Replace long per-LUN queue of blocked commands, scanned on each command
completion and sometimes even twice, causing up to O(n^^2) processing cost,
by much shorter per-command blocked queues, scanned only when respective
command completes, and check only commands before the previous blocker,
reducing cost to O(n).
While there, unblock aborted commands to make them "complete" ASAP to be
removed from the OOA queue and so not waste time ordering other commands
against them. Aborted commands that were not sent to execution yet should
have no visible side effects, so this is safe and easy optimization now,
comparing to commands already in processing, which are a still pain.
Together those two optimizations should fix quite pathological case, when
due to backend slowness CTL accumulated many thousands of blocked requests,
partially aborted by initiator and so supposedly not even existing, but
still wasting CTL CPU time.
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
It was not only disabled for quite a while, but also appeared to be broken
at r325517, when maximum number of ports was made configurable.
MFC after: 1 week
Mainly focus on files that use BSD 2-Clause license, however the tool I
was using misidentified many licenses so this was mostly a manual - error
prone - task.
The Software Package Data Exchange (SPDX) group provides a specification
to make it easier for automated tools to detect and summarize well known
opensource licenses. We are gradually adopting the specification, noting
that the tags are considered only advisory and do not, in any way,
superceed or replace the license texts.
The biggest change is that ctl_remove_initiator() now generates I_T NEXUS
LOSS event, cleaning part of LUs state related to the initiator.
MFC after: 2 weeks
This code was disabled due to its high memory usage. But now we need this
functionality for cfumass(4) frontend, since USB MS BBB transport does not
support autosense.
MFC after: 2 weeks
It is only a first step and not perfect, but better then nothing.
The main blocker is CAM target frontend, that can not be unloaded,
since CAM does not have mechanism to unregister periph driver now.
MFC after: 2 weeks
Instead of collecting statistics for each combination of ports and logical
units, that consumed ~45KB per LU with present number of ports, collect
separate statistics for every port and every logical unit separately, that
consume only 176 bytes per each single LU/port. This reduces struct
ctl_lun size down to just 6KB.
Also new IOCTL API/ABI does not hardcode number of LUs/ports, and should
allow handling of very large quantities.
MFC after: 2 weeks (probably keeping old API enabled for some time)
This array takes 64KB of RAM now, that was more then half of struct ctl_lun
size. If at some point we support more ports, this may need another tune.
MFC after: 2 weeks
- Allow maximal sense size limitation via Control Extension mode page.
- When sense size limited, include descriptors atomically: whole or none.
- Set new SDAT_OVFL bit if some descriptors don't fit the limit.
- Report real written sense length instead of static maximal 252 bytes.
MFC after: 2 weeks
VMware tries to enable this bit to avoid multiple threshold notifications
in case of multiple initiators connected to the same LUN. Unfortunately
their code sends MODE SELECT(6) request with parameter length hardcoded
for the page without any thresholds. Since we have four threshold and our
page is bigger, this attempt fails, that is correct in my understanding.
So all we can do about this now is to report proper error code and hope
VMware fix their code one day.
MFC after: 2 weeks
While CTL still has no real events to report in this way (like SMART),
it is possible to trigger false event by manually setting TEST bit in
Informational Exceptions Control mode page, that can be useful for
initiator testing. This code supports all flavours of IE reporting:
UNIT ATTENTION, RECOVERED ERROR and NO SENSE sense keys, REQUEST SENSE
command and Informational Exceptions log page.
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
- Introduce "ha_shared" port option, which being set to "on" moves the
port into separate port group, shared between HA nodes. This allows to
better handle cases when iSCSI portals are bound to CARP address that can
dynamically move between nodes. Some initiators (at least VMware) don't
detect that after iSCSI reconnect they've attached to different SCSI port
from different port group, that totally breakes ALUA status parsing.
In theory, I believe, it should be enough to have different iSCSI portal
group tags on different nodes to make initiators detect this condition,
but it seems like VMware ignores those values, and even full LUN retaste
forced by UA does not help.
- Make CTL report up to three port groups: 1 -- non-HA mode or ports
with "ha_shared" option set, 2 -- HA node 1, 3 -- HA node 2.
- Report Transitioning state for all port groups when HA interlink is
connected, but neither of nodes is primary for the LUN.
MFC after: 2 weeks
CTL HA functionality was originally implemented by Copan many years ago,
but large part of the sources was never published. This change includes
clean room implementation of the missing code and fixes for many bugs.
This code supports dual-node HA with ALUA in four modes:
- Active/Unavailable without interlink between nodes;
- Active/Standby with second node handling only basic LUN discovery and
reservation, synchronizing with the first node through the interlink;
- Active/Active with both nodes processing commands and accessing the
backing storage, synchronizing with the first node through the interlink;
- Active/Active with second node working as proxy, transfering all
commands to the first node for execution through the interlink.
