On 10/30/19 2:49 AM, Simon Glass wrote:
Hi Heinrich,
On Fri, 25 Oct 2019 at 04:15, Heinrich Schuchardt <xypron.g...@gmx.de> wrote:
The ext4 file system requires log2blksz to be set. So when setting the
block size on the block descriptor we should fill this field too.
This fixes a problem with EFI block devices providing ext4 partitions, cf.
https://lists.denx.de/pipermail/u-boot/2019-October/387702.html.
Signed-off-by: Heinrich Schuchardt <xypron.g...@gmx.de>
---
drivers/block/blk-uclass.c | 1 +
1 file changed, 1 insertion(+)
Reviewed-by: Simon Glass <s...@chromium.org>
But I wonder if blk_create_device() should change to take log2blksz as
its parameter and calculate blksz?
CD-ROM mode 2 has 2336 data bytes,
CD-ROM XA Mode 2, Form 2 has 2324 data bytes
according to https://en.wikipedia.org/wiki/CD-ROM.
So we cannot assume that blksz is a power of two in the general case.
LOG2() is behaving a bit erratically for numbers that are not a power of
two (in contrast to what ilog2() does):
LOG2(2048) = 11
LOG2(2049) = 11
LOG2(2324) = 15
LOG2(2336) = 15
LOG2(3072) = 11
LOG2(4096) = 12
ilog2(2048) = 11
ilog2(2049) = 11
ilog2(2324) = 11
ilog2(2336) = 11
ilog2(3072) = 11
ilog2(4096) = 12
But if we ever add an ISO 9660 file system driver, we will not be able
to use log2blksz anyway. So I think we can stick with LOG2().
Best regards
Heinrich
diff --git a/drivers/block/blk-uclass.c b/drivers/block/blk-uclass.c
index e8f58b3f5e..ca8978f0e1 100644
--- a/drivers/block/blk-uclass.c
+++ b/drivers/block/blk-uclass.c
@@ -580,6 +580,7 @@ int blk_create_device(struct udevice *parent, const char
*drv_name,
desc = dev_get_uclass_platdata(dev);
desc->if_type = if_type;
desc->blksz = blksz;
+ desc->log2blksz = LOG2(desc->blksz);
desc->lba = lba;
desc->part_type = PART_TYPE_UNKNOWN;
desc->bdev = dev;
--
2.23.0
Regards,
Simon
_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de
https://lists.denx.de/listinfo/u-boot
