Re: [PATCH v2 04/17] staging/rdma/hfi1: Fix qp.h comments

On Tue, Dec 01, 2015 at 03:38:13PM -0500, Jubin John wrote: > From: Kaike Wan > > This patch fixes a few incorrect header file comments in qp.h FWIW, within drivers/infiniband we've been moving these comments to the implementation, not the function prototype. Jason -- To

[PATCH v2 13/17] staging/rdma/hfi1: remove SPC freeze error messages

From: Dean Luick An SPC freeze is not an error. Remove the messages. Reviewed-by: Dennis Dalessandro Signed-off-by: Dean Luick Signed-off-by: Jubin John --- Changes in v2: - No changes

[PATCH v2 10/17] staging/rdma/hfi1: Adds software counters for bitfields within various error status fields

From: Joel Rosenzweig Provides error status counters for CceErrStatus, Send*ErrStatus, RcvErrStatus and MISC_ERR_STATUS Reviewed-by: Mitko Haralanov Reviewed-by: Mike Marciniszyn Signed-off-by: Joel Rosenzweig

[PATCH v2 05/17] staging/rdma/hfi1: Clean up comments

From: Edward Mascarenhas Clean up comments by deleting numbering and terms internal to Intel. The information on the actual bugs is not deleted. Reviewed-by: Mike Marciniszyn Signed-off-by: Edward Mascarenhas

[PATCH v2 00/17] Driver cleanup and misc fixes series 3

Changes in v2: - Added more information in commit messages of patches 01, 02 03, 04, 05, 07 and 09 - Fixed driver name to hfi1 in subject line of patch 06 - Refreshed patch 10 on top of staging-next - Dropped patch 18 (staging/rdma/hfi1: Workaround

[PATCH v2 02/17] staging/rdma/hfi1: Decode CNP opcode

From: Dean Luick Add CNP opcode decode. Prior to this patch the trace appeared like: -0 [001] d.h. 94062.578932: input_ibhdr: [:05:00.0] vl 0 lver 0 sl 0 lnh 2,LRH_BTH dlid 0003 len 6 slid 0001 op 0x80,0x80 se 0 m 0 pad 0 tver 0 pkey 0x8001 f 0 b 0 qpn 0x001234 a

[PATCH v2 03/17] staging/rdma/hfi1: Add aeth name syndrome decode

From: Dean Luick Add aeth name syndrome decode to enhance debugging. The IBTA RC ACK contains an ACK extended transport header. Part of that header is the syndrome field that qualifies the RC ACK as an ACK, NAK, or RNR NAK. Without the patch here is the syndrome decode:

[PATCH v2 09/17] staging/rdma/hfi1: Correctly limit VLs against SDMA engines

From: Dean Luick Correctly reduce the number of VLs when limited by the number of SDMA engines. The hardware has multiple egress mechanisms, SDMA and pio, and multiples of those. These mechanisms are chosen using the VL (8) The fix corrects a panic issue with one of the

[PATCH v2 17/17] staging/rdma/hfi1: Adding counter resolutions for DataPortCounters

From: Andrea Lowe Changing the 32-bit reserved field in opa_port_data_counters_msg to the new 'resolution' field. PMA will use resolutions to right- shift values for LocalLinkIntegrity and LinkErrorRecovery when computing the ErrorCounterSummary for a DataPortCounters

[PATCH v2 16/17] staging/rdma/hfi1: Fix Xmit Wait calculation

From: Ira Weiny Total XMIT wait needs to sum the xmit wait values of all the VLs not just those requested in the query. Also, make the algorithm used for both PortStatus and PortDataCounters the same. Reviewed-by: Arthur Kepner Reviewed-by: Breyer,

[PATCH v2 06/17] staging/rdma/hfi1: Add one-time LCB reset

From: Dean Luick Add one-time LCB reset on driver load to pre-emptively work around any LCB power cycle issues. Reviewed-by: Easwar Hariharan Signed-off-by: Dean Luick Signed-off-by: Jubin John ---

[PATCH v2 11/17] staging/rdma/hfi1: Destroy workqueues if hfi1_register_ib_device() call returns error

From: Harish Chegondi Currently, if hfi1_register_ib_device() call is unsuccessful, workqueues are not being destroyed before bailing out. This patch fixes this issue. Reviewed-by: Dennis Dalessandro Signed-off-by: Harish Chegondi

[PATCH v2 01/17] staging/rdma/hfi1: Support alternate firmware names

From: Dean Luick Add support for an automatic fallback for firmware names to support debug-signed and production-signed firmware images. Reviewed-by: Dennis Dalessandro Signed-off-by: Dean Luick Signed-off-by: Jubin

[PATCH v2 07/17] staging/rdma/hfi1: Extend quiet timeout

From: Dean Luick The longest quiet timeout is now 6s. Extend the driver wait to 6s. The driver wasn't following our internal specification: 6 seconds. This patch corrects that issue. Reviewed-by: Dennis Dalessandro Signed-off-by: Dean Luick

[PATCH v2 08/17] staging/rdma/hfi1: Add a credit push on diagpkt allocate fail

From: Dean Luick When sending a diagnostic packet, if the send context does not have enough room, force a credit return and try again. Reviewed-by: Dennis Dalessandro Signed-off-by: Dean Luick Signed-off-by: Jubin John

[PATCH v2 15/17] staging/rdma/hfi1: Consider VL15 MTU also when calculating the maximum VL MTU

From: Harish Chegondi Currently, only MTUs of VLs 0-7 are checked when calculating the maximum VL MTU which is used to set the port MTU capability in DCC_CFG_PORT_CONFIG CSR This can cause a port MTU capability to be set to 0 if MTUs of VLs 0-7 is 0 This would affect

[PATCH v2 12/17] staging/rdma/hfi1: Unexpected link up pkey values are not an error

From: Dean Luick Only warn when link up pkeys are not what we expect. Also, allow for the pkey to already be initialized. Reviewed-by: Arthur Kepner Signed-off-by: Dean Luick Signed-off-by: Jubin John

[PATCH v2 04/17] staging/rdma/hfi1: Fix qp.h comments

From: Kaike Wan This patch fixes a few incorrect header file comments in qp.h Reviewed-by: Mike Marciniszyn Signed-off-by: Kaike Wan Signed-off-by: Jubin John --- Changes in v2: - Added more

[PATCH v2 14/17] staging/rdma/hfi1: unknown frame messages are not errors

From: Dean Luick Change reported unknown frame messages into a counter. These are informational, no errors. Reviewed-by: Dennis Dalessandro Signed-off-by: Dean Luick Signed-off-by: Jubin John ---

Re: stalled again

On Tue, 1 Dec 2015, Or Gerlitz wrote: > We're against into this... upstream is on 4.4-rc3 while your latest branch in > kernel.org (the one that carries thefor-next tag) is rebased to 4.3-rc3... Seems that everything was merged? So you can directly use 4.4-rc1 until he comes up with a for-next

[PATCH 6/6] IB/srp: Fix srp_map_sg_fr()

After dma_map_sg() has been called the return value of that function must be used as the number of elements in the scatterlist instead of scsi_sg_count(). Fixes: commit f7f7aab1a5c0 ("IB/srp: Convert to new registration API") Reported-by: Christoph Hellwig Signed-off-by: Bart Van

[PATCH 0/6] SRP initiator related bug fixes

This patch series contains six bug fixes either for the SRP initiator itself or for IB core functionality used by the SRP initiator. The order of these patches matches the order in which the corresponding bugs have been introduced. The patches in this series are:

[PATCH 5/6] IB core: Fix ib_sg_to_pages()

Fix the code for detecting gaps and disable the code for chunking. A gap occurs not only if the second or later scatterlist element is not aligned but also if any scatterlist element other than the last does not end at a page boundary. Disable the code for chunking. Ensure that this function

