| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
ok stsp@
|
|
Since asynchronous commands can be submitted from interrupt context
it was possible to race with a process waiting for the completion of
a previously submitted command. So stop relying on the per-softc
TRB pointer for asynchronous commands and simply get the address of
the command TRB from the event TRB.
|
|
time payload. Super speed companion descriptor are still not used but
at least we can properly initialize super speed interrupt pipes.
|
|
|
|
external USB 3.0 hub support.
|
|
transfer stalled, report that a stall happen because umass(4) relies
on this behavior...
|
|
|
|
committing for jsg@, ok reyk@ tedu@ guenther@
|
|
While the index variables were correct the arrays of
dma handles they indexed were swapped for rx and tx.
As there are a mismatched number of rx and tx descriptors
we'd walk off the end of the rx handle array by 30 items.
ok deraadt@
|
|
ok deraadt@
|
|
opened with a callback.
If a driver opens an interrupt pipe without callback function, like
umct(4) does with one of its bulk in endpoints being reported as an
interrupt endpoint, then we can end up aborting a transfer which is
different from the interrupt one.
Issue reported by Roberto E. Vargas Caballero, ok deraadt@
|
|
|
|
|
|
It seems mpi introduced a problem into ehci.c rev 1.162
two weeks ago. An error check that used to return NULL
now jumps to the end of the function, but sqtd isn't
initialised at that point.
ok miod@ deraadt@
|
|
that used to always initialise the error variable is no longer run.
And at the end of bwi_encap() there is:
if (error)
m_freem(m);
return (error);
Fixing this prevents packet loss stsp was seeing.
ok stsp@ miod@ deraadt@
|
|
from the volume metadata rather than the currently defined data offset.
This allows rebuilds to work correctly when the volume metadata has a
different data offset to that currently in use (for example, volumes
created prior to softraid gaining boot support).
Found the hard way by henning@
ok deraadt@
|
|
Damn those memcpy-wrapping macros!
"do it" deraadt
|
|
ok deraadt@ sthen@
|
|
Reported by Roman Yakovlev, thanks!
"do it now" deraadt
|
|
|
|
|
|
Even if in the end we would like to be more strict about what userland
can do with kernel-managed route entries, most of the tools out there
are not yet ready for this. Since RTF_LOCAL routes are for the moment
just like RTF_LLINFO routes without expire timer, allow userland tools
to remove/modify them. In case they are missing, the good old cloning
mechanism will recreate what you need.
bluhm@ and deraadt@ agree.
|
|
iov_len to 0, we need a propper length calculation. While there,
use -1 for the file descriptor because 0 is reserved for stdin.
OK deraadt@ guenther@
|
|
specifying which media the on-board interface uses. We already query it to
set up proper register values; extend this to be able to pass a default
media to the le(4) child.
This makes SPARCbook system default to AUI without needing for a manual media
change.
tested by sebastia@
|
|
Originally with SMALL_KERNEL until sebastia@ pointed out that not
all ramdisks are built with SMALL_KERNEL.
ok deraadt@ kettenis@
|
|
print a message to that extent such that the install scripts can do their
magic.
ok miod@, deraadt@
|
|
and return the current current CPU, otherwise sched_stop_secondary_cpus()
will spin forever trying to empty its run queues. Fixes hangs during suspend
that many people reported over the last couple of days.
ok bcook@, guenther@
|
|
If you were wondering why there's a pause of 5 seconds after loading the
kernel and before seeing the copyright in blue when booting some of your
machines, well this is for you!
mlarkin@ found that this delay is caused by some code probing for xt scan
code tables when pckbd(4) initialize the console keyboard. So this diff
implements an idea from deraadt@ to only do the scan once, either during
*attach() or if we enter ddb> or ukc> before that!
ok miod@, mlarkin@, deraadt@, shadchin@
|
|
|
|
ok miod@, deraadt@, shadchin@
|
|
ok dlg@
|
|
Some of them aren't too difficult to find and fix, but others are turning
out to be deeply hidden. The timing is poor -- disable this for now.
We need to revisit this right after tree unlock.
|
|
|
|
- move a declaration before code
- sync with wd.c -r1.101 from may 2011 and pass dk_openmask to setdisklabel
|
|
|
|
|
|
- octcfsize() returns daddr_t, so calculate the return value in a daddr_t
|
|
the same functionality.
|
|
Note about workaround for 8bpp frame buffer:
Current LUNA wscons touches only first 4 planes (plane #0-#3), but
other program (e.g. mlterm-fb) can use all 8 planes on an 8bpp frame
buffer. When such program exits, it may not clear all planes, so
there may be some visible garbage data on the other 4 planes (plane
#4-#7) when we use default 256 rasops_cmap directly.
We should manage all 8 planes on LUNA, but that will be too much
overhead for 16 colors wscons. So, by repeating 16 colors in 256
colormap, we can ignore the values on the other 4 planes.
ok miod@
|
|
|
|
|
|
|
|
|
|
|
|
Problem noticed/fix tested by sebastia@ on macppc.
sparc64 problems in earlier diff pointed out by deraadt@
'looks good' kettenis@ 'commit' deraadt@
|
|
|
|
- minor code shuffling and rename a variable to reduce diff with wd.c
no binary change
|
|
- Always use either: ((x & RB_XXX) != 0) or ((x & RB_XXX) == 0) in boolean
context (mostly if (x), or x ? y : z)
- prom_halt() in alpha is confirmed to take int as boolean
Converted by coccinelle. No functional change intended.
|
