Commit Graph

19 Commits

Author SHA1 Message Date
Konstantin Belousov
27007c6576 Put the aesni_cipher_setup() and aesni_cipher_process() functions into
the file which is compiled with SSE disabled.  The functions set up
the FPU context for kernel, and compiler optimizations which could
lead to use of XMM registers before the fpu_kern_enter(9) is called or
after fpu_kern_leave(9), panic the machine.

Discussed with:	jmg
Sponsored by:	The FreeBSD Foundation
MFC after:	1 week
2014-06-24 06:55:49 +00:00
Konstantin Belousov
633034fe0e Add FPU_KERN_KTHR flag to fpu_kern_enter(9), which avoids saving FPU
context into memory for the kernel threads which called
fpu_kern_thread(9).  This allows the fpu_kern_enter() callers to not
check for is_fpu_kern_thread() to get the optimization.

Apply the flag to padlock(4) and aesni(4).  In aesni_cipher_process(),
do not leak FPU context state on error.

Sponsored by:	The FreeBSD Foundation
MFC after:	1 week
2014-06-23 07:37:54 +00:00
John-Mark Gurney
9164a239ea fix broken style(9) in r258399
Pointed out by:	brd
2013-11-23 00:28:18 +00:00
John-Mark Gurney
23fc6e1652 flag that the aesni driver is sync... This means we don't waste a
context switch just to call the done callback...  On my machine, this
improves geli/gzero decrypt performance by ~27% from 550MB/sec to
~700MB/sec...

MFC after:	3 days
2013-11-20 20:25:27 +00:00
John-Mark Gurney
038ffd3e43 make it so that from/to can be missaligned as it can happen (the geli
regression manages to do it)...  We use a packed struct to coerce
gcc/clang into producing unaligned loads (there is not packed pointer
attribute, otherwise this would be easier)...

use _storeu_ and _loadu_ when using the structure is overkill...

be better at using types properly...  Since we allocate our own key
schedule and make sure it's aligned, use the __m128i type in various
arguments to functions...

clang ignores __aligned on prototypes and gcc errors on them, leave them
in comments to document that these function arguments are require to be
aligned...

about all that changes is movdqa -> movdqu from reading the diff of the
disassembly output...

Noticed by:	symbolics at gmx.com
MFC after:	3 days
2013-11-06 19:14:49 +00:00
John-Mark Gurney
ff6c7bf5ca Use the fact that the AES-NI instructions can be pipelined to improve
performance... Use SSE2 instructions for calculating the XTS tweek
factor...  Let the compiler do more work and handle register allocation
by using intrinsics, now only the key schedule is in assembly...

Replace .byte hard coded instructions w/ the proper instructions now
that both clang and gcc support them...

On my machine, pulling the code to userland I saw performance go from
~150MB/sec to 2GB/sec in XTS mode.  GELI on GNOP saw a more modest
increase of about 3x due to other system overhead (geom and
opencrypto)...

These changes allow almost full disk io rate w/ geli...

Reviewed by:	-current, -security
Thanks to:	Mike Hamburg for the XTS tweek algorithm
2013-09-03 18:31:23 +00:00
Pawel Jakub Dawidek
45b56a6ba2 When porting XTS-related code from OpenBSD I forgot to update copyright (only
OpenBSD was credited in one of two commits). Fix it.

Reported by:	Theo de Raadt <deraadt@cvs.openbsd.org>
Reviewed by:	Damien Miller <djm@mindrot.org>
2013-02-20 22:59:53 +00:00
Konstantin Belousov
8c6f8f3d5b Add support for the extended FPU states on amd64, both for native
64bit and 32bit ABIs.  As a side-effect, it enables AVX on capable
CPUs.

In particular:

- Query the CPU support for XSAVE, list of the supported extensions
  and the required size of FPU save area. The hw.use_xsave tunable is
  provided for disabling XSAVE, and hw.xsave_mask may be used to
  select the enabled extensions.

- Remove the FPU save area from PCB and dynamically allocate the
  (run-time sized) user save area on the top of the kernel stack,
  right above the PCB. Reorganize the thread0 PCB initialization to
  postpone it after BSP is queried for save area size.

- The dumppcb, stoppcbs and susppcbs now do not carry the FPU state as
  well. FPU state is only useful for suspend, where it is saved in
  dynamically allocated suspfpusave area.

- Use XSAVE and XRSTOR to save/restore FPU state, if supported and
  enabled.

- Define new mcontext_t flag _MC_HASFPXSTATE, indicating that
  mcontext_t has a valid pointer to out-of-struct extended FPU
  state. Signal handlers are supplied with stack-allocated fpu
  state. The sigreturn(2) and setcontext(2) syscall honour the flag,
  allowing the signal handlers to inspect and manipilate extended
  state in the interrupted context.

