V4L/DVB updates
in [Kernel]
|
Prev: [PATCH] pata_rdc: MWDMA0 is unsupported
Next: Problem? intel_iommu=off; perf top shows acpi_os_read_port as extremely busy
From: Mauro Carvalho Chehab on 30 Nov 2009 09:30 Em Mon, 30 Nov 2009 12:03:32 -0200 Mauro Carvalho Chehab <mchehab(a)redhat.com> escreveu: > Em Sat, 28 Nov 2009 03:57:58 +0100 > Hans Verkuil <hverkuil(a)xs4all.nl> escreveu: > > > On Friday 27 November 2009 22:40:01 Stefan Lippers-Hollmann wrote: > > > Hi > > > > > > On Friday 27 November 2009, Mauro Carvalho Chehab wrote: > > > > Hi Linus, > > > > > > > > Please pull from: > > > > ssh://master.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6.git for_linus > > > > > > > > For the following drivers and building fixes: > > > > > > > > - radio-gemtek-pci: fix double mutex_lock > > > > - v4l: add more missing linux/sched.h includes > > > > - soc-camera: properly initialise the device object when reusing > > > > - soc-camera: sh_mobile_ceu_camera: call pm_runtime_disable > > > > - em28xx: fix Reddo DVB-C USB TV Box GPIO > > > > - davinci: remove stray duplicate config pointer > > > > - SMS_SIANO_MDTV should depend on HAS_DMA > > > > - cxusb: Fix hang on DViCO FusionHDTV DVB-T Dual Digital 4 (rev 1) > > > > - sh_mobile_ceu_camera: fix compile warning > > > > - Fix wrong parameter order in memset > > > [...] > > > > > > Please consider cherry picking the following two patches from Hans > > > Verkuil [1]: > > > - add the missing s2250-loader.h I'm not finding the above patch. I remember I got one patch at the go7007 conversion series to V4L2 that were removing this patch that didn't exist upstream. I had to manually fix the patch to apply at my linux-next tree. The patches I actually merged from hans v4l-dvb-staging (after the tree rebase) were: changeset: 13455:0eb2e476ac2a parent: 13372:8bff7e6c44d4 user: Hans Verkuil <hverkuil(a)xs4all.nl> date: Mon Nov 16 19:07:42 2009 +0100 summary: go7007: Add struct v4l2_device. changeset: 13456:f6f0c5647657 user: Hans Verkuil <hverkuil(a)xs4all.nl> date: Mon Nov 16 19:13:51 2009 +0100 summary: s2250: Change module structure changeset: 13457:701aead61759 user: Hans Verkuil <hverkuil(a)xs4all.nl> date: Mon Nov 16 19:15:07 2009 +0100 summary: s2250: subdev conversion changeset: 13458:2ca6a9dc112b user: Hans Verkuil <hverkuil(a)xs4all.nl> date: Mon Nov 16 19:16:00 2009 +0100 summary: go7007: subdev conversion changeset: 13459:28dc36f9dfdc parent: 13454:0ba581813fde parent: 13458:2ca6a9dc112b user: Mauro Carvalho Chehab <mchehab(a)redhat.com> date: Tue Nov 24 14:45:53 2009 -0200 summary: merge: http://www.linuxtv.org/hg/~hverkuil/v4l-dvb-staging Unfortunately, the original tree were already removed. I suspect that hans didn't merge that patch on the latest version of v4l-dvb-stating tree. > > > - s2250 mutex patch This one were merged on my tree: http://linuxtv.org/hg/v4l-dvb/rev/8bff7e6c44d4 However, I suspect that just applying it won't solve the compilation breakage. > > > > Good catch. Mauro, can you please add these two as well for 2.6.32? I'm sure > > I marked these two as high-prio patches. This staging driver is actively > > being used and developed so this regression should be fixed. > > There were a merge problem when applying those patches at -git. As this is a > staging driver, I opted to just move all patches to linux-next, as there were > no mention on the patch subject mentioned that they are, in fact, fixes. > > I'll see what can we do to solve the s2250 compilation breakage. > > As Linus didn't merge from this pull request yet, I'll work on a fix, based > on one of the Stefan suggestions, preferably backporting the patches. > Cheers, Mauro -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Hans Verkuil on 30 Nov 2009 12:50 > Em Mon, 30 Nov 2009 12:03:32 -0200 > Mauro Carvalho Chehab <mchehab(a)redhat.com> escreveu: > >> Em Sat, 28 Nov 2009 03:57:58 +0100 >> Hans Verkuil <hverkuil(a)xs4all.nl> escreveu: >> >> > On Friday 27 November 2009 22:40:01 Stefan Lippers-Hollmann wrote: >> > > Hi >> > > >> > > On Friday 27 November 2009, Mauro Carvalho Chehab wrote: >> > > > Hi Linus, >> > > > >> > > > Please pull from: >> > > > ssh://master.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6.git >> for_linus >> > > > >> > > > For the following drivers and building fixes: >> > > > >> > > > - radio-gemtek-pci: fix double mutex_lock >> > > > - v4l: add more missing linux/sched.h includes >> > > > - soc-camera: properly initialise the device object when >> reusing >> > > > - soc-camera: sh_mobile_ceu_camera: call pm_runtime_disable >> > > > - em28xx: fix Reddo DVB-C USB TV Box GPIO >> > > > - davinci: remove stray duplicate config pointer >> > > > - SMS_SIANO_MDTV should depend on HAS_DMA >> > > > - cxusb: Fix hang on DViCO FusionHDTV DVB-T Dual Digital 4 (rev >> 1) >> > > > - sh_mobile_ceu_camera: fix compile warning >> > > > - Fix wrong parameter order in memset >> > > [...] >> > > >> > > Please consider cherry picking the following two patches from Hans >> > > Verkuil [1]: >> > > - add the missing s2250-loader.h > > I'm not finding the above patch. I remember I got one patch at the go7007 > conversion series to V4L2 that were removing this patch that didn't exist > upstream. I had to manually fix the patch to apply at my linux-next tree. > > The patches I actually merged from hans v4l-dvb-staging (after the tree > rebase) > were: > > changeset: 13455:0eb2e476ac2a > parent: 13372:8bff7e6c44d4 > user: Hans Verkuil <hverkuil(a)xs4all.nl> > date: Mon Nov 16 19:07:42 2009 +0100 > summary: go7007: Add struct v4l2_device. > > changeset: 13456:f6f0c5647657 > user: Hans Verkuil <hverkuil(a)xs4all.nl> > date: Mon Nov 16 19:13:51 2009 +0100 > summary: s2250: Change module structure > > changeset: 13457:701aead61759 > user: Hans Verkuil <hverkuil(a)xs4all.nl> > date: Mon Nov 16 19:15:07 2009 +0100 > summary: s2250: subdev conversion > > changeset: 13458:2ca6a9dc112b > user: Hans Verkuil <hverkuil(a)xs4all.nl> > date: Mon Nov 16 19:16:00 2009 +0100 > summary: go7007: subdev conversion > > changeset: 13459:28dc36f9dfdc > parent: 13454:0ba581813fde > parent: 13458:2ca6a9dc112b > user: Mauro Carvalho Chehab <mchehab(a)redhat.com> > date: Tue Nov 24 14:45:53 2009 -0200 > summary: merge: http://www.linuxtv.org/hg/~hverkuil/v4l-dvb-staging > > Unfortunately, the original tree were already removed. > > I suspect that hans didn't merge that patch on the latest version of > v4l-dvb-stating tree. > >> > > - s2250 mutex patch > > This one were merged on my tree: > http://linuxtv.org/hg/v4l-dvb/rev/8bff7e6c44d4 > > However, I suspect that just applying it won't solve the compilation > breakage. There wasn't a patch in my tree for the header because it is only missing in the git tree. The hg repo does have it. So that header should be copied from our master v4l-dvb tree to Linus' git tree. Together with the mutex patch this driver should compile again. Regards, Hans >> > >> > Good catch. Mauro, can you please add these two as well for 2.6.32? >> I'm sure >> > I marked these two as high-prio patches. This staging driver is >> actively >> > being used and developed so this regression should be fixed. >> >> There were a merge problem when applying those patches at -git. As this >> is a >> staging driver, I opted to just move all patches to linux-next, as there >> were >> no mention on the patch subject mentioned that they are, in fact, fixes. >> >> I'll see what can we do to solve the s2250 compilation breakage. >> >> As Linus didn't merge from this pull request yet, I'll work on a fix, >> based >> on one of the Stefan suggestions, preferably backporting the patches. >> > > > > > Cheers, > Mauro > -- Hans Verkuil - video4linux developer - sponsored by TANDBERG Telecom -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Mike Isely on 30 Nov 2009 15:00 On Mon, 30 Nov 2009, Mauro Carvalho Chehab wrote: > Em Sat, 28 Nov 2009 14:33:30 -0600 (CST) > Mike Isely <isely(a)isely.net> escreveu: > > > > > Mauro: > > > > I had also posted up two high priority pvrusb2 patches that should > > really be cherry-picked for 2.6.32. You've already pulled them into > > v4l/dvb and I did mark them as high priority at the time. > > > > These patches enable use of FX2 microcontroller firmware that is 16KB in > > size. Hauppauge is no longer shipping 8KB firmware for HVR-1950 and > > HVR-1900 and without these changes then those devices won't work AT ALL > > in kernel 2.6.32. > > > > You can find these within the v4l-dvb Mercurial repository here: > > > > Changeset 13495:87c3853fe2b3 > > Subject: pvrusb2: Support 16KB FX2 firmware > > http://linuxtv.org/hg/v4l-dvb/rev/87c3853fe2b3 > > > > Changeset 13500:d4c418d4b25c > > Subject: pvrusb2: Fix lingering 16KB FX2 Firmware issues > > http://linuxtv.org/hg/v4l-dvb/rev/d4c418d4b25c > > > > I do not believe these patches have any ordering dependencies with other > > patches, though between the two the second one technically should come > > after the first. > > There are. Picking just those patches broke compilation. Mauro: Please forward to me the compilation errors. Right now I am just not seeing how a patch this trivial could have any compilation dependencies. And unfortunately I will not be able to reproduce your build setup until at least Tuesday night. I must be blind. > > Also, it seemed too late for adding support for newer boards/firmware when Linus > is about to release a kernel. This is not a new feature. It's a bug fix due to something that Hauppauge recently did. Hauppauge is NO LONGER officially distributing FX2 firmware with their hardware which the driver can use. This simply prevents any new HVR-1950 / HVR-1900 users from working under Linux. This fixes breakage for previously working hardware. The root cause is simple - the firmware blob is just larger now - and the fix is trivial. It absolutely needs to go in. In fact, this should go back to a 2.6.31.x and a 2.6.27.x release as well, though in those cases I have to figure out if driver source code is still close enough for the same patches to still work. I am sorry this is showing up late for you. There are multiple reasons for this. However I did mark these patches as "high priority", following your v4l-dvb changeset process. I did comment on the pull request that these were important but I guess I needed to also specifically call these out in the pull request text as well. If these don't get in now as part of the official 2.6.32 release, these absolutely need to be queued for 2.6.32.1. -Mike -- Mike Isely isely @ isely (dot) net PGP: 03 54 43 4D 75 E5 CC 92 71 16 01 E2 B5 F5 C1 E8 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Mauro Carvalho Chehab on 30 Nov 2009 15:10 Hi Linus, Em 27-11-2009 19:40, Stefan Lippers-Hollmann escreveu: > Please consider cherry picking the following two patches from Hans > Verkuil [1]: > - add the missing s2250-loader.h > - s2250 mutex patch > > or revert the patch which broke compiling go7007-usb (imho it would be > better to re-add the 2 externs [2], than throwing away the needed mutex > conversion, but it would be an option) > Commit: fd9a40da1db372833e1af6397d2f6c94ceff3dad > V4L/DVB (12859): go7007: semaphore -> mutex conversion > > or mark CONFIG_VIDEO_GO7007_USB_S2250_BOARD as BROKEN, as it fails to > compile since the afforementioned patch was merged on september 19th (this > is a 2.6.31 --> 2.6.32 regression) [2,3]. Please pull from: ssh://master.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6.git for_linus It is basically the same pull request as before with two more patches added, in order to fix the staging/g07007 compilation/work. The two new patches are: staging/go7007: Fix compilation by re-adding the missing s2250-loader.h staging/go7007: fix mutex function usage for s2250 (I replaced the subjects for a more adequate one, to explicitly say that those are fixes) Thanks! Mauro --- MAINTAINERS | 7 +++++++ drivers/media/common/ir-functions.c | 2 +- drivers/media/dvb/dvb-usb/cxusb.c | 10 +++++++++- drivers/media/dvb/siano/Kconfig | 2 +- drivers/media/radio/radio-gemtek-pci.c | 2 -- drivers/media/video/davinci/vpif_display.c | 1 - drivers/media/video/em28xx/em28xx-cards.c | 2 +- drivers/media/video/mx1_camera.c | 1 + drivers/media/video/mx3_camera.c | 1 + drivers/media/video/sh_mobile_ceu_camera.c | 7 +++++-- drivers/media/video/soc_camera.c | 17 ++++++++++++----- drivers/media/video/videobuf-dma-contig.c | 1 + drivers/staging/go7007/s2250-board.c | 4 ++-- drivers/staging/go7007/s2250-loader.h | 24 ++++++++++++++++++++++++ 14 files changed, 65 insertions(+), 16 deletions(-) create mode 100644 drivers/staging/go7007/s2250-loader.h Alan Cox (1): V4L/DVB (13530): Fix wrong parameter order in memset Antti Palosaari (1): V4L/DVB (13366): em28xx: fix Reddo DVB-C USB TV Box GPIO Geert Uytterhoeven (1): V4L/DVB (13412): SMS_SIANO_MDTV should depend on HAS_DMA Guennadi Liakhovetski (2): V4L/DVB (13343): v4l: add more missing linux/sched.h includes V4L/DVB (13344): soc-camera: properly initialise the device object when reusing Hans Verkuil (3): V4L/DVB (13321): radio-gemtek-pci: fix double mutex_lock V4L/DVB (13371): davinci: remove stray duplicate config pointer V4L/DVB (13481): sh_mobile_ceu_camera: fix compile warning Kuninori Morimoto (1): V4L/DVB (13345): soc-camera: sh_mobile_ceu_camera: call pm_runtime_disable Mauro Carvalho Chehab (1): staging/go7007: Fix compilation by re-adding the missing s2250-loader.h Olivier Lorin (1): V4L/DVB (13372a): MAINTAINERS: addition of gspca_gl860 driver Pete Eberlein (1): V4L/DVB (13372): staging/go7007: fix mutex function usage for s2250 Robert Lowery (1): V4L/DVB (13436): cxusb: Fix hang on DViCO FusionHDTV DVB-T Dual Digital 4 (rev 1) --------------------------------------------------- V4L/DVB development is hosted at http://linuxtv.org -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Mauro Carvalho Chehab on 1 Dec 2009 14:30
Mike Isely wrote: > On Mon, 30 Nov 2009, Mauro Carvalho Chehab wrote: > >> Em Sat, 28 Nov 2009 14:33:30 -0600 (CST) >> Mike Isely <isely(a)isely.net> escreveu: >> >>> Mauro: >>> >>> I had also posted up two high priority pvrusb2 patches that should >>> really be cherry-picked for 2.6.32. You've already pulled them into >>> v4l/dvb and I did mark them as high priority at the time. >>> >>> These patches enable use of FX2 microcontroller firmware that is 16KB in >>> size. Hauppauge is no longer shipping 8KB firmware for HVR-1950 and >>> HVR-1900 and without these changes then those devices won't work AT ALL >>> in kernel 2.6.32. >>> >>> You can find these within the v4l-dvb Mercurial repository here: >>> >>> Changeset 13495:87c3853fe2b3 >>> Subject: pvrusb2: Support 16KB FX2 firmware >>> http://linuxtv.org/hg/v4l-dvb/rev/87c3853fe2b3 >>> >>> Changeset 13500:d4c418d4b25c >>> Subject: pvrusb2: Fix lingering 16KB FX2 Firmware issues >>> http://linuxtv.org/hg/v4l-dvb/rev/d4c418d4b25c >>> >>> I do not believe these patches have any ordering dependencies with other >>> patches, though between the two the second one technically should come >>> after the first. > > > >> There are. Picking just those patches broke compilation. > > Mauro: > > Please forward to me the compilation errors. Right now I am just not > seeing how a patch this trivial could have any compilation dependencies. > And unfortunately I will not be able to reproduce your build setup until > at least Tuesday night. I must be blind. > > >> Also, it seemed too late for adding support for newer boards/firmware when Linus >> is about to release a kernel. > > This is not a new feature. It's a bug fix due to something that > Hauppauge recently did. Hauppauge is NO LONGER officially distributing > FX2 firmware with their hardware which the driver can use. This simply > prevents any new HVR-1950 / HVR-1900 users from working under Linux. > This fixes breakage for previously working hardware. The root cause is > simple - the firmware blob is just larger now - and the fix is trivial. > It absolutely needs to go in. In fact, this should go back to a > 2.6.31.x and a 2.6.27.x release as well, though in those cases I have to > figure out if driver source code is still close enough for the same > patches to still work. > > I am sorry this is showing up late for you. There are multiple reasons > for this. However I did mark these patches as "high priority", > following your v4l-dvb changeset process. I did comment on the pull > request that these were important but I guess I needed to also > specifically call these out in the pull request text as well. > > If these don't get in now as part of the official 2.6.32 release, these > absolutely need to be queued for 2.6.32.1. We are very late for 2.6.32. I'm not sure if are there still time for it. I'll seek for some time during this week to add those patches at the upstream tree and removing them from the development tree and see what compilation issues arise. Cheers, Mauro. > > -Mike > > -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |