903 microSD write fails on PocketBook 903

[LoneTech]

I absolutely agree. It's in the SD driver, as the usb gadget storage interface works fine. I'll check the dmesg after I do more testing, just haven't gotten to it yet. I might even be able to fix the problem, given root on the device, but the one thing that would really help is the Samsung technical documentation. Right now my hands are a bit tied by lack of time, PB's withholding root, and Samsung's withholding documentation.

[LoneTech]

For what it's worth, here is the dmesg output from trying to make two trivial modifications in poterm (touch foo, mv ebooks books):

<3>mmcblk1: error -84 transferring data, sector 7664, nr 1, card status 0x900
<3>end_request: I/O error, dev mmcblk1, sector 7664
<3>Buffer I/O error on device mmcblk1, logical block 7664
<4>lost page write due to I/O error on mmcblk1
<4>------------[ cut here ]------------
<4>WARNING: at fs/buffer.c:1247 mark_buffer_dirty+0x38/0xd4()
<1>Modules linked in:
<1>[<c0303260>] (dump_stack+0x0/0x14) from [<c004e00c>] (warn_slowpath+0x68/0x9c)
<1>[<c004dfa4>] (warn_slowpath+0x0/0x9c) from [<c00d8644>] (mark_buffer_dirty+0x38/0xd4)
<1> r3:00000828 r2:00000000
<1> r7:00000000 r6:cf223e00 r5:c04552cc r4:cf497c48
<1>[<c00d860c>] (mark_buffer_dirty+0x0/0xd4) from [<c0125bec>] (fat_remove_entries+0x68/0x14c)
<1> r5:cf497c48 r4:cf495c80
<1>[<c0125b84>] (fat_remove_entries+0x0/0x14c) from [<c012c1f0>] (vfat_rename+0x38c/0x3c8)
<1> r8:00000000 r7:cf495c80 r6:cf495c80 r5:00000000 r4:00000001
<1>[<c012be64>] (vfat_rename+0x0/0x3c8) from [<c00be898>] (vfs_rename+0x138/0x374)
<1>[<c00be760>] (vfs_rename+0x0/0x374) from [<c00c0940>] (sys_renameat+0x160/0x1d8)
<1>[<c00c07e0>] (sys_renameat+0x0/0x1d8) from [<c00c09dc>] (sys_rename+0x24/0x28)
<1>[<c00c09b8>] (sys_rename+0x0/0x28) from [<c002c9c0>] (ret_fast_syscall+0x0/0x2c)
<4>---[ end trace 852b6eab636ec444 ]---
<3>mmcblk1: error -84 transferring data, sector 7664, nr 1, card status 0x900
<3>end_request: I/O error, dev mmcblk1, sector 7664
<3>Buffer I/O error on device mmcblk1, logical block 7664
<4>lost page write due to I/O error on mmcblk1
<3>FAT: Filesystem error (dev mmcblk1)
<3>    vfat_rename: Filesystem corrupted (i_pos 122629)
<3>    File system has been set read-only

The card was left with a corrupt filesystem where both book and ebooks pointed to the same directory. That it ever wrote such a state indicates an oddity in the vfat filesystem driver, but the sd/mmc driver is what failed.

Notably, these writes should both be only a single sector (updating entries in the root directory). It should be entirely unaffected by card class. The reason it triggers the fault is probably rapid rewrites done by the vfat driver.

[bogdaniosif]

Strange for you to be having these problems LoneTech. As you can see in the link below, Mr Maksym Prudeus, who is no less than director of international sales at Pocketbook and thus not a sad joke of an emo with an '80s hairstyle, announced that they have 60 geniuses working on the firmware and their product QA is top notch. So, again, strange for you to be having these problems. Are you sure you read the manual?

http://www.youtube.com/watch?v=fYvZgK0QDSA

[LoneTech]

Of course I have read the manual. I mentioned it in the review, including little niggles like how it shows the wrong model, wrong keyboard, menus that do not exist etc. It's also poorly translated. Thanks for suggesting it, though, as it allowed me an opportunity to find that PDF text search in pdfviewer is useless. In AdobeViewer, it merely hangs when I ask it to continue searching, but did find words. I really did not expect that.

The manual mentions the slot exists, that it takes MicroSD up to 32Gb (some of us would spell that GiB to avoid confusion, as b is frequently used for bit and G is 10^9).. which of course means microSDHC, a different protocol. SD only goes up to 2GB, formally. My card happens to be SDHC although it is only 2GB, but supports both protocols. There's no mention of the speed class there, nor on the box - only in this forum.

Another document I have been reading is what little of the SD specification that is public. Conveniently, this includes no way of confirming what the Class ratings even mean. The common hypothesis is that it means linear recording speed in MB/s, which in the specification is carefully outlined; that means writing entire eraseblocks, a behaviour a video camera can easily apply. A computer, not so much.. unless you pass low level knowledge of the card to it and accept abysmal performance (none of the major OSes do this). The only other difference with higher class cards appears to be an optional write rate informational command, that could be used to detect slow cards. If you use that, naturally you'd assume a slow card if the command fails, so if they do, and require faster cards (which they should not), there's no excuse to not show a warning.

The fact of it is, the SD slot isn't working as advertised, and instead of escalating this issue to the maker of that part, PocketBook is filing it as "probably doesn't happen with another card". That the failure mode is unacceptable as well is also glossed over.

I'm sorry if I misunderstood your post. I still can't quite figure out what its purpose was.

[rkomar]

Originally Posted by LoneTech

I'm sorry if I misunderstood your post. I still can't quite figure out what its purpose was.

He was making a dig at PocketBook. I don't think he was expecting any kind of reply.

I've been googling to see if similar problems to this have turned up elsewhere (as I'm sure you have), and it looks like they have. I haven't seen anything pointing to a solution yet (certainly not anything like "you only have to mount the card using the --avoid-sdhd-horrors switch"). The kernel used on the 902 is rather old (18 months), so there's a faint hope that this problem has been fixed in a more recent kernel and PocketBook can "just" fix it by upgrading.

[bogdaniosif]

Originally Posted by rkomar

He was making a dig at PocketBook. I don't think he was expecting any kind of reply.

Right you are rkomar.

I was just watching a CES press release from Pocketbook and when I heard that joke of a director bragging about how they can ask for more money than Sony because they have experts writing their own firmware and they have amazing QA, I just snapped.

Sorry to have steered this thread a little off topic LoneTech. I watched, in disappointment, other people having serious problems with their new pocketbooks but I just found your problem to be the most outrageous.

[Earl Grey]

Originally Posted by LoneTech

There's no mention of the speed class there, nor on the box - only in this forum.

The fact of it is, the SD slot isn't working as advertised, and instead of escalating this issue to the maker of that part, PocketBook is filing it as "probably doesn't happen with another card". That the failure mode is unacceptable as well is also glossed over.

I have to agree, there is no mention anywhere in the manual concerning speed class.

I also have found that problems happened with a 2Gb Dane_Elec card, presumably class 2, but also a 8Gb Sony class 4 card.