You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Ruediger Pluem <rp...@apache.org> on 2010/07/20 08:57:37 UTC

Re: svn commit: r965408 - in /httpd/httpd/trunk: include/ap_mmn.h include/http_config.h modules/core/mod_so.c server/config.c


On 07/19/2010 12:06 PM, sf@apache.org wrote:
> Author: sf
> Date: Mon Jul 19 10:06:15 2010
> New Revision: 965408
> 
> URL: http://svn.apache.org/viewvc?rev=965408&view=rev
> Log:
> Add ap_find_module_short_name() to quickly get the module short name
> (i.e. symbol name with trailing "_module" removed) from the module_index.
> To be used for logging.
> 
> Modified:
>     httpd/httpd/trunk/include/ap_mmn.h
>     httpd/httpd/trunk/include/http_config.h
>     httpd/httpd/trunk/modules/core/mod_so.c
>     httpd/httpd/trunk/server/config.c
> 

> Modified: httpd/httpd/trunk/server/config.c
> URL: http://svn.apache.org/viewvc/httpd/httpd/trunk/server/config.c?rev=965408&r1=965407&r2=965408&view=diff
> ==============================================================================
> --- httpd/httpd/trunk/server/config.c (original)
> +++ httpd/httpd/trunk/server/config.c Mon Jul 19 10:06:15 2010

> @@ -549,8 +555,30 @@ AP_DECLARE(const char *) ap_add_module(m
>                                  "reached. Please increase "
>                                  "DYNAMIC_MODULE_LIMIT and recompile.", m->name);
>          }
> +
> +    }
> +    else if (!sym_name) {
> +        while (sym->modp != NULL) {
> +            if (sym->modp == m) {
> +                sym_name = sym->name;
> +                break;
> +            }
> +            sym++;
> +        }
>      }
>  
> +    if (sym_name) {
> +        int len = strlen(sym_name);
> +        int slen = strlen("_module");
> +        if (len > slen && !strcmp(sym_name + len - slen, "_module")) {
> +            len -= slen;
> +        }
> +
> +        ap_module_short_names[m->module_index] = strdup(sym_name);

Why not using pools here instead of malloc / free?

> +        ap_module_short_names[m->module_index][len] = '\0';
> +    }
> +
> +
>      /* Some C compilers put a complete path into __FILE__, but we want
>       * only the filename (e.g. mod_includes.c). So check for path
>       * components (Unix and DOS), and remove them.
> @@ -623,13 +651,17 @@ AP_DECLARE(void) ap_remove_module(module
>          modp->next = modp->next->next;
>      }
>  
> +    free(ap_module_short_names[m->module_index]);

Why not using pools here instead of malloc / free?

> +    ap_module_short_names[m->module_index] = NULL;
> +
>      m->module_index = -1; /* simulate being unloaded, should
>                             * be unnecessary */
>      dynamic_modules--;
>      total_modules--;
>  }
>  
> -AP_DECLARE(const char *) ap_add_loaded_module(module *mod, apr_pool_t *p)
> +AP_DECLARE(const char *) ap_add_loaded_module(module *mod, apr_pool_t *p,
> +                                              const char *short_name)
>  {
>      module **m;
>      const char *error;

Regards

RĂ¼diger

Re: svn commit: r965408 - in /httpd/httpd/trunk: include/ap_mmn.h include/http_config.h modules/core/mod_so.c server/config.c

Posted by Ruediger Pluem <rp...@apache.org>.

On 07/20/2010 08:43 PM, Stefan Fritsch wrote:
> On Tuesday 20 July 2010, Ruediger Pluem wrote:
>> On 07/19/2010 12:06 PM, sf@apache.org wrote:

>>> +
>>> +        ap_module_short_names[m->module_index] =
>>> strdup(sym_name);
>> Why not using pools here instead of malloc / free?
> 
> Because it didn't work. But looking at the code again, I believe this 
> is due to ap_setup_prelinked_modules calling ap_add_module with the 
> wrong pool. ap_setup_prelinked_modules is called only once and should 
> therefore use process->pool instead of process->pconf. Do you agree?
> 
> 

Not sure, because ap_add_module calls

    ap_add_module_commands(m, p);
    /*  FIXME: is this the right place to call this?
     *  It doesn't appear to be
     */
    ap_register_hooks(m, p);

with this pool. So this would need closer investigation if
process->pool is correct for them as well in this case.
But yes, if this fits process->pool is the correct choice IMHO.

Regards

RĂ¼diger



Re: svn commit: r965408 - in /httpd/httpd/trunk: include/ap_mmn.h include/http_config.h modules/core/mod_so.c server/config.c

Posted by Stefan Fritsch <sf...@sfritsch.de>.
On Tuesday 20 July 2010, Ruediger Pluem wrote:
> On 07/19/2010 12:06 PM, sf@apache.org wrote:
> > Author: sf
> > Date: Mon Jul 19 10:06:15 2010
> > New Revision: 965408
> > 
> > URL: http://svn.apache.org/viewvc?rev=965408&view=rev
> > Log:
> > Add ap_find_module_short_name() to quickly get the module short
> > name (i.e. symbol name with trailing "_module" removed) from the
> > module_index. To be used for logging.
> > 
> > Modified:
> >     httpd/httpd/trunk/include/ap_mmn.h
> >     httpd/httpd/trunk/include/http_config.h
> >     httpd/httpd/trunk/modules/core/mod_so.c
> >     httpd/httpd/trunk/server/config.c
> > 
> > Modified: httpd/httpd/trunk/server/config.c
> > URL:
> > http://svn.apache.org/viewvc/httpd/httpd/trunk/server/config.c?r
> > ev=965408&r1=965407&r2=965408&view=diff
> > ================================================================
> > ============== --- httpd/httpd/trunk/server/config.c (original)
> > +++ httpd/httpd/trunk/server/config.c Mon Jul 19 10:06:15 2010
> > 
> > @@ -549,8 +555,30 @@ AP_DECLARE(const char *) ap_add_module(m
> > 
> >                                  "reached. Please increase "
> >                                  "DYNAMIC_MODULE_LIMIT and
> >                                  recompile.", m->name);
> >          
> >          }
> > 
> > +
> > +    }
> > +    else if (!sym_name) {
> > +        while (sym->modp != NULL) {
> > +            if (sym->modp == m) {
> > +                sym_name = sym->name;
> > +                break;
> > +            }
> > +            sym++;
> > +        }
> > 
> >      }
> > 
> > +    if (sym_name) {
> > +        int len = strlen(sym_name);
> > +        int slen = strlen("_module");
> > +        if (len > slen && !strcmp(sym_name + len - slen,
> > "_module")) { +            len -= slen;
> > +        }
> > +
> > +        ap_module_short_names[m->module_index] =
> > strdup(sym_name);
> 
> Why not using pools here instead of malloc / free?

Because it didn't work. But looking at the code again, I believe this 
is due to ap_setup_prelinked_modules calling ap_add_module with the 
wrong pool. ap_setup_prelinked_modules is called only once and should 
therefore use process->pool instead of process->pconf. Do you agree?