linux-kernel - Re: [tip:x86/cpu] x86/cpu: Strip any /proc/ cpuinfo model name field whitespace

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

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <2A88ADE0-90EF-4A94-BF4D-463E478B2FDA@zytor.com>
Date:	Thu, 28 May 2015 09:57:15 -0700
From:	"H. Peter Anvin" <hpa@...or.com>
To:	Borislav Petkov <bp@...en8.de>, Prarit Bhargava <prarit@...hat.com>
CC:	Joe Perches <joe@...ches.com>, luto@...capital.net,
	peterz@...radead.org, dvlasenk@...hat.com,
	torvalds@...ux-foundation.org, imammedo@...hat.com,
	brgerst@...il.com, mingo@...nel.org, dave.hansen@...ux.intel.com,
	fenghua.yu@...el.com, linux-kernel@...r.kernel.org,
	tglx@...utronix.de, bp@...e.de, linux-tip-commits@...r.kernel.org
Subject: Re: [tip:x86/cpu] x86/cpu: Strip any /proc/ cpuinfo model name field whitespace

Why?!

We are taking about 48 bytes run once per cpu.  It isn't worth it to optimize, in fact the extra code size hurts more.

On May 28, 2015 5:58:19 AM PDT, Borislav Petkov <bp@...en8.de> wrote:
>On Thu, May 28, 2015 at 01:32:29PM +0200, Borislav Petkov wrote:
>> +	while (*p) {
>> +		/* Note the last non-whitespace index */
>> +		if (!isspace(*p))
>> +			s = q;
>> +
>> +		*q++ = *p++;
>
>This should be optimized to not copy if there's no preceding whitespace
>and p == q:
>
>From: Borislav Petkov <bp@...e.de>
>Date: Tue, 26 May 2015 10:28:17 +0200
>Subject: [PATCH] x86/cpu: Trim model id whitespace
>
>We did try trimming whitespace surrounding the 'model name' field
>in /proc/cpuinfo since reportedly some userspace uses it in string
>comparisons and there were discrepancies:
>
>[thetango@...rit ~]# grep "^model name" /proc/cpuinfo | uniq -c | sed
>'s/\ /_/g'
>  ______1_model_name      :_AMD_Opteron(TM)_Processor_6272
>_____63_model_name     
>:_AMD_Opteron(TM)_Processor_6272_________________
>
>However, there were issues with overlapping buffers, string sizes and
>non-byte-sized copies in the previous proposed solutions; see Link tags
>below for the whole farce.
>
>So, instead of diddling with this more, let's simply extend what was
>there originally with trimming any present trailing whitespace. Final
>result is really simple and obvious.
>
>Testing with the most insane model IDs qemu can generate, looks good:
>
>  .model_id = "            My funny model ID CPU          ",
>  ______4_model_name      :_My_funny_model_ID_CPU
>
>  .model_id = "My funny model ID CPU          ",
>  ______4_model_name      :_My_funny_model_ID_CPU
>
>  .model_id = "            My funny model ID CPU",
>  ______4_model_name      :_My_funny_model_ID_CPU
>
>  .model_id = "            ",
>  ______4_model_name      :__
>
>  .model_id = "",
>  ______4_model_name      :_15/02
>
>Cc: Andy Lutomirski <luto@...capital.net>
>Cc: Brian Gerst <brgerst@...il.com>
>Cc: Dave Hansen <dave.hansen@...ux.intel.com>
>Cc: Denys Vlasenko <dvlasenk@...hat.com>
>Cc: Fenghua Yu <fenghua.yu@...el.com>
>Cc: H. Peter Anvin <hpa@...or.com>
>Cc: Igor Mammedov <imammedo@...hat.com>
>Cc: Linus Torvalds <torvalds@...ux-foundation.org>
>Cc: Peter Zijlstra <peterz@...radead.org>
>Cc: Thomas Gleixner <tglx@...utronix.de>
>Link:
>http://lkml.kernel.org/r/1432050210-32036-1-git-send-email-prarit@redhat.com
>Link:
>http://lkml.kernel.org/r/1432628901-18044-15-git-send-email-bp@alien8.de
>Signed-off-by: Borislav Petkov <bp@...e.de>
>---
> arch/x86/kernel/cpu/common.c | 27 ++++++++++++++++++++++-----
> 1 file changed, 22 insertions(+), 5 deletions(-)
>
>diff --git a/arch/x86/kernel/cpu/common.c
>b/arch/x86/kernel/cpu/common.c
>index 41a8e9cb30bc..18120a33a2c1 100644
>--- a/arch/x86/kernel/cpu/common.c
>+++ b/arch/x86/kernel/cpu/common.c
>@@ -5,6 +5,7 @@
> #include <linux/module.h>
> #include <linux/percpu.h>
> #include <linux/string.h>
>+#include <linux/ctype.h>
> #include <linux/delay.h>
> #include <linux/sched.h>
> #include <linux/init.h>
>@@ -419,6 +420,7 @@ static const struct cpu_dev
>*cpu_devs[X86_VENDOR_NUM] = {};
> static void get_model_name(struct cpuinfo_x86 *c)
> {
> 	unsigned int *v;
>+	char *p, *q, *s;
> 
> 	if (c->extended_cpuid_level < 0x80000004)
> 		return;
>@@ -429,11 +431,26 @@ static void get_model_name(struct cpuinfo_x86 *c)
> 	cpuid(0x80000004, &v[8], &v[9], &v[10], &v[11]);
> 	c->x86_model_id[48] = 0;
> 
>-	/*
>-	 * Remove leading whitespace on Intel processors and trailing
>-	 * whitespace on AMD processors.
>-	 */
>-	memmove(c->x86_model_id, strim(c->x86_model_id), 48);
>+	/* Trim whitespace */
>+	p = q = s = &c->x86_model_id[0];
>+
>+	while (*p == ' ')
>+		p++;
>+
>+	while (*p) {
>+		/* Note the last non-whitespace index: */
>+		if (!isspace(*p))
>+			s = q;
>+
>+		/* Only copy if p advanced due to whitespace: */
>+		if (p != q)
>+			*q = *p;
>+
>+		p++;
>+		q++;
>+	}
>+
>+	*(s + 1) = '\0';
> }
> 
> void cpu_detect_cache_sizes(struct cpuinfo_x86 *c)

-- 
Sent from my mobile phone.  Please pardon brevity and lack of formatting.
--
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/