On Mon, Dec 11, 2017 at 11:30 AM, Guillaume Tucker <guillaume.tucker@xxxxxxxxxxxxx> wrote: > Hi Daniel, > > Please see below, I've had several bisection results pointing at > that commit over the week-end on mainline but also on linux-next > and net-next. While the peach-pi is a bit flaky at the moment > and is likely to have more than one issue, it does seem like this > commit is causing some well reproducible kernel hang. > > Here's a re-run with v4.15-rc3 showing the issue: > > https://lava.collabora.co.uk/scheduler/job/1018478 > > and here's another one with the change mentioned below reverted: > > https://lava.collabora.co.uk/scheduler/job/1018479 > > They both show a warning about "unbalanced disables for lcd_vdd", > I don't know if this is related as I haven't investigated any > further. It does appear to reliably hang with v4.15-rc3 and > boot most of the time with the commit reverted though. > > The automated kernelci.org bisection is still an experimental > tool and it may well be a false positive, so please take this > result with a pinch of salt... The patch just very minimal moves the connector cleanup around (so timing change), but except when you unload a driver (or maybe that funny EPROBE_DEFER stuff) it shouldn't matter. So if you don't have more info than "seems to hang a bit more" I have no idea what's wrong. The patch itself should work, at least it survived quite some serious testing we do on everything. -Daniel > Hope this helps! > > Best wishes, > Guillaume > > > -------- Forwarded Message -------- > Subject: mainline/master boot bisection: v4.15-rc3 on peach-pi #3228-staging > Date: Mon, 11 Dec 2017 08:25:55 +0000 (UTC) > From: kernelci.org bot <bot@xxxxxxxxxxxx> > To: guillaume.tucker@xxxxxxxxxxxxx > > Bisection result for mainline/master (v4.15-rc3) on peach-pi > > Good known revision: > > c6b3e96 Merge branch 'for-linus' of > git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux > > Bad known revision: > > 50c4c4e Linux 4.15-rc3 > > Extra parameters: > > Tree: mainline > Branch: master > Target: peach-pi > Lab: lab-collabora > Defconfig: exynos_defconfig > Plan: boot > > > Breaking commit found: > > ------------------------------------------------------------------------------- > commit a703c55004e1c5076d57e43771b3e11117796ea0 > Author: Daniel Vetter <daniel.vetter@xxxxxxxx> > Date: Mon Dec 4 21:48:18 2017 +0100 > > drm: safely free connectors from connector_iter > In > commit 613051dac40da1751ab269572766d3348d45a197 > Author: Daniel Vetter <daniel.vetter@xxxxxxxx> > Date: Wed Dec 14 00:08:06 2016 +0100 > drm: locking&new iterators for connector_list > we've went to extreme lengths to make sure connector iterations > works > in any context, without introducing any additional locking context. > This worked, except for a small fumble in the implementation: > When we actually race with a concurrent connector unplug event, and > our temporary connector reference turns out to be the final one, then > everything breaks: We call the connector release function from > whatever context we happen to be in, which can be an irq/atomic > context. And connector freeing grabs all kinds of locks and stuff. > Fix this by creating a specially safe put function for > connetor_iter, > which (in this rare case) punts the cleanup to a worker. > Reported-by: Ben Widawsky <ben@xxxxxxxxxxxx> > Cc: Ben Widawsky <ben@xxxxxxxxxxxx> > Fixes: 613051dac40d ("drm: locking&new iterators for connector_list") > Cc: Dave Airlie <airlied@xxxxxxxxx> > Cc: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> > Cc: Sean Paul <seanpaul@xxxxxxxxxxxx> > Cc: <stable@xxxxxxxxxxxxxxx> # v4.11+ > Reviewed-by: Dave Airlie <airlied@xxxxxxxxx> > Signed-off-by: Daniel Vetter <daniel.vetter@xxxxxxxxx> > Link: > https://patchwork.freedesktop.org/patch/msgid/20171204204818.24745-1-daniel.vetter@xxxxxxxx > > diff --git a/drivers/gpu/drm/drm_connector.c > b/drivers/gpu/drm/drm_connector.c > index 25f4b2e..4820141 100644 > --- a/drivers/gpu/drm/drm_connector.c > +++ b/drivers/gpu/drm/drm_connector.c > @@ -152,6 +152,16 @@ static void drm_connector_free(struct kref *kref) > connector->funcs->destroy(connector); > } > +static void drm_connector_free_work_fn(struct work_struct *work) > +{ > + struct drm_connector *connector = > + container_of(work, struct drm_connector, free_work); > + struct drm_device *dev = connector->dev; > + > + drm_mode_object_unregister(dev, &connector->base); > + connector->funcs->destroy(connector); > +} > + > /** > * drm_connector_init - Init a preallocated connector > * @dev: DRM device > @@ -181,6 +191,8 @@ int drm_connector_init(struct drm_device *dev, > if (ret) > return ret; > + INIT_WORK(&connector->free_work, drm_connector_free_work_fn); > + > connector->base.properties = &connector->properties; > connector->dev = dev; > connector->funcs = funcs; > @@ -529,6 +541,18 @@ void drm_connector_list_iter_begin(struct drm_device > *dev, > } > EXPORT_SYMBOL(drm_connector_list_iter_begin); > +/* > + * Extra-safe connector put function that works in any context. Should only > be > + * used from the connector_iter functions, where we never really expect to > + * actually release the connector when dropping our final reference. > + */ > +static void > +drm_connector_put_safe(struct drm_connector *conn) > +{ > + if (refcount_dec_and_test(&conn->base.refcount.refcount)) > + schedule_work(&conn->free_work); > +} > + > /** > * drm_connector_list_iter_next - return next connector > * @iter: connectr_list iterator > @@ -561,7 +585,7 @@ drm_connector_list_iter_next(struct > drm_connector_list_iter *iter) > spin_unlock_irqrestore(&config->connector_list_lock, flags); > if (old_conn) > - drm_connector_put(old_conn); > + drm_connector_put_safe(old_conn); > return iter->conn; > } > @@ -580,7 +604,7 @@ void drm_connector_list_iter_end(struct > drm_connector_list_iter *iter) > { > iter->dev = NULL; > if (iter->conn) > - drm_connector_put(iter->conn); > + drm_connector_put_safe(iter->conn); > lock_release(&connector_list_iter_dep_map, 0, _RET_IP_); > } > EXPORT_SYMBOL(drm_connector_list_iter_end); > diff --git a/drivers/gpu/drm/drm_mode_config.c > b/drivers/gpu/drm/drm_mode_config.c > index cda8bfa..cc78b3d 100644 > --- a/drivers/gpu/drm/drm_mode_config.c > +++ b/drivers/gpu/drm/drm_mode_config.c > @@ -431,6 +431,8 @@ void drm_mode_config_cleanup(struct drm_device *dev) > drm_connector_put(connector); > } > drm_connector_list_iter_end(&conn_iter); > + /* connector_iter drops references in a work item. */ > + flush_scheduled_work(); > if (WARN_ON(!list_empty(&dev->mode_config.connector_list))) { > drm_connector_list_iter_begin(dev, &conn_iter); > drm_for_each_connector_iter(connector, &conn_iter) > diff --git a/include/drm/drm_connector.h b/include/drm/drm_connector.h > index df9807a..a4649c5 100644 > --- a/include/drm/drm_connector.h > +++ b/include/drm/drm_connector.h > @@ -916,6 +916,14 @@ struct drm_connector { > uint8_t num_h_tile, num_v_tile; > uint8_t tile_h_loc, tile_v_loc; > uint16_t tile_h_size, tile_v_size; > + > + /** > + * @free_work: > + * > + * Work used only by &drm_connector_iter to be able to clean up a > + * connector from any context. > + */ > + struct work_struct free_work; > }; > #define obj_to_connector(x) container_of(x, struct drm_connector, base) > ------------------------------------------------------------------------------- > > > Git bisection log: > > ------------------------------------------------------------------------------- > git bisect start > # good: [c6b3e9693f8a32ba3b07e2f2723886ea2aff4e94] Merge branch 'for-linus' > of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux > git bisect good c6b3e9693f8a32ba3b07e2f2723886ea2aff4e94 > # bad: [50c4c4e268a2d7a3e58ebb698ac74da0de40ae36] Linux 4.15-rc3 > git bisect bad 50c4c4e268a2d7a3e58ebb698ac74da0de40ae36 > # bad: [e9ef1fe312b533592e39cddc1327463c30b0ed8d] Merge > git://git.kernel.org/pub/scm/linux/kernel/git/davem/net > git bisect bad e9ef1fe312b533592e39cddc1327463c30b0ed8d > # bad: [77071bc6c472bb0b36818f3e9595114cdf98c86d] Merge tag 'media/v4.15-2' > of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media > git bisect bad 77071bc6c472bb0b36818f3e9595114cdf98c86d > # bad: [4066aa72f9f2886105c6f747d7f9bd4f14f53c12] Merge tag > 'drm-fixes-for-v4.15-rc3' of git://people.freedesktop.org/~airlied/linux > git bisect bad 4066aa72f9f2886105c6f747d7f9bd4f14f53c12 > # bad: [96980844bb4b74d2e7ce93d907670658e39a3992] Merge tag > 'drm-intel-fixes-2017-12-07' of git://anongit.freedesktop.org/drm/drm-intel > into drm-fixes > git bisect bad 96980844bb4b74d2e7ce93d907670658e39a3992 > # bad: [120a264f9c2782682027d931d83dcbd22e01da80] drm/exynos: gem: Drop > NONCONTIG flag for buffers allocated without IOMMU > git bisect bad 120a264f9c2782682027d931d83dcbd22e01da80 > # good: [2bf257d662509553ae226239e7dc1c3d00636ca6] drm/ttm: roundup the > shrink request to prevent skip huge pool > git bisect good 2bf257d662509553ae226239e7dc1c3d00636ca6 > # good: [db8f884ca7fe6af64d443d1510464efe23826131] Merge branch > 'drm-fixes-4.15' of git://people.freedesktop.org/~agd5f/linux into drm-fixes > git bisect good db8f884ca7fe6af64d443d1510464efe23826131 > # bad: [bd3a3a2e92624942a143e485c83e641b2492d828] Merge tag > 'drm-misc-fixes-2017-12-06' of git://anongit.freedesktop.org/drm/drm-misc > into drm-fixes > git bisect bad bd3a3a2e92624942a143e485c83e641b2492d828 > # bad: [a703c55004e1c5076d57e43771b3e11117796ea0] drm: safely free > connectors from connector_iter > git bisect bad a703c55004e1c5076d57e43771b3e11117796ea0 > # first bad commit: [a703c55004e1c5076d57e43771b3e11117796ea0] drm: safely > free connectors from connector_iter > ------------------------------------------------------------------------------- -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch -- To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
 |