[PATCH] coresight: cti: Fix remove sysfs link error.
Mike Leach
mike.leach at linaro.org
Tue Jul 21 15:50:13 EDT 2020
Hi Mathieu,
On Tue, 21 Jul 2020 at 18:51, Mathieu Poirier
<mathieu.poirier at linaro.org> wrote:
>
> Hi Mike,
>
> On Tue, Jul 21, 2020 at 12:17:36PM +0100, Mike Leach wrote:
> > CTI code to remove sysfs link to other devices on shutdown, incorrectly
> > tries to remove a single ended link when these are all double ended. This
> > implementation leaves elements in the link info structure undefined which
> > results in a crash in recent tests for driver module unload.
> >
> > This patch corrects the link removal code.
> >
> > Fixes: 73274abb6557 (coresight: cti: Add in sysfs links to other coresight devices)
> >
>
> Please remove the extra new line, add a "Reported-by:" for Tingwei and remove the
> '.' at the end of the title. Up to now I've been removing those manually but it
> causes problems when Greg has to pickup a set directly from the mailing list.
>
Will do.
> Also, I just noticed the other problem you shared with Tingwei regarding the
> call to cti_remove_conn_xrefs() in cti_remove(). I think it will be simpler if
> you send a 2 patch serie with those fixes.
>
The issue here is that this patch is completely independent of
Tingwei's coresight module set, and applies immediately to
coresight\next.
The additional change I suggested to completely fix the problem is
entirely dependent on changes (i.e. the introduction of cti_remove())
made by Tingwei and thus dependent on that set.
Thus I cannot issue a 2 patch set without Tingwei re-working the
module set, and the resulting set would be dependent on that set,
leaving the coresight module set broken until a second set is applied.
My expectation was that Tingwei would include the cti_remove()
suggested change as part of his re-work while rebasing on
coresight/next, leaving this patch to be applied before / afterwards.
Alternatively, this patch could simply become another patch in
Tingwei's coresight module series. (the issue it fixes doesn't appear
to have adverse effects when coresight drivers are built in).
Regards
Mike
> Thanks,
> Mathieu
>
> > Signed-off-by: Mike Leach <mike.leach at linaro.org>
> > ---
> > drivers/hwtracing/coresight/coresight-cti.c | 9 ++++++---
> > 1 file changed, 6 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/hwtracing/coresight/coresight-cti.c b/drivers/hwtracing/coresight/coresight-cti.c
> > index 3ccc703dc940..19d0b031a7df 100644
> > --- a/drivers/hwtracing/coresight/coresight-cti.c
> > +++ b/drivers/hwtracing/coresight/coresight-cti.c
> > @@ -511,12 +511,15 @@ static bool cti_add_sysfs_link(struct cti_drvdata *drvdata,
> > return !link_err;
> > }
> >
> > -static void cti_remove_sysfs_link(struct cti_trig_con *tc)
> > +static void cti_remove_sysfs_link(struct cti_drvdata *drvdata,
> > + struct cti_trig_con *tc)
> > {
> > struct coresight_sysfs_link link_info;
> >
> > + link_info.orig = drvdata->csdev;
> > link_info.orig_name = tc->con_dev_name;
> > link_info.target = tc->con_dev;
> > + link_info.target_name = dev_name(&drvdata->csdev->dev);
> > coresight_remove_sysfs_link(&link_info);
> > }
> >
> > @@ -607,7 +610,7 @@ void cti_remove_assoc_from_csdev(struct coresight_device *csdev)
> > ctidev = &ctidrv->ctidev;
> > list_for_each_entry(tc, &ctidev->trig_cons, node) {
> > if (tc->con_dev == csdev->ect_dev) {
> > - cti_remove_sysfs_link(tc);
> > + cti_remove_sysfs_link(ctidrv, tc);
> > tc->con_dev = NULL;
> > break;
> > }
> > @@ -651,7 +654,7 @@ static void cti_remove_conn_xrefs(struct cti_drvdata *drvdata)
> > if (tc->con_dev) {
> > coresight_set_assoc_ectdev_mutex(tc->con_dev,
> > NULL);
> > - cti_remove_sysfs_link(tc);
> > + cti_remove_sysfs_link(drvdata, tc);
> > tc->con_dev = NULL;
> > }
> > }
> > --
> > 2.17.1
> >
--
Mike Leach
Principal Engineer, ARM Ltd.
Manchester Design Centre. UK
More information about the linux-arm-kernel
mailing list