| Message ID | 20200226052934.9221-1-oohall@gmail.com |
|---|---|
| State | Accepted |
| Headers | show |
| Series | firenze-pci: Fix infinite loop in firenze_pci_add_loc_code() | expand |
| Context | Check | Description |
|---|---|---|
| snowpatch_ozlabs/apply_patch | success | Successfully applied on branch master (f123417068e51842004bdc047c8c5107b70442ef) |
| snowpatch_ozlabs/snowpatch_job_snowpatch-skiboot | success | Test snowpatch/job/snowpatch-skiboot on branch master |
| snowpatch_ozlabs/snowpatch_job_snowpatch-skiboot-dco | success | Signed-off-by present |
On 2/26/2020 2:29 AM, Oliver O'Halloran wrote: > */ > - for (p = np->parent; p; p = np->parent) { > + for (p = np->parent; p; p = p->parent) { > blcode = dt_prop_get_def(p, "ibm,slot-location-code", NULL); Good catch! I should have seen it the first time.. -Klaus
On Thu, Feb 27, 2020 at 10:46 PM Klaus Heinrich Kiwi <klaus@linux.vnet.ibm.com> wrote: > > > > On 2/26/2020 2:29 AM, Oliver O'Halloran wrote: > > > */ > > - for (p = np->parent; p; p = np->parent) { > > + for (p = np->parent; p; p = p->parent) { > > blcode = dt_prop_get_def(p, "ibm,slot-location-code", NULL); > > Good catch! Less of a catch and more blundering into it after It bricked my Tuleta :( > I should have seen it the first time.. > > -Klaus I should probably not tell people to use bad code, oh well.
On Wed, Feb 26, 2020 at 4:29 PM Oliver O'Halloran <oohall@gmail.com> wrote: > > If ibm,slot-location-code isn't in a PCI device's parent node the loop > to search for it will never terminate since p = np->parent is always > going to return the same result. > > Fixes: ab1b05d29f5e ("PCI: create optional loc-code platform callback") > Signed-off-by: Oliver O'Halloran <oohall@gmail.com> Merged as 82aed17a5468aff6b600ee1694a10a60f942c018
diff --git a/platforms/ibm-fsp/firenze-pci.c b/platforms/ibm-fsp/firenze-pci.c index efbe5525da4b..12e0e98991aa 100644 --- a/platforms/ibm-fsp/firenze-pci.c +++ b/platforms/ibm-fsp/firenze-pci.c @@ -987,7 +987,7 @@ void firenze_pci_add_loc_code(struct dt_node *np, struct pci_device *pd) * prefer fully-qualified slot-location-code, walk-up parent tree * to find one */ - for (p = np->parent; p; p = np->parent) { + for (p = np->parent; p; p = p->parent) { blcode = dt_prop_get_def(p, "ibm,slot-location-code", NULL); if (blcode) break;
If ibm,slot-location-code isn't in a PCI device's parent node the loop to search for it will never terminate since p = np->parent is always going to return the same result. Fixes: ab1b05d29f5e ("PCI: create optional loc-code platform callback") Signed-off-by: Oliver O'Halloran <oohall@gmail.com> --- platforms/ibm-fsp/firenze-pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)