[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Chicken-users] Using -accumulate-profile
From: |
Felix |
Subject: |
Re: [Chicken-users] Using -accumulate-profile |
Date: |
Tue, 18 Jan 2011 09:36:55 +0100 (CET) |
From: Taylor Venable <address@hidden>
Subject: [Chicken-users] Using -accumulate-profile
Date: Mon, 17 Jan 2011 21:58:12 -0500
> Hi there, I'm building a program with csc and it's running slower than I'd
> hoped so I'm trying to do some profiling. I tried using the
> -accumulate-profile option to concatenate information from multiple runs of
> the program but it seems to produce the same result as just using -profile
> (i.e. multiple PROFILE.<pid> files, each containing information from a
> single run). I thought maybe I had to explicitly set the profile file name
> using -profile-name but that doesn't help (it changes the stem (i.e.
> <name>.<pid> with -profile-name <name>) but still multiple executions create
> multiple files). I'm using 4.6.3 on 64-bit Linux: "linux-unix-gnu-x86-64 [
> 64bit manyargs dload ptables ]". I had thought from reading the doc that
> using -accumulate-profile would build up a single profile output file
> containing information for numerous successive executions; is that not
> correct? Thanks for any advice (pointers to any general tips on profiling
> and optimizing in Chicken also greatly appreciated).
>
This was indeed broken. I have checked in a fix in the "experimental"
branch - `-accumulate-profile' now requires that `-profile-name' is
used. This name given will be used as the target for the profile
data (without appending the PID, the name is used unchanged).
Thanks for reporting this problem.
cheers,
felix