[Touch-packages] [Bug 1952083] Re: Add support for Beige Goby

2022-01-03 Thread Alex Hung
** Changed in: amd
   Status: New => Fix Released

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to mesa in Ubuntu.
https://bugs.launchpad.net/bugs/1952083

Title:
  Add support for Beige Goby

Status in amd:
  Fix Released
Status in OEM Priority Project:
  Fix Released
Status in mesa package in Ubuntu:
  Fix Released
Status in mesa source package in Focal:
  Fix Released

Bug description:
  Impact]

  New hardware support for AMD's Beige Goby needs a commit backported to
  mesa. This is only needed on focal, skipping hirsute (no kernel
  support there). Fixed in impish and up.

  [Test plan]

  Install updates, boot a BG machine and check that the desktop has full
  hardware acceleration.

  [Where problems could occur]

  Mesa adds a single commit for enabling BEIGE_GOBY, hard to see what
  could go wrong.

To manage notifications about this bug go to:
https://bugs.launchpad.net/amd/+bug/1952083/+subscriptions


-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1888352] Re: use builtin dump_acpi_tables.py in hookutils

2020-09-08 Thread Alex Hung
After using apport-unpack for #8, acpidump looks good

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apport in Ubuntu.
https://bugs.launchpad.net/bugs/1888352

Title:
  use builtin dump_acpi_tables.py in hookutils

Status in Apport:
  New
Status in OEM Priority Project:
  In Progress
Status in apport package in Ubuntu:
  Fix Released
Status in apport source package in Focal:
  Fix Committed
Status in apport source package in Groovy:
  Fix Released

Bug description:
  1. add apcidump to hookutils.py:attach_hardware and use the builtin 
dump_acpi_tables.py.
  2. remove explicit acpidump from oem-getlogs to use the builtin one.
  3. refine usage string in oem-getlogs.

  To SRU to focal:

  [Impact]

   * for OEM project, lts is been used. And collect log is important.
 With built-in tool to get acpidump, we don't need to install
 extra tool in the end-customer's machine. That make it much
 easier for the oem process.

   * By call built-in utility, we no lounger need to install extra
 tools to collect data that's both complete and convenient for
 HWE people to work on.

  [Test Case]

   * before this applied, run oem-getlog without install acpidump. 
 we can't get the dump data. After this applied, we can just dump
 the data HWE need.

   * test step: install the new package, run

 sudo -E oem-getlogs [-c case_id] to get the

 use apport-unpack to unpack the apport file.

 check the acpidump file. HWE people know much better on check
 the acpidump file. Give the dump_acpi_tables.py just updated and SRUed
 by HWE/ACPI/UEFI expert, it's pretty safe to do so.

  [Regression Potential]

   * Given the modification change the way to collect log, even it
 failed, it won't break apport itself. Just the collected log
 might contain data not so valid.

   * Given acpidump mostly used by HWE/ACIP/UEFI expert, and they
 just reviewed and updated the dump_acpi_tables.py script, I
 believe it will have good quality.

To manage notifications about this bug go to:
https://bugs.launchpad.net/apport/+bug/1888352/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1888352] Re: use builtin dump_acpi_tables.py in hookutils

2020-09-08 Thread Alex Hung
@ycheng-twn,

#10 and #11 are basically the same except orders of some tables; however
this is as expected.

the gz file of #8 contains a "single" plain text file. Shouldn't it be
one file for each log file?

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apport in Ubuntu.
https://bugs.launchpad.net/bugs/1888352

Title:
  use builtin dump_acpi_tables.py in hookutils

Status in Apport:
  New
Status in OEM Priority Project:
  In Progress
Status in apport package in Ubuntu:
  Fix Released
Status in apport source package in Focal:
  Fix Committed
Status in apport source package in Groovy:
  Fix Released

Bug description:
  1. add apcidump to hookutils.py:attach_hardware and use the builtin 
dump_acpi_tables.py.
  2. remove explicit acpidump from oem-getlogs to use the builtin one.
  3. refine usage string in oem-getlogs.

  To SRU to focal:

  [Impact]

   * for OEM project, lts is been used. And collect log is important.
 With built-in tool to get acpidump, we don't need to install
 extra tool in the end-customer's machine. That make it much
 easier for the oem process.

   * By call built-in utility, we no lounger need to install extra
 tools to collect data that's both complete and convenient for
 HWE people to work on.

  [Test Case]

   * before this applied, run oem-getlog without install acpidump. 
 we can't get the dump data. After this applied, we can just dump
 the data HWE need.

   * test step: install the new package, run

 sudo -E oem-getlogs [-c case_id] to get the

 use apport-unpack to unpack the apport file.

 check the acpidump file. HWE people know much better on check
 the acpidump file. Give the dump_acpi_tables.py just updated and SRUed
 by HWE/ACPI/UEFI expert, it's pretty safe to do so.

  [Regression Potential]

   * Given the modification change the way to collect log, even it
 failed, it won't break apport itself. Just the collected log
 might contain data not so valid.

   * Given acpidump mostly used by HWE/ACIP/UEFI expert, and they
 just reviewed and updated the dump_acpi_tables.py script, I
 believe it will have good quality.

To manage notifications about this bug go to:
https://bugs.launchpad.net/apport/+bug/1888352/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1888352] Re: use builtin dump_acpi_tables.py in hookutils

2020-09-08 Thread Alex Hung
@ycheng-twn,

1. Is it normal that everything is in a plain text file?
2. There is an extra space for acpidump. This causes acpixtract to fail to 
extract ACPI tables.
3. The content looks "reasonable", but /sys/firmware/acpi/tables/* are required 
to byte-to-byte compare after 1 & 2 are fixed.

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apport in Ubuntu.
https://bugs.launchpad.net/bugs/1888352

Title:
  use builtin dump_acpi_tables.py in hookutils

Status in Apport:
  New
Status in OEM Priority Project:
  In Progress
Status in apport package in Ubuntu:
  Fix Released
Status in apport source package in Focal:
  Fix Committed
Status in apport source package in Groovy:
  Fix Released

Bug description:
  1. add apcidump to hookutils.py:attach_hardware and use the builtin 
dump_acpi_tables.py.
  2. remove explicit acpidump from oem-getlogs to use the builtin one.
  3. refine usage string in oem-getlogs.

  To SRU to focal:

  [Impact]

   * for OEM project, lts is been used. And collect log is important.
 With built-in tool to get acpidump, we don't need to install
 extra tool in the end-customer's machine. That make it much
 easier for the oem process.

   * By call built-in utility, we no lounger need to install extra
 tools to collect data that's both complete and convenient for
 HWE people to work on.

  [Test Case]

   * before this applied, run oem-getlog without install acpidump. 
 we can't get the dump data. After this applied, we can just dump
 the data HWE need.

   * test step: install the new package, run

 sudo -E oem-getlogs [-c case_id] to get the

 use apport-unpack to unpack the apport file.

 check the acpidump file. HWE people know much better on check
 the acpidump file. Give the dump_acpi_tables.py just updated and SRUed
 by HWE/ACPI/UEFI expert, it's pretty safe to do so.

  [Regression Potential]

   * Given the modification change the way to collect log, even it
 failed, it won't break apport itself. Just the collected log
 might contain data not so valid.

   * Given acpidump mostly used by HWE/ACIP/UEFI expert, and they
 just reviewed and updated the dump_acpi_tables.py script, I
 believe it will have good quality.

To manage notifications about this bug go to:
https://bugs.launchpad.net/apport/+bug/1888352/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1883027] Re: dump_acpi_tables.py: fix incorrect output and change format

2020-07-02 Thread Alex Hung
Proposed package is tested (version below) and this fixes
dump_acpi_tables.py.

~$ lsb_release -c ; apt-cache policy apport
Codename:   focal
apport:
  Installed: 2.20.11-0ubuntu27.4
  Candidate: 2.20.11-0ubuntu27.4



** Tags removed: verification-needed verification-needed-focal
** Tags added: verification-done verification-done-focal

** Tags removed: verification-done
** Tags added: verification-needed

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apport in Ubuntu.
https://bugs.launchpad.net/bugs/1883027

Title:
  dump_acpi_tables.py: fix incorrect output and change format

Status in Apport:
  Confirmed
Status in apport package in Ubuntu:
  Fix Released
Status in apport source package in Focal:
  Fix Committed
Status in apport source package in Groovy:
  Fix Released

Bug description:
  [Impact]

    dump_acpi_tables.py does not always generate 4-char table names
  (which is defined by ACPI spec), and it can miss or duplicate data of
  ACPI tables in /sys/firmware/acpi/tables.

  [Test Case]
    Tested with ACPI tables from Dell's Precision 3530. The outputs are almost 
the same as ones from "acpidump". The difference is the order of ACPI tables in 
the output but the order is not important.

1. run dump_acpi_tables.py
2. run acpidump 
3. compare the outputs

Note: acpidump is from acpica-tools

  [Regression Potential]

    Formats were changed to be the same as outputs from "acpidump", but
  known tools, such as acpiexec and fwts, by default reads outputs from
  acpidump and they do not reply on spaces and upper/lower cases.

  [Other Info]

    See original bug reports as below:

  dump_acpi_tables.py generates log files that is similar as the output
  from "acpidump" in acpica-tools. The output can be passed to other
  utilities in acpica-tools and fwts.

  It, however, has some subtle differences and some errors.

  Summary:

  1. ACPI tables have 4-char signatures - meaning SSDT's are SSDT, not
  SSDT1, SSDT2 and so on. Each table is unique by its table ID.

  Original:
    SSDT1, SSDT2, SSDT3 ...

  Changed:
    SSDT, SSDT, SSDT ...

  2. (Minor) acpidump outputs are all in upper cases

  3. (Minor) acpidump offset are aligned by data, not address

  Original:
    FFF0: 53 41 56 43 52 44 43 41 4E 43 52 4E 72 4E 50 4D  SAVCRDCANCRNrNPM
    1: 56 0A 04 00 0C FC FF FF FF 0A 03 0A 03 52 44 43  VRDC

  Changed:
  FFF0: 53 41 56 43 52 44 43 41 4E 43 52 4E 72 4E 50 4D  SAVCRDCANCRNrNPM
     1: 56 0A 04 00 0C FC FF FF FF 0A 03 0A 03 52 44 43  VRDC

  4. (Bug) dump_acpi_tables.py generates an extra line when data sizes
  are multiple of 16 bytes

  Original:
  05F0: 30 30 0B 00 08 00 A4 48 50 53 44 A4 53 50 53 44  00.HPSD.SPSD
  05F0: 30 30 0B 00 08 00 A4 48 50 53 44 A4 53 50 53 44 
 00.HPSD.SPSD

  Corrected:
  05F0: 30 30 0B 00 08 00 A4 48 50 53 44 A4 53 50 53 44  00.HPSD.SPSD

  5. (Bug) dump_acpi_tables.py misses an line when data sizes are
  multiple of 15 bytes

  Original:
  07E0: 42 0A 87 5C 2F 03 5F 53 42 5F 55 42 54 43 43 43  B..\/._SB_UBTCCC

  Corrected:
  07E0: 42 0A 87 5C 2F 03 5F 53 42 5F 55 42 54 43 43 43  B..\/._SB_UBTCCC
  07F0: 49 33 86 5C 2E 5F 53 42 5F 55 42 54 43 0A 80 I3.\._SB_UBTC..

To manage notifications about this bug go to:
https://bugs.launchpad.net/apport/+bug/1883027/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp


[Touch-packages] [Bug 1883027] Re: dump_acpi_tables.py: fix incorrect output and change format

2020-06-24 Thread Alex Hung
@brian-murray,

Bug description is updated. Please keep me posted if it needs more
information. Thanks

** Description changed:

+ [Impact]
+ 
+   dump_acpi_tables.py does not always generate 4-char table names (which
+ is defined by ACPI spec) and can miss or duplicate data of ACPI tables
+ in /sys/firmware/acpi/tables.
+ 
+ 
+ [Test Case]
+   Tested with ACPI tables from Dell's Precision 3530. The outputs are almost 
the same as ones from "acpidump". The difference is the order of ACPI tables in 
the output but the order is not important.
+ 
+ [Regression Potential]
+ 
+   Formats were changed to be the same as outputs from "acpidump", but
+ known tools, such as acpiexec and fwts, by default reads outputs from
+ acpidump and they do not reply on spaces and upper/lower cases.
+ 
+ [Other Info]
+ 
+   See original bug reports as below:
+ 
  dump_acpi_tables.py generates log files that is similar as the output
  from "acpidump" in acpica-tools. The output can be passed to other
  utilities in acpica-tools and fwts.
  
  It, however, has some subtle differences and some errors.
  
  Summary:
  
  1. ACPI tables have 4-char signatures - meaning SSDT's are SSDT, not
  SSDT1, SSDT2 and so on. Each table is unique by its table ID.
  
  Original:
    SSDT1, SSDT2, SSDT3 ...
  
  Changed:
    SSDT, SSDT, SSDT ...
  
  2. (Minor) acpidump outputs are all in upper cases
  
  3. (Minor) acpidump offset are aligned by data, not address
  
  Original:
    FFF0: 53 41 56 43 52 44 43 41 4E 43 52 4E 72 4E 50 4D  SAVCRDCANCRNrNPM
    1: 56 0A 04 00 0C FC FF FF FF 0A 03 0A 03 52 44 43  VRDC
  
  Changed:
  FFF0: 53 41 56 43 52 44 43 41 4E 43 52 4E 72 4E 50 4D  SAVCRDCANCRNrNPM
     1: 56 0A 04 00 0C FC FF FF FF 0A 03 0A 03 52 44 43  VRDC
  
  4. (Bug) dump_acpi_tables.py generates an extra line when data sizes are
  multiple of 16 bytes
  
  Original:
  05F0: 30 30 0B 00 08 00 A4 48 50 53 44 A4 53 50 53 44  00.HPSD.SPSD
  05F0: 30 30 0B 00 08 00 A4 48 50 53 44 A4 53 50 53 44 
 00.HPSD.SPSD
  
  Corrected:
  05F0: 30 30 0B 00 08 00 A4 48 50 53 44 A4 53 50 53 44  00.HPSD.SPSD
  
  5. (Bug) dump_acpi_tables.py misses an line when data sizes are multiple
  of 15 bytes
  
  Original:
  07E0: 42 0A 87 5C 2F 03 5F 53 42 5F 55 42 54 43 43 43  B..\/._SB_UBTCCC
  
  Corrected:
  07E0: 42 0A 87 5C 2F 03 5F 53 42 5F 55 42 54 43 43 43  B..\/._SB_UBTCCC
  07F0: 49 33 86 5C 2E 5F 53 42 5F 55 42 54 43 0A 80 I3.\._SB_UBTC..

** Description changed:

  [Impact]
  
-   dump_acpi_tables.py does not always generate 4-char table names (which
- is defined by ACPI spec) and can miss or duplicate data of ACPI tables
- in /sys/firmware/acpi/tables.
- 
+   dump_acpi_tables.py does not always generate 4-char table names (which
+ is defined by ACPI spec), and it can miss or duplicate data of ACPI
+ tables in /sys/firmware/acpi/tables.
  
  [Test Case]
-   Tested with ACPI tables from Dell's Precision 3530. The outputs are almost 
the same as ones from "acpidump". The difference is the order of ACPI tables in 
the output but the order is not important.
+   Tested with ACPI tables from Dell's Precision 3530. The outputs are almost 
the same as ones from "acpidump". The difference is the order of ACPI tables in 
the output but the order is not important.
+ 
+   1. run dump_acpi_tables.py
+   2. run acpidump 
+   3. compare the outputs
+ 
+   Note: acpidump is from acpica-tools
  
  [Regression Potential]
  
-   Formats were changed to be the same as outputs from "acpidump", but
+   Formats were changed to be the same as outputs from "acpidump", but
  known tools, such as acpiexec and fwts, by default reads outputs from
  acpidump and they do not reply on spaces and upper/lower cases.
  
  [Other Info]
  
-   See original bug reports as below:
+   See original bug reports as below:
  
  dump_acpi_tables.py generates log files that is similar as the output
  from "acpidump" in acpica-tools. The output can be passed to other
  utilities in acpica-tools and fwts.
  
  It, however, has some subtle differences and some errors.
  
  Summary:
  
  1. ACPI tables have 4-char signatures - meaning SSDT's are SSDT, not
  SSDT1, SSDT2 and so on. Each table is unique by its table ID.
  
  Original:
    SSDT1, SSDT2, SSDT3 ...
  
  Changed:
    SSDT, SSDT, SSDT ...
  
  2. (Minor) acpidump outputs are all in upper cases
  
  3. (Minor) acpidump offset are aligned by data, not address
  
  Original:
    FFF0: 53 41 56 43 52 44 43 41 4E 43 52 4E 72 4E 50 4D  SAVCRDCANCRNrNPM
    1: 56 0A 04 00 0C FC FF FF FF 0A 03 0A 03 52 44 43  VRDC
  
  Changed:
  FFF0: 53 41 56 43 52 44 43 41 4E 43 52 4E 72 4E 50 4D  SAVCRDCANCRNrNPM
     1: 56 0A 04 00 0C FC FF FF FF 0A 03 0A 03 52 44 43  VRDC
  
  4. (Bug) dump_acpi_tables.py generates an extra line when data sizes are
  multiple of 16 bytes
  
  Original:
  05F0: 30 30 

[Touch-packages] [Bug 1883027] Re: dump_acpi_tables.py: fix incorrect output and change format

2020-06-16 Thread Alex Hung
** Also affects: apport (Ubuntu)
   Importance: Undecided
   Status: New

-- 
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to apport in Ubuntu.
https://bugs.launchpad.net/bugs/1883027

Title:
  dump_acpi_tables.py: fix incorrect output and change format

Status in Apport:
  New
Status in apport package in Ubuntu:
  New

Bug description:
  dump_acpi_tables.py generates log files that is similar as the output
  from "acpidump" in acpica-tools. The output can be past to other
  utilities in acpica-tools and / or fwts.

  It, however, has some subtle differences and some errors.

  Summary:

  1. ACPI tables have 4-char signatures - meaning SSDT's are SSDT, not
  SSDT1, SSDT2 and so on. Each table is unique by its table ID.

  Original:
    SSDT1, SSDT2, SSDT3 ...

  Changed:
    SSDT, SSDT, SSDT ...

  2. (Minor) acpidump outputs are all in upper cases

  3. (Minor) acpidump offset are aligned by data, not address

  Original:
    FFF0: 53 41 56 43 52 44 43 41 4E 43 52 4E 72 4E 50 4D  SAVCRDCANCRNrNPM
    1: 56 0A 04 00 0C FC FF FF FF 0A 03 0A 03 52 44 43  VRDC

  Changed:
  FFF0: 53 41 56 43 52 44 43 41 4E 43 52 4E 72 4E 50 4D  SAVCRDCANCRNrNPM
     1: 56 0A 04 00 0C FC FF FF FF 0A 03 0A 03 52 44 43  VRDC

  4. (Bug) dump_acpi_tables.py generates an extra line when data sizes
  are multiple of 16 bytes

  Original:
  05F0: 30 30 0B 00 08 00 A4 48 50 53 44 A4 53 50 53 44  00.HPSD.SPSD
  05F0: 30 30 0B 00 08 00 A4 48 50 53 44 A4 53 50 53 44 
 00.HPSD.SPSD

  Corrected:
  05F0: 30 30 0B 00 08 00 A4 48 50 53 44 A4 53 50 53 44  00.HPSD.SPSD

  5. (Bug) dump_acpi_tables.py misses an line when data sizes are
  multiple of 15 bytes

  Original:
  07E0: 42 0A 87 5C 2F 03 5F 53 42 5F 55 42 54 43 43 43  B..\/._SB_UBTCCC

  Corrected:
  07E0: 42 0A 87 5C 2F 03 5F 53 42 5F 55 42 54 43 43 43  B..\/._SB_UBTCCC
  07F0: 49 33 86 5C 2E 5F 53 42 5F 55 42 54 43 0A 80 I3.\._SB_UBTC..

To manage notifications about this bug go to:
https://bugs.launchpad.net/apport/+bug/1883027/+subscriptions

-- 
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help   : https://help.launchpad.net/ListHelp