digitalmars.D - Clang with SAFECode
- Jacob Carlborg (6/6) Aug 18 2011 "We have a new release of Clang with SAFECode technology for detecting
- Andrei Alexandrescu (5/9) Aug 18 2011 Thanks for the info. Walter and my perception is that safety is big and
- bearophile (6/8) Aug 18 2011 becoming huge, so SafeD was a good bet (props to Bartosz Milewski for
- Walter Bright (3/4) Aug 19 2011 Very little code in a D program needs to be @system, so I see little
- bearophile (4/5) Aug 20 2011 I'd like to know what percentage of code of a high-performance indie vid...
- Walter Bright (2/4) Aug 22 2011 You're not going to use those runtime checks in high performance video c...
- bearophile (4/5) Aug 22 2011 I see it as infrastructure useful for debugging and code test. The advan...
- Josh Simmons (6/11) Aug 22 2011 Not to mention while games are performance sensitive, unless it's an
"We have a new release of Clang with SAFECode technology for detecting memory safety errors." I thought that someone might be interested: http://clang-developers.42468.n3.nabble.com/Clang-SAFECode-Release-Announcement-td3265143.html -- /Jacob Carlborg
Aug 18 2011
On 8/18/11 12:41 PM, Jacob Carlborg wrote:"We have a new release of Clang with SAFECode technology for detecting memory safety errors." I thought that someone might be interested: http://clang-developers.42468.n3.nabble.com/Clang-SAFECode-Release-Announcement-td3265143.htmlThanks for the info. Walter and my perception is that safety is big and becoming huge, so SafeD was a good bet (props to Bartosz Milewski for first proposing it). Andrei
Aug 18 2011
Andrei Alexandrescu:http://clang-developers.42468.n3.nabble.com/Clang-SAFECode-Release-Announcement-td3265143.htmlThanks for the info. Walter and my perception is that safety is big andbecoming huge, so SafeD was a good bet (props to Bartosz Milewski for first proposing it).< The purpose of that SAFECode is very different from the purpose of D safe: it's a tool to help find bugs in unsafe code, while safe merely disallows some memory-unsafe operations. I presume both things are needed, because sometimes you need to write unsafe code too. I'd like some optional functionality like SAFECode in the D compiler too. Bye, bearophile
Aug 18 2011
bearophile wrote:I'd like some optional functionality like SAFECode in the D compiler too.Very little code in a D program needs to be system, so I see little need for this.
Aug 19 2011
Walter:Very little code in a D program needs to be system, so I see little need for this.I'd like to know what percentage of code of a high-performance indie video game is system :-) Bye, bearophile
Aug 20 2011
On 8/20/2011 11:48 PM, bearophile wrote:You're not going to use those runtime checks in high performance video code.Very little code in a D program needs to be system, so I see little need for this.I'd like to know what percentage of code of a high-performance indie video game is system :-)
Aug 22 2011
Walter:You're not going to use those runtime checks in high performance video code.I see it as infrastructure useful for debugging and code test. The advantage over Valgrind is that it doesn't slow down the code 10 times, but only 1.2-1.5 times, so you are able to play the game during debugging too. Bye, bearophile
Aug 22 2011
On Mon, Aug 22, 2011 at 6:11 PM, bearophile <bearophileHUGS lycos.com> wrote:Walter:Not to mention while games are performance sensitive, unless it's an AAA title it's probably not _that_ sensitive in the majority of areas. Most games are loaded with assertions in debug mode anyway. Cheers, JoshYou're not going to use those runtime checks in high performance video code.I see it as infrastructure useful for debugging and code test. The advantage over Valgrind is that it doesn't slow down the code 10 times, but only 1.2-1.5 times, so you are able to play the game during debugging too. Bye, bearophile
Aug 22 2011