Revision 3b6b7813b198b578aa7e04e4047ddb8225c37b7f authored by Mikulas Patocka on 20 March 2013, 17:21:25 UTC, committed by Alasdair G Kergon on 20 March 2013, 17:21:25 UTC
A deadlock was found in the prefetch code in the dm verity map
function.  This patch fixes this by transferring the prefetch
to a worker thread and skipping it completely if kmalloc fails.

If generic_make_request is called recursively, it queues the I/O
request on the current->bio_list without making the I/O request
and returns. The routine making the recursive call cannot wait
for the I/O to complete.

The deadlock occurs when one thread grabs the bufio_client
mutex and waits for an I/O to complete but the I/O is queued
on another thread's current->bio_list and is waiting to get
the mutex held by the first thread.

The fix recognises that prefetching is not essential.  If memory
can be allocated, it queues the prefetch request to the worker thread,
but if not, it does nothing.

Signed-off-by: Paul Taysom <taysom@chromium.org>
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Cc: stable@kernel.org
1 parent 58051b9
Raw File
cpumask.c
#include <linux/slab.h>
#include <linux/kernel.h>
#include <linux/bitops.h>
#include <linux/cpumask.h>
#include <linux/export.h>
#include <linux/bootmem.h>

int __first_cpu(const cpumask_t *srcp)
{
	return min_t(int, NR_CPUS, find_first_bit(srcp->bits, NR_CPUS));
}
EXPORT_SYMBOL(__first_cpu);

int __next_cpu(int n, const cpumask_t *srcp)
{
	return min_t(int, NR_CPUS, find_next_bit(srcp->bits, NR_CPUS, n+1));
}
EXPORT_SYMBOL(__next_cpu);

#if NR_CPUS > 64
int __next_cpu_nr(int n, const cpumask_t *srcp)
{
	return min_t(int, nr_cpu_ids,
				find_next_bit(srcp->bits, nr_cpu_ids, n+1));
}
EXPORT_SYMBOL(__next_cpu_nr);
#endif

/**
 * cpumask_next_and - get the next cpu in *src1p & *src2p
 * @n: the cpu prior to the place to search (ie. return will be > @n)
 * @src1p: the first cpumask pointer
 * @src2p: the second cpumask pointer
 *
 * Returns >= nr_cpu_ids if no further cpus set in both.
 */
int cpumask_next_and(int n, const struct cpumask *src1p,
		     const struct cpumask *src2p)
{
	while ((n = cpumask_next(n, src1p)) < nr_cpu_ids)
		if (cpumask_test_cpu(n, src2p))
			break;
	return n;
}
EXPORT_SYMBOL(cpumask_next_and);

/**
 * cpumask_any_but - return a "random" in a cpumask, but not this one.
 * @mask: the cpumask to search
 * @cpu: the cpu to ignore.
 *
 * Often used to find any cpu but smp_processor_id() in a mask.
 * Returns >= nr_cpu_ids if no cpus set.
 */
int cpumask_any_but(const struct cpumask *mask, unsigned int cpu)
{
	unsigned int i;

	cpumask_check(cpu);
	for_each_cpu(i, mask)
		if (i != cpu)
			break;
	return i;
}

/* These are not inline because of header tangles. */
#ifdef CONFIG_CPUMASK_OFFSTACK
/**
 * alloc_cpumask_var_node - allocate a struct cpumask on a given node
 * @mask: pointer to cpumask_var_t where the cpumask is returned
 * @flags: GFP_ flags
 *
 * Only defined when CONFIG_CPUMASK_OFFSTACK=y, otherwise is
 * a nop returning a constant 1 (in <linux/cpumask.h>)
 * Returns TRUE if memory allocation succeeded, FALSE otherwise.
 *
 * In addition, mask will be NULL if this fails.  Note that gcc is
 * usually smart enough to know that mask can never be NULL if
 * CONFIG_CPUMASK_OFFSTACK=n, so does code elimination in that case
 * too.
 */
bool alloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags, int node)
{
	*mask = kmalloc_node(cpumask_size(), flags, node);

#ifdef CONFIG_DEBUG_PER_CPU_MAPS
	if (!*mask) {
		printk(KERN_ERR "=> alloc_cpumask_var: failed!\n");
		dump_stack();
	}
#endif
	/* FIXME: Bandaid to save us from old primitives which go to NR_CPUS. */
	if (*mask) {
		unsigned char *ptr = (unsigned char *)cpumask_bits(*mask);
		unsigned int tail;
		tail = BITS_TO_LONGS(NR_CPUS - nr_cpumask_bits) * sizeof(long);
		memset(ptr + cpumask_size() - tail, 0, tail);
	}

	return *mask != NULL;
}
EXPORT_SYMBOL(alloc_cpumask_var_node);

bool zalloc_cpumask_var_node(cpumask_var_t *mask, gfp_t flags, int node)
{
	return alloc_cpumask_var_node(mask, flags | __GFP_ZERO, node);
}
EXPORT_SYMBOL(zalloc_cpumask_var_node);

/**
 * alloc_cpumask_var - allocate a struct cpumask
 * @mask: pointer to cpumask_var_t where the cpumask is returned
 * @flags: GFP_ flags
 *
 * Only defined when CONFIG_CPUMASK_OFFSTACK=y, otherwise is
 * a nop returning a constant 1 (in <linux/cpumask.h>).
 *
 * See alloc_cpumask_var_node.
 */
bool alloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
{
	return alloc_cpumask_var_node(mask, flags, NUMA_NO_NODE);
}
EXPORT_SYMBOL(alloc_cpumask_var);

bool zalloc_cpumask_var(cpumask_var_t *mask, gfp_t flags)
{
	return alloc_cpumask_var(mask, flags | __GFP_ZERO);
}
EXPORT_SYMBOL(zalloc_cpumask_var);

/**
 * alloc_bootmem_cpumask_var - allocate a struct cpumask from the bootmem arena.
 * @mask: pointer to cpumask_var_t where the cpumask is returned
 *
 * Only defined when CONFIG_CPUMASK_OFFSTACK=y, otherwise is
 * a nop (in <linux/cpumask.h>).
 * Either returns an allocated (zero-filled) cpumask, or causes the
 * system to panic.
 */
void __init alloc_bootmem_cpumask_var(cpumask_var_t *mask)
{
	*mask = alloc_bootmem(cpumask_size());
}

/**
 * free_cpumask_var - frees memory allocated for a struct cpumask.
 * @mask: cpumask to free
 *
 * This is safe on a NULL mask.
 */
void free_cpumask_var(cpumask_var_t mask)
{
	kfree(mask);
}
EXPORT_SYMBOL(free_cpumask_var);

/**
 * free_bootmem_cpumask_var - frees result of alloc_bootmem_cpumask_var
 * @mask: cpumask to free
 */
void __init free_bootmem_cpumask_var(cpumask_var_t mask)
{
	free_bootmem(__pa(mask), cpumask_size());
}
#endif
back to top