Remote register I/O amounts to a measurably significant portion of CPU
time due to how frequently this function is used. Cache the value of
each register on-demand and use this value in future invocations to
mitigate the expensive I/O.
Co-authored-by: Sultan Alsawaf <sultan@kerneltoast.com>
[@0ctobot: Adapted for msm-4.19]
Change-Id: I6846ee58c476a125076d0c3ad351f9e15eba180a
A measurably significant amount of CPU time is spent in these routines
while the camera is open. These are also responsible for a grotesque
amount of dmesg spam, so let's nuke them.
Signed-off-by: Sultan Alsawaf <sultan@kerneltoast.com>
Change-Id: I1711e1e815e7c5abebdf888b752823bded926ebd
Signed-off-by: starlight5234 <starlight5234@protonmail.ch>
The Patch based on QualComm release TAG:LA.UM.8.15.r1-06600-KAMORTA.0
Change-Id: I1f98410528de75bb7c87b12ae7666769945e9ad9
Signed-off-by: starlight5234 <starlight5234@protonmail.ch>
* CAF never deattaches the DMA buffer, creating a memory leak
every time this function is run. This issue becomes increasingly
problematic as it's run a metric ton every time camera is open,
which on Coral is pretty much every time you unlock the phone
and often while you run it too. Fix it by unmaping and detaching
the dma buffer.
Reported-by: celtare21 <celtare21@gmail.com>
Change-Id: Id21b7d8c77b24ad3c198f46a68f1408c5bd9f104
Signed-off-by: starlight5234 <starlight5234@protonmail.ch>
error return type defaults to 'int'
Signed-off-by: Carlos Jimenez(JavaShin-X)<javashin1986@gmail.com>
Change-Id: I134153e945af4d70ab679af062a35e2b6e0f75ea
Signed-off-by: starlight5234 <starlight5234@protonmail.ch>
This feature implemented for power optimization
with OMX but not well compliant with Codec 2.0.
Disable it to unblock VT frame drop issue.
Bug: 149071324
Test: VT frame rate test
Change-Id: I187958f4da10d5936c0f0fbd5060301e55ac7f29
Signed-off-by: Wen Chang Liu <wenchangliu@google.com>
Signed-off-by: starlight5234 <starlight5234@protonmail.ch>
This patch fixes an issue with all the functions defined in
scripts/Makefile.lib gets called everytime when this makefile is
included thus causing the build to slow down.
Bug: 118818737
Change-Id: I4929e64d53a0bc865b12a27cce37520be52c9744
Signed-off-by: Thierry Strudel <tstrudel@google.com>
Signed-off-by: starlight5234 <starlight5234@protonmail.ch>
We already have CONFIG_CC_WERROR to enable -Werror globally when we want
it.
Signed-off-by: Danny Lin <danny@kdrag0n.dev>
Signed-off-by: KenHV <yo@kenharris.xyz>
Change-Id: I00c42b42a750dd35181a718c248d2aab7dcab1aa
Signed-off-by: starlight5234 <starlight5234@protonmail.ch>
* These symlinks were meant for the out-of-tree module. Now that we
build them in-tree, we need to correct them.
Change-Id: If2f506138c7e08123a90e0f897f0e899bb47ef0c
Signed-off-by: starlight5234 <starlight5234@protonmail.ch>
Support build environments where find -xtype is not supported.
This reverts commit 5081f16b3d.
Change-Id: Ibb4a9f2429bad5e0bae690f60605887f01623a16
Signed-off-by: Adam Bickett <abickett@codeaurora.org>
Due to path being inaccurate, the techpack folder was not being
compiled completely. Using techpack fixes this issue.
Change-Id: Id06927e5b4bfad0d5c8208b199d9d217abbf6306
Signed-off-by: Prakruthi Deepak Heragu <pheragu@codeaurora.org>
Signed-off-by: Adam Bickett <abickett@codeaurora.org>
Tech packages default configuration should be on.
Change-Id: I3cdf9276103f08db7130535aab1206a95a03b5a7
Signed-off-by: Bryan Huntsman <bryanh@codeaurora.org>
Enable tech packages to have kernel drivers that are physically
located in the kernel source tree but not present in the kernel
repository. This is done via a manifest line item to fetch a
techpackage-module into $KERNEL/techpack/$techpackage-module.
$KERNEL/techpack/Kbuild will automatically detect any
subdirectories and link them to the kernel's kbuild system.
The resulting layout of techpack within kernel source would be as
follows :-
kernel/
├── techpack/
├── Kbuild
└── stub
│ ├── Makefile
│ ├── include
│ │ └── uapi
│ │ └── Kbuild
└── stub.c
├── techpackage-module
├── Makefile
├── include
│ └── uapi
│ ├── Kbuild
$KERNEL/techpack only contains Kbuild (no Makefile) at the
toplevel and this Kbuild takes care of both compiling the
subdirectories and of exporting the needed header files therein.
The reason for having only Kbuild at the top is that Kbuild
and Makefile cannot exist together in same directory and Makefile
doesn't cater to uapi header installation.
stub is an empty techpackage-module which serves both as a sample
layout and satisfies the requirement of kernel build system, by
providing necessary buit-in.o, when no other techpackage-module has
been pulled under techpack/.
$KERNEL/techpack/techpackage-module should have a Makefile at the
top and Kbuild under $KERNEL/techpack/techpackage-module/include/uapi
directory. The uapi headers pertaining to a techpackage-module
should reside under $KERNEL/techpack/techpackage-module/include/uapi
and associated Kbuild should have necessary rules to export it.
Change-Id: I0d0ced38566907d2074831edde0934833f666eff
Signed-off-by: Bryan Huntsman <bryanh@codeaurora.org>
Signed-off-by: Imran Khan <kimran@codeaurora.org>
[satyap@codeaurora.org: fix trivial merge conflict and add SPDX-License-Identifier]
Signed-off-by: Satya Durga Srinivasu Prabhala <satyap@codeaurora.org>
