summaryrefslogtreecommitdiff
path: root/drivers
AgeCommit message (Collapse)Author
2017-04-26drm/i915: Apply a cond_resched() to the saturated signalerChris Wilson
If the engine is continually completing nops, we can saturate the signaler and keep it working indefinitely. This angers the NMI watchdog! A good example is to disable semaphores on snb and run igt/gem_exec_nop - the parallel, multi-engine workloads are more than sufficient to hog the CPU, preventing the system from even processing ICMP echo replies. v2: Tvrtko dug into cond_resched() on x86 and found that it only depended upon preempt_count and not tif_need_resched() - which means that we would always call schedule() at that point. Fixes: c81d46138da6 ("drm/i915: Convert trace-irq to the breadcrumb waiter") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20170404120531.10737-1-chris@chris-wilson.co.uk Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> (cherry picked from commit a7980a640cbd339aa80f406d1786a275a2c320bc) Signed-off-by: Jani Nikula <jani.nikula@intel.com>
2017-04-26drm/i915: Park the signaler before sleepingChris Wilson
If the signal to park arrives before we sleep, then we need to check kthread_should_park() before sleeping to avoid missing the signal. Otherwise, if the signal arrives whilst we are processing completed requests, we will reset the current->state back to TASK_INTERRUPTIBLE and so miss the wakeup. Fixes: fe3288b5da2c ("drm/i915: Park the breadcrumbs signaler across a GPU reset") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20170403105124.8969-1-chris@chris-wilson.co.uk Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> (cherry picked from commit b1becb88268beb72df6495e35d3d76c138d215bb) Signed-off-by: Jani Nikula <jani.nikula@intel.com>
2017-04-26s390/trng: Introduce s390 TRNG device driver.Harald Freudenberger
This patch introduces a new device driver s390-trng for the s390 platform which exploits the new PRNO TRNG cpacf subfunction. The true-random-number-generator is accessible from userspace, by default visible as /dev/trng. The driver also registers at the kernel build-in hwrng API to feed the hwrng with fresh entropy data. This generic device driver for hardware random data is visible from userspace as /dev/hwrng. Signed-off-by: Harald Freudenberger <freude@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2017-04-26Merge tag 'iwlwifi-next-for-kalle-2017-04-26' of ↵Kalle Valo
git://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/iwlwifi-next One more pull-request intended for 4.12. These are the changes: * The firmware for 7265D and 3168 NICs is frozen at version 29; * Sari continues working heavily on support for A000 series; * A bunch of fixes; * Some cleanups here and there;
2017-04-26iommu/arm-smmu: Return IOVA in iova_to_phys when SMMU is bypassedSunil Goutham
For software initiated address translation, when domain type is IOMMU_DOMAIN_IDENTITY i.e SMMU is bypassed, mimic HW behavior i.e return the same IOVA as translated address. This patch is an extension to Will Deacon's patchset "Implement SMMU passthrough using the default domain". Signed-off-by: Sunil Goutham <sgoutham@cavium.com> Acked-by: Will Deacon <will.deacon@arm.com> Signed-off-by: Joerg Roedel <jroedel@suse.de>
2017-04-26goldfish_pipe: fix build warning about using too much stack.Greg Kroah-Hartman
The new goldfish_pipe code used too much stack space in the transfer_max_buffers() call. As the function is serialized with a lock, let's make the buffer static to not use the stack for the large buffer. Reported-by: kbuild test robot <fengguang.wu@intel.com> Cc: Yurii Zubrytskyi <zyy@google.com> Cc: Jin Qian <jinqian@android.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-04-26goldfish_pipe: An implementation of more parallel pipeJin Qian
This is a driver code for a redesigned android pipe. Currently it works for x86 and x64 emulators with the following performance results: ADB push to /dev/null, Ubuntu, 400 MB file, times are for (1 / 10 / 100) parallel adb commands x86 adb push: (4.4s / 11.5s / 2m10s) -> (2.8s / 6s / 51s) x64 adb push: (7s / 15s / (too long, 6m+) -> (2.7s / 6.2s / 52s) ADB pull and push to /data/ have the same %% of speedup More importantly, I don't see any signs of slowdowns when run in parallel with Antutu benchmark, so it is definitely making much better job at multithreading. Signed-off-by: Yurii Zubrytskyi <zyy@google.com> Signed-off-by: Jin Qian <jinqian@android.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-04-26fpga fr br: update supported version numbersMatthew Gerlach
The value in the version register of the altera freeze bridge controller changed from the beta value of 2 to the value of 0xad000003 in the official release of the IP. This patch supports the old and new version numbers, and the driver's probe function will fail if neither of the supported versions is found. Signed-off-by: Matthew Gerlach <matthew.gerlach@linux.intel.com> Reviewed-by: Moritz Fischer <mdf@kernel.org> Signed-off-by: Alan Tull <atull@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-04-26fpga: region: release FPGA region reference in error pathTobias Klauser
If fpga_region_get_manager() fails in fpga_region_program_fpga(), a reference to the fpga_manager instance previously acquired through fpga_region_get() is retained. Make sure to properly release it in the error case by using a separate jump label which will call fpga_region_put() in before returning. Signed-off-by: Tobias Klauser <tklauser@distanz.ch> Acked-by: Moritz Fischer <mdf@kernel.org> Signed-off-by: Alan Tull <atull@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-04-26fpga altera-hps2fpga: disable/unprepare clock on error in ↵Tobias Klauser
alt_fpga_bridge_probe() If either _alt_hps2fpga_enable_set() or fpga_bridge_register() fail in alt_fpga_bridge_probe(), the clock remains enabled and prepared. Also, in the error path for _alt_hps2fpga_enable_set() a call to fpga_bridge_unregister() is made even though the bridge was not registered yet. Remove the unnecessary call to fpga_bridge_unregister() and call clk_disable_unprepare() in both error paths in order to make sure the clock gets properly disabled and unprepared. Signed-off-by: Tobias Klauser <tklauser@distanz.ch> Acked-by: Moritz Fischer <mdf@kernel.org> Signed-off-by: Alan Tull <atull@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-04-26usb: musb: don't mark of_dev_auxdata as initdataArnd Bergmann
The probe function is not __init since it can be called for deferred probing or when unbinding/rebinding the device, and therefore it must not reference objects in __initdata, as pointed out by this link time warning: WARNING: drivers/usb/musb/da8xx.o(.text+0x9d4): Section mismatch in reference from the function da8xx_probe() to the (unknown reference) .init.data:(unknown) This removes the annotation. Reported-by: Olof's autobuilder <olof@lixom.net> Fixes: d6299b6efbf6 ("usb: musb: Add support of CPPI 4.1 DMA controller to DA8xx") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-04-26usb: misc: legousbtower: Fix buffers on stackMaksim Salau
Allocate buffers on HEAP instead of STACK for local structures that are to be received using usb_control_msg(). Signed-off-by: Maksim Salau <maksim.salau@gmail.com> Tested-by: Alfredo Rafael Vicente Boix <alviboi@gmail.com>; Cc: stable <stable@vger.kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-04-26Merge tag 'gvt-next-fixes-2017-04-20' of https://github.com/01org/gvt-linux ↵Jani Nikula
into drm-intel-next-fixes gvt-next-fixes-2017-04-20 - some code optimization from Changbin - debug message cleanup after QoS merge - misc fixes for display mmio init, reset vgpu warning, etc. Signed-off-by: Jani Nikula <jani.nikula@intel.com>
2017-04-26brcmfmac: Make skb header writable before useJames Hughes
The driver was making changes to the skb_header without ensuring it was writable (i.e. uncloned). This patch also removes some boiler plate header size checking/adjustment code as that is also handled by the skb_cow_header function used to make header writable. Signed-off-by: James Hughes <james.hughes@raspberrypi.org> Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-04-26brcmfmac: Ensure pointer correctly set if skb data location changesJames Hughes
The incoming skb header may be resized if header space is insufficient, which might change the data adddress in the skb. Ensure that a cached pointer to that data is correctly set by moving assignment to after any possible changes. Signed-off-by: James Hughes <james.hughes@raspberrypi.org> Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-04-26rndis_wlan: add return value validationPan Bian
Function create_singlethread_workqueue() will return a NULL pointer if there is no enough memory, and its return value should be validated before using. However, in function rndis_wlan_bind(), its return value is not checked. This may cause NULL dereference bugs. This patch fixes it. Signed-off-by: Pan Bian <bianpan2016@163.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-04-26libertas: check return value of alloc_workqueuePan Bian
Function alloc_workqueue() will return a NULL pointer if there is no enough memory, and its return value should be validated before using. However, in function if_spi_probe(), its return value is not checked. This may result in a NULL dereference bug. This patch fixes the bug. Signed-off-by: Pan Bian <bianpan2016@163.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-04-26mt7601u: check return value of alloc_skbPan Bian
Function alloc_skb() will return a NULL pointer if there is no enough memory. However, in function mt7601u_mcu_msg_alloc(), its return value is not validated before it is used. This patch fixes it. Signed-off-by: Pan Bian <bianpan2016@163.com> Acked-by: Jakub Kicinski <kubakici@wp.pl> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-04-26orinoco_usb: Fix buffer on stackMaksim Salau
Allocate buffer on HEAP instead of STACK for a local variable that is to be sent using usb_control_msg(). Signed-off-by: Maksim Salau <maksim.salau@gmail.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-04-26orinoco: fix spelling mistake: "Registerred" -> "Registered"Colin Ian King
trivial fix to spelling mistake in dbg_dbg message Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-04-26rtlwifi: rtl8821ae: setup 8812ae RFE according to device typeLarry Finger
Current channel switch implementation sets 8812ae RFE reg value assuming that device always has type 2. Extend possible RFE types set and write corresponding reg values. Source for new code is http://dlcdnet.asus.com/pub/ASUS/wireless/PCE-AC51/DR_PCE_AC51_20232801152016.zip Signed-off-by: Maxim Samoylov <max7255@gmail.com> Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Cc: Stable <stable@vger.kernel.org> Cc: Yan-Hsuan Chuang <yhchuang@realtek.com> Cc: Pkshih <pkshih@realtek.com> Cc: Birming Chiu <birming@realtek.com> Cc: Shaofu <shaofu@realtek.com> Cc: Steven Ting <steventing@realtek.com> Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
2017-04-26HID: usbhid: Add HID_QUIRK_NOGET for Aten CS-1758 KVM switchVasilis Liaskovitis
Like other switches, the Aten CS-1758 KVM switch needs a quirk to avoid spewing errors: [12599018.071059] usb 5-2: input irq status -75 received [12599018.079053] usb 5-2: input irq status -75 received Signed-off-by: Vasilis Liaskovitis <vliaskovitis@suse.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2017-04-26dmaengine: virt-dma: Convert to use list_for_each_entry_safe()Andy Shevchenko
Use list_for_each_entry_safe() instead of open coding variants. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
2017-04-25x86, dax, pmem: remove indirection around memcpy_from_pmem()Dan Williams
memcpy_from_pmem() maps directly to memcpy_mcsafe(). The wrapper serves no real benefit aside from affording a more generic function name than the x86-specific 'mcsafe'. However this would not be the first time that x86 terminology leaked into the global namespace. For lack of better name, just use memcpy_mcsafe() directly. This conversion also catches a place where we should have been using plain memcpy, acpi_nfit_blk_single_io(). Cc: <x86@kernel.org> Cc: Jan Kara <jack@suse.cz> Cc: Jeff Moyer <jmoyer@redhat.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Christoph Hellwig <hch@lst.de> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Matthew Wilcox <mawilcox@microsoft.com> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Acked-by: Tony Luck <tony.luck@intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2017-04-25block: remove block_device_operations ->direct_access()Dan Williams
Now that all the producers and consumers of dax interfaces have been converted to using dax_operations on a dax_device, remove the block device direct_access enabling. Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2017-04-25dm: teach dm-targets to use a dax_device + dax_operationsDan Williams
Arrange for dm to lookup the dax services available from member devices. Update the dax-capable targets, linear and stripe, to route dax operations to the underlying device. Changes the target-internal ->direct_access() method to more closely align with the dax_operations ->direct_access() calling convention. Cc: Toshi Kani <toshi.kani@hpe.com> Reviewed-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2017-04-25dm crypt: remove obsolete references to per-CPU stateEric Biggers
dm-crypt used to use separate crypto transforms for each CPU, but this is no longer the case. To avoid confusion, fix up obsolete comments and rename setup_essiv_cpu(). Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Mike Snitzer <snitzer@redhat.com>
2017-04-25iwlwifi: adjust NVM parsing APIs for new a000 methodSara Sharon
In a000 devices we will get all nvm data from the firmware, and can save most of the parsing. Export two APIs that op mode will still use. Adjust API of init_sbands to be independent of NVM file structure so it can be used by op mode as well. Signed-off-by: Sara Sharon <sara.sharon@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-04-25iwlwifi: pcie: apply no-reclaim logic only to group 0Johannes Berg
When applying no-reclaim logic to commands other than the group zero for legacy commands, commands such as 0x1c (TX_CMD in group 0) can't be used in any other group. Fix that by applying this logic only for group 0 - it's not and should never be needed for any other groups. Reported-by: Sharon Dvir <sharon.dvir@intel.com> Reported-by: Shaul Triebitz <shaul.triebitz@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-04-25iwlwifi: mvm: memset binding before setting valuesSara Sharon
The changes in commit 9415af7f306b ("iwlwifi: mvm: support new binding API") assigned values that were later memset to 0. Move the memset earlier. Fixes: 9415af7f306b ("iwlwifi: mvm: support new binding API") Signed-off-by: Sara Sharon <sara.sharon@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-04-25iwlwifi: rename wait_for_tx_queues_emptySara Sharon
Rename current wait_tx_queue_empty to wait_tx_queues_empty since it waits for multiple queues (up to 32). Next patch will add a wait for single TX queue which is needed for gen2 to be scalable for 512. Signed-off-by: Sara Sharon <sara.sharon@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-04-25iwlwifi: move to 512 queuesSara Sharon
Avoid using the old define since it will enlarge necessary structs for previous HW. Signed-off-by: Sara Sharon <sara.sharon@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-04-25iwlwifi: mvm: support station type APISara Sharon
Support change to ADD_STA API to support station types. Each station is assigned its type. This simplifies FW handling of the broadcast and multicast stations: * broadcast station is identified by its type and not the mac address. * multicast queue is no longer treated differently. The opening and closing of it is done by referring to its station. There is no need to specify it in the MAC command. * When disabling TX to all station driver can disable the traffic on multicast station, so FW doesn't have to do it. Change is backward compatible. Change the order of adding and removing the stations according to FW requirements. Signed-off-by: Sara Sharon <sara.sharon@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-04-25iwlwifi: mvm: remove references to queue_info in new TX pathSara Sharon
Most of the fields aren't needed in new TX path. Enlarging the struct to 512 queues will consume a lot of memory. Remove all references to the struct in the new TX path. Move mac80211 queue mapping outside, since it will be needed per queue for TVQM mode. Add warning in paths that shouldn't be hit. Signed-off-by: Sara Sharon <sara.sharon@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-04-25iwlwifi: gen2: support nmi triggering from hostLiad Kaufman
For gen2 there is a new register. Signed-off-by: Liad Kaufman <liad.kaufman@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-04-25iwlwifi: remove module loading failure messageJohannes Berg
When CONFIG_DEBUG_TEST_DRIVER_REMOVE is set, iwlwifi crashes when the opmode module cannot be loaded, due to completing the completion before using drv->dev, which can then already be freed. Fix this by removing the (fairly useless) message. Moving the completion later causes a deadlock instead, so that's not an option. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-04-25iwlwifi: don't leak memory on allocation failureJohannes Berg
If we fail to allocate the small chunk of memory for the pieces of the firmware file, we leak the whole firmware image instead... Since the allocation failure is really unlikely, just bail out at that point instead. Remove the error message at the label since we now (and actually have been) use it for various reasons. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-04-25iwlwifi: pcie: remove superfluous trans->dev assignmentJohannes Berg
This struct member is already assigned in the previous call to iwl_trans_alloc(), so assigning the same value again is superfluous - remove it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-04-25iwlwifi: mvm: use defines instead of variables for shared dwell timesSara Sharon
Most of the dwells are constant across different scan types. Use defines instead of depending on scan type. This is needed as preparation to having different scan type per band. Signed-off-by: Sara Sharon <sara.sharon@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-04-25iwlwifi: mvm: remove color definitionSara Sharon
We use the full station field as sta_id, and never use the color. FW are about to clean the color out, so those defines are incorrect now (and were redundant before). Signed-off-by: Sara Sharon <sara.sharon@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-04-25iwlwifi: mvm: move internally to use bigger INVALID_TXQSara Sharon
We can't use IEEE80211_INVAL_HW_QUEUE to mark a queue as invalid since 255 will be a valid value for a TVQM queue index. Use IWL_MVM_INVALID_QUEUE instead for accessing txq_id. reserved_queue can stay a u8 since reserved_queue is not used when TVQM is enabled. Signed-off-by: Sara Sharon <sara.sharon@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
2017-04-25IB/mlx5: Add ODP support to MWArtemy Kovalyov
Internally MW implemented as KLM MKey and filled by userspace UMR postsends. Handle pagefault trigered by operations on this MKeys. Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by: Leon Romanovsky <leon@kernel.org> Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-04-25IB/mlx5: Extract page fault codeArtemy Kovalyov
To make page fault handling code more flexible split pagefault_single_data_segment() function. Keep MR resolution in pagefault_single_data_segment() and move actual updates into pagefault_single_mr(). Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by: Leon Romanovsky <leon@kernel.org> Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-04-25IB/umem: Add support to huge ODPArtemy Kovalyov
Add IB_ACCESS_HUGETLB ib_reg_mr flag. Hugetlb region registered with this flag will use single translation entry per huge page. Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by: Leon Romanovsky <leon@kernel.org> Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-04-25IB/mlx5: Add contiguous ODP supportArtemy Kovalyov
Currenlty ODP supports only regular MMU pages. Add ODP support for regions consisting of physically contiguous chunks of arbitrary order (huge pages for instance) to improve performance. Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by: Leon Romanovsky <leon@kernel.org> Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-04-25IB/umem: Add contiguous ODP supportArtemy Kovalyov
Currenlty ODP supports only regular MMU pages. Add ODP support for regions consisting of physically contiguous chunks of arbitrary order (huge pages for instance) to improve performance. Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by: Leon Romanovsky <leon@kernel.org> Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-04-25IB/mlx5: Decrease verbosity level of ODP errorsArtemy Kovalyov
Decrease verbosity level of ODP error flows messages to debug level. Remove one redundant print since debug level message already exists in this flow. Fixes: d9aaed838765 ('{net,IB}/mlx5: Refactor page fault handling') Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by: Leon Romanovsky <leon@kernel.org> Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-04-25IB/mlx5: Fix implicit MR GCArtemy Kovalyov
When implicit MR's leaf MKey becomes unused, i.e. when it's last page being released my MMU invalidation it is marked as "dying" and scheduled for release by garbage collector. Currentle consequent page fault may remove "dying" flag. Treat leaf MKey as non-existent once it was scheduled to removal by GC. Fixes: 81713d3788d2 ('IB/mlx5: Add implicit MR support') Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by: Leon Romanovsky <leon@kernel.org> Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-04-25IB/mlx5: Fix UMR size calculationArtemy Kovalyov
Translation table updates of large UMR may require multiple post send operations. The last operations can be in various lengths, but current code set them to be the same length. Fixes: 7d0cc6edcc70 ('IB/mlx5: Add MR cache for large UMR regions') Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by: Leon Romanovsky <leon@kernel.org> Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-04-25IB/mlx5: Fix function updating xlt emergency pathArtemy Kovalyov
In memory shortage path we fall back to use spare buffer. mlx5_ib_update_xlt() called from ib_uverbs_reg_mr when ibmr.ucontext not initialized yet. Scenario how to test it: 1. trigger memory exhaustion so __get_free_pages(GFP_KERNEL, 4) will fail 2. register MR 3. there should be no kernel oops Fixes: 7d0cc6edcc70 ('IB/mlx5: Add MR cache for large UMR regions') Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by: Leon Romanovsky <leon@kernel.org> Signed-off-by: Doug Ledford <dledford@redhat.com>
generated by cgit (git 2.34.1) at 2025-09-18 06:24:06 +0000