All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	patches@lists.linux.dev, Liam Howlett <liam.howlett@oracle.com>,
	Suren Baghdasaryan <surenb@google.com>,
	Carlos Llamas <cmllamas@google.com>
Subject: [PATCH 5.15 39/42] binder: add lockless binder_alloc_(set|get)_vma()
Date: Thu,  1 Jun 2023 14:21:26 +0100	[thread overview]
Message-ID: <20230601131938.459836957@linuxfoundation.org> (raw)
In-Reply-To: <20230601131936.699199833@linuxfoundation.org>

From: Carlos Llamas <cmllamas@google.com>

commit 0fa53349c3acba0239369ba4cd133740a408d246 upstream.

Bring back the original lockless design in binder_alloc to determine
whether the buffer setup has been completed by the ->mmap() handler.
However, this time use smp_load_acquire() and smp_store_release() to
wrap all the ordering in a single macro call.

Also, add comments to make it evident that binder uses alloc->vma to
determine when the binder_alloc has been fully initialized. In these
scenarios acquiring the mmap_lock is not required.

Fixes: a43cfc87caaf ("android: binder: stop saving a pointer to the VMA")
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: stable@vger.kernel.org
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Link: https://lore.kernel.org/r/20230502201220.1756319-3-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[cmllamas: fixed minor merge conflict in binder_alloc_set_vma()]
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/android/binder_alloc.c |   22 +++++++---------------
 1 file changed, 7 insertions(+), 15 deletions(-)

--- a/drivers/android/binder_alloc.c
+++ b/drivers/android/binder_alloc.c
@@ -309,29 +309,18 @@ err_no_vma:
 	return vma ? -ENOMEM : -ESRCH;
 }
 
-
 static inline void binder_alloc_set_vma(struct binder_alloc *alloc,
 		struct vm_area_struct *vma)
 {
-	/*
-	 * If we see alloc->vma is not NULL, buffer data structures set up
-	 * completely. Look at smp_rmb side binder_alloc_get_vma.
-	 */
-	smp_wmb();
-	alloc->vma = vma;
+	/* pairs with smp_load_acquire in binder_alloc_get_vma() */
+	smp_store_release(&alloc->vma, vma);
 }
 
 static inline struct vm_area_struct *binder_alloc_get_vma(
 		struct binder_alloc *alloc)
 {
-	struct vm_area_struct *vma = NULL;
-
-	if (alloc->vma) {
-		/* Look at description in binder_alloc_set_vma */
-		smp_rmb();
-		vma = alloc->vma;
-	}
-	return vma;
+	/* pairs with smp_store_release in binder_alloc_set_vma() */
+	return smp_load_acquire(&alloc->vma);
 }
 
 static bool debug_low_async_space_locked(struct binder_alloc *alloc, int pid)
@@ -394,6 +383,7 @@ static struct binder_buffer *binder_allo
 	size_t size, data_offsets_size;
 	int ret;
 
+	/* Check binder_alloc is fully initialized */
 	if (!binder_alloc_get_vma(alloc)) {
 		binder_alloc_debug(BINDER_DEBUG_USER_ERROR,
 				   "%d: binder_alloc_buf, no vma\n",
@@ -789,6 +779,8 @@ int binder_alloc_mmap_handler(struct bin
 	buffer->free = 1;
 	binder_insert_free_buffer(alloc, buffer);
 	alloc->free_async_space = alloc->buffer_size / 2;
+
+	/* Signal binder_alloc is fully initialized */
 	binder_alloc_set_vma(alloc, vma);
 
 	return 0;



  parent reply	other threads:[~2023-06-01 13:25 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-06-01 13:20 [PATCH 5.15 00/42] 5.15.115-rc1 review Greg Kroah-Hartman
2023-06-01 13:20 ` [PATCH 5.15 01/42] power: supply: bq27xxx: expose battery data when CI=1 Greg Kroah-Hartman
2023-06-01 13:20 ` [PATCH 5.15 02/42] power: supply: bq27xxx: Move bq27xxx_battery_update() down Greg Kroah-Hartman
2023-06-01 13:20 ` [PATCH 5.15 03/42] power: supply: bq27xxx: Ensure power_supply_changed() is called on current sign changes Greg Kroah-Hartman
2023-06-01 13:20 ` [PATCH 5.15 04/42] power: supply: bq27xxx: After charger plug in/out wait 0.5s for things to stabilize Greg Kroah-Hartman
2023-06-01 13:20 ` [PATCH 5.15 05/42] power: supply: core: Refactor power_supply_set_input_current_limit_from_supplier() Greg Kroah-Hartman
2023-06-01 13:20 ` [PATCH 5.15 06/42] power: supply: bq24190: Call power_supply_changed() after updating input current Greg Kroah-Hartman
2023-06-01 13:20 ` [PATCH 5.15 07/42] bpf: fix a memory leak in the LRU and LRU_PERCPU hash maps Greg Kroah-Hartman
2023-06-01 13:20 ` [PATCH 5.15 08/42] net/mlx5: devcom only supports 2 ports Greg Kroah-Hartman
2023-06-01 13:20 ` [PATCH 5.15 09/42] net/mlx5e: Fix deadlock in tc route query code Greg Kroah-Hartman
2023-06-01 13:20 ` [PATCH 5.15 10/42] net/mlx5: Devcom, serialize devcom registration Greg Kroah-Hartman
2023-06-01 13:20 ` [PATCH 5.15 11/42] platform/x86: ISST: PUNIT device mapping with Sub-NUMA clustering Greg Kroah-Hartman
2023-06-01 13:20 ` [PATCH 5.15 12/42] platform/x86: ISST: Remove 8 socket limit Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 13/42] net: phy: mscc: enable VSC8501/2 RGMII RX clock Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 14/42] dmaengine: at_xdmac: Move the free desc to the tail of the desc list Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 15/42] dmaengine: at_xdmac: Remove a level of indentation in at_xdmac_tasklet() Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 16/42] dmaengine: at_xdmac: disable/enable clock directly on suspend/resume Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 17/42] dmaengine: at_xdmac: do not resume channels paused by consumers Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 18/42] dmaengine: at_xdmac: restore the content of grws register Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 19/42] KVM: s390: pv: add export before import Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 20/42] KVM: s390: fix race in gmap_make_secure() Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 21/42] net: dsa: introduce helpers for iterating through ports using dp Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 22/42] net: dsa: mt7530: rework mt753[01]_setup Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 23/42] net: dsa: mt7530: split-off common parts from mt7531_setup Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 24/42] net: dsa: mt7530: fix network connectivity with multiple CPU ports Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 25/42] Bonding: add arp_missed_max option Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 26/42] bonding: fix send_peer_notif overflow Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 27/42] binder: fix UAF caused by faulty buffer cleanup Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 28/42] irqchip/mips-gic: Get rid of the reliance on irq_cpu_online() Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 29/42] irqchip/mips-gic: Use raw spinlock for gic_lock Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 30/42] net/mlx5e: Fix SQ wake logic in ptp napi_poll context Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 31/42] xdp: Allow registering memory model without rxq reference Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 32/42] net: page_pool: use in_softirq() instead Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 33/42] page_pool: fix inconsistency for page_pool_ring_[un]lock() Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 34/42] irqchip/mips-gic: Dont touch vl_map if a local interrupt is not routable Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 35/42] xdp: xdp_mem_allocator can be NULL in trace_mem_connect() Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 36/42] bluetooth: Add cmd validity checks at the start of hci_sock_ioctl() Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 37/42] Revert "binder_alloc: add missing mmap_lock calls when using the VMA" Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 38/42] Revert "android: binder: stop saving a pointer to " Greg Kroah-Hartman
2023-06-01 13:21 ` Greg Kroah-Hartman [this message]
2023-06-01 13:21 ` [PATCH 5.15 40/42] binder: fix UAF of alloc->vma in race with munmap() Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 41/42] ipv{4,6}/raw: fix output xfrm lookup wrt protocol Greg Kroah-Hartman
2023-06-01 13:21 ` [PATCH 5.15 42/42] netfilter: ctnetlink: Support offloaded conntrack entry deletion Greg Kroah-Hartman
2023-06-01 14:12 ` [PATCH 5.15 00/42] 5.15.115-rc1 review Naresh Kamboju

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230601131938.459836957@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=cmllamas@google.com \
    --cc=liam.howlett@oracle.com \
    --cc=patches@lists.linux.dev \
    --cc=stable@vger.kernel.org \
    --cc=surenb@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.