On Sat, 13 Jul 2002, Tom Hughes wrote:
> Of course... The attached patch should handle that I think...
This patch is breaking several Solaris 32-bit tests. The following
assembly (from t/pmc/perlarray1.pbc):
new P0,.PerlArray
set P0,0
set I0,P0
print I0
print "\n"
set P0,1
set I0,P0
print I0
print "\n"
set P0,5
set I0,P0
print I0
print "\n"
end
Is assembling to this (with this patch installed):
#
# Disassembly of Parrot Byte Code from 't/pmc/perlarray1.pbc'
#
# Segments:
#
# * Wordsize: 1 bytes
# * Byteorder: 1 bytes
# * Major: 1 bytes
# * Minor: 1 bytes
# * Flags: 1 bytes
# * FloatType: 1 bytes
# * Unused: 10 bytes
# * Magic Number: 8 bytes
# * Opcode Type: 8 bytes
# * Fixup Table: 0 bytes
# * Const Table: 64 bytes
# * Byte Code: 272 bytes (34 opcode_ts)
#
# Constant Type Data
# -------- ------------ ------------------------------
# 0 PFC_STRING "\n"
#
# WORD BYTE BYTE CODE LABEL
OPERATION ARGUMENTS
# -------- ---------- ------------------------------------------------ ------
--------------- --------------------
00000000 [00000000]: 00000751 00000000 00000005
loadlib P0, S5
00000006 [00000024]: 00000088 00000000 00000000 set
P0, 0
00000012 [00000048]: 00000093 00000000 00000000 set
I0, P0
00000018 [00000072]: 00000019 00000000
print I0
00000022 [00000088]: 00000024 00000000
print [sc:0]
00000026 [00000104]: 00000088 00000000 00000001 set
P0, 1
00000032 [00000128]: 00000093 00000000 00000000 set
I0, P0
00000038 [00000152]: 00000019 00000000
print I0
00000042 [00000168]: 00000024 00000000
print [sc:0]
00000046 [00000184]: 00000088 00000000 00000005 set
P0, 5
00000052 [00000208]: 00000093 00000000 00000000 set
I0, P0
00000058 [00000232]: 00000019 00000000
print I0
00000062 [00000248]: 00000024 00000000
print [sc:0]
00000066 [00000264]: 00000000 end
When I roll back the patch, this is how it (correctly) disassembles:
#
# Disassembly of Parrot Byte Code from 't/pmc/perlarray1.pbc'
#
# Segments:
#
# * Wordsize: 1 bytes
# * Byteorder: 1 bytes
# * Major: 1 bytes
# * Minor: 1 bytes
# * Flags: 1 bytes
# * FloatType: 1 bytes
# * Unused: 10 bytes
# * Magic Number: 8 bytes
# * Opcode Type: 8 bytes
# * Fixup Table: 0 bytes
# * Const Table: 64 bytes
# * Byte Code: 272 bytes (34 opcode_ts)
#
# Constant Type Data
# -------- ------------ ------------------------------
# 0 PFC_STRING "\n"
#
# WORD BYTE BYTE CODE LABEL
OPERATION ARGUMENTS
# -------- ---------- ------------------------------------------------ ------
--------------- --------------------
00000000 [00000000]: 00000725 00000000 00000005 new
P0, 5
00000006 [00000024]: 00000088 00000000 00000000 set
P0, 0
00000012 [00000048]: 00000093 00000000 00000000 set
I0, P0
00000018 [00000072]: 00000019 00000000
print I0
00000022 [00000088]: 00000024 00000000
print [sc:0]
00000026 [00000104]: 00000088 00000000 00000001 set
P0, 1
00000032 [00000128]: 00000093 00000000 00000000 set
I0, P0
00000038 [00000152]: 00000019 00000000
print I0
00000042 [00000168]: 00000024 00000000
print [sc:0]
00000046 [00000184]: 00000088 00000000 00000005 set
P0, 5
00000052 [00000208]: 00000093 00000000 00000000 set
I0, P0
00000058 [00000232]: 00000019 00000000
print I0
00000062 [00000248]: 00000024 00000000
print [sc:0]
00000066 [00000264]: 00000000 end
I didn't roll it back in CVS (yet) because it's probably a small glitch
but I didn't investigate to thouroghly (yet).
- D
<[EMAIL PROTECTED]>
