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,
	Bjorn Andersson <bjorn.andersson@linaro.org>,
	Sai Prakash Ranjan <saiprakash.ranjan@codeaurora.org>,
	Stephen Boyd <sboyd@kernel.org>,
	Naresh Kamboju <naresh.kamboju@linaro.org>
Subject: [PATCH 4.19 71/71] clk: Evict unregistered clks from parent caches
Date: Mon, 24 Aug 2020 10:32:02 +0200	[thread overview]
Message-ID: <20200824082359.534322751@linuxfoundation.org> (raw)
In-Reply-To: <20200824082355.848475917@linuxfoundation.org>

From: Stephen Boyd <sboyd@kernel.org>

commit bdcf1dc253248542537a742ae1e7ccafdd03f2d3 upstream.

We leave a dangling pointer in each clk_core::parents array that has an
unregistered clk as a potential parent when that clk_core pointer is
freed by clk{_hw}_unregister(). It is impossible for the true parent of
a clk to be set with clk_set_parent() once the dangling pointer is left
in the cache because we compare parent pointers in
clk_fetch_parent_index() instead of checking for a matching clk name or
clk_hw pointer.

Before commit ede77858473a ("clk: Remove global clk traversal on fetch
parent index"), we would check clk_hw pointers, which has a higher
chance of being the same between registration and unregistration, but it
can still be allocated and freed by the clk provider. In fact, this has
been a long standing problem since commit da0f0b2c3ad2 ("clk: Correct
lookup logic in clk_fetch_parent_index()") where we stopped trying to
compare clk names and skipped over entries in the cache that weren't
NULL.

There are good (performance) reasons to not do the global tree lookup in
cases where the cache holds dangling pointers to parents that have been
unregistered. Let's take the performance hit on the uncommon
registration path instead. Loop through all the clk_core::parents arrays
when a clk is unregistered and set the entry to NULL when the parent
cache entry and clk being unregistered are the same pointer. This will
fix this problem and avoid the overhead for the "normal" case.

Based on a patch by Bjorn Andersson.

Fixes: da0f0b2c3ad2 ("clk: Correct lookup logic in clk_fetch_parent_index()")
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Tested-by: Sai Prakash Ranjan <saiprakash.ranjan@codeaurora.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Link: https://lkml.kernel.org/r/20190828181959.204401-1-sboyd@kernel.org
Tested-by: Naresh Kamboju <naresh.kamboju@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/clk/clk.c |   52 +++++++++++++++++++++++++++++++++++++++++-----------
 1 file changed, 41 insertions(+), 11 deletions(-)

--- a/drivers/clk/clk.c
+++ b/drivers/clk/clk.c
@@ -40,6 +40,17 @@ static HLIST_HEAD(clk_root_list);
 static HLIST_HEAD(clk_orphan_list);
 static LIST_HEAD(clk_notifier_list);
 
+static struct hlist_head *all_lists[] = {
+	&clk_root_list,
+	&clk_orphan_list,
+	NULL,
+};
+
+static struct hlist_head *orphan_list[] = {
+	&clk_orphan_list,
+	NULL,
+};
+
 /***    private data structures    ***/
 
 struct clk_core {
@@ -2618,17 +2629,6 @@ static int inited = 0;
 static DEFINE_MUTEX(clk_debug_lock);
 static HLIST_HEAD(clk_debug_list);
 
-static struct hlist_head *all_lists[] = {
-	&clk_root_list,
-	&clk_orphan_list,
-	NULL,
-};
-
-static struct hlist_head *orphan_list[] = {
-	&clk_orphan_list,
-	NULL,
-};
-
 static void clk_summary_show_one(struct seq_file *s, struct clk_core *c,
 				 int level)
 {
@@ -3328,6 +3328,34 @@ static const struct clk_ops clk_nodrv_op
 	.set_parent	= clk_nodrv_set_parent,
 };
 
+static void clk_core_evict_parent_cache_subtree(struct clk_core *root,
+						struct clk_core *target)
+{
+	int i;
+	struct clk_core *child;
+
+	for (i = 0; i < root->num_parents; i++)
+		if (root->parents[i] == target)
+			root->parents[i] = NULL;
+
+	hlist_for_each_entry(child, &root->children, child_node)
+		clk_core_evict_parent_cache_subtree(child, target);
+}
+
+/* Remove this clk from all parent caches */
+static void clk_core_evict_parent_cache(struct clk_core *core)
+{
+	struct hlist_head **lists;
+	struct clk_core *root;
+
+	lockdep_assert_held(&prepare_lock);
+
+	for (lists = all_lists; *lists; lists++)
+		hlist_for_each_entry(root, *lists, child_node)
+			clk_core_evict_parent_cache_subtree(root, core);
+
+}
+
 /**
  * clk_unregister - unregister a currently registered clock
  * @clk: clock to unregister
@@ -3366,6 +3394,8 @@ void clk_unregister(struct clk *clk)
 			clk_core_set_parent_nolock(child, NULL);
 	}
 
+	clk_core_evict_parent_cache(clk->core);
+
 	hlist_del_init(&clk->core->child_node);
 
 	if (clk->core->prepare_count)



  parent reply	other threads:[~2020-08-24  8:58 UTC|newest]

Thread overview: 79+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-08-24  8:30 [PATCH 4.19 00/71] 4.19.142-rc1 review Greg Kroah-Hartman
2020-08-24  8:30 ` [PATCH 4.19 01/71] drm/vgem: Replace opencoded version of drm_gem_dumb_map_offset() Greg Kroah-Hartman
2020-08-24  8:30 ` [PATCH 4.19 02/71] perf probe: Fix memory leakage when the probe point is not found Greg Kroah-Hartman
2020-08-24  8:30 ` [PATCH 4.19 03/71] khugepaged: khugepaged_test_exit() check mmget_still_valid() Greg Kroah-Hartman
2020-08-24  8:30 ` [PATCH 4.19 04/71] khugepaged: adjust VM_BUG_ON_MM() in __khugepaged_enter() Greg Kroah-Hartman
2020-08-24  8:30 ` [PATCH 4.19 05/71] btrfs: export helpers for subvolume name/id resolution Greg Kroah-Hartman
2020-08-24  8:30 ` [PATCH 4.19 06/71] btrfs: dont show full path of bind mounts in subvol= Greg Kroah-Hartman
2020-08-24  8:30 ` [PATCH 4.19 07/71] btrfs: Move free_pages_out label in inline extent handling branch in compress_file_range Greg Kroah-Hartman
2020-08-24  8:30 ` [PATCH 4.19 08/71] btrfs: inode: fix NULL pointer dereference if inode doesnt need compression Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 09/71] btrfs: sysfs: use NOFS for device creation Greg Kroah-Hartman
2020-08-25 18:19   ` Pavel Machek
2020-08-25 23:58     ` Sasha Levin
2020-08-24  8:31 ` [PATCH 4.19 10/71] romfs: fix uninitialized memory leak in romfs_dev_read() Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 11/71] kernel/relay.c: fix memleak on destroy relay channel Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 12/71] mm: include CMA pages in lowmem_reserve at boot Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 13/71] mm, page_alloc: fix core hung in free_pcppages_bulk() Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 14/71] ext4: fix checking of directory entry validity for inline directories Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 15/71] jbd2: add the missing unlock_buffer() in the error path of jbd2_write_superblock() Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 16/71] scsi: zfcp: Fix use-after-free in request timeout handlers Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 17/71] mm/memory.c: skip spurious TLB flush for retried page fault Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 18/71] drm/amd/display: fix pow() crashing when given base 0 Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 19/71] kthread: Do not preempt current task if it is going to call schedule() Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 20/71] spi: Prevent adding devices below an unregistering controller Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 21/71] scsi: ufs: Add DELAY_BEFORE_LPM quirk for Micron devices Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 22/71] scsi: target: tcmu: Fix crash in tcmu_flush_dcache_range on ARM Greg Kroah-Hartman
2020-08-25 19:46   ` Pavel Machek
2020-08-24  8:31 ` [PATCH 4.19 23/71] media: budget-core: Improve exception handling in budget_register() Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 24/71] rtc: goldfish: Enable interrupt in set_alarm() when necessary Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 25/71] media: vpss: clean up resources in init Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 26/71] Input: psmouse - add a newline when printing proto by sysfs Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 27/71] m68knommu: fix overwriting of bits in ColdFire V3 cache control Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 28/71] svcrdma: Fix another Receive buffer leak Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 29/71] xfs: fix inode quota reservation checks Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 30/71] jffs2: fix UAF problem Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 31/71] ceph: fix use-after-free for fsc->mdsc Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 32/71] cpufreq: intel_pstate: Fix cpuinfo_max_freq when MSR_TURBO_RATIO_LIMIT is 0 Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 33/71] scsi: libfc: Free skb in fc_disc_gpn_id_resp() for valid cases Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 34/71] virtio_ring: Avoid loop when vq is broken in virtqueue_poll Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 35/71] tools/testing/selftests/cgroup/cgroup_util.c: cg_read_strcmp: fix null pointer dereference Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 36/71] xfs: Fix UBSAN null-ptr-deref in xfs_sysfs_init Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 37/71] alpha: fix annotation of io{read,write}{16,32}be() Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 38/71] fs/signalfd.c: fix inconsistent return codes for signalfd4 Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 39/71] ext4: fix potential negative array index in do_split() Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 40/71] ext4: dont allow overlapping system zones Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 41/71] ASoC: q6routing: add dummy register read/write function Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 42/71] i40e: Set RX_ONLY mode for unicast promiscuous on VLAN Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 43/71] i40e: Fix crash during removing i40e driver Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 44/71] net: fec: correct the error path for regulator disable in probe Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 45/71] bonding: show saner speed for broadcast mode Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 46/71] bonding: fix a potential double-unregister Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 47/71] s390/runtime_instrumentation: fix storage key handling Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 48/71] s390/ptrace: " Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 49/71] ASoC: msm8916-wcd-analog: fix register Interrupt offset Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 50/71] ASoC: intel: Fix memleak in sst_media_open Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 51/71] vfio/type1: Add proper error unwind for vfio_iommu_replay() Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 52/71] kvm: x86: Toggling CR4.SMAP does not load PDPTEs in PAE mode Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 53/71] kvm: x86: Toggling CR4.PKE " Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 54/71] kconfig: qconf: do not limit the pop-up menu to the first row Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 55/71] kconfig: qconf: fix signal connection to invalid slots Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 56/71] efi: avoid error message when booting under Xen Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 57/71] Fix build error when CONFIG_ACPI is not set/enabled: Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 58/71] RDMA/bnxt_re: Do not add user qps to flushlist Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 59/71] afs: Fix NULL deref in afs_dynroot_depopulate() Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 60/71] bonding: fix active-backup failover for current ARP slave Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 61/71] net: ena: Prevent reset after device destruction Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 62/71] net: gemini: Fix missing free_netdev() in error path of gemini_ethernet_port_probe() Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 63/71] hv_netvsc: Fix the queue_mapping in netvsc_vf_xmit() Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 64/71] net: dsa: b53: check for timeout Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 65/71] powerpc/pseries: Do not initiate shutdown when system is running on UPS Greg Kroah-Hartman
2020-08-25 19:56   ` Pavel Machek
2020-08-26 11:14     ` Vasant Hegde
2020-08-24  8:31 ` [PATCH 4.19 66/71] efi: add missed destroy_workqueue when efisubsys_init fails Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 67/71] epoll: Keep a reference on files added to the check list Greg Kroah-Hartman
2020-08-24  8:31 ` [PATCH 4.19 68/71] do_epoll_ctl(): clean the failure exits up a bit Greg Kroah-Hartman
2020-08-24  8:32 ` [PATCH 4.19 69/71] mm/hugetlb: fix calculation of adjust_range_if_pmd_sharing_possible Greg Kroah-Hartman
2020-08-24  8:32 ` [PATCH 4.19 70/71] xen: dont reschedule in preemption off sections Greg Kroah-Hartman
2020-08-24  8:32 ` Greg Kroah-Hartman [this message]
2020-08-26  8:07 ` [PATCH 4.19 00/71] 4.19.142-rc1 review Pavel Machek
2020-08-26  8:28   ` Greg Kroah-Hartman

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=20200824082359.534322751@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=bjorn.andersson@linaro.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=naresh.kamboju@linaro.org \
    --cc=saiprakash.ranjan@codeaurora.org \
    --cc=sboyd@kernel.org \
    --cc=stable@vger.kernel.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).