nvidia-tegra-modules.git/drivers/firmware, branch gpu-paging NVIDIA's kernel modules to support tegra chips (used in Jetson boards) firmware: tegra: fix channel reclaim feature 2020-09-02T13:24:43+00:00 Timo Alho talho@nvidia.com 2020-02-19T11:09:40+00:00 0626681ac8f0b309b1b8b248fa715b4bc22b741e Commit 4710a191d45f9889d204bc06d40ee4cec077eef4 (firmware: tegra: reclaim timedout channels) introduced a feature to reclaim channels that timed out. The implementation has a logical bug that upon receiving a message, and if there was any channel to be reclaimed (regardless of the current channel), the channel was freed forcefully. Fix the issue by correcting the boolean statement to use bit-wise-and of bit-masks. Bug 200588424 Bug 200644486 Change-Id: Ifcca42e1d1a8a9ef534af1e421d8340e93eea878 Signed-off-by: Timo Alho <talho@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2299496 (cherry picked from commit a4943add3e4c119b09b65f6c727586c0b7ba4e60) Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2394660 Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> Reviewed-by: automaticguardword <automaticguardword@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> GVS: Gerrit_Virtual_Submit 
Commit 4710a191d45f9889d204bc06d40ee4cec077eef4 (firmware: tegra:
reclaim timedout channels) introduced a feature to reclaim channels
that timed out. The implementation has a logical bug that upon
receiving a message, and if there was any channel to be reclaimed
(regardless of the current channel), the channel was freed forcefully.

Fix the issue by correcting the boolean statement to use bit-wise-and
of bit-masks.

Bug 200588424
Bug 200644486

Change-Id: Ifcca42e1d1a8a9ef534af1e421d8340e93eea878
Signed-off-by: Timo Alho <talho@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2299496
(cherry picked from commit a4943add3e4c119b09b65f6c727586c0b7ba4e60)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2394660
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: automaticguardword <automaticguardword@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
tegra/mail: fix 'out of bound access' 2020-03-20T16:47:48+00:00 Achal Verma achalv@nvidia.com 2020-03-18T11:30:08+00:00 44eb35c1842bc2ec2709819eb38c8f2ef00ece7e correct range in for loop as per the channel_area array size. This fixes coverity issue: 7043029 Bug 2028892 Change-Id: Ia6cd74f50f4cf954f408b61d09e2434b30d57d7f Signed-off-by: Achal Verma <achalv@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2314674 Reviewed-by: automaticguardword <automaticguardword@nvidia.com> Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Dmitry Pervushin <dpervushin@nvidia.com> Reviewed-by: Phoenix Jung <pjung@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> GVS: Gerrit_Virtual_Submit 
correct range in for loop as per the
channel_area array size.

This fixes coverity issue: 7043029

Bug 2028892

Change-Id: Ia6cd74f50f4cf954f408b61d09e2434b30d57d7f
Signed-off-by: Achal Verma <achalv@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2314674
Reviewed-by: automaticguardword <automaticguardword@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Dmitry Pervushin <dpervushin@nvidia.com>
Reviewed-by: Phoenix Jung <pjung@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: Gerrit_Virtual_Submit
firmware: tegra: reclaim timedout channels 2018-09-14T00:55:18+00:00 Sivaram Nair sivaramn@nvidia.com 2018-09-10T22:52:07+00:00 4710a191d45f9889d204bc06d40ee4cec077eef4 Rather than loosing IPC channels on a timeout, reclaim them for future use. Bug 200446874 Change-Id: I4b2f3c95822f0418ced21fb073f3f10f4c38b04b Signed-off-by: Sivaram Nair <sivaramn@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1818472 Reviewed-by: Stefan Kristiansson <stefank@nvidia.com> Reviewed-by: Vishruth Jain <vishruthj@nvidia.com> Reviewed-by: Timo Alho <talho@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
Rather than loosing IPC channels on a timeout, reclaim them for future
use.

