Hello Rob, Sorry for the late reply ... Rob Herring wrote: > On 12/06/2011 04:06 PM, Wolfram Sang wrote: >>>> bootcount itself is not a device. It is a feature of certain devices. And that >>>> needs to be implemented; possibly generic enough that it can work for register >>>> based, i2c based, and so forth, accesses. >>> If "boot counter" is not a good name for such a device, then what name >>> would you suggest? >> None. >> >>> Or do you think a counter (which can be implemented in a number of >>> different ways, depending on hardware specifics) is not a device? >> Yes. >> >>> What would be such a device, then? >> "maxim,ds1338" >> >> Please have a look at the devicetree.org-wiki-page I just mentioned: >> >> http://devicetree.org/Device_Tree_Usage >> > > Perhaps fs/pstore would be a good choice for the user space interface > (defining a new file bootcount). This can support any arbitrary backing > device although pretty much only ACPI is implemented. I tried to use fs/pstore for the bootcount feature, and I can mount and read the bootcount value from the file, I created. But I could not write to that file ... I only see the callback from include/linux/pstore.h struct pstore_info { [...] int (*write)(enum pstore_type_id type, enum kmsg_dump_reason reason, u64 *id, unsigned int part, size_t size, struct pstore_info *psi); called, if I reboot ... do I miss something? Or is it not possible to write to the files created through fs/pstore? bye, Heiko -- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany -- To unsubscribe from this list: send the line "unsubscribe linux-watchdog" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
