| Age | Commit message (Collapse) | Author |
|---|
|
rx descriptor.
slightly tweak by matthew
tested on alpha and amd64
|
|
right value out of the softc. then its assigned the same value again
after the rest of the var decls.
well, it used to be. not after this commit.
|
|
set. em_rxeof doesnt have to check that flag again.
|
|
entire tx path. if we try to bus_dmamap_load a very fragmented packet
m_defrag it and try again.
this is just like if_bge.c r1.355.
|
|
request and ok deraadt
|
|
print a message and ignore the interrupt. There are BIOSen out there with
random garbage in NMI entries for the non-BP CPUs.
|
|
From Jerome Glisse
16d30071c6cd4b72a950d265b0d7d08fb4220b92 in ubuntu 3.8
24f47acc78b0ab5e2201f859fe1f693ae90c7c83 in mainline linux
|
|
From Alex Deucher
6a6170926060797bdf25e5d6609a3168466f2e6c in ubuntu 3.8
d8ade3526b2aa0505132c404c05a38b73ea15490 in mainline linux
|
|
From Christian Koenig
72568eb7e4e40096fd80c7c8abd04ac630e45bb9 in ubuntu 3.8
544092596e8ac269f70e70961b5e9381909c9b1e in mainline linux
|
|
ok jsg@
|
|
in bge_encap to fail because the dmamap lacks space, not necessarily
because the ring is full. however, bge_encap failure sets the OACTIVE
flag on the interface and keeps the packet at the start of the send
queue.
the next time we try to fill the tx ring we'll try to load the same
packet and fail. an empty tx ring means bge_txeof hasnt got anything
which is where the OACTIVE condition is cleared.
this diff adds handling of fragmented packets via m_defrag.
this might fix the issues landry@ has been complaining about on his
bulk build machines. i can reproduce the above problem in contrived
circumstances here and this diff fixes it, so its going in so landry@
is forced to test it.
|
|
ok guenther
|
|
during autoconf makes no sense.
ok deraadt@
|
|
This is required by the generic loader code in newer versions of Mesa
even when we aren't building any nouveau drivers.
The way these headers are used on Linux is with a kernel version
and a libdrm version which are largely identical with the *_drm.h
headers always installed by libdrm when libdrm is built with autoconf.
In OpenBSD we have a single version of the _drm.h headers that
are shared by the kernel and userspace and have our own simple
Makefile for the libdrm build.
|
|
(we need to make them signed to spot controller overruns), and fix format
strings accordingly.
While there, make sure every runtime printf is prefixed by either the complete
target information, if available, or at least the driver name with the
proper instance number - supported systems with > 1 wdsc are quite common.
|
|
ok sf@
|
|
device detection if the port is still busy after our normal timeout.
based on a diff by chris@, based on a change in dragonflybsd
tested by Peter J Philipp
ok dlg@
|
|
width that isn't compatible with the alignment of the region:
OperationRegion (SATC, SystemMemory, 0xE008800A, 0x04)
Field (SATC, DWordAcc, NoLock, Preserve)
{
STCL, 16
}
DWordAcc means 4-byte access, but 0xE008800A is only 2-byte aligned.
Moreover, the region itself is 4 bytes long. From this I conclude
that 4-byte wide access to a 2-byte aligned address is really wanted
in cases like this.
ok deraadt@
|
|
early during attach if it signals that the device isn't present.
Reduces the noise on machines where the AML provides an (incomplete)
description of the Embedded Controller but doesn't actually have one.
ok mlarkin@, deraadt@
|
|
must use the bus_space(9) API to access it instead of memcpy(9). Also make
sure we properly align access to fields that need more than byte alignment.
ok guenther@
|
|
From Egbert Eich
9e68a1156da093385f057b9e4576652203f1248e in ubuntu 3.8
7f1950fbb989e8fc5463b307e062b4529d51c862 in mainline linux
|
|
From Alex Deucher
eed35bff1fe05f3a32989ace86107a1402f18537 in ubuntu 3.8
8c79bae6a30f606b7a4e17c994bc5f72f8fdaf11 in mainline linux
|
|
From Quentin Casasnovas
2b9149f17e7135c31c9df1c80ab6d912c4265126 in ubuntu 3.8
74073c9dd29905645feb6dee03c144657a9844cd in mainline linux
|
|
|
|
this breaks when compiling a kernel with DEBUG defined for additional
printfs elsewhere. Instead of renaming the macro, undefine DEBUG
before the new definition to minimise the diff the linux.
|
|
DP 1.2 compatible displays may report a 5.4Gbps maximum bandwidth which
the driver will treat as an invalid value and use 1.62Gbps instead. Fix
this by capping to 2.7Gbps for sinks reporting a 5.4Gbps max bw.
