Message ID | f93e6b2393301df6ac960ef6891b1b2812da67f3.1731005223.git.leonro@nvidia.com |
---|---|
State | New |
Headers | show |
Series | Fix read permissions for VPD attributes | expand |
On Thu, Nov 07, 2024 at 08:56:56PM +0200, Leon Romanovsky wrote: > From: Leon Romanovsky <leonro@nvidia.com> > > The Vital Product Data (VPD) attribute is not readable by regular > user without root permissions. Such restriction is not really needed > for many devices in the world, as data presented in that VPD is not > sensitive and access to the HW is safe and tested. > > This change aligns the permissions of the VPD attribute to be accessible > for read by all users, while write being restricted to root only. > > For the driver, there is a need to opt-in in order to allow this > functionality. I don't think the use case is very strong (and not included at all here). If we do need to do this, I think it's a property of the device, not the driver. > Signed-off-by: Leon Romanovsky <leonro@nvidia.com> > --- > drivers/pci/vpd.c | 9 ++++++++- > include/linux/pci.h | 7 ++++++- > 2 files changed, 14 insertions(+), 2 deletions(-) > > diff --git a/drivers/pci/vpd.c b/drivers/pci/vpd.c > index e4300f5f304f..7c70930abaa0 100644 > --- a/drivers/pci/vpd.c > +++ b/drivers/pci/vpd.c > @@ -156,6 +156,7 @@ static ssize_t pci_vpd_read(struct pci_dev *dev, loff_t pos, size_t count, > void *arg, bool check_size) > { > struct pci_vpd *vpd = &dev->vpd; > + struct pci_driver *drv; > unsigned int max_len; > int ret = 0; > loff_t end = pos + count; > @@ -167,6 +168,12 @@ static ssize_t pci_vpd_read(struct pci_dev *dev, loff_t pos, size_t count, > if (pos < 0) > return -EINVAL; > > + if (!capable(CAP_SYS_ADMIN)) { > + drv = to_pci_driver(dev->dev.driver); > + if (!drv || !drv->downgrade_vpd_read) > + return -EPERM; > + } > + > max_len = check_size ? vpd->len : PCI_VPD_MAX_SIZE; > > if (pos >= max_len) > @@ -317,7 +324,7 @@ static ssize_t vpd_write(struct file *filp, struct kobject *kobj, > > return ret; > } > -static BIN_ATTR(vpd, 0600, vpd_read, vpd_write, 0); > +static BIN_ATTR_RW(vpd, 0); > > static struct bin_attribute *vpd_attrs[] = { > &bin_attr_vpd, > diff --git a/include/linux/pci.h b/include/linux/pci.h > index 573b4c4c2be6..b8fed74e742e 100644 > --- a/include/linux/pci.h > +++ b/include/linux/pci.h > @@ -943,6 +943,10 @@ struct module; > * how to manage the DMA themselves and set this flag so that > * the IOMMU layer will allow them to setup and manage their > * own I/O address space. > + * @downgrade_vpd_read: Device doesn't require root permissions from the users > + * to read VPD information. The driver doesn't expose any sensitive > + * information through that interface and safe to be accessed by > + * unprivileged users. > */ > struct pci_driver { > const char *name; > @@ -960,7 +964,8 @@ struct pci_driver { > const struct attribute_group **dev_groups; > struct device_driver driver; > struct pci_dynids dynids; > - bool driver_managed_dma; > + bool driver_managed_dma : 1; > + bool downgrade_vpd_read : 1; > }; > > #define to_pci_driver(__drv) \ > -- > 2.47.0 >
On 2024-11-07 20:56:56+0200, Leon Romanovsky wrote: > From: Leon Romanovsky <leonro@nvidia.com> > > The Vital Product Data (VPD) attribute is not readable by regular > user without root permissions. Such restriction is not really needed > for many devices in the world, as data presented in that VPD is not > sensitive and access to the HW is safe and tested. > > This change aligns the permissions of the VPD attribute to be accessible > for read by all users, while write being restricted to root only. > > For the driver, there is a need to opt-in in order to allow this > functionality. > > Signed-off-by: Leon Romanovsky <leonro@nvidia.com> > --- > drivers/pci/vpd.c | 9 ++++++++- > include/linux/pci.h | 7 ++++++- > 2 files changed, 14 insertions(+), 2 deletions(-) > > diff --git a/drivers/pci/vpd.c b/drivers/pci/vpd.c > index e4300f5f304f..7c70930abaa0 100644 > --- a/drivers/pci/vpd.c > +++ b/drivers/pci/vpd.c > @@ -156,6 +156,7 @@ static ssize_t pci_vpd_read(struct pci_dev *dev, loff_t pos, size_t count, > void *arg, bool check_size) > { > struct pci_vpd *vpd = &dev->vpd; > + struct pci_driver *drv; > unsigned int max_len; > int ret = 0; > loff_t end = pos + count; > @@ -167,6 +168,12 @@ static ssize_t