On Tue, Jun 22, 2021 at 10:03:03PM +0200, Thomas Gleixner wrote:
On Tue, Jun 22 2021 at 12:17, Greg KH wrote:
On Mon, Jun 21, 2021 at 07:45:02PM +0200, Borislav Petkov wrote:
On Mon, Jun 21, 2021 at 12:52:12PM +0200, gregkh@linuxfoundation.org wrote:
The patch below does not apply to the 4.4-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to stable@vger.kernel.org.
Ok, how's this below?
It should at least capture the gist of what this commit is trying to achieve as the FPU mess has changed substantially since 4.4 so I'm really cautious here not to break any existing setups.
I've boot-tested this in a VM but Greg, I'd appreciate running it through some sort of stable testing framework if you're using one.
This applied to 4.4.y and 4.9.y, but we still need a 4.14.y and 4.19.y version if at all possible.
Everything is possible :)
Subject: x86/fpu: Reset state for all signal restore failures From: Thomas Gleixner tglx@linutronix.de Date: Wed Jun 9 21:18:00 2021 +0200
From: Thomas Gleixner tglx@linutronix.de
commit efa165504943f2128d50f63de0c02faf6dcceb0d upstream
If access_ok() or fpregs_soft_set() fails in __fpu__restore_sig() then the function just returns but does not clear the FPU state as it does for all other fatal failures.
Clear the FPU state for these failures as well.
Fixes: 72a671ced66d ("x86, fpu: Unify signal handling code paths for x86 and x86_64 kernels") Signed-off-by: Thomas Gleixner tglx@linutronix.de Signed-off-by: Borislav Petkov bp@suse.de Signed-off-by: Thomas Gleixner tglx@linutronix.de Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/87mtryyhhz.ffs@nanos.tec.linutronix.de
arch/x86/kernel/fpu/signal.c | 30 +++++++++++++++++++----------- 1 file changed, 19 insertions(+), 11 deletions(-)
--- a/arch/x86/kernel/fpu/signal.c +++ b/arch/x86/kernel/fpu/signal.c @@ -281,15 +281,21 @@ static int __fpu__restore_sig(void __use return 0; }
- if (!access_ok(VERIFY_READ, buf, size))
return -EACCES;
- if (!access_ok(VERIFY_READ, buf, size)) {
ret = -EACCES;goto out_err;- }
fpu__initialize(fpu);
- if (!static_cpu_has(X86_FEATURE_FPU))
return fpregs_soft_set(current, NULL,0, sizeof(struct user_i387_ia32_struct),NULL, buf) != 0;
- if (!static_cpu_has(X86_FEATURE_FPU)) {
ret = fpregs_soft_set(current, NULL,0, sizeof(struct user_i387_ia32_struct),NULL, buf) != 0;if (ret)goto out_err;return 0;- }
if (use_xsave()) { struct _fpx_sw_bytes fx_sw_user; @@ -349,6 +355,7 @@ static int __fpu__restore_sig(void __use fpu__restore(fpu); local_bh_enable();
/* Failure is already handled */@@ -356,13 +363,14 @@ static int __fpu__restore_sig(void __use * state to the registers directly (with exceptions handled). */ user_fpu_begin();
if (copy_user_to_fpregs_zeroing(buf_fx, xfeatures, fx_only)) {fpu__clear(fpu);return -1;}
if (!copy_user_to_fpregs_zeroing(buf_fx, xfeatures, fx_only))return 0;ret = -1;
- return 0;
+out_err:
- fpu__clear(fpu);
- return ret;
} static inline int xstate_sigframe_size(void)
Hm, did you build this?
I get the following build error: arch/x86/kernel/fpu/signal.c: In function ‘__fpu__restore_sig’: arch/x86/kernel/fpu/signal.c:285:17: error: ‘ret’ undeclared (first use in this function); did you mean ‘net’? 285 | ret = -EACCES; | ^~~ | net arch/x86/kernel/fpu/signal.c:285:17: note: each undeclared identifier is reported only once for each function it appears in arch/x86/kernel/fpu/signal.c:374:1: warning: control reaches end of non-void function [-Wreturn-type] 374 | } | ^
I'll fix it up, it's an "obvious" change :)
thanks,
greg k-h