Re: [yocto] [PATCH][meta-intel] grub: remove the dependency on freetype, disable grub-mkfont
Seeing this change reminded me that we really should move grub_1.98 to
oe-core, is there any objection to this? Any reason this should not be
part of oe-core to support general x86_64 machines?
Sau!
On 08/31/2011 07:34 PM, Darren Hart wrote:
Grub's configure task was picking up the host freetype libraries if
bitbake hadn't gotten around building freetype yet. We could add a
dependency on freetype, but it's only used for the optional grub-mkfont
utility which we don't really need.
Disable grub-mkfont via EXTRA_OECONF, removing the dependency on freetype.
Testing: core-image-minimal build and install on sugarbay.
Signed-off-by: Darren Hart
Tested-by: Tom Zanussi
---
common/recipes-bsp/grub/grub_1.98.bb |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/common/recipes-bsp/grub/grub_1.98.bb
b/common/recipes-bsp/grub/grub_1.98.bb
index 0cd3a40..802733f 100644
--- a/common/recipes-bsp/grub/grub_1.98.bb
+++ b/common/recipes-bsp/grub/grub_1.98.bb
@@ -24,7 +24,7 @@ SRC_URI = "ftp://alpha.gnu.org/gnu/grub/grub-${PV}.tar.gz \
inherit autotools
inherit gettext
-EXTRA_OECONF = "--with-platform=pc --target=i386"
+EXTRA_OECONF = "--with-platform=pc --target=i386 --disable-grub-mkfont"
do_configure() {
oe_runconf
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto
Re: [yocto] e100 driver does not build
Yes, the problem was due to missing dependent options.. I had incorrectly assumed that dependent options would get pulled in or perhaps a error would be generated. It worked after I added following 3 options stated as dependent options in a Kernel configuration database that I saw: CONFIG_ISA CONFIG_EISA CONFIG_NET_PCI I am still iterating to check if all three of the above are really needed or or if just the CONFIG_NET_PCI option is needed. I had enabled the option using a .cfg file. Also checking again to see if a warning gets generated when the dependent options are not included. BTW what does a message such as following in the .config file mean ? "CONFIG_x is not set" Thanks Rahul -Original Message- From: Bruce Ashfield [mailto:bruce.ashfi...@windriver.com] Sent: Wednesday, August 31, 2011 6:27 PM To: Zanussi, Tom Cc: Saxena, Rahul; yocto@yoctoproject.org Subject: Re: [yocto] e100 driver does not build On 11-08-31 7:18 PM, Tom Zanussi wrote: > > On Wed, 2011-08-31 at 16:04 -0700, Saxena, Rahul wrote: >> I am unable to get the e100 driver to build for my BSP. >> >> I have configured to build it as a module. I do not see the e100.ko >> file being generated nor does >> >> CONFIG_E100 appear in the generated .config file in my build area. >> >> >> >> However I do see the driver source file e100.c in >>tmp/work/bspname-poky-linux/linux-yocto-3.0.3+git1 >> + ../linux/drivers/net >> >> >> >> I do not have any problems with other drivers (igb, E1000 etc) that I >> have configured in exactly similar way. >> >> >> >> Any insight on this issue ? >> > > It's probably missing another option it depends on e.g. is > CONFIG_NET_PCI and CONFIG_MII enabled in your .config? This is most likely what it is. Out of curiosity, how did you enable the option ? Via a config fragement (.cfg file) or via menuconfig .. or something else? If you tried to set something via a fragment and it doesn't make it to the final .config, you should see a warning during the kernel build phase. Cheers, Bruce > > Tom > >> >> >> Thanks >> >> Rahul >> >> > > > ___ > yocto mailing list > yocto@yoctoproject.org > https://lists.yoctoproject.org/listinfo/yocto ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
Re: [yocto] e100 driver does not build
On 11-09-01 2:11 PM, Saxena, Rahul wrote: Yes, the problem was due to missing dependent options.. I had incorrectly assumed that dependent options would get pulled in or perhaps a error would be generated. A warning can (and will) be generated for options that are set by a BSP and don't end up in a final .config, but that's only if you've placed them in a .cfg file and fed them through the config auditing steps of the linux-yocto build (this is automatic). dependencies won't be set automatically, it's the opposite, you must ensure they are set or a particular Kconfig value must select them. It worked after I added following 3 options stated as dependent options in a Kernel configuration database that I saw: CONFIG_ISA CONFIG_EISA CONFIG_NET_PCI I am still iterating to check if all three of the above are really needed or or if just the CONFIG_NET_PCI option is needed. I had enabled the option using a .cfg file. Also checking again to see if a warning gets generated when the dependent options are not included. BTW what does a message such as following in the .config file mean ? "CONFIG_x is not set" It should be: # CONFIG_foo is not set And that is should be read as "config foo is disabled" Bruce Thanks Rahul -Original Message- From: Bruce Ashfield [mailto:bruce.ashfi...@windriver.com] Sent: Wednesday, August 31, 2011 6:27 PM To: Zanussi, Tom Cc: Saxena, Rahul; yocto@yoctoproject.org Subject: Re: [yocto] e100 driver does not build On 11-08-31 7:18 PM, Tom Zanussi wrote: On Wed, 2011-08-31 at 16:04 -0700, Saxena, Rahul wrote: I am unable to get the e100 driver to build for my BSP. I have configured to build it as a module. I do not see the e100.ko file being generated nor does CONFIG_E100 appear in the generated .config file in my build area. However I do see the driver source file e100.c in tmp/work/bspname-poky-linux/linux-yocto-3.0.3+git1 + ../linux/drivers/net I do not have any problems with other drivers (igb, E1000 etc) that I have configured in exactly similar way. Any insight on this issue ? It's probably missing another option it depends on e.g. is CONFIG_NET_PCI and CONFIG_MII enabled in your .config? This is most likely what it is. Out of curiosity, how did you enable the option ? Via a config fragement (.cfg file) or via menuconfig .. or something else? If you tried to set something via a fragment and it doesn't make it to the final .config, you should see a warning during the kernel build phase. Cheers, Bruce Tom Thanks Rahul ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
Re: [yocto] e100 driver does not build
On Thu, 2011-09-01 at 11:11 -0700, Saxena, Rahul wrote: > Yes, the problem was due to missing dependent options.. > I had incorrectly assumed that dependent options would get pulled in > or perhaps a error would be generated. > It worked after I added following 3 options stated as dependent options > in a Kernel configuration database that I saw: > > CONFIG_ISA > CONFIG_EISA > CONFIG_NET_PCI > > I am still iterating to check if all three of the above are really needed or > or if just the CONFIG_NET_PCI option is needed. > > I had enabled the option using a .cfg file. Also checking again to see if a > warning > gets generated when the dependent options are not included. > Note that it's also part of the intel-1 feature, which is basically there for reusability - they were all together in a bunch of BSPs so I moved them into a feature. I'm wondering whether we want to break them up or at least create a separate e100 feature. > BTW what does a message such as following in the .config file mean ? > > "CONFIG_x is not set" > It means that config option was never set or got turned off because of unmet depencies. Tom > > Thanks > Rahul > > -Original Message- > From: Bruce Ashfield [mailto:bruce.ashfi...@windriver.com] > Sent: Wednesday, August 31, 2011 6:27 PM > To: Zanussi, Tom > Cc: Saxena, Rahul; yocto@yoctoproject.org > Subject: Re: [yocto] e100 driver does not build > > On 11-08-31 7:18 PM, Tom Zanussi wrote: > > > > On Wed, 2011-08-31 at 16:04 -0700, Saxena, Rahul wrote: > >> I am unable to get the e100 driver to build for my BSP. > >> > >> I have configured to build it as a module. I do not see the e100.ko > >> file being generated nor does > >> > >> CONFIG_E100 appear in the generated .config file in my build area. > >> > >> > >> > >> However I do see the driver source file e100.c in > >>tmp/work/bspname-poky-linux/linux-yocto-3.0.3+git1 > >> + ../linux/drivers/net > >> > >> > >> > >> I do not have any problems with other drivers (igb, E1000 etc) that I > >> have configured in exactly similar way. > >> > >> > >> > >> Any insight on this issue ? > >> > > > > It's probably missing another option it depends on e.g. is > > CONFIG_NET_PCI and CONFIG_MII enabled in your .config? > > This is most likely what it is. > > Out of curiosity, how did you enable the option ? Via a config fragement > (.cfg file) or via menuconfig .. or something else? > > If you tried to set something via a fragment and it doesn't make it to > the final .config, you should see a warning during the kernel build > phase. > > Cheers, > > Bruce > > > > > Tom > > > >> > >> > >> Thanks > >> > >> Rahul > >> > >> > > > > > > ___ > > yocto mailing list > > yocto@yoctoproject.org > > https://lists.yoctoproject.org/listinfo/yocto > > ___ > yocto mailing list > yocto@yoctoproject.org > https://lists.yoctoproject.org/listinfo/yocto ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
Re: [yocto] RT kernel with core-image-sato build failed for sugarbay
On 08/31/2011 04:49 PM, Bodke, Kishore K wrote: Hi, I am trying to build core-image-sato with rt kernel for sugarbay and it fails for the do_rootfs . Attached is the build log. Could you please let me know what went wrong? Did you save a log of the build itself by any chance? It seems strange that you would have this dependencies are not being met. Sau! Thanks Kishore. ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
Re: [yocto] e100 driver does not build
Yes, I had noticed that it is part of intel-1 feature also. I tend to think that it is better to separate out CONFIG_E100 option from intel-1 I think that platforms that want 1G/10G enabled would not need 10/100 driver >> It means that config option was never set or got turned off because of unmet depencies << However I am not sure if this works in all cases as I did not see "# CONFIG_E100 is not set" even when its dependent configurations were not satisfied Rahul -Original Message- From: Zanussi, Tom Sent: Thursday, September 01, 2011 11:22 AM To: Saxena, Rahul Cc: yocto@yoctoproject.org Subject: Re: [yocto] e100 driver does not build On Thu, 2011-09-01 at 11:11 -0700, Saxena, Rahul wrote: > Yes, the problem was due to missing dependent options.. > I had incorrectly assumed that dependent options would get pulled in > or perhaps a error would be generated. > It worked after I added following 3 options stated as dependent options > in a Kernel configuration database that I saw: > > CONFIG_ISA > CONFIG_EISA > CONFIG_NET_PCI > > I am still iterating to check if all three of the above are really needed or > or if just the CONFIG_NET_PCI option is needed. > > I had enabled the option using a .cfg file. Also checking again to see if a > warning > gets generated when the dependent options are not included. > Note that it's also part of the intel-1 feature, which is basically there for reusability - they were all together in a bunch of BSPs so I moved them into a feature. I'm wondering whether we want to break them up or at least create a separate e100 feature. > BTW what does a message such as following in the .config file mean ? > > "CONFIG_x is not set" > It means that config option was never set or got turned off because of unmet depencies. Tom > > Thanks > Rahul > > -Original Message- > From: Bruce Ashfield [mailto:bruce.ashfi...@windriver.com] > Sent: Wednesday, August 31, 2011 6:27 PM > To: Zanussi, Tom > Cc: Saxena, Rahul; yocto@yoctoproject.org > Subject: Re: [yocto] e100 driver does not build > > On 11-08-31 7:18 PM, Tom Zanussi wrote: > > > > On Wed, 2011-08-31 at 16:04 -0700, Saxena, Rahul wrote: > >> I am unable to get the e100 driver to build for my BSP. > >> > >> I have configured to build it as a module. I do not see the e100.ko > >> file being generated nor does > >> > >> CONFIG_E100 appear in the generated .config file in my build area. > >> > >> > >> > >> However I do see the driver source file e100.c in > >>tmp/work/bspname-poky-linux/linux-yocto-3.0.3+git1 > >> + ../linux/drivers/net > >> > >> > >> > >> I do not have any problems with other drivers (igb, E1000 etc) that I > >> have configured in exactly similar way. > >> > >> > >> > >> Any insight on this issue ? > >> > > > > It's probably missing another option it depends on e.g. is > > CONFIG_NET_PCI and CONFIG_MII enabled in your .config? > > This is most likely what it is. > > Out of curiosity, how did you enable the option ? Via a config fragement > (.cfg file) or via menuconfig .. or something else? > > If you tried to set something via a fragment and it doesn't make it to > the final .config, you should see a warning during the kernel build > phase. > > Cheers, > > Bruce > > > > > Tom > > > >> > >> > >> Thanks > >> > >> Rahul > >> > >> > > > > > > ___ > > yocto mailing list > > yocto@yoctoproject.org > > https://lists.yoctoproject.org/listinfo/yocto > > ___ > yocto mailing list > yocto@yoctoproject.org > https://lists.yoctoproject.org/listinfo/yocto ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
Re: [yocto] e100 driver does not build
On Thu, 2011-09-01 at 11:34 -0700, Saxena, Rahul wrote: > Yes, I had noticed that it is part of intel-1 feature also. > > I tend to think that it is better to separate out CONFIG_E100 option from > intel-1 > I think that platforms that want 1G/10G enabled would not need 10/100 driver > > Yeah, makes sense. I can do that if you want, or you can - either way is fine with me... Tom > >> It means that config option was never set or got turned off because of > unmet depencies << > > However I am not sure if this works in all cases as I did not see > "# CONFIG_E100 is not set" even when its dependent configurations were not > satisfied > > Rahul > > -Original Message- > From: Zanussi, Tom > Sent: Thursday, September 01, 2011 11:22 AM > To: Saxena, Rahul > Cc: yocto@yoctoproject.org > Subject: Re: [yocto] e100 driver does not build > > On Thu, 2011-09-01 at 11:11 -0700, Saxena, Rahul wrote: > > Yes, the problem was due to missing dependent options.. > > I had incorrectly assumed that dependent options would get pulled in > > or perhaps a error would be generated. > > It worked after I added following 3 options stated as dependent options > > in a Kernel configuration database that I saw: > > > > CONFIG_ISA > > CONFIG_EISA > > CONFIG_NET_PCI > > > > I am still iterating to check if all three of the above are really needed or > > or if just the CONFIG_NET_PCI option is needed. > > > > I had enabled the option using a .cfg file. Also checking again to see if > > a warning > > gets generated when the dependent options are not included. > > > > Note that it's also part of the intel-1 feature, which is basically > there for reusability - they were all together in a bunch of BSPs so I > moved them into a feature. I'm wondering whether we want to break them > up or at least create a separate e100 feature. > > > BTW what does a message such as following in the .config file mean ? > > > > "CONFIG_x is not set" > > > > It means that config option was never set or got turned off because of > unmet depencies. > > Tom > > > > > Thanks > > Rahul > > > > -Original Message- > > From: Bruce Ashfield [mailto:bruce.ashfi...@windriver.com] > > Sent: Wednesday, August 31, 2011 6:27 PM > > To: Zanussi, Tom > > Cc: Saxena, Rahul; yocto@yoctoproject.org > > Subject: Re: [yocto] e100 driver does not build > > > > On 11-08-31 7:18 PM, Tom Zanussi wrote: > > > > > > On Wed, 2011-08-31 at 16:04 -0700, Saxena, Rahul wrote: > > >> I am unable to get the e100 driver to build for my BSP. > > >> > > >> I have configured to build it as a module. I do not see the e100.ko > > >> file being generated nor does > > >> > > >> CONFIG_E100 appear in the generated .config file in my build area. > > >> > > >> > > >> > > >> However I do see the driver source file e100.c in > > >>tmp/work/bspname-poky-linux/linux-yocto-3.0.3+git1 > > >> + ../linux/drivers/net > > >> > > >> > > >> > > >> I do not have any problems with other drivers (igb, E1000 etc) that I > > >> have configured in exactly similar way. > > >> > > >> > > >> > > >> Any insight on this issue ? > > >> > > > > > > It's probably missing another option it depends on e.g. is > > > CONFIG_NET_PCI and CONFIG_MII enabled in your .config? > > > > This is most likely what it is. > > > > Out of curiosity, how did you enable the option ? Via a config fragement > > (.cfg file) or via menuconfig .. or something else? > > > > If you tried to set something via a fragment and it doesn't make it to > > the final .config, you should see a warning during the kernel build > > phase. > > > > Cheers, > > > > Bruce > > > > > > > > Tom > > > > > >> > > >> > > >> Thanks > > >> > > >> Rahul > > >> > > >> > > > > > > > > > ___ > > > yocto mailing list > > > yocto@yoctoproject.org > > > https://lists.yoctoproject.org/listinfo/yocto > > > > ___ > > yocto mailing list > > yocto@yoctoproject.org > > https://lists.yoctoproject.org/listinfo/yocto > > ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
[yocto] [PATCH 0/2] [KERNEL] Cover letter for the delete duplicate statement and delete unused cfg fragment patches
Cover letter for the delete duplicate statement and delete unused cfg fragment patches Rahul Saxena (2): Removed duplicate statement Removed unused cfg file meta/cfg/kernel-cache/bsp/fishriver/eg20t.cfg | 39 - meta/cfg/kernel-cache/bsp/fishriver/fishriver.scc |1 - 2 files changed, 0 insertions(+), 40 deletions(-) delete mode 100644 meta/cfg/kernel-cache/bsp/fishriver/eg20t.cfg ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
[yocto] [PATCH 1/2] [KERNEL] Removed duplicate statement
Removed duplicate statement --- meta/cfg/kernel-cache/bsp/fishriver/fishriver.scc |1 - 1 files changed, 0 insertions(+), 1 deletions(-) diff --git a/meta/cfg/kernel-cache/bsp/fishriver/fishriver.scc b/meta/cfg/kernel-cache/bsp/fishriver/fishriver.scc index 0abde0d..6e3d385 100644 --- a/meta/cfg/kernel-cache/bsp/fishriver/fishriver.scc +++ b/meta/cfg/kernel-cache/bsp/fishriver/fishriver.scc @@ -1,5 +1,4 @@ kconf hardware fishriver.cfg -kconf hardware fishriver.cfg git merge yocto/emgd -- 1.7.0.4 ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
