[OE-core] [PATCH 1/2] glibc: split glibc and libcrypt
According to Fedora 28 change
(https://fedoraproject.org/wiki/Changes/Replace_glibc_libcrypt_with_libxcrypt)
and probably glibc upstream change, the nativesdk provides libcrypt from
libxcrypt instead of pproviding from glibc itself.
The purpose is to provide one uninative file compatible with Fedora >= 28 and
other distros. libxcrypt is backwards compatible with libcrypt but not the
converse.
Signed-off-by: Charles-Antoine Couret <charles-antoine.cou...@essensium.com>
---
meta/recipes-core/glibc/glibc-package.inc | 1 +
.../0031-nativesdk-deprecate-libcrypt.patch | 418 ++
meta/recipes-core/glibc/glibc_2.27.bb | 1 +
.../recipes-core/libxcrypt/libxcrypt_4.0.0.bb | 24 +
meta/recipes-core/meta/uninative-tarball.bb | 1 +
5 files changed, 445 insertions(+)
create mode 100644
meta/recipes-core/glibc/glibc/0031-nativesdk-deprecate-libcrypt.patch
create mode 100644 meta/recipes-core/libxcrypt/libxcrypt_4.0.0.bb
diff --git a/meta/recipes-core/glibc/glibc-package.inc
b/meta/recipes-core/glibc/glibc-package.inc
index ff45dfe35a..1b54dd8880 100644
--- a/meta/recipes-core/glibc/glibc-package.inc
+++ b/meta/recipes-core/glibc/glibc-package.inc
@@ -16,6 +16,7 @@ RPROVIDES_${PN}-pcprofile = "eglibc-pcprofile"
RPROVIDES_${PN}-dbg = "eglibc-dbg"
libc_baselibs = "${base_libdir}/libcrypt*.so.* ${base_libdir}/libcrypt-*.so
${base_libdir}/libc.so.* ${base_libdir}/libc-*.so ${base_libdir}/libm*.so.*
${base_libdir}/libm-*.so ${base_libdir}/libmvec-*.so ${base_libdir}/ld*.so.*
${base_libdir}/ld-*.so ${base_libdir}/libpthread*.so.*
${base_libdir}/libpthread-*.so ${base_libdir}/libresolv*.so.*
${base_libdir}/libresolv-*.so ${base_libdir}/librt*.so.*
${base_libdir}/librt-*.so ${base_libdir}/libutil*.so.*
${base_libdir}/libutil-*.so ${base_libdir}/libnsl*.so.*
${base_libdir}/libnsl-*.so ${base_libdir}/libnss_files*.so.*
${base_libdir}/libnss_files-*.so ${base_libdir}/libnss_compat*.so.*
${base_libdir}/libnss_compat-*.so ${base_libdir}/libnss_dns*.so.*
${base_libdir}/libnss_dns-*.so ${base_libdir}/libdl*.so.*
${base_libdir}/libdl-*.so ${base_libdir}/libanl*.so.*
${base_libdir}/libanl-*.so ${base_libdir}/libBrokenLocale*.so.*
${base_libdir}/libBrokenLocale-*.so"
libc_baselibs_append_aarch64 = " /lib/ld-linux-aarch64*.so.1"
+libc_baselibs_remove_class_sdk = "${base_libdir}/libcrypt*.so.*
${base_libdir}/libcrypt-*.so"
INSANE_SKIP_${PN}_append_aarch64 = " libdir"
FILES_${PN} = "${libc_baselibs} ${libexecdir}/* ${base_sbindir}/ldconfig
${sysconfdir}/ld.so.conf"
diff --git
a/meta/recipes-core/glibc/glibc/0031-nativesdk-deprecate-libcrypt.patch
b/meta/recipes-core/glibc/glibc/0031-nativesdk-deprecate-libcrypt.patch
new file mode 100644
index 00..6b34c70829
--- /dev/null
+++ b/meta/recipes-core/glibc/glibc/0031-nativesdk-deprecate-libcrypt.patch
@@ -0,0 +1,418 @@
+Background information:
+
+https://sourceware.org/ml/libc-alpha/2017-08/msg01257.html
+https://fedoraproject.org/wiki/Changes/Replace_glibc_libcrypt_with_libxcrypt
+
+Upstream-Status: Submitted [libc-al...@sourceware.org]
+
+From: Zack Weinberg <za...@panix.com>
+Subject: [PATCH] Deprecate libcrypt and don't build it by default
+
+Back in June, Björn Esser proposed to add OpenBSD-compatible bcrypt
+support to our implementation of crypt(3), and Zack Weinberg replied
+that it might actually make more sense to _remove_ libcrypt from
+glibc, freeing up libcrypt.so.1 and crypt.h to be provided by a
+separate project that could move faster. (For instance, libxcrypt:
+https://github.com/besser82/libxcrypt)
+
+This patch disables build and installation of libcrypt by default. It
+can be re-enabled with --enable-obsolete-crypt to configure. Unlike
+libnsl, we do *not* install a runtime shared library; that's left to
+the replacement. (Unlike the SunRPC situation, I think we can
+probably drop this code altogether in a release or two.)
+
+The function prototypes for crypt and encrypt are removed from
+unistd.h, and the function prototype for setkey is removed from
+stdlib.h; they do *not* come back with --enable-obsolete-crypt. This
+means glibc no longer provides the POSIX CRYPT option, and the macro
+_XOPEN_CRYPT is also removed from unistd.h to indicate that.
+(_SC_XOPEN_CRYPT is still defined, but sysconf(_SC_XOPEN_CRYPT) will
+return -1 at runtime.) These functions are also unconditionally
+removed from conform/data/{stdlib,unistd}.h-data.
+
+ * posix/unistd.h (_XOPEN_CRYPT, crypt, encrypt): Don't declare.
+ * stdlib/stdlib.h (setkey): Don't declare.
+
+ * configure.ac (--enable-obsolete-crypt): New configure option.
+ * configure: Regenerate.
+ * config.make.in (build-obsolete-crypt): New makefile variable.
+ * crypt/Banner: Delete file.
+ * crypt/Makefile: Don't build anything unless
+ $(build-obsolete-crypt) is 'yes'.
+ * sysdeps/sparc/sparc32/sparcv9
[OE-core] [PATCH 2/2] perl: add patch to solve libcrypt incompatibility
Add Perl's patch submitted to upstream to be compiled along with glibc with
libcrypt split.
Signed-off-by: Charles-Antoine Couret <charles-antoine.cou...@essensium.com>
---
.../perl/perl-native_5.24.1.bb| 1 +
.../perl-5.26.1-guard_old_libcrypt_fix.patch | 27 +++
meta/recipes-devtools/perl/perl_5.24.1.bb | 1 +
3 files changed, 29 insertions(+)
create mode 100644
meta/recipes-devtools/perl/perl/perl-5.26.1-guard_old_libcrypt_fix.patch
diff --git a/meta/recipes-devtools/perl/perl-native_5.24.1.bb
b/meta/recipes-devtools/perl/perl-native_5.24.1.bb
index 48113f05a9..a9ab17d16c 100644
--- a/meta/recipes-devtools/perl/perl-native_5.24.1.bb
+++ b/meta/recipes-devtools/perl/perl-native_5.24.1.bb
@@ -15,6 +15,7 @@ SRC_URI += "\
file://dynaloaderhack.patch \
file://perl-PathTools-don-t-filter-out-blib-from-INC.patch \
file://0001-Configure-Remove-fstack-protector-strong-for-native-.patch \
+ file://perl-5.26.1-guard_old_libcrypt_fix.patch \
"
SRC_URI[md5sum] = "af6a84c7c3e2b8b269c105a5db2f6d53"
diff --git
a/meta/recipes-devtools/perl/perl/perl-5.26.1-guard_old_libcrypt_fix.patch
b/meta/recipes-devtools/perl/perl/perl-5.26.1-guard_old_libcrypt_fix.patch
new file mode 100644
index 00..bdf8e32dfc
--- /dev/null
+++ b/meta/recipes-devtools/perl/perl/perl-5.26.1-guard_old_libcrypt_fix.patch
@@ -0,0 +1,27 @@
+commit 13e70b397dcb0d1bf4a869b670f041c1d7b730d0
+Author: Bjรถrn Esser <besse...@fedoraproject.org>
+Date: Sat Jan 20 20:22:53 2018 +0100
+
+pp: Guard fix for really old bug in glibc libcrypt
+
+Upstream-Status: Pending
+
+diff --git a/pp.c b/pp.c
+index d50ad7ddbf..6510c7b15c 100644
+--- a/pp.c
b/pp.c
+@@ -3650,8 +3650,12 @@ PP(pp_crypt)
+ #if defined(__GLIBC__) || defined(__EMX__)
+ if (PL_reentrant_buffer->_crypt_struct_buffer) {
+ PL_reentrant_buffer->_crypt_struct_buffer->initialized = 0;
+- /* work around glibc-2.2.5 bug */
++#if (defined(__GLIBC__) && __GLIBC__ == 2) && \
++(defined(__GLIBC_MINOR__) && __GLIBC_MINOR__ >= 2 && __GLIBC_MINOR__ < 4)
++ /* work around glibc-2.2.5 bug, has been fixed at some
++ * time in glibc-2.3.X */
+ PL_reentrant_buffer->_crypt_struct_buffer->current_saltbits = 0;
++#endif
+ }
+ #endif
+ }
+
diff --git a/meta/recipes-devtools/perl/perl_5.24.1.bb
b/meta/recipes-devtools/perl/perl_5.24.1.bb
index e5b4886a95..66ccd7f0be 100644
--- a/meta/recipes-devtools/perl/perl_5.24.1.bb
+++ b/meta/recipes-devtools/perl/perl_5.24.1.bb
@@ -63,6 +63,7 @@ SRC_URI += " \
file://perl-errno-generation-gcc5.patch \
file://perl-fix-conflict-between-skip_all-and-END.patch \
file://perl-test-customized.patch \
+file://perl-5.26.1-guard_old_libcrypt_fix.patch \
"
# Fix test case issues
--
2.17.0
--
___
Openembedded-core mailing list
Openembedded-core@lists.openembedded.org
http://lists.openembedded.org/mailman/listinfo/openembedded-core
[OE-core] [PATCH 0/2] Split glibc and libcrypt
Fedora 28 is introducing a breaking change where glibc and libcrypt are split. libcrypt is now provided by libxcrypt as external library. It is backwards compatible, but not the converse. Currently Poky could not be compiled with Fedora 28 and the official yocto-uninative. The purpose of these changes is to split glibc and libcrypt to use libxcrypt only to generate the SDK and the uninative tarball. With that, all distros are able to build Poky with official uninative. The time to glibc to apply this change upstream. Charles-Antoine Couret (2): glibc: split glibc and libcrypt perl: add patch to solve libcrypt incompatibility meta/recipes-core/glibc/glibc-package.inc | 1 + .../0031-nativesdk-deprecate-libcrypt.patch | 418 ++ meta/recipes-core/glibc/glibc_2.27.bb | 1 + .../recipes-core/libxcrypt/libxcrypt_4.0.0.bb | 24 + meta/recipes-core/meta/uninative-tarball.bb | 1 + .../perl/perl-native_5.24.1.bb| 1 + .../perl-5.26.1-guard_old_libcrypt_fix.patch | 27 ++ meta/recipes-devtools/perl/perl_5.24.1.bb | 1 + 8 files changed, 474 insertions(+) create mode 100644 meta/recipes-core/glibc/glibc/0031-nativesdk-deprecate-libcrypt.patch create mode 100644 meta/recipes-core/libxcrypt/libxcrypt_4.0.0.bb create mode 100644 meta/recipes-devtools/perl/perl/perl-5.26.1-guard_old_libcrypt_fix.patch -- 2.17.0 -- ___ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.openembedded.org/mailman/listinfo/openembedded-core
[OE-core] [PATCH] utils: fix gcc 10 version detection
Utils can not detect GCC 10 correctly due to wrong regex.
It generates this error "ERROR: Can't get compiler version from gcc --version
output"
The major version should be 1 or more digits instead of 1 only.
Signed-off-by: Charles-Antoine Couret
---
meta/lib/oe/utils.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta/lib/oe/utils.py b/meta/lib/oe/utils.py
index aee4336482..a4cdce6ff0 100644
--- a/meta/lib/oe/utils.py
+++ b/meta/lib/oe/utils.py
@@ -391,7 +391,7 @@ def host_gcc_version(d, taskcontextonly=False):
except subprocess.CalledProcessError as e:
bb.fatal("Error running %s --version: %s" % (compiler,
e.output.decode("utf-8")))
-match = re.match(r".* (\d\.\d)\.\d.*", output.split('\n')[0])
+match = re.match(r".* (\d+\.\d)\.\d.*", output.split('\n')[0])
if not match:
bb.fatal("Can't get compiler version from %s --version output" %
compiler)
--
2.26.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#136772):
https://lists.openembedded.org/g/openembedded-core/message/136772
Mute This Topic: https://lists.openembedded.org/mt/72570953/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCHv2] utils: fix gcc 10 version detection
Utils can not detect GCC 10 correctly due to wrong regex.
It generates this error "ERROR: Can't get compiler version from gcc --version
output"
Sub-version numbers should be 1 or more digits instead of 1 only.
Signed-off-by: Charles-Antoine Couret
---
meta/lib/oe/utils.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta/lib/oe/utils.py b/meta/lib/oe/utils.py
index aee4336482..9042b370f7 100644
--- a/meta/lib/oe/utils.py
+++ b/meta/lib/oe/utils.py
@@ -391,7 +391,7 @@ def host_gcc_version(d, taskcontextonly=False):
except subprocess.CalledProcessError as e:
bb.fatal("Error running %s --version: %s" % (compiler,
e.output.decode("utf-8")))
-match = re.match(r".* (\d\.\d)\.\d.*", output.split('\n')[0])
+match = re.match(r".* (\d+\.\d+)\.\d+.*", output.split('\n')[0])
if not match:
bb.fatal("Can't get compiler version from %s --version output" %
compiler)
--
2.26.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#136781):
https://lists.openembedded.org/g/openembedded-core/message/136781
Mute This Topic: https://lists.openembedded.org/mt/72573049/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 0/5 v3] image_types: use IMAGE_FILE_MAXSIZE variable to create fixed partition size
Details are described in the first patch. Difference from v2: * Added working and failing tests in meta/lib/oeqa/selftest/cases/imagefeatures.py * Split commit to send documentation to right mailing list Difference from v1: * Added documentation for IMAGE_FILE_MAXSIZE variable * Added Python function to get the value of this variable from shell functions otherwise parsing issue can happen * Added an additional task to check the final result which works for all filesystems and not only those created with dd command. Charles-Antoine Couret (6): image_types: add python function to get the IMAGE_FILE_MAXSIZE:fstype value image_types: use IMAGE_FILE_MAXSIZE variable for ext2/3/4 image types image_types: use IMAGE_FILE_MAXSIZE variable for btrfs image types image_types: use IMAGE_FILE_MAXSIZE variable for f2fs image types image: add check_image_max_size as post function to check file size against IMAGE_FILE_MAXSIZE ref-manual: document IMAGE_FILE_MAXSIZE variable documentation/ref-manual/variables.rst| 14 + meta/classes-recipe/image.bbclass | 30 +++ meta/classes-recipe/image_types.bbclass | 43 --- meta/lib/oeqa/selftest/cases/imagefeatures.py | 52 +++ 4 files changed, 131 insertions(+), 8 deletions(-) -- 2.41.0 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#190048): https://lists.openembedded.org/g/openembedded-core/message/190048 Mute This Topic: https://lists.openembedded.org/mt/102331599/21656 Group Owner: openembedded-core+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 1/5 v3] image_types: add python function to get the IMAGE_FILE_MAXSIZE:fstype value
It returns 0 if the variable is not set for this filesystem.
