Bug 1368300
Summary: | Sata disk can not be attached to ahci controller device in q35 guest by libvirt automaticly | ||
---|---|---|---|
Product: | Red Hat Enterprise Linux 7 | Reporter: | Jingjing Shao <jishao> |
Component: | libvirt | Assignee: | Libvirt Maintainers <libvirt-maint> |
Status: | CLOSED NOTABUG | QA Contact: | Virtualization Bugs <virt-bugs> |
Severity: | unspecified | Docs Contact: | |
Priority: | unspecified | ||
Version: | 7.3 | CC: | dyuan, laine, lmen, rbalakri |
Target Milestone: | rc | ||
Target Release: | --- | ||
Hardware: | Unspecified | ||
OS: | Unspecified | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | If docs needed, set a value | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2016-08-19 14:17:36 UTC | Type: | Bug |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: |
Description
Jingjing Shao
2016-08-19 03:13:07 UTC
This is an invalid test. The Q35 machinetype in qemu has an integrated ACHI controller in slot 0x1f function 2 of the root complex (bus 0). This controller is there no matter what you do - no commandline argument is necessary in order for it to be there, and it is not possible to remove it. Additionally, the name of this controller (qemu's "id"), which is defined by qemu and is also not changeable, is inexplicably "ide" (the same as used for the integrated IDE controller on i440fx machinetypes). Since the commandline syntax for adding a disk device is that the controller id and unit number within the controller are given together as "bus=$id.$unit", the string "bus=ide.1" thus makes sense. So there is nothing wrong here. The odd naming of the integrated primary ahci controller as "ide" just makes it a bit confusing. (BTW, the Q35 machinetype doesn't have any integrated IDE controller, and libvirt doesn't support adding additional IDE controllers, so it's not even possible to have a disk connected to an IDE controller on a Q35 virtual machine.) (In reply to Laine Stump from comment #2) > This is an invalid test. > > The Q35 machinetype in qemu has an integrated ACHI controller in slot 0x1f > function 2 of the root complex (bus 0). This controller is there no matter > what you do - no commandline argument is necessary in order for it to be > there, and it is not possible to remove it. > > Additionally, the name of this controller (qemu's "id"), which is defined by > qemu and is also not changeable, is inexplicably "ide" (the same as used for > the integrated IDE controller on i440fx machinetypes). > > Since the commandline syntax for adding a disk device is that the controller > id and unit number within the controller are given together as > "bus=$id.$unit", the string "bus=ide.1" thus makes sense. > > So there is nothing wrong here. The odd naming of the integrated primary > ahci controller as "ide" just makes it a bit confusing. > > (BTW, the Q35 machinetype doesn't have any integrated IDE controller, and > libvirt doesn't support adding additional IDE controllers, so it's not even > possible to have a disk connected to an IDE controller on a Q35 virtual > machine.) Hi laine, I want to double check that the controller device "ide.1" attached automaticly is the integrated ACHI controller,just with a confused name, is it right? If so, should we update this name? I think it is really confused. Yes, "ide" is the name qemu uses internally for the integrated sata controller ("1" is the unit number on that controller). We *can't* change this name; it's not up to us, it's qemu's name and they hardcoded it. It's been there for several years with that name, and changing the name would break untold other software (libvirt included). It's silly, but that's the way it is and we have to live with it. At least libvirt shows the controller for what it is (a sata controller). (In reply to Laine Stump from comment #4) > Yes, "ide" is the name qemu uses internally for the integrated sata > controller ("1" is the unit number on that controller). > > We *can't* change this name; it's not up to us, it's qemu's name and they > hardcoded it. It's been there for several years with that name, and changing > the name would break untold other software (libvirt included). > > It's silly, but that's the way it is and we have to live with it. At least > libvirt shows the controller for what it is (a sata controller). OK,thanks for your detailed explanation |