[yocto] [PATCH 2/2] [KERNEL] Removed unused cfg file
Removed unused cfg file --- meta/cfg/kernel-cache/bsp/fishriver/eg20t.cfg | 39 - 1 files changed, 0 insertions(+), 39 deletions(-) delete mode 100644 meta/cfg/kernel-cache/bsp/fishriver/eg20t.cfg diff --git a/meta/cfg/kernel-cache/bsp/fishriver/eg20t.cfg b/meta/cfg/kernel-cache/bsp/fishriver/eg20t.cfg deleted file mode 100644 index 225581c..000 --- a/meta/cfg/kernel-cache/bsp/fishriver/eg20t.cfg +++ /dev/null @@ -1,39 +0,0 @@ -# Hardware support for the Platform Controller Hub EG20T - -CONFIG_PCH_DMA=y -CONFIG_PCH_UART_DMA=y -CONFIG_SERIAL_8250_PCI=y -CONFIG_SERIAL_8250_PCH=y -CONFIG_SERIAL_8250_PCH_DMA=y - -CONFIG_SERIAL_PCH_UART=y -CONFIG_SERIAL_CORE=y - -CONFIG_SPI=y -CONFIG_SPI_BITBANG=y -CONFIG_SPI_GPIO=y -CONFIG_SPI_TOPCLIFF_PCH=y - -CONFIG_MISC_DEVICES=y -CONFIG_PCH_PHUB=y - -CONFIG_NETDEVICES=y -CONFIG_NETDEV_1000=y -CONFIG_PCH_GBE=y - -CONFIG_PCH_USBDEV=y -CONFIG_USB_GADGET=y -CONFIG_USB_GADGET_EG20T=y -CONFIG_USB_GADGET_DUALSPEED=y -CONFIG_USB_GADGET_SELECTED=y - -CONFIG_I2C_EG20T=y - -CONFIG_GPIOLIB=y -CONFIG_GPIO_PCH=y - -CONFIG_PCH_IEEE1588=y - -CONFIG_CAN=y -CONFIG_CAN_DEV=y -CONFIG_PCH_CAN=y -- 1.7.0.4 ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
Re: [yocto] [PATCH][meta-intel] grub: remove the dependency on freetype, disable grub-mkfont
On 08/31/2011 07:34 PM, Darren Hart wrote:
> Grub's configure task was picking up the host freetype libraries if
> bitbake hadn't gotten around building freetype yet. We could add a
> dependency on freetype, but it's only used for the optional grub-mkfont
> utility which we don't really need.
>
> Disable grub-mkfont via EXTRA_OECONF, removing the dependency on freetype.
>
> Testing: core-image-minimal build and install on sugarbay.
>
> Signed-off-by: Darren Hart
> Tested-by: Tom Zanussi
> ---
> common/recipes-bsp/grub/grub_1.98.bb |2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/common/recipes-bsp/grub/grub_1.98.bb
> b/common/recipes-bsp/grub/grub_1.98.bb
> index 0cd3a40..802733f 100644
> --- a/common/recipes-bsp/grub/grub_1.98.bb
> +++ b/common/recipes-bsp/grub/grub_1.98.bb
> @@ -24,7 +24,7 @@ SRC_URI = "ftp://alpha.gnu.org/gnu/grub/grub-${PV}.tar.gz \
> inherit autotools
> inherit gettext
>
> -EXTRA_OECONF = "--with-platform=pc --target=i386"
> +EXTRA_OECONF = "--with-platform=pc --target=i386 --disable-grub-mkfont"
>
> do_configure() {
> oe_runconf
Merged to meta-intel:master.
--
Darren Hart
Intel Open Source Technology Center
Yocto Project - Linux Kernel
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto
Re: [yocto] RT kernel with core-image-sato build failed for sugarbay
On 08/31/2011 09:19 PM, Tom Zanussi wrote: > > On Wed, 2011-08-31 at 16:49 -0700, Bodke, Kishore K wrote: >> Hi, >> >> >> >> I am trying to build core-image-sato with rt kernel for sugarbay and >> it fails for the do_rootfs . >> > > FWIW, I was able to build and boot core-image-sato with the rt kernel: > > Linux sugarbay 3.0.3-rt8-yocto-preempt-rt+ #1 SMP PREEMPT RT Wed Aug 31 > 21:33:27 CDT 2011 x86_64 GNU/Linux > > using the following commits: > > poky/master 705ec50edda7f7fb8c132238fe3f1477384a470e > meta/intel 902d2f3c919cd1d870dc8d35a7521cbba42e53ea > > Actually, meta/intel was that commit with Darren's grub patch on top > (meta-intel/dvhart/grub-freetype). > Kishore, would you try with a clean build and see if this works for you? I'm going to try and build first without -rt and then with rt and see if I can reproduce the issue. Thanks, Darren > Tom > >> >> >> Attached is the build log. >> >> >> >> Could you please let me know what went wrong? >> >> >> >> Thanks >> >> Kishore. >> >> > > -- Darren Hart Intel Open Source Technology Center Yocto Project - Linux Kernel ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
Re: [yocto] RT kernel with core-image-sato build failed for sugarbay
