linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, tglx@linutronix.de, bp@alien8.de,
	luto@kernel.org
Cc: hpa@zytor.com, dave.hansen@intel.com, tony.luck@intel.com,
	ak@linux.intel.com, ravi.v.shankar@intel.com,
	chang.seok.bae@intel.com, Sasha Levin <sashal@kernel.org>
Subject: [PATCH v10 11/18] x86/fsgsbase/64: Use FSGSBASE in switch_to() if available
Date: Thu, 23 Apr 2020 19:22:00 -0400	[thread overview]
Message-ID: <20200423232207.5797-12-sashal@kernel.org> (raw)
In-Reply-To: <20200423232207.5797-1-sashal@kernel.org>

From: Andy Lutomirski <luto@kernel.org>

With the new FSGSBASE instructions, FS/GS base can be efficiently read
and written in __switch_to(). Use that capability to preserve the full
state.

This will enable user code to do whatever it wants with the new
instructions without any kernel-induced gotchas.  (There can still be
architectural gotchas: movl %gs,%eax; movl %eax,%gs may change GS base
if WRGSBASE was used, but users are expected to read the CPU manual
before doing things like that.)

This is a considerable speedup. It seems to save about 100 cycles per
context switch compared to the baseline 4.6-rc1 behavior on a Skylake
laptop.

[ chang: 5~10% performance improvements were seen by a context switch
  benchmark that ran threads with different FS/GS base values (to the
  baseline 4.16). ]

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Chang S. Bae <chang.seok.bae@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 arch/x86/kernel/process_64.c | 34 ++++++++++++++++++++++++++++------
 1 file changed, 28 insertions(+), 6 deletions(-)

diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index aaa65f284b9b9..e066750be89a0 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -199,8 +199,18 @@ static __always_inline void save_fsgs(struct task_struct *task)
 {
 	savesegment(fs, task->thread.fsindex);
 	savesegment(gs, task->thread.gsindex);
-	save_base_legacy(task, task->thread.fsindex, FS);
-	save_base_legacy(task, task->thread.gsindex, GS);
+	if (static_cpu_has(X86_FEATURE_FSGSBASE)) {
+		/*
+		 * If FSGSBASE is enabled, we can't make any useful guesses
+		 * about the base, and user code expects us to save the current
+		 * value.  Fortunately, reading the base directly is efficient.
+		 */
+		task->thread.fsbase = rdfsbase();
+		task->thread.gsbase = x86_gsbase_read_cpu_inactive();
+	} else {
+		save_base_legacy(task, task->thread.fsindex, FS);
+		save_base_legacy(task, task->thread.gsindex, GS);
+	}
 }
 
 #if IS_ENABLED(CONFIG_KVM)
@@ -279,10 +289,22 @@ static __always_inline void load_seg_legacy(unsigned short prev_index,
 static __always_inline void x86_fsgsbase_load(struct thread_struct *prev,
 					      struct thread_struct *next)
 {
-	load_seg_legacy(prev->fsindex, prev->fsbase,
-			next->fsindex, next->fsbase, FS);
-	load_seg_legacy(prev->gsindex, prev->gsbase,
-			next->gsindex, next->gsbase, GS);
+	if (static_cpu_has(X86_FEATURE_FSGSBASE)) {
+		/* Update the FS and GS selectors if they could have changed. */
+		if (unlikely(prev->fsindex || next->fsindex))
+			loadseg(FS, next->fsindex);
+		if (unlikely(prev->gsindex || next->gsindex))
+			loadseg(GS, next->gsindex);
+
+		/* Update the bases. */
+		wrfsbase(next->fsbase);
+		x86_gsbase_write_cpu_inactive(next->gsbase);
+	} else {
+		load_seg_legacy(prev->fsindex, prev->fsbase,
+				next->fsindex, next->fsbase, FS);
+		load_seg_legacy(prev->gsindex, prev->gsbase,
+				next->gsindex, next->gsbase, GS);
+	}
 }
 
 static unsigned long x86_fsgsbase_read_task(struct task_struct *task,
-- 
2.20.1


  parent reply	other threads:[~2020-04-23 23:23 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-23 23:21 [PATCH v10 00/18] Enable FSGSBASE instructions Sasha Levin
2020-04-23 23:21 ` [PATCH v10 01/18] x86/ptrace: Prevent ptrace from clearing the FS/GS selector Sasha Levin
2020-04-25 22:46   ` Andy Lutomirski
2020-04-23 23:21 ` [PATCH v10 02/18] selftests/x86/fsgsbase: Test GS selector on ptracer-induced GS base write Sasha Levin
2020-04-23 23:21 ` [PATCH v10 03/18] x86/cpu: Add 'unsafe_fsgsbase' to enable CR4.FSGSBASE Sasha Levin
2020-04-23 23:21 ` [PATCH v10 04/18] x86/entry/64: Clean up paranoid exit Sasha Levin
2020-04-23 23:21 ` [PATCH v10 05/18] x86/entry/64: Switch CR3 before SWAPGS in paranoid entry Sasha Levin
2020-04-23 23:21 ` [PATCH v10 06/18] x86/entry/64: Introduce the FIND_PERCPU_BASE macro Sasha Levin
2020-04-23 23:21 ` [PATCH v10 07/18] x86/entry/64: Handle FSGSBASE enabled paranoid entry/exit Sasha Levin
2020-04-23 23:21 ` [PATCH v10 08/18] x86/entry/64: Document GSBASE handling in the paranoid path Sasha Levin
2020-04-23 23:21 ` [PATCH v10 09/18] x86/fsgsbase/64: Add intrinsics for FSGSBASE instructions Sasha Levin
2020-04-23 23:21 ` [PATCH v10 10/18] x86/fsgsbase/64: Enable FSGSBASE instructions in helper functions Sasha Levin
2020-04-23 23:22 ` Sasha Levin [this message]
2020-04-23 23:22 ` [PATCH v10 12/18] x86/fsgsbase/64: move save_fsgs to header file Sasha Levin
2020-04-23 23:22 ` [PATCH v10 13/18] x86/fsgsbase/64: Use FSGSBASE instructions on thread copy and ptrace Sasha Levin
2020-04-23 23:22 ` [PATCH v10 14/18] x86/speculation/swapgs: Check FSGSBASE in enabling SWAPGS mitigation Sasha Levin
2020-04-23 23:22 ` [PATCH v10 15/18] selftests/x86/fsgsbase: Test ptracer-induced GS base write with FSGSBASE Sasha Levin
2020-04-23 23:22 ` [PATCH v10 16/18] x86/fsgsbase/64: Enable FSGSBASE on 64bit by default and add a chicken bit Sasha Levin
2020-04-23 23:22 ` [PATCH v10 17/18] x86/elf: Enumerate kernel FSGSBASE capability in AT_HWCAP2 Sasha Levin
2020-04-23 23:22 ` [PATCH v10 18/18] Documentation/x86/64: Add documentation for GS/FS addressing mode Sasha Levin
2020-05-10  8:09 ` [PATCH v10 00/18] Enable FSGSBASE instructions Vegard Nossum
2020-05-10  8:29   ` Vegard Nossum
2020-05-10 10:15     ` Thomas Gleixner
2020-05-10 14:17       ` Sasha Levin
2020-05-11  0:48     ` Andi Kleen
2020-05-11  0:50       ` Andi Kleen
2020-05-11  5:03         ` Sasha Levin

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=20200423232207.5797-12-sashal@kernel.org \
    --to=sashal@kernel.org \
    --cc=ak@linux.intel.com \
    --cc=bp@alien8.de \
    --cc=chang.seok.bae@intel.com \
    --cc=dave.hansen@intel.com \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luto@kernel.org \
    --cc=ravi.v.shankar@intel.com \
    --cc=tglx@linutronix.de \
    --cc=tony.luck@intel.com \
    --subject='Re: [PATCH v10 11/18] x86/fsgsbase/64: Use FSGSBASE in switch_to() if available' \
    /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

This is a public inbox, see mirroring instructions
on how to clone and mirror all data and code used for this inbox