From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx0a-001b2d01.pphosted.com (mx0b-001b2d01.pphosted.com [148.163.158.5]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id 40nNWX26rFzF29S for ; Fri, 18 May 2018 19:42:47 +1000 (AEST) Received: from pps.filterd (m0098421.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.16.0.22/8.16.0.22) with SMTP id w4I9cfXe124435 for ; Fri, 18 May 2018 05:42:44 -0400 Received: from e06smtp10.uk.ibm.com (e06smtp10.uk.ibm.com [195.75.94.106]) by mx0a-001b2d01.pphosted.com with ESMTP id 2j1tr0vqa1-1 (version=TLSv1.2 cipher=AES256-GCM-SHA384 bits=256 verify=NOT) for ; Fri, 18 May 2018 05:42:44 -0400 Received: from localhost by e06smtp10.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Fri, 18 May 2018 10:42:43 +0100 From: Vaibhav Jain To: linuxppc-dev@lists.ozlabs.org, Frederic Barrat Cc: Vaibhav Jain , Andrew Donnellan , Christophe Lombard , Philippe Bergheaud , "Alastair D'Silva" , Vaibhav Jain , stable@vger.kernel.org Subject: [PATCH-RESEND] cxl: Disable prefault_mode in Radix mode Date: Fri, 18 May 2018 15:12:23 +0530 Message-Id: <20180518094223.786-1-vaibhav@linux.vnet.ibm.com> List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , From: Vaibhav Jain Currently we see a kernel-oops reported on Power-9 while attaching a context to an AFU, with radix-mode and sysfs attr 'prefault_mode' set to anything other than 'none'. The backtrace of the oops is of this form: Unable to handle kernel paging request for data at address 0x00000080 Faulting instruction address: 0xc00800000bcf3b20 cpu 0x1: Vector: 300 (Data Access) at [c00000037f003800] pc: c00800000bcf3b20: cxl_load_segment+0x178/0x290 [cxl] lr: c00800000bcf39f0: cxl_load_segment+0x48/0x290 [cxl] sp: c00000037f003a80 msr: 9000000000009033 dar: 80 dsisr: 40000000 current = 0xc00000037f280000 paca = 0xc0000003ffffe600 softe: 3 irq_happened: 0x01 pid = 3529, comm = afp_no_int [c00000037f003af0] c00800000bcf4424 cxl_prefault+0xfc/0x248 [cxl] [c00000037f003b50] c00800000bcf8a40 process_element_entry_psl9+0xd8/0x1a0 [cxl] [c00000037f003b90] c00800000bcf944c cxl_attach_dedicated_process_psl9+0x44/0x130 [cxl] [c00000037f003bd0] c00800000bcf5448 native_attach_process+0xc0/0x130 [cxl] [c00000037f003c50] c00800000bcf16cc afu_ioctl+0x3f4/0x5e0 [cxl] [c00000037f003d00] c00000000039d98c do_vfs_ioctl+0xdc/0x890 [c00000037f003da0] c00000000039e1a8 ksys_ioctl+0x68/0xf0 [c00000037f003df0] c00000000039e270 sys_ioctl+0x40/0xa0 [c00000037f003e30] c00000000000b320 system_call+0x58/0x6c --- Exception: c01 (System Call) at 0000000010053bb0 The issue is caused as on Power-8 the AFU attr 'prefault_mode' was used to improve initial storage fault performance by prefaulting process segments. However on Power-9 with radix mode we don't have Storage-Segments that we can prefault. Also prefaulting process Pages will be too costly and fine-grained. Hence, since the prefaulting mechanism doesn't makes sense of radix-mode, this patch updates prefault_mode_store() to not allow any other value apart from CXL_PREFAULT_NONE when radix mode is enabled. Cc: Fixes: f24be42aab37 ("cxl: Add psl9 specific code") Signed-off-by: Vaibhav Jain --- Change-log: Resend -> Updated the commit description to add more info on the issue seen [Andrew] --- Documentation/ABI/testing/sysfs-class-cxl | 4 +++- drivers/misc/cxl/sysfs.c | 16 ++++++++++++---- 2 files changed, 15 insertions(+), 5 deletions(-) diff --git a/Documentation/ABI/testing/sysfs-class-cxl b/Documentation/ABI/testing/sysfs-class-cxl index 640f65e79ef1..267920a1874b 100644 --- a/Documentation/ABI/testing/sysfs-class-cxl +++ b/Documentation/ABI/testing/sysfs-class-cxl @@ -69,7 +69,9 @@ Date: September 2014 Contact: linuxppc-dev@lists.ozlabs.org Description: read/write Set the mode for prefaulting in segments into the segment table - when performing the START_WORK ioctl. Possible values: + when performing the START_WORK ioctl. Only applicable when + running under hashed page table mmu. + Possible values: none: No prefaulting (default) work_element_descriptor: Treat the work element descriptor as an effective address and diff --git a/drivers/misc/cxl/sysfs.c b/drivers/misc/cxl/sysfs.c index 4b5a4c5d3c01..629e2e156412 100644 --- a/drivers/misc/cxl/sysfs.c +++ b/drivers/misc/cxl/sysfs.c @@ -353,12 +353,20 @@ static ssize_t prefault_mode_store(struct device *device, struct cxl_afu *afu = to_cxl_afu(device); enum prefault_modes mode = -1; - if (!strncmp(buf, "work_element_descriptor", 23)) - mode = CXL_PREFAULT_WED; - if (!strncmp(buf, "all", 3)) - mode = CXL_PREFAULT_ALL; if (!strncmp(buf, "none", 4)) mode = CXL_PREFAULT_NONE; + else { + if (!radix_enabled()) { + + /* only allowed when not in radix mode */ + if (!strncmp(buf, "work_element_descriptor", 23)) + mode = CXL_PREFAULT_WED; + if (!strncmp(buf, "all", 3)) + mode = CXL_PREFAULT_ALL; + } else { + dev_err(device, "Cannot prefault with radix enabled\n"); + } + } if (mode == -1) return -EINVAL; -- 2.17.0