zador.blood.stained
-
Posts
3633 -
Joined
-
Last visited
Content Type
Forums
Store
Crowdfunding
Applications
Events
Raffles
Community Map
Posts posted by zador.blood.stained
-
-
-
I tested it on OPi Zero (basically the same device) with u-boot master branch and it passed the loopback test.
defconfig changes:
diff --git a/configs/orangepi_zero_defconfig b/configs/orangepi_zero_defconfig index e5a4c1d9fc..39528026a6 100644 --- a/configs/orangepi_zero_defconfig +++ b/configs/orangepi_zero_defconfig @@ -9,10 +9,19 @@ CONFIG_DRAM_ODT_EN=y CONFIG_SPL_SPI_SUNXI=y CONFIG_NR_DRAM_BANKS=1 # CONFIG_SYS_MALLOC_CLEAR_ON_INIT is not set +# CONFIG_USE_BOOTCOMMAND is not set CONFIG_CONSOLE_MUX=y # CONFIG_CMD_FLASH is not set +CONFIG_CMD_SF=y +CONFIG_CMD_SF_TEST=y +CONFIG_CMD_SPI=y CONFIG_DEFAULT_DEVICE_TREE="sun8i-h2-plus-orangepi-zero" +CONFIG_MTD=y +CONFIG_DM_SPI_FLASH=y CONFIG_SUN8I_EMAC=y -CONFIG_USB_OHCI_HCD=y +CONFIG_SPI=y +CONFIG_DM_SPI=y +CONFIG_SOFT_SPI=y CONFIG_USB_EHCI_HCD=y +CONFIG_USB_OHCI_HCD=y CONFIG_SYS_USB_EVENT_POLL_VIA_INT_QUEUE=y
DT changes:
diff --git a/arch/arm/dts/sun8i-h2-plus-orangepi-zero.dts b/arch/arm/dts/sun8i-h2-plus-orangepi-zero.dts index 0bc031fe4c..40b078723d 100644 --- a/arch/arm/dts/sun8i-h2-plus-orangepi-zero.dts +++ b/arch/arm/dts/sun8i-h2-plus-orangepi-zero.dts @@ -59,6 +59,8 @@ /* ethernet0 is the H3 emac, defined in sun8i-h3.dtsi */ ethernet0 = &emac; ethernet1 = &xr819; + spi0 = &softspi0; + spi1 = &softspi1; }; chosen { @@ -89,6 +91,26 @@ gpio = <&pio 0 20 GPIO_ACTIVE_HIGH>; }; + softspi0: soft-spi0 { + compatible = "spi-gpio"; + status = "okay"; + gpio-sck = <&pio 2 2 0>; + gpio-mosi = <&pio 2 0 0>; + gpio-miso = <&pio 2 1 0>; + cs-gpios = <&pio 2 3 0>; + num-chipselects = <1>; + }; + + softspi1: soft-spi1 { + compatible = "spi-gpio"; + status = "okay"; + gpio-sck = <&pio 0 14 0>; + gpio-mosi = <&pio 0 15 0>; + gpio-miso = <&pio 0 16 0>; + cs-gpios = <&pio 0 13 0>; + num-chipselects = <1>; + }; + wifi_pwrseq: wifi_pwrseq { compatible = "mmc-pwrseq-simple"; reset-gpios = <&r_pio 0 7 GPIO_ACTIVE_LOW>;Partial console log:
=> echo loopback wire disconnected loopback wire disconnected => sspi 1:0 16 1234 0000 => echo loopback wire connected loopback wire connected => sspi 1:0 16 1234 1234 => dm tree Class Index Probed Driver Name ----------------------------------------------------------- root 0 [ + ] root_driver root_driver simple_bus 0 [ + ] generic_simple_bus |-- soc mmc 0 [ + ] sunxi_mmc | |-- mmc@1c0f000 blk 0 [ ] mmc_blk | | `-- mmc@1c0f000.blk mmc 1 [ + ] sunxi_mmc | |-- mmc@1c10000 blk 1 [ ] mmc_blk | | `-- mmc@1c10000.blk phy 0 [ + ] sun4i_usb_phy | |-- phy@1c19400 usb 0 [ + ] ehci_generic | |-- usb@1c1a000 usb_hub 0 [ + ] usb_hub | | `-- usb_hub usb 1 [ + ] ohci_generic | |-- usb@1c1a400 usb_hub 1 [ + ] usb_hub | | `-- usb_hub usb 2 [ + ] ehci_generic | |-- usb@1c1b000 usb_hub 2 [ + ] usb_hub | | `-- usb_hub usb 3 [ + ] ohci_generic | |-- usb@1c1b400 usb_hub 3 [ + ] usb_hub | | `-- usb_hub clk 0 [ + ] sun8i_h3_ccu | |-- clock@1c20000 reset 0 [ + ] sunxi_reset | | `-- reset gpio 0 [ + ] gpio_sunxi | |-- pinctrl@1c20800 gpio 1 [ + ] gpio_sunxi | | |-- PA gpio 2 [ + ] gpio_sunxi | | |-- PB gpio 3 [ + ] gpio_sunxi | | |-- PC gpio 4 [ + ] gpio_sunxi | | |-- PD gpio 5 [ + ] gpio_sunxi | | |-- PE gpio 6 [ + ] gpio_sunxi | | |-- PF gpio 7 [ + ] gpio_sunxi | | |-- PG gpio 8 [ + ] gpio_sunxi | | |-- PH gpio 9 [ + ] gpio_sunxi | | `-- PI eth 0 [ + ] eth_sun8i_emac | |-- ethernet@1c30000 serial 0 [ + ] ns16550_serial | |-- serial@1c28000 gpio 10 [ + ] gpio_sunxi | `-- pinctrl@1f02c00 gpio 11 [ + ] gpio_sunxi | `-- PL spi 0 [ + ] soft_spi |-- soft-spi0 spi_generi 0 [ + ] spi_generic_drv | `-- generic_0:0 spi 1 [ + ] soft_spi |-- soft-spi1 spi_generi 1 [ + ] spi_generic_drv | `-- generic_1:0 clk 1 [ + ] fixed_rate_clock |-- osc24M_clk clk 2 [ + ] fixed_rate_clock |-- osc32k_clk clk 3 [ ] fixed_rate_clock `-- internal-osc-clk =>
-
There is no SPI driver for H3 in the "main" u-boot (CONFIG_SUN4I_SPI is for older SoCs), but software SPI (CONFIG_SOFT_SPI) should work if configured correctly.
-
On 1/29/2019 at 2:34 AM, TonyMac32 said:
They expect the user to know too much, and get grouchy when they don't.
I think you meant "expect the user to read and not just mindlessly tick checkboxes that prevent them from posting their question"?
-
16 minutes ago, martinayotte said:
Probably one image is older than the other, and in the older one there is no "sun4i_hdmi_i2c" defined in DT ...
It's not defined in DT and is created dynamically by the DRM subsystem.
-
There is an old upstream mainline kernel issue that never was fixed: https://groups.google.com/forum/#!topic/linux-sunxi/Ar9A_OYzk1Q
And we can't pull random patches without proper long term testing because they may break more things than they fix.
-
10 hours ago, Igor said:
This means adding armbianmonitor -u mandatory to all subforums in Technical support and adding a new one for issues related when you can't supply it? That sounds even better.
First I would change and reword the current implementation - use something like "I understand that not providing requested information will reduce the chance to solve my issue" (current one with the possibility of getting banned doesn't correlate with the forum rules), explain why providing armbianmonitor -u info is needed (I checked a few new new threads and they don't have it), deal with old images that try to upload the info to sprunge.us (i.e. by linking an instructions for updating the script without updating the BSP), ideally deal with the possibility that ix.io may stop to provide its free service one day too, etc.
2 hours ago, rooted said:Running a site such as Armbian isn't (necessarily) expensive so why is so much money needed? I mean $100/month USD can run a site with larger volume.
I'll just leave the public download statistics link here. I'm not an expert on hosting prices in EU, but you need to take into account both the storage space and monthly traffic.
Edit: clarification - this is daily stats that don't include torrent traffic.
-
Quote
fatal: dumb http transport does not support shallow capabilities
This is not a networking issue but a repository and protocol limitation, as I remember git.denx.de supports shallow clones only via the git:// protocol
-
Don't get me wrong, I'm not a fan of the current form implementation, but it's only the first iteration. Since "Technical support" section is mostly used as a bug tracker and we wouldn't be able to support a standalone bug tracker / task management system, adding data collection form to the forum is a step in the right direction.
-
6 hours ago, NicoD said:
You know I love this forum, it's the main spot to get my information about SBC's. And I love Armbian, nothing in the forum can change that.
Unfortunately "getting information" != "answering the same questions and requesting even the basic info again and again and again".
For example, threads like this one would have 1 reply with a simple answer - eMMC CSD checks for revision 8 were fixed more than a year ago. It took 5 posts to confirm that this is the culprit and it is still impossible to tell which kernel version (I mean kernel compilation date) is used due to missing dmesg from "armbianmonitor -u".
Funny thing is that this thread was created after the addition of the new information collection form, so we may as well make providing armbianmonitor -u output mandatory with 2 exceptions in dedicated subforums - devices without network connection and devices that failed to boot completely.
6 hours ago, NicoD said:But when I first came to the forum it was a lot more lively, and more fun. That's not the purpose of a forum.
Applying your knowledge and experience while at the same time learning something new is fun, and it can happen when you are talking with a person who at least tries to understand what he is doing and who respects your time by trying to provide requested (and even more than requested) information.
Reading 90% of threads in the "Allwinner H2/H3" section made by people who expect performance and software quality of at least a $200 PC like a J5005 based NUC7 from a $20 board like Orange Pi PC is definitely not fun for me. YMMV.
Unfortunately supporting low end (by price) boards attracts their target audience aka people for whom even the Raspberry Pi is too expensive.
-
23 minutes ago, Igor said:57 minutes ago, Werner said:
What about simply redirecting "Armbian build framework" to https://github.com/armbian/build as all stuff regarding the building process, script and so on is discussed (or should be at least) discussed there?
Perhaps. @zador.blood.stained?Not sure about this.
- the second tab on Github project is called "Issues", not "Discussions" or "all stuff", and it has "Open"/"Closed" statuses more suitable for tracking actual bugs/issues than discussing build script usage
- this would make it impossible to easily move framework related discussions from the forum to the right place
-
Looks like we got an upstream DT for the T4: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/arch/arm64/boot/dts/rockchip?id=e7a095908227fb3ccc86d001d9e13c9ae2bef8e6
-
Just now, g40 said:
Thanks. Have you had any 64bit boards working with NFS?
No.
Root on NFS should not be hard by itself (similar to H3, loading u-boot and kernel from SD).
FEL support, on the other hand, is not implemented in the mainline u-boot for 64-bit chips, so it requires either using prebuilt u-boot binaries or adding patches to u-boot and rewriting the build script.
-
There is no FEL support for 64-bit targets in the build script.
-
According to the SoC HW specifications MPP pins are not 5V tolerant.
-
Last time I tried it I didn't get it past the low FPS issue described in this thread. The only documentation I can point is this repository which contains some instructions that may or may not work with the kernel currently available in Armbian.
-
1 hour ago, Nazar Gerasymchuk said:
My main confusion came from situation that there are lots of H3 boards that works quite well with mainline kernel, and there is no option to use GPU. from customer side it looks very strange.
Who is the customer of who here?
