summaryrefslogtreecommitdiffstats
path: root/share/man (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Add refcnt_take_if_gt()Matt Dunwoodie2021-04-131-0/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This function (or of similar nature) is required to safely use a refcnt and smr_entry together. Such functions exist on other platforms as kref_get_unless_zero (on Linux) and refcount_acquire_if_gt (on FreeBSD). The following diagram details the following situation with and without refcnt_take_if_gt in 3 cases, with the first showing the "invalid" use of refcnt_take. Situation: Thread #1 is removing the global referenc (o). Thread #2 wants to reference an object (r), using a thread pointer (t). Case: 1) refcnt_take after Thread #1 has released "o" 2) refcnt_take_if_gt before Thread #1 has released "o" 3) refcnt_take_if_gt after Thread #1 has released "o" Data: struct obj { struct smr_entry smr; struct refcnt refcnt; } *o, *r, *t1, *t2; Thread #1 | Thread #2 ---------------------------------+------------------------------------ | r = NULL; rw_enter_write(&lock); | smr_read_enter(); | t1 = SMR_PTR_GET_LOCKED(&o); | t2 = SMR_PTR_GET(&o); SMR_PTR_SET_LOCKED(&o, NULL); | | if (refcnt_rele(&t1->refcnt) | smr_call(&t1->smr, free, t1); | | if (t2 != NULL) { | refcnt_take(&t2->refcnt); | r = t2; | } rw_exit_write(&lock); | smr_read_exit(); ..... // called by smr_thread | free(t1); | ..... | // use after free | *r ---------------------------------+------------------------------------ | r = NULL; rw_enter_write(&lock); | smr_read_enter(); | t1 = SMR_PTR_GET_LOCKED(&o); | t2 = SMR_PTR_GET(&o); SMR_PTR_SET_LOCKED(&o, NULL); | | if (refcnt_rele(&t1->refcnt) | smr_call(&t1->smr, free, t1); | | if (t2 != NULL && | refcnt_take_if_gt(&t2->refcnt, 0)) | r = t2; rw_exit_write(&lock); | smr_read_exit(); ..... // called by smr_thread | // we don't have a valid reference free(t1); | assert(r == NULL); ---------------------------------+------------------------------------ | r = NULL; rw_enter_write(&lock); | smr_read_enter(); | t1 = SMR_PTR_GET_LOCKED(&o); | t2 = SMR_PTR_GET(&o); SMR_PTR_SET_LOCKED(&o, NULL); | | if (t2 != NULL && | refcnt_take_if_gt(&t2->refcnt, 0)) | r = t2; if (refcnt_rele(&t1->refcnt) | smr_call(&t1->smr, free, t1); | rw_exit_write(&lock); | smr_read_exit(); ..... | // we need to put our reference | if (refcnt_rele(&t2->refcnt)) | smr_call(&t2->smr, free, t2); ..... // called by smr_thread | free(t1); | ---------------------------------+------------------------------------ Currently it uses atomic_add_int_nv to atomically read the refcnt, but I'm open to suggestions for better ways. The atomic_cas_uint is used to ensure that refcnt hasn't been modified since reading `old`.
* tweak previous;jmc2021-04-021-3/+4
|
* Document ioctl(2)'s for vmm(4). OK kn@.dv2021-04-021-2/+43
|
* document trusted_snapshotespie2021-03-311-2/+8
|
* list Dell DW5821e as supported for umb(4)sthen2021-03-281-2/+4
|
* Push kernel lock within rtable_add(9) and rework it to return 0 in themvs2021-03-261-6/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | case when requested table is already exists. Except initialization time, route_output() and if_createrdomain() are the only paths where we call rtable_add(9). We check requested table existence by rtable_exists(9) and it's not the error condition if the table exists. Otherwise we are trying to create requested table by rtable_add(9). Those paths are kernel locked so concurrent thread can't create requested table just after rtable_exists(9) check. Also rtable_add(9) has internal rtable_exists(9) check and in this case the table existence assumed as EEXIST error. This error path is never reached. We are going to unlock PF_ROUTE sockets. This means route_output() will not be serialized with if_createrdomain() and concurrent thread could create requested table. Table existence check and creation should be serialized and it makes sense to do this within rtable_add(9). This time kernel lock is used for this so it pushed down to rtable_add(9). The internal rtable_exists(9) check was modified and table existence is not error now. Since the external rtable_exists(9) check is useless it was removed from if_createrdomain(). It still exists in route_output() path because the logic is more complicated here. ok mpi@
* Document SIOCGIFADDR and SIOCSIFADDR ioctls for tap devices.claudio2021-03-231-2/+6
| | | | OK kn@
* Document the @version suffix that can be added when running 'portgen go'abieber2021-03-231-2/+11
| | | | Patch from Josh Rickmar. Ty jrick!
* document NEVER_CLEANespie2021-03-221-2/+6
|
* s/struft/struct/; thanks James Hastingssthen2021-03-202-6/+6
|
* Sync with apm(4/macppc) to document which ioctls are not supportedkn2021-03-201-2/+13
| | | | Suspend/resume and other power events are NOT YET SUPPORTED.
* End sentence and add .Pp after (all) "NOT SUPPORTED" lineskn2021-03-201-7/+13
|
* Update go-module docs for recent changes, from Josh Rickmar, small tweaksthen2021-03-181-6/+4
| | | | from me
* Document SIOCDELLABEL, link among MPLS driverskn2021-03-183-9/+23
| | | | | | | | | | Use of the IOCTL section losely adopted from bridge(4), the list of ioctls however is still incomplete. mpw(4) and mpip(4) could use a reference to mpe(4) IOCTL or so, but this is good enough for starters. Feedback OK claudio
* List the Qu-c0-hr-b0-48 firmware image in the iwx(4) man page.stsp2021-03-171-2/+3
|
* Describe what happens when RFC 4638 is not supported.millert2021-03-161-7/+16
| | | | With help from sthen@. OK sthen@ jmc@
* add HISTORY; from maxim vuetsjmc2021-03-141-2/+7
|
* sync usb_device_info with usb.h, spotted by Enrik Berkhan, and millert@sthen2021-03-131-10/+8
| | | | noticed another change
* Link to sh(1) and use the same wording and markup for EXTRACT_CASES codekn2021-03-111-3/+6
| | | | "good idea" sthen
* Typofix previouskn2021-03-101-2/+2
|
* cleanup the 3 ways (2 for ipv4, 1 for ipv6) of doing dynamic addressderaadt2021-03-101-10/+21
| | | | | allocation. not the perfect text yet, but it is better. discussed with florian and jmc
* Remove documentation for all port modules no longer supported. ok sthen@tracey2021-03-091-153/+12
|
* document zstd support in EXTRACT_CASES and change from a big code samplesthen2021-03-091-30/+32
| | | | | that isn't directly usable to a shorter table and example of what one might need to add to EXTRACT_CASES when making a change; ok espie@
* Note that slaacd(8) is enabled by default. Also note that if youflorian2021-03-081-5/+2
| | | | | | disabled it in the past and then note that IPv6 autoconfiguration doesn't work for you, you will quickly note that we will point and laugh.
* Add RK809 to list of supported chips for this driver.kurt2021-03-081-3/+3
|
* correct paper titlejsg2021-03-081-3/+3
|
* Add some references, most of these were removed when we stopped buildingjsg2021-03-082-4/+30
| | | | | | and installing USD/SMM/PSD docs. jmc@ agrees with the direction, ok millert@ on an earlier diff
* new sentence, new line;jmc2021-03-041-9/+9
|
* Add a configurable button mapping for tap gestures on touchpads.bru2021-03-031-9/+20
| | | | | | Thanks to RJ Johnson for this work! ok mpi@
* remove unneccessary Pp;jmc2021-03-011-3/+2
|
* document log entry W, reminded by naddy@espie2021-03-011-2/+8
|
* fix Nd and remove trailing whitespace;jmc2021-03-011-4/+4
|
* document pkg_check-versionespie2021-03-012-3/+58
|
* document the move of FIX_CRLF_FILES to the end of distpatchespie2021-02-281-6/+6
|
* removal of "must" missed in previous: spotted by george brown, who alsojmc2021-02-271-2/+2
| | | | submitted the diff for the previous commit and who i forgot to credit...
* remove some info related to compat;jmc2021-02-271-5/+4
| | | | | | tweaked by deraadt ok kettenis deraadt
* fix NAME;jmc2021-02-271-2/+2
|
* apldart(4)kettenis2021-02-271-0/+42
|
* aplpcie(4)kettenis2021-02-271-0/+44
|
* aipldart(4)kettenis2021-02-271-2/+2
|
* aplpcie(4)kettenis2021-02-271-2/+3
|
* Add logger(1) support for daemons that are logging to stdout/stderr (mostly fromajacoutot2021-02-272-4/+16
| | | | | | | | | | | | | the go ecosystem). Properly handle failing daemon startup now that we have pipefail. To take advantage of this new feature, just add foo_logger=facility to the daemon rc.d(8) script or in rc.conf.local(8) or use rcctl: rcctl set foo logger daemon.info tweak for checking flags in rcctl(8) from martijn@ "this looks pretty good" deraadt@ ok sthen@
* Fix Typo, ok jcadanj2021-02-261-2/+2
|
* Add my copyright since I added most of the contentdanj2021-02-261-2/+3
|
* mention that FIX_CRLF_FILES is rooted in WRKDIST, pointed out by espie@.sthen2021-02-251-1/+3
| | | | mention that shell wildcards may be used.
* wrong annotationespie2021-02-251-2/+2
|
* make the timing of FIX_CRLF_FILES explicitespie2021-02-251-2/+4
|
* document FIX_CRLF_FILES.sthen2021-02-251-2/+7
|
* split off the large module documentations, so that port-modules is less ofespie2021-02-257-573/+791
| | | | | | a monster. okay sthen@
* let m_copydata use a void * instead of caddr_tdlg2021-02-251-4/+4
| | | | | | | i'm not a fan of having to cast to caddr_t when we have modern inventions like void *s we can take advantage of. ok claudio@ mvs@ bluhm@
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.