From patchwork Fri Oct 20 16:57:58 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Khalid Aziz X-Patchwork-Id: 828735 X-Patchwork-Delegate: davem@davemloft.net Return-Path: X-Original-To: patchwork-incoming@ozlabs.org Delivered-To: patchwork-incoming@ozlabs.org Authentication-Results: ozlabs.org; spf=none (mailfrom) smtp.mailfrom=vger.kernel.org (client-ip=209.132.180.67; helo=vger.kernel.org; envelope-from=sparclinux-owner@vger.kernel.org; receiver=) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 3yJXBl3tpNz9t5l for ; Sat, 21 Oct 2017 04:01:35 +1100 (AEDT) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753057AbdJTRBV (ORCPT ); Fri, 20 Oct 2017 13:01:21 -0400 Received: from aserp1040.oracle.com ([141.146.126.69]:26831 "EHLO aserp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753378AbdJTQ7F (ORCPT ); Fri, 20 Oct 2017 12:59:05 -0400 Received: from aserv0021.oracle.com (aserv0021.oracle.com [141.146.126.233]) by aserp1040.oracle.com (Sentrion-MTA-4.3.2/Sentrion-MTA-4.3.2) with ESMTP id v9KGwSbs011565 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Fri, 20 Oct 2017 16:58:28 GMT Received: from userv0122.oracle.com (userv0122.oracle.com [156.151.31.75]) by aserv0021.oracle.com (8.14.4/8.14.4) with ESMTP id v9KGwRAQ025439 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Fri, 20 Oct 2017 16:58:28 GMT Received: from abhmp0019.oracle.com (abhmp0019.oracle.com [141.146.116.25]) by userv0122.oracle.com (8.14.4/8.14.4) with ESMTP id v9KGwRdL017981; Fri, 20 Oct 2017 16:58:27 GMT Received: from concerto.us.oracle.com (/24.9.64.241) by default (Oracle Beehive Gateway v4.0) with ESMTP ; Fri, 20 Oct 2017 09:58:26 -0700 From: Khalid Aziz To: davem@davemloft.net, dave.hansen@linux.intel.com Cc: Khalid Aziz , dan.carpenter@oracle.com, Liam.Howlett@oracle.com, mhocko@suse.com, mingo@kernel.org, paul.gortmaker@windriver.com, vegard.nossum@oracle.com, jane.chu@oracle.com, sparclinux@vger.kernel.org, linux-kernel@vger.kernel.org, Khalid Aziz Subject: [PATCH v9 05/10] sparc64: Add handler for "Memory Corruption Detected" trap Date: Fri, 20 Oct 2017 10:57:58 -0600 Message-Id: X-Mailer: git-send-email 2.11.0 In-Reply-To: References: In-Reply-To: References: X-Source-IP: aserv0021.oracle.com [141.146.126.233] Sender: sparclinux-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: sparclinux@vger.kernel.org M7 and newer processors add a "Memory corruption Detected" trap with the addition of ADI feature. This trap is vectored into kernel by HV through resumable error trap with error attribute for the resumable error set to 0x00000800. Signed-off-by: Khalid Aziz Cc: Khalid Aziz --- v7: - new patch split off from patch 4/4 in v6 arch/sparc/kernel/traps_64.c | 59 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 59 insertions(+) diff --git a/arch/sparc/kernel/traps_64.c b/arch/sparc/kernel/traps_64.c index 7775d011691c..565e3c2836aa 100644 --- a/arch/sparc/kernel/traps_64.c +++ b/arch/sparc/kernel/traps_64.c @@ -1825,6 +1825,7 @@ struct sun4v_error_entry { #define SUN4V_ERR_ATTRS_ASI 0x00000080 #define SUN4V_ERR_ATTRS_PRIV_REG 0x00000100 #define SUN4V_ERR_ATTRS_SPSTATE_MSK 0x00000600 +#define SUN4V_ERR_ATTRS_MCD 0x00000800 #define SUN4V_ERR_ATTRS_SPSTATE_SHFT 9 #define SUN4V_ERR_ATTRS_MODE_MSK 0x03000000 #define SUN4V_ERR_ATTRS_MODE_SHFT 24 @@ -2022,6 +2023,56 @@ static void sun4v_log_error(struct pt_regs *regs, struct sun4v_error_entry *ent, } } +/* Handle memory corruption detected error which is vectored in + * through resumable error trap. + */ +void do_mcd_err(struct pt_regs *regs, struct sun4v_error_entry ent) +{ + siginfo_t info; + + if (notify_die(DIE_TRAP, "MCD error", regs, 0, 0x34, + SIGSEGV) == NOTIFY_STOP) + return; + + if (regs->tstate & TSTATE_PRIV) { + /* MCD exception could happen because the task was + * running a system call with MCD enabled and passed a + * non-versioned pointer or pointer with bad version + * tag to the system call. In such cases, hypervisor + * places the address of offending instruction in the + * resumable error report. This is a deferred error, + * so the read/write that caused the trap was potentially + * retired long time back and we may have no choice + * but to send SIGSEGV to the process. + */ + const struct exception_table_entry *entry; + + entry = search_exception_tables(regs->tpc); + if (entry) { + /* Looks like a bad syscall parameter */ +#ifdef DEBUG_EXCEPTIONS + pr_emerg("Exception: PC<%016lx> faddr\n", + regs->tpc); + pr_emerg("EX_TABLE: insn<%016lx> fixup<%016lx>\n", + ent.err_raddr, entry->fixup); +#endif + regs->tpc = entry->fixup; + regs->tnpc = regs->tpc + 4; + return; + } + } + + /* Send SIGSEGV to the userspace process with the right signal + * code + */ + info.si_signo = SIGSEGV; + info.si_errno = 0; + info.si_code = SEGV_ADIDERR; + info.si_addr = (void __user *)ent.err_raddr; + info.si_trapno = 0; + force_sig_info(SIGSEGV, &info, current); +} + /* We run with %pil set to PIL_NORMAL_MAX and PSTATE_IE enabled in %pstate. * Log the event and clear the first word of the entry. */ @@ -2059,6 +2110,14 @@ void sun4v_resum_error(struct pt_regs *regs, unsigned long offset) goto out; } + /* If this is a memory corruption detected error vectored in + * by HV through resumable error trap, call the handler + */ + if (local_copy.err_attrs & SUN4V_ERR_ATTRS_MCD) { + do_mcd_err(regs, local_copy); + return; + } + sun4v_log_error(regs, &local_copy, cpu, KERN_ERR "RESUMABLE ERROR", &sun4v_resum_oflow_cnt);