In case of fixed partitionning where the rootfs partition can't exceed an
amount of bytes, there is currently no automatic and no generic way to have
this requirement met in any case.
Until now, ROOTFS_SIZE value got from directory_size() does not takes into
account
the size of required metadata for the filesystem itself (and does not work well
for other block size than 4k BTW).
Obviously it's a difficult task which depends on rootfs size and filesystem
type.
The workaround was to set IMAGE_OVERHEAD_FACTOR and IMAGE_ROOTFS_EXTRA_SPACE
to add the required extra margins. But when the final rootfs is closed to the
maximum size, it's difficult to adjust them correctly. And if you remove
or add new recipes in your image, you've to recompute these margins to have
enough
space for these metadata when the rootfs is small, and to not have too big final
image when the rootfs is big.
It's cumbersome and error prone to just have a build failure when the final
output
can't be flashed into the partition.
The solution is to follow how it's implemented in buildroot by having a
specific variable, here IMAGE_FILE_MAXSIZE, to create the final sparse file
and trying to fill it with the content of rootfs. If there is enough space,
margins are well compressed and does not consume space in the filesystem.
If there is no enough space, an error is triggered to warm the developer before
trying to use it in the device.
If IMAGE_FILE_MAXSIZE is not set, the idea is to keep the previous behaviour
for compatibility reason and to met other requirements.
Signed-off-by: Charles-Antoine Couret
---
meta/classes-recipe/image_types.bbclass | 7 +++
1 file changed, 7 insertions(+)
diff --git a/meta/classes-recipe/image_types.bbclass
b/meta/classes-recipe/image_types.bbclass
index 4aed64e27f..be8197f1f6 100644
--- a/meta/classes-recipe/image_types.bbclass
+++ b/meta/classes-recipe/image_types.bbclass
@@ -54,6 +54,13 @@ def imagetypes_getdepends(d):
# Sort the set so that ordering is consistant
return " ".join(sorted(deps))
+def get_max_image_size(d, fs):
+max_size = d.getVar("IMAGE_FILE_MAXSIZE:%s" % fs)
+if max_size is not None:
+return max_size
+
+return 0
+
XZ_COMPRESSION_LEVEL ?= "-9"
XZ_INTEGRITY_CHECK ?= "crc32"
--
2.41.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#190049):
https://lists.openembedded.org/g/openembedded-core/message/190049
Mute This Topic: https://lists.openembedded.org/mt/102331600/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 2/5 v3] image_types: use IMAGE_FILE_MAXSIZE variable for ext2/3/4 image types
If defined, this variable value overrides the size of ext* partition file
created by mkfs.
Otherwise previous logic based on ROOTFS_SIZE variable is used.
It should be set when the final file size would not be above a specific value
due to fixed
partitionning for example.
Signed-off-by: Charles-Antoine Couret
---
meta/classes-recipe/image_types.bbclass | 24
1 file changed, 16 insertions(+), 8 deletions(-)
diff --git a/meta/classes-recipe/image_types.bbclass
b/meta/classes-recipe/image_types.bbclass
index be8197f1f6..fb1e33cf3e 100644
--- a/meta/classes-recipe/image_types.bbclass
+++ b/meta/classes-recipe/image_types.bbclass
@@ -79,24 +79,32 @@ IMAGE_CMD:cramfs = "mkfs.cramfs ${IMAGE_ROOTFS}
${IMGDEPLOYDIR}/${IMAGE_NAME}.cr
oe_mkext234fs () {
fstype=$1
+ image_file_maxsize=$2
extra_imagecmd=""
+ rootfs_file_size=$ROOTFS_SIZE
- if [ $# -gt 1 ]; then
- shift
+ if [ $# -gt 2 ]; then
+ shift 2
extra_imagecmd=$@
fi
+
+ if [[ "${image_file_maxsize}" -ne 0 ]]; then
+ rootfs_file_size=${image_file_maxsize}
+ fi
+
# If generating an empty image the size of the sparse block should be
large
# enough to allocate an ext4 filesystem using 4096 bytes per inode,
this is
# about 60K, so dd needs a minimum count of 60, with bs=1024 (bytes per
IO)
eval local COUNT=\"0\"
eval local MIN_COUNT=\"60\"
- if [ $ROOTFS_SIZE -lt $MIN_COUNT ]; then
+ if [ $rootfs_file_size -lt $MIN_COUNT ]; then
eval COUNT=\"$MIN_COUNT\"
fi
+
# Create a sparse image block
- bbdebug 1 Executing "dd if=/dev/zero
of=${IMGDEPLOYDIR}/${IMAGE_NAME}.$fstype seek=$ROOTFS_SIZE count=$COUNT bs=1024"
- dd if=/dev/zero of=${IMGDEPLOYDIR}/${IMAGE_NAME}.$fstype
seek=$ROOTFS_SIZE count=$COUNT bs=1024
+ bbdebug 1 Executing "dd if=/dev/zero
of=${IMGDEPLOYDIR}/${IMAGE_NAME}.$fstype seek=$rootfs_file_size count=$COUNT
bs=1024"
+ dd if=/dev/zero of=${IMGDEPLOYDIR}/${IMAGE_NAME}.$fstype
seek=$rootfs_file_size count=$COUNT bs=1024
bbdebug 1 "Actual Rootfs size: `du -s ${IMAGE_ROOTFS}`"
bbdebug 1 "Actual Partition size: `stat -c '%s'
${IMGDEPLOYDIR}/${IMAGE_NAME}.$fstype`"
bbdebug 1 Executing "mkfs.$fstype -F $extra_imagecmd
${IMGDEPLOYDIR}/${IMAGE_NAME}.$fstype -d ${IMAGE_ROOTFS}"
@@ -105,9 +113,9 @@ oe_mkext234fs () {
fsck.$fstype -pvfD ${IMGDEPLOYDIR}/${IMAGE_NAME}.$fstype || [ $? -le 3 ]
}
-IMAGE_CMD:ext2 = "oe_mkext234fs ext2 ${EXTRA_IMAGECMD}"
-IMAGE_CMD:ext3 = "oe_mkext234fs ext3 ${EXTRA_IMAGECMD}"
-IMAGE_CMD:ext4 = "oe_mkext234fs ext4 ${EXTRA_IMAGECMD}"
+IMAGE_CMD:ext2 = "oe_mkext234fs ext2 ${@get_max_image_size(d, 'ext2')}
${EXTRA_IMAGECMD}"
+IMAGE_CMD:ext3 = "oe_mkext234fs ext3 ${@get_max_image_size(d, 'ext3')}
${EXTRA_IMAGECMD}"
+IMAGE_CMD:ext4 = "oe_mkext234fs ext4 ${@get_max_image_size(d, 'ext4')}
${EXTRA_IMAGECMD}"
MIN_BTRFS_SIZE ?= "16384"
IMAGE_CMD:btrfs () {
--
2.41.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#190050):
https://lists.openembedded.org/g/openembedded-core/message/190050
Mute This Topic: https://lists.openembedded.org/mt/102331601/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 4/5 v3] image_types: use IMAGE_FILE_MAXSIZE variable for f2fs image types
If defined, this variable value overrides the size of f2fs partition file
created by mkfs.
Otherwise previous logic based on ROOTFS_SIZE variable is used.
It should be set when the final file size would not be above a specific value
due to fixed
partitionning for example.
Signed-off-by: Charles-Antoine Couret
---
meta/classes-recipe/image_types.bbclass | 6 ++
1 file changed, 6 insertions(+)
diff --git a/meta/classes-recipe/image_types.bbclass
b/meta/classes-recipe/image_types.bbclass
index 1356552445..6e3190fa3f 100644
--- a/meta/classes-recipe/image_types.bbclass
+++ b/meta/classes-recipe/image_types.bbclass
@@ -272,6 +272,12 @@ IMAGE_CMD:f2fs () {
# 500M the standard IMAGE_OVERHEAD_FACTOR does not work, so add
additional
# space here when under 500M
size=${ROOTFS_SIZE}
+ image_file_maxsize=${@get_max_image_size(d, "f2fs")}
+
+ if [[ "${image_file_maxsize}" -ne 0 ]]; then
+ size=${image_file_maxsize}
+ fi
+
if [ ${size} -lt ${MIN_F2FS_SIZE} ] ; then
size=${MIN_F2FS_SIZE}
bbwarn "Rootfs size is too small for F2FS. Filesystem will be
extended to ${size}K"
--
2.41.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#190052):
https://lists.openembedded.org/g/openembedded-core/message/190052
Mute This Topic: https://lists.openembedded.org/mt/102331603/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 5/5 v3] image: add check_image_max_size as post function to check file size against IMAGE_FILE_MAXSIZE
Trigger an error if the final size is above IMAGE_FILE_MAXSIZE value. Which is
relevant if the
partition size is fixed and the user wants to be sure that the image can be
entirely installed
into its partition.
If the variable is not set, no error is trigger. It works for all filesystems.
Signed-off-by: Charles-Antoine Couret
---
meta/classes-recipe/image.bbclass | 30 +++
meta/lib/oeqa/selftest/cases/imagefeatures.py | 52 +++
2 files changed, 82 insertions(+)
diff --git a/meta/classes-recipe/image.bbclass
b/meta/classes-recipe/image.bbclass
index 7231fad940..d2fcf078c9 100644
--- a/meta/classes-recipe/image.bbclass
+++ b/meta/classes-recipe/image.bbclass
@@ -509,6 +509,7 @@ python () {
d.appendVarFlag(task, 'prefuncs', ' ' + debug + ' set_image_size')
d.prependVarFlag(task, 'postfuncs', 'create_symlinks ')
+d.prependVarFlag(task, 'postfuncs', 'check_image_max_size ')
d.appendVarFlag(task, 'subimages', ' ' + ' '.join(subimages))
d.appendVarFlag(task, 'vardeps', ' ' + ' '.join(vardeps))
d.appendVarFlag(task, 'vardepsexclude', ' DATETIME DATE ' + '
'.join(vardepsexclude))
@@ -609,6 +610,35 @@ python create_symlinks() {
bb.note("Skipping symlink, source does not exist: %s -> %s" %
(dst, src))
}
+#
+# Check if image size is lighter than maximum size
+#
+python check_image_max_size() {
+def get_max_image_size(d, fs):
+max_size = d.getVar("IMAGE_FILE_MAXSIZE:%s" % fs)
+if max_size is not None:
+return int(max_size)
+
+return None
+
+deploy_dir = d.getVar('IMGDEPLOYDIR')
+img_name = d.getVar('IMAGE_NAME')
+taskname = d.getVar("BB_CURRENTTASK")
+subimages = (d.getVarFlag("do_" + taskname, 'subimages', False) or
"").split()
+imgsuffix = d.getVarFlag("do_" + taskname, 'imgsuffix') or
d.expand("${IMAGE_NAME_SUFFIX}.")
+
+for type in subimages:
+file_name = os.path.join(deploy_dir, img_name + imgsuffix + type)
+
+if os.path.exists(file_name):
+file_size = os.stat(file_name).st_size / 1024
+max_size = get_max_image_size(d, type)
+if max_size is not None:
+if file_size > max_size:
+bb.error("The image %s size %d(K) exceeds
IMAGE_FILE_MAXSIZE: %d(K)" % \
+(file_name, file_size, max_size))
+}
+
MULTILIBRE_ALLOW_REP =.
"${base_bindir}|${base_sbindir}|${bindir}|${sbindir}|${libexecdir}|${sysconfdir}|${nonarch_base_libdir}/udev|/lib/modules/[^/]*/modules.*|"
MULTILIB_CHECK_FILE = "${WORKDIR}/multilib_check.py"
MULTILIB_TEMP_ROOTFS = "${WORKDIR}/multilib"
diff --git a/meta/lib/oeqa/selftest/cases/imagefeatures.py
b/meta/lib/oeqa/selftest/cases/imagefeatures.py
index dc88c222bd..afdc7a72fa 100644
--- a/meta/lib/oeqa/selftest/cases/imagefeatures.py
+++ b/meta/lib/oeqa/selftest/cases/imagefeatures.py
@@ -234,6 +234,58 @@ UBINIZE_ARGS_mtd_4_256 ?= "-m 4096 -p 256KiB"
self.assertTrue(os.path.exists(image_path),
"%s image %s doesn't exist" % (itype,
image_path))
+def test_image_maxsize_success(self):
+"""
+Summary: Check if the image built is below maximum size if set
+ Expected:image is built, size below the limit and no error are
reported
+Product: oe-core
+Author: Charles-Antoine Couret
+"""
+image = 'core-image-minimal'
+
+config = """
+IMAGE_FSTYPES += "ext4"
+IMAGE_FILE_MAXSIZE:ext4 = "30"
+"""
+self.write_config(config)
+
+res = bitbake(image)
+bb_vars = get_bb_vars(['DEPLOY_DIR_IMAGE', 'IMAGE_LINK_NAME'], image)
+
+image_path = os.path.join(bb_vars['DEPLOY_DIR_IMAGE'], "%s.%s" %
(bb_vars['IMAGE_LINK_NAME'], "ext4"))
+# check if result image is in deploy directory
+self.assertTrue(os.path.exists(image_path),
+"%s image %s doesn't exist" % ("ext4", image_path))
+# check if result image size is below than maximum value
+self.assertTrue(os.stat(image_path).st_size / 1024 <= 30)
+# No error during execution
+self.assertEqual(res.status, 0)
+
+def test_image_maxsize_failure(self):
+"""
+Summary: Check if the image built with size above limit is
triggering error
+Expected:the image size is above limit and triggers error
+Product: oe-core
+Author: Charles-Antoine Couret
+"""
+image = 'core-image-minimal'
+
+config = """
+IMAGE_FSTYPES += "ext4"
+IMAGE_FILE_MAXSIZE:ext4 = "1000"
+"&quo
[OE-core] [PATCH 3/5 v3] image_types: use IMAGE_FILE_MAXSIZE variable for btrfs image types
If defined, this variable value overrides the size of btrfs partition file
created by mkfs.
Otherwise previous logic based on ROOTFS_SIZE variable is used.
It should be set when the final file size would not be above a specific value
due to fixed
partitionning for example.
Signed-off-by: Charles-Antoine Couret
---
meta/classes-recipe/image_types.bbclass | 6 ++
1 file changed, 6 insertions(+)
diff --git a/meta/classes-recipe/image_types.bbclass
b/meta/classes-recipe/image_types.bbclass
index fb1e33cf3e..1356552445 100644
--- a/meta/classes-recipe/image_types.bbclass
+++ b/meta/classes-recipe/image_types.bbclass
@@ -120,6 +120,12 @@ IMAGE_CMD:ext4 = "oe_mkext234fs ext4
${@get_max_image_size(d, 'ext4')} ${EXTRA_I
MIN_BTRFS_SIZE ?= "16384"
IMAGE_CMD:btrfs () {
size=${ROOTFS_SIZE}
+ image_file_maxsize=${@get_max_image_size(d, "btrfs")}
+
+ if [[ ${image_file_maxsize} -ne 0 ]]; then
+ size=${image_file_maxsize}
+ fi
+
if [ ${size} -lt ${MIN_BTRFS_SIZE} ] ; then
size=${MIN_BTRFS_SIZE}
bbwarn "Rootfs size is too small for BTRFS. Filesystem will be
extended to ${size}K"
--
2.41.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#190051):
https://lists.openembedded.org/g/openembedded-core/message/190051
Mute This Topic: https://lists.openembedded.org/mt/102331602/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH] systemd-boot-cfg: add .conf suffix to default entry label
Since systemd v245 (commit 6cd12ebcfe459466257ea63022a32515d756e719),
systemd-boot
expects default entry to have the complete filename as value.
LABELS from poky are by default without any suffixes like "boot install", so
default entry
does not have the .conf suffix as well and systemd-boot is not able to use this
information
and it's starting in any case the first entry. To be able to start another
entry by default,
.conf suffix is required.
With this change, LABELS variable can still be used by other bootloaders and
being used as description
field.
Signed-off-by: Charles-Antoine Couret
---
meta/classes-recipe/systemd-boot-cfg.bbclass | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta/classes-recipe/systemd-boot-cfg.bbclass
b/meta/classes-recipe/systemd-boot-cfg.bbclass
index 366dd23738..12da41ebad 100644
--- a/meta/classes-recipe/systemd-boot-cfg.bbclass
+++ b/meta/classes-recipe/systemd-boot-cfg.bbclass
@@ -35,7 +35,7 @@ python build_efi_cfg() {
bb.fatal('Unable to open %s' % cfile)
cfgfile.write('# Automatically created by OE\n')
-cfgfile.write('default %s\n' % (labels.split()[0]))
+cfgfile.write('default %s.conf\n' % (labels.split()[0]))
timeout = d.getVar('SYSTEMD_BOOT_TIMEOUT')
if timeout:
cfgfile.write('timeout %s\n' % timeout)
--
2.41.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#188167):
https://lists.openembedded.org/g/openembedded-core/message/188167
Mute This Topic: https://lists.openembedded.org/mt/101565200/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 1/3] image_types: use IMAGE_FILE_MAXSIZE variable for ext2/3/4 image types
If defined, this variable value overrides the size of ext* partition file
created by mkfs.
Otherwise previous logic based on ROOTFS_SIZE variable is used.
It should be set when the final file size would not be above a specific value
due to fixed
partitionning for example.
Signed-off-by: Charles-Antoine Couret
---
meta/classes-recipe/image_types.bbclass | 25 +
1 file changed, 17 insertions(+), 8 deletions(-)
diff --git a/meta/classes-recipe/image_types.bbclass
b/meta/classes-recipe/image_types.bbclass
index e4939af459..cebbb61545 100644
--- a/meta/classes-recipe/image_types.bbclass
+++ b/meta/classes-recipe/image_types.bbclass
@@ -68,24 +68,33 @@ IMAGE_CMD:cramfs = "mkfs.cramfs ${IMAGE_ROOTFS}
${IMGDEPLOYDIR}/${IMAGE_NAME}${I
oe_mkext234fs () {
fstype=$1
+ image_file_maxsize=$2
extra_imagecmd=""
+ rootfs_file_size=$ROOTFS_SIZE
- if [ $# -gt 1 ]; then
- shift
+ if [ $# -gt 2 ]; then
+ shift 2
extra_imagecmd=$@
fi
+
+ if [[ "${image_file_maxsize}" != "\"\"" ]]; then
+ # Remove quotes to get numbers only
+ rootfs_file_size=$(echo "${image_file_maxsize}" | tr -d '"')
+ fi
+
# If generating an empty image the size of the sparse block should be
large
# enough to allocate an ext4 filesystem using 4096 bytes per inode,
this is
# about 60K, so dd needs a minimum count of 60, with bs=1024 (bytes per
IO)
eval local COUNT=\"0\"
eval local MIN_COUNT=\"60\"
- if [ $ROOTFS_SIZE -lt $MIN_COUNT ]; then
+ if [ $rootfs_file_size -lt $MIN_COUNT ]; then
eval COUNT=\"$MIN_COUNT\"
fi
+
# Create a sparse image block
- bbdebug 1 Executing "dd if=/dev/zero
of=${IMGDEPLOYDIR}/${IMAGE_NAME}${IMAGE_NAME_SUFFIX}.$fstype seek=$ROOTFS_SIZE
count=$COUNT bs=1024"
- dd if=/dev/zero
of=${IMGDEPLOYDIR}/${IMAGE_NAME}${IMAGE_NAME_SUFFIX}.$fstype seek=$ROOTFS_SIZE
count=$COUNT bs=1024
+ bbdebug 1 Executing "dd if=/dev/zero
of=${IMGDEPLOYDIR}/${IMAGE_NAME}${IMAGE_NAME_SUFFIX}.$fstype
seek=$rootfs_file_size count=$COUNT bs=1024"
+ dd if=/dev/zero
of=${IMGDEPLOYDIR}/${IMAGE_NAME}${IMAGE_NAME_SUFFIX}.$fstype
seek=$rootfs_file_size count=$COUNT bs=1024
bbdebug 1 "Actual Rootfs size: `du -s ${IMAGE_ROOTFS}`"
bbdebug 1 "Actual Partition size: `stat -c '%s'
${IMGDEPLOYDIR}/${IMAGE_NAME}${IMAGE_NAME_SUFFIX}.$fstype`"
bbdebug 1 Executing "mkfs.$fstype -F $extra_imagecmd
${IMGDEPLOYDIR}/${IMAGE_NAME}${IMAGE_NAME_SUFFIX}.$fstype -d ${IMAGE_ROOTFS}"
@@ -94,9 +103,9 @@ oe_mkext234fs () {
fsck.$fstype -pvfD
${IMGDEPLOYDIR}/${IMAGE_NAME}${IMAGE_NAME_SUFFIX}.$fstype || [ $? -le 3 ]
}
-IMAGE_CMD:ext2 = "oe_mkext234fs ext2 ${EXTRA_IMAGECMD}"
-IMAGE_CMD:ext3 = "oe_mkext234fs ext3 ${EXTRA_IMAGECMD}"
-IMAGE_CMD:ext4 = "oe_mkext234fs ext4 ${EXTRA_IMAGECMD}"
+IMAGE_CMD:ext2 = "oe_mkext234fs ext2 \"${IMAGE_FILE_MAXSIZE:ext2}\"
${EXTRA_IMAGECMD}"
+IMAGE_CMD:ext3 = "oe_mkext234fs ext3 \"${IMAGE_FILE_MAXSIZE:ext3}\"
${EXTRA_IMAGECMD}"
+IMAGE_CMD:ext4 = "oe_mkext234fs ext4 \"${IMAGE_FILE_MAXSIZE:ext4}\"
${EXTRA_IMAGECMD}"
MIN_BTRFS_SIZE ?= "16384"
IMAGE_CMD:btrfs () {
--
2.40.1
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#182360):
https://lists.openembedded.org/g/openembedded-core/message/182360
Mute This Topic: https://lists.openembedded.org/mt/99320004/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 2/3] image_types: use IMAGE_FILE_MAXSIZE variable for btrfs image types
If defined, this variable value overrides the size of btrfs partition file
created by mkfs.
Otherwise previous logic based on ROOTFS_SIZE variable is used.
It should be set when the final file size would not be above a specific value
due to fixed
partitionning for example.
Signed-off-by: Charles-Antoine Couret
---
meta/classes-recipe/image_types.bbclass | 4
1 file changed, 4 insertions(+)
diff --git a/meta/classes-recipe/image_types.bbclass
b/meta/classes-recipe/image_types.bbclass
index cebbb61545..f157a84b2e 100644
--- a/meta/classes-recipe/image_types.bbclass
+++ b/meta/classes-recipe/image_types.bbclass
@@ -110,6 +110,10 @@ IMAGE_CMD:ext4 = "oe_mkext234fs ext4
\"${IMAGE_FILE_MAXSIZE:ext4}\" ${EXTRA_IMAG
MIN_BTRFS_SIZE ?= "16384"
IMAGE_CMD:btrfs () {
size=${ROOTFS_SIZE}
+ if [ -n "${IMAGE_FILE_MAXSIZE:btrfs}" ]; then
+ size=${IMAGE_FILE_MAXSIZE:btrfs}
+ fi
+
if [ ${size} -lt ${MIN_BTRFS_SIZE} ] ; then
size=${MIN_BTRFS_SIZE}
bbwarn "Rootfs size is too small for BTRFS. Filesystem will be
extended to ${size}K"
--
2.40.1
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#182361):
https://lists.openembedded.org/g/openembedded-core/message/182361
Mute This Topic: https://lists.openembedded.org/mt/99320005/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 3/3] image_types: use IMAGE_FILE_MAXSIZE variable for f2fs image types
If defined, this variable value overrides the size of f2fs partition file
created by mkfs.
Otherwise previous logic based on ROOTFS_SIZE variable is used.
It should be set when the final file size would not be above a specific value
due to fixed
partitionning for example.
Signed-off-by: Charles-Antoine Couret
---
meta/classes-recipe/image_types.bbclass | 5 +
1 file changed, 5 insertions(+)
diff --git a/meta/classes-recipe/image_types.bbclass
b/meta/classes-recipe/image_types.bbclass
index f157a84b2e..b5d32b7622 100644
--- a/meta/classes-recipe/image_types.bbclass
+++ b/meta/classes-recipe/image_types.bbclass
@@ -250,6 +250,11 @@ IMAGE_CMD:f2fs () {
# 500M the standard IMAGE_OVERHEAD_FACTOR does not work, so add
additional
# space here when under 500M
size=${ROOTFS_SIZE}
+
+ if [ -n "${IMAGE_FILE_MAXSIZE:f2fs}" ]; then
+ size=${IMAGE_FILE_MAXSIZE:f2fs}
+ fi
+
if [ ${size} -lt ${MIN_F2FS_SIZE} ] ; then
size=${MIN_F2FS_SIZE}
bbwarn "Rootfs size is too small for F2FS. Filesystem will be
extended to ${size}K"
--
2.40.1
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#182362):
https://lists.openembedded.org/g/openembedded-core/message/182362
Mute This Topic: https://lists.openembedded.org/mt/99320006/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 0/3] image_types: use IMAGE_FILE_MAXSIZE variable to create fixed partition size
In case of fixed partitionning where the rootfs partition can't exceed an amount of bytes, there is currently no automatic and no generic way to have this requirement met in any case. Until now, ROOTFS_SIZE value got from directory_size() does not takes into account the size of required metadata for the filesystem itself (and does not work well for other block size than 4k BTW). Obviously it's a difficult task which depends on rootfs size and filesystem type. The workaround was to set IMAGE_OVERHEAD_FACTOR and IMAGE_ROOTFS_EXTRA_SPACE to add the required extra margins. But when the final rootfs is closed to the maximum size, it's difficult to adjust them correctly And if you remove or add new recipes in your image, you've to recompute these margins to have enough space for these metadata when the rootfs is small, and to not have too big final file when the rootfs is big. It's cumbersome and error prone to just have a build failure when the final output can't be flashed into the partition. The solution is to follow how it's implemented in buildroot by having a specific variable, here IMAGE_FILE_MAXSIZE, to create the final sparse file and trying to fill it with the content of rootfs. If there is enough space, margins are well compressed and does not consume space in the filesystem. If there is no enough space, an error is triggered to warm the developer before trying to use it in the device. If IMAGE_FILE_MAXSIZE is not set, the idea is to keep the previous behaviour for compatibility reason and to met other requirements. Charles-Antoine Couret (3): image_types: use IMAGE_FILE_MAXSIZE variable for ext2/3/4 image types image_types: use IMAGE_FILE_MAXSIZE variable for btrfs image types image_types: use IMAGE_FILE_MAXSIZE variable for f2fs image types meta/classes-recipe/image_types.bbclass | 34 +++-- 1 file changed, 26 insertions(+), 8 deletions(-) -- 2.40.1 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#182359): https://lists.openembedded.org/g/openembedded-core/message/182359 Mute This Topic: https://lists.openembedded.org/mt/99320002/21656 Group Owner: openembedded-core+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 1/5 v2] image_types: add python function to get the IMAGE_FILE_MAXSIZE:fstype value
It returns 0 if the variable is not set for this filesystem.
In case of fixed partitionning where the rootfs partition can't exceed an
amount of bytes, there is currently no automatic and no generic way to have
this requirement met in any case.
Until now, ROOTFS_SIZE value got from directory_size() does not takes into
account
the size of required metadata for the filesystem itself (and does not work well
for other block size than 4k BTW).
Obviously it's a difficult task which depends on rootfs size and filesystem
type.
The workaround was to set IMAGE_OVERHEAD_FACTOR and IMAGE_ROOTFS_EXTRA_SPACE
to add the required extra margins. But when the final rootfs is closed to the
maximum size, it's difficult to adjust them correctly And if you remove
or add new recipes in your image, you've to recompute these margins to have
enough
space for these metadata when the rootfs is small, and to not have too big final
file when the rootfs is big.
It's cumbersome and error prone to just have a build failure when the final
output
can't be flashed into the partition.
The solution is to follow how it's implemented in buildroot by having a
specific variable, here IMAGE_FILE_MAXSIZE, to create the final sparse file
and trying to fill it with the content of rootfs. If there is enough space,
margins are well compressed and does not consume space in the filesystem.
If there is no enough space, an error is triggered to warm the developer before
trying to use it in the device.
If IMAGE_FILE_MAXSIZE is not set, the idea is to keep the previous behaviour
for compatibility reason and to met other requirements.
