summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* MFC r308247: MFV r308222: 6051 lzc_receive: allow the caller to read theavg2016-11-172-8/+41
| | | | begin record
* MFC r308218: Add support for microcode update on newer AMD CPUs (10h+)avg2016-11-175-30/+374
|
* MFC r308101: hwpmc: fix a race between amd_stop_pmc and amd_intravg2016-11-171-1/+2
|
* MFC r307195: convert iicsmb to use iicbus_transfer for all operationsavg2016-11-171-168/+142
|
* MFC r308040,308479: nap time between pats is forced to be at most halfavg2016-11-172-25/+30
| | | | | | | of the timeout Note that in this branch the default nap period is 1 second unlike the head where the period is 10 seconds.
* MFC r308464, r308471: Add some device IDs found in my new laptop.mav2016-11-165-0/+16
|
* MFC r308348:hrs2016-11-161-0/+4
| | | | | | Add link-layer address option in RA even for IFT_L2VLAN and IFT_BRIDGE. Reported by: philip
* MFC r308347:hrs2016-11-161-0/+3
| | | | | | | | Fix an infinite loop at an non-responding hop when other echo replies are kept arriving in the waittime time window. Submitted by: Denny Page PR: 210286
* MFC r308416:hselasky2016-11-153-2/+12
| | | | | | | | | | | Add timer to watch the RQ when we are out of mbufs. The firmware/hardware does not generate additional completion events unless we post new buffers. Use a timer to try to post more buffers in case we are temporarily out of mbufs. Else the receive schedule completely stops. Sponsored by: Mellanox Technologies
* MFC r308414:hselasky2016-11-155-43/+467
| | | | | | | Add more firmware related structures and update existing ones in the MLX5 core module. Update the set and query diagnostics counter API. Sponsored by: Mellanox Technologies
* MFC r308413:hselasky2016-11-151-1/+5
| | | | | | | Query flow table capabilities according to the correct capability bit for infiniband. Sponsored by: Mellanox Technologies
* MFC r308412:hselasky2016-11-151-4/+5
| | | | | | | Correct checksum fields in the "mlx5_mini_cqe8" structure. The fields in question are currently not used. Sponsored by: Mellanox Technologies
* MFC r308411:hselasky2016-11-151-5/+5
| | | | | | | | Ensure the firmware is notified of any host memory allocation failures. Else firmware commands may time out waiting for host memory. Sponsored by: Mellanox Technologies
* MFC r308409:hselasky2016-11-151-1/+4
| | | | | | | When a firmware command times out do not free the command structure to avoid use after free. Sponsored by: Mellanox Technologies
* MFC r308133, r308134: Fix wrong copy/paste in error message.mav2016-11-141-2/+2
|
* MFC r308173:mav2016-11-123-5/+14
| | | | | | | | | Fix ZIL records ordering when ZVOL opened both with and without FSYNC. Before this an earlier writes to a ZVOL opened without FSYNC could get to ZIL after later writes to the same ZVOL opened with FSYNC. Fix this by replicating functionality of ZPL (zv_sync_cnt equivalent to z_sync_cnt), marking all log records sync if anybody opened the ZVOL with FSYNC.
* MFC r308169:mav2016-11-121-5/+7
| | | | | | | Pass to zvol_log_truncate() same sync values as to zvol_log_write(). Surplus marking of TX_TRUNCATE records as sync could result in putting them into ZIL before previous writes if ones were async.
* MFC r308055: Add vdev_reopening support to vdev_geom.mav2016-11-121-1/+12
| | | | | | | It allows to avoid extra GEOM providers flapping without significant need. Since GEOM got resize support, we don't need to reopen provider to get new size. If provider was orphaned and no longer valid, ZFS should already know that, and in such case reopen should be done in full as expected.
* MFC r308051: Matching GUIDs, handle possible race on vdev detach.mav2016-11-121-56/+63
| | | | | | | | | | | | | | | | In case of vdev detach, causing top level mirror vdev destruction, leaf vdev changes its GUID to one of the destroyed mirror, that creates race condition when GUID in vdev label may not match one in the pool config. This change replicates logic nuance of vdev_validate() by adding special exception, matching the vdev GUID against the top level vdev GUID. Since this exception is not completely reliable (may give false positives if we fail to erase label on detached vdev), use it only as last resort. Quick way to reproduce this scenario now is detach vdev from a pool with enabled autoextend. During vdev detach autoextend logic tries to reopen remaining vdev, that always fails now since in-memory configuration is already updated, while on-disk labels are not yet.
* MFC r308049: Improve few debugging log messages.mav2016-11-121-3/+3
|
* MFC r307318: MFV r307314:mav2016-11-121-20/+86
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 6988 spa_sync() spends half its time in dmu_objset_do_userquota_updates Using a benchmark which creates 2 million files in one TXG, I observe that the thread running spa_sync() is on CPU almost the entire time we are syncing, and therefore can be a performance bottleneck. About 50% of the time in spa_sync() is in dmu_objset_do_userquota_updates(). The problem is that dmu_objset_do_userquota_updates() calls zap_increment_int(DMU_USERUSED_OBJECT) once for every file that was modified (or created). In this benchmark, all the files are owned by the same user/group, so all 2 million calls to zap_increment_int() are modifying the same entry in the zap. The same issue exists for the DMU_GROUPUSED_OBJECT. We should keep an in-memory map from user to space delta while we are syncing, and when we finish, iterate over the in-memory map and modify the ZAP once per entry. This reduces the number of calls to zap_increment_int() from "number of objects modified" to "number of owners/groups of modified files". This reduced the time spent in spa_sync() in the file create benchmark by ~33%, from 11 seconds to 7 seconds. Closes #107 Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Steve Gonczi <steve.gonczi@delphix.com> Reviewed by: Ned Bass <bass6@llnl.gov> Reviewed by: Jinshan Xiong <jinshan.xiong@intel.com> Author: Matthew Ahrens <mahrens@delphix.com> openzfs/openzfs@5fc46359c569369d87728ca09f8705cdff6cc8e2
* MFC r308437 and r308461:hselasky2016-11-121-2/+16
| | | | | | | | | | | Range check the jitter values to avoid bogus sample rate adjustments. The expected deviation should not be more than 1Hz per second. The USB v2.0 specification also mandates this requirement. Refer to chapter 5.12.4.2 about feedback. Allow higher sample rates to have more jitter than lower ones. PR: 208791
* MFC r308025:kib2016-11-112-3/+9
| | | | Enable vn_io_fault() deadlock avoidance for msdosfs.
* MFC r308024:kib2016-11-111-4/+26
| | | | | Ensure that cluster allocations never allocate clusters outside the volume limits.
* MFC r308023:kib2016-11-111-1/+4
| | | | | If the fatchain() call in chainalloc() returned an error, revert marking the cluster run as in-use.
* MFC r308022:kib2016-11-111-5/+7
| | | | Use symbolic name for the value of fully free word in pm_inusemap.
* MFC r308021:kib2016-11-111-1/+1
| | | | Use symbolic name for the free cluster number.
* MFC r308020:kib2016-11-111-6/+8
| | | | Fix comment formatting.
* MFC r308019:kib2016-11-111-2/+1
| | | | Remove useless NULL check.
* MFC 308201sephe2016-11-114-69/+71
| | | | | | | | | | | hyperv/kvp: Don't mix message status codes and function return values. While I'm here, move message status codes to hv_utilreg.h, since they will be used by the upcoming VSS stuffs. Submitted by: Hongjiang Zhang <honzhan microsoft com> Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8391
* MFC 308194sephe2016-11-113-72/+0
| | | | | | | hyperv: GC unused functions. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8406
* MFC 308168sephe2016-11-111-11/+15
| | | | | | | hyperv/vmbus: Avoid extra header copy. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8395
* MFC 308166,308167sephe2016-11-111-50/+132
| | | | | | | | | | | | | | | | 308166 hyperv/hn: Move TSO packet fixup to an earlier place for if_transmit. While TSO packet header may be still cache-hot. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8393 308167 hyperv/hn: Directly fill chimney sending buffer for small packets. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8394
* MFC 308164sephe2016-11-111-125/+153
| | | | | | | | | | hyperv/hn: Regroup if_start related functions. And put them under HN_IFSTART_SUPPORT, which is by default on until we whack the if_start related bits from base system. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8392
* MFC 308163sephe2016-11-114-3/+3
| | | | | | | hyperv/hn: Rename cleaned up file. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8390
* MFC 308162sephe2016-11-111-297/+271
| | | | | | | hyperv/hn: Cosmetic cleanup; no functional changes. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8389
* MFC 308117-308120sephe2016-11-112-78/+114
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 308117 hyperv/hn: Rework temporary channel packet buffer expanding. And use large default temporary channel packer buffer; we really don't want it to be expanded at run time. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8367 308118 hyperv/hn: Cleanup RXBUF ack processing. - Increase the # of retries. - Add comment. - Log error, if RXBUF ack fails. - Add stat for RXBUF ack failures. RXBUF ack really should _not_ fail... Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8368 308119 hyperv/hn: Reset do_lro, if the hash types are not TCP related. Mainly because the host side only set TCPCS and IPCS even for UDP datagrams. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8369 308120 hyperv/hn: Don't start shared TX taskq, if the hypervisor is not Hyper-V. - Move the SYSINIT to DRIVER/SECOND, i.e. after the vm_guest becomes determistic. - Minor style changes. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8370
* MFC 308018,308116sephe2016-11-116-7/+5
| | | | | | | | | | | | | | 308018 hyeprv/hn: Rename cleaned up RNDIS header file. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8360 308116 hyperv/hn: Rename cleaned up RNDIS source file. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8361
* hyperv/hn: Fix i386 build; if_baudrate is 32bits on i386 on stable/10sephe2016-11-111-0/+5
| | | | | | This is a direct commit. Sponsored by: Microsoft
* MFC 308013-308017sephe2016-11-114-461/+437
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 308013 hyperv/hn: Nuke unnecessary indirection. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8355 308014 hyperv/hn: Reorganize RX path; mainly pull non-control code path up Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8356 308015 hyperv/hn: Pull data path code up. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8357 308016 hyperv/hn: Cleanup RNDIS related files. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8358 308017 hyperv/hn: Change header guardian; in preparation for the upcoming rename. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8359
* MFC 308011,308012sephe2016-11-117-6/+6
| | | | | | | | | | | | | | 308011 hyperv/hn: Rename cleaned up NVS header file. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8353 308012 hyperv/hn: Rename cleaned up NVS source file. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8354
* MFC 307989-307991,308010sephe2016-11-115-397/+360
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 307989 hyperv/hn: Move hn_softc to if_hnvar.h While I'm here, use consistent macro names. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8345 307990 hyperv/hn: Move send context to NVS domain. Since all sends are encapsulated in NVS messages. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8346 307991 hyperv/hn: NVS inclusion cleanup and forward declare functions. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8347 308010 hyperv/hn: Change header guardian; in preparation for the upcoming rename. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8352
* MFC 307985-307988sephe2016-11-115-85/+74
| | | | | | | | | | | | | | | | | | | | | | | | | | | | 307985 hyperv/hn: Nuke unnecessary M_NETVSC Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8340 307986 hyperv/hn: Move %b format string for capabilities near their definition. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8341 307987 hyperv/hn: Define empty packet filter. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8342 307988 hyperv/hn: Shuffle chimney sending buffer alloc/free around. This paves way for more chimney sending buffer reorganization. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8343
* MFC 307983sephe2016-11-114-16/+113
| | | | | | | hyperv/hn: Properly configure RSS according to RSS capabilities Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8338
* MFC 307952,307953,308278sephe2016-11-112-58/+81
| | | | | | | | | | | | | | | | | | | | | | | | | | 307952 hyperv/vmbus: Add missing white space. Submitted by: QianYue You <t-youqi microsoft com> Sponsored by: Microsoft 307953 hyperv/vmbus: Implement vmbus_chan_printf. And use it for vmbus channel logging, which can log the channel owner's name properly, instead of vmbus0. Submitted by: QianYue You <t-youqi microsoft com> Sponsored by: Microsoft 308278 hyperv/vmbus: Reset ch_dev, once the child is deleted. So it will not be mis-used later on, e.g. in vmbus_chan_printf(). Submitted by: dexuan Reported by: dexuan Sponsored by: Microsoft
* MFC 307893sephe2016-11-111-0/+1
| | | | | | | | | hyperv/hn: Set baudrate properly PR: 208931 Submitted by: Eugene Grosbein <ports grosbein net> Reported by: Eugene Grosbein <ports grosbein net> Sponsored by: Microsoft
* MFC 307845sephe2016-11-119-88/+207
| | | | | | | | | hyperv/ic: Rework framework/message version negotiation. Submitted by: Hongjiang Zhang <honzhan microsoft com> Modified by: sephe Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8280
* MFC 307844sephe2016-11-111-5/+0
| | | | | | | hyperv/hn: Nuke unused forward declaration. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8314
* MFC 307843sephe2016-11-113-31/+94
| | | | | | | hyperv/hn: Fix RX filter settings. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8313
* MFC 307840,307842sephe2016-11-111-20/+31
| | | | | | | | | | | | | | | | 307840 hyperv/hn: Properly handle synthetic parts reattach failure. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8310 307842 hyperv/hn: Start link status check, if no network changes were pending. Link status check is much more lightweight than network change detection. Sponsored by: Microsoft Differential Revision: https://reviews.freebsd.org/D8311
OpenPOWER on IntegriCloud