- 21 Dec, 2015 1 commit
-
-
Dean Luick authored
A link downgrade can race with link up. Avoid the race in two ways. First, by having the downgrade application logic take the link state mutex for all of its checking. Second, by waiting for the link to move out of the going up state. Reviewed-by:
Dennis Dalessandro <dennis.dalessandro@intel.com> Signed-off-by:
Dean Luick <dean.luick@intel.com> Signed-off-by:
Jubin John <jubin.john@intel.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
- 20 Nov, 2015 3 commits
-
-
Niranjana Vishwanathapura authored
Context 0 (which handles the error packets) can potentially receive an invalid rhf. Hence, it can not depend on RHF sequence number and can only use DMA_RTAIL mechanism. Detect such packets with invalid rhf using rhf sequence counting mechanism and drop them. As DMA_RTAIL mechanism has performance penalties, do not use context 0 for performance critical verbs path. Use context 0 for VL15 (MAD), multicast and error packets. Reviewed-by:
Arthur Kepner <arthur.kepner@intel.com> Reviewed-by:
Mike Marciniszyn <mike.marciniszyn@intel.com> Reviewed-by:
Mitko Haralanov <mitko.haralanov@intel.com> Signed-off-by:
Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com> Signed-off-by:
Ira Weiny <ira.weiny@intel.com> Signed-off-by:
-
Mark F. Brown authored
Disabling one receive context when RX_DMA is receiving a packet can cause incorrect packet delivery for a subsequent packet on another receive context. This is resolved by doing the following: 1. Programming dummy tail address for every receive context before enabling it 2. While deallocating receive context resetting tail address to dummy address 3. Leaving the dummy address in when disabling tail update 4. When disabling receive context leaving tail update enabled Reviewed-by:
Mark F. Brown <mark.f.brown@intel.com> Signed-off-by:
-
Mike Marciniszyn authored
The workqueue is currently single threaded per port which for a small number of SDMA engines is ok. For hfi1, the there are up to 16 SDMA engines that can be fed descriptors in parallel. Use alloc_workqueue with a workqueue limit of the number of sdma engines and with WQ_CPU_INTENSIVE and WQ_HIGHPRI specified. Then change send to use the new scheduler which no longer needs to get the s_lock Reviewed-by:
-
- 16 Nov, 2015 5 commits
-
-
Dean Luick authored
Use the one-shot LCB write implemented in the 8051 firmware. This speeds up 8051 LCB writes by 2x. Use old method for older firmwares. Reviewed-by:
-
Jareer Abdel-Qader authored
During driver load the thermal sensor needs to be reset prior to initialization of the sensor. This prevents a possible sensor lock up which can cause the wrong temperature value to be reported. This fix leads to remove disabling thermal polling from reset_asic_csrs() function. Reviewed by: Dennis Dalessandro <dennis.dalessandro@intel.com> Reviewed by: Easwar Hariharan <easwar.hariharan@intel.com> Signed-off-by:
Jareer Abdel-Qader <jareer.h.abdel-qader@intel.com> Signed-off-by:
-
Easwar Hariharan authored
The FLR on driver load asserts the QSFP reset pin and the driver does not deassert it after. This patch allows the external QSFP cable to exit reset by writing 1 to all the QSFP pins. Reviewed-by:
Easwar Hariharan <easwar.hariharan@intel.com> Signed-off-by:
-
Amitoj Kaur Chawla authored
Drop wrapper function remap_receive_available_interrupt() that wraps a call to remap_intr() with the only difference being the addition of macro IS_RCVAVAIL_START to the second argument of remap_intr(). Both the function names give the same information so the wrapper function can be dropped. Signed-off-by:
Amitoj Kaur Chawla <amitoj1606@gmail.com> Signed-off-by:
-
Amitoj Kaur Chawla authored
