/freebsd/sys/dev/beri/virtio/ |
H A D | virtio_block.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/share/man/man9/ |
H A D | vgone.9 | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/fs/autofs/ |
H A D | autofs_vnops.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/fs/ext2fs/ |
H A D | ext2_lookup.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/ufs/ffs/ |
H A D | ffs_rawread.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/fs/fuse/ |
H A D | fuse_io.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/kern/ |
H A D | kern_lockf.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/fs/smbfs/ |
H A D | smbfs_io.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
H A D | smbfs_vnops.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/fs/nullfs/ |
H A D | null_subr.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
H A D | null_vfsops.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
H A D | null_vnops.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/cam/ctl/ |
H A D | ctl_backend_block.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/dev/xen/blkback/ |
H A D | blkback.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/ufs/ufs/ |
H A D | ufs_inode.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/fs/pseudofs/ |
H A D | pseudofs_vnops.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/fs/nfsserver/ |
H A D | nfs_nfsdserv.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
H A D | nfs_nfsdstate.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/fs/nfsclient/ |
H A D | nfs_clport.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
H A D | nfs_clvnops.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/fs/cd9660/ |
H A D | cd9660_vnops.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/fs/tmpfs/ |
H A D | tmpfs_subr.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
H A D | tmpfs_vnops.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/fs/fdescfs/ |
H A D | fdesc_vnops.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|
/freebsd/sys/fs/unionfs/ |
H A D | union_subr.c | diff abd80ddb9474948fb291becc395d72b40927a32b Sun Dec 08 22:30:04 CET 2019 Mateusz Guzik <mjg@FreeBSD.org> vfs: introduce v_irflag and make v_type smaller
The current vnode layout is not smp-friendly by having frequently read data avoidably sharing cachelines with very frequently modified fields. In particular v_iflag inspected for VI_DOOMED can be found in the same line with v_usecount. Instead make it available in the same cacheline as the v_op, v_data and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1. Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff Differential Revision: https://reviews.freebsd.org/D22715
|