[tip:x86/fpu] x86/fpu: Add a fastpath to __fpu__restore_sig()

From: tip-bot for Sebastian Andrzej Siewior
Date: Sat Apr 13 2019 - 17:04:20 EST


Commit-ID: 1d731e731c4cd7cbd3b1aa295f0932e7610da82f
Gitweb: https://git.kernel.org/tip/1d731e731c4cd7cbd3b1aa295f0932e7610da82f
Author: Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx>
AuthorDate: Wed, 3 Apr 2019 18:41:53 +0200
Committer: Borislav Petkov <bp@xxxxxxx>
CommitDate: Fri, 12 Apr 2019 20:04:49 +0200

x86/fpu: Add a fastpath to __fpu__restore_sig()

The previous commits refactor the restoration of the FPU registers so
that they can be loaded from in-kernel memory. This overhead can be
avoided if the load can be performed without a pagefault.

Attempt to restore FPU registers by invoking
copy_user_to_fpregs_zeroing(). If it fails try the slowpath which can
handle pagefaults.

[ bp: Add a comment over the fastpath to be able to find one's way
around the function. ]

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx>
Signed-off-by: Borislav Petkov <bp@xxxxxxx>
Reviewed-by: Dave Hansen <dave.hansen@xxxxxxxxx>
Reviewed-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: Andy Lutomirski <luto@xxxxxxxxxx>
Cc: "H. Peter Anvin" <hpa@xxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: Jann Horn <jannh@xxxxxxxxxx>
Cc: "Jason A. Donenfeld" <Jason@xxxxxxxxx>
Cc: kvm ML <kvm@xxxxxxxxxxxxxxx>
Cc: Paolo Bonzini <pbonzini@xxxxxxxxxx>
Cc: Radim KrÄmÃÅ <rkrcmar@xxxxxxxxxx>
Cc: Rik van Riel <riel@xxxxxxxxxxx>
Cc: x86-ml <x86@xxxxxxxxxx>
Link: https://lkml.kernel.org/r/20190403164156.19645-25-bigeasy@xxxxxxxxxxxxx
---
arch/x86/kernel/fpu/signal.c | 23 +++++++++++++++++++++--
1 file changed, 21 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kernel/fpu/signal.c b/arch/x86/kernel/fpu/signal.c
index 6df1f15e0cd5..a1bd7be70206 100644
--- a/arch/x86/kernel/fpu/signal.c
+++ b/arch/x86/kernel/fpu/signal.c
@@ -242,10 +242,10 @@ sanitize_restored_xstate(union fpregs_state *state,
/*
* Restore the extended state if present. Otherwise, restore the FP/SSE state.
*/
-static inline int copy_user_to_fpregs_zeroing(void __user *buf, u64 xbv, int fx_only)
+static int copy_user_to_fpregs_zeroing(void __user *buf, u64 xbv, int fx_only)
{
if (use_xsave()) {
- if ((unsigned long)buf % 64 || fx_only) {
+ if (fx_only) {
u64 init_bv = xfeatures_mask & ~XFEATURE_MASK_FPSSE;
copy_kernel_to_xregs(&init_fpstate.xsave, init_bv);
return copy_user_to_fxregs(buf);
@@ -327,8 +327,27 @@ static int __fpu__restore_sig(void __user *buf, void __user *buf_fx, int size)
if (ret)
goto err_out;
envp = &env;
+ } else {
+ /*
+ * Attempt to restore the FPU registers directly from user
+ * memory. For that to succeed, the user access cannot cause
+ * page faults. If it does, fall back to the slow path below,
+ * going through the kernel buffer with the enabled pagefault
+ * handler.
+ */
+ fpregs_lock();
+ pagefault_disable();
+ ret = copy_user_to_fpregs_zeroing(buf_fx, xfeatures, fx_only);
+ pagefault_enable();
+ if (!ret) {
+ fpregs_mark_activate();
+ fpregs_unlock();
+ return 0;
+ }
+ fpregs_unlock();
}

+
if (use_xsave() && !fx_only) {
u64 init_bv = xfeatures_mask & ~xfeatures;