Drivers: hv: hv_balloon
cover Drivers: hv: hv_balloon
Some bug fixes for the balloon driver.
K. Y. Srinivasan (3):
Drivers: hv: hv_balloon: Make adjustments in computing the floor
Drivers: hv: hv_balloon: Fix a locking bug in the balloon driver
Drivers: hv: hv_balloon: Don't post pressure status from interrupt
context
drivers/hv/hv_balloon.c | 89 ++++++++++++++++++++++++++++++++++++++-------- 1 files changed, 73 insertions(+), 16 deletions(-)
Signature · 1 line
1.7.4.1
PATCH 1/3 Drivers: hv: hv_balloon: Make adjustments in computing the floor
Make adjustments in computing the balloon floor.
--- drivers/hv/hv_balloon.c | 9 +++++---- 1 files changed, 5 insertions(+), 4 deletions(-)
drivers/hv/hv_balloon.c +5 -4
@@ -928,9 +928,8 @@ static unsigned long compute_balloon_floor(void)* 128 72 (1/2) * 512 168 (1/4) * 2048 360 (1/8) -* 8192 552 (1/32) -* 32768 1320 -* 131072 4392 +* 8192 768 (1/16) +* 32768 1536 (1/32) */ if (totalram_pages <MB2PAGES (128 )) min_pages =MB2PAGES (8 ) + (totalram_pages >>1 );
@@ -938,8 +937,10 @@ static unsigned long compute_balloon_floor(void) min_pages =MB2PAGES (40 ) + (totalram_pages >>2 ); else if (totalram_pages < MB2PAGES(2048)) min_pages =MB2PAGES (104 ) + (totalram_pages >>3 ); + else if (totalram_pages < MB2PAGES(8192)) + min_pages =MB2PAGES (256 ) + (totalram_pages >>4 ); else - min_pages =MB2PAGES (296 ) + (totalram_pages >>5 ); + min_pages =MB2PAGES (512 ) + (totalram_pages >>5 );#undef MB2PAGESreturn min_pages; }
Signature · 1 line
1.7.4.1
PATCH 2/3 Drivers: hv: hv_balloon: Fix a locking bug in the balloon driver
We support memory hot-add in the Hyper-V balloon driver by hot adding an appropriately sized and aligned region and controlling the on-lining of pages within that region based on the pages that the host wants us to online. We do this because the granularity and alignment requirements in Linux are different from what Windows expects. The state to manage the onlining of pages needs to be correctly protected. Fix this bug.
--- drivers/hv/hv_balloon.c | 69 +++++++++++++++++++++++++++++++++++++++++++--- 1 files changed, 64 insertions(+), 5 deletions(-)
drivers/hv/hv_balloon.c +64 -5
@@ -533,6 +533,9 @@ struct hv_dynmem_device {*/ struct task_struct *thread; +struct mutex ha_region_mutex; +struct completion waiter_event; +/* * A list of hot-add regions. */
@@ -549,7 +552,60 @@ struct hv_dynmem_device {static struct hv_dynmem_device dm_device;static void post_status (struct hv_dynmem_device *dm); + +static void acquire_region_mutex (bool trylock) +{ +if (trylock) { +reinit_completion (&dm_device.waiter_event); +while (!mutex_trylock (&dm_device.ha_region_mutex)) +wait_for_completion (&dm_device.waiter_event); + }else { +mutex_lock (&dm_device.ha_region_mutex); + } +} + +static void release_region_mutex (bool trylock) +{ +if (trylock) { +mutex_unlock (&dm_device.ha_region_mutex); + }else { +mutex_unlock (&dm_device.ha_region_mutex); +complete (&dm_device.waiter_event); + } +} + +#ifdef CONFIG_MEMORY_HOTPLUG +static int hv_memory_notifier (struct notifier_block *nb,unsigned long val, +void *v) +{ +switch (val) { +case MEM_GOING_ONLINE : +acquire_region_mutex (true); +break ; + +case MEM_ONLINE : +case MEM_CANCEL_ONLINE : +release_region_mutex (true); +if (dm_device.ha_waiting) { + dm_device.ha_waiting = false; +complete (&dm_device.ol_waitevent); + } +break ; + +case MEM_GOING_OFFLINE : +case MEM_OFFLINE : +case MEM_CANCEL_OFFLINE : +break ; + } +return NOTIFY_OK ; +} + +static struct notifier_block hv_memory_nb = { + .notifier_call = hv_memory_notifier, + .priority =0 +}; +static void hv_bring_pgs_online (unsigned long start_pfn,unsigned long size) {
@@ -591,6 +647,7 @@ static void hv_mem_hot_add(unsigned long start, unsigned long size,init_completion (&dm_device.ol_waitevent); dm_device.ha_waiting = true; +release_region_mutex (false); nid =memory_add_physaddr_to_nid (PFN_PHYS (start_pfn)); ret =add_memory (nid,PFN_PHYS ((start_pfn)), (HA_CHUNK <<PAGE_SHIFT ));
@@ -619,6 +676,7 @@ static void hv_mem_hot_add(unsigned long start, unsigned long size,* have not been "onlined" within the allowed time. */ wait_for_completion_timeout (&dm_device.ol_waitevent,5 *HZ ); +acquire_region_mutex (false);post_status (&dm_device); }
@@ -632,11 +690,6 @@ static void hv_online_page(struct page *pg)unsigned long cur_start_pgp;unsigned long cur_end_pgp; -if (dm_device.ha_waiting) { - dm_device.ha_waiting = false; -complete (&dm_device.ol_waitevent); - } -list_for_each (cur, &dm_device.ha_region_list) { has = list_entry(cur, struct hv_hotadd_state, list); cur_start_pgp = (unsigned long)
@@ -834,6 +887,7 @@ static void hot_add_req(struct work_struct *dummy) resp.hdr.size =sizeof (struct dm_hot_add_response );#ifdef CONFIG_MEMORY_HOTPLUG +acquire_region_mutex (false); pg_start = dm->ha_wrk.ha_page_range.finfo.start_page; pfn_cnt = dm->ha_wrk.ha_page_range.finfo.page_cnt;
@@ -865,6 +919,7 @@ static void hot_add_req(struct work_struct *dummy)if (do_hot_add) resp.page_count =process_hot_add (pg_start, pfn_cnt, rg_start, rg_sz); +release_region_mutex (false);#endif /* * The result field of the response structure has the
@@ -1388,7 +1443,9 @@ static int balloon_probe(struct hv_device *dev, dm_device.next_version =DYNMEM_PROTOCOL_VERSION_WIN7 ;init_completion (&dm_device.host_event);init_completion (&dm_device.config_event); +init_completion (&dm_device.waiter_event);INIT_LIST_HEAD (&dm_device.ha_region_list); +mutex_init (&dm_device.ha_region_mutex);INIT_WORK (&dm_device.balloon_wrk.wrk, balloon_up);INIT_WORK (&dm_device.ha_wrk.wrk, hot_add_req); dm_device.host_specified_ha_region = false;
@@ -1402,6 +1459,7 @@ static int balloon_probe(struct hv_device *dev,#ifdef CONFIG_MEMORY_HOTPLUG set_online_page_callback (&hv_online_page); +register_memory_notifier (&hv_memory_nb);#endif hv_set_drvdata (dev, &dm_device);
@@ -1520,6 +1578,7 @@ static int balloon_remove(struct hv_device *dev)kfree (send_buffer);#ifdef CONFIG_MEMORY_HOTPLUG restore_online_page_callback (&hv_online_page); +unregister_memory_notifier (&hv_memory_nb);#endif list_for_each_safe (cur, tmp, &dm->ha_region_list) { has = list_entry(cur, struct hv_hotadd_state, list);
Signature · 1 line
1.7.4.1
PATCH 3/3 Drivers: hv: hv_balloon: Don't post pressure status from interrupt context
In reply to K. Y. SrinivasanWe currently release memory (balloon down) in the interrupt context and we also post memory status while releasing memory. Rather than posting the status in the interrupt context, wakeup the status posting thread to post the status. This will address the inconsistent lock state that Sitsofe Wheeler <[email protected]> reported:
http://lkml.iu.edu/hypermail/linux/kernel/1411.1/00075.html
--- drivers/hv/hv_balloon.c | 11 ++++------- 1 files changed, 4 insertions(+), 7 deletions(-)
drivers/hv/hv_balloon.c +4 -7
@@ -1227,7 +1227,7 @@ static void balloon_down(struct hv_dynmem_device *dm,for (i =0 ; i < range_count; i++) {free_balloon_pages (dm, &range_array[i]); -post_status (&dm_device); +complete (&dm_device.config_event); }if (req->more_pages ==1 )
@@ -1251,19 +1251,16 @@ static void balloon_onchannelcallback(void *context);static int dm_thread_func (void *dm_dev) {struct hv_dynmem_device *dm = dm_dev; -int t;while (!kthread_should_stop ()) { - t =wait_for_completion_interruptible_timeout ( +wait_for_completion_interruptible_timeout ( &dm_device.config_event,1 *HZ );/* * The host expects us to post information on the memory * pressure every second. */ - -if (t ==0 ) -post_status (dm); - +reinit_completion (&dm_device.config_event); +post_status (dm); }return 0 ;
Signature · 1 line
1.7.4.1
Re: [PATCH 3/3] Drivers: hv: hv_balloon: Don't post pressure status from interrupt context
Quoted 5 lines · K. Y. Srinivasan
We currently release memory (balloon down) in the interrupt context and we also post memory status while releasing memory. Rather than posting the status in the interrupt context, wakeup the status posting thread to post the status. This will address the inconsistent lock state that Sitsofe Wheeler <[email protected]> reported:
Sorry I didn't follow up on this. Things looked better after the patch but I was seeing a strange issue where the memory was being reduced to some minimum and the balloon would never shrink so the available memory was seemingly locked from then on. I ran out of time to investigate this later behaviour and didn't have the time to check if it had always been the case but I think I see other patches saying you nailed it and other issues. Do you have a published git tree with all these pieces in or have they already been mainlined?
Signature · 1 line
Sitsofe | http://sucs.org/~sits/
RE: [PATCH 3/3] Drivers: hv: hv_balloon: Don't post pressure status from interrupt context
Quoted 21 lines
-----Original Message-----
From: Sitsofe Wheeler [mailto:[email protected]]
Sent: Sunday, December 7, 2014 12:05 AM
To: KY Srinivasan
Cc: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH 3/3] Drivers: hv:hv_balloon: Don't post pressure status from interrupt contextQuoted 3 lines · K. Y. Srinivasan
We currently release memory (balloon down) in the interrupt context and we also post memory status while releasing memory. Rather than posting the status in the interrupt context, wakeup the status posting
thread to post the status.
Quoted 3 lines
This will address the inconsistent lock state that Sitsofe Wheeler <[email protected]>
reported:
Quoted 8 lines
Sorry I didn't follow up on this. Things looked better after the patch but I was seeing a strange issue where the memory was being reduced to some minimum and the balloon would never shrink so the available memory was seemingly locked from then on. I ran out of time to investigate this later behaviour and didn't have the time to check if it had always been the case but I think I see other patches saying you nailed it and other issues. Do you have a published git tree with all these pieces in or have they already been mainlined?
Greg has not committed these patches yet. One of the patches changes the balloon floor. This means that the guest will not be ballooned down below the floor. Is this what you are seeing? In our testing we did not see anything unusual other than the floor being elevated (as per the design).
K. Y
Quoted 3 lines
--
Sitsofe | http://sucs.org/~sits/
Re: [PATCH 3/3] Drivers: hv: hv_balloon: Don't post pressure status from interrupt context
Quoted 5 lines · KY Srinivasan
Greg has not committed these patches yet. One of the patches changes the balloon floor. This means that the guest will not be ballooned down below the floor. Is this what you are seeing? In our testing we did not see anything unusual other than the floor being elevated (as per the design).
