From: Ashish Kumar Mishra <git-patches@bmwtechworks.in>
To: openembedded-core@lists.openembedded.org
Cc: Ashish Kumar Mishra <git-patches@bmwtechworks.in>,
AshishKumar Mishra <ashishkumar.mishra@bmwtechworks.in>
Subject: [PATCH] shared/fdset: add detailed debug logging to fdset_new_fill()
Date: Tue, 10 Mar 2026 13:39:42 +0530 [thread overview]
Message-ID: <20260310080942.47334-1-git-patches@bmwtechworks.in> (raw)
Currently, when fdset_new_fill() fails to open /proc/self/fd or
encounters an error while processing individual file descriptors
(such as fcntl or fstat failures), it returns a silent error code.
For debugging rarely reproducible failures it becomes difficult to
know the exact cause of failure
This commit updates the function to use log_debug_errno() for all
error paths and hence provides better visibility into why FD collection
failed, including the path of the problematic FD (via fd_get_path)
and its inode type.
Upstream-Status: Backport [https://github.com/systemd/systemd/pull/40385]
Signed-off-by: AshishKumar Mishra <ashishkumar.mishra@bmwtechworks.in>
---
...-detailed-debug-logging-to-fdset_new.patch | 97 +++++++++++++++++++
meta/recipes-core/systemd/systemd_259.3.bb | 1 +
2 files changed, 98 insertions(+)
create mode 100644 meta/recipes-core/systemd/systemd/0018-shared-fdset-add-detailed-debug-logging-to-fdset_new.patch
diff --git a/meta/recipes-core/systemd/systemd/0018-shared-fdset-add-detailed-debug-logging-to-fdset_new.patch b/meta/recipes-core/systemd/systemd/0018-shared-fdset-add-detailed-debug-logging-to-fdset_new.patch
new file mode 100644
index 0000000000..849f5cfda1
--- /dev/null
+++ b/meta/recipes-core/systemd/systemd/0018-shared-fdset-add-detailed-debug-logging-to-fdset_new.patch
@@ -0,0 +1,97 @@
+From 0565f9f27323a8f9e62d85f2add542af99cea06a Mon Sep 17 00:00:00 2001
+From: AshishKumar Mishra <ashishkumar.mishra@bmwtechworks.in>
+Date: Wed, 21 Jan 2026 14:13:29 +0530
+Subject: [PATCH] shared/fdset: add detailed debug logging to fdset_new_fill()
+
+Currently, when fdset_new_fill() fails to open /proc/self/fd or
+encounters an error while processing individual file descriptors
+(such as fcntl or fstat failures), it returns a silent error code.
+
+For debugging rarely reproducible failures it becomes difficult to
+know the exact cause of failure
+This commit updates the function to use log_debug_errno() for all
+error paths and hence provides better visibility into why FD collection
+failed, including the path of the problematic FD (via fd_get_path)
+and its inode type.
+
+Upstream-Status: Backport [https://github.com/systemd/systemd/pull/40385]
+
+Signed-off-by: AshishKumar Mishra <ashishkumar.mishra@bmwtechworks.in>
+---
+ src/shared/fdset.c | 35 ++++++++++++++++++++++++++---------
+ 1 file changed, 26 insertions(+), 9 deletions(-)
+
+diff --git a/src/shared/fdset.c b/src/shared/fdset.c
+index 832e7fda60..f340f41b0e 100644
+--- a/src/shared/fdset.c
++++ b/src/shared/fdset.c
+@@ -8,6 +8,7 @@
+ #include "alloc-util.h"
+ #include "async.h"
+ #include "dirent-util.h"
++#include "errno-util.h"
+ #include "fd-util.h"
+ #include "fdset.h"
+ #include "log.h"
+@@ -179,9 +180,10 @@ int fdset_new_fill(
+ d = opendir("/proc/self/fd");
+ if (!d) {
+ if (errno == ENOENT && proc_mounted() == 0)
+- return -ENOSYS;
++ return log_debug_errno(SYNTHETIC_ERRNO(ENOSYS),
++ "Failed to open /proc/self/fd/, /proc/ is not mounted.");
+
+- return -errno;
++ return log_debug_errno(errno, "Failed to open /proc/self/fd/: %m ");
+ }
+
+ s = fdset_new();
+@@ -210,9 +212,14 @@ int fdset_new_fill(
+ * been passed in can be collected and fds which have been created locally can be
+ * ignored, under the assumption that only the latter have O_CLOEXEC set. */
+
+- fl = fcntl(fd, F_GETFD);
+- if (fl < 0)
+- return -errno;
++ fl = RET_NERRNO(fcntl(fd, F_GETFD));
++ if (fl < 0) {
++ _cleanup_free_ char *path = NULL;
++ (void) fd_get_path(fd, &path);
++ return log_debug_errno(fl,
++ "Failed to get flag of fd=%d (%s): %m ",
++ fd, strna(path));
++ }
+
+ if (FLAGS_SET(fl, FD_CLOEXEC) != !!filter_cloexec)
+ continue;
+@@ -221,13 +228,23 @@ int fdset_new_fill(
+ /* We need to set CLOEXEC manually only if we're collecting non-CLOEXEC fds. */
+ if (filter_cloexec <= 0) {
+ r = fd_cloexec(fd, true);
+- if (r < 0)
+- return r;
++ if (r < 0) {
++ _cleanup_free_ char *path = NULL;
++ (void) fd_get_path(fd, &path);
++ return log_debug_errno(r,
++ "Failed to set CLOEXEC flag fd=%d (%s): %m ",
++ fd, strna(path));
++ }
+ }
+
+ r = fdset_put(s, fd);
+- if (r < 0)
+- return r;
++ if (r < 0) {
++ _cleanup_free_ char *path = NULL;
++ (void) fd_get_path(fd, &path);
++ return log_debug_errno(r,
++ "Failed to put fd=%d (%s) into fdset: %m ",
++ fd, strna(path));
++ }
+ }
+
+ *ret = TAKE_PTR(s);
+--
+2.34.1
+
diff --git a/meta/recipes-core/systemd/systemd_259.3.bb b/meta/recipes-core/systemd/systemd_259.3.bb
index bcb8fe3d9c..44d319d0c5 100644
--- a/meta/recipes-core/systemd/systemd_259.3.bb
+++ b/meta/recipes-core/systemd/systemd_259.3.bb
@@ -33,6 +33,7 @@ SRC_URI += " \
file://0001-binfmt-Don-t-install-dependency-links-at-install-tim.patch \
file://0002-implment-systemd-sysv-install-for-OE.patch \
file://0003-Do-not-create-var-log-README.patch \
+ file://0018-shared-fdset-add-detailed-debug-logging-to-fdset_new.patch \
"
PAM_PLUGINS = " \
--
2.34.1
next reply other threads:[~2026-03-10 8:29 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-10 8:09 Ashish Kumar Mishra [this message]
2026-03-12 7:45 ` [OE-core] [PATCH] shared/fdset: add detailed debug logging to fdset_new_fill() Yoann Congal
2026-03-12 8:11 ` Ashish Mishra
2026-03-13 17:03 ` [OE-core] " Peter Kjellerstedt
2026-03-18 10:13 ` Ashish Mishra
-- strict thread matches above, loose matches on Subject: below --
2026-03-03 8:12 Ashish Kumar Mishra
2026-03-09 4:30 ` AshishKumar Mishra
2026-03-09 7:26 ` [OE-core] " Mathieu Dubois-Briand
2026-03-10 8:51 ` AshishKumar Mishra
2026-03-12 7:33 ` Ashish Mishra
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20260310080942.47334-1-git-patches@bmwtechworks.in \
--to=git-patches@bmwtechworks.in \
--cc=ashishkumar.mishra@bmwtechworks.in \
--cc=openembedded-core@lists.openembedded.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox