[Bug dyninst/23511] Weird slowness in the dyninst runtime when tracing simplest C programs

["agentzh at gmail dot com" <sourceware-bugzilla at sourceware dot org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=23511

--- Comment #4 from agentzh <agentzh at gmail dot com> ---
But even for long running process, it is not uncommon to run the same tool for
multiple times or run multiple different tools against the same target process.
And every time we run a tool, it costs more than 3 seconds of precious CPU time
(100% on a CPU core), which looks very unreasonable, which may well outweigh
the overhead of the real useful work of the tool itself. (I'd rather it sleeps
for 3 seconds without using so much CPU resources.)

Also, it looks unreasonable to have to do such expensive analyses on an *empty*
C program since there's really nothing worth analyzing.

Is it technically possible to cache and reuse the result of the "decompilation
analysis" thing?

-- 
You are receiving this mail because:
You are the assignee for the bug.