linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Nick Desaulniers <nick.desaulniers@gmail.com>
To: unlisted-recipients:; (no To-header on input)
Cc: "Nick Desaulniers" <nick.desaulniers@gmail.com>,
	"Paolo Bonzini" <pbonzini@redhat.com>,
	"Radim Krčmář" <rkrcmar@redhat.com>,
	"Thomas Gleixner" <tglx@linutronix.de>,
	"Ingo Molnar" <mingo@redhat.com>,
	"H. Peter Anvin" <hpa@zytor.com>,
	x86@kernel.org, kvm@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: [PATCH v4] KVM: x86: avoid large stack allocations in em_fxrstor
Date: Mon, 29 May 2017 15:48:58 -0700	[thread overview]
Message-ID: <20170529224858.9942-1-nick.desaulniers@gmail.com> (raw)
In-Reply-To: <20170529203908.10775-1-nick.desaulniers@gmail.com>

em_fxstor previously called fxstor_fixup.  Both created instances of
struct fxregs_state on the stack, which triggered the warning:

arch/x86/kvm/emulate.c:4018:12: warning: stack frame size of 1080 bytes
in function
      'em_fxrstor' [-Wframe-larger-than=]
static int em_fxrstor(struct x86_emulate_ctxt *ctxt)
           ^
with CONFIG_FRAME_WARN set to 1024.

This patch does the fixup in em_fxstor now, avoiding one additional
struct fxregs_state, and now fxstor_fixup can be removed as it has no
other call sites.

Signed-off-by: Nick Desaulniers <nick.desaulniers@gmail.com>
---
New in V4:
* undo changes for V3, prefer to only call segmented_read_std() when
  size has been initialized.

New in V3:
* initialized size to 0 to avoid maybe-uninitialized warning.  Check
  that it gets set to something other than 0 before passing size to
  segmented_read_std().

New in V2:
* reworked patch to do what was recommended by maintainers in:
  https://lkml.org/lkml/2017/5/25/391

 arch/x86/kvm/emulate.c | 64 ++++++++++++++++++++------------------------------
 1 file changed, 26 insertions(+), 38 deletions(-)

diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index 0816ab2e8adc..0367f7f81792 100644
--- a/arch/x86/kvm/emulate.c
+++ b/arch/x86/kvm/emulate.c
@@ -3985,57 +3985,45 @@ static int em_fxsave(struct x86_emulate_ctxt *ctxt)
 	return segmented_write_std(ctxt, ctxt->memop.addr.mem, &fx_state, size);
 }
 
-static int fxrstor_fixup(struct x86_emulate_ctxt *ctxt,
-		struct fxregs_state *new)
-{
-	int rc = X86EMUL_CONTINUE;
-	struct fxregs_state old;
-
-	rc = asm_safe("fxsave %[fx]", , [fx] "+m"(old));
-	if (rc != X86EMUL_CONTINUE)
-		return rc;
-
-	/*
-	 * 64 bit host will restore XMM 8-15, which is not correct on non-64
-	 * bit guests.  Load the current values in order to preserve 64 bit
-	 * XMMs after fxrstor.
-	 */
-#ifdef CONFIG_X86_64
-	/* XXX: accessing XMM 8-15 very awkwardly */
-	memcpy(&new->xmm_space[8 * 16/4], &old.xmm_space[8 * 16/4], 8 * 16);
-#endif
-
-	/*
-	 * Hardware doesn't save and restore XMM 0-7 without CR4.OSFXSR, but
-	 * does save and restore MXCSR.
-	 */
-	if (!(ctxt->ops->get_cr(ctxt, 4) & X86_CR4_OSFXSR))
-		memcpy(new->xmm_space, old.xmm_space, 8 * 16);
-
-	return rc;
-}
-
 static int em_fxrstor(struct x86_emulate_ctxt *ctxt)
 {
 	struct fxregs_state fx_state;
 	int rc;
+	unsigned int size;
 
 	rc = check_fxsr(ctxt);
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 
-	rc = segmented_read_std(ctxt, ctxt->memop.addr.mem, &fx_state, 512);
-	if (rc != X86EMUL_CONTINUE)
-		return rc;
+	ctxt->ops->get_fpu(ctxt);
+
+	if (ctxt->mode < X86EMUL_MODE_PROT64) {
+		rc = asm_safe("fxsave %[fx]", , [fx] "+m"(fx_state));
+		if (rc != X86EMUL_CONTINUE)
+			return rc;
+		/*
+		 * Hardware doesn't save and restore XMM 0-7 without
+		 * CR4.OSFXSR, but does save and restore MXCSR.
+		 */
+		if (ctxt->ops->get_cr(ctxt, 4) & X86_CR4_OSFXSR)
+			size = offsetof(struct fxregs_state, xmm_space[8]);
+		else
+			size = offsetof(struct fxregs_state, xmm_space[0]);
+		rc = segmented_read_std(ctxt, ctxt->memop.addr.mem, &fx_state,
+			size);
+		if (rc != X86EMUL_CONTINUE)
+			return rc;
+	} else if (ctxt->mode == X86EMUL_MODE_PROT64) {
+		size = offsetof(struct fxregs_state, xmm_space[16]);
+		rc = segmented_read_std(ctxt, ctxt->memop.addr.mem, &fx_state,
+			size);
+		if (rc != X86EMUL_CONTINUE)
+			return rc;
+	}
 
 	if (fx_state.mxcsr >> 16)
 		return emulate_gp(ctxt, 0);
 
-	ctxt->ops->get_fpu(ctxt);
-
-	if (ctxt->mode < X86EMUL_MODE_PROT64)
-		rc = fxrstor_fixup(ctxt, &fx_state);
-
 	if (rc == X86EMUL_CONTINUE)
 		rc = asm_safe("fxrstor %[fx]", : [fx] "m"(fx_state));
 
-- 
2.11.0

  parent reply	other threads:[~2017-05-29 22:49 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-24  6:24 [PATCH] KVM: x86: dynamically allocate large struct in em_fxrstor Nick Desaulniers
2017-05-24 14:19 ` Radim Krčmář
2017-05-25  1:36   ` Nick Desaulniers
2017-05-25 14:07     ` Paolo Bonzini
2017-05-26  4:13       ` Nick Desaulniers
2017-05-26  7:18         ` Paolo Bonzini
2017-05-29 19:55           ` [PATCH v2] KVM: x86: avoid large stack allocations " Nick Desaulniers
2017-05-29 20:14             ` kbuild test robot
2017-05-29 20:29               ` Nick Desaulniers
2017-05-29 20:39             ` [PATCH v3] " Nick Desaulniers
2017-05-29 22:40               ` Nick Desaulniers
2017-05-29 22:48               ` Nick Desaulniers [this message]
2017-05-30 10:15                 ` [PATCH v4] " Paolo Bonzini
2017-05-30 14:05                   ` Radim Krčmář
2017-05-31  3:08                 ` [PATCH v5] " Nick Desaulniers
2017-05-31 11:01                   ` Paolo Bonzini
2017-06-01  1:05                     ` Nick Desaulniers
2017-06-01  7:36                       ` Paolo Bonzini
2017-06-02  2:10                         ` Nick Desaulniers

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=20170529224858.9942-1-nick.desaulniers@gmail.com \
    --to=nick.desaulniers@gmail.com \
    --cc=hpa@zytor.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=rkrcmar@redhat.com \
    --cc=tglx@linutronix.de \
    --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).