linux-kernel - [PATCH 2/5] x86: PAT: fix ambiguous paranoia check in pat

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-next>] [day] [month] [year] [list]

Message-ID: <20080610140518.GF5024@alberich.amd.com>
Date:	Tue, 10 Jun 2008 16:05:18 +0200
From:	Andreas Herrmann <andreas.herrmann3@....com>
To:	Thomas Gleixner <tglx@...utronix.de>, Ingo Molnar <mingo@...e.hu>,
	"H. Peter Anvin" <hpa@...or.com>
Cc:	linux-kernel@...r.kernel.org,
	Venkatesh Pallipadi <venkatesh.pallipadi@...el.com>,
	Suresh B Siddha <suresh.b.siddha@...el.com>
Subject: [PATCH 2/5] x86: PAT: fix ambiguous paranoia check in pat_init()

Starting with commit 8d4a4300854f3971502e81dacd930704cb88f606 (x86:
cleanup PAT cpu validation) the PAT CPU feature flag is not cleared
anymore. Now the error message

  "PAT enabled, but CPU feature cleared"

in pat_init() is misleading.

Furthermore the current code does not check for existence of the PAT
CPU feature flag if a CPU is whitelisted in validate_pat_support.

This patch clears pat_wc_enabled if boot CPU has no PAT feature flag
and adapts the paranoia check.

Signed-off-by: Andreas Herrmann <andreas.herrmann3@....com>
---
 arch/x86/kernel/cpu/addon_cpuid_features.c |    7 ++++---
 arch/x86/mm/pat.c                          |   11 ++++++-----
 2 files changed, 10 insertions(+), 8 deletions(-)

diff --git a/arch/x86/kernel/cpu/addon_cpuid_features.c b/arch/x86/kernel/cpu/addon_cpuid_features.c
index 0fbd062..2df461f 100644
--- a/arch/x86/kernel/cpu/addon_cpuid_features.c
+++ b/arch/x86/kernel/cpu/addon_cpuid_features.c
@@ -53,6 +53,9 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c)
 #ifdef CONFIG_X86_PAT
 void __cpuinit validate_pat_support(struct cpuinfo_x86 *c)
 {
+	if (!cpu_has_pat)
+		pat_disable("PAT not supported by CPU.");
+
 	switch (c->x86_vendor) {
 	case X86_VENDOR_INTEL:
 		if (c->x86 == 0xF || (c->x86 == 6 && c->x86_model >= 15))
@@ -64,8 +67,6 @@ void __cpuinit validate_pat_support(struct cpuinfo_x86 *c)
 		return;
 	}
 
-	pat_disable(cpu_has_pat ?
-		    "PAT disabled. Not yet verified on this CPU type." :
-		    "PAT not supported by CPU.");
+	pat_disable("PAT disabled. Not yet verified on this CPU type.");
 }
 #endif
diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c
index e83b770..8f288f8 100644
--- a/arch/x86/mm/pat.c
+++ b/arch/x86/mm/pat.c
@@ -76,14 +76,15 @@ void pat_init(void)
 		return;
 
 	/* Paranoia check. */
-	if (!cpu_has_pat) {
-		printk(KERN_ERR "PAT enabled, but CPU feature cleared\n");
+	if (!cpu_has_pat && boot_pat_state) {
 		/*
-		 * Panic if this happens on the secondary CPU, and we
+		 * If this happens we are on a secondary CPU, but
 		 * switched to PAT on the boot CPU. We have no way to
 		 * undo PAT.
-		*/
-		BUG_ON(boot_pat_state);
+		 */
+		printk(KERN_ERR "PAT enabled, "
+		       "but not supported by secondary CPU\n");
+		BUG();
 	}
 
 	/* Set PWT to Write-Combining. All other bits stay the same */
-- 
1.5.5.3



--
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/