Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755694AbaA1Vgq (ORCPT ); Tue, 28 Jan 2014 16:36:46 -0500 Received: from userp1040.oracle.com ([156.151.31.81]:20080 "EHLO userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754497AbaA1Vgo (ORCPT ); Tue, 28 Jan 2014 16:36:44 -0500 Message-ID: <52E8230F.7080300@oracle.com> Date: Tue, 28 Jan 2014 16:37:19 -0500 From: Boris Ostrovsky User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130805 Thunderbird/17.0.8 MIME-Version: 1.0 To: Borislav Petkov CC: "H. Peter Anvin" , linux-kernel@vger.kernel.org, Ingo Molnar Subject: Re: AMD microcode loading broken on 32 bit References: <20140121161438.GD9004@pd.tnic> <52DEB4A9.4070707@oracle.com> <20140121182554.GE9004@pd.tnic> <52E15AAE.4090106@oracle.com> <20140123192939.GB11190@pd.tnic> <52E17087.2060301@oracle.com> <20140128162431.GH815@pd.tnic> <52E8168D.9060805@oracle.com> <20140128205246.GL815@pd.tnic> <52E81B90.8040604@oracle.com> <20140128213030.GM815@pd.tnic> In-Reply-To: <20140128213030.GM815@pd.tnic> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit X-Source-IP: acsinet22.oracle.com [141.146.126.238] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 01/28/2014 04:30 PM, Borislav Petkov wrote: > On Tue, Jan 28, 2014 at 04:05:20PM -0500, Boris Ostrovsky wrote: >> Yes. This was a bug in our build environment. It is fixed now but I >> put this bug back when I was testing your patch. > Right, that was the deal. Oh well, we'd need to verify the container > too, I guess: I'll give it a spin. Apparently the problem is currently in install_equiv_cpu_table() when we are trying to index into buf and buf (which is container) doesn't appear to be there. Are you sure that container = (u8 *)(__va((u32)relocated_ramdisk) + ((u32)container - boot_params.hdr.ramdisk_image)); in save_microcode_in_initrd_amd() always results in a valid pointer? It is non-NULL but it points to address that looks to be not mapped. -boris > > --- > diff --git a/arch/x86/kernel/cpu/microcode/amd_early.c b/arch/x86/kernel/cpu/microcode/amd_early.c > index 8384c0fa206f..c568b6c9af95 100644 > --- a/arch/x86/kernel/cpu/microcode/amd_early.c > +++ b/arch/x86/kernel/cpu/microcode/amd_early.c > @@ -37,10 +37,12 @@ static __initdata char ucode_path[] = "kernel/x86/microcode/AuthenticAMD.bin"; > > static struct cpio_data __init find_ucode_in_initrd(void) > { > + struct cpio_data ret; > long offset = 0; > char *path; > void *start; > size_t size; > + u32 *hdr; > > #ifdef CONFIG_X86_32 > struct boot_params *p; > @@ -59,7 +61,18 @@ static struct cpio_data __init find_ucode_in_initrd(void) > size = boot_params.hdr.ramdisk_size; > #endif > > - return find_cpio_data(path, start, size, &offset); > + ret = find_cpio_data(path, start, size, &offset); > + > + /* Verify we didn't get some garbage from the initrd. */ > + hdr = (u32 *)ret.data; > + > + if (hdr[0] != UCODE_MAGIC || > + hdr[1] != UCODE_EQUIV_CPU_TABLE_TYPE || > + hdr[2] == 0) { > + ret.data = NULL; > + ret.size = 0; > + } > + return ret; > } > > static size_t compute_container_size(u8 *data, u32 total_size) > @@ -359,7 +372,7 @@ int __init save_microcode_in_initrd_amd(void) > pr_info("microcode: updated early to new patch_level=0x%08x\n", > ucode_new_rev); > > - if (!container) > + if (!container || !ucode_cpio.data) > return -EINVAL; > > eax = cpuid_eax(0x00000001); > -- > -- 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/