- Sep 12, 2022
-
-
Borislav Petkov authored
There are cases where the IP pointer in a Code: line in an oops doesn't point at the beginning of an instruction: Code: 0f bd c2 e9 a0 cd b5 e4 48 0f bd c2 e9 97 cd b5 e4 0f 1f 80 00 00 00 00 \ e9 8b cd b5 e4 0f 1f 00 66 0f a3 d0 e9 7f cd b5 e4 0f 1f <80> 00 00 00 \ 00 0f a3 d0 e9 70 cd b5 e4 48 0f a3 d0 e9 67 cd b5 e9 7f cd b5 e4 jmp 0xffffffffe4b5cda8 0f 1f 80 00 00 00 00 nopl 0x0(%rax) ^^ and the current way of determining the faulting instruction line doesn't work because disassembled instructions are counted from the IP byte to the end and when that thing points in the middle, the trailing bytes can be interpreted as different insns: Code starting with the faulting instruction =========================================== 0: 80 00 00 addb $0x0,(%rax) 3: 00 00 add %al,(%rax) whereas, this is part of 0f 1f 80 00 00 00 00 nopl 0x0(%rax) 5: 0f a3 d0 bt %edx,%eax ... leading to: 1d: 0f 1f 00 nopl (%rax) 20: 66 0f a3 d0 bt %dx,%ax 24:* e9 7f cd b5 e4 jmp 0xffffffffe4b5cda8 <-- trapping instruction 29: 0f 1f 80 00 00 00 00 nopl 0x0(%rax) 30: 0f a3 d0 bt %edx,%eax which is the wrong faulting instruction. Change the way the faulting line number is determined by matching the opcode bytes from the beginning, leading to correct output: 1d: 0f 1f 00 nopl (%rax) 20: 66 0f a3 d0 bt %dx,%ax 24: e9 7f cd b5 e4 jmp 0xffffffffe4b5cda8 29:* 0f 1f 80 00 00 00 00 nopl 0x0(%rax) <-- trapping instruction 30: 0f a3 d0 bt %edx,%eax While at it, make decodecode use bash as the interpreter - that thing should be present on everything by now. It simplifies the code a lot too. Link: https://lkml.kernel.org/r/20220808085928.29840-1-bp@alien8.de Signed-off-by:Borislav Petkov <bp@suse.de> Cc: Marc Zyngier <maz@kernel.org> Cc: Will Deacon <will@kernel.org> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org>
-
- Nov 06, 2021
-
-
weidonghui authored
If opps.file is in DOS format, faulting instruction cannot be printed: / # ARCH=arm64 CROSS_COMPILE=aarch64-linux-gnu- / # ./scripts/decodecode < oops.file [ 0.734345] Code: d0002881 912f9c21 94067e68 d2800001 (b900003f) aarch64-linux-gnu-strip: '/tmp/tmp.5Y9eybnnSi.o': No such file aarch64-linux-gnu-objdump: '/tmp/tmp.5Y9eybnnSi.o': No such file All code ======== 0: d0002881 adrp x1, 0x512000 4: 912f9c21 add x1, x1, #0xbe7 8: 94067e68 bl 0x19f9a8 c: d2800001 mov x1, #0x0 // #0 10: b900003f str wzr, [x1] Code starting with the faulting instruction =========================================== Background: The compilation environment is Ubuntu, and the test environment is Windows. Most logs are generated in the Windows environment. In this way, CR (carriage return) will inevitably appear, which will affect the use of decodecode in the Ubuntu environment. The repaired effect is as follows: / # ARCH=arm64 CROSS_COMPILE=aarch64-linux-gnu- / # ./scripts/decodecode < oops.file [ 0.734345] Code: d0002881 912f9c21 94067e68 d2800001 (b900003f) All code ======== 0: d0002881 adrp x1, 0x512000 4: 912f9c21 add x1, x1, #0xbe7 8: 94067e68 bl 0x19f9a8 c: d2800001 mov x1, #0x0 // #0 10:* b900003f str wzr, [x1] <-- trapping instruction Code starting with the faulting instruction =========================================== 0: b900003f str wzr, [x1] Link: https://lkml.kernel.org/r/20211008064712.926-1-weidonghui@allwinnertech.com Signed-off-by:weidonghui <weidonghui@allwinnertech.com> Acked-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
- Oct 14, 2020
-
-
Borislav Petkov authored
So that comparing with objdump output from vmlinux can ease pinpointing where the trapping instruction actually is. An example is better than a thousand words: $ PC=0xffffffff8329a927 ./scripts/decodecode < ~/tmp/syz/gfs2.splat [ 477.379104][T23917] Code: 48 83 ec 28 48 89 3c 24 48 89 54 24 08 e8 c1 b4 4a fe 48 8d bb 00 01 00 00 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f 85 97 05 00 00 48 8b 9b 00 01 00 00 48 85 db 0f 84 All code ======== ffffffff8329a8fd: 48 83 ec 28 sub $0x28,%rsp ffffffff8329a901: 48 89 3c 24 mov %rdi,(%rsp) ffffffff8329a905: 48 89 54 24 08 mov %rdx,0x8(%rsp) ffffffff8329a90a: e8 c1 b4 4a fe callq 0xffffffff81745dd0 ffffffff8329a90f: 48 8d bb 00 01 00 00 lea 0x100(%rbx),%rdi ffffffff8329a916: 48 b8 00 00 00 00 00 movabs $0xdffffc0000000000,%rax ffffffff8329a91d: fc ff df ffffffff8329a920: 48 89 fa mov %rdi,%rdx ffffffff8329a923: 48 c1 ea 03 shr $0x3,%rdx ffffffff8329a927:* 80 3c 02 00 cmpb $0x0,(%rdx,%rax,1) <-- trapping instruction ffffffff8329a92b: 0f 85 97 05 00 00 jne 0xffffffff8329aec8 ffffffff8329a931: 48 8b 9b 00 01 00 00 mov 0x100(%rbx),%rbx ffffffff8329a938: 48 85 db test %rbx,%rbx ffffffff8329a93b: 0f .byte 0xf ffffffff8329a93c: 84 .byte 0x84 Signed-off-by:
-
- May 08, 2020
-
-
Ivan Delalande authored
If the trapping instruction contains a ':', for a memory access through segment registers for example, the sed substitution will insert the '*' marker in the middle of the instruction instead of the line address: 2b: 65 48 0f c7 0f cmpxchg16b %gs:*(%rdi) <-- trapping instruction I started to think I had forgotten some quirk of the assembly syntax before noticing that it was actually coming from the script. Fix it to add the address marker at the right place for these instructions: 28: 49 8b 06 mov (%r14),%rax 2b:* 65 48 0f c7 0f cmpxchg16b %gs:(%rdi) <-- trapping instruction 30: 0f 94 c0 sete %al Fixes: 18ff44b1 ("scripts/decodecode: make faulting insn ptr more robust") Signed-off-by:
Ivan Delalande <colona@arista.com> Signed-off-by:
-
- Dec 28, 2018
-
-
Marc Zyngier authored
When running decodecode natively on arm64, ARCH is likely not to be set, and we end-up with .4byte instead of .inst when generating the disassembly. Similar effects would occur if running natively on a 32bit ARM platform, although that's even less popular. A simple workaround is to populate ARCH when it is not set and that we're running on an arm/arm64 system. Link: http://lkml.kernel.org/r/20181210174659.31054-2-marc.zyngier@arm.com Signed-off-by:
Marc Zyngier <marc.zyngier@arm.com> Acked-by:
Will Deacon <will.deacon@arm.com> Signed-off-by:
-
- Feb 01, 2018
-
-
Andy Shevchenko authored
In case of running scripts/decodecode without any parameters in order to give a copy'n'pasted Code line from, for example, email it would parse only first line of it, while in emails it's split to few. ie, when you have a file out of oops the Code line looks like Code: hh hh ... <hh> ... hh\n When copy'n'paste from, for example, email where sender or some middle MTA split it, the line looks like: Code: hh hh ... hh\n hh ... <hh> ... hh\n hh hh ... hh\n The Code line followed by another oops line usually contains characters out of hex digit + space + < + > set. So add logic to join this split back if and only if the following lines have hex digits, or spaces, or '<', or '>' characters. It will be quite unlikely to have a broken input in well formed Oops or dmesg, thus a simple regex is being used. Link: http://lkml.kernel.org/r/20171212100323.33201-1-andriy.shevchenko@linux.intel.com Signed-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Will Deacon <will.deacon@arm.com> Cc: Dave Martin <Dave.Martin@arm.com> Cc: Philippe Ombredanne <pombredanne@nexb.com> Signed-off-by:
-
- Jan 19, 2018
-
-
Will Deacon authored
There are a couple of problems with the decodecode script and arm64: 1. AArch64 objdump refuses to disassemble .4byte directives as instructions, insisting that they are data values and displaying them as: a94153f3 .word 0xa94153f3 <-- trapping instruction This is resolved by using the .inst directive instead. 2. Disassembly of branch instructions attempts to provide the target as an offset from a symbol, e.g.: 0: 34000082 cbz w2, 10 <.text+0x10> however this falls foul of the grep -v, which matches lines containing ".text" and ends up removing all branch instructions from the dump. This patch resolves both issues by using the .inst directive for 4-byte quantities on arm64 and stripping the resulting binaries (as is done on arm already) to remove the mapping symbols. Link: http://lkml.kernel.org/r/1506596147-23630-1-git-send-email-will.deacon@arm.com Signed-off-by:
Dave Martin <Dave.Martin@arm.com> Cc: Michal Marek <mmarek@suse.cz> Signed-off-by:
-
- Nov 02, 2017
-
-
Greg Kroah-Hartman authored
Many source files in the tree are missing licensing information, which makes it harder for compliance tools to determine the correct license. By default all files without license information are under the default license of the kernel, which is GPL version 2. Update the files which contain no license information with the 'GPL-2.0' SPDX license identifier. The SPDX identifier is a legally binding shorthand, which can be used instead of the full boiler plate text. This patch is based on work done by Thomas Gleixner and Kate Stewart and Philippe Ombredanne. How this work was done: Patches were generated and checked against linux-4.14-rc6 for a subset of the use cases: - file had no licensing information it it. - file was a */uapi/* one with no licensing information in it, - file was a */uapi/* one with existing licensing information, Further patches will be generated in subsequent months to fix up cases where non-standard license headers were used, and references to license had to be inferred by heuristics based on keywords. The analysis to determine which SPDX License Identifier to be applied to a file was done in a spreadsheet of side by side results from of the output of two independent scanners (ScanCode & Windriver) producing SPDX tag:value files created by Philippe Ombredanne. Philippe prepared the base worksheet, and did an initial spot review of a few 1000 files. The 4.13 kernel was the starting point of the analysis with 60,537 files assessed. Kate Stewart did a file by file comparison of the scanner results in the spreadsheet to determine which SPDX license identifier(s) to be applied to the file. She confirmed any determination that was not immediately clear with lawyers working with the Linux Foundation. Criteria used to select files for SPDX license identifier tagging was: - Files considered eligible had to be source code files. - Make and config files were included as candidates if they contained >5 lines of source - File already had some variant of a license header in it (even if <5 lines). All documentation files were explicitly excluded. The following heuristics were used to determine which SPDX license identifiers to apply. - when both scanners couldn't find any license traces, file was considered to have no license information in it, and the top level COPYING file license applied. For non */uapi/* files that summary was: SPDX license identifier # files ---------------------------------------------------|------- GPL-2.0 11139 and resulted in the first patch in this series. If that file was a */uapi/* path one, it was "GPL-2.0 WITH Linux-syscall-note" otherwise it was "GPL-2.0". Results of that was: SPDX license identifier # files ---------------------------------------------------|------- GPL-2.0 WITH Linux-syscall-note 930 and resulted in the second patch in this series. - if a file had some form of licensing information in it, and was one of the */uapi/* ones, it was denoted with the Linux-syscall-note if any GPL family license was found in the file or had no licensing in it (per prior point). Results summary: SPDX license identifier # files ---------------------------------------------------|------ GPL-2.0 WITH Linux-syscall-note 270 GPL-2.0+ WITH Linux-syscall-note 169 ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) 21 ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) 17 LGPL-2.1+ WITH Linux-syscall-note 15 GPL-1.0+ WITH Linux-syscall-note 14 ((GPL-2.0+ WITH Linux-syscall-note) OR BSD-3-Clause) 5 LGPL-2.0+ WITH Linux-syscall-note 4 LGPL-2.1 WITH Linux-syscall-note 3 ((GPL-2.0 WITH Linux-syscall-note) OR MIT) 3 ((GPL-2.0 WITH Linux-syscall-note) AND MIT) 1 and that resulted in the third patch in this series. - when the two scanners agreed on the detected license(s), that became the concluded license(s). - when there was disagreement between the two scanners (one detected a license but the other didn't, or they both detected different licenses) a manual inspection of the file occurred. - In most cases a manual inspection of the information in the file resulted in a clear resolution of the license that should apply (and which scanner probably needed to revisit its heuristics). - When it was not immediately clear, the license identifier was confirmed with lawyers working with the Linux Foundation. - If there was any question as to the appropriate license identifier, the file was flagged for further research and to be revisited later in time. In total, over 70 hours of logged manual review was done on the spreadsheet to determine the SPDX license identifiers to apply to the source files by Kate, Philippe, Thomas and, in some cases, confirmation by lawyers working with the Linux Foundation. Kate also obtained a third independent scan of the 4.13 code base from FOSSology, and compared selected files where the other two scanners disagreed against that SPDX file, to see if there was new insights. The Windriver scanner is based on an older version of FOSSology in part, so they are related. Thomas did random spot checks in about 500 files from the spreadsheets for the uapi headers and agreed with SPDX license identifier in the files he inspected. For the non-uapi files Thomas did random spot checks in about 15000 files. In initial set of patches against 4.14-rc6, 3 files were found to have copy/paste license identifier errors, and have been fixed to reflect the correct identifier. Additionally Philippe spent 10 hours this week doing a detailed manual inspection and review of the 12,461 patched files from the initial patch version early this week with: - a full scancode scan run, collecting the matched texts, detected license ids and scores - reviewing anything where there was a license detected (about 500+ files) to ensure that the applied SPDX license was correct - reviewing anything where there was no detection but the patch license was not GPL-2.0 WITH Linux-syscall-note to ensure that the applied SPDX license was correct This produced a worksheet with 20 files needing minor correction. This worksheet was then exported into 3 different .csv files for the different types of files to be modified. These .csv files were then reviewed by Greg. Thomas wrote a script to parse the csv files and add the proper SPDX tag to the file, in the format that the file expected. This script was further refined by Greg based on the output to detect more types of files automatically and to distinguish between header and source .c files (which need different comment types.) Finally Greg ran the script using the .csv files to generate the patches. Reviewed-by:
Kate Stewart <kstewart@linuxfoundation.org> Reviewed-by:
Philippe Ombredanne <pombredanne@nexb.com> Reviewed-by:
Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
- Apr 29, 2013
-
-
Borislav Petkov authored
It can accidentally happen that the faulting insn (the exact instruction bytes) is repeated a little further on in the trace. This causes that same instruction to be tagged twice, see example below. What we want to do, however, is to track back from the end of the whole disassembly so many lines as the slice which starts with the faulting instruction is long. This leads us to the actual faulting instruction and *then* we tag it. While we're at it, we can drop the sed "g" flag because we address only this one line. Also, if we point to an instruction which changes decoding depending on the slice being objdumped, like a Jcc insn, for example, we do not even tag it as a faulting instruction because the instruction decode changes in the second slice but we use that second format as a regex on the fsrst disassembled buffer and more often than not that instruction doesn't match. Again, simply tag the line which is deduced from the original "<>" marking we've received from the kernel. This also solves the pathologic issue of multiple tagging like this: 29:* 0f 0b ud2 <-- trapping instruction 2b:* 0f 0b ud2 <-- trapping instruction 2d:* 0f 0b ud2 <-- trapping instruction Double tagging example: Code: 34 dd 40 30 ad 81 48 c7 c0 80 f6 00 00 48 8b 3c 30 48 01 c6 b8 ff ff ff ff 48 8d 57 f0 48 39 f7 74 2f 49 8b 4c 24 08 48 8b 47 f0 <48> 39 48 08 75 0e eb 2a 66 90 48 8b 40 f0 48 39 48 08 74 1e 48 All code ======== 0: 34 dd xor $0xdd,%al 2: 40 30 ad 81 48 c7 c0 xor %bpl,-0x3f38b77f(%rbp) 9: 80 f6 00 xor $0x0,%dh c: 00 48 8b add %cl,-0x75(%rax) f: 3c 30 cmp $0x30,%al 11: 48 01 c6 add %rax,%rsi 14: b8 ff ff ff ff mov $0xffffffff,%eax 19: 48 8d 57 f0 lea -0x10(%rdi),%rdx 1d: 48 39 f7 cmp %rsi,%rdi 20: 74 2f je 0x51 22: 49 8b 4c 24 08 mov 0x8(%r12),%rcx 27: 48 8b 47 f0 mov -0x10(%rdi),%rax 2b:* 48 39 48 08 cmp %rcx,0x8(%rax) <-- trapping instruction 2f: 75 0e jne 0x3f 31: eb 2a jmp 0x5d 33: 66 90 xchg %ax,%ax 35: 48 8b 40 f0 mov -0x10(%rax),%rax 39:* 48 39 48 08 cmp %rcx,0x8(%rax) <-- trapping instruction 3d: 74 1e je 0x5d 3f: 48 rex.W Signed-off-by:
-
- Aug 16, 2012
-
-
Borislav Petkov authored
When dumping "Code: " sections from an oops, the trapping instruction %rip points to can be a string copy 2b:* f3 a5 rep movsl %ds:(%rsi),%es:(%rdi) and the line contain a bunch of ":". Current "cut" selects only the and the second field output looks funnily overlaid this: 2b:* f3 a5 rep movsl %ds <-- trapping instruction:(%rsi),%es:(%rdi Fix this by selecting the remaining fields too. Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: linux-kbuild@vger.kernel.org Signed-off-by:
Borislav Petkov <borislav.petkov@amd.com> Signed-off-by:
-
- Jun 04, 2010
-
-
Rabin Vincent authored
Remove bashisms to make scripts/decodecode work with other shells. Signed-off-by:
Rabin Vincent <rabin@rab.in> Reviewed-by:
WANG Cong <xiyou.wangcong@gmail.com> Signed-off-by:
Michal Marek <mmarek@suse.cz>
-
- Feb 02, 2010
-
-
Rabin Vincent authored
This patch adds support for decoding ARM oopses to scripts/decodecode. The following things are handled: - ARCH and CROSS_COMPILE environment variables are respected. - The Code: in x86 oopses is in bytes, while it is in either words (4 bytes) or halfwords for ARM. - Some versions of ARM objdump refuse to disassemble instructions generated by literal constants (".word 0x..."). The workaround is to strip the object file first. - The faulting instruction is marked (liked so) in ARM, but <like so> in x86. - ARM mnemonics may include characters such as [] which need to be escaped before being passed to sed for the "<- trapping instruction" substitution. Signed-off-by:
-
- Dec 03, 2008
-
-
kerneloops.org has been using an improved "decodecode" script, specifically it has a special marker that shows which line in the assembly the oops happened at, like this: 20: 83 e0 03 and $0x3,%eax 23: 09 d8 or %ebx,%eax 25: 85 db test %ebx,%ebx 27: 89 02 mov %eax,(%edx) 29: 74 0f je 0x3a 2b:* 3b 73 04 cmp 0x4(%ebx),%esi <-- trapping instruction 2e: 75 05 jne 0x35 30: 89 53 04 mov %edx,0x4(%ebx) 33: eb 07 jmp 0x3c 35: 89 53 08 mov %edx,0x8(%ebx) this patch updates the kernel copy to also have this functionality. Signed-off-by:
Arjan van de Ven <arjan@linux.intel.com> Reviewed-by:
WANG Cong <wangcong@zeuux.org> Signed-off-by:
Sam Ravnborg <sam@ravnborg.org>
-
- May 30, 2008
-
-
Linus Torvalds authored
.. because it is. Signed-off-by:
-
- Jan 28, 2008
-
-
Remove the tmp file when exiting. Noticed by Arjan van de Ven. Catch mktemp failure and exit with message. Trap kill or other signals and exit cleanly. Signed-off-by:
Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by:
-
- Jul 16, 2007
-
-
Randy Dunlap authored
Add info that the Code: bytes line contains <xy> or (wxyz) in some architecture oops reports and what that means. Add a script by Andi Kleen that reads the Code: line from an Oops report file and generates assembly code from the hex bytes. Signed-off-by:
-
