Re: [PATCH] mod_cache fixes: #3

2004-08-02 Thread Bill Stoddard
Justin Erenkrantz wrote: --On Sunday, August 1, 2004 11:25 AM -0400 Bill Stoddard [EMAIL PROTECTED] wrote: Too many changes in one patch. Break this up into multiple consumable in 15 minute patches and I'll review them. * modules/experimental/mod_disk_cache.c: Allow sendfile on cache bodies.

Re: [PATCH] mod_cache fixes: #3

2004-08-02 Thread Brian Akins
Bill Stoddard wrote: /* Open the headers file */ -rc = apr_file_open(hfd, headers, APR_READ|APR_BINARY, 0, r-pool); +rc = apr_file_open(hfd, headers, flags, 0, r-pool); Should be something like this adapted from core: core_dir_config *core_config; core_config = (core_dir_config *)

Re: [PATCH] mod_cache fixes: #3

2004-08-02 Thread Bill Stoddard
Brian Akins wrote: Bill Stoddard wrote: /* Open the headers file */ -rc = apr_file_open(hfd, headers, APR_READ|APR_BINARY, 0, r-pool); +rc = apr_file_open(hfd, headers, flags, 0, r-pool); Should be something like this adapted from core: core_dir_config *core_config; core_config =

Re: [PATCH] mod_cache fixes: #3

2004-08-02 Thread William A. Rowe, Jr.
At 10:46 AM 8/2/2004, Bill Stoddard wrote: EnableSendfile off is a global directive (?) so only need to check it once at startup and save it in a static variable? Oh, no! Directory /fs1/shared/www EnableSendfile Off /Directory kills sendfile for that mount. Bill

Re: [PATCH] mod_cache fixes: #3

2004-08-02 Thread Brian Akins
William A. Rowe, Jr. wrote: Oh, no! Directory /fs1/shared/www EnableSendfile Off /Directory kills sendfile for that mount. Bill However, in a quick handler, only the global config matters. Am I correct? Because the per_dir has not been merged until map to storage and Location is later.

Re: [PATCH] mod_cache fixes: #3

2004-08-02 Thread Justin Erenkrantz
--On Monday, August 2, 2004 10:35 AM -0400 Bill Stoddard [EMAIL PROTECTED] wrote: * modules/experimental/mod_disk_cache.c: Allow sendfile on cache bodies. -1, Need to check for EnableSendfile off. No, core_output_filter does that check. Modules don't have that information whether

Re: [PATCH] mod_cache fixes: #3

2004-08-02 Thread Bill Stoddard
Justin Erenkrantz wrote: --On Monday, August 2, 2004 10:35 AM -0400 Bill Stoddard [EMAIL PROTECTED] wrote: * modules/experimental/mod_disk_cache.c: Allow sendfile on cache bodies. -1, Need to check for EnableSendfile off. No, core_output_filter does that check. Modules don't have that

Re: [PATCH] mod_cache fixes: #3

2004-08-02 Thread Justin Erenkrantz
--On Monday, August 2, 2004 1:05 PM -0400 Bill Stoddard [EMAIL PROTECTED] wrote: I should amend my vote a -.5. The patch should work as you've coded it but opening a file for use with apr_sendfile causes the file to be opened for overlapped i/o on Windows. I expect some of the codepaths will

Re: [PATCH] mod_cache fixes: #3

2004-08-02 Thread Bill Stoddard
Justin Erenkrantz wrote: --On Monday, August 2, 2004 1:05 PM -0400 Bill Stoddard [EMAIL PROTECTED] wrote: I should amend my vote a -.5. The patch should work as you've coded it but opening a file for use with apr_sendfile causes the file to be opened for overlapped i/o on Windows. I expect

[PATCH] mod_cache fixes: #3

2004-08-01 Thread Justin Erenkrantz
--On Sunday, August 1, 2004 11:25 AM -0400 Bill Stoddard [EMAIL PROTECTED] wrote: Too many changes in one patch. Break this up into multiple consumable in 15 minute patches and I'll review them. * modules/experimental/mod_disk_cache.c: Allow sendfile on cache bodies. Index: