/usr/bin/g++ -DHAVE_CONFIG_H -I..... -DHAVE_ERRNO_AS_DEFINE=1 -g -pthread -O2 -M..... -c -o sched_customize.o sched_customize.cpp

sched_customize.cpp: In function ‘int app_plan(SCHEDULER_REQUEST&, char*, HOST_USAGE&)’:
sched_customize.cpp:111: error: ‘struct COPROC_ATI’ has no member named ‘flops’
 sched_customize.cpp:190: error: ‘struct COPROC_CUDA’ has no member named ‘flops_estimate’

My videocard is NVIDIA, not ATI. And it does not support CUDA. So these messages seem strange a bit. Maybe something is wrong?

I can see you're trying to build the client in Linux, but do you mind telling which Linux, which client? Have you downloaded the source code from trunk or a branch?

I just had a quick chat with a developer, he says if you are just trying to build the client to pass the --disable-server parameter to the build script. It now is failing on the server component.

---

My configure line is:

./configure --disable-server --enable-client --enable-unicode 

---

Kathryn :o)

My videocard is NVIDIA, not ATI. And it does not support CUDA. So these messages seem strange a bit. Maybe something is wrong?

Your video cards are detected when the client runs, not when you compile it. That would be silly. Support for both ATI and NVIDIA is always compiled.

Imagine getting a new card and having to recompile the client from source...

---