treewide: use kv[mz]alloc* rather than opencoded variants
authorMichal Hocko <mhocko@suse.com>
Mon, 8 May 2017 22:57:27 +0000 (15:57 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Tue, 9 May 2017 00:15:13 +0000 (17:15 -0700)
There are many code paths opencoding kvmalloc.  Let's use the helper
instead.  The main difference to kvmalloc is that those users are
usually not considering all the aspects of the memory allocator.  E.g.
allocation requests <= 32kB (with 4kB pages) are basically never failing
and invoke OOM killer to satisfy the allocation.  This sounds too
disruptive for something that has a reasonable fallback - the vmalloc.
On the other hand those requests might fallback to vmalloc even when the
memory allocator would succeed after several more reclaim/compaction
attempts previously.  There is no guarantee something like that happens
though.

This patch converts many of those places to kv[mz]alloc* helpers because
they are more conservative.

Link: http://lkml.kernel.org/r/20170306103327.2766-2-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> # Xen bits
Acked-by: Kees Cook <keescook@chromium.org>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Andreas Dilger <andreas.dilger@intel.com> # Lustre
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com> # KVM/s390
Acked-by: Dan Williams <dan.j.williams@intel.com> # nvdim
Acked-by: David Sterba <dsterba@suse.com> # btrfs
Acked-by: Ilya Dryomov <idryomov@gmail.com> # Ceph
Acked-by: Tariq Toukan <tariqt@mellanox.com> # mlx4
Acked-by: Leon Romanovsky <leonro@mellanox.com> # mlx5
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Anton Vorontsov <anton@enomsg.org>
Cc: Colin Cross <ccross@android.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Cc: Ben Skeggs <bskeggs@redhat.com>
Cc: Kent Overstreet <kent.overstreet@gmail.com>
Cc: Santosh Raspatur <santosh@chelsio.com>
Cc: Hariprasad S <hariprasad@chelsio.com>
Cc: Yishai Hadas <yishaih@mellanox.com>
Cc: Oleg Drokin <oleg.drokin@intel.com>
Cc: "Yan, Zheng" <zyan@redhat.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
44 files changed:
arch/s390/kvm/kvm-s390.c
crypto/lzo.c
drivers/acpi/apei/erst.c
drivers/char/agp/generic.c
drivers/gpu/drm/nouveau/nouveau_gem.c
drivers/md/bcache/util.h
drivers/net/ethernet/chelsio/cxgb3/cxgb3_defs.h
drivers/net/ethernet/chelsio/cxgb3/cxgb3_offload.c
drivers/net/ethernet/chelsio/cxgb3/l2t.c
drivers/net/ethernet/chelsio/cxgb3/l2t.h
drivers/net/ethernet/chelsio/cxgb4/clip_tbl.c
drivers/net/ethernet/chelsio/cxgb4/cxgb4.h
drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_u32.c
drivers/net/ethernet/chelsio/cxgb4/l2t.c
drivers/net/ethernet/chelsio/cxgb4/sched.c
drivers/net/ethernet/mellanox/mlx4/en_tx.c
drivers/net/ethernet/mellanox/mlx4/mr.c
drivers/nvdimm/dimm_devs.c
drivers/staging/lustre/lnet/libcfs/linux/linux-mem.c
drivers/xen/evtchn.c
fs/btrfs/ctree.c
fs/btrfs/ioctl.c
fs/btrfs/send.c
fs/ceph/file.c
fs/select.c
fs/xattr.c
include/linux/mlx5/driver.h
include/linux/mm.h
lib/iov_iter.c
mm/frame_vector.c
net/ipv4/inet_hashtables.c
net/ipv4/tcp_metrics.c
net/mpls/af_mpls.c
net/netfilter/x_tables.c
net/netfilter/xt_recent.c
net/sched/sch_choke.c
net/sched/sch_fq_codel.c
net/sched/sch_hhf.c
net/sched/sch_netem.c
net/sched/sch_sfq.c
security/keys/keyctl.c

index d5c5c91..323297e 100644 (file)
@@ -1166,10 +1166,7 @@ static long kvm_s390_get_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
        if (args->count < 1 || args->count > KVM_S390_SKEYS_MAX)
                return -EINVAL;
 
-       keys = kmalloc_array(args->count, sizeof(uint8_t),
-                            GFP_KERNEL | __GFP_NOWARN);
-       if (!keys)
-               keys = vmalloc(sizeof(uint8_t) * args->count);
+       keys = kvmalloc_array(args->count, sizeof(uint8_t), GFP_KERNEL);
        if (!keys)
                return -ENOMEM;
 
@@ -1211,10 +1208,7 @@ static long kvm_s390_set_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
        if (args->count < 1 || args->count > KVM_S390_SKEYS_MAX)
                return -EINVAL;
 
-       keys = kmalloc_array(args->count, sizeof(uint8_t),
-                            GFP_KERNEL | __GFP_NOWARN);
-       if (!keys)
-               keys = vmalloc(sizeof(uint8_t) * args->count);
+       keys = kvmalloc_array(args->count, sizeof(uint8_t), GFP_KERNEL);
        if (!keys)
                return -ENOMEM;
 
index 168df78..218567d 100644 (file)
@@ -32,9 +32,7 @@ static void *lzo_alloc_ctx(struct crypto_scomp *tfm)
 {
        void *ctx;
 
-       ctx = kmalloc(LZO1X_MEM_COMPRESS, GFP_KERNEL | __GFP_NOWARN);
-       if (!ctx)
-               ctx = vmalloc(LZO1X_MEM_COMPRESS);
+       ctx = kvmalloc(LZO1X_MEM_COMPRESS, GFP_KERNEL);
        if (!ctx)
                return ERR_PTR(-ENOMEM);
 
index 7207e5f..2c462be 100644 (file)
@@ -513,7 +513,7 @@ retry:
        if (i < erst_record_id_cache.len)
                goto retry;
        if (erst_record_id_cache.len >= erst_record_id_cache.size) {
-               int new_size, alloc_size;
+               int new_size;
                u64 *new_entries;
 
                new_size = erst_record_id_cache.size * 2;
@@ -524,11 +524,7 @@ retry:
                                pr_warn(FW_WARN "too many record IDs!\n");
                        return 0;
                }
-               alloc_size = new_size * sizeof(entries[0]);
-               if (alloc_size < PAGE_SIZE)
-                       new_entries = kmalloc(alloc_size, GFP_KERNEL);
-               else
-                       new_entries = vmalloc(alloc_size);
+               new_entries = kvmalloc(new_size * sizeof(entries[0]), GFP_KERNEL);
                if (!new_entries)
                        return -ENOMEM;
                memcpy(new_entries, entries,
index f002fa5..bdf418c 100644 (file)
@@ -88,13 +88,7 @@ static int agp_get_key(void)
 
 void agp_alloc_page_array(size_t size, struct agp_memory *mem)
 {
-       mem->pages = NULL;
-
-       if (size <= 2*PAGE_SIZE)
-               mem->pages = kmalloc(size, GFP_KERNEL | __GFP_NOWARN);
-       if (mem->pages == NULL) {
-               mem->pages = vmalloc(size);
-       }
+       mem->pages = kvmalloc(size, GFP_KERNEL);
 }
 EXPORT_SYMBOL(agp_alloc_page_array);
 
index ca5397b..2170534 100644 (file)
@@ -568,9 +568,7 @@ u_memcpya(uint64_t user, unsigned nmemb, unsigned size)
 
        size *= nmemb;
 
-       mem = kmalloc(size, GFP_KERNEL | __GFP_NOWARN);
-       if (!mem)
-               mem = vmalloc(size);
+       mem = kvmalloc(size, GFP_KERNEL);
        if (!mem)
                return ERR_PTR(-ENOMEM);
 
index 5d13930..cb8d2cc 100644 (file)
@@ -43,11 +43,7 @@ struct closure;
        (heap)->used = 0;                                               \
        (heap)->size = (_size);                                         \
        _bytes = (heap)->size * sizeof(*(heap)->data);                  \
-       (heap)->data = NULL;                                            \
-       if (_bytes < KMALLOC_MAX_SIZE)                                  \
-               (heap)->data = kmalloc(_bytes, (gfp));                  \
-       if ((!(heap)->data) && ((gfp) & GFP_KERNEL))                    \
-               (heap)->data = vmalloc(_bytes);                         \
+       (heap)->data = kvmalloc(_bytes, (gfp) & GFP_KERNEL);            \
        (heap)->data;                                                   \
 })
 
@@ -136,12 +132,8 @@ do {                                                                       \
                                                                        \
        (fifo)->mask = _allocated_size - 1;                             \
        (fifo)->front = (fifo)->back = 0;                               \
-       (fifo)->data = NULL;                                            \
                                                                        \
-       if (_bytes < KMALLOC_MAX_SIZE)                                  \
-               (fifo)->data = kmalloc(_bytes, (gfp));                  \
-       if ((!(fifo)->data) && ((gfp) & GFP_KERNEL))                    \
-               (fifo)->data = vmalloc(_bytes);                         \
+       (fifo)->data = kvmalloc(_bytes, (gfp) & GFP_KERNEL);            \
        (fifo)->data;                                                   \
 })
 
index 920d918..f04e81f 100644 (file)
@@ -41,9 +41,6 @@
 
 #define VALIDATE_TID 1
 
-void *cxgb_alloc_mem(unsigned long size);
-void cxgb_free_mem(void *addr);
-
 /*
  * Map an ATID or STID to their entries in the corresponding TID tables.
  */
index 76684dc..fa81445 100644 (file)
@@ -1152,27 +1152,6 @@ static void cxgb_redirect(struct dst_entry *old, struct dst_entry *new,
 }
 
 /*
- * Allocate a chunk of memory using kmalloc or, if that fails, vmalloc.
- * The allocated memory is cleared.
- */
-void *cxgb_alloc_mem(unsigned long size)
-{
-       void *p = kzalloc(size, GFP_KERNEL | __GFP_NOWARN);
-
-       if (!p)
-               p = vzalloc(size);
-       return p;
-}
-
-/*
- * Free memory allocated through t3_alloc_mem().
- */
-void cxgb_free_mem(void *addr)
-{
-       kvfree(addr);
-}
-
-/*
  * Allocate and initialize the TID tables.  Returns 0 on success.
  */
 static int init_tid_tabs(struct tid_info *t, unsigned int ntids,
@@ -1182,7 +1161,7 @@ static int init_tid_tabs(struct tid_info *t, unsigned int ntids,
        unsigned long size = ntids * sizeof(*t->tid_tab) +
            natids * sizeof(*t->atid_tab) + nstids * sizeof(*t->stid_tab);
 
-       t->tid_tab = cxgb_alloc_mem(size);
+       t->tid_tab = kvzalloc(size, GFP_KERNEL);
        if (!t->tid_tab)
                return -ENOMEM;
 
@@ -1218,7 +1197,7 @@ static int init_tid_tabs(struct tid_info *t, unsigned int ntids,
 
 static void free_tid_maps(struct tid_info *t)
 {
-       cxgb_free_mem(t->tid_tab);
+       kvfree(t->tid_tab);
 }
 
 static inline void add_adapter(struct adapter *adap)
@@ -1293,7 +1272,7 @@ int cxgb3_offload_activate(struct adapter *adapter)
        return 0;
 
 out_free_l2t:
-       t3_free_l2t(l2td);
+       kvfree(l2td);
 out_free:
        kfree(t);
        return err;
@@ -1302,7 +1281,7 @@ out_free:
 static void clean_l2_data(struct rcu_head *head)
 {
        struct l2t_data *d = container_of(head, struct l2t_data, rcu_head);
-       t3_free_l2t(d);
+       kvfree(d);
 }
 
 
index 5206358..2626412 100644 (file)
@@ -444,7 +444,7 @@ struct l2t_data *t3_init_l2t(unsigned int l2t_capacity)
        struct l2t_data *d;
        int i, size = sizeof(*d) + l2t_capacity * sizeof(struct l2t_entry);
 
-       d = cxgb_alloc_mem(size);
+       d = kvzalloc(size, GFP_KERNEL);
        if (!d)
                return NULL;
 
@@ -462,9 +462,3 @@ struct l2t_data *t3_init_l2t(unsigned int l2t_capacity)
        }
        return d;
 }
-
-void t3_free_l2t(struct l2t_data *d)
-{
-       cxgb_free_mem(d);
-}
-
index 8cffcdf..c2fd323 100644 (file)
@@ -115,7 +115,6 @@ int t3_l2t_send_slow(struct t3cdev *dev, struct sk_buff *skb,
                     struct l2t_entry *e);
 void t3_l2t_send_event(struct t3cdev *dev, struct l2t_entry *e);
 struct l2t_data *t3_init_l2t(unsigned int l2t_capacity);
-void t3_free_l2t(struct l2t_data *d);
 
 int cxgb3_ofld_send(struct t3cdev *dev, struct sk_buff *skb);
 
index 7ad43af..3103ef9 100644 (file)
@@ -290,8 +290,8 @@ struct clip_tbl *t4_init_clip_tbl(unsigned int clipt_start,
        if (clipt_size < CLIPT_MIN_HASH_BUCKETS)
                return NULL;
 
