linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: LKML <linux-kernel@vger.kernel.org>
Cc: Andy Lutomirski <luto@kernel.org>,
	Dave Hansen <dave.hansen@linux.intel.com>,
	Fenghua Yu <fenghua.yu@intel.com>,
	Tony Luck <tony.luck@intel.com>,
	Yu-cheng Yu <yu-cheng.yu@intel.com>,
	Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
	Borislav Petkov <bp@suse.de>,
	Peter Zijlstra <peterz@infradead.org>,
	Kan Liang <kan.liang@linux.intel.com>
Subject: [patch V2 41/52] x86/fpu: Clean up the fpu__clear() variants
Date: Mon, 14 Jun 2021 17:44:49 +0200	[thread overview]
Message-ID: <20210614155357.866249780@linutronix.de> (raw)
In-Reply-To: 20210614154408.673478623@linutronix.de

From: Andy Lutomirski <luto@kernel.org>

fpu__clear() currently resets both register state and kernel XSAVE buffer
state.  It has two modes: one for all state (supervisor and user) and
another for user state only.  fpu__clear_all() uses the "all state"
(user_only=0) mode, while a number of signal paths use the user_only=1
mode.

Make fpu__clear() work only for user state (user_only=1) and remove the
"all state" (user_only=0) code.  Rename it to match so it can be used by
the signal paths.

Replace the "all state" (user_only=0) fpu__clear() functionality.  Use the
TIF_NEED_FPU_LOAD functionality instead of making any actual hardware
registers changes in this path.

Instead of invoking fpu__initialize() just memcpy() init_fpstate into the
tasks FPU state because that has already the correct format and in case of
PKRU also contains the default PKRU value. Move the actual PKRU write out
into flush_thread() where it belongs and where it will end up anyway when
PKRU and XSTATE have been distangled.

For bisectability a workaround is required which stores the PKRU value in
the xstate memory until PKRU is distangled from XSTATE for context
switching and return to user.

[ Dave Hansen: Polished changelog ]
[ tglx: Fixed the PKRU fallout ]

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/fpu/core.c |  113 ++++++++++++++++++++++++++++++---------------
 arch/x86/kernel/process.c  |   10 +++
 2 files changed, 87 insertions(+), 36 deletions(-)

--- a/arch/x86/kernel/fpu/core.c
+++ b/arch/x86/kernel/fpu/core.c
@@ -283,19 +283,6 @@ int fpu__copy(struct task_struct *dst, s
 }
 
 /*
- * Activate the current task's in-memory FPU context,
- * if it has not been used before:
- */
-static void fpu__initialize(struct fpu *fpu)
-{
-	WARN_ON_FPU(fpu != &current->thread.fpu);
-
-	set_thread_flag(TIF_NEED_FPU_LOAD);
-	fpstate_init(&fpu->state);
-	trace_x86_fpu_init_state(fpu);
-}
-
-/*
  * This function must be called before we read a task's fpstate.
  *
  * There's two cases where this gets called:
@@ -381,46 +368,100 @@ static inline void load_fpregs_from_init
 	pkru_write_default();
 }
 
+static inline unsigned int init_fpstate_copy_size(void)
+{
+	if (!use_xsave())
+		return fpu_kernel_xstate_size;
+
+	/* XSAVE(S) just needs the legacy and the xstate header part */
+	return sizeof(init_fpstate.xsave);
+}
+
+/* Temporary workaround. Will be removed once PKRU and XSTATE are distangled. */
+static inline void pkru_set_default_in_xstate(struct xregs_state *xsave)
+{
+	struct pkru_state *pk;
+
+	if (!cpu_feature_enabled(X86_FEATURE_OSPKE))
+		return;
+	/*
+	 * Force XFEATURE_PKRU to be set in the header otherwise
+	 * get_xsave_addr() does not work and it also needs to be set to
+	 * make XRSTOR(S) load it.
+	 */
+	xsave->header.xfeatures |= XFEATURE_MASK_PKRU;
+	pk = get_xsave_addr(xsave, XFEATURE_PKRU);
+	pk->pkru = pkru_get_init_value();
+}
+
 /*
- * Clear the FPU state back to init state.
- *
- * Called by sys_execve(), by the signal handler code and by various
- * error paths.
+ * Reset current->fpu memory state to the init values.
  */
-static void fpu__clear(struct fpu *fpu, bool user_only)
+static void fpu_reset_fpstate(void)
+{
+	struct fpu *fpu= &current->thread.fpu;
+
+	fpregs_lock();
+	fpu__drop(fpu);
+	/*
+	 * This does not change the actual hardware registers. It just
+	 * resets the memory image and sets TIF_NEED_FPU_LOAD so a
+	 * subsequent return to usermode will reload the registers from the
+	 * tasks memory image.
+	 *
+	 * Do not use fpstate_init() here. Just copy init_fpstate which has
+	 * the correct content already except for PKRU.
+	 */
+	memcpy(&fpu->state, &init_fpstate, init_fpstate_copy_size());
+	pkru_set_default_in_xstate(&fpu->state.xsave);
+	set_thread_flag(TIF_NEED_FPU_LOAD);
+	fpregs_unlock();
+}
+
+/*
+ * Reset current's user FPU states to the init states.  current's
+ * supervisor states, if any, are not modified by this function.  The
+ * caller guarantees that the XSTATE header in memory is intact.
+ */
+void fpu__clear_user_states(struct fpu *fpu)
 {
 	WARN_ON_FPU(fpu != &current->thread.fpu);
 
+	fpregs_lock();
 	if (!static_cpu_has(X86_FEATURE_FPU)) {
-		fpu__drop(fpu);
-		fpu__initialize(fpu);
+		fpu_reset_fpstate();
+		fpregs_unlock();
 		return;
 	}
 
-	fpregs_lock();
-
-	if (user_only) {
-		if (!fpregs_state_valid(fpu, smp_processor_id()) &&
-		    xfeatures_mask_supervisor())
-			xrstor_from_kernel(&fpu->state.xsave,
-					     xfeatures_mask_supervisor());
-		load_fpregs_from_init_fpstate(xfeatures_mask_user());
-	} else {
-		load_fpregs_from_init_fpstate(xfeatures_mask_all);
+	/*
+	 * Ensure that current's supervisor states are loaded into their
+	 * corresponding registers.
+	 */
+	if (xfeatures_mask_supervisor() &&
+	    !fpregs_state_valid(fpu, smp_processor_id())) {
+		xrstor_from_kernel(&fpu->state.xsave,
+				   xfeatures_mask_supervisor());
 	}
 
+	/* Reset user states in registers. */
+	load_fpregs_from_init_fpstate(xfeatures_mask_user());
+
+	/*
+	 * Now all FPU registers have their desired values.  Inform the FPU
+	 * state machine that current's FPU registers are in the hardware
+	 * registers. The memory image does not need to be updated because
+	 * any operation relying on it has to save the registers first when
+	 * currents FPU is marked active.
+	 */
 	fpregs_mark_activate();
-	fpregs_unlock();
-}
 
-void fpu__clear_user_states(struct fpu *fpu)
-{
-	fpu__clear(fpu, true);
+	fpregs_unlock();
 }
 
 void fpu_flush_thread(void)
 {
-	fpu__clear(&current->thread.fpu, false);
+	fpu_reset_fpstate();
 }
 
 /*
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -199,6 +199,15 @@ int copy_thread(unsigned long clone_flag
 	return ret;
 }
 
+static void pkru_flush_thread(void)
+{
+	/*
+	 * If PKRU is enabled the default PKRU value has to be loaded into
+	 * the hardware right here (similar to context switch).
+	 */
+	pkru_write_default();
+}
+
 void flush_thread(void)
 {
 	struct task_struct *tsk = current;
@@ -207,6 +216,7 @@ void flush_thread(void)
 	memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array));
 
 	fpu_flush_thread();
+	pkru_flush_thread();
 }
 
 void disable_TSC(void)


  parent reply	other threads:[~2021-06-14 16:40 UTC|newest]