Also add a warning for reserved values.
From Imre Deak
d4eead50eb206b875f54f66cc0f6ec7d54122c28 in mainline linux
One of two backported commits to prevent a black screen
on a Lenovo IdeaPad Yoga 2 Pro as reported/debugged by Steven Wilson.
|
|
Use drm_dp_bw_code_to_link_rate insead. It's the same thing, but
supports DP_LINK_BW_5_4 and is also used by the other drivers.
From Paulo Zanoni
9fa5f6522e6eecb5ab20192a264a29ba4f2f4e85 in mainline linux
One of two backported commits to prevent a black screen
on a Lenovo IdeaPad Yoga 2 Pro as reported/debugged by Steven Wilson.
|
|
from working.
|
|
im sick of fixing this by hand on all my boxes while hacking on
other stuff and having it pollute my diffs.
no functional change.
|
|
without an intervening STRIOCRESCWD, release the current values instead
of the saved values, so as to effectively emulate a STRIOCRESCWD.
While here, make sure the fdc->fd_[cr]dir members are updated before
vrele()ing their old values, so that they can't be caught as dangling refs.
ok matthew@
|
|
ok gcc sha256
|
|
If a user calls STRIOCGETCWD two or more times in a row without an
intervening STRIOCRESCWD, then the references held on fst->fd_{c,r}dir
would leak.
ok miod
|
|
|
|
When using the RING_EVENT_IDX feature, we must first call publish_avail_idx()
and then read VQ_AVAIL_EVENT(vq), or there is a race condition that may cause
us to miss that the host needs to be notified. This resulted in an occasional
hang of network in vio(4).
|
|
Heads-up from Xen Li (delphij@FreeBSD), thanks!
Tested on Sony VGN-P530H.
Okay mlarkin@, matthieu@
|
|
Heads-up from Xen Li (delphij@FreeBSD), thanks!
Tested on Sony VGN-P530H.
Okay mlarkin@, matthieu@
|
|
Heads-up from Xen Li (delphij@FreeBSD), thanks!
Tested on Sony VGN-P530H.
Okay mlarkin@, matthieu@
|
|
buffersize is enforced strictly, this supplies sufficient entropy
payload to act as seed material. Discourage general use of this
API, but lock down this function name as the go-to for userland
PRNG seeding. Improve documentation.
ok miod matthew
|
|
Tested with "MAC/BBP RT5592 (rev 0x0222), RF RT5592 (MIMO 2T2R)" device.
|
|
|
|
|
|
I accepted that he's right (again) to seperate this out from heavy
sysctl API and this will simply a variety of things. Functionname
is not used by anyone in the ports tree, so we guess we can use it.
Shocking that no application has a function called this.
ok matthew & others who pushed him to start this early on
|
|
ok deraadt, kettenis
|
|
together, so to receive a full 9000 byte frame, it needs 5 descriptors.
its current mclgeti low watermark is 4.
it appears that the chip will block if it has rxed a packet into
its internal buffers waiting for descriptors. if you're at the lwm,
that means you're DoSed.
this raises the lwm so we can get at least two jumbos, ie, 10
descriptors.
ok deraadt@
|
|
Makes RT5592 devices work, except for an 11a scanning issue, according
to Brendan MacDonell who submitted this patch on tech@. Thanks!
ok claudio@
|
|
instead of magic numbers. Constants found in Linux brcmsmac.
ok claudio@
|
|
since this function is optional get rid of the no-op ones.
ok ratchov@
|
|
Apart from improving readability, this will help us reduce the number of
no-op functions now that some of them are optional.
ok ratchov@
|
|
is not part of the address. From Remco <remco at d-compu.dyndns.org>,
thanks!
ok mpi@
|
|
ok krw@ kettenis@ mpi@
|