-       ctbl = t4_alloc_mem(sizeof(*ctbl) +
-                           clipt_size*sizeof(struct list_head));
+       ctbl = kvzalloc(sizeof(*ctbl) +
+                           clipt_size*sizeof(struct list_head), GFP_KERNEL);
        if (!ctbl)
                return NULL;
 
@@ -305,9 +305,9 @@ struct clip_tbl *t4_init_clip_tbl(unsigned int clipt_start,
        for (i = 0; i < ctbl->clipt_size; ++i)
                INIT_LIST_HEAD(&ctbl->hash_list[i]);
 
-       cl_list = t4_alloc_mem(clipt_size*sizeof(struct clip_entry));
+       cl_list = kvzalloc(clipt_size*sizeof(struct clip_entry), GFP_KERNEL);
        if (!cl_list) {
-               t4_free_mem(ctbl);
+               kvfree(ctbl);
                return NULL;
        }
        ctbl->cl_list = (void *)cl_list;
@@ -326,8 +326,8 @@ void t4_cleanup_clip_tbl(struct adapter *adap)
 
        if (ctbl) {
                if (ctbl->cl_list)
-                       t4_free_mem(ctbl->cl_list);
-               t4_free_mem(ctbl);
+                       kvfree(ctbl->cl_list);
+               kvfree(ctbl);
        }
 }
 EXPORT_SYMBOL(t4_cleanup_clip_tbl);
index 163543b..1d2be2d 100644 (file)
@@ -1184,8 +1184,6 @@ extern const char cxgb4_driver_version[];
 void t4_os_portmod_changed(const struct adapter *adap, int port_id);
 void t4_os_link_changed(struct adapter *adap, int port_id, int link_stat);
 
-void *t4_alloc_mem(size_t size);
-
 void t4_free_sge_resources(struct adapter *adap);
 void t4_free_ofld_rxqs(struct adapter *adap, int n, struct sge_ofld_rxq *q);
 irq_handler_t t4_intr_handler(struct adapter *adap);
@@ -1557,7 +1555,6 @@ int t4_sched_params(struct adapter *adapter, int type, int level, int mode,
                    int rateunit, int ratemode, int channel, int class,
                    int minrate, int maxrate, int weight, int pktsize);
 void t4_sge_decode_idma_state(struct adapter *adapter, int state);
-void t4_free_mem(void *addr);
 void t4_idma_monitor_init(struct adapter *adapter,
                          struct sge_idma_monitor_state *idma);
 void t4_idma_monitor(struct adapter *adapter,
index f6e739d..1fa34b0 100644 (file)
@@ -2634,7 +2634,7 @@ static ssize_t mem_read(struct file *file, char __user *buf, size_t count,
        if (count > avail - pos)
                count = avail - pos;
 
-       data = t4_alloc_mem(count);
+       data = kvzalloc(count, GFP_KERNEL);
        if (!data)
                return -ENOMEM;
 
@@ -2642,12 +2642,12 @@ static ssize_t mem_read(struct file *file, char __user *buf, size_t count,
        ret = t4_memory_rw(adap, 0, mem, pos, count, data, T4_MEMORY_READ);
        spin_unlock(&adap->win0_lock);
        if (ret) {
-               t4_free_mem(data);
+               kvfree(data);
                return ret;
        }
        ret = copy_to_user(buf, data, count);
 
-       t4_free_mem(data);
+       kvfree(data);
        if (ret)
                return -EFAULT;
 
@@ -2753,7 +2753,7 @@ static ssize_t blocked_fl_read(struct file *filp, char __user *ubuf,
                       adap->sge.egr_sz, adap->sge.blocked_fl);
        len += sprintf(buf + len, "\n");
        size = simple_read_from_buffer(ubuf, count, ppos, buf, len);
-       t4_free_mem(buf);
+       kvfree(buf);
        return size;
 }
 
@@ -2773,7 +2773,7 @@ static ssize_t blocked_fl_write(struct file *filp, const char __user *ubuf,
                return err;
 
        bitmap_copy(adap->sge.blocked_fl, t, adap->sge.egr_sz);
-       t4_free_mem(t);
+       kvfree(t);
        return count;
 }
 
index 02f80fe..0ba7866 100644 (file)
@@ -969,7 +969,7 @@ static int get_eeprom(struct net_device *dev, struct ethtool_eeprom *e,
 {
        int i, err = 0;
        struct adapter *adapter = netdev2adap(dev);
-       u8 *buf = t4_alloc_mem(EEPROMSIZE);
+       u8 *buf = kvzalloc(EEPROMSIZE, GFP_KERNEL);
 
        if (!buf)
                return -ENOMEM;
@@ -980,7 +980,7 @@ static int get_eeprom(struct net_device *dev, struct ethtool_eeprom *e,
 
        if (!err)
                memcpy(data, buf + e->offset, e->len);
-       t4_free_mem(buf);
+       kvfree(buf);
        return err;
 }
 
@@ -1009,7 +1009,7 @@ static int set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom,
        if (aligned_offset != eeprom->offset || aligned_len != eeprom->len) {
                /* RMW possibly needed for first or last words.
                 */
-               buf = t4_alloc_mem(aligned_len);
+               buf = kvzalloc(aligned_len, GFP_KERNEL);
                if (!buf)
                        return -ENOMEM;
                err = eeprom_rd_phys(adapter, aligned_offset, (u32 *)buf);
@@ -1037,7 +1037,7 @@ static int set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom,
                err = t4_seeprom_wp(adapter, true);
 out:
        if (buf != data)
-               t4_free_mem(buf);
+               kvfree(buf);
        return err;
 }
 
index c12c4a3..38a5c67 100644 (file)
@@ -880,27 +880,6 @@ freeout:
        return err;
 }
 
-/*
- * Allocate a chunk of memory using kmalloc or, if that fails, vmalloc.
- * The allocated memory is cleared.
- */
-void *t4_alloc_mem(size_t size)
-{
-       void *p = kzalloc(size, GFP_KERNEL | __GFP_NOWARN);
-
-       if (!p)
-               p = vzalloc(size);
-       return p;
-}
-
-/*
- * Free memory allocated through alloc_mem().
- */
-void t4_free_mem(void *addr)
-{
-       kvfree(addr);
-}
-
 static u16 cxgb_select_queue(struct net_device *dev, struct sk_buff *skb,
                             void *accel_priv, select_queue_fallback_t fallback)
 {
@@ -1299,7 +1278,7 @@ static int tid_init(struct tid_info *t)
               max_ftids * sizeof(*t->ftid_tab) +
               ftid_bmap_size * sizeof(long);
 
-       t->tid_tab = t4_alloc_mem(size);
+       t->tid_tab = kvzalloc(size, GFP_KERNEL);
        if (!t->tid_tab)
                return -ENOMEM;
 
@@ -3445,7 +3424,7 @@ static int adap_init0(struct adapter *adap)
                /* allocate memory to read the header of the firmware on the
                 * card
                 */
-               card_fw = t4_alloc_mem(sizeof(*card_fw));
+               card_fw = kvzalloc(sizeof(*card_fw), GFP_KERNEL);
 
                /* Get FW from from /lib/firmware/ */
                ret = request_firmware(&fw, fw_info->fw_mod_name,
@@ -3465,7 +3444,7 @@ static int adap_init0(struct adapter *adap)
 
                /* Cleaning up */
                release_firmware(fw);
-               t4_free_mem(card_fw);
+               kvfree(card_fw);
 
                if (ret < 0)
                        goto bye;
@@ -4470,9 +4449,9 @@ static void free_some_resources(struct adapter *adapter)
 {
        unsigned int i;
 
-       t4_free_mem(adapter->l2t);
+       kvfree(adapter->l2t);
        t4_cleanup_sched(adapter);
-       t4_free_mem(adapter->tids.tid_tab);
+       kvfree(adapter->tids.tid_tab);
        cxgb4_cleanup_tc_u32(adapter);
        kfree(adapter->sge.egr_map);
        kfree(adapter->sge.ingr_map);
index a1b1942..ef06ce8 100644 (file)
@@ -432,9 +432,9 @@ void cxgb4_cleanup_tc_u32(struct adapter *adap)
        for (i = 0; i < t->size; i++) {
                struct cxgb4_link *link = &t->table[i];
 
-               t4_free_mem(link->tid_map);
+               kvfree(link->tid_map);
        }
-       t4_free_mem(adap->tc_u32);
+       kvfree(adap->tc_u32);
 }
 
 struct cxgb4_tc_u32_table *cxgb4_init_tc_u32(struct adapter *adap)
@@ -446,8 +446,8 @@ struct cxgb4_tc_u32_table *cxgb4_init_tc_u32(struct adapter *adap)
        if (!max_tids)
                return NULL;
 
-       t = t4_alloc_mem(sizeof(*t) +
-                        (max_tids * sizeof(struct cxgb4_link)));
+       t = kvzalloc(sizeof(*t) +
+                        (max_tids * sizeof(struct cxgb4_link)), GFP_KERNEL);
        if (!t)
                return NULL;
 
@@ -458,7 +458,7 @@ struct cxgb4_tc_u32_table *cxgb4_init_tc_u32(struct adapter *adap)
                unsigned int bmap_size;
 
                bmap_size = BITS_TO_LONGS(max_tids);
-               link->tid_map = t4_alloc_mem(sizeof(unsigned long) * bmap_size);
+               link->tid_map = kvzalloc(sizeof(unsigned long) * bmap_size, GFP_KERNEL);
                if (!link->tid_map)
                        goto out_no_mem;
                bitmap_zero(link->tid_map, max_tids);
@@ -471,11 +471,11 @@ out_no_mem:
                struct cxgb4_link *link = &t->table[i];
 
                if (link->tid_map)
-                       t4_free_mem(link->tid_map);
+                       kvfree(link->tid_map);
        }
 
        if (t)
-               t4_free_mem(t);
+               kvfree(t);
 
        return NULL;
 }
index 7c8c5b9..6f3692d 100644 (file)
@@ -646,7 +646,7 @@ struct l2t_data *t4_init_l2t(unsigned int l2t_start, unsigned int l2t_end)
        if (l2t_size < L2T_MIN_HASH_BUCKETS)
                return NULL;
 
-       d = t4_alloc_mem(sizeof(*d) + l2t_size * sizeof(struct l2t_entry));
+       d = kvzalloc(sizeof(*d) + l2t_size * sizeof(struct l2t_entry), GFP_KERNEL);
        if (!d)
                return NULL;
 
index c902635..02acff7 100644 (file)
@@ -177,7 +177,7 @@ static int t4_sched_queue_unbind(struct port_info *pi, struct ch_sched_queue *p)
                }
 
                list_del(&qe->list);
-               t4_free_mem(qe);
+               kvfree(qe);
                if (atomic_dec_and_test(&e->refcnt)) {
                        e->state = SCHED_STATE_UNUSED;
                        memset(&e->info, 0, sizeof(e->info));
@@ -201,7 +201,7 @@ static int t4_sched_queue_bind(struct port_info *pi, struct ch_sched_queue *p)
        if (p->queue < 0 || p->queue >= pi->nqsets)
                return -ERANGE;
 
-       qe = t4_alloc_mem(sizeof(struct sched_queue_entry));
+       qe = kvzalloc(sizeof(struct sched_queue_entry), GFP_KERNEL);
        if (!qe)
                return -ENOMEM;
 
@@ -211,7 +211,7 @@ static int t4_sched_queue_bind(struct port_info *pi, struct ch_sched_queue *p)
        /* Unbind queue from any existing class */
        err = t4_sched_queue_unbind(pi, p);
        if (err) {
-               t4_free_mem(qe);
+               kvfree(qe);
                goto out;
        }
 
@@ -224,7 +224,7 @@ static int t4_sched_queue_bind(struct port_info *pi, struct ch_sched_queue *p)
        spin_lock(&e->lock);
        err = t4_sched_bind_unbind_op(pi, (void *)qe, SCHED_QUEUE, true);
        if (err) {
-               t4_free_mem(qe);
+               kvfree(qe);
                spin_unlock(&e->lock);
                goto out;
        }
@@ -512,7 +512,7 @@ struct sched_table *t4_init_sched(unsigned int sched_size)
        struct sched_table *s;
        unsigned int i;
 
-       s = t4_alloc_mem(sizeof(*s) + sched_size * sizeof(struct sched_class));
+       s = kvzalloc(sizeof(*s) + sched_size * sizeof(struct sched_class), GFP_KERNEL);
        if (!s)
                return NULL;
 
@@ -548,6 +548,6 @@ void t4_cleanup_sched(struct adapter *adap)
                                t4_sched_class_free(pi, e);
                        write_unlock(&s->rw_lock);
                }
-               t4_free_mem(s);
+               kvfree(s);
        }
 }
index 3ba89bc..6ffd184 100644 (file)
@@ -70,13 +70,10 @@ int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv,
        ring->full_size = ring->size - HEADROOM - MAX_DESC_TXBBS;
 
        tmp = size * sizeof(struct mlx4_en_tx_info);
-       ring->tx_info = kmalloc_node(tmp, GFP_KERNEL | __GFP_NOWARN, node);
+       ring->tx_info = kvmalloc_node(tmp, GFP_KERNEL, node);
        if (!ring->tx_info) {
-               ring->tx_info = vmalloc(tmp);
-               if (!ring->tx_info) {
-                       err = -ENOMEM;
-                       goto err_ring;
-               }
+               err = -ENOMEM;
+               goto err_ring;
        }
 
        en_dbg(DRV, priv, "Allocated tx_info ring at addr:%p size:%d\n",
index db65f72..ce852ca 100644 (file)
@@ -115,12 +115,9 @@ static int mlx4_buddy_init(struct mlx4_buddy *buddy, int max_order)
 
        for (i = 0; i <= buddy->max_order; ++i) {
                s = BITS_TO_LONGS(1 << (buddy->max_order - i));
-               buddy->bits[i] = kcalloc(s, sizeof (long), GFP_KERNEL | __GFP_NOWARN);
-               if (!buddy->bits[i]) {
-                       buddy->bits[i] = vzalloc(s * sizeof(long));
-                       if (!buddy->bits[i])
-                               goto err_out_free;
-               }
+               buddy->bits[i] = kvmalloc_array(s, sizeof(long), GFP_KERNEL | __GFP_ZERO);
+               if (!buddy->bits[i])
+                       goto err_out_free;
        }
 
        set_bit(0, buddy->bits[buddy->max_order]);
index fac1e9f..9852a33 100644 (file)
@@ -106,10 +106,7 @@ int nvdimm_init_config_data(struct nvdimm_drvdata *ndd)
                return -ENXIO;
        }
 
-       ndd->data = kmalloc(ndd->nsarea.config_size, GFP_KERNEL);
-       if (!ndd->data)
-               ndd->data = vmalloc(ndd->nsarea.config_size);
-
+       ndd->data = kvmalloc(ndd->nsarea.config_size, GFP_KERNEL);
        if (!ndd->data)
                return -ENOMEM;
 
index a6a76a6..8f63826 100644 (file)
@@ -45,15 +45,6 @@ EXPORT_SYMBOL(libcfs_kvzalloc);
 void *libcfs_kvzalloc_cpt(struct cfs_cpt_table *cptab, int cpt, size_t size,
                          gfp_t flags)
 {
-       void *ret;
-
-       ret = kzalloc_node(size, flags | __GFP_NOWARN,
-                          cfs_cpt_spread_node(cptab, cpt));
-       if (!ret) {
-               WARN_ON(!(flags & (__GFP_FS | __GFP_HIGH)));
-               ret = vmalloc_node(size, cfs_cpt_spread_node(cptab, cpt));
-       }
-
-       return ret;
+       return kvzalloc_node(size, flags, cfs_cpt_spread_node(cptab, cpt));
 }
 EXPORT_SYMBOL(libcfs_kvzalloc_cpt);
index 6890897..10f1ef5 100644 (file)
@@ -87,18 +87,6 @@ struct user_evtchn {
        bool enabled;
 };
 
-static evtchn_port_t *evtchn_alloc_ring(unsigned int size)
-{
-       evtchn_port_t *ring;
-       size_t s = size * sizeof(*ring);
-
-       ring = kmalloc(s, GFP_KERNEL);
-       if (!ring)
-               ring = vmalloc(s);
-
-       return ring;
-}
-
 static void evtchn_free_ring(evtchn_port_t *ring)
 {
        kvfree(ring);
@@ -334,7 +322,7 @@ static int evtchn_resize_ring(struct per_user_data *u)
        else
                new_size = 2 * u->ring_size;
 
-       new_ring = evtchn_alloc_ring(new_size);
+       new_ring = kvmalloc(new_size * sizeof(*new_ring), GFP_KERNEL);
        if (!new_ring)
                return -ENOMEM;
 
index 7dc8844..1c3b6c5 100644 (file)
@@ -5392,13 +5392,10 @@ int btrfs_compare_trees(struct btrfs_root *left_root,
                goto out;
        }
 
-       tmp_buf = kmalloc(fs_info->nodesize, GFP_KERNEL | __GFP_NOWARN);
+       tmp_buf = kvmalloc(fs_info->nodesize, GFP_KERNEL);
        if (!tmp_buf) {
-               tmp_buf = vmalloc(fs_info->nodesize);
-               if (!tmp_buf) {
-                       ret = -ENOMEM;
-                       goto out;
-               }
+               ret = -ENOMEM;
+               goto out;
        }
 
        left_path->search_commit_root = 1;
index dabfc7a..922a66f 100644 (file)
@@ -3539,12 +3539,9 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
        u64 last_dest_end = destoff;
 
        ret = -ENOMEM;
-       buf = kmalloc(fs_info->nodesize, GFP_KERNEL | __GFP_NOWARN);
-       if (!buf) {
-               buf = vmalloc(fs_info->nodesize);
-               if (!buf)
-                       return ret;
-       }
+       buf = kvmalloc(fs_info->nodesize, GFP_KERNEL);
+       if (!buf)
+               return ret;
 
        path = btrfs_alloc_path();
        if (!path) {
index a60d5bf..3f645cd 100644 (file)
@@ -6360,22 +6360,16 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
        sctx->clone_roots_cnt = arg->clone_sources_count;
 
        sctx->send_max_size = BTRFS_SEND_BUF_SIZE;
-       sctx->send_buf = kmalloc(sctx->send_max_size, GFP_KERNEL | __GFP_NOWARN);
+       sctx->send_buf = kvmalloc(sctx->send_max_size, GFP_KERNEL);
        if (!sctx->send_buf) {
-               sctx->send_buf = vmalloc(sctx->send_max_size);
-               if (!sctx->send_buf) {
-                       ret = -ENOMEM;
-                       goto out;
-               }
+               ret = -ENOMEM;
+               goto out;
        }
 
-       sctx->read_buf = kmalloc(BTRFS_SEND_READ_SIZE, GFP_KERNEL | __GFP_NOWARN);
+       sctx->read_buf = kvmalloc(BTRFS_SEND_READ_SIZE, GFP_KERNEL);
        if (!sctx->read_buf) {
-               sctx->read_buf = vmalloc(BTRFS_SEND_READ_SIZE);
-               if (!sctx->read_buf) {
-                       ret = -ENOMEM;
-                       goto out;
-               }
+               ret = -ENOMEM;
+               goto out;
        }
 
        sctx->pending_dir_moves = RB_ROOT;
@@ -6396,13 +6390,10 @@ long btrfs_ioctl_send(struct file *mnt_file, void __user *arg_)
        alloc_size = arg->clone_sources_count * sizeof(*arg->clone_sources);
 
        if (arg->clone_sources_count) {
-               clone_sources_tmp = kmalloc(alloc_size, GFP_KERNEL | __GFP_NOWARN);
+               clone_sources_tmp = kvmalloc(alloc_size, GFP_KERNEL);
                if (!clone_sources_tmp) {
-                       clone_sources_tmp = vmalloc(alloc_size);
-                       if (!clone_sources_tmp) {
-                               ret = -ENOMEM;
-                               goto out;
-                       }
+                       ret = -ENOMEM;
+                       goto out;
                }
 
                ret = copy_from_user(clone_sources_tmp, arg->clone_sources,
index 26cc954..18c045e 100644 (file)
@@ -74,12 +74,9 @@ dio_get_pages_alloc(const struct iov_iter *it, size_t nbytes,
        align = (unsigned long)(it->iov->iov_base + it->iov_offset) &
                (PAGE_SIZE - 1);
        npages = calc_pages_for(align, nbytes);
-       pages = kmalloc(sizeof(*pages) * npages, GFP_KERNEL);
-       if (!pages) {
-               pages = vmalloc(sizeof(*pages) * npages);
-               if (!pages)
-                       return ERR_PTR(-ENOMEM);
-       }
+       pages = kvmalloc(sizeof(*pages) * npages, GFP_KERNEL);
+       if (!pages)
+               return ERR_PTR(-ENOMEM);
 
        for (idx = 0; idx < npages; ) {
                size_t start;
index bd4b2cc..d6c652a 100644 (file)
@@ -633,10 +633,7 @@ int core_sys_select(int n, fd_set __user *inp, fd_set __user *outp,
                        goto out_nofds;
 
                alloc_size = 6 * size;
-               bits = kmalloc(alloc_size, GFP_KERNEL|__GFP_NOWARN);
-               if (!bits && alloc_size > PAGE_SIZE)
-                       bits = vmalloc(alloc_size);
-
+               bits = kvmalloc(alloc_size, GFP_KERNEL);
                if (!bits)
                        goto out_nofds;
        }
index 94f49a0..464c94b 100644 (file)
@@ -431,12 +431,9 @@ setxattr(struct dentry *d, const char __user *name, const void __user *value,
        if (size) {
                if (size > XATTR_SIZE_MAX)
                        return -E2BIG;
-               kvalue = kmalloc(size, GFP_KERNEL | __GFP_NOWARN);
-               if (!kvalue) {
-                       kvalue = vmalloc(size);
-                       if (!kvalue)
-                               return -ENOMEM;
-               }
+               kvalue = kvmalloc(size, GFP_KERNEL);
+               if (!kvalue)
+                       return -ENOMEM;
                if (copy_from_user(kvalue, value, size)) {
                        error = -EFAULT;
                        goto out;
@@ -528,12 +525,9 @@ getxattr(struct dentry *d, const char __user *name, void __user *value,
        if (size) {
                if (size > XATTR_SIZE_MAX)
                        size = XATTR_SIZE_MAX;
-               kvalue = kzalloc(size, GFP_KERNEL | __GFP_NOWARN);
-               if (!kvalue) {
-                       kvalue = vzalloc(size);
-                       if (!kvalue)
-                               return -ENOMEM;
-               }
+               kvalue = kvzalloc(size, GFP_KERNEL);
+               if (!kvalue)
+                       return -ENOMEM;
        }
 
        error = vfs_getxattr(d, kname, kvalue, size);
@@ -611,12 +605,9 @@ listxattr(struct dentry *d, char __user *list, size_t size)
        if (size) {
                if (size > XATTR_LIST_MAX)
                        size = XATTR_LIST_MAX;
-               klist = kmalloc(size, __GFP_NOWARN | GFP_KERNEL);
-               if (!klist) {
-                       klist = vmalloc(size);
-                       if (!klist)
-                               return -ENOMEM;
-               }
+               klist = kvmalloc(size, GFP_KERNEL);
+               if (!klist)
+                       return -ENOMEM;
        }
 
        error = vfs_listxattr(d, klist, size);
index 3fece51..18fc65b 100644 (file)
@@ -892,12 +892,7 @@ static inline u16 cmdif_rev(struct mlx5_core_dev *dev)
 
 static inline void *mlx5_vzalloc(unsigned long size)
 {
-       void *rtn;
-
-       rtn = kzalloc(size, GFP_KERNEL | __GFP_NOWARN);
-       if (!rtn)
-               rtn = vzalloc(size);
-       return rtn;
+       return kvzalloc(size, GFP_KERNEL);
 }
 
 static inline u32 mlx5_base_mkey(const u32 key)
index 08e2849..7cb17c6 100644 (file)
@@ -532,6 +532,14 @@ static inline void *kvzalloc(size_t size, gfp_t flags)
        return kvmalloc(size, flags | __GFP_ZERO);
 }
 
+static inline void *kvmalloc_array(size_t n, size_t size, gfp_t flags)
+{
+       if (size != 0 && n > SIZE_MAX / size)
+               return NULL;
+
+       return kvmalloc(n * size, flags);
+}
+
 extern void kvfree(const void *addr);
 
 static inline atomic_t *compound_mapcount_ptr(struct page *page)
index 4952311..ae82d9c 100644 (file)
@@ -1028,10 +1028,7 @@ EXPORT_SYMBOL(iov_iter_get_pages);
 
 static struct page **get_pages_array(size_t n)
 {
-       struct page **p = kmalloc(n * sizeof(struct page *), GFP_KERNEL);
-       if (!p)
-               p = vmalloc(n * sizeof(struct page *));
-       return p;
+       return kvmalloc_array(n, sizeof(struct page *), GFP_KERNEL);
 }
 
 static ssize_t pipe_get_pages_alloc(struct iov_iter *i,
index db77dcb..72ebec1 100644 (file)
@@ -200,10 +200,7 @@ struct frame_vector *frame_vector_create(unsigned int nr_frames)
         * Avoid higher order allocations, use vmalloc instead. It should
         * be rare anyway.
         */
-       if (size <= PAGE_SIZE)
-               vec = kmalloc(size, GFP_KERNEL);
-       else
-               vec = vmalloc(size);
+       vec = kvmalloc(size, GFP_KERNEL);
        if (!vec)
                return NULL;
        vec->nr_allocated = nr_frames;
index 8bea742..e9a59d2 100644 (file)
@@ -678,11 +678,7 @@ int inet_ehash_locks_alloc(struct inet_hashinfo *hashinfo)
                /* no more locks than number of hash buckets */
                nblocks = min(nblocks, hashinfo->ehash_mask + 1);
 
-               hashinfo->ehash_locks = kmalloc_array(nblocks, locksz,
-                                                     GFP_KERNEL | __GFP_NOWARN);
-               if (!hashinfo->ehash_locks)
-                       hashinfo->ehash_locks = vmalloc(nblocks * locksz);
-
+               hashinfo->ehash_locks = kvmalloc_array(nblocks, locksz, GFP_KERNEL);
                if (!hashinfo->ehash_locks)
                        return -ENOMEM;
 
index 9d0d4f3..653bbd6 100644 (file)
@@ -1011,10 +1011,7 @@ static int __net_init tcp_net_metrics_init(struct net *net)
        tcp_metrics_hash_log = order_base_2(slots);
        size = sizeof(struct tcpm_hash_bucket) << tcp_metrics_hash_log;
 
-       tcp_metrics_hash = kzalloc(size, GFP_KERNEL | __GFP_NOWARN);
-       if (!tcp_metrics_hash)
-               tcp_metrics_hash = vzalloc(size);
-
+       tcp_metrics_hash = kvzalloc(size, GFP_KERNEL);
        if (!tcp_metrics_hash)
                return -ENOMEM;
 
index 088e2b4..257ec66 100644 (file)
@@ -2005,10 +2005,7 @@ static int resize_platform_label_table(struct net *net, size_t limit)
        unsigned index;
 
        if (size) {
-               labels = kzalloc(size, GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY);
-               if (!labels)
-                       labels = vzalloc(size);
-
+               labels = kvzalloc(size, GFP_KERNEL);
                if (!labels)
                        goto nolabels;
        }
index f134d38..3d05846 100644 (file)
@@ -763,17 +763,8 @@ EXPORT_SYMBOL(xt_check_entry_offsets);
  */
 unsigned int *xt_alloc_entry_offsets(unsigned int size)
 {
-       unsigned int *off;
+       return kvmalloc_array(size, sizeof(unsigned int), GFP_KERNEL | __GFP_ZERO);
 
-       off = kcalloc(size, sizeof(unsigned int), GFP_KERNEL | __GFP_NOWARN);
-
-       if (off)
-               return off;
-
-       if (size < (SIZE_MAX / sizeof(unsigned int)))
-               off = vmalloc(size * sizeof(unsigned int));
-
-       return off;
 }
 EXPORT_SYMBOL(xt_alloc_entry_offsets);
 
@@ -1116,7 +1107,7 @@ static int xt_jumpstack_alloc(struct xt_table_info *i)
 
        size = sizeof(void **) * nr_cpu_ids;
        if (size > PAGE_SIZE)
-               i->jumpstack = vzalloc(size);
+               i->jumpstack = kvzalloc(size, GFP_KERNEL);
        else
                i->jumpstack = kzalloc(size, GFP_KERNEL);
        if (i->jumpstack == NULL)
@@ -1138,12 +1129,8 @@ static int xt_jumpstack_alloc(struct xt_table_info *i)
         */
        size = sizeof(void *) * i->stacksize * 2u;
        for_each_possible_cpu(cpu) {
-               if (size > PAGE_SIZE)
-                       i->jumpstack[cpu] = vmalloc_node(size,
-                               cpu_to_node(cpu));
-               else
-                       i->jumpstack[cpu] = kmalloc_node(size,
-                               GFP_KERNEL, cpu_to_node(cpu));
+               i->jumpstack[cpu] = kvmalloc_node(size, GFP_KERNEL,
+                       cpu_to_node(cpu));
                if (i->jumpstack[cpu] == NULL)
                        /*
                         * Freeing will be done later on by the callers. The
index 37d581a..3f6c4fa 100644 (file)
@@ -388,10 +388,7 @@ static int recent_mt_check(const struct xt_mtchk_param *par,
        }
 
        sz = sizeof(*t) + sizeof(t->iphash[0]) * ip_list_hash_size;
-       if (sz <= PAGE_SIZE)
-               t = kzalloc(sz, GFP_KERNEL);
-       else
-               t = vzalloc(sz);
+       t = kvzalloc(sz, GFP_KERNEL);
        if (t == NULL) {
                ret = -ENOMEM;
                goto out;
index d00f4c7..b30a2c7 100644 (file)
@@ -376,10 +376,7 @@ static int choke_change(struct Qdisc *sch, struct nlattr *opt)
        if (mask != q->tab_mask) {
                struct sk_buff **ntab;
 
-               ntab = kcalloc(mask + 1, sizeof(struct sk_buff *),
-                              GFP_KERNEL | __GFP_NOWARN);
-               if (!ntab)
-                       ntab = vzalloc((mask + 1) * sizeof(struct sk_buff *));
+               ntab = kvmalloc_array((mask + 1), sizeof(struct sk_buff *), GFP_KERNEL | __GFP_ZERO);
                if (!ntab)
                        return -ENOMEM;
 
index 18bbb54..9201abc 100644 (file)
@@ -446,27 +446,13 @@ static int fq_codel_change(struct Qdisc *sch, struct nlattr *opt)
        return 0;
 }
 
-static void *fq_codel_zalloc(size_t sz)
-{
-       void *ptr = kzalloc(sz, GFP_KERNEL | __GFP_NOWARN);
-
-       if (!ptr)
-               ptr = vzalloc(sz);
-       return ptr;
-}
-
-static void fq_codel_free(void *addr)
-{
-       kvfree(addr);
-}
-
 static void fq_codel_destroy(struct Qdisc *sch)
 {
        struct fq_codel_sched_data *q = qdisc_priv(sch);
 
        tcf_destroy_chain(&q->filter_list);
-       fq_codel_free(q->backlogs);
-       fq_codel_free(q->flows);
+       kvfree(q->backlogs);
+       kvfree(q->flows);
 }
 
 static int fq_codel_init(struct Qdisc *sch, struct nlattr *opt)
@@ -493,13 +479,13 @@ static int fq_codel_init(struct Qdisc *sch, struct nlattr *opt)
        }
 
        if (!q->flows) {
-               q->flows = fq_codel_zalloc(q->flows_cnt *
-                                          sizeof(struct fq_codel_flow));
+               q->flows = kvzalloc(q->flows_cnt *
+                                          sizeof(struct fq_codel_flow), GFP_KERNEL);
                if (!q->flows)
                        return -ENOMEM;
-               q->backlogs = fq_codel_zalloc(q->flows_cnt * sizeof(u32));
+               q->backlogs = kvzalloc(q->flows_cnt * sizeof(u32), GFP_KERNEL);
                if (!q->backlogs) {
-                       fq_codel_free(q->flows);
+                       kvfree(q->flows);
                        return -ENOMEM;
                }
                for (i = 0; i < q->flows_cnt; i++) {
index c19d346..51d3ba6 100644 (file)
@@ -467,29 +467,14 @@ static void hhf_reset(struct Qdisc *sch)
                rtnl_kfree_skbs(skb, skb);
 }
 
