linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Marco Elver <elver@google.com>,
	Alexander Potapenko <glider@google.com>,
	Dmitry Vyukov <dvyukov@google.com>, Jann Horn <jannh@google.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Linus Torvalds <torvalds@linux-foundation.org>
Subject: [PATCH 5.15 12/26] kfence: always use static branches to guard kfence_alloc()
Date: Wed, 10 Nov 2021 19:44:11 +0100	[thread overview]
Message-ID: <20211110182004.092421111@linuxfoundation.org> (raw)
In-Reply-To: <20211110182003.700594531@linuxfoundation.org>

From: Marco Elver <elver@google.com>

commit 07e8481d3c38f461d7b79c1d5c9afe013b162b0c upstream.

Regardless of KFENCE mode (CONFIG_KFENCE_STATIC_KEYS: either using
static keys to gate allocations, or using a simple dynamic branch),
always use a static branch to avoid the dynamic branch in kfence_alloc()
if KFENCE was disabled at boot.

For CONFIG_KFENCE_STATIC_KEYS=n, this now avoids the dynamic branch if
KFENCE was disabled at boot.

To simplify, also unifies the location where kfence_allocation_gate is
read-checked to just be inline in kfence_alloc().

Link: https://lkml.kernel.org/r/20211019102524.2807208-1-elver@google.com
Signed-off-by: Marco Elver <elver@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Jann Horn <jannh@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 include/linux/kfence.h |   21 +++++++++++----------
 mm/kfence/core.c       |   16 +++++++---------
 2 files changed, 18 insertions(+), 19 deletions(-)

--- a/include/linux/kfence.h
+++ b/include/linux/kfence.h
@@ -14,6 +14,9 @@
 
 #ifdef CONFIG_KFENCE
 
+#include <linux/atomic.h>
+#include <linux/static_key.h>
+
 /*
  * We allocate an even number of pages, as it simplifies calculations to map
  * address to metadata indices; effectively, the very first page serves as an
@@ -22,13 +25,8 @@
 #define KFENCE_POOL_SIZE ((CONFIG_KFENCE_NUM_OBJECTS + 1) * 2 * PAGE_SIZE)
 extern char *__kfence_pool;
 
-#ifdef CONFIG_KFENCE_STATIC_KEYS
-#include <linux/static_key.h>
 DECLARE_STATIC_KEY_FALSE(kfence_allocation_key);
-#else
-#include <linux/atomic.h>
 extern atomic_t kfence_allocation_gate;
-#endif
 
 /**
  * is_kfence_address() - check if an address belongs to KFENCE pool
@@ -116,13 +114,16 @@ void *__kfence_alloc(struct kmem_cache *
  */
 static __always_inline void *kfence_alloc(struct kmem_cache *s, size_t size, gfp_t flags)
 {
-#ifdef CONFIG_KFENCE_STATIC_KEYS
-	if (static_branch_unlikely(&kfence_allocation_key))
+#if defined(CONFIG_KFENCE_STATIC_KEYS) || CONFIG_KFENCE_SAMPLE_INTERVAL == 0
+	if (!static_branch_unlikely(&kfence_allocation_key))
+		return NULL;
 #else
-	if (unlikely(!atomic_read(&kfence_allocation_gate)))
+	if (!static_branch_likely(&kfence_allocation_key))
+		return NULL;
 #endif
-		return __kfence_alloc(s, size, flags);
-	return NULL;
+	if (likely(atomic_read(&kfence_allocation_gate)))
+		return NULL;
+	return __kfence_alloc(s, size, flags);
 }
 
 /**
--- a/mm/kfence/core.c
+++ b/mm/kfence/core.c
@@ -97,10 +97,11 @@ struct kfence_metadata kfence_metadata[C
 static struct list_head kfence_freelist = LIST_HEAD_INIT(kfence_freelist);
 static DEFINE_RAW_SPINLOCK(kfence_freelist_lock); /* Lock protecting freelist. */
 
-#ifdef CONFIG_KFENCE_STATIC_KEYS
-/* The static key to set up a KFENCE allocation. */
+/*
+ * The static key to set up a KFENCE allocation; or if static keys are not used
+ * to gate allocations, to avoid a load and compare if KFENCE is disabled.
+ */
 DEFINE_STATIC_KEY_FALSE(kfence_allocation_key);
-#endif
 
 /* Gates the allocation, ensuring only one succeeds in a given period. */
 atomic_t kfence_allocation_gate = ATOMIC_INIT(1);
@@ -668,6 +669,8 @@ void __init kfence_init(void)
 		return;
 	}
 
+	if (!IS_ENABLED(CONFIG_KFENCE_STATIC_KEYS))
+		static_branch_enable(&kfence_allocation_key);
 	WRITE_ONCE(kfence_enabled, true);
 	queue_delayed_work(system_unbound_wq, &kfence_timer, 0);
 	pr_info("initialized - using %lu bytes for %d objects at 0x%p-0x%p\n", KFENCE_POOL_SIZE,
@@ -752,12 +755,7 @@ void *__kfence_alloc(struct kmem_cache *
 	    (s->flags & (SLAB_CACHE_DMA | SLAB_CACHE_DMA32)))
 		return NULL;
 
-	/*
-	 * allocation_gate only needs to become non-zero, so it doesn't make
-	 * sense to continue writing to it and pay the associated contention
-	 * cost, in case we have a large number of concurrent allocations.
-	 */
-	if (atomic_read(&kfence_allocation_gate) || atomic_inc_return(&kfence_allocation_gate) > 1)
+	if (atomic_inc_return(&kfence_allocation_gate) > 1)
 		return NULL;
 #ifdef CONFIG_KFENCE_STATIC_KEYS
 	/*



  parent reply	other threads:[~2021-11-10 18:56 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-11-10 18:43 [PATCH 5.15 00/26] 5.15.2-rc1 review Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 01/26] KVM: x86: avoid warning with -Wbitwise-instead-of-logical Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 02/26] Revert "x86/kvm: fix vcpu-id indexed array sizes" Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 03/26] usb: ehci: handshake CMD_RUN instead of STS_HALT Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 04/26] usb: gadget: Mark USB_FSL_QE broken on 64-bit Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 05/26] usb: musb: Balance list entry in musb_gadget_queue Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 06/26] usb-storage: Add compatibility quirk flags for iODD 2531/2541 Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 07/26] Revert "proc/wchan: use printk format instead of lookup_symbol_name()" Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 08/26] binder: use euid from cred instead of using task Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 09/26] binder: use cred instead of task for selinux checks Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 10/26] binder: use cred instead of task for getsecid Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 11/26] binder: dont detect sender/target during buffer cleanup Greg Kroah-Hartman
2021-11-10 18:44 ` Greg Kroah-Hartman [this message]
2021-11-10 18:44 ` [PATCH 5.15 13/26] kfence: default to dynamic branch instead of static keys mode Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 14/26] btrfs: fix lzo_decompress_bio() kmap leakage Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 15/26] staging: rtl8712: fix use-after-free in rtl8712_dl_fw Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 16/26] isofs: Fix out of bound access for corrupted isofs image Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 17/26] comedi: dt9812: fix DMA buffers on stack Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 18/26] comedi: ni_usb6501: fix NULL-deref in command paths Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 19/26] comedi: vmk80xx: fix transfer-buffer overflows Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 20/26] comedi: vmk80xx: fix bulk-buffer overflow Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 21/26] comedi: vmk80xx: fix bulk and interrupt message timeouts Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 22/26] staging: r8712u: fix control-message timeout Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 23/26] staging: rtl8192u: fix control-message timeouts Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 24/26] staging: r8188eu: fix memleak in rtw_wx_set_enc_ext Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 25/26] media: staging/intel-ipu3: css: Fix wrong size comparison imgu_css_fw_init Greg Kroah-Hartman
2021-11-10 18:44 ` [PATCH 5.15 26/26] rsi: fix control-message timeout Greg Kroah-Hartman
2021-11-10 23:43 ` [PATCH 5.15 00/26] 5.15.2-rc1 review Florian Fainelli
2021-11-11  9:57 ` Naresh Kamboju
2021-11-11 16:26 ` Shuah Khan
2021-11-11 16:37 ` Fox Chen
2021-11-12  0:59 ` Guenter Roeck
2021-11-12 15:46 ` Jon Hunter

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=20211110182004.092421111@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=akpm@linux-foundation.org \
    --cc=dvyukov@google.com \
    --cc=elver@google.com \
    --cc=glider@google.com \
    --cc=jannh@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=torvalds@linux-foundation.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).