[v3,12/26] compat_ioctl: move more drivers to compat_ptr_ioctl

  The .ioctl and .compat_ioctl file operations have the same prototype so
they can both point to the same function, which works great almost all
the time when all the commands are compatible.

One exception is the s390 architecture, where a compat pointer is only
31 bit wide, and converting it into a 64-bit pointer requires calling
compat_ptr(). Most drivers here will ever run in s390, but since we now
have a generic helper for it, it's easy enough to use it consistently.

I double-checked all these drivers to ensure that all ioctl arguments
are used as pointers or are ignored, but are not interpreted as integer
values.

Acked-by: Jason Gunthorpe <jgg@mellanox.com>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Acked-by: David Sterba <dsterba@suse.com>
Acked-by: Darren Hart (VMware) <dvhart@infradead.org>
Acked-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Acked-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 drivers/android/binder.c                    | 2 +-
 drivers/crypto/qat/qat_common/adf_ctl_drv.c | 2 +-
 drivers/dma-buf/dma-buf.c                   | 4 +---
 drivers/dma-buf/sw_sync.c                   | 2 +-
 drivers/dma-buf/sync_file.c                 | 2 +-
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c    | 2 +-
 drivers/hid/hidraw.c                        | 4 +---
 drivers/iio/industrialio-core.c             | 2 +-
 drivers/infiniband/core/uverbs_main.c       | 4 ++--
 drivers/media/rc/lirc_dev.c                 | 4 +---
 drivers/mfd/cros_ec_dev.c                   | 4 +---
 drivers/misc/vmw_vmci/vmci_host.c           | 2 +-
 drivers/nvdimm/bus.c                        | 4 ++--
 drivers/nvme/host/core.c                    | 2 +-
 drivers/pci/switch/switchtec.c              | 2 +-
 drivers/platform/x86/wmi.c                  | 2 +-
 drivers/rpmsg/rpmsg_char.c                  | 4 ++--
 drivers/sbus/char/display7seg.c             | 2 +-
 drivers/sbus/char/envctrl.c                 | 4 +---
 drivers/scsi/3w-xxxx.c                      | 4 +---
 drivers/scsi/cxlflash/main.c                | 2 +-
 drivers/scsi/esas2r/esas2r_main.c           | 2 +-
 drivers/scsi/pmcraid.c                      | 4 +---
 drivers/staging/android/ion/ion.c           | 4 +---
 drivers/staging/vme/devices/vme_user.c      | 2 +-
 drivers/tee/tee_core.c                      | 2 +-
 drivers/usb/class/cdc-wdm.c                 | 2 +-
 drivers/usb/class/usbtmc.c                  | 4 +---
 drivers/virt/fsl_hypervisor.c               | 2 +-
 fs/btrfs/super.c                            | 2 +-
 fs/ceph/dir.c                               | 2 +-
 fs/ceph/file.c                              | 2 +-
 fs/fuse/dev.c                               | 2 +-
 fs/notify/fanotify/fanotify_user.c          | 2 +-
 fs/userfaultfd.c                            | 2 +-
 net/rfkill/core.c                           | 2 +-
 36 files changed, 39 insertions(+), 57 deletions(-)
  

Em Tue, 16 Apr 2019 22:25:33 +0200
Arnd Bergmann <arnd@arndb.de> escreveu:

> The .ioctl and .compat_ioctl file operations have the same prototype so
> they can both point to the same function, which works great almost all
> the time when all the commands are compatible.
> 
> One exception is the s390 architecture, where a compat pointer is only
> 31 bit wide, and converting it into a 64-bit pointer requires calling
> compat_ptr(). Most drivers here will ever run in s390, but since we now
> have a generic helper for it, it's easy enough to use it consistently.
> 
> I double-checked all these drivers to ensure that all ioctl arguments
> are used as pointers or are ignored, but are not interpreted as integer
> values.
> 
> Acked-by: Jason Gunthorpe <jgg@mellanox.com>
> Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
> Acked-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
> Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> Acked-by: David Sterba <dsterba@suse.com>
> Acked-by: Darren Hart (VMware) <dvhart@infradead.org>
> Acked-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> Acked-by: Bjorn Andersson <bjorn.andersson@linaro.org>
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>  drivers/android/binder.c                    | 2 +-
>  drivers/crypto/qat/qat_common/adf_ctl_drv.c | 2 +-
>  drivers/dma-buf/dma-buf.c                   | 4 +---
>  drivers/dma-buf/sw_sync.c                   | 2 +-
>  drivers/dma-buf/sync_file.c                 | 2 +-
>  drivers/gpu/drm/amd/amdkfd/kfd_chardev.c    | 2 +-
>  drivers/hid/hidraw.c                        | 4 +---
>  drivers/iio/industrialio-core.c             | 2 +-
>  drivers/infiniband/core/uverbs_main.c       | 4 ++--
>  drivers/media/rc/lirc_dev.c                 | 4 +---

If I understand your patch description well, using compat_ptr_ioctl
only works if the driver is not for s390, right?

In thesis, nothing prevents to use LIRC API on s390 - as this isn't
a driver but, instead, RC core feature to expose raw remote controller
codes to userspace.

Yet, lirc_dev will only work if the system has a remote controller driver.

Well, we don't have any for s390. Despite we don't have such driver, 
I can't possible see why someone would use a remote controller for a
mainframe :-p

Anyway, if someone ever come with such driver/usecase, reverting this
change (or adding an #ifdef to check if arch is 390) should be
pretty straight forward.

So:

Acked-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>



>  drivers/mfd/cros_ec_dev.c                   | 4 +---
>  drivers/misc/vmw_vmci/vmci_host.c           | 2 +-
>  drivers/nvdimm/bus.c                        | 4 ++--
>  drivers/nvme/host/core.c                    | 2 +-
>  drivers/pci/switch/switchtec.c              | 2 +-
>  drivers/platform/x86/wmi.c                  | 2 +-
>  drivers/rpmsg/rpmsg_char.c                  | 4 ++--
>  drivers/sbus/char/display7seg.c             | 2 +-
>  drivers/sbus/char/envctrl.c                 | 4 +---
>  drivers/scsi/3w-xxxx.c                      | 4 +---
>  drivers/scsi/cxlflash/main.c                | 2 +-
>  drivers/scsi/esas2r/esas2r_main.c           | 2 +-
>  drivers/scsi/pmcraid.c                      | 4 +---
>  drivers/staging/android/ion/ion.c           | 4 +---
>  drivers/staging/vme/devices/vme_user.c      | 2 +-
>  drivers/tee/tee_core.c                      | 2 +-
>  drivers/usb/class/cdc-wdm.c                 | 2 +-
>  drivers/usb/class/usbtmc.c                  | 4 +---
>  drivers/virt/fsl_hypervisor.c               | 2 +-
>  fs/btrfs/super.c                            | 2 +-
>  fs/ceph/dir.c                               | 2 +-
>  fs/ceph/file.c                              | 2 +-
>  fs/fuse/dev.c                               | 2 +-
>  fs/notify/fanotify/fanotify_user.c          | 2 +-
>  fs/userfaultfd.c                            | 2 +-
>  net/rfkill/core.c                           | 2 +-
>  36 files changed, 39 insertions(+), 57 deletions(-)
> 
> diff --git a/drivers/android/binder.c b/drivers/android/binder.c
> index 4b9c7ca492e6..48109ade7234 100644
> --- a/drivers/android/binder.c
> +++ b/drivers/android/binder.c
> @@ -5998,7 +5998,7 @@ const struct file_operations binder_fops = {
>  	.owner = THIS_MODULE,
>  	.poll = binder_poll,
>  	.unlocked_ioctl = binder_ioctl,
> -	.compat_ioctl = binder_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  	.mmap = binder_mmap,
>  	.open = binder_open,
>  	.flush = binder_flush,
> diff --git a/drivers/crypto/qat/qat_common/adf_ctl_drv.c b/drivers/crypto/qat/qat_common/adf_ctl_drv.c
> index abc7a7f64d64..ef0e482ee04f 100644
> --- a/drivers/crypto/qat/qat_common/adf_ctl_drv.c
> +++ b/drivers/crypto/qat/qat_common/adf_ctl_drv.c
> @@ -68,7 +68,7 @@ static long adf_ctl_ioctl(struct file *fp, unsigned int cmd, unsigned long arg);
>  static const struct file_operations adf_ctl_ops = {
>  	.owner = THIS_MODULE,
>  	.unlocked_ioctl = adf_ctl_ioctl,
> -	.compat_ioctl = adf_ctl_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  };
>  
>  struct adf_ctl_drv_info {
> diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c
> index 7c858020d14b..0cb336fe6324 100644
> --- a/drivers/dma-buf/dma-buf.c
> +++ b/drivers/dma-buf/dma-buf.c
> @@ -325,9 +325,7 @@ static const struct file_operations dma_buf_fops = {
>  	.llseek		= dma_buf_llseek,
>  	.poll		= dma_buf_poll,
>  	.unlocked_ioctl	= dma_buf_ioctl,
> -#ifdef CONFIG_COMPAT
> -	.compat_ioctl	= dma_buf_ioctl,
> -#endif
> +	.compat_ioctl	= compat_ptr_ioctl,
>  };
>  
>  /*
> diff --git a/drivers/dma-buf/sw_sync.c b/drivers/dma-buf/sw_sync.c
> index 32dcf7b4c935..411de6a8a0ad 100644
> --- a/drivers/dma-buf/sw_sync.c
> +++ b/drivers/dma-buf/sw_sync.c
> @@ -419,5 +419,5 @@ const struct file_operations sw_sync_debugfs_fops = {
>  	.open           = sw_sync_debugfs_open,
>  	.release        = sw_sync_debugfs_release,
>  	.unlocked_ioctl = sw_sync_ioctl,
> -	.compat_ioctl	= sw_sync_ioctl,
> +	.compat_ioctl	= compat_ptr_ioctl,
>  };
> diff --git a/drivers/dma-buf/sync_file.c b/drivers/dma-buf/sync_file.c
> index 4f6305ca52c8..0949f91eb85f 100644
> --- a/drivers/dma-buf/sync_file.c
> +++ b/drivers/dma-buf/sync_file.c
> @@ -488,5 +488,5 @@ static const struct file_operations sync_file_fops = {
>  	.release = sync_file_release,
>  	.poll = sync_file_poll,
>  	.unlocked_ioctl = sync_file_ioctl,
> -	.compat_ioctl = sync_file_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  };
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> index 083bd8114db1..5d6ac7885aa7 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> @@ -49,7 +49,7 @@ static const char kfd_dev_name[] = "kfd";
>  static const struct file_operations kfd_fops = {
>  	.owner = THIS_MODULE,
>  	.unlocked_ioctl = kfd_ioctl,
> -	.compat_ioctl = kfd_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  	.open = kfd_open,
>  	.mmap = kfd_mmap,
>  };
> diff --git a/drivers/hid/hidraw.c b/drivers/hid/hidraw.c
> index 9fc51eff1079..e7284d38b66d 100644
> --- a/drivers/hid/hidraw.c
> +++ b/drivers/hid/hidraw.c
> @@ -476,9 +476,7 @@ static const struct file_operations hidraw_ops = {
>  	.release =      hidraw_release,
>  	.unlocked_ioctl = hidraw_ioctl,
>  	.fasync =	hidraw_fasync,
> -#ifdef CONFIG_COMPAT
> -	.compat_ioctl   = hidraw_ioctl,
> -#endif
> +	.compat_ioctl   = compat_ptr_ioctl,
>  	.llseek =	noop_llseek,
>  };
>  
> diff --git a/drivers/iio/industrialio-core.c b/drivers/iio/industrialio-core.c
> index 4700fd5d8c90..eed1bea257b4 100644
> --- a/drivers/iio/industrialio-core.c
> +++ b/drivers/iio/industrialio-core.c
> @@ -1635,7 +1635,7 @@ static const struct file_operations iio_buffer_fileops = {
>  	.owner = THIS_MODULE,
>  	.llseek = noop_llseek,
>  	.unlocked_ioctl = iio_ioctl,
> -	.compat_ioctl = iio_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  };
>  
>  static int iio_check_unique_scan_index(struct iio_dev *indio_dev)
> diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c
> index 70b7d80431a9..ac4321d7c800 100644
> --- a/drivers/infiniband/core/uverbs_main.c
> +++ b/drivers/infiniband/core/uverbs_main.c
> @@ -1120,7 +1120,7 @@ static const struct file_operations uverbs_fops = {
>  	.release = ib_uverbs_close,
>  	.llseek	 = no_llseek,
>  	.unlocked_ioctl = ib_uverbs_ioctl,
> -	.compat_ioctl = ib_uverbs_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  };
>  
>  static const struct file_operations uverbs_mmap_fops = {
> @@ -1131,7 +1131,7 @@ static const struct file_operations uverbs_mmap_fops = {
>  	.release = ib_uverbs_close,
>  	.llseek	 = no_llseek,
>  	.unlocked_ioctl = ib_uverbs_ioctl,
> -	.compat_ioctl = ib_uverbs_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  };
>  
>  static struct ib_client uverbs_client = {
> diff --git a/drivers/media/rc/lirc_dev.c b/drivers/media/rc/lirc_dev.c
> index f862f1b7f996..9ccc7e9cbc8e 100644
> --- a/drivers/media/rc/lirc_dev.c
> +++ b/drivers/media/rc/lirc_dev.c
> @@ -730,9 +730,7 @@ static const struct file_operations lirc_fops = {
>  	.owner		= THIS_MODULE,
>  	.write		= ir_lirc_transmit_ir,
>  	.unlocked_ioctl	= ir_lirc_ioctl,
> -#ifdef CONFIG_COMPAT
> -	.compat_ioctl	= ir_lirc_ioctl,
> -#endif
> +	.compat_ioctl	= compat_ptr_ioctl,
>  	.read		= ir_lirc_read,
>  	.poll		= ir_lirc_poll,
>  	.open		= ir_lirc_open,
> diff --git a/drivers/mfd/cros_ec_dev.c b/drivers/mfd/cros_ec_dev.c
> index d275deaecb12..4a602a40d75c 100644
> --- a/drivers/mfd/cros_ec_dev.c
> +++ b/drivers/mfd/cros_ec_dev.c
> @@ -251,9 +251,7 @@ static const struct file_operations fops = {
>  	.release = ec_device_release,
>  	.read = ec_device_read,
>  	.unlocked_ioctl = ec_device_ioctl,
> -#ifdef CONFIG_COMPAT
> -	.compat_ioctl = ec_device_ioctl,
> -#endif
> +	.compat_ioctl = compat_ptr_ioctl,
>  };
>  
>  static void cros_ec_class_release(struct device *dev)
> diff --git a/drivers/misc/vmw_vmci/vmci_host.c b/drivers/misc/vmw_vmci/vmci_host.c
> index 997f92543dd4..5bb406dabe85 100644
> --- a/drivers/misc/vmw_vmci/vmci_host.c
> +++ b/drivers/misc/vmw_vmci/vmci_host.c
> @@ -969,7 +969,7 @@ static const struct file_operations vmuser_fops = {
>  	.release	= vmci_host_close,
>  	.poll		= vmci_host_poll,
>  	.unlocked_ioctl	= vmci_host_unlocked_ioctl,
> -	.compat_ioctl	= vmci_host_unlocked_ioctl,
> +	.compat_ioctl	= compat_ptr_ioctl,
>  };
>  
>  static struct miscdevice vmci_host_miscdev = {
> diff --git a/drivers/nvdimm/bus.c b/drivers/nvdimm/bus.c
> index 7bbff0af29b2..065ebd584482 100644
> --- a/drivers/nvdimm/bus.c
> +++ b/drivers/nvdimm/bus.c
> @@ -1167,7 +1167,7 @@ static const struct file_operations nvdimm_bus_fops = {
>  	.owner = THIS_MODULE,
>  	.open = nd_open,
>  	.unlocked_ioctl = nd_ioctl,
> -	.compat_ioctl = nd_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  	.llseek = noop_llseek,
>  };
>  
> @@ -1175,7 +1175,7 @@ static const struct file_operations nvdimm_fops = {
>  	.owner = THIS_MODULE,
>  	.open = nd_open,
>  	.unlocked_ioctl = nvdimm_ioctl,
> -	.compat_ioctl = nvdimm_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  	.llseek = noop_llseek,
>  };
>  
> diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
> index 2c43e12b70af..560929bee5ce 100644
> --- a/drivers/nvme/host/core.c
> +++ b/drivers/nvme/host/core.c
> @@ -2739,7 +2739,7 @@ static const struct file_operations nvme_dev_fops = {
>  	.owner		= THIS_MODULE,
>  	.open		= nvme_dev_open,
>  	.unlocked_ioctl	= nvme_dev_ioctl,
> -	.compat_ioctl	= nvme_dev_ioctl,
> +	.compat_ioctl	= compat_ptr_ioctl,
>  };
>  
>  static ssize_t nvme_sysfs_reset(struct device *dev,
> diff --git a/drivers/pci/switch/switchtec.c b/drivers/pci/switch/switchtec.c
> index e22766c79fe9..3a54b4b616e2 100644
> --- a/drivers/pci/switch/switchtec.c
> +++ b/drivers/pci/switch/switchtec.c
> @@ -1006,7 +1006,7 @@ static const struct file_operations switchtec_fops = {
>  	.read = switchtec_dev_read,
>  	.poll = switchtec_dev_poll,
>  	.unlocked_ioctl = switchtec_dev_ioctl,
> -	.compat_ioctl = switchtec_dev_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  };
>  
>  static void link_event_work(struct work_struct *work)
> diff --git a/drivers/platform/x86/wmi.c b/drivers/platform/x86/wmi.c
> index 7b26b6ccf1a0..dded9cef42f4 100644
> --- a/drivers/platform/x86/wmi.c
> +++ b/drivers/platform/x86/wmi.c
> @@ -889,7 +889,7 @@ static const struct file_operations wmi_fops = {
>  	.read		= wmi_char_read,
>  	.open		= wmi_char_open,
>  	.unlocked_ioctl	= wmi_ioctl,
> -	.compat_ioctl	= wmi_ioctl,
> +	.compat_ioctl	= compat_ptr_ioctl,
>  };
>  
>  static int wmi_dev_probe(struct device *dev)
> diff --git a/drivers/rpmsg/rpmsg_char.c b/drivers/rpmsg/rpmsg_char.c
> index eea5ebbb5119..507bfe163883 100644
> --- a/drivers/rpmsg/rpmsg_char.c
> +++ b/drivers/rpmsg/rpmsg_char.c
> @@ -290,7 +290,7 @@ static const struct file_operations rpmsg_eptdev_fops = {
>  	.write_iter = rpmsg_eptdev_write_iter,
>  	.poll = rpmsg_eptdev_poll,
>  	.unlocked_ioctl = rpmsg_eptdev_ioctl,
> -	.compat_ioctl = rpmsg_eptdev_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  };
>  
>  static ssize_t name_show(struct device *dev, struct device_attribute *attr,
> @@ -451,7 +451,7 @@ static const struct file_operations rpmsg_ctrldev_fops = {
>  	.open = rpmsg_ctrldev_open,
>  	.release = rpmsg_ctrldev_release,
>  	.unlocked_ioctl = rpmsg_ctrldev_ioctl,
> -	.compat_ioctl = rpmsg_ctrldev_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  };
>  
>  static void rpmsg_ctrldev_release_device(struct device *dev)
> diff --git a/drivers/sbus/char/display7seg.c b/drivers/sbus/char/display7seg.c
> index a36e4cf1841d..c9f60656f54d 100644
> --- a/drivers/sbus/char/display7seg.c
> +++ b/drivers/sbus/char/display7seg.c
> @@ -155,7 +155,7 @@ static long d7s_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
>  static const struct file_operations d7s_fops = {
>  	.owner =		THIS_MODULE,
>  	.unlocked_ioctl =	d7s_ioctl,
> -	.compat_ioctl =		d7s_ioctl,
> +	.compat_ioctl =		compat_ptr_ioctl,
>  	.open =			d7s_open,
>  	.release =		d7s_release,
>  	.llseek = noop_llseek,
> diff --git a/drivers/sbus/char/envctrl.c b/drivers/sbus/char/envctrl.c
> index 1a6e7224017c..dd2dfa85fc68 100644
> --- a/drivers/sbus/char/envctrl.c
> +++ b/drivers/sbus/char/envctrl.c
> @@ -714,9 +714,7 @@ static const struct file_operations envctrl_fops = {
>  	.owner =		THIS_MODULE,
>  	.read =			envctrl_read,
>  	.unlocked_ioctl =	envctrl_ioctl,
> -#ifdef CONFIG_COMPAT
> -	.compat_ioctl =		envctrl_ioctl,
> -#endif
> +	.compat_ioctl =		compat_ptr_ioctl,
>  	.open =			envctrl_open,
>  	.release =		envctrl_release,
>  	.llseek =		noop_llseek,
> diff --git a/drivers/scsi/3w-xxxx.c b/drivers/scsi/3w-xxxx.c
> index 2b1e0d503020..fb6444d0409c 100644
> --- a/drivers/scsi/3w-xxxx.c
> +++ b/drivers/scsi/3w-xxxx.c
> @@ -1049,9 +1049,7 @@ static int tw_chrdev_open(struct inode *inode, struct file *file)
>  static const struct file_operations tw_fops = {
>  	.owner		= THIS_MODULE,
>  	.unlocked_ioctl	= tw_chrdev_ioctl,
> -#ifdef CONFIG_COMPAT
> -	.compat_ioctl   = tw_chrdev_ioctl,
> -#endif
> +	.compat_ioctl   = compat_ptr_ioctl,
>  	.open		= tw_chrdev_open,
>  	.release	= NULL,
>  	.llseek		= noop_llseek,
> diff --git a/drivers/scsi/cxlflash/main.c b/drivers/scsi/cxlflash/main.c
> index 7096810fd222..e13d5de1d76e 100644
> --- a/drivers/scsi/cxlflash/main.c
> +++ b/drivers/scsi/cxlflash/main.c
> @@ -3589,7 +3589,7 @@ static const struct file_operations cxlflash_chr_fops = {
>  	.owner          = THIS_MODULE,
>  	.open           = cxlflash_chr_open,
>  	.unlocked_ioctl	= cxlflash_chr_ioctl,
> -	.compat_ioctl	= cxlflash_chr_ioctl,
> +	.compat_ioctl	= compat_ptr_ioctl,
>  };
>  
>  /**
> diff --git a/drivers/scsi/esas2r/esas2r_main.c b/drivers/scsi/esas2r/esas2r_main.c
> index fdbda5c05aa0..80c5a235d193 100644
> --- a/drivers/scsi/esas2r/esas2r_main.c
> +++ b/drivers/scsi/esas2r/esas2r_main.c
> @@ -613,7 +613,7 @@ static int __init esas2r_init(void)
>  
>  /* Handle ioctl calls to "/proc/scsi/esas2r/ATTOnode" */
>  static const struct file_operations esas2r_proc_fops = {
> -	.compat_ioctl	= esas2r_proc_ioctl,
> +	.compat_ioctl	= compat_ptr_ioctl,
>  	.unlocked_ioctl = esas2r_proc_ioctl,
>  };
>  
> diff --git a/drivers/scsi/pmcraid.c b/drivers/scsi/pmcraid.c
> index e338d7a4f571..c0a1a1218c56 100644
> --- a/drivers/scsi/pmcraid.c
> +++ b/drivers/scsi/pmcraid.c
> @@ -3988,9 +3988,7 @@ static const struct file_operations pmcraid_fops = {
>  	.open = pmcraid_chr_open,
>  	.fasync = pmcraid_chr_fasync,
>  	.unlocked_ioctl = pmcraid_chr_ioctl,
> -#ifdef CONFIG_COMPAT
> -	.compat_ioctl = pmcraid_chr_ioctl,
> -#endif
> +	.compat_ioctl = compat_ptr_ioctl,
>  	.llseek = noop_llseek,
>  };
>  
> diff --git a/drivers/staging/android/ion/ion.c b/drivers/staging/android/ion/ion.c
> index 92c2914239e3..1663c163edca 100644
> --- a/drivers/staging/android/ion/ion.c
> +++ b/drivers/staging/android/ion/ion.c
> @@ -567,9 +567,7 @@ static long ion_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
>  static const struct file_operations ion_fops = {
>  	.owner          = THIS_MODULE,
>  	.unlocked_ioctl = ion_ioctl,
> -#ifdef CONFIG_COMPAT
> -	.compat_ioctl	= ion_ioctl,
> -#endif
> +	.compat_ioctl	= compat_ptr_ioctl,
>  };
>  
>  static int debug_shrink_set(void *data, u64 val)
> diff --git a/drivers/staging/vme/devices/vme_user.c b/drivers/staging/vme/devices/vme_user.c
> index 6a33aaa1a49f..fd0ea4dbcb91 100644
> --- a/drivers/staging/vme/devices/vme_user.c
> +++ b/drivers/staging/vme/devices/vme_user.c
> @@ -494,7 +494,7 @@ static const struct file_operations vme_user_fops = {
>  	.write = vme_user_write,
>  	.llseek = vme_user_llseek,
>  	.unlocked_ioctl = vme_user_unlocked_ioctl,
> -	.compat_ioctl = vme_user_unlocked_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  	.mmap = vme_user_mmap,
>  };
>  
> diff --git a/drivers/tee/tee_core.c b/drivers/tee/tee_core.c
> index 17c64fccbb10..eb97acf09868 100644
> --- a/drivers/tee/tee_core.c
> +++ b/drivers/tee/tee_core.c
> @@ -684,7 +684,7 @@ static const struct file_operations tee_fops = {
>  	.open = tee_open,
>  	.release = tee_release,
>  	.unlocked_ioctl = tee_ioctl,
> -	.compat_ioctl = tee_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  };
>  
>  static void tee_release_device(struct device *dev)
> diff --git a/drivers/usb/class/cdc-wdm.c b/drivers/usb/class/cdc-wdm.c
> index 9e9caff905d5..d48c032580d0 100644
> --- a/drivers/usb/class/cdc-wdm.c
> +++ b/drivers/usb/class/cdc-wdm.c
> @@ -724,7 +724,7 @@ static const struct file_operations wdm_fops = {
>  	.release =	wdm_release,
>  	.poll =		wdm_poll,
>  	.unlocked_ioctl = wdm_ioctl,
> -	.compat_ioctl = wdm_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  	.llseek =	noop_llseek,
>  };
>  
> diff --git a/drivers/usb/class/usbtmc.c b/drivers/usb/class/usbtmc.c
> index 4942122b2346..bbd0308b13f5 100644
> --- a/drivers/usb/class/usbtmc.c
> +++ b/drivers/usb/class/usbtmc.c
> @@ -2220,9 +2220,7 @@ static const struct file_operations fops = {
>  	.release	= usbtmc_release,
>  	.flush		= usbtmc_flush,
>  	.unlocked_ioctl	= usbtmc_ioctl,
> -#ifdef CONFIG_COMPAT
> -	.compat_ioctl	= usbtmc_ioctl,
> -#endif
> +	.compat_ioctl	= compat_ptr_ioctl,
>  	.fasync         = usbtmc_fasync,
>  	.poll           = usbtmc_poll,
>  	.llseek		= default_llseek,
> diff --git a/drivers/virt/fsl_hypervisor.c b/drivers/virt/fsl_hypervisor.c
> index 8ba726e600e9..fbf02bf60f62 100644
> --- a/drivers/virt/fsl_hypervisor.c
> +++ b/drivers/virt/fsl_hypervisor.c
> @@ -703,7 +703,7 @@ static const struct file_operations fsl_hv_fops = {
>  	.poll = fsl_hv_poll,
>  	.read = fsl_hv_read,
>  	.unlocked_ioctl = fsl_hv_ioctl,
> -	.compat_ioctl = fsl_hv_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  };
>  
>  static struct miscdevice fsl_hv_misc_dev = {
> diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
> index 120e4340792a..162ea4b6b417 100644
> --- a/fs/btrfs/super.c
> +++ b/fs/btrfs/super.c
> @@ -2307,7 +2307,7 @@ static const struct super_operations btrfs_super_ops = {
>  static const struct file_operations btrfs_ctl_fops = {
>  	.open = btrfs_control_open,
>  	.unlocked_ioctl	 = btrfs_control_ioctl,
> -	.compat_ioctl = btrfs_control_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  	.owner	 = THIS_MODULE,
>  	.llseek = noop_llseek,
>  };
> diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
> index 7c060cb22aa3..a493b957713f 100644
> --- a/fs/ceph/dir.c
> +++ b/fs/ceph/dir.c
> @@ -1785,7 +1785,7 @@ const struct file_operations ceph_dir_fops = {
>  	.open = ceph_open,
>  	.release = ceph_release,
>  	.unlocked_ioctl = ceph_ioctl,
> -	.compat_ioctl = ceph_ioctl,
> +	.compat_ioctl = compat_ptr_ioctl,
>  	.fsync = ceph_fsync,
>  	.lock = ceph_lock,
>  	.flock = ceph_flock,
> diff --git a/fs/ceph/file.c b/fs/ceph/file.c
> index 9f53c3d99304..9b5fe7eee3c1 100644
> --- a/fs/ceph/file.c
> +++ b/fs/ceph/file.c
> @@ -2112,7 +2112,7 @@ const struct file_operations ceph_file_fops = {
>  	.splice_read = generic_file_splice_read,
>  	.splice_write = iter_file_splice_write,
>  	.unlocked_ioctl = ceph_ioctl,
> -	.compat_ioctl	= ceph_ioctl,
> +	.compat_ioctl	= compat_ptr_ioctl,
>  	.fallocate	= ceph_fallocate,
>  	.copy_file_range = ceph_copy_file_range,
>  };
> diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
> index 9971a35cf1ef..dcdb26068b71 100644
> --- a/fs/fuse/dev.c
> +++ b/fs/fuse/dev.c
> @@ -2354,7 +2354,7 @@ const struct file_operations fuse_dev_operations = {
>  	.release	= fuse_dev_release,
>  	.fasync		= fuse_dev_fasync,
>  	.unlocked_ioctl = fuse_dev_ioctl,
> -	.compat_ioctl   = fuse_dev_ioctl,
> +	.compat_ioctl   = compat_ptr_ioctl,
>  };
>  EXPORT_SYMBOL_GPL(fuse_dev_operations);
>  
> diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
> index a90bb19dcfa2..a55aa029a308 100644
> --- a/fs/notify/fanotify/fanotify_user.c
> +++ b/fs/notify/fanotify/fanotify_user.c
> @@ -523,7 +523,7 @@ static const struct file_operations fanotify_fops = {
>  	.fasync		= NULL,
>  	.release	= fanotify_release,
>  	.unlocked_ioctl	= fanotify_ioctl,
> -	.compat_ioctl	= fanotify_ioctl,
> +	.compat_ioctl	= compat_ptr_ioctl,
>  	.llseek		= noop_llseek,
>  };
>  
> diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c
> index 89800fc7dc9d..f93dcf8c996f 100644
> --- a/fs/userfaultfd.c
> +++ b/fs/userfaultfd.c
> @@ -1901,7 +1901,7 @@ static const struct file_operations userfaultfd_fops = {
>  	.poll		= userfaultfd_poll,
>  	.read		= userfaultfd_read,
>  	.unlocked_ioctl = userfaultfd_ioctl,
> -	.compat_ioctl	= userfaultfd_ioctl,
> +	.compat_ioctl	= compat_ptr_ioctl,
>  	.llseek		= noop_llseek,
>  };
>  
> diff --git a/net/rfkill/core.c b/net/rfkill/core.c
> index abca57040f37..3b2f6ea44397 100644
> --- a/net/rfkill/core.c
> +++ b/net/rfkill/core.c
> @@ -1323,7 +1323,7 @@ static const struct file_operations rfkill_fops = {
>  	.release	= rfkill_fop_release,
>  #ifdef CONFIG_RFKILL_INPUT
>  	.unlocked_ioctl	= rfkill_fop_ioctl,
> -	.compat_ioctl	= rfkill_fop_ioctl,
> +	.compat_ioctl	= compat_ptr_ioctl,
>  #endif
>  	.llseek		= no_llseek,
>  };



Thanks,
Mauro
  

On Thu, Apr 25, 2019 at 5:22 PM Mauro Carvalho Chehab
<mchehab+samsung@kernel.org> wrote:
> Em Tue, 16 Apr 2019 22:25:33 +0200 Arnd Bergmann <arnd@arndb.de> escreveu:
>
> If I understand your patch description well, using compat_ptr_ioctl
> only works if the driver is not for s390, right?

No, the purpose of compat_ptr_ioctl() is to make sure it works
everywhere including s390.

Even on s390 it tends to work most of the time, but for correctness
the upper bit of a 32-bit pointer needs to be cleared, as
compat_ptr_ioctl does, in case some application passes a pointer
with that bit set. [IIRC, in the instruction pointer, the high bit is set, in
data references it is ignored but usually cleared, but it may be left
on for IP-relative address generation]

       Arnd
  

On Thu, Apr 25, 2019 at 12:21:53PM -0300, Mauro Carvalho Chehab wrote:

> If I understand your patch description well, using compat_ptr_ioctl
> only works if the driver is not for s390, right?

No; s390 is where "oh, just set ->compat_ioctl same as ->unlocked_ioctl
and be done with that; compat_ptr() is a no-op anyway" breaks.  IOW,
s390 is the reason for having compat_ptr_ioctl() in the first place;
that thing works on all biarch architectures, as long as all stuff
handled by ->ioctl() takes pointer to arch-independent object as
argument.  IOW,
	argument ignored => OK
	any arithmetical type => no go, compat_ptr() would bugger it
	pointer to int => OK
	pointer to string => OK
	pointer to u64 => OK
	pointer to struct {u64 addr; char s[11];} => OK
	pointer to long => needs explicit handler
	pointer to struct {void *addr; char s[11];} => needs explicit handler
	pointer to struct {int x; u64 y;} => needs explicit handler on amd64
For "just use ->unlocked_ioctl for ->ioctl" we have
	argument ignored => OK
	any arithmetical type => OK
	any pointer => instant breakage on s390, in addtion to cases that break
with compat_ptr_ioctl().

Probably some form of that ought to go into commit message for compat_ptr_ioctl()
introduction...
  

Em Thu, 25 Apr 2019 16:35:34 +0100
Al Viro <viro@zeniv.linux.org.uk> escreveu:

> On Thu, Apr 25, 2019 at 12:21:53PM -0300, Mauro Carvalho Chehab wrote:
> 
> > If I understand your patch description well, using compat_ptr_ioctl
> > only works if the driver is not for s390, right?  
> 
> No; s390 is where "oh, just set ->compat_ioctl same as ->unlocked_ioctl
> and be done with that; compat_ptr() is a no-op anyway" breaks.  IOW,
> s390 is the reason for having compat_ptr_ioctl() in the first place;
> that thing works on all biarch architectures, as long as all stuff
> handled by ->ioctl() takes pointer to arch-independent object as
> argument.  IOW,
> 	argument ignored => OK
> 	any arithmetical type => no go, compat_ptr() would bugger it
> 	pointer to int => OK

That's the case for all LIRC ioctls: they all use a pointer to u32
argument.

> 	pointer to string => OK
> 	pointer to u64 => OK
> 	pointer to struct {u64 addr; char s[11];} => OK
> 	pointer to long => needs explicit handler
> 	pointer to struct {void *addr; char s[11];} => needs explicit handler
> 	pointer to struct {int x; u64 y;} => needs explicit handler on amd64
> For "just use ->unlocked_ioctl for ->ioctl" we have
> 	argument ignored => OK
> 	any arithmetical type => OK
> 	any pointer => instant breakage on s390, in addtion to cases that break
> with compat_ptr_ioctl().
> 
> Probably some form of that ought to go into commit message for compat_ptr_ioctl()
> introduction...

Agreed.

Thanks,
Mauro
  

On Thu, Apr 25, 2019 at 5:35 PM Al Viro <viro@zeniv.linux.org.uk> wrote:
>
> On Thu, Apr 25, 2019 at 12:21:53PM -0300, Mauro Carvalho Chehab wrote:
>
> > If I understand your patch description well, using compat_ptr_ioctl
> > only works if the driver is not for s390, right?
>
> No; s390 is where "oh, just set ->compat_ioctl same as ->unlocked_ioctl
> and be done with that; compat_ptr() is a no-op anyway" breaks.  IOW,
> s390 is the reason for having compat_ptr_ioctl() in the first place;
> that thing works on all biarch architectures, as long as all stuff
> handled by ->ioctl() takes pointer to arch-independent object as
> argument.  IOW,
>         argument ignored => OK
>         any arithmetical type => no go, compat_ptr() would bugger it
>         pointer to int => OK
>         pointer to string => OK
>         pointer to u64 => OK
>         pointer to struct {u64 addr; char s[11];} => OK

To be extra pedantic, the 'struct {u64 addr; char s[11];} '
case is also broken on x86, because sizeof (obj) is smaller
on i386, even though the location of the members are
the same. i.e. you can copy_from_user() this, but not
copy_to_user(), which overwrites 4 bytes after the end of
the 20-byte user structure.

       Arnd
  

On Thu, Apr 25, 2019 at 05:55:23PM +0200, Arnd Bergmann wrote:
> On Thu, Apr 25, 2019 at 5:35 PM Al Viro <viro@zeniv.linux.org.uk> wrote:
> >
> > On Thu, Apr 25, 2019 at 12:21:53PM -0300, Mauro Carvalho Chehab wrote:
> >
> > > If I understand your patch description well, using compat_ptr_ioctl
> > > only works if the driver is not for s390, right?
> >
> > No; s390 is where "oh, just set ->compat_ioctl same as ->unlocked_ioctl
> > and be done with that; compat_ptr() is a no-op anyway" breaks.  IOW,
> > s390 is the reason for having compat_ptr_ioctl() in the first place;
> > that thing works on all biarch architectures, as long as all stuff
> > handled by ->ioctl() takes pointer to arch-independent object as
> > argument.  IOW,
> >         argument ignored => OK
> >         any arithmetical type => no go, compat_ptr() would bugger it
> >         pointer to int => OK
> >         pointer to string => OK
> >         pointer to u64 => OK
> >         pointer to struct {u64 addr; char s[11];} => OK
> 
> To be extra pedantic, the 'struct {u64 addr; char s[11];} '
> case is also broken on x86, because sizeof (obj) is smaller
> on i386, even though the location of the members are
> the same. i.e. you can copy_from_user() this, but not
> copy_to_user(), which overwrites 4 bytes after the end of
> the 20-byte user structure.

D'oh!  FWIW, it might be worth putting into Documentation/ somewhere;
basically, what is and what isn't biarch-neutral.

Or arch-neutral, for that matter - it's very close.  The only real
exception, IIRC, is an extra twist on m68k, where int behaves
like x86 long long - its alignment is only half its size, so
sizeof(struct {char c; int x;}) is 6, not 8 as everywhere
else.  Irrelevant for biarch, thankfully (until somebody gets insane
enough to implement 64bit coldfire, kernel port for it *and* biarch
support for m68k binaries on that thing, that is)...
  

On Thu, 2019-04-25 at 17:55 +0200, Arnd Bergmann wrote:
> On Thu, Apr 25, 2019 at 5:35 PM Al Viro <viro@zeniv.linux.org.uk> wrote:
> > 
> > On Thu, Apr 25, 2019 at 12:21:53PM -0300, Mauro Carvalho Chehab wrote:
> > 
> > > If I understand your patch description well, using compat_ptr_ioctl
> > > only works if the driver is not for s390, right?
> > 
> > No; s390 is where "oh, just set ->compat_ioctl same as ->unlocked_ioctl
> > and be done with that; compat_ptr() is a no-op anyway" breaks.  IOW,
> > s390 is the reason for having compat_ptr_ioctl() in the first place;
> > that thing works on all biarch architectures, as long as all stuff
> > handled by ->ioctl() takes pointer to arch-independent object as
> > argument.  IOW,
> >         argument ignored => OK
> >         any arithmetical type => no go, compat_ptr() would bugger it
> >         pointer to int => OK
> >         pointer to string => OK
> >         pointer to u64 => OK
> >         pointer to struct {u64 addr; char s[11];} => OK
> 
> To be extra pedantic, the 'struct {u64 addr; char s[11];} '
> case is also broken on x86, because sizeof (obj) is smaller
> on i386, even though the location of the members are
> the same. i.e. you can copy_from_user() this

Actually, you can't even do that because the struct might sit at the end
of a page and then you'd erroneously fault in this case.

We had this a while ago with struct ifreq, see commit 98406133dd and its
parents.

johannes
  

On Thu, Apr 25, 2019 at 11:25 PM Johannes Berg
<johannes@sipsolutions.net> wrote:
> On Thu, 2019-04-25 at 17:55 +0200, Arnd Bergmann wrote:
> > On Thu, Apr 25, 2019 at 5:35 PM Al Viro <viro@zeniv.linux.org.uk> wrote:
> > >
> > > On Thu, Apr 25, 2019 at 12:21:53PM -0300, Mauro Carvalho Chehab wrote:
> > >
> > > > If I understand your patch description well, using compat_ptr_ioctl
> > > > only works if the driver is not for s390, right?
> > >
> > > No; s390 is where "oh, just set ->compat_ioctl same as ->unlocked_ioctl
> > > and be done with that; compat_ptr() is a no-op anyway" breaks.  IOW,
> > > s390 is the reason for having compat_ptr_ioctl() in the first place;
> > > that thing works on all biarch architectures, as long as all stuff
> > > handled by ->ioctl() takes pointer to arch-independent object as
> > > argument.  IOW,
> > >         argument ignored => OK
> > >         any arithmetical type => no go, compat_ptr() would bugger it
> > >         pointer to int => OK
> > >         pointer to string => OK
> > >         pointer to u64 => OK
> > >         pointer to struct {u64 addr; char s[11];} => OK
> >
> > To be extra pedantic, the 'struct {u64 addr; char s[11];} '
> > case is also broken on x86, because sizeof (obj) is smaller
> > on i386, even though the location of the members are
> > the same. i.e. you can copy_from_user() this
>
> Actually, you can't even do that because the struct might sit at the end
> of a page and then you'd erroneously fault in this case.
>
> We had this a while ago with struct ifreq, see commit 98406133dd and its
> parents.

Yes, you are right. Very rare to hit with real-life code, but easily
reproduced by intentionally hitting it and clearly a bug.

As the saying goes

  | the difference between "always works" and "almost always works"
  | is called data corruption

here the difference is an -EFAULT.

      Arnd