[PATCH] x86: ia32_setup_rt_frame(): propagate __user annotations properly

[PATCH] x86: ia32_setup_rt_frame(): propagate __user annotations properly

[Lukas Bulwahn]

Commit 57d563c82925 ("x86: ia32_setup_rt_frame(): consolidate uaccess
areas") dropped a __user annotation in a cast when refactoring __put_user()
to unsafe_put_user().

Hence, since then, sparse warns in arch/x86/ia32/ia32_signal.c:350:9:

  warning: cast removes address space '__user' of expression
  warning: incorrect type in argument 1 (different address spaces)
    expected void const volatile [noderef] __user *ptr
    got unsigned long long [usertype] *

Add the __user annotation to restore the propagation of address spaces.

Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
---
applies cleanly on current master (v5.10-rc7) and next-20201204

Thomas, Ingo, Boris, please pick this minor non-urgent clean-up patch.

 arch/x86/ia32/ia32_signal.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c
index 81cf22398cd1..5e3d9b7fd5fb 100644
--- a/arch/x86/ia32/ia32_signal.c
+++ b/arch/x86/ia32/ia32_signal.c
@@ -347,7 +347,7 @@ int ia32_setup_rt_frame(int sig, struct ksignal *ksig,
 	 */
 	unsafe_put_user(*((u64 *)&code), (u64 __user *)frame->retcode, Efault);
 	unsafe_put_sigcontext32(&frame->uc.uc_mcontext, fp, regs, set, Efault);
-	unsafe_put_user(*(__u64 *)set, (__u64 *)&frame->uc.uc_sigmask, Efault);
+	unsafe_put_user(*(__u64 *)set, (__u64 __user *)&frame->uc.uc_sigmask, Efault);
 	user_access_end();
 
 	if (__copy_siginfo_to_user32(&frame->info, &ksig->info))
-- 
2.17.1

[PATCH] x86: ia32_setup_rt_frame(): propagate __user annotations properly

[Lukas Bulwahn]

Commit 57d563c82925 ("x86: ia32_setup_rt_frame(): consolidate uaccess
areas") dropped a __user annotation in a cast when refactoring __put_user()
to unsafe_put_user().

Hence, since then, sparse warns in arch/x86/ia32/ia32_signal.c:350:9:

  warning: cast removes address space '__user' of expression
  warning: incorrect type in argument 1 (different address spaces)
    expected void const volatile [noderef] __user *ptr
    got unsigned long long [usertype] *

Add the __user annotation to restore the propagation of address spaces.

Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
---
applies cleanly on current master (v5.10-rc7) and next-20201204

Thomas, Ingo, Boris, please pick this minor non-urgent clean-up patch.

 arch/x86/ia32/ia32_signal.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c
index 81cf22398cd1..5e3d9b7fd5fb 100644
--- a/arch/x86/ia32/ia32_signal.c
+++ b/arch/x86/ia32/ia32_signal.c
@@ -347,7 +347,7 @@ int ia32_setup_rt_frame(int sig, struct ksignal *ksig,
 	 */
 	unsafe_put_user(*((u64 *)&code), (u64 __user *)frame->retcode, Efault);
 	unsafe_put_sigcontext32(&frame->uc.uc_mcontext, fp, regs, set, Efault);
-	unsafe_put_user(*(__u64 *)set, (__u64 *)&frame->uc.uc_sigmask, Efault);
+	unsafe_put_user(*(__u64 *)set, (__u64 __user *)&frame->uc.uc_sigmask, Efault);
 	user_access_end();
 
 	if (__copy_siginfo_to_user32(&frame->info, &ksig->info))
-- 
2.17.1

Re: [PATCH] x86: ia32_setup_rt_frame(): propagate __user annotations properly

[Borislav Petkov]

On Mon, Dec 07, 2020 at 01:41:41PM +0100, Lukas Bulwahn wrote:
> Thomas, Ingo, Boris, please pick this minor non-urgent clean-up patch.

Why?

Isn't it obvious that when you send a patch to us, the final goal is for
it to be applied. Eventually.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: [PATCH] x86: ia32_setup_rt_frame(): propagate __user annotations properly

[Borislav Petkov]

On Mon, Dec 07, 2020 at 01:41:41PM +0100, Lukas Bulwahn wrote:
> Thomas, Ingo, Boris, please pick this minor non-urgent clean-up patch.

Why?

Isn't it obvious that when you send a patch to us, the final goal is for
it to be applied. Eventually.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: [PATCH] x86: ia32_setup_rt_frame(): propagate __user annotations properly

[Lukas Bulwahn]

On Fri, Dec 11, 2020 at 7:47 PM Borislav Petkov <bp@alien8.de> wrote:
>
> On Mon, Dec 07, 2020 at 01:41:41PM +0100, Lukas Bulwahn wrote:
> > Thomas, Ingo, Boris, please pick this minor non-urgent clean-up patch.
>
> Why?
>
> Isn't it obvious that when you send a patch to us, the final goal is for
> it to be applied. Eventually.
>

Yes, agree. Other maintainers noted that I should point out that the
patch is only a minor clean-up and it is not urgent to be considered.

So, I add this remark to make clear that it is not top priority to
apply just that the maintainers know.

You will sure review it eventually, and hopefully accept it then.

If that comment disturbs you, please ignore it.

Lukas

Re: [PATCH] x86: ia32_setup_rt_frame(): propagate __user annotations properly

[Lukas Bulwahn]

On Fri, Dec 11, 2020 at 7:47 PM Borislav Petkov <bp@alien8.de> wrote:
>
> On Mon, Dec 07, 2020 at 01:41:41PM +0100, Lukas Bulwahn wrote:
> > Thomas, Ingo, Boris, please pick this minor non-urgent clean-up patch.
>
> Why?
>
> Isn't it obvious that when you send a patch to us, the final goal is for
> it to be applied. Eventually.
>

Yes, agree. Other maintainers noted that I should point out that the
patch is only a minor clean-up and it is not urgent to be considered.

So, I add this remark to make clear that it is not top priority to
apply just that the maintainers know.

You will sure review it eventually, and hopefully accept it then.

If that comment disturbs you, please ignore it.

Lukas

Re: [PATCH] x86: ia32_setup_rt_frame(): propagate __user annotations properly

[Borislav Petkov]

On Fri, Dec 11, 2020 at 07:55:50PM +0100, Lukas Bulwahn wrote:
> Yes, agree. Other maintainers noted that I should point out that the
> patch is only a minor clean-up and it is not urgent to be considered.
> 
> So, I add this remark to make clear that it is not top priority to
> apply just that the maintainers know.
> 
> You will sure review it eventually, and hopefully accept it then.

Well, after I review it, I probably should know what the priority is,
right?

> If that comment disturbs you, please ignore it.

I was just wondering why you're writing something which is not really
needed. Notes under "---" are usually used to clarify aspects of the
patch handling, versioning, etc.

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: [PATCH] x86: ia32_setup_rt_frame(): propagate __user annotations properly

[Borislav Petkov]

On Fri, Dec 11, 2020 at 07:55:50PM +0100, Lukas Bulwahn wrote:
> Yes, agree. Other maintainers noted that I should point out that the
> patch is only a minor clean-up and it is not urgent to be considered.
> 
> So, I add this remark to make clear that it is not top priority to
> apply just that the maintainers know.
> 
> You will sure review it eventually, and hopefully accept it then.

Well, after I review it, I probably should know what the priority is,
right?

> If that comment disturbs you, please ignore it.

I was just wondering why you're writing something which is not really
needed. Notes under "---" are usually used to clarify aspects of the
patch handling, versioning, etc.

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

[tip: x86/cleanups] x86/ia32_signal: Propagate __user annotation properly

[tip-bot2 for Lukas Bulwahn]

The following commit has been merged into the x86/cleanups branch of tip:

Commit-ID:     9a02fd8b19247e80e2354a227b6e2392e8fae78a
Gitweb:        https://git.kernel.org/tip/9a02fd8b19247e80e2354a227b6e2392e8fae78a
Author:        Lukas Bulwahn <lukas.bulwahn@gmail.com>
AuthorDate:    Mon, 07 Dec 2020 13:41:41 +01:00
Committer:     Borislav Petkov <bp@suse.de>
CommitterDate: Fri, 11 Dec 2020 19:44:31 +01:00

x86/ia32_signal: Propagate __user annotation properly

Commit

  57d563c82925 ("x86: ia32_setup_rt_frame(): consolidate uaccess areas")

dropped a __user annotation in a cast when refactoring __put_user() to
unsafe_put_user().

Hence, since then, sparse warns in arch/x86/ia32/ia32_signal.c:350:9:

  warning: cast removes address space '__user' of expression
  warning: incorrect type in argument 1 (different address spaces)
    expected void const volatile [noderef] __user *ptr
    got unsigned long long [usertype] *

Add the __user annotation to restore the propagation of address spaces.

Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20201207124141.21859-1-lukas.bulwahn@gmail.com
---
 arch/x86/ia32/ia32_signal.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c
index 81cf223..5e3d9b7 100644
--- a/arch/x86/ia32/ia32_signal.c
+++ b/arch/x86/ia32/ia32_signal.c
@@ -347,7 +347,7 @@ int ia32_setup_rt_frame(int sig, struct ksignal *ksig,
 	 */
 	unsafe_put_user(*((u64 *)&code), (u64 __user *)frame->retcode, Efault);
 	unsafe_put_sigcontext32(&frame->uc.uc_mcontext, fp, regs, set, Efault);
-	unsafe_put_user(*(__u64 *)set, (__u64 *)&frame->uc.uc_sigmask, Efault);
+	unsafe_put_user(*(__u64 *)set, (__u64 __user *)&frame->uc.uc_sigmask, Efault);
 	user_access_end();
 
 	if (__copy_siginfo_to_user32(&frame->info, &ksig->info))