You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Before commit bc1239b it was possible to compress using legacy mode, and decompress using the normal reader. After the commit, not possible anymore. I wrote a test case based on some tests failing in other repos. I'm not exactly sure if this is supposed to be supported or possible, or even really used, but my coworkers wrote a test case for it here.
I applied the above patch and git bisected from v4.1.14 to v4.1.17. bc1239b is what fell out.
It seems like the decoding is not broken, but the encoding in legacy mode is given the commit that broke things. So technically speaking, our use case in u-root is still working, just the test is broken for the time being.
The text was updated successfully, but these errors were encountered:
Patch with test to reproduce: https://gist.github.com/hugelgupf/fc7ea7c735017acfbcad61ef36632d55
Before commit bc1239b it was possible to compress using legacy mode, and decompress using the normal reader. After the commit, not possible anymore. I wrote a test case based on some tests failing in other repos. I'm not exactly sure if this is supposed to be supported or possible, or even really used, but my coworkers wrote a test case for it here.
I applied the above patch and git bisected from v4.1.14 to v4.1.17. bc1239b is what fell out.
It seems like the decoding is not broken, but the encoding in legacy mode is given the commit that broke things. So technically speaking, our use case in u-root is still working, just the test is broken for the time being.
The text was updated successfully, but these errors were encountered: