Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
203,132 Commits 606 Branches 0 Tags 3.1 GiB
5b854a1a07
Commit Graph

203132 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
dholland
5b854a1a07 ffs and ffslog are no longer xfail. 2011-07-18 06:47:08 +00:00
dholland
36c3688f7b Remove some unneeded rename-related static const data. 
(Why didn't gcc warn that this was unused?)
 2011-07-18 06:46:05 +00:00
dholland
8b45aa4ea4 Eliminate the old ufs_rename. The only reason the WAPBL one was 
different is that in order to avoid issues with the WAPBL journal lock
the wrong locking had to be changed to different wrong locking. This
is now moot.

I have not hand-validated that the current two copies of rename are
equivalent, or that the locking fixes merged with the old rename
produce code that is textually identical (modulo WAPBL calls that do
nothing when WAPBL is turned off) to the WAPBL rename... but I did
this check when preparing my previous round of rename patches last
year and all updates since have been applied to both.
 2011-07-18 06:45:47 +00:00
dholland
ffbed3d146 Move ufs_wapbl_rename to ufs_vnops.c next to the old ufs_rename. 2011-07-18 06:45:27 +00:00
jruoho
b4152f8512 Regen. 2011-07-18 06:31:59 +00:00
jruoho
d019b19341 Add MSI MegaSky DVB-T dongle. 2011-07-18 06:30:49 +00:00
jruoho
cc6f1c900e Use __arraycount(3). 2011-07-18 05:57:40 +00:00
jruoho
4c0e40e4b4 Use a list for clarity, update standards, etc. 2011-07-18 05:17:16 +00:00
dholland
6afd4e5087 typo 2011-07-18 04:39:28 +00:00
jruoho
650c5cb762 Verify that dup2(2) and dup3(2) fail with EBADF if the "new descriptor" is 
specified to be larger than RLIMIT_NOFILE.
 2011-07-18 04:29:37 +00:00
christos
9b1cdf6a14 better fix for gcc uninitialized. 2011-07-18 02:47:52 +00:00
dholland
091c1fe61a ok, it is clear that at least vput(vp) needs to go before fstrans_done(). 
I hoping vput(dvp) doesn't, because if it does that will vastly complicate
future vfs locking cleanup.
 2011-07-18 02:35:11 +00:00
manu
7541af315c ftruncate(2) cause a SETATTR with only va_size set, and some filesystems 
(e.g.: glusterfs) will do a custom handling in such a situation. This
breaks because libpuffs folds a metadata (va_atime and va_mtime) update
in each SETATTR. We try to identify SETATTR caused by ftruncate(2) and
remove va_atime and va_mtime in such situation.

This fixes a bug with glusterfs, where parts of a file downloaded by
FTP was filled with zeros because of a ftruncate(2) sent out of order
with write(2) requests. glusterfs behavior depends on the undocumented
FUSE rule that ftruncate(2) will only set va_size in SETATTR.
 2011-07-18 02:14:01 +00:00
dholland
c751831386 Add the long essay on rename locking from my earlier patch set as a 
big comment, and expand it some for clarity.
 2011-07-18 01:52:55 +00:00
dholland
5c6ce29c40 More rename tidying. 2011-07-18 01:14:27 +00:00
dholland
fe126fdad8 In ufs_wapbl_rename, remove #if 0 blocks and remove code for 
now-impossible cases.
 2011-07-18 01:14:04 +00:00
dholland
964d208c77 Fix error path for UFS_WAPBL_BEGIN failure to avoid leaking an vnode 
(in memory) and inode (on disk). Caution: untested; I have no idea how
to provoke such a failure.
 2011-07-18 01:13:33 +00:00
mrg
d4bff0e86b convert the remaining dev/sbus drivers to CFATTACH_DECL_NEW, cfdata_t 
and device_t, including the pci frontend and backend for en(4).
 2011-07-18 00:58:51 +00:00
christos
6e03a05a94 fix gcc uninitialized warnings 2011-07-18 00:43:35 +00:00
christos
6c10492aaa fix uninitialized gcc variables 2011-07-18 00:42:40 +00:00
mrg
142a02103b convert the remaining sparc drivers to CFATTACH_DECL_NEW/cfdata_t/device_t. 
(cgsix_obio.c was only partially converted with the rest of the cgsix code
when it was changed some time ago.)
 2011-07-18 00:31:13 +00:00
mrg
9c508f54bd convert to use device_t, cfdata_t and CFATTACH_DECL_NEW(). 
XXX: compile time tested only.
 2011-07-18 00:05:35 +00:00
christos
1360a27d17 fail early on socket domain+type combinations we don't support 2011-07-17 23:59:54 +00:00
dyoung
106ff60f41 Do not install bus.h. 2011-07-17 23:52:12 +00:00
dholland
99199b5b1a fix typo 2011-07-17 23:50:05 +00:00
dyoung
8a61060442 Switch to new-style <sys/bus.h>. 2011-07-17 23:48:35 +00:00
dholland
99f3bafa7f prune stray commas 2011-07-17 23:46:53 +00:00
dyoung
35546ae908 On ARM-ish ports having PCI support, use new-style <sys/bus.h>. Do not 
install <machine/bus.h>.  Remove various <machine/bus.h> sources.
 2011-07-17 23:46:49 +00:00
dyoung
90bc18e758 Increase SYMTAB_SPACE. 2011-07-17 23:40:43 +00:00
dyoung
fb90a131da Switch to new-style <sys/bus.h>. 2011-07-17 23:40:16 +00:00
dyoung
e6b05a5d9b Increase SYMTAB_SPACE. How do we automate this? 2011-07-17 23:39:46 +00:00
dyoung
f9a358cfdb Good-bye bus.h. Don't install <machine/bus.h>. 2011-07-17 23:38:32 +00:00
dyoung
e08fcd397b Update set lists for changes related to new-style <sys/bus.h>. 
While I'm here, repair sparc64 set list.
 2011-07-17 23:37:32 +00:00
mrg
210f684c14 actually use CFATTACH_DECL_NEW() and device_private() for mspcic. 2011-07-17 23:34:17 +00:00
mrg
794d81a9dd convert all the "empty softc" (just "struct device") drivers to using 
device_t/cfdata_t and CFATTACH_DECL_NEW().

the mspcic driver was also converted since it does not ever touch the
sc_dev it used to have in there.
 2011-07-17 23:32:37 +00:00
dyoung
bcbb44b510 Switch to new-style <sys/bus.h>. hp700 releases are unbuildable for 
reasons not related to my changes here, so I'm not 100% that a release
build will run to completion with these changes in place.
 2011-07-17 23:32:21 +00:00
dyoung
9c93e701a6 Switch sparc and sparc64 to new-style <sys/bus.h>. 2011-07-17 23:29:10 +00:00
dyoung
7d2843a545 Switch more PowerPC ports having PCI support to new-style <sys/bus.h>. 2011-07-17 23:27:02 +00:00
dyoung
d181d57cd1 Enable new-style <sys/bus.h> on several PowerPC ports supporting PCI 
buses.  Make non-inline implementations of bus_space(9) and bus_dma(9)
routines and move them to appropriate .c files.

This may leave amigappc in a bad state, sorry.  Fortunately, it will be
easy to repair by imitating the bus.h -> bus_{defs,funcs}.h split in
some other PowerPC port.
 2011-07-17 23:23:53 +00:00
mrg
ace9f0e4af convert all the sparc drivers that attach on my ss20 to use device_t, 
cfdata_t and CFATTACH_DECL_NEW().

fd/fdc could use better testing, but i'm pretty sure i got it right.
 2011-07-17 23:18:23 +00:00
dyoung
6e7a8e528f #include <sys/bus.h> not <machine/bus.h>. 2011-07-17 23:08:56 +00:00
dholland
84a8bd42ce minor amendment to previous 2011-07-17 22:14:47 +00:00
dholland
d1917ffca2 Provide correct locking for ufs_wapbl_rename. Note that this does not 
fix the non-wapbl rename; that will be coming soon. This patch also
leaves a lot of the older locking-related code around in #if 0 blocks,
and there's a lot of leftover redundant logic. All that will be going
away later.

Relates to at least these PRs:

  PR kern/24887
  PR kern/41417
  PR kern/42093
  PR kern/43626

and possibly others.
 2011-07-17 22:07:59 +00:00
dholland
9ae442b9dd At the end of ufs_rmdir, don't use a dangling vnode pointer to call 
fstrans_done. Ok hannken@
 2011-07-17 22:02:26 +00:00
dholland
3d3932fe1c Fix typo in ufs_rmdir that causes locking botches. This code should be 
unreachable because the FS-independent code contains the same test... but
I'm not sure if that applies if nfsd is involved.
 2011-07-17 22:00:38 +00:00
joerg
3eb244d801 Retire varargs.h support. Move machine/stdarg.h logic into MI 
sys/stdarg.h and expect compiler to provide proper builtins, defaulting
to the GCC interface. lint still has a special fallback.
Reduce abuse of _BSD_VA_LIST_ by defining __va_list by default and
derive va_list as required by standards.
 2011-07-17 20:54:30 +00:00
christos
ab0103dee5 fix type punned warnings in debugging code. 2011-07-17 20:47:39 +00:00
joerg
9777e8c193 Update to Clang/LLVM r135360. Most importantly, this adjusts the 
cc -print-search-dirs output to expand sysroot-prefixed library paths to
make live easier for libtool.
 2011-07-17 20:07:56 +00:00
joerg
5158e28f3b Disable Clang's integrated assembler for the AES-NI files for now. 
Somewhere in this mess of .byte streams, corruption happens. Disassembly
only shows slightly different filling of alignment sequences, further
analysis is needed.

XXX This should be rewritten to be proper assembler code
 2011-07-17 19:48:31 +00:00
jym
4b70f9c001 CR4_PAE is always set to 1 under amd64, so indicate that PAE mode is 
enabled. Can be useful for 32-bits test runs on amd64 hosts.
 2011-07-17 15:16:58 +00:00