# [RFC] Cache optimizations 2 messages from 2005-05-09 to 2005-05-09. Participants: Daniel Barkalow, Junio C Hamano. Thread: https://gitlist.dev/t/547 ## Daniel Barkalow, 2005-05-09 05:15 Subject: [RFC] Cache optimizations Message-ID: URL: https://gitlist.dev/e/Pine.LNX.4.21.0505090058240.30848-100000%40iabervon.org ``` Does anyone know of a reason not to refresh each entry in the cache as it's checked out? It seems to me like this would eliminate the need to call update-cache --refresh after checkout-cache and it would update only the entries for files which are actually changed in the working directory. Is it just that the code to lock and write the cache is (currently) inconvenient to add to different programs? -Daniel *This .sig left intentionally blank* ``` ## Junio C Hamano, 2005-05-09 05:49 Subject: Re: [RFC] Cache optimizations Message-ID: <7vzmv5ymn6.fsf@assigned-by-dhcp.cox.net> URL: https://gitlist.dev/e/7vzmv5ymn6.fsf%40assigned-by-dhcp.cox.net In-Reply-To: ``` >>>>> "DB" == Daniel Barkalow writes: DB> Does anyone know of a reason not to refresh each entry in DB> the cache as it's checked out? I vaguely recall Linus saying something about wanting to keep checkout-cache being a read-only consumer of the index file; I've checked mailing list archive but I cannot not find a reference to point you at, though. ```