87405 Commits

Author SHA1 Message Date
Konstantin Belousov
4596f12de7 MFamd64 r238597:
Add stmxcsr.

MFC after:	3 weeks
2012-07-21 21:39:23 +00:00
Konstantin Belousov
1a3f27687e MFamd64 r238669:
Force clean FPU state in PCB user FPU save area for PT_I386_{GET,SET}XMMREGS.

Reported by:	bde
MFC after:	1 week
2012-07-21 21:39:02 +00:00
Alexander Motin
b8aff31b25 Use 16bit PIO instead of 32bit in case of misaligned buffer.
It fixes kernel panic during CD write with cdrecord on sparc64.
2012-07-21 14:59:43 +00:00
Gleb Smirnoff
1fc2ea61bf Fix typo in comment, should be MHz here.
Submitted by:	Daan Vreeken <Daan vitsch.nl>
2012-07-21 14:07:43 +00:00
Konstantin Belousov
7c80fcfdba Constently use 2-space sentence breaks.
Submitted by:	 bde
MFC after:	 1 week
2012-07-21 13:53:00 +00:00
Konstantin Belousov
1965c139f1 Stop caching curpcb in the local variable.
Requested by:	    bde
MFC after:	    1 week
2012-07-21 13:47:37 +00:00
Konstantin Belousov
700de5109a The PT_I386_{GET,SET}XMMREGS and PT_{GET,SET}XSTATE operate on the
stopped threads. Implementation assumes that the thread's FPU context
is spilled into the PCB due to stop. This is mostly true, except when
FPU state for the thread is not initialized. Then the requests operate
on the garbage state which is currently left in the PCB, causing
confusion.

The situation is indeed observed after a signal delivery and before
#NM fault on execution of any FPU instruction in the signal handler,
since sendsig(9) drops FPU state for current thread, clearing
PCB_FPUINITDONE. When inspecting context state for the signal handler,
debugger sees the FPU state of the main program context instead of the
clear state supposed to be provided to handler.

Fix this by forcing clean FPU state in PCB user FPU save area by
performing getfpuregs(9) before accessing user FPU save area in
ptrace_machdep.c.

Note: this change will be merged to i386 kernel as well, where it is
much more important, since e.g. gdb on i386 uses PT_I386_GETXMMREGS to
inspect FPU context on CPUs that support SSE. Amd64 version of gdb
uses PT_GETFPREGS to inspect both 64 and 32 bit processes, which does
not exhibit the bug.

Reported by:	bde
MFC after:	1 week
2012-07-21 13:06:37 +00:00
Konstantin Belousov
dfa8a51288 Stop clearing x87 exceptions in the #MF handler on amd64. If user code
understands FPU hardware enough to catch SIGFPE and unmask exceptions
in control word, then it may as well properly handle return from
SIGFPE without causing an infinite loop of #MF exceptions due to
faulting instruction restart, when needed.

Clearing exceptions causes information loss for handlers which do
understand FPU hardware, and struct siginfo si_code member cannot be
considered adequate replacement for en_sw content due to translation.

Supposed reason for clearing the exceptions, which is IRQ13 handling
oddities, were never applicable to amd64.

Note: this change will be merged to i386 kernel as well, since we do
not support IRQ13 delivery of #MF notifications for some time.

Requested by:	bde
MFC after:	1 week
2012-07-21 13:05:34 +00:00
Konstantin Belousov
a53cab2c6c (Incomplete) fixes for symbols visibility issues and style in fcntl.h.
Append '__' prefix to the tag of struct oflock, and put it under BSD
namespace. Structure is needed both by libc and kernel, thus cannot be
hidden under #ifdef _KERNEL.

Move a set of non-standard F_* and O_* constants into BSD namespace.
SUSv4 explicitely allows implemenation to pollute F_* and O_* names
after fcntl.h is included, but it costs us nothing to adhere
to the specification if exact POSIX compliance level is requested by
user code.

Change some spaces after #define to tabs.

Noted by and discussed with:	     bde
MFC after:   1 week
2012-07-21 13:02:11 +00:00
Alexander Motin
d17996abb0 Fix typo in bzero length argument during sense fetching.
For me it at least fixed CD burning in PIO mode.

MFC after:	3 days
2012-07-21 08:19:43 +00:00
Edward Tomasz Napierala
a1cf7f75a6 Make it possible to resize opened partitions.
Sponsored by:	FreeBSD Foundation
2012-07-20 17:51:20 +00:00
Edward Tomasz Napierala
b75ca29147 Make ZVOL resizing ('zfs set volsize') properly resize the GEOM provider.
Sponsored by:	FreeBSD Foundation
2012-07-20 16:56:34 +00:00
Adrian Chadd
be4f96a6b7 Introduce a rate table TLV so rate table statistics consumers
know how to map rix -> rate code.
2012-07-20 02:17:48 +00:00
Adrian Chadd
42420dccd5 Bump this up to match what the HAL is at now. 2012-07-20 01:41:18 +00:00
Adrian Chadd
b8f2a85349 Enable the basic node-based rate control statistics via an ioctl(). 2012-07-20 01:36:46 +00:00
Adrian Chadd
2d20d6559d Add a per-node rate control routine for each rate control module.
For now, the only module implement is 'sample', and that's only partially
implemented.  The main issue here with reusing this structure in userland
is that it uses 'rix' everywhere, which requires the userland code to
have access to the current HAL rate table.

For now, this is a very large work in progress.

Specific details:

* The rate control information is per-node at the moment and wrapped
  in a TLV, to ease parsing and backwards compatibility.
* .. but so I can be slack for now, the userland statistics are just
  a copy of the kernel-land sample node state.
* However, for now use a temporary copy and change the rix entries
  to dot11rate entries to make it slightly easier to eyeball.

Problems:

* The actual rate information table is unfortunately indexed by rix
  and it doesn't contain a rate code.  So the userland side of this
  currently has no way to extract out a mapping.

TODO:

* Add a TLV payload to dump out the rate control table mapping so
  'rix' can be turned into a dot11 / MCS rate.
* .. then remove the temporary copy.
2012-07-20 01:36:02 +00:00
Adrian Chadd
9e38f70840 Create an ioctl API for fetching the current rate control information. 2012-07-20 01:27:20 +00:00
Adrian Chadd
dd9f5bba52 Prepare for (re)using this header file in userland.
Remove the inlined code from the header file if it's compiled in userland.
It's not required and it shouldn't be there in the first place.
2012-07-20 00:47:23 +00:00
Konstantin Belousov
eb3d975443 Remove line which was accidentally kept in r238614.
Submitted by:	pjd
Pointy hat to:	kib
MFC after:	1 week
2012-07-19 20:38:03 +00:00
Konstantin Belousov
83b22b05e6 Introduce curpcb magic variable, similar to curthread, which is MD
amd64.  It is implemented as __pure2 inline with non-volatile asm read
from pcpu, which allows a compiler to cache its results.

Convert most PCPU_GET(pcb) and curthread->td_pcb accesses into curpcb.

Note that __curthread() uses magic value 0 as an offsetof(struct pcpu,
pc_curthread). It seems to be done this way due to machine/pcpu.h
needs to be processed before sys/pcpu.h, because machine/pcpu.h
contributes machine-depended fields to the struct pcpu definition. As
result, machine/pcpu.h cannot use struct pcpu yet.

The __curpcb() also uses a magic constant instead of offsetof(struct
pcpu, pc_curpcb) for the same reason. The constants are now defined as
symbols and CTASSERTs are added to ensure that future KBI changes do
not break the code.

Requested and reviewed by: bde
MFC after:    3 weeks
2012-07-19 19:09:12 +00:00
Marius Strobl
6d36859855 Revert the use of BUS_DMA_ALLOCNOW when creating the DMA tag for user
data introduced in r236061. Using that flag doesn't make that much
sense on this case as the DMA maps using it are also created during
sym_pci_attach(). Moreover, due to the maxsegsz parameter used, doing
so may exhaust the bounce pages pool on architectures requiring
bounce pages. [1]
While at it, use a slightly more appropriate maxsegsz parameter.

PR:		169526
Submitted by:	Mike Watters [1]
MFC after:	3 days
2012-07-19 14:43:46 +00:00
Konstantin Belousov
d1ae5c8337 Fix several reads beyond the mapped first page of the binary in the
ELF parser. Specifically, do not allow note reader and interpreter
path comparision in the brandelf code to read past end of the page.
This may happen if specially crafter ELF image is activated.

Submitted by:	Lukasz Wojcik <lukasz.wojcik zoho com>
MFC after:	3 days
2012-07-19 11:15:53 +00:00
Konstantin Belousov
49d02b13bc Implement F_DUPFD_CLOEXEC command for fcntl(2), specified by SUSv4.
PR:	  standards/169962
Submitted by:	Jukka A. Ukkonen <jau iki fi>
MFC after:	1 week
2012-07-19 10:22:54 +00:00
Alan Cox
3088e08c4b Don't unnecessarily set PGA_REFERENCED in pmap_enter(). 2012-07-19 05:34:19 +00:00
Adrian Chadd
bb06995571 Convert the TX path to use the new HAL methods for accessing the
TX descriptor link pointers.

This is required for the AR93xx and later chipsets.

The RX path is slightly different - the legacy RX path directly
accesses ath_desc->ds_link for now, however this isn't at all done
for EDMA (FIFO) RX.

Now, for those performing a little software archeology here:

This is all a bit sub-optimal. "struct ath_desc" is only really relevant
for the pre-AR93xx NICs - where ds_link and ds_data is always in the
same location.

The AR93xx and later NICs have different descriptor layouts altogether.

Now, for AR93xx and later NICs, you should never directly reference
ds_link and ds_data, as:

* the RX descriptors don't have either - the data is _after_ the RX
  descriptor.  They're just one large buffer.  There's also no need for
  a per-descriptor RX buffer size as they're all fixed sizes.

* the TX descriptors have 4 buffer and 4 length fields _and_ a link
  pointer.  Each frame takes up one TX FIFO pointer, but it can contain
  multiple subframes (either multiple frames in a buffer, and/or
  multiple frames in an aggregate/RIFS burst.)

* .. so, when TX frames are queued to a hardware queue, the link
  pointer is ONLY for buffers in that frame/aggregate.  The next frame
  starts in a new FIFO pointer.

* Finally, descriptor completion status is in a different ring.
  I'll write something up about that when its time to do so.

This was inspired by Linux ath9k and the reference driver but is a
reimplementation.

Obtained from:	Linux ath9k, Qualcomm Atheros
2012-07-19 03:51:16 +00:00
Adrian Chadd
661deb68d5 Use HAL_NUM_RX_QUEUES rather than a magic constant. 2012-07-19 03:18:15 +00:00
Adrian Chadd
ad3e6dcd37 Break out the TX descriptor link field into HAL methods.
The DMA FIFO chips (AR93xx and later) differ slightly to th elegacy
chips:

* The RX DMA descriptors don't have a ds_link field;
* The TX DMA descriptors have a ds_link field however at a different
  offset.

This is a reimplementation based on what the reference driver and ath9k
does.

A subsequent commit will enable it in the TX and beacon paths.

Obtained from:	Linux ath9k, Qualcomm Atheros
2012-07-19 02:25:14 +00:00
Konstantin Belousov
d4961bcb3a Do not restart scan of the inactive queue when non-inactive page is
found. Rather, we shall not find such pages on inactive queue at all.

Requested and reviewed by:	alc
MFC after:    2 weeks
2012-07-18 21:47:50 +00:00
Sean Bruno
ce61627a2d On BIO_ERROR, set bio_resid to stop losing data in the error case.
Submitted by:	Mark Johnston <markjdb@gmail.com>
Reviewed by:	scottl@freebsd.org
MFC after:	2 weeks
2012-07-18 18:10:27 +00:00
Konstantin Belousov
bc84db6267 On AMD64, provide siginfo.si_code for floating point errors when error
occurs using the SSE math processor.  Update comments describing the
handling of the exception status bits in coprocessors control words.

Remove GET_FPU_CW and GET_FPU_SW macros which were used only once.
Prefer to use curpcb to access pcb_save over the longer path of
referencing pcb through the thread structure.

Based on the submission by:	Ed Alley <wea llnl gov>
PR:	  amd64/169927
Reviewed by:	bde
MFC after:	3 weeks
2012-07-18 15:43:47 +00:00
Konstantin Belousov
a81f9fed5d Add stmxcsr.
Submitted by:	Ed Alley <wea llnl gov>
PR:	  amd64/169927
MFC after:	3 weeks
2012-07-18 15:36:03 +00:00
Alexander Motin
9eb97fed6b Fix some typos in r238595.
Reported by:	brueffer
2012-07-18 12:41:09 +00:00
Alexander Motin
cdcb782817 Add bunch of new ASC/ASCQ values from T10 site. 2012-07-18 12:23:45 +00:00
Christian Brueffer
1d50dd1f08 Fix a small memory leak in mpssas_get_sata_identify(). The change has been
submitted upstream as well.

Reviewed by:	ken, scottl
Obtained from:	DragonFly BSD (change df8658e030226dd015cff9749452666d8fe1e87b)
MFC after:	5 days
2012-07-18 09:06:07 +00:00
Gleb Smirnoff
3c73180f55 Plug a reference leak: before doing 'goto again' we need to unref
ia->ia_ifa if there is any.

Submitted by:	Andrey Zonov <andrey zonov.org>
2012-07-18 08:58:30 +00:00
Gleb Smirnoff
b9abeb9d99 When traversing global in_ifaddr list in the IFP_TO_IA() macro, we need
to obtain IN_IFADDR_RLOCK().
2012-07-18 08:41:00 +00:00
Edward Tomasz Napierala
3a3ef28e15 Add missing free. 2012-07-18 07:26:20 +00:00
Alan Cox
85eeca35b9 Move what remains of vm/vm_contig.c into vm/vm_pageout.c, where similar
code resides.  Rename vm_contig_grow_cache() to vm_pageout_grow_cache().

Reviewed by:	kib
2012-07-18 05:21:34 +00:00
Kenneth D. Merry
edad9799e8 Add back spare fields consumed in r237545. It seems that these should only
be consumed to maintain backward compatibility in stable, but should not be
consumed in head.

Submitted by:	trasz, attilio (indirectly)
2012-07-17 22:16:10 +00:00
George V. Neville-Neil
d638e8dcde Change UL to ULL since time is 32 bits.
Pointed out by: avg@
MFC after:	2 weeks
2012-07-17 14:36:40 +00:00
Alexander Motin
f4f6d5e0ab For Intel Panther/Lynx Point USB 3.0 xHCI controllers enable SuperSpeed USB
capability and reroute USB 2.0 ports to the xHCI controller.

Reviewed by:	hselasky
2012-07-17 14:03:04 +00:00
Michael Tuexen
d07b2ac6c6 Fix a refcount bug when freeing an association.
While there: Change code to be consistent.
Discussed with rrs@.
MFC after: 3 days
2012-07-17 13:03:47 +00:00
Oleksandr Tymoshenko
86bce74937 Move unmask IRQ function call up to nexus device level.
FDT-enabled targets were broken after r238043 that relies
on device up the hierarchy to properly setup interrupt.
nexus device for ARM platforms did job only partially:
setting handler but not unmasking interrupt. Unmasking
was performed by platform code.

Reviewed by:	andrew@
2012-07-17 03:18:12 +00:00
Alan Cox
da1ab8a4a0 Correct vm_page_alloc_contig()'s implementation of VM_ALLOC_NODUMP. 2012-07-17 02:36:59 +00:00
Christian Brueffer
01cc0b6531 Simply error handling by moving the allocation of np down to where it is
actually used.  While here, improve style a little.

Submitted by:	mjg
MFC after:	2 weeks
2012-07-16 22:07:29 +00:00
George V. Neville-Neil
57d025c338 Add support for walltimestamp in DTrace.
Submitted by:	Fabian Keil
MFC after:	2 weeks
2012-07-16 20:17:19 +00:00
Alan Cox
907e4524dc Various improvements to vm_contig_grow_cache(). Most notably, even when
it can't sleep, it can still move clean pages from the inactive queue to
the cache.  Also, when a page is cached, there is no need to restart the
scan.  The "next" page pointer held by vm_contig_launder() is still
valid.  Finally, add a comment summarizing what vm_contig_grow_cache()
does based upon the value of "tries".

MFC after:	3 weeks
2012-07-16 18:13:43 +00:00
Edward Tomasz Napierala
9e9d445ed1 The resize GEOM event has no references, thus cannot be canceled. 2012-07-16 17:41:38 +00:00
Edward Tomasz Napierala
8fe7677998 Add back spare fields reused in r238213. According to Attilio, the rule
is to use reuse spares only when MFC-ing, not in CURRENT.
2012-07-16 16:50:28 +00:00
Hans Petter Selasky
c6b5a9a02b Add new USB device ID.
PR:		usb/169789
Submitted by:	Ruslan Bukin
MFC after:	1 week
2012-07-16 10:12:07 +00:00