summaryrefslogtreecommitdiffstats
path: root/sys/i386/include/atomic.h
Commit message (Expand)AuthorAgeFilesLines
* Implement atomic_cmpset_64() and atomic_swap_64() for i386.jkim2013-08-211-0/+113
* Reimplement atomic_load_acq_64() and atomic_store_rel_64() for i386. Thesejkim2013-08-211-82/+110
* Remove empty lines before return statements for style consistency.jkim2013-08-211-6/+0
* Implement atomic_swap() and atomic_testandset().jkim2013-08-211-19/+43
* - Remove the "a" constraint from main output operand for atomic_cmpset().jkim2013-08-211-5/+5
* Use '+' modifier for a memory operand that is both an input and an output.jkim2013-08-211-29/+21
* Remove bogus labels. No functional change.jkim2013-08-211-1/+0
* Use consistent style. No functional change.jkim2013-08-211-10/+10
* Add an unified macro to deny ability from the compiler to reorderattilio2012-10-091-2/+2
* Use plain store for atomic_store_rel on x86, instead of implicitlykib2012-06-021-39/+39
* Implement atomic_load_acq_64(9) and atomic_store_rel_64(9) for i386. Thesejkim2011-04-061-0/+86
* Inform a compiler which asm statements in the x86 implementation ofkib2010-12-181-6/+7
* Rename an argument from "exp" to "expect" since the former makes FlexeLintphk2010-05-201-8/+8
* atomic_cmpset_barr_* was added in order to cope with compilers willing toattilio2009-10-091-64/+48
* - All the functions in atomic.h needs to be in "physical" form (likeattilio2009-10-061-45/+50
* Per their definition, atomic instructions used in conjuction withattilio2009-10-061-46/+73
* Restore memory clobber, to cause mb on the compiler level too.kib2008-12-061-3/+3
* Unconditionally use locked addition of zero to tip of the stack forkib2008-12-051-14/+3
* - bump __FreeBSD version to reflect added buf_ring, memory barriers,kmacy2008-11-221-0/+15
* Implement atomic_fetchadd_long() for all architectures and document it.pjd2008-03-161-0/+7
* Fix oops in previous commit.bde2006-12-291-1/+1
* Fixed some style bugs (mainly assorted errors in comments, and inconsistentbde2006-12-291-26/+26
* Fixed some style bugs (whitespace only).bde2006-12-291-31/+34
* Try harder to garbage-collect the "LOCORE" (really asm) version ofbde2006-12-291-5/+5
* Avoid an instruction in atomic_cmpset_{int_long)() in most cases.bde2006-12-271-11/+11
* Use wrapper macros for atomic pointer operations in order to perform thedes2006-03-281-18/+36
* Add a new atomic_fetchadd() primitive that atomically adds a value to ajhb2005-09-271-0/+21
* Stop using the '+' constraint modifier with inline assembly. The '+'jhb2005-09-151-22/+27
* Convert the atomic_ptr() operations over to operating on uintptr_tjhb2005-07-151-52/+18
* Some cleanups and tweaks to some of the atomic.h files in preparation forjhb2005-07-091-83/+72
* netchild's mega-patch to isolate compiler dependencies into a centraljoerg2005-03-021-11/+15
* Initiate deorbit burn sequence for 80386 support in FreeBSD: Removejhb2004-11-161-3/+3
* Spell _KERNEL correctly so that UP kernels are actually optimized again.jhb2004-11-121-1/+1
* - Use the SMP style ops for atomic_load/store() in userland so thatjhb2004-11-121-2/+1
* - Place the gcc memory barrier hint in the right place in the 80386 versionjhb2004-11-111-8/+9
* These are changes to allow to use the Intel C/C++ compiler (lang/icc)trhodes2004-03-121-11/+11
* Fixed pedantic syntax errors. Many macros didn't permit a semicolon afterbde2003-11-171-3/+6
* Avoid a warning for compiling with `gcc -Wbad-function cast'. (Thisbde2003-11-171-1/+5
* Add a knob to turn on and off the CMPXCHG instruction on > i386 IA32 systems.pirzyk2002-10-141-1/+1
* Beautify. This has the side effect of improving portability andmarkm2002-07-181-39/+31
* Clean up the syntax WRT semicolons at the end of function-like-macros, and pr...markm2002-07-171-23/+74
* Make MPLOCKED work again in asm files and stringify it explicitlybmilekic2002-02-281-4/+4
* Garbage-collect the "LOCORE" version of MPLOCKED.bde2002-02-111-10/+2
* Allow the ATOMIC_ASM() macro to pass in the constraints on the V parameterjhb2001-12-181-23/+23
* Use newer constraints for atomic_cmpset().jhb2001-11-121-11/+9
* Use newer constraints for inline assembly for an operand that is both anjhb2001-11-121-21/+21
* Allow atomic ops to be somewhat safely used in userland. We always usejhb2001-10-081-3/+7
* RIP <machine/lock.h>.markm2001-02-111-4/+9
* - Sort of lie and say that %eax is an output only and not an input for thejhb2001-01-171-2/+2
* - Fix atomic_load_* and atomic_store_* to generate functions for atomic.cjhb2001-01-161-168/+123
OpenPOWER on IntegriCloud