Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Updating mpi-vector-mode branch with main #1595

Closed
wants to merge 63 commits into from
Closed
Changes from 1 commit
Commits
Show all changes
63 commits
Select commit Hold shift + click to select a range
7508ad4
Fix SCEV memory error (#1524)
wsmoses Nov 4, 2023
a1d95f5
[Blas] fix temporary allocation (and deallocation) (#1525)
wsmoses Nov 6, 2023
3339d74
Fix Blas diffuse on P^3 computation (#1526)
wsmoses Nov 6, 2023
7d2b630
Fix stability of minCut (#1527)
wsmoses Nov 7, 2023
9559b6e
Add nv_atan2 explicitly (#1530)
wsmoses Nov 8, 2023
35614bf
Additional differential use stability (#1529)
wsmoses Nov 8, 2023
82318b8
Ring of springs integration test using modulo (#1521)
martinjm97 Nov 8, 2023
91d70ef
Improve single type of store info for print (#1531)
wsmoses Nov 8, 2023
fd2a2ed
Add support for multityped store operation (#1532)
wsmoses Nov 8, 2023
7b629b8
acos and asin explicit nv (#1536)
jgreener64 Nov 8, 2023
42aaaa6
Add flag to disable activity analysis (#1537)
wsmoses Nov 9, 2023
09cedd9
Update to latest llvm (#1534)
wsmoses Nov 9, 2023
2d71fd7
Place load at top of build file (#1538)
wsmoses Nov 10, 2023
72695c4
Add werror for unused vars (#1539)
wsmoses Nov 10, 2023
c4736f3
Forbid dangling else (#1540)
wsmoses Nov 10, 2023
ab705bf
Fix mlir warning to errors (#1541)
wsmoses Nov 10, 2023
74a25dc
permit unused in scev (#1542)
wsmoses Nov 11, 2023
f59fcf0
Fix MPI differential use analysis (#1548)
michel2323 Nov 15, 2023
0655023
Fix blas diff use (#1551)
wsmoses Nov 16, 2023
10f0847
Make nicer error for insert multitype (#1553)
wsmoses Nov 17, 2023
99e5f7b
drop outdated SCEV backports (#1554)
ZuseZ4 Nov 17, 2023
a6e607d
Add error about taping non-instruction value (#1550)
wsmoses Nov 17, 2023
227b443
Add nv isnan/isinf (#1557)
wsmoses Nov 20, 2023
7f28ab0
potential store (#1558)
wsmoses Nov 21, 2023
f18cf5b
[mlir] Update EnzymeMLIR pieces to LLVM 18 (#1556)
pengmai Nov 22, 2023
6732d3d
Add flag to disable recursive activity analysis (#1559)
pengmai Nov 23, 2023
1135f76
Handle insert value of multi type (#1560)
wsmoses Nov 27, 2023
0fd562c
Fix atomic error message (#1562)
wsmoses Nov 28, 2023
2b0f8c2
Dont compile xerbla (#1561)
wsmoses Nov 28, 2023
1f1e996
handle non integral mustexitscev (#1563)
wsmoses Nov 28, 2023
77b4fff
[forwardmode] handle multi store on active store (#1564)
wsmoses Nov 28, 2023
5701321
Jmp/dataflow activity analysis (#1549)
wsmoses Dec 1, 2023
7094012
Fix crash on no return use (#1575)
wsmoses Dec 9, 2023
cbb9701
add postadprint (#1576)
wsmoses Dec 9, 2023
233a736
Fix alwaysinlinediff flag
wsmoses Dec 9, 2023
c27e7bc
Add more version bounds to max/min (#1566)
wsmoses Dec 10, 2023
f1eba59
use early_inc_range (#1567)
tgymnich Dec 11, 2023
a59cb89
Add getReturnDiffeType api specifying a mode (#1579)
wsmoses Dec 12, 2023
79c4322
Adapt to llvm main abi update (#1577)
wsmoses Dec 12, 2023
ee3ffb8
use early_inc_range 2 (#1578)
tgymnich Dec 12, 2023
409cbb8
Fix needsshadow in call handler (#1580)
wsmoses Dec 13, 2023
7284903
Make nicer sparse arg error (#1582)
wsmoses Dec 14, 2023
f940df8
shadow alloc rewriter (#1583)
wsmoses Dec 15, 2023
0570dfa
Embarassing bugfix for shadowalloc rewrite (#1584)
wsmoses Dec 15, 2023
03ff726
Add sparse prints (#1545)
wsmoses Dec 15, 2023
d22dd49
Use static clang plugin registration (#1544)
wsmoses Dec 15, 2023
a4f429f
Add nvidia reciprocal (#1586)
wsmoses Dec 17, 2023
405fac0
Zero sret copy (#1585)
wsmoses Dec 17, 2023
48cbcca
Nicer error for getindex (#1587)
wsmoses Dec 17, 2023
b5b81f8
[TypeAnalysis] improve memtransfer error handler (#1588)
wsmoses Dec 17, 2023
247f09e
[TypeAnalysis] handle arg count mismatch (#1589)
wsmoses Dec 17, 2023
022b3e5
[TypeAnalysis] improve arg error message (#1590)
wsmoses Dec 17, 2023
7c0f813
Ring of springs integration test using modulo
martinjm97 Nov 3, 2023
95a7a43
new bugs
martinjm97 Nov 7, 2023
6e26376
add increasingly hard tests ringspring3Dextenddata.cpp, ringspring3D…
martinjm97 Nov 13, 2023
fb6a1c6
[Sparse] fix embarassing bugfix
wsmoses Dec 17, 2023
2d0b3bb
move away from using deprecated StringRef API (#1593)
ftynse Dec 18, 2023
686a7bb
Better error messages for invalid erasure (#1592)
wsmoses Dec 18, 2023
7736245
update TypeAnalysis for recent LLVMs (#1594)
ftynse Dec 19, 2023
bad7df0
Add NoFree cmd line fallback (#1591)
wsmoses Dec 20, 2023
a6e8dc0
[mlir] support unknown (top) state in aliasing and activity analysis …
ftynse Jan 2, 2024
119f2f1
Fix printing bug #1599 (#1601)
maxaehle Jan 4, 2024
5422797
Mark floor as inactive (#1598)
wsmoses Jan 4, 2024
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Add more version bounds to max/min (#1566)
  • Loading branch information
wsmoses authored Dec 10, 2023
commit c27e7bc4b2ec7c9fee90936ee79f4cfc14a2bc4e
4 changes: 2 additions & 2 deletions enzyme/Enzyme/InstructionDerivatives.td
Original file line number Diff line number Diff line change
@@ -777,7 +777,7 @@ def : IntrPattern<(Op $x, $y),
>;

def : IntrPattern<(Op $x, $y),
[["minnum"], ["nvvm_fmin_f"], ["nvvm_fmin_d"], ["nvvm_fmin_ftz_f"], ["x86_sse_min_ss", "", "9"], ["x86_sse_min_ps", "", "9"], ["x86_sse2_min_pd", "", ""], ["minimum", "15", ""]],
[["minnum"], ["nvvm_fmin_f"], ["nvvm_fmin_d"], ["nvvm_fmin_ftz_f"], ["x86_sse_min_ss", "", "9"], ["x86_sse_min_ps", "", "9"], ["x86_sse2_min_pd", "", ""], ["minimum", "11", ""]],
[
(Select (FCmpOLT $x, $y), (DiffeRet), (ConstantFP<"0"> $x)),
(Select (FCmpOLT $x, $y), (ConstantFP<"0"> $x), (DiffeRet))
@@ -786,7 +786,7 @@ def : IntrPattern<(Op $x, $y),
>;

def : IntrPattern<(Op $x, $y),
[["maxnum"], ["nvvm_fmax_f"], ["nvvm_fmax_d"], ["nvvm_fmax_ftz_f"], ["x86_sse_max_ss", "", "9"], ["x86_sse_max_ps", "", "9"], ["x86_sse2_max_pd", "", ""], ["maximum", "15", ""]],
[["maxnum"], ["nvvm_fmax_f"], ["nvvm_fmax_d"], ["nvvm_fmax_ftz_f"], ["x86_sse_max_ss", "", "9"], ["x86_sse_max_ps", "", "9"], ["x86_sse2_max_pd", "", ""], ["maximum", "11", ""]],
[
(Select (FCmpOLT $x, $y), (ConstantFP<"0"> $x), (DiffeRet)),
(Select (FCmpOLT $x, $y), (DiffeRet), (ConstantFP<"0"> $x))
Loading