I applied the following:
drivers-scsi-storvsc-Fix-a-bug-in-handling-ring-buffer-failures-that-may-result-in-I-O-freeze.patch V2-1-3-Drivers-hv-hv_balloon-Make-adjustments-in-computing-the-floor.patch V2-2-3-Drivers-hv-hv_balloon-Fix-a-locking-bug-in-the-balloon-driver.patch V2-3-3-Drivers-hv-hv_balloon-Don-t-post-pressure-status-from-interrupt-context.patch
Initially things looked OK but now I'm starting to see the following which is rather worrying:
Dec 10 20:37:11 a kernel: BUG: unable to handle kernel NULL pointer dereference at (null) Dec 10 20:37:11 a kernel: IP: [<ffffffff811c30a0>] commit_charge+0x20/0x90 Dec 10 20:37:11 a kernel: PGD e44cb067 PUD e4495067 PMD 0 Dec 10 20:37:11 a kernel: Oops: 0000 [#1] SMP DEBUG_PAGEALLOC Dec 10 20:37:11 a kernel: CPU: 5 PID: 1490 Comm: ruby Not tainted 3.18.0.x86_64-01967-g86c6a2f-dirty #163 Dec 10 20:37:11 a kernel: Hardware name: Microsoft Corporation Virtual Machine/Virtual Machine, BIOS 090006 05/23/2012 Dec 10 20:37:11 a kernel: task: ffff8800e9bce040 ti: ffff880003890000 task.ti: ffff880003890000 Dec 10 20:37:11 a kernel: RIP: 0010:[<ffffffff811c30a0>] [<ffffffff811c30a0>] commit_charge+0x20/0x90 Dec 10 20:37:11 a kernel: RSP: 0018:ffff880003893a88 EFLAGS: 00010246 Dec 10 20:37:11 a kernel: RAX: 0000000000000000 RBX: ffffea00048d0380 RCX: 0000000000000006 Dec 10 20:37:11 a kernel: RDX: 0000000000000480 RSI: ffff880108829bd8 RDI: 000000000012340e Dec 10 20:37:11 a kernel: RBP: ffff880003893ac8 R08: 0000000000000000 R09: 0000000000000000 Dec 10 20:37:11 a kernel: R10: 0000000000000001 R11: 0000000000000000 R12: 0000000000000000 Dec 10 20:37:11 a kernel: R13: ffff880108829bd8 R14: ffff880017669c58 R15: 0000000000000000 Dec 10 20:37:11 a kernel: FS: 00007f4dc62fa740(0000) GS:ffff88010d4a0000(0000) knlGS:0000000000000000 Dec 10 20:37:11 a kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 Dec 10 20:37:11 a kernel: CR2: 0000000000000000 CR3: 00000000f1459000 CR4: 00000000000406e0 Dec 10 20:37:11 a kernel: Stack: Dec 10 20:37:11 a kernel: ffff8800e9bce040 ffffffff816f3950 0000000000000000 ffff880017669c58 Dec 10 20:37:11 a kernel: ffff880003893ac8 ffffea00048d0380 ffff880108829bd8 0000000000000000 Dec 10 20:37:11 a kernel: ffff880003893af8 ffffffff811c6b36 ffff880003893af8 ffffea00048d0380 Dec 10 20:37:11 a kernel: Call Trace: Dec 10 20:37:11 a kernel: [<ffffffff816f3950>] ? _raw_spin_unlock_irq+0x30/0x50 Dec 10 20:37:11 a kernel: [<ffffffff811c6b36>] mem_cgroup_commit_charge+0x76/0x140 Dec 10 20:37:11 a kernel: [<ffffffff8115d8d5>] __add_to_page_cache_locked+0x1e5/0x2d0 Dec 10 20:37:11 a kernel: [<ffffffff8115dfb8>] add_to_page_cache_lru+0x28/0x80 Dec 10 20:37:11 a kernel: [<ffffffff8115f347>] pagecache_get_page+0x197/0x220 Dec 10 20:37:11 a kernel: [<ffffffff81160cc3>] grab_cache_page_write_begin+0x33/0x50 Dec 10 20:37:11 a kernel: [<ffffffff81254fd7>] ext4_da_write_begin+0x157/0x340 Dec 10 20:37:11 a kernel: [<ffffffff81160da1>] generic_perform_write+0xc1/0x1d0 Dec 10 20:37:11 a kernel: [<ffffffff81161138>] __generic_file_write_iter+0x288/0x340 Dec 10 20:37:11 a kernel: [<ffffffff8124a693>] ext4_file_write_iter+0x2f3/0x3b0 Dec 10 20:37:11 a kernel: [<ffffffff811cde47>] ? vfs_write+0xa7/0x1d0 Dec 10 20:37:11 a kernel: [<ffffffff811cdc31>] new_sync_write+0x81/0xb0 Dec 10 20:37:11 a kernel: [<ffffffff811cde6b>] vfs_write+0xcb/0x1d0 Dec 10 20:37:11 a kernel: [<ffffffff811ce069>] SyS_write+0x49/0xb0 Dec 10 20:37:11 a kernel: [<ffffffff816f45a9>] system_call_fastpath+0x12/0x17 Dec 10 20:37:11 a kernel: Code: 5d c3 66 0f 1f 84 00 00 00 00 00 66 66 66 66 90 55 48 89 e5 41 55 49 89 f5 41 54 41 89 d4 53 48 89 fb 48 83 ec 28 e8 90 3e 00 00 <f6> 00 01 74 1b 48 c7 c6 e0 f1 9e 81 48 89 df e8 cc 4f fc ff 0f Dec 10 20:37:11 a kernel: RIP [<ffffffff811c30a0>] commit_charge+0x20/0x90 Dec 10 20:37:11 a kernel: RSP <ffff880003893a88> Dec 10 20:37:11 a kernel: CR2: 0000000000000000 Dec 10 20:37:11 a kernel: BUG: unable to handle kernel Dec 10 20:37:11 a kernel: ---[ end trace 0ae405bbdfb1f416 ]--- Dec 10 20:37:11 a kernel: NULL pointer dereference Dec 10 20:37:11 a kernel: at (null) Dec 10 20:37:11 a kernel: IP: [<ffffffff811c30a0>] commit_charge+0x20/0x90 Dec 10 20:37:11 a kernel: PGD f17d4067 PUD f1567067 PMD 0 Dec 10 20:37:12 a kernel: Oops: 0000 [#2] SMP DEBUG_PAGEALLOC Dec 10 20:37:12 a kernel: CPU: 2 PID: 25465 Comm: ruby Tainted: G D 3.18.0.x86_64-01967-g86c6a2f-dirty #163 Dec 10 20:37:12 a kernel: Hardware name: Microsoft Corporation Virtual Machine/Virtual Machine, BIOS 090006 05/23/2012 Dec 10 20:37:12 a kernel: task: ffff880011a16040 ti: ffff880098754000 task.ti: ffff880098754000 Dec 10 20:37:12 a kernel: init_memory_mapping: [mem 0x128000000-0x12fffffff] Dec 10 20:37:12 a kernel: [mem 0x128000000-0x12fffffff] page 4k Dec 10 20:37:12 a kernel: [ffffea0004800000-ffffea00049fffff] PMD -> [ffff8800c7400000-ffff8800c75fffff] on node 0 Dec 10 20:37:12 a kernel: RIP: 0010:[<ffffffff811c30a0>] [<ffffffff811c30a0>] commit_charge+0x20/0x90 Dec 10 20:37:12 a kernel: RSP: 0000:ffff880098757d18 EFLAGS: 00010246 Dec 10 20:37:12 a kernel: RAX: 0000000000000000 RBX: ffffea0004915300 RCX: 0000000000000001 Dec 10 20:37:12 a kernel: RDX: 0000000000000480 RSI: ffff880108829bd8 RDI: 000000000012454c Dec 10 20:37:12 a kernel: RBP: ffff880098757d58 R08: 0000000000000006 R09: 0000000000000000 Dec 10 20:37:12 a kernel: R10: ffff880011a16040 R11: 0000000000000000 R12: 0000000000000000 Dec 10 20:37:12 a kernel: R13: ffff880108829bd8 R14: ffff8800f159a5f0 R15: ffff88006b3bc600 Dec 10 20:37:12 a kernel: FS: 00007f0836edf700(0000) GS:ffff88010d440000(0000) knlGS:0000000000000000 Dec 10 20:37:12 a kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 Dec 10 20:37:12 a kernel: CR2: 0000000000000000 CR3: 00000000b8bfd000 CR4: 00000000000406e0 Dec 10 20:37:12 a kernel: Stack: Dec 10 20:37:12 a kernel: 00000000811bf285 ffff88000723e118 ffff880108829bd8 ffff88000723e100 Dec 10 20:37:12 a kernel: ffffea0004915300 ffffea0004915300 ffff880108829bd8 ffff88000613a280 Dec 10 20:37:12 a kernel: ffff880098757d88 ffffffff811c6b36 ffffffff8118d6fc 00007f08200bea58 Dec 10 20:37:12 a kernel: Call Trace: Dec 10 20:37:12 a kernel: [<ffffffff811c6b36>] mem_cgroup_commit_charge+0x76/0x140 Dec 10 20:37:12 a kernel: [<ffffffff8118d6fc>] ? handle_mm_fault+0x62c/0x12a0 Dec 10 20:37:12 a kernel: [<ffffffff8118d742>] handle_mm_fault+0x672/0x12a0 Dec 10 20:37:12 a kernel: [<ffffffff81041a13>] ? __do_page_fault+0x1c3/0x4f0 Dec 10 20:37:12 a kernel: [<ffffffff81041ce0>] __do_page_fault+0x490/0x4f0 Dec 10 20:37:12 a kernel: [<ffffffff810bf2cd>] ? trace_hardirqs_on+0xd/0x10 Dec 10 20:37:12 a kernel: [<ffffffff816f3950>] ? _raw_spin_unlock_irq+0x30/0x50 Dec 10 20:37:12 a kernel: [<ffffffff81097a88>] ? finish_task_switch+0x88/0x100 Dec 10 20:37:12 a kernel: [<ffffffff81097a4a>] ? finish_task_switch+0x4a/0x100 Dec 10 20:37:12 a kernel: [<ffffffff816ee380>] ? __schedule+0x6a0/0x830 Dec 10 20:37:12 a kernel: [<ffffffff813b24ed>] ? trace_hardirqs_off_thunk+0x3a/0x3c Dec 10 20:37:12 a kernel: [<ffffffff81041d92>] do_page_fault+0x22/0x30 Dec 10 20:37:12 a kernel: [<ffffffff816f6398>] page_fault+0x28/0x30 Dec 10 20:37:12 a kernel: Code: 5d c3 66 0f 1f 84 00 00 00 00 00 66 66 66 66 90 55 48 89 e5 41 55 49 89 f5 41 54 41 89 d4 53 48 89 fb 48 83 ec 28 e8 90 3e 00 00 <f6> 00 01 74 1b 48 c7 c6 e0 f1 9e 81 48 89 df e8 cc 4f fc ff 0f Dec 10 20:37:12 a kernel: RIP [<ffffffff811c30a0>] commit_charge+0x20/0x90 Dec 10 20:37:12 a kernel: RSP <ffff880098757d18> Dec 10 20:37:12 a kernel: CR2: 0000000000000000 Dec 10 20:37:12 a kernel: ---[ end trace 0ae405bbdfb1f417 ]--- Dec 10 20:37:12 a kernel: BUG: sleeping function called from invalid context at kernel/locking/rwsem.c:41 Dec 10 20:37:12 a kernel: in_atomic(): 1, irqs_disabled(): 1, pid: 25465, name: ruby Dec 10 20:37:12 a kernel: INFO: lockdep is turned off. Dec 10 20:37:12 a kernel: irq event stamp: 2431342 Dec 10 20:37:12 a kernel: hardirqs last enabled at (2431341): [<ffffffff816f38fd>] _raw_spin_unlock_irqrestore+0x4d/0x70 Dec 10 20:37:12 a kernel: hardirqs last disabled at (2431342): [<ffffffff816f37dd>] _raw_spin_lock_irq+0x1d/0x60 Dec 10 20:37:12 a kernel: softirqs last enabled at (2431322): [<ffffffff81078458>] __do_softirq+0x298/0x340 Dec 10 20:37:12 a kernel: softirqs last disabled at (2431317): [<ffffffff810787c8>] irq_exit+0x58/0xc0 Dec 10 20:37:12 a kernel: CPU: 2 PID: 25465 Comm: ruby Tainted: G D 3.18.0.x86_64-01967-g86c6a2f-dirty #163 Dec 10 20:37:12 a kernel: Hardware name: Microsoft Corporation Virtual Machine/Virtual Machine, BIOS 090006 05/23/2012 Dec 10 20:37:12 a kernel: 0000000000000029 ffff8800987578f8 ffffffff816ea99f 0000000000000000 Dec 10 20:37:12 a kernel: ffff880011a16040 ffff880098757918 ffffffff810a2dc5 ffff880098757948 Dec 10 20:37:12 a kernel: ffffffff819d796f ffff880098757948 ffffffff810a2e46 ffffffff82b828c2 Dec 10 20:37:12 a kernel: Call Trace: Dec 10 20:37:12 a kernel: [<ffffffff816ea99f>] dump_stack+0x4e/0x68 Dec 10 20:37:12 a kernel: [<ffffffff810a2dc5>] ___might_sleep+0x115/0x120 Dec 10 20:37:12 a kernel: [<ffffffff810a2e46>] __might_sleep+0x76/0xa0 Dec 10 20:37:12 a kernel: [<ffffffff816f1f04>] down_read+0x24/0x70 Dec 10 20:37:12 a kernel: [<ffffffff81082de4>] exit_signals+0x24/0x140 Dec 10 20:37:12 a kernel: [<ffffffff81076714>] do_exit+0x134/0xa80 Dec 10 20:37:12 a kernel: [<ffffffff810cb8cc>] ? kmsg_dump+0xfc/0x110 Dec 10 20:37:12 a kernel: [<ffffffff810cb7f5>] ? kmsg_dump+0x25/0x110 Dec 10 20:37:12 a kernel: [<ffffffff810064e8>] oops_end+0xa8/0xc0 Dec 10 20:37:12 a kernel: [<ffffffff816e53bc>] no_context+0x319/0x362 Dec 10 20:37:12 a kernel: [<ffffffff816e55d0>] __bad_area_nosemaphore+0x1cb/0x1ea Dec 10 20:37:12 a kernel: [<ffffffff816e5602>] bad_area_nosemaphore+0x13/0x15 Dec 10 20:37:12 a kernel: [<ffffffff81041a3e>] __do_page_fault+0x1ee/0x4f0 Dec 10 20:37:12 a kernel: [<ffffffff811680d5>] ? __alloc_pages_nodemask+0x225/0xaf0 Dec 10 20:37:12 a kernel: [<ffffffff813b24ed>] ? trace_hardirqs_off_thunk+0x3a/0x3c Dec 10 20:37:12 a kernel: [<ffffffff81041d92>] do_page_fault+0x22/0x30 Dec 10 20:37:12 a kernel: [<ffffffff816f6398>] page_fault+0x28/0x30 Dec 10 20:37:12 a kernel: [<ffffffff811c30a0>] ? commit_charge+0x20/0x90 Dec 10 20:37:12 a kernel: [<ffffffff811c30a0>] ? commit_charge+0x20/0x90 Dec 10 20:37:12 a kernel: [<ffffffff811c6b36>] mem_cgroup_commit_charge+0x76/0x140 Dec 10 20:37:12 a kernel: [<ffffffff8118d6fc>] ? handle_mm_fault+0x62c/0x12a0 Dec 10 20:37:12 a kernel: [<ffffffff8118d742>] handle_mm_fault+0x672/0x12a0 Dec 10 20:37:12 a kernel: [<ffffffff81041a13>] ? __do_page_fault+0x1c3/0x4f0 Dec 10 20:37:12 a kernel: [<ffffffff81041ce0>] __do_page_fault+0x490/0x4f0 Dec 10 20:37:12 a kernel: [<ffffffff810bf2cd>] ? trace_hardirqs_on+0xd/0x10 Dec 10 20:37:12 a kernel: [<ffffffff816f3950>] ? _raw_spin_unlock_irq+0x30/0x50 Dec 10 20:37:12 a kernel: [<ffffffff81097a88>] ? finish_task_switch+0x88/0x100 Dec 10 20:37:12 a kernel: [<ffffffff81097a4a>] ? finish_task_switch+0x4a/0x100 Dec 10 20:37:12 a kernel: [<ffffffff816ee380>] ? __schedule+0x6a0/0x830 Dec 10 20:37:12 a kernel: [<ffffffff813b24ed>] ? trace_hardirqs_off_thunk+0x3a/0x3c Dec 10 20:37:12 a kernel: [<ffffffff81041d92>] do_page_fault+0x22/0x30 Dec 10 20:37:12 a kernel: [<ffffffff816f6398>] page_fault+0x28/0x30 Dec 10 20:37:12 a kernel: note: ruby[25465] exited with preempt_count 1 Dec 10 20:37:16 a kernel: init_memory_mapping: [mem 0x130000000-0x137ffffff] Dec 10 20:37:16 a kernel: [mem 0x130000000-0x137ffffff] page 4k Dec 10 20:37:16 a kernel: [ffffea0004a00000-ffffea0004bfffff] PMD -> [ffff880093200000-ffff8800933fffff] on node 0 Dec 10 20:37:17 a kernel: BUG: unable to handle kernel NULL pointer dereference at (null) Dec 10 20:37:17 a kernel: IP: [<ffffffff811c30a0>] commit_charge+0x20/0x90
Are these Hyper-V related?
Signature · 1 line
Sitsofe | http://sucs.org/~sits/
RE: [PATCH 3/3] Drivers: hv: hv_balloon: Don't post pressure status from interrupt context
Quoted 208 lines
-----Original Message-----
From: Sitsofe Wheeler [mailto:[email protected]]
Sent: Wednesday, December 10, 2014 12:50 PM
To: KY Srinivasan
Cc: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH 3/3] Drivers: hv:hv_balloon: Don't post pressure status from interrupt contextQuoted 2 lines · KY Srinivasan
Greg has not committed these patches yet. One of the patches changes
the balloon floor.
Quoted 3 lines
This means that the guest will not be ballooned down below the floor. Is this what you are seeing? In our testing we did not see anything unusual other than the floor being elevated (as per the design).
I applied the following:
drivers-scsi-storvsc-Fix-a-bug-in-handling-ring-buffer-failures-that-may- result-in-I-O-freeze.patch V2-1-3-Drivers-hv-hv_balloon-Make-adjustments-in-computing-the- floor.patch V2-2-3-Drivers-hv-hv_balloon-Fix-a-locking-bug-in-the-balloon-driver.patch V2-3-3-Drivers-hv-hv_balloon-Don-t-post-pressure-status-from-interrupt- context.patch
Initially things looked OK but now I'm starting to see the following which is rather worrying:
Dec 10 20:37:11 a kernel: BUG: unable to handle kernel NULL pointer dereference at (null) Dec 10 20:37:11 a kernel: IP: [<ffffffff811c30a0>] commit_charge+0x20/0x90 Dec 10 20:37:11 a kernel: PGD e44cb067 PUD e4495067 PMD 0 Dec 10 20:37:11 a kernel: Oops: 0000 [#1] SMP DEBUG_PAGEALLOC Dec 10 20:37:11 a kernel: CPU: 5 PID: 1490 Comm: ruby Not tainted 3.18.0.x86_64-01967-g86c6a2f-dirty#163 Dec 10 20:37:11 a kernel: Hardware name: Microsoft Corporation Virtual Machine/Virtual Machine, BIOS 090006 05/23/2012 Dec 10 20:37:11 a kernel:
task:ffff8800e9bce040ti:ffff880003890000task.ti:ffff880003890000Dec 10 20:37:11 a kernel: RIP: 0010:[<ffffffff811c30a0>] [<ffffffff811c30a0>]commit_charge+0x20/0x90Dec 10 20:37:11 a kernel: RSP:
0018:ffff880003893a88 EFLAGS: 00010246 Dec 10 20:37:11 a kernel: RAX:
0000000000000000 RBX:ffffea00048d0380RCX: 0000000000000006 Dec 10 20:37:11 a kernel: RDX: 0000000000000480 RSI:ffff880108829bd8RDI:000000000012340eDec 10 20:37:11 a kernel: RBP:ffff880003893ac8R08:
0000000000000000 R09: 0000000000000000 Dec 10 20:37:11 a kernel: R10:
0000000000000001 R11: 0000000000000000 R12: 0000000000000000 Dec 10 20:37:11 a kernel: R13:ffff880108829bd8R14:ffff880017669c58R15:
0000000000000000 Dec 10 20:37:11 a kernel: FS: 00007f4dc62fa740(0000) GS:ffff88010d4a0000(0000) knlGS:0000000000000000 Dec 10 20:37:11 a kernel:CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 Dec 10 20:37:11 a kernel: CR2: 0000000000000000 CR3: 00000000f1459000 CR4: 00000000000406e0 Dec 10 20:37:11 a kernel: Stack: Dec 10 20:37:11 a kernel: ffff8800e9bce040 ffffffff816f39500000000000000000
ffff880017669c58Dec 10 20:37:11 a kernel:ffff880003893ac8ffffea00048d0380ffff880108829bd80000000000000000 Dec 10 20:37:11 a kernel:ffff880003893af8ffffffff811c6b36ffff880003893af8ffffea00048d0380Dec 10 20:37:11 a kernel: Call Trace:Dec 10 20:37:11 a kernel: [<ffffffff816f3950>] ? _raw_spin_unlock_irq+0x30/0x50 Dec 10 20:37:11 a kernel: [<ffffffff811c6b36>] mem_cgroup_commit_charge+0x76/0x140 Dec 10 20:37:11 a kernel: [<ffffffff8115d8d5>] __add_to_page_cache_locked+0x1e5/0x2d0 Dec 10 20:37:11 a kernel: [<ffffffff8115dfb8>] add_to_page_cache_lru+0x28/0x80 Dec 10 20:37:11 a kernel: [<ffffffff8115f347>] pagecache_get_page+0x197/0x220 Dec 10 20:37:11 a kernel: [<ffffffff81160cc3>] grab_cache_page_write_begin+0x33/0x50 Dec 10 20:37:11 a kernel: [<ffffffff81254fd7>] ext4_da_write_begin+0x157/0x340 Dec 10 20:37:11 a kernel: [<ffffffff81160da1>] generic_perform_write+0xc1/0x1d0 Dec 10 20:37:11 a kernel: [<ffffffff81161138>] __generic_file_write_iter+0x288/0x340 Dec 10 20:37:11 a kernel: [<ffffffff8124a693>] ext4_file_write_iter+0x2f3/0x3b0 Dec 10 20:37:11 a kernel: [<ffffffff811cde47>] ? vfs_write+0xa7/0x1d0 Dec 10 20:37:11 a kernel: [<ffffffff811cdc31>] new_sync_write+0x81/0xb0 Dec 10 20:37:11 a kernel: [<ffffffff811cde6b>] vfs_write+0xcb/0x1d0 Dec 10 20:37:11 a kernel: [<ffffffff811ce069>] SyS_write+0x49/0xb0 Dec 10 20:37:11 a kernel: [<ffffffff816f45a9>] system_call_fastpath+0x12/0x17 Dec 10 20:37:11 akernel: Code: 5d c3 66 0f 1f 84 00 00 00 00 00 66 66 66 66 90 55 48 89 e5 41 55 49 89 f5 41 54 41 89 d4 53 48 89 fb 48 83 ec 28 e8 90 3e 00 00 <f6> 00 01 74 1b 48 c7 c6 e0 f1 9e 81 48 89 df e8 cc 4f fc ff 0f Dec 10 20:37:11 a kernel: RIP [<
ffffffff811c30a0>]commit_charge+0x20/0x90Dec 10 20:37:11 a kernel: RSP <ffff880003893a88> Dec 10 20:37:11 a kernel: CR2: 0000000000000000 Dec 10 20:37:11 a kernel: BUG: unable to handle kernel Dec 10 20:37:11 a kernel: ---[ end trace0ae405bbdfb1f416]--- Dec 10 20:37:11 a kernel: NULL pointer dereferenceDec 10 20:37:11 a kernel: at (null) Dec 10 20:37:11 a kernel: IP: [<ffffffff811c30a0>] commit_charge+0x20/0x90 Dec 10 20:37:11 a kernel: PGD f17d4067 PUD f1567067 PMD 0 Dec 10 20:37:12 a kernel: Oops: 0000 [#2] SMP DEBUG_PAGEALLOC Dec 10 20:37:12 a kernel: CPU: 2 PID: 25465 Comm: ruby Tainted: G D 3.18.0.x86_64-01967-g86c6a2f-dirty #163 Dec 10 20:37:12 a kernel: Hardware name: Microsoft Corporation Virtual Machine/Virtual Machine, BIOS 090006 05/23/2012 Dec 10 20:37:12 a kernel: task: ffff880011a16040 ti: ffff880098754000 task.ti: ffff880098754000 Dec 10 20:37:12 a kernel: init_memory_mapping: [mem 0x128000000-0x12fffffff] Dec 10 20:37:12 a kernel: [mem 0x128000000-0x12fffffff] page 4k Dec 10 20:37:12 a kernel: [ffffea0004800000-ffffea00049fffff] PMD -> [ffff8800c7400000-ffff8800c75fffff] on node 0 Dec 10 20:37:12 a kernel: RIP: 0010:[<ffffffff811c30a0>] [<ffffffff811c30a0>] commit_charge+0x20/0x90 Dec 10 20:37:12 a kernel: RSP: 0000:ffff880098757d18 EFLAGS: 00010246 Dec10 20:37:12 a kernel: RAX: 0000000000000000 RBX:
ffffea0004915300RCX:
0000000000000001 Dec 10 20:37:12 a kernel: RDX: 0000000000000480 RSI:ffff880108829bd8RDI:000000000012454cDec 10 20:37:12 a kernel: RBP:ffff880098757d58R08: 0000000000000006 R09: 0000000000000000 Dec 10 20:37:12 a kernel: R10:ffff880011a16040R11: 0000000000000000 R12:
0000000000000000 Dec 10 20:37:12 a kernel: R13:ffff880108829bd8R14:ffff8800f159a5f0R15:ffff88006b3bc600Dec 10 20:37:12 a kernel: FS:
00007f0836edf700(0000) GS:ffff88010d440000(0000) knlGS:0000000000000000Dec 10 20:37:12 a kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 Dec 10 20:37:12 a kernel: CR2: 0000000000000000 CR3: 00000000b8bfd000 CR4: 00000000000406e0 Dec 10 20:37:12 a kernel: Stack: Dec 10 20:37:12 a kernel: 00000000811bf285 ffff88000723e118
ffff880108829bd8ffff88000723e100Dec 10 20:37:12 a kernel:ffffea0004915300ffffea0004915300ffff880108829bd8ffff88000613a280Dec 10 20:37:12 a kernel:ffff880098757d88ffffffff811c6b36ffffffff8118d6fc00007f08200bea58Dec 10 20:37:12 a kernel: Call Trace:Dec 10 20:37:12 a kernel: [<ffffffff811c6b36>] mem_cgroup_commit_charge+0x76/0x140 Dec 10 20:37:12 a kernel: [<ffffffff8118d6fc>] ? handle_mm_fault+0x62c/0x12a0 Dec 10 20:37:12 a kernel: [<ffffffff8118d742>] handle_mm_fault+0x672/0x12a0 Dec 10 20:37:12 a kernel: [<ffffffff81041a13>] ? __do_page_fault+0x1c3/0x4f0 Dec 10 20:37:12 a kernel: [<ffffffff81041ce0>] __do_page_fault+0x490/0x4f0 Dec 10 20:37:12 a kernel: [<ffffffff810bf2cd>] ? trace_hardirqs_on+0xd/0x10 Dec 10 20:37:12 a kernel: [<ffffffff816f3950>] ? _raw_spin_unlock_irq+0x30/0x50 Dec 10 20:37:12 a kernel: [<ffffffff81097a88>] ? finish_task_switch+0x88/0x100 Dec 10 20:37:12 a kernel: [<ffffffff81097a4a>] ? finish_task_switch+0x4a/0x100 Dec 10 20:37:12 a kernel: [<ffffffff816ee380>] ? __schedule+0x6a0/0x830 Dec 10 20:37:12 a kernel: [<ffffffff813b24ed>] ? trace_hardirqs_off_thunk+0x3a/0x3c Dec 10 20:37:12 a kernel: [<ffffffff81041d92>] do_page_fault+0x22/0x30 Dec 10 20:37:12 a kernel: [<ffffffff816f6398>] page_fault+0x28/0x30 Dec 1020:37:12 a kernel: Code: 5d c3 66 0f 1f 84 00 00 00 00 00 66 66 66 66 90 55 48 89 e5 41 55 49 89 f5 41 54 41 89 d4 53 48 89 fb 48 83 ec 28 e8 90 3e 00 00 <f6> 00 01 74 1b 48 c7 c6 e0 f1 9e 81 48 89 df e8 cc 4f fc ff 0f Dec 10 20:37:12 a kernel:
RIP [<ffffffff811c30a0>]commit_charge+0x20/0x90Dec 10 20:37:12 a kernel:
RSP <ffff880098757d18> Dec 10 20:37:12 a kernel: CR2: 0000000000000000 Dec 10 20:37:12 a kernel: ---[ end trace0ae405bbdfb1f417]--- Dec 10 20:37:12 a
kernel: BUG: sleeping function called from invalid context atkernel/locking/rwsem.c:41Dec 10 20:37:12 a kernel:in_atomic(): 1,irqs_disabled(): 1, pid: 25465, name: ruby Dec 10 20:37:12 a kernel: INFO:
lockdep is turned off.Dec 10 20:37:12 a kernel: irq event stamp: 2431342 Dec 10 20:37:12 a kernel: hardirqs last enabled at (2431341): [<ffffffff816f38fd>] _raw_spin_unlock_irqrestore+0x4d/0x70 Dec 10 20:37:12 a kernel: hardirqs last disabled at (2431342): [<ffffffff816f37dd>] _raw_spin_lock_irq+0x1d/0x60 Dec 10 20:37:12 a kernel: softirqs last enabled at (2431322): [<ffffffff81078458>] __do_softirq+0x298/0x340 Dec 10 20:37:12 a kernel: softirqs last disabled at (2431317): [<ffffffff810787c8>] irq_exit+0x58/0xc0 Dec 10 20:37:12 a kernel: CPU: 2 PID: 25465 Comm: ruby Tainted: G D 3.18.0.x86_64-01967-g86c6a2f-dirty #163 Dec 10 20:37:12 a kernel: Hardware name: Microsoft Corporation VirtualMachine/Virtual Machine, BIOS 090006 05/23/2012 Dec 10 20:37:12 a kernel:
0000000000000029ffff8800987578f8ffffffff816ea99f0000000000000000 Dec 10 20:37:12 a kernel:ffff880011a16040ffff880098757918ffffffff810a2dc5ffff880098757948Dec 10 20:37:12 a kernel:ffffffff819d796fffff880098757948ffffffff810a2e46ffffffff82b828c2Dec 10 20:37:12 a kernel: Call Trace:Dec 10 20:37:12 a kernel: [<ffffffff816ea99f>] dump_stack+0x4e/0x68 Dec 10 20:37:12 a kernel: [<ffffffff810a2dc5>] ___might_sleep+0x115/0x120 Dec 10 20:37:12 a kernel: [<ffffffff810a2e46>] __might_sleep+0x76/0xa0 Dec 10 20:37:12 a kernel: [<ffffffff816f1f04>] down_read+0x24/0x70 Dec 10 20:37:12 a kernel: [<ffffffff81082de4>] exit_signals+0x24/0x140 Dec 10 20:37:12 a kernel: [<ffffffff81076714>] do_exit+0x134/0xa80 Dec 10 20:37:12 a kernel: [<ffffffff810cb8cc>] ? kmsg_dump+0xfc/0x110 Dec 10 20:37:12 a kernel: [<ffffffff810cb7f5>] ? kmsg_dump+0x25/0x110 Dec 10 20:37:12 a kernel: [<ffffffff810064e8>] oops_end+0xa8/0xc0 Dec 10 20:37:12 a kernel: [<ffffffff816e53bc>] no_context+0x319/0x362 Dec 10 20:37:12 a kernel: [<ffffffff816e55d0>] __bad_area_nosemaphore+0x1cb/0x1ea Dec 10 20:37:12 a kernel: [<ffffffff816e5602>] bad_area_nosemaphore+0x13/0x15 Dec 10 20:37:12 a kernel: [<ffffffff81041a3e>] __do_page_fault+0x1ee/0x4f0 Dec 10 20:37:12 a kernel: [<ffffffff811680d5>] ? __alloc_pages_nodemask+0x225/0xaf0 Dec 10 20:37:12 a kernel: [<ffffffff813b24ed>] ? trace_hardirqs_off_thunk+0x3a/0x3c Dec 10 20:37:12 a kernel: [<ffffffff81041d92>] do_page_fault+0x22/0x30 Dec 10 20:37:12 a kernel: [<ffffffff816f6398>] page_fault+0x28/0x30 Dec 10 20:37:12 a kernel: [<ffffffff811c30a0>] ? commit_charge+0x20/0x90 Dec 10 20:37:12 a kernel: [<ffffffff811c30a0>] ? commit_charge+0x20/0x90 Dec 10 20:37:12 a kernel: [<ffffffff811c6b36>] mem_cgroup_commit_charge+0x76/0x140 Dec 10 20:37:12 a kernel: [<ffffffff8118d6fc>] ? handle_mm_fault+0x62c/0x12a0 Dec 10 20:37:12 a kernel: [<ffffffff8118d742>] handle_mm_fault+0x672/0x12a0 Dec 10 20:37:12 a kernel: [<ffffffff81041a13>] ? __do_page_fault+0x1c3/0x4f0 Dec 10 20:37:12 a kernel: [<ffffffff81041ce0>] __do_page_fault+0x490/0x4f0 Dec 10 20:37:12 a kernel: [<ffffffff810bf2cd>] ? trace_hardirqs_on+0xd/0x10 Dec 10 20:37:12 a kernel: [<ffffffff816f3950>] ? _raw_spin_unlock_irq+0x30/0x50 Dec 10 20:37:12 a kernel: [<ffffffff81097a88>] ? finish_task_switch+0x88/0x100 Dec 10 20:37:12 a kernel: [<ffffffff81097a4a>] ? finish_task_switch+0x4a/0x100 Dec 10 20:37:12 a kernel: [<ffffffff816ee380>] ? __schedule+0x6a0/0x830 Dec 10 20:37:12 a kernel: [<ffffffff813b24ed>] ? trace_hardirqs_off_thunk+0x3a/0x3c Dec 10 20:37:12 a kernel: [<ffffffff81041d92>] do_page_fault+0x22/0x30 Dec 10 20:37:12 a kernel: [<ffffffff816f6398>] page_fault+0x28/0x30 Dec 10 20:37:12 a kernel: note: ruby[25465] exited with preempt_count 1 Dec 10 20:37:16 a kernel: init_memory_mapping: [mem 0x130000000-0x137ffffff] Dec 10 20:37:16 a kernel: [mem 0x130000000-0x137ffffff] page 4k Dec 10 20:37:16 a kernel: [ffffea0004a00000-ffffea0004bfffff] PMD -> [ffff880093200000-ffff8800933fffff] on node 0 Dec 10 20:37:17 a kernel: BUG: unable to handle kernel NULL pointer dereference at (null) Dec 10 20:37:17 a kernel: IP: [<ffffffff811c30a0>] commit_charge+0x20/0x90
Quoted 1 line
Are these Hyper-V related?
I cannot see how the patches that you have applied can cause this problem. What do you need to run to trigger this problem. Do you see this issue even without these patches.
Regards,
K. Y
Re: [PATCH 1/3] Drivers: hv: hv_balloon: Make adjustments in computing the floor
In reply to K. Y. SrinivasanQuoted 2 lines · K. Y. Srinivasan
Make adjustments in computing the balloon floor.
The changelog should say what the user visible effects of the bug are.
regards,
dan carpenter
RE: [PATCH 1/3] Drivers: hv: hv_balloon: Make adjustments in computing the floor
Quoted 12 lines
-----Original Message-----
From: Dan Carpenter [mailto:[email protected]]
Sent: Tuesday, December 2, 2014 1:51 PM
To: KY Srinivasan
Cc: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH 1/3] Drivers: hv:hv_balloon: Make adjustments in computing the floor
Quoted 5 lines
Quoted 2 lines · K. Y. Srinivasan
Make adjustments in computing the balloon floor.
The changelog should say what the user visible effects of the bug are.
Thanks Dan; I will make the log more descriptive.
K. Y
Quoted 3 lines
regards,
dan carpenter