Contributed a patch (driver-core) to the driver core)

I have been busy recently, so I did not care about writing logs.

Starting from the last time, UBUNTU 4.28 came out in 11.04 and changed to a system without live writer. Therefore, sometimes I cannot remember to write a blog by myself. 11.04 is very easy to use. First of all, I like the sidebar and the toolbar, which greatly improves the desktop usage and looks at the code. Compared with using a virtual machine on Windows, it is too convenient to change the code and issue patches. Haha.

Then there were a few small patches that were applied before, so I didn't need to talk about it. Later I was busy with writing the kerneldoc of driver-core for about a week. From that time.

On that day, I took a look at the kernel document. When I saw the driver-model, I found that those documents had expired for a long time. Even the data structure was quite different from the current driver-core, so I changed them, changed them for several days, and then sent a patch. After the result was sent, Greg said that such a thing should actually be included in the kernel Doc to facilitate maintenance, let me move them over. I have never played with kerneldoc before. I only replied to I'll try. Then let's look at the kerneldoc and try to modify it. then compile the kerneldoc in the device. in H, the data structures of several core components, such as bus_type, class, device, and device_driver, are not clear about member comments and explanations. I can only write them in my own understanding, after reading some of the previous modification records of driver-core, there are also a lot of development directions of driver-core written by Greg in the future kernel. A patch is sent when it is almost done, greg helped to change the syntax, then sent it again, and then changed it. Finally, it was almost the same as V4. Finally, Randy helped to change the syntax. There are still some comments, sent a V5, and finally sent the patches of those documents removed from documentation to Greg. Later, Greg added them to the driver-core tree.

Although they are troublesome, they have gained a lot. On the one hand, they have deepened their understanding of driver-core, and on the other hand, although they only contributed dozens of lines of comments, however, it is also the largest patch for yourself, cheers.

Http://git.kernel.org /? P = linux/kernel/git/gregkh/driver-core-2.6.git; a = commitdiff; h = 880ffb5c6c5c8c8c6efd9efe9355317322b4603b

Http://git.kernel.org /? P = linux/kernel/git/gregkh/driver-core-2.6.git; a = commitdiff; h = 63dc355a5a8cf296e2b1cc2e4192108dca221129

PS: I am talking about a problem when I used the tcard download program in uboot. Previously I used hynix nand flash, and sometimes some board downloads often encountered the re-write problem, I modified the timing sequence and changed the timing sequence to a large one. Then I found the problem disappeared, but then I encountered another problem with the board. It would not help me to change the timing sequence again, I checked this problem for a long time. Later, I accidentally found that the problem was missing in the new version. After several days of back-to-version and verification, I found that an engineer had previously increased the core voltage by 100mV, as a result, I increased the kernel voltage of the problematic version by. The problem was indeed gone. The previous problems all occurred when the core voltage was too low, and the voltage of some board cores was unstable, as a result, an error occurred while writing the nand. I will try another hardware engineer tomorrow to test the core voltage to see if it is unstable. We are lucky to solve this problem.