qemu-stable
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH v2] hw/audio/virtio-snd: Use device endianness instead of tar


From: Philippe Mathieu-Daudé
Subject: Re: [PATCH v2] hw/audio/virtio-snd: Use device endianness instead of target one
Date: Mon, 22 Apr 2024 16:25:44 +0200
User-agent: Mozilla Thunderbird

On 22/4/24 15:45, Manos Pitsidianakis wrote:
On Mon, 22 Apr 2024 16:13, Philippe Mathieu-Daudé <philmd@linaro.org> wrote:
Since VirtIO devices can change endianness at runtime,
we need to use the device endianness, not the target
one.

Hey Philippe, can you clarify what do you mean by they can change endianness at runtime?

The target's one is used because that's the one it will be using to do I/O with its kernel's audio interface.

See for example in SysemuCPUOps:

    /**
     * @virtio_is_big_endian: Callback to return %true if a CPU which
     * supports runtime configurable endianness is currently big-endian.
     * Non-configurable CPUs can use the default implementation of this
     * method.
     * This method should not be used by any callers other than the
     * pre-1.0 virtio devices.
     */
    bool (*virtio_is_big_endian)(CPUState *cpu);

and:

/**
 * cpu_virtio_is_big_endian:
 * @cpu: CPU

 * Returns %true if a CPU which supports runtime configurable endianness
 * is currently big-endian.
 */
bool cpu_virtio_is_big_endian(CPUState *cpu);

From commit 616a655219 ("virtio: add endian-ambivalent support to VirtIODevice"):

    Some CPU families can dynamically change their endianness. This
    means we can have little endian ppc or big endian arm guests for
    example. This has an impact on legacy virtio data structures since
    they are target endian.
    We hence introduce a new property to track the endianness of each
    virtio device. It is reasonnably assumed that endianness won't
    change while the device is in use : we hence capture the device
    endianness when it gets reset.



Cc: qemu-stable@nongnu.org
Fixes: eb9ad377bb ("virtio-sound: handle control messages and streams")
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
---
hw/audio/virtio-snd.c | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/hw/audio/virtio-snd.c b/hw/audio/virtio-snd.c
index c80b58bf5d..82c5647660 100644
--- a/hw/audio/virtio-snd.c
+++ b/hw/audio/virtio-snd.c
@@ -395,13 +395,15 @@ static uint32_t virtio_snd_get_qemu_freq(uint32_t rate)  * Get QEMU Audiosystem compatible audsettings from virtio based pcm stream
 * params.
 */
-static void virtio_snd_get_qemu_audsettings(audsettings *as,
+static void virtio_snd_get_qemu_audsettings(VirtIOSound *s, audsettings *as,                                             virtio_snd_pcm_set_params *params)
{
+    VirtIODevice *vdev = VIRTIO_DEVICE(s);
+
    as->nchannels = MIN(AUDIO_MAX_CHANNELS, params->channels);
    as->fmt = virtio_snd_get_qemu_format(params->format);
    as->freq = virtio_snd_get_qemu_freq(params->rate);
-    as->endianness = target_words_bigendian() ? 1 : 0;
+    as->endianness = virtio_is_big_endian(vdev) ? 1 : 0;
}

/*
@@ -464,7 +466,7 @@ static uint32_t virtio_snd_pcm_prepare(VirtIOSound *s, uint32_t stream_id)
        s->pcm->streams[stream_id] = stream;
    }

-    virtio_snd_get_qemu_audsettings(&as, params);
+    virtio_snd_get_qemu_audsettings(s, &as, params);
    stream->info.direction = stream_id < s->snd_conf.streams / 2 +
        (s->snd_conf.streams & 1) ? VIRTIO_SND_D_OUTPUT : VIRTIO_SND_D_INPUT;
    stream->info.hdr.hda_fn_nid = VIRTIO_SOUND_HDA_FN_NID;
--
2.41.0





reply via email to

[Prev in Thread] Current Thread [Next in Thread]