[OE-core] [PATCH v2] ninja: upate to 1.9.0
Oleksandr Kravchuk
open.source at oleksandr-kravchuk.com
Wed Apr 24 14:31:11 UTC 2019
Sure. It's already in master though.
Should I send a new patch or just a diff with master?
On 23/04/2019 11:26, Burton, Ross wrote:
> On Sat, 23 Mar 2019 at 16:39, Oleksandr Kravchuk
> <open.source at oleksandr-kravchuk.com> wrote:
>> +++ b/meta/recipes-devtools/ninja/ninja/fix-musl.patch
>> @@ -0,0 +1,39 @@
>> +2ff54ad7478a90bd75c91e434236a Mon Sep 17 00:00:00 2001
>> +From: makepost <makepost at firemail.cc>
>> +Date: Mon, 24 Dec 2018 03:13:16 +0200
>> +Subject: [PATCH] Use st_mtim if st_mtime is macro, fix #1510
>> +
>> +In POSIX.1-2008, sys_stat has a st_mtim member and a st_mtime backward
>> +compatibility macro. Should help avoid hardcoding platform detection.
>> +---
>> + src/disk_interface.cc | 14 ++++----------
>> + 1 file changed, 4 insertions(+), 10 deletions(-)
>> +
>> +diff --git a/src/disk_interface.cc b/src/disk_interface.cc
>> +index d4c2fb087..dc297c449 100644
>> +--- src/disk_interface.cc
>> ++++ /src/disk_interface.cc
>> +@@ -202,19 +202,13 @@ TimeStamp RealDiskInterface::Stat(const string& path, string* err) const {
>> + // that it doesn't exist.
>> + if (st.st_mtime == 0)
>> + return 1;
>> +-#if defined(__APPLE__) && !defined(_POSIX_C_SOURCE)
>> ++#if defined(_AIX)
>> ++ return (int64_t)st.st_mtime * 1000000000LL + st.st_mtime_n;
>> ++#elif defined(__APPLE__)
>> + return ((int64_t)st.st_mtimespec.tv_sec * 1000000000LL +
>> + st.st_mtimespec.tv_nsec);
>> +-#elif (_POSIX_C_SOURCE >= 200809L || _XOPEN_SOURCE >= 700 || defined(_BSD_SOURCE) || defined(_SVID_SOURCE) || \
>> +- defined(__BIONIC__) || (defined (__SVR4) && defined (__sun)) || defined(__FreeBSD__))
>> +- // For glibc, see "Timestamp files" in the Notes of http://www.kernel.org/doc/man-pages/online/pages/man2/stat.2.html
>> +- // newlib, uClibc and musl follow the kernel (or Cygwin) headers and define the right macro values above.
>> +- // For bsd, see https://github.com/freebsd/freebsd/blob/master/sys/sys/stat.h and similar
>> +- // For bionic, C and POSIX API is always enabled.
>> +- // For solaris, see https://docs.oracle.com/cd/E88353_01/html/E37841/stat-2.html.
>> ++#elif defined(st_mtime) // A macro, so we're likely on modern POSIX.
>> + return (int64_t)st.st_mtim.tv_sec * 1000000000LL + st.st_mtim.tv_nsec;
>> +-#elif defined(_AIX)
>> +- return (int64_t)st.st_mtime * 1000000000LL + st.st_mtime_n;
>> + #else
>> + return (int64_t)st.st_mtime * 1000000000LL + st.st_mtimensec;
>> + #endif
> This patch has no Signed-off-by or Upstream-Status. Can you add those please?
>
> Ross
More information about the Openembedded-core
mailing list