Commit Graph

27495 Commits

Author SHA1 Message Date
Luigi Rizzo
0dc809c034 move netmap_getna() to a freebsd-specific file 2014-06-06 16:23:08 +00:00
Luigi Rizzo
89cc25561c align comments with the ones in our development trunk 2014-06-06 14:58:25 +00:00
Luigi Rizzo
d8e1c53b15 rate limit some error messages 2014-06-06 14:57:40 +00:00
Luigi Rizzo
5899a007ae remove two debugging messages, align comments with the code
in our development trunk
2014-06-06 14:57:16 +00:00
Kevin Lo
ec74a03d01 Replace deprecated M_DONTWAIT with M_NOWAIT.
Reviewed by:	jfv
2014-06-06 11:12:19 +00:00
Luigi Rizzo
e31c6ec7e2 add checks for invalid buffer pointers and lengths 2014-06-06 10:50:14 +00:00
Luigi Rizzo
441ab64f52 prevent a panic when the netdev/ifp is not set in attach
(internal  c63a7b85)

MFC after:	3 days
2014-06-06 10:40:20 +00:00
Andrey Zonov
dc8a95e62b Use mtx_lock_spin/mtx_unlock_spin primitives on spin lock
Reviewed by:	luigi
MFC after:	1 week
2014-06-06 00:24:04 +00:00
Luigi Rizzo
43ed1d3c76 whitespace change: remove trailing whitespace 2014-06-05 21:12:41 +00:00
Hans Petter Selasky
408b9d7cfb - Verify that we don't load more data into a FIFO than it is
configured to handle.
- Add code to handle suspend and resume.

MFC after:	3 days
2014-06-05 18:23:51 +00:00
Hans Petter Selasky
1b1df1a91f Need to check the transfer cache field in the device done function
to be sure all allocated channels are freed and not the transfer first
one.
2014-06-05 18:19:48 +00:00
Hans Petter Selasky
08aa4c94b8 Try to fix DWC OTG regression issues with full and low speed devices:
- Remove double buffering interrupt and isochronous traffic via the
transaction translator. It can be avoided because the DWC OTG will
always delay the start split transactions for interrupt and
isochronous traffic, but will not delay the complete split
transactions, if we set the odd frame bit correctly.
- Need to check the transfer cache field in the device done function
to be sure all allocated channels are freed and not the transfer first
one. This seems to resolve the control endpoint transfer type quirk
which is now removed.
- Make sure any received data upon TX is dumped else RX path will
stop.
- Transmit isochronous data before receiving isochronous data as a
means to optimise the TT schedule.
- Implement a simple TT bandwidth scheduler.
- Cleanup use of old "td->error" variable.
- On interrupt IN traffic via the transaction translator we simply
ignore missed transfer opportunities and silently retry the
transaction upon next available time slot.

MFC after:	3 days
2014-06-05 18:17:40 +00:00
Ed Maste
29743a4117 Update vt(4) "Terminus BSD Console" font
"Terminus BSD Console" is a derivative of Terminus that is provided
by Mr. Dimitar Zhekov under the 2-clause BSD license for use by the             FreeBSD vt(4) console.

Reviewed by:	jhb
MFC after:	1 week
Sponsored by:	The FreeBSD Foundation
2014-06-05 13:49:37 +00:00
Kevin Lo
a080f71521 Remove D-Link DWA140 rev D1 by accident.
Pointed out by Anders Larsson in private email.
2014-06-05 01:55:23 +00:00
Navdeep Parhar
861e42b209 cxgbe(4): Properly account for the freelist buffers used when returning
early from service_iq due to a budget restriction.  This fixes a potential
rx hang when using INTx.

MFC after:	3 days
2014-06-05 00:38:32 +00:00
Ed Maste
733301bf97 Remove extra copy of old email address.
MFC after:	3 days
2014-06-04 20:05:52 +00:00
Ed Maste
6f7ba220bc Update vt(4) console font author's email address
MFC after:	3 days
2014-06-04 19:42:47 +00:00
Hans Petter Selasky
bb03cd6fa1 More RSU fixes:
- Properly align temporary buffer to 32-bit.
- Add an extra parenthesis to make expression clear.
- Range check the association ID received from hardware.

MFC after:	1 week
2014-06-04 16:58:35 +00:00
Luigi Rizzo
c26e5fc2ed make sure ifp->if_transmit returns 0 if a buffer is enqueued.
A similar fix should be applied to vmxnet, ixgbe, igb, i40e.
(some of them previously reported by Michael Tuexen)

Drivers using if_transmit are correct, and so are most of the
other drivers that reassing if_transmit.

Among other things, this bug causes panics when using netmap emulation
on top of generic drivers.

Approved by:	bryanv
MFC after:	3 days
2014-06-04 16:57:05 +00:00
Konstantin Belousov
49fe48ab0c For Xeon 7500 and 48XX (Nehalem EX and Westmere EX) variants of the
Core i7 and Westmere processors, the uncore PMC subsystem is
completely different from the uncore PMC on smaller versions of CPUs.
Disable existing uncore hwpmc code for EX, otherwise non-existing MSRs
are accessed.

The cores PMCs seems to be identical for non-EX and EX, according to
the SDM.

Reviewed by:	davide, fabient
Sponsored by:	The FreeBSD Foundation
MFC after:	2 weeks
2014-06-04 16:06:38 +00:00
Hans Petter Selasky
a31f80290e The external USB HUB in the RPI-B rejects control endpoint traffic
using the BULK endpoint type, while other USB HUBs do not. Disable
endpoint type workaround for TT traffic.

MFC after:	3 days
2014-06-04 10:29:01 +00:00
Hans Petter Selasky
910593b55c Fixes for the RSU driver:
- The R92S_TCR register is an 8-bit register. Don't access it like a
16-bit register.
- Disable parsing the delete station event, due to many false events.
- Ensure that there is only one transfer queue for each endpoint, so
that packets transmitted don't get out of order.

MFC after:	1 week
2014-06-04 09:18:13 +00:00
Xin LI
f0dc762a56 Don't leak Giant.
Noticed by:	Sascha Wildner (DragonFlyBSD) via IRC
Reviewed by:	mav
MFC after:	3 days
2014-06-03 21:31:28 +00:00
Aleksandr Rybalko
bfc00339de Fix case when vt(4) started w/o driver assigned.
o Always init locks and cv ASAP.
o Initialize driver-independent parts even if driver probing fail.
o Allow to call vt_upgrade anytime, for later loaded drivers.
o New window flag VWF_READY, to track if window already initialized.
Other updates:
o Pass vd as a cookie for kbd_allocate.
o Do not blank window on driver replacement.

Tested by:	hselasky (RPi), emaste(VGA, EFIFB, KMS), me

MFC after:	7 days
Sponsored by:	The FreeBSD Foundation
2014-06-03 13:33:43 +00:00
George V. Neville-Neil
5c37e5dacf Add missing Ivy Bridge and Haswell events.
Submitted by: Anton Rang <rang@mac.com>
MFC: 2 weeks
2014-06-02 20:50:08 +00:00
Marcel Moolenaar
717568ad8d Convert bxe(4) to use the driver API.
Submitted by:   Sreekanth Rupavatharam <rupavath@juniper.net>
Obtained from:  Juniper Networks, Inc.
2014-06-02 18:54:45 +00:00
Marcel Moolenaar
9e11529015 Convert em(4) to use the driver API.
Submitted by:   Anuranjan Shukla <anshukla@juniper.net>
Obtained from:  Juniper Networks, Inc.
2014-06-02 18:52:03 +00:00
Marcel Moolenaar
41eb5ac39d Convert fxp(4) to use the driver API.
Submitted by:	Anuranjan Shukla <anshukla@juniper.net>
Obtained from:	Juniper Networks, Inc.
2014-06-02 18:45:36 +00:00
Marcel Moolenaar
62d76917b8 Introduce a procedural interface to the ifnet structure. The new
interface allows the ifnet structure to be defined as an opaque
type in NIC drivers.  This then allows the ifnet structure to be
changed without a need to change or recompile NIC drivers.

Put differently, NIC drivers can be written and compiled once and
be used with different network stack implementations, provided of
course that those network stack implementations have an API and
ABI compatible interface.

This commit introduces the 'if_t' type to replace 'struct ifnet *'
as the type of a network interface. The 'if_t' type is defined as
'void *' to enable the compiler to perform type conversion to
'struct ifnet *' and vice versa where needed and without warnings.
The functions that implement the API are the only functions that
need to have an explicit cast.

The MII code has been converted to use the driver API to avoid
unnecessary code churn. Code churn comes from having to work with
both converted and unconverted drivers in correlation with having
callback functions that take an interface. By converting the MII
code first, the callback functions can be defined so that the
compiler will perform the typecasts automatically.

As soon as all drivers have been converted, the if_t type can be
redefined as needed and the API functions can be fix to not need
an explicit cast.

The immediate benefactors of this change are:
1.  Juniper Networks - The network stack implementation in Junos
    is entirely different from FreeBSD's one and this change
    allows Juniper to build "stock" NIC drivers that can be used
    in combination with both the FreeBSD and Junos stacks.
2.  FreeBSD - This change opens the door towards changing ifnet
    and implementing new features and optimizations in the network
    stack without it requiring a change in the many NIC drivers
    FreeBSD has.

Submitted by:	Anuranjan Shukla <anshukla@juniper.net>
Reviewed by:	glebius@
Obtained from:	Juniper Networks, Inc.
2014-06-02 17:54:39 +00:00
Hans Petter Selasky
b2fa7d8104 Change type of the DMA address so that on PAE platforms we get can
64-bit DMA addresses even though the USB stack currently only uses
32-bit DMA.

Suggested by:	Kohji Okuno <okuno.kohji@jp.panasonic.com>
2014-06-02 07:08:34 +00:00
Bryan Venteicher
9a73216696 Split the virtio.h header file into multiple files
Reorganize the previous contexts of the file as it is in Linux. The
eventual goal is to install the header files and share them between
the kernel and bhyve.

MFC after:	1 week
2014-06-01 18:16:01 +00:00
Hans Petter Selasky
dd5a01badc Add basic support for isochronous transfers in host mode to the
ISP/SAF1761 driver.

Sponsored by:	DARPA, AFRL
2014-06-01 10:22:18 +00:00
Luiz Otavio O Souza
34aa2211e7 Ignore IIC_ENOADDR from iicbus_reset() as it only means we have a
master-only controller.

This fixes the iic bus scan with i2c(8) (on supported controllers).

Tested with gpioiic(4).
2014-05-31 14:38:05 +00:00
Luiz Otavio O Souza
e72f32cd0f Add a bounds verification to the SCL and SDA pin values.
At attach, print the SCL and SDA pin numbers.

Remove a stray blank line.

Remove the GPIOBUS locking from gpioiic_reset(), it is already called with
this lock held.  This fixes a crash when you try to scan the iicbus with
i2c(8).
2014-05-31 14:27:50 +00:00
Navdeep Parhar
368541ba1e cxgbe(4): Fix a NULL dereference when the very first call to
get_scatter_segment() in get_fl_payload() fails.  While here,
fix the code to adjust fl_bufs_used when a failure occurs for
any other scatter segment.

MFC after:	3 days
2014-05-30 22:59:45 +00:00
Hans Petter Selasky
4be86ee126 It appears the ISP/SAF1761 sometimes is busy when reading the status
word of the PTD and simply returns a value of zero. Retry a few times
before giving up.

