Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261171AbVCaIQ1 (ORCPT ); Thu, 31 Mar 2005 03:16:27 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261173AbVCaIQ1 (ORCPT ); Thu, 31 Mar 2005 03:16:27 -0500 Received: from e35.co.us.ibm.com ([32.97.110.133]:35767 "EHLO e35.co.us.ibm.com") by vger.kernel.org with ESMTP id S261171AbVCaIP5 (ORCPT ); Thu, 31 Mar 2005 03:15:57 -0500 Date: Thu, 31 Mar 2005 13:47:14 +0530 From: Prasanna S Panchamukhi To: SystemTAP , akpm@osdl.org, Andi Kleen , davem@davemloft.net, ananth@in.ibm.com, linux-kernel@vger.kernel.org Subject: [PATCH] Kprobes: Allow/deny probes on int3/breakpoint instruction? Message-ID: <20050331081714.GA32299@in.ibm.com> Reply-To: prasanna@in.ibm.com References: <20050331080714.GB31660@in.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20050331080714.GB31660@in.ibm.com> User-Agent: Mutt/1.4i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 6675 Lines: 176 Hi, Kprobes did an improper exit when a probe is inserted on an int3 instruction. In case of normal execution of int3/breakpoint instruction, it oops!. Probe on an int3 instruction was not handled properly by the kprobes, it generated faults after oops! doing an improper exit with holding the lock. This fix employes a bit different method to handle probe on an int3/breakpoint instruction. On execution of an int3/breakpoint instruction (placed by kprobe), kprobes_handler() is called which sets it for single stepping in-line(it does not matter whether we single step out-of-line/inline since the single stepping instruction is same). Now it single steps on int3/breakpoint instruction here, entering kprobes_handler() once again. Kprobes now check's the status that it is single stepping and avoids the recursion. It runs down through the trap handler and oops messages is seen on the console since it executed int3/breakpoint instruction. Here the kprobes single stepping handler never gets called. Is this behaviour acceptable ? Or should we avoid putting probes on an int3 /breakpoint instruction ? How should it handle such situations? Below is the patch to allow probes on an int3/breakpoint instruction. This patch fixes the above problem by doing a proper exit while avoiding recursion. Any pointers/suggestions on the above issues will be helpful. Signed-off-by: Prasanna S Panchamukhi --- linux-2.6.12-rc1-prasanna/arch/i386/kernel/kprobes.c | 12 +++++++++++- linux-2.6.12-rc1-prasanna/arch/ppc64/kernel/kprobes.c | 12 +++++++++++- linux-2.6.12-rc1-prasanna/arch/sparc64/kernel/kprobes.c | 16 ++++++++++++++-- linux-2.6.12-rc1-prasanna/arch/x86_64/kernel/kprobes.c | 13 +++++++++++-- 4 files changed, 47 insertions(+), 6 deletions(-) diff -puN arch/i386/kernel/kprobes.c~kprobes-allow-probes-on-int3 arch/i386/kernel/kprobes.c --- linux-2.6.12-rc1/arch/i386/kernel/kprobes.c~kprobes-allow-probes-on-int3 2005-03-30 16:47:42.000000000 +0530 +++ linux-2.6.12-rc1-prasanna/arch/i386/kernel/kprobes.c 2005-03-30 16:51:43.000000000 +0530 @@ -84,7 +84,11 @@ static inline void prepare_singlestep(st { regs->eflags |= TF_MASK; regs->eflags &= ~IF_MASK; - regs->eip = (unsigned long)&p->ainsn.insn; + /*single step inline if the instruction is an int3*/ + if (p->opcode == BREAKPOINT_INSTRUCTION) + regs->eip = (unsigned long)p->addr; + else + regs->eip = (unsigned long)&p->ainsn.insn; } /* @@ -117,6 +121,12 @@ static int kprobe_handler(struct pt_regs Disarm the probe we just hit, and ignore it. */ p = get_kprobe(addr); if (p) { + if (kprobe_status == KPROBE_HIT_SS) { + regs->eflags &= ~TF_MASK; + regs->eflags |= kprobe_saved_eflags; + unlock_kprobes(); + goto no_kprobe; + } disarm_kprobe(p, regs); ret = 1; } else { diff -puN arch/x86_64/kernel/kprobes.c~kprobes-allow-probes-on-int3 arch/x86_64/kernel/kprobes.c --- linux-2.6.12-rc1/arch/x86_64/kernel/kprobes.c~kprobes-allow-probes-on-int3 2005-03-30 20:55:23.000000000 +0530 +++ linux-2.6.12-rc1-prasanna/arch/x86_64/kernel/kprobes.c 2005-03-31 12:19:53.000000000 +0530 @@ -108,8 +108,11 @@ static void prepare_singlestep(struct kp { regs->eflags |= TF_MASK; regs->eflags &= ~IF_MASK; - - regs->rip = (unsigned long)p->ainsn.insn; + /*single step inline if the instruction is an int3*/ + if (p->opcode == BREAKPOINT_INSTRUCTION) + regs->rip = (unsigned long)p->addr; + else + regs->rip = (unsigned long)p->ainsn.insn; } /* @@ -131,6 +134,12 @@ int kprobe_handler(struct pt_regs *regs) Disarm the probe we just hit, and ignore it. */ p = get_kprobe(addr); if (p) { + if (kprobe_status == KPROBE_HIT_SS) { + regs->eflags &= ~TF_MASK; + regs->eflags |= kprobe_saved_rflags; + unlock_kprobes(); + goto no_kprobe; + } disarm_kprobe(p, regs); ret = 1; } else { diff -puN arch/ppc64/kernel/kprobes.c~kprobes-allow-probes-on-int3 arch/ppc64/kernel/kprobes.c --- linux-2.6.12-rc1/arch/ppc64/kernel/kprobes.c~kprobes-allow-probes-on-int3 2005-03-30 21:03:14.000000000 +0530 +++ linux-2.6.12-rc1-prasanna/arch/ppc64/kernel/kprobes.c 2005-03-31 10:46:16.000000000 +0530 @@ -71,7 +71,11 @@ static inline void disarm_kprobe(struct static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs) { regs->msr |= MSR_SE; - regs->nip = (unsigned long)&p->ainsn.insn; + /*single step inline if it a breakpoint instruction*/ + if (p->opcode == BREAKPOINT_INSTRUCTION) + regs->nip = (unsigned long)p->addr; + else + regs->nip = (unsigned long)&p->ainsn.insn; } static inline int kprobe_handler(struct pt_regs *regs) @@ -89,6 +93,12 @@ static inline int kprobe_handler(struct Disarm the probe we just hit, and ignore it. */ p = get_kprobe(addr); if (p) { + if (kprobe_status == KPROBE_HIT_SS) { + regs->msr &= ~MSR_SE; + regs->msr |= kprobe_saved_msr; + unlock_kprobes(); + goto no_kprobe; + } disarm_kprobe(p, regs); ret = 1; } else { diff -puN arch/sparc64/kernel/kprobes.c~kprobes-allow-probes-on-int3 arch/sparc64/kernel/kprobes.c --- linux-2.6.12-rc1/arch/sparc64/kernel/kprobes.c~kprobes-allow-probes-on-int3 2005-03-31 10:34:50.000000000 +0530 +++ linux-2.6.12-rc1-prasanna/arch/sparc64/kernel/kprobes.c 2005-03-31 10:44:19.000000000 +0530 @@ -68,8 +68,14 @@ static inline void prepare_singlestep(st current_kprobe_orig_tstate_pil = (regs->tstate & TSTATE_PIL); regs->tstate |= TSTATE_PIL; - regs->tpc = (unsigned long) &p->ainsn.insn[0]; - regs->tnpc = (unsigned long) &p->ainsn.insn[1]; + /*single step inline, if it a breakpoint instruction*/ + if (p->opcode == BREAKPOINT_INSTRUCTION) { + regs->tpc = (unsigned long) p->addr; + regs->tnpc = current_kprobe_orig_tnpc; + } else { + regs->tpc = (unsigned long) &p->ainsn.insn[0]; + regs->tnpc = (unsigned long) &p->ainsn.insn[1]; + } } static inline void disarm_kprobe(struct kprobe *p, struct pt_regs *regs) @@ -97,6 +103,12 @@ static int kprobe_handler(struct pt_regs */ p = get_kprobe(addr); if (p) { + if (kprobe_status == KPROBE_HIT_SS) { + regs->tstate = ((regs->tstate & ~TSTATE_PIL) | + current_kprobe_orig_tstate_pil); + unlock_kprobes(); + goto no_kprobe; + } disarm_kprobe(p, regs); ret = 1; } else { _ Thanks Prasanna -- Prasanna S Panchamukhi Linux Technology Center India Software Labs, IBM Bangalore Ph: 91-80-25044636 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/