On 5/8/25 10:52 AM, Eric Sandeen wrote:
>>> +
>>> + if (f2fs_test_compress_extension(F2FS_CTX_INFO(ctx).noextensions,
>>> + F2FS_CTX_INFO(ctx).nocompress_ext_cnt,
>>> + F2FS_CTX_INFO(ctx).extensions,
>>> + F2FS_CTX_INFO(ctx).compress_ext_cnt)) {
>>> + f2fs_err(sbi, "invalid compress or nocompress extension");
>> Can you please describe what is detailed confliction in the log? e.g. new
>> noext conflicts w/ new ext...
> Hmm, let me think about this. I had not noticed it was calling
> f2fs_test_compress_extension 3 times, I wonder if there is a better option.
> I need to understand this approach better. Maybe Hongbo has thoughts.
(taking linux-fsdevel off cc: to reduce noise)
Looking at this some more, I don't think any information is lost with this
change. The messages are exactly the same as they were before, in the error
cases.
(I don't love the "check 3 times" logic, but I guess we have to check internal
ctx consistency, as well as ctx vs. sbi, and sbi vs. ctx).
I think that if you would like to see clearer error messages, that's outside
the scope of the mount API conversion.
(If you have an example of a kernel message difference under this mount API
conversion vs current upstream, I'd be happy to look in more detail.)
Thanks,
-Eric
>>> + return -EINVAL;
>>> + }
>>> + if (f2fs_test_compress_extension(F2FS_CTX_INFO(ctx).noextensions,
>>> + F2FS_CTX_INFO(ctx).nocompress_ext_cnt,
>>> + F2FS_OPTION(sbi).extensions,
>>> + F2FS_OPTION(sbi).compress_ext_cnt)) {
>>> + f2fs_err(sbi, "invalid compress or nocompress extension");
>> Ditto,
>>
>>> + return -EINVAL;
>>> + }
>>> + if (f2fs_test_compress_extension(F2FS_OPTION(sbi).noextensions,
>>> + F2FS_OPTION(sbi).nocompress_ext_cnt,
>>> + F2FS_CTX_INFO(ctx).extensions,
>>> + F2FS_CTX_INFO(ctx).compress_ext_cnt)) {
>>> + f2fs_err(sbi, "invalid compress or nocompress extension");
>> Ditto,
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
