summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAge
* gpu: nvgpu: Delete more Linux includesAlex Waterman2018-02-15
| | | | | | | | | | | | | | Delete yet more Linux includes in gv11b code. Change-Id: I96f2c053cb903ac7cc51ad8cdb12f4a75ac95ae1 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1657742 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Delete unused Linux headers in gv11b_hal.cAlex Waterman2018-02-15
| | | | | | | | | | | | | Not sure how these even made it in in the first place... Change-Id: Ibca4d525926d4dacc7f8b41609dd147f14dd1a0d Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1657733 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: limit min freq to 216.75Mhzseshendra Gadagottu2018-02-15
| | | | | | | | | | | | | | | | | Until issue related to low frequencies root caused, limit min frequency to known safe value: 216.75Mhz. This change needs to be reverted, once orginal issue root-caused and fixed. Bug 2051863 Bug 2056266 Change-Id: If6e56f59ee5fa06967fde1128b58a7fc97be74e9 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1657595 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Remove the use of READ_ONLY for DMA APITerje Bergstrom2018-02-15
| | | | | | | | | | | | | | READ_ONLY flag for dma API is a Tegra specific API. We use it only to prevent accidental writes to non-secure ACR bootloader. Its use is marginal, so remove the flag. JIRA NVGPU-4 Change-Id: I887dc04aee8f7ace40220294851b210375dfde98 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1648174 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Use preallocated VPR bufferTerje Bergstrom2018-02-15
| | | | | | | | | | | | To prevent deadlock while allocating VPR in nvgpu, allocate all the needed VPR memory at probe time and use an internal allocator to hand out space for VPR buffers. Change-Id: I584b9a0f746d5d1dec021cdfbd6f26b4b92e4412 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1655324 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: delete nvgpu_semaphore_int listKonsta Holtta2018-02-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | The hw semas in a sema pool are stored in a list. All elements in this list are freed in a loop when a semaphore pool is destroyed. However, each hw sema is always owned by a channel, and each such channel frees its hw sema during channel closure before putting a ref to the VM which holds a ref to the sema pool, so the lifetime of all the hw semas is shorter than that of the pool and this list is always empty when freeing the pool. Delete the list and this freeing loop. Meanwhile delete also the nr_incrs member in nvgpu_semaphore_int that is never accessed. Jira NVGPU-512 Change-Id: Ie072029f9e7cc749141e9f02ef45fdf64358ad96 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1653540 Reviewed-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: Alex Waterman <alexw@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: update subctx pdb infoseshendra Gadagottu2018-02-13
| | | | | | | | | | | | | | | | | | | | | | | | | | Updated subctx pdb info for all instblks created. Earlier subctx pdb info was getting updated during instblk commit. But some instblks like pmu instblk are never committed. Missing subctx pdb info in instblk is creating issues accessing subctx info. So, by filling subctx pdb info during instblk creation fixed all these issues. Also as part of re-org of the function gv11b_init_subcontext_pdb, moved setting subctx info in ram_in_engine_wfi_veid_w() to channel_gv11b_setup_ramfc. Bug 2051863 Change-Id: Ida96118e8f86b638fa6a8586d026ad2617ebbf64 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1654678 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Alex Waterman <alexw@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: enable elpgseshendra Gadagottu2018-02-13
| | | | | | | | | | | | | | | | Enabled Engine Level Power Gating for gv11b. Bug 2051863 Change-Id: I59a51dbe8fa9f13e4b8be03f02e1571093fdaeb0 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1646322 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Use #define for log masksTerje Bergstrom2018-02-09
| | | | | | | | | | | | | | | | | | Log masks are a bitmask, and passed as u32 through the API calls. They were still defined as enums, which causes unnecessary implicit conversions. Convert the log masks to be defined as u32. JIRA NVGPU-52 Change-Id: I4b20f0ad2a9f18056502940ea677b3ea8526d830 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1649816 GVS: Gerrit_Virtual_Submit Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: call nvgpu_init_mm_vars just after probeSeema Khowala2018-02-09
| | | | | | | | | | | | | | | It is good to init mm vars right after probe as driver is heavily dependent on enabled flags for all kinds of memory related needs Change-Id: I62ca280ff9240649798faa34767f7dc9ea3c0db1 Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1649724 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: Fix CBC base calculusSami Kiminki2018-02-09
| | | | | | | | | | | | | | On GV11B, CBC base is calculated in similar fashion than it's calculated on dGPUs. Thus, remove gv11b_ltc_cbc_fix_config() as it would incorrectly multiply the CBC base by the LTC count. Bug 2054860 Change-Id: Iaed717161547468c17e12236149d970c497885b3 Signed-off-by: Sami Kiminki <skiminki@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1654506 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: handle pm_prepare_poweroff failureseshendra Gadagottu2018-02-09
| | | | | | | | | | | | | | As part of gk20a_pm_prepare_poweroff, gpu hw state is destroyed even in case of any errors. So try to recover from that situation by calling gk20a_pm_finalize_poweron. Bug 200380708 Change-Id: Ibff656cda67241ad111fd22701e05871f20d6f70 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1653750 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Rely on own dma attribute handlingTerje Bergstrom2018-02-08
| | | | | | | | | | | | | | Tegra kernel abstracts Linux 4.4 vs Linux 4.9 differences from drivers. Upstream kernel does not provide that facility, so add nvgpu internal way of dealing with the differences. JIRA NVGPU-4 Change-Id: I8289fdcf98873de14398bffc808d89a675f2aa15 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1648160 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add vpr flag in gpu characteristicsAparna Das2018-02-08
| | | | | | | | | | | | | | | | | | | VPR is currently not supported in virtualized configuration. Allow reporting VPR capability in gpu characteristics Jira EVLR-2236 Change-Id: Id61a0045577e4add0d9cdfddcefcedd5b20eb1dd Signed-off-by: Aparna Das <aparnad@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1639798 (cherry picked from commit 4136b74fd4435966ee2e69ec88fb66424382a7c0) Reviewed-on: https://git-master.nvidia.com/r/1640712 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add user API to get read-only syncpoint address mapDeepak Nibade2018-02-07
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add User space API NVGPU_AS_IOCTL_GET_SYNC_RO_MAP to get read-only syncpoint address map in user space We already map whole syncpoint shim to each address space with base address being vm->syncpt_ro_map_gpu_va This new API exposes this base GPU_VA address of syncpoint map, and unit size of each syncpoint to user space. User space can then calculate address of each syncpoint as syncpoint_address = base_gpu_va + (syncpoint_id * syncpoint_unit_size) Note that this syncpoint address is read_only, and should be only used for inserting semaphore acquires. Adding semaphore release with this address would result in MMU_FAULT Define new HAL g->ops.fifo.get_sync_ro_map and set this for all GPUs supported on Xavier SoC Bug 200327559 Change-Id: Ica0db48fc28fdd0ff2a5eb09574dac843dc5e4fd Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1649365 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add characteristic flag for syncpoint address supportDeepak Nibade2018-02-07
| | | | | | | | | | | | | | | | | | | | | Add characteristic flag NVGPU_GPU_FLAGS_SUPPORT_SYNCPOINT_ADDRESS to indicate if platform supports semaphore GPU_VA address for a syncpoint Define NVGPU_SUPPORT_SYNCPOINT_ADDRESS for core driver book keeping Set this flag for both GV100 and GV11B since Xavier SoC supports a semaphore GPU_VA address for a syncpoint through syncpoint SHIM Bug 200327559 Change-Id: I1f31673c9fd59f493d0b35a80d23151fc063ae06 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1649364 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add speculative load barrier (ctrl IOCTLs)Aparna Das2018-02-06
| | | | | | | | | | | | | | | | Data can be speculatively loaded from memory and stay in cache even when bound check fails. This can lead to unintended information disclosure via side-channel analysis. To mitigate this problem insert a speculation barrier. bug 2041355 Change-Id: I03fbc81c105632e788d8a0f97fb4e233d1124b49 Signed-off-by: Aparna Das <aparnad@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1652230 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: fix out of bounds accessDeepak Goyal2018-02-06
| | | | | | | | | | | | | | | | | | | lsf_ucode_desc_v1 has more size than signature bin. In memcpy(dest, src, size_to_copy) usage, "size_to_copy" is more than "size of the src" which is causing out of bounds access. Bug 2051856 NVGPU-507 Change-Id: I0aad34df39f95f7e95ccb10539e1fae9f65361a8 Signed-off-by: Deepak Goyal <dgoyal@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1650140 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: disable SWDX spill buffer invalidatesSami Kiminki2018-02-06
| | | | | | | | | | | | | | | | Disable SWDX spill buffer invalidates as is required by HW. Since this register is context-switched, add these in the GR init sequence. Bug 2040262 Change-Id: I0be10d12516bce6ce6f8fb0e8af5b67f8af92257 Signed-off-by: Sami Kiminki <skiminki@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1650563 Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: disable SCC pagepool invalidatesSami Kiminki2018-02-06
| | | | | | | | | | | | | | | | Disable SCC pagepool invalidates as is required by HW. Since this register is context-switched, add these in the GR init sequence. Bug 2040262 Change-Id: I8dd1b7c7c4b0544878ca57b1261f9c85fa380d47 Signed-off-by: Sami Kiminki <skiminki@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1649719 Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Build iGPU files only on TegraTerje Bergstrom2018-02-06
| | | | | | | | | | | | | | | | | Build files specific to iGPU in common/linux only when iGPU is supported. JIRa NVGPU-4 Change-Id: I227fc6fd41729088281d5c0c8cde5a963ba18f6a Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1648064 Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Konsta Holtta <kholtta@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: add scg support info in gpu characteristicsseshendra Gadagottu2018-02-02
| | | | | | | | | | | | | Indicated support for Simultaneous Compute and Graphics(SCG) in gpu characteristics for gv11b. Bug 2053932 Change-Id: I788e22242083dff775dd4cc5b9aa73c938028536 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1649805 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Cleanup usage of bypass_smmuAlex Waterman2018-02-02
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The GPU has multiple different operating modes in respect to IOMMU'ability. As such there needs to be a clean way to tell the driver whether it is IOMMU'able or not. This state also does not always reflect what is possible: all becasue the GPU can generate IOMMU'ed memory requests doesn't mean it wants to. The nvgpu_iommuable() API has now existed for a little while which is a useful way to convey whether nvgpu should consider the GPU as IOMMU'able. However, there is also the g->mm.bypass_smmu flag which used to be able to override what the GPU decided it should do. Typically it was assigned the same value as nvgpu_iommuable() but that was not necessarily a requirment. This patch removes all the usages of g->mm.bypass_smmu and instead uses the nvgpu_iommuable() function. All places where the check against g->mm.bypass_smmu have been replaced with nvgpu_iommuable(). The code should now be much cleaner. Subsequently other checks can also be placed in the nvgpu_iommuable() function. For example, when NVLINK comes online and the GPU should no longer consider DMA addresses and instead use scatter-gather lists directly the ngpu_iommuable() function will be able to check the state of NVLINK and then act accordingly. Change-Id: I0da6262386de15709decac89d63d3eecfec20cd7 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1648332 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Remove init_state initialization codeTejal Kudav2018-02-02
| | | | | | | | | | | | | | | | | | nvlink core library no longer exposes the set_init_state() interface as it wishes to block init_state changes from endpoint drivers. Now, the core driver is responsible for initializing init_state variables using set_init_state() interface. Hence, we remove this redundant code. Change-Id: I81c4922cf48f7918e69795579b39b7fa0c299644 Signed-off-by: Tejal Kudav <tkudav@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1646437 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Adeel Raza <araza@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Set DMA mask to 34 bitsAlex Waterman2018-02-01
| | | | | | | | | | | | | | | | | | Set the DMA mask to 34 bits so that large DMA allocs can be done. Currently the DMA mask is left unset which limits the size of the maximum DMA allocation to 32 bits. The 34 bit mask was chosen because it works for all chips (even gm20b supports 34 bit physical addresses). However, newer chips could use larger masks in the future if they desire. Bug 200377221 Change-Id: Iaa0543f77ff4e2bd6616f38e4464240375bb37b6 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1641762 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: enable devfreqseshendra Gadagottu2018-02-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | After moving devfreq enable to end of finalize power on, intermittent issues related to gpu booting with devfreq enabled are fixed. Enabled devfreq for gv11b by enabling ""nvhost_podgov" governor in platform data. Reused scaling functions from gp10b/gk20a. Removed emc floor on railgate for power saving. Added max emc frequency as floor in rail-ungate for faster gpu boot. Bug 2049965 Bug 2039013 Bug 200377508 Change-Id: Ia1dec278b663b9f7ed859dd953a60f3eae7ef9a0 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1644702 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Add tracking of dma_buf_attachmentTerje Bergstrom2018-02-01
| | | | | | | | | | | | | | | | | VM and CDE code assumes that dma_buf_attachment is stored as a pointer in the private dma_buf_drvdata, so it is not tracked. In Linux trees without dma_buf_*_drvdata() support this is not true, so change the code to explicitly track dma_buf_attachment. JIRA NVGPU-4 Change-Id: I692f05a19a6469195d5444a7e5ff6e92f77ae272 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1648004 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Remove NVGPU_IOCTL_GET_BUFFER_INFOTerje Bergstrom2018-01-31
| | | | | | | | | | | | | | | | | | | The IOCTL was introduced for making efficient query of buffer identity and size. It was never taken into use, and it adds a dependency to Tegra specific dma_buf API, so remove it. JIRA NVGPU-4 Change-Id: I194d7bb1f54997900a3be8d39c93331befa225c7 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1648001 Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Konsta Holtta <kholtta@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: enable devfreq for silicon onlyseshendra Gadagottu2018-01-31
| | | | | | | | | | | | | | | | | | | | gpu frequency scaling is available only on silicon platforms. Added check for silicon platform before enabling scaling init. Bug 2049965 Bug 2039013 Bug 200377508 Change-Id: Ie780147cee904137e4618e17162e5cedba4987ee Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1642529 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: enable devfreq after finalize poweronseshendra Gadagottu2018-01-31
| | | | | | | | | | | | | | | | | | | Enabling gpu scaling driver after finalize poweron, will make gpu booting happen at initially set frequency(1GHz). Also doing platform specific init scale after enabling scaling driver. Bug 2049965 Bug 2039013 Bug 200377508 Change-Id: I633f8f5a25d9de18cbb3a022913b8b725ccd87e5 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1644703 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: enable more gr exceptionsSeema Khowala2018-01-31
| | | | | | | | | | | | | | | | -pd, scc, ds, ssync, mme and sked exceptions are enabled. This will be useful for debugging -Handle enabled interrupts -Add gr ops to handle ssync hww. For legacy chips, ssync hww_esr register is gpcs_ppcs_ssync_hww_esr. Since ssync hww is not enabled on legacy chips, added ssync hww exception handling for volta only. Change-Id: I63ba2eb51fa82e74832df26ee4cf3546458e5669 Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1644751 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add fecs_host_int_enable halSeema Khowala2018-01-31
| | | | | | | | | | | This will be used to enable fecs interrupts per chip. Change-Id: Id99412ca1a9c4caad999c3458b0e9701515db4b9 Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1642554 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: disable cbm alpha/beta cache invalidatesseshendra Gadagottu2018-01-31
| | | | | | | | | | | | | | | | | | | | | Disabled CBM alpha and beta cache invalidates as required by hw. Since these registers are context switched out, added these invalidates as part of gr init sequence, so golden context restore these settings for all contexts. Bug 2040262 Change-Id: Iffdd03f2ac6440ddd615899c407cfee692460918 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1648948 Reviewed-by: Sami Kiminki <skiminki@nvidia.com> Tested-by: Sami Kiminki <skiminki@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Tested-by: Chris Dragan <kdragan@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Unify querying stream idTerje Bergstrom2018-01-31
| | | | | | | | | | | | | | Stream ID for gp10b is retrieved directly from DT headers in common code. Introduce instead a variable to store the stream ID and move the query to platform_gp10b_tegra.c. JIRA NVGPU-4 Change-Id: I123024e13e470283bb691883f8f963eb72c997d8 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1648013 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: vgpu: fix resource leakAingara Paramakuru2018-01-31
| | | | | | | | | | | | | gr_ctx->tsgid needs to be set to ensure that the GR ctx free sequence will target the correct TSG's GR ctx. Bug 200341631 Change-Id: I83c57597f10ce3af572f114d28312376cea55c2a Signed-off-by: Aingara Paramakuru <aparamakuru@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1646790 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* scripts: rfr: support https://git-master.nvidia.com/r/#/c/(\d+)Richard Zhao2018-01-31
| | | | | | | | | | | | | | | | | https://git-master.nvidia.com/r/#/c/(\d+) is acceptable format for gerrit website and it won't auto-correct by adding '/' at endof the url. Jira EVLR-2364 Change-Id: I1a33578058d23fbb3eaf76d260a1a659c628fdc3 Signed-off-by: Richard Zhao <rizhao@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1646315 Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: enable clock gating featuresseshendra Gadagottu2018-01-31
| | | | | | | | | | | | | | | Enable ELCG, BLCG and SLCG features. Bug 2051863 Change-Id: Id2c67c94c7b2dd0517d4ee4b0280aeb19f3fe35a Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1646302 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: vgpu: add vgpu_ivc_* wrappersRichard Zhao2018-01-31
| | | | | | | | | | | | | | | | | tegra_gr_comm_* are wrapped as vgpu_ivc_*, which helps make vgpu code more common. Jira EVLR-2364 Change-Id: Id49462ed6c176c73ceee8c6bc41104447748e187 Signed-off-by: Richard Zhao <rizhao@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1645656 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Aingara Paramakuru <aparamakuru@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: make .tsg_unbind_channel one layer lowerRichard Zhao2018-01-31
| | | | | | | | | | | | | | | | | | | | | | | | | The message to tell RM server to unbind channel has to be sent after client unbinds the channel and before client calls tsg release. The channel has to belong to a tsg on RM server before client submit a runlist to remove the channel. Or there's a bare channel problem. By moving .tsg_unbind_channl one layer lower, gk20a_tsg_unbind_channel() will be common functions for all chip, and it'll call tsg release after call .tsg_unbind_channel. So vgpu won't need to worry about tsg was released before sending msg to RM server. Bug 200382695 Bug 200382785 Change-Id: I32acc122f3f9d5d0628049ccf673225f9e90c87a Signed-off-by: Richard Zhao <rizhao@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1645383 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: cause early VPR resize for gv11bDeepak Nibade2018-01-30
| | | | | | | | | | | | | | | | | | | | | | | | | | | Patch 7240b3c2 enabled secure allocation for gv11b But since we allocate secure buffers in poweron path, and secure allocation needs GPU to be in off state, this results in deadlock in poweron path To solve this, we already cause early VPR resize for older chips by calling gk20a_tegra_secure_page_alloc() from late_probe Implement same for gv11b. Add late_probe callback and add a call to gk20a_tegra_secure_page_alloc() Bug 2038249 Change-Id: I8c17b069962b26edbd0639a7c0d6c2fdaa352935 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1648831 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Tested-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Seema Khowala <seemaj@nvidia.com>
* gpu: nvgpu: Update gv10x regops whitelistThomas Fleury2018-01-27
| | | | | | | | | | | | | | | Update regops whitelists for GV100 Bug 2046928 Change-Id: If07450dd43f7c7e5177831a645460644c7d86018 Signed-off-by: Thomas Fleury <tfleury@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1640386 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: vgpu: remove virt_ctx from tegra_gr_commRichard Zhao2018-01-26
| | | | | | | | | | | | | | | | | | | queue index can already index the queues. It also help make the api more common. Jira EVLR-2364 Change-Id: I98a5014ba0510a2687fdf096a160c497bd1f6985 Signed-off-by: Richard Zhao <rizhao@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1646197 Reviewed-by: Damian Halas <dhalas@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Aingara Paramakuru <aparamakuru@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: Nirav Patel <nipatel@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: ce: store fences in a separate arrayKonsta Holtta2018-01-26
| | | | | | | | | | | | | | | | | | | | Simplify the copyengine code massively by storing the job post fence pointers in an array of fences instead of mixing them up in the command buffer memory. The post fences are used when the ring buffer of a context gets full and we need to wait for the oldest slot to free up. NVGPU-43 NVGPU-52 Change-Id: I36969e19676bec0f38de9a6357767a8d5cbcd329 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1646037 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Alex Waterman <alexw@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: ce: drop prefence supportKonsta Holtta2018-01-26
| | | | | | | | | | | | | | | | | Delete the gk20a_fence_in argument in gk20a_ce_execute_ops. It has never been used and is in the way of some upcoming code cleanup. NVGPU-43 Change-Id: Ie61e1a2f4945b1e34d64880044c265d26fa822d7 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1646036 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Alex Waterman <alexw@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: enable IO coherence characteristics for dGPUsDeepak Nibade2018-01-26
| | | | | | | | | | | | | Enable NVGPU_SUPPORT_IO_COHERENCE characteristics for dGPUs which support DMA_COHERENCE e.g. GV100 Bug 200383034 Change-Id: If12d2ef6c642f7c4cce83dbf05f492100ee1c7e0 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1644277 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: dump more mmu fault infoSeema Khowala2018-01-26
| | | | | | | | | | | | | | | | | | | | | -Dump client type/id and protected mode as error prints. This will help figuring out which client is causing mmu fault. -Removed extra print for unbound instance block fault as it is already printed as fault type -Changed few extra prints from info prints to prints protected by gpu_dbg_intr Change-Id: I9e87e2a701372b47200f85149e040176365bd71c Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1643817 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Initial Nvlink driver skeletonDavid Nieto2018-01-25
| | | | | | | | | | | | | | | | | | | | | | Adds the skeleton and integration of the GV100 endpoint driver to NVGPU (1) Adds a OS abstraction layer for the internal nvlink structure. (2) Adds linux specific integration with Nvlink core driver. (3) Adds function pointers for nvlink api, initialization and isr process. (4) Adds initial support for minion. (5) Adds new GPU enable properties to handle NVLINK presence (6) Adds new GPU enable properties for SG_PHY bypass (required for NVLINK over PCI) (7) Adds parsing of nvlink vbios structures. (8) Adds logging defines for NVGPU JIRA: EVLR-2328 Change-Id: I0720a165a15c7187892c8c1a0662ec598354ac06 Signed-off-by: David Nieto <dmartineznie@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1644708 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add speculative load barrier (sched IOCTLs)Alex Waterman2018-01-25
| | | | | | | | | | | | | | | | | | | | | | Data can be speculatively loaded from memory and stay in cache even when bound check fails. This can lead to unintended information disclosure via side-channel analysis. To mitigate this problem insert a speculation barrier. bug 2039126 CVE-2017-5753 Change-Id: Iec23eb75ce2a9251c8a5c8cbdd21a32910e1a71a Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1640502 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Richard Zhao <rizhao@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add speculative load barrier (dbg IOCTLs)Alex Waterman2018-01-25
| | | | | | | | | | | | | | | | | Data can be speculatively loaded from memory and stay in cache even when bound check fails. This can lead to unintended information disclosure via side-channel analysis. To mitigate this problem insert a speculation barrier. bug 2039126 CVE-2017-5753 Change-Id: I982225e754cc5d430c19f4cc542302e52243bd38 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1640501 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add speculative load barrier (ctrl IOCTLs)Alex Waterman2018-01-25
| | | | | | | | | | | | | | | | | Data can be speculatively loaded from memory and stay in cache even when bound check fails. This can lead to unintended information disclosure via side-channel analysis. To mitigate this problem insert a speculation barrier. bug 2039126 CVE-2017-5753 Change-Id: Ib6c4b2f99b85af3119cce3882fe35ab47509c76f Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1640500 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
generated by cgit v1.2.2 (git 2.25.0) at 2024-11-05 11:02:29 -0500