projects / sbcl.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: a3caf31) | patch
micro-optimize allocation sequences, special variable binding on x86-64
authorNathan Froyd <froydnj@gmail.com>
Fri, 13 Apr 2012 17:58:41 +0000 (13:58 -0400)
committerNathan Froyd <froydnj@gmail.com>
Fri, 13 Apr 2012 17:58:41 +0000 (13:58 -0400)
commit1dd5275c69e1260d0a049052b99938fb043355fd
treede8d70675a188c4d4b4c0deb0d8e70f4415753edtree | snapshot
parenta3caf3155958590af9614705770358c0c8bdd8a8commit | diff
micro-optimize allocation sequences, special variable binding on x86-64

Move the ALLOC-REGION, PSEUDO-ATOMIC-BITS, and BINDING-STACK-* slots
closer to the beginning of the thread structure.  This change ensures
that the offsets for those slots are < 128 bytes, which in turns enables
shorter encodings for all accesses to this structure from Lisp code.

Code size of the C runtime was negligibly affected by this change.
src/compiler/generic/objdef.lisp diff | blob | history
Mirror of Steel Bank Common Lisp SBCL's official repository