From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Mon, 14 Aug 2017 22:36:19 -0700 Subject: [PATCH] x86/xen/64: Fix the reported SS and CS in SYSCALL MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit CVE-2017-5754 When I cleaned up the Xen SYSCALL entries, I inadvertently changed the reported segment registers. Before my patch, regs->ss was __USER(32)_DS and regs->cs was __USER(32)_CS. After the patch, they are FLAT_USER_CS/DS(32). This had a couple unfortunate effects. It confused the opportunistic fast return logic. It also significantly increased the risk of triggering a nasty glibc bug: https://sourceware.org/bugzilla/show_bug.cgi?id=21269 Update the Xen entry code to change it back. Reported-by: Brian Gerst Signed-off-by: Andy Lutomirski Cc: Andrew Cooper Cc: Boris Ostrovsky Cc: Borislav Petkov Cc: Juergen Gross Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: xen-devel@lists.xenproject.org Fixes: 8a9949bc71a7 ("x86/xen/64: Rearrange the SYSCALL entries") Link: http://lkml.kernel.org/r/daba8351ea2764bb30272296ab9ce08a81bd8264.1502775273.git.luto@kernel.org Signed-off-by: Ingo Molnar (cherry picked from commit fa2016a8e7d846b306e431646d250500e1da0c33) Signed-off-by: Andy Whitcroft Signed-off-by: Kleber Sacilotto de Souza (cherry picked from commit 69a6ef3aeb274efe86fd74771830354f303ccc2f) Signed-off-by: Fabian Grünbichler --- arch/x86/xen/xen-asm_64.S | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/arch/x86/xen/xen-asm_64.S b/arch/x86/xen/xen-asm_64.S index a8a4f4c460a6..c5fee2680abc 100644 --- a/arch/x86/xen/xen-asm_64.S +++ b/arch/x86/xen/xen-asm_64.S @@ -88,6 +88,15 @@ RELOC(xen_sysret64, 1b+1) ENTRY(xen_syscall_target) popq %rcx popq %r11 + + /* + * Neither Xen nor the kernel really knows what the old SS and + * CS were. The kernel expects __USER_DS and __USER_CS, so + * report those values even though Xen will guess its own values. + */ + movq $__USER_DS, 4*8(%rsp) + movq $__USER_CS, 1*8(%rsp) + jmp entry_SYSCALL_64_after_hwframe ENDPROC(xen_syscall_target) @@ -97,6 +106,15 @@ ENDPROC(xen_syscall_target) ENTRY(xen_syscall32_target) popq %rcx popq %r11 + + /* + * Neither Xen nor the kernel really knows what the old SS and + * CS were. The kernel expects __USER32_DS and __USER32_CS, so + * report those values even though Xen will guess its own values. + */ + movq $__USER32_DS, 4*8(%rsp) + movq $__USER32_CS, 1*8(%rsp) + jmp entry_SYSCALL_compat_after_hwframe ENDPROC(xen_syscall32_target) -- 2.14.2