linux/kernel/dma/debug.c

1610 lines
42 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* Copyright (C) 2008 Advanced Micro Devices, Inc.
*
* Author: Joerg Roedel <joerg.roedel@amd.com>
*/
#define pr_fmt(fmt) "DMA-API: " fmt
#include <linux/sched/task_stack.h>
#include <linux/scatterlist.h>
#include <linux/dma-map-ops.h>
#include <linux/sched/task.h>
#include <linux/stacktrace.h>
#include <linux/spinlock.h>
#include <linux/vmalloc.h>
#include <linux/debugfs.h>
#include <linux/uaccess.h>
#include <linux/export.h>
#include <linux/device.h>
#include <linux/types.h>
#include <linux/sched.h>
#include <linux/ctype.h>
#include <linux/list.h>
#include <linux/slab.h>
#include <asm/sections.h>
#include "debug.h"
#define HASH_SIZE 16384ULL
#define HASH_FN_SHIFT 13
#define HASH_FN_MASK (HASH_SIZE - 1)
#define PREALLOC_DMA_DEBUG_ENTRIES (1 << 16)
/* If the pool runs out, add this many new entries at once */
2018-12-10 17:00:33 +03:00
#define DMA_DEBUG_DYNAMIC_ENTRIES (PAGE_SIZE / sizeof(struct dma_debug_entry))
enum {
dma_debug_single,
dma_debug_sg,
dma_debug_coherent,
dma_debug_resource,
};
enum map_err_types {
MAP_ERR_CHECK_NOT_APPLICABLE,
MAP_ERR_NOT_CHECKED,
MAP_ERR_CHECKED,
};
#define DMA_DEBUG_STACKTRACE_ENTRIES 5
/**
* struct dma_debug_entry - track a dma_map* or dma_alloc_coherent mapping
* @list: node on pre-allocated free_entries list
* @dev: 'dev' argument to dma_map_{page|single|sg} or dma_alloc_coherent
* @dev_addr: dma address
* @size: length of the mapping
* @type: single, page, sg, coherent
* @direction: enum dma_data_direction
* @sg_call_ents: 'nents' from dma_map_sg
* @sg_mapped_ents: 'mapped_ents' from dma_map_sg
* @pfn: page frame of the start address
* @offset: offset of mapping relative to pfn
* @map_err_type: track whether dma_mapping_error() was checked
* @stack_len: number of backtrace entries in @stack_entries
* @stack_entries: stack of backtrace history
*/
struct dma_debug_entry {
struct list_head list;
struct device *dev;
u64 dev_addr;
u64 size;
int type;
int direction;
int sg_call_ents;
int sg_mapped_ents;
unsigned long pfn;
size_t offset;
enum map_err_types map_err_type;
#ifdef CONFIG_STACKTRACE
dma/debug: Simplify stracktrace retrieval Replace the indirection through struct stack_trace with an invocation of the storage array based interface. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: iommu@lists.linux-foundation.org Cc: Robin Murphy <robin.murphy@arm.com> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Alexander Potapenko <glider@google.com> Cc: Alexey Dobriyan <adobriyan@gmail.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: linux-mm@kvack.org Cc: David Rientjes <rientjes@google.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Andrey Ryabinin <aryabinin@virtuozzo.com> Cc: kasan-dev@googlegroups.com Cc: Mike Rapoport <rppt@linux.vnet.ibm.com> Cc: Akinobu Mita <akinobu.mita@gmail.com> Cc: Johannes Thumshirn <jthumshirn@suse.de> Cc: David Sterba <dsterba@suse.com> Cc: Chris Mason <clm@fb.com> Cc: Josef Bacik <josef@toxicpanda.com> Cc: linux-btrfs@vger.kernel.org Cc: dm-devel@redhat.com Cc: Mike Snitzer <snitzer@redhat.com> Cc: Alasdair Kergon <agk@redhat.com> Cc: Daniel Vetter <daniel@ffwll.ch> Cc: intel-gfx@lists.freedesktop.org Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: dri-devel@lists.freedesktop.org Cc: David Airlie <airlied@linux.ie> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Tom Zanussi <tom.zanussi@linux.intel.com> Cc: Miroslav Benes <mbenes@suse.cz> Cc: linux-arch@vger.kernel.org Link: https://lkml.kernel.org/r/20190425094802.248658135@linutronix.de
2019-04-25 12:45:05 +03:00
unsigned int stack_len;
unsigned long stack_entries[DMA_DEBUG_STACKTRACE_ENTRIES];
#endif
} ____cacheline_aligned_in_smp;
dma-debug: hash_bucket_find needs to allow for offsets within an entry Summary: Users of the pci_dma_sync_single_* api allow users to sync address ranges within the range of a mapped entry (i.e. you can dma map address X to dma_addr_t A and then pci_dma_sync_single on dma_addr_t A+1. The dma-debug library however assume dma syncs will always occur using the base address of a mapped region, and uses that assumption to find entries in its hash table. Since thats often (but not always the case), the dma debug library can give us false errors about missing entries, which are reported as syncing of memory not allocated by the driver. This was noted in the cxgb3 driver as this error: WARNING: at lib/dma-debug.c:902 check_sync+0xdd/0x48c() Hardware name: To be filled by O.E.M. cxgb3 0000:01:00.0: DMA-API: device driver tries to sync DMA memory it has not allocated [device address=0x00000000fff97800] [size=1984 bytes] Modules linked in: autofs4 sunrpc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 uinput snd_hda_codec_intelhdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_timer e1000e snd soundcore r8169 cxgb3 iTCO_wdt snd_page_alloc mii shpchp i2c_i801 iTCO_vendor_support mdio microcode firewire_ohci firewire_core crc_itu_t ata_generic pata_acpi i915 drm_kms_helper drm i2c_algo_bit i2c_core video output [last unloaded: scsi_wait_scan] Pid: 1818, comm: ifconfig Not tainted 2.6.35-0.23.rc3.git6.fc14.x86_64 #1 Call Trace: [<ffffffff81050f71>] warn_slowpath_common+0x85/0x9d [<ffffffff8105102c>] warn_slowpath_fmt+0x46/0x48 [<ffffffff8124658e>] ? check_sync+0x39/0x48c [<ffffffff8107c470>] ? trace_hardirqs_on+0xd/0xf [<ffffffff81246632>] check_sync+0xdd/0x48c [<ffffffff81246ca6>] debug_dma_sync_single_for_device+0x3f/0x41 [<ffffffffa011615c>] ? pci_map_page+0x84/0x97 [cxgb3] [<ffffffffa0117bc3>] pci_dma_sync_single_for_device.clone.0+0x65/0x6e [cxgb3] [<ffffffffa0117ed1>] refill_fl+0x305/0x30a [cxgb3] [<ffffffffa011857d>] t3_sge_alloc_qset+0x6a7/0x821 [cxgb3] [<ffffffffa010a07b>] cxgb_up+0x4d0/0xe62 [cxgb3] [<ffffffff81086037>] ? __module_text_address+0x12/0x58 [<ffffffffa010aa4c>] cxgb_open+0x3f/0x309 [cxgb3] [<ffffffff813e9f6c>] __dev_open+0x8e/0xbc [<ffffffff813e7ca5>] __dev_change_flags+0xbe/0x142 [<ffffffff813e9ea8>] dev_change_flags+0x21/0x57 [<ffffffff81445937>] devinet_ioctl+0x29a/0x54b [<ffffffff811f9a87>] ? inode_has_perm+0xaa/0xce [<ffffffff81446ed2>] inet_ioctl+0x8f/0xa7 [<ffffffff813d683a>] sock_do_ioctl+0x29/0x48 [<ffffffff813d6c83>] sock_ioctl+0x213/0x222 [<ffffffff81137f78>] vfs_ioctl+0x32/0xa6 [<ffffffff811384e2>] do_vfs_ioctl+0x47a/0x4b3 [<ffffffff81138571>] sys_ioctl+0x56/0x79 [<ffffffff81009c32>] system_call_fastpath+0x16/0x1b ---[ end trace 69a4d4cc77b58004 ]--- (some edits by Joerg Roedel) Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Jay Fenalson <fenlason@redhat.com> CC: Divy LeRay <divy@chelsio.com> CC: Stanislaw Gruszka <sgruszka@redhat.com> CC: Joerg Roedel <joerg.roedel@amd.com> CC: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
2011-08-08 23:13:54 +04:00
typedef bool (*match_fn)(struct dma_debug_entry *, struct dma_debug_entry *);
struct hash_bucket {
struct list_head list;
spinlock_t lock;
};
/* Hash list to save the allocated dma addresses */
static struct hash_bucket dma_entry_hash[HASH_SIZE];
/* List of pre-allocated dma_debug_entry's */
static LIST_HEAD(free_entries);
/* Lock for the list above */
static DEFINE_SPINLOCK(free_entries_lock);
/* Global disable flag - will be set in case of an error */
static bool global_disable __read_mostly;
dma-debug: prevent early callers from crashing dma_debug_init() is called by architecture specific code at different levels, but typically as a fs_initcall due to the debugfs initialization. Some platforms may have early callers of the DMA-API, running prior to the fs_initcall() level, which is not much of an issue unless CONFIG_DMA_API_DEBUG is set. When the DMA-API debugging facilities are turned on a caller will go through: debug_dma_map_{single,page} -> dma_mapping_error (inline function usually) -> debug_dma_mapping_error -> get_hash_bucket Calling get_hash_bucket() returns a valid hash value since we hash on high bits of the dma_addr cookie, but we will grab an unitialized spinlock, which typically won't crash but produce a warning, the real crash will however happen during the bucket list traversal because the list has not been initialized yet. An obvious solution is of course to move some of the offenders to run after the fs_initcall level, but since this might not always be an option, we add a flag "dma_debug_initialized" which is set to false by default, and set to true once dma_debug_init() has had a chance to run. The dma_debug_disabled() helper function previously introduced just needs to check for dma_debug_initialized to allow the caller to proceed or not. Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Cc: Dan Williams <dan.j.williams@intel.com> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Horia Geanta <horia.geanta@freescale.com> Cc: Brian Norris <computersforpeace@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-11 02:41:25 +03:00
/* Early initialization disable flag, set at the end of dma_debug_init */
static bool dma_debug_initialized __read_mostly;
static inline bool dma_debug_disabled(void)
{
dma-debug: prevent early callers from crashing dma_debug_init() is called by architecture specific code at different levels, but typically as a fs_initcall due to the debugfs initialization. Some platforms may have early callers of the DMA-API, running prior to the fs_initcall() level, which is not much of an issue unless CONFIG_DMA_API_DEBUG is set. When the DMA-API debugging facilities are turned on a caller will go through: debug_dma_map_{single,page} -> dma_mapping_error (inline function usually) -> debug_dma_mapping_error -> get_hash_bucket Calling get_hash_bucket() returns a valid hash value since we hash on high bits of the dma_addr cookie, but we will grab an unitialized spinlock, which typically won't crash but produce a warning, the real crash will however happen during the bucket list traversal because the list has not been initialized yet. An obvious solution is of course to move some of the offenders to run after the fs_initcall level, but since this might not always be an option, we add a flag "dma_debug_initialized" which is set to false by default, and set to true once dma_debug_init() has had a chance to run. The dma_debug_disabled() helper function previously introduced just needs to check for dma_debug_initialized to allow the caller to proceed or not. Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Cc: Dan Williams <dan.j.williams@intel.com> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Horia Geanta <horia.geanta@freescale.com> Cc: Brian Norris <computersforpeace@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-11 02:41:25 +03:00
return global_disable || !dma_debug_initialized;
}
/* Global error count */
static u32 error_count;
/* Global error show enable*/
static u32 show_all_errors __read_mostly;
/* Number of errors to show */
static u32 show_num_errors = 1;
static u32 num_free_entries;
static u32 min_free_entries;
static u32 nr_total_entries;
/* number of preallocated entries requested by kernel cmdline */
static u32 nr_prealloc_entries = PREALLOC_DMA_DEBUG_ENTRIES;
/* per-driver filter related state */
#define NAME_MAX_LEN 64
static char current_driver_name[NAME_MAX_LEN] __read_mostly;
static struct device_driver *current_driver __read_mostly;
static DEFINE_RWLOCK(driver_name_lock);
static const char *const maperr2str[] = {
[MAP_ERR_CHECK_NOT_APPLICABLE] = "dma map error check not applicable",
[MAP_ERR_NOT_CHECKED] = "dma map error not checked",
[MAP_ERR_CHECKED] = "dma map error checked",
};
static const char *type2name[] = {
[dma_debug_single] = "single",
[dma_debug_sg] = "scatter-gather",
[dma_debug_coherent] = "coherent",
[dma_debug_resource] = "resource",
};
static const char *dir2name[] = {
[DMA_BIDIRECTIONAL] = "DMA_BIDIRECTIONAL",
[DMA_TO_DEVICE] = "DMA_TO_DEVICE",
[DMA_FROM_DEVICE] = "DMA_FROM_DEVICE",
[DMA_NONE] = "DMA_NONE",
};
/*
* The access to some variables in this macro is racy. We can't use atomic_t
* here because all these variables are exported to debugfs. Some of them even
* writeable. This is also the reason why a lock won't help much. But anyway,
* the races are no big deal. Here is why:
*
* error_count: the addition is racy, but the worst thing that can happen is
* that we don't count some errors
* show_num_errors: the subtraction is racy. Also no big deal because in
* worst case this will result in one warning more in the
* system log than the user configured. This variable is
* writeable via debugfs.
*/
static inline void dump_entry_trace(struct dma_debug_entry *entry)
{
#ifdef CONFIG_STACKTRACE
if (entry) {
pr_warn("Mapped at:\n");
dma/debug: Simplify stracktrace retrieval Replace the indirection through struct stack_trace with an invocation of the storage array based interface. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: iommu@lists.linux-foundation.org Cc: Robin Murphy <robin.murphy@arm.com> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Alexander Potapenko <glider@google.com> Cc: Alexey Dobriyan <adobriyan@gmail.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: linux-mm@kvack.org Cc: David Rientjes <rientjes@google.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Andrey Ryabinin <aryabinin@virtuozzo.com> Cc: kasan-dev@googlegroups.com Cc: Mike Rapoport <rppt@linux.vnet.ibm.com> Cc: Akinobu Mita <akinobu.mita@gmail.com> Cc: Johannes Thumshirn <jthumshirn@suse.de> Cc: David Sterba <dsterba@suse.com> Cc: Chris Mason <clm@fb.com> Cc: Josef Bacik <josef@toxicpanda.com> Cc: linux-btrfs@vger.kernel.org Cc: dm-devel@redhat.com Cc: Mike Snitzer <snitzer@redhat.com> Cc: Alasdair Kergon <agk@redhat.com> Cc: Daniel Vetter <daniel@ffwll.ch> Cc: intel-gfx@lists.freedesktop.org Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: dri-devel@lists.freedesktop.org Cc: David Airlie <airlied@linux.ie> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Tom Zanussi <tom.zanussi@linux.intel.com> Cc: Miroslav Benes <mbenes@suse.cz> Cc: linux-arch@vger.kernel.org Link: https://lkml.kernel.org/r/20190425094802.248658135@linutronix.de
2019-04-25 12:45:05 +03:00
stack_trace_print(entry->stack_entries, entry->stack_len, 0);
}
#endif
}
static bool driver_filter(struct device *dev)
{
struct device_driver *drv;
unsigned long flags;
bool ret;
/* driver filter off */
if (likely(!current_driver_name[0]))
return true;
/* driver filter on and initialized */
if (current_driver && dev && dev->driver == current_driver)
return true;
/* driver filter on, but we can't filter on a NULL device... */
if (!dev)
return false;
if (current_driver || !current_driver_name[0])
return false;
/* driver filter on but not yet initialized */
drv = dev->driver;
if (!drv)
return false;
/* lock to protect against change of current_driver_name */
read_lock_irqsave(&driver_name_lock, flags);
ret = false;
if (drv->name &&
strncmp(current_driver_name, drv->name, NAME_MAX_LEN - 1) == 0) {
current_driver = drv;
ret = true;
}
read_unlock_irqrestore(&driver_name_lock, flags);
return ret;
}
#define err_printk(dev, entry, format, arg...) do { \
error_count += 1; \
if (driver_filter(dev) && \
(show_all_errors || show_num_errors > 0)) { \
WARN(1, pr_fmt("%s %s: ") format, \
dev ? dev_driver_string(dev) : "NULL", \
dev ? dev_name(dev) : "NULL", ## arg); \
dump_entry_trace(entry); \
} \
if (!show_all_errors && show_num_errors > 0) \
show_num_errors -= 1; \
} while (0);
/*
* Hash related functions
*
* Every DMA-API request is saved into a struct dma_debug_entry. To
* have quick access to these structs they are stored into a hash.
*/
static int hash_fn(struct dma_debug_entry *entry)
{
/*
* Hash function is based on the dma address.
* We use bits 20-27 here as the index into the hash
*/
return (entry->dev_addr >> HASH_FN_SHIFT) & HASH_FN_MASK;
}
/*
* Request exclusive access to a hash bucket for a given dma_debug_entry.
*/
static struct hash_bucket *get_hash_bucket(struct dma_debug_entry *entry,
unsigned long *flags)
__acquires(&dma_entry_hash[idx].lock)
{
int idx = hash_fn(entry);
unsigned long __flags;
spin_lock_irqsave(&dma_entry_hash[idx].lock, __flags);
*flags = __flags;
return &dma_entry_hash[idx];
}
/*
* Give up exclusive access to the hash bucket
*/
static void put_hash_bucket(struct hash_bucket *bucket,
unsigned long flags)
__releases(&bucket->lock)
{
spin_unlock_irqrestore(&bucket->lock, flags);
}
dma-debug: hash_bucket_find needs to allow for offsets within an entry Summary: Users of the pci_dma_sync_single_* api allow users to sync address ranges within the range of a mapped entry (i.e. you can dma map address X to dma_addr_t A and then pci_dma_sync_single on dma_addr_t A+1. The dma-debug library however assume dma syncs will always occur using the base address of a mapped region, and uses that assumption to find entries in its hash table. Since thats often (but not always the case), the dma debug library can give us false errors about missing entries, which are reported as syncing of memory not allocated by the driver. This was noted in the cxgb3 driver as this error: WARNING: at lib/dma-debug.c:902 check_sync+0xdd/0x48c() Hardware name: To be filled by O.E.M. cxgb3 0000:01:00.0: DMA-API: device driver tries to sync DMA memory it has not allocated [device address=0x00000000fff97800] [size=1984 bytes] Modules linked in: autofs4 sunrpc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 uinput snd_hda_codec_intelhdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_timer e1000e snd soundcore r8169 cxgb3 iTCO_wdt snd_page_alloc mii shpchp i2c_i801 iTCO_vendor_support mdio microcode firewire_ohci firewire_core crc_itu_t ata_generic pata_acpi i915 drm_kms_helper drm i2c_algo_bit i2c_core video output [last unloaded: scsi_wait_scan] Pid: 1818, comm: ifconfig Not tainted 2.6.35-0.23.rc3.git6.fc14.x86_64 #1 Call Trace: [<ffffffff81050f71>] warn_slowpath_common+0x85/0x9d [<ffffffff8105102c>] warn_slowpath_fmt+0x46/0x48 [<ffffffff8124658e>] ? check_sync+0x39/0x48c [<ffffffff8107c470>] ? trace_hardirqs_on+0xd/0xf [<ffffffff81246632>] check_sync+0xdd/0x48c [<ffffffff81246ca6>] debug_dma_sync_single_for_device+0x3f/0x41 [<ffffffffa011615c>] ? pci_map_page+0x84/0x97 [cxgb3] [<ffffffffa0117bc3>] pci_dma_sync_single_for_device.clone.0+0x65/0x6e [cxgb3] [<ffffffffa0117ed1>] refill_fl+0x305/0x30a [cxgb3] [<ffffffffa011857d>] t3_sge_alloc_qset+0x6a7/0x821 [cxgb3] [<ffffffffa010a07b>] cxgb_up+0x4d0/0xe62 [cxgb3] [<ffffffff81086037>] ? __module_text_address+0x12/0x58 [<ffffffffa010aa4c>] cxgb_open+0x3f/0x309 [cxgb3] [<ffffffff813e9f6c>] __dev_open+0x8e/0xbc [<ffffffff813e7ca5>] __dev_change_flags+0xbe/0x142 [<ffffffff813e9ea8>] dev_change_flags+0x21/0x57 [<ffffffff81445937>] devinet_ioctl+0x29a/0x54b [<ffffffff811f9a87>] ? inode_has_perm+0xaa/0xce [<ffffffff81446ed2>] inet_ioctl+0x8f/0xa7 [<ffffffff813d683a>] sock_do_ioctl+0x29/0x48 [<ffffffff813d6c83>] sock_ioctl+0x213/0x222 [<ffffffff81137f78>] vfs_ioctl+0x32/0xa6 [<ffffffff811384e2>] do_vfs_ioctl+0x47a/0x4b3 [<ffffffff81138571>] sys_ioctl+0x56/0x79 [<ffffffff81009c32>] system_call_fastpath+0x16/0x1b ---[ end trace 69a4d4cc77b58004 ]--- (some edits by Joerg Roedel) Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Jay Fenalson <fenlason@redhat.com> CC: Divy LeRay <divy@chelsio.com> CC: Stanislaw Gruszka <sgruszka@redhat.com> CC: Joerg Roedel <joerg.roedel@amd.com> CC: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
2011-08-08 23:13:54 +04:00
static bool exact_match(struct dma_debug_entry *a, struct dma_debug_entry *b)
{
return ((a->dev_addr == b->dev_addr) &&
dma-debug: hash_bucket_find needs to allow for offsets within an entry Summary: Users of the pci_dma_sync_single_* api allow users to sync address ranges within the range of a mapped entry (i.e. you can dma map address X to dma_addr_t A and then pci_dma_sync_single on dma_addr_t A+1. The dma-debug library however assume dma syncs will always occur using the base address of a mapped region, and uses that assumption to find entries in its hash table. Since thats often (but not always the case), the dma debug library can give us false errors about missing entries, which are reported as syncing of memory not allocated by the driver. This was noted in the cxgb3 driver as this error: WARNING: at lib/dma-debug.c:902 check_sync+0xdd/0x48c() Hardware name: To be filled by O.E.M. cxgb3 0000:01:00.0: DMA-API: device driver tries to sync DMA memory it has not allocated [device address=0x00000000fff97800] [size=1984 bytes] Modules linked in: autofs4 sunrpc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 uinput snd_hda_codec_intelhdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_timer e1000e snd soundcore r8169 cxgb3 iTCO_wdt snd_page_alloc mii shpchp i2c_i801 iTCO_vendor_support mdio microcode firewire_ohci firewire_core crc_itu_t ata_generic pata_acpi i915 drm_kms_helper drm i2c_algo_bit i2c_core video output [last unloaded: scsi_wait_scan] Pid: 1818, comm: ifconfig Not tainted 2.6.35-0.23.rc3.git6.fc14.x86_64 #1 Call Trace: [<ffffffff81050f71>] warn_slowpath_common+0x85/0x9d [<ffffffff8105102c>] warn_slowpath_fmt+0x46/0x48 [<ffffffff8124658e>] ? check_sync+0x39/0x48c [<ffffffff8107c470>] ? trace_hardirqs_on+0xd/0xf [<ffffffff81246632>] check_sync+0xdd/0x48c [<ffffffff81246ca6>] debug_dma_sync_single_for_device+0x3f/0x41 [<ffffffffa011615c>] ? pci_map_page+0x84/0x97 [cxgb3] [<ffffffffa0117bc3>] pci_dma_sync_single_for_device.clone.0+0x65/0x6e [cxgb3] [<ffffffffa0117ed1>] refill_fl+0x305/0x30a [cxgb3] [<ffffffffa011857d>] t3_sge_alloc_qset+0x6a7/0x821 [cxgb3] [<ffffffffa010a07b>] cxgb_up+0x4d0/0xe62 [cxgb3] [<ffffffff81086037>] ? __module_text_address+0x12/0x58 [<ffffffffa010aa4c>] cxgb_open+0x3f/0x309 [cxgb3] [<ffffffff813e9f6c>] __dev_open+0x8e/0xbc [<ffffffff813e7ca5>] __dev_change_flags+0xbe/0x142 [<ffffffff813e9ea8>] dev_change_flags+0x21/0x57 [<ffffffff81445937>] devinet_ioctl+0x29a/0x54b [<ffffffff811f9a87>] ? inode_has_perm+0xaa/0xce [<ffffffff81446ed2>] inet_ioctl+0x8f/0xa7 [<ffffffff813d683a>] sock_do_ioctl+0x29/0x48 [<ffffffff813d6c83>] sock_ioctl+0x213/0x222 [<ffffffff81137f78>] vfs_ioctl+0x32/0xa6 [<ffffffff811384e2>] do_vfs_ioctl+0x47a/0x4b3 [<ffffffff81138571>] sys_ioctl+0x56/0x79 [<ffffffff81009c32>] system_call_fastpath+0x16/0x1b ---[ end trace 69a4d4cc77b58004 ]--- (some edits by Joerg Roedel) Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Jay Fenalson <fenlason@redhat.com> CC: Divy LeRay <divy@chelsio.com> CC: Stanislaw Gruszka <sgruszka@redhat.com> CC: Joerg Roedel <joerg.roedel@amd.com> CC: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
2011-08-08 23:13:54 +04:00
(a->dev == b->dev)) ? true : false;
}
static bool containing_match(struct dma_debug_entry *a,
struct dma_debug_entry *b)
{
if (a->dev != b->dev)
return false;
if ((b->dev_addr <= a->dev_addr) &&
((b->dev_addr + b->size) >= (a->dev_addr + a->size)))
return true;
return false;
}
/*
* Search a given entry in the hash bucket list
*/
dma-debug: hash_bucket_find needs to allow for offsets within an entry Summary: Users of the pci_dma_sync_single_* api allow users to sync address ranges within the range of a mapped entry (i.e. you can dma map address X to dma_addr_t A and then pci_dma_sync_single on dma_addr_t A+1. The dma-debug library however assume dma syncs will always occur using the base address of a mapped region, and uses that assumption to find entries in its hash table. Since thats often (but not always the case), the dma debug library can give us false errors about missing entries, which are reported as syncing of memory not allocated by the driver. This was noted in the cxgb3 driver as this error: WARNING: at lib/dma-debug.c:902 check_sync+0xdd/0x48c() Hardware name: To be filled by O.E.M. cxgb3 0000:01:00.0: DMA-API: device driver tries to sync DMA memory it has not allocated [device address=0x00000000fff97800] [size=1984 bytes] Modules linked in: autofs4 sunrpc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 uinput snd_hda_codec_intelhdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_timer e1000e snd soundcore r8169 cxgb3 iTCO_wdt snd_page_alloc mii shpchp i2c_i801 iTCO_vendor_support mdio microcode firewire_ohci firewire_core crc_itu_t ata_generic pata_acpi i915 drm_kms_helper drm i2c_algo_bit i2c_core video output [last unloaded: scsi_wait_scan] Pid: 1818, comm: ifconfig Not tainted 2.6.35-0.23.rc3.git6.fc14.x86_64 #1 Call Trace: [<ffffffff81050f71>] warn_slowpath_common+0x85/0x9d [<ffffffff8105102c>] warn_slowpath_fmt+0x46/0x48 [<ffffffff8124658e>] ? check_sync+0x39/0x48c [<ffffffff8107c470>] ? trace_hardirqs_on+0xd/0xf [<ffffffff81246632>] check_sync+0xdd/0x48c [<ffffffff81246ca6>] debug_dma_sync_single_for_device+0x3f/0x41 [<ffffffffa011615c>] ? pci_map_page+0x84/0x97 [cxgb3] [<ffffffffa0117bc3>] pci_dma_sync_single_for_device.clone.0+0x65/0x6e [cxgb3] [<ffffffffa0117ed1>] refill_fl+0x305/0x30a [cxgb3] [<ffffffffa011857d>] t3_sge_alloc_qset+0x6a7/0x821 [cxgb3] [<ffffffffa010a07b>] cxgb_up+0x4d0/0xe62 [cxgb3] [<ffffffff81086037>] ? __module_text_address+0x12/0x58 [<ffffffffa010aa4c>] cxgb_open+0x3f/0x309 [cxgb3] [<ffffffff813e9f6c>] __dev_open+0x8e/0xbc [<ffffffff813e7ca5>] __dev_change_flags+0xbe/0x142 [<ffffffff813e9ea8>] dev_change_flags+0x21/0x57 [<ffffffff81445937>] devinet_ioctl+0x29a/0x54b [<ffffffff811f9a87>] ? inode_has_perm+0xaa/0xce [<ffffffff81446ed2>] inet_ioctl+0x8f/0xa7 [<ffffffff813d683a>] sock_do_ioctl+0x29/0x48 [<ffffffff813d6c83>] sock_ioctl+0x213/0x222 [<ffffffff81137f78>] vfs_ioctl+0x32/0xa6 [<ffffffff811384e2>] do_vfs_ioctl+0x47a/0x4b3 [<ffffffff81138571>] sys_ioctl+0x56/0x79 [<ffffffff81009c32>] system_call_fastpath+0x16/0x1b ---[ end trace 69a4d4cc77b58004 ]--- (some edits by Joerg Roedel) Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Jay Fenalson <fenlason@redhat.com> CC: Divy LeRay <divy@chelsio.com> CC: Stanislaw Gruszka <sgruszka@redhat.com> CC: Joerg Roedel <joerg.roedel@amd.com> CC: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
2011-08-08 23:13:54 +04:00
static struct dma_debug_entry *__hash_bucket_find(struct hash_bucket *bucket,
struct dma_debug_entry *ref,
match_fn match)
{
struct dma_debug_entry *entry, *ret = NULL;
int matches = 0, match_lvl, last_lvl = -1;
list_for_each_entry(entry, &bucket->list, list) {
dma-debug: hash_bucket_find needs to allow for offsets within an entry Summary: Users of the pci_dma_sync_single_* api allow users to sync address ranges within the range of a mapped entry (i.e. you can dma map address X to dma_addr_t A and then pci_dma_sync_single on dma_addr_t A+1. The dma-debug library however assume dma syncs will always occur using the base address of a mapped region, and uses that assumption to find entries in its hash table. Since thats often (but not always the case), the dma debug library can give us false errors about missing entries, which are reported as syncing of memory not allocated by the driver. This was noted in the cxgb3 driver as this error: WARNING: at lib/dma-debug.c:902 check_sync+0xdd/0x48c() Hardware name: To be filled by O.E.M. cxgb3 0000:01:00.0: DMA-API: device driver tries to sync DMA memory it has not allocated [device address=0x00000000fff97800] [size=1984 bytes] Modules linked in: autofs4 sunrpc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 uinput snd_hda_codec_intelhdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_timer e1000e snd soundcore r8169 cxgb3 iTCO_wdt snd_page_alloc mii shpchp i2c_i801 iTCO_vendor_support mdio microcode firewire_ohci firewire_core crc_itu_t ata_generic pata_acpi i915 drm_kms_helper drm i2c_algo_bit i2c_core video output [last unloaded: scsi_wait_scan] Pid: 1818, comm: ifconfig Not tainted 2.6.35-0.23.rc3.git6.fc14.x86_64 #1 Call Trace: [<ffffffff81050f71>] warn_slowpath_common+0x85/0x9d [<ffffffff8105102c>] warn_slowpath_fmt+0x46/0x48 [<ffffffff8124658e>] ? check_sync+0x39/0x48c [<ffffffff8107c470>] ? trace_hardirqs_on+0xd/0xf [<ffffffff81246632>] check_sync+0xdd/0x48c [<ffffffff81246ca6>] debug_dma_sync_single_for_device+0x3f/0x41 [<ffffffffa011615c>] ? pci_map_page+0x84/0x97 [cxgb3] [<ffffffffa0117bc3>] pci_dma_sync_single_for_device.clone.0+0x65/0x6e [cxgb3] [<ffffffffa0117ed1>] refill_fl+0x305/0x30a [cxgb3] [<ffffffffa011857d>] t3_sge_alloc_qset+0x6a7/0x821 [cxgb3] [<ffffffffa010a07b>] cxgb_up+0x4d0/0xe62 [cxgb3] [<ffffffff81086037>] ? __module_text_address+0x12/0x58 [<ffffffffa010aa4c>] cxgb_open+0x3f/0x309 [cxgb3] [<ffffffff813e9f6c>] __dev_open+0x8e/0xbc [<ffffffff813e7ca5>] __dev_change_flags+0xbe/0x142 [<ffffffff813e9ea8>] dev_change_flags+0x21/0x57 [<ffffffff81445937>] devinet_ioctl+0x29a/0x54b [<ffffffff811f9a87>] ? inode_has_perm+0xaa/0xce [<ffffffff81446ed2>] inet_ioctl+0x8f/0xa7 [<ffffffff813d683a>] sock_do_ioctl+0x29/0x48 [<ffffffff813d6c83>] sock_ioctl+0x213/0x222 [<ffffffff81137f78>] vfs_ioctl+0x32/0xa6 [<ffffffff811384e2>] do_vfs_ioctl+0x47a/0x4b3 [<ffffffff81138571>] sys_ioctl+0x56/0x79 [<ffffffff81009c32>] system_call_fastpath+0x16/0x1b ---[ end trace 69a4d4cc77b58004 ]--- (some edits by Joerg Roedel) Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Jay Fenalson <fenlason@redhat.com> CC: Divy LeRay <divy@chelsio.com> CC: Stanislaw Gruszka <sgruszka@redhat.com> CC: Joerg Roedel <joerg.roedel@amd.com> CC: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
2011-08-08 23:13:54 +04:00
if (!match(ref, entry))
continue;
/*
* Some drivers map the same physical address multiple
* times. Without a hardware IOMMU this results in the
* same device addresses being put into the dma-debug
* hash multiple times too. This can result in false
* positives being reported. Therefore we implement a
* best-fit algorithm here which returns the entry from
* the hash which fits best to the reference value
* instead of the first-fit.
*/
matches += 1;
match_lvl = 0;
entry->size == ref->size ? ++match_lvl : 0;
entry->type == ref->type ? ++match_lvl : 0;
entry->direction == ref->direction ? ++match_lvl : 0;
entry->sg_call_ents == ref->sg_call_ents ? ++match_lvl : 0;
if (match_lvl == 4) {
/* perfect-fit - return the result */
return entry;
} else if (match_lvl > last_lvl) {
/*
* We found an entry that fits better then the
* previous one or it is the 1st match.
*/
last_lvl = match_lvl;
ret = entry;
}
}
/*
* If we have multiple matches but no perfect-fit, just return
* NULL.
*/
ret = (matches == 1) ? ret : NULL;
return ret;
}
dma-debug: hash_bucket_find needs to allow for offsets within an entry Summary: Users of the pci_dma_sync_single_* api allow users to sync address ranges within the range of a mapped entry (i.e. you can dma map address X to dma_addr_t A and then pci_dma_sync_single on dma_addr_t A+1. The dma-debug library however assume dma syncs will always occur using the base address of a mapped region, and uses that assumption to find entries in its hash table. Since thats often (but not always the case), the dma debug library can give us false errors about missing entries, which are reported as syncing of memory not allocated by the driver. This was noted in the cxgb3 driver as this error: WARNING: at lib/dma-debug.c:902 check_sync+0xdd/0x48c() Hardware name: To be filled by O.E.M. cxgb3 0000:01:00.0: DMA-API: device driver tries to sync DMA memory it has not allocated [device address=0x00000000fff97800] [size=1984 bytes] Modules linked in: autofs4 sunrpc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 uinput snd_hda_codec_intelhdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_timer e1000e snd soundcore r8169 cxgb3 iTCO_wdt snd_page_alloc mii shpchp i2c_i801 iTCO_vendor_support mdio microcode firewire_ohci firewire_core crc_itu_t ata_generic pata_acpi i915 drm_kms_helper drm i2c_algo_bit i2c_core video output [last unloaded: scsi_wait_scan] Pid: 1818, comm: ifconfig Not tainted 2.6.35-0.23.rc3.git6.fc14.x86_64 #1 Call Trace: [<ffffffff81050f71>] warn_slowpath_common+0x85/0x9d [<ffffffff8105102c>] warn_slowpath_fmt+0x46/0x48 [<ffffffff8124658e>] ? check_sync+0x39/0x48c [<ffffffff8107c470>] ? trace_hardirqs_on+0xd/0xf [<ffffffff81246632>] check_sync+0xdd/0x48c [<ffffffff81246ca6>] debug_dma_sync_single_for_device+0x3f/0x41 [<ffffffffa011615c>] ? pci_map_page+0x84/0x97 [cxgb3] [<ffffffffa0117bc3>] pci_dma_sync_single_for_device.clone.0+0x65/0x6e [cxgb3] [<ffffffffa0117ed1>] refill_fl+0x305/0x30a [cxgb3] [<ffffffffa011857d>] t3_sge_alloc_qset+0x6a7/0x821 [cxgb3] [<ffffffffa010a07b>] cxgb_up+0x4d0/0xe62 [cxgb3] [<ffffffff81086037>] ? __module_text_address+0x12/0x58 [<ffffffffa010aa4c>] cxgb_open+0x3f/0x309 [cxgb3] [<ffffffff813e9f6c>] __dev_open+0x8e/0xbc [<ffffffff813e7ca5>] __dev_change_flags+0xbe/0x142 [<ffffffff813e9ea8>] dev_change_flags+0x21/0x57 [<ffffffff81445937>] devinet_ioctl+0x29a/0x54b [<ffffffff811f9a87>] ? inode_has_perm+0xaa/0xce [<ffffffff81446ed2>] inet_ioctl+0x8f/0xa7 [<ffffffff813d683a>] sock_do_ioctl+0x29/0x48 [<ffffffff813d6c83>] sock_ioctl+0x213/0x222 [<ffffffff81137f78>] vfs_ioctl+0x32/0xa6 [<ffffffff811384e2>] do_vfs_ioctl+0x47a/0x4b3 [<ffffffff81138571>] sys_ioctl+0x56/0x79 [<ffffffff81009c32>] system_call_fastpath+0x16/0x1b ---[ end trace 69a4d4cc77b58004 ]--- (some edits by Joerg Roedel) Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Jay Fenalson <fenlason@redhat.com> CC: Divy LeRay <divy@chelsio.com> CC: Stanislaw Gruszka <sgruszka@redhat.com> CC: Joerg Roedel <joerg.roedel@amd.com> CC: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
2011-08-08 23:13:54 +04:00
static struct dma_debug_entry *bucket_find_exact(struct hash_bucket *bucket,
struct dma_debug_entry *ref)
{
return __hash_bucket_find(bucket, ref, exact_match);
}
static struct dma_debug_entry *bucket_find_contain(struct hash_bucket **bucket,
struct dma_debug_entry *ref,
unsigned long *flags)
{
struct dma_debug_entry *entry, index = *ref;
int limit = min(HASH_SIZE, (index.dev_addr >> HASH_FN_SHIFT) + 1);
dma-debug: hash_bucket_find needs to allow for offsets within an entry Summary: Users of the pci_dma_sync_single_* api allow users to sync address ranges within the range of a mapped entry (i.e. you can dma map address X to dma_addr_t A and then pci_dma_sync_single on dma_addr_t A+1. The dma-debug library however assume dma syncs will always occur using the base address of a mapped region, and uses that assumption to find entries in its hash table. Since thats often (but not always the case), the dma debug library can give us false errors about missing entries, which are reported as syncing of memory not allocated by the driver. This was noted in the cxgb3 driver as this error: WARNING: at lib/dma-debug.c:902 check_sync+0xdd/0x48c() Hardware name: To be filled by O.E.M. cxgb3 0000:01:00.0: DMA-API: device driver tries to sync DMA memory it has not allocated [device address=0x00000000fff97800] [size=1984 bytes] Modules linked in: autofs4 sunrpc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 uinput snd_hda_codec_intelhdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_timer e1000e snd soundcore r8169 cxgb3 iTCO_wdt snd_page_alloc mii shpchp i2c_i801 iTCO_vendor_support mdio microcode firewire_ohci firewire_core crc_itu_t ata_generic pata_acpi i915 drm_kms_helper drm i2c_algo_bit i2c_core video output [last unloaded: scsi_wait_scan] Pid: 1818, comm: ifconfig Not tainted 2.6.35-0.23.rc3.git6.fc14.x86_64 #1 Call Trace: [<ffffffff81050f71>] warn_slowpath_common+0x85/0x9d [<ffffffff8105102c>] warn_slowpath_fmt+0x46/0x48 [<ffffffff8124658e>] ? check_sync+0x39/0x48c [<ffffffff8107c470>] ? trace_hardirqs_on+0xd/0xf [<ffffffff81246632>] check_sync+0xdd/0x48c [<ffffffff81246ca6>] debug_dma_sync_single_for_device+0x3f/0x41 [<ffffffffa011615c>] ? pci_map_page+0x84/0x97 [cxgb3] [<ffffffffa0117bc3>] pci_dma_sync_single_for_device.clone.0+0x65/0x6e [cxgb3] [<ffffffffa0117ed1>] refill_fl+0x305/0x30a [cxgb3] [<ffffffffa011857d>] t3_sge_alloc_qset+0x6a7/0x821 [cxgb3] [<ffffffffa010a07b>] cxgb_up+0x4d0/0xe62 [cxgb3] [<ffffffff81086037>] ? __module_text_address+0x12/0x58 [<ffffffffa010aa4c>] cxgb_open+0x3f/0x309 [cxgb3] [<ffffffff813e9f6c>] __dev_open+0x8e/0xbc [<ffffffff813e7ca5>] __dev_change_flags+0xbe/0x142 [<ffffffff813e9ea8>] dev_change_flags+0x21/0x57 [<ffffffff81445937>] devinet_ioctl+0x29a/0x54b [<ffffffff811f9a87>] ? inode_has_perm+0xaa/0xce [<ffffffff81446ed2>] inet_ioctl+0x8f/0xa7 [<ffffffff813d683a>] sock_do_ioctl+0x29/0x48 [<ffffffff813d6c83>] sock_ioctl+0x213/0x222 [<ffffffff81137f78>] vfs_ioctl+0x32/0xa6 [<ffffffff811384e2>] do_vfs_ioctl+0x47a/0x4b3 [<ffffffff81138571>] sys_ioctl+0x56/0x79 [<ffffffff81009c32>] system_call_fastpath+0x16/0x1b ---[ end trace 69a4d4cc77b58004 ]--- (some edits by Joerg Roedel) Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Jay Fenalson <fenlason@redhat.com> CC: Divy LeRay <divy@chelsio.com> CC: Stanislaw Gruszka <sgruszka@redhat.com> CC: Joerg Roedel <joerg.roedel@amd.com> CC: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
2011-08-08 23:13:54 +04:00
for (int i = 0; i < limit; i++) {
entry = __hash_bucket_find(*bucket, ref, containing_match);
dma-debug: hash_bucket_find needs to allow for offsets within an entry Summary: Users of the pci_dma_sync_single_* api allow users to sync address ranges within the range of a mapped entry (i.e. you can dma map address X to dma_addr_t A and then pci_dma_sync_single on dma_addr_t A+1. The dma-debug library however assume dma syncs will always occur using the base address of a mapped region, and uses that assumption to find entries in its hash table. Since thats often (but not always the case), the dma debug library can give us false errors about missing entries, which are reported as syncing of memory not allocated by the driver. This was noted in the cxgb3 driver as this error: WARNING: at lib/dma-debug.c:902 check_sync+0xdd/0x48c() Hardware name: To be filled by O.E.M. cxgb3 0000:01:00.0: DMA-API: device driver tries to sync DMA memory it has not allocated [device address=0x00000000fff97800] [size=1984 bytes] Modules linked in: autofs4 sunrpc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 uinput snd_hda_codec_intelhdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_timer e1000e snd soundcore r8169 cxgb3 iTCO_wdt snd_page_alloc mii shpchp i2c_i801 iTCO_vendor_support mdio microcode firewire_ohci firewire_core crc_itu_t ata_generic pata_acpi i915 drm_kms_helper drm i2c_algo_bit i2c_core video output [last unloaded: scsi_wait_scan] Pid: 1818, comm: ifconfig Not tainted 2.6.35-0.23.rc3.git6.fc14.x86_64 #1 Call Trace: [<ffffffff81050f71>] warn_slowpath_common+0x85/0x9d [<ffffffff8105102c>] warn_slowpath_fmt+0x46/0x48 [<ffffffff8124658e>] ? check_sync+0x39/0x48c [<ffffffff8107c470>] ? trace_hardirqs_on+0xd/0xf [<ffffffff81246632>] check_sync+0xdd/0x48c [<ffffffff81246ca6>] debug_dma_sync_single_for_device+0x3f/0x41 [<ffffffffa011615c>] ? pci_map_page+0x84/0x97 [cxgb3] [<ffffffffa0117bc3>] pci_dma_sync_single_for_device.clone.0+0x65/0x6e [cxgb3] [<ffffffffa0117ed1>] refill_fl+0x305/0x30a [cxgb3] [<ffffffffa011857d>] t3_sge_alloc_qset+0x6a7/0x821 [cxgb3] [<ffffffffa010a07b>] cxgb_up+0x4d0/0xe62 [cxgb3] [<ffffffff81086037>] ? __module_text_address+0x12/0x58 [<ffffffffa010aa4c>] cxgb_open+0x3f/0x309 [cxgb3] [<ffffffff813e9f6c>] __dev_open+0x8e/0xbc [<ffffffff813e7ca5>] __dev_change_flags+0xbe/0x142 [<ffffffff813e9ea8>] dev_change_flags+0x21/0x57 [<ffffffff81445937>] devinet_ioctl+0x29a/0x54b [<ffffffff811f9a87>] ? inode_has_perm+0xaa/0xce [<ffffffff81446ed2>] inet_ioctl+0x8f/0xa7 [<ffffffff813d683a>] sock_do_ioctl+0x29/0x48 [<ffffffff813d6c83>] sock_ioctl+0x213/0x222 [<ffffffff81137f78>] vfs_ioctl+0x32/0xa6 [<ffffffff811384e2>] do_vfs_ioctl+0x47a/0x4b3 [<ffffffff81138571>] sys_ioctl+0x56/0x79 [<ffffffff81009c32>] system_call_fastpath+0x16/0x1b ---[ end trace 69a4d4cc77b58004 ]--- (some edits by Joerg Roedel) Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Jay Fenalson <fenlason@redhat.com> CC: Divy LeRay <divy@chelsio.com> CC: Stanislaw Gruszka <sgruszka@redhat.com> CC: Joerg Roedel <joerg.roedel@amd.com> CC: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
2011-08-08 23:13:54 +04:00
if (entry)
return entry;
/*
* Nothing found, go back a hash bucket
*/
put_hash_bucket(*bucket, *flags);
dma-debug: hash_bucket_find needs to allow for offsets within an entry Summary: Users of the pci_dma_sync_single_* api allow users to sync address ranges within the range of a mapped entry (i.e. you can dma map address X to dma_addr_t A and then pci_dma_sync_single on dma_addr_t A+1. The dma-debug library however assume dma syncs will always occur using the base address of a mapped region, and uses that assumption to find entries in its hash table. Since thats often (but not always the case), the dma debug library can give us false errors about missing entries, which are reported as syncing of memory not allocated by the driver. This was noted in the cxgb3 driver as this error: WARNING: at lib/dma-debug.c:902 check_sync+0xdd/0x48c() Hardware name: To be filled by O.E.M. cxgb3 0000:01:00.0: DMA-API: device driver tries to sync DMA memory it has not allocated [device address=0x00000000fff97800] [size=1984 bytes] Modules linked in: autofs4 sunrpc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 uinput snd_hda_codec_intelhdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_timer e1000e snd soundcore r8169 cxgb3 iTCO_wdt snd_page_alloc mii shpchp i2c_i801 iTCO_vendor_support mdio microcode firewire_ohci firewire_core crc_itu_t ata_generic pata_acpi i915 drm_kms_helper drm i2c_algo_bit i2c_core video output [last unloaded: scsi_wait_scan] Pid: 1818, comm: ifconfig Not tainted 2.6.35-0.23.rc3.git6.fc14.x86_64 #1 Call Trace: [<ffffffff81050f71>] warn_slowpath_common+0x85/0x9d [<ffffffff8105102c>] warn_slowpath_fmt+0x46/0x48 [<ffffffff8124658e>] ? check_sync+0x39/0x48c [<ffffffff8107c470>] ? trace_hardirqs_on+0xd/0xf [<ffffffff81246632>] check_sync+0xdd/0x48c [<ffffffff81246ca6>] debug_dma_sync_single_for_device+0x3f/0x41 [<ffffffffa011615c>] ? pci_map_page+0x84/0x97 [cxgb3] [<ffffffffa0117bc3>] pci_dma_sync_single_for_device.clone.0+0x65/0x6e [cxgb3] [<ffffffffa0117ed1>] refill_fl+0x305/0x30a [cxgb3] [<ffffffffa011857d>] t3_sge_alloc_qset+0x6a7/0x821 [cxgb3] [<ffffffffa010a07b>] cxgb_up+0x4d0/0xe62 [cxgb3] [<ffffffff81086037>] ? __module_text_address+0x12/0x58 [<ffffffffa010aa4c>] cxgb_open+0x3f/0x309 [cxgb3] [<ffffffff813e9f6c>] __dev_open+0x8e/0xbc [<ffffffff813e7ca5>] __dev_change_flags+0xbe/0x142 [<ffffffff813e9ea8>] dev_change_flags+0x21/0x57 [<ffffffff81445937>] devinet_ioctl+0x29a/0x54b [<ffffffff811f9a87>] ? inode_has_perm+0xaa/0xce [<ffffffff81446ed2>] inet_ioctl+0x8f/0xa7 [<ffffffff813d683a>] sock_do_ioctl+0x29/0x48 [<ffffffff813d6c83>] sock_ioctl+0x213/0x222 [<ffffffff81137f78>] vfs_ioctl+0x32/0xa6 [<ffffffff811384e2>] do_vfs_ioctl+0x47a/0x4b3 [<ffffffff81138571>] sys_ioctl+0x56/0x79 [<ffffffff81009c32>] system_call_fastpath+0x16/0x1b ---[ end trace 69a4d4cc77b58004 ]--- (some edits by Joerg Roedel) Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Jay Fenalson <fenlason@redhat.com> CC: Divy LeRay <divy@chelsio.com> CC: Stanislaw Gruszka <sgruszka@redhat.com> CC: Joerg Roedel <joerg.roedel@amd.com> CC: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
2011-08-08 23:13:54 +04:00
index.dev_addr -= (1 << HASH_FN_SHIFT);
*bucket = get_hash_bucket(&index, flags);
}
return NULL;
}
/*
* Add an entry to a hash bucket
*/
static void hash_bucket_add(struct hash_bucket *bucket,
struct dma_debug_entry *entry)
{
list_add_tail(&entry->list, &bucket->list);
}
/*
* Remove entry from a hash bucket list
*/
static void hash_bucket_del(struct dma_debug_entry *entry)
{
list_del(&entry->list);
}
static unsigned long long phys_addr(struct dma_debug_entry *entry)
{
if (entry->type == dma_debug_resource)
return __pfn_to_phys(entry->pfn) + entry->offset;
return page_to_phys(pfn_to_page(entry->pfn)) + entry->offset;
}
/*
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
* For each mapping (initial cacheline in the case of
* dma_alloc_coherent/dma_map_page, initial cacheline in each page of a
* scatterlist, or the cacheline specified in dma_map_single) insert
* into this tree using the cacheline as the key. At
* dma_unmap_{single|sg|page} or dma_free_coherent delete the entry. If
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
* the entry already exists at insertion time add a tag as a reference
* count for the overlapping mappings. For now, the overlap tracking
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
* just ensures that 'unmaps' balance 'maps' before marking the
* cacheline idle, but we should also be flagging overlaps as an API
* violation.
*
* Memory usage is mostly constrained by the maximum number of available
* dma-debug entries in that we need a free dma_debug_entry before
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
* inserting into the tree. In the case of dma_map_page and
* dma_alloc_coherent there is only one dma_debug_entry and one
* dma_active_cacheline entry to track per event. dma_map_sg(), on the
* other hand, consumes a single dma_debug_entry, but inserts 'nents'
* entries into the tree.
*/
static RADIX_TREE(dma_active_cacheline, GFP_ATOMIC);
static DEFINE_SPINLOCK(radix_lock);
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
#define ACTIVE_CACHELINE_MAX_OVERLAP ((1 << RADIX_TREE_MAX_TAGS) - 1)
#define CACHELINE_PER_PAGE_SHIFT (PAGE_SHIFT - L1_CACHE_SHIFT)
#define CACHELINES_PER_PAGE (1 << CACHELINE_PER_PAGE_SHIFT)
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
static phys_addr_t to_cacheline_number(struct dma_debug_entry *entry)
{
return (entry->pfn << CACHELINE_PER_PAGE_SHIFT) +
(entry->offset >> L1_CACHE_SHIFT);
}
static int active_cacheline_read_overlap(phys_addr_t cln)
{
int overlap = 0, i;
for (i = RADIX_TREE_MAX_TAGS - 1; i >= 0; i--)
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
if (radix_tree_tag_get(&dma_active_cacheline, cln, i))
overlap |= 1 << i;
return overlap;
}
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
static int active_cacheline_set_overlap(phys_addr_t cln, int overlap)
{
int i;
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
if (overlap > ACTIVE_CACHELINE_MAX_OVERLAP || overlap < 0)
return overlap;
for (i = RADIX_TREE_MAX_TAGS - 1; i >= 0; i--)
if (overlap & 1 << i)
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
radix_tree_tag_set(&dma_active_cacheline, cln, i);
else
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
radix_tree_tag_clear(&dma_active_cacheline, cln, i);
return overlap;
}
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
static void active_cacheline_inc_overlap(phys_addr_t cln)
{
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
int overlap = active_cacheline_read_overlap(cln);
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
overlap = active_cacheline_set_overlap(cln, ++overlap);
/* If we overflowed the overlap counter then we're potentially
* leaking dma-mappings.
*/
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
WARN_ONCE(overlap > ACTIVE_CACHELINE_MAX_OVERLAP,
pr_fmt("exceeded %d overlapping mappings of cacheline %pa\n"),
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
ACTIVE_CACHELINE_MAX_OVERLAP, &cln);
}
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
static int active_cacheline_dec_overlap(phys_addr_t cln)
{
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
int overlap = active_cacheline_read_overlap(cln);
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
return active_cacheline_set_overlap(cln, --overlap);
}
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
static int active_cacheline_insert(struct dma_debug_entry *entry)
{
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
phys_addr_t cln = to_cacheline_number(entry);
unsigned long flags;
int rc;
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
/* If the device is not writing memory then we don't have any
* concerns about the cpu consuming stale data. This mitigates
* legitimate usages of overlapping mappings.
*/
if (entry->direction == DMA_TO_DEVICE)
return 0;
spin_lock_irqsave(&radix_lock, flags);
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
rc = radix_tree_insert(&dma_active_cacheline, cln, entry);
if (rc == -EEXIST)
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
active_cacheline_inc_overlap(cln);
spin_unlock_irqrestore(&radix_lock, flags);
return rc;
}
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
static void active_cacheline_remove(struct dma_debug_entry *entry)
{
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
phys_addr_t cln = to_cacheline_number(entry);
unsigned long flags;
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
/* ...mirror the insert case */
if (entry->direction == DMA_TO_DEVICE)
return;
spin_lock_irqsave(&radix_lock, flags);
/* since we are counting overlaps the final put of the
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
* cacheline will occur when the overlap count is 0.
* active_cacheline_dec_overlap() returns -1 in that case
*/
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
if (active_cacheline_dec_overlap(cln) < 0)
radix_tree_delete(&dma_active_cacheline, cln);
spin_unlock_irqrestore(&radix_lock, flags);
}
/*
* Dump mappings entries on kernel space for debugging purposes
*/
void debug_dma_dump_mappings(struct device *dev)
{
int idx;
phys_addr_t cln;
for (idx = 0; idx < HASH_SIZE; idx++) {
struct hash_bucket *bucket = &dma_entry_hash[idx];
struct dma_debug_entry *entry;
unsigned long flags;
spin_lock_irqsave(&bucket->lock, flags);
list_for_each_entry(entry, &bucket->list, list) {
if (!dev || dev == entry->dev) {
cln = to_cacheline_number(entry);
dev_info(entry->dev,
"%s idx %d P=%llx N=%lx D=%llx L=%llx cln=%pa %s %s\n",
type2name[entry->type], idx,
phys_addr(entry), entry->pfn,
entry->dev_addr, entry->size,
&cln, dir2name[entry->direction],
maperr2str[entry->map_err_type]);
}
}
spin_unlock_irqrestore(&bucket->lock, flags);
cond_resched();
}
}
/*
* Dump mappings entries on user space via debugfs
*/
static int dump_show(struct seq_file *seq, void *v)
{
int idx;
phys_addr_t cln;
for (idx = 0; idx < HASH_SIZE; idx++) {
struct hash_bucket *bucket = &dma_entry_hash[idx];
struct dma_debug_entry *entry;
unsigned long flags;
spin_lock_irqsave(&bucket->lock, flags);
list_for_each_entry(entry, &bucket->list, list) {
cln = to_cacheline_number(entry);
seq_printf(seq,
"%s %s %s idx %d P=%llx N=%lx D=%llx L=%llx cln=%pa %s %s\n",
dev_driver_string(entry->dev),
dev_name(entry->dev),
type2name[entry->type], idx,
phys_addr(entry), entry->pfn,
entry->dev_addr, entry->size,
&cln, dir2name[entry->direction],
maperr2str[entry->map_err_type]);
}
spin_unlock_irqrestore(&bucket->lock, flags);
}
return 0;
}
DEFINE_SHOW_ATTRIBUTE(dump);
/*
* Wrapper function for adding an entry to the hash.
* This function takes care of locking itself.
*/
static void add_dma_entry(struct dma_debug_entry *entry, unsigned long attrs)
{
struct hash_bucket *bucket;
unsigned long flags;
int rc;
bucket = get_hash_bucket(entry, &flags);
hash_bucket_add(bucket, entry);
put_hash_bucket(bucket, flags);
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
rc = active_cacheline_insert(entry);
if (rc == -ENOMEM) {
pr_err_once("cacheline tracking ENOMEM, dma-debug disabled\n");
global_disable = true;
} else if (rc == -EEXIST && !(attrs & DMA_ATTR_SKIP_CPU_SYNC)) {
err_printk(entry->dev, entry,
"cacheline tracking EEXIST, overlapping mappings aren't supported\n");
}
}
2018-12-10 17:00:33 +03:00
static int dma_debug_create_entries(gfp_t gfp)
{
2018-12-10 17:00:33 +03:00
struct dma_debug_entry *entry;
int i;
2018-12-10 17:00:33 +03:00
entry = (void *)get_zeroed_page(gfp);
if (!entry)
return -ENOMEM;
2018-12-10 17:00:33 +03:00
for (i = 0; i < DMA_DEBUG_DYNAMIC_ENTRIES; i++)
list_add_tail(&entry[i].list, &free_entries);
2018-12-10 17:00:33 +03:00
num_free_entries += DMA_DEBUG_DYNAMIC_ENTRIES;
nr_total_entries += DMA_DEBUG_DYNAMIC_ENTRIES;
return 0;
}
static struct dma_debug_entry *__dma_entry_alloc(void)
{
struct dma_debug_entry *entry;
entry = list_entry(free_entries.next, struct dma_debug_entry, list);
list_del(&entry->list);
memset(entry, 0, sizeof(*entry));
num_free_entries -= 1;
if (num_free_entries < min_free_entries)
min_free_entries = num_free_entries;
return entry;
}
dma-debug: don't call __dma_entry_alloc_check_leak() under free_entries_lock __dma_entry_alloc_check_leak() calls into printk -> serial console output (qcom geni) and grabs port->lock under free_entries_lock spin lock, which is a reverse locking dependency chain as qcom_geni IRQ handler can call into dma-debug code and grab free_entries_lock under port->lock. Move __dma_entry_alloc_check_leak() call out of free_entries_lock scope so that we don't acquire serial console's port->lock under it. Trimmed-down lockdep splat: The existing dependency chain (in reverse order) is: -> #2 (free_entries_lock){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x60/0x80 dma_entry_alloc+0x38/0x110 debug_dma_map_page+0x60/0xf8 dma_map_page_attrs+0x1e0/0x230 dma_map_single_attrs.constprop.0+0x6c/0xc8 geni_se_rx_dma_prep+0x40/0xcc qcom_geni_serial_isr+0x310/0x510 __handle_irq_event_percpu+0x110/0x244 handle_irq_event_percpu+0x20/0x54 handle_irq_event+0x50/0x88 handle_fasteoi_irq+0xa4/0xcc handle_irq_desc+0x28/0x40 generic_handle_domain_irq+0x24/0x30 gic_handle_irq+0xc4/0x148 do_interrupt_handler+0xa4/0xb0 el1_interrupt+0x34/0x64 el1h_64_irq_handler+0x18/0x24 el1h_64_irq+0x64/0x68 arch_local_irq_enable+0x4/0x8 ____do_softirq+0x18/0x24 ... -> #1 (&port_lock_key){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x60/0x80 qcom_geni_serial_console_write+0x184/0x1dc console_flush_all+0x344/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 register_console+0x230/0x38c uart_add_one_port+0x338/0x494 qcom_geni_serial_probe+0x390/0x424 platform_probe+0x70/0xc0 really_probe+0x148/0x280 __driver_probe_device+0xfc/0x114 driver_probe_device+0x44/0x100 __device_attach_driver+0x64/0xdc bus_for_each_drv+0xb0/0xd8 __device_attach+0xe4/0x140 device_initial_probe+0x1c/0x28 bus_probe_device+0x44/0xb0 device_add+0x538/0x668 of_device_add+0x44/0x50 of_platform_device_create_pdata+0x94/0xc8 of_platform_bus_create+0x270/0x304 of_platform_populate+0xac/0xc4 devm_of_platform_populate+0x60/0xac geni_se_probe+0x154/0x160 platform_probe+0x70/0xc0 ... -> #0 (console_owner){-...}-{0:0}: __lock_acquire+0xdf8/0x109c lock_acquire+0x234/0x284 console_flush_all+0x330/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 dma_entry_alloc+0xb4/0x110 debug_dma_map_sg+0xdc/0x2f8 __dma_map_sg_attrs+0xac/0xe4 dma_map_sgtable+0x30/0x4c get_pages+0x1d4/0x1e4 [msm] msm_gem_pin_pages_locked+0x38/0xac [msm] msm_gem_pin_vma_locked+0x58/0x88 [msm] msm_ioctl_gem_submit+0xde4/0x13ac [msm] drm_ioctl_kernel+0xe0/0x15c drm_ioctl+0x2e8/0x3f4 vfs_ioctl+0x30/0x50 ... Chain exists of: console_owner --> &port_lock_key --> free_entries_lock Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(free_entries_lock); lock(&port_lock_key); lock(free_entries_lock); lock(console_owner); *** DEADLOCK *** Call trace: dump_backtrace+0xb4/0xf0 show_stack+0x20/0x30 dump_stack_lvl+0x60/0x84 dump_stack+0x18/0x24 print_circular_bug+0x1cc/0x234 check_noncircular+0x78/0xac __lock_acquire+0xdf8/0x109c lock_acquire+0x234/0x284 console_flush_all+0x330/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 dma_entry_alloc+0xb4/0x110 debug_dma_map_sg+0xdc/0x2f8 __dma_map_sg_attrs+0xac/0xe4 dma_map_sgtable+0x30/0x4c get_pages+0x1d4/0x1e4 [msm] msm_gem_pin_pages_locked+0x38/0xac [msm] msm_gem_pin_vma_locked+0x58/0x88 [msm] msm_ioctl_gem_submit+0xde4/0x13ac [msm] drm_ioctl_kernel+0xe0/0x15c drm_ioctl+0x2e8/0x3f4 vfs_ioctl+0x30/0x50 ... Reported-by: Rob Clark <robdclark@chromium.org> Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org> Acked-by: Robin Murphy <robin.murphy@arm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2023-08-16 05:32:21 +03:00
/*
* This should be called outside of free_entries_lock scope to avoid potential
* deadlocks with serial consoles that use DMA.
*/
static void __dma_entry_alloc_check_leak(u32 nr_entries)
{
dma-debug: don't call __dma_entry_alloc_check_leak() under free_entries_lock __dma_entry_alloc_check_leak() calls into printk -> serial console output (qcom geni) and grabs port->lock under free_entries_lock spin lock, which is a reverse locking dependency chain as qcom_geni IRQ handler can call into dma-debug code and grab free_entries_lock under port->lock. Move __dma_entry_alloc_check_leak() call out of free_entries_lock scope so that we don't acquire serial console's port->lock under it. Trimmed-down lockdep splat: The existing dependency chain (in reverse order) is: -> #2 (free_entries_lock){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x60/0x80 dma_entry_alloc+0x38/0x110 debug_dma_map_page+0x60/0xf8 dma_map_page_attrs+0x1e0/0x230 dma_map_single_attrs.constprop.0+0x6c/0xc8 geni_se_rx_dma_prep+0x40/0xcc qcom_geni_serial_isr+0x310/0x510 __handle_irq_event_percpu+0x110/0x244 handle_irq_event_percpu+0x20/0x54 handle_irq_event+0x50/0x88 handle_fasteoi_irq+0xa4/0xcc handle_irq_desc+0x28/0x40 generic_handle_domain_irq+0x24/0x30 gic_handle_irq+0xc4/0x148 do_interrupt_handler+0xa4/0xb0 el1_interrupt+0x34/0x64 el1h_64_irq_handler+0x18/0x24 el1h_64_irq+0x64/0x68 arch_local_irq_enable+0x4/0x8 ____do_softirq+0x18/0x24 ... -> #1 (&port_lock_key){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x60/0x80 qcom_geni_serial_console_write+0x184/0x1dc console_flush_all+0x344/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 register_console+0x230/0x38c uart_add_one_port+0x338/0x494 qcom_geni_serial_probe+0x390/0x424 platform_probe+0x70/0xc0 really_probe+0x148/0x280 __driver_probe_device+0xfc/0x114 driver_probe_device+0x44/0x100 __device_attach_driver+0x64/0xdc bus_for_each_drv+0xb0/0xd8 __device_attach+0xe4/0x140 device_initial_probe+0x1c/0x28 bus_probe_device+0x44/0xb0 device_add+0x538/0x668 of_device_add+0x44/0x50 of_platform_device_create_pdata+0x94/0xc8 of_platform_bus_create+0x270/0x304 of_platform_populate+0xac/0xc4 devm_of_platform_populate+0x60/0xac geni_se_probe+0x154/0x160 platform_probe+0x70/0xc0 ... -> #0 (console_owner){-...}-{0:0}: __lock_acquire+0xdf8/0x109c lock_acquire+0x234/0x284 console_flush_all+0x330/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 dma_entry_alloc+0xb4/0x110 debug_dma_map_sg+0xdc/0x2f8 __dma_map_sg_attrs+0xac/0xe4 dma_map_sgtable+0x30/0x4c get_pages+0x1d4/0x1e4 [msm] msm_gem_pin_pages_locked+0x38/0xac [msm] msm_gem_pin_vma_locked+0x58/0x88 [msm] msm_ioctl_gem_submit+0xde4/0x13ac [msm] drm_ioctl_kernel+0xe0/0x15c drm_ioctl+0x2e8/0x3f4 vfs_ioctl+0x30/0x50 ... Chain exists of: console_owner --> &port_lock_key --> free_entries_lock Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(free_entries_lock); lock(&port_lock_key); lock(free_entries_lock); lock(console_owner); *** DEADLOCK *** Call trace: dump_backtrace+0xb4/0xf0 show_stack+0x20/0x30 dump_stack_lvl+0x60/0x84 dump_stack+0x18/0x24 print_circular_bug+0x1cc/0x234 check_noncircular+0x78/0xac __lock_acquire+0xdf8/0x109c lock_acquire+0x234/0x284 console_flush_all+0x330/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 dma_entry_alloc+0xb4/0x110 debug_dma_map_sg+0xdc/0x2f8 __dma_map_sg_attrs+0xac/0xe4 dma_map_sgtable+0x30/0x4c get_pages+0x1d4/0x1e4 [msm] msm_gem_pin_pages_locked+0x38/0xac [msm] msm_gem_pin_vma_locked+0x58/0x88 [msm] msm_ioctl_gem_submit+0xde4/0x13ac [msm] drm_ioctl_kernel+0xe0/0x15c drm_ioctl+0x2e8/0x3f4 vfs_ioctl+0x30/0x50 ... Reported-by: Rob Clark <robdclark@chromium.org> Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org> Acked-by: Robin Murphy <robin.murphy@arm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2023-08-16 05:32:21 +03:00
u32 tmp = nr_entries % nr_prealloc_entries;
/* Shout each time we tick over some multiple of the initial pool */
if (tmp < DMA_DEBUG_DYNAMIC_ENTRIES) {
pr_info("dma_debug_entry pool grown to %u (%u00%%)\n",
dma-debug: don't call __dma_entry_alloc_check_leak() under free_entries_lock __dma_entry_alloc_check_leak() calls into printk -> serial console output (qcom geni) and grabs port->lock under free_entries_lock spin lock, which is a reverse locking dependency chain as qcom_geni IRQ handler can call into dma-debug code and grab free_entries_lock under port->lock. Move __dma_entry_alloc_check_leak() call out of free_entries_lock scope so that we don't acquire serial console's port->lock under it. Trimmed-down lockdep splat: The existing dependency chain (in reverse order) is: -> #2 (free_entries_lock){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x60/0x80 dma_entry_alloc+0x38/0x110 debug_dma_map_page+0x60/0xf8 dma_map_page_attrs+0x1e0/0x230 dma_map_single_attrs.constprop.0+0x6c/0xc8 geni_se_rx_dma_prep+0x40/0xcc qcom_geni_serial_isr+0x310/0x510 __handle_irq_event_percpu+0x110/0x244 handle_irq_event_percpu+0x20/0x54 handle_irq_event+0x50/0x88 handle_fasteoi_irq+0xa4/0xcc handle_irq_desc+0x28/0x40 generic_handle_domain_irq+0x24/0x30 gic_handle_irq+0xc4/0x148 do_interrupt_handler+0xa4/0xb0 el1_interrupt+0x34/0x64 el1h_64_irq_handler+0x18/0x24 el1h_64_irq+0x64/0x68 arch_local_irq_enable+0x4/0x8 ____do_softirq+0x18/0x24 ... -> #1 (&port_lock_key){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x60/0x80 qcom_geni_serial_console_write+0x184/0x1dc console_flush_all+0x344/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 register_console+0x230/0x38c uart_add_one_port+0x338/0x494 qcom_geni_serial_probe+0x390/0x424 platform_probe+0x70/0xc0 really_probe+0x148/0x280 __driver_probe_device+0xfc/0x114 driver_probe_device+0x44/0x100 __device_attach_driver+0x64/0xdc bus_for_each_drv+0xb0/0xd8 __device_attach+0xe4/0x140 device_initial_probe+0x1c/0x28 bus_probe_device+0x44/0xb0 device_add+0x538/0x668 of_device_add+0x44/0x50 of_platform_device_create_pdata+0x94/0xc8 of_platform_bus_create+0x270/0x304 of_platform_populate+0xac/0xc4 devm_of_platform_populate+0x60/0xac geni_se_probe+0x154/0x160 platform_probe+0x70/0xc0 ... -> #0 (console_owner){-...}-{0:0}: __lock_acquire+0xdf8/0x109c lock_acquire+0x234/0x284 console_flush_all+0x330/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 dma_entry_alloc+0xb4/0x110 debug_dma_map_sg+0xdc/0x2f8 __dma_map_sg_attrs+0xac/0xe4 dma_map_sgtable+0x30/0x4c get_pages+0x1d4/0x1e4 [msm] msm_gem_pin_pages_locked+0x38/0xac [msm] msm_gem_pin_vma_locked+0x58/0x88 [msm] msm_ioctl_gem_submit+0xde4/0x13ac [msm] drm_ioctl_kernel+0xe0/0x15c drm_ioctl+0x2e8/0x3f4 vfs_ioctl+0x30/0x50 ... Chain exists of: console_owner --> &port_lock_key --> free_entries_lock Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(free_entries_lock); lock(&port_lock_key); lock(free_entries_lock); lock(console_owner); *** DEADLOCK *** Call trace: dump_backtrace+0xb4/0xf0 show_stack+0x20/0x30 dump_stack_lvl+0x60/0x84 dump_stack+0x18/0x24 print_circular_bug+0x1cc/0x234 check_noncircular+0x78/0xac __lock_acquire+0xdf8/0x109c lock_acquire+0x234/0x284 console_flush_all+0x330/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 dma_entry_alloc+0xb4/0x110 debug_dma_map_sg+0xdc/0x2f8 __dma_map_sg_attrs+0xac/0xe4 dma_map_sgtable+0x30/0x4c get_pages+0x1d4/0x1e4 [msm] msm_gem_pin_pages_locked+0x38/0xac [msm] msm_gem_pin_vma_locked+0x58/0x88 [msm] msm_ioctl_gem_submit+0xde4/0x13ac [msm] drm_ioctl_kernel+0xe0/0x15c drm_ioctl+0x2e8/0x3f4 vfs_ioctl+0x30/0x50 ... Reported-by: Rob Clark <robdclark@chromium.org> Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org> Acked-by: Robin Murphy <robin.murphy@arm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2023-08-16 05:32:21 +03:00
nr_entries,
(nr_entries / nr_prealloc_entries));
}
}
/* struct dma_entry allocator
*
* The next two functions implement the allocator for
* struct dma_debug_entries.
*/
static struct dma_debug_entry *dma_entry_alloc(void)
{
dma-debug: don't call __dma_entry_alloc_check_leak() under free_entries_lock __dma_entry_alloc_check_leak() calls into printk -> serial console output (qcom geni) and grabs port->lock under free_entries_lock spin lock, which is a reverse locking dependency chain as qcom_geni IRQ handler can call into dma-debug code and grab free_entries_lock under port->lock. Move __dma_entry_alloc_check_leak() call out of free_entries_lock scope so that we don't acquire serial console's port->lock under it. Trimmed-down lockdep splat: The existing dependency chain (in reverse order) is: -> #2 (free_entries_lock){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x60/0x80 dma_entry_alloc+0x38/0x110 debug_dma_map_page+0x60/0xf8 dma_map_page_attrs+0x1e0/0x230 dma_map_single_attrs.constprop.0+0x6c/0xc8 geni_se_rx_dma_prep+0x40/0xcc qcom_geni_serial_isr+0x310/0x510 __handle_irq_event_percpu+0x110/0x244 handle_irq_event_percpu+0x20/0x54 handle_irq_event+0x50/0x88 handle_fasteoi_irq+0xa4/0xcc handle_irq_desc+0x28/0x40 generic_handle_domain_irq+0x24/0x30 gic_handle_irq+0xc4/0x148 do_interrupt_handler+0xa4/0xb0 el1_interrupt+0x34/0x64 el1h_64_irq_handler+0x18/0x24 el1h_64_irq+0x64/0x68 arch_local_irq_enable+0x4/0x8 ____do_softirq+0x18/0x24 ... -> #1 (&port_lock_key){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x60/0x80 qcom_geni_serial_console_write+0x184/0x1dc console_flush_all+0x344/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 register_console+0x230/0x38c uart_add_one_port+0x338/0x494 qcom_geni_serial_probe+0x390/0x424 platform_probe+0x70/0xc0 really_probe+0x148/0x280 __driver_probe_device+0xfc/0x114 driver_probe_device+0x44/0x100 __device_attach_driver+0x64/0xdc bus_for_each_drv+0xb0/0xd8 __device_attach+0xe4/0x140 device_initial_probe+0x1c/0x28 bus_probe_device+0x44/0xb0 device_add+0x538/0x668 of_device_add+0x44/0x50 of_platform_device_create_pdata+0x94/0xc8 of_platform_bus_create+0x270/0x304 of_platform_populate+0xac/0xc4 devm_of_platform_populate+0x60/0xac geni_se_probe+0x154/0x160 platform_probe+0x70/0xc0 ... -> #0 (console_owner){-...}-{0:0}: __lock_acquire+0xdf8/0x109c lock_acquire+0x234/0x284 console_flush_all+0x330/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 dma_entry_alloc+0xb4/0x110 debug_dma_map_sg+0xdc/0x2f8 __dma_map_sg_attrs+0xac/0xe4 dma_map_sgtable+0x30/0x4c get_pages+0x1d4/0x1e4 [msm] msm_gem_pin_pages_locked+0x38/0xac [msm] msm_gem_pin_vma_locked+0x58/0x88 [msm] msm_ioctl_gem_submit+0xde4/0x13ac [msm] drm_ioctl_kernel+0xe0/0x15c drm_ioctl+0x2e8/0x3f4 vfs_ioctl+0x30/0x50 ... Chain exists of: console_owner --> &port_lock_key --> free_entries_lock Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(free_entries_lock); lock(&port_lock_key); lock(free_entries_lock); lock(console_owner); *** DEADLOCK *** Call trace: dump_backtrace+0xb4/0xf0 show_stack+0x20/0x30 dump_stack_lvl+0x60/0x84 dump_stack+0x18/0x24 print_circular_bug+0x1cc/0x234 check_noncircular+0x78/0xac __lock_acquire+0xdf8/0x109c lock_acquire+0x234/0x284 console_flush_all+0x330/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 dma_entry_alloc+0xb4/0x110 debug_dma_map_sg+0xdc/0x2f8 __dma_map_sg_attrs+0xac/0xe4 dma_map_sgtable+0x30/0x4c get_pages+0x1d4/0x1e4 [msm] msm_gem_pin_pages_locked+0x38/0xac [msm] msm_gem_pin_vma_locked+0x58/0x88 [msm] msm_ioctl_gem_submit+0xde4/0x13ac [msm] drm_ioctl_kernel+0xe0/0x15c drm_ioctl+0x2e8/0x3f4 vfs_ioctl+0x30/0x50 ... Reported-by: Rob Clark <robdclark@chromium.org> Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org> Acked-by: Robin Murphy <robin.murphy@arm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2023-08-16 05:32:21 +03:00
bool alloc_check_leak = false;
struct dma_debug_entry *entry;
unsigned long flags;
dma-debug: don't call __dma_entry_alloc_check_leak() under free_entries_lock __dma_entry_alloc_check_leak() calls into printk -> serial console output (qcom geni) and grabs port->lock under free_entries_lock spin lock, which is a reverse locking dependency chain as qcom_geni IRQ handler can call into dma-debug code and grab free_entries_lock under port->lock. Move __dma_entry_alloc_check_leak() call out of free_entries_lock scope so that we don't acquire serial console's port->lock under it. Trimmed-down lockdep splat: The existing dependency chain (in reverse order) is: -> #2 (free_entries_lock){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x60/0x80 dma_entry_alloc+0x38/0x110 debug_dma_map_page+0x60/0xf8 dma_map_page_attrs+0x1e0/0x230 dma_map_single_attrs.constprop.0+0x6c/0xc8 geni_se_rx_dma_prep+0x40/0xcc qcom_geni_serial_isr+0x310/0x510 __handle_irq_event_percpu+0x110/0x244 handle_irq_event_percpu+0x20/0x54 handle_irq_event+0x50/0x88 handle_fasteoi_irq+0xa4/0xcc handle_irq_desc+0x28/0x40 generic_handle_domain_irq+0x24/0x30 gic_handle_irq+0xc4/0x148 do_interrupt_handler+0xa4/0xb0 el1_interrupt+0x34/0x64 el1h_64_irq_handler+0x18/0x24 el1h_64_irq+0x64/0x68 arch_local_irq_enable+0x4/0x8 ____do_softirq+0x18/0x24 ... -> #1 (&port_lock_key){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x60/0x80 qcom_geni_serial_console_write+0x184/0x1dc console_flush_all+0x344/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 register_console+0x230/0x38c uart_add_one_port+0x338/0x494 qcom_geni_serial_probe+0x390/0x424 platform_probe+0x70/0xc0 really_probe+0x148/0x280 __driver_probe_device+0xfc/0x114 driver_probe_device+0x44/0x100 __device_attach_driver+0x64/0xdc bus_for_each_drv+0xb0/0xd8 __device_attach+0xe4/0x140 device_initial_probe+0x1c/0x28 bus_probe_device+0x44/0xb0 device_add+0x538/0x668 of_device_add+0x44/0x50 of_platform_device_create_pdata+0x94/0xc8 of_platform_bus_create+0x270/0x304 of_platform_populate+0xac/0xc4 devm_of_platform_populate+0x60/0xac geni_se_probe+0x154/0x160 platform_probe+0x70/0xc0 ... -> #0 (console_owner){-...}-{0:0}: __lock_acquire+0xdf8/0x109c lock_acquire+0x234/0x284 console_flush_all+0x330/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 dma_entry_alloc+0xb4/0x110 debug_dma_map_sg+0xdc/0x2f8 __dma_map_sg_attrs+0xac/0xe4 dma_map_sgtable+0x30/0x4c get_pages+0x1d4/0x1e4 [msm] msm_gem_pin_pages_locked+0x38/0xac [msm] msm_gem_pin_vma_locked+0x58/0x88 [msm] msm_ioctl_gem_submit+0xde4/0x13ac [msm] drm_ioctl_kernel+0xe0/0x15c drm_ioctl+0x2e8/0x3f4 vfs_ioctl+0x30/0x50 ... Chain exists of: console_owner --> &port_lock_key --> free_entries_lock Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(free_entries_lock); lock(&port_lock_key); lock(free_entries_lock); lock(console_owner); *** DEADLOCK *** Call trace: dump_backtrace+0xb4/0xf0 show_stack+0x20/0x30 dump_stack_lvl+0x60/0x84 dump_stack+0x18/0x24 print_circular_bug+0x1cc/0x234 check_noncircular+0x78/0xac __lock_acquire+0xdf8/0x109c lock_acquire+0x234/0x284 console_flush_all+0x330/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 dma_entry_alloc+0xb4/0x110 debug_dma_map_sg+0xdc/0x2f8 __dma_map_sg_attrs+0xac/0xe4 dma_map_sgtable+0x30/0x4c get_pages+0x1d4/0x1e4 [msm] msm_gem_pin_pages_locked+0x38/0xac [msm] msm_gem_pin_vma_locked+0x58/0x88 [msm] msm_ioctl_gem_submit+0xde4/0x13ac [msm] drm_ioctl_kernel+0xe0/0x15c drm_ioctl+0x2e8/0x3f4 vfs_ioctl+0x30/0x50 ... Reported-by: Rob Clark <robdclark@chromium.org> Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org> Acked-by: Robin Murphy <robin.murphy@arm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2023-08-16 05:32:21 +03:00
u32 nr_entries;
spin_lock_irqsave(&free_entries_lock, flags);
if (num_free_entries == 0) {
2018-12-10 17:00:33 +03:00
if (dma_debug_create_entries(GFP_ATOMIC)) {
global_disable = true;
spin_unlock_irqrestore(&free_entries_lock, flags);
pr_err("debugging out of memory - disabling\n");
return NULL;
}
dma-debug: don't call __dma_entry_alloc_check_leak() under free_entries_lock __dma_entry_alloc_check_leak() calls into printk -> serial console output (qcom geni) and grabs port->lock under free_entries_lock spin lock, which is a reverse locking dependency chain as qcom_geni IRQ handler can call into dma-debug code and grab free_entries_lock under port->lock. Move __dma_entry_alloc_check_leak() call out of free_entries_lock scope so that we don't acquire serial console's port->lock under it. Trimmed-down lockdep splat: The existing dependency chain (in reverse order) is: -> #2 (free_entries_lock){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x60/0x80 dma_entry_alloc+0x38/0x110 debug_dma_map_page+0x60/0xf8 dma_map_page_attrs+0x1e0/0x230 dma_map_single_attrs.constprop.0+0x6c/0xc8 geni_se_rx_dma_prep+0x40/0xcc qcom_geni_serial_isr+0x310/0x510 __handle_irq_event_percpu+0x110/0x244 handle_irq_event_percpu+0x20/0x54 handle_irq_event+0x50/0x88 handle_fasteoi_irq+0xa4/0xcc handle_irq_desc+0x28/0x40 generic_handle_domain_irq+0x24/0x30 gic_handle_irq+0xc4/0x148 do_interrupt_handler+0xa4/0xb0 el1_interrupt+0x34/0x64 el1h_64_irq_handler+0x18/0x24 el1h_64_irq+0x64/0x68 arch_local_irq_enable+0x4/0x8 ____do_softirq+0x18/0x24 ... -> #1 (&port_lock_key){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x60/0x80 qcom_geni_serial_console_write+0x184/0x1dc console_flush_all+0x344/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 register_console+0x230/0x38c uart_add_one_port+0x338/0x494 qcom_geni_serial_probe+0x390/0x424 platform_probe+0x70/0xc0 really_probe+0x148/0x280 __driver_probe_device+0xfc/0x114 driver_probe_device+0x44/0x100 __device_attach_driver+0x64/0xdc bus_for_each_drv+0xb0/0xd8 __device_attach+0xe4/0x140 device_initial_probe+0x1c/0x28 bus_probe_device+0x44/0xb0 device_add+0x538/0x668 of_device_add+0x44/0x50 of_platform_device_create_pdata+0x94/0xc8 of_platform_bus_create+0x270/0x304 of_platform_populate+0xac/0xc4 devm_of_platform_populate+0x60/0xac geni_se_probe+0x154/0x160 platform_probe+0x70/0xc0 ... -> #0 (console_owner){-...}-{0:0}: __lock_acquire+0xdf8/0x109c lock_acquire+0x234/0x284 console_flush_all+0x330/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 dma_entry_alloc+0xb4/0x110 debug_dma_map_sg+0xdc/0x2f8 __dma_map_sg_attrs+0xac/0xe4 dma_map_sgtable+0x30/0x4c get_pages+0x1d4/0x1e4 [msm] msm_gem_pin_pages_locked+0x38/0xac [msm] msm_gem_pin_vma_locked+0x58/0x88 [msm] msm_ioctl_gem_submit+0xde4/0x13ac [msm] drm_ioctl_kernel+0xe0/0x15c drm_ioctl+0x2e8/0x3f4 vfs_ioctl+0x30/0x50 ... Chain exists of: console_owner --> &port_lock_key --> free_entries_lock Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(free_entries_lock); lock(&port_lock_key); lock(free_entries_lock); lock(console_owner); *** DEADLOCK *** Call trace: dump_backtrace+0xb4/0xf0 show_stack+0x20/0x30 dump_stack_lvl+0x60/0x84 dump_stack+0x18/0x24 print_circular_bug+0x1cc/0x234 check_noncircular+0x78/0xac __lock_acquire+0xdf8/0x109c lock_acquire+0x234/0x284 console_flush_all+0x330/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 dma_entry_alloc+0xb4/0x110 debug_dma_map_sg+0xdc/0x2f8 __dma_map_sg_attrs+0xac/0xe4 dma_map_sgtable+0x30/0x4c get_pages+0x1d4/0x1e4 [msm] msm_gem_pin_pages_locked+0x38/0xac [msm] msm_gem_pin_vma_locked+0x58/0x88 [msm] msm_ioctl_gem_submit+0xde4/0x13ac [msm] drm_ioctl_kernel+0xe0/0x15c drm_ioctl+0x2e8/0x3f4 vfs_ioctl+0x30/0x50 ... Reported-by: Rob Clark <robdclark@chromium.org> Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org> Acked-by: Robin Murphy <robin.murphy@arm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2023-08-16 05:32:21 +03:00
alloc_check_leak = true;
nr_entries = nr_total_entries;
}
entry = __dma_entry_alloc();
spin_unlock_irqrestore(&free_entries_lock, flags);
dma-debug: don't call __dma_entry_alloc_check_leak() under free_entries_lock __dma_entry_alloc_check_leak() calls into printk -> serial console output (qcom geni) and grabs port->lock under free_entries_lock spin lock, which is a reverse locking dependency chain as qcom_geni IRQ handler can call into dma-debug code and grab free_entries_lock under port->lock. Move __dma_entry_alloc_check_leak() call out of free_entries_lock scope so that we don't acquire serial console's port->lock under it. Trimmed-down lockdep splat: The existing dependency chain (in reverse order) is: -> #2 (free_entries_lock){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x60/0x80 dma_entry_alloc+0x38/0x110 debug_dma_map_page+0x60/0xf8 dma_map_page_attrs+0x1e0/0x230 dma_map_single_attrs.constprop.0+0x6c/0xc8 geni_se_rx_dma_prep+0x40/0xcc qcom_geni_serial_isr+0x310/0x510 __handle_irq_event_percpu+0x110/0x244 handle_irq_event_percpu+0x20/0x54 handle_irq_event+0x50/0x88 handle_fasteoi_irq+0xa4/0xcc handle_irq_desc+0x28/0x40 generic_handle_domain_irq+0x24/0x30 gic_handle_irq+0xc4/0x148 do_interrupt_handler+0xa4/0xb0 el1_interrupt+0x34/0x64 el1h_64_irq_handler+0x18/0x24 el1h_64_irq+0x64/0x68 arch_local_irq_enable+0x4/0x8 ____do_softirq+0x18/0x24 ... -> #1 (&port_lock_key){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x60/0x80 qcom_geni_serial_console_write+0x184/0x1dc console_flush_all+0x344/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 register_console+0x230/0x38c uart_add_one_port+0x338/0x494 qcom_geni_serial_probe+0x390/0x424 platform_probe+0x70/0xc0 really_probe+0x148/0x280 __driver_probe_device+0xfc/0x114 driver_probe_device+0x44/0x100 __device_attach_driver+0x64/0xdc bus_for_each_drv+0xb0/0xd8 __device_attach+0xe4/0x140 device_initial_probe+0x1c/0x28 bus_probe_device+0x44/0xb0 device_add+0x538/0x668 of_device_add+0x44/0x50 of_platform_device_create_pdata+0x94/0xc8 of_platform_bus_create+0x270/0x304 of_platform_populate+0xac/0xc4 devm_of_platform_populate+0x60/0xac geni_se_probe+0x154/0x160 platform_probe+0x70/0xc0 ... -> #0 (console_owner){-...}-{0:0}: __lock_acquire+0xdf8/0x109c lock_acquire+0x234/0x284 console_flush_all+0x330/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 dma_entry_alloc+0xb4/0x110 debug_dma_map_sg+0xdc/0x2f8 __dma_map_sg_attrs+0xac/0xe4 dma_map_sgtable+0x30/0x4c get_pages+0x1d4/0x1e4 [msm] msm_gem_pin_pages_locked+0x38/0xac [msm] msm_gem_pin_vma_locked+0x58/0x88 [msm] msm_ioctl_gem_submit+0xde4/0x13ac [msm] drm_ioctl_kernel+0xe0/0x15c drm_ioctl+0x2e8/0x3f4 vfs_ioctl+0x30/0x50 ... Chain exists of: console_owner --> &port_lock_key --> free_entries_lock Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(free_entries_lock); lock(&port_lock_key); lock(free_entries_lock); lock(console_owner); *** DEADLOCK *** Call trace: dump_backtrace+0xb4/0xf0 show_stack+0x20/0x30 dump_stack_lvl+0x60/0x84 dump_stack+0x18/0x24 print_circular_bug+0x1cc/0x234 check_noncircular+0x78/0xac __lock_acquire+0xdf8/0x109c lock_acquire+0x234/0x284 console_flush_all+0x330/0x454 console_unlock+0x94/0xf0 vprintk_emit+0x238/0x24c vprintk_default+0x3c/0x48 vprintk+0xb4/0xbc _printk+0x68/0x90 dma_entry_alloc+0xb4/0x110 debug_dma_map_sg+0xdc/0x2f8 __dma_map_sg_attrs+0xac/0xe4 dma_map_sgtable+0x30/0x4c get_pages+0x1d4/0x1e4 [msm] msm_gem_pin_pages_locked+0x38/0xac [msm] msm_gem_pin_vma_locked+0x58/0x88 [msm] msm_ioctl_gem_submit+0xde4/0x13ac [msm] drm_ioctl_kernel+0xe0/0x15c drm_ioctl+0x2e8/0x3f4 vfs_ioctl+0x30/0x50 ... Reported-by: Rob Clark <robdclark@chromium.org> Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org> Acked-by: Robin Murphy <robin.murphy@arm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2023-08-16 05:32:21 +03:00
if (alloc_check_leak)
__dma_entry_alloc_check_leak(nr_entries);
#ifdef CONFIG_STACKTRACE
dma/debug: Simplify stracktrace retrieval Replace the indirection through struct stack_trace with an invocation of the storage array based interface. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: iommu@lists.linux-foundation.org Cc: Robin Murphy <robin.murphy@arm.com> Cc: Marek Szyprowski <m.szyprowski@samsung.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Alexander Potapenko <glider@google.com> Cc: Alexey Dobriyan <adobriyan@gmail.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: linux-mm@kvack.org Cc: David Rientjes <rientjes@google.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Andrey Ryabinin <aryabinin@virtuozzo.com> Cc: kasan-dev@googlegroups.com Cc: Mike Rapoport <rppt@linux.vnet.ibm.com> Cc: Akinobu Mita <akinobu.mita@gmail.com> Cc: Johannes Thumshirn <jthumshirn@suse.de> Cc: David Sterba <dsterba@suse.com> Cc: Chris Mason <clm@fb.com> Cc: Josef Bacik <josef@toxicpanda.com> Cc: linux-btrfs@vger.kernel.org Cc: dm-devel@redhat.com Cc: Mike Snitzer <snitzer@redhat.com> Cc: Alasdair Kergon <agk@redhat.com> Cc: Daniel Vetter <daniel@ffwll.ch> Cc: intel-gfx@lists.freedesktop.org Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: dri-devel@lists.freedesktop.org Cc: David Airlie <airlied@linux.ie> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Tom Zanussi <tom.zanussi@linux.intel.com> Cc: Miroslav Benes <mbenes@suse.cz> Cc: linux-arch@vger.kernel.org Link: https://lkml.kernel.org/r/20190425094802.248658135@linutronix.de
2019-04-25 12:45:05 +03:00
entry->stack_len = stack_trace_save(entry->stack_entries,
ARRAY_SIZE(entry->stack_entries),
1);
#endif
return entry;
}
static void dma_entry_free(struct dma_debug_entry *entry)
{
unsigned long flags;
dma debug: account for cachelines and read-only mappings in overlap tracking While debug_dma_assert_idle() checks if a given *page* is actively undergoing dma the valid granularity of a dma mapping is a *cacheline*. Sander's testing shows that the warning message "DMA-API: exceeded 7 overlapping mappings of pfn..." is falsely triggering. The test is simply mapping multiple cachelines in a given page. Ultimately we want overlap tracking to be valid as it is a real api violation, so we need to track active mappings by cachelines. Update the active dma tracking to use the page-frame-relative cacheline of the mapping as the key, and update debug_dma_assert_idle() to check for all possible mapped cachelines for a given page. However, the need to track active mappings is only relevant when the dma-mapping is writable by the device. In fact it is fairly standard for read-only mappings to have hundreds or thousands of overlapping mappings at once. Limiting the overlap tracking to writable (!DMA_TO_DEVICE) eliminates this class of false-positive overlap reports. Note, the radix gang lookup is sub-optimal. It would be best if it stopped fetching entries once the search passed a page boundary. Nevertheless, this implementation does not perturb the original net_dma failing case. That is to say the extra overhead does not show up in terms of making the failing case pass due to a timing change. References: http://marc.info/?l=linux-netdev&m=139232263419315&w=2 http://marc.info/?l=linux-netdev&m=139217088107122&w=2 Signed-off-by: Dan Williams <dan.j.williams@intel.com> Reported-by: Sander Eikelenboom <linux@eikelenboom.it> Reported-by: Dave Jones <davej@redhat.com> Tested-by: Dave Jones <davej@redhat.com> Tested-by: Sander Eikelenboom <linux@eikelenboom.it> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Cc: Francois Romieu <romieu@fr.zoreil.com> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wei Liu <wei.liu2@citrix.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-03-04 03:38:21 +04:00
active_cacheline_remove(entry);
/*
* add to beginning of the list - this way the entries are
* more likely cache hot when they are reallocated.
*/
spin_lock_irqsave(&free_entries_lock, flags);
list_add(&entry->list, &free_entries);
num_free_entries += 1;
spin_unlock_irqrestore(&free_entries_lock, flags);
}
/*
* DMA-API debugging init code
*
* The init code does two things:
* 1. Initialize core data structures
* 2. Preallocate a given number of dma_debug_entry structs
*/
static ssize_t filter_read(struct file *file, char __user *user_buf,
size_t count, loff_t *ppos)
{
char buf[NAME_MAX_LEN + 1];
unsigned long flags;
int len;
if (!current_driver_name[0])
return 0;
/*
* We can't copy to userspace directly because current_driver_name can
* only be read under the driver_name_lock with irqs disabled. So
* create a temporary copy first.
*/
read_lock_irqsave(&driver_name_lock, flags);
len = scnprintf(buf, NAME_MAX_LEN + 1, "%s\n", current_driver_name);
read_unlock_irqrestore(&driver_name_lock, flags);
return simple_read_from_buffer(user_buf, count, ppos, buf, len);
}
static ssize_t filter_write(struct file *file, const char __user *userbuf,
size_t count, loff_t *ppos)
{
char buf[NAME_MAX_LEN];
unsigned long flags;
size_t len;
int i;
/*
* We can't copy from userspace directly. Access to
* current_driver_name is protected with a write_lock with irqs
* disabled. Since copy_from_user can fault and may sleep we
* need to copy to temporary buffer first
*/
len = min(count, (size_t)(NAME_MAX_LEN - 1));
if (copy_from_user(buf, userbuf, len))
return -EFAULT;
buf[len] = 0;
write_lock_irqsave(&driver_name_lock, flags);
/*
* Now handle the string we got from userspace very carefully.
* The rules are:
* - only use the first token we got
* - token delimiter is everything looking like a space
* character (' ', '\n', '\t' ...)
*
*/
if (!isalnum(buf[0])) {
/*
* If the first character userspace gave us is not
* alphanumerical then assume the filter should be
* switched off.
*/
if (current_driver_name[0])
pr_info("switching off dma-debug driver filter\n");
current_driver_name[0] = 0;
current_driver = NULL;
goto out_unlock;
}
/*
* Now parse out the first token and use it as the name for the
* driver to filter for.
*/
for (i = 0; i < NAME_MAX_LEN - 1; ++i) {
current_driver_name[i] = buf[i];
if (isspace(buf[i]) || buf[i] == ' ' || buf[i] == 0)
break;
}
current_driver_name[i] = 0;
current_driver = NULL;
pr_info("enable driver filter for driver [%s]\n",
current_driver_name);
out_unlock:
write_unlock_irqrestore(&driver_name_lock, flags);
return count;
}
static const struct file_operations filter_fops = {
.read = filter_read,
.write = filter_write,
llseek: automatically add .llseek fop All file_operations should get a .llseek operation so we can make nonseekable_open the default for future file operations without a .llseek pointer. The three cases that we can automatically detect are no_llseek, seq_lseek and default_llseek. For cases where we can we can automatically prove that the file offset is always ignored, we use noop_llseek, which maintains the current behavior of not returning an error from a seek. New drivers should normally not use noop_llseek but instead use no_llseek and call nonseekable_open at open time. Existing drivers can be converted to do the same when the maintainer knows for certain that no user code relies on calling seek on the device file. The generated code is often incorrectly indented and right now contains comments that clarify for each added line why a specific variant was chosen. In the version that gets submitted upstream, the comments will be gone and I will manually fix the indentation, because there does not seem to be a way to do that using coccinelle. Some amount of new code is currently sitting in linux-next that should get the same modifications, which I will do at the end of the merge window. Many thanks to Julia Lawall for helping me learn to write a semantic patch that does all this. ===== begin semantic patch ===== // This adds an llseek= method to all file operations, // as a preparation for making no_llseek the default. // // The rules are // - use no_llseek explicitly if we do nonseekable_open // - use seq_lseek for sequential files // - use default_llseek if we know we access f_pos // - use noop_llseek if we know we don't access f_pos, // but we still want to allow users to call lseek // @ open1 exists @ identifier nested_open; @@ nested_open(...) { <+... nonseekable_open(...) ...+> } @ open exists@ identifier open_f; identifier i, f; identifier open1.nested_open; @@ int open_f(struct inode *i, struct file *f) { <+... ( nonseekable_open(...) | nested_open(...) ) ...+> } @ read disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ read_no_fpos disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { ... when != off } @ write @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ write_no_fpos @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { ... when != off } @ fops0 @ identifier fops; @@ struct file_operations fops = { ... }; @ has_llseek depends on fops0 @ identifier fops0.fops; identifier llseek_f; @@ struct file_operations fops = { ... .llseek = llseek_f, ... }; @ has_read depends on fops0 @ identifier fops0.fops; identifier read_f; @@ struct file_operations fops = { ... .read = read_f, ... }; @ has_write depends on fops0 @ identifier fops0.fops; identifier write_f; @@ struct file_operations fops = { ... .write = write_f, ... }; @ has_open depends on fops0 @ identifier fops0.fops; identifier open_f; @@ struct file_operations fops = { ... .open = open_f, ... }; // use no_llseek if we call nonseekable_open //////////////////////////////////////////// @ nonseekable1 depends on !has_llseek && has_open @ identifier fops0.fops; identifier nso ~= "nonseekable_open"; @@ struct file_operations fops = { ... .open = nso, ... +.llseek = no_llseek, /* nonseekable */ }; @ nonseekable2 depends on !has_llseek @ identifier fops0.fops; identifier open.open_f; @@ struct file_operations fops = { ... .open = open_f, ... +.llseek = no_llseek, /* open uses nonseekable */ }; // use seq_lseek for sequential files ///////////////////////////////////// @ seq depends on !has_llseek @ identifier fops0.fops; identifier sr ~= "seq_read"; @@ struct file_operations fops = { ... .read = sr, ... +.llseek = seq_lseek, /* we have seq_read */ }; // use default_llseek if there is a readdir /////////////////////////////////////////// @ fops1 depends on !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier readdir_e; @@ // any other fop is used that changes pos struct file_operations fops = { ... .readdir = readdir_e, ... +.llseek = default_llseek, /* readdir is present */ }; // use default_llseek if at least one of read/write touches f_pos ///////////////////////////////////////////////////////////////// @ fops2 depends on !fops1 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read.read_f; @@ // read fops use offset struct file_operations fops = { ... .read = read_f, ... +.llseek = default_llseek, /* read accesses f_pos */ }; @ fops3 depends on !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, ... + .llseek = default_llseek, /* write accesses f_pos */ }; // Use noop_llseek if neither read nor write accesses f_pos /////////////////////////////////////////////////////////// @ fops4 depends on !fops1 && !fops2 && !fops3 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; identifier write_no_fpos.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, .read = read_f, ... +.llseek = noop_llseek, /* read and write both use no f_pos */ }; @ depends on has_write && !has_read && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write_no_fpos.write_f; @@ struct file_operations fops = { ... .write = write_f, ... +.llseek = noop_llseek, /* write uses no f_pos */ }; @ depends on has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; @@ struct file_operations fops = { ... .read = read_f, ... +.llseek = noop_llseek, /* read uses no f_pos */ }; @ depends on !has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; @@ struct file_operations fops = { ... +.llseek = noop_llseek, /* no read or write fn */ }; ===== End semantic patch ===== Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Julia Lawall <julia@diku.dk> Cc: Christoph Hellwig <hch@infradead.org>
2010-08-15 20:52:59 +04:00
.llseek = default_llseek,
};
static int __init dma_debug_fs_init(void)
{
struct dentry *dentry = debugfs_create_dir("dma-api", NULL);
debugfs_create_bool("disabled", 0444, dentry, &global_disable);
debugfs_create_u32("error_count", 0444, dentry, &error_count);
debugfs_create_u32("all_errors", 0644, dentry, &show_all_errors);
debugfs_create_u32("num_errors", 0644, dentry, &show_num_errors);
debugfs_create_u32("num_free_entries", 0444, dentry, &num_free_entries);
debugfs_create_u32("min_free_entries", 0444, dentry, &min_free_entries);
debugfs_create_u32("nr_total_entries", 0444, dentry, &nr_total_entries);
debugfs_create_file("driver_filter", 0644, dentry, NULL, &filter_fops);
debugfs_create_file("dump", 0444, dentry, NULL, &dump_fops);
return 0;
}
core_initcall_sync(dma_debug_fs_init);
static int device_dma_allocations(struct device *dev, struct dma_debug_entry **out_entry)
{
struct dma_debug_entry *entry;
unsigned long flags;
int count = 0, i;
for (i = 0; i < HASH_SIZE; ++i) {
spin_lock_irqsave(&dma_entry_hash[i].lock, flags);
list_for_each_entry(entry, &dma_entry_hash[i].list, list) {
if (entry->dev == dev) {
count += 1;
*out_entry = entry;
}
}
spin_unlock_irqrestore(&dma_entry_hash[i].lock, flags);
}
return count;
}
static int dma_debug_device_change(struct notifier_block *nb, unsigned long action, void *data)
{
struct device *dev = data;
treewide: Remove uninitialized_var() usage Using uninitialized_var() is dangerous as it papers over real bugs[1] (or can in the future), and suppresses unrelated compiler warnings (e.g. "unused variable"). If the compiler thinks it is uninitialized, either simply initialize the variable or make compiler changes. In preparation for removing[2] the[3] macro[4], remove all remaining needless uses with the following script: git grep '\buninitialized_var\b' | cut -d: -f1 | sort -u | \ xargs perl -pi -e \ 's/\buninitialized_var\(([^\)]+)\)/\1/g; s:\s*/\* (GCC be quiet|to make compiler happy) \*/$::g;' drivers/video/fbdev/riva/riva_hw.c was manually tweaked to avoid pathological white-space. No outstanding warnings were found building allmodconfig with GCC 9.3.0 for x86_64, i386, arm64, arm, powerpc, powerpc64le, s390x, mips, sparc64, alpha, and m68k. [1] https://lore.kernel.org/lkml/20200603174714.192027-1-glider@google.com/ [2] https://lore.kernel.org/lkml/CA+55aFw+Vbj0i=1TGqCR5vQkCzWJ0QxK6CernOU6eedsudAixw@mail.gmail.com/ [3] https://lore.kernel.org/lkml/CA+55aFwgbgqhbp1fkxvRKEpzyR5J8n1vKT1VZdz9knmPuXhOeg@mail.gmail.com/ [4] https://lore.kernel.org/lkml/CA+55aFz2500WfbKXAx8s67wrm9=yVJu65TpLgN_ybYNv0VEOKA@mail.gmail.com/ Reviewed-by: Leon Romanovsky <leonro@mellanox.com> # drivers/infiniband and mlx4/mlx5 Acked-by: Jason Gunthorpe <jgg@mellanox.com> # IB Acked-by: Kalle Valo <kvalo@codeaurora.org> # wireless drivers Reviewed-by: Chao Yu <yuchao0@huawei.com> # erofs Signed-off-by: Kees Cook <keescook@chromium.org>
2020-06-03 23:09:38 +03:00
struct dma_debug_entry *entry;
int count;
if (dma_debug_disabled())
return 0;
switch (action) {
case BUS_NOTIFY_UNBOUND_DRIVER:
count = device_dma_allocations(dev, &entry);
if (count == 0)
break;
err_printk(dev, entry, "device driver has pending "
"DMA allocations while released from device "
"[count=%d]\n"
"One of leaked entries details: "
"[device address=0x%016llx] [size=%llu bytes] "
"[mapped with %s] [mapped as %s]\n",
count, entry->dev_addr, entry->size,
dir2name[entry->direction], type2name[entry->type]);
break;
default:
break;
}
return 0;
}
void dma_debug_add_bus(const struct bus_type *bus)
{
struct notifier_block *nb;
if (dma_debug_disabled())
return;
nb = kzalloc(sizeof(struct notifier_block), GFP_KERNEL);
if (nb == NULL) {
pr_err("dma_debug_add_bus: out of memory\n");
return;
}
nb->notifier_call = dma_debug_device_change;
bus_register_notifier(bus, nb);
}
static int dma_debug_init(void)
{
2018-12-10 17:00:33 +03:00
int i, nr_pages;
dma-debug: prevent early callers from crashing dma_debug_init() is called by architecture specific code at different levels, but typically as a fs_initcall due to the debugfs initialization. Some platforms may have early callers of the DMA-API, running prior to the fs_initcall() level, which is not much of an issue unless CONFIG_DMA_API_DEBUG is set. When the DMA-API debugging facilities are turned on a caller will go through: debug_dma_map_{single,page} -> dma_mapping_error (inline function usually) -> debug_dma_mapping_error -> get_hash_bucket Calling get_hash_bucket() returns a valid hash value since we hash on high bits of the dma_addr cookie, but we will grab an unitialized spinlock, which typically won't crash but produce a warning, the real crash will however happen during the bucket list traversal because the list has not been initialized yet. An obvious solution is of course to move some of the offenders to run after the fs_initcall level, but since this might not always be an option, we add a flag "dma_debug_initialized" which is set to false by default, and set to true once dma_debug_init() has had a chance to run. The dma_debug_disabled() helper function previously introduced just needs to check for dma_debug_initialized to allow the caller to proceed or not. Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Cc: Dan Williams <dan.j.williams@intel.com> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Horia Geanta <horia.geanta@freescale.com> Cc: Brian Norris <computersforpeace@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-11 02:41:25 +03:00
/* Do not use dma_debug_initialized here, since we really want to be
* called to set dma_debug_initialized
*/
if (global_disable)
return 0;
for (i = 0; i < HASH_SIZE; ++i) {
INIT_LIST_HEAD(&dma_entry_hash[i].list);
spin_lock_init(&dma_entry_hash[i].lock);
}
2018-12-10 17:00:33 +03:00
nr_pages = DIV_ROUND_UP(nr_prealloc_entries, DMA_DEBUG_DYNAMIC_ENTRIES);
for (i = 0; i < nr_pages; ++i)
dma_debug_create_entries(GFP_KERNEL);
if (num_free_entries >= nr_prealloc_entries) {
pr_info("preallocated %d debug entries\n", nr_total_entries);
} else if (num_free_entries > 0) {
pr_warn("%d debug entries requested but only %d allocated\n",
nr_prealloc_entries, nr_total_entries);
} else {
pr_err("debugging out of memory error - disabled\n");
global_disable = true;
return 0;
}
min_free_entries = num_free_entries;
dma-debug: prevent early callers from crashing dma_debug_init() is called by architecture specific code at different levels, but typically as a fs_initcall due to the debugfs initialization. Some platforms may have early callers of the DMA-API, running prior to the fs_initcall() level, which is not much of an issue unless CONFIG_DMA_API_DEBUG is set. When the DMA-API debugging facilities are turned on a caller will go through: debug_dma_map_{single,page} -> dma_mapping_error (inline function usually) -> debug_dma_mapping_error -> get_hash_bucket Calling get_hash_bucket() returns a valid hash value since we hash on high bits of the dma_addr cookie, but we will grab an unitialized spinlock, which typically won't crash but produce a warning, the real crash will however happen during the bucket list traversal because the list has not been initialized yet. An obvious solution is of course to move some of the offenders to run after the fs_initcall level, but since this might not always be an option, we add a flag "dma_debug_initialized" which is set to false by default, and set to true once dma_debug_init() has had a chance to run. The dma_debug_disabled() helper function previously introduced just needs to check for dma_debug_initialized to allow the caller to proceed or not. Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Cc: Dan Williams <dan.j.williams@intel.com> Cc: Jiri Kosina <jkosina@suse.cz> Cc: Horia Geanta <horia.geanta@freescale.com> Cc: Brian Norris <computersforpeace@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-11 02:41:25 +03:00
dma_debug_initialized = true;
pr_info("debugging enabled by kernel config\n");
return 0;
}
core_initcall(dma_debug_init);
static __init int dma_debug_cmdline(char *str)
{
if (!str)
return -EINVAL;
if (strncmp(str, "off", 3) == 0) {
pr_info("debugging disabled on kernel command line\n");
global_disable = true;
}
return 1;
}
static __init int dma_debug_entries_cmdline(char *str)
{
if (!str)
return -EINVAL;
if (!get_option(&str, &nr_prealloc_entries))
nr_prealloc_entries = PREALLOC_DMA_DEBUG_ENTRIES;
return 1;
}
__setup("dma_debug=", dma_debug_cmdline);
__setup("dma_debug_entries=", dma_debug_entries_cmdline);
static void check_unmap(struct dma_debug_entry *ref)
{
struct dma_debug_entry *entry;
struct hash_bucket *bucket;
unsigned long flags;
bucket = get_hash_bucket(ref, &flags);
dma-debug: hash_bucket_find needs to allow for offsets within an entry Summary: Users of the pci_dma_sync_single_* api allow users to sync address ranges within the range of a mapped entry (i.e. you can dma map address X to dma_addr_t A and then pci_dma_sync_single on dma_addr_t A+1. The dma-debug library however assume dma syncs will always occur using the base address of a mapped region, and uses that assumption to find entries in its hash table. Since thats often (but not always the case), the dma debug library can give us false errors about missing entries, which are reported as syncing of memory not allocated by the driver. This was noted in the cxgb3 driver as this error: WARNING: at lib/dma-debug.c:902 check_sync+0xdd/0x48c() Hardware name: To be filled by O.E.M. cxgb3 0000:01:00.0: DMA-API: device driver tries to sync DMA memory it has not allocated [device address=0x00000000fff97800] [size=1984 bytes] Modules linked in: autofs4 sunrpc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 uinput snd_hda_codec_intelhdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_timer e1000e snd soundcore r8169 cxgb3 iTCO_wdt snd_page_alloc mii shpchp i2c_i801 iTCO_vendor_support mdio microcode firewire_ohci firewire_core crc_itu_t ata_generic pata_acpi i915 drm_kms_helper drm i2c_algo_bit i2c_core video output [last unloaded: scsi_wait_scan] Pid: 1818, comm: ifconfig Not tainted 2.6.35-0.23.rc3.git6.fc14.x86_64 #1 Call Trace: [<ffffffff81050f71>] warn_slowpath_common+0x85/0x9d [<ffffffff8105102c>] warn_slowpath_fmt+0x46/0x48 [<ffffffff8124658e>] ? check_sync+0x39/0x48c [<ffffffff8107c470>] ? trace_hardirqs_on+0xd/0xf [<ffffffff81246632>] check_sync+0xdd/0x48c [<ffffffff81246ca6>] debug_dma_sync_single_for_device+0x3f/0x41 [<ffffffffa011615c>] ? pci_map_page+0x84/0x97 [cxgb3] [<ffffffffa0117bc3>] pci_dma_sync_single_for_device.clone.0+0x65/0x6e [cxgb3] [<ffffffffa0117ed1>] refill_fl+0x305/0x30a [cxgb3] [<ffffffffa011857d>] t3_sge_alloc_qset+0x6a7/0x821 [cxgb3] [<ffffffffa010a07b>] cxgb_up+0x4d0/0xe62 [cxgb3] [<ffffffff81086037>] ? __module_text_address+0x12/0x58 [<ffffffffa010aa4c>] cxgb_open+0x3f/0x309 [cxgb3] [<ffffffff813e9f6c>] __dev_open+0x8e/0xbc [<ffffffff813e7ca5>] __dev_change_flags+0xbe/0x142 [<ffffffff813e9ea8>] dev_change_flags+0x21/0x57 [<ffffffff81445937>] devinet_ioctl+0x29a/0x54b [<ffffffff811f9a87>] ? inode_has_perm+0xaa/0xce [<ffffffff81446ed2>] inet_ioctl+0x8f/0xa7 [<ffffffff813d683a>] sock_do_ioctl+0x29/0x48 [<ffffffff813d6c83>] sock_ioctl+0x213/0x222 [<ffffffff81137f78>] vfs_ioctl+0x32/0xa6 [<ffffffff811384e2>] do_vfs_ioctl+0x47a/0x4b3 [<ffffffff81138571>] sys_ioctl+0x56/0x79 [<ffffffff81009c32>] system_call_fastpath+0x16/0x1b ---[ end trace 69a4d4cc77b58004 ]--- (some edits by Joerg Roedel) Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Jay Fenalson <fenlason@redhat.com> CC: Divy LeRay <divy@chelsio.com> CC: Stanislaw Gruszka <sgruszka@redhat.com> CC: Joerg Roedel <joerg.roedel@amd.com> CC: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
2011-08-08 23:13:54 +04:00
entry = bucket_find_exact(bucket, ref);
if (!entry) {
/* must drop lock before calling dma_mapping_error */
put_hash_bucket(bucket, flags);
if (dma_mapping_error(ref->dev, ref->dev_addr)) {
err_printk(ref->dev, NULL,
"device driver tries to free an "
"invalid DMA memory address\n");
} else {
err_printk(ref->dev, NULL,
"device driver tries to free DMA "
"memory it has not allocated [device "
"address=0x%016llx] [size=%llu bytes]\n",
ref->dev_addr, ref->size);
}
return;
}
if (ref->size != entry->size) {
err_printk(ref->dev, entry, "device driver frees "
"DMA memory with different size "
"[device address=0x%016llx] [map size=%llu bytes] "
"[unmap size=%llu bytes]\n",
ref->dev_addr, entry->size, ref->size);
}
if (ref->type != entry->type) {
err_printk(ref->dev, entry, "device driver frees "
"DMA memory with wrong function "
"[device address=0x%016llx] [size=%llu bytes] "
"[mapped as %s] [unmapped as %s]\n",
ref->dev_addr, ref->size,
type2name[entry->type], type2name[ref->type]);
} else if ((entry->type == dma_debug_coherent) &&
(phys_addr(ref) != phys_addr(entry))) {
err_printk(ref->dev, entry, "device driver frees "
"DMA memory with different CPU address "
"[device address=0x%016llx] [size=%llu bytes] "
"[cpu alloc address=0x%016llx] "
"[cpu free address=0x%016llx]",
ref->dev_addr, ref->size,
phys_addr(entry),
phys_addr(ref));
}
if (ref->sg_call_ents && ref->type == dma_debug_sg &&
ref->sg_call_ents != entry->sg_call_ents) {
err_printk(ref->dev, entry, "device driver frees "
"DMA sg list with different entry count "
"[map count=%d] [unmap count=%d]\n",
entry->sg_call_ents, ref->sg_call_ents);
}
/*
* This may be no bug in reality - but most implementations of the
* DMA API don't handle this properly, so check for it here
*/
if (ref->direction != entry->direction) {
err_printk(ref->dev, entry, "device driver frees "
"DMA memory with different direction "
"[device address=0x%016llx] [size=%llu bytes] "
"[mapped with %s] [unmapped with %s]\n",
ref->dev_addr, ref->size,
dir2name[entry->direction],
dir2name[ref->direction]);
}
/*
* Drivers should use dma_mapping_error() to check the returned
* addresses of dma_map_single() and dma_map_page().
* If not, print this warning message. See Documentation/core-api/dma-api.rst.
*/
if (entry->map_err_type == MAP_ERR_NOT_CHECKED) {
err_printk(ref->dev, entry,
"device driver failed to check map error"
"[device address=0x%016llx] [size=%llu bytes] "
"[mapped as %s]",
ref->dev_addr, ref->size,
type2name[entry->type]);
}
hash_bucket_del(entry);
dma_entry_free(entry);
put_hash_bucket(bucket, flags);
}
static void check_for_stack(struct device *dev,
struct page *page, size_t offset)
{
void *addr;
struct vm_struct *stack_vm_area = task_stack_vm_area(current);
if (!stack_vm_area) {
/* Stack is direct-mapped. */
if (PageHighMem(page))
return;
addr = page_address(page) + offset;
if (object_is_on_stack(addr))
err_printk(dev, NULL, "device driver maps memory from stack [addr=%p]\n", addr);
} else {
/* Stack is vmalloced. */
int i;
for (i = 0; i < stack_vm_area->nr_pages; i++) {
if (page != stack_vm_area->pages[i])
continue;
addr = (u8 *)current->stack + i * PAGE_SIZE + offset;
err_printk(dev, NULL, "device driver maps memory from stack [probable addr=%p]\n", addr);
break;
}
}
}
static void check_for_illegal_area(struct device *dev, void *addr, unsigned long len)
{
if (memory_intersects(_stext, _etext, addr, len) ||
memory_intersects(__start_rodata, __end_rodata, addr, len))
err_printk(dev, NULL, "device driver maps memory from kernel text or rodata [addr=%p] [len=%lu]\n", addr, len);
}
static void check_sync(struct device *dev,
struct dma_debug_entry *ref,
bool to_cpu)
{
struct dma_debug_entry *entry;
struct hash_bucket *bucket;
unsigned long flags;
bucket = get_hash_bucket(ref, &flags);
dma-debug: hash_bucket_find needs to allow for offsets within an entry Summary: Users of the pci_dma_sync_single_* api allow users to sync address ranges within the range of a mapped entry (i.e. you can dma map address X to dma_addr_t A and then pci_dma_sync_single on dma_addr_t A+1. The dma-debug library however assume dma syncs will always occur using the base address of a mapped region, and uses that assumption to find entries in its hash table. Since thats often (but not always the case), the dma debug library can give us false errors about missing entries, which are reported as syncing of memory not allocated by the driver. This was noted in the cxgb3 driver as this error: WARNING: at lib/dma-debug.c:902 check_sync+0xdd/0x48c() Hardware name: To be filled by O.E.M. cxgb3 0000:01:00.0: DMA-API: device driver tries to sync DMA memory it has not allocated [device address=0x00000000fff97800] [size=1984 bytes] Modules linked in: autofs4 sunrpc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 uinput snd_hda_codec_intelhdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_timer e1000e snd soundcore r8169 cxgb3 iTCO_wdt snd_page_alloc mii shpchp i2c_i801 iTCO_vendor_support mdio microcode firewire_ohci firewire_core crc_itu_t ata_generic pata_acpi i915 drm_kms_helper drm i2c_algo_bit i2c_core video output [last unloaded: scsi_wait_scan] Pid: 1818, comm: ifconfig Not tainted 2.6.35-0.23.rc3.git6.fc14.x86_64 #1 Call Trace: [<ffffffff81050f71>] warn_slowpath_common+0x85/0x9d [<ffffffff8105102c>] warn_slowpath_fmt+0x46/0x48 [<ffffffff8124658e>] ? check_sync+0x39/0x48c [<ffffffff8107c470>] ? trace_hardirqs_on+0xd/0xf [<ffffffff81246632>] check_sync+0xdd/0x48c [<ffffffff81246ca6>] debug_dma_sync_single_for_device+0x3f/0x41 [<ffffffffa011615c>] ? pci_map_page+0x84/0x97 [cxgb3] [<ffffffffa0117bc3>] pci_dma_sync_single_for_device.clone.0+0x65/0x6e [cxgb3] [<ffffffffa0117ed1>] refill_fl+0x305/0x30a [cxgb3] [<ffffffffa011857d>] t3_sge_alloc_qset+0x6a7/0x821 [cxgb3] [<ffffffffa010a07b>] cxgb_up+0x4d0/0xe62 [cxgb3] [<ffffffff81086037>] ? __module_text_address+0x12/0x58 [<ffffffffa010aa4c>] cxgb_open+0x3f/0x309 [cxgb3] [<ffffffff813e9f6c>] __dev_open+0x8e/0xbc [<ffffffff813e7ca5>] __dev_change_flags+0xbe/0x142 [<ffffffff813e9ea8>] dev_change_flags+0x21/0x57 [<ffffffff81445937>] devinet_ioctl+0x29a/0x54b [<ffffffff811f9a87>] ? inode_has_perm+0xaa/0xce [<ffffffff81446ed2>] inet_ioctl+0x8f/0xa7 [<ffffffff813d683a>] sock_do_ioctl+0x29/0x48 [<ffffffff813d6c83>] sock_ioctl+0x213/0x222 [<ffffffff81137f78>] vfs_ioctl+0x32/0xa6 [<ffffffff811384e2>] do_vfs_ioctl+0x47a/0x4b3 [<ffffffff81138571>] sys_ioctl+0x56/0x79 [<ffffffff81009c32>] system_call_fastpath+0x16/0x1b ---[ end trace 69a4d4cc77b58004 ]--- (some edits by Joerg Roedel) Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Jay Fenalson <fenlason@redhat.com> CC: Divy LeRay <divy@chelsio.com> CC: Stanislaw Gruszka <sgruszka@redhat.com> CC: Joerg Roedel <joerg.roedel@amd.com> CC: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
2011-08-08 23:13:54 +04:00
entry = bucket_find_contain(&bucket, ref, &flags);
if (!entry) {
err_printk(dev, NULL, "device driver tries "
"to sync DMA memory it has not allocated "
"[device address=0x%016llx] [size=%llu bytes]\n",
(unsigned long long)ref->dev_addr, ref->size);
goto out;
}
if (ref->size > entry->size) {
err_printk(dev, entry, "device driver syncs"
" DMA memory outside allocated range "
"[device address=0x%016llx] "
"[allocation size=%llu bytes] "
"[sync offset+size=%llu]\n",
entry->dev_addr, entry->size,
ref->size);
}
if (entry->direction == DMA_BIDIRECTIONAL)
goto out;
if (ref->direction != entry->direction) {
err_printk(dev, entry, "device driver syncs "
"DMA memory with different direction "
"[device address=0x%016llx] [size=%llu bytes] "
"[mapped with %s] [synced with %s]\n",
(unsigned long long)ref->dev_addr, entry->size,
dir2name[entry->direction],
dir2name[ref->direction]);
}
if (to_cpu && !(entry->direction == DMA_FROM_DEVICE) &&
!(ref->direction == DMA_TO_DEVICE))
err_printk(dev, entry, "device driver syncs "
"device read-only DMA memory for cpu "
"[device address=0x%016llx] [size=%llu bytes] "
"[mapped with %s] [synced with %s]\n",
(unsigned long long)ref->dev_addr, entry->size,
dir2name[entry->direction],
dir2name[ref->direction]);
if (!to_cpu && !(entry->direction == DMA_TO_DEVICE) &&
!(ref->direction == DMA_FROM_DEVICE))
err_printk(dev, entry, "device driver syncs "
"device write-only DMA memory to device "
"[device address=0x%016llx] [size=%llu bytes] "
"[mapped with %s] [synced with %s]\n",
(unsigned long long)ref->dev_addr, entry->size,
dir2name[entry->direction],
dir2name[ref->direction]);
if (ref->sg_call_ents && ref->type == dma_debug_sg &&
ref->sg_call_ents != entry->sg_call_ents) {
err_printk(ref->dev, entry, "device driver syncs "
"DMA sg list with different entry count "
"[map count=%d] [sync count=%d]\n",
entry->sg_call_ents, ref->sg_call_ents);
}
out:
put_hash_bucket(bucket, flags);
}
static void check_sg_segment(struct device *dev, struct scatterlist *sg)
{
#ifdef CONFIG_DMA_API_DEBUG_SG
unsigned int max_seg = dma_get_max_seg_size(dev);
u64 start, end, boundary = dma_get_seg_boundary(dev);
/*
* Either the driver forgot to set dma_parms appropriately, or
* whoever generated the list forgot to check them.
*/
if (sg->length > max_seg)
err_printk(dev, NULL, "mapping sg segment longer than device claims to support [len=%u] [max=%u]\n",
sg->length, max_seg);
/*
* In some cases this could potentially be the DMA API
* implementation's fault, but it would usually imply that
* the scatterlist was built inappropriately to begin with.
*/
start = sg_dma_address(sg);
end = start + sg_dma_len(sg) - 1;
if ((start ^ end) & ~boundary)
err_printk(dev, NULL, "mapping sg segment across boundary [start=0x%016llx] [end=0x%016llx] [boundary=0x%016llx]\n",
start, end, boundary);
#endif
}
void debug_dma_map_single(struct device *dev, const void *addr,
unsigned long len)
{
if (unlikely(dma_debug_disabled()))
return;
if (!virt_addr_valid(addr))
err_printk(dev, NULL, "device driver maps memory from invalid area [addr=%p] [len=%lu]\n",
addr, len);
if (is_vmalloc_addr(addr))
err_printk(dev, NULL, "device driver maps memory from vmalloc area [addr=%p] [len=%lu]\n",
addr, len);
}
EXPORT_SYMBOL(debug_dma_map_single);
void debug_dma_map_page(struct device *dev, struct page *page, size_t offset,
size_t size, int direction, dma_addr_t dma_addr,
unsigned long attrs)
{
struct dma_debug_entry *entry;
if (unlikely(dma_debug_disabled()))
return;
if (dma_mapping_error(dev, dma_addr))
return;
entry = dma_entry_alloc();
if (!entry)
return;
entry->dev = dev;
entry->type = dma_debug_single;
entry->pfn = page_to_pfn(page);
entry->offset = offset;
entry->dev_addr = dma_addr;
entry->size = size;
entry->direction = direction;
entry->map_err_type = MAP_ERR_NOT_CHECKED;
check_for_stack(dev, page, offset);
if (!PageHighMem(page)) {
void *addr = page_address(page) + offset;
check_for_illegal_area(dev, addr, size);
}
add_dma_entry(entry, attrs);
}
void debug_dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
{
struct dma_debug_entry ref;
struct dma_debug_entry *entry;
struct hash_bucket *bucket;
unsigned long flags;
if (unlikely(dma_debug_disabled()))
return;
ref.dev = dev;
ref.dev_addr = dma_addr;
bucket = get_hash_bucket(&ref, &flags);
dma-debug: update DMA debug API to better handle multiple mappings of a buffer There were reports of the igb driver unmapping buffers without calling dma_mapping_error. On closer inspection issues were found in the DMA debug API and how it handled multiple mappings of the same buffer. The issue I found is the fact that the debug_dma_mapping_error would only set the map_err_type to MAP_ERR_CHECKED in the case that the was only one match for device and device address. However in the case of non-IOMMU, multiple addresses existed and as a result it was not setting this field once a second mapping was instantiated. I have resolved this by changing the search so that it instead will now set MAP_ERR_CHECKED on the first buffer that matches the device and DMA address that is currently in the state MAP_ERR_NOT_CHECKED. A secondary side effect of this patch is that in the case of multiple buffers using the same address only the last mapping will have a valid map_err_type. The previous mappings will all end up with map_err_type set to MAP_ERR_CHECKED because of the dma_mapping_error call in debug_dma_map_page. However this behavior may be preferable as it means you will likely only see one real error per multi-mapped buffer, versus the current behavior of multiple false errors mer multi-mapped buffer. Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com> Cc: Joerg Roedel <joro@8bytes.org> Reviewed-by: Shuah Khan <shuah.khan@hp.com> Tested-by: Shuah Khan <shuah.khan@hp.com> Cc: Jakub Kicinski <kubakici@wp.pl> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-03-23 02:04:49 +04:00
list_for_each_entry(entry, &bucket->list, list) {
if (!exact_match(&ref, entry))
continue;
/*
* The same physical address can be mapped multiple
* times. Without a hardware IOMMU this results in the
* same device addresses being put into the dma-debug
* hash multiple times too. This can result in false
* positives being reported. Therefore we implement a
* best-fit algorithm here which updates the first entry
* from the hash which fits the reference value and is
* not currently listed as being checked.
*/
if (entry->map_err_type == MAP_ERR_NOT_CHECKED) {
entry->map_err_type = MAP_ERR_CHECKED;
break;
}
}
put_hash_bucket(bucket, flags);
}
EXPORT_SYMBOL(debug_dma_mapping_error);
void debug_dma_unmap_page(struct device *dev, dma_addr_t dma_addr,
size_t size, int direction)
{
struct dma_debug_entry ref = {
.type = dma_debug_single,
.dev = dev,
.dev_addr = dma_addr,
.size = size,
.direction = direction,
};
if (unlikely(dma_debug_disabled()))
return;
check_unmap(&ref);
}
void debug_dma_map_sg(struct device *dev, struct scatterlist *sg,
int nents, int mapped_ents, int direction,
unsigned long attrs)
{
struct dma_debug_entry *entry;
struct scatterlist *s;
int i;
if (unlikely(dma_debug_disabled()))
return;
dma-debug: fix sg checks in debug_dma_map_sg() The following warning occurred sporadically on s390: DMA-API: nvme 0006:00:00.0: device driver maps memory from kernel text or rodata [addr=0000000048cc5e2f] [len=131072] WARNING: CPU: 4 PID: 825 at kernel/dma/debug.c:1083 check_for_illegal_area+0xa8/0x138 It is a false-positive warning, due to broken logic in debug_dma_map_sg(). check_for_illegal_area() checks for overlay of sg elements with kernel text or rodata. It is called with sg_dma_len(s) instead of s->length as parameter. After the call to ->map_sg(), sg_dma_len() will contain the length of possibly combined sg elements in the DMA address space, and not the individual sg element length, which would be s->length. The check will then use the physical start address of an sg element, and add the DMA length for the overlap check, which could result in the false warning, because the DMA length can be larger than the actual single sg element length. In addition, the call to check_for_illegal_area() happens in the iteration over mapped_ents, which will not include all individual sg elements if any of them were combined in ->map_sg(). Fix this by using s->length instead of sg_dma_len(s). Also put the call to check_for_illegal_area() in a separate loop, iterating over all the individual sg elements ("nents" instead of "mapped_ents"). While at it, as suggested by Robin Murphy, also move check_for_stack() inside the new loop, as it is similarly concerned with validating the individual sg elements. Link: https://lore.kernel.org/lkml/20210705185252.4074653-1-gerald.schaefer@linux.ibm.com Fixes: 884d05970bfb ("dma-debug: use sg_dma_len accessor") Signed-off-by: Gerald Schaefer <gerald.schaefer@linux.ibm.com> Signed-off-by: Christoph Hellwig <hch@lst.de>
2021-10-06 23:19:43 +03:00
for_each_sg(sg, s, nents, i) {
check_for_stack(dev, sg_page(s), s->offset);
if (!PageHighMem(sg_page(s)))
check_for_illegal_area(dev, sg_virt(s), s->length);
}
for_each_sg(sg, s, mapped_ents, i) {
entry = dma_entry_alloc();
if (!entry)
return;
entry->type = dma_debug_sg;
entry->dev = dev;
entry->pfn = page_to_pfn(sg_page(s));
entry->offset = s->offset;
entry->size = sg_dma_len(s);
entry->dev_addr = sg_dma_address(s);
entry->direction = direction;
entry->sg_call_ents = nents;
entry->sg_mapped_ents = mapped_ents;
check_sg_segment(dev, s);
add_dma_entry(entry, attrs);
}
}
static int get_nr_mapped_entries(struct device *dev,
struct dma_debug_entry *ref)
{
struct dma_debug_entry *entry;
struct hash_bucket *bucket;
unsigned long flags;
int mapped_ents;
bucket = get_hash_bucket(ref, &flags);
dma-debug: hash_bucket_find needs to allow for offsets within an entry Summary: Users of the pci_dma_sync_single_* api allow users to sync address ranges within the range of a mapped entry (i.e. you can dma map address X to dma_addr_t A and then pci_dma_sync_single on dma_addr_t A+1. The dma-debug library however assume dma syncs will always occur using the base address of a mapped region, and uses that assumption to find entries in its hash table. Since thats often (but not always the case), the dma debug library can give us false errors about missing entries, which are reported as syncing of memory not allocated by the driver. This was noted in the cxgb3 driver as this error: WARNING: at lib/dma-debug.c:902 check_sync+0xdd/0x48c() Hardware name: To be filled by O.E.M. cxgb3 0000:01:00.0: DMA-API: device driver tries to sync DMA memory it has not allocated [device address=0x00000000fff97800] [size=1984 bytes] Modules linked in: autofs4 sunrpc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 uinput snd_hda_codec_intelhdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm snd_timer e1000e snd soundcore r8169 cxgb3 iTCO_wdt snd_page_alloc mii shpchp i2c_i801 iTCO_vendor_support mdio microcode firewire_ohci firewire_core crc_itu_t ata_generic pata_acpi i915 drm_kms_helper drm i2c_algo_bit i2c_core video output [last unloaded: scsi_wait_scan] Pid: 1818, comm: ifconfig Not tainted 2.6.35-0.23.rc3.git6.fc14.x86_64 #1 Call Trace: [<ffffffff81050f71>] warn_slowpath_common+0x85/0x9d [<ffffffff8105102c>] warn_slowpath_fmt+0x46/0x48 [<ffffffff8124658e>] ? check_sync+0x39/0x48c [<ffffffff8107c470>] ? trace_hardirqs_on+0xd/0xf [<ffffffff81246632>] check_sync+0xdd/0x48c [<ffffffff81246ca6>] debug_dma_sync_single_for_device+0x3f/0x41 [<ffffffffa011615c>] ? pci_map_page+0x84/0x97 [cxgb3] [<ffffffffa0117bc3>] pci_dma_sync_single_for_device.clone.0+0x65/0x6e [cxgb3] [<ffffffffa0117ed1>] refill_fl+0x305/0x30a [cxgb3] [<ffffffffa011857d>] t3_sge_alloc_qset+0x6a7/0x821 [cxgb3] [<ffffffffa010a07b>] cxgb_up+0x4d0/0xe62 [cxgb3] [<ffffffff81086037>] ? __module_text_address+0x12/0x58 [<ffffffffa010aa4c>] cxgb_open+0x3f/0x309 [cxgb3] [<ffffffff813e9f6c>] __dev_open+0x8e/0xbc [<ffffffff813e7ca5>] __dev_change_flags+0xbe/0x142 [<ffffffff813e9ea8>] dev_change_flags+0x21/0x57 [<ffffffff81445937>] devinet_ioctl+0x29a/0x54b [<ffffffff811f9a87>] ? inode_has_perm+0xaa/0xce [<ffffffff81446ed2>] inet_ioctl+0x8f/0xa7 [<ffffffff813d683a>] sock_do_ioctl+0x29/0x48 [<ffffffff813d6c83>] sock_ioctl+0x213/0x222 [<ffffffff81137f78>] vfs_ioctl+0x32/0xa6 [<ffffffff811384e2>] do_vfs_ioctl+0x47a/0x4b3 [<ffffffff81138571>] sys_ioctl+0x56/0x79 [<ffffffff81009c32>] system_call_fastpath+0x16/0x1b ---[ end trace 69a4d4cc77b58004 ]--- (some edits by Joerg Roedel) Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Jay Fenalson <fenlason@redhat.com> CC: Divy LeRay <divy@chelsio.com> CC: Stanislaw Gruszka <sgruszka@redhat.com> CC: Joerg Roedel <joerg.roedel@amd.com> CC: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
2011-08-08 23:13:54 +04:00
entry = bucket_find_exact(bucket, ref);
mapped_ents = 0;
if (entry)
mapped_ents = entry->sg_mapped_ents;
put_hash_bucket(bucket, flags);
return mapped_ents;
}
void debug_dma_unmap_sg(struct device *dev, struct scatterlist *sglist,
int nelems, int dir)
{
struct scatterlist *s;
int mapped_ents = 0, i;
if (unlikely(dma_debug_disabled()))
return;
for_each_sg(sglist, s, nelems, i) {
struct dma_debug_entry ref = {
.type = dma_debug_sg,
.dev = dev,
.pfn = page_to_pfn(sg_page(s)),
.offset = s->offset,
.dev_addr = sg_dma_address(s),
.size = sg_dma_len(s),
.direction = dir,
.sg_call_ents = nelems,
};
if (mapped_ents && i >= mapped_ents)
break;
if (!i)
mapped_ents = get_nr_mapped_entries(dev, &ref);
check_unmap(&ref);
}
}
void debug_dma_alloc_coherent(struct device *dev, size_t size,
dma_addr_t dma_addr, void *virt,
unsigned long attrs)
{
struct dma_debug_entry *entry;
if (unlikely(dma_debug_disabled()))
return;
if (unlikely(virt == NULL))
return;
/* handle vmalloc and linear addresses */
if (!is_vmalloc_addr(virt) && !virt_addr_valid(virt))
return;
entry = dma_entry_alloc();
if (!entry)
2017-11-18 02:26:19 +03:00
return;
entry->type = dma_debug_coherent;
entry->dev = dev;
entry->offset = offset_in_page(virt);
entry->size = size;
entry->dev_addr = dma_addr;
entry->direction = DMA_BIDIRECTIONAL;
2017-11-18 02:26:19 +03:00
if (is_vmalloc_addr(virt))
entry->pfn = vmalloc_to_pfn(virt);
else
entry->pfn = page_to_pfn(virt_to_page(virt));
add_dma_entry(entry, attrs);
}
void debug_dma_free_coherent(struct device *dev, size_t size,
void *virt, dma_addr_t dma_addr)
{
struct dma_debug_entry ref = {
.type = dma_debug_coherent,
.dev = dev,
.offset = offset_in_page(virt),
.dev_addr = dma_addr,
.size = size,
.direction = DMA_BIDIRECTIONAL,
};
2017-11-18 02:26:19 +03:00
/* handle vmalloc and linear addresses */
if (!is_vmalloc_addr(virt) && !virt_addr_valid(virt))
2017-11-18 02:26:19 +03:00
return;
if (is_vmalloc_addr(virt))
ref.pfn = vmalloc_to_pfn(virt);
else
ref.pfn = page_to_pfn(virt_to_page(virt));
if (unlikely(dma_debug_disabled()))
return;
check_unmap(&ref);
}
void debug_dma_map_resource(struct device *dev, phys_addr_t addr, size_t size,
int direction, dma_addr_t dma_addr,
unsigned long attrs)
{
struct dma_debug_entry *entry;
if (unlikely(dma_debug_disabled()))
return;
entry = dma_entry_alloc();
if (!entry)
return;
entry->type = dma_debug_resource;
entry->dev = dev;
entry->pfn = PHYS_PFN(addr);
entry->offset = offset_in_page(addr);
entry->size = size;
entry->dev_addr = dma_addr;
entry->direction = direction;
entry->map_err_type = MAP_ERR_NOT_CHECKED;
add_dma_entry(entry, attrs);
}
void debug_dma_unmap_resource(struct device *dev, dma_addr_t dma_addr,
size_t size, int direction)
{
struct dma_debug_entry ref = {
.type = dma_debug_resource,
.dev = dev,
.dev_addr = dma_addr,
.size = size,
.direction = direction,
};
if (unlikely(dma_debug_disabled()))
return;
check_unmap(&ref);
}
void debug_dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
size_t size, int direction)
{
struct dma_debug_entry ref;
if (unlikely(dma_debug_disabled()))
return;
ref.type = dma_debug_single;
ref.dev = dev;
ref.dev_addr = dma_handle;
ref.size = size;
ref.direction = direction;
ref.sg_call_ents = 0;
check_sync(dev, &ref, true);
}
void debug_dma_sync_single_for_device(struct device *dev,
dma_addr_t dma_handle, size_t size,
int direction)
{
struct dma_debug_entry ref;
if (unlikely(dma_debug_disabled()))
return;
ref.type = dma_debug_single;
ref.dev = dev;
ref.dev_addr = dma_handle;
ref.size = size;
ref.direction = direction;
ref.sg_call_ents = 0;
check_sync(dev, &ref, false);
}
void debug_dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
int nelems, int direction)
{
struct scatterlist *s;
int mapped_ents = 0, i;
if (unlikely(dma_debug_disabled()))
return;
for_each_sg(sg, s, nelems, i) {
struct dma_debug_entry ref = {
.type = dma_debug_sg,
.dev = dev,
.pfn = page_to_pfn(sg_page(s)),
.offset = s->offset,
.dev_addr = sg_dma_address(s),
.size = sg_dma_len(s),
.direction = direction,
.sg_call_ents = nelems,
};
if (!i)
mapped_ents = get_nr_mapped_entries(dev, &ref);
if (i >= mapped_ents)
break;
check_sync(dev, &ref, true);
}
}
void debug_dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
int nelems, int direction)
{
struct scatterlist *s;
int mapped_ents = 0, i;
if (unlikely(dma_debug_disabled()))
return;
for_each_sg(sg, s, nelems, i) {
struct dma_debug_entry ref = {
.type = dma_debug_sg,
.dev = dev,
.pfn = page_to_pfn(sg_page(s)),
.offset = s->offset,
.dev_addr = sg_dma_address(s),
.size = sg_dma_len(s),
.direction = direction,
.sg_call_ents = nelems,
};
if (!i)
mapped_ents = get_nr_mapped_entries(dev, &ref);
if (i >= mapped_ents)
break;
check_sync(dev, &ref, false);
}
}
static int __init dma_debug_driver_setup(char *str)
{
int i;
for (i = 0; i < NAME_MAX_LEN - 1; ++i, ++str) {
current_driver_name[i] = *str;
if (*str == 0)
break;
}
if (current_driver_name[0])
pr_info("enable driver filter for driver [%s]\n",
current_driver_name);
return 1;
}
__setup("dma_debug_driver=", dma_debug_driver_setup);