* vmlinux.lds.h: Bug report: unable to handle page fault when start the virtual machine with qemu
@ 2022-11-07 3:27 zhaowenhui (A)
2022-11-07 16:00 ` Ard Biesheuvel
0 siblings, 1 reply; 6+ messages in thread
From: zhaowenhui (A) @ 2022-11-07 3:27 UTC (permalink / raw)
To: Arnd Bergmann, Kees Cook, Nathan Chancellor,
open list:GENERIC INCLUDE/ASM HEADER FILES, open list
Cc: xiafukun, yusongping
Hello,
We compiled the kernel with x86_64_defconfig and the following configs
from commit d4c6399900364facd84c9e35ce1540b6046c345f (vmlinux.lds.h:
Avoid orphan section with !SMP ):
CONFIG_SMP=n
CONFIG_AMD_MEM_ENCRYPT=y
CONFIG_HYPERVISOR_GUEST=y
CONFIG_KVM=y
CONFIG_PARAVIRT=y
Then start virtual machine with the following command (OS: Ubuntu; Arch:
x86-64):
qemu-system-x86_64 -enable-kvm -cpu Skylake-Server -smp 10 -m 8192
-boot menu=on,splash-time=1000 \
-device virtio-scsi-pci \
-initrd ${initramfs} \
-kernel ./linux/arch/x86/boot/bzImage \
-append "root=/dev/ram rw rdinit=/sbin/init console=tty0
console=ttyS0,115200 earlyprintk=ttyS0 debug " \
-nographic -vnc :18
(Note: ./linux/arch/x86/boot/bzImage is the compiled kernel bzImage path
On my machine, initramfs=./x86_procfs.cpio.gz_1 )
QEMU reports an error: BUG: unable to handle page fault for address:
ffffffff8ad01040
The bug was introduced by commit d4c6399900, and the problem can be
avoided by rolling back the patch.
Patch link:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=d4c6399900364facd84c9e35ce1540b6046c345f.
We speculate that the problem is related to the hardware memory
encryption feature in the virtualization scenario of the AMD platform.
# Error log:
[ 0.000000] Linux version 6.1.0-rc3+ (root@cgsubuntu13) (gcc (Ubuntu
7.5.0-6ubuntu2) 2
[ 0.000000] DMI: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
rel-1.15.0-0-g2dd4b9b3f4
[ 0.000000] Hypervisor detected: KVM
[ 0.017457] found SMP MP-table at [mem 0x000f5b90-0x000f5b9f]
[ 0.018300] Using GB pages for direct mapping
[ 0.162066] Memory: 7783780K/8388088K available (16398K kernel code,
2819K rwdata, 40)
[ 0.164323] SLUB: HWalign=64, Order=0-3, MinObjects=0, CPUs=1, Nodes=1
[ 0.165253] Kernel/User page tables isolation: enabled
[ 0.252783] CPU: Intel Xeon Processor (Skylake) (family: 0x6, model:
0x55, stepping: )
[ 1.535968] PM: Magic number: 14:957:648
[ 1.537003] acpi LNXCPU:01: hash matches
[ 1.537857] printk: console [netcon0] enabled
[ 1.538766] netconsole: network logging started
[ 1.540923] cfg80211: Loading compiled-in X.509 certificates for
regulatory database
[ 2.078792] input: ImExPS/2 Generic Explorer Mouse as
/devices/platform/i8042/serio1/3
[ 6.876322] Freeing initrd memory: 358780K
[ 6.878474] modprobe (63) used greatest stack depth: 13192 bytes left
[ 6.880931] cfg80211: Loaded X.509 cert 'sforshee: 00b28ddf47aef9cea7'
[ 6.882256] platform regulatory.0: Direct firmware load for
regulatory.db failed with2
[ 6.883983] ALSA device list:
[ 6.884646] No soundcards found.
