thaks all for generating ideas , but i didnt get answer for my question, does transistor buffer works for spi or not?
If I understand you correctly, you're not just looking to solve an actual problem, but also curious if your thinking and possible solution is correct.
I think part of the reason that you haven't gotten a really good answer to that might be that there's a lot of variables at play, and hard to give an answer that's correct for all of them.
As a general answer, yes, it should be a workable to use a transistor-buffer for SPI, as long as you keep an eye on not getting an inverted signal etc.
There is however a lot of factors to consider. Take speed for example. How fast will the SPI-bus be? Will bandwidth/rise-time be fast enough?
Figuring out if everything will always behave the right way - with margins - is quite a bit of work, and virtually impossible without diving into details.
Depending on specific, it's quite possible that a 3.3V signal can register correctly as HIGH on a 5V micro.
For the other direction, you might be able to simply cut down the 5V signal to 3.3V-safe levels using a resistor divider.