[PATCH 3/6] IB/srp: Initialize dma_length in srp_map_idb

From: Christoph Hellwig Without this sg_dma_len will return 0 on architectures tha have the dma_length field. Fixes: commit f7f7aab1a5c0 ("IB/srp: Convert to new registration API") Signed-off-by: Christoph Hellwig --- drivers/infiniband/ulp/srp/ib_srp.c | 3 +++ 1

[PATCH 4/6] IB/srp: Fix indirect data buffer rkey endianness

Detected by sparse. Fixes: commit 330179f2fa93 ("IB/srp: Register the indirect data buffer descriptor") Signed-off-by: Bart Van Assche Cc: stable # v4.3+ Cc: Sagi Grimberg Cc: Christoph Hellwig Cc: Sebastian

Re: [PATCH 5/6] IB core: Fix ib_sg_to_pages()

Hi Bart, Fix the code for detecting gaps and disable the code for chunking. A gap occurs not only if the second or later scatterlist element is not aligned but also if any scatterlist element other than the last does not end at a page boundary. Disable the code for chunking. So can you

Re: stalled again

On Tue, Dec 01, 2015 at 06:30:28PM +0200, Or Gerlitz wrote: > Doug, > > We're against into this... upstream is on 4.4-rc3 while your latest branch > in kernel.org (the one that carries thefor-next tag) is rebased to > 4.3-rc3... > > --> our internal build and review systems for patches to

[PATCH 1/6] IB/srp: Fix a memory leak

If srp_connect_ch() returns a positive value then that is considered by its caller as a connection failure but this does not result in a scsi_host_put() call and additionally causes the srp_create_target() function to return a positive value while it should return a negative value. Avoid all this

[PATCH 2/6] IB/srp: Fix possible send queue overflow

From: Sagi Grimberg When using work request based memory registration (fast_reg) we must reserve SQ entries for registration and invalidation in addition to send operations. Each IO consumes 3 SQ entries (registration, send, invalidation) so we need to allocate 3x larger

[PATCH v3 01/13] staging/rdma/hfi1: diag.c use BIT macros

From: Ira Weiny Use BIT macros rather than shifts. Signed-off-by: Ira Weiny --- drivers/staging/rdma/hfi1/diag.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/staging/rdma/hfi1/diag.c

[PATCH v3 03/13] staging/rdma/hfi1: diag.c fix logical continuations

From: Ira Weiny Place logical operators at the end of the previous line when using a multi-line statement. Found by checkpatch --strict Signed-off-by: Ira Weiny --- drivers/staging/rdma/hfi1/diag.c | 6 +++--- 1 file changed, 3 insertions(+), 3

[PATCH v3 00/13] Fix hfi1_ioctl locking

From: Ira Weiny It was identified that hfi1_ioctl may sleep with a spin lock held. This was identified publicly here: http://www.spinics.net/lists/linux-rdma/msg29926.html As well as by our internal development. This series cleans up the code and parameter checks, as

[PATCH v3 07/13] staging/rdma/hfi1: diag.c correct sizeof parameter

From: Ira Weiny sizeof should use the variable rather than the struct definition to ensure that type changes are properly accounted for. Signed-off-by: Ira Weiny --- drivers/staging/rdma/hfi1/diag.c | 2 +- 1 file changed, 1 insertion(+), 1

[PATCH v3 08/13] staging/rdma/hfi1: Fix camel case variables

From: Ira Weiny physState, linkState, and devState should be phys_state, link_state, and dev_state Signed-off-by: Dennis Dalessandro Signed-off-by: Ira Weiny --- drivers/staging/rdma/hfi1/diag.c | 24

[PATCH v3 06/13] staging/rdma/hfi1: diag.c add missing braces

From: Ira Weiny Else statements should continue using braces even if there is only 1 line in the block. Found by checkpatch --strict Signed-off-by: Ira Weiny --- drivers/staging/rdma/hfi1/diag.c | 6 -- 1 file changed, 4 insertions(+), 2

