| 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
| ||
|
Date: Thu, 26 Jan 2012 22:38:18 +0900
From: Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>
To: Denys Vlasenko <vda.linux@...glemail.com>
Cc: Srikar Dronamraju <srikar@...ux.vnet.ibm.com>,
Peter Zijlstra <peterz@...radead.org>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Oleg Nesterov <oleg@...hat.com>, Ingo Molnar <mingo@...e.hu>,
Andrew Morton <akpm@...ux-foundation.org>,
LKML <linux-kernel@...r.kernel.org>,
Linux-mm <linux-mm@...ck.org>, Andi Kleen <andi@...stfloor.org>,
Christoph Hellwig <hch@...radead.org>,
Steven Rostedt <rostedt@...dmis.org>,
Roland McGrath <roland@...k.frob.com>,
Thomas Gleixner <tglx@...utronix.de>,
Arnaldo Carvalho de Melo <acme@...radead.org>,
Anton Arapov <anton@...hat.com>,
Ananth N Mavinakayanahalli <ananth@...ibm.com>,
Jim Keniston <jkenisto@...ux.vnet.ibm.com>,
Stephen Rothwell <sfr@...b.auug.org.au>,
yrl.pp-manager.tt@...achi.com
Subject: Re: [PATCH v9 3.2 1/9] uprobes: Install and remove breakpoints.
(2012/01/26 0:13), Denys Vlasenko wrote:
> On Tue, Jan 10, 2012 at 12:48 PM, Srikar Dronamraju
> <srikar@...ux.vnet.ibm.com> wrote:
>> +/*
>> + * If uprobe->insn doesn't use rip-relative addressing, return
>> + * immediately. Otherwise, rewrite the instruction so that it accesses
>> + * its memory operand indirectly through a scratch register. Set
>> + * uprobe->arch_info.fixups and uprobe->arch_info.rip_rela_target_address
>> + * accordingly. (The contents of the scratch register will be saved
>> + * before we single-step the modified instruction, and restored
>> + * afterward.)
>> + *
>> + * We do this because a rip-relative instruction can access only a
>> + * relatively small area (+/- 2 GB from the instruction), and the XOL
>> + * area typically lies beyond that area. At least for instructions
>> + * that store to memory, we can't execute the original instruction
>> + * and "fix things up" later, because the misdirected store could be
>> + * disastrous.
>> + *
>> + * Some useful facts about rip-relative instructions:
>> + * - There's always a modrm byte.
>> + * - There's never a SIB byte.
>> + * - The displacement is always 4 bytes.
>> + */
>> +static void handle_riprel_insn(struct mm_struct *mm, struct uprobe *uprobe,
>> + struct insn *insn)
>> +{
>> + u8 *cursor;
>> + u8 reg;
>> +
>> + if (mm->context.ia32_compat)
>> + return;
>> +
>> + uprobe->arch_info.rip_rela_target_address = 0x0;
>> + if (!insn_rip_relative(insn))
>> + return;
>> +
>> + /*
>> + * Point cursor at the modrm byte. The next 4 bytes are the
>> + * displacement. Beyond the displacement, for some instructions,
>> + * is the immediate operand.
>> + */
>> + cursor = uprobe->insn + insn->prefixes.nbytes
>> + + insn->rex_prefix.nbytes + insn->opcode.nbytes;
>> + insn_get_length(insn);
>> +
>> + /*
>> + * Convert from rip-relative addressing to indirect addressing
>> + * via a scratch register. Change the r/m field from 0x5 (%rip)
>> + * to 0x0 (%rax) or 0x1 (%rcx), and squeeze out the offset field.
>> + */
>> + reg = MODRM_REG(insn);
>> + if (reg == 0) {
>> + /*
>> + * The register operand (if any) is either the A register
>> + * (%rax, %eax, etc.) or (if the 0x4 bit is set in the
>> + * REX prefix) %r8. In any case, we know the C register
>> + * is NOT the register operand, so we use %rcx (register
>> + * #1) for the scratch register.
>> + */
>> + uprobe->arch_info.fixups = UPROBES_FIX_RIP_CX;
>> + /* Change modrm from 00 000 101 to 00 000 001. */
>> + *cursor = 0x1;
>> + } else {
>> + /* Use %rax (register #0) for the scratch register. */
>> + uprobe->arch_info.fixups = UPROBES_FIX_RIP_AX;
>> + /* Change modrm from 00 xxx 101 to 00 xxx 000 */
>> + *cursor = (reg << 3);
>> + }
>> +
>> + /* Target address = address of next instruction + (signed) offset */
>> + uprobe->arch_info.rip_rela_target_address = (long)insn->length
>> + + insn->displacement.value;
>> + /* Displacement field is gone; slide immediate field (if any) over. */
>> + if (insn->immediate.nbytes) {
>> + cursor++;
>> + memmove(cursor, cursor + insn->displacement.nbytes,
>> + insn->immediate.nbytes);
>> + }
>> + return;
>> +}
>
> It seems to be possible to store RIP value *without displacement*
> into AX/CX and convert rip-relative instruction into AX/CX *relative* one.
> Example:
> c7 05 78 56 34 12 2a 00 00 00 movl $0x2a,0x12345678(%rip)
> converts to:
> c7 81 78 56 34 12 2a 00 00 00 movl $0x2a,0x12345678(%rcx)
>
> This way instruction size stays the same and you don't need
> to memmove immediate value.
Right, I agree there is a possibility of optimizing.
However, for ease of review, I think it's better to be
a separate patch.
Thank you,
--
Masami HIRAMATSU
Software Platform Research Dept. Linux Technology Center
Hitachi, Ltd., Yokohama Research Laboratory
E-mail: masami.hiramatsu.pt@...achi.com
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists