Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 21 May 2009 18:34:26 -0400
From:      Jung-uk Kim <jkim@FreeBSD.org>
To:        freebsd-acpi@FreeBSD.org
Cc:        Andriy Gapon <avg@freebsd.org>
Subject:   Re: ACPICA version 20090521 released
Message-ID:  <200905211834.43324.jkim@FreeBSD.org>
In-Reply-To: <4911F71203A09E4D9981D27F9D8308582E0E5016@orsmsx503.amr.corp.intel.com>
References:  <4911F71203A09E4D9981D27F9D8308582E0E5016@orsmsx503.amr.corp.intel.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Thursday 21 May 2009 11:56 am, Moore, Robert wrote:
> 21 May 2009. Summary of changes for version 20090521:
>
> This release is available at www.acpica.org/downloads
>
> 1) ACPI CA Core Subsystem:
>
> Disabled the preservation of the SCI enable bit in the PM1 control
> register. The SCI enable bit (bit 0, SCI_EN) is defined by the ACPI
> specification to be a "preserved" bit - "OSPM always preserves this
> bit position", section 4.7.3.2.1. However, some machines fail if
> this bit is in fact preserved because the bit needs to be
> explicitly set by the OS as a workaround. No machines fail if the
> bit is not preserved. Therefore, ACPICA no longer attempts to
> preserve this bit.
>
> Fixed a problem in AcpiRsGetPciRoutingTableLength where an invalid
> or incorrectly formed _PRT package could cause a fault. Added
> validation to ensure that each package element is actually a
> sub-package.
>
> Implemented a new interface to install or override a single control
> method, AcpiInstallMethod. This interface is useful when debugging
> in order to repair an existing method or to install a missing
> method without having to override the entire ACPI table. See the
> ACPICA Programmer Reference for use and examples. Lin Ming, Bob
> Moore.
>
> Fixed several reference count issues with the DdbHandle object that
> is created from a Load or LoadTable operator. Prevent premature
> deletion of the object. Also, mark the object as invalid once the
> table has been unloaded. This is needed because the handle itself
> may not be deleted after the table unload, depending on whether it
> has been stored in a named object by the caller. Lin Ming.
>
> Fixed a problem with Mutex Sync Levels. Fixed a problem where if
> multiple mutexes of the same sync level are acquired but then not
> released in strict opposite order, the internally maintained
> Current Sync Level becomes confused and can cause subsequent
> execution errors. ACPICA BZ 471.
>
> Changed the allowable release order for ASL mutex objects. The ACPI
> 4.0 specification has been changed to make the SyncLevel for mutex
> objects more useful. When releasing a mutex, the SyncLevel of the
> mutex must now be the same as the current sync level. This makes
> more sense than the previous rule (SyncLevel less than or equal).
> This change updates the code to match the specification.
>
> Fixed a problem with the local version of the AcpiOsPurgeCache
> function. The (local) cache must be locked during all cache object
> deletions. Andrew Baumann.
>
> Updated the Load operator to use operation region interfaces. This
> replaces direct memory mapping with region access calls. Now, all
> region accesses go through the installed region handler as they
> should.
>
> Simplified and optimized the NsGetNextNode function. Reduced
> parameter count and reduced code for this frequently used function.
>
> Example Code and Data Size: These are the sizes for the
> OS-independent acpica.lib produced by the Microsoft Visual C++ 6.0
> 32-bit compiler. The debug version of the code includes the debug
> output trace mechanism and has a much larger code and data size.
>
>   Previous Release:
>     Non-Debug Version:  82.8K Code, 17.5K Data, 100.3K Total
>     Debug Version:     158.0K Code, 49.9K Data, 207.9K Total
>   Current Release:
>     Non-Debug Version:  83.4K Code, 17.5K Data, 100.9K Total
>     Debug Version:     158.9K Code, 50.0K Data, 208.9K Total
>
> 2) iASL Compiler/Disassembler and Tools:
>
> Disassembler: Fixed some issues with DMAR, HEST, MADT tables. Some
> problems with sub-table disassembly and handling invalid
> sub-tables. Attempt recovery after an invalid sub-table ID.

I just uploaded a new patchset against -CURRENT:

http://people.freebsd.org/~jkim/acpica-import-20090521.diff.gz

Please note that I am planning on merging it before 8.0 release.  
Therefore, it needs more regression tests and serious code reviews 
now.

Thanks,

Jung-uk Kim



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200905211834.43324.jkim>