Bug 173115 - [zfs] [patch] Teach ZFS about geom stripe size so zpools are created with optimum ashift
Summary: [zfs] [patch] Teach ZFS about geom stripe size so zpools are created with opt...
Status: Closed FIXED
Alias: None
Product: Base System
Classification: Unclassified
Component: kern (show other bugs)
Version: 8.3-RELEASE
Hardware: Any Any
: Normal Affects Only Me
Assignee: Steven Hartland
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2012-10-26 09:50 UTC by Steven Hartland
Modified: 2016-01-31 15:02 UTC (History)
1 user (show)

See Also:

Attachments
file.txt (8.25 KB, text/plain)
2012-10-26 09:50 UTC, Steven Hartland 		no flags Details
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Steven Hartland 2012-10-26 09:50:01 UTC 
Changes zfs zpool initial / desired ashift to be based off stripesize instead of sectorsize making it compatible with drives marked with the 4k sector size quirk.

Without the correct min block size BIO_DELETE requests passed to a large number of current SSD's via TRIM don't actually perform any LBA TRIM so its vital for the correct operation of TRIM to get the correct min block size.

To do this we added the additional dashift (desired ashift) to vdev_open_func_t calls. This was needed as just updating ashift to be based off stripesize would mean that a devices reported minimum transfer size (ashift) could increase and that in turn would cause member devices to be unusable and hence break pools with error ZFS-8000-5E.

Fix: Apply the attached patch

Patch attached with submission follows:
How-To-Repeat: N/A
Comment 1 Steven Hartland freebsd_committer freebsd_triage 2012-12-11 14:10:03 UTC 
Responsible Changed
From-To: freebsd-bugs->smh

I'll take it.
Comment 2 Allan Jude freebsd_committer freebsd_triage 2016-01-31 14:41:36 UTC 
is this done?
Comment 3 Steven Hartland freebsd_committer freebsd_triage 2016-01-31 15:02:48 UTC 
Yes I did this ages ago, closing.