Add experimental::sample and experimental::shuffle from N4531

Message ID 20180625162338.GA15218@redhat.com
State New
Headers show
Series
  • Add experimental::sample and experimental::shuffle from N4531
Related show

Commit Message

Jonathan Wakely June 25, 2018, 4:23 p.m.
The additions to <experimental/random> were added in 2015 but the new
algorithms in <experimental/algorithm> were not. This adds them.

	* include/experimental/algorithm (sample, shuffle): Add new overloads
	using per-thread random number engine.
	* testsuite/experimental/algorithm/sample.cc: Simpify and reduce
	dependencies by using __gnu_test::test_container.
	* testsuite/experimental/algorithm/sample-2.cc: New.
	* testsuite/experimental/algorithm/shuffle.cc: New.

Tested x86_64-linux, committed to trunk.

This would be safe to backport, but nobody has noticed the algos are
missing or complained, so it doesn't seem very important to backport.
commit 3dd31954a53f74b1faa1b5a6dcb0b3d355738931
Author: Jonathan Wakely <jwakely@redhat.com>
Date:   Mon Jun 25 16:41:13 2018 +0100

    Add experimental::sample and experimental::shuffle from N4531
    
    The additions to <experimental/random> were added in 2015 but the new
    algorithms in <experimental/algorithm> were not. This adds them.
    
            * include/experimental/algorithm (sample, shuffle): Add new overloads
            using per-thread random number engine.
            * testsuite/experimental/algorithm/sample.cc: Simpify and reduce
            dependencies by using __gnu_test::test_container.
            * testsuite/experimental/algorithm/sample-2.cc: New.
            * testsuite/experimental/algorithm/shuffle.cc: New.

Comments

Jonathan Wakely June 25, 2018, 4:42 p.m. | #1
On 25/06/18 17:23 +0100, Jonathan Wakely wrote:
>The additions to <experimental/random> were added in 2015 but the new

>algorithms in <experimental/algorithm> were not. This adds them.

>

>	* include/experimental/algorithm (sample, shuffle): Add new overloads

>	using per-thread random number engine.

>	* testsuite/experimental/algorithm/sample.cc: Simpify and reduce

>	dependencies by using __gnu_test::test_container.

>	* testsuite/experimental/algorithm/sample-2.cc: New.

>	* testsuite/experimental/algorithm/shuffle.cc: New.

>

>Tested x86_64-linux, committed to trunk.


And this documents it in the manual.
commit 0e350380ee69c6b719362fbd9fbb6a6d0854f6ec
Author: Jonathan Wakely <jwakely@redhat.com>
Date:   Mon Jun 25 17:42:09 2018 +0100

            * doc/xml/manual/status_cxx2017.xml: Document N4531 status.

diff --git a/libstdc++-v3/doc/xml/manual/status_cxx2017.xml b/libstdc++-v3/doc/xml/manual/status_cxx2017.xml
index aa0914cff72..a77653a3ab4 100644
--- a/libstdc++-v3/doc/xml/manual/status_cxx2017.xml
+++ b/libstdc++-v3/doc/xml/manual/status_cxx2017.xml
@@ -954,6 +954,16 @@ and test for <code>__STDCPP_MATH_SPEC_FUNCS__ >= 201003L</code>.
       <entry>Library Fundamentals 2 TS</entry>
     </row>
 
+    <row>
+      <entry>
+	<link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2015/n4531.html">
+	  N4531
+	</link>
+      </entry>
+      <entry>std::rand replacement, revision 3</entry>
+      <entry>Y</entry>
+      <entry>Library Fundamentals 2 TS</entry>
+    </row>
 
   </tbody>
 </tgroup>
Christophe Lyon June 29, 2018, 6:55 a.m. | #2
On Mon, 25 Jun 2018 at 18:23, Jonathan Wakely <jwakely@redhat.com> wrote:
>

> The additions to <experimental/random> were added in 2015 but the new

> algorithms in <experimental/algorithm> were not. This adds them.

>

>         * include/experimental/algorithm (sample, shuffle): Add new overloads

>         using per-thread random number engine.

>         * testsuite/experimental/algorithm/sample.cc: Simpify and reduce

>         dependencies by using __gnu_test::test_container.

>         * testsuite/experimental/algorithm/sample-2.cc: New.

>         * testsuite/experimental/algorithm/shuffle.cc: New.

>

> Tested x86_64-linux, committed to trunk.

>

> This would be safe to backport, but nobody has noticed the algos are

> missing or complained, so it doesn't seem very important to backport.

>

>


Hi,

On bare-metal targets (aarch64 and arm + newlib), I've noticed that
the two new tests fail:
PASS: experimental/algorithm/shuffle.cc (test for excess errors)
spawn /aci-gcc-fsf/builds/gcc-fsf-gccsrc/obj-arm-none-eabi/gcc3/utils/bin/qemu-wrapper.sh
./shuffle.exe
terminate called after throwing an instance of 'std::runtime_error'
  what():  random_device::random_device(const std::string&)

*** EXIT code 4242
FAIL: experimental/algorithm/shuffle.cc execution test

PASS: experimental/algorithm/sample-2.cc (test for excess errors)
spawn /aci-gcc-fsf/builds/gcc-fsf-gccsrc/obj-arm-none-eabi/gcc3/utils/bin/qemu-wrapper.sh
./sample-2.exe
terminate called after throwing an instance of 'std::runtime_error'
  what():  random_device::random_device(const std::string&)

*** EXIT code 4242
FAIL: experimental/algorithm/sample-2.cc execution test

Does this ring a bell?

Thanks,

Christophe
Jonathan Wakely June 29, 2018, 7:21 a.m. | #3
On 29/06/18 08:55 +0200, Christophe Lyon wrote:
>On Mon, 25 Jun 2018 at 18:23, Jonathan Wakely <jwakely@redhat.com> wrote:

>>

>> The additions to <experimental/random> were added in 2015 but the new

>> algorithms in <experimental/algorithm> were not. This adds them.

>>

>>         * include/experimental/algorithm (sample, shuffle): Add new overloads

>>         using per-thread random number engine.

>>         * testsuite/experimental/algorithm/sample.cc: Simpify and reduce

>>         dependencies by using __gnu_test::test_container.

>>         * testsuite/experimental/algorithm/sample-2.cc: New.

>>         * testsuite/experimental/algorithm/shuffle.cc: New.

>>

>> Tested x86_64-linux, committed to trunk.

>>

>> This would be safe to backport, but nobody has noticed the algos are

>> missing or complained, so it doesn't seem very important to backport.

>>

>>

>

>Hi,

>

>On bare-metal targets (aarch64 and arm + newlib), I've noticed that

>the two new tests fail:

>PASS: experimental/algorithm/shuffle.cc (test for excess errors)

>spawn /aci-gcc-fsf/builds/gcc-fsf-gccsrc/obj-arm-none-eabi/gcc3/utils/bin/qemu-wrapper.sh

>./shuffle.exe

>terminate called after throwing an instance of 'std::runtime_error'

>  what():  random_device::random_device(const std::string&)

>

>*** EXIT code 4242

>FAIL: experimental/algorithm/shuffle.cc execution test

>

>PASS: experimental/algorithm/sample-2.cc (test for excess errors)

>spawn /aci-gcc-fsf/builds/gcc-fsf-gccsrc/obj-arm-none-eabi/gcc3/utils/bin/qemu-wrapper.sh

>./sample-2.exe

>terminate called after throwing an instance of 'std::runtime_error'

>  what():  random_device::random_device(const std::string&)

>

>*** EXIT code 4242

>FAIL: experimental/algorithm/sample-2.cc execution test

>

>Does this ring a bell?


Does the existing testsuite/experimental/random/randint.cc file fail
in the same way?
Christophe Lyon June 29, 2018, 7:39 a.m. | #4
On Fri, 29 Jun 2018 at 09:21, Jonathan Wakely <jwakely@redhat.com> wrote:
>

> On 29/06/18 08:55 +0200, Christophe Lyon wrote:

> >On Mon, 25 Jun 2018 at 18:23, Jonathan Wakely <jwakely@redhat.com> wrote:

> >>

> >> The additions to <experimental/random> were added in 2015 but the new

> >> algorithms in <experimental/algorithm> were not. This adds them.

> >>

> >>         * include/experimental/algorithm (sample, shuffle): Add new overloads

> >>         using per-thread random number engine.

> >>         * testsuite/experimental/algorithm/sample.cc: Simpify and reduce

> >>         dependencies by using __gnu_test::test_container.

> >>         * testsuite/experimental/algorithm/sample-2.cc: New.

> >>         * testsuite/experimental/algorithm/shuffle.cc: New.

> >>

> >> Tested x86_64-linux, committed to trunk.

> >>

> >> This would be safe to backport, but nobody has noticed the algos are

> >> missing or complained, so it doesn't seem very important to backport.

> >>

> >>

> >

> >Hi,

> >

> >On bare-metal targets (aarch64 and arm + newlib), I've noticed that

> >the two new tests fail:

> >PASS: experimental/algorithm/shuffle.cc (test for excess errors)

> >spawn /aci-gcc-fsf/builds/gcc-fsf-gccsrc/obj-arm-none-eabi/gcc3/utils/bin/qemu-wrapper.sh

> >./shuffle.exe

> >terminate called after throwing an instance of 'std::runtime_error'

> >  what():  random_device::random_device(const std::string&)

> >

> >*** EXIT code 4242

> >FAIL: experimental/algorithm/shuffle.cc execution test

> >

> >PASS: experimental/algorithm/sample-2.cc (test for excess errors)

> >spawn /aci-gcc-fsf/builds/gcc-fsf-gccsrc/obj-arm-none-eabi/gcc3/utils/bin/qemu-wrapper.sh

> >./sample-2.exe

> >terminate called after throwing an instance of 'std::runtime_error'

> >  what():  random_device::random_device(const std::string&)

> >

> >*** EXIT code 4242

> >FAIL: experimental/algorithm/sample-2.cc execution test

> >

> >Does this ring a bell?

>

> Does the existing testsuite/experimental/random/randint.cc file fail

> in the same way?

>


Yes it does.

And so do:
25_algorithms/make_heap/complexity.cc
23_containers/array/element_access/at_neg.cc
26_numerics/random/random_device/cons/default.cc
Jonathan Wakely June 29, 2018, 9:45 a.m. | #5
On 29/06/18 09:39 +0200, Christophe Lyon wrote:
>On Fri, 29 Jun 2018 at 09:21, Jonathan Wakely <jwakely@redhat.com> wrote:

>>

>> On 29/06/18 08:55 +0200, Christophe Lyon wrote:

>> >On Mon, 25 Jun 2018 at 18:23, Jonathan Wakely <jwakely@redhat.com> wrote:

>> >>

>> >> The additions to <experimental/random> were added in 2015 but the new

>> >> algorithms in <experimental/algorithm> were not. This adds them.

>> >>

>> >>         * include/experimental/algorithm (sample, shuffle): Add new overloads

>> >>         using per-thread random number engine.

>> >>         * testsuite/experimental/algorithm/sample.cc: Simpify and reduce

>> >>         dependencies by using __gnu_test::test_container.

>> >>         * testsuite/experimental/algorithm/sample-2.cc: New.

>> >>         * testsuite/experimental/algorithm/shuffle.cc: New.

>> >>

>> >> Tested x86_64-linux, committed to trunk.

>> >>

>> >> This would be safe to backport, but nobody has noticed the algos are

>> >> missing or complained, so it doesn't seem very important to backport.

>> >>

>> >>

>> >

>> >Hi,

>> >

>> >On bare-metal targets (aarch64 and arm + newlib), I've noticed that

>> >the two new tests fail:

>> >PASS: experimental/algorithm/shuffle.cc (test for excess errors)

>> >spawn /aci-gcc-fsf/builds/gcc-fsf-gccsrc/obj-arm-none-eabi/gcc3/utils/bin/qemu-wrapper.sh

>> >./shuffle.exe

>> >terminate called after throwing an instance of 'std::runtime_error'

>> >  what():  random_device::random_device(const std::string&)

>> >

>> >*** EXIT code 4242

>> >FAIL: experimental/algorithm/shuffle.cc execution test

>> >

>> >PASS: experimental/algorithm/sample-2.cc (test for excess errors)

>> >spawn /aci-gcc-fsf/builds/gcc-fsf-gccsrc/obj-arm-none-eabi/gcc3/utils/bin/qemu-wrapper.sh

>> >./sample-2.exe

>> >terminate called after throwing an instance of 'std::runtime_error'

>> >  what():  random_device::random_device(const std::string&)

>> >

>> >*** EXIT code 4242

>> >FAIL: experimental/algorithm/sample-2.cc execution test

>> >

>> >Does this ring a bell?

>>

>> Does the existing testsuite/experimental/random/randint.cc file fail

>> in the same way?

>>

>

>Yes it does.

>

>And so do:

>25_algorithms/make_heap/complexity.cc


This one also uses std::random_device.

>23_containers/array/element_access/at_neg.cc


Hmm,

  // Expected behavior is to either throw and have the uncaught
  // exception end up in a terminate handler which eventually exits,
  // or abort. (Depending on -fno-exceptions.)

So this is expected to XFAIL.

>26_numerics/random/random_device/cons/default.cc


We should XFAIL the ones that use std::random_device, if we can
identify an effective target to describe them.
Jonathan Wakely July 4, 2018, 4:56 p.m. | #6
On 29/06/18 10:45 +0100, Jonathan Wakely wrote:
>On 29/06/18 09:39 +0200, Christophe Lyon wrote:

>>On Fri, 29 Jun 2018 at 09:21, Jonathan Wakely <jwakely@redhat.com> wrote:

>>>

>>>On 29/06/18 08:55 +0200, Christophe Lyon wrote:

>>>>On Mon, 25 Jun 2018 at 18:23, Jonathan Wakely <jwakely@redhat.com> wrote:

>>>>>

>>>>> The additions to <experimental/random> were added in 2015 but the new

>>>>> algorithms in <experimental/algorithm> were not. This adds them.

>>>>>

>>>>>         * include/experimental/algorithm (sample, shuffle): Add new overloads

>>>>>         using per-thread random number engine.

>>>>>         * testsuite/experimental/algorithm/sample.cc: Simpify and reduce

>>>>>         dependencies by using __gnu_test::test_container.

>>>>>         * testsuite/experimental/algorithm/sample-2.cc: New.

>>>>>         * testsuite/experimental/algorithm/shuffle.cc: New.

>>>>>

>>>>> Tested x86_64-linux, committed to trunk.

>>>>>

>>>>> This would be safe to backport, but nobody has noticed the algos are

>>>>> missing or complained, so it doesn't seem very important to backport.

>>>>>

>>>>>

>>>>

>>>>Hi,

>>>>

>>>>On bare-metal targets (aarch64 and arm + newlib), I've noticed that

>>>>the two new tests fail:

>>>>PASS: experimental/algorithm/shuffle.cc (test for excess errors)

>>>>spawn /aci-gcc-fsf/builds/gcc-fsf-gccsrc/obj-arm-none-eabi/gcc3/utils/bin/qemu-wrapper.sh

>>>>./shuffle.exe

>>>>terminate called after throwing an instance of 'std::runtime_error'

>>>>  what():  random_device::random_device(const std::string&)

>>>>

>>>>*** EXIT code 4242

>>>>FAIL: experimental/algorithm/shuffle.cc execution test

>>>>

>>>>PASS: experimental/algorithm/sample-2.cc (test for excess errors)

>>>>spawn /aci-gcc-fsf/builds/gcc-fsf-gccsrc/obj-arm-none-eabi/gcc3/utils/bin/qemu-wrapper.sh

>>>>./sample-2.exe

>>>>terminate called after throwing an instance of 'std::runtime_error'

>>>>  what():  random_device::random_device(const std::string&)

>>>>

>>>>*** EXIT code 4242

>>>>FAIL: experimental/algorithm/sample-2.cc execution test

>>>>

>>>>Does this ring a bell?

>>>

>>>Does the existing testsuite/experimental/random/randint.cc file fail

>>>in the same way?

>>>

>>

>>Yes it does.

>>

>>And so do:

>>25_algorithms/make_heap/complexity.cc

>

>This one also uses std::random_device.

>

>>23_containers/array/element_access/at_neg.cc

>

>Hmm,

>

> // Expected behavior is to either throw and have the uncaught

> // exception end up in a terminate handler which eventually exits,

> // or abort. (Depending on -fno-exceptions.)

>

>So this is expected to XFAIL.

>

>>26_numerics/random/random_device/cons/default.cc

>

>We should XFAIL the ones that use std::random_device, if we can

>identify an effective target to describe them.


This adds a new "random_device" effective-target, so the tests are
disabled when the random_device isn't usable.

Tested powerpc64le-linux, committed to trunk. If this works for
Christophe's bare metal targets I'll backport it to gcc-8-branch too.
commit b32bcdc1dc7ff6d483a4a7223d78198b7522cbe4
Author: Jonathan Wakely <jwakely@redhat.com>
Date:   Wed Jul 4 15:44:45 2018 +0100

    Define "random_device" effective target
    
    Currently only matches targets where _GLIBCXX_USE_RANDOM_TR1 is defined,
    which means /dev/random and /dev/urandom are usable.
    
            * testsuite/25_algorithms/make_heap/complexity.cc: Require effective
            target for std::random_device.
            * testsuite/26_numerics/random/random_device/cons/default.cc:
            Likewise.
            * testsuite/experimental/algorithm/sample-2.cc: Likewise.
            * testsuite/experimental/algorithm/shuffle.cc: Likewise.
            * testsuite/experimental/random/randint.cc: Likewise.
            * testsuite/lib/libstdc++.exp
            (check_effective_target_random_device): New proc.

diff --git a/libstdc++-v3/testsuite/25_algorithms/make_heap/complexity.cc b/libstdc++-v3/testsuite/25_algorithms/make_heap/complexity.cc
index cca48f61e0a..069d2d0433d 100644
--- a/libstdc++-v3/testsuite/25_algorithms/make_heap/complexity.cc
+++ b/libstdc++-v3/testsuite/25_algorithms/make_heap/complexity.cc
@@ -16,6 +16,7 @@
 // <http://www.gnu.org/licenses/>.
 
 // { dg-do run { target c++11 } }
+// { dg-require-effective-target random_device }
 
 #include <random>
 #include <vector>
diff --git a/libstdc++-v3/testsuite/26_numerics/random/random_device/cons/default.cc b/libstdc++-v3/testsuite/26_numerics/random/random_device/cons/default.cc
index 38210963f7e..5a34526a5f7 100644
--- a/libstdc++-v3/testsuite/26_numerics/random/random_device/cons/default.cc
+++ b/libstdc++-v3/testsuite/26_numerics/random/random_device/cons/default.cc
@@ -1,4 +1,5 @@
 // { dg-do run { target c++11 } }
+// { dg-require-effective-target random_device }
 // { dg-require-cstdint "" }
 //
 // 2008-11-24  Edward M. Smith-Rowland <3dw4rd@verizon.net>
diff --git a/libstdc++-v3/testsuite/experimental/algorithm/sample-2.cc b/libstdc++-v3/testsuite/experimental/algorithm/sample-2.cc
index 541d17e08a0..ef3f7daa14c 100644
--- a/libstdc++-v3/testsuite/experimental/algorithm/sample-2.cc
+++ b/libstdc++-v3/testsuite/experimental/algorithm/sample-2.cc
@@ -16,6 +16,7 @@
 // <http://www.gnu.org/licenses/>.
 
 // { dg-do run { target c++14 } }
+// { dg-require-effective-target random_device }
 // { dg-require-effective-target tls_runtime }
 // { dg-add-options tls }
 
diff --git a/libstdc++-v3/testsuite/experimental/algorithm/shuffle.cc b/libstdc++-v3/testsuite/experimental/algorithm/shuffle.cc
index ee06fbd996a..db958f600d4 100644
--- a/libstdc++-v3/testsuite/experimental/algorithm/shuffle.cc
+++ b/libstdc++-v3/testsuite/experimental/algorithm/shuffle.cc
@@ -1,4 +1,5 @@
 // { dg-do run { target c++14 } }
+// { dg-require-effective-target random_device }
 // { dg-require-effective-target tls_runtime }
 // { dg-add-options tls }
 
diff --git a/libstdc++-v3/testsuite/experimental/random/randint.cc b/libstdc++-v3/testsuite/experimental/random/randint.cc
index e80bd858f29..90ba72ac2fc 100644
--- a/libstdc++-v3/testsuite/experimental/random/randint.cc
+++ b/libstdc++-v3/testsuite/experimental/random/randint.cc
@@ -1,4 +1,5 @@
 // { dg-do run { target c++14 } }
+// { dg-require-effective-target random_device }
 // { dg-require-effective-target tls_runtime }
 // { dg-add-options tls }
 
diff --git a/libstdc++-v3/testsuite/lib/libstdc++.exp b/libstdc++-v3/testsuite/lib/libstdc++.exp
index 7af3266f855..d8717e402df 100644
--- a/libstdc++-v3/testsuite/lib/libstdc++.exp
+++ b/libstdc++-v3/testsuite/lib/libstdc++.exp
@@ -2061,6 +2061,34 @@ proc check_effective_target_cxx11-abi { } {
     return 0
 }
 
+# Return 1 if std::random_device should be usable using the current flags, 0 otherwise.
+proc check_effective_target_random_device { } {
+    global cxxflags
+
+    # Set up and preprocess a C++ test program that depends
+    # on std::random_device being usable.
+    set src random_device[pid].cc
+
+    set f [open $src "w"]
+    puts $f "#include <bits/c++config.h>"
+    puts $f "#if ! _GLIBCXX_USE_RANDOM_TR1"
+    puts $f "#  error No working std::random_device available"
+    puts $f "#endif"
+    close $f
+
+    set lines [v3_target_compile $src /dev/null preprocess ""]
+    file delete $src
+
+    if [string match "" $lines] {
+	# No error message, preprocessing succeeded.
+	verbose "check_v3_random_device: `1'" 2
+	return 1
+    }
+
+    verbose "check_v3_random_device: `0'" 2
+    return 0
+}
+
 set additional_prunes ""
 
 if { [info exists env(GCC_RUNTEST_PARALLELIZE_DIR)] \
Christophe Lyon July 5, 2018, 8:23 a.m. | #7
On Wed, 4 Jul 2018 at 18:56, Jonathan Wakely <jwakely@redhat.com> wrote:
>

> On 29/06/18 10:45 +0100, Jonathan Wakely wrote:

> >On 29/06/18 09:39 +0200, Christophe Lyon wrote:

> >>On Fri, 29 Jun 2018 at 09:21, Jonathan Wakely <jwakely@redhat.com> wrote:

> >>>

> >>>On 29/06/18 08:55 +0200, Christophe Lyon wrote:

> >>>>On Mon, 25 Jun 2018 at 18:23, Jonathan Wakely <jwakely@redhat.com> wrote:

> >>>>>

> >>>>> The additions to <experimental/random> were added in 2015 but the new

> >>>>> algorithms in <experimental/algorithm> were not. This adds them.

> >>>>>

> >>>>>         * include/experimental/algorithm (sample, shuffle): Add new overloads

> >>>>>         using per-thread random number engine.

> >>>>>         * testsuite/experimental/algorithm/sample.cc: Simpify and reduce

> >>>>>         dependencies by using __gnu_test::test_container.

> >>>>>         * testsuite/experimental/algorithm/sample-2.cc: New.

> >>>>>         * testsuite/experimental/algorithm/shuffle.cc: New.

> >>>>>

> >>>>> Tested x86_64-linux, committed to trunk.

> >>>>>

> >>>>> This would be safe to backport, but nobody has noticed the algos are

> >>>>> missing or complained, so it doesn't seem very important to backport.

> >>>>>

> >>>>>

> >>>>

> >>>>Hi,

> >>>>

> >>>>On bare-metal targets (aarch64 and arm + newlib), I've noticed that

> >>>>the two new tests fail:

> >>>>PASS: experimental/algorithm/shuffle.cc (test for excess errors)

> >>>>spawn /aci-gcc-fsf/builds/gcc-fsf-gccsrc/obj-arm-none-eabi/gcc3/utils/bin/qemu-wrapper.sh

> >>>>./shuffle.exe

> >>>>terminate called after throwing an instance of 'std::runtime_error'

> >>>>  what():  random_device::random_device(const std::string&)

> >>>>

> >>>>*** EXIT code 4242

> >>>>FAIL: experimental/algorithm/shuffle.cc execution test

> >>>>

> >>>>PASS: experimental/algorithm/sample-2.cc (test for excess errors)

> >>>>spawn /aci-gcc-fsf/builds/gcc-fsf-gccsrc/obj-arm-none-eabi/gcc3/utils/bin/qemu-wrapper.sh

> >>>>./sample-2.exe

> >>>>terminate called after throwing an instance of 'std::runtime_error'

> >>>>  what():  random_device::random_device(const std::string&)

> >>>>

> >>>>*** EXIT code 4242

> >>>>FAIL: experimental/algorithm/sample-2.cc execution test

> >>>>

> >>>>Does this ring a bell?

> >>>

> >>>Does the existing testsuite/experimental/random/randint.cc file fail

> >>>in the same way?

> >>>

> >>

> >>Yes it does.

> >>

> >>And so do:

> >>25_algorithms/make_heap/complexity.cc

> >

> >This one also uses std::random_device.

> >

> >>23_containers/array/element_access/at_neg.cc

> >

> >Hmm,

> >

> > // Expected behavior is to either throw and have the uncaught

> > // exception end up in a terminate handler which eventually exits,

> > // or abort. (Depending on -fno-exceptions.)

> >

> >So this is expected to XFAIL.

> >

> >>26_numerics/random/random_device/cons/default.cc

> >

> >We should XFAIL the ones that use std::random_device, if we can

> >identify an effective target to describe them.

>

> This adds a new "random_device" effective-target, so the tests are

> disabled when the random_device isn't usable.

>

> Tested powerpc64le-linux, committed to trunk. If this works for

> Christophe's bare metal targets I'll backport it to gcc-8-branch too.

>

Yes, that works for me: the tests are now UNSUPPORTED on aarch64*-elf
and arm-none-eabi.

Thanks!

Patch

diff --git a/libstdc++-v3/include/experimental/algorithm b/libstdc++-v3/include/experimental/algorithm
index fde4f347f88..4c51efb1c97 100644
--- a/libstdc++-v3/include/experimental/algorithm
+++ b/libstdc++-v3/include/experimental/algorithm
@@ -35,6 +35,7 @@ 
 
 #include <algorithm>
 #include <experimental/bits/lfts_config.h>
+#include <experimental/random>
 
 namespace std _GLIBCXX_VISIBILITY(default)
 {
@@ -42,7 +43,7 @@  _GLIBCXX_BEGIN_NAMESPACE_VERSION
 
 namespace experimental
 {
-inline namespace fundamentals_v1
+inline namespace fundamentals_v2
 {
   template<typename _ForwardIterator, typename _Searcher>
     inline _ForwardIterator
@@ -79,7 +80,23 @@  inline namespace fundamentals_v1
 			   __d,
 			   std::forward<_UniformRandomNumberGenerator>(__g));
     }
-} // namespace fundamentals_v1
+
+  template<typename _PopulationIterator, typename _SampleIterator,
+           typename _Distance>
+    inline _SampleIterator
+    sample(_PopulationIterator __first, _PopulationIterator __last,
+	   _SampleIterator __out, _Distance __n)
+    {
+      return experimental::sample(__first, __last, __out, __n,
+				  _S_randint_engine());
+    }
+
+  template<typename _RandomAccessIterator>
+    inline void
+    shuffle(_RandomAccessIterator __first, _RandomAccessIterator __last)
+    { return std::shuffle(__first, __last, _S_randint_engine()); }
+
+} // namespace fundamentals_v2
 } // namespace experimental
 
 _GLIBCXX_END_NAMESPACE_VERSION
diff --git a/libstdc++-v3/testsuite/experimental/algorithm/sample-2.cc b/libstdc++-v3/testsuite/experimental/algorithm/sample-2.cc
new file mode 100644
index 00000000000..4ef9a7c77e4
--- /dev/null
+++ b/libstdc++-v3/testsuite/experimental/algorithm/sample-2.cc
@@ -0,0 +1,98 @@ 
+// Copyright (C) 2018 Free Software Foundation, Inc.
+//
+// This file is part of the GNU ISO C++ Library.  This library is free
+// software; you can redistribute it and/or modify it under the
+// terms of the GNU General Public License as published by the
+// Free Software Foundation; either version 3, or (at your option)
+// any later version.
+
+// This library is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License for more details.
+
+// You should have received a copy of the GNU General Public License along
+// with this library; see the file COPYING3.  If not see
+// <http://www.gnu.org/licenses/>.
+
+// { dg-do run { target c++14 } }
+
+#include <experimental/algorithm>
+#include <algorithm>
+#include <testsuite_hooks.h>
+#include <testsuite_iterators.h>
+
+using __gnu_test::test_container;
+using __gnu_test::input_iterator_wrapper;
+using __gnu_test::output_iterator_wrapper;
+using __gnu_test::forward_iterator_wrapper;
+
+void
+test01()
+{
+  const int pop[] = { 1, 2 };
+  int samp[10] = { };
+
+  // population smaller than desired sample size
+  auto it = std::experimental::sample(pop, pop + 2, samp, 10);
+  VERIFY( it == samp + 2 );
+  VERIFY( std::accumulate(samp, samp + 10, 0) == 3 );
+}
+
+void
+test02()
+{
+  const int pop[] = { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 };
+  int samp[10] = { };
+
+  auto it = std::experimental::sample(pop, std::end(pop), samp, 10);
+  VERIFY( it == samp + 10 );
+
+  std::sort(samp, it);
+  auto it2 = std::unique(samp, it);
+  VERIFY( it2 == it );
+}
+
+void
+test03()
+{
+  const int pop[] = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, };
+  int samp[5] = { };
+
+  // input iterator for population
+  test_container<const int, input_iterator_wrapper> pop_in{pop};
+  auto it = std::experimental::sample(pop_in.begin(), pop_in.end(), samp, 5);
+  VERIFY( it == samp + 5 );
+
+  std::sort(samp, it);
+  auto it2 = std::unique(samp, it);
+  VERIFY( it2 == it );
+}
+
+void
+test04()
+{
+  const int pop[] = { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 };
+  int samp[5] = { };
+
+  // forward iterator for population and output iterator for result
+  test_container<const int, forward_iterator_wrapper> pop_fwd{pop};
+  test_container<int, output_iterator_wrapper> samp_out{samp};
+  auto it = std::experimental::sample(pop_fwd.begin(), pop_fwd.end(),
+				      samp_out.begin(), 5);
+
+  VERIFY( std::distance(samp, it.ptr) == 5 );
+
+  std::sort(samp, it.ptr);
+  auto it2 = std::unique(samp, it.ptr);
+  VERIFY( it2 == it.ptr );
+}
+
+int
+main()
+{
+  test01();
+  test02();
+  test03();
+  test04();
+}
diff --git a/libstdc++-v3/testsuite/experimental/algorithm/sample.cc b/libstdc++-v3/testsuite/experimental/algorithm/sample.cc
index 543a6efc461..b2373706f04 100644
--- a/libstdc++-v3/testsuite/experimental/algorithm/sample.cc
+++ b/libstdc++-v3/testsuite/experimental/algorithm/sample.cc
@@ -18,19 +18,17 @@ 
 // { dg-do run { target c++14 } }
 
 #include <experimental/algorithm>
-#include <iterator>
-#include <sstream>
-#include <forward_list>
-#include <vector>
 #include <random>
-#include <algorithm>
 #include <testsuite_hooks.h>
+#include <testsuite_iterators.h>
+
+using __gnu_test::test_container;
+using __gnu_test::input_iterator_wrapper;
+using __gnu_test::output_iterator_wrapper;
+using __gnu_test::forward_iterator_wrapper;
 
 std::mt19937 rng;
 
-using std::istream_iterator;
-using std::ostream_iterator;
-
 void
 test01()
 {
@@ -60,11 +58,12 @@  test02()
 void
 test03()
 {
-  std::istringstream pop("0 1 2 3 4 5 6 7 8 9");
+  const int pop[] = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, };
   int samp[5] = { };
 
   // input iterator for population
-  auto it = std::experimental::sample(istream_iterator<int>{pop}, {},
+  test_container<const int, input_iterator_wrapper> pop_in{pop};
+  auto it = std::experimental::sample(pop_in.begin(), pop_in.end(),
                                       samp,
                                       5, rng);
   VERIFY( it == samp + 5 );
@@ -77,21 +76,20 @@  test03()
 void
 test04()
 {
-  std::forward_list<int> pop{ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 };
-  std::stringstream samp;
+  const int pop[] = { 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 };
+  int samp[5] = { };
 
   // forward iterator for population and output iterator for result
-  std::experimental::sample(pop.begin(), pop.end(),
-                            ostream_iterator<int>{samp, " "},
-                            5, rng);
+  test_container<const int, forward_iterator_wrapper> pop_fwd{pop};
+  test_container<int, output_iterator_wrapper> samp_out{samp};
+  auto it = std::experimental::sample(pop_fwd.begin(), pop_fwd.end(),
+				      samp_out.begin(), 5, rng);
 
-  // samp.rdbuf()->pubseekoff(0, std::ios::beg);
-  std::vector<int> v(istream_iterator<int>{samp}, {});
-  VERIFY( v.size() == 5 );
+  VERIFY( std::distance(samp, it.ptr) == 5 );
 
-  std::sort(v.begin(), v.end());
-  auto it = std::unique(v.begin(), v.end());
-  VERIFY( it == v.end() );
+  std::sort(samp, it.ptr);
+  auto it2 = std::unique(samp, it.ptr);
+  VERIFY( it2 == it.ptr );
 }
 
 int
diff --git a/libstdc++-v3/testsuite/experimental/algorithm/shuffle.cc b/libstdc++-v3/testsuite/experimental/algorithm/shuffle.cc
new file mode 100644
index 00000000000..5c4e7fbd433
--- /dev/null
+++ b/libstdc++-v3/testsuite/experimental/algorithm/shuffle.cc
@@ -0,0 +1,61 @@ 
+// { dg-do run { target c++14 } }
+
+// Derived from: 2010-03-19  Paolo Carlini  <paolo.carlini@oracle.com>
+
+// Copyright (C) 2010-2018 Free Software Foundation, Inc.
+//
+// This file is part of the GNU ISO C++ Library.  This library is free
+// software; you can redistribute it and/or modify it under the
+// terms of the GNU General Public License as published by the
+// Free Software Foundation; either version 3, or (at your option)
+// any later version.
+
+// This library is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License for more details.
+
+// You should have received a copy of the GNU General Public License along
+// with this library; see the file COPYING3.  If not see
+// <http://www.gnu.org/licenses/>.
+
+#include <experimental/algorithm>
+#include <vector>
+#include <testsuite_hooks.h>
+
+void test01()
+{
+  for (unsigned size = 0; size < 50; ++size)
+    {
+      std::vector<int> vref(size);
+      std::iota(vref.begin(), vref.end(), 0);
+      std::vector<int> v1(vref), v2(vref);
+
+      std::experimental::shuffle(v1.begin(), v1.end());
+      std::experimental::shuffle(v2.begin(), v2.end());
+
+      if (size >= 10)
+	{
+	  VERIFY( !std::equal(v1.begin(), v1.end(), vref.begin()) );
+	  VERIFY( !std::equal(v2.begin(), v2.end(), vref.begin()) );
+	  VERIFY( !std::equal(v1.begin(), v1.end(), v2.begin()) );
+	}
+
+      for (unsigned ind = 0; ind < size; ++ind)
+	{
+	  auto it1 = std::find(v1.begin(), v1.end(), vref[ind]);
+	  auto it2 = std::find(v2.begin(), v2.end(), vref[ind]);
+	  VERIFY( it1 != v1.end() );
+	  VERIFY( it2 != v2.end() );
+	  v1.erase(it1);
+	  v2.erase(it2);
+	}
+      VERIFY( v1.empty() );
+      VERIFY( v2.empty() );
+    }
+}
+
+int main()
+{
+  test01();
+}