https://gcc.gnu.org/bugzilla/show_bug.cgi?id=107801
--- Comment #5 from jdx at o2 dot pl --- (In reply to CVS Commits from comment #4) > The releases/gcc-12 branch has been updated by Jonathan Wakely > <r...@gcc.gnu.org>: > > https://gcc.gnu.org/g:691012693aaa831e4cd1ab5180195792e32baceb > > commit r12-8926-g691012693aaa831e4cd1ab5180195792e32baceb > Author: Jonathan Wakely <jwak...@redhat.com> > Date: Tue Nov 22 09:53:36 2022 +0000 > > libstdc++: Fix pool resource build errors for H8 [PR107801] > > The array of pool sizes was previously adjusted to work for msp430-elf > which has 16-bit int and either 16-bit size_t or 20-bit size_t. The > largest pool sizes were disabled unless size_t has more than 20 bits. > > The H8 family has 16-bit int but 32-bit size_t, which means that the > largest sizes are enabled, but 1<<15 produces a negative number that > then cannot be narrowed to size_t. Jonathan, a bit of explanation regarding H8 family – it has a few "subfamilies": 1. H8/300 and H8/300L (L from "low power" AFAIR) – these are pure 16-bit MCUs, it has 16-bit general purpose registers, sizeof(int) = sizeof(size_t) = 2; AFAIK they are not manufactured anymore; I have never used any of them. 2. H8/300H, H8S and H8SX – Renesas/Hitachi calls them 16-bit but I call them quasi 32-bit MCUs because they have 32-bit GPRs and instructions, although data bus width is 16 bit. Here things are more complicated – sizeof(size_t) = 4 but sizeof(int) may be 2 or 4 depending on -mint32 command line option. To get things even more complicated, these MCUs have also so called "normal mode" (-mn option) – it is compatibility mode with H8/300 and obviously in this mode sizeof(int) = sizeof(size_t) = 2; __NORMAL_MODE__ macro is defined in this mode.
