Commit-ID: 86cc35109029b7f1b195cef6c74654bad95e81af Gitweb: https://git.kernel.org/tip/86cc35109029b7f1b195cef6c74654bad95e81af Author: Ricardo Neri <ricardo.neri-calde...@linux.intel.com> AuthorDate: Sun, 5 Nov 2017 18:27:49 -0800 Committer: Ingo Molnar <mi...@kernel.org> CommitDate: Wed, 8 Nov 2017 11:16:20 +0100
x86/insn-eval: Handle 32-bit address encodings in virtual-8086 mode It is possible to utilize 32-bit address encodings in virtual-8086 mode via an address override instruction prefix. However, the range of the effective address is still limited to [0x-0xffff]. In such a case, return error. Also, linear addresses in virtual-8086 mode are limited to 20 bits. Enforce such limit by truncating the most significant bytes of the computed linear address. Signed-off-by: Ricardo Neri <ricardo.neri-calde...@linux.intel.com> Reviewed-by: Thomas Gleixner <t...@linutronix.de> Cc: Adam Buchbinder <adam.buchbin...@gmail.com> Cc: Adrian Hunter <adrian.hun...@intel.com> Cc: Andrew Morton <a...@linux-foundation.org> Cc: Andy Lutomirski <l...@kernel.org> Cc: Arnaldo Carvalho de Melo <a...@redhat.com> Cc: Borislav Petkov <b...@alien8.de> Cc: Borislav Petkov <b...@suse.de> Cc: Brian Gerst <brge...@gmail.com> Cc: Chen Yucong <sla...@gmail.com> Cc: Chris Metcalf <cmetc...@mellanox.com> Cc: Colin Ian King <colin.k...@canonical.com> Cc: Dave Hansen <dave.han...@linux.intel.com> Cc: Denys Vlasenko <dvlas...@redhat.com> Cc: Dmitry Vyukov <dvyu...@google.com> Cc: H. Peter Anvin <h...@zytor.com> Cc: Huang Rui <ray.hu...@amd.com> Cc: Jiri Slaby <jsl...@suse.cz> Cc: Jonathan Corbet <cor...@lwn.net> Cc: Josh Poimboeuf <jpoim...@redhat.com> Cc: Kees Cook <keesc...@chromium.org> Cc: Linus Torvalds <torva...@linux-foundation.org> Cc: Lorenzo Stoakes <lstoa...@gmail.com> Cc: Masami Hiramatsu <mhira...@kernel.org> Cc: Michael S. Tsirkin <m...@redhat.com> Cc: Paolo Bonzini <pbonz...@redhat.com> Cc: Paul Gortmaker <paul.gortma...@windriver.com> Cc: Peter Zijlstra <pet...@infradead.org> Cc: Qiaowei Ren <qiaowei....@intel.com> Cc: Ravi V. Shankar <ravi.v.shan...@intel.com> Cc: Shuah Khan <sh...@kernel.org> Cc: Thomas Garnier <thgar...@google.com> Cc: Tony Luck <tony.l...@intel.com> Cc: Vlastimil Babka <vba...@suse.cz> Cc: ricardo.n...@intel.com Link: http://lkml.kernel.org/r/1509935277-22138-5-git-send-email-ricardo.neri-calde...@linux.intel.com Signed-off-by: Ingo Molnar <mi...@kernel.org> --- arch/x86/lib/insn-eval.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/arch/x86/lib/insn-eval.c b/arch/x86/lib/insn-eval.c index 1ac3973..ef102db 100644 --- a/arch/x86/lib/insn-eval.c +++ b/arch/x86/lib/insn-eval.c @@ -1042,12 +1042,23 @@ static void __user *get_addr_ref_32(struct insn *insn, struct pt_regs *regs) goto out; /* + * Even though 32-bit address encodings are allowed in virtual-8086 + * mode, the address range is still limited to [0x-0xffff]. + */ + if (v8086_mode(regs) && (eff_addr & ~0xffff)) + goto out; + + /* * Data type long could be 64 bits in size. Ensure that our 32-bit * effective address is not sign-extended when computing the linear * address. */ linear_addr = (unsigned long)(eff_addr & 0xffffffff) + seg_base; + /* Limit linear address to 20 bits */ + if (v8086_mode(regs)) + linear_addr &= 0xfffff; + out: return (void __user *)linear_addr; }
