summaryrefslogtreecommitdiffstats
path: root/sys/netatm
Commit message (Collapse)AuthorAgeFilesLines
* Update netisr handling; Each SWI now registers its queue, and all queuejlemon2003-03-044-49/+30
| | | | | | | | | | drain routines are done by swi_net, which allows for better queue control at some future point. Packets may also be directly dispatched to a netisr instead of queued, this may be of interest at some installations, but currently defaults to off. Reviewed by: hsu, silby, jayanth, sam Sponsored by: DARPA, NAI Labs
* There is no reason to be cute with ntohl(). Just call it directly ratherobrien2003-02-235-22/+8
| | | | | | than use a macro that tries to do conversions in place. Compile tested on: sparc64
* Back out M_* changes, per decision of the TRB.imp2003-02-1922-71/+71
| | | | Approved by: trb
* Band-XXX-aid an easy to provoke panic.phk2003-01-281-0/+7
| | | | MFC: 2 weeks
* Remove M_TRYWAIT/M_WAITOK/M_WAIT. Callers should use 0.alfred2003-01-2123-83/+83
| | | | Merge M_NOWAIT/M_DONTWAIT into a single flag M_NOWAIT.
* Correct typos, mostly s/ a / an / where appropriate. Some whitespace cleanup,schweikh2003-01-013-3/+3
| | | | especially in troff files.
* Fix typos, mostly s/ an / a / where appropriate and a few s/an/and/schweikh2002-12-303-5/+5
| | | | Add FreeBSD Id tag where missing.
* Correct mbuf packet header propagation. Previously, packet headerssam2002-12-301-2/+1
| | | | | | | | | | | | | | | | | | | | | | were sometimes propagated using M_COPY_PKTHDR which actually did something between a "move" and a "copy" operation. This is replaced by M_MOVE_PKTHDR (which copies the pkthdr contents and "removes" it from the source mbuf) and m_dup_pkthdr which copies the packet header contents including any m_tag chain. This corrects numerous problems whereby mbuf tags could be lost during packet manipulations. These changes also introduce arguments to m_tag_copy and m_tag_copy_chain to specify if the tag copy work should potentially block. This introduces an incompatibility with openbsd which we may want to revisit. Note that move/dup of packet headers does not handle target mbufs that have a cluster bound to them. We may want to support this; for now we watch for it with an assert. Finally, M_COPYFLAGS was updated to include M_FIRSTFRAG|M_LASTFRAG. Supported by: Vernier Networks Reviewed by: Robert Watson <rwatson@FreeBSD.org>
* Indirectly pull in declaration for M_IFADDR.hsu2002-12-281-0/+1
|
* SMP locking for radix nodes.hsu2002-12-241-0/+2
|
* SMP locking for ifnet list.hsu2002-12-222-0/+5
|
* - Change the ATM stack functions to use intptr_t instead of int for opaquejhb2002-11-0831-202/+205
| | | | | | | arguments. - Fix a few other places that assumed that sizeof(int) == sizeof(void *). Reviewed by: mdodd
* Add a field to struct cmn_unit to hold a pointer to the driver's softc.mdodd2002-10-011-0/+2
|
* Include <sys/systm.h> instead of depending on namespace pollution inbde2002-09-292-16/+2
| | | | | <net/if_var.h>. But depend on the standard pollution in <sys/param.h>. Removed unused includes.
* Include <sys/systm.h> instead of depending on namespace pollution inbde2002-09-291-2/+1
| | | | <net/if_var.h>. But depend on the standard pollution in <sys/param.h>.
* Make netatm/spans compile in the kernel without depending on userlandpeter2002-09-174-4/+381
| | | | include files to provide functions for kernel source (spans_kxdr.c)
* Replace various spelling with FALLTHROUGH which is lint()ablecharnier2002-08-251-1/+1
|
* Increase size of ifnet.if_flags from 16 bits (short) to 32 bits (int). To avoidsobomax2002-08-181-1/+1
| | | | | | | breaking application ABI use unused ifreq.ifru_flags[1] for upper 16 bits in SIOCSIFFLAGS and SIOCGIFFLAGS ioctl's. Reviewed by: -hackers, -net
* - Remove UM_* user land memory macros since they are no longer used.arr2002-06-241-17/+1
|
* - Remove DMA_* macros as only one, DMA_GET_ADDR, was used.arr2002-06-241-8/+0
| | | | | - Convert DMA_GET_ADDR() calls to vtophys() calls in order to finish removing the DMA_* macros.
* - Remove the definition for SCHED_ATMarr2002-06-241-1/+0
| | | | - Change SCHED_ATM statements into schednetisr() calls.
* - Turn a DIAGNOSTIC check into a KASSERT(9).arr2002-06-171-4/+1
|
* - Turn a DIAGNOSTIC check into a KASSERT(9) macro.arr2002-06-171-6/+2
|
* - Explicitly pass M_WAITOK in a few calls to uma_zalloc(9).arr2002-06-171-7/+7
|
* - Chainsaw the storage pool code. This was being used by a bunch of codearr2002-06-143-404/+1
| | | | | | within the HARP atm stack and the hea and hfa device drivers, but since all of these systems were changed to use UMA zones, there is no use for the api any longer.
* - Turn the hea and hfa HARP storage pools into UMA zones and insertarr2002-06-144-19/+22
| | | | | | | | | | | | | | | | the necesary uma_zcreate() and uma_zdestroy calls into module loading handler and the device attach handling. - Change the related HARP netatm code to use UMA zone functions when dealing with the zones that were formerly the ATM interface (hea, hfa) storage pools. - Have atm_physif_freenifs() now get passed an uma_zone_t so that we can properly free the allocated NIF's back to their zone. This should be the last commit to remove any code that makes use of the netatm storage pool api. I will be removing the api code within the near future. Reviewed by: mdodd
* - Finish the uni part of the storage pool cleanup. There should now onlyarr2002-06-1318-200/+195
| | | | | | | | | | be a few bits left to clean from the HARP code in terms of what is using the storage pools; once that's done, the memory management code can be removed entirely. This commit effectively changes the use of dynamic memory routines from atm_allocate, atm_free, atm_release_pool to uma_zcreate, uma_zalloc, uma_zfree, uma_zdestroy.
* Use a larger data type to prevent counters wrapping so quickly.mdodd2002-06-072-14/+14
| | | | Silence a warning.
* Back out my lats commit of locking down a socket, it conflicts with hsu's work.tanimura2002-05-312-33/+3
| | | | Requested by: hsu
* Pacify gcc by preinitializing a variable.peter2002-05-241-2/+2
|
* Fix a few warnings by adding a missing prototypepeter2002-05-241-0/+1
|
* - Remove a few storage pools and replace them with UMA zones. The spansarr2002-05-247-78/+90
| | | | | code is now storage pool free, so I believe this only leaves the uni base not cleaned.
* - Turn a couple of storage pools into UMA zones and their associated callsarr2002-05-242-10/+13
| | | | to the appropriate UMA api calls.
* - Fix uma_zcreate parameters -- don't pass M_* flags here, doh.arr2002-05-231-2/+2
|
* - Turn two more storage pools into UMA zones and make the related memoryarr2002-05-234-30/+31
| | | | allocations and frees use the UMA api.
* Lock down a socket, milestone 1.tanimura2002-05-202-3/+33
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | o Add a mutex (sb_mtx) to struct sockbuf. This protects the data in a socket buffer. The mutex in the receive buffer also protects the data in struct socket. o Determine the lock strategy for each members in struct socket. o Lock down the following members: - so_count - so_options - so_linger - so_state o Remove *_locked() socket APIs. Make the following socket APIs touching the members above now require a locked socket: - sodisconnect() - soisconnected() - soisconnecting() - soisdisconnected() - soisdisconnecting() - sofree() - soref() - sorele() - sorwakeup() - sotryfree() - sowakeup() - sowwakeup() Reviewed by: alfred
* - Remove the init routine for loop -- it broke something and I don't havearr2002-05-071-11/+4
| | | | the time to fix it properly at the moment.
* - Use an array of init functions to be used in a for loop in ourarr2002-05-071-4/+11
| | | | initialization routine.
* - Add atm_init_fnarr2002-05-071-0/+2
|
* - Add atm_sock_init()arr2002-05-073-9/+16
| | | | - Move the Atm_pcb storage pool (atm_pcb_pool) to be an UMA zone.
* - Change the atm_cm_init function to take no argument.arr2002-05-072-2/+2
|
* - Replaced the Atm_connection storage pool with an uma_zone ofarr2002-05-023-29/+31
| | | | | | | | | | Atm_connection items. - Replaced the Atm_connvc storage pool with an uma_zone of Atm_connvc items. - Created void atm_cm_init(void *) and added it to the netatm init code. I'm thinking that there will definetly be more ``subsystem'' init functions to be added so I'll probably change these calls to be a for loop through init routines (or something).
* - Take advantage of the M_ZERO flag that can now be passed to uma_zalloc.arr2002-04-303-15/+4
| | | | - Remove atm_uma_ctor() as the M_ZERO will zero out the memory for us.
* Revert the change of #includes in sys/filedesc.h and sys/socketvar.h.tanimura2002-04-301-2/+5
| | | | | | | | | | Requested by: bde Since locking sigio_lock is usually followed by calling pgsigio(), move the declaration of sigio_lock and the definitions of SIGIO_*() to sys/signalvar.h. While I am here, sort include files alphabetically, where possible.
* - Tweak wording of panic message to make more sense.arr2002-04-241-2/+2
|
* - Turn the atm_stackq_pool into a uma_zone (and change it's name toarr2002-04-241-9/+9
| | | | | | atm_stackq_zone). - Change the related atm_allocate() and atm_free() calls into uma_zalloc() and uma_zfree() calls.
* - Remove atm_attributes_pool and the relating atm_allocate() and atm_free()arr2002-04-243-11/+15
| | | | | | | | calls associated with the pool and the objects allocated out from the pool. - Insert atm_attributes_zone which is a uma_zone that is used just as the atm_attributes_pool was (including the max objects value). Also, used the appropriate zalloc and zfree's where necesary.
* - Revert previous change of atm storage pools -> uma_zones until a solutionarr2002-04-223-19/+17
| | | | to atm_free() is written.
* - Change the atm_attributes_pool and atm_stackq_pool to be uma_zone'sarr2002-04-223-17/+19
| | | | | (with the appropriate set_max) rather than using the HARP storage pool code.
* - Nuke some more not needed #ifdef cruft.arr2002-04-2114-319/+0
|
OpenPOWER on IntegriCloud