From: Simon Marchi <simon.marchi@ericsson.com>
To: Alan Hayward <alan.hayward@arm.com>, <gdb-patches@sourceware.org>
Cc: <nd@arm.com>
Subject: Re: [PATCH 1/8] Add Aarch64 SVE target description
Date: Thu, 31 May 2018 11:56:00 -0000 [thread overview]
Message-ID: <e5a7a3c9-7dcc-0103-f2a9-61a35b8b8401@ericsson.com> (raw)
In-Reply-To: <20180511105256.27388-2-alan.hayward@arm.com>
On 2018-05-11 06:52 AM, Alan Hayward wrote:
> This patch adds the SVE target description. However, no code will
> yet use it - that comes in the later patches.
>
> The create_feature_aarch64_sve function is not generated from XML.
> This is because we need to know the sve vector size (VQ) in order
> to size the registers correctly.
>
> A VQ of 0 is used when the hardware does not support SVE.
> (SVE hardware will always have a valid vector size). I considered
> using a bool to indicate SVE in addition to the VQ. Whilst this
> may be slightly more readable initially, I think it's a little
> odd to have two variables, eg:
> aarch64_create_target_description (bool sve_supported, long vq)
>
> Alan.
Hi Alan,
This patch LGTM, I just noted some nits.
> /* Initialize the current architecture based on INFO. If possible,
> @@ -2864,7 +2875,8 @@ aarch64_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
>
> /* Ensure we always have a target descriptor. */
> if (!tdesc_has_registers (tdesc))
> - tdesc = aarch64_read_description ();
> + /* SVE is not yet supported. */
> + tdesc = aarch64_read_description (0);
When there there is a comment above the single statement branch, braces become required:
if (!tdesc_has_registers (tdesc))
{
/* SVE is not yet supported. */
tdesc = aarch64_read_description (0);
}
> diff --git a/gdb/arch/aarch64.c b/gdb/arch/aarch64.c
> index b85e460b6b..d1ec5cedf8 100644
> --- a/gdb/arch/aarch64.c
> +++ b/gdb/arch/aarch64.c
> @@ -21,11 +21,13 @@
>
> #include "../features/aarch64-core.c"
> #include "../features/aarch64-fpu.c"
> +#include "../features/aarch64-sve.c"
>
> -/* Create the aarch64 target description. */
> +/* Create the aarch64 target description. A non zero VQ value indicates both
> + the presence of SVE and the SVE vector quotient. */
What does "SVE vector quotient" mean? Is there maybe a simpler way to say it?
Could you move this comment to the .h and put
/* See arch/aarch64.h. */
here?
> diff --git a/gdb/features/aarch64-sve.c b/gdb/features/aarch64-sve.c
> new file mode 100644
> index 0000000000..6442640a73
> --- /dev/null
> +++ b/gdb/features/aarch64-sve.c
> @@ -0,0 +1,158 @@
> +/* Copyright (C) 2017 Free Software Foundation, Inc.
2018
Thanks,
Simon
next prev parent reply other threads:[~2018-05-31 11:10 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-05-11 10:53 [PATCH 0/8] Add SVE support for Aarch64 GDB Alan Hayward
2018-05-11 10:53 ` [PATCH 4/8] Enable SVE for GDB Alan Hayward
2018-05-31 12:22 ` Simon Marchi
2018-05-31 14:58 ` Pedro Alves
2018-05-31 16:13 ` Pedro Alves
2018-05-31 16:20 ` Alan Hayward
[not found] ` <2c87cd8d-c608-4ccf-b16a-635168dbb250@redhat.com>
2018-05-31 18:06 ` Alan Hayward
2018-05-11 10:53 ` [PATCH 6/8] Aarch64 SVE pseudo register support Alan Hayward
2018-05-31 13:26 ` Simon Marchi
2018-05-31 14:59 ` Pedro Alves
2018-05-11 10:53 ` [PATCH 2/8] Function for reading the Aarch64 SVE vector length Alan Hayward
2018-05-31 12:06 ` Simon Marchi
2018-05-31 14:57 ` Pedro Alves
2018-06-05 20:01 ` Sergio Durigan Junior
2018-06-05 22:06 ` [PATCH] Guard declarations of 'sve_*_from_*' macros on Aarch64 (and unbreak build) Sergio Durigan Junior
2018-06-05 23:37 ` Sergio Durigan Junior
2018-06-06 7:34 ` Alan Hayward
2018-06-06 21:19 ` Simon Marchi
2018-06-06 21:36 ` Sergio Durigan Junior
2018-05-11 10:53 ` [PATCH 8/8] Ptrace support for Aarch64 SVE Alan Hayward
2018-05-31 13:40 ` Simon Marchi
2018-05-31 14:56 ` Alan Hayward
2018-06-01 15:17 ` Simon Marchi
2018-06-04 15:49 ` Alan Hayward
2018-05-31 20:17 ` Simon Marchi
2018-05-11 10:53 ` [PATCH 1/8] Add Aarch64 SVE target description Alan Hayward
2018-05-11 14:56 ` Eli Zaretskii
2018-05-11 16:46 ` Alan Hayward
2018-05-31 11:56 ` Simon Marchi [this message]
2018-05-31 14:12 ` Alan Hayward
2018-05-11 10:53 ` [PATCH 7/8] Add methods to gdbserver regcache and raw_compare Alan Hayward
2018-05-11 11:52 ` [PATCH 5/8] Add aarch64 psuedo help functions Alan Hayward
2018-05-31 13:22 ` Simon Marchi
2018-05-31 15:20 ` Pedro Alves
2018-06-04 13:13 ` Alan Hayward
2018-05-11 12:12 ` [PATCH 3/8] Add SVE register defines Alan Hayward
2018-06-01 8:33 ` Alan Hayward
2018-06-01 15:18 ` Simon Marchi
2018-05-22 11:00 ` [PATCH 0/8] Add SVE support for Aarch64 GDB Alan Hayward
2018-05-29 12:09 ` [PING 2][PATCH " Alan Hayward
2018-05-29 14:35 ` Omair Javaid
2018-05-29 14:59 ` Alan Hayward
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=e5a7a3c9-7dcc-0103-f2a9-61a35b8b8401@ericsson.com \
--to=simon.marchi@ericsson.com \
--cc=alan.hayward@arm.com \
--cc=gdb-patches@sourceware.org \
--cc=nd@arm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox