* [PATCH] PCI: hv: Fix double ida_free in hv_pci_probe error path
@ 2026-04-03 12:09 Sahil Chandna
2026-04-03 17:16 ` Saurabh Singh Sengar
2026-04-04 2:42 ` Manivannan Sadhasivam
0 siblings, 2 replies; 4+ messages in thread
From: Sahil Chandna @ 2026-04-03 12:09 UTC (permalink / raw)
To: kys, haiyangz, wei.liu, decui, longli, lpieralisi, kwilczynski,
mani, robh, bhelgaas, dan.j.williams, mhklinux, linux-hyperv,
linux-pci, linux-kernel
If hv_pci_probe() fails after storing the domain number in
hbus->bridge->domain_nr, there is a call to free this domain_nr via
pci_bus_release_emul_domain_nr(), however, during cleanup, the bridge
release callback pci_release_host_bridge_dev() also frees the domain_nr
causing ida_free to be called on same ID twice and triggering following
warning:
ida_free called for id=28971 which is not allocated.
WARNING: lib/idr.c:594 at ida_free+0xdf/0x160, CPU#0: kworker/0:2/198
Call Trace:
pci_bus_release_emul_domain_nr+0x17/0x20
pci_release_host_bridge_dev+0x4b/0x60
device_release+0x3b/0xa0
kobject_put+0x8e/0x220
devm_pci_alloc_host_bridge_release+0xe/0x20
devres_release_all+0x9a/0xd0
device_unbind_cleanup+0x12/0xa0
really_probe+0x1c5/0x3f0
vmbus_add_channel_work+0x135/0x1a0
Fix this by letting pci core handle the free domain_nr and remove
the explicit free called in pci-hyperv driver.
Fixes: bcce8c74f1ce ("PCI: Enable host bridge emulation for PCI_DOMAINS_GENERIC platforms")
Signed-off-by: Sahil Chandna <sahilchandna@linux.microsoft.com>
---
drivers/pci/controller/pci-hyperv.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/pci/controller/pci-hyperv.c b/drivers/pci/controller/pci-hyperv.c
index 2c7a406b4ba8..5616ad5d2a8f 100644
--- a/drivers/pci/controller/pci-hyperv.c
+++ b/drivers/pci/controller/pci-hyperv.c
@@ -3778,7 +3778,7 @@ static int hv_pci_probe(struct hv_device *hdev,
hbus->bridge->domain_nr);
if (!hbus->wq) {
ret = -ENOMEM;
- goto free_dom;
+ goto free_bus;
}
hdev->channel->next_request_id_callback = vmbus_next_request_id;
@@ -3874,8 +3874,6 @@ static int hv_pci_probe(struct hv_device *hdev,
vmbus_close(hdev->channel);
destroy_wq:
destroy_workqueue(hbus->wq);
-free_dom:
- pci_bus_release_emul_domain_nr(hbus->bridge->domain_nr);
free_bus:
kfree(hbus);
return ret;
--
2.53.0
^ permalink raw reply related [flat|nested] 4+ messages in thread* Re: [PATCH] PCI: hv: Fix double ida_free in hv_pci_probe error path
2026-04-03 12:09 [PATCH] PCI: hv: Fix double ida_free in hv_pci_probe error path Sahil Chandna
@ 2026-04-03 17:16 ` Saurabh Singh Sengar
2026-04-04 2:42 ` Manivannan Sadhasivam
1 sibling, 0 replies; 4+ messages in thread
From: Saurabh Singh Sengar @ 2026-04-03 17:16 UTC (permalink / raw)
To: Sahil Chandna
Cc: kys, haiyangz, wei.liu, decui, longli, lpieralisi, kwilczynski,
mani, robh, bhelgaas, dan.j.williams, mhklinux, linux-hyperv,
linux-pci, linux-kernel
On Fri, Apr 03, 2026 at 05:09:29AM -0700, Sahil Chandna wrote:
> If hv_pci_probe() fails after storing the domain number in
> hbus->bridge->domain_nr, there is a call to free this domain_nr via
> pci_bus_release_emul_domain_nr(), however, during cleanup, the bridge
> release callback pci_release_host_bridge_dev() also frees the domain_nr
> causing ida_free to be called on same ID twice and triggering following
> warning:
>
> ida_free called for id=28971 which is not allocated.
> WARNING: lib/idr.c:594 at ida_free+0xdf/0x160, CPU#0: kworker/0:2/198
> Call Trace:
> pci_bus_release_emul_domain_nr+0x17/0x20
> pci_release_host_bridge_dev+0x4b/0x60
> device_release+0x3b/0xa0
> kobject_put+0x8e/0x220
> devm_pci_alloc_host_bridge_release+0xe/0x20
> devres_release_all+0x9a/0xd0
> device_unbind_cleanup+0x12/0xa0
> really_probe+0x1c5/0x3f0
> vmbus_add_channel_work+0x135/0x1a0
>
> Fix this by letting pci core handle the free domain_nr and remove
> the explicit free called in pci-hyperv driver.
>
> Fixes: bcce8c74f1ce ("PCI: Enable host bridge emulation for PCI_DOMAINS_GENERIC platforms")
> Signed-off-by: Sahil Chandna <sahilchandna@linux.microsoft.com>
> ---
> drivers/pci/controller/pci-hyperv.c | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
>
> diff --git a/drivers/pci/controller/pci-hyperv.c b/drivers/pci/controller/pci-hyperv.c
> index 2c7a406b4ba8..5616ad5d2a8f 100644
> --- a/drivers/pci/controller/pci-hyperv.c
> +++ b/drivers/pci/controller/pci-hyperv.c
> @@ -3778,7 +3778,7 @@ static int hv_pci_probe(struct hv_device *hdev,
> hbus->bridge->domain_nr);
> if (!hbus->wq) {
> ret = -ENOMEM;
> - goto free_dom;
> + goto free_bus;
> }
>
> hdev->channel->next_request_id_callback = vmbus_next_request_id;
> @@ -3874,8 +3874,6 @@ static int hv_pci_probe(struct hv_device *hdev,
> vmbus_close(hdev->channel);
> destroy_wq:
> destroy_workqueue(hbus->wq);
> -free_dom:
> - pci_bus_release_emul_domain_nr(hbus->bridge->domain_nr);
> free_bus:
> kfree(hbus);
> return ret;
> --
> 2.53.0
>
LGTM,
Reviewed-by: Saurabh Sengar <ssengar@linux.microsoft.com>
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] PCI: hv: Fix double ida_free in hv_pci_probe error path
2026-04-03 12:09 [PATCH] PCI: hv: Fix double ida_free in hv_pci_probe error path Sahil Chandna
2026-04-03 17:16 ` Saurabh Singh Sengar
@ 2026-04-04 2:42 ` Manivannan Sadhasivam
2026-04-04 5:19 ` Wei Liu
1 sibling, 1 reply; 4+ messages in thread
From: Manivannan Sadhasivam @ 2026-04-04 2:42 UTC (permalink / raw)
To: Sahil Chandna
Cc: kys, haiyangz, wei.liu, decui, longli, lpieralisi, kwilczynski,
robh, bhelgaas, dan.j.williams, mhklinux, linux-hyperv, linux-pci,
linux-kernel
On Fri, Apr 03, 2026 at 05:09:29AM -0700, Sahil Chandna wrote:
> If hv_pci_probe() fails after storing the domain number in
> hbus->bridge->domain_nr, there is a call to free this domain_nr via
> pci_bus_release_emul_domain_nr(), however, during cleanup, the bridge
> release callback pci_release_host_bridge_dev() also frees the domain_nr
> causing ida_free to be called on same ID twice and triggering following
> warning:
>
> ida_free called for id=28971 which is not allocated.
> WARNING: lib/idr.c:594 at ida_free+0xdf/0x160, CPU#0: kworker/0:2/198
> Call Trace:
> pci_bus_release_emul_domain_nr+0x17/0x20
> pci_release_host_bridge_dev+0x4b/0x60
> device_release+0x3b/0xa0
> kobject_put+0x8e/0x220
> devm_pci_alloc_host_bridge_release+0xe/0x20
> devres_release_all+0x9a/0xd0
> device_unbind_cleanup+0x12/0xa0
> really_probe+0x1c5/0x3f0
> vmbus_add_channel_work+0x135/0x1a0
>
> Fix this by letting pci core handle the free domain_nr and remove
> the explicit free called in pci-hyperv driver.
>
> Fixes: bcce8c74f1ce ("PCI: Enable host bridge emulation for PCI_DOMAINS_GENERIC platforms")
> Signed-off-by: Sahil Chandna <sahilchandna@linux.microsoft.com>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
Wei: I see that you've taken previous hyperv patches. So feel free to take this
one also.
We will be proactive from next release onwards ;)
- Mani
> ---
> drivers/pci/controller/pci-hyperv.c | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
>
> diff --git a/drivers/pci/controller/pci-hyperv.c b/drivers/pci/controller/pci-hyperv.c
> index 2c7a406b4ba8..5616ad5d2a8f 100644
> --- a/drivers/pci/controller/pci-hyperv.c
> +++ b/drivers/pci/controller/pci-hyperv.c
> @@ -3778,7 +3778,7 @@ static int hv_pci_probe(struct hv_device *hdev,
> hbus->bridge->domain_nr);
> if (!hbus->wq) {
> ret = -ENOMEM;
> - goto free_dom;
> + goto free_bus;
> }
>
> hdev->channel->next_request_id_callback = vmbus_next_request_id;
> @@ -3874,8 +3874,6 @@ static int hv_pci_probe(struct hv_device *hdev,
> vmbus_close(hdev->channel);
> destroy_wq:
> destroy_workqueue(hbus->wq);
> -free_dom:
> - pci_bus_release_emul_domain_nr(hbus->bridge->domain_nr);
> free_bus:
> kfree(hbus);
> return ret;
> --
> 2.53.0
>
--
மணிவண்ணன் சதாசிவம்
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] PCI: hv: Fix double ida_free in hv_pci_probe error path
2026-04-04 2:42 ` Manivannan Sadhasivam
@ 2026-04-04 5:19 ` Wei Liu
0 siblings, 0 replies; 4+ messages in thread
From: Wei Liu @ 2026-04-04 5:19 UTC (permalink / raw)
To: Manivannan Sadhasivam
Cc: Sahil Chandna, kys, haiyangz, wei.liu, decui, longli, lpieralisi,
kwilczynski, robh, bhelgaas, dan.j.williams, mhklinux,
linux-hyperv, linux-pci, linux-kernel
On Sat, Apr 04, 2026 at 08:12:36AM +0530, Manivannan Sadhasivam wrote:
> On Fri, Apr 03, 2026 at 05:09:29AM -0700, Sahil Chandna wrote:
> > If hv_pci_probe() fails after storing the domain number in
> > hbus->bridge->domain_nr, there is a call to free this domain_nr via
> > pci_bus_release_emul_domain_nr(), however, during cleanup, the bridge
> > release callback pci_release_host_bridge_dev() also frees the domain_nr
> > causing ida_free to be called on same ID twice and triggering following
> > warning:
> >
> > ida_free called for id=28971 which is not allocated.
> > WARNING: lib/idr.c:594 at ida_free+0xdf/0x160, CPU#0: kworker/0:2/198
> > Call Trace:
> > pci_bus_release_emul_domain_nr+0x17/0x20
> > pci_release_host_bridge_dev+0x4b/0x60
> > device_release+0x3b/0xa0
> > kobject_put+0x8e/0x220
> > devm_pci_alloc_host_bridge_release+0xe/0x20
> > devres_release_all+0x9a/0xd0
> > device_unbind_cleanup+0x12/0xa0
> > really_probe+0x1c5/0x3f0
> > vmbus_add_channel_work+0x135/0x1a0
> >
> > Fix this by letting pci core handle the free domain_nr and remove
> > the explicit free called in pci-hyperv driver.
> >
> > Fixes: bcce8c74f1ce ("PCI: Enable host bridge emulation for PCI_DOMAINS_GENERIC platforms")
> > Signed-off-by: Sahil Chandna <sahilchandna@linux.microsoft.com>
>
> Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
>
> Wei: I see that you've taken previous hyperv patches. So feel free to take this
> one also.
Sure. I will pick this up.
Wei
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-04-04 5:19 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-04-03 12:09 [PATCH] PCI: hv: Fix double ida_free in hv_pci_probe error path Sahil Chandna
2026-04-03 17:16 ` Saurabh Singh Sengar
2026-04-04 2:42 ` Manivannan Sadhasivam
2026-04-04 5:19 ` Wei Liu
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox