From: Pierrick Bouvier <pierrick.bouvier@linaro.org>
To: "Philippe Mathieu-Daudé" <philmd@linaro.org>, qemu-devel@nongnu.org
Cc: "Alex Bennée" <alex.bennee@linaro.org>,
"Markus Armbruster" <armbru@redhat.com>,
"Richard Henderson" <richard.henderson@linaro.org>
Subject: Re: [RFC PATCH-for-10.1 19/19] system/vl: Filter machine list for binary using machine_binary_filter()
Date: Fri, 4 Apr 2025 10:10:23 -0700 [thread overview]
Message-ID: <5fa51205-4172-41d4-8bd9-5a95e8da24c2@linaro.org> (raw)
In-Reply-To: <20250403234914.9154-20-philmd@linaro.org>
On 4/3/25 16:49, Philippe Mathieu-Daudé wrote:
> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
> ---
> system/vl.c | 24 ++++++++++++++++++++++++
> 1 file changed, 24 insertions(+)
>
> diff --git a/system/vl.c b/system/vl.c
> index d8a0fe713c9..554f5f2a467 100644
> --- a/system/vl.c
> +++ b/system/vl.c
> @@ -27,6 +27,8 @@
> #include "qemu/datadir.h"
> #include "qemu/units.h"
> #include "qemu/module.h"
> +#include "qemu/target_info.h"
> +#include "qemu/target_info-qom.h"
> #include "exec/cpu-common.h"
> #include "exec/page-vary.h"
> #include "hw/qdev-properties.h"
> @@ -833,14 +835,29 @@ static bool usb_parse(const char *cmdline, Error **errp)
> /***********************************************************/
> /* machine registration */
>
> +static char *machine_binary_filter(void)
> +{
> + if (target_info_is_stub()) {
> + return NULL;
> + }
> + return g_strconcat(TYPE_LEGACY_BINARY_PREFIX,
> + "qemu-system-", target_name(), NULL);
No, we should not have such things.
We can make it work with proper QOM types, defined by target, instead of
relying on string construction/compare like this.
> +}
> +
> static MachineClass *find_machine(const char *name, GSList *machines)
> {
> GSList *el;
> + g_autofree char *binary_filter = machine_binary_filter();
>
> for (el = machines; el; el = el->next) {
> MachineClass *mc = el->data;
>
> if (!strcmp(mc->name, name) || !g_strcmp0(mc->alias, name)) {
> + if (binary_filter && !object_class_dynamic_cast(el->data,
> + binary_filter)) {
> + /* Machine is not for this binary: fail */
> + return NULL;
> + }
> return mc;
> }
> }
> @@ -1563,6 +1580,7 @@ static void machine_help_func(const QDict *qdict)
> g_autoptr(GSList) machines = NULL;
> GSList *el;
> const char *type = qdict_get_try_str(qdict, "type");
> + g_autofree char *binary_filter = machine_binary_filter();
>
> machines = object_class_get_list(TYPE_MACHINE, false);
If we define a proper TYPE_TARGET_MACHINE per target, and we add this to
TargetInfo, this can become:
machines = object_class_get_list(target_machine_type(), false);
And we don't need any other string hack to detect what is the correct type.
> if (type) {
> @@ -1577,6 +1595,12 @@ static void machine_help_func(const QDict *qdict)
> machines = g_slist_sort(machines, machine_class_cmp);
> for (el = machines; el; el = el->next) {
> MachineClass *mc = el->data;
> +
> + if (binary_filter && !object_class_dynamic_cast(el->data,
> + binary_filter)) {
> + /* Machine is not for this binary: skip */
> + continue;
> + }
With the approach above, this is not needed anymore.
> if (mc->alias) {
> printf("%-20s %s (alias of %s)\n", mc->alias, mc->desc, mc->name);
> }
I think we are missing a commit here, defining a proper
TYPE_TARGET_MACHINE_PREFIX, that is target dependent, instead of the
TYPE_LEGACY_BINARY_PREFIX.
And we should include in this type in TargetInfo, the same way it was
done for cpus.
next prev parent reply other threads:[~2025-04-04 17:13 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-04-03 23:48 [RFC PATCH-for-10.1 00/19] qemu: Introduce TargetInfo API (for single binary) Philippe Mathieu-Daudé
2025-04-03 23:48 ` [RFC PATCH-for-10.1 01/19] qemu: Introduce TargetInfo API in 'target_info.h' Philippe Mathieu-Daudé
2025-04-04 16:41 ` Pierrick Bouvier
2025-04-03 23:48 ` [RFC PATCH-for-10.1 02/19] qemu: Convert target_name() to TargetInfo API Philippe Mathieu-Daudé
2025-04-04 16:42 ` Pierrick Bouvier
2025-04-03 23:48 ` [RFC PATCH-for-10.1 03/19] qemu: Factor target_system_arch() out Philippe Mathieu-Daudé
2025-04-04 16:44 ` Pierrick Bouvier
2025-04-03 23:48 ` [RFC PATCH-for-10.1 04/19] qemu: Convert target_words_bigendian() to TargetInfo API Philippe Mathieu-Daudé
2025-04-04 16:45 ` Pierrick Bouvier
2025-04-03 23:49 ` [RFC PATCH-for-10.1 05/19] qemu: Introduce target_long_bits() Philippe Mathieu-Daudé
2025-04-04 16:46 ` Pierrick Bouvier
2025-04-03 23:49 ` [RFC PATCH-for-10.1 06/19] target/tricore: Replace TARGET_LONG_BITS -> target_long_bits() Philippe Mathieu-Daudé
2025-04-04 16:48 ` Pierrick Bouvier
2025-04-04 17:53 ` Philippe Mathieu-Daudé
2025-04-17 18:00 ` Paolo Bonzini
2025-04-17 18:32 ` Philippe Mathieu-Daudé
2025-04-03 23:49 ` [RFC PATCH-for-10.1 07/19] target/hppa: " Philippe Mathieu-Daudé
2025-04-04 16:48 ` Pierrick Bouvier
2025-04-04 17:54 ` Philippe Mathieu-Daudé
2025-04-17 17:27 ` Paolo Bonzini
2025-04-03 23:49 ` [RFC PATCH-for-10.1 08/19] target/riscv: " Philippe Mathieu-Daudé
2025-04-04 16:48 ` Pierrick Bouvier
2025-04-04 17:54 ` Philippe Mathieu-Daudé
2025-04-03 23:49 ` [RFC PATCH-for-10.1 09/19] qemu: Introduce target_cpu_type() Philippe Mathieu-Daudé
2025-04-04 16:48 ` Pierrick Bouvier
2025-04-03 23:49 ` [RFC PATCH-for-10.1 10/19] cpus: Replace CPU_RESOLVING_TYPE -> target_cpu_type() Philippe Mathieu-Daudé
2025-04-04 16:51 ` Pierrick Bouvier
2025-04-03 23:49 ` [RFC PATCH-for-10.1 11/19] accel/tcg: " Philippe Mathieu-Daudé
2025-04-04 16:51 ` Pierrick Bouvier
2025-04-04 17:56 ` Philippe Mathieu-Daudé
2025-04-04 18:04 ` Pierrick Bouvier
2025-04-03 23:49 ` [RFC PATCH-for-10.1 12/19] cpus: Move target-agnostic methods out of cpu-target.c Philippe Mathieu-Daudé
2025-04-04 16:53 ` Pierrick Bouvier
2025-04-03 23:49 ` [RFC PATCH-for-10.1 13/19] accel: Replace CPU_RESOLVING_TYPE -> target_cpu_type() Philippe Mathieu-Daudé
2025-04-04 16:52 ` Pierrick Bouvier
2025-04-03 23:49 ` [RFC PATCH-for-10.1 14/19] accel: Implement accel_init_ops_interfaces() for both system/user mode Philippe Mathieu-Daudé
2025-04-04 16:56 ` Pierrick Bouvier
2025-04-03 23:49 ` [RFC PATCH-for-10.1 15/19] accel: Include missing 'qemu/accel.h' header in accel-internal.h Philippe Mathieu-Daudé
2025-04-04 16:56 ` Pierrick Bouvier
2025-04-03 23:49 ` [RFC PATCH-for-10.1 16/19] accel: Make AccelCPUClass structure target-agnostic Philippe Mathieu-Daudé
2025-04-04 16:57 ` Pierrick Bouvier
2025-04-03 23:49 ` [RFC PATCH-for-10.1 17/19] accel: Move target-agnostic code from accel-target.c -> accel-common.c Philippe Mathieu-Daudé
2025-04-04 16:59 ` Pierrick Bouvier
2025-04-17 16:42 ` Philippe Mathieu-Daudé
2025-04-03 23:49 ` [RFC PATCH-for-10.1 18/19] qemu: Prepare per-binary QOM filter via TYPE_BINARY_PREFIX Philippe Mathieu-Daudé
2025-04-04 17:04 ` Pierrick Bouvier
2025-04-03 23:49 ` [RFC PATCH-for-10.1 19/19] system/vl: Filter machine list for binary using machine_binary_filter() Philippe Mathieu-Daudé
2025-04-04 17:10 ` Pierrick Bouvier [this message]
2025-04-04 18:01 ` Philippe Mathieu-Daudé
2025-04-04 18:08 ` Pierrick Bouvier
2025-04-04 18:11 ` Pierrick Bouvier
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=5fa51205-4172-41d4-8bd9-5a95e8da24c2@linaro.org \
--to=pierrick.bouvier@linaro.org \
--cc=alex.bennee@linaro.org \
--cc=armbru@redhat.com \
--cc=philmd@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=richard.henderson@linaro.org \
/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;
as well as URLs for NNTP newsgroup(s).