httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "William A. Rowe, Jr." <wr...@apache.org>
Subject Re: cvs commit: httpd-2.0/modules/ssl mod_ssl.c mod_ssl.h ssl_engine_config.c ssl_engine_mutex.c
Date Wed, 19 Mar 2003 01:47:42 GMT
At 06:19 PM 3/18/2003, André Malo wrote:
>* jim@apache.org wrote:
>
>>        if (strcEQ(arg, "none") || strcEQ(arg, "no")) {
>>            mc->nMutexMode  = SSL_MUTEXMODE_NONE;
>>        }
>>   +    /* NOTE: previously, 'yes' implied 'sem' */
>>   +    else if (strcEQ(arg, "default") || strcEQ(arg, "yes")) {
>>   +        mc->nMutexMode  = SSL_MUTEXMODE_USED;
>>   +        mc->nMutexMech = APR_LOCK_DEFAULT;
>
>ehm, sorry for that late response to that ;-)
>But shouldn't we follow our own conventions and (at least also) recognize 
>On|Off here (instead of yes/no)?

I was about to say that 'on' implies that 'yes' is a correct setting going
forward, and that we only hoped to ensure continued success of existing
configs.  But now that I think about it, we are saying 'default locking if
simply turned on' - so I suppose 'on' would also be alright in that context.

On the other hand, no, none and off all seem like reasonable flags to disable
the mutex altogether.  Adding off wouldn't be a bad thing.

Bill


Mime
View raw message