Re: arm64 crashkernel fails to boot on acpi-only machines due to ACPI regions being no longer mapped as NOMAP

From: AKASHI Takahiro
Date: Tue Dec 19 2017 - 00:21:33 EST


On Tue, Dec 19, 2017 at 02:58:20AM +0530, Bhupesh Sharma wrote:
> Hi Dave,
>
> On Mon, Dec 18, 2017 at 10:46 AM, Dave Young <dyoung@xxxxxxxxxx> wrote:
> > kexec@xxxxxxxxxxxxxxxx is for Fedora kexec scripts discussion, changed it
> > to kexec@xxxxxxxxxxxxxxxxxxx
> >
> > Also add linux-acpi list
> > On 12/18/17 at 02:31am, Bhupesh Sharma wrote:
> >> On Fri, Dec 15, 2017 at 3:05 PM, Ard Biesheuvel
> >> <ard.biesheuvel@xxxxxxxxxx> wrote:
> >> > On 15 December 2017 at 09:59, AKASHI Takahiro
> >> > <takahiro.akashi@xxxxxxxxxx> wrote:
> >> >> On Wed, Dec 13, 2017 at 12:17:22PM +0000, Ard Biesheuvel wrote:
> >> >>> On 13 December 2017 at 12:16, AKASHI Takahiro
> >> >>> <takahiro.akashi@xxxxxxxxxx> wrote:
> >> >>> > On Wed, Dec 13, 2017 at 10:49:27AM +0000, Ard Biesheuvel wrote:
> >> >>> >> On 13 December 2017 at 10:26, AKASHI Takahiro
> >> >>> >> <takahiro.akashi@xxxxxxxxxx> wrote:
> >> >>> >> > Bhupesh, Ard,
> >> >>> >> >
> >> >>> >> > On Wed, Dec 13, 2017 at 03:21:59AM +0530, Bhupesh Sharma wrote:
> >> >>> >> >> Hi Ard, Akashi
> >> >>> >> >>
> >> >>> >> > (snip)
> >> >>> >> >
> >> >>> >> >> Looking deeper into the issue, since the arm64 kexec-tools uses the
> >> >>> >> >> 'linux,usable-memory-range' dt property to allow crash dump kernel to
> >> >>> >> >> identify its own usable memory and exclude, at its boot time, any
> >> >>> >> >> other memory areas that are part of the panicked kernel's memory.
> >> >>> >> >> (see https://www.kernel.org/doc/Documentation/devicetree/bindings/chosen.txt
> >> >>> >> >> , for details)
> >> >>> >> >
> >> >>> >> > Right.
> >> >>> >> >
> >> >>> >> >> 1). Now when 'kexec -p' is executed, this node is patched up only
> >> >>> >> >> with the crashkernel memory range:
> >> >>> >> >>
> >> >>> >> >> /* add linux,usable-memory-range */
> >> >>> >> >> nodeoffset = fdt_path_offset(new_buf, "/chosen");
> >> >>> >> >> result = fdt_setprop_range(new_buf, nodeoffset,
> >> >>> >> >> PROP_USABLE_MEM_RANGE, &crash_reserved_mem,
> >> >>> >> >> address_cells, size_cells);
> >> >>> >> >>
> >> >>> >> >> (see https://git.kernel.org/pub/scm/utils/kernel/kexec/kexec-tools.git/tree/kexec/arch/arm64/kexec-arm64.c#n465
> >> >>> >> >> , for details)
> >> >>> >> >>
> >> >>> >> >> 2). This excludes the ACPI reclaim regions irrespective of whether
> >> >>> >> >> they are marked as System RAM or as RESERVED. As,
> >> >>> >> >> 'linux,usable-memory-range' dt node is patched up only with
> >> >>> >> >> 'crash_reserved_mem' and not 'system_memory_ranges'
> >> >>> >> >>
> >> >>> >> >> 3). As a result when the crashkernel boots up it doesn't find this
> >> >>> >> >> ACPI memory and crashes while trying to access the same:
> >> >>> >> >>
> >> >>> >> >> # kexec -p /boot/vmlinuz-`uname -r` --initrd=/boot/initramfs-`uname
> >> >>> >> >> -r`.img --reuse-cmdline -d
> >> >>> >> >>
> >> >>> >> >> [snip..]
> >> >>> >> >>
> >> >>> >> >> Reserved memory range
> >> >>> >> >> 000000000e800000-000000002e7fffff (0)
> >> >>> >> >>
> >> >>> >> >> Coredump memory ranges
> >> >>> >> >> 0000000000000000-000000000e7fffff (0)
> >> >>> >> >> 000000002e800000-000000003961ffff (0)
> >> >>> >> >> 0000000039d40000-000000003ed2ffff (0)
> >> >>> >> >> 000000003ed60000-000000003fbfffff (0)
> >> >>> >> >> 0000001040000000-0000001ffbffffff (0)
> >> >>> >> >> 0000002000000000-0000002ffbffffff (0)
> >> >>> >> >> 0000009000000000-0000009ffbffffff (0)
> >> >>> >> >> 000000a000000000-000000affbffffff (0)
> >> >>> >> >>
> >> >>> >> >> 4). So if we revert Ard's patch or just comment the fixing up of the
> >> >>> >> >> memory cap'ing passed to the crash kernel inside
> >> >>> >> >> 'arch/arm64/mm/init.c' (see below):
> >> >>> >> >>
> >> >>> >> >> static void __init fdt_enforce_memory_region(void)
> >> >>> >> >> {
> >> >>> >> >> struct memblock_region reg = {
> >> >>> >> >> .size = 0,
> >> >>> >> >> };
> >> >>> >> >>
> >> >>> >> >> of_scan_flat_dt(early_init_dt_scan_usablemem, &reg);
> >> >>> >> >>
> >> >>> >> >> if (reg.size)
> >> >>> >> >> //memblock_cap_memory_range(reg.base, reg.size); /*
> >> >>> >> >> comment this out */
> >> >>> >> >> }
> >> >>> >> >
> >> >>> >> > Please just don't do that. It can cause a fatal damage on
> >> >>> >> > memory contents of the *crashed* kernel.
> >> >>> >> >
> >> >>> >> >> 5). Both the above temporary solutions fix the problem.
> >> >>> >> >>
> >> >>> >> >> 6). However exposing all System RAM regions to the crashkernel is not
> >> >>> >> >> advisable and may cause the crashkernel or some crashkernel drivers to
> >> >>> >> >> fail.
> >> >>> >> >>
> >> >>> >> >> 6a). I am trying an approach now, where the ACPI reclaim regions are
> >> >>> >> >> added to '/proc/iomem' separately as ACPI reclaim regions by the
> >> >>> >> >> kernel code and on the other hand the user-space 'kexec-tools' will
> >> >>> >> >> pick up the ACPI reclaim regions from '/proc/iomem' and add it to the
> >> >>> >> >> dt node 'linux,usable-memory-range'
> >> >>> >> >
> >> >>> >> > I still don't understand why we need to carry over the information
> >> >>> >> > about "ACPI Reclaim memory" to crash dump kernel. In my understandings,
> >> >>> >> > such regions are free to be reused by the kernel after some point of
> >> >>> >> > initialization. Why does crash dump kernel need to know about them?
> >> >>> >> >
> >> >>> >>
> >> >>> >> Not really. According to the UEFI spec, they can be reclaimed after
> >> >>> >> the OS has initialized, i.e., when it has consumed the ACPI tables and
> >> >>> >> no longer needs them. Of course, in order to be able to boot a kexec
> >> >>> >> kernel, those regions needs to be preserved, which is why they are
> >> >>> >> memblock_reserve()'d now.
> >> >>> >
> >> >>> > For my better understandings, who is actually accessing such regions
> >> >>> > during boot time, uefi itself or efistub?
> >> >>> >
> >> >>>
> >> >>> No, only the kernel. This is where the ACPI tables are stored. For
> >> >>> instance, on QEMU we have
> >> >>>
> >> >>> ACPI: RSDP 0x0000000078980000 000024 (v02 BOCHS )
> >> >>> ACPI: XSDT 0x0000000078970000 000054 (v01 BOCHS BXPCFACP 00000001
> >> >>> 01000013)
> >> >>> ACPI: FACP 0x0000000078930000 00010C (v05 BOCHS BXPCFACP 00000001
> >> >>> BXPC 00000001)
> >> >>> ACPI: DSDT 0x0000000078940000 0011DA (v02 BOCHS BXPCDSDT 00000001
> >> >>> BXPC 00000001)
> >> >>> ACPI: APIC 0x0000000078920000 000140 (v03 BOCHS BXPCAPIC 00000001
> >> >>> BXPC 00000001)
> >> >>> ACPI: GTDT 0x0000000078910000 000060 (v02 BOCHS BXPCGTDT 00000001
> >> >>> BXPC 00000001)
> >> >>> ACPI: MCFG 0x0000000078900000 00003C (v01 BOCHS BXPCMCFG 00000001
> >> >>> BXPC 00000001)
> >> >>> ACPI: SPCR 0x00000000788F0000 000050 (v02 BOCHS BXPCSPCR 00000001
> >> >>> BXPC 00000001)
> >> >>> ACPI: IORT 0x00000000788E0000 00007C (v00 BOCHS BXPCIORT 00000001
> >> >>> BXPC 00000001)
> >> >>>
> >> >>> covered by
> >> >>>
> >> >>> efi: 0x0000788e0000-0x00007894ffff [ACPI Reclaim Memory ...]
> >> >>> ...
> >> >>> efi: 0x000078970000-0x00007898ffff [ACPI Reclaim Memory ...]
> >> >>
> >> >> OK. I mistakenly understood those regions could be freed after exiting
> >> >> UEFI boot services.
> >> >>
> >> >>>
> >> >>> >> So it seems that kexec does not honour the memblock_reserve() table
> >> >>> >> when booting the next kernel.
> >> >>> >
> >> >>> > not really.
> >> >>> >
> >> >>> >> > (In other words, can or should we skip some part of ACPI-related init code
> >> >>> >> > on crash dump kernel?)
> >> >>> >> >
> >> >>> >>
> >> >>> >> I don't think so. And the change to the handling of ACPI reclaim
> >> >>> >> regions only revealed the bug, not created it (given that other
> >> >>> >> memblock_reserve regions may be affected as well)
> >> >>> >
> >> >>> > As whether we should honor such reserved regions over kexec'ing
> >> >>> > depends on each one's specific nature, we will have to take care one-by-one.
> >> >>> > As a matter of fact, no information about "reserved" memblocks is
> >> >>> > exposed to user space (via proc/iomem).
> >> >>> >
> >> >>>
> >> >>> That is why I suggested (somewhere in this thread?) to not expose them
> >> >>> as 'System RAM'. Do you think that could solve this?
> >> >>
> >> >> Memblock-reserv'ing them is necessary to prevent their corruption and
> >> >> marking them under another name in /proc/iomem would also be good in order
> >> >> not to allocate them as part of crash kernel's memory.
> >> >>
> >> >
> >> > I agree. However, this may not be entirely trivial, since iterating
> >> > over the memblock_reserved table and creating iomem entries may result
> >> > in collisions.
> >>
> >> I found a method (using the patch I shared earlier in this thread) to mark these
> >> entries as 'ACPI reclaim memory' ranges rather than System RAM or
> >> reserved regions.
> >>
> >> >> But I'm not still convinced that we should export them in useable-
> >> >> memory-range to crash dump kernel. They will be accessed through
> >> >> acpi_os_map_memory() and so won't be required to be part of system ram
> >> >> (or memblocks), I guess.
> >> >
> >> > Agreed. They will be covered by the linear mapping in the boot kernel,
> >> > and be mapped explicitly via ioremap_cache() in the kexec kernel,
> >> > which is exactly what we want in this case.
> >>
> >> Now this is what is confusing me. I don't see the above happening.
> >>
> >> I see that the primary kernel boots up and adds the ACPI regions via:
> >> acpi_os_ioremap
> >> -> ioremap_cache
> >>
> >> But during the crashkernel boot, ''acpi_os_ioremap' calls
> >> 'ioremap' for the ACPI Reclaim Memory regions and not the _cache
> >> variant.
> >>
> >> And it fails while accessing the ACPI tables:
> >>
> >> [ 0.039205] ACPI: Core revision 20170728
> >> pud=000000002e7d0003, *pmd=000000002e7c0003, *pte=00e8000039710707
> >> [ 0.095098] Internal error: Oops: 96000021 [#1] SMP
> >> [ 0.100022] Modules linked in:
> >> [ 0.103102] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.14.0-rc6 #1
> >> [ 0.109432] task: ffff000008d05180 task.stack: ffff000008cc0000
> >> [ 0.115414] PC is at acpi_ns_lookup+0x25c/0x3c0
> >> [ 0.119987] LR is at acpi_ds_load1_begin_op+0xa4/0x294
> >> [ 0.125175] pc : [<ffff0000084a6764>] lr : [<ffff00000849b4f8>]
> >> pstate: 60000045
> >> [ 0.132647] sp : ffff000008ccfb40
> >> [ 0.135989] x29: ffff000008ccfb40 x28: ffff000008a9f2a4
> >> [ 0.141354] x27: ffff0000088be820 x26: 0000000000000000
> >> [ 0.146718] x25: 000000000000001b x24: 0000000000000001
> >> [ 0.152083] x23: 0000000000000001 x22: ffff000009710027
> >> [ 0.157447] x21: ffff000008ccfc50 x20: 0000000000000001
> >> [ 0.162812] x19: 000000000000001b x18: 0000000000000005
> >> [ 0.168176] x17: 0000000000000000 x16: 0000000000000000
> >> [ 0.173541] x15: 0000000000000000 x14: 000000000000038e
> >> [ 0.178905] x13: ffffffff00000000 x12: ffffffffffffffff
> >> [ 0.184270] x11: 0000000000000006 x10: 00000000ffffff76
> >> [ 0.189634] x9 : 000000000000005f x8 : ffff8000126d0140
> >> [ 0.194998] x7 : 0000000000000000 x6 : ffff000008ccfc50
> >> [ 0.200362] x5 : ffff80000fe62c00 x4 : 0000000000000001
> >> [ 0.205727] x3 : ffff000008ccfbe0 x2 : ffff0000095e3980
> >> [ 0.211091] x1 : ffff000009710027 x0 : 0000000000000000
> >> [ 0.216456] Process swapper/0 (pid: 0, stack limit = 0xffff000008cc0000)
> >> [ 0.223224] Call trace:
> >> [ 0.225688] Exception stack(0xffff000008ccfa00 to 0xffff000008ccfb40)
> >> [ 0.232194] fa00: 0000000000000000 ffff000009710027
> >> ffff0000095e3980 ffff000008ccfbe0
> >> [ 0.240106] fa20: 0000000000000001 ffff80000fe62c00
> >> ffff000008ccfc50 0000000000000000
> >> [ 0.248018] fa40: ffff8000126d0140 000000000000005f
> >> 00000000ffffff76 0000000000000006
> >> [ 0.255931] fa60: ffffffffffffffff ffffffff00000000
> >> 000000000000038e 0000000000000000
> >> [ 0.263843] fa80: 0000000000000000 0000000000000000
> >> 0000000000000005 000000000000001b
> >> [ 0.271754] faa0: 0000000000000001 ffff000008ccfc50
> >> ffff000009710027 0000000000000001
> >> [ 0.279667] fac0: 0000000000000001 000000000000001b
> >> 0000000000000000 ffff0000088be820
> >> [ 0.287579] fae0: ffff000008a9f2a4 ffff000008ccfb40
> >> ffff00000849b4f8 ffff000008ccfb40
> >> [ 0.295491] fb00: ffff0000084a6764 0000000060000045
> >> ffff000008ccfb40 ffff000008260a18
> >> [ 0.303403] fb20: ffffffffffffffff ffff0000087f3fb0
> >> ffff000008ccfb40 ffff0000084a6764
> >> [ 0.311316] [<ffff0000084a6764>] acpi_ns_lookup+0x25c/0x3c0
> >> [ 0.316943] [<ffff00000849b4f8>] acpi_ds_load1_begin_op+0xa4/0x294
> >> [ 0.323186] [<ffff0000084ad4ac>] acpi_ps_build_named_op+0xc4/0x198
> >> [ 0.329428] [<ffff0000084ad6cc>] acpi_ps_create_op+0x14c/0x270
> >> [ 0.335319] [<ffff0000084acfa8>] acpi_ps_parse_loop+0x188/0x5c8
> >> [ 0.341298] [<ffff0000084ae048>] acpi_ps_parse_aml+0xb0/0x2b8
> >> [ 0.347101] [<ffff0000084a8e10>] acpi_ns_one_complete_parse+0x144/0x184
> >> [ 0.353783] [<ffff0000084a8e98>] acpi_ns_parse_table+0x48/0x68
> >> [ 0.359675] [<ffff0000084a82cc>] acpi_ns_load_table+0x4c/0xdc
> >> [ 0.365479] [<ffff0000084b32f8>] acpi_tb_load_namespace+0xe4/0x264
> >> [ 0.371723] [<ffff000008baf9b4>] acpi_load_tables+0x48/0xc0
> >> [ 0.377350] [<ffff000008badc20>] acpi_early_init+0x9c/0xd0
> >> [ 0.382891] [<ffff000008b70d50>] start_kernel+0x3b4/0x43c
> >> [ 0.388343] Code: b9008fb9 2a000318 36380054 32190318 (b94002c0)
> >> [ 0.394500] ---[ end trace c46ed37f9651c58e ]---
> >> [ 0.399160] Kernel panic - not syncing: Fatal exception
> >> [ 0.404437] Rebooting in 10 seconds.
> >>
> >> So, I think the linear mapping done by the primary kernel does not
> >> make these accessible in the crash kernel directly.
> >>
> >> Any pointers?
> >
> > Can you get the code line number for acpi_ns_lookup+0x25c?
>
> gdb points to the following code line number:
>
> (gdb) list *(acpi_ns_lookup+0x25c)
> 0xffff0000084aa250 is in acpi_ns_lookup (drivers/acpi/acpica/nsaccess.c:577).
> 572 }
> 573 }
> 574
> 575 /* Extract one ACPI name from the front of the pathname */
> 576
> 577 ACPI_MOVE_32_TO_32(&simple_name, path);
> 578
> 579 /* Try to find the single (4 character) ACPI name */
> 580
> 581 status =
> (gdb)
>
> i.e. ACPI_MOVE_32_TO_32(&simple_name, path);

This macro can be defined in two ways depending on
ACPI_MISALIGNMENT_NOT_SUPPORTED in drivers/acpi/acpica/acmarcos.h.
So, in principle, any use of ioremap() in acpi_os_ioremap() may be
in conflict with those definitions here.

This suggests that, under the current code base, we must expose
ACPI reclaim regions as memblocks (i.e. via usable-memory-range)
in order to avoid the reported issue.

Thanks,
-Takahiro AKASHI

> addr2line also confirms the same:
>
> # addr2line -e vmlinux ffff0000084aa250
> /root/git/kernel-alt/drivers/acpi/acpica/nsaccess.c:577
>
>
> Regards,
> Bhupesh
>
>
> >>
> >> Regards,
> >> Bhupesh
> >>
> >> >> Just FYI, on x86, ACPI tables seems to be exposed to crash dump kernel
> >> >> via a kernel command line parameter, "memmap=".
> >> >>
> >> _______________________________________________
> >> kexec mailing list -- kexec@xxxxxxxxxxxxxxxxxxxxxxx
> >> To unsubscribe send an email to kexec-leave@xxxxxxxxxxxxxxxxxxxxxxx