WARNING: at block/blk-cgroup.c:867

Hi Tejun,

for !CONFIG_CFQ_GROUP_IOSCHED the following warning triggered on s390:

------------[ cut here ]------------
WARNING: at block/blk-cgroup.c:867
Modules linked in:
Modules linked in:
CPU: 3 Not tainted 3.4.0-09547-gfb21aff #1
Process swapper/0 (pid: 1, task: 000000003ff80000, ksp: 000000003ff7f8b8)
Krnl PSW : 0704100180000000 00000000003d76ca (blkcg_policy_register+0xca/0xe0)
           R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:0 CC:1 PM:0 EA:3
Krnl GPRS: 0000000000000000 00000000014b85ec 00000000014b85b0 0000000000000000
           000000000096fb60 0000000000000000 00000000009a8e78 0000000000000048
           000000000099c070 0000000000b6f000 0000000000000000 000000000099c0b8
           00000000014b85b0 0000000000667580 000000003ff7fd98 000000003ff7fd70
Krnl Code: 00000000003d76be: a7280001           lhi     %r2,1
           00000000003d76c2: a7f4ffdf           brc     15,3d7680
          #00000000003d76c6: a7f40001           brc     15,3d76c8
          >00000000003d76ca: a7c8ffea           lhi     %r12,-22
           00000000003d76ce: a7f4ffce           brc     15,3d766a
           00000000003d76d2: a7f40001           brc     15,3d76d4
           00000000003d76d6: a7c80000           lhi     %r12,0
           00000000003d76da: a7f4ffc2           brc     15,3d765e
Call Trace:
([<0000000000b6f000>] initcall_debug+0x0/0x4)
 [<0000000000989e8a>] cfq_init+0x62/0xd4
 [<00000000001000ba>] do_one_initcall+0x3a/0x170
 [<000000000096fb60>] kernel_init+0x214/0x2bc
 [<0000000000623202>] kernel_thread_starter+0x6/0xc
 [<00000000006231fc>] kernel_thread_starter+0x0/0xc
no locks held by swapper/0/1.
Last Breaking-Event-Address:
 [<00000000003d76c6>] blkcg_policy_register+0xc6/0xe0
---[ end trace b8ef4903fcbf9dd3 ]---

for !CONFIG_CFQ_GROUP_IOSCHED the blkcg_policy_cfq that cfq_init() tries to register with blkcg_policy_register() is in the bss section and therefore initialized with 0.

So the warning in blkcg_policy_register()

	if (WARN_ON(pol->pd_size < sizeof(struct blkg_policy_data)))
		return -EINVAL;

will trigger since pd_size == 0.

cfq_init() would return zero after kmem cache creation failure. Fix so that it returns -ENOMEM.

---
 block/cfq-iosched.c |    1 +
 1 file changed, 1 insertion(+)

@@ -4202,6 +4202,7 @@ static int __init cfq_init(void)
 	if (ret)
 		return ret;
 
+	ret = -ENOMEM;
 	cfq_pool = KMEM_CACHE(cfq_queue, 0);
 	if (!cfq_pool)
 		goto err_pol_unreg;

cfq may be built w/ or w/o blkcg support depending on CONFIG_CFQ_CGROUP_IOSCHED. If blkcg support is disabled, most of related code is ifdef'd out but some part is left dangling - blkcg_policy_cfq is left zero-filled and blkcg_policy_[un]register() calls are made on it.

Feeding zero filled policy to blkcg_policy_register() is incorrect and triggers the following WARN_ON() if CONFIG_BLK_CGROUP && !CONFIG_CFQ_GROUP_IOSCHED.

 ------------[ cut here ]------------
 WARNING: at block/blk-cgroup.c:867
 Modules linked in:
 Modules linked in:
 CPU: 3 Not tainted 3.4.0-09547-gfb21aff #1
 Process swapper/0 (pid: 1, task: 000000003ff80000, ksp: 000000003ff7f8b8)
 Krnl PSW : 0704100180000000 00000000003d76ca (blkcg_policy_register+0xca/0xe0)
	    R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:0 CC:1 PM:0 EA:3
 Krnl GPRS: 0000000000000000 00000000014b85ec 00000000014b85b0 0000000000000000
	    000000000096fb60 0000000000000000 00000000009a8e78 0000000000000048
	    000000000099c070 0000000000b6f000 0000000000000000 000000000099c0b8
	    00000000014b85b0 0000000000667580 000000003ff7fd98 000000003ff7fd70
 Krnl Code: 00000000003d76be: a7280001           lhi     %r2,1
	    00000000003d76c2: a7f4ffdf           brc     15,3d7680
	   #00000000003d76c6: a7f40001           brc     15,3d76c8
	   >00000000003d76ca: a7c8ffea           lhi     %r12,-22
	    00000000003d76ce: a7f4ffce           brc     15,3d766a
	    00000000003d76d2: a7f40001           brc     15,3d76d4
	    00000000003d76d6: a7c80000           lhi     %r12,0
	    00000000003d76da: a7f4ffc2           brc     15,3d765e
 Call Trace:
 ([<0000000000b6f000>] initcall_debug+0x0/0x4)
  [<0000000000989e8a>] cfq_init+0x62/0xd4
  [<00000000001000ba>] do_one_initcall+0x3a/0x170
  [<000000000096fb60>] kernel_init+0x214/0x2bc
  [<0000000000623202>] kernel_thread_starter+0x6/0xc
  [<00000000006231fc>] kernel_thread_starter+0x0/0xc
 no locks held by swapper/0/1.
 Last Breaking-Event-Address:
  [<00000000003d76c6>] blkcg_policy_register+0xc6/0xe0
 ---[ end trace b8ef4903fcbf9dd3 ]---

This patch fixes the problem by ensuring all blkcg support code is inside CONFIG_CFQ_GROUP_IOSCHED.

* blkcg_policy_cfq declaration and blkg_to_cfqg() definition are moved inside the first CONFIG_CFQ_GROUP_IOSCHED block. __maybe_unused is dropped from blkcg_policy_cfq decl.

* blkcg_deactivate_poilcy() invocation is moved inside ifdef. This also makes the activation logic match cfq_init_queue().

* All blkcg_policy_[un]register() invocations are moved inside ifdef.

LKML-Reference: <[email protected]>

---
 block/cfq-iosched.c |   29 +++++++++++++++++------------
 1 file changed, 17 insertions(+), 12 deletions(-)

@@ -17,8 +17,6 @@
 #include "blk.h"
 #include "blk-cgroup.h"
 
-static struct blkcg_policy blkcg_policy_cfq __maybe_unused;
-
 /*
  * tunables
  */

@@ -418,11 +416,6 @@ static inline struct cfq_group *pd_to_cf
 	return pd ? container_of(pd, struct cfq_group, pd) : NULL;
 }
 
-static inline struct cfq_group *blkg_to_cfqg(struct blkcg_gq *blkg)
-{
-	return pd_to_cfqg(blkg_to_pd(blkg, &blkcg_policy_cfq));
-}
-
 static inline struct blkcg_gq *cfqg_to_blkg(struct cfq_group *cfqg)
 {
 	return pd_to_blkg(&cfqg->pd);

@@ -572,6 +565,13 @@ static inline void cfqg_stats_update_avg
 
 #ifdef CONFIG_CFQ_GROUP_IOSCHED
 
+static struct blkcg_policy blkcg_policy_cfq;
+
+static inline struct cfq_group *blkg_to_cfqg(struct blkcg_gq *blkg)
+{
+	return pd_to_cfqg(blkg_to_pd(blkg, &blkcg_policy_cfq));
+}
+
 static inline void cfqg_get(struct cfq_group *cfqg)
 {
 	return blkg_get(cfqg_to_blkg(cfqg));

@@ -3951,10 +3951,11 @@ static void cfq_exit_queue(struct elevat
 
 	cfq_shutdown_timer_wq(cfqd);
 
-#ifndef CONFIG_CFQ_GROUP_IOSCHED
+#ifdef CONFIG_CFQ_GROUP_IOSCHED
+	blkcg_deactivate_policy(q, &blkcg_policy_cfq);
+#else
 	kfree(cfqd->root_group);
 #endif
-	blkcg_deactivate_policy(q, &blkcg_policy_cfq);
 	kfree(cfqd);
 }
 

@@ -4194,13 +4195,13 @@ static int __init cfq_init(void)
 #ifdef CONFIG_CFQ_GROUP_IOSCHED
 	if (!cfq_group_idle)
 		cfq_group_idle = 1;
-#else
-		cfq_group_idle = 0;
-#endif
 
 	ret = blkcg_policy_register(&blkcg_policy_cfq);
 	if (ret)
 		return ret;
+#else
+	cfq_group_idle = 0;
+#endif
 
 	ret = -ENOMEM;
 	cfq_pool = KMEM_CACHE(cfq_queue, 0);

@@ -4216,13 +4217,17 @@ static int __init cfq_init(void)
 err_free_pool:
 	kmem_cache_destroy(cfq_pool);
 err_pol_unreg:
+#ifdef CONFIG_CFQ_GROUP_IOSCHED
 	blkcg_policy_unregister(&blkcg_policy_cfq);
+#endif
 	return ret;
 }
 
 static void __exit cfq_exit(void)
 {
+#ifdef CONFIG_CFQ_GROUP_IOSCHED
 	blkcg_policy_unregister(&blkcg_policy_cfq);
+#endif
 	elv_unregister(&iosched_cfq);
 	kmem_cache_destroy(cfq_pool);
 }

Thanks Tejun, applied this one and 1/2 as well as a standalone bug fix.

Jens Axboe