linuxcompressed-devel

[lc-devel] UML swap bug fixed

[Rodrigo S. de C. <rc...@im...>]

Hi there,

Finally Jeff seems to have fixed that bug which screwed up our vswap
addresses. I think I can continue developing under UML :-)

----- Forwarded message from Jeff Dike <jd...@ka...> -----

Date: Thu, 10 Jan 2002 22:16:18 -0500
From: Jeff Dike <jd...@ka...>
Subject: Re: [uml-devel] swap entry and newpage bit 
To: Rodrigo Souza de Castro <rc...@im...>
Cc: use...@li...

rc...@im... said:
> The problem I noticed is that, when setting a pte to a swap entry, the
> NEWPAGE bit is turned on to tell fix_range() that this address should
> be unmapped. It turns out that, for swap adresses that are shared by
> many ptes, the address is not unmapped and the NEWPAGE bit is not
> turned off soon enough before any pte faults. Then, in fault handlers
> (handle_pte_fault() and do_swap_page()), the pte has a wrong value, so
> I can't find the page in swap cache. For example, if the swap cache
> page is set to 0x3307c and the wrong pte is set to 0x3307e (due to
> newpage bit), I won't be able to find it.

This is an excellent description of the bug that I just spend 5 days chasing.

Unfortunately, I didn't until recently (yesterday) understand enough of the
swap code to tell whether you were describing something real or whether
you were on crack.

It is now fixed.

pte_to_swp_entry now masks off the arch bits.

				Jeff

----- End forwarded message -----

-- 
Rodrigo S. de Castro <rc...@im...>

Re: [lc-devel] UML swap bug fixed

[David C. <dav...@rc...>]

Is this the bug that I point out in my previous mail to the list? Or it 
is caused by a different issue? Please clarify, since I don't have 
enough time to understand all the code at the moment. Thanks.

David

Rodrigo Souza de Castro wrote:

>Hi there,
>
>Finally Jeff seems to have fixed that bug which screwed up our vswap
>addresses. I think I can continue developing under UML :-)
>
>----- Forwarded message from Jeff Dike <jd...@ka...> -----
>
>Date: Thu, 10 Jan 2002 22:16:18 -0500
>From: Jeff Dike <jd...@ka...>
>Subject: Re: [uml-devel] swap entry and newpage bit 
>To: Rodrigo Souza de Castro <rc...@im...>
>Cc: use...@li...
>
>rc...@im... said:
>
>>The problem I noticed is that, when setting a pte to a swap entry, the
>>NEWPAGE bit is turned on to tell fix_range() that this address should
>>be unmapped. It turns out that, for swap adresses that are shared by
>>many ptes, the address is not unmapped and the NEWPAGE bit is not
>>turned off soon enough before any pte faults. Then, in fault handlers
>>(handle_pte_fault() and do_swap_page()), the pte has a wrong value, so
>>I can't find the page in swap cache. For example, if the swap cache
>>page is set to 0x3307c and the wrong pte is set to 0x3307e (due to
>>newpage bit), I won't be able to find it.
>>
>
>This is an excellent description of the bug that I just spend 5 days chasing.
>
>Unfortunately, I didn't until recently (yesterday) understand enough of the
>swap code to tell whether you were describing something real or whether
>you were on crack.
>
>It is now fixed.
>
>pte_to_swp_entry now masks off the arch bits.
>
>				Jeff
>
>----- End forwarded message -----
>

Re: [lc-devel] UML swap bug fixed

[Rodrigo S. de C. <rc...@im...>]

On Sat, Jan 12, 2002 at 05:50:54PM +0800, David Chow wrote:
> Is this the bug that I point out in my previous mail to the list? Or it 
> is caused by a different issue? Please clarify, since I don't have 
> enough time to understand all the code at the moment. Thanks.

I forwarded an email to this list on Dec 12th concerning a bug in
User-Mode-Linux that made compressed cache not work. To be clear, with
compressed cache code this bug was triggered easier. And now it's
finally fixed and we can continue using UML for development.

Regarding the bug your pointed out, I'd like you to test 0.21pre5
version I released this week. The swapout path code was rewritten and
it's much simpler and efficient now. I tested the last code to trigger
this bug and couldn't make it hang. Neither could Livio. He tested
under UML yesterday and got disappointed that he couldn't make the
code crash. :-) In case you still have problems with the this newer
version, please let me know.

> Rodrigo Souza de Castro wrote:
> 
> >Hi there,
> >
> >Finally Jeff seems to have fixed that bug which screwed up our vswap
> >addresses. I think I can continue developing under UML :-)

Best regards,
-- 
Rodrigo S. de Castro <rc...@im...>

Re: [lc-devel] UML swap bug fixed

[David C. <dav...@sh...>]

=A6b =B6g=A4=BB, 2002-01-12 23:20, Rodrigo Souza de Castro =BCg=B9D=A1G
> On Sat, Jan 12, 2002 at 05:50:54PM +0800, David Chow wrote:
> > Is this the bug that I point out in my previous mail to the list? Or it=
=20
> > is caused by a different issue? Please clarify, since I don't have=20
> > enough time to understand all the code at the moment. Thanks.
>=20
> I forwarded an email to this list on Dec 12th concerning a bug in
> User-Mode-Linux that made compressed cache not work. To be clear, with
> compressed cache code this bug was triggered easier. And now it's
> finally fixed and we can continue using UML for development.
>=20
> Regarding the bug your pointed out, I'd like you to test 0.21pre5
> version I released this week. The swapout path code was rewritten and
> it's much simpler and efficient now. I tested the last code to trigger
> this bug and couldn't make it hang. Neither could Livio. He tested
> under UML yesterday and got disappointed that he couldn't make the
> code crash. :-) In case you still have problems with the this newer
> version, please let me know.

I have downloaded the latest 0.21pre6 .=20

It still have problem nears out of memory when I have not turned on
swap. Do you think this is something not related with compressed cache
or a vm bug from Linux?

Usually when really out of memory, the oom killer should be triggered
and will kill that process... my experiecne is the machine freezes.

I will do more testing with swap devices.
>=20
> > Rodrigo Souza de Castro wrote:
> >=20
> > >Hi there,
> > >
> > >Finally Jeff seems to have fixed that bug which screwed up our vswap
> > >addresses. I think I can continue developing under UML :-)
>=20
> Best regards,
> --=20
> Rodrigo S. de Castro <rc...@im...>

David

Re: [lc-devel] UML swap bug fixed

[Rodrigo S. de C. <rc...@im...>]

On Thu, Jan 17, 2002 at 01:00:08PM +0800, David Chow wrote:
> ?b ?g??, 2002-01-12 23:20, Rodrigo Souza de Castro ?g?D?G
> > Regarding the bug your pointed out, I'd like you to test 0.21pre5
> > version I released this week. The swapout path code was rewritten and
> > it's much simpler and efficient now. I tested the last code to trigger
> > this bug and couldn't make it hang. Neither could Livio. He tested
> > under UML yesterday and got disappointed that he couldn't make the
> > code crash. :-) In case you still have problems with the this newer
> > version, please let me know.
> 
> I have downloaded the latest 0.21pre6 . 
> 
> It still have problem nears out of memory when I have not turned on
> swap. Do you think this is something not related with compressed cache
> or a vm bug from Linux?

Not sure. Can you reproduce this problem without compressed cache
patch?

> Usually when really out of memory, the oom killer should be triggered
> and will kill that process... my experiecne is the machine freezes.

Still hard to say. If you can't reproduce with vanilla, please test
again with 0.21pre7 (there's a small fix for vswap). In the case the
problem is still there, tell me exactly what's your configuration like
and what you are trying to do (only allocation, for example, or
running a normal desktop with browser, etc) so I may be able to
reproduce it here.

Regards,
-- 
Rodrigo S. de Castro <rc...@im...>