Sponsored by:	DARPA, AFRL
2014-05-30 16:44:03 +00:00
John Baldwin
f1f7034d11 Use callout_*() rather than timeout()/untimeout().
Reviewed by:	imp
2014-05-30 15:46:10 +00:00
Hans Petter Selasky
ce19294c0e Make driver detach code in USB mass storage test optional.
Sponsored by:	DARPA, AFRL
2014-05-30 13:45:20 +00:00
Hans Petter Selasky
74c1f36628 Add ISP/SAF1761 bootloader bus interface file.
Sponsored by:	DARPA, AFRL
2014-05-30 13:42:11 +00:00
Hans Petter Selasky
e045de0740 Add support for basic read, write and read capacity disk operations to
the USB mass storage test utility file.

Sponsored by:	DARPA, AFRL
2014-05-30 07:30:24 +00:00
Ed Maste
c6317b8376 Correct vt(4) border calculations on font switch
If a vt(4) font does not exactly fit the screen dimensions, the console
window is offset so that it is centered.  A rectangle is drawn at the
top, left, right, and bottom of the screen, to erase any leftovers that
are outside of the new usable console area.

If the x offset or y offset is 0 then the left border or top border
respectively is not drawn.  The right and bottom borders may be one
pixel larger than necessary due to rounding, and are always drawn.

Prior to this change a 0 offset would result in a panic when calling
vt_drawrect with an x or y coordinate of -1.

Sponsored by:	The FreeBSD Foundation
2014-05-29 21:52:42 +00:00
Ed Maste
7cd5ec40cf Update size in vt_grow if we don't have to reallocate
vt_grow may be called with a new size that's larger than previous but
does not require reallocation - for example, when the number of columns
is the same and new number of rows is less than the history size.
Prior to this change we would fail to update vb_scr_size, and then hit
a KASSERT when trying to write to the newly visible rows.

Sponsored by:	The FreeBSD Foundation
2014-05-29 21:10:33 +00:00
Olivier Houchard
8bc9a079ef In uart_bus_grab(), use the ier_mask instead of a custom hack for XScale.
Suggested by:	jmg
2014-05-29 21:00:06 +00:00
Ed Maste
ba23b43576 Correct KASSERT strings
Rows refer to height and cols to width.
2014-05-29 20:26:01 +00:00
Olivier Houchard
caf6d6b491 In the grab function, keep the bit 6 on in the IER, on XScale, using 0
turns the UART off, which is unfortunate if one want to use it as a console.
2014-05-29 19:57:51 +00:00
Konstantin Belousov
c33ce50343 Add support for the unmapped i/o to mfi(4).
Tested by:	Nicholas Esborn <nick@desert.net>
Sponsored by:	The FreeBSD Foundation
MFC after:	1 week
2014-05-29 16:20:34 +00:00
Aleksandr Rybalko
bf03576620 Remove driver as unused.
MFC after:	7 days
Sponsored by:	The FreeBSD Foundation
2014-05-29 12:29:58 +00:00
Hans Petter Selasky
2994bac4ba Add some more spinlocks to protect the state of the USB transfer
queue. Rename some functions to indicate locking requirements.

MFC after:	1 week
2014-05-29 11:11:19 +00:00
Hans Petter Selasky
0b4dc07df8 Hook the ISP/SAF1761 driver into MIPS kernel builds.
- Update FDT file for BERI DE4 boards.
- Add needed kernel configuration keywords.
- Rename module to saf1761otg so that the device unit number does not
interfere with the hardware ID in dmesg.

Sponsored by:	DARPA, AFRL
2014-05-29 10:46:09 +00:00
Hans Petter Selasky
580bc2220a Optimise the ISP/SAF1761 driver:
- Use an interrupt filter for handling the data path interrupts. This
increases the throughput significantly.
- Implement support for USB suspend and resume in USB host mode.

Sponsored by:	DARPA, AFRL
2014-05-29 10:06:18 +00:00