Neal Cardwell [Mon, 25 Jan 2016 22:01:53 +0000 (14:01 -0800)]
tcp: fix tcp_mark_head_lost to check skb len before fragmenting
This commit fixes a corner case in tcp_mark_head_lost() which was
causing the WARN_ON(len > skb->len) in tcp_fragment() to fire.
tcp_mark_head_lost() was assuming that if a packet has
tcp_skb_pcount(skb) of N, then it's safe to fragment off a prefix of
M*mss bytes, for any M < N. But with the tricky way TCP pcounts are
maintained, this is not always true.
For example, suppose the sender sends 4 1-byte packets and have the
last 3 packet sacked. It will merge the last 3 packets in the write
queue into an skb with pcount = 3 and len = 3 bytes. If another
recovery happens after a sack reneging event, tcp_mark_head_lost()
may attempt to split the skb assuming it has more than 2*MSS bytes.
This sounds very counterintuitive, but as the commit description for
the related commit c0638c247f55 ("tcp: don't fragment SACKed skbs in
tcp_mark_head_lost()") notes, this is because tcp_shifted_skb()
coalesces adjacent regions of SACKed skbs, and when doing this it
preserves the sum of their packet counts in order to reflect the
real-world dynamics on the wire. The c0638c247f55 commit tried to
avoid problems by not fragmenting SACKed skbs, since SACKed skbs are
where the non-proportionality between pcount and skb->len/mss is known
to be possible. However, that commit did not handle the case where
during a reneging event one of these weird SACKed skbs becomes an
un-SACKed skb, which tcp_mark_head_lost() can then try to fragment.
The fix is to simply mark the entire skb lost when this happens.
This makes the recovery slightly more aggressive in such corner
cases before we detect reordering. But once we detect reordering
this code path is by-passed because FACK is disabled.
Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit d88270eef4b56bd7973841dd1fed387ccfa83709)
Orabug: 26646104
Conflicts:
tcp_skb_mss is not used in UEK4. Hence, skb_shinfo()
is used to get the mss size.
Signed-off-by: Ashok Vairavan <ashok.vairavan@oracle.com> Reviewed-by: Jack Vogel <jack.vogel@oracle.com>
Jim Mattson [Tue, 12 Sep 2017 20:02:54 +0000 (13:02 -0700)]
kvm: nVMX: Don't allow L2 to access the hardware CR8
If L1 does not specify the "use TPR shadow" VM-execution control in
vmcs12, then L0 must specify the "CR8-load exiting" and "CR8-store
exiting" VM-execution controls in vmcs02. Failure to do so will give
the L2 VM unrestricted read/write access to the hardware CR8.
This fixes CVE-2017-12154.
Signed-off-by: Jim Mattson <jmattson@google.com> Reviewed-by: David Hildenbrand <david@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(cherry picked from commit 51aa68e7d57e3217192d88ce90fd5b8ef29ec94f)
OraBug: 26868769 CVE-2017-12154 kvm: nVMX: L2 guest could access hardware(L0) CR8 register Tested-by: Krish Sadhukhan <krish.sadhukhan@oracle.com> Signed-off-by: Krish Sadhukhan <krish.sadhukhan@oracle.com> Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
The SDT stub function is used during the kernel boot process (prior to
the patching of SDT probe points). Since it is used for both regular
SDT probes and is-enabled SDT probes, it should return 0 to be a no-op
before call patching takes place.
Orabug: 26909775 Signed-off-by: Kris Van Hees <kris.van.hees@oracle.com> Reviewed-by: Nick Alcock <nick.alcock@oracle.com>
Wei Wang [Thu, 18 May 2017 18:22:33 +0000 (11:22 -0700)]
tcp: initialize rcv_mss to TCP_MIN_MSS instead of 0
When tcp_disconnect() is called, inet_csk_delack_init() sets
icsk->icsk_ack.rcv_mss to 0.
This could potentially cause tcp_recvmsg() => tcp_cleanup_rbuf() =>
__tcp_select_window() call path to have division by 0 issue.
So this patch initializes rcv_mss to TCP_MIN_MSS instead of 0.
Reported-by: Andrey Konovalov <andreyknvl@google.com> Signed-off-by: Wei Wang <weiwan@google.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 499350a5a6e7512d9ed369ed63a4244b6536f4f8)
Sabrina Dubroca [Wed, 3 May 2017 14:43:19 +0000 (16:43 +0200)]
xfrm: fix stack access out of bounds with CONFIG_XFRM_SUB_POLICY
When CONFIG_XFRM_SUB_POLICY=y, xfrm_dst stores a copy of the flowi for
that dst. Unfortunately, the code that allocates and fills this copy
doesn't care about what type of flowi (flowi, flowi4, flowi6) gets
passed. In multiple code paths (from raw_sendmsg, from TCP when
replying to a FIN, in vxlan, geneve, and gre), the flowi that gets
passed to xfrm is actually an on-stack flowi4, so we end up reading
stuff from the stack past the end of the flowi4 struct.
Since xfrm_dst->origin isn't used anywhere following commit ca116922afa8 ("xfrm: Eliminate "fl" and "pol" args to
xfrm_bundle_ok()."), just get rid of it. xfrm_dst->partner isn't used
either, so get rid of that too.
