Re: Disabling updating of makefiles

[Colm Aengus Murphy]

Hi Greg, I use DEPENDS_FILES = $(OBJ_DIR)/dummy.P $(wildcard $(OBJ_DIR)/*.P) -include $(DEPENDS_FILES) %.P: I use .P instead of .d as I do some post processing on the .d files from gcc. It appears after some testing that most of the timing problems relate to file io. Things are much faster on a local disk compared with a clearcase dynamic view. I'll send in a bug report for the make -R issue. Regards Colm A Greg Chicares wrote: On 2006-7-11 10:20 UTC, Colm Aengus Murphy wrote: We are using Paul Smith's "Advanced Auto-Dependancy Generation" (http://make.paulandlesley.org/autodep.html) to great effect. However as the number of source files increase make slows down. The reason is that when you launch make it reads in an make file for each source file. This can take some time. Then it tried to update them. I have added a rule to speed up its decision to remake them but it still slows make down. What rule did you add? I'm using -include *.d *.d:: ; but I don't think I ever actually looked into whether the double-colon rule has any real effect, so I'd be glad to learn the "right" way to do this. Is there any way to disable the updating of makefiles altogether ? Has anyone else solved the scaleability issue with having a make file for each source file ? The speed issue is most apparent on windows when make can take 20 seconds before it even begins to check if it needs to update any object files. I don't see anything like a twenty-second overhead on the msw platform, but I only have a few hundred source files. Do you have more than that? My '.d' files average about 20K each; are yours larger? If you're using '-MD' instead of '-MMD' with gcc tools to generate the dependency files, that might make things a lot slower. We are already running with make -R to avoid make trying its own builtin rules (as an aside this option doesn't work in a MAKEFLAGS variable for some reason). The manual says: | You can also set MAKEFLAGS in a makefile, to specify additional flags | that should also be in effect for that makefile. Can you post a testcase that shows that to be incorrect? -- --------------------------------------------------------------------- Colm Aengus Murphy, Tel : +353 1 2911000 Senior Hardware Design Engineer, Direct Tel: +353 1 2911373 Silicon & Software Systems, Fax : +353 1 2911001 South County Business Park, Leopardstown, E-mail: [EMAIL PROTECTED] Dublin 18. WWW : www.s3group.com Ireland --------------------------------------------------------------------- The information contained in this e-mail and in any attachments is confidential and is designated solely for the attention of the intended recipient(s). If you are not an intended recipient, you must not use, disclose, copy, distribute or retain this e-mail or any part thereof. If you have received this e-mail in error, please notify the sender by return e-mail and delete all copies of this e-mail from your computer system(s). Please direct any additional queries to: [EMAIL PROTECTED] Thank You.

_______________________________________________
Help-make mailing list
Help-make@gnu.org
http://lists.gnu.org/mailman/listinfo/help-make