[Bug middle-end/110148] [14 Regression] TSVC s242 regression between g:c0df96b3cda5738afbba3a65bb054183c5cd5530 and g:e4c986fde56a6248f8fbe6cf0704e1da34b055d8
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110148 Martin Jambor changed: What|Removed |Added CC||jamborm at gcc dot gnu.org Status|UNCONFIRMED |RESOLVED Resolution|--- |FIXED --- Comment #8 from Martin Jambor --- (In reply to cuilili from comment #7) > (In reply to Martin Jambor from comment #6) > > I believe this has been fixed? > > Yes. Closing the bug then.
[Bug middle-end/110148] [14 Regression] TSVC s242 regression between g:c0df96b3cda5738afbba3a65bb054183c5cd5530 and g:e4c986fde56a6248f8fbe6cf0704e1da34b055d8
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110148 --- Comment #7 from cuilili --- (In reply to Martin Jambor from comment #6) > I believe this has been fixed? Yes.
[Bug middle-end/110148] [14 Regression] TSVC s242 regression between g:c0df96b3cda5738afbba3a65bb054183c5cd5530 and g:e4c986fde56a6248f8fbe6cf0704e1da34b055d8
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110148 --- Comment #6 from Martin Jambor --- I believe this has been fixed?
[Bug middle-end/110148] [14 Regression] TSVC s242 regression between g:c0df96b3cda5738afbba3a65bb054183c5cd5530 and g:e4c986fde56a6248f8fbe6cf0704e1da34b055d8
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110148
--- Comment #5 from CVS Commits ---
The master branch has been updated by Lili Cui :
https://gcc.gnu.org/g:4633e38cd22c5e51fac984124c7627be912d0999
commit r14-2185-g4633e38cd22c5e51fac984124c7627be912d0999
Author: Lili Cui
Date: Thu Jun 29 06:51:56 2023 +
Avoid adding loop-carried ops to long chains
Avoid adding loop-carried ops to long chains, otherwise the whole chain
will
have dependencies across the loop iteration. Just keep loop-carried ops in
a
separate chain.
E.g.
x_1 = phi(x_0, x_2)
y_1 = phi(y_0, y_2)
a + b + c + d + e + x1 + y1
SSA1 = a + b;
SSA2 = c + d;
SSA3 = SSA1 + e;
SSA4 = SSA3 + SSA2;
SSA5 = x1 + y1;
SSA6 = SSA4 + SSA5;
With the patch applied, these test cases improved by 32%~100%.
S242:
for (int i = 1; i < LEN_1D; ++i) {
a[i] = a[i - 1] + s1 + s2 + b[i] + c[i] + d[i];}
Case 1:
for (int i = 1; i < LEN_1D; ++i) {
a[i] = a[i - 1] + s1 + s2 + b[i] + c[i] + d[i] + e[i];}
Case 2:
for (int i = 1; i < LEN_1D; ++i) {
a[i] = a[i - 1] + b[i - 1] + s1 + s2 + b[i] + c[i] + d[i] + e[i];}
The value is the execution time
A: original version
B: with FMA patch g:e5405f065bace0685cb3b8878d1dfc7a6e7ef409(base on A)
C: with current patch(base on B)
A B C B/A C/A
s2422.859 5.152 2.859 1.802028681 1
case 1 5.489 5.488 3.511 0.9998180.64
case 2 7.216 7.499 4.885 1.0392180.68
gcc/ChangeLog:
PR tree-optimization/110148
* tree-ssa-reassoc.cc (rewrite_expr_tree_parallel): Handle
loop-carried
ops in this function.
[Bug middle-end/110148] [14 Regression] TSVC s242 regression between g:c0df96b3cda5738afbba3a65bb054183c5cd5530 and g:e4c986fde56a6248f8fbe6cf0704e1da34b055d8
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110148 --- Comment #4 from Jan Hubicka --- zen3 fma requires all inputs to be ready to start execution, separate multiply+add can start multiplication earlier. Not sure if that explains the difference.
[Bug middle-end/110148] [14 Regression] TSVC s242 regression between g:c0df96b3cda5738afbba3a65bb054183c5cd5530 and g:e4c986fde56a6248f8fbe6cf0704e1da34b055d8
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110148
--- Comment #3 from cuilili ---
I reproduced S1244 regression on znver3.
Src code:
for (int i = 0; i < LEN_1D-1; i++)
{
a[i] = b[i] + c[i] * c[i] + b[i] * b[i] + c[i];
d[i] = a[i] + a[i+1];
}
Base version: Base + commit version:
Assembler Assembler
Loop1:Loop1:
vmovsd 0x60c400(%rax),%xmm2 vmovsd 0x60ba00(%rax),%xmm2
vmovsd 0x60ba00(%rax),%xmm1 vmovsd 0x60c400(%rax),%xmm1
add$0x8,%rax add$0x8,%rax
vaddsd %xmm1,%xmm2,%xmm0 vmovsd %xmm2,%xmm2,%xmm0
vmulsd %xmm2,%xmm2,%xmm2 vfmadd132sd %xmm2,%xmm1,%xmm0
vfmadd132sd %xmm1,%xmm2,%xmm1 vfmadd132sd %xmm1,%xmm2,%xmm1
vaddsd %xmm1,%xmm0,%xmm0 vaddsd %xmm1,%xmm0,%xmm0
vmovsd %xmm0,0x60cdf8(%rax) vmovsd %xmm0,0x60cdf8(%rax)
vaddsd 0x60ce00(%rax),%xmm0,%xmm0 vaddsd 0x60ce00(%rax),%xmm0,%xmm0
vmovsd %xmm0,0x60aff8(%rax) vmovsd %xmm0,0x60aff8(%rax)
cmp$0x9f8,%raxcmp$0x9f8,%rax
jneLoop1: jneLoop1
For the Base version, mult and FMA have dependencies, which increases the
latency of the critical dependency chain. I didn't find out why znver3 has
regression. Same binary running on ICX has 11% gain (with #define iterations
1).
[Bug middle-end/110148] [14 Regression] TSVC s242 regression between g:c0df96b3cda5738afbba3a65bb054183c5cd5530 and g:e4c986fde56a6248f8fbe6cf0704e1da34b055d8
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110148
cuilili changed:
What|Removed |Added
CC||lili.cui at intel dot com
--- Comment #2 from cuilili ---
The commit changed the break dependency chain function, in order to generate
more FMA. S242 has a chain that needs to be broken. The chain is in a small
loop and related with the loop reduction variable a[i-1].
Src code:
for (int i = 1; i < LEN_1D; ++i)
{
a[i] = a[i - 1] + s1 + s2 + b[i] + c[i] + d[i];
}
--
Base version:
SSA tree
ssa1 = (s1+s2) + b[i];
ssa2 = c[i] + d[i];
ssa3 = ssa1+ssa2;
ssa4 = ssa3 + a[i-1]
a[i-1] uses xmm1, there are 2 instructions using xmm0 have dependencies across
iterations
Assembler
Loop1:
vmovsd 0x60c400(%rax),%xmm0
vaddsd 0x60b000(%rax),%xmm3,%xmm2
add$0x8,%rax
vaddsd 0x60b9f8(%rax),%xmm0,%xmm0
vaddsd %xmm2,%xmm0,%xmm0
vaddsd %xmm0,%xmm1,%xmm1 ---> 1
vmovsd %xmm1,0x60cdf8(%rax) ---> 2
cmp$0xa00,%rdx
jneLoop1
--
Base + commit g:e5405f065bace0685cb3b8878d1dfc7a6e7ef409 version:
a[i-1] uses xmm0, there are 4 instructions using xmm0 have dependencies across
iterations
SSA tree
ssa1 = (s1+s2) + b[i];
ssa2 = c[i] + d[i];
ssa3 = ssa1 + a[i-1]
ssa3 = ssa2 + ssa3;
Assembler
Loop1:
vaddsdq 0x60b000(%rax), %xmm0, %xmm0 ---> 1
vmovsdq 0x60c400(%rax), %xmm1
add $0x8, %rax
vaddsdq 0x60b9f8(%rax), %xmm1, %xmm1
vaddsd %xmm2, %xmm0, %xmm0 ---> 2
vaddsd %xmm1, %xmm0, %xmm0 ---> 3
vmovsdq %xmm0, 0x60cdf8(%rax) ---> 4
cmp$0xa00,%rdx
jneLoop1
[Bug middle-end/110148] [14 Regression] TSVC s242 regression between g:c0df96b3cda5738afbba3a65bb054183c5cd5530 and g:e4c986fde56a6248f8fbe6cf0704e1da34b055d8
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110148 Richard Biener changed: What|Removed |Added Blocks||53947 Target||x86_64-*-* Version|13.1.0 |14.0 Summary|TSVC s242 regression|[14 Regression] TSVC s242 |between |regression between |g:c0df96b3cda5738afbba3a65b |g:c0df96b3cda5738afbba3a65b |b054183c5cd5530 and |b054183c5cd5530 and |g:e4c986fde56a6248f8fbe6cf0 |g:e4c986fde56a6248f8fbe6cf0 |704e1da34b055d8 |704e1da34b055d8 Target Milestone|--- |14.0 Keywords||needs-bisection Referenced Bugs: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=53947 [Bug 53947] [meta-bug] vectorizer missed-optimizations
