• Open Menu Close Menu
  • Apple
  • Shopping Bag
  • Apple
  • Mac
  • iPad
  • iPhone
  • Watch
  • TV
  • Music
  • Support
  • Search apple.com
  • Shopping Bag
 

Lists

Open Menu Close Menu
  • Terms and Conditions
  • Lists hosted on this site
  • Email the Postmaster
  • Tips for posting to public mailing lists
Re: Shark attributing time oddly to source
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Shark attributing time oddly to source


  • Subject: Re: Shark attributing time oddly to source
  • From: Rick Altherr <email@hidden>
  • Date: Tue, 14 Apr 2009 16:36:32 -0700

For the archives: the code was being compiled with -ggdb. Compiling with -g instead resolved the problems.

Rick


On Apr 14, 2009, at 12:10 AM, Rick Altherr wrote:


On Apr 13, 2009, at 6:40 PM, Benson Margulies wrote:

Oddly, while Rick's post helped me see what's slow about this code, it
doesn't explain much about the results themselves. The function in
question really distributed all of its time to a set of static
functions called by function pointer, and so certainly not inlined.
None of them show up in the profile at all.



If you can reproduce this with a test case, please file a bug report.

I replaced the function pointery with calls to inlinable functions.
And now I'm a bit stumped because Shark has cached an old copy of the
source somewhere and I don't know how to get it to see the current
source.


Restart Shark. We are aware of this problem.

On Mon, Apr 13, 2009 at 9:29 PM, Greg Guerin <email@hidden> wrote:
Rick Altherr wrote:

- This function is called _very_ frequently but isn't inlined. Thus, the
time spent on the return is actually time spent in the epilog of the
function. You options are to reduce the number of calls to the function or
to inline it.
- The function runs very quickly but is also called frequently. In this
case, Nyquist might require a faster sampling rate. The default is 1ms, but
100us is still reasonable. If the profile changes when using 100us, you were
seeing aliasing.

Or maybe if the function is approximately synchronized to the sampling
timer?


-- GG

_______________________________________________
Do not post admin requests to the list. They will be ignored.
Xcode-users mailing list      (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden

_______________________________________________
Do not post admin requests to the list. They will be ignored.
Xcode-users mailing list      (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden

-- Rick Altherr Performance Tools Development Technologies email@hidden




_______________________________________________
Do not post admin requests to the list. They will be ignored.
Xcode-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
@apple.com


This email sent to email@hidden

-- Rick Altherr Performance Tools Development Technologies email@hidden




Attachment: smime.p7s
Description: S/MIME cryptographic signature

 _______________________________________________
Do not post admin requests to the list. They will be ignored.
Xcode-users mailing list      (email@hidden)
Help/Unsubscribe/Update your Subscription:

This email sent to email@hidden

References: 
 >Re: Shark attributing time oddly to source (From: Greg Guerin <email@hidden>)
 >Re: Shark attributing time oddly to source (From: Benson Margulies <email@hidden>)
 >Re: Shark attributing time oddly to source (From: Rick Altherr <email@hidden>)

  • Prev by Date: Re: Xcode 3.1.2 Crash
  • Next by Date: RE: Xcode 3.1.2 ( downloaded from iPhone SDK 9m2621afinal )Debugger bugged : Won't stop at break points ?
  • Previous by thread: Re: Shark attributing time oddly to source
  • Next by thread: Re: Shark attributing time oddly to source
  • Index(es):
    • Date
    • Thread