Bug 241269
Summary: | local_save_flags behaves differently from bare-metal kernel | ||
---|---|---|---|
Product: | Red Hat Enterprise Linux 5 | Reporter: | Kimball Murray <kmurray> |
Component: | kernel-xen | Assignee: | Kimball Murray <kmurray> |
Status: | CLOSED CANTFIX | QA Contact: | Martin Jenner <mjenner> |
Severity: | medium | Docs Contact: | |
Priority: | medium | ||
Version: | 5.0 | CC: | chas.horvath, clalance, riel, smcgrath, xen-maint |
Target Milestone: | --- | ||
Target Release: | --- | ||
Hardware: | x86_64 | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2008-01-29 19:31:45 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: |
Description
Kimball Murray
2007-05-24 17:57:52 UTC
Nice find. I agree that this inconsistency probably should be fixed. However, this is not a place where we can deviate from upstream easily. The change you suggest really needs to be merged in the upstream Xen tree before we can take it in RHEL. It might be better to have an abstraction like "is_interrupt_enabled()"; that way the number returned by "local_save_flags()" would be opaque to the caller. Chris Lalancette Eduardo points out that we already have "irqs_disabled()", which is modified appropriately for Xen. Would that work for you, Kimball? Chris Lalancette Cool. Yes it does, and I'll use that. I can do this because I've stumbled into the problem head-first and can now change my code. But what about all the other 3rd party modules that may stumble into this? How do we give them a warning to avoid the X86_EFLAGS_IF bit? Hm, we could do something where we #define X86_EFLAGS_IF bit to a piece of invalid code in the Xen case, so that compilation would fail. That might be a bit heavy handed, though. I think we would need to consult upstream about something like that. Chris Lalancette change QA contact Kimball is no longer the onsite engineer, please reopen (Simon) if you'd like to pick this up. Neither of these bug btw have been prioritized correctly. |