Date: Tue, 12 Jul 2016 11:45:28 +0300 From: Andrey Chernov <ache@freebsd.org> To: Andrew Turner <andrew@fubar.geek.nz> Cc: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: Re: svn commit: r302601 - in head/sys: arm/include arm64/include Message-ID: <03ee3549-563d-6ee1-dab7-4818e8376dd1@freebsd.org> In-Reply-To: <20160712093509.69540801@zapp> References: <201607120037.u6C0bmNF054452@repo.freebsd.org> <20160712093509.69540801@zapp>
next in thread | previous in thread | raw e-mail | index | archive | help
On 12.07.2016 11:35, Andrew Turner wrote: > On Tue, 12 Jul 2016 00:37:48 +0000 (UTC) > "Andrey A. Chernov" <ache@FreeBSD.org> wrote: > >> Author: ache >> Date: Tue Jul 12 00:37:48 2016 >> New Revision: 302601 >> URL: https://svnweb.freebsd.org/changeset/base/302601 >> >> Log: >> I don't know why unsigned int is choosed for wchar_t here, but >> WCHAR_MAX should be <= WINT_MAX. It is bigger, __UINT_MAX > INT32_MAX > > Because the ABI either requires us to use an unsigned int [1], or the > preferred type is unsigned int [2]. In the latter case the other choice > is unsigned short, it would seem this is for Windows. Thanx for explanation. Perhaps we need to use 32bit unsigned int for other architectures too (instead of 32bit signed int), because no L'<char>' literals produce negative value and locale enumerates positive values only. BTW, this commit is already backed out. > > Andrew > > [1] > http://infocenter.arm.com/help/topic/com.arm.doc.ihi0055c/IHI0055C_beta_aapcs64.pdf > [2] > http://infocenter.arm.com/help/topic/com.arm.doc.ihi0042f/IHI0042F_aapcs.pdf >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?03ee3549-563d-6ee1-dab7-4818e8376dd1>