Minor updates to comments and README files
This commit is contained in:
parent
d5e7644e23
commit
45f46b4815
@ -1358,9 +1358,9 @@ NAND Support
|
||||
verified, but there is no way to use it in the current NuttX architecture
|
||||
other than through the raw MTD interface.
|
||||
|
||||
NAND should be considered a work in progress. You will not want to use
|
||||
NAND unless you are interested in investing a little effort, particularly
|
||||
in infrastructure. See the "STATUS SUMMARY" section below.
|
||||
NAND should still be considered a work in progress. You will not want to
|
||||
use NAND unless you are interested in investing a little effort,
|
||||
particularly in infrastructure. See the "STATUS SUMMARY" section below.
|
||||
|
||||
NAND Support
|
||||
------------
|
||||
@ -1652,13 +1652,12 @@ NAND Support
|
||||
work correctly: The MBR is written and read back correctly, but gets
|
||||
corrupted later for unknown reasons.
|
||||
|
||||
2. DMA works (with software ECC), but I have seen occasional failurs. I
|
||||
believe that these issue have been resolved but I recommend enabling
|
||||
DMA with caution.
|
||||
2. DMA works (at least with software ECC), but I have seen occasional
|
||||
failures. I recommend enabling DMA with caution.
|
||||
|
||||
In NuttX, DMA will also cost two context switches (and, hence, four
|
||||
register state transfers). With smaller NAND page sizes (say 2KiB and
|
||||
below), I would not expect a great performance improvement with DMA
|
||||
below), I would expect little or no performance improvement with DMA
|
||||
for this reason.
|
||||
|
||||
3. NXFFS does not work with NAND. NAND differs from other other FLASH
|
||||
|
Loading…
Reference in New Issue
Block a user