R=r
OCL=29646
CL=29646

1. ar reports names of objects with duplicate text symbols.
2. gobuild only shows first line of error output for each failed command.
3. gobuild ignores files that begin with ascii non-alphanumeric non _.

; gobuild
$ 6g -I _obj gobuild.go
  gobuild.go:150: PackageImports: undefined
$ 6g -I _obj makefile.go
  makefile.go:102: ShellString: undefined
$ 6g -I _obj util.go
  util.go:114: syntax error near zzz
gobuild: stalemate
;

; gobuild
$ 6ar grc _obj/gobuild.a util.6 util1.6
  duplicate text symbol: util1.6 and util.6: gobuild·Build
$ 6g -I _obj gobuild.go
  gobuild.go:150: PackageImports: undefined
$ 6g -I _obj makefile.go
  makefile.go:102: ShellString: undefined
gobuild: stalemate
;

R=r
DELTA=95  (49 added, 9 deleted, 37 changed)
OCL=29625
CL=29640

bug157 and bug158

R=rsc
DELTA=59  (59 added, 0 deleted, 0 changed)
OCL=29632
CL=29634

[]int -> string is already in the document.

DELTA=7  (1 added, 1 deleted, 5 changed)
OCL=29622
CL=29631

*** Reason for rollback ***

  these semicolons are necessary only because of a bug in 6g

*** Original change description ***

add missing semicolons in example

TBR=gri
OCL=29624
CL=29624

R=r
OCL=29623
CL=29623

R=r
OCL=29619
CL=29619

R=r
DELTA=46  (46 added, 0 deleted, 0 changed)
OCL=29569
CL=29614

R=r
OCL=29612
CL=29612

The code was already careful not to use malloc/free
for stack growth during calls to malloc.
Avoid them during calls to free too.

R=r
DELTA=9  (7 added, 0 deleted, 2 changed)
OCL=29606
CL=29610

R=ken
OCL=29577
CL=29589

passes ridiculous test from CL 29569.

R=ken
OCL=29571
CL=29573

R=ken
OCL=29568
CL=29570

R=r
OCL=29544
CL=29544

fewer moves, fewer stupid LEALs.
powser1 runs (with evaln commented out).
beginnings of floating point.

R=ken
OCL=29540
CL=29543

get the code going.

R=rsc
APPROVED=rsc
DELTA=4752  (1723 added, 948 deleted, 2081 changed)
OCL=29403
CL=29530

R=rsc
APPROVED=rsc
DELTA=132  (16 added, 45 deleted, 71 changed)
OCL=29468
CL=29497

	* conversions all in one place.
	* no separate load, store phases;
	  direct memory addressing when possible
	  (this is the x86 after all!).
	  avoids extra registers, extra MOVQs.
	* fixes int32 -> uint64 bug
	  (was zero-extending)

R=ken
OCL=29482
CL=29484

R=r
OCL=29479
CL=29479

R=rsc
DELTA=5  (0 added, 0 deleted, 5 changed)
OCL=29475
CL=29478

R=r
DELTA=1  (0 added, 1 deleted, 0 changed)
OCL=29473
CL=29477

between first and second pass of converting
[]int to string.

R=r
OCL=29467
CL=29467

R=r
OCL=29466
CL=29466

R=ken
OCL=29448
CL=29448

R=dsymonds
DELTA=5  (5 added, 0 deleted, 0 changed)
OCL=29424
CL=29447

R=rsc
DELTA=35  (35 added, 0 deleted, 0 changed)
OCL=29442
CL=29444

TBR=rsc
DELTA=14  (14 added, 0 deleted, 0 changed)
OCL=29441
CL=29441

R=ken
OCL=29425
CL=29427

R=ken
OCL=29426
CL=29426

can run 64-bit sieve and powser.
interfaces are limping along.
next hurdle is floating point.

R=ken
OCL=29418
CL=29423

R=rsc
APPROVED=rsc
DELTA=40  (38 added, 0 deleted, 2 changed)
OCL=29351
CL=29422

R=r
OCL=29421
CL=29421

R=r
OCL=29419
CL=29419

of getenv by low-level runtime easier.
fix 32-bit bug in gc (there are still more).

R=ken
OCL=29415
CL=29415

R=r
DELTA=129  (78 added, 1 deleted, 50 changed)
OCL=29411
CL=29413

R=ken
OCL=29412
CL=29412

R=r
OCL=29409
CL=29409

test/turing.go runs if you move the big array off its stack.

finally remembered to g4 add cgen.c gsubr.c

R=ken
OCL=29408
CL=29408

R=ken
OCL=29407
CL=29407

R=r
OCL=29405
CL=29405