Since we already disable this mechanism in B5/R3.
So we remove it.
Bug: 179620905
Test: Build pass. Basic function work.
Test with generate bugreport success.
Change-Id: Iaa9932928afc93934e570cb9f4adf694ce68cbb3
Let's remove segment_info which seems not useful.
Bug: 184964835
Signed-off-by: Jaegeuk Kim <jaegeuk@google.com>
Change-Id: Ib9826f536681d9f36f9d55dab3801a05c47f27a4
Bug: 169662280
The timing of dumpstate to enable diag_mdlog is too marginal to
diag_mdlog process exiting. The starting of diag_mdlog would fail if
diag_mdlog process still exists. Change is to leave some time for
diag_mdlog process to exit before it is enabled again by modifying the
system property vendor.sys.modem.diag.mdlog to true.
Change-Id: I825cd77e3f6e8fe6ce395cbe2c21974f1cb29d02
Originally dumpstate copies files by calling /vendor/bin/cp with execvp.
We are seeing some stuck issues while copying files and taring files.
Change is to copy and tar files in place instead of calling system shell.
Bug: 169465773
Change-Id: I9878bde8e540d41c1b498c841079af38f4ec502c
Dump fastrpc dma buffer size so we could understand
the memory usage of fastrpc.
Bug: 168084904
Test: boot
Signed-off-by: Martin Liu <liumartin@google.com>
Merged-In: Iefa6795ded8d8abc377ec6a0f2237649d0164efa
Change-Id: Iefa6795ded8d8abc377ec6a0f2237649d0164efa
Dump fastrpc dma buffer size so we could understand
the memory usage of fastrpc.
Bug: 168084904
Test: boot
Signed-off-by: Martin Liu <liumartin@google.com>
Change-Id: Iefa6795ded8d8abc377ec6a0f2237649d0164efa
We have only 30sec budget to finish dunmpstate HAL so rollback to
30sec and then find a way to improve it whatever.
Bug: 168344541
Signed-off-by: Minchan Kim <minchan@google.com>
Change-Id: I4f6eff5130ca04ede47c503f00c3d76a660c1ce8
Page Owner dump would take long time if memory size would be bigger.
To be sure, let's increase the timeout to 60sec.
Bug: 168344541
Signed-off-by: Minchan Kim <minchan@google.com>
Change-Id: I0932839b2047916f9ebcb5ecfe6009afb90d5e52
If kernel supports page_owner, dump it at bugreport so that it could
help memory leak problem.
Bug: 168344541
Signed-off-by: Minchan Kim <minchan@google.com>
Change-Id: I183ec4e9a510d52f838db811237417ce3846274f
There is a risk that if we force active the kernel by touch driver but
fail to set it back, it will cause the kernel cannot suspend and imapct
power consumption.
Bug: 159576114
Test: trigger butreport.
Signed-off-by: davidycchen <davidycchen@google.com>
Change-Id: I58eb8661737ada43dcec13bbbe4ae30dde347f9d
In order to clarify and improve video jank problem, we plan to add
some CamNode log into bugreport
Bug: 158243741
Change-Id: Id93e956472286cf8969ec5e96f762334957a50b8
convert traffic stats log(extended_log_datastall_pkt.txt) to pcap
when dump bugreport.
Bug: 154492175
Test: 1. I do see extended_log_datastall.pcap file
in dumstate_board.bin when I trigger bugreport.
(data/vendor/radio/extended_logs/extended_log_datastall.txt)
2. I have do the test about incomplete/truncated logs, pcap just
can parse to wrong information but not impact the dumpstate
process.
Change-Id: Iea892bc66bbce8d7f219ca1cde3701cebb35ff80
add panel information to dumpstate_board.txt
Bug: 143331062
Test: adb bugreport and verify logs in dumpstate_board.txt
Change-Id: I056ab3a2d3079ea256da5b4bbb59ff9adcbff0fd
The DumpUFS function cannot work with the real sepolicy rules. So
refactor the function to fix the problem.
Bug: 149468879
Test: run bugreport
Change-Id: I9b260d18f4cc768cba98b64d76a149241e17c22b
Signed-off-by: Leo Liou <leoliou@google.com>
The PROTO mode was added after initial HAL upgrade and is causing VTS
failures due to the way the DumpstateMode param is being validated.
Bug: 150873571
Test: make (no device run VTS, older devices pass with identical change)
Change-Id: Icc4e93c48e65244463db6313de7dee3af098831c
(cherry picked from commit d1ee906172)
The /data/vendor/display/mdp_xlog is for mdp driver but not for DRM/KMS.
The bugreport should not need dump it.
Bug: 149467103
Test: There is no failed selinux log of display_vendor_data_file.
Change-Id: Iae9758234c3a1eb97c2fd1b6de5a9415fa991bcd
Change touch ITO test to the last test item because ITO test will
trigger touch hardware reset which may affect the rawdata value of other
test items.
Test: trigger bugreport and check the touch information is correct in
dumpstate.
Bug: 147336072
Change-Id: Iad903e70584f28e824e9a12da442fae553d97cfb
Signed-off-by: davidycchen <davidycchen@google.com>
Gather Citadel's firmware version, uptime stats, and motherboard type in all
bug reports. None of this can be used to identify a specific phone, so it's
safe to run in production builds.
Previously we were running --id and --selftest. Because these could be used
to indentify a specific Citadel chip, we only gathered it on userdebug and
test builds. We used that to isolate problems with early Citadel samples in
prototype phones, which we've done, fixed, and replaced. We don't need that
info anymore.
Bug: 141235263
Test: manual
Change-Id: I6e4d3817fc7f3793956c60f6189ac1ff84b2ae55
Merged-In: I6e4d3817fc7f3793956c60f6189ac1ff84b2ae55
Signed-off-by: Bill Richardson <wfrichar@google.com>
(cherry picked from commit ea5ca7374c)
