Commits on Mar 24, 2002
  1. 0.7.2:

    	no real changes except making NEWS more complete
    William Harold Newman committed Mar 24, 2002
Commits on Mar 23, 2002
  1. @telent
  2. @telent
Commits on Mar 21, 2002
  1. @csrhodes

    	Compile the runtime on SPARC/Linux with -O3.
    csrhodes committed Mar 21, 2002
Commits on Mar 20, 2002
  1. @telent

    	Make UNIX-NAMESTRING merge pathnames with
    	*default-pathname-defaults* to get rid of ugly special case for
    	"", and hopefully to fix all the remaining bits of code that
    	don't do pathname merging before they access the filesystem
    telent committed Mar 20, 2002
Commits on Mar 19, 2002

    	reviewed OPTIMIZE declarations, (mostly trying to avoid
    		inadvertently imposing large %DETECT-STACK-EXHAUSTION
    	...bumped SAFETY 1 to SAFETY 2 in warm.lisp (since
    		(> SPEED SAFETY) isn't what we want)
    	...changed (SPEED 3) (SAFETY 3) to (SPEED 2) (SAFETY 2) in
    		SORT logic
    	...removed OPTIMIZE declarations in time.lisp, since there's
    		no up-to-date compelling reason to use other than the
    		default optimization there.
    		since there seems to be no reason for it
    	rearranged insert-%DETECT-STACK-EXHAUSTION code to make
    		indentation less ridiculous
    William Harold Newman committed Mar 19, 2002
Commits on Mar 18, 2002
  1. @csrhodes

    	Merged patches from CSR "pseudo-atomic magic constant propagation"
    	sbcl-devel 2002-03-12
    	... first patch for PSEUDO-ATOMIC-TRAP propagation on SPARC
    	... second patch for various #+ -> #!+ changes in library and
    		alpha port
    	Also other SPARC cleanups
    	... implemented suggestion from DB for ldso-stubs.S, allowing
    		the SPARC port to be dynamically linked
    	... fprintf -> FSHOW
    csrhodes committed Mar 18, 2002
  2. @telent

    need these too

    telent committed Mar 18, 2002
  3. @telent

    Merge PPC port

    	... new directories src/compiler/ppc, src/assembly/ppc
    	... other new files
    	... new clause in genesis for PPC fixups
    	... new files in runtime, PPC conditionals added in other .[ch] files
    	Small Makefile cleanups in runtime
    	... actually _use_ the dependency information
    	... regenerate depends on source changes
    	We don't actually use sigreturn() in _any_ present port: conditionals
    	changed to make this obvious
    telent committed Mar 18, 2002

    	minor comment fixes which want to be synchronized in CVS (the
    		BUGS change because bugs have incrementing numbers --
    		which will become a nuisance with multiple committers,
    		dunno what the long term fix is -- and the comments because DB and CSR have
    		been fixing this on sbcl-internals and if I don't
    		commit my corresponding fix soon I'll find it collides
    		with theirs)
    William Harold Newman committed Mar 18, 2002
Commits on Mar 17, 2002
  1. @telent

    	In OPEN, the pathname is now merged against
    	*default-pathname-defaults* before access to the filesystem.
    	Failure to do this breaks OPEN on relative pathnames when
    	*default-pathname-defaults* is not the Unix cwd
    telent committed Mar 17, 2002
Commits on Mar 16, 2002
  1. @csrhodes

    	Merged DB "alpha floating point traps and infinities" patch
    	sbcl-devel 2002-03-14
    	... added a comment about the commentary nature of the x86/linux
    		implementation of os_context_fp_control
    	... moved the "interpreter" stubs back to src/code/float.lisp,
    		protected by #!-alpha
    csrhodes committed Mar 16, 2002

    	more ordinary name (one *FEATURES* symbol, not two)
    		for the opposed pair of stack-grows-upward and
    		stack-grows-downward concepts
    	...replaced old pair of features with new
    	...#!+STACK-GROWS-DOWNWARD becomes
    	...#!+STACK-GROWS-UPWARD becomes
    	Since references to the new STACK-EXHAUSTION stuff can be
    		automatically written into fasl files, the fasl file
    		version needs to be incremented.
    William Harold Newman committed Mar 16, 2002

    	more-explicit names for new stack stuff (thanks to Dan Barlow
    		sbcl-devel a week or so ago for pointing out
    		unnoticed ambiguity)...
    	...mostly s/stack-exhaustion/control-stack-exhaustion/
    	related name clarification in old code...
    	...s/cstack/control-stack/ (since evidently the C stack is, on
    		non-X86en, distinct from the Lisp control stack)
    William Harold Newman committed Mar 16, 2002
  4. @csrhodes

    	(on the basis that the first checkin should probably be something
    	simple and straightforward):
    	Fix for wrapped_readlink, allowing TRUENAME to return sensible
    	results on SPARC.
    csrhodes committed Mar 16, 2002
Commits on Mar 14, 2002

    	catching stack overflow, part III...
    	...redid *STACK-EXHAUSTION* value as SAP, because although
    		FIXNUM is cute and should be implementable efficiently,
    		it seems to be awfully annoying to work with
    	...made %DETECT-STACK-EXHAUSTION actually check for the
    	...set up machinery to try to handle the problem reasonably
    	...added basic regression test
    		EXPORT of '*SHEBANG-FEATURES* so that chill.lisp
    		works again
    William Harold Newman committed Mar 14, 2002
Commits on Mar 12, 2002

    	(oops, not really a version at all, just a vehicle for the
    		comment that the previous commit should've had:)
    	merged CSR "format ~/ /" patch sbcl-devel 2002-03-12
    William Harold Newman committed Mar 12, 2002

    	merged CSR
    William Harold Newman committed Mar 12, 2002
Commits on Mar 9, 2002

    	merged Robert Strandh "patch for format-dollars" (sbcl-devel
    William Harold Newman committed Mar 9, 2002

    	responding to CSR "src/compiler/main.lisp" in sbcl-devel
    	...fixed argument list for COMPILE-LOAD-TIME-STUFF (broken in
    	...added some test cases for MAKE-LOAD-FORM (to catch problems
    		like this in the future)
    	split the MAKE-LOAD-FORM tests, and some other tests, into
    		new tests/dump.impure-cload.lisp
    William Harold Newman committed Mar 9, 2002
Commits on Mar 8, 2002

    	merged patch: CSR "mostly types" (sbcl-devel 2002-03-07)
    William Harold Newman committed Mar 8, 2002
Commits on Mar 7, 2002

    	merged APD "bug 152" patch sbcl-devel 2002-03-06
    William Harold Newman committed Mar 7, 2002

    	merged MNA cleanups patch (sbcl-devel 2002-03-05)...
    	...correct FTYPE for SAPINT-TO-CORE, which has been
    		renamed to SAP-INT-TO-CORE
    	...dead code removal
    	...*TARGET-MOST-NEGATIVE-FIXNUM*" is renamed to
    		+TARGET-MOST-NEGATIVE-FIXNUM+, as well as being defined
    		as a constant now (same for the POSITIVE).
    	...SB!KERNEL::ARG-COUNT-ERROR is defined as a stub, to silence
    		the flood of style-warnings a little bit.
    	...some other FIXMEs (renaming DEFINE-STORAGE-CLASSES to
    	further cleanups of MOST-FOOATIVE-FIXNUM stuff...
    	...renamed again, this time to SB!XC:MOST-FOOATIVE-FIXNUM
    	...removed now-redundant DEFCONSTANTs in toplevel.lisp
    	...removed now-redundant #. wrappers
    	other tweaks to the patch...
    	...Instead of defining a second version of ARG-COUNT-ERROR
    		for use at compile time, move the definition of
    		ARG-COUNT-ERROR earlier so it'll be visible in more
    		of the code which uses it.
    	moved src/code/globals.lisp much earlier in build-order.lisp,
    		since there's no such thing as too early and since the
    		previous location was too late for some uses of
    William Harold Newman committed Mar 7, 2002
Commits on Mar 5, 2002

    	belatedly deleted BUGS entry 150
    William Harold Newman committed Mar 5, 2002

    	merged APD bug 150 patch sbcl-devel 2002-03-05
    William Harold Newman committed Mar 5, 2002

    	(Note that the STACK-FOO symbols introduced in this version
    		probably need to change, since as DB pointed out
    		on sbcl-devel 2002-03-04, they assume an identity
    		between C stack and SBCL control stack which doesn't
    		exist on the non-X86 ports.)
    	catching stack overflow, part II...
    	...defined *STACK-EXHAUSTION* and arranged for it to be
    		initialized in cold init
    	...added STACK-GROWS-FOOWARD to *FEATURES*, and used it to
    		reexpress some of the old #!+X86 conditionalization
    	merged APD "unbound condition slot" patch (sbcl-devel
    		2002-03-04), so that PARSE-DEFMACRO creates
    		ARG-COUNT-ERROR conditions correctly in the
    		(NOT (EQ ERROR-FUN 'ERROR)) case
    	made parse-defmacro-errors.lisp use pretty-printer
    	added SB-DEBUG::*NESTED-DEBUG-CONDITION* to succor those of us
    		who like to divert ourselves by trashing print methods
    	bumped *DEBUG-PRINT-LEVEL* and *DEBUG-PRINT-LENGTH* so that
    		the new highly-structured pretty-printing :REPORT
    		methods for parse-defmacro-errors.lisp don't truncate
    		output too ludicrously
    William Harold Newman committed Mar 5, 2002
Commits on Mar 2, 2002

    	merged the third (of 3) patches from CSR "x86, format, types"
    		patches (sbcl-devel 2002-02-27)
    	worried about the special case of (SPECIFIER-TYPE T), tried to
    		protect the patched code from it
    William Harold Newman committed Mar 2, 2002
Commits on Mar 1, 2002

    	merged the first 2 (of 3) patches from CSR "x86, format, types"
    		patches (sbcl-devel 2002-02-27)
    William Harold Newman committed Mar 1, 2002
Commits on Feb 27, 2002

    	merged CSR backend-subfeatures patch (sbcl-devel 2002-02-22)
    William Harold Newman committed Feb 27, 2002
Commits on Feb 26, 2002
  1. *** empty log message ***

    William Harold Newman committed Feb 26, 2002
Commits on Feb 22, 2002

    	merged CSR "build patch" 2002-02-21
    William Harold Newman committed Feb 22, 2002
Commits on Feb 21, 2002

    	catching stack overflow, part I...
    	...defined placeholder %DETECT-STACK-EXHAUSTION
    	...arranged for it to be called on entry to lambdas when
    	...downgraded SAFETY to 2 (= SPEED) in SBCL's own code, since
    		SBCL's own code seems to be at little risk of infinite
    William Harold Newman committed Feb 21, 2002
Commits on Feb 20, 2002

    	merged CSR patch I and II: "CATCH and gc" 2002-02-18
    	merged CSR patch III: "lowtag horribleness" 2002-02-19
    William Harold Newman committed Feb 20, 2002
Commits on Feb 19, 2002

    	added CSR test code from "precedence/sequence gotchas" message
    		sbcl-devel 2002-02-13 (but didn't try to figure out,
    		much less fix, the problems it complains about, just
    		wrote a FIXME instead)
    	bumped default #-GENCGC BYTES-CONSED-BETWEEN-GCS to 20M
    William Harold Newman committed Feb 19, 2002
Commits on Feb 15, 2002

    	merged CSR SPARC port patch (sbcl-devel 2002-02-12, plus
    		lotso new files through FTP)
    	tweaking patch...
    	...updated sbcl.1
    	...added 'typedef os_context_register' for OpenBSD
    	...added ARCH_HAS_STACK_POINTER for x86 (and added
    		x86-arch.h to hold it)
    	...renamed fpregister and fp.register to float.register (to
    		avoid the ambiguity with "frame pointer" that I
    		experienced when first trying to figure this out,
    		since even though for a given architecture it's pretty
    		unambiguous, in architecture-neutral code it's not)
    	...added x86-bsd-os.h
    William Harold Newman committed Feb 15, 2002