mthca_cq_clean() updates the CQ consumer index without moving CQEs
back to HW ownership. As a result, the same WRID might get reported
twice, resulting in a use-after-free. This was observed in IPoIB CM.
Fix by moving all freed CQEs to HW ownership.
This fixes <https://bugs.openfabrics.org/show_bug.cgi?id=617>
Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Fix posting lists of > 255 receive WRs for Tavor: rq.next_ind must
be updated each doorbell, otherwise the next doorbell will use an
incorrect index.
Found by Ronni Zimmermann at Mellanox.
Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Export ib_umem_get()/ib_umem_release() and put low-level drivers in
control of when to call ib_umem_get() to pin and DMA map userspace,
rather than always calling it in ib_uverbs_reg_mr() before calling the
low-level driver's reg_user_mr method.
Also move these functions to be in the ib_core module instead of
ib_uverbs, so that driver modules using them do not depend on
ib_uverbs.
This has a number of advantages:
- It is better design from the standpoint of making generic code a
library that can be used or overridden by device-specific code as
the details of specific devices dictate.
- Drivers that do not need to pin userspace memory regions do not
need to take the performance hit of calling ib_mem_get(). For
example, although I have not tried to implement it in this patch,
the ipath driver should be able to avoid pinning memory and just
use copy_{to,from}_user() to access userspace memory regions.
- Buffers that need special mapping treatment can be identified by
the low-level driver. For example, it may be possible to solve
some Altix-specific memory ordering issues with mthca CQs in
userspace by mapping CQ buffers with extra flags.
- Drivers that need to pin and DMA map userspace memory for things
other than memory regions can use ib_umem_get() directly, instead
of hacks using extra parameters to their reg_phys_mr method. For
example, the mlx4 driver that is pending being merged needs to pin
and DMA map QP and CQ buffers, but it does not need to create a
memory key for these buffers. So the cleanest solution is for mlx4
to call ib_umem_get() in the create_qp and create_cq methods.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
* 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband:
IPoIB: Convert to NAPI
IB: Return "maybe missed event" hint from ib_req_notify_cq()
IB: Add CQ comp_vector support
IB/ipath: Fix a race condition when generating ACKs
IB/ipath: Fix two more spin lock problems
IB/fmr_pool: Add prefix to all printks
IB/srp: Set proc_name
IB/srp: Add orig_dgid sysfs attribute to scsi_host
IPoIB/cm: Don't crash if remote side uses one QP for both directions
RDMA/cxgb3: Support for new abort logic
RDMA/cxgb3: Initialize cpu_idx field in cpl_close_listserv_req message
RDMA/cxgb3: Fail qp creation if the requested max_inline is too large
RDMA/cxgb3: Fix TERM codes
IPoIB/cm: Fix error handling in ipoib_cm_dev_open()
IB/ipath: Don't corrupt pending mmap list when unmapped objects are freed
IB/mthca: Work around kernel QP starvation
IB/ipath: Don't put QP in timeout queue if waiting to send
IB/ipath: Don't call spin_lock_irq() from interrupt context
The semantics defined by the InfiniBand specification say that
completion events are only generated when a completions is added to a
completion queue (CQ) after completion notification is requested. In
other words, this means that the following race is possible:
while (CQ is not empty)
ib_poll_cq(CQ);
// new completion is added after while loop is exited
ib_req_notify_cq(CQ);
// no event is generated for the existing completion
To close this race, the IB spec recommends doing another poll of the
CQ after requesting notification.
However, it is not always possible to arrange code this way (for
example, we have found that NAPI for IPoIB cannot poll after
requesting notification). Also, some hardware (eg Mellanox HCAs)
actually will generate an event for completions added before the call
to ib_req_notify_cq() -- which is allowed by the spec, since there's
no way for any upper-layer consumer to know exactly when a completion
was really added -- so the extra poll of the CQ is just a waste.
Motivated by this, we add a new flag "IB_CQ_REPORT_MISSED_EVENTS" for
ib_req_notify_cq() so that it can return a hint about whether the a
completion may have been added before the request for notification.
The return value of ib_req_notify_cq() is extended so:
< 0 means an error occurred while requesting notification
== 0 means notification was requested successfully, and if
IB_CQ_REPORT_MISSED_EVENTS was passed in, then no
events were missed and it is safe to wait for another
event.
> 0 is only returned if IB_CQ_REPORT_MISSED_EVENTS was
passed in. It means that the consumer must poll the
CQ again to make sure it is empty to avoid the race
described above.
We add a flag to enable this behavior rather than turning it on
unconditionally, because checking for missed events may incur
significant overhead for some low-level drivers, and consumers that
don't care about the results of this test shouldn't be forced to pay
for the test.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add a num_comp_vectors member to struct ib_device and extend
ib_create_cq() to pass in a comp_vector parameter -- this parallels
the userspace libibverbs API. Update all hardware drivers to set
num_comp_vectors to 1 and have all ULPs pass 0 for the comp_vector
value. Pass the value of num_comp_vectors to userspace rather than
hard-coding a value of 1.
We want multiple CQ event vector support (via MSI-X or similar for
adapters that can generate multiple interrupts), but it's not clear
how many vectors we want, or how we want to deal with policy issues
such as how to decide which vector to use or how to set up interrupt
affinity. This patch is useful for experimenting, since no core
changes will be necessary when updating a driver to support multiple
vectors, and we know that we want to make at least these changes
anyway.
Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
I noticed that many source files include <linux/pci.h> while they do
not appear to need it. Here is an attempt to clean it all up.
In order to find all possibly affected files, I searched for all
files including <linux/pci.h> but without any other occurence of "pci"
or "PCI". I removed the include statement from all of these, then I
compiled an allmodconfig kernel on both i386 and x86_64 and fixed the
false positives manually.
My tests covered 66% of the affected files, so there could be false
positives remaining. Untested files are:
arch/alpha/kernel/err_common.c
arch/alpha/kernel/err_ev6.c
arch/alpha/kernel/err_ev7.c
arch/ia64/sn/kernel/huberror.c
arch/ia64/sn/kernel/xpnet.c
arch/m68knommu/kernel/dma.c
arch/mips/lib/iomap.c
arch/powerpc/platforms/pseries/ras.c
arch/ppc/8260_io/enet.c
arch/ppc/8260_io/fcc_enet.c
arch/ppc/8xx_io/enet.c
arch/ppc/syslib/ppc4xx_sgdma.c
arch/sh64/mach-cayman/iomap.c
arch/xtensa/kernel/xtensa_ksyms.c
arch/xtensa/platform-iss/setup.c
drivers/i2c/busses/i2c-at91.c
drivers/i2c/busses/i2c-mpc.c
drivers/media/video/saa711x.c
drivers/misc/hdpuftrs/hdpu_cpustate.c
drivers/misc/hdpuftrs/hdpu_nexus.c
drivers/net/au1000_eth.c
drivers/net/fec_8xx/fec_main.c
drivers/net/fec_8xx/fec_mii.c
drivers/net/fs_enet/fs_enet-main.c
drivers/net/fs_enet/mac-fcc.c
drivers/net/fs_enet/mac-fec.c
drivers/net/fs_enet/mac-scc.c
drivers/net/fs_enet/mii-bitbang.c
drivers/net/fs_enet/mii-fec.c
drivers/net/ibm_emac/ibm_emac_core.c
drivers/net/lasi_82596.c
drivers/parisc/hppb.c
drivers/sbus/sbus.c
drivers/video/g364fb.c
drivers/video/platinumfb.c
drivers/video/stifb.c
drivers/video/valkyriefb.c
include/asm-arm/arch-ixp4xx/dma.h
sound/oss/au1550_ac97.c
I would welcome test reports for these files. I am fine with removing
the untested files from the patch if the general opinion is that these
changes aren't safe. The tested part would still be nice to have.
Note that this patch depends on another header fixup patch I submitted
to LKML yesterday:
[PATCH] scatterlist.h needs types.h
http://lkml.org/lkml/2007/3/01/141
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
With mthca, RC QPs can starve each other and even UD QPs on the same
hardware schedule queue. As a result, userspace MPI can starve
e.g. IPoIB traffic, with netdev watchdog warnings getting printed out,
and TCP connections getting stuck or failing.
Reduce the chance of this happening by using three separate hardware
schedule queues: one for userspace RC QPs, one for kernel RC QPs, and
one for all other QPs.
Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
All RDMA drivers except ehca set class_dev->dev to their dma_device
value (ehca leaves this unset). dma_device is the only value that
makes any sense, so move this assignment to core/sysfs.c. This reduce
the duplicated code in the rest of the drivers and gives ehca a nice
/sys/class/infiniband/ehcaX/device symlink.
Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
mthca_free_qp() already has local variables to hold the QP's send_cq
and recv_cq, so we can slightly clean up the calls to mthca_cq_clean()
by using those local variables instead of expressions like
to_mcq(qp->ibqp.send_cq).
Also, by cleaning the recv_cq first, we can avoid worrying about
whether the QP is attached to an SRQ for the second call, because we
would only clean send_cq if send_cq is not equal to recv_cq, and that
means send_cq cannot have any receive completions from the QP being
destroyed.
All this work even improves the generated code a bit:
add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-5 (-5)
function old new delta
mthca_free_qp 510 505 -5
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Commit b2875d4c ("IB/mthca: Always fill MTTs from CPU") causes a crash
in mthca_write_mtt() with non-memfree HCAs that have their memory
hidden (that is, have only two PCI BARs instead of having a third BAR
that allows access to the RAM attached to the HCA) on 64-bit
architectures. This is because the commit just before, c20e20ab
("IB/mthca: Merge MR and FMR space on 64-bit systems") makes
dev->mr_table.fmr_mtt_buddy equal to &dev->mr_table.mtt_buddy and
hence mthca_write_mtt() tries to write directly into the HCA's MTT
table. However, since that table is in the HCA's memory, this is
impossible without the PCI BAR that gives access to that memory.
This causes a crash because mthca_tavor_write_mtt_seg() basically
tries to dereference some offset of a NULL pointer. Fix this by
adding a test of MTHCA_FLAG_FMR in mthca_write_mtt() so that we always
use the WRITE_MTT firmware command rather than writing directly if
FMRs are not enabled.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
In mthca_arbel_fmr_unmap(), the high bits of the key are masked off.
This gets rid of the effect of adjust_key(), which makes sure that
bits 3 and 23 of the key are equal when the Sinai throughput
optimization is enabled, and so it may happen that an FMR will end up
with bits 3 and 23 in the key being different. This causes data
corruption, because when enabling the throughput optimization, the
driver promises the HCA firmware that bits 3 and 23 of all memory keys
will always be equal.
Fix by re-applying adjust_key() after masking the key.
Thanks to Or Gerlitz for reproducing the problem, and Ariel Shahar for
help in debug.
Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Commit c20e20ab ("IB/mthca: Merge MR and FMR space on 64-bit systems")
swapped the number of MTTs and MPTs when initializing the MR table. As
a result, we get a kernel oops when the number of MTT segments
allocated exceeds 0x20000.
Noted by Troy Benjegerdes <troy@scl.ameslab.gov>, and reproduced by
Dotan Barak <dotanb@mellanox.co.il>. This fixes
https://bugs.openfabrics.org/show_bug.cgi?id=490
Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The garbled logic in mthca_alloc_memfree() causes it to return 0, even
if it fails to allocate all doorbell records. Fix it to return -ENOMEM
when it fails.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
This patch makes the needlessly global functions mthca_tavor_write_mtt_seg()
and mthca_arbel_write_mtt_seg() static.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The change to allow allocating ICM chunks from coherent memory did not
increment the count of sg entries properly, so a chunk that required
more than allocation would not be mapped properly by the HCA.
Fix this by adding the missing increment of chunk->nsg.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
RESET->RESET is an allowed QP state transition, so mthca should handle
it correctly, by just returning success without involving the firmware.
Signed-off-by: Dotan Barak <dotanb@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Speed up memory registration by filling in MTTs directly when the CPU
can write directly to the whole table (all mem-free cards, and to
Tavor mode on 64-bit systems with the patch I posted earlier). This
reduces the number of FW commands needed to register an MR by at least
a factor of 2 and speeds up memory registration significantly.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
For Tavor, we currently reserve separate MPT and MTT space for FMRs to
avoid abusing the vmalloc space on 32 bit kernels. No such problem
exists on 64 bit kernels so let's not do it there.
This way we have a shared pool for MR and FMR resources, used on
demand. This will also make it possible to write MTTs for regular
regions directly from driver.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
We allocate the MTT table with alloc_pages() and then do pci_map_sg(),
so we must call pci_dma_sync_sg() after the CPU writes to the MTT
table. This works since the device will never write MTTs on mem-free
HCAs, once we get rid of the use of the WRITE_MTT firmware command.
This change is needed to make that work, and is an improvement for
now, since it gives FMRs a chance at working.
For MPTs, both the device and CPU might write there, so we must
allocate DMA coherent memory for these.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
MTTs are allocated in non-cache-coherent memory, so we must give
reserved MTTs their own cache line, to prevent both device and
CPU from writing into the same cache line at the same time.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The reserved_mtts field has different meaning in Tavor and Arbel, so
we are wasting mtt entries on memfree. Fix the Arbel case to match
Tavor semantics.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
For some reason gcc-3.4.5 on sparc64 does:
WARNING: "____ilog2_NaN" [drivers/infiniband/hw/mthca/ib_mthca.ko] undefined!
Points to note:
(1) The asm volatile flush/flushw are just markers for viewing what comes out
in the assembly; removing them has no effect on the result.
(2) Changing almost anything else in dwh__mthca_arbel_init_srq_context() or
dwh__mthca_alloc_srq() causes the problem to go away.
The compiler command line issued by the kernel build is:
/opt/crosstool/gcc-3.4.5-glibc-2.3.6/sparc64-unknown-linux-gnu/bin/sparc64-unknown-linux-gnu-gcc -fno-strict-aliasing -fno-common -Os -m64 -mno-fpu -mcpu=ultrasparc -mcmodel=medlow -ffixed-g4 -ffixed-g5 -fcall-used-g7 -Wa,--undeclared-regs -pg -fno-omit-frame-pointer -fno-optimize-sibling-calls -fasynchronous-unwind-tables -g -c -o drivers/infiniband/hw/mthca/.tmp_mthca_srq.o drivers/infiniband/hw/mthca/mthca_srq.c
This can be reduced to this whilst still retaining the problem:
/opt/crosstool/gcc-3.4.5-glibc-2.3.6/sparc64-unknown-linux-gnu/bin/sparc64-unknown-linux-gnu-gcc -m64 -c -o drivers/infiniband/hw/mthca/mthca_srq.o drivers/infiniband/hw/mthca/mthca_srq.c -Os
Removing -Os or changing it to -O or -O0 thru -O6 gets rid of the problem.
This patch to the kernel code fixes the problem:
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
When clearing the ib_ah_attr parameter in to_ib_ah_attr(), use sizeof
*ib_ah_attr instead of sizeof *path.
Pointed out by Jack Morgenstein <jackm@mellanox.co.il>.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
struct ib_wc currently only includes the local QP number: this matches
the IB spec, but seems mostly useless. The following patch replaces
this with the pointer to qp itself, and updates all low level drivers
and all users.
This has the following advantages:
- Ability to get a per-qp context through wc->qp->qp_context
- Existing drivers already have the qp pointer ready in poll cq, so
this change actually saves a tiny bit (extra memory read) on data path
(for ehca it would actually be expensive to find the QP pointer when
polling a CQ, but ehca does not support SRQ so we can leave wc->qp as
NULL for ehca)
- Users that need the QP number can still get it through wc->qp->qp_num
Use case:
In IPoIB connected mode code, I have a common CQ shared by multiple
QPs. To track connection usage, I need a way to get at some per-QP
context upon the completion, and I would like to avoid allocating
context object per work request just to stick a QP pointer into it.
With this code, I can just use wc->qp->qp_context.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
If a QP being queried is in the RESET state, don't execute the
QUERY_QP firmware command (because it will fail).
Signed-off-by: Dotan Barak <dotanb@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
According to the Tavor and Arbel programmer's reference manuals, the
number of bytes transferred is not provided in the byte_cnt field of
the CQ entry for atomic operation completions. For atomic operations,
the number of bytes transferred is always 8 (when the status is
"success"), and this constant value should always be used by the
driver in the ib_wc entry returned, rather than using the CQE.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
mthca_table_find() will return the wrong address when the table entry
being searched for is exactly at the beginning of a sglist entry
(other than the first), because it uses >= when it should use >.
Example: assume we have 2 entries in scatterlist, 4K each, offset is
4K. The current code will return first entry + 4K when we really want
the second entry.
In particular this means mapping an FMR on a memfree HCA may end up
writing the page table into the wrong place, leading to memory
corruption and also causing the HCA to use an incorrect address
translation table.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Commit bed8bdfd ("IB: kmemdup() cleanup") introduced one bad conversion to
kmemdup() in mthca_alloc_fmr(), where the structure allocated and the
structure copied are not the same size. Revert this back to the original
kmalloc()/memcpy() code.
Reported-by: Dotan Barak <dotanb@mellanox.co.il>.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <roland@digitalvampire.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
mthca_device_mutex() can be initialized automatically with
DEFINE_MUTEX() rather than explicitly calling mutex_init(). This
saves a bit of text and shrinks the source by a line, so we may as
well do it....
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add module parameters that enable settting some of the HCA
profile values, such as the number of QPs, CQs, etc.
Signed-off-by: Leonid Arsh <leonida@voltaire.com>
Signed-off-by: Moni Shoua <monis@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
This facility provides three entry points:
ilog2() Log base 2 of unsigned long
ilog2_u32() Log base 2 of u32
ilog2_u64() Log base 2 of u64
These facilities can either be used inside functions on dynamic data:
int do_something(long q)
{
...;
y = ilog2(x)
...;
}
Or can be used to statically initialise global variables with constant values:
unsigned n = ilog2(27);
When performing static initialisation, the compiler will report "error:
initializer element is not constant" if asked to take a log of zero or of
something not reducible to a constant. They treat negative numbers as
unsigned.
When not dealing with a constant, they fall back to using fls() which permits
them to use arch-specific log calculation instructions - such as BSR on
x86/x86_64 or SCAN on FRV - if available.
[akpm@osdl.org: MMC fix]
Signed-off-by: David Howells <dhowells@redhat.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: David Howells <dhowells@redhat.com>
Cc: Wojtek Kaniewski <wojtekka@toxygen.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
SLAB_ATOMIC is an alias of GFP_ATOMIC
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Conflicts:
drivers/infiniband/core/iwcm.c
drivers/net/chelsio/cxgb2.c
drivers/net/wireless/bcm43xx/bcm43xx_main.c
drivers/net/wireless/prism54/islpci_eth.c
drivers/usb/core/hub.h
drivers/usb/input/hid-core.c
net/core/netpoll.c
Fix up merge failures with Linus's head and fix new compilation failures.
Signed-Off-By: David Howells <dhowells@redhat.com>
When initializing an mthca SRQ, the log_srq_size field should be the
log of the number of SRQ WQEs, not the log of the number of bytes in
the SRQ.
This affects only mthca drivers for memfree HCAs which set the initial
srq wqe counter (in the SW2HW transition) to a non-zero value.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Commit b3b30f5e ("IB/mthca: Recover from catastrophic errors")
introduced some section mismatch breakage, because the error recovery
code tears down and reinitializes the device, which calls into lots of
code originally marked __devinit and __devexit from regular .text.
Fix this by getting rid of these now-incorrect section markers.
Reported by Randy Dunlap <randy.dunlap@oracle.com>.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Replace open coded kmemdup() to save some screen space, and allow
inlining/not inlining to be triggered by gcc.
Signed-off-by: Eric Sesterhenn <snakebyte@gmx.de>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Several fields in an incoming MAD extended info header were passed
into the MAD_IFC firmware command at incorrect offsets (mostly off by
4 bytes). As the result, the HCA will fail to generate traps in which
this info is needed (e.g. traps which include the GRH of the incoming
packet), in violation of the IB spec.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
We discovered a problem when running IPoIB applications on multiple
CPUs on an Altix system. Many messages such as:
ib_mthca 0002:01:00.0: SQ 000014 full (19941644 head, 19941707 tail, 64 max, 0 nreq)
appear in syslog, and the driver wedges up.
Apparently this is because writes to the doorbells from different CPUs
reach the device out of order. The following patch adds mmiowb() calls
after doorbell rings to ensure the doorbell writes are ordered.
Signed-off-by: Arthur Kepner <akepner@sgi.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
All HCAs (not just mem-free) need a spare SRQ entry, so bump srq->max
by 1 in all cases.
Noted by Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Fill in "max_vl_num" (encoded according to VLCap field in the PortInfo MAD)
and "init_type_reply" values in the ib_query_port() verb.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Maintain a per-CPU global "struct pt_regs *" variable which can be used instead
of passing regs around manually through all ~1800 interrupt handlers in the
Linux kernel.
The regs pointer is used in few places, but it potentially costs both stack
space and code to pass it around. On the FRV arch, removing the regs parameter
from all the genirq function results in a 20% speed up of the IRQ exit path
(ie: from leaving timer_interrupt() to leaving do_IRQ()).
Where appropriate, an arch may override the generic storage facility and do
something different with the variable. On FRV, for instance, the address is
maintained in GR28 at all times inside the kernel as part of general exception
handling.
Having looked over the code, it appears that the parameter may be handed down
through up to twenty or so layers of functions. Consider a USB character
device attached to a USB hub, attached to a USB controller that posts its
interrupts through a cascaded auxiliary interrupt controller. A character
device driver may want to pass regs to the sysrq handler through the input
layer which adds another few layers of parameter passing.
I've build this code with allyesconfig for x86_64 and i386. I've runtested the
main part of the code on FRV and i386, though I can't test most of the drivers.
I've also done partial conversion for powerpc and MIPS - these at least compile
with minimal configurations.
This will affect all archs. Mostly the changes should be relatively easy.
Take do_IRQ(), store the regs pointer at the beginning, saving the old one:
struct pt_regs *old_regs = set_irq_regs(regs);
And put the old one back at the end:
set_irq_regs(old_regs);
Don't pass regs through to generic_handle_irq() or __do_IRQ().
In timer_interrupt(), this sort of change will be necessary:
- update_process_times(user_mode(regs));
- profile_tick(CPU_PROFILING, regs);
+ update_process_times(user_mode(get_irq_regs()));
+ profile_tick(CPU_PROFILING);
I'd like to move update_process_times()'s use of get_irq_regs() into itself,
except that i386, alone of the archs, uses something other than user_mode().
Some notes on the interrupt handling in the drivers:
(*) input_dev() is now gone entirely. The regs pointer is no longer stored in
the input_dev struct.
(*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does
something different depending on whether it's been supplied with a regs
pointer or not.
(*) Various IRQ handler function pointers have been moved to type
irq_handler_t.
Signed-Off-By: David Howells <dhowells@redhat.com>
(cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
If a QP has separate send and receive CQs, then the send CQ will never
have receive completions from that QP in it. So when cleaning the
send CQ, there's no need to pass in an SRQ pointer, even if the QP is
attached to an SRQ.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Trigger device remove and then add when a catastrophic error is
detected in hardware. This, in turn, will cause a device reset, which
we hope will recover from the catastrophic condition.
Since this might interefere with debugging the root cause, add a
module option to suppress this behaviour.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Modifications to the existing rdma header files, core files, drivers,
and ulp files to support iWARP, including:
- Hook iWARP CM into the build system and use it in rdma_cm.
- Convert enum ib_node_type to enum rdma_node_type, which includes
the possibility of RDMA_NODE_RNIC, and update everything for this.
