On 27/11/2019 17:48, Luc Van Oostenryck wrote:
> On Wed, Nov 27, 2019 at 05:15:20PM +0000, Ramsay Jones wrote:
>>
>>
>> On 27/11/2019 02:23, Luc Van Oostenryck wrote:
>>> These are small tweaks going on top of all the previous ones.
>>
>> Hi Luc,
>>
>> I only had minor comments on this (and the previous) series, but I
>> haven't had a chance to apply and test them yet.
>>
>> I was going to just test 'luc/int128_ctype', but looking at the output
>> of 'git log --oneline --graph master..luc/int128_ctype' I can see that
>> several other branches are merged into the result. So, I will apply
>> the patches from the mailing list directly to test them (hmm, _hopefully_
>> tonight).
>
> Hi Ramsay,
>
> Sorry, I should have added in the cover letter that the series
> with all related patches is available at:
> git://github.com/lucvoo/sparse-dev.git next
Hmm, but that _also_ contains several other branches not contained
in 'master', thus:
$ git --no-pager log --oneline --graph master..luc/next
* 4a8aa8d1 (luc/tip, luc/next) cgcc: add support for riscv64
* df14d253 arch: fix wchar_t & wint_t for openbsd
* 662ed740 arch: add missing predefines for PPC
* 1ff97d3a arch: add missing predfines: __amd64 & __amd64__
* d8754f2c cgcc: filter-out -msize-long & -msize-llp64
* 7efa090f (luc/int128_ctype) spec: replace lllong_ctype by int128_ctype
* 28f31494 Merge branch 'arch-cleanup' into master
|\
| * b0b47346 arch: add note for 128-bit long double on mips64
| * 66101fc5 arch: sparc32 on SunOS/Solaris uses 128-bit long doubles
| * 9c091c77 arch: fix wchar_t & wint_t on SunOS/Solaris
| * c79584fa arch: use a variable for the OS
| * eb87ce1b arch: add predefines for INT128 only on supported archs
| * a39cfb99 arch: (almost) all platforms simply use int for int32
| * 12abe926 arch: char32_t should be the same as uint32_t, not uint
| * 832d7a76 arch: remove impossible cases with 64-bit arch not being lp64
| * 6096ede6 arch: do not needlessly set bitness on 32-bit archs
* | dbcc589f Merge branch 'arm-hf' into master
|\ \
| * | fd59a5e8 cgcc: use -mfloat-abi=hard for armhf
| * | 7729ee4a fp-abi: teach sparse about -m{hard,soft}-float
| * | 532f6621 fp-abi: teach sparse about -mfloat-abi on ARM
| * | 16df5da6 fp-abi: add tests for ARM's -mfloat-abi=... & -msoft-float
| * | ec743e7f option: simpler handling of key-value for suboptions
| * | 95620eed option: use handle_switches() for -m flags
| * | 13391893 option: allow handle_switches() to set non-boolean values
| * | ed7b9191 option: move all option parsing helpers before their potential uses
| /
* | 50438f45 Merge branch 'arch-cygwin' into master
|\ \
| * | 16a575e1 (luc/arch-cygwin) cgcc: let cygwin use -fshort-wchar
| * | 0789c9a3 arch: teach sparse about -fshort-wchar
| * | 47628528 arch: keep cygwin specifics with i386/x86-64 specifics
| |/
| * c67fc56d arch: keep BSD & Darwin specifics with i386/x86-64 specifics
| * ecf39861 arch: simplify i386/x86-64 specifics
| * ebe08622 arch: fix the signedness of plain chars
* 467d8d02 Merge branch 'parse-spec' into master
* 10b89d60 spec: get rid of all specifier MOD_XXX
* 3d515131 spec: make ctype_declare[] more readable
* 570e295b spec: s/size/rank/
* c2d67bb0 spec: KW_LONG is not needed
* be1c60ca spec: KW_SHORT is not needed
* c04f113b spec: improve parsing of __int128
* 4c3567c2 spec: add '.class = CInt'
* 30ca8e0b spec: types[] is indexed by the specifier class
* 91c4ac10 spec: process chars like other integer types
$
Yes, I think most of the 'extra' commits are still in my email
client waiting to be looked at (sorry for being slow). :-D
ATB,
Ramsay Jones
