Messages in this thread | | | Subject | RE: Slow down across kernels | Date | Mon, 24 May 2004 10:12:14 -0400 | From | "Piszcz, Justin Michael" <> |
| |
This is unreal! Have you tried 2.6.5 by the way? What is it you are benchmarking if you can say?
-----Original Message----- From: linux-kernel-owner@vger.kernel.org [mailto:linux-kernel-owner@vger.kernel.org] On Behalf Of Phy Prabab Sent: Sunday, May 23, 2004 3:24 PM To: linux-kernel@vger.kernel.org Subject: Slow down across kernels
Hello,
Please cc me as I am not on this mailing list.
I have an issue with the 2.6 kernels that I am unable to understand and correct.
I have a custom binary that I have used to compile across different kernel versions and I am getting vildly different numbers: (as reported by time):
2.4.21-9 ~45s 2.4.22 ~55s 2.6.1 ~3m.40s 2.6.2 ~4m.00 2.6.3 ~4m.00 2.6.6 ~3m.15s 2.6.6-mm4 ~1m.30s
It is most shocking how much more time is spent in user/kernel in comparsion: 2.4.21-9
5.09user 9.04system 0:43.79elapsed 32%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (7378946major+1139090minor)pagefaults 0swaps
VS.
2.6.6-mm4
13.06user 23.80system 1:32.00elapsed 40%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+7537027minor)pagefaults 0swaps
The system is AMD Opteron 246 w/12G RAM and broadcom Ge. To rule out network issues, I have moved the files to the local files system, ext2 on a UW320 drive.
I have also tested this on a dual Xeon 3.06GHz w/8G RAM ge with very similar results in differences across the kernels.
strace and make -d show no major differences.
Kernels are built with gcc323/340 and bintuils 2.13.2.1.
Could someone help me understand what is at issue here?
Thank you and appreciate the help. Phy
__________________________________ Do you Yahoo!? Yahoo! Domains - Claim yours for only $14.70/year http://smallbusiness.promotions.yahoo.com/offer - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |