From a9c667f8f0656631ee5438baaf21bf30d5f67375 Mon Sep 17 00:00:00 2001 From: Lukas Czerner Date: Mon, 6 Jun 2011 09:51:52 -0400 Subject: ext4: fixed tracepoints cleanup While creating fixed tracepoints for ext3, basically by porting them from ext4, I found a lot of useless retyping, wrong type usage, useless variable passing and other inconsistencies in the ext4 fixed tracepoint code. This patch cleans the fixed tracepoint code for ext4 and also simplify some of them. Signed-off-by: Lukas Czerner Signed-off-by: "Theodore Ts'o" --- include/trace/events/ext4.h | 179 +++++++++++++++++++------------------------- 1 file changed, 76 insertions(+), 103 deletions(-) (limited to 'include') diff --git a/include/trace/events/ext4.h b/include/trace/events/ext4.h index e09592d2f916..5ce2b2f5f524 100644 --- a/include/trace/events/ext4.h +++ b/include/trace/events/ext4.h @@ -26,7 +26,7 @@ TRACE_EVENT(ext4_free_inode, __field( umode_t, mode ) __field( uid_t, uid ) __field( gid_t, gid ) - __field( blkcnt_t, blocks ) + __field( __u64, blocks ) ), TP_fast_assign( @@ -40,9 +40,8 @@ TRACE_EVENT(ext4_free_inode, TP_printk("dev %d,%d ino %lu mode 0%o uid %u gid %u blocks %llu", MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long) __entry->ino, - __entry->mode, __entry->uid, __entry->gid, - (unsigned long long) __entry->blocks) + (unsigned long) __entry->ino, __entry->mode, + __entry->uid, __entry->gid, __entry->blocks) ); TRACE_EVENT(ext4_request_inode, @@ -178,7 +177,7 @@ TRACE_EVENT(ext4_begin_ordered_truncate, TP_printk("dev %d,%d ino %lu new_size %lld", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - (long long) __entry->new_size) + __entry->new_size) ); DECLARE_EVENT_CLASS(ext4__write_begin, @@ -204,7 +203,7 @@ DECLARE_EVENT_CLASS(ext4__write_begin, __entry->flags = flags; ), - TP_printk("dev %d,%d ino %lu pos %llu len %u flags %u", + TP_printk("dev %d,%d ino %lu pos %lld len %u flags %u", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, __entry->pos, __entry->len, __entry->flags) @@ -248,7 +247,7 @@ DECLARE_EVENT_CLASS(ext4__write_end, __entry->copied = copied; ), - TP_printk("dev %d,%d ino %lu pos %llu len %u copied %u", + TP_printk("dev %d,%d ino %lu pos %lld len %u copied %u", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, __entry->pos, __entry->len, __entry->copied) @@ -286,29 +285,6 @@ DEFINE_EVENT(ext4__write_end, ext4_da_write_end, TP_ARGS(inode, pos, len, copied) ); -TRACE_EVENT(ext4_writepage, - TP_PROTO(struct inode *inode, struct page *page), - - TP_ARGS(inode, page), - - TP_STRUCT__entry( - __field( dev_t, dev ) - __field( ino_t, ino ) - __field( pgoff_t, index ) - - ), - - TP_fast_assign( - __entry->dev = inode->i_sb->s_dev; - __entry->ino = inode->i_ino; - __entry->index = page->index; - ), - - TP_printk("dev %d,%d ino %lu page_index %lu", - MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long) __entry->ino, __entry->index) -); - TRACE_EVENT(ext4_da_writepages, TP_PROTO(struct inode *inode, struct writeback_control *wbc), @@ -341,7 +317,7 @@ TRACE_EVENT(ext4_da_writepages, ), TP_printk("dev %d,%d ino %lu nr_to_write %ld pages_skipped %ld " - "range_start %llu range_end %llu sync_mode %d" + "range_start %lld range_end %lld sync_mode %d" "for_kupdate %d range_cyclic %d writeback_index %lu", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, __entry->nr_to_write, @@ -449,7 +425,14 @@ DECLARE_EVENT_CLASS(ext4__page_op, TP_printk("dev %d,%d ino %lu page_index %lu", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - __entry->index) + (unsigned long) __entry->index) +); + +DEFINE_EVENT(ext4__page_op, ext4_writepage, + + TP_PROTO(struct page *page), + + TP_ARGS(page) ); DEFINE_EVENT(ext4__page_op, ext4_readpage, @@ -489,7 +472,7 @@ TRACE_EVENT(ext4_invalidatepage, TP_printk("dev %d,%d ino %lu page_index %lu offset %lu", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - __entry->index, __entry->offset) + (unsigned long) __entry->index, __entry->offset) ); TRACE_EVENT(ext4_discard_blocks, @@ -562,12 +545,10 @@ DEFINE_EVENT(ext4__mb_new_pa, ext4_mb_new_group_pa, ); TRACE_EVENT(ext4_mb_release_inode_pa, - TP_PROTO(struct super_block *sb, - struct inode *inode, - struct ext4_prealloc_space *pa, + TP_PROTO(struct ext4_prealloc_space *pa, unsigned long long block, unsigned int count), - TP_ARGS(sb, inode, pa, block, count), + TP_ARGS(pa, block, count), TP_STRUCT__entry( __field( dev_t, dev ) @@ -578,8 +559,8 @@ TRACE_EVENT(ext4_mb_release_inode_pa, ), TP_fast_assign( - __entry->dev = sb->s_dev; - __entry->ino = inode->i_ino; + __entry->dev = pa->pa_inode->i_sb->s_dev; + __entry->ino = pa->pa_inode->i_ino; __entry->block = block; __entry->count = count; ), @@ -591,10 +572,9 @@ TRACE_EVENT(ext4_mb_release_inode_pa, ); TRACE_EVENT(ext4_mb_release_group_pa, - TP_PROTO(struct super_block *sb, - struct ext4_prealloc_space *pa), + TP_PROTO(struct ext4_prealloc_space *pa), - TP_ARGS(sb, pa), + TP_ARGS(pa), TP_STRUCT__entry( __field( dev_t, dev ) @@ -604,7 +584,7 @@ TRACE_EVENT(ext4_mb_release_group_pa, ), TP_fast_assign( - __entry->dev = sb->s_dev; + __entry->dev = pa->pa_inode->i_sb->s_dev; __entry->pa_pstart = pa->pa_pstart; __entry->pa_len = pa->pa_len; ), @@ -666,10 +646,10 @@ TRACE_EVENT(ext4_request_blocks, __field( ino_t, ino ) __field( unsigned int, flags ) __field( unsigned int, len ) - __field( __u64, logical ) + __field( __u32, logical ) + __field( __u32, lleft ) + __field( __u32, lright ) __field( __u64, goal ) - __field( __u64, lleft ) - __field( __u64, lright ) __field( __u64, pleft ) __field( __u64, pright ) ), @@ -687,17 +667,13 @@ TRACE_EVENT(ext4_request_blocks, __entry->pright = ar->pright; ), - TP_printk("dev %d,%d ino %lu flags %u len %u lblk %llu goal %llu " - "lleft %llu lright %llu pleft %llu pright %llu ", + TP_printk("dev %d,%d ino %lu flags %u len %u lblk %u goal %llu " + "lleft %u lright %u pleft %llu pright %llu ", MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long) __entry->ino, - __entry->flags, __entry->len, - (unsigned long long) __entry->logical, - (unsigned long long) __entry->goal, - (unsigned long long) __entry->lleft, - (unsigned long long) __entry->lright, - (unsigned long long) __entry->pleft, - (unsigned long long) __entry->pright) + (unsigned long) __entry->ino, __entry->flags, + __entry->len, __entry->logical, __entry->goal, + __entry->lleft, __entry->lright, __entry->pleft, + __entry->pright) ); TRACE_EVENT(ext4_allocate_blocks, @@ -711,10 +687,10 @@ TRACE_EVENT(ext4_allocate_blocks, __field( __u64, block ) __field( unsigned int, flags ) __field( unsigned int, len ) - __field( __u64, logical ) + __field( __u32, logical ) + __field( __u32, lleft ) + __field( __u32, lright ) __field( __u64, goal ) - __field( __u64, lleft ) - __field( __u64, lright ) __field( __u64, pleft ) __field( __u64, pright ) ), @@ -733,17 +709,13 @@ TRACE_EVENT(ext4_allocate_blocks, __entry->pright = ar->pright; ), - TP_printk("dev %d,%d ino %lu flags %u len %u block %llu lblk %llu " - "goal %llu lleft %llu lright %llu pleft %llu pright %llu", + TP_printk("dev %d,%d ino %lu flags %u len %u block %llu lblk %u " + "goal %llu lleft %u lright %u pleft %llu pright %llu", MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long) __entry->ino, - __entry->flags, __entry->len, __entry->block, - (unsigned long long) __entry->logical, - (unsigned long long) __entry->goal, - (unsigned long long) __entry->lleft, - (unsigned long long) __entry->lright, - (unsigned long long) __entry->pleft, - (unsigned long long) __entry->pright) + (unsigned long) __entry->ino, __entry->flags, + __entry->len, __entry->block, __entry->logical, + __entry->goal, __entry->lleft, __entry->lright, + __entry->pleft, __entry->pright) ); TRACE_EVENT(ext4_free_blocks, @@ -755,10 +727,10 @@ TRACE_EVENT(ext4_free_blocks, TP_STRUCT__entry( __field( dev_t, dev ) __field( ino_t, ino ) - __field( umode_t, mode ) + __field( umode_t, mode ) __field( __u64, block ) __field( unsigned long, count ) - __field( int, flags ) + __field( int, flags ) ), TP_fast_assign( @@ -798,7 +770,7 @@ TRACE_EVENT(ext4_sync_file_enter, __entry->parent = dentry->d_parent->d_inode->i_ino; ), - TP_printk("dev %d,%d ino %ld parent %ld datasync %d ", + TP_printk("dev %d,%d ino %lu parent %lu datasync %d ", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, (unsigned long) __entry->parent, __entry->datasync) @@ -821,7 +793,7 @@ TRACE_EVENT(ext4_sync_file_exit, __entry->dev = inode->i_sb->s_dev; ), - TP_printk("dev %d,%d ino %ld ret %d", + TP_printk("dev %d,%d ino %lu ret %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, __entry->ret) @@ -1005,7 +977,7 @@ DECLARE_EVENT_CLASS(ext4__mballoc, __entry->result_len = len; ), - TP_printk("dev %d,%d inode %lu extent %u/%d/%u ", + TP_printk("dev %d,%d inode %lu extent %u/%d/%d ", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, __entry->result_group, __entry->result_start, @@ -1093,7 +1065,7 @@ TRACE_EVENT(ext4_da_update_reserve_space, "allocated_meta_blocks %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - __entry->mode, (unsigned long long) __entry->i_blocks, + __entry->mode, __entry->i_blocks, __entry->used_blocks, __entry->reserved_data_blocks, __entry->reserved_meta_blocks, __entry->allocated_meta_blocks) ); @@ -1127,7 +1099,7 @@ TRACE_EVENT(ext4_da_reserve_space, "reserved_data_blocks %d reserved_meta_blocks %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - __entry->mode, (unsigned long long) __entry->i_blocks, + __entry->mode, __entry->i_blocks, __entry->md_needed, __entry->reserved_data_blocks, __entry->reserved_meta_blocks) ); @@ -1164,7 +1136,7 @@ TRACE_EVENT(ext4_da_release_space, "allocated_meta_blocks %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - __entry->mode, (unsigned long long) __entry->i_blocks, + __entry->mode, __entry->i_blocks, __entry->freed_blocks, __entry->reserved_data_blocks, __entry->reserved_meta_blocks, __entry->allocated_meta_blocks) ); @@ -1239,14 +1211,15 @@ TRACE_EVENT(ext4_direct_IO_enter, __entry->rw = rw; ), - TP_printk("dev %d,%d ino %lu pos %llu len %lu rw %d", + TP_printk("dev %d,%d ino %lu pos %lld len %lu rw %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - (unsigned long long) __entry->pos, __entry->len, __entry->rw) + __entry->pos, __entry->len, __entry->rw) ); TRACE_EVENT(ext4_direct_IO_exit, - TP_PROTO(struct inode *inode, loff_t offset, unsigned long len, int rw, int ret), + TP_PROTO(struct inode *inode, loff_t offset, unsigned long len, + int rw, int ret), TP_ARGS(inode, offset, len, rw, ret), @@ -1268,10 +1241,10 @@ TRACE_EVENT(ext4_direct_IO_exit, __entry->ret = ret; ), - TP_printk("dev %d,%d ino %lu pos %llu len %lu rw %d ret %d", + TP_printk("dev %d,%d ino %lu pos %lld len %lu rw %d ret %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - (unsigned long long) __entry->pos, __entry->len, + __entry->pos, __entry->len, __entry->rw, __entry->ret) ); @@ -1296,15 +1269,15 @@ TRACE_EVENT(ext4_fallocate_enter, __entry->mode = mode; ), - TP_printk("dev %d,%d ino %ld pos %llu len %llu mode %d", + TP_printk("dev %d,%d ino %lu pos %lld len %lld mode %d", MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long) __entry->ino, - (unsigned long long) __entry->pos, - (unsigned long long) __entry->len, __entry->mode) + (unsigned long) __entry->ino, __entry->pos, + __entry->len, __entry->mode) ); TRACE_EVENT(ext4_fallocate_exit, - TP_PROTO(struct inode *inode, loff_t offset, unsigned int max_blocks, int ret), + TP_PROTO(struct inode *inode, loff_t offset, + unsigned int max_blocks, int ret), TP_ARGS(inode, offset, max_blocks, ret), @@ -1312,7 +1285,7 @@ TRACE_EVENT(ext4_fallocate_exit, __field( ino_t, ino ) __field( dev_t, dev ) __field( loff_t, pos ) - __field( unsigned, blocks ) + __field( unsigned int, blocks ) __field( int, ret ) ), @@ -1324,10 +1297,10 @@ TRACE_EVENT(ext4_fallocate_exit, __entry->ret = ret; ), - TP_printk("dev %d,%d ino %ld pos %llu blocks %d ret %d", + TP_printk("dev %d,%d ino %lu pos %lld blocks %u ret %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - (unsigned long long) __entry->pos, __entry->blocks, + __entry->pos, __entry->blocks, __entry->ret) ); @@ -1350,7 +1323,7 @@ TRACE_EVENT(ext4_unlink_enter, __entry->dev = dentry->d_inode->i_sb->s_dev; ), - TP_printk("dev %d,%d ino %ld size %lld parent %ld", + TP_printk("dev %d,%d ino %lu size %lld parent %lu", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, __entry->size, (unsigned long) __entry->parent) @@ -1373,7 +1346,7 @@ TRACE_EVENT(ext4_unlink_exit, __entry->ret = ret; ), - TP_printk("dev %d,%d ino %ld ret %d", + TP_printk("dev %d,%d ino %lu ret %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, __entry->ret) @@ -1387,7 +1360,7 @@ DECLARE_EVENT_CLASS(ext4__truncate, TP_STRUCT__entry( __field( ino_t, ino ) __field( dev_t, dev ) - __field( blkcnt_t, blocks ) + __field( __u64, blocks ) ), TP_fast_assign( @@ -1396,9 +1369,9 @@ DECLARE_EVENT_CLASS(ext4__truncate, __entry->blocks = inode->i_blocks; ), - TP_printk("dev %d,%d ino %lu blocks %lu", + TP_printk("dev %d,%d ino %lu blocks %llu", MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long) __entry->ino, (unsigned long) __entry->blocks) + (unsigned long) __entry->ino, __entry->blocks) ); DEFINE_EVENT(ext4__truncate, ext4_truncate_enter, @@ -1417,7 +1390,7 @@ DEFINE_EVENT(ext4__truncate, ext4_truncate_exit, DECLARE_EVENT_CLASS(ext4__map_blocks_enter, TP_PROTO(struct inode *inode, ext4_lblk_t lblk, - unsigned len, unsigned flags), + unsigned int len, unsigned int flags), TP_ARGS(inode, lblk, len, flags), @@ -1425,8 +1398,8 @@ DECLARE_EVENT_CLASS(ext4__map_blocks_enter, __field( ino_t, ino ) __field( dev_t, dev ) __field( ext4_lblk_t, lblk ) - __field( unsigned, len ) - __field( unsigned, flags ) + __field( unsigned int, len ) + __field( unsigned int, flags ) ), TP_fast_assign( @@ -1440,7 +1413,7 @@ DECLARE_EVENT_CLASS(ext4__map_blocks_enter, TP_printk("dev %d,%d ino %lu lblk %u len %u flags %u", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - (unsigned) __entry->lblk, __entry->len, __entry->flags) + __entry->lblk, __entry->len, __entry->flags) ); DEFINE_EVENT(ext4__map_blocks_enter, ext4_ext_map_blocks_enter, @@ -1459,7 +1432,7 @@ DEFINE_EVENT(ext4__map_blocks_enter, ext4_ind_map_blocks_enter, DECLARE_EVENT_CLASS(ext4__map_blocks_exit, TP_PROTO(struct inode *inode, ext4_lblk_t lblk, - ext4_fsblk_t pblk, unsigned len, int ret), + ext4_fsblk_t pblk, unsigned int len, int ret), TP_ARGS(inode, lblk, pblk, len, ret), @@ -1468,7 +1441,7 @@ DECLARE_EVENT_CLASS(ext4__map_blocks_exit, __field( dev_t, dev ) __field( ext4_lblk_t, lblk ) __field( ext4_fsblk_t, pblk ) - __field( unsigned, len ) + __field( unsigned int, len ) __field( int, ret ) ), @@ -1484,7 +1457,7 @@ DECLARE_EVENT_CLASS(ext4__map_blocks_exit, TP_printk("dev %d,%d ino %lu lblk %u pblk %llu len %u ret %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - (unsigned) __entry->lblk, (unsigned long long) __entry->pblk, + __entry->lblk, __entry->pblk, __entry->len, __entry->ret) ); @@ -1524,7 +1497,7 @@ TRACE_EVENT(ext4_ext_load_extent, TP_printk("dev %d,%d ino %lu lblk %u pblk %llu", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - (unsigned) __entry->lblk, (unsigned long long) __entry->pblk) + __entry->lblk, __entry->pblk) ); TRACE_EVENT(ext4_load_inode, -- cgit v1.2.3 From b084f598df36b62dfae83c10ed17f0b66b50f442 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Tue, 31 May 2011 12:24:58 -0400 Subject: nfsd: fix dependency of nfsd on auth_rpcgss Commit b0b0c0a26e84 "nfsd: add proc file listing kernel's gss_krb5 enctypes" added an nunnecessary dependency of nfsd on the auth_rpcgss module. It's a little ad hoc, but since the only piece of information nfsd needs from rpcsec_gss_krb5 is a single static string, one solution is just to share it with an include file. Cc: stable@kernel.org Reported-by: Michael Guntsche Cc: Kevin Coffman Signed-off-by: J. Bruce Fields --- fs/nfsd/nfsctl.c | 19 ++++++------------- include/linux/sunrpc/gss_krb5_enctypes.h | 4 ++++ net/sunrpc/auth_gss/gss_krb5_mech.c | 3 ++- 3 files changed, 12 insertions(+), 14 deletions(-) create mode 100644 include/linux/sunrpc/gss_krb5_enctypes.h (limited to 'include') diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 1f5eae40f34e..2b1449dd2f49 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -13,6 +13,7 @@ #include #include #include +#include #include "idmap.h" #include "nfsd.h" @@ -189,18 +190,10 @@ static struct file_operations export_features_operations = { .release = single_release, }; -#ifdef CONFIG_SUNRPC_GSS +#if defined(CONFIG_SUNRPC_GSS) || defined(CONFIG_SUNRPC_GSS_MODULE) static int supported_enctypes_show(struct seq_file *m, void *v) { - struct gss_api_mech *k5mech; - - k5mech = gss_mech_get_by_name("krb5"); - if (k5mech == NULL) - goto out; - if (k5mech->gm_upcall_enctypes != NULL) - seq_printf(m, k5mech->gm_upcall_enctypes); - gss_mech_put(k5mech); -out: + seq_printf(m, KRB5_SUPPORTED_ENCTYPES); return 0; } @@ -215,7 +208,7 @@ static struct file_operations supported_enctypes_ops = { .llseek = seq_lseek, .release = single_release, }; -#endif /* CONFIG_SUNRPC_GSS */ +#endif /* CONFIG_SUNRPC_GSS or CONFIG_SUNRPC_GSS_MODULE */ extern int nfsd_pool_stats_open(struct inode *inode, struct file *file); extern int nfsd_pool_stats_release(struct inode *inode, struct file *file); @@ -1427,9 +1420,9 @@ static int nfsd_fill_super(struct super_block * sb, void * data, int silent) [NFSD_Versions] = {"versions", &transaction_ops, S_IWUSR|S_IRUSR}, [NFSD_Ports] = {"portlist", &transaction_ops, S_IWUSR|S_IRUGO}, [NFSD_MaxBlkSize] = {"max_block_size", &transaction_ops, S_IWUSR|S_IRUGO}, -#ifdef CONFIG_SUNRPC_GSS +#if defined(CONFIG_SUNRPC_GSS) || defined(CONFIG_SUNRPC_GSS_MODULE) [NFSD_SupportedEnctypes] = {"supported_krb5_enctypes", &supported_enctypes_ops, S_IRUGO}, -#endif /* CONFIG_SUNRPC_GSS */ +#endif /* CONFIG_SUNRPC_GSS or CONFIG_SUNRPC_GSS_MODULE */ #ifdef CONFIG_NFSD_V4 [NFSD_Leasetime] = {"nfsv4leasetime", &transaction_ops, S_IWUSR|S_IRUSR}, [NFSD_Gracetime] = {"nfsv4gracetime", &transaction_ops, S_IWUSR|S_IRUSR}, diff --git a/include/linux/sunrpc/gss_krb5_enctypes.h b/include/linux/sunrpc/gss_krb5_enctypes.h new file mode 100644 index 000000000000..ec6234eee89c --- /dev/null +++ b/include/linux/sunrpc/gss_krb5_enctypes.h @@ -0,0 +1,4 @@ +/* + * Dumb way to share this static piece of information with nfsd + */ +#define KRB5_SUPPORTED_ENCTYPES "18,17,16,23,3,1,2" diff --git a/net/sunrpc/auth_gss/gss_krb5_mech.c b/net/sunrpc/auth_gss/gss_krb5_mech.c index 0a9a2ec2e469..c3b75333b821 100644 --- a/net/sunrpc/auth_gss/gss_krb5_mech.c +++ b/net/sunrpc/auth_gss/gss_krb5_mech.c @@ -43,6 +43,7 @@ #include #include #include +#include #ifdef RPC_DEBUG # define RPCDBG_FACILITY RPCDBG_AUTH @@ -750,7 +751,7 @@ static struct gss_api_mech gss_kerberos_mech = { .gm_ops = &gss_kerberos_ops, .gm_pf_num = ARRAY_SIZE(gss_kerberos_pfs), .gm_pfs = gss_kerberos_pfs, - .gm_upcall_enctypes = "18,17,16,23,3,1,2", + .gm_upcall_enctypes = KRB5_SUPPORTED_ENCTYPES, }; static int __init init_kerberos_module(void) -- cgit v1.2.3 From a685e08987d1edf1995b76511d4c98ea0e905377 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 8 Jun 2011 21:13:01 -0400 Subject: Delay struct net freeing while there's a sysfs instance refering to it * new refcount in struct net, controlling actual freeing of the memory * new method in kobj_ns_type_operations (->drop_ns()) * ->current_ns() semantics change - it's supposed to be followed by corresponding ->drop_ns(). For struct net in case of CONFIG_NET_NS it bumps the new refcount; net_drop_ns() decrements it and calls net_free() if the last reference has been dropped. Method renamed to ->grab_current_ns(). * old net_free() callers call net_drop_ns() instead. * sysfs_exit_ns() is gone, along with a large part of callchain leading to it; now that the references stored in ->ns[...] stay valid we do not need to hunt them down and replace them with NULL. That fixes problems in sysfs_lookup() and sysfs_readdir(), along with getting rid of sb->s_instances abuse. Note that struct net *shutdown* logics has not changed - net_cleanup() is called exactly when it used to be called. The only thing postponed by having a sysfs instance refering to that struct net is actual freeing of memory occupied by struct net. Signed-off-by: Al Viro --- fs/sysfs/mount.c | 37 +++++++++++-------------------------- fs/sysfs/sysfs.h | 2 +- include/linux/kobject_ns.h | 10 ++++++---- include/linux/sysfs.h | 7 ------- include/net/net_namespace.h | 10 +++++++++- lib/kobject.c | 26 +++++++++----------------- net/core/net-sysfs.c | 23 +++++++++-------------- net/core/net_namespace.c | 12 ++++++++++-- 8 files changed, 55 insertions(+), 72 deletions(-) (limited to 'include') diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index 266895783b47..e34f0d99ea4e 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -95,6 +95,14 @@ static int sysfs_set_super(struct super_block *sb, void *data) return error; } +static void free_sysfs_super_info(struct sysfs_super_info *info) +{ + int type; + for (type = KOBJ_NS_TYPE_NONE; type < KOBJ_NS_TYPES; type++) + kobj_ns_drop(type, info->ns[type]); + kfree(info); +} + static struct dentry *sysfs_mount(struct file_system_type *fs_type, int flags, const char *dev_name, void *data) { @@ -108,11 +116,11 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type, return ERR_PTR(-ENOMEM); for (type = KOBJ_NS_TYPE_NONE; type < KOBJ_NS_TYPES; type++) - info->ns[type] = kobj_ns_current(type); + info->ns[type] = kobj_ns_grab_current(type); sb = sget(fs_type, sysfs_test_super, sysfs_set_super, info); if (IS_ERR(sb) || sb->s_fs_info != info) - kfree(info); + free_sysfs_super_info(info); if (IS_ERR(sb)) return ERR_CAST(sb); if (!sb->s_root) { @@ -131,12 +139,11 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type, static void sysfs_kill_sb(struct super_block *sb) { struct sysfs_super_info *info = sysfs_info(sb); - /* Remove the superblock from fs_supers/s_instances * so we can't find it, before freeing sysfs_super_info. */ kill_anon_super(sb); - kfree(info); + free_sysfs_super_info(info); } static struct file_system_type sysfs_fs_type = { @@ -145,28 +152,6 @@ static struct file_system_type sysfs_fs_type = { .kill_sb = sysfs_kill_sb, }; -void sysfs_exit_ns(enum kobj_ns_type type, const void *ns) -{ - struct super_block *sb; - - mutex_lock(&sysfs_mutex); - spin_lock(&sb_lock); - list_for_each_entry(sb, &sysfs_fs_type.fs_supers, s_instances) { - struct sysfs_super_info *info = sysfs_info(sb); - /* - * If we see a superblock on the fs_supers/s_instances - * list the unmount has not completed and sb->s_fs_info - * points to a valid struct sysfs_super_info. - */ - /* Ignore superblocks with the wrong ns */ - if (info->ns[type] != ns) - continue; - info->ns[type] = NULL; - } - spin_unlock(&sb_lock); - mutex_unlock(&sysfs_mutex); -} - int __init sysfs_init(void) { int err = -ENOMEM; diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h index 3d28af31d863..2ed2404f3113 100644 --- a/fs/sysfs/sysfs.h +++ b/fs/sysfs/sysfs.h @@ -136,7 +136,7 @@ struct sysfs_addrm_cxt { * instance). */ struct sysfs_super_info { - const void *ns[KOBJ_NS_TYPES]; + void *ns[KOBJ_NS_TYPES]; }; #define sysfs_info(SB) ((struct sysfs_super_info *)(SB->s_fs_info)) extern struct sysfs_dirent sysfs_root; diff --git a/include/linux/kobject_ns.h b/include/linux/kobject_ns.h index 82cb5bf461fb..f66b065a8b5f 100644 --- a/include/linux/kobject_ns.h +++ b/include/linux/kobject_ns.h @@ -32,15 +32,17 @@ enum kobj_ns_type { /* * Callbacks so sysfs can determine namespaces - * @current_ns: return calling task's namespace + * @grab_current_ns: return a new reference to calling task's namespace * @netlink_ns: return namespace to which a sock belongs (right?) * @initial_ns: return the initial namespace (i.e. init_net_ns) + * @drop_ns: drops a reference to namespace */ struct kobj_ns_type_operations { enum kobj_ns_type type; - const void *(*current_ns)(void); + void *(*grab_current_ns)(void); const void *(*netlink_ns)(struct sock *sk); const void *(*initial_ns)(void); + void (*drop_ns)(void *); }; int kobj_ns_type_register(const struct kobj_ns_type_operations *ops); @@ -48,9 +50,9 @@ int kobj_ns_type_registered(enum kobj_ns_type type); const struct kobj_ns_type_operations *kobj_child_ns_ops(struct kobject *parent); const struct kobj_ns_type_operations *kobj_ns_ops(struct kobject *kobj); -const void *kobj_ns_current(enum kobj_ns_type type); +void *kobj_ns_grab_current(enum kobj_ns_type type); const void *kobj_ns_netlink(enum kobj_ns_type type, struct sock *sk); const void *kobj_ns_initial(enum kobj_ns_type type); -void kobj_ns_exit(enum kobj_ns_type type, const void *ns); +void kobj_ns_drop(enum kobj_ns_type type, void *ns); #endif /* _LINUX_KOBJECT_NS_H */ diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h index c3acda60eee0..e2696d76a599 100644 --- a/include/linux/sysfs.h +++ b/include/linux/sysfs.h @@ -177,9 +177,6 @@ struct sysfs_dirent *sysfs_get_dirent(struct sysfs_dirent *parent_sd, struct sysfs_dirent *sysfs_get(struct sysfs_dirent *sd); void sysfs_put(struct sysfs_dirent *sd); -/* Called to clear a ns tag when it is no longer valid */ -void sysfs_exit_ns(enum kobj_ns_type type, const void *tag); - int __must_check sysfs_init(void); #else /* CONFIG_SYSFS */ @@ -338,10 +335,6 @@ static inline void sysfs_put(struct sysfs_dirent *sd) { } -static inline void sysfs_exit_ns(int type, const void *tag) -{ -} - static inline int __must_check sysfs_init(void) { return 0; diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index 2bf9ed9ef26b..aef430d779bd 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -35,8 +35,11 @@ struct netns_ipvs; #define NETDEV_HASHENTRIES (1 << NETDEV_HASHBITS) struct net { + atomic_t passive; /* To decided when the network + * namespace should be freed. + */ atomic_t count; /* To decided when the network - * namespace should be freed. + * namespace should be shut down. */ #ifdef NETNS_REFCNT_DEBUG atomic_t use_count; /* To track references we @@ -154,6 +157,9 @@ int net_eq(const struct net *net1, const struct net *net2) { return net1 == net2; } + +extern void net_drop_ns(void *); + #else static inline struct net *get_net(struct net *net) @@ -175,6 +181,8 @@ int net_eq(const struct net *net1, const struct net *net2) { return 1; } + +#define net_drop_ns NULL #endif diff --git a/lib/kobject.c b/lib/kobject.c index 82dc34c095c2..640bd98a4c8a 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -948,14 +948,14 @@ const struct kobj_ns_type_operations *kobj_ns_ops(struct kobject *kobj) } -const void *kobj_ns_current(enum kobj_ns_type type) +void *kobj_ns_grab_current(enum kobj_ns_type type) { - const void *ns = NULL; + void *ns = NULL; spin_lock(&kobj_ns_type_lock); if ((type > KOBJ_NS_TYPE_NONE) && (type < KOBJ_NS_TYPES) && kobj_ns_ops_tbl[type]) - ns = kobj_ns_ops_tbl[type]->current_ns(); + ns = kobj_ns_ops_tbl[type]->grab_current_ns(); spin_unlock(&kobj_ns_type_lock); return ns; @@ -987,23 +987,15 @@ const void *kobj_ns_initial(enum kobj_ns_type type) return ns; } -/* - * kobj_ns_exit - invalidate a namespace tag - * - * @type: the namespace type (i.e. KOBJ_NS_TYPE_NET) - * @ns: the actual namespace being invalidated - * - * This is called when a tag is no longer valid. For instance, - * when a network namespace exits, it uses this helper to - * make sure no sb's sysfs_info points to the now-invalidated - * netns. - */ -void kobj_ns_exit(enum kobj_ns_type type, const void *ns) +void kobj_ns_drop(enum kobj_ns_type type, void *ns) { - sysfs_exit_ns(type, ns); + spin_lock(&kobj_ns_type_lock); + if ((type > KOBJ_NS_TYPE_NONE) && (type < KOBJ_NS_TYPES) && + kobj_ns_ops_tbl[type] && kobj_ns_ops_tbl[type]->drop_ns) + kobj_ns_ops_tbl[type]->drop_ns(ns); + spin_unlock(&kobj_ns_type_lock); } - EXPORT_SYMBOL(kobject_get); EXPORT_SYMBOL(kobject_put); EXPORT_SYMBOL(kobject_del); diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 11b98bc2aa8f..33d2a1fba131 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -1179,9 +1179,14 @@ static void remove_queue_kobjects(struct net_device *net) #endif } -static const void *net_current_ns(void) +static void *net_grab_current_ns(void) { - return current->nsproxy->net_ns; + struct net *ns = current->nsproxy->net_ns; +#ifdef CONFIG_NET_NS + if (ns) + atomic_inc(&ns->passive); +#endif + return ns; } static const void *net_initial_ns(void) @@ -1196,22 +1201,13 @@ static const void *net_netlink_ns(struct sock *sk) struct kobj_ns_type_operations net_ns_type_operations = { .type = KOBJ_NS_TYPE_NET, - .current_ns = net_current_ns, + .grab_current_ns = net_grab_current_ns, .netlink_ns = net_netlink_ns, .initial_ns = net_initial_ns, + .drop_ns = net_drop_ns, }; EXPORT_SYMBOL_GPL(net_ns_type_operations); -static void net_kobj_ns_exit(struct net *net) -{ - kobj_ns_exit(KOBJ_NS_TYPE_NET, net); -} - -static struct pernet_operations kobj_net_ops = { - .exit = net_kobj_ns_exit, -}; - - #ifdef CONFIG_HOTPLUG static int netdev_uevent(struct device *d, struct kobj_uevent_env *env) { @@ -1339,6 +1335,5 @@ EXPORT_SYMBOL(netdev_class_remove_file); int netdev_kobject_init(void) { kobj_ns_type_register(&net_ns_type_operations); - register_pernet_subsys(&kobj_net_ops); return class_register(&net_class); } diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index 6c6b86d0da15..cdcbc3cb00a9 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -128,6 +128,7 @@ static __net_init int setup_net(struct net *net) LIST_HEAD(net_exit_list); atomic_set(&net->count, 1); + atomic_set(&net->passive, 1); #ifdef NETNS_REFCNT_DEBUG atomic_set(&net->use_count, 0); @@ -210,6 +211,13 @@ static void net_free(struct net *net) kmem_cache_free(net_cachep, net); } +void net_drop_ns(void *p) +{ + struct net *ns = p; + if (ns && atomic_dec_and_test(&ns->passive)) + net_free(ns); +} + struct net *copy_net_ns(unsigned long flags, struct net *old_net) { struct net *net; @@ -230,7 +238,7 @@ struct net *copy_net_ns(unsigned long flags, struct net *old_net) } mutex_unlock(&net_mutex); if (rv < 0) { - net_free(net); + net_drop_ns(net); return ERR_PTR(rv); } return net; @@ -286,7 +294,7 @@ static void cleanup_net(struct work_struct *work) /* Finally it is safe to free my network namespace structure */ list_for_each_entry_safe(net, tmp, &net_exit_list, exit_list) { list_del_init(&net->exit_list); - net_free(net); + net_drop_ns(net); } } static DECLARE_WORK(net_cleanup_work, cleanup_net); -- cgit v1.2.3 From 08e8138adebdd511e0955e8d6c051904bb4082af Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 13 Jun 2011 10:42:49 +0200 Subject: block: Add __attribute__((format(printf...) and fix fallout Use the compiler to verify format strings and arguments. Fix fallout. Signed-off-by: Joe Perches Signed-off-by: Jens Axboe --- block/blk-throttle.c | 4 ++-- block/cfq-iosched.c | 11 ++++++----- include/linux/blktrace_api.h | 3 ++- 3 files changed, 10 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/block/blk-throttle.c b/block/blk-throttle.c index a62be8d0dc1b..3689f833afdc 100644 --- a/block/blk-throttle.c +++ b/block/blk-throttle.c @@ -927,7 +927,7 @@ static int throtl_dispatch(struct request_queue *q) bio_list_init(&bio_list_on_stack); - throtl_log(td, "dispatch nr_queued=%lu read=%u write=%u", + throtl_log(td, "dispatch nr_queued=%d read=%u write=%u", total_nr_queued(td), td->nr_queued[READ], td->nr_queued[WRITE]); @@ -1204,7 +1204,7 @@ int blk_throtl_bio(struct request_queue *q, struct bio **biop) } queue_bio: - throtl_log_tg(td, tg, "[%c] bio. bdisp=%u sz=%u bps=%llu" + throtl_log_tg(td, tg, "[%c] bio. bdisp=%llu sz=%u bps=%llu" " iodisp=%u iops=%u queued=%d/%d", rw == READ ? 'R' : 'W', tg->bytes_disp[rw], bio->bi_size, tg->bps[rw], diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 545b8d4c829d..f3799432676d 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -988,9 +988,10 @@ static void cfq_group_served(struct cfq_data *cfqd, struct cfq_group *cfqg, cfq_log_cfqg(cfqd, cfqg, "served: vt=%llu min_vt=%llu", cfqg->vdisktime, st->min_vdisktime); - cfq_log_cfqq(cfqq->cfqd, cfqq, "sl_used=%u disp=%u charge=%u iops=%u" - " sect=%u", used_sl, cfqq->slice_dispatch, charge, - iops_mode(cfqd), cfqq->nr_sectors); + cfq_log_cfqq(cfqq->cfqd, cfqq, + "sl_used=%u disp=%u charge=%u iops=%u sect=%lu", + used_sl, cfqq->slice_dispatch, charge, + iops_mode(cfqd), cfqq->nr_sectors); cfq_blkiocg_update_timeslice_used(&cfqg->blkg, used_sl, unaccounted_sl); cfq_blkiocg_set_start_empty_time(&cfqg->blkg); @@ -2023,8 +2024,8 @@ static void cfq_arm_slice_timer(struct cfq_data *cfqd) */ if (sample_valid(cic->ttime_samples) && (cfqq->slice_end - jiffies < cic->ttime_mean)) { - cfq_log_cfqq(cfqd, cfqq, "Not idling. think_time:%d", - cic->ttime_mean); + cfq_log_cfqq(cfqd, cfqq, "Not idling. think_time:%lu", + cic->ttime_mean); return; } diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h index b22fb0d3db0f..8c7c2de7631a 100644 --- a/include/linux/blktrace_api.h +++ b/include/linux/blktrace_api.h @@ -169,7 +169,8 @@ extern void blk_trace_shutdown(struct request_queue *); extern int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, struct block_device *bdev, struct blk_user_trace_setup *buts); -extern void __trace_note_message(struct blk_trace *, const char *fmt, ...); +extern __attribute__((format(printf, 2, 3))) +void __trace_note_message(struct blk_trace *, const char *fmt, ...); /** * blk_add_trace_msg - Add a (simple) message to the blktrace stream -- cgit v1.2.3 From e9c039052be59753e6bcc7c8b59763899dc1161c Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Mon, 13 Jun 2011 19:05:58 +0100 Subject: ASoC: Remove unused and about to be broken SND_SOC_CUSTOM I/O bus This will be removed in -next so let's drop it from mainline as soon as we can in order to minimise surprises. Signed-off-by: Mark Brown --- include/sound/soc.h | 3 +-- sound/soc/soc-cache.c | 3 --- 2 files changed, 1 insertion(+), 5 deletions(-) (limited to 'include') diff --git a/include/sound/soc.h b/include/sound/soc.h index f1de3e0c75bc..3a4bd3a3c68d 100644 --- a/include/sound/soc.h +++ b/include/sound/soc.h @@ -248,8 +248,7 @@ typedef int (*hw_write_t)(void *,const char* ,int); extern struct snd_ac97_bus_ops soc_ac97_ops; enum snd_soc_control_type { - SND_SOC_CUSTOM = 1, - SND_SOC_I2C, + SND_SOC_I2C = 1, SND_SOC_SPI, }; diff --git a/sound/soc/soc-cache.c b/sound/soc/soc-cache.c index c005ceb70c9d..039b9532b270 100644 --- a/sound/soc/soc-cache.c +++ b/sound/soc/soc-cache.c @@ -409,9 +409,6 @@ int snd_soc_codec_set_cache_io(struct snd_soc_codec *codec, codec->bulk_write_raw = snd_soc_hw_bulk_write_raw; switch (control) { - case SND_SOC_CUSTOM: - break; - case SND_SOC_I2C: #if defined(CONFIG_I2C) || (defined(CONFIG_I2C_MODULE) && defined(MODULE)) codec->hw_write = (hw_write_t)i2c_master_send; -- cgit v1.2.3 From de1b794130b130e77ffa975bb58cb843744f9ae5 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 13 Jun 2011 15:38:22 -0400 Subject: jbd2: Fix oops in jbd2_journal_remove_journal_head() jbd2_journal_remove_journal_head() can oops when trying to access journal_head returned by bh2jh(). This is caused for example by the following race: TASK1 TASK2 jbd2_journal_commit_transaction() ... processing t_forget list __jbd2_journal_refile_buffer(jh); if (!jh->b_transaction) { jbd_unlock_bh_state(bh); jbd2_journal_try_to_free_buffers() jbd2_journal_grab_journal_head(bh) jbd_lock_bh_state(bh) __journal_try_to_free_buffer() jbd2_journal_put_journal_head(jh) jbd2_journal_remove_journal_head(bh); jbd2_journal_put_journal_head() in TASK2 sees that b_jcount == 0 and buffer is not part of any transaction and thus frees journal_head before TASK1 gets to doing so. Note that even buffer_head can be released by try_to_free_buffers() after jbd2_journal_put_journal_head() which adds even larger opportunity for oops (but I didn't see this happen in reality). Fix the problem by making transactions hold their own journal_head reference (in b_jcount). That way we don't have to remove journal_head explicitely via jbd2_journal_remove_journal_head() and instead just remove journal_head when b_jcount drops to zero. The result of this is that [__]jbd2_journal_refile_buffer(), [__]jbd2_journal_unfile_buffer(), and __jdb2_journal_remove_checkpoint() can free journal_head which needs modification of a few callers. Also we have to be careful because once journal_head is removed, buffer_head might be freed as well. So we have to get our own buffer_head reference where it matters. Signed-off-by: Jan Kara Signed-off-by: "Theodore Ts'o" --- fs/jbd2/checkpoint.c | 28 +++++++++------- fs/jbd2/commit.c | 33 +++++++++++-------- fs/jbd2/journal.c | 91 ++++++++++++++++----------------------------------- fs/jbd2/transaction.c | 67 +++++++++++++++++++------------------ include/linux/jbd2.h | 2 -- 5 files changed, 99 insertions(+), 122 deletions(-) (limited to 'include') diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 6a79fd0a1a32..2c62c5aae82f 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -97,10 +97,14 @@ static int __try_to_free_cp_buf(struct journal_head *jh) if (jh->b_jlist == BJ_None && !buffer_locked(bh) && !buffer_dirty(bh) && !buffer_write_io_error(bh)) { + /* + * Get our reference so that bh cannot be freed before + * we unlock it + */ + get_bh(bh); JBUFFER_TRACE(jh, "remove from checkpoint list"); ret = __jbd2_journal_remove_checkpoint(jh) + 1; jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); BUFFER_TRACE(bh, "release"); __brelse(bh); } else { @@ -223,8 +227,8 @@ restart: spin_lock(&journal->j_list_lock); goto restart; } + get_bh(bh); if (buffer_locked(bh)) { - atomic_inc(&bh->b_count); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); wait_on_buffer(bh); @@ -243,7 +247,6 @@ restart: */ released = __jbd2_journal_remove_checkpoint(jh); jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); __brelse(bh); } @@ -284,7 +287,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, int ret = 0; if (buffer_locked(bh)) { - atomic_inc(&bh->b_count); + get_bh(bh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); wait_on_buffer(bh); @@ -316,12 +319,12 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, ret = 1; if (unlikely(buffer_write_io_error(bh))) ret = -EIO; + get_bh(bh); J_ASSERT_JH(jh, !buffer_jbddirty(bh)); BUFFER_TRACE(bh, "remove from checkpoint"); __jbd2_journal_remove_checkpoint(jh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); __brelse(bh); } else { /* @@ -554,7 +557,8 @@ int jbd2_cleanup_journal_tail(journal_t *journal) /* * journal_clean_one_cp_list * - * Find all the written-back checkpoint buffers in the given list and release them. + * Find all the written-back checkpoint buffers in the given list and + * release them. * * Called with the journal locked. * Called with j_list_lock held. @@ -663,8 +667,8 @@ out: * checkpoint lists. * * The function returns 1 if it frees the transaction, 0 otherwise. + * The function can free jh and bh. * - * This function is called with the journal locked. * This function is called with j_list_lock held. * This function is called with jbd_lock_bh_state(jh2bh(jh)) */ @@ -684,13 +688,14 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh) } journal = transaction->t_journal; + JBUFFER_TRACE(jh, "removing from transaction"); __buffer_unlink(jh); jh->b_cp_transaction = NULL; + jbd2_journal_put_journal_head(jh); if (transaction->t_checkpoint_list != NULL || transaction->t_checkpoint_io_list != NULL) goto out; - JBUFFER_TRACE(jh, "transaction has no more buffers"); /* * There is one special case to worry about: if we have just pulled the @@ -701,10 +706,8 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh) * The locking here around t_state is a bit sleazy. * See the comment at the end of jbd2_journal_commit_transaction(). */ - if (transaction->t_state != T_FINISHED) { - JBUFFER_TRACE(jh, "belongs to running/committing transaction"); + if (transaction->t_state != T_FINISHED) goto out; - } /* OK, that was the last buffer for the transaction: we can now safely remove this transaction from the log */ @@ -723,7 +726,6 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh) wake_up(&journal->j_wait_logspace); ret = 1; out: - JBUFFER_TRACE(jh, "exit"); return ret; } @@ -742,6 +744,8 @@ void __jbd2_journal_insert_checkpoint(struct journal_head *jh, J_ASSERT_JH(jh, buffer_dirty(jh2bh(jh)) || buffer_jbddirty(jh2bh(jh))); J_ASSERT_JH(jh, jh->b_cp_transaction == NULL); + /* Get reference for checkpointing transaction */ + jbd2_journal_grab_journal_head(jh2bh(jh)); jh->b_cp_transaction = transaction; if (!transaction->t_checkpoint_list) { diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 7f21cf3aaf92..eef6979821a4 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -848,10 +848,16 @@ restart_loop: while (commit_transaction->t_forget) { transaction_t *cp_transaction; struct buffer_head *bh; + int try_to_free = 0; jh = commit_transaction->t_forget; spin_unlock(&journal->j_list_lock); bh = jh2bh(jh); + /* + * Get a reference so that bh cannot be freed before we are + * done with it. + */ + get_bh(bh); jbd_lock_bh_state(bh); J_ASSERT_JH(jh, jh->b_transaction == commit_transaction); @@ -914,28 +920,27 @@ restart_loop: __jbd2_journal_insert_checkpoint(jh, commit_transaction); if (is_journal_aborted(journal)) clear_buffer_jbddirty(bh); - JBUFFER_TRACE(jh, "refile for checkpoint writeback"); - __jbd2_journal_refile_buffer(jh); - jbd_unlock_bh_state(bh); } else { J_ASSERT_BH(bh, !buffer_dirty(bh)); - /* The buffer on BJ_Forget list and not jbddirty means + /* + * The buffer on BJ_Forget list and not jbddirty means * it has been freed by this transaction and hence it * could not have been reallocated until this * transaction has committed. *BUT* it could be * reallocated once we have written all the data to * disk and before we process the buffer on BJ_Forget - * list. */ - JBUFFER_TRACE(jh, "refile or unfile freed buffer"); - __jbd2_journal_refile_buffer(jh); - if (!jh->b_transaction) { - jbd_unlock_bh_state(bh); - /* needs a brelse */ - jbd2_journal_remove_journal_head(bh); - release_buffer_page(bh); - } else - jbd_unlock_bh_state(bh); + * list. + */ + if (!jh->b_next_transaction) + try_to_free = 1; } + JBUFFER_TRACE(jh, "refile or unfile buffer"); + __jbd2_journal_refile_buffer(jh); + jbd_unlock_bh_state(bh); + if (try_to_free) + release_buffer_page(bh); /* Drops bh reference */ + else + __brelse(bh); cond_resched_lock(&journal->j_list_lock); } spin_unlock(&journal->j_list_lock); diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 9a7826990304..0dfa5b598e68 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -2078,10 +2078,9 @@ static void journal_free_journal_head(struct journal_head *jh) * When a buffer has its BH_JBD bit set it is immune from being released by * core kernel code, mainly via ->b_count. * - * A journal_head may be detached from its buffer_head when the journal_head's - * b_transaction, b_cp_transaction and b_next_transaction pointers are NULL. - * Various places in JBD call jbd2_journal_remove_journal_head() to indicate that the - * journal_head can be dropped if needed. + * A journal_head is detached from its buffer_head when the journal_head's + * b_jcount reaches zero. Running transaction (b_transaction) and checkpoint + * transaction (b_cp_transaction) hold their references to b_jcount. * * Various places in the kernel want to attach a journal_head to a buffer_head * _before_ attaching the journal_head to a transaction. To protect the @@ -2094,17 +2093,16 @@ static void journal_free_journal_head(struct journal_head *jh) * (Attach a journal_head if needed. Increments b_jcount) * struct journal_head *jh = jbd2_journal_add_journal_head(bh); * ... + * (Get another reference for transaction) + * jbd2_journal_grab_journal_head(bh); * jh->b_transaction = xxx; + * (Put original reference) * jbd2_journal_put_journal_head(jh); - * - * Now, the journal_head's b_jcount is zero, but it is safe from being released - * because it has a non-zero b_transaction. */ /* * Give a buffer_head a journal_head. * - * Doesn't need the journal lock. * May sleep. */ struct journal_head *jbd2_journal_add_journal_head(struct buffer_head *bh) @@ -2168,61 +2166,29 @@ static void __journal_remove_journal_head(struct buffer_head *bh) struct journal_head *jh = bh2jh(bh); J_ASSERT_JH(jh, jh->b_jcount >= 0); - - get_bh(bh); - if (jh->b_jcount == 0) { - if (jh->b_transaction == NULL && - jh->b_next_transaction == NULL && - jh->b_cp_transaction == NULL) { - J_ASSERT_JH(jh, jh->b_jlist == BJ_None); - J_ASSERT_BH(bh, buffer_jbd(bh)); - J_ASSERT_BH(bh, jh2bh(jh) == bh); - BUFFER_TRACE(bh, "remove journal_head"); - if (jh->b_frozen_data) { - printk(KERN_WARNING "%s: freeing " - "b_frozen_data\n", - __func__); - jbd2_free(jh->b_frozen_data, bh->b_size); - } - if (jh->b_committed_data) { - printk(KERN_WARNING "%s: freeing " - "b_committed_data\n", - __func__); - jbd2_free(jh->b_committed_data, bh->b_size); - } - bh->b_private = NULL; - jh->b_bh = NULL; /* debug, really */ - clear_buffer_jbd(bh); - __brelse(bh); - journal_free_journal_head(jh); - } else { - BUFFER_TRACE(bh, "journal_head was locked"); - } + J_ASSERT_JH(jh, jh->b_transaction == NULL); + J_ASSERT_JH(jh, jh->b_next_transaction == NULL); + J_ASSERT_JH(jh, jh->b_cp_transaction == NULL); + J_ASSERT_JH(jh, jh->b_jlist == BJ_None); + J_ASSERT_BH(bh, buffer_jbd(bh)); + J_ASSERT_BH(bh, jh2bh(jh) == bh); + BUFFER_TRACE(bh, "remove journal_head"); + if (jh->b_frozen_data) { + printk(KERN_WARNING "%s: freeing b_frozen_data\n", __func__); + jbd2_free(jh->b_frozen_data, bh->b_size); } + if (jh->b_committed_data) { + printk(KERN_WARNING "%s: freeing b_committed_data\n", __func__); + jbd2_free(jh->b_committed_data, bh->b_size); + } + bh->b_private = NULL; + jh->b_bh = NULL; /* debug, really */ + clear_buffer_jbd(bh); + journal_free_journal_head(jh); } /* - * jbd2_journal_remove_journal_head(): if the buffer isn't attached to a transaction - * and has a zero b_jcount then remove and release its journal_head. If we did - * see that the buffer is not used by any transaction we also "logically" - * decrement ->b_count. - * - * We in fact take an additional increment on ->b_count as a convenience, - * because the caller usually wants to do additional things with the bh - * after calling here. - * The caller of jbd2_journal_remove_journal_head() *must* run __brelse(bh) at some - * time. Once the caller has run __brelse(), the buffer is eligible for - * reaping by try_to_free_buffers(). - */ -void jbd2_journal_remove_journal_head(struct buffer_head *bh) -{ - jbd_lock_bh_journal_head(bh); - __journal_remove_journal_head(bh); - jbd_unlock_bh_journal_head(bh); -} - -/* - * Drop a reference on the passed journal_head. If it fell to zero then try to + * Drop a reference on the passed journal_head. If it fell to zero then * release the journal_head from the buffer_head. */ void jbd2_journal_put_journal_head(struct journal_head *jh) @@ -2232,11 +2198,12 @@ void jbd2_journal_put_journal_head(struct journal_head *jh) jbd_lock_bh_journal_head(bh); J_ASSERT_JH(jh, jh->b_jcount > 0); --jh->b_jcount; - if (!jh->b_jcount && !jh->b_transaction) { + if (!jh->b_jcount) { __journal_remove_journal_head(bh); + jbd_unlock_bh_journal_head(bh); __brelse(bh); - } - jbd_unlock_bh_journal_head(bh); + } else + jbd_unlock_bh_journal_head(bh); } /* diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 547b101049e5..2d7109414cdd 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -30,6 +30,7 @@ #include static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh); +static void __jbd2_journal_unfile_buffer(struct journal_head *jh); /* * jbd2_get_transaction: obtain a new transaction_t object. @@ -764,7 +765,6 @@ repeat: if (!jh->b_transaction) { JBUFFER_TRACE(jh, "no transaction"); J_ASSERT_JH(jh, !jh->b_next_transaction); - jh->b_transaction = transaction; JBUFFER_TRACE(jh, "file as BJ_Reserved"); spin_lock(&journal->j_list_lock); __jbd2_journal_file_buffer(jh, transaction, BJ_Reserved); @@ -895,8 +895,6 @@ int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh) * committed and so it's safe to clear the dirty bit. */ clear_buffer_dirty(jh2bh(jh)); - jh->b_transaction = transaction; - /* first access by this transaction */ jh->b_modified = 0; @@ -1230,8 +1228,6 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh) __jbd2_journal_file_buffer(jh, transaction, BJ_Forget); } else { __jbd2_journal_unfile_buffer(jh); - jbd2_journal_remove_journal_head(bh); - __brelse(bh); if (!buffer_jbd(bh)) { spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); @@ -1554,19 +1550,32 @@ void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh) mark_buffer_dirty(bh); /* Expose it to the VM */ } -void __jbd2_journal_unfile_buffer(struct journal_head *jh) +/* + * Remove buffer from all transactions. + * + * Called with bh_state lock and j_list_lock + * + * jh and bh may be already freed when this function returns. + */ +static void __jbd2_journal_unfile_buffer(struct journal_head *jh) { __jbd2_journal_temp_unlink_buffer(jh); jh->b_transaction = NULL; + jbd2_journal_put_journal_head(jh); } void jbd2_journal_unfile_buffer(journal_t *journal, struct journal_head *jh) { - jbd_lock_bh_state(jh2bh(jh)); + struct buffer_head *bh = jh2bh(jh); + + /* Get reference so that buffer cannot be freed before we unlock it */ + get_bh(bh); + jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); __jbd2_journal_unfile_buffer(jh); spin_unlock(&journal->j_list_lock); - jbd_unlock_bh_state(jh2bh(jh)); + jbd_unlock_bh_state(bh); + __brelse(bh); } /* @@ -1593,8 +1602,6 @@ __journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh) if (jh->b_jlist == BJ_None) { JBUFFER_TRACE(jh, "remove from checkpoint list"); __jbd2_journal_remove_checkpoint(jh); - jbd2_journal_remove_journal_head(bh); - __brelse(bh); } } spin_unlock(&journal->j_list_lock); @@ -1657,7 +1664,6 @@ int jbd2_journal_try_to_free_buffers(journal_t *journal, /* * We take our own ref against the journal_head here to avoid * having to add tons of locking around each instance of - * jbd2_journal_remove_journal_head() and * jbd2_journal_put_journal_head(). */ jh = jbd2_journal_grab_journal_head(bh); @@ -1695,10 +1701,9 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction) int may_free = 1; struct buffer_head *bh = jh2bh(jh); - __jbd2_journal_unfile_buffer(jh); - if (jh->b_cp_transaction) { JBUFFER_TRACE(jh, "on running+cp transaction"); + __jbd2_journal_temp_unlink_buffer(jh); /* * We don't want to write the buffer anymore, clear the * bit so that we don't confuse checks in @@ -1709,8 +1714,7 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction) may_free = 0; } else { JBUFFER_TRACE(jh, "on running transaction"); - jbd2_journal_remove_journal_head(bh); - __brelse(bh); + __jbd2_journal_unfile_buffer(jh); } return may_free; } @@ -1988,6 +1992,8 @@ void __jbd2_journal_file_buffer(struct journal_head *jh, if (jh->b_transaction) __jbd2_journal_temp_unlink_buffer(jh); + else + jbd2_journal_grab_journal_head(bh); jh->b_transaction = transaction; switch (jlist) { @@ -2039,9 +2045,10 @@ void jbd2_journal_file_buffer(struct journal_head *jh, * already started to be used by a subsequent transaction, refile the * buffer on that transaction's metadata list. * - * Called under journal->j_list_lock - * + * Called under j_list_lock * Called under jbd_lock_bh_state(jh2bh(jh)) + * + * jh and bh may be already free when this function returns */ void __jbd2_journal_refile_buffer(struct journal_head *jh) { @@ -2065,6 +2072,11 @@ void __jbd2_journal_refile_buffer(struct journal_head *jh) was_dirty = test_clear_buffer_jbddirty(bh); __jbd2_journal_temp_unlink_buffer(jh); + /* + * We set b_transaction here because b_next_transaction will inherit + * our jh reference and thus __jbd2_journal_file_buffer() must not + * take a new one. + */ jh->b_transaction = jh->b_next_transaction; jh->b_next_transaction = NULL; if (buffer_freed(bh)) @@ -2081,30 +2093,21 @@ void __jbd2_journal_refile_buffer(struct journal_head *jh) } /* - * For the unlocked version of this call, also make sure that any - * hanging journal_head is cleaned up if necessary. - * - * __jbd2_journal_refile_buffer is usually called as part of a single locked - * operation on a buffer_head, in which the caller is probably going to - * be hooking the journal_head onto other lists. In that case it is up - * to the caller to remove the journal_head if necessary. For the - * unlocked jbd2_journal_refile_buffer call, the caller isn't going to be - * doing anything else to the buffer so we need to do the cleanup - * ourselves to avoid a jh leak. - * - * *** The journal_head may be freed by this call! *** + * __jbd2_journal_refile_buffer() with necessary locking added. We take our + * bh reference so that we can safely unlock bh. + * + * The jh and bh may be freed by this call. */ void jbd2_journal_refile_buffer(journal_t *journal, struct journal_head *jh) { struct buffer_head *bh = jh2bh(jh); + /* Get reference so that buffer cannot be freed before we unlock it */ + get_bh(bh); jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); - __jbd2_journal_refile_buffer(jh); jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); - spin_unlock(&journal->j_list_lock); __brelse(bh); } diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 4ecb7b16b278..d087c2e7b2aa 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1024,7 +1024,6 @@ struct journal_s /* Filing buffers */ extern void jbd2_journal_unfile_buffer(journal_t *, struct journal_head *); -extern void __jbd2_journal_unfile_buffer(struct journal_head *); extern void __jbd2_journal_refile_buffer(struct journal_head *); extern void jbd2_journal_refile_buffer(journal_t *, struct journal_head *); extern void __jbd2_journal_file_buffer(struct journal_head *, transaction_t *, int); @@ -1165,7 +1164,6 @@ extern void jbd2_journal_release_jbd_inode(journal_t *journal, struct jbd2_in */ struct journal_head *jbd2_journal_add_journal_head(struct buffer_head *bh); struct journal_head *jbd2_journal_grab_journal_head(struct buffer_head *bh); -void jbd2_journal_remove_journal_head(struct buffer_head *bh); void jbd2_journal_put_journal_head(struct journal_head *jh); /* -- cgit v1.2.3 From 09223371deac67d08ca0b70bd18787920284c967 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Tue, 14 Jun 2011 13:26:25 +0800 Subject: rcu: Use softirq to address performance regression Commit a26ac2455ffcf3(rcu: move TREE_RCU from softirq to kthread) introduced performance regression. In an AIM7 test, this commit degraded performance by about 40%. The commit runs rcu callbacks in a kthread instead of softirq. We observed high rate of context switch which is caused by this. Out test system has 64 CPUs and HZ is 1000, so we saw more than 64k context switch per second which is caused by RCU's per-CPU kthread. A trace showed that most of the time the RCU per-CPU kthread doesn't actually handle any callbacks, but instead just does a very small amount of work handling grace periods. This means that RCU's per-CPU kthreads are making the scheduler do quite a bit of work in order to allow a very small amount of RCU-related processing to be done. Alex Shi's analysis determined that this slowdown is due to lock contention within the scheduler. Unfortunately, as Peter Zijlstra points out, the scheduler's real-time semantics require global action, which means that this contention is inherent in real-time scheduling. (Yes, perhaps someone will come up with a workaround -- otherwise, -rt is not going to do well on large SMP systems -- but this patch will work around this issue in the meantime. And "the meantime" might well be forever.) This patch therefore re-introduces softirq processing to RCU, but only for core RCU work. RCU callbacks are still executed in kthread context, so that only a small amount of RCU work runs in softirq context in the common case. This should minimize ksoftirqd execution, allowing us to skip boosting of ksoftirqd for CONFIG_RCU_BOOST=y kernels. Signed-off-by: Shaohua Li Tested-by: "Alex,Shi" Signed-off-by: Paul E. McKenney --- Documentation/filesystems/proc.txt | 1 + include/linux/interrupt.h | 1 + include/trace/events/irq.h | 3 ++- kernel/rcutree.c | 23 +++++++++++++++++++---- kernel/rcutree.h | 1 + kernel/rcutree_plugin.h | 9 +++++++++ kernel/softirq.c | 2 +- tools/perf/util/trace-event-parse.c | 1 + 8 files changed, 35 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt index f48178024067..db3b1aba32a3 100644 --- a/Documentation/filesystems/proc.txt +++ b/Documentation/filesystems/proc.txt @@ -843,6 +843,7 @@ Provides counts of softirq handlers serviced since boot time, for each cpu. TASKLET: 0 0 0 290 SCHED: 27035 26983 26971 26746 HRTIMER: 0 0 0 0 + RCU: 1678 1769 2178 2250 1.3 IDE devices in /proc/ide diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index 6c12989839d9..f6efed0039ed 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -414,6 +414,7 @@ enum TASKLET_SOFTIRQ, SCHED_SOFTIRQ, HRTIMER_SOFTIRQ, + RCU_SOFTIRQ, /* Preferable RCU should always be the last softirq */ NR_SOFTIRQS }; diff --git a/include/trace/events/irq.h b/include/trace/events/irq.h index ae045ca7d356..1c09820df585 100644 --- a/include/trace/events/irq.h +++ b/include/trace/events/irq.h @@ -20,7 +20,8 @@ struct softirq_action; softirq_name(BLOCK_IOPOLL), \ softirq_name(TASKLET), \ softirq_name(SCHED), \ - softirq_name(HRTIMER)) + softirq_name(HRTIMER), \ + softirq_name(RCU)) /** * irq_handler_entry - called immediately before the irq action handler diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 0a8ec5b2e208..ae5c9ea68662 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -100,6 +100,7 @@ static char rcu_kthreads_spawnable; static void rcu_node_kthread_setaffinity(struct rcu_node *rnp, int outgoingcpu); static void invoke_rcu_cpu_kthread(void); +static void __invoke_rcu_cpu_kthread(void); #define RCU_KTHREAD_PRIO 1 /* RT priority for per-CPU kthreads. */ @@ -1442,13 +1443,21 @@ __rcu_process_callbacks(struct rcu_state *rsp, struct rcu_data *rdp) } /* If there are callbacks ready, invoke them. */ - rcu_do_batch(rsp, rdp); + if (cpu_has_callbacks_ready_to_invoke(rdp)) + __invoke_rcu_cpu_kthread(); +} + +static void rcu_kthread_do_work(void) +{ + rcu_do_batch(&rcu_sched_state, &__get_cpu_var(rcu_sched_data)); + rcu_do_batch(&rcu_bh_state, &__get_cpu_var(rcu_bh_data)); + rcu_preempt_do_callbacks(); } /* * Do softirq processing for the current CPU. */ -static void rcu_process_callbacks(void) +static void rcu_process_callbacks(struct softirq_action *unused) { __rcu_process_callbacks(&rcu_sched_state, &__get_cpu_var(rcu_sched_data)); @@ -1465,7 +1474,7 @@ static void rcu_process_callbacks(void) * the current CPU with interrupts disabled, the rcu_cpu_kthread_task * cannot disappear out from under us. */ -static void invoke_rcu_cpu_kthread(void) +static void __invoke_rcu_cpu_kthread(void) { unsigned long flags; @@ -1479,6 +1488,11 @@ static void invoke_rcu_cpu_kthread(void) local_irq_restore(flags); } +static void invoke_rcu_cpu_kthread(void) +{ + raise_softirq(RCU_SOFTIRQ); +} + /* * Wake up the specified per-rcu_node-structure kthread. * Because the per-rcu_node kthreads are immortal, we don't need @@ -1613,7 +1627,7 @@ static int rcu_cpu_kthread(void *arg) *workp = 0; local_irq_restore(flags); if (work) - rcu_process_callbacks(); + rcu_kthread_do_work(); local_bh_enable(); if (*workp != 0) spincnt++; @@ -2387,6 +2401,7 @@ void __init rcu_init(void) rcu_init_one(&rcu_sched_state, &rcu_sched_data); rcu_init_one(&rcu_bh_state, &rcu_bh_data); __rcu_init_preempt(); + open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); /* * We don't need protection against CPU-hotplug here because diff --git a/kernel/rcutree.h b/kernel/rcutree.h index 7b9a08b4aaea..0fed6b934d2a 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -439,6 +439,7 @@ static void rcu_preempt_offline_cpu(int cpu); #endif /* #ifdef CONFIG_HOTPLUG_CPU */ static void rcu_preempt_check_callbacks(int cpu); static void rcu_preempt_process_callbacks(void); +static void rcu_preempt_do_callbacks(void); void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)); #if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_TREE_PREEMPT_RCU) static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp); diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index ea2e2fb79e81..38d09c5f2b41 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -602,6 +602,11 @@ static void rcu_preempt_process_callbacks(void) &__get_cpu_var(rcu_preempt_data)); } +static void rcu_preempt_do_callbacks(void) +{ + rcu_do_batch(&rcu_preempt_state, &__get_cpu_var(rcu_preempt_data)); +} + /* * Queue a preemptible-RCU callback for invocation after a grace period. */ @@ -997,6 +1002,10 @@ static void rcu_preempt_process_callbacks(void) { } +static void rcu_preempt_do_callbacks(void) +{ +} + /* * Wait for an rcu-preempt grace period, but make it happen quickly. * But because preemptible RCU does not exist, map to rcu-sched. diff --git a/kernel/softirq.c b/kernel/softirq.c index 13960170cad4..40cf63ddd4b3 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -58,7 +58,7 @@ DEFINE_PER_CPU(struct task_struct *, ksoftirqd); char *softirq_to_name[NR_SOFTIRQS] = { "HI", "TIMER", "NET_TX", "NET_RX", "BLOCK", "BLOCK_IOPOLL", - "TASKLET", "SCHED", "HRTIMER" + "TASKLET", "SCHED", "HRTIMER", "RCU" }; /* diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c index 1e88485c16a0..0a7ed5b5e281 100644 --- a/tools/perf/util/trace-event-parse.c +++ b/tools/perf/util/trace-event-parse.c @@ -2187,6 +2187,7 @@ static const struct flag flags[] = { { "TASKLET_SOFTIRQ", 6 }, { "SCHED_SOFTIRQ", 7 }, { "HRTIMER_SOFTIRQ", 8 }, + { "RCU_SOFTIRQ", 9 }, { "HRTIMER_NORESTART", 0 }, { "HRTIMER_RESTART", 1 }, -- cgit v1.2.3 From 0b760113a3a155269a3fba93a409c640031dd68f Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 31 May 2011 15:15:34 -0400 Subject: NLM: Don't hang forever on NLM unlock requests If the NLM daemon is killed on the NFS server, we can currently end up hanging forever on an 'unlock' request, instead of aborting. Basically, if the rpcbind request fails, or the server keeps returning garbage, we really want to quit instead of retrying. Tested-by: Vasily Averin Signed-off-by: Trond Myklebust Cc: stable@kernel.org --- fs/lockd/clntproc.c | 8 +++++++- include/linux/sunrpc/sched.h | 3 ++- net/sunrpc/clnt.c | 3 +++ net/sunrpc/sched.c | 1 + 4 files changed, 13 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index adb45ec9038c..e374050a911c 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -708,7 +708,13 @@ static void nlmclnt_unlock_callback(struct rpc_task *task, void *data) if (task->tk_status < 0) { dprintk("lockd: unlock failed (err = %d)\n", -task->tk_status); - goto retry_rebind; + switch (task->tk_status) { + case -EACCES: + case -EIO: + goto die; + default: + goto retry_rebind; + } } if (status == NLM_LCK_DENIED_GRACE_PERIOD) { rpc_delay(task, NLMCLNT_GRACE_WAIT); diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h index f73c482ec9c6..fe2d8e6b923b 100644 --- a/include/linux/sunrpc/sched.h +++ b/include/linux/sunrpc/sched.h @@ -84,7 +84,8 @@ struct rpc_task { #endif unsigned char tk_priority : 2,/* Task priority */ tk_garb_retry : 2, - tk_cred_retry : 2; + tk_cred_retry : 2, + tk_rebind_retry : 2; }; #define tk_xprt tk_client->cl_xprt diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index b84d7395535e..566bcfd067f6 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -1175,6 +1175,9 @@ call_bind_status(struct rpc_task *task) status = -EOPNOTSUPP; break; } + if (task->tk_rebind_retry == 0) + break; + task->tk_rebind_retry--; rpc_delay(task, 3*HZ); goto retry_timeout; case -ETIMEDOUT: diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index 6b43ee7221d5..a27406b1654f 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -792,6 +792,7 @@ static void rpc_init_task(struct rpc_task *task, const struct rpc_task_setup *ta /* Initialize retry counters */ task->tk_garb_retry = 2; task->tk_cred_retry = 2; + task->tk_rebind_retry = 2; task->tk_priority = task_setup_data->priority - RPC_PRIORITY_LOW; task->tk_owner = current->tgid; -- cgit v1.2.3 From c9c30dd5f73dccaa326a54dfcf490316946aea87 Mon Sep 17 00:00:00 2001 From: Benny Halevy Date: Sat, 11 Jun 2011 17:08:39 -0400 Subject: NFSv4.1: deprecate headerpadsz in CREATE_SESSION We don't support header padding yet so better off ditching it Reported-by: Sid Moore Signed-off-by: Benny Halevy Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 4 ---- fs/nfs/nfs4xdr.c | 10 ++++++---- include/linux/nfs_xdr.h | 1 - 3 files changed, 6 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index d2c4b59c896d..3ca449789545 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -5098,7 +5098,6 @@ static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args) if (mxresp_sz == 0) mxresp_sz = NFS_MAX_FILE_IO_SIZE; /* Fore channel attributes */ - args->fc_attrs.headerpadsz = 0; args->fc_attrs.max_rqst_sz = mxrqst_sz; args->fc_attrs.max_resp_sz = mxresp_sz; args->fc_attrs.max_ops = NFS4_MAX_OPS; @@ -5111,7 +5110,6 @@ static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args) args->fc_attrs.max_ops, args->fc_attrs.max_reqs); /* Back channel attributes */ - args->bc_attrs.headerpadsz = 0; args->bc_attrs.max_rqst_sz = PAGE_SIZE; args->bc_attrs.max_resp_sz = PAGE_SIZE; args->bc_attrs.max_resp_sz_cached = 0; @@ -5131,8 +5129,6 @@ static int nfs4_verify_fore_channel_attrs(struct nfs41_create_session_args *args struct nfs4_channel_attrs *sent = &args->fc_attrs; struct nfs4_channel_attrs *rcvd = &session->fc_attrs; - if (rcvd->headerpadsz > sent->headerpadsz) - return -EINVAL; if (rcvd->max_resp_sz > sent->max_resp_sz) return -EINVAL; /* diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index d869a5e5464b..c4b7d6c04948 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -1725,7 +1725,7 @@ static void encode_create_session(struct xdr_stream *xdr, *p++ = cpu_to_be32(args->flags); /*flags */ /* Fore Channel */ - *p++ = cpu_to_be32(args->fc_attrs.headerpadsz); /* header padding size */ + *p++ = cpu_to_be32(0); /* header padding size */ *p++ = cpu_to_be32(args->fc_attrs.max_rqst_sz); /* max req size */ *p++ = cpu_to_be32(args->fc_attrs.max_resp_sz); /* max resp size */ *p++ = cpu_to_be32(max_resp_sz_cached); /* Max resp sz cached */ @@ -1734,7 +1734,7 @@ static void encode_create_session(struct xdr_stream *xdr, *p++ = cpu_to_be32(0); /* rdmachannel_attrs */ /* Back Channel */ - *p++ = cpu_to_be32(args->fc_attrs.headerpadsz); /* header padding size */ + *p++ = cpu_to_be32(0); /* header padding size */ *p++ = cpu_to_be32(args->bc_attrs.max_rqst_sz); /* max req size */ *p++ = cpu_to_be32(args->bc_attrs.max_resp_sz); /* max resp size */ *p++ = cpu_to_be32(args->bc_attrs.max_resp_sz_cached); /* Max resp sz cached */ @@ -4997,12 +4997,14 @@ static int decode_chan_attrs(struct xdr_stream *xdr, struct nfs4_channel_attrs *attrs) { __be32 *p; - u32 nr_attrs; + u32 nr_attrs, val; p = xdr_inline_decode(xdr, 28); if (unlikely(!p)) goto out_overflow; - attrs->headerpadsz = be32_to_cpup(p++); + val = be32_to_cpup(p++); /* headerpadsz */ + if (val) + return -EINVAL; /* no support for header padding yet */ attrs->max_rqst_sz = be32_to_cpup(p++); attrs->max_resp_sz = be32_to_cpup(p++); attrs->max_resp_sz_cached = be32_to_cpup(p++); diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 5e8444a11adf..00848d86ffb2 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -158,7 +158,6 @@ struct nfs_seqid; /* nfs41 sessions channel attributes */ struct nfs4_channel_attrs { - u32 headerpadsz; u32 max_rqst_sz; u32 max_resp_sz; u32 max_resp_sz_cached; -- cgit v1.2.3 From a59ec1e7ff98cc4365d5b1bff4e7102e86b5716b Mon Sep 17 00:00:00 2001 From: Michael Hennerich Date: Wed, 15 Jun 2011 15:08:11 -0700 Subject: backlight: new driver for the ADP8870 backlight devices Signed-off-by: Michael Hennerich Signed-off-by: Mike Frysinger Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- .../testing/sysfs-class-backlight-driver-adp8870 | 56 ++ drivers/video/backlight/Kconfig | 12 + drivers/video/backlight/Makefile | 1 + drivers/video/backlight/adp8870_bl.c | 1011 ++++++++++++++++++++ include/linux/i2c/adp8870.h | 153 +++ 5 files changed, 1233 insertions(+) create mode 100644 Documentation/ABI/testing/sysfs-class-backlight-driver-adp8870 create mode 100644 drivers/video/backlight/adp8870_bl.c create mode 100644 include/linux/i2c/adp8870.h (limited to 'include') diff --git a/Documentation/ABI/testing/sysfs-class-backlight-driver-adp8870 b/Documentation/ABI/testing/sysfs-class-backlight-driver-adp8870 new file mode 100644 index 000000000000..aa11dbdd794b --- /dev/null +++ b/Documentation/ABI/testing/sysfs-class-backlight-driver-adp8870 @@ -0,0 +1,56 @@ +What: /sys/class/backlight//_max +What: /sys/class/backlight//l1_daylight_max +What: /sys/class/backlight//l2_bright_max +What: /sys/class/backlight//l3_office_max +What: /sys/class/backlight//l4_indoor_max +What: /sys/class/backlight//l5_dark_max +Date: Mai 2011 +KernelVersion: 2.6.40 +Contact: device-drivers-devel@blackfin.uclinux.org +Description: + Control the maximum brightness for + on this . Values are between 0 and 127. This file + will also show the brightness level stored for this + . + +What: /sys/class/backlight//_dim +What: /sys/class/backlight//l2_bright_dim +What: /sys/class/backlight//l3_office_dim +What: /sys/class/backlight//l4_indoor_dim +What: /sys/class/backlight//l5_dark_dim +Date: Mai 2011 +KernelVersion: 2.6.40 +Contact: device-drivers-devel@blackfin.uclinux.org +Description: + Control the dim brightness for + on this . Values are between 0 and 127, typically + set to 0. Full off when the backlight is disabled. + This file will also show the dim brightness level stored for + this . + +What: /sys/class/backlight//ambient_light_level +Date: Mai 2011 +KernelVersion: 2.6.40 +Contact: device-drivers-devel@blackfin.uclinux.org +Description: + Get conversion value of the light sensor. + This value is updated every 80 ms (when the light sensor + is enabled). Returns integer between 0 (dark) and + 8000 (max ambient brightness) + +What: /sys/class/backlight//ambient_light_zone +Date: Mai 2011 +KernelVersion: 2.6.40 +Contact: device-drivers-devel@blackfin.uclinux.org +Description: + Get/Set current ambient light zone. Reading returns + integer between 1..5 (1 = daylight, 2 = bright, ..., 5 = dark). + Writing a value between 1..5 forces the backlight controller + to enter the corresponding ambient light zone. + Writing 0 returns to normal/automatic ambient light level + operation. The ambient light sensing feature on these devices + is an extension to the API documented in + Documentation/ABI/stable/sysfs-class-backlight. + It can be enabled by writing the value stored in + /sys/class/backlight//max_brightness to + /sys/class/backlight//brightness. \ No newline at end of file diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig index 0c9373bedd1f..2d93c8d61ad5 100644 --- a/drivers/video/backlight/Kconfig +++ b/drivers/video/backlight/Kconfig @@ -302,6 +302,18 @@ config BACKLIGHT_ADP8860 To compile this driver as a module, choose M here: the module will be called adp8860_bl. +config BACKLIGHT_ADP8870 + tristate "Backlight Driver for ADP8870 using WLED" + depends on BACKLIGHT_CLASS_DEVICE && I2C + select NEW_LEDS + select LEDS_CLASS + help + If you have a LCD backlight connected to the ADP8870, + say Y here to enable this driver. + + To compile this driver as a module, choose M here: the module will + be called adp8870_bl. + config BACKLIGHT_88PM860X tristate "Backlight Driver for 88PM8606 using WLED" depends on MFD_88PM860X diff --git a/drivers/video/backlight/Makefile b/drivers/video/backlight/Makefile index b9ca8490df87..ee72adb8786e 100644 --- a/drivers/video/backlight/Makefile +++ b/drivers/video/backlight/Makefile @@ -34,6 +34,7 @@ obj-$(CONFIG_BACKLIGHT_WM831X) += wm831x_bl.o obj-$(CONFIG_BACKLIGHT_ADX) += adx_bl.o obj-$(CONFIG_BACKLIGHT_ADP5520) += adp5520_bl.o obj-$(CONFIG_BACKLIGHT_ADP8860) += adp8860_bl.o +obj-$(CONFIG_BACKLIGHT_ADP8870) += adp8870_bl.o obj-$(CONFIG_BACKLIGHT_88PM860X) += 88pm860x_bl.o obj-$(CONFIG_BACKLIGHT_PCF50633) += pcf50633-backlight.o diff --git a/drivers/video/backlight/adp8870_bl.c b/drivers/video/backlight/adp8870_bl.c new file mode 100644 index 000000000000..e320b62576e2 --- /dev/null +++ b/drivers/video/backlight/adp8870_bl.c @@ -0,0 +1,1011 @@ +/* + * Backlight driver for Analog Devices ADP8870 Backlight Devices + * + * Copyright 2009-2011 Analog Devices Inc. + * + * Licensed under the GPL-2 or later. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#define ADP8870_EXT_FEATURES +#define ADP8870_USE_LEDS + + +#define ADP8870_MFDVID 0x00 /* Manufacturer and device ID */ +#define ADP8870_MDCR 0x01 /* Device mode and status */ +#define ADP8870_INT_STAT 0x02 /* Interrupts status */ +#define ADP8870_INT_EN 0x03 /* Interrupts enable */ +#define ADP8870_CFGR 0x04 /* Configuration register */ +#define ADP8870_BLSEL 0x05 /* Sink enable backlight or independent */ +#define ADP8870_PWMLED 0x06 /* PWM Enable Selection Register */ +#define ADP8870_BLOFF 0x07 /* Backlight off timeout */ +#define ADP8870_BLDIM 0x08 /* Backlight dim timeout */ +#define ADP8870_BLFR 0x09 /* Backlight fade in and out rates */ +#define ADP8870_BLMX1 0x0A /* Backlight (Brightness Level 1-daylight) maximum current */ +#define ADP8870_BLDM1 0x0B /* Backlight (Brightness Level 1-daylight) dim current */ +#define ADP8870_BLMX2 0x0C /* Backlight (Brightness Level 2-bright) maximum current */ +#define ADP8870_BLDM2 0x0D /* Backlight (Brightness Level 2-bright) dim current */ +#define ADP8870_BLMX3 0x0E /* Backlight (Brightness Level 3-office) maximum current */ +#define ADP8870_BLDM3 0x0F /* Backlight (Brightness Level 3-office) dim current */ +#define ADP8870_BLMX4 0x10 /* Backlight (Brightness Level 4-indoor) maximum current */ +#define ADP8870_BLDM4 0x11 /* Backlight (Brightness Level 4-indoor) dim current */ +#define ADP8870_BLMX5 0x12 /* Backlight (Brightness Level 5-dark) maximum current */ +#define ADP8870_BLDM5 0x13 /* Backlight (Brightness Level 5-dark) dim current */ +#define ADP8870_ISCLAW 0x1A /* Independent sink current fade law register */ +#define ADP8870_ISCC 0x1B /* Independent sink current control register */ +#define ADP8870_ISCT1 0x1C /* Independent Sink Current Timer Register LED[7:5] */ +#define ADP8870_ISCT2 0x1D /* Independent Sink Current Timer Register LED[4:1] */ +#define ADP8870_ISCF 0x1E /* Independent sink current fade register */ +#define ADP8870_ISC1 0x1F /* Independent Sink Current LED1 */ +#define ADP8870_ISC2 0x20 /* Independent Sink Current LED2 */ +#define ADP8870_ISC3 0x21 /* Independent Sink Current LED3 */ +#define ADP8870_ISC4 0x22 /* Independent Sink Current LED4 */ +#define ADP8870_ISC5 0x23 /* Independent Sink Current LED5 */ +#define ADP8870_ISC6 0x24 /* Independent Sink Current LED6 */ +#define ADP8870_ISC7 0x25 /* Independent Sink Current LED7 (Brightness Level 1-daylight) */ +#define ADP8870_ISC7_L2 0x26 /* Independent Sink Current LED7 (Brightness Level 2-bright) */ +#define ADP8870_ISC7_L3 0x27 /* Independent Sink Current LED7 (Brightness Level 3-office) */ +#define ADP8870_ISC7_L4 0x28 /* Independent Sink Current LED7 (Brightness Level 4-indoor) */ +#define ADP8870_ISC7_L5 0x29 /* Independent Sink Current LED7 (Brightness Level 5-dark) */ +#define ADP8870_CMP_CTL 0x2D /* ALS Comparator Control Register */ +#define ADP8870_ALS1_EN 0x2E /* Main ALS comparator level enable */ +#define ADP8870_ALS2_EN 0x2F /* Second ALS comparator level enable */ +#define ADP8870_ALS1_STAT 0x30 /* Main ALS Comparator Status Register */ +#define ADP8870_ALS2_STAT 0x31 /* Second ALS Comparator Status Register */ +#define ADP8870_L2TRP 0x32 /* L2 comparator reference */ +#define ADP8870_L2HYS 0x33 /* L2 hysteresis */ +#define ADP8870_L3TRP 0x34 /* L3 comparator reference */ +#define ADP8870_L3HYS 0x35 /* L3 hysteresis */ +#define ADP8870_L4TRP 0x36 /* L4 comparator reference */ +#define ADP8870_L4HYS 0x37 /* L4 hysteresis */ +#define ADP8870_L5TRP 0x38 /* L5 comparator reference */ +#define ADP8870_L5HYS 0x39 /* L5 hysteresis */ +#define ADP8870_PH1LEVL 0x40 /* First phototransistor ambient light level-low byte register */ +#define ADP8870_PH1LEVH 0x41 /* First phototransistor ambient light level-high byte register */ +#define ADP8870_PH2LEVL 0x42 /* Second phototransistor ambient light level-low byte register */ +#define ADP8870_PH2LEVH 0x43 /* Second phototransistor ambient light level-high byte register */ + +#define ADP8870_MANUFID 0x3 /* Analog Devices AD8870 Manufacturer and device ID */ +#define ADP8870_DEVID(x) ((x) & 0xF) +#define ADP8870_MANID(x) ((x) >> 4) + +/* MDCR Device mode and status */ +#define D7ALSEN (1 << 7) +#define INT_CFG (1 << 6) +#define NSTBY (1 << 5) +#define DIM_EN (1 << 4) +#define GDWN_DIS (1 << 3) +#define SIS_EN (1 << 2) +#define CMP_AUTOEN (1 << 1) +#define BLEN (1 << 0) + +/* ADP8870_ALS1_EN Main ALS comparator level enable */ +#define L5_EN (1 << 3) +#define L4_EN (1 << 2) +#define L3_EN (1 << 1) +#define L2_EN (1 << 0) + +#define CFGR_BLV_SHIFT 3 +#define CFGR_BLV_MASK 0x7 +#define ADP8870_FLAG_LED_MASK 0xFF + +#define FADE_VAL(in, out) ((0xF & (in)) | ((0xF & (out)) << 4)) +#define BL_CFGR_VAL(law, blv) ((((blv) & CFGR_BLV_MASK) << CFGR_BLV_SHIFT) | ((0x3 & (law)) << 1)) +#define ALS_CMPR_CFG_VAL(filt) ((0x7 & (filt)) << 1) + +struct adp8870_bl { + struct i2c_client *client; + struct backlight_device *bl; + struct adp8870_led *led; + struct adp8870_backlight_platform_data *pdata; + struct mutex lock; + unsigned long cached_daylight_max; + int id; + int revid; + int current_brightness; +}; + +struct adp8870_led { + struct led_classdev cdev; + struct work_struct work; + struct i2c_client *client; + enum led_brightness new_brightness; + int id; + int flags; +}; + +static int adp8870_read(struct i2c_client *client, int reg, uint8_t *val) +{ + int ret; + + ret = i2c_smbus_read_byte_data(client, reg); + if (ret < 0) { + dev_err(&client->dev, "failed reading at 0x%02x\n", reg); + return ret; + } + + *val = ret; + return 0; +} + + +static int adp8870_write(struct i2c_client *client, u8 reg, u8 val) +{ + int ret = i2c_smbus_write_byte_data(client, reg, val); + if (ret) + dev_err(&client->dev, "failed to write\n"); + + return ret; +} + +static int adp8870_set_bits(struct i2c_client *client, int reg, uint8_t bit_mask) +{ + struct adp8870_bl *data = i2c_get_clientdata(client); + uint8_t reg_val; + int ret; + + mutex_lock(&data->lock); + + ret = adp8870_read(client, reg, ®_val); + + if (!ret && ((reg_val & bit_mask) == 0)) { + reg_val |= bit_mask; + ret = adp8870_write(client, reg, reg_val); + } + + mutex_unlock(&data->lock); + return ret; +} + +static int adp8870_clr_bits(struct i2c_client *client, int reg, uint8_t bit_mask) +{ + struct adp8870_bl *data = i2c_get_clientdata(client); + uint8_t reg_val; + int ret; + + mutex_lock(&data->lock); + + ret = adp8870_read(client, reg, ®_val); + + if (!ret && (reg_val & bit_mask)) { + reg_val &= ~bit_mask; + ret = adp8870_write(client, reg, reg_val); + } + + mutex_unlock(&data->lock); + return ret; +} + +/* + * Independent sink / LED + */ +#if defined(ADP8870_USE_LEDS) +static void adp8870_led_work(struct work_struct *work) +{ + struct adp8870_led *led = container_of(work, struct adp8870_led, work); + adp8870_write(led->client, ADP8870_ISC1 + led->id - 1, + led->new_brightness >> 1); +} + +static void adp8870_led_set(struct led_classdev *led_cdev, + enum led_brightness value) +{ + struct adp8870_led *led; + + led = container_of(led_cdev, struct adp8870_led, cdev); + led->new_brightness = value; + /* + * Use workqueue for IO since I2C operations can sleep. + */ + schedule_work(&led->work); +} + +static int adp8870_led_setup(struct adp8870_led *led) +{ + struct i2c_client *client = led->client; + int ret = 0; + + ret = adp8870_write(client, ADP8870_ISC1 + led->id - 1, 0); + if (ret) + return ret; + + ret = adp8870_set_bits(client, ADP8870_ISCC, 1 << (led->id - 1)); + if (ret) + return ret; + + if (led->id > 4) + ret = adp8870_set_bits(client, ADP8870_ISCT1, + (led->flags & 0x3) << ((led->id - 5) * 2)); + else + ret = adp8870_set_bits(client, ADP8870_ISCT2, + (led->flags & 0x3) << ((led->id - 1) * 2)); + + return ret; +} + +static int __devinit adp8870_led_probe(struct i2c_client *client) +{ + struct adp8870_backlight_platform_data *pdata = + client->dev.platform_data; + struct adp8870_bl *data = i2c_get_clientdata(client); + struct adp8870_led *led, *led_dat; + struct led_info *cur_led; + int ret, i; + + + led = kcalloc(pdata->num_leds, sizeof(*led), GFP_KERNEL); + if (led == NULL) { + dev_err(&client->dev, "failed to alloc memory\n"); + return -ENOMEM; + } + + ret = adp8870_write(client, ADP8870_ISCLAW, pdata->led_fade_law); + if (ret) + goto err_free; + + ret = adp8870_write(client, ADP8870_ISCT1, + (pdata->led_on_time & 0x3) << 6); + if (ret) + goto err_free; + + ret = adp8870_write(client, ADP8870_ISCF, + FADE_VAL(pdata->led_fade_in, pdata->led_fade_out)); + if (ret) + goto err_free; + + for (i = 0; i < pdata->num_leds; ++i) { + cur_led = &pdata->leds[i]; + led_dat = &led[i]; + + led_dat->id = cur_led->flags & ADP8870_FLAG_LED_MASK; + + if (led_dat->id > 7 || led_dat->id < 1) { + dev_err(&client->dev, "Invalid LED ID %d\n", + led_dat->id); + goto err; + } + + if (pdata->bl_led_assign & (1 << (led_dat->id - 1))) { + dev_err(&client->dev, "LED %d used by Backlight\n", + led_dat->id); + goto err; + } + + led_dat->cdev.name = cur_led->name; + led_dat->cdev.default_trigger = cur_led->default_trigger; + led_dat->cdev.brightness_set = adp8870_led_set; + led_dat->cdev.brightness = LED_OFF; + led_dat->flags = cur_led->flags >> FLAG_OFFT_SHIFT; + led_dat->client = client; + led_dat->new_brightness = LED_OFF; + INIT_WORK(&led_dat->work, adp8870_led_work); + + ret = led_classdev_register(&client->dev, &led_dat->cdev); + if (ret) { + dev_err(&client->dev, "failed to register LED %d\n", + led_dat->id); + goto err; + } + + ret = adp8870_led_setup(led_dat); + if (ret) { + dev_err(&client->dev, "failed to write\n"); + i++; + goto err; + } + } + + data->led = led; + + return 0; + + err: + for (i = i - 1; i >= 0; --i) { + led_classdev_unregister(&led[i].cdev); + cancel_work_sync(&led[i].work); + } + + err_free: + kfree(led); + + return ret; +} + +static int __devexit adp8870_led_remove(struct i2c_client *client) +{ + struct adp8870_backlight_platform_data *pdata = + client->dev.platform_data; + struct adp8870_bl *data = i2c_get_clientdata(client); + int i; + + for (i = 0; i < pdata->num_leds; i++) { + led_classdev_unregister(&data->led[i].cdev); + cancel_work_sync(&data->led[i].work); + } + + kfree(data->led); + return 0; +} +#else +static int __devinit adp8870_led_probe(struct i2c_client *client) +{ + return 0; +} + +static int __devexit adp8870_led_remove(struct i2c_client *client) +{ + return 0; +} +#endif + +static int adp8870_bl_set(struct backlight_device *bl, int brightness) +{ + struct adp8870_bl *data = bl_get_data(bl); + struct i2c_client *client = data->client; + int ret = 0; + + if (data->pdata->en_ambl_sens) { + if ((brightness > 0) && (brightness < ADP8870_MAX_BRIGHTNESS)) { + /* Disable Ambient Light auto adjust */ + ret = adp8870_clr_bits(client, ADP8870_MDCR, + CMP_AUTOEN); + if (ret) + return ret; + ret = adp8870_write(client, ADP8870_BLMX1, brightness); + if (ret) + return ret; + } else { + /* + * MAX_BRIGHTNESS -> Enable Ambient Light auto adjust + * restore daylight l1 sysfs brightness + */ + ret = adp8870_write(client, ADP8870_BLMX1, + data->cached_daylight_max); + if (ret) + return ret; + + ret = adp8870_set_bits(client, ADP8870_MDCR, + CMP_AUTOEN); + if (ret) + return ret; + } + } else { + ret = adp8870_write(client, ADP8870_BLMX1, brightness); + if (ret) + return ret; + } + + if (data->current_brightness && brightness == 0) + ret = adp8870_set_bits(client, + ADP8870_MDCR, DIM_EN); + else if (data->current_brightness == 0 && brightness) + ret = adp8870_clr_bits(client, + ADP8870_MDCR, DIM_EN); + + if (!ret) + data->current_brightness = brightness; + + return ret; +} + +static int adp8870_bl_update_status(struct backlight_device *bl) +{ + int brightness = bl->props.brightness; + if (bl->props.power != FB_BLANK_UNBLANK) + brightness = 0; + + if (bl->props.fb_blank != FB_BLANK_UNBLANK) + brightness = 0; + + return adp8870_bl_set(bl, brightness); +} + +static int adp8870_bl_get_brightness(struct backlight_device *bl) +{ + struct adp8870_bl *data = bl_get_data(bl); + + return data->current_brightness; +} + +static const struct backlight_ops adp8870_bl_ops = { + .update_status = adp8870_bl_update_status, + .get_brightness = adp8870_bl_get_brightness, +}; + +static int adp8870_bl_setup(struct backlight_device *bl) +{ + struct adp8870_bl *data = bl_get_data(bl); + struct i2c_client *client = data->client; + struct adp8870_backlight_platform_data *pdata = data->pdata; + int ret = 0; + + ret = adp8870_write(client, ADP8870_BLSEL, ~pdata->bl_led_assign); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_PWMLED, pdata->pwm_assign); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_BLMX1, pdata->l1_daylight_max); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_BLDM1, pdata->l1_daylight_dim); + if (ret) + return ret; + + if (pdata->en_ambl_sens) { + data->cached_daylight_max = pdata->l1_daylight_max; + ret = adp8870_write(client, ADP8870_BLMX2, + pdata->l2_bright_max); + if (ret) + return ret; + ret = adp8870_write(client, ADP8870_BLDM2, + pdata->l2_bright_dim); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_BLMX3, + pdata->l3_office_max); + if (ret) + return ret; + ret = adp8870_write(client, ADP8870_BLDM3, + pdata->l3_office_dim); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_BLMX4, + pdata->l4_indoor_max); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_BLDM4, + pdata->l4_indor_dim); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_BLMX5, + pdata->l5_dark_max); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_BLDM5, + pdata->l5_dark_dim); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_L2TRP, pdata->l2_trip); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_L2HYS, pdata->l2_hyst); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_L3TRP, pdata->l3_trip); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_L3HYS, pdata->l3_hyst); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_L4TRP, pdata->l4_trip); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_L4HYS, pdata->l4_hyst); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_L5TRP, pdata->l5_trip); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_L5HYS, pdata->l5_hyst); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_ALS1_EN, L5_EN | L4_EN | + L3_EN | L2_EN); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_CMP_CTL, + ALS_CMPR_CFG_VAL(pdata->abml_filt)); + if (ret) + return ret; + } + + ret = adp8870_write(client, ADP8870_CFGR, + BL_CFGR_VAL(pdata->bl_fade_law, 0)); + if (ret) + return ret; + + ret = adp8870_write(client, ADP8870_BLFR, FADE_VAL(pdata->bl_fade_in, + pdata->bl_fade_out)); + if (ret) + return ret; + /* + * ADP8870 Rev0 requires GDWN_DIS bit set + */ + + ret = adp8870_set_bits(client, ADP8870_MDCR, BLEN | DIM_EN | NSTBY | + (data->revid == 0 ? GDWN_DIS : 0)); + + return ret; +} + +static ssize_t adp8870_show(struct device *dev, char *buf, int reg) +{ + struct adp8870_bl *data = dev_get_drvdata(dev); + int error; + uint8_t reg_val; + + mutex_lock(&data->lock); + error = adp8870_read(data->client, reg, ®_val); + mutex_unlock(&data->lock); + + if (error < 0) + return error; + + return sprintf(buf, "%u\n", reg_val); +} + +static ssize_t adp8870_store(struct device *dev, const char *buf, + size_t count, int reg) +{ + struct adp8870_bl *data = dev_get_drvdata(dev); + unsigned long val; + int ret; + + ret = strict_strtoul(buf, 10, &val); + if (ret) + return ret; + + mutex_lock(&data->lock); + adp8870_write(data->client, reg, val); + mutex_unlock(&data->lock); + + return count; +} + +static ssize_t adp8870_bl_l5_dark_max_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return adp8870_show(dev, buf, ADP8870_BLMX5); +} + +static ssize_t adp8870_bl_l5_dark_max_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + return adp8870_store(dev, buf, count, ADP8870_BLMX5); +} +static DEVICE_ATTR(l5_dark_max, 0664, adp8870_bl_l5_dark_max_show, + adp8870_bl_l5_dark_max_store); + + +static ssize_t adp8870_bl_l4_indoor_max_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return adp8870_show(dev, buf, ADP8870_BLMX4); +} + +static ssize_t adp8870_bl_l4_indoor_max_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + return adp8870_store(dev, buf, count, ADP8870_BLMX4); +} +static DEVICE_ATTR(l4_indoor_max, 0664, adp8870_bl_l4_indoor_max_show, + adp8870_bl_l4_indoor_max_store); + + +static ssize_t adp8870_bl_l3_office_max_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return adp8870_show(dev, buf, ADP8870_BLMX3); +} + +static ssize_t adp8870_bl_l3_office_max_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + return adp8870_store(dev, buf, count, ADP8870_BLMX3); +} + +static DEVICE_ATTR(l3_office_max, 0664, adp8870_bl_l3_office_max_show, + adp8870_bl_l3_office_max_store); + +static ssize_t adp8870_bl_l2_bright_max_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return adp8870_show(dev, buf, ADP8870_BLMX2); +} + +static ssize_t adp8870_bl_l2_bright_max_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + return adp8870_store(dev, buf, count, ADP8870_BLMX2); +} +static DEVICE_ATTR(l2_bright_max, 0664, adp8870_bl_l2_bright_max_show, + adp8870_bl_l2_bright_max_store); + +static ssize_t adp8870_bl_l1_daylight_max_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return adp8870_show(dev, buf, ADP8870_BLMX1); +} + +static ssize_t adp8870_bl_l1_daylight_max_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct adp8870_bl *data = dev_get_drvdata(dev); + int ret = strict_strtoul(buf, 10, &data->cached_daylight_max); + if (ret) + return ret; + + return adp8870_store(dev, buf, count, ADP8870_BLMX1); +} +static DEVICE_ATTR(l1_daylight_max, 0664, adp8870_bl_l1_daylight_max_show, + adp8870_bl_l1_daylight_max_store); + +static ssize_t adp8870_bl_l5_dark_dim_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return adp8870_show(dev, buf, ADP8870_BLDM5); +} + +static ssize_t adp8870_bl_l5_dark_dim_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + return adp8870_store(dev, buf, count, ADP8870_BLDM5); +} +static DEVICE_ATTR(l5_dark_dim, 0664, adp8870_bl_l5_dark_dim_show, + adp8870_bl_l5_dark_dim_store); + +static ssize_t adp8870_bl_l4_indoor_dim_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return adp8870_show(dev, buf, ADP8870_BLDM4); +} + +static ssize_t adp8870_bl_l4_indoor_dim_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + return adp8870_store(dev, buf, count, ADP8870_BLDM4); +} +static DEVICE_ATTR(l4_indoor_dim, 0664, adp8870_bl_l4_indoor_dim_show, + adp8870_bl_l4_indoor_dim_store); + + +static ssize_t adp8870_bl_l3_office_dim_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return adp8870_show(dev, buf, ADP8870_BLDM3); +} + +static ssize_t adp8870_bl_l3_office_dim_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + return adp8870_store(dev, buf, count, ADP8870_BLDM3); +} +static DEVICE_ATTR(l3_office_dim, 0664, adp8870_bl_l3_office_dim_show, + adp8870_bl_l3_office_dim_store); + +static ssize_t adp8870_bl_l2_bright_dim_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return adp8870_show(dev, buf, ADP8870_BLDM2); +} + +static ssize_t adp8870_bl_l2_bright_dim_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + return adp8870_store(dev, buf, count, ADP8870_BLDM2); +} +static DEVICE_ATTR(l2_bright_dim, 0664, adp8870_bl_l2_bright_dim_show, + adp8870_bl_l2_bright_dim_store); + +static ssize_t adp8870_bl_l1_daylight_dim_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + return adp8870_show(dev, buf, ADP8870_BLDM1); +} + +static ssize_t adp8870_bl_l1_daylight_dim_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + return adp8870_store(dev, buf, count, ADP8870_BLDM1); +} +static DEVICE_ATTR(l1_daylight_dim, 0664, adp8870_bl_l1_daylight_dim_show, + adp8870_bl_l1_daylight_dim_store); + +#ifdef ADP8870_EXT_FEATURES +static ssize_t adp8870_bl_ambient_light_level_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct adp8870_bl *data = dev_get_drvdata(dev); + int error; + uint8_t reg_val; + uint16_t ret_val; + + mutex_lock(&data->lock); + error = adp8870_read(data->client, ADP8870_PH1LEVL, ®_val); + if (error < 0) { + mutex_unlock(&data->lock); + return error; + } + ret_val = reg_val; + error = adp8870_read(data->client, ADP8870_PH1LEVH, ®_val); + mutex_unlock(&data->lock); + + if (error < 0) + return error; + + /* Return 13-bit conversion value for the first light sensor */ + ret_val += (reg_val & 0x1F) << 8; + + return sprintf(buf, "%u\n", ret_val); +} +static DEVICE_ATTR(ambient_light_level, 0444, + adp8870_bl_ambient_light_level_show, NULL); + +static ssize_t adp8870_bl_ambient_light_zone_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct adp8870_bl *data = dev_get_drvdata(dev); + int error; + uint8_t reg_val; + + mutex_lock(&data->lock); + error = adp8870_read(data->client, ADP8870_CFGR, ®_val); + mutex_unlock(&data->lock); + + if (error < 0) + return error; + + return sprintf(buf, "%u\n", + ((reg_val >> CFGR_BLV_SHIFT) & CFGR_BLV_MASK) + 1); +} + +static ssize_t adp8870_bl_ambient_light_zone_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct adp8870_bl *data = dev_get_drvdata(dev); + unsigned long val; + uint8_t reg_val; + int ret; + + ret = strict_strtoul(buf, 10, &val); + if (ret) + return ret; + + if (val == 0) { + /* Enable automatic ambient light sensing */ + adp8870_set_bits(data->client, ADP8870_MDCR, CMP_AUTOEN); + } else if ((val > 0) && (val < 6)) { + /* Disable automatic ambient light sensing */ + adp8870_clr_bits(data->client, ADP8870_MDCR, CMP_AUTOEN); + + /* Set user supplied ambient light zone */ + mutex_lock(&data->lock); + adp8870_read(data->client, ADP8870_CFGR, ®_val); + reg_val &= ~(CFGR_BLV_MASK << CFGR_BLV_SHIFT); + reg_val |= (val - 1) << CFGR_BLV_SHIFT; + adp8870_write(data->client, ADP8870_CFGR, reg_val); + mutex_unlock(&data->lock); + } + + return count; +} +static DEVICE_ATTR(ambient_light_zone, 0664, + adp8870_bl_ambient_light_zone_show, + adp8870_bl_ambient_light_zone_store); +#endif + +static struct attribute *adp8870_bl_attributes[] = { + &dev_attr_l5_dark_max.attr, + &dev_attr_l5_dark_dim.attr, + &dev_attr_l4_indoor_max.attr, + &dev_attr_l4_indoor_dim.attr, + &dev_attr_l3_office_max.attr, + &dev_attr_l3_office_dim.attr, + &dev_attr_l2_bright_max.attr, + &dev_attr_l2_bright_dim.attr, + &dev_attr_l1_daylight_max.attr, + &dev_attr_l1_daylight_dim.attr, +#ifdef ADP8870_EXT_FEATURES + &dev_attr_ambient_light_level.attr, + &dev_attr_ambient_light_zone.attr, +#endif + NULL +}; + +static const struct attribute_group adp8870_bl_attr_group = { + .attrs = adp8870_bl_attributes, +}; + +static int __devinit adp8870_probe(struct i2c_client *client, + const struct i2c_device_id *id) +{ + struct backlight_properties props; + struct backlight_device *bl; + struct adp8870_bl *data; + struct adp8870_backlight_platform_data *pdata = + client->dev.platform_data; + uint8_t reg_val; + int ret; + + if (!i2c_check_functionality(client->adapter, + I2C_FUNC_SMBUS_BYTE_DATA)) { + dev_err(&client->dev, "SMBUS Byte Data not Supported\n"); + return -EIO; + } + + if (!pdata) { + dev_err(&client->dev, "no platform data?\n"); + return -EINVAL; + } + + ret = adp8870_read(client, ADP8870_MFDVID, ®_val); + if (ret < 0) + return -EIO; + + if (ADP8870_MANID(reg_val) != ADP8870_MANUFID) { + dev_err(&client->dev, "failed to probe\n"); + return -ENODEV; + } + + data = kzalloc(sizeof(*data), GFP_KERNEL); + if (data == NULL) + return -ENOMEM; + + data->revid = ADP8870_DEVID(reg_val); + data->client = client; + data->pdata = pdata; + data->id = id->driver_data; + data->current_brightness = 0; + i2c_set_clientdata(client, data); + + mutex_init(&data->lock); + + memset(&props, 0, sizeof(props)); + props.max_brightness = props.brightness = ADP8870_MAX_BRIGHTNESS; + bl = backlight_device_register(dev_driver_string(&client->dev), + &client->dev, data, &adp8870_bl_ops, &props); + if (IS_ERR(bl)) { + dev_err(&client->dev, "failed to register backlight\n"); + ret = PTR_ERR(bl); + goto out2; + } + + data->bl = bl; + + if (pdata->en_ambl_sens) + ret = sysfs_create_group(&bl->dev.kobj, + &adp8870_bl_attr_group); + + if (ret) { + dev_err(&client->dev, "failed to register sysfs\n"); + goto out1; + } + + ret = adp8870_bl_setup(bl); + if (ret) { + ret = -EIO; + goto out; + } + + backlight_update_status(bl); + + dev_info(&client->dev, "Rev.%d Backlight\n", data->revid); + + if (pdata->num_leds) + adp8870_led_probe(client); + + return 0; + +out: + if (data->pdata->en_ambl_sens) + sysfs_remove_group(&data->bl->dev.kobj, + &adp8870_bl_attr_group); +out1: + backlight_device_unregister(bl); +out2: + i2c_set_clientdata(client, NULL); + kfree(data); + + return ret; +} + +static int __devexit adp8870_remove(struct i2c_client *client) +{ + struct adp8870_bl *data = i2c_get_clientdata(client); + + adp8870_clr_bits(client, ADP8870_MDCR, NSTBY); + + if (data->led) + adp8870_led_remove(client); + + if (data->pdata->en_ambl_sens) + sysfs_remove_group(&data->bl->dev.kobj, + &adp8870_bl_attr_group); + + backlight_device_unregister(data->bl); + i2c_set_clientdata(client, NULL); + kfree(data); + + return 0; +} + +#ifdef CONFIG_PM +static int adp8870_i2c_suspend(struct i2c_client *client, pm_message_t message) +{ + adp8870_clr_bits(client, ADP8870_MDCR, NSTBY); + + return 0; +} + +static int adp8870_i2c_resume(struct i2c_client *client) +{ + adp8870_set_bits(client, ADP8870_MDCR, NSTBY); + + return 0; +} +#else +#define adp8870_i2c_suspend NULL +#define adp8870_i2c_resume NULL +#endif + +static const struct i2c_device_id adp8870_id[] = { + { "adp8870", 0 }, + { } +}; +MODULE_DEVICE_TABLE(i2c, adp8870_id); + +static struct i2c_driver adp8870_driver = { + .driver = { + .name = KBUILD_MODNAME, + }, + .probe = adp8870_probe, + .remove = __devexit_p(adp8870_remove), + .suspend = adp8870_i2c_suspend, + .resume = adp8870_i2c_resume, + .id_table = adp8870_id, +}; + +static int __init adp8870_init(void) +{ + return i2c_add_driver(&adp8870_driver); +} +module_init(adp8870_init); + +static void __exit adp8870_exit(void) +{ + i2c_del_driver(&adp8870_driver); +} +module_exit(adp8870_exit); + +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Michael Hennerich "); +MODULE_DESCRIPTION("ADP8870 Backlight driver"); +MODULE_ALIAS("platform:adp8870-backlight"); diff --git a/include/linux/i2c/adp8870.h b/include/linux/i2c/adp8870.h new file mode 100644 index 000000000000..624dceccbd5b --- /dev/null +++ b/include/linux/i2c/adp8870.h @@ -0,0 +1,153 @@ +/* + * Definitions and platform data for Analog Devices + * Backlight drivers ADP8870 + * + * Copyright 2009-2010 Analog Devices Inc. + * + * Licensed under the GPL-2 or later. + */ + +#ifndef __LINUX_I2C_ADP8870_H +#define __LINUX_I2C_ADP8870_H + +#define ID_ADP8870 8870 + +#define ADP8870_MAX_BRIGHTNESS 0x7F +#define FLAG_OFFT_SHIFT 8 + +/* + * LEDs subdevice platform data + */ + +#define ADP8870_LED_DIS_BLINK (0 << FLAG_OFFT_SHIFT) +#define ADP8870_LED_OFFT_600ms (1 << FLAG_OFFT_SHIFT) +#define ADP8870_LED_OFFT_1200ms (2 << FLAG_OFFT_SHIFT) +#define ADP8870_LED_OFFT_1800ms (3 << FLAG_OFFT_SHIFT) + +#define ADP8870_LED_ONT_200ms 0 +#define ADP8870_LED_ONT_600ms 1 +#define ADP8870_LED_ONT_800ms 2 +#define ADP8870_LED_ONT_1200ms 3 + +#define ADP8870_LED_D7 (7) +#define ADP8870_LED_D6 (6) +#define ADP8870_LED_D5 (5) +#define ADP8870_LED_D4 (4) +#define ADP8870_LED_D3 (3) +#define ADP8870_LED_D2 (2) +#define ADP8870_LED_D1 (1) + +/* + * Backlight subdevice platform data + */ + +#define ADP8870_BL_D7 (1 << 6) +#define ADP8870_BL_D6 (1 << 5) +#define ADP8870_BL_D5 (1 << 4) +#define ADP8870_BL_D4 (1 << 3) +#define ADP8870_BL_D3 (1 << 2) +#define ADP8870_BL_D2 (1 << 1) +#define ADP8870_BL_D1 (1 << 0) + +#define ADP8870_FADE_T_DIS 0 /* Fade Timer Disabled */ +#define ADP8870_FADE_T_300ms 1 /* 0.3 Sec */ +#define ADP8870_FADE_T_600ms 2 +#define ADP8870_FADE_T_900ms 3 +#define ADP8870_FADE_T_1200ms 4 +#define ADP8870_FADE_T_1500ms 5 +#define ADP8870_FADE_T_1800ms 6 +#define ADP8870_FADE_T_2100ms 7 +#define ADP8870_FADE_T_2400ms 8 +#define ADP8870_FADE_T_2700ms 9 +#define ADP8870_FADE_T_3000ms 10 +#define ADP8870_FADE_T_3500ms 11 +#define ADP8870_FADE_T_4000ms 12 +#define ADP8870_FADE_T_4500ms 13 +#define ADP8870_FADE_T_5000ms 14 +#define ADP8870_FADE_T_5500ms 15 /* 5.5 Sec */ + +#define ADP8870_FADE_LAW_LINEAR 0 +#define ADP8870_FADE_LAW_SQUARE 1 +#define ADP8870_FADE_LAW_CUBIC1 2 +#define ADP8870_FADE_LAW_CUBIC2 3 + +#define ADP8870_BL_AMBL_FILT_80ms 0 /* Light sensor filter time */ +#define ADP8870_BL_AMBL_FILT_160ms 1 +#define ADP8870_BL_AMBL_FILT_320ms 2 +#define ADP8870_BL_AMBL_FILT_640ms 3 +#define ADP8870_BL_AMBL_FILT_1280ms 4 +#define ADP8870_BL_AMBL_FILT_2560ms 5 +#define ADP8870_BL_AMBL_FILT_5120ms 6 +#define ADP8870_BL_AMBL_FILT_10240ms 7 /* 10.24 sec */ + +/* + * Blacklight current 0..30mA + */ +#define ADP8870_BL_CUR_mA(I) ((I * 127) / 30) + +/* + * L2 comparator current 0..1106uA + */ +#define ADP8870_L2_COMP_CURR_uA(I) ((I * 255) / 1106) + +/* + * L3 comparator current 0..551uA + */ +#define ADP8870_L3_COMP_CURR_uA(I) ((I * 255) / 551) + +/* + * L4 comparator current 0..275uA + */ +#define ADP8870_L4_COMP_CURR_uA(I) ((I * 255) / 275) + +/* + * L5 comparator current 0..138uA + */ +#define ADP8870_L5_COMP_CURR_uA(I) ((I * 255) / 138) + +struct adp8870_backlight_platform_data { + u8 bl_led_assign; /* 1 = Backlight 0 = Individual LED */ + u8 pwm_assign; /* 1 = Enables PWM mode */ + + u8 bl_fade_in; /* Backlight Fade-In Timer */ + u8 bl_fade_out; /* Backlight Fade-Out Timer */ + u8 bl_fade_law; /* fade-on/fade-off transfer characteristic */ + + u8 en_ambl_sens; /* 1 = enable ambient light sensor */ + u8 abml_filt; /* Light sensor filter time */ + + u8 l1_daylight_max; /* use BL_CUR_mA(I) 0 <= I <= 30 mA */ + u8 l1_daylight_dim; /* typ = 0, use BL_CUR_mA(I) 0 <= I <= 30 mA */ + u8 l2_bright_max; /* use BL_CUR_mA(I) 0 <= I <= 30 mA */ + u8 l2_bright_dim; /* typ = 0, use BL_CUR_mA(I) 0 <= I <= 30 mA */ + u8 l3_office_max; /* use BL_CUR_mA(I) 0 <= I <= 30 mA */ + u8 l3_office_dim; /* typ = 0, use BL_CUR_mA(I) 0 <= I <= 30 mA */ + u8 l4_indoor_max; /* use BL_CUR_mA(I) 0 <= I <= 30 mA */ + u8 l4_indor_dim; /* typ = 0, use BL_CUR_mA(I) 0 <= I <= 30 mA */ + u8 l5_dark_max; /* use BL_CUR_mA(I) 0 <= I <= 30 mA */ + u8 l5_dark_dim; /* typ = 0, use BL_CUR_mA(I) 0 <= I <= 30 mA */ + + u8 l2_trip; /* use L2_COMP_CURR_uA(I) 0 <= I <= 1106 uA */ + u8 l2_hyst; /* use L2_COMP_CURR_uA(I) 0 <= I <= 1106 uA */ + u8 l3_trip; /* use L3_COMP_CURR_uA(I) 0 <= I <= 551 uA */ + u8 l3_hyst; /* use L3_COMP_CURR_uA(I) 0 <= I <= 551 uA */ + u8 l4_trip; /* use L4_COMP_CURR_uA(I) 0 <= I <= 275 uA */ + u8 l4_hyst; /* use L4_COMP_CURR_uA(I) 0 <= I <= 275 uA */ + u8 l5_trip; /* use L5_COMP_CURR_uA(I) 0 <= I <= 138 uA */ + u8 l5_hyst; /* use L6_COMP_CURR_uA(I) 0 <= I <= 138 uA */ + + /** + * Independent Current Sinks / LEDS + * Sinks not assigned to the Backlight can be exposed to + * user space using the LEDS CLASS interface + */ + + int num_leds; + struct led_info *leds; + u8 led_fade_in; /* LED Fade-In Timer */ + u8 led_fade_out; /* LED Fade-Out Timer */ + u8 led_fade_law; /* fade-on/fade-off transfer characteristic */ + u8 led_on_time; +}; + +#endif /* __LINUX_I2C_ADP8870_H */ -- cgit v1.2.3 From a433658c30974fc87ba3ff52d7e4e6299762aa3d Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Wed, 15 Jun 2011 15:08:13 -0700 Subject: vmscan,memcg: memcg aware swap token Currently, memcg reclaim can disable swap token even if the swap token mm doesn't belong in its memory cgroup. It's slightly risky. If an admin creates very small mem-cgroup and silly guy runs contentious heavy memory pressure workload, every tasks are going to lose swap token and then system may become unresponsive. That's bad. This patch adds 'memcg' parameter into disable_swap_token(). and if the parameter doesn't match swap token, VM doesn't disable it. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: KOSAKI Motohiro Reviewed-by: KAMEZAWA Hiroyuki Reviewed-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 6 ++++ include/linux/swap.h | 8 ++--- mm/memcontrol.c | 16 ++++----- mm/thrash.c | 87 ++++++++++++++++++++++++++++++++++++---------- mm/vmscan.c | 4 +-- 5 files changed, 85 insertions(+), 36 deletions(-) (limited to 'include') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 9724a38ee69d..50940da6adf3 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -84,6 +84,7 @@ int task_in_mem_cgroup(struct task_struct *task, const struct mem_cgroup *mem); extern struct mem_cgroup *try_get_mem_cgroup_from_page(struct page *page); extern struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p); +extern struct mem_cgroup *try_get_mem_cgroup_from_mm(struct mm_struct *mm); static inline int mm_match_cgroup(const struct mm_struct *mm, const struct mem_cgroup *cgroup) @@ -246,6 +247,11 @@ static inline struct mem_cgroup *try_get_mem_cgroup_from_page(struct page *page) return NULL; } +static inline struct mem_cgroup *try_get_mem_cgroup_from_mm(struct mm_struct *mm) +{ + return NULL; +} + static inline int mm_match_cgroup(struct mm_struct *mm, struct mem_cgroup *mem) { return 1; diff --git a/include/linux/swap.h b/include/linux/swap.h index 384eb5fe530b..e70564647039 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -358,6 +358,7 @@ struct backing_dev_info; extern struct mm_struct *swap_token_mm; extern void grab_swap_token(struct mm_struct *); extern void __put_swap_token(struct mm_struct *); +extern void disable_swap_token(struct mem_cgroup *memcg); static inline int has_swap_token(struct mm_struct *mm) { @@ -370,11 +371,6 @@ static inline void put_swap_token(struct mm_struct *mm) __put_swap_token(mm); } -static inline void disable_swap_token(void) -{ - put_swap_token(swap_token_mm); -} - #ifdef CONFIG_CGROUP_MEM_RES_CTLR extern void mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent, bool swapout); @@ -500,7 +496,7 @@ static inline int has_swap_token(struct mm_struct *mm) return 0; } -static inline void disable_swap_token(void) +static inline void disable_swap_token(struct mem_cgroup *memcg) { } diff --git a/mm/memcontrol.c b/mm/memcontrol.c index bd9052a5d3ad..e37c44d88d46 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -735,7 +735,7 @@ struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p) struct mem_cgroup, css); } -static struct mem_cgroup *try_get_mem_cgroup_from_mm(struct mm_struct *mm) +struct mem_cgroup *try_get_mem_cgroup_from_mm(struct mm_struct *mm) { struct mem_cgroup *mem = NULL; @@ -5414,18 +5414,16 @@ static void mem_cgroup_move_task(struct cgroup_subsys *ss, struct cgroup *old_cont, struct task_struct *p) { - struct mm_struct *mm; + struct mm_struct *mm = get_task_mm(p); - if (!mc.to) - /* no need to move charge */ - return; - - mm = get_task_mm(p); if (mm) { - mem_cgroup_move_charge(mm); + if (mc.to) + mem_cgroup_move_charge(mm); + put_swap_token(mm); mmput(mm); } - mem_cgroup_clear_mc(); + if (mc.to) + mem_cgroup_clear_mc(); } #else /* !CONFIG_MMU */ static int mem_cgroup_can_attach(struct cgroup_subsys *ss, diff --git a/mm/thrash.c b/mm/thrash.c index 2372d4ed5dd8..6cdf86511385 100644 --- a/mm/thrash.c +++ b/mm/thrash.c @@ -21,11 +21,31 @@ #include #include #include +#include static DEFINE_SPINLOCK(swap_token_lock); struct mm_struct *swap_token_mm; +struct mem_cgroup *swap_token_memcg; static unsigned int global_faults; +#ifdef CONFIG_CGROUP_MEM_RES_CTLR +static struct mem_cgroup *swap_token_memcg_from_mm(struct mm_struct *mm) +{ + struct mem_cgroup *memcg; + + memcg = try_get_mem_cgroup_from_mm(mm); + if (memcg) + css_put(mem_cgroup_css(memcg)); + + return memcg; +} +#else +static struct mem_cgroup *swap_token_memcg_from_mm(struct mm_struct *mm) +{ + return NULL; +} +#endif + void grab_swap_token(struct mm_struct *mm) { int current_interval; @@ -38,40 +58,69 @@ void grab_swap_token(struct mm_struct *mm) return; /* First come first served */ - if (swap_token_mm == NULL) { - mm->token_priority = mm->token_priority + 2; - swap_token_mm = mm; + if (!swap_token_mm) + goto replace_token; + + if (mm == swap_token_mm) { + mm->token_priority += 2; goto out; } - if (mm != swap_token_mm) { - if (current_interval < mm->last_interval) - mm->token_priority++; - else { - if (likely(mm->token_priority > 0)) - mm->token_priority--; - } - /* Check if we deserve the token */ - if (mm->token_priority > swap_token_mm->token_priority) { - mm->token_priority += 2; - swap_token_mm = mm; - } - } else { - /* Token holder came in again! */ - mm->token_priority += 2; + if (current_interval < mm->last_interval) + mm->token_priority++; + else { + if (likely(mm->token_priority > 0)) + mm->token_priority--; } + /* Check if we deserve the token */ + if (mm->token_priority > swap_token_mm->token_priority) + goto replace_token; + out: mm->faultstamp = global_faults; mm->last_interval = current_interval; spin_unlock(&swap_token_lock); + return; + +replace_token: + mm->token_priority += 2; + swap_token_mm = mm; + swap_token_memcg = swap_token_memcg_from_mm(mm); + goto out; } /* Called on process exit. */ void __put_swap_token(struct mm_struct *mm) { spin_lock(&swap_token_lock); - if (likely(mm == swap_token_mm)) + if (likely(mm == swap_token_mm)) { swap_token_mm = NULL; + swap_token_memcg = NULL; + } spin_unlock(&swap_token_lock); } + +static bool match_memcg(struct mem_cgroup *a, struct mem_cgroup *b) +{ + if (!a) + return true; + if (!b) + return true; + if (a == b) + return true; + return false; +} + +void disable_swap_token(struct mem_cgroup *memcg) +{ + /* memcg reclaim don't disable unrelated mm token. */ + if (match_memcg(memcg, swap_token_memcg)) { + spin_lock(&swap_token_lock); + if (match_memcg(memcg, swap_token_memcg)) { + swap_token_mm = NULL; + swap_token_memcg = NULL; + } + spin_unlock(&swap_token_lock); + } +} diff --git a/mm/vmscan.c b/mm/vmscan.c index faa0a088f9cc..dbe6ea321df4 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2081,7 +2081,7 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist, for (priority = DEF_PRIORITY; priority >= 0; priority--) { sc->nr_scanned = 0; if (!priority) - disable_swap_token(); + disable_swap_token(sc->mem_cgroup); total_scanned += shrink_zones(priority, zonelist, sc); /* * Don't shrink slabs when reclaiming memory from @@ -2407,7 +2407,7 @@ loop_again: /* The swap token gets in the way of swapout... */ if (!priority) - disable_swap_token(); + disable_swap_token(NULL); all_zones_ok = 1; balanced = 0; -- cgit v1.2.3 From 83cd81a34357a632509f7491eec81e62e71d65f7 Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Wed, 15 Jun 2011 15:08:14 -0700 Subject: vmscan: implement swap token trace This is useful for observing swap token activity. example output: zsh-1845 [000] 598.962716: update_swap_token_priority: mm=ffff88015eaf7700 old_prio=1 new_prio=0 memtoy-1830 [001] 602.033900: update_swap_token_priority: mm=ffff880037a45880 old_prio=947 new_prio=949 memtoy-1830 [000] 602.041509: update_swap_token_priority: mm=ffff880037a45880 old_prio=949 new_prio=951 memtoy-1830 [000] 602.051959: update_swap_token_priority: mm=ffff880037a45880 old_prio=951 new_prio=953 memtoy-1830 [000] 602.052188: update_swap_token_priority: mm=ffff880037a45880 old_prio=953 new_prio=955 memtoy-1830 [001] 602.427184: put_swap_token: token_mm=ffff880037a45880 zsh-1789 [000] 602.427281: replace_swap_token: old_token_mm= (null) old_prio=0 new_token_mm=ffff88015eaf7018 new_prio=2 zsh-1789 [001] 602.433456: update_swap_token_priority: mm=ffff88015eaf7018 old_prio=2 new_prio=4 zsh-1789 [000] 602.437613: update_swap_token_priority: mm=ffff88015eaf7018 old_prio=4 new_prio=6 zsh-1789 [000] 602.443924: update_swap_token_priority: mm=ffff88015eaf7018 old_prio=6 new_prio=8 zsh-1789 [000] 602.451873: update_swap_token_priority: mm=ffff88015eaf7018 old_prio=8 new_prio=10 zsh-1789 [001] 602.462639: update_swap_token_priority: mm=ffff88015eaf7018 old_prio=10 new_prio=12 Signed-off-by: KOSAKI Motohiro Acked-by: Rik van Riel Reviewed-by: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/trace/events/vmscan.h | 77 +++++++++++++++++++++++++++++++++++++++++++ mm/thrash.c | 11 ++++++- 2 files changed, 87 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/trace/events/vmscan.h b/include/trace/events/vmscan.h index ea422aaa23e1..1798e0cee2a9 100644 --- a/include/trace/events/vmscan.h +++ b/include/trace/events/vmscan.h @@ -6,6 +6,8 @@ #include #include +#include +#include #include "gfpflags.h" #define RECLAIM_WB_ANON 0x0001u @@ -310,6 +312,81 @@ TRACE_EVENT(mm_vmscan_lru_shrink_inactive, show_reclaim_flags(__entry->reclaim_flags)) ); +TRACE_EVENT(replace_swap_token, + TP_PROTO(struct mm_struct *old_mm, + struct mm_struct *new_mm), + + TP_ARGS(old_mm, new_mm), + + TP_STRUCT__entry( + __field(struct mm_struct*, old_mm) + __field(unsigned int, old_prio) + __field(struct mm_struct*, new_mm) + __field(unsigned int, new_prio) + ), + + TP_fast_assign( + __entry->old_mm = old_mm; + __entry->old_prio = old_mm ? old_mm->token_priority : 0; + __entry->new_mm = new_mm; + __entry->new_prio = new_mm->token_priority; + ), + + TP_printk("old_token_mm=%p old_prio=%u new_token_mm=%p new_prio=%u", + __entry->old_mm, __entry->old_prio, + __entry->new_mm, __entry->new_prio) +); + +DECLARE_EVENT_CLASS(put_swap_token_template, + TP_PROTO(struct mm_struct *swap_token_mm), + + TP_ARGS(swap_token_mm), + + TP_STRUCT__entry( + __field(struct mm_struct*, swap_token_mm) + ), + + TP_fast_assign( + __entry->swap_token_mm = swap_token_mm; + ), + + TP_printk("token_mm=%p", __entry->swap_token_mm) +); + +DEFINE_EVENT(put_swap_token_template, put_swap_token, + TP_PROTO(struct mm_struct *swap_token_mm), + TP_ARGS(swap_token_mm) +); + +DEFINE_EVENT_CONDITION(put_swap_token_template, disable_swap_token, + TP_PROTO(struct mm_struct *swap_token_mm), + TP_ARGS(swap_token_mm), + TP_CONDITION(swap_token_mm != NULL) +); + +TRACE_EVENT_CONDITION(update_swap_token_priority, + TP_PROTO(struct mm_struct *mm, + unsigned int old_prio), + + TP_ARGS(mm, old_prio), + + TP_CONDITION(mm->token_priority != old_prio), + + TP_STRUCT__entry( + __field(struct mm_struct*, mm) + __field(unsigned int, old_prio) + __field(unsigned int, new_prio) + ), + + TP_fast_assign( + __entry->mm = mm; + __entry->old_prio = old_prio; + __entry->new_prio = mm->token_priority; + ), + + TP_printk("mm=%p old_prio=%u new_prio=%u", + __entry->mm, __entry->old_prio, __entry->new_prio) +); #endif /* _TRACE_VMSCAN_H */ diff --git a/mm/thrash.c b/mm/thrash.c index 6cdf86511385..17d9e29e4c9c 100644 --- a/mm/thrash.c +++ b/mm/thrash.c @@ -23,6 +23,8 @@ #include #include +#include + static DEFINE_SPINLOCK(swap_token_lock); struct mm_struct *swap_token_mm; struct mem_cgroup *swap_token_memcg; @@ -49,6 +51,7 @@ static struct mem_cgroup *swap_token_memcg_from_mm(struct mm_struct *mm) void grab_swap_token(struct mm_struct *mm) { int current_interval; + unsigned int old_prio = mm->token_priority; global_faults++; @@ -63,7 +66,7 @@ void grab_swap_token(struct mm_struct *mm) if (mm == swap_token_mm) { mm->token_priority += 2; - goto out; + goto update_priority; } if (current_interval < mm->last_interval) @@ -77,6 +80,9 @@ void grab_swap_token(struct mm_struct *mm) if (mm->token_priority > swap_token_mm->token_priority) goto replace_token; +update_priority: + trace_update_swap_token_priority(mm, old_prio); + out: mm->faultstamp = global_faults; mm->last_interval = current_interval; @@ -85,6 +91,7 @@ out: replace_token: mm->token_priority += 2; + trace_replace_swap_token(swap_token_mm, mm); swap_token_mm = mm; swap_token_memcg = swap_token_memcg_from_mm(mm); goto out; @@ -95,6 +102,7 @@ void __put_swap_token(struct mm_struct *mm) { spin_lock(&swap_token_lock); if (likely(mm == swap_token_mm)) { + trace_put_swap_token(swap_token_mm); swap_token_mm = NULL; swap_token_memcg = NULL; } @@ -118,6 +126,7 @@ void disable_swap_token(struct mem_cgroup *memcg) if (match_memcg(memcg, swap_token_memcg)) { spin_lock(&swap_token_lock); if (match_memcg(memcg, swap_token_memcg)) { + trace_disable_swap_token(swap_token_mm); swap_token_mm = NULL; swap_token_memcg = NULL; } -- cgit v1.2.3 From d7911ef30cb7bec52234c2b7a5c275ac8f07905a Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Wed, 15 Jun 2011 15:08:15 -0700 Subject: vmscan: implement swap token priority aging While testing for memcg aware swap token, I observed a swap token was often grabbed an intermittent running process (eg init, auditd) and they never release a token. Why? Some processes (eg init, auditd, audispd) wake up when a process exiting. And swap token can be get first page-in process when a process exiting makes no swap token owner. Thus such above intermittent running process often get a token. And currently, swap token priority is only decreased at page fault path. Then, if the process sleep immediately after to grab swap token, the swap token priority never be decreased. That's obviously undesirable. This patch implement very poor (and lightweight) priority aging. It only be affect to the above corner case and doesn't change swap tendency workload performance (eg multi process qsbench load) Signed-off-by: KOSAKI Motohiro Reviewed-by: Rik van Riel Reviewed-by: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/trace/events/vmscan.h | 20 +++++++++++++------- mm/thrash.c | 11 ++++++++++- 2 files changed, 23 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/trace/events/vmscan.h b/include/trace/events/vmscan.h index 1798e0cee2a9..b2c33bd955fa 100644 --- a/include/trace/events/vmscan.h +++ b/include/trace/events/vmscan.h @@ -366,9 +366,10 @@ DEFINE_EVENT_CONDITION(put_swap_token_template, disable_swap_token, TRACE_EVENT_CONDITION(update_swap_token_priority, TP_PROTO(struct mm_struct *mm, - unsigned int old_prio), + unsigned int old_prio, + struct mm_struct *swap_token_mm), - TP_ARGS(mm, old_prio), + TP_ARGS(mm, old_prio, swap_token_mm), TP_CONDITION(mm->token_priority != old_prio), @@ -376,16 +377,21 @@ TRACE_EVENT_CONDITION(update_swap_token_priority, __field(struct mm_struct*, mm) __field(unsigned int, old_prio) __field(unsigned int, new_prio) + __field(struct mm_struct*, swap_token_mm) + __field(unsigned int, swap_token_prio) ), TP_fast_assign( - __entry->mm = mm; - __entry->old_prio = old_prio; - __entry->new_prio = mm->token_priority; + __entry->mm = mm; + __entry->old_prio = old_prio; + __entry->new_prio = mm->token_priority; + __entry->swap_token_mm = swap_token_mm; + __entry->swap_token_prio = swap_token_mm ? swap_token_mm->token_priority : 0; ), - TP_printk("mm=%p old_prio=%u new_prio=%u", - __entry->mm, __entry->old_prio, __entry->new_prio) + TP_printk("mm=%p old_prio=%u new_prio=%u swap_token_mm=%p token_prio=%u", + __entry->mm, __entry->old_prio, __entry->new_prio, + __entry->swap_token_mm, __entry->swap_token_prio) ); #endif /* _TRACE_VMSCAN_H */ diff --git a/mm/thrash.c b/mm/thrash.c index 17d9e29e4c9c..fabf2d0f5169 100644 --- a/mm/thrash.c +++ b/mm/thrash.c @@ -25,10 +25,13 @@ #include +#define TOKEN_AGING_INTERVAL (0xFF) + static DEFINE_SPINLOCK(swap_token_lock); struct mm_struct *swap_token_mm; struct mem_cgroup *swap_token_memcg; static unsigned int global_faults; +static unsigned int last_aging; #ifdef CONFIG_CGROUP_MEM_RES_CTLR static struct mem_cgroup *swap_token_memcg_from_mm(struct mm_struct *mm) @@ -64,6 +67,11 @@ void grab_swap_token(struct mm_struct *mm) if (!swap_token_mm) goto replace_token; + if ((global_faults - last_aging) > TOKEN_AGING_INTERVAL) { + swap_token_mm->token_priority /= 2; + last_aging = global_faults; + } + if (mm == swap_token_mm) { mm->token_priority += 2; goto update_priority; @@ -81,7 +89,7 @@ void grab_swap_token(struct mm_struct *mm) goto replace_token; update_priority: - trace_update_swap_token_priority(mm, old_prio); + trace_update_swap_token_priority(mm, old_prio, swap_token_mm); out: mm->faultstamp = global_faults; @@ -94,6 +102,7 @@ replace_token: trace_replace_swap_token(swap_token_mm, mm); swap_token_mm = mm; swap_token_memcg = swap_token_memcg_from_mm(mm); + last_aging = global_faults; goto out; } -- cgit v1.2.3 From ac5622418bbff9cd3dc607aa57dfb4f62a7f2043 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 15 Jun 2011 15:08:17 -0700 Subject: kmsg_dump.h: fix build when CONFIG_PRINTK is disabled Fix when CONFIG_PRINTK is not enabled: include/linux/kmsg_dump.h:56: error: 'EINVAL' undeclared (first use in this function) include/linux/kmsg_dump.h:61: error: 'EINVAL' undeclared (first use in this function) Looks like commit 595dd3d8bf95 ("kmsg_dump: fix build for CONFIG_PRINTK=n") uses EINVAL without having the needed header file(s), but I'm sure that I build tested that patch also. oh well. Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kmsg_dump.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/kmsg_dump.h b/include/linux/kmsg_dump.h index 2a0d7d651dc3..ee0c952188de 100644 --- a/include/linux/kmsg_dump.h +++ b/include/linux/kmsg_dump.h @@ -12,6 +12,7 @@ #ifndef _LINUX_KMSG_DUMP_H #define _LINUX_KMSG_DUMP_H +#include #include enum kmsg_dump_reason { -- cgit v1.2.3 From 32e45ff43eaf5c17f5a82c9ad358d515622c2562 Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Wed, 15 Jun 2011 15:08:20 -0700 Subject: mm: increase RECLAIM_DISTANCE to 30 Recently, Robert Mueller reported (http://lkml.org/lkml/2010/9/12/236) that zone_reclaim_mode doesn't work properly on his new NUMA server (Dual Xeon E5520 + Intel S5520UR MB). He is using Cyrus IMAPd and it's built on a very traditional single-process model. * a master process which reads config files and manages the other process * multiple imapd processes, one per connection * multiple pop3d processes, one per connection * multiple lmtpd processes, one per connection * periodical "cleanup" processes. There are thousands of independent processes. The problem is, recent Intel motherboard turn on zone_reclaim_mode by default and traditional prefork model software don't work well on it. Unfortunatelly, such models are still typical even in the 21st century. We can't ignore them. This patch raises the zone_reclaim_mode threshold to 30. 30 doesn't have any specific meaning. but 20 means that one-hop QPI/Hypertransport and such relatively cheap 2-4 socket machine are often used for traditional servers as above. The intention is that these machines don't use zone_reclaim_mode. Note: ia64 and Power have arch specific RECLAIM_DISTANCE definitions. This patch doesn't change such high-end NUMA machine behavior. Dave Hansen said: : I know specifically of pieces of x86 hardware that set the information : in the BIOS to '21' *specifically* so they'll get the zone_reclaim_mode : behavior which that implies. : : They've done performance testing and run very large and scary benchmarks : to make sure that they _want_ this turned on. What this means for them : is that they'll probably be de-optimized, at least on newer versions of : the kernel. : : If you want to do this for particular systems, maybe _that_'s what we : should do. Have a list of specific configurations that need the : defaults overridden either because they're buggy, or they have an : unusual hardware configuration not really reflected in the distance : table. And later said: : The original change in the hardware tables was for the benefit of a : benchmark. Said benchmark isn't going to get run on mainline until the : next batch of enterprise distros drops, at which point the hardware where : this was done will be irrelevant for the benchmark. I'm sure any new : hardware will just set this distance to another yet arbitrary value to : make the kernel do what it wants. :) : : Also, when the hardware got _set_ to this initially, I complained. So, I : guess I'm getting my way now, with this patch. I'm cool with it. Reported-by: Robert Mueller Signed-off-by: KOSAKI Motohiro Acked-by: Christoph Lameter Acked-by: David Rientjes Reviewed-by: KAMEZAWA Hiroyuki Cc: Benjamin Herrenschmidt Cc: "Luck, Tony" Acked-by: Dave Hansen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/topology.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/topology.h b/include/linux/topology.h index b91a40e847d2..fc839bfa7935 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -60,7 +60,7 @@ int arch_update_cpu_topology(void); * (in whatever arch specific measurement units returned by node_distance()) * then switch on zone reclaim on boot. */ -#define RECLAIM_DISTANCE 20 +#define RECLAIM_DISTANCE 30 #endif #ifndef PENALTY_FOR_NODE_WITH_CPUS #define PENALTY_FOR_NODE_WITH_CPUS (1) -- cgit v1.2.3 From ca39599c633fb02aceac31a7e67563612e4fe347 Mon Sep 17 00:00:00 2001 From: "Dr. David Alan Gilbert" Date: Wed, 15 Jun 2011 15:08:27 -0700 Subject: BUILD_BUG_ON_ZERO: fix sparse breakage BUILD_BUG_ON_ZERO and BUILD_BUG_ON_NULL must return values, even in the CHECKER case otherwise various users of it become syntactically invalid. Signed-off-by: Dr. David Alan Gilbert Reviewed-by: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kernel.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index fb0e7329fee1..953352a88336 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -671,8 +671,8 @@ struct sysinfo { #ifdef __CHECKER__ #define BUILD_BUG_ON_NOT_POWER_OF_2(n) -#define BUILD_BUG_ON_ZERO(e) -#define BUILD_BUG_ON_NULL(e) +#define BUILD_BUG_ON_ZERO(e) (0) +#define BUILD_BUG_ON_NULL(e) ((void*)0) #define BUILD_BUG_ON(condition) #else /* __CHECKER__ */ -- cgit v1.2.3 From bd5dc17be87b3a3073d50b23802647db3ae3fa8e Mon Sep 17 00:00:00 2001 From: Josh Triplett Date: Wed, 15 Jun 2011 15:08:28 -0700 Subject: uts: make default hostname configurable, rather than always using "(none)" The "hostname" tool falls back to setting the hostname to "localhost" if /etc/hostname does not exist. Distribution init scripts have the same fallback. However, if userspace never calls sethostname, such as when booting with init=/bin/sh, or otherwise booting a minimal system without the usual init scripts, the default hostname of "(none)" remains, unhelpfully appearing in various places such as prompts ("root@(none):~#") and logs. Furthermore, "(none)" doesn't typically resolve to anything useful. Make the default hostname configurable. This removes the need for the standard fallback, provides a useful default for systems that never call sethostname, and makes minimal systems that much more useful with less configuration. Distributions could choose to use "localhost" here to avoid the fallback, while embedded systems may wish to use a specific target hostname. Signed-off-by: Josh Triplett Acked-by: Linus Torvalds Acked-by: David Miller Cc: Serge Hallyn Cc: Kel Modderman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/uts.h | 2 +- init/Kconfig | 9 +++++++++ 2 files changed, 10 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/uts.h b/include/linux/uts.h index 73eb1ed36ec4..6ddbd86377de 100644 --- a/include/linux/uts.h +++ b/include/linux/uts.h @@ -9,7 +9,7 @@ #endif #ifndef UTS_NODENAME -#define UTS_NODENAME "(none)" /* set by sethostname() */ +#define UTS_NODENAME CONFIG_DEFAULT_HOSTNAME /* set by sethostname() */ #endif #ifndef UTS_DOMAINNAME diff --git a/init/Kconfig b/init/Kconfig index ebafac4231ee..e0a22e42d39a 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -204,6 +204,15 @@ config KERNEL_LZO endchoice +config DEFAULT_HOSTNAME + string "Default hostname" + default "(none)" + help + This option determines the default system hostname before userspace + calls sethostname(2). The kernel traditionally uses "(none)" here, + but you may wish to use a different default here to make a minimal + system more usable with less configuration. + config SWAP bool "Support for paging of anonymous memory (swap)" depends on MMU && BLOCK -- cgit v1.2.3 From 49b24d6b41c576ba43153fc94695f871cce139a5 Mon Sep 17 00:00:00 2001 From: Nicolas Kaiser Date: Wed, 15 Jun 2011 15:08:34 -0700 Subject: include/asm-generic/pgtable.h: fix unbalanced parenthesis Signed-off-by: Nicolas Kaiser Reviewed-by: Andrea Arcangeli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-generic/pgtable.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h index e9b8e5926bef..76bff2bff15e 100644 --- a/include/asm-generic/pgtable.h +++ b/include/asm-generic/pgtable.h @@ -88,7 +88,7 @@ static inline pmd_t pmdp_get_and_clear(struct mm_struct *mm, pmd_t pmd = *pmdp; pmd_clear(mm, address, pmdp); return pmd; -}) +} #endif /* CONFIG_TRANSPARENT_HUGEPAGE */ #endif -- cgit v1.2.3 From 4a9a8b71e12d41abb71c4e741bff524f016cfef4 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Tue, 14 Jun 2011 06:13:55 +0000 Subject: drm/radeon: workaround a hw bug on some radeon chipsets with all-0 EDIDs. Some RS690 chipsets seem to end up with floating connectors, either a DVI connector isn't actually populated, or an add-in HDMI card is available but not installed. In this case we seem to get a NULL byte response for each byte of the i2c transaction, so we detect this case and if we see it we don't do anymore DDC transactions on this connector. I've tested this on my RS690 without the HDMI card installed and it seems to work fine. Signed-off-by: Dave Airlie Reviewed-by: Alex Deucher --- drivers/gpu/drm/drm_edid.c | 15 +++++++++++++++ drivers/gpu/drm/radeon/radeon_connectors.c | 7 +++++++ include/drm/drm_crtc.h | 2 ++ 3 files changed, 24 insertions(+) (limited to 'include') diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index 3618d29c79a2..09292193dafe 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -258,6 +258,17 @@ drm_do_probe_ddc_edid(struct i2c_adapter *adapter, unsigned char *buf, return ret == 2 ? 0 : -1; } +static bool drm_edid_is_zero(u8 *in_edid, int length) +{ + int i; + u32 *raw_edid = (u32 *)in_edid; + + for (i = 0; i < length / 4; i++) + if (*(raw_edid + i) != 0) + return false; + return true; +} + static u8 * drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter) { @@ -273,6 +284,10 @@ drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter) goto out; if (drm_edid_block_valid(block)) break; + if (i == 0 && drm_edid_is_zero(block, EDID_LENGTH)) { + connector->null_edid_counter++; + goto carp; + } } if (i == 4) goto carp; diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index 9c2929c7e79f..c04e18ee8a87 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -836,6 +836,13 @@ radeon_dvi_detect(struct drm_connector *connector, bool force) if (!radeon_connector->edid) { DRM_ERROR("%s: probed a monitor but no|invalid EDID\n", drm_get_connector_name(connector)); + /* rs690 seems to have a problem with connectors not existing and always + * return a block of 0's. If we see this just stop polling on this output */ + if ((rdev->family == CHIP_RS690 || rdev->family == CHIP_RS740) && radeon_connector->base.null_edid_counter) { + ret = connector_status_disconnected; + DRM_ERROR("%s: detected RS690 floating bus bug, stopping ddc detect\n", drm_get_connector_name(connector)); + radeon_connector->ddc_bus = NULL; + } } else { radeon_connector->use_digital = !!(radeon_connector->edid->input & DRM_EDID_INPUT_DIGITAL); diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h index 9573e0ce3120..33d12f87f0e0 100644 --- a/include/drm/drm_crtc.h +++ b/include/drm/drm_crtc.h @@ -520,6 +520,8 @@ struct drm_connector { uint32_t encoder_ids[DRM_CONNECTOR_MAX_ENCODER]; uint32_t force_encoder_id; struct drm_encoder *encoder; /* currently active encoder */ + + int null_edid_counter; /* needed to workaround some HW bugs where we get all 0s */ }; /** -- cgit v1.2.3 From c001fb72a7b705f902bdfdd05b5d2408efe6f848 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 14 Jun 2011 17:05:11 -0700 Subject: gpio: add GPIOF_ values regardless on kconfig settings Make GPIOF_ defined values available even when GPIOLIB nor GENERIC_GPIO is enabled by moving them to . Fixes these build errors in linux-next: sound/soc/codecs/ak4641.c:524: error: 'GPIOF_OUT_INIT_LOW' undeclared (first use in this function) sound/soc/codecs/wm8915.c:2921: error: 'GPIOF_OUT_INIT_LOW' undeclared (first use in this function) Signed-off-by: Randy Dunlap Signed-off-by: Grant Likely --- include/asm-generic/gpio.h | 10 ---------- include/linux/gpio.h | 11 +++++++++++ 2 files changed, 11 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/asm-generic/gpio.h b/include/asm-generic/gpio.h index fcdcb5d5c995..d494001b1226 100644 --- a/include/asm-generic/gpio.h +++ b/include/asm-generic/gpio.h @@ -170,16 +170,6 @@ extern int __gpio_cansleep(unsigned gpio); extern int __gpio_to_irq(unsigned gpio); -#define GPIOF_DIR_OUT (0 << 0) -#define GPIOF_DIR_IN (1 << 0) - -#define GPIOF_INIT_LOW (0 << 1) -#define GPIOF_INIT_HIGH (1 << 1) - -#define GPIOF_IN (GPIOF_DIR_IN) -#define GPIOF_OUT_INIT_LOW (GPIOF_DIR_OUT | GPIOF_INIT_LOW) -#define GPIOF_OUT_INIT_HIGH (GPIOF_DIR_OUT | GPIOF_INIT_HIGH) - /** * struct gpio - a structure describing a GPIO with configuration * @gpio: the GPIO number diff --git a/include/linux/gpio.h b/include/linux/gpio.h index 32d47e710661..17b5a0d80e42 100644 --- a/include/linux/gpio.h +++ b/include/linux/gpio.h @@ -3,6 +3,17 @@ /* see Documentation/gpio.txt */ +/* make these flag values available regardless of GPIO kconfig options */ +#define GPIOF_DIR_OUT (0 << 0) +#define GPIOF_DIR_IN (1 << 0) + +#define GPIOF_INIT_LOW (0 << 1) +#define GPIOF_INIT_HIGH (1 << 1) + +#define GPIOF_IN (GPIOF_DIR_IN) +#define GPIOF_OUT_INIT_LOW (GPIOF_DIR_OUT | GPIOF_INIT_LOW) +#define GPIOF_OUT_INIT_HIGH (GPIOF_DIR_OUT | GPIOF_INIT_HIGH) + #ifdef CONFIG_GENERIC_GPIO #include -- cgit v1.2.3 From 42c1edd345c8412d96e7a362ee06feb7be73bb6c Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Thu, 16 Jun 2011 17:29:22 +0200 Subject: netfilter: nf_nat: avoid double seq_adjust for loopback Avoid double seq adjustment for loopback traffic because it causes silent repetition of TCP data. One example is passive FTP with DNAT rule and difference in the length of IP addresses. This patch adds check if packet is sent and received via loopback device. As the same conntrack is used both for outgoing and incoming direction, we restrict seq adjustment to happen only in POSTROUTING. Signed-off-by: Julian Anastasov Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack.h | 6 ++++++ net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 4 +++- 2 files changed, 9 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index c7c42e7acc31..5d4f8e586e32 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -307,6 +307,12 @@ static inline int nf_ct_is_untracked(const struct nf_conn *ct) return test_bit(IPS_UNTRACKED_BIT, &ct->status); } +/* Packet is received from loopback */ +static inline bool nf_is_loopback_packet(const struct sk_buff *skb) +{ + return skb->dev && skb->skb_iif && skb->dev->flags & IFF_LOOPBACK; +} + extern int nf_conntrack_set_hashsize(const char *val, struct kernel_param *kp); extern unsigned int nf_conntrack_htable_size; extern unsigned int nf_conntrack_max; diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index db10075dd88e..de9da21113a1 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -121,7 +121,9 @@ static unsigned int ipv4_confirm(unsigned int hooknum, return ret; } - if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status)) { + /* adjust seqs for loopback traffic only in outgoing direction */ + if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status) && + !nf_is_loopback_packet(skb)) { typeof(nf_nat_seq_adjust_hook) seq_adjust; seq_adjust = rcu_dereference(nf_nat_seq_adjust_hook); -- cgit v1.2.3 From b5199515c25cca622495eb9c6a8a1d275e775088 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 16 Jun 2011 16:22:08 +0200 Subject: clocksource: Make watchdog robust vs. interruption The clocksource watchdog code is interruptible and it has been observed that this can trigger false positives which disable the TSC. The reason is that an interrupt storm or a long running interrupt handler between the read of the watchdog source and the read of the TSC brings the two far enough apart that the delta is larger than the unstable treshold. Move both reads into a short interrupt disabled region to avoid that. Reported-and-tested-by: Vernon Mauery Signed-off-by: Thomas Gleixner Cc: stable@kernel.org --- include/linux/clocksource.h | 1 + kernel/time/clocksource.c | 24 +++++++++++++----------- 2 files changed, 14 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h index d4646b48dc4a..18a1baf31f2d 100644 --- a/include/linux/clocksource.h +++ b/include/linux/clocksource.h @@ -188,6 +188,7 @@ struct clocksource { #ifdef CONFIG_CLOCKSOURCE_WATCHDOG /* Watchdog related data, used by the framework */ struct list_head wd_list; + cycle_t cs_last; cycle_t wd_last; #endif } ____cacheline_aligned; diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index 1c95fd677328..e0980f0d9a0a 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -185,7 +185,6 @@ static struct clocksource *watchdog; static struct timer_list watchdog_timer; static DECLARE_WORK(watchdog_work, clocksource_watchdog_work); static DEFINE_SPINLOCK(watchdog_lock); -static cycle_t watchdog_last; static int watchdog_running; static int clocksource_watchdog_kthread(void *data); @@ -254,11 +253,6 @@ static void clocksource_watchdog(unsigned long data) if (!watchdog_running) goto out; - wdnow = watchdog->read(watchdog); - wd_nsec = clocksource_cyc2ns((wdnow - watchdog_last) & watchdog->mask, - watchdog->mult, watchdog->shift); - watchdog_last = wdnow; - list_for_each_entry(cs, &watchdog_list, wd_list) { /* Clocksource already marked unstable? */ @@ -268,19 +262,28 @@ static void clocksource_watchdog(unsigned long data) continue; } + local_irq_disable(); csnow = cs->read(cs); + wdnow = watchdog->read(watchdog); + local_irq_enable(); /* Clocksource initialized ? */ if (!(cs->flags & CLOCK_SOURCE_WATCHDOG)) { cs->flags |= CLOCK_SOURCE_WATCHDOG; - cs->wd_last = csnow; + cs->wd_last = wdnow; + cs->cs_last = csnow; continue; } - /* Check the deviation from the watchdog clocksource. */ - cs_nsec = clocksource_cyc2ns((csnow - cs->wd_last) & + wd_nsec = clocksource_cyc2ns((wdnow - cs->wd_last) & watchdog->mask, + watchdog->mult, watchdog->shift); + + cs_nsec = clocksource_cyc2ns((csnow - cs->cs_last) & cs->mask, cs->mult, cs->shift); - cs->wd_last = csnow; + cs->cs_last = csnow; + cs->wd_last = wdnow; + + /* Check the deviation from the watchdog clocksource. */ if (abs(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD) { clocksource_unstable(cs, cs_nsec - wd_nsec); continue; @@ -318,7 +321,6 @@ static inline void clocksource_start_watchdog(void) return; init_timer(&watchdog_timer); watchdog_timer.function = clocksource_watchdog; - watchdog_last = watchdog->read(watchdog); watchdog_timer.expires = jiffies + WATCHDOG_INTERVAL; add_timer_on(&watchdog_timer, cpumask_first(cpu_online_mask)); watchdog_running = 1; -- cgit v1.2.3 From d8ad7d1123a960cc9f276bd499f9325c6f5e1bd1 Mon Sep 17 00:00:00 2001 From: Takao Indoh Date: Tue, 29 Mar 2011 12:35:04 -0400 Subject: generic-ipi: Fix kexec boot crash by initializing call_single_queue before enabling interrupts There is a problem that kdump(2nd kernel) sometimes hangs up due to a pending IPI from 1st kernel. Kernel panic occurs because IPI comes before call_single_queue is initialized. To fix the crash, rename init_call_single_data() to call_function_init() and call it in start_kernel() so that call_single_queue can be initialized before enabling interrupts. The details of the crash are: (1) 2nd kernel boots up (2) A pending IPI from 1st kernel comes when irqs are first enabled in start_kernel(). (3) Kernel tries to handle the interrupt, but call_single_queue is not initialized yet at this point. As a result, in the generic_smp_call_function_single_interrupt(), NULL pointer dereference occurs when list_replace_init() tries to access &q->list.next. Therefore this patch changes the name of init_call_single_data() to call_function_init() and calls it before local_irq_enable() in start_kernel(). Signed-off-by: Takao Indoh Reviewed-by: WANG Cong Acked-by: Neil Horman Acked-by: Vivek Goyal Acked-by: Peter Zijlstra Cc: Milton Miller Cc: Jens Axboe Cc: Paul E. McKenney Cc: kexec@lists.infradead.org Link: http://lkml.kernel.org/r/D6CBEE2F420741indou.takao@jp.fujitsu.com Signed-off-by: Ingo Molnar --- include/linux/smp.h | 5 ++++- init/main.c | 1 + kernel/smp.c | 5 +---- 3 files changed, 6 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/smp.h b/include/linux/smp.h index 7ad824d510a2..8cc38d3bab0c 100644 --- a/include/linux/smp.h +++ b/include/linux/smp.h @@ -85,12 +85,15 @@ int smp_call_function_any(const struct cpumask *mask, * Generic and arch helpers */ #ifdef CONFIG_USE_GENERIC_SMP_HELPERS +void __init call_function_init(void); void generic_smp_call_function_single_interrupt(void); void generic_smp_call_function_interrupt(void); void ipi_call_lock(void); void ipi_call_unlock(void); void ipi_call_lock_irq(void); void ipi_call_unlock_irq(void); +#else +static inline void call_function_init(void) { } #endif /* @@ -134,7 +137,7 @@ static inline void smp_send_reschedule(int cpu) { } #define smp_prepare_boot_cpu() do {} while (0) #define smp_call_function_many(mask, func, info, wait) \ (up_smp_call_function(func, info)) -static inline void init_call_single_data(void) { } +static inline void call_function_init(void) { } static inline int smp_call_function_any(const struct cpumask *mask, smp_call_func_t func, diff --git a/init/main.c b/init/main.c index cafba67c13bf..d7211faed2ad 100644 --- a/init/main.c +++ b/init/main.c @@ -542,6 +542,7 @@ asmlinkage void __init start_kernel(void) timekeeping_init(); time_init(); profile_init(); + call_function_init(); if (!irqs_disabled()) printk(KERN_CRIT "start_kernel(): bug: interrupts were " "enabled early\n"); diff --git a/kernel/smp.c b/kernel/smp.c index 73a195193558..fb67dfa8394e 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -74,7 +74,7 @@ static struct notifier_block __cpuinitdata hotplug_cfd_notifier = { .notifier_call = hotplug_cfd, }; -static int __cpuinit init_call_single_data(void) +void __init call_function_init(void) { void *cpu = (void *)(long)smp_processor_id(); int i; @@ -88,10 +88,7 @@ static int __cpuinit init_call_single_data(void) hotplug_cfd(&hotplug_cfd_notifier, CPU_UP_PREPARE, cpu); register_cpu_notifier(&hotplug_cfd_notifier); - - return 0; } -early_initcall(init_call_single_data); /* * csd_lock/csd_unlock used to serialize access to per-cpu csd resources -- cgit v1.2.3 From 879669961b11e7f40b518784863a259f735a72bf Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 17 Jun 2011 11:25:59 +0100 Subject: KEYS/DNS: Fix ____call_usermodehelper() to not lose the session keyring ____call_usermodehelper() now erases any credentials set by the subprocess_inf::init() function. The problem is that commit 17f60a7da150 ("capabilites: allow the application of capability limits to usermode helpers") creates and commits new credentials with prepare_kernel_cred() after the call to the init() function. This wipes all keyrings after umh_keys_init() is called. The best way to deal with this is to put the init() call just prior to the commit_creds() call, and pass the cred pointer to init(). That means that umh_keys_init() and suchlike can modify the credentials _before_ they are published and potentially in use by the rest of the system. This prevents request_key() from working as it is prevented from passing the session keyring it set up with the authorisation token to /sbin/request-key, and so the latter can't assume the authority to instantiate the key. This causes the in-kernel DNS resolver to fail with ENOKEY unconditionally. Signed-off-by: David Howells Acked-by: Eric Paris Tested-by: Jeff Layton Signed-off-by: Linus Torvalds --- fs/exec.c | 2 +- include/linux/kmod.h | 8 ++++---- kernel/kmod.c | 16 +++++++++------- security/keys/request_key.c | 3 +-- 4 files changed, 15 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/fs/exec.c b/fs/exec.c index 97e0d52d72fd..6075a1e727ae 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1996,7 +1996,7 @@ static void wait_for_dump_helpers(struct file *file) * is a special value that we use to trap recursive * core dumps */ -static int umh_pipe_setup(struct subprocess_info *info) +static int umh_pipe_setup(struct subprocess_info *info, struct cred *new) { struct file *rp, *wp; struct fdtable *fdt; diff --git a/include/linux/kmod.h b/include/linux/kmod.h index d4a5c84c503d..0da38cf7db7b 100644 --- a/include/linux/kmod.h +++ b/include/linux/kmod.h @@ -45,7 +45,7 @@ static inline int request_module_nowait(const char *name, ...) { return -ENOSYS; #endif -struct key; +struct cred; struct file; enum umh_wait { @@ -62,7 +62,7 @@ struct subprocess_info { char **envp; enum umh_wait wait; int retval; - int (*init)(struct subprocess_info *info); + int (*init)(struct subprocess_info *info, struct cred *new); void (*cleanup)(struct subprocess_info *info); void *data; }; @@ -73,7 +73,7 @@ struct subprocess_info *call_usermodehelper_setup(char *path, char **argv, /* Set various pieces of state into the subprocess_info structure */ void call_usermodehelper_setfns(struct subprocess_info *info, - int (*init)(struct subprocess_info *info), + int (*init)(struct subprocess_info *info, struct cred *new), void (*cleanup)(struct subprocess_info *info), void *data); @@ -87,7 +87,7 @@ void call_usermodehelper_freeinfo(struct subprocess_info *info); static inline int call_usermodehelper_fns(char *path, char **argv, char **envp, enum umh_wait wait, - int (*init)(struct subprocess_info *info), + int (*init)(struct subprocess_info *info, struct cred *new), void (*cleanup)(struct subprocess_info *), void *data) { struct subprocess_info *info; diff --git a/kernel/kmod.c b/kernel/kmod.c index ad6a81c58b44..47613dfb7b28 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -156,12 +156,6 @@ static int ____call_usermodehelper(void *data) */ set_user_nice(current, 0); - if (sub_info->init) { - retval = sub_info->init(sub_info); - if (retval) - goto fail; - } - retval = -ENOMEM; new = prepare_kernel_cred(current); if (!new) @@ -173,6 +167,14 @@ static int ____call_usermodehelper(void *data) new->cap_inheritable); spin_unlock(&umh_sysctl_lock); + if (sub_info->init) { + retval = sub_info->init(sub_info, new); + if (retval) { + abort_creds(new); + goto fail; + } + } + commit_creds(new); retval = kernel_execve(sub_info->path, @@ -388,7 +390,7 @@ EXPORT_SYMBOL(call_usermodehelper_setup); * context in which call_usermodehelper_exec is called. */ void call_usermodehelper_setfns(struct subprocess_info *info, - int (*init)(struct subprocess_info *info), + int (*init)(struct subprocess_info *info, struct cred *new), void (*cleanup)(struct subprocess_info *info), void *data) { diff --git a/security/keys/request_key.c b/security/keys/request_key.c index d31862e0aa1c..8e319a416eec 100644 --- a/security/keys/request_key.c +++ b/security/keys/request_key.c @@ -71,9 +71,8 @@ EXPORT_SYMBOL(complete_request_key); * This is called in context of freshly forked kthread before kernel_execve(), * so we can simply install the desired session_keyring at this point. */ -static int umh_keys_init(struct subprocess_info *info) +static int umh_keys_init(struct subprocess_info *info, struct cred *cred) { - struct cred *cred = (struct cred*)current_cred(); struct key *keyring = info->data; return install_session_keyring_to_cred(cred, keyring); -- cgit v1.2.3 From cca23d0b5350c9ca0473625c3f5879422ba534a6 Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Sat, 18 Jun 2011 02:51:52 -0700 Subject: Input: sh_keysc - 8x8 MODE_6 fix According to the data sheet for G4, AP4 and AG5 KEYSC MODE_6 is 8x8 keys. Bump up MAXKEYS to 64 too. Signed-off-by: Magnus Damm Reviewed-by: Simon Horman Signed-off-by: Dmitry Torokhov --- drivers/input/keyboard/sh_keysc.c | 2 +- include/linux/input/sh_keysc.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/input/keyboard/sh_keysc.c b/drivers/input/keyboard/sh_keysc.c index 834cf98e7efb..6876700a4469 100644 --- a/drivers/input/keyboard/sh_keysc.c +++ b/drivers/input/keyboard/sh_keysc.c @@ -32,7 +32,7 @@ static const struct { [SH_KEYSC_MODE_3] = { 2, 4, 7 }, [SH_KEYSC_MODE_4] = { 3, 6, 6 }, [SH_KEYSC_MODE_5] = { 4, 6, 7 }, - [SH_KEYSC_MODE_6] = { 5, 7, 7 }, + [SH_KEYSC_MODE_6] = { 5, 8, 8 }, }; struct sh_keysc_priv { diff --git a/include/linux/input/sh_keysc.h b/include/linux/input/sh_keysc.h index 649dc7f12925..5d253cd93691 100644 --- a/include/linux/input/sh_keysc.h +++ b/include/linux/input/sh_keysc.h @@ -1,7 +1,7 @@ #ifndef __SH_KEYSC_H__ #define __SH_KEYSC_H__ -#define SH_KEYSC_MAXKEYS 49 +#define SH_KEYSC_MAXKEYS 64 struct sh_keysc_info { enum { SH_KEYSC_MODE_1, SH_KEYSC_MODE_2, SH_KEYSC_MODE_3, -- cgit v1.2.3 From be98ca652faa6468916a9b7608befff215a8ca70 Mon Sep 17 00:00:00 2001 From: Manoj Iyer Date: Thu, 26 May 2011 11:19:05 -0500 Subject: mmc: Add PCI fixup quirks for Ricoh 1180:e823 reader Signed-off-by: Manoj Iyer Cc: Signed-off-by: Chris Ball --- drivers/pci/quirks.c | 2 ++ include/linux/pci_ids.h | 1 + 2 files changed, 3 insertions(+) (limited to 'include') diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index e8a140669f90..02145e9697a9 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -2761,6 +2761,8 @@ static void ricoh_mmc_fixup_r5c832(struct pci_dev *dev) } DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_RICOH, PCI_DEVICE_ID_RICOH_R5C832, ricoh_mmc_fixup_r5c832); DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_RICOH, PCI_DEVICE_ID_RICOH_R5C832, ricoh_mmc_fixup_r5c832); +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_RICOH, PCI_DEVICE_ID_RICOH_R5CE823, ricoh_mmc_fixup_r5c832); +DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_RICOH, PCI_DEVICE_ID_RICOH_R5CE823, ricoh_mmc_fixup_r5c832); #endif /*CONFIG_MMC_RICOH_MMC*/ #if defined(CONFIG_DMAR) || defined(CONFIG_INTR_REMAP) diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index a311008af5e1..f8910e155566 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1537,6 +1537,7 @@ #define PCI_DEVICE_ID_RICOH_RL5C476 0x0476 #define PCI_DEVICE_ID_RICOH_RL5C478 0x0478 #define PCI_DEVICE_ID_RICOH_R5C822 0x0822 +#define PCI_DEVICE_ID_RICOH_R5CE823 0xe823 #define PCI_DEVICE_ID_RICOH_R5C832 0x0832 #define PCI_DEVICE_ID_RICOH_R5C843 0x0843 -- cgit v1.2.3 From 155d109b5f52ffd749219b27702462dcd9cf4f8d Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Mon, 20 Jun 2011 13:23:14 +0200 Subject: block: add REQ_SECURE to REQ_COMMON_MASK Add REQ_SECURE flag to REQ_COMMON_MASK so that init_request_from_bio() can pass it to @req->cmd_flags. Signed-off-by: Namhyung Kim Acked-by: Adrian Hunter Cc: stable@kernel.org # 2.6.36 and newer Signed-off-by: Jens Axboe --- include/linux/blk_types.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index 2a7cea53ca0d..6395692b2e7a 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -167,7 +167,7 @@ enum rq_flag_bits { (REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT | REQ_FAILFAST_DRIVER) #define REQ_COMMON_MASK \ (REQ_WRITE | REQ_FAILFAST_MASK | REQ_SYNC | REQ_META | REQ_DISCARD | \ - REQ_NOIDLE | REQ_FLUSH | REQ_FUA) + REQ_NOIDLE | REQ_FLUSH | REQ_FUA | REQ_SECURE) #define REQ_CLONE_MASK REQ_COMMON_MASK #define REQ_RAHEAD (1 << __REQ_RAHEAD) -- cgit v1.2.3 From 482e0cd3dbaa70f2a2bead4b5f2c0d203ef654ba Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 19 Jun 2011 13:01:04 -0400 Subject: devcgroup_inode_permission: take "is it a device node" checks to inlined wrapper inode_permission() calls devcgroup_inode_permission() and almost all such calls are _not_ for device nodes; let's at least keep the common path straight... Signed-off-by: Al Viro --- include/linux/device_cgroup.h | 10 +++++++++- security/device_cgroup.c | 8 +------- 2 files changed, 10 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/linux/device_cgroup.h b/include/linux/device_cgroup.h index 0b0d9c39ed67..7aad1f440867 100644 --- a/include/linux/device_cgroup.h +++ b/include/linux/device_cgroup.h @@ -2,8 +2,16 @@ #include #ifdef CONFIG_CGROUP_DEVICE -extern int devcgroup_inode_permission(struct inode *inode, int mask); +extern int __devcgroup_inode_permission(struct inode *inode, int mask); extern int devcgroup_inode_mknod(int mode, dev_t dev); +static inline int devcgroup_inode_permission(struct inode *inode, int mask) +{ + if (likely(!inode->i_rdev)) + return 0; + if (!S_ISBLK(inode->i_mode) && !S_ISCHR(inode->i_mode)) + return 0; + return __devcgroup_inode_permission(inode, mask); +} #else static inline int devcgroup_inode_permission(struct inode *inode, int mask) { return 0; } diff --git a/security/device_cgroup.c b/security/device_cgroup.c index cd1f779fa51d..1be68269e1c2 100644 --- a/security/device_cgroup.c +++ b/security/device_cgroup.c @@ -474,17 +474,11 @@ struct cgroup_subsys devices_subsys = { .subsys_id = devices_subsys_id, }; -int devcgroup_inode_permission(struct inode *inode, int mask) +int __devcgroup_inode_permission(struct inode *inode, int mask) { struct dev_cgroup *dev_cgroup; struct dev_whitelist_item *wh; - dev_t device = inode->i_rdev; - if (!device) - return 0; - if (!S_ISBLK(inode->i_mode) && !S_ISCHR(inode->i_mode)) - return 0; - rcu_read_lock(); dev_cgroup = task_devcgroup(current); -- cgit v1.2.3 From 19345cb299e8234006c5125151ab723e851a1d24 Mon Sep 17 00:00:00 2001 From: Benny Halevy Date: Sun, 19 Jun 2011 18:33:46 -0400 Subject: NFSv4.1: file layout must consider pg_bsize for coalescing Otherwise we end up overflowing the rpc buffer size on the receive end. Signed-off-by: Benny Halevy Signed-off-by: Trond Myklebust --- fs/nfs/nfs4filelayout.c | 6 ++++-- fs/nfs/pagelist.c | 3 ++- include/linux/nfs_page.h | 3 +++ 3 files changed, 9 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c index 5d6f369b15d0..0bafcc91c27f 100644 --- a/fs/nfs/nfs4filelayout.c +++ b/fs/nfs/nfs4filelayout.c @@ -30,6 +30,7 @@ */ #include +#include #include "internal.h" #include "nfs4filelayout.h" @@ -666,8 +667,9 @@ filelayout_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, u64 p_stripe, r_stripe; u32 stripe_unit; - if (!pnfs_generic_pg_test(pgio, prev, req)) - return 0; + if (!pnfs_generic_pg_test(pgio, prev, req) || + !nfs_generic_pg_test(pgio, prev, req)) + return false; if (!pgio->pg_lseg) return 1; diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 7913961aff22..009855716286 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -204,7 +204,7 @@ nfs_wait_on_request(struct nfs_page *req) TASK_UNINTERRUPTIBLE); } -static bool nfs_generic_pg_test(struct nfs_pageio_descriptor *desc, struct nfs_page *prev, struct nfs_page *req) +bool nfs_generic_pg_test(struct nfs_pageio_descriptor *desc, struct nfs_page *prev, struct nfs_page *req) { /* * FIXME: ideally we should be able to coalesce all requests @@ -218,6 +218,7 @@ static bool nfs_generic_pg_test(struct nfs_pageio_descriptor *desc, struct nfs_p return desc->pg_count + req->wb_bytes <= desc->pg_bsize; } +EXPORT_SYMBOL_GPL(nfs_generic_pg_test); /** * nfs_pageio_init - initialise a page io descriptor diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h index 3a34e80ae92f..25311b3bedf8 100644 --- a/include/linux/nfs_page.h +++ b/include/linux/nfs_page.h @@ -92,6 +92,9 @@ extern int nfs_pageio_add_request(struct nfs_pageio_descriptor *, struct nfs_page *); extern void nfs_pageio_complete(struct nfs_pageio_descriptor *desc); extern void nfs_pageio_cond_complete(struct nfs_pageio_descriptor *, pgoff_t); +extern bool nfs_generic_pg_test(struct nfs_pageio_descriptor *desc, + struct nfs_page *prev, + struct nfs_page *req); extern int nfs_wait_on_request(struct nfs_page *); extern void nfs_unlock_request(struct nfs_page *req); extern int nfs_set_page_tag_locked(struct nfs_page *req); -- cgit v1.2.3 From 79568f5be06c91071697c065f01f3ebfbeb25a61 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 20 Jun 2011 20:13:49 -0700 Subject: vfs: i_state needs to be 'unsigned long' for now Commit 13e12d14e2dc ("vfs: reorganize 'struct inode' layout a bit") moved things around a bit changed i_state to be unsigned int instead of unsigned long. That was to help structure layout for the 64-bit case, and shrink 'struct inode' a bit (admittedly that only happened when spinlock debugging was on and i_flags didn't pack with i_lock). However, Meelis Roos reports that this results in unaligned exceptions on sprc, and it turns out that the bit-locking primitives that we use for the I_NEW bit want to use the bitops. Which want 'unsigned long', not 'unsigned int'. We really should fix the bit locking code to not have that kind of requirement, but that's a much bigger change. So for now, revert that field back to 'unsigned long' (but keep the other re-ordering changes from the commit that caused this). Andi points out that we have played games with this in 'struct page', so it's solvable with other hacks too, but since right now the struct inode size advantage only happens with some rare config options, it's not worth fighting. It _would_ be worth fixing the bitlocking code, though. Especially since there is no type safety in the bitlocking code (this never caused any warnings, and worked fine on x86-64, because the bitlocks take a 'void *' and x86-64 doesn't care that deeply about alignment). So it's currently a very easy problem to trigger by mistake and never notice. Reported-by: Meelis Roos Cc: Andi Kleen Cc: David Miller Signed-off-by: Linus Torvalds --- include/linux/fs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/fs.h b/include/linux/fs.h index 1c777878f1ea..6e73e2e9ae33 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -744,7 +744,7 @@ struct inode { spinlock_t i_lock; /* i_blocks, i_bytes, maybe i_size */ unsigned int i_flags; - unsigned int i_state; + unsigned long i_state; #ifdef CONFIG_SECURITY void *i_security; #endif -- cgit v1.2.3 From f76b168b6f117a49d36307053e1acbe30580ea5b Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Sat, 18 Jun 2011 20:22:23 +0200 Subject: PM: Rename dev_pm_info.in_suspend to is_prepared This patch (as1473) renames the "in_suspend" field in struct dev_pm_info to "is_prepared", in preparation for an upcoming change. The new name is more descriptive of what the field really means. Signed-off-by: Alan Stern Signed-off-by: Rafael J. Wysocki Cc: stable@kernel.org --- drivers/base/power/main.c | 14 +++++++++----- drivers/usb/core/driver.c | 6 +++--- include/linux/device.h | 4 ++-- include/linux/pm.h | 2 +- 4 files changed, 15 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index aa6320207745..bf5a59ac1957 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -57,7 +57,7 @@ static int async_error; */ void device_pm_init(struct device *dev) { - dev->power.in_suspend = false; + dev->power.is_prepared = false; init_completion(&dev->power.completion); complete_all(&dev->power.completion); dev->power.wakeup = NULL; @@ -91,7 +91,7 @@ void device_pm_add(struct device *dev) pr_debug("PM: Adding info for %s:%s\n", dev->bus ? dev->bus->name : "No Bus", dev_name(dev)); mutex_lock(&dpm_list_mtx); - if (dev->parent && dev->parent->power.in_suspend) + if (dev->parent && dev->parent->power.is_prepared) dev_warn(dev, "parent %s should not be sleeping\n", dev_name(dev->parent)); list_add_tail(&dev->power.entry, &dpm_list); @@ -511,7 +511,11 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) dpm_wait(dev->parent, async); device_lock(dev); - dev->power.in_suspend = false; + /* + * This is a fib. But we'll allow new children to be added below + * a resumed device, even if the device hasn't been completed yet. + */ + dev->power.is_prepared = false; if (dev->pwr_domain) { pm_dev_dbg(dev, state, "power domain "); @@ -670,7 +674,7 @@ void dpm_complete(pm_message_t state) struct device *dev = to_device(dpm_prepared_list.prev); get_device(dev); - dev->power.in_suspend = false; + dev->power.is_prepared = false; list_move(&dev->power.entry, &list); mutex_unlock(&dpm_list_mtx); @@ -1042,7 +1046,7 @@ int dpm_prepare(pm_message_t state) put_device(dev); break; } - dev->power.in_suspend = true; + dev->power.is_prepared = true; if (!list_empty(&dev->power.entry)) list_move_tail(&dev->power.entry, &dpm_prepared_list); put_device(dev); diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c index e35a17687c05..aa3cc465a601 100644 --- a/drivers/usb/core/driver.c +++ b/drivers/usb/core/driver.c @@ -375,7 +375,7 @@ static int usb_unbind_interface(struct device *dev) * Just re-enable it without affecting the endpoint toggles. */ usb_enable_interface(udev, intf, false); - } else if (!error && !intf->dev.power.in_suspend) { + } else if (!error && !intf->dev.power.is_prepared) { r = usb_set_interface(udev, intf->altsetting[0]. desc.bInterfaceNumber, 0); if (r < 0) @@ -960,7 +960,7 @@ void usb_rebind_intf(struct usb_interface *intf) } /* Try to rebind the interface */ - if (!intf->dev.power.in_suspend) { + if (!intf->dev.power.is_prepared) { intf->needs_binding = 0; rc = device_attach(&intf->dev); if (rc < 0) @@ -1107,7 +1107,7 @@ static int usb_resume_interface(struct usb_device *udev, if (intf->condition == USB_INTERFACE_UNBOUND) { /* Carry out a deferred switch to altsetting 0 */ - if (intf->needs_altsetting0 && !intf->dev.power.in_suspend) { + if (intf->needs_altsetting0 && !intf->dev.power.is_prepared) { usb_set_interface(udev, intf->altsetting[0]. desc.bInterfaceNumber, 0); intf->needs_altsetting0 = 0; diff --git a/include/linux/device.h b/include/linux/device.h index c66111affca9..553fd37b173b 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -654,13 +654,13 @@ static inline int device_is_registered(struct device *dev) static inline void device_enable_async_suspend(struct device *dev) { - if (!dev->power.in_suspend) + if (!dev->power.is_prepared) dev->power.async_suspend = true; } static inline void device_disable_async_suspend(struct device *dev) { - if (!dev->power.in_suspend) + if (!dev->power.is_prepared) dev->power.async_suspend = false; } diff --git a/include/linux/pm.h b/include/linux/pm.h index 3160648ccdda..cc536bd80984 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -425,7 +425,7 @@ struct dev_pm_info { pm_message_t power_state; unsigned int can_wakeup:1; unsigned int async_suspend:1; - unsigned int in_suspend:1; /* Owned by the PM core */ + bool is_prepared:1; /* Owned by the PM core */ spinlock_t lock; #ifdef CONFIG_PM_SLEEP struct list_head entry; -- cgit v1.2.3 From 6d0e0e84f66d32c33511984dd3badd32364b863c Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Sat, 18 Jun 2011 22:42:09 +0200 Subject: PM: Fix async resume following suspend failure The PM core doesn't handle suspend failures correctly when it comes to asynchronously suspended devices. These devices are moved onto the dpm_suspended_list as soon as the corresponding async thread is started up, and they remain on the list even if they fail to suspend or the sleep transition is cancelled before they get suspended. As a result, when the PM core unwinds the transition, it tries to resume the devices even though they were never suspended. This patch (as1474) fixes the problem by adding a new "is_suspended" flag to dev_pm_info. Devices are resumed only if the flag is set. [rjw: * Moved the dev->power.is_suspended check into device_resume(), because we need to complete dev->power.completion and clear dev->power.is_prepared too for devices whose dev->power.is_suspended flags are unset. * Fixed __device_suspend() to avoid setting dev->power.is_suspended if async_error is different from zero.] Signed-off-by: Alan Stern Signed-off-by: Rafael J. Wysocki Cc: stable@kernel.org --- drivers/base/power/main.c | 14 ++++++++++++-- include/linux/pm.h | 1 + 2 files changed, 13 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index bf5a59ac1957..06f09bf89cb2 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -58,6 +58,7 @@ static int async_error; void device_pm_init(struct device *dev) { dev->power.is_prepared = false; + dev->power.is_suspended = false; init_completion(&dev->power.completion); complete_all(&dev->power.completion); dev->power.wakeup = NULL; @@ -517,6 +518,9 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) */ dev->power.is_prepared = false; + if (!dev->power.is_suspended) + goto Unlock; + if (dev->pwr_domain) { pm_dev_dbg(dev, state, "power domain "); error = pm_op(dev, &dev->pwr_domain->ops, state); @@ -552,6 +556,9 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) } End: + dev->power.is_suspended = false; + + Unlock: device_unlock(dev); complete_all(&dev->power.completion); @@ -839,11 +846,11 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) device_lock(dev); if (async_error) - goto End; + goto Unlock; if (pm_wakeup_pending()) { async_error = -EBUSY; - goto End; + goto Unlock; } if (dev->pwr_domain) { @@ -881,6 +888,9 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) } End: + dev->power.is_suspended = !error; + + Unlock: device_unlock(dev); complete_all(&dev->power.completion); diff --git a/include/linux/pm.h b/include/linux/pm.h index cc536bd80984..411e4f4be52b 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -426,6 +426,7 @@ struct dev_pm_info { unsigned int can_wakeup:1; unsigned int async_suspend:1; bool is_prepared:1; /* Owned by the PM core */ + bool is_suspended:1; /* Ditto */ spinlock_t lock; #ifdef CONFIG_PM_SLEEP struct list_head entry; -- cgit v1.2.3 From c6830c22603aaecf65405af23f6da2d55892f9cb Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Thu, 16 Jun 2011 17:28:07 +0900 Subject: Fix node_start/end_pfn() definition for mm/page_cgroup.c commit 21a3c96 uses node_start/end_pfn(nid) for detection start/end of nodes. But, it's not defined in linux/mmzone.h but defined in /arch/???/include/mmzone.h which is included only under CONFIG_NEED_MULTIPLE_NODES=y. Then, we see mm/page_cgroup.c: In function 'page_cgroup_init': mm/page_cgroup.c:308: error: implicit declaration of function 'node_start_pfn' mm/page_cgroup.c:309: error: implicit declaration of function 'node_end_pfn' So, fixiing page_cgroup.c is an idea... But node_start_pfn()/node_end_pfn() is a very generic macro and should be implemented in the same manner for all archs. (m32r has different implementation...) This patch removes definitions of node_start/end_pfn() in each archs and defines a unified one in linux/mmzone.h. It's not under CONFIG_NEED_MULTIPLE_NODES, now. A result of macro expansion is here (mm/page_cgroup.c) for !NUMA start_pfn = ((&contig_page_data)->node_start_pfn); end_pfn = ({ pg_data_t *__pgdat = (&contig_page_data); __pgdat->node_start_pfn + __pgdat->node_spanned_pages;}); for NUMA (x86-64) start_pfn = ((node_data[nid])->node_start_pfn); end_pfn = ({ pg_data_t *__pgdat = (node_data[nid]); __pgdat->node_start_pfn + __pgdat->node_spanned_pages;}); Changelog: - fixed to avoid using "nid" twice in node_end_pfn() macro. Reported-and-acked-by: Randy Dunlap Reported-and-tested-by: Ingo Molnar Acked-by: Mel Gorman Signed-off-by: KAMEZAWA Hiroyuki Signed-off-by: Linus Torvalds --- arch/alpha/include/asm/mmzone.h | 1 - arch/m32r/include/asm/mmzone.h | 8 +------- arch/parisc/include/asm/mmzone.h | 7 ------- arch/powerpc/include/asm/mmzone.h | 7 ------- arch/sh/include/asm/mmzone.h | 4 ---- arch/sparc/include/asm/mmzone.h | 2 -- arch/tile/include/asm/mmzone.h | 11 ----------- arch/x86/include/asm/mmzone_32.h | 11 ----------- arch/x86/include/asm/mmzone_64.h | 3 --- include/linux/mmzone.h | 7 +++++++ 10 files changed, 8 insertions(+), 53 deletions(-) (limited to 'include') diff --git a/arch/alpha/include/asm/mmzone.h b/arch/alpha/include/asm/mmzone.h index 8af56ce346ad..445dc42e0334 100644 --- a/arch/alpha/include/asm/mmzone.h +++ b/arch/alpha/include/asm/mmzone.h @@ -56,7 +56,6 @@ PLAT_NODE_DATA_LOCALNR(unsigned long p, int n) * Given a kernel address, find the home node of the underlying memory. */ #define kvaddr_to_nid(kaddr) pa_to_nid(__pa(kaddr)) -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) /* * Given a kaddr, LOCAL_BASE_ADDR finds the owning node of the memory diff --git a/arch/m32r/include/asm/mmzone.h b/arch/m32r/include/asm/mmzone.h index 9f3b5accda88..115ced33febd 100644 --- a/arch/m32r/include/asm/mmzone.h +++ b/arch/m32r/include/asm/mmzone.h @@ -14,12 +14,6 @@ extern struct pglist_data *node_data[]; #define NODE_DATA(nid) (node_data[nid]) #define node_localnr(pfn, nid) ((pfn) - NODE_DATA(nid)->node_start_pfn) -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) \ -({ \ - pg_data_t *__pgdat = NODE_DATA(nid); \ - __pgdat->node_start_pfn + __pgdat->node_spanned_pages - 1; \ -}) #define pmd_page(pmd) (pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT)) /* @@ -44,7 +38,7 @@ static __inline__ int pfn_to_nid(unsigned long pfn) int node; for (node = 0 ; node < MAX_NUMNODES ; node++) - if (pfn >= node_start_pfn(node) && pfn <= node_end_pfn(node)) + if (pfn >= node_start_pfn(node) && pfn < node_end_pfn(node)) break; return node; diff --git a/arch/parisc/include/asm/mmzone.h b/arch/parisc/include/asm/mmzone.h index 9608d2cf214a..e67eb9c3d1bf 100644 --- a/arch/parisc/include/asm/mmzone.h +++ b/arch/parisc/include/asm/mmzone.h @@ -14,13 +14,6 @@ extern struct node_map_data node_data[]; #define NODE_DATA(nid) (&node_data[nid].pg_data) -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) \ -({ \ - pg_data_t *__pgdat = NODE_DATA(nid); \ - __pgdat->node_start_pfn + __pgdat->node_spanned_pages; \ -}) - /* We have these possible memory map layouts: * Astro: 0-3.75, 67.75-68, 4-64 * zx1: 0-1, 257-260, 4-256 diff --git a/arch/powerpc/include/asm/mmzone.h b/arch/powerpc/include/asm/mmzone.h index fd3fd58bad84..7b589178be46 100644 --- a/arch/powerpc/include/asm/mmzone.h +++ b/arch/powerpc/include/asm/mmzone.h @@ -38,13 +38,6 @@ u64 memory_hotplug_max(void); #define memory_hotplug_max() memblock_end_of_DRAM() #endif -/* - * Following are macros that each numa implmentation must define. - */ - -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) (NODE_DATA(nid)->node_end_pfn) - #else #define memory_hotplug_max() memblock_end_of_DRAM() #endif /* CONFIG_NEED_MULTIPLE_NODES */ diff --git a/arch/sh/include/asm/mmzone.h b/arch/sh/include/asm/mmzone.h index 8887baff5eff..15a8496960e6 100644 --- a/arch/sh/include/asm/mmzone.h +++ b/arch/sh/include/asm/mmzone.h @@ -9,10 +9,6 @@ extern struct pglist_data *node_data[]; #define NODE_DATA(nid) (node_data[nid]) -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) (NODE_DATA(nid)->node_start_pfn + \ - NODE_DATA(nid)->node_spanned_pages) - static inline int pfn_to_nid(unsigned long pfn) { int nid; diff --git a/arch/sparc/include/asm/mmzone.h b/arch/sparc/include/asm/mmzone.h index e8c648741ed4..99d9b9f577bf 100644 --- a/arch/sparc/include/asm/mmzone.h +++ b/arch/sparc/include/asm/mmzone.h @@ -8,8 +8,6 @@ extern struct pglist_data *node_data[]; #define NODE_DATA(nid) (node_data[nid]) -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) (NODE_DATA(nid)->node_end_pfn) extern int numa_cpu_lookup_table[]; extern cpumask_t numa_cpumask_lookup_table[]; diff --git a/arch/tile/include/asm/mmzone.h b/arch/tile/include/asm/mmzone.h index c6344c4f32ac..9d3dbce8f953 100644 --- a/arch/tile/include/asm/mmzone.h +++ b/arch/tile/include/asm/mmzone.h @@ -40,17 +40,6 @@ static inline int pfn_to_nid(unsigned long pfn) return highbits_to_node[__pfn_to_highbits(pfn)]; } -/* - * Following are macros that each numa implmentation must define. - */ - -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) \ -({ \ - pg_data_t *__pgdat = NODE_DATA(nid); \ - __pgdat->node_start_pfn + __pgdat->node_spanned_pages; \ -}) - #define kern_addr_valid(kaddr) virt_addr_valid((void *)kaddr) static inline int pfn_valid(int pfn) diff --git a/arch/x86/include/asm/mmzone_32.h b/arch/x86/include/asm/mmzone_32.h index 5e83a416eca8..224e8c5eb307 100644 --- a/arch/x86/include/asm/mmzone_32.h +++ b/arch/x86/include/asm/mmzone_32.h @@ -48,17 +48,6 @@ static inline int pfn_to_nid(unsigned long pfn) #endif } -/* - * Following are macros that each numa implmentation must define. - */ - -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) \ -({ \ - pg_data_t *__pgdat = NODE_DATA(nid); \ - __pgdat->node_start_pfn + __pgdat->node_spanned_pages; \ -}) - static inline int pfn_valid(int pfn) { int nid = pfn_to_nid(pfn); diff --git a/arch/x86/include/asm/mmzone_64.h b/arch/x86/include/asm/mmzone_64.h index b3f88d7867c7..129d9aa3ceb3 100644 --- a/arch/x86/include/asm/mmzone_64.h +++ b/arch/x86/include/asm/mmzone_64.h @@ -13,8 +13,5 @@ extern struct pglist_data *node_data[]; #define NODE_DATA(nid) (node_data[nid]) -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) (NODE_DATA(nid)->node_start_pfn + \ - NODE_DATA(nid)->node_spanned_pages) #endif #endif /* _ASM_X86_MMZONE_64_H */ diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index c928dac6cad0..9f7c3ebcbbad 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -647,6 +647,13 @@ typedef struct pglist_data { #endif #define nid_page_nr(nid, pagenr) pgdat_page_nr(NODE_DATA(nid),(pagenr)) +#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) + +#define node_end_pfn(nid) ({\ + pg_data_t *__pgdat = NODE_DATA(nid);\ + __pgdat->node_start_pfn + __pgdat->node_spanned_pages;\ +}) + #include extern struct mutex zonelists_mutex; -- cgit v1.2.3 From 4d258b25d947521c8b913154db61ec55198243f8 Mon Sep 17 00:00:00 2001 From: Vitaliy Ivanov Date: Mon, 27 Jun 2011 19:07:08 +0300 Subject: Fix some kernel-doc warnings Fix 'make htmldocs' warnings: Warning(/include/linux/hrtimer.h:153): No description found for parameter 'clockid' Warning(/include/linux/device.h:604): Excess struct/union/enum/typedef member 'of_match' description in 'device' Warning(/include/net/sock.h:349): Excess struct/union/enum/typedef member 'sk_rmem_alloc' description in 'sock' Signed-off-by: Vitaliy Ivanov Acked-by: Grant Likely Acked-by: David S. Miller Signed-off-by: Linus Torvalds --- include/linux/device.h | 1 - include/linux/hrtimer.h | 1 + include/net/sock.h | 1 - 3 files changed, 1 insertion(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/device.h b/include/linux/device.h index 553fd37b173b..e4f62d8896b7 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -530,7 +530,6 @@ struct device_dma_parameters { * @dma_mem: Internal for coherent mem override. * @archdata: For arch-specific additions. * @of_node: Associated device tree node. - * @of_match: Matching of_device_id from driver. * @devt: For creating the sysfs "dev". * @devres_lock: Spinlock to protect the resource of the device. * @devres_head: The resources list of the device. diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h index 51932e5acf7c..fd0dc30c9f15 100644 --- a/include/linux/hrtimer.h +++ b/include/linux/hrtimer.h @@ -135,6 +135,7 @@ struct hrtimer_sleeper { * @cpu_base: per cpu clock base * @index: clock type index for per_cpu support when moving a * timer to a base on another cpu. + * @clockid: clock id for per_cpu support * @active: red black tree root node for the active timers * @resolution: the resolution of the clock, in nanoseconds * @get_time: function to retrieve the current time of the clock diff --git a/include/net/sock.h b/include/net/sock.h index f2046e404a61..c0b938cb4b1a 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -178,7 +178,6 @@ struct sock_common { * @sk_dst_cache: destination cache * @sk_dst_lock: destination cache lock * @sk_policy: flow policy - * @sk_rmem_alloc: receive queue bytes committed * @sk_receive_queue: incoming packets * @sk_wmem_alloc: transmit queue bytes committed * @sk_write_queue: Packet sending queue -- cgit v1.2.3 From 072441e21ddcd1140606b7d4ef6eab579a86b0b3 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 27 Jun 2011 16:18:02 -0700 Subject: mm: move shmem prototypes to shmem_fs.h Before adding any more global entry points into shmem.c, gather such prototypes into shmem_fs.h. Remove mm's own declarations from swap.h, but for now leave the ones in mm.h: because shmem_file_setup() and shmem_zero_setup() are called from various places, and we should not force other subsystems to update immediately. Signed-off-by: Hugh Dickins Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/shmem_fs.h | 17 +++++++++++++++++ include/linux/swap.h | 10 ---------- mm/memcontrol.c | 1 + mm/swapfile.c | 2 +- 4 files changed, 19 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h index 2b7fec840517..cae65dc42bcc 100644 --- a/include/linux/shmem_fs.h +++ b/include/linux/shmem_fs.h @@ -5,6 +5,13 @@ #include #include +struct page; +struct file; +struct inode; +struct super_block; +struct user_struct; +struct vm_area_struct; + /* inode in-kernel data */ #define SHMEM_NR_DIRECT 16 @@ -45,7 +52,17 @@ static inline struct shmem_inode_info *SHMEM_I(struct inode *inode) return container_of(inode, struct shmem_inode_info, vfs_inode); } +/* + * Functions in mm/shmem.c called directly from elsewhere: + */ extern int init_tmpfs(void); extern int shmem_fill_super(struct super_block *sb, void *data, int silent); +extern struct file *shmem_file_setup(const char *name, + loff_t size, unsigned long flags); +extern int shmem_zero_setup(struct vm_area_struct *); +extern int shmem_lock(struct file *file, int lock, struct user_struct *user); +extern int shmem_unuse(swp_entry_t entry, struct page *page); +extern void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, + struct page **pagep, swp_entry_t *ent); #endif diff --git a/include/linux/swap.h b/include/linux/swap.h index e70564647039..a273468f8285 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -300,16 +300,6 @@ static inline void scan_unevictable_unregister_node(struct node *node) extern int kswapd_run(int nid); extern void kswapd_stop(int nid); -#ifdef CONFIG_MMU -/* linux/mm/shmem.c */ -extern int shmem_unuse(swp_entry_t entry, struct page *page); -#endif /* CONFIG_MMU */ - -#ifdef CONFIG_CGROUP_MEM_RES_CTLR -extern void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, - struct page **pagep, swp_entry_t *ent); -#endif - #ifdef CONFIG_SWAP /* linux/mm/page_io.c */ extern int swap_readpage(struct page *); diff --git a/mm/memcontrol.c b/mm/memcontrol.c index cf7d027a8844..ddffc74cdebe 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -35,6 +35,7 @@ #include #include #include +#include #include #include #include diff --git a/mm/swapfile.c b/mm/swapfile.c index d537d29e9b7b..ff8dc1a18cb4 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -14,7 +14,7 @@ #include #include #include -#include +#include #include #include #include -- cgit v1.2.3 From 94c1e62df4494b79782cb9c7279f827212d1de70 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 27 Jun 2011 16:18:03 -0700 Subject: tmpfs: take control of its truncate_range 2.6.35's new truncate convention gave tmpfs the opportunity to control its file truncation, no longer enforced from outside by vmtruncate(). We shall want to build upon that, to handle pagecache and swap together. Slightly redefine the ->truncate_range interface: let it now be called between the unmap_mapping_range()s, with the filesystem responsible for doing the truncate_inode_pages_range() from it - just as the filesystem is nowadays responsible for doing that from its ->setattr. Let's rename shmem_notify_change() to shmem_setattr(). Instead of calling the generic truncate_setsize(), bring that code in so we can call shmem_truncate_range() - which will later be updated to perform its own variant of truncate_inode_pages_range(). Remove the punch_hole unmap_mapping_range() from shmem_truncate_range(): now that the COW's unmap_mapping_range() comes after ->truncate_range, there is no need to call it a third time. Export shmem_truncate_range() and add it to the list in shmem_fs.h, so that i915_gem_object_truncate() can call it explicitly in future; get this patch in first, then update drm/i915 once this is available (until then, i915 will just be doing the truncate_inode_pages() twice). Though introduced five years ago, no other filesystem is implementing ->truncate_range, and its only other user is madvise(,,MADV_REMOVE): we expect to convert it to fallocate(,FALLOC_FL_PUNCH_HOLE,,) shortly, whereupon ->truncate_range can be removed from inode_operations - shmem_truncate_range() will help i915 across that transition too. Signed-off-by: Hugh Dickins Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/shmem_fs.h | 1 + mm/shmem.c | 51 +++++++++++++++++++++++++++--------------------- mm/truncate.c | 4 ++-- 3 files changed, 32 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h index cae65dc42bcc..22a20af4d785 100644 --- a/include/linux/shmem_fs.h +++ b/include/linux/shmem_fs.h @@ -61,6 +61,7 @@ extern struct file *shmem_file_setup(const char *name, loff_t size, unsigned long flags); extern int shmem_zero_setup(struct vm_area_struct *); extern int shmem_lock(struct file *file, int lock, struct user_struct *user); +extern void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end); extern int shmem_unuse(swp_entry_t entry, struct page *page); extern void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, struct page **pagep, swp_entry_t *ent); diff --git a/mm/shmem.c b/mm/shmem.c index d221a1cfd7b1..f1714758ea96 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -539,7 +539,7 @@ static void shmem_free_pages(struct list_head *next) } while (next); } -static void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end) +void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end) { struct shmem_inode_info *info = SHMEM_I(inode); unsigned long idx; @@ -562,6 +562,8 @@ static void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end) spinlock_t *punch_lock; unsigned long upper_limit; + truncate_inode_pages_range(inode->i_mapping, start, end); + inode->i_ctime = inode->i_mtime = CURRENT_TIME; idx = (start + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; if (idx >= info->next_index) @@ -738,16 +740,8 @@ done2: * lowered next_index. Also, though shmem_getpage checks * i_size before adding to cache, no recheck after: so fix the * narrow window there too. - * - * Recalling truncate_inode_pages_range and unmap_mapping_range - * every time for punch_hole (which never got a chance to clear - * SHMEM_PAGEIN at the start of vmtruncate_range) is expensive, - * yet hardly ever necessary: try to optimize them out later. */ truncate_inode_pages_range(inode->i_mapping, start, end); - if (punch_hole) - unmap_mapping_range(inode->i_mapping, start, - end - start, 1); } spin_lock(&info->lock); @@ -766,22 +760,23 @@ done2: shmem_free_pages(pages_to_free.next); } } +EXPORT_SYMBOL_GPL(shmem_truncate_range); -static int shmem_notify_change(struct dentry *dentry, struct iattr *attr) +static int shmem_setattr(struct dentry *dentry, struct iattr *attr) { struct inode *inode = dentry->d_inode; - loff_t newsize = attr->ia_size; int error; error = inode_change_ok(inode, attr); if (error) return error; - if (S_ISREG(inode->i_mode) && (attr->ia_valid & ATTR_SIZE) - && newsize != inode->i_size) { + if (S_ISREG(inode->i_mode) && (attr->ia_valid & ATTR_SIZE)) { + loff_t oldsize = inode->i_size; + loff_t newsize = attr->ia_size; struct page *page = NULL; - if (newsize < inode->i_size) { + if (newsize < oldsize) { /* * If truncating down to a partial page, then * if that page is already allocated, hold it @@ -810,12 +805,19 @@ static int shmem_notify_change(struct dentry *dentry, struct iattr *attr) spin_unlock(&info->lock); } } - - /* XXX(truncate): truncate_setsize should be called last */ - truncate_setsize(inode, newsize); + if (newsize != oldsize) { + i_size_write(inode, newsize); + inode->i_ctime = inode->i_mtime = CURRENT_TIME; + } + if (newsize < oldsize) { + loff_t holebegin = round_up(newsize, PAGE_SIZE); + unmap_mapping_range(inode->i_mapping, holebegin, 0, 1); + shmem_truncate_range(inode, newsize, (loff_t)-1); + /* unmap again to remove racily COWed private pages */ + unmap_mapping_range(inode->i_mapping, holebegin, 0, 1); + } if (page) page_cache_release(page); - shmem_truncate_range(inode, newsize, (loff_t)-1); } setattr_copy(inode, attr); @@ -832,7 +834,6 @@ static void shmem_evict_inode(struct inode *inode) struct shmem_xattr *xattr, *nxattr; if (inode->i_mapping->a_ops == &shmem_aops) { - truncate_inode_pages(inode->i_mapping, 0); shmem_unacct_size(info->flags, inode->i_size); inode->i_size = 0; shmem_truncate_range(inode, 0, (loff_t)-1); @@ -2706,7 +2707,7 @@ static const struct file_operations shmem_file_operations = { }; static const struct inode_operations shmem_inode_operations = { - .setattr = shmem_notify_change, + .setattr = shmem_setattr, .truncate_range = shmem_truncate_range, #ifdef CONFIG_TMPFS_XATTR .setxattr = shmem_setxattr, @@ -2739,7 +2740,7 @@ static const struct inode_operations shmem_dir_inode_operations = { .removexattr = shmem_removexattr, #endif #ifdef CONFIG_TMPFS_POSIX_ACL - .setattr = shmem_notify_change, + .setattr = shmem_setattr, .check_acl = generic_check_acl, #endif }; @@ -2752,7 +2753,7 @@ static const struct inode_operations shmem_special_inode_operations = { .removexattr = shmem_removexattr, #endif #ifdef CONFIG_TMPFS_POSIX_ACL - .setattr = shmem_notify_change, + .setattr = shmem_setattr, .check_acl = generic_check_acl, #endif }; @@ -2908,6 +2909,12 @@ int shmem_lock(struct file *file, int lock, struct user_struct *user) return 0; } +void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end) +{ + truncate_inode_pages_range(inode->i_mapping, start, end); +} +EXPORT_SYMBOL_GPL(shmem_truncate_range); + #ifdef CONFIG_CGROUP_MEM_RES_CTLR /** * mem_cgroup_get_shmem_target - find a page or entry assigned to the shmem file diff --git a/mm/truncate.c b/mm/truncate.c index 5b4c3a4847e9..29a9b8a5a31a 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -619,9 +619,9 @@ int vmtruncate_range(struct inode *inode, loff_t offset, loff_t end) mutex_lock(&inode->i_mutex); down_write(&inode->i_alloc_sem); unmap_mapping_range(mapping, offset, (end - offset), 1); - truncate_inode_pages_range(mapping, offset, end); - unmap_mapping_range(mapping, offset, (end - offset), 1); inode->i_op->truncate_range(inode, offset, end); + /* unmap again to remove racily COWed private pages */ + unmap_mapping_range(mapping, offset, (end - offset), 1); up_write(&inode->i_alloc_sem); mutex_unlock(&inode->i_mutex); -- cgit v1.2.3 From d9d90e5eb70e09903dadff42099b6c948f814050 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 27 Jun 2011 16:18:04 -0700 Subject: tmpfs: add shmem_read_mapping_page_gfp Although it is used (by i915) on nothing but tmpfs, read_cache_page_gfp() is unsuited to tmpfs, because it inserts a page into pagecache before calling the filesystem's ->readpage: tmpfs may have pages in swapcache which only it knows how to locate and switch to filecache. At present tmpfs provides a ->readpage method, and copes with this by copying pages; but soon we can simplify it by removing its ->readpage. Provide shmem_read_mapping_page_gfp() now, ready for that transition, Export shmem_read_mapping_page_gfp() and add it to list in shmem_fs.h, with shmem_read_mapping_page() inline for the common mapping_gfp case. (shmem_read_mapping_page_gfp or shmem_read_cache_page_gfp? Generally the read_mapping_page functions use the mapping's ->readpage, and the read_cache_page functions use the supplied filler, so I think read_cache_page_gfp was slightly misnamed.) Signed-off-by: Hugh Dickins Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/shmem_fs.h | 17 ++++++++++------- mm/shmem.c | 23 +++++++++++++++++++++++ 2 files changed, 33 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h index 22a20af4d785..aa08fa8fd79b 100644 --- a/include/linux/shmem_fs.h +++ b/include/linux/shmem_fs.h @@ -3,15 +3,9 @@ #include #include +#include #include -struct page; -struct file; -struct inode; -struct super_block; -struct user_struct; -struct vm_area_struct; - /* inode in-kernel data */ #define SHMEM_NR_DIRECT 16 @@ -61,9 +55,18 @@ extern struct file *shmem_file_setup(const char *name, loff_t size, unsigned long flags); extern int shmem_zero_setup(struct vm_area_struct *); extern int shmem_lock(struct file *file, int lock, struct user_struct *user); +extern struct page *shmem_read_mapping_page_gfp(struct address_space *mapping, + pgoff_t index, gfp_t gfp_mask); extern void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end); extern int shmem_unuse(swp_entry_t entry, struct page *page); extern void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, struct page **pagep, swp_entry_t *ent); +static inline struct page *shmem_read_mapping_page( + struct address_space *mapping, pgoff_t index) +{ + return shmem_read_mapping_page_gfp(mapping, index, + mapping_gfp_mask(mapping)); +} + #endif diff --git a/mm/shmem.c b/mm/shmem.c index f1714758ea96..fcedf5464eb7 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -3035,3 +3035,26 @@ int shmem_zero_setup(struct vm_area_struct *vma) vma->vm_flags |= VM_CAN_NONLINEAR; return 0; } + +/** + * shmem_read_mapping_page_gfp - read into page cache, using specified page allocation flags. + * @mapping: the page's address_space + * @index: the page index + * @gfp: the page allocator flags to use if allocating + * + * This behaves as a tmpfs "read_cache_page_gfp(mapping, index, gfp)", + * with any new page allocations done using the specified allocation flags. + * But read_cache_page_gfp() uses the ->readpage() method: which does not + * suit tmpfs, since it may have pages in swapcache, and needs to find those + * for itself; although drivers/gpu/drm i915 and ttm rely upon this support. + * + * Provide a stub for those callers to start using now, then later + * flesh it out to call shmem_getpage() with additional gfp mask, when + * shmem_file_splice_read() is added and shmem_readpage() is removed. + */ +struct page *shmem_read_mapping_page_gfp(struct address_space *mapping, + pgoff_t index, gfp_t gfp) +{ + return read_cache_page_gfp(mapping, index, gfp); +} +EXPORT_SYMBOL_GPL(shmem_read_mapping_page_gfp); -- cgit v1.2.3 From 507c5f1224014f9956e604ee8703b3bbea7da4a4 Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Mon, 27 Jun 2011 16:18:07 -0700 Subject: include/linux/compat.h: declare compat_sys_sendmmsg() This is required for tilegx to be able to use the compat unistd.h header where compat_sys_sendmmsg() is now mentioned. Signed-off-by: Chris Metcalf Cc: Arnd Bergmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/compat.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/linux/compat.h b/include/linux/compat.h index ddcb7db38e67..846bb1792572 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -467,6 +467,8 @@ asmlinkage long compat_sys_setsockopt(int fd, int level, int optname, char __user *optval, unsigned int optlen); asmlinkage long compat_sys_sendmsg(int fd, struct compat_msghdr __user *msg, unsigned flags); +asmlinkage long compat_sys_sendmmsg(int fd, struct compat_mmsghdr __user *mmsg, + unsigned vlen, unsigned int flags); asmlinkage long compat_sys_recvmsg(int fd, struct compat_msghdr __user *msg, unsigned int flags); asmlinkage long compat_sys_recv(int fd, void __user *buf, size_t len, -- cgit v1.2.3 From 08142579b6ca35883c1ed066a2681de6f6917062 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 27 Jun 2011 16:18:10 -0700 Subject: mm: fix assertion mapping->nrpages == 0 in end_writeback() Under heavy memory and filesystem load, users observe the assertion mapping->nrpages == 0 in end_writeback() trigger. This can be caused by page reclaim reclaiming the last page from a mapping in the following race: CPU0 CPU1 ... shrink_page_list() __remove_mapping() __delete_from_page_cache() radix_tree_delete() evict_inode() truncate_inode_pages() truncate_inode_pages_range() pagevec_lookup() - finds nothing end_writeback() mapping->nrpages != 0 -> BUG page->mapping = NULL mapping->nrpages-- Fix the problem by doing a reliable check of mapping->nrpages under mapping->tree_lock in end_writeback(). Analyzed by Jay , lost in LKML, and dug out by Miklos Szeredi . Cc: Jay Cc: Miklos Szeredi Signed-off-by: Jan Kara Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/inode.c | 7 +++++++ include/linux/fs.h | 1 + mm/truncate.c | 5 +++++ 3 files changed, 13 insertions(+) (limited to 'include') diff --git a/fs/inode.c b/fs/inode.c index 0f7e88a7803f..43566d17d1b8 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -423,7 +423,14 @@ EXPORT_SYMBOL(remove_inode_hash); void end_writeback(struct inode *inode) { might_sleep(); + /* + * We have to cycle tree_lock here because reclaim can be still in the + * process of removing the last page (in __delete_from_page_cache()) + * and we must not free mapping under it. + */ + spin_lock_irq(&inode->i_data.tree_lock); BUG_ON(inode->i_data.nrpages); + spin_unlock_irq(&inode->i_data.tree_lock); BUG_ON(!list_empty(&inode->i_data.private_list)); BUG_ON(!(inode->i_state & I_FREEING)); BUG_ON(inode->i_state & I_CLEAR); diff --git a/include/linux/fs.h b/include/linux/fs.h index 6e73e2e9ae33..b5b979247863 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -639,6 +639,7 @@ struct address_space { struct prio_tree_root i_mmap; /* tree of private and shared mappings */ struct list_head i_mmap_nonlinear;/*list VM_NONLINEAR mappings */ struct mutex i_mmap_mutex; /* protect tree, count, list */ + /* Protected by tree_lock together with the radix tree */ unsigned long nrpages; /* number of total pages */ pgoff_t writeback_index;/* writeback starts here */ const struct address_space_operations *a_ops; /* methods */ diff --git a/mm/truncate.c b/mm/truncate.c index 29a9b8a5a31a..e13f22efaad7 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -304,6 +304,11 @@ EXPORT_SYMBOL(truncate_inode_pages_range); * @lstart: offset from which to truncate * * Called under (and serialised by) inode->i_mutex. + * + * Note: When this function returns, there can be a page in the process of + * deletion (inside __delete_from_page_cache()) in the specified range. Thus + * mapping->nrpages can be non-zero when this function returns even after + * truncation of the whole mapping. */ void truncate_inode_pages(struct address_space *mapping, loff_t lstart) { -- cgit v1.2.3