Re: How is the PostgreSQL debuginfo file generated

From: 高增琦 <pgf00a(at)gmail(dot)com>
To: Craig Ringer <craig(at)2ndquadrant(dot)com>
Cc: Rui Hai Jiang <ruihaijiang(at)msn(dot)com>, "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: How is the PostgreSQL debuginfo file generated
Date: 2017-11-24 07:17:24
Message-ID: CAFmBtr1k7yVJih-48wtOMk+jDuoZgh9OH9YFSXhZdLcMH5AHNA@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

A little trick:

1. configure with —enable-debug, run “make install”, use this build result
as debuginfo

2. then run “make install-strip”, use this result as release

However the it is not the regular debuginfo, you still can call gdb with
it. Additionally, you can dump the real debuginfo from it later.

Craig Ringer <craig(at)2ndquadrant(dot)com>于2017年11月24日 周五10:04写道:

> On 23 November 2017 at 18:38, Rui Hai Jiang <ruihaijiang(at)msn(dot)com> wrote:
>
>> Hello hackers,
>>
>>
>>
>> I’m wondering how to build the debuginfo package from the PostgreSQL
>> source.
>>
>>
>>
>> For example to generate something like this one :
>> postgresql91-debuginfo.x86_64.
>>
>>
>>
>> Is there existing support in the current PostgreSQL Makefiles to generate
>> such debuginfo? I have searched in the source code and could find anything.
>> How were the existing debuginfo packages created?
>>
>>
>>
> When you're building from source, no separate debuginfo is produced. If
> you build with --enable-debug, the debuginfo is embedded in the binaries.
> The resulting binaries are very large but not generally any slower.
>
> Packaging systems have helper programs that invoke 'strip' (or sometimes
> objcopy) to split out the debuginfo into a separate file and remove it from
> the original executable. See
> https://sourceware.org/gdb/onlinedocs/gdb/Separate-Debug-Files.html and
> https://fedoraproject.org/wiki/Packaging:Debuginfo?rd=Packaging/Debuginfo
> .
>
> You cannot generally re-create debuginfo to match a given binary package.
> You'd have to rebuild in an absolutely identical environment: exact same
> library versions, compiler version, configure flags, etc etc etc. Otherwise
> you'll get broken/misleading debuginfo.
>
>
> --
> Craig Ringer http://www.2ndQuadrant.com/
> PostgreSQL Development, 24x7 Support, Training & Services
>
--
GaoZengqi
pgf00a(at)gmail(dot)com
zengqigao(at)gmail(dot)com

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Ildus Kurbangaliev 2017-11-24 09:38:00 Re: [HACKERS] Custom compression methods
Previous Message Craig Ringer 2017-11-24 06:41:22 Re: [HACKERS] logical decoding of two-phase transactions