| ??? 08/29/02 09:42 Read: times |
#28133 - RE: 89c66x & keil & ERAM |
Hi Stephane,
I agree fully with you, that using the newest compiler software was definitively not your problem. On the contrary a very new version can contain bugs, which are not present on earlier versions. E.g. on a posting was reported, that using all 256 bytes of pdata causes an error, but on my old version I can use all 256 bytes successful. But I suggest to read also the C51-Primer. Its an excellent document, which describes the different memory areas especially in view of programming in C. If tightness and speed was never an issue you can select the LARGE memory model and all of the 768 bytes of xdata can be used without any special specifier of the memory area was needed. But I use always another approach: I select the SMALL model and so all local variables placed into the data space. This causes, that most of the code was tight and fast. And only large data structures and seldom used variables are specified as xdata. Then after the project was almost completed, I view on the *.m51 file to see the real memory usage. Then if enough space free I move some xdata variables into data, idata and pdata. Especially interrupt handlers can profit a lot, e.g. on implementing a serial receive buffer. This can give about 10% and more of code space reduction. Peter |
| Topic | Author | Date |
| 89c66x & keil & ERAM | 01/01/70 00:00 | |
| RE: 89c66x & keil & ERAM | 01/01/70 00:00 | |
| RE: 89c66x & keil & ERAM | 01/01/70 00:00 | |
| RE: 89c66x & keil & ERAM | 01/01/70 00:00 | |
| RE: 89c66x & keil & ERAM | 01/01/70 00:00 | |
| RE: 89c66x & keil & ERAM | 01/01/70 00:00 | |
| RE: 89c66x & keil & ERAM | 01/01/70 00:00 | |
| RE: 89c66x & keil & ERAM | 01/01/70 00:00 | |
| RE: 89c66x & keil & ERAM | 01/01/70 00:00 | |
| RE: old crock software | 01/01/70 00:00 | |
| RE: 89c66x & keil & ERAM | 01/01/70 00:00 | |
| RE: 89c66x & keil & ERAM | 01/01/70 00:00 | |
| RE: IAR | 01/01/70 00:00 | |
| RE: 89c66x & keil & ERAM | 01/01/70 00:00 | |
| RE: IAR & Zilog | 01/01/70 00:00 | |
| RE: 89c66x & keil & ERAM | 01/01/70 00:00 | |
RE: 89c66x & keil & ERAM | 01/01/70 00:00 |



