summaryrefslogtreecommitdiff
path: root/drivers/gpu/arm/midgard
AgeCommit message (Collapse)Author
2019-04-30gpu: mali: Fix signature of kbase_get_[static|dynamic]_power functions.Christoph Muellner
This addresses a type warning, which elevates to a broken build because of -Werror: CC drivers/gpu/arm/midgard/ipa/mali_kbase_ipa.o drivers/gpu/arm/midgard/ipa/mali_kbase_ipa.c:582:22: warning: initialization from incompatible pointer type [-Wincompatible-pointer-types] error, forbidden warning:mali_kbase_ipa.c:582 .get_static_power = &kbase_get_static_power, ^ drivers/gpu/arm/midgard/ipa/mali_kbase_ipa.c:582:22: note: (near initialization for ‘kbase_ipa_power_model_ops.get_static_power’) drivers/gpu/arm/midgard/ipa/mali_kbase_ipa.c:583:23: warning: initialization from incompatible pointer type [-Wincompatible-pointer-types] error, forbidden warning:mali_kbase_ipa.c:583 .get_dynamic_power = &kbase_get_dynamic_power, ^ drivers/gpu/arm/midgard/ipa/mali_kbase_ipa.c:583:23: note: (near initialization for ‘kbase_ipa_power_model_ops.get_dynamic_power’) scripts/Makefile.build:277: recipe for target 'drivers/gpu/arm/midgard/ipa/mali_kbase_ipa.o' failed Signed-off-by: Christoph Muellner <christoph.muellner@theobroma-systems.com>
2019-04-12Mali: midgard: fix Spectre v1 vulnerabilityZhen Chen
user_atom.atom_number can be indirectly controlled by user-space, hence leading to a potential exploitation of the Spectre variant 1 vulnerability. This issue was detected with the help of Smatch: drivers/gpu/arm/midgard/mali_kbase_jd.c:1397 kbase_jd_submit() warn: potential spectre issue 'jctx->atoms' [r] katom = &jctx->atoms[user_atom.atom_number]; Fix this by sanitizing user_atom.atom_number before 'katom = &jctx->atoms[user_atom.atom_number];'. Notice that given that speculation windows are large, the policy is to kill the speculation on the first load and not worry if it can be completed with a dependent load/store [1]. [1] https://marc.info/?l=linux-kernel&m=152449131114778&w=2 Fixes: 5cf27d0b6090 ("Mali: midgard: changes to enlarge BASE_JD_ATOM_COUNT to 512, for defect 184210") Change-Id: If52f30d29a80a06c6693ddadd5947ab9fe8fbc25 Signed-off-by: Zhen Chen <chenzhen@rock-chips.com>
2019-04-12Mali: midgard: changes to enlarge BASE_JD_ATOM_COUNT to 512, for defect 184210Zhen Chen
The source code of mali_so must be modified correspondingly. Change-Id: I3f4bd03fa2d369d912e6bc05c53d2d3abefb92d3 Signed-off-by: Zhen Chen <chenzhen@rock-chips.com>
2018-08-07Mali: midgard: set "FMODE_UNSIGNED_OFFSET" bit in file flags of mali device fileZhen Chen
According to commit "mmap: introduce sane default mmap limits" Change-Id: I96bc78053c06fdeaef178662dda36770da867436 Signed-off-by: Zhen Chen <chenzhen@rock-chips.com>
2018-07-23Mali: midgard: add an error handling pass in kbase_mmu_interrupt()Zhen Chen
For RK redmine Defect #168230. Change-Id: I3cd6544dd23b833138e4cc700a8f2cdd627ff592 Signed-off-by: Zhen Chen <chenzhen@rock-chips.com>
2018-07-06MALI: midgard: Add wide-temperature controlFinley Xiao
Change-Id: Ibf1c6ebd91d2e793fa8efe0383df346dd8c67b45 Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>
2018-06-11MALI: rockchip: remove the out of time midgardRandy Li
The midgard for linux is used for mali r9 library, which is not used and update anymore. Change-Id: I446c9b223b5235b40eb28cd8001d318578b16ba2 Signed-off-by: Randy Li <randy.li@rock-chips.com>
2018-06-01soc: rockchip: opp_select: Add support to adjust power scaleFinley Xiao
Change-Id: I2358d75c2fdada7cfe385e85d2106370f9aa5ea3 Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>
2018-05-09MALI: rockchip: Ensure that set voltage when update devfreq for the first timeFinley Xiao
Sometimes the regulator is shared between several devices, if target rate and target voltage are equal to initial rate and iniital voltage , the min_uV and max_uV of regulator will be always equal to zero, other devices may set a low voltage. Change-Id: Ibf82f335ce0739776286e3733be5415f84bf035b Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>
2018-05-02MALI: rockchip: upgrade midgard DDK to r18p0-01rel0chenzhen
