linux-kernel - Re: [PATCH v6 04/12] drm/panfrost: Handle error when allocating AS number

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: <p252igaqigmqo6axea35lqsy6ht67xgpu32r3hxn2t7eiqt3qd@vx6ngti3ljvw>
Date: Fri, 17 Oct 2025 14:31:10 +0100
From: Adrián Larumbe <adrian.larumbe@...labora.com>
To: Steven Price <steven.price@....com>
Cc: linux-kernel@...r.kernel.org, dri-devel@...ts.freedesktop.org, 
	Boris Brezillon <boris.brezillon@...labora.com>, kernel@...labora.com, Rob Herring <robh@...nel.org>, 
	Maarten Lankhorst <maarten.lankhorst@...ux.intel.com>, Maxime Ripard <mripard@...nel.org>, 
	Thomas Zimmermann <tzimmermann@...e.de>, David Airlie <airlied@...il.com>, 
	Simona Vetter <simona@...ll.ch>
Subject: Re: [PATCH v6 04/12] drm/panfrost: Handle error when allocating AS
 number

On 17.10.2025 14:25, Steven Price wrote:
> On 15/10/2025 01:09, Adrián Larumbe wrote:
> > If we reach the beginning of the LRU AS list, then return an error.
> >
> > Reviewed-by: Steven Price <steven.price@....com>
> > Reviewed-by: Boris Brezillon <boris.brezillon@...labora.com>
> > Signed-off-by: Adrián Larumbe <adrian.larumbe@...labora.com>
> > ---
> >  drivers/gpu/drm/panfrost/panfrost_job.c     | 6 +++++-
> >  drivers/gpu/drm/panfrost/panfrost_mmu.c     | 5 +++--
> >  drivers/gpu/drm/panfrost/panfrost_mmu.h     | 2 +-
> >  drivers/gpu/drm/panfrost/panfrost_perfcnt.c | 8 ++++++--
> >  4 files changed, 15 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/panfrost/panfrost_job.c b/drivers/gpu/drm/panfrost/panfrost_job.c
> > index 0722f297d142..d53e9db945ad 100644
> > --- a/drivers/gpu/drm/panfrost/panfrost_job.c
> > +++ b/drivers/gpu/drm/panfrost/panfrost_job.c
> > @@ -217,7 +217,11 @@ static int panfrost_job_hw_submit(struct panfrost_job *job, int js)
> >  		goto err_hwsubmit;
> >  	}
> >
> > -	cfg = panfrost_mmu_as_get(pfdev, job->mmu);
> > +	ret = panfrost_mmu_as_get(pfdev, job->mmu);
> > +	if (ret < 0)
> > +		goto err_hwsubmit;
> > +
> > +	cfg = ret;
> >
> >  	panfrost_devfreq_record_busy(&pfdev->pfdevfreq);
> >
> > diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.c b/drivers/gpu/drm/panfrost/panfrost_mmu.c
> > index 1d696eeea2fa..cf272b167feb 100644
> > --- a/drivers/gpu/drm/panfrost/panfrost_mmu.c
> > +++ b/drivers/gpu/drm/panfrost/panfrost_mmu.c
> > @@ -258,7 +258,7 @@ static int panfrost_mmu_cfg_init(struct panfrost_mmu *mmu,
> >  	}
> >  }
> >
> > -u32 panfrost_mmu_as_get(struct panfrost_device *pfdev, struct panfrost_mmu *mmu)
> > +int panfrost_mmu_as_get(struct panfrost_device *pfdev, struct panfrost_mmu *mmu)
> >  {
> >  	int as;
> >
> > @@ -300,7 +300,8 @@ u32 panfrost_mmu_as_get(struct panfrost_device *pfdev, struct panfrost_mmu *mmu)
> >  			if (!atomic_read(&lru_mmu->as_count))
> >  				break;
> >  		}
> > -		WARN_ON(&lru_mmu->list == &pfdev->as_lru_list);
> > +		if (WARN_ON(&lru_mmu->list == &pfdev->as_lru_list))
> > +			return -EBUSY;
>
> I was about to apply this series, but sparse told me that this is
> broken. We're holding the as_lock here, so we need to drop that before
> returning.

Thanks for catching this, I think going forward I should make a point of always running
sparse on a patch series before sending it out to the ML.

Adrian

> Thanks,
>
> Steve
>
> >
> >  		list_del_init(&lru_mmu->list);
> >  		as = lru_mmu->as;
> > diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.h b/drivers/gpu/drm/panfrost/panfrost_mmu.h
> > index 022a9a74a114..e6e6966a0cca 100644
> > --- a/drivers/gpu/drm/panfrost/panfrost_mmu.h
> > +++ b/drivers/gpu/drm/panfrost/panfrost_mmu.h
> > @@ -16,7 +16,7 @@ void panfrost_mmu_fini(struct panfrost_device *pfdev);
> >  void panfrost_mmu_reset(struct panfrost_device *pfdev);
> >  void panfrost_mmu_suspend_irq(struct panfrost_device *pfdev);
> >
> > -u32 panfrost_mmu_as_get(struct panfrost_device *pfdev, struct panfrost_mmu *mmu);
> > +int panfrost_mmu_as_get(struct panfrost_device *pfdev, struct panfrost_mmu *mmu);
> >  void panfrost_mmu_as_put(struct panfrost_device *pfdev, struct panfrost_mmu *mmu);
> >
> >  struct panfrost_mmu *panfrost_mmu_ctx_get(struct panfrost_mmu *mmu);
> > diff --git a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
> > index 718eb44b40f8..7020c0192e18 100644
> > --- a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
> > +++ b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
> > @@ -130,9 +130,11 @@ static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev,
> >  		goto err_vunmap;
> >  	}
> >
> > -	perfcnt->user = user;
> > +	ret = panfrost_mmu_as_get(pfdev, perfcnt->mapping->mmu);
> > +	if (ret < 0)
> > +		goto err_vunmap;
> >
> > -	as = panfrost_mmu_as_get(pfdev, perfcnt->mapping->mmu);
> > +	as = ret;
> >  	cfg = GPU_PERFCNT_CFG_AS(as) |
> >  	      GPU_PERFCNT_CFG_MODE(GPU_PERFCNT_CFG_MODE_MANUAL);
> >
> > @@ -164,6 +166,8 @@ static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev,
> >  	/* The BO ref is retained by the mapping. */
> >  	drm_gem_object_put(&bo->base);
> >
> > +	perfcnt->user = user;
> > +
> >  	return 0;
> >
> >  err_vunmap: