Bug 76231
| Summary: | loadkeys gives error on dead key codes | ||||||
|---|---|---|---|---|---|---|---|
| Product: | [Retired] Red Hat Linux | Reporter: | Stian Sletner <stian> | ||||
| Component: | kbd | Assignee: | Eido Inoue <havill> | ||||
| Status: | CLOSED DEFERRED | QA Contact: | |||||
| Severity: | medium | Docs Contact: | |||||
| Priority: | medium | ||||||
| Version: | 8.0 | ||||||
| Target Milestone: | --- | ||||||
| Target Release: | --- | ||||||
| Hardware: | All | ||||||
| OS: | Linux | ||||||
| Whiteboard: | |||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||
| Doc Text: | Story Points: | --- | |||||
| Clone Of: | Environment: | ||||||
| Last Closed: | 2003-07-14 16:33:18 UTC | Type: | --- | ||||
| Regression: | --- | Mount Type: | --- | ||||
| Documentation: | --- | CRM: | |||||
| Verified Versions: | Category: | --- | |||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||
| Embargoed: | |||||||
| Attachments: |
|
||||||
This went away when compiling a custom kernel. Can RH kernel patches have broken something? Anyway, bump. I see there's a new component owner. Maybe this one actually reads bugzilla. :-) Question about your custom config: are you running in Unicode mode? (Unicode no locale, kbd_mode -u and using unicode_start?) If not, can you try with your kernel and tell me the results? No, I'm not using Unicode mode. I'm not exactly sure what you're asking now, Unicode mode on the console in RH8 is pretty broken with or without my custom map (which is clearly an ISO 8859-1 map). I do not use a "no" locale at all, btw, and I remove the utf8 part in /etc/sysconfig/i18n to get out of Unicode mode. I did do a little test now though, and Unicode mode does indeed seem to break all the dead keys (and not only the ones I've added, either). By the way, how would I go about submitting my map to be included in RH? Ideally, it would replace the map called "no", which is useless. Please submit it (preferrably gzipped) as a patch to this bug report. It probably won't make it into the next release, but will be sent via RHN as an errata and the next release. Please make sure it's right-- I don't have a no keyboard (have to rely on scan codes) and nobody speaks no either (we used to have a dev, but not anymore. :( ) RH8 and beyond has standardized on Unicode for all single-byte locales... CJK are on the roadmap. While we will be looking at backwards compatiblity, 8+ will ship with Unicode enabled on the console by default. Created attachment 90787 [details]
Norwegian console keymap
Hello again (I've been away for a while). Attachment created. You should
verify that this loads without error on whatever kernel version it will be
bundled with, because as mentioned the stock 8.0 kernel chokes on it for some
reason. About it being bugfree and perfect, well, all I can say is that I've
been using this myself for a couple of years and have ironed out anything I've
found. It should be fine.
So, on closer thought it seems the kmap description language is far from perfect, as it makes me hardcode ISO 8859-1 things at certain spots. I guess I could whip up an UTF version of the same map. Would that be desired? non-ASCII input into the console won't be supported on UTF-8 based systems (RHL 8+) until the kernel supports it. See the release notes in the next beta for more info. Of course, input through X will always be supported. |
From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.1) Gecko/20020830 Description of problem: I have a custom keymap for Norwegian (that I hope to submit to be the default eventually, as it matches the keyboard better than the old, crufty `no' and `no-latin1' maps that are bundled). It uses the feature of making dead keys using 0x0dxx codes in the keymaps. Lines like these: keycode 27 = 0x0da8 dead_circumflex dead_tilde keycode 13 = backslash dead_grave 0x0db4 notsign keycode 51 = comma semicolon 0x0db8 guillemotleft I've used this successfully for several versions (at least all of 7.x), but in 8.0 it craps: KDSKBENT: Invalid argument failed to bind key 27 to value 3496 KDSKBENT: Invalid argument failed to bind key 13 to value 3508 KDSKBENT: Invalid argument failed to bind key 51 to value 3512 I talked to Brouwer, the kbd guy, and he said it worked for him (and did not mention having done any changes that would make it not work), so I assume this is something inflicted by RH. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1. Try the keymap lines in the Description, or download and loadkeys http://home.powertech.no/sletner/no-standard.kmap.gz 2. Watch. 3. Weep. Actual Results: KDSKBENT: Invalid argument failed to bind key 27 to value 3496 KDSKBENT: Invalid argument failed to bind key 13 to value 3508 KDSKBENT: Invalid argument failed to bind key 51 to value 3512 Additional info: