-
Notifications
You must be signed in to change notification settings - Fork 1.8k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Version 1.2.0 #1202
Version 1.2.0 #1202
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is good, just minor define/date touchups.
ABI compatibility is 100% so there are no issues there.
cc @yossigo Do you mind weighing in on whether or not we should bump the Let me know if I'm missing something, but I think we can maintain the SONAME at v1.1.0 so downstream packages that currently link with it don't need to be rebuilt. @chayim Once we iron that detail out we can move forward with a release. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@michael-grunder The right thing to do is bump the minor version to indicate we've extended the API/ABI in a backward-compatible manner.
In the case of Linux, this shouldn't require recompilation because shared object dependencies are expressed using the major version alone. IIRC, some other platforms do link against a specific minor version but have a SemVer-aware dynamic linker that will pick newer minor versions.
@michael-grunder The complaint was about changing SONAME for no reason since this was an internal fix with no ABI impact. Now, we are extending the ABI, so bumping the minor versions seems justified and required. BTW, while technically it was not necessary to change the version in 1.0.1, my understanding is that it should not have a negative impact on any system that assumes versioning adheres to SemVer. But I guess this is implementation specific and, on some platforms, may come with a price. Maybe @jfeep or @rgacogne can chime in and share their perspective with us. |
I'm not an expert on the subject at all, but my understanding is that the
Now in theory changing the minor version could be done without triggering a rebuild of binaries linked against $(INSTALL) $(DYLIBNAME) $(INSTALL_LIBRARY_PATH)/$(DYLIB_MINOR_NAME)
cd $(INSTALL_LIBRARY_PATH) && ln -sf $(DYLIB_MINOR_NAME) $(DYLIBNAME) I have not looked at what other distributions are doing, but on Arch it means we currently have $ objdump -p /usr/bin/unbound | grep -F NEEDED | grep -F libhiredis
NEEDED libhiredis.so.1.1.0 Therefore I would personally prefer the |
with minor too Co-authored-by: Michael Grunder <[email protected]>
@rgacogne Happy to revert the SONAME side. To be fair, I thought that was part and parcel of these minor bumps. Reverted SONAME here back to 1.1.1-dev to match. |
Thanks @rgacogne that's very useful info. I vote we maintain the |
This change addresses the issue discussed in redis#1202 and should make it possible in the future to update minor versions without requiring re-linking binaries.
Co-authored-by: Michael Grunder <[email protected]>
Co-authored-by: Michael Grunder <[email protected]>
This change addresses the issue discussed in #1202 and should make it possible in the future to update minor versions without requiring re-linking binaries.
@michael-grunder I see it just got merged... Assuming we're good with this, and there are no changes as discussed...
WDYT? |
@michael-grunder last bit of this, to ensure no scope creep. Confirming merge of #1205 in here (please thumbs up/down) and then that gets merged, and this plan followed. Thumbs down if you disagree with the merge proposal, and then we release as is. Last step @yossigo before release. |
No description provided.