08.12.2015 14:53, Dimitry Sibiryakov пишет: > 08.12.2015 13:39, Vlad Khorsun wrote: >> I don't know how to guarantee atomic execution of sequence of two >> instructions on Intel HW. > > Don't be confused by pseudocode in function description.
I'm not confused >> You have a proof of course... > > Test program from your link. > MSVC assembler: > >> ; Line 7 >> mov DWORD PTR _cur$[esp+8], 15 ; 0000000fH >> push esi >> ; Line 9 - old = _InterlockedOr(&cur, 0xF0); >> mov ecx, 240 ; 000000f0H >> lea edx, DWORD PTR _cur$[esp+12] >> mov eax, DWORD PTR [edx] >> $LN3@main: >> mov esi, eax >> or esi, ecx >> lock cmpxchg DWORD PTR [edx], esi >> jne SHORT $LN3@main >> mov DWORD PTR _old$[esp+12], eax >> ; Line 10 - _InterlockedOr(&cur, 0xF0); >> mov eax, edx >> lock or DWORD PTR [eax], ecx > > GCC assembler: > >> movl $15, 44(%rsp) >> ; old = __atomic_fetch_or(&cur, 0xF0, __ATOMIC_SEQ_CST); >> movl 44(%rsp), %eax >> .L2: >> movl %eax, %edx >> movl %eax, %ecx >> orb $-16, %dl >> lock cmpxchgl %edx, 44(%rsp) >> jne .L2 >> movl %ecx, 40(%rsp) >> ; __atomic_fetch_or(&cur, 0xF0, __ATOMIC_SEQ_CST); >> lock orl $240, 44(%rsp) > > Do you see the principal difference with Firebird asm code i show ? Do you see the loop with compare_exchange ? Do you see the "lock or" operation ? Do you still have questions ? Regards, Vlad ------------------------------------------------------------------------------ Go from Idea to Many App Stores Faster with Intel(R) XDK Give your users amazing mobile app experiences with Intel(R) XDK. Use one codebase in this all-in-one HTML5 development environment. Design, debug & build mobile apps & 2D/3D high-impact games for multiple OSs. http://pubads.g.doubleclick.net/gampad/clk?id=254741911&iu=/4140 Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
