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
cy82c693.c
/*
 *  Copyright (C) 1998-2000 Andreas S. Krebs (akrebs@altavista.net), Maintainer
 *  Copyright (C) 1998-2002 Andre Hedrick <andre@linux-ide.org>, Integrator
 *  Copyright (C) 2007-2011 Bartlomiej Zolnierkiewicz
 *
 * CYPRESS CY82C693 chipset IDE controller
 *
 * The CY82C693 chipset is used on Digital's PC-Alpha 164SX boards.
 */

#include <linux/module.h>
#include <linux/types.h>
#include <linux/pci.h>
#include <linux/ide.h>
#include <linux/init.h>

#include <asm/io.h>

#define DRV_NAME "cy82c693"

/*
 *	NOTE: the value for busmaster timeout is tricky and I got it by
 *	trial and error!  By using a to low value will cause DMA timeouts
 *	and drop IDE performance, and by using a to high value will cause
 *	audio playback to scatter.
 *	If you know a better value or how to calc it, please let me know.
 */

/* twice the value written in cy82c693ub datasheet */
#define BUSMASTER_TIMEOUT	0x50
/*
 * the value above was tested on my machine and it seems to work okay
 */

/* here are the offset definitions for the registers */
#define CY82_IDE_CMDREG		0x04
#define CY82_IDE_ADDRSETUP	0x48
#define CY82_IDE_MASTER_IOR	0x4C
#define CY82_IDE_MASTER_IOW	0x4D
#define CY82_IDE_SLAVE_IOR	0x4E
#define CY82_IDE_SLAVE_IOW	0x4F
#define CY82_IDE_MASTER_8BIT	0x50
#define CY82_IDE_SLAVE_8BIT	0x51

#define CY82_INDEX_PORT		0x22
#define CY82_DATA_PORT		0x23

#define CY82_INDEX_CHANNEL0	0x30
#define CY82_INDEX_CHANNEL1	0x31
#define CY82_INDEX_TIMEOUT	0x32

/*
 * set DMA mode a specific channel for CY82C693
 */

static void cy82c693_set_dma_mode(ide_hwif_t *hwif, ide_drive_t *drive)
{
	const u8 mode = drive->dma_mode;
	u8 single = (mode & 0x10) >> 4, index = 0, data = 0;

	index = hwif->channel ? CY82_INDEX_CHANNEL1 : CY82_INDEX_CHANNEL0;

	data = (mode & 3) | (single << 2);

	outb(index, CY82_INDEX_PORT);
	outb(data, CY82_DATA_PORT);

	/*
	 * note: below we set the value for Bus Master IDE TimeOut Register
	 * I'm not absolutely sure what this does, but it solved my problem
	 * with IDE DMA and sound, so I now can play sound and work with
	 * my IDE driver at the same time :-)
	 *
	 * If you know the correct (best) value for this register please
	 * let me know - ASK
	 */

	data = BUSMASTER_TIMEOUT;
	outb(CY82_INDEX_TIMEOUT, CY82_INDEX_PORT);
	outb(data, CY82_DATA_PORT);
}

static void cy82c693_set_pio_mode(ide_hwif_t *hwif, ide_drive_t *drive)
{
	struct pci_dev *dev = to_pci_dev(hwif->dev);
	int bus_speed = ide_pci_clk ? ide_pci_clk : 33;
	const unsigned long T = 1000000 / bus_speed;
	unsigned int addrCtrl;
	struct ide_timing t;
	u8 time_16, time_8;

	/* select primary or secondary channel */
	if (drive->dn > 1) {  /* drive is on the secondary channel */
		dev = pci_get_slot(dev->bus, dev->devfn+1);
		if (!dev) {
			printk(KERN_ERR "%s: tune_drive: "
				"Cannot find secondary interface!\n",
				drive->name);
			return;
		}
	}

	ide_timing_compute(drive, drive->pio_mode, &t, T, 1);

	time_16 = clamp_val(t.recover - 1, 0, 15) |
		  (clamp_val(t.active - 1, 0, 15) << 4);
	time_8 = clamp_val(t.act8b - 1, 0, 15) |
		 (clamp_val(t.rec8b - 1, 0, 15) << 4);

	/* now let's write  the clocks registers */
	if ((drive->dn & 1) == 0) {
		/*
		 * set master drive
		 * address setup control register
		 * is 32 bit !!!
		 */
		pci_read_config_dword(dev, CY82_IDE_ADDRSETUP, &addrCtrl);

		addrCtrl &= (~0xF);
		addrCtrl |= clamp_val(t.setup - 1, 0, 15);
		pci_write_config_dword(dev, CY82_IDE_ADDRSETUP, addrCtrl);

		/* now let's set the remaining registers */
		pci_write_config_byte(dev, CY82_IDE_MASTER_IOR, time_16);
		pci_write_config_byte(dev, CY82_IDE_MASTER_IOW, time_16);
		pci_write_config_byte(dev, CY82_IDE_MASTER_8BIT, time_8);
	} else {
		/*
		 * set slave drive
		 * address setup control register
		 * is 32 bit !!!
		 */
		pci_read_config_dword(dev, CY82_IDE_ADDRSETUP, &addrCtrl);

		addrCtrl &= (~0xF0);
		addrCtrl |= (clamp_val(t.setup - 1, 0, 15) << 4);
		pci_write_config_dword(dev, CY82_IDE_ADDRSETUP, addrCtrl);

		/* now let's set the remaining registers */
		pci_write_config_byte(dev, CY82_IDE_SLAVE_IOR, time_16);
		pci_write_config_byte(dev, CY82_IDE_SLAVE_IOW, time_16);
		pci_write_config_byte(dev, CY82_IDE_SLAVE_8BIT, time_8);
	}
	if (drive->dn > 1)
		pci_dev_put(dev);
}

static void init_iops_cy82c693(ide_hwif_t *hwif)
{
	static ide_hwif_t *primary;
	struct pci_dev *dev = to_pci_dev(hwif->dev);

	if (PCI_FUNC(dev->devfn) == 1)
		primary = hwif;
	else {
		hwif->mate = primary;
		hwif->channel = 1;
	}
}

static const struct ide_port_ops cy82c693_port_ops = {
	.set_pio_mode		= cy82c693_set_pio_mode,
	.set_dma_mode		= cy82c693_set_dma_mode,
};

static const struct ide_port_info cy82c693_chipset = {
	.name		= DRV_NAME,
	.init_iops	= init_iops_cy82c693,
	.port_ops	= &cy82c693_port_ops,
	.host_flags	= IDE_HFLAG_SINGLE,
	.pio_mask	= ATA_PIO4,
	.swdma_mask	= ATA_SWDMA2,
	.mwdma_mask	= ATA_MWDMA2,
};

static int cy82c693_init_one(struct pci_dev *dev,
			     const struct pci_device_id *id)
{
	struct pci_dev *dev2;
	int ret = -ENODEV;

	/* CY82C693 is more than only a IDE controller.
	   Function 1 is primary IDE channel, function 2 - secondary. */
	if ((dev->class >> 8) == PCI_CLASS_STORAGE_IDE &&
	    PCI_FUNC(dev->devfn) == 1) {
		dev2 = pci_get_slot(dev->bus, dev->devfn + 1);
		ret = ide_pci_init_two(dev, dev2, &cy82c693_chipset, NULL);
		if (ret)
			pci_dev_put(dev2);
	}
	return ret;
}

static void cy82c693_remove(struct pci_dev *dev)
{
	struct ide_host *host = pci_get_drvdata(dev);
	struct pci_dev *dev2 = host->dev[1] ? to_pci_dev(host->dev[1]) : NULL;

	ide_pci_remove(dev);
	pci_dev_put(dev2);
}

static const struct pci_device_id cy82c693_pci_tbl[] = {
	{ PCI_VDEVICE(CONTAQ, PCI_DEVICE_ID_CONTAQ_82C693), 0 },
	{ 0, },
};
MODULE_DEVICE_TABLE(pci, cy82c693_pci_tbl);

static struct pci_driver cy82c693_pci_driver = {
	.name		= "Cypress_IDE",
	.id_table	= cy82c693_pci_tbl,
	.probe		= cy82c693_init_one,
	.remove		= cy82c693_remove,
	.suspend	= ide_pci_suspend,
	.resume		= ide_pci_resume,
};

static int __init cy82c693_ide_init(void)
{
	return ide_pci_register_driver(&cy82c693_pci_driver);
}

static void __exit cy82c693_ide_exit(void)
{
	pci_unregister_driver(&cy82c693_pci_driver);
}

module_init(cy82c693_ide_init);
module_exit(cy82c693_ide_exit);

MODULE_AUTHOR("Andreas Krebs, Andre Hedrick, Bartlomiej Zolnierkiewicz");
MODULE_DESCRIPTION("PCI driver module for the Cypress CY82C693 IDE");
MODULE_LICENSE("GPL");
back to top