Fix docg3 read align bug
Robert Jarzmik
maintainerHi Artem,
I sent not long ago a patch for unaligned reads. And ... I made an error. I have a patch to patch my previous submission. I wonder if this can be squashed in, or requires an incremental commit. Tell me what you want to do, the patch is at the end of the mail.
I hope I have not overseen another corner case, I'll recheck tomorrow evening (GMT+1).
Cheers.
Signature · 3 lines
Robert ---8>---
Quoted 1 line
From
c013bc50a456c3c127eae3f1751369dec1e32221Mon Sep 17 00:00:00 2001
From: Robert Jarzmik <[email protected]>
Date: Sun, 25 Mar 2012 23:26:05 +0200
Subject: [PATCH] mtd: docg3 refix docg3 lessen read alignement burden
After the fix "mtd: docg3 less read alignement burden", a corner case where read is unaligned an length read is small was left. As a full page must be read, the remaining data bytes to read number was miscalculated, leading to unreadable LEBs by UBIFS.
--- drivers/mtd/devices/docg3.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-)
drivers/mtd/devices/docg3.c +2 -1
@@ -893,7 +893,8 @@ static int doc_read_oob(struct mtd_info *mtd, loff_t from, ret =doc_read_page_getbytes (docg3, nbdata, buf,0 );if (ret < nbdata) goto err_in_read; -doc_read_page_getbytes (docg3,DOC_LAYOUT_PAGE_SIZE - nbdata, +doc_read_page_getbytes (docg3, +DOC_LAYOUT_PAGE_SIZE - nbdata - skip,NULL ,0 ); ret =doc_read_page_getbytes (docg3, nboob, oobbuf,0 );if (ret < nboob)
Signature · 1 line
1.7.5.4
Quoted 9 lines · Robert Jarzmik
Hi Artem,
I sent not long ago a patch for unaligned reads. And ... I made an error. I have a patch to patch my previous submission. I wonder if this can be squashed in, or requires an incremental commit. Tell me what you want to do, the patch is at the end of the mail.
I hope I have not overseen another corner case, I'll recheck tomorrow evening (GMT+1).
Shoved in, thanks!
Signature · 2 lines
Best Regards, Artem Bityutskiy