| Message ID | 6d425cce-dc62-4221-a87d-c15598911e84@siemens.com |
|---|---|
| State | Accepted, archived |
| Headers | show |
| Series | image: Restrict coredumps to *.core files again | expand |
Make sense to me. ---------------- With best regards, Zhibin Dong Siemens Ltd., China DI FA CTR SVC&AI CN Tianyuan road No.99 611731 CHENGDU, China Mobile: +86 15520735269 mailto:zhibin.dong@siemens.com www.siemens.com -----Original Message----- From: Kiszka, Jan (T CED) <jan.kiszka@siemens.com> Sent: Friday, June 28, 2024 1:05 AM To: isar-users <isar-users@googlegroups.com> Cc: Dong, Zhi Bin (DI FA CTR SVC&AI CN) <ZhiBin.Dong@siemens.com>; Moessbauer, Felix (T CED OES-DE) <felix.moessbauer@siemens.com> Subject: [PATCH] image: Restrict coredumps to *.core files again From: Jan Kiszka <jan.kiszka@siemens.com> Looking at every file's mime-type in a rootfs is simply way too costly and also not needed in practice. Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> --- To make my point more concrete. Please speak up if you think that coredumps with different endings do play a role. But make sure that case is strong: - 7.5 min. do_rootfs_finalize for meta-iot2050 so far - 0.5 min. with this change meta/classes/image.bbclass | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/meta/classes/image.bbclass b/meta/classes/image.bbclass index 0a80273f..c29d9e26 100644 --- a/meta/classes/image.bbclass +++ b/meta/classes/image.bbclass @@ -436,7 +436,7 @@ EOSUDO # Sometimes qemu-user-static generates coredumps in chroot, move them # to work temporary directory and inform user about it. - for f in $(sudo find ${ROOTFSDIR} -type f -exec file --mime-type {} \; | grep 'application/x-coredump' | cut -d: -f1); do + for f in $(sudo find ${ROOTFSDIR} -type f -name *.core -exec file + --mime-type {} \; | grep 'application/x-coredump' | cut -d: -f1); do sudo mv "${f}" "${WORKDIR}/temp/" bbwarn "found core dump in rootfs, check it in ${WORKDIR}/temp/${f##*/}" done -- 2.43.0
On Thu, 2024-06-27 at 19:04 +0200, 'Jan Kiszka' via isar-users wrote: > From: Jan Kiszka <jan.kiszka@siemens.com> > > Looking at every file's mime-type in a rootfs is simply way too > costly > and also not needed in practice. > > Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> > --- > > To make my point more concrete. Please speak up if you think that > coredumps with different endings do play a role. But make sure that > case > is strong: > > - 7.5 min. do_rootfs_finalize for meta-iot2050 so far > - 0.5 min. with this change > > meta/classes/image.bbclass | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/meta/classes/image.bbclass b/meta/classes/image.bbclass > index 0a80273f..c29d9e26 100644 > --- a/meta/classes/image.bbclass > +++ b/meta/classes/image.bbclass > @@ -436,7 +436,7 @@ EOSUDO > > # Sometimes qemu-user-static generates coredumps in chroot, move > them > # to work temporary directory and inform user about it. > - for f in $(sudo find ${ROOTFSDIR} -type f -exec file --mime-type > {} \; | grep 'application/x-coredump' | cut -d: -f1); do > + for f in $(sudo find ${ROOTFSDIR} -type f -name *.core -exec > file --mime-type {} \; | grep 'application/x-coredump' | cut -d: - > f1); do > sudo mv "${f}" "${WORKDIR}/temp/" > bbwarn "found core dump in rootfs, check it in > ${WORKDIR}/temp/${f##*/}" > done > -- > 2.43.0 > Applied to next, thanks.
diff --git a/meta/classes/image.bbclass b/meta/classes/image.bbclass index 0a80273f..c29d9e26 100644 --- a/meta/classes/image.bbclass +++ b/meta/classes/image.bbclass @@ -436,7 +436,7 @@ EOSUDO # Sometimes qemu-user-static generates coredumps in chroot, move them # to work temporary directory and inform user about it. - for f in $(sudo find ${ROOTFSDIR} -type f -exec file --mime-type {} \; | grep 'application/x-coredump' | cut -d: -f1); do + for f in $(sudo find ${ROOTFSDIR} -type f -name *.core -exec file --mime-type {} \; | grep 'application/x-coredump' | cut -d: -f1); do sudo mv "${f}" "${WORKDIR}/temp/" bbwarn "found core dump in rootfs, check it in ${WORKDIR}/temp/${f##*/}" done