scst_lib,scst_sysfs: Add aen_disabled setting

[bmeagherix]

Add a setting to scst_tgt that can prevent scst_gen_aen_or_ua from generating an AEN, even if the underlying transport is capable of transmitting them. It will instead generate a UA.

This could prove useful in different situations, including when the target port is also a forward_dst.

(The new sysfs interface was modeled on the code for forward_dst.)

[lnocturno]

Hi Brian,

Sorry for the long response, but I still need more time to review your patch.
I'm trying to get it done by the end of this week.

Thanks,
Gleb

[bmeagherix]

Sorry for the long response, but I still need more time to review your patch.
I'm trying to get it done by the end of this week.

Hi Gleb,

No problem, I understand and appreciate all your effort & consideration!

FWIW, as you may notice the new checkpatch requirement means that some code doesn't match the rest of the codebase, e.g.

WARNING:SYMBOLIC_PERMS: Symbolic permissions 'S_IRUGO | S_IWUSR' are not preferred. Consider using octal permissions '0644'.
#192: FILE: scst/src/scst_sysfs.c:2694:
+       __ATTR(aen_disabled, S_IRUGO | S_IWUSR, scst_tgt_aen_disabled_show,

so now this PR is using 0644. Please let me know if that should have been handled differently.

Best regards,
-Brian.

[lnocturno]

Hi Brian,

FWIW, as you may notice the new checkpatch requirement means that some code doesn't match the rest of the codebase

The idea behind the new checkpatch requirement is to check new patches for errors/warnings. The current codebase still contains a lot of checkpatch errors/warnings, but I think I will fix them step by step wherever possible.

Not every error/warning needs to be fixed, I have already added an ignore rule for some of them and will continue to do so, depending on the type. In your case, you did everything right:

https://lkml.org/lkml/2021/9/4/55
https://lore.kernel.org/lkml/CA+55aFw5v23T-zvDZp-MmD_EYxF8WbafwwB59934FV7g21uMGQ@mail.gmail.com/

[bmeagherix]

BTW, this PR deliberately does not include suppressing AENs from scst_report_luns_changed_sess ... less obviously useful to do so there.

[lnocturno]

Hi Brian,

Thank you for the patch!

Thanks,
Gleb