This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [patch v4 1/3] Create remove-symbol-file command.
- From: Luis Machado <lgustavo at codesourcery dot com>
- To: "Blanc, Nicolas" <nicolas dot blanc at intel dot com>
- Cc: "gdb-patches at sourceware dot org" <gdb-patches at sourceware dot org>, "palves at redhat dot com" <palves at redhat dot com>, "tromey at redhat dot com" <tromey at redhat dot com>, "eliz at gnu dot org" <eliz at gnu dot org>, "yao at codesourcery dot com" <yao at codesourcery dot com>
- Date: Wed, 29 May 2013 14:11:38 +0200
- Subject: Re: [patch v4 1/3] Create remove-symbol-file command.
- References: <1369818805-14288-1-git-send-email-nicolas dot blanc at intel dot com> <1369818805-14288-2-git-send-email-nicolas dot blanc at intel dot com> <51A5CC72 dot 2090000 at codesourcery dot com> <388084C8C1E6A64FA36AD1D656E485661A799D20 at IRSMSX106 dot ger dot corp dot intel dot com>
- Reply-to: lgustavo at codesourcery dot com
Hi,
On 05/29/2013 02:08 PM, Blanc, Nicolas wrote:
Hi Luis,
Thanks for your feedback. I'll fix the nits.
I'm thinking, with a command called "remove-symbol-file" i would expect to provide some kind of filename to this command. Should the user also be able to state a DSO name here and have it unloaded?
Maybe have the name translated to the base address used to unload the library internally?
The first address parameter of the add-symbol-file command is a better way to identify the file to remove than the file name because a file can be loaded multiple times at different addresses -- this case is important for me. The user knows the "start address" very well since he typed it in for adding the file. So I see no real benefit of adding a file-name parameter.
I see. That was mostly a suggestion. If it problematic to implement it
that way, then i suppose it is OK to use solely the base address.
Keeping the current syntax would be ok with you?
Sure, that looks good to me.