On Thu, Jan 15, 2015 at 10:39 AM, Yuri Rumyantsev wrote:
> Hi All,
>
> I did a change proposed by Richard - unconditionally allocate from the heap.
>
> Bootstrap and regression testing did not show any new failures.
>
> Is it OK for trunk?
+ if (!is_gimple_assign (stmt)
+ || gimple_h
Hi All,
I did a change proposed by Richard - unconditionally allocate from the heap.
Bootstrap and regression testing did not show any new failures.
Is it OK for trunk?
ChangeLog
2015-01-15 Yuri Rumyantsev
PR tree-optimization/64434
* cfgexpand.c (reorder_operands): New function.
(expand_g
Jakub,
I did all changes requested by you.
Here is updated patch.
BTW I thought that gcc performs splitting of blocks with huge size.
2015-01-14 16:33 GMT+03:00 Jakub Jelinek :
> On Wed, Jan 14, 2015 at 04:28:42PM +0300, Yuri Rumyantsev wrote:
>> Hi All,
>>
>> I did all changes proposed by Ri
On Wed, Jan 14, 2015 at 2:33 PM, Jakub Jelinek wrote:
> On Wed, Jan 14, 2015 at 04:28:42PM +0300, Yuri Rumyantsev wrote:
>> Hi All,
>>
>> I did all changes proposed by Richard and delete check on def in the
>> same block as Jakub proposed.
>> I also moved check on optimization to call site..
>>
>
On Wed, Jan 14, 2015 at 04:28:42PM +0300, Yuri Rumyantsev wrote:
> Hi All,
>
> I did all changes proposed by Richard and delete check on def in the
> same block as Jakub proposed.
> I also moved check on optimization to call site..
>
> I also checked that bootstrap and regression testing did not
Hi All,
I did all changes proposed by Richard and delete check on def in the
same block as Jakub proposed.
I also moved check on optimization to call site..
I also checked that bootstrap and regression testing did not show any
new failures.
Is it OK for trunk?
2015-01-14 14:02 GMT+03:00 Jakub
On Wed, Jan 14, 2015 at 11:58:50AM +0100, Richard Biener wrote:
> >> + /* Swap operands if the second one is more expensive. */
> >> + def0 = get_gimple_for_ssa_name (op0);
> >> + if (!def0)
> >> + continue;
> >> + def1 = get_gimple_for_ssa_name (op1);
> >> + if (!def1
On Wed, Jan 14, 2015 at 11:45 AM, Jakub Jelinek
wrote:
> On Wed, Jan 14, 2015 at 01:32:13PM +0300, Yuri Rumyantsev wrote:
>> Sorry, I resend correct patch.
>
>> +reorder_operands (basic_block bb)
>> +{
>> + unsigned int *lattice; /* Hold cost of each statement. */
>> + unsigned int i = 0, n =
On Wed, Jan 14, 2015 at 01:32:13PM +0300, Yuri Rumyantsev wrote:
> Sorry, I resend correct patch.
> +reorder_operands (basic_block bb)
> +{
> + unsigned int *lattice; /* Hold cost of each statement. */
> + unsigned int i = 0, n = 0;
> + gimple_stmt_iterator gsi;
> + gimple_seq stmts;
> + gi
Sorry, I resend correct patch.
Yuri.
2015-01-14 13:23 GMT+03:00 Jakub Jelinek :
> On Wed, Jan 14, 2015 at 01:12:20PM +0300, Yuri Rumyantsev wrote:
>> Hi All,
>>
>> Here is updated patch which was redesigned accordingly to Richard review.
>> It performs swapping operands of commutative operations
On Wed, Jan 14, 2015 at 01:12:20PM +0300, Yuri Rumyantsev wrote:
> Hi All,
>
> Here is updated patch which was redesigned accordingly to Richard review.
> It performs swapping operands of commutative operations to expand the
> expensive one first.
>
> Bootstrap and regression testing did not show
11 matches
Mail list logo
