| Commit message (Expand) | Author | Age | Files | Lines |
* | MFC r311169,r311898,r312925,r312973,r312975,r313007,r313040,r313080, | mjg | 2017-03-16 | 1 | -0/+51 |
* | atomic: Add testandclear on i386/amd64 | sephe | 2016-05-16 | 1 | -0/+38 |
* | Add missing atomic wrapper macro. | hselasky | 2016-01-21 | 1 | -0/+1 |
* | Improve comments. | kib | 2015-07-30 | 1 | -4/+4 |
* | Remove full barrier from the amd64 atomic_load_acq_*(). Strong | kib | 2015-07-28 | 1 | -17/+7 |
* | Add a comment discussing the appropriate use of the atomic_*() functions | alc | 2015-07-24 | 1 | -0/+19 |
* | Add the atomic_thread_fence() family of functions with intent to | kib | 2015-07-08 | 1 | -0/+32 |
* | Add a comment about too strong semantic of atomic_load_acq() on x86. | kib | 2015-06-29 | 1 | -0/+9 |
* | Remove unneeded data dependency, currently imposed by | kib | 2015-06-28 | 1 | -50/+75 |
* | Remove empty lines before return statements for style consistency. | jkim | 2013-08-21 | 1 | -5/+0 |
* | Implement atomic_swap() and atomic_testandset(). | jkim | 2013-08-21 | 1 | -15/+57 |
* | - Remove the "a" constraint from main output operand for atomic_cmpset(). | jkim | 2013-08-21 | 1 | -10/+10 |
* | Use '+' modifier for a memory operand that is both an input and an output. | jkim | 2013-08-21 | 1 | -23/+16 |
* | Remove bogus labels. No functional change. | jkim | 2013-08-21 | 1 | -2/+0 |
* | Use consistent style. No functional change. | jkim | 2013-08-21 | 1 | -12/+12 |
* | Add an unified macro to deny ability from the compiler to reorder | attilio | 2012-10-09 | 1 | -2/+2 |
* | Use plain store for atomic_store_rel on x86, instead of implicitly | kib | 2012-06-02 | 1 | -37/+37 |
* | Inform a compiler which asm statements in the x86 implementation of | kib | 2010-12-18 | 1 | -9/+10 |
* | Rename an argument from "exp" to "expect" since the former makes FlexeLint | phk | 2010-05-20 | 1 | -7/+7 |
* | atomic_cmpset_barr_* was added in order to cope with compilers willing to | attilio | 2009-10-09 | 1 | -32/+44 |
* | - All the functions in atomic.h needs to be in "physical" form (like | attilio | 2009-10-06 | 1 | -46/+29 |
* | Per their definition, atomic instructions used in conjuction with | attilio | 2009-10-06 | 1 | -46/+67 |
* | - bump __FreeBSD version to reflect added buf_ring, memory barriers, | kmacy | 2008-11-22 | 1 | -0/+4 |
* | Implement atomic_fetchadd_long() for all architectures and document it. | pjd | 2008-03-16 | 1 | -0/+20 |
* | Fixed some style bugs (mainly assorted errors in comments, and inconsistent | bde | 2006-12-29 | 1 | -23/+23 |
* | Fixed some style bugs (whitespace only). | bde | 2006-12-29 | 1 | -30/+31 |
* | Try harder to garbage-collect the "LOCORE" (really asm) version of | bde | 2006-12-29 | 1 | -6/+6 |
* | Removed gratuitous cosmetic differences with the i386 version. This | bde | 2006-12-28 | 1 | -7/+7 |
* | Avoid an instruction in atomic_cmpset_{int_long)() in most cases. | bde | 2006-12-27 | 1 | -10/+10 |
* | Add a new atomic_fetchadd() primitive that atomically adds a value to a | jhb | 2005-09-27 | 1 | -0/+21 |
* | Stop using the '+' constraint modifier with inline assembly. The '+' | jhb | 2005-09-15 | 1 | -22/+27 |
* | Add aliases for atomic operations on 64-bit integers just like other | jhb | 2005-08-18 | 1 | -0/+20 |
* | Like on i386, bypass lock prefix for atomic ops on !SMP kernels. | peter | 2005-07-21 | 1 | -0/+27 |
* | Convert the atomic_ptr() operations over to operating on uintptr_t | jhb | 2005-07-15 | 1 | -52/+18 |
* | Some cleanups and tweaks to some of the atomic.h files in preparation for | jhb | 2005-07-09 | 1 | -79/+63 |
* | netchild's mega-patch to isolate compiler dependencies into a central | joerg | 2005-03-02 | 1 | -11/+15 |
* | Cosmetic and/or trivial sync up with i386. | peter | 2003-11-21 | 1 | -4/+10 |
* | Initial landing of SMP support for FreeBSD/amd64. | peter | 2003-11-17 | 1 | -1/+1 |
* | Commit MD parts of a loosely functional AMD64 port. This is based on | peter | 2003-05-01 | 1 | -67/+29 |
* | Add a knob to turn on and off the CMPXCHG instruction on > i386 IA32 systems. | pirzyk | 2002-10-14 | 1 | -1/+1 |
* | Beautify. This has the side effect of improving portability and | markm | 2002-07-18 | 1 | -39/+31 |
* | Clean up the syntax WRT semicolons at the end of function-like-macros, and pr... | markm | 2002-07-17 | 1 | -23/+74 |
* | Make MPLOCKED work again in asm files and stringify it explicitly | bmilekic | 2002-02-28 | 1 | -4/+4 |
* | Garbage-collect the "LOCORE" version of MPLOCKED. | bde | 2002-02-11 | 1 | -10/+2 |
* | Allow the ATOMIC_ASM() macro to pass in the constraints on the V parameter | jhb | 2001-12-18 | 1 | -23/+23 |
* | Use newer constraints for atomic_cmpset(). | jhb | 2001-11-12 | 1 | -11/+9 |
* | Use newer constraints for inline assembly for an operand that is both an | jhb | 2001-11-12 | 1 | -21/+21 |
* | Allow atomic ops to be somewhat safely used in userland. We always use | jhb | 2001-10-08 | 1 | -3/+7 |
* | RIP <machine/lock.h>. | markm | 2001-02-11 | 1 | -4/+9 |
* | - Sort of lie and say that %eax is an output only and not an input for the | jhb | 2001-01-17 | 1 | -2/+2 |