[PATCH v3 10/13] staging/rdma/hfi1: hfi1_ioctl remove setlink state

From: Ira Weiny Set link state is not supported remove from the switch statement and allow the default to return -ENOTTY Signed-off-by: Dennis Dalessandro Signed-off-by: Ira Weiny --- Changes from v1: Just remove

[PATCH v3 09/13] staging/rdma/hfi1: Return early from hfi1_ioctl parameter errors

From: Ira Weiny Rather than have a switch in a large else clause make the parameter checks return immediately. Signed-off-by: Dennis Dalessandro Signed-off-by: Ira Weiny --- drivers/staging/rdma/hfi1/diag.c | 348

[PATCH v3 13/13] staging/rdma/hfi1: Return immediately on error

From: Dennis Dalessandro Now that the spinlock is not taken throughout hfi1_ioctl it is safe to return early rather than setting a variable and falling through the switch. Reviewed-by: Mike Marciniszyn Signed-off-by: Dennis Dalessandro

[PATCH v3 02/13] staging/rdma/hfi1: diag.c fix alignment

From: Ira Weiny Fix line alignment in various places as caught by checkpatch --strict. Signed-off-by: Ira Weiny --- drivers/staging/rdma/hfi1/diag.c | 16 1 file changed, 8 insertions(+), 8 deletions(-) diff --git

[PATCH v3 11/13] staging/rdma/hfi1: Further clean up hfi1_ioctl parameter checks

From: Ira Weiny Final clean up of the if/then/else clause for the parameter checks of hfi1_ioctl Signed-off-by: Dennis Dalessandro Signed-off-by: Ira Weiny --- Changes from v1: SETLINKSTATE is removed

[PATCH v3 05/13] staging/rdma/hfi1: diag.c change null comparisons

From: Ira Weiny Use !foo rather than (foo == NULL) as recommended by checkpatch --strict Signed-off-by: Ira Weiny --- drivers/staging/rdma/hfi1/diag.c | 20 ++-- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git

[PATCH v3 12/13] staging/rdma/hfi1: Reduce snoop locking scope in IOCTL handler.

From: Dennis Dalessandro This patch avoids issues while calling into copy from/to user while holding the lock by only taking the lock when it is absolutely required. The only commands which require the snoop lock are: *Set Filter *Clear Filter *Clear Queue

Re: [PATCH 6/6] IB/srp: Fix srp_map_sg_fr()

After dma_map_sg() has been called the return value of that function must be used as the number of elements in the scatterlist instead of scsi_sg_count(). Umm, but ib_map_mr_sg iterates on the sg list. Say you have sg_nents=3 and after mapping you got dma_nents=2 (2 entries are contig) and

Re: [PATCH 3/6] IB/srp: Initialize dma_length in srp_map_idb

Reviewed-by: Sagi Grimberg -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: [PATCH 4/6] IB/srp: Fix indirect data buffer rkey endianness

On 01/12/2015 20:18, Bart Van Assche wrote: Detected by sparse. Fixes: commit 330179f2fa93 ("IB/srp: Register the indirect data buffer descriptor") Signed-off-by: Bart Van Assche Cc: stable # v4.3+ Cc: Sagi Grimberg

Re: [PATCH 6/6] IB/srp: Fix srp_map_sg_fr()

On 12/01/2015 10:35 AM, Sagi Grimberg wrote: After dma_map_sg() has been called the return value of that function must be used as the number of elements in the scatterlist instead of scsi_sg_count(). Umm, but ib_map_mr_sg iterates on the sg list. Say you have sg_nents=3 and after mapping you

Re: [PATCH 5/6] IB core: Fix ib_sg_to_pages()

On 12/01/2015 10:32 AM, Sagi Grimberg wrote: Fix the code for detecting gaps and disable the code for chunking. A gap occurs not only if the second or later scatterlist element is not aligned but also if any scatterlist element other than the last does not end at a page boundary. Disable the

Re: [PATCH 4/6] IB/srp: Fix indirect data buffer rkey endianness

On 12/01/2015 10:37 AM, Sagi Grimberg wrote: On 01/12/2015 20:18, Bart Van Assche wrote: Detected by sparse. Fixes: commit 330179f2fa93 ("IB/srp: Register the indirect data buffer descriptor") Signed-off-by: Bart Van Assche Cc: stable #

[PATCH v4 1/2] staging/rdma/hf1: add dd_dev_dbg

From: Ira Weiny To be used in future patches add dd_dev_dbg. dd_* functions properly decode the hfi1_devdata structure used throughout the driver Signed-off-by: Ira Weiny --- drivers/staging/rdma/hfi1/hfi.h | 4 1 file changed, 4 insertions(+)

[PATCH v4 2/2] staging/rdma/hfi1: set Gen3 half-swing for integrated devices

From: Dean Luick Correctly set half-swing for integrated devices. A0 needs all fields set for CcePcieCtrl. B0 and later only need a few fields set. Reviewed-by: Stuart Summers Signed-off-by: Dean Luick Signed-off-by: Ira

[PATCH v4 0/2] staging/rdma/hfi1: set Gen 3 half-swing for integrated devices.

From: Ira Weiny This was a single patch before. The change to dev_dbg required a precursor patch to add the dd_dev_dbg which is consistent with the other dev_* macros which automatically use struct hfi1_devdata. Dean Luick (1): staging/rdma/hfi1: set Gen3 half-swing for

Re: [PATCH v1 10/10] IB/iser: Support the remote invalidation exception

On 30/11/2015 23:36, Or Gerlitz wrote: On Tue, Nov 24, 2015 at 6:23 PM, Sagi Grimberg wrote: From: Jenny Derzhavetz Declare that we support remote invalidation in case we are: 1. using Fastreg method 2. always registering memory. decide if you

[PATCH V3 1/2] IB/core: Fix user mode post wr corruption

Commit e622f2f4ad21 ("IB: split struct ib_send_wr") introduced a regression for HCAs whose user mode post sends go through ib_uverbs_post_send(). The code didn't account for the fact that the first sge is offset by an operation dependent length. The allocation did, but the pointer to the

Re: Future of FMR support, was: Re: [PATCH v1 5/9] xprtrdma: Add ro_unmap_sync method for FMR

> On Nov 24, 2015, at 2:12 AM, Jason Gunthorpe > wrote: > > On Mon, Nov 23, 2015 at 10:52:26PM -0800, Christoph Hellwig wrote: >> >> So at lest for 4.5 we're unlikely to be able to get rid of it alone >> due to the RDS issue. We'll then need performance

Re: [PATCH v2 6/7] xprtrdma: Add class for RDMA backwards direction transport

> On Dec 1, 2015, at 8:38 AM, Tom Talpey wrote: > > On 11/30/2015 5:25 PM, Chuck Lever wrote: >> To support the server-side of an NFSv4.1 backchannel on RDMA >> connections, add a transport class that enables backward >> direction messages on an existing forward channel

Re: [PATCH for-next V1 5/9] IB/core: Add rdma_network_type to wc

On Mon, Nov 30, 2015 at 10:56 PM, Liran Liss wrote: >> From: linux-rdma-ow...@vger.kernel.org [mailto:linux-rdma- > >> >> The abstraction at the gid cache is making it too easy to make this mistake. >> It >> is enabling callers to do direct gid lookups without a route

Re: [PATCH v2 6/7] xprtrdma: Add class for RDMA backwards direction transport

On 11/30/2015 5:25 PM, Chuck Lever wrote: To support the server-side of an NFSv4.1 backchannel on RDMA connections, add a transport class that enables backward direction messages on an existing forward channel connection. +static void * +xprt_rdma_bc_allocate(struct rpc_task *task, size_t

stalled again

Doug, We're against into this... upstream is on 4.4-rc3 while your latest branch in kernel.org (the one that carries thefor-next tag) is rebased to 4.3-rc3... --> our internal build and review systems for patches to linux-rdma can'tmake any use of your tree. People here have to rebase their