summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2008-09-06Kill the stats data structures and noop some other parts. Nothing inOwain Ainsworth
userland asks for these stats, and we stopped recording anything interesting a while back.
2008-09-05The code for cleaning up errored buffers and for cleaning up at the endOwain Ainsworth
is the same. Factor them into one function. Saves another 450 bytes on amd64.
2008-09-05The helper functions in this file are only used in one place andOwain Ainsworth
relatively small, so inline them. This shrinks the kernel by about 500 bytes and saves a tree lookup.
2008-09-05syncOkan Demirmen
2008-09-05- clarify Radeon HD 2600 Pro entryOkan Demirmen
- add Mobility Radeon HD 2600 found in a hp 8510p ok brad
2008-09-05fix vblank interrupt mask. Unbreaks sync-to-vblank and anything thatOwain Ainsworth
needs the actual interrupts. Oops!
2008-09-05Back out previous. Art realised a problem with it.Owain Ainsworth
2008-09-05Don't overwrite the old ipl in msleep if PNORELOCK was set.Artur Grabowski
2008-09-05When munging the WANTIPL of the mutex to prevent undoing the sched_lock,Owain Ainsworth
use the constant for IPL_SCHED, and not splsched(), which doesn't do what we want. ok art@. Tested by Paul de Weerd.
2008-09-04mostly cosmetic.Damien Bergamini
also, do not set the privacy bit in the capinfo field of (re)assoc requests, even for RSNA.
2008-09-03Replace -s (sector size) option with more general -t (disktype)Kenneth R Westerback
option which makes the vnd device emulate the geometry of the specified disktab(5) entry. No change in behaviour or geometry unless -t is used. API for vnd configuration ioctl (VNDIOCSET) changes, so mount_vnd must be in sync with kernel. Tested & ok jsing@ 'Lovely' deraadt@
2008-09-03third parameter of ieee80211_get_assoc_req() is a management frameDamien Bergamini
subtype, not a boolean indicating assoc or reassoc. rename the parameter and use if (type == IEEE80211_FC0_SUBTYPE_REASSOC_REQ) instead of if (reassoc). it worked only because IEEE80211_FC0_SUBTYPE_ASSOC_REQ is equal to 0.
2008-09-03(Re)Association requests should contain a QoS Capability element,Damien Bergamini
not an EDCA Parameter Set element (spotted by IEEE80211_STA_ONLY).
2008-09-03redefine ic_send_mgmt() as a no-op instead of calling IF_PURGE inDamien Bergamini
{ipw,iwi}_start which is wrong (node reference is not released). from pgt(4).
2008-09-03In dc_setcfg() suppress printing a warning when forcing the receiverBrad Smith
and transmitter to idle state times out for chips where the status bits in question never change (observed in detail with DM9102A) and therefore the warning would highly likely be a false positive. From FreeBSD
2008-09-03regenStuart Henderson
2008-09-03housekeeping:Stuart Henderson
- remove extraneous spaces - better column alignment - sort vendors by vendor-id - sort products by vendor-id,device-id - group products by vendor-id rather than vendor-name - fix vendor-id for Motorola SB4100 (checked with vendor driver) ok brad@
2008-09-03before linking state keys compare them to verify they actually are theHenning Brauer
exact reverse of each other. if there is a mismatch don't erstablish the link and print enough so that we should be able to find the culprit.
2008-09-03do not set the pkthdr mbuf state key pointer to the state key saved in theHenning Brauer
pcb. the state key ptr in the pcb is the one that had to be used by pf outbound. but by convention the state key pointer in the pkthdr is the one used INbound, so pf follows its reverse pointer to find the sk to use, and since a reverse doesn't exist for locally terminated connections the reverse pointer is null and thus the whole game a noop. note that this only affects packets FROM local udp/tcp sockets, for the other direction everything works as expected.
2008-09-03Use & not && when masking bits.Jonathan Gray
2008-09-03Remove dead code: ip6_copypktopts() is not used anywhere.Marco Pfatschbacher
OK deraadt@, henning@
2008-09-03Prevent a possible overflow when the sum of all demotion countersMarco Pfatschbacher
gets bigger than 255. OK henning@
2008-09-02If we need a physical hardware status page, initialise it at device attachOwain Ainsworth
instead of putting up and tearing down on open and close, some chips got unstable with it being done repeatedly. From drm git. Tested by several.
2008-09-02Track progress inside of batchbuffers so we know the hardware isn'tOwain Ainsworth
wedged. This avoids early temination of long-running commands. From Keith Packard, via drm git. Tested by several on various chipsets.
2008-09-02remove dead stores and newly created unused variables.Charles Longeau
Found by LLVM/Clang Static Analyzer. ok henning@ mpf@
2008-09-02missing break;Jonathan Gray
ok oga@
2008-09-02detypedef some more. No functional change.Owain Ainsworth
2008-09-01use C99 style struct initializer instead of GCC specific one (from NetBSD).Damien Bergamini
some whitespace cleanup too.
2008-09-01add ic_tid_noack bitmap to indicate a per-TID ACK policy (1=no ack,Damien Bergamini
0=normal ack). all bits are currently set to 0. use this bitmap to set the ACK policy of the QoS control field of outgoing QoS frames.
2008-09-01extend ieee80211_classify() to extract DSCP field from IPv6 packets too.Damien Bergamini
2008-09-01rework previously unused ieee80211_pwrsave() function, call it fromDamien Bergamini
ieee80211_mgmt_output() and ieee80211_encap(). use new IEEE80211_C_APPMGT capability flag where appropriate. rename ic_tim_mcast to ic_tim_mcast_pending.
2008-09-01avoid loosing return value from copyin, and properly return it.Charles Longeau
initial patch from me, reworked by oga@. found by LLVM/Clang Static Analyzer. ok oga@
2008-09-01fix null dereference.Charles Longeau
initial patch from me, reworked by oga@. found by LLVM/Clang Static Analyzer. ok oga@
2008-09-01convert a tsleep back to msleep that got lost in a previous upstream merge.Owain Ainsworth
Won't sleep with a spinlock here anymore.
2008-09-01Add Interrupt mitigation for the i9XX user_irq, should save a bit of power;Owain Ainsworth
from drm git.
2008-09-01Avoid #pragma pack(1) and unify everything towards using __packed.Theo de Raadt
This requires that structures defined within __packed structures must independently request that they themselves become __packed, too. worked on with toby CVS: ----------------------------------------------------------------------
2008-09-01Bring comment into line with reality. We use sgi label info but don'tKenneth R Westerback
create just partitions 'i' -> 'l'. jsing@ had noted same discord.
2008-09-01Finish the rest... ATL2 -> LIIBrad Smith
ok jsing@
2008-09-01Allow sgi to spoof the native label. But don't spoof when justKenneth R Westerback
looking for the place to write the disklabel, i.e. partoffp != NULL. tested & ok jsing@
2008-09-01Switch ath(4) to softcrypto to enable support for WPA/WPA2.Reyk Floeter
Tested by ckuethe@ and many users, thanks!
2008-08-31Fix redundant device name and/or target printing after callingMiod Vallat
sc_print_addr(); ok krw@
2008-08-31remove the #if 0'ed NFS_ATTRTIMEO() macro since it wasThordur I. Bjornsson
replaced by a function long ago.. ok blambert@
2008-08-31remove http links to dead domain.Jonathan Gray
2008-08-31remove http links to dead/changed domains.Jonathan Gray
2008-08-30replace TRUNC_PAGE by trunc_page and ROUND_PAGE by round_pageMartin Reindl
plus cast to vaddr_t where necassary from Miod's todo list ok miod@
2008-08-30Code refactorization and state structure changes in preparation of asynchronousMiod Vallat
abort commands, needed to eventually support command timeouts for non-polled commands... no functional change.
2008-08-30no need to include uvm/uvm.h twiceMartin Reindl
2008-08-30Move the link state check back up to the top of the various statusBrad Smith
functions right after reading the BMSR as it was before the addition of the fiber support and simplify brgphy_fiber_status() a bit. ok dlg@
2008-08-30garbage collect the no longer used GEM_IS_APPLE macro.Brad Smith
2008-08-30Fix an issue initially reported by drahn@ with gem(4) where theseBrad Smith
drivers can report 2-3x times the number of actual packets being transmitted by only incrementing the counter for descriptors with buffers being freed. ok kettenis@
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-23 21:39:31 +0000