Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
On Mon, Jun 29, 2009 at 07:25:43PM +0300, Yusuf Caglar AKYUZ wrote: Denys Dmytriyenko wrote: Another heads up - the next branch is now rebased with vpfe v3 patches. Thanks again. Now I'm using your next branch for my own development. FYI, I've rebased the next branch of linux-davinci-staging tree on arago with Kevin's latest 2.6.31-rc2 and VPFE from Hans' pull request. The master is set aside to track 2.6.30 for now. BTW, does this tree includes musb patches which was submitted to usb mailing list. but yet not included in mainline and davinci trees? As David and Kevin mentioned, those are in upstream 2.6.31-rcX now. -- Denys ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
On Monday 29 June 2009, Yusuf Caglar AKYUZ wrote: BTW, does this tree includes musb patches which was submitted to usb mailing list. but yet not included in mainline and davinci trees? Can you please be more specific which musb patches you are referring to? I must admit that I lost track of musb patches. Currently I see: * One keyboard and mouse at the same time does not work. I remember seeing some mentioning of improvements to the interrupt end points on the list. * DMA is not usable and transfer speeds are so slow. I guess I need to check some archives. That stuff is in mainline, but hasn't yet been pulled down into the DaVinci tree. ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
David Brownell davi...@pacbell.net writes: On Monday 29 June 2009, Yusuf Caglar AKYUZ wrote: BTW, does this tree includes musb patches which was submitted to usb mailing list. but yet not included in mainline and davinci trees? Can you please be more specific which musb patches you are referring to? I must admit that I lost track of musb patches. Currently I see: * One keyboard and mouse at the same time does not work. I remember seeing some mentioning of improvements to the interrupt end points on the list. * DMA is not usable and transfer speeds are so slow. I guess I need to check some archives. That stuff is in mainline, but hasn't yet been pulled down into the DaVinci tree. By next week, DaVinci git will be merged with 2.6.31-rcX, so whould have these changes. Kevin ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
Denys Dmytriyenko wrote: On Fri, Jun 26, 2009 at 12:46:31AM +0300, Yusuf Caglar AKYUZ wrote: Karicheri, Muralidharan wrote: We have been using the arago git tree and I don't see any of the issues that you are seeing. Could you use arago tree and Let me know if you see the same issue? This tree is being tested internally by TI. So it is known to work. http://arago-project.org/git/people/sneha/linux-davinci-staging.git P.S. AFAIR, arago tree is based on 2.6.30-rc2. Eventually we have to test these patches with davinci tree. That's the confusion I was afraid of - the latest patches against 2.6.30 final were tested in the next branch, while master stayed slightly old. One of the reasons was to achieve feature-parity of the next branch before making it the new master, which happened today. But, as I explained in the previous email, that still has v2 vpfe patches, which I'll start rebasing tomorrow... Thanks for the heads up. This tree boots fine, without any issues. A minor problem is asoc code is not building without CONFIG_SND_DM6467_SOC_EVM=y. BTW, does this tree includes musb patches which was submitted to usb mailing list. but yet not included in mainline and davinci trees? Thanks, Caglar ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
On Fri, Jun 26, 2009 at 05:09:12PM +0300, Yusuf Caglar AKYUZ wrote: Denys Dmytriyenko wrote: On Fri, Jun 26, 2009 at 12:46:31AM +0300, Yusuf Caglar AKYUZ wrote: Karicheri, Muralidharan wrote: We have been using the arago git tree and I don't see any of the issues that you are seeing. Could you use arago tree and Let me know if you see the same issue? This tree is being tested internally by TI. So it is known to work. http://arago-project.org/git/people/sneha/linux-davinci-staging.git P.S. AFAIR, arago tree is based on 2.6.30-rc2. Eventually we have to test these patches with davinci tree. That's the confusion I was afraid of - the latest patches against 2.6.30 final were tested in the next branch, while master stayed slightly old. One of the reasons was to achieve feature-parity of the next branch before making it the new master, which happened today. But, as I explained in the previous email, that still has v2 vpfe patches, which I'll start rebasing tomorrow... Thanks for the heads up. This tree boots fine, without any issues. A minor problem is asoc code is not building without CONFIG_SND_DM6467_SOC_EVM=y. Another heads up - the next branch is now rebased with vpfe v3 patches. BTW, does this tree includes musb patches which was submitted to usb mailing list. but yet not included in mainline and davinci trees? Can you please be more specific which musb patches you are referring to? Thanks. -- Denys ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
Kevin Hilman wrote: m-kariche...@ti.com writes: From: Muralidharan Karicheri a0868...@gt516km11.gt.design.ti.com Big Thanks to all reviewers who have contributed to this driver by reviewing and offering valuable comments. VPFE Capture driver for DaVinci Media SOCs :- DM355 and DM6446 This is the version v3 of the patch series. This is the reworked version of the driver based on comments received against the last version (v2) of the patch and is expected to be final version candidate for merge to upstream kernel FYI... I've updated the staging/vpfe branch of davinci git with this series and the tvp514x v3 patch. Also, I'll be pushing the arch/arm/* patches of this series to DaVinci git master and queueing them for upstream merge. Has anyone using this staging tree? I'm having hard time to making this work. My problems are: 1) This branch does not built without attached patch. 2) tvp514x throws attached kernel oops. I guess this due missing platform data master definition in board support file. 3) Even if I correct these two and compile drivers built-in then I see an oops message related to asoc code on boot. 4) If I make these drivers kernel modules, then I can install all of them but dm6446_ccdc. Don't know why, just no messages on dmesg. I tried v1 and v2 of these patches previously, they were have the same problem of (3) but not the others. I was able to make v1 and v2 working but this one is a dead end due to (4). Thanks, Caglar diff --git a/arch/arm/mach-davinci/board-dm644x-evm.c b/arch/arm/mach-davinci/board-dm644x-evm.c index 13b73a7..c1b8747 100644 --- a/arch/arm/mach-davinci/board-dm644x-evm.c +++ b/arch/arm/mach-davinci/board-dm644x-evm.c @@ -29,6 +29,7 @@ #include linux/phy.h #include linux/clk.h #include linux/videodev2.h +#include media/v4l2-int-device.h #include media/tvp514x.h #include asm/setup.h #include asm/mach-types.h diff --git a/drivers/media/video/davinci/vpfe_capture.c b/drivers/media/video/davinci/vpfe_capture.c index 003450b..4ffb164 100644 --- a/drivers/media/video/davinci/vpfe_capture.c +++ b/drivers/media/video/davinci/vpfe_capture.c @@ -70,7 +70,7 @@ #include linux/platform_device.h #include linux/interrupt.h #include linux/version.h -#include media/v4l2-common.h +#include media/v4l2-int-device.h #include linux/io.h #include media/davinci/vpfe_capture.h #include media/tvp514x.h Unable to handle kernel NULL pointer dereference at virtual address pgd = c5dd8000 [] *pgd=85dce031, *pte=, *ppte= Internal error: Oops: 17 [#1] PREEMPT Modules linked in: tvp514x(+) videobuf_dma_contig videobuf_core ipv6 CPU: 0Not tainted (2.6.30-davinci1 #105) PC is at strcpy+0x4/0x18 LR is at tvp514x_probe+0xf0/0x18c [tvp514x] pc : [c01efd5c]lr : [bf04fce8]psr: 4013 sp : c6309e50 ip : c048579c fp : 0008b480 r10: r9 : c6308000 r8 : c049e6c4 r7 : bf051724 r6 : bf0500e8 r5 : c5c15e00 r4 : c5de1c00 r3 : 000c r2 : c5de1c3c r1 : r0 : c5de1c3c Flags: nZcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user Control: 0005317f Table: 85dd8000 DAC: 0015 Process insmod (pid: 993, stack limit = 0xc6308268) Stack: (0xc6309e50 to 0xc630a000) 9e40: c5c15e20 bf04fbf8 c5c15e00 9e60: c5c15e20 c0291fcc c5c15e20 c5c15e54 bf051724 c022e7c0 c5c56b70 c5c15e20 9e80: c5c15e54 bf051724 bf051724 c022e8d8 c6309ea0 c022e878 c022dca4 9ea0: c5c2efb8 c5c56b70 c5c2ef88 bf0516f8 bf051724 c632f900 bf054000 c022e258 9ec0: bf0507e0 0001 c6920088 bf0516f8 bf051724 000ac018 bf054000 c04a74cc 9ee0: c022ecb4 0021 bf0516f8 bf051aa8 000ac018 bf054000 c04a74cc 9f00: c0292f94 4a57 bf051aa8 c00292fc c691ff45 c57e3de0 9f20: c6921558 001f 0021 c69205b0 bf051ab4 c6920588 00b5 9f40: 9f60: 4a57 bf051aa8 000ac018 4a57 bf051aa8 000ac018 c0029f64 9f80: 000a96ac c0063bc0 c6308000 000ac018 4a57 01b0 be9c8ecd 9fa0: 0080 c0029de0 4a57 01b0 000ac018 4a57 000ac008 01b0 9fc0: 4a57 01b0 be9c8ecd 0080 00099f58 0046 000a96ac 0008b480 9fe0: be9c8c40 be9c8c30 00017110 40182ff0 6010 000ac018 e28dd004 e8bd4030 [c01efd5c] (strcpy+0x4/0x18) from [bf04fce8] (tvp514x_probe+0xf0/0x18c [tvp514x]) [bf04fce8] (tvp514x_probe+0xf0/0x18c [tvp514x]) from [c0291fcc] (i2c_device_probe+0x68/0x80) [c0291fcc] (i2c_device_probe+0x68/0x80) from [c022e7c0] (driver_probe_device+0xac/0x164) [c022e7c0] (driver_probe_device+0xac/0x164) from [c022e8d8] (__driver_attach+0x60/0x84) [c022e8d8] (__driver_attach+0x60/0x84) from [c022dca4] (bus_for_each_dev+0x4c/0x80) [c022dca4] (bus_for_each_dev+0x4c/0x80) from [c022e258]
Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
Karicheri, Muralidharan wrote: Yusuf, I am not sure what is the state of vpfe branch maintained by Kevin. We have been using the arago git tree and I don't see any of the issues that you are seeing. Could you use arago tree and Let me know if you see the same issue? This tree is being tested internally by TI. So it is known to work. http://arago-project.org/git/people/sneha/linux-davinci-staging.git I tried this one before and the results were the same. However, that time I imported vpfe stuff from arago tree into davinci tree. But I don't remember what was the result of arago tree. I guess I'll give this another try. Thanks, Caglar P.S. AFAIR, arago tree is based on 2.6.30-rc2. Eventually we have to test these patches with davinci tree. ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
RE: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
Yusuf, I am not sure what is the state of vpfe branch maintained by Kevin. We have been using the arago git tree and I don't see any of the issues that you are seeing. Could you use arago tree and Let me know if you see the same issue? This tree is being tested internally by TI. So it is known to work. http://arago-project.org/git/people/sneha/linux-davinci-staging.git Murali Karicheri Software Design Engineer Texas Instruments Inc. Germantown, MD 20874 Phone : 301-515-3736 email: m-kariche...@ti.com -Original Message- From: Yusuf Caglar AKYUZ [mailto:caglarak...@gmail.com] Sent: Thursday, June 25, 2009 3:37 PM To: Kevin Hilman Cc: Karicheri, Muralidharan; davinci-linux-open-source@linux.davincidsp.com Subject: Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver Kevin Hilman wrote: m-kariche...@ti.com writes: From: Muralidharan Karicheri a0868...@gt516km11.gt.design.ti.com Big Thanks to all reviewers who have contributed to this driver by reviewing and offering valuable comments. VPFE Capture driver for DaVinci Media SOCs :- DM355 and DM6446 This is the version v3 of the patch series. This is the reworked version of the driver based on comments received against the last version (v2) of the patch and is expected to be final version candidate for merge to upstream kernel FYI... I've updated the staging/vpfe branch of davinci git with this series and the tvp514x v3 patch. Also, I'll be pushing the arch/arm/* patches of this series to DaVinci git master and queueing them for upstream merge. Has anyone using this staging tree? I'm having hard time to making this work. My problems are: 1) This branch does not built without attached patch. 2) tvp514x throws attached kernel oops. I guess this due missing platform data master definition in board support file. 3) Even if I correct these two and compile drivers built-in then I see an oops message related to asoc code on boot. 4) If I make these drivers kernel modules, then I can install all of them but dm6446_ccdc. Don't know why, just no messages on dmesg. I tried v1 and v2 of these patches previously, they were have the same problem of (3) but not the others. I was able to make v1 and v2 working but this one is a dead end due to (4). Thanks, Caglar ___ _ diff --git a/arch/arm/mach-davinci/board-dm644x-evm.c b/arch/arm/mach-davinci/board-dm644x-evm.c index 13b73a7..c1b8747 100644 --- a/arch/arm/mach-davinci/board-dm644x-evm.c +++ b/arch/arm/mach-davinci/board-dm644x-evm.c @@ -29,6 +29,7 @@ #include linux/phy.h #include linux/clk.h #include linux/videodev2.h +#include media/v4l2-int-device.h #include media/tvp514x.h #include asm/setup.h #include asm/mach-types.h diff --git a/drivers/media/video/davinci/vpfe_capture.c b/drivers/media/video/davinci/vpfe_capture.c index 003450b..4ffb164 100644 --- a/drivers/media/video/davinci/vpfe_capture.c +++ b/drivers/media/video/davinci/vpfe_capture.c @@ -70,7 +70,7 @@ #include linux/platform_device.h #include linux/interrupt.h #include linux/version.h -#include media/v4l2-common.h +#include media/v4l2-int-device.h #include linux/io.h #include media/davinci/vpfe_capture.h #include media/tvp514x.h ___ _ Unable to handle kernel NULL pointer dereference at virtual address pgd = c5dd8000 [] *pgd=85dce031, *pte=, *ppte= Internal error: Oops: 17 [#1] PREEMPT Modules linked in: tvp514x(+) videobuf_dma_contig videobuf_core ipv6 CPU: 0Not tainted (2.6.30-davinci1 #105) PC is at strcpy+0x4/0x18 LR is at tvp514x_probe+0xf0/0x18c [tvp514x] pc : [c01efd5c]lr : [bf04fce8]psr: 4013 sp : c6309e50 ip : c048579c fp : 0008b480 r10: r9 : c6308000 r8 : c049e6c4 r7 : bf051724 r6 : bf0500e8 r5 : c5c15e00 r4 : c5de1c00 r3 : 000c r2 : c5de1c3c r1 : r0 : c5de1c3c Flags: nZcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user Control: 0005317f Table: 85dd8000 DAC: 0015 Process insmod (pid: 993, stack limit = 0xc6308268) Stack: (0xc6309e50 to 0xc630a000) 9e40: c5c15e20 bf04fbf8 c5c15e00 9e60: c5c15e20 c0291fcc c5c15e20 c5c15e54 bf051724 c022e7c0 c5c56b70 c5c15e20 9e80: c5c15e54 bf051724 bf051724 c022e8d8 c6309ea0 c022e878 c022dca4 9ea0: c5c2efb8 c5c56b70 c5c2ef88 bf0516f8 bf051724 c632f900 bf054000 c022e258 9ec0: bf0507e0 0001 c6920088 bf0516f8 bf051724 000ac018 bf054000 c04a74cc 9ee0: c022ecb4 0021 bf0516f8 bf051aa8 000ac018 bf054000 c04a74cc 9f00: c0292f94 4a57 bf051aa8 c00292fc c691ff45 c57e3de0 9f20: c6921558 001f 0021 c69205b0 bf051ab4 c6920588 00b5 9f40: 9f60: 4a57 bf051aa8 000ac018 4a57 bf051aa8 000ac018 c0029f64 9f80
Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
On Thu, Jun 25, 2009 at 04:38:13PM -0500, Karicheri, Muralidharan wrote: Yusuf, I am not sure what is the state of vpfe branch maintained by Kevin. In one of his posts few days ago Kevin said he updated it to v3. We have been using the arago git tree and I don't see any of the issues that you are seeing. Could you use arago tree and Let me know if you see the same issue? This tree is being tested internally by TI. So it is known to work. http://arago-project.org/git/people/sneha/linux-davinci-staging.git To avoid confusion, I moved away the previous old -rc2 based state into old-rc2 branch and made the next branch the current master. But it still has the v2 vpfe patches - I'm planning to rebase the staging tree from Kevin's tomorrow and will pick the v3 patches then. -- Denys Murali Karicheri Software Design Engineer Texas Instruments Inc. Germantown, MD 20874 Phone : 301-515-3736 email: m-kariche...@ti.com -Original Message- From: Yusuf Caglar AKYUZ [mailto:caglarak...@gmail.com] Sent: Thursday, June 25, 2009 3:37 PM To: Kevin Hilman Cc: Karicheri, Muralidharan; davinci-linux-open-source@linux.davincidsp.com Subject: Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver Kevin Hilman wrote: m-kariche...@ti.com writes: From: Muralidharan Karicheri a0868...@gt516km11.gt.design.ti.com Big Thanks to all reviewers who have contributed to this driver by reviewing and offering valuable comments. VPFE Capture driver for DaVinci Media SOCs :- DM355 and DM6446 This is the version v3 of the patch series. This is the reworked version of the driver based on comments received against the last version (v2) of the patch and is expected to be final version candidate for merge to upstream kernel FYI... I've updated the staging/vpfe branch of davinci git with this series and the tvp514x v3 patch. Also, I'll be pushing the arch/arm/* patches of this series to DaVinci git master and queueing them for upstream merge. Has anyone using this staging tree? I'm having hard time to making this work. My problems are: 1) This branch does not built without attached patch. 2) tvp514x throws attached kernel oops. I guess this due missing platform data master definition in board support file. 3) Even if I correct these two and compile drivers built-in then I see an oops message related to asoc code on boot. 4) If I make these drivers kernel modules, then I can install all of them but dm6446_ccdc. Don't know why, just no messages on dmesg. I tried v1 and v2 of these patches previously, they were have the same problem of (3) but not the others. I was able to make v1 and v2 working but this one is a dead end due to (4). Thanks, Caglar ___ _ diff --git a/arch/arm/mach-davinci/board-dm644x-evm.c b/arch/arm/mach-davinci/board-dm644x-evm.c index 13b73a7..c1b8747 100644 --- a/arch/arm/mach-davinci/board-dm644x-evm.c +++ b/arch/arm/mach-davinci/board-dm644x-evm.c @@ -29,6 +29,7 @@ #include linux/phy.h #include linux/clk.h #include linux/videodev2.h +#include media/v4l2-int-device.h #include media/tvp514x.h #include asm/setup.h #include asm/mach-types.h diff --git a/drivers/media/video/davinci/vpfe_capture.c b/drivers/media/video/davinci/vpfe_capture.c index 003450b..4ffb164 100644 --- a/drivers/media/video/davinci/vpfe_capture.c +++ b/drivers/media/video/davinci/vpfe_capture.c @@ -70,7 +70,7 @@ #include linux/platform_device.h #include linux/interrupt.h #include linux/version.h -#include media/v4l2-common.h +#include media/v4l2-int-device.h #include linux/io.h #include media/davinci/vpfe_capture.h #include media/tvp514x.h ___ _ Unable to handle kernel NULL pointer dereference at virtual address pgd = c5dd8000 [] *pgd=85dce031, *pte=, *ppte= Internal error: Oops: 17 [#1] PREEMPT Modules linked in: tvp514x(+) videobuf_dma_contig videobuf_core ipv6 CPU: 0Not tainted (2.6.30-davinci1 #105) PC is at strcpy+0x4/0x18 LR is at tvp514x_probe+0xf0/0x18c [tvp514x] pc : [c01efd5c]lr : [bf04fce8]psr: 4013 sp : c6309e50 ip : c048579c fp : 0008b480 r10: r9 : c6308000 r8 : c049e6c4 r7 : bf051724 r6 : bf0500e8 r5 : c5c15e00 r4 : c5de1c00 r3 : 000c r2 : c5de1c3c r1 : r0 : c5de1c3c Flags: nZcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user Control: 0005317f Table: 85dd8000 DAC: 0015 Process insmod (pid: 993, stack limit = 0xc6308268) Stack: (0xc6309e50 to 0xc630a000) 9e40: c5c15e20 bf04fbf8 c5c15e00 9e60: c5c15e20 c0291fcc c5c15e20 c5c15e54 bf051724 c022e7c0 c5c56b70 c5c15e20 9e80: c5c15e54 bf051724 bf051724 c022e8d8 c6309ea0 c022e878 c022dca4 9ea0
Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
On Fri, Jun 26, 2009 at 12:46:31AM +0300, Yusuf Caglar AKYUZ wrote: Karicheri, Muralidharan wrote: We have been using the arago git tree and I don't see any of the issues that you are seeing. Could you use arago tree and Let me know if you see the same issue? This tree is being tested internally by TI. So it is known to work. http://arago-project.org/git/people/sneha/linux-davinci-staging.git P.S. AFAIR, arago tree is based on 2.6.30-rc2. Eventually we have to test these patches with davinci tree. That's the confusion I was afraid of - the latest patches against 2.6.30 final were tested in the next branch, while master stayed slightly old. One of the reasons was to achieve feature-parity of the next branch before making it the new master, which happened today. But, as I explained in the previous email, that still has v2 vpfe patches, which I'll start rebasing tomorrow... -- Denys ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
RE: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
Hi, Check the following log... The crash doesn't seem to directly related to this patch. Did you get any error while applying the patch. You seem to have not build tvp514x. So no sub device available to register in vpfe capture. So enable it. It is part of encoder/decoder menu... vpfe_init vpfe-capture: vpfe vpss master clock enabled vpfe-capture: vpfe vpss slave clock enabled vpfe-capture vpfe-capture: v4l2 device registered vpfe-capture vpfe-capture: video device registered vpfe-capture vpfe-capture: No Subdevice data vpfe-capture: vpfe vpss master slave clocks disabled Murali Karicheri Software Design Engineer Texas Instruments Inc. Germantown, MD 20874 Phone : 301-515-3736 email: m-kariche...@ti.com -Original Message- From: davinci-linux-open-source-boun...@linux.davincidsp.com [mailto:davinci-linux-open-source-boun...@linux.davincidsp.com] On Behalf Of Scott Sweeny Sent: Friday, June 19, 2009 2:57 PM To: davinci-linux-open-source@linux.davincidsp.com Subject: Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver On Wed, 2009-06-17 at 16:11 -0400, m-kariche...@ti.com wrote: From: Muralidharan Karicheri a0868...@gt516km11.gt.design.ti.com Big Thanks to all reviewers who have contributed to this driver by reviewing and offering valuable comments. VPFE Capture driver for DaVinci Media SOCs :- DM355 and DM6446 This is the version v3 of the patch series. This is the reworked version of the driver based on comments received against the last version (v2) of the patch and is expected to be final version candidate for merge to upstream kernel +++ These patches add support for VPFE (Video Processing Front End) based video capture on DM355 and DM6446 EVMs. For more details on the hardware configuration and capabilities, please refer the vpfe_capture.c header. This patch set consists of following:- Patch 1: VPFE Capture bridge driver Patch 2: CCDC hw device header file Patch 3: DM355 CCDC hw module Patch 4: DM644x CCDC hw module Patch 5: ccdc types used across CCDC modules Patch 6: Makefile and config files for the driver Patch 7: DM355 platform and board setup Patch 8: DM644x platform and board setup Patch 9: common vpss hw module for video drivers Patch 10: Remove outdated driver files from davinci git tree Patch 11: Makefile and config files for the davinci git tree (New from v2) NOTE: 1. Patches 10-11 are only for DaVinci GIT tree. Others applies to DaVinci GIT and v4l-dvb 2. Dependent on the TVP514x decoder driver patch for migrating the driver to sub device model from Vaibhav Hiremath. I am sending the reworked version of this patch instead of Vaibhav. Following tests are performed. 1) Capture and display video (PAL NTSC) from tvp5146 decoder. Displayed using fbdev device driver available on davinci git tree 2) Tested with driver built statically and dynamically Muralidhara Karicheri Reviewed by: Hans Verkuil hverk...@xs4all.nl Reviewed by: Laurent Pinchart laurent.pinch...@skynet.be Reviewed by: Alexey Klimov klimov.li...@gmail.com Reviewed by: Kevin Hilman khil...@deeprootsystems.com Reviewed by: David Brownell davi...@pacbell.net Signed-off-by: Muralidharan Karicheri m-kariche...@ti.com ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source After applying this patch and enabling the driver I get a kernel panic on every boot up on my DM355 EVM from Spectrum Digital. With some experimentation I found that the panic only happens when I enable CONFIG_VIDEO_DM355_CCDC. I have applied all but #6 in this series to the davinci-linux git tree as well as v3 of the TVP514x-Migration-to-sub-device-framework patch (which I believe is the one mentioned in the second note. Does anything obvious that I might be missing jump out at anyone? I've attached the log of the kernel panic. Thanks in advance, ~Scott -- Scott Sweeny Linux Engineer Timesys Corporation ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
RE: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
Steve, I have responded to the issue. All CCDC modules will try to register with vpfe capture, but only one will succeed. This is done to allow same build works on all platforms as per review comments received on the driver. In this case, nothing succeeded since vpfe capture fail to initialize due to no sub device available to use. Murali Karicheri Software Design Engineer Texas Instruments Inc. Germantown, MD 20874 Phone : 301-515-3736 email: m-kariche...@ti.com -Original Message- From: davinci-linux-open-source-bounces+m- karicheri2=ti@linux.davincidsp.com [mailto:davinci-linux-open-source- bounces+m-karicheri2=ti@linux.davincidsp.com] On Behalf Of Steve Chen Sent: Friday, June 19, 2009 3:46 PM To: davinci-linux-open-source@linux.davincidsp.com Subject: Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver On Fri, 2009-06-19 at 14:56 -0400, Scott Sweeny wrote: ... vpfe-capture: vpfe vpss master slave clocks disabled dm644x_ccdc_init vpfe_register_ccdc_device: DM6446 CCDC dm355_ccdc_init vpfe_register_ccdc_device: DM355 CCDC TCP cubic registered ... Noticed above lines in the kernel boot message. Seems strange having dm6446 ccdc initialized on dm355. May want to take a look. Regards, Steve ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
RE: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
On Mon, 2009-06-22 at 10:21 -0400, Karicheri, Muralidharan wrote: Hi, Check the following log... The crash doesn't seem to directly related to this patch. Did you get any error while applying the patch. You seem to have not build tvp514x. So no sub device available to register in vpfe capture. So enable it. It is part of encoder/decoder menu... vpfe_init vpfe-capture: vpfe vpss master clock enabled vpfe-capture: vpfe vpss slave clock enabled vpfe-capture vpfe-capture: v4l2 device registered vpfe-capture vpfe-capture: video device registered vpfe-capture vpfe-capture: No Subdevice data vpfe-capture: vpfe vpss master slave clocks disabled Murali Karicheri Software Design Engineer Texas Instruments Inc. Germantown, MD 20874 Phone : 301-515-3736 email: m-kariche...@ti.com Actually building the tvp514x driver fixed the problem. I didn't see the encoder/decoder menu because the Autoselect pertinent encoder/decoders... option was enabled by default. Thanks for all the help! Cheers, ~Scott -- Scott Sweeny Linux Engineer Timesys Corporation ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
On Friday 19 June 2009 22:49:07 Kevin Hilman wrote: m-kariche...@ti.com writes: From: Muralidharan Karicheri a0868...@gt516km11.gt.design.ti.com Big Thanks to all reviewers who have contributed to this driver by reviewing and offering valuable comments. VPFE Capture driver for DaVinci Media SOCs :- DM355 and DM6446 This is the version v3 of the patch series. This is the reworked version of the driver based on comments received against the last version (v2) of the patch and is expected to be final version candidate for merge to upstream kernel FYI... I've updated the staging/vpfe branch of davinci git with this series and the tvp514x v3 patch. Also, I'll be pushing the arch/arm/* patches of this series to DaVinci git master and queueing them for upstream merge. Hi Kevin, It's better to leave that to Mauro (the v4l-dvb maintainer): that will ensure that the v4l driver changes and the arch/arm changes go into the mainline in the correct order. If you prefer to do this yourself, then you should contact Mauro first. Also note that my pull request for this driver contains one additional patch fixing some missing newlines and a wrong error code. Here is the link to that diff in my repository: http://linuxtv.org/hg/%7Ehverkuil/v4l-dvb-vpfe-cap/raw-diff/ae2ca5c20d85/linux/drivers/media/video/davinci/vpfe_capture.c Regards, Hans Kevin +++ These patches add support for VPFE (Video Processing Front End) based video capture on DM355 and DM6446 EVMs. For more details on the hardware configuration and capabilities, please refer the vpfe_capture.c header. This patch set consists of following:- Patch 1: VPFE Capture bridge driver Patch 2: CCDC hw device header file Patch 3: DM355 CCDC hw module Patch 4: DM644x CCDC hw module Patch 5: ccdc types used across CCDC modules Patch 6: Makefile and config files for the driver Patch 7: DM355 platform and board setup Patch 8: DM644x platform and board setup Patch 9: common vpss hw module for video drivers Patch 10: Remove outdated driver files from davinci git tree Patch 11: Makefile and config files for the davinci git tree (New from v2) NOTE: 1. Patches 10-11 are only for DaVinci GIT tree. Others applies to DaVinci GIT and v4l-dvb 2. Dependent on the TVP514x decoder driver patch for migrating the driver to sub device model from Vaibhav Hiremath. I am sending the reworked version of this patch instead of Vaibhav. Following tests are performed. 1) Capture and display video (PAL NTSC) from tvp5146 decoder. Displayed using fbdev device driver available on davinci git tree 2) Tested with driver built statically and dynamically Muralidhara Karicheri Reviewed by: Hans Verkuil hverk...@xs4all.nl Reviewed by: Laurent Pinchart laurent.pinch...@skynet.be Reviewed by: Alexey Klimov klimov.li...@gmail.com Reviewed by: Kevin Hilman khil...@deeprootsystems.com Reviewed by: David Brownell davi...@pacbell.net Signed-off-by: Muralidharan Karicheri m-kariche...@ti.com ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source -- Hans Verkuil - video4linux developer - sponsored by TANDBERG Telecom ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
On Wed, 2009-06-17 at 16:11 -0400, m-kariche...@ti.com wrote: From: Muralidharan Karicheri a0868...@gt516km11.gt.design.ti.com Big Thanks to all reviewers who have contributed to this driver by reviewing and offering valuable comments. VPFE Capture driver for DaVinci Media SOCs :- DM355 and DM6446 This is the version v3 of the patch series. This is the reworked version of the driver based on comments received against the last version (v2) of the patch and is expected to be final version candidate for merge to upstream kernel +++ These patches add support for VPFE (Video Processing Front End) based video capture on DM355 and DM6446 EVMs. For more details on the hardware configuration and capabilities, please refer the vpfe_capture.c header. This patch set consists of following:- Patch 1: VPFE Capture bridge driver Patch 2: CCDC hw device header file Patch 3: DM355 CCDC hw module Patch 4: DM644x CCDC hw module Patch 5: ccdc types used across CCDC modules Patch 6: Makefile and config files for the driver Patch 7: DM355 platform and board setup Patch 8: DM644x platform and board setup Patch 9: common vpss hw module for video drivers Patch 10: Remove outdated driver files from davinci git tree Patch 11: Makefile and config files for the davinci git tree (New from v2) NOTE: 1. Patches 10-11 are only for DaVinci GIT tree. Others applies to DaVinci GIT and v4l-dvb 2. Dependent on the TVP514x decoder driver patch for migrating the driver to sub device model from Vaibhav Hiremath. I am sending the reworked version of this patch instead of Vaibhav. Following tests are performed. 1) Capture and display video (PAL NTSC) from tvp5146 decoder. Displayed using fbdev device driver available on davinci git tree 2) Tested with driver built statically and dynamically Muralidhara Karicheri Reviewed by: Hans Verkuil hverk...@xs4all.nl Reviewed by: Laurent Pinchart laurent.pinch...@skynet.be Reviewed by: Alexey Klimov klimov.li...@gmail.com Reviewed by: Kevin Hilman khil...@deeprootsystems.com Reviewed by: David Brownell davi...@pacbell.net Signed-off-by: Muralidharan Karicheri m-kariche...@ti.com ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source After applying this patch and enabling the driver I get a kernel panic on every boot up on my DM355 EVM from Spectrum Digital. With some experimentation I found that the panic only happens when I enable CONFIG_VIDEO_DM355_CCDC. I have applied all but #6 in this series to the davinci-linux git tree as well as v3 of the TVP514x-Migration-to-sub-device-framework patch (which I believe is the one mentioned in the second note. Does anything obvious that I might be missing jump out at anyone? I've attached the log of the kernel panic. Thanks in advance, ~Scott -- Scott Sweeny Linux Engineer Timesys Corporation Bytes transferred = 1551820 (17adcc hex) ## Booting image at 8070 ... Image Name: Linux-2.6.30-davinci1-05987-g3f6 Image Type: ARM Linux Kernel Image (uncompressed) Data Size:1551756 Bytes = 1.5 MB Load Address: 80008000 Entry Point: 80008000 Verifying Checksum ... OK OK Starting kernel ... Uncompressing Linux... done, booting the kernel. Linux version 2.6.30-davinci1-05987-g3f69672-dirty (sc...@scotbox) (gcc version 4.3.3 (GCC) ) #10 PREEMPT Fri Jun 19 14:42:58 EDT 2009 CPU: ARM926EJ-S [41069265] revision 5 (ARMv5TEJ), cr=00053177 CPU: VIVT data cache, VIVT instruction cache Machine: DaVinci DM355 EVM Memory policy: ECC disabled, Data cache writeback DaVinci dm355 variant 0x0 Built 1 zonelists in Zone order, mobility grouping on. Total pages: 29464 Kernel command line: mem=116M console=ttyS0,115200n8 root=/dev/nfs rw ip=dhcp video=davincifb:vid0=720x576x16,2500K:vid1=720x576x16,2500K:osd0=720x576x16,2025K davinci_enc_mngr.ch0_output=COMPOSITE davinci_enc_mngr.ch0_mode=ntsc Unknown boot option `davinci_enc_mngr.ch0_output=COMPOSITE': ignoring Unknown boot option `davinci_enc_mngr.ch0_mode=ntsc': ignoring NR_IRQS:224 PID hash table entries: 512 (order: 9, 2048 bytes) Console: colour dummy device 80x30 Dentry cache hash table entries: 16384 (order: 4, 65536 bytes) Inode-cache hash table entries: 8192 (order: 3, 32768 bytes) Memory: 116MB = 116MB total Memory: 114332KB available (2724K code, 298K data, 120K init, 0K highmem) SLUB: Genslabs=11, HWalign=32, Order=0-3, MinObjects=0, CPUs=1, Nodes=1 Calibrating delay loop... 107.72 BogoMIPS (lpj=538624) Mount-cache hash table entries: 512 CPU: Testing write buffer coherency: ok DaVinci: 104 gpio irqs net_namespace: 928 bytes NET: Registered protocol family 16 MUX: initialized INT_EDMA_CC MUX:
Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
On Fri, 2009-06-19 at 14:56 -0400, Scott Sweeny wrote: ... vpfe-capture: vpfe vpss master slave clocks disabled dm644x_ccdc_init vpfe_register_ccdc_device: DM6446 CCDC dm355_ccdc_init vpfe_register_ccdc_device: DM355 CCDC TCP cubic registered ... Noticed above lines in the kernel boot message. Seems strange having dm6446 ccdc initialized on dm355. May want to take a look. Regards, Steve ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
On Fri, 2009-06-19 at 15:45 -0400, Steve Chen wrote: On Fri, 2009-06-19 at 14:56 -0400, Scott Sweeny wrote: ... vpfe-capture: vpfe vpss master slave clocks disabled dm644x_ccdc_init vpfe_register_ccdc_device: DM6446 CCDC dm355_ccdc_init vpfe_register_ccdc_device: DM355 CCDC TCP cubic registered ... Noticed above lines in the kernel boot message. Seems strange having dm6446 ccdc initialized on dm355. May want to take a look. Regards, Steve ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source Steve, I thought of that as well but I'm afraid disabling the 6446 driver has no effect. Cheers, ~Scott -- Scott Sweeny Linux Engineer Timesys Corporation ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
Re: [PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
m-kariche...@ti.com writes: From: Muralidharan Karicheri a0868...@gt516km11.gt.design.ti.com Big Thanks to all reviewers who have contributed to this driver by reviewing and offering valuable comments. VPFE Capture driver for DaVinci Media SOCs :- DM355 and DM6446 This is the version v3 of the patch series. This is the reworked version of the driver based on comments received against the last version (v2) of the patch and is expected to be final version candidate for merge to upstream kernel FYI... I've updated the staging/vpfe branch of davinci git with this series and the tvp514x v3 patch. Also, I'll be pushing the arch/arm/* patches of this series to DaVinci git master and queueing them for upstream merge. Kevin +++ These patches add support for VPFE (Video Processing Front End) based video capture on DM355 and DM6446 EVMs. For more details on the hardware configuration and capabilities, please refer the vpfe_capture.c header. This patch set consists of following:- Patch 1: VPFE Capture bridge driver Patch 2: CCDC hw device header file Patch 3: DM355 CCDC hw module Patch 4: DM644x CCDC hw module Patch 5: ccdc types used across CCDC modules Patch 6: Makefile and config files for the driver Patch 7: DM355 platform and board setup Patch 8: DM644x platform and board setup Patch 9: common vpss hw module for video drivers Patch 10: Remove outdated driver files from davinci git tree Patch 11: Makefile and config files for the davinci git tree (New from v2) NOTE: 1. Patches 10-11 are only for DaVinci GIT tree. Others applies to DaVinci GIT and v4l-dvb 2. Dependent on the TVP514x decoder driver patch for migrating the driver to sub device model from Vaibhav Hiremath. I am sending the reworked version of this patch instead of Vaibhav. Following tests are performed. 1) Capture and display video (PAL NTSC) from tvp5146 decoder. Displayed using fbdev device driver available on davinci git tree 2) Tested with driver built statically and dynamically Muralidhara Karicheri Reviewed by: Hans Verkuil hverk...@xs4all.nl Reviewed by: Laurent Pinchart laurent.pinch...@skynet.be Reviewed by: Alexey Klimov klimov.li...@gmail.com Reviewed by: Kevin Hilman khil...@deeprootsystems.com Reviewed by: David Brownell davi...@pacbell.net Signed-off-by: Muralidharan Karicheri m-kariche...@ti.com ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
[PATCH 0/11 - v3] ARM: DaVinci: Video: DM355/DM6446 VPFE Capture driver
From: Muralidharan Karicheri a0868...@gt516km11.gt.design.ti.com Big Thanks to all reviewers who have contributed to this driver by reviewing and offering valuable comments. VPFE Capture driver for DaVinci Media SOCs :- DM355 and DM6446 This is the version v3 of the patch series. This is the reworked version of the driver based on comments received against the last version (v2) of the patch and is expected to be final version candidate for merge to upstream kernel +++ These patches add support for VPFE (Video Processing Front End) based video capture on DM355 and DM6446 EVMs. For more details on the hardware configuration and capabilities, please refer the vpfe_capture.c header. This patch set consists of following:- Patch 1: VPFE Capture bridge driver Patch 2: CCDC hw device header file Patch 3: DM355 CCDC hw module Patch 4: DM644x CCDC hw module Patch 5: ccdc types used across CCDC modules Patch 6: Makefile and config files for the driver Patch 7: DM355 platform and board setup Patch 8: DM644x platform and board setup Patch 9: common vpss hw module for video drivers Patch 10: Remove outdated driver files from davinci git tree Patch 11: Makefile and config files for the davinci git tree (New from v2) NOTE: 1. Patches 10-11 are only for DaVinci GIT tree. Others applies to DaVinci GIT and v4l-dvb 2. Dependent on the TVP514x decoder driver patch for migrating the driver to sub device model from Vaibhav Hiremath. I am sending the reworked version of this patch instead of Vaibhav. Following tests are performed. 1) Capture and display video (PAL NTSC) from tvp5146 decoder. Displayed using fbdev device driver available on davinci git tree 2) Tested with driver built statically and dynamically Muralidhara Karicheri Reviewed by: Hans Verkuil hverk...@xs4all.nl Reviewed by: Laurent Pinchart laurent.pinch...@skynet.be Reviewed by: Alexey Klimov klimov.li...@gmail.com Reviewed by: Kevin Hilman khil...@deeprootsystems.com Reviewed by: David Brownell davi...@pacbell.net Signed-off-by: Muralidharan Karicheri m-kariche...@ti.com ___ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
