[BUG] v6.12.4 copy_page_to_iter WARNING

[BUG] v6.12.4 copy_page_to_iter WARNING

[Vincent Donnefort]

Hi,

Running v6.12.4, I can trigger the following WARNING

[   62.374530] WARNING: CPU: 0 PID: 133 at lib/iov_iter.c:344 copy_page_to_iter+0x44/0x148
[   62.375388] Modules linked in:
[   62.375747] CPU: 0 UID: 0 PID: 133 Comm: grep Tainted: G      D 6.12.4 #51
[   62.376452] Tainted: [D]=DIE
[   62.376732] Hardware name: linux,dummy-virt (DT)
[   62.377147] pstate: 21400005 (nzCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
[   62.377794] pc : copy_page_to_iter+0x44/0x148
[   62.378196] lr : pipe_read+0x104/0x3d0
[   62.378530] sp : ffff800083b7bc00
[   62.378816] x29: ffff800083b7bc00 x28: 0000000000001000 x27: fff00000c21f73c0
[   62.379494] x26: ffff800083b7bd60 x25: 0000000000000014 x24: 0000000000001000
[   62.380171] x23: 0000000000000000 x22: fff00000c21f73e0 x21: 0000000000000015
[   62.380891] x20: 0000000000001000 x19: 0000000000000000 x18: 0000000000000000
[   62.381536] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000
[   62.382185] x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000000
[   62.382831] x11: 0000000000000000 x10: 0000000000000000 x9 : 0000000000000000
[   62.383478] x8 : fff00000c1603400 x7 : 0000000000000000 x6 : 00000000000000a0
[   62.384153] x5 : 0000000000000000 x4 : ffffc1ffc30867c0 x3 : 0000000000001000
[   62.384798] x2 : 000000000000103f x1 : 000000000000003f x0 : ffffc1ffc30867c0
[   62.385441] Call trace:
[   62.385684]  copy_page_to_iter+0x44/0x148
[   62.386091]  pipe_read+0x104/0x3d0
[   62.386413]  vfs_read+0x294/0x2fc
[   62.386728]  ksys_read+0xdc/0xf4
[   62.387025]  __arm64_sys_read+0x1c/0x28
[   62.387400]  invoke_syscall+0x48/0x110
[   62.387759]  el0_svc_common.constprop.0+0x40/0xe0
[   62.388178]  do_el0_svc+0x1c/0x28
[   62.388525]  el0_svc+0x30/0xd0
[   62.388818]  el0t_64_sync_handler+0xc0/0xc4
[   62.389233]  el0t_64_sync+0x190/0x194
[   62.389663] ---[ end trace 0000000000000000 ]---                 

With trace_pipe and grep.

 $ cat trace_pipe | grep foo

   and 

 $ while true; do echo "foo" > trace_marker; done

Is that a known problem? Eventually this ends up in a more serious BUG such as

   BUG: Bad page map in process cat  pte:72616d5f676e6963 pmd:800000102279003

Or.

   Unable to handle kernel paging request at virtual address fff82e2dee2e2d80

Re: [BUG] v6.12.4 copy_page_to_iter WARNING

[Steven Rostedt]

On Tue, 10 Dec 2024 15:20:25 +0000
Vincent Donnefort <vdonnefort@google.com> wrote:

> Hi,
> 
> Running v6.12.4, I can trigger the following WARNING

Hi Vincent,

This is a very old email, but I just found it buried in my inbox ;-)

Are you still able to hit this? I just tried, and can't on the latest
kernel.

-- Steve

> 
> [   62.374530] WARNING: CPU: 0 PID: 133 at lib/iov_iter.c:344 copy_page_to_iter+0x44/0x148
> [   62.375388] Modules linked in:
> [   62.375747] CPU: 0 UID: 0 PID: 133 Comm: grep Tainted: G      D 6.12.4 #51
> [   62.376452] Tainted: [D]=DIE
> [   62.376732] Hardware name: linux,dummy-virt (DT)
> [   62.377147] pstate: 21400005 (nzCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
> [   62.377794] pc : copy_page_to_iter+0x44/0x148
> [   62.378196] lr : pipe_read+0x104/0x3d0
> [   62.378530] sp : ffff800083b7bc00
> [   62.378816] x29: ffff800083b7bc00 x28: 0000000000001000 x27: fff00000c21f73c0
> [   62.379494] x26: ffff800083b7bd60 x25: 0000000000000014 x24: 0000000000001000
> [   62.380171] x23: 0000000000000000 x22: fff00000c21f73e0 x21: 0000000000000015
> [   62.380891] x20: 0000000000001000 x19: 0000000000000000 x18: 0000000000000000
> [   62.381536] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000
> [   62.382185] x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000000
> [   62.382831] x11: 0000000000000000 x10: 0000000000000000 x9 : 0000000000000000
> [   62.383478] x8 : fff00000c1603400 x7 : 0000000000000000 x6 : 00000000000000a0
> [   62.384153] x5 : 0000000000000000 x4 : ffffc1ffc30867c0 x3 : 0000000000001000
> [   62.384798] x2 : 000000000000103f x1 : 000000000000003f x0 : ffffc1ffc30867c0
> [   62.385441] Call trace:
> [   62.385684]  copy_page_to_iter+0x44/0x148
> [   62.386091]  pipe_read+0x104/0x3d0
> [   62.386413]  vfs_read+0x294/0x2fc
> [   62.386728]  ksys_read+0xdc/0xf4
> [   62.387025]  __arm64_sys_read+0x1c/0x28
> [   62.387400]  invoke_syscall+0x48/0x110
> [   62.387759]  el0_svc_common.constprop.0+0x40/0xe0
> [   62.388178]  do_el0_svc+0x1c/0x28
> [   62.388525]  el0_svc+0x30/0xd0
> [   62.388818]  el0t_64_sync_handler+0xc0/0xc4
> [   62.389233]  el0t_64_sync+0x190/0x194
> [   62.389663] ---[ end trace 0000000000000000 ]---                 
> 
> With trace_pipe and grep.
> 
>  $ cat trace_pipe | grep foo
> 
>    and 
> 
>  $ while true; do echo "foo" > trace_marker; done
> 
> Is that a known problem? Eventually this ends up in a more serious BUG such as
> 
>    BUG: Bad page map in process cat  pte:72616d5f676e6963 pmd:800000102279003
> 
> Or.
> 
>    Unable to handle kernel paging request at virtual address fff82e2dee2e2d80