On Thu, Jun 12, 2025 at 9:08 AM Konrad Dybcio
<konrad.dybcio@xxxxxxxxxxxxxxxx> wrote:
>
> On 6/12/25 6:04 PM, Alexey Klimov wrote:
> > On Mon Jun 9, 2025 at 7:24 PM BST, Rob Clark wrote:
> >> If we have a newer dtb than kernel, we could end up in a situation where
> >> the GPU device is present in the dtb, but not in the drivers device
> >> table. We don't want this to prevent the display from probing. So
> >> check that we recognize the GPU before adding the GPU component.
> >>
> >> v2: use %pOF
> >>
> >> Signed-off-by: Rob Clark <robin.clark@xxxxxxxxxxxxxxxx>
> >> ---
> >> drivers/gpu/drm/msm/adreno/adreno_device.c | 29 ++++++++++++++++++----
> >> drivers/gpu/drm/msm/msm_drv.c | 2 +-
> >> drivers/gpu/drm/msm/msm_gpu.h | 1 +
> >> 3 files changed, 26 insertions(+), 6 deletions(-)
> >>
> >> diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c
> >> index 778e6ae7f137..0d12454b1f2e 100644
> >> --- a/drivers/gpu/drm/msm/adreno/adreno_device.c
> >> +++ b/drivers/gpu/drm/msm/adreno/adreno_device.c
> >> @@ -178,6 +178,26 @@ static int find_chipid(struct device_node *node, uint32_t *chipid)
> >> return 0;
> >> }
> >>
> >> +bool adreno_has_gpu(struct device_node *node)
> >> +{
> >> + const struct adreno_info *info;
> >> + uint32_t chip_id;
> >> + int ret;
> >> +
> >> + ret = find_chipid(node, &chip_id);
> >> + if (ret)
> >> + return false;
> >> +
> >> + info = adreno_info(chip_id);
> >> + if (!info) {
> >> + pr_warn("%pOF: Unknown GPU revision: %"ADRENO_CHIPID_FMT"\n",
> >> + node, ADRENO_CHIPID_ARGS(chip_id));
> >> + return false;
> >> + }
> >> +
> >> + return true;
> >> +}
> >> +
> >> static int adreno_bind(struct device *dev, struct device *master, void *data)
> >> {
> >> static struct adreno_platform_config config = {};
> >> @@ -188,18 +208,17 @@ static int adreno_bind(struct device *dev, struct device *master, void *data)
> >> int ret;
> >>
> >> ret = find_chipid(dev->of_node, &config.chip_id);
> >> - if (ret)
> >> + /* We shouldn't have gotten this far if we can't parse the chip_id */
> >> + if (WARN_ON(ret))
> >> return ret;
> >
> > I just hit this with linux-next on qrb2210 RB1 [1].
> >
> > Is it expected an warning now or do we miss some device tree updates on
> > linux-next for RB1?
> >
> > I don't recall seeing such warnings previously.
> >
> > Thanks,
> > Alexey
> >
> > [1]:
> >
> > msm_dpu 5e01000.display-controller: bound 5e94000.dsi (ops dsi_ops [msm])
> > ------------[ cut here ]------------
> > WARNING: CPU: 0 PID: 242 at drivers/gpu/drm/msm/adreno/adreno_device.c:224 adreno_bind+0x90/0x120 [msm]
> > Modules linked in: q6asm_dai q6routing q6afe_dai q6adm q6asm q6afe_clocks snd_q6dsp_common q6afe q6core apr pdr_interface qrtr_smd qcom_pd_mapper qcom_pdr_msg mcp251xfd ath10k_snoc snd_soc_wsa881x_i2c snd_soc_wsa881x_common can_dev lontium_lt9611uxc(+) ath10k_core ath mac80211 hci_uart btqca btbcm libarc4 msm snd_soc_sm8250 qrtr bluetooth drm_exec snd_soc_qcom_sdw qcom_q6v5_pas llcc_qcom snd_soc_qcom_common lmh qcom_wdt ocmem cfg80211 ecdh_generic qcom_pil_info pinctrl_sm6115_lpass_lpi gpu_sched ecc drm_display_helper rfkill qcom_q6v5 pinctrl_lpass_lpi qcom_sysmon pwrseq_core lpasscc_sm6115 dispcc_qcm2290 qcom_common snd_soc_lpass_va_macro cec snd_soc_lpass_rx_macro drm_dp_aux_bus snd_soc_lpass_tx_macro qcom_glink_smem gpucc_qcm2290 snd_soc_pm4125 mdt_loader snd_soc_lpass_macro_common qmi_helpers snd_soc_pm4125_sdw soundwire_qcom regmap_sdw slimbus qcom_pmic_tcpm qcom_usb_vbus_regulator drm_client_lib tcpm rtc_pm8xxx snd_soc_wcd_mbhc aux_hpd_bridge qcom_pon qcrypto soundwire_bus sha256 qcom_stats gpi
> > spi_geni_qcom i2c_qcom_geni rpmsg_ctrl libsha256_generic libsha256 rpmsg_char qcom_rng sha256_arm64 authenc icc_bwmon phy_qcom_qmp_usbc libdes typec phy_qcom_qusb2 display_connector i2c_gpio rmtfs_mem drm_kms_helper socinfo fuse drm backlight dm_mod ip_tables x_tables ipv6
> > CPU: 0 UID: 0 PID: 242 Comm: (udev-worker) Not tainted 6.16.0-rc1-next-20250612-00025-g0ce0d3974333-dirty #2 PREEMPT
> > Hardware name: Qualcomm Technologies, Inc. Robotics RB1 (DT)
> > pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> > pc : adreno_bind+0x90/0x120 [msm]
> > lr : adreno_bind+0x50/0x120 [msm]
> > sp : ffff8000813f3580
> > x29: ffff8000813f3580 x28: ffff000008ae8800 x27: ffff000007c63700
> > x26: ffffca4a2814b860 x25: ffff000008b26880 x24: ffffca4a24922000
> > x23: ffffca4a249229d8 x22: ffff000009838800 x21: ffff000008b26880
> > x20: ffff000002ce4410 x19: ffffca4a2495a710 x18: 0000000000000006
> > x17: 6f5f697364207370 x16: 6f28206973642e30 x15: 0720072007200720
> > x14: 0000000000000000 x13: 0000000000000000 x12: 0101010101010101
> > x11: 7f7f7f7f7f7f7f7f x10: ffffca4a2866e1b2 x9 : 0000000000000002
> > x8 : 0101010101010101 x7 : 7f7f7f7f7f7f7f7f x6 : fefefeff2f383131
> > x5 : ffff00007fc1b968 x4 : 0000000000000000 x3 : 000000000000001e
> > x2 : 0000000000000001 x1 : 0000000007000200 x0 : ffffca4a2495a710
> > Call trace:
> > adreno_bind+0x90/0x120 [msm] (P)
> > component_bind_all+0x100/0x23c
> > msm_drm_bind+0x148/0x3e0 [msm]
> > try_to_bring_up_aggregate_device+0x16c/0x1e0
> > __component_add+0xa4/0x174
> > component_add+0x14/0x20
> > dsi_dev_attach+0x20/0x34 [msm]
> > dsi_host_attach+0x58/0x98 [msm]
> > devm_mipi_dsi_attach+0x34/0x90
> > lt9611uxc_attach_dsi.isra.0+0x94/0x124 [lontium_lt9611uxc]
> > lt9611uxc_probe+0x568/0x604 [lontium_lt9611uxc]
> > i2c_device_probe+0x158/0x32c
> > really_probe+0xbc/0x2b4
> > __driver_probe_device+0x78/0x120
> > driver_probe_device+0x3c/0x154
> > __driver_attach+0x90/0x1a0
> > bus_for_each_dev+0x68/0xb8
> > driver_attach+0x24/0x30
> > bus_add_driver+0xe4/0x208
> > driver_register+0x68/0x124
> > i2c_register_driver+0x48/0xcc
> > lt9611uxc_driver_init+0x20/0x1000 [lontium_lt9611uxc]
> > do_one_initcall+0x60/0x1d4
> > do_init_module+0x54/0x23c
> > load_module+0x1730/0x1cc0
> > init_module_from_file+0x74/0xa0
> > __arm64_sys_finit_module+0x130/0x2f8
> > invoke_syscall+0x48/0x104
> > el0_svc_common.constprop.0+0xc0/0xe0
> > do_el0_svc+0x1c/0x28
> > el0_svc+0x2c/0x80
> > el0t_64_sync_handler+0x10c/0x138
> > el0t_64_sync+0x198/0x19c
> > ---[ end trace 0000000000000000 ]---
> > adreno 5900000.gpu: supply vdd not found, using dummy regulator
> > adreno 5900000.gpu: supply vddcx not found, using dummy regulator
> > msm_dpu 5e01000.display-controller: bound 5900000.gpu (ops a3xx_ops [msm])
> > [drm:dpu_kms_hw_init:1173] dpu hardware revision:0x60050000
> > dummy 1-0045: No cache used with register defaults set!
> > [drm] Initialized msm 1.12.0 for 5e01000.display-controller on minor 0
> > msm_dpu 5e01000.display-controller: [drm:adreno_request_fw [msm]] loaded qcom/a702_sqe.fw from new location
>
> Looks like we should be doing this instead
>
> diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c
> index 5e7307567239..16e7ac444efd 100644
> --- a/drivers/gpu/drm/msm/adreno/adreno_device.c
> +++ b/drivers/gpu/drm/msm/adreno/adreno_device.c
> @@ -221,7 +221,7 @@ static int adreno_bind(struct device *dev, struct device *master, void *data)
>
> info = adreno_info(config.chip_id);
> /* We shouldn't have gotten this far if we don't recognize the GPU: */
> - if (!WARN_ON(info))
> + if (WARN_ON(!info))
> return -ENXIO;
>
> config.info = info;
Oh, yes, indeed
BR,
-R
