Commit Graph

106981 Commits

Author SHA1 Message Date
Rick Macklem
b179878dde Revert r283330 since it broke directory caching in the client.
At this time I cannot see a way to fix directory caching when it
has partial blocks in the buffer cache, due to the fact that the
syscall's uio_offset won't stay the same as the lblkno * NFS_DIRBLKSIZ
offset.

Reported by:	bde
MFC after:	2 weeks
2015-11-21 00:15:41 +00:00
Gleb Smirnoff
09c837b897 Remove remnants of the old NFS from vnode pager.
Reviewed by:	kib
Sponsored by:	Netflix
2015-11-20 23:52:27 +00:00
Ed Maste
380392f2f9 Disconnect unused kgzldr from sys/boot/i386
Sponsored by:	The FreeBSD Foundation
2015-11-20 21:56:20 +00:00
Warner Losh
c192c5d300 Makeoption ARM_LITLE_ENDIAN does nothing. Remove it since it isn't
consistently used. It was a carry over from NetBSD that FreeBSD
doesn't use.
2015-11-20 21:49:46 +00:00
Andrew Turner
805471578e Remove bus_space_asm_generic.S from the per-SoC files.* files, it's already
in files.arm.

Sponsored by:	ABT Systems Ltd
2015-11-20 19:48:32 +00:00
Bryan Drewery
d9b8630978 Remove disconnected boot0ext.
This was [re-]added in r127458 (apparently with disconnected history as well)
and never connected to the build.

It is no longer relevant for modern systems from the past 20 years.

Discussed with:	jhb
Sponsored by:	EMC / Isilon Storage Division
2015-11-20 18:10:02 +00:00
Andrew Turner
09f449399b Remove pl310.c from the SoC std.* files, it's in files.arm
Sponsored by:	ABT Systems Ltd
2015-11-20 16:43:21 +00:00
Andrew Turner
c92ef78663 DTrace is known to work on armv6, enable building it as a module.
Reviewed by:	imp
Sponsored by:	ABT Systems Ltd
Differential Revision:	https://reviews.freebsd.org/D4221
2015-11-20 16:18:27 +00:00
Andrew Turner
aeef645f92 Stop setting {KERN,}PHYSADDR on armv6, it's unneeded.
Sponsored by:	ABT Systems Ltd
2015-11-20 16:12:22 +00:00
Alexander Motin
1fd7e8637c Some cosmetics for ancient cards. 2015-11-20 14:20:24 +00:00
Edward Tomasz Napierala
a8723fb881 The freebsd4_getfsstat() was broken in r281551 to always return 0 on success.
All versions of getfsstat(3) are supposed to return the number of [o]statfs
structs in the array that was copied out.

Also fix missing bounds checking and signed comparison of unsigned types.

Submitted by:	bde@
MFC after:	1 month
Sponsored by:	The FreeBSD Foundation
2015-11-20 14:08:12 +00:00
Svatopluk Kraus
8ac1f62137 Fix build when KTR is defined but not KTR_TULIP.
Approved by:	kib (mentor)
2015-11-20 09:37:04 +00:00
Svatopluk Kraus
3ee1f8b1b7 Add usermode variable to KTR output. Fix style.
Approved by:	kib (mentor)
2015-11-20 09:34:59 +00:00
Svatopluk Kraus
a7d84eda93 Fix style and argument count for KTR.
Approved by:	kib (mentor)
2015-11-20 09:33:22 +00:00
Alexander Motin
1e0e8c83b5 Optimize SNS_GID_FT request scratch memory usage.
Now with present 4K of scratch we can fetch up to 508 ports (16 more).
2015-11-20 08:54:19 +00:00
Marius Strobl
8fd47ac11c Avoid a NULL pointer dereference in bounce_bus_dmamap_sync() when the
map has been created via bounce_bus_dmamem_alloc(). Even for coherent
DMA - which bus_dmamem_alloc(9) typically is used for -, calling of
bus_dmamap_sync(9) isn't optional.

PR:		188899 (non-original problem)
MFC after:	3 days
2015-11-20 02:23:35 +00:00
Conrad Meyer
70bca634d0 if_ntb: Initialize if_mtu to the correct MTU
Lower the payload data (IP) portion of the MTU from 0x10000 to
IP_MAXPACKET (0xFFFF) to avoid panicing the IP stack.

Sponsored by:	EMC / Isilon Storage Division
2015-11-19 19:53:19 +00:00
Conrad Meyer
f51a1fe048 if_ntb: Add Xeon link watchdog register writes
This feature is disabled by default.  To enable it, tune
hw.if_ntb.enable_xeon_watchdog to non-zero.

If enabled, writes an unused NTB register every second to demonstrate to
a hardware watchdog that the NTB device is still alive.  Most machines
with NTB will not need this -- you know who you are.

Sponsored by:	EMC / Isilon Storage Division
2015-11-19 19:53:09 +00:00
Alexander Motin
eea52482f9 Another round of port scanner rewrite.
This change simplifies and unifies port adding/updating for loop and
fabric scanners.  It also fixes problems with scanning restarts due to
concurrent port databases changes.  It also fixes many cosmetic issues.
2015-11-19 17:43:47 +00:00
Michael Tuexen
bd5b567213 Fix a bug where an SCTP association was moved back to SHUTDOWN_SENT
state when the user issued a shutdown() call.

MFC after: 	1 week
2015-11-19 16:46:00 +00:00
Jonathan T. Looney
1067a2ba68 Consistently enforce the restriction against calling malloc/free when in a
critical section.

uma_zalloc_arg()/uma_zalloc_free() may acquire a sleepable lock on the
zone. The malloc() family of functions may call uma_zalloc_arg() or
uma_zalloc_free().

The malloc(9) man page currently claims that free() will never sleep.
It also implies that the malloc() family of functions will not sleep
when called with M_NOWAIT. However, it is more correct to say that
these functions will not sleep indefinitely. Indeed, they may acquire
a sleepable lock. However, a developer may overlook this restriction
because the WITNESS check that catches attempts to call the malloc()
family of functions within a critical section is inconsistenly
applied.

This change clarifies the language of the malloc(9) man page to clarify
the restriction against calling the malloc() family of functions
while in a critical section or holding a spin lock. It also adds
KASSERTs at appropriate points to make the enforcement of this
restriction more consistent.

PR:		204633
Differential Revision:	https://reviews.freebsd.org/D4197
Reviewed by:	markj
Approved by:	gnn (mentor)
Sponsored by:	Juniper Networks
2015-11-19 14:04:53 +00:00
Hans Petter Selasky
22f2c49ab1 Add the mlx5 and mlx5en modules to the i386 and amd64 kernel builds by
default and add a manual page for mlx5en. The mlx5 module contains
shared code for both infiniband and ethernet. The mlx5en module
contains specific code for ethernet functionality only. A mlx5ib
module is in the works for infiniband support.

Supported hardware:
- ConnectX-4: 10/20/25/40/50/56/100Gb/s speeds.
- ConnectX-4 LX: 10/25/40/50Gb/s speeds (low power consumption)

Refer to the mlx5en(4) manual page for a comprehensive list.

The team porting the mlx5 driver(s) to FreeBSD:
- Hans Petter Selasky <hselasky@freebsd.org>
- Oded Shanoon <odeds@mellanox.com>
- Meny Yossefi <menyy@mellanox.com>
- Shany Michaely <shanim@mellanox.com>
- Shahar Klein <shahark@mellanox.com>
- Daria Genzel <dariaz@mellanox.com>
- Mark Bloch <markb@mellanox.com>

Differential Revision:	https://reviews.freebsd.org/D4163
Submitted by:	Mark Block <markb@mellanox.com>
Sponsored by:	Mellanox Technologies
Reviewed by:	gnn @
MFC after:	3 days
2015-11-19 12:55:43 +00:00
Hans Petter Selasky
bb3853c6bd Style changes, mostly automated.
Differential Revision:	https://reviews.freebsd.org/D4179
Submitted by:	Daria Genzel <dariaz@mellanox.com>
Sponsored by:	Mellanox Technologies
MFC after:	3 days
2015-11-19 10:28:51 +00:00
Hans Petter Selasky
ee09079968 Accumulate out of RX buffers into a 64-bit value and subtract out of
RX buffers from number of received packets.

Differential Revision:	https://reviews.freebsd.org/D4178
Submitted by:	Drew Gallatin <gallatin@freebsd.org>
Sponsored by:	Mellanox Technologies
MFC after:	3 days
2015-11-19 10:23:10 +00:00
Hans Petter Selasky
36c1007d35 Maintain the "hw_lro" configuration variable correctly.
Setting sysctl dev....conf.hw_lro may fail if the net device lro is
turned off. Due to the nature of our sysctl handler we need to set the
values back to 0 and issue an error.

Differential Revision:	https://reviews.freebsd.org/D4177
Submitted by:	Shahar Klein <shahark@mellanox.com>
Sponsored by:	Mellanox Technologies
MFC after:	3 days
2015-11-19 10:18:13 +00:00
Hans Petter Selasky
7e1b8bc0c9 Print cable name, if cable type is not recognized.
Differential Revision:	https://reviews.freebsd.org/D4180
Submitted by:	Mark Bloch <markb@mellanox.com>
Sponsored by:	Mellanox Technologies
MFC after:	3 days
2015-11-19 10:10:52 +00:00
Alexander Motin
ece9d8b702 Improve locking of sg_threadcount.
MFC after:	1 week
2015-11-19 08:04:05 +00:00
Mark Johnston
150b709793 Remove a commented-out debug print.
MFC after:	1 week
2015-11-19 05:58:51 +00:00
Mark Johnston
1b9254f885 Add support for a configurable output channel to witness(4).
This is useful in environments where system configuration is performed by
automated interaction with the system console, since unexpected witness
output makes such automation difficult. With this change, the new
debug.witness.output_channel sysctl allows one to specify that witness
output is to be printed to the kernel log (using log(9)) rather than the
console.

Reviewed by:	cem, jhb
MFC after:	2 weeks
Relnotes:	yes
Sponsored by:	EMC / Isilon Storage Division
Differential Revision:	https://reviews.freebsd.org/D4183
2015-11-19 05:56:59 +00:00
Mark Johnston
07713dde22 Add vlog(9).
Reviewed by:	cem, jhb
MFC after:	1 week
Differential Revision:	https://reviews.freebsd.org/D4183
2015-11-19 05:50:22 +00:00
Mark Johnston
4d7296f9aa Fix a bug in the amd64 dtrace_getarg() implementation: when unwinding the
stack, take into account the copy of rsi pushed between the breakpoint
trapframe and the dtrace_invop frame. Prior to r287644, this was covered
by the fact that sizeof(struct amd64_frame) was 24 rather than 16.

Reported by:	smh
2015-11-19 05:33:15 +00:00
Justin Hibbits
d12c556d32 Revert r291009 until rman changes go in.
Pointy-hat to:	jhibbits
2015-11-19 04:41:16 +00:00
Warner Losh
29d1144aeb Mark the mostly redundant kernels that just pull
in something from _BASE as NO_UNIVERSE

Differential Revision: https://reviews.freebsd.org/D4200
2015-11-19 01:58:12 +00:00
Nathan Whitehorn
0df013a065 The Linux kexec boot loader doesn't need a font built in to it. This got
copied-and-pasted from the PS3 loader.
2015-11-19 01:57:23 +00:00
Enji Cooper
ae9356f143 Don't leak work if __mlx4_register_vlan(..) fails in
mlx4_master_immediate_activate_vlan_qos(..)

MFC after: 3 days
Differential Revision: https://reviews.freebsd.org/D4203
Submitted by: Miles Olrich <miles.olrich@isilon.com>
Sponsored by: EMC / Isilon Storage Division
2015-11-19 01:08:16 +00:00
Conrad Meyer
55faae77fa in6_mc_get: Fix recursion on if_addr_lock on malloc failure
Analogously to r291040, in6_mc_get recurses on if_addr_lock if the
M_NOWAIT allocation fails.  The fix is the same.

Suggested by:	Andrey V. Elsukov
Reviewed by:	jhb (ip4 version)
Sponsored by:	EMC / Isilon Storage Division
Differential Revision:	https://reviews.freebsd.org/D4138 (ip4 version)
2015-11-19 00:27:26 +00:00
Conrad Meyer
3fbd30d495 in_getmulti: Fix recursion on if_addr_lock on malloc failure
When the M_NOWAIT allocation fails, we recurse the if_addr_lock trying
to clean up.  Reorder the cleanup after dropping the if_addr_lock.  The
obvious race is already possible between if_addmulti and IF_ADDR_WLOCK
above, so it must be ok.

Submitted by:	Ryan Libby <rlibby@gmail.com>
Reviewed by:	jhb
Found with:	M_NOWAIT failure injection testing
Sponsored by:	EMC / Isilon Storage Division
Differential Revision:	https://reviews.freebsd.org/D4138
2015-11-18 23:53:13 +00:00
Rick Macklem
69527b11bb The problem report was for a crash that happened when smbfs was
trying to do a mount. Given the backtrace,
it appears that the crash occurred when smb_vc_create() failed and then
called smb_vc_put() with vcp->vc_iod == NULL. smb_vc_put() subsequently
called smb_vc_disconnect() with vcp->vc_iod == NULL, causing the crash.
This patch adds a check for vcp->vc_iod != NULL in smb_vc_disconnect() to
avoid the crash. It also fixes the case in smb_vc_create() where
kproc_create() fails so that it destroys the mutexes and sets
vcp->vc_iod == NULL before free()'ing the iod structure.
The person who reported the PR tested the patch, but was not able
to reproduce the crash with or without the patch.

PR:		201912
Reviewed by:	jhb
MFC after:	2 weeks
2015-11-18 23:04:01 +00:00
Conrad Meyer
e90df06388 if_ntb: Reuse receive buffers correctly
Discard the unused rx_free_q.  Instead, reuse inputed packets by putting
them back on the *pend* queue after reinitialization.

If tx or rx handlers are unavailable, free mbufs rather than leaking
them.

With this change, if_ntb can receive more than 100
(NTB_QP_DEF_NUM_ENTRIES) packets.

Sponsored by:	EMC / Isilon Storage Division
2015-11-18 22:20:49 +00:00
Conrad Meyer
c95b032ec6 NTB: Expose 32-bit BAR limits to consumers
32-bit BARs can only address memory mapped in the low 32 bits of
physical RAM.  Expose this as a 'plimit' out parameter from
ntb_mw_get_range().

Fix if_ntb to allocate memory within this limit.

Sponsored by:	EMC / Isilon Storage Division
2015-11-18 22:20:40 +00:00
Conrad Meyer
4e1eea459a NTB: Mask off the low 12 bits of address/range registers
Sometimes they'll read spurious values (observed: 0xc on Broadwell-DE),
failing link negotiation.

Discussed with:	Dave Jiang, Allen Hubbe
Sponsored by:	EMC / Isilon Storage Division
2015-11-18 22:20:31 +00:00
Conrad Meyer
57ba792a67 ntb_hw: Add programmatic interface to enable/disable WC
Enable users to enable/disable WC on memory windows programmatically.

Sponsored by:	EMC / Isilon Storage Division
2015-11-18 22:20:21 +00:00
Conrad Meyer
3062870897 ntb_hw: Add tunable to disable write-combining
The tunable 'hw.ntb.enable_writecombine' may be set to zero to
administratively disable write combining the mapped NTB region.

Sponsored by:	EMC / Isilon Storage Division
2015-11-18 22:20:13 +00:00
Conrad Meyer
552573e11c NTB: Fix 32-bit BAR size validation
Sponsored by:	EMC / Isilon Storage Division
2015-11-18 22:20:04 +00:00
Conrad Meyer
30994d9192 if_ntb: Diff reduce with Linux
Use bus_space_write instead of (non-volatile) C pointer writes via an
iowrite32() shim in the same places as the Dual BSD/GPL Linux driver.

Update some types to fixed 32-bit sizes.

Sponsored by:	EMC / Isilon Storage Division
2015-11-18 22:19:55 +00:00
Roger Pau Monné
1522652230 xen: fix dropping bitmap IPIs during resume
Current Xen resume code clears all pending bitmap IPIs on resume, which is
not correct. Instead re-inject bitmap IPI vectors on resume to all CPUs in
order to acknowledge any pending bitmap IPIs.

Sponsored by:		Citrix Systems R&D
MFC after:		2 weeks
2015-11-18 18:11:19 +00:00
Roger Pau Monné
ea64b86f94 xen/intr: properly dispose event channels on resume
All event channels are torn down when performing a migration on Xen, make
sure all handlers are also removed and the event channel structure is
properly disposed so it can be reused.

Sponsored by:		Citrix Systems R&D
MFC after:		2 weeks
2015-11-18 18:10:28 +00:00
Roger Pau Monné
531cfe55e2 x86/intr: allow mutex recursion in intr_remove_handler
This is needed so interrupt handlers can be removed while the PIC is
resuming, it was previously not possible due to intr_resume holding the
intr_table_lock and intr_remove_handler recursing on it.

Sponsored by:		Citrix Systems R&D
Reviewed by:		kib (previous version)
MFC after:		2 weeks
Differential Revision:	https://reviews.freebsd.org/D4114
2015-11-18 18:09:49 +00:00
Nathan Whitehorn
686d2f317a Extend r270123 to run the brand info's header_supported() routine for
branded as well as unbranded binaries. This will be required to add
support for the new ELFv2 ABI on powerpc64, which is distinguished from
ELFv1 by the contents of the ELF header's flags field.

Reviewed by:	imp
MFC after:	2 weeks
2015-11-18 17:03:22 +00:00
Michal Meloun
bb12cf0137 ARM: Fix dma_dcache_sync() for early allocated memory.
Drivers can request DMA to buffers that are not in memory represented
in the vm page arrays. Because of this, store KVA of already mapped
buffer to synclist and use it in dma_dcache_sync().

Reviewed by:	jah
Approved by:	kib (mentor)
Differential Revision: https://reviews.freebsd.org/D4120
2015-11-18 16:07:01 +00:00