From: Alexandre Courbot <acourbot@xxxxxxxxxxxx> When running memcpy_toio: memcpy_toio(send_obj->share_buf, buf, len); it was found that errors appear if len is not a multiple of 8: [58.350841] mtk-mdp 14001000.rdma: processing failed: -22 This is because in ARM64, memcpy_toio does byte-size access when the length is not a multiple of 8 while access to the vpu iomem must be 4 aligned. This patch ensures the copy of a multiple of 8 size by calling round_up(len, 8) when copying Fixes: e6599adfad30 ("media: mtk-vpu: avoid unaligned access to DTCM buffer.") Signed-off-by: Alexandre Courbot <acourbot@xxxxxxxxxxxx> Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@xxxxxxxxxxxxx> Reviewed-by: Houlong Wei <houlong.wei@xxxxxxxxxxxx> --- changes since v4: 1. replace data[SHARE_BUF_SIZE]; with data[sizeof(send_obj->share_buf)] 2. Add the explanation for the failure in commit log and inline doc changes since v3: 1. multile -> multiple 2. add inline doc changes since v2: 1. do the extra copy only if len is not multiple of 8 changes since v1: 1. change sign-off-by tags 2. change values to memset drivers/media/platform/mtk-vpu/mtk_vpu.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/drivers/media/platform/mtk-vpu/mtk_vpu.c b/drivers/media/platform/mtk-vpu/mtk_vpu.c index 7f1647da0ade..2ef93db239b9 100644 --- a/drivers/media/platform/mtk-vpu/mtk_vpu.c +++ b/drivers/media/platform/mtk-vpu/mtk_vpu.c @@ -349,7 +349,22 @@ int vpu_ipi_send(struct platform_device *pdev, } } while (vpu_cfg_readl(vpu, HOST_TO_VPU)); - memcpy_toio(send_obj->share_buf, buf, len); + /* + * On Arm64, the memcpy_toio does byte-size access as long as address or length are + * not 8 aligned. Access to the vpu iomem must be 4 aligned. Byte-size access is + * not allowed and cause processing to fail. Therefore make sure the length + * sent to memcpy_toio is a multiply of 8. + */ + if (len % 8 != 0) { + unsigned char data[sizeof(send_obj->share_buf)]; + + memset(data + len, 0, sizeof(data) - len); + memcpy(data, buf, len); + memcpy_toio(send_obj->share_buf, data, round_up(len, 8)); + } else { + memcpy_toio(send_obj->share_buf, buf, len); + } + writel(len, &send_obj->len); writel(id, &send_obj->id); -- 2.17.1