From b67b4b117746aef686e527c3205792db0f2c9e16 Mon Sep 17 00:00:00 2001 From: Dominic Curran Date: Mon, 27 Oct 2008 22:30:53 -0400 Subject: Input: gpio-keys - add flag to allow auto repeat This patch adds a flag to gpio-key driver to turn on the input subsystems auto repeat feature if needed. Signed-off-by: Dominic Curran Signed-off-by: Dmitry Torokhov --- include/linux/gpio_keys.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/gpio_keys.h b/include/linux/gpio_keys.h index ec6ecd74781d..1289fa7623ca 100644 --- a/include/linux/gpio_keys.h +++ b/include/linux/gpio_keys.h @@ -15,6 +15,7 @@ struct gpio_keys_button { struct gpio_keys_platform_data { struct gpio_keys_button *buttons; int nbuttons; + unsigned int rep:1; /* enable input subsystem auto repeat */ }; #endif -- cgit v1.2.3 From 5e1f8c9e20a92743eefc9a82c2db835213905e26 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Tue, 28 Oct 2008 13:21:55 -0400 Subject: ext3: Add support for non-native signed/unsigned htree hash algorithms The original ext3 hash algorithms assumed that variables of type char were signed, as God and K&R intended. Unfortunately, this assumption is not true on some architectures. Userspace support for marking filesystems with non-native signed/unsigned chars was added two years ago, but the kernel-side support was never added (until now). Signed-off-by: "Theodore Ts'o" Cc: akpm@linux-foundation.org Cc: linux-kernel@vger.kernel.org --- fs/ext3/hash.c | 77 ++++++++++++++++++++++++++++++++++++++++------ fs/ext3/namei.c | 7 +++++ fs/ext3/super.c | 12 ++++++++ include/linux/ext3_fs.h | 28 ++++++++++++++++- include/linux/ext3_fs_sb.h | 1 + 5 files changed, 114 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/fs/ext3/hash.c b/fs/ext3/hash.c index c30e149fbd2e..7d215b4d4f2e 100644 --- a/fs/ext3/hash.c +++ b/fs/ext3/hash.c @@ -35,23 +35,71 @@ static void TEA_transform(__u32 buf[4], __u32 const in[]) /* The old legacy hash */ -static __u32 dx_hack_hash (const char *name, int len) +static __u32 dx_hack_hash_unsigned(const char *name, int len) { - __u32 hash0 = 0x12a3fe2d, hash1 = 0x37abe8f9; + __u32 hash, hash0 = 0x12a3fe2d, hash1 = 0x37abe8f9; + const unsigned char *ucp = (const unsigned char *) name; + + while (len--) { + hash = hash1 + (hash0 ^ (((int) *ucp++) * 7152373)); + + if (hash & 0x80000000) + hash -= 0x7fffffff; + hash1 = hash0; + hash0 = hash; + } + return hash0 << 1; +} + +static __u32 dx_hack_hash_signed(const char *name, int len) +{ + __u32 hash, hash0 = 0x12a3fe2d, hash1 = 0x37abe8f9; + const signed char *scp = (const signed char *) name; + while (len--) { - __u32 hash = hash1 + (hash0 ^ (*name++ * 7152373)); + hash = hash1 + (hash0 ^ (((int) *scp++) * 7152373)); - if (hash & 0x80000000) hash -= 0x7fffffff; + if (hash & 0x80000000) + hash -= 0x7fffffff; hash1 = hash0; hash0 = hash; } - return (hash0 << 1); + return hash0 << 1; } -static void str2hashbuf(const char *msg, int len, __u32 *buf, int num) +static void str2hashbuf_signed(const char *msg, int len, __u32 *buf, int num) { __u32 pad, val; int i; + const signed char *scp = (const signed char *) msg; + + pad = (__u32)len | ((__u32)len << 8); + pad |= pad << 16; + + val = pad; + if (len > num*4) + len = num * 4; + for (i = 0; i < len; i++) { + if ((i % 4) == 0) + val = pad; + val = ((int) scp[i]) + (val << 8); + if ((i % 4) == 3) { + *buf++ = val; + val = pad; + num--; + } + } + if (--num >= 0) + *buf++ = val; + while (--num >= 0) + *buf++ = pad; +} + +static void str2hashbuf_unsigned(const char *msg, int len, __u32 *buf, int num) +{ + __u32 pad, val; + int i; + const unsigned char *ucp = (const unsigned char *) msg; pad = (__u32)len | ((__u32)len << 8); pad |= pad << 16; @@ -62,7 +110,7 @@ static void str2hashbuf(const char *msg, int len, __u32 *buf, int num) for (i=0; i < len; i++) { if ((i % 4) == 0) val = pad; - val = msg[i] + (val << 8); + val = ((int) ucp[i]) + (val << 8); if ((i % 4) == 3) { *buf++ = val; val = pad; @@ -95,6 +143,8 @@ int ext3fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo) const char *p; int i; __u32 in[8], buf[4]; + void (*str2hashbuf)(const char *, int, __u32 *, int) = + str2hashbuf_signed; /* Initialize the default seed for the hash checksum functions */ buf[0] = 0x67452301; @@ -113,13 +163,18 @@ int ext3fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo) } switch (hinfo->hash_version) { + case DX_HASH_LEGACY_UNSIGNED: + hash = dx_hack_hash_unsigned(name, len); + break; case DX_HASH_LEGACY: - hash = dx_hack_hash(name, len); + hash = dx_hack_hash_signed(name, len); break; + case DX_HASH_HALF_MD4_UNSIGNED: + str2hashbuf = str2hashbuf_unsigned; case DX_HASH_HALF_MD4: p = name; while (len > 0) { - str2hashbuf(p, len, in, 8); + (*str2hashbuf)(p, len, in, 8); half_md4_transform(buf, in); len -= 32; p += 32; @@ -127,10 +182,12 @@ int ext3fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo) minor_hash = buf[2]; hash = buf[1]; break; + case DX_HASH_TEA_UNSIGNED: + str2hashbuf = str2hashbuf_unsigned; case DX_HASH_TEA: p = name; while (len > 0) { - str2hashbuf(p, len, in, 4); + (*str2hashbuf)(p, len, in, 4); TEA_transform(buf, in); len -= 16; p += 16; diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index 1dd2abe6313e..287b304d42a7 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -368,6 +368,8 @@ dx_probe(struct qstr *entry, struct inode *dir, goto fail; } hinfo->hash_version = root->info.hash_version; + if (hinfo->hash_version <= DX_HASH_TEA) + hinfo->hash_version += EXT3_SB(dir->i_sb)->s_hash_unsigned; hinfo->seed = EXT3_SB(dir->i_sb)->s_hash_seed; if (entry) ext3fs_dirhash(entry->name, entry->len, hinfo); @@ -636,6 +638,9 @@ int ext3_htree_fill_tree(struct file *dir_file, __u32 start_hash, dir = dir_file->f_path.dentry->d_inode; if (!(EXT3_I(dir)->i_flags & EXT3_INDEX_FL)) { hinfo.hash_version = EXT3_SB(dir->i_sb)->s_def_hash_version; + if (hinfo.hash_version <= DX_HASH_TEA) + hinfo.hash_version += + EXT3_SB(dir->i_sb)->s_hash_unsigned; hinfo.seed = EXT3_SB(dir->i_sb)->s_hash_seed; count = htree_dirblock_to_tree(dir_file, dir, 0, &hinfo, start_hash, start_minor_hash); @@ -1398,6 +1403,8 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, /* Initialize as for dx_probe */ hinfo.hash_version = root->info.hash_version; + if (hinfo.hash_version <= DX_HASH_TEA) + hinfo.hash_version += EXT3_SB(dir->i_sb)->s_hash_unsigned; hinfo.seed = EXT3_SB(dir->i_sb)->s_hash_seed; ext3fs_dirhash(name, namelen, &hinfo); frame = frames; diff --git a/fs/ext3/super.c b/fs/ext3/super.c index f6c94f232ec1..541d5e4f7f6e 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -1744,6 +1744,18 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) for (i=0; i < 4; i++) sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]); sbi->s_def_hash_version = es->s_def_hash_version; + i = le32_to_cpu(es->s_flags); + if (i & EXT2_FLAGS_UNSIGNED_HASH) + sbi->s_hash_unsigned = 3; + else if ((i & EXT2_FLAGS_SIGNED_HASH) == 0) { +#ifdef __CHAR_UNSIGNED__ + es->s_flags |= cpu_to_le32(EXT2_FLAGS_UNSIGNED_HASH); + sbi->s_hash_unsigned = 3; +#else + es->s_flags |= cpu_to_le32(EXT2_FLAGS_SIGNED_HASH); +#endif + sb->s_dirt = 1; + } if (sbi->s_blocks_per_group > blocksize * 8) { printk (KERN_ERR diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h index d14f02918483..9004794a35fe 100644 --- a/include/linux/ext3_fs.h +++ b/include/linux/ext3_fs.h @@ -353,6 +353,13 @@ struct ext3_inode { #define EXT3_ERROR_FS 0x0002 /* Errors detected */ #define EXT3_ORPHAN_FS 0x0004 /* Orphans being recovered */ +/* + * Misc. filesystem flags + */ +#define EXT2_FLAGS_SIGNED_HASH 0x0001 /* Signed dirhash in use */ +#define EXT2_FLAGS_UNSIGNED_HASH 0x0002 /* Unsigned dirhash in use */ +#define EXT2_FLAGS_TEST_FILESYS 0x0004 /* to test development code */ + /* * Mount flags */ @@ -489,7 +496,23 @@ struct ext3_super_block { __u16 s_reserved_word_pad; __le32 s_default_mount_opts; __le32 s_first_meta_bg; /* First metablock block group */ - __u32 s_reserved[190]; /* Padding to the end of the block */ + __le32 s_mkfs_time; /* When the filesystem was created */ + __le32 s_jnl_blocks[17]; /* Backup of the journal inode */ + /* 64bit support valid if EXT4_FEATURE_COMPAT_64BIT */ +/*150*/ __le32 s_blocks_count_hi; /* Blocks count */ + __le32 s_r_blocks_count_hi; /* Reserved blocks count */ + __le32 s_free_blocks_count_hi; /* Free blocks count */ + __le16 s_min_extra_isize; /* All inodes have at least # bytes */ + __le16 s_want_extra_isize; /* New inodes should reserve # bytes */ + __le32 s_flags; /* Miscellaneous flags */ + __le16 s_raid_stride; /* RAID stride */ + __le16 s_mmp_interval; /* # seconds to wait in MMP checking */ + __le64 s_mmp_block; /* Block for multi-mount protection */ + __le32 s_raid_stripe_width; /* blocks on all data disks (N*stride)*/ + __u8 s_log_groups_per_flex; /* FLEX_BG group size */ + __u8 s_reserved_char_pad2; + __le16 s_reserved_pad; + __u32 s_reserved[162]; /* Padding to the end of the block */ }; #ifdef __KERNEL__ @@ -694,6 +717,9 @@ static inline __le16 ext3_rec_len_to_disk(unsigned len) #define DX_HASH_LEGACY 0 #define DX_HASH_HALF_MD4 1 #define DX_HASH_TEA 2 +#define DX_HASH_LEGACY_UNSIGNED 3 +#define DX_HASH_HALF_MD4_UNSIGNED 4 +#define DX_HASH_TEA_UNSIGNED 5 #ifdef __KERNEL__ diff --git a/include/linux/ext3_fs_sb.h b/include/linux/ext3_fs_sb.h index e024e38248ff..a4e9216b3a6d 100644 --- a/include/linux/ext3_fs_sb.h +++ b/include/linux/ext3_fs_sb.h @@ -57,6 +57,7 @@ struct ext3_sb_info { u32 s_next_generation; u32 s_hash_seed[4]; int s_def_hash_version; + int s_hash_unsigned; /* 3 if hash should be signed, 0 if not */ struct percpu_counter s_freeblocks_counter; struct percpu_counter s_freeinodes_counter; struct percpu_counter s_dirs_counter; -- cgit v1.2.3 From a20c7ab570ffdce1d6f67c7acf8c1c502a3b3839 Mon Sep 17 00:00:00 2001 From: Dmitry Baryshkov Date: Thu, 16 Oct 2008 18:43:48 +0400 Subject: [MTD] sharpsl-nand: use platform_data for model-specific values Add platform_data which holds all model-specific values, like badblocks pattern, oobinfo, partitions. Signed-off-by: Dmitry Baryshkov --- drivers/mtd/nand/sharpsl.c | 158 +++++++++++++++++++++++--------------------- include/linux/mtd/sharpsl.h | 20 ++++++ 2 files changed, 104 insertions(+), 74 deletions(-) create mode 100644 include/linux/mtd/sharpsl.h (limited to 'include/linux') diff --git a/drivers/mtd/nand/sharpsl.c b/drivers/mtd/nand/sharpsl.c index 17625c0a8f61..698378ca8e0a 100644 --- a/drivers/mtd/nand/sharpsl.c +++ b/drivers/mtd/nand/sharpsl.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include @@ -53,31 +54,6 @@ struct sharpsl_nand { #define FLCLE (1 << 1) #define FLCE0 (1 << 0) -#ifdef CONFIG_MTD_PARTITIONS -/* - * Define partitions for flash device - */ -#define DEFAULT_NUM_PARTITIONS 3 - -static struct mtd_partition sharpsl_nand_default_partition_info[] = { - { - .name = "System Area", - .offset = 0, - .size = 7 * 1024 * 1024, - }, - { - .name = "Root Filesystem", - .offset = 7 * 1024 * 1024, - .size = 30 * 1024 * 1024, - }, - { - .name = "Home Filesystem", - .offset = MTDPART_OFS_APPEND, - .size = MTDPART_SIZ_FULL, - }, -}; -#endif - /* * hardware specific access to control-lines * ctrl: @@ -106,31 +82,6 @@ static void sharpsl_nand_hwcontrol(struct mtd_info *mtd, int cmd, writeb(cmd, chip->IO_ADDR_W); } -static uint8_t scan_ff_pattern[] = { 0xff, 0xff }; - -static struct nand_bbt_descr sharpsl_bbt = { - .options = 0, - .offs = 4, - .len = 2, - .pattern = scan_ff_pattern -}; - -static struct nand_bbt_descr sharpsl_akita_bbt = { - .options = 0, - .offs = 4, - .len = 1, - .pattern = scan_ff_pattern -}; - -static struct nand_ecclayout akita_oobinfo = { - .eccbytes = 24, - .eccpos = { - 0x5, 0x1, 0x2, 0x3, 0x6, 0x7, 0x15, 0x11, - 0x12, 0x13, 0x16, 0x17, 0x25, 0x21, 0x22, 0x23, - 0x26, 0x27, 0x35, 0x31, 0x32, 0x33, 0x36, 0x37}, - .oobfree = {{0x08, 0x09}} -}; - static int sharpsl_nand_dev_ready(struct mtd_info *mtd) { struct sharpsl_nand *sharpsl = mtd_to_sharpsl(mtd); @@ -169,6 +120,12 @@ static int __devinit sharpsl_nand_probe(struct platform_device *pdev) struct resource *r; int err = 0; struct sharpsl_nand *sharpsl; + struct sharpsl_nand_platform_data *data = pdev->dev.platform_data; + + if (!data) { + dev_err(&pdev->dev, "no platform data!\n"); + return -EINVAL; + } /* Allocate memory for MTD device structure and private data */ sharpsl = kzalloc(sizeof(struct sharpsl_nand), GFP_KERNEL); @@ -218,11 +175,8 @@ static int __devinit sharpsl_nand_probe(struct platform_device *pdev) this->ecc.mode = NAND_ECC_HW; this->ecc.size = 256; this->ecc.bytes = 3; - this->badblock_pattern = &sharpsl_bbt; - if (machine_is_akita() || machine_is_borzoi()) { - this->badblock_pattern = &sharpsl_akita_bbt; - this->ecc.layout = &akita_oobinfo; - } + this->badblock_pattern = data->badblock_pattern; + this->ecc.layout = data->ecc_layout; this->ecc.hwctl = sharpsl_nand_enable_hwecc; this->ecc.calculate = sharpsl_nand_calculate_ecc; this->ecc.correct = nand_correct_data; @@ -236,29 +190,16 @@ static int __devinit sharpsl_nand_probe(struct platform_device *pdev) sharpsl->mtd.name = "sharpsl-nand"; #ifdef CONFIG_MTD_PARTITIONS nr_partitions = parse_mtd_partitions(&sharpsl->mtd, part_probes, &sharpsl_partition_info, 0); - if (nr_partitions <= 0) { - nr_partitions = ARRAY_SIZE(sharpsl_nand_default_partition_info); - sharpsl_partition_info = sharpsl_nand_default_partition_info; - if (machine_is_poodle()) { - sharpsl_partition_info[1].size = 22 * 1024 * 1024; - } else if (machine_is_corgi() || machine_is_shepherd()) { - sharpsl_partition_info[1].size = 25 * 1024 * 1024; - } else if (machine_is_husky()) { - sharpsl_partition_info[1].size = 53 * 1024 * 1024; - } else if (machine_is_spitz()) { - sharpsl_partition_info[1].size = 5 * 1024 * 1024; - } else if (machine_is_akita()) { - sharpsl_partition_info[1].size = 58 * 1024 * 1024; - } else if (machine_is_borzoi()) { - sharpsl_partition_info[1].size = 32 * 1024 * 1024; - } + nr_partitions = data->nr_partitions; + sharpsl_partition_info = data->partitions; } - err = add_mtd_partitions(&sharpsl->mtd, sharpsl_partition_info, nr_partitions); -#else - err = add_mtd_device(&sharpsl->mtd); + if (nr_partitions > 0) + err = add_mtd_partitions(&sharpsl->mtd, sharpsl_partition_info, nr_partitions); + else #endif + err = add_mtd_device(&sharpsl->mtd); if (err) goto err_add; @@ -306,6 +247,58 @@ static struct platform_driver sharpsl_nand_driver = { .remove = __devexit_p(sharpsl_nand_remove), }; +/* + * Define partitions for flash device + */ +static struct mtd_partition sharpsl_nand_partitions[] = { + { + .name = "System Area", + .offset = 0, + .size = 7 * 1024 * 1024, + }, + { + .name = "Root Filesystem", + .offset = 7 * 1024 * 1024, + .size = 30 * 1024 * 1024, + }, + { + .name = "Home Filesystem", + .offset = MTDPART_OFS_APPEND, + .size = MTDPART_SIZ_FULL, + }, +}; + +static uint8_t scan_ff_pattern[] = { 0xff, 0xff }; + +static struct nand_bbt_descr sharpsl_bbt = { + .options = 0, + .offs = 4, + .len = 2, + .pattern = scan_ff_pattern +}; + +static struct nand_bbt_descr sharpsl_akita_bbt = { + .options = 0, + .offs = 4, + .len = 1, + .pattern = scan_ff_pattern +}; + +static struct nand_ecclayout akita_oobinfo = { + .eccbytes = 24, + .eccpos = { + 0x5, 0x1, 0x2, 0x3, 0x6, 0x7, 0x15, 0x11, + 0x12, 0x13, 0x16, 0x17, 0x25, 0x21, 0x22, 0x23, + 0x26, 0x27, 0x35, 0x31, 0x32, 0x33, 0x36, 0x37}, + .oobfree = {{0x08, 0x09}} +}; + +static struct sharpsl_nand_platform_data sharpsl_nand_platform_data = { + .badblock_pattern = &sharpsl_bbt, + .partitions = sharpsl_nand_partitions, + .nr_partitions = ARRAY_SIZE(sharpsl_nand_partitions), +}; + static struct resource sharpsl_nand_resources[] = { { .start = 0x0C000000, @@ -319,10 +312,27 @@ static struct platform_device sharpsl_nand_device = { .id = -1, .resource = sharpsl_nand_resources, .num_resources = ARRAY_SIZE(sharpsl_nand_resources), + .dev.platform_data = &sharpsl_nand_platform_data, }; static int __init sharpsl_nand_init(void) { + if (machine_is_poodle()) { + sharpsl_nand_partitions[1].size = 22 * 1024 * 1024; + } else if (machine_is_corgi() || machine_is_shepherd()) { + sharpsl_nand_partitions[1].size = 25 * 1024 * 1024; + } else if (machine_is_husky()) { + sharpsl_nand_partitions[1].size = 53 * 1024 * 1024; + } else if (machine_is_spitz()) { + sharpsl_nand_partitions[1].size = 5 * 1024 * 1024; + } else if (machine_is_akita()) { + sharpsl_nand_partitions[1].size = 58 * 1024 * 1024; + sharpsl_nand_platform_data.badblock_pattern = &sharpsl_akita_bbt; + sharpsl_nand_platform_data.ecc_layout = &akita_oobinfo; + } else if (machine_is_borzoi()) { + sharpsl_nand_partitions[1].size = 32 * 1024 * 1024; + } + platform_device_register(&sharpsl_nand_device); return platform_driver_register(&sharpsl_nand_driver); } diff --git a/include/linux/mtd/sharpsl.h b/include/linux/mtd/sharpsl.h new file mode 100644 index 000000000000..25f4d2a845c1 --- /dev/null +++ b/include/linux/mtd/sharpsl.h @@ -0,0 +1,20 @@ +/* + * SharpSL NAND support + * + * Copyright (C) 2008 Dmitry Baryshkov + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include + +struct sharpsl_nand_platform_data { + struct nand_bbt_descr *badblock_pattern; + struct nand_ecclayout *ecc_layout; + struct mtd_partition *partitions; + unsigned int nr_partitions; +}; -- cgit v1.2.3 From 171bbfbeab7730031eec8025341401fabe540bd5 Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Tue, 25 Nov 2008 17:42:31 -0500 Subject: jbd2: Add BH_JBDPrivateStart Add this so that file systems using JBD2 can safely allocate unused b_state bits. In this case, we add it so that Ocfs2 can define a single bit for tracking the validation state of a buffer. Signed-off-by: Mark Fasheh Signed-off-by: "Theodore Ts'o" --- include/linux/jbd2.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index c7d106ef22e2..f36645745489 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -329,6 +329,7 @@ enum jbd_state_bits { BH_State, /* Pins most journal_head state */ BH_JournalHead, /* Pins bh->b_private and jh->b_bh */ BH_Unshadow, /* Dummy bit, for BJ_Shadow wakeup filtering */ + BH_JBDPrivateStart, /* First bit available for private use by FS */ }; BUFFER_FNS(JBD, jbd) -- cgit v1.2.3 From e07f7183a486cf9783d1f8c9d2997b5b39eeb2d4 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 26 Nov 2008 01:14:26 -0500 Subject: jbd2: improve jbd2 fsync batching This patch removes the static sleep time in favor of a more self optimizing approach where we measure the average amount of time it takes to commit a transaction to disk and the ammount of time a transaction has been running. If somebody does a sync write or an fsync() traditionally we would sleep for 1 jiffies, which depending on the value of HZ could be a significant amount of time compared to how long it takes to commit a transaction to the underlying storage. With this patch instead of sleeping for a jiffie, we check to see if the amount of time this transaction has been running is less than the average commit time, and if it is we sleep for the delta using schedule_hrtimeout to give us a higher precision sleep time. This greatly benefits high end storage where you could end up sleeping for longer than it takes to commit the transaction and therefore sitting idle instead of allowing the transaction to be committed by keeping the sleep time to a minimum so you are sure to always be doing something. Signed-off-by: Josef Bacik Signed-off-by: "Theodore Ts'o" --- fs/jbd2/commit.c | 14 +++++++++++++ fs/jbd2/transaction.c | 58 ++++++++++++++++++++++++++++++++++++++------------- include/linux/jbd2.h | 15 +++++++++++++ 3 files changed, 73 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 6393fd0d804e..f22d1828ea85 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -355,6 +355,8 @@ void jbd2_journal_commit_transaction(journal_t *journal) int flags; int err; unsigned long long blocknr; + ktime_t start_time; + u64 commit_time; char *tagp = NULL; journal_header_t *header; journal_block_tag_t *tag = NULL; @@ -481,6 +483,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) commit_transaction->t_state = T_FLUSH; journal->j_committing_transaction = commit_transaction; journal->j_running_transaction = NULL; + start_time = ktime_get(); commit_transaction->t_log_start = journal->j_head; wake_up(&journal->j_wait_transaction_locked); spin_unlock(&journal->j_state_lock); @@ -995,6 +998,17 @@ restart_loop: J_ASSERT(commit_transaction == journal->j_committing_transaction); journal->j_commit_sequence = commit_transaction->t_tid; journal->j_committing_transaction = NULL; + commit_time = ktime_to_ns(ktime_sub(ktime_get(), start_time)); + + /* + * weight the commit time higher than the average time so we don't + * react too strongly to vast changes in the commit time + */ + if (likely(journal->j_average_commit_time)) + journal->j_average_commit_time = (commit_time + + journal->j_average_commit_time*3) / 4; + else + journal->j_average_commit_time = commit_time; spin_unlock(&journal->j_state_lock); if (journal->j_commit_callback) diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 39b7805a599a..13dcbc990f41 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -25,6 +25,7 @@ #include #include #include +#include static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh); @@ -48,6 +49,7 @@ jbd2_get_transaction(journal_t *journal, transaction_t *transaction) { transaction->t_journal = journal; transaction->t_state = T_RUNNING; + transaction->t_start_time = ktime_get(); transaction->t_tid = journal->j_transaction_sequence++; transaction->t_expires = jiffies + journal->j_commit_interval; spin_lock_init(&transaction->t_handle_lock); @@ -1193,7 +1195,7 @@ int jbd2_journal_stop(handle_t *handle) { transaction_t *transaction = handle->h_transaction; journal_t *journal = transaction->t_journal; - int old_handle_count, err; + int err; pid_t pid; J_ASSERT(journal_current_handle() == handle); @@ -1216,24 +1218,52 @@ int jbd2_journal_stop(handle_t *handle) /* * Implement synchronous transaction batching. If the handle * was synchronous, don't force a commit immediately. Let's - * yield and let another thread piggyback onto this transaction. - * Keep doing that while new threads continue to arrive. - * It doesn't cost much - we're about to run a commit and sleep - * on IO anyway. Speeds up many-threaded, many-dir operations - * by 30x or more... + * yield and let another thread piggyback onto this + * transaction. Keep doing that while new threads continue to + * arrive. It doesn't cost much - we're about to run a commit + * and sleep on IO anyway. Speeds up many-threaded, many-dir + * operations by 30x or more... + * + * We try and optimize the sleep time against what the + * underlying disk can do, instead of having a static sleep + * time. This is useful for the case where our storage is so + * fast that it is more optimal to go ahead and force a flush + * and wait for the transaction to be committed than it is to + * wait for an arbitrary amount of time for new writers to + * join the transaction. We achieve this by measuring how + * long it takes to commit a transaction, and compare it with + * how long this transaction has been running, and if run time + * < commit time then we sleep for the delta and commit. This + * greatly helps super fast disks that would see slowdowns as + * more threads started doing fsyncs. * - * But don't do this if this process was the most recent one to - * perform a synchronous write. We do this to detect the case where a - * single process is doing a stream of sync writes. No point in waiting - * for joiners in that case. + * But don't do this if this process was the most recent one + * to perform a synchronous write. We do this to detect the + * case where a single process is doing a stream of sync + * writes. No point in waiting for joiners in that case. */ pid = current->pid; if (handle->h_sync && journal->j_last_sync_writer != pid) { + u64 commit_time, trans_time; + journal->j_last_sync_writer = pid; - do { - old_handle_count = transaction->t_handle_count; - schedule_timeout_uninterruptible(1); - } while (old_handle_count != transaction->t_handle_count); + + spin_lock(&journal->j_state_lock); + commit_time = journal->j_average_commit_time; + spin_unlock(&journal->j_state_lock); + + trans_time = ktime_to_ns(ktime_sub(ktime_get(), + transaction->t_start_time)); + + commit_time = min_t(u64, commit_time, + 1000*jiffies_to_usecs(1)); + + if (trans_time < commit_time) { + ktime_t expires = ktime_add_ns(ktime_get(), + commit_time); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_hrtimeout(&expires, HRTIMER_MODE_ABS); + } } current->journal_info = NULL; diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index f36645745489..ab8cef130c28 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -637,6 +637,11 @@ struct transaction_s */ unsigned long t_expires; + /* + * When this transaction started, in nanoseconds [no locking] + */ + ktime_t t_start_time; + /* * How many handles used this transaction? [t_handle_lock] */ @@ -939,8 +944,18 @@ struct journal_s struct buffer_head **j_wbuf; int j_wbufsize; + /* + * this is the pid of hte last person to run a synchronous operation + * through the journal + */ pid_t j_last_sync_writer; + /* + * the average amount of time in nanoseconds it takes to commit a + * transaction to disk. [j_state_lock] + */ + u64 j_average_commit_time; + /* This function is called when a transaction is closed */ void (*j_commit_callback)(journal_t *, transaction_t *); -- cgit v1.2.3 From 30773840c19cea60dcef39545960d541b1ac1cf8 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Sat, 3 Jan 2009 20:27:38 -0500 Subject: ext4: add fsync batch tuning knobs Add new mount options, min_batch_time and max_batch_time, which controls how long the jbd2 layer should wait for additional filesystem operations to get batched with a synchronous write transaction. Signed-off-by: "Theodore Ts'o" --- Documentation/filesystems/ext4.txt | 29 +++++++++++++++++++++++ fs/ext4/ext4.h | 7 ++++++ fs/ext4/ext4_sb.h | 2 ++ fs/ext4/super.c | 47 ++++++++++++++++++++++++++++++++------ fs/jbd2/journal.c | 2 ++ fs/jbd2/transaction.c | 4 +++- include/linux/jbd2.h | 8 +++++++ 7 files changed, 91 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt index f75ab101c00a..e3fcbea3ec8c 100644 --- a/Documentation/filesystems/ext4.txt +++ b/Documentation/filesystems/ext4.txt @@ -283,6 +283,35 @@ delalloc (*) Deferring block allocation until write-out time. nodelalloc Disable delayed allocation. Blocks are allocation when data is copied from user to page cache. +max_batch_time=usec Maximum amount of time ext4 should wait for + additional filesystem operations to be batch + together with a synchronous write operation. + Since a synchronous write operation is going to + force a commit and then a wait for the I/O + complete, it doesn't cost much, and can be a + huge throughput win, we wait for a small amount + of time to see if any other transactions can + piggyback on the synchronous write. The + algorithm used is designed to automatically tune + for the speed of the disk, by measuring the + amount of time (on average) that it takes to + finish committing a transaction. Call this time + the "commit time". If the time that the + transactoin has been running is less than the + commit time, ext4 will try sleeping for the + commit time to see if other operations will join + the transaction. The commit time is capped by + the max_batch_time, which defaults to 15000us + (15ms). This optimization can be turned off + entirely by setting max_batch_time to 0. + +min_batch_time=usec This parameter sets the commit time (as + described above) to be at least min_batch_time. + It defaults to zero microseconds. Increasing + this parameter may improve the throughput of + multi-threaded, synchronous workloads on very + fast disks, at the cost of increasing latency. + Data Mode ========= There are 3 different data modes: diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index ac8551e0b70a..9ba9fd6d14da 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -328,6 +328,7 @@ struct ext4_mount_options { uid_t s_resuid; gid_t s_resgid; unsigned long s_commit_interval; + u32 s_min_batch_time, s_max_batch_time; #ifdef CONFIG_QUOTA int s_jquota_fmt; char *s_qf_names[MAXQUOTAS]; @@ -805,6 +806,12 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino) #define EXT4_DEFM_JMODE_ORDERED 0x0040 #define EXT4_DEFM_JMODE_WBACK 0x0060 +/* + * Default journal batch times + */ +#define EXT4_DEF_MIN_BATCH_TIME 0 +#define EXT4_DEF_MAX_BATCH_TIME 15000 /* 15ms */ + /* * Structure of a directory entry */ diff --git a/fs/ext4/ext4_sb.h b/fs/ext4/ext4_sb.h index 3db800f399a6..039b6ea1a042 100644 --- a/fs/ext4/ext4_sb.h +++ b/fs/ext4/ext4_sb.h @@ -74,6 +74,8 @@ struct ext4_sb_info { struct journal_s *s_journal; struct list_head s_orphan; unsigned long s_commit_interval; + u32 s_max_batch_time; + u32 s_min_batch_time; struct block_device *journal_bdev; #ifdef CONFIG_JBD2_DEBUG struct timer_list turn_ro_timer; /* For turning read-only (crash simulation) */ diff --git a/fs/ext4/super.c b/fs/ext4/super.c index dc27d4c613c0..da377f9521bb 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -705,10 +705,19 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) #endif if (!test_opt(sb, RESERVATION)) seq_puts(seq, ",noreservation"); - if (sbi->s_commit_interval) { + if (sbi->s_commit_interval != JBD2_DEFAULT_MAX_COMMIT_AGE*HZ) { seq_printf(seq, ",commit=%u", (unsigned) (sbi->s_commit_interval / HZ)); } + if (sbi->s_min_batch_time != EXT4_DEF_MIN_BATCH_TIME) { + seq_printf(seq, ",min_batch_time=%u", + (unsigned) sbi->s_min_batch_time); + } + if (sbi->s_max_batch_time != EXT4_DEF_MAX_BATCH_TIME) { + seq_printf(seq, ",max_batch_time=%u", + (unsigned) sbi->s_min_batch_time); + } + /* * We're changing the default of barrier mount option, so * let's always display its mount state so it's clear what its @@ -874,7 +883,8 @@ enum { Opt_nouid32, Opt_debug, Opt_oldalloc, Opt_orlov, Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl, Opt_reservation, Opt_noreservation, Opt_noload, Opt_nobh, Opt_bh, - Opt_commit, Opt_journal_update, Opt_journal_inum, Opt_journal_dev, + Opt_commit, Opt_min_batch_time, Opt_max_batch_time, + Opt_journal_update, Opt_journal_inum, Opt_journal_dev, Opt_journal_checksum, Opt_journal_async_commit, Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback, Opt_data_err_abort, Opt_data_err_ignore, @@ -913,6 +923,8 @@ static const match_table_t tokens = { {Opt_nobh, "nobh"}, {Opt_bh, "bh"}, {Opt_commit, "commit=%u"}, + {Opt_min_batch_time, "min_batch_time=%u"}, + {Opt_max_batch_time, "max_batch_time=%u"}, {Opt_journal_update, "journal=update"}, {Opt_journal_inum, "journal=%u"}, {Opt_journal_dev, "journal_dev=%u"}, @@ -1131,6 +1143,22 @@ static int parse_options(char *options, struct super_block *sb, option = JBD2_DEFAULT_MAX_COMMIT_AGE; sbi->s_commit_interval = HZ * option; break; + case Opt_max_batch_time: + if (match_int(&args[0], &option)) + return 0; + if (option < 0) + return 0; + if (option == 0) + option = EXT4_DEF_MAX_BATCH_TIME; + sbi->s_max_batch_time = option; + break; + case Opt_min_batch_time: + if (match_int(&args[0], &option)) + return 0; + if (option < 0) + return 0; + sbi->s_min_batch_time = option; + break; case Opt_data_journal: data_opt = EXT4_MOUNT_JOURNAL_DATA; goto datacheck; @@ -1979,6 +2007,9 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) sbi->s_resuid = le16_to_cpu(es->s_def_resuid); sbi->s_resgid = le16_to_cpu(es->s_def_resgid); + sbi->s_commit_interval = JBD2_DEFAULT_MAX_COMMIT_AGE * HZ; + sbi->s_min_batch_time = EXT4_DEF_MIN_BATCH_TIME; + sbi->s_max_batch_time = EXT4_DEF_MAX_BATCH_TIME; set_opt(sbi->s_mount_opt, RESERVATION); set_opt(sbi->s_mount_opt, BARRIER); @@ -2524,11 +2555,9 @@ static void ext4_init_journal_params(struct super_block *sb, journal_t *journal) { struct ext4_sb_info *sbi = EXT4_SB(sb); - if (sbi->s_commit_interval) - journal->j_commit_interval = sbi->s_commit_interval; - /* We could also set up an ext4-specific default for the commit - * interval here, but for now we'll just fall back to the jbd - * default. */ + journal->j_commit_interval = sbi->s_commit_interval; + journal->j_min_batch_time = sbi->s_min_batch_time; + journal->j_max_batch_time = sbi->s_max_batch_time; spin_lock(&journal->j_state_lock); if (test_opt(sb, BARRIER)) @@ -3042,6 +3071,8 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data) old_opts.s_resuid = sbi->s_resuid; old_opts.s_resgid = sbi->s_resgid; old_opts.s_commit_interval = sbi->s_commit_interval; + old_opts.s_min_batch_time = sbi->s_min_batch_time; + old_opts.s_max_batch_time = sbi->s_max_batch_time; #ifdef CONFIG_QUOTA old_opts.s_jquota_fmt = sbi->s_jquota_fmt; for (i = 0; i < MAXQUOTAS; i++) @@ -3178,6 +3209,8 @@ restore_opts: sbi->s_resuid = old_opts.s_resuid; sbi->s_resgid = old_opts.s_resgid; sbi->s_commit_interval = old_opts.s_commit_interval; + sbi->s_min_batch_time = old_opts.s_min_batch_time; + sbi->s_max_batch_time = old_opts.s_max_batch_time; #ifdef CONFIG_QUOTA sbi->s_jquota_fmt = old_opts.s_jquota_fmt; for (i = 0; i < MAXQUOTAS; i++) { diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 74d87290381c..fd1d7557a098 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -964,6 +964,8 @@ static journal_t * journal_init_common (void) spin_lock_init(&journal->j_state_lock); journal->j_commit_interval = (HZ * JBD2_DEFAULT_MAX_COMMIT_AGE); + journal->j_min_batch_time = 0; + journal->j_max_batch_time = 15000; /* 15ms */ /* The journal is marked for error until we succeed with recovery! */ journal->j_flags = JBD2_ABORT; diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 13dcbc990f41..48c21bac5a56 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -1255,8 +1255,10 @@ int jbd2_journal_stop(handle_t *handle) trans_time = ktime_to_ns(ktime_sub(ktime_get(), transaction->t_start_time)); + commit_time = max_t(u64, commit_time, + 1000*journal->j_min_batch_time); commit_time = min_t(u64, commit_time, - 1000*jiffies_to_usecs(1)); + 1000*journal->j_max_batch_time); if (trans_time < commit_time) { ktime_t expires = ktime_add_ns(ktime_get(), diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index ab8cef130c28..a3cd647ea1bc 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -956,6 +956,14 @@ struct journal_s */ u64 j_average_commit_time; + /* + * minimum and maximum times that we should wait for + * additional filesystem operations to get batched into a + * synchronous handle in microseconds + */ + u32 j_min_batch_time; + u32 j_max_batch_time; + /* This function is called when a transaction is closed */ void (*j_commit_callback)(journal_t *, transaction_t *); -- cgit v1.2.3 From 1a0d3786dd57dbd74f340322054c3d618b999dcf Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Wed, 5 Nov 2008 00:09:22 -0500 Subject: jbd2: Remove a large array of bh's from the stack of the checkpoint routine jbd2_log_do_checkpoint()n is one of the kernel's largest stack users. Move the array of buffer head's from the stack of jbd2_log_do_checkpoint() to the in-core journal structure. Signed-off-by: "Theodore Ts'o" --- fs/jbd2/checkpoint.c | 22 +++++++++------------- fs/jbd2/journal.c | 2 ++ include/linux/jbd2.h | 10 ++++++++++ 3 files changed, 21 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 9497718fe920..adc08ec875ed 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -249,16 +249,14 @@ restart: return ret; } -#define NR_BATCH 64 - static void -__flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count) +__flush_batch(journal_t *journal, int *batch_count) { int i; - ll_rw_block(SWRITE, *batch_count, bhs); + ll_rw_block(SWRITE, *batch_count, journal->j_chkpt_bhs); for (i = 0; i < *batch_count; i++) { - struct buffer_head *bh = bhs[i]; + struct buffer_head *bh = journal->j_chkpt_bhs[i]; clear_buffer_jwrite(bh); BUFFER_TRACE(bh, "brelse"); __brelse(bh); @@ -277,8 +275,7 @@ __flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count) * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it */ static int __process_buffer(journal_t *journal, struct journal_head *jh, - struct buffer_head **bhs, int *batch_count, - transaction_t *transaction) + int *batch_count, transaction_t *transaction) { struct buffer_head *bh = jh2bh(jh); int ret = 0; @@ -325,14 +322,14 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, get_bh(bh); J_ASSERT_BH(bh, !buffer_jwrite(bh)); set_buffer_jwrite(bh); - bhs[*batch_count] = bh; + journal->j_chkpt_bhs[*batch_count] = bh; __buffer_relink_io(jh); jbd_unlock_bh_state(bh); transaction->t_chp_stats.cs_written++; (*batch_count)++; - if (*batch_count == NR_BATCH) { + if (*batch_count == JBD2_NR_BATCH) { spin_unlock(&journal->j_list_lock); - __flush_batch(journal, bhs, batch_count); + __flush_batch(journal, batch_count); ret = 1; } } @@ -388,7 +385,6 @@ restart: if (journal->j_checkpoint_transactions == transaction && transaction->t_tid == this_tid) { int batch_count = 0; - struct buffer_head *bhs[NR_BATCH]; struct journal_head *jh; int retry = 0, err; @@ -402,7 +398,7 @@ restart: retry = 1; break; } - retry = __process_buffer(journal, jh, bhs, &batch_count, + retry = __process_buffer(journal, jh, &batch_count, transaction); if (retry < 0 && !result) result = retry; @@ -419,7 +415,7 @@ restart: spin_unlock(&journal->j_list_lock); retry = 1; } - __flush_batch(journal, bhs, &batch_count); + __flush_batch(journal, &batch_count); } if (retry) { diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index fd1d7557a098..34ef98057202 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1477,7 +1477,9 @@ int jbd2_journal_destroy(journal_t *journal) spin_lock(&journal->j_list_lock); while (journal->j_checkpoint_transactions != NULL) { spin_unlock(&journal->j_list_lock); + mutex_lock(&journal->j_checkpoint_mutex); jbd2_log_do_checkpoint(journal); + mutex_unlock(&journal->j_checkpoint_mutex); spin_lock(&journal->j_list_lock); } diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index a3cd647ea1bc..004c9a8d63ed 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -687,6 +687,8 @@ jbd2_time_diff(unsigned long start, unsigned long end) return end + (MAX_JIFFY_OFFSET - start); } +#define JBD2_NR_BATCH 64 + /** * struct journal_s - The journal_s type is the concrete type associated with * journal_t. @@ -830,6 +832,14 @@ struct journal_s /* Semaphore for locking against concurrent checkpoints */ struct mutex j_checkpoint_mutex; + /* + * List of buffer heads used by the checkpoint routine. This + * was moved from jbd2_log_do_checkpoint() to reduce stack + * usage. Access to this array is controlled by the + * j_checkpoint_mutex. [j_checkpoint_mutex] + */ + struct buffer_head *j_chkpt_bhs[JBD2_NR_BATCH]; + /* * Journal head: identifies the first unused block in the journal. * [j_state_lock] -- cgit v1.2.3 From fb68407b0d9efba962c03f55009c797e22f024bc Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Thu, 6 Nov 2008 17:50:21 -0500 Subject: jbd2: Call journal commit callback without holding j_list_lock Avoid freeing the transaction in __jbd2_journal_drop_transaction() so the journal commit callback can run without holding j_list_lock, to avoid lock contention on this spinlock. Signed-off-by: Aneesh Kumar K.V Signed-off-by: "Theodore Ts'o" --- fs/jbd2/checkpoint.c | 2 +- fs/jbd2/commit.c | 13 ++++++++----- include/linux/jbd2.h | 4 ++-- 3 files changed, 11 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index adc08ec875ed..17159cacbd9e 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -682,6 +682,7 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh) safely remove this transaction from the log */ __jbd2_journal_drop_transaction(journal, transaction); + kfree(transaction); /* Just in case anybody was waiting for more transactions to be checkpointed... */ @@ -756,5 +757,4 @@ void __jbd2_journal_drop_transaction(journal_t *journal, transaction_t *transact J_ASSERT(journal->j_running_transaction != transaction); jbd_debug(1, "Dropping transaction %d, all done\n", transaction->t_tid); - kfree(transaction); } diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index f22d1828ea85..0ad84162c425 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -363,7 +363,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) int space_left = 0; int first_tag = 0; int tag_flag; - int i; + int i, to_free = 0; int tag_bytes = journal_tag_bytes(journal); struct buffer_head *cbh = NULL; /* For transactional checksums */ __u32 crc32_sum = ~0; @@ -1011,12 +1011,10 @@ restart_loop: journal->j_average_commit_time = commit_time; spin_unlock(&journal->j_state_lock); - if (journal->j_commit_callback) - journal->j_commit_callback(journal, commit_transaction); - if (commit_transaction->t_checkpoint_list == NULL && commit_transaction->t_checkpoint_io_list == NULL) { __jbd2_journal_drop_transaction(journal, commit_transaction); + to_free = 1; } else { if (journal->j_checkpoint_transactions == NULL) { journal->j_checkpoint_transactions = commit_transaction; @@ -1035,11 +1033,16 @@ restart_loop: } spin_unlock(&journal->j_list_lock); + if (journal->j_commit_callback) + journal->j_commit_callback(journal, commit_transaction); + trace_mark(jbd2_end_commit, "dev %s transaction %d head %d", - journal->j_devname, journal->j_commit_sequence, + journal->j_devname, commit_transaction->t_tid, journal->j_tail_sequence); jbd_debug(1, "JBD: commit %d complete, head %d\n", journal->j_commit_sequence, journal->j_tail_sequence); + if (to_free) + kfree(commit_transaction); wake_up(&journal->j_wait_done_commit); } diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 004c9a8d63ed..9d82084a1605 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1179,8 +1179,8 @@ int jbd2_log_wait_commit(journal_t *journal, tid_t tid); int jbd2_log_do_checkpoint(journal_t *journal); void __jbd2_log_wait_for_space(journal_t *journal); -extern void __jbd2_journal_drop_transaction(journal_t *, transaction_t *); -extern int jbd2_cleanup_journal_tail(journal_t *); +extern void __jbd2_journal_drop_transaction(journal_t *, transaction_t *); +extern int jbd2_cleanup_journal_tail(journal_t *); /* Debugging code only: */ -- cgit v1.2.3 From 3eb1aa43ef5cb871ba3fb2f08633675eca374d2e Mon Sep 17 00:00:00 2001 From: Jaya Kumar Date: Wed, 19 Nov 2008 16:58:50 -0500 Subject: Input: add support for Wacom W8001 penabled serial touchscreen The Wacom W8001 sensor is a sensor device (uses electromagnetic resonance) and it is interfaced via its serial microcontroller to the host. Signed-off-by: Jaya Kumar Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/Kconfig | 13 ++ drivers/input/touchscreen/Makefile | 1 + drivers/input/touchscreen/wacom_w8001.c | 325 ++++++++++++++++++++++++++++++++ include/linux/serio.h | 1 + 4 files changed, 340 insertions(+) create mode 100644 drivers/input/touchscreen/wacom_w8001.c (limited to 'include/linux') diff --git a/drivers/input/touchscreen/Kconfig b/drivers/input/touchscreen/Kconfig index 3d1ab8fa9acc..20eb52ed176d 100644 --- a/drivers/input/touchscreen/Kconfig +++ b/drivers/input/touchscreen/Kconfig @@ -95,6 +95,19 @@ config TOUCHSCREEN_ELO To compile this driver as a module, choose M here: the module will be called elo. +config TOUCHSCREEN_WACOM_W8001 + tristate "Wacom W8001 penabled serial touchscreen" + select SERIO + help + Say Y here if you have an Wacom W8001 penabled serial touchscreen + connected to your system. + + If unsure, say N. + + To compile this driver as a module, choose M here: the + module will be called wacom_w8001. + + config TOUCHSCREEN_MTOUCH tristate "MicroTouch serial touchscreens" select SERIO diff --git a/drivers/input/touchscreen/Makefile b/drivers/input/touchscreen/Makefile index 15cf29079489..3dc84d3846c1 100644 --- a/drivers/input/touchscreen/Makefile +++ b/drivers/input/touchscreen/Makefile @@ -26,6 +26,7 @@ obj-$(CONFIG_TOUCHSCREEN_TOUCHIT213) += touchit213.o obj-$(CONFIG_TOUCHSCREEN_TOUCHRIGHT) += touchright.o obj-$(CONFIG_TOUCHSCREEN_TOUCHWIN) += touchwin.o obj-$(CONFIG_TOUCHSCREEN_UCB1400) += ucb1400_ts.o +obj-$(CONFIG_TOUCHSCREEN_WACOM_W8001) += wacom_w8001.o obj-$(CONFIG_TOUCHSCREEN_WM97XX) += wm97xx-ts.o wm97xx-ts-$(CONFIG_TOUCHSCREEN_WM9705) += wm9705.o wm97xx-ts-$(CONFIG_TOUCHSCREEN_WM9712) += wm9712.o diff --git a/drivers/input/touchscreen/wacom_w8001.c b/drivers/input/touchscreen/wacom_w8001.c new file mode 100644 index 000000000000..2f33a0167644 --- /dev/null +++ b/drivers/input/touchscreen/wacom_w8001.c @@ -0,0 +1,325 @@ +/* + * Wacom W8001 penabled serial touchscreen driver + * + * Copyright (c) 2008 Jaya Kumar + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file COPYING in the main directory of this archive for + * more details. + * + * Layout based on Elo serial touchscreen driver by Vojtech Pavlik + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#define DRIVER_DESC "Wacom W8001 serial touchscreen driver" + +MODULE_AUTHOR("Jaya Kumar "); +MODULE_DESCRIPTION(DRIVER_DESC); +MODULE_LICENSE("GPL"); + +/* + * Definitions & global arrays. + */ + +#define W8001_MAX_LENGTH 11 +#define W8001_PACKET_LEN 11 +#define W8001_LEAD_MASK 0x80 +#define W8001_LEAD_BYTE 0x80 +#define W8001_TAB_MASK 0x40 +#define W8001_TAB_BYTE 0x40 + +#define W8001_QUERY_PACKET 0x20 + +struct w8001_coord { + u8 rdy; + u8 tsw; + u8 f1; + u8 f2; + u16 x; + u16 y; + u16 pen_pressure; + u8 tilt_x; + u8 tilt_y; +}; + +/* + * Per-touchscreen data. + */ + +struct w8001 { + struct input_dev *dev; + struct serio *serio; + struct mutex cmd_mutex; + struct completion cmd_done; + int id; + int idx; + unsigned char expected_packet; + unsigned char data[W8001_MAX_LENGTH]; + unsigned char response[W8001_PACKET_LEN]; + char phys[32]; +}; + +static int parse_data(u8 *data, struct w8001_coord *coord) +{ + coord->rdy = data[0] & 0x20; + coord->tsw = data[0] & 0x01; + coord->f1 = data[0] & 0x02; + coord->f2 = data[0] & 0x04; + + coord->x = (data[1] & 0x7F) << 9; + coord->x |= (data[2] & 0x7F) << 2; + coord->x |= (data[6] & 0x60) >> 5; + + coord->y = (data[3] & 0x7F) << 9; + coord->y |= (data[4] & 0x7F) << 2; + coord->y |= (data[6] & 0x18) >> 3; + + coord->pen_pressure = data[5] & 0x7F; + coord->pen_pressure |= (data[6] & 0x07) << 7 ; + + coord->tilt_x = data[7] & 0x7F; + coord->tilt_y = data[8] & 0x7F; + + return 0; +} + +static void w8001_process_data(struct w8001 *w8001, unsigned char data) +{ + struct input_dev *dev = w8001->dev; + u8 tmp; + struct w8001_coord coord; + + w8001->data[w8001->idx] = data; + switch (w8001->idx++) { + case 0: + if ((data & W8001_LEAD_MASK) != W8001_LEAD_BYTE) { + pr_debug("w8001: unsynchronized data: 0x%02x\n", data); + w8001->idx = 0; + } + break; + case 8: + tmp = w8001->data[0] & W8001_TAB_MASK; + if (unlikely(tmp == W8001_TAB_BYTE)) + break; + w8001->idx = 0; + memset(&coord, 0, sizeof(coord)); + parse_data(w8001->data, &coord); + input_report_abs(dev, ABS_X, coord.x); + input_report_abs(dev, ABS_Y, coord.y); + input_report_abs(dev, ABS_PRESSURE, coord.pen_pressure); + input_report_key(dev, BTN_TOUCH, coord.tsw); + input_sync(dev); + break; + case 10: + w8001->idx = 0; + memcpy(w8001->response, &w8001->data, W8001_PACKET_LEN); + w8001->expected_packet = W8001_QUERY_PACKET; + complete(&w8001->cmd_done); + break; + } +} + + +static irqreturn_t w8001_interrupt(struct serio *serio, + unsigned char data, unsigned int flags) +{ + struct w8001 *w8001 = serio_get_drvdata(serio); + + w8001_process_data(w8001, data); + + return IRQ_HANDLED; +} + +static int w8001_async_command(struct w8001 *w8001, unsigned char *packet, + int len) +{ + int rc = -1; + int i; + + mutex_lock(&w8001->cmd_mutex); + + for (i = 0; i < len; i++) { + if (serio_write(w8001->serio, packet[i])) + goto out; + } + rc = 0; + +out: + mutex_unlock(&w8001->cmd_mutex); + return rc; +} + +static int w8001_command(struct w8001 *w8001, unsigned char *packet, int len) +{ + int rc = -1; + int i; + + mutex_lock(&w8001->cmd_mutex); + + serio_pause_rx(w8001->serio); + init_completion(&w8001->cmd_done); + serio_continue_rx(w8001->serio); + + for (i = 0; i < len; i++) { + if (serio_write(w8001->serio, packet[i])) + goto out; + } + + wait_for_completion_timeout(&w8001->cmd_done, HZ); + + if (w8001->expected_packet == W8001_QUERY_PACKET) { + /* We are back in reporting mode, the query was ACKed */ + memcpy(packet, w8001->response, W8001_PACKET_LEN); + rc = 0; + } + +out: + mutex_unlock(&w8001->cmd_mutex); + return rc; +} + +static int w8001_setup(struct w8001 *w8001) +{ + struct w8001_coord coord; + struct input_dev *dev = w8001->dev; + unsigned char start[1] = { '1' }; + unsigned char query[11] = { '*' }; + + if (w8001_command(w8001, query, 1)) + return -1; + + memset(&coord, 0, sizeof(coord)); + parse_data(query, &coord); + + input_set_abs_params(dev, ABS_X, 0, coord.x, 0, 0); + input_set_abs_params(dev, ABS_Y, 0, coord.y, 0, 0); + input_set_abs_params(dev, ABS_PRESSURE, 0, coord.pen_pressure, 0, 0); + input_set_abs_params(dev, ABS_TILT_X, 0, coord.tilt_x, 0, 0); + input_set_abs_params(dev, ABS_TILT_Y, 0, coord.tilt_y, 0, 0); + + if (w8001_async_command(w8001, start, 1)) + return -1; + + return 0; +} + +/* + * w8001_disconnect() is the opposite of w8001_connect() + */ + +static void w8001_disconnect(struct serio *serio) +{ + struct w8001 *w8001 = serio_get_drvdata(serio); + + input_get_device(w8001->dev); + input_unregister_device(w8001->dev); + serio_close(serio); + serio_set_drvdata(serio, NULL); + input_put_device(w8001->dev); + kfree(w8001); +} + +/* + * w8001_connect() is the routine that is called when someone adds a + * new serio device that supports the w8001 protocol and registers it as + * an input device. + */ + +static int w8001_connect(struct serio *serio, struct serio_driver *drv) +{ + struct w8001 *w8001; + struct input_dev *input_dev; + int err; + + w8001 = kzalloc(sizeof(struct w8001), GFP_KERNEL); + input_dev = input_allocate_device(); + if (!w8001 || !input_dev) { + err = -ENOMEM; + goto fail1; + } + + w8001->serio = serio; + w8001->id = serio->id.id; + w8001->dev = input_dev; + mutex_init(&w8001->cmd_mutex); + init_completion(&w8001->cmd_done); + snprintf(w8001->phys, sizeof(w8001->phys), "%s/input0", serio->phys); + + input_dev->name = "Wacom W8001 Penabled Serial TouchScreen"; + input_dev->phys = w8001->phys; + input_dev->id.bustype = BUS_RS232; + input_dev->id.vendor = SERIO_W8001; + input_dev->id.product = w8001->id; + input_dev->id.version = 0x0100; + input_dev->dev.parent = &serio->dev; + + input_dev->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_ABS); + input_dev->keybit[BIT_WORD(BTN_TOUCH)] = BIT_MASK(BTN_TOUCH); + + serio_set_drvdata(serio, w8001); + err = serio_open(serio, drv); + if (err) + goto fail2; + + if (w8001_setup(w8001)) + goto fail3; + + err = input_register_device(w8001->dev); + if (err) + goto fail3; + + return 0; + +fail3: + serio_close(serio); +fail2: + serio_set_drvdata(serio, NULL); +fail1: + input_free_device(input_dev); + kfree(w8001); + return err; +} + +static struct serio_device_id w8001_serio_ids[] = { + { + .type = SERIO_RS232, + .proto = SERIO_W8001, + .id = SERIO_ANY, + .extra = SERIO_ANY, + }, + { 0 } +}; + +MODULE_DEVICE_TABLE(serio, w8001_serio_ids); + +static struct serio_driver w8001_drv = { + .driver = { + .name = "w8001", + }, + .description = DRIVER_DESC, + .id_table = w8001_serio_ids, + .interrupt = w8001_interrupt, + .connect = w8001_connect, + .disconnect = w8001_disconnect, +}; + +static int __init w8001_init(void) +{ + return serio_register_driver(&w8001_drv); +} + +static void __exit w8001_exit(void) +{ + serio_unregister_driver(&w8001_drv); +} + +module_init(w8001_init); +module_exit(w8001_exit); diff --git a/include/linux/serio.h b/include/linux/serio.h index 25641d9e0ea8..1bcb357a01a1 100644 --- a/include/linux/serio.h +++ b/include/linux/serio.h @@ -213,5 +213,6 @@ static inline void serio_unpin_driver(struct serio *serio) #define SERIO_ZHENHUA 0x36 #define SERIO_INEXIO 0x37 #define SERIO_TOUCHIT213 0x37 +#define SERIO_W8001 0x39 #endif -- cgit v1.2.3 From a2d781fc8d9b16113dd9440107d73c0f21d7cbef Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Wed, 19 Nov 2008 17:02:24 -0500 Subject: Input: libps2 - handle 0xfc responses from devices Signed-off-by: Dmitry Torokhov --- drivers/input/serio/libps2.c | 20 ++++++++++++++++---- include/linux/libps2.h | 2 ++ 2 files changed, 18 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/input/serio/libps2.c b/drivers/input/serio/libps2.c index 2b304c22c200..67248c31e19a 100644 --- a/drivers/input/serio/libps2.c +++ b/drivers/input/serio/libps2.c @@ -262,9 +262,17 @@ int ps2_handle_ack(struct ps2dev *ps2dev, unsigned char data) break; case PS2_RET_NAK: - ps2dev->nak = 1; + ps2dev->flags |= PS2_FLAG_NAK; + ps2dev->nak = PS2_RET_NAK; break; + case PS2_RET_ERR: + if (ps2dev->flags & PS2_FLAG_NAK) { + ps2dev->flags &= ~PS2_FLAG_NAK; + ps2dev->nak = PS2_RET_ERR; + break; + } + /* * Workaround for mice which don't ACK the Get ID command. * These are valid mouse IDs that we recognize. @@ -282,8 +290,11 @@ int ps2_handle_ack(struct ps2dev *ps2dev, unsigned char data) } - if (!ps2dev->nak && ps2dev->cmdcnt) - ps2dev->flags |= PS2_FLAG_CMD | PS2_FLAG_CMD1; + if (!ps2dev->nak) { + ps2dev->flags &= ~PS2_FLAG_NAK; + if (ps2dev->cmdcnt) + ps2dev->flags |= PS2_FLAG_CMD | PS2_FLAG_CMD1; + } ps2dev->flags &= ~PS2_FLAG_ACK; wake_up(&ps2dev->wait); @@ -329,6 +340,7 @@ void ps2_cmd_aborted(struct ps2dev *ps2dev) if (ps2dev->flags & (PS2_FLAG_ACK | PS2_FLAG_CMD)) wake_up(&ps2dev->wait); - ps2dev->flags = 0; + /* reset all flags except last nack */ + ps2dev->flags &= PS2_FLAG_NAK; } EXPORT_SYMBOL(ps2_cmd_aborted); diff --git a/include/linux/libps2.h b/include/linux/libps2.h index afc413369101..b94534b7e266 100644 --- a/include/linux/libps2.h +++ b/include/linux/libps2.h @@ -18,11 +18,13 @@ #define PS2_RET_ID 0x00 #define PS2_RET_ACK 0xfa #define PS2_RET_NAK 0xfe +#define PS2_RET_ERR 0xfc #define PS2_FLAG_ACK 1 /* Waiting for ACK/NAK */ #define PS2_FLAG_CMD 2 /* Waiting for command to finish */ #define PS2_FLAG_CMD1 4 /* Waiting for the first byte of command response */ #define PS2_FLAG_WAITID 8 /* Command execiting is GET ID */ +#define PS2_FLAG_NAK 16 /* Last transmission was NAKed */ struct ps2dev { struct serio *serio; -- cgit v1.2.3 From 193da6092764ab693da7170c5badbf60d7758c1d Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 26 Nov 2008 12:03:54 +0100 Subject: fuse: move FUSE_MINOR to miscdevice.h Move FUSE_MINOR to miscdevice.h. While at it, de-uglify the file. Signed-off-by: Tejun Heo Signed-off-by: Miklos Szeredi --- include/linux/fuse.h | 3 --- include/linux/miscdevice.h | 42 +++++++++++++++++++++--------------------- 2 files changed, 21 insertions(+), 24 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fuse.h b/include/linux/fuse.h index 350fe9767bbc..7caa473306e4 100644 --- a/include/linux/fuse.h +++ b/include/linux/fuse.h @@ -40,9 +40,6 @@ /** The major number of the fuse character device */ #define FUSE_MAJOR MISC_MAJOR -/** The minor number of the fuse character device */ -#define FUSE_MINOR 229 - /* Make sure all structures are padded to 64bit boundary, so 32bit userspace works under 64bit kernels */ diff --git a/include/linux/miscdevice.h b/include/linux/miscdevice.h index 26433ec520b3..a820f816a49e 100644 --- a/include/linux/miscdevice.h +++ b/include/linux/miscdevice.h @@ -3,33 +3,33 @@ #include #include -#define PSMOUSE_MINOR 1 -#define MS_BUSMOUSE_MINOR 2 -#define ATIXL_BUSMOUSE_MINOR 3 -/*#define AMIGAMOUSE_MINOR 4 FIXME OBSOLETE */ -#define ATARIMOUSE_MINOR 5 -#define SUN_MOUSE_MINOR 6 -#define APOLLO_MOUSE_MINOR 7 -#define PC110PAD_MINOR 9 -/*#define ADB_MOUSE_MINOR 10 FIXME OBSOLETE */ +#define PSMOUSE_MINOR 1 +#define MS_BUSMOUSE_MINOR 2 +#define ATIXL_BUSMOUSE_MINOR 3 +/*#define AMIGAMOUSE_MINOR 4 FIXME OBSOLETE */ +#define ATARIMOUSE_MINOR 5 +#define SUN_MOUSE_MINOR 6 +#define APOLLO_MOUSE_MINOR 7 +#define PC110PAD_MINOR 9 +/*#define ADB_MOUSE_MINOR 10 FIXME OBSOLETE */ #define WATCHDOG_MINOR 130 /* Watchdog timer */ #define TEMP_MINOR 131 /* Temperature Sensor */ -#define RTC_MINOR 135 +#define RTC_MINOR 135 #define EFI_RTC_MINOR 136 /* EFI Time services */ -#define SUN_OPENPROM_MINOR 139 +#define SUN_OPENPROM_MINOR 139 #define DMAPI_MINOR 140 /* DMAPI */ -#define NVRAM_MINOR 144 -#define SGI_MMTIMER 153 +#define NVRAM_MINOR 144 +#define SGI_MMTIMER 153 #define STORE_QUEUE_MINOR 155 -#define I2O_MINOR 166 +#define I2O_MINOR 166 #define MICROCODE_MINOR 184 -#define MWAVE_MINOR 219 /* ACP/Mwave Modem */ -#define MPT_MINOR 220 -#define MISC_DYNAMIC_MINOR 255 - -#define TUN_MINOR 200 -#define HPET_MINOR 228 -#define KVM_MINOR 232 +#define TUN_MINOR 200 +#define MWAVE_MINOR 219 /* ACP/Mwave Modem */ +#define MPT_MINOR 220 +#define HPET_MINOR 228 +#define FUSE_MINOR 229 +#define KVM_MINOR 232 +#define MISC_DYNAMIC_MINOR 255 struct device; -- cgit v1.2.3 From 59efec7b903987dcb60b9ebc85c7acd4443a11a1 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 26 Nov 2008 12:03:55 +0100 Subject: fuse: implement ioctl support Generic ioctl support is tricky to implement because only the ioctl implementation itself knows which memory regions need to be read and/or written. To support this, fuse client can request retry of ioctl specifying memory regions to read and write. Deep copying (nested pointers) can be implemented by retrying multiple times resolving one depth of dereference at a time. For security and cleanliness considerations, ioctl implementation has restricted mode where the kernel determines data transfer directions and sizes using the _IOC_*() macros on the ioctl command. In this mode, retry is not allowed. For all FUSE servers, restricted mode is enforced. Unrestricted ioctl will be used by CUSE. Plese read the comment on top of fs/fuse/file.c::fuse_file_do_ioctl() for more information. Signed-off-by: Tejun Heo Signed-off-by: Miklos Szeredi --- fs/fuse/file.c | 280 +++++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/fuse.h | 32 ++++++ 2 files changed, 312 insertions(+) (limited to 'include/linux') diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 617269803913..baed06ea7622 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -1469,6 +1469,282 @@ static loff_t fuse_file_llseek(struct file *file, loff_t offset, int origin) return retval; } +static int fuse_ioctl_copy_user(struct page **pages, struct iovec *iov, + unsigned int nr_segs, size_t bytes, bool to_user) +{ + struct iov_iter ii; + int page_idx = 0; + + if (!bytes) + return 0; + + iov_iter_init(&ii, iov, nr_segs, bytes, 0); + + while (iov_iter_count(&ii)) { + struct page *page = pages[page_idx++]; + size_t todo = min_t(size_t, PAGE_SIZE, iov_iter_count(&ii)); + void *kaddr, *map; + + kaddr = map = kmap(page); + + while (todo) { + char __user *uaddr = ii.iov->iov_base + ii.iov_offset; + size_t iov_len = ii.iov->iov_len - ii.iov_offset; + size_t copy = min(todo, iov_len); + size_t left; + + if (!to_user) + left = copy_from_user(kaddr, uaddr, copy); + else + left = copy_to_user(uaddr, kaddr, copy); + + if (unlikely(left)) + return -EFAULT; + + iov_iter_advance(&ii, copy); + todo -= copy; + kaddr += copy; + } + + kunmap(map); + } + + return 0; +} + +/* + * For ioctls, there is no generic way to determine how much memory + * needs to be read and/or written. Furthermore, ioctls are allowed + * to dereference the passed pointer, so the parameter requires deep + * copying but FUSE has no idea whatsoever about what to copy in or + * out. + * + * This is solved by allowing FUSE server to retry ioctl with + * necessary in/out iovecs. Let's assume the ioctl implementation + * needs to read in the following structure. + * + * struct a { + * char *buf; + * size_t buflen; + * } + * + * On the first callout to FUSE server, inarg->in_size and + * inarg->out_size will be NULL; then, the server completes the ioctl + * with FUSE_IOCTL_RETRY set in out->flags, out->in_iovs set to 1 and + * the actual iov array to + * + * { { .iov_base = inarg.arg, .iov_len = sizeof(struct a) } } + * + * which tells FUSE to copy in the requested area and retry the ioctl. + * On the second round, the server has access to the structure and + * from that it can tell what to look for next, so on the invocation, + * it sets FUSE_IOCTL_RETRY, out->in_iovs to 2 and iov array to + * + * { { .iov_base = inarg.arg, .iov_len = sizeof(struct a) }, + * { .iov_base = a.buf, .iov_len = a.buflen } } + * + * FUSE will copy both struct a and the pointed buffer from the + * process doing the ioctl and retry ioctl with both struct a and the + * buffer. + * + * This time, FUSE server has everything it needs and completes ioctl + * without FUSE_IOCTL_RETRY which finishes the ioctl call. + * + * Copying data out works the same way. + * + * Note that if FUSE_IOCTL_UNRESTRICTED is clear, the kernel + * automatically initializes in and out iovs by decoding @cmd with + * _IOC_* macros and the server is not allowed to request RETRY. This + * limits ioctl data transfers to well-formed ioctls and is the forced + * behavior for all FUSE servers. + */ +static long fuse_file_do_ioctl(struct file *file, unsigned int cmd, + unsigned long arg, unsigned int flags) +{ + struct inode *inode = file->f_dentry->d_inode; + struct fuse_file *ff = file->private_data; + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_ioctl_in inarg = { + .fh = ff->fh, + .cmd = cmd, + .arg = arg, + .flags = flags + }; + struct fuse_ioctl_out outarg; + struct fuse_req *req = NULL; + struct page **pages = NULL; + struct page *iov_page = NULL; + struct iovec *in_iov = NULL, *out_iov = NULL; + unsigned int in_iovs = 0, out_iovs = 0, num_pages = 0, max_pages; + size_t in_size, out_size, transferred; + int err; + + /* assume all the iovs returned by client always fits in a page */ + BUILD_BUG_ON(sizeof(struct iovec) * FUSE_IOCTL_MAX_IOV > PAGE_SIZE); + + if (!fuse_allow_task(fc, current)) + return -EACCES; + + err = -EIO; + if (is_bad_inode(inode)) + goto out; + + err = -ENOMEM; + pages = kzalloc(sizeof(pages[0]) * FUSE_MAX_PAGES_PER_REQ, GFP_KERNEL); + iov_page = alloc_page(GFP_KERNEL); + if (!pages || !iov_page) + goto out; + + /* + * If restricted, initialize IO parameters as encoded in @cmd. + * RETRY from server is not allowed. + */ + if (!(flags & FUSE_IOCTL_UNRESTRICTED)) { + struct iovec *iov = page_address(iov_page); + + iov->iov_base = (void *)arg; + iov->iov_len = _IOC_SIZE(cmd); + + if (_IOC_DIR(cmd) & _IOC_WRITE) { + in_iov = iov; + in_iovs = 1; + } + + if (_IOC_DIR(cmd) & _IOC_READ) { + out_iov = iov; + out_iovs = 1; + } + } + + retry: + inarg.in_size = in_size = iov_length(in_iov, in_iovs); + inarg.out_size = out_size = iov_length(out_iov, out_iovs); + + /* + * Out data can be used either for actual out data or iovs, + * make sure there always is at least one page. + */ + out_size = max_t(size_t, out_size, PAGE_SIZE); + max_pages = DIV_ROUND_UP(max(in_size, out_size), PAGE_SIZE); + + /* make sure there are enough buffer pages and init request with them */ + err = -ENOMEM; + if (max_pages > FUSE_MAX_PAGES_PER_REQ) + goto out; + while (num_pages < max_pages) { + pages[num_pages] = alloc_page(GFP_KERNEL | __GFP_HIGHMEM); + if (!pages[num_pages]) + goto out; + num_pages++; + } + + req = fuse_get_req(fc); + if (IS_ERR(req)) { + err = PTR_ERR(req); + req = NULL; + goto out; + } + memcpy(req->pages, pages, sizeof(req->pages[0]) * num_pages); + req->num_pages = num_pages; + + /* okay, let's send it to the client */ + req->in.h.opcode = FUSE_IOCTL; + req->in.h.nodeid = get_node_id(inode); + req->in.numargs = 1; + req->in.args[0].size = sizeof(inarg); + req->in.args[0].value = &inarg; + if (in_size) { + req->in.numargs++; + req->in.args[1].size = in_size; + req->in.argpages = 1; + + err = fuse_ioctl_copy_user(pages, in_iov, in_iovs, in_size, + false); + if (err) + goto out; + } + + req->out.numargs = 2; + req->out.args[0].size = sizeof(outarg); + req->out.args[0].value = &outarg; + req->out.args[1].size = out_size; + req->out.argpages = 1; + req->out.argvar = 1; + + request_send(fc, req); + err = req->out.h.error; + transferred = req->out.args[1].size; + fuse_put_request(fc, req); + req = NULL; + if (err) + goto out; + + /* did it ask for retry? */ + if (outarg.flags & FUSE_IOCTL_RETRY) { + char *vaddr; + + /* no retry if in restricted mode */ + err = -EIO; + if (!(flags & FUSE_IOCTL_UNRESTRICTED)) + goto out; + + in_iovs = outarg.in_iovs; + out_iovs = outarg.out_iovs; + + /* + * Make sure things are in boundary, separate checks + * are to protect against overflow. + */ + err = -ENOMEM; + if (in_iovs > FUSE_IOCTL_MAX_IOV || + out_iovs > FUSE_IOCTL_MAX_IOV || + in_iovs + out_iovs > FUSE_IOCTL_MAX_IOV) + goto out; + + err = -EIO; + if ((in_iovs + out_iovs) * sizeof(struct iovec) != transferred) + goto out; + + /* okay, copy in iovs and retry */ + vaddr = kmap_atomic(pages[0], KM_USER0); + memcpy(page_address(iov_page), vaddr, transferred); + kunmap_atomic(vaddr, KM_USER0); + + in_iov = page_address(iov_page); + out_iov = in_iov + in_iovs; + + goto retry; + } + + err = -EIO; + if (transferred > inarg.out_size) + goto out; + + err = fuse_ioctl_copy_user(pages, out_iov, out_iovs, transferred, true); + out: + if (req) + fuse_put_request(fc, req); + if (iov_page) + __free_page(iov_page); + while (num_pages) + __free_page(pages[--num_pages]); + kfree(pages); + + return err ? err : outarg.result; +} + +static long fuse_file_ioctl(struct file *file, unsigned int cmd, + unsigned long arg) +{ + return fuse_file_do_ioctl(file, cmd, arg, 0); +} + +static long fuse_file_compat_ioctl(struct file *file, unsigned int cmd, + unsigned long arg) +{ + return fuse_file_do_ioctl(file, cmd, arg, FUSE_IOCTL_COMPAT); +} + static const struct file_operations fuse_file_operations = { .llseek = fuse_file_llseek, .read = do_sync_read, @@ -1483,6 +1759,8 @@ static const struct file_operations fuse_file_operations = { .lock = fuse_file_lock, .flock = fuse_file_flock, .splice_read = generic_file_splice_read, + .unlocked_ioctl = fuse_file_ioctl, + .compat_ioctl = fuse_file_compat_ioctl, }; static const struct file_operations fuse_direct_io_file_operations = { @@ -1495,6 +1773,8 @@ static const struct file_operations fuse_direct_io_file_operations = { .fsync = fuse_fsync, .lock = fuse_file_lock, .flock = fuse_file_flock, + .unlocked_ioctl = fuse_file_ioctl, + .compat_ioctl = fuse_file_compat_ioctl, /* no mmap and splice_read */ }; diff --git a/include/linux/fuse.h b/include/linux/fuse.h index 7caa473306e4..608e300ae883 100644 --- a/include/linux/fuse.h +++ b/include/linux/fuse.h @@ -148,6 +148,21 @@ struct fuse_file_lock { */ #define FUSE_READ_LOCKOWNER (1 << 1) +/** + * Ioctl flags + * + * FUSE_IOCTL_COMPAT: 32bit compat ioctl on 64bit machine + * FUSE_IOCTL_UNRESTRICTED: not restricted to well-formed ioctls, retry allowed + * FUSE_IOCTL_RETRY: retry with new iovecs + * + * FUSE_IOCTL_MAX_IOV: maximum of in_iovecs + out_iovecs + */ +#define FUSE_IOCTL_COMPAT (1 << 0) +#define FUSE_IOCTL_UNRESTRICTED (1 << 1) +#define FUSE_IOCTL_RETRY (1 << 2) + +#define FUSE_IOCTL_MAX_IOV 256 + enum fuse_opcode { FUSE_LOOKUP = 1, FUSE_FORGET = 2, /* no reply */ @@ -185,6 +200,7 @@ enum fuse_opcode { FUSE_INTERRUPT = 36, FUSE_BMAP = 37, FUSE_DESTROY = 38, + FUSE_IOCTL = 39, }; /* The read buffer is required to be at least 8k, but may be much larger */ @@ -385,6 +401,22 @@ struct fuse_bmap_out { __u64 block; }; +struct fuse_ioctl_in { + __u64 fh; + __u32 flags; + __u32 cmd; + __u64 arg; + __u32 in_size; + __u32 out_size; +}; + +struct fuse_ioctl_out { + __s32 result; + __u32 flags; + __u32 in_iovs; + __u32 out_iovs; +}; + struct fuse_in_header { __u32 len; __u32 opcode; -- cgit v1.2.3 From 8599396b5062bf6bd2a0b433503849e2322df1c2 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 26 Nov 2008 12:03:55 +0100 Subject: fuse: implement unsolicited notification Clients always used to write only in response to read requests. To implement poll efficiently, clients should be able to issue unsolicited notifications. This patch implements basic notification support. Zero fuse_out_header.unique is now accepted and considered unsolicited notification and the error field contains notification code. This patch doesn't implement any actual notification. Signed-off-by: Tejun Heo Signed-off-by: Miklos Szeredi --- fs/fuse/dev.c | 27 +++++++++++++++++++++++++-- include/linux/fuse.h | 4 ++++ 2 files changed, 29 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 225388f54ae7..ffd670bb8c8c 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -816,6 +816,15 @@ static ssize_t fuse_dev_read(struct kiocb *iocb, const struct iovec *iov, return err; } +static int fuse_notify(struct fuse_conn *fc, enum fuse_notify_code code, + unsigned int size, struct fuse_copy_state *cs) +{ + switch (code) { + default: + return -EINVAL; + } +} + /* Look up request on processing list by unique ID */ static struct fuse_req *request_find(struct fuse_conn *fc, u64 unique) { @@ -879,9 +888,23 @@ static ssize_t fuse_dev_write(struct kiocb *iocb, const struct iovec *iov, err = fuse_copy_one(&cs, &oh, sizeof(oh)); if (err) goto err_finish; + + err = -EINVAL; + if (oh.len != nbytes) + goto err_finish; + + /* + * Zero oh.unique indicates unsolicited notification message + * and error contains notification code. + */ + if (!oh.unique) { + err = fuse_notify(fc, oh.error, nbytes - sizeof(oh), &cs); + fuse_copy_finish(&cs); + return err ? err : nbytes; + } + err = -EINVAL; - if (!oh.unique || oh.error <= -1000 || oh.error > 0 || - oh.len != nbytes) + if (oh.error <= -1000 || oh.error > 0) goto err_finish; spin_lock(&fc->lock); diff --git a/include/linux/fuse.h b/include/linux/fuse.h index 608e300ae883..abde9949e2c0 100644 --- a/include/linux/fuse.h +++ b/include/linux/fuse.h @@ -203,6 +203,10 @@ enum fuse_opcode { FUSE_IOCTL = 39, }; +enum fuse_notify_code { + FUSE_NOTIFY_CODE_MAX, +}; + /* The read buffer is required to be at least 8k, but may be much larger */ #define FUSE_MIN_READ_BUFFER 8192 -- cgit v1.2.3 From 95668a69a4bb862063c4d28a746e55107dee7b98 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 26 Nov 2008 12:03:55 +0100 Subject: fuse: implement poll support Implement poll support. Polled files are indexed using kh in a RB tree rooted at fuse_conn->polled_files. Client should send FUSE_NOTIFY_POLL notification once after processing FUSE_POLL which has FUSE_POLL_SCHEDULE_NOTIFY set. Sending notification unconditionally after the latest poll or everytime file content might have changed is inefficient but won't cause malfunction. fuse_file_poll() can sleep and requires patches from the following thread which allows f_op->poll() to sleep. http://thread.gmane.org/gmane.linux.kernel/726176 Signed-off-by: Tejun Heo Signed-off-by: Miklos Szeredi --- fs/fuse/dev.c | 19 ++++++++ fs/fuse/file.c | 132 +++++++++++++++++++++++++++++++++++++++++++++++++++ fs/fuse/fuse_i.h | 20 ++++++++ fs/fuse/inode.c | 1 + include/linux/fuse.h | 25 ++++++++++ 5 files changed, 197 insertions(+) (limited to 'include/linux') diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index ffd670bb8c8c..6176e444c76e 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -816,10 +816,29 @@ static ssize_t fuse_dev_read(struct kiocb *iocb, const struct iovec *iov, return err; } +static int fuse_notify_poll(struct fuse_conn *fc, unsigned int size, + struct fuse_copy_state *cs) +{ + struct fuse_notify_poll_wakeup_out outarg; + int err; + + if (size != sizeof(outarg)) + return -EINVAL; + + err = fuse_copy_one(cs, &outarg, sizeof(outarg)); + if (err) + return err; + + return fuse_notify_poll_wakeup(fc, &outarg); +} + static int fuse_notify(struct fuse_conn *fc, enum fuse_notify_code code, unsigned int size, struct fuse_copy_state *cs) { switch (code) { + case FUSE_NOTIFY_POLL: + return fuse_notify_poll(fc, size, cs); + default: return -EINVAL; } diff --git a/fs/fuse/file.c b/fs/fuse/file.c index a28ced678d38..b3a944e4bb9c 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -62,6 +62,8 @@ struct fuse_file *fuse_file_alloc(struct fuse_conn *fc) ff->kh = ++fc->khctr; spin_unlock(&fc->lock); } + RB_CLEAR_NODE(&ff->polled_node); + init_waitqueue_head(&ff->poll_wait); } return ff; } @@ -170,7 +172,11 @@ int fuse_release_common(struct inode *inode, struct file *file, int isdir) spin_lock(&fc->lock); list_del(&ff->write_entry); + if (!RB_EMPTY_NODE(&ff->polled_node)) + rb_erase(&ff->polled_node, &fc->polled_files); spin_unlock(&fc->lock); + + wake_up_interruptible_sync(&ff->poll_wait); /* * Normally this will send the RELEASE request, * however if some asynchronous READ or WRITE requests @@ -1749,6 +1755,130 @@ static long fuse_file_compat_ioctl(struct file *file, unsigned int cmd, return fuse_file_do_ioctl(file, cmd, arg, FUSE_IOCTL_COMPAT); } +/* + * All files which have been polled are linked to RB tree + * fuse_conn->polled_files which is indexed by kh. Walk the tree and + * find the matching one. + */ +static struct rb_node **fuse_find_polled_node(struct fuse_conn *fc, u64 kh, + struct rb_node **parent_out) +{ + struct rb_node **link = &fc->polled_files.rb_node; + struct rb_node *last = NULL; + + while (*link) { + struct fuse_file *ff; + + last = *link; + ff = rb_entry(last, struct fuse_file, polled_node); + + if (kh < ff->kh) + link = &last->rb_left; + else if (kh > ff->kh) + link = &last->rb_right; + else + return link; + } + + if (parent_out) + *parent_out = last; + return link; +} + +/* + * The file is about to be polled. Make sure it's on the polled_files + * RB tree. Note that files once added to the polled_files tree are + * not removed before the file is released. This is because a file + * polled once is likely to be polled again. + */ +static void fuse_register_polled_file(struct fuse_conn *fc, + struct fuse_file *ff) +{ + spin_lock(&fc->lock); + if (RB_EMPTY_NODE(&ff->polled_node)) { + struct rb_node **link, *parent; + + link = fuse_find_polled_node(fc, ff->kh, &parent); + BUG_ON(*link); + rb_link_node(&ff->polled_node, parent, link); + rb_insert_color(&ff->polled_node, &fc->polled_files); + } + spin_unlock(&fc->lock); +} + +static unsigned fuse_file_poll(struct file *file, poll_table *wait) +{ + struct inode *inode = file->f_dentry->d_inode; + struct fuse_file *ff = file->private_data; + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_poll_in inarg = { .fh = ff->fh, .kh = ff->kh }; + struct fuse_poll_out outarg; + struct fuse_req *req; + int err; + + if (fc->no_poll) + return DEFAULT_POLLMASK; + + poll_wait(file, &ff->poll_wait, wait); + + /* + * Ask for notification iff there's someone waiting for it. + * The client may ignore the flag and always notify. + */ + if (waitqueue_active(&ff->poll_wait)) { + inarg.flags |= FUSE_POLL_SCHEDULE_NOTIFY; + fuse_register_polled_file(fc, ff); + } + + req = fuse_get_req(fc); + if (IS_ERR(req)) + return PTR_ERR(req); + + req->in.h.opcode = FUSE_POLL; + req->in.h.nodeid = get_node_id(inode); + req->in.numargs = 1; + req->in.args[0].size = sizeof(inarg); + req->in.args[0].value = &inarg; + req->out.numargs = 1; + req->out.args[0].size = sizeof(outarg); + req->out.args[0].value = &outarg; + request_send(fc, req); + err = req->out.h.error; + fuse_put_request(fc, req); + + if (!err) + return outarg.revents; + if (err == -ENOSYS) { + fc->no_poll = 1; + return DEFAULT_POLLMASK; + } + return POLLERR; +} + +/* + * This is called from fuse_handle_notify() on FUSE_NOTIFY_POLL and + * wakes up the poll waiters. + */ +int fuse_notify_poll_wakeup(struct fuse_conn *fc, + struct fuse_notify_poll_wakeup_out *outarg) +{ + u64 kh = outarg->kh; + struct rb_node **link; + + spin_lock(&fc->lock); + + link = fuse_find_polled_node(fc, kh, NULL); + if (*link) { + struct fuse_file *ff; + + ff = rb_entry(*link, struct fuse_file, polled_node); + wake_up_interruptible_sync(&ff->poll_wait); + } + + spin_unlock(&fc->lock); + return 0; +} + static const struct file_operations fuse_file_operations = { .llseek = fuse_file_llseek, .read = do_sync_read, @@ -1765,6 +1895,7 @@ static const struct file_operations fuse_file_operations = { .splice_read = generic_file_splice_read, .unlocked_ioctl = fuse_file_ioctl, .compat_ioctl = fuse_file_compat_ioctl, + .poll = fuse_file_poll, }; static const struct file_operations fuse_direct_io_file_operations = { @@ -1779,6 +1910,7 @@ static const struct file_operations fuse_direct_io_file_operations = { .flock = fuse_file_flock, .unlocked_ioctl = fuse_file_ioctl, .compat_ioctl = fuse_file_compat_ioctl, + .poll = fuse_file_poll, /* no mmap and splice_read */ }; diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 86f013303828..986fbd4c1ff5 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -19,6 +19,8 @@ #include #include #include +#include +#include /** Max number of pages that can be used in a single read request */ #define FUSE_MAX_PAGES_PER_REQ 32 @@ -111,6 +113,12 @@ struct fuse_file { /** Entry on inode's write_files list */ struct list_head write_entry; + + /** RB node to be linked on fuse_conn->polled_files */ + struct rb_node polled_node; + + /** Wait queue head for poll */ + wait_queue_head_t poll_wait; }; /** One input argument of a request */ @@ -328,6 +336,9 @@ struct fuse_conn { /** The next unique kernel file handle */ u64 khctr; + /** rbtree of fuse_files waiting for poll events indexed by ph */ + struct rb_root polled_files; + /** Number of requests currently in the background */ unsigned num_background; @@ -416,6 +427,9 @@ struct fuse_conn { /** Is bmap not implemented by fs? */ unsigned no_bmap:1; + /** Is poll not implemented by fs? */ + unsigned no_poll:1; + /** Do multi-page cached writes */ unsigned big_writes:1; @@ -524,6 +538,12 @@ int fuse_release_common(struct inode *inode, struct file *file, int isdir); int fuse_fsync_common(struct file *file, struct dentry *de, int datasync, int isdir); +/** + * Notify poll wakeup + */ +int fuse_notify_poll_wakeup(struct fuse_conn *fc, + struct fuse_notify_poll_wakeup_out *outarg); + /** * Initialize file operations on a regular file */ diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 0e15bc221d23..ba7256128084 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -486,6 +486,7 @@ static struct fuse_conn *new_conn(struct super_block *sb) /* fuse does it's own writeback accounting */ fc->bdi.capabilities = BDI_CAP_NO_ACCT_WB; fc->khctr = 0; + fc->polled_files = RB_ROOT; fc->dev = sb->s_dev; err = bdi_init(&fc->bdi); if (err) diff --git a/include/linux/fuse.h b/include/linux/fuse.h index abde9949e2c0..5650cf033e73 100644 --- a/include/linux/fuse.h +++ b/include/linux/fuse.h @@ -163,6 +163,13 @@ struct fuse_file_lock { #define FUSE_IOCTL_MAX_IOV 256 +/** + * Poll flags + * + * FUSE_POLL_SCHEDULE_NOTIFY: request poll notify + */ +#define FUSE_POLL_SCHEDULE_NOTIFY (1 << 0) + enum fuse_opcode { FUSE_LOOKUP = 1, FUSE_FORGET = 2, /* no reply */ @@ -201,9 +208,11 @@ enum fuse_opcode { FUSE_BMAP = 37, FUSE_DESTROY = 38, FUSE_IOCTL = 39, + FUSE_POLL = 40, }; enum fuse_notify_code { + FUSE_NOTIFY_POLL = 1, FUSE_NOTIFY_CODE_MAX, }; @@ -421,6 +430,22 @@ struct fuse_ioctl_out { __u32 out_iovs; }; +struct fuse_poll_in { + __u64 fh; + __u64 kh; + __u32 flags; + __u32 padding; +}; + +struct fuse_poll_out { + __u32 revents; + __u32 padding; +}; + +struct fuse_notify_poll_wakeup_out { + __u64 kh; +}; + struct fuse_in_header { __u32 len; __u32 opcode; -- cgit v1.2.3 From 1f55ed06cf0c361b293b32e5947d35d173eff2aa Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Mon, 1 Dec 2008 19:14:02 +0100 Subject: fuse: update interface version Change interface version to 7.11 after adding the IOCTL and POLL messages. Also clean up the header a bit: - update copyright date to 2008 - fix checkpatch warning: WARNING: Use #include instead of - remove FUSE_MAJOR define, which is not being used any more Signed-off-by: Miklos Szeredi --- include/linux/fuse.h | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fuse.h b/include/linux/fuse.h index 5650cf033e73..162e5defe683 100644 --- a/include/linux/fuse.h +++ b/include/linux/fuse.h @@ -1,6 +1,6 @@ /* FUSE: Filesystem in Userspace - Copyright (C) 2001-2006 Miklos Szeredi + Copyright (C) 2001-2008 Miklos Szeredi This program can be distributed under the terms of the GNU GPL. See the file COPYING. @@ -20,26 +20,27 @@ * * 7.10 * - add nonseekable open flag + * + * 7.11 + * - add IOCTL message + * - add unsolicited notification support + * - add POLL message and NOTIFY_POLL notification */ #ifndef _LINUX_FUSE_H #define _LINUX_FUSE_H -#include -#include +#include /** Version number of this interface */ #define FUSE_KERNEL_VERSION 7 /** Minor version number of this interface */ -#define FUSE_KERNEL_MINOR_VERSION 10 +#define FUSE_KERNEL_MINOR_VERSION 11 /** The node ID of the root inode */ #define FUSE_ROOT_ID 1 -/** The major number of the fuse character device */ -#define FUSE_MAJOR MISC_MAJOR - /* Make sure all structures are padded to 64bit boundary, so 32bit userspace works under 64bit kernels */ -- cgit v1.2.3 From 69423d99fc182a81f3c5db3eb5c140acc6fc64be Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 10 Dec 2008 13:37:21 +0000 Subject: [MTD] update internal API to support 64-bit device size MTD internal API presently uses 32-bit values to represent device size. This patch updates them to 64-bits but leaves the external API unchanged. Extending the external API is a separate issue for several reasons. First, no one needs it at the moment. Secondly, whether the implementation is done with IOCTLs, sysfs or both is still debated. Thirdly external API changes require the internal API to be accepted first. Note that although the MTD API will be able to support 64-bit device sizes, existing drivers do not and are not required to do so, although NAND base has been updated. In general, changing from 32-bit to 64-bit values cause little or no changes to the majority of the code with the following exceptions: - printk message formats - division and modulus of 64-bit values - NAND base support - 32-bit local variables used by mtdpart and mtdconcat - naughtily assuming one structure maps to another in MEMERASE ioctl Signed-off-by: Adrian Hunter Signed-off-by: Artem Bityutskiy Signed-off-by: David Woodhouse --- drivers/mtd/chips/cfi_cmdset_0001.c | 12 ++++---- drivers/mtd/chips/cfi_cmdset_0002.c | 8 +++--- drivers/mtd/chips/cfi_cmdset_0020.c | 14 ++++----- drivers/mtd/chips/fwh_lock.h | 4 +-- drivers/mtd/inftlcore.c | 2 +- drivers/mtd/inftlmount.c | 4 +-- drivers/mtd/maps/amd76xrom.c | 4 +-- drivers/mtd/maps/ck804xrom.c | 4 +-- drivers/mtd/maps/esb2rom.c | 4 +-- drivers/mtd/maps/ichxrom.c | 4 +-- drivers/mtd/maps/nettel.c | 2 +- drivers/mtd/maps/scb2_flash.c | 8 ++++-- drivers/mtd/mtdchar.c | 6 +++- drivers/mtd/mtdconcat.c | 33 ++++++++++++--------- drivers/mtd/mtdcore.c | 16 ++++++++++- drivers/mtd/mtdoops.c | 9 ++++-- drivers/mtd/mtdpart.c | 34 +++++++++++----------- drivers/mtd/nand/nand_base.c | 24 ++++++++++------ drivers/mtd/nand/nand_bbt.c | 31 ++++++++++---------- drivers/mtd/nftlcore.c | 2 +- drivers/mtd/nftlmount.c | 4 +-- drivers/mtd/onenand/onenand_base.c | 8 +++--- drivers/mtd/rfd_ftl.c | 23 ++++++++------- drivers/mtd/ssfdc.c | 7 +++-- drivers/mtd/ubi/build.c | 2 +- drivers/mtd/ubi/gluebi.c | 17 +++++------ fs/jffs2/erase.c | 5 ++-- include/linux/mtd/mtd.h | 57 +++++++++++++++++++++++++++++++------ include/linux/mtd/nand.h | 2 +- include/linux/mtd/partitions.h | 4 +-- 30 files changed, 216 insertions(+), 138 deletions(-) (limited to 'include/linux') diff --git a/drivers/mtd/chips/cfi_cmdset_0001.c b/drivers/mtd/chips/cfi_cmdset_0001.c index c93a8be5d5f1..f5ab6fa1057b 100644 --- a/drivers/mtd/chips/cfi_cmdset_0001.c +++ b/drivers/mtd/chips/cfi_cmdset_0001.c @@ -58,8 +58,8 @@ static int cfi_intelext_write_buffers(struct mtd_info *, loff_t, size_t, size_t static int cfi_intelext_writev(struct mtd_info *, const struct kvec *, unsigned long, loff_t, size_t *); static int cfi_intelext_erase_varsize(struct mtd_info *, struct erase_info *); static void cfi_intelext_sync (struct mtd_info *); -static int cfi_intelext_lock(struct mtd_info *mtd, loff_t ofs, size_t len); -static int cfi_intelext_unlock(struct mtd_info *mtd, loff_t ofs, size_t len); +static int cfi_intelext_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len); +static int cfi_intelext_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len); #ifdef CONFIG_MTD_OTP static int cfi_intelext_read_fact_prot_reg (struct mtd_info *, loff_t, size_t, size_t *, u_char *); static int cfi_intelext_read_user_prot_reg (struct mtd_info *, loff_t, size_t, size_t *, u_char *); @@ -558,8 +558,8 @@ static struct mtd_info *cfi_intelext_setup(struct mtd_info *mtd) } for (i=0; inumeraseregions;i++){ - printk(KERN_DEBUG "erase region %d: offset=0x%x,size=0x%x,blocks=%d\n", - i,mtd->eraseregions[i].offset, + printk(KERN_DEBUG "erase region %d: offset=0x%llx,size=0x%x,blocks=%d\n", + i,(unsigned long long)mtd->eraseregions[i].offset, mtd->eraseregions[i].erasesize, mtd->eraseregions[i].numblocks); } @@ -2058,7 +2058,7 @@ out: put_chip(map, chip, adr); return ret; } -static int cfi_intelext_lock(struct mtd_info *mtd, loff_t ofs, size_t len) +static int cfi_intelext_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { int ret; @@ -2082,7 +2082,7 @@ static int cfi_intelext_lock(struct mtd_info *mtd, loff_t ofs, size_t len) return ret; } -static int cfi_intelext_unlock(struct mtd_info *mtd, loff_t ofs, size_t len) +static int cfi_intelext_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { int ret; diff --git a/drivers/mtd/chips/cfi_cmdset_0002.c b/drivers/mtd/chips/cfi_cmdset_0002.c index d74ec46aa032..f9c435a42670 100644 --- a/drivers/mtd/chips/cfi_cmdset_0002.c +++ b/drivers/mtd/chips/cfi_cmdset_0002.c @@ -71,8 +71,8 @@ static int get_chip(struct map_info *map, struct flchip *chip, unsigned long adr static void put_chip(struct map_info *map, struct flchip *chip, unsigned long adr); #include "fwh_lock.h" -static int cfi_atmel_lock(struct mtd_info *mtd, loff_t ofs, size_t len); -static int cfi_atmel_unlock(struct mtd_info *mtd, loff_t ofs, size_t len); +static int cfi_atmel_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len); +static int cfi_atmel_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len); static struct mtd_chip_driver cfi_amdstd_chipdrv = { .probe = NULL, /* Not usable directly */ @@ -1774,12 +1774,12 @@ out_unlock: return ret; } -static int cfi_atmel_lock(struct mtd_info *mtd, loff_t ofs, size_t len) +static int cfi_atmel_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { return cfi_varsize_frob(mtd, do_atmel_lock, ofs, len, NULL); } -static int cfi_atmel_unlock(struct mtd_info *mtd, loff_t ofs, size_t len) +static int cfi_atmel_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { return cfi_varsize_frob(mtd, do_atmel_unlock, ofs, len, NULL); } diff --git a/drivers/mtd/chips/cfi_cmdset_0020.c b/drivers/mtd/chips/cfi_cmdset_0020.c index d4714dd9f7ab..6c740f346f91 100644 --- a/drivers/mtd/chips/cfi_cmdset_0020.c +++ b/drivers/mtd/chips/cfi_cmdset_0020.c @@ -42,8 +42,8 @@ static int cfi_staa_writev(struct mtd_info *mtd, const struct kvec *vecs, unsigned long count, loff_t to, size_t *retlen); static int cfi_staa_erase_varsize(struct mtd_info *, struct erase_info *); static void cfi_staa_sync (struct mtd_info *); -static int cfi_staa_lock(struct mtd_info *mtd, loff_t ofs, size_t len); -static int cfi_staa_unlock(struct mtd_info *mtd, loff_t ofs, size_t len); +static int cfi_staa_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len); +static int cfi_staa_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len); static int cfi_staa_suspend (struct mtd_info *); static void cfi_staa_resume (struct mtd_info *); @@ -221,8 +221,8 @@ static struct mtd_info *cfi_staa_setup(struct map_info *map) } for (i=0; inumeraseregions;i++){ - printk(KERN_DEBUG "%d: offset=0x%x,size=0x%x,blocks=%d\n", - i,mtd->eraseregions[i].offset, + printk(KERN_DEBUG "%d: offset=0x%llx,size=0x%x,blocks=%d\n", + i, (unsigned long long)mtd->eraseregions[i].offset, mtd->eraseregions[i].erasesize, mtd->eraseregions[i].numblocks); } @@ -964,7 +964,7 @@ static int cfi_staa_erase_varsize(struct mtd_info *mtd, adr += regions[i].erasesize; len -= regions[i].erasesize; - if (adr % (1<< cfi->chipshift) == ((regions[i].offset + (regions[i].erasesize * regions[i].numblocks)) %( 1<< cfi->chipshift))) + if (adr % (1<< cfi->chipshift) == (((unsigned long)regions[i].offset + (regions[i].erasesize * regions[i].numblocks)) %( 1<< cfi->chipshift))) i++; if (adr >> cfi->chipshift) { @@ -1135,7 +1135,7 @@ retry: spin_unlock_bh(chip->mutex); return 0; } -static int cfi_staa_lock(struct mtd_info *mtd, loff_t ofs, size_t len) +static int cfi_staa_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; @@ -1284,7 +1284,7 @@ retry: spin_unlock_bh(chip->mutex); return 0; } -static int cfi_staa_unlock(struct mtd_info *mtd, loff_t ofs, size_t len) +static int cfi_staa_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { struct map_info *map = mtd->priv; struct cfi_private *cfi = map->fldrv_priv; diff --git a/drivers/mtd/chips/fwh_lock.h b/drivers/mtd/chips/fwh_lock.h index ab44f2b996f8..57e0e4e921f9 100644 --- a/drivers/mtd/chips/fwh_lock.h +++ b/drivers/mtd/chips/fwh_lock.h @@ -77,7 +77,7 @@ static int fwh_xxlock_oneblock(struct map_info *map, struct flchip *chip, } -static int fwh_lock_varsize(struct mtd_info *mtd, loff_t ofs, size_t len) +static int fwh_lock_varsize(struct mtd_info *mtd, loff_t ofs, uint64_t len) { int ret; @@ -88,7 +88,7 @@ static int fwh_lock_varsize(struct mtd_info *mtd, loff_t ofs, size_t len) } -static int fwh_unlock_varsize(struct mtd_info *mtd, loff_t ofs, size_t len) +static int fwh_unlock_varsize(struct mtd_info *mtd, loff_t ofs, uint64_t len) { int ret; diff --git a/drivers/mtd/inftlcore.c b/drivers/mtd/inftlcore.c index 50ce13887f63..73f05227dc8c 100644 --- a/drivers/mtd/inftlcore.c +++ b/drivers/mtd/inftlcore.c @@ -50,7 +50,7 @@ static void inftl_add_mtd(struct mtd_blktrans_ops *tr, struct mtd_info *mtd) struct INFTLrecord *inftl; unsigned long temp; - if (mtd->type != MTD_NANDFLASH) + if (mtd->type != MTD_NANDFLASH || mtd->size > UINT_MAX) return; /* OK, this is moderately ugly. But probably safe. Alternatives? */ if (memcmp(mtd->name, "DiskOnChip", 10)) diff --git a/drivers/mtd/inftlmount.c b/drivers/mtd/inftlmount.c index 9113628ed1ef..f751dd97c549 100644 --- a/drivers/mtd/inftlmount.c +++ b/drivers/mtd/inftlmount.c @@ -63,7 +63,7 @@ static int find_boot_record(struct INFTLrecord *inftl) * otherwise. */ inftl->EraseSize = inftl->mbd.mtd->erasesize; - inftl->nb_blocks = inftl->mbd.mtd->size / inftl->EraseSize; + inftl->nb_blocks = (u32)inftl->mbd.mtd->size / inftl->EraseSize; inftl->MediaUnit = BLOCK_NIL; @@ -187,7 +187,7 @@ static int find_boot_record(struct INFTLrecord *inftl) mh->BlockMultiplierBits); inftl->EraseSize = inftl->mbd.mtd->erasesize << mh->BlockMultiplierBits; - inftl->nb_blocks = inftl->mbd.mtd->size / inftl->EraseSize; + inftl->nb_blocks = (u32)inftl->mbd.mtd->size / inftl->EraseSize; block >>= mh->BlockMultiplierBits; } diff --git a/drivers/mtd/maps/amd76xrom.c b/drivers/mtd/maps/amd76xrom.c index d1eec7d3243f..237733d094c4 100644 --- a/drivers/mtd/maps/amd76xrom.c +++ b/drivers/mtd/maps/amd76xrom.c @@ -232,8 +232,8 @@ static int __devinit amd76xrom_init_one (struct pci_dev *pdev, /* Trim the size if we are larger than the map */ if (map->mtd->size > map->map.size) { printk(KERN_WARNING MOD_NAME - " rom(%u) larger than window(%lu). fixing...\n", - map->mtd->size, map->map.size); + " rom(%llu) larger than window(%lu). fixing...\n", + (unsigned long long)map->mtd->size, map->map.size); map->mtd->size = map->map.size; } if (window->rsrc.parent) { diff --git a/drivers/mtd/maps/ck804xrom.c b/drivers/mtd/maps/ck804xrom.c index 1a6feb4474de..5f7a245ed132 100644 --- a/drivers/mtd/maps/ck804xrom.c +++ b/drivers/mtd/maps/ck804xrom.c @@ -263,8 +263,8 @@ static int __devinit ck804xrom_init_one (struct pci_dev *pdev, /* Trim the size if we are larger than the map */ if (map->mtd->size > map->map.size) { printk(KERN_WARNING MOD_NAME - " rom(%u) larger than window(%lu). fixing...\n", - map->mtd->size, map->map.size); + " rom(%llu) larger than window(%lu). fixing...\n", + (unsigned long long)map->mtd->size, map->map.size); map->mtd->size = map->map.size; } if (window->rsrc.parent) { diff --git a/drivers/mtd/maps/esb2rom.c b/drivers/mtd/maps/esb2rom.c index bbbcdd4c8d13..11a2f57df9cf 100644 --- a/drivers/mtd/maps/esb2rom.c +++ b/drivers/mtd/maps/esb2rom.c @@ -324,8 +324,8 @@ static int __devinit esb2rom_init_one(struct pci_dev *pdev, /* Trim the size if we are larger than the map */ if (map->mtd->size > map->map.size) { printk(KERN_WARNING MOD_NAME - " rom(%u) larger than window(%lu). fixing...\n", - map->mtd->size, map->map.size); + " rom(%llu) larger than window(%lu). fixing...\n", + (unsigned long long)map->mtd->size, map->map.size); map->mtd->size = map->map.size; } if (window->rsrc.parent) { diff --git a/drivers/mtd/maps/ichxrom.c b/drivers/mtd/maps/ichxrom.c index aeb6c916e23f..c32bc28920b3 100644 --- a/drivers/mtd/maps/ichxrom.c +++ b/drivers/mtd/maps/ichxrom.c @@ -258,8 +258,8 @@ static int __devinit ichxrom_init_one (struct pci_dev *pdev, /* Trim the size if we are larger than the map */ if (map->mtd->size > map->map.size) { printk(KERN_WARNING MOD_NAME - " rom(%u) larger than window(%lu). fixing...\n", - map->mtd->size, map->map.size); + " rom(%llu) larger than window(%lu). fixing...\n", + (unsigned long long)map->mtd->size, map->map.size); map->mtd->size = map->map.size; } if (window->rsrc.parent) { diff --git a/drivers/mtd/maps/nettel.c b/drivers/mtd/maps/nettel.c index 965e6c6d6ab0..a8723a6b7457 100644 --- a/drivers/mtd/maps/nettel.c +++ b/drivers/mtd/maps/nettel.c @@ -226,7 +226,7 @@ static int __init nettel_init(void) if ((amd_mtd = do_map_probe("jedec_probe", &nettel_amd_map))) { printk(KERN_NOTICE "SNAPGEAR: AMD flash device size = %dK\n", - amd_mtd->size>>10); + (int)(amd_mtd->size>>10)); amd_mtd->owner = THIS_MODULE; diff --git a/drivers/mtd/maps/scb2_flash.c b/drivers/mtd/maps/scb2_flash.c index 21169e6d646c..7e329f09a548 100644 --- a/drivers/mtd/maps/scb2_flash.c +++ b/drivers/mtd/maps/scb2_flash.c @@ -118,7 +118,8 @@ scb2_fixup_mtd(struct mtd_info *mtd) struct mtd_erase_region_info *region = &mtd->eraseregions[i]; if (region->numblocks * region->erasesize > mtd->size) { - region->numblocks = (mtd->size / region->erasesize); + region->numblocks = ((unsigned long)mtd->size / + region->erasesize); done = 1; } else { region->numblocks = 0; @@ -187,8 +188,9 @@ scb2_flash_probe(struct pci_dev *dev, const struct pci_device_id *ent) return -ENODEV; } - printk(KERN_NOTICE MODNAME ": chip size 0x%x at offset 0x%x\n", - scb2_mtd->size, SCB2_WINDOW - scb2_mtd->size); + printk(KERN_NOTICE MODNAME ": chip size 0x%llx at offset 0x%llx\n", + (unsigned long long)scb2_mtd->size, + (unsigned long long)(SCB2_WINDOW - scb2_mtd->size)); add_mtd_device(scb2_mtd); diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c index bcffeda2df3d..e9ec59e9a566 100644 --- a/drivers/mtd/mtdchar.c +++ b/drivers/mtd/mtdchar.c @@ -450,16 +450,20 @@ static int mtd_ioctl(struct inode *inode, struct file *file, if (!erase) ret = -ENOMEM; else { + struct erase_info_user einfo; + wait_queue_head_t waitq; DECLARE_WAITQUEUE(wait, current); init_waitqueue_head(&waitq); - if (copy_from_user(&erase->addr, argp, + if (copy_from_user(&einfo, argp, sizeof(struct erase_info_user))) { kfree(erase); return -EFAULT; } + erase->addr = einfo.start; + erase->len = einfo.length; erase->mtd = mtd; erase->callback = mtdchar_erase_callback; erase->priv = (unsigned long)&waitq; diff --git a/drivers/mtd/mtdconcat.c b/drivers/mtd/mtdconcat.c index 789842d0e6f2..ff8c14bcac68 100644 --- a/drivers/mtd/mtdconcat.c +++ b/drivers/mtd/mtdconcat.c @@ -197,7 +197,7 @@ concat_writev(struct mtd_info *mtd, const struct kvec *vecs, continue; } - size = min(total_len, (size_t)(subdev->size - to)); + size = min_t(uint64_t, total_len, subdev->size - to); wsize = size; /* store for future use */ entry_high = entry_low; @@ -385,7 +385,7 @@ static int concat_erase(struct mtd_info *mtd, struct erase_info *instr) struct mtd_concat *concat = CONCAT(mtd); struct mtd_info *subdev; int i, err; - u_int32_t length, offset = 0; + uint64_t length, offset = 0; struct erase_info *erase; if (!(mtd->flags & MTD_WRITEABLE)) @@ -518,7 +518,7 @@ static int concat_erase(struct mtd_info *mtd, struct erase_info *instr) return 0; } -static int concat_lock(struct mtd_info *mtd, loff_t ofs, size_t len) +static int concat_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { struct mtd_concat *concat = CONCAT(mtd); int i, err = -EINVAL; @@ -528,7 +528,7 @@ static int concat_lock(struct mtd_info *mtd, loff_t ofs, size_t len) for (i = 0; i < concat->num_subdev; i++) { struct mtd_info *subdev = concat->subdev[i]; - size_t size; + uint64_t size; if (ofs >= subdev->size) { size = 0; @@ -556,7 +556,7 @@ static int concat_lock(struct mtd_info *mtd, loff_t ofs, size_t len) return err; } -static int concat_unlock(struct mtd_info *mtd, loff_t ofs, size_t len) +static int concat_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { struct mtd_concat *concat = CONCAT(mtd); int i, err = 0; @@ -566,7 +566,7 @@ static int concat_unlock(struct mtd_info *mtd, loff_t ofs, size_t len) for (i = 0; i < concat->num_subdev; i++) { struct mtd_info *subdev = concat->subdev[i]; - size_t size; + uint64_t size; if (ofs >= subdev->size) { size = 0; @@ -842,12 +842,14 @@ struct mtd_info *mtd_concat_create(struct mtd_info *subdev[], /* subdevices to c concat->mtd.erasesize = curr_erasesize; concat->mtd.numeraseregions = 0; } else { + uint64_t tmp64; + /* * erase block size varies across the subdevices: allocate * space to store the data describing the variable erase regions */ struct mtd_erase_region_info *erase_region_p; - u_int32_t begin, position; + uint64_t begin, position; concat->mtd.erasesize = max_erasesize; concat->mtd.numeraseregions = num_erase_region; @@ -879,8 +881,9 @@ struct mtd_info *mtd_concat_create(struct mtd_info *subdev[], /* subdevices to c erase_region_p->offset = begin; erase_region_p->erasesize = curr_erasesize; - erase_region_p->numblocks = - (position - begin) / curr_erasesize; + tmp64 = position - begin; + do_div(tmp64, curr_erasesize); + erase_region_p->numblocks = tmp64; begin = position; curr_erasesize = subdev[i]->erasesize; @@ -897,9 +900,9 @@ struct mtd_info *mtd_concat_create(struct mtd_info *subdev[], /* subdevices to c erase_region_p->offset = begin; erase_region_p->erasesize = curr_erasesize; - erase_region_p->numblocks = - (position - - begin) / curr_erasesize; + tmp64 = position - begin; + do_div(tmp64, curr_erasesize); + erase_region_p->numblocks = tmp64; begin = position; curr_erasesize = @@ -909,14 +912,16 @@ struct mtd_info *mtd_concat_create(struct mtd_info *subdev[], /* subdevices to c } position += subdev[i]->eraseregions[j]. - numblocks * curr_erasesize; + numblocks * (uint64_t)curr_erasesize; } } } /* Now write the final entry */ erase_region_p->offset = begin; erase_region_p->erasesize = curr_erasesize; - erase_region_p->numblocks = (position - begin) / curr_erasesize; + tmp64 = position - begin; + do_div(tmp64, curr_erasesize); + erase_region_p->numblocks = tmp64; } return &concat->mtd; diff --git a/drivers/mtd/mtdcore.c b/drivers/mtd/mtdcore.c index a9d246949820..76fe0a1e7a5e 100644 --- a/drivers/mtd/mtdcore.c +++ b/drivers/mtd/mtdcore.c @@ -57,6 +57,19 @@ int add_mtd_device(struct mtd_info *mtd) mtd->index = i; mtd->usecount = 0; + if (is_power_of_2(mtd->erasesize)) + mtd->erasesize_shift = ffs(mtd->erasesize) - 1; + else + mtd->erasesize_shift = 0; + + if (is_power_of_2(mtd->writesize)) + mtd->writesize_shift = ffs(mtd->writesize) - 1; + else + mtd->writesize_shift = 0; + + mtd->erasesize_mask = (1 << mtd->erasesize_shift) - 1; + mtd->writesize_mask = (1 << mtd->writesize_shift) - 1; + /* Some chips always power up locked. Unlock them now */ if ((mtd->flags & MTD_WRITEABLE) && (mtd->flags & MTD_POWERUP_LOCK) && mtd->unlock) { @@ -344,7 +357,8 @@ static inline int mtd_proc_info (char *buf, int i) if (!this) return 0; - return sprintf(buf, "mtd%d: %8.8x %8.8x \"%s\"\n", i, this->size, + return sprintf(buf, "mtd%d: %8.8llx %8.8x \"%s\"\n", i, + (unsigned long long)this->size, this->erasesize, this->name); } diff --git a/drivers/mtd/mtdoops.c b/drivers/mtd/mtdoops.c index aebb3b27edbd..1a6b3beabe8d 100644 --- a/drivers/mtd/mtdoops.c +++ b/drivers/mtd/mtdoops.c @@ -80,9 +80,9 @@ static int mtdoops_erase_block(struct mtd_info *mtd, int offset) if (ret) { set_current_state(TASK_RUNNING); remove_wait_queue(&wait_q, &wait); - printk (KERN_WARNING "mtdoops: erase of region [0x%x, 0x%x] " + printk (KERN_WARNING "mtdoops: erase of region [0x%llx, 0x%llx] " "on \"%s\" failed\n", - erase.addr, erase.len, mtd->name); + (unsigned long long)erase.addr, (unsigned long long)erase.len, mtd->name); return ret; } @@ -289,7 +289,10 @@ static void mtdoops_notify_add(struct mtd_info *mtd) } cxt->mtd = mtd; - cxt->oops_pages = mtd->size / OOPS_PAGE_SIZE; + if (mtd->size > INT_MAX) + cxt->oops_pages = INT_MAX / OOPS_PAGE_SIZE; + else + cxt->oops_pages = (int)mtd->size / OOPS_PAGE_SIZE; find_next_position(cxt); diff --git a/drivers/mtd/mtdpart.c b/drivers/mtd/mtdpart.c index 3728913fa5fa..144e6b613a77 100644 --- a/drivers/mtd/mtdpart.c +++ b/drivers/mtd/mtdpart.c @@ -26,7 +26,7 @@ static LIST_HEAD(mtd_partitions); struct mtd_part { struct mtd_info mtd; struct mtd_info *master; - u_int32_t offset; + uint64_t offset; int index; struct list_head list; int registered; @@ -235,7 +235,7 @@ void mtd_erase_callback(struct erase_info *instr) } EXPORT_SYMBOL_GPL(mtd_erase_callback); -static int part_lock(struct mtd_info *mtd, loff_t ofs, size_t len) +static int part_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { struct mtd_part *part = PART(mtd); if ((len + ofs) > mtd->size) @@ -243,7 +243,7 @@ static int part_lock(struct mtd_info *mtd, loff_t ofs, size_t len) return part->master->lock(part->master, ofs + part->offset, len); } -static int part_unlock(struct mtd_info *mtd, loff_t ofs, size_t len) +static int part_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { struct mtd_part *part = PART(mtd); if ((len + ofs) > mtd->size) @@ -317,7 +317,7 @@ EXPORT_SYMBOL(del_mtd_partitions); static struct mtd_part *add_one_partition(struct mtd_info *master, const struct mtd_partition *part, int partno, - u_int32_t cur_offset) + uint64_t cur_offset) { struct mtd_part *slave; @@ -395,19 +395,19 @@ static struct mtd_part *add_one_partition(struct mtd_info *master, slave->offset = cur_offset; if (slave->offset == MTDPART_OFS_NXTBLK) { slave->offset = cur_offset; - if ((cur_offset % master->erasesize) != 0) { + if (mtd_mod_by_eb(cur_offset, master) != 0) { /* Round up to next erasesize */ - slave->offset = ((cur_offset / master->erasesize) + 1) * master->erasesize; + slave->offset = (mtd_div_by_eb(cur_offset, master) + 1) * master->erasesize; printk(KERN_NOTICE "Moving partition %d: " - "0x%08x -> 0x%08x\n", partno, - cur_offset, slave->offset); + "0x%012llx -> 0x%012llx\n", partno, + (unsigned long long)cur_offset, (unsigned long long)slave->offset); } } if (slave->mtd.size == MTDPART_SIZ_FULL) slave->mtd.size = master->size - slave->offset; - printk(KERN_NOTICE "0x%08x-0x%08x : \"%s\"\n", slave->offset, - slave->offset + slave->mtd.size, slave->mtd.name); + printk(KERN_NOTICE "0x%012llx-0x%012llx : \"%s\"\n", (unsigned long long)slave->offset, + (unsigned long long)(slave->offset + slave->mtd.size), slave->mtd.name); /* let's do some sanity checks */ if (slave->offset >= master->size) { @@ -420,13 +420,13 @@ static struct mtd_part *add_one_partition(struct mtd_info *master, } if (slave->offset + slave->mtd.size > master->size) { slave->mtd.size = master->size - slave->offset; - printk(KERN_WARNING"mtd: partition \"%s\" extends beyond the end of device \"%s\" -- size truncated to %#x\n", - part->name, master->name, slave->mtd.size); + printk(KERN_WARNING"mtd: partition \"%s\" extends beyond the end of device \"%s\" -- size truncated to %#llx\n", + part->name, master->name, (unsigned long long)slave->mtd.size); } if (master->numeraseregions > 1) { /* Deal with variable erase size stuff */ int i, max = master->numeraseregions; - u32 end = slave->offset + slave->mtd.size; + u64 end = slave->offset + slave->mtd.size; struct mtd_erase_region_info *regions = master->eraseregions; /* Find the first erase regions which is part of this @@ -449,7 +449,7 @@ static struct mtd_part *add_one_partition(struct mtd_info *master, } if ((slave->mtd.flags & MTD_WRITEABLE) && - (slave->offset % slave->mtd.erasesize)) { + mtd_mod_by_eb(slave->offset, &slave->mtd)) { /* Doesn't start on a boundary of major erase size */ /* FIXME: Let it be writable if it is on a boundary of * _minor_ erase size though */ @@ -458,7 +458,7 @@ static struct mtd_part *add_one_partition(struct mtd_info *master, part->name); } if ((slave->mtd.flags & MTD_WRITEABLE) && - (slave->mtd.size % slave->mtd.erasesize)) { + mtd_mod_by_eb(slave->mtd.size, &slave->mtd)) { slave->mtd.flags &= ~MTD_WRITEABLE; printk(KERN_WARNING"mtd: partition \"%s\" doesn't end on an erase block -- force read-only\n", part->name); @@ -466,7 +466,7 @@ static struct mtd_part *add_one_partition(struct mtd_info *master, slave->mtd.ecclayout = master->ecclayout; if (master->block_isbad) { - uint32_t offs = 0; + uint64_t offs = 0; while (offs < slave->mtd.size) { if (master->block_isbad(master, @@ -501,7 +501,7 @@ int add_mtd_partitions(struct mtd_info *master, int nbparts) { struct mtd_part *slave; - u_int32_t cur_offset = 0; + uint64_t cur_offset = 0; int i; printk(KERN_NOTICE "Creating %d MTD partitions on \"%s\":\n", nbparts, master->name); diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c index 0a9c9cd33f96..ff2d33e4d6d6 100644 --- a/drivers/mtd/nand/nand_base.c +++ b/drivers/mtd/nand/nand_base.c @@ -2014,13 +2014,14 @@ static int nand_erase(struct mtd_info *mtd, struct erase_info *instr) int nand_erase_nand(struct mtd_info *mtd, struct erase_info *instr, int allowbbt) { - int page, len, status, pages_per_block, ret, chipnr; + int page, status, pages_per_block, ret, chipnr; struct nand_chip *chip = mtd->priv; - int rewrite_bbt[NAND_MAX_CHIPS]={0}; + loff_t rewrite_bbt[NAND_MAX_CHIPS]={0}; unsigned int bbt_masked_page = 0xffffffff; + loff_t len; - DEBUG(MTD_DEBUG_LEVEL3, "nand_erase: start = 0x%08x, len = %i\n", - (unsigned int)instr->addr, (unsigned int)instr->len); + DEBUG(MTD_DEBUG_LEVEL3, "nand_erase: start = 0x%012llx, len = %llu\n", + (unsigned long long)instr->addr, (unsigned long long)instr->len); /* Start address must align on block boundary */ if (instr->addr & ((1 << chip->phys_erase_shift) - 1)) { @@ -2116,7 +2117,8 @@ int nand_erase_nand(struct mtd_info *mtd, struct erase_info *instr, DEBUG(MTD_DEBUG_LEVEL0, "nand_erase: " "Failed erase, page 0x%08x\n", page); instr->state = MTD_ERASE_FAILED; - instr->fail_addr = (page << chip->page_shift); + instr->fail_addr = + ((loff_t)page << chip->page_shift); goto erase_exit; } @@ -2126,7 +2128,8 @@ int nand_erase_nand(struct mtd_info *mtd, struct erase_info *instr, */ if (bbt_masked_page != 0xffffffff && (page & BBT_PAGE_MASK) == bbt_masked_page) - rewrite_bbt[chipnr] = (page << chip->page_shift); + rewrite_bbt[chipnr] = + ((loff_t)page << chip->page_shift); /* Increment page address and decrement length */ len -= (1 << chip->phys_erase_shift); @@ -2173,7 +2176,7 @@ int nand_erase_nand(struct mtd_info *mtd, struct erase_info *instr, continue; /* update the BBT for chip */ DEBUG(MTD_DEBUG_LEVEL0, "nand_erase_nand: nand_update_bbt " - "(%d:0x%0x 0x%0x)\n", chipnr, rewrite_bbt[chipnr], + "(%d:0x%0llx 0x%0x)\n", chipnr, rewrite_bbt[chipnr], chip->bbt_td->pages[chipnr]); nand_update_bbt(mtd, rewrite_bbt[chipnr]); } @@ -2365,7 +2368,7 @@ static struct nand_flash_dev *nand_get_flash_type(struct mtd_info *mtd, if (!mtd->name) mtd->name = type->name; - chip->chipsize = type->chipsize << 20; + chip->chipsize = (uint64_t)type->chipsize << 20; /* Newer devices have all the information in additional id bytes */ if (!type->pagesize) { @@ -2423,7 +2426,10 @@ static struct nand_flash_dev *nand_get_flash_type(struct mtd_info *mtd, chip->bbt_erase_shift = chip->phys_erase_shift = ffs(mtd->erasesize) - 1; - chip->chip_shift = ffs(chip->chipsize) - 1; + if (chip->chipsize & 0xffffffff) + chip->chip_shift = ffs((unsigned)chip->chipsize) - 1; + else + chip->chip_shift = ffs((unsigned)(chip->chipsize >> 32)) + 32 - 1; /* Set the bad block position */ chip->badblockpos = mtd->writesize > 512 ? diff --git a/drivers/mtd/nand/nand_bbt.c b/drivers/mtd/nand/nand_bbt.c index 0b1c48595f12..55c23e5cd210 100644 --- a/drivers/mtd/nand/nand_bbt.c +++ b/drivers/mtd/nand/nand_bbt.c @@ -171,16 +171,16 @@ static int read_bbt(struct mtd_info *mtd, uint8_t *buf, int page, int num, if (tmp == msk) continue; if (reserved_block_code && (tmp == reserved_block_code)) { - printk(KERN_DEBUG "nand_read_bbt: Reserved block at 0x%08x\n", - ((offs << 2) + (act >> 1)) << this->bbt_erase_shift); + printk(KERN_DEBUG "nand_read_bbt: Reserved block at 0x%012llx\n", + (loff_t)((offs << 2) + (act >> 1)) << this->bbt_erase_shift); this->bbt[offs + (act >> 3)] |= 0x2 << (act & 0x06); mtd->ecc_stats.bbtblocks++; continue; } /* Leave it for now, if its matured we can move this * message to MTD_DEBUG_LEVEL0 */ - printk(KERN_DEBUG "nand_read_bbt: Bad block at 0x%08x\n", - ((offs << 2) + (act >> 1)) << this->bbt_erase_shift); + printk(KERN_DEBUG "nand_read_bbt: Bad block at 0x%012llx\n", + (loff_t)((offs << 2) + (act >> 1)) << this->bbt_erase_shift); /* Factory marked bad or worn out ? */ if (tmp == 0) this->bbt[offs + (act >> 3)] |= 0x3 << (act & 0x06); @@ -284,7 +284,7 @@ static int read_abs_bbts(struct mtd_info *mtd, uint8_t *buf, /* Read the primary version, if available */ if (td->options & NAND_BBT_VERSION) { - scan_read_raw(mtd, buf, td->pages[0] << this->page_shift, + scan_read_raw(mtd, buf, (loff_t)td->pages[0] << this->page_shift, mtd->writesize); td->version[0] = buf[mtd->writesize + td->veroffs]; printk(KERN_DEBUG "Bad block table at page %d, version 0x%02X\n", @@ -293,7 +293,7 @@ static int read_abs_bbts(struct mtd_info *mtd, uint8_t *buf, /* Read the mirror version, if available */ if (md && (md->options & NAND_BBT_VERSION)) { - scan_read_raw(mtd, buf, md->pages[0] << this->page_shift, + scan_read_raw(mtd, buf, (loff_t)md->pages[0] << this->page_shift, mtd->writesize); md->version[0] = buf[mtd->writesize + md->veroffs]; printk(KERN_DEBUG "Bad block table at page %d, version 0x%02X\n", @@ -411,7 +411,7 @@ static int create_bbt(struct mtd_info *mtd, uint8_t *buf, numblocks = this->chipsize >> (this->bbt_erase_shift - 1); startblock = chip * numblocks; numblocks += startblock; - from = startblock << (this->bbt_erase_shift - 1); + from = (loff_t)startblock << (this->bbt_erase_shift - 1); } for (i = startblock; i < numblocks;) { @@ -428,8 +428,8 @@ static int create_bbt(struct mtd_info *mtd, uint8_t *buf, if (ret) { this->bbt[i >> 3] |= 0x03 << (i & 0x6); - printk(KERN_WARNING "Bad eraseblock %d at 0x%08x\n", - i >> 1, (unsigned int)from); + printk(KERN_WARNING "Bad eraseblock %d at 0x%012llx\n", + i >> 1, (unsigned long long)from); mtd->ecc_stats.badblocks++; } @@ -495,7 +495,7 @@ static int search_bbt(struct mtd_info *mtd, uint8_t *buf, struct nand_bbt_descr for (block = 0; block < td->maxblocks; block++) { int actblock = startblock + dir * block; - loff_t offs = actblock << this->bbt_erase_shift; + loff_t offs = (loff_t)actblock << this->bbt_erase_shift; /* Read first page */ scan_read_raw(mtd, buf, offs, mtd->writesize); @@ -719,7 +719,7 @@ static int write_bbt(struct mtd_info *mtd, uint8_t *buf, memset(&einfo, 0, sizeof(einfo)); einfo.mtd = mtd; - einfo.addr = (unsigned long)to; + einfo.addr = to; einfo.len = 1 << this->bbt_erase_shift; res = nand_erase_nand(mtd, &einfo, 1); if (res < 0) @@ -729,8 +729,8 @@ static int write_bbt(struct mtd_info *mtd, uint8_t *buf, if (res < 0) goto outerr; - printk(KERN_DEBUG "Bad block table written to 0x%08x, version " - "0x%02X\n", (unsigned int)to, td->version[chip]); + printk(KERN_DEBUG "Bad block table written to 0x%012llx, version " + "0x%02X\n", (unsigned long long)to, td->version[chip]); /* Mark it as used */ td->pages[chip] = page; @@ -910,7 +910,7 @@ static void mark_bbt_region(struct mtd_info *mtd, struct nand_bbt_descr *td) newval = oldval | (0x2 << (block & 0x06)); this->bbt[(block >> 3)] = newval; if ((oldval != newval) && td->reserved_block_code) - nand_update_bbt(mtd, block << (this->bbt_erase_shift - 1)); + nand_update_bbt(mtd, (loff_t)block << (this->bbt_erase_shift - 1)); continue; } update = 0; @@ -931,7 +931,7 @@ static void mark_bbt_region(struct mtd_info *mtd, struct nand_bbt_descr *td) new ones have been marked, then we need to update the stored bbts. This should only happen once. */ if (update && td->reserved_block_code) - nand_update_bbt(mtd, (block - 2) << (this->bbt_erase_shift - 1)); + nand_update_bbt(mtd, (loff_t)(block - 2) << (this->bbt_erase_shift - 1)); } } @@ -1027,7 +1027,6 @@ int nand_update_bbt(struct mtd_info *mtd, loff_t offs) if (!this->bbt || !td) return -EINVAL; - len = mtd->size >> (this->bbt_erase_shift + 2); /* Allocate a temporary buffer for one eraseblock incl. oob */ len = (1 << this->bbt_erase_shift); len += (len >> this->page_shift) * mtd->oobsize; diff --git a/drivers/mtd/nftlcore.c b/drivers/mtd/nftlcore.c index 320b929abe79..d1c4546513f7 100644 --- a/drivers/mtd/nftlcore.c +++ b/drivers/mtd/nftlcore.c @@ -39,7 +39,7 @@ static void nftl_add_mtd(struct mtd_blktrans_ops *tr, struct mtd_info *mtd) struct NFTLrecord *nftl; unsigned long temp; - if (mtd->type != MTD_NANDFLASH) + if (mtd->type != MTD_NANDFLASH || mtd->size > UINT_MAX) return; /* OK, this is moderately ugly. But probably safe. Alternatives? */ if (memcmp(mtd->name, "DiskOnChip", 10)) diff --git a/drivers/mtd/nftlmount.c b/drivers/mtd/nftlmount.c index ccc4f209fbb5..8b22b1836e9f 100644 --- a/drivers/mtd/nftlmount.c +++ b/drivers/mtd/nftlmount.c @@ -51,7 +51,7 @@ static int find_boot_record(struct NFTLrecord *nftl) the mtd device accordingly. We could even get rid of nftl->EraseSize if there were any point in doing so. */ nftl->EraseSize = nftl->mbd.mtd->erasesize; - nftl->nb_blocks = nftl->mbd.mtd->size / nftl->EraseSize; + nftl->nb_blocks = (u32)nftl->mbd.mtd->size / nftl->EraseSize; nftl->MediaUnit = BLOCK_NIL; nftl->SpareMediaUnit = BLOCK_NIL; @@ -168,7 +168,7 @@ device is already correct. printk(KERN_NOTICE "WARNING: Support for NFTL with UnitSizeFactor 0x%02x is experimental\n", mh->UnitSizeFactor); nftl->EraseSize = nftl->mbd.mtd->erasesize << (0xff - mh->UnitSizeFactor); - nftl->nb_blocks = nftl->mbd.mtd->size / nftl->EraseSize; + nftl->nb_blocks = (u32)nftl->mbd.mtd->size / nftl->EraseSize; } #endif nftl->nb_boot_blocks = le16_to_cpu(mh->FirstPhysicalEUN); diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c index 90ed319f26e6..529af271db17 100644 --- a/drivers/mtd/onenand/onenand_base.c +++ b/drivers/mtd/onenand/onenand_base.c @@ -1772,7 +1772,7 @@ static int onenand_erase(struct mtd_info *mtd, struct erase_info *instr) int len; int ret = 0; - DEBUG(MTD_DEBUG_LEVEL3, "onenand_erase: start = 0x%08x, len = %i\n", (unsigned int) instr->addr, (unsigned int) instr->len); + DEBUG(MTD_DEBUG_LEVEL3, "onenand_erase: start = 0x%012llx, len = %llu\n", (unsigned long long) instr->addr, (unsigned long long) instr->len); block_size = (1 << this->erase_shift); @@ -1810,7 +1810,7 @@ static int onenand_erase(struct mtd_info *mtd, struct erase_info *instr) /* Check if we have a bad block, we do not erase bad blocks */ if (onenand_block_isbad_nolock(mtd, addr, 0)) { - printk (KERN_WARNING "onenand_erase: attempt to erase a bad block at addr 0x%08x\n", (unsigned int) addr); + printk (KERN_WARNING "onenand_erase: attempt to erase a bad block at addr 0x%012llx\n", (unsigned long long) addr); instr->state = MTD_ERASE_FAILED; goto erase_exit; } @@ -2029,7 +2029,7 @@ static int onenand_do_lock_cmd(struct mtd_info *mtd, loff_t ofs, size_t len, int * * Lock one or more blocks */ -static int onenand_lock(struct mtd_info *mtd, loff_t ofs, size_t len) +static int onenand_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { int ret; @@ -2047,7 +2047,7 @@ static int onenand_lock(struct mtd_info *mtd, loff_t ofs, size_t len) * * Unlock one or more blocks */ -static int onenand_unlock(struct mtd_info *mtd, loff_t ofs, size_t len) +static int onenand_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { int ret; diff --git a/drivers/mtd/rfd_ftl.c b/drivers/mtd/rfd_ftl.c index e538c0a72abb..490b4742ce3a 100644 --- a/drivers/mtd/rfd_ftl.c +++ b/drivers/mtd/rfd_ftl.c @@ -156,7 +156,7 @@ static int scan_header(struct partition *part) size_t retlen; sectors_per_block = part->block_size / SECTOR_SIZE; - part->total_blocks = part->mbd.mtd->size / part->block_size; + part->total_blocks = (u32)part->mbd.mtd->size / part->block_size; if (part->total_blocks < 2) return -ENOENT; @@ -276,16 +276,17 @@ static void erase_callback(struct erase_info *erase) part = (struct partition*)erase->priv; - i = erase->addr / part->block_size; - if (i >= part->total_blocks || part->blocks[i].offset != erase->addr) { - printk(KERN_ERR PREFIX "erase callback for unknown offset %x " - "on '%s'\n", erase->addr, part->mbd.mtd->name); + i = (u32)erase->addr / part->block_size; + if (i >= part->total_blocks || part->blocks[i].offset != erase->addr || + erase->addr > UINT_MAX) { + printk(KERN_ERR PREFIX "erase callback for unknown offset %llx " + "on '%s'\n", (unsigned long long)erase->addr, part->mbd.mtd->name); return; } if (erase->state != MTD_ERASE_DONE) { - printk(KERN_WARNING PREFIX "erase failed at 0x%x on '%s', " - "state %d\n", erase->addr, + printk(KERN_WARNING PREFIX "erase failed at 0x%llx on '%s', " + "state %d\n", (unsigned long long)erase->addr, part->mbd.mtd->name, erase->state); part->blocks[i].state = BLOCK_FAILED; @@ -345,9 +346,9 @@ static int erase_block(struct partition *part, int block) rc = part->mbd.mtd->erase(part->mbd.mtd, erase); if (rc) { - printk(KERN_ERR PREFIX "erase of region %x,%x on '%s' " - "failed\n", erase->addr, erase->len, - part->mbd.mtd->name); + printk(KERN_ERR PREFIX "erase of region %llx,%llx on '%s' " + "failed\n", (unsigned long long)erase->addr, + (unsigned long long)erase->len, part->mbd.mtd->name); kfree(erase); } @@ -763,7 +764,7 @@ static void rfd_ftl_add_mtd(struct mtd_blktrans_ops *tr, struct mtd_info *mtd) { struct partition *part; - if (mtd->type != MTD_NORFLASH) + if (mtd->type != MTD_NORFLASH || mtd->size > UINT_MAX) return; part = kzalloc(sizeof(struct partition), GFP_KERNEL); diff --git a/drivers/mtd/ssfdc.c b/drivers/mtd/ssfdc.c index 33a5d6ed6f18..3f67e00d98e0 100644 --- a/drivers/mtd/ssfdc.c +++ b/drivers/mtd/ssfdc.c @@ -294,7 +294,8 @@ static void ssfdcr_add_mtd(struct mtd_blktrans_ops *tr, struct mtd_info *mtd) int cis_sector; /* Check for small page NAND flash */ - if (mtd->type != MTD_NANDFLASH || mtd->oobsize != OOB_SIZE) + if (mtd->type != MTD_NANDFLASH || mtd->oobsize != OOB_SIZE || + mtd->size > UINT_MAX) return; /* Check for SSDFC format by reading CIS/IDI sector */ @@ -316,7 +317,7 @@ static void ssfdcr_add_mtd(struct mtd_blktrans_ops *tr, struct mtd_info *mtd) ssfdc->cis_block = cis_sector / (mtd->erasesize >> SECTOR_SHIFT); ssfdc->erase_size = mtd->erasesize; - ssfdc->map_len = mtd->size / mtd->erasesize; + ssfdc->map_len = (u32)mtd->size / mtd->erasesize; DEBUG(MTD_DEBUG_LEVEL1, "SSFDC_RO: cis_block=%d,erase_size=%d,map_len=%d,n_zones=%d\n", @@ -327,7 +328,7 @@ static void ssfdcr_add_mtd(struct mtd_blktrans_ops *tr, struct mtd_info *mtd) ssfdc->heads = 16; ssfdc->sectors = 32; get_chs(mtd->size, NULL, &ssfdc->heads, &ssfdc->sectors); - ssfdc->cylinders = (unsigned short)((mtd->size >> SECTOR_SHIFT) / + ssfdc->cylinders = (unsigned short)(((u32)mtd->size >> SECTOR_SHIFT) / ((long)ssfdc->sectors * (long)ssfdc->heads)); DEBUG(MTD_DEBUG_LEVEL1, "SSFDC_RO: using C:%d H:%d S:%d == %ld sects\n", diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index c7630a228310..634e2e86525f 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -561,7 +561,7 @@ static int io_init(struct ubi_device *ubi) */ ubi->peb_size = ubi->mtd->erasesize; - ubi->peb_count = ubi->mtd->size / ubi->mtd->erasesize; + ubi->peb_count = mtd_div_by_eb(ubi->mtd->size, ubi->mtd); ubi->flash_size = ubi->mtd->size; if (ubi->mtd->block_isbad && ubi->mtd->block_markbad) diff --git a/drivers/mtd/ubi/gluebi.c b/drivers/mtd/ubi/gluebi.c index 605812bb0b1a..6dd4f5e77f82 100644 --- a/drivers/mtd/ubi/gluebi.c +++ b/drivers/mtd/ubi/gluebi.c @@ -215,7 +215,8 @@ static int gluebi_erase(struct mtd_info *mtd, struct erase_info *instr) struct ubi_volume *vol; struct ubi_device *ubi; - dbg_gen("erase %u bytes at offset %u", instr->len, instr->addr); + dbg_gen("erase %llu bytes at offset %llu", (unsigned long long)instr->len, + (unsigned long long)instr->addr); if (instr->addr < 0 || instr->addr > mtd->size - mtd->erasesize) return -EINVAL; @@ -223,11 +224,11 @@ static int gluebi_erase(struct mtd_info *mtd, struct erase_info *instr) if (instr->len < 0 || instr->addr + instr->len > mtd->size) return -EINVAL; - if (instr->addr % mtd->writesize || instr->len % mtd->writesize) + if (mtd_mod_by_ws(instr->addr, mtd) || mtd_mod_by_ws(instr->len, mtd)) return -EINVAL; - lnum = instr->addr / mtd->erasesize; - count = instr->len / mtd->erasesize; + lnum = mtd_div_by_eb(instr->addr, mtd); + count = mtd_div_by_eb(instr->len, mtd); vol = container_of(mtd, struct ubi_volume, gluebi_mtd); ubi = vol->ubi; @@ -255,7 +256,7 @@ static int gluebi_erase(struct mtd_info *mtd, struct erase_info *instr) out_err: instr->state = MTD_ERASE_FAILED; - instr->fail_addr = lnum * mtd->erasesize; + instr->fail_addr = (long long)lnum * mtd->erasesize; return err; } @@ -294,7 +295,7 @@ int ubi_create_gluebi(struct ubi_device *ubi, struct ubi_volume *vol) * bytes. */ if (vol->vol_type == UBI_DYNAMIC_VOLUME) - mtd->size = vol->usable_leb_size * vol->reserved_pebs; + mtd->size = (long long)vol->usable_leb_size * vol->reserved_pebs; else mtd->size = vol->used_bytes; @@ -304,8 +305,8 @@ int ubi_create_gluebi(struct ubi_device *ubi, struct ubi_volume *vol) return -ENFILE; } - dbg_gen("added mtd%d (\"%s\"), size %u, EB size %u", - mtd->index, mtd->name, mtd->size, mtd->erasesize); + dbg_gen("added mtd%d (\"%s\"), size %llu, EB size %u", + mtd->index, mtd->name, (unsigned long long)mtd->size, mtd->erasesize); return 0; } diff --git a/fs/jffs2/erase.c b/fs/jffs2/erase.c index 259461b910af..c32b4a1ad6cf 100644 --- a/fs/jffs2/erase.c +++ b/fs/jffs2/erase.c @@ -175,7 +175,7 @@ static void jffs2_erase_failed(struct jffs2_sb_info *c, struct jffs2_eraseblock { /* For NAND, if the failure did not occur at the device level for a specific physical page, don't bother updating the bad block table. */ - if (jffs2_cleanmarker_oob(c) && (bad_offset != MTD_FAIL_ADDR_UNKNOWN)) { + if (jffs2_cleanmarker_oob(c) && (bad_offset != (uint32_t)MTD_FAIL_ADDR_UNKNOWN)) { /* We had a device-level failure to erase. Let's see if we've failed too many times. */ if (!jffs2_write_nand_badblock(c, jeb, bad_offset)) { @@ -209,7 +209,8 @@ static void jffs2_erase_callback(struct erase_info *instr) struct erase_priv_struct *priv = (void *)instr->priv; if(instr->state != MTD_ERASE_DONE) { - printk(KERN_WARNING "Erase at 0x%08x finished, but state != MTD_ERASE_DONE. State is 0x%x instead.\n", instr->addr, instr->state); + printk(KERN_WARNING "Erase at 0x%08llx finished, but state != MTD_ERASE_DONE. State is 0x%x instead.\n", + (unsigned long long)instr->addr, instr->state); jffs2_erase_failed(priv->c, priv->jeb, instr->fail_addr); } else { jffs2_erase_succeeded(priv->c, priv->jeb); diff --git a/include/linux/mtd/mtd.h b/include/linux/mtd/mtd.h index eae26bb6430a..95e585ecc297 100644 --- a/include/linux/mtd/mtd.h +++ b/include/linux/mtd/mtd.h @@ -15,6 +15,8 @@ #include #include +#include + #define MTD_CHAR_MAJOR 90 #define MTD_BLOCK_MAJOR 31 #define MAX_MTD_DEVICES 32 @@ -25,16 +27,16 @@ #define MTD_ERASE_DONE 0x08 #define MTD_ERASE_FAILED 0x10 -#define MTD_FAIL_ADDR_UNKNOWN 0xffffffff +#define MTD_FAIL_ADDR_UNKNOWN -1LL /* If the erase fails, fail_addr might indicate exactly which block failed. If fail_addr = MTD_FAIL_ADDR_UNKNOWN, the failure was not at the device level or was not specific to any particular block. */ struct erase_info { struct mtd_info *mtd; - u_int32_t addr; - u_int32_t len; - u_int32_t fail_addr; + uint64_t addr; + uint64_t len; + uint64_t fail_addr; u_long time; u_long retries; u_int dev; @@ -46,7 +48,7 @@ struct erase_info { }; struct mtd_erase_region_info { - u_int32_t offset; /* At which this region starts, from the beginning of the MTD */ + uint64_t offset; /* At which this region starts, from the beginning of the MTD */ u_int32_t erasesize; /* For this region */ u_int32_t numblocks; /* Number of blocks of erasesize in this region */ unsigned long *lockmap; /* If keeping bitmap of locks */ @@ -101,7 +103,7 @@ struct mtd_oob_ops { struct mtd_info { u_char type; u_int32_t flags; - u_int32_t size; // Total size of the MTD + uint64_t size; // Total size of the MTD /* "Major" erase size for the device. Naïve users may take this * to be the only erase size available, or may use the more detailed @@ -120,6 +122,16 @@ struct mtd_info { u_int32_t oobsize; // Amount of OOB data per block (e.g. 16) u_int32_t oobavail; // Available OOB bytes per block + /* + * If erasesize is a power of 2 then the shift is stored in + * erasesize_shift otherwise erasesize_shift is zero. Ditto writesize. + */ + unsigned int erasesize_shift; + unsigned int writesize_shift; + /* Masks based on erasesize_shift and writesize_shift */ + unsigned int erasesize_mask; + unsigned int writesize_mask; + // Kernel-only stuff starts here. const char *name; int index; @@ -190,8 +202,8 @@ struct mtd_info { void (*sync) (struct mtd_info *mtd); /* Chip-supported device locking */ - int (*lock) (struct mtd_info *mtd, loff_t ofs, size_t len); - int (*unlock) (struct mtd_info *mtd, loff_t ofs, size_t len); + int (*lock) (struct mtd_info *mtd, loff_t ofs, uint64_t len); + int (*unlock) (struct mtd_info *mtd, loff_t ofs, uint64_t len); /* Power Management functions */ int (*suspend) (struct mtd_info *mtd); @@ -221,6 +233,35 @@ struct mtd_info { void (*put_device) (struct mtd_info *mtd); }; +static inline u_int32_t mtd_div_by_eb(uint64_t sz, struct mtd_info *mtd) +{ + if (mtd->erasesize_shift) + return sz >> mtd->erasesize_shift; + do_div(sz, mtd->erasesize); + return sz; +} + +static inline u_int32_t mtd_mod_by_eb(uint64_t sz, struct mtd_info *mtd) +{ + if (mtd->erasesize_shift) + return sz & mtd->erasesize_mask; + return do_div(sz, mtd->erasesize); +} + +static inline u_int32_t mtd_div_by_ws(uint64_t sz, struct mtd_info *mtd) +{ + if (mtd->writesize_shift) + return sz >> mtd->writesize_shift; + do_div(sz, mtd->writesize); + return sz; +} + +static inline u_int32_t mtd_mod_by_ws(uint64_t sz, struct mtd_info *mtd) +{ + if (mtd->writesize_shift) + return sz & mtd->writesize_mask; + return do_div(sz, mtd->writesize); +} /* Kernel-side ioctl definitions */ diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h index 733d3f3b4eb8..c0677b8082be 100644 --- a/include/linux/mtd/nand.h +++ b/include/linux/mtd/nand.h @@ -399,7 +399,7 @@ struct nand_chip { int bbt_erase_shift; int chip_shift; int numchips; - unsigned long chipsize; + uint64_t chipsize; int pagemask; int pagebuf; int subpagesize; diff --git a/include/linux/mtd/partitions.h b/include/linux/mtd/partitions.h index c92b4d439609..164c7d78687d 100644 --- a/include/linux/mtd/partitions.h +++ b/include/linux/mtd/partitions.h @@ -36,8 +36,8 @@ struct mtd_partition { char *name; /* identifier string */ - u_int32_t size; /* partition size */ - u_int32_t offset; /* offset within the master MTD space */ + uint64_t size; /* partition size */ + uint64_t offset; /* offset within the master MTD space */ u_int32_t mask_flags; /* master MTD flags to mask out for this partition */ struct nand_ecclayout *ecclayout; /* out of band layout for this partition (NAND only)*/ struct mtd_info **mtdp; /* pointer to store the MTD object */ -- cgit v1.2.3 From 3854be7712f7b4bdcaed14664fc7c7124b3fef0d Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 10 Dec 2008 14:06:42 +0000 Subject: [MTD] Remove strange u_int32_t types from FTL Signed-off-by: David Woodhouse --- drivers/mtd/ftl.c | 100 ++++++++++++++++++++++++------------------------ include/linux/mtd/ftl.h | 38 +++++++++--------- 2 files changed, 69 insertions(+), 69 deletions(-) (limited to 'include/linux') diff --git a/drivers/mtd/ftl.c b/drivers/mtd/ftl.c index 9bf581c4f740..a790c062af1f 100644 --- a/drivers/mtd/ftl.c +++ b/drivers/mtd/ftl.c @@ -109,25 +109,25 @@ module_param(shuffle_freq, int, 0); /* Each memory region corresponds to a minor device */ typedef struct partition_t { struct mtd_blktrans_dev mbd; - u_int32_t state; - u_int32_t *VirtualBlockMap; - u_int32_t *VirtualPageMap; - u_int32_t FreeTotal; + uint32_t state; + uint32_t *VirtualBlockMap; + uint32_t *VirtualPageMap; + uint32_t FreeTotal; struct eun_info_t { - u_int32_t Offset; - u_int32_t EraseCount; - u_int32_t Free; - u_int32_t Deleted; + uint32_t Offset; + uint32_t EraseCount; + uint32_t Free; + uint32_t Deleted; } *EUNInfo; struct xfer_info_t { - u_int32_t Offset; - u_int32_t EraseCount; - u_int16_t state; + uint32_t Offset; + uint32_t EraseCount; + uint16_t state; } *XferInfo; - u_int16_t bam_index; - u_int32_t *bam_cache; - u_int16_t DataUnits; - u_int32_t BlocksPerUnit; + uint16_t bam_index; + uint32_t *bam_cache; + uint16_t DataUnits; + uint32_t BlocksPerUnit; erase_unit_header_t header; } partition_t; @@ -199,8 +199,8 @@ static int scan_header(partition_t *part) static int build_maps(partition_t *part) { erase_unit_header_t header; - u_int16_t xvalid, xtrans, i; - u_int blocks, j; + uint16_t xvalid, xtrans, i; + unsigned blocks, j; int hdr_ok, ret = -1; ssize_t retval; loff_t offset; @@ -269,14 +269,14 @@ static int build_maps(partition_t *part) /* Set up virtual page map */ blocks = le32_to_cpu(header.FormattedSize) >> header.BlockSize; - part->VirtualBlockMap = vmalloc(blocks * sizeof(u_int32_t)); + part->VirtualBlockMap = vmalloc(blocks * sizeof(uint32_t)); if (!part->VirtualBlockMap) goto out_XferInfo; - memset(part->VirtualBlockMap, 0xff, blocks * sizeof(u_int32_t)); + memset(part->VirtualBlockMap, 0xff, blocks * sizeof(uint32_t)); part->BlocksPerUnit = (1 << header.EraseUnitSize) >> header.BlockSize; - part->bam_cache = kmalloc(part->BlocksPerUnit * sizeof(u_int32_t), + part->bam_cache = kmalloc(part->BlocksPerUnit * sizeof(uint32_t), GFP_KERNEL); if (!part->bam_cache) goto out_VirtualBlockMap; @@ -290,7 +290,7 @@ static int build_maps(partition_t *part) offset = part->EUNInfo[i].Offset + le32_to_cpu(header.BAMOffset); ret = part->mbd.mtd->read(part->mbd.mtd, offset, - part->BlocksPerUnit * sizeof(u_int32_t), &retval, + part->BlocksPerUnit * sizeof(uint32_t), &retval, (unsigned char *)part->bam_cache); if (ret) @@ -332,7 +332,7 @@ out: ======================================================================*/ static int erase_xfer(partition_t *part, - u_int16_t xfernum) + uint16_t xfernum) { int ret; struct xfer_info_t *xfer; @@ -408,7 +408,7 @@ static int prepare_xfer(partition_t *part, int i) erase_unit_header_t header; struct xfer_info_t *xfer; int nbam, ret; - u_int32_t ctl; + uint32_t ctl; ssize_t retlen; loff_t offset; @@ -430,15 +430,15 @@ static int prepare_xfer(partition_t *part, int i) } /* Write the BAM stub */ - nbam = (part->BlocksPerUnit * sizeof(u_int32_t) + + nbam = (part->BlocksPerUnit * sizeof(uint32_t) + le32_to_cpu(part->header.BAMOffset) + SECTOR_SIZE - 1) / SECTOR_SIZE; offset = xfer->Offset + le32_to_cpu(part->header.BAMOffset); ctl = cpu_to_le32(BLOCK_CONTROL); - for (i = 0; i < nbam; i++, offset += sizeof(u_int32_t)) { + for (i = 0; i < nbam; i++, offset += sizeof(uint32_t)) { - ret = part->mbd.mtd->write(part->mbd.mtd, offset, sizeof(u_int32_t), + ret = part->mbd.mtd->write(part->mbd.mtd, offset, sizeof(uint32_t), &retlen, (u_char *)&ctl); if (ret) @@ -461,18 +461,18 @@ static int prepare_xfer(partition_t *part, int i) ======================================================================*/ -static int copy_erase_unit(partition_t *part, u_int16_t srcunit, - u_int16_t xferunit) +static int copy_erase_unit(partition_t *part, uint16_t srcunit, + uint16_t xferunit) { u_char buf[SECTOR_SIZE]; struct eun_info_t *eun; struct xfer_info_t *xfer; - u_int32_t src, dest, free, i; - u_int16_t unit; + uint32_t src, dest, free, i; + uint16_t unit; int ret; ssize_t retlen; loff_t offset; - u_int16_t srcunitswap = cpu_to_le16(srcunit); + uint16_t srcunitswap = cpu_to_le16(srcunit); eun = &part->EUNInfo[srcunit]; xfer = &part->XferInfo[xferunit]; @@ -486,7 +486,7 @@ static int copy_erase_unit(partition_t *part, u_int16_t srcunit, offset = eun->Offset + le32_to_cpu(part->header.BAMOffset); ret = part->mbd.mtd->read(part->mbd.mtd, offset, - part->BlocksPerUnit * sizeof(u_int32_t), + part->BlocksPerUnit * sizeof(uint32_t), &retlen, (u_char *) (part->bam_cache)); /* mark the cache bad, in case we get an error later */ @@ -503,7 +503,7 @@ static int copy_erase_unit(partition_t *part, u_int16_t srcunit, offset = xfer->Offset + 20; /* Bad! */ unit = cpu_to_le16(0x7fff); - ret = part->mbd.mtd->write(part->mbd.mtd, offset, sizeof(u_int16_t), + ret = part->mbd.mtd->write(part->mbd.mtd, offset, sizeof(uint16_t), &retlen, (u_char *) &unit); if (ret) { @@ -560,7 +560,7 @@ static int copy_erase_unit(partition_t *part, u_int16_t srcunit, /* All clear? Then update the LogicalEUN again */ - ret = part->mbd.mtd->write(part->mbd.mtd, xfer->Offset + 20, sizeof(u_int16_t), + ret = part->mbd.mtd->write(part->mbd.mtd, xfer->Offset + 20, sizeof(uint16_t), &retlen, (u_char *)&srcunitswap); if (ret) { @@ -605,8 +605,8 @@ static int copy_erase_unit(partition_t *part, u_int16_t srcunit, static int reclaim_block(partition_t *part) { - u_int16_t i, eun, xfer; - u_int32_t best; + uint16_t i, eun, xfer; + uint32_t best; int queued, ret; DEBUG(0, "ftl_cs: reclaiming space...\n"); @@ -723,10 +723,10 @@ static void dump_lists(partition_t *part) } #endif -static u_int32_t find_free(partition_t *part) +static uint32_t find_free(partition_t *part) { - u_int16_t stop, eun; - u_int32_t blk; + uint16_t stop, eun; + uint32_t blk; size_t retlen; int ret; @@ -749,7 +749,7 @@ static u_int32_t find_free(partition_t *part) ret = part->mbd.mtd->read(part->mbd.mtd, part->EUNInfo[eun].Offset + le32_to_cpu(part->header.BAMOffset), - part->BlocksPerUnit * sizeof(u_int32_t), + part->BlocksPerUnit * sizeof(uint32_t), &retlen, (u_char *) (part->bam_cache)); if (ret) { @@ -786,7 +786,7 @@ static u_int32_t find_free(partition_t *part) static int ftl_read(partition_t *part, caddr_t buffer, u_long sector, u_long nblocks) { - u_int32_t log_addr, bsize; + uint32_t log_addr, bsize; u_long i; int ret; size_t offset, retlen; @@ -829,14 +829,14 @@ static int ftl_read(partition_t *part, caddr_t buffer, ======================================================================*/ -static int set_bam_entry(partition_t *part, u_int32_t log_addr, - u_int32_t virt_addr) +static int set_bam_entry(partition_t *part, uint32_t log_addr, + uint32_t virt_addr) { - u_int32_t bsize, blk, le_virt_addr; + uint32_t bsize, blk, le_virt_addr; #ifdef PSYCHO_DEBUG - u_int32_t old_addr; + uint32_t old_addr; #endif - u_int16_t eun; + uint16_t eun; int ret; size_t retlen, offset; @@ -845,11 +845,11 @@ static int set_bam_entry(partition_t *part, u_int32_t log_addr, bsize = 1 << part->header.EraseUnitSize; eun = log_addr / bsize; blk = (log_addr % bsize) / SECTOR_SIZE; - offset = (part->EUNInfo[eun].Offset + blk * sizeof(u_int32_t) + + offset = (part->EUNInfo[eun].Offset + blk * sizeof(uint32_t) + le32_to_cpu(part->header.BAMOffset)); #ifdef PSYCHO_DEBUG - ret = part->mbd.mtd->read(part->mbd.mtd, offset, sizeof(u_int32_t), + ret = part->mbd.mtd->read(part->mbd.mtd, offset, sizeof(uint32_t), &retlen, (u_char *)&old_addr); if (ret) { printk(KERN_WARNING"ftl: Error reading old_addr in set_bam_entry: %d\n",ret); @@ -886,7 +886,7 @@ static int set_bam_entry(partition_t *part, u_int32_t log_addr, #endif part->bam_cache[blk] = le_virt_addr; } - ret = part->mbd.mtd->write(part->mbd.mtd, offset, sizeof(u_int32_t), + ret = part->mbd.mtd->write(part->mbd.mtd, offset, sizeof(uint32_t), &retlen, (u_char *)&le_virt_addr); if (ret) { @@ -900,7 +900,7 @@ static int set_bam_entry(partition_t *part, u_int32_t log_addr, static int ftl_write(partition_t *part, caddr_t buffer, u_long sector, u_long nblocks) { - u_int32_t bsize, log_addr, virt_addr, old_addr, blk; + uint32_t bsize, log_addr, virt_addr, old_addr, blk; u_long i; int ret; size_t retlen, offset; diff --git a/include/linux/mtd/ftl.h b/include/linux/mtd/ftl.h index 0be442f881dd..0555f7a0b9ed 100644 --- a/include/linux/mtd/ftl.h +++ b/include/linux/mtd/ftl.h @@ -32,25 +32,25 @@ #define _LINUX_FTL_H typedef struct erase_unit_header_t { - u_int8_t LinkTargetTuple[5]; - u_int8_t DataOrgTuple[10]; - u_int8_t NumTransferUnits; - u_int32_t EraseCount; - u_int16_t LogicalEUN; - u_int8_t BlockSize; - u_int8_t EraseUnitSize; - u_int16_t FirstPhysicalEUN; - u_int16_t NumEraseUnits; - u_int32_t FormattedSize; - u_int32_t FirstVMAddress; - u_int16_t NumVMPages; - u_int8_t Flags; - u_int8_t Code; - u_int32_t SerialNumber; - u_int32_t AltEUHOffset; - u_int32_t BAMOffset; - u_int8_t Reserved[12]; - u_int8_t EndTuple[2]; + uint8_t LinkTargetTuple[5]; + uint8_t DataOrgTuple[10]; + uint8_t NumTransferUnits; + uint32_t EraseCount; + uint16_t LogicalEUN; + uint8_t BlockSize; + uint8_t EraseUnitSize; + uint16_t FirstPhysicalEUN; + uint16_t NumEraseUnits; + uint32_t FormattedSize; + uint32_t FirstVMAddress; + uint16_t NumVMPages; + uint8_t Flags; + uint8_t Code; + uint32_t SerialNumber; + uint32_t AltEUHOffset; + uint32_t BAMOffset; + uint8_t Reserved[12]; + uint8_t EndTuple[2]; } erase_unit_header_t; /* Flags in erase_unit_header_t */ -- cgit v1.2.3 From 26cdb67c74aedc22367e6d0271f7f955220cca65 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 10 Dec 2008 14:08:12 +0000 Subject: [MTD] Remove more strange u_intxx_t types Signed-off-by: David Woodhouse --- drivers/mtd/mtdconcat.c | 2 +- include/linux/mtd/mtd.h | 26 +++++++++++++------------- include/linux/mtd/partitions.h | 2 +- 3 files changed, 15 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/drivers/mtd/mtdconcat.c b/drivers/mtd/mtdconcat.c index ff8c14bcac68..c26dd528d094 100644 --- a/drivers/mtd/mtdconcat.c +++ b/drivers/mtd/mtdconcat.c @@ -696,7 +696,7 @@ struct mtd_info *mtd_concat_create(struct mtd_info *subdev[], /* subdevices to c int i; size_t size; struct mtd_concat *concat; - u_int32_t max_erasesize, curr_erasesize; + uint32_t max_erasesize, curr_erasesize; int num_erase_region; printk(KERN_NOTICE "Concatenating MTD devices:\n"); diff --git a/include/linux/mtd/mtd.h b/include/linux/mtd/mtd.h index 95e585ecc297..adef674855f3 100644 --- a/include/linux/mtd/mtd.h +++ b/include/linux/mtd/mtd.h @@ -39,8 +39,8 @@ struct erase_info { uint64_t fail_addr; u_long time; u_long retries; - u_int dev; - u_int cell; + unsigned dev; + unsigned cell; void (*callback) (struct erase_info *self); u_long priv; u_char state; @@ -49,8 +49,8 @@ struct erase_info { struct mtd_erase_region_info { uint64_t offset; /* At which this region starts, from the beginning of the MTD */ - u_int32_t erasesize; /* For this region */ - u_int32_t numblocks; /* Number of blocks of erasesize in this region */ + uint32_t erasesize; /* For this region */ + uint32_t numblocks; /* Number of blocks of erasesize in this region */ unsigned long *lockmap; /* If keeping bitmap of locks */ }; @@ -102,14 +102,14 @@ struct mtd_oob_ops { struct mtd_info { u_char type; - u_int32_t flags; + uint32_t flags; uint64_t size; // Total size of the MTD /* "Major" erase size for the device. Naïve users may take this * to be the only erase size available, or may use the more detailed * information below if they desire */ - u_int32_t erasesize; + uint32_t erasesize; /* Minimal writable flash unit size. In case of NOR flash it is 1 (even * though individual bits can be cleared), in case of NAND flash it is * one NAND page (or half, or one-fourths of it), in case of ECC-ed NOR @@ -117,10 +117,10 @@ struct mtd_info { * Any driver registering a struct mtd_info must ensure a writesize of * 1 or larger. */ - u_int32_t writesize; + uint32_t writesize; - u_int32_t oobsize; // Amount of OOB data per block (e.g. 16) - u_int32_t oobavail; // Available OOB bytes per block + uint32_t oobsize; // Amount of OOB data per block (e.g. 16) + uint32_t oobavail; // Available OOB bytes per block /* * If erasesize is a power of 2 then the shift is stored in @@ -233,7 +233,7 @@ struct mtd_info { void (*put_device) (struct mtd_info *mtd); }; -static inline u_int32_t mtd_div_by_eb(uint64_t sz, struct mtd_info *mtd) +static inline uint32_t mtd_div_by_eb(uint64_t sz, struct mtd_info *mtd) { if (mtd->erasesize_shift) return sz >> mtd->erasesize_shift; @@ -241,14 +241,14 @@ static inline u_int32_t mtd_div_by_eb(uint64_t sz, struct mtd_info *mtd) return sz; } -static inline u_int32_t mtd_mod_by_eb(uint64_t sz, struct mtd_info *mtd) +static inline uint32_t mtd_mod_by_eb(uint64_t sz, struct mtd_info *mtd) { if (mtd->erasesize_shift) return sz & mtd->erasesize_mask; return do_div(sz, mtd->erasesize); } -static inline u_int32_t mtd_div_by_ws(uint64_t sz, struct mtd_info *mtd) +static inline uint32_t mtd_div_by_ws(uint64_t sz, struct mtd_info *mtd) { if (mtd->writesize_shift) return sz >> mtd->writesize_shift; @@ -256,7 +256,7 @@ static inline u_int32_t mtd_div_by_ws(uint64_t sz, struct mtd_info *mtd) return sz; } -static inline u_int32_t mtd_mod_by_ws(uint64_t sz, struct mtd_info *mtd) +static inline uint32_t mtd_mod_by_ws(uint64_t sz, struct mtd_info *mtd) { if (mtd->writesize_shift) return sz & mtd->writesize_mask; diff --git a/include/linux/mtd/partitions.h b/include/linux/mtd/partitions.h index 164c7d78687d..a45dd831b3f8 100644 --- a/include/linux/mtd/partitions.h +++ b/include/linux/mtd/partitions.h @@ -38,7 +38,7 @@ struct mtd_partition { char *name; /* identifier string */ uint64_t size; /* partition size */ uint64_t offset; /* offset within the master MTD space */ - u_int32_t mask_flags; /* master MTD flags to mask out for this partition */ + uint32_t mask_flags; /* master MTD flags to mask out for this partition */ struct nand_ecclayout *ecclayout; /* out of band layout for this partition (NAND only)*/ struct mtd_info **mtdp; /* pointer to store the MTD object */ }; -- cgit v1.2.3 From d3af0f048c114dd53713d5920c54f6d5b6b12139 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Mon, 1 Dec 2008 14:23:38 -0800 Subject: [MTD] [NAND] remove excess kernel-doc notation Delete extra kernel-doc notation for struct fields and function parameters that don't exist: Warning(include/linux/mtd/nand.h:428): Excess struct/union/enum/typedef member 'wq' description in 'nand_chip' Warning(include/linux/mtd/nand.h:428): Excess struct/union/enum/typedef member 'datbuf' description in 'nand_chip' Warning(include/linux/mtd/nand.h:428): Excess struct/union/enum/typedef member 'oobbuf' description in 'nand_chip' Warning(include/linux/mtd/nand.h:428): Excess struct/union/enum/typedef member 'oobdirty' description in 'nand_chip' Warning(include/linux/mtd/nand.h:428): Excess struct/union/enum/typedef member 'data_poi' description in 'nand_chip' Warning(drivers/mtd/nand/nand_base.c:2527): Excess function parameter 'maxchips' description in 'nand_scan_tail' Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: David Woodhouse --- drivers/mtd/nand/nand_base.c | 1 - include/linux/mtd/nand.h | 5 ----- 2 files changed, 6 deletions(-) (limited to 'include/linux') diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c index ff2d33e4d6d6..0c3afccde8a2 100644 --- a/drivers/mtd/nand/nand_base.c +++ b/drivers/mtd/nand/nand_base.c @@ -2523,7 +2523,6 @@ int nand_scan_ident(struct mtd_info *mtd, int maxchips) /** * nand_scan_tail - [NAND Interface] Scan for the NAND device * @mtd: MTD device structure - * @maxchips: Number of chips to scan for * * This is the second phase of the normal nand_scan() function. It * fills out all the uninitialized function pointers with the defaults diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h index c0677b8082be..db5b63da2a7e 100644 --- a/include/linux/mtd/nand.h +++ b/include/linux/mtd/nand.h @@ -335,17 +335,12 @@ struct nand_buffers { * @erase_cmd: [INTERN] erase command write function, selectable due to AND support * @scan_bbt: [REPLACEABLE] function to scan bad block table * @chip_delay: [BOARDSPECIFIC] chip dependent delay for transfering data from array to read regs (tR) - * @wq: [INTERN] wait queue to sleep on if a NAND operation is in progress * @state: [INTERN] the current state of the NAND device * @oob_poi: poison value buffer * @page_shift: [INTERN] number of address bits in a page (column address bits) * @phys_erase_shift: [INTERN] number of address bits in a physical eraseblock * @bbt_erase_shift: [INTERN] number of address bits in a bbt entry * @chip_shift: [INTERN] number of address bits in one chip - * @datbuf: [INTERN] internal buffer for one page + oob - * @oobbuf: [INTERN] oob buffer for one eraseblock - * @oobdirty: [INTERN] indicates that oob_buf must be reinitialized - * @data_poi: [INTERN] pointer to a data buffer * @options: [BOARDSPECIFIC] various chip options. They can partly be set to inform nand_scan about * special functionality. See the defines for further explanation * @badblockpos: [INTERN] position of the bad block marker in the oob area -- cgit v1.2.3 From 3f4b0ef7f2899c91b1d6958779f084b44dd59d32 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 26 Oct 2008 20:52:15 +0100 Subject: ACPI hibernate: Add a mechanism to save/restore ACPI NVS memory According to the ACPI Specification 3.0b, Section 15.3.2, "OSPM will call the _PTS control method some time before entering a sleeping state, to allow the platform's AML code to update this memory image before entering the sleeping state. After the system awakes from an S4 state, OSPM will restore this memory area and call the _WAK control method to enable the BIOS to reclaim its memory image." For this reason, implement a mechanism allowing us to save the NVS memory during hibernation and to restore it during the subsequent resume. Based on a patch by Zhang Rui. Signed-off-by: Rafael J. Wysocki Acked-by: Nigel Cunningham Cc: Zhang Rui Signed-off-by: Len Brown --- drivers/acpi/sleep/main.c | 53 +++++++++++++++++--- include/linux/suspend.h | 13 +++++ kernel/power/swsusp.c | 122 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 180 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/drivers/acpi/sleep/main.c b/drivers/acpi/sleep/main.c index 28a691cc625e..45a8015e4217 100644 --- a/drivers/acpi/sleep/main.c +++ b/drivers/acpi/sleep/main.c @@ -394,9 +394,25 @@ void __init acpi_no_s4_hw_signature(void) static int acpi_hibernation_begin(void) { - acpi_target_sleep_state = ACPI_STATE_S4; - acpi_sleep_tts_switch(acpi_target_sleep_state); - return 0; + int error; + + error = hibernate_nvs_alloc(); + if (!error) { + acpi_target_sleep_state = ACPI_STATE_S4; + acpi_sleep_tts_switch(acpi_target_sleep_state); + } + + return error; +} + +static int acpi_hibernation_pre_snapshot(void) +{ + int error = acpi_pm_prepare(); + + if (!error) + hibernate_nvs_save(); + + return error; } static int acpi_hibernation_enter(void) @@ -417,6 +433,12 @@ static int acpi_hibernation_enter(void) return ACPI_SUCCESS(status) ? 0 : -EFAULT; } +static void acpi_hibernation_finish(void) +{ + hibernate_nvs_free(); + acpi_pm_finish(); +} + static void acpi_hibernation_leave(void) { /* @@ -432,6 +454,8 @@ static void acpi_hibernation_leave(void) "cannot resume!\n"); panic("ACPI S4 hardware signature mismatch"); } + /* Restore the NVS memory area */ + hibernate_nvs_restore(); } static void acpi_pm_enable_gpes(void) @@ -442,8 +466,8 @@ static void acpi_pm_enable_gpes(void) static struct platform_hibernation_ops acpi_hibernation_ops = { .begin = acpi_hibernation_begin, .end = acpi_pm_end, - .pre_snapshot = acpi_pm_prepare, - .finish = acpi_pm_finish, + .pre_snapshot = acpi_hibernation_pre_snapshot, + .finish = acpi_hibernation_finish, .prepare = acpi_pm_prepare, .enter = acpi_hibernation_enter, .leave = acpi_hibernation_leave, @@ -469,8 +493,21 @@ static int acpi_hibernation_begin_old(void) error = acpi_sleep_prepare(ACPI_STATE_S4); + if (!error) { + error = hibernate_nvs_alloc(); + if (!error) + acpi_target_sleep_state = ACPI_STATE_S4; + } + return error; +} + +static int acpi_hibernation_pre_snapshot_old(void) +{ + int error = acpi_pm_disable_gpes(); + if (!error) - acpi_target_sleep_state = ACPI_STATE_S4; + hibernate_nvs_save(); + return error; } @@ -481,8 +518,8 @@ static int acpi_hibernation_begin_old(void) static struct platform_hibernation_ops acpi_hibernation_ops_old = { .begin = acpi_hibernation_begin_old, .end = acpi_pm_end, - .pre_snapshot = acpi_pm_disable_gpes, - .finish = acpi_pm_finish, + .pre_snapshot = acpi_hibernation_pre_snapshot_old, + .finish = acpi_hibernation_finish, .prepare = acpi_pm_disable_gpes, .enter = acpi_hibernation_enter, .leave = acpi_hibernation_leave, diff --git a/include/linux/suspend.h b/include/linux/suspend.h index 2ce8207686e2..2b409c44db83 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -232,6 +232,11 @@ extern unsigned long get_safe_page(gfp_t gfp_mask); extern void hibernation_set_ops(struct platform_hibernation_ops *ops); extern int hibernate(void); +extern int hibernate_nvs_register(unsigned long start, unsigned long size); +extern int hibernate_nvs_alloc(void); +extern void hibernate_nvs_free(void); +extern void hibernate_nvs_save(void); +extern void hibernate_nvs_restore(void); #else /* CONFIG_HIBERNATION */ static inline int swsusp_page_is_forbidden(struct page *p) { return 0; } static inline void swsusp_set_page_free(struct page *p) {} @@ -239,6 +244,14 @@ static inline void swsusp_unset_page_free(struct page *p) {} static inline void hibernation_set_ops(struct platform_hibernation_ops *ops) {} static inline int hibernate(void) { return -ENOSYS; } +static inline int hibernate_nvs_register(unsigned long a, unsigned long b) +{ + return 0; +} +static inline int hibernate_nvs_alloc(void) { return 0; } +static inline void hibernate_nvs_free(void) {} +static inline void hibernate_nvs_save(void) {} +static inline void hibernate_nvs_restore(void) {} #endif /* CONFIG_HIBERNATION */ #ifdef CONFIG_PM_SLEEP diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c index 023ff2a31d89..a92c91451559 100644 --- a/kernel/power/swsusp.c +++ b/kernel/power/swsusp.c @@ -262,3 +262,125 @@ int swsusp_shrink_memory(void) return 0; } + +/* + * Platforms, like ACPI, may want us to save some memory used by them during + * hibernation and to restore the contents of this memory during the subsequent + * resume. The code below implements a mechanism allowing us to do that. + */ + +struct nvs_page { + unsigned long phys_start; + unsigned int size; + void *kaddr; + void *data; + struct list_head node; +}; + +static LIST_HEAD(nvs_list); + +/** + * hibernate_nvs_register - register platform NVS memory region to save + * @start - physical address of the region + * @size - size of the region + * + * The NVS region need not be page-aligned (both ends) and we arrange + * things so that the data from page-aligned addresses in this region will + * be copied into separate RAM pages. + */ +int hibernate_nvs_register(unsigned long start, unsigned long size) +{ + struct nvs_page *entry, *next; + + while (size > 0) { + unsigned int nr_bytes; + + entry = kzalloc(sizeof(struct nvs_page), GFP_KERNEL); + if (!entry) + goto Error; + + list_add_tail(&entry->node, &nvs_list); + entry->phys_start = start; + nr_bytes = PAGE_SIZE - (start & ~PAGE_MASK); + entry->size = (size < nr_bytes) ? size : nr_bytes; + + start += entry->size; + size -= entry->size; + } + return 0; + + Error: + list_for_each_entry_safe(entry, next, &nvs_list, node) { + list_del(&entry->node); + kfree(entry); + } + return -ENOMEM; +} + +/** + * hibernate_nvs_free - free data pages allocated for saving NVS regions + */ +void hibernate_nvs_free(void) +{ + struct nvs_page *entry; + + list_for_each_entry(entry, &nvs_list, node) + if (entry->data) { + free_page((unsigned long)entry->data); + entry->data = NULL; + if (entry->kaddr) { + iounmap(entry->kaddr); + entry->kaddr = NULL; + } + } +} + +/** + * hibernate_nvs_alloc - allocate memory necessary for saving NVS regions + */ +int hibernate_nvs_alloc(void) +{ + struct nvs_page *entry; + + list_for_each_entry(entry, &nvs_list, node) { + entry->data = (void *)__get_free_page(GFP_KERNEL); + if (!entry->data) { + hibernate_nvs_free(); + return -ENOMEM; + } + } + return 0; +} + +/** + * hibernate_nvs_save - save NVS memory regions + */ +void hibernate_nvs_save(void) +{ + struct nvs_page *entry; + + printk(KERN_INFO "PM: Saving platform NVS memory\n"); + + list_for_each_entry(entry, &nvs_list, node) + if (entry->data) { + entry->kaddr = ioremap(entry->phys_start, entry->size); + memcpy(entry->data, entry->kaddr, entry->size); + } +} + +/** + * hibernate_nvs_restore - restore NVS memory regions + * + * This function is going to be called with interrupts disabled, so it + * cannot iounmap the virtual addresses used to access the NVS region. + */ +void hibernate_nvs_restore(void) +{ + struct nvs_page *entry; + + printk(KERN_INFO "PM: Restoring platform NVS memory\n"); + + list_for_each_entry(entry, &nvs_list, node) + if (entry->data) + memcpy(entry->kaddr, entry->data, entry->size); +} -- cgit v1.2.3 From ba84ed9546e91348fdf3ff2bff859b0ee53b407a Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 26 Oct 2008 20:56:30 +0100 Subject: ACPI hibernate: Introduce new kernel parameter acpi_sleep=s4_nonvs On some machines it may be necessary to disable the saving/restoring of the ACPI NVS memory region during hibernation/resume. For this purpose, introduce new ACPI kernel command line option acpi_sleep=s4_nonvs. Based on a patch by Zhang Rui. Signed-off-by: Rafael J. Wysocki Acked-by: Nigel Cunningham Acked-by: Pavel Machek Signed-off-by: Len Brown --- Documentation/kernel-parameters.txt | 5 ++++- arch/x86/kernel/acpi/sleep.c | 2 ++ drivers/acpi/sleep/main.c | 18 ++++++++++++++++-- include/linux/acpi.h | 1 + 4 files changed, 23 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index e0f346d201ed..1d089eeff3cf 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -149,7 +149,8 @@ and is between 256 and 4096 characters. It is defined in the file default: 0 acpi_sleep= [HW,ACPI] Sleep options - Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig, old_ordering } + Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig, + old_ordering, s4_nonvs } See Documentation/power/video.txt for s3_bios and s3_mode. s3_beep is for debugging; it makes the PC's speaker beep as soon as the kernel's real-mode entry point is called. @@ -159,6 +160,8 @@ and is between 256 and 4096 characters. It is defined in the file control method, wrt putting devices into low power states, to be enforced (the ACPI 2.0 ordering of _PTS is used by default). + s4_nonvs prevents the kernel from saving/restoring the + ACPI NVS memory during hibernation. acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode Format: { level | edge | high | low } diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index 806b4e9051b4..707c1f6f95fa 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -159,6 +159,8 @@ static int __init acpi_sleep_setup(char *str) #endif if (strncmp(str, "old_ordering", 12) == 0) acpi_old_suspend_ordering(); + if (strncmp(str, "s4_nonvs", 8) == 0) + acpi_s4_no_nvs(); str = strchr(str, ','); if (str != NULL) str += strspn(str, ", \t"); diff --git a/drivers/acpi/sleep/main.c b/drivers/acpi/sleep/main.c index 45a8015e4217..bef41fd4c877 100644 --- a/drivers/acpi/sleep/main.c +++ b/drivers/acpi/sleep/main.c @@ -101,6 +101,19 @@ void __init acpi_old_suspend_ordering(void) * cases. */ static bool set_sci_en_on_resume; +/* + * The ACPI specification wants us to save NVS memory regions during hibernation + * and to restore them during the subsequent resume. However, it is not certain + * if this mechanism is going to work on all machines, so we allow the user to + * disable this mechanism using the 'acpi_sleep=s4_nonvs' kernel command line + * option. + */ +static bool s4_no_nvs; + +void __init acpi_s4_no_nvs(void) +{ + s4_no_nvs = true; +} /** * acpi_pm_disable_gpes - Disable the GPEs. @@ -396,7 +409,7 @@ static int acpi_hibernation_begin(void) { int error; - error = hibernate_nvs_alloc(); + error = s4_no_nvs ? 0 : hibernate_nvs_alloc(); if (!error) { acpi_target_sleep_state = ACPI_STATE_S4; acpi_sleep_tts_switch(acpi_target_sleep_state); @@ -494,7 +507,8 @@ static int acpi_hibernation_begin_old(void) error = acpi_sleep_prepare(ACPI_STATE_S4); if (!error) { - error = hibernate_nvs_alloc(); + if (!s4_no_nvs) + error = hibernate_nvs_alloc(); if (!error) acpi_target_sleep_state = ACPI_STATE_S4; } diff --git a/include/linux/acpi.h b/include/linux/acpi.h index fba8051fb297..dfa0a5356c53 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -270,6 +270,7 @@ int acpi_check_mem_region(resource_size_t start, resource_size_t n, #ifdef CONFIG_PM_SLEEP void __init acpi_no_s4_hw_signature(void); void __init acpi_old_suspend_ordering(void); +void __init acpi_s4_no_nvs(void); #endif /* CONFIG_PM_SLEEP */ #else /* CONFIG_ACPI */ -- cgit v1.2.3 From 50b6f1f4a430608f7345f66ecd68a129bff11649 Mon Sep 17 00:00:00 2001 From: Kwangwoo Lee Date: Sat, 20 Dec 2008 04:26:01 -0500 Subject: Input: add tsc2007 based touchscreen driver This drive has been tested on ARM9 based SoC - MV86XX. Signed-off-by: Kwangwoo Lee Acked-by: Jean Delvare Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/Kconfig | 11 ++ drivers/input/touchscreen/Makefile | 1 + drivers/input/touchscreen/tsc2007.c | 381 ++++++++++++++++++++++++++++++++++++ include/linux/i2c/tsc2007.h | 17 ++ 4 files changed, 410 insertions(+) create mode 100644 drivers/input/touchscreen/tsc2007.c create mode 100644 include/linux/i2c/tsc2007.h (limited to 'include/linux') diff --git a/drivers/input/touchscreen/Kconfig b/drivers/input/touchscreen/Kconfig index 20eb52ed176d..83747dc9ba05 100644 --- a/drivers/input/touchscreen/Kconfig +++ b/drivers/input/touchscreen/Kconfig @@ -389,4 +389,15 @@ config TOUCHSCREEN_TOUCHIT213 To compile this driver as a module, choose M here: the module will be called touchit213. +config TOUCHSCREEN_TSC2007 + tristate "TSC2007 based touchscreens" + depends on I2C + help + Say Y here if you have a TSC2007 based touchscreen. + + If unsure, say N. + + To compile this driver as a module, choose M here: the + module will be called tsc2007. + endif diff --git a/drivers/input/touchscreen/Makefile b/drivers/input/touchscreen/Makefile index 3dc84d3846c1..127f87cc2e2e 100644 --- a/drivers/input/touchscreen/Makefile +++ b/drivers/input/touchscreen/Makefile @@ -25,6 +25,7 @@ obj-$(CONFIG_TOUCHSCREEN_PENMOUNT) += penmount.o obj-$(CONFIG_TOUCHSCREEN_TOUCHIT213) += touchit213.o obj-$(CONFIG_TOUCHSCREEN_TOUCHRIGHT) += touchright.o obj-$(CONFIG_TOUCHSCREEN_TOUCHWIN) += touchwin.o +obj-$(CONFIG_TOUCHSCREEN_TSC2007) += tsc2007.o obj-$(CONFIG_TOUCHSCREEN_UCB1400) += ucb1400_ts.o obj-$(CONFIG_TOUCHSCREEN_WACOM_W8001) += wacom_w8001.o obj-$(CONFIG_TOUCHSCREEN_WM97XX) += wm97xx-ts.o diff --git a/drivers/input/touchscreen/tsc2007.c b/drivers/input/touchscreen/tsc2007.c new file mode 100644 index 000000000000..b75dc2990574 --- /dev/null +++ b/drivers/input/touchscreen/tsc2007.c @@ -0,0 +1,381 @@ +/* + * drivers/input/touchscreen/tsc2007.c + * + * Copyright (c) 2008 MtekVision Co., Ltd. + * Kwangwoo Lee + * + * Using code from: + * - ads7846.c + * Copyright (c) 2005 David Brownell + * Copyright (c) 2006 Nokia Corporation + * - corgi_ts.c + * Copyright (C) 2004-2005 Richard Purdie + * - omap_ts.[hc], ads7846.h, ts_osk.c + * Copyright (C) 2002 MontaVista Software + * Copyright (C) 2004 Texas Instruments + * Copyright (C) 2005 Dirk Behme + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include +#include + +#define TS_POLL_DELAY (10 * 1000) /* ns delay before the first sample */ +#define TS_POLL_PERIOD (5 * 1000) /* ns delay between samples */ + +#define TSC2007_MEASURE_TEMP0 (0x0 << 4) +#define TSC2007_MEASURE_AUX (0x2 << 4) +#define TSC2007_MEASURE_TEMP1 (0x4 << 4) +#define TSC2007_ACTIVATE_XN (0x8 << 4) +#define TSC2007_ACTIVATE_YN (0x9 << 4) +#define TSC2007_ACTIVATE_YP_XN (0xa << 4) +#define TSC2007_SETUP (0xb << 4) +#define TSC2007_MEASURE_X (0xc << 4) +#define TSC2007_MEASURE_Y (0xd << 4) +#define TSC2007_MEASURE_Z1 (0xe << 4) +#define TSC2007_MEASURE_Z2 (0xf << 4) + +#define TSC2007_POWER_OFF_IRQ_EN (0x0 << 2) +#define TSC2007_ADC_ON_IRQ_DIS0 (0x1 << 2) +#define TSC2007_ADC_OFF_IRQ_EN (0x2 << 2) +#define TSC2007_ADC_ON_IRQ_DIS1 (0x3 << 2) + +#define TSC2007_12BIT (0x0 << 1) +#define TSC2007_8BIT (0x1 << 1) + +#define MAX_12BIT ((1 << 12) - 1) + +#define ADC_ON_12BIT (TSC2007_12BIT | TSC2007_ADC_ON_IRQ_DIS0) + +#define READ_Y (ADC_ON_12BIT | TSC2007_MEASURE_Y) +#define READ_Z1 (ADC_ON_12BIT | TSC2007_MEASURE_Z1) +#define READ_Z2 (ADC_ON_12BIT | TSC2007_MEASURE_Z2) +#define READ_X (ADC_ON_12BIT | TSC2007_MEASURE_X) +#define PWRDOWN (TSC2007_12BIT | TSC2007_POWER_OFF_IRQ_EN) + +struct ts_event { + u16 x; + u16 y; + u16 z1, z2; +}; + +struct tsc2007 { + struct input_dev *input; + char phys[32]; + struct hrtimer timer; + struct ts_event tc; + + struct i2c_client *client; + + spinlock_t lock; + + u16 model; + u16 x_plate_ohms; + + unsigned pendown; + int irq; + + int (*get_pendown_state)(void); + void (*clear_penirq)(void); +}; + +static inline int tsc2007_xfer(struct tsc2007 *tsc, u8 cmd) +{ + s32 data; + u16 val; + + data = i2c_smbus_read_word_data(tsc->client, cmd); + if (data < 0) { + dev_err(&tsc->client->dev, "i2c io error: %d\n", data); + return data; + } + + /* The protocol and raw data format from i2c interface: + * S Addr Wr [A] Comm [A] S Addr Rd [A] [DataLow] A [DataHigh] NA P + * Where DataLow has [D11-D4], DataHigh has [D3-D0 << 4 | Dummy 4bit]. + */ + val = swab16(data) >> 4; + + dev_dbg(&tsc->client->dev, "data: 0x%x, val: 0x%x\n", data, val); + + return val; +} + +static void tsc2007_send_event(void *tsc) +{ + struct tsc2007 *ts = tsc; + u32 rt; + u16 x, y, z1, z2; + + x = ts->tc.x; + y = ts->tc.y; + z1 = ts->tc.z1; + z2 = ts->tc.z2; + + /* range filtering */ + if (x == MAX_12BIT) + x = 0; + + if (likely(x && z1)) { + /* compute touch pressure resistance using equation #1 */ + rt = z2; + rt -= z1; + rt *= x; + rt *= ts->x_plate_ohms; + rt /= z1; + rt = (rt + 2047) >> 12; + } else + rt = 0; + + /* Sample found inconsistent by debouncing or pressure is beyond + * the maximum. Don't report it to user space, repeat at least + * once more the measurement + */ + if (rt > MAX_12BIT) { + dev_dbg(&ts->client->dev, "ignored pressure %d\n", rt); + + hrtimer_start(&ts->timer, ktime_set(0, TS_POLL_PERIOD), + HRTIMER_MODE_REL); + return; + } + + /* NOTE: We can't rely on the pressure to determine the pen down + * state, even this controller has a pressure sensor. The pressure + * value can fluctuate for quite a while after lifting the pen and + * in some cases may not even settle at the expected value. + * + * The only safe way to check for the pen up condition is in the + * timer by reading the pen signal state (it's a GPIO _and_ IRQ). + */ + if (rt) { + struct input_dev *input = ts->input; + + if (!ts->pendown) { + dev_dbg(&ts->client->dev, "DOWN\n"); + + input_report_key(input, BTN_TOUCH, 1); + ts->pendown = 1; + } + + input_report_abs(input, ABS_X, x); + input_report_abs(input, ABS_Y, y); + input_report_abs(input, ABS_PRESSURE, rt); + + input_sync(input); + + dev_dbg(&ts->client->dev, "point(%4d,%4d), pressure (%4u)\n", + x, y, rt); + } + + hrtimer_start(&ts->timer, ktime_set(0, TS_POLL_PERIOD), + HRTIMER_MODE_REL); +} + +static int tsc2007_read_values(struct tsc2007 *tsc) +{ + /* y- still on; turn on only y+ (and ADC) */ + tsc->tc.y = tsc2007_xfer(tsc, READ_Y); + + /* turn y- off, x+ on, then leave in lowpower */ + tsc->tc.x = tsc2007_xfer(tsc, READ_X); + + /* turn y+ off, x- on; we'll use formula #1 */ + tsc->tc.z1 = tsc2007_xfer(tsc, READ_Z1); + tsc->tc.z2 = tsc2007_xfer(tsc, READ_Z2); + + /* power down */ + tsc2007_xfer(tsc, PWRDOWN); + + return 0; +} + +static enum hrtimer_restart tsc2007_timer(struct hrtimer *handle) +{ + struct tsc2007 *ts = container_of(handle, struct tsc2007, timer); + + spin_lock_irq(&ts->lock); + + if (unlikely(!ts->get_pendown_state() && ts->pendown)) { + struct input_dev *input = ts->input; + + dev_dbg(&ts->client->dev, "UP\n"); + + input_report_key(input, BTN_TOUCH, 0); + input_report_abs(input, ABS_PRESSURE, 0); + input_sync(input); + + ts->pendown = 0; + enable_irq(ts->irq); + } else { + /* pen is still down, continue with the measurement */ + dev_dbg(&ts->client->dev, "pen is still down\n"); + + tsc2007_read_values(ts); + tsc2007_send_event(ts); + } + + spin_unlock_irq(&ts->lock); + + return HRTIMER_NORESTART; +} + +static irqreturn_t tsc2007_irq(int irq, void *handle) +{ + struct tsc2007 *ts = handle; + unsigned long flags; + + spin_lock_irqsave(&ts->lock, flags); + + if (likely(ts->get_pendown_state())) { + disable_irq(ts->irq); + hrtimer_start(&ts->timer, ktime_set(0, TS_POLL_DELAY), + HRTIMER_MODE_REL); + } + + if (ts->clear_penirq) + ts->clear_penirq(); + + spin_unlock_irqrestore(&ts->lock, flags); + + return IRQ_HANDLED; +} + +static int tsc2007_probe(struct i2c_client *client, + const struct i2c_device_id *id) +{ + struct tsc2007 *ts; + struct tsc2007_platform_data *pdata = pdata = client->dev.platform_data; + struct input_dev *input_dev; + int err; + + if (!pdata) { + dev_err(&client->dev, "platform data is required!\n"); + return -EINVAL; + } + + if (!i2c_check_functionality(client->adapter, + I2C_FUNC_SMBUS_READ_WORD_DATA)) + return -EIO; + + ts = kzalloc(sizeof(struct tsc2007), GFP_KERNEL); + input_dev = input_allocate_device(); + if (!ts || !input_dev) { + err = -ENOMEM; + goto err_free_mem; + } + + ts->client = client; + i2c_set_clientdata(client, ts); + + ts->input = input_dev; + + hrtimer_init(&ts->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + ts->timer.function = tsc2007_timer; + + spin_lock_init(&ts->lock); + + ts->model = pdata->model; + ts->x_plate_ohms = pdata->x_plate_ohms; + ts->get_pendown_state = pdata->get_pendown_state; + ts->clear_penirq = pdata->clear_penirq; + + pdata->init_platform_hw(); + + snprintf(ts->phys, sizeof(ts->phys), "%s/input0", client->dev.bus_id); + + input_dev->name = "TSC2007 Touchscreen"; + input_dev->phys = ts->phys; + input_dev->id.bustype = BUS_I2C; + + input_dev->evbit[0] = BIT_MASK(EV_KEY) | BIT_MASK(EV_ABS); + input_dev->keybit[BIT_WORD(BTN_TOUCH)] = BIT_MASK(BTN_TOUCH); + + input_set_abs_params(input_dev, ABS_X, 0, MAX_12BIT, 0, 0); + input_set_abs_params(input_dev, ABS_Y, 0, MAX_12BIT, 0, 0); + input_set_abs_params(input_dev, ABS_PRESSURE, 0, MAX_12BIT, 0, 0); + + tsc2007_read_values(ts); + + ts->irq = client->irq; + + err = request_irq(ts->irq, tsc2007_irq, 0, + client->dev.driver->name, ts); + if (err < 0) { + dev_err(&client->dev, "irq %d busy?\n", ts->irq); + goto err_free_mem; + } + + err = input_register_device(input_dev); + if (err) + goto err_free_irq; + + dev_info(&client->dev, "registered with irq (%d)\n", ts->irq); + + return 0; + + err_free_irq: + free_irq(ts->irq, ts); + hrtimer_cancel(&ts->timer); + err_free_mem: + input_free_device(input_dev); + kfree(ts); + return err; +} + +static int tsc2007_remove(struct i2c_client *client) +{ + struct tsc2007 *ts = i2c_get_clientdata(client); + struct tsc2007_platform_data *pdata; + + pdata = client->dev.platform_data; + pdata->exit_platform_hw(); + + free_irq(ts->irq, ts); + hrtimer_cancel(&ts->timer); + input_unregister_device(ts->input); + kfree(ts); + + return 0; +} + +static struct i2c_device_id tsc2007_idtable[] = { + { "tsc2007", 0 }, + { } +}; + +MODULE_DEVICE_TABLE(i2c, tsc2007_idtable); + +static struct i2c_driver tsc2007_driver = { + .driver = { + .owner = THIS_MODULE, + .name = "tsc2007" + }, + .id_table = tsc2007_idtable, + .probe = tsc2007_probe, + .remove = tsc2007_remove, +}; + +static int __init tsc2007_init(void) +{ + return i2c_add_driver(&tsc2007_driver); +} + +static void __exit tsc2007_exit(void) +{ + i2c_del_driver(&tsc2007_driver); +} + +module_init(tsc2007_init); +module_exit(tsc2007_exit); + +MODULE_AUTHOR("Kwangwoo Lee "); +MODULE_DESCRIPTION("TSC2007 TouchScreen Driver"); +MODULE_LICENSE("GPL"); diff --git a/include/linux/i2c/tsc2007.h b/include/linux/i2c/tsc2007.h new file mode 100644 index 000000000000..c6361fbb7bf9 --- /dev/null +++ b/include/linux/i2c/tsc2007.h @@ -0,0 +1,17 @@ +#ifndef __LINUX_I2C_TSC2007_H +#define __LINUX_I2C_TSC2007_H + +/* linux/i2c/tsc2007.h */ + +struct tsc2007_platform_data { + u16 model; /* 2007. */ + u16 x_plate_ohms; + + int (*get_pendown_state)(void); + void (*clear_penirq)(void); /* If needed, clear 2nd level + interrupt source */ + int (*init_platform_hw)(void); + void (*exit_platform_hw)(void); +}; + +#endif -- cgit v1.2.3 From 160bbab3000dafccbe43688e48208cecf4deb879 Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Tue, 23 Dec 2008 10:00:14 +0000 Subject: [MTD] struct device - replace bus_id with dev_name(), dev_set_name() Signed-off-by: Kay Sievers Signed-off-by: Greg Kroah-Hartman Signed-off-by: David Woodhouse --- drivers/mtd/devices/m25p80.c | 20 ++++++++++---------- drivers/mtd/devices/mtd_dataflash.c | 32 ++++++++++++++++---------------- drivers/mtd/maps/integrator-flash.c | 2 +- drivers/mtd/maps/ixp2000.c | 4 ++-- drivers/mtd/maps/ixp4xx.c | 2 +- drivers/mtd/maps/omap_nor.c | 2 +- drivers/mtd/maps/physmap.c | 6 +++--- drivers/mtd/maps/physmap_of.c | 4 ++-- drivers/mtd/mtdconcat.c | 2 +- drivers/mtd/nand/fsl_upm.c | 2 +- drivers/mtd/nand/plat_nand.c | 2 +- drivers/mtd/nand/tmio_nand.c | 2 +- drivers/mtd/onenand/generic.c | 2 +- drivers/mtd/onenand/omap2.c | 2 +- drivers/mtd/ubi/build.c | 2 +- drivers/mtd/ubi/vmt.c | 4 ++-- include/linux/mtd/concat.h | 2 +- 17 files changed, 46 insertions(+), 46 deletions(-) (limited to 'include/linux') diff --git a/drivers/mtd/devices/m25p80.c b/drivers/mtd/devices/m25p80.c index 9be0229c3d30..7c3fc766dcf1 100644 --- a/drivers/mtd/devices/m25p80.c +++ b/drivers/mtd/devices/m25p80.c @@ -171,8 +171,8 @@ static int wait_till_ready(struct m25p *flash) static int erase_chip(struct m25p *flash) { DEBUG(MTD_DEBUG_LEVEL3, "%s: %s %lldKiB\n", - flash->spi->dev.bus_id, __func__, - (long long)(flash->mtd.size >> 10)); + dev_name(&flash->spi->dev), __func__, + (long long)(flash->mtd.size >> 10)); /* Wait until finished previous write command. */ if (wait_till_ready(flash)) @@ -198,7 +198,7 @@ static int erase_chip(struct m25p *flash) static int erase_sector(struct m25p *flash, u32 offset) { DEBUG(MTD_DEBUG_LEVEL3, "%s: %s %dKiB at 0x%08x\n", - flash->spi->dev.bus_id, __func__, + dev_name(&flash->spi->dev), __func__, flash->mtd.erasesize / 1024, offset); /* Wait until finished previous write command. */ @@ -236,8 +236,8 @@ static int m25p80_erase(struct mtd_info *mtd, struct erase_info *instr) uint32_t rem; DEBUG(MTD_DEBUG_LEVEL2, "%s: %s %s 0x%llx, len %lld\n", - flash->spi->dev.bus_id, __func__, "at", - (long long)instr->addr, (long long)instr->len); + dev_name(&flash->spi->dev), __func__, "at", + (long long)instr->addr, (long long)instr->len); /* sanity checks */ if (instr->addr + instr->len > flash->mtd.size) @@ -296,7 +296,7 @@ static int m25p80_read(struct mtd_info *mtd, loff_t from, size_t len, struct spi_message m; DEBUG(MTD_DEBUG_LEVEL2, "%s: %s %s 0x%08x, len %zd\n", - flash->spi->dev.bus_id, __func__, "from", + dev_name(&flash->spi->dev), __func__, "from", (u32)from, len); /* sanity checks */ @@ -368,7 +368,7 @@ static int m25p80_write(struct mtd_info *mtd, loff_t to, size_t len, struct spi_message m; DEBUG(MTD_DEBUG_LEVEL2, "%s: %s %s 0x%08x, len %zd\n", - flash->spi->dev.bus_id, __func__, "to", + dev_name(&flash->spi->dev), __func__, "to", (u32)to, len); if (retlen) @@ -564,7 +564,7 @@ static struct flash_info *__devinit jedec_probe(struct spi_device *spi) tmp = spi_write_then_read(spi, &code, 1, id, 5); if (tmp < 0) { DEBUG(MTD_DEBUG_LEVEL0, "%s: error %d reading JEDEC ID\n", - spi->dev.bus_id, tmp); + dev_name(&spi->dev), tmp); return NULL; } jedec = id[0]; @@ -618,7 +618,7 @@ static int __devinit m25p_probe(struct spi_device *spi) /* unrecognized chip? */ if (i == ARRAY_SIZE(m25p_data)) { DEBUG(MTD_DEBUG_LEVEL0, "%s: unrecognized id %s\n", - spi->dev.bus_id, data->type); + dev_name(&spi->dev), data->type); info = NULL; /* recognized; is that chip really what's there? */ @@ -659,7 +659,7 @@ static int __devinit m25p_probe(struct spi_device *spi) if (data && data->name) flash->mtd.name = data->name; else - flash->mtd.name = spi->dev.bus_id; + flash->mtd.name = dev_name(&spi->dev); flash->mtd.type = MTD_NORFLASH; flash->mtd.writesize = 1; diff --git a/drivers/mtd/devices/mtd_dataflash.c b/drivers/mtd/devices/mtd_dataflash.c index 68068975940b..d44f741ae229 100644 --- a/drivers/mtd/devices/mtd_dataflash.c +++ b/drivers/mtd/devices/mtd_dataflash.c @@ -129,7 +129,7 @@ static int dataflash_waitready(struct spi_device *spi) status = dataflash_status(spi); if (status < 0) { DEBUG(MTD_DEBUG_LEVEL1, "%s: status %d?\n", - spi->dev.bus_id, status); + dev_name(&spi->dev), status); status = 0; } @@ -156,8 +156,8 @@ static int dataflash_erase(struct mtd_info *mtd, struct erase_info *instr) uint32_t rem; DEBUG(MTD_DEBUG_LEVEL2, "%s: erase addr=0x%llx len 0x%llx\n", - spi->dev.bus_id, (long long)instr->addr, - (long long)instr->len); + dev_name(&spi->dev), (long long)instr->addr, + (long long)instr->len); /* Sanity checks */ if (instr->addr + instr->len > mtd->size) @@ -203,7 +203,7 @@ static int dataflash_erase(struct mtd_info *mtd, struct erase_info *instr) if (status < 0) { printk(KERN_ERR "%s: erase %x, err %d\n", - spi->dev.bus_id, pageaddr, status); + dev_name(&spi->dev), pageaddr, status); /* REVISIT: can retry instr->retries times; or * giveup and instr->fail_addr = instr->addr; */ @@ -245,7 +245,7 @@ static int dataflash_read(struct mtd_info *mtd, loff_t from, size_t len, int status; DEBUG(MTD_DEBUG_LEVEL2, "%s: read 0x%x..0x%x\n", - priv->spi->dev.bus_id, (unsigned)from, (unsigned)(from + len)); + dev_name(&priv->spi->dev), (unsigned)from, (unsigned)(from + len)); *retlen = 0; @@ -294,7 +294,7 @@ static int dataflash_read(struct mtd_info *mtd, loff_t from, size_t len, status = 0; } else DEBUG(MTD_DEBUG_LEVEL1, "%s: read %x..%x --> %d\n", - priv->spi->dev.bus_id, + dev_name(&priv->spi->dev), (unsigned)from, (unsigned)(from + len), status); return status; @@ -321,7 +321,7 @@ static int dataflash_write(struct mtd_info *mtd, loff_t to, size_t len, uint8_t *command; DEBUG(MTD_DEBUG_LEVEL2, "%s: write 0x%x..0x%x\n", - spi->dev.bus_id, (unsigned)to, (unsigned)(to + len)); + dev_name(&spi->dev), (unsigned)to, (unsigned)(to + len)); *retlen = 0; @@ -380,7 +380,7 @@ static int dataflash_write(struct mtd_info *mtd, loff_t to, size_t len, status = spi_sync(spi, &msg); if (status < 0) DEBUG(MTD_DEBUG_LEVEL1, "%s: xfer %u -> %d \n", - spi->dev.bus_id, addr, status); + dev_name(&spi->dev), addr, status); (void) dataflash_waitready(priv->spi); } @@ -402,7 +402,7 @@ static int dataflash_write(struct mtd_info *mtd, loff_t to, size_t len, spi_transfer_del(x + 1); if (status < 0) DEBUG(MTD_DEBUG_LEVEL1, "%s: pgm %u/%u -> %d \n", - spi->dev.bus_id, addr, writelen, status); + dev_name(&spi->dev), addr, writelen, status); (void) dataflash_waitready(priv->spi); @@ -422,14 +422,14 @@ static int dataflash_write(struct mtd_info *mtd, loff_t to, size_t len, status = spi_sync(spi, &msg); if (status < 0) DEBUG(MTD_DEBUG_LEVEL1, "%s: compare %u -> %d \n", - spi->dev.bus_id, addr, status); + dev_name(&spi->dev), addr, status); status = dataflash_waitready(priv->spi); /* Check result of the compare operation */ if (status & (1 << 6)) { printk(KERN_ERR "%s: compare page %u, err %d\n", - spi->dev.bus_id, pageaddr, status); + dev_name(&spi->dev), pageaddr, status); remaining = 0; status = -EIO; break; @@ -785,7 +785,7 @@ static struct flash_info *__devinit jedec_probe(struct spi_device *spi) tmp = spi_write_then_read(spi, &code, 1, id, 3); if (tmp < 0) { DEBUG(MTD_DEBUG_LEVEL0, "%s: error %d reading JEDEC ID\n", - spi->dev.bus_id, tmp); + dev_name(&spi->dev), tmp); return ERR_PTR(tmp); } if (id[0] != 0x1f) @@ -875,7 +875,7 @@ static int __devinit dataflash_probe(struct spi_device *spi) status = dataflash_status(spi); if (status <= 0 || status == 0xff) { DEBUG(MTD_DEBUG_LEVEL1, "%s: status error %d\n", - spi->dev.bus_id, status); + dev_name(&spi->dev), status); if (status == 0 || status == 0xff) status = -ENODEV; return status; @@ -911,13 +911,13 @@ static int __devinit dataflash_probe(struct spi_device *spi) /* obsolete AT45DB1282 not (yet?) supported */ default: DEBUG(MTD_DEBUG_LEVEL1, "%s: unsupported device (%x)\n", - spi->dev.bus_id, status & 0x3c); + dev_name(&spi->dev), status & 0x3c); status = -ENODEV; } if (status < 0) DEBUG(MTD_DEBUG_LEVEL1, "%s: add_dataflash --> %d\n", - spi->dev.bus_id, status); + dev_name(&spi->dev), status); return status; } @@ -927,7 +927,7 @@ static int __devexit dataflash_remove(struct spi_device *spi) struct dataflash *flash = dev_get_drvdata(&spi->dev); int status; - DEBUG(MTD_DEBUG_LEVEL1, "%s: remove\n", spi->dev.bus_id); + DEBUG(MTD_DEBUG_LEVEL1, "%s: remove\n", dev_name(&spi->dev)); if (mtd_has_partitions() && flash->partitioned) status = del_mtd_partitions(&flash->mtd); diff --git a/drivers/mtd/maps/integrator-flash.c b/drivers/mtd/maps/integrator-flash.c index 7100ee3c7b01..d2ec262666c7 100644 --- a/drivers/mtd/maps/integrator-flash.c +++ b/drivers/mtd/maps/integrator-flash.c @@ -105,7 +105,7 @@ static int armflash_probe(struct platform_device *dev) info->map.bankwidth = plat->width; info->map.phys = res->start; info->map.virt = base; - info->map.name = dev->dev.bus_id; + info->map.name = dev_name(&dev->dev); info->map.set_vpp = armflash_set_vpp; simple_map_init(&info->map); diff --git a/drivers/mtd/maps/ixp2000.c b/drivers/mtd/maps/ixp2000.c index dcdb1f17577d..d76880d91bdb 100644 --- a/drivers/mtd/maps/ixp2000.c +++ b/drivers/mtd/maps/ixp2000.c @@ -188,7 +188,7 @@ static int ixp2000_flash_probe(struct platform_device *dev) */ info->map.map_priv_2 = (unsigned long) ixp_data->bank_setup; - info->map.name = dev->dev.bus_id; + info->map.name = dev_name(&dev->dev); info->map.read = ixp2000_flash_read8; info->map.write = ixp2000_flash_write8; info->map.copy_from = ixp2000_flash_copy_from; @@ -196,7 +196,7 @@ static int ixp2000_flash_probe(struct platform_device *dev) info->res = request_mem_region(dev->resource->start, dev->resource->end - dev->resource->start + 1, - dev->dev.bus_id); + dev_name(&dev->dev)); if (!info->res) { dev_err(&dev->dev, "Could not reserve memory region\n"); err = -ENOMEM; diff --git a/drivers/mtd/maps/ixp4xx.c b/drivers/mtd/maps/ixp4xx.c index 9c7a5fbd4e51..4d0be2f1503f 100644 --- a/drivers/mtd/maps/ixp4xx.c +++ b/drivers/mtd/maps/ixp4xx.c @@ -218,7 +218,7 @@ static int ixp4xx_flash_probe(struct platform_device *dev) * handle that. */ info->map.bankwidth = 2; - info->map.name = dev->dev.bus_id; + info->map.name = dev_name(&dev->dev); info->map.read = ixp4xx_read16, info->map.write = ixp4xx_probe_write16, info->map.copy_from = ixp4xx_copy_from, diff --git a/drivers/mtd/maps/omap_nor.c b/drivers/mtd/maps/omap_nor.c index 05f276af15da..7e50e9b1b781 100644 --- a/drivers/mtd/maps/omap_nor.c +++ b/drivers/mtd/maps/omap_nor.c @@ -101,7 +101,7 @@ static int __init omapflash_probe(struct platform_device *pdev) err = -ENOMEM; goto out_release_mem_region; } - info->map.name = pdev->dev.bus_id; + info->map.name = dev_name(&pdev->dev); info->map.phys = res->start; info->map.size = size; info->map.bankwidth = pdata->width; diff --git a/drivers/mtd/maps/physmap.c b/drivers/mtd/maps/physmap.c index 58207b3b9411..d3a2acc7e9be 100644 --- a/drivers/mtd/maps/physmap.c +++ b/drivers/mtd/maps/physmap.c @@ -106,13 +106,13 @@ static int physmap_flash_probe(struct platform_device *dev) if (!devm_request_mem_region(&dev->dev, dev->resource[i].start, dev->resource[i].end - dev->resource[i].start + 1, - dev->dev.bus_id)) { + dev_name(&dev->dev))) { dev_err(&dev->dev, "Could not reserve memory region\n"); err = -ENOMEM; goto err_out; } - info->map[i].name = dev->dev.bus_id; + info->map[i].name = dev_name(&dev->dev); info->map[i].phys = dev->resource[i].start; info->map[i].size = dev->resource[i].end - dev->resource[i].start + 1; info->map[i].bankwidth = physmap_data->width; @@ -148,7 +148,7 @@ static int physmap_flash_probe(struct platform_device *dev) * We detected multiple devices. Concatenate them together. */ #ifdef CONFIG_MTD_CONCAT - info->cmtd = mtd_concat_create(info->mtd, devices_found, dev->dev.bus_id); + info->cmtd = mtd_concat_create(info->mtd, devices_found, dev_name(&dev->dev)); if (info->cmtd == NULL) err = -ENXIO; #else diff --git a/drivers/mtd/maps/physmap_of.c b/drivers/mtd/maps/physmap_of.c index 5fcfec034a94..fbf0ca939d72 100644 --- a/drivers/mtd/maps/physmap_of.c +++ b/drivers/mtd/maps/physmap_of.c @@ -183,7 +183,7 @@ static int __devinit of_flash_probe(struct of_device *dev, err = -EBUSY; info->res = request_mem_region(res.start, res.end - res.start + 1, - dev->dev.bus_id); + dev_name(&dev->dev)); if (!info->res) goto err_out; @@ -194,7 +194,7 @@ static int __devinit of_flash_probe(struct of_device *dev, goto err_out; } - info->map.name = dev->dev.bus_id; + info->map.name = dev_name(&dev->dev); info->map.phys = res.start; info->map.size = res.end - res.start + 1; info->map.bankwidth = *width; diff --git a/drivers/mtd/mtdconcat.c b/drivers/mtd/mtdconcat.c index c26dd528d094..3dbb1b38db66 100644 --- a/drivers/mtd/mtdconcat.c +++ b/drivers/mtd/mtdconcat.c @@ -691,7 +691,7 @@ static int concat_block_markbad(struct mtd_info *mtd, loff_t ofs) */ struct mtd_info *mtd_concat_create(struct mtd_info *subdev[], /* subdevices to concatenate */ int num_devs, /* number of subdevices */ - char *name) + const char *name) { /* name for the new device */ int i; size_t size; diff --git a/drivers/mtd/nand/fsl_upm.c b/drivers/mtd/nand/fsl_upm.c index a83192f80eba..7815a404a632 100644 --- a/drivers/mtd/nand/fsl_upm.c +++ b/drivers/mtd/nand/fsl_upm.c @@ -222,7 +222,7 @@ static int __devinit fun_probe(struct of_device *ofdev, fun->rnb_gpio = of_get_gpio(ofdev->node, 0); if (fun->rnb_gpio >= 0) { - ret = gpio_request(fun->rnb_gpio, ofdev->dev.bus_id); + ret = gpio_request(fun->rnb_gpio, dev_name(&ofdev->dev)); if (ret) { dev_err(&ofdev->dev, "can't request RNB gpio\n"); goto err2; diff --git a/drivers/mtd/nand/plat_nand.c b/drivers/mtd/nand/plat_nand.c index f674c5427b17..75f9f4874ecf 100644 --- a/drivers/mtd/nand/plat_nand.c +++ b/drivers/mtd/nand/plat_nand.c @@ -54,7 +54,7 @@ static int __init plat_nand_probe(struct platform_device *pdev) data->chip.priv = &data; data->mtd.priv = &data->chip; data->mtd.owner = THIS_MODULE; - data->mtd.name = pdev->dev.bus_id; + data->mtd.name = dev_name(&pdev->dev); data->chip.IO_ADDR_R = data->io_base; data->chip.IO_ADDR_W = data->io_base; diff --git a/drivers/mtd/nand/tmio_nand.c b/drivers/mtd/nand/tmio_nand.c index edb1e322113d..daa6a4c3b8ce 100644 --- a/drivers/mtd/nand/tmio_nand.c +++ b/drivers/mtd/nand/tmio_nand.c @@ -433,7 +433,7 @@ static int tmio_probe(struct platform_device *dev) nand_chip->chip_delay = 15; retval = request_irq(irq, &tmio_irq, - IRQF_DISABLED, dev->dev.bus_id, tmio); + IRQF_DISABLED, dev_name(&dev->dev), tmio); if (retval) { dev_err(&dev->dev, "request_irq error %d\n", retval); goto err_irq; diff --git a/drivers/mtd/onenand/generic.c b/drivers/mtd/onenand/generic.c index ad81ab8e95e2..5b69e7773c6c 100644 --- a/drivers/mtd/onenand/generic.c +++ b/drivers/mtd/onenand/generic.c @@ -63,7 +63,7 @@ static int __devinit generic_onenand_probe(struct device *dev) info->onenand.mmcontrol = pdata->mmcontrol; info->onenand.irq = platform_get_irq(pdev, 0); - info->mtd.name = pdev->dev.bus_id; + info->mtd.name = dev_name(&pdev->dev); info->mtd.priv = &info->onenand; info->mtd.owner = THIS_MODULE; diff --git a/drivers/mtd/onenand/omap2.c b/drivers/mtd/onenand/omap2.c index a7e4d985f5ef..710edee790b5 100644 --- a/drivers/mtd/onenand/omap2.c +++ b/drivers/mtd/onenand/omap2.c @@ -668,7 +668,7 @@ static int __devinit omap2_onenand_probe(struct platform_device *pdev) c->onenand.base); c->pdev = pdev; - c->mtd.name = pdev->dev.bus_id; + c->mtd.name = dev_name(&pdev->dev); c->mtd.priv = &c->onenand; c->mtd.owner = THIS_MODULE; diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index 634e2e86525f..84a134ead7cc 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -280,7 +280,7 @@ static int ubi_sysfs_init(struct ubi_device *ubi) ubi->dev.release = dev_release; ubi->dev.devt = ubi->cdev.dev; ubi->dev.class = ubi_class; - sprintf(&ubi->dev.bus_id[0], UBI_NAME_STR"%d", ubi->ubi_num); + dev_set_name(&ubi->dev, UBI_NAME_STR"%d", ubi->ubi_num); err = device_register(&ubi->dev); if (err) return err; diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c index 3531ca9a1e24..22e1d7398fce 100644 --- a/drivers/mtd/ubi/vmt.c +++ b/drivers/mtd/ubi/vmt.c @@ -329,7 +329,7 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req) vol->dev.devt = dev; vol->dev.class = ubi_class; - sprintf(&vol->dev.bus_id[0], "%s_%d", ubi->ubi_name, vol->vol_id); + dev_set_name(&vol->dev, "%s_%d", ubi->ubi_name, vol->vol_id); err = device_register(&vol->dev); if (err) { ubi_err("cannot register device"); @@ -678,7 +678,7 @@ int ubi_add_volume(struct ubi_device *ubi, struct ubi_volume *vol) vol->dev.parent = &ubi->dev; vol->dev.devt = dev; vol->dev.class = ubi_class; - sprintf(&vol->dev.bus_id[0], "%s_%d", ubi->ubi_name, vol->vol_id); + dev_set_name(&vol->dev, "%s_%d", ubi->ubi_name, vol->vol_id); err = device_register(&vol->dev); if (err) goto out_gluebi; diff --git a/include/linux/mtd/concat.h b/include/linux/mtd/concat.h index c02f3d264ecf..e80c674daeb3 100644 --- a/include/linux/mtd/concat.h +++ b/include/linux/mtd/concat.h @@ -13,7 +13,7 @@ struct mtd_info *mtd_concat_create( struct mtd_info *subdev[], /* subdevices to concatenate */ int num_devs, /* number of subdevices */ - char *name); /* name for the new device */ + const char *name); /* name for the new device */ void mtd_concat_destroy(struct mtd_info *mtd); -- cgit v1.2.3 From 70a7d3cc1308a55104fbe505d76f2aca8a4cf53e Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Mon, 22 Dec 2008 10:26:05 -0800 Subject: swiotlb: add hwdev to swiotlb_phys_to_bus() / swiotlb_sg_to_bus() Impact: extend functions with a (yet unused) parameter, update callsites Some architectures need it - in preparation for highmem swiotlb. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- arch/x86/kernel/pci-swiotlb_64.c | 2 +- include/linux/swiotlb.h | 3 ++- lib/swiotlb.c | 53 +++++++++++++++++----------------------- 3 files changed, 25 insertions(+), 33 deletions(-) (limited to 'include/linux') diff --git a/arch/x86/kernel/pci-swiotlb_64.c b/arch/x86/kernel/pci-swiotlb_64.c index 242c3440687f..6cf8a816dc29 100644 --- a/arch/x86/kernel/pci-swiotlb_64.c +++ b/arch/x86/kernel/pci-swiotlb_64.c @@ -23,7 +23,7 @@ void *swiotlb_alloc(unsigned order, unsigned long nslabs) return (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, order); } -dma_addr_t swiotlb_phys_to_bus(phys_addr_t paddr) +dma_addr_t swiotlb_phys_to_bus(struct device *hwdev, phys_addr_t paddr) { return paddr; } diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h index 325af1de0351..dedd3c0cfe30 100644 --- a/include/linux/swiotlb.h +++ b/include/linux/swiotlb.h @@ -27,7 +27,8 @@ swiotlb_init(void); extern void *swiotlb_alloc_boot(size_t bytes, unsigned long nslabs); extern void *swiotlb_alloc(unsigned order, unsigned long nslabs); -extern dma_addr_t swiotlb_phys_to_bus(phys_addr_t address); +extern dma_addr_t swiotlb_phys_to_bus(struct device *hwdev, + phys_addr_t address); extern phys_addr_t swiotlb_bus_to_phys(dma_addr_t address); extern int swiotlb_arch_range_needs_mapping(void *ptr, size_t size); diff --git a/lib/swiotlb.c b/lib/swiotlb.c index fa2dc4e5f9ba..3657da8ebbc3 100644 --- a/lib/swiotlb.c +++ b/lib/swiotlb.c @@ -126,7 +126,7 @@ void * __weak swiotlb_alloc(unsigned order, unsigned long nslabs) return (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, order); } -dma_addr_t __weak swiotlb_phys_to_bus(phys_addr_t paddr) +dma_addr_t __weak swiotlb_phys_to_bus(struct device *hwdev, phys_addr_t paddr) { return paddr; } @@ -136,9 +136,10 @@ phys_addr_t __weak swiotlb_bus_to_phys(dma_addr_t baddr) return baddr; } -static dma_addr_t swiotlb_virt_to_bus(volatile void *address) +static dma_addr_t swiotlb_virt_to_bus(struct device *hwdev, + volatile void *address) { - return swiotlb_phys_to_bus(virt_to_phys(address)); + return swiotlb_phys_to_bus(hwdev, virt_to_phys(address)); } static void *swiotlb_bus_to_virt(dma_addr_t address) @@ -151,35 +152,23 @@ int __weak swiotlb_arch_range_needs_mapping(void *ptr, size_t size) return 0; } -static dma_addr_t swiotlb_sg_to_bus(struct scatterlist *sg) +static dma_addr_t swiotlb_sg_to_bus(struct device *hwdev, struct scatterlist *sg) { - return swiotlb_phys_to_bus(page_to_phys(sg_page(sg)) + sg->offset); + return swiotlb_phys_to_bus(hwdev, page_to_phys(sg_page(sg)) + sg->offset); } static void swiotlb_print_info(unsigned long bytes) { phys_addr_t pstart, pend; - dma_addr_t bstart, bend; pstart = virt_to_phys(io_tlb_start); pend = virt_to_phys(io_tlb_end); - bstart = swiotlb_phys_to_bus(pstart); - bend = swiotlb_phys_to_bus(pend); - printk(KERN_INFO "Placing %luMB software IO TLB between %p - %p\n", bytes >> 20, io_tlb_start, io_tlb_end); - if (pstart != bstart || pend != bend) - printk(KERN_INFO "software IO TLB at phys %#llx - %#llx" - " bus %#llx - %#llx\n", - (unsigned long long)pstart, - (unsigned long long)pend, - (unsigned long long)bstart, - (unsigned long long)bend); - else - printk(KERN_INFO "software IO TLB at phys %#llx - %#llx\n", - (unsigned long long)pstart, - (unsigned long long)pend); + printk(KERN_INFO "software IO TLB at phys %#llx - %#llx\n", + (unsigned long long)pstart, + (unsigned long long)pend); } /* @@ -406,7 +395,7 @@ map_single(struct device *hwdev, struct swiotlb_phys_addr buffer, size_t size, i struct swiotlb_phys_addr slot_buf; mask = dma_get_seg_boundary(hwdev); - start_dma_addr = swiotlb_virt_to_bus(io_tlb_start) & mask; + start_dma_addr = swiotlb_virt_to_bus(hwdev, io_tlb_start) & mask; offset_slots = ALIGN(start_dma_addr, 1 << IO_TLB_SHIFT) >> IO_TLB_SHIFT; @@ -585,7 +574,9 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size, dma_mask = hwdev->coherent_dma_mask; ret = (void *)__get_free_pages(flags, order); - if (ret && !is_buffer_dma_capable(dma_mask, swiotlb_virt_to_bus(ret), size)) { + if (ret && + !is_buffer_dma_capable(dma_mask, swiotlb_virt_to_bus(hwdev, ret), + size)) { /* * The allocated memory isn't reachable by the device. * Fall back on swiotlb_map_single(). @@ -609,7 +600,7 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size, } memset(ret, 0, size); - dev_addr = swiotlb_virt_to_bus(ret); + dev_addr = swiotlb_virt_to_bus(hwdev, ret); /* Confirm address can be DMA'd by device */ if (!is_buffer_dma_capable(dma_mask, dev_addr, size)) { @@ -669,7 +660,7 @@ dma_addr_t swiotlb_map_single_attrs(struct device *hwdev, void *ptr, size_t size, int dir, struct dma_attrs *attrs) { - dma_addr_t dev_addr = swiotlb_virt_to_bus(ptr); + dma_addr_t dev_addr = swiotlb_virt_to_bus(hwdev, ptr); void *map; struct swiotlb_phys_addr buffer; @@ -694,7 +685,7 @@ swiotlb_map_single_attrs(struct device *hwdev, void *ptr, size_t size, map = io_tlb_overflow_buffer; } - dev_addr = swiotlb_virt_to_bus(map); + dev_addr = swiotlb_virt_to_bus(hwdev, map); /* * Ensure that the address returned is DMA'ble @@ -840,7 +831,7 @@ swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl, int nelems, BUG_ON(dir == DMA_NONE); for_each_sg(sgl, sg, nelems, i) { - dev_addr = swiotlb_sg_to_bus(sg); + dev_addr = swiotlb_sg_to_bus(hwdev, sg); if (range_needs_mapping(sg_virt(sg), sg->length) || address_needs_mapping(hwdev, dev_addr, sg->length)) { void *map; @@ -856,7 +847,7 @@ swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl, int nelems, sgl[0].dma_length = 0; return 0; } - sg->dma_address = swiotlb_virt_to_bus(map); + sg->dma_address = swiotlb_virt_to_bus(hwdev, map); } else sg->dma_address = dev_addr; sg->dma_length = sg->length; @@ -886,7 +877,7 @@ swiotlb_unmap_sg_attrs(struct device *hwdev, struct scatterlist *sgl, BUG_ON(dir == DMA_NONE); for_each_sg(sgl, sg, nelems, i) { - if (sg->dma_address != swiotlb_sg_to_bus(sg)) + if (sg->dma_address != swiotlb_sg_to_bus(hwdev, sg)) unmap_single(hwdev, swiotlb_bus_to_virt(sg->dma_address), sg->dma_length, dir); else if (dir == DMA_FROM_DEVICE) @@ -919,7 +910,7 @@ swiotlb_sync_sg(struct device *hwdev, struct scatterlist *sgl, BUG_ON(dir == DMA_NONE); for_each_sg(sgl, sg, nelems, i) { - if (sg->dma_address != swiotlb_sg_to_bus(sg)) + if (sg->dma_address != swiotlb_sg_to_bus(hwdev, sg)) sync_single(hwdev, swiotlb_bus_to_virt(sg->dma_address), sg->dma_length, dir, target); else if (dir == DMA_FROM_DEVICE) @@ -944,7 +935,7 @@ swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg, int swiotlb_dma_mapping_error(struct device *hwdev, dma_addr_t dma_addr) { - return (dma_addr == swiotlb_virt_to_bus(io_tlb_overflow_buffer)); + return (dma_addr == swiotlb_virt_to_bus(hwdev, io_tlb_overflow_buffer)); } /* @@ -956,7 +947,7 @@ swiotlb_dma_mapping_error(struct device *hwdev, dma_addr_t dma_addr) int swiotlb_dma_supported(struct device *hwdev, u64 mask) { - return swiotlb_virt_to_bus(io_tlb_end - 1) <= mask; + return swiotlb_virt_to_bus(hwdev, io_tlb_end - 1) <= mask; } EXPORT_SYMBOL(swiotlb_map_single); -- cgit v1.2.3 From ea319518ba3de282c13ae1cf4bf2215c5e03e67e Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 26 Dec 2008 15:08:55 +0100 Subject: locking, percpu counters: introduce separate lock classes Impact: fix lockdep false positives Classify percpu_counter instances similar to regular lock objects -- that is, per instantiation site. The networking code has increased its use of percpu_counters, which leads to false positives if they are treated as a single class. Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- include/linux/percpu_counter.h | 14 ++++++++++---- lib/percpu_counter.c | 18 ++++-------------- lib/proportions.c | 6 +++--- mm/backing-dev.c | 2 +- 4 files changed, 18 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/include/linux/percpu_counter.h b/include/linux/percpu_counter.h index 9007ccdfc112..96bdde36599f 100644 --- a/include/linux/percpu_counter.h +++ b/include/linux/percpu_counter.h @@ -30,8 +30,16 @@ struct percpu_counter { #define FBC_BATCH (NR_CPUS*4) #endif -int percpu_counter_init(struct percpu_counter *fbc, s64 amount); -int percpu_counter_init_irq(struct percpu_counter *fbc, s64 amount); +int __percpu_counter_init(struct percpu_counter *fbc, s64 amount, + struct lock_class_key *key); + +#define percpu_counter_init(fbc, value) \ + ({ \ + static struct lock_class_key __key; \ + \ + __percpu_counter_init(fbc, value, &__key); \ + }) + void percpu_counter_destroy(struct percpu_counter *fbc); void percpu_counter_set(struct percpu_counter *fbc, s64 amount); void __percpu_counter_add(struct percpu_counter *fbc, s64 amount, s32 batch); @@ -85,8 +93,6 @@ static inline int percpu_counter_init(struct percpu_counter *fbc, s64 amount) return 0; } -#define percpu_counter_init_irq percpu_counter_init - static inline void percpu_counter_destroy(struct percpu_counter *fbc) { } diff --git a/lib/percpu_counter.c b/lib/percpu_counter.c index a8663890a88c..c7fe2e4e8ed1 100644 --- a/lib/percpu_counter.c +++ b/lib/percpu_counter.c @@ -71,11 +71,11 @@ s64 __percpu_counter_sum(struct percpu_counter *fbc) } EXPORT_SYMBOL(__percpu_counter_sum); -static struct lock_class_key percpu_counter_irqsafe; - -int percpu_counter_init(struct percpu_counter *fbc, s64 amount) +int __percpu_counter_init(struct percpu_counter *fbc, s64 amount, + struct lock_class_key *key) { spin_lock_init(&fbc->lock); + lockdep_set_class(&fbc->lock, key); fbc->count = amount; fbc->counters = alloc_percpu(s32); if (!fbc->counters) @@ -87,17 +87,7 @@ int percpu_counter_init(struct percpu_counter *fbc, s64 amount) #endif return 0; } -EXPORT_SYMBOL(percpu_counter_init); - -int percpu_counter_init_irq(struct percpu_counter *fbc, s64 amount) -{ - int err; - - err = percpu_counter_init(fbc, amount); - if (!err) - lockdep_set_class(&fbc->lock, &percpu_counter_irqsafe); - return err; -} +EXPORT_SYMBOL(__percpu_counter_init); void percpu_counter_destroy(struct percpu_counter *fbc) { diff --git a/lib/proportions.c b/lib/proportions.c index 4f387a643d72..7367f2b727d0 100644 --- a/lib/proportions.c +++ b/lib/proportions.c @@ -83,11 +83,11 @@ int prop_descriptor_init(struct prop_descriptor *pd, int shift) pd->index = 0; pd->pg[0].shift = shift; mutex_init(&pd->mutex); - err = percpu_counter_init_irq(&pd->pg[0].events, 0); + err = percpu_counter_init(&pd->pg[0].events, 0); if (err) goto out; - err = percpu_counter_init_irq(&pd->pg[1].events, 0); + err = percpu_counter_init(&pd->pg[1].events, 0); if (err) percpu_counter_destroy(&pd->pg[0].events); @@ -191,7 +191,7 @@ int prop_local_init_percpu(struct prop_local_percpu *pl) spin_lock_init(&pl->lock); pl->shift = 0; pl->period = 0; - return percpu_counter_init_irq(&pl->events, 0); + return percpu_counter_init(&pl->events, 0); } void prop_local_destroy_percpu(struct prop_local_percpu *pl) diff --git a/mm/backing-dev.c b/mm/backing-dev.c index f2e574dbc300..f3b125857827 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -220,7 +220,7 @@ int bdi_init(struct backing_dev_info *bdi) bdi->max_prop_frac = PROP_FRAC_BASE; for (i = 0; i < NR_BDI_STAT_ITEMS; i++) { - err = percpu_counter_init_irq(&bdi->bdi_stat[i], 0); + err = percpu_counter_init(&bdi->bdi_stat[i], 0); if (err) goto err; } -- cgit v1.2.3 From 34a4c5eb421dab6fe8381aa12c990f9d6f645b17 Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Mon, 29 Dec 2008 04:00:23 -0800 Subject: Input: map_to_7segment.h - convert to __inline__ for userspace Use __inline__ rather than inline for map_to_seg7() since it is exported to userspace. Signed-off-by: Mike Frysinger Signed-off-by: Dmitry Torokhov --- include/linux/map_to_7segment.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/map_to_7segment.h b/include/linux/map_to_7segment.h index 7df8432c4402..12d62a54d470 100644 --- a/include/linux/map_to_7segment.h +++ b/include/linux/map_to_7segment.h @@ -75,7 +75,7 @@ struct seg7_conversion_map { unsigned char table[128]; }; -static inline int map_to_seg7(struct seg7_conversion_map *map, int c) +static __inline__ int map_to_seg7(struct seg7_conversion_map *map, int c) { return c >= 0 && c < sizeof(map->table) ? map->table[c] : -EINVAL; } -- cgit v1.2.3 From 47fea2adfc9e16846bc57c2f64ff233b354fef39 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Rajput Date: Mon, 29 Dec 2008 23:39:17 +0530 Subject: sched: sched.c declare variables before they get used Impact: cleanup, avoid sparse warnings In linux/sched.h moved out sysctl_sched_latency, sysctl_sched_min_granularity, sysctl_sched_wakeup_granularity, sysctl_sched_shares_ratelimit and sysctl_sched_shares_thresh from #ifdef CONFIG_SCHED_DEBUG as these variables are common for both. Fixes these sparse warnings: kernel/sched.c:825:14: warning: symbol 'sysctl_sched_shares_ratelimit' was not declared. Should it be static? kernel/sched.c:832:14: warning: symbol 'sysctl_sched_shares_thresh' was not declared. Should it be static? kernel/sched_fair.c:37:14: warning: symbol 'sysctl_sched_latency' was not declared. Should it be static? kernel/sched_fair.c:43:14: warning: symbol 'sysctl_sched_min_granularity' was not declared. Should it be static? kernel/sched_fair.c:72:14: warning: symbol 'sysctl_sched_wakeup_granularity' was not declared. Should it be static? Signed-off-by: Jaswinder Singh Rajput Signed-off-by: Ingo Molnar --- include/linux/sched.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 8395e715809d..01d9fd268eb0 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1651,16 +1651,16 @@ extern void wake_up_idle_cpu(int cpu); static inline void wake_up_idle_cpu(int cpu) { } #endif -#ifdef CONFIG_SCHED_DEBUG extern unsigned int sysctl_sched_latency; extern unsigned int sysctl_sched_min_granularity; extern unsigned int sysctl_sched_wakeup_granularity; +extern unsigned int sysctl_sched_shares_ratelimit; +extern unsigned int sysctl_sched_shares_thresh; +#ifdef CONFIG_SCHED_DEBUG extern unsigned int sysctl_sched_child_runs_first; extern unsigned int sysctl_sched_features; extern unsigned int sysctl_sched_migration_cost; extern unsigned int sysctl_sched_nr_migrate; -extern unsigned int sysctl_sched_shares_ratelimit; -extern unsigned int sysctl_sched_shares_thresh; int sched_nr_latency_handler(struct ctl_table *table, int write, struct file *file, void __user *buffer, size_t *length, -- cgit v1.2.3 From f748bafa3ca1fb056e63afdeecacc1c68d8104df Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Mon, 8 Dec 2008 21:30:31 -0700 Subject: ACPI: PCI: move struct acpi_prt_entry declaration out of public header file The struct acpi_prt_entry is used only in pci_irq.c, so there's no need for the declaration to be public. This patch moves it into pci_irq.c. Signed-off-by: Bjorn Helgaas Signed-off-by: Len Brown --- drivers/acpi/pci_irq.c | 16 ++++++++++++++++ include/linux/acpi.h | 16 ---------------- 2 files changed, 16 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/drivers/acpi/pci_irq.c b/drivers/acpi/pci_irq.c index 9302f4bb89e2..ea003bab7ecd 100644 --- a/drivers/acpi/pci_irq.c +++ b/drivers/acpi/pci_irq.c @@ -41,6 +41,22 @@ #define _COMPONENT ACPI_PCI_COMPONENT ACPI_MODULE_NAME("pci_irq"); +struct acpi_prt_entry { + struct list_head node; + struct acpi_pci_id id; + u8 pin; + struct { + acpi_handle handle; + u32 index; + } link; + u32 irq; +}; + +struct acpi_prt_list { + int count; + struct list_head entries; +}; + static struct acpi_prt_list acpi_prt; static DEFINE_SPINLOCK(acpi_prt_lock); diff --git a/include/linux/acpi.h b/include/linux/acpi.h index fba8051fb297..813f937b3ab4 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -131,22 +131,6 @@ extern int acpi_get_override_irq(int bus_irq, int *trigger, int *polarity); */ void acpi_unregister_gsi (u32 gsi); -struct acpi_prt_entry { - struct list_head node; - struct acpi_pci_id id; - u8 pin; - struct { - acpi_handle handle; - u32 index; - } link; - u32 irq; -}; - -struct acpi_prt_list { - int count; - struct list_head entries; -}; - struct pci_dev; int acpi_pci_irq_enable (struct pci_dev *dev); -- cgit v1.2.3 From ea7e96e0f2277107d9ea14c3f16c86ba82b2e560 Mon Sep 17 00:00:00 2001 From: Lin Ming Date: Tue, 16 Dec 2008 16:28:17 +0800 Subject: ACPI: remove private acpica headers from driver files External driver files should not include any private acpica headers. Signed-off-by: Lin Ming Signed-off-by: Len Brown --- arch/ia64/include/asm/acpi-ext.h | 1 - arch/ia64/include/asm/sn/acpi.h | 2 -- arch/ia64/sn/kernel/io_acpi_init.c | 1 - arch/ia64/sn/kernel/io_common.c | 1 - drivers/acpi/cm_sbs.c | 3 --- drivers/acpi/debug.c | 1 - drivers/acpi/ec.c | 1 - drivers/acpi/numa.c | 1 - drivers/acpi/sbshc.c | 1 - drivers/acpi/scan.c | 1 - drivers/acpi/sleep/wakeup.c | 1 - drivers/ata/libata-acpi.c | 6 ------ drivers/ata/pata_acpi.c | 6 ------ drivers/char/tpm/tpm_bios.c | 2 -- drivers/ide/ide-acpi.c | 6 ------ drivers/misc/tc1100-wmi.c | 1 - drivers/misc/thinkpad_acpi.c | 1 - drivers/pci/hotplug/acpi_pcihp.c | 1 - drivers/pci/hotplug/pciehp.h | 1 - drivers/pci/pci-acpi.c | 2 -- drivers/pnp/pnpacpi/core.c | 1 - include/linux/pci_hotplug.h | 1 - 22 files changed, 42 deletions(-) (limited to 'include/linux') diff --git a/arch/ia64/include/asm/acpi-ext.h b/arch/ia64/include/asm/acpi-ext.h index 734d137dda6e..7f8362b379eb 100644 --- a/arch/ia64/include/asm/acpi-ext.h +++ b/arch/ia64/include/asm/acpi-ext.h @@ -14,7 +14,6 @@ #define _ASM_IA64_ACPI_EXT_H #include -#include extern acpi_status hp_acpi_csr_space (acpi_handle, u64 *base, u64 *length); diff --git a/arch/ia64/include/asm/sn/acpi.h b/arch/ia64/include/asm/sn/acpi.h index 9ce2801cbd57..fd480db25565 100644 --- a/arch/ia64/include/asm/sn/acpi.h +++ b/arch/ia64/include/asm/sn/acpi.h @@ -9,8 +9,6 @@ #ifndef _ASM_IA64_SN_ACPI_H #define _ASM_IA64_SN_ACPI_H -#include "acpi/acglobal.h" - extern int sn_acpi_rev; #define SN_ACPI_BASE_SUPPORT() (sn_acpi_rev >= 0x20101) diff --git a/arch/ia64/sn/kernel/io_acpi_init.c b/arch/ia64/sn/kernel/io_acpi_init.c index 4c8bc8eee5ad..c5a214026a77 100644 --- a/arch/ia64/sn/kernel/io_acpi_init.c +++ b/arch/ia64/sn/kernel/io_acpi_init.c @@ -13,7 +13,6 @@ #include #include "xtalk/hubdev.h" #include -#include /* diff --git a/arch/ia64/sn/kernel/io_common.c b/arch/ia64/sn/kernel/io_common.c index e1917ede0d60..0d4ffa4da1da 100644 --- a/arch/ia64/sn/kernel/io_common.c +++ b/arch/ia64/sn/kernel/io_common.c @@ -26,7 +26,6 @@ #include #include #include -#include "acpi/acglobal.h" extern void sn_init_cpei_timer(void); extern void register_sn_procfs(void); diff --git a/drivers/acpi/cm_sbs.c b/drivers/acpi/cm_sbs.c index 307963bd1043..332fe4b21708 100644 --- a/drivers/acpi/cm_sbs.c +++ b/drivers/acpi/cm_sbs.c @@ -27,9 +27,6 @@ #include #include #include -#include -#include -#include ACPI_MODULE_NAME("cm_sbs"); #define ACPI_AC_CLASS "ac_adapter" diff --git a/drivers/acpi/debug.c b/drivers/acpi/debug.c index c48396892008..20223cbd0d1c 100644 --- a/drivers/acpi/debug.c +++ b/drivers/acpi/debug.c @@ -9,7 +9,6 @@ #include #include #include -#include #define _COMPONENT ACPI_SYSTEM_COMPONENT ACPI_MODULE_NAME("debug"); diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c index 2cbc2c9c07ac..3105e0410e9b 100644 --- a/drivers/acpi/ec.c +++ b/drivers/acpi/ec.c @@ -42,7 +42,6 @@ #include #include #include -#include #define ACPI_EC_CLASS "embedded_controller" #define ACPI_EC_DEVICE_NAME "Embedded Controller" diff --git a/drivers/acpi/numa.c b/drivers/acpi/numa.c index 25ceae9191ef..c5e292aab0e3 100644 --- a/drivers/acpi/numa.c +++ b/drivers/acpi/numa.c @@ -29,7 +29,6 @@ #include #include #include -#include #define ACPI_NUMA 0x80000000 #define _COMPONENT ACPI_NUMA diff --git a/drivers/acpi/sbshc.c b/drivers/acpi/sbshc.c index e53e590252c0..0619734895b2 100644 --- a/drivers/acpi/sbshc.c +++ b/drivers/acpi/sbshc.c @@ -10,7 +10,6 @@ #include #include -#include #include #include #include diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c index 39b7233c3485..c54d7b6c4066 100644 --- a/drivers/acpi/scan.c +++ b/drivers/acpi/scan.c @@ -10,7 +10,6 @@ #include #include -#include /* for acpi_ex_eisa_id_to_string() */ #define _COMPONENT ACPI_BUS_COMPONENT ACPI_MODULE_NAME("scan"); diff --git a/drivers/acpi/sleep/wakeup.c b/drivers/acpi/sleep/wakeup.c index db325c28a1fb..2d34806d45dd 100644 --- a/drivers/acpi/sleep/wakeup.c +++ b/drivers/acpi/sleep/wakeup.c @@ -8,7 +8,6 @@ #include #include #include -#include #include "sleep.h" #define _COMPONENT ACPI_SYSTEM_COMPONENT diff --git a/drivers/ata/libata-acpi.c b/drivers/ata/libata-acpi.c index c012307d0ba6..246987f0b88c 100644 --- a/drivers/ata/libata-acpi.c +++ b/drivers/ata/libata-acpi.c @@ -19,12 +19,6 @@ #include "libata.h" #include -#include -#include -#include -#include -#include -#include enum { ATA_ACPI_FILTER_SETXFER = 1 << 0, diff --git a/drivers/ata/pata_acpi.c b/drivers/ata/pata_acpi.c index e2e332d8ff95..8b77a9802df1 100644 --- a/drivers/ata/pata_acpi.c +++ b/drivers/ata/pata_acpi.c @@ -13,12 +13,6 @@ #include #include #include -#include -#include -#include -#include -#include -#include #include #include diff --git a/drivers/char/tpm/tpm_bios.c b/drivers/char/tpm/tpm_bios.c index 68f052b42ed7..ed306eb1057f 100644 --- a/drivers/char/tpm/tpm_bios.c +++ b/drivers/char/tpm/tpm_bios.c @@ -23,8 +23,6 @@ #include #include #include -#include -#include #include "tpm.h" #define TCG_EVENT_NAME_LEN_MAX 255 diff --git a/drivers/ide/ide-acpi.c b/drivers/ide/ide-acpi.c index 244a8a052ce8..9e8d52a4f306 100644 --- a/drivers/ide/ide-acpi.c +++ b/drivers/ide/ide-acpi.c @@ -18,12 +18,6 @@ #include #include -#include -#include -#include -#include -#include -#include #define REGS_PER_GTF 7 struct taskfile_array { diff --git a/drivers/misc/tc1100-wmi.c b/drivers/misc/tc1100-wmi.c index f25e4c974dcf..b4a4aa9ee482 100644 --- a/drivers/misc/tc1100-wmi.c +++ b/drivers/misc/tc1100-wmi.c @@ -30,7 +30,6 @@ #include #include #include -#include #include #include #include diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c index 899766e16fa8..3478453eba7a 100644 --- a/drivers/misc/thinkpad_acpi.c +++ b/drivers/misc/thinkpad_acpi.c @@ -76,7 +76,6 @@ #include #include -#include #include diff --git a/drivers/pci/hotplug/acpi_pcihp.c b/drivers/pci/hotplug/acpi_pcihp.c index e17ef54f0efc..2c981cbb0719 100644 --- a/drivers/pci/hotplug/acpi_pcihp.c +++ b/drivers/pci/hotplug/acpi_pcihp.c @@ -33,7 +33,6 @@ #include #include #include -#include #define MY_NAME "acpi_pcihp" diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h index b2801a7ee37f..7072952ea1d2 100644 --- a/drivers/pci/hotplug/pciehp.h +++ b/drivers/pci/hotplug/pciehp.h @@ -217,7 +217,6 @@ struct hpc_ops { #ifdef CONFIG_ACPI #include #include -#include #include static inline int pciehp_get_hp_hw_control_from_firmware(struct pci_dev *dev) diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c index ae5ec76dca77..9d976d51d406 100644 --- a/drivers/pci/pci-acpi.c +++ b/drivers/pci/pci-acpi.c @@ -13,8 +13,6 @@ #include #include #include -#include -#include #include #include diff --git a/drivers/pnp/pnpacpi/core.c b/drivers/pnp/pnpacpi/core.c index 383e47c392a4..2834846a185d 100644 --- a/drivers/pnp/pnpacpi/core.c +++ b/drivers/pnp/pnpacpi/core.c @@ -23,7 +23,6 @@ #include #include #include -#include #include "../base.h" #include "pnpacpi.h" diff --git a/include/linux/pci_hotplug.h b/include/linux/pci_hotplug.h index a00bd1a0f156..c2d1a7d1886a 100644 --- a/include/linux/pci_hotplug.h +++ b/include/linux/pci_hotplug.h @@ -223,7 +223,6 @@ struct hotplug_params { #ifdef CONFIG_ACPI #include #include -#include extern acpi_status acpi_get_hp_params_from_firmware(struct pci_bus *bus, struct hotplug_params *hpp); int acpi_get_hp_hw_control_from_firmware(struct pci_dev *dev, u32 flags); -- cgit v1.2.3 From 1c5745aa380efb6417b5681104b007c8612fb496 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 22 Dec 2008 23:05:28 +0100 Subject: sched_clock: prevent scd->clock from moving backwards, take #2 Redo: 5b7dba4: sched_clock: prevent scd->clock from moving backwards which had to be reverted due to s2ram hangs: ca7e716: Revert "sched_clock: prevent scd->clock from moving backwards" ... this time with resume restoring GTOD later in the sequence taken into account as well. The "timekeeping_suspended" flag is not very nice but we cannot call into GTOD before it has been properly resumed and the scheduler will run very early in the resume sequence. Cc: Signed-off-by: Ingo Molnar --- include/linux/time.h | 1 + kernel/sched_clock.c | 5 ++++- kernel/time/timekeeping.c | 7 +++++-- 3 files changed, 10 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/time.h b/include/linux/time.h index ce321ac5c8f8..fbbd2a1c92ba 100644 --- a/include/linux/time.h +++ b/include/linux/time.h @@ -105,6 +105,7 @@ extern unsigned long read_persistent_clock(void); extern int update_persistent_clock(struct timespec now); extern int no_sync_cmos_clock __read_mostly; void timekeeping_init(void); +extern int timekeeping_suspended; unsigned long get_seconds(void); struct timespec current_kernel_time(void); diff --git a/kernel/sched_clock.c b/kernel/sched_clock.c index e8ab096ddfe3..a0b0852414cc 100644 --- a/kernel/sched_clock.c +++ b/kernel/sched_clock.c @@ -124,7 +124,7 @@ static u64 __update_sched_clock(struct sched_clock_data *scd, u64 now) clock = scd->tick_gtod + delta; min_clock = wrap_max(scd->tick_gtod, scd->clock); - max_clock = scd->tick_gtod + TICK_NSEC; + max_clock = wrap_max(scd->clock, scd->tick_gtod + TICK_NSEC); clock = wrap_max(clock, min_clock); clock = wrap_min(clock, max_clock); @@ -227,6 +227,9 @@ EXPORT_SYMBOL_GPL(sched_clock_idle_sleep_event); */ void sched_clock_idle_wakeup_event(u64 delta_ns) { + if (timekeeping_suspended) + return; + sched_clock_tick(); touch_softlockup_watchdog(); } diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index fa05e88aa76f..900f1b6598d1 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -46,6 +46,9 @@ struct timespec xtime __attribute__ ((aligned (16))); struct timespec wall_to_monotonic __attribute__ ((aligned (16))); static unsigned long total_sleep_time; /* seconds */ +/* flag for if timekeeping is suspended */ +int __read_mostly timekeeping_suspended; + static struct timespec xtime_cache __attribute__ ((aligned (16))); void update_xtime_cache(u64 nsec) { @@ -92,6 +95,8 @@ void getnstimeofday(struct timespec *ts) unsigned long seq; s64 nsecs; + WARN_ON(timekeeping_suspended); + do { seq = read_seqbegin(&xtime_lock); @@ -299,8 +304,6 @@ void __init timekeeping_init(void) write_sequnlock_irqrestore(&xtime_lock, flags); } -/* flag for if timekeeping is suspended */ -static int timekeeping_suspended; /* time in seconds when suspend began */ static unsigned long timekeeping_suspend_time; -- cgit v1.2.3 From 56c451f4b583ccdf80c9e676179c9cb49de86745 Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Thu, 18 Dec 2008 14:49:37 +0900 Subject: [SCSI] block: fix the partial mappings with struct rq_map_data This fixes bio_copy_user_iov to properly handle the partial mappings with struct rq_map_data (which only sg uses for now but st and osst will shortly). It adds the offset member to struct rq_map_data and changes blk_rq_map_user to update it so that bio_copy_user_iov can add an appropriate page frame via bio_add_pc_page(). Signed-off-by: FUJITA Tomonori Acked-by: Jens Axboe Signed-off-by: James Bottomley --- block/blk-map.c | 3 +++ drivers/scsi/sg.c | 1 + fs/bio.c | 12 +++++++++--- include/linux/blkdev.h | 1 + 4 files changed, 14 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/block/blk-map.c b/block/blk-map.c index 2990447f45e9..c7e55b23a2bc 100644 --- a/block/blk-map.c +++ b/block/blk-map.c @@ -150,6 +150,9 @@ int blk_rq_map_user(struct request_queue *q, struct request *rq, bio = rq->bio; bytes_read += ret; ubuf += ret; + + if (map_data) + map_data->offset += ret; } if (!bio_flagged(bio, BIO_USER_MAPPED)) diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c index 5103855242ae..7d0b3d9ee43b 100644 --- a/drivers/scsi/sg.c +++ b/drivers/scsi/sg.c @@ -1669,6 +1669,7 @@ static int sg_start_req(Sg_request *srp, unsigned char *cmd) md->pages = req_schp->pages; md->page_order = req_schp->page_order; md->nr_entries = req_schp->k_use_sg; + md->offset = 0; } if (iov_count) diff --git a/fs/bio.c b/fs/bio.c index 356e7423b923..13be075806b6 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -788,6 +788,7 @@ struct bio *bio_copy_user_iov(struct request_queue *q, int i, ret; int nr_pages = 0; unsigned int len = 0; + unsigned int offset = map_data ? map_data->offset & ~PAGE_MASK : 0; for (i = 0; i < iov_count; i++) { unsigned long uaddr; @@ -814,12 +815,16 @@ struct bio *bio_copy_user_iov(struct request_queue *q, bio->bi_rw |= (!write_to_vm << BIO_RW); ret = 0; - i = 0; - if (map_data) + + if (map_data) { nr_pages = 1 << map_data->page_order; + i = map_data->offset / PAGE_SIZE; + } while (len) { unsigned int bytes = PAGE_SIZE; + bytes -= offset; + if (bytes > len) bytes = len; @@ -841,10 +846,11 @@ struct bio *bio_copy_user_iov(struct request_queue *q, } } - if (bio_add_pc_page(q, bio, page, bytes, 0) < bytes) + if (bio_add_pc_page(q, bio, page, bytes, offset) < bytes) break; len -= bytes; + offset = 0; } if (ret) diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 7035cec583b6..811e5342c452 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -690,6 +690,7 @@ struct rq_map_data { struct page **pages; int page_order; int nr_entries; + unsigned long offset; }; struct req_iterator { -- cgit v1.2.3 From 97ae77a1cd332c7b011d71315c8faabce6840c72 Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Thu, 18 Dec 2008 14:49:38 +0900 Subject: [SCSI] block: make blk_rq_map_user take a NULL user-space buffer for WRITE The commit 818827669d85b84241696ffef2de485db46b0b5e (block: make blk_rq_map_user take a NULL user-space buffer) extended blk_rq_map_user to accept a NULL user-space buffer with a READ command. It was necessary to convert sg to use the block layer mapping API. This patch extends blk_rq_map_user again for a WRITE command. It is necessary to convert st and osst drivers to use the block layer apping API. Signed-off-by: FUJITA Tomonori Acked-by: Jens Axboe Signed-off-by: James Bottomley --- block/blk-map.c | 16 +++++++--------- drivers/scsi/sg.c | 1 + fs/bio.c | 2 +- include/linux/blkdev.h | 1 + 4 files changed, 10 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/block/blk-map.c b/block/blk-map.c index c7e55b23a2bc..f103729b462f 100644 --- a/block/blk-map.c +++ b/block/blk-map.c @@ -42,7 +42,7 @@ static int __blk_rq_unmap_user(struct bio *bio) static int __blk_rq_map_user(struct request_queue *q, struct request *rq, struct rq_map_data *map_data, void __user *ubuf, - unsigned int len, int null_mapped, gfp_t gfp_mask) + unsigned int len, gfp_t gfp_mask) { unsigned long uaddr; struct bio *bio, *orig_bio; @@ -63,7 +63,7 @@ static int __blk_rq_map_user(struct request_queue *q, struct request *rq, if (IS_ERR(bio)) return PTR_ERR(bio); - if (null_mapped) + if (map_data && map_data->null_mapped) bio->bi_flags |= (1 << BIO_NULL_MAPPED); orig_bio = bio; @@ -114,17 +114,15 @@ int blk_rq_map_user(struct request_queue *q, struct request *rq, { unsigned long bytes_read = 0; struct bio *bio = NULL; - int ret, null_mapped = 0; + int ret; if (len > (q->max_hw_sectors << 9)) return -EINVAL; if (!len) return -EINVAL; - if (!ubuf) { - if (!map_data || rq_data_dir(rq) != READ) - return -EINVAL; - null_mapped = 1; - } + + if (!ubuf && (!map_data || !map_data->null_mapped)) + return -EINVAL; while (bytes_read != len) { unsigned long map_len, end, start; @@ -143,7 +141,7 @@ int blk_rq_map_user(struct request_queue *q, struct request *rq, map_len -= PAGE_SIZE; ret = __blk_rq_map_user(q, rq, map_data, ubuf, map_len, - null_mapped, gfp_mask); + gfp_mask); if (ret < 0) goto unmap_rq; if (!bio) diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c index 7d0b3d9ee43b..8f0bd3f7a59f 100644 --- a/drivers/scsi/sg.c +++ b/drivers/scsi/sg.c @@ -1670,6 +1670,7 @@ static int sg_start_req(Sg_request *srp, unsigned char *cmd) md->page_order = req_schp->page_order; md->nr_entries = req_schp->k_use_sg; md->offset = 0; + md->null_mapped = hp->dxferp ? 0 : 1; } if (iov_count) diff --git a/fs/bio.c b/fs/bio.c index 13be075806b6..062299acbccd 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -859,7 +859,7 @@ struct bio *bio_copy_user_iov(struct request_queue *q, /* * success */ - if (!write_to_vm) { + if (!write_to_vm && (!map_data || !map_data->null_mapped)) { ret = __bio_copy_iov(bio, bio->bi_io_vec, iov, iov_count, 0, 0); if (ret) goto cleanup; diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 811e5342c452..044467ef7b11 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -691,6 +691,7 @@ struct rq_map_data { int page_order; int nr_entries; unsigned long offset; + int null_mapped; }; struct req_iterator { -- cgit v1.2.3 From 87d8fe1ee6b8d2f95076142d58c440dba4e7bdc2 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Sat, 3 Jan 2009 09:47:09 -0500 Subject: add releasepage hooks to block devices which can be used by file systems Implement blkdev_releasepage() to release the buffer_heads and pages after we release private data belonging to a mounted filesystem. Cc: Toshiyuki Okajima Cc: linux-fsdevel@vger.kernel.org Signed-off-by: "Theodore Ts'o" --- fs/block_dev.c | 15 +++++++++++++++ fs/super.c | 2 ++ include/linux/fs.h | 2 ++ 3 files changed, 19 insertions(+) (limited to 'include/linux') diff --git a/fs/block_dev.c b/fs/block_dev.c index 349a26c10001..1dd07e66e98a 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -1220,6 +1220,20 @@ static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg) return blkdev_ioctl(bdev, mode, cmd, arg); } +/* + * Try to release a page associated with block device when the system + * is under memory pressure. + */ +static int blkdev_releasepage(struct page *page, gfp_t wait) +{ + struct super_block *super = BDEV_I(page->mapping->host)->bdev.bd_super; + + if (super && super->s_op->bdev_try_to_free_page) + return super->s_op->bdev_try_to_free_page(super, page, wait); + + return try_to_free_buffers(page); +} + static const struct address_space_operations def_blk_aops = { .readpage = blkdev_readpage, .writepage = blkdev_writepage, @@ -1227,6 +1241,7 @@ static const struct address_space_operations def_blk_aops = { .write_begin = blkdev_write_begin, .write_end = blkdev_write_end, .writepages = generic_writepages, + .releasepage = blkdev_releasepage, .direct_IO = blkdev_direct_IO, }; diff --git a/fs/super.c b/fs/super.c index ddba069d7a99..d5fd4498548a 100644 --- a/fs/super.c +++ b/fs/super.c @@ -800,6 +800,7 @@ int get_sb_bdev(struct file_system_type *fs_type, } s->s_flags |= MS_ACTIVE; + bdev->bd_super = s; } return simple_set_mnt(mnt, s); @@ -819,6 +820,7 @@ void kill_block_super(struct super_block *sb) struct block_device *bdev = sb->s_bdev; fmode_t mode = sb->s_mode; + bdev->bd_super = 0; generic_shutdown_super(sb); sync_blockdev(bdev); close_bdev_exclusive(bdev, mode); diff --git a/include/linux/fs.h b/include/linux/fs.h index f2a3010140e3..0f54ae0f0ccd 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -565,6 +565,7 @@ struct address_space { struct block_device { dev_t bd_dev; /* not a kdev_t - it's a search key */ struct inode * bd_inode; /* will die */ + struct super_block * bd_super; int bd_openers; struct mutex bd_mutex; /* open/close mutex */ struct semaphore bd_mount_sem; @@ -1385,6 +1386,7 @@ struct super_operations { ssize_t (*quota_read)(struct super_block *, int, char *, size_t, loff_t); ssize_t (*quota_write)(struct super_block *, int, const char *, size_t, loff_t); #endif + int (*bdev_try_to_free_page)(struct super_block*, struct page*, gfp_t); }; /* -- cgit v1.2.3 From c31910672376dfb8d020e32afa7249763bcd924a Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Tue, 6 Jan 2009 11:14:25 -0500 Subject: ext4: Remove code to create the journal inode This code has been obsolete in quite some time, since the supported method for adding a journal inode is to use tune2fs (or to creating new filesystem with a journal via mke2fs or mkfs.ext4). Signed-off-by: "Theodore Ts'o" --- Documentation/filesystems/ext4.txt | 4 --- fs/ext4/super.c | 68 +++-------------------------------- fs/jbd2/journal.c | 72 -------------------------------------- include/linux/jbd2.h | 1 - 4 files changed, 4 insertions(+), 141 deletions(-) (limited to 'include/linux') diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt index e3fcbea3ec8c..9ec29d86ff8b 100644 --- a/Documentation/filesystems/ext4.txt +++ b/Documentation/filesystems/ext4.txt @@ -149,10 +149,6 @@ journal_async_commit Commit block can be written to disk without waiting journal=update Update the ext4 file system's journal to the current format. -journal=inum When a journal already exists, this option is ignored. - Otherwise, it specifies the number of the inode which - will represent the ext4 file system's journal file. - journal_dev=devnum When the external journal device's major/minor numbers have changed, this option allows the user to specify the new journal location. The journal device is diff --git a/fs/ext4/super.c b/fs/ext4/super.c index e5ab520724da..8036392b2121 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -51,8 +51,6 @@ struct proc_dir_entry *ext4_proc_root; static int ext4_load_journal(struct super_block *, struct ext4_super_block *, unsigned long journal_devnum); -static int ext4_create_journal(struct super_block *, struct ext4_super_block *, - unsigned int); static void ext4_commit_super(struct super_block *sb, struct ext4_super_block *es, int sync); static void ext4_mark_recovery_complete(struct super_block *sb, @@ -1006,7 +1004,7 @@ enum { Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl, Opt_reservation, Opt_noreservation, Opt_noload, Opt_nobh, Opt_bh, Opt_commit, Opt_min_batch_time, Opt_max_batch_time, - Opt_journal_update, Opt_journal_inum, Opt_journal_dev, + Opt_journal_update, Opt_journal_dev, Opt_journal_checksum, Opt_journal_async_commit, Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback, Opt_data_err_abort, Opt_data_err_ignore, @@ -1048,7 +1046,6 @@ static const match_table_t tokens = { {Opt_min_batch_time, "min_batch_time=%u"}, {Opt_max_batch_time, "max_batch_time=%u"}, {Opt_journal_update, "journal=update"}, - {Opt_journal_inum, "journal=%u"}, {Opt_journal_dev, "journal_dev=%u"}, {Opt_journal_checksum, "journal_checksum"}, {Opt_journal_async_commit, "journal_async_commit"}, @@ -1102,7 +1099,7 @@ static ext4_fsblk_t get_sb_block(void **data) } static int parse_options(char *options, struct super_block *sb, - unsigned int *inum, unsigned long *journal_devnum, + unsigned long *journal_devnum, ext4_fsblk_t *n_blocks_count, int is_remount) { struct ext4_sb_info *sbi = EXT4_SB(sb); @@ -1226,16 +1223,6 @@ static int parse_options(char *options, struct super_block *sb, } set_opt(sbi->s_mount_opt, UPDATE_JOURNAL); break; - case Opt_journal_inum: - if (is_remount) { - printk(KERN_ERR "EXT4-fs: cannot specify " - "journal on remount\n"); - return 0; - } - if (match_int(&args[0], &option)) - return 0; - *inum = option; - break; case Opt_journal_dev: if (is_remount) { printk(KERN_ERR "EXT4-fs: cannot specify " @@ -2035,7 +2022,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) ext4_fsblk_t sb_block = get_sb_block(&data); ext4_fsblk_t logical_sb_block; unsigned long offset = 0; - unsigned int journal_inum = 0; unsigned long journal_devnum = 0; unsigned long def_mount_opts; struct inode *root; @@ -2155,8 +2141,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) set_opt(sbi->s_mount_opt, DELALLOC); - if (!parse_options((char *) data, sb, &journal_inum, &journal_devnum, - NULL, 0)) + if (!parse_options((char *) data, sb, &journal_devnum, NULL, 0)) goto failed_mount; sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | @@ -2460,9 +2445,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) goto failed_mount4; } } - } else if (journal_inum) { - if (ext4_create_journal(sb, es, journal_inum)) - goto failed_mount3; } else if (test_opt(sb, NOLOAD) && !(sb->s_flags & MS_RDONLY) && EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER)) { printk(KERN_ERR "EXT4-fs: required journal recovery " @@ -2926,48 +2908,6 @@ static int ext4_load_journal(struct super_block *sb, return 0; } -static int ext4_create_journal(struct super_block *sb, - struct ext4_super_block *es, - unsigned int journal_inum) -{ - journal_t *journal; - int err; - - if (sb->s_flags & MS_RDONLY) { - printk(KERN_ERR "EXT4-fs: readonly filesystem when trying to " - "create journal.\n"); - return -EROFS; - } - - journal = ext4_get_journal(sb, journal_inum); - if (!journal) - return -EINVAL; - - printk(KERN_INFO "EXT4-fs: creating new journal on inode %u\n", - journal_inum); - - err = jbd2_journal_create(journal); - if (err) { - printk(KERN_ERR "EXT4-fs: error creating journal.\n"); - jbd2_journal_destroy(journal); - return -EIO; - } - - EXT4_SB(sb)->s_journal = journal; - - ext4_update_dynamic_rev(sb); - EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); - EXT4_SET_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL); - - es->s_journal_inum = cpu_to_le32(journal_inum); - sb->s_dirt = 1; - - /* Make sure we flush the recovery flag to disk. */ - ext4_commit_super(sb, es, 1); - - return 0; -} - static void ext4_commit_super(struct super_block *sb, struct ext4_super_block *es, int sync) { @@ -3209,7 +3149,7 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data) /* * Allow the "check" option to be passed as a remount option. */ - if (!parse_options(data, sb, NULL, NULL, &n_blocks_count, 1)) { + if (!parse_options(data, sb, NULL, &n_blocks_count, 1)) { err = -EINVAL; goto restore_opts; } diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 34ef98057202..b10d7283ba5b 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -66,7 +66,6 @@ EXPORT_SYMBOL(jbd2_journal_update_format); EXPORT_SYMBOL(jbd2_journal_check_used_features); EXPORT_SYMBOL(jbd2_journal_check_available_features); EXPORT_SYMBOL(jbd2_journal_set_features); -EXPORT_SYMBOL(jbd2_journal_create); EXPORT_SYMBOL(jbd2_journal_load); EXPORT_SYMBOL(jbd2_journal_destroy); EXPORT_SYMBOL(jbd2_journal_abort); @@ -1162,77 +1161,6 @@ static int journal_reset(journal_t *journal) return jbd2_journal_start_thread(journal); } -/** - * int jbd2_journal_create() - Initialise the new journal file - * @journal: Journal to create. This structure must have been initialised - * - * Given a journal_t structure which tells us which disk blocks we can - * use, create a new journal superblock and initialise all of the - * journal fields from scratch. - **/ -int jbd2_journal_create(journal_t *journal) -{ - unsigned long long blocknr; - struct buffer_head *bh; - journal_superblock_t *sb; - int i, err; - - if (journal->j_maxlen < JBD2_MIN_JOURNAL_BLOCKS) { - printk (KERN_ERR "Journal length (%d blocks) too short.\n", - journal->j_maxlen); - journal_fail_superblock(journal); - return -EINVAL; - } - - if (journal->j_inode == NULL) { - /* - * We don't know what block to start at! - */ - printk(KERN_EMERG - "%s: creation of journal on external device!\n", - __func__); - BUG(); - } - - /* Zero out the entire journal on disk. We cannot afford to - have any blocks on disk beginning with JBD2_MAGIC_NUMBER. */ - jbd_debug(1, "JBD: Zeroing out journal blocks...\n"); - for (i = 0; i < journal->j_maxlen; i++) { - err = jbd2_journal_bmap(journal, i, &blocknr); - if (err) - return err; - bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize); - lock_buffer(bh); - memset (bh->b_data, 0, journal->j_blocksize); - BUFFER_TRACE(bh, "marking dirty"); - mark_buffer_dirty(bh); - BUFFER_TRACE(bh, "marking uptodate"); - set_buffer_uptodate(bh); - unlock_buffer(bh); - __brelse(bh); - } - - sync_blockdev(journal->j_dev); - jbd_debug(1, "JBD: journal cleared.\n"); - - /* OK, fill in the initial static fields in the new superblock */ - sb = journal->j_superblock; - - sb->s_header.h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); - sb->s_header.h_blocktype = cpu_to_be32(JBD2_SUPERBLOCK_V2); - - sb->s_blocksize = cpu_to_be32(journal->j_blocksize); - sb->s_maxlen = cpu_to_be32(journal->j_maxlen); - sb->s_first = cpu_to_be32(1); - - journal->j_transaction_sequence = 1; - - journal->j_flags &= ~JBD2_ABORT; - journal->j_format_version = 2; - - return journal_reset(journal); -} - /** * void jbd2_journal_update_superblock() - Update journal sb on disk. * @journal: The journal to update. diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 9d82084a1605..adef1c9940d3 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1104,7 +1104,6 @@ extern int jbd2_journal_set_features (journal_t *, unsigned long, unsigned long, unsigned long); extern void jbd2_journal_clear_features (journal_t *, unsigned long, unsigned long, unsigned long); -extern int jbd2_journal_create (journal_t *); extern int jbd2_journal_load (journal_t *journal); extern int jbd2_journal_destroy (journal_t *); extern int jbd2_journal_recover (journal_t *journal); -- cgit v1.2.3 From 14eaddc967b16017d4a1a24d2be6c28ecbe06ed8 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 31 Dec 2008 15:15:42 +0000 Subject: CRED: Fix regression in cap_capable() as shown up by sys_faccessat() [ver #2] Fix a regression in cap_capable() due to: commit 5ff7711e635b32f0a1e558227d030c7e45b4a465 Author: David Howells Date: Wed Dec 31 02:52:28 2008 +0000 CRED: Differentiate objective and effective subjective credentials on a task The problem is that the above patch allows a process to have two sets of credentials, and for the most part uses the subjective credentials when accessing current's creds. There is, however, one exception: cap_capable(), and thus capable(), uses the real/objective credentials of the target task, whether or not it is the current task. Ordinarily this doesn't matter, since usually the two cred pointers in current point to the same set of creds. However, sys_faccessat() makes use of this facility to override the credentials of the calling process to make its test, without affecting the creds as seen from other processes. One of the things sys_faccessat() does is to make an adjustment to the effective capabilities mask, which cap_capable(), as it stands, then ignores. The affected capability check is in generic_permission(): if (!(mask & MAY_EXEC) || execute_ok(inode)) if (capable(CAP_DAC_OVERRIDE)) return 0; This change splits capable() from has_capability() down into the commoncap and SELinux code. The capable() security op now only deals with the current process, and uses the current process's subjective creds. A new security op - task_capable() - is introduced that can check any task's objective creds. strictly the capable() security op is superfluous with the presence of the task_capable() op, however it should be faster to call the capable() op since two fewer arguments need be passed down through the various layers. This can be tested by compiling the following program from the XFS testsuite: /* * t_access_root.c - trivial test program to show permission bug. * * Written by Michael Kerrisk - copyright ownership not pursued. * Sourced from: http://linux.derkeiler.com/Mailing-Lists/Kernel/2003-10/6030.html */ #include #include #include #include #include #include #define UID 500 #define GID 100 #define PERM 0 #define TESTPATH "/tmp/t_access" static void errExit(char *msg) { perror(msg); exit(EXIT_FAILURE); } /* errExit */ static void accessTest(char *file, int mask, char *mstr) { printf("access(%s, %s) returns %d\n", file, mstr, access(file, mask)); } /* accessTest */ int main(int argc, char *argv[]) { int fd, perm, uid, gid; char *testpath; char cmd[PATH_MAX + 20]; testpath = (argc > 1) ? argv[1] : TESTPATH; perm = (argc > 2) ? strtoul(argv[2], NULL, 8) : PERM; uid = (argc > 3) ? atoi(argv[3]) : UID; gid = (argc > 4) ? atoi(argv[4]) : GID; unlink(testpath); fd = open(testpath, O_RDWR | O_CREAT, 0); if (fd == -1) errExit("open"); if (fchown(fd, uid, gid) == -1) errExit("fchown"); if (fchmod(fd, perm) == -1) errExit("fchmod"); close(fd); snprintf(cmd, sizeof(cmd), "ls -l %s", testpath); system(cmd); if (seteuid(uid) == -1) errExit("seteuid"); accessTest(testpath, 0, "0"); accessTest(testpath, R_OK, "R_OK"); accessTest(testpath, W_OK, "W_OK"); accessTest(testpath, X_OK, "X_OK"); accessTest(testpath, R_OK | W_OK, "R_OK | W_OK"); accessTest(testpath, R_OK | X_OK, "R_OK | X_OK"); accessTest(testpath, W_OK | X_OK, "W_OK | X_OK"); accessTest(testpath, R_OK | W_OK | X_OK, "R_OK | W_OK | X_OK"); exit(EXIT_SUCCESS); } /* main */ This can be run against an Ext3 filesystem as well as against an XFS filesystem. If successful, it will show: [root@andromeda src]# ./t_access_root /tmp/xxx 0 4043 4043 ---------- 1 dhowells dhowells 0 2008-12-31 03:00 /tmp/xxx access(/tmp/xxx, 0) returns 0 access(/tmp/xxx, R_OK) returns 0 access(/tmp/xxx, W_OK) returns 0 access(/tmp/xxx, X_OK) returns -1 access(/tmp/xxx, R_OK | W_OK) returns 0 access(/tmp/xxx, R_OK | X_OK) returns -1 access(/tmp/xxx, W_OK | X_OK) returns -1 access(/tmp/xxx, R_OK | W_OK | X_OK) returns -1 If unsuccessful, it will show: [root@andromeda src]# ./t_access_root /tmp/xxx 0 4043 4043 ---------- 1 dhowells dhowells 0 2008-12-31 02:56 /tmp/xxx access(/tmp/xxx, 0) returns 0 access(/tmp/xxx, R_OK) returns -1 access(/tmp/xxx, W_OK) returns -1 access(/tmp/xxx, X_OK) returns -1 access(/tmp/xxx, R_OK | W_OK) returns -1 access(/tmp/xxx, R_OK | X_OK) returns -1 access(/tmp/xxx, W_OK | X_OK) returns -1 access(/tmp/xxx, R_OK | W_OK | X_OK) returns -1 I've also tested the fix with the SELinux and syscalls LTP testsuites. Signed-off-by: David Howells Signed-off-by: James Morris --- include/linux/capability.h | 17 ++++++++++++++-- include/linux/security.h | 49 +++++++++++++++++++++++++++++++++++++--------- kernel/capability.c | 2 +- security/capability.c | 1 + security/commoncap.c | 42 +++++++++++++++++++++++++++------------ security/root_plug.c | 1 + security/security.c | 25 +++++++++++++++++++---- security/selinux/hooks.c | 26 ++++++++++++++++++------ security/smack/smack_lsm.c | 1 + 9 files changed, 129 insertions(+), 35 deletions(-) (limited to 'include/linux') diff --git a/include/linux/capability.h b/include/linux/capability.h index e22f48c2a46f..5b8a13214451 100644 --- a/include/linux/capability.h +++ b/include/linux/capability.h @@ -529,8 +529,21 @@ extern const kernel_cap_t __cap_init_eff_set; * * Note that this does not set PF_SUPERPRIV on the task. */ -#define has_capability(t, cap) (security_capable((t), (cap)) == 0) -#define has_capability_noaudit(t, cap) (security_capable_noaudit((t), (cap)) == 0) +#define has_capability(t, cap) (security_task_capable((t), (cap)) == 0) + +/** + * has_capability_noaudit - Determine if a task has a superior capability available (unaudited) + * @t: The task in question + * @cap: The capability to be tested for + * + * Return true if the specified task has the given superior capability + * currently in effect, false if not, but don't write an audit message for the + * check. + * + * Note that this does not set PF_SUPERPRIV on the task. + */ +#define has_capability_noaudit(t, cap) \ + (security_task_capable_noaudit((t), (cap)) == 0) extern int capable(int cap); diff --git a/include/linux/security.h b/include/linux/security.h index 3416cb85e77b..76989b8bc34f 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -48,7 +48,9 @@ struct audit_krule; * These functions are in security/capability.c and are used * as the default capabilities functions */ -extern int cap_capable(struct task_struct *tsk, int cap, int audit); +extern int cap_capable(int cap, int audit); +extern int cap_task_capable(struct task_struct *tsk, const struct cred *cred, + int cap, int audit); extern int cap_settime(struct timespec *ts, struct timezone *tz); extern int cap_ptrace_may_access(struct task_struct *child, unsigned int mode); extern int cap_ptrace_traceme(struct task_struct *parent); @@ -1195,9 +1197,18 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts) * @permitted contains the permitted capability set. * Return 0 and update @new if permission is granted. * @capable: - * Check whether the @tsk process has the @cap capability. + * Check whether the current process has the @cap capability in its + * subjective/effective credentials. + * @cap contains the capability . + * @audit: Whether to write an audit message or not + * Return 0 if the capability is granted for @tsk. + * @task_capable: + * Check whether the @tsk process has the @cap capability in its + * objective/real credentials. * @tsk contains the task_struct for the process. + * @cred contains the credentials to use. * @cap contains the capability . + * @audit: Whether to write an audit message or not * Return 0 if the capability is granted for @tsk. * @acct: * Check permission before enabling or disabling process accounting. If @@ -1290,7 +1301,9 @@ struct security_operations { const kernel_cap_t *effective, const kernel_cap_t *inheritable, const kernel_cap_t *permitted); - int (*capable) (struct task_struct *tsk, int cap, int audit); + int (*capable) (int cap, int audit); + int (*task_capable) (struct task_struct *tsk, const struct cred *cred, + int cap, int audit); int (*acct) (struct file *file); int (*sysctl) (struct ctl_table *table, int op); int (*quotactl) (int cmds, int type, int id, struct super_block *sb); @@ -1556,8 +1569,9 @@ int security_capset(struct cred *new, const struct cred *old, const kernel_cap_t *effective, const kernel_cap_t *inheritable, const kernel_cap_t *permitted); -int security_capable(struct task_struct *tsk, int cap); -int security_capable_noaudit(struct task_struct *tsk, int cap); +int security_capable(int cap); +int security_task_capable(struct task_struct *tsk, int cap); +int security_task_capable_noaudit(struct task_struct *tsk, int cap); int security_acct(struct file *file); int security_sysctl(struct ctl_table *table, int op); int security_quotactl(int cmds, int type, int id, struct super_block *sb); @@ -1754,14 +1768,31 @@ static inline int security_capset(struct cred *new, return cap_capset(new, old, effective, inheritable, permitted); } -static inline int security_capable(struct task_struct *tsk, int cap) +static inline int security_capable(int cap) { - return cap_capable(tsk, cap, SECURITY_CAP_AUDIT); + return cap_capable(cap, SECURITY_CAP_AUDIT); } -static inline int security_capable_noaudit(struct task_struct *tsk, int cap) +static inline int security_task_capable(struct task_struct *tsk, int cap) { - return cap_capable(tsk, cap, SECURITY_CAP_NOAUDIT); + int ret; + + rcu_read_lock(); + ret = cap_task_capable(tsk, __task_cred(tsk), cap, SECURITY_CAP_AUDIT); + rcu_read_unlock(); + return ret; +} + +static inline +int security_task_capable_noaudit(struct task_struct *tsk, int cap) +{ + int ret; + + rcu_read_lock(); + ret = cap_task_capable(tsk, __task_cred(tsk), cap, + SECURITY_CAP_NOAUDIT); + rcu_read_unlock(); + return ret; } static inline int security_acct(struct file *file) diff --git a/kernel/capability.c b/kernel/capability.c index 36b4b4daebec..df62f53f84ac 100644 --- a/kernel/capability.c +++ b/kernel/capability.c @@ -308,7 +308,7 @@ int capable(int cap) BUG(); } - if (has_capability(current, cap)) { + if (security_capable(cap) == 0) { current->flags |= PF_SUPERPRIV; return 1; } diff --git a/security/capability.c b/security/capability.c index 2dce66fcb992..fd1493da4f8d 100644 --- a/security/capability.c +++ b/security/capability.c @@ -826,6 +826,7 @@ void security_fixup_ops(struct security_operations *ops) set_to_cap_if_null(ops, capset); set_to_cap_if_null(ops, acct); set_to_cap_if_null(ops, capable); + set_to_cap_if_null(ops, task_capable); set_to_cap_if_null(ops, quotactl); set_to_cap_if_null(ops, quota_on); set_to_cap_if_null(ops, sysctl); diff --git a/security/commoncap.c b/security/commoncap.c index 79713545cd63..7f0b2a68717d 100644 --- a/security/commoncap.c +++ b/security/commoncap.c @@ -43,28 +43,44 @@ int cap_netlink_recv(struct sk_buff *skb, int cap) EXPORT_SYMBOL(cap_netlink_recv); /** - * cap_capable - Determine whether a task has a particular effective capability - * @tsk: The task to query + * cap_capable - Determine whether current has a particular effective capability * @cap: The capability to check for * @audit: Whether to write an audit message or not * * Determine whether the nominated task has the specified capability amongst - * its effective set, returning 0 if it does, -ve if it does not. + * its effective set, returning 0 if it does, -ve if it does not. Note that + * this uses current's subjective/effective credentials. * * NOTE WELL: cap_capable() cannot be used like the kernel's capable() * function. That is, it has the reverse semantics: cap_capable() returns 0 * when a task has a capability, but the kernel's capable() returns 1 for this * case. */ -int cap_capable(struct task_struct *tsk, int cap, int audit) +int cap_capable(int cap, int audit) { - __u32 cap_raised; + return cap_raised(current_cap(), cap) ? 0 : -EPERM; +} - /* Derived from include/linux/sched.h:capable. */ - rcu_read_lock(); - cap_raised = cap_raised(__task_cred(tsk)->cap_effective, cap); - rcu_read_unlock(); - return cap_raised ? 0 : -EPERM; +/** + * cap_has_capability - Determine whether a task has a particular effective capability + * @tsk: The task to query + * @cred: The credentials to use + * @cap: The capability to check for + * @audit: Whether to write an audit message or not + * + * Determine whether the nominated task has the specified capability amongst + * its effective set, returning 0 if it does, -ve if it does not. Note that + * this uses the task's objective/real credentials. + * + * NOTE WELL: cap_has_capability() cannot be used like the kernel's + * has_capability() function. That is, it has the reverse semantics: + * cap_has_capability() returns 0 when a task has a capability, but the + * kernel's has_capability() returns 1 for this case. + */ +int cap_task_capable(struct task_struct *tsk, const struct cred *cred, int cap, + int audit) +{ + return cap_raised(cred->cap_effective, cap) ? 0 : -EPERM; } /** @@ -160,7 +176,7 @@ static inline int cap_inh_is_capped(void) /* they are so limited unless the current task has the CAP_SETPCAP * capability */ - if (cap_capable(current, CAP_SETPCAP, SECURITY_CAP_AUDIT) == 0) + if (cap_capable(CAP_SETPCAP, SECURITY_CAP_AUDIT) == 0) return 0; #endif return 1; @@ -869,7 +885,7 @@ int cap_task_prctl(int option, unsigned long arg2, unsigned long arg3, & (new->securebits ^ arg2)) /*[1]*/ || ((new->securebits & SECURE_ALL_LOCKS & ~arg2)) /*[2]*/ || (arg2 & ~(SECURE_ALL_LOCKS | SECURE_ALL_BITS)) /*[3]*/ - || (cap_capable(current, CAP_SETPCAP, SECURITY_CAP_AUDIT) != 0) /*[4]*/ + || (cap_capable(CAP_SETPCAP, SECURITY_CAP_AUDIT) != 0) /*[4]*/ /* * [1] no changing of bits that are locked * [2] no unlocking of locks @@ -950,7 +966,7 @@ int cap_vm_enough_memory(struct mm_struct *mm, long pages) { int cap_sys_admin = 0; - if (cap_capable(current, CAP_SYS_ADMIN, SECURITY_CAP_NOAUDIT) == 0) + if (cap_capable(CAP_SYS_ADMIN, SECURITY_CAP_NOAUDIT) == 0) cap_sys_admin = 1; return __vm_enough_memory(mm, pages, cap_sys_admin); } diff --git a/security/root_plug.c b/security/root_plug.c index 40fb4f15e27b..559578f8ac66 100644 --- a/security/root_plug.c +++ b/security/root_plug.c @@ -77,6 +77,7 @@ static struct security_operations rootplug_security_ops = { .capget = cap_capget, .capset = cap_capset, .capable = cap_capable, + .task_capable = cap_task_capable, .bprm_set_creds = cap_bprm_set_creds, diff --git a/security/security.c b/security/security.c index d85dbb37c972..9bbc8e57b8c6 100644 --- a/security/security.c +++ b/security/security.c @@ -154,14 +154,31 @@ int security_capset(struct cred *new, const struct cred *old, effective, inheritable, permitted); } -int security_capable(struct task_struct *tsk, int cap) +int security_capable(int cap) { - return security_ops->capable(tsk, cap, SECURITY_CAP_AUDIT); + return security_ops->capable(cap, SECURITY_CAP_AUDIT); } -int security_capable_noaudit(struct task_struct *tsk, int cap) +int security_task_capable(struct task_struct *tsk, int cap) { - return security_ops->capable(tsk, cap, SECURITY_CAP_NOAUDIT); + const struct cred *cred; + int ret; + + cred = get_task_cred(tsk); + ret = security_ops->task_capable(tsk, cred, cap, SECURITY_CAP_AUDIT); + put_cred(cred); + return ret; +} + +int security_task_capable_noaudit(struct task_struct *tsk, int cap) +{ + const struct cred *cred; + int ret; + + cred = get_task_cred(tsk); + ret = security_ops->task_capable(tsk, cred, cap, SECURITY_CAP_NOAUDIT); + put_cred(cred); + return ret; } int security_acct(struct file *file) diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index df30a7555d8a..eb6c45107a05 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -1433,12 +1433,13 @@ static int current_has_perm(const struct task_struct *tsk, /* Check whether a task is allowed to use a capability. */ static int task_has_capability(struct task_struct *tsk, + const struct cred *cred, int cap, int audit) { struct avc_audit_data ad; struct av_decision avd; u16 sclass; - u32 sid = task_sid(tsk); + u32 sid = cred_sid(cred); u32 av = CAP_TO_MASK(cap); int rc; @@ -1865,15 +1866,27 @@ static int selinux_capset(struct cred *new, const struct cred *old, return cred_has_perm(old, new, PROCESS__SETCAP); } -static int selinux_capable(struct task_struct *tsk, int cap, int audit) +static int selinux_capable(int cap, int audit) +{ + int rc; + + rc = secondary_ops->capable(cap, audit); + if (rc) + return rc; + + return task_has_capability(current, current_cred(), cap, audit); +} + +static int selinux_task_capable(struct task_struct *tsk, + const struct cred *cred, int cap, int audit) { int rc; - rc = secondary_ops->capable(tsk, cap, audit); + rc = secondary_ops->task_capable(tsk, cred, cap, audit); if (rc) return rc; - return task_has_capability(tsk, cap, audit); + return task_has_capability(tsk, cred, cap, audit); } static int selinux_sysctl_get_sid(ctl_table *table, u16 tclass, u32 *sid) @@ -2037,7 +2050,7 @@ static int selinux_vm_enough_memory(struct mm_struct *mm, long pages) { int rc, cap_sys_admin = 0; - rc = selinux_capable(current, CAP_SYS_ADMIN, SECURITY_CAP_NOAUDIT); + rc = selinux_capable(CAP_SYS_ADMIN, SECURITY_CAP_NOAUDIT); if (rc == 0) cap_sys_admin = 1; @@ -2880,7 +2893,7 @@ static int selinux_inode_getsecurity(const struct inode *inode, const char *name * and lack of permission just means that we fall back to the * in-core context value, not a denial. */ - error = selinux_capable(current, CAP_MAC_ADMIN, SECURITY_CAP_NOAUDIT); + error = selinux_capable(CAP_MAC_ADMIN, SECURITY_CAP_NOAUDIT); if (!error) error = security_sid_to_context_force(isec->sid, &context, &size); @@ -5568,6 +5581,7 @@ static struct security_operations selinux_ops = { .capset = selinux_capset, .sysctl = selinux_sysctl, .capable = selinux_capable, + .task_capable = selinux_task_capable, .quotactl = selinux_quotactl, .quota_on = selinux_quota_on, .syslog = selinux_syslog, diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c index 6bfaba6177c2..7f12cc7015b6 100644 --- a/security/smack/smack_lsm.c +++ b/security/smack/smack_lsm.c @@ -2827,6 +2827,7 @@ struct security_operations smack_ops = { .capget = cap_capget, .capset = cap_capset, .capable = cap_capable, + .task_capable = cap_task_capable, .syslog = smack_syslog, .settime = cap_settime, .vm_enough_memory = cap_vm_enough_memory, -- cgit v1.2.3 From ea7d3fef4222cd98556a0b386598268d4dbf6670 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 4 Jan 2009 13:03:02 -0800 Subject: rcu: eliminate synchronize_rcu_xxx macro Impact: cleanup Expand macro into two files. The synchronize_rcu_xxx macro is quite ugly and it's only used by two callers, so expand it instead. This makes this code easier to change. Signed-off-by: Andi Kleen Signed-off-by: Paul E. McKenney Signed-off-by: Ingo Molnar --- include/linux/rcupdate.h | 12 ------------ kernel/rcupdate.c | 11 +++++++++-- kernel/rcupreempt.c | 11 ++++++++++- 3 files changed, 19 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 1168fbcea8d4..921340a7b71c 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -204,18 +204,6 @@ struct rcu_synchronize { extern void wakeme_after_rcu(struct rcu_head *head); -#define synchronize_rcu_xxx(name, func) \ -void name(void) \ -{ \ - struct rcu_synchronize rcu; \ - \ - init_completion(&rcu.completion); \ - /* Will wake me after RCU finished. */ \ - func(&rcu.head, wakeme_after_rcu); \ - /* Wait for it. */ \ - wait_for_completion(&rcu.completion); \ -} - /** * synchronize_sched - block until all CPUs have exited any non-preemptive * kernel code sequences. diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index ad63af8b2521..d92a76a881aa 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -77,8 +77,15 @@ void wakeme_after_rcu(struct rcu_head *head) * sections are delimited by rcu_read_lock() and rcu_read_unlock(), * and may be nested. */ -void synchronize_rcu(void); /* Makes kernel-doc tools happy */ -synchronize_rcu_xxx(synchronize_rcu, call_rcu) +void synchronize_rcu(void) +{ + struct rcu_synchronize rcu; + init_completion(&rcu.completion); + /* Will wake me after RCU finished. */ + call_rcu(&rcu.head, wakeme_after_rcu); + /* Wait for it. */ + wait_for_completion(&rcu.completion); +} EXPORT_SYMBOL_GPL(synchronize_rcu); static void rcu_barrier_callback(struct rcu_head *notused) diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c index f9dc8f3720f6..33cfc50781f9 100644 --- a/kernel/rcupreempt.c +++ b/kernel/rcupreempt.c @@ -1177,7 +1177,16 @@ EXPORT_SYMBOL_GPL(call_rcu_sched); * in -rt this does -not- necessarily result in all currently executing * interrupt -handlers- having completed. */ -synchronize_rcu_xxx(__synchronize_sched, call_rcu_sched) +void __synchronize_sched(void) +{ + struct rcu_synchronize rcu; + + init_completion(&rcu.completion); + /* Will wake me after RCU finished. */ + call_rcu_sched(&rcu.head, wakeme_after_rcu); + /* Wait for it. */ + wait_for_completion(&rcu.completion); +} EXPORT_SYMBOL_GPL(__synchronize_sched); /* -- cgit v1.2.3 From a6037b61c2f5fc99c57c15b26d7cfa58bbb34008 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 5 Jan 2009 11:28:22 +0100 Subject: hrtimer: fix recursion deadlock by re-introducing the softirq Impact: fix rare runtime deadlock There are a few sites that do: spin_lock_irq(&foo) hrtimer_start(&bar) __run_hrtimer(&bar) func() spin_lock(&foo) which obviously deadlocks. In order to avoid this, never call __run_hrtimer() from hrtimer_start*() context, but instead defer this to softirq context. Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- include/linux/interrupt.h | 3 ++- kernel/hrtimer.c | 60 +++++++++++++++++++++-------------------------- 2 files changed, 29 insertions(+), 34 deletions(-) (limited to 'include/linux') diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index 0702c4d7bdf0..2062833f5f7a 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -253,7 +253,8 @@ enum BLOCK_SOFTIRQ, TASKLET_SOFTIRQ, SCHED_SOFTIRQ, - RCU_SOFTIRQ, /* Preferable RCU should always be the last softirq */ + HRTIMER_SOFTIRQ, + RCU_SOFTIRQ, /* Preferable RCU should always be the last softirq */ NR_SOFTIRQS }; diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index 8010a67cead0..b68e98f4e4c1 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -634,7 +634,6 @@ static inline void hrtimer_init_timer_hres(struct hrtimer *timer) { } -static void __run_hrtimer(struct hrtimer *timer); /* * When High resolution timers are active, try to reprogram. Note, that in case @@ -646,13 +645,9 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer, struct hrtimer_clock_base *base) { if (base->cpu_base->hres_active && hrtimer_reprogram(timer, base)) { - /* - * XXX: recursion check? - * hrtimer_forward() should round up with timer granularity - * so that we never get into inf recursion here, - * it doesn't do that though - */ - __run_hrtimer(timer); + spin_unlock(&base->cpu_base->lock); + raise_softirq_irqoff(HRTIMER_SOFTIRQ); + spin_lock(&base->cpu_base->lock); return 1; } return 0; @@ -705,11 +700,6 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer, } static inline void hrtimer_init_hres(struct hrtimer_cpu_base *base) { } static inline void hrtimer_init_timer_hres(struct hrtimer *timer) { } -static inline int hrtimer_reprogram(struct hrtimer *timer, - struct hrtimer_clock_base *base) -{ - return 0; -} #endif /* CONFIG_HIGH_RES_TIMERS */ @@ -780,9 +770,11 @@ EXPORT_SYMBOL_GPL(hrtimer_forward); * * The timer is inserted in expiry order. Insertion into the * red black tree is O(log(n)). Must hold the base lock. + * + * Returns 1 when the new timer is the leftmost timer in the tree. */ -static void enqueue_hrtimer(struct hrtimer *timer, - struct hrtimer_clock_base *base, int reprogram) +static int enqueue_hrtimer(struct hrtimer *timer, + struct hrtimer_clock_base *base) { struct rb_node **link = &base->active.rb_node; struct rb_node *parent = NULL; @@ -814,20 +806,8 @@ static void enqueue_hrtimer(struct hrtimer *timer, * Insert the timer to the rbtree and check whether it * replaces the first pending timer */ - if (leftmost) { - /* - * Reprogram the clock event device. When the timer is already - * expired hrtimer_enqueue_reprogram has either called the - * callback or added it to the pending list and raised the - * softirq. - * - * This is a NOP for !HIGHRES - */ - if (reprogram && hrtimer_enqueue_reprogram(timer, base)) - return; - + if (leftmost) base->first = &timer->node; - } rb_link_node(&timer->node, parent, link); rb_insert_color(&timer->node, &base->active); @@ -836,6 +816,8 @@ static void enqueue_hrtimer(struct hrtimer *timer, * state of a possibly running callback. */ timer->state |= HRTIMER_STATE_ENQUEUED; + + return leftmost; } /* @@ -912,7 +894,7 @@ hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, unsigned long delta_n { struct hrtimer_clock_base *base, *new_base; unsigned long flags; - int ret; + int ret, leftmost; base = lock_hrtimer_base(timer, &flags); @@ -940,12 +922,16 @@ hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, unsigned long delta_n timer_stats_hrtimer_set_start_info(timer); + leftmost = enqueue_hrtimer(timer, new_base); + /* * Only allow reprogramming if the new base is on this CPU. * (it might still be on another CPU if the timer was pending) + * + * XXX send_remote_softirq() ? */ - enqueue_hrtimer(timer, new_base, - new_base->cpu_base == &__get_cpu_var(hrtimer_bases)); + if (leftmost && new_base->cpu_base == &__get_cpu_var(hrtimer_bases)) + hrtimer_enqueue_reprogram(timer, new_base); unlock_hrtimer_base(timer, &flags); @@ -1163,7 +1149,7 @@ static void __run_hrtimer(struct hrtimer *timer) */ if (restart != HRTIMER_NORESTART) { BUG_ON(timer->state != HRTIMER_STATE_CALLBACK); - enqueue_hrtimer(timer, base, 0); + enqueue_hrtimer(timer, base); } timer->state &= ~HRTIMER_STATE_CALLBACK; } @@ -1277,6 +1263,11 @@ void hrtimer_peek_ahead_timers(void) local_irq_restore(flags); } +static void run_hrtimer_softirq(struct softirq_action *h) +{ + hrtimer_peek_ahead_timers(); +} + #endif /* CONFIG_HIGH_RES_TIMERS */ /* @@ -1532,7 +1523,7 @@ static void migrate_hrtimer_list(struct hrtimer_clock_base *old_base, * is done, which will run all expired timers and re-programm * the timer device. */ - enqueue_hrtimer(timer, new_base, 0); + enqueue_hrtimer(timer, new_base); /* Clear the migration state bit */ timer->state &= ~HRTIMER_STATE_MIGRATE; @@ -1610,6 +1601,9 @@ void __init hrtimers_init(void) hrtimer_cpu_notify(&hrtimers_nb, (unsigned long)CPU_UP_PREPARE, (void *)(long)smp_processor_id()); register_cpu_notifier(&hrtimers_nb); +#ifdef CONFIG_HIGH_RES_TIMERS + open_softirq(HRTIMER_SOFTIRQ, run_hrtimer_softirq); +#endif } /** -- cgit v1.2.3 From c70f22d203fc02c805b6ed4a3483b740dc36786b Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Mon, 5 Jan 2009 19:07:50 +0800 Subject: sched: clean up arch_reinit_sched_domains() - Make arch_reinit_sched_domains() static. It was exported to be used in s390, but now rebuild_sched_domains() is used instead. - Make it return void. Signed-off-by: Li Zefan Signed-off-by: Ingo Molnar --- include/linux/sched.h | 1 - kernel/sched.c | 9 +++------ 2 files changed, 3 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 38a3f4b15394..91207df702e8 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -912,7 +912,6 @@ static inline struct cpumask *sched_domain_span(struct sched_domain *sd) extern void partition_sched_domains(int ndoms_new, struct cpumask *doms_new, struct sched_domain_attr *dattr_new); -extern int arch_reinit_sched_domains(void); /* Test a flag in parent sched domain */ static inline int test_sd_parent(struct sched_domain *sd, int flag) diff --git a/kernel/sched.c b/kernel/sched.c index 9a8e296959c1..c5019a5dcaa4 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -7987,7 +7987,7 @@ match2: } #if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) -int arch_reinit_sched_domains(void) +static void arch_reinit_sched_domains(void) { get_online_cpus(); @@ -7996,13 +7996,10 @@ int arch_reinit_sched_domains(void) rebuild_sched_domains(); put_online_cpus(); - - return 0; } static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt) { - int ret; unsigned int level = 0; if (sscanf(buf, "%u", &level) != 1) @@ -8023,9 +8020,9 @@ static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt) else sched_mc_power_savings = level; - ret = arch_reinit_sched_domains(); + arch_reinit_sched_domains(); - return ret ? ret : count; + return count; } #ifdef CONFIG_SCHED_MC -- cgit v1.2.3 From 922ab535bbe73975ce62f71ab9bf8ec9bce71c29 Mon Sep 17 00:00:00 2001 From: Alexey Korolev Date: Tue, 16 Dec 2008 18:13:58 +0000 Subject: [MTD] LPDDR QINFO records definitions There are declaraton of structures and macros definitions necessary for operations with QINFO in this patch. Signed-off-by: Alexey Korolev Acked-by: Jared Hulbert Signed-off-by: David Woodhouse --- include/linux/mtd/qinfo.h | 91 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 91 insertions(+) create mode 100644 include/linux/mtd/qinfo.h (limited to 'include/linux') diff --git a/include/linux/mtd/qinfo.h b/include/linux/mtd/qinfo.h new file mode 100644 index 000000000000..7b3d487d8b3f --- /dev/null +++ b/include/linux/mtd/qinfo.h @@ -0,0 +1,91 @@ +#ifndef __LINUX_MTD_QINFO_H +#define __LINUX_MTD_QINFO_H + +#include +#include +#include +#include +#include +#include +#include + +/* lpddr_private describes lpddr flash chip in memory map + * @ManufactId - Chip Manufacture ID + * @DevId - Chip Device ID + * @qinfo - pointer to qinfo records describing the chip + * @numchips - number of chips including virual RWW partitions + * @chipshift - Chip/partiton size 2^chipshift + * @chips - per-chip data structure + */ +struct lpddr_private { + uint16_t ManufactId; + uint16_t DevId; + struct qinfo_chip *qinfo; + int numchips; + unsigned long chipshift; + struct flchip chips[0]; +}; + +/* qinfo_query_info structure contains request information for + * each qinfo record + * @major - major number of qinfo record + * @major - minor number of qinfo record + * @id_str - descriptive string to access the record + * @desc - detailed description for the qinfo record + */ +struct qinfo_query_info { + uint8_t major; + uint8_t minor; + char *id_str; + char *desc; +}; + +/* + * qinfo_chip structure contains necessary qinfo records data + * @DevSizeShift - Device size 2^n bytes + * @BufSizeShift - Program buffer size 2^n bytes + * @TotalBlocksNum - Total number of blocks + * @UniformBlockSizeShift - Uniform block size 2^UniformBlockSizeShift bytes + * @HWPartsNum - Number of hardware partitions + * @SuspEraseSupp - Suspend erase supported + * @SingleWordProgTime - Single word program 2^SingleWordProgTime u-sec + * @ProgBufferTime - Program buffer write 2^ProgBufferTime u-sec + * @BlockEraseTime - Block erase 2^BlockEraseTime m-sec + */ +struct qinfo_chip { + /* General device info */ + uint16_t DevSizeShift; + uint16_t BufSizeShift; + /* Erase block information */ + uint16_t TotalBlocksNum; + uint16_t UniformBlockSizeShift; + /* Partition information */ + uint16_t HWPartsNum; + /* Optional features */ + uint16_t SuspEraseSupp; + /* Operation typical time */ + uint16_t SingleWordProgTime; + uint16_t ProgBufferTime; + uint16_t BlockEraseTime; +}; + +/* defines for fixup usage */ +#define LPDDR_MFR_ANY 0xffff +#define LPDDR_ID_ANY 0xffff +#define NUMONYX_MFGR_ID 0x0089 +#define R18_DEVICE_ID_1G 0x893c + +static inline map_word lpddr_build_cmd(u_long cmd, struct map_info *map) +{ + map_word val = { {0} }; + val.x[0] = cmd; + return val; +} + +#define CMD(x) lpddr_build_cmd(x, map) +#define CMDVAL(cmd) cmd.x[0] + +struct mtd_info *lpddr_cmdset(struct map_info *); + +#endif + -- cgit v1.2.3 From eb3db27507f74b99241abfa11824d8b6d92b84ef Mon Sep 17 00:00:00 2001 From: Alexey Korolev Date: Tue, 16 Dec 2008 18:15:33 +0000 Subject: [MTD] LPDDR PFOW definition LPDDR chips use PFOW window for sending commands, reading status and capabilites requesting. This pfow.h - contains definitions for PFOW window fileds, possible commands, error flags and some common macro function to avoid code duplications. Signed-off-by: Alexey Korolev Acked-by: Jared Hulbert Signed-off-by: David Woodhouse --- include/linux/mtd/pfow.h | 159 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 159 insertions(+) create mode 100644 include/linux/mtd/pfow.h (limited to 'include/linux') diff --git a/include/linux/mtd/pfow.h b/include/linux/mtd/pfow.h new file mode 100644 index 000000000000..b730d4f84655 --- /dev/null +++ b/include/linux/mtd/pfow.h @@ -0,0 +1,159 @@ +/* Primary function overlay window definitions + * and service functions used by LPDDR chips + */ +#ifndef __LINUX_MTD_PFOW_H +#define __LINUX_MTD_PFOW_H + +#include + +/* PFOW registers addressing */ +/* Address of symbol "P" */ +#define PFOW_QUERY_STRING_P 0x0000 +/* Address of symbol "F" */ +#define PFOW_QUERY_STRING_F 0x0002 +/* Address of symbol "O" */ +#define PFOW_QUERY_STRING_O 0x0004 +/* Address of symbol "W" */ +#define PFOW_QUERY_STRING_W 0x0006 +/* Identification info for LPDDR chip */ +#define PFOW_MANUFACTURER_ID 0x0020 +#define PFOW_DEVICE_ID 0x0022 +/* Address in PFOW where prog buffer can can be found */ +#define PFOW_PROGRAM_BUFFER_OFFSET 0x0040 +/* Size of program buffer in words */ +#define PFOW_PROGRAM_BUFFER_SIZE 0x0042 +/* Address command code register */ +#define PFOW_COMMAND_CODE 0x0080 +/* command data register */ +#define PFOW_COMMAND_DATA 0x0084 +/* command address register lower address bits */ +#define PFOW_COMMAND_ADDRESS_L 0x0088 +/* command address register upper address bits */ +#define PFOW_COMMAND_ADDRESS_H 0x008a +/* number of bytes to be proggrammed lower address bits */ +#define PFOW_DATA_COUNT_L 0x0090 +/* number of bytes to be proggrammed higher address bits */ +#define PFOW_DATA_COUNT_H 0x0092 +/* command execution register, the only possible value is 0x01 */ +#define PFOW_COMMAND_EXECUTE 0x00c0 +/* 0x01 should be written at this address to clear buffer */ +#define PFOW_CLEAR_PROGRAM_BUFFER 0x00c4 +/* device program/erase suspend register */ +#define PFOW_PROGRAM_ERASE_SUSPEND 0x00c8 +/* device status register */ +#define PFOW_DSR 0x00cc + +/* LPDDR memory device command codes */ +/* They are possible values of PFOW command code register */ +#define LPDDR_WORD_PROGRAM 0x0041 +#define LPDDR_BUFF_PROGRAM 0x00E9 +#define LPDDR_BLOCK_ERASE 0x0020 +#define LPDDR_LOCK_BLOCK 0x0061 +#define LPDDR_UNLOCK_BLOCK 0x0062 +#define LPDDR_READ_BLOCK_LOCK_STATUS 0x0065 +#define LPDDR_INFO_QUERY 0x0098 +#define LPDDR_READ_OTP 0x0097 +#define LPDDR_PROG_OTP 0x00C0 +#define LPDDR_RESUME 0x00D0 + +/* Defines possible value of PFOW command execution register */ +#define LPDDR_START_EXECUTION 0x0001 + +/* Defines possible value of PFOW program/erase suspend register */ +#define LPDDR_SUSPEND 0x0001 + +/* Possible values of PFOW device status register */ +/* access R - read; RC read & clearable */ +#define DSR_DPS (1<<1) /* RC; device protect status + * 0 - not protected 1 - locked */ +#define DSR_PSS (1<<2) /* R; program suspend status; + * 0-prog in progress/completed, + * 1- prog suspended */ +#define DSR_VPPS (1<<3) /* RC; 0-Vpp OK, * 1-Vpp low */ +#define DSR_PROGRAM_STATUS (1<<4) /* RC; 0-successful, 1-error */ +#define DSR_ERASE_STATUS (1<<5) /* RC; erase or blank check status; + * 0-success erase/blank check, + * 1 blank check error */ +#define DSR_ESS (1<<6) /* R; erase suspend status; + * 0-erase in progress/complete, + * 1 erase suspended */ +#define DSR_READY_STATUS (1<<7) /* R; Device status + * 0-busy, + * 1-ready */ +#define DSR_RPS (0x3<<8) /* RC; region program status + * 00 - Success, + * 01-re-program attempt in region with + * object mode data, + * 10-object mode program w attempt in + * region with control mode data + * 11-attempt to program invalid half + * with 0x41 command */ +#define DSR_AOS (1<<12) /* RC; 1- AO related failure */ +#define DSR_AVAILABLE (1<<15) /* R; Device availbility + * 1 - Device available + * 0 - not available */ + +/* The superset of all possible error bits in DSR */ +#define DSR_ERR 0x133A + +static inline void send_pfow_command(struct map_info *map, + unsigned long cmd_code, unsigned long adr, + unsigned long len, map_word *datum) +{ + int bits_per_chip = map_bankwidth(map) * 8; + int chipnum; + struct lpddr_private *lpddr = map->fldrv_priv; + chipnum = adr >> lpddr->chipshift; + + map_write(map, CMD(cmd_code), map->pfow_base + PFOW_COMMAND_CODE); + map_write(map, CMD(adr & ((1<pfow_base + PFOW_COMMAND_ADDRESS_L); + map_write(map, CMD(adr>>bits_per_chip), + map->pfow_base + PFOW_COMMAND_ADDRESS_H); + if (len) { + map_write(map, CMD(len & ((1<pfow_base + PFOW_DATA_COUNT_L); + map_write(map, CMD(len>>bits_per_chip), + map->pfow_base + PFOW_DATA_COUNT_H); + } + if (datum) + map_write(map, *datum, map->pfow_base + PFOW_COMMAND_DATA); + + /* Command execution start */ + map_write(map, CMD(LPDDR_START_EXECUTION), + map->pfow_base + PFOW_COMMAND_EXECUTE); +} + +static inline void print_drs_error(unsigned dsr) +{ + int prog_status = (dsr & DSR_RPS) >> 8; + + if (!(dsr & DSR_AVAILABLE)) + printk(KERN_NOTICE"DSR.15: (0) Device not Available\n"); + if (prog_status & 0x03) + printk(KERN_NOTICE"DSR.9,8: (11) Attempt to program invalid " + "half with 41h command\n"); + else if (prog_status & 0x02) + printk(KERN_NOTICE"DSR.9,8: (10) Object Mode Program attempt " + "in region with Control Mode data\n"); + else if (prog_status & 0x01) + printk(KERN_NOTICE"DSR.9,8: (01) Program attempt in region " + "with Object Mode data\n"); + if (!(dsr & DSR_READY_STATUS)) + printk(KERN_NOTICE"DSR.7: (0) Device is Busy\n"); + if (dsr & DSR_ESS) + printk(KERN_NOTICE"DSR.6: (1) Erase Suspended\n"); + if (dsr & DSR_ERASE_STATUS) + printk(KERN_NOTICE"DSR.5: (1) Erase/Blank check error\n"); + if (dsr & DSR_PROGRAM_STATUS) + printk(KERN_NOTICE"DSR.4: (1) Program Error\n"); + if (dsr & DSR_VPPS) + printk(KERN_NOTICE"DSR.3: (1) Vpp low detect, operation " + "aborted\n"); + if (dsr & DSR_PSS) + printk(KERN_NOTICE"DSR.2: (1) Program suspended\n"); + if (dsr & DSR_DPS) + printk(KERN_NOTICE"DSR.1: (1) Aborted Erase/Program attempt " + "on locked block\n"); +} +#endif /* __LINUX_MTD_PFOW_H */ -- cgit v1.2.3 From d13e51e747fee301b404dffcf4a7e1bdc558969b Mon Sep 17 00:00:00 2001 From: Alexey Korolev Date: Tue, 16 Dec 2008 18:21:10 +0000 Subject: [MTD] LPDDR added new pfow_base parameter We need to supply additional parameter to mapping driver and tell LPDDR drivers where PFOW window is in chip mapping. It leads to necessity of map_info structure extendoing. Signed-off-by: Alexey Korolev Acked-by: Jared Hulbert Signed-off-by: David Woodhouse --- include/linux/mtd/map.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/mtd/map.h b/include/linux/mtd/map.h index aa30244492c6..b981b8772217 100644 --- a/include/linux/mtd/map.h +++ b/include/linux/mtd/map.h @@ -223,6 +223,7 @@ struct map_info { must leave it enabled. */ void (*set_vpp)(struct map_info *, int); + unsigned long pfow_base; unsigned long map_priv_1; unsigned long map_priv_2; void *fldrv_priv; -- cgit v1.2.3 From d81408304b06a71c28417445202af9cd6673168d Mon Sep 17 00:00:00 2001 From: Alexey Korolev Date: Tue, 16 Dec 2008 18:22:39 +0000 Subject: [MTD] LPDDR extended physmap driver to support LPDDR flash Physmap is a generic map driver for different platforms and flash types. We added support of LPDDR to physmap. All changes here are related to introduction of new pfow_base parameter. This parameter is valid in case of LPDDR chips only. Signed-off-by: Alexey Korolev Acked-by: Jared Hulbert Signed-off-by: David Woodhouse --- drivers/mtd/maps/Kconfig | 4 ++-- drivers/mtd/maps/physmap.c | 8 +++++++- include/linux/mtd/physmap.h | 1 + 3 files changed, 10 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/drivers/mtd/maps/Kconfig b/drivers/mtd/maps/Kconfig index 3788a548336c..0225cbbf22de 100644 --- a/drivers/mtd/maps/Kconfig +++ b/drivers/mtd/maps/Kconfig @@ -10,8 +10,8 @@ config MTD_COMPLEX_MAPPINGS paged mappings of flash chips. config MTD_PHYSMAP - tristate "CFI Flash device in physical memory map" - depends on MTD_CFI || MTD_JEDECPROBE || MTD_ROM + tristate "Flash device in physical memory map" + depends on MTD_CFI || MTD_JEDECPROBE || MTD_ROM || MTD_LPDDR help This provides a 'mapping' driver which allows the NOR Flash and ROM driver code to communicate with chips which are mapped diff --git a/drivers/mtd/maps/physmap.c b/drivers/mtd/maps/physmap.c index d3a2acc7e9be..87743661d48e 100644 --- a/drivers/mtd/maps/physmap.c +++ b/drivers/mtd/maps/physmap.c @@ -68,7 +68,12 @@ static int physmap_flash_remove(struct platform_device *dev) return 0; } -static const char *rom_probe_types[] = { "cfi_probe", "jedec_probe", "map_rom", NULL }; +static const char *rom_probe_types[] = { + "cfi_probe", + "jedec_probe", + "qinfo_probe", + "map_rom", + NULL }; #ifdef CONFIG_MTD_PARTITIONS static const char *part_probe_types[] = { "cmdlinepart", "RedBoot", NULL }; #endif @@ -117,6 +122,7 @@ static int physmap_flash_probe(struct platform_device *dev) info->map[i].size = dev->resource[i].end - dev->resource[i].start + 1; info->map[i].bankwidth = physmap_data->width; info->map[i].set_vpp = physmap_data->set_vpp; + info->map[i].pfow_base = physmap_data->pfow_base; info->map[i].virt = devm_ioremap(&dev->dev, info->map[i].phys, info->map[i].size); diff --git a/include/linux/mtd/physmap.h b/include/linux/mtd/physmap.h index c8e63a5ee72e..76f7cabf07d3 100644 --- a/include/linux/mtd/physmap.h +++ b/include/linux/mtd/physmap.h @@ -24,6 +24,7 @@ struct physmap_flash_data { unsigned int width; void (*set_vpp)(struct map_info *, int); unsigned int nr_parts; + unsigned int pfow_base; struct mtd_partition *parts; }; -- cgit v1.2.3 From be92d7af38fb8a91f8575ab2272e00f2e51667ff Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Mon, 5 Jan 2009 14:34:42 +0100 Subject: genirq: provide irq_to_desc() to non-genirq architectures too Impact: build fix on non-genirq architectures Sam Ravnborg reported this build failure on sparc32 allmodconfig, the GPIO drivers assume the presence of irq_to_desc(): drivers/gpio/gpiolib.c: In function `gpiolib_dbg_show': drivers/gpio/gpiolib.c:1146: error: implicit declaration of function 'irq_to_desc' Add it in the !genirq case too. Reported-by: Sam Ravnborg Signed-off-by: Ingo Molnar Tested-by: Sam Ravnborg --- include/linux/irqnr.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/irqnr.h b/include/linux/irqnr.h index 5504a5c97836..86af92e9e84c 100644 --- a/include/linux/irqnr.h +++ b/include/linux/irqnr.h @@ -8,7 +8,12 @@ #ifndef CONFIG_GENERIC_HARDIRQS #include -# define nr_irqs NR_IRQS + +/* + * Wrappers for non-genirq architectures: + */ +#define nr_irqs NR_IRQS +#define irq_to_desc(irq) (&irq_desc[irq]) # define for_each_irq_desc(irq, desc) \ for (irq = 0; irq < nr_irqs; irq++) -- cgit v1.2.3 From c42aa775cc8a8ca558db0cc75979fb8e16667447 Mon Sep 17 00:00:00 2001 From: Nicolas Ferre Date: Thu, 20 Nov 2008 15:59:12 +0100 Subject: atmel-mci: move atmel-mci.h file to include/linux Needed to use the atmel-mci driver in an architecture independant maner. Signed-off-by: Nicolas Ferre Signed-off-by: Haavard Skinnemoen --- arch/avr32/boards/atngw100/setup.c | 2 +- arch/avr32/boards/atstk1000/atstk1002.c | 2 +- arch/avr32/boards/atstk1000/atstk1003.c | 2 +- arch/avr32/boards/atstk1000/atstk1004.c | 2 +- arch/avr32/boards/mimc200/setup.c | 2 +- arch/avr32/include/asm/atmel-mci.h | 39 --------------------------------- arch/avr32/mach-at32ap/at32ap700x.c | 2 +- drivers/mmc/host/atmel-mci.c | 2 +- include/linux/atmel-mci.h | 39 +++++++++++++++++++++++++++++++++ 9 files changed, 46 insertions(+), 46 deletions(-) delete mode 100644 arch/avr32/include/asm/atmel-mci.h create mode 100644 include/linux/atmel-mci.h (limited to 'include/linux') diff --git a/arch/avr32/boards/atngw100/setup.c b/arch/avr32/boards/atngw100/setup.c index 32fb9ba0fbdf..05d3722fff18 100644 --- a/arch/avr32/boards/atngw100/setup.c +++ b/arch/avr32/boards/atngw100/setup.c @@ -19,8 +19,8 @@ #include #include #include +#include -#include #include #include diff --git a/arch/avr32/boards/atstk1000/atstk1002.c b/arch/avr32/boards/atstk1000/atstk1002.c index 5c5cdf3b464f..6d94f74bc5c7 100644 --- a/arch/avr32/boards/atstk1000/atstk1002.c +++ b/arch/avr32/boards/atstk1000/atstk1002.c @@ -16,12 +16,12 @@ #include #include #include +#include #include