DCT Upgrade teststand requests ============================== Requester : Stephen, 14 Nov 02 Problem Description : Several memory read/write requests: - It would be very helpful if the block read and write file formats were the same so that the file output from a read operation could be used to write the memory later. The initial text is helpful -- maybe a comment character could be introduced so that the text can remain with the read output but be igored by the write? - Include 0x when displaying the data in the GUI after doing a read. Without the 0x, if I try to write back what I just read, I might get an invalid number since it will try to interpret the hex number as decimal. - hide writer/reader details : if I do a single read, it should use the one word reader. If I do a block write, it should use a block writer. Radio buttons will be needed to select between variable/fixed for new/old boards, but otherwise the user shouldn't need to know or care about different reader/writer types. - when doing a single word read, the text always says "block address 0" instead of the actual block address Implementation status: Closed Implementor's notes : Implemented in L1DctTeststand V00-00-02 Requester : Stephen, 9 Oct 02 Request Description : When doing a "read" operation on the newwin GUI, the checkboxes, etc. should be updated. Currently there is text output giving the results of the read, but the GUI itself is not updated with the new information. Implementation status: Open Implementor's notes : This works for memory enable/disable and play/record, but not for: - oneshot/continuous >>Done - load/run >> Done - LEDs >> Done - Data format (even the text window is broken on this one) Requester : Stephen, 28 Oct 02 Problem Description : Please add some documentation to the teststand webpage: - Where the GUI executable is located >> done - Updated command line documentation >> sort of done - Command line summary document with one line per command so that we don't have to scroll through pages of text to remember if it is "dctSetLeds" vs. "dctSetLEDs" or "dctWriteMemoryShort" vs. "dctWriteShort" or what their arguments are. >> done Implementation status: almost closed Implementor's notes : awaiting updated complete command line doc; low priority Requester : Stephen, 9 Oct 02 Problem Description : Newwin GUI: single reads and writes don't consistently work. Sometimes it appears that they don't actual read/write, sometimes they get into a confused state with errors on the screen, sometimes it confused the ROM bad enough that a reboot is required. Implementation status: Open 15 Oct 02, ROM fixed, GUI needs to be updated. Implementor's notes : Gerald. This was a conceptual error in the design of the "generic memory buffer". The software is fixed from the point of view of the ROM. However it implied some modifications in the interface. The newwin GUI needs to be updated accordingly. >> closed, 31 Oct 02 Requester : Stephen, 9 Oct 02 Problem Description : GUI block read does nothing. GUI block write does something, but apparently not a write. (If these do work, then the interface needs to be improved so that users can figure out how to use it...) Implementation status: Open (hopefully closed & date later) Implementor's notes : ! We are figuring how to write file to nfs disk. Once it is done, It will be implemented. Sorry for the inconvience, Xuedong >> closed, 31 Oct 02 Requester : Stephen, 9 Oct 02 Problem Description : newwin GUI diagnostic memory control: we find the non-standard implementation of checkboxes plus changing text to be confusing. e.g. when the checkbox next to "input record" is not checked, I would interpret this to mean that the input memory is not recording since the checkbox is not checked. But in this GUI, it means that it *is* recording. The standard implementation of a checkbox is that it is a true/false indicator of whatever the text next to it says. (This is somewhat a matter of taste, and once you know the system you know how to use it. But why?) Implementation status: Open (hopefully closed & date later) Implementor's notes : close --- Now, I agree with you, and have redone them. Thank you. >> closed Requester : Stephen, 9 Oct 02 Request Description : A several GUI (newwin) items: - make 16 bit Read/Write the default >> done - keep address field as hex when doing a read and increment >> done - block read/write directory should start with a reasonable default >> done - directory and file names should not require quotes >> done - can the slot mask be set to select multiple boards? (the drop down list is nice for a single board) >> done - Selecting a slot that was previously selected causes a complaint that the slot is already occupied >> done - In the text output field, there shouldn't be so many extra carriage returns. >> done - Diagnostic mem panel : is the "read back" portion necessary? Don't we always just want to read back the dmem control status and update the GUI? >> done - User reset and sync buttons seem broken. >> done Implementation status: Open (hopefully closed & date later) Implementor's notes : --- "make 16 bit Read/Write the default" is done --- keeping address field as hex is done --- Now, quotes is not needed ? But which directory do we should use? ! "HardwareBoard" can only work with one board --- currently you can ignore this complaint if you only have one board ! I am checking where "extra carriage" come from -- I think it is nessary to have a readback operation >> closed, 6 Nov 02 Requester : Stephen, 9 Oct 02 Problem Description : Could the generic board functionality of L1DHardwareBoard be put into L1DBoard so that other boards like L1DZpdBoard can inherit those features? Implementation status: Closed, Gerald, 9 Oct 02 Implementor's notes : This demands a redesign of the L1DBoard and derivative classes. The present design should be able to handle most of the use cases. This request will not be pursued until we find a use case that can't be done with the current design. Requester : name, date Problem Description : Implementation status: Open (hopefully closed & date later) Implementor's notes : Requester : name, date Problem Description : Implementation status: Open (hopefully closed & date later) Implementor's notes : Requester : name, date Problem Description : Implementation status: Open (hopefully closed & date later) Implementor's notes :