hi all, Can anyone please tell me the usage of 'loops_per_jiffy'. Where it is used?, how value of this variable affects the CPU usage and performance of a application. I am using a ARM11 (400 MHz, 32k I cache, 8 K D cache) with 2.6.11 kernel. caliberate_delay() function calculates loops_per_jiffy = 1331200, which turns out to be 266.8 BogoMips. Now i bridge TCP traffic from a wireless interface to ethernet, and i get throughput about 45Mbps at cpu usage of 95%. Whereas when i use same wireless and ethernet driver on a Xscale processor (400MHz, 32k I cache, 32K D cache, 399.6 BogoMips) using 2.6.13 kernel, i get throughput of 92Mbps at CPU usage of 25%. I tried to tweek the value of loops_per_jiffy manually (just for testing, i have no intensions of replacing calibration function :)) on my ARM11. For example setting loops_per_jiffiy value such tthat i get 140 BogoMips, the throughput incereased considerably to 66Mbps, but CPU usage remained around 85%. And using diffrent hard coded values of loops_per_jiffy i got diffrenent Throughput values and CPU usages. please tell me the reason for variation of throughput with diffrenet values loops_per_jiffy. Also if possible tell me what should be ideal value of loops_per_jiffy for the ARM11 with details i mentioned above. Also note that my ARM11 platform has 64M of SDRAM , and Xscale platform has 32k SDRAM. Thanks pankaj Send free SMS to your Friends on Mobile from your Yahoo! Messenger. Download Now! http://messenger.yahoo.com/download.php -- Kernelnewbies: Help each other learn about the Linux kernel. Archive: http://mail.nl.linux.org/kernelnewbies/ FAQ: http://kernelnewbies.org/faq/
