[Issue 16503] New: [ndslice] prevents fastmath LDC attribute
https://issues.dlang.org/show_bug.cgi?id=16503
Issue ID: 16503
Summary: [ndslice] prevents fastmath LDC attribute
Product: D
Version: D2
Hardware: All
OS: All
Status: NEW
Severity: major
Priority: P1
Component: phobos
Assignee: nob...@puremagic.com
Reporter: ilyayaroshe...@gmail.com
Slice
{
front
opIndex
ptr
...
}
auto sl1 = slice!double(...);
auto sl2 = slice!double(...);
@fastmath ... {
// loop for strides == 1
c += sl1.ptr[i] * sl2.ptr[i];
}
c's computation can not be fully optimized because Slice primitives has not
fastmath attribute.
In the same time this is safe TODO:
Slice
{
@fastmath:
front
opIndex
...
}
auto sl = slice!double(...);
... { // no fastmath
// loop for strides == 1
c += sl1.ptr[i] * sl2.ptr[i];
}
No optimisations in the last case.
So, we can just mark all ndslice except mapSlice as @fastmath :-)
--
[Issue 16502] New: spawnProcess does not throw on exec errors
https://issues.dlang.org/show_bug.cgi?id=16502
Issue ID: 16502
Summary: spawnProcess does not throw on exec errors
Product: D
Version: D2
Hardware: All
OS: Linux
Status: NEW
Severity: enhancement
Priority: P1
Component: phobos
Assignee: nob...@puremagic.com
Reporter: freeslav...@gmail.com
I chose Linux as OS for this issue, but it's legitimate for all Posix.
spawnProcess does not implement proper reporting of errors from exec.
This leads to the situations when execve fails, but spawnProcess does not
throw.
spawnProcess has some checks before fork, but these are not enough.
E.g. create empty file and mark it as executable
touch notreallyexecutable
chmod +x notreallyexecutable
Then write D program:
import std.process;
void main(string[] args)
{
spawnProcess(args[1]);
}
Compile and run:
dmd test.d
./test ./notreallyexecutable
It will not throw.
Error reporting can be implemented via pipe. Open pipe on parent side with
CLOEXEC. It will be inherited in fork. In case of exec errors write some code
to the pipe, e.g. errno code. On success pipe will be closed automatically.
On parent side read from pipe. In case of errors there will be errno code,
otherwise nothing.
--
[Issue 15341] segfault with std.signals slots
https://issues.dlang.org/show_bug.cgi?id=15341 --- Comment #3 from github-bugzi...@puremagic.com --- Commits pushed to master at https://github.com/dlang/phobos https://github.com/dlang/phobos/commit/978574006c8b3c79cdf08eaecac90e45d215fa9f Fixes issue 15341 - detachDisposeEvent called even with remaining connected slots for that instance https://github.com/dlang/phobos/commit/2ccfbcccab6f121681e4c6727389e469cfe850db Merge pull request #4441 from Dechcaudron/15341 Fixes issue 15341 --
[Issue 15341] segfault with std.signals slots
https://issues.dlang.org/show_bug.cgi?id=15341 github-bugzi...@puremagic.com changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --
[Issue 16501] New: packed ndslices does not compile
https://issues.dlang.org/show_bug.cgi?id=16501
Issue ID: 16501
Summary: packed ndslices does not compile
Product: D
Version: D2
Hardware: All
OS: All
Status: NEW
Severity: normal
Priority: P1
Component: phobos
Assignee: nob...@puremagic.com
Reporter: ilyayaroshe...@gmail.com
unittest
{
import std.experimental.ndslice;
alias sab = sliced!("a", "b");
auto sl = sab(new double[12], new double[12], 3, 4);
auto psl = sl.pack!1;
}
--
[Issue 15857] incorrect checkimports mismatch for overload sets
https://issues.dlang.org/show_bug.cgi?id=15857 --- Comment #14 from github-bugzi...@puremagic.com --- Commits pushed to stable at https://github.com/dlang/dmd https://github.com/dlang/dmd/commit/c5822394c89c98a8855302164433acf6db09208b fix Issue 15857 - incorrect checkimports mismatch for overloadsets - properly fix Issue 15857 by comparing overload sets for equality (see 082e1199c258e99f7867ca33509db9ae9a31f3ae or #5681 for the previous fix) - caching them in the local symbol table doesn't fully fix the issue b/c the table is not checked w/ SearchImportsOnly (see added test case) - also with -transition=checkimports adding an overload set to the local symbol table will return that overload set for locals only searches, even though it might contain imported overloads (see readded dip22e deprecation) - might still make sense to correctly cache overload sets at some point, but they shouldn't be that common https://github.com/dlang/dmd/commit/9ef10b73175ffb5ba6c638eef77bcdb243115026 Merge pull request #6136 from MartinNowak/fix15857 fix Issue 15857 - incorrect checkimports mismatch for overloadsets --
[Issue 16072] std.container.binaryheap should be extendable for arrays
https://issues.dlang.org/show_bug.cgi?id=16072 --- Comment #1 from github-bugzi...@puremagic.com --- Commits pushed to master at https://github.com/dlang/phobos https://github.com/dlang/phobos/commit/71f01c961de1e710250aee77e00c07839e7b68e3 fix issue 16072 - container.binaryheap should be extendable for arrays https://github.com/dlang/phobos/commit/6ca3017b9188caca204ed43dade7c851f86e6866 Merge pull request #4359 from wilzbach/fix_16072 fix issue 16072 - container.binaryheap should be extendable for arrays --
[Issue 3573] pure and nothrow allow function return type to be inferred
https://issues.dlang.org/show_bug.cgi?id=3573 Bolpat changed: What|Removed |Added CC||qs.il.paperi...@gmail.com --- Comment #3 from Bolpat --- Isn't 1. a feature? --
[Issue 16500] New: inferred != annotated pure
https://issues.dlang.org/show_bug.cgi?id=16500
Issue ID: 16500
Summary: inferred != annotated pure
Product: D
Version: D2
Hardware: All
OS: All
Status: NEW
Severity: normal
Priority: P1
Component: dmd
Assignee: nob...@puremagic.com
Reporter: qs.il.paperi...@gmail.com
struct X(T)
{
this(S)(in S[] arr) // inferred pure
{ }
}
void main() pure
{
X!uint mut = [ 1, 2 ]; // proves inference (cf. main is pure)
// immutable X!uint imm1 = [ 1, 2 ];
// auto imm2 = immutable X!uint([1, 2]);
}
The last lines don't compile. However, they do if the inferred pure is
explicitly annotated.
>From the discussion
https://forum.dlang.org/post/seuujamznrzqygrrp...@forum.dlang.org
--
[Issue 15857] incorrect checkimports mismatch for overload sets
https://issues.dlang.org/show_bug.cgi?id=15857 --- Comment #13 from Martin Nowak --- https://github.com/dlang/dmd/pull/6136 --
