| Message ID | 0470293e16f424ba4ee6505f3ef68e5ee13df483.1685672395.git.steve@sakoman.com |
|---|---|
| State | New |
| Headers | show |
| Series | [mickledore,01/13] linux-yocto/5.15: update to v5.15.109 | expand |
Hi, it looks like this change is causing mkfs.ext4 to segfault on some builders. I'm seeing it on ubuntu-18.04 (I know it's unsupported now). It might be related to older docker or libseccomp2 there, because there is similar issue with gdk-pixbuf-native which fails when buildtools-extended is used together with this and might be related to https://github.com/moby/moby/issues/43595 because when I try very similar 18.04 ubuntu image on my gentoo host I cannot reproduce neither the gdk-pixbuf-native failure nor the mkfs.ext4 segfaults. I'm trying to get access to some of the affected boxes to find out which version docker/libseccomp2 is used there and to double check that this is root cause for these failures (now I can only confirm that it started in Nanbield when this uninative upgrade was merged and now I've reproduced the same with kirkstone-nut with the same uninative upgrade. It might be useful to delay backporting this for a bit longer (and I'll soon confirm if 3.10 was already affected or not). Regards, On Fri, Jun 2, 2023 at 4:22 AM Steve Sakoman <steve@sakoman.com> wrote: > From: Michael Halstead <mhalstead@linuxfoundation.org> > > Signed-off-by: Michael Halstead <mhalstead@linuxfoundation.org> > Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org> > (cherry picked from commit f87becb69e02bdf055dffb633ed4f6d36b36f7a7) > Signed-off-by: Steve Sakoman <steve@sakoman.com> > --- > meta/conf/distro/include/yocto-uninative.inc | 8 ++++---- > 1 file changed, 4 insertions(+), 4 deletions(-) > > diff --git a/meta/conf/distro/include/yocto-uninative.inc > b/meta/conf/distro/include/yocto-uninative.inc > index f3c05b7e20..ad4816a1f3 100644 > --- a/meta/conf/distro/include/yocto-uninative.inc > +++ b/meta/conf/distro/include/yocto-uninative.inc > @@ -7,9 +7,9 @@ > # > > UNINATIVE_MAXGLIBCVERSION = "2.37" > -UNINATIVE_VERSION = "3.10" > +UNINATIVE_VERSION = "4.0" > > UNINATIVE_URL ?= " > http://downloads.yoctoproject.org/releases/uninative/${UNINATIVE_VERSION}/ > " > -UNINATIVE_CHECKSUM[aarch64] ?= > "b11f7dff0ff5e3c971f22e89a105efb0f8d2006fd71d1ea3e5da771e416b0e01" > -UNINATIVE_CHECKSUM[i686] ?= > "ad7daa7fca7693ca07eea5279f830333530171d357759b380954a25254edd8dd" > -UNINATIVE_CHECKSUM[x86_64] ?= > "ce8ff7e3247da312fbf84008f415bc1b4d113ac9fe4ec91581e4a20cd8b475d3" > +UNINATIVE_CHECKSUM[aarch64] ?= > "7baa8418a302df52e00916193b0a04f318356d9d2670c9a2bce3e966efefd738" > +UNINATIVE_CHECKSUM[i686] ?= > "83114d36883d43a521e280742b9849bf85d039b2f83d8e21d480659babe75ee8" > +UNINATIVE_CHECKSUM[x86_64] ?= > "fd75b2a1a67a10f6b7d65afb7d0f3e71a63b0038e428f34dfe420bb37716558a" > -- > 2.34.1 > > > -=-=-=-=-=-=-=-=-=-=-=- > Links: You receive all messages sent to this group. > View/Reply Online (#182136): > https://lists.openembedded.org/g/openembedded-core/message/182136 > Mute This Topic: https://lists.openembedded.org/mt/99279066/3617156 > Group Owner: openembedded-core+owner@lists.openembedded.org > Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [ > Martin.Jansa@gmail.com] > -=-=-=-=-=-=-=-=-=-=-=- > >
On Sun, Jun 4, 2023 at 7:39 AM Martin Jansa <martin.jansa@gmail.com> wrote: > > Hi, > > it looks like this change is causing mkfs.ext4 to segfault on some builders. > > I'm seeing it on ubuntu-18.04 (I know it's unsupported now). It might be related to older docker or libseccomp2 there, because there is similar issue with gdk-pixbuf-native which fails when buildtools-extended is used together with this and might be related to https://github.com/moby/moby/issues/43595 because when I try very similar 18.04 ubuntu image on my gentoo host I cannot reproduce neither the gdk-pixbuf-native failure nor the mkfs.ext4 segfaults. > > I'm trying to get access to some of the affected boxes to find out which version docker/libseccomp2 is used there and to double check that this is root cause for these failures (now I can only confirm that it started in Nanbield when this uninative upgrade was merged and now I've reproduced the same with kirkstone-nut with the same uninative upgrade. > > It might be useful to delay backporting this for a bit longer (and I'll soon confirm if 3.10 was already affected or not). Agreed, I'm struggling with kirkstone too, so will delay taking the uninative update until we learn more. Thanks for the review! Steve > On Fri, Jun 2, 2023 at 4:22 AM Steve Sakoman <steve@sakoman.com> wrote: >> >> From: Michael Halstead <mhalstead@linuxfoundation.org> >> >> Signed-off-by: Michael Halstead <mhalstead@linuxfoundation.org> >> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org> >> (cherry picked from commit f87becb69e02bdf055dffb633ed4f6d36b36f7a7) >> Signed-off-by: Steve Sakoman <steve@sakoman.com> >> --- >> meta/conf/distro/include/yocto-uninative.inc | 8 ++++---- >> 1 file changed, 4 insertions(+), 4 deletions(-) >> >> diff --git a/meta/conf/distro/include/yocto-uninative.inc b/meta/conf/distro/include/yocto-uninative.inc >> index f3c05b7e20..ad4816a1f3 100644 >> --- a/meta/conf/distro/include/yocto-uninative.inc >> +++ b/meta/conf/distro/include/yocto-uninative.inc >> @@ -7,9 +7,9 @@ >> # >> >> UNINATIVE_MAXGLIBCVERSION = "2.37" >> -UNINATIVE_VERSION = "3.10" >> +UNINATIVE_VERSION = "4.0" >> >> UNINATIVE_URL ?= "http://downloads.yoctoproject.org/releases/uninative/${UNINATIVE_VERSION}/" >> -UNINATIVE_CHECKSUM[aarch64] ?= "b11f7dff0ff5e3c971f22e89a105efb0f8d2006fd71d1ea3e5da771e416b0e01" >> -UNINATIVE_CHECKSUM[i686] ?= "ad7daa7fca7693ca07eea5279f830333530171d357759b380954a25254edd8dd" >> -UNINATIVE_CHECKSUM[x86_64] ?= "ce8ff7e3247da312fbf84008f415bc1b4d113ac9fe4ec91581e4a20cd8b475d3" >> +UNINATIVE_CHECKSUM[aarch64] ?= "7baa8418a302df52e00916193b0a04f318356d9d2670c9a2bce3e966efefd738" >> +UNINATIVE_CHECKSUM[i686] ?= "83114d36883d43a521e280742b9849bf85d039b2f83d8e21d480659babe75ee8" >> +UNINATIVE_CHECKSUM[x86_64] ?= "fd75b2a1a67a10f6b7d65afb7d0f3e71a63b0038e428f34dfe420bb37716558a" >> -- >> 2.34.1 >> >> >> -=-=-=-=-=-=-=-=-=-=-=- >> Links: You receive all messages sent to this group. >> View/Reply Online (#182136): https://lists.openembedded.org/g/openembedded-core/message/182136 >> Mute This Topic: https://lists.openembedded.org/mt/99279066/3617156 >> Group Owner: openembedded-core+owner@lists.openembedded.org >> Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [Martin.Jansa@gmail.com] >> -=-=-=-=-=-=-=-=-=-=-=- >>
> It might be useful to delay backporting this for a bit longer (and I'll soon confirm if 3.10 was already affected or not). I can confirm that 3.10 is also affected. I did build with https://git.openembedded.org/openembedded-core-contrib/commit/?h=stable/kirkstone-nut&id=75e3870ea124321b81f856c4e5170c891f0bd050 and mkfs.ext4 didn't segfault. With uninative 3.10 from https://git.openembedded.org/openembedded-core-contrib/commit/?h=stable/kirkstone-nut&id=680eecc3ae09ea5585e46ff0e67f7c6704221806 it segfaults again. I haven't reproduced the gdk-pixbuf-native issue with buildtools-extended 4.0.6 with kirkstone yet, so that one might be only with nanbield, but mkfs.ext4 segfaults seem related to uninative for sure. Will debug more when I get access. Cheers, On Sun, Jun 4, 2023 at 9:14 PM Steve Sakoman <steve@sakoman.com> wrote: > On Sun, Jun 4, 2023 at 7:39 AM Martin Jansa <martin.jansa@gmail.com> > wrote: > > > > Hi, > > > > it looks like this change is causing mkfs.ext4 to segfault on some > builders. > > > > I'm seeing it on ubuntu-18.04 (I know it's unsupported now). It might be > related to older docker or libseccomp2 there, because there is similar > issue with gdk-pixbuf-native which fails when buildtools-extended is used > together with this and might be related to > https://github.com/moby/moby/issues/43595 because when I try very similar > 18.04 ubuntu image on my gentoo host I cannot reproduce neither the > gdk-pixbuf-native failure nor the mkfs.ext4 segfaults. > > > > I'm trying to get access to some of the affected boxes to find out which > version docker/libseccomp2 is used there and to double check that this is > root cause for these failures (now I can only confirm that it started in > Nanbield when this uninative upgrade was merged and now I've reproduced the > same with kirkstone-nut with the same uninative upgrade. > > > > It might be useful to delay backporting this for a bit longer (and I'll > soon confirm if 3.10 was already affected or not). > > Agreed, I'm struggling with kirkstone too, so will delay taking the > uninative update until we learn more. > > Thanks for the review! > > Steve > > > On Fri, Jun 2, 2023 at 4:22 AM Steve Sakoman <steve@sakoman.com> wrote: > >> > >> From: Michael Halstead <mhalstead@linuxfoundation.org> > >> > >> Signed-off-by: Michael Halstead <mhalstead@linuxfoundation.org> > >> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org> > >> (cherry picked from commit f87becb69e02bdf055dffb633ed4f6d36b36f7a7) > >> Signed-off-by: Steve Sakoman <steve@sakoman.com> > >> --- > >> meta/conf/distro/include/yocto-uninative.inc | 8 ++++---- > >> 1 file changed, 4 insertions(+), 4 deletions(-) > >> > >> diff --git a/meta/conf/distro/include/yocto-uninative.inc > b/meta/conf/distro/include/yocto-uninative.inc > >> index f3c05b7e20..ad4816a1f3 100644 > >> --- a/meta/conf/distro/include/yocto-uninative.inc > >> +++ b/meta/conf/distro/include/yocto-uninative.inc > >> @@ -7,9 +7,9 @@ > >> # > >> > >> UNINATIVE_MAXGLIBCVERSION = "2.37" > >> -UNINATIVE_VERSION = "3.10" > >> +UNINATIVE_VERSION = "4.0" > >> > >> UNINATIVE_URL ?= " > http://downloads.yoctoproject.org/releases/uninative/${UNINATIVE_VERSION}/ > " > >> -UNINATIVE_CHECKSUM[aarch64] ?= > "b11f7dff0ff5e3c971f22e89a105efb0f8d2006fd71d1ea3e5da771e416b0e01" > >> -UNINATIVE_CHECKSUM[i686] ?= > "ad7daa7fca7693ca07eea5279f830333530171d357759b380954a25254edd8dd" > >> -UNINATIVE_CHECKSUM[x86_64] ?= > "ce8ff7e3247da312fbf84008f415bc1b4d113ac9fe4ec91581e4a20cd8b475d3" > >> +UNINATIVE_CHECKSUM[aarch64] ?= > "7baa8418a302df52e00916193b0a04f318356d9d2670c9a2bce3e966efefd738" > >> +UNINATIVE_CHECKSUM[i686] ?= > "83114d36883d43a521e280742b9849bf85d039b2f83d8e21d480659babe75ee8" > >> +UNINATIVE_CHECKSUM[x86_64] ?= > "fd75b2a1a67a10f6b7d65afb7d0f3e71a63b0038e428f34dfe420bb37716558a" > >> -- > >> 2.34.1 > >> > >> > >> -=-=-=-=-=-=-=-=-=-=-=- > >> Links: You receive all messages sent to this group. > >> View/Reply Online (#182136): > https://lists.openembedded.org/g/openembedded-core/message/182136 > >> Mute This Topic: https://lists.openembedded.org/mt/99279066/3617156 > >> Group Owner: openembedded-core+owner@lists.openembedded.org > >> Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [ > Martin.Jansa@gmail.com] > >> -=-=-=-=-=-=-=-=-=-=-=- > >> >
On Sun, Jun 4, 2023 at 10:34 AM Martin Jansa <martin.jansa@gmail.com> wrote: > > > It might be useful to delay backporting this for a bit longer (and I'll soon confirm if 3.10 was already affected or not). > > I can confirm that 3.10 is also affected. > > I did build with https://git.openembedded.org/openembedded-core-contrib/commit/?h=stable/kirkstone-nut&id=75e3870ea124321b81f856c4e5170c891f0bd050 and mkfs.ext4 didn't segfault. With uninative 3.10 from https://git.openembedded.org/openembedded-core-contrib/commit/?h=stable/kirkstone-nut&id=680eecc3ae09ea5585e46ff0e67f7c6704221806 it segfaults again. > > I haven't reproduced the gdk-pixbuf-native issue with buildtools-extended 4.0.6 with kirkstone yet, so that one might be only with nanbield, but mkfs.ext4 segfaults seem related to uninative for sure. Will debug more when I get access. Have you learned anything more on this issue? I've continued testing mickledore both locally and on the autobuilder and I'm not seeing any failures :-( Steve > On Sun, Jun 4, 2023 at 9:14 PM Steve Sakoman <steve@sakoman.com> wrote: >> >> On Sun, Jun 4, 2023 at 7:39 AM Martin Jansa <martin.jansa@gmail.com> wrote: >> > >> > Hi, >> > >> > it looks like this change is causing mkfs.ext4 to segfault on some builders. >> > >> > I'm seeing it on ubuntu-18.04 (I know it's unsupported now). It might be related to older docker or libseccomp2 there, because there is similar issue with gdk-pixbuf-native which fails when buildtools-extended is used together with this and might be related to https://github.com/moby/moby/issues/43595 because when I try very similar 18.04 ubuntu image on my gentoo host I cannot reproduce neither the gdk-pixbuf-native failure nor the mkfs.ext4 segfaults. >> > >> > I'm trying to get access to some of the affected boxes to find out which version docker/libseccomp2 is used there and to double check that this is root cause for these failures (now I can only confirm that it started in Nanbield when this uninative upgrade was merged and now I've reproduced the same with kirkstone-nut with the same uninative upgrade. >> > >> > It might be useful to delay backporting this for a bit longer (and I'll soon confirm if 3.10 was already affected or not). >> >> Agreed, I'm struggling with kirkstone too, so will delay taking the >> uninative update until we learn more. >> >> Thanks for the review! >> >> Steve >> >> > On Fri, Jun 2, 2023 at 4:22 AM Steve Sakoman <steve@sakoman.com> wrote: >> >> >> >> From: Michael Halstead <mhalstead@linuxfoundation.org> >> >> >> >> Signed-off-by: Michael Halstead <mhalstead@linuxfoundation.org> >> >> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org> >> >> (cherry picked from commit f87becb69e02bdf055dffb633ed4f6d36b36f7a7) >> >> Signed-off-by: Steve Sakoman <steve@sakoman.com> >> >> --- >> >> meta/conf/distro/include/yocto-uninative.inc | 8 ++++---- >> >> 1 file changed, 4 insertions(+), 4 deletions(-) >> >> >> >> diff --git a/meta/conf/distro/include/yocto-uninative.inc b/meta/conf/distro/include/yocto-uninative.inc >> >> index f3c05b7e20..ad4816a1f3 100644 >> >> --- a/meta/conf/distro/include/yocto-uninative.inc >> >> +++ b/meta/conf/distro/include/yocto-uninative.inc >> >> @@ -7,9 +7,9 @@ >> >> # >> >> >> >> UNINATIVE_MAXGLIBCVERSION = "2.37" >> >> -UNINATIVE_VERSION = "3.10" >> >> +UNINATIVE_VERSION = "4.0" >> >> >> >> UNINATIVE_URL ?= "http://downloads.yoctoproject.org/releases/uninative/${UNINATIVE_VERSION}/" >> >> -UNINATIVE_CHECKSUM[aarch64] ?= "b11f7dff0ff5e3c971f22e89a105efb0f8d2006fd71d1ea3e5da771e416b0e01" >> >> -UNINATIVE_CHECKSUM[i686] ?= "ad7daa7fca7693ca07eea5279f830333530171d357759b380954a25254edd8dd" >> >> -UNINATIVE_CHECKSUM[x86_64] ?= "ce8ff7e3247da312fbf84008f415bc1b4d113ac9fe4ec91581e4a20cd8b475d3" >> >> +UNINATIVE_CHECKSUM[aarch64] ?= "7baa8418a302df52e00916193b0a04f318356d9d2670c9a2bce3e966efefd738" >> >> +UNINATIVE_CHECKSUM[i686] ?= "83114d36883d43a521e280742b9849bf85d039b2f83d8e21d480659babe75ee8" >> >> +UNINATIVE_CHECKSUM[x86_64] ?= "fd75b2a1a67a10f6b7d65afb7d0f3e71a63b0038e428f34dfe420bb37716558a" >> >> -- >> >> 2.34.1 >> >> >> >> >> >> -=-=-=-=-=-=-=-=-=-=-=- >> >> Links: You receive all messages sent to this group. >> >> View/Reply Online (#182136): https://lists.openembedded.org/g/openembedded-core/message/182136 >> >> Mute This Topic: https://lists.openembedded.org/mt/99279066/3617156 >> >> Group Owner: openembedded-core+owner@lists.openembedded.org >> >> Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [Martin.Jansa@gmail.com] >> >> -=-=-=-=-=-=-=-=-=-=-=- >> >>
Hi Steve, I still didn't get shell access to those builders, feel free to merge it as is. Hopefully other people don't have so outdated build infrastructure :). Regards, On Thu, Jun 8, 2023 at 8:50 PM Steve Sakoman <steve@sakoman.com> wrote: > On Sun, Jun 4, 2023 at 10:34 AM Martin Jansa <martin.jansa@gmail.com> > wrote: > > > > > It might be useful to delay backporting this for a bit longer (and > I'll soon confirm if 3.10 was already affected or not). > > > > I can confirm that 3.10 is also affected. > > > > I did build with > https://git.openembedded.org/openembedded-core-contrib/commit/?h=stable/kirkstone-nut&id=75e3870ea124321b81f856c4e5170c891f0bd050 > and mkfs.ext4 didn't segfault. With uninative 3.10 from > https://git.openembedded.org/openembedded-core-contrib/commit/?h=stable/kirkstone-nut&id=680eecc3ae09ea5585e46ff0e67f7c6704221806 > it segfaults again. > > > > I haven't reproduced the gdk-pixbuf-native issue with > buildtools-extended 4.0.6 with kirkstone yet, so that one might be only > with nanbield, but mkfs.ext4 segfaults seem related to uninative for sure. > Will debug more when I get access. > > Have you learned anything more on this issue? I've continued testing > mickledore both locally and on the autobuilder and I'm not seeing any > failures :-( > > Steve > > > On Sun, Jun 4, 2023 at 9:14 PM Steve Sakoman <steve@sakoman.com> wrote: > >> > >> On Sun, Jun 4, 2023 at 7:39 AM Martin Jansa <martin.jansa@gmail.com> > wrote: > >> > > >> > Hi, > >> > > >> > it looks like this change is causing mkfs.ext4 to segfault on some > builders. > >> > > >> > I'm seeing it on ubuntu-18.04 (I know it's unsupported now). It might > be related to older docker or libseccomp2 there, because there is similar > issue with gdk-pixbuf-native which fails when buildtools-extended is used > together with this and might be related to > https://github.com/moby/moby/issues/43595 because when I try very similar > 18.04 ubuntu image on my gentoo host I cannot reproduce neither the > gdk-pixbuf-native failure nor the mkfs.ext4 segfaults. > >> > > >> > I'm trying to get access to some of the affected boxes to find out > which version docker/libseccomp2 is used there and to double check that > this is root cause for these failures (now I can only confirm that it > started in Nanbield when this uninative upgrade was merged and now I've > reproduced the same with kirkstone-nut with the same uninative upgrade. > >> > > >> > It might be useful to delay backporting this for a bit longer (and > I'll soon confirm if 3.10 was already affected or not). > >> > >> Agreed, I'm struggling with kirkstone too, so will delay taking the > >> uninative update until we learn more. > >> > >> Thanks for the review! > >> > >> Steve > >> > >> > On Fri, Jun 2, 2023 at 4:22 AM Steve Sakoman <steve@sakoman.com> > wrote: > >> >> > >> >> From: Michael Halstead <mhalstead@linuxfoundation.org> > >> >> > >> >> Signed-off-by: Michael Halstead <mhalstead@linuxfoundation.org> > >> >> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org> > >> >> (cherry picked from commit f87becb69e02bdf055dffb633ed4f6d36b36f7a7) > >> >> Signed-off-by: Steve Sakoman <steve@sakoman.com> > >> >> --- > >> >> meta/conf/distro/include/yocto-uninative.inc | 8 ++++---- > >> >> 1 file changed, 4 insertions(+), 4 deletions(-) > >> >> > >> >> diff --git a/meta/conf/distro/include/yocto-uninative.inc > b/meta/conf/distro/include/yocto-uninative.inc > >> >> index f3c05b7e20..ad4816a1f3 100644 > >> >> --- a/meta/conf/distro/include/yocto-uninative.inc > >> >> +++ b/meta/conf/distro/include/yocto-uninative.inc > >> >> @@ -7,9 +7,9 @@ > >> >> # > >> >> > >> >> UNINATIVE_MAXGLIBCVERSION = "2.37" > >> >> -UNINATIVE_VERSION = "3.10" > >> >> +UNINATIVE_VERSION = "4.0" > >> >> > >> >> UNINATIVE_URL ?= " > http://downloads.yoctoproject.org/releases/uninative/${UNINATIVE_VERSION}/ > " > >> >> -UNINATIVE_CHECKSUM[aarch64] ?= > "b11f7dff0ff5e3c971f22e89a105efb0f8d2006fd71d1ea3e5da771e416b0e01" > >> >> -UNINATIVE_CHECKSUM[i686] ?= > "ad7daa7fca7693ca07eea5279f830333530171d357759b380954a25254edd8dd" > >> >> -UNINATIVE_CHECKSUM[x86_64] ?= > "ce8ff7e3247da312fbf84008f415bc1b4d113ac9fe4ec91581e4a20cd8b475d3" > >> >> +UNINATIVE_CHECKSUM[aarch64] ?= > "7baa8418a302df52e00916193b0a04f318356d9d2670c9a2bce3e966efefd738" > >> >> +UNINATIVE_CHECKSUM[i686] ?= > "83114d36883d43a521e280742b9849bf85d039b2f83d8e21d480659babe75ee8" > >> >> +UNINATIVE_CHECKSUM[x86_64] ?= > "fd75b2a1a67a10f6b7d65afb7d0f3e71a63b0038e428f34dfe420bb37716558a" > >> >> -- > >> >> 2.34.1 > >> >> > >> >> > >> >> -=-=-=-=-=-=-=-=-=-=-=- > >> >> Links: You receive all messages sent to this group. > >> >> View/Reply Online (#182136): > https://lists.openembedded.org/g/openembedded-core/message/182136 > >> >> Mute This Topic: https://lists.openembedded.org/mt/99279066/3617156 > >> >> Group Owner: openembedded-core+owner@lists.openembedded.org > >> >> Unsubscribe: > https://lists.openembedded.org/g/openembedded-core/unsub [ > Martin.Jansa@gmail.com] > >> >> -=-=-=-=-=-=-=-=-=-=-=- > >> >> >
On Sun, Jun 4, 2023 at 10:34 PM Martin Jansa <martin.jansa@gmail.com> wrote: > > It might be useful to delay backporting this for a bit longer (and I'll > soon confirm if 3.10 was already affected or not). > > I can confirm that 3.10 is also affected. > > I did build with > https://git.openembedded.org/openembedded-core-contrib/commit/?h=stable/kirkstone-nut&id=75e3870ea124321b81f856c4e5170c891f0bd050 > and mkfs.ext4 didn't segfault. With uninative 3.10 from > https://git.openembedded.org/openembedded-core-contrib/commit/?h=stable/kirkstone-nut&id=680eecc3ae09ea5585e46ff0e67f7c6704221806 > it segfaults again. > > I haven't reproduced the gdk-pixbuf-native issue with buildtools-extended > 4.0.6 with kirkstone yet, so that one might be only with nanbield, but > mkfs.ext4 segfaults seem related to uninative for sure. Will debug more > when I get access. > I got access to one of the servers where this is still reproducible, but the real root cause is still a mystery for me. I'll share what I've found so far in case someone else is seeing the same or can spot what I'm missing. It's reproducible on ubuntu-18.06, but not everywhere, I've tried older docker image and latest 18.04 on my host and both work fine. I've tried to narrow the difference from the 18.04 image where it is failing but without any luck. The same happens with nanbield (where I've noticed it first), but now the same happens with mickledore and kirkstone. No special config needed, default poky configuration reproduces the same. Easiest test is to build just e2fsprogs-native and try to execute mkfs.ext4.real, if it immediately segfaults it's a "bad" binary, if it complains about missing e2* symbols or BLKID_2_36 symbol missing in libblkid.so.1, then it's "good" binary and works as long as you let it load the right libraries from native sysroot. Surprisingly whatever breaks the binary happens between "image" and "sysroot-destdir" so first I was suspecting strip from host's binutils, but downgrading binutils in different 18.04 system to match the one where it is failing didn't reproduce it. Downgrading libgcc1 and some other packages to match didn't reproduce it as well. Next I've tried to inhibit strip completely (assuming it will be usable work around to keep the binary in sysroot-destdir working with new uninative and narrow down what I need to debug), but surprisingly now both the binary in "image" and "sysroot-destdir" fail the same. Strace doesn't show anything useful, other than segfault triggered very early: $ strace tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/sysroot-destdir/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real execve("tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/sysroot-destdir/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real", ["tmp/work/x86_64-linux/e2fsprogs-"...], 0x7fffbec05ac0 /* 55 vars */) = 0 brk(NULL) = 0x55b91bc0e000 arch_prctl(0x3001 /* ARCH_??? */, 0x7ffc146b5340) = -1 EINVAL (Invalid argument) --- SIGSEGV {si_signo=SIGSEGV, si_code=SEGV_MAPERR, si_addr=0x55b91ba016e0} --- +++ killed by SIGSEGV (core dumped) +++ Segmentation fault (core dumped) $ strace tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/image/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real 2>&1 | head execve("tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/image/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real", ["tmp/work/x86_64-linux/e2fsprogs-"...], 0x7ffd387a7f60 /* 55 vars */) = 0 brk(NULL) = 0x5641b084f000 arch_prctl(0x3001 /* ARCH_??? */, 0x7ffe271263f0) = -1 EINVAL (Invalid argument) --- SIGSEGV {si_signo=SIGSEGV, si_code=SEGV_MAPERR, si_addr=0x5641b00016e0} --- +++ killed by SIGSEGV (core dumped) +++ While "good" binary: $ strace tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/image/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real | head execve("tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/image/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real", ["tmp/work/x86_64-linux/e2fsprogs-"...], 0x7ffd9f569ae0 /* 55 vars */) = 0 brk(NULL) = 0x5569fad4e000 arch_prctl(0x3001 /* ARCH_??? */, 0x7ffe1d8ce2f0) = -1 EINVAL (Invalid argument) mmap(NULL, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f210b337000 access("/etc/ld.so.preload", R_OK) = -1 ENOENT (No such file or directory) openat(AT_FDCWD, "/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/usr/lib/glibc-hwcaps/x86-64-v4/libext2fs.so.2", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) newfstatat(AT_FDCWD, "/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/usr/lib/glibc-hwcaps/x86-64-v4", 0x7ffe1d8cd520, 0) = -1 ENOENT (No such file or directory) The diffoscope output shows that dynsym section is second to last in "bad" binary and first in "good" binary, which might relate to the warnings shown by gdb: Reading symbols from ./recipe-sysroot-native/sbin/mkfs.ext4.real... warning: Loadable section ".dynsym" outside of ELF segments warning: Loadable section ".interp" outside of ELF segments (No debugging symbols found in ./recipe-sysroot-native/sbin/mkfs.ext4.real) (gdb) r Starting program: /data001/jansa/build-webos/BUILD/work/qemux86_64-webos-linux/webos-image/1.0-r3normal1/recipe-sysroot-native/sbin/mkfs.ext4.real warning: Error disabling address space randomization: Operation not permitted BFD: /data001/jansa/build-webos/BUILD/sysroots-uninative/x86_64-linux/lib/ld-linux-x86-64.so.2: unknown type [0x13] section `.relr.dyn' warning: `/data001/jansa/build-webos/BUILD/sysroots-uninative/x86_64-linux/lib/ld-linux-x86-64.so.2': Shared library architecture unknown is not compatible with target architecture i386:x86-64. warning: `/data001/jansa/build-webos/BUILD/sysroots-uninative/x86_64-linux/lib/ld-linux-x86-64.so.2': Shared library architecture unknown is not compatible with target architecture i386:x86-64. Program received signal SIGSEGV, Segmentation fault. 0x00007f08c9d9232a in strcmp () from /data001/jansa/build-webos/BUILD/sysroots-uninative/x86_64-linux/lib/ld-linux-x86-64.so.2 And ldd also says that the "bad" binary isn't dynamic executable while "good" on is (objdump -x works for both of them): the "bad" one: ldd tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/sysroot-destdir/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real not a dynamic executable and the "good" one: ldd tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/image/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real linux-vdso.so.1 (0x00007ffce83fc000) libext2fs.so.2 => /lib/x86_64-linux-gnu/libext2fs.so.2 (0x00007fd1ba5a0000) libcom_err.so.2 => /lib/x86_64-linux-gnu/libcom_err.so.2 (0x00007fd1ba39c000) libblkid.so.1 => /home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/usr/lib/libblkid.so.1 (0x00007fd1ba149000) libuuid.so.1 => /home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/usr/lib/libuuid.so.1 (0x00007fd1b9f42000) libe2p.so.2 => /lib/x86_64-linux-gnu/libe2p.so.2 (0x00007fd1b9d39000) libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007fd1b9b1a000) libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007fd1b9729000) /home/worker/poky-kirkstone-jansa/build/tmp/sysroots-uninative/x86_64-linux/lib/ld-linux-x86-64.so.2 => /lib64/ld-linux-x86-64.so.2 (0x00007fd1baa0f000) Once the "bad" binary is built (with either uninative 4.0 or 3.9) it will always fail with uninative 4.0 even when reused from sstate on different server (if it's the server producing "good" mkfs.ext4.real binaries, then just rebuilding e2fsprogs-native will fix it). Does this make any sense to anyone? I'll continue to look into it, but if you have any hints, please let me know. I know 18.04 is EOL and nobody should be using it, I'm trying to get our build infra upgraded for very long time. The diffoscope output (mkfs.4.0-18.04-sstate is the "bad" binary from sstate while mkfs.4.0-18.04-rebuilt is the "good" one). --- mkfs.4.0-18.04-rebuilt +++ mkfs.4.0-18.04-sstate │┄ File has been modified after NT_GNU_BUILD_ID has been applied. ├── readelf --wide --file-header {} │ @@ -4,17 +4,17 @@ │ Data: 2's complement, little endian │ Version: 1 (current) │ OS/ABI: UNIX - System V │ ABI Version: 0 │ Type: DYN (Position-Independent Executable file) │ Machine: Advanced Micro Devices X86-64 │ Version: 0x1 │ - Entry point address: 0x8350 │ + Entry point address: 0x8380 │ Start of program headers: 64 (bytes into file) │ Start of section headers: 127520 (bytes into file) │ Flags: 0x0 │ Size of this header: 64 (bytes) │ Size of program headers: 56 (bytes) │ Number of program headers: 13 │ Size of section headers: 64 (bytes) │ Number of section headers: 27 │ - Section header string table index: 22 │ + Section header string table index: 21 ├── readelf --wide --program-header {} │ @@ -1,37 +1,37 @@ │ │ Elf file type is DYN (Position-Independent Executable file) │ -Entry point 0x8350 │ +Entry point 0x8380 │ There are 13 program headers, starting at offset 64 │ │ Program Headers: │ Type Offset VirtAddr PhysAddr FileSiz MemSiz Flg Align │ PHDR 0x000040 0x0000000000000040 0x0000000000000040 0x0002d8 0x0002d8 R 0x8 │ - LOAD 0x000000 0x0000000000000000 0x0000000000000000 0x01dcb0 0x01dcb0 R E 0x200000 │ + LOAD 0x000000 0x0000000000000000 0x0000000000000000 0x01dcf0 0x01dcf0 R E 0x200000 │ GNU_STACK 0x000000 0x0000000000000000 0x0000000000000000 0x000000 0x000000 RW 0x10 │ - GNU_EH_FRAME 0x01b794 0x000000000001b794 0x000000000001b794 0x0004fc 0x0004fc R 0x4 │ + GNU_EH_FRAME 0x01b7d4 0x000000000001b7d4 0x000000000001b7d4 0x0004fc 0x0004fc R 0x4 │ LOAD 0x01e450 0x000000000021e450 0x000000000021e450 0x000cd8 0x0013d8 RW 0x200000 │ GNU_RELRO 0x01e450 0x000000000021e450 0x000000000021e450 0x000bb0 0x000bb0 R 0x1 │ DYNAMIC 0x01e678 0x000000000021e678 0x000000000021e678 0x000260 0x000260 RW 0x8 │ - LOAD 0x020000 0x0000000000400000 0x0000000000400000 0x000178 0x000178 RW 0x200000 │ + LOAD 0x020000 0x0000000000400000 0x0000000000400000 0x0000a8 0x0000a8 RW 0x200000 │ NOTE 0x020060 0x0000000000400060 0x0000000000400060 0x000020 0x000020 R 0x4 │ NOTE 0x020080 0x0000000000400080 0x0000000000400080 0x000024 0x000024 R 0x4 │ - LOAD 0x021000 0x0000000000600000 0x0000000000600000 0x001060 0x001060 RW 0x200000 │ - INTERP 0x023000 0x0000000000a00000 0x0000000000a00000 0x00005a 0x00005a R 0x1 │ + LOAD 0x021000 0x0000000000600000 0x0000000000600000 0x000130 0x000130 RW 0x200000 │ + LOAD 0x022000 0x0000000000800000 0x0000000000800000 0x002740 0x002740 RW 0x200000 │ + INTERP 0x0246e0 0x0000000000a016e0 0x0000000000a016e0 0x00005a 0x00005a R 0x1 │ [Requesting program interpreter: /data001/jansa/build-webos/BUILD/sysroots-uninative/x86_64-linux/lib/ld-linux-x86-64.so.2] │ - LOAD 0x023000 0x0000000000a00000 0x0000000000a00000 0x000060 0x000060 RW 0x200000 │ │ Section to Segment mapping: │ Segment Sections... │ 00 │ - 01 .dynsym .dynstr .gnu.version .gnu.version_r .rela.dyn .rela.plt .init .plt .plt.got .text .fini .rodata .eh_frame_hdr .eh_frame │ + 01 .dynstr .gnu.version .gnu.version_r .rela.dyn .rela.plt .init .plt .plt.got .text .fini .rodata .eh_frame_hdr .eh_frame │ 02 │ 03 .eh_frame_hdr │ 04 .init_array .fini_array .data.rel.ro .dynamic .got .data .bss │ 05 .init_array .fini_array .data.rel.ro .dynamic .got │ 06 .dynamic │ - 07 .note.ABI-tag .note.gnu.build-id .gnu.hash │ + 07 .note.ABI-tag .note.gnu.build-id │ 08 .note.ABI-tag │ 09 .note.gnu.build-id │ - 10 │ - 11 .interp │ + 10 .gnu.hash │ + 11 │ 12 .interp ├── readelf --wide --sections {} │ @@ -1,36 +1,36 @@ │ There are 27 section headers, starting at offset 0x1f220: │ │ Section Headers: │ [Nr] Name Type Address Off Size ES Flg Lk Inf Al │ [ 0] NULL 0000000000000000 000000 000000 00 0 0 0 │ - [ 1] .dynsym DYNSYM 00000000000003a8 0003a8 0016e0 18 A 2 1 8 │ - [ 2] .dynstr STRTAB 0000000000001a88 001a88 001182 00 A 0 0 1 │ - [ 3] .gnu.version VERSYM 0000000000002c0a 002c0a 0001e8 02 A 1 0 2 │ - [ 4] .gnu.version_r VERNEED 0000000000002df8 002df8 000100 00 A 2 4 8 │ - [ 5] .rela.dyn RELA 0000000000002ef8 002ef8 0008a0 18 A 1 0 8 │ - [ 6] .rela.plt RELA 0000000000003798 003798 001488 18 AI 1 19 8 │ - [ 7] .init PROGBITS 0000000000004c20 004c20 000017 00 AX 0 0 4 │ - [ 8] .plt PROGBITS 0000000000004c40 004c40 000dc0 10 AX 0 0 16 │ - [ 9] .plt.got PROGBITS 0000000000005a00 005a00 000008 08 AX 0 0 8 │ - [10] .text PROGBITS 0000000000005a10 005a10 010cd2 00 AX 0 0 16 │ - [11] .fini PROGBITS 00000000000166e4 0166e4 000009 00 AX 0 0 4 │ - [12] .rodata PROGBITS 00000000000166f0 0166f0 0050a2 00 A 0 0 16 │ - [13] .eh_frame_hdr PROGBITS 000000000001b794 01b794 0004fc 00 A 0 0 4 │ - [14] .eh_frame PROGBITS 000000000001bc90 01bc90 002020 00 A 0 0 8 │ - [15] .init_array INIT_ARRAY 000000000021e450 01e450 000008 08 WA 0 0 8 │ - [16] .fini_array FINI_ARRAY 000000000021e458 01e458 000008 08 WA 0 0 8 │ - [17] .data.rel.ro PROGBITS 000000000021e460 01e460 000218 00 WA 0 0 32 │ - [18] .dynamic DYNAMIC 000000000021e678 01e678 000260 10 WA 2 0 8 │ - [19] .got PROGBITS 000000000021e8d8 01e8d8 000720 08 WA 0 0 8 │ - [20] .data PROGBITS 000000000021f000 01f000 000128 00 WA 0 0 32 │ - [21] .bss NOBITS 000000000021f140 01f128 0006e8 00 WA 0 0 32 │ - [22] .shstrtab STRTAB 0000000000000000 01f128 0000f2 00 0 0 1 │ - [23] .note.ABI-tag NOTE 0000000000400060 020060 000020 00 A 0 0 4 │ - [24] .note.gnu.build-id NOTE 0000000000400080 020080 000024 00 A 0 0 4 │ - [25] .gnu.hash GNU_HASH 00000000004000a8 0200a8 0000cc 00 A 1 0 8 │ - [26] .interp PROGBITS 0000000000a00000 023000 00005a 00 A 0 0 8 │ + [ 1] .dynstr STRTAB 0000000000001aa0 001aa0 0011a8 00 A 0 0 1 │ + [ 2] .gnu.version VERSYM 0000000000002c48 002c48 0001e8 02 A 25 0 2 │ + [ 3] .gnu.version_r VERNEED 0000000000002e30 002e30 000100 00 A 1 4 8 │ + [ 4] .rela.dyn RELA 0000000000002f30 002f30 0008a0 18 A 25 0 8 │ + [ 5] .rela.plt RELA 00000000000037d0 0037d0 001488 18 AI 25 18 8 │ + [ 6] .init PROGBITS 0000000000004c58 004c58 000017 00 AX 0 0 4 │ + [ 7] .plt PROGBITS 0000000000004c70 004c70 000dc0 10 AX 0 0 16 │ + [ 8] .plt.got PROGBITS 0000000000005a30 005a30 000008 08 AX 0 0 8 │ + [ 9] .text PROGBITS 0000000000005a40 005a40 010cd2 00 AX 0 0 16 │ + [10] .fini PROGBITS 0000000000016714 016714 000009 00 AX 0 0 4 │ + [11] .rodata PROGBITS 0000000000016720 016720 0050b2 00 A 0 0 16 │ + [12] .eh_frame_hdr PROGBITS 000000000001b7d4 01b7d4 0004fc 00 A 0 0 4 │ + [13] .eh_frame PROGBITS 000000000001bcd0 01bcd0 002020 00 A 0 0 8 │ + [14] .init_array INIT_ARRAY 000000000021e450 01e450 000008 08 WA 0 0 8 │ + [15] .fini_array FINI_ARRAY 000000000021e458 01e458 000008 08 WA 0 0 8 │ + [16] .data.rel.ro PROGBITS 000000000021e460 01e460 000218 00 WA 0 0 32 │ + [17] .dynamic DYNAMIC 000000000021e678 01e678 000260 10 WA 1 0 8 │ + [18] .got PROGBITS 000000000021e8d8 01e8d8 000720 08 WA 0 0 8 │ + [19] .data PROGBITS 000000000021f000 01f000 000128 00 WA 0 0 32 │ + [20] .bss NOBITS 000000000021f140 01f128 0006e8 00 WA 0 0 32 │ + [21] .shstrtab STRTAB 0000000000000000 01f128 0000f2 00 0 0 1 │ + [22] .note.ABI-tag NOTE 0000000000400060 020060 000020 00 A 0 0 4 │ + [23] .note.gnu.build-id NOTE 0000000000400080 020080 000024 00 A 0 0 4 │ + [24] .gnu.hash GNU_HASH 0000000000600000 021000 0000cc 00 A 25 0 8 │ + [25] .dynsym DYNSYM 0000000000a00000 023000 0016e0 18 A 1 1 8 │ + [26] .interp PROGBITS 0000000000a016e0 0246e0 00005a 00 A 0 0 8 │ Key to Flags: │ W (write), A (alloc), X (execute), M (merge), S (strings), I (info), │ L (link order), O (extra OS processing required), G (group), T (TLS), │ C (compressed), x (unknown), o (OS specific), E (exclude), │ D (mbind), l (large), p (processor specific) ├── readelf --wide --symbols {} │ @@ -223,25 +223,25 @@ │ 219: 0000000000000000 0 FUNC GLOBAL DEFAULT UND ext2fs_get_device_size2 │ 220: 0000000000000000 0 FUNC GLOBAL DEFAULT UND ext2fs_free_blocks_count │ 221: 0000000000000000 0 FUNC GLOBAL DEFAULT UND ext2fs_inode_table_loc │ 222: 0000000000000000 0 FUNC GLOBAL DEFAULT UND ext2fs_badblocks_list_iterate │ 223: 0000000000000000 0 FUNC GLOBAL DEFAULT UND __ctype_b_loc@GLIBC_2.3 (8) │ 224: 0000000000000000 0 FUNC GLOBAL DEFAULT UND ext2fs_open_file │ 225: 0000000000000000 0 FUNC GLOBAL DEFAULT UND __sprintf_chk@GLIBC_2.3.4 (4) │ - 226: 000000000021f128 0 NOTYPE GLOBAL DEFAULT 20 _edata │ - 227: 000000000021f1b0 8 OBJECT GLOBAL DEFAULT 21 _et_list │ - 228: 000000000021f140 8 OBJECT GLOBAL DEFAULT 21 stdout@GLIBC_2.2.5 (3) │ - 229: 00000000000166e4 0 FUNC GLOBAL DEFAULT 11 _fini │ - 230: 000000000021e660 24 OBJECT GLOBAL DEFAULT 17 et_ext2_error_table │ - 231: 000000000021f168 8 OBJECT GLOBAL DEFAULT 21 optarg@GLIBC_2.2.5 (3) │ - 232: 0000000000004c20 0 FUNC GLOBAL DEFAULT 7 _init │ - 233: 000000000021f158 4 OBJECT GLOBAL DEFAULT 21 optind@GLIBC_2.2.5 (3) │ - 234: 000000000021f190 8 OBJECT GLOBAL DEFAULT 21 unix_io_manager │ - 235: 000000000021f1a8 8 OBJECT GLOBAL DEFAULT 21 test_io_backing_manager │ - 236: 000000000021f128 0 NOTYPE GLOBAL DEFAULT 21 __bss_start │ - 237: 000000000021f150 8 OBJECT GLOBAL DEFAULT 21 stdin@GLIBC_2.2.5 (3) │ - 238: 000000000021f1c0 8 OBJECT GLOBAL DEFAULT 21 stderr@GLIBC_2.2.5 (3) │ - 239: 000000000021f170 24 OBJECT GLOBAL DEFAULT 21 ext2fs_numeric_progress_ops │ - 240: 000000000021f828 0 NOTYPE GLOBAL DEFAULT 21 _end │ - 241: 000000000021f1a0 8 OBJECT GLOBAL DEFAULT 21 undo_io_manager │ - 242: 000000000021f160 8 OBJECT GLOBAL DEFAULT 21 test_io_manager │ - 243: 000000000021f1b8 8 OBJECT GLOBAL DEFAULT 21 sparse_io_manager │ + 226: 000000000021f128 0 NOTYPE GLOBAL DEFAULT 19 _edata │ + 227: 000000000021f1b0 8 OBJECT GLOBAL DEFAULT 20 _et_list │ + 228: 000000000021f140 8 OBJECT GLOBAL DEFAULT 20 stdout@GLIBC_2.2.5 (3) │ + 229: 0000000000016714 0 FUNC GLOBAL DEFAULT 10 _fini │ + 230: 000000000021e660 24 OBJECT GLOBAL DEFAULT 16 et_ext2_error_table │ + 231: 000000000021f168 8 OBJECT GLOBAL DEFAULT 20 optarg@GLIBC_2.2.5 (3) │ + 232: 0000000000004c58 0 FUNC GLOBAL DEFAULT 6 _init │ + 233: 000000000021f158 4 OBJECT GLOBAL DEFAULT 20 optind@GLIBC_2.2.5 (3) │ + 234: 000000000021f190 8 OBJECT GLOBAL DEFAULT 20 unix_io_manager │ + 235: 000000000021f1a8 8 OBJECT GLOBAL DEFAULT 20 test_io_backing_manager │ + 236: 000000000021f128 0 NOTYPE GLOBAL DEFAULT 20 __bss_start │ + 237: 000000000021f150 8 OBJECT GLOBAL DEFAULT 20 stdin@GLIBC_2.2.5 (3) │ + 238: 000000000021f1c0 8 OBJECT GLOBAL DEFAULT 20 stderr@GLIBC_2.2.5 (3) │ + 239: 000000000021f170 24 OBJECT GLOBAL DEFAULT 20 ext2fs_numeric_progress_ops │ + 240: 000000000021f828 0 NOTYPE GLOBAL DEFAULT 20 _end │ + 241: 000000000021f1a0 8 OBJECT GLOBAL DEFAULT 20 undo_io_manager │ + 242: 000000000021f160 8 OBJECT GLOBAL DEFAULT 20 test_io_manager │ + 243: 000000000021f1b8 8 OBJECT GLOBAL DEFAULT 20 sparse_io_manager ├── readelf --wide --relocs {} │ @@ -1,83 +1,83 @@ │ │ -Relocation section '.rela.dyn' at offset 0x2ef8 contains 92 entries: │ +Relocation section '.rela.dyn' at offset 0x2f30 contains 92 entries: │ Offset Info Type Symbol's Value Symbol's Name + Addend │ -000000000021e450 0000000000000008 R_X86_64_RELATIVE 8430 │ -000000000021e458 0000000000000008 R_X86_64_RELATIVE 83f0 │ -000000000021e460 0000000000000008 R_X86_64_RELATIVE 1aaad │ -000000000021e468 0000000000000008 R_X86_64_RELATIVE 1a9c1 │ -000000000021e470 0000000000000008 R_X86_64_RELATIVE 1a9c4 │ -000000000021e478 0000000000000008 R_X86_64_RELATIVE 1a9ca │ -000000000021e480 0000000000000008 R_X86_64_RELATIVE 1aa17 │ -000000000021e488 0000000000000008 R_X86_64_RELATIVE 1a9ce │ -000000000021e4a0 0000000000000008 R_X86_64_RELATIVE 190bd │ -000000000021e4a8 0000000000000008 R_X86_64_RELATIVE 1a9d2 │ -000000000021e4b0 0000000000000008 R_X86_64_RELATIVE 1a9d6 │ -000000000021e4b8 0000000000000008 R_X86_64_RELATIVE 1aa87 │ -000000000021e4c0 0000000000000008 R_X86_64_RELATIVE 18ebd │ -000000000021e4c8 0000000000000008 R_X86_64_RELATIVE 1904e │ +000000000021e450 0000000000000008 R_X86_64_RELATIVE 8460 │ +000000000021e458 0000000000000008 R_X86_64_RELATIVE 8420 │ +000000000021e460 0000000000000008 R_X86_64_RELATIVE 1aaed │ +000000000021e468 0000000000000008 R_X86_64_RELATIVE 1aa01 │ +000000000021e470 0000000000000008 R_X86_64_RELATIVE 1aa04 │ +000000000021e478 0000000000000008 R_X86_64_RELATIVE 1aa0a │ +000000000021e480 0000000000000008 R_X86_64_RELATIVE 1aa57 │ +000000000021e488 0000000000000008 R_X86_64_RELATIVE 1aa0e │ +000000000021e4a0 0000000000000008 R_X86_64_RELATIVE 19100 │ +000000000021e4a8 0000000000000008 R_X86_64_RELATIVE 1aa12 │ +000000000021e4b0 0000000000000008 R_X86_64_RELATIVE 1aa16 │ +000000000021e4b8 0000000000000008 R_X86_64_RELATIVE 1aac7 │ +000000000021e4c0 0000000000000008 R_X86_64_RELATIVE 18f00 │ +000000000021e4c8 0000000000000008 R_X86_64_RELATIVE 19091 ...
Well, after typing all of that I've realized that it might be related to: https://github.com/NixOS/patchelf/issues/244 as we had similar issue before: https://git.openembedded.org/openembedded-core/commit/?id=cccd4bcaf381c2729adc000381bd89906003e72a On Thu, Jun 15, 2023 at 5:06 PM Martin Jansa via lists.openembedded.org <Martin.Jansa=gmail.com@lists.openembedded.org> wrote: > On Sun, Jun 4, 2023 at 10:34 PM Martin Jansa <martin.jansa@gmail.com> > wrote: > >> > It might be useful to delay backporting this for a bit longer (and I'll >> soon confirm if 3.10 was already affected or not). >> >> I can confirm that 3.10 is also affected. >> >> I did build with >> https://git.openembedded.org/openembedded-core-contrib/commit/?h=stable/kirkstone-nut&id=75e3870ea124321b81f856c4e5170c891f0bd050 >> and mkfs.ext4 didn't segfault. With uninative 3.10 from >> https://git.openembedded.org/openembedded-core-contrib/commit/?h=stable/kirkstone-nut&id=680eecc3ae09ea5585e46ff0e67f7c6704221806 >> it segfaults again. >> >> I haven't reproduced the gdk-pixbuf-native issue with buildtools-extended >> 4.0.6 with kirkstone yet, so that one might be only with nanbield, but >> mkfs.ext4 segfaults seem related to uninative for sure. Will debug more >> when I get access. >> > > I got access to one of the servers where this is still reproducible, but > the real root cause is still a mystery for me. > > I'll share what I've found so far in case someone else is seeing the same > or can spot what I'm missing. > > It's reproducible on ubuntu-18.06, but not everywhere, I've tried older > docker image and latest 18.04 on my host and both work fine. I've tried to > narrow the difference from the 18.04 image where it is failing but without > any luck. > > The same happens with nanbield (where I've noticed it first), but now the > same happens with mickledore and kirkstone. > > No special config needed, default poky configuration reproduces the same. > > Easiest test is to build just e2fsprogs-native and try to execute > mkfs.ext4.real, if it immediately segfaults it's a "bad" binary, if it > complains about missing e2* symbols or BLKID_2_36 symbol missing > in libblkid.so.1, then it's "good" binary and works as long as you let it > load the right libraries from native sysroot. > > Surprisingly whatever breaks the binary happens between "image" and > "sysroot-destdir" so first I was suspecting strip from host's binutils, but > downgrading binutils in different 18.04 system to match the one where it is > failing didn't reproduce it. Downgrading libgcc1 and some other packages to > match didn't reproduce it as well. > > Next I've tried to inhibit strip completely (assuming it will be usable > work around to keep the binary in sysroot-destdir working with new > uninative and narrow down what I need to debug), but surprisingly now both > the binary in "image" and "sysroot-destdir" fail the same. > > Strace doesn't show anything useful, other than segfault triggered very > early: > > $ strace > tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/sysroot-destdir/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real > execve("tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/sysroot-destdir/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real", > ["tmp/work/x86_64-linux/e2fsprogs-"...], 0x7fffbec05ac0 /* 55 vars */) = 0 > brk(NULL) = 0x55b91bc0e000 > arch_prctl(0x3001 /* ARCH_??? */, 0x7ffc146b5340) = -1 EINVAL (Invalid > argument) > --- SIGSEGV {si_signo=SIGSEGV, si_code=SEGV_MAPERR, > si_addr=0x55b91ba016e0} --- > +++ killed by SIGSEGV (core dumped) +++ > Segmentation fault (core dumped) > > $ strace > tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/image/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real > 2>&1 | head > execve("tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/image/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real", > ["tmp/work/x86_64-linux/e2fsprogs-"...], 0x7ffd387a7f60 /* 55 vars */) = 0 > brk(NULL) = 0x5641b084f000 > arch_prctl(0x3001 /* ARCH_??? */, 0x7ffe271263f0) = -1 EINVAL (Invalid > argument) > --- SIGSEGV {si_signo=SIGSEGV, si_code=SEGV_MAPERR, > si_addr=0x5641b00016e0} --- > +++ killed by SIGSEGV (core dumped) +++ > > While "good" binary: > $ strace > tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/image/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real > | head > execve("tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/image/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real", > ["tmp/work/x86_64-linux/e2fsprogs-"...], 0x7ffd9f569ae0 /* 55 vars */) = 0 > brk(NULL) = 0x5569fad4e000 > arch_prctl(0x3001 /* ARCH_??? */, 0x7ffe1d8ce2f0) = -1 EINVAL (Invalid > argument) > mmap(NULL, 8192, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = > 0x7f210b337000 > access("/etc/ld.so.preload", R_OK) = -1 ENOENT (No such file or > directory) > openat(AT_FDCWD, > "/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/usr/lib/glibc-hwcaps/x86-64-v4/libext2fs.so.2", > O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) > newfstatat(AT_FDCWD, > "/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/usr/lib/glibc-hwcaps/x86-64-v4", > 0x7ffe1d8cd520, 0) = -1 ENOENT (No such file or directory) > > The diffoscope output shows that dynsym section is second to last in "bad" > binary and first in "good" binary, which might relate to the warnings shown > by gdb: > > Reading symbols from ./recipe-sysroot-native/sbin/mkfs.ext4.real... > > warning: Loadable section ".dynsym" outside of ELF segments > > warning: Loadable section ".interp" outside of ELF segments > (No debugging symbols found in ./recipe-sysroot-native/sbin/mkfs.ext4.real) > (gdb) r > Starting program: > /data001/jansa/build-webos/BUILD/work/qemux86_64-webos-linux/webos-image/1.0-r3normal1/recipe-sysroot-native/sbin/mkfs.ext4.real > > warning: Error disabling address space randomization: Operation not > permitted > BFD: > /data001/jansa/build-webos/BUILD/sysroots-uninative/x86_64-linux/lib/ld-linux-x86-64.so.2: > unknown type [0x13] section `.relr.dyn' > warning: > `/data001/jansa/build-webos/BUILD/sysroots-uninative/x86_64-linux/lib/ld-linux-x86-64.so.2': > Shared library architecture unknown is not compatible with target > architecture i386:x86-64. > warning: > `/data001/jansa/build-webos/BUILD/sysroots-uninative/x86_64-linux/lib/ld-linux-x86-64.so.2': > Shared library architecture unknown is not compatible with target > architecture i386:x86-64. > > Program received signal SIGSEGV, Segmentation fault. > 0x00007f08c9d9232a in strcmp () from > /data001/jansa/build-webos/BUILD/sysroots-uninative/x86_64-linux/lib/ld-linux-x86-64.so.2 > > And ldd also says that the "bad" binary isn't dynamic executable while > "good" on is (objdump -x works for both of them): > > the "bad" one: > ldd > tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/sysroot-destdir/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real > not a dynamic executable > > and the "good" one: > ldd > tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/image/home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/sbin/mkfs.ext4.real > > linux-vdso.so.1 (0x00007ffce83fc000) > libext2fs.so.2 => /lib/x86_64-linux-gnu/libext2fs.so.2 (0x00007fd1ba5a0000) > libcom_err.so.2 => /lib/x86_64-linux-gnu/libcom_err.so.2 > (0x00007fd1ba39c000) > libblkid.so.1 => > /home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/usr/lib/libblkid.so.1 > (0x00007fd1ba149000) > libuuid.so.1 => > /home/worker/poky-kirkstone-jansa/build/tmp/work/x86_64-linux/e2fsprogs-native/1.46.5-r0/recipe-sysroot-native/usr/lib/libuuid.so.1 > (0x00007fd1b9f42000) > libe2p.so.2 => /lib/x86_64-linux-gnu/libe2p.so.2 (0x00007fd1b9d39000) > libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 > (0x00007fd1b9b1a000) > libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007fd1b9729000) > /home/worker/poky-kirkstone-jansa/build/tmp/sysroots-uninative/x86_64-linux/lib/ld-linux-x86-64.so.2 > => /lib64/ld-linux-x86-64.so.2 (0x00007fd1baa0f000) > > Once the "bad" binary is built (with either uninative 4.0 or 3.9) it will > always fail with uninative 4.0 even when reused from sstate on different > server (if it's the server producing "good" mkfs.ext4.real binaries, then > just rebuilding e2fsprogs-native will fix it). > > Does this make any sense to anyone? I'll continue to look into it, but if > you have any hints, please let me know. > > I know 18.04 is EOL and nobody should be using it, I'm trying to get our > build infra upgraded for very long time. > > The diffoscope output (mkfs.4.0-18.04-sstate is the "bad" binary from > sstate while mkfs.4.0-18.04-rebuilt is the "good" one). > > --- mkfs.4.0-18.04-rebuilt > +++ mkfs.4.0-18.04-sstate > │┄ File has been modified after NT_GNU_BUILD_ID has been applied. > ├── readelf --wide --file-header {} > │ @@ -4,17 +4,17 @@ > │ Data: 2's complement, little endian > │ Version: 1 (current) > │ OS/ABI: UNIX - System V > │ ABI Version: 0 > │ Type: DYN (Position-Independent > Executable file) > │ Machine: Advanced Micro Devices X86-64 > │ Version: 0x1 > │ - Entry point address: 0x8350 > │ + Entry point address: 0x8380 > │ Start of program headers: 64 (bytes into file) > │ Start of section headers: 127520 (bytes into file) > │ Flags: 0x0 > │ Size of this header: 64 (bytes) > │ Size of program headers: 56 (bytes) > │ Number of program headers: 13 > │ Size of section headers: 64 (bytes) > │ Number of section headers: 27 > │ - Section header string table index: 22 > │ + Section header string table index: 21 > ├── readelf --wide --program-header {} > │ @@ -1,37 +1,37 @@ > │ > │ Elf file type is DYN (Position-Independent Executable file) > │ -Entry point 0x8350 > │ +Entry point 0x8380 > │ There are 13 program headers, starting at offset 64 > │ > │ Program Headers: > │ Type Offset VirtAddr PhysAddr FileSiz > MemSiz Flg Align > │ PHDR 0x000040 0x0000000000000040 0x0000000000000040 > 0x0002d8 0x0002d8 R 0x8 > │ - LOAD 0x000000 0x0000000000000000 0x0000000000000000 > 0x01dcb0 0x01dcb0 R E 0x200000 > │ + LOAD 0x000000 0x0000000000000000 0x0000000000000000 > 0x01dcf0 0x01dcf0 R E 0x200000 > │ GNU_STACK 0x000000 0x0000000000000000 0x0000000000000000 > 0x000000 0x000000 RW 0x10 > │ - GNU_EH_FRAME 0x01b794 0x000000000001b794 0x000000000001b794 > 0x0004fc 0x0004fc R 0x4 > │ + GNU_EH_FRAME 0x01b7d4 0x000000000001b7d4 0x000000000001b7d4 > 0x0004fc 0x0004fc R 0x4 > │ LOAD 0x01e450 0x000000000021e450 0x000000000021e450 > 0x000cd8 0x0013d8 RW 0x200000 > │ GNU_RELRO 0x01e450 0x000000000021e450 0x000000000021e450 > 0x000bb0 0x000bb0 R 0x1 > │ DYNAMIC 0x01e678 0x000000000021e678 0x000000000021e678 > 0x000260 0x000260 RW 0x8 > │ - LOAD 0x020000 0x0000000000400000 0x0000000000400000 > 0x000178 0x000178 RW 0x200000 > │ + LOAD 0x020000 0x0000000000400000 0x0000000000400000 > 0x0000a8 0x0000a8 RW 0x200000 > │ NOTE 0x020060 0x0000000000400060 0x0000000000400060 > 0x000020 0x000020 R 0x4 > │ NOTE 0x020080 0x0000000000400080 0x0000000000400080 > 0x000024 0x000024 R 0x4 > │ - LOAD 0x021000 0x0000000000600000 0x0000000000600000 > 0x001060 0x001060 RW 0x200000 > │ - INTERP 0x023000 0x0000000000a00000 0x0000000000a00000 > 0x00005a 0x00005a R 0x1 > │ + LOAD 0x021000 0x0000000000600000 0x0000000000600000 > 0x000130 0x000130 RW 0x200000 > │ + LOAD 0x022000 0x0000000000800000 0x0000000000800000 > 0x002740 0x002740 RW 0x200000 > │ + INTERP 0x0246e0 0x0000000000a016e0 0x0000000000a016e0 > 0x00005a 0x00005a R 0x1 > │ [Requesting program interpreter: > /data001/jansa/build-webos/BUILD/sysroots-uninative/x86_64-linux/lib/ld-linux-x86-64.so.2] > │ - LOAD 0x023000 0x0000000000a00000 0x0000000000a00000 > 0x000060 0x000060 RW 0x200000 > │ > │ Section to Segment mapping: > │ Segment Sections... > │ 00 > │ - 01 .dynsym .dynstr .gnu.version .gnu.version_r .rela.dyn > .rela.plt .init .plt .plt.got .text .fini .rodata .eh_frame_hdr .eh_frame > │ + 01 .dynstr .gnu.version .gnu.version_r .rela.dyn .rela.plt .init > .plt .plt.got .text .fini .rodata .eh_frame_hdr .eh_frame > │ 02 > │ 03 .eh_frame_hdr > │ 04 .init_array .fini_array .data.rel.ro .dynamic .got .data .bss > │ 05 .init_array .fini_array .data.rel.ro .dynamic .got > │ 06 .dynamic > │ - 07 .note.ABI-tag .note.gnu.build-id .gnu.hash > │ + 07 .note.ABI-tag .note.gnu.build-id > │ 08 .note.ABI-tag > │ 09 .note.gnu.build-id > │ - 10 > │ - 11 .interp > │ + 10 .gnu.hash > │ + 11 > │ 12 .interp > ├── readelf --wide --sections {} > │ @@ -1,36 +1,36 @@ > │ There are 27 section headers, starting at offset 0x1f220: > │ > │ Section Headers: > │ [Nr] Name Type Address Off Size > ES Flg Lk Inf Al > │ [ 0] NULL 0000000000000000 000000 000000 > 00 0 0 0 > │ - [ 1] .dynsym DYNSYM 00000000000003a8 0003a8 0016e0 > 18 A 2 1 8 > │ - [ 2] .dynstr STRTAB 0000000000001a88 001a88 001182 > 00 A 0 0 1 > │ - [ 3] .gnu.version VERSYM 0000000000002c0a 002c0a 0001e8 > 02 A 1 0 2 > │ - [ 4] .gnu.version_r VERNEED 0000000000002df8 002df8 000100 > 00 A 2 4 8 > │ - [ 5] .rela.dyn RELA 0000000000002ef8 002ef8 0008a0 > 18 A 1 0 8 > │ - [ 6] .rela.plt RELA 0000000000003798 003798 001488 > 18 AI 1 19 8 > │ - [ 7] .init PROGBITS 0000000000004c20 004c20 000017 > 00 AX 0 0 4 > │ - [ 8] .plt PROGBITS 0000000000004c40 004c40 000dc0 > 10 AX 0 0 16 > │ - [ 9] .plt.got PROGBITS 0000000000005a00 005a00 000008 > 08 AX 0 0 8 > │ - [10] .text PROGBITS 0000000000005a10 005a10 010cd2 > 00 AX 0 0 16 > │ - [11] .fini PROGBITS 00000000000166e4 0166e4 000009 > 00 AX 0 0 4 > │ - [12] .rodata PROGBITS 00000000000166f0 0166f0 0050a2 > 00 A 0 0 16 > │ - [13] .eh_frame_hdr PROGBITS 000000000001b794 01b794 0004fc > 00 A 0 0 4 > │ - [14] .eh_frame PROGBITS 000000000001bc90 01bc90 002020 > 00 A 0 0 8 > │ - [15] .init_array INIT_ARRAY 000000000021e450 01e450 000008 > 08 WA 0 0 8 > │ - [16] .fini_array FINI_ARRAY 000000000021e458 01e458 000008 > 08 WA 0 0 8 > │ - [17] .data.rel.ro PROGBITS 000000000021e460 01e460 > 000218 00 WA 0 0 32 > │ - [18] .dynamic DYNAMIC 000000000021e678 01e678 000260 > 10 WA 2 0 8 > │ - [19] .got PROGBITS 000000000021e8d8 01e8d8 000720 > 08 WA 0 0 8 > │ - [20] .data PROGBITS 000000000021f000 01f000 000128 > 00 WA 0 0 32 > │ - [21] .bss NOBITS 000000000021f140 01f128 0006e8 > 00 WA 0 0 32 > │ - [22] .shstrtab STRTAB 0000000000000000 01f128 0000f2 > 00 0 0 1 > │ - [23] .note.ABI-tag NOTE 0000000000400060 020060 000020 > 00 A 0 0 4 > │ - [24] .note.gnu.build-id NOTE 0000000000400080 020080 > 000024 00 A 0 0 4 > │ - [25] .gnu.hash GNU_HASH 00000000004000a8 0200a8 0000cc > 00 A 1 0 8 > │ - [26] .interp PROGBITS 0000000000a00000 023000 00005a > 00 A 0 0 8 > │ + [ 1] .dynstr STRTAB 0000000000001aa0 001aa0 0011a8 > 00 A 0 0 1 > │ + [ 2] .gnu.version VERSYM 0000000000002c48 002c48 0001e8 > 02 A 25 0 2 > │ + [ 3] .gnu.version_r VERNEED 0000000000002e30 002e30 000100 > 00 A 1 4 8 > │ + [ 4] .rela.dyn RELA 0000000000002f30 002f30 0008a0 > 18 A 25 0 8 > │ + [ 5] .rela.plt RELA 00000000000037d0 0037d0 001488 > 18 AI 25 18 8 > │ + [ 6] .init PROGBITS 0000000000004c58 004c58 000017 > 00 AX 0 0 4 > │ + [ 7] .plt PROGBITS 0000000000004c70 004c70 000dc0 > 10 AX 0 0 16 > │ + [ 8] .plt.got PROGBITS 0000000000005a30 005a30 000008 > 08 AX 0 0 8 > │ + [ 9] .text PROGBITS 0000000000005a40 005a40 010cd2 > 00 AX 0 0 16 > │ + [10] .fini PROGBITS 0000000000016714 016714 000009 > 00 AX 0 0 4 > │ + [11] .rodata PROGBITS 0000000000016720 016720 0050b2 > 00 A 0 0 16 > │ + [12] .eh_frame_hdr PROGBITS 000000000001b7d4 01b7d4 0004fc > 00 A 0 0 4 > │ + [13] .eh_frame PROGBITS 000000000001bcd0 01bcd0 002020 > 00 A 0 0 8 > │ + [14] .init_array INIT_ARRAY 000000000021e450 01e450 000008 > 08 WA 0 0 8 > │ + [15] .fini_array FINI_ARRAY 000000000021e458 01e458 000008 > 08 WA 0 0 8 > │ + [16] .data.rel.ro PROGBITS 000000000021e460 01e460 > 000218 00 WA 0 0 32 > │ + [17] .dynamic DYNAMIC 000000000021e678 01e678 000260 > 10 WA 1 0 8 > │ + [18] .got PROGBITS 000000000021e8d8 01e8d8 000720 > 08 WA 0 0 8 > │ + [19] .data PROGBITS 000000000021f000 01f000 000128 > 00 WA 0 0 32 > │ + [20] .bss NOBITS 000000000021f140 01f128 0006e8 > 00 WA 0 0 32 > │ + [21] .shstrtab STRTAB 0000000000000000 01f128 0000f2 > 00 0 0 1 > │ + [22] .note.ABI-tag NOTE 0000000000400060 020060 000020 > 00 A 0 0 4 > │ + [23] .note.gnu.build-id NOTE 0000000000400080 020080 > 000024 00 A 0 0 4 > │ + [24] .gnu.hash GNU_HASH 0000000000600000 021000 0000cc > 00 A 25 0 8 > │ + [25] .dynsym DYNSYM 0000000000a00000 023000 0016e0 > 18 A 1 1 8 > │ + [26] .interp PROGBITS 0000000000a016e0 0246e0 00005a > 00 A 0 0 8 > │ Key to Flags: > │ W (write), A (alloc), X (execute), M (merge), S (strings), I (info), > │ L (link order), O (extra OS processing required), G (group), T (TLS), > │ C (compressed), x (unknown), o (OS specific), E (exclude), > │ D (mbind), l (large), p (processor specific) > ├── readelf --wide --symbols {} > │ @@ -223,25 +223,25 @@ > │ 219: 0000000000000000 0 FUNC GLOBAL DEFAULT UND > ext2fs_get_device_size2 > │ 220: 0000000000000000 0 FUNC GLOBAL DEFAULT UND > ext2fs_free_blocks_count > │ 221: 0000000000000000 0 FUNC GLOBAL DEFAULT UND > ext2fs_inode_table_loc > │ 222: 0000000000000000 0 FUNC GLOBAL DEFAULT UND > ext2fs_badblocks_list_iterate > │ 223: 0000000000000000 0 FUNC GLOBAL DEFAULT UND > __ctype_b_loc@GLIBC_2.3 (8) > │ 224: 0000000000000000 0 FUNC GLOBAL DEFAULT UND > ext2fs_open_file > │ 225: 0000000000000000 0 FUNC GLOBAL DEFAULT UND > __sprintf_chk@GLIBC_2.3.4 (4) > │ - 226: 000000000021f128 0 NOTYPE GLOBAL DEFAULT 20 _edata > │ - 227: 000000000021f1b0 8 OBJECT GLOBAL DEFAULT 21 _et_list > │ - 228: 000000000021f140 8 OBJECT GLOBAL DEFAULT 21 > stdout@GLIBC_2.2.5 (3) > │ - 229: 00000000000166e4 0 FUNC GLOBAL DEFAULT 11 _fini > │ - 230: 000000000021e660 24 OBJECT GLOBAL DEFAULT 17 > et_ext2_error_table > │ - 231: 000000000021f168 8 OBJECT GLOBAL DEFAULT 21 > optarg@GLIBC_2.2.5 (3) > │ - 232: 0000000000004c20 0 FUNC GLOBAL DEFAULT 7 _init > │ - 233: 000000000021f158 4 OBJECT GLOBAL DEFAULT 21 > optind@GLIBC_2.2.5 (3) > │ - 234: 000000000021f190 8 OBJECT GLOBAL DEFAULT 21 > unix_io_manager > │ - 235: 000000000021f1a8 8 OBJECT GLOBAL DEFAULT 21 > test_io_backing_manager > │ - 236: 000000000021f128 0 NOTYPE GLOBAL DEFAULT 21 __bss_start > │ - 237: 000000000021f150 8 OBJECT GLOBAL DEFAULT 21 > stdin@GLIBC_2.2.5 (3) > │ - 238: 000000000021f1c0 8 OBJECT GLOBAL DEFAULT 21 > stderr@GLIBC_2.2.5 (3) > │ - 239: 000000000021f170 24 OBJECT GLOBAL DEFAULT 21 > ext2fs_numeric_progress_ops > │ - 240: 000000000021f828 0 NOTYPE GLOBAL DEFAULT 21 _end > │ - 241: 000000000021f1a0 8 OBJECT GLOBAL DEFAULT 21 > undo_io_manager > │ - 242: 000000000021f160 8 OBJECT GLOBAL DEFAULT 21 > test_io_manager > │ - 243: 000000000021f1b8 8 OBJECT GLOBAL DEFAULT 21 > sparse_io_manager > │ + 226: 000000000021f128 0 NOTYPE GLOBAL DEFAULT 19 _edata > │ + 227: 000000000021f1b0 8 OBJECT GLOBAL DEFAULT 20 _et_list > │ + 228: 000000000021f140 8 OBJECT GLOBAL DEFAULT 20 > stdout@GLIBC_2.2.5 (3) > │ + 229: 0000000000016714 0 FUNC GLOBAL DEFAULT 10 _fini > │ + 230: 000000000021e660 24 OBJECT GLOBAL DEFAULT 16 > et_ext2_error_table > │ + 231: 000000000021f168 8 OBJECT GLOBAL DEFAULT 20 > optarg@GLIBC_2.2.5 (3) > │ + 232: 0000000000004c58 0 FUNC GLOBAL DEFAULT 6 _init > │ + 233: 000000000021f158 4 OBJECT GLOBAL DEFAULT 20 > optind@GLIBC_2.2.5 (3) > │ + 234: 000000000021f190 8 OBJECT GLOBAL DEFAULT 20 > unix_io_manager > │ + 235: 000000000021f1a8 8 OBJECT GLOBAL DEFAULT 20 > test_io_backing_manager > │ + 236: 000000000021f128 0 NOTYPE GLOBAL DEFAULT 20 __bss_start > │ + 237: 000000000021f150 8 OBJECT GLOBAL DEFAULT 20 > stdin@GLIBC_2.2.5 (3) > │ + 238: 000000000021f1c0 8 OBJECT GLOBAL DEFAULT 20 > stderr@GLIBC_2.2.5 (3) > │ + 239: 000000000021f170 24 OBJECT GLOBAL DEFAULT 20 > ext2fs_numeric_progress_ops > │ + 240: 000000000021f828 0 NOTYPE GLOBAL DEFAULT 20 _end > │ + 241: 000000000021f1a0 8 OBJECT GLOBAL DEFAULT 20 > undo_io_manager > │ + 242: 000000000021f160 8 OBJECT GLOBAL DEFAULT 20 > test_io_manager > │ + 243: 000000000021f1b8 8 OBJECT GLOBAL DEFAULT 20 > sparse_io_manager > ├── readelf --wide --relocs {} > │ @@ -1,83 +1,83 @@ > │ > │ -Relocation section '.rela.dyn' at offset 0x2ef8 contains 92 entries: > │ +Relocation section '.rela.dyn' at offset 0x2f30 contains 92 entries: > │ Offset Info Type Symbol's > Value Symbol's Name + Addend > │ -000000000021e450 0000000000000008 R_X86_64_RELATIVE > 8430 > │ -000000000021e458 0000000000000008 R_X86_64_RELATIVE > 83f0 > │ -000000000021e460 0000000000000008 R_X86_64_RELATIVE > 1aaad > │ -000000000021e468 0000000000000008 R_X86_64_RELATIVE > 1a9c1 > │ -000000000021e470 0000000000000008 R_X86_64_RELATIVE > 1a9c4 > │ -000000000021e478 0000000000000008 R_X86_64_RELATIVE > 1a9ca > │ -000000000021e480 0000000000000008 R_X86_64_RELATIVE > 1aa17 > │ -000000000021e488 0000000000000008 R_X86_64_RELATIVE > 1a9ce > │ -000000000021e4a0 0000000000000008 R_X86_64_RELATIVE > 190bd > │ -000000000021e4a8 0000000000000008 R_X86_64_RELATIVE > 1a9d2 > │ -000000000021e4b0 0000000000000008 R_X86_64_RELATIVE > 1a9d6 > │ -000000000021e4b8 0000000000000008 R_X86_64_RELATIVE > 1aa87 > │ -000000000021e4c0 0000000000000008 R_X86_64_RELATIVE > 18ebd > │ -000000000021e4c8 0000000000000008 R_X86_64_RELATIVE > 1904e > │ +000000000021e450 0000000000000008 R_X86_64_RELATIVE > 8460 > │ +000000000021e458 0000000000000008 R_X86_64_RELATIVE > 8420 > │ +000000000021e460 0000000000000008 R_X86_64_RELATIVE > 1aaed > │ +000000000021e468 0000000000000008 R_X86_64_RELATIVE > 1aa01 > │ +000000000021e470 0000000000000008 R_X86_64_RELATIVE > 1aa04 > │ +000000000021e478 0000000000000008 R_X86_64_RELATIVE > 1aa0a > │ +000000000021e480 0000000000000008 R_X86_64_RELATIVE > 1aa57 > │ +000000000021e488 0000000000000008 R_X86_64_RELATIVE > 1aa0e > │ +000000000021e4a0 0000000000000008 R_X86_64_RELATIVE > 19100 > │ +000000000021e4a8 0000000000000008 R_X86_64_RELATIVE > 1aa12 > │ +000000000021e4b0 0000000000000008 R_X86_64_RELATIVE > 1aa16 > │ +000000000021e4b8 0000000000000008 R_X86_64_RELATIVE > 1aac7 > │ +000000000021e4c0 0000000000000008 R_X86_64_RELATIVE > 18f00 > │ +000000000021e4c8 0000000000000008 R_X86_64_RELATIVE > 19091 > ... > > -=-=-=-=-=-=-=-=-=-=-=- > Links: You receive all messages sent to this group. > View/Reply Online (#182862): > https://lists.openembedded.org/g/openembedded-core/message/182862 > Mute This Topic: https://lists.openembedded.org/mt/99550848/3617156 > Group Owner: openembedded-core+owner@lists.openembedded.org > Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [ > Martin.Jansa@gmail.com] > -=-=-=-=-=-=-=-=-=-=-=- > >
diff --git a/meta/conf/distro/include/yocto-uninative.inc b/meta/conf/distro/include/yocto-uninative.inc index f3c05b7e20..ad4816a1f3 100644 --- a/meta/conf/distro/include/yocto-uninative.inc +++ b/meta/conf/distro/include/yocto-uninative.inc @@ -7,9 +7,9 @@ # UNINATIVE_MAXGLIBCVERSION = "2.37" -UNINATIVE_VERSION = "3.10" +UNINATIVE_VERSION = "4.0" UNINATIVE_URL ?= "http://downloads.yoctoproject.org/releases/uninative/${UNINATIVE_VERSION}/" -UNINATIVE_CHECKSUM[aarch64] ?= "b11f7dff0ff5e3c971f22e89a105efb0f8d2006fd71d1ea3e5da771e416b0e01" -UNINATIVE_CHECKSUM[i686] ?= "ad7daa7fca7693ca07eea5279f830333530171d357759b380954a25254edd8dd" -UNINATIVE_CHECKSUM[x86_64] ?= "ce8ff7e3247da312fbf84008f415bc1b4d113ac9fe4ec91581e4a20cd8b475d3" +UNINATIVE_CHECKSUM[aarch64] ?= "7baa8418a302df52e00916193b0a04f318356d9d2670c9a2bce3e966efefd738" +UNINATIVE_CHECKSUM[i686] ?= "83114d36883d43a521e280742b9849bf85d039b2f83d8e21d480659babe75ee8" +UNINATIVE_CHECKSUM[x86_64] ?= "fd75b2a1a67a10f6b7d65afb7d0f3e71a63b0038e428f34dfe420bb37716558a"