Skip to contentSkip to conversation

Patch review

PATCH In review Steve Sakoman 4 participants 14 messages Oct 15, 2020 – Oct 22, 2020 1 message not in the archive

lore
Share & tools
lore
#0 lore raw

cover Patch review

Please review this next set of patches for dunfell and have comments back by end of day Monday.

Passed a-full on autobuilder:

https://autobuilder.yoctoproject.org/typhoon/#/builders/83/builds/1478

The following changes since commit 656d2070f8448681cb69a3d43dbae84a681c1a75:

linux-yocto/5.4: update to v5.4.69 (2020-10-07 13:41:09 -1000)

are available in the Git repository at:

git://git.openembedded.org/openembedded-core-contrib stable/dunfell-nut
http://cgit.openembedded.org/openembedded-core-contrib/log/?h=stable/dunfell-nut

Armin Kuster (1):
timezone: update to 2020b

Naoki Hayama (1):
uninative: Fix typo in error message

Richard Purdie (2):
scripts/oe-build-perf-report: Allow operation with no buildstats
oe-build-perf-report: Ensure correct data is shown for multiple branch
options

Steve Sakoman (1):
Revert "package: get_package_mapping: avoid dependency mapping if
renamed package provides original name"

Victor Kamensky (2):
qemu: add 34Kf-64tlb fictitious cpu type
qemumips: use 34Kf-64tlb CPU emulation

Yoann Congal (1):
bitbake-bblayers/create: Make the example recipe print its message

 meta/classes/package.bbclass                  |  16 +--
 meta/classes/uninative.bbclass                |   2 +-
 meta/conf/machine/qemumips.conf               |   2 +-
 meta/lib/bblayers/templates/example.bb        |   4 +-
 meta/recipes-devtools/qemu/qemu.inc           |   1 +
 ...tlb-fictitious-cpu-type-like-34Kf-bu.patch | 118 ++++++++++++++++++
 meta/recipes-extended/timezone/timezone.inc   |   8 +-
 meta/recipes-extended/timezone/tzdata.bb      |   8 +-
 scripts/oe-build-perf-report                  |   6 +-
 9 files changed, 137 insertions(+), 28 deletions(-)
 create mode 100644 meta/recipes-devtools/qemu/qemu/0001-mips-add-34Kf-64tlb-fictitious-cpu-type-like-34Kf-bu.patch
Signature · 1 line
2.17.1
#1 lore raw

PATCH 1/8 Revert "package: get_package_mapping: avoid dependency mapping if renamed package provides original name"

Breaks wireguard

This reverts commit 210ebed1e9c2285d6e457bf03d1f1a1f3ddc7fda.

---
 meta/classes/package.bbclass | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)
meta/classes/package.bbclass +3 -13
index cc64ddffc3..23ff4772fa 100644
@@ -535,7 +535,7 @@ def copydebugsources(debugsrcdir, sources, d):
 # Package data handling routines
 #
 
-def get_package_mapping (pkg, basepkg, d, depversions=None):
+def get_package_mapping (pkg, basepkg, d):
     import oe.packagedata
 
     data = oe.packagedata.read_subpkgdata(pkg, d)
@@ -546,14 +546,6 @@ def get_package_mapping (pkg, basepkg, d, depversions=None):
         if bb.data.inherits_class('allarch', d) and not d.getVar('MULTILIB_VARIANTS') \
             and data[key] == basepkg:
             return pkg
-        if depversions == []:
-            # Avoid returning a mapping if the renamed package rprovides its original name
-            rprovkey = "RPROVIDES_%s" % pkg
-            if rprovkey in data:
-                if pkg in bb.utils.explode_dep_versions2(data[rprovkey]):
-                    bb.note("%s rprovides %s, not replacing the latter" % (data[key], pkg))
-                    return pkg
-        # Do map to rewritten package name
         return data[key]
 
     return pkg
@@ -574,10 +566,8 @@ def runtime_mapping_rename (varname, pkg, d):
 
     new_depends = {}
     deps = bb.utils.explode_dep_versions2(d.getVar(varname) or "")
-    for depend, depversions in deps.items():
-        new_depend = get_package_mapping(depend, pkg, d, depversions)
-        if depend != new_depend:
-            bb.note("package name mapping done: %s -> %s" % (depend, new_depend))
+    for depend in deps:
+        new_depend = get_package_mapping(depend, pkg, d)
         new_depends[new_depend] = deps[depend]
 
     d.setVar(varname, bb.utils.join_deps(new_depends, commasep=False))
Signature · 1 line
2.17.1
#2 lore raw

PATCH 2/8 timezone: update to 2020b

In reply to Steve Sakoman

From: Armin Kuster <[email protected]>

For more info see: https://github.com/eggert/tz/blob/master/NEWS

(cherry picked from commit b19bc7ae71a59673bd725e1de3251667c2026ed5)
---
 meta/recipes-extended/timezone/timezone.inc | 8 +++-----
 meta/recipes-extended/timezone/tzdata.bb    | 8 ++++----
 2 files changed, 7 insertions(+), 9 deletions(-)
meta/recipes-extended/timezone/timezone.inc +3 -5
index 32723f3dec..8eb17c5eaf 100644
@@ -6,7 +6,7 @@ SECTION = "base"
 LICENSE = "PD & BSD & BSD-3-Clause"
 LIC_FILES_CHKSUM = "file://LICENSE;md5=c679c9d6b02bc2757b3eaf8f53c43fba"
 
-PV = "2020a"
+PV = "2020b"
 
 SRC_URI =" http://www.iana.org/time-zones/repository/releases/tzcode${PV}.tar.gz;name=tzcode \
            http://www.iana.org/time-zones/repository/releases/tzdata${PV}.tar.gz;name=tzdata \
@@ -14,7 +14,5 @@ SRC_URI =" http://www.iana.org/time-zones/repository/releases/tzcode${PV}.tar.gz
 
 UPSTREAM_CHECK_URI = "http://www.iana.org/time-zones"
 
-SRC_URI[tzcode.md5sum] = "f87c3477e85a5c4b00df0def6c6a0055"
-SRC_URI[tzcode.sha256sum] = "7d2af7120ee03df71fbca24031ccaf42404752e639196fe93c79a41b38a6d669"
-SRC_URI[tzdata.md5sum] = "96a985bb8eeab535fb8aa2132296763a"
-SRC_URI[tzdata.sha256sum] = "547161eca24d344e0b5f96aff6a76b454da295dc14ed4ca50c2355043fb899a2"
+SRC_URI[tzcode.sha256sum] = "47eff8944de4a64f7629b851e4a32338ab12c9b73edd62063795167ff1fe43da"
+SRC_URI[tzdata.sha256sum] = "9b053f951d245ce89d850b96ee4711d82d833559b1fc96ba19f90bc4d745e809"
meta/recipes-extended/timezone/tzdata.bb +4 -4
index 1e2b440fb0..e6a0655afe 100644
@@ -15,18 +15,18 @@ DEFAULT_TIMEZONE ?= "Universal"
 INSTALL_TIMEZONE_FILE ?= "1"
 
 TZONES= "africa antarctica asia australasia europe northamerica southamerica  \
-         factory etcetera backward systemv \
+         factory etcetera backward \
         "
 # pacificnew 
 
 do_compile () {
         for zone in ${TZONES}; do \
             ${STAGING_BINDIR_NATIVE}/zic -d ${WORKDIR}${datadir}/zoneinfo -L /dev/null \
-                -y ${S}/yearistype.sh ${S}/${zone} ; \
+                ${S}/${zone} ; \
             ${STAGING_BINDIR_NATIVE}/zic -d ${WORKDIR}${datadir}/zoneinfo/posix -L /dev/null \
-                -y ${S}/yearistype.sh ${S}/${zone} ; \
+                ${S}/${zone} ; \
             ${STAGING_BINDIR_NATIVE}/zic -d ${WORKDIR}${datadir}/zoneinfo/right -L ${S}/leapseconds \
-                -y ${S}/yearistype.sh ${S}/${zone} ; \
+                ${S}/${zone} ; \
         done
 }
 
Signature · 1 line
2.17.1
#3 lore raw

PATCH 3/8 scripts/oe-build-perf-report: Allow operation with no buildstats

In reply to Steve Sakoman

From: Richard Purdie <[email protected]>

If buildstats is missing, avoid a backtrace.

(cherry picked from commit 23c4b21f947c7aae1303ca6526fc5aaaa9fc7bb8)
---
 scripts/oe-build-perf-report | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
scripts/oe-build-perf-report +1 -1
index 7ed86a72f6..b46d61869c 100755
@@ -353,7 +353,7 @@ def print_html_report(data, id_comp, buildstats):
             bs_key = test + '.' + meas
             rev = str(metadata['commit_num']['value'])
             comp_rev = str(metadata['commit_num']['value_old'])
-            if (rev in buildstats and bs_key in buildstats[rev] and
+            if (buildstats and rev in buildstats and bs_key in buildstats[rev] and
                     comp_rev in buildstats and bs_key in buildstats[comp_rev]):
                 new_meas['buildstats'] = BSSummary(buildstats[comp_rev][bs_key],
                                                    buildstats[rev][bs_key])
Signature · 1 line
2.17.1
#4 lore raw

PATCH 4/8 oe-build-perf-report: Ensure correct data is shown for multiple branch options

In reply to Steve Sakoman

From: Richard Purdie <[email protected]>

If --branch and --branch2 are the same, there is extra confusing data plotted onto the graph. Only do that if the branches are actually different.

(cherry picked from commit c263d810b7eb47ee90f2adeb5ab6decf7332bec2)
---
 scripts/oe-build-perf-report | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
scripts/oe-build-perf-report +2 -2
index b46d61869c..7812ea4540 100755
@@ -509,7 +509,7 @@ def main(argv=None):
 
     revs = gitarchive.get_test_revs(log, repo, args.tag_name, hostname=args.hostname,
                                     branch=args.branch, machine=args.machine)
-    if args.branch2:
+    if args.branch2 and args.branch2 != args.branch:
         revs2 = gitarchive.get_test_revs(log, repo, args.tag_name, hostname=args.hostname,
                                          branch=args.branch2, machine=args.machine)
         if not len(revs2):
@@ -533,7 +533,7 @@ def main(argv=None):
     else:
         index1 = len(revs) - 1
 
-    if args.branch2:
+    if args.branch2 and args.branch2 != args.branch:
         revs2.append(revs[index1])
         index1 = len(revs2) - 1
         revs = revs2
Signature · 1 line
2.17.1
#5 lore raw

PATCH 5/8 bitbake-bblayers/create: Make the example recipe print its message

In reply to Steve Sakoman

From: Yoann Congal <[email protected]>

The example recipe is setup to print a message using bb.plain() in the "do_build" task but this task is "noexec" so the message never prints. This might be confusing.

This moves the message printing into another "do_display_banner" task and add it to the do_build "before" list.

(cherry picked from commit 252385bef9b226f32691b8513869ea3e41813b40)
---
 meta/lib/bblayers/templates/example.bb | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
meta/lib/bblayers/templates/example.bb +3 -1
index c4b873d593..facaae35d2 100644
@@ -2,10 +2,12 @@ SUMMARY = "bitbake-layers recipe"
 DESCRIPTION = "Recipe created by bitbake-layers"
 LICENSE = "MIT"
 
-python do_build() {
+python do_display_banner() {
     bb.plain("***********************************************");
     bb.plain("*                                             *");
     bb.plain("*  Example recipe created by bitbake-layers   *");
     bb.plain("*                                             *");
     bb.plain("***********************************************");
 }
+
+addtask display_banner before do_build
Signature · 1 line
2.17.1
#6 lore raw

PATCH 6/8 uninative: Fix typo in error message

In reply to Steve Sakoman

From: Naoki Hayama <[email protected]>

Fix typo in an error message.
s/verson/version/

(cherry picked from commit bc96db2e0b5b8a9cc2c909ea70df290e03a50b94)
---
 meta/classes/uninative.bbclass | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
meta/classes/uninative.bbclass +1 -1
index 316c0f0616..1e19917a97 100644
@@ -89,7 +89,7 @@ python uninative_event_fetchloader() {
         # ldd output is "ldd (Ubuntu GLIBC 2.23-0ubuntu10) 2.23", extract last option from first line
         glibcver = subprocess.check_output(["ldd", "--version"]).decode('utf-8').split('\n')[0].split()[-1]
         if bb.utils.vercmp_string(d.getVar("UNINATIVE_MAXGLIBCVERSION"), glibcver) < 0:
-            raise RuntimeError("Your host glibc verson (%s) is newer than that in uninative (%s). Disabling uninative so that sstate is not corrupted." % (glibcver, d.getVar("UNINATIVE_MAXGLIBCVERSION")))
+            raise RuntimeError("Your host glibc version (%s) is newer than that in uninative (%s). Disabling uninative so that sstate is not corrupted." % (glibcver, d.getVar("UNINATIVE_MAXGLIBCVERSION")))
 
         cmd = d.expand("\
 mkdir -p ${UNINATIVE_STAGING_DIR}-uninative; \
Signature · 1 line
2.17.1
#7 lore raw

PATCH 7/8 qemu: add 34Kf-64tlb fictitious cpu type

In reply to Steve Sakoman

From: Victor Kamensky <[email protected]>

In Yocto Project PR 13992 it was reported that qemumips in autobuilder runs almost twice slower then qemumips64 and some times hit time out.

Upon investigations of qemu-system with perf, gdb, and SystemTap and comparing qemumips and qemumips64 machines behavior it was noticed that qemu soft mmu code behaves quite different and in case if qemumips tlbwr instruction called 16 times more oftern. It happens that in qemumips64 case qemu runs with cpu type that contains 64 TLB, but in case of qemumips qemu runs with cpu type that contains only 16 TLBs.

The idea of proposed qemu patch is to introduce fictitious 34Kf-64tlb cpu type that defined exactly as 34Kf but has 64 TLBs, instead of original 16 TLBs.

Testing of core-image-full-cmdline:do_testimage with 34Kf-64tlb shows 40% or so test execution real time improvement.

Note for future porters of the patch: easiest way to update the patch and be in sync with 34Kf definition is to copy 34Kf machine definition and apply the following changes to it (just change 15 to 63 of CP0C1_MMU bits value)

[kamensky@coreos-lnx2 qemu]$ diff ~/34Kf.c ~/34Kf-64tlb.c 2c2
< .name = "34Kf",

Quoted 1 line
        .name = "34Kf-64tlb",

6c6

<         .CP0_Config1 = MIPS_CONFIG1 | (1 << CP0C1_FP) | (15 << CP0C1_MMU) |
>         .CP0_Config1 = MIPS_CONFIG1 | (1 << CP0C1_FP) | (63 << CP0C1_MMU) |

Fixes https://bugzilla.yoctoproject.org/show_bug.cgi?id=13992

Upstream Status: Inappropriate

(cherry picked from commit 4470a04943352224955f17e004962f0f9e1c9b0c)
---
 meta/recipes-devtools/qemu/qemu.inc           |   1 +
 ...tlb-fictitious-cpu-type-like-34Kf-bu.patch | 118 ++++++++++++++++++
 2 files changed, 119 insertions(+)
 create mode 100644 meta/recipes-devtools/qemu/qemu/0001-mips-add-34Kf-64tlb-fictitious-cpu-type-like-34Kf-bu.patch
meta/recipes-devtools/qemu/qemu.inc +1 -0
index 7ce89c0023..7c21b66a0c 100644
@@ -48,6 +48,7 @@ SRC_URI = "https://download.qemu.org/${BPN}-${PV}.tar.xz \
 	   file://CVE-2020-14364.patch \
 	   file://CVE-2020-14415.patch \
 	   file://CVE-2020-16092.patch \
+	   file://0001-mips-add-34Kf-64tlb-fictitious-cpu-type-like-34Kf-bu.patch \
 	   "
 UPSTREAM_CHECK_REGEX = "qemu-(?P<pver>\d+(\.\d+)+)\.tar"
 
meta/recipes-devtools/qemu/qemu/0001-mips-add-34Kf-64tlb-fictitious-cpu-type-like-34Kf-bu.patch new file +118 -0

118 lines not on this page · raw message

Signature · 1 line
2.17.1
#8 lore raw

Re: [OE-core][dunfell 7/8] qemu: add 34Kf-64tlb fictitious cpu type

Quoted 19 lines · Steve Sakoman

From: Victor Kamensky <[email protected]>

In Yocto Project PR 13992 it was reported that qemumips in autobuilder runs almost twice slower then qemumips64 and some times hit time out.

Upon investigations of qemu-system with perf, gdb, and SystemTap and comparing qemumips and qemumips64 machines behavior it was noticed that qemu soft mmu code behaves quite different and in case if qemumips tlbwr instruction called 16 times more oftern. It happens that in qemumips64 case qemu runs with cpu type that contains 64 TLB, but in case of qemumips qemu runs with cpu type that contains only 16 TLBs.

The idea of proposed qemu patch is to introduce fictitious 34Kf-64tlb cpu type that defined exactly as 34Kf but has 64 TLBs, instead of original 16 TLBs.

Quoted 8 lines

Testing of core-image-full-cmdline:do_testimage with 34Kf-64tlb shows 40% or so test execution real time improvement.

Note for future porters of the patch: easiest way to update the patch and be in sync with 34Kf definition is to copy 34Kf machine definition and apply the following changes to it (just change 15 to 63 of CP0C1_MMU bits value)

There is discussion upstream about just changing the tlb to 64 for this machine. I'd suggest we hold off this for dunfell until its resolved there.

Cheers,

Richard

#9 lore raw

Re: [OE-core][dunfell 7/8] qemu: add 34Kf-64tlb fictitious cpu type

Quoted 25 lines · Richard Purdie
Quoted 23 lines · Steve Sakoman

From: Victor Kamensky <[email protected]>

In Yocto Project PR 13992 it was reported that qemumips in autobuilder runs almost twice slower then qemumips64 and some times hit time out.

Upon investigations of qemu-system with perf, gdb, and SystemTap and comparing qemumips and qemumips64 machines behavior it was noticed that qemu soft mmu code behaves quite different and in case if qemumips tlbwr instruction called 16 times more oftern. It happens that in qemumips64 case qemu runs with cpu type that contains 64 TLB, but in case of qemumips qemu runs with cpu type that contains only 16 TLBs.

The idea of proposed qemu patch is to introduce fictitious 34Kf-64tlb cpu type that defined exactly as 34Kf but has 64 TLBs, instead of original 16 TLBs.

Testing of core-image-full-cmdline:do_testimage with 34Kf-64tlb shows 40% or so test execution real time improvement.

Quoted 8 lines
Quoted 4 lines

Note for future porters of the patch: easiest way to update the patch and be in sync with 34Kf definition is to copy 34Kf machine definition and apply the following changes to it (just change 15 to 63 of CP0C1_MMU bits value)

There is discussion upstream about just changing the tlb to 64 for this machine. I'd suggest we hold off this for dunfell until its resolved there.

OK, I'll remove these two patches from the pull request.

Steve

#10 lore raw

PATCH 8/8 qemumips: use 34Kf-64tlb CPU emulation

In reply to Steve Sakoman

From: Victor Kamensky <[email protected]>

In order to improve performance of qemumips autobuilder test runs, let's use 34Kf-64tlb cpu type that was introduced in OE version of qemu. 34Kf-64tlb cpu type is identical to 34Kf but it has 64 TLBs configure vs original 16 TLBs.

The change in number of TLBs in emulated CPU reduces TLB trash and reduces number of times TLB refill kernel code runs, and therefore siginificantly improves tests execution time.

Note 34Kf-64tlb qemu cpu type does not exist in upstream, so far it is added as patch in OE. See qemu 34Kf-64tlb cpu addition commit for more details.

Fixes https://bugzilla.yoctoproject.org/show_bug.cgi?id=13992

(cherry picked from commit 6a9568401cbf8bed89d1f63dae61a07a83a20b50)
---
 meta/conf/machine/qemumips.conf | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
meta/conf/machine/qemumips.conf +1 -1
index 1373e4cba0..b8c80f02ef 100644
@@ -15,4 +15,4 @@ SERIAL_CONSOLES ?= "115200;ttyS0 115200;ttyS1"
 
 QB_SYSTEM_NAME = "qemu-system-mips"
 
-QB_CPU = "-cpu 34Kf"
+QB_CPU = "-cpu 34Kf-64tlb"
Signature · 1 line
2.17.1

A message that is not in the archive (sent off-list or not yet received): [email protected]

#12 lore raw

Re: [OE-core][dunfell 1/8] Revert "package: get_package_mapping: avoid dependency mapping if renamed package provides original name"

I've included this revert patch for comment because I have gotten several requests to revert and I'd like wider discussion of the issue.

At the moment wireguard is the only known recipe that is failing and there are no proposed fixes.

If you are aware of recipes in other layers that are affected, please let me know.

And if someone would like to tackle fixing wireguard in meta-openembedded that would be much preferred to a revert.

Thanks!

Steve

Quoted 57 lines · Steve Sakoman via lists.openembedded.org

Breaks wireguard

This reverts commit 210ebed1e9c2285d6e457bf03d1f1a1f3ddc7fda.

---
 meta/classes/package.bbclass | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)
meta/classes/package.bbclass +3 -14
index cc64ddffc3..23ff4772fa 100644
@@ -535,7 +535,7 @@ def copydebugsources(debugsrcdir, sources, d):
 # Package data handling routines
 #
 
-def get_package_mapping (pkg, basepkg, d, depversions=None):
+def get_package_mapping (pkg, basepkg, d):
     import oe.packagedata
 
     data = oe.packagedata.read_subpkgdata(pkg, d)
@@ -546,14 +546,6 @@ def get_package_mapping (pkg, basepkg, d, depversions=None):
         if bb.data.inherits_class('allarch', d) and not d.getVar('MULTILIB_VARIANTS') \
             and data[key] == basepkg:
             return pkg
-        if depversions == []:
-            # Avoid returning a mapping if the renamed package rprovides its original name
-            rprovkey = "RPROVIDES_%s" % pkg
-            if rprovkey in data:
-                if pkg in bb.utils.explode_dep_versions2(data[rprovkey]):
-                    bb.note("%s rprovides %s, not replacing the latter" % (data[key], pkg))
-                    return pkg
-        # Do map to rewritten package name
         return data[key]
 
     return pkg
@@ -574,10 +566,8 @@ def runtime_mapping_rename (varname, pkg, d):
 
     new_depends = {}
     deps = bb.utils.explode_dep_versions2(d.getVar(varname) or "")
-    for depend, depversions in deps.items():
-        new_depend = get_package_mapping(depend, pkg, d, depversions)
-        if depend != new_depend:
-            bb.note("package name mapping done: %s -> %s" % (depend, new_depend))
+    for depend in deps:
+        new_depend = get_package_mapping(depend, pkg, d)
         new_depends[new_depend] = deps[depend]
 
     d.setVar(varname, bb.utils.join_deps(new_depends, commasep=False))
--

2.17.1

#13 lore raw

Re: [OE-core][dunfell 1/8] Revert "package: get_package_mapping: avoid dependency mapping if renamed package provides original name"

Quoted 4 lines · Steve Sakoman

I've included this revert patch for comment because I have gotten several requests to revert and I'd like wider discussion of the issue.

Quoted 8 lines

At the moment wireguard is the only known recipe that is failing and there are no proposed fixes.

If you are aware of recipes in other layers that are affected, please let me know.

And if someone would like to tackle fixing wireguard in meta-openembedded that would be much preferred to a revert.

It would also be good to know if the issue exists on master or not. I'm worried we may have the issue there as well which is a reason I'd favour looking into and understanding this rather than blindly reverting. I've not seen anyone have a full understanding of what the issue is as yet.

Cheers,

Richard

#14 lore raw

Re: [OE-core][dunfell 1/8] Revert "package: get_package_mapping: avoid dependency mapping if renamed package provides original name"

Looks like it's caused by wireguard setting PKG variable as:
PKG_${PN} = "kernel-module-${MODULE_NAME}-${KERNEL_VERSION}"

as discussed in:
https://github.com/openembedded/meta-openembedded/issues/285#issuecomment-714401682

it was added in
https://git.openembedded.org/meta-openembedded/commit/?id=ad0db129c7c8484520cd20cdfbb4670489125280
but the explanation there doesn't make much sense to me.

Quoted 30 lines · Richard Purdie
Quoted 4 lines · Steve Sakoman

I've included this revert patch for comment because I have gotten several requests to revert and I'd like wider discussion of the issue.

Quoted 8 lines

At the moment wireguard is the only known recipe that is failing and there are no proposed fixes.

If you are aware of recipes in other layers that are affected, please let me know.

And if someone would like to tackle fixing wireguard in meta-openembedded that would be much preferred to a revert.

It would also be good to know if the issue exists on master or not. I'm worried we may have the issue there as well which is a reason I'd favour looking into and understanding this rather than blindly reverting. I've not seen anyone have a full understanding of what the issue is as yet.

Cheers,

Richard