summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* style(9) and cleanup fixes.mdf2011-07-081-80/+69
| | | | MFC after: 1 week
* Add device ID for the Davicom 56PDV PCI Modem.jhb2011-07-081-0/+1
| | | | | | PR: kern/75132 Submitted by: Mike Tancsa @ Sentex (older patch against puc(4)) MFC after: 1 week
* Implement basic support for memory attributes. At this time we onlymarcel2011-07-083-27/+121
| | | | | | | | | | | | | distinguish between UC and WB memory so that we can map the page to either a region 6 address (for UC) or a region 7 address (for WB). This change is only now possible, because previously we would map regions 6 and 7 with 256MB translations and on top of that had the kernel mapped in region 7 using a wired translation. The introduction of the PBVM moved the kernel into its own region and freed up region 7 and allowed us to revert to standard page-sized translations. This commit inroduces pmap_page_to_va() that respects the attribute.
* Note that -a, -C, -H, -j, and -z are also toggles.jhb2011-07-081-1/+6
| | | | | | PR: bin/158682 Reported by: arundel MFC after: 3 days
* Fix the "passability" test in fdcopy().jonathan2011-07-081-2/+2
| | | | | | | | | | | | | | | Rather than checking to see if a descriptor is a kqueue, check to see if its fileops flags include DFLAG_PASSABLE. At the moment, these two tests are equivalent, but this will change with the addition of capabilities that wrap kqueues but are themselves of type DTYPE_CAPABILITY. We already have the DFLAG_PASSABLE abstraction, so let's use it. This change has been tested with [the newly improved] tools/regression/kqueue. Approved by: mentor (rwatson), re (Capsicum blanket) Sponsored by: Google Inc
* Clarify the meaning of a test.jonathan2011-07-082-4/+4
| | | | | | | | | | | | Rather than using err() if either of two failure conditions fires (which can produce spurious error messages), just use errx() if the one condition that really matters fires. In practice, this single test is enough to detect the failure mode we're looking for (kqueue being inherited across fork). Approved by: mentor (rwatson), re (Capsicum blanket) Sponsored by: Google Inc
* Add new USB 3G driver.hselasky2011-07-086-3/+1883
| | | | | Submitted by: PseudoCylon <moonlightakkiy@yahoo.ca> MFC after: 14 days
* In the experimental soreceive_stream():andre2011-07-081-13/+9
| | | | | | | | | | | o Move the non-blocking socket test below the SBS_CANTRCVMORE so that EOF is correctly returned on a remote connection close. o In the non-blocking socket test compare SS_NBIO against the so->so_state field instead of the incorrect sb->sb_state field. o Simplify the ENOTCONN test by removing cases that can't occur. Submitted by: trociny (with some further tweaks by committer) Tested by: trociny
* Permit ARP to proceed for IPv4 host routes for which the gateway is thezec2011-07-083-7/+12
| | | | | | | | | same as the host address. This already works fine for INET6 and ND6. While here, remove two function pointers from struct lltable which are only initialized but never used. MFC after: 3 days
* Fixing building bustage on 32 bits platforms when WARNS >= 2. Note thatavatar2011-07-082-5/+10
| | | | | | | | | this fix only applies to zalloc.c, the other part of libstand such like qdivrem.c still gives compilation warnings on sparc64 tinderbox builds; therefore, WARNS level isn't changed for now. Submitted by: Garrett Cooper <yanegomi@gmail.com> Reviewed by: bde
* Delete the /etc/rc.d/nfsserver script, since it is normacklem2011-07-086-23/+6
| | | | | | | | longer used by /etc/rc.d/nfsd and it is no longer necessary to load the old nfs server by default, when nfs_server_enable="YES". Tested by: sgk at troutmask.apl.washington.edu Reviewed by: rc (Andrzej Tobola)
* Note the PF version.obrien2011-07-071-2/+7
| | | | Discussed with: bz
* Fix the /etc/rc.d/nfsd script so that it no longer usesrmacklem2011-07-071-3/+3
| | | | | | | | the /etc/rc.d/nfsserver script to load the old nfs server module. Tested by: sgk at troutmask.apl.washington.edu Reviewed by: rc (hrs)
* Grab the rlock before checking if our interface is enabled, it could bethompsa2011-07-071-1/+2
| | | | | | | | possible to hit a dead pointer when changing interfaces. PR: kern/156978 Submitted by: Andrew Boyer MFC after: 1 week
* Ensure that kqueue is not inherited across fork().jonathan2011-07-073-6/+21
| | | | | | | | | | | | | | | Modify the existing unit test (from libkqueue) which already exercises process events via fork() and kill(). Now, the child process simply checks that the 'kqfd' descriptor is invalid. Some minor modifications were required to make err() work correctly. It seems that this test was imported using the output of a configure script, but config.h was not included in key places, nor was its syntax correct (need '#define HAVE_FOO 1' rather than '#define HAVE_FOO'). Finally, change main() to run the "proc" suite by default, but widened the '#if TODO' in proc.c to include the non-functioning test event_trigger(). Approved by: mentor (rwatson), re (Capsicum blanket) Sponsored by: Google Inc
* Style fix - macros are supposed to be uppercase.trasz2011-07-073-23/+23
|
* Make a comment more accurate.jonathan2011-07-071-1/+1
| | | | | | | | | This comment refers to CAP_NT_SMBS, which does not exist; it should refer to SMB_CAP_NT_SMBS. Fixing this comment makes it easier for people interested in Capsicum to grep around for capability rights, whose identifiers are of the form 'CAP_[A-Z_]'. Approved by: mentor (rwatson), re (Capsicum blanket) Sponsored by: Google Inc
* Fix memset sizeofskevlo2011-07-072-2/+2
| | | | Reviewed by: bschmidt
* Always skip the kernel idle process if requested, it is not specific tojhb2011-07-071-5/+4
| | | | | | | | the 'CPU' mode. PR: bin/158677 Reported by: arundel MFC after: 3 days
* Add again the checking for log_arp_permanent_modify that was by accidentae2011-07-071-5/+7
| | | | | | | removed in the r186119. PR: kern/154831 MFC after: 1 week
* Remove the TCP_SORECEIVE_STREAM compile time option. The use ofandre2011-07-073-7/+2
| | | | | | | soreceive_stream() for TCP still has to be enabled with the loader tuneable net.inet.tcp.soreceive_stream. Suggested by: trociny and others
* Add an extra tab between fs_file and fs_vfstypekevlo2011-07-071-1/+1
| | | | Reviewed by: nwhitehorn
* A fix to make the LINT-NOINET build happy, if thisjfv2011-07-071-3/+3
| | | | works out the ixgbe driver should be changed as well.
* All the racct_*() calls need to happen with the proc locked. Fixing thistrasz2011-07-0614-1/+86
| | | | | | won't happen before 9.0. This commit adds "#ifdef RACCT" around all the "PROC_LOCK(p); racct_whatever(p, ...); PROC_UNLOCK(p)" instances, in order to avoid useless locking/unlocking in kernels built without "options RACCT".
* Conditionally compile in the IPv4/IPv6 address family support.pluknet2011-07-062-4/+24
| | | | | | | Test it with feature_present(3), so that we do not end up with passing an unsupported version to NgSendMsg(3). Approved by: glebius
* Handle a race between device_pager and devsw in a more graceful manner:attilio2011-07-061-2/+4
| | | | | | | | | return an error code rather than panic the kernel. Sponsored by: Sandvine Incorporated Reviewed by: kib Tested by: pho MFC after: 2 weeks
* Add missing unlocks.glebius2011-07-061-0/+2
|
* Only increment the ypcnt once per data record. The old implementationgordon2011-07-061-4/+4
| | | | | | | was incrementing it twice making it impossible to iterate the table since the records were 1, 3, 5, 7 (or 2, 4, 6, 8 for the v3 records). MFC after: 10 days
* Update to version 9.6-ESV-R4-P3dougb2011-07-0613-64/+96
| | | | | | | | | | | | | | | | | | | | | | ALL BIND USERS ARE ENCOURAGED TO UPGRADE IMMEDIATELY This update addresses the following vulnerability: CVE-2011-2464 ============= Severity: High Exploitable: Remotely Description: A defect in the affected BIND 9 versions allows an attacker to remotely cause the "named" process to exit using a specially crafted packet. This defect affects both recursive and authoritative servers. The code location of the defect makes it impossible to protect BIND using ACLs configured within named.conf or by disabling any features at compile-time or run-time. https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2011-2464 https://www.isc.org/software/bind/advisories/cve-2011-2464
* Apply bug fixesdougb2011-07-061-10/+9
| | | | Submitted by: marius
* Remove the IDR_CHEETAH_MAX_BN_PAIRS limit from cheetah_ipi_selected().marius2011-07-051-2/+2
| | | | | | This is just a simple approach. For reasons unknown OpenSolaris uses a more sophisticated one involving IPIing the remaining CPUs in reverse order after the first batch of 32.
* Delete files made obsolete because they were moved to sys/nfsrmacklem2011-07-051-0/+3
| | | | in order to be shared by both NFS clients.
* It can be useful to know which page still has mappings.marius2011-07-051-1/+1
|
* - pmap_cache_remove() and pmap_protect_tte() are only used within pmap.cmarius2011-07-052-6/+6
| | | | | so static'ize them. - Correct a typo.
* Remove #ifdef notyet code dating back to 4.3BSD Net/2 (and possibly earlier).cperciva2011-07-051-18/+1
| | | | | | | | I think the benefit of making the code cleaner and easier to understand outweighs the humour of leaving this intact (or possibly changing it to #ifdef not_yet_and_probably_never). MFC after: 2 weeks
* In pmap_remove_all() assert that the page is neither fictitious normarius2011-07-051-0/+2
| | | | | | unmanaged as also done on other architectures. Reviewed by: alc
* Don't allow lro->len to exceed 65535, as this will result in overflowcperciva2011-07-051-0/+8
| | | | | | | | | | | when len is inserted back into the synthetic IP packet and cause a multiple of 2^16 bytes of TCP "packet loss". This improves Linux->FreeBSD netperf bandwidth by a factor of 300 in testing on Amazon EC2. Reviewed by: jfv MFC after: 2 weeks
* Correct cpu_monitor() and cpu_mwait() for amd64. These instructions takejkim2011-07-052-10/+14
| | | | | | %rcx as "extensions" in long mode. If any unused bit is set in %rcx, these instructions cause general protection fault. Fix style nits and synchronize i386 with amd64.
* Call pmap_qremove() before freeing or unwiring the pages, otherwisemarius2011-07-052-4/+6
| | | | | | | | there's a window during which a page can be re-used before its previous mapping is removed. Reviewed by: alc MFC after: 1 week
* Delete files made obsolete by the change in default NFS clientrmacklem2011-07-051-0/+4
| | | | and the move of nfs_kdtrace.h to sys/nfs.
* Follow Linux by unconditionally stripping the RX vlan tag from incomingnwhitehorn2011-07-051-4/+7
| | | | | | | packets. It turns out that all firmware versions insert it, whether or not they support VLAN tagging. Submitted by: glevand <geoffrey.levand at mail dot ru>
* Update manual page. Mention IPv6 support, and notice that mainglebius2011-07-051-3/+8
| | | | argument isn't actually a node, but a netgraph path.
* Rewrite the flowctl utility to add it support for displayingglebius2011-07-052-53/+168
| | | | both IPv4 and IPv4 flows.
* o Eliminate flow6_hash_entry in favor of flow_hash_entry. We don't needglebius2011-07-053-107/+152
| | | | | | | | | | a separate struct to start a slist of semi-opaque structs. This makes some code more compact. o Rewrite ng_netflow_flow_show() and its API/ABI: - Support for IPv6 is added. - Request and response now use same struct. Structure specifies version (6 or 4), index of last retrieved hash, and also index of last retrieved entry in the hash entry.
* Only print entries for which ut_host points to a character device.ed2011-07-051-1/+1
| | | | | | Now that we use utmpx, we more often have entries for which the ut_line is left blank. To prevent us from returning struct stat for "/dev/", check that the resulting stat structure belongs to a character device.
* Rework _fget to accept capability parameters.jonathan2011-07-052-6/+76
| | | | | | | | | | | | | | | | | | | | | This new version of _fget() requires new parameters: - cap_rights_t needrights the rights that we expect the capability's rights mask to include (e.g. CAP_READ if we are going to read from the file) - cap_rights_t *haverights used to return the capability's rights mask (ignored if NULL) - u_char *maxprotp the maximum mmap() rights (e.g. VM_PROT_READ) that can be permitted (only used if we are going to mmap the file; ignored if NULL) - int fget_flags FGET_GETCAP if we want to return the capability itself, rather than the underlying object which it wraps Approved by: mentor (rwatson), re (Capsicum blanket) Sponsored by: Google Inc
* Remove useless initialization.trociny2011-07-051-2/+1
| | | | | Approved by: pjd (mentor) MFC after: 3 days
* The algorithm used by nfscl_getopen() could have resulted inrmacklem2011-07-041-81/+94
| | | | | | | | | | | | | | | | multiple instances of the same lock_owner when a process both inherited an open file descriptor plus opened the same file itself. Since some NFSv4 servers cannot handle multiple instances of the same lock_owner string, this patch changes the algorithm used by nfscl_getopen() in the new NFSv4 client to keep that from happening. The new algorithm is simpler, since there is no longer any need to ascend the process's parentage tree because all NFSv4 Closes for a file are done at VOP_INACTIVE()/VOP_RECLAIM(), making the Opens indistinct w.r.t. use with Lock Ops. This problem was discovered at the recent NFSv4 interoperability Bakeathon. MFC after: 2 weeks
* - General grammar and mdoc(7) fixes. [1] [2]gjb2011-07-041-65/+66
| | | | | | | | | | - While here, remove a paragraph about userspace operation that has been outdated for some time. [2] PR: 158623 Submitted by: Ben Kudak (kaduk % mit!edu) [1] Reviewed by: glebius [2] MFC after: 1 week
* - Speed up pendingblock processing again. Having too much delay betweenjeff2011-07-042-15/+41
| | | | | ffs_blkfree() and the pending adjustment causes all kinds of space related problems.
OpenPOWER on IntegriCloud