review request
PATCH 3/15 oeqa/qemurunner: Fix binary vs str issue
In reply to Anuj MittalFrom: Richard Purdie <[email protected]>
The recent logging changes for qemurunner showed up as errors on the autobuilder where decode couldn't be called on the returned string. Since the code returns binary data, return b'' instead of '' to match to avoid tracebacks.
One of these cases was newly added, copied from the other which has been there for a long time, always broken.
--- meta/lib/oeqa/utils/qemurunner.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
meta/lib/oeqa/utils/qemurunner.py +2 -2
@@ -650,7 +650,7 @@ class LoggingThread(threading.Thread): data = self.readsock.recv(count) except socket.error as e: if e.errno == errno.EAGAIN or e.errno == errno.EWOULDBLOCK: - return '' + return b'' else: raise
@@ -663,7 +663,7 @@ class LoggingThread(threading.Thread): # until qemu exits. if not self.canexit: raise Exception("Console connection closed unexpectedly") - return '' + return b'' return data
Signature · 1 line
2.31.1
cover review request
Please review these changes for hardknott. No problems while testing on autobuilder.
Thanks,
Anuj
The following changes since commit d5d88c2293e8ebc958d1bce9af8f796024443be9:
libevent: Increase ptest timing tolerance 50 ms -> 100 ms (2021-05-07 08:51:54 +0800)
are available in the Git repository at:
git://push.openembedded.org/openembedded-core-contrib anujm/hardknott
Alexander Kanavin (1):
linux-firmware: upgrade 20210208 -> 20210315
Anuj Mittal (1):
lsb-release: fix reproducibility failure
Bruce Ashfield (1):
linux-yocto/5.4: qemuppc32: reduce serial shutdown issues
Chen Qi (1):
db: update CVE_PRODUCT
Kai Kang (1):
grub2.inc: remove '-O2' from CFLAGS
Michael Opdenacker (1):
sanity.bbclass: mention CONNECTIVITY_CHECK_URIS in network failure
message
Peter Kjellerstedt (1):
libcap: Configure Make variables correctly without a horrible hack
Richard Purdie (4):
oeqa/qemurunner: Improve logging thread exit handling for qemu
shutdown test
oeqa/qemurunner: Fix binary vs str issue
oeqa/qemurunner: Improve handling of run_serial for shutdown commands
puzzles: Upstream changed to main branch for development
Romain Naour (1):
dejagnu: needs expect at runtime
Vinícius Ossanes Aquino (1):
lttng-modules: backport patches to fix build against 5.12+ kernel
Yann Dirson (1):
linux-firmware: include all relevant files in -bcm4356
zhengruoqin (1):
ruby: upgrade 3.0.0 -> 3.0.1
meta/classes/sanity.bbclass | 9 +-
meta/lib/oeqa/selftest/cases/runqemu.py | 9 +-
meta/lib/oeqa/utils/qemurunner.py | 21 +-
meta/recipes-bsp/grub/grub2.inc | 2 +
.../recipes-devtools/dejagnu/dejagnu_1.6.2.bb | 1 +
.../ruby/{ruby_3.0.0.bb => ruby_3.0.1.bb} | 2 +-
.../help2man-reproducibility.patch | 27 ++
meta/recipes-extended/lsb/lsb-release_1.4.bb | 1 +
...20210208.bb => linux-firmware_20210315.bb} | 8 +-
.../linux/linux-yocto-rt_5.4.bb | 2 +-
.../linux/linux-yocto-tiny_5.4.bb | 2 +-
meta/recipes-kernel/linux/linux-yocto_5.4.bb | 2 +-
...block-add-a-disk_uevent-helper-v5.12.patch | 305 ++++++++++++++++++
...block-add-a-disk_uevent-helper-v5.12.patch | 48 +++
...free-event-name-mismatching-with-pro.patch | 71 ++++
.../lttng/lttng-modules_2.12.5.bb | 3 +
meta/recipes-sato/puzzles/puzzles_git.bb | 2 +-
meta/recipes-support/db/db_5.3.28.bb | 2 +-
meta/recipes-support/libcap/libcap_2.48.bb | 21 +-
19 files changed, 504 insertions(+), 34 deletions(-)
rename meta/recipes-devtools/ruby/{ruby_3.0.0.bb => ruby_3.0.1.bb} (97%)
create mode 100644 meta/recipes-extended/lsb/lsb-release/help2man-reproducibility.patch
rename meta/recipes-kernel/linux-firmware/{linux-firmware_20210208.bb => linux-firmware_20210315.bb} (99%)
create mode 100644 meta/recipes-kernel/lttng/lttng-modules/0005-fix-block-add-a-disk_uevent-helper-v5.12.patch
create mode 100644 meta/recipes-kernel/lttng/lttng-modules/0006-fix-backport-block-add-a-disk_uevent-helper-v5.12.patch
create mode 100644 meta/recipes-kernel/lttng/lttng-modules/0007-fix-mm-tracing-kfree-event-name-mismatching-with-pro.patchSignature · 1 line
2.31.1
PATCH 1/15 sanity.bbclass: mention CONNECTIVITY_CHECK_URIS in network failure message
From: Michael Opdenacker <[email protected]>
This expands the error message when a network failure is detected. It happens that some ISPs or networks block the default example.com domain. Therefore, instead of disabling network access, it lets the user know how to modify the test URL.
--- meta/classes/sanity.bbclass | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)
meta/classes/sanity.bbclass +6 -3
@@ -392,9 +392,12 @@ def check_connectivity(d): msg = data.getVar('CONNECTIVITY_CHECK_MSG') or "" if len(msg) == 0: msg = "%s.\n" % err - msg += " Please ensure your host's network is configured correctly,\n" - msg += " or set BB_NO_NETWORK = \"1\" to disable network access if\n" - msg += " all required sources are on local disk.\n" + msg += " Please ensure your host's network is configured correctly.\n" + msg += " If your ISP or network is blocking the above URL,\n" + msg += " try with another domain name, for example by setting:\n" + msg += " CONNECTIVITY_CHECK_URIS = \"https://www.yoctoproject.org/\"" + msg += " You could also set BB_NO_NETWORK = \"1\" to disable network\n" + msg += " access if all required sources are on local disk.\n" retval = msg return retval
Signature · 1 line
2.31.1
PATCH 2/15 oeqa/qemurunner: Improve logging thread exit handling for qemu shutdown test
In reply to Anuj MittalFrom: Richard Purdie <[email protected]>
Rather than totally disabling the logging, inform it we're about to exit so we can log messages over the exit cleanly too. This aids debugging. It also avoids a race where the logging handler could still error whilst shutting down.
Also remove a race window by notificing the handler of the shutdown first, before triggering it. This removes a race window I watched in local testing.
--- meta/lib/oeqa/selftest/cases/runqemu.py | 9 ++++----- meta/lib/oeqa/utils/qemurunner.py | 12 +++++++++++- 2 files changed, 15 insertions(+), 6 deletions(-)
meta/lib/oeqa/selftest/cases/runqemu.py +4 -5
@@ -163,12 +163,11 @@ class QemuTest(OESelftestTestCase): bitbake(cls.recipe) def _start_qemu_shutdown_check_if_shutdown_succeeded(self, qemu, timeout): + # Allow the runner's LoggingThread instance to exit without errors + # (such as the exception "Console connection closed unexpectedly") + # as qemu will disappear when we shut it down + qemu.runner.allowexit() qemu.run_serial("shutdown -h now") - # Stop thread will stop the LoggingThread instance used for logging - # qemu through serial console, stop thread will prevent this code - # from facing exception (Console connection closed unexpectedly) - # when qemu was shutdown by the above shutdown command - qemu.runner.stop_thread() time_track = 0 try: while True:
meta/lib/oeqa/utils/qemurunner.py +11 -1
@@ -471,6 +471,10 @@ class QemuRunner: self.thread.stop() self.thread.join() + def allowexit(self): + if self.thread: + self.thread.allowexit() + def restart(self, qemuparams = None): self.logger.warning("Restarting qemu process") if self.runqemu.poll() is None:
@@ -564,6 +568,7 @@ class LoggingThread(threading.Thread): self.logger = logger self.readsock = None self.running = False + self.canexit = False self.errorevents = select.POLLERR | select.POLLHUP | select.POLLNVAL self.readevents = select.POLLIN | select.POLLPRI
@@ -597,6 +602,9 @@ class LoggingThread(threading.Thread): self.close_ignore_error(self.writepipe) self.running = False + def allowexit(self): + self.canexit = True + def eventloop(self): poll = select.poll() event_read_mask = self.errorevents | self.readevents
@@ -653,7 +661,9 @@ class LoggingThread(threading.Thread): # happened. But for this code it counts as an # error since the connection shouldn't go away # until qemu exits. - raise Exception("Console connection closed unexpectedly") + if not self.canexit: + raise Exception("Console connection closed unexpectedly") + return '' return data
Signature · 1 line
2.31.1
PATCH 4/15 oeqa/qemurunner: Improve handling of run_serial for shutdown commands
In reply to Anuj MittalFrom: Richard Purdie <[email protected]>
When running a shutdown command, the serial port can close without the command returning. This is seen as the socket being readable but having no data. Change the way this case is handled in the code to avoid tracebacks.
--- meta/lib/oeqa/utils/qemurunner.py | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
meta/lib/oeqa/utils/qemurunner.py +6 -1
@@ -71,6 +71,8 @@ class QemuRunner: self.monitorpipe = None self.logger = logger + # Whether we're expecting an exit and should show related errors + self.canexit = False # Enable testing other OS's # Set commands for target communication, and default to Linux ALWAYS
@@ -472,6 +474,7 @@ class QemuRunner: self.thread.join() def allowexit(self): + self.canexit = True if self.thread: self.thread.allowexit()
@@ -530,7 +533,9 @@ class QemuRunner: if re.search(self.boot_patterns['search_cmd_finished'], data): break else: - raise Exception("No data on serial console socket") + if self.canexit: + return (1, "") + raise Exception("No data on serial console socket, connection closed?") if data: if raw:
Signature · 1 line
2.31.1
PATCH 5/15 db: update CVE_PRODUCT
In reply to Anuj MittalFrom: Chen Qi <[email protected]>
Update CVE_PRODUCT to also include 'berkeley_db'. For example, CVE-2020-2981 uses 'berkeley_db'.
--- meta/recipes-support/db/db_5.3.28.bb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
meta/recipes-support/db/db_5.3.28.bb +1 -1
@@ -15,7 +15,7 @@ HOMEPAGE = "https://www.oracle.com/database/technologies/related/berkeleydb.html LICENSE = "Sleepycat" RCONFLICTS_${PN} = "db3" -CVE_PRODUCT = "oracle_berkeley_db" +CVE_PRODUCT = "oracle_berkeley_db berkeley_db" CVE_VERSION = "11.2.${PV}" PR = "r1"
Signature · 1 line
2.31.1
PATCH 6/15 linux-firmware: upgrade 20210208 -> 20210315
In reply to Anuj MittalFrom: Alexander Kanavin <[email protected]>
License-Update: additional firmware files, version changes
---
...{linux-firmware_20210208.bb => linux-firmware_20210315.bb} | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
rename meta/recipes-kernel/linux-firmware/{linux-firmware_20210208.bb => linux-firmware_20210315.bb} (99%)meta/recipes-kernel/linux-firmware/linux-firmware_20210315.bb renamed from meta/recipes-kernel/linux-firmware/linux-firmware_20210208.bb +2 -2
@@ -132,7 +132,7 @@ LIC_FILES_CHKSUM = "file://LICENCE.Abilis;md5=b5ee3f410780e56711ad48eadc22b8bc \ file://LICENCE.xc4000;md5=0ff51d2dc49fce04814c9155081092f0 \ file://LICENCE.xc5000;md5=1e170c13175323c32c7f4d0998d53f66 \ file://LICENCE.xc5000c;md5=12b02efa3049db65d524aeb418dd87ca \ - file://WHENCE;md5=ef0565762eac313c409567b59dff00b2 \ + file://WHENCE;md5=e21a8cbddc1612bce56f06fe154a0743 \ " # These are not common licenses, set NO_GENERIC_LICENSE for them
@@ -205,7 +205,7 @@ PE = "1" SRC_URI = "${KERNELORG_MIRROR}/linux/kernel/firmware/${BPN}-${PV}.tar.xz" -SRC_URI[sha256sum] = "1bcb1a3944c361507754a7d26ccff40ffc28d1fb93bce711d67da26b33e785b7" +SRC_URI[sha256sum] = "a2348f03492713dca9aef202496c6e58f5e63ee5bec6a7bdfcf8b18ce7155e70" inherit allarch
Signature · 1 line
2.31.1
PATCH 7/15 linux-yocto/5.4: qemuppc32: reduce serial shutdown issues
In reply to Anuj MittalFrom: Bruce Ashfield <[email protected]>
Integrating the following commit(s) to linux-yocto/5.4:
qemuppc32: reduce serial issues seen on shutdown
Richard reported:
We've been seeing a lot of the qemuppc shutdown issue and I decided to
look into it. The really worrying thing looking at the logs locally is the
serial ports are showing irq issues and becoming disabled as nobody would
handle them.
Errors like:
[ 9.194886] irq 36: nobody cared (try booting with the "irqpoll" option)
[ 9.198712] CPU: 0 PID: 127 Comm: bootlogd Not tainted
[ 9.202283] Call Trace:
[ 9.205611] [d1005f00] [c00a0da8] __report_bad_irq+0x50/0x138 (unreliable)
[ 9.209347] [d1005f30] [c00a0cc0] note_interrupt+0x324/0x378
[ 9.212855] [d1005f70] [c009d138] handle_irq_event+0xe8/0x104
[ 9.216353] [d1005fa0] [c00a1d9c] handle_fasteoi_irq+0xc0/0x29c
[ 9.219960] [d1005fc0] [c009b798] generic_handle_irq+0x40/0x5c
[ 9.223496] [d1005fd0] [c00075d0] __do_irq+0x58/0x188
[ 9.226948] [d1005ff0] [c0010040] call_do_irq+0x20/0x38
[ 9.230391] [d29eda60] [c0007788] do_IRQ+0x88/0xfc
[ 9.233860] [d29eda90] [c0016454] ret_from_except+0x0/0x14
[ 9.237288] --- interrupt: 501 at __setup_irq+0x3c4/0x838
[ 9.237288] LR = __setup_irq+0x790/0x838
[ 9.244155] [d29edb88] [c009f0a4] request_threaded_irq+0x114/0x1c8
[ 9.247672] [d29edbb8] [c07a5a18] pmz_startup+0x17c/0x32c
[ 9.251203] [d29edbd8] [c07a1140] uart_port_startup+0x184/0x2f8
[ 9.254651] [d29edc08] [c07a1974] uart_port_activate+0x78/0xf4
[ 9.258141] [d29edc28] [c07839f8] tty_port_open+0xd4/0x170
[ 9.261579] [d29edc58] [c079db74] uart_open+0x2c/0x48
[ 9.265116] [d29edc68] [c077a288] tty_open+0x168/0x640
[ 9.268574] [d29edcd8] [c0280be8] chrdev_open+0x138/0x2a4
[ 9.272123] [d29edd18] [c027421c] do_dentry_open+0x228/0x410
[ 9.275643] [d29edd48] [c028e9f4] path_openat+0xb04/0xf28
[ 9.279184] [d29eddd8] [c02917e4] do_filp_open+0x120/0x164
[ 9.282535] [d29ede98] [c0276238] do_sys_openat2+0xd8/0x19c
[ 9.285790] [d29edee8] [c0276574] sys_openat+0x88/0xdc
[ 9.289096] [d29edf38] [c00160d8] ret_from_syscall+0x0/0x34
[ 9.292620] --- interrupt: c01 at 0xfec3738
[ 9.292620] LR = 0xfec36e0
[ 9.299035] handlers:
[ 9.302312] [<7f7f7da8>] pmz_interrupt
[ 9.305541] Disabling IRQ #36
(and the irqpoll option does not help)
This is problematic as the shutdown test uses the serial interface to
shut down the system. If the serial interface fails to login or run the command,
game over for the test.
CONFIG_SERIAL_PMACZILOG_CONSOLE complicates that handling, but doesn't provide
any output or capabilities that we need. So we disable it here, and
reduce the chances of issues during shutdown.--- meta/recipes-kernel/linux/linux-yocto-rt_5.4.bb | 2 +- meta/recipes-kernel/linux/linux-yocto-tiny_5.4.bb | 2 +- meta/recipes-kernel/linux/linux-yocto_5.4.bb | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-)
meta/recipes-kernel/linux/linux-yocto-rt_5.4.bb +1 -1
@@ -12,7 +12,7 @@ python () { } SRCREV_machine ?= "b62ae8bedb024e67e7c5cda51840454a4170c858" -SRCREV_meta ?= "cf20a3c44f5ef181b16ed80e9c2683f76b1ead20" +SRCREV_meta ?= "b89df7433ea8124d3092805391b78808df4147a7" SRC_URI = "git://git.yoctoproject.org/linux-yocto.git;branch=${KBRANCH};name=machine \ git://git.yoctoproject.org/yocto-kernel-cache;type=kmeta;name=meta;branch=yocto-5.4;destsuffix=${KMETA}"
meta/recipes-kernel/linux/linux-yocto-tiny_5.4.bb +1 -1
@@ -17,7 +17,7 @@ KCONF_BSP_AUDIT_LEVEL = "2" SRCREV_machine_qemuarm ?= "80bd6016a9bdaed4b66ddffffa8c8e62d7c1f8a6" SRCREV_machine ?= "ea7a54fa402727f3c4bc4a1904d4a9590e7c8b85" -SRCREV_meta ?= "cf20a3c44f5ef181b16ed80e9c2683f76b1ead20" +SRCREV_meta ?= "b89df7433ea8124d3092805391b78808df4147a7" PV = "${LINUX_VERSION}+git${SRCPV}"
meta/recipes-kernel/linux/linux-yocto_5.4.bb +1 -1
@@ -21,7 +21,7 @@ SRCREV_machine_qemux86 ?= "ea7a54fa402727f3c4bc4a1904d4a9590e7c8b85" SRCREV_machine_qemux86-64 ?= "ea7a54fa402727f3c4bc4a1904d4a9590e7c8b85" SRCREV_machine_qemumips64 ?= "b36d79d6f2aaf9dadec352f611e7b9becf2b9a55" SRCREV_machine ?= "ea7a54fa402727f3c4bc4a1904d4a9590e7c8b85" -SRCREV_meta ?= "cf20a3c44f5ef181b16ed80e9c2683f76b1ead20" +SRCREV_meta ?= "b89df7433ea8124d3092805391b78808df4147a7" # remap qemuarm to qemuarma15 for the 5.4 kernel # KMACHINE_qemuarm ?= "qemuarma15"
Signature · 1 line
2.31.1
PATCH 8/15 lttng-modules: backport patches to fix build against 5.12+ kernel
In reply to Anuj MittalFrom: Vinícius Ossanes Aquino <[email protected]>
Add the following patches from stable-2.12 branch of lttng repository to fix errors when building lttng-modules against 5.12+ kernel since they are not present on the release 2.12.5:
- 17cd2dc9 fix: block: add a disk_uevent helper (v5.12)
- 127135b6 fix backport: block: add a disk_uevent helper (v5.12)
- 853d5903 fix: mm, tracing: kfree event name mismatching with provider kmem (v5.12)
--- ...block-add-a-disk_uevent-helper-v5.12.patch | 305 ++++++++++++++++++ ...block-add-a-disk_uevent-helper-v5.12.patch | 48 +++ ...free-event-name-mismatching-with-pro.patch | 71 ++++ .../lttng/lttng-modules_2.12.5.bb | 3 + 4 files changed, 427 insertions(+) create mode 100644 meta/recipes-kernel/lttng/lttng-modules/0005-fix-block-add-a-disk_uevent-helper-v5.12.patch create mode 100644 meta/recipes-kernel/lttng/lttng-modules/0006-fix-backport-block-add-a-disk_uevent-helper-v5.12.patch create mode 100644 meta/recipes-kernel/lttng/lttng-modules/0007-fix-mm-tracing-kfree-event-name-mismatching-with-pro.patch
meta/recipes-kernel/lttng/lttng-modules/0005-fix-block-add-a-disk_uevent-helper-v5.12.patch new file +305 -0
@@ -0,0 +1,305 @@ +From 17cd2dc91cb82ed342b0da699f2b1a70c1bf6a03 Mon Sep 17 00:00:00 2001 +From: Michael Jeanson <[email protected]> +Date: Mon, 15 Mar 2021 14:54:02 -0400 +Subject: [PATCH 2/4] fix: block: add a disk_uevent helper (v5.12) + +See upstream commit: + + commit bc359d03c7ec1bf3b86d03bafaf6bbb21e6414fd + Author: Christoph Hellwig <[email protected]> + Date: Sun Jan 24 11:02:39 2021 +0100 + + block: add a disk_uevent helper + + Add a helper to call kobject_uevent for the disk and all partitions, and + unexport the disk_part_iter_* helpers that are now only used in the core + block code. + +Upstream-status: Backport [2.12.6] + +Change-Id: If6e8797049642ab382d5699660ee1dd734e92c90 +Signed-off-by: Michael Jeanson <[email protected]> +Signed-off-by: Mathieu Desnoyers <[email protected]> +--- + Makefile | 1 + + lttng-statedump-impl.c | 34 +++++++++---- + src/wrapper/genhd.c | 111 +++++++++++++++++++++++++++++++++++++++++ + wrapper/genhd.h | 62 +++++++++++++++++++++++ + 4 files changed, 198 insertions(+), 10 deletions(-) + create mode 100644 src/wrapper/genhd.c + +diff --git a/Makefile b/Makefile +index a9aff3f1..34043cfb 100644 +--- a/Makefile ++++ b/Makefile +@@ -80,6 +80,7 @@ ifneq ($(KERNELRELEASE),) + wrapper/kallsyms.o \ + wrapper/irqdesc.o \ + wrapper/fdtable.o \ ++ wrapper/genhd.o \ + lttng-wrapper-impl.o + + ifneq ($(CONFIG_HAVE_SYSCALL_TRACEPOINTS),) +diff --git a/lttng-statedump-impl.c b/lttng-statedump-impl.c +index 60b937c9..5511c7e8 100644 +--- a/lttng-statedump-impl.c ++++ b/lttng-statedump-impl.c +@@ -250,13 +250,17 @@ int lttng_enumerate_block_devices(struct lttng_session *session) + struct device_type *ptr_disk_type; + struct class_dev_iter iter; + struct device *dev; ++ int ret = 0; + + ptr_block_class = wrapper_get_block_class(); +- if (!ptr_block_class) +- return -ENOSYS; ++ if (!ptr_block_class) { ++ ret = -ENOSYS; ++ goto end; ++ } + ptr_disk_type = wrapper_get_disk_type(); + if (!ptr_disk_type) { +- return -ENOSYS; ++ ret = -ENOSYS; ++ goto end; + } + class_dev_iter_init(&iter, ptr_block_class, NULL, ptr_disk_type); + while ((dev = class_dev_iter_next(&iter))) { +@@ -272,22 +276,32 @@ int lttng_enumerate_block_devices(struct lttng_session *session) + (disk->flags & GENHD_FL_SUPPRESS_PARTITION_INFO)) + continue; + +- disk_part_iter_init(&piter, disk, DISK_PITER_INCL_PART0); +- while ((part = disk_part_iter_next(&piter))) { ++ /* ++ * The original 'disk_part_iter_init' returns void, but our ++ * wrapper can fail to lookup the original symbol. ++ */ ++ if (wrapper_disk_part_iter_init(&piter, disk, DISK_PITER_INCL_PART0) < 0) { ++ ret = -ENOSYS; ++ goto iter_exit; ++ } ++ ++ while ((part = wrapper_disk_part_iter_next(&piter))) { + char name_buf[BDEVNAME_SIZE]; + + if (lttng_get_part_name(disk, part, name_buf) == -ENOSYS) { +- disk_part_iter_exit(&piter); +- class_dev_iter_exit(&iter); +- return -ENOSYS; ++ wrapper_disk_part_iter_exit(&piter); ++ ret = -ENOSYS; ++ goto iter_exit; + } + trace_lttng_statedump_block_device(session, + lttng_get_part_devt(part), name_buf); + } +- disk_part_iter_exit(&piter); ++ wrapper_disk_part_iter_exit(&piter); + } ++iter_exit: + class_dev_iter_exit(&iter); +- return 0; ++end: ++ return ret; + } + + #ifdef CONFIG_INET +diff --git a/src/wrapper/genhd.c b/src/wrapper/genhd.c +new file mode 100644 +index 00000000..a5a6c410 +--- /dev/null ++++ b/src/wrapper/genhd.c +@@ -0,0 +1,111 @@ ++/* SPDX-License-Identifier: (GPL-2.0-only OR LGPL-2.1-only) ++ * ++ * wrapper/genhd.c ++ * ++ * Wrapper around disk_part_iter_(init|next|exit). Using KALLSYMS to get the ++ * addresses when available, else we need to have a kernel that exports this ++ * function to GPL modules. This export was removed in 5.12. ++ * ++ * Copyright (C) 2021 Michael Jeanson <[email protected]> ++ */ ++ ++#include <lttng/kernel-version.h> ++#include <linux/module.h> ++#include <wrapper/genhd.h> ++ ++#if (defined(CONFIG_KALLSYMS) && \ ++ (LTTNG_LINUX_VERSION_CODE >= LTTNG_KERNEL_VERSION(5,12,0))) ++ ++#include <wrapper/kallsyms.h> ++ ++static ++void (*disk_part_iter_init_sym)(struct disk_part_iter *piter, struct gendisk *disk, ++ unsigned int flags); ++ ++static ++LTTNG_DISK_PART_TYPE *(*disk_part_iter_next_sym)(struct disk_part_iter *piter); ++ ++static ++void (*disk_part_iter_exit_sym)(struct disk_part_iter *piter); ++ ++/* ++ * This wrapper has an 'int' return type instead of the original 'void', to be ++ * able to report the symbol lookup failure to the caller. ++ * ++ * Return 0 on success, -1 on error. ++ */ ++int wrapper_disk_part_iter_init(struct disk_part_iter *piter, struct gendisk *disk, ++ unsigned int flags) ++{ ++ if (!disk_part_iter_init_sym) ++ disk_part_iter_init_sym = (void *) kallsyms_lookup_funcptr("disk_part_iter_init"); ++ ++ if (disk_part_iter_init_sym) { ++ disk_part_iter_init_sym(piter, disk, flags); ++ } else { ++ printk_once(KERN_WARNING "LTTng: disk_part_iter_init symbol lookup failed.\n"); ++ return -1; ++ } ++ return 0; ++} ++EXPORT_SYMBOL_GPL(wrapper_disk_part_iter_init); ++ ++LTTNG_DISK_PART_TYPE *wrapper_disk_part_iter_next(struct disk_part_iter *piter) ++{ ++ if (!disk_part_iter_next_sym) ++ disk_part_iter_next_sym = (void *) kallsyms_lookup_funcptr("disk_part_iter_next"); ++ ++ if (disk_part_iter_next_sym) { ++ return disk_part_iter_next_sym(piter); ++ } else { ++ printk_once(KERN_WARNING "LTTng: disk_part_iter_next symbol lookup failed.\n"); ++ return NULL; ++ } ++} ++EXPORT_SYMBOL_GPL(wrapper_disk_part_iter_next); ++ ++/* ++ * We don't return an error on symbol lookup failure here because there is ++ * nothing the caller can do to cleanup the iterator. ++ */ ++void wrapper_disk_part_iter_exit(struct disk_part_iter *piter) ++{ ++ if (!disk_part_iter_exit_sym) ++ disk_part_iter_exit_sym = (void *) kallsyms_lookup_funcptr("disk_part_iter_exit"); ++ ++ if (disk_part_iter_exit_sym) { ++ disk_part_iter_exit_sym(piter); ++ } else { ++ printk_once(KERN_WARNING "LTTng: disk_part_iter_exit symbol lookup failed.\n"); ++ } ++} ++EXPORT_SYMBOL_GPL(wrapper_disk_part_iter_exit); ++ ++#else ++ ++/* ++ * This wrapper has an 'int' return type instead of the original 'void', so the ++ * kallsyms variant can report the symbol lookup failure to the caller. ++ * ++ * This variant always succeeds and returns 0. ++ */ ++int wrapper_disk_part_iter_init(struct disk_part_iter *piter, struct gendisk *disk, ++ unsigned int flags) ++{ ++ disk_part_iter_init(piter, disk, flags); ++ return 0; ++} ++EXPORT_SYMBOL_GPL(wrapper_disk_part_iter_init); ++ ++LTTNG_DISK_PART_TYPE *wrapper_disk_part_iter_next(struct disk_part_iter *piter) ++{ ++ return disk_part_iter_next(piter); ++} ++EXPORT_SYMBOL_GPL(wrapper_disk_part_iter_next); ++ ++void wrapper_disk_part_iter_exit(struct disk_part_iter *piter) ++{ ++ disk_part_iter_exit(piter); ++} ++EXPORT_SYMBOL_GPL(wrapper_disk_part_iter_exit); ++#endif +diff --git a/wrapper/genhd.h b/wrapper/genhd.h +index 98feb57b..6bae239d 100644 +--- a/wrapper/genhd.h ++++ b/wrapper/genhd.h +@@ -13,6 +13,13 @@ + #define _LTTNG_WRAPPER_GENHD_H + + #include <linux/genhd.h> ++#include <lttng/kernel-version.h> ++ ++#if (LTTNG_LINUX_VERSION_CODE >= LTTNG_KERNEL_VERSION(5,11,0)) ++#define LTTNG_DISK_PART_TYPE struct block_device ++#else ++#define LTTNG_DISK_PART_TYPE struct hd_struct ++#endif + + #ifdef CONFIG_KALLSYMS_ALL + +@@ -94,4 +101,59 @@ struct device_type *wrapper_get_disk_type(void) + + #endif + ++/* ++ * This wrapper has an 'int' return type instead of the original 'void', to be ++ * able to report the symbol lookup failure to the caller. ++ * ++ * Return 0 on success, -1 on error. ++ */ ++int wrapper_disk_part_iter_init(struct disk_part_iter *piter, struct gendisk *disk, ++ unsigned int flags); ++LTTNG_DISK_PART_TYPE *wrapper_disk_part_iter_next(struct disk_part_iter *piter); ++void wrapper_disk_part_iter_exit(struct disk_part_iter *piter); ++ ++/* ++ * Canary function to check for 'disk_part_iter_init()' at compile time. ++ * ++ * From 'include/linux/genhd.h': ++ * ++ * extern void disk_part_iter_init(struct disk_part_iter *piter, ++ * struct gendisk *disk, unsigned int flags); ++ * ++ */ ++static inline ++void __canary__disk_part_iter_init(struct disk_part_iter *piter, struct gendisk *disk, ++ unsigned int flags) ++{ ++ disk_part_iter_init(piter, disk, flags); ++} ++ ++/* ++ * Canary function to check for 'disk_part_iter_next()' at compile time. ++ * ++ * From 'include/linux/genhd.h': ++ * ++ * struct block_device *disk_part_iter_next(struct disk_part_iter *piter); ++ * ++ */ ++static inline ++LTTNG_DISK_PART_TYPE *__canary__disk_part_iter_next(struct disk_part_iter *piter) ++{ ++ return disk_part_iter_next(piter); ++} ++ ++/* ++ * Canary function to check for 'disk_part_iter_exit()' at compile time. ++ * ++ * From 'include/linux/genhd.h': ++ * ++ * extern void disk_part_iter_exit(struct disk_part_iter *piter); ++ * ++ */ ++static inline ++void __canary__disk_part_iter_exit(struct disk_part_iter *piter) ++{ ++ return disk_part_iter_exit(piter); ++} ++ + #endif /* _LTTNG_WRAPPER_GENHD_H */ +-- +2.25.1 +
meta/recipes-kernel/lttng/lttng-modules/0006-fix-backport-block-add-a-disk_uevent-helper-v5.12.patch new file +48 -0
48 lines not on this page · raw message
meta/recipes-kernel/lttng/lttng-modules/0007-fix-mm-tracing-kfree-event-name-mismatching-with-pro.patch new file +71 -0
71 lines not on this page · raw message
meta/recipes-kernel/lttng/lttng-modules_2.12.5.bb +3 -0
@@ -15,6 +15,9 @@ SRC_URI = "https://lttng.org/files/${BPN}/${BPN}-${PV}.tar.bz2 \ file://0002-Fix-filter-interpreter-early-exits-on-uninitialized-.patch \ file://0003-fix-mm-tracing-record-slab-name-for-kmem_cache_free-.patch \ file://0004-Fix-kretprobe-null-ptr-deref-on-session-destroy.patch \ + file://0005-fix-block-add-a-disk_uevent-helper-v5.12.patch \ + file://0006-fix-backport-block-add-a-disk_uevent-helper-v5.12.patch \ + file://0007-fix-mm-tracing-kfree-event-name-mismatching-with-pro.patch \ " SRC_URI[sha256sum] = "c4d1a1b42c728e37b6b7947ae16563a011c4b297311aa04d56f9a1791fb5a30a"
Signature · 1 line
2.31.1
PATCH 9/15 libcap: Configure Make variables correctly without a horrible hack
In reply to Anuj MittalFrom: Peter Kjellerstedt <[email protected]>
Occasionally, the build would fail with:
make[2]: execvp: mkdir: Argument list too long
This turned out to be due to a hacky solution used in the recipe to modify the Makefile, which resulted in one more $(BUILD_CFLAGS) being added to the immediately expanded BUILD_CFLAGS Make variable each time do_configure was executed. After a couple of times, this lead to an environment with a 140 kB BUILD_CFLAGS when mkdir should execute, which resulted in the E2BIG.
--- meta/recipes-support/libcap/libcap_2.48.bb | 21 ++++++++------------- 1 file changed, 8 insertions(+), 13 deletions(-)
meta/recipes-support/libcap/libcap_2.48.bb +8 -13
@@ -20,15 +20,6 @@ UPSTREAM_CHECK_URI = "https://www.kernel.org/pub/linux/libs/security/linux-privs inherit lib_package -# do NOT pass target cflags to host compilations -# -do_configure() { - # libcap uses := for compilers, fortunately, it gives us a hint - # on what should be replaced with ?= - sed -e 's,:=,?=,g' -i Make.Rules - sed -e 's,^BUILD_CFLAGS ?= ,BUILD_CFLAGS := $(BUILD_CFLAGS) ,' -i Make.Rules -} - PACKAGECONFIG ??= "${@bb.utils.filter('DISTRO_FEATURES', 'pam', d)}" PACKAGECONFIG_class-native ??= ""
@@ -44,11 +35,15 @@ EXTRA_OEMAKE = " \ EXTRA_OEMAKE_append_class-target = " SYSTEM_HEADERS=${STAGING_INCDIR}" -# these are present in the libcap defaults, so include in our CFLAGS too -CFLAGS += "-D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64" - do_compile() { - oe_runmake ${PACKAGECONFIG_CONFARGS} + unset CFLAGS BUILD_CFLAGS + oe_runmake \ + ${PACKAGECONFIG_CONFARGS} \ + AR="${AR}" \ + CC="${CC}" \ + RANLIB="${RANLIB}" \ + COPTS="${CFLAGS}" \ + BUILD_COPTS="${BUILD_CFLAGS}" } do_install() {
Signature · 1 line
2.31.1
PATCH 10/15 dejagnu: needs expect at runtime
In reply to Anuj MittalFrom: Romain Naour <[email protected]>
runtest return an error due to missing expect on the target. Add expect as runtime dependency.
--- meta/recipes-devtools/dejagnu/dejagnu_1.6.2.bb | 1 + 1 file changed, 1 insertion(+)
meta/recipes-devtools/dejagnu/dejagnu_1.6.2.bb +1 -0
@@ -7,6 +7,7 @@ LIC_FILES_CHKSUM = "file://COPYING;md5=d32239bcb673463ab874e80d47fae504" SECTION = "devel" DEPENDS += "expect-native" +RDEPENDS_${PN} = "expect" inherit autotools
Signature · 1 line
2.31.1
PATCH 11/15 grub2.inc: remove '-O2' from CFLAGS
In reply to Anuj MittalFrom: Kai Kang <[email protected]>
It fails to boot grub after upgrade grub to 2.06. According to description in
https://bugzilla.yoctoproject.org/show_bug.cgi?id=14367
it is introduced by a commit to fix CVE. So remove option '-O2' from CFLAGS rather than revert the commit to avoid the failure.
[YOCTO #14367]
--- meta/recipes-bsp/grub/grub2.inc | 2 ++ 1 file changed, 2 insertions(+)
meta/recipes-bsp/grub/grub2.inc +2 -0
@@ -49,6 +49,8 @@ GRUBPLATFORM ??= "pc" inherit autotools gettext texinfo pkgconfig +CFLAGS_remove = "-O2" + EXTRA_OECONF = "--with-platform=${GRUBPLATFORM} \ --disable-grub-mkfont \ --program-prefix="" \
Signature · 1 line
2.31.1
PATCH 12/15 ruby: upgrade 3.0.0 -> 3.0.1
In reply to Anuj MittalFrom: zhengruoqin <[email protected]>
---
meta/recipes-devtools/ruby/{ruby_3.0.0.bb => ruby_3.0.1.bb} | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
rename meta/recipes-devtools/ruby/{ruby_3.0.0.bb => ruby_3.0.1.bb} (97%)meta/recipes-devtools/ruby/ruby_3.0.1.bb renamed from meta/recipes-devtools/ruby/ruby_3.0.0.bb +1 -1
@@ -8,7 +8,7 @@ SRC_URI += " \ file://0001-template-Makefile.in-do-not-write-host-cross-cc-item.patch \ " -SRC_URI[sha256sum] = "a13ed141a1c18eb967aac1e33f4d6ad5f21be1ac543c344e0d6feeee54af8e28" +SRC_URI[sha256sum] = "369825db2199f6aeef16b408df6a04ebaddb664fb9af0ec8c686b0ce7ab77727" PACKAGECONFIG ??= "" PACKAGECONFIG += "${@bb.utils.filter('DISTRO_FEATURES', 'ipv6', d)}"
Signature · 1 line
2.31.1
PATCH 13/15 lsb-release: fix reproducibility failure
In reply to Anuj MittalMake sure help2man output is reproducible. Fixes:
| .\"·DO·NOT·MODIFY·THIS·FILE!··It·was·generated·by·help2man·1.022. .\"·DO·NOT·MODIFY·THIS·FILE!··It·was·generated·by·help2man·1.022. | .TH·FSG·"1"·"April·2021"·"FSG·lsb_release·v1.4"·FSG .TH·FSG·"1"·"May·2021"·"FSG·lsb_release·v1.4"·FSG | .SH·NAME 3 .SH·NAME
--- .../help2man-reproducibility.patch | 27 +++++++++++++++++++ meta/recipes-extended/lsb/lsb-release_1.4.bb | 1 + 2 files changed, 28 insertions(+) create mode 100644 meta/recipes-extended/lsb/lsb-release/help2man-reproducibility.patch
meta/recipes-extended/lsb/lsb-release/help2man-reproducibility.patch new file +27 -0
@@ -0,0 +1,27 @@ +lsb-release maintains it's own copy of help2man. Include the support +for specifying SOURCE_DATE_EPOCH from upstream. + +Upstream-Status: Pending + +Signed-off-by: Anuj Mittal <[email protected]> + +diff --git a/help2man b/help2man +index 13015c2..63439db 100755 +--- a/help2man ++++ b/help2man +@@ -173,7 +173,14 @@ my ($help_text, $version_text) = map { + or die "$this_program: can't get `--$_' info from $ARGV[0]\n" + } qw(help), $opt_version_key; + +-my $date = strftime "%B %Y", localtime; ++my $epoch_secs = time; ++if (exists $ENV{SOURCE_DATE_EPOCH} and $ENV{SOURCE_DATE_EPOCH} =~ /^(\d+)$/) ++{ ++ $epoch_secs = $1; ++ $ENV{TZ} = 'UTC0'; ++} ++ ++my $date = strftime "%B %Y", localtime $epoch_secs; + (my $program = $ARGV[0]) =~ s!.*/!!; + my $package = $program; + my $version;
meta/recipes-extended/lsb/lsb-release_1.4.bb +1 -0
@@ -11,6 +11,7 @@ LIC_FILES_CHKSUM = "file://README;md5=12da544b1a3a5a1795a21160b49471cf" SRC_URI = "${SOURCEFORGE_MIRROR}/project/lsb/lsb_release/1.4/lsb-release-1.4.tar.gz \ file://0001-fix-lsb_release-to-work-with-busybox-head-and-find.patch \ file://0001-Remove-timestamp-from-manpage.patch \ + file://help2man-reproducibility.patch \ " SRC_URI[md5sum] = "30537ef5a01e0ca94b7b8eb6a36bb1e4"
Signature · 1 line
2.31.1
PATCH 14/15 linux-firmware: include all relevant files in -bcm4356
In reply to Anuj MittalFrom: Yann Dirson <[email protected]>
This currently catches the .clb_blob and .vamrs,rock960.txt, and other .txt files may come in future upstream releases.
--- meta/recipes-kernel/linux-firmware/linux-firmware_20210315.bb | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
meta/recipes-kernel/linux-firmware/linux-firmware_20210315.bb +2 -2
@@ -645,8 +645,8 @@ FILES_${PN}-bcm43455 = "${nonarch_base_libdir}/firmware/brcm/brcmfmac43455-sdio. " FILES_${PN}-bcm4350c2 = "${nonarch_base_libdir}/firmware/brcm/brcmfmac4350c2-pcie.bin" FILES_${PN}-bcm4350 = "${nonarch_base_libdir}/firmware/brcm/brcmfmac4350-pcie.bin" -FILES_${PN}-bcm4356 = "${nonarch_base_libdir}/firmware/brcm/brcmfmac4356-sdio.bin \ - ${nonarch_base_libdir}/firmware/cypress/cyfmac4356-sdio.bin \ +FILES_${PN}-bcm4356 = "${nonarch_base_libdir}/firmware/brcm/brcmfmac4356-sdio.* \ + ${nonarch_base_libdir}/firmware/cypress/cyfmac4356-sdio.* \ " FILES_${PN}-bcm43569 = "${nonarch_base_libdir}/firmware/brcm/brcmfmac43569.bin" FILES_${PN}-bcm43570 = "${nonarch_base_libdir}/firmware/brcm/brcmfmac43570-pcie.bin \
Signature · 1 line
2.31.1
PATCH 15/15 puzzles: Upstream changed to main branch for development
In reply to Anuj MittalFrom: Richard Purdie <[email protected]>
--- meta/recipes-sato/puzzles/puzzles_git.bb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
meta/recipes-sato/puzzles/puzzles_git.bb +1 -1
@@ -9,7 +9,7 @@ DEPENDS = "libxt" # The libxt requires x11 in DISTRO_FEATURES REQUIRED_DISTRO_FEATURES = "x11" -SRC_URI = "git://git.tartarus.org/simon/puzzles.git \ +SRC_URI = "git://git.tartarus.org/simon/puzzles.git;branch=main \ file://fix-compiling-failure-with-option-g-O.patch \ file://0001-palisade-Fix-warnings-with-clang-on-arm.patch \ file://0001-Use-Wno-error-format-overflow-if-the-compiler-suppor.patch \
Signature · 1 line
2.31.1