linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH V2 1/2] csky: don't let sigreturn play with priveleged bits of status register
@ 2021-09-25  5:16 guoren
  2021-09-25  5:16 ` [PATCH V2 2/2] csky: Fixup regs.sr broken in ptrace guoren
  0 siblings, 1 reply; 2+ messages in thread
From: guoren @ 2021-09-25  5:16 UTC (permalink / raw)
  To: guoren; +Cc: linux-kernel, linux-csky, Al Viro, stable

From: Al Viro <viro@zeniv.linux.org.uk>

csky restore_sigcontext() blindly overwrites regs->sr with the value
it finds in sigcontext.  Attacker can store whatever they want in there,
which includes things like S-bit.  Userland shouldn't be able to set
that, or anything other than C flag (bit 0).

Do the same thing other architectures with protected bits in flags
register do - preserve everything that shouldn't be settable in
user mode, picking the rest from the value saved is sigcontext.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Guo Ren <guoren@kernel.org>
Cc: stable@vger.kernel.org
---
 arch/csky/kernel/signal.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/csky/kernel/signal.c b/arch/csky/kernel/signal.c
index 312f046d452d..6ba3969ec175 100644
--- a/arch/csky/kernel/signal.c
+++ b/arch/csky/kernel/signal.c
@@ -52,10 +52,14 @@ static long restore_sigcontext(struct pt_regs *regs,
 	struct sigcontext __user *sc)
 {
 	int err = 0;
+	unsigned long sr = regs->sr;
 
 	/* sc_pt_regs is structured the same as the start of pt_regs */
 	err |= __copy_from_user(regs, &sc->sc_pt_regs, sizeof(struct pt_regs));
 
+	/* BIT(0) of regs->sr is Condition Code/Carry bit */
+	regs->sr = (sr & ~1) | (regs->sr & 1);
+
 	/* Restore the floating-point state. */
 	err |= restore_fpu_state(sc);
 
-- 
2.25.1


^ permalink raw reply related	[flat|nested] 2+ messages in thread

* [PATCH V2 2/2] csky: Fixup regs.sr broken in ptrace
  2021-09-25  5:16 [PATCH V2 1/2] csky: don't let sigreturn play with priveleged bits of status register guoren
@ 2021-09-25  5:16 ` guoren
  0 siblings, 0 replies; 2+ messages in thread
From: guoren @ 2021-09-25  5:16 UTC (permalink / raw)
  To: guoren; +Cc: linux-kernel, linux-csky, Guo Ren, Al Viro, stable

From: Guo Ren <guoren@linux.alibaba.com>

gpr_get() return the entire pt_regs (include sr) to userspace, if we
don't restore the C bit in gpr_set, it may break the ALU result in
that context. So the C flag bit is part of gpr context, that's why
riscv totally remove the C bit in the ISA. That makes sr reg clear
from userspace to supervisor privilege.

Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: stable@vger.kernel.org
---
 arch/csky/kernel/ptrace.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/csky/kernel/ptrace.c b/arch/csky/kernel/ptrace.c
index 0105ac81b432..1a5f54e0d272 100644
--- a/arch/csky/kernel/ptrace.c
+++ b/arch/csky/kernel/ptrace.c
@@ -99,7 +99,8 @@ static int gpr_set(struct task_struct *target,
 	if (ret)
 		return ret;
 
-	regs.sr = task_pt_regs(target)->sr;
+	/* BIT(0) of regs.sr is Condition Code/Carry bit */
+	regs.sr = (regs.sr & BIT(0)) | (task_pt_regs(target)->sr & ~BIT(0));
 #ifdef CONFIG_CPU_HAS_HILO
 	regs.dcsr = task_pt_regs(target)->dcsr;
 #endif
-- 
2.25.1


^ permalink raw reply related	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2021-09-25  5:17 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-09-25  5:16 [PATCH V2 1/2] csky: don't let sigreturn play with priveleged bits of status register guoren
2021-09-25  5:16 ` [PATCH V2 2/2] csky: Fixup regs.sr broken in ptrace guoren

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).