Hojun-Cho/ww
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

selfhost: mirror writable .data + R_X86_64_64 across the wwstage

Browse Source 
Bring the wwstage toolchain to parity with C-side DATAW / DATAR /
.data / .rela.data support. With this, w6c_ww + w6a_ww + w6l_ww can
compile, assemble and link `let g: str = "lit";` (and the scalar /
str / slice / struct globals that landed earlier) end-to-end, with
output that's byte-identical to the C-side pipeline.

w6a (types.ww / parse.ww / asm.ww / obj.ww):
  - A_DATAW + A_DATAR opcodes; parser learns `name+disp(SB)`;
    A_DATAR records an R_X86_64_64 reloc in .data via the new
    addrelocdata helper; areloc gains a `section` flag and asym
    an `isdata` flag; obj.ww splits relocs into .rela.text /
    .rela.data, emits .data PROGBITS + .rela.data conditionally,
    and shuffles section indices the same way cmd/w6a/obj.c does
    so byte output stays identical when no DATAW/DATAR are used.

w6l (sym.ww / obj.ww / pass.ww / out.ww / dynout.ww / main.ww):
  - lrel grows `section`; lsym grows `indata`; lobj tracks
    dataoff / datasize; lnk grows combined .data buffer;
  - obj.ww loads .data and .rela.data, registers data symbols
    with indata=1 and val shifted by the input's data_off, and
    the archive scanner includes both .text and .data globals;
  - pass.ww adds R_X86_64_64 (patch 8 bytes in .text or .data
    with sym_va + addend); relocate's signature becomes
    (textva, datava);
  - out.ww emits a second PT_LOAD (R+W) when datalen > 0, with
    .data at the page-aligned offset after .text;
  - dynout.ww refuses .data + -l/-L cleanly (matches the C-side
    error message);
  - main.ww computes text_va / data_va and passes both to
    relocate.

wcc cgen (cgen.ww / cgendecl.ww):
  - letpreintern walks top-level str-lets and interns the strlit
    BEFORE emitdatasection emits its DATA row, so emitletdataw
    can later look up the same label;
  - emitletdataw's 16B branch detects non-empty strlit init and
    emits the 8-zero + 8-LE-len DATAW plus a DATAR slot+0,strlit
    reloc, mirroring cmd/w6c/cgen.c.

Verified: `wwdump_ww -c` byte-matches `w6c` on a `let g: str =
"hello world\n";` fixture; `w6a_ww` and `w6l_ww` produce a
binary byte-identical to the C-side pipeline that runs and
prints "hello world". Bootstrap fixed-point holds (ww2 == ww3 ==
ww4), 26/26 tests green.
This commit is contained in:
Hojun-Cho
2026-05-12 13:03:03 +09:00
parent 003f707618
commit 1ac9980f7e
16 changed files with 1132 additions and 145 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
selfhost/cmd/w6l/dynout.ww
View File

@@ -163,6 +163,14 @@ fn streqd(a: str, b: str) bool = {
// ---- main entry --------------------------------------------------------
export fn emitdynelf(l: *lnk, fd: i32, base: u64, entry: u64) i32 = {
// Writable globals on the dynamic-link path need their own R+W
// segment that the existing gotplt + .dynamic layout doesn't
// account for. Refuse rather than silently miscompile; folding
// .data into the existing R+W block is a follow-up.
if (l.datalen > 0u64) {
os.write(2, "w6l: top-level mutable globals (.data) not yet supported with -l/-L\n".ptr, 67u64);
return 1;
};
let a: *arena = l.a;
let n: i32 = l.dynn;
let nu: u64 = n: u64;

244
selfhost/cmd/w6l/main.combined.ww
View File

@@ -345,8 +345,10 @@ use mem;
type lsym = struct {
name: str,
val: u64, // offset within combined .text once linked
val: u64, // offset within combined .text (or .data when
// indata=1) once linked
defined: i32, // 1 if some lobj defines this symbol
indata: i32, // 1 if defined in .data (writable globals)
owner: *lobj,
idxinowner: i32,
// Dynamic-linking fields. Set by resolve when an undefined sym
@@ -361,7 +363,8 @@ type lsym = struct {
};
type lrel = struct {
off: u64, // offset within combined .text
off: u64, // offset within the relocation's section
section: i32, // 0 = .text, 1 = .data
kind: i32, // R_X86_64_*
sym: *lsym,
addend: i64,
@@ -374,6 +377,8 @@ type lobj = struct {
len: u64,
textoff: u64, // offset of .text in combined output
textsize: u64,
dataoff: u64, // offset of .data in combined output
datasize: u64, // bytes contributed to combined .data (0 if none)
onext: *lobj,
};
@@ -402,6 +407,11 @@ type lnk = struct {
text: *u8, // combined .text
textcap: u64,
textlen: u64,
// Combined .data (writable). Empty unless any input .o has a
// .data PROGBITS section.
data: *u8,
datacap: u64,
datalen: u64,
errs: i32,
dynn: i32, // number of syms routed through PLT
};
@@ -564,6 +574,28 @@ fn emittext(l: *lnk, src: *u8, n: u64) void = {
l.textlen += n;
};
fn emitdata(l: *lnk, src: *u8, n: u64) void = {
if (l.datalen + n > l.datacap) {
let nc: u64 = l.datacap;
if (nc == 0u64) { nc = 256u64; };
for (nc < l.datalen + n) { nc = nc * 2u64; };
let nb: *u8 = os.alloc(nc): *u8;
let i: u64 = 0u64;
for (i < l.datalen) {
nb[i] = l.data[i];
i += 1u64;
};
l.data = nb;
l.datacap = nc;
};
let i: u64 = 0u64;
for (i < n) {
l.data[l.datalen + i] = src[i];
i += 1u64;
};
l.datalen += n;
};
// ---- C-string helpers --------------------------------------------------
fn cstrlen(p: *u8) u64 = {
@@ -654,6 +686,7 @@ fn elfglobals(a: *arena, buf: *u8, len: u64) *defent = {
let shstr: *u8 = buf + shstrshoff;
let idxtext: i32 = -1;
let idxdata: i32 = -1;
let idxsymtab: i32 = -1;
let i: u32 = 0u32;
for (i < shnum) {
@@ -663,6 +696,7 @@ fn elfglobals(a: *arena, buf: *u8, len: u64) *defent = {
let nm: *u8 = shstr + (shname: u64);
if (shtype == SHT_PROGBITS: u32) {
if (cstreq(nm, ".text")) { idxtext = i: i32; };
if (cstreq(nm, ".data")) { idxdata = i: i32; };
};
if (shtype == SHT_SYMTAB: u32) { idxsymtab = i: i32; };
i += 1u32;
@@ -688,10 +722,17 @@ fn elfglobals(a: *arena, buf: *u8, len: u64) *defent = {
let stinfo: u8 = buf[symp + SYM_INFO];
let stshndx: u16 = rdu16(buf, symp + SYM_SHNDX);
let bind: u32 = (stinfo: u32) >> 4u32;
// STB_GLOBAL = 1; defined in .text.
// STB_GLOBAL = 1; defined in .text or .data. Both are
// included so an archive member that owns a data global
// gets pulled in when something references it.
if (bind == 1u32) {
if (stshndx != 0u16) {
if ((stshndx: i32) == idxtext) {
let intext: bool = (stshndx: i32) == idxtext;
let indt: bool = false;
if (idxdata >= 0) {
indt = (stshndx: i32) == idxdata;
};
if (intext || indt) {
let nmp: *u8 = strtab + (stname: u64);
if (nmp[0u64] != 0u8) {
let nm: str = cstrtostr(a, nmp);
@@ -813,10 +854,12 @@ fn loadimage(l: *lnk, path: *u8, buf: *u8, len: u64) i32 = {
let shstrshoff: u64 = rdu64(buf, shoff + (shstrndx: u64) * SHDR_SIZE + SHDR_OFFSET);
let shstr: *u8 = buf + shstrshoff;
// find .text, .symtab, .rela.text
// find .text, .data, .symtab, .rela.text, .rela.data
let idxtext: i32 = -1;
let idxdata: i32 = -1;
let idxsymtab: i32 = -1;
let idxrela: i32 = -1;
let idxrelad: i32 = -1;
let i: u32 = 0u32;
for (i < shnum) {
let secoff: u64 = shoff + (i: u64) * SHDR_SIZE;
@@ -825,10 +868,12 @@ fn loadimage(l: *lnk, path: *u8, buf: *u8, len: u64) i32 = {
let nm: *u8 = shstr + (shname: u64);
if (shtype == SHT_PROGBITS: u32) {
if (cstreq(nm, ".text")) { idxtext = i: i32; };
if (cstreq(nm, ".data")) { idxdata = i: i32; };
};
if (shtype == SHT_SYMTAB: u32) { idxsymtab = i: i32; };
if (shtype == SHT_RELA: u32) {
if (cstreq(nm, ".rela.text")) { idxrela = i: i32; };
if (cstreq(nm, ".rela.data")) { idxrelad = i: i32; };
};
i += 1u32;
};
@@ -855,18 +900,34 @@ fn loadimage(l: *lnk, path: *u8, buf: *u8, len: u64) i32 = {
let stroff: u64 = rdu64(buf, strsh + SHDR_OFFSET);
let strtab: *u8 = buf + stroff;
let datasize: u64 = 0u64;
let dataoff: u64 = 0u64;
if (idxdata >= 0) {
let datash: u64 = shoff + (idxdata: u64) * SHDR_SIZE;
dataoff = rdu64(buf, datash + SHDR_OFFSET);
datasize = rdu64(buf, datash + SHDR_SIZE_F);
};
// Track this object.
let ob: *lobj = amalloc(l.a, 64u64): *lobj;
let ob: *lobj = amalloc(l.a, 96u64): *lobj;
ob.path = cstrtostr(l.a, path);
ob.buf = buf;
ob.len = len;
ob.textoff = l.textlen;
ob.textsize = textsize;
ob.dataoff = l.datalen;
ob.datasize = datasize;
ob.onext = l.objs;
l.objs = ob;
// Append .text bytes to the combined image.
emittext(l, buf + textoff, textsize);
// Append .data bytes (if present) to the combined .data buffer.
if (idxdata >= 0) {
if (datasize > 0u64) {
emitdata(l, buf + dataoff, datasize);
};
};
// Walk symbols. We don't keep a per-object map[] of *lsym. Instead
// the reloc loop re-walks symtab and re-interns by name. Simpler
@@ -882,7 +943,13 @@ fn loadimage(l: *lnk, path: *u8, buf: *u8, len: u64) i32 = {
let nm: str = cstrtostr(l.a, nmp);
let gs: *lsym = intern(l, nm);
if (stshndx != 0u16) {
if ((stshndx: i32) == idxtext) {
let intext: bool = (stshndx: i32) == idxtext;
let indt: bool = false;
if (idxdata >= 0) {
indt = (stshndx: i32) == idxdata;
};
if (intext) { if (indt) { indt = false; }; };
if (intext) {
if (gs.defined != 0) {
os.write(2, "w6l: duplicate symbol\n".ptr, 21u64);
l.errs += 1;
@@ -893,6 +960,18 @@ fn loadimage(l: *lnk, path: *u8, buf: *u8, len: u64) i32 = {
gs.val = ob.textoff + stvalue;
};
};
if (indt) {
if (gs.defined != 0) {
os.write(2, "w6l: duplicate symbol\n".ptr, 21u64);
l.errs += 1;
} else {
gs.defined = 1;
gs.indata = 1;
gs.owner = ob;
gs.idxinowner = si: i32;
gs.val = ob.dataoff + stvalue;
};
};
};
};
si += 1u64;
@@ -912,8 +991,9 @@ fn loadimage(l: *lnk, path: *u8, buf: *u8, len: u64) i32 = {
let raddend: u64 = rdu64(buf, rp + RELA_ADDEND);
let rsymidx: u32 = (rinfo >> 32u64): u32;
let rkind: i32 = ((rinfo & 4294967295u64): u32): i32;
let nr: *lrel = amalloc(l.a, 48u64): *lrel;
let nr: *lrel = amalloc(l.a, 64u64): *lrel;
nr.off = ob.textoff + roff;
nr.section = 0;
nr.kind = rkind;
nr.addend = raddend: i64;
// Look up the referenced sym by name (re-walk symtab).
@@ -932,6 +1012,41 @@ fn loadimage(l: *lnk, path: *u8, buf: *u8, len: u64) i32 = {
};
};
// Data-reloc collection. Offsets land in .data, shifted by
// this object's data_off so they index the combined buffer.
if (idxrelad >= 0) {
let relash: u64 = shoff + (idxrelad: u64) * SHDR_SIZE;
let relaoff: u64 = rdu64(buf, relash + SHDR_OFFSET);
let relasize: u64 = rdu64(buf, relash + SHDR_SIZE_F);
let nrel: u64 = relasize / RELA_SIZE;
let ri: u64 = 0u64;
for (ri < nrel) {
let rp: u64 = relaoff + ri * RELA_SIZE;
let roff: u64 = rdu64(buf, rp + RELA_OFFSET);
let rinfo: u64 = rdu64(buf, rp + RELA_INFO);
let raddend: u64 = rdu64(buf, rp + RELA_ADDEND);
let rsymidx: u32 = (rinfo >> 32u64): u32;
let rkind: i32 = ((rinfo & 4294967295u64): u32): i32;
let nr: *lrel = amalloc(l.a, 64u64): *lrel;
nr.off = ob.dataoff + roff;
nr.section = 1;
nr.kind = rkind;
nr.addend = raddend: i64;
if ((rsymidx: u64) < nsyms) {
let sp: u64 = symoff + (rsymidx: u64) * SYM_SIZE;
let sname: u32 = rdu32(buf, sp + SYM_NAME);
let snm: *u8 = strtab + (sname: u64);
if (snm[0u64] != 0u8) {
let nm: str = cstrtostr(l.a, snm);
nr.sym = intern(l, nm);
};
};
nr.rnext = l.rels;
l.rels = nr;
ri += 1u64;
};
};
return 0;
};
@@ -1361,6 +1476,7 @@ use os;
use sym;
use dyn;
def R_X86_64_64: i32 = 1;
def R_X86_64_PC32: i32 = 2;
def R_X86_64_PLT32: i32 = 4;
@@ -1427,7 +1543,15 @@ fn patchu32(p: *u8, v: u32) void = {
p[3] = ((v >> 24u32) & 255u32): u8;
};
export fn relocate(l: *lnk, base: u64) i32 = {
fn patchu64(p: *u8, v: u64) void = {
let i: i32 = 0;
for (i < 8) {
p[i] = ((v >> (i: u64 * 8u64)) & 255u64): u8;
i += 1;
};
};
export fn relocate(l: *lnk, textva: u64, datava: u64) i32 = {
let r: *lrel = l.rels;
for (r != nil) {
if (r.sym != nil) {
@@ -1438,21 +1562,30 @@ export fn relocate(l: *lnk, base: u64) i32 = {
continue;
};
if (r.sym.defined != 0) {
let symva: u64 = textva + r.sym.val;
if (r.sym.indata != 0) { symva = datava + r.sym.val; };
let k: i32 = r.kind;
if (k == R_X86_64_PC32) {
let site: u64 = base + r.off;
let target: i64 = (base + r.sym.val): i64;
let rel: i64 = (target - site: i64) + r.addend;
let site: u64 = textva + r.off;
let rel: i64 = (symva: i64 - site: i64) + r.addend;
patchu32(l.text + r.off, rel: u32);
} else { if (k == R_X86_64_PLT32) {
let site: u64 = base + r.off;
let target: i64 = (base + r.sym.val): i64;
let rel: i64 = (target - site: i64) + r.addend;
let site: u64 = textva + r.off;
let rel: i64 = (symva: i64 - site: i64) + r.addend;
patchu32(l.text + r.off, rel: u32);
} else { if (k == R_X86_64_64) {
// Absolute 64-bit. Currently used only
// for DATAR slots in .data.
let v: u64 = (symva: i64 + r.addend): u64;
if (r.section == 1) {
patchu64(l.data + r.off, v);
} else {
patchu64(l.text + r.off, v);
};
} else {
os.write(2, "w6l: unsupported reloc kind\n".ptr, 27u64);
l.errs += 1;
};};
};};};
};
};
r = r.rnext;
@@ -1626,6 +1759,14 @@ fn streqd(a: str, b: str) bool = {
// ---- main entry --------------------------------------------------------
export fn emitdynelf(l: *lnk, fd: i32, base: u64, entry: u64) i32 = {
// Writable globals on the dynamic-link path need their own R+W
// segment that the existing gotplt + .dynamic layout doesn't
// account for. Refuse rather than silently miscompile; folding
// .data into the existing R+W block is a follow-up.
if (l.datalen > 0u64) {
os.write(2, "w6l: top-level mutable globals (.data) not yet supported with -l/-L\n".ptr, 67u64);
return 1;
};
let a: *arena = l.a;
let n: i32 = l.dynn;
let nu: u64 = n: u64;
@@ -2212,9 +2353,11 @@ def ELFCLASS64: u8 = 2u8;
def ELFDATA2LSB: u8 = 1u8;
def PT_LOAD: u32 = 1u32;
def PF_X: u32 = 1u32;
def PF_W: u32 = 2u32;
def PF_R: u32 = 4u32;
def TEXT_OFF: u64 = 4096u64; // 0x1000
def PAGE_SZ: u64 = 4096u64;
// ---- little-endian byte writers ----------------------------------------
@@ -2246,7 +2389,14 @@ export fn emitelf(l: *lnk, fd: i32, base: u64, entry: u64) i32 = {
};
};
let filesz: u64 = TEXT_OFF + l.textlen;
let hasdata: bool = l.datalen > 0u64;
let rxend: u64 = TEXT_OFF + l.textlen;
// .data lands at the next page boundary so the loader can give
// it fresh R+W permissions without overlapping the R+X mapping.
let dataoff: u64 = 0u64;
if (hasdata) {
dataoff = (rxend + PAGE_SZ - 1u64) & ~(PAGE_SZ - 1u64);
};
// One contiguous header buffer covering [0..0x1000), then .text.
let hdr: *u8 = os.alloc(TEXT_OFF): *u8; // zero-initialised by mmap
@@ -2268,21 +2418,34 @@ export fn emitelf(l: *lnk, fd: i32, base: u64, entry: u64) i32 = {
wru32(hdr, 48u64, 0u32); // e_flags
wru16(hdr, 52u64, 64u16); // e_ehsize
wru16(hdr, 54u64, 56u16); // e_phentsize
wru16(hdr, 56u64, 1u16); // e_phnum
if (hasdata) { wru16(hdr, 56u64, 2u16); }
else { wru16(hdr, 56u64, 1u16); };
wru16(hdr, 58u64, 0u16); // e_shentsize
wru16(hdr, 60u64, 0u16); // e_shnum
wru16(hdr, 62u64, 0u16); // e_shstrndx
// --- Phdr (56 bytes) at offset 64 ---
wru32(hdr, 64u64, PT_LOAD); // p_type
wru32(hdr, 68u64, PF_R | PF_X); // p_flags
// --- Phdr #1 (R+X) at offset 64 ---
wru32(hdr, 64u64, PT_LOAD);
wru32(hdr, 68u64, PF_R | PF_X);
wru64(hdr, 72u64, 0u64); // p_offset
wru64(hdr, 80u64, base); // p_vaddr
wru64(hdr, 88u64, base); // p_paddr
wru64(hdr, 96u64, filesz); // p_filesz
wru64(hdr, 104u64, filesz); // p_memsz
wru64(hdr, 96u64, rxend); // p_filesz
wru64(hdr, 104u64, rxend); // p_memsz
wru64(hdr, 112u64, TEXT_OFF); // p_align
if (hasdata) {
// --- Phdr #2 (R+W) at offset 64+56=120 ---
wru32(hdr, 120u64, PT_LOAD);
wru32(hdr, 124u64, PF_R | PF_W);
wru64(hdr, 128u64, dataoff); // p_offset
wru64(hdr, 136u64, base + dataoff); // p_vaddr
wru64(hdr, 144u64, base + dataoff); // p_paddr
wru64(hdr, 152u64, l.datalen); // p_filesz
wru64(hdr, 160u64, l.datalen); // p_memsz
wru64(hdr, 168u64, PAGE_SZ); // p_align
};
// Write [0..0x1000) then .text.
let r1: (i64 | os.oserror) = os.writeall(fd, hdr, TEXT_OFF);
let n1: i64 = 0i64;
@@ -2300,6 +2463,26 @@ export fn emitelf(l: *lnk, fd: i32, base: u64, entry: u64) i32 = {
};
if (n2 != l.textlen: i64) { return -1; };
};
if (hasdata) {
// Pad to the page-aligned data offset, then write .data.
let here: u64 = TEXT_OFF + l.textlen;
let zero: u8 = 0u8;
for (here < dataoff) {
let r3: (i64 | os.oserror) = os.writeall(fd, &zero, 1u64);
match (r3) {
case let v: i64 => { };
case let e: os.oserror => return -1;
};
here += 1u64;
};
let r4: (i64 | os.oserror) = os.writeall(fd, l.data, l.datalen);
let n4: i64 = 0i64;
match (r4) {
case let v: i64 => n4 = v;
case let e: os.oserror => return -1;
};
if (n4 != l.datalen: i64) { return -1; };
};
return 0;
};
@@ -2324,7 +2507,7 @@ def BASE: u64 = 4194304u64; // 0x400000
def CODE_VA_OFF: u64 = 4096u64; // .text starts at base + 0x1000
fn mklnk(a: *arena) *lnk = {
let l: *lnk = amalloc(a, 112u64): *lnk;
let l: *lnk = amalloc(a, 144u64): *lnk;
l.a = a;
return l;
};
@@ -2604,7 +2787,18 @@ export fn main(argc: i32, argv: **u8) i32 = {
};
if (resolve(l) != 0) { return 1; };
if (relocate(l, BASE + CODE_VA_OFF) != 0) { return 1; };
// Static layout for relocation purposes. .data lands at the
// next page after .text; the dyn path overrides if its layout
// is different (text→text PC32 cancels the absolute VA so
// only the data delta matters here).
let textva: u64 = BASE + CODE_VA_OFF;
let datava: u64 = 0u64;
if (l.datalen > 0u64) {
let page: u64 = 4096u64;
let dataoff: u64 = (CODE_VA_OFF + l.textlen + page - 1u64) & ~(page - 1u64);
datava = BASE + dataoff;
};
if (relocate(l, textva, datava) != 0) { return 1; };
let entrysym: *lsym = lookup(l, "_start");
if (entrysym == nil) { entrysym = lookup(l, "main"); }

15
selfhost/cmd/w6l/main.ww
View File

@@ -18,7 +18,7 @@ def BASE: u64 = 4194304u64; // 0x400000
def CODE_VA_OFF: u64 = 4096u64; // .text starts at base + 0x1000
fn mklnk(a: *arena) *lnk = {
let l: *lnk = amalloc(a, 112u64): *lnk;
let l: *lnk = amalloc(a, 144u64): *lnk;
l.a = a;
return l;
};
@@ -298,7 +298,18 @@ export fn main(argc: i32, argv: **u8) i32 = {
};
if (resolve(l) != 0) { return 1; };
if (relocate(l, BASE + CODE_VA_OFF) != 0) { return 1; };
// Static layout for relocation purposes. .data lands at the
// next page after .text; the dyn path overrides if its layout
// is different (text→text PC32 cancels the absolute VA so
// only the data delta matters here).
let textva: u64 = BASE + CODE_VA_OFF;
let datava: u64 = 0u64;
if (l.datalen > 0u64) {
let page: u64 = 4096u64;
let dataoff: u64 = (CODE_VA_OFF + l.textlen + page - 1u64) & ~(page - 1u64);
datava = BASE + dataoff;
};
if (relocate(l, textva, datava) != 0) { return 1; };
let entrysym: *lsym = lookup(l, "_start");
if (entrysym == nil) { entrysym = lookup(l, "main"); }

117
selfhost/cmd/w6l/obj.ww
View File

@@ -123,6 +123,28 @@ fn emittext(l: *lnk, src: *u8, n: u64) void = {
l.textlen += n;
};
fn emitdata(l: *lnk, src: *u8, n: u64) void = {
if (l.datalen + n > l.datacap) {
let nc: u64 = l.datacap;
if (nc == 0u64) { nc = 256u64; };
for (nc < l.datalen + n) { nc = nc * 2u64; };
let nb: *u8 = os.alloc(nc): *u8;
let i: u64 = 0u64;
for (i < l.datalen) {
nb[i] = l.data[i];
i += 1u64;
};
l.data = nb;
l.datacap = nc;
};
let i: u64 = 0u64;
for (i < n) {
l.data[l.datalen + i] = src[i];
i += 1u64;
};
l.datalen += n;
};
// ---- C-string helpers --------------------------------------------------
fn cstrlen(p: *u8) u64 = {
@@ -213,6 +235,7 @@ fn elfglobals(a: *arena, buf: *u8, len: u64) *defent = {
let shstr: *u8 = buf + shstrshoff;
let idxtext: i32 = -1;
let idxdata: i32 = -1;
let idxsymtab: i32 = -1;
let i: u32 = 0u32;
for (i < shnum) {
@@ -222,6 +245,7 @@ fn elfglobals(a: *arena, buf: *u8, len: u64) *defent = {
let nm: *u8 = shstr + (shname: u64);
if (shtype == SHT_PROGBITS: u32) {
if (cstreq(nm, ".text")) { idxtext = i: i32; };
if (cstreq(nm, ".data")) { idxdata = i: i32; };
};
if (shtype == SHT_SYMTAB: u32) { idxsymtab = i: i32; };
i += 1u32;
@@ -247,10 +271,17 @@ fn elfglobals(a: *arena, buf: *u8, len: u64) *defent = {
let stinfo: u8 = buf[symp + SYM_INFO];
let stshndx: u16 = rdu16(buf, symp + SYM_SHNDX);
let bind: u32 = (stinfo: u32) >> 4u32;
// STB_GLOBAL = 1; defined in .text.
// STB_GLOBAL = 1; defined in .text or .data. Both are
// included so an archive member that owns a data global
// gets pulled in when something references it.
if (bind == 1u32) {
if (stshndx != 0u16) {
if ((stshndx: i32) == idxtext) {
let intext: bool = (stshndx: i32) == idxtext;
let indt: bool = false;
if (idxdata >= 0) {
indt = (stshndx: i32) == idxdata;
};
if (intext || indt) {
let nmp: *u8 = strtab + (stname: u64);
if (nmp[0u64] != 0u8) {
let nm: str = cstrtostr(a, nmp);
@@ -372,10 +403,12 @@ fn loadimage(l: *lnk, path: *u8, buf: *u8, len: u64) i32 = {
let shstrshoff: u64 = rdu64(buf, shoff + (shstrndx: u64) * SHDR_SIZE + SHDR_OFFSET);
let shstr: *u8 = buf + shstrshoff;
// find .text, .symtab, .rela.text
// find .text, .data, .symtab, .rela.text, .rela.data
let idxtext: i32 = -1;
let idxdata: i32 = -1;
let idxsymtab: i32 = -1;
let idxrela: i32 = -1;
let idxrelad: i32 = -1;
let i: u32 = 0u32;
for (i < shnum) {
let secoff: u64 = shoff + (i: u64) * SHDR_SIZE;
@@ -384,10 +417,12 @@ fn loadimage(l: *lnk, path: *u8, buf: *u8, len: u64) i32 = {
let nm: *u8 = shstr + (shname: u64);
if (shtype == SHT_PROGBITS: u32) {
if (cstreq(nm, ".text")) { idxtext = i: i32; };
if (cstreq(nm, ".data")) { idxdata = i: i32; };
};
if (shtype == SHT_SYMTAB: u32) { idxsymtab = i: i32; };
if (shtype == SHT_RELA: u32) {
if (cstreq(nm, ".rela.text")) { idxrela = i: i32; };
if (cstreq(nm, ".rela.data")) { idxrelad = i: i32; };
};
i += 1u32;
};
@@ -414,18 +449,34 @@ fn loadimage(l: *lnk, path: *u8, buf: *u8, len: u64) i32 = {
let stroff: u64 = rdu64(buf, strsh + SHDR_OFFSET);
let strtab: *u8 = buf + stroff;
let datasize: u64 = 0u64;
let dataoff: u64 = 0u64;
if (idxdata >= 0) {
let datash: u64 = shoff + (idxdata: u64) * SHDR_SIZE;
dataoff = rdu64(buf, datash + SHDR_OFFSET);
datasize = rdu64(buf, datash + SHDR_SIZE_F);
};
// Track this object.
let ob: *lobj = amalloc(l.a, 64u64): *lobj;
let ob: *lobj = amalloc(l.a, 96u64): *lobj;
ob.path = cstrtostr(l.a, path);
ob.buf = buf;
ob.len = len;
ob.textoff = l.textlen;
ob.textsize = textsize;
ob.dataoff = l.datalen;
ob.datasize = datasize;
ob.onext = l.objs;
l.objs = ob;
// Append .text bytes to the combined image.
emittext(l, buf + textoff, textsize);
// Append .data bytes (if present) to the combined .data buffer.
if (idxdata >= 0) {
if (datasize > 0u64) {
emitdata(l, buf + dataoff, datasize);
};
};
// Walk symbols. We don't keep a per-object map[] of *lsym. Instead
// the reloc loop re-walks symtab and re-interns by name. Simpler
@@ -441,7 +492,13 @@ fn loadimage(l: *lnk, path: *u8, buf: *u8, len: u64) i32 = {
let nm: str = cstrtostr(l.a, nmp);
let gs: *lsym = intern(l, nm);
if (stshndx != 0u16) {
if ((stshndx: i32) == idxtext) {
let intext: bool = (stshndx: i32) == idxtext;
let indt: bool = false;
if (idxdata >= 0) {
indt = (stshndx: i32) == idxdata;
};
if (intext) { if (indt) { indt = false; }; };
if (intext) {
if (gs.defined != 0) {
os.write(2, "w6l: duplicate symbol\n".ptr, 21u64);
l.errs += 1;
@@ -452,6 +509,18 @@ fn loadimage(l: *lnk, path: *u8, buf: *u8, len: u64) i32 = {
gs.val = ob.textoff + stvalue;
};
};
if (indt) {
if (gs.defined != 0) {
os.write(2, "w6l: duplicate symbol\n".ptr, 21u64);
l.errs += 1;
} else {
gs.defined = 1;
gs.indata = 1;
gs.owner = ob;
gs.idxinowner = si: i32;
gs.val = ob.dataoff + stvalue;
};
};
};
};
si += 1u64;
@@ -471,8 +540,9 @@ fn loadimage(l: *lnk, path: *u8, buf: *u8, len: u64) i32 = {
let raddend: u64 = rdu64(buf, rp + RELA_ADDEND);
let rsymidx: u32 = (rinfo >> 32u64): u32;
let rkind: i32 = ((rinfo & 4294967295u64): u32): i32;
let nr: *lrel = amalloc(l.a, 48u64): *lrel;
let nr: *lrel = amalloc(l.a, 64u64): *lrel;
nr.off = ob.textoff + roff;
nr.section = 0;
nr.kind = rkind;
nr.addend = raddend: i64;
// Look up the referenced sym by name (re-walk symtab).
@@ -491,5 +561,40 @@ fn loadimage(l: *lnk, path: *u8, buf: *u8, len: u64) i32 = {
};
};
// Data-reloc collection. Offsets land in .data, shifted by
// this object's data_off so they index the combined buffer.
if (idxrelad >= 0) {
let relash: u64 = shoff + (idxrelad: u64) * SHDR_SIZE;
let relaoff: u64 = rdu64(buf, relash + SHDR_OFFSET);
let relasize: u64 = rdu64(buf, relash + SHDR_SIZE_F);
let nrel: u64 = relasize / RELA_SIZE;
let ri: u64 = 0u64;
for (ri < nrel) {
let rp: u64 = relaoff + ri * RELA_SIZE;
let roff: u64 = rdu64(buf, rp + RELA_OFFSET);
let rinfo: u64 = rdu64(buf, rp + RELA_INFO);
let raddend: u64 = rdu64(buf, rp + RELA_ADDEND);
let rsymidx: u32 = (rinfo >> 32u64): u32;
let rkind: i32 = ((rinfo & 4294967295u64): u32): i32;
let nr: *lrel = amalloc(l.a, 64u64): *lrel;
nr.off = ob.dataoff + roff;
nr.section = 1;
nr.kind = rkind;
nr.addend = raddend: i64;
if ((rsymidx: u64) < nsyms) {
let sp: u64 = symoff + (rsymidx: u64) * SYM_SIZE;
let sname: u32 = rdu32(buf, sp + SYM_NAME);
let snm: *u8 = strtab + (sname: u64);
if (snm[0u64] != 0u8) {
let nm: str = cstrtostr(l.a, snm);
nr.sym = intern(l, nm);
};
};
nr.rnext = l.rels;
l.rels = nr;
ri += 1u64;
};
};
return 0;
};

56
selfhost/cmd/w6l/out.ww
View File

@@ -18,9 +18,11 @@ def ELFCLASS64: u8 = 2u8;
def ELFDATA2LSB: u8 = 1u8;
def PT_LOAD: u32 = 1u32;
def PF_X: u32 = 1u32;
def PF_W: u32 = 2u32;
def PF_R: u32 = 4u32;
def TEXT_OFF: u64 = 4096u64; // 0x1000
def PAGE_SZ: u64 = 4096u64;
// ---- little-endian byte writers ----------------------------------------
@@ -52,7 +54,14 @@ export fn emitelf(l: *lnk, fd: i32, base: u64, entry: u64) i32 = {
};
};
let filesz: u64 = TEXT_OFF + l.textlen;
let hasdata: bool = l.datalen > 0u64;
let rxend: u64 = TEXT_OFF + l.textlen;
// .data lands at the next page boundary so the loader can give
// it fresh R+W permissions without overlapping the R+X mapping.
let dataoff: u64 = 0u64;
if (hasdata) {
dataoff = (rxend + PAGE_SZ - 1u64) & ~(PAGE_SZ - 1u64);
};
// One contiguous header buffer covering [0..0x1000), then .text.
let hdr: *u8 = os.alloc(TEXT_OFF): *u8; // zero-initialised by mmap
@@ -74,21 +83,34 @@ export fn emitelf(l: *lnk, fd: i32, base: u64, entry: u64) i32 = {
wru32(hdr, 48u64, 0u32); // e_flags
wru16(hdr, 52u64, 64u16); // e_ehsize
wru16(hdr, 54u64, 56u16); // e_phentsize
wru16(hdr, 56u64, 1u16); // e_phnum
if (hasdata) { wru16(hdr, 56u64, 2u16); }
else { wru16(hdr, 56u64, 1u16); };
wru16(hdr, 58u64, 0u16); // e_shentsize
wru16(hdr, 60u64, 0u16); // e_shnum
wru16(hdr, 62u64, 0u16); // e_shstrndx
// --- Phdr (56 bytes) at offset 64 ---
wru32(hdr, 64u64, PT_LOAD); // p_type
wru32(hdr, 68u64, PF_R | PF_X); // p_flags
// --- Phdr #1 (R+X) at offset 64 ---
wru32(hdr, 64u64, PT_LOAD);
wru32(hdr, 68u64, PF_R | PF_X);
wru64(hdr, 72u64, 0u64); // p_offset
wru64(hdr, 80u64, base); // p_vaddr
wru64(hdr, 88u64, base); // p_paddr
wru64(hdr, 96u64, filesz); // p_filesz
wru64(hdr, 104u64, filesz); // p_memsz
wru64(hdr, 96u64, rxend); // p_filesz
wru64(hdr, 104u64, rxend); // p_memsz
wru64(hdr, 112u64, TEXT_OFF); // p_align
if (hasdata) {
// --- Phdr #2 (R+W) at offset 64+56=120 ---
wru32(hdr, 120u64, PT_LOAD);
wru32(hdr, 124u64, PF_R | PF_W);
wru64(hdr, 128u64, dataoff); // p_offset
wru64(hdr, 136u64, base + dataoff); // p_vaddr
wru64(hdr, 144u64, base + dataoff); // p_paddr
wru64(hdr, 152u64, l.datalen); // p_filesz
wru64(hdr, 160u64, l.datalen); // p_memsz
wru64(hdr, 168u64, PAGE_SZ); // p_align
};
// Write [0..0x1000) then .text.
let r1: (i64 | os.oserror) = os.writeall(fd, hdr, TEXT_OFF);
let n1: i64 = 0i64;
@@ -106,5 +128,25 @@ export fn emitelf(l: *lnk, fd: i32, base: u64, entry: u64) i32 = {
};
if (n2 != l.textlen: i64) { return -1; };
};
if (hasdata) {
// Pad to the page-aligned data offset, then write .data.
let here: u64 = TEXT_OFF + l.textlen;
let zero: u8 = 0u8;
for (here < dataoff) {
let r3: (i64 | os.oserror) = os.writeall(fd, &zero, 1u64);
match (r3) {
case let v: i64 => { };
case let e: os.oserror => return -1;
};
here += 1u64;
};
let r4: (i64 | os.oserror) = os.writeall(fd, l.data, l.datalen);
let n4: i64 = 0i64;
match (r4) {
case let v: i64 => n4 = v;
case let e: os.oserror => return -1;
};
if (n4 != l.datalen: i64) { return -1; };
};
return 0;
};

34
selfhost/cmd/w6l/pass.ww
View File

@@ -14,6 +14,7 @@ use os;
use sym;
use dyn;
def R_X86_64_64: i32 = 1;
def R_X86_64_PC32: i32 = 2;
def R_X86_64_PLT32: i32 = 4;
@@ -80,7 +81,15 @@ fn patchu32(p: *u8, v: u32) void = {
p[3] = ((v >> 24u32) & 255u32): u8;
};
export fn relocate(l: *lnk, base: u64) i32 = {
fn patchu64(p: *u8, v: u64) void = {
let i: i32 = 0;
for (i < 8) {
p[i] = ((v >> (i: u64 * 8u64)) & 255u64): u8;
i += 1;
};
};
export fn relocate(l: *lnk, textva: u64, datava: u64) i32 = {
let r: *lrel = l.rels;
for (r != nil) {
if (r.sym != nil) {
@@ -91,21 +100,30 @@ export fn relocate(l: *lnk, base: u64) i32 = {
continue;
};
if (r.sym.defined != 0) {
let symva: u64 = textva + r.sym.val;
if (r.sym.indata != 0) { symva = datava + r.sym.val; };
let k: i32 = r.kind;
if (k == R_X86_64_PC32) {
let site: u64 = base + r.off;
let target: i64 = (base + r.sym.val): i64;
let rel: i64 = (target - site: i64) + r.addend;
let site: u64 = textva + r.off;
let rel: i64 = (symva: i64 - site: i64) + r.addend;
patchu32(l.text + r.off, rel: u32);
} else { if (k == R_X86_64_PLT32) {
let site: u64 = base + r.off;
let target: i64 = (base + r.sym.val): i64;
let rel: i64 = (target - site: i64) + r.addend;
let site: u64 = textva + r.off;
let rel: i64 = (symva: i64 - site: i64) + r.addend;
patchu32(l.text + r.off, rel: u32);
} else { if (k == R_X86_64_64) {
// Absolute 64-bit. Currently used only
// for DATAR slots in .data.
let v: u64 = (symva: i64 + r.addend): u64;
if (r.section == 1) {
patchu64(l.data + r.off, v);
} else {
patchu64(l.text + r.off, v);
};
} else {
os.write(2, "w6l: unsupported reloc kind\n".ptr, 27u64);
l.errs += 1;
};};
};};};
};
};
r = r.rnext;

14
selfhost/cmd/w6l/sym.ww
View File

@@ -7,8 +7,10 @@ use mem;
type lsym = struct {
name: str,
val: u64, // offset within combined .text once linked
val: u64, // offset within combined .text (or .data when
// indata=1) once linked
defined: i32, // 1 if some lobj defines this symbol
indata: i32, // 1 if defined in .data (writable globals)
owner: *lobj,
idxinowner: i32,
// Dynamic-linking fields. Set by resolve when an undefined sym
@@ -23,7 +25,8 @@ type lsym = struct {
};
type lrel = struct {
off: u64, // offset within combined .text
off: u64, // offset within the relocation's section
section: i32, // 0 = .text, 1 = .data
kind: i32, // R_X86_64_*
sym: *lsym,
addend: i64,
@@ -36,6 +39,8 @@ type lobj = struct {
len: u64,
textoff: u64, // offset of .text in combined output
textsize: u64,
dataoff: u64, // offset of .data in combined output
datasize: u64, // bytes contributed to combined .data (0 if none)
onext: *lobj,
};
@@ -64,6 +69,11 @@ type lnk = struct {
text: *u8, // combined .text
textcap: u64,
textlen: u64,
// Combined .data (writable). Empty unless any input .o has a
// .data PROGBITS section.
data: *u8,
datacap: u64,
datalen: u64,
errs: i32,
dynn: i32, // number of syms routed through PLT
};