| Message ID | 1467382327-9726-1-git-send-email-cov@codeaurora.org (mailing list archive) |
|---|---|
| State | Superseded, archived |
| Delegated to: | Andy Gross |
| Headers | show |
Christopher Covington wrote: > arch/arm64/configs/defconfig | 4 ++++ > scripts/patch-details.sh | 21 ++++++++++++++++++--- I don't think these two files should be combined. > CONFIG_PINCTRL_SINGLE=y > CONFIG_PINCTRL_MSM8916=y > +CONFIG_PINCTRL_QDF2XXX=m > CONFIG_PINCTRL_QCOM_SPMI_PMIC=y > CONFIG_GPIO_SYSFS=y > CONFIG_GPIO_DWAPB=y > @@ -245,6 +246,7 @@ CONFIG_MMC=y > CONFIG_MMC_BLOCK_MINORS=32 > CONFIG_MMC_ARMMMCI=y > CONFIG_MMC_SDHCI=y > +CONFIG_MMC_SDHCI_ACPI=m > CONFIG_MMC_SDHCI_PLTFM=y Why are these =m when all the others are =y?
On 07/01/2016 10:14 AM, Timur Tabi wrote: > Christopher Covington wrote: >> arch/arm64/configs/defconfig | 4 ++++ >> scripts/patch-details.sh | 21 ++++++++++++++++++--- > > I don't think these two files should be combined. Oops, sorry. Fixed in v2. >> CONFIG_PINCTRL_SINGLE=y >> CONFIG_PINCTRL_MSM8916=y >> +CONFIG_PINCTRL_QDF2XXX=m >> CONFIG_PINCTRL_QCOM_SPMI_PMIC=y >> CONFIG_GPIO_SYSFS=y >> CONFIG_GPIO_DWAPB=y >> @@ -245,6 +246,7 @@ CONFIG_MMC=y >> CONFIG_MMC_BLOCK_MINORS=32 >> CONFIG_MMC_ARMMMCI=y >> CONFIG_MMC_SDHCI=y >> +CONFIG_MMC_SDHCI_ACPI=m >> CONFIG_MMC_SDHCI_PLTFM=y > > Why are these =m when all the others are =y? Due to distribution differences [1][2], I see =y built-in as the default on mobile platforms and =m modular as the default on server platforms. But thinking about it, we could get a poweroff, reset, over-temperature, or other interrupt before we get to the initramfs. So the pin controller driver should be =y built-in. One might make rootfs media =y built-in to make testing easier, but I prefer to put my rootfs on SATA, as SD/MMC is much slower and smaller by comparison. Thanks, Cov Examples I found to spot check my recollection: 1. https://github.com/CyanogenMod/android_kernel_lge_hammerhead/blob/cm-13.0/arch/arm/configs/cyanogenmod_hammerhead_defconfig 2. https://git.centos.org/blob/sig-altarch!kernel.git/c3cb6cc6edb5400321fd9d83413add4cf61d942e/SOURCES!config-arm64
Christopher Covington wrote: > Due to distribution differences [1][2], I see =y built-in as the default > on mobile platforms and =m modular as the default on server platforms. I don't think we should mix "server" defconfing entries with "mobile" defconfig entries. It's a arm64 defconfig, neither server nor mobile. The other entries are =y, so these should be =y No one uses this defconfig as-is for any platform. > But thinking about it, we could get a poweroff, reset, over-temperature, > or other interrupt before we get to the initramfs. So the pin controller > driver should be =y built-in. > > One might make rootfs media =y built-in to make testing easier, but I > prefer to put my rootfs on SATA, as SD/MMC is much slower and smaller by > comparison. Let's not overthink things, and just make *everything* =y by default.
On Fri, Jul 01, 2016 at 10:11:00AM -0500, Timur Tabi wrote: > Christopher Covington wrote: > >Due to distribution differences [1][2], I see =y built-in as the default > >on mobile platforms and =m modular as the default on server platforms. > > I don't think we should mix "server" defconfing entries with "mobile" > defconfig entries. It's a arm64 defconfig, neither server nor mobile. The > other entries are =y, so these should be =y > > No one uses this defconfig as-is for any platform. > > >But thinking about it, we could get a poweroff, reset, over-temperature, > >or other interrupt before we get to the initramfs. So the pin controller > >driver should be =y built-in. > > > >One might make rootfs media =y built-in to make testing easier, but I > >prefer to put my rootfs on SATA, as SD/MMC is much slower and smaller by > >comparison. > > Let's not overthink things, and just make *everything* =y by default. Do you guys want me to fix this up so it can hit this window? I can just change the m to y before I send it off. Andy -- To unsubscribe from this list: send the line "unsubscribe linux-arm-msm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Andy Gross wrote: > Do you guys want me to fix this up so it can hit this window? I can just change > the m to y before I send it off. Yes, please.
diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig index fd2d74d..c138425 100644 --- a/arch/arm64/configs/defconfig +++ b/arch/arm64/configs/defconfig @@ -186,6 +186,7 @@ CONFIG_SPI_SPIDEV=m CONFIG_SPMI=y CONFIG_PINCTRL_SINGLE=y CONFIG_PINCTRL_MSM8916=y +CONFIG_PINCTRL_QDF2XXX=m CONFIG_PINCTRL_QCOM_SPMI_PMIC=y CONFIG_GPIO_SYSFS=y CONFIG_GPIO_DWAPB=y @@ -245,6 +246,7 @@ CONFIG_MMC=y CONFIG_MMC_BLOCK_MINORS=32 CONFIG_MMC_ARMMMCI=y CONFIG_MMC_SDHCI=y +CONFIG_MMC_SDHCI_ACPI=m CONFIG_MMC_SDHCI_PLTFM=y CONFIG_MMC_SDHCI_OF_ESDHC=y CONFIG_MMC_SDHCI_TEGRA=y @@ -272,6 +274,8 @@ CONFIG_DMADEVICES=y CONFIG_PL330_DMA=y CONFIG_TEGRA20_APB_DMA=y CONFIG_QCOM_BAM_DMA=y +CONFIG_QCOM_HIDMA_MGMT=m +CONFIG_QCOM_HIDMA=m CONFIG_RCAR_DMAC=y CONFIG_VFIO=y CONFIG_VFIO_PCI=y diff --git a/scripts/patch-details.sh b/scripts/patch-details.sh index ec7d163..4069855 100755 --- a/scripts/patch-details.sh +++ b/scripts/patch-details.sh @@ -35,7 +35,7 @@ submaint() { N; s/THE REST,//; s/,THE REST//; s/\n/ /' } -echo "URL RPK Style Tag Version Checkpatch Errors Checkpatch Warnings Rebases Builds Last Updated Author Subject File Updates Lines Added Lines Removed Maintainer(s) Subsystem(s)" +echo "URL RPK Style Tag Version Checkpatch Errors Checkpatch Warnings Rebases Builds Boots Last Updated Author Subject File Updates Lines Added Lines Removed Maintainer(s) Subsystem(s)" while read u; do echo -n "$u " p= @@ -121,12 +121,27 @@ while read u; do if $build; then if (make defconfig && make) &> buildlog; then echo -n 'yes ' + if [ "$QEMU" ]; then + timeout 10 \ + $QEMU \ + -nodefaults \ + -nographic \ + -M virt \ + -cpu cortex-a57 \ + -m 512 \ + -serial stdio \ + -kernel arch/arm64/boot/Image \ + -append "debug earlycon=pl011,0x9000000 rdinit=/bin/busybox poweroff -f" \ + -initrd ../fs/busybox64.cpio.gz + else + echo -n ' ' + fi else git reset --hard HEAD~1 &> /dev/null - echo -n 'no ' + echo -n 'no ' fi else - echo -n ' ' + echo -n ' ' fi awk -f scripts/parsepatch.awk /tmp/patch.mbox || diffstat -s /tmp/patch.mbox | awk -f scripts/parsepatch.awk submaint /tmp/patch.mbox
Now that ACPI is on in the defconfig, the time seems right to enable drivers for the SD/MMC, DMA, and pin control hardware described in the ACPI tables of the QDF2432 server platform. Signed-off-by: Christopher Covington <cov@codeaurora.org> --- arch/arm64/configs/defconfig | 4 ++++ scripts/patch-details.sh | 21 ++++++++++++++++++--- 2 files changed, 22 insertions(+), 3 deletions(-)