On Wed, Jan 20, 2016 at 03:14:16PM -0800, Roland McGrath wrote:
> > Other addresses in this code are printed with ("%#" PRIx64).
> > (AFAIK '#' here will always give "0x", but lets keep it consistent).
>
> %#x prints "0" for zero (not "0x0"). Indeed it is the right thing to use
> here.
>
> > It's also curious that nothing is done with op1 after printing it.
> > Shouldn't this set pc, at least?
>
> I think it should, but that is unrelated to this change.
> Arguably, it should be:
>
> get_uleb128 (op1, readp, endp);
> op1 *= code_align;
> printf (" set_loc %#" PRIx64 " to %#" PRIx64 "\n",
> op1, pc = vma_base + op1);
>
> so that it displays the literal value (like the other cases do) as well as
> the final value (which in this case is just with VMA_BASE added in).
I agree with using %# for printing the address and setting the pc using
vma_base and op1. But the argument of a DW_CFA_set_loc is an address, not
a uleb128 and it should not be adjusted by code_align, which is only used
for CFA_advance operations as far as I understand it. To read an address
we need to know the FDE address encoding and use read_encoded (which
isn't defined till a bit later, so we need to move it up).
So I believe the full change should be:
diff --git a/src/readelf.c b/src/readelf.c
index 0db192e..a25e4ac 100644
--- a/src/readelf.c
+++ b/src/readelf.c
@@ -5044,11 +5044,68 @@ register_info (Ebl *ebl, unsigned int regno, const
Ebl_Register_Location *loc,
return set;
}
+static const unsigned char *
+read_encoded (unsigned int encoding, const unsigned char *readp,
+ const unsigned char *const endp, uint64_t *res, Dwarf *dbg)
+{
+ if ((encoding & 0xf) == DW_EH_PE_absptr)
+ encoding = gelf_getclass (dbg->elf) == ELFCLASS32
+ ? DW_EH_PE_udata4 : DW_EH_PE_udata8;
+
+ switch (encoding & 0xf)
+ {
+ case DW_EH_PE_uleb128:
+ get_uleb128 (*res, readp, endp);
+ break;
+ case DW_EH_PE_sleb128:
+ get_sleb128 (*res, readp, endp);
+ break;
+ case DW_EH_PE_udata2:
+ if (readp + 2 > endp)
+ goto invalid;
+ *res = read_2ubyte_unaligned_inc (dbg, readp);
+ break;
+ case DW_EH_PE_udata4:
+ if (readp + 4 > endp)
+ goto invalid;
+ *res = read_4ubyte_unaligned_inc (dbg, readp);
+ break;
+ case DW_EH_PE_udata8:
+ if (readp + 8 > endp)
+ goto invalid;
+ *res = read_8ubyte_unaligned_inc (dbg, readp);
+ break;
+ case DW_EH_PE_sdata2:
+ if (readp + 2 > endp)
+ goto invalid;
+ *res = read_2sbyte_unaligned_inc (dbg, readp);
+ break;
+ case DW_EH_PE_sdata4:
+ if (readp + 4 > endp)
+ goto invalid;
+ *res = read_4sbyte_unaligned_inc (dbg, readp);
+ break;
+ case DW_EH_PE_sdata8:
+ if (readp + 8 > endp)
+ goto invalid;
+ *res = read_8sbyte_unaligned_inc (dbg, readp);
+ break;
+ default:
+ invalid:
+ error (1, 0,
+ gettext ("invalid encoding"));
+ }
+
+ return readp;
+}
+
+
static void
print_cfa_program (const unsigned char *readp, const unsigned char *const endp,
Dwarf_Word vma_base, unsigned int code_align,
int data_align,
unsigned int version, unsigned int ptr_size,
+ unsigned int encoding,
Dwfl_Module *dwflmod, Ebl *ebl, Dwarf *dbg)
{
char regnamebuf[REGNAMESZ];
@@ -5079,9 +5136,9 @@ print_cfa_program (const unsigned char *readp, const
unsigned char *const endp,
case DW_CFA_set_loc:
if ((uint64_t) (endp - readp) < 1)
goto invalid;
- get_uleb128 (op1, readp, endp);
- op1 += vma_base;
- printf (" set_loc %" PRIu64 "\n", op1 * code_align);
+ readp = read_encoded (encoding, readp, endp, &op1, dbg);
+ printf (" set_loc %#" PRIx64 " to %#" PRIx64 "\n",
+ op1, pc = vma_base + op1);
break;
case DW_CFA_advance_loc1:
if ((uint64_t) (endp - readp) < 1)
@@ -5421,62 +5478,6 @@ print_encoding_base (const char *pfx, unsigned int
fde_encoding)
}
-static const unsigned char *
-read_encoded (unsigned int encoding, const unsigned char *readp,
- const unsigned char *const endp, uint64_t *res, Dwarf *dbg)
-{
- if ((encoding & 0xf) == DW_EH_PE_absptr)
- encoding = gelf_getclass (dbg->elf) == ELFCLASS32
- ? DW_EH_PE_udata4 : DW_EH_PE_udata8;
-
- switch (encoding & 0xf)
- {
- case DW_EH_PE_uleb128:
- get_uleb128 (*res, readp, endp);
- break;
- case DW_EH_PE_sleb128:
- get_sleb128 (*res, readp, endp);
- break;
- case DW_EH_PE_udata2:
- if (readp + 2 > endp)
- goto invalid;
- *res = read_2ubyte_unaligned_inc (dbg, readp);
- break;
- case DW_EH_PE_udata4:
- if (readp + 4 > endp)
- goto invalid;
- *res = read_4ubyte_unaligned_inc (dbg, readp);
- break;
- case DW_EH_PE_udata8:
- if (readp + 8 > endp)
- goto invalid;
- *res = read_8ubyte_unaligned_inc (dbg, readp);
- break;
- case DW_EH_PE_sdata2:
- if (readp + 2 > endp)
- goto invalid;
- *res = read_2sbyte_unaligned_inc (dbg, readp);
- break;
- case DW_EH_PE_sdata4:
- if (readp + 4 > endp)
- goto invalid;
- *res = read_4sbyte_unaligned_inc (dbg, readp);
- break;
- case DW_EH_PE_sdata8:
- if (readp + 8 > endp)
- goto invalid;
- *res = read_8sbyte_unaligned_inc (dbg, readp);
- break;
- default:
- invalid:
- error (1, 0,
- gettext ("invalid encoding"));
- }
-
- return readp;
-}
-
-
static void
print_debug_frame_section (Dwfl_Module *dwflmod, Ebl *ebl, GElf_Ehdr *ehdr,
Elf_Scn *scn, GElf_Shdr *shdr, Dwarf *dbg)
@@ -5851,7 +5852,7 @@ print_debug_frame_section (Dwfl_Module *dwflmod, Ebl
*ebl, GElf_Ehdr *ehdr,
else
print_cfa_program (readp, cieend, vma_base, code_alignment_factor,
data_alignment_factor, version, ptr_size,
- dwflmod, ebl, dbg);
+ fde_encoding, dwflmod, ebl, dbg);
readp = cieend;
}
}
Ben, do you happen to have a testcase around for this?
Thanks,
Mark
