This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: PR13984 - gdb stops controlling a thread after "Remote 'g' packet reply is too long: ..." error message
- From: Pedro Alves <palves at redhat dot com>
- To: Orgad Shaneh <orgads at gmail dot com>
- Cc: Luis Machado <lgustavo at codesourcery dot com>, gdb-patches at sourceware dot org, "Maciej W. Rozycki" <macro at linux-mips dot org>
- Date: Thu, 14 Apr 2016 10:47:36 +0100
- Subject: Re: PR13984 - gdb stops controlling a thread after "Remote 'g' packet reply is too long: ..." error message
- Authentication-results: sourceware.org; auth=none
- References: <CAGHpTBJLq0RhLo=tTP_BmDzNAaVFCHebBvoom21x-ETT9t4ZQw at mail dot gmail dot com> <570C14D4 dot 3030600 at codesourcery dot com> <CAGHpTBLM5r5xVWzxYv=qZpRz_o-d9=gp_yg+RSYqoEXFBywgfQ at mail dot gmail dot com> <570CFA04 dot 3070109 at codesourcery dot com> <CAGHpTBKTr=DcosvWbObCpFPGjNqgMqhQwBgRn3MOz6voBLGkkw at mail dot gmail dot com> <570E99DC dot 8080902 at codesourcery dot com> <CAGHpTB+dZ3a86nFFN8CAyTxzTBMk6eReWZpg03HYVqQO8-TVyA at mail dot gmail dot com> <570EABAA dot 50908 at redhat dot com> <CAGHpTB+zOJwfNaWjsvWTu9ek7qpgdNC1KOR336zN09uySJ9Bew at mail dot gmail dot com> <570EC0C0 dot 8030500 at redhat dot com> <CAGHpTB+y=1Rst7H9BsSX_enxraXAy_YpzPYRSCfoa=8QXvEM-w at mail dot gmail dot com> <CAGHpTBLAuUi5WdtFdbuVGav8GmQCA=aP=F+88fXsBRTKRvnCxQ at mail dot gmail dot com> <570F613A dot 8060708 at redhat dot com> <CAGHpTB+1vzHu=-AQhDRK808HiiM41wc+60UpFJXK8XoMe_ApMA at mail dot gmail dot com>
On 04/14/2016 10:39 AM, Orgad Shaneh wrote:
> On Thu, Apr 14, 2016 at 12:22 PM, Pedro Alves <palves@redhat.com> wrote:
>> On 04/14/2016 10:06 AM, Orgad Shaneh wrote:
>>
>>> Ok, autoconf helped. FSF gdbserver 6.5 has the same issue when the gdb
>>> client is >=7.5: "Remote 'g' packet reply is too long".
>>>
>>> So this doesn't seem to be related to Cavium patches.
>>
>> Since you can build gdb now, you should be able to use "git bisect"
>> to find the culprit.
>>
>> Thanks,
>> Pedro Alves
>>
>
> Good point. The first bad commit is 1faeff08.
>
Thanks. That's:
Author: Maciej W. Rozycki <macro@linux-mips.org>
AuthorDate: Thu Mar 1 22:19:48 2012 +0000
...
* mips-tdep.c (mips_generic_reg_names): Remove trailing null
strings.
(mips_tx39_reg_names): Likewise.
(mips_linux_reg_names): New array of register names for Linux
targets.
(mips_register_name): Check for a null pointer in
mips_processor_reg_names and return an empty string.
(mips_register_type): Exclude embedded registers for the IRIX
and Linux ABIs.
(mips_pseudo_register_type): Likewise. Use dynamic numbers to
refer to FP registers, LO, HI, BadVAddr, Cause and PC. Handle
DSP registers.
(mips_stab_reg_to_regnum): Handle DSP accumulators.
(mips_dwarf_dwarf2_ecoff_reg_to_regnum): Likewise.
(mips_gdbarch_init): Likewise. Initialize internal register
indices for the Linux ABI. Use dynamic numbers to refer to
registers, as applicable, while parsing the target description.
Maciej, see https://sourceware.org/ml/gdb-patches/2016-04/msg00310.html .
Thanks,
Pedro Alves