On 12/14/2017 01:17 AM, Ram Pai wrote:
On Wed, Dec 13, 2017 at 04:40:11PM +0100, Florian Weimer wrote:
On 12/13/2017 04:22 PM, Dave Hansen wrote:
On 12/13/2017 07:08 AM, Florian Weimer wrote:
Okay, this model is really quite different from x86. Is there a
good reason for the difference?
Yes, both implementations are simple and take the "natural" behavior.
x86 changes XSAVE-controlled register values on entering a signal, so we
let them be changed (including PKRU). POWER hardware does not do this
to its PKRU-equivalent, so we do not force it to.
Whuy? Is there a technical reason not have fully-aligned behavior?
Can POWER at least implement the original PKEY_ALLOC_SETSIGNAL
semantics (reset the access rights for certain keys before switching
to the signal handler) in a reasonably efficient manner?
This can be done on POWER. I can also change the behavior on POWER
to exactly match x86; i.e reset the value to init value before
calling the signal handler.
Maybe we can implement a compromise?
Assuming I got the attached patch right, it implements PKRU inheritance
in signal handlers, similar to what you intend to implement for POWER.
It still restores the PKRU register value upon regular exit from the
signal handler, which I think is something we should keep.
I think we still should add a flag, so that applications can easily
determine if a kernel has this patch. Setting up a signal handler,
sending the signal, and thus checking for inheritance is a bit involved,
and we'd have to do this in the dynamic linker before we can use pkeys
to harden lazy binding. The flag could just be a no-op, apart from the
lack of an EINVAL failure if it is specified.
But I think, we should clearly define the default behavior, the behavior
when no flag is specified. Applications tend to rely on default behavior
and expect the most intuitive behavior to be the default behavior.
Because this feature already shipped on x86, we already have the
unspecified signal handler behavior in the wild, and if applications
need the new, clearly defined semantics, there has to be a way to detect
that the kernel makes this guarantee.
I tend to think; keeping my biases aside, that the most intuitive
behavior is to preserve access/write permissions of any key, i.e not
reset to the init value. If the application has set the permissions of
a key to some value, it would'nt expect anyone to change them,
irrespective of which context it is in.
Sure, it also fixes the siglongjmp issue:
https://sourceware.org/bugzilla/show_bug.cgi?id=22396
If we do not reset the PKRU register on x86 anymore, a non-pkeys-aware
signal handler will not clobber it.
Thanks,
Florian
commit 148de78dc299b87ea3dcf13d73c92431f94643a0
Author: Florian Weimer <fweimer@xxxxxxxxxx>
Date: Thu Dec 14 11:12:02 2017 -0500
x86 pkeys: Use PKRU register of interrupted thread in signal handlers
pkeys support for IBM POWER intends to inherited the access rights of
the current thread in signal handlers. The advantage is that this
preserves access to memory regions associated with non-default keys,
enabling additional usage scenarios for memory protection keys which
currently do not work due to the unconditional reset to the
(configurable) default key in signal handlers.
This change does not affect the init_pkru optimization because if the
thread's PKRU register is zero due to the init_pkru setting, it will
remain zero in the signal handler through inheritance.
Signed-off-by: Florian Weimer <fweimer@xxxxxxxxxx>
diff --git a/arch/x86/include/asm/fpu/internal.h b/arch/x86/include/asm/fpu/internal.h
index a38bf5a..a87e99f7 100644
--- a/arch/x86/include/asm/fpu/internal.h
+++ b/arch/x86/include/asm/fpu/internal.h
@@ -33,6 +33,7 @@
extern void fpu__drop(struct fpu *fpu);
extern int fpu__copy(struct fpu *dst_fpu, struct fpu *src_fpu);
extern void fpu__clear(struct fpu *fpu);
+extern void fpu__clear_signal(struct fpu *fpu);
extern int fpu__exception_code(struct fpu *fpu, int trap_nr);
extern int dump_fpu(struct pt_regs *ptregs, struct user_i387_struct *fpstate);
diff --git a/arch/x86/kernel/fpu/core.c b/arch/x86/kernel/fpu/core.c
index f92a659..a3b3048 100644
--- a/arch/x86/kernel/fpu/core.c
+++ b/arch/x86/kernel/fpu/core.c
@@ -370,21 +370,16 @@ static inline void copy_init_fpstate_to_fpregs(void)
copy_kernel_to_fxregs(&init_fpstate.fxsave);
else
copy_kernel_to_fregs(&init_fpstate.fsave);
-
- if (boot_cpu_has(X86_FEATURE_OSPKE))
- copy_init_pkru_to_fpregs();
}
-/*
- * Clear the FPU state back to init state.
- *
- * Called by sys_execve(), by the signal handler code and by various
- * error paths.
- */
-void fpu__clear(struct fpu *fpu)
+static void __fpu_clear(struct fpu *fpu, bool for_signal)
{
+ u32 pkru;
+
WARN_ON_FPU(fpu != ¤t->thread.fpu); /* Almost certainly an anomaly */
+ if (for_signal)
+ pkru = read_pkru();
fpu__drop(fpu);
/*
@@ -395,11 +390,43 @@ void fpu__clear(struct fpu *fpu)
fpu__initialize(fpu);
user_fpu_begin();
copy_init_fpstate_to_fpregs();
+ if (boot_cpu_has(X86_FEATURE_OSPKE)) {
+ /* A signal handler inherits the original PKRU
+ * value of the interrupted thread.
+ */
+ if (for_signal)
+ __write_pkru(pkru);
+ else
+ copy_init_pkru_to_fpregs();
+ }
preempt_enable();
}
}
/*
+ * Clear the FPU state back to init state.
+ *
+ * Called by sys_execve(), the signal handler return code, and by
+ * various error paths.
+ */
+void fpu__clear(struct fpu *fpu)
+{
+ return __fpu_clear(fpu, false);
+}
+
+/*
+ * Prepare the FPU for invoking a signal handler.
+ *
+ * This is like fpu__clear(), but some CPU registers are inherited
+ * from the current thread and not restored to their initial values,
+ * to match behavior on other architectures.
+ */
+void fpu__clear_signal(struct fpu *fpu)
+{
+ return __fpu_clear(fpu, true);
+}
+
+/*
* x87 math exception handling:
*/
diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c
index b9e00e8..4263f18 100644
--- a/arch/x86/kernel/signal.c
+++ b/arch/x86/kernel/signal.c
@@ -757,7 +757,7 @@ static inline int is_x32_frame(struct ksignal *ksig)
* Ensure the signal handler starts with the new fpu state.
*/
if (fpu->initialized)
- fpu__clear(fpu);
+ fpu__clear_signal(fpu);
}
signal_setup_done(failed, ksig, stepping);
}
