d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
106,847 Commits 4 Branches 49 Tags
Commit Graph

49079 Commits

Author SHA1 Message Date
rwatson
8c9dc6fcfb Since if_tx doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:52:33 +00:00
rwatson
e85007f8b5 Since if_sr doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:49:48 +00:00
rwatson
c5bde45b43 Since if_snc doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:47:01 +00:00
rwatson
85f7f92072 Since if_sbni doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:41:00 +00:00
rwatson
58dec551ba Since if_ray doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:39:07 +00:00
rwatson
d9942f2ea2 Since if_plip doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:32:17 +00:00
rwatson
5f924be0bb Since if_nge doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.

Note: this driver does declare and occasionally reference mutexes,
but I believe not nearly enough to provide safety.
 2004-08-13 23:22:53 +00:00
rwatson
75ac991de3 Since if_lnc doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:20:50 +00:00
rwatson
7e9810d11e Since if_lge doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:18:01 +00:00
rwatson
9878d4a745 Since if_ic doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:16:44 +00:00
rwatson
1087b18ddf Since if_ie doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:15:44 +00:00
rwatson
04cbaba6eb Since if_hme doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:14:50 +00:00
rwatson
d12331cbbb Since if_gem doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:11:24 +00:00
rwatson
b702c717f3 Since if_fwip doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.

Approved by:	dfr
 2004-08-13 23:09:41 +00:00
rwatson
2ab0b531c9 Since if_fe doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:08:08 +00:00
rwatson
037ce3f3f4 Since if_ex doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.

Note: there are locking macros in if_exreg.h, but they appear to be
unused.
 2004-08-13 23:06:55 +00:00
rwatson
6370834221 Since if_ed doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:04:23 +00:00
rwatson
048a0f3416 Since if_cs doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:03:11 +00:00
rwatson
0564d2c810 Since if_cp doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 23:02:17 +00:00
rwatson
af8028e5e2 Since if_cm doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 22:57:44 +00:00
rwatson
5b97e47f5b Since if_awi doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 22:55:25 +00:00
rwatson
c02722ec8f Since if_arl doesn't contain locking or run with INTR_MPSAFE, mark 
the interface as IFF_NEEDSGIANT so if_start is run holding Giant.
 2004-08-13 22:54:19 +00:00
rwatson
f9261894f4 As the if_ar driver doesn't contain locking or run its interrupt 
MPSAFE, mark it as IFF_NEEDSGIANT so that its if_start routine is
run holding Giant.
 2004-08-13 22:52:11 +00:00
rwatson
f57f7a3212 Since the if_de driver doesn't contain locking, mark it as 
IFF_NEEDSGIANT so that ifp->if_start won't be called without Giant
when running debug.mpsafenet=1.
 2004-08-13 22:48:05 +00:00
ambrisko
30b952654a Fix the memory scaling bug when basemem was converted to Kbytes from 
bytes for AMD64.  Otherwise the AP will be started at 640K which
won't work.  Bug found on a Xeon 64bit system.
 2004-08-13 22:30:55 +00:00
rwatson
927adfff57 Use IFQ_SET_MAXLEN() to set the maximum queue depth of the routing 
socket netisr queue.

Pointed out by:	winter
 2004-08-13 22:23:21 +00:00
gibbs
c2e804d77e Extend critical section protection around portions of selection processing 
that cannot tolerate changes to the waiting for selection queue by the
host or the host canceling an active selection.
 2004-08-13 21:41:23 +00:00
gibbs
2e970a2641 Fix an off by one in the critical section clearing code. The 
code was adjusting twice for the instruction pointer indicating
the *next* instruction to execute.  The aic79xx driver had a similar
bug, but was fixed some time ago.
 2004-08-13 21:39:14 +00:00
rwatson
6994ab16ad Commit a work-around for a more general bug involving process state: 
check whether p_ucred is NULL or not in pfs_getattr() before
dereferencing the credential, and return ENOENT if there wasn't one.

This is a symptom of a larger problem, wherein pfind() can return
references to incompletely initialized processes, and we instead ought
to not return them, or check the process state before acting on the
process.

Reported by:	kris
Discussed with:	tjr, others
 2004-08-13 20:27:56 +00:00
njl
03ab910ba0 Fix building for the ACPI_DEBUG case. 2004-08-13 19:27:21 +00:00
alc
f19ef6db53 Add pmap locking to pmap_remove_all(). 2004-08-13 18:54:21 +00:00
mjacob
5f72b8eba7 When printing out an unknown sense code we should print it in hex, not 
decimal.

Reviewed by: gibbs, nate, kdm
 2004-08-13 18:45:04 +00:00
njl
7d472ce18d Record the new status after checking if it has changed, not before. This 
fixes lost AC line transition events.

Bug report:	Kevin Oberman
 2004-08-13 17:47:40 +00:00
jmg
bea28d4a04 clean up whitespace... 2004-08-13 17:43:53 +00:00
ru
98087389a3 Removed COPTS support from kmod.mk and kern.pre.mk. 
COPTS support in bsd.prog.mk is preserved but discouraged.
 2004-08-13 14:30:26 +00:00
rik
7cdae66efa Fix resource check while autodetection. 2004-08-13 12:35:52 +00:00
rik
1500156262 White space cleanup. 2004-08-13 12:27:24 +00:00
rik
83e85d7393 White space cleanup. 2004-08-13 12:22:01 +00:00
rik
d5629fc00d White space cleanup. 2004-08-13 12:14:39 +00:00
fjoe
50b0c5e0ac Add geom_uzip -- geom class that implements read-only compressed disks. 
Currently supports cloop V2.0 disk compression format.
May support more formats in future.
 2004-08-13 09:40:58 +00:00
sos
6f119c28ba Allow the use of a supplied function to set the PRD table. This is 
needed for new chips that supports 64bit addressing.
 2004-08-13 08:14:27 +00:00
alc
482b6818af Replace the linear search in vm_map_findspace() with an O(log n) 
algorithm built into the map entry splay tree.  This replaces the
first_free hint in struct vm_map with two fields in vm_map_entry:
adj_free, the amount of free space following a map entry, and
max_free, the maximum amount of free space in the entry's subtree.
These fields make it possible to find a first-fit free region of a
given size in one pass down the tree, so O(log n) amortized using
splay trees.

This significantly reduces the overhead in vm_map_findspace() for
applications that mmap() many hundreds or thousands of regions, and
has a negligible slowdown (0.1%) on buildworld.  See, for example, the
discussion of a micro-benchmark titled "Some mmap observations
compared to Linux 2.6/OpenBSD" on -hackers in late October 2003.

OpenBSD adopted this approach in March 2002, and NetBSD added it in
November 2003, both with Red-Black trees.

Submitted by: Mark W. Krentel
 2004-08-13 08:06:34 +00:00
jmg
d2ff11056b looks like rwatson forgot tabs... :) 2004-08-13 07:38:58 +00:00
imp
462c08381f Move PNP IDs back into oldcard files 2004-08-13 06:57:31 +00:00
njl
5e14731c01 MPSAFE locking 
* Serialize access to the sysctl routines and the notify handler
* Assert that the sx lock is held in any functions they call.
* Note that recursively calling to re-enable the hotkeys is sub-optimal.
 2004-08-13 06:22:35 +00:00
njl
86267b0197 MPSAFE locking 
* Serialize access to the sysctl routines and the notify handler
* Assert that the sx lock is held in any functions they call.
 2004-08-13 06:22:31 +00:00
njl
8fcef8f766 MPSAFE locking 
* Serialize access to the sysctl routines and the notify handler.
 2004-08-13 06:22:29 +00:00
njl
e7ca6bbdad MPSAFE locking 
* Simplify taskqueue locking.
* Don't acquire Giant around calls to the taskqueue function.
* Remove 4.x compatibility routines.
 2004-08-13 06:22:26 +00:00
njl
d49f14c7b2 MPSAFE locking 
* Remove the interrupt wrapper that locked Giant and call the handler
  directly.  Mark the handler as MPSAFE.
* Don't attempt to detect if a handler is installed.  Leave that to the
  bus_alloc_resource() function.
 2004-08-13 06:22:24 +00:00
njl
77ba746d3a MPSAFE locking 
* Serialize operations in acpi_video_bind_outputs(), acpi_video_detach(),
  acpi_video_notify_handler(), acpi_video_power_profile(), and the sysctls.
  The main goal is to protect the shared device list and prevent conflicting
  settings.
* Add assertions that the sx lock is held in the leaf functions.
 2004-08-13 06:22:20 +00:00