Signed-off-by: Charles-Antoine Couret
---
documentation/ref-manual/variables.rst | 14 ++
meta/classes-recipe/image_types.bbclass | 7 +++
2 files changed, 21 insertions(+)
diff --git a/documentation/ref-manual/variables.rst
b/documentation/ref-manual/variables.rst
index fc29e476cd..d0e476d2eb 100644
--- a/documentation/ref-manual/variables.rst
+++ b/documentation/ref-manual/variables.rst
@@ -3476,6 +3476,20 @@ system and gives an overview of their function and
contents.
variable, see the ":ref:`dev-manual/customizing-images:customizing
images using custom \`\`image_features\`\` and \`\`extra_image_features\`\``"
section in the Yocto Project Development Tasks Manual.
+ :term:`IMAGE_FILE_MAXSIZE`
+ Specifies the maximum size in kilobytes to create the image file for a
+ specific image type, which corresponds to the value set in
+ :term:`IMAGE_FSTYPES`, (e.g. ``ext3``,
+ ``btrfs``, and so forth). When setting this variable, you should use
+ an override for the associated type. Here is an example::
+
+ IMAGE_FILE_MAXSIZE:ext4 = "8192"
+
+ It overrides the :term:`IMAGE_OVERHEAD_FACTOR` and
+ :term:`IMAGE_ROOTFS_EXTRA_SPACE` mechanism for some filesystems.
+ If the maximum size is below the required size to store the rootfs
content,
+ the operation will fail.
+
:term:`IMAGE_FSTYPES`
Specifies the formats the OpenEmbedded build system uses during the
build when creating the root filesystem. For example, setting
diff --git a/meta/classes-recipe/image_types.bbclass
b/meta/classes-recipe/image_types.bbclass
index 023eb87537..33c65e8282 100644
--- a/meta/classes-recipe/image_types.bbclass
+++ b/meta/classes-recipe/image_types.bbclass
@@ -54,6 +54,13 @@ def imagetypes_getdepends(d):
# Sort the set so that ordering is consistant
return " ".join(sorted(deps))
+def get_max_image_size(d, fs):
+max_size = d.getVar("IMAGE_FILE_MAXSIZE:%s" % fs)
+if max_size is not None:
+return max_size
+
+return 0
+
XZ_COMPRESSION_LEVEL ?= "-9"
XZ_INTEGRITY_CHECK ?= "crc32"
--
2.41.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#185575):
https://lists.openembedded.org/g/openembedded-core/message/185575
Mute This Topic: https://lists.openembedded.org/mt/100588378/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 2/5 v2] image_types: use IMAGE_FILE_MAXSIZE variable for ext2/3/4 image types
If defined, this variable value overrides the size of ext* partition file
created by mkfs.
Otherwise previous logic based on ROOTFS_SIZE variable is used.
It should be set when the final file size would not be above a specific value
due to fixed
partitionning for example.
Signed-off-by: Charles-Antoine Couret
---
meta/classes-recipe/image_types.bbclass | 24
1 file changed, 16 insertions(+), 8 deletions(-)
diff --git a/meta/classes-recipe/image_types.bbclass
b/meta/classes-recipe/image_types.bbclass
index 33c65e8282..2ec41c6e54 100644
--- a/meta/classes-recipe/image_types.bbclass
+++ b/meta/classes-recipe/image_types.bbclass
@@ -79,24 +79,32 @@ IMAGE_CMD:cramfs = "mkfs.cramfs ${IMAGE_ROOTFS}
${IMGDEPLOYDIR}/${IMAGE_NAME}${I
oe_mkext234fs () {
fstype=$1
+ image_file_maxsize=$2
extra_imagecmd=""
+ rootfs_file_size=$ROOTFS_SIZE
- if [ $# -gt 1 ]; then
- shift
+ if [ $# -gt 2 ]; then
+ shift 2
extra_imagecmd=$@
fi
+
+ if [[ "${image_file_maxsize}" -ne 0 ]]; then
+ rootfs_file_size=${image_file_maxsize}
+ fi
+
# If generating an empty image the size of the sparse block should be
large
# enough to allocate an ext4 filesystem using 4096 bytes per inode,
this is
# about 60K, so dd needs a minimum count of 60, with bs=1024 (bytes per
IO)
eval local COUNT=\"0\"
eval local MIN_COUNT=\"60\"
- if [ $ROOTFS_SIZE -lt $MIN_COUNT ]; then
+ if [ $rootfs_file_size -lt $MIN_COUNT ]; then
eval COUNT=\"$MIN_COUNT\"
fi
+
# Create a sparse image block
- bbdebug 1 Executing "dd if=/dev/zero
of=${IMGDEPLOYDIR}/${IMAGE_NAME}${IMAGE_NAME_SUFFIX}.$fstype seek=$ROOTFS_SIZE
count=$COUNT bs=1024"
- dd if=/dev/zero
of=${IMGDEPLOYDIR}/${IMAGE_NAME}${IMAGE_NAME_SUFFIX}.$fstype seek=$ROOTFS_SIZE
count=$COUNT bs=1024
+ bbdebug 1 Executing "dd if=/dev/zero
of=${IMGDEPLOYDIR}/${IMAGE_NAME}${IMAGE_NAME_SUFFIX}.$fstype
seek=$rootfs_file_size count=$COUNT bs=1024"
+ dd if=/dev/zero
of=${IMGDEPLOYDIR}/${IMAGE_NAME}${IMAGE_NAME_SUFFIX}.$fstype
seek=$rootfs_file_size count=$COUNT bs=1024
bbdebug 1 "Actual Rootfs size: `du -s ${IMAGE_ROOTFS}`"
bbdebug 1 "Actual Partition size: `stat -c '%s'
${IMGDEPLOYDIR}/${IMAGE_NAME}${IMAGE_NAME_SUFFIX}.$fstype`"
bbdebug 1 Executing "mkfs.$fstype -F $extra_imagecmd
${IMGDEPLOYDIR}/${IMAGE_NAME}${IMAGE_NAME_SUFFIX}.$fstype -d ${IMAGE_ROOTFS}"
@@ -105,9 +113,9 @@ oe_mkext234fs () {
fsck.$fstype -pvfD
${IMGDEPLOYDIR}/${IMAGE_NAME}${IMAGE_NAME_SUFFIX}.$fstype || [ $? -le 3 ]
}
-IMAGE_CMD:ext2 = "oe_mkext234fs ext2 ${EXTRA_IMAGECMD}"
-IMAGE_CMD:ext3 = "oe_mkext234fs ext3 ${EXTRA_IMAGECMD}"
-IMAGE_CMD:ext4 = "oe_mkext234fs ext4 ${EXTRA_IMAGECMD}"
+IMAGE_CMD:ext2 = "oe_mkext234fs ext2 ${@get_max_image_size(d, 'ext2')}
${EXTRA_IMAGECMD}"
+IMAGE_CMD:ext3 = "oe_mkext234fs ext3 ${@get_max_image_size(d, 'ext3')}
${EXTRA_IMAGECMD}"
+IMAGE_CMD:ext4 = "oe_mkext234fs ext4 ${@get_max_image_size(d, 'ext4')}
${EXTRA_IMAGECMD}"
MIN_BTRFS_SIZE ?= "16384"
IMAGE_CMD:btrfs () {
--
2.41.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#185576):
https://lists.openembedded.org/g/openembedded-core/message/185576
Mute This Topic: https://lists.openembedded.org/mt/100588379/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 3/5 v2] image_types: use IMAGE_FILE_MAXSIZE variable for btrfs image types
If defined, this variable value overrides the size of btrfs partition file
created by mkfs.
Otherwise previous logic based on ROOTFS_SIZE variable is used.
It should be set when the final file size would not be above a specific value
due to fixed
partitionning for example.
Signed-off-by: Charles-Antoine Couret
---
meta/classes-recipe/image_types.bbclass | 6 ++
1 file changed, 6 insertions(+)
diff --git a/meta/classes-recipe/image_types.bbclass
b/meta/classes-recipe/image_types.bbclass
index 2ec41c6e54..6e54f2f13a 100644
--- a/meta/classes-recipe/image_types.bbclass
+++ b/meta/classes-recipe/image_types.bbclass
@@ -120,6 +120,12 @@ IMAGE_CMD:ext4 = "oe_mkext234fs ext4
${@get_max_image_size(d, 'ext4')} ${EXTRA_I
MIN_BTRFS_SIZE ?= "16384"
IMAGE_CMD:btrfs () {
size=${ROOTFS_SIZE}
+ image_file_maxsize=${@get_max_image_size(d, "btrfs")}
+
+ if [[ ${image_file_maxsize} -ne 0 ]]; then
+ size=${image_file_maxsize}
+ fi
+
if [ ${size} -lt ${MIN_BTRFS_SIZE} ] ; then
size=${MIN_BTRFS_SIZE}
bbwarn "Rootfs size is too small for BTRFS. Filesystem will be
extended to ${size}K"
--
2.41.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#185577):
https://lists.openembedded.org/g/openembedded-core/message/185577
Mute This Topic: https://lists.openembedded.org/mt/100588381/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 4/5 v2] image_types: use IMAGE_FILE_MAXSIZE variable for f2fs image types
If defined, this variable value overrides the size of f2fs partition file
created by mkfs.
Otherwise previous logic based on ROOTFS_SIZE variable is used.
It should be set when the final file size would not be above a specific value
due to fixed
partitionning for example.
Signed-off-by: Charles-Antoine Couret
---
meta/classes-recipe/image_types.bbclass | 6 ++
1 file changed, 6 insertions(+)
diff --git a/meta/classes-recipe/image_types.bbclass
b/meta/classes-recipe/image_types.bbclass
index 6e54f2f13a..51b2b6ea1d 100644
--- a/meta/classes-recipe/image_types.bbclass
+++ b/meta/classes-recipe/image_types.bbclass
@@ -263,6 +263,12 @@ IMAGE_CMD:f2fs () {
# 500M the standard IMAGE_OVERHEAD_FACTOR does not work, so add
additional
# space here when under 500M
size=${ROOTFS_SIZE}
+ image_file_maxsize=${@get_max_image_size(d, "f2fs")}
+
+ if [[ "${image_file_maxsize}" -ne 0 ]]; then
+ size=${image_file_maxsize}
+ fi
+
if [ ${size} -lt ${MIN_F2FS_SIZE} ] ; then
size=${MIN_F2FS_SIZE}
bbwarn "Rootfs size is too small for F2FS. Filesystem will be
extended to ${size}K"
--
2.41.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#185578):
https://lists.openembedded.org/g/openembedded-core/message/185578
Mute This Topic: https://lists.openembedded.org/mt/100588383/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 5/5 v2] image: add check_image_max_size as post function to check file size against IMAGE_FILE_MAXSIZE
Trigger an error if the final size is above IMAGE_FILE_MAXSIZE value. Which is
relevant if the
partition size is fixed and the user wants to be sure that the image can be
entirely installed
into its partition.
If the variable is not set, no error is trigger. It works for all filesystems.
Signed-off-by: Charles-Antoine Couret
---
meta/classes-recipe/image.bbclass | 30 ++
1 file changed, 30 insertions(+)
diff --git a/meta/classes-recipe/image.bbclass
b/meta/classes-recipe/image.bbclass
index e0dfba4a42..bf47f3dea3 100644
--- a/meta/classes-recipe/image.bbclass
+++ b/meta/classes-recipe/image.bbclass
@@ -509,6 +509,7 @@ python () {
d.appendVarFlag(task, 'prefuncs', ' ' + debug + ' set_image_size')
d.prependVarFlag(task, 'postfuncs', 'create_symlinks ')
+d.prependVarFlag(task, 'postfuncs', 'check_image_max_size ')
d.appendVarFlag(task, 'subimages', ' ' + ' '.join(subimages))
d.appendVarFlag(task, 'vardeps', ' ' + ' '.join(vardeps))
d.appendVarFlag(task, 'vardepsexclude', ' DATETIME DATE ' + '
'.join(vardepsexclude))
@@ -610,6 +611,35 @@ python create_symlinks() {
bb.note("Skipping symlink, source does not exist: %s -> %s" %
(dst, src))
}
+#
+# Check if image size is lighter than maximum size
+#
+python check_image_max_size() {
+def get_max_image_size(d, fs):
+max_size = d.getVar("IMAGE_FILE_MAXSIZE:%s" % fs)
+if max_size is not None:
+return int(max_size)
+
+return None
+
+deploy_dir = d.getVar('IMGDEPLOYDIR')
+img_name = d.getVar('IMAGE_NAME')
+taskname = d.getVar("BB_CURRENTTASK")
+subimages = (d.getVarFlag("do_" + taskname, 'subimages', False) or
"").split()
+imgsuffix = d.getVarFlag("do_" + taskname, 'imgsuffix') or
d.expand("${IMAGE_NAME_SUFFIX}.")
+
+for type in subimages:
+file_name = os.path.join(deploy_dir, img_name + imgsuffix + type)
+
+if os.path.exists(file_name):
+file_size = os.stat(file_name).st_size / 1024
+max_size = get_max_image_size(d, type)
+if max_size is not None:
+if file_size > max_size:
+bb.error("The image %s size %d(K) exceeds
IMAGE_FILE_MAXSIZE: %d(K)" % \
+(file_name, file_size, max_size))
+}
+
MULTILIBRE_ALLOW_REP =.
"${base_bindir}|${base_sbindir}|${bindir}|${sbindir}|${libexecdir}|${sysconfdir}|${nonarch_base_libdir}/udev|/lib/modules/[^/]*/modules.*|"
MULTILIB_CHECK_FILE = "${WORKDIR}/multilib_check.py"
MULTILIB_TEMP_ROOTFS = "${WORKDIR}/multilib"
--
2.41.0
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#185579):
https://lists.openembedded.org/g/openembedded-core/message/185579
Mute This Topic: https://lists.openembedded.org/mt/100588384/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-
[OE-core] [PATCH 0/5 v2] image_types: use IMAGE_FILE_MAXSIZE variable to create fixed partition size
Details are described in the first patch now. Difference from v1: * Added documentation for IMAGE_FILE_MAXSIZE variable * Added Python function to get the value of this variable from shell functions otherwise parsing issue can happen * Added an additional task to check the final result which works for all filesystems and not only those created with dd command. Charles-Antoine Couret (5): image_types: add python function to get the IMAGE_FILE_MAXSIZE:fstype value image_types: use IMAGE_FILE_MAXSIZE variable for ext2/3/4 image types image_types: use IMAGE_FILE_MAXSIZE variable for btrfs image types image_types: use IMAGE_FILE_MAXSIZE variable for f2fs image types image: add check_image_max_size as post function to check file size against IMAGE_FILE_MAXSIZE documentation/ref-manual/variables.rst | 14 meta/classes-recipe/image.bbclass | 30 + meta/classes-recipe/image_types.bbclass | 43 - 3 files changed, 79 insertions(+), 8 deletions(-) -- 2.41.0 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#185574): https://lists.openembedded.org/g/openembedded-core/message/185574 Mute This Topic: https://lists.openembedded.org/mt/100588377/21656 Group Owner: openembedded-core+ow...@lists.openembedded.org Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
