linux-kernel - [tip:perfcounters/core] perf_counter: per user mlock gift

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

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <tip-789f90fcf6b0b54e655740e9396c954378542c79@git.kernel.org>
Date:	Fri, 15 May 2009 14:43:24 GMT
From:	tip-bot for Peter Zijlstra <a.p.zijlstra@...llo.nl>
To:	linux-tip-commits@...r.kernel.org
Cc:	linux-kernel@...r.kernel.org, acme@...hat.com, paulus@...ba.org,
	hpa@...or.com, mingo@...hat.com, a.p.zijlstra@...llo.nl,
	tglx@...utronix.de, cjashfor@...ux.vnet.ibm.com, mingo@...e.hu
Subject: [tip:perfcounters/core] perf_counter: per user mlock gift

Commit-ID:  789f90fcf6b0b54e655740e9396c954378542c79
Gitweb:     http://git.kernel.org/tip/789f90fcf6b0b54e655740e9396c954378542c79
Author:     Peter Zijlstra <a.p.zijlstra@...llo.nl>
AuthorDate: Fri, 15 May 2009 15:19:27 +0200
Committer:  Ingo Molnar <mingo@...e.hu>
CommitDate: Fri, 15 May 2009 15:26:56 +0200

perf_counter: per user mlock gift

Instead of a per-process mlock gift for perf-counters, use a
per-user gift so that there is less of a DoS potential.

[ Impact: allow less worst-case unprivileged memory consumption ]

Signed-off-by: Peter Zijlstra <a.p.zijlstra@...llo.nl>
Cc: Paul Mackerras <paulus@...ba.org>
Cc: Corey Ashford <cjashfor@...ux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@...hat.com>
LKML-Reference: <20090515132018.496182835@...llo.nl>
Signed-off-by: Ingo Molnar <mingo@...e.hu>


---
 include/linux/sched.h |    4 ++++
 kernel/perf_counter.c |   22 +++++++++++++++-------
 2 files changed, 19 insertions(+), 7 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index d185758..ff59d12 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -674,6 +674,10 @@ struct user_struct {
 	struct work_struct work;
 #endif
 #endif
+
+#ifdef CONFIG_PERF_COUNTERS
+	atomic_long_t locked_vm;
+#endif
 };
 
 extern int uids_sysfs_init(void);
diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c
index 0173738..93f4a0e 100644
--- a/kernel/perf_counter.c
+++ b/kernel/perf_counter.c
@@ -45,7 +45,7 @@ static atomic_t nr_munmap_tracking __read_mostly;
 static atomic_t nr_comm_tracking __read_mostly;
 
 int sysctl_perf_counter_priv __read_mostly; /* do we need to be privileged */
-int sysctl_perf_counter_mlock __read_mostly = 128; /* 'free' kb per counter */
+int sysctl_perf_counter_mlock __read_mostly = 512; /* 'free' kb per user */
 
 /*
  * Lock for (sysadmin-configurable) counter reservations:
@@ -1522,6 +1522,9 @@ static void perf_mmap_close(struct vm_area_struct *vma)
 
 	if (atomic_dec_and_mutex_lock(&counter->mmap_count,
 				      &counter->mmap_mutex)) {
+		struct user_struct *user = current_user();
+
+		atomic_long_sub(counter->data->nr_pages + 1, &user->locked_vm);
 		vma->vm_mm->locked_vm -= counter->data->nr_locked;
 		perf_mmap_data_free(counter);
 		mutex_unlock(&counter->mmap_mutex);
@@ -1537,11 +1540,13 @@ static struct vm_operations_struct perf_mmap_vmops = {
 static int perf_mmap(struct file *file, struct vm_area_struct *vma)
 {
 	struct perf_counter *counter = file->private_data;
+	struct user_struct *user = current_user();
 	unsigned long vma_size;
 	unsigned long nr_pages;
+	unsigned long user_locked, user_lock_limit;
 	unsigned long locked, lock_limit;
+	long user_extra, extra;
 	int ret = 0;
-	long extra;
 
 	if (!(vma->vm_flags & VM_SHARED) || (vma->vm_flags & VM_WRITE))
 		return -EINVAL;
@@ -1569,15 +1574,17 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma)
 		goto unlock;
 	}
 
-	extra = nr_pages /* + 1 only account the data pages */;
-	extra -= sysctl_perf_counter_mlock >> (PAGE_SHIFT - 10);
-	if (extra < 0)
-		extra = 0;
+	user_extra = nr_pages + 1;
+	user_lock_limit = sysctl_perf_counter_mlock >> (PAGE_SHIFT - 10);
+	user_locked = atomic_long_read(&user->locked_vm) + user_extra;
 
-	locked = vma->vm_mm->locked_vm + extra;
+	extra = 0;
+	if (user_locked > user_lock_limit)
+		extra = user_locked - user_lock_limit;
 
 	lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur;
 	lock_limit >>= PAGE_SHIFT;
+	locked = vma->vm_mm->locked_vm + extra;
 
 	if ((locked > lock_limit) && !capable(CAP_IPC_LOCK)) {
 		ret = -EPERM;
@@ -1590,6 +1597,7 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma)
 		goto unlock;
 
 	atomic_set(&counter->mmap_count, 1);
+	atomic_long_add(user_extra, &user->locked_vm);
 	vma->vm_mm->locked_vm += extra;
 	counter->data->nr_locked = extra;
 unlock:
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/