Changes in lima_device.c

a78422e9 - drm/sched: implement dynamic job-flow control

Fri, 10 Nov 2023 00:16:33 +0000

drm/sched: implement dynamic job-flow controlCurrently, job flow control is implemented simply by limiting the numberof jobs in flight. Therefore, a scheduler is initialized with a creditlimit that corresponds to the number of jobs which can be sent to thehardware.This implies that for each job, drivers need to account for the maximumjob size possible in order to not overflow the ring buffer.However, there are drivers, such as Nouveau, where the job size has arather large range. For such drivers it can easily happen that jobsubmissions not even filling the ring by 1% can block subsequentsubmissions, which, in the worst case, can lead to the ring run dry.In order to overcome this issue, allow for tracking the actual job sizeinstead of the number of jobs. Therefore, add a field to track a job'scredit count, which represents the number of credits a job contributesto the scheduler's credit limit.Signed-off-by: Danilo Krummrich Reviewed-by: Luben Tuikov Link: https://patchwork.freedesktop.org/patch/msgid/20231110001638.71750-1-dakr@redhat.com List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

8581fd40 - treewide: Add missing includes masked by cgroup -> bpf dependency

Thu, 02 Dec 2021 20:34:00 +0000

treewide: Add missing includes masked by cgroup -> bpf dependencycgroup.h (therefore swap.h, therefore half of the universe)includes bpf.h which in turn includes module.h and slab.h.Since we're about to get rid of that dependency we needto clean things up.v2: drop the cpu.h include from cacheinfo.h, it's not necessaryand it makes riscv sensitive to ordering of include files.Signed-off-by: Jakub Kicinski Signed-off-by: Alexei Starovoitov Reviewed-by: Christoph Hellwig Acked-by: Krzysztof Wilczyński Acked-by: Peter Chen Acked-by: SeongJae Park Acked-by: Jani Nikula Acked-by: Greg Kroah-Hartman Link: https://lore.kernel.org/all/20211120035253.72074-1-kuba@kernel.org/ # v1Link: https://lore.kernel.org/all/20211120165528.197359-1-kuba@kernel.org/ # cacheinfo discussionLink: https://lore.kernel.org/bpf/20211202203400.1208663-1-kuba@kernel.org List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

89636a06 - drm/lima: fix warning when CONFIG_DEBUG_SG=y & CONFIG_DMA_API_DEBUG=y

Sun, 31 Oct 2021 04:16:04 +0000

drm/lima: fix warning when CONFIG_DEBUG_SG=y & CONFIG_DMA_API_DEBUG=yOtherwise get following warning:DMA-API: lima 1c40000.gpu: mapping sg segment longer than device claims to support [len=4149248] [max=65536]See: https://gitlab.freedesktop.org/mesa/mesa/-/issues/5496Reviewed-by: Vasily Khoruzhick Reported-by: Roman Stratiienko Signed-off-by: Qiang Yu Link: https://patchwork.freedesktop.org/patch/msgid/20211031041604.187216-1-yuq825@gmail.com List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

63945d51 - drm/lima: add pm resume/suspend ops

Tue, 21 Apr 2020 13:35:50 +0000

drm/lima: add pm resume/suspend opsAdd driver pm system and runtime hardware resume/suspend ops.Note this won't enable runtime pm of the device yet.v2:Do clock and power gating when suspend/resume.Tested-by: Bhushan Shah Reviewed-by: Vasily Khoruzhick Signed-off-by: Qiang Yu Link: https://patchwork.freedesktop.org/patch/msgid/20200421133551.31481-10-yuq825@gmail.com List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

7a475eb4 - drm/lima: separate clk/regulator enable/disable function

Tue, 21 Apr 2020 13:35:49 +0000

drm/lima: separate clk/regulator enable/disable functionFor being used by both device init/fini and suspend/resume.Tested-by: Bhushan Shah Reviewed-by: Vasily Khoruzhick Signed-off-by: Qiang Yu Link: https://patchwork.freedesktop.org/patch/msgid/20200421133551.31481-9-yuq825@gmail.com List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

3446d7e9 - drm/lima: add resume/suspend callback for each ip

Tue, 21 Apr 2020 13:35:48 +0000

drm/lima: add resume/suspend callback for each ipFor called when PM do resume/suspend.Tested-by: Bhushan Shah Reviewed-by: Vasily Khoruzhick Signed-off-by: Qiang Yu Link: https://patchwork.freedesktop.org/patch/msgid/20200421133551.31481-8-yuq825@gmail.com List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

