What does it cost to try?

[Steven T. Hatton]

What is the runtime cost of wrapping a function call in a try/catch? I
assume it's something like an if/else when it goes on the stack. I believe
that means an additional compare and an additional memory location to
branch to. Would this be significant enough to make a difference between
say putting the try{}catch{} inside the body of a loop as opposed to
wrapping the loop in the try/catch?
--
STH
Hatton's Law: "There is only One inviolable Law"
KDevelop: http://www.kdevelop.org SuSE: http://www.suse.com
Mozilla: http://www.mozilla.org

[David Hilsee]

"Steven T. Hatton" <su******@setidava.kushan.aa> wrote in message
news:2v********************@speakeasy.net...

What is the runtime cost of wrapping a function call in a try/catch? I
assume it's something like an if/else when it goes on the stack. I believe that means an additional compare and an additional memory location to
branch to. Would this be significant enough to make a difference between
say putting the try{}catch{} inside the body of a loop as opposed to
wrapping the loop in the try/catch?

This is obviously implementation-dependent. The g++ compiler even offers
flags to control the "cost" of using exceptions
(http://gcc.gnu.org/onlinedocs/gnat_u...g-Control.html)
.. You might get better answers in a newsgroup that discusses your
implementation.

--
David Hilsee

[David Hilsee]

"David Hilsee" <da*************@yahoo.com> wrote in message
news:Cs********************@comcast.com...

"Steven T. Hatton" <su******@setidava.kushan.aa> wrote in message
news:2v********************@speakeasy.net...

What is the runtime cost of wrapping a function call in a try/catch? I
assume it's something like an if/else when it goes on the stack. I believe

that means an additional compare and an additional memory location to
branch to. Would this be significant enough to make a difference between say putting the try{}catch{} inside the body of a loop as opposed to
wrapping the loop in the try/catch?

This is obviously implementation-dependent. The g++ compiler even offers
flags to control the "cost" of using exceptions

(http://gcc.gnu.org/onlinedocs/gnat_u...g-Control.html) . You might get better answers in a newsgroup that discusses your
implementation.

Sorry, I linked to an Ada compiler, but the message is still the same.

--
David Hilsee