From 832ce4b3c23c24275c979c951bd444a28b032ef7 Mon Sep 17 00:00:00 2001 From: Jared Hoberock Date: Fri, 17 Nov 2017 15:11:53 -0600 Subject: [PATCH 1/3] Apply P0075R2 --- algorithms.html | 360 ++++++++++++++++++++++++++++++++++++++++++++++++ general.html | 8 ++ 2 files changed, 368 insertions(+) diff --git a/algorithms.html b/algorithms.html index 275c08a..554032f 100644 --- a/algorithms.html +++ b/algorithms.html @@ -548,6 +548,64 @@

Header <experimental/algorithm> synopsis

template<class T> ordered_update_t<T> ordered_update(T& ref) noexcept; } + + +// Exposition only: Suppress template argument deduction. +template<class T> struct no_deduce { using type = T; }; +template<class T> struct no_dedude_t = typename no_deduce<T>::type; + + Support for reductions +template<class T, class BinaryOperation> + unspecified reduction(T& var, const T& identity, BinaryOperation combiner); +template<class T> + unspecified reduction_plus(T& var); +template<class T> + unspecified reduction_multiplies(T& var); +template<class T> + unspecified reduction_bit_and(T& var); +template<class T> + unspecified reduction_bit_or(T& var); +template<class T> + unspecified reduction_bit_xor(T& var); +template<class T> + unspecified reduction_min(T& var); +template<class T> + unspecified reduction_max(T& var); + + Support for inductions +template<class T> + unspecified induction(T&& var); +template<class T> + unspecified induction(T&& var, S stride); + + for_loop +template<class I, class... Rest> + void for_loop(no_deduce_t<I> start, I finish, Rest&&... rest); +template<class ExecutionPolicy, + class I, class... Rest> + void for_loop(ExecutionPolicy&& exec, + no_deduce_t<I> start, I finish, Rest&&... rest); +template<class I, class S, class... Rest> + void for_loop_strided(no_deduce_t<I> start, I finish, + S stride, Rest&&... rest); +template<class ExecutionPolicy, + class I, class S, class... Rest> + void for_loop_strided(ExecutionPolicy&& exec, + no_deduce_t<I> start, I finish, + S stride, Rest&&... rest); +template<class I, class Size, class... Rest> + void for_loop_n(I start, Size n, Rest&&... rest); +template<class ExecutionPolicy, + class I, class Size, class... Rest> + void for_loop_n(ExecutionPolicy&& exec, + I start, Size n, Rest&&... rest); +template<class I, class Size, class S, class... Rest> + void for_loop_n_strided(I start, Size n, S stride, Rest&&... rest); +template<class ExecutionPolicy, + class I, class Size, class S, class... Rest> + void for_loop_n_strided(ExecutionPolicy&& exec, + I start, Size n, S stride, Rest&&... rest); + } } } @@ -555,6 +613,308 @@

Header <experimental/algorithm> synopsis

+ +

Reductions

+ + +

+ Each of the function templates in this subclause ([parallel.alg.reductions]) returns a reduction object + of unspecified type having a reduction value type and encapsulating a reduction identity value for the reduction, a + combiner function object, and a live-out object from which the initial value is obtained and into which the final + value is stored. +

+ +

+ An algorithm uses reduction objects by allocating an unspecified number of instances, known as accumulators, of the reduction value + type. An implementation might, for example, allocate an accumulator for each thread in its private thread pool. + Each accumulator is initialized with the object’s reduction identity, except that the live-out object (which was initialized by the + caller) comprises one of the accumulators. The algorithm passes a reference to an accumulator to each application of an element-access + function, ensuring that no two concurrently executing invocations share the same accumulator. An accumulator can be shared between two + applications that do not execute concurrently, but initialization is performed only once per accumulator. +

+ +

+ Modifications to the accumulator by the application of element access functions accrue as partial results. At some point before the algorithm + returns, the partial results are combined, two at a time, using the reduction object’s combiner operation until a single value remains, which + is then assigned back to the live-out object. in order to produce useful results, modifications to the accumulator should be limited + to commutative operations closely related to the combiner operation. For example if the combiner is plus<T>, incrementing + the accumulator would be consistent with the combiner but doubling it or assigning to it would not. +

+ + + + template<class T, class BinaryOperation> +unspecified reduction(T& var, const T& identity, BinaryOperation combiner); + + + T shall meet the requirements of CopyConstructible and MoveAssignable. The expression var = combiner(var, var) shall be well-formed. + + + + a reduction object of unspecified type having reduction value type T, reduction identity identity, combiner function object combiner, and using the object referenced by var as its live-out object. + + + + + template<class T> +unspecified reduction_plus(T& var); + template<class T> +unspecified reduction_multiplies(T& var); + template<class T> +unspecified reduction_bit_and(T& var); + template<class T> +unspecified reduction_bit_or(T& var); + template<class T> +unspecified reduction_bit_xor(T& var); + template<class T> +unspecified reduction_min(T& var); + template<class T> +unspecified reduction_max(T& var); + + + T shall meet the requirements of CopyConstructible and MoveAssignable. + + + + a reduction object of unspecified type having reduction value type T, reduction identity and combiner operation as specified in table and using the object referenced by var as its live-out object. + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Reduction identities and combiner operations
FunctionReduction IdentityCombiner Operation
reduction_plusT()x + y
reduction_multipliesT(1)x * y
reduction_bit_and(~T())X & y
reduction_bit_orT()x | y
reduction_bit_xorT()x ^ y
reduction_minvarmin(x, y)
reduction_maxvarmax(x, y)
+ + + The following code updates each element of y and sets s ot the sum of the squares. +
+extern int n;
+extern float x[], y[], a;
+float s = 0;
+for_loop(execution::vec, 0, n,
+    reduction(s, 0.0f, plus<>()),
+    [&](int i, float& accum) {
+            y[i] += a*x[i];
+            accum += y[i]*y[i];
+    }
+);
+
+ + + + + + +

Inductions

+ + +

+ Each of the function templates in this section return an induction object of unspecified type having an induction + value type and encapsulating an initial value i of that type and, optionally, a stride. +

+ +

+ For each element in the input range, an algorithm over input sequence S computes an induction value from an induction variable + and ordinal position p within S by the formula i + p * stride if a stride was specified or i + p otherwise. This induction value is + passed to the element access function. +

+ +

+ An induction object may refer to a live-out object to hold the final value of the induction sequence. When the algorithm using the induction + object completes, the live-out object is assigned the value i + n * stride, where n is the number of elements in the input range. +

+ + + + template<class T> +unspecified induction(T&& var); + + template<class T, class S> +unspecified induction(T&& var, S stride); + + + + + an induction object with induction value type remove_cv_t>remove_reference_t>T<<, + initial value var, and (if specified) stride stride. If T is an lvalue reference + to non-const type, then the object referenced by var becomes the live-out object for the + induction object; otherwise there is no live-out object. + + + + + + + +

For loop

+ + + template<class I, class... Rest> +void for_loop(no_deduce_t<I> start, I finish, Rest&&... rest); + + template<class ExecutionPolicy, + class I, class... Rest> +void for_loop(ExecutionPolicy&& exec, + no_deduce_t<I> start, I finish, Rest&&... rest); + + + + template<class I, class S, class... Rest> +void for_loop_strided(no_deduce_t<I> start, I finish, + S stride, Rest&&... rest); + + template<class ExecutionPolicy, + class I, class S, class... Rest> +void for_loop_strided(ExecutionPolicy&& exec, + no_deduce_t<I> start, I finish, + S stride, Rest&&... rest); + + + + template<class I, class Size, class... Rest> +void for_loop_n(I start, Size n, Rest&&... rest); + + template<class ExecutionPolicy, + class I, class Size, class... Rest> +void for_loop_n(ExecutionPolicy&& exec, + I start, Size n, Rest&&... rest); + + + + template<class I, class Size, class S, class... Rest> +void for_loop_n_strided(I start, Size n, S stride, Rest&&... rest); + + template<class ExecutionPolicy, + class I, class Size, class S, class... Rest> +void for_loop_n_strided(ExecutionPolicy&& exec, + I start, Size n, S stride, Rest&&... rest); + + + + + For the overloads with an ExecutionPolicy, I shall be an integral type + or meet the requirements of a forward iterator type; otherwise, I shall be an integral + type or meet the requirements of an input iterator type. Size shall be an integral type + and n shall be non-negative. S shall have integral type and stride + shall have non-zero value. stride shall be negative only if I has integral + type or meets the requirements of a bidirectional iterator. The rest parameter pack shall + have at least one element, comprising objects returned by invocations of reduction + ([parallel.alg.reduction]) and/or induction ([parallel.alg.induction]) function templates + followed by exactly one invocable element-access function, f. For the overloads with an + ExecutionPolicy, f shall meet the requirements of CopyConstructible; + otherwise, f shall meet the requirements of MoveConstructible. + + + + + + + + Applies f to each element in the input sequence, as described below, with additional + arguments corresponding to the reductions and inductions in the rest parameter pack. The + length of the input sequence is: + +
    +
  • + n, if specified, +
    • + +
  • + otherwise finish - start if neither n nor stride is specified, +
    • + +
  • + otherwise 1 + (finish-start-1)/stride if stride is positive, +
    • + +
  • + otherwise 1 + (start-finish-1)/-stride. +
    • +
+ + The first element in the input sequence is start. Each subsequent element is generated by adding + stride to the previous element, if stride is specified, otherwise by incrementing + the previous element. As described in the C++ standard, section [algorithms.general], arithmetic + on non-random-access iterators is performed using advance and distance. The order of the + elements of the input sequence is important for determining ordinal position of an application of f, + even though the applications themselves may be unordered.

+ + The first argument to f is an element from the input sequence. if I is an + iterator type, the iterators in the input sequence are not dereferenced before + being passed to f. For each member of the rest parameter pack + excluding f, an additional argument is passed to each application of f as follows: + +
    +
  • + If the pack member is an object returned by a call to a reduction function listed in section + [parallel.alg.reductions], then the additional argument is a reference to an accumulator of that reduction + object. +
    • + +
  • + If the pack member is an object returned by a call to induction, then the additional argument is the + induction value for that induction object corresponding to the position of the application of f in the input + sequence. +
    • +
+ + + + + + + Applies f exactly once for each element of the input sequence. + + + + + + If f returns a result, the result is ignored. + + + + +

For each

diff --git a/general.html b/general.html index 6c82000..bda89ac 100644 --- a/general.html +++ b/general.html @@ -80,6 +80,14 @@

Feature-testing recommendations

<experimental/execution>
+ + P0075R2 + Template Library for Parallel For Loops + , , + __cpp_lib_experimental_parallel_for_loop + 201711 + <experimental/algorithm> + From 02367c4680eee86c865df0c6cfc13c7cf45b46fc Mon Sep 17 00:00:00 2001 From: Jared Hoberock Date: Fri, 17 Nov 2017 15:26:11 -0600 Subject: [PATCH 2/3] Update README for post-Albuquerque mailing --- README.md | 20 ++++++++------------ 1 file changed, 8 insertions(+), 12 deletions(-) diff --git a/README.md b/README.md index 9e1e52a..006bfb5 100644 --- a/README.md +++ b/README.md @@ -1,5 +1,5 @@ - Document Number: N4699 - Date: 2017-10-16 + Document Number: N4XXX+1 + Date: 2017-11-XX Revises: Project: Programming Language C++ Project Number: TS 19570 @@ -9,20 +9,16 @@ # Parallelism TS Editor's Report, pre-Albuquerque mailing -N4698 is the proposed working draft of Parallelism TS Version 2. It contains changes to the Parallelism TS as directed by the committee at the Toronto meeting, and editorial changes. +NXXXX is the proposed working draft of Parallelism TS Version 2. It contains changes to the Parallelism TS as directed by the committee at the Albuquerque meeting. -N4698 updates the previous draft, N4669, published in the pre-Toronto mailing. +NXXXX updates the previous draft, N4696, published in the pre-Toronto mailing. # Technical Changes -* Apply P0076R4 - Vector and Wavefront Policies. +* Apply P0776R1 - Rebase the Parallelism TS onto the C++17 Standard +* Apply P0075R2 - Template Library for Parallel For Loops -# Editorial Changes +# Acknowledgements -* Reformat Table 1 - Feature Test Macro(s), to match the style of the Library Fundamentals TS. - -# Notes - -* The pre-existing content of N4698 has not yet been harmonized with C++17. As a result, this content is named and namespaced inconsistently with the newly applied content of P0076R4. We anticipate that these inconsistencies will be harmonized by a future revision. -* N4698 contains forward references to `for_loop` and `for_loop_strided`. We anticipate their introduction in a future revision. +Thanks to Alisdair Meredith and Pablo Halpern for reviewing these changes. From 255967e557c54faf090318dd5c46a88e10098003 Mon Sep 17 00:00:00 2001 From: Jared Hoberock Date: Mon, 20 Nov 2017 14:16:22 -0600 Subject: [PATCH 3/3] Add paper numbers for post-Albuquerque mailing --- README.md | 6 +++--- front_matter.html | 4 ++-- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/README.md b/README.md index 006bfb5..5b5c497 100644 --- a/README.md +++ b/README.md @@ -1,4 +1,4 @@ - Document Number: N4XXX+1 + Document Number: N4707 Date: 2017-11-XX Revises: Project: Programming Language C++ @@ -9,9 +9,9 @@ # Parallelism TS Editor's Report, pre-Albuquerque mailing -NXXXX is the proposed working draft of Parallelism TS Version 2. It contains changes to the Parallelism TS as directed by the committee at the Albuquerque meeting. +N4706 is the proposed working draft of Parallelism TS Version 2. It contains changes to the Parallelism TS as directed by the committee at the Albuquerque meeting. -NXXXX updates the previous draft, N4696, published in the pre-Toronto mailing. +N4706 updates the previous draft, N4696, published in the pre-Toronto mailing. # Technical Changes diff --git a/front_matter.html b/front_matter.html index 9a1ad15..5498f05 100644 --- a/front_matter.html +++ b/front_matter.html @@ -1,8 +1,8 @@ -N4698 +N4706 19570 - N4669 + N4698 Jared Hoberock
NVIDIA Corporation