pci_vpd_read(struct pci_dev *dev, loff_t pos, size_t count, > if (pos < 0) > return -EINVAL; > > + if (!capable(CAP_SYS_ADMIN)) { > + drv = to_pci_driver(dev->dev.driver); > + if (!drv || !drv->downgrade_vpd_read) > + return -EPERM; > + } If you move the check into vpd_attr_is_visible() then the sysfs core will enforce the permissions and it's obvious for the user if they can or can't read/write the file. > + > max_len = check_size ? vpd->len : PCI_VPD_MAX_SIZE; > > if (pos >= max_len) > @@ -317,7 +324,7 @@ static ssize_t vpd_write(struct file *filp, struct kobject *kobj, > > return ret; > } > -static BIN_ATTR(vpd, 0600, vpd_read, vpd_write, 0); > +static BIN_ATTR_RW(vpd, 0); > > static struct bin_attribute *vpd_attrs[] = { > &bin_attr_vpd, > diff --git a/include/linux/pci.h b/include/linux/pci.h > index 573b4c4c2be6..b8fed74e742e 100644 > --- a/include/linux/pci.h > +++ b/include/linux/pci.h > @@ -943,6 +943,10 @@ struct module; > * how to manage the DMA themselves and set this flag so that > * the IOMMU layer will allow them to setup and manage their > * own I/O address space. > + * @downgrade_vpd_read: Device doesn't require root permissions from the users > + * to read VPD information. The driver doesn't expose any sensitive > + * information through that interface and safe to be accessed by > + * unprivileged users. > */ > struct pci_driver { > const char *name; > @@ -960,7 +964,8 @@ struct pci_driver { > const struct attribute_group **dev_groups; > struct device_driver driver; > struct pci_dynids dynids; > - bool driver_managed_dma; > + bool driver_managed_dma : 1; > + bool downgrade_vpd_read : 1; > }; > > #define to_pci_driver(__drv) \ > -- > 2.47.0 >
On Mon, Nov 11, 2024 at 02:41:04PM -0600, Bjorn Helgaas wrote: > On Thu, Nov 07, 2024 at 08:56:56PM +0200, Leon Romanovsky wrote: > > From: Leon Romanovsky <leonro@nvidia.com> > > > > The Vital Product Data (VPD) attribute is not readable by regular > > user without root permissions. Such restriction is not really needed > > for many devices in the world, as data presented in that VPD is not > > sensitive and access to the HW is safe and tested. > > > > This change aligns the permissions of the VPD attribute to be accessible > > for read by all users, while write being restricted to root only. > > > > For the driver, there is a need to opt-in in order to allow this > > functionality. > > I don't think the use case is very strong (and not included at all > here). I will add the use case, which is running monitoring application without need to be root. IMHO reducing number of applications that require privileged access is a very strong case. I personally try to avoid applications with root/setuid privileges. > > If we do need to do this, I think it's a property of the device, not > the driver. But how will device inform PCI core about safe VPD read? Should I add new field to struct pci_device_id? Add a quirk? Otherwise, I will need to add a line "pci_dev->downgrade_vpd_read=true" to mlx5 probe function and it won't change a lot from current implementation. > > > Signed-off-by: Leon Romanovsky <leonro@nvidia.com> > > --- > > drivers/pci/vpd.c | 9 ++++++++- > > include/linux/pci.h | 7 ++++++- > > 2 files changed, 14 insertions(+), 2 deletions(-) > > > > diff --git a/drivers/pci/vpd.c b/drivers/pci/vpd.c > > index e4300f5f304f..7c70930abaa0 100644 > > --- a/drivers/pci/vpd.c > > +++ b/drivers/pci/vpd.c > > @@ -156,6 +156,7 @@ static ssize_t pci_vpd_read(struct pci_dev *dev, loff_t pos, size_t count, > > void *arg, bool check_size) > > { > > struct pci_vpd *vpd = &dev->vpd; > > + struct pci_driver *drv; > > unsigned int max_len; > > int ret = 0; > > loff_t end = pos + count; > > @@ -167,6 +168,12 @@ static ssize_t pci_vpd_read(struct pci_dev *dev, loff_t pos, size_t count, > > if (pos < 0) > > return -EINVAL; > > > > + if (!capable(CAP_SYS_ADMIN)) { > > + drv = to_pci_driver(dev->dev.driver); > > + if (!drv || !drv->downgrade_vpd_read) > > + return -EPERM; > > + } > > + > > max_len = check_size ? vpd->len : PCI_VPD_MAX_SIZE; > > > > if (pos >= max_len) > > @@ -317,7 +324,7 @@ static ssize_t vpd_write(struct file *filp, struct kobject *kobj, > > > > return ret; > > } > > -static BIN_ATTR(vpd, 0600, vpd_read, vpd_write, 0); > > +static BIN_ATTR_RW(vpd, 0); > > > > static struct