[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20250515155811.114392-4-iii@linux.ibm.com>
Date: Thu, 15 May 2025 17:52:13 +0200
From: Ilya Leoshkevich <iii@...ux.ibm.com>
To: Jan Kiszka <jan.kiszka@...mens.com>, Kieran Bingham <kbingham@...nel.org>,
Andrew Morton <akpm@...ux-foundation.org>
Cc: linux-kernel@...r.kernel.org, Heiko Carstens <hca@...ux.ibm.com>,
Vasily Gorbik <gor@...ux.ibm.com>,
Alexander Gordeev <agordeev@...ux.ibm.com>,
Ilya Leoshkevich <iii@...ux.ibm.com>
Subject: [PATCH 3/3] scripts/gdb/symbols: determine KASLR offset on s390 during early boot
Using lx-symbols during s390 early boot fails with:
Error occurred in Python: 'utf-8' codec can't decode byte 0xcb in position 0: invalid continuation byte
The reason is that s390 decompressor's startup_kernel() does not create
vmcoreinfo note, and sets vmcore_info to kernel's physical base. This
confuses get_vmcore_s390().
Fix by handling this special case. Extract vm_layout.kaslr_offset from
the kernel image in physical memory, which is placed there by the
decompressor using the __bootdata_preserved mechanism, and generate
a synthetic vmcoreinfo note from it.
Signed-off-by: Ilya Leoshkevich <iii@...ux.ibm.com>
---
scripts/gdb/linux/symbols.py | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/scripts/gdb/linux/symbols.py b/scripts/gdb/linux/symbols.py
index 0c7af712c44c..2332bd8eddf1 100644
--- a/scripts/gdb/linux/symbols.py
+++ b/scripts/gdb/linux/symbols.py
@@ -54,6 +54,18 @@ def get_vmcore_s390():
vmcore_info = 0x0e0c
paddr_vmcoreinfo_note = gdb.parse_and_eval("*(unsigned long long *)" +
hex(vmcore_info))
+ if paddr_vmcoreinfo_note == 0 or paddr_vmcoreinfo_note & 1:
+ # In the early boot case, extract vm_layout.kaslr_offset from the
+ # vmlinux image in physical memory.
+ if paddr_vmcoreinfo_note == 0:
+ kaslr_offset_phys = 0
+ else:
+ kaslr_offset_phys = paddr_vmcoreinfo_note - 1
+ with utils.pagination_off():
+ gdb.execute("symbol-file {0} -o {1}".format(
+ utils.get_vmlinux(), hex(kaslr_offset_phys)))
+ kaslr_offset = gdb.parse_and_eval("vm_layout.kaslr_offset")
+ return "KERNELOFFSET=" + hex(kaslr_offset)[2:]
inferior = gdb.selected_inferior()
elf_note = inferior.read_memory(paddr_vmcoreinfo_note, 12)
n_namesz, n_descsz, n_type = struct.unpack(">III", elf_note)
--
2.49.0
Powered by blists - more mailing lists