Contribute
Register

DPCIManager 1.5 Open Beta

Status
Not open for further replies.
Joined
Dec 3, 2010
Messages
460
Motherboard
Gigabyte GA-H55M-S2V
CPU
Intel i3-530
Graphics
HIS HD 6570
Mac
  1. iMac
Mobile Phone
  1. Android
So, I'd like to move this project along as well, though I don't have a solution for dmidecode on Clover yet. BIOS flashing still works, but it will not detect your motherboard until I find a way to make minimal edits to the normal package to support it. Otherwise, this build adds:
  • "MSRDumper" functionality through DirectHW and the log panel (PStates)
  • Detection of "stub" MAC addresses from incomplete ASUS BIOS flashes
  • Built in reporter for flashrom tasks (please submit your reports!)
  • Motherboard and bios version detection for ROM flashing
  • Chimera and Clover boot log support
  • Better Status page failure recovery
  • Includes superiotool, dmidecode, and fd44copier
  • Better kext installation (still not perfect)
  • VBios ROM fetching
  • Better CMOS method (not implemented in GUI)
  • Better unCAPing of ASUS ROMs, and FD44 copy
  • Better NSTask and logging methods to prevent thread and timing issues
  • Stricter detection of PCI IDs, unaffected by DSDT device-id injection
  • Faster PCI.ID database reader
  • Bundled `dscpi` binary to implement scriptable `lspci` functionality (no root or other files required!)
  • Kext permission repair for DirectHW (caused problems on FAT32 filesystems)
The reporter is currently "off", but I will be turning it on again as reports come in.

DPCIManager 1.5 available in source and binary at SourceForge
Three screenshots
 
I forgot to mention, PStates is currently in the toolbar "drawer". Customize the toolbar to add the item and use it.
 
This is an interesting and very useful app. And I know it's in beta and all but I've got some suggestions you may like (or not):

- Group the "Repair Permissions", "Rebuild Cache" and "Install Kext" functions into another tab, just like "Status", "PCI List" and "Misc".
- The 2 "Update" and "Submit List" buttons should be in the "PCI List" tab (or just avoid the update buttons by auto updating the DB at boot or once a day; I'm also assuming the "Submit List" button refers to the PCI List only)
- The PStates function is awesome, but it is not clear if you're "unloading" it as you can only close the log window (in fact it is still active even after closing the log window)
- The app doesn't remember windows size, position or toolbar customization (I guess you didn't work on that yet)

Hope to see this app grow up into a great information and utility center.

EDIT: I would also like to suggest you to post your work at the InsanelyMac forum too, because I believe there are a lot more developers and advanced users there that could help you improve the app in its beta stage.

EDIT2: Looks like pressing the P-States button again stops it
 
ok, next beta #229 is up. This fixes:
  • CFtype convenience method to handle incorrectly-injected DSDT properties
  • Better handling of nonexistent PCI devices (says 0000 0000)
  • Sortable PCI List
  • UI element autosaving
  • flashrom updated to r1646
  • PMPatch updated to 5.10
  • Clearer indication PStates is running (button will depress)
  • Clearer indication PStates has exited
  • Unified OF matching
  • More semantic alert messages
  • dmidecode protection in case of failure
I haven't found the root cause of a very rare dmidecode error, but I can't reproduce it in any Clover installation I have access to, so I've simply put up guardrails. If dmidecode returns "" or contains "Stop.", DPCI will warn and prevent flashrom access to the program.
 
to those wondering (and commenting) on PStates' refresh rate, it doesn't echo to the log at every request, to keep usage down and avoid skewing the results. The current internal refresh rate is "usleep(kMillisecondScale/5);", and echoes to the log every 25th refresh. The list of states is kept in an NSSet. If you find PStates exits prematurely (as I did when trying iMac13,1), please file a report with the exit code.
 
next beta #230 is up. This fixes:
  • Not all fd44 pathways reported success/failure, now much better at halting after failure
  • Updated PCI.IDs
 
next beta #231 is up. This fixes:
  • Strange fileSystemRepresentation issue with NULLs
  • flashrom tasks now output to a single pipe, should keep output aligned even in rare cases
 
can you share a link of this new version?
 
P-States, i get:

P States: 24
Current State: 24
P States: 12, 24
Current State: 24
P-State method failed, exiting 0
 
Status
Not open for further replies.
Back
Top