On Tue, Mar 6, 2012 at 8:10 PM, H.J. Lu <hjl.to...@gmail.com> wrote:
> On Tue, Mar 6, 2012 at 7:37 AM, H.J. Lu <hjl.to...@gmail.com> wrote:
>> On Mon, Mar 5, 2012 at 9:11 AM, H.J. Lu <hjl.to...@gmail.com> wrote:
>>> On Sun, Mar 4, 2012 at 11:47 PM, Uros Bizjak <ubiz...@gmail.com> wrote:
>>>> On Mon, Mar 5, 2012 at 4:53 AM, H.J. Lu <hjl.to...@gmail.com> wrote:
>>>>
>>>>> and compiler does generate the same output. i386.c also has
>>>>>
>>>>> xasm = "jmp\t%A0";
>>>>> xasm = "call\t%A0";
>>>>>
>>>>> for calls. There are no separate indirect call patterns. For x32,
>>>>> only indirect register calls have to be in DImode. The direct call
>>>>> should be in Pmode (SImode).
>>>>
>>>> Direct call just expects label to some abolute address that is assumed
>>>> to fit in 32 bits (see constant_call_address_operand_p).
>>>>
>>>> call and jmp insn expect word_mode operands, so please change
>>>> ix86_expand_call and call patterns in the same way as jump
>>>> instructions above.
>>>>
>>>>> Since x86-64 hardware always zero-extends upper 32bits of 64bit
>>>>> registers when loading its lower 32bits, it is safe and easier to just
>>>>> to output 64bit registers for %A than zero-extend it by hand for all
>>>>> jump/call patterns.
>>>>
>>>> No, the instruction expects word_mode operands, so we have to extend
>>>> values to expected mode. I don't think that patching at insn output
>>>> time is acceptable.
>>>
>>> You are right. I found a testcase to show problem:
>>>
>>> struct foo
>>> {
>>> void (*f) (void);
>>> int i;
>>> };
>>>
>>> void
>>> __attribute__ ((noinline))
>>> bar (struct foo x)
>>> {
>>> x.f ();
>>> }
>>>
>>> "x" is passed in RDI and the uppper 32bits of RDI is "int i".
>>>
>>
>
> Hi,
>
> This is what I come up with. This patch enforces word_mode for
> non-SYMBOL_REF call/jmp address while allowing Pmode for
> SYMBOL_REF call/jmp address. I added W for word_mode
> used on indirect branches and added C for Pmode/word_mode
> used on calls. For calls, I added check for SYMBOL_REF address
> or address in word_mode since non-SYMBOL_REF address must
> be in word_mode. Tested on Linux/x86-64.
>
> OK for trunk?
(define_insn_and_split "*sibcall_value_vzeroupper"
[(set (match_operand 0 "" "")
- (call (mem:QI (match_operand:P 1 "sibcall_insn_operand" "Uz"))
+ (call (mem:QI (match_operand:C 1 "sibcall_insn_operand" "Uz"))
(match_operand 2 "" "")))
(unspec [(match_operand 3 "const_int_operand" "")]
UNSPEC_CALL_NEEDS_VZEROUPPER)]
- "TARGET_VZEROUPPER && SIBLING_CALL_P (insn)"
+ "TARGET_VZEROUPPER
+ && SIBLING_CALL_P (insn)
+ && ((GET_CODE (operands[1]) == SYMBOL_REF
+ && GET_MODE (operands[1]) == Pmode)
+ || (GET_CODE (operands[1]) != SYMBOL_REF
+ && GET_MODE (operands[1]) == word_mode))"
"#"
"&& reload_completed"
[(const_int 0)]
Why does this patterh have different insn constraitn than all others?
BTW - please do not split existing "TARGET_VZEROUPPER &&
SIBLING_CALL_P (insn)" lines ...
Uros.
