linux-kernel - Re: linux-next: build failure after merge of the drm-misc tree

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] [thread-next>] [day] [month] [year] [list]
Date:   Fri, 05 Nov 2021 13:03:43 +0200
From:   Jani Nikula <jani.nikula@...el.com>
To:     Stephen Rothwell <sfr@...b.auug.org.au>,
        Dave Airlie <airlied@...ux.ie>
Cc:     Daniel Vetter <daniel.vetter@...ll.ch>,
        Intel Graphics <intel-gfx@...ts.freedesktop.org>,
        DRI <dri-devel@...ts.freedesktop.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Linux Next Mailing List <linux-next@...r.kernel.org>,
        Thomas Zimmermann <tzimmermann@...e.de>,
        Maarten Lankhorst <maarten.lankhorst@...ux.intel.com>,
        Maxime Ripard <mripard@...nel.org>
Subject: Re: linux-next: build failure after merge of the drm-misc tree

On Fri, 05 Nov 2021, Stephen Rothwell <sfr@...b.auug.org.au> wrote:
> Hi all,
>
> On Mon, 1 Nov 2021 19:42:23 +1100 Stephen Rothwell <sfr@...b.auug.org.au> wrote:
>>
>> On Fri, 15 Oct 2021 20:26:48 +1100 Stephen Rothwell <sfr@...b.auug.org.au> wrote:
>> >
>> > After merging the drm-misc tree, today's linux-next build (arm
>> > multi_v7_defconfig) failed like this:
>> > 
>> > drivers/gpu/drm/drm_modeset_lock.c:111:29: error: conflicting types for '__stack_depot_save'
>> >   111 | static depot_stack_handle_t __stack_depot_save(void)
>> >       |                             ^~~~~~~~~~~~~~~~~~
>> > In file included from include/linux/page_ext.h:7,
>> >                  from include/linux/mm.h:25,
>> >                  from include/linux/kallsyms.h:13,
>> >                  from include/linux/bpf.h:20,
>> >                  from include/linux/bpf-cgroup.h:5,
>> >                  from include/linux/cgroup-defs.h:22,
>> >                  from include/linux/cgroup.h:28,
>> >                  from include/linux/memcontrol.h:13,
>> >                  from include/linux/swap.h:9,
>> >                  from include/linux/suspend.h:5,
>> >                  from include/linux/regulator/consumer.h:35,
>> >                  from include/linux/i2c.h:18,
>> >                  from include/drm/drm_crtc.h:28,
>> >                  from include/drm/drm_atomic.h:31,
>> >                  from drivers/gpu/drm/drm_modeset_lock.c:24:
>> > include/linux/stackdepot.h:18:22: note: previous declaration of '__stack_depot_save' was here
>> >    18 | depot_stack_handle_t __stack_depot_save(unsigned long *entries,
>> >       |                      ^~~~~~~~~~~~~~~~~~
>> > 
>> > Caused by commit
>> > 
>> >   cd06ab2fd48f ("drm/locking: add backtrace for locking contended locks without backoff")
>> > 
>> > This may only have been revealed because of another fix I have had to
>> > apply today.
>> > 
>> > I have applied the following patch for today.
>> > 
>> > From: Stephen Rothwell <sfr@...b.auug.org.au>
>> > Date: Fri, 15 Oct 2021 20:17:52 +1100
>> > Subject: [PATCH] drm/locking: fix for name conflict
>> > 
>> > Fixes: cd06ab2fd48f ("drm/locking: add backtrace for locking contended locks without backoff")
>> > Signed-off-by: Stephen Rothwell <sfr@...b.auug.org.au>
>> > ---
>> >  drivers/gpu/drm/drm_modeset_lock.c | 6 +++---
>> >  1 file changed, 3 insertions(+), 3 deletions(-)
>> > 
>> > diff --git a/drivers/gpu/drm/drm_modeset_lock.c b/drivers/gpu/drm/drm_modeset_lock.c
>> > index 4d32b61fa1fd..ee36dd20900d 100644
>> > --- a/drivers/gpu/drm/drm_modeset_lock.c
>> > +++ b/drivers/gpu/drm/drm_modeset_lock.c
>> > @@ -79,7 +79,7 @@
>> >  static DEFINE_WW_CLASS(crtc_ww_class);
>> >  
>> >  #if IS_ENABLED(CONFIG_DRM_DEBUG_MODESET_LOCK)
>> > -static noinline depot_stack_handle_t __stack_depot_save(void)
>> > +static noinline depot_stack_handle_t __drm_stack_depot_save(void)
>> >  {
>> >  	unsigned long entries[8];
>> >  	unsigned int n;
>> > @@ -108,7 +108,7 @@ static void __stack_depot_print(depot_stack_handle_t stack_depot)
>> >  	kfree(buf);
>> >  }
>> >  #else /* CONFIG_DRM_DEBUG_MODESET_LOCK */
>> > -static depot_stack_handle_t __stack_depot_save(void)
>> > +static depot_stack_handle_t __drm_stack_depot_save(void)
>> >  {
>> >  	return 0;
>> >  }
>> > @@ -317,7 +317,7 @@ static inline int modeset_lock(struct drm_modeset_lock *lock,
>> >  		ret = 0;
>> >  	} else if (ret == -EDEADLK) {
>> >  		ctx->contended = lock;
>> > -		ctx->stack_depot = __stack_depot_save();
>> > +		ctx->stack_depot = __drm_stack_depot_save();
>> >  	}
>> >  
>> >  	return ret;
>> 
>> This has reappeared today.  I don't know what happened to the drm-misc
>> tree over the weeked :-(
>> 
>> I have reapplied the above fix.
>
> So the above drm-misc commit is now in the drm tree, but its fix up
> commit vanished from the drm-misc tree over the past weekend :-(

Cc: drm-misc maintainers.

We normally point drm-misc/for-linux-next at drm-misc-next, *except* to
drm-misc-next-fixes during the merge window. This is because
drm-misc-next already starts accumulating stuff that's headed to one
release later, e.g. currently v5.17. I think that's part of the reason.

I probably should have pushed c4f08d7246a5 ("drm/locking: fix
__stack_depot_* name conflict") to drm-misc-next-fixes.

There's still something funny going on, because the drm-misc-next pull
request [1] isn't part of the drm pull request for v5.16 [2]. Is there
going to be another drm pull?

BR,
Jani.


[1] https://lore.kernel.org/r/20211014120452.2wicnt6hobu3kbwb@gilmour
[2] https://lore.kernel.org/r/CAPM=9tyOyz4_-OdjDduFkponSXycO6maBDFsWGTLv+j=_Vp6ww@mail.gmail.com



-- 
Jani Nikula, Intel Open Source Graphics Center