nvidia-tegra-modules.git/drivers/pci/host, branch gpu-paging NVIDIA's kernel modules to support tegra chips (used in Jetson boards) pcie-tegra: fix 'out of bound access' 2020-03-31T15:39:13+00:00 Achal Verma achalv@nvidia.com 2020-03-27T11:34:22+00:00 6e3f7c2ca9fa86f0c1603e888d7e9234f4cb7d12 correct range in find_first_bit. This fixes coverity issue: 5755193 Bug 2028892 Change-Id: I1eb677c760cc6fc2c62bb7fdcac1341f539643d5 Signed-off-by: Achal Verma <achalv@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2319414 Reviewed-by: automaticguardword <automaticguardword@nvidia.com> Reviewed-by: Dmitry Pervushin <dpervushin@nvidia.com> Reviewed-by: Phoenix Jung <pjung@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> GVS: Gerrit_Virtual_Submit Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
correct range in find_first_bit.

This fixes coverity issue: 5755193

Bug 2028892

Change-Id: I1eb677c760cc6fc2c62bb7fdcac1341f539643d5
Signed-off-by: Achal Verma <achalv@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2319414
Reviewed-by: automaticguardword <automaticguardword@nvidia.com>
Reviewed-by: Dmitry Pervushin <dpervushin@nvidia.com>
Reviewed-by: Phoenix Jung <pjung@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
pci: fix memory access 2020-03-18T21:24:45+00:00 Dmitry Pervushin dpervushin@nvidia.com 2019-07-16T11:06:39+00:00 4b53b0708e9549567871af3f2d865ca0edc9b3c7 KASAN has reported the following trace: [ 8.802446] Call trace: [ 8.802460] [<ffffff900808dbd0>] dump_backtrace+0x0/0x290 [ 8.802471] [<ffffff900808e1ac>] show_stack+0x24/0x30 [ 8.802483] [<ffffff9008566e38>] dump_stack+0xa0/0xc8 [ 8.802496] [<ffffff90082cd830>] print_address_description+0x68/0x258 [ 8.802506] [<ffffff90082cdd84>] kasan_report+0x264/0x338 [ 8.802514] [<ffffff90082cc3b4>] __asan_load8+0x84/0x98 [ 8.802523] [<ffffff900858a578>] _find_next_bit.part.0+0x30/0xc8 [ 8.802532] [<ffffff900858a638>] find_next_bit+0x28/0x30 [ 8.802569] [<ffffff9000ef87a0>] tegra_pcie_dw_scan_bus+0x128/0x518 [pcie_tegra_dw] [ 8.802578] [<ffffff9008624394>] dw_pcie_host_init+0x474/0x8e0 [ 8.802612] [<ffffff9000efeb64>] tegra_pcie_dw_runtime_resume+0x37c/0x510 [pcie_tegra_dw] [ 8.802625] [<ffffff900879e188>] pm_generic_runtime_resume+0x50/0x70 [ 8.802636] [<ffffff90087ac360>] __genpd_runtime_resume+0x50/0xe0 [ 8.802646] [<ffffff90087af690>] genpd_runtime_resume+0x1b0/0x278 [ 8.802656] [<ffffff90087a0c88>] __rpm_callback+0x88/0xb8 [ 8.802666] [<ffffff90087a0cfc>] rpm_callback+0x44/0xe0 [ 8.802673] [<ffffff90087a2b84>] rpm_resume+0x744/0xa38 [ 8.802681] [<ffffff90087a2ec4>] __pm_runtime_resume+0x4c/0x70 [ 8.802714] [<ffffff9000efb1f0>] tegra_pcie_dw_probe+0xca8/0x1388 [pcie_tegra_dw] [ 8.802726] [<ffffff9008793458>] platform_drv_probe+0x78/0xf8 [ 8.802737] [<ffffff900878f938>] driver_probe_device+0x128/0x598 [ 8.802747] [<ffffff900878ffec>] __device_attach_driver+0xec/0x1a8 [ 8.802758] [<ffffff900878c47c>] bus_for_each_drv+0xec/0x158 [ 8.802777] [<ffffff900878f654>] __device_attach+0x164/0x1f0 [ 8.802790] [<ffffff900879012c>] device_initial_probe+0x24/0x30 [ 8.802800] [<ffffff900878e0d8>] bus_probe_device+0xe8/0xf8 [ 8.802810] [<ffffff900878e9c8>] deferred_probe_work_func+0xc8/0x128 [ 8.802819] [<ffffff90080db8f8>] process_one_work+0x388/0x810 [ 8.802827] [<ffffff90080dbe1c>] worker_thread+0x9c/0x778 [ 8.802837] [<ffffff90080e5660>] kthread+0x170/0x190 [ 8.802845] [<ffffff9008083420>] ret_from_fork+0x10/0x30 [ 8.802850] The buggy address belongs to the page: [ 8.802861] page:ffffffbf0c1e78c0 count:0 mapcount:0 mapping: (null) index:0x0 [ 8.802865] flags: 0x4000000000000000() [ 8.802868] page dumped because: kasan: bad access detected find_first_bit (and therefore find_next_bit) is trying to dereference unsigned long So, pass pointer to 8-byte integer instead of u32 Bug 200535399 Change-Id: I5301896645e7c8fc68b2fceacbc61a3712d88c3f Signed-off-by: Dmitry Pervushin <dpervushin@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/2154067 GVS: Gerrit_Virtual_Submit Reviewed-by: Rahul Jain (SW-TEGRA) <rahuljain@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2313454 Reviewed-by: automaticguardword <automaticguardword@nvidia.com> Reviewed-by: Achal Verma <achalv@nvidia.com> Reviewed-by: Phoenix Jung <pjung@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: Achal Verma <achalv@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
KASAN  has reported the following trace:
[    8.802446] Call trace:
[    8.802460] [<ffffff900808dbd0>] dump_backtrace+0x0/0x290
[    8.802471] [<ffffff900808e1ac>] show_stack+0x24/0x30
[    8.802483] [<ffffff9008566e38>] dump_stack+0xa0/0xc8
[    8.802496] [<ffffff90082cd830>] print_address_description+0x68/0x258
[    8.802506] [<ffffff90082cdd84>] kasan_report+0x264/0x338
[    8.802514] [<ffffff90082cc3b4>] __asan_load8+0x84/0x98
[    8.802523] [<ffffff900858a578>] _find_next_bit.part.0+0x30/0xc8
[    8.802532] [<ffffff900858a638>] find_next_bit+0x28/0x30
[    8.802569] [<ffffff9000ef87a0>] tegra_pcie_dw_scan_bus+0x128/0x518 [pcie_tegra_dw]
[    8.802578] [<ffffff9008624394>] dw_pcie_host_init+0x474/0x8e0
[    8.802612] [<ffffff9000efeb64>] tegra_pcie_dw_runtime_resume+0x37c/0x510 [pcie_tegra_dw]
[    8.802625] [<ffffff900879e188>] pm_generic_runtime_resume+0x50/0x70
[    8.802636] [<ffffff90087ac360>] __genpd_runtime_resume+0x50/0xe0
[    8.802646] [<ffffff90087af690>] genpd_runtime_resume+0x1b0/0x278
[    8.802656] [<ffffff90087a0c88>] __rpm_callback+0x88/0xb8
[    8.802666] [<ffffff90087a0cfc>] rpm_callback+0x44/0xe0
[    8.802673] [<ffffff90087a2b84>] rpm_resume+0x744/0xa38
[    8.802681] [<ffffff90087a2ec4>] __pm_runtime_resume+0x4c/0x70
[    8.802714] [<ffffff9000efb1f0>] tegra_pcie_dw_probe+0xca8/0x1388 [pcie_tegra_dw]
[    8.802726] [<ffffff9008793458>] platform_drv_probe+0x78/0xf8
[    8.802737] [<ffffff900878f938>] driver_probe_device+0x128/0x598
[    8.802747] [<ffffff900878ffec>] __device_attach_driver+0xec/0x1a8
[    8.802758] [<ffffff900878c47c>] bus_for_each_drv+0xec/0x158
[    8.802777] [<ffffff900878f654>] __device_attach+0x164/0x1f0
[    8.802790] [<ffffff900879012c>] device_initial_probe+0x24/0x30
[    8.802800] [<ffffff900878e0d8>] bus_probe_device+0xe8/0xf8
[    8.802810] [<ffffff900878e9c8>] deferred_probe_work_func+0xc8/0x128
[    8.802819] [<ffffff90080db8f8>] process_one_work+0x388/0x810
[    8.802827] [<ffffff90080dbe1c>] worker_thread+0x9c/0x778
[    8.802837] [<ffffff90080e5660>] kthread+0x170/0x190
[    8.802845] [<ffffff9008083420>] ret_from_fork+0x10/0x30

[    8.802850] The buggy address belongs to the page:
[    8.802861] page:ffffffbf0c1e78c0 count:0 mapcount:0 mapping:          (null) index:0x0
[    8.802865] flags: 0x4000000000000000()
[    8.802868] page dumped because: kasan: bad access detected

find_first_bit (and therefore find_next_bit) is trying to dereference unsigned long
So, pass pointer to 8-byte integer instead of u32

Bug 200535399

Change-Id: I5301896645e7c8fc68b2fceacbc61a3712d88c3f
Signed-off-by: Dmitry Pervushin <dpervushin@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2154067
GVS: Gerrit_Virtual_Submit
Reviewed-by: Rahul Jain (SW-TEGRA) <rahuljain@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2313454
Reviewed-by: automaticguardword <automaticguardword@nvidia.com>
Reviewed-by: Achal Verma <achalv@nvidia.com>
Reviewed-by: Phoenix Jung <pjung@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: Achal Verma <achalv@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
PCI: tegra: Fix programming for CLKREQ absence 2019-09-21T06:38:50+00:00 Vidya Sagar vidyas@nvidia.com 2019-09-20T04:39:15+00:00 379a249726ccf9eaacaf14327272859945ad3770 Fixes programming to correctly remove the dependency of CLKREQ signal on providing REFCLK to endpoints. Bug 200553607 Change-Id: I387ee5f2429e993d465ced9b2c871d64c07ea114 Signed-off-by: Vidya Sagar <vidyas@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/2202082 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Manikanta Maddireddy <mmaddireddy@nvidia.com> Reviewed-by: Bibek Basu <bbasu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
Fixes programming to correctly remove the dependency of CLKREQ signal
on providing REFCLK to endpoints.

Bug 200553607

Change-Id: I387ee5f2429e993d465ced9b2c871d64c07ea114
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2202082
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Manikanta Maddireddy <mmaddireddy@nvidia.com>
Reviewed-by: Bibek Basu <bbasu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
PCI: tegra: Program core registers only after its reset de-assertion 2019-07-02T23:14:59+00:00 Vidya Sagar vidyas@nvidia.com 2019-07-01T13:50:23+00:00 ad44d44962d31b93addb24296dde9e8aee533c74 When "nvidia,disable-clock-request" device tree flag is passed, driver attempts to disable ASPM L1 sub-states by configuring corresponding DBI registers before even the reset to DBI registers is asserted resulting in system crash. This patch fixes that by moving DBI configuration code after de-assertion of core reset Bug 200531572 Change-Id: I7ee961a88676bea3dd5bc6be81161fb69612b89e Signed-off-by: Vidya Sagar <vidyas@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/2146277 (cherry picked from commit 2c985b55a20c25913e75c4c940095729047e578d) Reviewed-on: https://git-master.nvidia.com/r/2146603 GVS: Gerrit_Virtual_Submit Reviewed-by: Bibek Basu <bbasu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
When "nvidia,disable-clock-request" device tree flag is passed, driver
attempts to disable ASPM L1 sub-states by configuring corresponding DBI
registers before even the reset to DBI registers is asserted resulting
in system crash. This patch fixes that by moving DBI configuration code
after de-assertion of core reset

Bug 200531572

Change-Id: I7ee961a88676bea3dd5bc6be81161fb69612b89e
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2146277
(cherry picked from commit 2c985b55a20c25913e75c4c940095729047e578d)
Reviewed-on: https://git-master.nvidia.com/r/2146603
GVS: Gerrit_Virtual_Submit
Reviewed-by: Bibek Basu <bbasu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
PCI: tegra: Fix controller power down code 2019-06-27T11:42:12+00:00 Achal Verma achalv@nvidia.com 2019-06-26T09:38:31+00:00 140d6722fd886b2acf9e04c0cdd5dd4dacbf0885 PCIe controller dt property "nvidia,enable-power-down", which flags to power down the controller in case of link down is not taken care for suspend resume funcion calls. This change fixes it. Bug 200503658 Change-Id: I12899631fcf8fd05adfd5766b39ee719ce547d9a Signed-off-by: Achal Verma <achalv@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/2143614 Reviewed-by: Vidya Sagar <vidyas@nvidia.com> Reviewed-by: Dmitry Pervushin <dpervushin@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Rahul Jain (SW-TEGRA) <rahuljain@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
PCIe controller dt property "nvidia,enable-power-down",
which flags to power down the controller in case of link
down is not taken care for suspend resume funcion calls.
This change fixes it.


Bug 200503658

Change-Id: I12899631fcf8fd05adfd5766b39ee719ce547d9a
Signed-off-by: Achal Verma <achalv@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2143614
Reviewed-by: Vidya Sagar <vidyas@nvidia.com>
Reviewed-by: Dmitry Pervushin <dpervushin@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Rahul Jain (SW-TEGRA) <rahuljain@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
PCI: tegra: fix iATU aperture programming 2019-04-16T10:58:01+00:00 Vidya Sagar vidyas@nvidia.com 2019-04-11T07:22:27+00:00 0f89604c2f9c5513f12874ca1a226912e4283432 fixes iATU outbound programming by correctly updating pci bus address without which any mapping other than one-to-one doesn't work Bug 2494098 Change-Id: Ia292adb3e28094aab529fcb83c9de1983a7b3cc5 Signed-off-by: Vidya Sagar <vidyas@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/2094983 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Manikanta Maddireddy <mmaddireddy@nvidia.com> Reviewed-by: Bibek Basu <bbasu@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
fixes iATU outbound programming by correctly updating pci bus address
without which any mapping other than one-to-one doesn't work

Bug 2494098

Change-Id: Ia292adb3e28094aab529fcb83c9de1983a7b3cc5
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2094983
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Manikanta Maddireddy <mmaddireddy@nvidia.com>
Reviewed-by: Bibek Basu <bbasu@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
pci: do not print EPROBE_DEFER 2019-03-11T17:59:01+00:00 Dmitry Pervushin dpervushin@nvidia.com 2019-01-10T16:29:22+00:00 931db742e4910b4b9617bcbc1fc124d4dd704d2d If some other driver requires us to defer the probe, we should not shout about it Bug 2330176 Change-Id: Ie5400493458f58d032bb0f715c6b44e0394d2d77 Signed-off-by: Dmitry Pervushin <dpervushin@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1992185 (cherry picked from commit bc84224796df76be8154173742c6fdd5e30accd9) Reviewed-on: https://git-master.nvidia.com/r/2023870 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Rahul Jain (SW-TEGRA) <rahuljain@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
If some other driver requires us to defer the probe,
we should not shout about it

Bug 2330176

Change-Id: Ie5400493458f58d032bb0f715c6b44e0394d2d77
Signed-off-by: Dmitry Pervushin <dpervushin@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1992185
(cherry picked from commit bc84224796df76be8154173742c6fdd5e30accd9)
Reviewed-on: https://git-master.nvidia.com/r/2023870
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Rahul Jain (SW-TEGRA) <rahuljain@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
PCI: designware: rename Kconfig symbol 2018-12-11T19:35:26+00:00 Stephen Warren swarren@nvidia.com 2018-11-30T00:02:50+00:00 1d0c96c98418773bb8dc2d3bce41c8e2a7e26d64 A future change will backport a new DWC PCIe driver from kernel 4.14. Unfortunately, this causes a Kconfig symbol collision, which in turn causes parts of both drivers to always be built when either is enabled, which causes link failures. Solve this by renaming the existing Kconfig symbol to a legacy name. Eventually, we'll get rid of the current driver and replace it with the new one, so renaming the current driver causes the least amount of churn. Bug 2312059 Change-Id: I57216e7f2e802364046cf00b24efae96b537c417 Signed-off-by: Stephen Warren <swarren@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1962220 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Manikanta Maddireddy <mmaddireddy@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Sachin Nikam <snikam@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
A future change will backport a new DWC PCIe driver from kernel 4.14.
Unfortunately, this causes a Kconfig symbol collision, which in turn
causes parts of both drivers to always be built when either is enabled,
which causes link failures. Solve this by renaming the existing Kconfig
symbol to a legacy name. Eventually, we'll get rid of the current driver
and replace it with the new one, so renaming the current driver causes
the least amount of churn.