-static void *hhf_zalloc(size_t sz)
-{
-       void *ptr = kzalloc(sz, GFP_KERNEL | __GFP_NOWARN);
-
-       if (!ptr)
-               ptr = vzalloc(sz);
-
-       return ptr;
-}
-
-static void hhf_free(void *addr)
-{
-       kvfree(addr);
-}
-
 static void hhf_destroy(struct Qdisc *sch)
 {
        int i;
        struct hhf_sched_data *q = qdisc_priv(sch);
 
        for (i = 0; i < HHF_ARRAYS_CNT; i++) {
-               hhf_free(q->hhf_arrays[i]);
-               hhf_free(q->hhf_valid_bits[i]);
+               kvfree(q->hhf_arrays[i]);
+               kvfree(q->hhf_valid_bits[i]);
        }
 
        for (i = 0; i < HH_FLOWS_CNT; i++) {
@@ -503,7 +488,7 @@ static void hhf_destroy(struct Qdisc *sch)
                        kfree(flow);
                }
        }
-       hhf_free(q->hh_flows);
+       kvfree(q->hh_flows);
 }
 
 static const struct nla_policy hhf_policy[TCA_HHF_MAX + 1] = {
@@ -609,8 +594,8 @@ static int hhf_init(struct Qdisc *sch, struct nlattr *opt)
 
        if (!q->hh_flows) {
                /* Initialize heavy-hitter flow table. */
-               q->hh_flows = hhf_zalloc(HH_FLOWS_CNT *
-                                        sizeof(struct list_head));
+               q->hh_flows = kvzalloc(HH_FLOWS_CNT *
+                                        sizeof(struct list_head), GFP_KERNEL);
                if (!q->hh_flows)
                        return -ENOMEM;
                for (i = 0; i < HH_FLOWS_CNT; i++)
@@ -624,8 +609,8 @@ static int hhf_init(struct Qdisc *sch, struct nlattr *opt)
 
                /* Initialize heavy-hitter filter arrays. */
                for (i = 0; i < HHF_ARRAYS_CNT; i++) {
-                       q->hhf_arrays[i] = hhf_zalloc(HHF_ARRAYS_LEN *
-                                                     sizeof(u32));
+                       q->hhf_arrays[i] = kvzalloc(HHF_ARRAYS_LEN *
+                                                     sizeof(u32), GFP_KERNEL);
                        if (!q->hhf_arrays[i]) {
                                /* Note: hhf_destroy() will be called
                                 * by our caller.
@@ -637,8 +622,8 @@ static int hhf_init(struct Qdisc *sch, struct nlattr *opt)
 
                /* Initialize valid bits of heavy-hitter filter arrays. */
                for (i = 0; i < HHF_ARRAYS_CNT; i++) {
-                       q->hhf_valid_bits[i] = hhf_zalloc(HHF_ARRAYS_LEN /
-                                                         BITS_PER_BYTE);
+                       q->hhf_valid_bits[i] = kvzalloc(HHF_ARRAYS_LEN /
+                                                         BITS_PER_BYTE, GFP_KERNEL);
                        if (!q->hhf_valid_bits[i]) {
                                /* Note: hhf_destroy() will be called
                                 * by our caller.
index f0ce478..1b3dd61 100644 (file)
@@ -702,15 +702,11 @@ static int get_dist_table(struct Qdisc *sch, const struct nlattr *attr)
        spinlock_t *root_lock;
        struct disttable *d;
        int i;
-       size_t s;
 
        if (n > NETEM_DIST_MAX)
                return -EINVAL;
 
-       s = sizeof(struct disttable) + n * sizeof(s16);
-       d = kmalloc(s, GFP_KERNEL | __GFP_NOWARN);
-       if (!d)
-               d = vmalloc(s);
+       d = kvmalloc(sizeof(struct disttable) + n * sizeof(s16), GFP_KERNEL);
        if (!d)
                return -ENOMEM;
 
index b00e02c..332d94b 100644 (file)
@@ -685,11 +685,7 @@ static int sfq_change(struct Qdisc *sch, struct nlattr *opt)
 
 static void *sfq_alloc(size_t sz)
 {
-       void *ptr = kmalloc(sz, GFP_KERNEL | __GFP_NOWARN);
-
-       if (!ptr)
-               ptr = vmalloc(sz);
-       return ptr;
+       return  kvmalloc(sz, GFP_KERNEL);
 }
 
 static void sfq_free(void *addr)
index 82a9e18..447a7d5 100644 (file)
@@ -101,14 +101,9 @@ SYSCALL_DEFINE5(add_key, const char __user *, _type,
 
        if (_payload) {
                ret = -ENOMEM;
-               payload = kmalloc(plen, GFP_KERNEL | __GFP_NOWARN);
-               if (!payload) {
-                       if (plen <= PAGE_SIZE)
-                               goto error2;
-                       payload = vmalloc(plen);
-                       if (!payload)
-                               goto error2;
-               }
+               payload = kvmalloc(plen, GFP_KERNEL);
+               if (!payload)
+                       goto error2;
 
                ret = -EFAULT;
                if (copy_from_user(payload, _payload, plen) != 0)
@@ -1071,14 +1066,9 @@ long keyctl_instantiate_key_common(key_serial_t id,
 
        if (from) {
                ret = -ENOMEM;
-               payload = kmalloc(plen, GFP_KERNEL);
-               if (!payload) {
-                       if (plen <= PAGE_SIZE)
-                               goto error;
-                       payload = vmalloc(plen);
-                       if (!payload)
-                               goto error;
-               }
+               payload = kvmalloc(plen, GFP_KERNEL);
+               if (!payload)
+                       goto error;
 
                ret = -EFAULT;
                if (!copy_from_iter_full(payload, plen, from))