linux-kernel - Re: [PATCH v3] KVM: x86: avoid large stack allocations in em

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20170529224010.slebvm7ts6ki7llg@lostoracle.net>
Date:   Mon, 29 May 2017 15:40:10 -0700
From:   Nick Desaulniers <nick.desaulniers@...il.com>
To:     Paolo Bonzini <pbonzini@...hat.com>,
        Radim Krčmář <rkrcmar@...hat.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>,
        "H. Peter Anvin" <hpa@...or.com>, x86@...nel.org,
        kvm@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3] KVM: x86: avoid large stack allocations in em_fxrstor

On Mon, May 29, 2017 at 01:39:08PM -0700, Nick Desaulniers wrote:
> +	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]);
> +	} else if (ctxt->mode == X86EMUL_MODE_PROT64)
> +		size = offsetof(struct fxregs_state, xmm_space[16]);
> +
> +	if (size == 0)
> +		return X86EMUL_UNHANDLEABLE;
> +
> +	rc = segmented_read_std(ctxt, ctxt->memop.addr.mem, &fx_state, size);
>  	if (rc != X86EMUL_CONTINUE)
>  		return rc;

Thinking more about this, I think it may be more elegant to move the
segmented_read_std into the then/else branches above, remove
initialization of size, and remove the size == 0 check.  Thoughts?