SED Opal Fixups
Subsystems (3)
So we have a few patches here, they're pretty small. First patch changes the sed-opal ioctl function parameters to take a void __user* instead of an unsigned long, this required a small cast in the nvme driver. Patch 2 is a UAPI fixup for the IOW to make an ioctl
the right size. Patch 3 fixes a compiliation error when building using KSAN due to the stack frame being too large. And lastly we move the Maintainers list from nvme to linux-block.
PATCH v5 1/4 block: sed-opal: change ioctl to take user pointer instead of unsinged long
--- block/sed-opal.c | 6 ++++-- drivers/nvme/host/core.c | 3 ++- include/linux/sed-opal.h | 4 ++-- 3 files changed, 8 insertions(+), 5 deletions(-)
block/sed-opal.c +4 -2
@@ -2344,9 +2344,11 @@ bool opal_unlock_from_suspend(struct opal_dev *dev) }EXPORT_SYMBOL (opal_unlock_from_suspend); -int sed_ioctl (struct opal_dev *dev,unsigned int cmd,unsigned long ptr) +int sed_ioctl (struct opal_dev *dev,unsigned int cmd,void __user *arg) { -void __user *arg = (void __user *)ptr; +void *ioctl_ptr; +int ret = -ENOTTY ; +unsigned int cmd_size =_IOC_SIZE (cmd);if (!capable (CAP_SYS_ADMIN ))return -EACCES ;
drivers/nvme/host/core.c +2 -1
@@ -811,7 +811,8 @@ static int nvme_ioctl(struct block_device *bdev, fmode_t mode,return nvme_nvm_ioctl (ns, cmd, arg);#endif if (is_sed_ioctl (cmd)) -return sed_ioctl (&ns->ctrl->opal_dev, cmd, arg); +return sed_ioctl (&ns->ctrl->opal_dev, cmd, + (void __user *) arg);return -ENOTTY ; } }
include/linux/sed-opal.h +2 -2
@@ -132,7 +132,7 @@ struct opal_dev {#ifdef CONFIG_BLK_SED_OPAL bool opal_unlock_from_suspend (struct opal_dev *dev);void init_opal_dev (struct opal_dev *opal_dev,sec_send_recv *send_recv); -int sed_ioctl (struct opal_dev *dev,unsigned int cmd,unsigned long ptr); +int sed_ioctl (struct opal_dev *dev,unsigned int cmd,void __user *ioctl_ptr);static inline bool is_sed_ioctl (unsigned int cmd) {
@@ -160,7 +160,7 @@ static inline bool is_sed_ioctl(unsigned int cmd) }static inline int sed_ioctl (struct opal_dev *dev,unsigned int cmd, -unsigned long ptr) +void __user *ioctl_ptr) {return 0 ; }
Signature · 7 lines
2.7.4 _______________________________________________ Linux-nvme mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-nvme
Re: [PATCH V5 1/4] block: sed-opal: change ioctl to take user pointer instead of unsinged long
Quoted 12 lines · Scott Bauer
Signed-off-by Scott Bauer <scott.bauer at intel.com>--- block/sed-opal.c | 6 ++++-- drivers/nvme/host/core.c | 3 ++- include/linux/sed-opal.h | 4 ++-- 3 files changed, 8 insertions(+), 5 deletions(-)block/sed-opal.c +0
-0
block/sed-opal.c bool opal_unlock_from_suspend(struct opal_dev *dev) · line 2351
Quoted 10 lines
block/sed-opal.c +4
-2@@ -2344,9 +2344,11 @@ bool opal_unlock_from_suspend(struct opal_dev *dev) }EXPORT_SYMBOL (opal_unlock_from_suspend); -int sed_ioctl (struct opal_dev *dev,unsigned int cmd,unsigned long ptr) +int sed_ioctl (struct opal_dev *dev,unsigned int cmd,void __user *arg) { -void __user *arg = (void __user *)ptr; +void *ioctl_ptr; +int ret = -ENOTTY ; +unsigned int cmd_size =_IOC_SIZE (cmd);
ugh, I apparently messed up my rebase these should be in patch 2 or maybe I should sqash p1 and p2 together.
Re: [PATCH V5 1/4] block: sed-opal: change ioctl to take user pointer instead of unsinged long
Quoted 16 lines · Scott Bauer
Quoted 15 lines · Scott Bauer
Signed-off-by Scott Bauer <scott.bauer at intel.com>--- block/sed-opal.c | 6 ++++-- drivers/nvme/host/core.c | 3 ++- include/linux/sed-opal.h | 4 ++-- 3 files changed, 8 insertions(+), 5 deletions(-)block/sed-opal.c +0
-0@@ -2344,9 +2344,11 @@ bool opal_unlock_from_suspend(struct opal_dev *dev) }EXPORT_SYMBOL (opal_unlock_from_suspend);
Quoted 10 lines
block/sed-opal.c bool opal_unlock_from_suspend(struct opal_dev *dev) · line 2351
Quoted 7 lines
block/sed-opal.c +4
-2-int sed_ioctl (struct opal_dev *dev,unsigned int cmd,unsigned long ptr) +int sed_ioctl (struct opal_dev *dev,unsigned int cmd,void __user *arg) { -void __user *arg = (void __user *)ptr; +void *ioctl_ptr; +int ret = -ENOTTY ; +unsigned int cmd_size =_IOC_SIZE (cmd);ugh, I apparently messed up my rebase these should be in patch 2 or maybe I should sqash p1 and p2 together.
I guess it should be 1 and 3. No real opinipon on that, but can you chose a simpler and more fitting name than ioctl_ptr? I'd suggest 'p'.
PATCH v5 2/4 uapi: sed-opal fix IOW for activate lsp to use correct struct
In reply to Scott Bauerthe IOW for the IOC_OPAL_ACTIVATE_LSP took the wrong strcure which would give us the wrong size when using _IOC_SIZE, switch it to the right structure.
--- include/uapi/linux/sed-opal.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/uapi/linux/sed-opal.h +1 -1
@@ -106,7 +106,7 @@ struct opal_mbr_data {#define IOC_OPAL_SAVE _IOW('p', 220, struct opal_lock_unlock)#define IOC_OPAL_LOCK_UNLOCK _IOW('p', 221, struct opal_lock_unlock)#define IOC_OPAL_TAKE_OWNERSHIP _IOW('p', 222, struct opal_key) -#define IOC_OPAL_ACTIVATE_LSP _IOW('p', 223, struct opal_key) +#define IOC_OPAL_ACTIVATE_LSP _IOW('p', 223, struct opal_lr_act)#define IOC_OPAL_SET_PW _IOW('p', 224, struct opal_new_pw)#define IOC_OPAL_ACTIVATE_USR _IOW('p', 225, struct opal_session_info)#define IOC_OPAL_REVERT_TPR _IOW('p', 226, struct opal_key)
Signature · 1 line
2.7.4
Re: [PATCH V5 2/4] uapi: sed-opal fix IOW for activate lsp to use correct struct
Let's get this one in ASAP while waiting for a respin of the KASAN fix.
PATCH v5 3/4 Move stack parameters for sed_ioctl to prevent oversized stack with CONFIG_KASAN
In reply to Scott BauerWhen CONFIG_KASAN is enabled, compilation fails:
block/sed-opal.c: In function 'sed_ioctl':block/sed-opal.c:2447:1: error: the frame size of 2256 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
Moved all the ioctl structures off the stack and dynamically activate using _IOC_SIZE()
--- block/sed-opal.c | 130 +++++++++++++++++++------------------------------------ 1 file changed, 45 insertions(+), 85 deletions(-)
block/sed-opal.c +45 -85
@@ -2348,7 +2348,6 @@ int sed_ioctl(struct opal_dev *dev, unsigned int cmd, void __user *arg) {void *ioctl_ptr;int ret = -ENOTTY ; -unsigned int cmd_size =_IOC_SIZE (cmd);if (!capable (CAP_SYS_ADMIN ))return -EACCES ;
@@ -2357,94 +2356,55 @@ int sed_ioctl(struct opal_dev *dev, unsigned int cmd, void __user *arg)return -ENOTSUPP ; } -switch (cmd) { -case IOC_OPAL_SAVE : { -struct opal_lock_unlock lk_unlk; - -if (copy_from_user (&lk_unlk, arg,sizeof (lk_unlk))) -return -EFAULT ; -return opal_save (dev, &lk_unlk); - } -case IOC_OPAL_LOCK_UNLOCK : { -struct opal_lock_unlock lk_unlk; - -if (copy_from_user (&lk_unlk, arg,sizeof (lk_unlk))) -return -EFAULT ; -return opal_lock_unlock (dev, &lk_unlk); - } -case IOC_OPAL_TAKE_OWNERSHIP : { -struct opal_key opal_key; - -if (copy_from_user (&opal_key, arg,sizeof (opal_key))) -return -EFAULT ; -return opal_take_ownership (dev, &opal_key); - } -case IOC_OPAL_ACTIVATE_LSP : { -struct opal_lr_act opal_lr_act; - -if (copy_from_user (&opal_lr_act, arg,sizeof (opal_lr_act))) -return -EFAULT ; -return opal_activate_lsp (dev, &opal_lr_act); - } -case IOC_OPAL_SET_PW : { -struct opal_new_pw opal_pw; - -if (copy_from_user (&opal_pw, arg,sizeof (opal_pw))) -return -EFAULT ; -return opal_set_new_pw (dev, &opal_pw); - } -case IOC_OPAL_ACTIVATE_USR : { -struct opal_session_info session; - -if (copy_from_user (&session, arg,sizeof (session))) -return -EFAULT ; -return opal_activate_user (dev, &session); - } -case IOC_OPAL_REVERT_TPR : { -struct opal_key opal_key; - -if (copy_from_user (&opal_key, arg,sizeof (opal_key))) -return -EFAULT ; -return opal_reverttper (dev, &opal_key); - } -case IOC_OPAL_LR_SETUP : { -struct opal_user_lr_setup lrs; - -if (copy_from_user (&lrs, arg,sizeof (lrs))) -return -EFAULT ; -return opal_setup_locking_range (dev, &lrs); - } -case IOC_OPAL_ADD_USR_TO_LR : { -struct opal_lock_unlock lk_unlk; - -if (copy_from_user (&lk_unlk, arg,sizeof (lk_unlk))) -return -EFAULT ; -return opal_add_user_to_lr (dev, &lk_unlk); - } -case IOC_OPAL_ENABLE_DISABLE_MBR : { -struct opal_mbr_data mbr; - -if (copy_from_user (&mbr, arg,sizeof (mbr))) -return -EFAULT ; -return opal_enable_disable_shadow_mbr (dev, &mbr); - } -case IOC_OPAL_ERASE_LR : { -struct opal_session_info session; - -if (copy_from_user (&session, arg,sizeof (session))) -return -EFAULT ; -return opal_erase_locking_range (dev, &session); + ioctl_ptr =memdup_user (arg,_IOC_SIZE (cmd)); +if (IS_ERR_OR_NULL (ioctl_ptr)) { + ret =PTR_ERR (ioctl_ptr); + goto out; } -case IOC_OPAL_SECURE_ERASE_LR : { -struct opal_session_info session; -if (copy_from_user (&session, arg,sizeof (session))) -return -EFAULT ; -return opal_secure_erase_locking_range (dev, &session); - } +switch (cmd) { +case IOC_OPAL_SAVE : + ret =opal_save (dev, ioctl_ptr); +break ; +case IOC_OPAL_LOCK_UNLOCK : + ret =opal_lock_unlock (dev, ioctl_ptr); +break ; +case IOC_OPAL_TAKE_OWNERSHIP : + ret =opal_take_ownership (dev, ioctl_ptr); +break ; +case IOC_OPAL_ACTIVATE_LSP : + ret =opal_activate_lsp (dev, ioctl_ptr); +break ; +case IOC_OPAL_SET_PW : + ret =opal_set_new_pw (dev, ioctl_ptr); +break ; +case IOC_OPAL_ACTIVATE_USR : + ret =opal_activate_user (dev, ioctl_ptr); +break ; +case IOC_OPAL_REVERT_TPR : + ret =opal_reverttper (dev, ioctl_ptr); +break ; +case IOC_OPAL_LR_SETUP : + ret =opal_setup_locking_range (dev, ioctl_ptr); +break ; +case IOC_OPAL_ADD_USR_TO_LR : + ret =opal_add_user_to_lr (dev, ioctl_ptr); +break ; +case IOC_OPAL_ENABLE_DISABLE_MBR : + ret =opal_enable_disable_shadow_mbr (dev, ioctl_ptr); +break ; +case IOC_OPAL_ERASE_LR : + ret =opal_erase_locking_range (dev, ioctl_ptr); +break ; +case IOC_OPAL_SECURE_ERASE_LR : + ret =opal_secure_erase_locking_range (dev, ioctl_ptr); +break ;default :pr_warn ("No such Opal Ioctl %u\n" , cmd); } -return -ENOTTY ; + + out: +kfree (ioctl_ptr); +return ret; }EXPORT_SYMBOL_GPL (sed_ioctl);
Signature · 1 line
2.7.4
RE: [PATCH V5 3/4] Move stack parameters for sed_ioctl to prevent oversized stack with CONFIG_KASAN
From: Scott Bauer Sent: 13 February 2017 16:11
Quoted 8 lines
When
CONFIG_KASANis enabled, compilation fails:
block/sed-opal.c: In function 'sed_ioctl':block/sed-opal.c:2447:1: error: the frame size of 2256 bytes is larger than 2048 bytes [-Werror=frame- larger-than=]Moved all the ioctl structures off the stack and dynamically activate using
_IOC_SIZE()
Think I'd not that this simplifies the code considerably. AFAICT CONFIG_KASAN is a just brainf*ck.
It at least needs annotation that copy_from_user() has properties similar to memset().
So if the size matches that of the type then no guard space (etc) is required.
...
Quoted 4 lines
block/sed-opal.c +4
-0+ ioctl_ptr =memdup_user (arg,_IOC_SIZE (cmd)); +if (IS_ERR_OR_NULL (ioctl_ptr)) { + ret =PTR_ERR (ioctl_ptr); + goto out;
...
block/sed-opal.c int sed_ioctl(struct opal_dev *dev, unsigned int cmd, void __user *arg) · line 2409
Quoted 4 lines
+ out:
block/sed-opal.c +2
-0+kfree (ioctl_ptr); +return ret; }
That error path doesn't look quite right to me.
David
Re: [PATCH V5 3/4] Move stack parameters for sed_ioctl to prevent oversized stack with CONFIG_KASAN
Quoted 10 lines · David Laight
From: Scott Bauer Sent: 13 February 2017 16:11
Quoted 8 lines
When
CONFIG_KASANis enabled, compilation fails:
block/sed-opal.c: In function 'sed_ioctl':block/sed-opal.c:2447:1: error: the frame size of 2256 bytes is larger than 2048 bytes [-Werror=frame- larger-than=]Moved all the ioctl structures off the stack and dynamically activate using
_IOC_SIZE()
Quoted 7 lines
Think I'd not that this simplifies the code considerably. AFAICT
CONFIG_KASANis a just brainf*ck.
It at least needs annotation thatcopy_from_user()has properties similar tomemset().
So if the size matches that of the type then no guard space (etc) is required.
I don't really follow what you're saying. Do you want me to just include that the patch cleans up the ioctl path a bit. I need to include the KASAN part since there was build breakage and the series does fix it even if it simplifies the path as well. As for the memset part, we never copy back to userland so there's no chance of data leakage which is what it seems you're hinting at.
Quoted 15 lines
...
Quoted 4 lines
block/sed-opal.c +4
-0+ ioctl_ptr =memdup_user (arg,_IOC_SIZE (cmd)); +if (IS_ERR_OR_NULL (ioctl_ptr)) { + ret =PTR_ERR (ioctl_ptr); + goto out;...
block/sed-opal.c int sed_ioctl(struct opal_dev *dev, unsigned int cmd, void __user *arg) · line 2409
Quoted 4 lines
+ out:
block/sed-opal.c +2
-0+kfree (ioctl_ptr); +return ret; }That error path doesn't look quite right to me.
Quoted 2 lines
David
good god, this is a mess this morning. Thanks for the catch, I'll review these more aggressively before sending out.
Re: [PATCH V5 3/4] Move stack parameters for sed_ioctl to prevent oversized stack with CONFIG_KASAN
Quoted 11 lines · Scott Bauer
Quoted 10 lines · David Laight
From: Scott Bauer Sent: 13 February 2017 16:11
Quoted 8 lines
When
CONFIG_KASANis enabled, compilation fails:
block/sed-opal.c: In function 'sed_ioctl':block/sed-opal.c:2447:1: error: the frame size of 2256 bytes is larger than 2048 bytes [-Werror=frame- larger-than=]Moved all the ioctl structures off the stack and dynamically activate using
_IOC_SIZE()
Quoted 6 lines
Quoted 6 lines
Think I'd not that this simplifies the code considerably. AFAICT
CONFIG_KASANis a just brainf*ck.
It at least needs annotation thatcopy_from_user()has properties similar tomemset().
So if the size matches that of the type then no guard space (etc) is required.
I think it still would, as the pointer to the local variable gets passed through dev->func_data[].
Quoted 13 lines
Quoted 13 lines
...
Quoted 4 lines
block/sed-opal.c +4
-0+ ioctl_ptr =memdup_user (arg,_IOC_SIZE (cmd)); +if (IS_ERR_OR_NULL (ioctl_ptr)) { + ret =PTR_ERR (ioctl_ptr); + goto out;...
Quoted 4 lines
+ out:
block/sed-opal.c +2
-0+kfree (ioctl_ptr); +return ret; }
Quoted 7 lines
Quoted 4 lines
That error path doesn't look quite right to me.
David
good god, this is a mess this morning. Thanks for the catch, I'll review these more aggressively before sending out.
memdup_user() never returns NULL, and generally speaking any use of IS_ERR_OR_NULL() indicates that there is something wrong with the interface, so aside from passing the right pointer (or NULL) into kfree() I think using IS_ERR() is the correct solution.
Arnd
RE: [PATCH V5 3/4] Move stack parameters for sed_ioctl to prevent oversized stack with CONFIG_KASAN
From: Arnd Bergmann
Quoted 1 line
Sent: 13 February 2017 17:02
...
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 4 lines
block/sed-opal.c +4
-0+ ioctl_ptr =memdup_user (arg,_IOC_SIZE (cmd)); +if (IS_ERR_OR_NULL (ioctl_ptr)) { + ret =PTR_ERR (ioctl_ptr); + goto out;...
Quoted 3 lines
+ out:
block/sed-opal.c +2
-0+kfree (ioctl_ptr); +return ret;
...
Quoted 1 line
Quoted 1 line
Quoted 1 line
That error path doesn't look quite right to me.
...
Quoted 7 lines
Quoted 2 lines
good god, this is a mess this morning. Thanks for the catch, I'll review these more aggressively before sending out.
memdup_user()never returns NULL, and generally speaking any use ofIS_ERR_OR_NULL()indicates that there is something wrong with the interface, so aside from passing the right pointer (or NULL) intokfree()I think usingIS_ERR()is the correct solution.
You missed the problem entirely.
Code needs to be:
if (IS_ERR(ioctl_ptr)) return PTR_ERR(ioctl_ptr); David
_______________________________________________
Linux-nvme mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-nvme
Re: [PATCH V5 3/4] Move stack parameters for sed_ioctl to prevent oversized stack with CONFIG_KASAN
Quoted 22 lines · David Laight
From: Arnd Bergmann
Quoted 1 line
Sent: 13 February 2017 17:02
...
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 4 lines
block/sed-opal.c +4
-0+ ioctl_ptr =memdup_user (arg,_IOC_SIZE (cmd)); +if (IS_ERR_OR_NULL (ioctl_ptr)) { + ret =PTR_ERR (ioctl_ptr); + goto out;...
Quoted 3 lines
+ out:
block/sed-opal.c +2
-0+kfree (ioctl_ptr); +return ret;...
Quoted 1 line
Quoted 1 line
Quoted 1 line
That error path doesn't look quite right to me.
...
Quoted 7 lines
Quoted 2 lines
good god, this is a mess this morning. Thanks for the catch, I'll review these more aggressively before sending out.
memdup_user()never returns NULL, and generally speaking any use ofIS_ERR_OR_NULL()indicates that there is something wrong with the interface, so aside from passing the right pointer (or NULL) intokfree()I think usingIS_ERR()is the correct solution.
Quoted 4 lines
You missed the problem entirely.
Code needs to be:if (IS_ERR(ioctl_ptr)) return PTR_ERR(ioctl_ptr);
Sorry if that wasn't clear, I expected the part about the kfree(errptr) to be obvious but was trying to avoid having Scott go through another revision for removing the IS_ERR_OR_NULL() after fixing the first problem.
Arnd
PATCH v5 4/4 Maintainers: Modify SED list from nvme to block
In reply to Scott Bauer--- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
MAINTAINERS +1 -1
@@ -11094,7 +11094,7 @@ SECURE ENCRYPTING DEVICE (SED) OPAL DRIVER M: Scott Bauer <scott.bauer at intel.com> M: Jonathan Derrick <jonathan.derrick at intel.com> M: Rafael Antognolli <rafael.antognolli at intel.com> -L: linux-nvme at lists.infradead.org +L: linux-block at vger.kernel.org S: Supported F: block/sed* F: block/opal_proto.h
Signature · 1 line
2.7.4
Re: [PATCH V5 4/4] Maintainers: Modify SED list from nvme to block
Let's get it in ASAP as well.
_______________________________________________
Linux-nvme mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-nvme