Commit Graph

234681 Commits

Author SHA1 Message Date
Warner Losh
71cb2dfcbc Remove special cases for armeb in the build.
Differential Revision:  https://reviews.freebsd.org/D16257
2018-07-17 23:23:54 +00:00
Warner Losh
ff9452772d Remove kernel support for armeb
Remove all the big-endian arm architectures (ixp425 and ixp435)
support in the kernel and associated drivers.

Differential Revision:  https://reviews.freebsd.org/D16257
2018-07-17 23:23:45 +00:00
Warner Losh
e9d6b13d25 Remove armeb from arch.7
Remove all the details about armeb from arch(7), except for its
release start and end, like alpha, ia64 and pc98.

Differential Revision:  https://reviews.freebsd.org/D16257
2018-07-17 23:23:39 +00:00
Warner Losh
8b6b96c93d As discussed several times on freebsd-arch, start to decommission armeb.
Remove armeb/arm from KNOWN_ARCHS.
Remove armeb from arm universe targets.

Differential Revision:	https://reviews.freebsd.org/D16257
2018-07-17 23:23:34 +00:00
Colin Percival
43fbd222cc Check that EC2SNSTOPIC and EC2PUBLIC are non-empty, along with being
defined, in order to avoid problems with thermite.sh.

Reported by:	gjb
2018-07-17 23:16:05 +00:00
Warner Losh
50b1a01ecb Fix compile error introduced in r336245.
Include sys/boot.h to pickup the prototypes for boot_parse_arg.
2018-07-17 23:00:52 +00:00
Warner Losh
5a399a1a39 Remove bogus attempt to simulate scrolling. It's not needed and messes
up serial output. Setting the cursor position after every character is
inefficient, and causes all lines to be over-printed in the serial
console for the boot loader. Allow the terminal to do the emulation.

This isn't completely perfect when the size of the terminal attached
to the serial port isn't the same as 80x25 to match the viedoe console
(or whatever the video console is). While imperfect still, these
changes make it much better.

This makes the serial port useful with UEFI.

Differential Revision: https://reviews.freebsd.org/D16309
2018-07-17 21:45:14 +00:00
Warner Losh
fc4d56e7a2 Fix a typo: change lists to list to match rest of sentence.
Also, I misspoke in r336428. Any devices on sparc64 machines on "isa"
that can do DMA can do 32-bit address DMA and aren't limited to
24-bits of address.
2018-07-17 21:18:49 +00:00
Warner Losh
61e8cc885b Remove VM_FREELIST_ISADMA
Sparc64 doesn't have any devices that do DMA that are attached to isa
or ebus, so we don't need a low-memory pool.
2018-07-17 21:08:01 +00:00
Warner Losh
9ecd7fdebe Remove VM_FREELIST_ISADMA. It's not needed on these architectures.
Differential Review: https://reviews.freebsd.org/D16290
2018-07-17 21:07:53 +00:00
Dexuan Cui
d76fb49fd8 hyperv/hn: Fix panic in hypervisor code upon device detach event
Submitted by:	hselasky
Reviewed by:	dexuan
MFC after:	3 days
Differential Revision:	https://reviews.freebsd.org/D16139
2018-07-17 21:05:08 +00:00
Warner Losh
11df3a2881 More typos
Noticed by: ian@
2018-07-17 20:26:17 +00:00
Warner Losh
debf8b839b Fix typo in the command summary.
Of course, I can't get the command to work, but it's a start...
2018-07-17 20:11:52 +00:00
Colin Percival
a6fe18b04b Teach EC2 AMI release code to send an SNS notification after publishing
new AMIs if EC2SNSTOPIC is defined.

Reviewed by:	gjb
Requested by:	Amazon
2018-07-17 19:05:30 +00:00
Kyle Evans
761d344c16 config(8): Invert checks; envmode/hintmode reflect "env provided"1 2018-07-17 15:16:34 +00:00
Alan Somers
405f09319f audit(4): add tests for _exit(2), cap_enter(2), and cap_getmode(2)
Also, fix a bug in common code that could cause other tests to fail: using
ppoll(2) in combination with buffered I/O for /dev/auditpipe.  Fix it by
disabling buffering.

Submitted by:	aniketp
MFC after:	2 weeks
Sponsored by:	Google, Inc. (GSoC 2018)
Differential Revision:	https://reviews.freebsd.org/D16099
2018-07-17 15:12:55 +00:00
Mark Johnston
9295517ac9 Add a FALLTHROUGH comment to kvprintf().
Submitted by:	Sebastian Huber <sebastian.huber@embedded-brains.de>
MFC after:	3 days
2018-07-17 14:56:54 +00:00
Kyle Evans
63975809ab Fix GCC 4.2 build after r336415, proper declaration and prototype 2018-07-17 14:34:54 +00:00
Kyle Evans
8e62839eb1 config(8): Add compatibility shims for r335998
Plumb the %VERSREQ from Makefile.<arch> through to the rest of config(8).
We've recorded the config(8) version that we're calling "the end of
envmode and hintmode," and we'll write them out for earlier versions. Later
kernel version bumps will remove envmode/hintmode from the kernel as needed,
which is OK since the current kernel does not use them at all.

These compatibility shims really need to go away when the major version
rolls over...

Discussed with:	imp
2018-07-17 14:14:53 +00:00
Kyle Evans
e47edf50dd Revert r336353 completely based on protest; compatibility shims incoming 2018-07-17 14:11:30 +00:00
Bruce Evans
1693fd03d9 Minor cleanups to csqrt*(), mostly in comments.
Remove the STDC CX_LIMITED_RANGE pragma and its verbose comment.  We still
don't have any C99 compilers (that support fenv pragmas), and if we did
then there are thousands of other places in libm that would need to use
them more than here.

The other cleanups are smaller.
2018-07-17 12:01:59 +00:00
Hans Petter Selasky
90c8e44125 Use a mbuf header instead of a mbuf cluster for debugging interrupts in mlx5en(4).
MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 11:53:37 +00:00
Hans Petter Selasky
a6b2d28d05 Add module parameter to limit number of MSIX EQ vectors in mlx5en(4).
For setups having a large amount of PCI devices, it makes sense to limit the
number of MSIX vectors per PCI device, in order to avoid running out of IRQ
vectors.

MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 11:47:56 +00:00
Hans Petter Selasky
aa9f073c9b Add missing newline.
MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 11:43:43 +00:00
Hans Petter Selasky
2f17f76aa4 Handle jumbo frames without requiring big clusters in mlx5en(4).
The scatter list is formed by the chunks of MCLBYTES each, and larger
than default packets are returned to the stack as the mbuf chain.

Submitted by:		kib@
MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 11:42:05 +00:00
Peter Jeremy
7bca8056ad Retrospectively document SVN branch point for stable-11
Approved by:	jhb (mentor)
Differential Revision:	https://reviews.freebsd.org/D16262
2018-07-17 11:35:40 +00:00
Andrey V. Elsukov
acf673edf0 Move invoking of callout_stop(&lle->lle_timer) into llentry_free().
This deduplicates the code a bit, and also implicitly adds missing
callout_stop() to in[6]_lltable_delete_entry() functions.

PR:		209682, 225927
Submitted by:	hselasky (previous version)
MFC after:	2 weeks
Differential Revision:	https://reviews.freebsd.org/D4605
2018-07-17 11:33:23 +00:00
Hans Petter Selasky
f8c3349737 Enable both receive and transmit pauseframes by default in mlx5en(4).
MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 11:21:02 +00:00
Hans Petter Selasky
f2b4782c81 Add context numbers for HW elements in mlx5en(4).
To access the data, set sysctl dev.mce.N.conf.debug_stats to 1.
This enables the sysctl node dev.mce.N.hw_ctx_debug.  Its content is
the mapping of each channel' number to used receive queue and associated
completion queue, set of the transmit queues numbers and corresponding
completion queues.

Trimmed example output:
channel 30 rq 188 cq 1085
channel 30 tc 0 sq 187 cq 1084
channel 31 rq 191 cq 1087
channel 31 tc 0 sq 190 cq 1086

MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 11:18:01 +00:00
Hans Petter Selasky
a880c1ff6a Do not hint about 'trust both' mode when the mlx5en(4) hardware does not support it.
MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 11:11:30 +00:00
Hans Petter Selasky
f0474ab919 Correctly write atomic variable in mlx5en(4).
MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 11:08:40 +00:00
Bruce Evans
9d7dc13615 Fix scaling bugs which gave innaccuracies and spurious underflows in csqrt()
and csqrtl().

When one component is huge and the other is tiny, scaling down the tiny
component gave spurious underflow.

When both components are denormal, not scaling them up gave inaccuracies
of 34+ ulps on not very carefully selected args.  Fixing this reduces the
maximum error to 1.6 ulps on the same set of args (mosly not denormal ones).

The scaling used multiplication of a complex variable by 2, but clang messes
this on amd64 up by losing the sign of -0.0.  Calculate the components
separately, as is well known to be needed for operations on more exceptional
values.
2018-07-17 10:44:16 +00:00
Hans Petter Selasky
52ff436841 Remove redundant call to mlx5_vsc_find_cap() in mlx5core.
MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 10:27:46 +00:00
Hans Petter Selasky
6d54b22db7 Make sure the state variable is set atomically instead of using a mutex in mlx5core.
Device detach and setting error state may deadlock over the interface mutex
like this:

a) Detach code in mlx5en waits until error state is set while the interface
mutex is locked.
b) The set error handler needs to lock the interface mutex before it can
set the error state.

The solution is to use atomics to set the error state.

MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 10:20:01 +00:00
Hans Petter Selasky
b575d8c850 Refactor access to CR-space into using VSC APIs in mlx5core.
Remove no longer used files and APIs.

MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 10:16:32 +00:00
Hans Petter Selasky
9fc929d2e2 Remove redundant newline character in mlx5core.
MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 10:11:00 +00:00
Hans Petter Selasky
18450a3b10 Update version information for the mlx5ib module.
MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 10:07:40 +00:00
Hans Petter Selasky
62bfa774ae Don't pass unsupported events to ibcore from mlx5ib.
MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 09:59:55 +00:00
Hans Petter Selasky
14a1b9bd3a Use static device naming instead of dynamic one in mlx5ib.
When resetting mlx5core instances it can happen that the order of attach and
detach for mlx5ib instances is changed. Take the unit number for mlx5_%d from
the parent PCI device, similarly to what is done in mlx5en(4), so that there
is a direct relationship between mce<N> and mlx5_<N>.

MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 09:58:11 +00:00
Hans Petter Selasky
ed0cee0bf4 Implement support for Differentiated Service Code Point, DSCP, in mlx5en(4).
The DSCP feature is controlled using a set of sysctl(8) fields under
the qos sysctl directory entry for mlx5en(4).

For Routable RoCE QPs, the DSCP should be set in the QP's address path.
The DSCP's value is derived from the traffic class.

Linux commit:
ed88451e1f2d400fd6a743d0a481631cf9f97550

MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 09:56:40 +00:00
Hans Petter Selasky
cda1e10ca0 Use __FBSDID() for RCS tags in ibcore.
MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 09:47:14 +00:00
Hans Petter Selasky
8ac2952539 Remove blank line.
MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 09:44:16 +00:00
Hans Petter Selasky
1bebf70fe7 Add support for IPv6 multicast in ibcore.
This change allows us to join IPv6 multicast networks.

MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 09:37:16 +00:00
Hans Petter Selasky
c68cb3a60a Add support for RoCEv2 multicast in ibcore.
When creating address handle from multicast GID, set MAC according to
the appropriate formula instead of searching for it in the GID table:
- For IPv4 multicast GID use ip_eth_mc_map().
- For IPv6 multicast GID use ipv6_eth_mc_map().

Linux commit:
9636a56fa864464896bf7d1272c701f2b9a57737

MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 09:36:04 +00:00
Hans Petter Selasky
e3e6820f69 Honor return status of ib_init_ah_from_mcmember() in ibcore.
The return status of ib_init_ah_from_mcmember() is ignored by
cma_ib_mc_handler().  Honor it and return error event if ah attribute
initialization failed.

Linux commit:
6d337179f28cc50ddd7e224f677b4cda70b275fc

MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 09:34:29 +00:00
Hans Petter Selasky
0f2b4041fe Honor port_num while resolving GID for IB link layer in ibcore.
ah_attr contains the port number to which cm_id is bound. However, while
searching for GID table for matching GID entry, the port number is
ignored.

This could cause the wrong GID to be used when the ah_attr is converted to
an AH.

Linux commit:
563c4ba3bd2b8b0b21c65669ec2226b1cfa1138b

MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 09:33:20 +00:00
Hans Petter Selasky
56acbf0932 Set IPv4 TOS and IPv6 traffic class field for RoCEv2 traffic in ibcore.
The current implementation assumes a static mapping between
the TOS bits and the priority code point, PCP bits.

MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 09:32:09 +00:00
Hans Petter Selasky
02825401aa Fix for loopback detection in address resolve logic in ibcore.
When a loopback address is detected use the network interface which
has the loopback flag set to trigger loopback logic in address resolve.

MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 09:30:32 +00:00
Hans Petter Selasky
f736cb92cf Check port number supplied by user verbs cmds in ibcore.
The ib_uverbs_create_ah() ind ib_uverbs_modify_qp() calls receive
the port number from user input as part of its attributes and assumes
it is valid. Down on the stack, that parameter is used to access kernel
data structures.  If the value is invalid, the kernel accesses memory
it should not.  To prevent this, verify the port number before using it.

Linux commit:
5ecce4c9b17bed4dc9cb58bfb10447307569b77b
a62ab66b13a0f9bcb17b7b761f6670941ed5cd62
5a7a88f1b488e4ee49eb3d5b82612d4d9ffdf2c3

MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 09:29:14 +00:00
Hans Petter Selasky
56326b4755 Depend on IPv6 stack to resolve link local address for RoCEv2 in ibcore.
RoCEv1 does not use the IPv6 stack to resolve the link local DGID since it
uses GID address. It forms the DMAC directly from the DGID.

Linux commit:
56d0a7d9a0f045ee27a001762deac28c7d28e2e4

MFC after:		1 week
Sponsored by:		Mellanox Technologies
2018-07-17 09:27:31 +00:00