Signed-off-by: Tom Tucker <tom@opengridcomputing.com>
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Remove some trailing whitespace that has snuck in despite the best
efforts of whitespace=error-all. Also fix a few other whitespace
bogosities.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Incorrect number of bits was taken for static_rate field.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
port_num was not being returned for unconnected QPs.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
When default static rate is returned for Tavor, need to translate it
to an ib rate value.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Pass a struct ib_udata to the low-level driver's ->modify_srq() and
->modify_qp() methods, so that it can get to the device-specific data
passed in by the userspace driver.
Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Make kernel use UAR2 instead of UAR1 for hardware access: this adds
sanity checking from the hardware side, without any performance cost.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The SM LID used to send traps to is incorrectly set to port LID. This
is a regression from 2.6.17 -- after a PortInfo MAD is received, no
traps are sent to the SM LID. The traps go to the loopback interface
instead, and are dropped there. The SM LID should be taken from the
sm_lid of the PortInfo response.
The bug was introduced by commit 12bbb2b7be:
IB/mthca: Add client reregister event generation
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
It is supposed to be OK to call mthca_create_ah() and mthca_destroy_ah()
from any context. However, for mem-full HCAs, these functions use the
mthca_alloc() and mthca_free() bitmap helpers, and those helpers use
non-IRQ-safe spin_lock() internally. Lockdep correctly warns that
this could lead to a deadlock. Fix this by changing mthca_alloc() and
mthca_free() to use spin_lock_irqsave().
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Update the driver's list of HCA firmware revisions to make sure people
running Sinai firmware older than 1.1.0 get a message suggesting a
firmware upgrade. Update the Arbel versions as well while we are at it.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Leave all SRQ methods out of the device's uverbs_cmd_mask if the
device doesn't have SRQ support (because of ancient firmware) so that
we don't allow userspace to call the driver's create_srq method. This
fixes a userspace-triggerable oops caused by ib_uverbs_create_srq()
following the device's ->create_srq function pointer, which will be
NULL if the device doesn't support SRQs.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
When destroying a QP, mthca locks both the QP's send CQ and receive
CQ. However, the following scenario is perfectly valid:
QP_a: send_cq == CQ_x, recv_cq == CQ_y
QP_b: send_cq == CQ_y, recv_cq == CQ_x
The old mthca code simply locked send_cq and then recv_cq, which in
this case could lead to an AB-BA deadlock if QP_a and QP_b were
destroyed simultaneously.
We can fix this by changing the locking code to lock the CQ with the
lower CQ number first, which will create a consistent lock ordering.
Also, the second CQ is locked with spin_lock_nested() to tell lockdep
that we know what we're doing with the lock nesting.
This bug was found by lockdep.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The fence bit needs to be set in the doorbell too, not just the WQE.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Define a constant MTHCA_ARRAY_MASK to replace repeated uses of
(PAGE_SIZE / sizeof (void *) - 1) in mthca array code.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
mthca_array_clear() does not clear the slot if the used count is
positive. This leads to crashes in mthca_qp_event() since that uses
mthca_array_get() to check that the qp is valid.
Discovered by Ali Ayoub.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Read the max_cmds value from the response to the QUERY_FW command
before printing out the value, so that the real value goes into the
debug output.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Mem-free HCAs always keep one spare SRQ WQE, so the SRQ limit cannot
be set beyond srq->max - 1.
Signed-off-by: Dotan Barak <dotanb@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
After recent changes, mthca_wq_init does not actually initialize the WQ as it
used to - it simply resets all index fields to their initial values. So,
let's rename it to mthca_wq_reset.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Cc: Roland Dreier <rolandd@cisco.com>
Acked-by: Zach Brown <zach.brown@oracle.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
mthca_ah_query returs the static rate of the address handle in internal mthc
format. fix it to use rate encoding from enum ib_rate, which is what users
expect.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
mthca: initialize send and receive queue locks separately
lockdep identifies a lock by the call site of its initialization. By
initializing the send and receive queue locks in mthca_wq_init() we confuse
lockdep. It warns that that the ordered acquiry of both locks in
mthca_modify_qp() is recursive acquiry of one lock:
=============================================
[ INFO: possible recursive locking detected ]
---------------------------------------------
modprobe/1192 is trying to acquire lock:
(&wq->lock){....}, at: [<f892b4db>] mthca_modify_qp+0x60/0xa7b [ib_mthca]
but task is already holding lock:
(&wq->lock){....}, at: [<f892b4ce>] mthca_modify_qp+0x53/0xa7b [ib_mthca]
Initializing the locks separately in mthca_alloc_qp_common() stops the
warning and will let lockdep enforce proper ordering on paths that acquire
both locks.
Signed-off-by: Zach Brown <zach.brown@oracle.com>
Cc: Roland Dreier <rolandd@cisco.com>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This is needed if we wish to change the size of the resource structures.
Based on an original patch from Vivek Goyal <vgoyal@in.ibm.com>
Cc: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Documentation/infiniband/core_locking.txt says:
All of the methods in struct ib_device exported by a low-level
driver must be fully reentrant. The low-level driver is required to
perform all synchronization necessary to maintain consistency, even
if multiple function calls using the same object are run
simultaneously.
However, mthca's modify_qp, modify_srq and resize_cq methods are
currently not reentrant. Add a mutex to the QP, SRQ and CQ structures
so that these calls can be properly serialized.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Some error paths after the mthca_alloc_mailbox() call in mthca_modify_qp()
just do a "return -EINVAL" without freeing the mailbox. Convert these
returns to "goto out" to avoid leaking the mailbox storage.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Report the true max_map_per_fmr value from mthca_query_device(),
taking into account the change in FMR remapping introduced by the
Sinai performance optimization.
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Change the mthca snoop of MADs that set PortInfo to check if the SM
has set the client reregister bit, and if it has, generate a client
reregister event. If the bit is not set, just generate a LID change
event as usual.
Signed-off-by: Leonid Arsh <leonida@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The kernel has had wait_for_completion_timeout() for a long time now.
mthca should use it to handle FW commands timing out, instead of
implementing the same thing in a much more complicated way by using
wait_for_completion() along with a timer that does complete().
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Memfree firmware is in rare cases reporting WQE index == base - 1 in
receive completion with error, instead of (rq size - 1); base is 0 in
mthca. Here is a patch to avoid kernel crash and report a correct WR
id in this case.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
mthca does not restore the following PCI-X/PCI Express registers after reset:
PCI-X device: PCI-X command register
PCI-X bridge: upstream and downstream split transaction registers
PCI Express : PCI Express device control and link control registers
This causes instability and/or bad performance on systems where one of
these registers is set to a non-default value by BIOS.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
If we post a list of length exactly a multiple of 256, nreq in
doorbell gets set to 256 which is wrong: it should be encoded by 0.
This is because we only zero it out on the next WR, which may not be
there. The solution is to ring the doorbell after posting a WQE, not
before posting the next one.
This is the same bug that we just fixed for QPs with non-shared RQ.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
If we post a list of length 256 exactly, nreq in doorbell gets set to
256 which is wrong: it should be encoded by 0. This is because we
only zero it out on the next WR, which may not be there. The solution
is to ring the doorbell after posting a WQE, not before posting the
next one.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Setting fw_cmd_doorbell allows FW command to be queued using posted
writes instead of requiring polling on a "go" bit, so it should be a
performance boost. However, the option causes problems with at least
some device/firmware combinations, so set the default to 0 until we
understand what's going on better.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Addresses for ioremap must be calculated off of pci_resource_start;
we can't directly use the bus address as seen by the HCA. Fix the
code that remaps device memory for FMR access.
Based on patch by Klaus Smolin.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Fix races in in destroying various objects. If a destroy routine
waits for an object to become free by doing
wait_event(&obj->wait, !atomic_read(&obj->refcount));
/* now clean up and destroy the object */
and another place drops a reference to the object by doing
if (atomic_dec_and_test(&obj->refcount))
wake_up(&obj->wait);
then this is susceptible to a race where the wait_event() and final
freeing of the object occur between the atomic_dec_and_test() and the
wake_up(). And this is a use-after-free, since wake_up() will be
called on part of the already-freed object.
Fix this in mthca by replacing the atomic_t refcounts with plain old
integers protected by a spinlock. This makes it possible to do the
decrement of the reference count and the wake_up() so that it appears
as a single atomic operation to the code waiting on the wait queue.
While touching this code, also simplify mthca_cq_clean(): the CQ being
cleaned cannot go away, because it still has a QP attached to it. So
there's no reason to be paranoid and look up the CQ by number; it's
perfectly safe to use the pointer that the callers already have.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
GuidInfo records have 8 byte GUIDs in them, so an index should be
multiplied by 8 to get an offset. mthca_query_gid() was incorrectly
multiplying by 16.
Noticed by Leonid Keller <leonid@mellanox.co.il>.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
This patch makes the needlessly global mthca_update_rate() static.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The driver allocates SRQ WQEs size with a power of 2 size both for
Tavor and for memfree. For Tavor, however, the hardware only requires
the WQE size to be a multiple of 16, not a power of 2, and the max
number of scatter-gather allowed is reported accordingly by the
firmware (and this is the value currently returned by
ib_query_device() and ibv_query_device()).
If the max number of scatter/gather entries reported by the FW is used
when creating an SRQ, the creation will fail for Tavor, since the
required WQE size will be increased to the next power of 2, which
turns out to be larger than the device permitted max WQE size (which
is not a power of 2).
This patch reduces the reported SRQ max wqe size so that it can be used
successfully in creating an SRQ on Tavor HCAs.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The PCI spec recommends against drivers playing with a device's PCI
read burst size, and says that systems software should configure it.
And we actually have users that report that changing it from the
default set by BIOS hurts performance and/or stability for them. On
the other hand, the Mellanox Programmer's Reference Manual recommends
turning it up all the way to the maximum value. Some tests conducted
here in the lab do not show performance improvement from this tuning,
but this might be just me.
As a work-around, make this tuning an option, off by default (safe
value), with an eye towards removing it completely one day if no one
complains.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Push translation of static rate to HCA format into low-level drivers,
where it belongs. For static rate encoding, use encoding of rate
field from IB standard PathRecord, with addition of value 0, for
backwards compatibility with current usage. The changes are:
- Add enum ib_rate to midlayer includes.
- Get rid of static rate translation in IPoIB; just use static rate
directly from Path and MulticastGroup records.
- Update mthca driver to translate absolute static rate into the
format used by hardware. This also fixes mthca's static rate
handling for HCAs that are capable of 4X DDR.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Change the mthca debugging trace output code so that it can enabled
and disabled at runtime with the debug_level module parameter in
sysfs. Also, don't allow CONFIG_INFINIBAND_MTHCA_DEBUG to be disabled
unless CONFIG_EMBEDDED is selected. We want users (and especially
distros) to have this turned on unless they really need to save space,
because by the time we want debugging output, it's usually too late to
rebuild a kernel.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Quite a few cleanup functions in mthca were marked as __devexit.
However, they could also be called from error paths during
initialization, so they cannot be marked that way. Just delete all of
the incorrect annotations.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The previous patch for Tavor broke MemFree logic.
The driver should perform limit check only for Tavor. For MemFree,
the check is incorrect, since ds (WQE stride) is always a power-of-2
(although the max_desc_size may not be).
In Tavor, however, WQE stride and desc_size are the same, and are not
necessarily power-of-2. The check was really for the WQE stride (and
it Tavor, we use max_desc_size for the stride).
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
If the call to mthca_MODIFY_QP() failed, then mthca_modify_qp() would
still do some things it shouldn't, such as store away attributes for
special QPs. Fix this, and simplify the code, by simply jumping to
the exit path if mthca_MODIFY_QP() fails.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
mthca_alloc_sqp() by mthca_set_qp_size() need to set qp->transport
before calling mthca_set_qp_size(), since the value is used there.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
When setting the shared receive queue (SRQ) watermark in a modify SRQ
operation, make sure that the supplied value is not larger than the
full size of the SRQ.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Guarantee the calculated work queue entry size does not exceed the max
allowable WQE size when creating an SRQ. This is a problem with Arbel
in Tavor-compatibility mode because the current WQE size computation
method rounds up to next power of 2.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add a check that the modify QP parameters sgid_index and path_mtu are
valid, since they might come from userspace.
Signed-off-by: Dotan Barak <dotanb@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Fix endianness handling of srq_limit: it is big-endian in the context
structure, so we need to swab it before returning it.
Also add support for srq_limit query for Tavor (non-MemFree) HCAs.
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
MemFree devices need to reserve one shared receive queue (SRQ) work
request for internal use, so the capacity returned from the create_srq
and query_srq methods should be srq->max - 1.
Signed-off-by: Dotan Barak <dotanb@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Fix bug found by coverity: the loop body never executed, because it
was doing for (i = 0; i < MTHCA_EQ_CMD; ++i), but MTHCA_EQ_CMD is 0.
The correct loop bound is MTHCA_NUM_EQ, to loop over all EQs.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Sinai (one-port PCI Express) HCAs get improved throughput for messages
bigger than 80 KB in DDR mode if memory keys are formatted in a
specific way. The enhancement only works if the memory key table is
smaller than 2^24 entries. For larger tables, the enhancement is off
and a warning is printed (to avoid silent performance loss).
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Michael Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Use a named enum for the HCA's internal page size, rather than having
magic values of 4096 and shifts by 12 all over the code. Also, fix
one minor bug in EQ handling: only one HCA page is mapped to the HCA
during initialization, but a full kernel page is unmapped during
cleanup. This might cause problems when PAGE_SIZE != 4096.
Signed-off-by: Ishai Rabinovitz <ishai@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Check that the alternate P_Key index is in range when setting the
alternate path for a QP. Also make a cosmetic touch up to the debug
message printed when the main P_Key index is out of range.
Signed-off-by: Dotan Barak <dotanb@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add support for IB_SEND_FENCE flag in post_send methods.
Signed-off-by: Dotan Barak <dotanb@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Implement query_ah (except for AVs which are in HCA memory). This is
needed to implement RMPP duplicate session detection on sending side
(extraction of DGID/DLID and GRH flag from address handle).
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
This patch is checks whether the HCA supports posting FW commands
through a doorbell page (user access region 0, or "UAR0"). If this is
supported, the driver maps UAR0 and uses it for FW commands. This can
be controlled by the value of a writable module parameter
fw_cmd_doorbell. When the parameter is 0, the commands are posted
through HCR using the old method; otherwise if HCA is capable commands
go through UAR0.
This use of UAR0 to post commands eliminates the need for polling the
"go" bit prior to posting a new command. Since reading from a PCI
device is much more expensive then issuing a posted write, it is
expected that issuing FW commands this way will provide better CPU
utilization.
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Have mthca's create_srq method return the actual capacity of the SRQ
that gets created. Also update comments in <rdma/ib_verbs.h> to
clarify that this is what is expected from ib_create_srq().
Signed-off-by: Dotan Barak <dotanb@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Use ib_modify_qp_is_ok() in mthca, and delete the big table of
attributes for queue pair state transitions.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add low-level driver support to ib_mthca so that consumers can request
a "send queue drained" event be generated when a transiton to the SQD
state completes.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
This patch allows the consumer to set the page size of "pages" mapped
by the pool FMRs, which is a feature already existing in the base
verbs API. On the cosmetic side it changes ib_fmr_attr.page_size field
to be named page_shift.
Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add a modify_device method to mthca, which implements setting the node
description. This makes the writable "node_desc" sysfs attribute work
for Mellanox HCAs.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The might_sleep() annotations in mthca are silly -- they all occur
shortly before calls that will end up in core functions like kmalloc()
that will print the same warning in an unsafe context anyway. In
fact, beyond cluttering the source, we're actually bloating text with
CONFIG_DEBUG_SPINLOCK_SLEEP and/or CONFIG_PREEMPT_VOLUNTARY set.
With both options set, getting rid of the might_sleep()s saves a lot:
add/remove: 0/0 grow/shrink: 0/7 up/down: 0/-171 (-171)
function old new delta
mthca_pd_alloc 132 109 -23
mthca_init_cq 969 946 -23
mthca_mr_alloc 592 568 -24
mthca_pd_free 67 42 -25
mthca_free_mr 219 194 -25
mthca_free_cq 570 545 -25
mthca_fmr_alloc 742 716 -26
Signed-off-by: Roland Dreier <rolandd@cisco.com>
The function mthca_free_err_wqe() can never fail, so get rid of its
return value. That means handle_error_cqe() doesn't have to check
what mthca_free_err_wqe() returns, which means it can't fail either
and doesn't have to return anything either. All this results in
simpler source code and a slight object code improvement:
add/remove: 0/0 grow/shrink: 0/2 up/down: 0/-10 (-10)
function old new delta
mthca_free_err_wqe 83 81 -2
mthca_poll_cq 1758 1750 -8
Signed-off-by: Roland Dreier <rolandd@cisco.com>
When debugging is enabled, the mthca_QUERY_DEV_LIM() firmware command
function prints out some of the device limits that it queries.
However the debugging prints happen before all of the fields are
extracted from the firmware response, so some of the values that get
printed are uninitialized junk. Move the prints to the end of the
function to fix this.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Convert semaphores to mutexes in mthca. Leave firmware command
interface poll_sem and event_sem as semaphores.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
We have run into the following problem: if a task receives a signal
while in the process of e.g. destroying a resource (which could be
because the relevant file was closed) mthca could bail out from trying
to take a command interface semaphore without performing the
appropriate command to tell hardware that the resource is being
destroyed.
As a result we see messages like
ib_mthca 0000:04:00.0: HW2SW_CQ failed (-4)
In this case, hardware could access the resource after the memory has
been freed, possibly causing memory corruption.
A simple solution is to replace down_interruptible() by down() in
command interface activation.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
[ It's also not safe to bail out on multicast table operations, since
they may be invoked on the cleanup path too. So use down() for
mcg_table.sem too. ]
Signed-off-by: Roland Dreier <rolandd@cisco.com>
There are some cards around that have UAR (user access region) size
different from 8 MB. Relax our sanity check to make sure that the PCI
BAR is big enough to access the UAR size reported by the device
firmware instead.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
mthca_create_ah() includes the port number in the GID index. The reverse
needs to be done in mthca_read_ah().
Noted by Hal Rosenstock.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
build_mlx_header() was using sqp->ud_header.grh_present before it was
initialized by mthca_read_ah(). Furthermore, header->grh_present is
set by ib_ud_header_init, so there's no need to set it again in
mthca_read_ah().
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Use the ALIGN macro to simplify some rounding code.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Fix memory leaks in mthca_create_qp() and mthca_create_srq()
error handling.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Convert "/ (1 << lg)" to ">> lg" for a slight code size reduction.
add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-24 (-24)
function old new delta
mthca_map_cmd 613 589 -24
Signed-off-by: Ishai Rabinovitz <ishai@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Add a node_guid field to struct ib_device. It is the responsibility
of the low-level driver to initialize this field before registering a
device with the midlayer. Convert everyone to looking at this field
instead of calling ib_query_device() when all they want is the node
GUID, and remove the node_guid field from struct ib_device_attr.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Factor out common code for initializing MAD packets, which is shared
by many query routines in mthca_provider.c, into init_query_mad().
add/remove: 1/0 grow/shrink: 0/4 up/down: 16/-44 (-28)
function old new delta
init_query_mad - 16 +16
mthca_query_port 521 518 -3
mthca_query_pkey 301 294 -7
mthca_query_device 648 641 -7
mthca_query_gid 453 426 -27
Signed-off-by: Roland Dreier <rolandd@cisco.com>
I am seeing EQ overruns in SDP stress tests: if the CQ completion
handler arms a CQ, this could generate more EQEs, so that EQ will
never get empty and consumer index will never get updated.
This is similiar to what we have with command interface:
/*
* cmd_event() may add more commands.
* The card will think the queue has overflowed if
* we don't tell it we've been processing events.
*/
However, for completion events, we *don't* want to update the consumer
index on each event. So, perform EQ doorbell coalescing: allocate EQs
with some spare EQEs, and update once we run out of them.
The value 0x80 was selected to avoid any performance impact.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
For all pages except possibly the last one, the byte beyond the buffer
end must be page aligned. Therefore, when computing the page shift to
use, OR the end addresses of the buffers as well as the start
addresses into the mask we check.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Include fixes for 2.6.14-git11. Should allow to remove sched.h from
module.h on i386, x86_64, arm, ia64, ppc, ppc64, and s390. Probably more
to come since I haven't yet checked the other archs.
Signed-off-by: Tim Schmielau <tim@physik3.uni-rostock.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Add code to modify QP operation to handle setting alternate paths for
connected QPs.
Signed-off-by: Dotan Barak <dotanb@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Fill vendor_err field in completion with error.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Multicast group management fixes:
. Fix leak of mailbox memory in error handling on multicast group operations.
. Free AMGM indices at detach and in attach error handling.
. Fix amount to shift for aligning next_gid_index in mailbox: it
starts at bit 6, not bit 5.
. Allocate AMGM index after end of MGM table, in the range num_mgms to
multicast table size - 1. Add some BUG_ON checks to catch cases
where the index falls in the MGM hash area.
. Initialize the list of QPs in a newly-allocated group from AMGM to 0
This is necessary since when a group is moved from AMGM to MGM (in the
case where the MGM entry has been emptied of QPs), the AMGM entry is
not reset to 0 (and we don't want an extra command to do that).
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
PKEY_INDEX is not a legal parameter in the RTR->RTS transition.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Fixes to SQEr->RTS transition in modify_qp:
1. The flag IB_QP_ACCESS_FLAGS is optional for UC qps
2. The SQEr state is not supported for RC qps
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Fix a case where copying max_inline_data from a successful create_qp
capabilities output to create_qp input could cause EINVAL error:
mthca_set_qp_size must check max_inline_data directly against
max_desc_sz; checking qp->sq.max_gs is wrong since max_inline_data
depends on the qp type and does not involve max_sg.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Fix mthca_create_eq for when the EQ size is not a power of 2.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Modify_qp should check that the physical port number provided
is a legal value.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Check error return on call to mthca_dev_lim for Tavor
(as is done for memfree).
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Thinko: 64 bytes is the minimum SRQ WQE size (not the maximum).
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
sae and sre bits should only be set when setting sra_max. Further, in
the old code, if the caller specifies max_rd_atomic = 0, the sre and
sae bits are still set, with the result that the QP ends up with
max_rd_atomic = 1 in effect.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
This patch corrects some corner cases in managing the RAE/RRE bits in
the mthca qp context. These bits need to be zero if the user requests
max_dest_rd_atomic of zero. The bits need to be restored to the value
implied by the qp access flags attribute in a previous (or the
current) modify-qp command if the dest_rd_atomic variable is changed
to non-zero.
In the current implementation, the following scenario will not work:
RESET-to-INIT set QP access flags to all disabled (zeroes)
INIT-to-RTR set max_dest_rd_atomic=10, AND
set qp_access_flags = IB_ACCESS_REMOTE_READ | IB_ACCESS_REMOTE_ATOMIC
The current code will incorrectly take the access-flags value set in
the RESET-to-INIT transition.
We can simplify, and correct, this IB_QP_ACCESS_FLAGS handling: it is
always safe to set qp access flags in the firmware command if either
of IB_QP_MAX_DEST_RD_ATOMIC or IB_QP_ACCESS_FLAGS is set, so let's
just set it to the correct value, always.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
When cleaning up a CQ for a QP attached to SRQ, need to free an SRQ
WQE only if the CQE is a receive completion.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
break only escapes from the innermost loop, and we want to escape both
loops and return an answer. Noticed by Ishai Rabinovitch.
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Only change the driver's copy of the QP attributes in modify QP after
checking the modify QP command completed successfully.
Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
