linux-kernel - Re: [PATCH v2 1/3] of: overlay: add overlay unittest data for node names and symbols

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <CAL_Jsq+trpQsmsst=A0kZnw0S-wZ7FaLQZuB0-9e7HurPnmyxQ@mail.gmail.com>
Date:   Tue, 11 Jul 2017 14:29:29 -0500
From:   Rob Herring <robh+dt@...nel.org>
To:     Frank Rowand <frowand.list@...il.com>
Cc:     Pantelis Antoniou <pantelis.antoniou@...sulko.com>,
        Pantelis Antoniou <panto@...oniou-consulting.com>,
        "devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2 1/3] of: overlay: add overlay unittest data for node
 names and symbols

On Mon, Jul 10, 2017 at 11:52 PM, Frank Rowand <frowand.list@...il.com> wrote:
> On 07/10/17 19:31, Rob Herring wrote:
>> On Mon, Jul 10, 2017 at 2:05 PM,  <frowand.list@...il.com> wrote:
>>> From: Frank Rowand <frank.rowand@...y.com>
>>>
>>> Add nodes and properties to overlay_base and overlay dts files to
>>> test for
>>>    - incorrect existing node name detection when overlay node name
>>>      has a unit-address
>>>    - adding overlay __symbols__ properties to live tree when an
>>>      overlay is added to the live tree
>>>
>>> Expected result from patch 2/3 is overlay will update the nodes and
>>> properties for /testcase-data-2/fairway-1/ride@.../
>>>
>>> Before patch 2/3 is applied:
>>
>> This is good information, but what is patch 2/3 is less clear when
>> this is committed.
>
> Yes, but that is the best way I've figured out to convey the information.
> I'm expecting the three patches to be three consecutive commits in the
> history.  I'd love to have a way to specify what the commit id of
> patch 3 will be in the patch 1 commit message.  Given the way that
> I think git works, I don't think there is any way the git wizards
> will be able to add that feature.  Maybe it would be clearer to
> reference the short description of patch 2 and patch 3 instead.

What's wrong with:

#1 - These additional tests have the following failures which are
fixed in subsequent commits: ...
#2 - This commit fixes these unittest failures: ...
#3 - This commit fixes these unittest failures: ...

2 and 3 already do that. So if you are assuming they are all applied
together, then why summarize the whole series in patch 1?

>> And 1 and 2 are probably stable material? I'd just
>> note in this patch what the failures are and show before and after
>> results in the patch that changes them.
>
> I consider overlays to be a not yet functional feature, that
> still needs a some more code before being usable. In that case,
> I don't think it is worth marking the patches for stable.

Well, people are using them "in production" though you do need
additional support. I'd hope that is all outside the core DT code
though.

But okay by me if not tagged for stable.

>>>    Console error message near end of unittest:
>>>       OF: Duplicate name in fairway-1, renamed to "ride@...#1"
>>>
>>>    $ cd /proc/device-tree/testcase-data-2/fairway-1/
>>>    $ # extra node: ride@...#1
>>>    $ ls
>>>    #address-cells  linux,phandle   phandle         ride@200
>>>    #size-cells     name            ride@100        status
>>>    compatible      orientation     ride@...#1
>>>    $ cd /proc/device-tree/testcase-data-2/fairway-1/ride@.../
>>>    $ ls track@...ncline_up
>>>    ls: track@...ncline_up: No such file or directory
>>>    $ ls track@...ncline_up
>>>    ls: track@...ncline_up: No such file or directory
>>
>> [...]
>>
>>> diff --git a/drivers/of/unittest-data/Makefile b/drivers/of/unittest-data/Makefile
>>> index 6e00a9c69e58..dae2fe23cd2e 100644
>>> --- a/drivers/of/unittest-data/Makefile
>>> +++ b/drivers/of/unittest-data/Makefile
>>> @@ -1,11 +1,13 @@
>>>  obj-y += testcases.dtb.o
>>>  obj-y += overlay.dtb.o
>>>  obj-y += overlay_bad_phandle.dtb.o
>>> +obj-y += overlay_bad_symbol.dtb.o
>>>  obj-y += overlay_base.dtb.o
>>
>> There's no reason for these all to be 1 per line.
>
> OK.  Do you prefer something like:
>
>    obj-y += testcases.dtb.o overlay.dtb.o overlay_bad_phandle.dtb.o \
>             overlay_bad_symbol.dtb.o overlay_base.dtb.o

I think this is more inline with other makefiles. Though it's probably
all over the map.

> My preference is to keep the objects in alphabetic order.  That might
> argue for something easier to read and update, like:
>
>    obj-y += testcases.dtb.o \
>             overlay.dtb.o \
>             overlay_bad_phandle.dtb.o \
>             overlay_bad_symbol.dtb.o \
>             overlay_base.dtb.o

Changing to this is probably not worth doing.

>> Also, should the
>> overlay dtb's be conditioned on CONFIG_OF_OVERLAY (or whatever we call
>> it)?
>
> I think so.  I'll verify that it doesn't break anything (and if so fix
> the breakage).
>
>
>>
>> But this is fine. That can all be a followup patch.
>>
>