ConnectX-3 devices can use either 64- or 32-byte completion queue
entries (CQEs) and event queue entries (EQEs). Using 64-byte
EQEs/CQEs performs better because each entry is aligned to a complete
cacheline. This patch queries the HCA's capabilities, and if it
supports 64-byte CQEs and EQES the driver will configure the HW to
work in 64-byte mode.
The 32-byte vs 64-byte mode is global per HCA and not per CQ or EQ.
Since this mode is global, userspace (libmlx4) must be updated to work
with the configured CQE size, and guests using SR-IOV virtual
functions need to know both EQE and CQE size.
In case one of the 64-byte CQE/EQE capabilities is activated, the
patch makes sure that older guest drivers that use the QUERY_DEV_FUNC
command (e.g as done in mlx4_core of Linux 3.3..3.6) will notice that
they need an update to be able to work with the PPF. This is done by
changing the returned pf_context_behaviour not to be zero any more. In
case none of these capabilities is activated that value remains zero
and older guest drivers can run OK.
The SRIOV related flow is as follows
1. the PPF does the detection of the new capabilities using
QUERY_DEV_CAP command.
2. the PPF activates the new capabilities using INIT_HCA.
3. the VF detects if the PPF activated the capabilities using
QUERY_HCA, and if this is the case activates them for itself too.
Note that the VF detects that it must be aware to the new PF behaviour
using QUERY_FUNC_CAP. Steps 1 and 2 apply also for native mode.
User space notification is done through a new field introduced in
struct mlx4_ib_ucontext which holds device capabilities for which user
space must take action. This changes the binary interface so the ABI
towards libmlx4 exposed through uverbs is bumped from 3 to 4 but only
when **needed** i.e. only when the driver does use 64-byte CQEs or
future device capabilities which must be in sync by user space. This
practice allows to work with unmodified libmlx4 on older devices (e.g
A0, B0) which don't support 64-byte CQEs.
In order to keep existing systems functional when they update to a
newer kernel that contains these changes in VF and userspace ABI, a
module parameter enable_64b_cqe_eqe must be set to enable 64-byte
mode; the default is currently false.
Signed-off-by: Eli Cohen <eli@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
1. Introduce the basic SR-IOV parvirtualization context objects for
multiplexing and demultiplexing MADs.
2. Introduce support for the new proxy and tunnel QP types.
This patch introduces the objects required by the master for managing
QP paravirtualization for guests.
struct mlx4_ib_sriov is created by the master only.
It is a container for the following:
1. All the info required by the PPF to multiplex and de-multiplex MADs
(including those from the PF). (struct mlx4_ib_demux_ctx demux)
2. All the info required to manage alias GUIDs (i.e., the GUID at
index 0 that each guest perceives. In fact, this is not the GUID
which is actually at index 0, but is, in fact, the GUID which is at
index[<VF number>] in the physical table.
3. structures which are used to manage CM paravirtualization
4. structures for managing the real special QPs when running in SR-IOV
mode. The real SQPs are controlled by the PPF in this case. All
SQPs created and controlled by the ib core layer are proxy SQP.
struct mlx4_ib_demux_ctx contains the information per port needed
to manage paravirtualization:
1. All multicast paravirt info
2. All tunnel-qp paravirt info for the port.
3. GUID-table and GUID-prefix for the port
4. work queues.
struct mlx4_ib_demux_pv_ctx contains all the info for managing the
paravirtualized QPs for one slave/port.
struct mlx4_ib_demux_pv_qp contains the info need to run an individual
QP (either tunnel qp or real SQP).
Note: We made use of the 2 most significant bits in enum
mlx4_ib_qp_flags (based on enum ib_qp_create_flags in ib_verbs.h).
We need these bits in the low-level driver for internal purposes.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Enable IB ULPs to use a larger portion of the device EQs (which map to
IRQs). The mlx4_ib driver follows the mlx4_core framework of the EQs
to be divided among the device ports. In this scheme, for each IB
port, the number of allocated EQs follows the number of cores, subject
to other system constraints, such as number available MSI-X vectors.
Signed-off-by: Shlomo Pongratz <shlomop@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Signed-off-by: Shlomo Pongratz <shlomop@mellanox.com>
[ Replace one more printk_once() with pr_info_once(). - Roland ]
Signed-off-by: Roland Dreier <roland@purestorage.com>
If an erroneous CQE is polled in the first iteration (i.e. npolled ==
0), we don't update the consumer index and hence the hardware could
get a wrong notion of how many CQEs software polled. Fix this by
unconditionally updating the doorbell record. We could change the
check to be something like
if (npolled || err != -EAGAIN)
...
but it does not seem worth the effort since a posted write to memory
should not cost too much.
Signed-off-by: Eli Cohen <eli@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Use a bit in wc_flags rather then a whole integer to hold the
"checksum OK" flag. By itself, this change doesn't reduce the size of
struct ib_wc on 64bit machines -- it stays on 56 bytes because of
padding. However, it will allow to add more fields in the future
without enlarging the struct. Also, it will let us have a unified
approach with future libibverbs checksum offload reporting, because a
bit flag doesn't break the library ABI.
This patch was suggested during conversation with Liran Liss
<liranl@mellanox.com>.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
For IBoE, SLs 0-7 are mapped to Ethernet 802.1Q user priority bits
(pbits) which are part of the VLAN tag, SLs 8-15 are reserved.
Under Ethernet, the ConnectX firmware treats (decode/encode) the four
bit SL field in various constructs such as QPC / UD WQE / CQE as PPP0
and not as 0PPP. This correlates well to the fact that within the
vlan tag the pbits are located in bits 15-13 and not 12-14.
The current code wasn't consistent around that area - the
encoding was correct for the IBoE QPC.path.schedule_queue field,
but was wrong for IBoE CQEs and when MLX header was built.
These inconsistencies resulted in wrong SL <--> wire 802.1Q pbits
mapping, which is fixed by using SL <--> PPP0 all around the place.
Signed-off-by: Oren Duer <oren@mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
mlx4_ib_free_cq_buf() should not be called under spin_lock_irq() since
it calls dma_free_coherent(), which needs irqs enabled. Fix this by
deferring the free to outside the locked region.
This was found due to the
WARN_ON(irqs_disabled());
in swiotlb_free_coherent().
Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add support for masked atomic operations (masked compare and swap,
masked fetch and add).
Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Commit f780a9f1 ("mlx4_core: Add ethernet fields to CQE struct")
introduced a bug in how wc->sl is set in mlx4_ib_poll_one() -- since
cqe->sl_vid is a big-endian value, the shift must be done after
converting to host endianness.
This bug was found using sparse endianness checking.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
When resizing a CQ, when copying over unpolled CQEs from the old CQE
buffer to the new buffer, the ownership bit must be set appropriately
for the new buffer, or the ownership bit in the new buffer gets
corrupted.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
When using MSI-X mode, create a completion event queue for each CPU.
Report the number of completion EQs in a new struct mlx4_caps member,
num_comp_vectors, and extend the mlx4_cq_alloc() interface with a
vector parameter so that consumers can specify which completion EQ
should be used to report events for the CQ being created.
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
When resizing a CQ, MTTs associated with the old CQE buffer were not
freed. As a result, if any app used resize CQ repeatedly, all MTTs
were eventually exhausted, which led to all memory registration
operations failing until the driver is reloaded.
Once the RESIZE_CQ command returns successfully from FW, FW no longer
accesses the old CQ buffer, so it is safe to deallocate the MTT
entries used by the old CQ buffer.
Finally, if the RESIZE_CQ command fails, the MTTs allocated for the
new CQEs buffer also need to be de-allocated.
This fixes <https://bugs.openfabrics.org/show_bug.cgi?id=1416>.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add ethernet-related fields to struct mlx4_cqe so that the mlx4_en
ethernet NIC driver can share the same definition.
