Timur Tabi wrote:
Tolunay Orkun wrote:
Here is the patch (attached) that handles top boot geometry reversal case on AMD flash.
This patch works for me, so I approve it. I tested it on the board that has this problem, as well as an 8360-based board that has only same-sized sectors.
Glad to hear. Hopefully I will get more positive feedback.
Just one nit:
Adds trailing whitespace. .dotest/patch:35: man_id = flash_read_uchar (info, Adds trailing whitespace. .dotest/patch:37: dev_id = flash_read_uchar(info, warning: 2 lines add trailing whitespaces.
I will fix that.
Questions for the maintainers:
1) I have some new variables for man_id, dev_id, cfi_version introduced by this patch (actually man_id is currently not used so could be removed). I can kept these local to the function for now or I can add them to the info structure along with ext_addr (to easily locate the Vendor Extended Query Structure in flash). What do you think?
In the minimum displaying man_id, dev_id, cfi_version when executing "flinfo" command could be useful for diagnostic purposes. We might need to use them to tweak the behavior of some other functionality in the future as well.
2) There are some outdated comments at the beginning of the cfi_flash.c. The modification history is not up to date and todo list stuff is mostly addressed over time. Since the actual history is already is the git, I propose to get rid of modification history and todo list from the comments.
Best regards, Tolunay