R=ken
OCL=33792
CL=33803

R=gri
DELTA=20  (17 added, 0 deleted, 3 changed)
OCL=33793
CL=33802

in a .6 file, an export line

	//ffi T localfib remotefib remote.so

means the dynamic linker should initialize
localfib, always a pointer, to the address of
remotefib, either text (T) or data (D) after
loading remote.so.

the C compiler will generate an export section
when given the pragmas

	#pragma package fib
	#pragma ffi T localfib remotefib remote.so

needing #pragma package is a bit of a kludge
and hopefully could go away later.

this is just the 6 tool chain support.
other architectures will happen once 6 settles down.

code using this to do FFI is in a later CL.

R=r
DELTA=161  (141 added, 14 deleted, 6 changed)
OCL=33783
CL=33795

R=rsc
DELTA=1  (0 added, 0 deleted, 1 changed)
OCL=33788
CL=33791

bug192.go:11:5: error: redefinition of ‘fmt’
bug192.go:9:8: note: previous definition of ‘fmt’ was here

R=r
DELTA=2  (0 added, 0 deleted, 2 changed)
OCL=33780
CL=33790

R=rsc
DELTA=9  (5 added, 0 deleted, 4 changed)
OCL=33768
CL=33785

R=ken
OCL=33778
CL=33781

R=rsc
APPROVED=rsc
DELTA=85  (35 added, 0 deleted, 50 changed)
OCL=33761
CL=33767

variables to be given initial values in some cases, to make
building global scopes more convenient.

R=rsc
APPROVED=rsc
DELTA=29  (17 added, 0 deleted, 12 changed)
OCL=33760
CL=33766

R=ken
OCL=33765
CL=33765

also fix echo bug that just surfaced in mkerrors.sh

R=r
DELTA=67  (11 added, 19 deleted, 37 changed)
OCL=33743
CL=33757

  * use //ffi comments in package import data
    to generate relocation entries and library loads.
  * call initffi in rt0.s if present

R=r
DELTA=117  (91 added, 3 deleted, 23 changed)
OCL=33739
CL=33750

R=ken
OCL=33745
CL=33745

R=ken
OCL=33730
CL=33730

  * more elf constants from FreeBSD
  * simple relocation demo

R=r
DELTA=410  (354 added, 49 deleted, 7 changed)
OCL=33689
CL=33729

R=rsc
DELTA=196  (118 added, 24 deleted, 54 changed)
OCL=33716
CL=33727

DELTA=11  (4 added, 0 deleted, 7 changed)
OCL=33698
CL=33712

6g's error for I4 is not quite accurate
(it calls I4 a non-interface) but also
not a crash or infinite loop.

R=iant
DELTA=15  (15 added, 0 deleted, 0 changed)
OCL=33701
CL=33709

R=rsc
APPROVED=rsc
DELTA=329  (301 added, 2 deleted, 26 changed)
OCL=33696
CL=33706

R=rsc
APPROVED=rsc
DELTA=23  (14 added, 8 deleted, 1 changed)
OCL=33699
CL=33705

PC/line/aline conversion methods.

R=rsc
APPROVED=rsc
DELTA=458  (434 added, 15 deleted, 9 changed)
OCL=33677
CL=33702

R=ken
OCL=33700
CL=33700

R=ken
OCL=33694
CL=33697

which turns out to be invalid.

R=rsc
DELTA=15  (15 added, 0 deleted, 0 changed)
OCL=33691
CL=33693

message with the full path of the errchk script.  Catch that
by wrapping the if statement which invokes the compiler in a
subshell.  Use the $TMPOUT file as a flag to let the main
shell know whether the subshell ran.  Since the compiler
stdout and stderr are redirected, if the if statement produces
any output, then the compiler crashed, and we report that.

R=r,rsc
DELTA=14  (11 added, 1 deleted, 2 changed)
OCL=33690
CL=33692

- fixed a number of places where we tried to allocate 64bit
  regs. added honeypot in regalloc to catch these in the future.
- implemented quad copying in sgen
- cgen64.c, add, mul
- clearfat done
- D_REGREG output from 5g (linker already knew about them)
- gmove done
- optoas almost done, last bit probably not needed
- D_SHIFT support in list.c

R=rsc
APPROVED=rsc
DELTA=963  (711 added, 112 deleted, 140 changed)
OCL=33619
CL=33688

gccgo in a few different places, and crashes 6g as of this
writing.  Note that the index in a composite literal must be
looked up in a different namespace if the composite literal
type turns out to be a struct.

R=rsc
DELTA=36  (36 added, 0 deleted, 0 changed)
OCL=33681
CL=33685

type switch on them despite their private fields.  Add some
tracing stuff.

R=rsc
APPROVED=rsc
DELTA=18  (7 added, 0 deleted, 11 changed)
OCL=33678
CL=33683

compile with the current compiler.  Adjust expected error
message in method1.go to match 6g output.

R=rsc
DELTA=9  (2 added, 5 deleted, 2 changed)
OCL=33674
CL=33682

DELTA=48  (22 added, 0 deleted, 26 changed)
OCL=33657
CL=33668

R=r
DELTA=20  (20 added, 0 deleted, 0 changed)
OCL=33646
CL=33667

 * remove now-unused D_SBIG (was for typestrings)
 * rename elf64.[ch] to elf.[ch]
 * pull in elf headers from FreeBSD instead of writing our own
 * emit non-header ELF data in data section
 * stub out a few more ELF sections needed for dynamic loading

R=r
DELTA=1928  (1237 added, 635 deleted, 56 changed)
OCL=33642
CL=33658

DELTA=39  (4 added, 15 deleted, 20 changed)
OCL=33639
CL=33649

R=rsc
APPROVED=rsc
DELTA=2  (2 added, 0 deleted, 0 changed)
OCL=33640
CL=33645

TBR=r
OCL=33623
CL=33623

do not emit unreachable data symbols.

R=austin
DELTA=103  (71 added, 4 deleted, 28 changed)
OCL=33325
CL=33622

introduced explicit "data" symbol instead of etext
to mark beginning of data, so that using larger
alignment (i.e. 4MB like GNU loader) doesn't
confuse garbage collector.

split dodata into dodata and dobss in preparation
for putting the dynamic data + headers in the data
segment instead of stuffed at the beginning of the binary.

R=r
DELTA=52  (37 added, 3 deleted, 12 changed)
OCL=33610
CL=33618

R=r
DELTA=1  (1 added, 0 deleted, 0 changed)
OCL=33615
CL=33617

length-distance pairs.

The new test data was generated by "gzip shesells.txt", which is
presumably what you (rsc) did before, for the other test cases in
gunzip_test.go.

R=rsc
APPROVED=rsc
DELTA=21  (17 added, 2 deleted, 2 changed)
OCL=33582
CL=33616

R=rsc
DELTA=96  (25 added, 5 deleted, 66 changed)
OCL=33607
CL=33612