Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756804AbcJRQcW (ORCPT ); Tue, 18 Oct 2016 12:32:22 -0400 Received: from mail-wm0-f54.google.com ([74.125.82.54]:38631 "EHLO mail-wm0-f54.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756551AbcJRQbr (ORCPT ); Tue, 18 Oct 2016 12:31:47 -0400 Date: Tue, 18 Oct 2016 17:31:42 +0100 From: Peter Griffin To: Kieran Bingham Cc: linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, kernel@stlinux.com, jan.kiszka@siemens.com, lee.jones@linaro.org, devicetree@vger.kernel.org Subject: Re: [PATCH 1/2] scripts/gdb: add lx-fdtdump command Message-ID: <20161018163142.GB28359@griffinp-ThinkPad-X1-Carbon-2nd> References: <1476803249-23328-1-git-send-email-peter.griffin@linaro.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 6220 Lines: 181 Hi Kieran, Thanks for reviewing. On Tue, 18 Oct 2016, Kieran Bingham wrote: > Hi Pete, > > Thanks for your patch. > > On 18/10/16 16:07, Peter Griffin wrote: > > lx-fdtdump dumps the flatenned device tree passed to the kernel > > s/flatenned/flattened/ > > > from the bootloader to a file called fdtdump.dtb to allow further > > post processing on the machine running GDB. The fdt header is also > > also printed in the GDB console. For example: > > > > Excellent - I've been looking forward to this > > General comment: It would be good to see the output filename > configurable as a parameter, defaulting to fdtdump.dtb if none provided. > > Is this something you have time to add? Have added this in v2 :) > > > (gdb) lx-fdtdump > > fdt_magic: 0xD00DFEED > > fdt_totalsize: 0xC108 > > off_dt_struct: 0x38 > > off_dt_strings: 0x3804 > > off_mem_rsvmap: 0x28 > > version: 17 > > last_comp_version: 16 > > Dumped fdt to fdtdump.dtb > > > >> fdtdump fdtdump.dtb | less > > > > This command is useful as the bootloader can often re-write parts > > of the device tree, and this can sometimes cause the kernel to not > > boot. > > > > Signed-off-by: Peter Griffin > > --- > > scripts/gdb/linux/constants.py.in | 8 +++++ > > scripts/gdb/linux/proc.py | 70 ++++++++++++++++++++++++++++++++++++++- > > 2 files changed, 77 insertions(+), 1 deletion(-) > > > > diff --git a/scripts/gdb/linux/constants.py.in b/scripts/gdb/linux/constants.py.in > > index 7986f4e..43c6241 100644 > > --- a/scripts/gdb/linux/constants.py.in > > +++ b/scripts/gdb/linux/constants.py.in > > @@ -14,6 +14,7 @@ > > > > #include > > #include > > +#include > > > > /* We need to stringify expanded macros so that they can be parsed */ > > > > @@ -50,3 +51,10 @@ LX_VALUE(MNT_NOEXEC) > > LX_VALUE(MNT_NOATIME) > > LX_VALUE(MNT_NODIRATIME) > > LX_VALUE(MNT_RELATIME) > > + > > +/* linux/of_fdt.h> */ > > +LX_VALUE(OF_DT_HEADER) > > + > > +/* Kernel Configs */ > > +LX_CONFIG(CONFIG_OF) > > + > > diff --git a/scripts/gdb/linux/proc.py b/scripts/gdb/linux/proc.py > > index 38b1f09..f20fcfa 100644 > > --- a/scripts/gdb/linux/proc.py > > +++ b/scripts/gdb/linux/proc.py > > @@ -16,7 +16,7 @@ from linux import constants > > from linux import utils > > from linux import tasks > > from linux import lists > > - > > +from struct import * > > > > class LxCmdLine(gdb.Command): > > """ Report the Linux Commandline used in the current kernel. > > @@ -195,3 +195,71 @@ values of that process namespace""" > > info_opts(MNT_INFO, m_flags))) > > > > LxMounts() > > + > > +class LxFdtDump(gdb.Command): > > + """Output Flattened Device Tree header and dump FDT blob to a file > > + Equivalent to 'cat /proc/fdt > fdtdump.dtb' on a running target""" > > + > > + def __init__(self): > > + super(LxFdtDump, self).__init__("lx-fdtdump", gdb.COMMAND_DATA) > > + > > + def fdthdr_to_cpu(self, fdt_header): > > + > > + fdt_header_be = ">IIIIIII" > > + fdt_header_le = " > + > > + if utils.get_target_endianness() == 1: > > + output_fmt = fdt_header_le > > + else: > > + output_fmt = fdt_header_be > > + > > + return unpack(output_fmt, pack(fdt_header_be, > > + fdt_header['magic'], > > + fdt_header['totalsize'], > > + fdt_header['off_dt_struct'], > > + fdt_header['off_dt_strings'], > > + fdt_header['off_mem_rsvmap'], > > + fdt_header['version'], > > + fdt_header['last_comp_version'])) > > + > > + def invoke(self, arg, from_tty): > > + > > + if constants.LX_CONFIG_OF: > > + > > + filename = "fdtdump.dtb" > > + > > You could parse the arguments here to allow users to write to their own > file / path ... arg is just a string, so you can either parse it as such > or convert to argv with > argv = gdb.string_to_argv(arg) > > > Aha - I see Jan beat me to it :) > > > > + py_fdt_header_ptr = gdb.parse_and_eval( > > + "(const struct fdt_header *) initial_boot_params") > > + py_fdt_header = py_fdt_header_ptr.dereference() > > + > > + fdt_header = self.fdthdr_to_cpu(py_fdt_header) > > + > > + if fdt_header[0] != constants.LX_OF_DT_HEADER: > > + raise gdb.GdbError("No flattened device tree magic found\n") > > + > > + gdb.write("fdt_magic: 0x{:02X}\n".format(fdt_header[0])) > > + gdb.write("fdt_totalsize: 0x{:02X}\n".format(fdt_header[1])) > > + gdb.write("off_dt_struct: 0x{:02X}\n".format(fdt_header[2])) > > + gdb.write("off_dt_strings: 0x{:02X}\n".format(fdt_header[3])) > > + gdb.write("off_mem_rsvmap: 0x{:02X}\n".format(fdt_header[4])) > > + gdb.write("version: {}\n".format(fdt_header[5])) > > + gdb.write("last_comp_version: {}\n".format(fdt_header[6])) > > + > > + inf = gdb.inferiors()[0] > > + fdt_buf = utils.read_memoryview(inf, py_fdt_header_ptr, > > + fdt_header[1]).tobytes() > > + > > + try: > > + f = open(filename, 'wb') > > + except: > > + raise gdb.GdbError("Could not open file to dump fdt") > > + > > + f.write(fdt_buf) > > + f.close() > > + > > + gdb.write("Dumped fdt to " + filename + "\n") > > + > > + else: > > + gdb.write("Kernel not compiled with CONFIG_OF\n") > > Would it be cleaner to write > if not constants.LX_CONFIG_OF: > raise gdb.GdbError("Kernel not compiled with CONFIG_OF\n") > > at the beginning, and reduce the indentation level required ? Good idea, have updated like you suggest in V2 and reduced the indentation. regards, Peter