This isn't everything, but it's a start.
Still producing bit-identical compiler output.
The semantics of the old back ends is preserved,
even when they are probably buggy.
There are some TODOs in gc/gsubr.c to
remove special cases to preserve bugs in 5g and 8g.
Change-Id: I28ae295fbfc94ef9df43e13ab96bd6fc2f194bc4
Reviewed-on: https://go-review.googlesource.com/3802
Reviewed-by: Austin Clements <austin@google.com>
Add Addr-checking for all Progs on input to liblink, in liblink/pass.c,
including requiring use of TYPE_ADDR, not TYPE_CONST.
Update compilers and assemblers to satisfy checks.
Change-Id: Idac36b9f6805f0451cb541d2338992ca5eaf3963
Reviewed-on: https://go-review.googlesource.com/3801
Reviewed-by: Austin Clements <austin@google.com>
This makes names like ANOP, ATEXT, AGLOBL, ACALL, AJMP, ARET
available for use by architecture-independent processing passes.
On arm and ppc64, the alternate names are now aliases for the
official ones (ABL for ACALL, AB or ABR for AJMP, ARETURN for ARET).
Change-Id: Id027771243795af2b3745199c645b6e1bedd7d18
Reviewed-on: https://go-review.googlesource.com/3577
Reviewed-by: Aram Hăvărneanu <aram@mgk.ro>
Reviewed-by: Austin Clements <austin@google.com>
Like the TEXT/GLOBL flags, this was split between from.scale and reg,
neither of which is appropriate.
Change-Id: I2a16ef066a53b6edb7afb16cce108c0d1d26389c
Reviewed-on: https://go-review.googlesource.com/3576
Reviewed-by: Aram Hăvărneanu <aram@mgk.ro>
Reviewed-by: Austin Clements <austin@google.com>
Use AXXX instead of AGOK (neither is a valid instruction but AXXX is zero)
for the initial setting of Prog.as, and now there are no non-zero default
field settings.
Remove the arch-specific zprog/zprg in favor of a single global zprog.
Remove the arch-specific prg constructor in favor of emallocz(sizeof(Prog)).
Change-Id: Ia73078726768333d7cdba296f548170c1bea9498
Reviewed-on: https://go-review.googlesource.com/3575
Reviewed-by: Aram Hăvărneanu <aram@mgk.ro>
Reviewed-by: Austin Clements <austin@google.com>
Before, amd64 and 386 stored the flags in p->from.scale
and arm and ppc64 stored the flags in p->reg.
Both caused special cases in printing and in handling of the
addresses.
To avoid possible conflicts with the real meaning of p->from
and to avoid storing a non-register value in a reg field,
use from3 to hold a TYPE_CONST value giving the flags.
There is still a special case for printing, because the flags
are specified without a $, and normally a TYPE_CONST prints
with a $. But that's much less special than what came before.
This allows us to remove the textflag and settextflag methods
from LinkArch. They are no longer architecture-specific.
Change-Id: I931da8e1ecd92e127cd9aa44ef5a73c42e730110
Reviewed-on: https://go-review.googlesource.com/3572
Reviewed-by: Austin Clements <austin@google.com>
Fourth time's the charm.
Actually this doesn't fix the build, there is a
crash after go_bootstrap is compiled which looks
like it is related to auxv parsing.
Change-Id: Id00e2dfbe7bae42856f996065d3fb90b820e29a8
Reviewed-on: https://go-review.googlesource.com/3610
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Another attempt to fix the arm build by moving the include of signal.h
to cmd/lex.c, unless we are building on plan9.
Obviously if we had a plan9/arm builder this would probably not work, but
this is only a temporary measure until the c2go transition is complete.
Change-Id: I7f8ae27349b2e7a09c55db03e02a01939159a268
Reviewed-on: https://go-review.googlesource.com/3566
Reviewed-by: Russ Cox <rsc@golang.org>
- Change forward reference to struct Node* to void* in liblink.
- Use explicit (Node*) casts in cmd/gc to get at that field.
- Define struct Array in go.h instead of hiding it in array.c.
- Remove some sizeof(uint32), sizeof(uint64) uses.
- Remove some ? : expressions.
- Rewrite some problematic mid-expression assignments.
Change-Id: I308c70140238a0cfffd90e133f86f442cd0e17d4
Reviewed-on: https://go-review.googlesource.com/3276
Reviewed-by: Rob Pike <r@golang.org>
The function getgohostos and getgohostarch
were declared in include/libc.h in CL 3042.
Change-Id: Ib4ff5182cb71cc79a99663ce727fa4c28d15d7ad
Reviewed-on: https://go-review.googlesource.com/3122
Reviewed-by: Ian Lance Taylor <iant@golang.org>
The function runcmd was declared in
include/libc.h in CL 7523043.
Change-Id: I3839b96b2ac0d63e5c2eb4c216710442d0962119
Reviewed-on: https://go-review.googlesource.com/3125
Reviewed-by: Ian Lance Taylor <iant@golang.org>
This CL enables moving the bulk of the object writing code
out of liblink and into translated Go libraries in cmd/internal/obj,
but it does not do the move.
This CL introduces two new environment variables,
$GOOBJ and $GOOBJWRITER, but both will be deleted along with
the rest of the liblink C code.
The default behavior of a build is unchanged by this CL:
the C version of liblink uses the C object layout and writing code.
If $GOOBJ=1, liblink invokes go tool objwriter instead.
If $GOOBJ=2, liblink does its own layout and then invokes
go tool objwriter, which checks that it gets the same answer.
That is, in $GOOBJ=2 mode, both the C and the Go version of
the code run, and the operation fails if the two produce different
answers. This provides a very strong check that the translation
is working correctly.
Change-Id: I56ab49b07ccb2c7b81085f1d6950131047c6aa3c
Reviewed-on: https://go-review.googlesource.com/3048
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Needed for invoking a Go subprocess in the C code.
The Go tools live in $GOROOT/pkg/tool/$GOHOSTARCH_$GOHOSTOS.
Change-Id: I961b6b8a07de912de174b758b2fb87d77080546d
Reviewed-on: https://go-review.googlesource.com/3042
Reviewed-by: Ian Lance Taylor <iant@golang.org>
This implements the ELF relocations and dynamic linking tables
necessary to support internal linking on ppc64. It also marks ppc64le
ELF files as ABI v2; failing to do this doesn't seem to confuse the
loader, but it does confuse libbfd (and hence gdb, objdump, etc).
Change-Id: I559dddf89b39052e1b6288a4dd5e72693b5355e4
Reviewed-on: https://go-review.googlesource.com/2006
Reviewed-by: Russ Cox <rsc@golang.org>
Most ppc64 relocations come in six or more variants where the basic
relocation formula is the same, but which bits of the computed value
are installed where changes. Introduce the concept of "variants" for
internal relocations to support this. Since this applies to
architecture-independent relocation types like R_PCREL, we do this in
relocsym.
Currently there is only an identity variant. A later CL that adds
support for ppc64 ELF relocations will introduce more.
Change-Id: I0c5f0e7dbe5beece79cd24fe36267d37c52f1a0c
Reviewed-on: https://go-review.googlesource.com/2005
Reviewed-by: Russ Cox <rsc@golang.org>
ppc64 function symbols have both a global entry point and a local
entry point, where the difference is stashed in sym.other. We'll need
this information to generate calls to ELF ABI functions.
Change-Id: Ibe343923f56801de7ebec29946c79690a9ffde57
Reviewed-on: https://go-review.googlesource.com/2002
Reviewed-by: Russ Cox <rsc@golang.org>
Reviewed-by: Minux Ma <minux@golang.org>
This reverts commit ab0535ae3f.
I think it will remain useful to distinguish code that must
run on a system stack from code that can run on either stack,
even if that distinction is no
longer based on the implementation language.
That is, I expect to add a //go:systemstack comment that,
in terms of the old implementation, tells the compiler,
to pretend this function was written in C.
Change-Id: I33d2ebb2f99ae12496484c6ec8ed07233d693275
Reviewed-on: https://go-review.googlesource.com/2275
Reviewed-by: Russ Cox <rsc@golang.org>
Our definition of struct timespec used to cause problems with
certain versions of mingw-rt. However, as it turns out, we don't
actually need those definitions and prototypes, so remove them.
Fixes#9472.
Change-Id: Ie0880f0d58be112625140f73d0bed71f98b7cf05
Reviewed-on: https://go-review.googlesource.com/2236
Reviewed-by: Alex Brainman <alex.brainman@gmail.com>
Now that we've removed all the C code in runtime and the C compilers,
there is no need to have a separate stackguard field to check for C
code on Go stack.
Remove field g.stackguard1 and rename g.stackguard0 to g.stackguard.
Adjust liblink and cmd/ld as necessary.
Change-Id: I54e75db5a93d783e86af5ff1a6cd497d669d8d33
Reviewed-on: https://go-review.googlesource.com/2144
Reviewed-by: Keith Randall <khr@golang.org>
warning: src/liblink/list6.c:94 set and not used: s
warning: src/liblink/list6.c:157 format mismatch ld VLONG, arg 3
warning: src/liblink/list6.c:157 format mismatch E UINT, arg 4
warning: src/liblink/list6.c:157 format mismatch d VLONG, arg 5
warning: src/liblink/list6.c:163 set and not used: s
warning: src/liblink/list9.c:105 set and not used: s
warning: src/liblink/list9.c:185 format mismatch ld VLONG, arg 3
warning: src/liblink/list9.c:185 format mismatch E UINT, arg 4
warning: src/liblink/list9.c:185 format mismatch d VLONG, arg 5
warning: src/liblink/list9.c:193 set and not used: s
LGTM=rsc
R=rsc
CC=austin, golang-codereviews, minux
https://golang.org/cl/176130043
This is more complicated than the other enums because the D_*
enums are full of explicit initializers and repeated values.
This tries its best. (This will get much cleaner once we
tease these constants apart better.)
LGTM=rsc
R=rsc
CC=golang-codereviews
https://golang.org/cl/166700043
This brings dev.power64 up-to-date with the current tip of
default. go_bootstrap is still panicking with a bad defer
when initializing the runtime (even on amd64).
LGTM=rsc
R=rsc
CC=golang-codereviews
https://golang.org/cl/152570049
The assembler could give a better error, but this one
is good enough for now.
Fixes#8880.
LGTM=r
R=r
CC=golang-codereviews
https://golang.org/cl/153610043
Corrections due to new strict type rules for data+bss.
Also disable misc/cgo/cdefstest since you can't compile C code anymore.
TBR=iant
CC=golang-codereviews
https://golang.org/cl/148050044
The liblink code to insert the FUNCDATA for a stack map
from the Go prototype was not correct for ARM
(different data structure layout).
Also, sync/atomic was missing some Go prototypes
for ARM-specific functions.
TBR=r
CC=golang-codereviews
https://golang.org/cl/143160045
Commit to stack copying for stack growth.
We're carrying around a surprising amount of cruft from older schemes.
I am confident that precise stack scans and stack copying are here to stay.
Delete fallback code for when precise stack info is disabled.
Delete fallback code for when copying stacks is disabled.
Delete fallback code for when StackCopyAlways is disabled.
Delete Stktop chain - there is only one stack segment now.
Delete M.moreargp, M.moreargsize, M.moreframesize, M.cret.
Delete G.writenbuf (unrelated, just dead).
Delete runtime.lessstack, runtime.oldstack.
Delete many amd64 morestack variants.
Delete initialization of morestack frame/arg sizes (shortens split prologue!).
Replace G's stackguard/stackbase/stack0/stacksize/
syscallstack/syscallguard/forkstackguard with simple stack
bounds (lo, hi).
Update liblink, runtime/cgo for adjustments to G.
LGTM=khr
R=khr, bradfitz
CC=golang-codereviews, iant, r
https://golang.org/cl/137410043
The helps certain diagnostics and also removed duplicated enums as a side effect.
LGTM=dave, rsc
R=rsc, dave
CC=golang-codereviews
https://golang.org/cl/115060044
Rewrite gotos that violate Go's stricter rules.
Use uchar* instead of char* in a few places that aren't strings.
Remove dead opcross code from asm5.c.
Declare regstr (in both list6 and list8) static.
LGTM=minux, dave
R=minux, dave
CC=golang-codereviews
https://golang.org/cl/113230043
The main changes fall into a few patterns:
1. Replace #define with enum.
2. Add /*c2go */ comment giving effect of #define.
This is necessary for function-like #defines and
non-enum-able #defined constants.
(Not all compilers handle negative or large enums.)
3. Add extra braces in struct initializer.
(c2go does not implement the full rules.)
This is enough to let c2go typecheck the source tree.
There may be more changes once it is doing
other semantic analyses.
LGTM=minux, iant
R=minux, dave, iant
CC=golang-codereviews
https://golang.org/cl/106860045
A TLS slot is reserved by _rt0_.*_plan9 as an automatic and
its address (which is static on Plan 9) is saved in the
global _privates symbol. The startup linkage now is exactly
like that from Plan 9 libc, and the way we access g is
exactly as if we'd have used privalloc(2).
Aside from making the code more standard, this change
drastically simplifies it, both for 386 and for amd64, and
makes the Plan 9 code in liblink common for both 386 and
amd64.
The amd64 runtime code was cleared of nxm assumptions, and
now runs on the standard Plan 9 kernel.
Note handling fixes will follow in a separate CL.
LGTM=rsc
R=golang-codereviews, rsc, bradfitz, dave
CC=0intro, ality, golang-codereviews, jas, minux.ma, mischief
https://golang.org/cl/101510049
The runtime has historically held two dedicated values g (current goroutine)
and m (current thread) in 'extern register' slots (TLS on x86, real registers
backed by TLS on ARM).
This CL removes the extern register m; code now uses g->m.
On ARM, this frees up the register that formerly held m (R9).
This is important for NaCl, because NaCl ARM code cannot use R9 at all.
The Go 1 macrobenchmarks (those with per-op times >= 10 µs) are unaffected:
BenchmarkBinaryTree17 5491374955 5471024381 -0.37%
BenchmarkFannkuch11 4357101311 4275174828 -1.88%
BenchmarkGobDecode 11029957 11364184 +3.03%
BenchmarkGobEncode 6852205 6784822 -0.98%
BenchmarkGzip 650795967 650152275 -0.10%
BenchmarkGunzip 140962363 141041670 +0.06%
BenchmarkHTTPClientServer 71581 73081 +2.10%
BenchmarkJSONEncode 31928079 31913356 -0.05%
BenchmarkJSONDecode 117470065 113689916 -3.22%
BenchmarkMandelbrot200 6008923 5998712 -0.17%
BenchmarkGoParse 6310917 6327487 +0.26%
BenchmarkRegexpMatchMedium_1K 114568 114763 +0.17%
BenchmarkRegexpMatchHard_1K 168977 169244 +0.16%
BenchmarkRevcomp 935294971 914060918 -2.27%
BenchmarkTemplate 145917123 148186096 +1.55%
Minux previous reported larger variations, but these were caused by
run-to-run noise, not repeatable slowdowns.
Actual code changes by Minux.
I only did the docs and the benchmarking.
LGTM=dvyukov, iant, minux
R=minux, josharian, iant, dave, bradfitz, dvyukov
CC=golang-codereviews
https://golang.org/cl/109050043
The USEFIELD instructions no longer make it to the linker,
so we have to do something else to pin the references
they were pinning. Emit a 0-length relocation of type R_USEFIELD.
Fixes#7486.
LGTM=iant
R=golang-codereviews, iant
CC=golang-codereviews, r
https://golang.org/cl/95530043
Since there is no Runemask constant in UTF-8 enum in
9atom, removing the other constants leads to an empty
enum, which is illegal. Thus, we remove enum entirely.
LGTM=r
R=golang-codereviews, r
CC=golang-codereviews
https://golang.org/cl/82660044
The new code is adapted from the Go 1.2 nosplit code,
but it does not have the bug reported in issue 7623:
g% go run nosplit.go
g% go1.2 run nosplit.go
BUG
rejected incorrectly:
main 0 call f; f 120
linker output:
# _/tmp/go-test-nosplit021064539
main.main: nosplit stack overflow
120 guaranteed after split check in main.main
112 on entry to main.f
-8 after main.f uses 120
g%
Fixes#6931.
Fixes#7623.
LGTM=iant
R=golang-codereviews, iant, ality
CC=golang-codereviews, r
https://golang.org/cl/88190043
The name linkwriteobj is misleading because it implies that
the function has something to do with the linker, which it
does not. The name is historical: the function performs an
operation that was previously performed by the linker, but no
longer is.
LGTM=rsc
R=rsc, minux.ma
CC=golang-codereviews
https://golang.org/cl/88210045
linklookup uses hash(name, v) as the hash table index but then
only compares name to find a symbol to return.
If hash(name, v1) == hash(name, v2) for v1 != v2, the lookup
for v2 will return the symbol with v1.
The input routines assume that each symbol is found only once,
and then each symbol is added to a linked list, with the list header
in the symbol. Adding a symbol to such a list multiple times
short-circuits the list the second time it is added, causing symbols
to be dropped.
The liblink rewrite introduced an elegant, if inefficient, handling
of duplicated symbols by creating a dummy symbol to read the
duplicate into. The dummy symbols are named .dup with
sequential version numbers. With many .dup symbols, eventually
there will be a conflict, causing a duplicate list add, causing elided
symbols, causing a crash when calling one of the elided symbols.
The bug is old (2011) but could not have manifested until the
liblink rewrite introduced this heavily duplicated symbol .dup.
(See History section below.)
1. Correct the lookup function.
2. Since we want all the .dup symbols to be different, there's no
point in inserting them into the table. Call linknewsym directly,
avoiding the lookup function entirely.
3. Since nothing can refer to the .dup symbols, do not bother
adding them to the list of functions (textp) at all.
4. In lieu of a unit test, introduce additional consistency checks to
detect adding a symbol to a list multiple times. This would have
caught the short-circuit more directly, and it will detect a variety
of double-use bugs, including the one arising from the bad lookup.
Fixes#7749.
History
On April 9, 2011, I submitted CL 4383047, making ld 25% faster.
Much of the focus was on the hash table lookup function, and
one of the changes was to remove the s->version == v comparison [1].
I don't know if this was a simple editing error or if I reasoned that
same name but different v would yield a different hash slot and
so the name test alone sufficed. It is tempting to claim the former,
but it was probably the latter.
Because the hash is an iterated multiply+add, the version ends up
adding v*3ⁿ to the hash, where n is the length of the name.
A collision would need x*3ⁿ ≡ y*3ⁿ (mod 2²⁴ mod 100003),
or equivalently x*3ⁿ ≡ x*3ⁿ + (y-x)*3ⁿ (mod 2²⁴ mod 100003),
so collisions will actually be periodic: versions x and y collide
when d = y-x satisfies d*3ⁿ ≡ 0 (mod 2²⁴ mod 100003).
Since we allocate version numbers sequentially, this is actually
about the best case one could imagine: the collision rate is
much lower than if the hash were more random.
http://play.golang.org/p/TScD41c_hA computes the collision
period for various name lengths.
The most common symbol in the new linker is .dup, and for n=4
the period is maximized: the 100004th symbol is the first collision.
Unfortunately, there are programs with more duplicated symbols
than that.
In Go 1.2 and before, duplicate symbols were handled without
creating a dummy symbol, so this particular case for generating
many duplicate symbols could not happen. Go does not use
versioned symbols. Only C does; each input file gives a different
version to its static declarations. There just aren't enough C files
for this to come up in that context.
So the bug is old but the realization of the bug is new.
[1] https://golang.org/cl/4383047/diff/5001/src/cmd/ld/lib.c
LGTM=minux.ma, iant, dave
R=golang-codereviews, minux.ma, bradfitz, iant, dave
CC=golang-codereviews, r
https://golang.org/cl/87910047
