* Lubos Lunak
Well the "claim" acutally comes from the bash maintaines, read bash(1) BUGS. One example is startup time, try libmicro's system benchmark /usr/lib/libMicro/bin/system -E -C 200 -L -S -W -N "system" -I 1000000 with /bin/sh as a link to ksh or dash and empty .profile/ENV. dash is 2,5 times and ksh93 is still 1,5 times faster. I have anecdotal evidence that pattern matching while processing textfiles is significantly slower in bash compared to ksh93/dash.
Speaking of anecdotes, there is one I remember from the times of the communist regime about plans set up by the party: "Our farm has completed the 5-year plan at 200%. We have four chickens instead of two."
Another area in which bash is bad at and which matters even more are subshells, for a rough estimate try time $shell -c 'i=0; while [ $i -lt 10000 ]; do $(a=$$); i=$((i+1)); done' with bash, dash, and ksh93. Here dash is 3 times faster and ksh93 37 times faster than bash (ksh93 shines here because it does not actually use a subprocess).
Let me give you a piece of advice: If you want to argue by technical arguments, then do so. "37 times faster", without anything else, is like "hair 87% more shiny" ads. If better performance should be a reason to avoid bashism, then say how big improvements you actually expect, and provide some real numbers to support that. "37 times faster" is completely unimpressive if it in practice may mean that this change will save quarter of a second of boot time.
And your point is? Firstly, the original question was whether there is evidence that bash is slow and where that matters, I think I have answered that. Secondly, this is not about boot speed, shell scripting has more use cases than simple boot scripts. And thirdly, speed advantages were only one, not the main rationale for my proposal. -- Guido Berhoerster -- To unsubscribe, e-mail: opensuse-packaging+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-packaging+help@opensuse.org