Bug 200446874

Change-Id: I4b2f3c95822f0418ced21fb073f3f10f4c38b04b
Signed-off-by: Sivaram Nair <sivaramn@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1818472
Reviewed-by: Stefan Kristiansson <stefank@nvidia.com>
Reviewed-by: Vishruth Jain <vishruthj@nvidia.com>
Reviewed-by: Timo Alho <talho@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
firmware: tegra: increase IPC wait time 2018-09-14T00:55:14+00:00 Sivaram Nair sivaramn@nvidia.com 2018-09-10T22:44:13+00:00 6930c4e303afceb820b09d3121b1d2f0804cf1d2 Due to combined UART contention, we can sometime take longer than the current timeout period (of one second) to complete a IPC transaction. Quadruple the deadline, but show a warning message when turnaround time is longer than one-fourth of it. Bug 200446874 Change-Id: I88d9f8d337c389f378e4778ebd856ca142bd40c2 Signed-off-by: Sivaram Nair <sivaramn@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1818471 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Timo Alho <talho@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
Due to combined UART contention, we can sometime take longer than the
current timeout period (of one second) to complete a IPC transaction.
Quadruple the deadline, but show a warning message when turnaround time
is longer than one-fourth of it.

Bug 200446874

Change-Id: I88d9f8d337c389f378e4778ebd856ca142bd40c2
Signed-off-by: Sivaram Nair <sivaramn@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1818471
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Timo Alho <talho@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
firmware: tegra: enhance mrq data output 2018-09-14T00:55:11+00:00 Sivaram Nair sivaramn@nvidia.com 2018-09-10T22:37:17+00:00 7e3a430b3e5d64308abf9cab161d36392c00e453 Show IPC failure message and MRQ data with a single pr_err() so that this does not get broken up and/or mangled (as they do today in some platforms). Bug 200446874 Change-Id: I2969fd8aa677e2ed32d86201fcb83cf26266413e Signed-off-by: Sivaram Nair <sivaramn@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1818470 Reviewed-by: Srikar Srimath Tirumala <srikars@nvidia.com> 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> 
Show IPC failure message and MRQ data with a single pr_err() so that
this does not get broken up and/or mangled (as they do today in some
platforms).

Bug 200446874

Change-Id: I2969fd8aa677e2ed32d86201fcb83cf26266413e
Signed-off-by: Sivaram Nair <sivaramn@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1818470
Reviewed-by: Srikar Srimath Tirumala <srikars@nvidia.com>
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>
firmware: tegra: cleanup channel indexing 2018-09-14T00:55:07+00:00 Sivaram Nair sivaramn@nvidia.com 2018-09-10T22:31:50+00:00 41be4438c2f21ac2c96f9a3ef6f352c6e1edcf36 Get rid of the confusing bpmp_thread_ch_index() and bpmp_thread_ch() functions and use channel number directly in tch_free, so that it also aligns with to_complete. Bug 200446874 Change-Id: I35ce9fd650af348b3760c38b3b6635e0d62126ed Signed-off-by: Sivaram Nair <sivaramn@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1818469 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Stefan Kristiansson <stefank@nvidia.com> Reviewed-by: Vishruth Jain <vishruthj@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
Get rid of the confusing bpmp_thread_ch_index() and bpmp_thread_ch()
functions and use channel number directly in tch_free, so that it also
aligns with to_complete.

Bug 200446874

Change-Id: I35ce9fd650af348b3760c38b3b6635e0d62126ed
Signed-off-by: Sivaram Nair <sivaramn@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1818469
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Stefan Kristiansson <stefank@nvidia.com>
Reviewed-by: Vishruth Jain <vishruthj@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
firmware: tegra: fix type error with memcpy_*io() 2018-09-10T00:22:53+00:00 Sivaram Nair sivaramn@nvidia.com 2018-09-04T22:58:22+00:00 517525c1f0edcb9764716c8c0d216eb3cdc23f77 Bug 200434802 Change-Id: I549b6ef0b4accfcc265e4d17eea7ffa6a55a2b3c Signed-off-by: Sivaram Nair <sivaramn@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1812810 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
Bug 200434802

Change-Id: I549b6ef0b4accfcc265e4d17eea7ffa6a55a2b3c
Signed-off-by: Sivaram Nair <sivaramn@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1812810
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
firmware: tegra: explicilty cast little-endians to uint32_t 2018-09-04T23:17:15+00:00 Timo Alho talho@nvidia.com 2018-09-03T12:47:13+00:00 233d8c34db47629c78f863ad29320dc4d404d005 The BPMP-FW firmware header uses little endian byte order in the command structs. However, the struct members are declared as uint32_t and sparse emits warnings as cpu_to_le32() return type is __le32. To avoid these warnings, cast the return value of cpu_to_le32() explicitly to uint32_t. It is arguable that more correct way to resolve the warnings would be to change the ABI header data types. However, this is unfavorable as the same header is shared across multiple operating systems and __le32 is not standard data type. Bug 200434802 Change-Id: I3b27b0086a7b6aeb9707e72c21e08bbf89fe8a7b Signed-off-by: Timo Alho <talho@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1811967 GVS: Gerrit_Virtual_Submit Reviewed-by: Sivaram Nair <sivaramn@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
The BPMP-FW firmware header uses little endian byte order in the
command structs. However, the struct members are declared as uint32_t
and sparse emits warnings as cpu_to_le32() return type is __le32.

To avoid these warnings, cast the return value of cpu_to_le32()
explicitly to uint32_t. It is arguable that more correct way to
resolve the warnings would be to change the ABI header data
types. However, this is unfavorable as the same header is shared
across multiple operating systems and __le32 is not standard data
type.

Bug 200434802

Change-Id: I3b27b0086a7b6aeb9707e72c21e08bbf89fe8a7b
Signed-off-by: Timo Alho <talho@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1811967
GVS: Gerrit_Virtual_Submit
Reviewed-by: Sivaram Nair <sivaramn@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
firmware: tegra: dump pending messages upon SC7 entry 2018-08-31T20:23:43+00:00 Timo Alho talho@nvidia.com 2018-08-20T09:58:38+00:00 e921157ed6c7851ea5bc0008c1ee3a125968e576 Bug 200441330 Change-Id: I36ceaf6bd927f56abe0904f0a61d5c1c6038d1ab Signed-off-by: Timo Alho <talho@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1803049 GVS: Gerrit_Virtual_Submit Reviewed-by: Sivaram Nair <sivaramn@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
Bug 200441330

Change-Id: I36ceaf6bd927f56abe0904f0a61d5c1c6038d1ab
Signed-off-by: Timo Alho <talho@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1803049
GVS: Gerrit_Virtual_Submit
Reviewed-by: Sivaram Nair <sivaramn@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
firmware: tegra: Make tegra_hv_ivc_disconnect a static 2018-08-27T11:53:14+00:00 sreenivasulu velpula svelpula@nvidia.com 2018-08-16T17:47:30+00:00 98c6b535a26596400995b513cb8cda9967528e93 Fix Sparse static code analysis warning. Bug 200408168 Change-Id: I46943fd906df2a015a0248c747b36ff09d911240 Signed-off-by: sreenivasulu velpula <svelpula@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1801262 Reviewed-by: Sachin Nikam <snikam@nvidia.com> Tested-by: Sachin Nikam <snikam@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> 
Fix Sparse static code analysis warning.

Bug 200408168

Change-Id: I46943fd906df2a015a0248c747b36ff09d911240
Signed-off-by: sreenivasulu velpula <svelpula@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1801262
Reviewed-by: Sachin Nikam <snikam@nvidia.com>
Tested-by: Sachin Nikam <snikam@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>