You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As the number of applets, many targeting very obscure hardware, grows, the need for applet regression tests is clear. In general, applets can be classified into three categories.
For the following applets, regression tests can be done by mocking the gateware interface wrapper (e.g. SPIMasterInterface for SPI flash and AVR flash, and so on). These are the easiest to test, have tests that run the fastest because it does not require simulating gateware, and have the most impact, because this is where most applet complexity lies.
i2c-bmp280
i2c-eeprom-24c
i2c-tps6598x
jtag-arc
jtag-mec16xx
jtag-mips
jtag-svf
jtag-xc9500
spi-flash-25c
spi-flash-avr
For the following applets, regression testing requires simulating gateware. This is slower, more complex and more fragile due to the need for non-cycle-accurate fuzzy matching, but is essential as these applets underlie many others. However, they both typically have less functionality and the hardware is easier to obtain.
hd44780
i2c-master
jtag
jtag-pinout
nand-flash
program-ice40
shugart-floppy
spi-master
swd
uart
The following applets are utility applets that only use Glasgow itself. They do not require regression testing or regression testing is meaningless for them due to their nature.
benchmark
selftest
The following applets cannot be regression tested within the generic framework due to reasons such as very high bit rate.
rgb_grabber
The text was updated successfully, but these errors were encountered:
As the number of applets, many targeting very obscure hardware, grows, the need for applet regression tests is clear. In general, applets can be classified into three categories.
For the following applets, regression tests can be done by mocking the gateware interface wrapper (e.g. SPIMasterInterface for SPI flash and AVR flash, and so on). These are the easiest to test, have tests that run the fastest because it does not require simulating gateware, and have the most impact, because this is where most applet complexity lies.
For the following applets, regression testing requires simulating gateware. This is slower, more complex and more fragile due to the need for non-cycle-accurate fuzzy matching, but is essential as these applets underlie many others. However, they both typically have less functionality and the hardware is easier to obtain.
The following applets are utility applets that only use Glasgow itself. They do not require regression testing or regression testing is meaningless for them due to their nature.
The following applets cannot be regression tested within the generic framework due to reasons such as very high bit rate.
The text was updated successfully, but these errors were encountered: