Comments for John Ford http://blog.johnford.org the things that I remembered to write about Sat, 04 Feb 2012 01:04:01 +0000 hourly 1 http://wordpress.org/?v=3.3 Comment on Writting a native rm program for Windows by John Ford http://blog.johnford.org/writting-a-native-rm-program-for-windows/comment-page-1/#comment-6504 John Ford Sat, 04 Feb 2012 01:04:01 +0000 http://blog.johnford.org/?p=380#comment-6504 absolutely no clue, but I found that just rmdir wasn't enough. absolutely no clue, but I found that just rmdir wasn’t enough.

]]> Comment on Writting a native rm program for Windows by Neil Rashbrook http://blog.johnford.org/writting-a-native-rm-program-for-windows/comment-page-1/#comment-6499 Neil Rashbrook Thu, 02 Feb 2012 11:23:27 +0000 http://blog.johnford.org/?p=380#comment-6499 I have never seen <code>rd /s /q</code> fail in the way you describe... why is the build system creating readonly/hidden/system files anyway? I have never seen rd /s /q fail in the way you describe… why is the build system creating readonly/hidden/system files anyway?

]]> Comment on Writting a native rm program for Windows by Alfred Kayser http://blog.johnford.org/writting-a-native-rm-program-for-windows/comment-page-1/#comment-6498 Alfred Kayser Wed, 01 Feb 2012 21:00:49 +0000 http://blog.johnford.org/?p=380#comment-6498 Couple of nits: 1. When the del_directory fails, it will both report the error and tell us that the directory has been deleted (in verbose mode). In the del_file there is the same problem. 2. Reversing arguments is not nice, one could also instead of using a linked list, allocate an array and put the filenames in there (the length of the array is at most argc) (even if with many options, using an array is simpler and smaller than the linked list). Couple of nits:
1. When the del_directory fails, it will both report the error and tell us that the directory has been deleted (in verbose mode). In the del_file there is the same problem.
2. Reversing arguments is not nice, one could also instead of using a linked list, allocate an array and put the filenames in there (the length of the array is at most argc) (even if with many options, using an array is simpler and smaller than the linked list).

]]> Comment on Writting a native rm program for Windows by John Ford http://blog.johnford.org/writting-a-native-rm-program-for-windows/comment-page-1/#comment-6497 John Ford Wed, 01 Feb 2012 19:32:33 +0000 http://blog.johnford.org/?p=380#comment-6497 Heh, I do have a copy of the mozilla-build builder vm and have been looking for a reason to learn how to make a release of mozilla-build. Heh, I do have a copy of the mozilla-build builder vm and have been looking for a reason to learn how to make a release of mozilla-build.

]]> Comment on Writting a native rm program for Windows by John Ford http://blog.johnford.org/writting-a-native-rm-program-for-windows/comment-page-1/#comment-6495 John Ford Wed, 01 Feb 2012 19:30:27 +0000 http://blog.johnford.org/?p=380#comment-6495 Sid, the benefit is that this is available outside of makefiles. I am also not sure how pymake natives work, but I heard they are done with a new python interpreter. If that's the case, this program would have lower startup overhead and likely slightly faster execution. Sid, the benefit is that this is available outside of makefiles. I am also not sure how pymake natives work, but I heard they are done with a new python interpreter. If that’s the case, this program would have lower startup overhead and likely slightly faster execution.

]]> Comment on Writting a native rm program for Windows by Sid http://blog.johnford.org/writting-a-native-rm-program-for-windows/comment-page-1/#comment-6494 Sid Wed, 01 Feb 2012 19:17:50 +0000 http://blog.johnford.org/?p=380#comment-6494 Does this have significant benefits over using pymake's native commands? Does this have significant benefits over using pymake’s native commands?

]]> Comment on Writting a native rm program for Windows by Dave Townsend http://blog.johnford.org/writting-a-native-rm-program-for-windows/comment-page-1/#comment-6493 Dave Townsend Wed, 01 Feb 2012 18:08:04 +0000 http://blog.johnford.org/?p=380#comment-6493 Wonderful! Can we upstream this to msys at all? Or at least get it in MozillaBuild? Wonderful! Can we upstream this to msys at all? Or at least get it in MozillaBuild?

]]> Comment on Evaluating new Mac Mini builders, SSDs and -j settings by Dan http://blog.johnford.org/new-mac-builders-ssds-j-settings/comment-page-1/#comment-6463 Dan Thu, 12 Jan 2012 03:05:38 +0000 http://blog.johnford.info/?p=350#comment-6463 "My data also suggests that the formula is (cores – 1) * 4" As glandium said, you can't really say that given that you didn't test with all intermediate values of -j. I agree though that this gives a good enough idea. Once you have chosen the hardware (especially if you go with the quad-core), I'd definitely be interested to see a graph of -j from 1 to 12 (including all values in between) for that machine. “My data also suggests that the formula is (cores – 1) * 4″

As glandium said, you can’t really say that given that you didn’t test with all intermediate values of -j. I agree though that this gives a good enough idea. Once you have chosen the hardware (especially if you go with the quad-core), I’d definitely be interested to see a graph of -j from 1 to 12 (including all values in between) for that machine.

]]> Comment on Evaluating new Mac Mini builders, SSDs and -j settings by IVB http://blog.johnford.org/new-mac-builders-ssds-j-settings/comment-page-1/#comment-6462 IVB Wed, 11 Jan 2012 20:07:40 +0000 http://blog.johnford.info/?p=350#comment-6462 Mozilla should wait for Ivy Bridge powered Mac Minis. Ivy Bridge uses less power(77w TDP for quad-cores vs Sandy Bridge 95w TDP quad-cores) and between 10-20% more performance. http://www.anandtech.com/show/4830/intels-ivy-bridge-architecture-exposed Mozilla should wait for Ivy Bridge powered Mac Minis. Ivy Bridge uses less power(77w TDP for quad-cores vs Sandy Bridge 95w TDP quad-cores) and between 10-20% more performance.

http://www.anandtech.com/show/4830/intels-ivy-bridge-architecture-exposed

]]> Comment on Evaluating new Mac Mini builders, SSDs and -j settings by John Ford http://blog.johnford.org/new-mac-builders-ssds-j-settings/comment-page-1/#comment-6461 John Ford Wed, 11 Jan 2012 19:56:48 +0000 http://blog.johnford.info/?p=350#comment-6461 Nope, we don't use the Mac Minis for building Windows or Linux. We currently have all of Windows and Linux testing on Mac Minis, but we have plans to change that. Nope, we don’t use the Mac Minis for building Windows or Linux. We currently have all of Windows and Linux testing on Mac Minis, but we have plans to change that.

]]>