Revision 85bd839983778fcd0c1c043327b14a046e979b39 authored by Gu Zheng on 10 June 2015, 18:14:43 UTC, committed by Linus Torvalds on 10 June 2015, 23:43:43 UTC
Izumi found the following oops when hot re-adding a node:

    BUG: unable to handle kernel paging request at ffffc90008963690
    IP: __wake_up_bit+0x20/0x70
    Oops: 0000 [#1] SMP
    CPU: 68 PID: 1237 Comm: rs:main Q:Reg Not tainted 4.1.0-rc5 #80
    Hardware name: FUJITSU PRIMEQUEST2800E/SB, BIOS PRIMEQUEST 2000 Series BIOS Version 1.87 04/28/2015
    task: ffff880838df8000 ti: ffff880017b94000 task.ti: ffff880017b94000
    RIP: 0010:[<ffffffff810dff80>]  [<ffffffff810dff80>] __wake_up_bit+0x20/0x70
    RSP: 0018:ffff880017b97be8  EFLAGS: 00010246
    RAX: ffffc90008963690 RBX: 00000000003c0000 RCX: 000000000000a4c9
    RDX: 0000000000000000 RSI: ffffea101bffd500 RDI: ffffc90008963648
    RBP: ffff880017b97c08 R08: 0000000002000020 R09: 0000000000000000
    R10: 0000000000000000 R11: 0000000000000000 R12: ffff8a0797c73800
    R13: ffffea101bffd500 R14: 0000000000000001 R15: 00000000003c0000
    FS:  00007fcc7ffff700(0000) GS:ffff880874800000(0000) knlGS:0000000000000000
    CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
    CR2: ffffc90008963690 CR3: 0000000836761000 CR4: 00000000001407e0
    Call Trace:
      unlock_page+0x6d/0x70
      generic_write_end+0x53/0xb0
      xfs_vm_write_end+0x29/0x80 [xfs]
      generic_perform_write+0x10a/0x1e0
      xfs_file_buffered_aio_write+0x14d/0x3e0 [xfs]
      xfs_file_write_iter+0x79/0x120 [xfs]
      __vfs_write+0xd4/0x110
      vfs_write+0xac/0x1c0
      SyS_write+0x58/0xd0
      system_call_fastpath+0x12/0x76
    Code: 5d c3 66 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 55 48 89 e5 48 83 ec 20 65 48 8b 04 25 28 00 00 00 48 89 45 f8 31 c0 48 8d 47 48 <48> 39 47 48 48 c7 45 e8 00 00 00 00 48 c7 45 f0 00 00 00 00 48
    RIP  [<ffffffff810dff80>] __wake_up_bit+0x20/0x70
     RSP <ffff880017b97be8>
    CR2: ffffc90008963690

Reproduce method (re-add a node)::
  Hot-add nodeA --> remove nodeA --> hot-add nodeA (panic)

This seems an use-after-free problem, and the root cause is
zone->wait_table was not set to *NULL* after free it in
try_offline_node.

When hot re-add a node, we will reuse the pgdat of it, so does the zone
struct, and when add pages to the target zone, it will init the zone
first (including the wait_table) if the zone is not initialized.  The
judgement of zone initialized is based on zone->wait_table:

	static inline bool zone_is_initialized(struct zone *zone)
	{
		return !!zone->wait_table;
	}

so if we do not set the zone->wait_table to *NULL* after free it, the
memory hotplug routine will skip the init of new zone when hot re-add
the node, and the wait_table still points to the freed memory, then we
will access the invalid address when trying to wake up the waiting
people after the i/o operation with the page is done, such as mentioned
above.

Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com>
Reported-by: Taku Izumi <izumi.taku@jp.fujitsu.com>
Reviewed by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Tang Chen <tangchen@cn.fujitsu.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent 5879ae5
Raw File
dtc2278.c
/*
 *  Copyright (C) 1996  Linus Torvalds & author (see below)
 */

#include <linux/module.h>
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/delay.h>
#include <linux/timer.h>
#include <linux/mm.h>
#include <linux/ioport.h>
#include <linux/blkdev.h>
#include <linux/ide.h>
#include <linux/init.h>

#include <asm/io.h>

#define DRV_NAME "dtc2278"

/*
 * Changing this #undef to #define may solve start up problems in some systems.
 */
#undef ALWAYS_SET_DTC2278_PIO_MODE

/*
 * From: andy@cercle.cts.com (Dyan Wile)
 *
 * Below is a patch for DTC-2278 - alike software-programmable controllers
 * The code enables the secondary IDE controller and the PIO4 (3?) timings on
 * the primary (EIDE). You may probably have to enable the 32-bit support to
 * get the full speed. You better get the disk interrupts disabled ( hdparm -u0
 * /dev/hd.. ) for the drives connected to the EIDE interface. (I get my
 * filesystem  corrupted with -u1, but under heavy disk load only :-)
 *
 * This card is now forced to use the "serialize" feature,
 * and irq-unmasking is disallowed.  If io_32bit is enabled,
 * it must be done for BOTH drives on each interface.
 *
 * This code was written for the DTC2278E, but might work with any of these:
 *
 * DTC2278S has only a single IDE interface.
 * DTC2278D has two IDE interfaces and is otherwise identical to the S version.
 * DTC2278E also has serial ports and a printer port
 * DTC2278EB: has onboard BIOS, and "works like a charm" -- Kent Bradford <kent@theory.caltech.edu>
 *
 * There may be a fourth controller type. The S and D versions use the
 * Winbond chip, and I think the E version does also.
 *
 */

static void sub22 (char b, char c)
{
	int i;

	for(i = 0; i < 3; ++i) {
		inb(0x3f6);
		outb_p(b,0xb0);
		inb(0x3f6);
		outb_p(c,0xb4);
		inb(0x3f6);
		if(inb(0xb4) == c) {
			outb_p(7,0xb0);
			inb(0x3f6);
			return;	/* success */
		}
	}
}

static DEFINE_SPINLOCK(dtc2278_lock);

static void dtc2278_set_pio_mode(ide_hwif_t *hwif, ide_drive_t *drive)
{
	unsigned long flags;

	if (drive->pio_mode >= XFER_PIO_3) {
		spin_lock_irqsave(&dtc2278_lock, flags);
		/*
		 * This enables PIO mode4 (3?) on the first interface
		 */
		sub22(1,0xc3);
		sub22(0,0xa0);
		spin_unlock_irqrestore(&dtc2278_lock, flags);
	} else {
		/* we don't know how to set it back again.. */
		/* Actually we do - there is a data sheet available for the
		   Winbond but does anyone actually care */
	}
}

static const struct ide_port_ops dtc2278_port_ops = {
	.set_pio_mode		= dtc2278_set_pio_mode,
};

static const struct ide_port_info dtc2278_port_info __initconst = {
	.name			= DRV_NAME,
	.chipset		= ide_dtc2278,
	.port_ops		= &dtc2278_port_ops,
	.host_flags		= IDE_HFLAG_SERIALIZE |
				  IDE_HFLAG_NO_UNMASK_IRQS |
				  IDE_HFLAG_IO_32BIT |
				  /* disallow ->io_32bit changes */
				  IDE_HFLAG_NO_IO_32BIT |
				  IDE_HFLAG_NO_DMA |
				  IDE_HFLAG_DTC2278,
	.pio_mask		= ATA_PIO4,
};

static int __init dtc2278_probe(void)
{
	unsigned long flags;

	local_irq_save(flags);
	/*
	 * This enables the second interface
	 */
	outb_p(4,0xb0);
	inb(0x3f6);
	outb_p(0x20,0xb4);
	inb(0x3f6);
#ifdef ALWAYS_SET_DTC2278_PIO_MODE
	/*
	 * This enables PIO mode4 (3?) on the first interface
	 * and may solve start-up problems for some people.
	 */
	sub22(1,0xc3);
	sub22(0,0xa0);
#endif
	local_irq_restore(flags);

	return ide_legacy_device_add(&dtc2278_port_info, 0);
}

static bool probe_dtc2278;

module_param_named(probe, probe_dtc2278, bool, 0);
MODULE_PARM_DESC(probe, "probe for DTC2278xx chipsets");

static int __init dtc2278_init(void)
{
	if (probe_dtc2278 == 0)
		return -ENODEV;

	if (dtc2278_probe()) {
		printk(KERN_ERR "dtc2278: ide interfaces already in use!\n");
		return -EBUSY;
	}
	return 0;
}

module_init(dtc2278_init);

MODULE_AUTHOR("See Local File");
MODULE_DESCRIPTION("support of DTC-2278 VLB IDE chipsets");
MODULE_LICENSE("GPL");
back to top