bin_attribute *vpd_attrs[] = { > > &bin_attr_vpd, > > diff --git a/include/linux/pci.h b/include/linux/pci.h > > index 573b4c4c2be6..b8fed74e742e 100644 > > --- a/include/linux/pci.h > > +++ b/include/linux/pci.h > > @@ -943,6 +943,10 @@ struct module; > > * how to manage the DMA themselves and set this flag so that > > * the IOMMU layer will allow them to setup and manage their > > * own I/O address space. > > + * @downgrade_vpd_read: Device doesn't require root permissions from the users > > + * to read VPD information. The driver doesn't expose any sensitive > > + * information through that interface and safe to be accessed by > > + * unprivileged users. > > */ > > struct pci_driver { > > const char *name; > > @@ -960,7 +964,8 @@ struct pci_driver { > > const struct attribute_group **dev_groups; > > struct device_driver driver; > > struct pci_dynids dynids; > > - bool driver_managed_dma; > > + bool driver_managed_dma : 1; > > + bool downgrade_vpd_read : 1; > > }; > > > > #define to_pci_driver(__drv) \ > > -- > > 2.47.0 > >
[+cc Thomas] On Mon, Nov 11, 2024 at 11:17:38PM +0200, Leon Romanovsky wrote: > On Mon, Nov 11, 2024 at 02:41:04PM -0600, Bjorn Helgaas wrote: > > On Thu, Nov 07, 2024 at 08:56:56PM +0200, Leon Romanovsky wrote: > > > From: Leon Romanovsky <leonro@nvidia.com> > > > > > > The Vital Product Data (VPD) attribute is not readable by regular > > > user without root permissions. Such restriction is not really needed > > > for many devices in the world, as data presented in that VPD is not > > > sensitive and access to the HW is safe and tested. > > > > > > This change aligns the permissions of the VPD attribute to be accessible > > > for read by all users, while write being restricted to root only. > > > > > > For the driver, there is a need to opt-in in order to allow this > > > functionality. > > > > I don't think the use case is very strong (and not included at all > > here). > > I will add the use case, which is running monitoring application without > need to be root. IMHO reducing number of applications that require > privileged access is a very strong case. I personally try to avoid > applications with root/setuid privileges. Avoiding root/setuid is a very good thing. I just don't think using VPD directly from userspace is a great idea because VPD is so slow and sometimes unreliable to read. And apparently this is a pretty unusual situation since I haven't heard similar requests for other devices. Sort of ironic that some vendors, especially Intel and AMD, add new Device IDs for devices that work exactly the same as their predecessors, so we are continually adding to the pci_device_id tables, while here we apparently the same Device ID is used for devices that differ in ways we actually want to know about. > > If we do need to do this, I think it's a property of the device, not > > the driver. > > But how will device inform PCI core about safe VPD read? > Should I add new field to struct pci_device_id? Add a quirk? > Otherwise, I will need to add a line "pci_dev->downgrade_vpd_read=true" > to mlx5 probe function and it won't change a lot from current > implementation. To me it looks like a pci_dev bit, not a pci_driver bit. I would set it .probe() so all the code is in one place. And it's not related to a device defect, as most quirks are. Bjorn
On Mon, 11 Nov 2024 14:41:04 -0600 Bjorn Helgaas <helgaas@kernel.org> wrote: > On Thu, Nov 07, 2024 at 08:56:56PM +0200, Leon Romanovsky wrote: > > From: Leon Romanovsky <leonro@nvidia.com> > > > > The Vital Product Data (VPD) attribute is not readable by regular > > user without root permissions. Such restriction is not really needed > > for many devices in the world, as data presented in that VPD is not > > sensitive and access to the HW is safe and tested. > > > > This change aligns the permissions of the VPD attribute to be accessible > > for read by all users, while write being restricted to root only. > > > > For the driver, there is a need to opt-in in order to allow this > > functionality. > > I don't think the use case is very strong (and not included at all > here). > > If we do need to do this, I think it's a property of the device, not > the driver. I remember some broken PCI devices, which will crash if VPD is read. Probably not worth opening this can of worms.
On Mon, Nov 11, 2024 at 04:34:30PM -0800, Stephen Hemminger wrote: > On Mon, 11 Nov 2024 14:41:04 -0600 > Bjorn Helgaas <helgaas@kernel.org> wrote: > > > On Thu, Nov 07, 2024 at 08:56:56PM +0200, Leon Romanovsky wrote: > > > From: Leon Romanovsky <leonro@nvidia.com> > > > > > > The Vital Product Data (VPD) attribute is not readable by regular > > > user without root permissions. Such restriction is not really needed > > > for many devices in the world, as data presented in that VPD is not > > > sensitive and access to the HW is safe and tested. > > > > > > This change aligns the permissions of the VPD attribute to be accessible > > > for read by all users, while write being restricted to root only. > > > > > > For the driver, there is a need to opt-in in order to allow this > > > functionality. > > > > I don't think the use case is very strong (and not included at all > > here). > > > > If we do need to do this, I think it's a property of the device, not > > the driver. > > I remember some broken PCI devices, which will crash if VPD is read. This is opt-in feature for devices which are known to be working. Broken devices will continue to be broken and will continue to require root permissions for read. Thanks
On Mon, Nov 11, 2024 at 03:48:04PM -0600, Bjorn Helgaas wrote: > [+cc Thomas] > > On Mon, Nov 11, 2024 at 11:17:38PM +0200, Leon Romanovsky wrote: > > On Mon, Nov 11, 2024 at 02:41:04PM -0600, Bjorn Helgaas wrote: > > > On Thu, Nov 07, 2024 at 08:56:56PM +0200, Leon Romanovsky wrote: > > > > From: Leon Romanovsky <leonro@nvidia.com> > > > > > > > > The Vital Product Data (VPD) attribute is not readable by regular > > > > user without root permissions. Such restriction is not really needed > > > > for many devices in the world, as data presented in that VPD is not > > > > sensitive and access to the HW is safe and tested. > > > > > > > > This change aligns the permissions of the VPD attribute to be accessible > > > > for read by all users, while write being restricted to root only. > > > > > > > > For the driver, there is a need to opt-in in order to allow this > > > > functionality. > > > > > > I don't think the use case is very strong (and not included at all > > > here). > > > > I will add the use case, which is running monitoring application without > > need to be root. IMHO reducing number of applications that require > > privileged access is a very strong case. I personally try to avoid > > applications with root/setuid privileges. > > Avoiding root/setuid is a very good thing. I just don't think using > VPD directly from userspace is a great idea because VPD is so slow and > sometimes unreliable to read. This is one time operation during application initialization, which is fast in our devices. It is used by the NVML https://developer.nvidia.com/management-library-nvml. > And apparently this is a pretty unusual situation since I haven't heard > similar requests for other devices. Maybe they didn't bother to ask. > > Sort of ironic that some vendors, especially Intel and AMD, add new > Device IDs for devices that work exactly the same as their > predecessors, so we are continually adding to the pci_device_id > tables, while here we apparently the same Device ID is used for > devices that differ in ways we actually want to know about. I'm not Intel or AMD employee and never worked there, but from what I heard it is not technical decision but outcome of how their development process is structured. > > > > If we do need to do this, I think it's a property of the device, not > > > the driver. > > > > But how will device inform PCI core about safe VPD read? > > Should I add new field to struct pci_device_id? Add a quirk? > > Otherwise, I will need to add a line "pci_dev->downgrade_vpd_read=true" > > to mlx5 probe function and it won't change a lot from current > > implementation. > > To me it looks like a pci_dev bit, not a pci_driver bit. > > I would set it .probe() so all the code is in one place. And it's not > related to a device defect, as most quirks are. The advantage of quirks is that we will be able to set proper file permissions from the beginning without need to load/bind driver. As Thomas suggested, the vpd_attr_is_visible() will be something like this, which is neat: if (pdev->downgrade_vpd_read) return 644; else return 600; Thanks > > Bjorn >
