# New Ticket Created by "Jesse Taylor"
# Please include the string: [perl #51634]
# in the subject line of all future correspondence about this issue.
# <URL: http://rt.perl.org/rt3/Ticket/Display.html?id=51634 >
fixup_unpack() in src/packfile.c contains the following code:
for (i = 0; i < self->fixup_count; i++) {
PackFile_FixupEntry * const entry =
self->fixups[i] =
mem_allocate_typed(PackFile_FixupEntry);
entry->type = PF_fetch_opcode(pf, &cursor);
switch (entry->type) {
case enum_fixup_label:
case enum_fixup_sub:
entry->name = PF_fetch_cstring(pf, &cursor);
entry->offset = PF_fetch_opcode(pf, &cursor);
break;
case enum_fixup_none:
break;
default:
PIO_eprintf(interp,
"PackFile_FixupTable_unpack: Unknown fixup type
%d!\n",
entry->type);
return NULL;
}
}
And include/parrot/packfile.h has the following:
typedef enum {
enum_fixup_none,
enum_fixup_label,
enum_fixup_sub
} enum_fixup_t;
When walking through fixup_unpack() in GDB, I get the following:
Breakpoint 4, fixup_unpack (interp=0x805f7e0, seg=0x825c5d8,
cursor=0x9d570120) at src/packfile.c:3043
...
...
3072 entry->type = PF_fetch_opcode(pf, &cursor);
(gdb) n
3073 switch (entry->type) {
(gdb) p entry->type
$13 = 2
However, the docs -- namely docs/pdds/pdd13_bytecode.pod (in the section
titled "Fixup Segment") and docs/parrotbyte.pod (in the section "Fixup
segment") say that the type for a subroutine fixup should be 0x01, not 0x02
...
This should either be corrected if wrong, or the docs should be cleared up a
bit otherwise to make the meaning more clear.
Thanks,
Jesse Taylor