From a683a5b2ba23598ad343e5ec10a4ef4077497fc9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 2 Jul 2025 06:34:37 +0100 Subject: fold fs_struct->{lock,seq} into a seqlock The combination of spinlock_t lock and seqcount_spinlock_t seq in struct fs_struct is an open-coded seqlock_t (see linux/seqlock_types.h). Combine and switch to equivalent seqlock_t primitives. AFAICS, that does end up with the same sequence of underlying operations in all cases. While we are at it, get_fs_pwd() is open-coded verbatim in get_path_from_fd(); rather than applying conversion to it, replace with the call of get_fs_pwd() there. Not worth splitting the commit for that, IMO... A bit of historical background - conversion of seqlock_t to use of seqcount_spinlock_t happened several months after the same had been done to struct fs_struct; switching fs_struct to seqlock_t could've been done immediately after that, but it looks like nobody had gotten around to that until now. Signed-off-by: Al Viro Link: https://lore.kernel.org/20250702053437.GC1880847@ZenIV Acked-by: Ahmed S. Darwish Acked-by: Peter Zijlstra (Intel) Reviewed-by: Christian Brauner Signed-off-by: Christian Brauner --- fs/exec.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/exec.c') diff --git a/fs/exec.c b/fs/exec.c index 1f5fdd2e096e..871078ddb220 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1510,7 +1510,7 @@ static void check_unsafe_exec(struct linux_binprm *bprm) * state is protected by cred_guard_mutex we hold. */ n_fs = 1; - spin_lock(&p->fs->lock); + read_seqlock_excl(&p->fs->seq); rcu_read_lock(); for_other_threads(p, t) { if (t->fs == p->fs) @@ -1523,7 +1523,7 @@ static void check_unsafe_exec(struct linux_binprm *bprm) bprm->unsafe |= LSM_UNSAFE_SHARE; else p->fs->in_exec = 1; - spin_unlock(&p->fs->lock); + read_sequnlock_excl(&p->fs->seq); } static void bprm_fill_uid(struct linux_binprm *bprm, struct file *file) -- cgit v1.2.3