* Re: [PATCH 2/2] PCI: Fix pci_host_bridge struct device release/free handling
2020-05-13 22:38 ` [PATCH 2/2] PCI: Fix pci_host_bridge struct device release/free handling Rob Herring
@ 2020-05-14 10:27 ` Anders Roxell
2020-05-14 10:30 ` Lorenzo Pieralisi
` (2 subsequent siblings)
3 siblings, 0 replies; 10+ messages in thread
From: Anders Roxell @ 2020-05-14 10:27 UTC (permalink / raw)
To: Rob Herring; +Cc: Bjorn Helgaas, linux-pci, Lorenzo Pieralisi, Arnd Bergmann
On Thu, 14 May 2020 at 00:39, Rob Herring <robh@kernel.org> wrote:
>
> The PCI code has several paths where the struct pci_host_bridge is freed
> directly. This is wrong because it contains a struct device which is
> refcounted and should be freed using put_device(). This can result in
> use-after-free errors. I think this problem has existed since 2012 with
> commit 7b5436635800 ("PCI: add generic device into pci_host_bridge
> struct"). It generally hasn't mattered as most host bridge drivers are
> still built-in and can't unbind.
>
> The problem is a struct device should never be freed directly once
> device_initialize() is called and a ref is held, but that doesn't happen
> until pci_register_host_bridge(). There's then a window between
> allocating the host bridge and pci_register_host_bridge() where kfree
> should be used. This is fragile and requires callers to do the right
> thing. To fix this, we need to split device_register() into
> device_initialize() and device_add() calls, so that the host bridge
> struct is always freed by using a put_device().
>
> devm_pci_alloc_host_bridge() is using devm_kzalloc() to allocate struct
> pci_host_bridge which will be freed directly. Instead, we can use a
> custom devres action to call put_device().
>
> Reported-by: Anders Roxell <anders.roxell@linaro.org>
> Cc: Arnd Bergmann <arnd@arndb.de>
> Cc: Bjorn Helgaas <bhelgaas@google.com>
> Signed-off-by: Rob Herring <robh@kernel.org>
Thank you Rob for fixing this so quickly!
I applied both patches and I couldn't see the use-after-free anymore.
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Cheers,
Anders
> ---
> drivers/pci/probe.c | 36 +++++++++++++++++++-----------------
> drivers/pci/remove.c | 2 +-
> 2 files changed, 20 insertions(+), 18 deletions(-)
>
> diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
> index e21dc71b1907..e064ded6fbec 100644
> --- a/drivers/pci/probe.c
> +++ b/drivers/pci/probe.c
> @@ -565,7 +565,7 @@ static struct pci_bus *pci_alloc_bus(struct pci_bus *parent)
> return b;
> }
>
> -static void devm_pci_release_host_bridge_dev(struct device *dev)
> +static void pci_release_host_bridge_dev(struct device *dev)
> {
> struct pci_host_bridge *bridge = to_pci_host_bridge(dev);
>
> @@ -574,12 +574,7 @@ static void devm_pci_release_host_bridge_dev(struct device *dev)
>
> pci_free_resource_list(&bridge->windows);
> pci_free_resource_list(&bridge->dma_ranges);
> -}
> -
> -static void pci_release_host_bridge_dev(struct device *dev)
> -{
> - devm_pci_release_host_bridge_dev(dev);
> - kfree(to_pci_host_bridge(dev));
> + kfree(bridge);
> }
>
> static void pci_init_host_bridge(struct pci_host_bridge *bridge)
> @@ -599,6 +594,8 @@ static void pci_init_host_bridge(struct pci_host_bridge *bridge)
> bridge->native_pme = 1;
> bridge->native_ltr = 1;
> bridge->native_dpc = 1;
> +
> + device_initialize(&bridge->dev);
> }
>
> struct pci_host_bridge *pci_alloc_host_bridge(size_t priv)
> @@ -616,17 +613,25 @@ struct pci_host_bridge *pci_alloc_host_bridge(size_t priv)
> }
> EXPORT_SYMBOL(pci_alloc_host_bridge);
>
> +static void devm_pci_alloc_host_bridge_release(void *data)
> +{
> + pci_free_host_bridge(data);
> +}
> +
> struct pci_host_bridge *devm_pci_alloc_host_bridge(struct device *dev,
> size_t priv)
> {
> + int ret;
> struct pci_host_bridge *bridge;
>
> - bridge = devm_kzalloc(dev, sizeof(*bridge) + priv, GFP_KERNEL);
> + bridge = pci_alloc_host_bridge(priv);
> if (!bridge)
> return NULL;
>
> - pci_init_host_bridge(bridge);
> - bridge->dev.release = devm_pci_release_host_bridge_dev;
> + ret = devm_add_action_or_reset(dev, devm_pci_alloc_host_bridge_release,
> + bridge);
> + if (ret)
> + return NULL;
>
> return bridge;
> }
> @@ -634,10 +639,7 @@ EXPORT_SYMBOL(devm_pci_alloc_host_bridge);
>
> void pci_free_host_bridge(struct pci_host_bridge *bridge)
> {
> - pci_free_resource_list(&bridge->windows);
> - pci_free_resource_list(&bridge->dma_ranges);
> -
> - kfree(bridge);
> + put_device(&bridge->dev);
> }
> EXPORT_SYMBOL(pci_free_host_bridge);
>
> @@ -908,7 +910,7 @@ static int pci_register_host_bridge(struct pci_host_bridge *bridge)
> if (err)
> goto free;
>
> - err = device_register(&bridge->dev);
> + err = device_add(&bridge->dev);
> if (err) {
> put_device(&bridge->dev);
> goto free;
> @@ -978,7 +980,7 @@ static int pci_register_host_bridge(struct pci_host_bridge *bridge)
>
> unregister:
> put_device(&bridge->dev);
> - device_unregister(&bridge->dev);
> + device_del(&bridge->dev);
>
> free:
> kfree(bus);
> @@ -2953,7 +2955,7 @@ struct pci_bus *pci_create_root_bus(struct device *parent, int bus,
> return bridge->bus;
>
> err_out:
> - kfree(bridge);
> + put_device(&bridge->dev);
> return NULL;
> }
> EXPORT_SYMBOL_GPL(pci_create_root_bus);
> diff --git a/drivers/pci/remove.c b/drivers/pci/remove.c
> index e9c6b120cf45..95dec03d9f2a 100644
> --- a/drivers/pci/remove.c
> +++ b/drivers/pci/remove.c
> @@ -160,6 +160,6 @@ void pci_remove_root_bus(struct pci_bus *bus)
> host_bridge->bus = NULL;
>
> /* remove the host bridge */
> - device_unregister(&host_bridge->dev);
> + device_del(&host_bridge->dev);
> }
> EXPORT_SYMBOL_GPL(pci_remove_root_bus);
> --
> 2.20.1
>
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 2/2] PCI: Fix pci_host_bridge struct device release/free handling
2020-05-13 22:38 ` [PATCH 2/2] PCI: Fix pci_host_bridge struct device release/free handling Rob Herring
2020-05-14 10:27 ` Anders Roxell
@ 2020-05-14 10:30 ` Lorenzo Pieralisi
2020-05-14 12:50 ` Rob Herring
2020-05-14 16:43 ` Lorenzo Pieralisi
2020-05-14 21:22 ` Arnd Bergmann
3 siblings, 1 reply; 10+ messages in thread
From: Lorenzo Pieralisi @ 2020-05-14 10:30 UTC (permalink / raw)
To: Rob Herring; +Cc: Bjorn Helgaas, linux-pci, Anders Roxell, Arnd Bergmann
On Wed, May 13, 2020 at 05:38:59PM -0500, Rob Herring wrote:
> The PCI code has several paths where the struct pci_host_bridge is freed
> directly. This is wrong because it contains a struct device which is
> refcounted and should be freed using put_device(). This can result in
> use-after-free errors. I think this problem has existed since 2012 with
> commit 7b5436635800 ("PCI: add generic device into pci_host_bridge
> struct"). It generally hasn't mattered as most host bridge drivers are
> still built-in and can't unbind.
>
> The problem is a struct device should never be freed directly once
> device_initialize() is called and a ref is held, but that doesn't happen
> until pci_register_host_bridge(). There's then a window between
> allocating the host bridge and pci_register_host_bridge() where kfree
> should be used. This is fragile and requires callers to do the right
> thing. To fix this, we need to split device_register() into
> device_initialize() and device_add() calls, so that the host bridge
> struct is always freed by using a put_device().
>
> devm_pci_alloc_host_bridge() is using devm_kzalloc() to allocate struct
> pci_host_bridge which will be freed directly. Instead, we can use a
> custom devres action to call put_device().
>
> Reported-by: Anders Roxell <anders.roxell@linaro.org>
> Cc: Arnd Bergmann <arnd@arndb.de>
> Cc: Bjorn Helgaas <bhelgaas@google.com>
> Signed-off-by: Rob Herring <robh@kernel.org>
> ---
> drivers/pci/probe.c | 36 +++++++++++++++++++-----------------
> drivers/pci/remove.c | 2 +-
> 2 files changed, 20 insertions(+), 18 deletions(-)
>
> diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
> index e21dc71b1907..e064ded6fbec 100644
> --- a/drivers/pci/probe.c
> +++ b/drivers/pci/probe.c
> @@ -565,7 +565,7 @@ static struct pci_bus *pci_alloc_bus(struct pci_bus *parent)
> return b;
> }
>
> -static void devm_pci_release_host_bridge_dev(struct device *dev)
> +static void pci_release_host_bridge_dev(struct device *dev)
> {
> struct pci_host_bridge *bridge = to_pci_host_bridge(dev);
>
> @@ -574,12 +574,7 @@ static void devm_pci_release_host_bridge_dev(struct device *dev)
>
> pci_free_resource_list(&bridge->windows);
> pci_free_resource_list(&bridge->dma_ranges);
> -}
> -
> -static void pci_release_host_bridge_dev(struct device *dev)
> -{
> - devm_pci_release_host_bridge_dev(dev);
> - kfree(to_pci_host_bridge(dev));
> + kfree(bridge);
> }
>
> static void pci_init_host_bridge(struct pci_host_bridge *bridge)
> @@ -599,6 +594,8 @@ static void pci_init_host_bridge(struct pci_host_bridge *bridge)
> bridge->native_pme = 1;
> bridge->native_ltr = 1;
> bridge->native_dpc = 1;
> +
> + device_initialize(&bridge->dev);
> }
>
> struct pci_host_bridge *pci_alloc_host_bridge(size_t priv)
> @@ -616,17 +613,25 @@ struct pci_host_bridge *pci_alloc_host_bridge(size_t priv)
> }
> EXPORT_SYMBOL(pci_alloc_host_bridge);
>
> +static void devm_pci_alloc_host_bridge_release(void *data)
> +{
> + pci_free_host_bridge(data);
> +}
> +
> struct pci_host_bridge *devm_pci_alloc_host_bridge(struct device *dev,
> size_t priv)
> {
> + int ret;
> struct pci_host_bridge *bridge;
>
> - bridge = devm_kzalloc(dev, sizeof(*bridge) + priv, GFP_KERNEL);
> + bridge = pci_alloc_host_bridge(priv);
> if (!bridge)
> return NULL;
>
> - pci_init_host_bridge(bridge);
> - bridge->dev.release = devm_pci_release_host_bridge_dev;
> + ret = devm_add_action_or_reset(dev, devm_pci_alloc_host_bridge_release,
> + bridge);
> + if (ret)
> + return NULL;
>
> return bridge;
> }
> @@ -634,10 +639,7 @@ EXPORT_SYMBOL(devm_pci_alloc_host_bridge);
>
> void pci_free_host_bridge(struct pci_host_bridge *bridge)
> {
> - pci_free_resource_list(&bridge->windows);
> - pci_free_resource_list(&bridge->dma_ranges);
> -
> - kfree(bridge);
> + put_device(&bridge->dev);
> }
> EXPORT_SYMBOL(pci_free_host_bridge);
>
> @@ -908,7 +910,7 @@ static int pci_register_host_bridge(struct pci_host_bridge *bridge)
> if (err)
> goto free;
>
> - err = device_register(&bridge->dev);
> + err = device_add(&bridge->dev);
> if (err) {
> put_device(&bridge->dev);
> goto free;
> @@ -978,7 +980,7 @@ static int pci_register_host_bridge(struct pci_host_bridge *bridge)
>
> unregister:
> put_device(&bridge->dev);
> - device_unregister(&bridge->dev);
> + device_del(&bridge->dev);
I think we need to execute device_del() first, then put_device().
Thank you for fixing this code path.
Lorenzo
> free:
> kfree(bus);
> @@ -2953,7 +2955,7 @@ struct pci_bus *pci_create_root_bus(struct device *parent, int bus,
> return bridge->bus;
>
> err_out:
> - kfree(bridge);
> + put_device(&bridge->dev);
> return NULL;
> }
> EXPORT_SYMBOL_GPL(pci_create_root_bus);
> diff --git a/drivers/pci/remove.c b/drivers/pci/remove.c
> index e9c6b120cf45..95dec03d9f2a 100644
> --- a/drivers/pci/remove.c
> +++ b/drivers/pci/remove.c
> @@ -160,6 +160,6 @@ void pci_remove_root_bus(struct pci_bus *bus)
> host_bridge->bus = NULL;
>
> /* remove the host bridge */
> - device_unregister(&host_bridge->dev);
> + device_del(&host_bridge->dev);
> }
> EXPORT_SYMBOL_GPL(pci_remove_root_bus);
> --
> 2.20.1
>
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 2/2] PCI: Fix pci_host_bridge struct device release/free handling
2020-05-14 10:30 ` Lorenzo Pieralisi
@ 2020-05-14 12:50 ` Rob Herring
0 siblings, 0 replies; 10+ messages in thread
From: Rob Herring @ 2020-05-14 12:50 UTC (permalink / raw)
To: Lorenzo Pieralisi; +Cc: Bjorn Helgaas, PCI, Anders Roxell, Arnd Bergmann
On Thu, May 14, 2020 at 5:30 AM Lorenzo Pieralisi
<lorenzo.pieralisi@arm.com> wrote:
>
> On Wed, May 13, 2020 at 05:38:59PM -0500, Rob Herring wrote:
> > The PCI code has several paths where the struct pci_host_bridge is freed
> > directly. This is wrong because it contains a struct device which is
> > refcounted and should be freed using put_device(). This can result in
> > use-after-free errors. I think this problem has existed since 2012 with
> > commit 7b5436635800 ("PCI: add generic device into pci_host_bridge
> > struct"). It generally hasn't mattered as most host bridge drivers are
> > still built-in and can't unbind.
> >
> > The problem is a struct device should never be freed directly once
> > device_initialize() is called and a ref is held, but that doesn't happen
> > until pci_register_host_bridge(). There's then a window between
> > allocating the host bridge and pci_register_host_bridge() where kfree
> > should be used. This is fragile and requires callers to do the right
> > thing. To fix this, we need to split device_register() into
> > device_initialize() and device_add() calls, so that the host bridge
> > struct is always freed by using a put_device().
> >
> > devm_pci_alloc_host_bridge() is using devm_kzalloc() to allocate struct
> > pci_host_bridge which will be freed directly. Instead, we can use a
> > custom devres action to call put_device().
> >
> > Reported-by: Anders Roxell <anders.roxell@linaro.org>
> > Cc: Arnd Bergmann <arnd@arndb.de>
> > Cc: Bjorn Helgaas <bhelgaas@google.com>
> > Signed-off-by: Rob Herring <robh@kernel.org>
> > ---
> > drivers/pci/probe.c | 36 +++++++++++++++++++-----------------
> > drivers/pci/remove.c | 2 +-
> > 2 files changed, 20 insertions(+), 18 deletions(-)
[...]
> > @@ -908,7 +910,7 @@ static int pci_register_host_bridge(struct pci_host_bridge *bridge)
> > if (err)
> > goto free;
> >
> > - err = device_register(&bridge->dev);
> > + err = device_add(&bridge->dev);
> > if (err) {
> > put_device(&bridge->dev);
> > goto free;
> > @@ -978,7 +980,7 @@ static int pci_register_host_bridge(struct pci_host_bridge *bridge)
> >
> > unregister:
> > put_device(&bridge->dev);
> > - device_unregister(&bridge->dev);
> > + device_del(&bridge->dev);
>
> I think we need to execute device_del() first, then put_device().
No, because after the device_add, there's a get_device() adding the
bridge ptr to the bus. So the put_device here is for that. The final
put_device() is in the bridge free or devm action.