Unlike original Copan's implementation, depending on specific hardware,
this code uses simple custom TCP-based protocol for interlink. It has
no authentication, so it should never be enabled on public interfaces.
The code may still need some polishing, but generally it is functional.
Relnotes: yes
Sponsored by: iXsystems, Inc.
This is preparation for possibility to open/close media several times
per LUN life cycle. While there, rename variables to reduce confusion.
As additional bonus this allows to open read-only media, such as ZFS
snapshots.
It has nothing to share with too huge ctl.c other then device descriptor,
but even that may be counted as design error that may be fixed later.
At some point we may even want to have several ioctl ports.
Its idea was to be a simple initiator and execute several commands from
kernel level, but FreeBSD never had consumer for that functionality,
while its implementation polluted many unrelated places..
Before this change SYNCHRONIZE CACHE commands were executed exclusively,
as if they had ORDERED tag. But looking through SCSI specs I've found
no any reason to be so strict. For reads this ordering seems pointless.
For writes it looks less obvious, so I left ordering against preceeding
write commands, while following ones are no longer required to wait.
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
- remove last remnants of never implemented multiple targets support;
- implement missing support for LUN mapping in this area.
Due to existing locking constraints LUN mapping code is practically
unlocked at this point. Hopefully it is not racy enough to live until
somebody get idea how to call sleeping fronend methods under lock also
taken by the same frontend in non-sleepable context. :(
Replace iSCSI-specific LUN mapping mechanism with new one, working for any
ports. By default all ports are created without LUN mapping, exposing all
CTL LUNs as before. But, if needed, LUN mapping can be manually set on
per-port basis via ctladm. For its iSCSI ports ctld does it via ioctl(2).
The next step will be to teach ctld to work with FibreChannel ports also.
Respecting additional flexibility of the new mechanism, ctl.conf now allows
alternative syntax for LUN definition. LUNs can now be defined in global
context, and then referenced from targets by unique name, as needed. It
allows same LUN to be exposed several times via multiple targets.
While there, increase limit for LUNs per target in ctld from 256 to 1024.
Some initiators do not support LUNs above 255, but that is not our problem.
Discussed with: trasz
MFC after: 2 weeks
Relnotes: yes
Sponsored by: iXsystems, Inc.
Technically read requests can be executed in any order or simultaneously
since they are not changing any data. But ZFS prefetcher goes crasy when
it receives consecutive requests from different threads. Since prefetcher
works on level of separate blocks, instead of two consecutive 128K requests
it may receive 32 8K requests in mixed order.
This patch is more workaround then a real fix, and it does not fix all of
prefetcher problems, but it improves sequential read speed by 3-4x times
in some configurations. On the other side it may hurt performance if
some backing store has no prefetch, that is why it is disabled by default
for raw devices.
MFC after: 2 weeks
It is implemented for LUNs backed by ZVOLs in "dev" mode and files.
GEOM has no such API, so for LUNs backed by raw devices all LBAs will
be reported as mapped/unknown.
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
Abusing ability of major UAs cover minor ones we may not account UAs for
inactive ports. Allocate UAs storage for port and start accounting only
after some initiator from that port fetched its first POWER ON OCCURRED.
This reduces per-LUN CTL memory usage from >1MB to less then 100K.
MFC after: 1 month
In configurations with many ports, like iSCSI, each LUN is typically
accessed only by limited subset of ports. Allocating that memory on
demand allows to reduce CTL memory usage from 5.3MB/LUN to 1.3MB/LUN.
MFC after: 1 month
Old allocator created significant lock congestion protecting its lists
of preallocated I/Os, while UMA provides much better SMP scalability.
The downside of UMA is lack of reliable preallocation, that could guarantee
successful allocation in non-sleepable environments. But careful code
review shown, that only CAM target frontend really has that requirement.
Fix that making that frontend preallocate and statically bind CTL I/O for
every ATIO/INOT it preallocates any way. That allows to avoid allocations
in hot I/O path. Other frontends either may sleep in allocation context
or can properly handle allocation errors.
On 40-core server with 6 ZVOL-backed LUNs and 7 iSCSI client connections
this change increases peak performance from ~700K to >1M IOPS! Yay! :)
MFC after: 1 month
Sponsored by: iXsystems, Inc.
In this mode one head is in Active state, supporting all commands, while
another is in Standby state, supporting only minimal LUN discovery subset.
It is still incomplete since Standby state requires reservation support,
which is impossible to do right without having interlink between heads.
But it allows to run some basic experiments.
For ZVOL-backed LUNs this allows to inform initiators if storage's used or
available spaces get above/below the configured thresholds.
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
This includes support for:
- Read-Write Error Recovery mode page;
- Informational Exceptions Control mode page;
- Logical Block Provisioning mode page;
- LOG SENSE command.
No real Informational Exceptions features yet. This is only a placeholder.
Sponsored by: iXsystems, Inc.
