HP IO Accelerator Driver and Management Software 2.2.0 Release Notes

Errata 15
This issue does not affect IO Accelerator Duo devices or HP IO Accelerators for BladeSystem c-Class.
Linux-specific issues
The following issues apply only to systems running Linux operating systems.
Linux MD RAID5 performs poorly
HP does not recommend using a RAID5 configuration. The Linux kernel RAID5 implementation performs
poorly at high data rates. This issue must be fixed in the Linux kernel and is not controlled by HP.
Alternatives include using RAID10 or possibly a third-party RAID stack. The cause of the issue is believed
to be that the RAID5 parity calculation is handled by a single kernel thread, which does not match the
performance of an array of IO Accelerators.
Drivers not installed after updating kernel
When the driver is installed for a specific kernel version and the kernel version is changed for any reason,
the IO Accelerator driver must be reinstalled to work with the new kernel version. RHEL5 has some
processes that minimize the need to reinstall the driver after a kernel upgrade.
Rare error on driver unload using kernels older than 2.6.24
A bug in Linux kernels prior to 2.6.24 can cause a general protection fault or other kernel error when the
IO Accelerator driver is unloaded. This bug also affects non-HP drivers. The bug has been resolved in
newer kernels. See the kernel 2.6.24 change log
(http://www.kernel.org/pub/linux/kernel/v2.6/ChangeLog-2.6.24) for more information. Search for
commit 5a622f2d0f86b316b07b55a4866ecb5518dd1cf7.
Because this is a bug in the Linux kernel, HP cannot resolve this issue for older kernels.
ext4 in Kernel.org 2.6.33 or earlier might silently corrupt data
when discard (trim) is enabled
NOTE: The IO Accelerator driver does not enable trim support by default at this time.
Therefore, with default options this data corruption bug cannot be triggered without explicitly
using an old kernel, enabling trim in the IO Accelerator driver and, for many instances of ext4,
explicitly enabling discard support in ext4 at mount time.
The ext4 filesystem in the Kernel.org kernel 2.6.33 and earlier contains a bug where the data in a portion
of a file might be improperly discarded (set to all 0x00) under some workloads. Use Version 2.6.34 or
newer to avoid this issue. For more information, see the patch
(http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=b90f687018e6d6 ) and
bug report (https://bugzilla.kernel.org/show_bug.cgi?id=15579).
The fix is included in RHEL6 pre-release kernel kernel-2.6.32-23.el6. The eventual release RHEL6 kernel
will not be affected by this issue.
Discard support was added to the kernel.org mainline ext4 in Version 2.6.28 and was enabled by
default. For fear of damaging some devices, discard was set to default to disabled in Version 2.6.33-rc1
and was back ported to 2.6.31.8 and 2.6.32.1. If you must use a kernel.org kernel Version 2.6.[28-30]