On 20/11/13 13:57, Richard Earnshaw wrote:
> On 19/11/13 17:48, Jeff Law wrote:
>> On 11/19/13 10:32, Steven Bosscher wrote:
>>>
>>> Yes. In the GCC3 days it was important for sincos on i386, and on mk68
>>> it used to be important for some of the funnier patterns. Not sure if
>>> it's still useful
On 19/11/13 17:48, Jeff Law wrote:
> On 11/19/13 10:32, Steven Bosscher wrote:
>>
>> Yes. In the GCC3 days it was important for sincos on i386, and on mk68
>> it used to be important for some of the funnier patterns. Not sure if
>> it's still useful today, though. Might be worth looking into, just
On 11/19/13 09:43, Richard Earnshaw wrote:
PR 54300 is a problem in regcprop where the compiler sees
(parallel [(set (x) (y)
(set (y) (x)]) (REG_UNUSED (y))
as a single-set insn (since the other operand, y, is not used) and
replaces a use of x with a use of y. However, it fails to
On 11/19/13 10:32, Steven Bosscher wrote:
Yes. In the GCC3 days it was important for sincos on i386, and on mk68
it used to be important for some of the funnier patterns. Not sure if
it's still useful today, though. Might be worth looking into, just to
avoid the confusion in the future.
I doubt
On Tue, Nov 19, 2013 at 6:27 PM, Jeff Law wrote:
> I don't remember the history here, but that may have been to better support
> divmod and similar insns.
Yes. In the GCC3 days it was important for sincos on i386, and on mk68
it used to be important for some of the funnier patterns. Not sure if
it
On Tue, Nov 19, 2013 at 6:22 PM, Richard Earnshaw wrote:
>>> as a single-set insn (since the other operand, y, is not used) and
>>> replaces a use of x with a use of y. However, it fails to take into
>>> account that y has been clobbered in the insn itself.
>>
>> Ah, wait. Incorrect use of single_
On 11/19/13 10:17, Steven Bosscher wrote:
On Tue, Nov 19, 2013 at 5:43 PM, Richard Earnshaw wrote:
PR 54300 is a problem in regcprop where the compiler sees
(parallel [(set (x) (y)
(set (y) (x)]) (REG_UNUSED (y))
as a single-set insn (since the other operand, y, is not used) and
re
On 19/11/13 17:17, Steven Bosscher wrote:
> On Tue, Nov 19, 2013 at 5:43 PM, Richard Earnshaw wrote:
>> PR 54300 is a problem in regcprop where the compiler sees
>> (parallel [(set (x) (y)
>>(set (y) (x)]) (REG_UNUSED (y))
>>
>> as a single-set insn (since the other operand, y, is not
On Tue, Nov 19, 2013 at 5:43 PM, Richard Earnshaw wrote:
> PR 54300 is a problem in regcprop where the compiler sees
> (parallel [(set (x) (y)
>(set (y) (x)]) (REG_UNUSED (y))
>
> as a single-set insn (since the other operand, y, is not used) and
> replaces a use of x with a use of y.
On Tue, Nov 19, 2013 at 5:43 PM, Richard Earnshaw wrote:
> The test is unfortunately ARM specific -- I'm not aware of any generic
> code that triggers this.
Isn't this an insn constraint issue, then? I'd expect recog to reject
the pattern if the right constraints are used.
Ciao!
Steven
PR 54300 is a problem in regcprop where the compiler sees
(parallel [(set (x) (y)
(set (y) (x)]) (REG_UNUSED (y))
as a single-set insn (since the other operand, y, is not used) and
replaces a use of x with a use of y. However, it fails to take into
account that y has been clobbered in
11 matches
Mail list logo
