linux-kernel - [RESEND PATCH v4] ARM: at91: fix at91_sysirq_mask

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: <1399646610-18816-1-git-send-email-boris.brezillon@free-electrons.com>
Date:	Fri,  9 May 2014 16:43:30 +0200
From:	Boris BREZILLON <boris.brezillon@...e-electrons.com>
To:	Bryan Evenson <bevenson@...inkcorp.com>
Cc:	Andrew Victor <linux@...im.org.za>,
	Nicolas Ferre <nicolas.ferre@...el.com>,
	Jean-Christophe Plagniol-Villard <plagnioj@...osoft.com>,
	linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
	Mark Roszko <mark.roszko@...il.com>,
	Johan Hovold <jhovold@...il.com>,
	Douglas Gilbert <dgilbert@...erlog.com>,
	Boris BREZILLON <boris.brezillon@...e-electrons.com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	<stable@...r.kernel.org>
Subject: [RESEND PATCH v4] ARM: at91: fix at91_sysirq_mask_rtc for sam9x5 SoCs

sam9x5 SoCs have the following errata:
"RTC: Interrupt Mask Register cannot be used
 Interrupt Mask Register read always returns 0."

Hence we should not rely on what IMR claims about already masked IRQs
and just disable all IRQs.

Cc: Andrew Morton <akpm@...ux-foundation.org>
Cc: <stable@...r.kernel.org> # v3.10+
Signed-off-by: Boris BREZILLON <boris.brezillon@...e-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@...el.com>
---
Hello,

Again, sorry for the noise, but I forgot to add Nicolas' ack and to keep
Andrew and the stable ML in cc.

Best Regards,

Boris

Changes since v3:
- fix commit message and comment
Changes since v2:
- removed unused variable 'mask'
Changes since v1:
- use a macro to define IRQs bitmask
- read IMR register to ensure the write to IDR has been flushed
- quote atmel's datasheet errata in commit message
- comment the code to describe why we're not using IMR to disable
  the interrupts

 arch/arm/mach-at91/sysirq_mask.c | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)

diff --git a/arch/arm/mach-at91/sysirq_mask.c b/arch/arm/mach-at91/sysirq_mask.c
index 2ba694f..f8bc351 100644
--- a/arch/arm/mach-at91/sysirq_mask.c
+++ b/arch/arm/mach-at91/sysirq_mask.c
@@ -25,24 +25,28 @@
 
 #include "generic.h"
 
-#define AT91_RTC_IDR	0x24	/* Interrupt Disable Register */
-#define AT91_RTC_IMR	0x28	/* Interrupt Mask Register */
+#define AT91_RTC_IDR		0x24	/* Interrupt Disable Register */
+#define AT91_RTC_IMR		0x28	/* Interrupt Mask Register */
+#define AT91_RTC_IRQ_MASK	0x1f	/* Available IRQs mask */
 
 void __init at91_sysirq_mask_rtc(u32 rtc_base)
 {
 	void __iomem *base;
-	u32 mask;
 
 	base = ioremap(rtc_base, 64);
 	if (!base)
 		return;
 
-	mask = readl_relaxed(base + AT91_RTC_IMR);
-	if (mask) {
-		pr_info("AT91: Disabling rtc irq\n");
-		writel_relaxed(mask, base + AT91_RTC_IDR);
-		(void)readl_relaxed(base + AT91_RTC_IMR);	/* flush */
-	}
+	/*
+	 * sam9x5 SoCs have the following errata:
+	 * "RTC: Interrupt Mask Register cannot be used
+	 *  Interrupt Mask Register read always returns 0."
+	 *
+	 * Hence we're not relying on IMR values to disable
+	 * interrupts.
+	 */
+	writel_relaxed(AT91_RTC_IRQ_MASK, base + AT91_RTC_IDR);
+	(void)readl_relaxed(base + AT91_RTC_IMR);	/* flush */
 
 	iounmap(base);
 }
-- 
1.8.3.2

--
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/