Re: New ac patch???

Re: New ac patch???

[Petr Vandrovec]

On 21 Nov 01 at 11:12, Alan Cox wrote:
> > > Not exaclty. It is a 48Gig drive in a dell inspiron 8000. I think it is
> > > IBM but the logs do not show a brandname. I can try open up the case tonight
> > > if you want to know for sure?
> > 
> > It's an IBM IC25T048ATDA05-0 to be precise.
> 
> Thanks. It seems IBM laptop drives are the ones that specifically need this
> fix. That ties in with the windows 98 reports/microsoft fixes.

If it will go into mainstream, please change it to config option or
something like that. I'm doing 

for a in /dev/hd[a-z]; do hdparm -Y $a; done

in my poweroff script. With -ac /sbin/halt then stops on each disk 
for couple of seconds, because of drive in sleep mode does not respond,
and so IDE driver has to timeout, reinit interface and flush cache
(fortunately drive does not spin up again).
                                        Thanks,
                                                Petr Vandrovec
                                                vandrove@vc.cvut.cz
                                                

Re: New ac patch???

[Alan Cox]

> > Thanks. It seems IBM laptop drives are the ones that specifically need this
> > fix. That ties in with the windows 98 reports/microsoft fixes.
> 
> If it will go into mainstream, please change it to config option or
> something like that. I'm doing 
> 
> for a in /dev/hd[a-z]; do hdparm -Y $a; done

Thats somewhat undefined in its behaviour. You might also page after the
hdparm -Y

Re: New ac patch???

[Petr Vandrovec]

On 26 Nov 01 at 20:39, Alan Cox wrote:
> > If it will go into mainstream, please change it to config option or
> > something like that. I'm doing 
> > 
> > for a in /dev/hd[a-z]; do hdparm -Y $a; done
> 
> Thats somewhat undefined in its behaviour. You might also page after the
> hdparm -Y

cat /sbin/halt > /dev/null
cat /bin/sleep > /dev/null
for a in /dev/hd[a-z]; do /sbin/hdparm -Y $a; done
/bin/sleep 1
/sbin/halt -d -f -i -p -h

works without a glitch - libc & co. is already in memory due to running
other code, hdparm is read during first pass through the loop, and cat 
brings /sbin/halt into the memory. And as root filesystem is mounted 
read-only at that time, there are of course no atime updates pending from 
these accesses.
                                        Best regards,
                                            Petr Vandrovec
                                            vandrove@vc.cvut.cz
                                            

New ac patch???

[Roy Sigurd Karlsbakk]

Hi all

When will the AC patches follow the linus tree?

thanx

--
Roy Sigurd Karlsbakk, MCSE, MCNE, CLS, LCA

Computers are like air conditioners.
They stop working when you open Windows.

Re: New ac patch???

[J Sloan]

Roy Sigurd Karlsbakk wrote:

> Hi all
>
> When will the AC patches follow the linus tree?

Good question -

Perhaps -mt patches will appear instead?

cu

jjs

Re: New ac patch???

[Alan Cox]

> When will the AC patches follow the linus tree?

Right now I've fed all the stuff I feel makes sense to Linus for 2.4.15.
Once 2.4.15 is out I'll send some more bits to Marcelo, and also some bits
to Linus that are 2.5 material (eg PnPBIOS). The only "-ac" patch as such
would be for 32bit quota and other oddments so I don't think its worth the
effort.

Alan

Re: New ac patch???

[Roy Sigurd Karlsbakk]

> Right now I've fed all the stuff I feel makes sense to Linus for 2.4.15.
> Once 2.4.15 is out I'll send some more bits to Marcelo, and also some bits
> to Linus that are 2.5 material (eg PnPBIOS). The only "-ac" patch as such
> would be for 32bit quota and other oddments so I don't think its worth the
> effort.

Will this include the patches to allow for /proc/sys/vm/(min|max)-readahead
soon?

thanks

roy
--
Roy Sigurd Karlsbakk, MCSE, MCNE, CLS, LCA

Computers are like air conditioners.
They stop working when you open Windows.

Re: New ac patch???

[Alan Cox]

> > to Linus that are 2.5 material (eg PnPBIOS). The only "-ac" patch as such
> > would be for 32bit quota and other oddments so I don't think its worth the
> > effort.
> 
> Will this include the patches to allow for /proc/sys/vm/(min|max)-readahead
> soon?

Thats pretty low on my priority list. Its actually not a hard patch to
extract although I'd prefer someone like Andrea who knows the new rather
undocumented VM did the merge

Re: New ac patch???

[Marcel J.E. Mol]

Hi,


2.4.13-ac will "flushing ide drives" on shutdown. This helped my laptop
from not '/dev/hdax no cleanly unmounted, checking' on startup. I'm sure
the system did not crash before that.

2.4.15-pre6 does not have this code and now sometimes some filesystems
seem not to be clean anymore on startup...

Will the ide_notify_reboot be included in 2.4.15 final?

-Marcel

On Tue, Nov 20, 2001 at 11:38:50PM +0000, Alan Cox wrote:
> > > to Linus that are 2.5 material (eg PnPBIOS). The only "-ac" patch as such
> > > would be for 32bit quota and other oddments so I don't think its worth the
> > > effort.
> > 
> > Will this include the patches to allow for /proc/sys/vm/(min|max)-readahead
> > soon?
> 
> Thats pretty low on my priority list. Its actually not a hard patch to
> extract although I'd prefer someone like Andrea who knows the new rather
> undocumented VM did the merge
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

-- 
     ======--------         Marcel J.E. Mol                MESA Consulting B.V.
    =======---------        ph. +31-(0)6-54724868          P.O. Box 112
    =======---------        marcel@mesa.nl                 2630 AC  Nootdorp
__==== www.mesa.nl ---____U_n_i_x______I_n_t_e_r_n_e_t____ The Netherlands ____
 They couldn't think of a number,           Linux user 1148  --  counter.li.org
    so they gave me a name!  -- Rupert Hine  --  www.ruperthine.com

Re: New ac patch???

[Alan Cox]

> 2.4.13-ac will "flushing ide drives" on shutdown. This helped my laptop
> from not '/dev/hdax no cleanly unmounted, checking' on startup. I'm sure
> the system did not crash before that.

You have a box with an IBM 20Gig 2.5" drive (just out of interest)

> 2.4.15-pre6 does not have this code and now sometimes some filesystems
> seem not to be clean anymore on startup...
> Will the ide_notify_reboot be included in 2.4.15 final?

Probably not - the taskfile/LBA48 code wants more testing first. I believe
you can pick up the relevant patch from www.linux-ide.org however, thanks
to Andre

Re: New ac patch???

[Marcel J.E. Mol]

On Wed, Nov 21, 2001 at 09:21:45AM +0000, Alan Cox wrote:
> > 2.4.13-ac will "flushing ide drives" on shutdown. This helped my laptop
> > from not '/dev/hdax no cleanly unmounted, checking' on startup. I'm sure
> > the system did not crash before that.
> 
> You have a box with an IBM 20Gig 2.5" drive (just out of interest)

Not exaclty. It is a 48Gig drive in a dell inspiron 8000. I think it is
IBM but the logs do not show a brandname. I can try open up the case tonight
if you want to know for sure?

-Marcel
-- 
     ======--------         Marcel J.E. Mol                MESA Consulting B.V.
    =======---------        ph. +31-(0)6-54724868          P.O. Box 112
    =======---------        marcel@mesa.nl                 2630 AC  Nootdorp
__==== www.mesa.nl ---____U_n_i_x______I_n_t_e_r_n_e_t____ The Netherlands ____
 They couldn't think of a number,           Linux user 1148  --  counter.li.org
    so they gave me a name!  -- Rupert Hine  --  www.ruperthine.com

Re: New ac patch???

[Rasmus Bøg Hansen]

On Wed, 21 Nov 2001, Marcel J.E. Mol wrote:

> > You have a box with an IBM 20Gig 2.5" drive (just out of interest)
> 
> Not exaclty. It is a 48Gig drive in a dell inspiron 8000. I think it is
> IBM but the logs do not show a brandname. I can try open up the case tonight
> if you want to know for sure?

'hdparm -i /dev/hdX' should tell you...

Rasmus

-- 
-- [ Rasmus 'Møffe' Bøg Hansen ] ---------------------------------------
God, root, what is difference?
God is more forgiving.
--------------------------------- [ moffe at amagerkollegiet dot dk ] --

Re: New ac patch???

[Dominik Kubla]

On Wed, Nov 21, 2001 at 10:33:54AM +0100, Marcel J.E. Mol wrote:
> Not exaclty. It is a 48Gig drive in a dell inspiron 8000. I think it is
> IBM but the logs do not show a brandname. I can try open up the case tonight
> if you want to know for sure?

It's an IBM IC25T048ATDA05-0 to be precise.

Dominik Kubla
-- 
ScioByte GmbH    Zum Schiersteiner Grund 2     55127 Mainz (Germany)
Phone: +49 700 724 629 83                    Fax: +49 700 724 629 84
1024D/717F16BB    A384 F5F1 F566 5716 5485  27EF 3B00 C007 717F 16BB

Re: New ac patch???

[Alan Cox]

> > Not exaclty. It is a 48Gig drive in a dell inspiron 8000. I think it is
> > IBM but the logs do not show a brandname. I can try open up the case tonight
> > if you want to know for sure?
> 
> It's an IBM IC25T048ATDA05-0 to be precise.

Thanks. It seems IBM laptop drives are the ones that specifically need this
fix. That ties in with the windows 98 reports/microsoft fixes.

Re: New ac patch???

[Miquel van Smoorenburg]

In article <E166VIr-0004ik-00@the-village.bc.nu>,
Alan Cox  <alan@lxorguk.ukuu.org.uk> wrote:
>> > Not exaclty. It is a 48Gig drive in a dell inspiron 8000. I think it is
>> > IBM but the logs do not show a brandname. I can try open up the case tonight
>> > if you want to know for sure?
>> 
>> It's an IBM IC25T048ATDA05-0 to be precise.
>
>Thanks. It seems IBM laptop drives are the ones that specifically need this
>fix. That ties in with the windows 98 reports/microsoft fixes.

I have Debian bugreports saying that it's not only IBM laptop drives.
It happens on Seagate, Samsung, Maxtor as well.
See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=110804&repeatmerged=yes

Note that the final fix (putting the drives in standby mode) is not
discussed in this bug report - I experimented first with turning off
the write cache in the shutdown scripts which kind of worked but
feels not quite right - you might need just enough write activity
after that to completely flush the on-disk cache.

Mike.
-- 
"Only two things are infinite, the universe and human stupidity,
 and I'm not sure about the former" -- Albert Einstein.

Re: New ac patch???

[Marcel J.E. Mol]

On Wed, Nov 21, 2001 at 11:12:28AM +0000, Alan Cox wrote:
> > > Not exaclty. It is a 48Gig drive in a dell inspiron 8000. I think it is
> > > IBM but the logs do not show a brandname. I can try open up the case tonight
> > > if you want to know for sure?
> > 
> > It's an IBM IC25T048ATDA05-0 to be precise.
> 
> Thanks. It seems IBM laptop drives are the ones that specifically need this
> fix. That ties in with the windows 98 reports/microsoft fixes.

Would that be enough reason to add only the specific flushing code of
the taskfile patch (if at all possible) to the kernel? Maybe Andre is
willing to extract the relevant code in a seperate patch...

-Marcel
-- 
     ======--------         Marcel J.E. Mol                MESA Consulting B.V.
    =======---------        ph. +31-(0)6-54724868          P.O. Box 112
    =======---------        marcel@mesa.nl                 2630 AC  Nootdorp
__==== www.mesa.nl ---____U_n_i_x______I_n_t_e_r_n_e_t____ The Netherlands ____
 They couldn't think of a number,           Linux user 1148  --  counter.li.org
    so they gave me a name!  -- Rupert Hine  --  www.ruperthine.com

Re: New ac patch???

[Jakob Kemi]

On Wednesdayen den 21 November 2001 13.23, Marcel J.E. Mol wrote:
> On Wed, Nov 21, 2001 at 11:12:28AM +0000, Alan Cox wrote:
> > > > Not exaclty. It is a 48Gig drive in a dell inspiron 8000. I think it
> > > > is IBM but the logs do not show a brandname. I can try open up the
> > > > case tonight if you want to know for sure?
> > >
> > > It's an IBM IC25T048ATDA05-0 to be precise.
> >
> > Thanks. It seems IBM laptop drives are the ones that specifically need
> > this fix. That ties in with the windows 98 reports/microsoft fixes.
>
> Would that be enough reason to add only the specific flushing code of
> the taskfile patch (if at all possible) to the kernel? Maybe Andre is
> willing to extract the relevant code in a seperate patch...
>
> -Marcel

This also affects my Desktop PC with two IBM Deskstar 60GXP 40GB drives. I'd 
like to see it in 2.4.15.

/Jakob

Re: New ac patch???

[Jeff Chua]

On Wed, 21 Nov 2001, Alan Cox wrote:

> > > Not exaclty. It is a 48Gig drive in a dell inspiron 8000. I think it is
> > > IBM but the logs do not show a brandname. I can try open up the case tonight
> > > if you want to know for sure?
> >
> > It's an IBM IC25T048ATDA05-0 to be precise.
>
> Thanks. It seems IBM laptop drives are the ones that specifically need this
> fix. That ties in with the windows 98 reports/microsoft fixes.

I'm using IBM 240Z with IBM-DJSA-220 (20GB, 9.5mm thickness) with
2.4.15-pre7 and it does not need any fixes. Disk clean after shutdown.

Jeff.

Re: New ac patch???

[Marcel J.E. Mol]

On Thu, Nov 22, 2001 at 11:03:27AM +0800, Jeff Chua wrote:
> On Wed, 21 Nov 2001, Alan Cox wrote:
> 
> > > > Not exaclty. It is a 48Gig drive in a dell inspiron 8000. I think it is
> > > > IBM but the logs do not show a brandname. I can try open up the case tonight
> > > > if you want to know for sure?
> > >
> > > It's an IBM IC25T048ATDA05-0 to be precise.
> >
> > Thanks. It seems IBM laptop drives are the ones that specifically need this
> > fix. That ties in with the windows 98 reports/microsoft fixes.
> 
> I'm using IBM 240Z with IBM-DJSA-220 (20GB, 9.5mm thickness) with
> 2.4.15-pre7 and it does not need any fixes. Disk clean after shutdown.

Does it have write caching enabled? check with hdparm ( I think you need one of
the latest hdparm versions...)

-Marcel
-- 
     ======--------         Marcel J.E. Mol                MESA Consulting B.V.
    =======---------        ph. +31-(0)6-54724868          P.O. Box 112
    =======---------        marcel@mesa.nl                 2630 AC  Nootdorp
__==== www.mesa.nl ---____U_n_i_x______I_n_t_e_r_n_e_t____ The Netherlands ____
 They couldn't think of a number,           Linux user 1148  --  counter.li.org
    so they gave me a name!  -- Rupert Hine  --  www.ruperthine.com

Re: New ac patch???

[Jeff Chua]

On Thu, 22 Nov 2001, Marcel J.E. Mol wrote:

> Does it have write caching enabled? check with hdparm ( I think you need one of
> the latest hdparm versions...)

# hdparm -i /dev/hda

/dev/hda:

 Model=IBM-DJSA-220, FwRev=JS4OAC2A, SerialNo=44T44P30559
 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
 RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
 BuffType=DualPortCache, BuffSize=1874kB, MaxMultSect=16, MultSect=16
 CurCHS=17475/15/63, CurSects=4216520955, LBA=yes, LBAsects=39070080
 IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
 PIO modes: pio0 pio1 pio2 pio3 pio4
 DMA modes: mdma0 mdma1 mdma2 udma0 udma1 *udma2 udma3 udma4
 AdvancedPM=yes: mode=0x80 (128) WriteCache=enabled
 Drive Supports : ATA/ATAPI-5 T13 1321D revision 1 : ATA-2 ATA-3 ATA-4
ATA-5

Jeff

Re: New ac patch???

[Marcel J.E. Mol]

On Wed, Nov 21, 2001 at 12:00:33PM +0100, Dominik Kubla wrote:
> On Wed, Nov 21, 2001 at 10:33:54AM +0100, Marcel J.E. Mol wrote:
> > Not exaclty. It is a 48Gig drive in a dell inspiron 8000. I think it is
> > IBM but the logs do not show a brandname. I can try open up the case tonight
> > if you want to know for sure?
> 
> It's an IBM IC25T048ATDA05-0 to be precise.

Exact. But noone reports the 'IBM' part, not even hdparm -i...
But hdparm does shows that writecaching is enabled.

-Marcel
-- 
     ======--------         Marcel J.E. Mol                MESA Consulting B.V.
    =======---------        ph. +31-(0)6-54724868          P.O. Box 112
    =======---------        marcel@mesa.nl                 2630 AC  Nootdorp
__==== www.mesa.nl ---____U_n_i_x______I_n_t_e_r_n_e_t____ The Netherlands ____
 They couldn't think of a number,           Linux user 1148  --  counter.li.org
    so they gave me a name!  -- Rupert Hine  --  www.ruperthine.com

Re: New ac patch???

[Miquel van Smoorenburg]

In article <E166TZh-0004T8-00@the-village.bc.nu>,
Alan Cox  <alan@lxorguk.ukuu.org.uk> wrote:
>> 2.4.13-ac will "flushing ide drives" on shutdown. This helped my laptop
>> from not '/dev/hdax no cleanly unmounted, checking' on startup. I'm sure
>> the system did not crash before that.
>
>You have a box with an IBM 20Gig 2.5" drive (just out of interest)
>
>> 2.4.15-pre6 does not have this code and now sometimes some filesystems
>> seem not to be clean anymore on startup...
>> Will the ide_notify_reboot be included in 2.4.15 final?
>
>Probably not - the taskfile/LBA48 code wants more testing first. I believe
>you can pick up the relevant patch from www.linux-ide.org however, thanks
>to Andre

As I posted here last week, it's trivial to put the IDE drives into
standby mode just before powerdown instead. It has the same effect
and it would take just a few lines of code.

The Debian /sbin/halt does this now just before it asks the kernel
to poweroff, and it fixes this problem just fine. It would be better
if the kernel did it though.

Mike.
-- 
"Only two things are infinite, the universe and human stupidity,
 and I'm not sure about the former" -- Albert Einstein.

Re: New ac patch???

[Arnvid Karstad]

On Wed, 21 Nov 2001 09:21:45 +0000 (GMT) Alan Cox <alan@lxorguk.ukuu.org.uk> wrote:

> > 2.4.13-ac will "flushing ide drives" on shutdown. This helped my laptop
> > from not '/dev/hdax no cleanly unmounted, checking' on startup. I'm sure
> > the system did not crash before that.
> 
> You have a box with an IBM 20Gig 2.5" drive (just out of interest)

I've got one of those..  IBM Thinkpad T22 running on Linux Redhat 7.2
and Kernel 2.4.9-13. And I haven't really been stumbeling into any such
problems lately.. could try to upgrade to 2.4.14 and see if it's still
stable ;)

Arnvid