Обсуждение: Compilation warning on 9.5

<div dir="ltr"><pre>Hello:<br /><br />I am a pgRouting developer.<br /><br />In my CmakeLists.txt I use this flags:<br
/>set(CMAKE_C_FLAGS  "${CMAKE_C_FLAGS}   -std=gnu99 -fPIC -O2 -g -Wall -Wconversion -pedantic -fmax-errors=10
-Wmissing-prototypes-frounding-math") 
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -std=c++0x -fPIC -O2 -g -Wconversion -Wall -pedantic -fmax-errors=10 -Wextra
-frounding-math-Wno-deprecated") 
<br /><br />for testing I use travis CI framework and test pgRouting using postgresql 9.1 to 9.5<br />I follow the
instructionsfor including:<br />#include "postgres.h"<br />#include "executor/spi.h"<br />#include "funcapi.h"<br
/>#include"catalog/pg_type.h"<br />#if PGSQL_VERSION > 92<br />#include "access/htup_details.h"<br />#endif<br
/>#include"fmgr.h"<br /><br /><br />I wonder if -std=gnu99 is the correct standard to include postgres.h etc. in 9.5
<br/>because that standard (and all the flags I am using to generate pgrouting code without warnings)<br />catches the
followingcatches warnings of type conversions on some postgresql included files.<br />This doesn't happen on postgresql
9.1to 9.4<br /><br />for example:<br /><br /></pre><span id="1-2270">In file included from
/usr/include/postgresql/9.5/server/postgres.h:47:0,</span><br/><span id="1-2271"> from
/home/travis/build/pgRouting/pgrouting/src/dijkstra/src/many_to_many_dijkstra.c:31:</span><br/><span
id="1-2272">/usr/include/postgresql/9.5/server/c.h:298:9:warning: ISO C does not support ‘__int128’ type
[-pedantic]</span><br/><span id="1-2273">/usr/include/postgresql/9.5/server/c.h:299:18: warning: ISO C does not support
‘__int128’type [-pedantic]</span><br /><br /><span id="1-2273"></span><br /><span id="1-2273"><br /></span><br /><span
id="1-2274">Infile included from /usr/include/postgresql/9.5/server/port/atomics.h:119:0,</span><br /><span
id="1-2275">from /usr/include/postgresql/9.5/server/storage/lwlock.h:19,</span><br /><span id="1-2276"> from
/usr/include/postgresql/9.5/server/storage/lock.h:18,</span><br/><span id="1-2277"> from
/usr/include/postgresql/9.5/server/access/genam.h:20,</span><br/><span id="1-2278"> from
/usr/include/postgresql/9.5/server/nodes/execnodes.h:17,</span><br/><span id="1-2279"> from
/usr/include/postgresql/9.5/server/executor/execdesc.h:18,</span><br/><span id="1-2280"> from
/usr/include/postgresql/9.5/server/utils/portal.h:50,</span><br/><span id="1-2281"> from
/usr/include/postgresql/9.5/server/executor/spi.h:18,</span><br/><span id="1-2282"> from
/home/travis/build/pgRouting/pgrouting/src/dijkstra/src/many_to_many_dijkstra.c:32:</span><br/><span
id="1-2283">/usr/include/postgresql/9.5/server/port/atomics/generic.h:In function
‘pg_atomic_add_fetch_u32_impl’:</span><br/><span
id="1-2284">/usr/include/postgresql/9.5/server/port/atomics/generic.h:238:2:warning: conversion to ‘uint32’ from
‘int32’may change the sign of the result [-Wsign-conversion]</span><br /><span
id="1-2285">/usr/include/postgresql/9.5/server/port/atomics/generic.h:In function
‘pg_atomic_sub_fetch_u32_impl’:</span><br/><span
id="1-2286">/usr/include/postgresql/9.5/server/port/atomics/generic.h:247:2:warning: conversion to ‘uint32’ from
‘int32’may change the sign of the result [-Wsign-conversion]</span><br /><span
id="1-2287">/usr/include/postgresql/9.5/server/port/atomics/generic.h:In function
‘pg_atomic_add_fetch_u64_impl’:</span><br/><span
id="1-2288">/usr/include/postgresql/9.5/server/port/atomics/generic.h:372:2:warning: conversion to ‘long unsigned int’
from‘int64’ may change the sign of the result [-Wsign-conversion]</span><br /><span
id="1-2289">/usr/include/postgresql/9.5/server/port/atomics/generic.h:In function
‘pg_atomic_sub_fetch_u64_impl’:</span><br/><span
id="1-2290">/usr/include/postgresql/9.5/server/port/atomics/generic.h:381:2:warning: conversion to ‘long unsigned int’
from‘int64’ may change the sign of the result [-Wsign-conversion]</span><br /><pre>of course, I can't go and modify
<spanid="1-2290">generic.h, </span><span id="1-2273">c.h which are included when I include postgres.h or spi.h, or any
ofthe files<br />that I include that are of the postgresql project.<br /><br />I already posted in this mailing list <a
href="http://www.postgresql.org/message-id/BAY177-W104EC0B93C9FC5453B04CD8AA90@phx.gbl"
target="_blank">http://www.postgresql.org/message-id/BAY177-W104EC0B93C9FC5453B04CD8AA90@phx.gbl</a><br/>But talking
withmy co-developer Steve Woodbri, he suggested this mailing list.<br /></span></pre> <pre> 
you can see a full travis test here:
<a href="https://travis-ci.org/pgRouting/pgrouting/builds/108791787"
rel="nofollow">https://travis-ci.org/pgRouting/pgrouting/builds/108791787</a>
(I have my own warnings which I am fixing and are very visible from 9.1 to 9.4)

Note1: when pgRouting gets released, all those flags:<br />-Wall -Wconversion -pedantic -fmax-errors=10
-Wmissing-prototypes<br/>will be removed, and of course those warnings won't show up.<br /><br /><br /><br /> 
Thanks
Vicky Vergara</pre></div>

Vicky Vergara <vicky_vergara@hotmail.com> writes:
> I wonder if -std=gnu99 is the correct standard to include postgres.h etc. in 9.5 
> because that standard (and all the flags I am using to generate pgrouting code without warnings)
> catches the following catches warnings of type conversions on some postgresql included files.

It's not -std=gnu99 that's causing those messages, it's -pedantic and
-Wconversion respectively.

> /usr/include/postgresql/9.5/server/c.h:298:9: warning: ISO C does not support �__int128� type [-pedantic]
> /usr/include/postgresql/9.5/server/c.h:299:18: warning: ISO C does not support �__int128� type [-pedantic]

We're not going to do anything about this one; certainly we won't stop
using int128 where it's available, and there isn't any other apparent way
to suppress the warning.  I doubt that -pedantic is a useful switch in
practice, and this seems to be a particularly unhelpful bit of pedantry.
Consider dropping that flag.

> /usr/include/postgresql/9.5/server/port/atomics/generic.h: In function �pg_atomic_add_fetch_u32_impl�:
> /usr/include/postgresql/9.5/server/port/atomics/generic.h:238:2: warning: conversion to �uint32� from �int32� may
changethe sign of the result [-Wsign-conversion]
 

According to the gcc manual, inserting explicit casts would silence these;
is that worth doing?  I doubt anyone cares about making all our code
-Wconversion clean, but maybe making the headers clean would be worth
the trouble.
        regards, tom lane