On 08/19/16 08:57, Allan Jude wrote:
On 2016-08-19 10:13, Warner Losh wrote:
On Fri, Aug 19, 2016 at 12:51 AM, Dag-Erling Smørgrav wrote:
Warner Losh writes:
Allan Jude writes:
Which makes more sense:
A) If stripesize == 0, use some
On 08/18/16 22:33, Warner Losh wrote:
On Aug 18, 2016, at 11:21 PM, Nathan Whitehorn wrote:
On 08/18/16 21:15, Warner Losh wrote:
On Thu, Aug 18, 2016 at 6:56 AM, Allan Jude wrote:
On 08/18/16 05:50 AM, Dag-Erling Smørgrav wrote:
Nathan
On 2016-08-19 10:13, Warner Losh wrote:
> On Fri, Aug 19, 2016 at 12:51 AM, Dag-Erling Smørgrav wrote:
>> Warner Losh writes:
>>> Allan Jude writes:
Which makes more sense:
A) If stripesize == 0, use some sane value like 4096
On 2016-08-19 00:15, Warner Losh wrote:
>> Which makes more sense:
>>
>> A) If stripesize == 0, use some sane value like 4096
>
> I don't like this.
>
>> B) Some other combination that uses the reported stripe size, unless it
>> is 0, in which case it uses 4096 (or some other value controlled by
On Fri, Aug 19, 2016 at 12:51 AM, Dag-Erling Smørgrav wrote:
> Warner Losh writes:
>> Allan Jude writes:
>> > Which makes more sense:
>> >
>> > A) If stripesize == 0, use some sane value like 4096
>>
>> I don't like this.
>>
>> > B) Some
Warner Losh writes:
> Allan Jude writes:
> > Which makes more sense:
> >
> > A) If stripesize == 0, use some sane value like 4096
>
> I don't like this.
>
> > B) Some other combination that uses the reported stripe size, unless it
> > is 0, in which case
On Fri, Aug 19, 2016 at 08:21:49AM +0200, Dag-Erling Smørgrav wrote:
> Nathan Whitehorn writes:
> > This is ridiculous. I've asked a series of technical questions about
> > generalizing a patch you made and that I think is a good idea. In
> > response, those questions have
Nathan Whitehorn writes:
> This is ridiculous. I've asked a series of technical questions about
> generalizing a patch you made and that I think is a good idea. In
> response, those questions have been met with a non-stop torrent of
> insults and abuse instead of answers,
> On Aug 18, 2016, at 11:21 PM, Nathan Whitehorn wrote:
>
>
>
> On 08/18/16 21:15, Warner Losh wrote:
>> On Thu, Aug 18, 2016 at 6:56 AM, Allan Jude wrote:
>>> On 08/18/16 05:50 AM, Dag-Erling Smørgrav wrote:
Nathan Whitehorn
On 08/18/16 21:15, Warner Losh wrote:
On Thu, Aug 18, 2016 at 6:56 AM, Allan Jude wrote:
On 08/18/16 05:50 AM, Dag-Erling Smørgrav wrote:
Nathan Whitehorn writes:
OK. In which configurations? My Dell servers, for instance, don't do
this. How
On Thu, Aug 18, 2016 at 6:56 AM, Allan Jude wrote:
> On 08/18/16 05:50 AM, Dag-Erling Smørgrav wrote:
>> Nathan Whitehorn writes:
>>> OK. In which configurations? My Dell servers, for instance, don't do
>>> this. How are they set up? What drivers
On 08/18/16 08:12, Dag-Erling Smørgrav wrote:
Nathan Whitehorn writes:
We have a mechanism (GEOM stripe size) for drivers to supply a default
alignment to userland. If we think we can get that right, great. If we
don't think we can get it right, the default system
On 08/18/16 05:56, Allan Jude wrote:
On 08/18/16 05:50 AM, Dag-Erling Smørgrav wrote:
Nathan Whitehorn writes:
OK. In which configurations? My Dell servers, for instance, don't do
this. How are they set up? What drivers are being used? Is this
something that affects
Nathan Whitehorn writes:
> We have a mechanism (GEOM stripe size) for drivers to supply a default
> alignment to userland. If we think we can get that right, great. If we
> don't think we can get it right, the default system policy in the
> absence of real information from
On 08/18/16 02:50, Dag-Erling Smørgrav wrote:
Nathan Whitehorn writes:
OK. In which configurations? My Dell servers, for instance, don't do
this. How are they set up? What drivers are being used? Is this
something that affects passthrough disks, RAIDs, disk images?