This patch removes hfi1_nomsix() wrapper function that is used to wrap pci_disable_msix() and so substituted the wrapper function by a direct call to pci_disable_msix(). Signed-off-by:
-
- 27 Oct, 2015 6 commits
-
-
Dean Luick authored
When under heavy load, the receive interrupt handler can run too long with IRQs disabled. Add a mixed-mode threading scheme. Initially process packets in the handler for quick responses (latency). If there are too many packets to process move to a thread to continue (bandwidth). Reviewed-by:
-
Easwar Hariharan authored
The DC firmware overwrites the enable_lane_tx register and does not update it on a host request to go to Poll. This causes an infinite loop through the LNI state machine if a link width downgrade occurs. This patch re-sets the enable_lane_tx register to all 4 lanes. Reviewed-by:
-
Dean Luick authored
When under heavy load, the send handler can run too long without allowing other tasks to run. Add a conditional resched to break this up. Reviewed-by:
-
Easwar Hariharan authored
The QSFP interface code has been running without issues and the flag is never set to off. This patch removes the QSFP_ENABLED bit from HFI1_CAP. Reviewed-by:
-
Dean Luick authored
The latest version of the 8051 firmware will wait longer when bringing the link down. Extend the driver's timeout to go with that. Reviewed-by:
-
Easwar Hariharan authored
The ASIC registers were not reset on FLR, and the code to protect the ASIC block against multiple initializations by peer HFIs did not extend to multiple ASICs in a system. This patch addresses this gap. Reviewed-by:
-
- 25 Oct, 2015 1 commit
-
-
Muhammad Falak R Wani authored
Use the timer API function setup_timer instead of init_timer, removing the structure field assignments. <smpl> @timer@ expression e1,e2,e3,fn_ptr; @@ -init_timer(&e1); +setup_timer(&e1, fn_ptr, e2); ... when != fn_ptr = e3 -e1.function = fn_ptr; -e1.data = e2; </smpl> Signed-off-by:
Muhammad Falak R Wani <falakreyaz@gmail.com> Signed-off-by:
-
- 17 Oct, 2015 1 commit
-
-
Shraddha Barke authored
void pointers do not need to be cast to other pointer types. The semantic patch used to find this: @r@ expression x; void* e; type T; identifier f; @@ ( *((T *)e) | ((T *)x)[...] | ((T *)x)->f | - (T *) e ) Signed-off-by:
Shraddha Barke <shraddha.6596@gmail.com> Signed-off-by:
-
- 13 Oct, 2015 1 commit
-
-
Alison Schofield authored
Out of memory messages are unnecssary in the drivers as they are reported by memory management. Addresses checkpatch.pl: WARNING: Possible unnecessary 'out of memory' message Signed-off-by:
Alison Schofield <amsfield22@gmail.com> Signed-off-by:
-
- 18 Sep, 2015 1 commit
-
-
Dan Carpenter authored
mutex_trylock() returns zero on failure, not EBUSY. Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
Doug Ledford <dledford@redhat.com>
-
- 29 Aug, 2015 1 commit
-
-
Mike Marciniszyn authored
Signed-off-by:
Andrew Friedley <andrew.friedley@intel.com> Signed-off-by:
Brendan Cunningham <brendan.cunningham@intel.com> Signed-off-by:
Brian Welty <brian.welty@intel.com> Signed-off-by:
Caz Yokoyama <caz.yokoyama@intel.com> Signed-off-by:
Harish Chegondi <harish.chegondi@intel.com> Signed-off-by:
Jim Snow <jim.m.snow@intel.com> Signed-off-by:
John Gregor <john.a.gregor@intel.com> Signed-off-by:
Kaike Wan <kaike.wan@intel.com> Signed-off-by:
Kevin Pine <kevin.pine@intel.com> Signed-off-by:
Kyle Liddell <kyle.liddell@intel.com> Signed-off-by:
Ravi Krishnaswamy <ravi.krishnaswamy@intel.com> Signed-off-by:
Sadanand Warrier <sadanand.warrier@intel.com> Signed-off-by:
Sanath Kumar <sanath.s.kumar@intel.com> Signed-off-by:
Sudeep Dutt <sudeep.dutt@intel.com> Signed-off-by:
Vlad Danushevsky <vladimir.danusevsky@intel.com> Signed-off-by:
-
