| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <20081031081509.GF4476@duck.suse.cz>
Date: Fri, 31 Oct 2008 09:15:09 +0100
From: Jan Kara <jack@...e.cz>
To: Mingming Cao <cmm@...ibm.com>
Cc: linux-ext4 <linux-ext4@...r.kernel.org>, tytso <tytso@....edu>,
linux-fsdevel <linux-fsdevel@...r.kernel.org>
Subject: Re: [RFC PATCH 1/3] quota: Add reservation support for delayed
block allocation
Hi Mingming,
I had a quick look at your patches and they look fine. I'm now leaving
on vacation for two weeks so I won't be responding for a while.
Honza
On Wed 29-10-08 17:59:15, Mingming Cao wrote:
> Delayed allocation defers the block allocation at the dirty pages
> flush-out time, doing quota charge/check at that time is too late.
> But we can't charge the quota blocks until blocks are really allocated,
> otherwise users could get overcharged after reboot from system crash.
>
> This patch adds quota reservation for delayed llocation. Quota blocks
> are reserved in memory, inode and quota won't gets dirtied until later
> block allocation time.
>
> Signed-off-by: Mingming Cao <cmm@...ibm.com>
>
> ---
> fs/dquot.c | 113 ++++++++++++++++++++++++++++++++++-------------
> include/linux/quota.h | 2
> include/linux/quotaops.h | 30 ++++++++++++
> 3 files changed, 115 insertions(+), 30 deletions(-)
>
> Index: linux-2.6.28-rc2/fs/dquot.c
> ===================================================================
> --- linux-2.6.28-rc2.orig/fs/dquot.c 2008-10-28 16:01:44.000000000 -0700
> +++ linux-2.6.28-rc2/fs/dquot.c 2008-10-28 16:34:36.000000000 -0700
> @@ -841,6 +841,11 @@ static inline void dquot_incr_space(stru
> dquot->dq_dqb.dqb_curspace += number;
> }
>
> +static inline void dquot_resv_space(struct dquot *dquot, qsize_t number)
> +{
> + dquot->dq_dqb.dqb_rsvspace += number;
> +}
> +
> static inline void dquot_decr_inodes(struct dquot *dquot, qsize_t number)
> {
> if (dquot->dq_dqb.dqb_curinodes > number)
> @@ -1069,13 +1074,18 @@ static int check_idq(struct dquot *dquot
> /* needs dq_data_lock */
> static int check_bdq(struct dquot *dquot, qsize_t space, int prealloc, char *warntype)
> {
> + qsize_t tspace;
> +
> *warntype = QUOTA_NL_NOWARN;
> if (!sb_has_quota_limits_enabled(dquot->dq_sb, dquot->dq_type) ||
> test_bit(DQ_FAKE_B, &dquot->dq_flags))
> return QUOTA_OK;
>
> + tspace = dquot->dq_dqb.dqb_curspace + dquot->dq_dqb.dqb_rsvspace
> + + space;
> +
> if (dquot->dq_dqb.dqb_bhardlimit &&
> - dquot->dq_dqb.dqb_curspace + space > dquot->dq_dqb.dqb_bhardlimit &&
> + tspace > dquot->dq_dqb.dqb_bhardlimit &&
> !ignore_hardlimit(dquot)) {
> if (!prealloc)
> *warntype = QUOTA_NL_BHARDWARN;
> @@ -1083,7 +1093,7 @@ static int check_bdq(struct dquot *dquot
> }
>
> if (dquot->dq_dqb.dqb_bsoftlimit &&
> - dquot->dq_dqb.dqb_curspace + space > dquot->dq_dqb.dqb_bsoftlimit &&
> + tspace > dquot->dq_dqb.dqb_bsoftlimit &&
> dquot->dq_dqb.dqb_btime && get_seconds() >= dquot->dq_dqb.dqb_btime &&
> !ignore_hardlimit(dquot)) {
> if (!prealloc)
> @@ -1092,7 +1102,7 @@ static int check_bdq(struct dquot *dquot
> }
>
> if (dquot->dq_dqb.dqb_bsoftlimit &&
> - dquot->dq_dqb.dqb_curspace + space > dquot->dq_dqb.dqb_bsoftlimit &&
> + tspace > dquot->dq_dqb.dqb_bsoftlimit &&
> dquot->dq_dqb.dqb_btime == 0) {
> if (!prealloc) {
> *warntype = QUOTA_NL_BSOFTWARN;
> @@ -1227,48 +1237,90 @@ void vfs_dq_drop(struct inode *inode)
> /*
> * This operation can block, but only after everything is updated
> */
> -int dquot_alloc_space(struct inode *inode, qsize_t number, int warn)
> +int __dquot_alloc_space(struct inode *inode, qsize_t number,
> + int warn, int reserve)
> {
> - int cnt, ret = NO_QUOTA;
> + int cnt, ret = QUOTA_OK;
> char warntype[MAXQUOTAS];
>
> - /* First test before acquiring mutex - solves deadlocks when we
> - * re-enter the quota code and are already holding the mutex */
> - if (IS_NOQUOTA(inode)) {
> -out_add:
> - inode_add_bytes(inode, number);
> - return QUOTA_OK;
> - }
> for (cnt = 0; cnt < MAXQUOTAS; cnt++)
> warntype[cnt] = QUOTA_NL_NOWARN;
>
> - down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
> - if (IS_NOQUOTA(inode)) { /* Now we can do reliable test... */
> - up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
> - goto out_add;
> - }
> spin_lock(&dq_data_lock);
> for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
> if (inode->i_dquot[cnt] == NODQUOT)
> continue;
> - if (check_bdq(inode->i_dquot[cnt], number, warn, warntype+cnt) == NO_QUOTA)
> - goto warn_put_all;
> + if (check_bdq(inode->i_dquot[cnt], number, warn, warntype+cnt)
> + == NO_QUOTA) {
> + ret = NO_QUOTA;
> + goto out_unlock;
> + }
> }
> for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
> if (inode->i_dquot[cnt] == NODQUOT)
> continue;
> - dquot_incr_space(inode->i_dquot[cnt], number);
> + if (reserve)
> + dquot_resv_space(inode->i_dquot[cnt], number);
> + else
> + dquot_incr_space(inode->i_dquot[cnt], number);
> }
> - inode_add_bytes(inode, number);
> - ret = QUOTA_OK;
> -warn_put_all:
> - spin_unlock(&dq_data_lock);
> - if (ret == QUOTA_OK)
> - /* Dirtify all the dquots - this can block when journalling */
> - for (cnt = 0; cnt < MAXQUOTAS; cnt++)
> - if (inode->i_dquot[cnt])
> - mark_dquot_dirty(inode->i_dquot[cnt]);
> +out_unlock:
> flush_warnings(inode->i_dquot, warntype);
> + spin_unlock(&dq_data_lock);
> + return ret;
> +}
> +
> +int dquot_alloc_space(struct inode *inode, qsize_t number, int warn)
> +{
> + int cnt, ret = QUOTA_OK;
> +
> + /*
> + * First test before acquiring mutex - solves deadlocks when we
> + * re-enter the quota code and are already holding the mutex
> + */
> + if (IS_NOQUOTA(inode)) {
> + inode_add_bytes(inode, number);
> + return ret;
> + }
> +
> + down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
> + if (IS_NOQUOTA(inode)) {
> + /* Now we can do reliable test... */
> + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
> + inode_add_bytes(inode, number);
> + return ret;
> + }
> +
> + ret = __dquot_alloc_space(inode, number, warn, 0);
> + if (ret == NO_QUOTA) {
> + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
> + return ret;
> + }
> +
> + /* Dirtify all the dquots - this can block when journalling */
> + for (cnt = 0; cnt < MAXQUOTAS; cnt++)
> + if (inode->i_dquot[cnt])
> + mark_dquot_dirty(inode->i_dquot[cnt]);
> + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
> + inode_add_bytes(inode, number);
> + return ret;
> +}
> +
> +int dquot_reserve_space(struct inode *inode, qsize_t number, int warn)
> +{
> + int ret = QUOTA_OK;
> +
> + if (IS_NOQUOTA(inode))
> + return ret;
> +
> + down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
> + if (IS_NOQUOTA(inode)) {
> + /* Now we can do reliable test... */
> + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
> + return ret;
> + }
> +
> + ret = __dquot_alloc_space(inode, number, warn, 1);
> up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
> return ret;
> }
> @@ -1958,7 +2010,7 @@ static void do_get_dqblk(struct dquot *d
> spin_lock(&dq_data_lock);
> di->dqb_bhardlimit = stoqb(dm->dqb_bhardlimit);
> di->dqb_bsoftlimit = stoqb(dm->dqb_bsoftlimit);
> - di->dqb_curspace = dm->dqb_curspace;
> + di->dqb_curspace = dm->dqb_curspace + dm->dqb_rsvspace;
> di->dqb_ihardlimit = dm->dqb_ihardlimit;
> di->dqb_isoftlimit = dm->dqb_isoftlimit;
> di->dqb_curinodes = dm->dqb_curinodes;
> @@ -2297,6 +2349,7 @@ EXPORT_SYMBOL(dquot_alloc_space);
> EXPORT_SYMBOL(dquot_alloc_inode);
> EXPORT_SYMBOL(dquot_free_space);
> EXPORT_SYMBOL(dquot_free_inode);
> +EXPORT_SYMBOL (dquot_reserve_space);
> EXPORT_SYMBOL(dquot_transfer);
> EXPORT_SYMBOL(vfs_dq_transfer);
> EXPORT_SYMBOL(vfs_dq_quota_on_remount);
> Index: linux-2.6.28-rc2/include/linux/quota.h
> ===================================================================
> --- linux-2.6.28-rc2.orig/include/linux/quota.h 2008-10-28 16:01:46.000000000 -0700
> +++ linux-2.6.28-rc2/include/linux/quota.h 2008-10-28 16:36:12.000000000 -0700
> @@ -186,6 +186,7 @@ struct mem_dqblk {
> qsize_t dqb_bhardlimit; /* absolute limit on disk blks alloc */
> qsize_t dqb_bsoftlimit; /* preferred limit on disk blks */
> qsize_t dqb_curspace; /* current used space */
> + qsize_t dqb_rsvspace; /* current reserved space for delalloc*/
> qsize_t dqb_ihardlimit; /* absolute limit on allocated inodes */
> qsize_t dqb_isoftlimit; /* preferred inode limit */
> qsize_t dqb_curinodes; /* current # allocated inodes */
> @@ -291,6 +292,7 @@ struct dquot_operations {
> int (*release_dquot) (struct dquot *); /* Quota is going to be deleted from disk */
> int (*mark_dirty) (struct dquot *); /* Dquot is marked dirty */
> int (*write_info) (struct super_block *, int); /* Write of quota "superblock" */
> + int (*reserve_space) (struct inode *, qsize_t, int); /* reserve quota for delayed block allocation */
> };
>
> /* Operations handling requests from userspace */
> Index: linux-2.6.28-rc2/include/linux/quotaops.h
> ===================================================================
> --- linux-2.6.28-rc2.orig/include/linux/quotaops.h 2008-10-28 16:01:43.000000000 -0700
> +++ linux-2.6.28-rc2/include/linux/quotaops.h 2008-10-28 16:39:38.000000000 -0700
> @@ -176,6 +176,16 @@ static inline int vfs_dq_alloc_space(str
> return ret;
> }
>
> +static inline int vfs_dq_reserve_space(struct inode *inode, qsize_t nr)
> +{
> + if (sb_any_quota_active(inode->i_sb)) {
> + /* Used space is updated in alloc_space() */
> + if (inode->i_sb->dq_op->reserve_space(inode, nr, 0) == NO_QUOTA)
> + return 1;
> + }
> + return 0;
> +}
> +
> static inline int vfs_dq_alloc_inode(struct inode *inode)
> {
> if (sb_any_quota_active(inode->i_sb)) {
> @@ -327,6 +337,11 @@ static inline int vfs_dq_alloc_space(str
> return 0;
> }
>
> +static inline int vfs_dq_reserve_space(struct inode *inode, qsize_t nr)
> +{
> + return 0;
> +}
> +
> static inline void vfs_dq_free_space_nodirty(struct inode *inode, qsize_t nr)
> {
> inode_sub_bytes(inode, nr);
> @@ -358,12 +373,19 @@ static inline int vfs_dq_alloc_block_nod
> nr << inode->i_sb->s_blocksize_bits);
> }
>
> +
> static inline int vfs_dq_alloc_block(struct inode *inode, qsize_t nr)
> {
> return vfs_dq_alloc_space(inode,
> nr << inode->i_sb->s_blocksize_bits);
> }
>
> +static inline int vfs_dq_reserve_block(struct inode *inode, qsize_t nr)
> +{
> + return vfs_dq_reserve_space(inode,
> + nr << inode->i_sb->s_blocksize_bits);
> +}
> +
> static inline void vfs_dq_free_block_nodirty(struct inode *inode, qsize_t nr)
> {
> vfs_dq_free_space_nodirty(inode, nr << inode->i_sb->s_blocksize_bits);
> @@ -392,6 +414,7 @@ static inline void vfs_dq_free_block(str
> #define DQUOT_ALLOC_BLOCK_NODIRTY(inode, nr) \
> vfs_dq_alloc_block_nodirty(inode, nr)
> #define DQUOT_ALLOC_BLOCK(inode, nr) vfs_dq_alloc_block(inode, nr)
> +#define DQUOT_RESERVE_BLOCK(inode, nr) vfs_dq_reserve_block(inode, nr)
> #define DQUOT_ALLOC_INODE(inode) vfs_dq_alloc_inode(inode)
> #define DQUOT_FREE_SPACE_NODIRTY(inode, nr) \
> vfs_dq_free_space_nodirty(inode, nr)
>
>
--
Jan Kara <jack@...e.cz>
SUSE Labs, CR
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists