All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andy Lutomirski <luto@kernel.org>
To: x86@kernel.org
Cc: Borislav Petkov <bp@alien8.de>,
	linux-kernel@vger.kernel.org,
	Rusty Russell <rusty@rustcorp.com.au>,
	Paolo Bonzini <pbonzini@redhat.com>,
	Rik van Riel <riel@redhat.com>, kvm list <kvm@vger.kernel.org>,
	Andy Lutomirski <luto@kernel.org>
Subject: [PATCH 4/8] x86/kvm: Remove host CR0.TS manipulation
Date: Mon, 31 Oct 2016 15:18:45 -0700	[thread overview]
Message-ID: <b212064922537c05d0c81d931fc4dbe769127ce7.1477951965.git.luto@kernel.org> (raw)
In-Reply-To: <cover.1477951965.git.luto@kernel.org>
In-Reply-To: <cover.1477951965.git.luto@kernel.org>

Now that x86 always uses eager fpu switching on the host, there's no
need for KVM to manipulate the host's CR0.TS.

This should be both simpler and faster.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
---
 arch/x86/kvm/vmx.c | 12 ++++--------
 arch/x86/kvm/x86.c |  5 -----
 2 files changed, 4 insertions(+), 13 deletions(-)

diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index cf1b16dbc98a..531c44633190 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -2144,12 +2144,6 @@ static void __vmx_load_host_state(struct vcpu_vmx *vmx)
 #endif
 	if (vmx->host_state.msr_host_bndcfgs)
 		wrmsrl(MSR_IA32_BNDCFGS, vmx->host_state.msr_host_bndcfgs);
-	/*
-	 * If the FPU is not active (through the host task or
-	 * the guest vcpu), then restore the cr0.TS bit.
-	 */
-	if (!fpregs_active() && !vmx->vcpu.guest_fpu_loaded)
-		stts();
 	load_gdt(this_cpu_ptr(&host_gdt));
 }
 
@@ -4871,9 +4865,11 @@ static void vmx_set_constant_host_state(struct vcpu_vmx *vmx)
 	u32 low32, high32;
 	unsigned long tmpl;
 	struct desc_ptr dt;
-	unsigned long cr4;
+	unsigned long cr0, cr4;
 
-	vmcs_writel(HOST_CR0, read_cr0() & ~X86_CR0_TS);  /* 22.2.3 */
+	cr0 = read_cr0();
+	WARN_ON(cr0 & X86_CR0_TS);
+	vmcs_writel(HOST_CR0, cr0);  /* 22.2.3 */
 	vmcs_writel(HOST_CR3, read_cr3());  /* 22.2.3  FIXME: shadow tables */
 
 	/* Save the most likely value for this task's CR4 in the VMCS. */
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index d5700263dad2..123d428d319a 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -5060,11 +5060,6 @@ static void emulator_get_fpu(struct x86_emulate_ctxt *ctxt)
 {
 	preempt_disable();
 	kvm_load_guest_fpu(emul_to_vcpu(ctxt));
-	/*
-	 * CR0.TS may reference the host fpu state, not the guest fpu state,
-	 * so it may be clear at this point.
-	 */
-	clts();
 }
 
 static void emulator_put_fpu(struct x86_emulate_ctxt *ctxt)
-- 
2.7.4

  parent reply	other threads:[~2016-10-31 22:19 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-10-31 22:18 [PATCH 0/8] x86/fpu: Remove CR0.TS support Andy Lutomirski
2016-10-31 22:18 ` [PATCH 1/8] fpu/init: Get rid of two redundant clts() calls Andy Lutomirski
2016-11-01  7:13   ` [tip:x86/fpu] x86/fpu: " tip-bot for Andy Lutomirski
2016-10-31 22:18 ` [PATCH 2/8] fpu/bugs: Stop saving and restoring CR0.TS in fpu__init_check_bugs() Andy Lutomirski
2016-11-01  7:14   ` [tip:x86/fpu] x86/fpu: " tip-bot for Andy Lutomirski
2016-10-31 22:18 ` [PATCH 3/8] x86/fpu: Remove irq_ts_save() and irq_ts_restore() Andy Lutomirski
2016-11-01  7:14   ` [tip:x86/fpu] " tip-bot for Andy Lutomirski
2016-10-31 22:18 ` Andy Lutomirski [this message]
2016-11-01  7:15   ` [tip:x86/fpu] x86/fpu, kvm: Remove host CR0.TS manipulation tip-bot for Andy Lutomirski
2016-10-31 22:18 ` [PATCH 5/8] lguest: Remove CR0.TS support Andy Lutomirski
2016-11-01  7:15   ` [tip:x86/fpu] x86/fpu, " tip-bot for Andy Lutomirski
2016-10-31 22:18 ` [PATCH 6/8] x86/fpu: #NM without FPU emulation is an error Andy Lutomirski
2016-11-01  7:16   ` [tip:x86/fpu] x86/fpu: Handle #NM without FPU emulation as " tip-bot for Andy Lutomirski
2016-10-31 22:18 ` [PATCH 7/8] x86/fpu: Remove stts() Andy Lutomirski
2016-11-01  7:16   ` [tip:x86/fpu] " tip-bot for Andy Lutomirski
2016-10-31 22:18 ` [PATCH 8/8] x86/fpu: Remove clts() Andy Lutomirski
2016-11-01  7:17   ` [tip:x86/fpu] " tip-bot for Andy Lutomirski
2016-10-31 22:41 ` [PATCH 0/8] x86/fpu: Remove CR0.TS support Paul Bolle
2016-10-31 23:04   ` Borislav Petkov
2016-10-31 23:10     ` Paul Bolle
2016-10-31 23:48       ` Borislav Petkov
2016-11-01  7:51         ` Paul Bolle
2016-11-01  8:50           ` Borislav Petkov

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=b212064922537c05d0c81d931fc4dbe769127ce7.1477951965.git.luto@kernel.org \
    --to=luto@kernel.org \
    --cc=bp@alien8.de \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=riel@redhat.com \
    --cc=rusty@rustcorp.com.au \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.