Various omap device tree usability fixes for v3.13 merge window
Subsystems (15)
Tony Lindgren
maintainercover Various omap device tree usability fixes for v3.13 merge window
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Hi all,
Here are some patches to fix various issue for device tree based booting for omaps that I've ran into recently while working on removing legacy platform data for mach-omap2.
Regards,
Tony
Tony Lindgren (8):
net: smc91x: Fix device tree based configuration so it's usable
mmc: omap: Fix DMA configuration to not rely on device id
mmc: omap: Fix I2C dependency and make driver usable with device tree
i2c: omap: Fix missing device tree flags for omap2
gpio: twl4030: Fix regression for twl gpio output
gpio: twl4030: Fix passing of pdata in the device tree case
ARM: OMAP2+: Fix GPMC and simplify bootloader timings for 8250 and
smc91x
ARM: dts: Fix omap2 specific dtsi files by adding the missing entries
.../devicetree/bindings/net/smsc-lan91c111.txt | 4 + arch/arm/boot/dts/omap-zoom-common.dtsi | 2 +- arch/arm/boot/dts/omap2.dtsi | 96 ++++++++++++++++++++++ arch/arm/boot/dts/omap2420.dtsi | 23 ++++++ arch/arm/boot/dts/omap2430.dtsi | 49 +++++++++++ arch/arm/mach-omap2/gpmc.c | 58 +++++-------- drivers/gpio/gpio-twl4030.c | 13 ++- drivers/i2c/busses/i2c-omap.c | 22 +++++ drivers/mmc/host/omap.c | 42 +++++----- drivers/net/ethernet/smsc/smc91x.c | 52 +++++++++--- 10 files changed, 285 insertions(+), 76 deletions(-)
Signature · 1 line
1.8.1.1
Tony Lindgren
maintainerPATCH 1/8 net: smc91x: Fix device tree based configuration so it's usable
Commit 89ce376c6bdc (drivers/net: Use of_match_ptr() macro in smc91x.c) added minimal device tree support to smc91x, but it's not working on many platforms because of the lack of some key configuration bits.
Fix the issue by parsing the necessary configuration like the smc911x driver is doing.
---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
--- .../devicetree/bindings/net/smsc-lan91c111.txt | 4 ++ drivers/net/ethernet/smsc/smc91x.c | 52 +++++++++++++++++----- 2 files changed, 46 insertions(+), 10 deletions(-)
Documentation/devicetree/bindings/net/smsc-lan91c111.txt +4 -0
@@ -8,3 +8,7 @@ Required properties: Optional properties: - phy-device : phandle to Ethernet phy - local-mac-address : Ethernet mac address to use +- reg-io-width : Specify the size (in bytes) of the IO accesses that + should be performed on the device. Valid value for SMSC LAN is + 1, 2 or 4. If it's omitted or invalid, the size would be 2. +- smsc,nowait : Setup for fast register access with no waits
drivers/net/ethernet/smsc/smc91x.c +42 -10
86 lines not on this page · raw message
Signature · 1 line
1.8.1.1
Re: [PATCH 1/8] net: smc91x: Fix device tree based configuration so it's usable
Quoted 28 lines · Tony Lindgren
Commit
89ce376c6bdc(drivers/net: Useof_match_ptr()macro insmc91x.c) added minimal device tree support to smc91x, but it's not working on many platforms because of the lack of some key configuration bits.Fix the issue by parsing the necessary configuration like the smc911x driver is doing.
Cc Nicolas Pitre <[email protected]>Cc "David S. Miller" <[email protected]>Cc netdev at vger.kernel.orgCc devicetree at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
--- .../devicetree/bindings/net/smsc-lan91c111.txt | 4 ++ drivers/net/ethernet/smsc/smc91x.c | 52 +++++++++++++++++----- 2 files changed, 46 insertions(+), 10 deletions(-)Documentation/devicetree/bindings/net/smsc-lan91c111.txt +0
-0
Documentation/devicetree/bindings/net/smsc-lan91c111.txt Required properties: · line 13
Quoted 6 lines
Documentation/devicetree/bindings/net/smsc-lan91c111.txt +3
-0@@ -8,3 +8,7 @@ Required properties: Optional properties: - phy-device : phandle to Ethernet phy - local-mac-address : Ethernet mac address to use +- reg-io-width : Specify the size (in bytes) of the IO accesses that + should be performed on the device. Valid value for SMSC LAN is + 1, 2 or 4. If it's omitted or invalid, the size would be 2.
In the driver the supported access sizes are not mutually exclusive. It would be nice for the binding to have the same property.
Documentation/devicetree/bindings/net/smsc-lan91c111.txt Required properties: · line 14
Quoted 1 line
Documentation/devicetree/bindings/net/smsc-lan91c111.txt +1
-0+- smsc,nowait : Setup for fast register access with no waits
I'm confused by what this means. When would this be selected, and when wouldn't it be?
Thanks,
Mark.
Tony Lindgren
maintainerRe: [PATCH 1/8] net: smc91x: Fix device tree based configuration so it's usable
Quoted 24 lines · Mark Rutland
Quoted 23 lines · Tony Lindgren
Commit
89ce376c6bdc(drivers/net: Useof_match_ptr()macro insmc91x.c) added minimal device tree support to smc91x, but it's not working on many platforms because of the lack of some key configuration bits.Fix the issue by parsing the necessary configuration like the smc911x driver is doing.
Cc Nicolas Pitre <[email protected]>Cc "David S. Miller" <[email protected]>Cc netdev at vger.kernel.orgCc devicetree at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
--- .../devicetree/bindings/net/smsc-lan91c111.txt | 4 ++ drivers/net/ethernet/smsc/smc91x.c | 52 +++++++++++++++++----- 2 files changed, 46 insertions(+), 10 deletions(-)
Quoted 14 lines
Documentation/devicetree/bindings/net/smsc-lan91c111.txt Required properties: · line 13
Quoted 11 lines
Documentation/devicetree/bindings/net/smsc-lan91c111.txt +3
-0@@ -8,3 +8,7 @@ Required properties: Optional properties: - phy-device : phandle to Ethernet phy - local-mac-address : Ethernet mac address to use +- reg-io-width : Specify the size (in bytes) of the IO accesses that + should be performed on the device. Valid value for SMSC LAN is + 1, 2 or 4. If it's omitted or invalid, the size would be 2.In the driver the supported access sizes are not mutually exclusive. It would be nice for the binding to have the same property.
Hmm indeed. How about we add reg-io-width-mask:
1 = 8-bit access 2 = 16-bit access 4 = 32-bit access ...
So for a driver to support 8, 16 and 32-bit access the mask would be:
reg-io-width-mask = <7>;
Although the values for reg-io-width would support masks too, it might be better to have reg-io-width-mask to avoid confusion.
Or do you have any better ideas?
Quoted 4 lines
Documentation/devicetree/bindings/net/smsc-lan91c111.txt Required properties: · line 14
Quoted 1 line
Documentation/devicetree/bindings/net/smsc-lan91c111.txt +1
-0+- smsc,nowait : Setup for fast register access with no waitsI'm confused by what this means. When would this be selected, and when wouldn't it be?
The driver has a module parameter for it and the comments say:
"nowait = 0 for normal wait states, 1 eliminates additional wait states"
Most platforms seem to set it, but the default is to not set it. I guess we could that be a module parameter for now as that's a timing optimization.
Regards,
Tony
Tony Lindgren
maintainerRe: [PATCH 1/8] net: smc91x: Fix device tree based configuration so it's usable
Quoted 26 lines · Tony Lindgren
Quoted 3 lines · Mark Rutland
In the driver the supported access sizes are not mutually exclusive. It would be nice for the binding to have the same property.
Hmm indeed. How about we add reg-io-width-mask:
1 = 8-bit access 2 = 16-bit access 4 = 32-bit access ...So for a driver to support 8, 16 and 32-bit access the mask would be:
reg-io-width-mask = <7>;Although the values for reg-io-width would support masks too, it might be better to have reg-io-width-mask to avoid confusion.
Or do you have any better ideas?
Quoted 4 lines
Quoted 1 line
+- smsc,nowait : Setup for fast register access with no waits
I'm confused by what this means. When would this be selected, and when wouldn't it be?
Quoted 7 lines
The driver has a module parameter for it and the comments say:
"nowait = 0 for normal wait states, 1 eliminates additional wait states"
Most platforms seem to set it, but the default is to not set it. I guess we could that be a module parameter for now as that's a timing optimization.
Here's what I was thinking with the reg-io-width-mask. Anybody have comments on using reg-io-width vs reg-io-width-mask?
Regards,
Tony
From: Tony Lindgren <[email protected]>
Date: Wed, 13 Nov 2013 16:36:37 -0800
Subject: [PATCH] net: smc91x: Fix device tree based configuration so it's usable
Commit 89ce376c6bdc (drivers/net: Use of_match_ptr() macro in smc91x.c) added minimal device tree support to smc91x, but it's not working on many platforms because of the lack of some key configuration bits.
Fix the issue by parsing the necessary configuration like the smc911x driver is doing.
Documentation/devicetree/bindings/net/smsc-lan91c111.txt +4 -0
@@ -8,3 +8,7 @@ Required properties: Optional properties: - phy-device : phandle to Ethernet phy - local-mac-address : Ethernet mac address to use +- reg-io-width-mask : Mask of sizes (in bytes) of the IO accesses that + are supported on the device. Valid value for SMSC LAN91c111 are + 1, 2 or 4. If it's omitted or invalid, the size would be 2 meaning + 16-bit access only.
drivers/net/ethernet/smsc/smc91x.c +33 -10
77 lines not on this page · raw message
Tony Lindgren
maintainerRe: [PATCH 1/8] net: smc91x: Fix device tree based configuration so it's usable
Quoted 3 lines · Tony Lindgren
Here's what I was thinking with the reg-io-width-mask. Anybody have comments on using reg-io-width vs reg-io-width-mask?
...
Quoted 25 lines
drivers/net/ethernet/smsc/smc91x.c +14
-1@@ -2222,11 +2234,31 @@ static int smc_drv_probe(struct platform_device *pdev)*/ lp =netdev_priv (ndev); + lp->cfg.flags =0 ;if (pd) {memcpy (&lp->cfg, pd,sizeof (lp->cfg)); lp->io_shift =SMC91X_IO_SHIFT (lp->cfg.flags); - }else { + } + +#if IS_BUILTIN (CONFIG_OF ) + match =of_match_device (of_match_ptr (smc91x_match), &pdev->dev); +if (match) { +u32 val; + +of_property_read_u32 (np,"reg-io-width" , &val); +if (val ==0 ) + lp->cfg.flags |=SMC91X_USE_16BIT ; +if (val &1 ) + lp->cfg.flags |=SMC91X_USE_8BIT ; +if (val &2 )
drivers/net/ethernet/smsc/smc91x.c static int smc_drv_probe(struct platform_device *pdev) · line 2264
Quoted 10 lines
drivers/net/ethernet/smsc/smc91x.c +7
-0+ lp->cfg.flags |=SMC91X_USE_16BIT ; +if (val &4 ) + lp->cfg.flags |= SMC91X_USE_32BIT; + } +#endif + +if (!pd && !match) { lp->cfg.flags |= (SMC_CAN_USE_8BIT ) ?SMC91X_USE_8BIT :0 ; lp->cfg.flags |= (SMC_CAN_USE_16BIT ) ?SMC91X_USE_16BIT :0 ; lp->cfg.flags |= (SMC_CAN_USE_32BIT ) ?SMC91X_USE_32BIT :0 ;
Looks this patch is missing the check for the return value for of_property_read_u32(), will repost this patch separately as the others in this series are out of the way now.
Regards,
Tony
Tony Lindgren
maintainerPATCH 2/8 mmc: omap: Fix DMA configuration to not rely on device id
In reply to Tony LindgrenWe are wrongly relying on device id for the DMA configuration which can lead to wrong DMA channel being selected.
Fix the issue by using the standard resources like we should.
---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
---
drivers/mmc/host/omap.c | 32 +++++++++++--------------------- 1 file changed, 11 insertions(+), 21 deletions(-)
drivers/mmc/host/omap.c +11 -21
@@ -90,17 +90,6 @@#define OMAP_MMC_CMDTYPE_AC 2#define OMAP_MMC_CMDTYPE_ADTC 3 -#define OMAP_DMA_MMC_TX 21 -#define OMAP_DMA_MMC_RX 22 -#define OMAP_DMA_MMC2_TX 54 -#define OMAP_DMA_MMC2_RX 55 - -#define OMAP24XX_DMA_MMC2_TX 47 -#define OMAP24XX_DMA_MMC2_RX 48 -#define OMAP24XX_DMA_MMC1_TX 61 -#define OMAP24XX_DMA_MMC1_RX 62 - -#define DRIVER_NAME "mmci-omap"/* Specifies how often in millisecs to poll for card status changes
@@ -1408,19 +1397,20 @@ static int mmc_omap_probe(struct platform_device *pdev) host->dma_tx_burst =-1 ; host->dma_rx_burst =-1 ; -if (mmc_omap2 ()) - sig = host->id ==0 ?OMAP24XX_DMA_MMC1_TX :OMAP24XX_DMA_MMC2_TX ; -else - sig = host->id ==0 ?OMAP_DMA_MMC_TX :OMAP_DMA_MMC2_TX ; - host->dma_tx =dma_request_channel (mask, omap_dma_filter_fn, &sig); + res =platform_get_resource_byname (pdev,IORESOURCE_DMA ,"tx" ); +if (res) + sig = res->start; + host->dma_tx =dma_request_slave_channel_compat (mask, + omap_dma_filter_fn, &sig, &pdev->dev,"tx" );if (!host->dma_tx)dev_warn (host->dev,"unable to obtain TX DMA engine channel %u\n" , sig); -if (mmc_omap2 ()) - sig = host->id ==0 ?OMAP24XX_DMA_MMC1_RX :OMAP24XX_DMA_MMC2_RX ; -else - sig = host->id ==0 ?OMAP_DMA_MMC_RX :OMAP_DMA_MMC2_RX ; - host->dma_rx =dma_request_channel (mask, omap_dma_filter_fn, &sig); + + res =platform_get_resource_byname (pdev,IORESOURCE_DMA ,"rx" ); +if (res) + sig = res->start; + host->dma_rx =dma_request_slave_channel_compat (mask, + omap_dma_filter_fn, &sig, &pdev->dev,"rx" );if (!host->dma_rx)dev_warn (host->dev,"unable to obtain RX DMA engine channel %u\n" , sig);
Signature · 1 line
1.8.1.1
Tony Lindgren
maintainerPATCH 2/8 mmc: omap: Fix DMA configuration to not rely on device id
Quoted 5 lines · Tony Lindgren
We are wrongly relying on device id for the DMA configuration which can lead to wrong DMA channel being selected.
Fix the issue by using the standard resources like we should.
Quoted 8 lines
Cc Chris Ball <[email protected]>Cc linux-mmc at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
Looks like we need sig initialized in this patch to avoid some uninitialized warnings.
Regards,
Tony
From: Tony Lindgren <[email protected]>
Date: Thu, 14 Nov 2013 15:25:06 -0800
Subject: [PATCH] mmc: omap: Fix DMA configuration to not rely on device id
We are wrongly relying on device id for the DMA configuration which can lead to wrong DMA channel being selected.
Fix the issue by using the standard resources like we should.
drivers/mmc/host/omap.c +12 -22
@@ -90,17 +90,6 @@#define OMAP_MMC_CMDTYPE_AC 2#define OMAP_MMC_CMDTYPE_ADTC 3 -#define OMAP_DMA_MMC_TX 21 -#define OMAP_DMA_MMC_RX 22 -#define OMAP_DMA_MMC2_TX 54 -#define OMAP_DMA_MMC2_RX 55 - -#define OMAP24XX_DMA_MMC2_TX 47 -#define OMAP24XX_DMA_MMC2_RX 48 -#define OMAP24XX_DMA_MMC1_TX 61 -#define OMAP24XX_DMA_MMC1_RX 62 - -#define DRIVER_NAME "mmci-omap"/* Specifies how often in millisecs to poll for card status changes
@@ -1331,7 +1320,7 @@ static int mmc_omap_probe(struct platform_device *pdev)struct mmc_omap_host *host =NULL ;struct resource *res;dma_cap_mask_t mask; -unsigned sig; +unsigned sig =0 ;int i, ret =0 ;int irq;
@@ -1408,19 +1397,20 @@ static int mmc_omap_probe(struct platform_device *pdev) host->dma_tx_burst =-1 ; host->dma_rx_burst =-1 ; -if (mmc_omap2 ()) - sig = host->id ==0 ?OMAP24XX_DMA_MMC1_TX :OMAP24XX_DMA_MMC2_TX ; -else - sig = host->id ==0 ?OMAP_DMA_MMC_TX :OMAP_DMA_MMC2_TX ; - host->dma_tx =dma_request_channel (mask, omap_dma_filter_fn, &sig); + res =platform_get_resource_byname (pdev,IORESOURCE_DMA ,"tx" ); +if (res) + sig = res->start; + host->dma_tx =dma_request_slave_channel_compat (mask, + omap_dma_filter_fn, &sig, &pdev->dev,"tx" );if (!host->dma_tx)dev_warn (host->dev,"unable to obtain TX DMA engine channel %u\n" , sig); -if (mmc_omap2 ()) - sig = host->id ==0 ?OMAP24XX_DMA_MMC1_RX :OMAP24XX_DMA_MMC2_RX ; -else - sig = host->id ==0 ?OMAP_DMA_MMC_RX :OMAP_DMA_MMC2_RX ; - host->dma_rx =dma_request_channel (mask, omap_dma_filter_fn, &sig); + + res =platform_get_resource_byname (pdev,IORESOURCE_DMA ,"rx" ); +if (res) + sig = res->start; + host->dma_rx =dma_request_slave_channel_compat (mask, + omap_dma_filter_fn, &sig, &pdev->dev,"rx" );if (!host->dma_rx)dev_warn (host->dev,"unable to obtain RX DMA engine channel %u\n" , sig);
PATCH 2/8 mmc: omap: Fix DMA configuration to not rely on device id
Hi Tony,
Quoted 8 lines · Tony Lindgren
We are wrongly relying on device id for the DMA configuration which can lead to wrong DMA channel being selected.
Fix the issue by using the standard resources like we should.
Cc Chris Ball <[email protected]>Cc linux-mmc at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>
Feel free to merge via your tree:
- Chris.
Signature · 1 line
Chris Ball <[email protected]> <http://printf.net/>
Tony Lindgren
maintainerPATCH 2/8 mmc: omap: Fix DMA configuration to not rely on device id
Quoted 8 lines · Chris Ball
Hi Tony,
Quoted 5 lines · Tony Lindgren
We are wrongly relying on device id for the DMA configuration which can lead to wrong DMA channel being selected.
Fix the issue by using the standard resources like we should.
Quoted 7 lines
Quoted 3 lines
Cc Chris Ball <[email protected]>Cc linux-mmc at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>Feel free to merge via your tree:
Acked-by Chris Ball <[email protected]>
OK thanks will merge both mmc fixes.
Tony
PATCH 2/8 mmc: omap: Fix DMA configuration to not rely on device id
In reply to Chris BallHi Chris
Quoted 8 lines · On Tue, 26 Nov 2013 18:33:50 -0500 Chris Ball <[email protected]> wrote:
Hi Tony,
Quoted 5 lines · Tony Lindgren
We are wrongly relying on device id for the DMA configuration which can lead to wrong DMA channel being selected.
Fix the issue by using the standard resources like we should.
Quoted 8 lines
Quoted 3 lines
Cc Chris Ball <[email protected]>Signed-off-by Tony Lindgren <[email protected]>Feel free to merge via your tree:
Acked-by Chris Ball <[email protected]>
Can you enlighten me what's the proper way to get patches to mmc since I cannot figure out working methodology from MAINTAINERS file?
I happened to notice that Tony had this similar patch in linux-omap list I sent first time already September to you and linux-mmc (sorry linux-omap folks, I didn't want to spam multiple lists):
http://www.spinics.net/lists/linux-mmc/msg22137.html
After that I've resend the set a few times including a fix to user triggable NULL pointer dereference:
http://www.spinics.net/lists/linux-mmc/msg22610.html
I'm fine if mmc patches should go through other subsystems but at least it should be documented in MAINTAINERS file.
Signature · 1 line
Jarkko
Tony Lindgren
maintainerPATCH 2/8 mmc: omap: Fix DMA configuration to not rely on device id
Quoted 30 lines · Jarkko Nikula
Hi Chris
Quoted 8 lines · On Tue, 26 Nov 2013 18:33:50 -0500 Chris Ball <[email protected]> wrote:
Hi Tony,
Quoted 5 lines · Tony Lindgren
We are wrongly relying on device id for the DMA configuration which can lead to wrong DMA channel being selected.
Fix the issue by using the standard resources like we should.
Quoted 8 lines
Quoted 3 lines
Cc Chris Ball <[email protected]>Cc linux-mmc at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>Feel free to merge via your tree:
Acked-by Chris Ball <[email protected]>Can you enlighten me what's the proper way to get patches to mmc since I cannot figure out working methodology from
MAINTAINERSfile?I happened to notice that Tony had this similar patch in linux-omap list I sent first time already September to you and linux-mmc (sorry linux-omap folks, I didn't want to spam multiple lists):
Quoted 7 lines
After that I've resend the set a few times including a fix to user triggable NULL pointer dereference:
http://www.spinics.net/lists/linux-mmc/msg22610.html
I'm fine if mmc patches should go through other subsystems but at least it should be documented in
MAINTAINERSfile.
Bummer, sounds like some duplicate work could have been avoided :( I suggest resend to Chris and linux-mmc one more time as Chris should pick up the MMC patches in general.
I've been just trying to get things working in general for v3.13-rc series for omaps with device tree based booting and patching all over the place. This is to make it easy for us to just drop the legacy platform based booting for v3.14 while keeping things working.
Chris, as far as I'm concerned, Aaro and Jarkko are good people to review and ack the drivers/mmc/omap.c patches, so adding Aaro to Cc as well.
Regards,
Tony
PATCH 2/8 mmc: omap: Fix DMA configuration to not rely on device id
Hi
Quoted 5 lines · Tony Lindgren
* Jarkko Nikula <[email protected]> [131127 12:55]:
Bummer, sounds like some duplicate work could have been avoided :( I suggest resend to Chris and linux-mmc one more time as Chris should pick up the MMC patches in general.
Before that should Chris merge your 2 patches in order to avoid couple trivial merge conflicts with my set when linux-omap and mmc are merged?
Signature · 1 line
Jarkko
Tony Lindgren
maintainerPATCH 2/8 mmc: omap: Fix DMA configuration to not rely on device id
Quoted 10 lines · Jarkko Nikula
Hi
Quoted 5 lines · Tony Lindgren
* Jarkko Nikula <[email protected]> [131127 12:55]:
Bummer, sounds like some duplicate work could have been avoided :( I suggest resend to Chris and linux-mmc one more time as Chris should pick up the MMC patches in general.Before that should Chris merge your 2 patches in order to avoid couple trivial merge conflicts with my set when linux-omap and mmc are merged?
Those are already queued to arm-soc/fixes branch and should hit the mainline tree within next few days, hopefully before -rc2 so it's probably to wait for that for that for further fixes.
Regards,
Tony
PATCH 2/8 mmc: omap: Fix DMA configuration to not rely on device id
In reply to Jarkko NikulaHi,
Quoted 2 lines · Jarkko Nikula
Can you enlighten me what's the proper way to get patches to mmc since I cannot figure out working methodology from
MAINTAINERSfile?
Sorry about this, Jarkko.
Since I don't have omap.c hardware, I'm generally going to wait for a Tested-by/Acked-by from someone else on patches like these. We actually have a maintainer already listed for the driver, Jarkko Lavinen <[email protected]>.
So, maybe the MAINTAINERS section for drivers/mmc/host/omap.c needs an update? Feel free to send me patches to it. Either way, I'll start treating Aaro and Jarkko (Nikula) as maintainers for this driver.
Thanks!
- Chris.
Signature · 1 line
Chris Ball <[email protected]> <http://printf.net/>
Tony Lindgren
maintainerPATCH 2/8 mmc: omap: Fix DMA configuration to not rely on device id
Quoted 13 lines · Chris Ball
Hi,
Quoted 2 lines · Jarkko Nikula
Can you enlighten me what's the proper way to get patches to mmc since I cannot figure out working methodology from
MAINTAINERSfile?Sorry about this, Jarkko.
Since I don't have
omap.chardware, I'm generally going to wait for a Tested-by/Acked-by from someone else on patches like these. We actually have a maintainer already listed for the driver, Jarkko Lavinen <[email protected]>.
Quoted 4 lines
So, maybe the
MAINTAINERSsection fordrivers/mmc/host/omap.cneeds an update? Feel free to send me patches to it. Either way, I'll start treating Aaro and Jarkko (Nikula) as maintainers for this driver.
Hmm looks like here's a reasonably recent update that we've missed from Jarkko Lavinen to update his email address:
http://lkml.org/lkml/2013/10/4/142
I've updated Jarkko Lavinen's address in this mail, we should probably also patch the MAINTAINERS file for that, and then see if Aaro and Jarkko Nikula are interested to be listed there as well.
Regards,
Tony
PATCH 2/8 mmc: omap: Fix DMA configuration to not rely on device id
Hi
Quoted 14 lines · Tony Lindgren
Quoted 13 lines · Chris Ball
Hi,
Quoted 2 lines · Jarkko Nikula
Can you enlighten me what's the proper way to get patches to mmc since I cannot figure out working methodology from
MAINTAINERSfile?Sorry about this, Jarkko.
Since I don't have
omap.chardware, I'm generally going to wait for a Tested-by/Acked-by from someone else on patches like these. We actually have a maintainer already listed for the driver, Jarkko Lavinen <[email protected]>.
Quoted 5 lines
Quoted 4 lines
So, maybe the
MAINTAINERSsection fordrivers/mmc/host/omap.cneeds an update? Feel free to send me patches to it. Either way, I'll start treating Aaro and Jarkko (Nikula) as maintainers for this driver.
Oh, I see. I even didn't check ./scripts/get_maintainer.pl nor check MAINTAINERS file further than main MMC entry as I knew Jarkko L. is not working anymore for Nokia. Sorry, we should have noticed that.
Quoted 9 lines
Hmm looks like here's a reasonably recent update that we've missed from Jarkko Lavinen to update his email address:
http://lkml.org/lkml/2013/10/4/142
I've updated Jarkko Lavinen's address in this mail, we should probably also patch the
MAINTAINERSfile for that, and then see if Aaro and Jarkko Nikula are interested to be listed there as well.
I can volunteer. Not that I know much about mmc but I like to keep my legacy hw still running :-)
Signature · 1 line
Jarkko
Tony Lindgren
maintainerPATCH 2/8 mmc: omap: Fix DMA configuration to not rely on device id
Quoted 29 lines · Jarkko Nikula
Hi
Quoted 14 lines · Tony Lindgren
Quoted 13 lines · Chris Ball
Hi,
Quoted 2 lines · Jarkko Nikula
Can you enlighten me what's the proper way to get patches to mmc since I cannot figure out working methodology from
MAINTAINERSfile?Sorry about this, Jarkko.
Since I don't have
omap.chardware, I'm generally going to wait for a Tested-by/Acked-by from someone else on patches like these. We actually have a maintainer already listed for the driver, Jarkko Lavinen <[email protected]>.Quoted 5 lines
Quoted 4 lines
So, maybe the
MAINTAINERSsection fordrivers/mmc/host/omap.cneeds an update? Feel free to send me patches to it. Either way, I'll start treating Aaro and Jarkko (Nikula) as maintainers for this driver.Oh, I see. I even didn't check
./scripts/get_maintainer.plnor checkMAINTAINERSfile further than main MMC entry as I knew Jarkko L. is not working anymore for Nokia. Sorry, we should have noticed that.Quoted 3 lines
Hmm looks like here's a reasonably recent update that we've missed from Jarkko Lavinen to update his email address:
Quoted 8 lines
Quoted 6 lines
http://lkml.org/lkml/2013/10/4/142
I've updated Jarkko Lavinen's address in this mail, we should probably also patch the
MAINTAINERSfile for that, and then see if Aaro and Jarkko Nikula are interested to be listed there as well.I can volunteer. Not that I know much about mmc but I like to keep my legacy hw still running :-)
Sounds good to me, good to hear :)
Regards,
Tony
PATCH 2/8 mmc: omap: Fix DMA configuration to not rely on device id
In reply to Tony LindgrenQuoted 45 lines · Tony Lindgren
We are wrongly relying on device id for the DMA configuration which can lead to wrong DMA channel being selected.
Fix the issue by using the standard resources like we should.
Cc Chris Ball <[email protected]>Cc linux-mmc at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
---
drivers/mmc/host/omap.c | 32 +++++++++++--------------------- 1 file changed, 11 insertions(+), 21 deletions(-)drivers/mmc/host/omap.c +0
-11@@ -90,17 +90,6 @@#define OMAP_MMC_CMDTYPE_AC 2#define OMAP_MMC_CMDTYPE_ADTC 3 -#define OMAP_DMA_MMC_TX 21 -#define OMAP_DMA_MMC_RX 22 -#define OMAP_DMA_MMC2_TX 54 -#define OMAP_DMA_MMC2_RX 55 - -#define OMAP24XX_DMA_MMC2_TX 47 -#define OMAP24XX_DMA_MMC2_RX 48 -#define OMAP24XX_DMA_MMC1_TX 61 -#define OMAP24XX_DMA_MMC1_RX 62 - -#define DRIVER_NAME "mmci-omap"/* Specifies how often in millisecs to poll for card status changes @@ -1408,19 +1397,20 @@ static int mmc_omap_probe(struct platform_device *pdev) host->dma_tx_burst =-1 ; host->dma_rx_burst =-1 ;
drivers/mmc/host/omap.c static int mmc_omap_probe(struct platform_device *pdev) · line 1404
Quoted 10 lines
drivers/mmc/host/omap.c +5
-5-if (mmc_omap2 ()) - sig = host->id ==0 ?OMAP24XX_DMA_MMC1_TX :OMAP24XX_DMA_MMC2_TX ; -else - sig = host->id ==0 ?OMAP_DMA_MMC_TX :OMAP_DMA_MMC2_TX ; - host->dma_tx =dma_request_channel (mask, omap_dma_filter_fn, &sig); + res =platform_get_resource_byname (pdev,IORESOURCE_DMA ,"tx" ); +if (res) + sig = res->start; + host->dma_tx =dma_request_slave_channel_compat (mask, + omap_dma_filter_fn, &sig, &pdev->dev,"tx" );
Minor comment, since we're moving to DT-only for platforms using this driver (hope I'm right about that), why not just do:
dma_request_slave_channel_(&pdev->dev, "tx");
IORESOURCE_DMA is not created by OF layer so I guess no need to call platform_get_resource either.
thanks,
-Joel
Tony Lindgren
maintainerPATCH 2/8 mmc: omap: Fix DMA configuration to not rely on device id
Quoted 38 lines · Joel Fernandes
Quoted 24 lines · Tony Lindgren
drivers/mmc/host/omap.c +0
-11@@ -90,17 +90,6 @@#define OMAP_MMC_CMDTYPE_AC 2#define OMAP_MMC_CMDTYPE_ADTC 3 -#define OMAP_DMA_MMC_TX 21 -#define OMAP_DMA_MMC_RX 22 -#define OMAP_DMA_MMC2_TX 54 -#define OMAP_DMA_MMC2_RX 55 - -#define OMAP24XX_DMA_MMC2_TX 47 -#define OMAP24XX_DMA_MMC2_RX 48 -#define OMAP24XX_DMA_MMC1_TX 61 -#define OMAP24XX_DMA_MMC1_RX 62 - -#define DRIVER_NAME "mmci-omap"/* Specifies how often in millisecs to poll for card status changes @@ -1408,19 +1397,20 @@ static int mmc_omap_probe(struct platform_device *pdev) host->dma_tx_burst =-1 ; host->dma_rx_burst =-1 ;drivers/mmc/host/omap.c static int mmc_omap_probe(struct platform_device *pdev) · line 1404
Quoted 10 lines
drivers/mmc/host/omap.c +5
-5-if (mmc_omap2 ()) - sig = host->id ==0 ?OMAP24XX_DMA_MMC1_TX :OMAP24XX_DMA_MMC2_TX ; -else - sig = host->id ==0 ?OMAP_DMA_MMC_TX :OMAP_DMA_MMC2_TX ; - host->dma_tx =dma_request_channel (mask, omap_dma_filter_fn, &sig); + res =platform_get_resource_byname (pdev,IORESOURCE_DMA ,"tx" ); +if (res) + sig = res->start; + host->dma_tx =dma_request_slave_channel_compat (mask, + omap_dma_filter_fn, &sig, &pdev->dev,"tx" );Minor comment, since we're moving to DT-only for platforms using this driver (hope I'm right about that), why not just do:
Quoted 4 lines
dma_request_slave_channel_(&pdev->dev, "tx");
IORESOURCE_DMAis not created by OF layer so I guess no need to callplatform_get_resourceeither.
Well this is not the omap_hsmmc.c driver, this is the one for earlier hardware that's available on omap1 and 2420. All the later ones use the omap_hsmmc.c driver instead. So this one we cannot make DT only unless we make mach-omap1 DT only, which is doable, but may never happen because of the effort needed.
Most of the effort would be quite trivial, except for the conversion to use the common clock framework.
Regards,
Tony
Tony Lindgren
maintainerPATCH 3/8 mmc: omap: Fix I2C dependency and make driver usable with device tree
In reply to Tony LindgrenSome features can be configured by the companion I2C chips, which may not be available at the probe time. Fix the issue by returning -EPROBE_DEFER when the MMC controller slots are not configured.
While at it, let's also add minimal device tree support so omap24xx platforms can use this driver without legacy mode since we claim to support device tree for mach-omap2 based systems.
Although adding the minimal device tree support is not strictly a fix, it does remove one of the last blockers for dropping a bunch of legacy platform data for mach-omap2.
---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
--- drivers/mmc/host/omap.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-)
drivers/mmc/host/omap.c +9 -1
@@ -22,6 +22,7 @@#include <linux/delay.h> #include <linux/spinlock.h> #include <linux/timer.h> +#include <linux/of.h> #include <linux/omap-dma.h> #include <linux/mmc/host.h> #include <linux/mmc/card.h>
@@ -1330,7 +1331,7 @@ static int mmc_omap_probe(struct platform_device *pdev) }if (pdata->nr_slots ==0 ) {dev_err (&pdev->dev,"no slots\n" ); -return -ENXIO ; +return -EPROBE_DEFER ; } res =platform_get_resource (pdev,IORESOURCE_MEM ,0 );
@@ -1553,6 +1554,12 @@ static int mmc_omap_resume(struct platform_device *pdev)#define mmc_omap_resume NULL#endif +#if IS_BUILTIN (CONFIG_OF ) +static const struct of_device_id mmc_omap_match[] = { + { .compatible ="ti,omap2420-mmc" , }, + { }, +}; +#endif static struct platform_driver mmc_omap_driver = { .probe = mmc_omap_probe, .remove = mmc_omap_remove,
@@ -1561,6 +1568,7 @@ static struct platform_driver mmc_omap_driver = { .driver = { .name = DRIVER_NAME, .owner = THIS_MODULE, + .of_match_table = of_match_ptr(mmc_omap_match), }, };
Signature · 1 line
1.8.1.1
PATCH 3/8 mmc: omap: Fix I2C dependency and make driver usable with device tree
Quoted 49 lines · Tony Lindgren
Some features can be configured by the companion I2C chips, which may not be available at the probe time. Fix the issue by returning -EPROBE_DEFER when the MMC controller slots are not configured.
While at it, let's also add minimal device tree support so omap24xx platforms can use this driver without legacy mode since we claim to support device tree for mach-omap2 based systems.
Although adding the minimal device tree support is not strictly a fix, it does remove one of the last blockers for dropping a bunch of legacy platform data for mach-omap2.
Cc Chris Ball <[email protected]>Cc linux-mmc at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
--- drivers/mmc/host/omap.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-)drivers/mmc/host/omap.c +2
-1@@ -22,6 +22,7 @@#include <linux/delay.h> #include <linux/spinlock.h> #include <linux/timer.h> +#include <linux/of.h> #include <linux/omap-dma.h> #include <linux/mmc/host.h> #include <linux/mmc/card.h> @@ -1330,7 +1331,7 @@ static int mmc_omap_probe(struct platform_device *pdev) }if (pdata->nr_slots ==0 ) {dev_err (&pdev->dev,"no slots\n" ); -return -ENXIO ; +return -EPROBE_DEFER ; } res =platform_get_resource (pdev,IORESOURCE_MEM ,0 );
drivers/mmc/host/omap.c static int mmc_omap_resume(struct platform_device *pdev) · line 1559
Quoted 6 lines
drivers/mmc/host/omap.c +3
-0@@ -1553,6 +1554,12 @@ static int mmc_omap_resume(struct platform_device *pdev)#define mmc_omap_resume NULL#endif +#if IS_BUILTIN (CONFIG_OF ) +static const struct of_device_id mmc_omap_match[] = { + { .compatible = "ti,omap2420-mmc", },
Missing binding document.
Thanks,
Mark.
Tony Lindgren
maintainerPATCH 3/8 mmc: omap: Fix I2C dependency and make driver usable with device tree
Quoted 2 lines · Mark Rutland
Missing binding document.
Here's this one updated with a minimal binding document. The ti,hwmods is still needed until we have removed the dependencies to hwmod code for omaps.
Regards,
Tony
From: Tony Lindgren <[email protected]>
Date: Wed, 13 Nov 2013 16:36:37 -0800
Subject: [PATCH] mmc: omap: Fix I2C dependency and make driver usable with device tree
Some features can be configured by the companion I2C chips, which may not be available at the probe time. Fix the issue by returning -EPROBE_DEFER when the MMC controller slots are not configured.
While at it, let's also add minimal device tree support so omap24xx platforms can use this driver without legacy mode since we claim to support device tree for mach-omap2 based systems.
Although adding the minimal device tree support is not strictly a fix, it does remove one of the last blockers for dropping a bunch of legacy platform data for mach-omap2.
+++ b/Documentation/devicetree/bindings/mmc/ti-omap.txt
@@ -0,0 +1,27 @@
+* TI MMC host controller for OMAP1 and 2420
+
+The MMC Host Controller on TI OMAP1 and 2420 family provides +an interface for MMC, SD, and SDIO types of memory cards. +
+This file documents differences between the core properties described +by mmc.txt and the properties used by the omap mmc driver. +
+Note that this driver will not work with omap2430 or later omaps, +please see the omap hsmmc driver for the current omaps.
+
+Required properties:
+- compatible: Must be "ti,omap2420-mmc", for OMAP2420 controllers +- ti,hwmods: For 2420, must be "msdi<n>", where n is controller + instance starting 1
+
+Examples:
+
+ msdi1: mmc at 4809c000 {
+ compatible = "ti,omap2420-mmc";
+ ti,hwmods = "msdi1";
+ reg = <0x4809c000 0x80>;
+ interrupts = <83>;
+ dmas = <&sdma 61 &sdma 62>;
+ dma-names = "tx", "rx";
+ };
+
drivers/mmc/host/omap.c +9 -1
@@ -22,6 +22,7 @@#include <linux/delay.h> #include <linux/spinlock.h> #include <linux/timer.h> +#include <linux/of.h> #include <linux/omap-dma.h> #include <linux/mmc/host.h> #include <linux/mmc/card.h>
@@ -1330,7 +1331,7 @@ static int mmc_omap_probe(struct platform_device *pdev) }if (pdata->nr_slots ==0 ) {dev_err (&pdev->dev,"no slots\n" ); -return -ENXIO ; +return -EPROBE_DEFER ; } res =platform_get_resource (pdev,IORESOURCE_MEM ,0 );
@@ -1553,6 +1554,12 @@ static int mmc_omap_resume(struct platform_device *pdev)#define mmc_omap_resume NULL#endif +#if IS_BUILTIN (CONFIG_OF ) +static const struct of_device_id mmc_omap_match[] = { + { .compatible ="ti,omap2420-mmc" , }, + { }, +}; +#endif static struct platform_driver mmc_omap_driver = { .probe = mmc_omap_probe, .remove = mmc_omap_remove,
@@ -1561,6 +1568,7 @@ static struct platform_driver mmc_omap_driver = { .driver = { .name = DRIVER_NAME, .owner = THIS_MODULE, + .of_match_table = of_match_ptr(mmc_omap_match), }, };
PATCH 3/8 mmc: omap: Fix I2C dependency and make driver usable with device tree
Hi Tony,
Quoted 10 lines · Tony Lindgren
Some features can be configured by the companion I2C chips, which may not be available at the probe time. Fix the issue by returning -EPROBE_DEFER when the MMC controller slots are not configured.
While at it, let's also add minimal device tree support so omap24xx platforms can use this driver without legacy mode since we claim to support device tree for mach-omap2 based systems.
Quoted 7 lines
Although adding the minimal device tree support is not strictly a fix, it does remove one of the last blockers for dropping a bunch of legacy platform data for mach-omap2.
Cc Chris Ball <[email protected]>Cc linux-mmc at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>
Feel free to merge via your tree:
- Chris.
Signature · 1 line
Chris Ball <[email protected]> <http://printf.net/>
Tony Lindgren
maintainerPATCH 4/8 i2c: omap: Fix missing device tree flags for omap2
In reply to Tony LindgrenAs we claim to support device tree for mach-omap2, we should have the necessary flags in the driver to make it usable.
---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
--- drivers/i2c/busses/i2c-omap.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+)
drivers/i2c/busses/i2c-omap.c +22 -0
@@ -1037,6 +1037,20 @@ static const struct i2c_algorithm omap_i2c_algo = { };#ifdef CONFIG_OF +static struct omap_i2c_bus_platform_data omap2420_pdata = { + .rev =OMAP_I2C_IP_VERSION_1 , + .flags =OMAP_I2C_FLAG_NO_FIFO | +OMAP_I2C_FLAG_SIMPLE_CLOCK | +OMAP_I2C_FLAG_16BIT_DATA_REG | +OMAP_I2C_FLAG_BUS_SHIFT_2 , +}; + +static struct omap_i2c_bus_platform_data omap2430_pdata = { + .rev =OMAP_I2C_IP_VERSION_1 , + .flags =OMAP_I2C_FLAG_BUS_SHIFT_2 | +OMAP_I2C_FLAG_FORCE_19200_INT_CLK , +}; +static struct omap_i2c_bus_platform_data omap3_pdata = { .rev =OMAP_I2C_IP_VERSION_1 , .flags = OMAP_I2C_FLAG_BUS_SHIFT_2,
@@ -1055,6 +1069,14 @@ static const struct of_device_id omap_i2c_of_match[] = { .compatible = "ti,omap3-i2c", .data = &omap3_pdata, }, + { + .compatible = "ti,omap2430-i2c", + .data = &omap2430_pdata, + }, + { + .compatible = "ti,omap2420-i2c", + .data = &omap2420_pdata, + }, { }, };MODULE_DEVICE_TABLE (of, omap_i2c_of_match);
Signature · 1 line
1.8.1.1
PATCH 4/8 i2c: omap: Fix missing device tree flags for omap2
Quoted 7 lines · Tony Lindgren
As we claim to support device tree for mach-omap2, we should have the necessary flags in the driver to make it usable.
Cc Wolfram Sang <[email protected]>Cc linux-i2c at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>
It would have been helpful if the message "PATCH [0/x]" would have been sent to the i2c-list also.
Thanks,
Wolfram
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20131114/2a1363a3/attachment-0001.sig>
Tony Lindgren
maintainerPATCH 4/8 i2c: omap: Fix missing device tree flags for omap2
Quoted 5 lines · Wolfram Sang
Quoted 4 lines · Tony Lindgren
As we claim to support device tree for mach-omap2, we should have the necessary flags in the driver to make it usable.
Quoted 8 lines
Quoted 3 lines
Cc Wolfram Sang <[email protected]>Cc linux-i2c at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>Acked-by Wolfram Sang <[email protected]>It would have been helpful if the message "PATCH [0/x]" would have been sent to the i2c-list also.
Thanks, next time I'll try check the cc list in the cover letter manually after running git format patch. I guess there's no way to deal with that in an automated way.
Here's a link to the whole thread for reference:
PATCH 4/8 i2c: omap: Fix missing device tree flags for omap2
Quoted 7 lines
Quoted 3 lines
It would have been helpful if the message "PATCH [0/x]" would have been sent to the i2c-list also.
Thanks, next time I'll try check the cc list in the cover letter manually after running git format patch. I guess there's no way to deal with that in an automated way.
I use this hackish script as --cc-cmd with git:
#! /bin/sh
#
# cocci_cc - send cover letter to all mailing lists referenced in a patch series # done by Wolfram Sang in 2012 - WTFPLv2
name=${1##*/}
num=${name%%-*}
if [ "$num" = "0000" ]; then
dir=${1%/*}
for f in $dir/*; do
patchname=${f##*/}
[ "${patchname%%-*}" = "0000" ] && continue
scripts/get_maintainer.pl --no-m $f
done | sort -uelse
scripts/get_maintainer.pl $1
fi
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20131114/147fd74a/attachment-0001.sig>
Tony Lindgren
maintainerPATCH 4/8 i2c: omap: Fix missing device tree flags for omap2
Quoted 25 lines · Wolfram Sang
Quoted 7 lines
Quoted 3 lines
It would have been helpful if the message "PATCH [0/x]" would have been sent to the i2c-list also.
Thanks, next time I'll try check the cc list in the cover letter manually after running git format patch. I guess there's no way to deal with that in an automated way.
I use this hackish script as --cc-cmd with git:
#!
/bin/sh
## cocci_cc - send cover letter to all mailing lists referenced in a patch series # done by Wolfram Sang in 2012 - WTFPLv2name=${1##*/}
num=${name%%-*}if [ "$num" = "0000" ]; then
dir=${1%/*} for f in $dir/*; do patchname=${f##*/} [ "${patchname%%-*}" = "0000" ] && continue scripts/get_maintainer.pl --no-m $f done | sort -u
Quoted 4 lines
else
scripts/get_maintainer.pl$1
fi
Cool thanks :) I'll give it a try next time.
Tony
PATCH 4/8 i2c: omap: Fix missing device tree flags for omap2
In reply to Tony LindgrenQuoted 16 lines · Tony Lindgren
As we claim to support device tree for mach-omap2, we should have the necessary flags in the driver to make it usable.
Cc Wolfram Sang <[email protected]>Cc linux-i2c at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
--- drivers/i2c/busses/i2c-omap.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+)
[...]
drivers/i2c/busses/i2c-omap.c static const struct of_device_id omap_i2c_of_match[] = { · line 1078
Quoted 7 lines
drivers/i2c/busses/i2c-omap.c +7
-0+ { + .compatible = "ti,omap2430-i2c", + .data = &omap2430_pdata, + }, + { + .compatible = "ti,omap2420-i2c", + .data = &omap2420_pdata,
Please update Documentation/devicetree/bindings/i2c/i2c-omap.txt
Otherwise, this is fine.
Thanks,
Mark.
Tony Lindgren
maintainerPATCH 4/8 i2c: omap: Fix missing device tree flags for omap2
Quoted 4 lines · Mark Rutland
Please update
Documentation/devicetree/bindings/i2c/i2c-omap.txtOtherwise, this is fine.
Here's this one with updated documentation.
Regards,
Tony
From: Tony Lindgren <[email protected]>
Date: Wed, 13 Nov 2013 16:36:37 -0800
Subject: [PATCH] i2c: omap: Fix missing device tree flags for omap2
As we claim to support device tree for mach-omap2, we should have the necessary flags in the driver to make it usable.
Documentation/devicetree/bindings/i2c/i2c-omap.txt +2 -1
@@ -1,7 +1,8 @@ I2C for OMAP platforms Required properties : -- compatible : Must be "ti,omap3-i2c" or "ti,omap4-i2c" +- compatible : Must be "ti,omap2420-i2c", "ti,omap2430-i2c", "ti,omap3-i2c" + or "ti,omap4-i2c" - ti,hwmods : Must be "i2c<n>", n being the instance number (1-based) - #address-cells = <1>; - #size-cells = <0>;
drivers/i2c/busses/i2c-omap.c +22 -0
@@ -1037,6 +1037,20 @@ static const struct i2c_algorithm omap_i2c_algo = { };#ifdef CONFIG_OF +static struct omap_i2c_bus_platform_data omap2420_pdata = { + .rev =OMAP_I2C_IP_VERSION_1 , + .flags =OMAP_I2C_FLAG_NO_FIFO | +OMAP_I2C_FLAG_SIMPLE_CLOCK | +OMAP_I2C_FLAG_16BIT_DATA_REG | +OMAP_I2C_FLAG_BUS_SHIFT_2 , +}; + +static struct omap_i2c_bus_platform_data omap2430_pdata = { + .rev =OMAP_I2C_IP_VERSION_1 , + .flags =OMAP_I2C_FLAG_BUS_SHIFT_2 | +OMAP_I2C_FLAG_FORCE_19200_INT_CLK , +}; +static struct omap_i2c_bus_platform_data omap3_pdata = { .rev =OMAP_I2C_IP_VERSION_1 , .flags = OMAP_I2C_FLAG_BUS_SHIFT_2,
@@ -1055,6 +1069,14 @@ static const struct of_device_id omap_i2c_of_match[] = { .compatible = "ti,omap3-i2c", .data = &omap3_pdata, }, + { + .compatible = "ti,omap2430-i2c", + .data = &omap2430_pdata, + }, + { + .compatible = "ti,omap2420-i2c", + .data = &omap2420_pdata, + }, { }, };MODULE_DEVICE_TABLE (of, omap_i2c_of_match);
Tony Lindgren
maintainerPATCH 5/8 gpio: twl4030: Fix regression for twl gpio output
In reply to Tony LindgrenCommit c111feabe2e2 (gpio: twl4030: Cache the direction and output states in private data) improved things in general, but caused a regression for setting the GPIO output direction.
The change reorganized twl_direction_out() and twl_set() and swapped the function names around in the process. While doing that, a bug got introduced that's not obvious while reading the patch as it appears as no change to the code.
The bug is we now call function twl4030_set_gpio_dataout() twice in both twl_direction_out() and twl_set(). Instead, we should first call twl_direction_out() in twl_direction_out() followed by twl4030_set_gpio_dataout() in twl_set().
This regression probably has gone unnoticed for a long time as the bootloader may have set the GPIO direction properly in many cases. This fixes at least the LCD panel not turning on omap3 LDP for example.
---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
--- drivers/gpio/gpio-twl4030.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/gpio/gpio-twl4030.c +3 -2
@@ -354,17 +354,18 @@ static void twl_set(struct gpio_chip *chip, unsigned offset, int value)static int twl_direction_out (struct gpio_chip *chip,unsigned offset,int value) {struct gpio_twl4030_priv *priv =to_gpio_twl4030 (chip); +int ret = -EINVAL ;mutex_lock (&priv->mutex);if (offset <TWL4030_GPIO_MAX ) -twl4030_set_gpio_dataout (offset, value); + ret =twl4030_set_gpio_direction (offset,0 ); priv->direction |=BIT (offset);mutex_unlock (&priv->mutex);twl_set (chip, offset, value); -return 0 ; +return ret; }static int twl_to_irq (struct gpio_chip *chip,unsigned offset)
Signature · 1 line
1.8.1.1
PATCH 5/8 gpio: twl4030: Fix regression for twl gpio output
Hi Tony,
Quoted 14 lines · Tony Lindgren
Commit
c111feabe2e2(gpio: twl4030: Cache the direction and output states in private data) improved things in general, but caused a regression for setting the GPIO output direction.The change reorganized
twl_direction_out()andtwl_set()and swapped the function names around in the process. While doing that, a bug got introduced that's not obvious while reading the patch as it appears as no change to the code.The bug is we now call function
twl4030_set_gpio_dataout()twice in bothtwl_direction_out()andtwl_set(). Instead, we should first calltwl_direction_out()intwl_direction_out()followed bytwl4030_set_gpio_dataout()intwl_set().
Quoted 4 lines
This regression probably has gone unnoticed for a long time as the bootloader may have set the GPIO direction properly in many cases. This fixes at least the LCD panel not turning on omap3 LDP for example.
Thanks for catching this. I have a recollection that I have tested the GPIO and it appeared to be working fine..
Quoted 40 lines
Cc Linus Walleij <[email protected]>Cc Peter Ujfalusi <[email protected]>Cc linux-gpio at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
--- drivers/gpio/gpio-twl4030.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)drivers/gpio/gpio-twl4030.c +3
-2@@ -354,17 +354,18 @@ static void twl_set(struct gpio_chip *chip, unsigned offset, int value)static int twl_direction_out (struct gpio_chip *chip,unsigned offset,int value) {struct gpio_twl4030_priv *priv =to_gpio_twl4030 (chip); +int ret = -EINVAL ;mutex_lock (&priv->mutex);if (offset <TWL4030_GPIO_MAX ) -twl4030_set_gpio_dataout (offset, value); + ret =twl4030_set_gpio_direction (offset,0 ); priv->direction |=BIT (offset);mutex_unlock (&priv->mutex);twl_set (chip, offset, value); -return 0 ; +return ret; }static int twl_to_irq (struct gpio_chip *chip,unsigned offset)
Signature · 1 line
P?ter
Tony Lindgren
maintainerPATCH 5/8 gpio: twl4030: Fix regression for twl gpio output
Quoted 17 lines · Peter Ujfalusi
Hi Tony,
Quoted 14 lines · Tony Lindgren
Commit
c111feabe2e2(gpio: twl4030: Cache the direction and output states in private data) improved things in general, but caused a regression for setting the GPIO output direction.The change reorganized
twl_direction_out()andtwl_set()and swapped the function names around in the process. While doing that, a bug got introduced that's not obvious while reading the patch as it appears as no change to the code.The bug is we now call function
twl4030_set_gpio_dataout()twice in bothtwl_direction_out()andtwl_set(). Instead, we should first calltwl_direction_out()intwl_direction_out()followed bytwl4030_set_gpio_dataout()intwl_set().
Quoted 7 lines
Quoted 4 lines
This regression probably has gone unnoticed for a long time as the bootloader may have set the GPIO direction properly in many cases. This fixes at least the LCD panel not turning on omap3 LDP for example.
Thanks for catching this. I have a recollection that I have tested the GPIO and it appeared to be working fine..
Yes it probably worked fine if the bootloader set the direction :) This is cc stable kind of patch for sure.
Regards,
Tony
PATCH 5/8 gpio: twl4030: Fix regression for twl gpio output
In reply to Tony LindgrenQuoted 23 lines · Tony Lindgren
Commit
c111feabe2e2(gpio: twl4030: Cache the direction and output states in private data) improved things in general, but caused a regression for setting the GPIO output direction.The change reorganized
twl_direction_out()andtwl_set()and swapped the function names around in the process. While doing that, a bug got introduced that's not obvious while reading the patch as it appears as no change to the code.The bug is we now call function
twl4030_set_gpio_dataout()twice in bothtwl_direction_out()andtwl_set(). Instead, we should first calltwl_direction_out()intwl_direction_out()followed bytwl4030_set_gpio_dataout()intwl_set().This regression probably has gone unnoticed for a long time as the bootloader may have set the GPIO direction properly in many cases. This fixes at least the LCD panel not turning on omap3 LDP for example.
Cc Linus Walleij <[email protected]>Cc Peter Ujfalusi <[email protected]>Cc linux-gpio at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>
Quoted 5 lines
---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
Sure:
Yours,
Linus Walleij
PATCH 5/8 gpio: twl4030: Fix regression for twl gpio output
In reply to Tony LindgrenHi Tony,
Quoted 23 lines · Tony Lindgren
Commit
c111feabe2e2(gpio: twl4030: Cache the direction and output states in private data) improved things in general, but caused a regression for setting the GPIO output direction.The change reorganized
twl_direction_out()andtwl_set()and swapped the function names around in the process. While doing that, a bug got introduced that's not obvious while reading the patch as it appears as no change to the code.The bug is we now call function
twl4030_set_gpio_dataout()twice in bothtwl_direction_out()andtwl_set(). Instead, we should first calltwl_direction_out()intwl_direction_out()followed bytwl4030_set_gpio_dataout()intwl_set().This regression probably has gone unnoticed for a long time as the bootloader may have set the GPIO direction properly in many cases. This fixes at least the LCD panel not turning on omap3 LDP for example.
Cc Linus Walleij <[email protected]>Cc Peter Ujfalusi <[email protected]>Cc linux-gpio at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>
Quoted 5 lines
---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
This patch causes a regression with LED outputs (GPO) on twl4030 on 3.13-rc2. As one of the LED GPO is used for USB host on beagleboard, it will cause failure of USB host probe.
I'll explain why below.
Quoted 10 lines
--- drivers/gpio/gpio-twl4030.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)drivers/gpio/gpio-twl4030.c +0
-0
drivers/gpio/gpio-twl4030.c static void twl_set(struct gpio_chip *chip, unsigned offset, int value) · line 361
Quoted 9 lines
drivers/gpio/gpio-twl4030.c +2
-1@@ -354,17 +354,18 @@ static void twl_set(struct gpio_chip *chip, unsigned offset, int value)static int twl_direction_out (struct gpio_chip *chip,unsigned offset,int value) {struct gpio_twl4030_priv *priv =to_gpio_twl4030 (chip); +int ret = -EINVAL ;mutex_lock (&priv->mutex);if (offset <TWL4030_GPIO_MAX ) -twl4030_set_gpio_dataout (offset, value); + ret =twl4030_set_gpio_direction (offset,0 );
twl_direction_out() is supposed to return 0 on success and non-zero only on failure.
for (offset >= TWL4030_GPIO_MAX) i.e. LED output case we now return -EINVAL, which causes the LED GPO set_direction_out to fail. LED outputs are always outputs so this function shouldn't fail.
drivers/gpio/gpio-twl4030.c static void twl_set(struct gpio_chip *chip, unsigned offset, int value) · line 371
Quoted 12 lines
drivers/gpio/gpio-twl4030.c +1
-1priv->direction |=BIT (offset);mutex_unlock (&priv->mutex);twl_set (chip, offset, value); -return 0 ; +return ret; }static int twl_to_irq (struct gpio_chip *chip,unsigned offset)
Below is a proposed fix for this.
Quoted 1 line
From
7c36c8952ee3c7220ea21396cd3f84a1f9e9ea73Mon Sep 17 00:00:00 2001
From: Roger Quadros <[email protected]>
Date: Tue, 3 Dec 2013 15:24:05 +0200
Subject: [PATCH] gpio: twl4030: Fix regression for twl gpio LED output
Commit 0b2aa8be introduced a regression that causes failure in setting LED GPO direction to OUT.
This causes USB host probe failures for Beagleboard C4.
[ 2.075469] platform usb_phy_gen_xceiv.2: Driver usb_phy_gen_xceiv requests probe deferral [ 2.090454] hsusb2_vcc: Failed to request enable GPIO510: -22 [ 2.100982] reg-fixed-voltage reg-fixed-voltage.0.auto: Failed to register regulator: -22 [ 2.109954] reg-fixed-voltage: probe of reg-fixed-voltage.0.auto failed with error -22
direction_out/direction_in must return 0 if the operation succeeded.
--- drivers/gpio/gpio-twl4030.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-)
drivers/gpio/gpio-twl4030.c +9 -6
@@ -294,13 +294,13 @@ out:static int twl_direction_in (struct gpio_chip *chip,unsigned offset) {struct gpio_twl4030_priv *priv =to_gpio_twl4030 (chip); -int ret; +int ret =0 ;mutex_lock (&priv->mutex);if (offset <TWL4030_GPIO_MAX ) - ret =twl4030_set_gpio_direction (offset,1 ); +twl4030_set_gpio_direction (offset,1 );else - ret = -EINVAL ; + ret = -EINVAL ;/* LED outputs can't be set as input */ if (!ret) priv->direction &= ~BIT (offset);
@@ -354,18 +354,21 @@ static void twl_set(struct gpio_chip *chip, unsigned offset, int value)static int twl_direction_out (struct gpio_chip *chip,unsigned offset,int value) {struct gpio_twl4030_priv *priv =to_gpio_twl4030 (chip); -int ret = -EINVAL ;mutex_lock (&priv->mutex);if (offset <TWL4030_GPIO_MAX ) - ret =twl4030_set_gpio_direction (offset,0 ); +twl4030_set_gpio_direction (offset,0 ); + +/* +* LED gpio's i.e. offset >= TWL4030_GPIO_MAX are always output +*/ priv->direction |=BIT (offset);mutex_unlock (&priv->mutex);twl_set (chip, offset, value); -return ret; +return 0 ; }static int twl_to_irq (struct gpio_chip *chip,unsigned offset)
Signature · 1 line
1.8.3.2
PATCH 5/8 gpio: twl4030: Fix regression for twl gpio output
On Tue, Dec 3, 2013 at 2:30 PM, Roger Quadros <[email protected]> wrote:
(...)
Quoted 3 lines
This patch causes a regression with LED outputs (GPO) on twl4030 on 3.13-rc2. As one of the LED GPO is used for USB host on beagleboard, it will cause failure of USB host probe.
(...)
Quoted 1 line
Below is a proposed fix for this.
Roger, Tony: what happened with this? Have you hashed this out? I ACK Roger's fixup too if that needs to be merged into the OMAP tree.
Yours,
Linus Walleij
Tony Lindgren
maintainerPATCH 5/8 gpio: twl4030: Fix regression for twl gpio output
Quoted 6 lines · Linus Walleij
On Tue, Dec 3, 2013 at 2:30 PM, Roger Quadros <[email protected]> wrote:
(...)
Quoted 3 lines
This patch causes a regression with LED outputs (GPO) on twl4030 on 3.13-rc2. As one of the LED GPO is used for USB host on beagleboard, it will cause failure of USB host probe.
Quoted 6 lines
(...)
Quoted 1 line
Below is a proposed fix for this.
Roger, Tony: what happened with this? Have you hashed this out? I ACK Roger's fixup too if that needs to be merged into the OMAP tree.
There's an updated version from Roger posted that I acked:
[PATCH v2 1/1] gpio: twl4030: Fix regression for twl gpio LED output
http://lkml.org/lkml/2013/12/5/65
I also commented on the dependencies there, but basically you can pick it up unless you want me to. Should be cc stable as well.
Regards,
Tony
PATCH 5/8 gpio: twl4030: Fix regression for twl gpio output
Quoted 14 lines · Tony Lindgren
Quoted 6 lines · Linus Walleij
On Tue, Dec 3, 2013 at 2:30 PM, Roger Quadros <[email protected]> wrote:
(...)
Quoted 3 lines
This patch causes a regression with LED outputs (GPO) on twl4030 on 3.13-rc2. As one of the LED GPO is used for USB host on beagleboard, it will cause failure of USB host probe.
Quoted 6 lines
(...)
Quoted 1 line
Below is a proposed fix for this.
Roger, Tony: what happened with this? Have you hashed this out? I ACK Roger's fixup too if that needs to be merged into the OMAP tree.
Quoted 8 lines
There's an updated version from Roger posted that I acked:
[PATCH v2 1/1] gpio: twl4030: Fix regression for twl gpio LED output
http://lkml.org/lkml/2013/12/5/65
I also commented on the dependencies there, but basically you can pick it up unless you want me to. Should be cc stable as well.
OK I have picked this now, I'll add a CC to stable.
Yours,
Linus Walleij
Tony Lindgren
maintainerPATCH 5/8 gpio: twl4030: Fix regression for twl gpio output
Quoted 17 lines · Linus Walleij
Quoted 16 lines · Tony Lindgren
Quoted 6 lines · Linus Walleij
On Tue, Dec 3, 2013 at 2:30 PM, Roger Quadros <[email protected]> wrote:
(...)
Quoted 3 lines
This patch causes a regression with LED outputs (GPO) on twl4030 on 3.13-rc2. As one of the LED GPO is used for USB host on beagleboard, it will cause failure of USB host probe.
Quoted 6 lines
(...)
Quoted 1 line
Below is a proposed fix for this.
Roger, Tony: what happened with this? Have you hashed this out? I ACK Roger's fixup too if that needs to be merged into the OMAP tree.
There's an updated version from Roger posted that I acked:
Quoted 8 lines
Quoted 6 lines
[PATCH v2 1/1] gpio: twl4030: Fix regression for twl gpio LED output
http://lkml.org/lkml/2013/12/5/65
I also commented on the dependencies there, but basically you can pick it up unless you want me to. Should be cc stable as well.
OK I have picked this now, I'll add a CC to stable.
Thanks!
Tony
Tony Lindgren
maintainerPATCH 6/8 gpio: twl4030: Fix passing of pdata in the device tree case
In reply to Tony LindgrenWe still have some legacy code needing the callback functions that won't work properly without platform data. To use platform data for twl4030-gpio, we need to not trash the possible data.
---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
--- drivers/gpio/gpio-twl4030.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/gpio/gpio-twl4030.c +6 -2
@@ -436,7 +436,8 @@ static int gpio_twl4030_debounce(u32 debounce, u8 mmc_cd)static int gpio_twl4030_remove (struct platform_device *pdev); -static struct twl4030_gpio_platform_data *of_gpio_twl4030 (struct device *dev) +static struct twl4030_gpio_platform_data *of_gpio_twl4030 (struct device *dev, +struct twl4030_gpio_platform_data *pdata) {struct twl4030_gpio_platform_data *omap_twl_info;
@@ -444,6 +445,9 @@ static struct twl4030_gpio_platform_data *of_gpio_twl4030(struct device *dev)if (!omap_twl_info)return NULL ; +if (pdata) +memcpy (omap_twl_info, pdata,sizeof (*omap_twl_info)); + omap_twl_info->use_leds =of_property_read_bool (dev->of_node,"ti,use-leds" );
@@ -501,7 +505,7 @@ no_irqs:mutex_init (&priv->mutex);if (node) - pdata =of_gpio_twl4030 (&pdev->dev); + pdata =of_gpio_twl4030 (&pdev->dev, pdata);if (pdata ==NULL ) {dev_err (&pdev->dev,"Platform data is missing\n" );
Signature · 1 line
1.8.1.1
Tony Lindgren
maintainerPATCH 6/8 gpio: twl4030: Fix passing of pdata in the device tree case
Quoted 4 lines · Tony Lindgren
We still have some legacy code needing the callback functions that won't work properly without platform data. To use platform data for twl4030-gpio, we need to not trash the possible data.
Quoted 8 lines
Cc Linus Walleij <[email protected]>Cc linux-gpio at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>---
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
Here's this one with a fix from Fengguang Wu to use struct assignment instead of memcpy folded in.
Regards,
Tony
From: Tony Lindgren <[email protected]>
Date: Thu, 14 Nov 2013 15:25:08 -0800
Subject: [PATCH] gpio: twl4030: Fix passing of pdata in the device tree case
We still have some legacy code needing the callback functions that won't work properly without platform data. To use platform data for twl4030-gpio, we need to not trash the possible data.
drivers/gpio/gpio-twl4030.c +6 -2
@@ -436,7 +436,8 @@ static int gpio_twl4030_debounce(u32 debounce, u8 mmc_cd)static int gpio_twl4030_remove (struct platform_device *pdev); -static struct twl4030_gpio_platform_data *of_gpio_twl4030 (struct device *dev) +static struct twl4030_gpio_platform_data *of_gpio_twl4030 (struct device *dev, +struct twl4030_gpio_platform_data *pdata) {struct twl4030_gpio_platform_data *omap_twl_info;
@@ -444,6 +445,9 @@ static struct twl4030_gpio_platform_data *of_gpio_twl4030(struct device *dev)if (!omap_twl_info)return NULL ; +if (pdata) + *omap_twl_info = *pdata; + omap_twl_info->use_leds =of_property_read_bool (dev->of_node,"ti,use-leds" );
@@ -501,7 +505,7 @@ no_irqs:mutex_init (&priv->mutex);if (node) - pdata =of_gpio_twl4030 (&pdev->dev); + pdata =of_gpio_twl4030 (&pdev->dev, pdata);if (pdata ==NULL ) {dev_err (&pdev->dev,"Platform data is missing\n" );
PATCH 6/8 gpio: twl4030: Fix passing of pdata in the device tree case
Quoted 10 lines · Tony Lindgren
Quoted 9 lines · Tony Lindgren
We still have some legacy code needing the callback functions that won't work properly without platform data. To use platform data for twl4030-gpio, we need to not trash the possible data.
Cc Linus Walleij <[email protected]>Cc linux-gpio at vger.kernel.orgSigned-off-by Tony Lindgren <[email protected]>---
Quoted 6 lines
Quoted 3 lines
If this looks OK, I'd like to merge this as a fix via arm-soc tree along with the other patches in this series as my later patches depend on patches in this series.
Here's this one with a fix from Fengguang Wu to use struct assignment instead of memcpy folded in.
Looks allright:
Yours,
Linus Walleij
Tony Lindgren
maintainerPATCH 7/8 ARM: OMAP2+: Fix GPMC and simplify bootloader timings for 8250 and smc91x
In reply to Tony LindgrenCommit f2bf0e72d000 (ARM: OMAP2+: Add minimal 8250 support for GPMC) added support for using bootloader timings for some devices. Turns out we can do the same by looking at the compatible flags of the child without adding a new function as smc91x has a similar issue as 8250 with the bootloader timings.
And let's fix the 8250 naming, we should use the device type as the name like uart instead of 8250 for zoom dts file.
--- arch/arm/boot/dts/omap-zoom-common.dtsi | 2 +- arch/arm/mach-omap2/gpmc.c | 58 +++++++++++---------------------- 2 files changed, 20 insertions(+), 40 deletions(-)
arch/arm/boot/dts/omap-zoom-common.dtsi +1 -1
@@ -13,7 +13,7 @@ * they probably share the same GPIO IRQ * REVISIT: Add timing support from slls644g.pdf */ - 8250 at 3,0 { + uart at 3,0 { compatible = "ns16550a"; reg = <3 0 0x100>; bank-width = <2>;
arch/arm/mach-omap2/gpmc.c +19 -39
83 lines not on this page · raw message
Signature · 1 line
1.8.1.1
Tony Lindgren
maintainerPATCH 8/8 ARM: dts: Fix omap2 specific dtsi files by adding the missing entries
In reply to Tony LindgrenLooks like we're missing few entries for omap2 and the drivers have only worked because of the omap hwmod building the devices for the missing entries.
Let's fix the missing entries so we don't need to rely on hwmod for the basic data and can then later on remove the duplicate data from hwmod. Otherwise device tree only drivers will not work properly.
--- arch/arm/boot/dts/omap2.dtsi | 96 +++++++++++++++++++++++++++++++++++++++++ arch/arm/boot/dts/omap2420.dtsi | 23 ++++++++++ arch/arm/boot/dts/omap2430.dtsi | 49 +++++++++++++++++++++ 3 files changed, 168 insertions(+)
arch/arm/boot/dts/omap2.dtsi +96 -0
@@ -9,6 +9,7 @@ */ #include <dt-bindings/gpio/gpio.h> +#include <dt-bindings/interrupt-controller/irq.h> #include <dt-bindings/pinctrl/omap.h> #include "skeleton.dtsi"
@@ -21,6 +22,8 @@ serial0 = &uart1; serial1 = &uart2; serial2 = &uart3; + i2c0 = &i2c1; + i2c1 = &i2c2; }; cpus {
@@ -53,6 +56,28 @@ ranges; ti,hwmods = "l3_main"; + aes: aes at 480a6000 { + compatible = "ti,omap2-aes"; + ti,hwmods = "aes"; + reg = <0x480a6000 0x50>; + dmas = <&sdma 9 &sdma 10>; + dma-names = "tx", "rx"; + }; + + hdq1w: 1w at 480b2000 { + compatible = "ti,omap2420-1w"; + ti,hwmods = "hdq1w"; + reg = <0x480b2000 0x1000>; + interrupts = <58>; + }; + + mailbox: mailbox at 48094000 { + compatible = "ti,omap2-mailbox"; + ti,hwmods = "mailbox"; + reg = <0x48094000 0x200>; + interrupts = <26>; + }; + intc: interrupt-controller at 1 { compatible = "ti,omap2-intc"; interrupt-controller;
@@ -63,6 +88,7 @@ sdma: dma-controller at 48056000 { compatible = "ti,omap2430-sdma", "ti,omap2420-sdma"; + ti,hwmods = "dma"; reg = <0x48056000 0x1000>; interrupts = <12>, <13>,
@@ -73,21 +99,91 @@ #dma-requests = <64>; }; + i2c1: i2c at 48070000 { + compatible = "ti,omap2-i2c"; + ti,hwmods = "i2c1"; + reg = <0x48070000 0x80>; + #address-cells = <1>; + #size-cells = <0>; + interrupts = <56>; + dmas = <&sdma 27 &sdma 28>; + dma-names = "tx", "rx"; + }; + + i2c2: i2c at 48072000 { + compatible = "ti,omap2-i2c"; + ti,hwmods = "i2c2"; + reg = <0x48072000 0x80>; + #address-cells = <1>; + #size-cells = <0>; + interrupts = <57>; + dmas = <&sdma 29 &sdma 30>; + dma-names = "tx", "rx"; + }; + + mcspi1: mcspi at 48098000 { + compatible = "ti,omap2-mcspi"; + ti,hwmods = "mcspi1"; + reg = <0x48098000 0x100>; + interrupts = <65>; + dmas = <&sdma 35 &sdma 36 &sdma 37 &sdma 38 + &sdma 39 &sdma 40 &sdma 41 &sdma 42>; + dma-names = "tx0", "rx0", "tx1", "rx1", + "tx2", "rx2", "tx3", "rx3"; + }; + + mcspi2: mcspi at 4809a000 { + compatible = "ti,omap2-mcspi"; + ti,hwmods = "mcspi2"; + reg = <0x4809a000 0x100>; + interrupts = <66>; + dmas = <&sdma 43 &sdma 44 &sdma 45 &sdma 46>; + dma-names = "tx0", "rx0", "tx1", "rx1"; + }; + + rng: rng at 480a0000 { + compatible = "ti,omap2-rng"; + ti,hwmods = "rng"; + reg = <0x480a0000 0x50>; + interrupts = <36>; + }; + + sham: sham at 480a4000 { + compatible = "ti,omap2-sham"; + ti,hwmods = "sham"; + reg = <0x480a4000 0x64>; + interrupts = <51>; + dmas = <&sdma 13>; + dma-names = "rx"; + }; + uart1: serial at 4806a000 { compatible = "ti,omap2-uart"; ti,hwmods = "uart1"; + reg = <0x4806a000 0x2000>; + interrupts = <72>; + dmas = <&sdma 49 &sdma 50>; + dma-names = "tx", "rx"; clock-frequency = <48000000>; }; uart2: serial at 4806c000 { compatible = "ti,omap2-uart"; ti,hwmods = "uart2"; + reg = <0x4806c000 0x400>; + interrupts = <73>; + dmas = <&sdma 51 &sdma 52>; + dma-names = "tx", "rx"; clock-frequency = <48000000>; }; uart3: serial at 4806e000 { compatible = "ti,omap2-uart"; ti,hwmods = "uart3"; + reg = <0x4806e000 0x400>; + interrupts = <74>; + dmas = <&sdma 53 &sdma 54>; + dma-names = "tx", "rx"; clock-frequency = <48000000>; };
arch/arm/boot/dts/omap2420.dtsi +23 -0
@@ -114,6 +114,15 @@ dma-names = "tx", "rx"; }; + msdi1: mmc at 4809c000 { + compatible = "ti,omap2420-mmc"; + ti,hwmods = "msdi1"; + reg = <0x4809c000 0x80>; + interrupts = <83>; + dmas = <&sdma 61 &sdma 62>; + dma-names = "tx", "rx"; + }; + timer1: timer at 48028000 { compatible = "ti,omap2420-timer"; reg = <0x48028000 0x400>;
@@ -121,5 +130,19 @@ ti,hwmods = "timer1"; ti,timer-alwon; }; + + wd_timer2: wdt at 48022000 { + compatible = "ti,omap2-wdt"; + ti,hwmods = "wd_timer2"; + reg = <0x48022000 0x80>; + }; }; }; + +&i2c1 { + compatible = "ti,omap2420-i2c"; +}; + +&i2c2 { + compatible = "ti,omap2420-i2c"; +};
arch/arm/boot/dts/omap2430.dtsi +49 -0
60 lines not on this page · raw message
Signature · 1 line
1.8.1.1
Tony Lindgren
maintainer[PATCH 9/8] i2c: Fix device tree binding for i2c-cbus-gpio
In reply to Tony LindgrenLooks like we're missing two lines needed to make it work properly with device tree.
---
Wolfram, I found one more bug booting omaps with device tree.
drivers/i2c/busses/i2c-cbus-gpio.c +2 -0
@@ -246,6 +246,7 @@ static int cbus_i2c_probe(struct platform_device *pdev) adapter->owner =THIS_MODULE ; adapter->class =I2C_CLASS_HWMON ; adapter->dev.parent = &pdev->dev; + adapter->dev.of_node = pdev->dev.of_node; adapter->nr = pdev->id; adapter->timeout =HZ ; adapter->algo = &cbus_i2c_algo;
@@ -289,6 +290,7 @@ static struct platform_driver cbus_i2c_driver = { .driver = { .owner = THIS_MODULE, .name = "i2c-cbus-gpio", + .of_match_table = of_match_ptr(i2c_cbus_dt_ids), }, };module_platform_driver (cbus_i2c_driver);
Aaro Koskinen
maintainer[PATCH 9/8] i2c: Fix device tree binding for i2c-cbus-gpio
Hi,
Quoted 5 lines · Tony Lindgren
Looks like we're missing two lines needed to make it work properly with device tree.
Cc linux-i2c at vger.kernel.orgCc Aaro Koskinen <[email protected]>
You can change this to: Tested-by: <[email protected]>
I booted DT-N800 with Tony's patches, and it works.
Thanks,
A.
Wolfram Sang
Applied[PATCH 9/8] i2c: Fix device tree binding for i2c-cbus-gpio
In reply to Tony LindgrenQuoted 7 lines · Tony Lindgren
Looks like we're missing two lines needed to make it work properly with device tree.
Cc linux-i2c at vger.kernel.orgCc Aaro Koskinen <[email protected]>Cc Wolfram Sang <[email protected]>Signed-off-by Tony Lindgren <[email protected]>
Applied to for-next, thanks!
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20131115/44668d17/attachment-0001.sig>
Tony Lindgren
maintainer[PATCH 9/8] i2c: Fix device tree binding for i2c-cbus-gpio
Quoted 10 lines · Wolfram Sang
Quoted 7 lines · Tony Lindgren
Looks like we're missing two lines needed to make it work properly with device tree.
Cc linux-i2c at vger.kernel.orgCc Aaro Koskinen <[email protected]>Cc Wolfram Sang <[email protected]>Signed-off-by Tony Lindgren <[email protected]>Applied to for-next, thanks!
Thanks, will drop this one from my fixes series.
Tony