glibc is about to effectively turn off ASLR by default on x86_64 because some broken Intel chips are 3% slower. https://sourceware.org/ml/libc-alpha/2015-12/msg00221.html …
-
-
Replying to @RichFelker
@RichFelker@solardiz The patch only affects Silvermont. Anything else in unaffected.2 replies 1 retweet 1 like -
Replying to @gcpascutto
@gcpascutto@RichFelker@solardiz which is recent enough that this hardly matters. Anyone wanting to do it in their app can do it themselves1 reply 0 retweets 0 likes -
Replying to @bofh453
@bofh453@gcpascutto@solardiz You can't do it in your app because the relevant PROT_EXEC mmaps are done by the dynamic linker.1 reply 0 retweets 0 likes -
Replying to @RichFelker
@bofh453@gcpascutto@solardiz Also I don't see how it even works since the initial mmap of the main program is done by kernel & not 32-bit.1 reply 0 retweets 1 like -
Replying to @RichFelker
@RichFelker@gcpascutto@solardiz you're right... sneaking suspicion the perf improvement comes only from libs loaded by ld.so, incl. glibc.1 reply 0 retweets 0 likes -
Replying to @bofh453
@bofh453@gcpascutto@solardiz Only for calls from one such lib to another. Or from non-PIE main program.2 replies 0 retweets 0 likes
@bofh453 @gcpascutto @solardiz Calls internal to a single lib never benefit unless they're forced thru main program's PLT.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.