Fixes: 9d6ec938019c ("ipv4: Use flowi4 in public route lookup interfaces.") Signed-off-by: Sabrina Dubroca <sd@queasysnail.net> Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
(cherry picked from commit 9b3eb54106cf6acd03f07cf0ab01c13676a226c2)
David Howells [Wed, 14 Jun 2017 23:12:24 +0000 (00:12 +0100)]
rxrpc: Fix several cases where a padded len isn't checked in ticket decode
This fixes CVE-2017-7482.
When a kerberos 5 ticket is being decoded so that it can be loaded into an
rxrpc-type key, there are several places in which the length of a
variable-length field is checked to make sure that it's not going to
overrun the available data - but the data is padded to the nearest
four-byte boundary and the code doesn't check for this extra. This could
lead to the size-remaining variable wrapping and the data pointer going
over the end of the buffer.
Fix this by making the various variable-length data checks use the padded
length.
Reported-by: 石磊 <shilei-c@360.cn> Signed-off-by: David Howells <dhowells@redhat.com> Reviewed-by: Marc Dionne <marc.c.dionne@auristor.com> Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(backported from commit 5f2f97656ada8d811d3c1bef503ced266fcd53a0)
Juergen Gross [Tue, 30 May 2017 18:52:26 +0000 (20:52 +0200)]
xen: don't print error message in case of missing Xenstore entry
When registering for the Xenstore watch of the node control/sysrq the
handler will be called at once. Don't issue an error message if the
Xenstore node isn't there, as it will be created only when an event
is being triggered.
(cherry picked from commit 4e93b6481c87ea5afde944a32b4908357ec58992) Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Reviewed-by: Shannon Nelson <shannon.nelson@oracle.com>
mlx4_core: calculate log_num_mtt based on total system memory
The SR-IOV shared-port mechanism has a limitation that all the resources
and qp contexts are proxied through the PF. In order to reflect the
supported mtt entries, the log_num_mtt must be calculated based on the host
system memory rather than the privileged domain system memory. Thus, this
patch performs a Xen specific call to obtain the total memory during the PF
driver loading and uses that info to determine the size of the mtt table.
Boris Ostrovsky [Fri, 15 Sep 2017 20:23:53 +0000 (16:23 -0400)]
xen/x86: Add interface for querying amount of host memory
A driver (or some other entity in the kernel) may need to know
amount of memory available on the host. Provide the interface (for
a privileged domain() to obtain this information.
rds: Fix non-atomic operation on shared flag variable
The bits in m_flags in struct rds_message are used for a plurality of
reasons, and from different contexts. To avoid any missing updates to
m_flags, use the atomic set_bit() instead of the non-atomic equivalent.
Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com> Reviewed-by: Knut Omang <knut.omang@oracle.com> Reviewed-by: Wei Lin Guay <wei.lin.guay@oracle.com> Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry-picked from upstream f530f39f5ff97209cc6f1bf66e634685954ad741)
In rds_send_xmit() there is logic to batch the sends. However, if
another thread has acquired the lock and has incremented the send_gen,
it is considered a race and we yield. The code incrementing the
s_send_lock_queue_raced statistics counter did not count this event
correctly.
This commit counts the race condition correctly.
Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com> Reviewed-by: Knut Omang <knut.omang@oracle.com> Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry-picked from upstream 126f760ca94dae77425695f9f9238b731de86e32)
Jacob Keller [Wed, 12 Jul 2017 09:46:05 +0000 (05:46 -0400)]
i40e: use cpumask_copy instead of direct assignment
According to the header file cpumask.h, we shouldn't be directly copying
a cpumask_t, since its a bitmap and might not be copied correctly. Lets
use the provided cpumask_copy() function instead.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Tested-by: Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Orabug: 26822609
(cherry picked from commit 7e4d01e7d3f7d4f7b0a768a1028cb26ea06c8694) Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Tested-by: Dib Chatterjee <dib.chatterjee@oracle.com> Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
THP defrag is enabled by default to direct reclaim/compact but not wake
kswapd in the event of a THP allocation failure. The problem is that
THP allocation requests potentially enter reclaim/compaction. This
potentially incurs a severe stall that is not guaranteed to be offset by
reduced TLB misses. While there has been considerable effort to reduce
the impact of reclaim/compaction, it is still a high cost and workloads
that should fit in memory fail to do so. Specifically, a simple
anon/file streaming workload will enter direct reclaim on NUMA at least
even though the working set size is 80% of RAM. It's been years and
it's time to throw in the towel.
First, this patch defines THP defrag as follows;
madvise: A failed allocation will direct reclaim/compact if the application requests it
never: Neither reclaim/compact nor wake kswapd
defer: A failed allocation will wake kswapd/kcompactd
always: A failed allocation will direct reclaim/compact (historical behaviour)
khugepaged defrag will enter direct/reclaim but not wake kswapd.
Next it sets the default defrag option to be "madvise" to only enter
direct reclaim/compaction for applications that specifically requested
it.
Lastly, it removes a check from the page allocator slowpath that is
related to __GFP_THISNODE to allow "defer" to work. The callers that
really cares are slub/slab and they are updated accordingly. The slab
one may be surprising because it also corrects a comment as kswapd was
never woken up by that path.
This means that a THP fault will no longer stall for most applications
by default and the ideal for most users that get THP if they are
immediately available. There are still options for users that prefer a
stall at startup of a new application by either restoring historical
behaviour with "always" or pick a half-way point with "defer" where
kswapd does some of the work in the background and wakes kcompactd if
necessary. THP defrag for khugepaged remains enabled and will enter
direct/reclaim but no wakeup kswapd or kcompactd.
After this patch a THP allocation failure will quickly fallback and rely
on khugepaged to recover the situation at some time in the future. In
some cases, this will reduce THP usage but the benefit of THP is hard to
measure and not a universal win where as a stall to reclaim/compaction
is definitely measurable and can be painful.
The first test for this is using "usemem" to read a large file and write
a large anonymous mapping (to avoid the zero page) multiple times. The
total size of the mappings is 80% of RAM and the benchmark simply
measures how long it takes to complete. It uses multiple threads to see
if that is a factor. On UMA, the performance is almost identical so is
not reported but on NUMA, we see this
For a single thread, the benchmark completes 43.23% faster with this
patch applied with smaller benefits as the thread increases. Similar,
notice the large reduction in most cases in system CPU usage. The
overall CPU time is
4.4.0 4.4.0
kcompactd-v1r1 nodefrag-v1r3
User 10357.65 10438.33
System 3988.88 3543.94
Elapsed 2203.01 1634.41
This patch eliminates almost all swapping and direct reclaim activity.
There is still overhead but it's from NUMA balancing which does not
identify that it's pointless trying to do anything with this workload.
I also tried the thpscale benchmark which forces a corner case where
compaction can be used heavily and measures the latency of whether base
or huge pages were used
The average time to fault pages is substantially reduced in the majority
of caseds but with the obvious caveat that fewer THPs are actually used
in this adverse workload
Note again that while this does swap as it's an aggressive workload, the
direct relcim activity and allocation stalls is substantially reduced.
There is some kswapd activity but ftrace showed that the kswapd activity
was due to normal wakeups from 4K pages being allocated.
Compaction-related stalls and activity are almost eliminated.
I also tried the stutter benchmark. For this, I do not have figures for
NUMA but it's something that does impact UMA so I'll report what is
available
This benchmark is trying to fault an anonymous mapping while there is a
heavy IO load -- a scenario that desktop users used to complain about
frequently. This shows a mix because the ideal case of mapping with THP
is not hit as often. However, note that 99% of the mappings complete
13.79% faster. The CPU usage here is particularly interesting
4.4.0 4.4.0
kcompactd-v1r1nodefrag-v1r3
User 67.50 0.99
System 1327.88 91.30
Elapsed 2079.00 2128.98
Allocation stalls and all direct reclaim activity is eliminated as well
as compaction-related stalls.
THP gives impressive gains in some cases but only if they are quickly
available. We're not going to reach the point where they are completely
free so lets take the costs out of the fast paths finally and defer the
cost to kswapd, kcompactd and khugepaged where it belongs.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Rik van Riel <riel@redhat.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Vlastimil Babka <vbabka@suse.cz> Cc: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Nitin Gupta <nitin.m.gupta@oracle.com> Reviewed-by: Bob Picco <bob.picco@oracle.com> Reviewed-by: Vijay Kumar <vijay.ac.kumar@oracle.com>
This problem is due to the test aes_gcm_enc/dec test templates have actual IV
size of 13 bytes, but alg copies 16 bytes which leads to out of bound access.
The fix is to initialize the iv member to MAX_IV_SIZE.
Fixes: b824b1aa827f ("crypto: testmgr - fix out of bound read in __test_aead()") Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com> Reviewed-by: John Haxby <john.haxby@oracle.com>
Nick Alcock [Wed, 6 Sep 2017 10:45:51 +0000 (11:45 +0100)]
SPEC: generate CTF when DTrace is enabled.
CTF is not yet generated for debug kernels, but this is purely because
the ctf target is unavailable because CONFIG_CTF is disabled in
debug kernels, despite with_dtrace being set. If and when CONFIG_DTRACE
(and thus CONFIG_CTF) are enabled in debug kernels, we can turn on CTF
building there without incident.
(Note: non-RPM builds are now much faster than before, since they don't
generate CTF unless you ask it to, but we cannot really avoid generating
CTF for RPM builds, since DTrace needs it. Future commits will speed up
CTF generation significantly, but for now we have to take the hit, just
as we have been before now.)
Signed-off-by: Nick Alcock <nick.alcock@oracle.com> Reviewed-by: Tomas Jedlicka <tomas.jedlicka@oracle.com> Reviewed-by: Victor Erminpour <victor.erminpour@oracle.com>
Orabug: 25815362
Nick Alcock [Tue, 5 Sep 2017 21:39:34 +0000 (22:39 +0100)]
SPEC: bump libdtrace-ctf requirement to 0.7+.
This version includes the CTF archive support needed to build
CTF into an archive rather than linking in into modules.
It is backwardly binary-, source-, and CTF-format-compatible with
current releases (0.5, 0.6).
Signed-off-by: Nick Alcock <nick.alcock@oracle.com> Reviewed-by: Tomas Jedlicka <tomas.jedlicka@oracle.com> Reviewed-by: Victor Erminpour <victor.erminpour@oracle.com>
Orabug: 25815362
Documentation: add watermark_scale_factor to the list of vm systcl file
Commit 795ae7a0de6b ("mm: scale kswapd watermarks in proportion to
memory") properly added the description of the new knob to
Documentation/sysctl/vm.txt, but forgot to add it to the list of files
in /proc/sys/vm. Let's fix that.
Signed-off-by: Jerome Marchand <jmarchan@redhat.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Jonathan Corbet <corbet@lwn.net>
(cherry picked from commit e6507a00fd08986ce003012a10af78cc7e47eee8)
Signed-off-by: Robert M. Harris <robert.m.harris@oracle.com> Reviewed-by: Khalid Aziz <khalid.aziz@oracle.com> Reviewed-by: Larry Bassel <larry.bassel@oracle.com> Reviewed-by: Herbert van den Bergh <herbert.van.den.bergh@oracle.com> Reviewed-by: Todd Vierling <todd.vierling@oracle.com>
Johannes Weiner [Thu, 17 Mar 2016 21:19:14 +0000 (14:19 -0700)]
mm: scale kswapd watermarks in proportion to memory
In machines with 140G of memory and enterprise flash storage, we have
seen read and write bursts routinely exceed the kswapd watermarks and
cause thundering herds in direct reclaim. Unfortunately, the only way
to tune kswapd aggressiveness is through adjusting min_free_kbytes - the
system's emergency reserves - which is entirely unrelated to the
system's latency requirements. In order to get kswapd to maintain a
250M buffer of free memory, the emergency reserves need to be set to 1G.
That is a lot of memory wasted for no good reason.
On the other hand, it's reasonable to assume that allocation bursts and
overall allocation concurrency scale with memory capacity, so it makes
sense to make kswapd aggressiveness a function of that as well.
Change the kswapd watermark scale factor from the currently fixed 25% of
the tunable emergency reserve to a tunable 0.1% of memory.
Beyond 1G of memory, this will produce bigger watermark steps than the
current formula in default settings. Ensure that the new formula never
chooses steps smaller than that, i.e. 25% of the emergency reserve.
On a 140G machine, this raises the default watermark steps - the
distance between min and low, and low and high - from 16M to 143M.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Acked-by: Mel Gorman <mgorman@suse.de> Acked-by: Rik van Riel <riel@redhat.com> Acked-by: David Rientjes <rientjes@google.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit 795ae7a0de6b834a0cc202aa55c190ef81496665)
Signed-off-by: Robert M. Harris <robert.m.harris@oracle.com> Reviewed-by: Khalid Aziz <khalid.aziz@oracle.com> Reviewed-by: Larry Bassel <larry.bassel@oracle.com> Reviewed-by: Herbert van den Bergh <herbert.van.den.bergh@oracle.com> Reviewed-by: Todd Vierling <todd.vierling@oracle.com>
Nick Alcock [Wed, 6 Sep 2017 10:45:51 +0000 (11:45 +0100)]
ctf: automate away the deduplication blacklist
The deduplication blacklist in scripts/dwarf2ctf/dedup.blacklist is a
great bit kludge. It contains a list of modules that cannot be
deduplicated because they contain structures which are defined in the
same location in different ways different kernel modules (usually
because the structure is modified by preprocessor conditionals). But
augmenting the blacklist is a pig, involving lots of poring over
debugging output to find the structure to focus on.
So automate the problem away, by augmenting type IDs for structures with
the sizeof() the structure in a new component (separated from the others
by //, a component invalid in POSIX pathnames, as usual). Helpfully
this is made available to us in the DW_AT_byte_size attribute, so it's
fast to obtain. (The component is optional because opaque structure
declarations obviously cannot include it.)
We adjust the one place that transforms transparent structure IDs into
opaque ones to take this tag into account.
This will still break for structures that are modified by preprocessor
conditionals in such a way that one member is replaced by another with a
different type but which has the same size as the one it replaces
(perhaps one pointer to a structure being replaced by a pointer to a
different structure), but in the interests of dwarf2ctf performance I'm
avoiding solving this for now, since we are not hitting it, and solving
it would require annotating structure IDs with some sort of hash of
their member names: the overhead of recursing over all members every
time we get an ID for a structure seems likely to be quite high, given
how often we look up type_id()s.
This change has no detectable effect on dwarf2ctf runtime, and shrinks
the CTF output by about 40KiB.
Signed-off-by: Nick Alcock <nick.alcock@oracle.com> Reviewed-by: Tomas Jedlicka <tomas.jedlicka@oracle.com>
Orabug: 26765112
Nick Alcock [Tue, 5 Sep 2017 21:25:34 +0000 (22:25 +0100)]
ctf: drop CONFIG_DT_DISABLE_CTF, ctf.ko, and all that it implies
Now that CTF is decoupled from the kernel build and built into a
separate archive, there is no longer any need to drag around a
fake ctf.ko module to contain the shared and built-in CTF info.
Drop it, and kernel/ctf/, and the code to autoload it when
dtrace.ko is loaded, and move its Kconfig contents into
lib/Kconfig (which used to include kernel/ctf/Kconfig).
Furthermore, now that CTF is built on demand and not unconditionally
built every time the kernel is, there is no longer any need for
the speedup hack CONFIG_DT_DISABLE_CTF. Drop it.
Signed-off-by: Nick Alcock <nick.alcock@oracle.com> Reviewed-by: Tomas Jedlicka <tomas.jedlicka@oracle.com> Reviewed-by: Victor Erminpour <victor.erminpour@oracle.com>
Orabug: 25815362
Nick Alcock [Wed, 19 Jul 2017 14:44:05 +0000 (15:44 +0100)]
ctf: do not allow dwarf2ctf to run as root
This is just insanely dangerous: with the addition of the CTF_DEBUGDIR
info it reads almost arbitrary DWARF. elfutils is not root-rated and
frankly neither is dwarf2ctf, valgrind or no valgrind. It's just too
complicated to risk that way.
Signed-off-by: Nick Alcock <nick.alcock@oracle.com> Reviewed-by: Tomas Jedlicka <tomas.jedlicka@oracle.com>
Orabug: 25815362
Nick Alcock [Wed, 19 Jul 2017 14:34:14 +0000 (15:34 +0100)]
ctf: decouple CTF building from the kernel build
This change causes CTF types for the core kernel and modules to be
generated only when the new 'ctf' make target is invoked. The CTF content
is emitted into a CTF archive with the default name of vmlinux.ctfa (the
name read by DTrace userspace): this can be changed via the CTF_FILENAME
makefile variable. If 'make ctf' has been run, 'make modules_install'
will install the generated CTF archive into the appropriate place. (If
CTF_FILENAME was specified on the 'make ctf' line, it needs to be passed
to 'make modules_install' as well for this to work.)
The existing link-into-modules machinery is still used for out-of-tree
modules, since these obviously cannot be visible when the vmlinux.ctfa
is built.
Usually the ctf target is invoked by kernel-uek.spec, but it can also
be invoked by developers if they know they have changed type or global
variable info while developing and would like DTrace to be able to
introspect the new data, or if they are building a kernel for the
first time and would like DTrace to be able to see its types at all.
(The archive format is fairly robust: you can often just copy
vmlinux.ctfa from one kernel to another, and types that have not
changed will continue to work with the new kernel.)
This depends on new machinery in libdtrace-ctf 0.7 or higher.
Signed-off-by: Nick Alcock <nick.alcock@oracle.com> Reviewed-by: Tomas Jedlicka <tomas.jedlicka@oracle.com> Reviewed-by: Victor Erminpour <victor.erminpour@oracle.com>
Orabug: 25815362
Martin K. Petersen [Fri, 11 Aug 2017 04:19:43 +0000 (00:19 -0400)]
oracleasm: Copy the integrity descriptor
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Divya Indi <divya.indi@oracle.com>
The original code made assumptions about the oracleasm integrity
descriptor hanging off of check_asm_ioc being mapped. Make sure we
properly copy and validate the descriptor before use.
RDS: IB: Add proxy qp to support FRWR through RDS_GET_MR
MR registration requested through RDS_GET_MR socket option will not have
any connection details. So, there isn't an appropriate qp to post the
registration/invalidation requests. This patch solves that issue by
using a proxy qp.
Avinash Repaka [Thu, 17 Aug 2017 21:02:47 +0000 (14:02 -0700)]
RDS: Add support for fast registration work request
This patch adds support for MR registration through work request in RDS,
commonly referred as FRWR/fastreg/FRMR.
With this patch added, RDS chooses the registration method, between FMR
and FRWR, based on the preference given through 'prefer_frwr' module
parameter and the support offered by the underlying device.
Please note that this patch is adding support for MR registration done
only through CMSG. Support for registrations through RDS_GET_MR socket
option will be added through another patch.
[qed_sp_iscsi_func_start:189(host_7-0)]Cannot satisfy CQ amount. Queues
requested 8, CQs available 4. Aborting function start
Above condition will resolve as management firmware is capable of
telling us the number of CQs available for a given PF, qed will
communicate the same number to qedi, So that qedi will know how much CQs
are allowed.
Signed-off-by: Manish Rangankar <manish.rangankar@cavium.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
Trivial fix to spelling mistake in QEDF_ERR message. I should have also
included this in a previous fix, but I only just spotted this one.
Signed-off-by: Colin Ian King <colin.king@canonical.com> Acked-by: Manish Rangankar <Manish.Rangankar@cavium.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
qedi uses iscsi_boot_sysfs to export the targets used for boot to
sysfs. Select the config option to make sure the module is built.
This addresses the compile time issue,
drivers/scsi/qedi/qedi_main.o: In function `qedi_remove':
qedi_main.c:(.text+0x3bbd): undefined reference to `iscsi_boot_destroy_kset'
drivers/scsi/qedi/qedi_main.o: In function `__qedi_probe.constprop.0':
qedi_main.c:(.text+0x577a): undefined reference to `iscsi_boot_create_target'
qedi_main.c:(.text+0x5807): undefined reference to `iscsi_boot_create_target'
qedi_main.c:(.text+0x587f): undefined reference to `iscsi_boot_create_initiator'
qedi_main.c:(.text+0x58f3): undefined reference to `iscsi_boot_create_ethernet'
qedi_main.c:(.text+0x5927): undefined reference to `iscsi_boot_destroy_kset'
qedi_main.c:(.text+0x5d7b): undefined reference to `iscsi_boot_create_host_kset'
[mkp: fixed whitespace]
Signed-off-by: Nilesh Javali <nilesh.javali@cavium.com> Fixes: c57ec8fb7c02 ("scsi: qedi: Add support for Boot from SAN over iSCSI offload") Reported-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
This patch adds support for Boot from SAN over iSCSI offload. The iSCSI
boot information in the NVRAM is populated under
/sys/firmware/iscsi_bootX/ using qed NVM-image reading API and further
exported to open-iscsi to perform iSCSI login enabling boot over offload
iSCSI interface in a Boot from SAN environment.
Signed-off-by: Arun Easi <arun.easi@cavium.com> Signed-off-by: Andrew Vasquez <andrew.vasquez@cavium.com> Signed-off-by: Manish Rangankar <manish.rangankar@cavium.com> Signed-off-by: Nilesh Javali <nilesh.javali@cavium.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
Signed-off-by: Manish Rangankar <manish.rangankar@cavium.com> Reviewed-by: Lee Duncan <lduncan@suse.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
Signed-off-by: Manish Rangankar <manish.rangankar@cavium.com> Reviewed-by: Lee Duncan <lduncan@suse.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
Variable idx is defined as u16 thus statement (idx < 0) is always false
and should be removed.
Signed-off-by: Christos Gkekas <chris.gekas@gmail.com> Acked-by: Manish Rangankar <Manish.Rangankar@cavium.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
We shouldn't be writing over the "ret" variable. It means we return
ERR_PTR(0) which is NULL and it results in a NULL dereference in the
caller.
Fixes: ace7f46ba5fd ("scsi: qedi: Add QLogic FastLinQ offload iSCSI driver framework.") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
Signed-off-by: Manish Rangankar <manish.rangankar@cavium.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
max_fin_rt is the maximum re-transmission of FIN packets
as part of the termination flow. After reaching this value
the FW will send a single RESET.
Signed-off-by: Nilesh Javali <nilesh.javali@cavium.com> Signed-off-by: Manish Rangankar <manish.rangankar@cavium.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
Signed-off-by: Manish Rangankar <manish.rangankar@cavium.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
Signed-off-by: Manish Rangankar <manish.rangankar@cavium.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
Signed-off-by: Manish Rangankar <manish.rangankar@cavium.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
Signed-off-by: Manish Rangankar <manish.rangankar@cavium.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
munmap done by iscsiuio during a stop of the service triggers a "bad
pte" warning sometimes. munmap kernel path goes through the mmapped
pages and has a validation check for mapcount (in struct page) to be
zero or above. kzalloc, which we had used to allocate udev->ctrl, uses
slab allocations, which re-uses mapcount (union) for other purposes that
can make the mapcount look negative. Avoid all these trouble by invoking
one of the __get_free_pages wrappers to be used instead of kzalloc for
udev->ctrl.
Signed-off-by: Arun Easi <arun.easi@cavium.com> Signed-off-by: Manish Rangankar <manish.rangankar@cavium.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Brian Maly <brian.maly@oracle.com>
Trivial fix to spelling mistake in DP_NOTICE message
Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
This patch adds support for adding and deleting rx flow
classification rules. Using this user can classify RX flow
constituting of TCP/UDP 4-tuples [src_ip/dst_ip and src_port/dst_port]
to be steered on a given RX queue
Signed-off-by: Manish Chopra <manish.chopra@cavium.com> Signed-off-by: Yuval Mintz <yuval.mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
The option "h" (host order ) exists for ipv4 only.
Remove the h when printing ipv6 addresses.
Lead to the following smatch warning:
drivers/net/ethernet/qlogic/qed/qed_iwarp.c:585 qed_iwarp_print_tcp_ramrod()
warn: '%pI6' can only be followed by c
drivers/net/ethernet/qlogic/qed/qed_iwarp.c:1521 qed_iwarp_print_cm_info()
warn: '%pI6' can only be followed by c
Fixes commit 456a584947d5 ("qed: iWARP CM add passive side connect")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Michal Kalderon <Michal.Kalderon@cavium.com> Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 91d1ae475b9833097e078c2581c9265d033cdbe4 ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
This patch takes care of active/passive disconnect flows.
Disconnect flows can be initiated remotely, in which case a async event
will arrive from peer and indicated to qedr driver. These
are referred to as exceptions. When a QP is destroyed, it needs to check
that it's associated ep has been closed.
Signed-off-by: Michal Kalderon <Michal.Kalderon@cavium.com> Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: Ariel Elior <Ariel.Elior@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit fc4c6065e661224df3db50780219ac53fee56e2b ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
This patch implements the active side connect.
Offload a connection, process MPA reply and send RTR.
In some of the common passive/active functions, the active side
will work in blocking mode.
Signed-off-by: Michal Kalderon <Michal.Kalderon@cavium.com> Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: Ariel Elior <Ariel.Elior@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 4b0fdd7c8b757125ac7996617d914bbdb9e0348c ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
This patch implements the passive side connect.
It addresses pre-allocating resources, creating a connection
element upon valid SYN packet received. Calling upper layer and
implementation of the accept/reject calls.
Error handling is not part of this patch.
Signed-off-by: Michal Kalderon <Michal.Kalderon@cavium.com> Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: Ariel Elior <Ariel.Elior@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 456a584947d5b92d5e5a62cc68125ab5f150aa8c ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
This patch adds the ability to add and remove listeners and identify
whether the SYN packet received is intended for iWARP or not. If
a listener is not found the SYN packet is posted back to the chip.
Signed-off-by: Michal Kalderon <Michal.Kalderon@cavium.com> Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: Ariel Elior <Ariel.Elior@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 65a91a6cdb868a28b919ca133c0f9d9dfd9a635a ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
iWARP handles incoming SYN packets using the ll2 interface. This patch
implements ll2 setup and teardown. Additional ll2 connections will
be used in the future which are not part of this patch series.
Signed-off-by: Michal Kalderon <Michal.Kalderon@cavium.com> Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: Ariel Elior <Ariel.Elior@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit b5c29ca7dab75f29a7df6e82285742f830d8ed1a ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
This patch adds iWARP support for flows that have common code
between RoCE and iWARP, such as initialization, teardown and
qp setup verbs: create, destroy, modify, query.
It introduces the iWARP specific files qed_iwarp.[ch] and
iwarp_common.h
Signed-off-by: Michal Kalderon <Michal.Kalderon@cavium.com> Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: Ariel Elior <Ariel.Elior@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 67b40dccc45ff5d488aad17114e80e00029fd854 ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
iWARP personality introduced the need for differentiating in several
places in the code whether we are RoCE, iWARP or either. This
leads to introducing new macros for querying the personality.
Signed-off-by: Michal Kalderon <Michal.Kalderon@cavium.com> Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: Ariel Elior <Ariel.Elior@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit c851a9dc4359c6b19722de568e9f543c1c23481c ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
Once we have iWARP support, the qede portion of the qedr<->qede would
serve all the RDMA protocols - so rename the file to be appropriate
to its function.
While we're at it, we're also moving a couple of inclusions to it into
.h files and adding includes to make sure it contains all type
definitions it requires.
Signed-off-by: Michal Kalderon <Michal.Kalderon@cavium.com> Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit b262a06e642cfb1eeb6c2c772f76dad674ada57e ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
The p_l2_info->pp_qid_usage[] array has "p_l2_info->queues" elements so
the > here should be a >= or we write beyond the end of the array.
Fixes: bbe3f233ec5e ("qed: Assign a unique per-queue index to queue-cid") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Acked-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 0331402aeaefe858709b0a4d44ade15f82d3a119 ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
When CONFIG_QED_SRIOV is disabled, we get a build error:
drivers/net/ethernet/qlogic/qed/qed_int.c: In function 'qed_int_sb_init':
drivers/net/ethernet/qlogic/qed/qed_int.c:1499:4: error: implicit declaration of function 'qed_vf_set_sb_info'; did you mean 'qed_mcp_get_resc_info'? [-Werror=implicit-function-declaration]
All the other declarations have a 'static inline' stub as an alternative
here, so this adds one more for qed_int_sb_init.
Fixes: 50a207147fce ("qed: Hold a single array for SBs") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Acked-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 2f3ca449a4f9a54d2bf39c873269e68ad5f34acb ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
LL2 today is interrupt driven - when tx/rx completion arrives [or any
other indication], qed needs to operate on the connection and pass
the information to the protocol-driver [or internal qed consumer].
Since we have several flavors of ll2 employeed by the driver,
each handler needs to do an if-else to determine the right functionality
to use based on the connection type.
In order to make things more scalable [given that we're going to add
additional types of ll2 flavors] move the infrastrucutre into using
a callback-based approach - the callbacks would be provided as part
of the connection's initialization parameters.
Signed-off-by: Michal Kalderon <Michal.Kalderon@cavium.com> Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 0518c12f1f79dc2f2020836974c577404e42ae89 ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
A LL2 connection [qed_ll2_info] has a sub-structure of type qed_ll2_conn
that contain various inputs for ll2 acquisition, but the connection also
utilizes a couple of other inputs.
Restructure the input structure to include all the inputs and refactor
the code necessary to populate those.
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 13c547717231aad7e1635004ae3f698e5e78d6d1 ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
First step in revising the LL2 interface, this declares
qed_ll2_tx_pkt_info as part of the ll2 interface, and uses it for
transmission instead of receiving lots of parameters.
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 7c7973b2ae277c6e89dceda2246fff2472c8ffdb ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
VFs are currently not mapping their doorbell bar, instead relying
on the small doorbell window they have in their limited regview bar.
In order to increase the number of possible Tx connections [queues]
employeed by VF past 16, we need to start using the doorbell bar if
one such is exposed - VF would communicate this fact to PF which would
return the size-bar internally configured into chip, according to
which the VF would decide whether to actually utilize the doorbell
bar.
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 1a850bfc9e71871599ddbc0d4d4cffa2dc409855 ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
Until now we used to have a single VF legacy compatibility mode,
one that affected the place of the Rx producers of those VFs [mostly].
As PF would soon support allocating CIDs for VFs instead of having
a static CID<->queue configuration for them, we'll need to have
an additional legacy mode since existing VFs would need to continue
on using the older mode of operation.
Change the infrastrucutre so that the legacy would be able to indicate
which of the legacy behaviors is needed for a given VF.
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 3b19f47820756f9905e7ef184747fbb3c8ed062f ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
When a queue-cid is allocated, assign an index inside that's
CID's queue-zone.
For PFs and VFS, this number is going to be unique and derive
from a per-queue-zone bitmap, while for PF's VFs queues the
number is currently going to constant; Later, we'd add the
capability of a VF to communicate such an index to its PF.
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit bbe3f233ec5ea99049f33471c0c0d0d2a78e2116 ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
We're going to need additional information for queue-cids
that a PF creates for its VFs, so start by refactoring existing
logic used for initializing said struct into receiving a structure
encapsulating the VF-specific information that needs to be provided.
This also introduces QED_QUEUE_CID_SELF - each queue-cid would hold
an indication to whether it belongs to the hw-function holding it
[whether that's a PF or a VF], or else what's the VF id it belongs
to.
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 3946497aff655b9bb1807ef7e2ecbe799e6d832a ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
Part of an effort of a cleaner seperation between qed and the protocol
drivers, the L2 interface is to use the SB structure for initialization
purposes opaquely.
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit f604b17d7fdef574792a7e0b39f1b926d6b43d9d ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
First step in allowing a single PF/VF to open multiple queues on
the same queue zone is to add per-hwfn database of queue-cids
as a two-dimensional array where entry would be according to
[queue zone][internal index].
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 0db711bb26209992da375730eab6b3cec1edee7a ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
Each PF has a bitmap for its own ranges of CIDs, to allow easy grabbing
of an available CID when such is needed. But VFs are not using the same
mechanism, instead relying on hard-coded CIDs [ queue-index == cid ].
As an infrastructure step toward increasing number of CIDs of VFs,
the PF is going to maintain bitmaps for the VF CIDs as well -
the bitmaps would be per-VF and the ranges would be the same [in HW all
VFs of a given PF have the same mapping of CIDs, and the HW is capable
of distinguishing between those according to the VF index]
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 6bea61da1716761c95cd32117be6004b0e14b4b2 ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
Storage drivers require images from the nvram in boot-from-SAN
scenarios. This provides the necessary API between qed and the
protocol drivers to perform such reads.
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 20675b37ee76d11430fd3d4da0851fc6a4e36abc ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
Driver reads values via HSI splitting this 8-byte into 2 32-bit
values and builds a single u64 field - but it does so by shifting
the lower field instead of the higher.
Luckily, we still don't use these fields for anything - but we're about
to start.
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 5779675912fa87d8d0af651537acc0e312f06c70 ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
The IGU CAM contains an assocaition between hardware SBs
and interrupt lines, and it can be dynamically configured
to allow more interrupts in one entity over another, specifically
for Re-distibution of SBs between a PF and its child VFs.
While we don't yet use this functionality, there are other
clients that do and as such its possible the information
passed from management firmware during initialization in
regard to the possible number of SBs doesn't accurately reflect
the current HW configuration.
The following changes are going to apply to the driver init sequence:
a. PF is going to re-configure all entries belonging to itself and
its child VFs in IGU CAM based on the management firmware info
regarding the number of SBs that are supposed to exist there.
b. PF is going to stop using the SB resource [management firmware
provided information] for anything but the initialization.
Instead, it would use the live-time counters it maintains for
the numbers.
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit ebbdcc669c7f9d8632d358a739d814485f8917dc ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
A PF today holds 2 different arrays - one holding information
about the HW configuration and one holding information about
the SBs that are used by the protocol drivers.
These arrays aren't really connected - e.g., protocol driver
initializing a given SB would not mark the same SB as occupied
in the HW shadow array.
Move into a single array [at least for PFs] - hold the mapping
of the driver-protocol SBs on the HW entry which they configure.
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 50a207147fceb64ad24c1e08e4a2a75535922e81 ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
IOV code is very intrusive in its manipulation of the status block
database.
Add a new auxiliary function to allow the PF to find an available unused
status block to configure for a specific VF's MSI-x vector.
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 09b6b14749523e3660b72be2ed91b3c0b852f58f ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
We already have an API struct that contains interrupt-related
numbers. Use it to encapsulate all information relating to the
status of SBs as (used|free).
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
[ Upstream commit 726fdbe9fa7ebccda1579716f68f8bae6fa9c87a ] Signed-off-by: Somasundaram Krishnasamy <somasundaram.krishnasamy@oracle.com>
projects / users / jedix / linux-maple.git / log