Reply by Brad Griffis October 20, 20062006-10-20
Borrall Wonnell wrote:
> Hi all, > > TI's DSP/BIOS generates a <projname>cfg.s55 file. When I build my > project, this file gets compiled. There are two problems: > 1) I get a number of big, ugly remarks that I don't understand > 2) It takes a long time to compile > > The remarks I get are all similar, and look something like this: > > "samplecfg.s55", REMARK at line 1118: [R5110] Instructions encoded in > an > interrupt vector > slot may > not be executed as > part of > the interrupt > routine > MOV #31, mmap(@bios) ; write the interrupt > number in > > I'm not sure why these remarks occur, but they are all related to > hardware interrupts (when a HWI_Obj is defined). I haven't explicitly > set up any interrupts yet. Has anyone seen this before? The TI > representative I have been in contact with seemed to think they weren't > really anything to worry about....but it sure is a nag, and I'm not > 100% convinced that everything is ok. > > Cheers, > Dave >
The remark is a bit scary sounding but I've never encountered a problem with it in the BIOS code. I think everyone using BIOS on the 55x gets that same warning. I believe the warning was introduced in a particular rev of the code generation tools so if you just updated to a new version of CCS you're probably seeing that warning for the first time. It's still the same assembly code and there are no know issues with that particular code. I wouldn't worry about it as that code is being used by many customers in many different applications. Brad
Reply by Borrall Wonnell October 20, 20062006-10-20
Hi all,

TI's DSP/BIOS generates a <projname>cfg.s55 file.  When I build my
project, this file gets compiled.  There are two problems:
1) I get a number of big, ugly remarks that I don't understand
2) It takes a long time to compile

The remarks I get are all similar, and look something like this:

"samplecfg.s55", REMARK   at line 1118: [R5110] Instructions encoded in
an
                                                    interrupt vector
slot may
                                                    not be executed as
part of
                                                    the interrupt
routine
	                   MOV #31, mmap(@bios)          ; write the interrupt
number in

I'm not sure why these remarks occur, but they are all related to
hardware interrupts (when a HWI_Obj is defined).  I haven't explicitly
set up any interrupts yet.  Has anyone seen this before?  The TI
representative I have been in contact with seemed to think they weren't
really anything to worry about....but it sure is a nag, and I'm not
100% convinced that everything is ok.  

Cheers,
Dave