Thread overview: 87+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-14 15:44 [patch V2 00/52] x86/fpu: Spring cleaning and PKRU sanitizing Thomas Gleixner
2021-06-14 15:44 ` [patch V2 01/52] x86/fpu: Make init_fpstate correct with optimized XSAVE Thomas Gleixner
2021-06-14 19:15   ` Borislav Petkov
2021-06-14 15:44 ` [patch V2 02/52] x86/fpu: Fix copy_xstate_to_kernel() gap handling Thomas Gleixner
2021-06-15 11:07   ` Borislav Petkov
2021-06-15 12:47     ` Thomas Gleixner
2021-06-15 12:59       ` Borislav Petkov
2021-06-16 22:02   ` Thomas Gleixner
2021-06-14 15:44 ` [patch V2 03/52] x86/pkeys: Revert a5eff7259790 ("x86/pkeys: Add PKRU value to init_fpstate") Thomas Gleixner
2021-06-15 13:15   ` Borislav Petkov
2021-06-14 15:44 ` [patch V2 04/52] x86/fpu: Mark various FPU states __ro_after_init Thomas Gleixner
2021-06-14 15:44 ` [patch V2 05/52] x86/fpu: Remove unused get_xsave_field_ptr() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 06/52] x86/fpu: Move inlines where they belong Thomas Gleixner
2021-06-14 15:44 ` [patch V2 07/52] x86/fpu: Limit xstate copy size in xstateregs_set() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 08/52] x86/fpu: Sanitize xstateregs_set() Thomas Gleixner
2021-06-15 17:40   ` Borislav Petkov
2021-06-15 21:32     ` Thomas Gleixner
2021-06-14 15:44 ` [patch V2 09/52] x86/fpu: Reject invalid MXCSR values in copy_kernel_to_xstate() Thomas Gleixner
2021-06-16 15:02   ` Borislav Petkov
2021-06-16 23:51     ` Thomas Gleixner
2021-06-14 15:44 ` [patch V2 10/52] x86/fpu: Simplify PTRACE_GETREGS code Thomas Gleixner
2021-06-14 15:44 ` [patch V2 11/52] x86/fpu: Rewrite xfpregs_set() Thomas Gleixner
2021-06-16 15:22   ` Borislav Petkov
2021-06-14 15:44 ` [patch V2 12/52] x86/fpu: Fail ptrace() requests that try to set invalid MXCSR values Thomas Gleixner
2021-06-16 15:31   ` Borislav Petkov
2021-06-14 15:44 ` [patch V2 13/52] x86/fpu: Clean up fpregs_set() Thomas Gleixner
2021-06-16 15:42   ` Borislav Petkov
2021-06-14 15:44 ` [patch V2 14/52] x86/fpu: Make copy_xstate_to_kernel() usable for [x]fpregs_get() Thomas Gleixner
2021-06-16 16:13   ` Borislav Petkov
2021-06-17 12:42     ` Thomas Gleixner
2021-06-14 15:44 ` [patch V2 15/52] x86/fpu: Use copy_uabi_xstate_to_membuf() in xfpregs_get() Thomas Gleixner
2021-06-17  8:59   ` Borislav Petkov
2021-06-18 11:19     ` Borislav Petkov
2021-06-18 13:25       ` [PATCH] selftests/x86/ptrace_syscall: Add a PTRACE_GETFPREGS test Borislav Petkov
2021-06-14 15:44 ` [patch V2 16/52] x86/fpu: Use copy_uabi_xstate_to_membuf() in fpregs_get() Thomas Gleixner
2021-06-17 11:50   ` Borislav Petkov
2021-06-17 12:43     ` Thomas Gleixner
2021-06-14 15:44 ` [patch V2 17/52] x86/fpu: Remove fpstate_sanitize_xstate() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 18/52] x86/fpu: Get rid of using_compacted_format() Thomas Gleixner
2021-06-17 11:59   ` Borislav Petkov
2021-06-14 15:44 ` [patch V2 19/52] x86/kvm: Avoid looking up PKRU in XSAVE buffer Thomas Gleixner
2021-06-17 12:09   ` Borislav Petkov
2021-06-14 15:44 ` [patch V2 20/52] x86/fpu: Cleanup arch_set_user_pkey_access() Thomas Gleixner
2021-06-17 12:22   ` Borislav Petkov
2021-06-17 12:49     ` Thomas Gleixner
2021-06-14 15:44 ` [patch V2 21/52] x86/fpu: Get rid of copy_supervisor_to_kernel() Thomas Gleixner
2021-06-17 12:41   ` Borislav Petkov
2021-06-14 15:44 ` [patch V2 22/52] x86/fpu: Rename copy_xregs_to_kernel() and copy_kernel_to_xregs() Thomas Gleixner
2021-06-17 12:48   ` Borislav Petkov
2021-06-14 15:44 ` [patch V2 23/52] x86/fpu: Rename copy_user_to_xregs() and copy_xregs_to_user() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 24/52] x86/fpu: Rename fxregs related copy functions Thomas Gleixner
2021-06-14 15:44 ` [patch V2 25/52] x86/fpu: Rename fregs " Thomas Gleixner
2021-06-14 15:44 ` [patch V2 26/52] x86/fpu: Rename xstate copy functions which are related to UABI Thomas Gleixner
2021-06-14 15:44 ` [patch V2 27/52] x86/fpu: Deduplicate copy_uabi_from_user/kernel_to_xstate() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 28/52] x86/fpu: Rename copy_fpregs_to_fpstate() to save_fpregs_to_fpstate() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 29/52] x86/fpu: Rename copy_kernel_to_fpregs() to restore_fpregs_from_kernel() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 30/52] x86/fpu: Rename initstate copy functions Thomas Gleixner
2021-06-14 15:44 ` [patch V2 31/52] x86/fpu: Rename "dynamic" XSTATEs to "independent" Thomas Gleixner
2021-06-14 15:44 ` [patch V2 32/52] x86/fpu/xstate: Sanitize handling of independent features Thomas Gleixner
2021-06-16 20:04   ` Liang, Kan
2021-06-17  7:15     ` Thomas Gleixner
2021-06-14 15:44 ` [patch V2 33/52] x86/pkeys: Move read_pkru() and write_pkru() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 34/52] x86/fpu: Differentiate "copy" versus "move" of fpregs Thomas Gleixner
2021-06-18 12:21   ` Thomas Gleixner
2021-06-14 15:44 ` [patch V2 35/52] x86/cpu: Sanitize X86_FEATURE_OSPKE Thomas Gleixner
2021-06-14 15:44 ` [patch V2 36/52] x86/pkru: Provide pkru_get_init_value() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 37/52] x86/pkru: Provide pkru_write_default() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 38/52] x86/cpu: Write the default PKRU value when enabling PKE Thomas Gleixner
2021-06-14 15:44 ` [patch V2 39/52] x86/fpu: Use pkru_write_default() in copy_init_fpstate_to_fpregs() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 40/52] x86/fpu: Rename fpu__clear_all() to fpu_flush_thread() Thomas Gleixner
2021-06-14 15:44 ` Thomas Gleixner [this message]
2021-06-14 15:44 ` [patch V2 42/52] x86/fpu: Rename __fpregs_load_activate() to fpregs_restore_userregs() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 43/52] x86/fpu: Move FXSAVE_LEAK quirk info __copy_kernel_to_fpregs() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 44/52] x86/fpu: Rename xfeatures_mask_user() to xfeatures_mask_uabi() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 45/52] x86/fpu: Dont restore PKRU in fpregs_restore_userspace() Thomas Gleixner
2021-06-16  0:52   ` Yu, Yu-cheng
2021-06-16  8:56     ` Thomas Gleixner
2021-06-14 15:44 ` [patch V2 46/52] x86/fpu: Add PKRU storage outside of task XSAVE buffer Thomas Gleixner
2021-06-14 15:44 ` [patch V2 47/52] x86/fpu: Hook up PKRU into ptrace() Thomas Gleixner
2021-06-14 19:29   ` [patch V2-A " Thomas Gleixner
2021-06-14 15:44 ` [patch V2 48/52] x86/fpu: Mask PKRU from kernel XRSTOR[S] operations Thomas Gleixner
2021-06-14 15:44 ` [patch V2 49/52] x86/fpu: Remove PKRU handling from switch_fpu_finish() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 50/52] x86/fpu: Dont store PKRU in xstate in fpu_reset_fpstate() Thomas Gleixner
2021-06-14 15:44 ` [patch V2 51/52] x86/pkru: Remove xstate fiddling from write_pkru() Thomas Gleixner
2021-06-14 15:45 ` [patch V2 52/52] x86/fpu: Mark init_fpstate __ro_after_init Thomas Gleixner
2021-06-14 20:15 ` [patch] x86/fpu: x86/fpu: Preserve supervisor states in sanitize_restored_user_xstate() Thomas Gleixner
2021-06-16  0:50 ` [patch V2 00/52] x86/fpu: Spring cleaning and PKRU sanitizing Yu, Yu-cheng

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=20210614155357.866249780@linutronix.de \
    --to=tglx@linutronix.de \
    --cc=bigeasy@linutronix.de \
    --cc=bp@suse.de \
    --cc=dave.hansen@linux.intel.com \
    --cc=fenghua.yu@intel.com \
    --cc=kan.liang@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luto@kernel.org \
    --cc=peterz@infradead.org \
    --cc=tony.luck@intel.com \
    --cc=yu-cheng.yu@intel.com \
    /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).