search for: striclty

Version: 1040 OS: Windows 98 SE Problem: I have installed R with all help files to a network read-only drive. help.start() requires rw on two files ../R/rw1040/doc/html/packages.html ../R/rw1040/doc/html/search/index.txt This is a minor issue as running R 1.4.0 help from the start menu works fine. Perhaps the next version may address this? Simon Bowman Systems Adminstrator University of Bath

I'm a neewbie in Linux, so please bear with me. I have a school assignment to make communication between 10 SIP softphones (kphone). So far I got trouble installing Asterisk. The information in asterisk web site seems to be a bit outdated because it's mentioned only kernel 2.4. Since Mandrake cooker (10.2beta3) witch I'm using is using kernel 2.6.10, I wander if I have a chance to

...it fail. In a 2nd step it can do the same thing by replacing functions in a "good" with functions from a "bad" file. In case of a compiler revisions that triggers a miscompile this is usually enough to track down the function that is miscompiled. Andys proposed scheme should be striclty more powerfull though as it is robust against different optimization decisions between the two compilations and even allows you to track down the pass that introduced the failure, but maybe my script is useful for somebody else in the meantime. - Matthias > On Mar 25, 2016, at 4:05 PM, Michae...

...an do the same thing by replacing functions in a "good" with functions from a "bad" file. >> In case of a compiler revisions that triggers a miscompile this is usually enough to track down the function that is miscompiled. >> >> Andys proposed scheme should be striclty more powerfull though as it is robust against different optimization decisions between the two compilations and even allows you to track down the pass that introduced the failure, but maybe my script is useful for somebody else in the meantime. > > Yes these sorts of scripts are useful (I ha...

...nd step it can do the same thing by replacing functions in a "good" with functions from a "bad" file. > In case of a compiler revisions that triggers a miscompile this is usually enough to track down the function that is miscompiled. > > Andys proposed scheme should be striclty more powerfull though as it is robust against different optimization decisions between the two compilations and even allows you to track down the pass that introduced the failure, but maybe my script is useful for somebody else in the meantime. Yes these sorts of scripts are useful (I have written...

...me thing by replacing functions in a "good" with functions from a "bad" file. >>> In case of a compiler revisions that triggers a miscompile this is usually enough to track down the function that is miscompiled. >>> >>> Andys proposed scheme should be striclty more powerfull though as it is robust against different optimization decisions between the two compilations and even allows you to track down the pass that introduced the failure, but maybe my script is useful for somebody else in the meantime. >> >> Yes these sorts of scripts are usef...

...it fail. In a 2nd step it can do the same thing by replacing functions in a "good" with functions from a "bad" file. In case of a compiler revisions that triggers a miscompile this is usually enough to track down the function that is miscompiled. Andys proposed scheme should be striclty more powerfull though as it is robust against different optimization decisions between the two compilations and even allows you to track down the pass that introduced the failure, but maybe my script is useful for somebody else in the meantime. Yes these sorts of scripts are useful (I have written...

...it fail. In a 2nd step it can do the same thing by replacing functions in a "good" with functions from a "bad" file. In case of a compiler revisions that triggers a miscompile this is usually enough to track down the function that is miscompiled. Andys proposed scheme should be striclty more powerfull though as it is robust against different optimization decisions between the two compilations and even allows you to track down the pass that introduced the failure, but maybe my script is useful for somebody else in the meantime. Yes these sorts of scripts are useful (I have written...

I will describe the complete process for completeness thus hopefully forestalling all questions [tell me if I did not ; )]. There is not much to it TBH. ./utils/bisect is a dumb python script that allows for arbitrary bisecting via the exit status of a script it runs . The way you use it is you write a script (lets call it test.sh). Then you invoke: ./utils/bisect --start=N --end=M ./test.sh

> In the swift-world we use utils/bisect + a single number all the time + extra verifications. It works really well. Can you describe to me what you mean by that exactly? Are you using the single number in the LLVM back end or somewhere else? From: llvm-dev [mailto:llvm-dev-bounces at lists.llvm.org] On Behalf Of Michael Gottesman via llvm-dev Sent: Friday, March 25, 2016 12:35 PM To: