Messages in this thread Patch in this message | | | Date | Tue, 22 Sep 2009 18:08:39 +0900 | From | Tejun Heo <> | Subject | Re: 2.6.31-07068-g43c1266 Early boot exception |
| |
Tony Vroon wrote: > Thanks Ingo. 11-step bisection reveals: > > chainsaw@prometheus /cvs/linux-2.6 $ git bisect bad > fd1e8a1fe2b54df6c185b4fa65f181f50b9c4d4e is the first bad commit > commit fd1e8a1fe2b54df6c185b4fa65f181f50b9c4d4e > Author: Tejun Heo <tj@kernel.org> > Date: Fri Aug 14 15:00:51 2009 +0900 > > percpu: introduce pcpu_alloc_info and pcpu_group_info > > I will repeat the early exception here: > IOAPIC[0]: apic_id 0, version 17, address 0xfec00000, GSI 0-23 > ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl) > ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 9 high level) > ACPI: INT_SRC_OVR (bus 0 bus_irq 14 global_irq 14 high edge) > ACPI: INT_SRC_OVR (bus 0 bus_irq 15 global_irq 15 high edge) > Using ACPI (MADT) for SMP configuration information > ACPI: HPET id: 0x10de8201 base: 0xfed00000 > SMP: allowing 12 CPUs, 0 hotplug CPUs > Allocating PCI resources starting at b00000000 (gap:b00000000:3000000 > NR_CPUS: 12 nr_cpumask_bits:12 nr_cpu_ids:12 nr_node_ids:2 > PERCPU: Embedded 23 pages/cpu @ffff880028200000 s73688 r0 d20520 u262 > PANIC: early exception 06 rip 10:ffffffff815a50e6 error 0 cr2 d0fff6 > Pid: 0, comm: swapper Not tained 2.6.31-07068-g43c266 #1 > Call Trace: > [<ffffffff81591175>] early_idt_handler+0x55/0x68 > [<ffffffff815a50e6>] ? pcpu_setup_first_chunk+0x40b/0x692 > [<ffffffff815a50ff>] ? pcpu_setup_first_chunk+0x424/0x692 > [<ffffffff815a5867>] pcpu_embed_first_chunk+0x1e2/0x241 > [<ffffffff81598c14>] ? pcpu_fc_alloc+0x0/0xac > [<ffffffff81598bf5>] ? pcpu_fc_free+0x0/0x1f > [<ffffffff81598a36>] setup_per_cpu_areas+0x65/0x219 > [<ffffffff815919bf>] start_kernel+0x124/0x2c5 > [<ffffffff81591252>] x86_64_start_reservations+0x82/0x86 > [<ffffffff8159133a>] x86_64_start_kernel+0xe4/0xeb > RIP pcpu_setup_first_chunk+0x40b/0x692
Can you please try the patch below? Also, please build the kernel with debug info and ask gdb which line the crash corresponds to? ie. l *pcpu_embed_first_chunk+0x1e2
Thanks.
diff --git a/mm/percpu.c b/mm/percpu.c index 43d8cac..66b7d5e 100644 --- a/mm/percpu.c +++ b/mm/percpu.c @@ -1812,6 +1812,9 @@ int __init pcpu_embed_first_chunk(size_t reserved_size, ssize_t dyn_size, /* allocate space for the whole group */ ptr = alloc_fn(cpu, gi->nr_units * ai->unit_size, atom_size); if (!ptr) { + printk("PERCPU: failed to allocate %zu bytes for " + "group %d from cpu%u\n", + gi->nr_units * ai->unit_size, group, cpu); rc = -ENOMEM; goto out_free_areas; } @@ -1844,8 +1847,9 @@ int __init pcpu_embed_first_chunk(size_t reserved_size, ssize_t dyn_size,
out_free_areas: for (group = 0; group < ai->nr_groups; group++) - free_fn(areas[group], - ai->groups[group].nr_units * ai->unit_size); + if (areas[group]) + free_fn(areas[group], + ai->groups[group].nr_units * ai->unit_size); out_free: pcpu_free_alloc_info(ai); if (areas) @@ -1956,7 +1960,8 @@ int __init pcpu_page_first_chunk(size_t reserved_size,
enomem: while (--j >= 0) - free_fn(page_address(pages[j]), PAGE_SIZE); + if (pages[j]) + free_fn(page_address(pages[j]), PAGE_SIZE); rc = -ENOMEM; out_free_ar: free_bootmem(__pa(pages), pages_size);
-- tejun
| |