All of lore.kernel.org
 help / color / mirror / Atom feed
From: john.stultz@linaro.org (John Stultz)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 3/4] process: Add display of memory around registers when displaying regs.
Date: Mon, 13 Dec 2010 20:57:38 -0800	[thread overview]
Message-ID: <1292302659-1863-4-git-send-email-john.stultz@linaro.org> (raw)
In-Reply-To: <1292302659-1863-1-git-send-email-john.stultz@linaro.org>

From: San Mehat <san@google.com>

This is extremely useful in diagnosing remote crashes, and is based heavily
on original work by <md@google.com>.

Signed-off-by: San Mehat <san@google.com>
Cc: Michael Davidson <md@google.com>

[ARM] process: Use uber-safe probe_kernel_address() to read mem when dumping.

This prevents the dump from taking pagefaults / external aborts.

CC: Nicolas Pitre <nicolas.pitre@linaro.org>
CC: Russell King <linux@arm.linux.org.uk>
Signed-off-by: San Mehat <san@google.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
---
 arch/arm/kernel/process.c |   73 +++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 73 insertions(+), 0 deletions(-)

diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index e76fcaa..82c37fe 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -251,6 +251,77 @@ void machine_restart(char *cmd)
 	arm_pm_restart(reboot_mode, cmd);
 }
 
+/*
+ * dump a block of kernel memory from around the given address
+ */
+static void show_data(unsigned long addr, int nbytes, const char *name)
+{
+	int	i, j;
+	int	nlines;
+	u32	*p;
+
+	/*
+	 * don't attempt to dump non-kernel addresses or
+	 * values that are probably just small negative numbers
+	 */
+	if (addr < PAGE_OFFSET || addr > -256UL)
+		return;
+
+	printk("\n%s: %#lx:\n", name, addr);
+
+	/*
+	 * round address down to a 32 bit boundary
+	 * and always dump a multiple of 32 bytes
+	 */
+	p = (u32 *)(addr & ~(sizeof(u32) - 1));
+	nbytes += (addr & (sizeof(u32) - 1));
+	nlines = (nbytes + 31) / 32;
+
+
+	for (i = 0; i < nlines; i++) {
+		/*
+		 * just display low 16 bits of address to keep
+		 * each line of the dump < 80 characters
+		 */
+		printk("%04lx ", (unsigned long)p & 0xffff);
+		for (j = 0; j < 8; j++) {
+			u32	data;
+			if (probe_kernel_address(p, data)) {
+				printk(" ********");
+			} else {
+				printk(" %08x", data);
+			}
+			++p;
+		}
+		printk("\n");
+	}
+}
+
+static void show_extra_register_data(struct pt_regs *regs, int nbytes)
+{
+	mm_segment_t fs;
+
+	fs = get_fs();
+	set_fs(KERNEL_DS);
+	show_data(regs->ARM_pc - nbytes, nbytes * 2, "PC");
+	show_data(regs->ARM_lr - nbytes, nbytes * 2, "LR");
+	show_data(regs->ARM_sp - nbytes, nbytes * 2, "SP");
+	show_data(regs->ARM_ip - nbytes, nbytes * 2, "IP");
+	show_data(regs->ARM_fp - nbytes, nbytes * 2, "FP");
+	show_data(regs->ARM_r0 - nbytes, nbytes * 2, "R0");
+	show_data(regs->ARM_r1 - nbytes, nbytes * 2, "R1");
+	show_data(regs->ARM_r2 - nbytes, nbytes * 2, "R2");
+	show_data(regs->ARM_r3 - nbytes, nbytes * 2, "R3");
+	show_data(regs->ARM_r4 - nbytes, nbytes * 2, "R4");
+	show_data(regs->ARM_r5 - nbytes, nbytes * 2, "R5");
+	show_data(regs->ARM_r6 - nbytes, nbytes * 2, "R6");
+	show_data(regs->ARM_r7 - nbytes, nbytes * 2, "R7");
+	show_data(regs->ARM_r8 - nbytes, nbytes * 2, "R8");
+	show_data(regs->ARM_r9 - nbytes, nbytes * 2, "R9");
+	show_data(regs->ARM_r10 - nbytes, nbytes * 2, "R10");
+	set_fs(fs);
+}
+
 void __show_regs(struct pt_regs *regs)
 {
 	unsigned long flags;
@@ -310,6 +381,8 @@ void __show_regs(struct pt_regs *regs)
 		printk("Control: %08x%s\n", ctrl, buf);
 	}
 #endif
+
+	show_extra_register_data(regs, 128);
 }
 
 void show_regs(struct pt_regs * regs)
-- 
1.7.3.2.146.gca209

  parent reply	other threads:[~2010-12-14  4:57 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-12-14  4:57 [PATCH 0/4][RFC] Trivial ARM related Android patches John Stultz
2010-12-14  4:57 ` [PATCH 1/4] avoid mis-detecting some V7 cores in the decompressor John Stultz
2010-12-14  4:57 ` [PATCH 2/4] Optionally flush entire dcache from v6_dma_flush_range John Stultz
2010-12-14  9:30   ` Russell King - ARM Linux
2010-12-14 10:58   ` Catalin Marinas
2010-12-14  4:57 ` John Stultz [this message]
2010-12-14  9:34   ` [PATCH 3/4] process: Add display of memory around registers when displaying regs Russell King - ARM Linux
2010-12-14  4:57 ` [PATCH 4/4] Do not call flush_cache_user_range with mmap_sem held John Stultz
2010-12-14  9:30   ` Russell King - ARM Linux
2010-12-14 17:51     ` Catalin Marinas
2010-12-14 19:05       ` Russell King - ARM Linux
2010-12-14 21:08         ` Catalin Marinas
2011-04-04 13:27           ` Catalin Marinas
2011-04-04 13:37             ` Russell King - ARM Linux
2011-04-04 13:43               ` Catalin Marinas
2011-08-26  7:32                 ` Jiejing.Zhang 
2011-09-05 11:21                   ` Catalin Marinas
2010-12-14 18:18   ` Catalin Marinas

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1292302659-1863-4-git-send-email-john.stultz@linaro.org \
    --to=john.stultz@linaro.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.