[bug report] drm: xlnx: DRM/KMS driver for Xilinx ZynqMP DisplayPort Subsystem
Hyun Kwon
hyun.kwon at xilinx.com
Wed Jul 29 18:21:16 UTC 2020
Hi Dan,
Thanks for sharing.
On Mon, Jul 27, 2020 at 04:18:25AM -0700, dan.carpenter at oracle.com wrote:
> Hello Hyun Kwon,
>
> The patch d76271d22694: "drm: xlnx: DRM/KMS driver for Xilinx ZynqMP
> DisplayPort Subsystem" from Jul 7, 2018, leads to the following
> static checker warning:
>
> drivers/gpu/drm/xlnx/zynqmp_dp.c:594 zynqmp_dp_mode_configure()
> error: iterator underflow 'bws' (-1)-2
>
> drivers/gpu/drm/xlnx/zynqmp_dp.c
> 566 static int zynqmp_dp_mode_configure(struct zynqmp_dp *dp, int pclock,
> 567 u8 current_bw)
> 568 {
> 569 int max_rate = dp->link_config.max_rate;
> 570 u8 bws[3] = { DP_LINK_BW_1_62, DP_LINK_BW_2_7, DP_LINK_BW_5_4 };
> 571 u8 max_lanes = dp->link_config.max_lanes;
> 572 u8 max_link_rate_code = drm_dp_link_rate_to_bw_code(max_rate);
> 573 u8 bpp = dp->config.bpp;
> 574 u8 lane_cnt;
> 575 s8 i;
> 576
> 577 if (current_bw == DP_LINK_BW_1_62) {
> 578 dev_err(dp->dev, "can't downshift. already lowest link rate\n");
> 579 return -EINVAL;
> 580 }
There are checks like this that ensure the 'i' to be not negative, but I agree
it should be made clearer. I've sumitted a fix for it [1].
[1] https://lists.freedesktop.org/archives/dri-devel/2020-July/274190.html
Thanks,
-hyun
> 581
> 582 for (i = ARRAY_SIZE(bws) - 1; i >= 0; i--) {
> ^^^^^^
> This exits with i == -1.
>
> 583 if (current_bw && bws[i] >= current_bw)
> 584 continue;
> 585
> 586 if (bws[i] <= max_link_rate_code)
> 587 break;
> 588 }
> 589
> 590 for (lane_cnt = 1; lane_cnt <= max_lanes; lane_cnt <<= 1) {
> 591 int bw;
> 592 u32 rate;
> 593
> 594 bw = drm_dp_bw_code_to_link_rate(bws[i]);
> ^^^^^^
> Potential negative array offset.
>
> 595 rate = zynqmp_dp_max_rate(bw, lane_cnt, bpp);
> 596 if (pclock <= rate) {
> 597 dp->mode.bw_code = bws[i];
> 598 dp->mode.lane_cnt = lane_cnt;
> 599 dp->mode.pclock = pclock;
> 600 return dp->mode.bw_code;
> 601 }
> 602 }
> 603
> 604 dev_err(dp->dev, "failed to configure link values\n");
> 605
> 606 return -EINVAL;
> 607 }
>
> regards,
> dan carpenter
More information about the dri-devel
mailing list