Recent libstdc++ build failure in hashtable_policy.h on alpha, r201522

2013-08-06 Thread Uros Bizjak
Hello!

Recent build failure on alpha-linux-gnu (r201522)

gmake[6]: Entering directory
`/space/homedirs/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/src/c++11'
/bin/sh ../../libtool --tag CXX --tag disable-shared   --mode=compile
/home/uros/gcc-build-xxx/./gcc/xgcc -shared-libgcc
-B/home/uros/gcc-build-xxx/./gcc -nostdinc++
-L/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/src
-L/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/src/.libs
-B/usr/local/alphaev68-unknown-linux-gnu/bin/
-B/usr/local/alphaev68-unknown-linux-gnu/lib/ -isystem
/usr/local/alphaev68-unknown-linux-gnu/include -isystem
/usr/local/alphaev68-unknown-linux-gnu/sys-include
-I/home/uros/gcc-svn/trunk/libstdc++-v3/../libgcc
-I/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/include/alphaev68-unknown-linux-gnu
-I/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/include
-I/home/uros/gcc-svn/trunk/libstdc++-v3/libsupc++  -std=gnu++11
-prefer-pic -D_GLIBCXX_SHARED -fno-implicit-templates -Wall -Wextra
-Wwrite-strings -Wcast-qual -Wabi  -fdiagnostics-show-location=once
-ffunction-sections -fdata-sections  -frandom-seed=hashtable_c++0x.lo
-g -O2 -D_GNU_SOURCE -mieee  -fimplicit-templates -c
../../../../../gcc-svn/trunk/libstdc++-v3/src/c++11/hashtable_c++0x.cc
libtool: compile:  /home/uros/gcc-build-xxx/./gcc/xgcc -shared-libgcc
-B/home/uros/gcc-build-xxx/./gcc -nostdinc++
-L/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/src
-L/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/src/.libs
-B/usr/local/alphaev68-unknown-linux-gnu/bin/
-B/usr/local/alphaev68-unknown-linux-gnu/lib/ -isystem
/usr/local/alphaev68-unknown-linux-gnu/include -isystem
/usr/local/alphaev68-unknown-linux-gnu/sys-include
-I/home/uros/gcc-svn/trunk/libstdc++-v3/../libgcc
-I/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/include/alphaev68-unknown-linux-gnu
-I/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/include
-I/home/uros/gcc-svn/trunk/libstdc++-v3/libsupc++ -std=gnu++11
-D_GLIBCXX_SHARED -fno-implicit-templates -Wall -Wextra
-Wwrite-strings -Wcast-qual -Wabi -fdiagnostics-show-location=once
-ffunction-sections -fdata-sections -frandom-seed=hashtable_c++0x.lo
-g -O2 -D_GNU_SOURCE -mieee -fimplicit-templates -c
../../../../../gcc-svn/trunk/libstdc++-v3/src/c++11/hashtable_c++0x.cc
 -fPIC -DPIC -D_GLIBCXX_SHARED -o hashtable_c++0x.o
In file included from
../../../../../gcc-svn/trunk/libstdc++-v3/src/c++11/hashtable_c++0x.cc:32:0:
/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/include/bits/hashtable_policy.h:700:11:
error: expected nested-name-specifier before ‘__alloctr_rebind’
  typename __alloctr_rebind_Alloc, __node_type::__type;
   ^
/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/include/bits/hashtable_policy.h:701:42:
error: ‘__node_alloc_type’ was not declared in this scope
   using __node_gen_type = _AllocNode__node_alloc_type;
  ^
/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/include/bits/hashtable_policy.h:701:59:
error: template argument 1 is invalid
   using __node_gen_type = _AllocNode__node_alloc_type;
   ^
/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/include/bits/hashtable_policy.h:
In member function ‘std::__detail::_Insert_base_Key, _Value, _Alloc,
_ExtractKey, _Equal, _H1, _H2, _Hash, _RehashPolicy,
_Traits::__ireturn_type std::__detail::_Insert_base_Key, _Value,
_Alloc, _ExtractKey, _Equal, _H1, _H2, _Hash, _RehashPolicy,
_Traits::insert(const value_type)’:
/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/include/bits/hashtable_policy.h:717:2:
error: ‘__node_gen_type’ was not declared in this scope
  __node_gen_type __node_gen(__h);
  ^
/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/include/bits/hashtable_policy.h:718:28:
error: ‘__node_gen’ was not declared in this scope
  return __h._M_insert(__v, __node_gen, __unique_keys());
^
/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/include/bits/hashtable_policy.h:
In member function ‘std::__detail::_Insert_base_Key, _Value, _Alloc,
_ExtractKey, _Equal, _H1, _H2, _Hash, _RehashPolicy,
_Traits::iterator std::__detail::_Insert_base_Key, _Value, _Alloc,
_ExtractKey, _Equal, _H1, _H2, _Hash, _RehashPolicy,
_Traits::insert(std::__detail::_Insert_base_Key, _Value, _Alloc,
_ExtractKey, _Equal, _H1, _H2, _Hash, _RehashPolicy,
_Traits::const_iterator, const value_type)’:
/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/include/bits/hashtable_policy.h:725:2:
error: ‘__node_gen_type’ was not declared in this scope
  __node_gen_type __node_gen(__h);
  ^
/home/uros/gcc-build-xxx/alphaev68-unknown-linux-gnu/libstdc++-v3/include/bits/hashtable_policy.h:726:36:
error: ‘__node_gen’ was not 

Re: Recent libstdc++ build failure in hashtable_policy.h on alpha, r201522

2013-08-06 Thread Paolo Carlini

On 08/06/2013 11:30 AM, Uros Bizjak wrote:

Hello!

Recent build failure on alpha-linux-gnu (r201522)
Uros, please revert it, thanks, we don't have the time to seriously look 
into it now.


Francois, please be more careful with testing, it's the second time in a 
few days.


Paolo.


Re: Recent libstdc++ build failure in hashtable_policy.h on alpha, r201522

2013-08-06 Thread Uros Bizjak
On Tue, Aug 6, 2013 at 11:33 AM, Paolo Carlini paolo.carl...@oracle.com wrote:

 Recent build failure on alpha-linux-gnu (r201522)

 Uros, please revert it, thanks, we don't have the time to seriously look
 into it now.

 Francois, please be more careful with testing, it's the second time in a few
 days.

It looks I wasn't clear enough: I found the regression when I built
revision r201522. I didn't yet bisect the failure, but if the
offending commit can be found from the data I provided, it will save
me a couple of hours trying to bisect recent commits.

Thanks,
Uros.


Re: Recent libstdc++ build failure in hashtable_policy.h on alpha, r201522

2013-08-06 Thread Paolo Carlini

On 08/06/2013 11:40 AM, Uros Bizjak wrote:

On Tue, Aug 6, 2013 at 11:33 AM, Paolo Carlini paolo.carl...@oracle.com wrote:


Recent build failure on alpha-linux-gnu (r201522)

Uros, please revert it, thanks, we don't have the time to seriously look
into it now.

Francois, please be more careful with testing, it's the second time in a few
days.

It looks I wasn't clear enough: I found the regression when I built
revision r201522.
Well r201522 corresponds to a serious change in exactly that code. I 
thought it was clear. And of course we all bootstrapped on x86_64-linux 
the recent revisions, immediately before that one.


I recommend reverting.

Thanks,
Paolo.