Commit Graph

205242 Commits

Author SHA1 Message Date
Alexander Motin
2626fa27ad Remove some unneeded code. 2015-10-29 20:43:13 +00:00
Alexander Motin
030eb8d0f2 Remove reset delays for which I see neither explanation nor need. 2015-10-29 20:34:01 +00:00
Conrad Meyer
217b098a1e ntb: Revert r290130 now that r290156 has landed
Nagged by:	vangyzen
Sponsored by:	EMC / Isilon Storage Division
2015-10-29 19:35:01 +00:00
Bryan Drewery
bd9302d862 Check archive_entry_new() result.
Coverity CID:	1331341
2015-10-29 19:28:21 +00:00
Conrad Meyer
0e5d2011ae pmap_change_attr: Only fixup DMAP for DMAPed ranges
pmap_change_attr must change the memory type of both the requested KVA
and the corresponding DMAP mappings (if such mappings exist), to satisfy
an Intel requirement that two or more mappings to the same physical
pages must have the same memory type.

However, not all kernel mapped pages have corresponding DMAP mappings --
for example, 64-bit BARs.  Skip fixing up the DMAP for out-of-bounds
addresses.

Submitted by:	Steve Wahl <steve_wahl@dell.com>
Reviewed by:	alc, jhb
Sponsored by:	Dell Compellent
Differential Revision:	https://reviews.freebsd.org/D4030
2015-10-29 19:07:00 +00:00
Bryan Drewery
156c04f793 getnewbuf: Initialize bp to avoid uninitialized pointer dereference and brelse().
This came in recently in r289279.

Coverity CID:	1331561
2015-10-29 19:02:24 +00:00
Bryan Drewery
2780ba06c7 Avoid passing an uninitialized 'i'. Currently nothing was depending on it
anyhow.

Coverity CID:	1331562
2015-10-29 18:58:18 +00:00
Bryan Drewery
b1585b5a06 Fix unlikely memory leak.
It is unlikely since the first check in the function is that dir[0] is '/',
but later code changes may make it real.

Coverity CID:	1332104
2015-10-29 18:29:28 +00:00
Warner Losh
4609a2a6fd PC Card and Cardbus are now in extended maintenance mode. No need to
have them cluttering up MAINTAINERS.
2015-10-29 16:50:28 +00:00
Alexander Motin
2e6beaf19e Fix and improve error masking and reporting. 2015-10-29 16:48:12 +00:00
John Baldwin
2219c44a1f Update for LINUX32 rename. The assembler didn't complain about undefined
symbols but just used 0 after the rename.
2015-10-29 15:20:47 +00:00
John Baldwin
6cea44a704 Fix build with DEBUG defined.
Reported by:	hselasky
2015-10-29 15:16:47 +00:00
Hans Petter Selasky
cb3450e26e Add missing NULL check in physio().
When destroying a character device the si_devsw field is set to NULL
before all references are gone, to indicate the character device is
going away. This can cause a NULL-dereference fault inside physio().

The callers of physio() should own a thread reference on the cdev and
if si_devsw is seen as non-NULL, it is usable during the execution of
the function. Else an ENXIO error code is returned.

Reviewed by:	kib
MFC after:	2 weeks
2015-10-29 13:53:37 +00:00
Alexander Motin
043fd51a74 Some minor additions to r290138, 2015-10-29 10:31:44 +00:00
Alexander Motin
0fd0e797e8 Some updates to isp(4) manual page. 2015-10-29 09:50:48 +00:00
Hans Petter Selasky
6fb116c111 Add myself to MAINTAINERS. 2015-10-29 08:45:56 +00:00
Hans Petter Selasky
8d59ecb214 Finish process of moving the LinuxKPI module into the default kernel build.
- Move all files related to the LinuxKPI into sys/compat/linuxkpi and
  its subfolders.
- Update sys/conf/files and some Makefiles to use new file locations.
- Added description of COMPAT_LINUXKPI to sys/conf/NOTES which in turn
  adds the LinuxKPI to all LINT builds.
- The LinuxKPI can be added to the kernel by setting the
  COMPAT_LINUXKPI option. The OFED kernel option no longer builds the
  LinuxKPI into the kernel. This was done to keep the build rules for
  the LinuxKPI in sys/conf/files simple.
- Extend the LinuxKPI module to include support for USB by moving the
  Linux USB compat from usb.ko to linuxkpi.ko.
- Bump the FreeBSD_version.
- A universe kernel build has been done.

