| Message ID | 20240124162857.111915-1-afd@ti.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [1/2] ARM: multi_v7_defconfig: Add more TI Keystone support | expand |
On 10:28-20240124, Andrew Davis wrote: > The Keystone platforms used their own keystone_defconfig mostly to > enable CONFIG_ARM_LPAE which could not be added to the multi_v7 config. > Now that we have multi_v7_lpae_defconfig/lpae.config target we can and > should use that defconfig for Keystone. Add the remaining must have > options for Keystone support to multi_v7_defconfig. > Please state bloat-o-meter impact on vmlinux. > Signed-off-by: Andrew Davis <afd@ti.com> > --- > arch/arm/configs/multi_v7_defconfig | 16 ++++++++++++++++ > 1 file changed, 16 insertions(+)
On 1/24/24 10:41 AM, Nishanth Menon wrote: > On 10:28-20240124, Andrew Davis wrote: >> The Keystone platforms used their own keystone_defconfig mostly to >> enable CONFIG_ARM_LPAE which could not be added to the multi_v7 config. >> Now that we have multi_v7_lpae_defconfig/lpae.config target we can and >> should use that defconfig for Keystone. Add the remaining must have >> options for Keystone support to multi_v7_defconfig. >> > > Please state bloat-o-meter impact on vmlinux. > $ ./scripts/bloat-o-meter build/vmlinux.baseline build/vmlinux add/remove: 387/0 grow/shrink: 6/0 up/down: 130945/0 (130945) ... Total: Before=22791863, After=22922808, chg +0.57% Seems half of that is TI_KEYSTONE_NETCP, if we don't think anyone will need nfs rootfs boot then we could make it a module: $ ./scripts/bloat-o-meter build/vmlinux.baseline build/vmlinux add/remove: 286/0 grow/shrink: 4/0 up/down: 64393/0 (64393) ... Total: Before=22791863, After=22856256, chg +0.28% Andrew >> Signed-off-by: Andrew Davis <afd@ti.com> >> --- >> arch/arm/configs/multi_v7_defconfig | 16 ++++++++++++++++ >> 1 file changed, 16 insertions(+)
On 10:59-20240124, Andrew Davis wrote: > On 1/24/24 10:41 AM, Nishanth Menon wrote: > > On 10:28-20240124, Andrew Davis wrote: > > > The Keystone platforms used their own keystone_defconfig mostly to > > > enable CONFIG_ARM_LPAE which could not be added to the multi_v7 config. > > > Now that we have multi_v7_lpae_defconfig/lpae.config target we can and > > > should use that defconfig for Keystone. Add the remaining must have > > > options for Keystone support to multi_v7_defconfig. > > > > > > > Please state bloat-o-meter impact on vmlinux. > > > > $ ./scripts/bloat-o-meter build/vmlinux.baseline build/vmlinux > add/remove: 387/0 grow/shrink: 6/0 up/down: 130945/0 (130945) > ... > Total: Before=22791863, After=22922808, chg +0.57% > > > Seems half of that is TI_KEYSTONE_NETCP, if we don't think > anyone will need nfs rootfs boot then we could make it a module: > > > $ ./scripts/bloat-o-meter build/vmlinux.baseline build/vmlinux > add/remove: 286/0 grow/shrink: 4/0 up/down: 64393/0 (64393) > ... > Total: Before=22791863, After=22856256, chg +0.28% Please add that to commit message - the delta and rationale for the config params you think we need to add in.
