Wireshark mailing list archives
Re: [Wireshark-commits] rev 35393: /trunk/epan/dissectors/ /trunk/epan/dissectors/: packet-catapult-dct2000.c
From: Jeff Morriss <jeff.morriss.ws () gmail com>
Date: Tue, 11 Jan 2011 21:33:41 -0500
On 01/07/2011 05:52 PM, Jeff Morriss wrote:
Martin Mathieson wrote:After remembering that profiling (at least in its easiest form with the 'time' command) isn't so hard, I played around a bit. After building a decent-sized dct2000 file (taking the sample from SampleCaptures and merging it with itself until I had a 276 Mb file), I tried before and after this change and I can't find a measurable difference in the CPU usage. I even tried forcing my (AMD) CPU down to 1 GHz to exaggerate the difference, but I still got only a couple of seconds CPU time difference out of over 5 minutes--and in that case rev 35393's code was faster. Maybe I'll try tomorrow on a SPARC: I know that memcpy()s are a lot more expensive there than on x86. I think you win, the difference isn't worth it and it'd be better not to leave unnecessary examples of tvb_get_ptr() use around.I tried with SPARC today and I do see a consistent 0.9% difference in CPU time before and after 35393. (In particular I see about 6-7 seconds of extra CPU time in a tshark job that takes around 11 minutes and 50 seconds.) Don't know if that difference is significant enough to matter.
(After entirely too much time trying to find a way to rebuild my NAS' disk array--shaking in my boots the whole time--I finally got back to this.)
I eventually decided to go ahead and check in tvb_get_const_stringz(): when using tshark I could eventually (once I got the capture file up to about 1 Gb) detect a measurable amount of CPU time saved, even on x86. (Before I had been using the GUI but the CPU usage seemed to vary too much.) Since there doesn't seem to be any harm in it and it's more efficient, I figured why not...
___________________________________________________________________________ Sent via: Wireshark-dev mailing list <wireshark-dev () wireshark org> Archives: http://www.wireshark.org/lists/wireshark-dev Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev mailto:wireshark-dev-request () wireshark org?subject=unsubscribe
Current thread:
- Re: [Wireshark-commits] rev 35393: /trunk/epan/dissectors/ /trunk/epan/dissectors/: packet-catapult-dct2000.c Martin Mathieson (Jan 06)
- Re: [Wireshark-commits] rev 35393: /trunk/epan/dissectors/ /trunk/epan/dissectors/: packet-catapult-dct2000.c Jeff Morriss (Jan 06)
- Re: [Wireshark-commits] rev 35393: /trunk/epan/dissectors/ /trunk/epan/dissectors/: packet-catapult-dct2000.c Martin Mathieson (Jan 06)
- Re: [Wireshark-commits] rev 35393: /trunk/epan/dissectors/ /trunk/epan/dissectors/: packet-catapult-dct2000.c Anders Broman (Jan 06)
- Re: [Wireshark-commits] rev 35393: /trunk/epan/dissectors/ /trunk/epan/dissectors/: packet-catapult-dct2000.c Jeff Morriss (Jan 06)
- Re: [Wireshark-commits] rev 35393: /trunk/epan/dissectors/ /trunk/epan/dissectors/: packet-catapult-dct2000.c Martin Mathieson (Jan 06)
- Re: [Wireshark-commits] rev 35393: /trunk/epan/dissectors/ /trunk/epan/dissectors/: packet-catapult-dct2000.c Jeff Morriss (Jan 06)
- Re: [Wireshark-commits] rev 35393: /trunk/epan/dissectors/ /trunk/epan/dissectors/: packet-catapult-dct2000.c Martin Mathieson (Jan 07)
- Re: [Wireshark-commits] rev 35393: /trunk/epan/dissectors/ /trunk/epan/dissectors/: packet-catapult-dct2000.c Jeff Morriss (Jan 07)
- Re: [Wireshark-commits] rev 35393: /trunk/epan/dissectors/ /trunk/epan/dissectors/: packet-catapult-dct2000.c Jeff Morriss (Jan 11)
- Re: [Wireshark-commits] rev 35393: /trunk/epan/dissectors/ /trunk/epan/dissectors/: packet-catapult-dct2000.c Martin Mathieson (Jan 06)
- Re: [Wireshark-commits] rev 35393: /trunk/epan/dissectors/ /trunk/epan/dissectors/: packet-catapult-dct2000.c Jeff Morriss (Jan 06)