More aggressive threading causing loop-interchange-9.c regression

[Jeff Law]

On 9/9/2021 2:14 AM, Aldy Hernandez wrote:
>
>
> On 9/8/21 8:13 PM, Michael Matz wrote:
>> Hello,
>>
>> [lame answer to self]
>>
>> On Wed, 8 Sep 2021, Michael Matz wrote:
>>
>>>>> The forward threader guards against this by simply disallowing
>>>>> threadings that involve different loops.  As I see
>>>>
>>>> The thread in question (5->9->3) is all within the same outer loop,
>>>> though. BTW, the backward threader also disallows threading across
>>>> different loops (see path_crosses_loops variable).
>> ...
>>> Maybe it's possible to not disable threading over latches 
>>> alltogether in
>>> the backward threader (like it's tried now), but I haven't looked at 
>>> the
>>> specific situation here in depth, so take my view only as opinion 
>>> from a
>>> large distance :-)
>>
>> I've now looked at the concrete situation.  So yeah, the whole path 
>> is in
>> the same loop, crosses the latch, _and there's code following the latch
>> on that path_.  (I.e. the latch isn't the last block in the path).  In
>> particular, after loop_optimizer_init() (before any threading) we have:
>>
>>    <bb 3> [local count: 118111600]:
>>    # j_19 = PHI <j_13(9), 0(7)>
>>    sum_11 = c[j_19];
>>    if (n_10(D) > 0)
>>      goto <bb 8>; [89.00%]
>>    else
>>      goto <bb 5>; [11.00%]
>>
>>       <bb 8> [local count: 105119324]:
>> ...
>>
>>    <bb 5> [local count: 118111600]:
>>    # sum_21 = PHI <sum_14(4), sum_11(3)>
>>    c[j_19] = sum_21;
>>    j_13 = j_19 + 1;
>>    if (n_10(D) > j_13)
>>      goto <bb 9>; [89.00%]
>>    else
>>      goto <bb 6>; [11.00%]
>>
>>    <bb 9> [local count: 105119324]:
>>    goto <bb 3>; [100.00%]
>>
>> With bb9 the outer (empty) latch, bb3 the outer header, and bb8 the
>> pre-header of inner loop, but more importantly something that's not 
>> at the
>> start of the outer loop.
>>
>> Now, any thread that includes the backedge 9->3 _including_ its
>> destination (i.e. where the backedge isn't the last to-be-redirected 
>> edge)
>> necessarily duplicates all code from that destination onto the back 
>> edge.
>> Here it's the load from c[j] into sum_11.
>>
>> The important part is the code is emitted onto the back edge,
>> conceptually; in reality it's simply included into the (new) latch block
>> (the duplicate of bb9, which is bb12 intermediately, then named bb7 
>> after
>> cfg_cleanup).
>>
>> That's what we can't have for some of our structural loop optimizers:
>> there must be no code executed after the exit test (e.g. in the latch
>> block).  (This requirement makes reasoning about which code is or isn't
>> executed completely for an iteration trivial; simply everything in the
>> body is always executed; e.g. loop interchange uses this to check that
>> there are no memory references after the exit test, because those would
>> then be only conditional and hence make loop interchange very awkward).
>>
>> Note that this situation can't be later rectified anymore: the 
>> duplicated
>> instructions (because they are memory refs) must remain after the exit
>> test.  Only by rerolling/unrotating the loop (i.e. noticing that the
>> memory refs on the loop-entry path and on the back edge are equivalent)
>> would that be possible, but that's something we aren't capable of.  Even
>> if we were that would simply just revert the whole work that the 
>> threader
>> did, so it's better to not even do that to start with.
>>
>> I believe something like below would be appropriate, it disables 
>> threading
>> if the path contains a latch at the non-last position (due to being
>> backwards on the non-first position in the array).  I.e. it disables
>> rotating the loop if there's danger of polluting the back edge. It might
>> be improved if the blocks following (preceding!) the latch are themself
>> empty because then no code is duplicated.  It might also be improved if
>> the latch is already non-empty.  That code should probably only be 
>> active
>> before the loop optimizers, but currently the backward threader isn't
>> differentiating between before/after loop-optims.
>>
>> I haven't tested this patch at all, except that it fixes the testcase :)
>
> Thanks for looking at this.
>
> I think you're onto something with this approach.  Perhaps in addition 
> to the loop header threading Richard mentions.
>
> Your patch causes some regressions, but I think most are noise from 
> FSM tests that must be adjusted.  However, there are some other ones 
> that are curious:
>
> > FAIL: gcc.dg/tree-ssa/ldist-22.c scan-tree-dump ldist "generated 
> memset zero"
> > FAIL: gcc.dg/tree-ssa/pr66752-3.c scan-tree-dump-not dce2 "if .flag"
> < XFAIL: gcc.dg/shrink-wrap-loop.c scan-rtl-dump pro_and_epilogue 
> "Performing shrink-wrapping"
> < XFAIL: gcc.dg/Warray-bounds-87.c pr101671 (test for bogus messages, 
> line 36)
> > FAIL: libgomp.graphite/force-parallel-4.c scan-tree-dump-times 
> graphite "1 loops carried no dependency" 1
> > FAIL: libgomp.graphite/force-parallel-4.c scan-tree-dump-times 
> optimized "loopfn.1" 4
> > FAIL: libgomp.graphite/force-parallel-8.c scan-tree-dump-times 
> graphite "5 loops carried no dependency" 1
>
> Interestingly your patch is fixing shrink-wrap-loop.c and 
> Warray-bounds-87, both of which were introduced by the backward 
> threader rewrite.  At least the Warray-bounds was the threader peeling 
> off an iteration that caused a bogus warning.
>
> The ldist-22 regression is interesting though:
>
> void foo ()
> {
>   int i;
>
>   <bb 2> :
>   goto <bb 6>; [INV]
>
>   <bb 3> :
>   a[i_1] = 0;
>   if (i_1 > 100)
>     goto <bb 4>; [INV]
>   else
>     goto <bb 5>; [INV]
>
>   <bb 4> :
>   b[i_1] = i_1;
>
>   <bb 5> :
>   i_8 = i_1 + 1;
>
>   <bb 6> :
>   # i_1 = PHI <0(2), i_8(5)>
>   if (i_1 <= 1023)
>     goto <bb 3>; [INV]
>   else
>     goto <bb 7>; [INV]
>
>   <bb 7> :
>   return;
>
> }
>
> Here we fail to look past 5->6 because BB5 is the latch and is not the 
> last block in the path.  So we fail to thread 3->5->6->3.  Doing so 
> would have split the function into two loops, one of which could use a 
> memset:
>
> void foo ()
> {
>   int i;
>
>   <bb 2> :
>   goto <bb 6>; [INV]
>
>   <bb 3> :
>   # i_12 = PHI <i_1(6), i_9(4)>
>   a[i_12] = 0;
>   if (i_12 > 100)
>     goto <bb 5>; [INV]
>   else
>     goto <bb 4>; [INV]
>
>   <bb 4> :
>   i_9 = i_12 + 1;
>   goto <bb 3>; [100.00%]
>
>   <bb 5> :
>   b[i_12] = i_12;
>   i_8 = i_12 + 1;
>
>   <bb 6> :
>   # i_1 = PHI <0(2), i_8(5)>
>   if (i_1 <= 1023)
>     goto <bb 3>; [INV]
>   else
>     goto <bb 7>; [INV]
>
>   <bb 7> :
>   return;
>
> }
>
> I would have to agree that threading through latches is problematic. 
> For that matter, the ldist-22 test shows that we're depending on the 
> threader to do work that seems to belong in the loop optimizer world.
>
> Would it be crazy to suggest that we disable threading through latches 
> altogether, and do whatever we're missing in the loop world?  It seems 
> loop has all the tools, cost model, and framework to do so.  Of 
> course, I know 0 about loop, and would hate to add work to other's 
> plates.
Threading through latches can be extremely problematical as they can 
change the underlying loop structure.    For example, it can take a 
simple loop and turn it into a loop nest.

I thought we'd already disabled threading through latches except for the 
case where doing so allows us to thread through an indirect jump (ie, 
the original motivation for the FSM threader).

As I mentioned in our private email, threading through headers, through 
latches, across loops, etc has a number of negative effects and we 
generally want to avoid that, at least in the instances before loop 
optimization and vectorization.  THe threader doesn't have any cost 
model for what is effectively loop peeling, loop rotatation, loop header 
copying and the like.  Zdenek argued those belong in the loop 
optimizers, not jump threading and I broadly agree with that assessment.

Jeff