On 12.11.2024 01:34, Stephen Hemminger wrote: > On Mon, 11 Nov 2024 14:41:04 -0600 > Bjorn Helgaas <helgaas@kernel.org> wrote: > >> On Thu, Nov 07, 2024 at 08:56:56PM +0200, Leon Romanovsky wrote: >>> From: Leon Romanovsky <leonro@nvidia.com> >>> >>> The Vital Product Data (VPD) attribute is not readable by regular >>> user without root permissions. Such restriction is not really needed >>> for many devices in the world, as data presented in that VPD is not >>> sensitive and access to the HW is safe and tested. >>> >>> This change aligns the permissions of the VPD attribute to be accessible >>> for read by all users, while write being restricted to root only. >>> >>> For the driver, there is a need to opt-in in order to allow this >>> functionality. >> >> I don't think the use case is very strong (and not included at all >> here). >> >> If we do need to do this, I think it's a property of the device, not >> the driver. > > I remember some broken PCI devices, which will crash if VPD is read. > Probably not worth opening this can of worms. These crashes shouldn't occur any longer. There are two problematic cases: 1. Reading past end of VPD This used to crash certain devices and was fixed by stop reading at the VPD end tag. 2. Accessing VPD if device firmware isn't correctly loaded and initialized This affects certain LSI devices, which are blacklisted so that PCI core prevents VPD access.
On Tue, Nov 12, 2024 at 07:44:09AM +0100, Heiner Kallweit wrote: > On 12.11.2024 01:34, Stephen Hemminger wrote: > > On Mon, 11 Nov 2024 14:41:04 -0600 > > Bjorn Helgaas <helgaas@kernel.org> wrote: > > > >> On Thu, Nov 07, 2024 at 08:56:56PM +0200, Leon Romanovsky wrote: > >>> From: Leon Romanovsky <leonro@nvidia.com> > >>> > >>> The Vital Product Data (VPD) attribute is not readable by regular > >>> user without root permissions. Such restriction is not really needed > >>> for many devices in the world, as data presented in that VPD is not > >>> sensitive and access to the HW is safe and tested. > >>> > >>> This change aligns the permissions of the VPD attribute to be accessible > >>> for read by all users, while write being restricted to root only. > >>> > >>> For the driver, there is a need to opt-in in order to allow this > >>> functionality. > >> > >> I don't think the use case is very strong (and not included at all > >> here). > >> > >> If we do need to do this, I think it's a property of the device, not > >> the driver. > > > > I remember some broken PCI devices, which will crash if VPD is read. > > Probably not worth opening this can of worms. > > These crashes shouldn't occur any longer. There are two problematic cases: > 1. Reading past end of VPD > This used to crash certain devices and was fixed by stop reading at > the VPD end tag. > 2. Accessing VPD if device firmware isn't correctly loaded and initialized > This affects certain LSI devices, which are blacklisted so that PCI core > prevents VPD access. Thanks for the information. Bjorn, After this response, do you still think that v0 [1] is not the right way to change the read permission? [1] https://lore.kernel.org/all/65791906154e3e5ea12ea49127cf7c707325ca56.1730102428.git.leonro@nvidia.com/ >
On Tue, Nov 12, 2024 at 09:26:04AM +0200, Leon Romanovsky wrote: > On Tue, Nov 12, 2024 at 07:44:09AM +0100, Heiner Kallweit wrote: > > On 12.11.2024 01:34, Stephen Hemminger wrote: > > > On Mon, 11 Nov 2024 14:41:04 -0600 > > > Bjorn Helgaas <helgaas@kernel.org> wrote: > > > > > >> On Thu, Nov 07, 2024 at 08:56:56PM +0200, Leon Romanovsky wrote: > > >>> From: Leon Romanovsky <leonro@nvidia.com> > > >>> > > >>> The Vital Product Data (VPD) attribute is not readable by regular > > >>> user without root permissions. Such restriction is not really needed > > >>> for many devices in the world, as data presented in that VPD is not > > >>> sensitive and access to the HW is safe and tested. > > >>> > > >>> This change aligns the permissions of the VPD attribute to be accessible > > >>> for read by all users, while write being restricted to root only. > > >>> > > >>> For the driver, there is a need to opt-in in order to allow this > > >>> functionality. > > >> > > >> I don't think the use case is very strong (and not included at all > > >> here). > > >> > > >> If we do need to do this, I think it's a property of the device, not > > >> the driver. > > > > > > I remember some broken PCI devices, which will crash if VPD is read. > > > Probably not worth opening this can of worms. > > > > These crashes shouldn't occur any longer. There are two problematic cases: > > 1. Reading past end of VPD > > This used to crash certain devices and was fixed by stop reading at > > the VPD end tag. > > 2. Accessing VPD if device firmware isn't correctly loaded and initialized > > This affects certain LSI devices, which are blacklisted so that PCI core > > prevents VPD access. > > Thanks for the information. > > Bjorn, > > After this response, do you still think that v0 [1] is not the right way > to change the read permission? > > [1] https://lore.kernel.org/all/65791906154e3e5ea12ea49127cf7c707325ca56.1730102428.git.leonro@nvidia.com/ Yes, I still think it's unnecessarily risky to make VPD readable by ordinary users. This is a pretty niche use case. Bjorn
diff --git a/drivers/pci/vpd.c b/drivers/pci/vpd.c index e4300f5f304f..7c70930abaa0 100644 --- a/drivers/pci/vpd.c +++ b/drivers/pci/vpd.c @@ -156,6 +156,7 @@ static ssize_t pci_vpd_read(struct pci_dev *dev, loff_t pos, size_t count, void *arg, bool check_size) { struct pci_vpd *vpd = &dev->vpd; + struct pci_driver *drv; unsigned int max_len; int ret = 0; loff_t end = pos + count; @@ -167,6 +168,12 @@ static ssize_t pci_vpd_read(struct pci_dev *dev, loff_t pos, size_t count, if (pos < 0) return -EINVAL; + if (!capable(CAP_SYS_ADMIN)) { + drv = to_pci_driver(dev->dev.driver); + if (!drv || !drv->downgrade_vpd_read) + return -EPERM; + } + max_len = check_size ? vpd->len : PCI_VPD_MAX_SIZE; if (pos >= max_len) @@ -317,7 +324,7 @@ static ssize_t vpd_write(struct file *filp, struct kobject *kobj, return ret; } -static BIN_ATTR(vpd, 0600, vpd_read, vpd_write, 0); +static BIN_ATTR_RW(vpd, 0); static struct bin_attribute *vpd_attrs[] = { &bin_attr_vpd, diff --git a/include/linux/pci.h b/include/linux/pci.h index 573b4c4c2be6..b8fed74e742e 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -943,6 +943,10 @@ struct module; * how to manage the DMA themselves and set this flag so that * the IOMMU layer will allow them to setup and manage their * own I/O address space. + * @downgrade_vpd_read: Device doesn't require root permissions from the users + * to read VPD information. The driver doesn't expose any sensitive + * information through that interface and safe to be accessed by + * unprivileged users. */ struct pci_driver { const char *name; @@ -960,7 +964,8 @@ struct pci_driver { const struct attribute_group **dev_groups; struct device_driver driver; struct pci_dynids dynids; - bool driver_managed_dma; + bool driver_managed_dma : 1; + bool downgrade_vpd_read : 1; }; #define to_pci_driver(__drv) \