[Issue 18223] New: std.experimental.allocator uninitializedFillDefault could use memset
https://issues.dlang.org/show_bug.cgi?id=18223
Issue ID: 18223
Summary: std.experimental.allocator uninitializedFillDefault
could use memset
Product: D
Version: D2
Hardware: All
OS: All
Status: NEW
Severity: enhancement
Priority: P1
Component: phobos
Assignee: nob...@puremagic.com
Reporter: n8sh.second...@hotmail.com
Current function in std.experimental.allocator.package:
```
private T[] uninitializedFillDefault(T)(T[] array) nothrow
{
T t = T.init;
fillWithMemcpy(array, t);
return array;
}
```
When we can statically determine that the representation of T.init consists of
nothing but zeroes we could instead use memset. char and wchar could also be
special-cased.
--
[Issue 18222] New: Illegal instruction for test_cdvecfill_1
https://issues.dlang.org/show_bug.cgi?id=18222 Issue ID: 18222 Summary: Illegal instruction for test_cdvecfill_1 Product: D Version: D2 Hardware: x86_64 OS: Linux Status: NEW Severity: normal Priority: P1 Component: dmd Assignee: nob...@puremagic.com Reporter: thomas.ma...@gmail.com I get the following when I run the dmd-testsuite of ldc 1.7.0 on NixOS: > 790: sh: line 1: 36610 Illegal instruction > /build/ldc-1.7.0-src/build/dmd-testsuite/runnable/test_cdvecfill_1 > > /build/ldc-1.7.0-src/build/dmd-testsuite/a0519aad7a20f86b87a15113f878a3d1815937849f5527e5ac5dbbdc24e22f05 > 2>&1 790: ... runnable/test_cdvecfill.d -O (-mcpu=avx -mcpu=avx2) -conf=/build/ldc-1.7.0-src/build/bin/ldc2.conf -O 790: Test failed. The logged output: 790: /build/ldc-1.7.0-src/build/bin/ldmd2 -conf= -m64 -Irunnable -O -od/build/ldc-1.7.0-src/build/dmd-testsuite/runnable -of/build/ldc-1.7.0-src/build/dmd-testsuite/runnable/test_cdvecfill_0 runnable/test_cdvecfill.d 790: /build/ldc-1.7.0-src/build/dmd-testsuite/runnable/test_cdvecfill_0 790: 790: /build/ldc-1.7.0-src/build/bin/ldmd2 -conf= -m64 -Irunnable -O -mcpu=avx -od/build/ldc-1.7.0-src/build/dmd-testsuite/runnable -of/build/ldc-1.7.0-src/build/dmd-testsuite/runnable/test_cdvecfill_1 runnable/test_cdvecfill.d 790: /build/ldc-1.7.0-src/build/dmd-testsuite/runnable/test_cdvecfill_1 790: 790: 790: == 790: Test failed: expected rc == 0, exited with rc == 132 See https://hydra.nixos.org/build/67414474/nixlog/1 for the full log. I built this package yesterday on my system without any problems though it only happens on these hydra builds. It works on MacOS X. --
[Issue 18188] rdmd ignores dependency in 'static this()'
https://issues.dlang.org/show_bug.cgi?id=18188 Seb changed: What|Removed |Added CC||greensunn...@gmail.com --- Comment #2 from Seb --- BTW are you aware of https://github.com/dlang/dmd/pull/7099 and https://github.com/dlang/tools/pull/271 ? It could as a side-effect fix your bug. --
[Issue 18188] rdmd ignores dependency in 'static this()'
https://issues.dlang.org/show_bug.cgi?id=18188 --- Comment #1 from Dragos Carp --- The problem is caused by `dmd -v` output, that is parsed and used by rdmd. $ dmd -v -o- foo.d predefs DigitalMars Posix linux ELFv1 LittleEndian D_Version2 all D_SIMD D_InlineAsm_X86_64 X86_64 CRuntime_Glibc D_LP64 D_PIC assert D_HardFloat binarydmd version v2.078.0 config/etc/dmd.conf parse foo importall foo importobject(/usr/include/dmd/druntime/import/object.d) semantic foo entry main foo.d semantic2 foo semantic3 foo importbar (bar.d) Calling dmd with -deps option fixes the problem: $ dmd -v -o- foo.d -deps=/dev/null predefs DigitalMars Posix linux ELFv1 LittleEndian D_Version2 all D_SIMD D_InlineAsm_X86_64 X86_64 CRuntime_Glibc D_LP64 D_PIC assert D_HardFloat binarydmd version v2.078.0 config/etc/dmd.conf parse foo importall foo importobject(/usr/include/dmd/druntime/import/object.d) semantic foo entry main foo.d semantic2 foo semantic3 foo importbar (bar.d) importbaz (baz.d) Consequently rdmd command also functions when -deps option is added: $ rdmd --build-only --chatty -deps=/dev/null foo.d mkdirRecurse /tmp/.rdmd-500 mkdirRecurse /tmp/.rdmd-500/rdmd-foo.d-2D7BE73D8EB3145BDB8A563EF51C3756 lock /tmp/.rdmd-500/rdmd-foo.d-2D7BE73D8EB3145BDB8A563EF51C3756/rdmd.lock mkdirRecurse /tmp/.rdmd-500/rdmd-foo.d-2D7BE73D8EB3145BDB8A563EF51C3756/objs stat /tmp/.rdmd-500/rdmd-foo.d-2D7BE73D8EB3145BDB8A563EF51C3756/rdmd.deps spawn ["/usr/bin/dmd", "-deps=/dev/null", "-v", "-o-", "foo.d", "-I."] read /tmp/.rdmd-500/rdmd-foo.d-2D7BE73D8EB3145BDB8A563EF51C3756/rdmd.deps which /usr/bin/dmd stat /etc/dmd.conf stat /tmp/.rdmd-500/rdmd-foo.d-2D7BE73D8EB3145BDB8A563EF51C3756/.built stat ./foo stat foo.d stat foo.d stat /etc/dmd.conf stat baz.d stat bar.d stat /usr/bin/dmd stat ./foo spawn ["/usr/bin/dmd", "-deps=/dev/null", "-of./foo.tmp", "-od/tmp/.rdmd-500/rdmd-foo.d-2D7BE73D8EB3145BDB8A563EF51C3756/objs", "-I.", "foo.d", "baz.d", "bar.d"] stat /tmp/.rdmd-500/rdmd-foo.d-2D7BE73D8EB3145BDB8A563EF51C3756/objs rmdirRecurse /tmp/.rdmd-500/rdmd-foo.d-2D7BE73D8EB3145BDB8A563EF51C3756/objs mv ./foo.tmp ./foo touch /tmp/.rdmd-500/rdmd-foo.d-2D7BE73D8EB3145BDB8A563EF51C3756/.built --
[Issue 18221] New: DMD64 2.078.0 compile time explodes with -inline for mir-algorithm
https://issues.dlang.org/show_bug.cgi?id=18221 Issue ID: 18221 Summary: DMD64 2.078.0 compile time explodes with -inline for mir-algorithm Product: D Version: D2 Hardware: x86 OS: All Status: NEW Severity: regression Priority: P1 Component: dmd Assignee: nob...@puremagic.com Reporter: timothee.co...@gmail.com details: http://forum.dlang.org/post/gevcbtueeiscqipqm...@forum.dlang.org as reported by Nathan S. --
[Issue 17819] static foreach segfaults on __traits(allMembers)
https://issues.dlang.org/show_bug.cgi?id=17819 Jean-Louis Leroy changed: What|Removed |Added CC||j...@leroy.nyc --
[Issue 17819] static foreach segfaults on __traits(allMembers)
https://issues.dlang.org/show_bug.cgi?id=17819 --- Comment #1 from Jean-Louis Leroy --- The bug is still present in 2.078. --
[Issue 18220] Allow rt_trapexceptions to be set from the CLI
https://issues.dlang.org/show_bug.cgi?id=18220 Seb changed: What|Removed |Added Summary|Allow trapexception to be |Allow rt_trapexceptions to |set from the CLI|be set from the CLI --
[Issue 18220] New: Allow trapexception to be set from the CLI
https://issues.dlang.org/show_bug.cgi?id=18220 Issue ID: 18220 Summary: Allow trapexception to be set from the CLI Product: D Version: D2 Hardware: x86_64 OS: Linux Status: NEW Severity: enhancement Priority: P1 Component: druntime Assignee: nob...@puremagic.com Reporter: greensunn...@gmail.com http://arsdnet.net/this-week-in-d/2016-aug-07.html > Hopefully, druntime will expose this variable in a more easier way in the > future. Currently, on Windows, it is automatically set if a debugger is > detected, but on Linux, debugger detection is quite a bit harder and hackier, > so it might not happen any time soon. Druntime could (and should!) also > potentially offer it as a command line switch, like it does with GC profiling > now. > DRUNTIME DEVS: if you read this, make it happen! --
[Issue 13632] Second argument for std.string.strip
https://issues.dlang.org/show_bug.cgi?id=13632 --- Comment #2 from Aravinda --- (In reply to Seb from comment #1) > @Aravinda VK: thanks a lot for taking this. Please don't forget to add a > link to your PR here. Otherwise people in the future will have a hard time > finding or even being aware of it: > > https://github.com/dlang/phobos/pull/6023 Noted. This is my first patch/bug in D lang community. I will make sure to add PR link in future. Thanks. --
[Issue 13632] Second argument for std.string.strip
https://issues.dlang.org/show_bug.cgi?id=13632 Seb changed: What|Removed |Added Keywords||pull CC||greensunn...@gmail.com --- Comment #1 from Seb --- @Aravinda VK: thanks a lot for taking this. Please don't forget to add a link to your PR here. Otherwise people in the future will have a hard time finding or even being aware of it: https://github.com/dlang/phobos/pull/6023 --
[Issue 18219] Private import inside struct leaks symbols when used as VarDeclaration types
https://issues.dlang.org/show_bug.cgi?id=18219 --- Comment #1 from RazvanN --- PR: https://github.com/dlang/dmd/pull/7668 --
[Issue 17961] std.uni does not compile with -unittest -dip1000
https://issues.dlang.org/show_bug.cgi?id=17961
Carsten Blüggel changed:
What|Removed |Added
CC||chi...@posteo.net
--- Comment #7 from Carsten Blüggel ---
Although -dip1000 is not ready, it's complaint seems to be justified in at
least 1 of 2 spots here:
The complaints originate in the function immediately preceding the reported
error line: ref intersect()(dchar ch)
(1) There are 2 assignments to this: this = This.init...
that raise the error, IMO legitimately: reference to local variable __tmp...
assigned to non-scope parameter this.
This could be calmed down using instead: data.length(0); I hope it's an
equivalent for This.init in this case.
(2) The other one (same error msg) stems from this.byInterval in the foreach
loop and seems to be questionable.
In the end, this works with -dip1000 enabled:
ref intersect()(dchar ch)
{
scope arr = data;
foreach (i; Intervals!(typeof(data))(arr))
if (i.a <= ch && ch < i.b)
{
data.length(0);
return this.add(ch, ch+1);
}
data.length(0);
return this;
}
If there are no objections, I'll file a PR Fix issue 17961 - ...
--
[Issue 13632] Second argument for std.string.strip
https://issues.dlang.org/show_bug.cgi?id=13632 Aravinda changed: What|Removed |Added Status|NEW |ASSIGNED CC||hallimanearav...@gmail.com Assignee|nob...@puremagic.com|hallimanearav...@gmail.com --
[Issue 18219] New: Private import inside struct leaks symbols when used as VarDeclaration types
https://issues.dlang.org/show_bug.cgi?id=18219
Issue ID: 18219
Summary: Private import inside struct leaks symbols when used
as VarDeclaration types
Product: D
Version: D2
Hardware: x86_64
OS: Linux
Status: NEW
Severity: normal
Priority: P1
Component: dmd
Assignee: nob...@puremagic.com
Reporter: razvan.nitu1...@gmail.com
// a.d
class Foobar
{
int a;
this(int a)
{
this.a = a;
}
static int smeth()
{
return 1;
}
}
void fun() {}
// b.d
struct AST
{
import a;
}
// c.d
void main()
{
import b;
AST.Foobar t;// compiles
AST.Foobar.smeth(); // does not compile
AST.fun(); // does not compile
}
`AST.Foobar t` should not compile.
--
