linux-kernel - Block request processing for MMC/SD over SPI bus

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <200608011037.27721.hans@rubico.se>
Date:	Tue, 1 Aug 2006 10:37:27 +0200
From:	Hans Eklund <hans@...ico.se>
To:	linux-kernel@...r.kernel.org
Subject: Block request processing for MMC/SD over SPI bus

(CC to hans@...ico.se)

After trying other forums and irc for help on this matter i hope someone on 
the LKML can help.

-Background(skip if short of time).
First linux driver for me and first post on LKML, but in short, i have been 
developing a MMC/SD card driver in SPI mode for uClinux on the Analog Devices 
Blackfin based platform. However, since the 2.6.15 version im moving over to 
the common SPI framework and all code will probably be platform independent 
soon. The driver is working quite well but need some tuning in the request 
processing.

The driver is also independent of any previous MMC related work at this point 
since MMC over SPI mode differs somewhat from the MMC mode.

I have implemeted the request processing from scratch. I have followed the 
Linux Device Drivers and the Understanding Linux Kernel as far as possible, 
but they are not very extensive on error handling. I have also had a look at 
the ATA and SCSI drivers.

-Questions
A snip of my current request processing can be found here: with comments and 
some questions. 

http://hasse.yohoo.nu/strat.txt

The big issue is error handling and how i report back to the user space 
processes that the device is dead(if hardware fails for example), basically 
how to gracefully handle critical issues. At this point, if retries on broken 
sectors fails im not sure how to report back to the I/O scheduler and block 
layer so the device can be halted. Now, the kernel actually hangs(or retries 
same sector forever). Im really at a dead end here. Could really use some 
help at this point.

Also, Is there any work done similar to mine?
Would there be interest in incorporating the driver in future kernels?

Answers/CC to hans@...ico.se since i did not dare to plunge into the lkml 
traffic just yet.

Best regards
Hans Eklund
Rubico AB
Sweden.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/