lkml.org 
[lkml]   [2008]   [Sep]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
SubjectRe: 2.6.27-rc5-mm1: 3 WARN_ON dumps during boot (acpi + vmap_pte_range)
Date
On Saturday 06 September 2008 16:50, Andrew Morton wrote:
> On Sat, 6 Sep 2008 08:45:58 +0200 Krzysztof Helt <krzysztof.h1@poczta.fm>
wrote:
> > Hi,
> >
> > There is a dmesg dump below from my Compaq AP550 workstation.
> > It has 3 WARN_ON() dumps: 1 from acpi layer and 2 from vmap_pte_range()
> > There is no such thing in 2.6.27-rc4 which I use daily so I assume
> > it is something in the -mm tree.
>
> yup thanks. The acpi guys and Rusty are still scratching each others
> heads over the acpi procfs one.
>
> > Intel 82802 RNG detected
> > Linux agpgart interface v0.103
> > agpgart-intel 0000:00:00.0: Intel i840 Chipset
> > ------------[ cut here ]------------
> > WARNING: at mm/vmalloc.c:105 vmap_pte_range+0xcb/0x100()
> > Modules linked in:
> > Pid: 1, comm: swapper Tainted: G W 2.6.27-rc5-mm1 #1
> > [<c012526f>] warn_on_slowpath+0x5f/0x90
> > [<c0217a5e>] fbcon_clear+0x13e/0x1a0
> > [<c0111da1>] xapic_wait_icr_idle+0x11/0x20
> > [<c01460cc>] generic_exec_single+0x9c/0xa0
> > [<c0109776>] read_tsc+0x6/0x30
> > [<c013e4f8>] getnstimeofday+0x38/0x150
> > [<c015496c>] buffered_rmqueue+0x12c/0x210
> > [<c011b53c>] update_curr+0x4c/0x70
> > [<c016768b>] vmap_pte_range+0xcb/0x100
> > [<c01676ed>] vmap_pmd_range+0x2d/0x40
> > [<c016772d>] vmap_pud_range+0x2d/0x40
> > [<c01677a6>] vmap_page_range+0x66/0x90
> > [<c016844f>] map_vm_area+0x2f/0x50
> > [<c0168996>] __vmalloc_area_node+0xb6/0x110
> > [<c0168a95>] __vmalloc_node+0x85/0xb0
> > [<c02603a2>] agp_backend_initialize+0x82/0x230
> > [<c0168b01>] vmalloc+0x21/0x30
> > [<c02603a2>] agp_backend_initialize+0x82/0x230
> > [<c02603a2>] agp_backend_initialize+0x82/0x230
> > [<c027f790>] __driver_attach+0x0/0xc0
> > [<c02606b5>] agp_add_bridge+0x55/0x1a0
> > [<c020955a>] pci_call_probe+0xa/0x10
> > [<c02095ae>] __pci_device_probe+0x4e/0x60
> > [<c02095e6>] pci_device_probe+0x26/0x60
> > [<c027f5cb>] really_probe+0x9b/0x130
> > [<c02094b0>] pci_match_device+0x10/0xb0
> > [<c027f6af>] driver_probe_device+0x3f/0x60
> > [<c027f819>] __driver_attach+0x89/0xc0
> > [<c027e519>] bus_for_each_dev+0x39/0x60
> > [<c027f866>] driver_attach+0x16/0x20
> > [<c027f790>] __driver_attach+0x0/0xc0
> > [<c027ed29>] bus_add_driver+0x109/0x1a0
> > [<c01f85ed>] kset_find_obj+0x2d/0x60
> > [<c0209660>] pci_device_shutdown+0x0/0x20
> > [<c0209660>] pci_device_shutdown+0x0/0x20
> > [<c027fcff>] driver_register+0x3f/0xd0
> > [<c0125b17>] printk+0x17/0x20
> > [<c0482250>] agp_intel_init+0x0/0x20
> > [<c02096c7>] __pci_register_driver+0x47/0x80
> > [<c0482250>] agp_intel_init+0x0/0x20
> > [<c0101032>] _stext+0x32/0x180
> > [<c0111da1>] xapic_wait_icr_idle+0x11/0x20
> > [<c011aeb4>] resched_task+0x54/0x60
> > [<c011e76e>] try_to_wake_up+0x12e/0x140
> > [<c0135211>] __create_workqueue_key+0x131/0x150
> > [<c046a8d3>] do_initcalls+0x53/0xd0
> > [<c046a9a0>] kernel_init+0x0/0xb0
> > [<c046a9f5>] kernel_init+0x55/0xb0
> > [<c0103f3b>] kernel_thread_helper+0x7/0x1c
> > =======================
> > ---[ end trace 4eaa2a86a8e2da22 ]---
> > agpgart-intel 0000:00:00.0: can't allocate memory for key lists
> > agpgart-intel 0000:00:00.0: agp_backend_initialize() failed
> > agpgart-intel: probe of 0000:00:00.0 failed with error -12
>
> This one I reported to Nick and Dave yesterday - no response as yet.
>
> > ------------[ cut here ]------------
> > WARNING: at mm/vmalloc.c:105 vmap_pte_range+0xcb/0x100()
> > Modules linked in:
> > Pid: 776, comm: modprobe Tainted: G W 2.6.27-rc5-mm1 #1
> > [<c012526f>] warn_on_slowpath+0x5f/0x90
> > [<c01617e9>] handle_mm_fault+0x1d9/0x210
> > [<c01bd0a0>] ext3_get_block+0x0/0x100
> > [<c011759b>] do_page_fault+0x12b/0x4f0
> > [<c0105dde>] do_IRQ+0x7e/0xd0
> > [<c0154ce4>] __alloc_pages_internal+0x94/0x410
> > [<c0117470>] do_page_fault+0x0/0x4f0
> > [<c038506a>] error_code+0x72/0x78
> > [<c015496c>] buffered_rmqueue+0x12c/0x210
> > [<c014fbab>] find_get_page+0x2b/0xa0
> > [<c016768b>] vmap_pte_range+0xcb/0x100
> > [<c01676ed>] vmap_pmd_range+0x2d/0x40
> > [<c016772d>] vmap_pud_range+0x2d/0x40
> > [<c01677a6>] vmap_page_range+0x66/0x90
> > [<c016844f>] map_vm_area+0x2f/0x50
> > [<c0168996>] __vmalloc_area_node+0xb6/0x110
> > [<c0168a95>] __vmalloc_node+0x85/0xb0
> > [<c0148eb9>] load_module+0x89/0xd90
> > [<c0168b01>] vmalloc+0x21/0x30
> > [<c0148eb9>] load_module+0x89/0xd90
> > [<c0148eb9>] load_module+0x89/0xd90
> > [<c01380e0>] autoremove_wake_function+0x0/0x50
> > [<c0177db0>] vfs_read+0x90/0x150
> > [<c0149c44>] sys_init_module+0x44/0x1a0
> > [<c0178111>] sys_read+0x41/0x70
> > [<c01031fa>] syscall_call+0x7/0xb
> > =======================
> > ---[ end trace 4eaa2a86a8e2da22 ]---
> > Intel ICH 0000:00:1f.5: PCI INT B -> GSI 17 (level, low) -> IRQ 17
> > Intel ICH 0000:00:1f.5: setting latency timer to 64
>
> That's coming out of the module loader and is a new one. It's the same
>
> warning as the agp one:
> : static int vmap_pte_range(pmd_t *pmd, unsigned long addr,
> : unsigned long end, pgprot_t prot, struct page **pages, int *nr)
> : {
> : pte_t *pte;
> :
> : /*
> : * nr is a running index into the array which helps higher level
> : * callers keep track of where we're up to.
> : */
> :
> : pte = pte_alloc_kernel(pmd, addr);
> : if (!pte)
> : return -ENOMEM;
> : do {
> : struct page *page = pages[*nr];
> :
> : -->> if (WARN_ON(!pte_none(*pte)))
> : return -EBUSY;
> : if (WARN_ON(!page))
> : return -ENOMEM;
> : set_pte_at(&init_mm, addr, pte, mk_pte(page, prot));
> : (*nr)++;
> : } while (pte++, addr += PAGE_SIZE, addr != end);
> : return 0;
> : }
>
> I'm suspecting an overactive assertion in the new vmap code?

OK, would it be possible to test the following patch on the failing
machine(s), and send me the complete dmesg trace afterwards, please?

The patch does a little bit of extra page table checking, and also
prints a trace of operations on the vmap-space.

Thanks,
Nick

Index: linux-2.6/mm/vmalloc.c
===================================================================
--- linux-2.6.orig/mm/vmalloc.c
+++ linux-2.6/mm/vmalloc.c
@@ -31,6 +31,61 @@

/*** Page table manipulation functions ***/

+static void check_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end)
+{
+ pte_t *pte;
+
+ pte = pte_offset_kernel(pmd, addr);
+ do {
+ WARN_ON_ONCE(!pte_none(*pte));
+ } while (pte++, addr += PAGE_SIZE, addr != end);
+}
+
+static void check_pmd_range(pud_t *pud, unsigned long addr, unsigned long end)
+{
+ pmd_t *pmd;
+ unsigned long next;
+
+ pmd = pmd_offset(pud, addr);
+ do {
+ next = pmd_addr_end(addr, end);
+ if (pmd_none_or_clear_bad(pmd))
+ continue;
+ check_pte_range(pmd, addr, next);
+ } while (pmd++, addr = next, addr != end);
+}
+
+static void check_pud_range(pgd_t *pgd, unsigned long addr, unsigned long end)
+{
+ pud_t *pud;
+ unsigned long next;
+
+ pud = pud_offset(pgd, addr);
+ do {
+ next = pud_addr_end(addr, end);
+ if (pud_none_or_clear_bad(pud))
+ continue;
+ check_pmd_range(pud, addr, next);
+ } while (pud++, addr = next, addr != end);
+}
+
+static void check_page_range(unsigned long addr, unsigned long end)
+{
+ pgd_t *pgd;
+ unsigned long next;
+
+ BUG_ON(addr >= end);
+ pgd = pgd_offset_k(addr);
+ flush_cache_vunmap(addr, end);
+ do {
+ next = pgd_addr_end(addr, end);
+ if (pgd_none_or_clear_bad(pgd))
+ continue;
+ check_pud_range(pgd, addr, next);
+ } while (pgd++, addr = next, addr != end);
+}
+
+
static void vunmap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end)
{
pte_t *pte;
@@ -75,6 +130,8 @@ static void vunmap_page_range(unsigned l
pgd_t *pgd;
unsigned long next;

+ printk("vunmap_page_range (%lx-%lx size=%lx)\n", addr, end, end-addr);
+
BUG_ON(addr >= end);
pgd = pgd_offset_k(addr);
flush_cache_vunmap(addr, end);
@@ -160,6 +217,8 @@ static int vmap_page_range(unsigned long
int err = 0;
int nr = 0;

+ printk("vmap_page_range (%lx-%lx size=%lx)\n", addr, end, end-addr);
+
BUG_ON(addr >= end);
pgd = pgd_offset_k(addr);
do {
@@ -371,6 +430,10 @@ found:
va->va_end = addr + size;
va->flags = 0;
__insert_vmap_area(va);
+
+ printk("alloc_vmap_area within(%lx-%lx) size=%lx returns=(%lx-%lx)\n", vstart, vend, size, addr, addr+size);
+ check_page_range(va->va_start, va->va_end);
+
spin_unlock(&vmap_area_lock);

return va;
@@ -385,6 +448,9 @@ static void rcu_free_va(struct rcu_head

static void __free_vmap_area(struct vmap_area *va)
{
+ printk("free_vmap_area (%lx-%lx size=%lx)\n", va->va_start, va->va_end, va->va_end - va->va_start);
+ check_page_range(va->va_start, va->va_end);
+
BUG_ON(RB_EMPTY_NODE(&va->rb_node));
rb_erase(&va->rb_node, &vmap_area_root);
RB_CLEAR_NODE(&va->rb_node);
\
 
 \ /
  Last update: 2008-09-08 11:39    [W:0.062 / U:0.076 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site