linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: linux-kernel@vger.kernel.org, x86@kernel.org,
	"Andy Lutomirski" <luto@kernel.org>,
	"Radim Krčmář" <rkrcmar@redhat.com>,
	kvm@vger.kernel.org, "Jason A. Donenfeld" <Jason@zx2c4.com>,
	"Rik van Riel" <riel@surriel.com>
Subject: [RFC PATCH 04/10 v2 ] x86/fpu: eager switch PKRU state
Date: Fri, 14 Sep 2018 22:35:01 +0200	[thread overview]
Message-ID: <20180914203501.qibhpmueosvkr74w@linutronix.de> (raw)
In-Reply-To: <8e5b64e4-b3e6-f884-beb6-b7b69ab2d8c1@redhat.com>

While most of a task's FPU state is only needed in user space,
the protection keys need to be in place immediately after a
context switch.

The reason is that any accesses to userspace memory while running
in kernel mode also need to abide by the memory permissions
specified in the protection keys.

The pkru info is put in its own cache line in the fpu struct because
that cache line is accessed anyway at context switch time.
Remove XFEATURE_MASK_PKRU from supported flags. This removes the PKRU
state from XSAVE/XRESTORE and so decouples it from the FPU state.

The initial state of pkru is updated in pkru_set_init_value() via
fpu__clear() - it is no longer affected by fpstate_init().

Signed-off-by: Rik van Riel <riel@surriel.com>
[bigeasy: load PKRU state only if we also load FPU content]
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---

v1…v2: remove PKRU from xsave/srestore.

On 2018-09-12 16:18:44 [+0200], Paolo Bonzini wrote:

> I think you can go a step further and exclude PKRU state from
> copy_kernel_to_fpregs altogether; you just use RDPKRU/WRPKRU.  This also
> means you don't need to call __fpregs_* functions in write_pkru.

something like this then? It looks like kvm excludes PKRU from
xsave/xrestore, too. This wouldn't be required then. This is (again)
untested since I have no box with this PKRU feature. This only available
on Intel's Xeon Scalable, right?

> Thanks,
> 
> Paolo

 arch/x86/include/asm/fpu/internal.h | 14 ++++++++++++--
 arch/x86/include/asm/fpu/types.h    |  7 +++++++
 arch/x86/include/asm/fpu/xstate.h   |  1 -
 arch/x86/include/asm/pgtable.h      |  7 +++++--
 arch/x86/include/asm/pkeys.h        |  2 +-
 arch/x86/kernel/fpu/core.c          |  2 +-
 arch/x86/kernel/fpu/xstate.c        |  4 ----
 arch/x86/mm/pkeys.c                 | 21 ++++-----------------
 include/linux/pkeys.h               |  2 +-
 9 files changed, 31 insertions(+), 29 deletions(-)

diff --git a/arch/x86/include/asm/fpu/internal.h b/arch/x86/include/asm/fpu/internal.h
index 16c4077ffc945..903ee77b6d5b0 100644
--- a/arch/x86/include/asm/fpu/internal.h
+++ b/arch/x86/include/asm/fpu/internal.h
@@ -573,8 +573,18 @@ static inline void switch_fpu_finish(struct fpu *new_fpu, int cpu)
 	bool preload = static_cpu_has(X86_FEATURE_FPU) &&
 		       new_fpu->initialized;
 
-	if (preload)
-		__fpregs_load_activate(new_fpu, cpu);
+	if (!preload)
+		return;
+
+	__fpregs_load_activate(new_fpu, cpu);
+
+#ifdef CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS
+	/* Protection keys need to be in place right at context switch time. */
+	if (boot_cpu_has(X86_FEATURE_OSPKE)) {
+		if (new_fpu->pkru != __read_pkru())
+			__write_pkru(new_fpu->pkru);
+	}
+#endif
 }
 
 /*
diff --git a/arch/x86/include/asm/fpu/types.h b/arch/x86/include/asm/fpu/types.h
index 202c53918ecfa..257b092bdaa4e 100644
--- a/arch/x86/include/asm/fpu/types.h
+++ b/arch/x86/include/asm/fpu/types.h
@@ -293,6 +293,13 @@ struct fpu {
 	 */
 	unsigned int			last_cpu;
 
+	/*
+	 * Protection key bits. The protection key needs to be switched out
+	 * immediately at context switch time, so it is in place for things
+	 * like copy_to_user.
+	 */
+	unsigned int			pkru;
+
 	/*
 	 * @initialized:
 	 *
diff --git a/arch/x86/include/asm/fpu/xstate.h b/arch/x86/include/asm/fpu/xstate.h
index 48581988d78c7..abe8793fa50f9 100644
--- a/arch/x86/include/asm/fpu/xstate.h
+++ b/arch/x86/include/asm/fpu/xstate.h
@@ -29,7 +29,6 @@
 				 XFEATURE_MASK_OPMASK | \
 				 XFEATURE_MASK_ZMM_Hi256 | \
 				 XFEATURE_MASK_Hi16_ZMM	 | \
-				 XFEATURE_MASK_PKRU | \
 				 XFEATURE_MASK_BNDREGS | \
 				 XFEATURE_MASK_BNDCSR)
 
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index 690c0307afed0..d87bdfaf45e56 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -134,8 +134,11 @@ static inline u32 read_pkru(void)
 
 static inline void write_pkru(u32 pkru)
 {
-	if (boot_cpu_has(X86_FEATURE_OSPKE))
-		__write_pkru(pkru);
+	if (!boot_cpu_has(X86_FEATURE_OSPKE))
+		return;
+
+	current->thread.fpu.pkru = pkru;
+	__write_pkru(pkru);
 }
 
 static inline int pte_young(pte_t pte)
diff --git a/arch/x86/include/asm/pkeys.h b/arch/x86/include/asm/pkeys.h
index 19b137f1b3beb..b184f916319e5 100644
--- a/arch/x86/include/asm/pkeys.h
+++ b/arch/x86/include/asm/pkeys.h
@@ -119,7 +119,7 @@ extern int arch_set_user_pkey_access(struct task_struct *tsk, int pkey,
 		unsigned long init_val);
 extern int __arch_set_user_pkey_access(struct task_struct *tsk, int pkey,
 		unsigned long init_val);
-extern void copy_init_pkru_to_fpregs(void);
+extern void pkru_set_init_value(void);
 
 static inline int vma_pkey(struct vm_area_struct *vma)
 {
diff --git a/arch/x86/kernel/fpu/core.c b/arch/x86/kernel/fpu/core.c
index 2ea85b32421a0..72cd2e2a07194 100644
--- a/arch/x86/kernel/fpu/core.c
+++ b/arch/x86/kernel/fpu/core.c
@@ -373,7 +373,7 @@ static inline void copy_init_fpstate_to_fpregs(void)
 		copy_kernel_to_fregs(&init_fpstate.fsave);
 
 	if (boot_cpu_has(X86_FEATURE_OSPKE))
-		copy_init_pkru_to_fpregs();
+		pkru_set_init_value();
 }
 
 /*
diff --git a/arch/x86/kernel/fpu/xstate.c b/arch/x86/kernel/fpu/xstate.c
index 87a57b7642d36..11014d841b9f7 100644
--- a/arch/x86/kernel/fpu/xstate.c
+++ b/arch/x86/kernel/fpu/xstate.c
@@ -920,10 +920,6 @@ int arch_set_user_pkey_access(struct task_struct *tsk, int pkey,
 	int pkey_shift = (pkey * PKRU_BITS_PER_PKEY);
 	u32 new_pkru_bits = 0;
 
-	/*
-	 * This check implies XSAVE support.  OSPKE only gets
-	 * set if we enable XSAVE and we enable PKU in XCR0.
-	 */
 	if (!boot_cpu_has(X86_FEATURE_OSPKE))
 		return -EINVAL;
 
diff --git a/arch/x86/mm/pkeys.c b/arch/x86/mm/pkeys.c
index 6e98e0a7c9231..2f9d95206c741 100644
--- a/arch/x86/mm/pkeys.c
+++ b/arch/x86/mm/pkeys.c
@@ -137,26 +137,13 @@ u32 init_pkru_value = PKRU_AD_KEY( 1) | PKRU_AD_KEY( 2) | PKRU_AD_KEY( 3) |
 		      PKRU_AD_KEY(10) | PKRU_AD_KEY(11) | PKRU_AD_KEY(12) |
 		      PKRU_AD_KEY(13) | PKRU_AD_KEY(14) | PKRU_AD_KEY(15);
 
-/*
- * Called from the FPU code when creating a fresh set of FPU
- * registers.  This is called from a very specific context where
- * we know the FPU regstiers are safe for use and we can use PKRU
- * directly.
- */
-void copy_init_pkru_to_fpregs(void)
+void pkru_set_init_value(void)
 {
 	u32 init_pkru_value_snapshot = READ_ONCE(init_pkru_value);
-	/*
-	 * Any write to PKRU takes it out of the XSAVE 'init
-	 * state' which increases context switch cost.  Avoid
-	 * writing 0 when PKRU was already 0.
-	 */
-	if (!init_pkru_value_snapshot && !read_pkru())
+
+	if (init_pkru_value_snapshot == read_pkru())
 		return;
-	/*
-	 * Override the PKRU state that came from 'init_fpstate'
-	 * with the baseline from the process.
-	 */
+
 	write_pkru(init_pkru_value_snapshot);
 }
 
diff --git a/include/linux/pkeys.h b/include/linux/pkeys.h
index 2955ba9760489..9a9efecc1388f 100644
--- a/include/linux/pkeys.h
+++ b/include/linux/pkeys.h
@@ -44,7 +44,7 @@ static inline bool arch_pkeys_enabled(void)
 	return false;
 }
 
-static inline void copy_init_pkru_to_fpregs(void)
+static inline void pkru_set_init_value(void)
 {
 }
 
-- 
2.19.0


  parent reply	other threads:[~2018-09-14 20:35 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-09-12 13:33 [RFC PATCH] x86: load FPU registers on return to userland Sebastian Andrzej Siewior
2018-09-12 13:33 ` [RFC PATCH 01/10] x86/entry: remove _TIF_ALLWORK_MASK Sebastian Andrzej Siewior
2018-09-27 14:21   ` Sebastian Andrzej Siewior
2018-09-12 13:33 ` [RFC PATCH 02/10] kvm: x86: make kvm_{load|put}_guest_fpu() static Sebastian Andrzej Siewior
2018-09-12 13:33 ` [RFC PATCH 03/10] x86/fpu: add (__)make_fpregs_active helpers Sebastian Andrzej Siewior
2018-09-12 13:33 ` [RFC PATCH 04/10] x86/fpu: eager switch PKRU state Sebastian Andrzej Siewior
2018-09-12 14:18   ` Paolo Bonzini
2018-09-12 15:24     ` Andy Lutomirski
2018-09-12 15:30       ` Paolo Bonzini
2018-09-14 20:35     ` Sebastian Andrzej Siewior [this message]
2018-09-17  8:37       ` [RFC PATCH 04/10 v2 ] " Paolo Bonzini
2018-09-18 14:27         ` Sebastian Andrzej Siewior
2018-09-18 15:07           ` Paolo Bonzini
2018-09-18 15:11             ` Rik van Riel
2018-09-18 15:29               ` Paolo Bonzini
2018-09-18 16:04                 ` Sebastian Andrzej Siewior
2018-09-18 17:29                   ` Rik van Riel
2018-09-19  5:55                     ` Paolo Bonzini
2018-09-19 16:57                       ` Sebastian Andrzej Siewior
2018-09-19 17:00                         ` Paolo Bonzini
2018-09-19 17:19                           ` Sebastian Andrzej Siewior
2018-09-19 19:38                           ` Rik van Riel
2018-09-19 19:49                           ` Andy Lutomirski
2018-09-12 15:20   ` [RFC PATCH 04/10] " Andy Lutomirski
2018-09-12 15:30     ` Rik van Riel
2018-09-12 15:49       ` Andy Lutomirski
2018-09-19 16:58         ` Sebastian Andrzej Siewior
2018-09-12 13:33 ` [RFC PATCH 05/10] x86/pkeys: Drop the preempt-disable section Sebastian Andrzej Siewior
2018-09-12 13:33 ` [RFC PATCH 06/10] x86/fpu: Always store the registers in copy_fpstate_to_sigframe() Sebastian Andrzej Siewior
2018-09-12 13:33 ` [RFC PATCH 07/10] x86/entry: add TIF_LOAD_FPU Sebastian Andrzej Siewior
2018-09-12 13:33 ` [RFC PATCH 08/10] x86/fpu: prepare copy_fpstate_to_sigframe for TIF_LOAD_FPU Sebastian Andrzej Siewior
2018-09-12 13:33 ` [RFC PATCH 09/10] x86/fpu: copy non-resident FPU state at fork time Sebastian Andrzej Siewior
2018-09-12 13:33 ` [RFC PATCH 10/10] x86/fpu: defer FPU state load until return to userspace Sebastian Andrzej Siewior
2018-09-12 15:47   ` Andy Lutomirski
2018-09-19 17:05     ` Sebastian Andrzej Siewior
2018-09-21  3:45       ` Andy Lutomirski
2018-09-21  4:15         ` Andy Lutomirski
2018-09-26 11:12           ` Sebastian Andrzej Siewior
2018-09-26 14:34             ` Andy Lutomirski
2018-09-26 15:32               ` Sebastian Andrzej Siewior
2018-09-26 16:24                 ` Andy Lutomirski

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20180914203501.qibhpmueosvkr74w@linutronix.de \
    --to=bigeasy@linutronix.de \
    --cc=Jason@zx2c4.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luto@kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=riel@surriel.com \
    --cc=rkrcmar@redhat.com \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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).