On Thu, Aug 18, 2016 at 08:56:55AM -0400, Allan Jude wrote:
> A) If stripesize == 0, use some sane value like 4096
>
> B) Some other combination that uses the reported stripe size, unless it
> is 0, in which case it uses 4096 (or some other value controlled by a
> different new sysctl)
>
> C)
On 08/18/16 05:50 AM, Dag-Erling Smørgrav wrote:
> Nathan Whitehorn writes:
>> OK. In which configurations? My Dell servers, for instance, don't do
>> this. How are they set up? What drivers are being used? Is this
>> something that affects passthrough disks, RAIDs, disk
Nathan Whitehorn writes:
> OK. In which configurations? My Dell servers, for instance, don't do
> this. How are they set up? What drivers are being used? Is this
> something that affects passthrough disks, RAIDs, disk images?
Most LSI MegaRAID controllers don't have real
On 17/08/2016 19:36, Nathan Whitehorn wrote:
> OK, so then what is the solution here? We have a number of tools that need to
> know this information: gpart, sade, bsdinstall, zfs, graid, etc. If we want to
> have a consistent set of defaults -- for example, to use 4K across the board,
> which I
On 08/17/16 10:56, Andriy Gapon wrote:
On 17/08/2016 19:36, Nathan Whitehorn wrote:
OK, so then what is the solution here? We have a number of tools that need to
know this information: gpart, sade, bsdinstall, zfs, graid, etc. If we want to
have a consistent set of defaults -- for example, to
On 08/17/16 08:57, Warner Losh wrote:
On Wed, Aug 17, 2016 at 9:26 AM, Nathan Whitehorn
wrote:
Not true at all. All modern disks report their physical sector size, as
distinct from the logical one, in their ATA IDENTIFY data and ata_da.c uses
that.
You are correct
On 08/17/16 08:45, Dag-Erling Smørgrav wrote:
Nathan Whitehorn writes:
Dag-Erling Smørgrav writes:
I have mentioned several examples to you, and even told you how to
confirm, by inspecting the source code, that most drivers do *not*
set the stripe size.
On Wed, Aug 17, 2016 at 10:04 AM, Warner Losh wrote:
> On Wed, Aug 17, 2016 at 9:45 AM, Dag-Erling Smørgrav wrote:
>> Nathan Whitehorn writes:
>>> Dag-Erling Smørgrav writes:
>>> > I have mentioned several examples to you, and
On Wed, Aug 17, 2016 at 9:45 AM, Dag-Erling Smørgrav wrote:
> Nathan Whitehorn writes:
>> Dag-Erling Smørgrav writes:
>> > I have mentioned several examples to you, and even told you how to
>> > confirm, by inspecting the source code, that most
On Wed, Aug 17, 2016 at 9:26 AM, Nathan Whitehorn
wrote:
> Not true at all. All modern disks report their physical sector size, as
> distinct from the logical one, in their ATA IDENTIFY data and ata_da.c uses
> that.
You are correct that there are two fields in the ATA
On Wed, Aug 17, 2016 at 8:36 AM, Nathan Whitehorn
wrote:
> As for grepping, the CAM disk drivers are all in sys/cam, not sys/dev, as
> I'm sure you know, and you will find all the code that handles this there.
There's at least a dozen disk drivers that aren't CAM.
The
Nathan Whitehorn writes:
> Dag-Erling Smørgrav writes:
> > I have mentioned several examples to you, and even told you how to
> > confirm, by inspecting the source code, that most drivers do *not*
> > set the stripe size.
> Most drivers? Yes, sure. Most
Hans Petter Selasky writes:
> My intention is not to install FreeBSD on a 3K disk. My question is
> pure mathematical, if the bsdinstall will segfault, division by zero
> or anything like that, if one should try to install FreeBSD on a 3K
> disk, because you round up the size of
On 08/17/16 08:03, Dag-Erling Smørgrav wrote:
Nathan Whitehorn writes:
Dag-Erling Smørgrav writes:
[...] And you keep refusing to address the fact that most drivers
don't report a stripe size, except by repeating your claim that they
do, with no
On 08/17/16 17:05, Dag-Erling Smørgrav wrote:
Hans Petter Selasky writes:
Not sure if it is an issue, but what will happen if a magic disk has a
size less than 4K and uses a block size of 512bytes and the disk
alignment gets rounded up to 4K. Will any of logic in this patch
On Wed, Aug 17, 2016 at 05:00:16PM +0200, Hans Petter Selasky wrote:
> On 08/17/16 16:49, Slawa Olhovchenkov wrote:
> > On Wed, Aug 17, 2016 at 07:36:00AM -0700, Nathan Whitehorn wrote:
> >
>
> > In long term, prefered aligment is forsing 4k (or may be more):
> > install system on 512b [mirror]
Hans Petter Selasky writes:
> Not sure if it is an issue, but what will happen if a magic disk has a
> size less than 4K and uses a block size of 512bytes and the disk
> alignment gets rounded up to 4K. Will any of logic in this patch fail
> or hang?
What is a magic disk, and
On 08/17/16 07:49, Slawa Olhovchenkov wrote:
On Wed, Aug 17, 2016 at 07:36:00AM -0700, Nathan Whitehorn wrote:
Your contention that the installer does not make policy decisions is
equally spurious. The installer makes many policy decisions, including
the disk layout, the size of the swap
Nathan Whitehorn writes:
> Dag-Erling Smørgrav writes:
> > [...] And you keep refusing to address the fact that most drivers
> > don't report a stripe size, except by repeating your claim that they
> > do, with no evidence to back it up. Feel free to 'grep
On 08/17/16 16:49, Slawa Olhovchenkov wrote:
On Wed, Aug 17, 2016 at 07:36:00AM -0700, Nathan Whitehorn wrote:
In long term, prefered aligment is forsing 4k (or may be more):
install system on 512b [mirror] disk aligment now may be need required replace
disk to 4k aligment. For more
On Wed, Aug 17, 2016 at 07:36:00AM -0700, Nathan Whitehorn wrote:
> > Your contention that the installer does not make policy decisions is
> > equally spurious. The installer makes many policy decisions, including
> > the disk layout, the size of the swap partition, the name of the pool,
> > the
On 08/17/16 03:07, Dag-Erling Smørgrav wrote:
Nathan Whitehorn writes:
As a note for people who weren't paying attention to the bug, we need
to fix this in a better way outside of the constraints of getting 11.0
out the door. The system (gpart, the installer, ZFS,
Nathan Whitehorn writes:
> As a note for people who weren't paying attention to the bug, we need
> to fix this in a better way outside of the constraints of getting 11.0
> out the door. The system (gpart, the installer, ZFS, etc.) uses the
> reported GEOM stripesize for
As a note for people who weren't paying attention to the bug, we need to
fix this in a better way outside of the constraints of getting 11.0 out
the door. The system (gpart, the installer, ZFS, etc.) uses the reported
GEOM stripesize for partition alignment and IO block size selection. If
that
39 matches
Mail list logo