[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <AM0PR04MB6434A133CB5ABA0286B0F487EE230@AM0PR04MB6434.eurprd04.prod.outlook.com>
Date: Tue, 23 Apr 2019 11:12:49 +0000
From: Leonard Crestez <leonard.crestez@....com>
To: Stephen Boyd <sboyd@...nel.org>
CC: Jan Kiszka <jan.kiszka@...mens.com>,
Kieran Bingham <kbingham@...nel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [PATCH 3/3] scripts/gdb: Add $lx_clk_core_lookup function
On 4/22/2019 11:18 PM, Stephen Boyd wrote:
> Quoting Leonard Crestez (2019-04-22 01:26:57)
>> diff --git a/scripts/gdb/linux/clk.py b/scripts/gdb/linux/clk.py
>> +class LxClkCoreLookup(gdb.Function):
>> + """Find struct clk_core by name"""
>> +
>> + def __init__(self):
>> + super(LxClkCoreLookup, self).__init__("lx_clk_core_lookup")
>> +
>> + def lookup_hlist(self, hlist_head, name):
>> + for child in clk_core_for_each_child(hlist_head):
>> + if child['name'].string() == name:
>
> Do you need to do the .string() for comparison? Or does it work just as
> well to compare a gdb.Value object to a python string? It would be nice
> if the gdb.Value object could figure out that they're not both gdb.Value
> objects so it can do a string comparison itself.
The gdb manual is not clear on how comparisons work on gdb.Value types.
Converting to a python string and comparing in python work well, using
== on string gdb.Values results in this:
gdb.error: evaluation of this expression requires the program to
have a function "malloc"
My guess is gdb attempts to convert both arguments to gdb.Value and do
the comparison via a call on the target? This is very undesirable here.
I get the same error if "name" is a gdb.Value instead of being converted
to a string in invoke().
--
Regards,
Leonard
Powered by blists - more mailing lists