Bug 2312059

Change-Id: I57216e7f2e802364046cf00b24efae96b537c417
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1962220
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Manikanta Maddireddy <mmaddireddy@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Sachin Nikam <snikam@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
PCI: tegra: Disable interrupts before entering L2 2018-11-10T00:05:45+00:00 Manikanta Maddireddy mmaddireddy@nvidia.com 2018-11-03T04:48:21+00:00 fe17829a12d57dd6a5f53a83f292703d0761581e In suspend_noirq() call if link doesn't goto L2, PERST# is asserted to bring link to detect state. However this is causing surprise link down AER error. Since Kernel is executing noirq suspend calls AER interrupt is not processed. PME and AER are shared interrupts and PCIe subsystem driver enable wake capability of PME irq during suspend. So this AER will cause suspend failure due to pending AER interrupt. After PCIe link is in L2, interrupts are not expected since PCIe controller will be in reset state. Disable PCIe interrupts before going to L2 state to avoid pending AER interrupt. bug 200464272 Change-Id: I3642714fed53f309360ab6f2c2d8668d55224136 Signed-off-by: Manikanta Maddireddy <mmaddireddy@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1945767 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Rajkumar Kasirajan <rkasirajan@nvidia.com> Tested-by: Rajkumar Kasirajan <rkasirajan@nvidia.com> Reviewed-by: Vidya Sagar <vidyas@nvidia.com> Reviewed-by: Bibek Basu <bbasu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
In suspend_noirq() call if link doesn't goto L2, PERST# is asserted
to bring link to detect state. However this is causing surprise
link down AER error. Since Kernel is executing noirq suspend calls
AER interrupt is not processed. PME and AER are shared interrupts
and PCIe subsystem driver enable wake capability of PME irq during
suspend. So this AER will cause suspend failure due to pending
AER interrupt. After PCIe link is in L2, interrupts are not expected
since PCIe controller will be in reset state. Disable PCIe interrupts
before going to L2 state to avoid pending AER interrupt.

bug 200464272

Change-Id: I3642714fed53f309360ab6f2c2d8668d55224136
Signed-off-by: Manikanta Maddireddy <mmaddireddy@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1945767
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit
Reviewed-by: Rajkumar Kasirajan <rkasirajan@nvidia.com>
Tested-by: Rajkumar Kasirajan <rkasirajan@nvidia.com>
Reviewed-by: Vidya Sagar <vidyas@nvidia.com>
Reviewed-by: Bibek Basu <bbasu@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
PCI: tegra: Deassert LTSSM_EN if link doesn't goto L2 2018-11-02T12:50:48+00:00 Manikanta Maddireddy mmaddireddy@nvidia.com 2018-10-26T08:53:20+00:00 6a2ca48110335f0a3c30e5129595e2473c3d1abb Few PCIe cards like Marvell eSATA controller doesn't support L2. If PCIe link doesn't goto L2, assert PERST# and deassert LTSSM to bring the link to detect state. This will reset the UPHY lane speed to Gen1. bug 200461591 Change-Id: I74894a16dd81079d9eb644af4faa60e1bf77e9e2 Signed-off-by: Manikanta Maddireddy <mmaddireddy@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1936284 Reviewed-by: Vidya Sagar <vidyas@nvidia.com> Reviewed-by: Ashutosh Jha <ajha@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
Few PCIe cards like Marvell eSATA controller doesn't support L2. If PCIe
link doesn't goto L2, assert PERST# and deassert LTSSM to bring the link
to detect state. This will reset the UPHY lane speed to Gen1.

bug 200461591

Change-Id: I74894a16dd81079d9eb644af4faa60e1bf77e9e2
Signed-off-by: Manikanta Maddireddy <mmaddireddy@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1936284
Reviewed-by: Vidya Sagar <vidyas@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>