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Update existing Mellanox copyright lines to 2008, and add such lines
to files where they are missing.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add support for the following operations to mlx4 when device firmware
supports them:
- Send with invalidate and local invalidate send queue work requests;
- Allocate/free fast register MRs;
- Allocate/free fast register MR page lists;
- Fast register MR send queue work requests;
- Local DMA L_Key.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
This patch adds support for the IB "base memory management extension"
(BMME) and the equivalent iWARP operations (which the iWARP verbs
mandates all devices must implement). The new operations are:
- Allocate an ib_mr for use in fast register work requests.
- Allocate/free a physical buffer lists for use in fast register work
requests. This allows device drivers to allocate this memory as
needed for use in posting send requests (eg via dma_alloc_coherent).
- New send queue work requests:
* send with remote invalidate
* fast register memory region
* local invalidate memory region
* RDMA read with invalidate local memory region (iWARP only)
Consumer interface details:
- A new device capability flag IB_DEVICE_MEM_MGT_EXTENSIONS is added
to indicate device support for these features.
- New send work request opcodes IB_WR_FAST_REG_MR, IB_WR_LOCAL_INV,
IB_WR_RDMA_READ_WITH_INV are added.
- A new consumer API function, ib_alloc_mr() is added to allocate
fast register memory regions.
- New consumer API functions, ib_alloc_fast_reg_page_list() and
ib_free_fast_reg_page_list() are added to allocate and free
device-specific memory for fast registration page lists.
- A new consumer API function, ib_update_fast_reg_key(), is added to
allow the key portion of the R_Key and L_Key of a fast registration
MR to be updated. Consumers call this if desired before posting
a IB_WR_FAST_REG_MR work request.
Consumers can use this as follows:
- MR is allocated with ib_alloc_mr().
- Page list memory is allocated with ib_alloc_fast_reg_page_list().
- MR R_Key/L_Key "key" field is updated with ib_update_fast_reg_key().
- MR made VALID and bound to a specific page list via
ib_post_send(IB_WR_FAST_REG_MR)
- MR made INVALID via ib_post_send(IB_WR_LOCAL_INV),
ib_post_send(IB_WR_RDMA_READ_WITH_INV) or an incoming send with
invalidate operation.
- MR is deallocated with ib_dereg_mr()
- page lists dealloced via ib_free_fast_reg_page_list().
Applications can allocate a fast register MR once, and then can
repeatedly bind the MR to different physical block lists (PBLs) via
posting work requests to a send queue (SQ). For each outstanding
MR-to-PBL binding in the SQ pipe, a fast_reg_page_list needs to be
allocated (the fast_reg_page_list is owned by the low-level driver
from the consumer posting a work request until the request completes).
Thus pipelining can be achieved while still allowing device-specific
page_list processing.
The 32-bit fast register memory key/STag is composed of a 24-bit index
and an 8-bit key. The application can change the key each time it
fast registers thus allowing more control over the peer's use of the
key/STag (ie it can effectively be changed each time the rkey is
rebound to a page list).
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
When I merged bbf8eed1 ("IB/mlx4: Add support for resizing CQs") I
changed things around so that mlx4_ib_alloc_cq_buf() and
mlx4_ib_free_cq_buf() were used everywhere they could be. However, I
screwed up the number of entries passed into mlx4_ib_alloc_cq_buf()
in a couple places -- the function bumps the number of entries
internally, so the caller shouldn't add 1 as well.
Passing a too-big value for the number of entries to mlx4_ib_free_cq_buf()
can cause the cleanup to go off the end of an array and corrupt
allocator state in interesting ways.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Extend the mlx4_cq_resize() API with a way to set the "collapsed" flag
for the CQ being created.
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add a new parameter, dmasync, to the ib_umem_get() prototype. Use dmasync = 1
when mapping user-allocated CQs with ib_umem_get().
Signed-off-by: Arthur Kepner <akepner@sgi.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: Jes Sorensen <jes@sgi.com>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Roland Dreier <rdreier@cisco.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: David Miller <davem@davemloft.net>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Grant Grundler <grundler@parisc-linux.org>
Cc: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In addition to mlx4_ib, there will be ethernet and FC consumers of
mlx4_core, so move the code for managing kernel doorbells into the
core module to avoid having to duplicate this multiple times.