with following modifications : fix some compile errors when ARCH=arm; replace "mali_base" with "midgard_base" in Kbuild files; fix a compile error when MALI_PWRSOFT_765 enabled; set "default n" for MALI_KUTF and MALI_IRQ_LATENCY. Change-Id: I858c72aab939002c8e6241cd75fa8877ebbf6c35 Signed-off-by: Zhen Chen <chenzhen@rock-chips.com>
2018-03-13MALI: midgard: rk: fix a memory leak in platform specific codeZhen Chen
Change-Id: I31ecc394c8ac971ed915b18b08d5b5cba5d440e2 Signed-off-by: Zhen Chen <chenzhen@rock-chips.com>
2018-03-08PM / devfreq: record opp rate instead of real clock rate for thermalLiang Chen
Opp rate is used to calc power in thermal framework, so we record this rate instead of real clock rate. Devfreq is not ready in target() when use performance governor, so we need record opp rate in probe(). Change-Id: Iec1918ad5d12124b9f112964f247339e0d50645f Signed-off-by: Liang Chen <cl@rock-chips.com>
2018-02-05MALI: midgard: RK: rename output mali_kbase.ko to midgard_kbase.koZhen Chen
Some optimizations on files of KBuild in addition. Change-Id: I1db012e116b8b69897a2791ae610da35365a1a61 Signed-off-by: Zhen Chen <chenzhen@rock-chips.com>
2018-01-31rk: add SPDX license identifier to files with no licenseTao Huang
Change-Id: I754250669891307b0deab2bdab1bd01512713f79 Signed-off-by: Tao Huang <huangtao@rock-chips.com>
2018-01-04BACKPORT: devfreq_cooling: pass a pointer to devfreq in the power model ↵Javi Merino
callbacks When the devfreq cooling device was designed, it was an oversight not to pass a pointer to the struct devfreq as the first parameters of the callbacks. The design patterns of the kernel suggest it for a good reason. By passing a pointer to struct devfreq, the driver can register one function that works with multiple devices. With the current implementation, a driver that can work with multiple devices has to create multiple copies of the same function with different parameters so that each devfreq_cooling_device can use the appropriate one. By passing a pointer to struct devfreq, the driver can identify which device it's referring to. Change-Id: I384bf9aafd2391eccab2ca6a76e4e57f2740aa6b Cc: Zhang Rui <rui.zhang@intel.com> Cc: Eduardo Valentin <edubezval@gmail.com> Reviewed-by: Punit Agrawal <punit.agrawal@arm.com> Reviewed-by: Ørjan Eide <orjan.eide@arm.com> Reviewed-by: Lukasz Luba <lukasz.luba@arm.com> Signed-off-by: Javi Merino <javi.merino@arm.com> Signed-off-by: Eduardo Valentin <edubezval@gmail.com> Signed-off-by: Tao Huang <huangtao@rock-chips.com> (cherry picked from commit 3aa5374376746f20a27be8682be7f91c8b71d1d8)
2018-01-02soc: rockchip: add functions to select opp level form device-treeLiang Chen
Some chips need adjust opp level by different chip-process, add common functions to select opp level from device-tree, so modules can select opp level easy. Change-Id: Ifbd5f720e6a52a68f13697bbb37ac01ff4a87e3e Signed-off-by: Liang Chen <cl@rock-chips.com>
2017-08-18MALI: rockchip: Add support to select voltage according to leakageFinley Xiao
The greater leakage, the lower the OPP's voltage. In order to reduce power consumption, it is necessary to adjust OPP's voltage according to leakage. Change-Id: I2bba04ac941cc6a0703b5208cb4e757ec2813bab Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>
2017-08-18MALI: rockchip: Add support to select voltage according to pvtm valueFinley Xiao
At same voltage and frequency, the greater the PVTM value, the lower the OPP's voltage. In order to reduce power consumption, it is necessary to adjust OPP's voltage according to PVTM value. Change-Id: Ifcac57fc965e3ea45523db2d9df19435479f6cee Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>
2017-04-27MALI: fix thermal crash with booting upCaesar Wang
If the temperature(sbs-battery) reaches the switch_on_temp, it would try to calculate requested power of all thermal instances. Then hit the crash[0] caused by the gpu thermal sensor, since the thermal driver had not registered in time. [0] [ 0.827943] Call trace: [ 0.827953] [< (null)>] (null) [ 0.827969] [<ffffffc00070af1c>] get_static_power+0xd8/0xe8 [ 0.827981] [<ffffffc00070b190>] devfreq_cooling_get_requested_power+0x94/0x170 [ 0.827997] [<ffffffc0007094c8>] power_allocator_throttle+0x270/0x804 .. Change-Id: I63f66e54d69115165a7b3ec798b9009c360daa62 Signed-off-by: Caesar Wang <wxt@rock-chips.com>
2017-03-15Revert "Revert "MALI: midgard: support sharing regulator with other devices""chenzhen
This reverts commit eb6c2ed7207f4dd086548dd58589ed4ea042504f. Change-Id: I09424d08bfac3457da24b10fe2b97d2856399e63 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2017-03-15Revert "Revert "MALI: midgard: avoid GPU voltage domain keeping the initial ↵chenzhen
voltage"" This reverts commit e7db50b51268386fd0e9f88a95c45a4ef3454c4e. Change-Id: I47f9c1d114b06264d6dd4098bc0e00228fd8f187 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2017-03-15MALI: midgard: RK: add wake_lock for delayed_work_to_power_off_gpuchenzhen
To ensure that the work is executed before system being suspended. Change-Id: Iec1bd114dfff53e2464540f09ced66cf6be81d1a Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2017-02-20MALI: midgard: RK: adapt cores_pm in DDK r14 for solution_1_for_glitchchenzhen
Change-Id: I383779bd39d6ae52f65ad25bf2e0eb0f1a25dd00 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2017-02-20MALI: rockchip: upgrade midgard DDK to r14p0-01rel0chenzhen
Along with a slight modification in mali_kbase_core_linux.c, for building in rk Linux 4.4: -#if KERNEL_VERSION(4, 6, 0) > LINUX_VERSION_CODE +#if KERNEL_VERSION(4, 4, 0) > LINUX_VERSION_CODE Change-Id: I34565cb975866b46c5e3a4d8e2ac5e350dcceb80 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2017-02-20Revert "Revert "MALI: midgard: RK: not to power off all the pm cores""chenzhen
This reverts commit d94880b547779baaaa9e9b733c38881cad8aa685. Change-Id: Iac64d84ff5a7ee3e5666ed2829c17de413fc9bcd Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2017-02-20Revert "MALI: midgard: RK: slowdown clk_gpu before poweroff cores"chenzhen
This reverts commit 89501d8dd3c214e4a162e94804f0d56c61c23237. Change-Id: I403b63847da10bc2c5536bd26f692bafc849588e Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2017-02-20Revert "MALI: midgard: avoid GPU voltage domain keeping the initial voltage"chenzhen
This reverts commit 57984d531806892a8e14bb7c1b42b1c4c406ddf9. Change-Id: If538c9bbeb5d3fc7302f9683cb85f8acdd309a09 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2017-02-20Revert "MALI: midgard: support sharing regulator with other devices"chenzhen
This reverts commit 85b4e1dffa2e7a0bbd092d294043f19f82417d74. Change-Id: Ie8fb980cb8a8b063dd6c9626d5b6c858b36f0976 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2017-01-13Revert "MALI: rockchip: upgrade midgard DDK to r14p0-01rel0"chenzhen
This reverts commit d1637ff80953fd46692f923f3ee7b656fb917081. Change-Id: Ib99bae99fe7246142bfa7369b8e79ebbfae1e736 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2017-01-06MALI: rockchip: upgrade midgard DDK to r14p0-01rel0chenzhen
Along with a slight modification in mali_kbase_core_linux.c, for building in rk Linux 4.4: -#if KERNEL_VERSION(4, 6, 0) > LINUX_VERSION_CODE +#if KERNEL_VERSION(4, 4, 0) > LINUX_VERSION_CODE Change-Id: I9a8f4696c603cf0be6ce7b9b31b23a928a21b2a6 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-11-29MALI: midgard: support sharing regulator with other devicesFinley Xiao
If the regulator is shared between several devices then the lowest request voltage that meets the system constraints will be used. Change-Id: I7dda43b24c7e19098db65b51ae0c4386b46ee0b7 Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>
2016-11-29MALI: midgard: avoid GPU voltage domain keeping the initial voltageFinley Xiao
If there is only one opp whose frequency is equal to the initial value in opp table list, the GPU voltage domain will keep the initial voltage, it may be too large. Change-Id: I87a5fb82eaac8466123b61e39a5d7587da3066da Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>
2016-11-14MALI: midgard: RK: slowdown clk_gpu before poweroff coreschenzhen
This is a workaround for the issue that "400M, 500M and 600M of clk_gpu needs high vdd_gpu", according to "6.2" of Mali Application Note "Potential glitches on Power Domain interfaces". Change-Id: I8b8eccd2079e21ac5e1db7b4552c8f998f676a9f Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-11-14Revert "MALI: midgard: RK: not to power off all the pm cores"chenzhen
This reverts commit 47a9fbf6c4de1dddaf82ed9159307f8ba039f85c. Change-Id: I4cfc36a5dab7e6cb4a766995d456301c54728498 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-11-07MALI: midgard: ARM: patch for fence timeout in monkey testchenzhen
comes from mali support case [TAC674871]. Change-Id: I5e7c54190e24eea0306a9d64545482c6209a2c0d Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-09-29MALI: midgard: RK: not to power off all the pm coreschenzhen
This is a workaround for the issue that "400M, 500M and 600M of clk_gpu needs high vdd_gpu", according to "6.1" of Mali Application Note "Potential glitches on Power Domain interfaces". Change-Id: I58daa3cf796802f073f67bacb62734516be76205 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-09-13MALI: rockchip: upgrade midgard DDK to r13p0-00rel0chenzhen
Conflicts: drivers/gpu/arm/midgard/backend/gpu/mali_kbase_devfreq.c drivers/gpu/arm/midgard/backend/gpu/mali_kbase_power_model_simple.c drivers/gpu/arm/midgard/backend/gpu/mali_kbase_power_model_simple.h drivers/gpu/arm/midgard/mali_kbase_defs.h Change-Id: Ia7b8004b09ce31a5af6414c27b8ec776c247835a Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-09-01MALI: midgard: RK: fix compiling fail without CONFIG_REGULATOR definedchenzhen
Change-Id: I4eb6fcae00d934c1fc2cbffc888a88c2fda9c22d Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-08-11MALI: midgard: rockchip: change devfreq_dvfs_interval to 20 mschenzhen
Change-Id: I2ba7988bd08bb05661a324b66a27cf2028ebd5db Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-08-02MALI: midgard: rockchip: add sysfs files to get GPU utilisationchenzhen
Change-Id: I758369bdd9ef945a89fd87fd7a69bd9f391f0880 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-07-19MALI: rockchip: upgrade midgard DDK to r12p0-04rel0chenzhen
Signed-off-by: chenzhen <chenzhen@rock-chips.com> Conflicts: drivers/gpu/arm/midgard/mali_kbase_core_linux.c Change-Id: I17d67d066f6e029c2d8f1711002c347002a48399 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-07-04MALI: midgard: rockchip: add .shutdown of GPU platform_driverchenzhen
Change-Id: I5af2a464db88b08530b063d90b3a7ce61e26f201 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-07-04MALI: midgard: rockchip: not to enable clk_gpu when probingchenzhen
Otherwise, clk_gpu won't be disabled actually in the runtime. Change-Id: If1e32061cbffc1564a5cf95fbf01aa91c827550d Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-06-14MALI: rockchip: add power-off-delay functionFinley Xiao
As gpu power on and off frequently, and the interval time is smaller than the polling time of devfreq, add power-off-delay function to ensure devfreq work fine. Change-Id: Iba2405c9ead91a437233f1fedf2f3555703aa9e1 Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>
2016-05-31MALI: rockchip: not to use sg_dma_len in midgard ddk r11p0-00rel0chenzhen
When CONFIG_NEED_SG_DMA_LENGTH is enabled, sg_dma_len is defined as follow : "#define sg_dma_len(sg) ((sg)->dma_length)" But, dma_length is not used by the framework indeed. Change-Id: I51d1adf9f2738b1036fdaf6172ae932c007fac76 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-05-31MALI: rockchip: upgrade midgard DDK to r11p0-00rel0chenzhen
Conflicts: drivers/gpu/arm/midgard/mali_kbase_core_linux.c drivers/gpu/arm/midgard/mali_kbase_jd.c Change-Id: I9c910f2b08ffd2e9101fbe85958030ac7bca1642 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-05-31MALI: rockchip: upgrade midgard DDK to r9p0-05rel0chenzhen
Conflicts: drivers/gpu/arm/midgard/Kconfig Change-Id: Ib7975ebe959624bedd92f126768987f2e2f0f84b Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-05-05MALI: rockchip: add "platform specific code" of rk platformchenzhen
Change-Id: Ia58cba15b43f875ac572a3b35807b5ec48e3df01 Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-04-22MALI: rockchip: adjust code about thermal for kernel 4.4chenzhen
Change-Id: Ic5f3947b032deaaa800ee316636a8cc61259ba5d Signed-off-by: chenzhen <chenzhen@rock-chips.com>
2016-03-09MALI: rockchip: not to use sg_dma_len.chenzhen
When CONFIG_NEED_SG_DMA_LENGTH is enabled, sg_dma_len is defined as follow : "#define sg_dma_len(sg) ((sg)->dma_length)" But, dma_length is not used by the framework indeed. Change-Id: Ibfd3223b38b96701f839cdc91207a49f20789fec Signed-off-by: chenzhen <chenzhen@rock-chips.com>
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-03 17:08:33 +0000