[Bug 63981] Bad: Buffer I/O errors make disk unusable

[bugzilla-daemon@bugzilla.kernel.org]

https://bugzilla.kernel.org/show_bug.cgi?id=63981

--- Comment #5 from Giuseppe Scalzi <scalg1@bfh.ch> ---
(In reply to Theodore Tso from comment #4)
> BTW, I'm using a 512GB Samsung 840 PRO (2.5" SATA SSD) and an 240GB Intel
> 525 SSD (mSata) on my Lenovo T430s, and they both work like a charm.
> 
> Hmm... I wasn't able to get detailed specs on your SAMSUNG
> MZNTD256HAGL-00000, but upon doing some further research, it appears to be a
> new-fangled M.2 PCIe interface.  So it's not a mSATA nor a 2.5" SATA
> interface, but Something New.
> 
> So whether or not this is a Linux bug, or an implementation bug in this new
> Samsung part (or a failure in the standardization of this new M.2 PCIe
> interface), I can't say, but this looks like the most likely cause is a
> problem with this new SSD or its new M.2 interface[1].
> 
> [1] http://en.wikipedia.org/wiki/Next_Generation_Form_Factor

Ok, thank you for you reply, I understand that isn't a problem related to EXT4.

I noticed from the archlinux wiki of my laptop model
(https://wiki.archlinux.org/index.php/Sony_Vaio_Pro_SVP-1x21) that they suggest
to use this option:

- When booting from USB, append libata.force=noncq to the kernel parameters to
avoid problems with the SSD.

Well they say "when booting from USB" but I'll try "libata.force=noncq" anyway.

We will see what happens.

-- 
You are receiving this mail because:
You are watching the assignee of the bug.