kill XFS_INOBT_IS_FREE_DISK
PATCH 1/1 kill XFS_INOBT_IS_FREE_DISK
This macro is unused an all other acros in this family operate on native types, so we most likely won't grow a user either.
fs/xfs/xfs_ialloc_btree.h +0 -2
@@ -81,8 +81,6 @@ typedef struct xfs_btree_sblock xfs_inob#define XFS_INOBT_MASK (i) ((xfs_inofree_t)1 << (i))#define XFS_INOBT_IS_FREE (rp,i) \ (((rp)->ir_free & XFS_INOBT_MASK(i)) != 0) -#define XFS_INOBT_IS_FREE_DISK (rp,i) \ - ((be64_to_cpu((rp)->ir_free) & XFS_INOBT_MASK(i)) != 0)#define XFS_INOBT_SET_FREE (rp,i) ((rp)->ir_free |= XFS_INOBT_MASK(i))#define XFS_INOBT_CLR_FREE (rp,i) ((rp)->ir_free &= ~XFS_INOBT_MASK(i))
Quoted 2 lines · Christoph Hellwig
This macro is unused an all other acros in this family operate on native types, so we most likely won't grow a user either.
Looks good to me... FWIW, here are the currently-unused macros in xfs cvs, if I'm massaging it right:
_ACL_ACCESS_EXISTS_ACL_GET_ACCESSATTR_DONTFOLLOWATTR_TRUSTBMBT_USE_64CE_CONTcurrent_clear_flags_nestedFSYNC_INVALFSYNC_NOWAITHAVE_DM_EVENTTYPE_THAVE_DM_RIGHT_THAVE_SPLICE
MODEMASK
NDPPnested_spinlocknested_spinunlockSGI_ACL_DEFAULT_SIZESV_KEYEDSV_LIFOSV_PRIOsv_timedwaitsv_timedwait_sigsv_wait_sig
TRACE0
TRACE1
TRACE3VN_ISBLKVN_ISCHRVN_ISLNK
VREAD
VSUID
VSVTX
VWRITEXB_GET_OWNERXFS_AT_ALLXFS_AT_SIZE_NOPERMXFS_AT_TIMESxfs_buf_ctobXFS_BUF_ISORDEREDXFS_BUF_ISWRITEXFS_BUF_OFFSETXFS_BUF_SET_OFFSETXFS_BUF_SET_SIZEXFS_BUF_UNBUSYXFS_BUF_VSEMAXFS_DA_MAXHASHXFS_DFORK_ASIZE_HOSTXFS_DFORK_DSIZE_HOSTXFS_ERRLEVEL_OFFXFS_ERRTAG_DIOWRITE_IOERRXFS_ERRTAG_MAXXFS_ERRTAG_NOERRORXFS_ERRTAG_STRATCMPL_IOERRXFS_ERRTAG_STRATREAD_IOERRXFS_EXTENT_TOKEN_WRXFS_IO_DCXVNXFS_IOMAP_WRITE_UNWRITTENXFS_IS_PQUOTA_RUNNINGXFS_LID_PINNEDXFS_MAX_LOG_BLOCKSXFS_MAX_LOG_BYTESxfs_probe_dmapixfs_probe_ioopsxfs_probe_quotaXFS_QMOPT_DOLOGXFS_RANDOM_DIOWRITE_IOERRXFS_RANDOM_STRATCMPL_IOERRXFS_RANDOM_STRATREAD_IOERRXFS_SBF_NOFLAGSXFS_SB_VERSION2_REALFBITSXFS_SB_VERSION2_RESERVED1BITXFS_SB_VERSION2_RESERVED4BITXFS_SB_VERSION_ADDDALIGNXFS_SB_VERSION_ADDEXTFLGBITXFS_SB_VERSION_ADDSHAREDXFS_SB_VERSION_SUBEXTFLGBITXFS_SB_VERSION_SUBSHAREDXFS_SIZE_TOKEN_WRXFS_SX_VERSIONxlog_exitXLOG_FMT_IRIX_BEXLOG_MIN_RECORD_BSHIFTXLOG_RHASH_BITSXLOG_SETXLOG_STATE_NOTUSED
-Eric
Quoted 11 lines · Eric Sandeen
Quoted 2 lines · Christoph Hellwig
This macro is unused an all other acros in this family operate on native types, so we most likely won't grow a user either.
Looks good to me... FWIW, here are the currently-unused macros in xfs cvs, if I'm massaging it right:
_ACL_ACCESS_EXISTS_ACL_GET_ACCESS
SGI_ACL_DEFAULT_SIZE
will go away with conversion to generic posix acl code.
Quoted 2 lines
ATTR_DONTFOLLOWATTR_TRUST
might go away with conversion to generic xattr code.
Quoted 1 line
BMBT_USE_64
I have a patch for this one.
Quoted 4 lines
CE_CONTcurrent_clear_flags_nestedFSYNC_INVALFSYNC_NOWAIT
We should probably kill these.
Quoted 2 lines
HAVE_DM_EVENTTYPE_THAVE_DM_RIGHT_T
no idea for these.
Quoted 1 line
HAVE_SPLICE
Backward compat?
Quoted 2 lines
MODEMASK
NDPP
kill?
Quoted 2 lines
nested_spinlocknested_spinunlock
don't your patches kill these?
Quoted 6 lines
SV_KEYEDSV_LIFOSV_PRIOsv_timedwaitsv_timedwait_sigsv_wait_sig
should probably go.
Quoted 3 lines
TRACE0
TRACE1
TRACE3
kill?
Quoted 7 lines
VN_ISBLKVN_ISCHRVN_ISLNK
VREAD
VSUID
VSVTX
VWRITE
I have a patch to kill all this gunk.
Quoted 3 lines
XFS_AT_ALLXFS_AT_SIZE_NOPERMXFS_AT_TIMES
I have some patches patch to kill XFS_AT_*
Quoted 2 lines
XFS_DFORK_ASIZE_HOSTXFS_DFORK_DSIZE_HOST
I have some patches to rework XFS_{C,D,I}FORK*
Quoted 9 lines
XB_GET_OWNERxfs_buf_ctobXFS_BUF_ISORDEREDXFS_BUF_ISWRITEXFS_BUF_OFFSETXFS_BUF_SET_OFFSETXFS_BUF_SET_SIZEXFS_BUF_UNBUSYXFS_BUF_VSEMA
not sure what to do with all the buf stuff.
Quoted 2 lines
XFS_IO_DCXVNXFS_IOMAP_WRITE_UNWRITTEN
should be gone with the patches I sent to the list.
Quoted 3 lines
xfs_probe_dmapixfs_probe_ioopsxfs_probe_quota
these are gone in a patch that I sent to the list already.
Quoted 9 lines
XFS_SBF_NOFLAGSXFS_SB_VERSION2_REALFBITSXFS_SB_VERSION2_RESERVED1BITXFS_SB_VERSION2_RESERVED4BITXFS_SB_VERSION_ADDDALIGNXFS_SB_VERSION_ADDEXTFLGBITXFS_SB_VERSION_ADDSHAREDXFS_SB_VERSION_SUBEXTFLGBITXFS_SB_VERSION_SUBSHARED
I suspect there are kind expected to be unused.
Quoted 5 lines · Christoph Hellwig
Quoted 2 lines
HAVE_SPLICEBackward compat?
ah, yeah, it's for 2.4:
0 xfs/linux-2.6/xfs_linux.h <global> 104 #define HAVE_SPLICE
1 fs/xfs/xfs_vnodeops.c <global> 4718 #ifdef HAVE_SPLICE
can probably go soon eh. HAVE_PERCPU_SB can too (not sure why it's not in my list) There may be others that are just for 2.4.
Quoted 5 lines
Quoted 3 lines
nested_spinlocknested_spinunlockdon't your patches kill these?
yeah, I just checked pristine cvs.
-Eric
Quoted 13 lines · Eric Sandeen
Quoted 11 lines
Christoph Hellwig wrote:
...XFS_SBF_NOFLAGSXFS_SB_VERSION2_REALFBITSXFS_SB_VERSION2_RESERVED1BITXFS_SB_VERSION2_RESERVED4BITXFS_SB_VERSION_ADDDALIGNXFS_SB_VERSION_ADDEXTFLGBITXFS_SB_VERSION_ADDSHAREDXFS_SB_VERSION_SUBEXTFLGBITXFS_SB_VERSION_SUBSHARED
Quoted 2 lines
I suspect there are kind expected to be unused.
They're used in userspace IIRC. (Well, some - probably not the RESERVED ones)
cheers.
--
Nathan
Quoted 2 lines · Christoph Hellwig
This macro is unused an all other acros in this family operate on native types, so we most likely won't grow a user either.
Could anyone put this trivial remove one line patch in?
Quoted 9 lines
Signed-off-by Christoph Hellwig <[email protected]>Index:
linux-2.6.23-rc6/fs/xfs/xfs_ialloc_btree.h
===================================================================
fs/xfs/xfs_ialloc_btree.h typedef struct xfs_btree_sblock xfs_inob · line 86
Quoted 8 lines
fs/xfs/xfs_ialloc_btree.h +0
-2@@ -81,8 +81,6 @@ typedef struct xfs_btree_sblock xfs_inob#define XFS_INOBT_MASK (i) ((xfs_inofree_t)1 << (i))#define XFS_INOBT_IS_FREE (rp,i) \ (((rp)->ir_free & XFS_INOBT_MASK(i)) != 0) -#define XFS_INOBT_IS_FREE_DISK (rp,i) \ - ((be64_to_cpu((rp)->ir_free) & XFS_INOBT_MASK(i)) != 0)#define XFS_INOBT_SET_FREE (rp,i) ((rp)->ir_free |= XFS_INOBT_MASK(i))#define XFS_INOBT_CLR_FREE (rp,i) ((rp)->ir_free &= ~XFS_INOBT_MASK(i))
---end quoted text---