SLOB breaks Crypto
in [Kernel]
|
Prev: [RELEASE] LTTng 0.214 for kernel 2.6.33.4
Next: vlynq: make whole Kconfig-menu dependant on architecture
From: Adrian-Ken Rueegsegger on 14 May 2010 11:00 Hi, Herbert Xu wrote: > On Thu, Mar 18, 2010 at 10:24:41PM +0100, michael-dev(a)fami-braun.de wrote: >> Pekka Enberg schrieb: >>> Even with CONFIG_DEBUG_SLAB enabled or with CONFIG_SLUB and >>> CONFIG_SLUB_DEBUG_ON? >> no, these options have not been / are not enabled. > > Can you please try it with those options enabled? That will tell > us if there is some latent bug in the crypto layer that only shows > up right away under SLOB. I was able reproduced the issue with the current crypto-2.6 tree 180ce7e... The issue does not show up with CONFIG_DEBUG_SLAB nor CONFIG_SLUB and CONFIG_SLUB_DEBUG_ON. It seems the issues is really related to hmac. Loading tcrypt with modes not using hmac seem to run fine. During my tests the system freezes eventually when doing multiple modprobes of tcrypt like this: modprobe tcrypt mode=100 I get multiple OOPses with a subsequent panic looking like this: [ 409.631551] BUG: unable to handle kernel NULL pointer dereference at 00000090 [ 409.631645] IP: [<f883a278>] hmac_setkey+0x38/0x140 [hmac] [ 409.631705] *pde = 00000000 [ 409.631757] Oops: 0000 [#1] PREEMPT DEBUG_PAGEALLOC [ 409.631857] last sysfs file: /sys/class/power_supply/AC/online [ 409.631891] Modules linked in: tcrypt(+) hmac netconsole loop ide_cd_mod cdrom ide_pci_generic ehci_hcd uhci_hcd piix usbcore [last unloaded: scsi_wait_scan] [ 409.632019] [ 409.632019] Pid: 5184, comm: modprobe Not tainted 2.6.33-g180ce7e #5 /Latitude D600 [ 409.632019] EIP: 0060:[<f883a278>] EFLAGS: 00010282 CPU: 0 [ 409.632019] EIP is at hmac_setkey+0x38/0x140 [hmac] [ 409.632019] EAX: e0254734 EBX: e025476c ECX: 00000090 EDX: e0254820 [ 409.632019] ESI: 00000040 EDI: e02547c4 EBP: e1dc3cb4 ESP: e1dc3c94 [ 409.632019] DS: 007b ES: 007b FS: 0000 GS: 0033 SS: 0068 [ 409.632019] Process modprobe (pid: 5184, ti=e1dc2000 task=e538b000 task.ti=e1dc2000) [ 409.632019] Stack: [ 409.632019] e0254668 e0254820 00000010 c043a348 00000010 e538b87c c043a348 [ 409.632019] <0> e1dc3cd8 c020f3d6 00000010 e0254734 e538b87c c04945b4 e1e47000 00000000 [ 409.632019] <0> c043a348 e1dc3ce0 c020f3fb e1dc3d00 c020eb4b 00000010 ebe26380 00000008 [ 409.632019] Call Trace: [ 409.632019] [<c020f3d6>] ? crypto_shash_setkey+0x96/0xa0 [ 409.632019] [<c020f3fb>] ? shash_async_setkey+0xb/0x10 [ 409.632019] [<c020eb4b>] ? crypto_ahash_setkey+0x8b/0x90 [ 409.632019] [<c0211056>] ? test_hash+0x176/0x620 [ 409.632019] [<c012a754>] ? _local_bh_enable+0x24/0x80 [ 409.632019] [<c0108441>] ? native_sched_clock+0x21/0x80 [ 409.632019] [<c0186fce>] ? slob_page_alloc+0x20e/0x240 [ 409.632019] [<c018758d>] ? slob_alloc+0x10d/0x1f0 [ 409.632019] [<c0187791>] ? __kmalloc_node+0xa1/0xc0 [ 409.632019] [<c02095e1>] ? crypto_create_tfm+0x41/0xc0 [ 409.632019] [<c020a77a>] ? crypto_spawn_tfm2+0x3a/0x60 [ 409.632019] [<f883a4f6>] ? hmac_init_tfm+0x26/0x5c [hmac] [ 409.632019] [<c0209620>] ? crypto_create_tfm+0x80/0xc0 [ 409.632019] [<c0211535>] ? alg_test_hash+0x35/0x70 [ 409.632019] [<c021320f>] ? alg_find_test+0x3f/0x70 [ 409.632019] [<c0213299>] ? alg_test+0x59/0x180 [ 409.632019] [<f8865a55>] ? do_test+0xec5/0x14ec [tcrypt] [ 409.632019] [<f886a0c1>] ? tcrypt_mod_init+0xc1/0xce [tcrypt] [ 409.632019] [<f886a000>] ? tcrypt_mod_init+0x0/0xce [tcrypt] [ 409.632019] [<c0101023>] ? do_one_initcall+0x23/0x170 [ 409.632019] [<c013f7fa>] ? blocking_notifier_call_chain+0x1a/0x20 [ 409.632019] [<c0151a02>] ? sys_init_module+0xb2/0x220 [ 409.632019] [<c017c974>] ? sys_mmap_pgoff+0xe4/0xf0 [ 409.632019] [<c0102ad0>] ? sysenter_do_call+0x12/0x26 [ 409.632019] Code: 55 ec 8b 50 34 8d 4a d4 8b 72 14 8b 59 24 89 5d f0 8b 52 1c 8b 49 28 8d 10 38 f7 d2 21 d3 8d 3c 0b 8d 54 0f 07 83 e2 f8 8b 0a <8b> 11 83 c2 0f 83 e2 f8 83 c2 10 29 d4 8d 54 24 13 83 e2 f0 89 [ 409.632019] EIP: [<f883a278>] hmac_setkey+0x38/0x140 [hmac] SS:ESP 0068:e1dc3c94 [ 409.632019] CR2: 0000000000000090 [ 474.594010] BUG: soft lockup - CPU#0 stuck for 61s! [modprobe:5184] [ 474.594010] Modules linked in: tcrypt(+) hmac netconsole loop ide_cd_mod cdrom ide_pci_generic ehci_hcd uhci_hcd piix usbcore [last unloaded: scsi_wait_scan] [ 474.594010] Modules linked in: tcrypt(+) hmac netconsole loop ide_cd_mod cdrom ide_pci_generic ehci_hcd uhci_hcd piix usbcore [last unloaded: scsi_wait_scan] [ 474.594010] [ 474.594010] Pid: 5184, comm: modprobe Tainted: G D 2.6.33-g180ce7e #5 /Latitude D600 [ 474.594010] EIP: 0060:[<c01180f0>] EFLAGS: 00000246 CPU: 0 [ 474.594010] EIP is at __bad_area_nosemaphore+0x20/0x180 [ 474.594010] EAX: e538b000 EBX: 00000000 ECX: 00000000 EDX: 00000000 [ 474.594010] ESI: e1dc39f0 EDI: 00000000 EBP: e1dc39a8 ESP: e1dc38f8 [ 474.594010] DS: 007b ES: 007b FS: 0000 GS: 0033 SS: 0068 [ 474.594010] Process modprobe (pid: 5184, ti=e1dc2000 task=e538b000 task.ti=e1dc2000) [ 474.594010] Stack: [ 474.594010] c0104c8e e1dc3920 e1dc3920 00000000 e1dc3920 c0105e5f [ 474.594010] <0> e538b000 c048dde0 e1dc3934 c0157bbd 0000193c c048dde0 00000000 e1dc394c [ 474.594010] <0> c0104784 e1dc3980 e1dc3980 e1dc394c c012aaaa e1dc3978 c01042a2 0000006e [ 474.594010] Call Trace: [ 474.594010] [<c0104c8e>] ? timer_interrupt+0x1e/0x60 [ 474.594010] [<c0155be6>] ? handle_IRQ_event+0x36/0xd0 [ 474.594010] [<c0105c6d>] ? mask_and_ack_8259A+0x7d/0x110 [ 474.594010] [<c0105e5f>] ? enable_8259A_irq+0x4f/0x70 [ 474.594010] [<c0157bbd>] ? handle_level_irq+0x10d/0x130 [ 474.594010] [<c0104784>] ? handle_irq+0x34/0x60 [ 474.594010] [<c012aaaa>] ? irq_exit+0x5a/0x90 [ 474.594010] [<c01042a2>] ? do_IRQ+0x42/0xb0 [ 474.594010] [<c0102fe9>] ? common_interrupt+0x29/0x30 [ 474.594010] [<c0118262>] ? bad_area_nosemaphore+0x12/0x20 [ 474.594010] [<c011862c>] ? do_page_fault+0x23c/0x2e0 [ 474.594010] [<c01042a2>] ? do_IRQ+0x42/0xb0 [ 474.594010] [<c01183f0>] ? do_page_fault+0x0/0x2e0 [ 474.594010] [<c03838fe>] ? error_code+0x5e/0x64 [ 474.594010] [<c0150000>] ? load_module+0x280/0x1890 [ 474.594010] [<c01183f0>] ? do_page_fault+0x0/0x2e0 [ 474.594010] [<c015928c>] ? __rcu_process_callbacks+0xdc/0x410 [ 474.594010] [<c01595d2>] ? rcu_process_callbacks+0x12/0x30 [ 474.594010] [<c012a827>] ? __do_softirq+0x77/0x100 [ 474.594010] [<c0105c5b>] ? mask_and_ack_8259A+0x6b/0x110 [ 474.594010] [<c0105e5f>] ? enable_8259A_irq+0x4f/0x70 [ 474.594010] [<c012a8dd>] ? do_softirq+0x2d/0x40 [ 474.594010] [<c012aabd>] ? irq_exit+0x6d/0x90 [ 474.594010] [<c01042a2>] ? do_IRQ+0x42/0xb0 [ 474.594010] [<c0102fe9>] ? common_interrupt+0x29/0x30 [ 474.594010] [<c0120000>] ? proc_sched_show_task+0x1f0/0x600 [ 474.594010] [<c010571b>] ? oops_end+0x4b/0xb0 [ 474.594010] [<c011802e>] ? no_context+0xbe/0x160 [ 474.594010] [<c011811f>] ? __bad_area_nosemaphore+0x4f/0x180 [ 474.594010] [<c0287a53>] ? extract_buf+0x73/0xc0 [ 474.594010] [<c0108441>] ? native_sched_clock+0x21/0x80 [ 474.594010] [<c0108441>] ? native_sched_clock+0x21/0x80 [ 474.594010] [<c01182aa>] ? bad_area+0x3a/0x50 [ 474.594010] [<c0118685>] ? do_page_fault+0x295/0x2e0 [ 474.594010] [<c01183f0>] ? do_page_fault+0x0/0x2e0 [ 474.594010] [<c03838fe>] ? error_code+0x5e/0x64 [ 474.594010] [<c018007b>] ? pcpu_get_vm_areas+0x4cb/0x500 [ 474.594010] [<c01183f0>] ? do_page_fault+0x0/0x2e0 [ 474.594010] [<f883a278>] ? hmac_setkey+0x38/0x140 [hmac] [ 474.594010] [<c020f3d6>] ? crypto_shash_setkey+0x96/0xa0 [ 474.594010] [<c020f3fb>] ? shash_async_setkey+0xb/0x10 [ 474.594010] [<c020eb4b>] ? crypto_ahash_setkey+0x8b/0x90 [ 474.594010] [<c0211056>] ? test_hash+0x176/0x620 [ 474.594010] [<c012a754>] ? _local_bh_enable+0x24/0x80 [ 474.594010] [<c0108441>] ? native_sched_clock+0x21/0x80 [ 474.594010] [<c018758d>] ? slob_alloc+0x10d/0x1f0 [ 474.594010] [<c0187791>] ? __kmalloc_node+0xa1/0xc0 [ 474.594010] [<c02095e1>] ? crypto_create_tfm+0x41/0xc0 [ 474.594010] [<c020a77a>] ? crypto_spawn_tfm2+0x3a/0x60 [ 474.594010] [<f883a4f6>] ? hmac_init_tfm+0x26/0x5c [hmac] [ 474.594010] [<c0209620>] ? crypto_create_tfm+0x80/0xc0 [ 474.594010] [<c0211535>] ? alg_test_hash+0x35/0x70 [ 474.594010] [<c021320f>] ? alg_find_test+0x3f/0x70 [ 474.594010] [<c0213299>] ? alg_test+0x59/0x180 [ 474.594010] [<f8865a55>] ? do_test+0xec5/0x14ec [tcrypt] [ 474.594010] [<f886a0c1>] ? tcrypt_mod_init+0xc1/0xce [tcrypt] [ 474.594010] [<f886a000>] ? tcrypt_mod_init+0x0/0xce [tcrypt] [ 474.594010] [<c0101023>] ? do_one_initcall+0x23/0x170 [ 474.594010] [<c013f7fa>] ? blocking_notifier_call_chain+0x1a/0x20 [ 474.594010] [<c0151a02>] ? sys_init_module+0xb2/0x220 [ 474.594010] [<c017c974>] ? sys_mmap_pgoff+0xe4/0xf0 [ 474.594010] [<c0102ad0>] ? sysenter_do_call+0x12/0x26 [ 474.594010] Code: 00 00 00 00 8d bc 27 00 00 00 00 55 89 e5 81 ec b0 00 00 00 89 75 f8 89 c6 a1 98 80 48 c0 f6 c2 04 89 5d f4 89 cb 89 7d fc 89 d7 <89> 85 70 ff ff ff 74 20 fb 89 f0 e8 80 fc ff ff 74 24 8b [ 474.594010] Call Trace: [ 474.594010] [<c0104c8e>] ? timer_interrupt+0x1e/0x60 [ 474.594010] [<c0155be6>] ? handle_IRQ_event+0x36/0xd0 [ 474.594010] [<c0105c6d>] ? mask_and_ack_8259A+0x7d/0x110 [ 474.594010] [<c0105e5f>] ? enable_8259A_irq+0x4f/0x70 [ 474.594010] [<c0157bbd>] ? handle_level_irq+0x10d/0x130 [ 474.594010] [<c0104784>] ? handle_irq+0x34/0x60 [ 474.594010] [<c012aaaa>] ? irq_exit+0x5a/0x90 [ 474.594010] [<c01042a2>] ? do_IRQ+0x42/0xb0 [ 474.594010] [<c0102fe9>] ? common_interrupt+0x29/0x30 [ 474.594010] [<c0118262>] bad_area_nosemaphore+0x12/0x20 [ 474.594010] [<c011862c>] do_page_fault+0x23c/0x2e0 [ 474.594010] [<c01042a2>] ? do_IRQ+0x42/0xb0 [ 474.594010] [<c01183f0>] ? do_page_fault+0x0/0x2e0 [ 474.594010] [<c03838fe>] error_code+0x5e/0x64 [ 474.594010] [<c0150000>] ? load_module+0x280/0x1890 [ 474.594010] [<c01183f0>] ? do_page_fault+0x0/0x2e0 [ 474.594010] [<c015928c>] ? __rcu_process_callbacks+0xdc/0x410 [ 474.594010] [<c01595d2>] rcu_process_callbacks+0x12/0x30 [ 474.594010] [<c012a827>] __do_softirq+0x77/0x100 [ 474.594010] [<c0105c5b>] ? mask_and_ack_8259A+0x6b/0x110 [ 474.594010] [<c0105e5f>] ? enable_8259A_irq+0x4f/0x70 [ 474.594010] [<c012a8dd>] do_softirq+0x2d/0x40 [ 474.594010] [<c012aabd>] irq_exit+0x6d/0x90 [ 474.594010] [<c01042a2>] do_IRQ+0x42/0xb0 [ 474.594010] [<c0102fe9>] common_interrupt+0x29/0x30 [ 474.594010] [<c0120000>] ? proc_sched_show_task+0x1f0/0x600 [ 474.594010] [<c010571b>] ? oops_end+0x4b/0xb0 [ 474.594010] [<c011802e>] no_context+0xbe/0x160 [ 474.594010] [<c011811f>] __bad_area_nosemaphore+0x4f/0x180 [ 474.594010] [<c0287a53>] ? extract_buf+0x73/0xc0 [ 474.594010] [<c0108441>] ? native_sched_clock+0x21/0x80 [ 474.594010] [<c0108441>] ? native_sched_clock+0x21/0x80 [ 474.594010] [<c01182aa>] bad_area+0x3a/0x50 [ 474.594010] [<c0118685>] do_page_fault+0x295/0x2e0 [ 474.594010] [<c01183f0>] ? do_page_fault+0x0/0x2e0 [ 474.594010] [<c03838fe>] error_code+0x5e/0x64 [ 474.594010] [<c018007b>] ? pcpu_get_vm_areas+0x4cb/0x500 [ 474.594010] [<c01183f0>] ? do_page_fault+0x0/0x2e0 [ 474.594010] [<f883a278>] ? hmac_setkey+0x38/0x140 [hmac] [ 474.594010] [<c020f3d6>] crypto_shash_setkey+0x96/0xa0 [ 474.594010] [<c020f3fb>] shash_async_setkey+0xb/0x10 [ 474.594010] [<c020eb4b>] crypto_ahash_setkey+0x8b/0x90 [ 474.594010] [<c0211056>] test_hash+0x176/0x620 [ 474.594010] [<c012a754>] ? _local_bh_enable+0x24/0x80 [ 474.594010] [<c0108441>] ? native_sched_clock+0x21/0x80 [ 474.594010] [<c0186fce>] ? slob_page_alloc+0x20e/0x240 [ 474.594010] [<c018758d>] ? slob_alloc+0x10d/0x1f0 [ 474.594010] [<c0187791>] ? __kmalloc_node+0xa1/0xc0 [ 474.594010] [<c02095e1>] ? crypto_create_tfm+0x41/0xc0 [ 474.594010] [<c020a77a>] ? crypto_spawn_tfm2+0x3a/0x60 [ 474.594010] [<f883a4f6>] ? hmac_init_tfm+0x26/0x5c [hmac] [ 474.594010] [<c0209620>] ? crypto_create_tfm+0x80/0xc0 [ 474.594010] [<c0211535>] alg_test_hash+0x35/0x70 [ 474.594010] [<c021320f>] ? alg_find_test+0x3f/0x70 [ 474.594010] [<c0213299>] alg_test+0x59/0x180 [ 474.594010] [<f8865a55>] do_test+0xec5/0x14ec [tcrypt] [ 474.594010] [<f886a0c1>] tcrypt_mod_init+0xc1/0xce [tcrypt] [ 474.594010] [<f886a000>] ? tcrypt_mod_init+0x0/0xce [tcrypt] [ 474.594010] [<c0101023>] do_one_initcall+0x23/0x170 [ 474.594010] [<c013f7fa>] ? blocking_notifier_call_chain+0x1a/0x20 [ 474.594010] [<c0151a02>] sys_init_module+0xb2/0x220 [ 474.594010] [<c0102ad0>] sysenter_do_call+0x12/0x26 [ 474.594010] Kernel panic - not syncing: softlockup: hung tasks [ 474.594010] Pid: 5184, comm: modprobe Tainted: G D 2.6.33-g180ce7e #5 [ 474.594010] Call Trace: [ 474.594010] [<c0381525>] ? printk+0x18/0x1b [ 474.594010] [<c0381452>] panic+0x4d/0x108 [ 474.594010] [<c015552e>] softlockup_tick+0x14e/0x1a0 [ 474.594010] [<c0130247>] run_local_timers+0x17/0x20 [ 474.594010] [<c013027e>] update_process_times+0x2e/0x70 [ 474.594010] [<c0148c4b>] tick_sched_timer+0x5b/0xb0 [ 474.594010] [<c013df0a>] ? __remove_hrtimer+0x2a/0x90 [ 474.594010] [<c0148bf0>] ? tick_sched_timer+0x0/0xb0 [ 474.594010] [<c013e070>] __run_hrtimer+0x50/0x120 [ 474.594010] [<c013e3d1>] hrtimer_interrupt+0x111/0x2f0 [ 474.594010] [<c0104c8e>] timer_interrupt+0x1e/0x60 [ 474.594010] [<c0155be6>] handle_IRQ_event+0x36/0xd0 [ 474.594010] [<c0105c6d>] ? mask_and_ack_8259A+0x7d/0x110 [ 474.594010] [<c0157b2d>] handle_level_irq+0x7d/0x130 [ 474.594010] [<c0104784>] handle_irq+0x34/0x60 [ 474.594010] [<c0104299>] do_IRQ+0x39/0xb0 [ 474.594010] [<c01487b7>] ? tick_program_event+0x17/0x20 [ 474.594010] [<c013e464>] ? hrtimer_interrupt+0x1a4/0x2f0 [ 474.594010] [<c0225720>] ? search_extable+0x30/0x40 [ 474.594010] [<c0102fe9>] common_interrupt+0x29/0x30 [ 474.594010] [<c01180f0>] ? __bad_area_nosemaphore+0x20/0x180 [ 474.594010] [<c0104c8e>] ? timer_interrupt+0x1e/0x60 [ 474.594010] [<c0155be6>] ? handle_IRQ_event+0x36/0xd0 [ 474.594010] [<c0105c6d>] ? mask_and_ack_8259A+0x7d/0x110 [ 474.594010] [<c0105e5f>] ? enable_8259A_irq+0x4f/0x70 [ 474.594010] [<c0157bbd>] ? handle_level_irq+0x10d/0x130 [ 474.594010] [<c0104784>] ? handle_irq+0x34/0x60 [ 474.594010] [<c012aaaa>] ? irq_exit+0x5a/0x90 [ 474.594010] [<c01042a2>] ? do_IRQ+0x42/0xb0 [ 474.594010] [<c0102fe9>] ? common_interrupt+0x29/0x30 [ 474.594010] [<c0118262>] bad_area_nosemaphore+0x12/0x20 [ 474.594010] [<c011862c>] do_page_fault+0x23c/0x2e0 [ 474.594010] [<c01042a2>] ? do_IRQ+0x42/0xb0 [ 474.594010] [<c01183f0>] ? do_page_fault+0x0/0x2e0 [ 474.594010] [<c03838fe>] error_code+0x5e/0x64 [ 474.594010] [<c0150000>] ? load_module+0x280/0x1890 [ 474.594010] [<c01183f0>] ? do_page_fault+0x0/0x2e0 [ 474.594010] [<c015928c>] ? __rcu_process_callbacks+0xdc/0x410 [ 474.594010] [<c01595d2>] rcu_process_callbacks+0x12/0x30 [ 474.594010] [<c012a827>] __do_softirq+0x77/0x100 [ 474.594010] [<c0105c5b>] ? mask_and_ack_8259A+0x6b/0x110 [ 474.594010] [<c0105e5f>] ? enable_8259A_irq+0x4f/0x70 [ 474.594010] [<c012a8dd>] do_softirq+0x2d/0x40 [ 474.594010] [<c012aabd>] irq_exit+0x6d/0x90 [ 474.594010] [<c01042a2>] do_IRQ+0x42/0xb0 [ 474.594010] [<c0102fe9>] common_interrupt+0x29/0x30 [ 474.594010] [<c0120000>] ? proc_sched_show_task+0x1f0/0x600 [ 474.594010] [<c010571b>] ? oops_end+0x4b/0xb0 [ 474.594010] [<c011802e>] no_context+0xbe/0x160 [ 474.594010] [<c011811f>] __bad_area_nosemaphore+0x4f/0x180 [ 474.594010] [<c0287a53>] ? extract_buf+0x73/0xc0 [ 474.594010] [<c0108441>] ? native_sched_clock+0x21/0x80 [ 474.594010] [<c0108441>] ? native_sched_clock+0x21/0x80 [ 474.594010] [<c01182aa>] bad_area+0x3a/0x50 [ 474.594010] [<c0118685>] do_page_fault+0x295/0x2e0 [ 474.594010] [<c01183f0>] ? do_page_fault+0x0/0x2e0 [ 474.594010] [<c03838fe>] error_code+0x5e/0x64 [ 474.594010] [<c018007b>] ? pcpu_get_vm_areas+0x4cb/0x500 [ 474.594010] [<c01183f0>] ? do_page_fault+0x0/0x2e0 [ 474.594010] [<f883a278>] ? hmac_setkey+0x38/0x140 [hmac] [ 474.594010] [<c020f3d6>] crypto_shash_setkey+0x96/0xa0 [ 474.594010] [<c020eb4b>] crypto_ahash_setkey+0x8b/0x90 [ 474.594010] [<c0211056>] test_hash+0x176/0x620 [ 474.594010] [<c012a754>] ? _local_bh_enable+0x24/0x80 [ 474.594010] [<c0108441>] ? native_sched_clock+0x21/0x80 [ 474.594010] [<c0186fce>] ? slob_page_alloc+0x20e/0x240 [ 474.594010] [<c018758d>] ? slob_alloc+0x10d/0x1f0 The offending line (hmac_setkey+0x38/0x140) in crypto/hmac.c is: (gdb) list *(hmac_setkey+0x38) 0x278 is in hmac_setkey (/home/ken/projects/kernel/crypto-2.6/crypto/hmac.c:59). 54 crypto_tfm_ctx_alignment()); 55 struct crypto_shash *hash = ctx->hash; 56 struct { 57 struct shash_desc shash; 58 char ctx[crypto_shash_descsize(hash)]; 59 } desc; 60 unsigned int i; 61 62 desc.shash.tfm = hash; 63 desc.shash.flags = crypto_shash_get_flags(parent) & (gdb) Regards, Adrian -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Geert Uytterhoeven on 17 May 2010 12:20 On Fri, Mar 19, 2010 at 02:33, Herbert Xu <herbert(a)gondor.apana.org.au> wrote: > On Thu, Mar 18, 2010 at 10:24:41PM +0100, michael-dev(a)fami-braun.de wrote: >> >> Pekka Enberg schrieb: >> > Even with CONFIG_DEBUG_SLAB enabled or with CONFIG_SLUB and >> > CONFIG_SLUB_DEBUG_ON? >> >> no, these options have not been / are not enabled. > > Can you please try it with those options enabled? That will tell > us if there is some latent bug in the crypto layer that only shows > up right away under SLOB. After seeing a posting from Nemoto-san on the linux-mips list (should show up soon on http://www.linux-mips.org/archives/linux-mips/2010-05/threads.html), I'm wondering if these defaults are the