You are viewing a plain text version of this content. The canonical link for it is here.
Posted to c-dev@axis.apache.org by "Jonathan Leffler (JIRA)" <ji...@apache.org> on 2008/07/09 01:12:31 UTC
[jira] Created: (AXIS2C-1231) Build on Solaris 10 requires minor
tweaks to get it to work
Build on Solaris 10 requires minor tweaks to get it to work
-----------------------------------------------------------
Key: AXIS2C-1231
URL: https://issues.apache.org/jira/browse/AXIS2C-1231
Project: Axis2-C
Issue Type: Bug
Components: build system (Unix/Linux)
Affects Versions: 1.4.0
Environment: Solaris 10 (UltraSPARC). GCC 4.3.1. Sun Assembler, Linker.
Reporter: Jonathan Leffler
In util/src/platforms/unix/thread_unix.c - lines 276-280:
#ifdef AXIS2_SOLARIS_but_not_for_Solaris_10
static const pthread_once_t once_init = { PTHREAD_ONCE_INIT };
#else
static const pthread_once_t once_init = PTHREAD_ONCE_INIT;
#endif
The compiler warned (and treated it as an error because of -Werror) about enclosing a scalar in braces.
The workaround I used was to change the #ifdef from AXIS2_SOLARIS to AXIS2_SOLARIS_but_not_for_Solaris_10.
Given the complaints I've seen in other bug entries for Solaris 9 issues, this problem is most likely in Solaris 10 only (or GCC 4.3.1 is fussier than other people's compilers).
In util/src/hash.c, GCC 4.3.1 spots that the static (recursive) function axutil_hash_entry_free() is unused by any other function. I commented it out with the #ifdef shown.
#ifdef USE_AXUTIL_HASH_ENTRY_FREE
static void
axutil_hash_entry_free(
const axutil_env_t *env,
axutil_hash_entry_t *hash_entry)
{
if (!hash_entry)
return;
if (hash_entry->next)
{
axutil_hash_entry_free(env, hash_entry->next);
}
AXIS2_FREE(env->allocator, hash_entry);
return;
}
#endif /* USE_AXUTIL_HASH_ENTRY_FREE */
I had linking problems - I needed to add -lsocket to a number of the programs' link lines. See a note appended to AXIS2C-638 and the related (but unannotated by me) AXIS2C-736.
Finally, reading tools/codegen/javatool/README, someone had severe problems with the 'r' key on their keyboard.
Typos noted include:
aguments -> arguments
geneation -> generation
geneate -> generate
expot -> export
diectoy -> directory
scipt -> script
I'm deeply suspicious that:
inteoptestdoclitpaametes --> interoptestdoclitparameters
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: axis-c-dev-unsubscribe@ws.apache.org
For additional commands, e-mail: axis-c-dev-help@ws.apache.org
[jira] Commented: (AXIS2C-1231) Build on Solaris 10 requires minor
tweaks to get it to work
Posted by "S.Uthaiyashankar (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/AXIS2C-1231?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12658519#action_12658519 ]
S.Uthaiyashankar commented on AXIS2C-1231:
------------------------------------------
(1) When including AXIS2_SOLARIS_but_not_for_Solaris_10 (util/src/platforms/unix/thread_unix.c), it fails in my system. I am using gcc version 3.4.6 on intel solaris 10. I suspect, this problem is there only on gcc 4 or more. So, haven't modify anything until somebody else confirms it.
(2) axutil_hash_entry_free is not used. For the moment, I have commented it out. We'll remove it later.
(3) Spelling error on tools/codegen/javatool/README is fixed. Thank you for pointing it out. Your suspicion is correct, (inteoptestdoclitpaametes --> interoptestdoclitparameters ) :D
> Build on Solaris 10 requires minor tweaks to get it to work
> -----------------------------------------------------------
>
> Key: AXIS2C-1231
> URL: https://issues.apache.org/jira/browse/AXIS2C-1231
> Project: Axis2-C
> Issue Type: Bug
> Components: build system (Unix/Linux)
> Affects Versions: 1.4.0
> Environment: Solaris 10 (UltraSPARC). GCC 4.3.1. Sun Assembler, Linker.
> Reporter: Jonathan Leffler
> Assignee: S.Uthaiyashankar
> Fix For: 1.6.0
>
> Original Estimate: 2h
> Remaining Estimate: 2h
>
> In util/src/platforms/unix/thread_unix.c - lines 276-280:
> #ifdef AXIS2_SOLARIS_but_not_for_Solaris_10
> static const pthread_once_t once_init = { PTHREAD_ONCE_INIT };
> #else
> static const pthread_once_t once_init = PTHREAD_ONCE_INIT;
> #endif
> The compiler warned (and treated it as an error because of -Werror) about enclosing a scalar in braces.
> The workaround I used was to change the #ifdef from AXIS2_SOLARIS to AXIS2_SOLARIS_but_not_for_Solaris_10.
> Given the complaints I've seen in other bug entries for Solaris 9 issues, this problem is most likely in Solaris 10 only (or GCC 4.3.1 is fussier than other people's compilers).
> In util/src/hash.c, GCC 4.3.1 spots that the static (recursive) function axutil_hash_entry_free() is unused by any other function. I commented it out with the #ifdef shown.
> #ifdef USE_AXUTIL_HASH_ENTRY_FREE
> static void
> axutil_hash_entry_free(
> const axutil_env_t *env,
> axutil_hash_entry_t *hash_entry)
> {
> if (!hash_entry)
> return;
> if (hash_entry->next)
> {
> axutil_hash_entry_free(env, hash_entry->next);
> }
> AXIS2_FREE(env->allocator, hash_entry);
> return;
> }
> #endif /* USE_AXUTIL_HASH_ENTRY_FREE */
> I had linking problems - I needed to add -lsocket to a number of the programs' link lines. See a note appended to AXIS2C-638 and the related (but unannotated by me) AXIS2C-736.
> Finally, reading tools/codegen/javatool/README, someone had severe problems with the 'r' key on their keyboard.
> Typos noted include:
> aguments -> arguments
> geneation -> generation
> geneate -> generate
> expot -> export
> diectoy -> directory
> scipt -> script
> I'm deeply suspicious that:
> inteoptestdoclitpaametes --> interoptestdoclitparameters
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Issue Comment Edited: (AXIS2C-1231) Build on Solaris 10
requires minor tweaks to get it to work
Posted by "S.Uthaiyashankar (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/AXIS2C-1231?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12658519#action_12658519 ]
shankar edited comment on AXIS2C-1231 at 12/22/08 3:54 AM:
--------------------------------------------------------------------
(1) When including AXIS2_SOLARIS_but_not_for_Solaris_10 (util/src/platforms/unix/thread_unix.c), it fails in my system. I am using gcc version 3.4.6 on intel solaris 10. I suspect, this problem is there only on gcc 4 or more. So, haven't modify anything until somebody else confirms it.
(2) axutil_hash_entry_free is not used. For the moment, I have commented it out. We'll remove it later.
(3) Spelling error on tools/codegen/javatool/README is fixed. Thank you for pointing it out. Your suspicion is correct, (inteoptestdoclitpaametes --> interoptestdoclitparameters ) :D
Last two fixes were there in revision 728656
was (Author: shankar):
(1) When including AXIS2_SOLARIS_but_not_for_Solaris_10 (util/src/platforms/unix/thread_unix.c), it fails in my system. I am using gcc version 3.4.6 on intel solaris 10. I suspect, this problem is there only on gcc 4 or more. So, haven't modify anything until somebody else confirms it.
(2) axutil_hash_entry_free is not used. For the moment, I have commented it out. We'll remove it later.
(3) Spelling error on tools/codegen/javatool/README is fixed. Thank you for pointing it out. Your suspicion is correct, (inteoptestdoclitpaametes --> interoptestdoclitparameters ) :D
> Build on Solaris 10 requires minor tweaks to get it to work
> -----------------------------------------------------------
>
> Key: AXIS2C-1231
> URL: https://issues.apache.org/jira/browse/AXIS2C-1231
> Project: Axis2-C
> Issue Type: Bug
> Components: build system (Unix/Linux)
> Affects Versions: 1.4.0
> Environment: Solaris 10 (UltraSPARC). GCC 4.3.1. Sun Assembler, Linker.
> Reporter: Jonathan Leffler
> Assignee: S.Uthaiyashankar
> Fix For: 1.6.0
>
> Original Estimate: 2h
> Remaining Estimate: 2h
>
> In util/src/platforms/unix/thread_unix.c - lines 276-280:
> #ifdef AXIS2_SOLARIS_but_not_for_Solaris_10
> static const pthread_once_t once_init = { PTHREAD_ONCE_INIT };
> #else
> static const pthread_once_t once_init = PTHREAD_ONCE_INIT;
> #endif
> The compiler warned (and treated it as an error because of -Werror) about enclosing a scalar in braces.
> The workaround I used was to change the #ifdef from AXIS2_SOLARIS to AXIS2_SOLARIS_but_not_for_Solaris_10.
> Given the complaints I've seen in other bug entries for Solaris 9 issues, this problem is most likely in Solaris 10 only (or GCC 4.3.1 is fussier than other people's compilers).
> In util/src/hash.c, GCC 4.3.1 spots that the static (recursive) function axutil_hash_entry_free() is unused by any other function. I commented it out with the #ifdef shown.
> #ifdef USE_AXUTIL_HASH_ENTRY_FREE
> static void
> axutil_hash_entry_free(
> const axutil_env_t *env,
> axutil_hash_entry_t *hash_entry)
> {
> if (!hash_entry)
> return;
> if (hash_entry->next)
> {
> axutil_hash_entry_free(env, hash_entry->next);
> }
> AXIS2_FREE(env->allocator, hash_entry);
> return;
> }
> #endif /* USE_AXUTIL_HASH_ENTRY_FREE */
> I had linking problems - I needed to add -lsocket to a number of the programs' link lines. See a note appended to AXIS2C-638 and the related (but unannotated by me) AXIS2C-736.
> Finally, reading tools/codegen/javatool/README, someone had severe problems with the 'r' key on their keyboard.
> Typos noted include:
> aguments -> arguments
> geneation -> generation
> geneate -> generate
> expot -> export
> diectoy -> directory
> scipt -> script
> I'm deeply suspicious that:
> inteoptestdoclitpaametes --> interoptestdoclitparameters
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (AXIS2C-1231) Build on Solaris 10 requires minor
tweaks to get it to work
Posted by "S.Uthaiyashankar (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/AXIS2C-1231?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12658522#action_12658522 ]
S.Uthaiyashankar commented on AXIS2C-1231:
------------------------------------------
Added
#if defined(AXIS2_SOLARIS) && (__GNUC__ <= 3)
static const pthread_once_t once_init = { PTHREAD_ONCE_INIT };
#else
static const pthread_once_t once_init = PTHREAD_ONCE_INIT;
#endif
Please check whether it fixed the problem (1)
> Build on Solaris 10 requires minor tweaks to get it to work
> -----------------------------------------------------------
>
> Key: AXIS2C-1231
> URL: https://issues.apache.org/jira/browse/AXIS2C-1231
> Project: Axis2-C
> Issue Type: Bug
> Components: build system (Unix/Linux)
> Affects Versions: 1.4.0
> Environment: Solaris 10 (UltraSPARC). GCC 4.3.1. Sun Assembler, Linker.
> Reporter: Jonathan Leffler
> Assignee: S.Uthaiyashankar
> Fix For: 1.6.0
>
> Original Estimate: 2h
> Remaining Estimate: 2h
>
> In util/src/platforms/unix/thread_unix.c - lines 276-280:
> #ifdef AXIS2_SOLARIS_but_not_for_Solaris_10
> static const pthread_once_t once_init = { PTHREAD_ONCE_INIT };
> #else
> static const pthread_once_t once_init = PTHREAD_ONCE_INIT;
> #endif
> The compiler warned (and treated it as an error because of -Werror) about enclosing a scalar in braces.
> The workaround I used was to change the #ifdef from AXIS2_SOLARIS to AXIS2_SOLARIS_but_not_for_Solaris_10.
> Given the complaints I've seen in other bug entries for Solaris 9 issues, this problem is most likely in Solaris 10 only (or GCC 4.3.1 is fussier than other people's compilers).
> In util/src/hash.c, GCC 4.3.1 spots that the static (recursive) function axutil_hash_entry_free() is unused by any other function. I commented it out with the #ifdef shown.
> #ifdef USE_AXUTIL_HASH_ENTRY_FREE
> static void
> axutil_hash_entry_free(
> const axutil_env_t *env,
> axutil_hash_entry_t *hash_entry)
> {
> if (!hash_entry)
> return;
> if (hash_entry->next)
> {
> axutil_hash_entry_free(env, hash_entry->next);
> }
> AXIS2_FREE(env->allocator, hash_entry);
> return;
> }
> #endif /* USE_AXUTIL_HASH_ENTRY_FREE */
> I had linking problems - I needed to add -lsocket to a number of the programs' link lines. See a note appended to AXIS2C-638 and the related (but unannotated by me) AXIS2C-736.
> Finally, reading tools/codegen/javatool/README, someone had severe problems with the 'r' key on their keyboard.
> Typos noted include:
> aguments -> arguments
> geneation -> generation
> geneate -> generate
> expot -> export
> diectoy -> directory
> scipt -> script
> I'm deeply suspicious that:
> inteoptestdoclitpaametes --> interoptestdoclitparameters
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Resolved: (AXIS2C-1231) Build on Solaris 10 requires minor
tweaks to get it to work
Posted by "S.Uthaiyashankar (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/AXIS2C-1231?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
S.Uthaiyashankar resolved AXIS2C-1231.
--------------------------------------
Resolution: Fixed
Fixed in revision 728663
> Build on Solaris 10 requires minor tweaks to get it to work
> -----------------------------------------------------------
>
> Key: AXIS2C-1231
> URL: https://issues.apache.org/jira/browse/AXIS2C-1231
> Project: Axis2-C
> Issue Type: Bug
> Components: build system (Unix/Linux)
> Affects Versions: 1.4.0
> Environment: Solaris 10 (UltraSPARC). GCC 4.3.1. Sun Assembler, Linker.
> Reporter: Jonathan Leffler
> Assignee: S.Uthaiyashankar
> Fix For: 1.6.0
>
> Original Estimate: 2h
> Remaining Estimate: 2h
>
> In util/src/platforms/unix/thread_unix.c - lines 276-280:
> #ifdef AXIS2_SOLARIS_but_not_for_Solaris_10
> static const pthread_once_t once_init = { PTHREAD_ONCE_INIT };
> #else
> static const pthread_once_t once_init = PTHREAD_ONCE_INIT;
> #endif
> The compiler warned (and treated it as an error because of -Werror) about enclosing a scalar in braces.
> The workaround I used was to change the #ifdef from AXIS2_SOLARIS to AXIS2_SOLARIS_but_not_for_Solaris_10.
> Given the complaints I've seen in other bug entries for Solaris 9 issues, this problem is most likely in Solaris 10 only (or GCC 4.3.1 is fussier than other people's compilers).
> In util/src/hash.c, GCC 4.3.1 spots that the static (recursive) function axutil_hash_entry_free() is unused by any other function. I commented it out with the #ifdef shown.
> #ifdef USE_AXUTIL_HASH_ENTRY_FREE
> static void
> axutil_hash_entry_free(
> const axutil_env_t *env,
> axutil_hash_entry_t *hash_entry)
> {
> if (!hash_entry)
> return;
> if (hash_entry->next)
> {
> axutil_hash_entry_free(env, hash_entry->next);
> }
> AXIS2_FREE(env->allocator, hash_entry);
> return;
> }
> #endif /* USE_AXUTIL_HASH_ENTRY_FREE */
> I had linking problems - I needed to add -lsocket to a number of the programs' link lines. See a note appended to AXIS2C-638 and the related (but unannotated by me) AXIS2C-736.
> Finally, reading tools/codegen/javatool/README, someone had severe problems with the 'r' key on their keyboard.
> Typos noted include:
> aguments -> arguments
> geneation -> generation
> geneate -> generate
> expot -> export
> diectoy -> directory
> scipt -> script
> I'm deeply suspicious that:
> inteoptestdoclitpaametes --> interoptestdoclitparameters
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (AXIS2C-1231) Build on Solaris 10 requires minor
tweaks to get it to work
Posted by "S.Uthaiyashankar (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/AXIS2C-1231?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
S.Uthaiyashankar updated AXIS2C-1231:
-------------------------------------
Fix Version/s: 1.6.0
Assignee: S.Uthaiyashankar
> Build on Solaris 10 requires minor tweaks to get it to work
> -----------------------------------------------------------
>
> Key: AXIS2C-1231
> URL: https://issues.apache.org/jira/browse/AXIS2C-1231
> Project: Axis2-C
> Issue Type: Bug
> Components: build system (Unix/Linux)
> Affects Versions: 1.4.0
> Environment: Solaris 10 (UltraSPARC). GCC 4.3.1. Sun Assembler, Linker.
> Reporter: Jonathan Leffler
> Assignee: S.Uthaiyashankar
> Fix For: 1.6.0
>
> Original Estimate: 2h
> Remaining Estimate: 2h
>
> In util/src/platforms/unix/thread_unix.c - lines 276-280:
> #ifdef AXIS2_SOLARIS_but_not_for_Solaris_10
> static const pthread_once_t once_init = { PTHREAD_ONCE_INIT };
> #else
> static const pthread_once_t once_init = PTHREAD_ONCE_INIT;
> #endif
> The compiler warned (and treated it as an error because of -Werror) about enclosing a scalar in braces.
> The workaround I used was to change the #ifdef from AXIS2_SOLARIS to AXIS2_SOLARIS_but_not_for_Solaris_10.
> Given the complaints I've seen in other bug entries for Solaris 9 issues, this problem is most likely in Solaris 10 only (or GCC 4.3.1 is fussier than other people's compilers).
> In util/src/hash.c, GCC 4.3.1 spots that the static (recursive) function axutil_hash_entry_free() is unused by any other function. I commented it out with the #ifdef shown.
> #ifdef USE_AXUTIL_HASH_ENTRY_FREE
> static void
> axutil_hash_entry_free(
> const axutil_env_t *env,
> axutil_hash_entry_t *hash_entry)
> {
> if (!hash_entry)
> return;
> if (hash_entry->next)
> {
> axutil_hash_entry_free(env, hash_entry->next);
> }
> AXIS2_FREE(env->allocator, hash_entry);
> return;
> }
> #endif /* USE_AXUTIL_HASH_ENTRY_FREE */
> I had linking problems - I needed to add -lsocket to a number of the programs' link lines. See a note appended to AXIS2C-638 and the related (but unannotated by me) AXIS2C-736.
> Finally, reading tools/codegen/javatool/README, someone had severe problems with the 'r' key on their keyboard.
> Typos noted include:
> aguments -> arguments
> geneation -> generation
> geneate -> generate
> expot -> export
> diectoy -> directory
> scipt -> script
> I'm deeply suspicious that:
> inteoptestdoclitpaametes --> interoptestdoclitparameters
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.