linux-kernel - [PATCH v3 0/4] KVM: x86: emulate FXSAVE and FXRSTOR

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-next>] [day] [month] [year] [list]

Message-Id: <20161108195419.4607-1-rkrcmar@redhat.com>
Date:   Tue,  8 Nov 2016 20:54:15 +0100
From:   Radim Krčmář <rkrcmar@...hat.com>
To:     linux-kernel@...r.kernel.org, kvm@...r.kernel.org
Cc:     Paolo Bonzini <pbonzini@...hat.com>, Bandan Das <bsd@...hat.com>,
        Nadav Amit <nadav.amit@...il.com>
Subject: [PATCH v3 0/4] KVM: x86: emulate FXSAVE and FXRSTOR

v2: http://www.spinics.net/lists/kvm/msg139681.html

v3 brings compatibility with old compilers and has been compile-tested
with GCC-4.4 on Debian Wheezy, GCC-4.4 on RHEL 6, and GCC-4.1 on RHEL 5.

[4/4] still has the hidden assumption that guest and host CPUID match.
Emulating a guest that does not deprecaste FCS and FDS on a host that
does (a modern host) will not necessarily cause a problem, which is why
patches don't handle the case.
Enforcing the equality in CPUID update would be best, but another series
could do that as Linux doesn't even have the CPUID bit defined yet.

Radim Krčmář (4):
  KVM: x86: add Align16 instruction flag
  KVM: x86: save one bit in ctxt->d
  KVM: x86: add asm_safe wrapper
  KVM: x86: emulate FXSAVE and FXRSTOR

 arch/x86/kvm/emulate.c | 184 +++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 157 insertions(+), 27 deletions(-)

-- 
2.10.2