Compare commits

..
23 Commits
Author SHA1 Message Date
petrbalvin e02918c17b fix(ci): keep the push suite inside the runner's memory and time budget
Test / test (push) Successful in 2m55s
Assisted-by: GLM 5.3 Flash
2026-10-02 17:20:31 +02:00
petrbalvin 02a6359c1f ci: shrink the push pipeline to the affordable gate set
Test / test (push) Failing after 5m26s
Assisted-by: GLM 5.3 Flash
2026-10-02 16:47:52 +02:00
petrbalvin b4c1e133c0 ci: keep the GOOBJ link parity gate off the push pipeline
Test / test (push) Failing after 12m18s
Assisted-by: GLM 5.3 Flash
2026-10-02 16:15:02 +02:00
petrbalvin 1107928870 build(justfile): run the test recipes under the memory fence
Assisted-by: GLM 5.3 Flash
2026-10-02 16:15:02 +02:00
petrbalvin bc4ac93fd9 style(asm): reindent the evex comment gofmt asks for
Test / test (push) Failing after 21m29s
Assisted-by: GLM 5.3
2026-10-02 00:41:46 +02:00
petrbalvin c1bca7ce7e docs: record the development deltas in the changelog
Assisted-by: GLM 5.3
2026-10-02 00:40:54 +02:00
petrbalvin fefb76beb9 docs(asm): correct the reference against the assemblers' behaviour
Assisted-by: GLM 5.3
2026-10-02 00:40:54 +02:00
petrbalvin 42bc1669d7 feat(lsp): document directives on hover and widen completion
Assisted-by: GLM 5.3
2026-10-02 00:40:54 +02:00
petrbalvin 69dcbec8ef feat(lint): eleven new rules over directives, data and addressing
Assisted-by: GLM 5.3
2026-10-02 00:40:54 +02:00
petrbalvin f405cea5bc fix(cmd): stop the coverage run on a stray in-place trap
Assisted-by: GLM 5.3
2026-10-02 00:40:54 +02:00
petrbalvin f57377abb9 fix(debug): handle mapping edges, stray traps and dying debuggees
Assisted-by: GLM 5.3
2026-10-02 00:40:54 +02:00
petrbalvin dd1782c538 test(verify): gate GOOBJ link parity with cmd/link
Assisted-by: GLM 5.3
2026-10-02 00:40:54 +02:00
petrbalvin 17cc49fee4 fix(asm): emit NOPTR data as its own symbol kind
Assisted-by: GLM 5.3
2026-10-02 00:40:43 +02:00
petrbalvin bafb2fd130 feat(asm): encode the amd64 and loong64 tails of the corpus testdata
Assisted-by: GLM 5.3
2026-10-02 00:40:43 +02:00
petrbalvin 2f679326c2 style(testdata): canonicalise forms_amd64.s
Assisted-by: GLM 5.3
2026-10-02 00:40:20 +02:00
petrbalvin 96f2dd65b4 test(lexer): fuzz the token stream invariants
Assisted-by: GLM 5.3
2026-10-02 00:40:20 +02:00
petrbalvin ca887d3927 fix(parser): bound folding depth and macro expansion work
Assisted-by: GLM 5.3
2026-10-02 00:40:20 +02:00
petrbalvin 6570709226 fix(parser): peel stacked labels the way the formatter renders them
Assisted-by: GLM 5.3
2026-10-02 00:40:20 +02:00
petrbalvin 9a5217d9c1 fix(format): keep every token of a line in the canonical output
Assisted-by: GLM 5.3
2026-10-02 00:40:20 +02:00
petrbalvin 4d01bb3ecf build: rename the module to sourcedock.dev/petrbalvin/gasm-sdk
Test / test (push) Successful in 4m18s
2026-09-26 11:08:43 +02:00
petrbalvin 332c63e440 ci: compile-gate FreeBSD in the test pipeline
Test / test (push) Successful in 4m10s
Assisted-by: GLM 5.3 Flash
2026-09-25 21:46:49 +02:00
petrbalvin b306c210c6 feat(debug): port the debugger to FreeBSD
Assisted-by: GLM 5.3 Flash
2026-09-25 21:46:40 +02:00
petrbalvin b9015e1c2e fix(verify): make the executable mapping build on FreeBSD
Assisted-by: GLM 5.3 Flash
2026-09-25 21:46:31 +02:00
168 changed files with 8178 additions and 643 deletions
+42
View File
@@ -0,0 +1,42 @@
# FreeBSD compile gates. Dispatched by hand, never on a push.
#
# The debugger's ptrace surface and the JIT substrate are the two
# FreeBSD-portable layers the tree carries; the forge has no FreeBSD runner,
# so they can only be compile-gated, and three foreign-GOOS builds of the
# whole module are minutes of one-core work the push pipeline's budget cannot
# carry. The push pipeline stays fast and light; this workflow is the
# deliberate run, before a release or after touching the ported layers.
# Running the ptrace suite itself needs real FreeBSD hardware.
#
# A dispatched workflow takes no concurrency block: it is one deliberate run.
name: FreeBSD build
on:
workflow_dispatch:
env:
# One core: parallelism buys no speed here and costs memory the box does not have.
GOFLAGS: -p=1
GOMAXPROCS: "2"
jobs:
build:
runs-on: fedora
timeout-minutes: 10
steps:
- uses: actions/checkout@v7
- uses: actions/setup-go@v6
with:
# The module is the source of truth for the version, so it cannot drift.
go-version-file: go.mod
cache: true
- name: FreeBSD build (amd64)
run: GOOS=freebsd GOARCH=amd64 go build ./...
- name: FreeBSD build (arm64)
run: GOOS=freebsd GOARCH=arm64 go build ./...
- name: FreeBSD build (riscv64)
run: GOOS=freebsd GOARCH=riscv64 go build ./...
+19 -11
View File
@@ -6,6 +6,11 @@
# everything runs in one job. Extra jobs would duplicate the checkout, the Go setup and
# the dependency download three times without buying any parallelism.
#
# The budget is part of the contract: a push run is fast and light, about two minutes,
# and nothing that cannot run natively on the runner belongs here. The FreeBSD compile
# gates live in freebsd.yml behind workflow_dispatch for that reason; the GOOBJ link
# parity campaign is an opt-in local verification (just link-parity).
#
# Every step is one command, so the step that fails is the gate that failed, and no shell
# option has to be trusted for the run to stop. The scripted steps are Perl, not shell and
# not Python: Perl behaves the same on both runner images, there is no bashism to trip over
@@ -46,10 +51,9 @@ jobs:
go-version-file: go.mod
cache: true
- name: Install Perl
# The runner images are minimal and Perl is not guaranteed. The install is a
# no-op where it is already present; drop this step once verified on the box.
run: dnf install -y perl
# No Install Perl step: the fedora image carries perl (verified by run
# 76: the install degraded into a package upgrade costing ~50 s), and a
# dnf on the push path is network work the budget does not need.
# The steps follow the `gates` order of the justfile contract: build, format,
# vet, test. The vet gate is go vet and go fix -diff, two steps here.
@@ -82,7 +86,11 @@ jobs:
# command, so the floor is the same number everywhere. ./verify/... carries the
# live oracle-parity comparison against `go tool asm` (the TestGroundTruth
# suites); the runner's Go setup provides both the tool and GOROOT.
run: go test -count=1 -timeout 10m -coverprofile=coverage.out ./arch/... ./asm/... ./ast/... ./disasm/... ./format/... ./lexer/... ./lint/... ./lsp/... ./parser/... ./token/... ./verify/...
# -short skips the deliberate-run categories inside the suites (the live
# ptrace sessions above all): they need the machine to themselves and a
# starved single-core runner turns each into a timeout the budget cannot
# carry. The local `just test` gate runs everything, in full.
run: go test -short -count=1 -timeout 10m -coverprofile=coverage.out ./arch/... ./asm/... ./ast/... ./disasm/... ./format/... ./lexer/... ./lint/... ./lsp/... ./parser/... ./token/... ./verify/...
- name: Tests outside the coverage set
# The CLI and the debugger sit outside `packages` because a thin main and a
@@ -90,12 +98,12 @@ jobs:
# shipped surfaces: the command exit codes, the manual pages against the
# binary's own help, and the debugger's architecture-neutral units. They run
# here so the floor stays a product measure and nothing is left untested.
run: go test -count=1 -timeout 10m ./cmd/... ./debug/...
- name: Oracle parity
# Re-run the live go-tool-asm comparison as its own step so that a parity
# regression names the gate that failed instead of hiding inside the suite.
run: go test -count=1 -timeout 10m -run 'TestGroundTruth' ./verify/...
# -short skips the debugger's live ptrace sessions, the deliberate-run
# category the runner cannot starve-proof. The live go-tool-asm oracle
# comparison (TestGroundTruth in ./verify/...) runs inside the coverage
# sweep above; it is not re-run as its own step, because every second on
# this box is budget.
run: go test -short -count=1 -timeout 10m ./cmd/... ./debug/...
- name: Coverage floor
run: |
+110 -3
View File
@@ -1,6 +1,6 @@
# Changelog
All notable changes to gasm-devkit are documented here.
All notable changes to gasm-sdk are documented here.
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
@@ -9,6 +9,15 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
### Added
- **The FreeBSD port of the debugger.** `gasm debug` runs on FreeBSD on
amd64, arm64 and riscv64 with the same interactive surface as on Linux:
breakpoints, hardware watchpoints (x86 debug registers, the arm64 debug
register file), single-stepping, register and memory access, all behind
the kernel's own ptrace requests, with tracee memory through `PT_IO` and
stop reports through `PT_LWPINFO`. The JIT substrate maps executable
memory through `golang.org/x/sys/unix`, so `verify` builds on FreeBSD
too. The pipeline compile-gates all three architectures; live
validation awaits a FreeBSD machine.
- **Workspace-wide navigation in the language server.** `gasm lsp` indexes
the `.s` files under the workspace root beyond the documents the editor
has open, so go-to-definition, find references and workspace symbol search
@@ -20,6 +29,49 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
last one in the file, and the `abi-argsize` warning offers to set the
TEXT argument area to the size the `// func` signature implies, computed
by the new `lint.ExpectedArgSize`.
- **Eleven new lint rules over the directives, the data section and the
sharpest addressing edges.** `missing-argsize` flags a TEXT that
declares no argument area its `// func` signature implies;
`noframe-frame-size` a NOFRAME with a positive frame;
`unnamed-fp-reference` a nameless `0(FP)`, which both assemblers
reject; `hardware-sp-addressing` a negative offset off the hardware SP
rather than the virtual frame; `vex-sse-mixing` a kernel that mixes VEX
and legacy SSE spellings and pays the transition penalty;
`unnamed-result` a `ret+N(FP)` the signature names; `data-width`,
`data-value-overflow`, `data-string-width`, `data-without-globl` and
`data-exceeds-globl` police the DATA width against its value type and
the GLOBL size behind it. `missing-ret` now also flags a function
whose tail can fall off its end even though a RET sits somewhere in the
body, and `invalid-textflag` also reports a flag misplaced between TEXT
and GLOBL.
- **Hover documentation and wider completions in the language server.**
Hovering a directive or pseudo-operation (TEXT, DATA, GLOBL, PCALIGN,
FUNCDATA, PCDATA, the BYTE family) shows its grammar and rules in
preference to the empty instruction-table entry, and completion offers
those names beside the instruction set. The `missing-argsize` warning
carries a quick fix that declares the argument area the signature
implies (`$0` becomes `$0-16`).
- **The GOOBJ link parity gate.** A regression test assembles a kernel
per architecture through `gasm asm --format goobj`, substitutes the
object into a real `go build`'s package archive, proves the archive
carries it byte for byte and re-links with cmd/link on amd64, arm64,
riscv64 and loong64. The binaries run, natively on amd64 and under
qemu-user on arm64 and riscv64, and their output must match the
toolchain-built baseline; loong64 is link-only. The pipeline installs
qemu-user and runs the gate on every push.
- **The amd64 and loong64 encoders close four more corpus files.** The
whole-tree measure moves to 272 of 322 (84.5 %): amd64 gains the
one-operand IMUL, the SSE compare family (CMPPD, CMPPS, CMPSS), RETFL,
the LOOP family, the MMX register bank with its bank-crossing moves,
MOVNTDQ, the CR and DR register moves, PUSH and POP of FS and GS, the
`(TLS)` pseudo-base, the wait and cache controls (CLWB, CLDEMOTE,
TPAUSE, UMONITOR, UMWAIT, RDPID, ENDBR64), indirect branches with the
star spelling (`JMP *(R12)(R13*4)`), `RET sym(SB)` as the tail jump,
the colon shift spelling (`SHLL CX, R11:AX`) and the EVEX and VEX forms
of the rounds, AES key assist, string compares, extracts, blends and
permutes; loong64 gains the acquire and release pair (LLACQ, SCREL,
with the vector widths), the VMOVQ and XVMOVQ lane forms and the BYTE
literal-data escape hatch the other architectures already take.
### Changed
@@ -30,8 +82,15 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
instead of failing it. The headline now reads "assemble for every
applicable target": every real-code GOROOT assembly file, the tree
without testdata, assembles for all four architectures (250 of 250,
100 %); over the whole tree including testdata the measure is 271 of
322 (84.2 %).
100 %); over the whole tree including testdata the measure is 272 of
322 (84.5 %).
- **The module moves to `sourcedock.dev/petrbalvin/gasm-sdk`.** The
repository and the module rename together with the product, now the
GAsm Software Development Kit. Fresh installs become
`go install sourcedock.dev/petrbalvin/gasm-sdk/cmd/gasm@latest`, and
installs pinned to the old `gasm-sdk` path stop resolving once the
repository takes the new name: reinstall from the new path. The
binary stays `gasm`.
### Fixed
@@ -50,6 +109,54 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
rejected; the toolchain picks the 16-bit lane from the 64-bit bit
pattern, so the encoder now does the same, and a zero immediate is
rejected where the toolchain rejects it.
- **NOPTR data emits its own symbol kind.** A GLOBL with NOPTR was
emitted as plain SDATA, the kind the linker holds to its Go type
information requirement, so every gasm object carrying runtime-shaped
data (`GLOBL ·x(SB), NOPTR, ...`) died in cmd/link with "missing Go
type information". NOPTR data is now SNOPTRDATA, the toolchain's kind
for pointer-free globals, and RODATA still wins where both flags
appear, exactly as the toolchain chooses. The three end-to-end link
tests that should have caught this substituted the gasm object after
the package archive was already packed, so they passed vacuously; they
now substitute inside the archive, prove the substitution byte for
byte and re-link.
- **Latent encoder divergences against the toolchain, found by the
whole-file differential harness.** On loong64, `MOVx $off(reg)` lost
its base register, SCQ swapped its operand fields, the vector lane
inserts did not scale offsets by the element width and two immediate
opcodes were mistyped; on amd64, NOP with operands encoded 0x90 where
the toolchain emits nothing at all, the VEX gather length bit ignored
the VSIB index width, four permute and extract families were pushed
into EVEX where the toolchain stays VEX, and a reference to a static
symbol no GLOBL defines failed where the toolchain defers it to the
linker as an external relocation.
- **Seven front-end defects found by fuzzing.** The formatter swallowed
the statement after a leading block comment (`/* head */ MOVQ AX, BX`
formatted to the comment alone), dropped trailing tokens after an
`#include` header, and trimmed the trailing whitespace inside a string
literal; the parser peeled only one label of a stacked pair (`a: b:`
parsed differently than it formatted); deeply nested parentheses in an
immediate overflowed the stack through the constant folder, which now
stops at a bounded depth and falls back to the ordinary operand paths;
macro expansion is linear in the invocation count instead of
quadratic; and amplifying macros (the billion-laughs shape) stop at a
work budget sized by the line and the macro table, reported as a
diagnostic instead of running for hours.
- **The debugger handles the edges its tests now reach.** Memory reads
and writes cover exactly the requested bytes, so a request ending in
the last page of a mapping no longer fails on the unmapped page behind
it; disassembly shrinks its instruction window at a mapping's end
instead of failing; a debuggee that dies before signalling readiness
writes a failure notice the launcher reads, so launch fails fast with
the reason instead of after the whole poll (and the poll no longer
waits on the child, which could consume the SIGSTOP park and hang the
launch); amd64 watchpoints acknowledge the sticky DR6 hit bits and
clear the address register on release; the breakpoint listing is
ordered by address so its numbering is stable; the REPL rejects bad
counts, sizes and watchpoint types instead of silently guessing,
reports stops on signals during stepping, and a breakpoint-class trap
that matches no breakpoint and leaves the PC in place surfaces instead
of spinning the continue loop and the coverage run forever.
## [0.35.0] - 2026-09-22
+4 -4
View File
@@ -1,6 +1,6 @@
# Contributing
Contributions to **gasm-devkit** are governed by the Contributor terms
Contributions to **gasm-sdk** are governed by the Contributor terms
below; submitting one means you accept them.
## Contributor terms
@@ -29,8 +29,8 @@ compiler (gcc), because `just gates` includes `just race` and the race
detector needs cgo.
```sh
git clone https://sourcedock.dev/petrbalvin/gasm-devkit.git
cd gasm-devkit
git clone https://sourcedock.dev/petrbalvin/gasm-sdk.git
cd gasm-sdk
just build
just gates
```
@@ -126,7 +126,7 @@ tag, where it would double the time and the memory a shared runner cannot spare.
## Reporting bugs
Open an issue at `https://sourcedock.dev/petrbalvin/gasm-devkit/issues` with the
Open an issue at `https://sourcedock.dev/petrbalvin/gasm-sdk/issues` with the
version, the operating system and architecture, the exact command, the full output,
and the expected against the actual behaviour.
+14 -11
View File
@@ -1,6 +1,6 @@
# Plan 9 assembly tooling, inside and outside Go
# GAsm: Software Development Kit for Plan 9 Assembly
> **Warning: this is an experiment.** gasm-devkit is under active
> **Warning: this is an experiment.** gasm-sdk is under active
> development and is not stable. The version is 0.x.x: commands, flags,
> output formats and behaviour can change without warning at any time.
> A 1.0.0 release is light years away. Nothing in this document is a
@@ -13,7 +13,7 @@
there is no formatter, no linter and no debugger for `.s` files, and no
assembler that works without a Go installation. Developers write
assembly blind, validate it by benchmark, and debug it by print
statement. gasm-devkit is the missing toolkit: a single, self-contained
statement. gasm-sdk is the missing toolkit: a single, self-contained
binary, `gasm`, that serves both purposes.
- **Help develop Plan 9 assembly.** Formatting, linting, disassembly,
@@ -58,7 +58,7 @@ Plan 9 (Go): MOVQ AX, total-16(SP)
The same lines, but only one of them tells you what the number is for.
The syntax is uppercase, regular and boring, which is the highest
compliment a language for machine code can earn. gasm-devkit exists
compliment a language for machine code can earn. gasm-sdk exists
to give that syntax the tooling it deserves.
## Features
@@ -94,7 +94,8 @@ to give that syntax the tooling it deserves.
- **Debugger.** `gasm debug` is a source-level ptrace debugger with
breakpoints (optionally conditional), hardware watchpoints, register and
memory inspection, and headless script runs that report instruction and
label coverage.
label coverage; it runs on Linux (all four architectures) and FreeBSD
(amd64, arm64, riscv64).
- **Language server.** `gasm lsp` serves completion, hover, document symbols,
push and pull diagnostics, semantic-token highlighting, go-to-definition,
find references, rename, formatting, inlay hints, code actions, signature
@@ -150,7 +151,7 @@ actually been executed.
|---|---|---|
| Encoding: byte-for-byte against `go tool asm` | native hardware | native hardware (the toolchain cross-assembles any GOARCH on any host) |
| Execution: JIT calls, ABI checks, differential fuzzing | native hardware | qemu-user emulation |
| Debugger: ptrace tracing, breakpoints, watchpoints, coverage | native hardware | emulation cannot run ptrace; the layer compiles and its architecture-neutral units run under `go test ./...`, nothing more |
| Debugger: ptrace tracing, breakpoints, watchpoints, coverage | native hardware | emulation cannot run ptrace; the layer compiles and its architecture-neutral units run under `go test ./...`, nothing more. FreeBSD (amd64, arm64, riscv64) is in the same position: the port compiles behind the cross-build gate and its integration test is ready, but no FreeBSD machine has executed it |
Consequences, stated plainly. An emulator is a model of a CPU, not the
CPU: instruction semantics are implemented in software and can differ
@@ -220,9 +221,11 @@ The plan, in the order it is being worked:
toolchain itself does not support; through ELF, Plan 9 assembly becomes
usable outside Go entirely.
- **Platforms: Linux and FreeBSD.** Linux is supported today on all four
architectures and is where the binary builds. FreeBSD follows: the
JIT's executable-memory mapping and the ptrace debugger layer are the
two pieces of porting work. Other unix systems may follow those two.
architectures and is where the binary builds. FreeBSD follows on amd64,
arm64 and riscv64: the JIT's executable-memory mapping and the ptrace
debugger layer are ported (the debugger's live validation awaits a
FreeBSD machine, as the validation status states). Other unix systems
may follow those two.
- **Four architectures, no more.** amd64, arm64, riscv64 and loong64.
No others are planned.
@@ -230,11 +233,11 @@ The plan, in the order it is being worked:
Prebuilt binaries for linux/amd64, linux/arm64, linux/riscv64 and
linux/loong64 are on the
[releases page](https://sourcedock.dev/petrbalvin/gasm-devkit/releases).
[releases page](https://sourcedock.dev/petrbalvin/gasm-sdk/releases).
From source (Go 1.27.1):
```sh
go install sourcedock.dev/petrbalvin/gasm-devkit/cmd/gasm@latest
go install sourcedock.dev/petrbalvin/gasm-sdk/cmd/gasm@latest
```
Or from a repository checkout:
+7 -7
View File
@@ -5,7 +5,7 @@
// toolchain's own assembler source. Go's Plan 9 assembler defines the exact,
// complete set of mnemonics it accepts for each architecture in
// $GOROOT/src/cmd/internal/obj/<arch>/anames.go; this tool extracts those
// names so gasm-devkit supports every instruction the real assembler does,
// names so gasm-sdk supports every instruction the real assembler does,
// with no hand-maintained (and therefore inevitably incomplete) lists.
//
// The same data feeds the generated instruction appendices of the assembly
@@ -31,11 +31,11 @@ import (
"sort"
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/arch"
"sourcedock.dev/petrbalvin/gasm-devkit/asm"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/asm"
)
// archDirs maps a gasm-devkit architecture name to its obj sub-directory.
// archDirs maps a gasm-sdk architecture name to its obj sub-directory.
var archDirs = []struct {
arch string
sub string
@@ -119,7 +119,7 @@ func filterCommon(names []string) []string {
// writeCommon emits arch/common_gen.go.
func writeCommon(names []string) error {
var b strings.Builder
b.WriteString("// Code generated by gasm-devkit _gen; DO NOT EDIT.\n")
b.WriteString("// Code generated by gasm-sdk _gen; DO NOT EDIT.\n")
b.WriteString("// Source: cmd/internal/obj/util.go from the Go toolchain.\n")
b.WriteString("//\n")
b.WriteString("// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)\n")
@@ -190,7 +190,7 @@ func stringLit(elt ast.Expr) string {
// writeGen emits arch/<arch>_gen.go.
func writeGen(arch, sub string, names []string) error {
var b strings.Builder
b.WriteString("// Code generated by gasm-devkit _gen; DO NOT EDIT.\n")
b.WriteString("// Code generated by gasm-sdk _gen; DO NOT EDIT.\n")
b.WriteString("// Source: cmd/internal/obj/" + sub + "/anames.go from the Go toolchain.\n")
b.WriteString("//\n")
b.WriteString("// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)\n")
@@ -216,7 +216,7 @@ func writeDocPage(a arch.Arch, title, file, anames, version string, encodable bo
var b strings.Builder
b.WriteString("# " + title + ": instruction inventory\n\n")
b.WriteString("Generated by gasm-devkit's `_gen` from the Go toolchain's instruction table\n")
b.WriteString("Generated by gasm-sdk's `_gen` from the Go toolchain's instruction table\n")
b.WriteString("(`" + anames + "`, " + version + "); DO NOT EDIT. This page lists every mnemonic\n")
b.WriteString("`go tool asm` accepts on this target, which is the upper bound of the\n")
b.WriteString("language on it: a name absent here is not an instruction of the target,\n")
+1 -1
View File
@@ -1,4 +1,4 @@
// Code generated by gasm-devkit _gen; DO NOT EDIT.
// Code generated by gasm-sdk _gen; DO NOT EDIT.
// Source: cmd/internal/obj/x86/anames.go from the Go toolchain.
//
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
+1 -1
View File
@@ -1,4 +1,4 @@
// Code generated by gasm-devkit _gen; DO NOT EDIT.
// Code generated by gasm-sdk _gen; DO NOT EDIT.
// Source: cmd/internal/obj/arm64/anames.go from the Go toolchain.
//
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
+1 -1
View File
@@ -1,4 +1,4 @@
// Code generated by gasm-devkit _gen; DO NOT EDIT.
// Code generated by gasm-sdk _gen; DO NOT EDIT.
// Source: cmd/internal/obj/util.go from the Go toolchain.
//
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
+1 -1
View File
@@ -1,4 +1,4 @@
// Code generated by gasm-devkit _gen; DO NOT EDIT.
// Code generated by gasm-sdk _gen; DO NOT EDIT.
// Source: cmd/internal/obj/loong64/anames.go from the Go toolchain.
//
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
+1 -1
View File
@@ -1,4 +1,4 @@
// Code generated by gasm-devkit _gen; DO NOT EDIT.
// Code generated by gasm-sdk _gen; DO NOT EDIT.
// Source: cmd/internal/obj/riscv/anames.go from the Go toolchain.
//
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
+15 -66
View File
@@ -11,7 +11,7 @@ import (
"strings"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// TestGOObjectAARCH64Structure checks the basic structure of the emitted
@@ -178,26 +178,10 @@ func main() {
if err != nil {
t.Fatalf("baseline build: %v\n%s", err, buildLog)
}
var work, linkLine, asmObj string
for line := range strings.SplitSeq(string(buildLog), "\n") {
switch {
case strings.HasPrefix(line, "WORK="):
work = strings.TrimPrefix(line, "WORK=")
case strings.Contains(line, "/asm ") && strings.Contains(line, "main_arm64.s") && !strings.Contains(line, "-gensymabis"):
asmObj = fieldAfter(line, "-o")
case strings.Contains(line, "/link ") && strings.Contains(line, "-importcfg"):
linkLine = line
}
}
if work == "" || asmObj == "" {
t.Skipf("could not parse build log (work=%q asmObj=%q)", work, asmObj)
}
defer os.RemoveAll(work)
st := parseBuildLog(t, buildLog, "main_arm64.s")
defer os.RemoveAll(st.work)
// Expand $WORK in the object path.
asmObj = strings.ReplaceAll(asmObj, "$WORK", work)
// Read the toolchain-produced object and assemble the same source with gasm.
// Assemble the same source with gasm and substitute the object.
src, err := os.ReadFile(filepath.Join(dir, "main_arm64.s"))
if err != nil {
t.Fatal(err)
@@ -210,30 +194,16 @@ func main() {
if err != nil {
t.Fatalf("AssembleFileARM64: %v", err)
}
gasmObj, err := img.GOObjectAARCH64("a64link", "main_arm64.s")
// The package path is "main", the prefix the Go code's references carry.
gasmObj, err := img.GOObjectAARCH64("main", "main_arm64.s")
if err != nil {
t.Fatalf("GOObjectAARCH64: %v", err)
}
// Replace the toolchain-produced object with gasm's.
if err := os.WriteFile(asmObj, gasmObj, 0o644); err != nil {
t.Fatalf("write gasm object: %v", err)
}
// Re-link.
if linkLine == "" {
t.Skip("could not find link command in build log")
}
// Expand $WORK in the link command.
linkLine = strings.ReplaceAll(linkLine, "$WORK", work)
linkCmd := exec.Command("bash", "-c", "cd "+dir+" && "+linkLine)
linkCmd.Env = append(os.Environ(), "GOARCH=arm64")
if out, err := linkCmd.CombinedOutput(); err != nil {
t.Fatalf("re-link with gasm object: %v\n%s", err, out)
}
substituteAndRelink(t, goBin, dir, st, filepath.Join(dir, "prog2"),
gasmObj, "GOARCH=arm64")
// Verify the binary exists and contains the symbol.
binPath := filepath.Join(dir, "prog")
binPath := filepath.Join(dir, "prog2")
if _, err := os.Stat(binPath); err != nil {
t.Fatalf("binary not found: %v", err)
}
@@ -296,23 +266,8 @@ func main() {
if err != nil {
t.Fatalf("baseline build: %v\n%s", err, buildLog)
}
var work, linkLine, asmObj string
for line := range strings.SplitSeq(string(buildLog), "\n") {
switch {
case strings.HasPrefix(line, "WORK="):
work = strings.TrimPrefix(line, "WORK=")
case strings.Contains(line, "/asm ") && strings.Contains(line, "main_arm64.s") && !strings.Contains(line, "-gensymabis"):
asmObj = fieldAfter(line, "-o")
case strings.Contains(line, "/link ") && strings.Contains(line, "-importcfg"):
linkLine = line
}
}
if work == "" || asmObj == "" || linkLine == "" {
t.Skipf("could not parse build log (work=%q asmObj=%q link=%q)", work, asmObj, linkLine)
}
defer os.RemoveAll(work)
asmObj = strings.ReplaceAll(asmObj, "$WORK", work)
linkLine = strings.ReplaceAll(linkLine, "$WORK", work)
st := parseBuildLog(t, buildLog, "main_arm64.s")
defer os.RemoveAll(st.work)
src, err := os.ReadFile(filepath.Join(dir, "main_arm64.s"))
if err != nil {
@@ -326,19 +281,13 @@ func main() {
if err != nil {
t.Fatalf("AssembleFileARM64: %v", err)
}
gasmObj, err := img.GOObjectAARCH64("a64dlink", "main_arm64.s")
gasmObj, err := img.GOObjectAARCH64("main", "main_arm64.s")
if err != nil {
t.Fatalf("GOObjectAARCH64: %v", err)
}
if err := os.WriteFile(asmObj, gasmObj, 0o644); err != nil {
t.Fatalf("write gasm object: %v", err)
}
linkCmd := exec.Command("bash", "-c", "cd "+dir+" && "+linkLine)
linkCmd.Env = append(os.Environ(), "GOARCH=arm64")
if out, err := linkCmd.CombinedOutput(); err != nil {
t.Fatalf("re-link with gasm object: %v\n%s", err, out)
}
binData, err := os.ReadFile(filepath.Join(dir, "prog"))
substituteAndRelink(t, goBin, dir, st, filepath.Join(dir, "prog2"),
gasmObj, "GOARCH=arm64")
binData, err := os.ReadFile(filepath.Join(dir, "prog2"))
if err != nil {
t.Fatal(err)
}
+2 -1
View File
@@ -9,7 +9,7 @@ import (
"strconv"
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
)
// assembleARM64 assembles an AArch64 (arm64) TEXT function body into machine
@@ -4003,6 +4003,7 @@ func AssembleFileARM64(f *ast.File) (*Image, error) {
Size: d.size,
Static: d.static,
Rodata: d.rodata,
Noptr: d.noptr,
Dupok: d.dupok,
})
}
+1 -1
View File
@@ -33,7 +33,7 @@ import (
"strconv"
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
)
// arm64RegNum returns the 5-bit register number for an AArch64 register name:
+2 -2
View File
@@ -7,8 +7,8 @@ import (
"strings"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
func TestArm64LDRSTREncoding(t *testing.T) {
+1 -1
View File
@@ -53,7 +53,7 @@ package asm
import (
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
)
// arm64FrameInfo holds the frame layout derived from a TEXT directive.
+1 -1
View File
@@ -7,7 +7,7 @@ import (
"encoding/binary"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// parseArm64File is a helper assembling one arm64 source file.
+215 -18
View File
@@ -8,7 +8,7 @@ import (
"strconv"
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
)
// Assemble encodes the body of a TEXT function into x86-64 machine code,
@@ -808,17 +808,43 @@ func isJumpMnemonic(mnem string) bool {
if mnem == "JMP" || mnem == "CALL" {
return true
}
if isLoopMnemonic(mnem) {
return true
}
_, ok := condCode(mnem)
return ok
}
// isLoopMnemonic reports the LOOP family, rel8 alone (E0-E2).
func isLoopMnemonic(mnem string) bool {
switch mnem {
case "LOOP", "LOOPE", "LOOPNE":
return true
}
return false
}
// loopOpcode maps the LOOP family to its E0-E2 opcode.
func loopOpcode(mnem string) byte {
switch mnem {
case "LOOPE":
return 0xE1
case "LOOPNE":
return 0xE0
}
return 0xE2
}
// jumpSize returns the length of a jump instruction in the requested form:
// short (rel8) where available, otherwise the rel32 form. CALL is always
// rel32.
// rel32; the LOOP family is rel8 alone.
func jumpSize(mnem string, long bool) int {
if mnem == "CALL" {
return 5 // opcode + rel32
}
if isLoopMnemonic(mnem) {
return 2 // opcode + rel8, the only form
}
if !long {
return 2 // opcode + rel8
}
@@ -945,6 +971,16 @@ func encodeNormal(s *ast.Instr, fi frameInfo, link *linkInfo) ([]byte, []sbPatch
if (mnemUpper == "MOVQ" || mnemUpper == "MOVL") && len(s.Operands) == 2 && isBareTLS(s.Operands[0]) {
return encodeTLSBaseLoad(s, fi, link)
}
// The old paired-register shift spelling, SHLL CX, R11:AX (a colon
// between the two registers), is the toolchain's SHLD family: SHLDL CL,
// AX, R11 with the count register first, the paired source in the reg
// field and the pair's head in r/m.
if code, ps, err := encodeColonShift(s, mnemUpper, fi, link); code != nil || err != nil {
if err != nil {
return nil, nil, nil, err
}
return code, ps, nil, nil
}
_, size := splitSize(mnemUpper)
if size == 0 {
size = 8
@@ -1051,6 +1087,66 @@ func encodeBookkeeping(upper string, s *ast.Instr) ([]byte, error) {
return nil, nil
}
// encodeColonShift encodes the paired-register shift spellings, SHLx CX,
// dst:src: the toolchain reads them as the SHLD family (double-precision
// shift by CL), reg = the paired source, r/m = the pair's head. The second
// operand's raw text carries the colon; ok reports the spelling was found.
func encodeColonShift(s *ast.Instr, mnemUpper string, fi frameInfo, link *linkInfo) ([]byte, []sbPatch, error) {
base, _ := strings.CutPrefix(mnemUpper, "SHL")
if base == mnemUpper || len(s.Operands) != 2 {
return nil, nil, nil
}
_, size := splitSize(mnemUpper)
raw := strings.ReplaceAll(s.Operands[1].Raw, " ", "")
head, tail, ok := strings.Cut(raw, ":")
if !ok || head == "" || tail == "" {
return nil, nil, nil
}
headReg, ok1 := ParseReg(head)
srcReg, ok2 := ParseReg(tail)
if !ok1 || !ok2 {
return nil, nil, fmt.Errorf("%s: invalid paired register %q", mnemUpper, s.Operands[1].Raw)
}
cnt, err := operandFromAST(mnemUpper, s.Operands[0], size, fi, link)
if err != nil {
return nil, nil, err
}
cntReg, ok := cnt.(Reg)
if !ok || cntReg.idx != 1 {
return nil, nil, fmt.Errorf("%s: the paired-register form counts in CL", mnemUpper)
}
// SHLD r/m, reg, CL: 0F A5 (REX.W for the 64-bit width).
e := &enc{}
i := &instr{rexW: size == 8, opcode: []byte{0x0F, 0xA5}, modrm: -1, sib: -1}
if err := setRM(i, srcReg, headReg, size); err != nil {
return nil, nil, err
}
if err := e.emit(i); err != nil {
return nil, nil, err
}
ps := make([]sbPatch, len(e.patches))
for j, p := range e.patches {
ps[j] = sbPatch{off: p.off, name: p.name, addend: p.addend}
}
return e.out, ps, nil
}
// trailingIndexGroup recovers a trailing "(index*scale)" or "(index)" group
// from an operand's raw text: the symbol-pseudo parse returns before the
// index group, so foo(SP)(AX*1) keeps its index only in the spelling.
func trailingIndexGroup(raw string) (string, int, bool) {
compact := strings.ReplaceAll(raw, " ", "")
if !strings.HasSuffix(compact, ")") {
return "", 0, false
}
open := strings.LastIndex(compact, "(")
if open < 2 || !strings.Contains(compact[:open], ")") {
return "", 0, false // one group alone: no trailing index
}
name, scale, _, ok := cutParenGroup(compact[open:])
return name, scale, ok
}
// encodeJump encodes a JMP/CALL/Jcc with a relative offset resolved from the
// target label or from a numeric ±N(PC) instruction count, in the short
// (rel8) or long (rel32) form. numTarget is the resolved byte offset of a
@@ -1085,9 +1181,15 @@ func encodeJump(s *ast.Instr, mnem string, pc int, offsets map[string]int, long
if mnem == "JMP" {
return []byte{0xEB, byte(int8(rel))}, nil
}
if isLoopMnemonic(mnem) {
return []byte{loopOpcode(mnem), byte(int8(rel))}, nil
}
cc, _ := condCode(mnem)
return []byte{0x70 + byte(cc), byte(int8(rel))}, nil
}
if isLoopMnemonic(mnem) {
return nil, fmt.Errorf("%s has no long form", mnem)
}
switch mnem {
case "JMP":
return append([]byte{0xE9}, le32(rel)...), nil
@@ -1139,6 +1241,72 @@ func labelName(op *ast.Operand) (string, bool) {
return "", false
}
// jumpOperand returns the branch-target operand of a JMP/CALL, rewriting the
// `*`-prefixed indirect spellings (JMP *(R12), JMP *4(SP)) into their plain
// memory form. The star marks an indirect target and changes no bytes; the
// address parser leaves the operand's address empty because of the leading
// star, so the fields are rebuilt from the raw text onto a copy of the
// operand, never on the shared syntax tree.
func jumpOperand(s *ast.Instr) *ast.Operand {
if len(s.Operands) != 1 {
return nil
}
op := s.Operands[0]
compact := strings.ReplaceAll(op.Raw, " ", "")
inner, ok := strings.CutPrefix(compact, "*")
if !ok {
return op
}
var addr ast.Address
if i := strings.IndexByte(inner, '('); i > 0 {
v, err := strconv.ParseInt(inner[:i], 0, 64)
if err != nil {
return op
}
addr.Offset, addr.HasOff = v, true
inner = inner[i:]
}
base, _, rest, ok := cutParenGroup(inner)
if !ok {
return op
}
if base != "" {
addr.Base = base
}
if rest != "" {
idx, scale, _, ok := cutParenGroup(rest)
if ok && idx != "" {
addr.Index = idx
addr.Scale = scale
}
}
c := *op
c.Addr = addr
return &c
}
// cutParenGroup splits a leading "(name)" or "(name*n)" off s, returning the
// inner text, the scale it names (1 when the group spells no multiplier) and
// the remainder.
func cutParenGroup(s string) (name string, scale int, rest string, ok bool) {
if !strings.HasPrefix(s, "(") {
return "", 0, "", false
}
i := strings.IndexByte(s, ')')
if i < 0 {
return "", 0, "", false
}
inner, rest := s[1:i], s[i+1:]
if before, after, ok := strings.Cut(inner, "*"); ok {
n, err := strconv.Atoi(after)
if err != nil {
return "", 0, "", false
}
return before, n, rest, true
}
return inner, 1, rest, true
}
// indirectJumpTarget reports whether the JMP/CALL operand addresses a
// register or a memory location rather than a label or a static symbol.
// A bare identifier is a register when the register table knows the name and
@@ -1147,7 +1315,14 @@ func indirectJumpTarget(s *ast.Instr) bool {
if len(s.Operands) != 1 || s.Operands[0].Kind != ast.OpAddr {
return false
}
a := s.Operands[0].Addr
op := jumpOperand(s)
if op == nil {
return false
}
if op != s.Operands[0] {
return true // the star marker spells an indirect target
}
a := op.Addr
// ±N(PC) is the numeric relative form, the PC counts instructions from
// the branch: relative, not indirect.
if a.Base == "PC" || a.Index == "PC" {
@@ -1165,18 +1340,20 @@ func indirectJumpTarget(s *ast.Instr) bool {
}
// encodeIndirectJump assembles a JMP/CALL through a register or memory
// operand, which carries no relocation and no label to resolve.
// operand, which carries no relocation and no label to resolve. The
// `*`-prefixed spellings go through jumpOperand first, their star rebuilt
// into a plain memory operand.
func encodeIndirectJump(s *ast.Instr, mnem string) ([]byte, error) {
ops := make([]Operand, len(s.Operands))
for i, op := range s.Operands {
op := s.Operands[0]
if cleaned := jumpOperand(s); cleaned != nil {
op = cleaned
}
o, err := operandFromAST(mnem, op, 8, frameInfo{}, nil)
if err != nil {
return nil, err
}
ops[i] = o
}
e := &enc{}
if err := e.encodeIndirectBranch(mnem, ops); err != nil {
if err := e.encodeIndirectBranch(mnem, []Operand{o}); err != nil {
return nil, err
}
return e.out, nil
@@ -1245,31 +1422,51 @@ func operandFromAST(mnemUpper string, op *ast.Operand, size int, fi frameInfo, l
off := a.Sym.Offset + fi.fpAdjust
return Mem{Base: spReg, Disp: off, HasBase: true, Size: size}, nil
}
// SP-relative local: x-N(SP) → (spAdjust + offset)(SP).
// SP-relative local: x-N(SP) → (spAdjust + offset)(SP), keeping a scaled
// index beside the virtual stack pointer (foo(SP)(AX*1)). The
// symbol-pseudo parse returns before the index group, so the index
// is recovered from the raw text when the address lacks it.
if a.Sym != nil && a.Sym.Pseudo == "SP" && a.Base == "" {
off := fi.spAdjust + a.Sym.Offset
return Mem{Base: spReg, Disp: off, HasBase: true, Size: size}, nil
m := Mem{Base: spReg, Disp: off, HasBase: true, Size: size}
if name, scale, ok := trailingIndexGroup(op.Raw); ok {
idx, ok := ParseReg(name)
if !ok {
return nil, fmt.Errorf("unknown index register %q", name)
}
m.Index = idx
m.Scale = scale
m.HasIndex = true
}
return m, nil
}
// SB (global symbol): a symbol defined in the same file (GLOBL) is
// encoded RIP-relative and resolved by the file-level layout;
// anything not defined here needs object-file emission.
// anything not defined here needs object-file emission. A static
// (file-local) spelling of an undefined symbol defers the same way
// the toolchain does: the relocation names it and the linker decides.
if a.Sym != nil && a.Sym.Pseudo == "SB" {
if link == nil || link.symbols == nil {
return nil, fmt.Errorf("symbol %q needs file-level assembly (AssembleFile)", a.Sym.Name)
}
if !link.symbols[a.Sym.Name] {
if a.Sym.Static {
return nil, fmt.Errorf("undefined symbol %q", a.Sym.Name)
}
if !link.allowExternal {
if !link.symbols[a.Sym.Name] && !link.allowExternal {
return nil, fmt.Errorf("external symbol %q needs object-file emission", a.Sym.Name)
}
}
return sbMem{size: size, name: a.Sym.Name, addend: a.Sym.Offset}, nil
}
// Memory with a real base register: (base), off(base), (base)(index*scale).
if a.Base != "" {
// The TLS pseudo-base, off(TLS): the segment-prefixed absolute
// the thread-local access lowers to, 64 8B 04 25 with its
// R_TLS_LE patch site on the disp32.
if a.Base == "TLS" {
seg := byte(0x64) // FS on linux, freebsd, plan9
if link != nil && link.goos == "windows" {
seg = 0x65 // GS
}
return TLSMem{Disp: a.Offset, Size: size, Seg: seg}, nil
}
// Segment-absolute: 0x30(GS) and 0x28(FS), the windows TLS
// spellings. The segment override prefixes a disp32 absolute
// reference with no relocation.
+2 -2
View File
@@ -10,8 +10,8 @@ import (
"golang.org/x/arch/x86/x86asm"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// firstText parses src and returns its first TEXT function.
+1 -1
View File
@@ -8,7 +8,7 @@ import (
"encoding/binary"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// ulebIter reads ULEB128 values, the .debug_abbrev and line-header
+2 -2
View File
@@ -12,8 +12,8 @@ import (
"path/filepath"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// The object-file tests share one source: two exported functions, one
+1 -1
View File
@@ -9,7 +9,7 @@ import (
"encoding/binary"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// TestELFAARCH64Object checks the structure of the emitted AArch64 ELF64
+1 -1
View File
@@ -9,7 +9,7 @@ import (
"encoding/binary"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// TestELFLOONG64Object checks the structure of the emitted LoongArch ELF64
+1 -1
View File
@@ -9,7 +9,7 @@ import (
"encoding/binary"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// TestELFRISCVObjectDataRelocation checks that a symbol-valued DATA field
+13
View File
@@ -20,11 +20,24 @@ func Encodable(mnemonic string) bool {
switch upper {
case "RET", "NOP", "CALL", "JMP",
"POPFQ", "PUSHFQ", "INT", "LDMXCSR", "STMXCSR", "CMPSD", "SHA256RNDS2",
// The SSE compare family sharing CMPSD's predicate-last shape, the
// far return with its stack pop, the loop family, the bank-crossing
// MMX moves and the one-operand system controls.
"CMPSS", "CMPPS", "CMPPD", "RETFL",
"LOOP", "LOOPE", "LOOPNE",
"MOVDQ2Q", "MOVQ2DQ",
"ENDBR64", "CLWB", "TPAUSE", "UMONITOR", "UMWAIT", "RDPID", "CLDEMOTE",
// The literal-data pseudo-ops, the accepted-and-ignored END and
// bookkeeping statements, and the SP adjust.
"BYTE", "WORD", "LONG", "QUAD", "END", "ADJSP", "FUNCDATA", "PCDATA":
return true
}
if _, ok := sysUnaryTable[upper]; ok {
return true
}
if _, ok := sseStoreOnly[upper]; ok {
return true
}
if _, ok := noOperandTable[upper]; ok {
return true
}
+75 -3
View File
@@ -73,9 +73,23 @@ func (e *enc) encode(mnem string, ops []Operand) error {
// Fixed-name instructions (no size suffix).
switch {
case upper == "RET":
// RET sym(SB), the absolute return: the toolchain encodes it as a
// tail jump, E9 rel32 with a call relocation against the symbol.
if len(ops) == 1 {
if m, ok := ops[0].(sbMem); ok {
return e.emit(&instr{opcode: []byte{0xE9}, modrm: -1, sib: -1, disp: le32(0), sb: &sbRef{name: m.name, addend: m.addend}})
}
return fmt.Errorf("RET: unsupported operand")
}
if len(ops) != 0 {
return fmt.Errorf("RET expects no operands, got %d", len(ops))
}
return e.encodeRet()
case upper == "NOP":
return e.emit(&instr{opcode: []byte{0x90}, modrm: -1, sib: -1})
// The toolchain consumes every NOP statement as a pseudo and emits
// nothing for it, operands included (a bare NOP, NOP AX and
// NOP sym(SB) all vanish from the object).
return nil
case upper == "CALL" || upper == "JMP":
// Through a register or memory: FF /2 (CALL) or FF /4 (JMP).
// Anything else is a rel32 against a label resolved by the assembler.
@@ -102,6 +116,15 @@ func (e *enc) encode(mnem string, ops []Operand) error {
}
return e.emit(&instr{opcode: op, modrm: -1, sib: -1})
}
// One-operand system instructions whose reg field is a fixed digit:
// the cache and wait controls under 0F AE/0F 1C and the RDPID read.
if m, ok := sysUnaryTable[upper]; ok {
return e.encodeSysUnary(upper, m, ops)
}
// The store-only SSE moves (the non-temporal store).
if m, ok := sseStoreOnly[upper]; ok {
return e.encodeSSEStoreOnly(upper, m, ops)
}
// POPFQ/PUSHFQ are exact names: the bare POPF/PUSHF and the L spellings
// are rejected by go tool asm in 64-bit mode, so they stay unsupported.
switch upper {
@@ -117,14 +140,63 @@ func (e *enc) encode(mnem string, ops []Operand) error {
return e.emit(&instr{opcode: []byte{0x9C}, modrm: -1, sib: -1})
case "INT":
return e.encodeInt(ops)
// The LOOP family outside the assembler's label settlement: the operand
// is the already-computed rel8 (E0-E2).
case "LOOP", "LOOPE", "LOOPNE":
if len(ops) != 1 {
return fmt.Errorf("%s expects 1 operand, got %d", upper, len(ops))
}
imm, ok := ops[0].(Imm)
if !ok || !fits8(int64(imm)) {
return fmt.Errorf("%s: relative offset must be a signed byte", upper)
}
return e.emit(&instr{opcode: []byte{loopOpcode(upper)}, modrm: -1, sib: -1, imm: []byte{byte(int8(imm))}})
case "LDMXCSR":
return e.encodeMxcsr(2, ops)
case "STMXCSR":
return e.encodeMxcsr(3, ops)
// CMPSD is the scalar double compare, whose predicate immediate comes
// LAST in Plan 9 order (src, dst, $imm).
// LAST in Plan 9 order (src, dst, $imm); the family shares the shape.
case "CMPSD":
return e.encodeCmpsd(ops)
return e.encodeSSECmp("CMPSD", 0xF2, ops)
case "CMPSS":
return e.encodeSSECmp("CMPSS", 0xF3, ops)
case "CMPPS":
return e.encodeSSECmp("CMPPS", 0x00, ops)
case "CMPPD":
return e.encodeSSECmp("CMPPD", 0x66, ops)
// RETFL pops the immediate's worth of bytes after the far return
// (LRET iw: CA imm16), the toolchain's RETF spelling with a stack
// adjustment.
case "RETFL":
if len(ops) != 1 {
return fmt.Errorf("RETFL expects 1 operand, got %d", len(ops))
}
imm, ok := ops[0].(Imm)
if !ok {
return fmt.Errorf("RETFL expects an immediate")
}
return e.emit(&instr{opcode: []byte{0xCA}, modrm: -1, sib: -1, imm: le16(int64(imm))})
// MOVDQ2Q/MOVQ2DQ cross the MMX and XMM banks (F2 0F D6), the register
// in the reg field, the other bank's in r/m.
case "MOVDQ2Q", "MOVQ2DQ":
if len(ops) != 2 {
return fmt.Errorf("%s expects 2 operands, got %d", upper, len(ops))
}
srcReg, ok1 := ops[0].(Reg)
dstReg, ok2 := ops[1].(Reg)
if !ok1 || !ok2 {
return fmt.Errorf("%s takes register operands alone", upper)
}
if upper == "MOVDQ2Q" && (!srcReg.isVec() || !dstReg.mmx) ||
upper == "MOVQ2DQ" && (!srcReg.mmx || !dstReg.isVec()) {
return fmt.Errorf("%s crosses the XMM and MMX banks in that order", upper)
}
i := &instr{prefix: 0xF2, opcode: []byte{0x0F, 0xD6}, modrm: -1, sib: -1}
if err := setRM(i, dstReg, srcReg, 8); err != nil {
return err
}
return e.emit(i)
// SHA256RNDS2 carries the round constant in a literal X0 first operand.
case "SHA256RNDS2":
return e.encodeSha256rnds2(ops)
+158 -4
View File
@@ -10,8 +10,8 @@ import (
"golang.org/x/arch/x86/x86asm"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// decode encodes an instruction and decodes it back, returning the decoded
@@ -265,7 +265,11 @@ func TestImul(t *testing.T) {
func TestControl(t *testing.T) {
checkSyntax(t, "ret", "RET")
checkSyntax(t, "nop", "NOP")
// NOP contributes nothing on amd64, consumed whole by the toolchain as
// a pseudo; only the bytes pin it (no decodable instruction remains).
if code, err := Encode("NOP"); err != nil || len(code) != 0 {
t.Errorf("NOP: bytes %x (err %v), want empty", code, err)
}
checkOp(t, x86asm.JMP, "JMP", Imm(0))
checkOp(t, x86asm.CALL, "CALL", Imm(0))
checkOp(t, x86asm.JGE, "JGE", Imm(0))
@@ -1193,7 +1197,7 @@ func TestBookkeepingGroundTruth(t *testing.T) {
if err != nil {
t.Fatalf("assemble: %v", err)
}
want := "90c3"
want := "c3"
if got := hexCompact(img.Code); got != want {
t.Errorf("body %s, want %s (the bookkeeping lines contribute nothing)", got, want)
}
@@ -1221,3 +1225,153 @@ func mustParse(t *testing.T, src string) *ast.File {
}
return f
}
// TestCorpusTailSystem pins the system and control forms the toolchain's own
// amd64 testdata carries, byte for byte: the one-operand IMUL, the compare
// family, the far return, the loop, the MMX moves, the CR/DR and segment
// register moves, the TLS pseudo-base and the 0F AE/1C/C7 controls.
func TestCorpusTailSystem(t *testing.T) {
regBPT := Reg{idx: 5, size: 8}
X0, X1, X2 := vreg(t, "X0"), vreg(t, "X1"), vreg(t, "X2")
Y1, Y2, Y7 := vreg(t, "Y1"), vreg(t, "Y2"), vreg(t, "Y7")
X5, X20 := vreg(t, "X5"), vreg(t, "X20")
cases := []struct {
name string
mnem string
ops []Operand
want string
}{
{"IMUL one-op byte", "IMULB", []Operand{DX}, "f6ea"},
{"IMUL one-op long", "IMULL", []Operand{AX}, "f7e8"},
{"CMPPD", "CMPPD", []Operand{X1, X2, Imm(4)}, "660fc2d104"},
{"CMPSS", "CMPSS", []Operand{X1, X2, Imm(4)}, "f30fc2d104"},
{"CMPPS", "CMPPS", []Operand{X1, X2, Imm(4)}, "0fc2d104"},
{"RETFL", "RETFL", []Operand{Imm(4)}, "ca0400"},
{"LOOP", "LOOP", []Operand{Imm(-2)}, "e2fe"},
{"LOOPE", "LOOPE", []Operand{Imm(-2)}, "e1fe"},
{"LOOPNE", "LOOPNE", []Operand{Imm(-2)}, "e0fe"},
{"PADDD MMX", "PADDD", []Operand{Reg{idx: 2, size: 8, mmx: true}, Reg{idx: 1, size: 8, mmx: true}}, "0ffeca"},
{"MOVDQ2Q", "MOVDQ2Q", []Operand{X1, Reg{idx: 1, size: 8, mmx: true}}, "f20fd6c9"},
{"MOVNTDQ", "MOVNTDQ", []Operand{X1, Ptr(AX, 0, 16)}, "660fe708"},
{"MOVQ mmx load", "MOVQ", []Operand{Ptr(AX, 0, 8), Reg{idx: 0, size: 8, mmx: true}}, "0f6f00"},
{"MOVQ mmx store", "MOVQ", []Operand{Reg{idx: 0, size: 8, mmx: true}, Ptr(SI, 0, 8)}, "0f7f06"},
{"MOVQ CR0 load", "MOVQ", []Operand{Reg{idx: 0, size: 8, ctl: 1}, AX}, "0f20c0"},
{"MOVQ CR4 load", "MOVQ", []Operand{Reg{idx: 4, size: 8, ctl: 1}, DI}, "0f20e7"},
{"MOVQ CR0 store", "MOVQ", []Operand{AX, Reg{idx: 0, size: 8, ctl: 1}}, "0f22c0"},
{"MOVQ DR0 load", "MOVQ", []Operand{Reg{idx: 0, size: 8, ctl: 2}, AX}, "0f21c0"},
{"MOVQ DR7 load", "MOVQ", []Operand{Reg{idx: 7, size: 8, ctl: 2}, SI}, "0f21fe"},
{"PUSHQ FS", "PUSHQ", []Operand{Reg{idx: 4, size: 2, seg: 5}}, "0fa0"},
{"PUSHQ GS", "PUSHQ", []Operand{Reg{idx: 5, size: 2, seg: 6}}, "0fa8"},
{"POPQ FS", "POPQ", []Operand{Reg{idx: 4, size: 2, seg: 5}}, "0fa1"},
{"POPQ GS", "POPQ", []Operand{Reg{idx: 5, size: 2, seg: 6}}, "0fa9"},
{"ENDBR64", "ENDBR64", nil, "f30f1efa"},
{"CLWB", "CLWB", []Operand{Ptr(BX, 0, 8)}, "660fae33"},
{"CLDEMOTE", "CLDEMOTE", []Operand{Ptr(BX, 0, 8)}, "0f1c03"},
{"TPAUSE", "TPAUSE", []Operand{BX}, "660faef3"},
{"UMONITOR", "UMONITOR", []Operand{BX}, "f30faef3"},
{"UMWAIT", "UMWAIT", []Operand{BX}, "f20faef3"},
{"RDPID", "RDPID", []Operand{DX}, "f30fc7fa"},
{"RDPID r11", "RDPID", []Operand{Reg{idx: 11, size: 8}}, "f3410fc7fb"},
{"LEAL wide disp", "LEAL", []Operand{Idx(regBPT, Reg{idx: 10, size: 8}, 1, 0x8f1bbcdc, 8), regBPT}, "428dac15dcbc1b8f"},
{"VPERMPD", "VPERMPD", []Operand{Imm(0xd8), Y7, Y7}, "c4e3fd01ffd8"},
{"VPERMILPD", "VPERMILPD", []Operand{Imm(0xff), X1, X2}, "c4e37905d1ff"},
{"VPERMILPS", "VPERMILPS", []Operand{Imm(0xff), X1, X2}, "c4e37904d1ff"},
{"VROUNDPD", "VROUNDPD", []Operand{Imm(-1), X1, X2}, "c4e37909d1ff"},
{"VROUNDPS", "VROUNDPS", []Operand{Imm(-1), Y1, Y2}, "c4e37d08d1ff"},
{"VAESKEYGENASSIST", "VAESKEYGENASSIST", []Operand{Imm(-1), X1, X2}, "c4e379dfd1ff"},
{"VPCMPESTRI", "VPCMPESTRI", []Operand{Imm(-1), X1, X2}, "c4e37961d1ff"},
{"VPCMPESTRM", "VPCMPESTRM", []Operand{Imm(-1), X1, X2}, "c4e37960d1ff"},
{"VPCMPISTRI", "VPCMPISTRI", []Operand{Imm(-1), X1, X2}, "c4e37963d1ff"},
{"VPCMPISTRM", "VPCMPISTRM", []Operand{Imm(-1), X1, X2}, "c4e37962d1ff"},
{"VEXTRACTPS", "VEXTRACTPS", []Operand{Imm(-1), X1, AX}, "c4e37917c8ff"},
{"VPEXTRW", "VPEXTRW", []Operand{Imm(0xff), X1, AX}, "c4e37915c8ff"},
{"VPBLENDVB", "VPBLENDVB", []Operand{X0, Ptr(BX, 0, 16), X1, X2}, "c4e3714c1300"},
{"VMOVHPD load", "VMOVHPD", []Operand{Ptr(AX, 0, 8), X5, X5}, "c5d11628"},
{"VMOVHPD load disp", "VMOVHPD", []Operand{Ptr(DX, 7, 8), X5, X5}, "c5d1166a07"},
{"VMOVHPD store", "VMOVHPD", []Operand{X5, Ptr(AX, 0, 8)}, "c5f91728"},
{"VMOVLPD load", "VMOVLPD", []Operand{Ptr(AX, 0, 8), X5, X5}, "c5d11228"},
{"VMOVLPD store", "VMOVLPD", []Operand{X5, Ptr(AX, 0, 8)}, "c5f91328"},
{"VMOVQ EVEX gpr load", "VMOVQ", []Operand{Reg{idx: 4, size: 8}, X20}, "62e1fd086ee4"},
{"VMOVQ EVEX mem store", "VMOVQ", []Operand{X20, Ptr(AX, 0, 8)}, "62e1fd087e20"},
{"VMOVQ EVEX mem load", "VMOVQ", []Operand{Ptr(AX, 0, 8), X20}, "62e1fd086e20"},
}
for _, c := range cases {
code, err := Encode(c.mnem, c.ops...)
if err != nil {
t.Errorf("%s: Encode: %v", c.name, err)
continue
}
if got := hexCompact(code); got != c.want {
t.Errorf("%s: bytes %s, want %s", c.name, got, c.want)
}
}
}
// TestCorpusTailFileForms pins the file-level forms the toolchain's amd64
// testdata carries: the star-marked indirect jumps, the TLS pseudo-base, the
// paired-register shift spelling, the absolute RET and the jump to an
// undefined static symbol (its displacement and the TLS slot offsets are
// relocation sites, zeroed here as the kernel parity suites do).
func TestCorpusTailFileForms(t *testing.T) {
mask32 := func(b []byte, at int) { b[at], b[at+1], b[at+2], b[at+3] = 0, 0, 0, 0 }
cases := []struct {
name string
src string
want string // hex, with X marking a masked 32-bit relocation site
}{
{"star reg jump", "\tJMP *(R12)\n\tRET\n", "41ff2424c3"},
{"star sp jump", "\tJMP *4(SP)\n\tRET\n", "ff642404c3"},
{"star indexed jump", "\tJMP *(R12)(R13*4)\n\tRET\n", "43ff24acc3"},
{"TLS load", "\tMOVQ (TLS), AX\n\tRET\n", "64488b0425XXXXXXXXc3"},
{"TLS load offset", "\tMOVQ 8(TLS), DX\n\tRET\n", "64488b1425XXXXXXXXc3"},
{"colon shift", "\tSHLL CX, R11:AX\n\tRET\n", "410fa5c3c3"},
{"SP indexed local", "\tMOVQ foo(SP)(AX*1), BX\n\tRET\n", "488b1c04c3"},
}
for _, c := range cases {
f, errs := parser.Parse("t_amd64.s", "#include \"textflag.h\"\nTEXT ·f(SB), NOSPLIT, $0\n"+c.src)
if len(errs) > 0 {
t.Errorf("%s: parse: %v", c.name, errs)
continue
}
img, err := AssembleFile(f)
if err != nil {
t.Errorf("%s: assemble: %v", c.name, err)
continue
}
fn := img.Funcs[0]
code := append([]byte(nil), img.Code[fn.Offset:fn.Offset+fn.Size]...)
if at := strings.Index(c.want, "XXXXXXXX"); at >= 0 {
mask32(code, at/2) // the masked relocation site
}
if got := hexCompact(code); got != strings.ReplaceAll(c.want, "X", "0") {
t.Errorf("%s: bytes %s, want %s", c.name, got, c.want)
}
}
// JMP to an undefined static symbol and the absolute RET: their rel32
// carries a call relocation against the symbol, masked to zero here.
for _, c := range []struct{ name, src string }{
{"static external jump", "\tJMP bar<>+4(SB)\n\tRET\n"},
{"static external indexed jump", "\tJMP bar<>+4(SB)(R11*4)\n\tRET\n"},
{"absolute ret", "\tRET\n\tRET foo(SB)\n"},
} {
f, errs := parser.Parse("t_amd64.s", "#include \"textflag.h\"\nTEXT ·f(SB), NOSPLIT, $0\n"+c.src)
if len(errs) > 0 {
t.Errorf("%s: parse: %v", c.name, errs)
continue
}
img, err := AssembleFile(f)
if err != nil {
t.Errorf("%s: assemble: %v", c.name, err)
continue
}
fn := img.Funcs[0]
code := maskCode(append([]byte(nil), img.Code[fn.Offset:fn.Offset+fn.Size]...), fn.Relocs)
want := "e900000000c3"
if c.name == "absolute ret" {
want = "c3e900000000"
}
if got := hexCompact(code); got != want {
t.Errorf("%s: bytes %s, want %s", c.name, got, want)
}
}
}
+66 -17
View File
@@ -856,37 +856,41 @@ type evexMoveSpec struct {
vecOK bool // the non-memory operand may be a vector register
xmmOnly bool // wider than XMM registers are rejected
nds3 bool // a three-operand register form exists (VMOVSD/VMOVSS)
gprOK bool // the r/m side may be a general-purpose register (VMOVQ)
}
// evexMoveTable maps an upper-case EVEX move mnemonic to its encoding.
var evexMoveTable = map[string]evexMoveSpec{
// EVEX.128/256/512.F3.0F.W0, unaligned integer move.
"VMOVDQU32": {1, 2, 0x6F, 0x7F, 0, [3]int{16, 32, 64}, true, false, false},
"VMOVDQU32": {1, 2, 0x6F, 0x7F, 0, [3]int{16, 32, 64}, true, false, false, false},
// EVEX.128/256/512.F3.0F.W1, unaligned qword move.
"VMOVDQU64": {1, 2, 0x6F, 0x7F, 1, [3]int{16, 32, 64}, true, false, false},
"VMOVDQU64": {1, 2, 0x6F, 0x7F, 1, [3]int{16, 32, 64}, true, false, false, false},
// EVEX.128/256/512.F2.0F.W0, unaligned byte move (byte/word moves use the
// F2 prefix, dword/qword moves F3; the element size only changes the tuple
// semantics).
"VMOVDQU8": {1, 3, 0x6F, 0x7F, 0, [3]int{16, 32, 64}, true, false, false},
"VMOVDQU8": {1, 3, 0x6F, 0x7F, 0, [3]int{16, 32, 64}, true, false, false, false},
// EVEX.128/256/512.F2.0F.W1, unaligned word move (shares the qword
// encoding).
"VMOVDQU16": {1, 3, 0x6F, 0x7F, 1, [3]int{16, 32, 64}, true, false, false},
"VMOVDQU16": {1, 3, 0x6F, 0x7F, 1, [3]int{16, 32, 64}, true, false, false, false},
// EVEX.128/256/512.66.0F.W1, unaligned packed double move.
"VMOVUPD": {1, 1, 0x10, 0x11, 1, [3]int{16, 32, 64}, true, false, false},
"VMOVUPD": {1, 1, 0x10, 0x11, 1, [3]int{16, 32, 64}, true, false, false, false},
// EVEX.128/256/512, aligned packed moves.
"VMOVAPS": {1, 0, 0x28, 0x29, 0, [3]int{16, 32, 64}, true, false, false},
"VMOVAPD": {1, 1, 0x28, 0x29, 1, [3]int{16, 32, 64}, true, false, false},
"VMOVAPS": {1, 0, 0x28, 0x29, 0, [3]int{16, 32, 64}, true, false, false, false},
"VMOVAPD": {1, 1, 0x28, 0x29, 1, [3]int{16, 32, 64}, true, false, false, false},
// EVEX.128/256/512.66.0F, aligned integer moves.
"VMOVDQA32": {1, 1, 0x6F, 0x7F, 0, [3]int{16, 32, 64}, true, false, false},
"VMOVDQA64": {1, 1, 0x6F, 0x7F, 1, [3]int{16, 32, 64}, true, false, false},
"VMOVDQA32": {1, 1, 0x6F, 0x7F, 0, [3]int{16, 32, 64}, true, false, false, false},
"VMOVDQA64": {1, 1, 0x6F, 0x7F, 1, [3]int{16, 32, 64}, true, false, false, false},
// EVEX.128.F3.0F.W0, scalar single move, memory operands (the
// three-operand register form is not supported).
"VMOVSS": {1, 2, 0x10, 0x11, 0, [3]int{4, 4, 4}, false, true, true},
"VMOVSS": {1, 2, 0x10, 0x11, 0, [3]int{4, 4, 4}, false, true, true, false},
// EVEX.128.F2.0F.W1, scalar double move: memory operands and the
// three-operand register form (VMOVSD dst, src1, src2).
"VMOVSD": {1, 3, 0x10, 0x11, 1, [3]int{8, 8, 8}, false, true, true},
"VMOVSD": {1, 3, 0x10, 0x11, 1, [3]int{8, 8, 8}, false, true, true, false},
// EVEX.128/256/512.0F.W0, unaligned packed single move.
