| Message ID | 20201112164844.18731-1-klaus@linux.vnet.ibm.com |
|---|---|
| State | Accepted |
| Headers | show |
| Series | [1/1] package/busybox: Fix hwclock for glibc 2.31+ | expand |
>>>>> "Klaus" == Klaus Heinrich Kiwi <klaus@linux.vnet.ibm.com> writes: > Pick the below patch from upstream, in order to fix > 'settimeofday: Invalid argument' introduced by using glibc v2.31+. > (busybox hasn't tagged a new version since). > See https://bugs.busybox.net/show_bug.cgi?id=12756 for more info. > --- > commit 1a5d6fcbb5e606ab4acdf22afa26361a25f1d43b > Author: Eddie James <eajames@linux.ibm.com> > hwclock: Fix settimeofday for glibc v2.31+ > --- > +++ b/package/busybox/0003-hwclock-Fix-settimeofday-for-glibc-v2.31.patch > @@ -0,0 +1,58 @@ > +From a4faa9af79c2a0f2b501c0fd507842e4966fc23b Mon Sep 17 00:00:00 2001 How did you end up with a different git hash here? Committed after fixing that, thanks.
On 11/12/2020 6:04 PM, Peter Korsgaard wrote: >>>>>> "Klaus" == Klaus Heinrich Kiwi <klaus@linux.vnet.ibm.com> writes: > > --- > > +++ b/package/busybox/0003-hwclock-Fix-settimeofday-for-glibc-v2.31.patch > > @@ -0,0 +1,58 @@ > > +From a4faa9af79c2a0f2b501c0fd507842e4966fc23b Mon Sep 17 00:00:00 2001 > > How did you end up with a different git hash here? I took busybox tag 1_32_0, applied the patches found on buildroot/package/busybox, then cherry-picked the above patch and git format-patch HEAD^ Guess I was being too cautious by applying the patches locally first? At any rate, I'm sending a second version where I format-patch'ed the patch itself (i.e., from busybox master), instead of applying it on top of a local 1_32_0 tree along with other buildroot build-time patches. > > Committed after fixing that, thanks. > Thanks, -Klaus
>>>>> "Klaus" == Klaus Heinrich Kiwi <klaus@linux.vnet.ibm.com> writes: > On 11/12/2020 6:04 PM, Peter Korsgaard wrote: >>>>>>> "Klaus" == Klaus Heinrich Kiwi <klaus@linux.vnet.ibm.com> writes: >> > --- >> > +++ b/package/busybox/0003-hwclock-Fix-settimeofday-for-glibc-v2.31.patch >> > @@ -0,0 +1,58 @@ >> > +From a4faa9af79c2a0f2b501c0fd507842e4966fc23b Mon Sep 17 00:00:00 2001 >> >> How did you end up with a different git hash here? > I took busybox tag 1_32_0, applied the patches found on buildroot/package/busybox, > then cherry-picked the above patch and git format-patch HEAD^ > Guess I was being too cautious by applying the patches locally first? That is also fine, but then use the -x option to git cherry-pick so the original git hash is tracked. > At any rate, I'm sending a second version where I format-patch'ed the patch itself > (i.e., from busybox master), instead of applying it on top of a local 1_32_0 tree > along with other buildroot build-time patches. That is not needed as I already applied this patch. Thanks!
On 11/13/2020 7:03 AM, Peter Korsgaard wrote: > > That is also fine, but then use the -x option to git cherry-pick so the > original git hash is tracked. Odd, I kinda always automatically cherry-pick with -xs, not sure how I missed that this time. > > > At any rate, I'm sending a second version where I format-patch'ed the patch itself > > (i.e., from busybox master), instead of applying it on top of a local 1_32_0 tree > > along with other buildroot build-time patches. > > That is not needed as I already applied this patch. > > Thanks! > Thanks! I'll send a similar patch to buildroot 2020.08.1 (hopefully this is still active?) and 2020.02.7 (in case it's really affected - need to double check) - but those will require a small change to apply on top of busybox 1_31_1.. -Klaus
>>>>> "Klaus" == Klaus Heinrich Kiwi <klaus@linux.vnet.ibm.com> writes: Hi, > Thanks! > I'll send a similar patch to buildroot 2020.08.1 (hopefully this is > still active?) and 2020.02.7 (in case it's really affected - need to > double check) - but those will require a small change to apply on top > of busybox 1_31_1.. Yes, 2020.02.x (until April 2021) and 2020.08.x (until January 2021) are still supported. 2020.02.x uses glibc 2.30 though, so it isn't needed there.
diff --git a/package/busybox/0003-hwclock-Fix-settimeofday-for-glibc-v2.31.patch b/package/busybox/0003-hwclock-Fix-settimeofday-for-glibc-v2.31.patch new file mode 100644 index 0000000000..98a0b01430 --- /dev/null +++ b/package/busybox/0003-hwclock-Fix-settimeofday-for-glibc-v2.31.patch @@ -0,0 +1,58 @@ +From a4faa9af79c2a0f2b501c0fd507842e4966fc23b Mon Sep 17 00:00:00 2001 +From: Eddie James <eajames@linux.ibm.com> +Date: Mon, 10 Aug 2020 09:59:02 -0500 +Subject: [PATCH] hwclock: Fix settimeofday for glibc v2.31+ + +The glibc implementation changed for settimeofday, resulting in "invalid +argument" error when attempting to set both timezone and time with a single +call. Fix this by calling settimeofday twice + +Signed-off-by: Eddie James <eajames@linux.ibm.com> +Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com> +--- + util-linux/hwclock.c | 14 +++++++++++--- + 1 file changed, 11 insertions(+), 3 deletions(-) + +diff --git a/util-linux/hwclock.c b/util-linux/hwclock.c +index dc97d8fb4..2479e7416 100644 +--- a/util-linux/hwclock.c ++++ b/util-linux/hwclock.c +@@ -122,16 +122,20 @@ static void to_sys_clock(const char **pp_rtcname, int utc) + struct timeval tv; + struct timezone tz; + +- tz.tz_minuteswest = timezone/60; ++ tz.tz_minuteswest = timezone / 60; + /* ^^^ used to also subtract 60*daylight, but it's wrong: + * daylight!=0 means "this timezone has some DST + * during the year", not "DST is in effect now". + */ + tz.tz_dsttime = 0; + ++ /* glibc v2.31+ returns an error if both args are non-NULL */ ++ if (settimeofday(NULL, &tz)) ++ bb_simple_perror_msg_and_die("settimeofday"); ++ + tv.tv_sec = read_rtc(pp_rtcname, NULL, utc); + tv.tv_usec = 0; +- if (settimeofday(&tv, &tz)) ++ if (settimeofday(&tv, NULL)) + bb_simple_perror_msg_and_die("settimeofday"); + } + +@@ -283,7 +287,11 @@ static void set_system_clock_timezone(int utc) + gettimeofday(&tv, NULL); + if (!utc) + tv.tv_sec += tz.tz_minuteswest * 60; +- if (settimeofday(&tv, &tz)) ++ ++ /* glibc v2.31+ returns an error if both args are non-NULL */ ++ if (settimeofday(NULL, &tz)) ++ bb_simple_perror_msg_and_die("settimeofday"); ++ if (settimeofday(&tv, NULL)) + bb_simple_perror_msg_and_die("settimeofday"); + } + +-- +2.17.1 +
Pick the below patch from upstream, in order to fix 'settimeofday: Invalid argument' introduced by using glibc v2.31+. (busybox hasn't tagged a new version since). See https://bugs.busybox.net/show_bug.cgi?id=12756 for more info. --- commit 1a5d6fcbb5e606ab4acdf22afa26361a25f1d43b Author: Eddie James <eajames@linux.ibm.com> hwclock: Fix settimeofday for glibc v2.31+ --- Signed-off-by: Klaus Heinrich Kiwi <klaus@linux.vnet.ibm.com> Signed-off-by: Klaus Heinrich Kiwi <klaus@linux.vnet.ibm.com> --- ...ock-Fix-settimeofday-for-glibc-v2.31.patch | 58 +++++++++++++++++++ 1 file changed, 58 insertions(+) create mode 100644 package/busybox/0003-hwclock-Fix-settimeofday-for-glibc-v2.31.patch