culprit; mm/slab.c:#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long long) mm/slob.c:#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long) mm/slub.c:#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long long) Note that some platforms override this. Gr{oetje,eeting}s, Geert -- Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert(a)linux-m68k.org In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that. -- Linus Torvalds -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Adrian-Ken Rueegsegger on 17 May 2010 18:00 Geert Uytterhoeven wrote: > On Fri, Mar 19, 2010 at 02:33, Herbert Xu <herbert(a)gondor.apana.org.au> wrote: >> On Thu, Mar 18, 2010 at 10:24:41PM +0100, michael-dev(a)fami-braun.de wrote: >>> Pekka Enberg schrieb: >>>> Even with CONFIG_DEBUG_SLAB enabled or with CONFIG_SLUB and >>>> CONFIG_SLUB_DEBUG_ON? >>> no, these options have not been / are not enabled. >> Can you please try it with those options enabled? That will tell >> us if there is some latent bug in the crypto layer that only shows >> up right away under SLOB. > > After seeing a posting from Nemoto-san on the linux-mips list (should > show up soon on > http://www.linux-mips.org/archives/linux-mips/2010-05/threads.html), > I'm wondering if these defaults are the culprit; > > mm/slab.c:#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long long) > mm/slob.c:#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long) > mm/slub.c:#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long long) Just a quick note: I changed this line in mm/slob.c to (unsigned long long) and was able to load tcrypt without the kernel oopsing or panic'ing. When running all test via modprobe tcrypt the tests seem to pass fine judging by /proc/crypto. Nevertheless module insertion fails (as expected) but does not return EAGAIN but EMULTIHOP instead... Regards, Adrian -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Matt Mackall on 17 May 2010 18:40 On Mon, 2010-05-17 at 23:50 +0200, Adrian-Ken Rueegsegger wrote: > Geert Uytterhoeven wrote: > > On Fri, Mar 19, 2010 at 02:33, Herbert Xu <herbert(a)gondor.apana.org.au> wrote: > >> On Thu, Mar 18, 2010 at 10:24:41PM +0100, michael-dev(a)fami-braun.de wrote: > >>> Pekka Enberg schrieb: > >>>> Even with CONFIG_DEBUG_SLAB enabled or with CONFIG_SLUB and > >>>> CONFIG_SLUB_DEBUG_ON? > >>> no, these options have not been / are not enabled. > >> Can you please try it with those options enabled? That will tell > >> us if there is some latent bug in the crypto layer that only shows > >> up right away under SLOB. > > > > After seeing a posting from Nemoto-san on the linux-mips list (should > > show up soon on > > http://www.linux-mips.org/archives/linux-mips/2010-05/threads.html), > > I'm wondering if these defaults are the culprit; > > > > mm/slab.c:#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long long) > > mm/slob.c:#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long) > > mm/slub.c:#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long long) Seems the target here is x86, which has no DMA or other alignment constraints. > Just a quick note: I changed this line in mm/slob.c to (unsigned long > long) and was able to load tcrypt without the kernel oopsing or > panic'ing. When running all test via > > modprobe tcrypt Hmmm, sounds like something in crypto has a bogus alignment/padding expectation. Surprised it wasn't caught by SLAB redzoning though. Perhaps we can 'bisect' the test suite to narrow it down to a particular test. Alternately, we can tweak SLAB to offset buffers by two bytes and see what breaks. Ages ago, I had a simple redzoning patch for SLOB, but I never submitted it: I wrote it to catch bugs in SLOB, not bugs in SLOB users. -- Mathematics is the supreme nostalgia of our time. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Adrian-Ken Rueegsegger on 18 May 2010 04:30
Matt Mackall schrieb: > On Mon, 2010-05-17 at 23:50 +0200, Adrian-Ken Rueegsegger wrote: >> Geert Uytterhoeven wrote: >>> On Fri, Mar 19, 2010 at 02:33, Herbert Xu <herbert(a)gondor.apana.org.au> wrote: >>>> On Thu, Mar 18, 2010 at 10:24:41PM +0100, michael-dev(a)fami-braun.de wrote: >>>>> Pekka Enberg schrieb: >>>>>> Even with CONFIG_DEBUG_SLAB enabled or with CONFIG_SLUB and >>>>>> CONFIG_SLUB_DEBUG_ON? >>>>> no, these options have not been / are not enabled. >>>> Can you please try it with those options enabled? That will tell >>>> us if there is some latent bug in the crypto layer that only shows >>>> up right away under SLOB. >>> After seeing a posting from Nemoto-san on the linux-mips list (should >>> show up soon on >>> http://www.linux-mips.org/archives/linux-mips/2010-05/threads.html), >>> I'm wondering if these defaults are the culprit; >>> >>> mm/slab.c:#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long long) >>> mm/slob.c:#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long) >>> mm/slub.c:#define ARCH_KMALLOC_MINALIGN __alignof__(unsigned long long) > > Seems the target here is x86, which has no DMA or other alignment > constraints. > >> Just a quick note: I changed this line in mm/slob.c to (unsigned long >> long) and was able to load tcrypt without the kernel oopsing or >> panic'ing. When running all test via >> >> modprobe tcrypt > > Hmmm, sounds like something in crypto has a bogus alignment/padding > expectation. Surprised it wasn't caught by SLAB redzoning though. > Perhaps we can 'bisect' the test suite to narrow it down to a particular > test. Alternately, we can tweak SLAB to offset buffers by two bytes and > see what breaks. As noted in my other mail [1] it seems like the HMAC tests trigger these errors. Regards, Adrian [1] - http://lkml.org/lkml/2010/5/14/199 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |