| Message ID | 20240531-x1e80100-phy-add-gen4x4-v1-2-5c841dae7850@linaro.org |
|---|---|
| State | New |
| Headers | show |
| Series | phy: qcom: qmp-pcie: Add support for Gen4 4-lane mode for X1E80100 | expand |
On Fri, May 31, 2024 at 07:06:45PM +0300, Abel Vesa wrote: > The PCIe 6th instance from X1E80100 can be used in both 4-lane mode or > 2-lane mode. Add the configuration and compatible for the 4-lane mode. > > Signed-off-by: Abel Vesa <abel.vesa@linaro.org> > --- > drivers/phy/qualcomm/phy-qcom-qmp-pcie.c | 42 ++++++++++++++++++++++++++++++++ > 1 file changed, 42 insertions(+) > Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
On Fri, May 31, 2024 at 07:06:45PM +0300, Abel Vesa wrote: > The PCIe 6th instance from X1E80100 can be used in both 4-lane mode or > 2-lane mode. Add the configuration and compatible for the 4-lane mode. Same language nits as for patch 1/1. > Signed-off-by: Abel Vesa <abel.vesa@linaro.org> I tried this patch along with the DT changes and the link on the CRD still comes up as 2-lane: qcom-pcie 1bf8000.pci: PCIe Gen.4 x2 link up so something appears to be wrong here. (I noticed the same with your next branch last week.) How did you test this? Does the link actually come up as 4-lane for you? Johan
On 24-06-03 14:46:12, Johan Hovold wrote: > On Fri, May 31, 2024 at 07:06:45PM +0300, Abel Vesa wrote: > > The PCIe 6th instance from X1E80100 can be used in both 4-lane mode or > > 2-lane mode. Add the configuration and compatible for the 4-lane mode. > > Same language nits as for patch 1/1. > > > Signed-off-by: Abel Vesa <abel.vesa@linaro.org> > > I tried this patch along with the DT changes and the link on the CRD > still comes up as 2-lane: > > qcom-pcie 1bf8000.pci: PCIe Gen.4 x2 link up > > so something appears to be wrong here. (I noticed the same with your > next branch last week.) > > How did you test this? Does the link actually come up as 4-lane for you? This is the PHY part. The controller needs some changes as well. Yes, as of yet, I'm not able to bring the link up in 4-lanes mode. This however doesn't mean the PHY sequence is incorrect. But, I agree, maybe I should hold on to the PHY changes as well until we get the controller side working as well. Thanks for reviewing. > > Johan
On Tue, Jun 04, 2024 at 03:25:52PM +0300, Abel Vesa wrote: > On 24-06-03 14:46:12, Johan Hovold wrote: > > I tried this patch along with the DT changes and the link on the CRD > > still comes up as 2-lane: > > > > qcom-pcie 1bf8000.pci: PCIe Gen.4 x2 link up > > > > so something appears to be wrong here. (I noticed the same with your > > next branch last week.) > > > > How did you test this? Does the link actually come up as 4-lane for you? > > This is the PHY part. The controller needs some changes as well. Are you sure? It wasn't needed on sc8280xp (except for updating num-lanes in DT). > Yes, as of yet, I'm not able to bring the link up in 4-lanes mode. Thanks for confirming. > This however doesn't mean the PHY sequence is incorrect. Not necessarily, but it means it hasn't been fully tested and that it could potentially be the reason for the failed x4 link up. Johan
diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-pcie.c b/drivers/phy/qualcomm/phy-qcom-qmp-pcie.c index 6c796723c8f5..4e0b28da69a7 100644 --- a/drivers/phy/qualcomm/phy-qcom-qmp-pcie.c +++ b/drivers/phy/qualcomm/phy-qcom-qmp-pcie.c @@ -1028,6 +1028,10 @@ static const struct qmp_phy_init_tbl x1e80100_qmp_gen4x2_pcie_ln_shrd_tbl[] = { QMP_PHY_INIT_CFG(QSERDES_V6_LN_SHRD_RX_MARG_COARSE_THRESH6_RATE3, 0x1f), }; +static const struct qmp_phy_init_tbl x1e80100_qmp_gen4x4_pcie_serdes_4ln_tbl[] = { + QMP_PHY_INIT_CFG(QSERDES_V6_COM_PLL_BIAS_EN_CLK_BUFLR_EN, 0x1c), +}; + static const struct qmp_phy_init_tbl x1e80100_qmp_gen4x2_pcie_tx_tbl[] = { QMP_PHY_INIT_CFG(QSERDES_V6_20_TX_RES_CODE_LANE_OFFSET_TX, 0x1d), QMP_PHY_INIT_CFG(QSERDES_V6_20_TX_RES_CODE_LANE_OFFSET_RX, 0x03), @@ -3342,6 +3346,41 @@ static const struct qmp_phy_cfg x1e80100_qmp_gen4x2_pciephy_cfg = { .ln_shrd = x1e80100_qmp_gen4x2_pcie_ln_shrd_tbl, .ln_shrd_num = ARRAY_SIZE(x1e80100_qmp_gen4x2_pcie_ln_shrd_tbl), }, + + .reset_list = sdm845_pciephy_reset_l, + .num_resets = ARRAY_SIZE(sdm845_pciephy_reset_l), + .vreg_list = sm8550_qmp_phy_vreg_l, + .num_vregs = ARRAY_SIZE(sm8550_qmp_phy_vreg_l), + .regs = pciephy_v6_regs_layout, + + .pwrdn_ctrl = SW_PWRDN | REFCLK_DRV_DSBL, + .phy_status = PHYSTATUS_4_20, + .has_nocsr_reset = true, +}; + +static const struct qmp_phy_cfg x1e80100_qmp_gen4x4_pciephy_cfg = { + .lanes = 4, + + .offsets = &qmp_pcie_offsets_v6_20, + + .tbls = { + .serdes = x1e80100_qmp_gen4x2_pcie_serdes_tbl, + .serdes_num = ARRAY_SIZE(x1e80100_qmp_gen4x2_pcie_serdes_tbl), + .tx = x1e80100_qmp_gen4x2_pcie_tx_tbl, + .tx_num = ARRAY_SIZE(x1e80100_qmp_gen4x2_pcie_tx_tbl), + .rx = x1e80100_qmp_gen4x2_pcie_rx_tbl, + .rx_num = ARRAY_SIZE(x1e80100_qmp_gen4x2_pcie_rx_tbl), + .pcs = x1e80100_qmp_gen4x2_pcie_pcs_tbl, + .pcs_num = ARRAY_SIZE(x1e80100_qmp_gen4x2_pcie_pcs_tbl), + .pcs_misc = x1e80100_qmp_gen4x2_pcie_pcs_misc_tbl, + .pcs_misc_num = ARRAY_SIZE(x1e80100_qmp_gen4x2_pcie_pcs_misc_tbl), + .ln_shrd = x1e80100_qmp_gen4x2_pcie_ln_shrd_tbl, + .ln_shrd_num = ARRAY_SIZE(x1e80100_qmp_gen4x2_pcie_ln_shrd_tbl), + }, + + .serdes_4ln_tbl = x1e80100_qmp_gen4x4_pcie_serdes_4ln_tbl, + .serdes_4ln_num = ARRAY_SIZE(x1e80100_qmp_gen4x4_pcie_serdes_4ln_tbl), + .reset_list = sdm845_pciephy_reset_l, .num_resets = ARRAY_SIZE(sdm845_pciephy_reset_l), .vreg_list = sm8550_qmp_phy_vreg_l, @@ -4108,6 +4147,9 @@ static const struct of_device_id qmp_pcie_of_match_table[] = { }, { .compatible = "qcom,x1e80100-qmp-gen4x2-pcie-phy", .data = &x1e80100_qmp_gen4x2_pciephy_cfg, + }, { + .compatible = "qcom,x1e80100-qmp-gen4x4-pcie-phy", + .data = &x1e80100_qmp_gen4x4_pciephy_cfg, }, { }, };
The PCIe 6th instance from X1E80100 can be used in both 4-lane mode or 2-lane mode. Add the configuration and compatible for the 4-lane mode. Signed-off-by: Abel Vesa <abel.vesa@linaro.org> --- drivers/phy/qualcomm/phy-qcom-qmp-pcie.c | 42 ++++++++++++++++++++++++++++++++ 1 file changed, 42 insertions(+)