Description of problem: Whilest compiling mono-debugger (and anjuta2), the build process throws up an error - expected specifier-qualifier-list before 'u64' Version-Release number of selected component (if applicable): 3.0-5.2 How reproducible: Always Steps to Reproduce: 1. Download the source to mono-debugger (http://www.smmp.salford.ac.uk/packages/mono-debugger-0.12-2.src.rpm) 2. Install and try to compile 3. Actual results: The build throws up In file included from /usr/include/asm/user.h:4, from x86_64-arch.h:10, from x86-arch.h:45, from x86-linux-ptrace.h:4, from x86-ptrace.c:35: /usr/include/asm-x86_64/user.h:56: error: expected specifier-qualifier-list before 'u64' Expected results: mono-debugger should compile Additional info: The compile works fine on an x86 box, just not on an x86_64
I went to take a look at this but the link above is invalid; I grabbed http://www.smmp.salford.ac.uk/packages/mono-debugger-0.12-3.src.rpm instead but it fails to build due to a missing BuildRequire: glib2-devel, mono-devel. I added that and handed it to mock (development, x86_64). Everything compiles fine but the package fails to build completely due to errors in the %file section. So at least this is not a problem in rawhide. It does indeed fail as advertised in FC5.
BTW, the rawhide version of /usr/include/asm-x86_64/user.h has double-underscores in front of the problematic types in that structure definition: rawhide: struct user_i387_struct { unsigned short cwd; unsigned short swd; unsigned short twd; /* Note this is not the same as the 32bit/x87/FSAVE twd */ unsigned short fop; __u64 rip; __u64 rdp; __u32 mxcsr; __u32 mxcsr_mask; __u32 st_space[32]; /* 8*16 bytes for each FP-reg = 128 bytes */ __u32 xmm_space[64]; /* 16*16 bytes for each XMM-reg = 256 bytes */ __u32 padding[24]; }; FC5: struct user_i387_struct { unsigned short cwd; unsigned short swd; unsigned short twd; /* Note this is not the same as the 32bit/x87/FSAVE twd */ unsigned short fop; u64 rip; u64 rdp; u32 mxcsr; u32 mxcsr_mask; u32 st_space[32]; /* 8*16 bytes for each FP-reg = 128 bytes */ u32 xmm_space[64]; /* 16*16 bytes for each XMM-reg = 256 bytes */ u32 padding[24]; };
Reopening -- still exists in FC5.
Um, I've not seen it when compiling on my vanilla box - there are certainly warnings, but mono-debugger is compiling fine (which is where I found this problem initially)
I don't think I ever fixed glibc-kernheaders though.
Something has... Dammit. I did! Just checked my logs!!!
Can this bug now be closed as the headers have been pushed to both rawhide and FC5?
They still haven't been pushed to FC5. See comment #5. And comment #6.
Ah. My apologies. I thought that they were going to be pushed shortly before the RH pow-wow.
There's still an occasional problem with them trickling in, and I want to ship the proper 2.6.17 headers, preferably after Linus has taken the 'make headers_install' thing into 2.6.18. So another week or two, I think.
Hey David. There also seems to be missing: /usr/include/linux/compiler.h file.
linux/compiler.h is intentionally absent and not relevant to this bug.
According to the comments this should be fixed in current Fedora releases.