interconnect: avoid memory allocation when 'icc_bw_lock' is held
Subsystems (2)
PATCH v4 1/1 interconnect: avoid memory allocation when 'icc_bw_lock' is held
From: Gabor Juhos <[email protected]>
The 'icc_bw_lock' mutex is introduced in commit af42269c3523 ("interconnect: Fix locking for runpm vs reclaim") in order to decouple serialization of bw aggregation from codepaths that require memory allocation.
However commit d30f83d278a9 ("interconnect: core: Add dynamic id allocation support") added a devm_kasprintf() call into a path protected by the 'icc_bw_lock' which causes the following lockdep warning on machines like the Lenovo ThinkPad X13s:
======================================================
WARNING: possible circular locking dependency detected
6.16.0-rc3 #15 Not tainted
------------------------------------------------------
(udev-worker)/342 is trying to acquire lock:
ffffb973f7ec4638 (fs_reclaim){+.+.}-{0:0}, at: __kmalloc_node_track_caller_noprof+0xa0/0x3e0
but task is already holding lock:
ffffb973f7f7f0e8 (icc_bw_lock){+.+.}-{4:4}, at: icc_node_add+0x44/0x154
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #1 (icc_bw_lock){+.+.}-{4:4}:
icc_init+0x48/0x108
do_one_initcall+0x64/0x30c
kernel_init_freeable+0x27c/0x500
kernel_init+0x20/0x1d8
ret_from_fork+0x10/0x20
-> #0 (fs_reclaim){+.+.}-{0:0}:
__lock_acquire+0x136c/0x2114
lock_acquire+0x1c8/0x354
fs_reclaim_acquire+0x74/0xa8
__kmalloc_node_track_caller_noprof+0xa0/0x3e0
devm_kmalloc+0x54/0x124
devm_kvasprintf+0x74/0xd4
devm_kasprintf+0x58/0x80
icc_node_add+0xb4/0x154
qcom_osm_l3_probe+0x20c/0x314 [icc_osm_l3]
platform_probe+0x68/0xd8
really_probe+0xc0/0x38c
__driver_probe_device+0x7c/0x160
driver_probe_device+0x40/0x110
__driver_attach+0xfc/0x208
bus_for_each_dev+0x74/0xd0
driver_attach+0x24/0x30
bus_add_driver+0x110/0x234
driver_register+0x60/0x128
__platform_driver_register+0x24/0x30
osm_l3_driver_init+0x20/0x1000 [icc_osm_l3]
do_one_initcall+0x64/0x30c
do_init_module+0x58/0x23c
load_module+0x1df8/0x1f70
init_module_from_file+0x88/0xc4
idempotent_init_module+0x188/0x280
__arm64_sys_finit_module+0x6c/0xd8
invoke_syscall+0x48/0x110
el0_svc_common.constprop.0+0xc0/0xe0
do_el0_svc+0x1c/0x28
el0_svc+0x4c/0x158
el0t_64_sync_handler+0xc8/0xcc
el0t_64_sync+0x198/0x19c
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0 CPU1
---- ----
lock(icc_bw_lock);
lock(fs_reclaim);
lock(icc_bw_lock);
lock(fs_reclaim);
*** DEADLOCK ***The icc_node_add() functions is not designed to fail, and as such it should not do any memory allocation. In order to avoid this, add a new helper function for the name generation to be called by drivers which are using the new dynamic id feature.
---
As Gabor pointed out, the reason he hit this on the IPQ9574 in the first place is due to a separate bug which is being fixed here:
https://lore.kernel.org/all/[email protected]/
Note that this one (along with the IPQ fix) should go into 6.16-final since this fixes a regression in 6.16-rc1 that prevents lockdep from being used on a number of Qualcomm platforms (and which risks introducing further locking issues).
Johan
Changes in v4:
- make sure to generate a name also for pre-allocated rpmh nodes
- add icc_node_set_name() helper
- do not use devres to allocate name
- use osm-l3 lockdep splat as example in commit message
- Link to v3: https://lore.kernel.org/r/[email protected]
Changes in v3:
- move memory allocation out from the icc_node_add() function directly into the users of the dynamic id feature
- adjust commit description according to the changes
- Link to v2: https://lore.kernel.org/r/[email protected]
Changes in v2:
- move memory allocation outside of icc_lock
- issue a warning and return without modifying the node name in case of memory allocation failure, and adjust the commit description
- remove offered tags from Johan and Bryan
Note: since I was not sure that that the added WARN_ON() is a substantial change or not, I have removed the offered tags intentionally to be on the safe side
- Link to v1: https://lore.kernel.org/r/[email protected]
drivers/interconnect/core.c | 29 +++++++++++++++++++++++---- drivers/interconnect/qcom/icc-rpmh.c | 7 ++++++- drivers/interconnect/qcom/osm-l3.c | 7 ++++++- include/linux/interconnect-provider.h | 6 ++++++ 4 files changed, 43 insertions(+), 6 deletions(-)
drivers/interconnect/core.c +25 -4
@@ -906,10 +906,35 @@ void icc_node_destroy(int id)return ;kfree (node->links); +if (node->id >=ICC_DYN_ID_START ) +kfree (node->name);kfree (node); }EXPORT_SYMBOL_GPL (icc_node_destroy); +/** +* icc_node_set_name() - set node name +* @node: node +* @provider: node provider +* @name: node name +* +* Return: 0 on success, or -ENOMEM on allocation failure +*/ +int icc_node_set_name (struct icc_node *node,const struct icc_provider *provider,const char *name) +{ +if (node->id >=ICC_DYN_ID_START ) { + node->name =kasprintf (GFP_KERNEL ,"%s@%s" , name, +dev_name (provider->dev)); +if (!node->name) +return -ENOMEM ; + }else { + node->name = name; + } + +return 0 ; +} +EXPORT_SYMBOL_GPL (icc_node_set_name); +/** * icc_link_nodes() - create link between two nodes * @src_node: source node
@@ -1038,10 +1063,6 @@ void icc_node_add(struct icc_node *node, struct icc_provider *provider) node->avg_bw = node->init_avg; node->peak_bw = node->init_peak; -if (node->id >=ICC_DYN_ID_START ) - node->name =devm_kasprintf (provider->dev,GFP_KERNEL ,"%s@%s" , - node->name,dev_name (provider->dev)); -if (node->avg_bw || node->peak_bw) {if (provider->pre_aggregate) provider->pre_aggregate (node);
drivers/interconnect/qcom/icc-rpmh.c +6 -1
@@ -293,7 +293,12 @@ int qcom_icc_rpmh_probe(struct platform_device *pdev) goto err_remove_nodes; } - node->name = qn->name; + ret =icc_node_set_name (node, provider, qn->name); +if (ret) { +icc_node_destroy (node->id); + goto err_remove_nodes; + } + node->data = qn;icc_node_add (node, provider);
drivers/interconnect/qcom/osm-l3.c +6 -1
@@ -236,7 +236,12 @@ static int qcom_osm_l3_probe(struct platform_device *pdev) goto err; } - node->name = qnodes[i]->name; + ret =icc_node_set_name (node, provider, qnodes[i]->name); +if (ret) { +icc_node_destroy (node->id); + goto err; + } +/* Cast away const and add it back in qcom_osm_l3_set() */ node->data = (void *)qnodes[i];icc_node_add (node, provider);
include/linux/interconnect-provider.h +6 -0
@@ -119,6 +119,7 @@ int icc_std_aggregate(struct icc_node *node, u32 tag, u32 avg_bw,struct icc_node *icc_node_create_dyn (void );struct icc_node *icc_node_create (int id);void icc_node_destroy (int id); +int icc_node_set_name (struct icc_node *node,const struct icc_provider *provider,const char *name);int icc_link_nodes (struct icc_node *src_node,struct icc_node **dst_node);int icc_link_create (struct icc_node *node,const int dst_id);void icc_node_add (struct icc_node *node,struct icc_provider *provider);
@@ -152,6 +153,11 @@ static inline void icc_node_destroy(int id) { } +static int icc_node_set_name (struct icc_node *node,const struct icc_provider *provider,const char *name) +{ +return -EOPNOTSUPP ; +} +static inline int icc_link_nodes (struct icc_node *src_node,struct icc_node **dst_node) {return -EOPNOTSUPP ;
Signature · 1 line
2.49.0