- The getcontext(2) never returns extended state, since there is no
  place in the fixed-sized mcontext_t to place variable-sized save
  area. And, since mcontext_t is embedded into ucontext_t, makes it
  impossible to fix in a reasonable way.  Instead of extending
  getcontext(2) syscall, provide a sysarch(2) facility to query
  extended FPU state.

- Add ptrace(2) support for getting and setting extended state; while
  there, implement missed PT_I386_{GET,SET}XMMREGS for 32bit binaries.

- Change fpu_kern KPI to not expose struct fpu_kern_ctx layout to
  consumers, making it opaque. Internally, struct fpu_kern_ctx now
  contains a space for the extended state. Convert in-kernel consumers
  of fpu_kern KPI both on i386 and amd64.

First version of the support for AVX was submitted by Tim Bird
<tim.bird am sony com> on behalf of Sony. This version was written
from scratch.

Tested by:	pho (previous version), Yamagi Burmeister <lists yamagi org>
MFC after:	1 month
2012-01-21 17:45:27 +00:00
Pawel Jakub Dawidek
14a0d24607 Update Copyright.
MFC after:	3 days
2011-10-27 14:15:26 +00:00
Pawel Jakub Dawidek
5fa1b35081 Improve AES-NI performance for AES-XTS:
- Operate on uint64_t types when doing XORing, etc. instead of uint8_t.
- Don't bzero() temporary block for every AES block. Do it once for entire
  data block.
- AES-NI is available only on little endian architectures. Simplify code
  that takes block number from IV.

Benchmarks:

Memory-backed md(4) device, software AES-XTS, 4kB sector:

	# dd if=/dev/md0.eli bs=1m
	59.61MB/s

Memory-backed md(4) device, old AES-NI AES-XTS, 4kB sector:

	# dd if=/dev/md0.eli bs=1m
	97.29MB/s

Memory-backed md(4) device, new AES-NI AES-XTS, 4kB sector:

	# dd if=/dev/md0.eli bs=1m
	221.26MB/s

127% performance improvement between old and new code.

Harddisk, raw speed:

	# dd if=/dev/ada0 bs=1m
	137.63MB/s

Harddisk, software AES-XTS, 4kB sector:

	# dd if=/dev/ada0.eli bs=1m
	47.83MB/s (34% of raw disk speed)

Harddisk, old AES-NI AES-XTS, 4kB sector:

	# dd if=/dev/ada0.eli bs=1m
	68.33MB/s (49% of raw disk speed)

Harddisk, new AES-NI AES-XTS, 4kB sector:

	# dd if=/dev/ada0.eli bs=1m
	108.35MB/s (78% of raw disk speed)

58% performance improvement between old and new code.

As a side-note, GELI with AES-NI using AES-CBC can achive native disk speed.

MFC after:	3 days
2011-10-27 14:07:57 +00:00
Konstantin Belousov
93ed70f9b4 Fix a bug in the result of manual assembly.
Reported by:	Stefan Grundmann <sg2342 googlemail com>
PR:	kern/155118
MFC after:	3 days
2011-03-02 14:56:58 +00:00
Konstantin Belousov
17ca6d98e1 Remove DEBUG sections.
MFC after:	3 days
2010-11-27 15:41:44 +00:00
Konstantin Belousov
1563ee36b9 Only save FPU context when not executing in the context of the crypto
thread.

Tested by:	Mike Tancsa
2010-11-17 16:17:15 +00:00
Dimitry Andric
235610273e Change two missed instances of 'retq' in aeskeys_i386.S to 'retl', which
makes it possible to assemble this file with gas from newer binutils.

Reviewed by:	kib
2010-10-13 17:55:53 +00:00
Pawel Jakub Dawidek
c979bd5e06 Fix two copy&paste bugs.
MFC after:	2 weeks
2010-09-25 10:32:52 +00:00
Pawel Jakub Dawidek
ac970319ff Add support for CRYPTO_AES_XTS.
MFC after:	1 week
2010-09-23 11:57:25 +00:00
Pawel Jakub Dawidek
30bd3bb07b Add support for CRD_F_KEY_EXPLICIT flag.
MFC after:	1 week
2010-09-23 11:46:53 +00:00
Pawel Jakub Dawidek
745eb0ccb6 Simplify code a bit.
MFC after:	1 week
2010-09-23 11:26:38 +00:00
Konstantin Belousov
5f270659fd Crypto(4) driver for AESNI.
The aeskeys_{amd64,i386}.S content was mostly obtained from OpenBSD,
no objections to the license from core.

Hardware provided by:	Sentex Communications
Tested by:	fabient, pho (previous versions)
MFC after:	1 month
2010-07-23 11:00:46 +00:00