[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4E0B71D5.9010204@linux.intel.com>
Date: Wed, 29 Jun 2011 11:41:25 -0700
From: Darren Hart <dvhart@...ux.intel.com>
To: Shawn Bohrer <sbohrer@...advisors.com>
CC: KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>,
peterz@...radead.org, eric.dumazet@...il.com,
david@...advisors.com, linux-kernel@...r.kernel.org,
zvonler@...advisors.com, hughd@...gle.com, tglx@...utronix.de,
mingo@...e.hu
Subject: Re: [PATCH v4] futex: Fix regression with read only mappings
On 06/29/2011 08:17 AM, Shawn Bohrer wrote:
> commit 7485d0d3758e8e6491a5c9468114e74dc050785d (futexes: Remove rw
> parameter from get_futex_key()) in 2.6.33 introduced a user-mode
> regression in that it additionally prevented futex operations on a
> region within a read only memory mapped file. For example this breaks
> workloads that have one or more reader processes doing a FUTEX_WAIT on a
> futex within a read only shared mapping, and a writer processes that has
> a writable mapping issuing the FUTEX_WAKE.
>
> This fixes the regression for futex operations that should be valid on
> RO mappings by trying a RO get_user_pages_fast() when the RW
> get_user_pages_fast() fails so as not to slow down the common path of
> writable anonymous maps and bailing when we used the RO path on
> anonymous memory.
>
> While fixing the regression this patch opens up a possible bad
> scenarios as identified by KOSAKI Motohiro:
>
> This patch also allows FUTEX_WAIT on RO private mappings which have
> the following corner case.
>
> Thread-A: call futex(FUTEX_WAIT, memory-region-A).
> get_futex_key() return inode based key.
> sleep on the key
> Thread-B: call mprotect(PROT_READ|PROT_WRITE, memory-region-A)
> Thread-B: write memory-region-A.
> COW happen. This process's memory-region-A become related
> to new COWed private (ie PageAnon=1) page.
> Thread-B: call futex(FUETX_WAKE, memory-region-A).
> get_futex_key() return mm based key.
> IOW, we fail to wake up Thread-A.
>
> This Patch is based on Peter Zijlstra's initial patch with modifications to
> only allow RO mappings for futex operations that need VERIFY_READ access.
>
> Reported-by: David Oliver <david@...advisors.com>
> Signed-off-by: Shawn Bohrer <sbohrer@...advisors.com>
> Acked-by: Peter Zijlstra <a.p.zijlstra@...llo.nl>
> ---
>
> Included Darren Hart's fix to infinite loop in kernel on zero page.
Thanks for sticking with it Shawn!
Signed-off-by: Darren Hart <dvhart@...ux.intel.com>
> kernel/futex.c | 57 ++++++++++++++++++++++++++++++++++++++++++-------------
> 1 files changed, 43 insertions(+), 14 deletions(-)
>
> diff --git a/kernel/futex.c b/kernel/futex.c
> index fe28dc2..d919cc0 100644
> --- a/kernel/futex.c
> +++ b/kernel/futex.c
> @@ -75,6 +75,7 @@ int __read_mostly futex_cmpxchg_enabled;
> #define FLAGS_SHARED 0x01
> #define FLAGS_CLOCKRT 0x02
> #define FLAGS_HAS_TIMEOUT 0x04
> +#define FLAGS_RO 0x08
>
> /*
> * Priority Inheritance state:
> @@ -216,7 +217,7 @@ static void drop_futex_key_refs(union futex_key *key)
> /**
> * get_futex_key() - Get parameters which are the keys for a futex
> * @uaddr: virtual address of the futex
> - * @fshared: 0 for a PROCESS_PRIVATE futex, 1 for PROCESS_SHARED
> + * @flags: futex flags: FLAGS_SHARED and FLAGS_RO are valid.
> * @key: address where result is stored.
> *
> * Returns a negative error code or 0
> @@ -229,12 +230,12 @@ static void drop_futex_key_refs(union futex_key *key)
> * lock_page() might sleep, the caller should not hold a spinlock.
> */
> static int
> -get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key)
> +get_futex_key(u32 __user *uaddr, unsigned int flags, union futex_key *key)
> {
> unsigned long address = (unsigned long)uaddr;
> struct mm_struct *mm = current->mm;
> struct page *page, *page_head;
> - int err;
> + int err, ro = 0;
>
> /*
> * The futex address must be "naturally" aligned.
> @@ -251,7 +252,7 @@ get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key)
> * Note : We do have to check 'uaddr' is a valid user address,
> * but access_ok() should be faster than find_vma()
> */
> - if (!fshared) {
> + if (!(flags & FLAGS_SHARED)) {
> if (unlikely(!access_ok(VERIFY_WRITE, uaddr, sizeof(u32))))
> return -EFAULT;
> key->private.mm = mm;
> @@ -262,8 +263,18 @@ get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key)
>
> again:
> err = get_user_pages_fast(address, 1, 1, &page);
> + /*
> + * If write access is not required (eg. FUTEX_WAIT), try
> + * and get read-only access.
> + */
> + if (err == -EFAULT && flags & FLAGS_RO) {
> + err = get_user_pages_fast(address, 1, 0, &page);
> + ro = 1;
> + }
> if (err < 0)
> return err;
> + else
> + err = 0;
>
> #ifdef CONFIG_TRANSPARENT_HUGEPAGE
> page_head = page;
> @@ -305,6 +316,13 @@ again:
> if (!page_head->mapping) {
> unlock_page(page_head);
> put_page(page_head);
> + /*
> + * ZERO_PAGE pages don't have a mapping. Avoid a busy loop
> + * trying to find one. RW mapping would have COW'd (and thus
> + * have a mapping) so this page is RO and won't ever change.
> + */
> + if ((page_head == ZERO_PAGE(address)))
> + return -EFAULT;
> goto again;
> }
>
> @@ -316,6 +334,15 @@ again:
> * the object not the particular process.
> */
> if (PageAnon(page_head)) {
> + /*
> + * A RO anonymous page will never change and thus doesn't make
> + * sense for futex operations.
> + */
> + if (ro) {
> + err = -EFAULT;
> + goto out;
> + }
> +
> key->both.offset |= FUT_OFF_MMSHARED; /* ref taken on mm */
> key->private.mm = mm;
> key->private.address = address;
> @@ -327,9 +354,10 @@ again:
>
> get_futex_key_refs(key);
>
> +out:
> unlock_page(page_head);
> put_page(page_head);
> - return 0;
> + return err;
> }
>
> static inline void put_futex_key(union futex_key *key)
> @@ -940,7 +968,7 @@ futex_wake(u32 __user *uaddr, unsigned int flags, int nr_wake, u32 bitset)
> if (!bitset)
> return -EINVAL;
>
> - ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &key);
> + ret = get_futex_key(uaddr, flags | FLAGS_RO, &key);
> if (unlikely(ret != 0))
> goto out;
>
> @@ -986,10 +1014,10 @@ futex_wake_op(u32 __user *uaddr1, unsigned int flags, u32 __user *uaddr2,
> int ret, op_ret;
>
> retry:
> - ret = get_futex_key(uaddr1, flags & FLAGS_SHARED, &key1);
> + ret = get_futex_key(uaddr1, flags | FLAGS_RO, &key1);
> if (unlikely(ret != 0))
> goto out;
> - ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2);
> + ret = get_futex_key(uaddr2, flags, &key2);
> if (unlikely(ret != 0))
> goto out_put_key1;
>
> @@ -1243,10 +1271,11 @@ retry:
> pi_state = NULL;
> }
>
> - ret = get_futex_key(uaddr1, flags & FLAGS_SHARED, &key1);
> + ret = get_futex_key(uaddr1, flags | FLAGS_RO, &key1);
> if (unlikely(ret != 0))
> goto out;
> - ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2);
> + ret = get_futex_key(uaddr2, requeue_pi ? flags : flags | FLAGS_RO,
> + &key2);
> if (unlikely(ret != 0))
> goto out_put_key1;
>
> @@ -1790,7 +1819,7 @@ static int futex_wait_setup(u32 __user *uaddr, u32 val, unsigned int flags,
> * while the syscall executes.
> */
> retry:
> - ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &q->key);
> + ret = get_futex_key(uaddr, flags | FLAGS_RO, &q->key);
> if (unlikely(ret != 0))
> return ret;
>
> @@ -1941,7 +1970,7 @@ static int futex_lock_pi(u32 __user *uaddr, unsigned int flags, int detect,
> }
>
> retry:
> - ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &q.key);
> + ret = get_futex_key(uaddr, flags, &q.key);
> if (unlikely(ret != 0))
> goto out;
>
> @@ -2060,7 +2089,7 @@ retry:
> if ((uval & FUTEX_TID_MASK) != vpid)
> return -EPERM;
>
> - ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &key);
> + ret = get_futex_key(uaddr, flags, &key);
> if (unlikely(ret != 0))
> goto out;
>
> @@ -2249,7 +2278,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags,
> debug_rt_mutex_init_waiter(&rt_waiter);
> rt_waiter.task = NULL;
>
> - ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2);
> + ret = get_futex_key(uaddr2, flags, &key2);
> if (unlikely(ret != 0))
> goto out;
>
--
Darren Hart
Intel Open Source Technology Center
Yocto Project - Linux Kernel
--
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/
Powered by blists - more mailing lists