Sitara Linux Upstream Status

OBSOLETE
Contents of this page are obsolete. It is no longer maintained.

Sitara Linux Upstream Status

Vaibhav Hiremath

Status Dashboard

Color Coding

Color	Status or Meaning
	TBD or Not planned yet OR Work has not started Yet
	Under Work OR 0 or more patch version submitted and cont effort is required
	Patches either acked or accepted under sub-maintainer tree/list
	Fully Upstreamed and 100% done

General Issues Status

General Upstream Status table

Module/Feature/Driver	Sub-feature or Patch	Submission Version	Current Status	Comments
U-Boot	SPL		In Progress	After posting patches for additional functionality in SPL, Wolfgang Denk (WD) has pushed back on adding more features until issues relating to the ease of other developers causing hard to track down bugs in SPL due to the resource constraints SPL has compared with full U-Boot. His request is a board-specific README (in boards/...) that defines SPL memory map (code, data, bss, stack, malloc area) and tools to verify this information. Some combination of gcc -fstack-usage or checkstack.pl and GNU cflow should allow for this to happen but the plumbing between these points does not exist yet. GNU global may also help here.
	Poor ext2load load times			ext2load has consistently and noticeably worse read times than fatload. Community has noticed and is taking multiple paths to fix (direct problem in ext2 code, ext4 code which doesn't have this problem).
	NAND and lib/bch.c support		Paused (trini)	With v2012.03 we have resynced with the kernel NAND subsystem and have lib/bch.c which we should show usage of for BCH4/8/16. For platforms that have no ROM requirement for BCH4+ we are able to utilize the library. We are not yet able to generate values that match what the ROM expects. HW Apps has been asked as of March 16th 2012.
	High-Speed MMC	N/A	Known Issues	In the community there are reports of problems with certain cards not working. Within the community, the omap3_pandora maintainer has been trying to reconcile the kernel driver and U-Boot drivers.
	multi-bus i2c support	N/A	Missing	The community i2c driver has no multi-bus support, and only i2c0 is visible to U-Boot on all boards.
	Host USB (EHCI)	N/A	Code needs cleanup	In addition to needing changes to be cache coherent, the ehci driver needs further work to generalize ehci_hcd_{init,stop} and the driver itself needs further updates to better utilize ULPI like the kernel driver does rather than GPIO banging.
	Common config files, more useful options	RFC v0		CONFIG_CMDLINE_EDITING, omap3_common.h ala omap4/5_common.h, ti_common.h, examples for redundant layout. The first step here is to make an omap3_common.h and consolidate all of the options. A v0 RFC has been posted to get feedback on the concepts used.
Linux Kernel	EMAC clocks for AM35x do not match driver requirements	Patch queued for 3.4. http://comments.gmane.org/gmane.linux.ports.arm.omap/68784	Tested on AM35x-EVM and Craneboard	N/A
	OMAP2+ does not provide proper regulators to smsc911x driver	Patch queued for 3.3 fixes in arm-soc tree. http://comments.gmane.org/gmane.linux.kernel/1257169	Tested on AM37x-EVM	N/A

AM335x U-Boot Status

AM335x U-Boot Upstream Status table

Module/Feature/Driver	Submission Version	Current Status	Comments
SPL	3rd	Upstreamed	Completed.
CPSW		Upstreamed	Completed.
NAND		Upstreamed	Completed.
USB (Host)	TBD		Hopeful that the now merged code for USB on OMAP3/4/5 will make enabling here easy.
CPU speed scaling	TBD		Possibly to be omitted now that Linux framework exists
EMIF conversion	TBD		Switch to common EMIF code introduced by OMAP4/5

AM335x Linux Status

AM335x Linux Kernel MainlineStatus table

Module/Feature/Driver	SubModule	Targeted Kernel Release	Owner	Current Status	Comments
Kernel Baseport	32KSYNC Timer dependency/Cleanup	v3.5	Vaibhav Hiremath	Upstreamed	Merged into mainline in the kernel version v3.5 (after 5 version)
	Basic AM335x DTS support	v3.4	Vaibhav Hiremath	Upstreamed	Queued for v3.6 (After 3 version)
	Control register offset addition to control.h	v3.6	Vaibhav Hiremath	Upstreamed	Completed.
	PRM, CM and clockdomain API interface	v3.6	Vaibhav Hiremath	Upstreamed	Completed.
	VoltageDomain, PowerDomain and ClockDomain data	v3.6	Vaibhav Hiremath	Upstreamed	Completed.
	hwmod API interface in omap_hwmod	v3.7	Vaibhav Hiremath	Upstreamed	Completed.
	Clock Tree (Common clock FW)	v3.6	Vaibhav Hiremath	Upstreamed	Completed.
	hwmod data	v3.7	Vaibhav Hiremath	Upstreamed	Completed.
	Support for Complete number of interrupts (128)	v3.12	Vaibhav Hiremath	Not yet Started	Submitted 1st version of patches, but KevinH wants existing code cleaned up before accepting any patch for irq. This will be taken up once Baseport support is complete.
	Enabled Misc modules in HWMOD	v3.10	Vaibhav Hiremath	Under Work	Modules like debugss needs to be handled and enabled in HWMOD.
	DT Update for Pad configuration of modules (default state)	v3.10	Vaibhav Hiremath	Under Work	This may not have functionality impact as of now, but required in order to get complete DT support.
GPMC	Existing Code Cleanup	v3.7	Mohammed Afzal	Upstreamed	Completed
	Complete Driver Migration	v3.7	Mohammed Afzal	Upstreamed	Completed.
	GPMC/MTD/NAND DT Support	v3.10	[1]	Accepted	Under Review - http://comments.gmane.org/gmane.linux.ports.arm.omap/90898
	AM335x Specific support	v3.10	Mohammed Afzal	Submitted	Under Review - https://lkml.org/lkml/2013/1/18/32
Basic Pin Mux	AM335x Pin Control driver (pin control single)	v3.7	AnilKumar Chimata	Upstreamed	Completed.
EDMA	EDMA driver migration to DMA Engine FW	v3.10	Matt Porter	v9 series posted	Under Work - https://lkml.org/lkml/2013/3/6/362
	Migrate private EDMA implementation to arm/common/	v3.10	Matt Porter	v9 series posted	Under Work - https://lkml.org/lkml/2013/3/6/362
	DT support for EDMA	v3.10	Matt Porter	v9 series posted	Under Review - https://lkml.org/lkml/2013/3/6/362
	EDMA DMA engine driver cyclic support	v3.8	Matt Porter	Not Started	Audio peripherals drivers that depend on periodic transfers accomplish this using the DMA Engine API's cyclic transfer type. This support is required for conversion of the McASP driver to DMA engine API.
MMCSD: AM335x specific changes (along with DT)		v3.10	Gururaja Hebbar	Under work	Dependency on EDMA support, handled as part of DMA Engine migration series
eHRPWM	eHRPWM driver	v3.7	Avinash Philip	Upstreamed	Completed.
	DT Support for eHRPWM driver	v3.10	Avinash Philip	Accpeted	Under review - https://lkml.org/lkml/2012/11/27/120
APWM	APWM driver	v3.7	Avinash Philip	Upstreamed	Completed.
	DT Support for APWM driver	v3.10	Avinash Philip	Submitted	Under Review - https://lkml.org/lkml/2012/11/27/120
TouchScreen	Basic Driver	v3.4	Rachna Patil	Upstreamed	Completed
	RunTime PM Support	v3.7	Rachna Patil	Upstreamed	Completed
	TouchScreen: DT Support	v3.7	Rachna Patil	Submitted	Under Review - https://lkml.org/lkml/2013/1/23/722
	ADC: TouchScreen Driver MFD Conversion	v3.7	Rachna Patil	Upstreamed	Completed
	ADC: IIO Framework - One-shot mode Support	v3.7	Rachna Patil	Upstreamed	Completed
	ADC: IIO Framework - Continuous mode support	v3.10	Rachna Patil	Under Work	Available with v3.2 kernel, need to forward port with bug fixes
CPSW	Ethernet: Basic Driver	v3.4	Mugunthan N	Upstreamed	Completed
	Ethernet: RunTime PM Support	v3.6	Mugunthan N	Upstreamed	completed
	Ethernet: DT Support	v3.7	Mugunthan N	Upstreamed	Completed
	Ethernet: Dual MAC Support	v3.7	Mugunthan N	Upstreamed	Completed - Merged in for v3.9
	Ethernet: VLAN Support	v3.8	Mugunthan N	Upstreamed	Completed.
	Ethernet: ALE Support	v3.11	Mugunthan N	Not Yet Started	Will have to start with RFC
USB	USB: Basic Glue Layer	v3.5	Ravi	Upstreamed	Completed
	USB: PIO Mode Support on musb0 port (with DT)	v3.5	Ajay Gupta	Upstreamed	Completed
	USB: Dual MUSB	v3.10	Ajay Gupta	Under Work	Patches submitted and is under review
	USB: PHY control driver	v3.10	Ravi	Not Yet Started	PHY driver is recently Merged for OMAP4 and same need to be done for am335x.
	USB: DT Support (dual instance)	v3.10	Ravi	Under work	Patches are submitted and is under review
	USB: CPPI 4.1 DMA Engine Support	v3.12	Visuvanadan Pasupathy	Under Work	RFC has already submitted to the community, we had healthy discussion and alignment. Implementation to be started yet. This may take its own time, so we will have custom patches available for TI releases.
SPI	SPI: DMA Engine API support	v3.6	[2]	Upstreamed	Completed (by Russell)
	SPI: AM335x DT Support (DMA mode)	v3.10	[3]	Under Work	Handled as part of DMA engine migration series.
I2C: AM335x DT Support		v3.6	Vaibhav Hiremath	Upstreamed	Completed
Basic GPIO: DT Support		v3.6	AnilKumar Chimata	Upstreamed	Completed
DCAN	DCAN: Basic Driver Support	v3.6	AnilKumar Chimata	Upstreamed	Completed
	DCAN: RunTime PM Support	v3.7	AnilKumar Chimata	Upstreamd	Completed
	DCAN: DT Support	v3.7	AnilKumar Chimata	Upstreamed	Completed.
Audio	Audio: DMA Engine API support	v3.10	Matt Porter	Not Yet Started	Dependency on EDMA DMA Engine driver cyclic transfer support
	Audio: RunTime PM Support	v3.7	Gururaja Hebbar	Upstreamed	Completed
	Audio: DT Support	v3.10	Gururaja Hebbar	Under Work.	Dependency on EDMA support in DMA Engine FW
	Audio: McASP Multimode (master/slave) support	v3.11	[mailto:]	Not started	Dependency on EDMA DMA Engine driver cyclic transfer support
WDT: DT and AM335x specific support		v3.6	Afzal Mohammed	Upstreamed
LCDC	LCD: MODEDB support (Dynamic Change of Panel Configuration)	v3.10	Rachana Patil	Submitted	under Review - https://lkml.org/lkml/2012/12/7/163
	LCD: RunTime PM Support	v3.7	Rachana Patil	Upstreamed	Completed
	LCD: DT Support	v3.10	Prakash Manjunathappa	Submitted	Under Review - https://lkml.org/lkml/2013/1/7/10
PM	PM: Suspend To RAM Support	v3.10	Vaibhav Bedia	Under Work	Under Review - http://www.spinics.net/lists/linux-omap/msg84343.html
	PM: CPUIDLE Support	v3.12	[mailto:]	Not Yet Started
	PM: CPUFREQ Support	v3.7	AnilKumar, Chimata	Upstreamed	Completed - Though Silicon related OPP support needs to be handled.
	PM: SmartReplex Support	v3.12	Greg	Under Work	Available with v3.2 kernel release
	PM: Suspend To Disk	v3.11	[mailto:]	Submitted	under Review https://patchwork.kernel.org/patch/1743731/
	PM: RTC+ Mode Support	v3.12	Russ dill	Not Yet Started
RTC	RunTime PM Support	v3.7	Mohammed Afzal	Upstreamed	Completed
	DT Support	v3.7	Mohammed Afzal	Upstreamed	Completed
Crypto Driver		v3.11	[mailto:]	Submitted	Under Review - http://www.spinics.net/lists/linux-omap/msg84189.html

AM35x Status

AM35x Upstream Status table (08:47, 7 February 2012 (CST) for kernel)

Module/Feature/Driver	Sub-feature or Patch	Submission Version	Current Status	Revision of tree at status	Comments
U-Boot	SPL			v2011.12+7cb30b1	NAND and SD/MMC only.
	Basic			v2011.12+7cb30b1	NAND and SD/MMC only for boot. UART for stdin/out
	CPU Scaling				Exists in AMSDK here
	Ethernet		Basic functionality works. Ethernet MAC detection patch needs rework to address comments. Ethernet driver (davinci_enet) needs to be converted to use PHYLIB.	d0b049d	v1 and comments are here, here, and here
	NOR Flash		NOR flash support exposed issues with gpmc_init() which need to be addressed first.	d0b049d	v1 and comments are here and here
	EMIF		Need to convert to using the EMIF support OMAP4/5 added.	d0b049d
	Improved config file		Easier customization, more robust, etc	d0b049d	v1 partially is here
	DDR	N/A	Code needs cleanup/refactoring	v2011.12+7cb30b1	mtest 0x80000000 0x8FF14000 0xdeadbeef 1000 passes
	USB Host	N/A	Non-functional	v2011.12+7cb30b1	Issuing usb start doesn't find any devices when USB storage device is inserted for example. Needs updating to current framework. See general note.
	USB Device	N/A	Non-functional	v2011.12+7cb30b1	Building with CONFIG_MUSB_UDC set does not result in the device being seen by the host with lsusb.
Kernel AM35x-EVM and omap2plus_defconfig	omap_hwmod: usbtll_fck: missing clockdomain for usbtll_fck. omap_hwmod: am35x_otg_hs: cannot be enabled (3) omap_hwmod: uart4: softreset failed (waited 10000 usec)			891003a
	_init_omap_device: could not find omap_hwmod for iva			891003a	Need to verify state of patches that were discussed related to this warning
	Address Hole seen by USB_OTG at address 46384544			891003a
	dpll3_m2_clk rate change failed: -22			891003a
	MMC/SD WP gpio mux Multiple paths error		Deferred until DT/pinctrl/pinmux conversion complete as current hsmmc init code has pinmuxing assumptions that are not valid on AM35x where multiple pins can provide the same GPIO signal. See old discussion at comments.gmane.org/gmane.linux.ports.arm.omap/52891	891003a
	Root from initrd ramdisk		This appears to be caused by the SDRC sleep code assuming it's on a true omap3 derivative. On AM35x we have EMIF registers and the existing sleep code is banging on SDRC/SMS registers which do not exist. Mark Greer is working on the root cause and fix to handle the different SDRC/EMIF integration on AM35x.	891003a	Hangs before reaching root mount when root=/dev/ram
	Root from NFS		Partially fails due to the SDRC issue above. Also requires the fix from General Issues for EMAC clocks and the .config must have EMAC support enabled.	891003a	Hangs before reaching root mount when root=/dev/nfs
	Root from MMC		Fails due to the SDRC issue above.	891003a	Hangs waiting for MMC card detection when root=/dev/mmc*

AM37x Status

AM37x Upstream Status table (08:48, 7 February 2012 (CST))

Module/Feature/Driver	Sub-feature or Patch	Submission Version	Current Status	Revision of tree at status	Comments
U-Boot	SPL	N/A		v2011.12+7cb30b1	MMC and NAND boot tested.
	DDR	N/A	Code needs cleanup/refactoring	v2011.12+7cb30b1	mtest 0x80000000 0x8FF14000 0xdeadbeef 1000 passes
	i2c0	N/A		v2011.12+7cb30b1	i2c probe ; i2c mw 4b 0 aa 1; i2c md 4b 0 ; i2c mw 4b 0 0 1; i2c md 4b 0 changes and displays values.
	Networking	N/A		v2011.12+7cb30b1	dhcp configured ethernet interface, detects MAC, is able to load a kernel image without packet loss.
	USB Host	N/A	Non-functional	v2011.12+7cb30b1	Issuing usb start doesn't find any devices when USB storage device is inserted for example. Needs updating to current framework. See general note.
	USB Device	N/A	Non-functional	v2011.12+7cb30b1	Building with CONFIG_MUSB_UDC set does not result in the device being seen by the host with lsusb.
Kernel AM37x-EVM and omap2plus_defconfig	omap_hwmod: usbtll_fck: missing clockdomain for usbtll_fck			23783f8
	_omap_mux_get_by_name: Could not find signal gpmc_wait2.uart4_tx omap_hwmod_mux_init: Could not allocate device mux entry			23783f8
	dpll3_m2_clk rate change failed: -22			23783f8
	smsc911x smsc911x.0: Failed to get supply 'vdd33a': -19 smsc911x smsc911x.0: (unregistered net_device): couldn't get regulators -19		Will retest when the v3 patch mentioned in General Issues hits mainline	23783f8
	twl4030_wdt twl4030_wdt: Failed to register misc device twl4030_wdt: probe of twl4030_wdt failed with error -16			23783f8	Possibly related? http://www.mail-archive.com/linux-omap@vger.kernel.org/msg47531.html
	Root from initrd ramdisk		Working	23783f8	Successfully mounts root=/dev/ram
	Root from NFS		Will retest when the v3 patch mentioned in General Issues hits mainline	23783f8	Hangs before reaching root mount when root=/dev/nfs. Problem is no network device as a result of smsc911x regulators not being registered
	Root from MMC			23783f8	Hangs waiting for MMC card detection when root=/dev/mmc*. After initial investigation of hanging on am35x, this appears to be a separate bug
	Extremely unresponsive userspace in interactive use		Need to retest interactive use now that serial PM fixes came into 3.3	23783f8	After filesystem root, system is painfully slow in interactive use. Symptom appears the same as http://www.spinics.net/lists/linux-omap/msg62760.html Verified that this is not specific to serial, occurs via shell use over network connection. Hacking PM off as mentioned in this post works around the issue so it appears to be a general PM issue. Needs investigation.

BeagleBoard xM Status

BeagleBoarx xM Upstream Status table (10:00, 24 February 2012 (MST))

Module/Feature/Driver	Sub-feature or Patch	Submission Version	Current Status	Revision of tree at status	Comments
U-Boot	SPL	N/A		v2011.12+6751b05	MMC boot tested.
	Networking	N/A	Known Issues	v2011.12	Without a network switch, no packets hit the wire (crossover or normal cable), with switch fine.
	DDR	N/A	Code needs cleanup/refactoring	v2011.12+g7cb30b1	mtest 0x80000000 0x9FF00000 0xdeadbeef 0x400 completes without error
	DVI-D	N/A		v2011.12+6751b05	DVI initialized. U-Boot splash screen support requires an image loaded in memory. Generic work needed to extend this support (possibly glue another ELF section into u-boot.img).
	i2c0	N/A		v2011.12+6751b05	i2c probe ; i2c mw 4b 0 aa 1; i2c md 4b 0 ; i2c mw 4b 0 0 1; i2c md 4b 0 changes and displays values.
	USB Host	N/A	Needs to be made cache coherent. General USB cache coherencey is being addressed in the community.	v2011.12+6751b05	While in previous versions dcache off needed to be issued prior to starting the USB stack, currently the ehci-omap driver will disable dcache automatically. Currently fatload/fatls work and with usbethaddr set prior to usb start or usb reset network commands work.
	USB Gadget	N/A	console only	v2011.12+7cb30b1	Issuing setenv stdout usbtty;setenv stdin usbtty switches to the ACM device. However, output acts like it's not being flushed properly.
Linux Kernel	Placeholder	?	Placeholder	N/A
	Placeholder	?	Placeholder	N/A