Greetings,
Enclosed you'll find a link to the agenda, minutes and actions from the Linaro kernel working group weekly meeting of Dec 06, 2010.
https://wiki.linaro.org/WorkingGroups/KernelConsolidation/Meetings/2010-12-0...https://wiki.linaro.org/WorkingGroups/KernelConsolidation/Meetings/2010-11-29
== Summary == * Dave Martin & Nico reviewed K2.x blueprints, bp's are Approved now. * Tool to measure and fix SD card performance. Found a case of excessive copy-and-paste, gaining unneeded abstraction layers not yet needed for this non-ARM Unicore architecture. * Mainlined Flash support, available in git.stericsson.org, also taken into the linux-arm-kernel tree 2.6.37. * Implemented USB transceiver driver as drivers/usb/otg. Cleaning up MUSB BSP for 8500 crossover code. Gadget code sent out for internal ST-Ericsson review, will post publicly after resolved. * Posted the five uboot patches to the list, built and tested the first two on all available platforms. Need additional testing, especially on big-endian platforms such as PowerPC. * Reviewing status. A few specifications and blueprints not yet complete, status lagging on some. Working on New Members wiki page. * ST-Ericsson Landing team going well. Finding problems, getting fixes. Waiting on some fixes from internal ST-Ericsson teams. Currently relying on to-be-written functionality. Team is under NDA, so can pull some patches from internal teams prior to upstreaming. Some issues with source-tree organization within ST-Ericsson, which is being resolved by having a single tree that functionality is added to via topic branches. * Thumb 2 work progressing. A couple of patches to be merged into 2.6.36 tree. Freescale BSP review progress. * OMAP4-EHCI support queued up in linux-usb tree for 2.6.38 * Document describing Linaro kernel, will publish on list and ask for comments. * Completed blocked-task-list reorganization for RCU_TREE priority boosting, and started coding up changes needed to move from softirq to kthread. Possible RCU energy-consumption issue appearing on lightly loaded systems with modest numbers of CPUs: RCU prevents entry into dyntick-idle mode: Examining possible approaches.
Please let me know if you have a question or comment,
Regards, Mounir