/* This will be used when we sort the dynamic relocation records. */
static bfd *reldyn_sorting_bfd;
+/* True if ABFD is for CPUs with load interlocking that include
+ non-MIPS1 CPUs and R3900. */
+#define LOAD_INTERLOCKS_P(abfd) \
+ ( ((elf_elfheader (abfd)->e_flags & EF_MIPS_ARCH) != E_MIPS_ARCH_1) \
+ || ((elf_elfheader (abfd)->e_flags & EF_MIPS_MACH) == E_MIPS_MACH_3900))
+
+/* True if ABFD is for CPUs that are faster if JAL is converted to BAL.
+ This should be safe for all architectures. We enable this predicate
+ for RM9000 for now. */
+#define JAL_TO_BAL_P(abfd) \
+ ((elf_elfheader (abfd)->e_flags & EF_MIPS_MACH) == E_MIPS_MACH_9000)
+
+/* True if ABFD is for CPUs that are faster if JALR is converted to BAL.
+ This should be safe for all architectures. We enable this predicate for
+ all CPUs. */
+#define JALR_TO_BAL_P(abfd) 1
+
/* True if ABFD is a PIC object. */
#define PIC_OBJECT_P(abfd) \
((elf_elfheader (abfd)->e_flags & EF_MIPS_PIC) != 0)
#define CALL_FP_STUB_P(name) CONST_STRNEQ (name, CALL_FP_STUB)
\f
/* The format of the first PLT entry in an O32 executable. */
-static const bfd_vma mips_o32_exec_plt0_entry[] = {
+static const bfd_vma mips_o32_exec_plt0_entry[] =
+{
0x3c1c0000, /* lui $28, %hi(&GOTPLT[0]) */
0x8f990000, /* lw $25, %lo(&GOTPLT[0])($28) */
0x279c0000, /* addiu $28, $28, %lo(&GOTPLT[0]) */
/* The format of the first PLT entry in an N32 executable. Different
because gp ($28) is not available; we use t2 ($14) instead. */
-static const bfd_vma mips_n32_exec_plt0_entry[] = {
+static const bfd_vma mips_n32_exec_plt0_entry[] =
+{
0x3c0e0000, /* lui $14, %hi(&GOTPLT[0]) */
0x8dd90000, /* lw $25, %lo(&GOTPLT[0])($14) */
0x25ce0000, /* addiu $14, $14, %lo(&GOTPLT[0]) */
/* The format of the first PLT entry in an N64 executable. Different
from N32 because of the increased size of GOT entries. */
-static const bfd_vma mips_n64_exec_plt0_entry[] = {
+static const bfd_vma mips_n64_exec_plt0_entry[] =
+{
0x3c0e0000, /* lui $14, %hi(&GOTPLT[0]) */
0xddd90000, /* ld $25, %lo(&GOTPLT[0])($14) */
0x25ce0000, /* addiu $14, $14, %lo(&GOTPLT[0]) */
};
/* The format of subsequent PLT entries. */
-static const bfd_vma mips_exec_plt_entry[] = {
+static const bfd_vma mips_exec_plt_entry[] =
+{
0x3c0f0000, /* lui $15, %hi(.got.plt entry) */
0x01f90000, /* l[wd] $25, %lo(.got.plt entry)($15) */
0x25f80000, /* addiu $24, $15, %lo(.got.plt entry) */
};
/* The format of the first PLT entry in a VxWorks executable. */
-static const bfd_vma mips_vxworks_exec_plt0_entry[] = {
+static const bfd_vma mips_vxworks_exec_plt0_entry[] =
+{
0x3c190000, /* lui t9, %hi(_GLOBAL_OFFSET_TABLE_) */
0x27390000, /* addiu t9, t9, %lo(_GLOBAL_OFFSET_TABLE_) */
0x8f390008, /* lw t9, 8(t9) */
};
/* The format of subsequent PLT entries. */
-static const bfd_vma mips_vxworks_exec_plt_entry[] = {
+static const bfd_vma mips_vxworks_exec_plt_entry[] =
+{
0x10000000, /* b .PLT_resolver */
0x24180000, /* li t8, <pltindex> */
0x3c190000, /* lui t9, %hi(<.got.plt slot>) */
};
/* The format of the first PLT entry in a VxWorks shared object. */
-static const bfd_vma mips_vxworks_shared_plt0_entry[] = {
+static const bfd_vma mips_vxworks_shared_plt0_entry[] =
+{
0x8f990008, /* lw t9, 8(gp) */
0x00000000, /* nop */
0x03200008, /* jr t9 */
};
/* The format of subsequent PLT entries. */
-static const bfd_vma mips_vxworks_shared_plt_entry[] = {
+static const bfd_vma mips_vxworks_shared_plt_entry[] =
+{
0x10000000, /* b .PLT_resolver */
0x24180000 /* li t8, <pltindex> */
};
x = (x & ~(0x3f << 26)) | (jalx_opcode << 26);
}
- /* On the RM9000, bal is faster than jal, because bal uses branch
- prediction hardware. If we are linking for the RM9000, and we
- see jal, and bal fits, use it instead. Note that this
- transformation should be safe for all architectures. */
- if (bfd_get_mach (input_bfd) == bfd_mach_mips9000
- && !info->relocatable
+ /* Try converting JAL and JALR to BAL, if the target is in range. */
+ if (!info->relocatable
&& !require_jalx
- && ((r_type == R_MIPS_26 && (x >> 26) == 0x3) /* jal addr */
- || (r_type == R_MIPS_JALR && x == 0x0320f809))) /* jalr t9 */
+ && ((JAL_TO_BAL_P (input_bfd)
+ && r_type == R_MIPS_26
+ && (x >> 26) == 0x3) /* jal addr */
+ || (JALR_TO_BAL_P (input_bfd)
+ && r_type == R_MIPS_JALR
+ && x == 0x0320f809))) /* jalr t9 */
{
bfd_vma addr;
bfd_vma dest;
{
const char *name = bfd_get_section_name (abfd, hdr->bfd_section);
+ /* .sbss is not handled specially here because the GNU/Linux
+ prelinker can convert .sbss from NOBITS to PROGBITS and
+ changing it back to NOBITS breaks the binary. The entry in
+ _bfd_mips_elf_special_sections will ensure the correct flags
+ are set on .sbss if BFD creates it without reading it from an
+ input file, and without special handling here the flags set
+ on it in an input file will be followed. */
if (strcmp (name, ".sdata") == 0
|| strcmp (name, ".lit8") == 0
|| strcmp (name, ".lit4") == 0)
hdr->sh_flags |= SHF_ALLOC | SHF_WRITE | SHF_MIPS_GPREL;
hdr->sh_type = SHT_PROGBITS;
}
- else if (strcmp (name, ".sbss") == 0)
- {
- hdr->sh_flags |= SHF_ALLOC | SHF_WRITE | SHF_MIPS_GPREL;
- hdr->sh_type = SHT_NOBITS;
- }
else if (strcmp (name, ".srdata") == 0)
{
hdr->sh_flags |= SHF_ALLOC | SHF_MIPS_GPREL;
symbol. We mark symbols as small common if appropriate. This is
also where we undo the increment of the value for a mips16 symbol. */
-bfd_boolean
+int
_bfd_mips_elf_link_output_symbol_hook
(struct bfd_link_info *info ATTRIBUTE_UNUSED,
const char *name ATTRIBUTE_UNUSED, Elf_Internal_Sym *sym,
if (ELF_ST_IS_MIPS16 (sym->st_other))
sym->st_value &= ~1;
- return TRUE;
+ return 1;
}
\f
/* Functions for the dynamic linker. */
else if (s == htab->splt)
{
/* If the last PLT entry has a branch delay slot, allocate
- room for an extra nop to fill the delay slot. */
- if (!htab->is_vxworks && s->size > 0)
+ room for an extra nop to fill the delay slot. This is
+ for CPUs without load interlocking. */
+ if (! LOAD_INTERLOCKS_P (output_bfd)
+ && ! htab->is_vxworks && s->size > 0)
s->size += 4;
}
else if (! CONST_STRNEQ (name, ".init")
plt_entry = mips_exec_plt_entry;
bfd_put_32 (output_bfd, plt_entry[0] | got_address_high, loc);
bfd_put_32 (output_bfd, plt_entry[1] | got_address_low | load, loc + 4);
- bfd_put_32 (output_bfd, plt_entry[2] | got_address_low, loc + 8);
- bfd_put_32 (output_bfd, plt_entry[3], loc + 12);
+
+ if (! LOAD_INTERLOCKS_P (output_bfd))
+ {
+ bfd_put_32 (output_bfd, plt_entry[2] | got_address_low, loc + 8);
+ bfd_put_32 (output_bfd, plt_entry[3], loc + 12);
+ }
+ else
+ {
+ bfd_put_32 (output_bfd, plt_entry[3], loc + 8);
+ bfd_put_32 (output_bfd, plt_entry[2] | got_address_low, loc + 12);
+ }
/* Emit an R_MIPS_JUMP_SLOT relocation against the .got.plt entry. */
mips_elf_output_dynamic_relocation (output_bfd, htab->srelplt,
projects / deliverable / binutils-gdb.git / blobdiff