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
ConnectX devices support checksum generation and verification of TCP
and UDP packets for UD IPoIB messages. This patch checks if the HCA
supports this and sets the IB_DEVICE_UD_IP_CSUM capability flag if it
does. It implements support for handling the IB_SEND_IP_CSUM send
flag and setting the csum_ok field in receive work completions.
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Ali Ayub <ali@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
ConnectX HCA supports shrinking WQEs, so that a single work request
can be made of multiple units of wqe_shift. This way, WRs can differ
in size, and do not have to be a power of 2 in size, saving memory and
speeding up send WR posting. Unfortunately, if we do this then the
wqe_index field in CQEs can't be used to look up the WR ID anymore, so
our implementation does this only if selective signaling is off.
Further, on 32-bit platforms, we can't use vmap() to make the QP
buffer virtually contigious. Thus we have to use constant-sized WRs to
make sure a WR is always fully within a single page-sized chunk.
Finally, we use WRs with the NOP opcode to avoid wrapping around the
queue buffer in the middle of posting a WR, and we set the
NoErrorCompletion bit to avoid getting completions with error for NOP
WRs. However, NEC is only supported starting with firmware 2.2.232,
so we use constant-sized WRs for older firmware. And, since MLX QPs
only support SEND, we use constant-sized WRs in this case.
When stamping during NOP posting, do stamping following setting of the
NOP WQE valid bit.
Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
We use struct mlx4_buf for kernel QP, CQ and SRQ buffers, and the code
to look up an entry is duplicated in get_cqe_from_buf() and the QP and
SRQ versions of get_wqe(). Factor this out into mlx4_buf_offset().
This will also make it easier to switch over to using vmap() for buffers.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Rather than byte-swapping cqe->g_mlpath_rqpn each time we extract a
field from it, byte-swap it once into a temporary variable. This
results in smaller, better code -- eg, on 32-bit x86:
add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-5 (-5)
function old new delta
mlx4_ib_poll_cq 1188 1183 -5
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Fix the value of pkey_index in completions to get a valid value for
GSI QPs. Without this fix, incoming GSI packets on port 2 get an
invalid P_Key index in the completion, which prevents the MAD layer
from sending back a response, which can make the second port of
ConnectX HCAs completely useless.
Signed-off-by: Dotan Barak <dotanb@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Current code has a cut-and-paste error and returns IB_WC_SEND when it
should return IB_WC_RDMA_READ.
Signed-off-by: Vu Pham <vu@mellanox.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
When compacting CQ entries, we need to set the correct value of the
ownership bit in case the value is different between the index we copy
the CQE from and the index we copy it to.
Found by Ronni Zimmerman of Mellanox.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
New ConnectX firmware introduces FW command interface revision 2,
which requires that for each QP, a chunk of send queue entries (the
"headroom") is kept marked as invalid, so that the HCA doesn't get
confused if it prefetches entries that haven't been posted yet. Add
code to the driver to do this, and also update the user ABI so that
userspace can request that the prefetcher be turned off for userspace
QPs (we just leave the prefetcher on for all kernel QPs).
Unfortunately, marking send queue entries this way is confuses older
firmware, so we change the driver to allow only FW command interface
revisions 2. This means that users will have to update their firmware
to work with the new driver, but the firmware is changing quickly and
the old firmware has lots of other bugs anyway, so this shouldn't be too
big a deal.
Based on a patch from Jack Morgenstein <jackm@dev.mellanox.co.il>.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Cast the increment added to wq->tail when send completions are
processed to u16 to avoid using wrong values caused by standard
integer promotions.
The same bug was fixed in libmlx4 by Eli Cohen <eli@mellanox.co.il>.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add an InfiniBand driver for Mellanox ConnectX adapters. Because
these adapters can also be used as ethernet NICs and Fibre Channel
HBAs, the driver is split into two modules:
mlx4_core: Handles low-level things like device initialization and
processing firmware commands. Also controls resource allocation
so that the InfiniBand, ethernet and FC functions can share a
device without stepping on each other.
mlx4_ib: Handles InfiniBand-specific things; plugs into the
InfiniBand midlayer.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Or Gerlitz