt0 char device and
userland poking using ioctl.
Testing by < jg ! rilk , com >, many thanks!
ok deraadt
|
|
-instead of staticaly declaring the heads, allocate memory for them
-add second dimension to the routing table head array, table ID
-space saver: do not use the AF as array index directly, but have an
AF->index translation array and only account for AFs actually attaching
a routing table
note that whil ethat allows for multiple routing table heads, nothing uses
taht yet.
tested by many, ok claudio reyk norby mcbride
|
|
|
|
|
|
|
|
to make this code cleaner, but I'm bloody tired of typing the root
device name
|
|
is most likely not coming back.
|
|
OK henning@ norby@
|
|
BIOS configured it in; from NetBSD via Andrew Daugherity.
|
|
that it can be attached properly
|
|
ok miod
|
|
ok miod
|
|
operation. Will allow a vax implementation of splassert().
|
|
|
|
|
|
the mac mini's ATI graphics card. The full 0xa0000-0xfffff range is
used here.
|
|
as 3DLabs Permedia and Intergraph Expert3D.
Also, if no mmio region is found, attach anyway, this only prevents accelerated
X11 from starting.
|
|
|
|
(/pci@8,600000/, 8 == upaid, 600000 == paddr of registers)
Also, match the mpi (LSILogic,sas)
|
|
we will randomly choose that stupid port, which NIC's are sometimes
programmed to eat invisibly; sthen@bootes.spacehopper.org, pr5139
|
|
USBD_FORCE_SHORT_XFER to ensure that we actually build and execute
a transfer.
Based on changes in FreeBSD rev1.47
|
|
|
|
msf@'s assistance with hexdumping, this bug was found.
do NOT access the wrong union member (sockaddr_in) when using v6 addresses
this fixes output errors when displaying v6 encap routes for ipsecctl, route,
and recently netstat
ok hshoexer@ markus@ claudio@ naddy@
|
