| Summary: | [PATCH] correct nanddump behaviour | ||
|---|---|---|---|
| Product: | Busybox | Reporter: | richard.genoud |
| Component: | Standard Compliance | Assignee: | unassigned |
| Status: | NEW --- | ||
| Severity: | normal | CC: | busybox-cvs, volchok |
| Priority: | P5 | ||
| Version: | 1.18.x | ||
| Target Milestone: | --- | ||
| Hardware: | Other | ||
| OS: | Linux | ||
| Host: | Target: | ||
| Build: | |||
| Attachments: |
patch correcting nanddump behaviour with bad blocks
[PATCH 1/2] nanddump: correct rounding to next page [PATCH 2/2] nanddump: dump rigth length when jumping bad blocks |
||
Comment on attachment 4076 [details]
patch correcting nanddump behaviour with bad blocks
patch forward ported on 1.21.0
Created attachment 4850 [details]
[PATCH 1/2] nanddump: correct rounding to next page
Created attachment 4856 [details]
[PATCH 2/2] nanddump: dump rigth length when jumping bad blocks
For an easier review, the original patch has been forward ported to v1.21.0 and split in 2 functional patches. Patch 1: Correct wrong page rounding formula on nanddump causing an (almost) infinite loop when dumping bad blocks. Patch 2: with the "skip bad blocks" options, nanddump is not dumping the right len when there's bad bocks. (It is dumping len - bad_block_number * erase_size). It should skip bad blocks and still dump the requested size. Richard. |
Created attachment 4076 [details] patch correcting nanddump behaviour with bad blocks There's an infinite loop in nanddump when dumping a bad block. Moreover, bad blocks are dumped even if the -b option is given. The attached patch corrects those points: - dump the specified length when -b is given - do not dump the bad blocks when -b is given - when dumping a bad block, len should be aligned on next page aligned only if it's not already page aligned in order to dump the right size.