This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH] Remove an instance of MAX_REGISTER_SIZE from record-full.c
- From: Alan Hayward <Alan dot Hayward at arm dot com>
- To: Yao Qi <qiyaoltc at gmail dot com>
- Cc: "gdb-patches at sourceware dot org" <gdb-patches at sourceware dot org>, nd <nd at arm dot com>
- Date: Wed, 21 Jun 2017 09:31:40 +0000
- Subject: Re: [PATCH] Remove an instance of MAX_REGISTER_SIZE from record-full.c
- Authentication-results: sourceware.org; auth=none
- Authentication-results: gmail.com; dkim=none (message not signed) header.d=none;gmail.com; dmarc=none action=none header.from=arm.com;
- Nodisclaimer: True
- References: <69EEE46A-D88C-4B4F-86A2-E35F6DAFD90A@arm.com> <86d1a9a2k9.fsf@gmail.com> <C7AECD2C-570A-4934-A153-FFA8B66E8185@arm.com>
- Spamdiagnosticmetadata: NSPM
- Spamdiagnosticoutput: 1:99
PING
> On 12 Jun 2017, at 14:59, Alan Hayward <Alan.Hayward@arm.com> wrote:
>
>
>> On 12 Jun 2017, at 12:05, Yao Qi <qiyaoltc@gmail.com> wrote:
>>
>> Alan Hayward <Alan.Hayward@arm.com> writes:
>>
>>> - regcache_cooked_read (regcache, entry->u.reg.num, reg);
>>> - regcache_cooked_write (regcache, entry->u.reg.num,
>>> - record_full_get_loc (entry));
>>> - memcpy (record_full_get_loc (entry), reg, entry->u.reg.len);
>>
>> The original code is about swapping contents of register REG in regcache
>> and record_full_get_loc (entry), and the length is known entry->u.reg.len.
>>
>
> Yes.
>
>>> + value = regcache->cooked_read_value (entry->u.reg.num);
>>> + gdb_assert (value != NULL);
>>> + regcache->cooked_write (entry->u.reg.num, record_full_get_loc (entry));
>>> + memcpy (record_full_get_loc (entry), value_contents_all (value),
>>> + entry->u.reg.len);
>>> + release_value (value);
>>> + value_free (value);
>>
>> It is a overkill to use value to swap these two buffers, IMO. How about
>> xmalloc "reg" instead?
>>
>
> Given that the code doesn’t use any of the error checking, then agreed.
>
>
> Tested on a --enable-targets=all build with board files unix and
> native-gdbserver.
>
>
> Ok to commit?
> Alan.
>
> 2017-06-12 Alan Hayward <alan.hayward@arm.com>
>
> * gdb/record-full.c (record_full_exec_insn): Allocate buffer.
>
>
> diff --git a/gdb/record-full.c b/gdb/record-full.c
> index 31ff558d2a633cff71d3e6082e42f5d6fb88bcf1..4f73e2a5ad0d4a2407b31a1d391e813147e15798 100644
> --- a/gdb/record-full.c
> +++ b/gdb/record-full.c
> @@ -698,7 +698,7 @@ record_full_exec_insn (struct regcache *regcache,
> {
> case record_full_reg: /* reg */
> {
> - gdb_byte reg[MAX_REGISTER_SIZE];
> + gdb_byte *reg = (gdb_byte *) xmalloc (entry->u.reg.len);
>
> if (record_debug > 1)
> fprintf_unfiltered (gdb_stdlog,
> @@ -711,6 +711,7 @@ record_full_exec_insn (struct regcache *regcache,
> regcache_cooked_write (regcache, entry->u.reg.num,
> record_full_get_loc (entry));
> memcpy (record_full_get_loc (entry), reg, entry->u.reg.len);
> + xfree (reg);
> }
> break;
>
>
>