2ce216ed - drm/lima: Clean up redundant pdev pointer

Tue, 21 Apr 2020 22:51:37 +0000

drm/lima: Clean up redundant pdev pointerThere's no point explicitly tracking the platform device when it can betrivially derived from the regular device pointer in the couple ofplaces it's ever used.Signed-off-by: Robin Murphy Signed-off-by: Qiang Yu Link: https://patchwork.freedesktop.org/patch/msgid/8d9073cc91c10fc70910587fd1794e0e8f32b467.1587509150.git.robin.murphy@arm.com List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

21d81f88 - drm/lima: Clean up IRQ warnings

Tue, 21 Apr 2020 22:51:36 +0000

drm/lima: Clean up IRQ warningsUse the optional form of platform_get_irq() for blocks that legitimatelymay not be present, to avoid getting an annoying barrage of spuriouswarnings for non-existent PPs on configurations like Mali-450 MP2.Signed-off-by: Robin Murphy Signed-off-by: Qiang Yu Link: https://patchwork.freedesktop.org/patch/msgid/de475904091400ef6c123285f221094654d96d35.1587509150.git.robin.murphy@arm.com List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

19969707 - drm/lima: Add optional devfreq and cooling device support

Thu, 19 Mar 2020 20:34:27 +0000

drm/lima: Add optional devfreq and cooling device supportMost platforms with a Mali-400 or Mali-450 GPU also have support forchanging the GPU clock frequency. Add devfreq support so the GPU clockrate is updated based on the actual GPU usage when the"operating-points-v2" property is present in the board.dts.The actual devfreq code is taken from panfrost_devfreq.c and modified soit matches what the lima hardware needs:- a call to dev_pm_opp_set_clkname() during initialization because there are two clocks on Mali-4x0 IPs. "core" is the one that actually clocks the GPU so we need to control it using devfreq.- locking when reading or writing the devfreq statistics because (unlike than panfrost) we have multiple PP and GP IRQs which may finish jobs concurrently.Signed-off-by: Martin Blumenstingl Signed-off-by: Qiang Yu Link: https://patchwork.freedesktop.org/patch/msgid/20200319203427.2259891-3-martin.blumenstingl@googlemail.com List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

b78edd46 - drm/lima: save task info dump when task fail

Sat, 07 Mar 2020 13:44:23 +0000

drm/lima: save task info dump when task failSave all information to start a task which can be exported to userfor debug usage. Dump file data format is specified in lima_dump.hv2:Add include header to address build robot complain.Tested-by: Andreas Baierl Reviewed-by: Vasily Khoruzhick Signed-off-by: Qiang Yu Link: https://patchwork.freedesktop.org/patch/msgid/20200307134423.24329-1-yuq825@gmail.com List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

e30b38b7 - drm/lima: add __GFP_NOWARN flag to all dma_alloc_wc

Thu, 10 Oct 2019 14:01:52 +0000

drm/lima: add __GFP_NOWARN flag to all dma_alloc_wcThis prevent CMA printing dumy "PFNs busy" info which iscaused by alloc fail re-try case.Reviewed-by: Vasily Khoruzhick Signed-off-by: Qiang Yu Link: https://patchwork.freedesktop.org/patch/msgid/20191010140152.17747-4-yuq825@gmail.com List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

30abc59b - drm/lima: Add support for multiple reset lines

Tue, 01 Oct 2019 18:29:27 +0000

drm/lima: Add support for multiple reset linesSome SoCs like HiKey have 2 reset lines, so updateto use the devm_reset_control_array_* variant of theAPI so that multiple resets can be specified in DT.Cc: Qiang Yu Cc: David Airlie Cc: Daniel Vetter Cc: dri-devel@lists.freedesktop.orgCc: lima@lists.freedesktop.orgSigned-off-by: Peter Griffin Signed-off-by: John Stultz Signed-off-by: Qiang Yu Link: https://patchwork.freedesktop.org/patch/msgid/20191001182927.70448-1-john.stultz@linaro.org List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

0e343b08 - drm/lima: Reduce the amount of logs on deferred probe of clocks and reset controller

Fri, 21 Jun 2019 16:21:17 +0000

drm/lima: Reduce the amount of logs on deferred probe of clocks and reset controllerThere is no point to print deferred probe messages as errors. Adjustthe printks for error paths of obtaining clocks and reset controller.This removes the error message of lima_clk_init() call in favor orspecific failure messages inside.Signed-off-by: Krzysztof Kozlowski Signed-off-by: Qiang Yu Link: https://patchwork.freedesktop.org/patch/msgid/20190621162117.22533-4-krzk@kernel.org List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

7b4e467d - drm/lima: Reduce number of PTR_ERR() calls

Fri, 21 Jun 2019 16:21:16 +0000

drm/lima: Reduce number of PTR_ERR() callsStore the PTR_ERR() result in local variable in clock init error path.This makes the code consistent with similar section in regulator initcode.Signed-off-by: Krzysztof Kozlowski Signed-off-by: Qiang Yu Link: https://patchwork.freedesktop.org/patch/msgid/20190621162117.22533-3-krzk@kernel.org List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

34e88f9e - drm/lima: Reduce the amount of logs on deferred probe

Fri, 21 Jun 2019 16:21:15 +0000

drm/lima: Reduce the amount of logs on deferred probeThere is no point to print deferred probe (and its failures to getresources) as an error. For example getting a regulator causes threeunneeded error messages: lima 13000000.gpu: failed to get regulator: -517 lima 13000000.gpu: regulator init fail -517 lima 13000000.gpu: Fatal error during GPU initAlso do not print clock rates before the initialization finishesbecause they will be duplicated after deferral. Each probe step alreadyprints error so remove the final error message "Fatal error during GPUinit".In case of multiple probe tries this would pollute the dmesg.Fixes: a1d2a6339961 ("drm/lima: driver for ARM Mali4xx GPUs")Signed-off-by: Krzysztof Kozlowski Signed-off-by: Qiang Yu Link: https://patchwork.freedesktop.org/patch/msgid/20190621162117.22533-2-krzk@kernel.org List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c

a1d2a633 - drm/lima: driver for ARM Mali4xx GPUs

Sat, 09 Mar 2019 12:20:12 +0000

drm/lima: driver for ARM Mali4xx GPUs- Mali 4xx GPUs have two kinds of processors GP and PP. GP is for OpenGL vertex shader processing and PP is for fragment shader processing. Each processor has its own MMU so prcessors work in virtual address space.- There's only one GP but multiple PP (max 4 for mali 400 and 8 for mali 450) in the same mali 4xx GPU. All PPs are grouped togather to handle a single fragment shader task divided by FB output tiled pixels. Mali 400 user space driver is responsible for assign target tiled pixels to each PP, but mali 450 has a HW module called DLBU to dynamically balance each PP's load.- User space driver allocate buffer object and map into GPU virtual address space, upload command stream and draw data with CPU mmap of the buffer object, then submit task to GP/PP with a register frame indicating where is the command stream and misc settings.- There's no command stream validation/relocation due to each user process has its own GPU virtual address space. GP/PP's MMU switch virtual address space before running two tasks from different user process. Error or evil user space code just get MMU fault or GP/PP error IRQ, then the HW/SW will be recovered.- Use GEM+shmem for MM. Currently just alloc and pin memory when gem object creation. GPU vm map of the buffer is also done in the alloc stage in kernel space. We may delay the memory allocation and real GPU vm map to command submission stage in the furture as improvement.- Use drm_sched for GPU task schedule. Each OpenGL context should have a lima context object in the kernel to distinguish tasks from different user. drm_sched gets task from each lima context in a fair way.mesa driver can be found here before upstreamed:https://gitlab.freedesktop.org/lima/mesav8:- add comments for in_sync- fix ctx free miss mutex unlockv7:- remove lima_fence_ops with default value- move fence slab create to device probe- check pad ioctl args to be zero- add comments for user/kernel interfacev6:- fix comments by checkpatch.plv5:- export gp/pp version to userspace- rebase on drm-misc-nextv4:- use get param interface to get info- separate context create/free ioctl- remove unused max sched task param- update copyright time- use xarray instead of idr- stop using drmP.hv3:- fix comments from kbuild robot- restrict supported arch to tested onesv2:- fix syscall argument check- fix job finish fence leak since kernel 5.0- use drm syncobj to replace native fence- move buffer object GPU va map into kernel- reserve syscall argument space for future info- remove kernel gem modifier- switch TTM back to GEM+shmem MM- use time based io poll- use whole register name- adopt gem reservation obj integration- use drm_timeout_abs_to_jiffiesCc: Eric Anholt Cc: Rob Herring Cc: Christian König Cc: Daniel Vetter Cc: Alex Deucher Cc: Sam Ravnborg Cc: Rob Clark Cc: Dave Airlie Signed-off-by: Andreas Baierl Signed-off-by: Erico Nunes Signed-off-by: Heiko Stuebner Signed-off-by: Marek Vasut Signed-off-by: Neil Armstrong Signed-off-by: Simon Shields Signed-off-by: Vasily Khoruzhick Signed-off-by: Qiang Yu Reviewed-by: Eric Anholt Reviewed-by: Rob Herring Signed-off-by: Eric Anholt Link: https://patchwork.freedesktop.org/patch/291200/ List of files: /linux-6.15/drivers/gpu/drm/lima/lima_device.c