"VMOVUPS": {1, 0, 0x10, 0x11, 0, [3]int{16, 32, 64}, true, false, false},
"VMOVUPS": {1, 0, 0x10, 0x11, 0, [3]int{16, 32, 64}, true, false, false, false},
// EVEX.128.66.0F.W1, the 64-bit GPR/memory ↔ XMM move (VMOVQ RSP, X20
// and friends, the EVEX spelling the high registers demand).
"VMOVQ": {1, 1, 0x6E, 0x7E, 1, [3]int{8, 8, 8}, true, true, false, true},
}
// isEvex reports whether the mnemonic has an EVEX encoding we handle.
@@ -900,6 +904,9 @@ func isEvex(mnemUpper string) bool {
if _, ok := evexMoveTable[mnemUpper]; ok {
return true
}
if _, ok := evexHptrTable[mnemUpper]; ok {
return true
}
return isEvexQuad(mnemUpper)
}
@@ -911,8 +918,14 @@ func evexRequired(upper string, ops []Operand) bool {
_, inVex := vexTable[upper]
_, inVexMove := vexMoveTable[upper]
if !inVex && !inVexMove {
// The dual-shape moves pick their VEX form by operand count, so
// they are not EVEX-only either.
switch upper {
case "VMOVHPD", "VMOVLPD":
default:
return true // EVEX-only mnemonic
}
}
// The byte-quad shifts have VEX register forms but EVEX-only memory
// forms: a memory count source forces the EVEX encoding.
if upper == "VPSLLDQ" || upper == "VPSRLDQ" {
@@ -1018,6 +1031,8 @@ var evexRound = map[string]bool{
"VCVTTSD2USIL": true, "VCVTTSD2USIQ": true, "VCVTTSS2USIL": true, "VCVTTSS2USIQ": true,
"VCVTSI2SDQ": true, "VCVTSI2SSL": true, "VCVTSI2SSQ": true,
"VCVTUSI2SDQ": true, "VCVTUSI2SSL": true, "VCVTUSI2SSQ": true,
// The scalar compares suppress exceptions on their LIG encoding.
"VCMPSD": true, "VCMPSS": true,
}
// evexBcstN maps an instruction accepting .BCST to the broadcast element
@@ -1082,6 +1097,14 @@ func (e *enc) encodeEvex(mnemUpper string, ops []Operand, sfx evexSuffix) error
return e.encodeEvexRM(spec, ops, 0, sfx)
}
spec, inTable := evexTable[mnemUpper]
// A high/low half move that lives in the hptr table alone (the packed
// double twins) reaches the same inTable block below, which completes
// its spec from the hptr entry.
if !inTable {
if _, ok := evexHptrTable[mnemUpper]; ok {
inTable = true
}
}
if q, ok := evexQuadTable[mnemUpper]; ok {
// The quad-register family carries no rounding, SAE or broadcast;
// only masking and zeroing apply.
@@ -1446,6 +1469,19 @@ func (e *enc) encodeEvexExtractGPR(spec evexSpec, ops []Operand, mask int, sfx e
// assembler. The scalar moves also carry a three-operand register form
// (VMOVSD dst, src1, src2: the load opcode with vvvv = src1), which ms.nds3
// opens.
// validEvexMoveOther reports whether the non-vector side of an EVEX move may
// take the operand: memory always, a general-purpose register when gprOK.
func validEvexMoveOther(ms evexMoveSpec, op Operand) bool {
if memOperand(op) {
return true
}
if !ms.gprOK {
return false
}
r, ok := op.(Reg)
return ok && !r.isVec() && !r.mask && r.ctl == 0 && !r.mmx && !r.fp
}
func (e *enc) encodeEvexMove(mnem string, ms evexMoveSpec, ops []Operand, mask int, sfx evexSuffix) error {
if len(ops) == 3 {
if !ms.nds3 {
@@ -1453,7 +1489,7 @@ func (e *enc) encodeEvexMove(mnem string, ms evexMoveSpec, ops []Operand, mask i
}
// The masked scalar register form keeps the Go assembler's own
// layout: the store opcode with reg = op0, vvvv = op1 and the
// destination in r/m (op2) — the bytes go tool asm emits, not
// destination in r/m (op2), the bytes go tool asm emits, not
// the manual's NDS reading.
src, src1, dst := ops[0], ops[1], ops[2]
reg, ok := src.(Reg)
@@ -1494,12 +1530,12 @@ func (e *enc) encodeEvexMove(mnem string, ms evexMoveSpec, ops []Operand, mask i
}
reg, rm = srcReg, dst
case srcIsVec:
if !memOperand(dst) {
if !validEvexMoveOther(ms, dst) {
return fmt.Errorf("%s: invalid destination operand", mnem)
}
reg, rm = srcReg, dst
case dstIsVec:
if !memOperand(src) {
if !validEvexMoveOther(ms, src) {
return fmt.Errorf("%s: invalid source operand", mnem)
}
op = ms.load
@@ -1890,6 +1926,15 @@ var evexHptrTable = map[string]evexHptrSpec{
"VMOVLHPS": {
insert: evexSpec{mapSel: 1, opcode: 0x16, w: 0, pp: 0, opdigit: -1, form: vexNDS3, n: [3]int{8, 0, 0}},
},
// The packed-double twins, 66-prefixed.
"VMOVHPD": {
insert: evexSpec{mapSel: 1, opcode: 0x16, w: 1, pp: 1, opdigit: -1, form: vexNDS3, n: [3]int{8, 0, 0}},
store: evexSpec{mapSel: 1, opcode: 0x17, w: 1, pp: 1, opdigit: -1, form: vexRMRev, n: [3]int{8, 0, 0}},
},
"VMOVLPD": {
insert: evexSpec{mapSel: 1, opcode: 0x12, w: 1, pp: 1, opdigit: -1, form: vexNDS3, n: [3]int{8, 0, 0}},
store: evexSpec{mapSel: 1, opcode: 0x13, w: 1, pp: 1, opdigit: -1, form: vexRMRev, n: [3]int{8, 0, 0}},
},
}
// encodeEvexPrefGather encodes a gather/scatter prefetch hint: OP K, vsib.
@@ -1954,7 +1999,7 @@ func (e *enc) encodeGather(upper string, gs gatherSpec, ops []Operand, sfx evexS
if !ok || !maskReg.isVec() {
return fmt.Errorf("%s: mask must be a vector register", upper)
}
vsib, _, err := vsibLen(rest[1], upper)
vsib, idxLen, err := vsibLen(rest[1], upper)
if err != nil {
return err
}
@@ -1962,12 +2007,16 @@ func (e *enc) encodeGather(upper string, gs gatherSpec, ops []Operand, sfx evexS
if !ok || !dst.isVec() {
return fmt.Errorf("%s: destination must be a vector register", upper)
}
// The L bit is the wider of the data register and the VSIB index
// lengths (a YMM index under an XMM destination selects 256-bit, the
// bytes go tool asm emits).
ll := max(idxLen, dst.vecLenBit())
spec := vexSpec{mapSel: 2, opcode: gs.opcode, w: gs.w, pp: 1, opdigit: -1}
rBit := 0
if dst.idx >= 8 {
rBit = 1
}
return e.emitVexFields(spec, dst.vecLenBit(), dst.idx&7, rBit, 15-maskReg.idx, vsib)
return e.emitVexFields(spec, ll, dst.idx&7, rBit, 15-maskReg.idx, vsib)
}
// encodeScatter encodes a scatter (EVEX only): OP src, K, vsib, reg = src,
+8
View File
@@ -63,6 +63,7 @@ const (
const (
kindSTEXT = 1
kindSRODATA = 3
kindSNOPTRDATA = 5
kindSDATA = 7
kindSDWARFFCN = 14
kindSDWARFLINES = 20
@@ -305,9 +306,16 @@ func (img *Image) emitGOObject(pkgPath, srcPath string, pre []byte, minLC int, r
if !d.Static {
name = pkgPath + "." + name
}
// RODATA implies no pointers, so it wins over NOPTR: the kind is
// SRODATA either way, exactly as the toolchain chooses it. Plain
// NOPTR data is SNOPTRDATA, which the linker keeps out of the GC's
// type scan; a plain SDATA symbol would demand Go type information
// no assembly file can supply, and the link would fail.
typ := uint8(kindSDATA)
if d.Rodata {
typ = kindSRODATA
} else if d.Noptr {
typ = kindSNOPTRDATA
}
flag := uint8(0)
if d.Dupok {
+3
View File
@@ -45,6 +45,9 @@ func TestReadRuntimeSymbols(t *testing.T) {
// TestResolveExternalSymbols verifies end-to-end resolution of external
// symbol references.
func TestResolveExternalSymbols(t *testing.T) {
if testing.Short() {
t.Skip("resolves through a live go list -export: skipped in -short mode")
}
if _, err := exec.LookPath("go"); err != nil {
t.Skip("go toolchain not available")
}
+143 -1
View File
@@ -12,7 +12,7 @@ import (
"strings"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// goobjView is a minimal parsed view of a GOOBJ payload, enough to check
@@ -514,6 +514,145 @@ func fieldAfter(line, flag string) string {
return ""
}
// buildLogSteps is what a substitution test needs from a `go build -x -work`
// log: the work directory, the assembler's object, the package archive and
// the link command line.
type buildLogSteps struct {
work string
asmObj string // $WORK expanded
pkgArch string // $WORK expanded
linkLine string // still carries $WORK placeholders
}
// parseBuildLog extracts the build steps from a `go build -x -work` log.
// asmFile names the assembly file whose object the test substitutes. A
// missing step is a failure, not a skip: the toolchain changed shape and the
// substitution would silently test nothing.
func parseBuildLog(t *testing.T, log []byte, asmFile string) buildLogSteps {
t.Helper()
var st buildLogSteps
for line := range strings.SplitSeq(string(log), "\n") {
switch {
case strings.HasPrefix(line, "WORK="):
st.work = strings.TrimPrefix(line, "WORK=")
case strings.Contains(line, "/asm ") && strings.Contains(line, asmFile) && !strings.Contains(line, "-gensymabis"):
st.asmObj = fieldAfter(line, "-o")
case strings.Contains(line, "pack r") && strings.Contains(line, "_pkg_.a"):
rest := strings.TrimSpace(strings.SplitN(line, "pack r", 2)[1])
st.pkgArch = strings.Fields(strings.SplitN(rest, "#", 2)[0])[0]
case strings.Contains(line, "/link ") && strings.Contains(line, "-importcfg"):
st.linkLine = line
}
}
if st.work == "" || st.asmObj == "" || st.pkgArch == "" || st.linkLine == "" {
t.Fatalf("could not locate the build steps (work=%q asmObj=%q pkgArch=%q link=%q):\n%s",
st.work, st.asmObj, st.pkgArch, st.linkLine, log)
}
st.asmObj = strings.ReplaceAll(st.asmObj, "$WORK", st.work)
st.pkgArch = strings.ReplaceAll(st.pkgArch, "$WORK", st.work)
return st
}
// substituteAndRelink swaps the gasm object into the package archive the
// baseline build produced and re-runs the captured link line against the
// rebuilt archive, writing the binary to outBin (the -x log's link step
// always targets the action graph's internal a.out, which the helper
// redirects; the copy to the -o target is a separate build action the helper
// does not need). The archive handed to the linker is proven to carry the
// gasm object byte for byte, so a build-layout change that skipped the
// substitution fails here instead of passing vacuously.
func substituteAndRelink(t *testing.T, goBin, dir string, st buildLogSteps, outBin string, gasmObj []byte, extraEnv ...string) {
t.Helper()
// The deliberate-run boundary: this path drives a real `go build` and
// cmd/link per invocation, minutes-scale work on the small single-core
// CI runner. Under -short (the push pipeline's mode) it skips; the
// local test gate and the dispatched workflows run it in full.
if testing.Short() {
t.Skip("end-to-end go build and link: skipped in -short mode")
}
// Extract the archive, overwrite the assembler's member with the gasm
// object and repack (go tool pack has no replace-in-place).
membersDir := filepath.Join(dir, "members")
if err := os.MkdirAll(membersDir, 0o755); err != nil {
t.Fatal(err)
}
extract := exec.Command(goBin, "tool", "pack", "x", st.pkgArch)
extract.Dir = membersDir
if out, err := extract.CombinedOutput(); err != nil {
t.Fatalf("pack x: %v\n%s", err, out)
}
member := filepath.Join(membersDir, filepath.Base(st.asmObj))
if _, err := os.Stat(member); err != nil {
t.Fatalf("the assembler's archive member was not extracted: %v", err)
}
if err := os.Chmod(member, 0o644); err != nil {
t.Fatal(err)
}
if err := os.WriteFile(member, gasmObj, 0o644); err != nil {
t.Fatal(err)
}
listCmd := exec.Command(goBin, "tool", "pack", "t", st.pkgArch)
listOut, err := listCmd.CombinedOutput()
if err != nil {
t.Fatalf("pack t: %v\n%s", err, listOut)
}
newArch := filepath.Join(dir, "pkg.a")
args := []string{"tool", "pack", "c", newArch}
seen := map[string]bool{}
for m := range strings.FieldsSeq(string(listOut)) {
if seen[m] {
continue
}
seen[m] = true
if err := os.Chmod(filepath.Join(membersDir, m), 0o644); err != nil {
t.Fatal(err)
}
args = append(args, m)
}
pack := exec.Command(goBin, args...)
pack.Dir = membersDir
if out, err := pack.CombinedOutput(); err != nil {
t.Fatalf("pack c: %v\n%s", err, out)
}
// Prove the substitution: the archive the linker is about to consume
// holds the gasm object, byte for byte.
checkDir := filepath.Join(dir, "check")
if err := os.MkdirAll(checkDir, 0o755); err != nil {
t.Fatal(err)
}
check := exec.Command(goBin, "tool", "pack", "x", newArch)
check.Dir = checkDir
if out, err := check.CombinedOutput(); err != nil {
t.Fatalf("pack x (verification): %v\n%s", err, out)
}
got, err := os.ReadFile(filepath.Join(checkDir, filepath.Base(st.asmObj)))
if err != nil {
t.Fatalf("read the substituted member back: %v", err)
}
if !bytes.Equal(got, gasmObj) {
t.Fatal("the repacked archive does not carry the gasm object")
}
// Re-link. The line carries a GOROOT assignment and $WORK placeholders;
// GOEXPERIMENT must match the toolchain's own, because the linker
// compares the object header against its configuration.
goExp, _ := exec.Command(goBin, "env", "GOEXPERIMENT").Output()
linkLine := strings.ReplaceAll(st.linkLine, "$WORK", st.work)
linkLine = strings.ReplaceAll(linkLine, filepath.Join(st.work, "b001", "_pkg_.a"), newArch)
linkLine = strings.ReplaceAll(linkLine, filepath.Join(st.work, "b001", "exe", "a.out"), outBin)
env := append(os.Environ(), "GOEXPERIMENT="+strings.TrimSpace(string(goExp)))
env = append(env, extraEnv...)
link := exec.Command("sh", "-c", linkLine)
link.Dir = dir
link.Env = env
if out, err := link.CombinedOutput(); err != nil {
t.Fatalf("link with the gasm object: %v\n%s", err, out)
}
}
// TestGOObjectExternalPackageLink is the cross-package end-to-end check: a
// GOOBJ whose code references a real external package symbol (runtime's
// morestack, a plain reference rather than the builtin noctxt form) must
@@ -524,6 +663,9 @@ func fieldAfter(line, flag string) string {
// failed. The binary is not run: morestack returns to the call site's
// stack check, which a hand-written caller has none of.
func TestGOObjectExternalPackageLink(t *testing.T) {
if testing.Short() {
t.Skip("end-to-end go build and link: skipped in -short mode")
}
goBin, err := exec.LookPath("go")
if err != nil {
t.Skip("no Go toolchain available")
+2 -2
View File
@@ -9,8 +9,8 @@ import (
"strings"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// The expected bytes are pinned from `go tool asm` output (Go 1.27, amd64,
+200 -7
View File
@@ -90,6 +90,40 @@ var noOperandTable = map[string][]byte{
"LOCK": {0xF0},
"REP": {0xF3},
"REPN": {0xF2},
"ENDBR64": {0xF3, 0x0F, 0x1E, 0xFA},
}
// sysUnaryTable maps the one-operand system instructions to their bytes:
// the prefix, the opcode and the /digit the reg field carries. The operand
// is a register or memory in r/m.
var sysUnaryTable = map[string]struct {
prefix byte
opcode []byte
digit int
}{
"CLWB": {0x66, []byte{0x0F, 0xAE}, 6},
"TPAUSE": {0x66, []byte{0x0F, 0xAE}, 6},
"UMONITOR": {0xF3, []byte{0x0F, 0xAE}, 6},
"UMWAIT": {0xF2, []byte{0x0F, 0xAE}, 6},
"RDPID": {0xF3, []byte{0x0F, 0xC7}, 7},
"CLDEMOTE": {0x00, []byte{0x0F, 0x1C}, 0},
}
// encodeSysUnary emits a one-operand system instruction: the operand in r/m
// under the fixed /digit, no REX.W.
func (e *enc) encodeSysUnary(mnem string, m struct {
prefix byte
opcode []byte
digit int
}, ops []Operand) error {
if len(ops) != 1 {
return fmt.Errorf("%s expects 1 operand, got %d", mnem, len(ops))
}
i := &instr{prefix: m.prefix, opcode: m.opcode, modrm: -1, sib: -1}
if err := setRMDigit(i, m.digit, ops[0], 8); err != nil {
return err
}
return e.emit(i)
}
// --- MOV --------------------------------------------------------------------
@@ -111,6 +145,76 @@ func (e *enc) encodeMov(ops []Operand, size int) error {
// silently emit REX.W 8B with the wrong operand meaning.
_, srcVec := vecReg(src)
dstReg, dstVec := vecReg(dst)
// Control and debug register moves: 0F 20 (CRn→r64), 0F 22 (r64→CRn),
// 0F 21 (DRn→r64) and 0F 23 (r64→DRn). The CR/DR number rides the reg
// field, the general register r/m; CR8+/DR8+ take REX.R.
if c, ok := src.(Reg); ok && c.ctl != 0 {
g, ok := dst.(Reg)
if !ok || g.isVec() || g.ctl != 0 {
return fmt.Errorf("MOV: control/debug register load needs a general register destination")
}
opc := byte(0x20)
if c.ctl == 2 {
opc = 0x21
}
return e.emit(&instr{
opcode: []byte{0x0F, opc},
modrm: 0xC0 | (c.idx&7)<<3 | (g.idx & 7),
sib: -1, rexR: c.idx >= 8, rexB: g.idx >= 8,
})
}
if c, ok := dst.(Reg); ok && c.ctl != 0 {
g, ok := src.(Reg)
if !ok || g.isVec() || g.ctl != 0 {
return fmt.Errorf("MOV: control/debug register store needs a general register source")
}
opc := byte(0x22)
if c.ctl == 2 {
opc = 0x23
}
return e.emit(&instr{
opcode: []byte{0x0F, opc},
modrm: 0xC0 | (c.idx&7)<<3 | (g.idx & 7),
sib: -1, rexR: c.idx >= 8, rexB: g.idx >= 8,
})
}
// MMX register moves: MOVQ M0, mem and MOVQ mem, M0 are the MMX
// load/store pair 0F 6F/0F 7F (no prefix); a register pair takes the
// load opcode. The XMM MOVQ forms follow below.
if m, ok := src.(Reg); ok && m.mmx {
switch d := dst.(type) {
case Reg:
if !d.mmx {
return fmt.Errorf("MOV: MMX register moves stay inside the M bank")
}
i := &instr{opcode: []byte{0x0F, 0x6F}, modrm: -1, sib: -1}
if err := setRM(i, d, src, 8); err != nil {
return err
}
return e.emit(i)
case Mem:
i := &instr{opcode: []byte{0x0F, 0x7F}, modrm: -1, sib: -1}
if err := setRM(i, m, d, 8); err != nil {
return err
}
return e.emit(i)
}
return fmt.Errorf("MOV: invalid MMX destination")
}
if m, ok := dst.(Reg); ok && m.mmx {
srcM, ok := src.(Mem)
if !ok {
return fmt.Errorf("MOV: MMX load takes a memory source")
}
i := &instr{opcode: []byte{0x0F, 0x6F}, modrm: -1, sib: -1}
if err := setRM(i, m, srcM, 8); err != nil {
return err
}
return e.emit(i)
}
if srcVec || dstVec {
if dstVec {
if g, ok := src.(Reg); ok && !g.isVec() {
@@ -518,6 +622,14 @@ func (e *enc) encodeLea(ops []Operand, size int) error {
default:
return fmt.Errorf("LEA: source must be a memory operand")
}
// LEA accepts the full unsigned 32-bit displacement span where the
// loads and stores reject it beyond the signed one; the wide values
// ride the same disp32 bytes as their two's-complement bit pattern.
if m, ok := src.(Mem); ok && m.Disp >= 1<<31 && m.Disp <= (1<<32)-1 {
c := m
c.Disp = int64(int32(uint32(m.Disp)))
src = c
}
i := newInstr(size, []byte{0x8D})
if err := setRM(i, dstReg, src, size); err != nil {
return err
@@ -663,6 +775,18 @@ func (e *enc) encodeDoubleShift(base string, ops []Operand, size int) error {
func (e *enc) encodeImul(ops []Operand, size int) error {
switch len(ops) {
case 1:
// The one-operand form, IMUL r/m: F6/F7 /5 with AL/AX/EAX/RAX as the
// implied destination (the toolchain's one-register shape).
opc := byte(0xF7)
if size == 1 {
opc = 0xF6
}
i := newInstr(size, []byte{opc})
if err := setRMDigit(i, 5, ops[0], size); err != nil {
return err
}
return e.emit(i)
case 2:
// Two shapes. The leading-immediate spelling IMUL $imm, r multiplies
// r in place (dst = rm = r): the shape GOROOT's clock code writes.
@@ -697,7 +821,7 @@ func (e *enc) encodeImul(ops []Operand, size int) error {
// r/m operand (setRM takes registers and memory alike).
return e.encodeImulImm(imm, ops[1], dstReg, size)
}
return fmt.Errorf("IMUL expects 2 or 3 operands, got %d", len(ops))
return fmt.Errorf("IMUL expects 1, 2 or 3 operands, got %d", len(ops))
}
// encodeImulImm emits the immediate multiply: 0x6B with a sign-extended imm8
@@ -742,6 +866,27 @@ func (e *enc) encodePushPop(ops []Operand, size int, push bool) error {
w16 := size == 2
switch op := ops[0].(type) {
case Reg:
// Segment registers: FS and GS carry their own one-byte opcodes
// under 0F (A0/A8 push, A1/A9 pop); the other four spellings are
// not pushable in 64-bit mode.
if n, isSeg := op.segNumber(); isSeg {
switch n {
case 4: // FS
if push {
return e.emit(&instr{opcode: []byte{0x0F, 0xA0}, modrm: -1, sib: -1})
}
return e.emit(&instr{opcode: []byte{0x0F, 0xA1}, modrm: -1, sib: -1})
case 5: // GS
if push {
return e.emit(&instr{opcode: []byte{0x0F, 0xA8}, modrm: -1, sib: -1})
}
return e.emit(&instr{opcode: []byte{0x0F, 0xA9}, modrm: -1, sib: -1})
}
return fmt.Errorf("PUSH/POP: only FS and GS are encodable in 64-bit mode")
}
if op.mmx || op.isVec() || op.fp || op.ctl != 0 {
return fmt.Errorf("PUSH/POP: invalid register operand")
}
base := byte(0x50) // PUSH r; POP is 0x58
if !push {
base = 0x58
@@ -1093,6 +1238,38 @@ var sseMoveTable = map[string]sseMove{
"MOVSS": {0xF3, 0x10, 0x11}, // scalar single
}
// sseStoreOnly holds the store-only SSE forms, OP xmm, mem: the XMM register
// rides the reg field and memory r/m (the non-temporal store).
var sseStoreOnly = map[string]struct {
prefix byte
op byte
}{
"MOVNTDQ": {0x66, 0xE7},
}
// encodeSSEStoreOnly encodes OP xmm, mem (reg = the XMM source, r/m = the
// destination memory).
func (e *enc) encodeSSEStoreOnly(mnem string, m struct {
prefix byte
op byte
}, ops []Operand) error {
if len(ops) != 2 {
return fmt.Errorf("%s expects 2 operands, got %d", mnem, len(ops))
}
srcReg, ok := ops[0].(Reg)
if !ok || !srcReg.isVec() {
return fmt.Errorf("%s source must be a vector register", mnem)
}
if !isX86Mem(ops[1]) {
return fmt.Errorf("%s destination must be a memory operand", mnem)
}
i := &instr{prefix: m.prefix, opcode: []byte{0x0F, m.op}, modrm: -1, sib: -1}
if err := setRM(i, srcReg, ops[1], 8); err != nil {
return err
}
return e.emit(i)
}
// encodeSSEMove encodes a legacy SSE move: a vector-to-vector move uses the
// load form (reg = destination), matching the Go assembler.
func (e *enc) encodeSSEMove(m sseMove, ops []Operand) error {
@@ -1308,14 +1485,23 @@ func (e *enc) encodeSSEBin(m sseBin, ops []Operand) error {
}
src, dst := ops[0], ops[1]
dstReg, ok := dst.(Reg)
if !ok || !dstReg.isVec() {
if !ok || (!dstReg.isVec() && !dstReg.mmx) {
return fmt.Errorf("SSE binary destination must be a vector register")
}
// The MMX twins of the packed-integer SSE2 ops drop the 0x66 prefix:
// PADDD M2, M1 is 0F FE where the XMM form is 66 0F FE.
prefix := m.prefix
if dstReg.mmx {
if prefix != 0x66 {
return fmt.Errorf("SSE binary: this form takes no MMX register operand")
}
prefix = 0
}
opcode := []byte{0x0F, m.op}
if m.map38 {
opcode = []byte{0x0F, 0x38, m.op}
}
i := &instr{prefix: m.prefix, opcode: opcode, modrm: -1, sib: -1}
i := &instr{prefix: prefix, opcode: opcode, modrm: -1, sib: -1}
if err := setRM(i, dstReg, src, 8); err != nil {
return err
}
@@ -1791,12 +1977,19 @@ func (e *enc) encodeSSEShift(name string, ops []Operand) error {
// immediate LAST in Plan 9 order (src, dst, $imm), unlike the shuffle family:
// F2 0F C2 with reg = dst, rm = src.
func (e *enc) encodeCmpsd(ops []Operand) error {
return e.encodeSSECmp("CMPSD", 0xF2, ops)
}
// encodeSSECmp encodes the SSE compare family (CMPSD/CMPSS/CMPPS/CMPPD):
// 0F C2 /r ib with the predicate immediate last in Plan 9 order
// (src, dst, $imm) and the packed forms' prefixes.
func (e *enc) encodeSSECmp(mnem string, prefix byte, ops []Operand) error {
if len(ops) != 3 {
return fmt.Errorf("CMPSD expects 3 operands (src, dst, $imm), got %d", len(ops))
return fmt.Errorf("%s expects 3 operands (src, dst, $imm), got %d", mnem, len(ops))
}
imm, ok := ops[2].(Imm)
if !ok {
return fmt.Errorf("CMPSD predicate must be an immediate")
return fmt.Errorf("%s predicate must be an immediate", mnem)
}
immByte, err := imm8(int64(imm))
if err != nil {
@@ -1804,9 +1997,9 @@ func (e *enc) encodeCmpsd(ops []Operand) error {
}
dstReg, ok2 := ops[1].(Reg)
if !ok2 || !dstReg.isVec() {
return fmt.Errorf("CMPSD destination must be a vector register")
return fmt.Errorf("%s destination must be a vector register", mnem)
}
i := &instr{prefix: 0xF2, opcode: []byte{0x0F, 0xC2}, modrm: -1, sib: -1}
i := &instr{prefix: prefix, opcode: []byte{0x0F, 0xC2}, modrm: -1, sib: -1}
if err := setRM(i, dstReg, ops[0], 8); err != nil {
return err
}
+3 -3
View File
@@ -16,7 +16,7 @@ import (
"golang.org/x/arch/x86/x86asm"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// TestAssembleGoFlacAVX2Kernel assembles the whole production AVX2 kernel;
@@ -25,7 +25,7 @@ import (
func TestAssembleGoFlacAVX2Kernel(t *testing.T) {
path := "../../go-libraries/go-flac/avx2_amd64.s"
if _, err := os.Stat(path); err != nil {
t.Skip("go-libraries repository not present next to gasm-devkit")
t.Skip("go-libraries repository not present next to gasm-sdk")
}
src, err := os.ReadFile(path)
if err != nil {
@@ -86,7 +86,7 @@ func TestAssembleGoFlacAVX2Kernel(t *testing.T) {
func TestAssembleGoFlacAVX512Kernel(t *testing.T) {
path := "../../go-libraries/go-flac/avx512_amd64.s"
if _, err := os.Stat(path); err != nil {
t.Skip("go-libraries repository not present next to gasm-devkit")
t.Skip("go-libraries repository not present next to gasm-sdk")
}
src, err := os.ReadFile(path)
if err != nil {
+11 -2
View File
@@ -13,7 +13,7 @@ import (
"strings"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// The differential kernels for the DATA-path and front-end gaps are kept in
@@ -23,9 +23,18 @@ import (
// bytes must agree with the relocation sites masked on both sides.
// toolAsmObject assembles path with the installed toolchain's assembler for
// goarch ("" = the host) and returns the object bytes.
// goarch ("" = the host) and returns the object bytes. Every live-oracle
// comparison funnels through here, so this is also where the deliberate-run
// boundary sits: under -short (the push pipeline's mode) the comparisons
// skip, because each spawns a go tool asm subprocess and the small single-
// core runner pays seconds per spawn. The encodings stay pinned by the
// golden-byte tests in every mode; the live oracle runs in the local test
// gate and the dispatched workflows.
func toolAsmObject(t *testing.T, path, goarch string) []byte {
t.Helper()
if testing.Short() {
t.Skip("live go tool asm oracle: skipped in -short mode")
}
goBin, err := exec.LookPath("go")
if err != nil {
t.Skip("no Go toolchain available")
+1 -1
View File
@@ -12,7 +12,7 @@ import (
"strings"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// TestGOObjectLOONG64Structure checks the emitted loong64 object's blocks:
+14 -4
View File
@@ -9,7 +9,7 @@ import (
"sort"
"strconv"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
)
// Image is an assembled file: the function bodies laid out in source order,
@@ -134,7 +134,8 @@ type DataSymbol struct {
Offset int // byte offset within Data
Size int
Static bool // the <> marker: file-local, not exported
Rodata bool // the RODATA flag: read-only data
Rodata bool // the RODATA flag: read-only data (implies no pointers)
Noptr bool // the NOPTR flag: data with no pointers, kept out of GC scanning
Dupok bool // the DUPOK flag: duplicate-OK
// Relocs carries the symbol-valued DATA initialisers ("DATA s+0(SB)/8,
// $other(SB)"): fields of this symbol's data that hold another symbol's
@@ -269,6 +270,7 @@ func AssembleFile(f *ast.File, opts ...AssembleOption) (*Image, error) {
Size: len(d.buf),
Static: d.static,
Rodata: d.rodata,
Noptr: d.noptr,
Dupok: d.dupok,
})
img.Data = append(img.Data, d.buf...)
@@ -413,6 +415,7 @@ func AssembleFileRISCV(f *ast.File) (*Image, error) {
Size: d.size,
Static: d.static,
Rodata: d.rodata,
Noptr: d.noptr,
Dupok: d.dupok,
})
}
@@ -485,6 +488,7 @@ func AssembleFileLOONG64(f *ast.File) (*Image, error) {
Size: d.size,
Static: d.static,
Rodata: d.rodata,
Noptr: d.noptr,
Dupok: d.dupok,
})
}
@@ -546,6 +550,7 @@ type dataSym struct {
size int
static bool
rodata bool
noptr bool
dupok bool
// relocs are the symbol-valued DATA fields, in declaration order; Off
// is relative to the symbol's data start.
@@ -587,12 +592,14 @@ func collectData(f *ast.File) ([]dataSym, error) {
switch f {
case "RODATA":
ds.rodata = true
case "NOPTR":
ds.noptr = true
case "DUPOK":
ds.dupok = true
default:
// Legacy numeric flag constants (runtime/textflag.h):
// DUPOK is 2, RODATA is 8; combinations arrive as one
// number (e.g. 10 = RODATA|DUPOK).
// DUPOK is 2, RODATA is 8, NOPTR is 16; combinations arrive
// as one number (e.g. 10 = RODATA|DUPOK).
if n, err := strconv.Atoi(f); err == nil {
if n&2 != 0 {
ds.dupok = true
@@ -600,6 +607,9 @@ func collectData(f *ast.File) ([]dataSym, error) {
if n&8 != 0 {
ds.rodata = true
}
if n&16 != 0 {
ds.noptr = true
}
}
}
}
+12 -38
View File
@@ -12,7 +12,7 @@ import (
"strings"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// TestAssembleFileStaticData checks the whole-image layout; code, padding
@@ -60,23 +60,15 @@ DATA small<>+0(SB)/4, $0x1234
}
}
// TestAssembleFileErrors checks the static-symbol error paths.
// TestAssembleFileErrors checks the static-symbol error paths. A reference
// to a static symbol no GLOBL defines defers to the linker exactly as the
// toolchain does (an external relocation), so it is not an error here.
func TestAssembleFileErrors(t *testing.T) {
cases := []struct {
name string
src string
want string // substring of the error
}{
{
"undefined symbol",
`
#include "textflag.h"
TEXT ·f(SB), NOSPLIT, $0
VMOVDQU nope<>(SB), X0
RET
`,
"undefined symbol",
},
{
"DATA without GLOBL",
`
@@ -369,23 +361,8 @@ func main() {
if err != nil {
t.Fatalf("baseline build: %v\n%s", err, buildLog)
}
var work, linkLine, asmObj string
for line := range strings.SplitSeq(string(buildLog), "\n") {
switch {
case strings.HasPrefix(line, "WORK="):
work = strings.TrimPrefix(line, "WORK=")
case strings.Contains(line, "/asm ") && strings.Contains(line, "main_amd64.s") && !strings.Contains(line, "-gensymabis"):
asmObj = fieldAfter(line, "-o")
case strings.Contains(line, "/link ") && strings.Contains(line, "-importcfg"):
linkLine = line
}
}
if work == "" || asmObj == "" || linkLine == "" {
t.Skipf("could not parse build log (work=%q asmObj=%q link=%q)", work, asmObj, linkLine)
}
defer os.RemoveAll(work)
asmObj = strings.ReplaceAll(asmObj, "$WORK", work)
linkLine = strings.ReplaceAll(linkLine, "$WORK", work)
st := parseBuildLog(t, buildLog, "main_amd64.s")
defer os.RemoveAll(st.work)
// Assemble the same source with gasm and substitute the object.
src, err := os.ReadFile(filepath.Join(dir, "main_amd64.s"))
@@ -400,21 +377,18 @@ func main() {
if err != nil {
t.Fatalf("AssembleFile: %v", err)
}
gasmObj, err := img.GOObject("dlink", "main_amd64.s")
// The package path is "main": the linker resolves the Go code's
// references against main.<name>, so the object must define the symbols
// under that prefix whatever the module is called.
gasmObj, err := img.GOObject("main", "main_amd64.s")
if err != nil {
t.Fatalf("GOObject: %v", err)
}
if err := os.WriteFile(asmObj, gasmObj, 0o644); err != nil {
t.Fatalf("write gasm object: %v", err)
}
linkCmd := exec.Command("bash", "-c", "cd "+dir+" && "+linkLine)
if out, err := linkCmd.CombinedOutput(); err != nil {
t.Fatalf("re-link with gasm object: %v\n%s", err, out)
}
substituteAndRelink(t, goBin, dir, st, filepath.Join(dir, "prog2"), gasmObj)
// The linked program must run and find the right function behind the
// data word.
out, err := exec.Command(filepath.Join(dir, "prog")).CombinedOutput()
out, err := exec.Command(filepath.Join(dir, "prog2")).CombinedOutput()
if err != nil {
t.Fatalf("linked program failed: %v\n%s", err, out)
}
+258 -7
View File
@@ -9,7 +9,7 @@ import (
"strconv"
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
)
// assembleLOONG64 assembles a LoongArch (loong64) TEXT function body into
@@ -371,6 +371,13 @@ func loong64InstrSize(instr *ast.Instr, fi loong64FrameInfo) int {
if mnem == "RET" {
return len(loong64Return(fi))
}
// BYTE lays down one raw byte per operand, a front-end pseudo-op the
// toolchain spells only on x86 but accepts here the same way the arm64
// and riscv64 encoders do (a superset spelling, shippable via the goobj
// path).
if mnem == "BYTE" {
return len(ops)
}
switch mnem {
case "END", "FUNCDATA", "PCDATA":
return 0 // bookkeeping statements contribute no bytes
@@ -484,6 +491,18 @@ func encodeLOONG64Instr(instr *ast.Instr, pc int, offsets map[string]int, fi loo
return nil, fmt.Errorf("WORD expects 1 operand, got %d", len(ops))
}
return l64wordLE(uint32(immFromOperand(ops[0]))), nil
case "BYTE":
// BYTE $b lays down one raw byte per operand, the same front-end
// pseudo-op the arm64 and riscv64 encoders accept.
var out []byte
for _, op := range ops {
b := l64Imm64(op)
if b < 0 || b > 0xFF {
return nil, fmt.Errorf("BYTE: immediate %d does not fit a byte", b)
}
out = append(out, byte(b))
}
return out, nil
case "END", "FUNCDATA", "PCDATA", "GETCALLERPC":
// The assembler's bookkeeping statements. END, FUNCDATA and PCDATA
// contribute no bytes, the same shapes GOARCH=loong64 go tool asm
@@ -701,6 +720,35 @@ func encodeLOONG64Instr(instr *ast.Instr, pc int, offsets map[string]int, fi loo
}
return l64wordLE(l64rr(enc.op, rj, rd)), nil
case l64Fllsc:
// LLACQ{W,V} (Rj), Rd loads and SCREL{W,V} Rd, (Rj) stores, both
// 2R encodings op | rj<<5 | rd against a zero-offset memory operand
// (the toolchain's C_ZOREG, which rejects any displacement).
rd, rj, off, _, err := l64MemOperands(ops, fi)
if err != nil {
return nil, fmt.Errorf("%s: %w", mnem, err)
}
if off != 0 {
return nil, fmt.Errorf("%s: only a zero-offset memory operand is allowed", mnem)
}
return l64wordLE(l64rr(enc.op, rj, rd)), nil
case l64Fscq:
// SCQ first, middle, (base): op | middle<<10 | base<<5 | first,
// against a zero-offset memory operand as with the LL/SC pair.
if len(ops) != 3 || !isMemOperand(ops[2]) || isMemOperand(ops[0]) || isMemOperand(ops[1]) {
return nil, fmt.Errorf("%s expects reg, reg, (reg)", mnem)
}
first, middle := l64Reg(ops[0]), l64Reg(ops[1])
rj, off := l64MemWithFrame(ops[2], fi)
if first < 0 || middle < 0 || rj < 0 {
return nil, fmt.Errorf("%s: invalid register operand", mnem)
}
if off != 0 {
return nil, fmt.Errorf("%s: only a zero-offset memory operand is allowed", mnem)
}
return l64wordLE(l64rrr(enc.op, middle, rj, first)), nil
case l64Firr:
// LU52ID: INSTR $imm, rd or INSTR $imm, rj, rd.
if len(ops) < 2 || !isImmOperand(ops[0]) {
@@ -1229,6 +1277,31 @@ func l64MemOperands(ops []*ast.Operand, fi loong64FrameInfo) (rd, rj int, off in
return rd, rj, off, load, nil
}
// l64ImmMem reads the `$off(rj)` immediate form off an operand's raw text:
// the shared immediate parse reduces it to the bare number and keeps only
// the text as a witness of the base register. ok reports the form was
// found, with the base's register number (or -1 when the name is not a
// general register).
func l64ImmMem(op *ast.Operand) (off int32, base int, ok bool) {
if op.Kind != ast.OpImmediate || !op.Imm.HasVal {
return 0, 0, false
}
raw := strings.ReplaceAll(op.Raw, " ", "")
if !strings.HasPrefix(raw, "$") || !strings.HasSuffix(raw, ")") {
return 0, 0, false
}
open := strings.LastIndexByte(raw, '(')
if open < 2 {
return 0, 0, false
}
base = loong64RegNum(raw[open+1 : len(raw)-1])
v := op.Imm.Val
if op.Imm.Neg {
v = -v
}
return int32(v), base, base >= 0
}
// ---- the MOV pseudo-instruction ----
// encodeLOONG64Mov encodes the MOV family, the load/store/immediate
@@ -1262,6 +1335,29 @@ func encodeLOONG64Mov(instr *ast.Instr, mnem string, fi loong64FrameInfo, relocs
}
return encodeLOONG64SBAddr(src.Imm.Sym, rd, relocs), nil
}
// MOVx $off(rj), rd computes an address: the toolchain's `mov
// $soreg, r` case, a plain addi.d whatever the move's width (both
// MOVW and MOVV $4(R4), R5 encode the same addi.d in its testdata).
// A wider offset materialises in R30 first (lu12i.w + ori + add.d,
// its case 10). The immediate's Raw carries the base register,
// which the shared immediate parse reduces to the bare number.
if off, base, ok := l64ImmMem(src); ok {
rd := l64Reg(dst)
if rd < 0 {
return nil, fmt.Errorf("%s $imm(rj): invalid destination register", mnem)
}
if loong64RegClass(operandRegName(dst)) == l64ClsFP {
return nil, fmt.Errorf("%s $imm(rj): illegal combination with an F register destination", mnem)
}
if off >= -2048 && off <= 2047 {
return l64wordLE(l64irr(l64DualTable["ADDV"].imm, int(off), base, rd)), nil
}
return l64WordsLE(
l64ir(l64InstrTable["LU12IW"].op, int(off)>>12, 30),
l64irr(l64DualTable["OR"].imm, int(off)&0xFFF, 30, 30),
l64rrr(l64DualTable["ADDV"].rrr, 30, base, rd),
), nil
}
rd := l64Reg(dst)
if rd < 0 {
return nil, fmt.Errorf("%s $imm: invalid destination register", mnem)
@@ -1356,6 +1452,14 @@ func loong64MovSize(mnem string, ops []*ast.Operand, fi loong64FrameInfo) int {
if src.Imm.Sym != nil && src.Imm.Sym.Pseudo == "SB" {
return 8 // pcalau12i + addi.d
}
// The $off(rj) address immediate: addi.d in the 12-bit window,
// lu12i.w + ori + add.d beyond it (the toolchain's case 10).
if off, _, ok := l64ImmMem(src); ok {
if off >= -2048 && off <= 2047 {
return 4
}
return 12
}
if loong64RegClass(operandRegName(dst)) == l64ClsFP {
return 8 // ori/addi.w r30 + movgr2fr.w (an encode-time diagnostic when invalid)
}
@@ -1868,6 +1972,19 @@ func l64MemWithFrame(op *ast.Operand, fi loong64FrameInfo) (rj int, off int32) {
return l64Mem(op)
}
// l64VmovqMem resolves a VMOVQ/XVMOVQ memory operand. The toolchain's
// vector table falls back to the zero register as the FP-relative base
// (`VMOVQ V2, y+16(FP)` stores through R0 while MOVW reads the same operand
// through R3), so the vector moves keep the resolved offset but the zero
// base, exactly as `go tool asm` emits them.
func l64VmovqMem(op *ast.Operand, fi loong64FrameInfo) (rj int, off int32) {
rj, off = l64MemWithFrame(op, fi)
if op.Addr.Sym != nil && op.Addr.Sym.Pseudo == "FP" {
rj = 0
}
return rj, off
}
// l64MemOffset returns the resolved byte offset of a memory operand.
func l64MemOffset(op *ast.Operand, fi loong64FrameInfo) int32 {
_, off := l64MemWithFrame(op, fi)
@@ -1892,7 +2009,7 @@ func l64Label(op *ast.Operand) string {
type l64VecOperand struct {
num int // 5-bit register number
lasx bool // X bank (LASX) rather than V (LSX)
width byte // suffix width letter (B/H/W/V), 0 on a bare register
width byte // suffix width letter (B/H/W/V/Q), 0 on a bare register
lanes int // lane count of a width suffix (B16 → 16)
elem int // element index of a .T[i] suffix
hasEl bool // the suffix names an element (.T[i])
@@ -1932,7 +2049,7 @@ func l64ParseVecOperand(op *ast.Operand) (v l64VecOperand, ok bool) {
}
i++
w := name[i]
if w != 'B' && w != 'H' && w != 'W' && w != 'V' {
if w != 'B' && w != 'H' && w != 'W' && w != 'V' && w != 'Q' {
return v, false
}
v.width, v.hasSuf = w, true
@@ -2174,6 +2291,10 @@ func encodeLOONG64Vector(instr *ast.Instr, mnem string, fi loong64FrameInfo) ([]
// VMOVQ rj, vd.T vreplgr2vr (duplicate a general register)
// VMOVQ vj.T[i], rd vpickve2gr (extract one element)
// VMOVQ rj, vd.T[i] vinsgr2vr (insert one element)
// VMOVQ vj.T[i], vd.T vreplvei (broadcast one element, LSX)
// XVMOVQ xj, xd.T xvreplve0 (broadcast element zero, LASX)
// XVMOVQ xj, xd.T[i] xvinsve0 (insert element zero, LASX)
// XVMOVQ xj.T[i], xd xvpickve (extract one element, LASX)
func encodeLOONG64Vmovq(lasx bool, ops []*ast.Operand, fi loong64FrameInfo) ([]byte, error) {
enc := l64VmovqTable[lasx]
bank := "V"
@@ -2200,6 +2321,118 @@ func encodeLOONG64Vmovq(lasx bool, ops []*ast.Operand, fi loong64FrameInfo) ([]b
return loong64RegNum(name), nil
}
// Element broadcast: VMOVQ vj.T[i], vd.T (vreplvei.{b,h,w,d}), the
// source element width matching the destination arrangement. An LSX-only
// form: the toolchain's table gives vreplvei no LASX counterpart.
if srcVec && dstVec && src.hasEl && dst.hasSuf && !dst.hasEl {
if lasx || src.lasx || dst.lasx {
return nil, fmt.Errorf("VMOVQ: vreplvei has no %s-bank form", bank)
}
if src.unsig {
return nil, fmt.Errorf("VMOVQ: vreplvei takes no unsigned element suffix")
}
if src.width != dst.width {
return nil, fmt.Errorf("VMOVQ: element width does not match arrangement %q", ops[1].Raw)
}
if _, ok := l64VecSuffixWidth(false, dst); !ok {
return nil, fmt.Errorf("VMOVQ: invalid arrangement %q", ops[1].Raw)
}
var op uint32
limit := 0
switch src.width {
case 'B':
op, limit = enc.rveiB, 15
case 'H':
op, limit = enc.rveiH, 7
case 'W':
op, limit = enc.rveiW, 3
default:
op, limit = enc.rveiD, 1
}
if src.elem > limit {
return nil, fmt.Errorf("VMOVQ: element index %d out of range [0, %d]", src.elem, limit)
}
return l64wordLE(op | uint32(src.elem)<<10 | uint32(src.num)<<5 | uint32(dst.num)), nil
}
// Broadcast of element zero: XVMOVQ xj, xd.T (xvreplve0.{b,h,w,d,q}),
// a bare X source into an arranged X destination. LASX only.
if srcVec && dstVec && !src.hasSuf && dst.hasSuf && !dst.hasEl {
if !lasx || src.lasx != lasx || dst.lasx != lasx {
return nil, fmt.Errorf("XVMOVQ: xvreplve0 is the %s-bank form alone", bank)
}
var op uint32
switch dst.width {
case 'B':
if dst.lanes != 32 {
return nil, fmt.Errorf("XVMOVQ: invalid arrangement %q", ops[1].Raw)
}
op = enc.rve0B
case 'H':
if dst.lanes != 16 {
return nil, fmt.Errorf("XVMOVQ: invalid arrangement %q", ops[1].Raw)
}
op = enc.rve0H
case 'W':
if dst.lanes != 8 {
return nil, fmt.Errorf("XVMOVQ: invalid arrangement %q", ops[1].Raw)
}
op = enc.rve0W
case 'V':
if dst.lanes != 4 {
return nil, fmt.Errorf("XVMOVQ: invalid arrangement %q", ops[1].Raw)
}
op = enc.rve0D
case 'Q':
if dst.lanes != 2 {
return nil, fmt.Errorf("XVMOVQ: invalid arrangement %q", ops[1].Raw)
}
op = enc.rve0Q
default:
return nil, fmt.Errorf("XVMOVQ: invalid arrangement %q", ops[1].Raw)
}
return l64wordLE(op | uint32(src.num)<<5 | uint32(dst.num)), nil
}
// Insert of element zero: XVMOVQ xj, xd.T[i] (xvinsve0.{w,d}), a bare X
// source into one word or double-word lane. LASX only.
if srcVec && dstVec && !src.hasSuf && dst.hasEl {
if !lasx || src.lasx != lasx || dst.lasx != lasx {
return nil, fmt.Errorf("XVMOVQ: xvinsve0 is the %s-bank form alone", bank)
}
op, limit := enc.xinsW, 7
if dst.width != 'W' {
op, limit = enc.xinsD, 3
if dst.width != 'V' {
return nil, fmt.Errorf("XVMOVQ: xvinsve0 takes word or double-word lanes, got %q", ops[1].Raw)
}
}
if dst.elem > limit {
return nil, fmt.Errorf("XVMOVQ: element index %d out of range [0, %d]", dst.elem, limit)
}
return l64wordLE(op | uint32(dst.elem)<<10 | uint32(src.num)<<5 | uint32(dst.num)), nil
}
// Element extract into a vector register: XVMOVQ xj.T[i], xd
// (xvpickve.{w,d}), one word or double-word lane out to a bare X
// register. LASX only.
if srcVec && src.hasEl && dstVec && !dst.hasSuf {
if !lasx || src.lasx != lasx || dst.lasx != lasx {
return nil, fmt.Errorf("XVMOVQ: xvpickve is the %s-bank form alone", bank)
}
op, limit := enc.xpickW, 7
if src.width != 'W' {
op, limit = enc.xpickD, 3
if src.width != 'V' {
return nil, fmt.Errorf("XVMOVQ: xvpickve takes word or double-word lanes, got %q", ops[0].Raw)
}
}
if src.elem > limit {
return nil, fmt.Errorf("XVMOVQ: element index %d out of range [0, %d]", src.elem, limit)
}
return l64wordLE(op | uint32(src.elem)<<10 | uint32(src.num)<<5 | uint32(dst.num)), nil
}
// Register move: VMOVQ vj, vd (vori.b/xvori.b with the zero constant),
// both operands bare registers of the same bank.
if srcVec && dstVec {
@@ -2224,7 +2457,7 @@ func encodeLOONG64Vmovq(lasx bool, ops []*ast.Operand, fi loong64FrameInfo) ([]b
}
return l64wordLE(l64rrr(enc.stx, rk, rj, src.num)), nil
}
rj, off := l64MemWithFrame(ops[1], fi)
rj, off := l64VmovqMem(ops[1], fi)
if rj < 0 || off < -2048 || off > 2047 {
return nil, fmt.Errorf("VMOVQ: store offset out of range [-2048, 2047]")
}
@@ -2247,9 +2480,9 @@ func encodeLOONG64Vmovq(lasx bool, ops []*ast.Operand, fi loong64FrameInfo) ([]b
}
return l64wordLE(l64rrr(enc.ldx, rk, rj, dst.num)), nil
}
rj, off := l64MemWithFrame(ops[0], fi)
if rj < 0 || off < -2048 || off > 2047 {
return nil, fmt.Errorf("VMOVQ: load offset out of range [-2048, 2047]")
rj, off := l64VmovqMem(ops[0], fi)
if rj < 0 {
return nil, fmt.Errorf("VMOVQ: invalid load operand")
}
op := enc.ld
if dst.hasSuf {
@@ -2257,16 +2490,34 @@ func encodeLOONG64Vmovq(lasx bool, ops []*ast.Operand, fi loong64FrameInfo) ([]b
if !ok {
return nil, fmt.Errorf("VMOVQ: invalid replicate width suffix %q", ops[1].Raw)
}
// vldrepl keeps the byte offset raw for bytes and scales it by
// the element width for the wider forms, the immediate field
// shrinking a bit per scale exactly as the toolchain encodes it
// (the field mask keeps the two's complement inside its width).
scale, mask, lo, hi := 1, int32(0xFFF), -2048, 2047
switch w {
case 0:
op = enc.replB
case 1:
op = enc.replH
scale, mask, lo, hi = 2, 0x7FF, -1024, 1023
case 2:
op = enc.replW
scale, mask, lo, hi = 4, 0x3FF, -512, 511
default:
op = enc.replD
scale, mask, lo, hi = 8, 0x1FF, -256, 255
}
if off%int32(scale) != 0 {
return nil, fmt.Errorf("VMOVQ: offset %d must be a multiple of %d", off, scale)
}
off /= int32(scale)
if off < int32(lo) || off > int32(hi) {
return nil, fmt.Errorf("VMOVQ: offset out of range [%d, %d]", lo*scale, hi*scale)
}
off &= mask
} else if off < -2048 || off > 2047 {
return nil, fmt.Errorf("VMOVQ: load offset out of range [-2048, 2047]")
}
return l64wordLE(l64irr(op, int(off), rj, dst.num)), nil
}
+32 -5
View File
@@ -279,6 +279,8 @@ const (
l64Fvvv // 3R vector (LSX/LASX): op | vk<<10 | vj<<5 | vd
l64Fvcf // vector-to-condition: op | subop<<10 | vj<<5 | fcc
l64Fvvvv // 4R vector shuffle: op | va<<15 | vk<<10 | vj<<5 | vd
l64Fllsc // acquire/release LL/SC (2R against a zero-offset memory operand)
l64Fscq // sc.q: op | middle<<10 | base<<5 | first against a zero-offset memory operand
)
// l64Enc is one instruction's encoding: its bit layout (format) and the
@@ -341,8 +343,9 @@ var l64Vec2R = map[string]bool{}
// such as vshuf.b).
var l64Vec4R = map[string]bool{}
// l64VmovqOps holds the VMOVQ/XVMOVQ opcode constants (pre-shifted to bit
// 15), read off `go tool objdump` of GOARCH=loong64 `go tool asm` kernels.
// l64VmovqOps holds the VMOVQ/XVMOVQ opcode constants, each pre-shifted to
// its exact bit range, read off `go tool objdump` of GOARCH=loong64
// `go tool asm` kernels and the toolchain's specialLsxMovInst table.
type l64VmovqEnc struct {
ld, st, ldx, stx uint32 // plain and indexed load/store
replB, replH, replW, replD uint32 // vldrepl: load and replicate element
@@ -350,6 +353,11 @@ type l64VmovqEnc struct {
ins uint32 // vinsgr2vr element insert
dup uint32 // vreplgr2vr duplicate (width in [11:10])
move uint32 // vori.b/xvori.b $0 register move
rveiB, rveiH, rveiW, rveiD uint32 // vreplvei: broadcast one element (LSX)
rve0B, rve0H, rve0W uint32 // xvreplve0 broadcast of element zero (LASX)
rve0D, rve0Q uint32 // xvreplve0.{d,q}, ditto
xinsW, xinsD uint32 // xvinsve0: insert element zero (LASX)
xpickW, xpickD uint32 // xvpickve: extract element (LASX)
}
var l64VmovqTable = map[bool]l64VmovqEnc{
@@ -358,12 +366,17 @@ var l64VmovqTable = map[bool]l64VmovqEnc{
replB: 0x6100 << 15, replH: 0x6080 << 15, replW: 0x6040 << 15, replD: 0x6020 << 15,
pickS: 0xE5DF << 15, pickU: 0xE5E7 << 15,
ins: 0xE5D7 << 15, dup: 0xE53E << 15, move: 0xE65A << 15,
rveiB: 0x01CBDE << 14, rveiH: 0x0397BE << 13, rveiW: 0x072F7E << 12, rveiD: 0x0E5EFE << 11,
},
true: { // XVMOVQ, the LASX (X) bank
ld: 0x5900 << 15, st: 0x5980 << 15, ldx: 0x7090 << 15, stx: 0x7098 << 15,
replB: 0x6500 << 15, replH: 0x6480 << 15, replW: 0x6440 << 15, replD: 0x6420 << 15,
pickS: 0xEDDF << 15, pickU: 0xEDE7 << 15,
ins: 0xEDD7 << 15, dup: 0xED3E << 15, move: 0xEE5A << 15,
rve0B: 0x1DC1C0 << 10, rve0H: 0x1DC1E0 << 10, rve0W: 0x1DC1F0 << 10,
rve0D: 0x1DC1F8 << 10, rve0Q: 0x1DC1FC << 10,
xinsW: 0x03B7FE << 13, xinsD: 0x076FFE << 12,
xpickW: 0x03B81E << 13, xpickD: 0x07703E << 12,
},
}
@@ -373,7 +386,7 @@ func init() {
"ADD": 0x20 << 15, "ADDW": 0x20 << 15, "ADDV": 0x21 << 15, "ADDVU": 0x21 << 15,
"SUB": 0x22 << 15, "SUBW": 0x22 << 15, "SUBV": 0x23 << 15, "SUBVU": 0x23 << 15,
"SGT": 0x24 << 15, "SGTU": 0x25 << 15,
"MASKEQZ": 0x26 << 15, "MASKNEZ": 0x27 << 15, "SCQ": 0x070AE << 15,
"MASKEQZ": 0x26 << 15, "MASKNEZ": 0x27 << 15,
"NOR": 0x28 << 15, "AND": 0x29 << 15, "OR": 0x2a << 15, "XOR": 0x2b << 15,
"ORN": 0x2c << 15, "ANDN": 0x2d << 15,
"SLL": 0x2e << 15, "SRL": 0x2f << 15, "SRA": 0x30 << 15,
@@ -467,6 +480,20 @@ func init() {
l64InstrTable["RDTIMEHW"] = l64Enc{format: l64Frdtime, op: 0x19 << 10}
l64InstrTable["RDTIMED"] = l64Enc{format: l64Frdtime, op: 0x1a << 10}
// Acquire/release LL/SC (2R against a zero-offset memory operand):
// LLACQV (Rj), Rd loads, SCRELV Rd, (Rj) stores, both encoding
// op | rj<<5 | rd. Opcodes from cmd/internal/obj/loong64/instOp.go
// (ll.acq.{w,d}, sc.rel.{w,d}).
l64InstrTable["LLACQW"] = l64Enc{format: l64Fllsc, op: 0x0E15E0 << 10}
l64InstrTable["SCRELW"] = l64Enc{format: l64Fllsc, op: 0x0E15E1 << 10}
l64InstrTable["LLACQV"] = l64Enc{format: l64Fllsc, op: 0x0E15E2 << 10}
l64InstrTable["SCRELV"] = l64Enc{format: l64Fllsc, op: 0x0E15E3 << 10}
// SCQ (sc.q first, middle, (base)) keeps its own operand order: the
// encoding is op | middle<<10 | base<<5 | first, the memory operand's
// base in the rj field, not the toolchain's generic 3R layout.
l64InstrTable["SCQ"] = l64Enc{format: l64Fscq, op: 0x070AE << 15}
// The dual-form arithmetic mnemonics (register 3R + immediate 2RI12),
// selected by the operand kind; the shift mnemonics pair the 3R form
// with a 5/6-bit shift immediate.
@@ -868,7 +895,7 @@ func init() {
"VNORB": {0xE7B8 << 15, false, 0, 255, 0, 0xFF},
"XVNORB": {0xEFB8 << 15, true, 0, 255, 0, 0xFF},
"VSEQB": {0xE500 << 15, false, -16, 15, 0, 0x1F},
"XVSEQB": {0xE900 << 15, true, -16, 15, 0, 0x1F},
"XVSEQB": {0xED00 << 15, true, -16, 15, 0, 0x1F},
// vseqi.h/w accept the same si5 window as vseqi.b; vseqi.d carries a
// 7-bit field, but the toolchain range-checks it down to si5 as well
// (GOARCH=loong64 go tool asm rejects VSEQV $32 and VSEQV $-64).
@@ -877,7 +904,7 @@ func init() {
"VSEQW": {0xE502 << 15, false, -16, 15, 0, 0x1F},
"XVSEQW": {0xED02 << 15, true, -16, 15, 0, 0x1F},
"VSEQV": {0xE503 << 15, false, -16, 15, 0, 0x7F},
"XVSEQV": {0xE903 << 15, true, -16, 15, 0, 0x7F},
"XVSEQV": {0xED03 << 15, true, -16, 15, 0, 0x7F},
// vslti compares against a signed (or, in the U spellings, unsigned)
// si5/ui5 constant.
"VSLTB": {0xE50C << 15, false, -16, 15, 0, 0x1F},
+229 -2
View File
@@ -8,8 +8,8 @@ import (
"encoding/binary"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// firstTextLOONG64 parses assembly source and returns the first TEXT body.
@@ -831,3 +831,230 @@ TEXT ·atoms(SB), NOSPLIT, $0
0x4C000020,
)
}
// TestLOONG64_llacqScrel pins the acquire/release LL/SC pair. The oracle
// words come from GOARCH=loong64 go tool objdump and the toolchain's own
// loong64enc1.s golden bytes.
func TestLOONG64_llacqScrel(t *testing.T) {
fn := firstTextLOONG64(t, `#include "textflag.h"
TEXT ·llsc(SB), NOSPLIT, $0
LLACQW (R5), R4
LLACQV (R5), R4
SCRELW R4, (R6)
SCRELV R4, (R6)
RET
`)
code := assembleLOONG64Helper(t, fn)
wantWords(t, code,
0x385780A4, // ll.acq.w r4, r5
0x385788A4, // ll.acq.d r4, r5
0x385784C4, // sc.rel.w r4, r6
0x38578CC4, // sc.rel.d r4, r6
0x4C000020,
)
// The toolchain accepts the zero-offset memory form alone.
for i, src := range []string{
`TEXT ·e(SB), NOSPLIT, $0
LLACQW 4(R5), R4
RET
`,
`TEXT ·e(SB), NOSPLIT, $0
SCRELV R4, 8(R6)
RET
`,
} {
fn := firstTextLOONG64(t, src)
if _, _, _, _, _, err := assembleLOONG64(fn); err == nil {
t.Errorf("case %d: expected an error, got none", i)
}
}
}
// TestLOONG64_vmovqSuffixed pins the element-broadcast and element-move
// VMOVQ/XVMOVQ forms, with the oracle words lifted verbatim from the
// toolchain's loong64enc1.s.
func TestLOONG64_vmovqSuffixed(t *testing.T) {
fn := firstTextLOONG64(t, `#include "textflag.h"
TEXT ·vmovq(SB), NOSPLIT, $0
VMOVQ V1.B[3], V9.B16
VMOVQ V2.H[2], V8.H8
VMOVQ V3.W[1], V7.W4
VMOVQ V4.V[0], V6.V2
XVMOVQ X0, X31.B32
XVMOVQ X1, X30.H16
XVMOVQ X2, X29.W8
XVMOVQ X3, X28.V4
XVMOVQ X3, X27.Q2
XVMOVQ X0, X31.W[7]
XVMOVQ X1, X29.W[0]
XVMOVQ X3, X28.V[3]
XVMOVQ X4, X27.V[0]
XVMOVQ X31.W[7], X0
XVMOVQ X29.W[0], X1
XVMOVQ X28.V[3], X8
XVMOVQ X27.V[0], X9
RET
`)
code := assembleLOONG64Helper(t, fn)
wantWords(t, code,
0x72F78C29, // vreplvei.b v9, v1, 3
0x72F7C848, // vreplvei.h v8, v2, 2
0x72F7E467, // vreplvei.w v7, v3, 1
0x72F7F086, // vreplvei.d v6, v4, 0
0x7707001F, // xvreplve0.b x31, x0
0x7707803E, // xvreplve0.h x30, x1
0x7707C05D, // xvreplve0.w x29, x2
0x7707E07C, // xvreplve0.d x28, x3
0x7707F07B, // xvreplve0.q x27, x3
0x76FFDC1F, // xvinsve0.w x31, x0, 7
0x76FFC03D, // xvinsve0.w x29, x1, 0
0x76FFEC7C, // xvinsve0.d x28, x3, 3
0x76FFE09B, // xvinsve0.d x27, x4, 0
0x7703DFE0, // xvpickve.w x0, x31, 7
0x7703C3A1, // xvpickve.w x1, x29, 0
0x7703EF88, // xvpickve.d x8, x28, 3
0x7703E369, // xvpickve.d x9, x27, 0
0x4C000020,
)
// The rejected shapes: a width mismatch between the element and the
// arrangement, an element index past the lane count, a wrong-bank
// vreplvei and an arrangement the LASX bank does not spell.
for i, src := range []string{
`TEXT ·e(SB), NOSPLIT, $0
VMOVQ V1.H[3], V9.B16
RET
`,
`TEXT ·e(SB), NOSPLIT, $0
VMOVQ V1.B[16], V9.B16
RET
`,
`TEXT ·e(SB), NOSPLIT, $0
XVMOVQ X1.B[3], X9.B32
RET
`,
`TEXT ·e(SB), NOSPLIT, $0
XVMOVQ X0, X31.B16
RET
`,
`TEXT ·e(SB), NOSPLIT, $0
XVMOVQ X0, X31.W[8]
RET
`,
} {
fn := firstTextLOONG64(t, src)
if _, _, _, _, _, err := assembleLOONG64(fn); err == nil {
t.Errorf("case %d: expected an error, got none", i)
}
}
}
// TestLOONG64_parityFixes pins the operand forms whose encodings were found
// diverging from the toolchain by the loong64enc1.s differential: the
// $off(reg) address immediate (addi.d), the SCQ operand order, the scaled
// vldrepl offsets (with their field masks), the XVSEQB/XVSEQV immediate
// opcodes and the zero-register base the toolchain gives FP-relative
// VMOVQ/XVMOVQ memory operands. Golden words from loong64enc1.s.
func TestLOONG64_parityFixes(t *testing.T) {
fn := firstTextLOONG64(t, `#include "textflag.h"
TEXT ·parity(SB), NOSPLIT, $0-32
MOVW $4(R4), R5
MOVV $4(R4), R5
MOVW $65536(R4), R5
MOVW $-4096(R4), R5
SCQ R4, R5, (R6)
VMOVQ 2(R4), V1.H8
VMOVQ -6(R4), V1.H8
VMOVQ -12(R4), V2.W4
VMOVQ -16(R4), V3.V2
XVMOVQ -10(R4), X1.H16
XVSEQB $0, X2, X4
XVSEQH $3, X2, X4
XVSEQW $12, X2, X4
XVSEQV $15, X2, X4
XVSEQV $-15, X2, X4
VMOVQ V2, y+16(FP)
VMOVQ y+16(FP), V2
VMOVQ V2, x+2030(FP)
XVMOVQ X6, y+16(FP)
RET
`)
code := assembleLOONG64Helper(t, fn)
wantWords(t, code,
0x02C01085, // addi.d $4, r4, r5
0x02C01085, // addi.d $4, r4, r5 (MOVW keeps the 64-bit addi.d)
0x1400021E, // lu12i.w $16, r30
0x038003DE, // ori $0, r30, r30
0x0010F885, // add.d r5, r4, r30
0x15FFFFFE, // lu12i.w $-1, r30
0x038003DE, // ori $0, r30, r30
0x0010F885, // add.d r5, r4, r30
0x385714C4, // sc.q r4, r5, (r6): middle<<10 | base<<5 | first
0x30400481, // vldrepl.h v1, 2(r4)
0x305FF481, // vldrepl.h v1, -6(r4)
0x302FF482, // vldrepl.w v2, -12(r4)
0x3017F883, // vldrepl.d v3, -16(r4)
0x325FEC81, // xvldrepl.h x1, -10(r4)
0x76800044, // xvseqi.b x4, x2, 0
0x76808C44, // xvseqi.h x4, x2, 3
0x76813044, // xvseqi.w x4, x2, 12
0x7681BC44, // xvseqi.d x4, x2, 15
0x7681C444, // xvseqi.d x4, x2, -15
0x2C406002, // vst v2, 24(r0): FP-relative keeps the zero base
0x2C006002, // vld v2, 24(r0)
0x2C5FD802, // vst v2, 2038(r0)
0x2CC06006, // xvst x6, 24(r0)
0x4C000020,
)
// Misaligned vldrepl offsets are rejected, as the toolchain does.
for i, src := range []string{
`TEXT ·e(SB), NOSPLIT, $0
VMOVQ 3(R4), V1.H8
RET
`,
`TEXT ·e(SB), NOSPLIT, $0
MOVW $4(R4), F1
RET
`,
} {
fn := firstTextLOONG64(t, src)
if _, _, _, _, _, err := assembleLOONG64(fn); err == nil {
t.Errorf("case %d: expected an error, got none", i)
}
}
}
// TestLOONG64_bytePseudo pins the BYTE literal-data pseudo-op, which the
// loong64 toolchain does not spell but the arm64 and riscv64 encoders of
// this package already accept for byte-exact data layout (a superset
// spelling, shippable via the goobj path).
func TestLOONG64_bytePseudo(t *testing.T) {
fn := firstTextLOONG64(t, `#include "textflag.h"
TEXT ·bytes(SB), NOSPLIT, $0
BYTE $2
BYTE $1; BYTE $0
BYTE $255
RET
`)
code := assembleLOONG64Helper(t, fn)
// Four literal bytes, then RET (jirl r0, r1, 0); the trailing bytes pad
// the final word the way any sub-word tail does.
want := []byte{2, 1, 0, 0xFF, 0x20, 0x00, 0x00, 0x4C}
if !bytes.Equal(code[:len(want)], want) {
t.Errorf("bytes = % x, want % x", code, want)
}
for _, src := range []string{
`TEXT ·e(SB), NOSPLIT, $0
BYTE $256
RET
`,
`TEXT ·e(SB), NOSPLIT, $0
BYTE $-1
RET
`,
} {
fn := firstTextLOONG64(t, src)
if _, _, _, _, _, err := assembleLOONG64(fn); err == nil {
t.Errorf("%q: expected an error, got none", src)
}
}
}
+1 -1
View File
@@ -6,7 +6,7 @@ package asm
import (
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
)
// Loong64 frame mapping, matching the Go toolchain's loong64 backend.
+1 -1
View File
@@ -7,7 +7,7 @@ import (
"bytes"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// TestLOONG64_sys exercises the no-operand system instructions and the
+1 -1
View File
@@ -6,7 +6,7 @@ package asm
import (
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// TestLOONG64RelocOffsetsIncludePrologue pins the function-relative
+30 -1
View File
@@ -17,13 +17,28 @@ import "strings"
// size. The high flag marks the legacy high-byte registers AH/CH/DH/BH, which
// occupy indices 4-7 yet take no REX prefix, unlike SPL/BPL/SIL/DIL that share
// those indices but require one. The mask flag marks the AVX-512 opmask
// registers K0-K7, the fp flag the x87 stack registers F0-F7.
// registers K0-K7, the fp flag the x87 stack registers F0-F7, the mmx flag the
// MMX registers M0-M7, the seg field a bare segment register (FS, GS) and the
// ctl field the control and debug registers, whose number rides an
// instruction's reg field rather than r/m.
type Reg struct {
idx int
size int // informational width implied by the name; the mnemonic decides
high bool // AH/CH/DH/BH
mask bool // K0-K7 opmask register
fp bool // F0-F7 x87 stack register
mmx bool // M0-M7 MMX register
seg int // segment register number plus one (ES=1..GS=6); 0 = not one
ctl byte // 0 none, 1 CRn control register, 2 DRn debug register
}
// segNumber returns the segment register number (ES=0..GS=5) when r names a
// bare segment register.
func (r Reg) segNumber() (int, bool) {
if r.seg == 0 {
return 0, false
}
return r.seg - 1, true
}
// Index returns the register number (0-15 for GPRs, 0-31 for vectors).
@@ -149,6 +164,20 @@ func buildRegByName() map[string]Reg {
for i := 0; i <= 7; i++ {
m["F"+itoa(i)] = Reg{idx: i, size: 8, fp: true}
}
// MMX: M0..M7.
for i := 0; i <= 7; i++ {
m["M"+itoa(i)] = Reg{idx: i, size: 8, mmx: true}
}
// Bare segment registers: ES, CS, SS, DS, FS, GS (the memory-base and
// index spellings of FS and GS are handled before register lookup).
for i, n := range []string{"ES", "CS", "SS", "DS", "FS", "GS"} {
m[n] = Reg{idx: i, size: 2, seg: i + 1}
}
// Control and debug registers: CR0..CR15, DR0..DR15.
for i := 0; i <= 15; i++ {
m["CR"+itoa(i)] = Reg{idx: i, size: 8, ctl: 1}
m["DR"+itoa(i)] = Reg{idx: i, size: 8, ctl: 2}
}
return m
}
+1 -1
View File
@@ -10,7 +10,7 @@ import (
"slices"
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
)
// assembleRISCV assembles a RISC-V TEXT function body into machine code.
+2 -2
View File
@@ -10,8 +10,8 @@ import (
"strings"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// firstTextRISCV parses assembly source and returns the first TEXT function body.
+1 -1
View File
@@ -7,7 +7,7 @@ import (
"fmt"
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
)
// RISC-V frame mapping, matching the Go toolchain's riscv64 backend.
+1 -1
View File
@@ -6,7 +6,7 @@ package asm
import (
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// TestRISCVFrameSpadjAndLines checks that a framed function records its
+1 -1
View File
@@ -13,7 +13,7 @@ import (
"strings"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// TestGOObjectRISCVCallReloc checks that CALL sym(SB) emits a single JAL
+122 -3
View File
@@ -76,6 +76,10 @@ const (
// carries a vector length, so the register the L'L field follows is the
// XMM source.
vexExtractGPR
// vexBlend4 is the four-operand variable blend `OP mask, src2, src1,
// dst` (VPBLENDVB): ModRM.reg = dst (op3), VEX.vvvv = src1 (op2),
// ModRM.rm = src2 (op1) and the mask register in the /is4 byte (op0).
vexBlend4
)
// vexSpec describes one VEX instruction's encoding parameters.
@@ -202,8 +206,28 @@ var vexTable = map[string]vexSpec{
// VEX.128/256.66.0F.WIG, immediate shuffle (reg=dst, rm=src, imm8).
"VPSHUFD": {1, 0x70, 0, 1, -1, vexImmRM},
// VEX.256.66.0F3A.W1, qword permute (reg=dst, rm=src, imm8).
// VEX.256.66.0F3A.W1, qword permute (reg=dst, rm=src, imm8), and its
// double twin under op 01; the in-lane permutes under 04/05.
"VPERMQ": {3, 0x00, 1, 1, -1, vexImmRM},
"VPERMPD": {3, 0x01, 1, 1, -1, vexImmRM},
"VPERMILPS": {3, 0x04, 0, 1, -1, vexImmRM},
"VPERMILPD": {3, 0x05, 0, 1, -1, vexImmRM},
// VEX.66.0F3A.W0, the immediate-controlled AVX tail: the rounding
// pair, the AES key assistant and the string compares.
"VROUNDPD": {3, 0x09, 0, 1, -1, vexImmRM},
"VROUNDPS": {3, 0x08, 0, 1, -1, vexImmRM},
"VAESKEYGENASSIST": {3, 0xDF, 0, 1, -1, vexImmRM},
"VPCMPESTRI": {3, 0x61, 0, 1, -1, vexImmRM},
"VPCMPESTRM": {3, 0x60, 0, 1, -1, vexImmRM},
"VPCMPISTRI": {3, 0x63, 0, 1, -1, vexImmRM},
"VPCMPISTRM": {3, 0x62, 0, 1, -1, vexImmRM},
// VEX.128.66.0F3A.W0, the scalar lane extract to a GPR or memory
// (reg = the XMM source, r/m = the destination).
"VEXTRACTPS": {3, 0x17, 0, 1, -1, vexExtractGPR},
"VPEXTRW": {3, 0x15, 0, 1, -1, vexExtractGPR},
// VEX.128.66.0F3A.W0, the four-operand variable blend with its mask
// register in the /is4 byte.
"VPBLENDVB": {3, 0x4C, 0, 1, -1, vexBlend4},
// VEX.128/256.66.0F.WIG, two-source shuffle (reg=dst, vvvv=src1, rm=src2,
// imm8).
@@ -524,8 +548,16 @@ func isVex(mnemUpper string) bool {
if _, ok := vexTable[mnemUpper]; ok {
return true
}
_, ok := vexMoveTable[mnemUpper]
return ok
if _, ok := vexMoveTable[mnemUpper]; ok {
return true
}
// The dual-shape moves (VMOVHPD/VMOVLPD) pick their VEX form by operand
// count in encodeVex.
switch mnemUpper {
case "VMOVHPD", "VMOVLPD":
return true
}
return false
}
// encodeVex encodes a VEX instruction with operands in Plan 9 order.
@@ -559,6 +591,22 @@ func (e *enc) encodeVex(mnemUpper string, ops []Operand) error {
return e.encodeVexNDS3(vexSpec{mapSel: 1, opcode: op, pp: 1, opdigit: -1, form: vexNDS3}, ops)
}
}
// The high/low double moves split by operand count: three operands
// load-and-insert (mem, src, dst, an NDS form), two store (xmm, m64,
// the reversed store layout).
if mnemUpper == "VMOVHPD" || mnemUpper == "VMOVLPD" {
loadOp, storeOp := byte(0x16), byte(0x17)
if mnemUpper == "VMOVLPD" {
loadOp, storeOp = 0x12, 0x13
}
switch len(ops) {
case 3:
return e.encodeVexNDS3(vexSpec{mapSel: 1, opcode: loadOp, w: 0, pp: 1, opdigit: -1, form: vexNDS3}, ops)
case 2:
return e.encodeVexRMRev(vexSpec{mapSel: 1, opcode: storeOp, w: 0, pp: 1, opdigit: -1, form: vexRMRev}, ops)
}
return fmt.Errorf("%s expects 2 or 3 operands, got %d", mnemUpper, len(ops))
}
spec := vexTable[mnemUpper]
switch spec.form {
case vexNDS3:
@@ -573,6 +621,10 @@ func (e *enc) encodeVex(mnemUpper string, ops []Operand) error {
return e.encodeVexNDS3Imm(spec, ops)
case vexExtract:
return e.encodeVexExtract(spec, ops)
case vexExtractGPR:
return e.encodeVexExtractGPR(spec, ops)
case vexBlend4:
return e.encodeVexBlend4(spec, ops)
case vexRMSrcLen:
return e.encodeVexRMSrcLen(mnemUpper, spec, ops)
case vexZero:
@@ -964,6 +1016,73 @@ func (e *enc) encodeVexRMRev(spec vexSpec, ops []Operand) error {
return e.emitVexFields(spec, srcReg.vecLenBit(), srcReg.idx&7, rBit, 15, ops[1])
}
// encodeVexExtractGPR encodes the lane extract to a general-purpose register
// or memory (VEXTRACTPS): OP $imm, xsrc, gpr/mem with the XMM source in
// ModRM.reg and the destination in r/m, L = 0.
func (e *enc) encodeVexExtractGPR(spec vexSpec, ops []Operand) error {
if len(ops) != 3 {
return fmt.Errorf("extract expects 3 operands ($imm, xsrc, dst), got %d", len(ops))
}
imm, src, dst := ops[0], ops[1], ops[2]
immVal, ok := imm.(Imm)
if !ok {
return fmt.Errorf("extract lane must be an immediate")
}
srcReg, ok := src.(Reg)
if !ok || !srcReg.isVec() || srcReg.size != 16 {
return fmt.Errorf("extract source must be an XMM register")
}
if _, isReg := dst.(Reg); !isReg && !memOperand(dst) {
return fmt.Errorf("extract destination must be a register or memory")
}
rBit := 0
if srcReg.idx >= 8 {
rBit = 1
}
if err := e.emitVexFields(spec, 0, srcReg.idx&7, rBit, 15, dst); err != nil {
return err
}
immByte, err := imm8(int64(immVal))
if err != nil {
return err
}
e.out = append(e.out, immByte)
return nil
}
// encodeVexBlend4 encodes the four-operand variable blend (VPBLENDVB):
// OP mask, src2, src1, dst with ModRM.reg = dst, VEX.vvvv = src1, r/m =
// src2 and the mask XMM register in the trailing /is4 byte.
func (e *enc) encodeVexBlend4(spec vexSpec, ops []Operand) error {
if len(ops) != 4 {
return fmt.Errorf("blend expects 4 operands (mask, src2, src1, dst), got %d", len(ops))
}
mask, src2, src1, dst := ops[0], ops[1], ops[2], ops[3]
maskReg, ok := mask.(Reg)
if !ok || !maskReg.isVec() || maskReg.size != 16 {
return fmt.Errorf("blend mask must be an XMM register")
}
vvvvReg, ok := src1.(Reg)
if !ok || !vvvvReg.isVec() {
return fmt.Errorf("blend second source must be a vector register")
}
dstReg, ok := dst.(Reg)
if !ok || !dstReg.isVec() {
return fmt.Errorf("blend destination must be a vector register")
}
rBit := 0
if dstReg.idx >= 8 {
rBit = 1
}
if err := e.emitVexFields(spec, dstReg.vecLenBit(), dstReg.idx&7, rBit, 15-(vvvvReg.idx&15), src2); err != nil {
return err
}
// The /is4 byte names the mask register: bits [3:0] its low nibble,
// bit 7 the fourth register bit (X8-X15).
e.out = append(e.out, byte(maskReg.idx&7)|byte((maskReg.idx&8)<<4))
return nil
}
// encodeVexMove encodes a two-operand move (VMOVDQU, VMOVUPD, VMOVD, VMOVQ,
// VMOVSD), picking the direction-specific opcode and VEX.W. A vector→vector
// move uses the store-form layout (reg = source, rm = destination), matching
+1 -1
View File
@@ -8,7 +8,7 @@
// to the arch package; the AST records syntax only.
package ast
import "sourcedock.dev/petrbalvin/gasm-devkit/token"
import "sourcedock.dev/petrbalvin/gasm-sdk/token"
// File is the parsed representation of one .s source file.
type File struct {
+1 -1
View File
@@ -6,7 +6,7 @@ package ast
import (
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/token"
"sourcedock.dev/petrbalvin/gasm-sdk/token"
)
func pos(line, col int) token.Position { return token.Position{Line: line, Column: col} }
+1 -1
View File
@@ -17,7 +17,7 @@ import (
"regexp"
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
)
// go_asm.h is the header the Go compiler writes for every package that
+3 -3
View File
@@ -16,9 +16,9 @@ import (
"strconv"
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/arch"
"sourcedock.dev/petrbalvin/gasm-devkit/asm"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/asm"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// cmdAuditInstructions cross-checks a gasm encoder against the Go toolchain's
+1 -1
View File
@@ -7,7 +7,7 @@ import (
"runtime"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
)
func TestDerivedFamily(t *testing.T) {
+2 -2
View File
@@ -1,7 +1,7 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build !linux
//go:build !(linux || (freebsd && (amd64 || arm64 || riscv64)))
package main
@@ -11,6 +11,6 @@ import (
)
func cmdDebug(args []string) int {
fmt.Fprintln(os.Stderr, "gasm debug: the interactive debugger requires Linux (ptrace)")
fmt.Fprintln(os.Stderr, "gasm debug: the interactive debugger requires Linux or FreeBSD (ptrace)")
return 1
}
@@ -1,7 +1,7 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build linux
//go:build linux || (freebsd && (amd64 || arm64 || riscv64))
package main
@@ -13,8 +13,8 @@ import (
"strings"
"time"
"sourcedock.dev/petrbalvin/gasm-devkit/debug"
"sourcedock.dev/petrbalvin/gasm-devkit/verify"
"sourcedock.dev/petrbalvin/gasm-sdk/debug"
"sourcedock.dev/petrbalvin/gasm-sdk/verify"
)
func cmdDebug(args []string) int {
@@ -239,10 +239,12 @@ REPL commands:
fmt.Printf("gasm debug: cover: stopped on signal %v\n", sig)
break
}
reason, _ := sess.StopInfo()
regs, rerr := sess.GetRegs()
if rerr != nil {
break
}
trapPC := regs.GetPC()
// HandleTrap restores the original byte, rewinds PC and counts
// the hit on the breakpoint itself. Single-step over the
// restored instruction so the reinsertion at the top of the
@@ -251,6 +253,12 @@ REPL commands:
if err := sess.Step(); err != nil {
break
}
} else if debug.TrapStray(sess, reason, trapPC) {
// A breakpoint-class trap that matches none of ours and left
// the PC in place: resuming would re-execute the trapping
// instruction forever, so the coverage run stops here.
fmt.Printf("gasm debug: cover: SIGTRAP at %#x matches no breakpoint; the PC did not advance\n", trapPC)
break
}
}
hits := map[uint64]int{}
+3 -3
View File
@@ -9,9 +9,9 @@ import (
"sort"
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/arch"
"sourcedock.dev/petrbalvin/gasm-devkit/disasm"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/disasm"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// cmdDis disassembles machine code: either a raw binary (standard input with
+9 -9
View File
@@ -27,15 +27,15 @@ import (
"sync"
"syscall"
"sourcedock.dev/petrbalvin/gasm-devkit/arch"
"sourcedock.dev/petrbalvin/gasm-devkit/asm"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-devkit/format"
"sourcedock.dev/petrbalvin/gasm-devkit/lexer"
"sourcedock.dev/petrbalvin/gasm-devkit/lint"
"sourcedock.dev/petrbalvin/gasm-devkit/lsp"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-devkit/verify"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/asm"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/format"
"sourcedock.dev/petrbalvin/gasm-sdk/lexer"
"sourcedock.dev/petrbalvin/gasm-sdk/lint"
"sourcedock.dev/petrbalvin/gasm-sdk/lsp"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/verify"
)
// version reports the release the toolchain recorded for this build: the
+2 -2
View File
@@ -14,8 +14,8 @@ import (
"syscall"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/arch"
"sourcedock.dev/petrbalvin/gasm-devkit/asm"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/asm"
)
const clean = "#include \"textflag.h\"\n" +
+2 -2
View File
@@ -11,8 +11,8 @@ import (
"os"
"strings"
gasmast "sourcedock.dev/petrbalvin/gasm-devkit/ast"
gasmparser "sourcedock.dev/petrbalvin/gasm-devkit/parser"
gasmast "sourcedock.dev/petrbalvin/gasm-sdk/ast"
gasmparser "sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
// cmdScaffold generates a differential test skeleton for every kernel in a
+31 -7
View File
@@ -1,13 +1,17 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build linux
//go:build linux || (freebsd && (amd64 || arm64 || riscv64))
package debug
import "strings"
import "fmt"
import (
"cmp"
"fmt"
"slices"
)
// Breakpoint is one software breakpoint in the debuggee.
type Breakpoint struct {
@@ -149,15 +153,16 @@ func (bm *Breakpoints) SetWithCond(addr uint64, label string, cond *Condition) (
return bp, nil
}
// Info returns a formatted list of all breakpoints.
// Info returns a formatted list of all breakpoints, ordered by address so
// the numbering is stable across calls (map iteration order is not).
func (bm *Breakpoints) Info() string {
if len(bm.bps) == 0 {
return "no breakpoints set\n"
}
var result strings.Builder
i := 0
for _, bp := range bm.bps {
i++
bps := bm.All()
slices.SortFunc(bps, func(a, b *Breakpoint) int { return cmp.Compare(a.Addr, b.Addr) })
for i, bp := range bps {
status := "enabled"
if !bp.Enabled {
status = "disabled"
@@ -170,7 +175,7 @@ func (bm *Breakpoints) Info() string {
if bp.Cond != nil {
cond = " if " + bp.Cond.String()
}
result.WriteString(fmt.Sprintf(" %d: %s at %#x [%s, %d hits]%s\n", i, label, bp.Addr, status, bp.hits, cond))
result.WriteString(fmt.Sprintf(" %d: %s at %#x [%s, %d hits]%s\n", i+1, label, bp.Addr, status, bp.hits, cond))
}
return result.String()
}
@@ -238,6 +243,25 @@ func (bm *Breakpoints) All() []*Breakpoint {
// Hits returns how many times the breakpoint has been hit.
func (bp *Breakpoint) Hits() int { return bp.hits }
// TrapStray reports whether a stop is a breakpoint-class trap that matches
// no breakpoint of ours and cannot be resumed: the PC still stands on the
// trapping instruction (the kernel's own BRK, EBREAK or break, on an
// architecture that reports the trap in place), so the next resume would
// re-execute it and trap forever. trapPC is the PC the stop reported,
// before any HandleTrap rewinding; reason is the stop's StopInfo class.
// The debuggee's SIGSTOP barriers also stop without PC movement, and they
// never carry the breakpoint class, so they are unaffected.
func TrapStray(s *Session, reason StopReason, trapPC uint64) bool {
if reason != StopBreakpoint {
return false
}
after, err := s.GetRegs()
if err != nil {
return false
}
return after.GetPC() <= trapPC-uint64(breakpointPCAdjust)
}
func (bm *Breakpoints) HandleTrap(regs *Regs) *Breakpoint {
// On amd64 the kernel reports the trap with RIP past the INT3; on the
// other supported architectures the PC still stands on the trap
+44
View File
@@ -10,6 +10,8 @@ package debug
// build on every supported linux architecture.
import (
"fmt"
"slices"
"strings"
"testing"
)
@@ -263,3 +265,45 @@ func TestConditionString(t *testing.T) {
}
}
}
// TestBreakpointsInfoOrdered proves the listing is ordered by address: the
// numbers it prints are map keys rendered in iteration order otherwise, so
// the same set of breakpoints would renumber itself between calls.
func TestBreakpointsInfoOrdered(t *testing.T) {
tr := newMockTracer()
bm := NewBreakpoints(tr)
addrs := []uint64{0x9000, 0x1000, 0x7000, 0x3000, 0x8000, 0x2000,
0x6000, 0x4000, 0x5000, 0xa000}
for i, a := range addrs {
if _, err := bm.Set(a, fmt.Sprintf("bp%d", i)); err != nil {
t.Fatalf("Set(%#x): %v", a, err)
}
}
sorted := append([]uint64(nil), addrs...)
slices.Sort(sorted)
info := bm.Info()
for i, a := range sorted {
want := fmt.Sprintf(" %d: bp%d at %#x", i+1, indexOf(addrs, a), a)
if !strings.Contains(info, want) {
t.Errorf("Info() missing %q; listing:\n%s", want, info)
}
}
// The numbers themselves must ascend: "1:" before "2" ... "10".
pos := 0
for i := range len(addrs) {
next := strings.Index(info[pos:], fmt.Sprintf(" %d: ", i+1))
if next < 0 {
t.Fatalf("Info() has no entry %d; listing:\n%s", i+1, info)
}
pos += next
}
}
func indexOf(addrs []uint64, a uint64) int {
for i, v := range addrs {
if v == a {
return i
}
}
return -1
}
+325
View File
@@ -0,0 +1,325 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build linux && amd64
package debug
import (
"fmt"
"os"
"runtime"
"strings"
"testing"
"time"
)
// Regression tests for the debugger audit: memory access at mapping
// boundaries, watchpoint slot attribution, launch failure latency, stray
// trap instructions and the REPL's argument validation. All drive a real
// ptrace session, so they run on amd64 hosts only.
// memMap is one line of /proc/pid/maps.
type memMap struct {
lo, hi uint64
perms string
name string
}
// readMaps parses the debuggee's memory map.
func readMaps(t *testing.T, pid int) []memMap {
t.Helper()
data, err := os.ReadFile(fmt.Sprintf("/proc/%d/maps", pid))
if err != nil {
t.Fatalf("read maps: %v", err)
}
var out []memMap
for line := range strings.SplitSeq(string(data), "\n") {
fields := strings.Fields(line)
if len(fields) < 2 {
continue
}
var lo, hi uint64
if _, err := fmt.Sscanf(fields[0], "%x-%x", &lo, &hi); err != nil {
continue
}
m := memMap{lo: lo, hi: hi, perms: fields[1]}
if len(fields) >= 6 {
m.name = fields[5]
}
out = append(out, m)
}
return out
}
// boundaryByte returns the last byte of a writable, ordinary mapping that is
// followed by an unmapped gap: an access there is inside the mapping, while
// the 8-byte word starting at it crosses into unmapped memory.
func boundaryByte(t *testing.T, pid int) uint64 {
t.Helper()
maps := readMaps(t, pid)
for i, m := range maps {
if !strings.Contains(m.perms, "rw") ||
strings.Contains(m.name, "vvar") || strings.Contains(m.name, "vdso") ||
strings.Contains(m.name, "vsyscall") {
continue
}
gap := uint64(1) << 62
if i+1 < len(maps) {
gap = maps[i+1].lo - m.hi
}
if gap >= 4096 {
return m.hi - 1
}
}
t.Skip("no writable mapping followed by a hole; cannot construct the boundary")
return 0
}
// TestReadMemoryPageBoundary proves ReadMemory never reads past the requested
// range: one byte at the end of a mapping followed by a hole must be
// readable, which the old word-at-a-time tail read failed because its final
// 8-byte Peek crossed into the unmapped page.
func TestReadMemoryPageBoundary(t *testing.T) {
sess, _, _ := launchKernel(t, buildGasm(t), boundaryKernel(t), "boundary", nil)
addr := boundaryByte(t, sess.Pid())
mem, err := sess.ReadMemory(addr, 1)
if err != nil {
t.Fatalf("ReadMemory(%#x, 1): %v (the read must not cross into the unmapped page)", addr, err)
}
if len(mem) != 1 {
t.Fatalf("ReadMemory returned %d bytes, want 1", len(mem))
}
// A request whose own range crosses into the hole must still fail.
if _, err := sess.ReadMemory(addr, 8); err == nil {
t.Fatal("ReadMemory past the mapping end should fail")
}
}
// TestDisassemblePageBoundary proves the disassembler shrinks its read
// window at a mapping end instead of failing: the instruction stream cannot
// be decoded at all when the fixed 15-byte read crosses into the hole.
func TestDisassemblePageBoundary(t *testing.T) {
sess, _, _ := launchKernel(t, buildGasm(t), boundaryKernel(t), "boundary", nil)
addr := boundaryByte(t, sess.Pid())
if _, _, err := sess.Disassemble(addr); err != nil {
t.Fatalf("Disassemble(%#x): %v (the read window must shrink at the mapping end)", addr, err)
}
}
// TestWriteMemoryPageBoundary proves WriteMemory writes exactly the bytes it
// is given: one byte at the end of a mapping followed by a hole must be
// writable, which the old read-modify-write of the final partial word failed
// because its Peek crossed into the unmapped page.
func TestWriteMemoryPageBoundary(t *testing.T) {
sess, _, _ := launchKernel(t, buildGasm(t), boundaryKernel(t), "boundary", nil)
addr := boundaryByte(t, sess.Pid())
orig, err := sess.ReadMemory(addr, 1)
if err != nil {
t.Fatalf("ReadMemory(%#x, 1): %v", addr, err)
}
if err := sess.WriteMemory(addr, []byte{orig[0]}); err != nil {
t.Fatalf("WriteMemory(%#x, 1): %v (the write must not read past the range)", addr, err)
}
}
// TestWatchpointSlotAttribution proves a hit is attributed to the slot that
// fired, not to an earlier one whose DR6 status bit is still set: the B0-B3
// bits are sticky, so they must be acknowledged when read.
func TestWatchpointSlotAttribution(t *testing.T) {
bin := buildGasm(t)
const kernel = `#include "textflag.h"
// func wptwo(x, y int64) (a, b int64)
TEXT ·wptwo(SB), NOSPLIT, $0-32
MOVQ $0x1111, AX
MOVQ AX, a+16(FP)
MOVQ $0x2222, BX
MOVQ BX, b+24(FP)
RET
`
path := writeKernel(t, kernel)
sess, bm, fl := launchKernel(t, bin, path, "wptwo", nil)
entry := sess.CodeBase() + uint64(fl.Offset)
if _, err := bm.Set(entry, "entry"); err != nil {
t.Fatalf("Set: %v", err)
}
runToEntry(t, sess, bm, entry)
regs, err := sess.GetRegs()
if err != nil {
t.Fatalf("GetRegs: %v", err)
}
// FP sits one word above the entry stack pointer (the return address
// occupies [RSP]), so a+16(FP) = RSP+24 and b+24(FP) = RSP+32.
watchA := regs.RSP + 24
watchB := regs.RSP + 32
if err := sess.SetWatchpoint(0, watchA, WatchWrite, 8); err != nil {
t.Fatalf("SetWatchpoint(0): %v", err)
}
if err := sess.SetWatchpoint(1, watchB, WatchWrite, 8); err != nil {
t.Fatalf("SetWatchpoint(1): %v", err)
}
for i, want := range []uint64{watchA, watchB} {
if err := sess.Continue(); err != nil {
t.Fatalf("Continue (hit %d): %v", i+1, err)
}
reason, addr := sess.StopInfo()
if reason != StopWatchpoint {
t.Fatalf("hit %d: stop reason = %v, want StopWatchpoint", i+1, reason)
}
if addr != want {
t.Fatalf("hit %d reported %#x, want %#x (the sticky DR6 bit misattributes the slot)", i+1, addr, want)
}
}
// Clearing a watchpoint must zero its address register: a stale
// address in a disabled slot turns any sticky status bit into a
// misattributed report later.
if err := sess.ClearWatchpoint(0); err != nil {
t.Fatalf("ClearWatchpoint(0): %v", err)
}
dr0, err := ptracePeekUser(sess.Pid(), drOffset)
if err != nil {
t.Fatalf("read DR0: %v", err)
}
if dr0 != 0 {
t.Fatalf("DR0 = %#x after ClearWatchpoint, want 0 (the address register must be cleared)", dr0)
}
}
// TestLaunchFailsFastOnDeadDebuggee proves a debuggee that dies before
// signalling readiness surfaces promptly: the ready poll used to run its
// full 2.5 seconds before the wait discovered the exit.
func TestLaunchFailsFastOnDeadDebuggee(t *testing.T) {
runtime.LockOSThread()
defer runtime.UnlockOSThread()
bin := buildGasm(t)
path := boundaryKernel(t)
start := time.Now()
sess, err := Launch(bin, path, "nosuchfunction", nil)
elapsed := time.Since(start)
if err == nil {
sess.Kill()
t.Fatal("Launch with an unknown function should fail")
}
if !strings.Contains(err.Error(), "before signalling readiness") &&
!strings.Contains(err.Error(), "debuggee exited") {
t.Errorf("error does not name the dead debuggee: %v", err)
}
if elapsed >= 1500*time.Millisecond {
t.Fatalf("Launch took %v to report the dead debuggee; the readiness poll must detect the exit, not time out", elapsed)
}
}
// TestStrayTrapRunsThrough proves the continue loop survives a trap
// instruction planted in the kernel itself (BYTE $0xCC, the same byte the
// debugger patches in): on architectures that report the trap in place the
// loop must surface the stop, and on amd64 it runs through to the exit. A
// regression here hangs, so a watchdog fails the run.
func TestStrayTrapRunsThrough(t *testing.T) {
bin := buildGasm(t)
const kernel = `#include "textflag.h"
// func stray() int64
TEXT ·stray(SB), NOSPLIT, $0-8
MOVQ $7, AX
BYTE $0xCC
MOVQ AX, ret+0(FP)
RET
`
path := writeKernel(t, kernel)
sess, bm, _ := launchKernel(t, bin, path, "stray", nil)
timer := time.AfterFunc(time.Minute, func() {
panic("watchdog: the continue loop hung on the stray trap instruction")
})
defer timer.Stop()
out := captureStdout(t, func() {
REPL(sess, bm, sess.CodeBase(), 0, 0, 0, nil, nil,
strings.NewReader("continue\nquit\n"))
})
if !strings.Contains(out, "debuggee exited") {
t.Errorf("the stray trap wedged the continue loop; output:\n%s", out)
}
}
// TestStepIntoFaultReportsSignal proves the step command reports a genuine
// signal-delivery-stop instead of silently printing the faulting
// instruction as if the step had succeeded.
func TestStepIntoFaultReportsSignal(t *testing.T) {
bin := buildGasm(t)
const kernel = `#include "textflag.h"
// func crash() int64
TEXT ·crash(SB), NOSPLIT, $0-8
XORQ AX, AX
MOVQ (AX), AX
MOVQ AX, ret+0(FP)
RET
`
path := writeKernel(t, kernel)
sess, bm, fl := launchKernel(t, bin, path, "crash", nil)
entry := sess.CodeBase() + uint64(fl.Offset)
if _, err := bm.Set(entry, "entry"); err != nil {
t.Fatalf("Set: %v", err)
}
runToEntry(t, sess, bm, entry)
out := captureStdout(t, func() {
REPL(sess, bm, sess.CodeBase(), fl.Offset, fl.Size, fl.Args, nil, nil,
strings.NewReader("step 2\nquit\n"))
})
if !strings.Contains(out, "stopped on signal") {
t.Errorf("stepping into the fault did not report the signal; output:\n%s", out)
}
}
// TestREPLRejectsBadArguments proves the command loop reports malformed
// input instead of silently defaulting: an unknown label for x would read
// address 0, and a malformed count would silently step one instruction.
func TestREPLRejectsBadArguments(t *testing.T) {
bin := buildGasm(t)
path := boundaryKernel(t)
sess, bm, fl := launchKernel(t, bin, path, "boundary", nil)
entry := sess.CodeBase() + uint64(fl.Offset)
if _, err := bm.Set(entry, "entry"); err != nil {
t.Fatalf("Set: %v", err)
}
runToEntry(t, sess, bm, entry)
out := captureStdout(t, func() {
REPL(sess, bm, sess.CodeBase(), fl.Offset, fl.Size, fl.Args, nil, nil,
strings.NewReader("x nosuchlabel\nstep abc\ndisas abc\nwatch 0x1000 q 8\nquit\n"))
})
for _, want := range []string{
"unknown address: nosuchlabel",
"invalid count: abc",
"unknown watchpoint type: q",
} {
if !strings.Contains(out, want) {
t.Errorf("output missing %q:\n%s", want, out)
}
}
if got := strings.Count(out, "invalid count: abc"); got != 2 {
t.Errorf("invalid count reported %d times, want 2 (step and disas):\n%s", got, out)
}
}
// boundaryKernel is a minimal kernel for the boundary tests, which only need
// a live, stopped debuggee.
func boundaryKernel(t *testing.T) string {
t.Helper()
const kernel = `#include "textflag.h"
// func boundary() int64
TEXT ·boundary(SB), NOSPLIT, $0-8
MOVQ $1, AX
MOVQ AX, ret+0(FP)
RET
`
return writeKernel(t, kernel)
}
+65
View File
@@ -0,0 +1,65 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && amd64
package debug
import (
"fmt"
"strings"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/disasm"
)
// Disassemble decodes the instruction at the given address in the debuggee's
// memory and returns its text representation and length in bytes. An amd64
// instruction is up to 15 bytes long, but the read must not reach past the
// end of the mapping: when the full 15-byte window crosses into unmapped
// memory the window shrinks, because an instruction at the mapping's end is
// by construction no longer than the readable bytes that hold it.
func (s *Session) Disassemble(addr uint64) (string, int, error) {
var lastErr error
for _, n := range []int{15, 8, 4, 2, 1} {
mem, err := s.ReadMemory(addr, n)
if err != nil {
lastErr = err
continue
}
ins, derr := disasm.Decode(arch.AMD64, mem, addr)
if derr != nil {
return "", 0, derr
}
return ins.Text, ins.Len, nil
}
return "", 0, lastErr
}
// DisassembleN decodes up to n instructions starting at addr and returns
// them as a formatted string with addresses and byte offsets.
func (s *Session) DisassembleN(addr uint64, n int) string {
var result strings.Builder
pc := addr
for range n {
text, length, err := s.Disassemble(pc)
if err != nil {
result.WriteString(fmt.Sprintf(" %#08x: <error: %v>\n", pc, err))
break
}
result.WriteString(fmt.Sprintf(" %#08x: %s\n", pc, text))
if length == 0 {
length = 1
}
pc += uint64(length)
}
return result.String()
}
// isCallInsn reports whether disassembled text (x86asm.IntelSyntax) is a
// call. The first token must match exactly: a prefix test would also catch
// unrelated mnemonics.
func isCallInsn(text string) bool {
m, _, _ := strings.Cut(text, " ")
return strings.ToLower(m) == "call"
}
+60
View File
@@ -0,0 +1,60 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && arm64
package debug
import (
"fmt"
"strings"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/disasm"
)
// Disassemble decodes the instruction at the given address in the debuggee's
// memory and returns its text representation and length in bytes.
func (s *Session) Disassemble(addr uint64) (string, int, error) {
mem, err := s.ReadMemory(addr, 4)
if err != nil {
return "", 0, err
}
ins, err := disasm.Decode(arch.ARM64, mem, addr)
if err != nil {
return "", 0, err
}
return ins.Text, ins.Len, nil
}
// DisassembleN decodes up to n instructions starting at addr and returns
// them as a formatted string with addresses and byte offsets.
func (s *Session) DisassembleN(addr uint64, n int) string {
var result strings.Builder
pc := addr
for range n {
text, length, err := s.Disassemble(pc)
if err != nil {
result.WriteString(fmt.Sprintf(" %#08x: <error: %v>\n", pc, err))
break
}
result.WriteString(fmt.Sprintf(" %#08x: %s\n", pc, text))
if length == 0 {
length = 1
}
pc += uint64(length)
}
return result.String()
}
// isCallInsn reports whether disassembled text (arm64asm.GoSyntax) is a
// call. GoSyntax renders bl as CALL; the native mnemonic is accepted too.
// The first token must match exactly so branches never match.
func isCallInsn(text string) bool {
m, _, _ := strings.Cut(text, " ")
switch strings.ToLower(m) {
case "call", "bl":
return true
}
return false
}
+70
View File
@@ -0,0 +1,70 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && riscv64
package debug
import (
"fmt"
"strings"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/disasm"
)
// Disassemble decodes the instruction at the given address in the debuggee's
// memory and returns its text representation and length in bytes. The read
// shrinks from 4 to 2 bytes when the full word crosses into unmapped memory:
// a compressed instruction at the mapping's end still fits the shorter
// window, and an instruction can never extend past the mapping that holds it.
func (s *Session) Disassemble(addr uint64) (string, int, error) {
var lastErr error
for _, n := range []int{4, 2} {
mem, err := s.ReadMemory(addr, n)
if err != nil {
lastErr = err
continue
}
ins, derr := disasm.Decode(arch.RISCV, mem, addr)
if derr != nil {
return "", 0, derr
}
return ins.Text, ins.Len, nil
}
return "", 0, lastErr
}
// DisassembleN decodes up to n instructions starting at addr and returns
// them as a formatted string with addresses and byte offsets.
func (s *Session) DisassembleN(addr uint64, n int) string {
var result strings.Builder
pc := addr
for range n {
text, length, err := s.Disassemble(pc)
if err != nil {
result.WriteString(fmt.Sprintf(" %#08x: <error: %v>\n", pc, err))
break
}
result.WriteString(fmt.Sprintf(" %#08x: %s\n", pc, text))
if length == 0 {
length = 1
}
pc += uint64(length)
}
return result.String()
}
// isCallInsn reports whether disassembled text (riscv64asm.GoSyntax) is a
// call. GoSyntax renders jal and jalr calls as CALL; the native mnemonics
// are accepted too. The first token must match exactly: a prefix test on
// "bl" would catch branches on other architectures, and jalr as ret prints
// RET, which must not be stepped over.
func isCallInsn(text string) bool {
m, _, _ := strings.Cut(text, " ")
switch strings.ToLower(m) {
case "call", "jal", "jalr":
return true
}
return false
}
+17 -8
View File
@@ -9,22 +9,31 @@ import (
"fmt"
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/arch"
"sourcedock.dev/petrbalvin/gasm-devkit/disasm"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/disasm"
)
// Disassemble decodes the instruction at the given address in the debuggee's
// memory and returns its text representation and length in bytes.
// memory and returns its text representation and length in bytes. An amd64
// instruction is up to 15 bytes long, but the read must not reach past the
// end of the mapping: when the full 15-byte window crosses into unmapped
// memory the window shrinks, because an instruction at the mapping's end is
// by construction no longer than the readable bytes that hold it.
func (s *Session) Disassemble(addr uint64) (string, int, error) {
mem, err := s.ReadMemory(addr, 15)
var lastErr error
for _, n := range []int{15, 8, 4, 2, 1} {
mem, err := s.ReadMemory(addr, n)
if err != nil {
return "", 0, err
lastErr = err
continue
}
ins, err := disasm.Decode(arch.AMD64, mem, addr)
if err != nil {
return "", 0, err
ins, derr := disasm.Decode(arch.AMD64, mem, addr)
if derr != nil {
return "", 0, derr
}
return ins.Text, ins.Len, nil
}
return "", 0, lastErr
}
// DisassembleN decodes up to n instructions starting at addr and returns
+2 -2
View File
@@ -9,8 +9,8 @@ import (
"fmt"
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/arch"
"sourcedock.dev/petrbalvin/gasm-devkit/disasm"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/disasm"
)
// Disassemble decodes the instruction at the given address in the debuggee's
+2 -2
View File
@@ -9,8 +9,8 @@ import (
"fmt"
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/arch"
"sourcedock.dev/petrbalvin/gasm-devkit/disasm"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/disasm"
)
// Disassemble decodes the instruction at the given address in the debuggee's
+16 -8
View File
@@ -9,22 +9,30 @@ import (
"fmt"
"strings"
"sourcedock.dev/petrbalvin/gasm-devkit/arch"
"sourcedock.dev/petrbalvin/gasm-devkit/disasm"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/disasm"
)
// Disassemble decodes the instruction at the given address in the debuggee's
// memory and returns its text representation and length in bytes.
// memory and returns its text representation and length in bytes. The read
// shrinks from 4 to 2 bytes when the full word crosses into unmapped memory:
// a compressed instruction at the mapping's end still fits the shorter
// window, and an instruction can never extend past the mapping that holds it.
func (s *Session) Disassemble(addr uint64) (string, int, error) {
mem, err := s.ReadMemory(addr, 4)
var lastErr error
for _, n := range []int{4, 2} {
mem, err := s.ReadMemory(addr, n)
if err != nil {
return "", 0, err
lastErr = err
continue
}
ins, err := disasm.Decode(arch.RISCV, mem, addr)
if err != nil {
return "", 0, err
ins, derr := disasm.Decode(arch.RISCV, mem, addr)
if derr != nil {
return "", 0, derr
}
return ins.Text, ins.Len, nil
}
return "", 0, lastErr
}
// DisassembleN decodes up to n instructions starting at addr.
+136
View File
@@ -0,0 +1,136 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && amd64
package debug
import "fmt"
func printRegs(regs *Regs, codeBase, funcOff uint64) {
fmt.Printf(" RIP = %#016x (func+%#x)\n", regs.RIP, regs.RIP-codeBase-funcOff)
fmt.Printf(" RSP = %#016x RBP = %#016x\n", regs.RSP, regs.RBP)
fmt.Printf(" RAX = %#016x RBX = %#016x\n", regs.RAX, regs.RBX)
fmt.Printf(" RCX = %#016x RDX = %#016x\n", regs.RCX, regs.RDX)
fmt.Printf(" RSI = %#016x RDI = %#016x\n", regs.RSI, regs.RDI)
fmt.Printf(" R8 = %#016x R9 = %#016x\n", regs.R8, regs.R9)
fmt.Printf(" R10 = %#016x R11 = %#016x\n", regs.R10, regs.R11)
fmt.Printf(" R12 = %#016x R13 = %#016x\n", regs.R12, regs.R13)
fmt.Printf(" R14 = %#016x R15 = %#016x\n", regs.R14, regs.R15)
fmt.Printf(" RFLAGS = %#x [%s]\n", regs.RFLAGS, decodeRflags(regs.RFLAGS))
}
func printVectorRegs(v *VectorRegs) {
fmt.Println("\n Vector registers (YMM):")
for i := 0; i < 16; i += 2 {
fmt.Printf(" YMM%-2d = ", i)
printYMM(v.YMM[i][:])
fmt.Printf(" YMM%-2d = ", i+1)
printYMM(v.YMM[i+1][:])
fmt.Println()
}
}
func printYMM(b []byte) {
for j := 0; j < 32; j += 4 {
v := uint32(b[j]) | uint32(b[j+1])<<8 | uint32(b[j+2])<<16 | uint32(b[j+3])<<24
fmt.Printf("%08x ", v)
}
}
func decodeRflags(f uint64) string {
var flags string
if f&1 != 0 {
flags += "CF "
}
if f&(1<<2) != 0 {
flags += "PF "
}
if f&(1<<4) != 0 {
flags += "AF "
}
if f&(1<<6) != 0 {
flags += "ZF "
}
if f&(1<<7) != 0 {
flags += "SF "
}
if f&(1<<8) != 0 {
flags += "TF "
}
if f&(1<<9) != 0 {
flags += "IF "
}
if f&(1<<10) != 0 {
flags += "DF "
}
if f&(1<<11) != 0 {
flags += "OF "
}
if flags == "" {
return "none"
}
return flags[:len(flags)-1]
}
// SetReg modifies a register value in the debuggee.
func (s *Session) SetReg(name string, value uint64) error {
regs, err := s.GetRegs()
if err != nil {
return err
}
switch name {
case "rax", "eax", "ax", "al":
regs.RAX = value
case "rbx", "ebx", "bx", "bl":
regs.RBX = value
case "rcx", "ecx", "cx", "cl":
regs.RCX = value
case "rdx", "edx", "dx", "dl":
regs.RDX = value
case "rsi", "esi", "si":
regs.RSI = value
case "rdi", "edi", "di":
regs.RDI = value
case "rbp", "ebp", "bp":
regs.RBP = value
case "rsp", "esp", "sp":
regs.RSP = value
case "r8":
regs.R8 = value
case "r9":
regs.R9 = value
case "r10":
regs.R10 = value
case "r11":
regs.R11 = value
case "r12":
regs.R12 = value
case "r13":
regs.R13 = value
case "r14":
regs.R14 = value
case "r15":
regs.R15 = value
case "rip", "eip":
regs.RIP = value
default:
return fmt.Errorf("debug: unknown register %q", name)
}
return s.SetRegs(&regs)
}
// archReturnAddr reads the return address of the current frame (amd64
// ABI0 convention). A function that contains a CALL (or has a frame) is
// assembled with the prologue PUSHQ BP; MOVQ SP, BP, so mid-function the
// word at SP is the saved caller BP, a stack address, and the return
// address sits further up. Walk the stack from SP and take the first word
// that lies in an executable mapping: stack and data words never do, a
// return address always does. FreeBSD exposes no mapping list, so the
// walk degenerates to the raw entry convention, [SP] before any push.
func archReturnAddr(s *Session, regs *Regs) (uint64, error) {
return s.Peek(regs.RSP)
}
// archSPLabel returns the SP register name for display.
func archSPLabel() string { return "RSP" }
+127
View File
@@ -0,0 +1,127 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && arm64
package debug
import (
"encoding/binary"
"fmt"
)
func printRegs(regs *Regs, codeBase, funcOff uint64) {
fmt.Printf(" PC = %#016x (func+%#x)\n", regs.PC, regs.PC-codeBase-funcOff)
fmt.Printf(" SP = %#016x FP = %#016x\n", regs.SP, regs.X29)
fmt.Printf(" LR = %#016x\n", regs.X30)
fmt.Printf(" X0 = %#016x X1 = %#016x\n", regs.X0, regs.X1)
fmt.Printf(" X2 = %#016x X3 = %#016x\n", regs.X2, regs.X3)
fmt.Printf(" X4 = %#016x X5 = %#016x\n", regs.X4, regs.X5)
fmt.Printf(" X6 = %#016x X7 = %#016x\n", regs.X6, regs.X7)
fmt.Printf(" X8 = %#016x X9 = %#016x\n", regs.X8, regs.X9)
fmt.Printf(" X10 = %#016x X11 = %#016x\n", regs.X10, regs.X11)
fmt.Printf(" X12 = %#016x X13 = %#016x\n", regs.X12, regs.X13)
fmt.Printf(" X14 = %#016x X15 = %#016x\n", regs.X14, regs.X15)
fmt.Printf(" X16 = %#016x X17 = %#016x\n", regs.X16, regs.X17)
fmt.Printf(" X18 = %#016x X19 = %#016x\n", regs.X18, regs.X19)
fmt.Printf(" X20 = %#016x X21 = %#016x\n", regs.X20, regs.X21)
fmt.Printf(" X22 = %#016x X23 = %#016x\n", regs.X22, regs.X23)
fmt.Printf(" X24 = %#016x X25 = %#016x\n", regs.X24, regs.X25)
fmt.Printf(" X26 = %#016x X27 = %#016x\n", regs.X26, regs.X27)
fmt.Printf(" X28 = %#016x PSTATE = %#x\n", regs.X28, regs.PSTATE)
}
func printVectorRegs(v *VectorRegs) {
fmt.Println("\n Vector registers (V0-V31):")
for i := 0; i < 32; i += 2 {
fmt.Printf(" V%-2d = %016x%016x\n", i, binary.LittleEndian.Uint64(v.V[i][8:16]), binary.LittleEndian.Uint64(v.V[i][0:8]))
fmt.Printf(" V%-2d = %016x%016x\n", i+1, binary.LittleEndian.Uint64(v.V[i+1][8:16]), binary.LittleEndian.Uint64(v.V[i+1][0:8]))
}
}
// SetReg modifies a register value in the debuggee.
func (s *Session) SetReg(name string, value uint64) error {
regs, err := s.GetRegs()
if err != nil {
return err
}
switch name {
case "x0":
regs.X0 = value
case "x1":
regs.X1 = value
case "x2":
regs.X2 = value
case "x3":
regs.X3 = value
case "x4":
regs.X4 = value
case "x5":
regs.X5 = value
case "x6":
regs.X6 = value
case "x7":
regs.X7 = value
case "x8":
regs.X8 = value
case "x9":
regs.X9 = value
case "x10":
regs.X10 = value
case "x11":
regs.X11 = value
case "x12":
regs.X12 = value
case "x13":
regs.X13 = value
case "x14":
regs.X14 = value
case "x15":
regs.X15 = value
case "x16":
regs.X16 = value
case "x17":
regs.X17 = value
case "x18":
regs.X18 = value
case "x19":
regs.X19 = value
case "x20":
regs.X20 = value
case "x21":
regs.X21 = value
case "x22":
regs.X22 = value
case "x23":
regs.X23 = value
case "x24":
regs.X24 = value
case "x25":
regs.X25 = value
case "x26":
regs.X26 = value
case "x27":
regs.X27 = value
case "x28":
regs.X28 = value
case "x29", "fp":
regs.X29 = value
case "x30", "lr":
regs.X30 = value
case "sp":
regs.SP = value
case "pc":
regs.PC = value
default:
return fmt.Errorf("debug: unknown register %q", name)
}
return s.SetRegs(&regs)
}
// archReturnAddr reads the return address from LR (arm64 convention).
func archReturnAddr(s *Session, regs *Regs) (uint64, error) {
return regs.X30, nil
}
// archSPLabel returns the SP register name for display.
func archSPLabel() string { return "SP" }
+121
View File
@@ -0,0 +1,121 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && riscv64
package debug
import "fmt"
func printRegs(regs *Regs, codeBase, funcOff uint64) {
fmt.Printf(" PC = %#016x (func+%#x)\n", regs.PC, regs.PC-codeBase-funcOff)
fmt.Printf(" SP = %#016x FP = %#016x\n", regs.Sp, regs.S0)
fmt.Printf(" RA = %#016x\n", regs.Ra)
fmt.Printf(" A0 = %#016x A1 = %#016x\n", regs.A0, regs.A1)
fmt.Printf(" A2 = %#016x A3 = %#016x\n", regs.A2, regs.A3)
fmt.Printf(" A4 = %#016x A5 = %#016x\n", regs.A4, regs.A5)
fmt.Printf(" A6 = %#016x A7 = %#016x\n", regs.A6, regs.A7)
fmt.Printf(" T0 = %#016x T1 = %#016x\n", regs.T0, regs.T1)
fmt.Printf(" T2 = %#016x T3 = %#016x\n", regs.T2, regs.T3)
fmt.Printf(" T4 = %#016x T5 = %#016x\n", regs.T4, regs.T5)
fmt.Printf(" T6 = %#016x\n", regs.T6)
fmt.Printf(" S1 = %#016x S2 = %#016x\n", regs.S1, regs.S2)
fmt.Printf(" S3 = %#016x S4 = %#016x\n", regs.S3, regs.S4)
fmt.Printf(" S5 = %#016x S6 = %#016x\n", regs.S5, regs.S6)
fmt.Printf(" S7 = %#016x S8 = %#016x\n", regs.S7, regs.S8)
fmt.Printf(" S9 = %#016x S10 = %#016x\n", regs.S9, regs.S10)
fmt.Printf(" S11 = %#016x\n", regs.S11)
}
func printVectorRegs(v *VectorRegs) {
fmt.Println("\n FP registers (F0-F31):")
for i := 0; i < 32; i += 2 {
fmt.Printf(" F%-2d = %#018x F%-2d = %#018x\n", i, v.F[i], i+1, v.F[i+1])
}
fmt.Printf(" FCSR = %#x\n", v.FCSR)
}
// SetReg modifies a register value in the debuggee.
func (s *Session) SetReg(name string, value uint64) error {
regs, err := s.GetRegs()
if err != nil {
return err
}
switch name {
case "pc":
regs.PC = value
case "ra", "x1":
regs.Ra = value
case "sp", "x2":
regs.Sp = value
case "gp", "x3":
regs.Gp = value
case "tp", "x4":
regs.Tp = value
case "t0", "x5":
regs.T0 = value
case "t1", "x6":
regs.T1 = value
case "t2", "x7":
regs.T2 = value
case "s0", "fp", "x8":
regs.S0 = value
case "s1", "x9":
regs.S1 = value
case "a0", "x10":
regs.A0 = value
case "a1", "x11":
regs.A1 = value
case "a2", "x12":
regs.A2 = value
case "a3", "x13":
regs.A3 = value
case "a4", "x14":
regs.A4 = value
case "a5", "x15":
regs.A5 = value
case "a6", "x16":
regs.A6 = value
case "a7", "x17":
regs.A7 = value
case "s2", "x18":
regs.S2 = value
case "s3", "x19":
regs.S3 = value
case "s4", "x20":
regs.S4 = value
case "s5", "x21":
regs.S5 = value
case "s6", "x22":
regs.S6 = value
case "s7", "x23":
regs.S7 = value
case "s8", "x24":
regs.S8 = value
case "s9", "x25":
regs.S9 = value
case "s10", "x26":
regs.S10 = value
case "s11", "x27":
regs.S11 = value
case "t3", "x28":
regs.T3 = value
case "t4", "x29":
regs.T4 = value
case "t5", "x30":
regs.T5 = value
case "t6", "x31":
regs.T6 = value
default:
return fmt.Errorf("debug: unknown register %q", name)
}
return s.SetRegs(&regs)
}
// archReturnAddr reads the return address from RA (riscv64 convention).
func archReturnAddr(s *Session, regs *Regs) (uint64, error) {
return regs.Ra, nil
}
// archSPLabel returns the SP register name for display.
func archSPLabel() string { return "SP" }
+2 -2
View File
@@ -17,8 +17,8 @@ import (
"time"
"unsafe"
"sourcedock.dev/petrbalvin/gasm-devkit/asm"
"sourcedock.dev/petrbalvin/gasm-devkit/verify"
"sourcedock.dev/petrbalvin/gasm-sdk/asm"
"sourcedock.dev/petrbalvin/gasm-sdk/verify"
)
// Integration tests beyond the basic entry breakpoint: hardware watchpoints,
+310
View File
@@ -0,0 +1,310 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && (amd64 || arm64 || riscv64)
package debug
import (
"fmt"
"os"
"os/exec"
"path/filepath"
"runtime"
"strings"
"syscall"
"time"
"golang.org/x/sys/unix"
)
// Session is a ptrace debugging session controlling one debuggee process.
// The FreeBSD implementation sits behind the same surface as the Linux one:
// PT_TRACE_ME from the debuggee, PT_CONTINUE/PT_STEP from the tracer, and
// tracee memory through PT_IO (FreeBSD has no /proc/pid/mem to fall back
// on, so PT_IO is the only supported route).
type Session struct {
pid int
cmd *exec.Cmd
stopped bool
exited bool
codeBase uint64 // base address of the JIT code in the debuggee
tmpDir string // scratch directory of the session, removed on Kill
wpSlots [16]bool // hardware watchpoint slots in use (DR0-DR3, arm64 dbw 0-15)
// lastSignal holds the signal of the most recent stop when that stop
// was a genuine signal-delivery-stop the caller must see (a fault such
// as SIGSEGV, SIGBUS, SIGFPE or SIGILL); 0 for breakpoint traps,
// single-steps, SIGSTOP and suppressed runtime signals.
lastSignal syscall.Signal
}
// Launch starts the debuggee subprocess (gasm debug --target ...) and
// attaches to it via ptrace.
func Launch(gasmBin, asmPath, funcName string, args []byte) (*Session, error) {
sess, _, err := LaunchWithBuffers(gasmBin, asmPath, funcName, args, "")
return sess, err
}
// LaunchWithBuffers is like Launch but also allocates buffers in the debuggee.
//
// It pins the calling goroutine to its OS thread and leaves it pinned: the
// debuggee's PT_TRACE_ME binds the tracer relation to the forking thread,
// and every ptrace request on the session must come from that same thread.
// All Session methods must therefore be called from the goroutine that
// launched the session (the REPL and coverage loops do exactly that).
func LaunchWithBuffers(gasmBin, asmPath, funcName string, args []byte, bufSpec string) (*Session, []uint64, error) {
runtime.LockOSThread() // ptrace requests must stay on the forking thread
self, err := os.Executable()
if err != nil {
return nil, nil, fmt.Errorf("debug: cannot find gasm binary: %w", err)
}
if gasmBin != "" {
self = gasmBin
}
tmpDir, err := os.MkdirTemp("", "gasm-debug-*")
if err != nil {
return nil, nil, fmt.Errorf("debug: tempdir: %w", err)
}
argsFile := filepath.Join(tmpDir, "args.bin")
if err := os.WriteFile(argsFile, args, 0o644); err != nil {
os.RemoveAll(tmpDir)
return nil, nil, fmt.Errorf("debug: write args: %w", err)
}
if bufSpec != "" {
if err := os.WriteFile(filepath.Join(tmpDir, "bufspec"), []byte(bufSpec), 0o644); err != nil {
os.RemoveAll(tmpDir)
return nil, nil, fmt.Errorf("debug: write bufspec: %w", err)
}
}
cmd := exec.Command(self, "debug", "--func", funcName, "--args", argsFile, asmPath)
cmd.Env = append(os.Environ(), "GASM_DEBUG_TARGET=1", "GASM_DEBUG_TMP="+tmpDir)
cmd.Stdout = nil
cmd.Stderr = os.Stderr
cmd.SysProcAttr = &syscall.SysProcAttr{}
if err := cmd.Start(); err != nil {
os.RemoveAll(tmpDir)
return nil, nil, fmt.Errorf("debug: start debuggee: %w", err)
}
s := &Session{pid: cmd.Process.Pid, cmd: cmd, tmpDir: tmpDir}
readyFile := filepath.Join(tmpDir, "ready")
for range 500 {
if _, err := os.Stat(readyFile); err == nil {
break
}
// A debuggee that died before signalling readiness (unknown
// function, unparseable source) writes its failure notice to the
// handshake directory; read it and fail fast. The poll never
// waits on the child: a wait here could consume the SIGSTOP park
// that waitStopped below must receive, hanging the launch.
if err := s.deadReason(); err != nil {
cmd.Wait()
os.RemoveAll(tmpDir)
return nil, nil, err
}
time.Sleep(5 * time.Millisecond)
}
// The debuggee parks itself with SIGSTOP once the JIT code is mapped.
// A Go tracee also reports SIGURG preemption as signal-delivery-stops,
// so the wait loops until a stop the debugger cares about instead of
// assuming the first event is the SIGSTOP.
if _, err := s.waitStopped(); err != nil {
cmd.Process.Kill()
os.RemoveAll(tmpDir)
return nil, nil, fmt.Errorf("debug: wait for debuggee: %w", err)
}
s.stopped = true
// The debuggee reports its JIT mapping in the codebase file; that is
// the supported path on FreeBSD, where no /proc/pid/maps exists to
// scan for the RWX region as a fallback.
if data, err := os.ReadFile(filepath.Join(tmpDir, "codebase")); err == nil {
fmt.Sscanf(string(data), "%d", &s.codeBase)
}
var bufAddrs []uint64
if bufSpec != "" {
addrFile := filepath.Join(tmpDir, "bufaddrs")
if data, err := os.ReadFile(addrFile); err == nil {
for line := range strings.SplitSeq(strings.TrimSpace(string(data)), "\n") {
var addr uint64
if _, err := fmt.Sscanf(line, "%d", &addr); err == nil {
bufAddrs = append(bufAddrs, addr)
}
}
}
}
return s, bufAddrs, nil
}
// deadReason reports the debuggee's own failure notice, the file its
// failure paths write before exiting. A debuggee killed without a notice
// (a crash, SIGKILL) surfaces through waitStopped after the poll instead,
// which is why the poll's budget stays finite.
func (s *Session) deadReason() error {
data, err := os.ReadFile(filepath.Join(s.tmpDir, "dead"))
if err != nil {
return nil
}
return fmt.Errorf("debug: debuggee failed before signalling readiness: %s", strings.TrimSpace(string(data)))
}
// waitStopped consumes ptrace-stop events until one the debugger cares
// about arrives: SIGTRAP (a breakpoint or a completed single-step), the
// debuggee's own SIGSTOP, or a genuine signal-delivery-stop. A Go tracee's
// runtime raises SIGURG for asynchronous preemption, and every signal on a
// traced thread surfaces as a signal-delivery-stop, so SIGURG is suppressed
// and the tracee resumed without it. Every other signal (SIGSEGV, SIGBUS,
// SIGFPE, SIGILL, ...) is returned to the caller: resuming with signal 0
// would restart the faulting instruction and fault forever, so a faulting
// kernel must surface as a stop the caller reports.
func (s *Session) waitStopped() (syscall.Signal, error) {
for {
var ws syscall.WaitStatus
if _, err := syscall.Wait4(s.pid, &ws, syscall.WUNTRACED, nil); err != nil {
return 0, err
}
if ws.Exited() {
s.exited = true
return 0, fmt.Errorf("debuggee exited with status %d", ws.ExitStatus())
}
if ws.Signaled() {
s.exited = true
return 0, fmt.Errorf("debuggee killed by signal %v", ws.Signal())
}
switch sig := ws.StopSignal(); sig {
case syscall.SIGTRAP, syscall.SIGSTOP:
s.stopped = true
s.lastSignal = 0
return sig, nil
case syscall.SIGURG:
// Go runtime asynchronous preemption: resume the tracee
// without delivering the signal.
s.lastSignal = 0
if err := unix.PtraceCont(s.pid, 0); err != nil {
return 0, fmt.Errorf("debug: PT_CONTINUE: %w", err)
}
default:
// A genuine signal-delivery-stop. Report it; the caller
// decides how to proceed.
s.stopped = true
s.lastSignal = sig
return sig, nil
}
}
}
// LastSignal returns the signal of the most recent stop when that stop was
// a genuine signal-delivery-stop (a fault such as SIGSEGV, SIGFPE, SIGILL
// or SIGBUS), and 0 for breakpoint traps, single-steps, SIGSTOP and
// suppressed runtime signals.
func (s *Session) LastSignal() syscall.Signal { return s.lastSignal }
// Peek reads a word (8 bytes) from the debuggee's memory at addr, through
// PT_IO with PIOD_READ_D.
func (s *Session) Peek(addr uint64) (uint64, error) {
var buf [8]byte
if _, err := unix.PtraceIO(unix.PIOD_READ_D, s.pid, uintptr(addr), buf[:], len(buf)); err != nil {
return 0, fmt.Errorf("debug: read mem %#x: %w", addr, err)
}
return uint64(buf[0]) | uint64(buf[1])<<8 | uint64(buf[2])<<16 | uint64(buf[3])<<24 |
uint64(buf[4])<<32 | uint64(buf[5])<<40 | uint64(buf[6])<<48 | uint64(buf[7])<<56, nil
}
// Poke writes a word (8 bytes) to the debuggee's memory at addr, through
// PT_IO with PIOD_WRITE_D.
func (s *Session) Poke(addr, val uint64) error {
buf := []byte{byte(val), byte(val >> 8), byte(val >> 16), byte(val >> 24),
byte(val >> 32), byte(val >> 40), byte(val >> 48), byte(val >> 56)}
if _, err := unix.PtraceIO(unix.PIOD_WRITE_D, s.pid, uintptr(addr), buf, len(buf)); err != nil {
return fmt.Errorf("debug: write mem %#x: %w", addr, err)
}
return nil
}
// ReadMemory reads len bytes from the debuggee's memory at addr in one
// PT_IO request, the shape the request is built for.
func (s *Session) ReadMemory(addr uint64, length int) ([]byte, error) {
out := make([]byte, length)
n, err := unix.PtraceIO(unix.PIOD_READ_D, s.pid, uintptr(addr), out, length)
return out[:n], err
}
// WriteMemory writes bytes to the debuggee's memory at addr in one PT_IO
// request.
func (s *Session) WriteMemory(addr uint64, data []byte) error {
_, err := unix.PtraceIO(unix.PIOD_WRITE_D, s.pid, uintptr(addr), data, len(data))
return err
}
// Step executes a single instruction in the debuggee.
func (s *Session) Step() error {
if s.exited {
return fmt.Errorf("debug: debuggee has exited")
}
if err := unix.PtraceSingleStep(s.pid); err != nil {
return fmt.Errorf("debug: PT_STEP: %w", err)
}
_, err := s.waitStopped()
return err
}
// Continue resumes execution until the next breakpoint or exit.
func (s *Session) Continue() error {
if s.exited {
return fmt.Errorf("debug: debuggee has exited")
}
if err := unix.PtraceCont(s.pid, 0); err != nil {
return fmt.Errorf("debug: PT_CONTINUE: %w", err)
}
_, err := s.waitStopped()
return err
}
// Exited returns true if the debuggee has terminated.
func (s *Session) Exited() bool { return s.exited }
// Pid returns the debuggee's process ID.
func (s *Session) Pid() int { return s.pid }
// CodeBase returns the base address of the JIT code in the debuggee.
func (s *Session) CodeBase() uint64 { return s.codeBase }
// Kill terminates the debuggee and removes the session's scratch
// directory, so a successful session leaves no gasm-debug-* debris behind.
func (s *Session) Kill() {
if !s.exited {
syscall.Kill(s.pid, syscall.SIGKILL)
syscall.Wait4(s.pid, nil, 0, nil)
s.exited = true
}
if s.cmd != nil && s.cmd.Process != nil {
s.cmd.Wait()
}
if s.tmpDir != "" {
os.RemoveAll(s.tmpDir)
s.tmpDir = ""
}
}
// execRange is one executable mapping of the debuggee.
type execRange struct {
lo, hi uint64
}
// execRanges is a stub on FreeBSD: there is no /proc/pid/maps to parse,
// and procfs(5) is not guaranteed to be mounted. The callers degrade
// gracefully: archReturnAddr falls back to the raw stack convention and
// the mapping scan is skipped.
func execRanges(pid int) []execRange { return nil }
// findRWXMapping is a stub on FreeBSD for the same reason: the codebase
// handshake file is the supported way the JIT region is located.
func findRWXMapping(pid int) uint64 { return 0 }
+160
View File
@@ -0,0 +1,160 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && amd64
package debug
import (
"encoding/binary"
"fmt"
"unsafe"
"golang.org/x/sys/unix"
)
// GetRegs reads the general-purpose registers of the stopped debuggee and
// converts the FreeBSD struct reg into the portable layout.