AFAIK most Chinese makers sell these boards not much above the BOM cost, and even Allwinner tries to cut their cost as much as possible, i.e. by not purchasing Mali driver licenses that are useless for them with Android: https://irclog.whitequark.org/linux-sunxi/2017-12-07#20730568; (timestamp 07:39)
1 hour ago, Nazar Gerasymchuk said:Doesn't it make usable Mali driver?
It only helps running the kernel-side driver. It still requires a userspace closed-source driver of a specific type (framebuffer, Xorg, Wayland) and applications that can use this driver, i.e. you need a special shim (xf86-video-armsoc) to make use of Mali with X.org.
-
26 minutes ago, lanefu said:
Quick suggestion --- also add donate link in plain site at top of forum....
And also Terms of Use / Community guidelines should be linked somewhere in plain sight. Right now it can be accessed when registering, and even there it is not clearly shown as a link. IMO it should be on the top menu row as it is more important than i.e. Privacy policy that exists mostly for legal reasons.
Edit: It's also on the bottom bar, but it will be hidden after you click "Accept", looks like by a specific cookie.
Edit 2: Now that I'm looking at those (Registration Terms specifically) - they should be a numbered list instead of a bullet list, and we should probably add a new rule "Try to stay on topic defined by the starting post or developer posts in the thread. Off-topic posts that derail the discussion may be hidden, moved or deleted."
-
6 minutes ago, martinayotte said:
You don't need to build everything, you only need to have DTC compiler, decompile the file /boot/dtb/sun8i-h3-orangepi-lite.dtb from SDCard into DTS, edit it to remove "cd-gpio" and add "cd-broken", and recompile it into DTB.
Changing DT will have no effect this early (in SPL), SPL relies on CONFIG_MMC0_CD_PIN configuration option.
And at least previous versions printed "mmc: no card present" when trying to access a card with broken CD pin.
-
2 hours ago, cabled said:
This means I clone the git and cross compile the kernel, then make that adjustment and retrieve the resultant uImage?
As I wrote in the issue, you need zImage and matching uInitrd with a custom premount script from the repository. There are no guides on how exactly to create them.
The only thing that makes it easier is that mentioned eMMC CSD revision patch is already included in the Armbian patchset, but this assumes that CSD revision 8 is the problem which is impossible to confirm without kernel logs.
-
This does not depend on u-boot (and it doesn't have much checks for eMMC revisions anyway).
Most likely this could be solved by patching the CSD check: https://github.com/zador-blood-stained/fel-mass-storage/issues/5
-
3 hours ago, TonyMac32 said:
Well, when the boot script says to apply it, u-boot spits out the "help text" for the command, reverts to original DTS, and boots without the adjustment.
At least the vanilla mainline u-boot "fdt apply" requires OF_LIBFDT_OVERLAY (which there is set on the per board basis, but it's easier to "select" or "imply" it from ARCH_ROCKCHIP like here: https://github.com/armbian/build/blob/master/patch/u-boot/u-boot-sunxi/enable-DT-overlays-support.patch
-
Or you would need to change the package names and installation paths for those libraries so they could coexist with upstream ones, and build mpv and plex against those.
Or use static linking to libav* and libsw* libraries.
-
On 12/7/2018 at 8:19 PM, jerryn said:
Also how would I add plexmedia player to the Armbian build ? I have it working.. just need to add hardware acceleration to ffmpeg and mpv before I package it up and get it into the Armbian build.
As long as you can make it work with ffmpeg versions that match the upstream distribution versions (Xenial, Stretch, Bionic), it could be done.
We tried building newer ffmpeg and mpv in extras-buildpackages and providing those packages (you can probably recover the files from git history), but it resulted in breaking more than improving (or you would need to rebuild every upstream Debian/Ubuntu package that depends on libav* and libsw* packages, and building ffmpeg by itself in a generic automated way creates a dependency nightmare already).
Testers wanted: sunxi Device Tree overlays
in Advanced users - Development
Posted
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/sound/soc/sunxi/sun4i-i2s.c