diff --git a/arch/arm/configs/multi_v7_defconfig b/arch/arm/configs/multi_v7_defconfig index ecb3e286107a4..04ba674e03f60 100644 --- a/arch/arm/configs/multi_v7_defconfig +++ b/arch/arm/configs/multi_v7_defconfig @@ -183,6 +183,7 @@ CONFIG_PCIE_RCAR_HOST=y CONFIG_PCI_RCAR_GEN2=y CONFIG_PCI_LAYERSCAPE=y CONFIG_PCI_DRA7XX_EP=y +CONFIG_PCI_KEYSTONE_HOST=y CONFIG_PCI_ENDPOINT=y CONFIG_PCI_ENDPOINT_CONFIGFS=y CONFIG_PCI_EPF_TEST=m @@ -191,6 +192,7 @@ CONFIG_DEVTMPFS_MOUNT=y CONFIG_OMAP_OCP2SCP=y CONFIG_ARM_SCMI_PROTOCOL=y CONFIG_RASPBERRYPI_FIRMWARE=y +CONFIG_TI_SCI_PROTOCOL=y CONFIG_TRUSTED_FOUNDATIONS=y CONFIG_BCM47XX_NVRAM=y CONFIG_BCM47XX_SPROM=y @@ -280,6 +282,8 @@ CONFIG_DWMAC_DWC_QOS_ETH=y CONFIG_TI_CPSW=y CONFIG_TI_CPSW_SWITCHDEV=y CONFIG_TI_CPTS=y +CONFIG_TI_KEYSTONE_NETCP=y +CONFIG_TI_KEYSTONE_NETCP_ETHSS=y CONFIG_XILINX_EMACLITE=y CONFIG_SFP=m CONFIG_BROADCOM_PHY=y @@ -292,6 +296,8 @@ CONFIG_CAN_AT91=m CONFIG_CAN_FLEXCAN=m CONFIG_CAN_SUN4I=y CONFIG_CAN_XILINXCAN=y +CONFIG_CAN_C_CAN=m +CONFIG_CAN_C_CAN_PLATFORM=m CONFIG_CAN_RCAR=m CONFIG_CAN_MCP251X=y CONFIG_MDIO_MSCC_MIIM=m @@ -335,6 +341,7 @@ CONFIG_INPUT_MISC=y CONFIG_INPUT_PM8941_PWRKEY=y CONFIG_INPUT_MAX77693_HAPTIC=m CONFIG_INPUT_MAX8997_HAPTIC=m +CONFIG_INPUT_GPIO_DECODER=m CONFIG_INPUT_CPCAP_PWRBUTTON=m CONFIG_INPUT_AXP20X_PEK=m CONFIG_INPUT_DA9063_ONKEY=m @@ -436,6 +443,7 @@ CONFIG_SPI_ATMEL_QUADSPI=m CONFIG_SPI_BCM2835=y CONFIG_SPI_BCM2835AUX=y CONFIG_SPI_CADENCE=y +CONFIG_SPI_CADENCE_QUADSPI=y CONFIG_SPI_DAVINCI=y CONFIG_SPI_FSL_QUADSPI=m CONFIG_SPI_GXP=m @@ -1073,6 +1081,7 @@ CONFIG_HWSPINLOCK_OMAP=y CONFIG_HWSPINLOCK_QCOM=y CONFIG_OMAP2PLUS_MBOX=y CONFIG_BCM2835_MBOX=y +CONFIG_TI_MESSAGE_MANAGER=y CONFIG_QCOM_APCS_IPC=y CONFIG_STM32_IPCC=m CONFIG_QCOM_IPCC=y @@ -1133,11 +1142,15 @@ CONFIG_ARCH_TEGRA_2x_SOC=y CONFIG_ARCH_TEGRA_3x_SOC=y CONFIG_ARCH_TEGRA_114_SOC=y CONFIG_ARCH_TEGRA_124_SOC=y +CONFIG_SOC_TI=y +CONFIG_KEYSTONE_NAVIGATOR_QMSS=y +CONFIG_KEYSTONE_NAVIGATOR_DMA=y CONFIG_RASPBERRYPI_POWER=y CONFIG_QCOM_CPR=y CONFIG_QCOM_RPMHPD=y CONFIG_QCOM_RPMPD=y CONFIG_ROCKCHIP_PM_DOMAINS=y +CONFIG_TI_SCI_PM_DOMAINS=y CONFIG_ARM_EXYNOS_BUS_DEVFREQ=m CONFIG_ARM_TEGRA_DEVFREQ=m CONFIG_DEVFREQ_EVENT_EXYNOS_NOCP=m @@ -1193,10 +1206,13 @@ CONFIG_PWM_STM32=m CONFIG_PWM_STM32_LP=m CONFIG_PWM_SUN4I=y CONFIG_PWM_TEGRA=y +CONFIG_PWM_TIECAP=m CONFIG_PWM_VT8500=y CONFIG_KEYSTONE_IRQ=y CONFIG_RESET_MCHP_SPARX5=y CONFIG_RESET_SCMI=y +CONFIG_RESET_TI_SCI=m +CONFIG_RESET_TI_SYSCON=m CONFIG_PHY_SUN4I_USB=y CONFIG_PHY_SUN9I_USB=y CONFIG_PHY_BRCM_USB=m
The Keystone platforms used their own keystone_defconfig mostly to enable CONFIG_ARM_LPAE which could not be added to the multi_v7 config. Now that we have multi_v7_lpae_defconfig/lpae.config target we can and should use that defconfig for Keystone. Add the remaining must have options for Keystone support to multi_v7_defconfig. Signed-off-by: Andrew Davis <afd@ti.com> --- arch/arm/configs/multi_v7_defconfig | 16 ++++++++++++++++ 1 file changed, 16 insertions(+)