Hi,
For reference, here is (I believe) the current list of Tegra patches required become submitting new patches. I note a few conflicts already, so this will take a little sorting out once the tegra tree is live.
31b9b64 tegra2: Move board_mmc_init into board files fbbb238 tegra2: Modify MMC driver to handle power and cd GPIOs 187643e tegra2: Add support for Ventana 7f27b57 tegra2: Use new GPIO APIs in gpio_config_uart() 6b61458 tegra2: Add arch_cpu_init() to fire up Cortex-A9 eda293e tegra2: Simplify tegra_start() boot path 34e9d46 arm: Move CP15 init out of cpu_init_crit() 3e75dfa tegra2: Enable instruction cache f9c505b tegra2: Remove unneeded boot code 788bfea tegra2: Remove unneeded config option ad525dc tegra2: Remove unused low-level Tegra2 UART code 29843f0 tegra2: Remove unneeded 'dynamic ram size' message 7c6c3ae tegra2: Don't use board pointer before it is set up 5cc0dad tegra2: Tidy UART selection 845f1c6 tegra2: Add UARTB support 33977b1 tegra2: spi: Add SPI driver for Tegra2 SOC 61d1db1 tegra2: config: Enable SPI flash on Seaboard 57e2005 tegra2: Enable SPI environment on Seaboard 707ae71 tegra2: Implement SPI / UART GPIO switch 024baf8 tegra2: spi: Support SPI / UART switch 930fe9f tegra2: Plumb in SPI/UART switch code 40b415f tegra2: Always build with USE_PRIVATE_LIBGCC=yes. 0ec937a tegra2: Change CONFIG_SYS_TEXT_BASE to 0x00108000. 2936c6b tegra2: Move tegra2_mmc_init() prototype to public header. d75bb9f tegra2: Add common Avionic Design Tamonten support. 6205647 tegra2: Add Avionic Design Plutux support. 0b526ff (HEAD, denx/us-tegra, us-tegra) tegra2: Add Avionic Design Medcom support
Regards, Simon
On 11/21/2011 04:07 PM, Simon Glass wrote:
Hi,
For reference, here is (I believe) the current list of Tegra patches required become submitting new patches. I note a few conflicts
s/required become/required in your local branch before your changes/ ?
already, so this will take a little sorting out once the tegra tree is live.
I assume that means we can assume these are all accepted and will be merged, so we should apply them locally and build on top of them?
Is there a status update for the creation of a Tegra git that we could fetch and branch from ?
Note that already, I think some of the later patches need rework given earlier ones, e.g. there was some change to board/nvidia/common/Makefile that required other boards using it to be updated, but the Ventana Makefile wasn't IIRC. I guess that's what you meant by conflicts.
Thanks for the list!
31b9b64 tegra2: Move board_mmc_init into board files fbbb238 tegra2: Modify MMC driver to handle power and cd GPIOs 187643e tegra2: Add support for Ventana 7f27b57 tegra2: Use new GPIO APIs in gpio_config_uart() 6b61458 tegra2: Add arch_cpu_init() to fire up Cortex-A9 eda293e tegra2: Simplify tegra_start() boot path 34e9d46 arm: Move CP15 init out of cpu_init_crit() 3e75dfa tegra2: Enable instruction cache f9c505b tegra2: Remove unneeded boot code 788bfea tegra2: Remove unneeded config option ad525dc tegra2: Remove unused low-level Tegra2 UART code 29843f0 tegra2: Remove unneeded 'dynamic ram size' message 7c6c3ae tegra2: Don't use board pointer before it is set up 5cc0dad tegra2: Tidy UART selection 845f1c6 tegra2: Add UARTB support 33977b1 tegra2: spi: Add SPI driver for Tegra2 SOC 61d1db1 tegra2: config: Enable SPI flash on Seaboard 57e2005 tegra2: Enable SPI environment on Seaboard 707ae71 tegra2: Implement SPI / UART GPIO switch 024baf8 tegra2: spi: Support SPI / UART switch 930fe9f tegra2: Plumb in SPI/UART switch code 40b415f tegra2: Always build with USE_PRIVATE_LIBGCC=yes. 0ec937a tegra2: Change CONFIG_SYS_TEXT_BASE to 0x00108000. 2936c6b tegra2: Move tegra2_mmc_init() prototype to public header. d75bb9f tegra2: Add common Avionic Design Tamonten support. 6205647 tegra2: Add Avionic Design Plutux support. 0b526ff (HEAD, denx/us-tegra, us-tegra) tegra2: Add Avionic Design Medcom support
Hi Stephen,
On Mon, Nov 21, 2011 at 3:30 PM, Stephen Warren swarren@nvidia.com wrote:
On 11/21/2011 04:07 PM, Simon Glass wrote:
Hi,
For reference, here is (I believe) the current list of Tegra patches required become submitting new patches. I note a few conflicts
s/required become/required in your local branch before your changes/ ?
s/become/before/
already, so this will take a little sorting out once the tegra tree is live.
I assume that means we can assume these are all accepted and will be merged, so we should apply them locally and build on top of them?
Is there a status update for the creation of a Tegra git that we could fetch and branch from ?
Well I could put up a tree, but it's a bit dangerous - the real place is patchwork.
Note that already, I think some of the later patches need rework given earlier ones, e.g. there was some change to board/nvidia/common/Makefile that required other boards using it to be updated, but the Ventana Makefile wasn't IIRC. I guess that's what you meant by conflicts.
Yes that's right. Hopefully the pain will reduce soon!
Thanks for the list!
You are welcome.
Regards, Simon
31b9b64 tegra2: Move board_mmc_init into board files fbbb238 tegra2: Modify MMC driver to handle power and cd GPIOs 187643e tegra2: Add support for Ventana 7f27b57 tegra2: Use new GPIO APIs in gpio_config_uart() 6b61458 tegra2: Add arch_cpu_init() to fire up Cortex-A9 eda293e tegra2: Simplify tegra_start() boot path 34e9d46 arm: Move CP15 init out of cpu_init_crit() 3e75dfa tegra2: Enable instruction cache f9c505b tegra2: Remove unneeded boot code 788bfea tegra2: Remove unneeded config option ad525dc tegra2: Remove unused low-level Tegra2 UART code 29843f0 tegra2: Remove unneeded 'dynamic ram size' message 7c6c3ae tegra2: Don't use board pointer before it is set up 5cc0dad tegra2: Tidy UART selection 845f1c6 tegra2: Add UARTB support 33977b1 tegra2: spi: Add SPI driver for Tegra2 SOC 61d1db1 tegra2: config: Enable SPI flash on Seaboard 57e2005 tegra2: Enable SPI environment on Seaboard 707ae71 tegra2: Implement SPI / UART GPIO switch 024baf8 tegra2: spi: Support SPI / UART switch 930fe9f tegra2: Plumb in SPI/UART switch code 40b415f tegra2: Always build with USE_PRIVATE_LIBGCC=yes. 0ec937a tegra2: Change CONFIG_SYS_TEXT_BASE to 0x00108000. 2936c6b tegra2: Move tegra2_mmc_init() prototype to public header. d75bb9f tegra2: Add common Avionic Design Tamonten support. 6205647 tegra2: Add Avionic Design Plutux support. 0b526ff (HEAD, denx/us-tegra, us-tegra) tegra2: Add Avionic Design Medcom support
-- nvpublic
On 11/21/2011 05:05 PM, Simon Glass wrote:
On Mon, Nov 21, 2011 at 3:30 PM, Stephen Warren swarren@nvidia.com wrote:
...
Is there a status update for the creation of a Tegra git that we could fetch and branch from ?
Well I could put up a tree, but it's a bit dangerous - the real place is patchwork.
I thought there was going to be a u-boot-tegra.git that a Tegra U-boot maintainer would put the patches into once reviewed/approved, and then send pull requests to Albert, like the other ARM SoCs? That's what I was expecting to rebase on top of. Sure some patches will still be waiting in patchwork, but the volume would presumably be a lot lower, and there'd be fewer conflicts.
Hi Stephen,
On Mon, Nov 21, 2011 at 4:12 PM, Stephen Warren swarren@nvidia.com wrote:
On 11/21/2011 05:05 PM, Simon Glass wrote:
On Mon, Nov 21, 2011 at 3:30 PM, Stephen Warren swarren@nvidia.com wrote:
...
Is there a status update for the creation of a Tegra git that we could fetch and branch from ?
Well I could put up a tree, but it's a bit dangerous - the real place is patchwork.
I thought there was going to be a u-boot-tegra.git that a Tegra U-boot maintainer would put the patches into once reviewed/approved, and then send pull requests to Albert, like the other ARM SoCs? That's what I was expecting to rebase on top of. Sure some patches will still be waiting in patchwork, but the volume would presumably be a lot lower, and there'd be fewer conflicts.
Yes and I think Tom volunteered to run that! Once it is up and going we should be able to find everything we need there.
But until then we need to rely on patchwork and keeping our own tree locally. My copy of these patches is at http://git.denx.de/u-boot-simonglass.git (branch us-tegra) but it includes all my 'patman' junk, and resolution of some patch conflicts, etc. So really only useful for comparing. I am using it to base further patches on, of which there are many as you know.
BTW my SPI series didn't really get a lot of comments - are you happy with the UART switch stuff? It is an incomplete implementation since the NS16550 buffering patch was NAKed. Mike suggested another option we could do to get around calling pinmux code in NS16650 and Tegra SPI, but in any case it might be dead in the water without serial buffering. I have put it down as something to look at 'later'.
Regards Simon
-- nvpublic
Dear Stephen Warren,
In message 4ECAE8DB.5070509@nvidia.com you wrote:
I thought there was going to be a u-boot-tegra.git that a Tegra U-boot maintainer would put the patches into once reviewed/approved, and then send pull requests to Albert, like the other ARM SoCs? That's what I was
Yes, there will be. I just didn't get to that yet.
expecting to rebase on top of. Sure some patches will still be waiting in patchwork, but the volume would presumably be a lot lower, and there'd be fewer conflicts.
Sorry.
Best regards,
Wolfgang Denk
participants (3)
-
Simon Glass -
Stephen Warren -
Wolfgang Denk