summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* MFC r290383:Luiz Otavio O Souza2015-11-069-127/+19
| | | | | | | | | | | | | | Replace the fastforward path with tryforward which does not require a sysctl and will always be on. The former split between default and fast forwarding is removed by this commit while preserving the ability to use all network stack features. Differential Revision: https://reviews.freebsd.org/D4042 Reviewed by: ae, melifaro, olivier, rwatson MFC after: 1 month Sponsored by: Rubicon Communications (Netgate) TAG: tryforward
* Revert "Replace the fastforward path with tryforward which does not require ↵Luiz Otavio O Souza2015-11-069-10/+124
| | | | | | | | a sysctl and will always be on." This reverts commit c58873dc9abc56028cc3435f692fd3583bd143af. TAG: tryforward
* Revert "If there are any IPSEC security policies skip the tryforward step."Luiz Otavio O Souza2015-11-061-9/+0
| | | | | | This reverts commit de3b422781821f809b436ced1af4fc3dd4dc0c1a. TAG: tryforward
* Fix a bad merge conflict resolution from c7275b999Renato Botelho2015-11-051-1/+0
|
* Merge branch 'stable/10' into develRenato Botelho2015-11-03727-26468/+22818
|\
| * MFC r285914, r289029 and r289560:hselasky2015-11-032-82/+81
| | | | | | | | | | | | | | | | | | | | | | | | | | | | - Move the remainder of host controller capability registers reading from xhci_start_controller() to xhci_init(). These values don't change at run- time so there's no point of acquiring them on every USB_HW_POWER_RESUME instead of only once during initialization. In r276717, reading the first couple of registers in question already had been moved as a prerequisite for the changes in that revision. - Identify ASMedia ASM1042A controllers. - Use NULL instead of 0 for pointers. - Add quirks for USB 3.0 PCI devices. PR: 203650
| * MFC r289661:kib2015-11-031-1/+1
| | | | | | | | | | Mark struct thread zone as type-stable, to prevent dereference of the freed memory in the locks spin loops.
| * MFC r289660,r289664:kib2015-11-031-2/+13
| | | | | | | | | | Do not allow to execute ptrace(PT_TRACE_ME) when the process is already traced or when there is no parent which can trace the process.
| * MFC r289658:kib2015-11-031-1/+1
| | | | | | | | | | No need to dereference struct proc to pids when comparing processes for equality.
| * MFC r287413: Minor code cleanups (no functional changes).dteske2015-11-021-14/+12
| |
| * MFC r287390: Bump version for altered long-opts processingdteske2015-11-021-1/+1
| |
| * MFC r287389: Simplify long-option processingdteske2015-11-021-14/+10
| |
| * MFC r287385: Bump version for prior fix (SVN r287381)dteske2015-11-021-1/+1
| |
| * MFC r287384:dteske2015-11-021-1/+1
| | | | | | | | Style: Remove whitespace around brackets from function syntax options
| * MFC r287383: Comment for escape() function.dteske2015-11-021-0/+8
| |
| * MFC r287382: Commentdteske2015-11-021-1/+1
| |
| * MFC r287381: Properly escape arguments when moving into jail or chrootdteske2015-11-021-1/+22
| |
| * MFC r287380: Style: commentsdteske2015-11-021-4/+4
| |
| * MFC r287379: Style consistency: add single space before each `;;' case entrydteske2015-11-021-19/+19
| |
| * MFC r287378:dteske2015-11-021-12/+12
| | | | | | | | | | Remove `SYSRC_' prefix from $SYSRC_VERBOSE (prefix unnecessary since this is a non-inheritable attribute; was previously).
| * MFC r287421: Fix mandoc(1) "WARNING: end of line whitespace"dteske2015-11-021-15/+15
| |
| * Snap to breuffer's r286660 (incorrectly merged)dteske2015-11-021-2/+0
| |
| * MFC r287420: Remove non-functional examples.dteske2015-11-021-22/+1
| |
| * MFC r287683: Ignore error results from newaliases(1)dteske2015-11-021-0/+2
| |
| * MFC r287685: Remove use of return outside of functiondteske2015-11-021-1/+1
| |
| * MFC r287686: Produce meaningful exit codedteske2015-11-021-1/+4
| |
| * MFC r287687: Update copyrightsdteske2015-11-023-2/+3
| |
| * MFC r287688: Explicitly exit with successdteske2015-11-021-1/+2
| |
| * MFC r287689: Explicitly exit with successdteske2015-11-021-1/+2
| |
| * MFC r287690: Produce meaningful exit codedteske2015-11-022-6/+14
| |
| * MFC r287691: Remove trailing newline at EOFdteske2015-11-021-1/+0
| |
| * MFC r287694: Use a variable for readabilitydteske2015-11-021-2/+2
| |
| * MFC r287695: Replace `return' outside of function with exitdteske2015-11-021-1/+1
| | | | | | | | Pointy hat to: me
| * MFC r287692: Better to reset trap and explicitly exit successdteske2015-11-021-1/+2
| |
| * MFC r287693: Update copyrightdteske2015-11-021-1/+1
| |
| * MFC r287803: Fix code typo (no functional change)dteske2015-11-021-3/+3
| |
| * MFC r287635: Fix argument quoting and remove unnecessary bracesdteske2015-11-025-63/+63
| |
| * MFC r287696:dteske2015-11-021-3/+4
| | | | | | | | | | | | | | | | | | | | | | The <arch>/mkisoimages.sh script in release knows how to add extra bits from an "xtra-bits-dir". This feature is unusable from release/Makefile. Add an XTRADIR setting to use it. MFC r287697: Whitespace alignment Differential Revision: https://reviews.freebsd.org/D3633 Reviewed by: kmacy Relnotes: yes
| * MFC r289790: dpv(3) merged to stable/10 before release/10.2.0dteske2015-11-021-1/+1
| |
| * MFC r289793: Bump date/copyright after correcting HISTORYdteske2015-11-021-2/+2
| |
| * MFC r289794: figpar(3) merged to stable/10 before release/10.2.0dteske2015-11-021-3/+3
| |
| * MFC r290163: Ignore per-mdN settings in mdconfig[2] startupdteske2015-11-022-0/+4
| | | | | | | | | | PR: base/189696 Submitted by: ganael.laplanche@martymac.org
| * MFC: r290121jkim2015-11-021-2/+10
| | | | | | | | Define endianness for non-x86 platforms.
| * MFC r289496:kib2015-11-021-5/+17
| | | | | | | | | | Modify the 'unchanged' calculation bu dereferencing the marker tailq pointers, which is known to belong to the queue.
| * MFC r289229:ngie2015-11-012-5/+5
| | | | | | | | | | | | | | | | | | Replace references to /dev/acd0 with /dev/cd0 atapicd(4) was replaced by cd(4) with the atacam work done by mav@ and then removed in r249083 X-MFC to: stable/10
| * MFC: 289879bapt2015-10-312-16/+48
| | | | | | | | | | | | | | | | newsyslog.conf: allow to configure the signal using the signal name. Submitted by: Alexandre Perrin <alex@kaworu.ch> Relnotes: yes Differential Revision: https://reviews.freebsd.org/D3961
| * Long-overdue MFC of r280930:wollman2015-10-302-26/+26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix overflow bugs in and remove obsolete limit from kernel RPC implementation. The kernel RPC code, which is responsible for the low-level scheduling of incoming NFS requests, contains a throttling mechanism that prevents too much kernel memory from being tied up by NFS requests that are being serviced. When the throttle is engaged, the RPC layer stops servicing incoming NFS sockets, resulting ultimately in backpressure on the clients (if they're using TCP). However, this is a very heavy-handed mechanism as it prevents all clients from making any requests, regardless of how heavy or light they are. (Thus, when engaged, the throttle often prevents clients from even mounting the filesystem.) The throttle mechanism applies specifically to requests that have been received by the RPC layer (from a TCP or UDP socket) and are queued waiting to be serviced by one of the nfsd threads; it does not limit the amount of backlog in the socket buffers. The original implementation limited the total bytes of queued requests to the minimum of a quarter of (nmbclusters * MCLBYTES) and 45 MiB. The former limit seems reasonable, since requests queued in the socket buffers and replies being constructed to the requests in progress will all require some amount of network memory, but the 45 MiB limit is plainly ridiculous for modern memory sizes: when running 256 service threads on a busy server, 45 MiB would result in just a single maximum-sized NFS3PROC_WRITE queued per thread before throttling. Removing this limit exposed integer-overflow bugs in the original computation, and related bugs in the routines that actually account for the amount of traffic enqueued for service threads. The old implementation also attempted to reduce accounting overhead by batching updates until each queue is fully drained, but this is prone to livelock, resulting in repeated accumulate-throttle-drain cycles on a busy server. Various data types are changed to long or unsigned long; explicit 64-bit types are not used due to the unavailability of 64-bit atomics on many 32-bit platforms, but those platforms also cannot support nmbclusters large enough to cause overflow. This code (in a 10.1 kernel) is presently running on production NFS servers at CSAIL. Summary of this revision: * Removes 45 MiB limit on requests queued for nfsd service threads * Fixes integer-overflow and signedness bugs * Avoids unnecessary throttling by not deferring accounting for completed requests Differential Revision: https://reviews.freebsd.org/D2165 Reviewed by: rmacklem, mav Relnotes: yes Sponsored by: MIT Computer Science & Artificial Intelligence Laboratory
| * MFC r289824:kib2015-10-302-0/+14
| | | | | | | | | | | | | | Add CLFLUSHOPT instruction wrappers. MFC r290188: Fix prefix on i386.
| * MFC r289823:kib2015-10-301-0/+12
| | | | | | | | | | Decode new values for CPUID leaf 2 cache and TLB descriptors, from the Intel SDM revision 56.
| * MFC r289819: Fix LUN disable in CAM broken at r285155.mav2015-10-301-1/+1
| |
OpenPOWER on IntegriCloud