> -----Original Message----- > > -----Original Message----- > > Thanks for the review, and yes it's definitely OK to change the patch > > in this way. I just took a brief look at the code, and modified it in > > the way that actually change nothing. And after a second look, indeed > > they are never used as variable, only used as parameters of sizeof(). > > > > So actually can we just get rid of them, and use sizeof(struct > > parallel_info) and sizeof(struct splitting_info) instead? It may be > > even simpler. > > Yes, let's do it. I thought to take the original intention and make the > patch simpler, but thinking it again, each of them is used only two times, > and as you say, it would be simpler. Applied. https://sourceforge.net/p/makedumpfile/code/ci/399f2c9a3acd5bd913e50a4dde52dee6527b297e/ I replaced "declarations" with "definitions", and added a little information to the commit message. (ah, this is the change.) https://gcc.gnu.org/gcc-10/changes.html | GCC now defaults to -fno-common. As a result, global variable accesses are | more efficient on various targets. In C, global variables with multiple | tentative definitions now result in linker errors. Thanks, Kazu _______________________________________________ kexec mailing list kexec@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/kexec