Reviewed by:	np @ (cxgb and cxgbe related changes only)
Sponsored by:	Mellanox Technologies
2015-10-29 08:28:39 +00:00
Kevin Lo
e3cf3d4428 Remove the static function declaration. 2015-10-29 04:51:27 +00:00
Kevin Lo
1c1cd920d7 - Add a missing prototype
- Fix typos
2015-10-29 04:21:34 +00:00
Conrad Meyer
1ffae6e80a ioat_test: Handled forced hardware resets gracefully
Sponsored by:	EMC / Isilon Storage Division
2015-10-29 04:16:52 +00:00
Conrad Meyer
5f77bd3e24 ioat: Drain/quiesce the device less racily
On detach and during a forced HW reset.

Sponsored by:	EMC / Isilon Storage Division
2015-10-29 04:16:39 +00:00
Conrad Meyer
79c1a0199f ntb: Do not attempt to set write-combining on MWs
AMD64 pmap assumes ranges will be in the DMAP, which isn't necessarily
true for NTB memory windows (especially 64-bit BARs).

Suggested by:	pmap_change_attr_locked -> kassert_panic
Sponsored by:	EMC / Isilon Storage Division
2015-10-29 04:16:28 +00:00
Conrad Meyer
e9497f9bbd ioatcontrol(8): Add and document "raw" testing mode
Allows DMA from/to arbitrary KVA or physical address.  /dev/ioat_test
must be enabled by root and is only R/W root, so this is approximately
as dangerous as /dev/mem and /dev/kmem.

Sponsored by:	EMC / Isilon Storage Division
2015-10-29 04:16:16 +00:00
Kevin Lo
52eab858c0 Add MLINKS for if_otus(4), if_rsu(4) and if_urtwn(4). 2015-10-29 03:28:28 +00:00
Kevin Lo
5909575ebf Xref otus(4). 2015-10-29 03:22:27 +00:00
Bryan Drewery
0edd283235 Fix regression from using .USEBEFORE in _SUBDIR in r289705.
Using .USEBEFORE had the unintended side-effect of changing the directory for
the real target ran in the current directory.  For example this meant that
the 'make clean' would run in one of the SUBDIR.

Sponsored by:	EMC / Isilon Storage Division
Pointyhat to:	bdrewery
2015-10-29 00:41:03 +00:00
Adrian Chadd
948457f1be Oops - use the wrong array offset. 2015-10-28 23:39:33 +00:00
Hiren Panchasara
12eeb81fc1 Calculate the correct amount of bytes that are in-flight for a connection as
suggested by RFC 6675.

Currently differnt places in the stack tries to guess this in suboptimal ways.
The main problem is that current calculations don't take sacked bytes into
account. Sacked bytes are the bytes receiver acked via SACK option. This is
suboptimal because it assumes that network has more outstanding (unacked) bytes
than the actual value and thus sends less data by setting congestion window
lower than what's possible which in turn may cause slower recovery from losses.

As an example, one of the current calculations looks something like this:
snd_nxt - snd_fack + sackhint.sack_bytes_rexmit
New proposal from RFC 6675 is:
snd_max - snd_una - sackhint.sacked_bytes + sackhint.sack_bytes_rexmit
which takes sacked bytes into account which is a new addition to the sackhint
struct. Only thing we are missing from RFC 6675 is isLost() i.e. segment being
considered lost and thus adjusting pipe based on that which makes this
calculation a bit on conservative side.

The approach is very simple. We already process each ack with sack info in
tcp_sack_doack() and extract sack blocks/holes out of it. We'd now also track
this new variable sacked_bytes which keeps track of total sacked bytes reported.

One downside to this approach is that we may get incorrect count of sacked_bytes
if the other end decides to drop sack info in the ack because of memory pressure
or some other reasons. But in this (not very likely) case also the pipe
calculation would be conservative which is okay as opposed to being aggressive
in sending packets into the network.

Next step is to use this more accurate pipe estimation to drive congestion
window adjustments.

In collaboration with:	rrs
Reviewed by:		jason_eggnet dot com, rrs
MFC after:		2 weeks
Sponsored by:		Limelight Networks
Differential Revision:	https://reviews.freebsd.org/D3971
2015-10-28 22:57:51 +00:00
Jung-uk Kim
9351ee8ccb Define endianness for non-x86 platforms.
MFC after:	3 days
2015-10-28 22:49:37 +00:00
Jason A. Harmening
d58d7ad4a8 Retire pmap_dmap_iscurrent(). It is only a wrapper around pmap_is_current(), and is no longer called. 2015-10-28 21:17:38 +00:00
Warner Losh
c64c9b8e46 BUS_ADD_CHILD calls device_add_child. device_add_child does not call
BUS_ADD_CHILD. Make it explicit since it follows the command paradigm
rather than the callback paradigm. Add other clarifying notes as well.
2015-10-28 19:11:06 +00:00
Alexander Motin
668c0ec64f Change the way how target mode is enabled on 23xx chips.
Without docs I am not completely sure about this, but on my tests new
method works better then previous, at least with our latest firmware.
2015-10-28 19:08:51 +00:00
Warner Losh
83a283cf62 Add a note to the effect that BUS_ADD_CHILD calls
device_add_child_ordered to add the child. device_add_child_ordered
doesn't call BUS_ADD_CHILD.
2015-10-28 18:53:18 +00:00
Andrey V. Elsukov
50bc87bc43 Check the size of data available in mbuf, before using them.
PR:		202667
MFC after:	1 week
2015-10-28 17:55:37 +00:00
Bryan Drewery
e6b0222d60 Include libutil's headers directly from src to avoid recording a dirdeps
dependency for META MODE.
2015-10-28 16:31:04 +00:00
Bryan Drewery
985dfe52e9 Connect mpsutil for META MODE. 2015-10-28 16:22:00 +00:00
Eric van Gyzen
9300da1d3d Fix spelling and grammer in tools/test/README.
Reviewed by:	gnn
2015-10-28 15:00:21 +00:00
Andrey A. Chernov
8b63538d89 Add _flags2 per jhb@ suggestion since no room left in _flags.
Rewrite O_APPEND flag checking using new __S2OAP flag.

MFC after:      3 weeks
2015-10-28 14:40:02 +00:00
Andrew Turner
4648ee9525 Remove the s3c2xx0 code, it's no longer used. As far as I know I as the
main user of this code, however I haven't used it in over two years, and
don't expect to in the future.
2015-10-28 13:14:16 +00:00
Andrew Turner
7c07418e51 Start to remove support for the XScale i80321. As far as I can tell nobody
uses this which makes it difficult to support.
2015-10-28 13:07:56 +00:00
Alexander Motin
b6983e5fc8 Improve/fix loop scanning routine.
For the most of chips (except anscient ones) port handlers have no relation
to port IDs.  In such situation old code scanning first 125 handlers was
quite naive.  Instead of doing that, send to chip single request to get full
list of port handlers available on specific virtual port and scan only them.

Old code had problems with case of several virtual ports enabled, when port
handlers allocated from global address space could easily go above 125.
This change was successfully tested on 23xx, 24xx and 25xx chips in loop
mode with 4 virtual initiator ports, each seing 50 virtual target ports.
2015-10-28 12:34:59 +00:00
Baptiste Daroussin
c94fbe2671 Connect mpsutil(8) to the build
Sponsored by:	Gandi.net
2015-10-28 12:31:29 +00:00
Baptiste Daroussin
7761e4e365 Merge mpsutil(8) branch
mpsutil(8)/mprutil(8) are new utilities for managing LSI Fusion-MPT
2/3 controllers (mps(4) and mpr(4))

For now only informational commands have been implemented.

This utility has been written by scottl@ [1] and polished by myself[2]

Submitted by:	scottl
Discussed with:	scottl
Relnotes:	yes
Sponsored by:	Netflix [1]
Sponsored by:	Gandi.net [2]
2015-10-28 12:30:29 +00:00
Hans Petter Selasky
8e9ef12d1e Build fix for i386/XBOX and pc98/GENERIC.
Reviewed by:	kib
2015-10-28 12:10:01 +00:00
Baptiste Daroussin
11d38a5764 Merge from head
Sponsored by:	Gandi.net
2015-10-28 11:58:18 +00:00
Baptiste Daroussin
4080404605 Fix a typo in the comment about the include guard
Sponsored by:	Gandi.net
2015-10-28 11:29:58 +00:00
Baptiste Daroussin
08a1f267fa Remove more of the mptutil unused code
Sponsored by:	Gandi.net
2015-10-28 11:28:46 +00:00
Baptiste Daroussin
5114c60e8a Remove some unused code from the mptutil(8)
The code from mptutil living in the same source tree the day those functions
will be implemented in the mpsutil(8) it can be copy/paste easily

Sponsored by:	Gandi.net
2015-10-28 11:25:28 +00:00
Baptiste Daroussin
76103694a0 Reduce overlinking, for now neither cam nor libutil are needed
Sponsored by:	Gandi.net
2015-10-28 11:20:55 +00:00
Baptiste Daroussin
64dbc39cd3 Do enforce "-g", DEBUG_FLAGS can be user if one want to set it
Sponsored by:	Gandi.net
2015-10-28 11:18:43 +00:00