]> git.proxmox.com Git - mirror_qemu.git/commit
hw/i2c/aspeed: Fix old reg slave receive
authorPeter Delevoryas <peter@pjd.dev>
Mon, 24 Oct 2022 09:20:15 +0000 (11:20 +0200)
committerCédric Le Goater <clg@kaod.org>
Mon, 24 Oct 2022 09:20:15 +0000 (11:20 +0200)
commit3648d31fa81c4a391b8cd74e9fcd410a74f72383
treed101baeb9afbc0a5537b1ea8756004421d4c4ccc
parent0529245488865038344d64fff7ee05864d3d17f6
hw/i2c/aspeed: Fix old reg slave receive

I think when Klaus ported his slave mode changes from the original patch
series to the rewritten I2C module, he changed the behavior of the first
byte that is received by the slave device.

What's supposed to happen is that the AspeedI2CBus's slave device's
i2c_event callback should run, and if the event is "send_async", then it
should populate the byte buffer with the 8-bit I2C address that is being
sent to. Since we only support "send_async", the lowest bit should
always be 0 (indicating that the master is requesting to send data).

This is the code Klaus had previously, for reference. [1]

    switch (event) {
    case I2C_START_SEND:
        bus->buf = bus->dev_addr << 1;

        bus->buf &= I2CD_BYTE_BUF_RX_MASK;
        bus->buf <<= I2CD_BYTE_BUF_RX_SHIFT;

        bus->intr_status |= (I2CD_INTR_SLAVE_ADDR_RX_MATCH | I2CD_INTR_RX_DONE);
        aspeed_i2c_set_state(bus, I2CD_STXD);

        break;

[1]: https://lore.kernel.org/qemu-devel/20220331165737.1073520-4-its@irrelevant.dk/

Fixes: a8d48f59cd021b25 ("hw/i2c/aspeed: add slave device in old register mode")
Signed-off-by: Peter Delevoryas <peter@pjd.dev>
Reviewed-by: Klaus Jensen <k.jensen@samsung.com>
Message-Id: <20220820225712.713209-2-peter@pjd.dev>
Signed-off-by: Cédric Le Goater <clg@kaod.org>
hw/i2c/aspeed_i2c.c
include/hw/i2c/aspeed_i2c.h