Clean build with the below TOM's two commit IDs worked for me, if I do not add IMAGE_INSTALL += "rttests" for sugarbay. It failed if I add the IMAGE_INSTALL += "rttests" Thanks Kishore. -Original Message- From: Darren Hart [mailto:dvh...@linux.intel.com] Sent: Thursday, September 01, 2011 1:57 PM To: Zanussi, Tom Cc: Bodke, Kishore K; yocto@yoctoproject.org; Hart, Darren; Saxena, Rahul Subject: Re: RT kernel with core-image-sato build failed for sugarbay On 08/31/2011 09:19 PM, Tom Zanussi wrote: > > On Wed, 2011-08-31 at 16:49 -0700, Bodke, Kishore K wrote: >> Hi, >> >> >> >> I am trying to build core-image-sato with rt kernel for sugarbay and >> it fails for the do_rootfs . >> > > FWIW, I was able to build and boot core-image-sato with the rt kernel: > > Linux sugarbay 3.0.3-rt8-yocto-preempt-rt+ #1 SMP PREEMPT RT Wed Aug 31 > 21:33:27 CDT 2011 x86_64 GNU/Linux > > using the following commits: > > poky/master 705ec50edda7f7fb8c132238fe3f1477384a470e > meta/intel 902d2f3c919cd1d870dc8d35a7521cbba42e53ea > > Actually, meta/intel was that commit with Darren's grub patch on top > (meta-intel/dvhart/grub-freetype). > Kishore, would you try with a clean build and see if this works for you? I'm going to try and build first without -rt and then with rt and see if I can reproduce the issue. Thanks, Darren > Tom > >> >> >> Attached is the build log. >> >> >> >> Could you please let me know what went wrong? >> >> >> >> Thanks >> >> Kishore. >> >> > > -- Darren Hart Intel Open Source Technology Center Yocto Project - Linux Kernel ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
Re: [yocto] [PATCH 0/2] [KERNEL] Cover letter for the delete duplicate statement and delete unused cfg fragment patches
On 11-09-01 4:41 PM, Saxena, Rahul wrote: Cover letter for the delete duplicate statement and delete unused cfg fragment patches Thanks. These look good. I'm preparing a pull request right now, so I'll merge these and include them in that update. Bruce Rahul Saxena (2): Removed duplicate statement Removed unused cfg file meta/cfg/kernel-cache/bsp/fishriver/eg20t.cfg | 39 - meta/cfg/kernel-cache/bsp/fishriver/fishriver.scc | 1 - 2 files changed, 0 insertions(+), 40 deletions(-) delete mode 100644 meta/cfg/kernel-cache/bsp/fishriver/eg20t.cfg ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
Re: [yocto] RT kernel with core-image-sato build failed for sugarbay
On 09/01/2011 02:01 PM, Bodke, Kishore K wrote: > Clean build with the below TOM's two commit IDs worked for me, if I > do not add IMAGE_INSTALL += "rttests" for sugarbay. > > It failed if I add the IMAGE_INSTALL += "rttests" Do you get the same error? If so, it's starting to sound like the second build is what fails since neither of your failure scenarios include a change that's related to the failure. Perhaps an issue with sstate... -- Darren > > Thanks Kishore. -Original Message- From: Darren Hart > [mailto:dvh...@linux.intel.com] Sent: Thursday, September 01, 2011 > 1:57 PM To: Zanussi, Tom Cc: Bodke, Kishore K; > yocto@yoctoproject.org; Hart, Darren; Saxena, Rahul Subject: Re: RT > kernel with core-image-sato build failed for sugarbay > > > > On 08/31/2011 09:19 PM, Tom Zanussi wrote: >> >> On Wed, 2011-08-31 at 16:49 -0700, Bodke, Kishore K wrote: >>> Hi, >>> >>> >>> >>> I am trying to build core-image-sato with rt kernel for sugarbay >>> and it fails for the do_rootfs . >>> >> >> FWIW, I was able to build and boot core-image-sato with the rt >> kernel: >> >> Linux sugarbay 3.0.3-rt8-yocto-preempt-rt+ #1 SMP PREEMPT RT Wed >> Aug 31 21:33:27 CDT 2011 x86_64 GNU/Linux >> >> using the following commits: >> >> poky/master 705ec50edda7f7fb8c132238fe3f1477384a470e meta/intel >> 902d2f3c919cd1d870dc8d35a7521cbba42e53ea >> >> Actually, meta/intel was that commit with Darren's grub patch on >> top (meta-intel/dvhart/grub-freetype). >> > > Kishore, would you try with a clean build and see if this works for > you? > > I'm going to try and build first without -rt and then with rt and see > if I can reproduce the issue. > > Thanks, > > Darren > >> Tom >> >>> >>> >>> Attached is the build log. >>> >>> >>> >>> Could you please let me know what went wrong? >>> >>> >>> >>> Thanks >>> >>> Kishore. >>> >>> >> >> > -- Darren Hart Intel Open Source Technology Center Yocto Project - Linux Kernel ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
Re: [yocto] RT kernel with core-image-sato build failed for sugarbay
Yeah, it is the same do_rootfs failure I get for both the builds. Thanks Kishore. -Original Message- From: Darren Hart [mailto:dvh...@linux.intel.com] Sent: Thursday, September 01, 2011 2:08 PM To: Bodke, Kishore K Cc: Zanussi, Tom; yocto@yoctoproject.org; Hart, Darren; Saxena, Rahul Subject: Re: RT kernel with core-image-sato build failed for sugarbay On 09/01/2011 02:01 PM, Bodke, Kishore K wrote: > Clean build with the below TOM's two commit IDs worked for me, if I > do not add IMAGE_INSTALL += "rttests" for sugarbay. > > It failed if I add the IMAGE_INSTALL += "rttests" Do you get the same error? If so, it's starting to sound like the second build is what fails since neither of your failure scenarios include a change that's related to the failure. Perhaps an issue with sstate... -- Darren > > Thanks Kishore. -Original Message- From: Darren Hart > [mailto:dvh...@linux.intel.com] Sent: Thursday, September 01, 2011 > 1:57 PM To: Zanussi, Tom Cc: Bodke, Kishore K; > yocto@yoctoproject.org; Hart, Darren; Saxena, Rahul Subject: Re: RT > kernel with core-image-sato build failed for sugarbay > > > > On 08/31/2011 09:19 PM, Tom Zanussi wrote: >> >> On Wed, 2011-08-31 at 16:49 -0700, Bodke, Kishore K wrote: >>> Hi, >>> >>> >>> >>> I am trying to build core-image-sato with rt kernel for sugarbay >>> and it fails for the do_rootfs . >>> >> >> FWIW, I was able to build and boot core-image-sato with the rt >> kernel: >> >> Linux sugarbay 3.0.3-rt8-yocto-preempt-rt+ #1 SMP PREEMPT RT Wed >> Aug 31 21:33:27 CDT 2011 x86_64 GNU/Linux >> >> using the following commits: >> >> poky/master 705ec50edda7f7fb8c132238fe3f1477384a470e meta/intel >> 902d2f3c919cd1d870dc8d35a7521cbba42e53ea >> >> Actually, meta/intel was that commit with Darren's grub patch on >> top (meta-intel/dvhart/grub-freetype). >> > > Kishore, would you try with a clean build and see if this works for > you? > > I'm going to try and build first without -rt and then with rt and see > if I can reproduce the issue. > > Thanks, > > Darren > >> Tom >> >>> >>> >>> Attached is the build log. >>> >>> >>> >>> Could you please let me know what went wrong? >>> >>> >>> >>> Thanks >>> >>> Kishore. >>> >>> >> >> > -- Darren Hart Intel Open Source Technology Center Yocto Project - Linux Kernel ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
Re: [yocto] RT kernel with core-image-sato build failed for sugarbay
Can you report the output of: $ bitbake -e core-image-sato | egrep "^IMAGE_INSTALL=" Thanks, Darren On 09/01/2011 02:14 PM, Bodke, Kishore K wrote: > Yeah, it is the same do_rootfs failure I get for both the builds. > Thanks > Kishore. > > -Original Message- > From: Darren Hart [mailto:dvh...@linux.intel.com] > Sent: Thursday, September 01, 2011 2:08 PM > To: Bodke, Kishore K > Cc: Zanussi, Tom; yocto@yoctoproject.org; Hart, Darren; Saxena, Rahul > Subject: Re: RT kernel with core-image-sato build failed for sugarbay > > On 09/01/2011 02:01 PM, Bodke, Kishore K wrote: >> Clean build with the below TOM's two commit IDs worked for me, if I >> do not add IMAGE_INSTALL += "rttests" for sugarbay. >> >> It failed if I add the IMAGE_INSTALL += "rttests" > > Do you get the same error? If so, it's starting to sound like the second > build is what fails since neither of your failure scenarios include > a change that's related to the failure. Perhaps an issue with sstate... > > -- > Darren > >> >> Thanks Kishore. -Original Message- From: Darren Hart >> [mailto:dvh...@linux.intel.com] Sent: Thursday, September 01, 2011 >> 1:57 PM To: Zanussi, Tom Cc: Bodke, Kishore K; >> yocto@yoctoproject.org; Hart, Darren; Saxena, Rahul Subject: Re: RT >> kernel with core-image-sato build failed for sugarbay >> >> >> >> On 08/31/2011 09:19 PM, Tom Zanussi wrote: >>> >>> On Wed, 2011-08-31 at 16:49 -0700, Bodke, Kishore K wrote: Hi, I am trying to build core-image-sato with rt kernel for sugarbay and it fails for the do_rootfs . >>> >>> FWIW, I was able to build and boot core-image-sato with the rt >>> kernel: >>> >>> Linux sugarbay 3.0.3-rt8-yocto-preempt-rt+ #1 SMP PREEMPT RT Wed >>> Aug 31 21:33:27 CDT 2011 x86_64 GNU/Linux >>> >>> using the following commits: >>> >>> poky/master 705ec50edda7f7fb8c132238fe3f1477384a470e meta/intel >>> 902d2f3c919cd1d870dc8d35a7521cbba42e53ea >>> >>> Actually, meta/intel was that commit with Darren's grub patch on >>> top (meta-intel/dvhart/grub-freetype). >>> >> >> Kishore, would you try with a clean build and see if this works for >> you? >> >> I'm going to try and build first without -rt and then with rt and see >> if I can reproduce the issue. >> >> Thanks, >> >> Darren >> >>> Tom >>> Attached is the build log. Could you please let me know what went wrong? Thanks Kishore. >>> >>> >> > -- Darren Hart Intel Open Source Technology Center Yocto Project - Linux Kernel ___ yocto mailing list yocto@yoctoproject.org https://lists.yoctoproject.org/listinfo/yocto
Re: [yocto] RT kernel with core-image-sato build failed for sugarbay
For the build I removed the rttests in the IMAGE_INSTALL , so it is not showing
up here.
bitbake -e core-image-sato | egrep "IMAGE_INSTALL="
# MULTILIB_IMAGE_INSTALL=
# IMAGE_INSTALL=${POKY_BASE_INSTALL}
IMAGE_INSTALL="task-core-boot task-base-extended"
I will rebuild with rttests and send the output.
Thanks
Kishore.
-Original Message-
From: Darren Hart [mailto:dvh...@linux.intel.com]
Sent: Thursday, September 01, 2011 2:52 PM
To: Bodke, Kishore K
Cc: Zanussi, Tom; yocto@yoctoproject.org; Hart, Darren; Saxena, Rahul
Subject: Re: RT kernel with core-image-sato build failed for sugarbay
Can you report the output of:
$ bitbake -e core-image-sato | egrep "^IMAGE_INSTALL="
Thanks,
Darren
On 09/01/2011 02:14 PM, Bodke, Kishore K wrote:
> Yeah, it is the same do_rootfs failure I get for both the builds.
> Thanks
> Kishore.
>
> -Original Message-
> From: Darren Hart [mailto:dvh...@linux.intel.com]
> Sent: Thursday, September 01, 2011 2:08 PM
> To: Bodke, Kishore K
> Cc: Zanussi, Tom; yocto@yoctoproject.org; Hart, Darren; Saxena, Rahul
> Subject: Re: RT kernel with core-image-sato build failed for sugarbay
>
> On 09/01/2011 02:01 PM, Bodke, Kishore K wrote:
>> Clean build with the below TOM's two commit IDs worked for me, if I
>> do not add IMAGE_INSTALL += "rttests" for sugarbay.
>>
>> It failed if I add the IMAGE_INSTALL += "rttests"
>
> Do you get the same error? If so, it's starting to sound like the second
> build is what fails since neither of your failure scenarios include
> a change that's related to the failure. Perhaps an issue with sstate...
>
> --
> Darren
>
>>
>> Thanks Kishore. -Original Message- From: Darren Hart
>> [mailto:dvh...@linux.intel.com] Sent: Thursday, September 01, 2011
>> 1:57 PM To: Zanussi, Tom Cc: Bodke, Kishore K;
>> yocto@yoctoproject.org; Hart, Darren; Saxena, Rahul Subject: Re: RT
>> kernel with core-image-sato build failed for sugarbay
>>
>>
>>
>> On 08/31/2011 09:19 PM, Tom Zanussi wrote:
>>>
>>> On Wed, 2011-08-31 at 16:49 -0700, Bodke, Kishore K wrote:
Hi,
I am trying to build core-image-sato with rt kernel for sugarbay
and it fails for the do_rootfs .
>>>
>>> FWIW, I was able to build and boot core-image-sato with the rt
>>> kernel:
>>>
>>> Linux sugarbay 3.0.3-rt8-yocto-preempt-rt+ #1 SMP PREEMPT RT Wed
>>> Aug 31 21:33:27 CDT 2011 x86_64 GNU/Linux
>>>
>>> using the following commits:
>>>
>>> poky/master 705ec50edda7f7fb8c132238fe3f1477384a470e meta/intel
>>> 902d2f3c919cd1d870dc8d35a7521cbba42e53ea
>>>
>>> Actually, meta/intel was that commit with Darren's grub patch on
>>> top (meta-intel/dvhart/grub-freetype).
>>>
>>
>> Kishore, would you try with a clean build and see if this works for
>> you?
>>
>> I'm going to try and build first without -rt and then with rt and see
>> if I can reproduce the issue.
>>
>> Thanks,
>>
>> Darren
>>
>>> Tom
>>>
Attached is the build log.
Could you please let me know what went wrong?
Thanks
Kishore.
>>>
>>>
>>
>
--
Darren Hart
Intel Open Source Technology Center
Yocto Project - Linux Kernel
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto
