summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* MFC r292638davidcs2016-01-123-4/+23
| | | | | | | | | | Check for packet_length is greater than 60 bytes as well as packet_length is greater than len_on_bd, before invoking the routine to handle jumbo over SGL (bxe_service_rxsgl()). Add counters for number of jumbo_over_SGL packets (rx_bxe_service_rxsgl) and erroneous jumbo_over_SGL packets (rx_erroneous_jumbo_sge_pkts) Fix formatting in bxe_sysctl_state()
* MFC r289199davidcs2016-01-121-1/+4
| | | | Add support for reading device temperature
* MFC r292950:dim2016-01-124-28/+7
| | | | | | Drop the clang patch which adds recognition of 'CC' suffixes as aliases for --driver-mode=g++, since this was never upstreamed. For backwards compatibility, add a wrapper shell script.
* Fix LINT-NOINET kernelsngie2016-01-121-1/+12
| | | | | | | | | | | | | | | | - Generate opt_inet.h appropriately according to MK_INET_SUPPORT - Only add in_gif.c to SRCS if MK_INET_SUPPORT != no This is a surgical fix since the build infrastructure has diverged sufficiently between stable/10 and head. This change will likely be reverted if and when imp's changes to config.mk have been backported to ^/stable/10 . Reviewed by: ae (no serious objections) Differential Revision: https://reviews.freebsd.org/D4864 MFC after: 1 week Sponsored by: EMCC / Isilon Storage Division
* MFC: r284589allanjude2016-01-126-20/+242
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add the ability to detect ZFS and GELI encrypted file systems to fstyp(8) MFC: r284644 Fix GCC Warnings MFC: r284728 Only build ZFS support in absense of WITHOUT_ZFS MFC: r285426 Remove excess copyrights MFC: r286569 Use GELI sentinel constant MFC: r287937 Eliminate unneeded copying of vdev data, goto, etc. and add a note that checksum of vdev label should be checked (which is not done currently). No functional change. While I'm there, raise WARNS to 2. MFC: r292757 Fix order of includes in usr.sbin/fstyp/zfs.c MFC: r292829 Eliminate unneeded includes.
* Hide the "unmount of /dev failed (BUSY)" warning at shutdown or reboot,trasz2016-01-121-1/+1
| | | | | | | | | | | introduced with r293742, just like it was hidden before that commit. This is a direct commit to 10-STABLE; this special case is not needed in 11-CURRENT, because devfs supports forced unmounts there. The forced unmount could be MFC-ed, but there are some LORs at shutdown, and I have a weird feelings about it. Sponsored by: The FreeBSD Foundation
* MFC r290689:trasz2016-01-121-8/+9
| | | | | | Fix resource leaks in error cases. Sponsored by: The FreeBSD Foundation
* MFC r290548:trasz2016-01-125-22/+316
| | | | | | | | | | | | Userspace part of reroot support. This makes it possible to change the root filesystem without full reboot, using "reboot -r". This can be used to to eg. boot from a temporary md_image preloaded by loader(8), setup an iSCSI session, and continue booting from rootfs mounted over iSCSI. Relnotes: yes Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D3693
* MFC r287964:trasz2016-01-124-21/+167
| | | | | | | | | | | | | Kernel part of reroot support - a way to change rootfs without reboot. Note that the mountlist manipulations are somewhat fragile, and not very pretty. The reason for this is to avoid changing vfs_mountroot(), which is (obviously) rather mission-critical, but not very well documented, and thus hard to test properly. It might be possible to rework it to use its own simple root mount mechanism instead of vfs_mountroot(). Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D2698
* MFC r287107:trasz2016-01-124-28/+38
| | | | | | | | | Make vfs_unmountall() unmount /dev after /, not before. The only reason this didn't result in an unclean shutdown is that devfs ignores MNT_FORCE flag. Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D3467
* MFC r285240:trasz2016-01-124-99/+2
| | | | | | | Remove reboot.S (part of libc). It's not needed and was actually broken - returning 0 from reboot(2) resulted in SIGBUS. Sponsored by: The FreeBSD Foundation
* MFC r289110:trasz2016-01-123-4/+47
| | | | | | | | | | Make geom_nop(4) collect statistics on all types of BIOs, not just reads and writes. PR: kern/198405 Submitted by: Matthew D. Fuller <fullermd at over-yonder dot net> Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D3679
* MFC r292989:hselasky2016-01-121-3/+4
| | | | | Handle when filedescriptors are closed before initialized. An early fdclose() call can cause fget_unlocked() to fail.
* MFC r293192:hselasky2016-01-121-6/+21
| | | | Fix for directly connected FULL or LOW speed USB devices.
* MFC: r287842allanjude2016-01-121-9/+6
| | | | | | Make ifconfig always exit with an error code if an important ioctl fails PR: 203062
* MFC: r287576allanjude2016-01-121-0/+2
| | | | | | | service(8) -e does not respect /etc/rc.conf.d/* entries PR: 173454 Submitted by: giantlock@gmail.com (original patch)
* MFC: r287528allanjude2016-01-121-13/+24
| | | | | | | Document the sctp blackhole sysctl MIB PR: 184110 Submitted by: Marie Helene Kvello-Aune <marieheleneka@gmail.com>
* MFC: r285552allanjude2016-01-127-10/+44
| | | | | | | | | | | | | | | | | | | Use strtonum(3) instead of atoi(3) MFC: r286289 Introduce -P0, creating as many concurrent processes as possible Relnotes: yes MFC: r286461 Fix regression in xargs -Px, add more regression tests MFC: r287004 MFC: r287005 Fix the racy xargs -P0 -n2 test added in r286289 PR: 199976 Submitted by: Nikolai Lifanov <lifanov@mail.lifanov.com>
* MFC r293188:gjb2016-01-124-0/+8
| | | | | | | | | | Prevent memstick installation medium from attempting to mount the root filesystem read-write. This causes problems booting the memstick installation medium from write-protected USB flash drives. PR: 187161, 205886 Sponsored by: The FreeBSD Foundation
* MFC: r292011hiren2016-01-112-8/+19
| | | | | | | MFC: r292012 Add an option to use rfc6675 based pipe/inflight bytes calculation in cubic and newreno.
* MFC: r292003hiren2016-01-113-13/+48
| | | | Improve tcp duplicate ack processing when SACK is present.
* MFC: r290122hiren2016-01-113-4/+40
| | | | | | | | | Calculate the correct amount of bytes that are in-flight for a connection as suggested by RFC 6675. MFC: r292046 r290122 added 4 bytes and removed 8 in struct sackhint. Add a pad entry of 4 bytes to restore the size.
* Merge from head r292743, r293627 by dchagin:glebius2016-01-111-0/+6
| | | | | | | | | | | | | ------------------------------------------------------------------------ r292743 | dchagin | 2015-12-26 01:04:47 -0800 (сб, 26 дек. 2015) | 5 lines Do not allow access to emuldata for non Linux processes. ------------------------------------------------------------------------ r293627 | dchagin | 2016-01-09 23:36:43 -0800 (сб, 09 янв. 2016) | 6 lines Unlock process lock when return error from getrobustlist call and add an forgotten dtrace probe when return the same error.
* MFC r292218asomers2016-01-112-6/+22
| | | | | | | | | | | | | | Don't retry SAS commands in response to protocol errors sys/dev/mpr/mpr_sas_lsi.c sys/dev/mps/mps_sas_lsi.c When mp[rs]sas_get_sata_identify returns MPI2_IOCSTATUS_SCSI_PROTOCOL_ERROR, don't bother retrying. Protocol errors aren't likely to be fixed by sleeping. Without this change, a system that generated may protocol errors due to signal integrity issues was taking more than an hour to boot, due to all the retries.
* MFC r293440:bdrewery2016-01-111-7/+4
| | | | Fix upgrading from OSVERSION 1000002-1000032 after r288829.
* MFC r292019asomers2016-01-111-2/+57
| | | | | | | When iostat(8) receives SIGINT while running with "-w" or "-c", it will now print statistics one more time before exiting. Also, it now implements the wait using setitimer instead of sleep, so the waits will be more consistent when the system is heavily loaded.
* MFC r292020asomers2016-01-111-6/+17
| | | | | | | Increase devd's client socket buffer size to 256KB. This is not as large as it looks, because we'll hit the sockbuf's mbuf limit long before hitting its data limit. A 256KB data limit allows creating a ZFS pool on about 450 drives without overflowing the client socket buffers.
* MFC r289060:trasz2016-01-111-1/+77
| | | | | | Document iSCSI and autofs(5) variables in rc.conf(5). Sponsored by: The FreeBSD Foundation
* MFC r287429:trasz2016-01-111-1/+0
| | | | | | | The dumpfs(8) utility doesn't actually use disktab(5); remove it from "SEE ALSO". Sponsored by: The FreeBSD Foundation
* MFC r287396:trasz2016-01-1115-24/+25
| | | | | | | | | | | It's 2015, and some people are still trying to use fdisk and then go asking what debug flags to set for GEOM to make it work. Advice them to use gpart(8) instead. Something similar should probably done with disklabel, but I need to rewrite the disklabel examples first. Sponsored by: The FreeBSD Foundation
* MFC r287704:trasz2016-01-111-1/+8
| | | | | | Point potential geom_fox(4) users to gmultipath(8). Sponsored by: The FreeBSD Foundation
* MFC 289113:trasz2016-01-112-6/+9
| | | | | | Add 'p' postfix to mdconfig(8). Sponsored by: The FreeBSD Foundation
* MFC r289111:trasz2016-01-111-6/+2
| | | | | | Provide better debug message on kernel module name clash. Sponsored by: The FreeBSD Foundation
* MFC r290408:trasz2016-01-111-1/+1
| | | | | | Make rctl(8) use more reasonable buffer size for retrieving the rules. Sponsored by: The FreeBSD Foundation
* MFC: r290153 (by bdrewery)bapt2016-01-111-1/+3
| | | | | | | | | Fix unlikely memory leak. It is unlikely since the first check in the function is that dir[0] is '/', but later code changes may make it real. Coverity CID: 1332104
* MFC: 278899bapt2016-01-111-1/+1
| | | | Regression: fix usershow -7
* MFC r290404:trasz2016-01-111-5/+5
| | | | | | Fix markup in rctl(8) - the 'rule' or 'filter' arguments are not optional. Sponsored by: The FreeBSD Foundation
* MFC r293369:jimharris2016-01-111-1/+1
| | | | ismt: fix ISMT_DESC_ADDR_RW macro
* MFC r266474:jimharris2016-01-117-1/+850
| | | | | | | | | Add ismt(4) driver. ismt(4) supports the SMBus Message Transport controller found on Intel C2000 series (Avoton) and S1200 series (Briarwood) Atom SoCs. Relnotes: Yes
* MFC r293354:jimharris2016-01-111-9/+3
| | | | nvme: replace NVME_CEILING macro with howmany()
* MFC r293352:jimharris2016-01-113-9/+41
| | | | | | | | | | | | | | | | | | | | | | nvme: add hw.nvme.min_cpus_per_ioq tunable Due to FreeBSD system-wide limits on number of MSI-X vectors (https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=199321), it may be desirable to allocate fewer than the maximum number of vectors for an NVMe device, in order to save vectors for other devices (usually Ethernet) that can take better advantage of them and may be probed after NVMe. This tunable is expressed in terms of minimum number of CPUs per I/O queue instead of max number of queues per controller, to allow for a more even distribution of CPUs per queue. This avoids cases where some number of CPUs have a dedicated queue, but other CPUs need to share queues. Ideally the PR referenced above will eventually be fixed and the mechanism implemented here becomes obsolete anyways. While here, fix a bug in the CPUs per I/O queue calculation to properly account for the admin queue's MSI-X vector.
* MFC r293328:jimharris2016-01-113-64/+106
| | | | | | | | | | nvme: do not revert to single I/O queue when per-CPU queues not available Previously nvme(4) would revert to a single I/O queue if it could not allocate enought interrupt vectors or NVMe submission/completion queues to have one I/O queue per core. This patch determines how to utilize a smaller number of available interrupt vectors, and assigns (as closely as possible) an equal number of cores to each associated I/O queue.
* MFC r293327:jimharris2016-01-111-66/+63
| | | | nvme: break out interrupt setup code into a separate function
* MFC r293326:jimharris2016-01-113-39/+3
| | | | | | | nvme: do not pre-allocate MSI-X IRQ resources The issue referenced here was resolved by other changes in recent commits, so this code is no longer needed.
* MFC r293325:jimharris2016-01-112-9/+3
| | | | | | | | | | nvme: remove per_cpu_io_queues from struct nvme_controller Instead just use num_io_queues to make this determination. This prepares for some future changes enabling use of multiple queues when we do not have enough queues or MSI-X vectors for one queue per CPU.
* MFC r293324:jimharris2016-01-111-16/+20
| | | | | | | nvme: simplify some of the nested ifs in interrupt setup code This prepares for some follow-up commits which do more work in this area.
* MFC r293323:jimharris2016-01-111-0/+18
| | | | | | | | | nvd: submit bios directly when BIO_ORDERED not set or in flight This significantly improves parallelism in the most common case. The taskqueue is still used whenever BIO_ORDERED bios are in flight. This patch is based heavily on a patch from gallatin@.
* MFC r293322:jimharris2016-01-111-12/+23
| | | | | | | nvd: break out submission logic into separate function This enables a future patch using this same logic to submit I/O directly bypassing the taskqueue.
* MFC r293321:jimharris2016-01-111-0/+1
| | | | | | | | nvd: skip BIO_ORDERED logic when bio fails submission This ensures the bio flags are not read after biodone(). The ordering will still be enforced, after the bio is submitted successfully.
* MFC r293320:jimharris2016-01-111-13/+0
| | | | | | | nvd: do not wait for previous bios before submitting ordered bio Still wait until all in-flight bios (including the ordered bio) complete before processing more bios from the queue.
OpenPOWER on IntegriCloud