Rob
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 2/2] PCI: Fix pci_host_bridge struct device release/free handling
2020-05-13 22:38 ` [PATCH 2/2] PCI: Fix pci_host_bridge struct device release/free handling Rob Herring
2020-05-14 10:27 ` Anders Roxell
2020-05-14 10:30 ` Lorenzo Pieralisi
@ 2020-05-14 16:43 ` Lorenzo Pieralisi
2020-05-14 21:22 ` Arnd Bergmann
3 siblings, 0 replies; 10+ messages in thread
From: Lorenzo Pieralisi @ 2020-05-14 16:43 UTC (permalink / raw)
To: Rob Herring; +Cc: Bjorn Helgaas, linux-pci, Anders Roxell, Arnd Bergmann
On Wed, May 13, 2020 at 05:38:59PM -0500, Rob Herring wrote:
> The PCI code has several paths where the struct pci_host_bridge is freed
> directly. This is wrong because it contains a struct device which is
> refcounted and should be freed using put_device(). This can result in
> use-after-free errors. I think this problem has existed since 2012 with
> commit 7b5436635800 ("PCI: add generic device into pci_host_bridge
> struct"). It generally hasn't mattered as most host bridge drivers are
> still built-in and can't unbind.
>
> The problem is a struct device should never be freed directly once
> device_initialize() is called and a ref is held, but that doesn't happen
> until pci_register_host_bridge(). There's then a window between
> allocating the host bridge and pci_register_host_bridge() where kfree
> should be used. This is fragile and requires callers to do the right
> thing. To fix this, we need to split device_register() into
> device_initialize() and device_add() calls, so that the host bridge
> struct is always freed by using a put_device().
>
> devm_pci_alloc_host_bridge() is using devm_kzalloc() to allocate struct
> pci_host_bridge which will be freed directly. Instead, we can use a
> custom devres action to call put_device().
>
> Reported-by: Anders Roxell <anders.roxell@linaro.org>
> Cc: Arnd Bergmann <arnd@arndb.de>
> Cc: Bjorn Helgaas <bhelgaas@google.com>
> Signed-off-by: Rob Herring <robh@kernel.org>
> ---
> drivers/pci/probe.c | 36 +++++++++++++++++++-----------------
> drivers/pci/remove.c | 2 +-
> 2 files changed, 20 insertions(+), 18 deletions(-)
Reviewed-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
> diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
> index e21dc71b1907..e064ded6fbec 100644
> --- a/drivers/pci/probe.c
> +++ b/drivers/pci/probe.c
> @@ -565,7 +565,7 @@ static struct pci_bus *pci_alloc_bus(struct pci_bus *parent)
> return b;
> }
>
> -static void devm_pci_release_host_bridge_dev(struct device *dev)
> +static void pci_release_host_bridge_dev(struct device *dev)
> {
> struct pci_host_bridge *bridge = to_pci_host_bridge(dev);
>
> @@ -574,12 +574,7 @@ static void devm_pci_release_host_bridge_dev(struct device *dev)
>
> pci_free_resource_list(&bridge->windows);
> pci_free_resource_list(&bridge->dma_ranges);
> -}
> -
> -static void pci_release_host_bridge_dev(struct device *dev)
> -{
> - devm_pci_release_host_bridge_dev(dev);
> - kfree(to_pci_host_bridge(dev));
> + kfree(bridge);
> }
>
> static void pci_init_host_bridge(struct pci_host_bridge *bridge)
> @@ -599,6 +594,8 @@ static void pci_init_host_bridge(struct pci_host_bridge *bridge)
> bridge->native_pme = 1;
> bridge->native_ltr = 1;
> bridge->native_dpc = 1;
> +
> + device_initialize(&bridge->dev);
> }
>
> struct pci_host_bridge *pci_alloc_host_bridge(size_t priv)
> @@ -616,17 +613,25 @@ struct pci_host_bridge *pci_alloc_host_bridge(size_t priv)
> }
> EXPORT_SYMBOL(pci_alloc_host_bridge);
>
> +static void devm_pci_alloc_host_bridge_release(void *data)
> +{
> + pci_free_host_bridge(data);
> +}
> +
> struct pci_host_bridge *devm_pci_alloc_host_bridge(struct device *dev,
> size_t priv)
> {
> + int ret;
> struct pci_host_bridge *bridge;
>
> - bridge = devm_kzalloc(dev, sizeof(*bridge) + priv, GFP_KERNEL);
> + bridge = pci_alloc_host_bridge(priv);
> if (!bridge)
> return NULL;
>
> - pci_init_host_bridge(bridge);
> - bridge->dev.release = devm_pci_release_host_bridge_dev;
> + ret = devm_add_action_or_reset(dev, devm_pci_alloc_host_bridge_release,
> + bridge);
> + if (ret)
> + return NULL;
>
> return bridge;
> }
> @@ -634,10 +639,7 @@ EXPORT_SYMBOL(devm_pci_alloc_host_bridge);
>
> void pci_free_host_bridge(struct pci_host_bridge *bridge)
> {
> - pci_free_resource_list(&bridge->windows);
> - pci_free_resource_list(&bridge->dma_ranges);
> -
> - kfree(bridge);
> + put_device(&bridge->dev);
> }
> EXPORT_SYMBOL(pci_free_host_bridge);
>
> @@ -908,7 +910,7 @@ static int pci_register_host_bridge(struct pci_host_bridge *bridge)
> if (err)
> goto free;
>
> - err = device_register(&bridge->dev);
> + err = device_add(&bridge->dev);
> if (err) {
> put_device(&bridge->dev);
> goto free;
> @@ -978,7 +980,7 @@ static int pci_register_host_bridge(struct pci_host_bridge *bridge)
>
> unregister:
> put_device(&bridge->dev);
> - device_unregister(&bridge->dev);
> + device_del(&bridge->dev);
>
> free:
> kfree(bus);
> @@ -2953,7 +2955,7 @@ struct pci_bus *pci_create_root_bus(struct device *parent, int bus,
> return bridge->bus;
>
> err_out:
> - kfree(bridge);
> + put_device(&bridge->dev);
> return NULL;
> }
> EXPORT_SYMBOL_GPL(pci_create_root_bus);
> diff --git a/drivers/pci/remove.c b/drivers/pci/remove.c
> index e9c6b120cf45..95dec03d9f2a 100644
> --- a/drivers/pci/remove.c
> +++ b/drivers/pci/remove.c
> @@ -160,6 +160,6 @@ void pci_remove_root_bus(struct pci_bus *bus)
> host_bridge->bus = NULL;
>
> /* remove the host bridge */
> - device_unregister(&host_bridge->dev);
> + device_del(&host_bridge->dev);
> }
> EXPORT_SYMBOL_GPL(pci_remove_root_bus);
> --
> 2.20.1
>
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 2/2] PCI: Fix pci_host_bridge struct device release/free handling
2020-05-13 22:38 ` [PATCH 2/2] PCI: Fix pci_host_bridge struct device release/free handling Rob Herring
` (2 preceding siblings ...)
2020-05-14 16:43 ` Lorenzo Pieralisi
@ 2020-05-14 21:22 ` Arnd Bergmann
3 siblings, 0 replies; 10+ messages in thread
From: Arnd Bergmann @ 2020-05-14 21:22 UTC (permalink / raw)
To: Rob Herring; +Cc: Bjorn Helgaas, linux-pci, Lorenzo Pieralisi, Anders Roxell
On Thu, May 14, 2020 at 12:39 AM Rob Herring <robh@kernel.org> wrote:
>
> The PCI code has several paths where the struct pci_host_bridge is freed
> directly. This is wrong because it contains a struct device which is
> refcounted and should be freed using put_device(). This can result in
> use-after-free errors. I think this problem has existed since 2012 with
> commit 7b5436635800 ("PCI: add generic device into pci_host_bridge
> struct"). It generally hasn't mattered as most host bridge drivers are
> still built-in and can't unbind.
>
> The problem is a struct device should never be freed directly once
> device_initialize() is called and a ref is held, but that doesn't happen
> until pci_register_host_bridge(). There's then a window between
> allocating the host bridge and pci_register_host_bridge() where kfree
> should be used. This is fragile and requires callers to do the right
> thing. To fix this, we need to split device_register() into
> device_initialize() and device_add() calls, so that the host bridge
> struct is always freed by using a put_device().
>
> devm_pci_alloc_host_bridge() is using devm_kzalloc() to allocate struct
> pci_host_bridge which will be freed directly. Instead, we can use a
> custom devres action to call put_device().
>
> Reported-by: Anders Roxell <anders.roxell@linaro.org>
> Cc: Arnd Bergmann <arnd@arndb.de>
> Cc: Bjorn Helgaas <bhelgaas@google.com>
> Signed-off-by: Rob Herring <robh@kernel.org>
Thanks for working your way through that bit of code
and fixing my mistakes!
Your description makes a lot of sense and the code looks
reasonable, but I don't understand my own work enough any
more to be sure I didn't miss anything.
Acked-by: Arnd Bergmann <arnd@arndb.de>
^ permalink raw reply [flat|nested] 10+ messages in thread