feat(elf): emit data relocations on arm64, riscv64 and loong64

Assisted-by: GLM 5.3 Flash

asm/elfriscv.go

@@ -24,11 +24,16 @@ const (
 	rRISCVPCRELHI20  = 23 // R_RISCV_PCREL_HI20
 	rRISCVPCRELLO12I = 24 // R_RISCV_PCREL_LO12_I
 	rRISCVPCRELLO12S = 25 // R_RISCV_PCREL_LO12_S
+	// R_RISCV_32 (debug/elf 1): the absolute 32-bit address of a symbol,
+	// the R_ADDR shape a 4-byte DATA field carries.  R_RISCV_64 (2) lives
+	// with the DWARF fixup constants as rRISCVAbs64.
+	rRISVCAbs32 = 1
 )
 
 // ELFRISCVObject returns the image as an ELF64 relocatable object file for
 // RISC-V (EM_RISCV, 64-bit, little-endian).  The structure mirrors the amd64
-// ELF emission: .text, .data, .symtab, .strtab and optional .rela.text.
+// ELF emission: .text, .data, .symtab, .strtab, an optional .rela.text and
+// an optional .rela.data.
 func (img *Image) ELFRISCVObject() ([]byte, error) {
 	le := binary.LittleEndian
 
@@ -129,6 +134,50 @@ func (img *Image) ELFRISCVObject() ([]byte, error) {
 		}
 	}
 
+	// The data symbols' symbol-valued DATA fields ("DATA s+0(SB)/8,
+	// $other(SB)") become .rela.data entries: an absolute relocation of the
+	// DATA line's width at the field's data-section offset, S + A with no
+	// PC term.  Widths 4 and 8 have ELF relocation shapes; narrower fields
+	// cannot hold an address, so they are refused rather than truncated.
+	var dataRelas []elfRela
+	for _, d := range img.DataSyms {
+		for _, r := range d.Relocs {
+			idx, ok := symIdx[r.Name]
+			if !ok {
+				return nil, fmt.Errorf("data relocation references unknown symbol %q", r.Name)
+			}
+			var typ uint32
+			switch r.Siz {
+			case 8:
+				typ = rRISCVAbs64
+			case 4:
+				typ = rRISVCAbs32
+			default:
+				return nil, fmt.Errorf("DATA %q: a symbol value of width %d has no ELF relocation", d.Name, r.Siz)
+			}
+			dataRelas = append(dataRelas, elfRela{
+				off:    uint64(d.Offset + r.Off),
+				sym:    idx,
+				typ:    typ,
+				addend: r.Addend,
+			})
+		}
+	}
+
+	// Section presence: .rela.text only when there are code relocations,
+	// .rela.data only when a DATA line holds a symbol value.
+	hasRela := len(relas) > 0
+	hasDataRela := len(dataRelas) > 0
+	nSections := 6
+	if hasRela {
+		nSections++
+	}
+	if hasDataRela {
+		nSections++
+	}
+	secSymtab, secStrtab := 3, 4
+	secShstr := nSections - 1
+
 	// String tables.
 	stNames := newElfStrtab()
 	for _, s := range syms {
@@ -138,18 +187,13 @@ func (img *Image) ELFRISCVObject() ([]byte, error) {
 	for _, n := range []string{".text", ".data", ".symtab", ".strtab", ".rela.text", ".shstrtab"} {
 		stSections.add(n)
 	}
+	if hasDataRela {
+		stSections.add(".rela.data")
+	}
 	for _, n := range dwarfSectionNames {
 		stSections.add(n)
 	}
 
-	hasRela := len(relas) > 0
-	nSections := 6
-	if hasRela {
-		nSections = 7
-	}
-	secSymtab, secStrtab := 3, 4
-	secShstr := nSections - 1
-
 	// Layout.
 	var out []byte
 	out = append(out, make([]byte, 64)...)
@@ -184,7 +228,7 @@ func (img *Image) ELFRISCVObject() ([]byte, error) {
 	strtabOff := len(out)
 	out = append(out, stNames.bytes()...)
 
-	var relaOff int
+	var relaOff, relaDataOff int
 	if hasRela {
 		align(8)
 		relaOff = len(out)
@@ -196,6 +240,17 @@ func (img *Image) ELFRISCVObject() ([]byte, error) {
 			out = append(out, b[:]...)
 		}
 	}
+	if hasDataRela {
+		align(8)
+		relaDataOff = len(out)
+		for _, r := range dataRelas {
+			var b [24]byte
+			le.PutUint64(b[0:], r.off)
+			le.PutUint64(b[8:], uint64(r.sym)<<32|uint64(r.typ))
+			le.PutUint64(b[16:], uint64(r.addend))
+			out = append(out, b[:]...)
+		}
+	}
 
 	shstrOff := len(out)
 	out = append(out, stSections.bytes()...)
@@ -251,6 +306,9 @@ func (img *Image) ELFRISCVObject() ([]byte, error) {
 	if hasRela {
 		putSh(".rela.text", shtRela, 0, relaOff, 24*len(relas), secSymtab, secText, 8, 24)
 	}
+	if hasDataRela {
+		putSh(".rela.data", shtRela, 0, relaDataOff, 24*len(dataRelas), secSymtab, secData, 8, 24)
+	}
 	putSh(".shstrtab", shtStrtab, 0, shstrOff, len(stSections.bytes()), 0, 0, 1, 0)
 	// DWARF section headers; their indices follow the write order.
 	if dw != nil {