Hi all, Another week, some more features!
A new version of fwts is now available in the PPA ppa:firmware-testing-team/ppa-firmware-test-suite-natty-stable and also it has been uploaded to the Natty 11.04 universe archive. I'm bumping the version to 0.21.0 as this now marks the end of the significant ACPICA core development changes found in 0.20.x Changes: * syntaxcheck test: This test disassembles and re-assembles the ACPI AML byte code and picks up syntactic and some semantic bugs. This picks up bugs that are introduced by the less strict Microsoft compiler. In this release I've expanded the error message checking so it provides detailed explanations of errors, for example: 00012 syntaxcheck Re-assemble DSDT and find syntax errors and warnings. 00013 syntaxcheck ----------------------------------------------------------------------------- 00014 syntaxcheck Test 1 of 2: Disassemble and reassemble DSDT 00015 syntaxcheck Line | AML source 00016 syntaxcheck ----------------------------------------------------------------------------- 00017 syntaxcheck 01067| 0x00000000, // Range Minimum 00018 syntaxcheck 01068| 0xDFFFFFFF, // Range Maximum 00019 syntaxcheck 01069| 0x00000000, // Translation Offset 00020 syntaxcheck 01070| 0x00000000, // Length 00021 syntaxcheck | ^ 00022 syntaxcheck | error 4122: Invalid combination of Length and Min/Max fixed flags 00023 syntaxcheck 01071| ,, _Y0D, AddressRangeMemory, TypeStatic) 00024 syntaxcheck 01072| DWordMemory (ResourceProducer, PosDecode, MinFixed, MaxFixed, Cacheable, ReadWrite, 00025 syntaxcheck 01073| 0x00000000, // Granularity 00026 syntaxcheck ============================================================================= 00027 syntaxcheck 00028 syntaxcheck ADVICE: (for error #4122): This occurs if the length is zero and just one of 00029 syntaxcheck the resource MIF/MAF flags are set, or the length is non-zero and resource MIF 00030 syntaxcheck /MAF flags are both set. These are illegal combinations and need to be fixed. 00031 syntaxcheck See section 6.4.3.5 Address Space Resource Descriptors of version 4.0a of the 00032 syntaxcheck ACPI specification for more details. Note that there is a ^ to point to where the error is (line 00021 of the above log), followed by the compiler error message, and then where appropriate an ADVICE line to explain this error in more details. ..I went through all the 120+ errors, located them in the compiler, figured out the reason for the error and added the appropriate advice text to the test. Hopefully this is a worth while exercise. Incidentally, a PDF copy of the ACPI spec can be found at: http://www.acpi.info/DOWNLOADS/ACPIspec40a.pdf * S4 extra option: --s4-sleep-delay=N has been added to specify the amount of time in seconds between initiating the hibernate and the wakeup from hibernate. We need to factor in the time it takes to write the hibernate image to disk, and this can be variable depending on the speed of the swap device and the amount of memory that needs to be dumped to disk. The default is 90 seconds, but with this option one can override the default. In theory we could automatically figure out the optimal sleep delay time, but since hibernate times can vary per S4 cycle, in practice it is more reliable for a human to specify this per use-case. * Disassemble AML option: --disassemble-aml is a new option that will disassemble all AML byte code in DSDT and SSDT ACPI tables and dump the output into files DSDT.dsl and SSDTn.dsl (where n is the nth SSDT table). This is quicker and more convenient than the normal way of dumping the assembler using the acpixtract, acpidump and iasl tools. Bug fixes: * Report month correctly in log. Off by one error. Doh. That's about it for this week. Please keep providing me we with feedback! Colin _______________________________________________ Mailing list: https://launchpad.net/~firmware-testing-team Post to : [email protected] Unsubscribe : https://launchpad.net/~firmware-testing-team More help : https://help.launchpad.net/ListHelp
