Variable p might've been optimized out. Try to qualify it as volatile.
[EDIT] If you can't display content of RAM as such, it might be also that the debugger does not work properly for whatever reason. "Blue pill" also sounds like a potential problem with not having a geniune STM32...
It's not clear from your screenshot if you're in debug mode at all. Even if you are, it does not appear execution is paused. The memory browser won't work unless you are in debug mode and execution is paused. It does not update continuously or when not connected to the chip.
You don't mention what chip you're on (STM32F103?), but unavailable addresses will show up as ???. If you only have 20kB of RAM, this is why the vast majority of the addresses show up as ???.
On top of those, the "Memory Browser" in Eclipse is not the most robust thing. Its functionality is the combination of several third-party tools that do not always work seamlessly together. Expect it to be finicky. You can also use the "Memory" tab with varying results.
"If you feel a post has answered your question, please click ""Accept as Solution""."
Variable p might've been optimized out. Try to qualify it as volatile.
[EDIT] If you can't display content of RAM as such, it might be also that the debugger does not work properly for whatever reason. "Blue pill" also sounds like a potential problem with not having a geniune STM32...