Please add my temporary MIPS fixes branch to linux-next

Please add my temporary MIPS fixes branch to linux-next

[James Hogan]

Hi Stephen & MIPS people

v3.16 is fast approaching and there are quite a few important MIPS
patches pending. Since Ralf appears to be unavailable at the moment I've
reviewed and applied some of those patches which are least controversial
to a fixes branch with the intention of sending a pull request to Ralf &
Linus so that one of them can hopefully merge it before the release.

Please could the following branch be added to linux-next:

git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/mips.git
branch: mips-fixes

Details of fixes below.

Thanks
James

The current shortlog looks like this:

Aaro Koskinen (1):
      MIPS: OCTEON: make get_system_type() thread-safe

Markos Chandras (5):
      MIPS: syscall: Fix AUDIT value for O32 processes on MIPS64
      MIPS: scall64-o32: Fix indirect syscall detection
      MIPS: EVA: Add new EVA header
      MIPS: Malta: EVA: Rename 'eva_entry' to 'platform_eva_init'
      MIPS: CPS: Initialize EVA before bringing up VPEs from secondary cores

Paul Burton (1):
      MIPS: prevent user from setting FCSR cause bits

Rob Kendrick (1):
      MIPS: math-emu: cp1emu: Fix typo when returning to register file


If I've missed any other critical fixes for v3.16 please let me know.



I haven't included the patches below, even though they are important, as
I'm less sure about them. Comments welcome.


This one fixes mips32 debian boot, but changes the layout of the
NT_PRSTATUS regset which is accessible through ptrace. I don't believe
this will break anything, but there are other patches pending in the
patchset to fix up the regset stuff properly anyway (as it is already
broken for core dumps) and I don't really want to take the risk without
Ralf's okay.

Alex Smith (1):
      MIPS: O32/32-bit: Fix bug which can cause incorrect system call
restarts


This one I'm not confident about, hasn't had any comments on the list,
and has been broken for quite a while anyway, so I haven't risked it.

Huacai Chen (1):
      MIPS: tlbex: fix a missing statement for HUGETLB

Re: Please add my temporary MIPS fixes branch to linux-next

[David Daney]

On 07/30/2014 09:00 AM, James Hogan wrote:
[...]

> ... Since Ralf appears to be unavailable at the moment ...
>
[...]

This doesn't seem to be a temporary situation.

If we want to keep the MIPS architecture in a functioning state, it may 
make sense to establish an alternate path for merging MIPS patches upstream.

David Daney.

Re: Please add my temporary MIPS fixes branch to linux-next

[Alex Smith]

On 30 July 2014 17:00, James Hogan <james.hogan@imgtec.com> wrote:
> This one fixes mips32 debian boot, but changes the layout of the
> NT_PRSTATUS regset which is accessible through ptrace. I don't believe
> this will break anything, but there are other patches pending in the
> patchset to fix up the regset stuff properly anyway (as it is already
> broken for core dumps) and I don't really want to take the risk without
> Ralf's okay.
>
> Alex Smith (1):
>       MIPS: O32/32-bit: Fix bug which can cause incorrect system call
> restarts

Right now the NT_PRSTATUS regset can't be relied upon to return the
same layout anyway, as it can differ depending on whether the kernel
is 32- or 64-bit, as well as with a couple of other Kconfig options.
Changing it in this patch shouldn't make things any worse. As you say
other patches in the series properly fix the regset/core dump
situation, though I suppose they aren't too critical for 3.16 as
nobody seems to have noticed/cared that core dumps aren't working
until now!

Thanks,
Alex

Re: Please add my temporary MIPS fixes branch to linux-next

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1507 bytes --]

Hi James,

On Wed, 30 Jul 2014 17:00:29 +0100 James Hogan <james.hogan@imgtec.com> wrote:
>
> v3.16 is fast approaching and there are quite a few important MIPS
> patches pending. Since Ralf appears to be unavailable at the moment I've
> reviewed and applied some of those patches which are least controversial
> to a fixes branch with the intention of sending a pull request to Ralf &
> Linus so that one of them can hopefully merge it before the release.
> 
> Please could the following branch be added to linux-next:
> 
> git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/mips.git
> branch: mips-fixes

Added from today.

Thanks for adding your subsystem tree as a participant of linux-next.  As
you may know, this is not a judgment of your code.  The purpose of
linux-next is for integration testing and to lower the impact of
conflicts between subsystems in the next merge window. 

You will need to ensure that the patches/commits in your tree/series have
been:
     * submitted under GPL v2 (or later) and include the Contributor's
        Signed-off-by,
     * posted to the relevant mailing list,
     * reviewed by you (or another maintainer of your subsystem tree),
     * successfully unit tested, and 
     * destined for the current or next Linux merge window.

Basically, this should be just what you would send to Linus (or ask him
to fetch).  It is allowed to be rebased if you deem it necessary.

-- 
Cheers,
Stephen Rothwell 
sfr@canb.auug.org.au

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: Please add my temporary MIPS fixes branch to linux-next

[James Hogan]

Hi Alex,

On 30/07/14 19:17, Alex Smith wrote:
> On 30 July 2014 17:00, James Hogan <james.hogan@imgtec.com> wrote:
>> This one fixes mips32 debian boot, but changes the layout of the
>> NT_PRSTATUS regset which is accessible through ptrace. I don't believe
>> this will break anything, but there are other patches pending in the
>> patchset to fix up the regset stuff properly anyway (as it is already
>> broken for core dumps) and I don't really want to take the risk without
>> Ralf's okay.
>>
>> Alex Smith (1):
>>       MIPS: O32/32-bit: Fix bug which can cause incorrect system call
>> restarts
> 
> Right now the NT_PRSTATUS regset can't be relied upon to return the
> same layout anyway, as it can differ depending on whether the kernel
> is 32- or 64-bit, as well as with a couple of other Kconfig options.
> Changing it in this patch shouldn't make things any worse.

Yeh, that is true, and it does fix a real regression.

Since Ralf does appear to be online I'll apply the patch unless he objects.

Cheers
James

Re: Please add my temporary MIPS fixes branch to linux-next

[James Hogan]

Hi Ralf,

On 30/07/14 17:00, James Hogan wrote:
> Hi Stephen & MIPS people
> 
> v3.16 is fast approaching and there are quite a few important MIPS
> patches pending. Since Ralf appears to be unavailable at the moment I've
> reviewed and applied some of those patches which are least controversial
> to a fixes branch with the intention of sending a pull request to Ralf &
> Linus so that one of them can hopefully merge it before the release.

I sent you a pull request for these fixes prior to v3.16 but
unfortunately they still missed the release, and only two of the patches
were applied in the main v3.17 MIPS merge.

What are you intentions for the remaining fixes from Markos & Aaro? I'm
happy for my branch to be dropped from linux-next if you want to apply
them yourself, or to drop the two patches you applied so you can merge
it cleanly.

Thanks
James

> 
> Please could the following branch be added to linux-next:
> 
> git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/mips.git
> branch: mips-fixes
> 
> Details of fixes below.
> 
> Thanks
> James
> 
> The current shortlog looks like this:
> 
> Aaro Koskinen (1):
>       MIPS: OCTEON: make get_system_type() thread-safe
> 
> Markos Chandras (5):
>       MIPS: syscall: Fix AUDIT value for O32 processes on MIPS64
>       MIPS: scall64-o32: Fix indirect syscall detection
>       MIPS: EVA: Add new EVA header
>       MIPS: Malta: EVA: Rename 'eva_entry' to 'platform_eva_init'
>       MIPS: CPS: Initialize EVA before bringing up VPEs from secondary cores
> 
> Paul Burton (1):
>       MIPS: prevent user from setting FCSR cause bits
> 
> Rob Kendrick (1):
>       MIPS: math-emu: cp1emu: Fix typo when returning to register file
> 
> 
> If I've missed any other critical fixes for v3.16 please let me know.
> 
> 
> 
> I haven't included the patches below, even though they are important, as
> I'm less sure about them. Comments welcome.
> 
> 
> This one fixes mips32 debian boot, but changes the layout of the
> NT_PRSTATUS regset which is accessible through ptrace. I don't believe
> this will break anything, but there are other patches pending in the
> patchset to fix up the regset stuff properly anyway (as it is already
> broken for core dumps) and I don't really want to take the risk without
> Ralf's okay.
> 
> Alex Smith (1):
>       MIPS: O32/32-bit: Fix bug which can cause incorrect system call
> restarts
> 
> 
> This one I'm not confident about, hasn't had any comments on the list,
> and has been broken for quite a while anyway, so I haven't risked it.
> 
> Huacai Chen (1):
>       MIPS: tlbex: fix a missing statement for HUGETLB
> 
> 

Re: Please add my temporary MIPS fixes branch to linux-next

[Aaro Koskinen]

Hi,

On Mon, Aug 11, 2014 at 12:09:36PM +0100, James Hogan wrote:
> On 30/07/14 17:00, James Hogan wrote:
> > Hi Stephen & MIPS people
> > 
> > v3.16 is fast approaching and there are quite a few important MIPS
> > patches pending. Since Ralf appears to be unavailable at the moment I've
> > reviewed and applied some of those patches which are least controversial
> > to a fixes branch with the intention of sending a pull request to Ralf &
> > Linus so that one of them can hopefully merge it before the release.
> 
> I sent you a pull request for these fixes prior to v3.16 but
> unfortunately they still missed the release, and only two of the patches
> were applied in the main v3.17 MIPS merge.
> 
> What are you intentions for the remaining fixes from Markos & Aaro?

Ralf, any comments? I think at least this fix is imporant,
because currently users can hang OCTEON systems simply by lauching
multiple readers of the proc file:

https://git.kernel.org/cgit/linux/kernel/git/jhogan/mips.git/commit/?id=08a9c3c9afcf6e8975fc1c9a97e871e6a039c25a

A.

Re: Please add my temporary MIPS fixes branch to linux-next

[Markos Chandras]

On 08/23/2014 09:24 PM, Aaro Koskinen wrote:
> Hi,
> 
> On Mon, Aug 11, 2014 at 12:09:36PM +0100, James Hogan wrote:
>> On 30/07/14 17:00, James Hogan wrote:
>>> Hi Stephen & MIPS people
>>>
>>> v3.16 is fast approaching and there are quite a few important MIPS
>>> patches pending. Since Ralf appears to be unavailable at the moment I've
>>> reviewed and applied some of those patches which are least controversial
>>> to a fixes branch with the intention of sending a pull request to Ralf &
>>> Linus so that one of them can hopefully merge it before the release.
>>
>> I sent you a pull request for these fixes prior to v3.16 but
>> unfortunately they still missed the release, and only two of the patches
>> were applied in the main v3.17 MIPS merge.
>>
>> What are you intentions for the remaining fixes from Markos & Aaro?
> 
> Ralf, any comments? I think at least this fix is imporant,
> because currently users can hang OCTEON systems simply by lauching
> multiple readers of the proc file:
> 
> https://git.kernel.org/cgit/linux/kernel/git/jhogan/mips.git/commit/?id=08a9c3c9afcf6e8975fc1c9a97e871e6a039c25a
> 
> A.
> 
Hi,

I plan to send GregKH another set of fixes to backport to 3.16, and
since this patch is now in mainline[1], I will include this one as well.

[1]
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=608308682addfdc7b8e2aee88f0e028331d88e4d

-- 
markos

Re: Please add my temporary MIPS fixes branch to linux-next

[James Hogan]

On 26/08/14 10:45, Markos Chandras wrote:
> On 08/23/2014 09:24 PM, Aaro Koskinen wrote:
>> Hi,
>>
>> On Mon, Aug 11, 2014 at 12:09:36PM +0100, James Hogan wrote:
>>> On 30/07/14 17:00, James Hogan wrote:
>>>> Hi Stephen & MIPS people
>>>>
>>>> v3.16 is fast approaching and there are quite a few important MIPS
>>>> patches pending. Since Ralf appears to be unavailable at the moment I've
>>>> reviewed and applied some of those patches which are least controversial
>>>> to a fixes branch with the intention of sending a pull request to Ralf &
>>>> Linus so that one of them can hopefully merge it before the release.
>>>
>>> I sent you a pull request for these fixes prior to v3.16 but
>>> unfortunately they still missed the release, and only two of the patches
>>> were applied in the main v3.17 MIPS merge.
>>>
>>> What are you intentions for the remaining fixes from Markos & Aaro?
>>
>> Ralf, any comments? I think at least this fix is imporant,
>> because currently users can hang OCTEON systems simply by lauching
>> multiple readers of the proc file:
>>
>> https://git.kernel.org/cgit/linux/kernel/git/jhogan/mips.git/commit/?id=08a9c3c9afcf6e8975fc1c9a97e871e6a039c25a
>>
>> A.
>>
> Hi,
> 
> I plan to send GregKH another set of fixes to backport to 3.16, and
> since this patch is now in mainline[1], I will include this one as well.
> 
> [1]
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=608308682addfdc7b8e2aee88f0e028331d88e4d

This particular fix is already marked for stable.

The ones I applied that aren't marked for stable (but are now needed in
linux-3.16.y) are:

f85b71ceabb9 MIPS: EVA: Add new EVA header
ca4d24f7954f MIPS: Malta: EVA: Rename 'eva_entry' to 'platform_eva_init'
6521d9a436a6 MIPS: CPS: Initialize EVA before bringing up VPEs from
secondary cores

Cheers
James

Re: Please add my temporary MIPS fixes branch to linux-next

[Markos Chandras]

On 08/26/2014 10:51 AM, James Hogan wrote:
> On 26/08/14 10:45, Markos Chandras wrote:
>> On 08/23/2014 09:24 PM, Aaro Koskinen wrote:
>>> Hi,
>>>
>>> On Mon, Aug 11, 2014 at 12:09:36PM +0100, James Hogan wrote:
>>>> On 30/07/14 17:00, James Hogan wrote:
>>>>> Hi Stephen & MIPS people
>>>>>
>>>>> v3.16 is fast approaching and there are quite a few important MIPS
>>>>> patches pending. Since Ralf appears to be unavailable at the moment I've
>>>>> reviewed and applied some of those patches which are least controversial
>>>>> to a fixes branch with the intention of sending a pull request to Ralf &
>>>>> Linus so that one of them can hopefully merge it before the release.
>>>>
>>>> I sent you a pull request for these fixes prior to v3.16 but
>>>> unfortunately they still missed the release, and only two of the patches
>>>> were applied in the main v3.17 MIPS merge.
>>>>
>>>> What are you intentions for the remaining fixes from Markos & Aaro?
>>>
>>> Ralf, any comments? I think at least this fix is imporant,
>>> because currently users can hang OCTEON systems simply by lauching
>>> multiple readers of the proc file:
>>>
>>> https://git.kernel.org/cgit/linux/kernel/git/jhogan/mips.git/commit/?id=08a9c3c9afcf6e8975fc1c9a97e871e6a039c25a
>>>
>>> A.
>>>
>> Hi,
>>
>> I plan to send GregKH another set of fixes to backport to 3.16, and
>> since this patch is now in mainline[1], I will include this one as well.
>>
>> [1]
>> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=608308682addfdc7b8e2aee88f0e028331d88e4d
> 
> This particular fix is already marked for stable.

True I didn't notice. Thanks!

-- 
markos