[ 6.885399] cfg80211: failed to load regulatory.db
[ 6.887176] Freeing unused decrypted memory: 2036K
[ 6.888195] Freeing unused kernel image (initmem) memory: 1360K
[ 6.889804] BUG: unable to handle page fault for address:
ffffffff8ad01040
[ 6.890856] #PF: supervisor write access in kernel mode
[ 6.891663] #PF: error_code(0x0002) - not-present page
[ 6.892459] PGD 1ee25067 P4D 1ee25067 PUD 1ee26063 PMD 100105063 PTE
800fffffe08fe062
[ 6.893668] Oops: 0002 [#1] PREEMPT PTI
[ 6.894309] CPU: 0 PID: 0 Comm: swapper Not tainted 6.1.0-rc3+ #4
[ 6.895244] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996),
BIOS rel-1.15.0-0-4
[ 6.896924] RIP: 0010:kvm_guest_apic_eoi_write+0x0/0x30
[ 6.897740] Code: 59 41 58 5f 5e 5a 59 c3 cc cc cc cc 66 66 2e 0f 1f
84 00 00 00 00 00
[ 6.900388] RSP: 0018:ffffab8680003fd8 EFLAGS: 00010046
[ 6.901198] RAX: ffffffff88eb4c80 RBX: 0000000000000000 RCX:
0000000000000017
[ 6.902253] RDX: 0000000000000000 RSI: 0000000000000000 RDI:
00000000000000b0
[ 6.903327] RBP: ffffffff8a403de8 R08: 0000000413a0c8af R09:
0000000000000000
[ 6.904398] R10: 0000000000000000 R11: ffffab8680003ff8 R12:
0000000000000000
[ 6.905466] R13: 0000000000000000 R14: 0000000000000000 R15:
0000000000000000
[ 6.906527] FS: 0000000000000000(0000) GS:ffffffff8a465000(0000)
knlGS:00000000000000
[ 6.907769] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 6.908648] CR2: ffffffff8ad01040 CR3: 000000001ee22001 CR4:
00000000007706f0
[ 6.909712] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[ 6.910778] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7:
0000000000000400
[ 6.911845] PKRU: 55555554
[ 6.912341] Call Trace:
[ 6.912808] <IRQ>
[ 6.913209] __sysvec_apic_timer_interrupt+0x41/0x120
[ 6.914013] sysvec_apic_timer_interrupt+0x82/0xb0
[ 6.914784] </IRQ>
[ 6.915203] <TASK>
[ 6.915624] asm_sysvec_apic_timer_interrupt+0x16/0x20
[ 6.916434] RIP: 0010:default_idle+0xb/0x10
[ 6.917114] Code: ff ff 4c 89 f7 e8 25 e1 89 ff e9 26 ff ff ff e8 9b
72 ff ff cc cc c2
[ 6.919764] RSP: 0018:ffffffff8a403e98 EFLAGS: 00000246
[ 6.920595] RAX: ffffffff89ac49c0 RBX: 0000000000000000 RCX:
0000000000000000
[ 6.921677] RDX: 0000000000000000 RSI: 0000000000000002 RDI:
0000000000000004
[ 6.922745] RBP: 0000000000000000 R08: 00000004139503a8 R09:
0000000000000001
[ 6.923880] R10: ffffab8680013e48 R11: ffffab8680013da0 R12:
ffffffffffffffff
[ 6.924948] R13: 0000000000000000 R14: ffffffff8a42aa00 R15:
0000000000000000
[ 6.926020] ? __cpuidle_text_start+0x8/0x8
[ 6.926705] ? __cpuidle_text_start+0x8/0x8
[ 6.927393] default_idle_call+0x28/0xb0
[ 6.928049] do_idle+0x172/0x200
[ 6.928613] cpu_startup_entry+0xa/0x10
[ 6.929250] rest_init+0xb9/0xc0
[ 6.929818] arch_call_rest_init+0x5/0xa
[ 6.930470] start_kernel+0x651/0x67c
[ 6.931092] secondary_startup_64_no_verify+0xe5/0xeb
[ 6.931893] </TASK>
[ 6.932321] Modules linked in:
[ 6.932859] CR2: ffffffff8ad01040
[ 6.933435] ---[ end trace 0000000000000000 ]---
[ 6.934179] RIP: 0010:kvm_guest_apic_eoi_write+0x0/0x30
[ 6.934998] Code: 59 41 58 5f 5e 5a 59 c3 cc cc cc cc 66 66 2e 0f 1f
84 00 00 00 00 00
[ 6.937626] RSP: 0018:ffffab8680003fd8 EFLAGS: 00010046
[ 6.938440] RAX: ffffffff88eb4c80 RBX: 0000000000000000 RCX:
0000000000000017
[ 6.939509] RDX: 0000000000000000 RSI: 0000000000000000 RDI:
00000000000000b0
[ 6.940575] RBP: ffffffff8a403de8 R08: 0000000413a0c8af R09:
0000000000000000
[ 6.941639] R10: 0000000000000000 R11: ffffab8680003ff8 R12:
0000000000000000
[ 6.942698] R13: 0000000000000000 R14: 0000000000000000 R15:
0000000000000000
[ 6.943789] FS: 0000000000000000(0000) GS:ffffffff8a465000(0000)
knlGS:00000000000000
[ 6.945026] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 6.945915] CR2: ffffffff8ad01040 CR3: 000000001ee22001 CR4:
00000000007706f0
[ 6.946991] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[ 6.948061] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7:
0000000000000400
[ 6.949122] PKRU: 55555554
[ 6.949613] Kernel panic - not syncing: Fatal exception in interrupt
[ 6.950575] Kernel Offset: 0x7e00000 from 0xffffffff81000000
(relocation range: 0xfff)
[ 6.952157] ---[ end Kernel panic - not syncing: Fatal exception in
interrupt ]---
(Note: In some environments, different errors may occur as follows:
[ C0] BUG: stack guard page was hit at 00000000eb1b929b (stack is
0000000034e7e985..000000005beaadc6)
[ C0] kernel stack overflow (double-fault): 0000 [#1] PTI )
Regards
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: vmlinux.lds.h: Bug report: unable to handle page fault when start the virtual machine with qemu
2022-11-07 3:27 vmlinux.lds.h: Bug report: unable to handle page fault when start the virtual machine with qemu zhaowenhui (A)
@ 2022-11-07 16:00 ` Ard Biesheuvel
2022-11-08 7:46 ` xiafukun
0 siblings, 1 reply; 6+ messages in thread
From: Ard Biesheuvel @ 2022-11-07 16:00 UTC (permalink / raw)
To: zhaowenhui (A)
Cc: Arnd Bergmann, Kees Cook, Nathan Chancellor,
open list:GENERIC INCLUDE/ASM HEADER FILES, open list, xiafukun,
yusongping
On Mon, 7 Nov 2022 at 04:27, zhaowenhui (A) <zhaowenhui8@huawei.com> wrote:
>
> Hello,
>
> We compiled the kernel with x86_64_defconfig and the following configs
> from commit d4c6399900364facd84c9e35ce1540b6046c345f (vmlinux.lds.h:
> Avoid orphan section with !SMP ):
>
> CONFIG_SMP=n
> CONFIG_AMD_MEM_ENCRYPT=y
> CONFIG_HYPERVISOR_GUEST=y
> CONFIG_KVM=y
> CONFIG_PARAVIRT=y
>
> Then start virtual machine with the following command (OS: Ubuntu; Arch:
> x86-64):
>
> qemu-system-x86_64 -enable-kvm -cpu Skylake-Server -smp 10 -m 8192
> -boot menu=on,splash-time=1000 \
> -device virtio-scsi-pci \
> -initrd ${initramfs} \
> -kernel ./linux/arch/x86/boot/bzImage \
> -append "root=/dev/ram rw rdinit=/sbin/init console=tty0
> console=ttyS0,115200 earlyprintk=ttyS0 debug " \
> -nographic -vnc :18
>
> (Note: ./linux/arch/x86/boot/bzImage is the compiled kernel bzImage path
> On my machine, initramfs=./x86_procfs.cpio.gz_1 )
>
> QEMU reports an error: BUG: unable to handle page fault for address:
> ffffffff8ad01040
>
> The bug was introduced by commit d4c6399900, and the problem can be
> avoided by rolling back the patch.
> Patch link:
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=d4c6399900364facd84c9e35ce1540b6046c345f.
> We speculate that the problem is related to the hardware memory
> encryption feature in the virtualization scenario of the AMD platform.
>
That patch looks incorrect to me. Without CONFIG_SMP, the PERCPU
sections are not instantiated, and the only copy of those variables is
created in the ordinary .data/.bss sections
Does the change below fix the issue for you?
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -347,6 +347,7 @@
#define DATA_DATA \
*(.xiptext) \
*(DATA_MAIN) \
+ *(.data..decrypted) \
*(.ref.data) \
*(.data..shared_aligned) /* percpu related */ \
MEM_KEEP(init.data*) \
@@ -995,7 +996,6 @@
#ifdef CONFIG_AMD_MEM_ENCRYPT
#define PERCPU_DECRYPTED_SECTION \
. = ALIGN(PAGE_SIZE); \
- *(.data..decrypted) \
*(.data..percpu..decrypted) \
. = ALIGN(PAGE_SIZE);
#else
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: vmlinux.lds.h: Bug report: unable to handle page fault when start the virtual machine with qemu
2022-11-07 16:00 ` Ard Biesheuvel
@ 2022-11-08 7:46 ` xiafukun
2022-11-08 14:35 ` Nathan Chancellor
0 siblings, 1 reply; 6+ messages in thread
From: xiafukun @ 2022-11-08 7:46 UTC (permalink / raw)
To: Ard Biesheuvel
Cc: arnd, keescook, nathan, linux-arch, linux-kernel, yusongping,
zhaowenhui8
Thank you for your reply.
We tested your changes to this patch and did fix the issue. Following the
solution you provided, we recompile the kernel and successfully start the
virtual machine.
在 2022/11/8 0:00, Ard Biesheuvel 写道:
>
> That patch looks incorrect to me. Without CONFIG_SMP, the PERCPU
> sections are not instantiated, and the only copy of those variables is
> created in the ordinary .data/.bss sections
>
> Does the change below fix the issue for you?
>
> --- a/include/asm-generic/vmlinux.lds.h
> +++ b/include/asm-generic/vmlinux.lds.h
> @@ -347,6 +347,7 @@
> #define DATA_DATA \
> *(.xiptext) \
> *(DATA_MAIN) \
> + *(.data..decrypted) \
> *(.ref.data) \
> *(.data..shared_aligned) /* percpu related */ \
> MEM_KEEP(init.data*) \
> @@ -995,7 +996,6 @@
> #ifdef CONFIG_AMD_MEM_ENCRYPT
> #define PERCPU_DECRYPTED_SECTION \
> . = ALIGN(PAGE_SIZE); \
> - *(.data..decrypted) \
> *(.data..percpu..decrypted) \
> . = ALIGN(PAGE_SIZE);
> #else
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: vmlinux.lds.h: Bug report: unable to handle page fault when start the virtual machine with qemu
2022-11-08 7:46 ` xiafukun
@ 2022-11-08 14:35 ` Nathan Chancellor
2022-11-08 15:14 ` Ard Biesheuvel
0 siblings, 1 reply; 6+ messages in thread
From: Nathan Chancellor @ 2022-11-08 14:35 UTC (permalink / raw)
To: xiafukun, Ard Biesheuvel
Cc: arnd, keescook, linux-arch, linux-kernel, yusongping, zhaowenhui8
On Tue, Nov 08, 2022 at 03:46:32PM +0800, xiafukun wrote:
> Thank you for your reply.
> We tested your changes to this patch and did fix the issue. Following the
> solution you provided, we recompile the kernel and successfully start the
> virtual machine.
Thank you a lot for testing and sorry about the breakage in the first
place :(
Ard, were you going to send a patch? Feel free to preemptively add:
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
if so; otherwise, I can send one later today.
> 在 2022/11/8 0:00, Ard Biesheuvel 写道:
> >
> > That patch looks incorrect to me. Without CONFIG_SMP, the PERCPU
> > sections are not instantiated, and the only copy of those variables is
> > created in the ordinary .data/.bss sections
> >
> > Does the change below fix the issue for you?
> >
> > --- a/include/asm-generic/vmlinux.lds.h
> > +++ b/include/asm-generic/vmlinux.lds.h
> > @@ -347,6 +347,7 @@
> > #define DATA_DATA \
> > *(.xiptext) \
> > *(DATA_MAIN) \
> > + *(.data..decrypted) \
> > *(.ref.data) \
> > *(.data..shared_aligned) /* percpu related */ \
> > MEM_KEEP(init.data*) \
> > @@ -995,7 +996,6 @@
> > #ifdef CONFIG_AMD_MEM_ENCRYPT
> > #define PERCPU_DECRYPTED_SECTION \
> > . = ALIGN(PAGE_SIZE); \
> > - *(.data..decrypted) \
> > *(.data..percpu..decrypted) \
> > . = ALIGN(PAGE_SIZE);
> > #else
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: vmlinux.lds.h: Bug report: unable to handle page fault when start the virtual machine with qemu
2022-11-08 14:35 ` Nathan Chancellor
@ 2022-11-08 15:14 ` Ard Biesheuvel
2022-11-08 15:16 ` Nathan Chancellor
0 siblings, 1 reply; 6+ messages in thread
From: Ard Biesheuvel @ 2022-11-08 15:14 UTC (permalink / raw)
To: Nathan Chancellor
Cc: xiafukun, arnd, keescook, linux-arch, linux-kernel, yusongping,
zhaowenhui8
On Tue, 8 Nov 2022 at 15:35, Nathan Chancellor <nathan@kernel.org> wrote:
>
> On Tue, Nov 08, 2022 at 03:46:32PM +0800, xiafukun wrote:
> > Thank you for your reply.
> > We tested your changes to this patch and did fix the issue. Following the
> > solution you provided, we recompile the kernel and successfully start the
> > virtual machine.
>
> Thank you a lot for testing and sorry about the breakage in the first
> place :(
>
> Ard, were you going to send a patch? Feel free to preemptively add:
>
> Reviewed-by: Nathan Chancellor <nathan@kernel.org>
>
> if so; otherwise, I can send one later today.
>
If you don't mind?
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: vmlinux.lds.h: Bug report: unable to handle page fault when start the virtual machine with qemu
2022-11-08 15:14 ` Ard Biesheuvel
@ 2022-11-08 15:16 ` Nathan Chancellor
0 siblings, 0 replies; 6+ messages in thread
From: Nathan Chancellor @ 2022-11-08 15:16 UTC (permalink / raw)
To: Ard Biesheuvel
Cc: xiafukun, arnd, keescook, linux-arch, linux-kernel, yusongping,
zhaowenhui8
On Tue, Nov 08, 2022 at 04:14:44PM +0100, Ard Biesheuvel wrote:
> On Tue, 8 Nov 2022 at 15:35, Nathan Chancellor <nathan@kernel.org> wrote:
> >
> > On Tue, Nov 08, 2022 at 03:46:32PM +0800, xiafukun wrote:
> > > Thank you for your reply.
> > > We tested your changes to this patch and did fix the issue. Following the
> > > solution you provided, we recompile the kernel and successfully start the
> > > virtual machine.
> >
> > Thank you a lot for testing and sorry about the breakage in the first
> > place :(
> >
> > Ard, were you going to send a patch? Feel free to preemptively add:
> >
> > Reviewed-by: Nathan Chancellor <nathan@kernel.org>
> >
> > if so; otherwise, I can send one later today.
> >
>
> If you don't mind?
Not at all, since it is my issue and you did the hard work for me :)
Will send it out in a little bit, thanks again for the assist!
Cheers,
Nathan
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2022-11-08 15:18 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-11-07 3:27 vmlinux.lds.h: Bug report: unable to handle page fault when start the virtual machine with qemu zhaowenhui (A)
2022-11-07 16:00 ` Ard Biesheuvel
2022-11-08 7:46 ` xiafukun
2022-11-08 14:35 ` Nathan Chancellor
2022-11-08 15:14 ` Ard Biesheuvel
2022-11-08 15:16 ` Nathan Chancellor
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.