- 17 Mar, 2016 17 commits
-
-
Kaike Wan authored
This patch fixed the problem where the driver might reschedule in atomic mode when sending packets. This is due to the fact that the call to cond_resched() in hfi1_do_send() might occur in atomic mode and a check is required to avoid the warning message: "kernel: BUG: scheduling while atomic: swapper/2/0/0x10000100." Reviewed-by:Dennis Dalessandro <dennis.dalessandro@intel.com> Signed-off-by:
Mike Marciniszyn <mike.marciniszyn@intel.com> Signed-off-by:
Kaike Wan <kaike.wan@intel.com> Signed-off-by:
Jubin John <jubin.john@intel.com> Signed-off-by:
Doug Ledford <dledford@redhat.com>
-
Dean Luick authored
The kernel memcpy is faster than a cacheless copy. However, if too much of the L3 cache is overwritten by one-time copies then overall bandwidth suffers. Implement an adaptive scheme where full page copies are tracked and if the number of unique entries are larger than a threshold, verbs will use a cacheless copy. Tracked entries are gradually cleaned, allowing memcpy to resume once the larger copies have stopped. Reviewed-by:
Dean Luick <dean.luick@intel.com> Signed-off-by:
-
Jubin John authored
Host handshake timeout can occur during the verify capability state. This is a LNI related failure and should be handled in the same way as other LNI failures. Reviewed-by:
Easwar Hariharan <easwar.hariharan@intel.com> Reviewed-by:
-
Dean Luick authored
Different OSes using parts of the same hardware may leave cross-device flags set. Export a debugfs file to view and clear these flags if needed. Reviewed-by:
Mitko Haralanov <mitko.haralanov@intel.com> Reviewed-by:
-
Dean Luick authored
External i2c firmware updates are done in multiple steps and cannot have other things done in between. For debugfs files, acquire the resource on open and release it on close. Reviewed-by:
-
Dean Luick authored
The hardware mutex is now held only long enough to set or clear flags. Reduce the timeout to something more reasonable. Reviewed-by:
-
Dean Luick authored
The flag HFI1_DO_INIT_ASIC flag is no longer used. Remove the flag and the code that sets it. Reviewed-by:
-
Dean Luick authored
Use the resource reservation system to flag that the ASIC thermal has been initialized. Reviewed-by:
-
Dean Luick authored
Remove the mutex guarding each operation in favor the ASIC resource acquire/release. Push the resource acquire/release, above each operation call to allow exclusive access across multiple operations. Reviewed-by:
-
Dean Luick authored
The SBus resource includes SBUS, PCIE, and THERM registers. Change SBus handling to use the new ASIC resource reservation system. Reviewed-by:
-
Dean Luick authored
Change EPROM handling to use the new ASIC resource reservation system. Reviewed-by:
-
Dean Luick authored
The ASIC block is a shared hardware resource between two devices on the chip. Add functions to acquire and release these resources in a way that is safe for both multiple users on the same OS and multiple users on different OSes, while holding the hardware mutex as little as possible. Reservations are noted in a scratch register in the shared region. There are two types of reservations: per-HFI dynamic and permanent. Reviewed-by:
-
Dean Luick authored
Create a shared structure to exist between devices that share the same ASIC. Reviewed-by:
-
Dean Luick authored
The ASIC block is shared between two HFIs. Individual devices should not initialize registers there. Retain the power-on values. Individual users set registers as needed with one exception. Clear sbus fast mode on "slow" calls. Reviewed-by:
-
Harish Chegondi authored
This change was recommended by Coccinelle tool when I ran the command: -bash-4.2$ make coccicheck MODE=patch M=drivers/infiniband/hw/hfi1/ Reviewed-by:
Harish Chegondi <harish.chegondi@intel.com> Signed-off-by:
-
Harish Chegondi authored
Implement changes recommended by the Coccinelle tool to move constant to the right in bitwise operations -bash-4.2$ make coccicheck MODE=report M=drivers/infiniband/hw/hfi1/ drivers/infiniband/hw/hfi1/pio.c:765:4-16: Move constant to right. drivers/infiniband/hw/hfi1/rc.c:2503:19-29: Move constant to right. drivers/infiniband/hw/hfi1/chip.c:9813:11-22: Move constant to right. drivers/infiniband/hw/hfi1/chip.c:14468:29-40: Move constant to right. Reviewed-by:
-
Harish Chegondi authored
The break statement was unintentionally removed in this patch commit 41ca419abc0ca7ee65d765408cdc1a7fed2897a3 ("staging/rdma/hfi1: Remove hfi1 MR and hfi1 specific qp type") Reviewed-by:
-
- 11 Mar, 2016 23 commits
-
-
Jubin John authored
Fix 3 memory leaks reported by the LeakCheck tool in the KEDR framework. The following resources were allocated memory during their respective initializations but not freed during cleanup: 1. SDMA map elements 2. PIO map elements 3. HW send context to SW index map This patch fixes the memory leaks by freeing the allocated memory in the cleanup path. Reviewed-by:
-
Easwar Hariharan authored
The LedInfo SMA attribute is redefined to control the LED beaconing state machine instead of the LED directly. In accordance, we now return the state of LED beaconing, represented by whether the beaconing timer is active, instead of the state of the LED itself for SMA queries Get(LedInfo) and Get(PortInfo). While we are at it, we fix the beaconing timer control code so that the state of the timer is accurately updated. Reviewed-by:
Ira Weiny <ira.weiny@intel.com> Reviewed-by:
-
Kaike Wan authored
This patch tests the interrupt registers when the driver has no access to its upstream component. In this case, it is highly likely that it is running in a virtual machine (eg, Qemu-kvm guest). If the interrupt registers are not mapped properly by the virtual machine monitor, an error message will be printed and the probing will be terminated. This will help the user identify the issue. On the other hand, if the driver is running in a host or has access to its upstream component in some other VM, it will do nothing. Reviewed-by:
-
Kaike Wan authored
When the hfi1 device is assigned to a VM (eg KVM), the hfi1 driver has no access to the upstream component and therefore cannot use it to perform some operations, such as secondary bus reset. As a result, the hfi1 driver cannot perform the pcie Gen3 transition. Instead, those operation should be done in the host environment, preferrably done during the Option ROM initialization. Similarly, the hfi1 driver cannot support ASPM and tune the pcie capability under this circumstance. Reviewed-by:
-
Jianxin Xiong authored
There is a header size counter in both the QP struture and the txreq structure. The counter in the txreq structure is not updated properly for RC and UC queue pairs with GRH enabled, and thus causing SDMA send to fail. This patch fixes the RC and UC path. Reviewed-by:
Jianxin Xiong <jianxin.xiong@intel.com> Signed-off-by:
-
Jubin John authored
The lkey_table_size driver specific parameter value is used before its value is sanity checked and restricted to RVT_MAX_LKEY_TABLE_BITS. This causes a vmalloc allocation failure for large values. Fix this by moving the value check before the first usage of the value. Reviewed-by:
-
Dean Luick authored
A cp or cat of /sys/kernel/debug/hfi1/hfi1_0/port1counters produces the following message: hfi1 0000:81:00.0: hfi1_0: index not supported hfi1 0000:81:00.0: hfi1_0: read_cntrs does not support indexing Fix by removing the file position logic and the associated messages and make the file positioning the responsibility of the caller. The port counter read function argument is changed to the per port data structure since the counters are relative to the port and not the device. Reviewed-by:
Sebastian Sanchez <sebastian.sanchez@intel.com> Signed-off-by:
-
Dean Luick authored
An attempt to cp or cat /sys/kernel/debug/hfi1/hfi1_0/i2c1 produces this message: hfi1 0000:81:00.0: hfi1_0: IB0:1 I2C failed even retrying Fix the issue by explicitly rejecting a simple cat/cp with an -EINVAL error return. Reviewed-by:
-
Mike Marciniszyn authored
The new check routine causes a larger than supported frame size on s390. Changing the check routine to noinline fixes the issue. Reviewed-by:
-
Dean Luick authored
Improve logging messages when there are i2c failures. Clean i2c read error handling. Reviewed-by:
-
Dean Luick authored
Debugfs access races with the driver being ready. Make sure the driver is ready before debugfs files appear and debufs files are gone before the driver starts tearing down. Reviewed-by:
-
Easwar Hariharan authored
This is a set of minor fixes including comment and log message cleanups and improvements to the PHY layer code. Reviewed-by:
-
Dean Luick authored
Count only the errors that apply to xmit discards. Update the comment to better explain the limitations of the count. Reviewed-by:
-
Mike Marciniszyn authored
Setting CONFIG_HFI1_DEBUG_SDMA_ORDER causes a syntax error: sdma.c: In function ‘complete_tx’: sdma.c:370: error: ‘txp’ undeclared (first use in this function) sdma.c:370: error: (Each undeclared identifier is reported only once sdma.c:370: error: for each function it appears in.) Adjust code under ifdef to reference the tx properly. Reviewed-by:
-
Jubin John authored
Fix the header by moving the copyright notice out of the license text and to the top of the header. Also, update the copyright date. Reviewed-by:
-
Jubin John authored
Remove else after break to fix checkpatch warning: WARNING: else is not generally useful after a break or return Reviewed-by:
-
Jubin John authored
Add braces on all arms of statements to fix checkpatch check: CHECK: braces {} should be used on all arms of this statement Reviewed-by: -
Jubin John authored
Fix code alignment to fix checkpatch check: CHECK: Alignment should match open parenthesis Reviewed-by:
-
Jubin John authored
Fix block comments with proper formatting to fix checkpatch warnings: WARNING: Block comments use * on subsequent lines WARNING: Block comments use a trailing */ on a separate line Reviewed-by:
-
Jubin John authored
Add comments describing the spinlock for spinlock_t definitions to fix checkpatch check: CHECK: spinlock_t definition without comment Reviewed-by:
-
Jubin John authored
Remove return statement at the end of a void function to fix checkpatch warning: WARNING: void function return statements are not generally useful Reviewed-by:
-
Jubin John authored
Use sizeof(*p) instead of sizeof(struct foo) to fix checkpatch check: CHECK: Prefer alloc(sizeof(*p)...) over alloc(sizeof(struct foo)...) Reviewed-by:
-
Jubin John authored
Remove CamelCase to fix checkpatch check: CHECK: Avoid CamelCase: <PLS_CONFIGPHY_ESTcOMM_LOCAL_COMPLETE> Reviewed-by:
-