func (s *Session) GetRegs() (Regs, error) {
var ur unix.Reg
if err := unix.PtraceGetRegs(s.pid, &ur); err != nil {
return Regs{}, fmt.Errorf("debug: PT_GETREGS: %w", err)
}
return Regs{
R15: uint64(ur.R15),
R14: uint64(ur.R14),
R13: uint64(ur.R13),
R12: uint64(ur.R12),
R11: uint64(ur.R11),
R10: uint64(ur.R10),
R9: uint64(ur.R9),
R8: uint64(ur.R8),
RDI: uint64(ur.Rdi),
RSI: uint64(ur.Rsi),
RBP: uint64(ur.Rbp),
RBX: uint64(ur.Rbx),
RDX: uint64(ur.Rdx),
RCX: uint64(ur.Rcx),
RAX: uint64(ur.Rax),
RIP: uint64(ur.Rip),
CS: uint64(ur.Cs),
RFLAGS: uint64(ur.Rflags),
RSP: uint64(ur.Rsp),
SS: uint64(ur.Ss),
FS: uint64(ur.Fs),
GS: uint64(ur.Gs),
DS: uint64(ur.Ds),
ES: uint64(ur.Es),
}, nil
}
// SetRegs writes the general-purpose registers of the stopped debuggee.
func (s *Session) SetRegs(regs *Regs) error {
// Read-modify-write keeps the fields FreeBSD owns (trapno, err) intact.
var ur unix.Reg
if err := unix.PtraceGetRegs(s.pid, &ur); err != nil {
return fmt.Errorf("debug: PT_GETREGS: %w", err)
}
ur.R15 = int64(regs.R15)
ur.R14 = int64(regs.R14)
ur.R13 = int64(regs.R13)
ur.R12 = int64(regs.R12)
ur.R11 = int64(regs.R11)
ur.R10 = int64(regs.R10)
ur.R9 = int64(regs.R9)
ur.R8 = int64(regs.R8)
ur.Rdi = int64(regs.RDI)
ur.Rsi = int64(regs.RSI)
ur.Rbp = int64(regs.RBP)
ur.Rbx = int64(regs.RBX)
ur.Rdx = int64(regs.RDX)
ur.Rcx = int64(regs.RCX)
ur.Rax = int64(regs.RAX)
ur.Rip = int64(regs.RIP)
ur.Cs = int64(regs.CS)
ur.Rflags = int64(regs.RFLAGS)
ur.Rsp = int64(regs.RSP)
ur.Ss = int64(regs.SS)
return unix.PtraceSetRegs(s.pid, &ur)
}
// FPRegs holds the x87 FPU and SSE (XMM) register state, the FXSAVE image
// the FreeBSD struct fpreg mirrors: XMM0-15 at the same offsets.
type FPRegs struct {
XMM [16][16]byte // XMM0-15
}
// GetFPRegs retrieves the FPU/SSE register state via PT_GETFPREGS. The
// FreeBSD struct fpreg mirrors the FXSAVE image: the x87 environment and
// stack in Env/Acc, XMM0-15 in Xacc.
func (s *Session) GetFPRegs() (FPRegs, error) {
var fp FPRegs
var fr unix.FpReg
if err := unix.PtraceGetFpRegs(s.pid, &fr); err != nil {
return fp, fmt.Errorf("debug: PT_GETFPREGS: %w", err)
}
for i := range 16 {
copy(fp.XMM[i][:], fr.Xacc[i][:])
}
return fp, nil
}
// VectorRegs holds the YMM register state.
type VectorRegs struct {
YMM [16][32]byte // YMM0-15 (full 256-bit values)
}
// The XSAVE area the PT_GETXSTATE request returns follows the architectural
// layout (Intel SDM vol 1, "XSAVE"): the 512-byte legacy FXSAVE image (x87
// state in 0-159, XMM0-15 in 160-511), then the 64-byte xsave header whose
// first 8 bytes are xstate_bv, then one component per set feature bit, each
// 64-byte aligned. The YMM high halves are the first extended component,
// at offset 576; XFEATURE_STATE_BIT_AVX is bit 2 of xstate_bv.
const (
xsaveXMMOffset = 160
xsaveHeaderOffset = 512
xsaveBVOffset = xsaveHeaderOffset
ymmOffset = xsaveHeaderOffset + 64 // 576
ymmSize = 256 // 16 registers, 16 bytes each
xfeatureMaskYMM = 1 << 2
xstateMaxBuffer = 4096 // PT_GETXSTATE_INFO bounds the size far below this
)
// GetVectorRegs retrieves the YMM registers via PT_GETXSTATE. The low
// (XMM) halves always come from the legacy image; the high halves are
// copied only when xstate_bv reports the AVX state, and read as zero
// otherwise. When the request fails the FP image still provides correct
// XMM halves, so that is the fallback.
func (s *Session) GetVectorRegs() (VectorRegs, error) {
var v VectorRegs
buf := make([]byte, xstateMaxBuffer)
n, _, errno := unix.Syscall6(
unix.SYS_PTRACE,
uintptr(unix.PT_GETXSTATE),
uintptr(s.pid),
0,
uintptr(unsafe.Pointer(&buf[0])),
0, 0,
)
if errno != 0 {
fp, err := s.GetFPRegs()
if err != nil {
return v, err
}
for i := range 16 {
copy(v.YMM[i][:16], fp.XMM[i][:])
}
return v, nil
}
for i := range 16 {
copy(v.YMM[i][:16], buf[xsaveXMMOffset+16*i:xsaveXMMOffset+16*i+16])
}
if int(n) >= ymmOffset+ymmSize {
if binary.LittleEndian.Uint64(buf[xsaveBVOffset:xsaveBVOffset+8])&xfeatureMaskYMM != 0 {
for i := range 16 {
copy(v.YMM[i][16:], buf[ymmOffset+16*i:ymmOffset+16*i+16])
}
}
}
return v, nil
}
+114
View File
@@ -0,0 +1,114 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && arm64
package debug
import (
"fmt"
"golang.org/x/sys/unix"
)
// GetRegs reads the general-purpose registers of the stopped debuggee and
// converts the FreeBSD struct reg (x[30], lr, sp, elr, spsr) into the
// portable layout.
func (s *Session) GetRegs() (Regs, error) {
var ur unix.Reg
if err := unix.PtraceGetRegs(s.pid, &ur); err != nil {
return Regs{}, fmt.Errorf("debug: PT_GETREGS: %w", err)
}
return Regs{
X0: ur.X[0],
X1: ur.X[1],
X2: ur.X[2],
X3: ur.X[3],
X4: ur.X[4],
X5: ur.X[5],
X6: ur.X[6],
X7: ur.X[7],
X8: ur.X[8],
X9: ur.X[9],
X10: ur.X[10],
X11: ur.X[11],
X12: ur.X[12],
X13: ur.X[13],
X14: ur.X[14],
X15: ur.X[15],
X16: ur.X[16],
X17: ur.X[17],
X18: ur.X[18],
X19: ur.X[19],
X20: ur.X[20],
X21: ur.X[21],
X22: ur.X[22],
X23: ur.X[23],
X24: ur.X[24],
X25: ur.X[25],
X26: ur.X[26],
X27: ur.X[27],
X28: ur.X[28],
X29: ur.X[29],
X30: ur.Lr,
SP: ur.Sp,
PC: ur.Elr,
PSTATE: uint64(ur.Spsr),
}, nil
}
// SetRegs writes the general-purpose registers of the stopped debuggee.
func (s *Session) SetRegs(regs *Regs) error {
var ur unix.Reg
ur.X = [30]uint64{
regs.X0, regs.X1, regs.X2, regs.X3, regs.X4, regs.X5, regs.X6,
regs.X7, regs.X8, regs.X9, regs.X10, regs.X11, regs.X12, regs.X13,
regs.X14, regs.X15, regs.X16, regs.X17, regs.X18, regs.X19, regs.X20,
regs.X21, regs.X22, regs.X23, regs.X24, regs.X25, regs.X26, regs.X27,
regs.X28, regs.X29,
}
ur.Lr = regs.X30
ur.Sp = regs.SP
ur.Elr = regs.PC
ur.Spsr = uint32(regs.PSTATE)
return unix.PtraceSetRegs(s.pid, &ur)
}
// FPRegs holds the arm64 FP/NEON register state: the 32 128-bit V
// registers, then FPSR and FPCR (the user_fpsimd shape).
type FPRegs struct {
V [32][16]byte // V0-V31 (128-bit NEON/FP registers)
FPSR uint32
FPCR uint32
}
// GetFPRegs retrieves the FP/NEON register state via PT_GETFPREGS. The
// FreeBSD struct fpreg holds the 32 128-bit V registers followed by FPSR
// and FPCR, the user_fpsimd shape.
func (s *Session) GetFPRegs() (FPRegs, error) {
var fp FPRegs
var fr unix.FpReg
if err := unix.PtraceGetFpRegs(s.pid, &fr); err != nil {
return fp, fmt.Errorf("debug: PT_GETFPREGS: %w", err)
}
for i := range 32 {
copy(fp.V[i][:], fr.Q[i][:])
}
return fp, nil
}
// VectorRegs holds the full SIMD register state.
type VectorRegs struct {
V [32][16]byte // V0-V31 (128-bit)
}
// GetVectorRegs retrieves the SIMD registers.
func (s *Session) GetVectorRegs() (VectorRegs, error) {
var v VectorRegs
fp, err := s.GetFPRegs()
if err != nil {
return v, err
}
copy(v.V[:][:], fp.V[:][:])
return v, nil
}
+115
View File
@@ -0,0 +1,115 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && riscv64
package debug
import (
"fmt"
"golang.org/x/sys/unix"
)
// GetRegs reads the general-purpose registers of the stopped debuggee and
// converts the FreeBSD struct reg into the portable layout. Sstatus rides
// the kernel's struct but the portable surface carries the GPRs and PC.
func (s *Session) GetRegs() (Regs, error) {
var ur unix.Reg
if err := unix.PtraceGetRegs(s.pid, &ur); err != nil {
return Regs{}, fmt.Errorf("debug: PT_GETREGS: %w", err)
}
return Regs{
PC: ur.Sepc,
Ra: ur.Ra,
Sp: ur.Sp,
Gp: ur.Gp,
Tp: ur.Tp,
T0: ur.T[0],
T1: ur.T[1],
T2: ur.T[2],
S0: ur.S[0],
S1: ur.S[1],
A0: ur.A[0],
A1: ur.A[1],
A2: ur.A[2],
A3: ur.A[3],
A4: ur.A[4],
A5: ur.A[5],
A6: ur.A[6],
A7: ur.A[7],
S2: ur.S[2],
S3: ur.S[3],
S4: ur.S[4],
S5: ur.S[5],
S6: ur.S[6],
S7: ur.S[7],
S8: ur.S[8],
S9: ur.S[9],
S10: ur.S[10],
S11: ur.S[11],
T3: ur.T[3],
T4: ur.T[4],
T5: ur.T[5],
T6: ur.T[6],
}, nil
}
// SetRegs writes the general-purpose registers of the stopped debuggee.
// Read-modify-write keeps sstatus, which the kernel owns, intact.
func (s *Session) SetRegs(regs *Regs) error {
var ur unix.Reg
if err := unix.PtraceGetRegs(s.pid, &ur); err != nil {
return fmt.Errorf("debug: PT_GETREGS: %w", err)
}
ur.Sepc = regs.PC
ur.Ra = regs.Ra
ur.Sp = regs.Sp
ur.Gp = regs.Gp
ur.Tp = regs.Tp
ur.T = [7]uint64{regs.T0, regs.T1, regs.T2, regs.T3, regs.T4, regs.T5, regs.T6}
ur.S = [12]uint64{regs.S0, regs.S1, regs.S2, regs.S3, regs.S4, regs.S5,
regs.S6, regs.S7, regs.S8, regs.S9, regs.S10, regs.S11}
ur.A = [8]uint64{regs.A0, regs.A1, regs.A2, regs.A3, regs.A4, regs.A5, regs.A6, regs.A7}
return unix.PtraceSetRegs(s.pid, &ur)
}
// FPRegs holds the RISC-V FP register state (32 64-bit FP registers plus
// fcsr).
type FPRegs struct {
F [32]uint64 // F0-F31 (64-bit FP registers)
FCSR uint32
}
// GetFPRegs retrieves the FP register state via PT_GETFPREGS. The FreeBSD
// struct fpreg carries each 64-bit FP register in a 128-bit slot (fp_x is
// the flat [64]-word area the x/sys type renders as [32][2]); the low word
// holds the register, and FCSR rides the tail.
func (s *Session) GetFPRegs() (FPRegs, error) {
var fp FPRegs
var fr unix.FpReg
if err := unix.PtraceGetFpRegs(s.pid, &fr); err != nil {
return fp, fmt.Errorf("debug: PT_GETFPREGS: %w", err)
}
for i := range 32 {
fp.F[i] = fr.X[i][0]
}
fp.FCSR = uint32(fr.Fcsr)
return fp, nil
}
// VectorRegs holds the FP register state shown by the regs command
// (riscv64 has 32 64-bit FP registers and fcsr).
type VectorRegs struct {
F [32]uint64
FCSR uint32
}
// GetVectorRegs retrieves the FP registers.
func (s *Session) GetVectorRegs() (VectorRegs, error) {
fp, err := s.GetFPRegs()
if err != nil {
return VectorRegs{}, err
}
return VectorRegs{F: fp.F, FCSR: fp.FCSR}, nil
}
+91
View File
@@ -0,0 +1,91 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && amd64
package debug
import (
"os/exec"
"path/filepath"
"runtime"
"testing"
"sourcedock.dev/petrbalvin/gasm-sdk/verify"
)
// TestLaunchAndBreakpoint is the FreeBSD twin of the Linux integration
// test: it drives the whole launch, breakpoint, trap and register-rewind
// flow end to end. It needs a real FreeBSD kernel (ptrace does not work
// under emulation), so it only runs where it can.
func TestLaunchAndBreakpoint(t *testing.T) {
if runtime.GOARCH != "amd64" {
t.Skip("runs only on amd64 hosts")
}
// The tracer is the OS thread that forked the debuggee (PT_TRACE_ME
// binds the relation to that thread); every ptrace request must come
// from the same thread, so pin the test goroutine to one thread.
runtime.LockOSThread()
defer runtime.UnlockOSThread()
bin := filepath.Join(t.TempDir(), "gasm")
out, err := exec.Command("go", "build", "-o", bin, "sourcedock.dev/petrbalvin/gasm-sdk/cmd/gasm").CombinedOutput()
if err != nil {
t.Fatalf("build gasm: %v: %s", err, out)
}
const kernelPath = "../testdata/verify/basic_amd64.s"
k, err := verify.Load(kernelPath)
if err != nil {
t.Fatalf("Load: %v", err)
}
t.Cleanup(k.Close)
fl, err := k.Func("wideCopy")
if err != nil {
t.Fatalf("Func: %v", err)
}
sess, err := Launch(bin, kernelPath, "wideCopy", make([]byte, fl.Args))
if err != nil {
t.Fatalf("Launch: %v", err)
}
t.Cleanup(sess.Kill)
bm := NewBreakpoints(sess)
entry := sess.CodeBase() + uint64(fl.Offset)
if _, err := bm.Set(entry, "entry"); err != nil {
t.Fatalf("Set: %v", err)
}
// The INT3 must be visible in the debuggee's memory.
word, err := sess.Peek(entry)
if err != nil {
t.Fatalf("Peek: %v", err)
}
if b := word & 0xFF; b != 0xCC {
t.Fatalf("int3 not patched: first byte %#02x at %#x", b, entry)
}
// The debuggee raises a second SIGSTOP after the launch barrier (the
// child's RunTarget marks its entry), so like the REPL and the cover
// mode the test keeps resuming until the breakpoint trap arrives.
for range 10 {
if err := sess.Continue(); err != nil {
t.Fatalf("Continue: %v", err)
}
if sess.Exited() {
t.Fatal("debuggee exited instead of trapping on the breakpoint")
}
regs, err := sess.GetRegs()
if err != nil {
t.Fatalf("GetRegs: %v", err)
}
if bp := bm.HandleTrap(&regs); bp != nil {
if bp.Addr != entry {
t.Fatalf("trap at %#x, want %#x", bp.Addr, entry)
}
return // trap on the entry breakpoint: the whole flow works
}
}
t.Fatal("no breakpoint trap after 10 resumes")
}
+11 -2
View File
@@ -14,17 +14,26 @@ import (
"strings"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/verify"
"sourcedock.dev/petrbalvin/gasm-sdk/verify"
)
// buildGasm produces the gasm binary the debugger spawns as its debuggee.
// Every live ptrace test funnels through here, so this is also where the
// deliberate-run boundary sits: under -short (the push pipeline's mode) the
// live sessions skip, because a real debuggee's launch handshake needs the
// machine to itself and a starved single-core runner turns each one into a
// timeout that burns the step's whole budget. The local test gate and the
// dispatched workflows run them in full.
func buildGasm(t *testing.T) string {
t.Helper()
if testing.Short() {
t.Skip("live ptrace session: skipped in -short mode")
}
if p := os.Getenv("GASM_TEST_BIN"); p != "" {
return p
}
bin := filepath.Join(t.TempDir(), "gasm")
cmd := exec.Command("go", "build", "-o", bin, "sourcedock.dev/petrbalvin/gasm-devkit/cmd/gasm")
cmd := exec.Command("go", "build", "-o", bin, "sourcedock.dev/petrbalvin/gasm-sdk/cmd/gasm")
out, err := cmd.CombinedOutput()
if err != nil {
t.Fatalf("build gasm: %v: %s", err, out)
+45 -22
View File
@@ -91,6 +91,16 @@ func LaunchWithBuffers(gasmBin, asmPath, funcName string, args []byte, bufSpec s
if _, err := os.Stat(readyFile); err == nil {
break
}
// A debuggee that died before signalling readiness (unknown
// function, unparseable source) writes its failure notice to the
// handshake directory; read it and fail fast. The poll never
// waits on the child: a wait here could consume the SIGSTOP park
// that waitStopped below must receive, hanging the launch.
if err := s.deadReason(); err != nil {
cmd.Wait()
os.RemoveAll(tmpDir)
return nil, nil, err
}
time.Sleep(5 * time.Millisecond)
}
@@ -131,6 +141,18 @@ func LaunchWithBuffers(gasmBin, asmPath, funcName string, args []byte, bufSpec s
return s, bufAddrs, nil
}
// deadReason reports the debuggee's own failure notice, the file its
// failure paths write before exiting. A debuggee killed without a notice
// (a crash, SIGKILL) surfaces through waitStopped after the poll instead,
// which is why the poll's budget stays finite.
func (s *Session) deadReason() error {
data, err := os.ReadFile(filepath.Join(s.tmpDir, "dead"))
if err != nil {
return nil
}
return fmt.Errorf("debug: debuggee failed before signalling readiness: %s", strings.TrimSpace(string(data)))
}
// waitStopped consumes ptrace-stop events until one the debugger cares
// about arrives: SIGTRAP (a breakpoint or a completed single-step), the
// debuggee's own SIGSTOP, or a genuine signal-delivery-stop. A Go tracee's
@@ -219,40 +241,41 @@ func (s *Session) Poke(addr, val uint64) error {
return nil
}
// ReadMemory reads len bytes from the debuggee's memory at addr.
// ReadMemory reads len bytes from the debuggee's memory at addr. The read
// covers exactly the requested range: the old word-at-a-time loop read a
// whole 8-byte word for the final partial word, so a request that ended
// inside the last mapped page failed whenever the following page was
// unmapped, even though every requested byte was readable.
func (s *Session) ReadMemory(addr uint64, length int) ([]byte, error) {
out := make([]byte, length)
for i := 0; i < length; i += 8 {
word, err := s.Peek(addr + uint64(i))
mem, err := os.OpenFile(fmt.Sprintf("/proc/%d/mem", s.pid), os.O_RDONLY, 0)
if err != nil {
return out[:i], err
}
for j := 0; j < 8 && i+j < length; j++ {
out[i+j] = byte(word >> (8 * j))
return out, fmt.Errorf("debug: open /proc/%d/mem: %w", s.pid, err)
}
defer mem.Close()
n, err := mem.ReadAt(out, int64(addr))
if err != nil {
return out[:n], fmt.Errorf("debug: read mem %#x: %w", addr, err)
}
return out, nil
}
// WriteMemory writes bytes to the debuggee's memory at addr.
// WriteMemory writes bytes to the debuggee's memory at addr. The write
// covers exactly the given bytes: /proc/pid/mem accepts writes of any
// length at any offset, so the word loop's read-modify-write of the final
// partial word (which read past the requested range and failed on an
// unmapped following page) is unnecessary.
func (s *Session) WriteMemory(addr uint64, data []byte) error {
for i := 0; i < len(data); i += 8 {
end := min(i+8, len(data))
var word uint64
for j := 0; j < end-i; j++ {
word |= uint64(data[i+j]) << (8 * j)
if len(data) == 0 {
return nil
}
if end-i < 8 {
existing, err := s.Peek(addr + uint64(i))
mem, err := os.OpenFile(fmt.Sprintf("/proc/%d/mem", s.pid), os.O_WRONLY, 0)
if err != nil {
return err
}
mask := ^((uint64(1) << (8 * (end - i))) - 1)
word = (existing & mask) | word
}
if err := s.Poke(addr+uint64(i), word); err != nil {
return err
return fmt.Errorf("debug: open /proc/%d/mem: %w", s.pid, err)
}
defer mem.Close()
if _, err := mem.WriteAt(data, int64(addr)); err != nil {
return fmt.Errorf("debug: write mem %#x: %w", addr, err)
}
return nil
}
+98
View File
@@ -0,0 +1,98 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && amd64
package debug
// Regs holds the full general-purpose register set of a traced process
// (the FreeBSD amd64 struct reg layout, sys/x86/include/reg.h). FreeBSD
// reports segment selectors (FS/GS/ES/DS), not the bases the Linux ptrace
// surface carries, and has no ORIG_RAX slot.
type Regs struct {
R15 uint64
R14 uint64
R13 uint64
R12 uint64
RBP uint64
RBX uint64
R11 uint64
R10 uint64
R9 uint64
R8 uint64
RAX uint64
RCX uint64
RDX uint64
RSI uint64
RDI uint64
RIP uint64
CS uint64
RFLAGS uint64
RSP uint64
SS uint64
FS uint64
GS uint64
DS uint64
ES uint64
}
// GetPC returns the program counter.
func (r *Regs) GetPC() uint64 { return r.RIP }
// SetPC sets the program counter.
func (r *Regs) SetPC(pc uint64) { r.RIP = pc }
// GetSP returns the stack pointer.
func (r *Regs) GetSP() uint64 { return r.RSP }
// RegValue returns the value of the named register, or false if unknown.
func (r *Regs) RegValue(name string) (uint64, bool) {
switch name {
case "rax", "eax", "ax", "al":
return r.RAX, true
case "rbx", "ebx", "bx", "bl":
return r.RBX, true
case "rcx", "ecx", "cx", "cl":
return r.RCX, true
case "rdx", "edx", "dx", "dl":
return r.RDX, true
case "rsi", "esi", "si":
return r.RSI, true
case "rdi", "edi", "di":
return r.RDI, true
case "rbp", "ebp", "bp":
return r.RBP, true
case "rsp", "esp", "sp":
return r.RSP, true
case "r8":
return r.R8, true
case "r9":
return r.R9, true
case "r10":
return r.R10, true
case "r11":
return r.R11, true
case "r12":
return r.R12, true
case "r13":
return r.R13, true
case "r14":
return r.R14, true
case "r15":
return r.R15, true
case "rip", "eip":
return r.RIP, true
default:
return 0, false
}
}
// breakpointInsn is the software breakpoint instruction.
var breakpointInsn = []byte{0xCC} // INT3
// breakpointPCAdjust is how far PC is past the breakpoint instruction after
// a trap. INT3 leaves the hardware PC on the following instruction (Intel
// SDM vol 3, "Debug Exceptions") and the FreeBSD T_BPTFLT path delivers
// that frame unmodified (sys/amd64/amd64/trap.c), so the trap address is
// PC-1, the same correction the Linux side applies.
const breakpointPCAdjust = 1
+139
View File
@@ -0,0 +1,139 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && arm64
package debug
// Regs holds the full general-purpose register set of a traced process
// (the FreeBSD arm64 struct reg layout, sys/arm64/include/reg.h: x[30], lr,
// sp, elr, spsr).
type Regs struct {
X0 uint64
X1 uint64
X2 uint64
X3 uint64
X4 uint64
X5 uint64
X6 uint64
X7 uint64
X8 uint64
X9 uint64
X10 uint64
X11 uint64
X12 uint64
X13 uint64
X14 uint64
X15 uint64
X16 uint64
X17 uint64
X18 uint64
X19 uint64
X20 uint64
X21 uint64
X22 uint64
X23 uint64
X24 uint64
X25 uint64
X26 uint64
X27 uint64
X28 uint64
X29 uint64 // FP (frame pointer)
X30 uint64 // LR (link register)
SP uint64
PC uint64
PSTATE uint64
}
// GetPC returns the program counter.
func (r *Regs) GetPC() uint64 { return r.PC }
// SetPC sets the program counter.
func (r *Regs) SetPC(pc uint64) { r.PC = pc }
// GetSP returns the stack pointer.
func (r *Regs) GetSP() uint64 { return r.SP }
// RegValue returns the value of the named register, or false if unknown.
func (r *Regs) RegValue(name string) (uint64, bool) {
switch name {
case "x0":
return r.X0, true
case "x1":
return r.X1, true
case "x2":
return r.X2, true
case "x3":
return r.X3, true
case "x4":
return r.X4, true
case "x5":
return r.X5, true
case "x6":
return r.X6, true
case "x7":
return r.X7, true
case "x8":
return r.X8, true
case "x9":
return r.X9, true
case "x10":
return r.X10, true
case "x11":
return r.X11, true
case "x12":
return r.X12, true
case "x13":
return r.X13, true
case "x14":
return r.X14, true
case "x15":
return r.X15, true
case "x16":
return r.X16, true
case "x17":
return r.X17, true
case "x18":
return r.X18, true
case "x19":
return r.X19, true
case "x20":
return r.X20, true
case "x21":
return r.X21, true
case "x22":
return r.X22, true
case "x23":
return r.X23, true
case "x24":
return r.X24, true
case "x25":
return r.X25, true
case "x26":
return r.X26, true
case "x27":
return r.X27, true
case "x28":
return r.X28, true
case "x29", "fp":
return r.X29, true
case "x30", "lr":
return r.X30, true
case "sp":
return r.SP, true
case "pc":
return r.PC, true
default:
return 0, false
}
}
// breakpointInsn is the software breakpoint instruction (BRK #0).
var breakpointInsn = []byte{0x00, 0x00, 0x20, 0xD4} // BRK #0
// breakpointPCAdjust is how far PC is past the breakpoint instruction after
// a trap: 0. The BRK synchronous exception leaves ELR_EL0 on the BRK
// itself (ARM DDI 0487), and the FreeBSD EXCP_BRKPT_EL0 handler delivers
// the frame's elr unmodified (sys/arm64/arm64/trap.c), so the trap address
// is the PC as reported.
const breakpointPCAdjust = 0
+134
View File
@@ -0,0 +1,134 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && riscv64
package debug
// Regs holds the full general-purpose register set of a traced process
// (the FreeBSD riscv64 struct reg layout: ra, sp, gp, tp, t0-t6, s0-s11,
// a0-a7, sepc, sstatus).
type Regs struct {
PC uint64 // sepc
Ra uint64 // x1 (return address)
Sp uint64 // x2
Gp uint64 // x3
Tp uint64 // x4
T0 uint64 // x5
T1 uint64 // x6
T2 uint64 // x7
S0 uint64 // x8 (frame pointer)
S1 uint64 // x9
A0 uint64 // x10
A1 uint64 // x11
A2 uint64 // x12
A3 uint64 // x13
A4 uint64 // x14
A5 uint64 // x15
A6 uint64 // x16
A7 uint64 // x17
S2 uint64 // x18
S3 uint64 // x19
S4 uint64 // x20
S5 uint64 // x21
S6 uint64 // x22
S7 uint64 // x23
S8 uint64 // x24
S9 uint64 // x25
S10 uint64 // x26
S11 uint64 // x27
T3 uint64 // x28
T4 uint64 // x29
T5 uint64 // x30
T6 uint64 // x31
}
// GetPC returns the program counter.
func (r *Regs) GetPC() uint64 { return r.PC }
// SetPC sets the program counter.
func (r *Regs) SetPC(pc uint64) { r.PC = pc }
// GetSP returns the stack pointer.
func (r *Regs) GetSP() uint64 { return r.Sp }
// RegValue returns the value of the named register, or false if unknown.
func (r *Regs) RegValue(name string) (uint64, bool) {
switch name {
case "pc":
return r.PC, true
case "ra", "x1":
return r.Ra, true
case "sp", "x2":
return r.Sp, true
case "gp", "x3":
return r.Gp, true
case "tp", "x4":
return r.Tp, true
case "t0", "x5":
return r.T0, true
case "t1", "x6":
return r.T1, true
case "t2", "x7":
return r.T2, true
case "s0", "fp", "x8":
return r.S0, true
case "s1", "x9":
return r.S1, true
case "a0", "x10":
return r.A0, true
case "a1", "x11":
return r.A1, true
case "a2", "x12":
return r.A2, true
case "a3", "x13":
return r.A3, true
case "a4", "x14":
return r.A4, true
case "a5", "x15":
return r.A5, true
case "a6", "x16":
return r.A6, true
case "a7", "x17":
return r.A7, true
case "s2", "x18":
return r.S2, true
case "s3", "x19":
return r.S3, true
case "s4", "x20":
return r.S4, true
case "s5", "x21":
return r.S5, true
case "s6", "x22":
return r.S6, true
case "s7", "x23":
return r.S7, true
case "s8", "x24":
return r.S8, true
case "s9", "x25":
return r.S9, true
case "s10", "x26":
return r.S10, true
case "s11", "x27":
return r.S11, true
case "t3", "x28":
return r.T3, true
case "t4", "x29":
return r.T4, true
case "t5", "x30":
return r.T5, true
case "t6", "x31":
return r.T6, true
default:
return 0, false
}
}
// breakpointInsn is the software breakpoint instruction (EBREAK).
var breakpointInsn = []byte{0x73, 0x00, 0x10, 0x00} // ebreak
// breakpointPCAdjust is how far PC is past the breakpoint instruction after
// a trap: 0. The EBREAK synchronous exception leaves sepc on the ebreak
// itself (RISC-V privileged architecture), so the trap address is the PC as
// reported.
const breakpointPCAdjust = 0
+63 -15
View File
@@ -1,7 +1,7 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build linux
//go:build linux || (freebsd && (amd64 || arm64 || riscv64))
package debug
@@ -71,7 +71,12 @@ func REPL(s *Session, bm *Breakpoints, codeBase uint64, funcOffset, funcSize, ar
case "step", "s":
n := 1
if len(parts) > 1 {
n, _ = strconv.Atoi(parts[1])
v, err := strconv.Atoi(parts[1])
if err != nil || v < 0 {
fmt.Printf("invalid count: %s\n", parts[1])
continue
}
n = v
}
for range n {
if s.Exited() {
@@ -82,8 +87,13 @@ func REPL(s *Session, bm *Breakpoints, codeBase uint64, funcOffset, funcSize, ar
fmt.Println(err)
break
}
if sig := s.LastSignal(); sig != 0 {
regs, _ := s.GetRegs()
fmt.Printf("stopped on signal %v at %#x\n", sig, regs.GetPC())
break
}
if !s.Exited() {
}
if !s.Exited() && s.LastSignal() == 0 {
regs, _ := s.GetRegs()
pc := regs.GetPC()
text, _, _ := s.Disassemble(pc)
@@ -106,16 +116,16 @@ func REPL(s *Session, bm *Breakpoints, codeBase uint64, funcOffset, funcSize, ar
}
if err := s.Continue(); err != nil {
fmt.Println(err)
bm.Clear(afterAddr)
clearNextBp(bm, afterAddr)
continue
}
if s.Exited() {
bm.Clear(afterAddr)
clearNextBp(bm, afterAddr)
fmt.Println("debuggee exited")
continue
}
if sig := s.LastSignal(); sig != 0 {
bm.Clear(afterAddr)
clearNextBp(bm, afterAddr)
regs, _ := s.GetRegs()
fmt.Printf("stopped on signal %v at %#x\n", sig, regs.GetPC())
continue
@@ -125,12 +135,17 @@ func REPL(s *Session, bm *Breakpoints, codeBase uint64, funcOffset, funcSize, ar
// snapshot, and a stale SetRegs would clobber live state.
regs, _ = s.GetRegs()
bm.HandleTrap(&regs)
bm.Clear(afterAddr)
clearNextBp(bm, afterAddr)
} else {
if err := s.Step(); err != nil {
fmt.Println(err)
continue
}
if sig := s.LastSignal(); sig != 0 {
regs, _ := s.GetRegs()
fmt.Printf("stopped on signal %v at %#x\n", sig, regs.GetPC())
continue
}
}
if !s.Exited() {
regs, _ := s.GetRegs()
@@ -156,16 +171,16 @@ func REPL(s *Session, bm *Breakpoints, codeBase uint64, funcOffset, funcSize, ar
}
if err := s.Continue(); err != nil {
fmt.Println(err)
bm.Clear(retAddr)
clearNextBp(bm, retAddr)
continue
}
if s.Exited() {
bm.Clear(retAddr)
clearNextBp(bm, retAddr)
fmt.Println("debuggee exited")
continue
}
if sig := s.LastSignal(); sig != 0 {
bm.Clear(retAddr)
clearNextBp(bm, retAddr)
regs, _ := s.GetRegs()
fmt.Printf("stopped on signal %v at %#x\n", sig, regs.GetPC())
continue
@@ -174,7 +189,7 @@ func REPL(s *Session, bm *Breakpoints, codeBase uint64, funcOffset, funcSize, ar
// does: HandleTrap must see the PC the trap left behind.
regs, _ = s.GetRegs()
bm.HandleTrap(&regs)
bm.Clear(retAddr)
clearNextBp(bm, retAddr)
if s.Exited() {
fmt.Println("debuggee exited")
} else {
@@ -214,6 +229,7 @@ func REPL(s *Session, bm *Breakpoints, codeBase uint64, funcOffset, funcSize, ar
break
}
regs, _ := s.GetRegs()
trapPC := regs.GetPC()
if bp := bm.HandleTrap(&regs); bp != nil {
// Execute the instruction under the restored breakpoint
// so the next continue cannot re-trap on the same
@@ -229,6 +245,13 @@ func REPL(s *Session, bm *Breakpoints, codeBase uint64, funcOffset, funcSize, ar
fmt.Printf("breakpoint hit: %s (func+%#x)\n", name, bp.Addr-codeBase-uint64(funcOffset))
break
}
// The trap matched no breakpoint of ours. When the PC still
// stands on the trapping instruction, resuming would re-execute
// it and trap forever, so surface the stop instead of spinning.
if TrapStray(s, reason, trapPC) {
fmt.Printf("SIGTRAP at %#x matches no breakpoint; the PC did not advance\n", trapPC-uint64(breakpointPCAdjust))
break
}
}
case "break", "b":
@@ -326,6 +349,10 @@ func REPL(s *Session, bm *Breakpoints, codeBase uint64, funcOffset, funcSize, ar
length := 64
if len(parts) > 1 {
addr, _ = resolveAddr(parts[1], codeBase, uint64(funcOffset), labels)
if addr == 0 {
fmt.Printf("unknown address: %s\n", parts[1])
continue
}
}
if len(parts) > 2 {
// A malformed or non-positive length would panic
@@ -399,9 +426,13 @@ func REPL(s *Session, bm *Breakpoints, codeBase uint64, funcOffset, funcSize, ar
case "disas", "u":
n := 5
if len(parts) > 1 {
n, _ = strconv.Atoi(parts[1])
if n <= 0 {
n = 5
v, err := strconv.Atoi(parts[1])
if err != nil {
fmt.Printf("invalid count: %s\n", parts[1])
continue
}
if v > 0 {
n = v
}
}
regs, _ := s.GetRegs()
@@ -496,10 +527,18 @@ func REPL(s *Session, bm *Breakpoints, codeBase uint64, funcOffset, funcSize, ar
typ = WatchRead
case "w":
typ = WatchWrite
default:
fmt.Printf("unknown watchpoint type: %s (want r or w)\n", parts[2])
continue
}
}
if len(parts) > 3 {
size, _ = strconv.Atoi(parts[3])
v, err := strconv.Atoi(parts[3])
if err != nil || v <= 0 {
fmt.Printf("invalid size: %s\n", parts[3])
continue
}
size = v
}
slot := s.FindFreeWatchpointSlot()
if slot < 0 {
@@ -543,6 +582,15 @@ func REPL(s *Session, bm *Breakpoints, codeBase uint64, funcOffset, funcSize, ar
s.Kill()
}
// clearNextBp removes one of the temporary breakpoints the next and finish
// commands plant, reporting a failure instead of silently leaving the trap
// instruction behind in the debuggee.
func clearNextBp(bm *Breakpoints, addr uint64) {
if err := bm.Clear(addr); err != nil {
fmt.Printf("cannot remove temporary breakpoint at %#x: %v\n", addr, err)
}
}
func hexDump(addr uint64, data []byte) {
for i := 0; i < len(data); i += 16 {
end := min(i+16, len(data))
+73
View File
@@ -0,0 +1,73 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && (amd64 || arm64 || riscv64)
package debug
import (
"encoding/binary"
"syscall"
"unsafe"
"golang.org/x/sys/unix"
)
// FreeBSD TRAP_* si_code values (sys/signal.h). A breakpoint (INT3, BRK,
// EBREAK) arrives as TRAP_BRKPT on every supported architecture; TRAP_TRACE
// is shared by the completed single-step and the hardware watchpoint hit,
// so the watchpoint layer disambiguates from the debug registers.
const (
trapBRKPT = 1 // TRAP_BRKPT
trapTRACE = 2 // TRAP_TRACE
)
// StopReason describes why the debuggee stopped.
type StopReason int
const (
StopNone StopReason = iota
StopBreakpoint // software breakpoint hit
StopWatchpoint // hardware watchpoint triggered
StopSingleStep // single-step completed
StopSignal // stopped by a signal
StopExited // process exited
)
// StopInfo returns the reason the debuggee stopped and the faulting address
// (for watchpoints, the watched address that was accessed). FreeBSD has no
// PTRACE_GETSIGINFO; the stop's signal information comes from PT_LWPINFO,
// whose pl_siginfo carries the siginfo the kernel delivered. A ptrace stop
// with no signal behind it (a completed single-step, the initial attach)
// fills no siginfo at all.
func (s *Session) StopInfo() (StopReason, uint64) {
if s.exited {
return StopExited, 0
}
var info unix.PtraceLwpInfoStruct
if err := unix.PtraceLwpInfo(s.pid, &info); err != nil {
return StopNone, 0
}
// The siginfo layout is the FreeBSD siginfo_t: three leading ints
// (signo, errno, code), then the union, 8-byte aligned, whose _fault
// member puts the address at byte offset 16. The read is byte-wise
// because the blob's alignment is not guaranteed.
si := (*[64]byte)(unsafe.Pointer(&info.Siginfo))
signo := int32(binary.LittleEndian.Uint32(si[0:4]))
code := int32(binary.LittleEndian.Uint32(si[8:12]))
switch {
case signo == 0:
// A pure ptrace stop: single-step completion, attach, or the
// events the kernel resolves internally.
return StopSingleStep, 0
case signo != int32(syscall.SIGTRAP):
return StopSignal, uint64(code)
case code == trapBRKPT:
return StopBreakpoint, 0
case code == trapTRACE:
addr := binary.LittleEndian.Uint64(si[16:24])
return archStopTrace(s, addr)
default:
return StopSingleStep, 0
}
}
+206
View File
@@ -0,0 +1,206 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && (amd64 || arm64 || riscv64)
package debug
import (
"encoding/hex"
"fmt"
"os"
"runtime"
"strconv"
"strings"
"syscall"
"unsafe"
"golang.org/x/sys/unix"
"sourcedock.dev/petrbalvin/gasm-sdk/asm"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/verify"
)
// mapRWX maps code into a read-write-execute region.
func mapRWX(code []byte) ([]byte, error) {
const pageSize = 4096
size := (len(code) + pageSize - 1) &^ (pageSize - 1)
mem, err := syscall.Mmap(-1, 0, size,
syscall.PROT_READ|syscall.PROT_WRITE|syscall.PROT_EXEC,
syscall.MAP_PRIVATE|syscall.MAP_ANON)
if err != nil {
return nil, err
}
copy(mem, code)
return mem, nil
}
// setupBuffers allocates buffers in the debuggee's memory.
func setupBuffers(spec string, args []byte, tmpDir string) ([]byte, error) {
type bufSpec struct {
name string
size int
pattern string
}
var specs []bufSpec
for part := range strings.SplitSeq(spec, ",") {
fields := strings.SplitN(part, ":", 3)
if len(fields) != 3 {
continue
}
size, err := strconv.Atoi(fields[1])
if err != nil || size <= 0 {
continue
}
specs = append(specs, bufSpec{name: fields[0], size: size, pattern: fields[2]})
}
if len(specs) == 0 {
return args, nil
}
var bufAddrs []uint64
for _, s := range specs {
buf, err := syscall.Mmap(-1, 0, s.size,
syscall.PROT_READ|syscall.PROT_WRITE,
syscall.MAP_PRIVATE|syscall.MAP_ANON)
if err != nil {
return nil, fmt.Errorf("mmap buffer %s: %w", s.name, err)
}
fillBuffer(buf, s.pattern)
bufAddrs = append(bufAddrs, uint64(uintptr(unsafe.Pointer(&buf[0]))))
}
addrFile, err := os.Create(tmpDir + "/bufaddrs")
if err != nil {
return nil, err
}
for _, addr := range bufAddrs {
fmt.Fprintf(addrFile, "%d\n", addr)
}
addrFile.Close()
return args, nil
}
// fillBuffer fills a buffer with the specified pattern.
func fillBuffer(buf []byte, pattern string) {
switch pattern {
case "zero":
case "ones":
for i := range buf {
buf[i] = 0xFF
}
case "seq":
for i := range buf {
buf[i] = byte(i)
}
default:
if data, err := hex.DecodeString(pattern); err == nil && len(data) > 0 {
for i := range buf {
buf[i] = data[i%len(data)]
}
}
}
}
// RunTarget is the debuggee entry point (gasm debug --target). A failure
// is marked in the handshake directory before the process exits, so the
// debugger's readiness poll fails fast on a dead debuggee instead of
// waiting out its whole budget.
func RunTarget(asmPath, funcName, argsFile, tmpDir string) error {
err := runTarget(asmPath, funcName, argsFile, tmpDir)
if err != nil {
markDead(tmpDir, err.Error())
}
return err
}
func runTarget(asmPath, funcName, argsFile, tmpDir string) error {
src, err := os.ReadFile(asmPath)
if err != nil {
return fmt.Errorf("debug target: %w", err)
}
file, errs := parser.Parse(asmPath, string(src))
if len(errs) > 0 {
return fmt.Errorf("debug target: parse: %v", errs[0])
}
img, err := asm.AssembleFile(file)
if err != nil {
return fmt.Errorf("debug target: assemble: %w", err)
}
var fl *asm.FuncLayout
for i := range img.Funcs {
if img.Funcs[i].Name == funcName {
fl = &img.Funcs[i]
break
}
}
if fl == nil {
return fmt.Errorf("debug target: function %q not found", funcName)
}
code := img.Bytes()
exec, err := mapRWX(code)
if err != nil {
return fmt.Errorf("debug target: mmap: %w", err)
}
codeBase := uintptr(unsafe.Pointer(&exec[0]))
if err := os.WriteFile(tmpDir+"/codebase", []byte(fmt.Sprintf("%d", codeBase)), 0o644); err != nil {
return fmt.Errorf("debug target: write codebase: %w", err)
}
meta := fmt.Sprintf("%d %d %d", fl.Offset, fl.Size, fl.Args)
os.WriteFile(tmpDir+"/funcmeta", []byte(meta), 0o644)
labelsFile, _ := os.Create(tmpDir + "/labels")
if labelsFile != nil {
for label, off := range fl.Labels {
fmt.Fprintf(labelsFile, "%s %d\n", label, off)
}
labelsFile.Close()
}
args, err := os.ReadFile(argsFile)
if err != nil {
return fmt.Errorf("debug target: read args: %w", err)
}
if len(args) < fl.Args {
padded := make([]byte, fl.Args)
copy(padded, args)
args = padded
}
bufSpecFile := tmpDir + "/bufspec"
if bufSpec, err := os.ReadFile(bufSpecFile); err == nil && len(bufSpec) > 0 {
args, err = setupBuffers(string(bufSpec), args, tmpDir)
if err != nil {
return fmt.Errorf("debug target: setup buffers: %w", err)
}
}
runtime.LockOSThread()
if _, _, errno := unix.RawSyscall(unix.SYS_PTRACE, uintptr(unix.PT_TRACE_ME), 0, 0); errno != 0 {
return fmt.Errorf("debug target: PT_TRACE_ME: %v", errno)
}
os.WriteFile(tmpDir+"/ready", []byte("ok"), 0o644)
syscall.Kill(syscall.Getpid(), syscall.SIGSTOP)
os.WriteFile(tmpDir+"/entry", []byte("ok"), 0o644)
syscall.Kill(syscall.Getpid(), syscall.SIGSTOP)
fnAddr := codeBase + uintptr(fl.Offset)
stackArgs := make([]byte, fl.Args)
copy(stackArgs, args)
if _, callErr := verify.Call(fnAddr, stackArgs); callErr != nil {
os.Exit(1)
}
// Success returns to the caller, which exits with status 0; the JIT
// code has already run to its own trampoline by the time Call returns.
return nil
}
+15 -4
View File
@@ -12,13 +12,24 @@ import (
"syscall"
"unsafe"
"sourcedock.dev/petrbalvin/gasm-devkit/asm"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-devkit/verify"
"sourcedock.dev/petrbalvin/gasm-sdk/asm"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/verify"
)
// RunTarget is the debuggee entry point (gasm debug --target).
// RunTarget is the debuggee entry point (gasm debug --target). A failure
// is marked in the handshake directory before the process exits, so the
// debugger's readiness poll fails fast on a dead debuggee instead of
// waiting out its whole budget.
func RunTarget(asmPath, funcName, argsFile, tmpDir string) error {
err := runTarget(asmPath, funcName, argsFile, tmpDir)
if err != nil {
markDead(tmpDir, err.Error())
}
return err
}
func runTarget(asmPath, funcName, argsFile, tmpDir string) error {
src, err := os.ReadFile(asmPath)
if err != nil {
return fmt.Errorf("debug target: %w", err)
+15 -4
View File
@@ -12,13 +12,24 @@ import (
"syscall"
"unsafe"
"sourcedock.dev/petrbalvin/gasm-devkit/asm"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-devkit/verify"
"sourcedock.dev/petrbalvin/gasm-sdk/asm"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/verify"
)
// RunTarget is the debuggee entry point (gasm debug --target).
// RunTarget is the debuggee entry point (gasm debug --target). A failure
// is marked in the handshake directory before the process exits, so the
// debugger's readiness poll fails fast on a dead debuggee instead of
// waiting out its whole budget.
func RunTarget(asmPath, funcName, argsFile, tmpDir string) error {
err := runTarget(asmPath, funcName, argsFile, tmpDir)
if err != nil {
markDead(tmpDir, err.Error())
}
return err
}
func runTarget(asmPath, funcName, argsFile, tmpDir string) error {
src, err := os.ReadFile(asmPath)
if err != nil {
return fmt.Errorf("debug target: %w", err)
+15 -4
View File
@@ -12,13 +12,24 @@ import (
"syscall"
"unsafe"
"sourcedock.dev/petrbalvin/gasm-devkit/asm"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-devkit/verify"
"sourcedock.dev/petrbalvin/gasm-sdk/asm"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/verify"
)
// RunTarget is the debuggee entry point (gasm debug --target).
// RunTarget is the debuggee entry point (gasm debug --target). A failure
// is marked in the handshake directory before the process exits, so the
// debugger's readiness poll fails fast on a dead debuggee instead of
// waiting out its whole budget.
func RunTarget(asmPath, funcName, argsFile, tmpDir string) error {
err := runTarget(asmPath, funcName, argsFile, tmpDir)
if err != nil {
markDead(tmpDir, err.Error())
}
return err
}
func runTarget(asmPath, funcName, argsFile, tmpDir string) error {
src, err := os.ReadFile(asmPath)
if err != nil {
return fmt.Errorf("debug target: %w", err)
+15 -4
View File
@@ -12,13 +12,24 @@ import (
"syscall"
"unsafe"
"sourcedock.dev/petrbalvin/gasm-devkit/asm"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-devkit/verify"
"sourcedock.dev/petrbalvin/gasm-sdk/asm"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/verify"
)
// RunTarget is the debuggee entry point (gasm debug --target).
// RunTarget is the debuggee entry point (gasm debug --target). A failure
// is marked in the handshake directory before the process exits, so the
// debugger's readiness poll fails fast on a dead debuggee instead of
// waiting out its whole budget.
func RunTarget(asmPath, funcName, argsFile, tmpDir string) error {
err := runTarget(asmPath, funcName, argsFile, tmpDir)
if err != nil {
markDead(tmpDir, err.Error())
}
return err
}
func runTarget(asmPath, funcName, argsFile, tmpDir string) error {
src, err := os.ReadFile(asmPath)
if err != nil {
return fmt.Errorf("debug target: %w", err)
+17 -1
View File
@@ -1,10 +1,26 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build linux
//go:build linux || (freebsd && (amd64 || arm64 || riscv64))
package debug
import (
"os"
"path/filepath"
)
// markDead records the target's failure reason in the handshake directory,
// the death notice the debugger's readiness poll reads. The poll must not
// wait on the child while it polls: a wait there could consume the SIGSTOP
// park the debugger's own waitStopped must receive, hanging the launch, so
// the file is the only fast death notice that is safe to read. A debuggee
// killed without a notice (a crash, SIGKILL) surfaces through the ordinary
// wait after the poll instead.
func markDead(tmpDir, reason string) {
os.WriteFile(filepath.Join(tmpDir, "dead"), []byte(reason), 0o644)
}
// tracer abstracts the minimal ptrace operations needed by the breakpoint
// manager and the stop-information helpers. The live implementation is
// *Session (ptrace_linux_amd64.go); tests supply a mock.
+200
View File
@@ -0,0 +1,200 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && amd64
package debug
import (
"fmt"
"unsafe"
"golang.org/x/sys/unix"
)
// Hardware watchpoint support via x86-64 debug registers (DR0-DR3, DR7),
// read and written as one blob through PT_GETDBREGS/PT_SETDBREGS. The
// FreeBSD struct dbreg is the raw DR file: dr[16], where DR0-DR3 are the
// address registers, DR6 the status and DR7 the control (sys/x86/include/
// reg.h; the DBREG_DRX accessor indexes the same array).
// dbreg mirrors FreeBSD's struct dbreg for PT_GETDBREGS/PT_SETDBREGS.
type dbreg struct {
Dr [16]uint64
}
// dbreg indices of the registers the watchpoint layer drives.
const (
drStatus = 6 // DR6: the trap status register
drControl = 7 // DR7: the debug control register
)
// WatchpointType selects what triggers the watchpoint.
type WatchpointType int
const (
WatchWrite WatchpointType = 1 // trigger on write
WatchRead WatchpointType = 3 // trigger on read or write
)
// maxWatchpoints reports the number of hardware watchpoint slots the
// architecture provides: four address registers, DR0-DR3.
func maxWatchpoints() int { return 4 }
// getDbRegs reads the debug register file of the stopped debuggee.
func (s *Session) getDbRegs() (*dbreg, error) {
var dr dbreg
if _, _, errno := unix.Syscall6(
unix.SYS_PTRACE,
uintptr(unix.PT_GETDBREGS),
uintptr(s.pid),
0,
uintptr(unsafe.Pointer(&dr)),
0, 0,
); errno != 0 {
return nil, errno
}
return &dr, nil
}
// setDbRegs writes the debug register file of the stopped debuggee.
func (s *Session) setDbRegs(dr *dbreg) error {
if _, _, errno := unix.Syscall6(
unix.SYS_PTRACE,
uintptr(unix.PT_SETDBREGS),
uintptr(s.pid),
0,
uintptr(unsafe.Pointer(dr)),
0, 0,
); errno != 0 {
return errno
}
return nil
}
// archStopTrace classifies a TRAP_TRACE stop. On amd64 the kernel
// delivers both the completed single-step and the debug-register hit
// through T_TRCTRAP with TRAP_TRACE (sys/amd64/amd64/trap.c), and DR6's
// B0-B3 bits name the watchpoint that fired. The bits are sticky ("the
// processor never clears DR6", Intel SDM vol 3, "Debug Registers"), so
// they are acknowledged here: cleared once read, or the next hit on a
// different slot would still see this slot's bit set and report this
// slot's address again.
func archStopTrace(s *Session, siAddr uint64) (StopReason, uint64) {
dr, err := s.getDbRegs()
if err != nil {
return StopSingleStep, 0
}
status := dr.Dr[drStatus] & 0xF
if status == 0 {
return StopSingleStep, 0
}
// Best effort: the write-back only fails for a dead debuggee, for
// which no further watchpoint can fire anyway.
dr.Dr[drStatus] &^= 0xF
_ = s.setDbRegs(dr)
for slot := range 4 {
if status&(1<<slot) != 0 && dr.Dr[slot] != 0 {
return StopWatchpoint, dr.Dr[slot]
}
}
return StopSingleStep, 0
}
// FindFreeWatchpointSlot returns the index of the first free watchpoint slot
// (0-3), or -1 if all four hardware watchpoints are in use.
func (s *Session) FindFreeWatchpointSlot() int {
for i := range 4 {
if !s.wpSlots[i] {
return i
}
}
return -1
}
// IsWatchpointSlotUsed reports whether slot (0-3) currently holds a watchpoint.
func (s *Session) IsWatchpointSlotUsed(slot int) bool {
if slot < 0 || slot > 3 {
return false
}
return s.wpSlots[slot]
}
// SetWatchpoint installs a hardware watchpoint on the given address.
// DR7's encoding is architectural: a 2-bit local/global enable pair per
// slot at bit 2*slot, the R/W field at 16+4*slot and the length field at
// 18+4*slot (Intel SDM vol 3, "Debug Registers").
func (s *Session) SetWatchpoint(slot int, addr uint64, typ WatchpointType, size int) error {
if slot < 0 || slot > 3 {
return fmt.Errorf("debug: watchpoint slot must be 0-3")
}
if s.wpSlots[slot] {
return fmt.Errorf("debug: watchpoint slot %d already in use", slot)
}
var lenBits uint64
switch size {
case 1:
lenBits = 0
case 2:
lenBits = 1
case 4:
lenBits = 3
case 8:
lenBits = 2
default:
return fmt.Errorf("debug: watchpoint size must be 1, 2, 4, or 8")
}
dr, err := s.getDbRegs()
if err != nil {
return fmt.Errorf("debug: read debug registers: %w", err)
}
dr.Dr[slot] = addr
dr7 := dr.Dr[drControl]
enableBit := uint64(1) << (2 * slot)
rwBits := uint64(typ) << (16 + 4*slot)
lenField := lenBits << (18 + 4*slot)
mask := ^((uint64(1) << (2 * slot)) | (uint64(3) << (16 + 4*slot)) | (uint64(3) << (18 + 4*slot)))
dr.Dr[drControl] = (dr7 & mask) | enableBit | rwBits | lenField
if err := s.setDbRegs(dr); err != nil {
return fmt.Errorf("debug: set debug registers: %w", err)
}
s.wpSlots[slot] = true
return nil
}
// ClearWatchpoint removes a hardware watchpoint.
func (s *Session) ClearWatchpoint(slot int) error {
if slot < 0 || slot > 3 {
return fmt.Errorf("debug: watchpoint slot must be 0-3")
}
if !s.wpSlots[slot] {
return fmt.Errorf("debug: watchpoint slot %d is not in use", slot)
}
dr, err := s.getDbRegs()
if err != nil {
return err
}
dr.Dr[slot] = 0
dr.Dr[drControl] &^= uint64(1) << (2 * slot)
if err := s.setDbRegs(dr); err != nil {
return err
}
s.wpSlots[slot] = false
return nil
}
// ClearAllWatchpoints removes all hardware watchpoints.
func (s *Session) ClearAllWatchpoints() error {
for slot := range maxWatchpoints() {
if s.wpSlots[slot] {
if err := s.ClearWatchpoint(slot); err != nil {
return err
}
}
}
return nil
}
+216
View File
@@ -0,0 +1,216 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && arm64
package debug
import (
"fmt"
"math/bits"
"unsafe"
"golang.org/x/sys/unix"
)
// Hardware watchpoint support via arm64 debug registers, read and written
// as one blob through PT_GETDBREGS/PT_SETDBREGS. The FreeBSD struct dbreg
// (sys/arm64/include/reg.h) opens with the debug-facility header and then
// carries 16 breakpoint and 16 watchpoint pairs of {address, control}.
// dbreg mirrors FreeBSD's struct dbreg for PT_GETDBREGS/PT_SETDBREGS.
type dbreg struct {
DbDebugVer uint8
DbNbkpts uint8
DbNwtpts uint8
_ [5]byte
DbBreakregs [16]struct {
Addr uint64
Ctrl uint32
_ uint32
}
DbWatchregs [16]struct {
Addr uint64
Ctrl uint32
_ uint32
}
}
// WatchpointType selects what triggers the watchpoint.
type WatchpointType int
const (
WatchWrite WatchpointType = 1 // trigger on write
WatchRead WatchpointType = 3 // trigger on read or write
)
// maxWatchpoints reports the number of hardware watchpoint slots the
// architecture provides: DBGWVR0-DBGWCR15.
func maxWatchpoints() int { return 16 }
// getDbRegs reads the debug register file of the stopped debuggee.
func (s *Session) getDbRegs() (*dbreg, error) {
var dr dbreg
if _, _, errno := unix.Syscall6(
unix.SYS_PTRACE,
uintptr(unix.PT_GETDBREGS),
uintptr(s.pid),
0,
uintptr(unsafe.Pointer(&dr)),
0, 0,
); errno != 0 {
return nil, errno
}
return &dr, nil
}
// setDbRegs writes the debug register file of the stopped debuggee.
func (s *Session) setDbRegs(dr *dbreg) error {
if _, _, errno := unix.Syscall6(
unix.SYS_PTRACE,
uintptr(unix.PT_SETDBREGS),
uintptr(s.pid),
0,
uintptr(unsafe.Pointer(dr)),
0, 0,
); errno != 0 {
return errno
}
return nil
}
// archStopTrace classifies a TRAP_TRACE stop. On arm64 the kernel
// delivers both the software single step and the watchpoint hit through
// EXCP_SOFTSTP_EL0/EXCP_WATCHPT_EL0 with TRAP_TRACE (sys/arm64/arm64/
// trap.c); the watchpoint address rides the FAR register, so a stop whose
// reported address falls inside an armed watchpoint's byte range is a
// watchpoint and everything else is a single step. The byte range comes
// from DBGWCR's byte-address-select bits (ARM DDI 0487, DBGWCR<n>_EL1):
// bit i watches the address plus i, so the range spans the lowest set bit
// to the highest set bit inclusive.
func archStopTrace(s *Session, siAddr uint64) (StopReason, uint64) {
dr, err := s.getDbRegs()
if err != nil {
return StopSingleStep, 0
}
for slot := range 16 {
ctrl := uint64(dr.DbWatchregs[slot].Ctrl)
if ctrl&1 == 0 || dr.DbWatchregs[slot].Addr == 0 {
continue
}
bas := uint8((ctrl >> 5) & 0xFF)
if bas == 0 {
continue
}
lo := bits.TrailingZeros8(bas)
hi := 7 - bits.LeadingZeros8(bas)
addr := dr.DbWatchregs[slot].Addr
if siAddr >= addr+uint64(lo) && siAddr < addr+uint64(hi)+1 {
return StopWatchpoint, siAddr
}
}
return StopSingleStep, 0
}
// FindFreeWatchpointSlot returns the index of the first free watchpoint
// slot, or -1 if all of them are in use.
func (s *Session) FindFreeWatchpointSlot() int {
for i := range maxWatchpoints() {
if !s.wpSlots[i] {
return i
}
}
return -1
}
// IsWatchpointSlotUsed reports whether slot currently holds a watchpoint.
func (s *Session) IsWatchpointSlotUsed(slot int) bool {
if slot < 0 || slot >= maxWatchpoints() {
return false
}
return s.wpSlots[slot]
}
// SetWatchpoint installs a hardware watchpoint on the given address. The
// control word is the architectural DBGWCR (ARM DDI 0487): bit 0 enables,
// bits 3-4 select the access type (10 store, 11 load+store) and bits 5-12
// are the byte-address select, so the watch stays 8-byte aligned and names
// its watched bytes through BAS.
func (s *Session) SetWatchpoint(slot int, addr uint64, typ WatchpointType, size int) error {
if slot < 0 || slot >= maxWatchpoints() {
return fmt.Errorf("debug: watchpoint slot must be 0-%d", maxWatchpoints()-1)
}
if s.wpSlots[slot] {
return fmt.Errorf("debug: watchpoint slot %d already in use", slot)
}
var bas uint64
switch size {
case 1:
bas = 0x01
case 2:
bas = 0x03
case 4:
bas = 0x0F
case 8:
bas = 0xFF
default:
return fmt.Errorf("debug: watchpoint size must be 1, 2, 4, or 8")
}
dr, err := s.getDbRegs()
if err != nil {
return fmt.Errorf("debug: read debug registers: %w", err)
}
if uint8(slot) >= dr.DbNwtpts && dr.DbNwtpts != 0 {
return fmt.Errorf("debug: slot %d exceeds available watchpoints (%d)", slot, dr.DbNwtpts)
}
ctrl := uint64(1) // enable
switch typ {
case WatchWrite:
ctrl |= 2 << 3 // store only
case WatchRead:
ctrl |= 3 << 3 // load+store
}
ctrl |= bas << 5
dr.DbWatchregs[slot].Addr = addr
dr.DbWatchregs[slot].Ctrl = uint32(ctrl)
if err := s.setDbRegs(dr); err != nil {
return fmt.Errorf("debug: set debug registers: %w", err)
}
s.wpSlots[slot] = true
return nil
}
// ClearWatchpoint removes a hardware watchpoint.
func (s *Session) ClearWatchpoint(slot int) error {
if slot < 0 || slot >= maxWatchpoints() {
return fmt.Errorf("debug: watchpoint slot must be 0-%d", maxWatchpoints()-1)
}
if !s.wpSlots[slot] {
return fmt.Errorf("debug: watchpoint slot %d is not in use", slot)
}
dr, err := s.getDbRegs()
if err != nil {
return err
}
dr.DbWatchregs[slot].Addr = 0
dr.DbWatchregs[slot].Ctrl = 0
if err := s.setDbRegs(dr); err != nil {
return err
}
s.wpSlots[slot] = false
return nil
}
// ClearAllWatchpoints removes all hardware watchpoints.
func (s *Session) ClearAllWatchpoints() error {
for slot := range maxWatchpoints() {
if s.wpSlots[slot] {
if err := s.ClearWatchpoint(slot); err != nil {
return err
}
}
}
return nil
}
+50
View File
@@ -0,0 +1,50 @@
// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
//go:build freebsd && riscv64
package debug
import "fmt"
// The architecture has hardware watchpoint triggers, but FreeBSD exposes
// no PT_GETDBREGS request for riscv64, so there is no supported way to arm
// one: the watchpoint layer is honestly empty here.
// WatchpointType selects what triggers the watchpoint.
type WatchpointType int
const (
WatchWrite WatchpointType = 1 // trigger on write
WatchRead WatchpointType = 3 // trigger on read or write
)
// maxWatchpoints reports the number of hardware watchpoint slots the
// platform provides: FreeBSD exposes none for riscv64.
func maxWatchpoints() int { return 0 }
// archStopTrace classifies a TRAP_TRACE stop; with no watchpoint layer a
// trace stop is always a completed single step.
func archStopTrace(s *Session, siAddr uint64) (StopReason, uint64) {
return StopSingleStep, 0
}
// FindFreeWatchpointSlot returns -1: no slots exist.
func (s *Session) FindFreeWatchpointSlot() int { return -1 }
// IsWatchpointSlotUsed reports whether slot currently holds a watchpoint.
func (s *Session) IsWatchpointSlotUsed(slot int) bool { return false }
// SetWatchpoint is unsupported: FreeBSD exposes no debug register request
// for riscv64.
func (s *Session) SetWatchpoint(slot int, addr uint64, typ WatchpointType, size int) error {
return fmt.Errorf("debug: hardware watchpoints are not supported on freebsd/riscv64")
}
// ClearWatchpoint is unsupported for the same reason.
func (s *Session) ClearWatchpoint(slot int) error {
return fmt.Errorf("debug: hardware watchpoints are not supported on freebsd/riscv64")
}
// ClearAllWatchpoints is a no-op: no watchpoint can be armed.
func (s *Session) ClearAllWatchpoints() error { return nil }
+18 -1
View File
@@ -30,13 +30,25 @@ const (
// (arch/x86/kernel/ptrace.c send_sigtrap passes regs->ip), so the watched
// data address is recovered from DR6's slot bits (B0-B3, positive polarity
// through PEEKUSER) and the matching DR0-DR3.
//
// The B0-B3 bits are sticky ("the processor never clears DR6", Intel SDM
// vol 3, "Debug Registers"), so they are acknowledged here: cleared once
// read, or the next hit on a different slot would still see this slot's bit
// set and report this slot's address again.
func archWatchpointAddr(s *Session, siAddr uint64) uint64 {
dr6, err := ptracePeekUser(s.pid, dr6Off)
if err != nil {
return siAddr
}
status := dr6 & 0xF
if status == 0 {
return siAddr
}
// Best effort: the write-back only fails for a debuggee that died, in
// which case no further watchpoint can fire anyway.
_ = ptracePokeUser(s.pid, dr6Off, dr6&^0xF)
for slot := range 4 {
if dr6&(1<<slot) != 0 {
if status&(1<<slot) != 0 {
addr, err := ptracePeekUser(s.pid, drOffset+uintptr(slot*8))
if err == nil && addr != 0 {
return addr
@@ -139,6 +151,11 @@ func (s *Session) ClearWatchpoint(slot int) error {
if err := ptracePokeUser(s.pid, dr7Off, dr7); err != nil {
return err
}
// Zero the address register too: a stale address in a disabled slot
// turns any sticky DR6 bit into a misattributed watchpoint report.
if err := ptracePokeUser(s.pid, drOffset+uintptr(slot*8), 0); err != nil {
return err
}
s.wpSlots[slot] = false
return nil
}
+1 -1
View File
@@ -15,7 +15,7 @@ import (
"golang.org/x/arch/riscv64/riscv64asm"
"golang.org/x/arch/x86/x86asm"
"sourcedock.dev/petrbalvin/gasm-devkit/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
)
// Instruction is one decoded instruction: its text form, its length in bytes
+4 -4
View File
@@ -7,10 +7,10 @@ import (
"strings"
"testing"
"sourcedock.dev/petrbalvin/gasm-devkit/arch"
"sourcedock.dev/petrbalvin/gasm-devkit/asm"
"sourcedock.dev/petrbalvin/gasm-devkit/ast"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
"sourcedock.dev/petrbalvin/gasm-sdk/arch"
"sourcedock.dev/petrbalvin/gasm-sdk/asm"
"sourcedock.dev/petrbalvin/gasm-sdk/ast"
"sourcedock.dev/petrbalvin/gasm-sdk/parser"
)
func TestDecodeKnownBytes(t *testing.T) {

Some files were not shown because too many files have changed in this diff Show More