ioemu: fix the devices loop. Signed-off-by: Jean Guyader <jean.guyader@eu.citrix.com> -- Jean Guyader _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Oops wrong patch, sorry about that .... ioemu: fix the devices loop. Signed-off-by: Jean Guyader <jean.guyader@eu.citrix.com> -- Jean Guyader _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Yuji Shimada
2008-Jul-17 07:05 UTC
[Xen-devel] [PATCH] ioemu: alternative patch fixing pt_chk_bar_overlap
I am submitting alternative patch fixing pt_chk_bar_overlap. The patch is for xen 3.3 unstable. Original issue is casting (PCIDevice **) to (PCIDevice *). The patch fix it. The difference between Jean''s patch and my patch is following. - check whether new resource overlaps with the resources allocated to other devices. - skip own devices'' resources. - check not only 1st function but also 2nd - 8th function. - loop from 0 to check all devices regardless of devfn_min. Thanks. Signed-off-by: Yuji Shimada <shimada-yxb@necst.nec.co.jp> -- Yuji Shimada _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Ian Jackson
2008-Jul-18 14:24 UTC
Re: [Xen-devel] [PATCH] ioemu: alternative patch fixing pt_chk_bar_overlap
Yuji Shimada writes ("[Xen-devel] [PATCH] ioemu: alternative patch fixing pt_chk_bar_overlap"):> I am submitting alternative patch fixing pt_chk_bar_overlap. > The patch is for xen 3.3 unstable.Is there any comparable bug in the merged qemu (aka ioemu-remote) ? I can''t seem to find any code which corresponds to these changes. Ian. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Jean Guyader
2008-Jul-18 14:27 UTC
Re: [Xen-devel] [PATCH] ioemu: alternative patch fixing pt_chk_bar_overlap
Ian Jackson wrote:> Yuji Shimada writes ("[Xen-devel] [PATCH] ioemu: alternative patch fixing pt_chk_bar_overlap"): >> I am submitting alternative patch fixing pt_chk_bar_overlap. >> The patch is for xen 3.3 unstable. > > Is there any comparable bug in the merged qemu (aka ioemu-remote) ? > I can''t seem to find any code which corresponds to these changes. >This code was parted of the FLR patch, I don''t think this patch has been merged with the ioemu-remote. -- Jean Guyader _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Ian Jackson
2008-Jul-18 14:54 UTC
Re: [Xen-devel] [PATCH] ioemu: alternative patch fixing pt_chk_bar_overlap
Jean Guyader writes ("Re: [Xen-devel] [PATCH] ioemu: alternative patch fixing pt_chk_bar_overlap"):> This code was parted of the FLR patch, I don''t think this patch has been > merged with the ioemu-remote.Ah, yes. I did try to apply it but it didn''t quite apply cleanly. If it''s important then we need to clean this up. Since people have started submitting patches to qemu (aka ioemu-remote) separately from xen-unstable, and similar things are getting committed to both, it''s not really practical now for me to do a bulk merge from xen-unstable. Ian. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Yuji Shimada
2008-Jul-22 08:14 UTC
Re: [Xen-devel] [PATCH] ioemu: alternative patch fixing pt_chk_bar_overlap
> Ah, yes. I did try to apply it but it didn''t quite apply cleanly. > > If it''s important then we need to clean this up.The reason why the patch can not be applied to ioemu-remote is "pci_config_passthrough_with_msi_msix.patch" I submitted has not been applied to ioemu-remote yet. I''ve created new patch for ioemu-remote. If you apply it to ioemu-remote, the implementation of passthrough device will become almost the same for both ioemu trees. The patch contains following patches which have been applied to Xen Unstable Changeset 17959-18076. pci_config_passthrough_with_msi_msix.patch [Xen-devel] [PATCH] Support more Capability Structures (including MSI/MSI-X) and Device Specific Registers for pt device. Yuji Shimada <shimada-yxb@necst.nec.co.jp> Fri, 04 Jul 2008 15:26:52 +0900 fix_pt_iomap.patch Re: [Xen-devel] [PATCH] Support more Capability Structures (including MSI/MSI-X) and Device Specific Registers for pt device. NISHIGUCHI Naoki <nisiguti@jp.fujitsu.com> Tue, 08 Jul 2008 14:14:55 +0900 fix_pt_bar_mapping.patch [Xen-devel] [PATCH] ioemu: pass-through: pt_bar_mapping fix NISHIGUCHI Naoki <nisiguti@jp.fujitsu.com> Wed, 09 Jul 2008 15:46:15 +0900 fix_hot-remove.patch [Xen-devel] [PATCH] Fix issues when passthrough device is hot-removed from HVM domain Yuji Shimada <shimada-yxb@necst.nec.co.jp> Wed, 16 Jul 2008 09:20:11 +0900 chk_bar_overlap_bugfix.patch [Xen-devel] [PATCH] ioemu: alternative patch fixing pt_chk_bar_overlap Yuji Shimada <shimada-yxb@necst.nec.co.jp> Thu, 17 Jul 2008 16:05:11 +0900 Thanks Signed-off-by: Yuji Shimada <shimada-yxb@necst.nec.co.jp> -- Yuji Shimada On Fri, 18 Jul 2008 15:54:14 +0100 Ian Jackson <Ian.Jackson@eu.citrix.com> wrote:> Jean Guyader writes ("Re: [Xen-devel] [PATCH] ioemu: alternative patch fixing pt_chk_bar_overlap"): > > This code was parted of the FLR patch, I don''t think this patch has been > > merged with the ioemu-remote. > > Ah, yes. I did try to apply it but it didn''t quite apply cleanly. > > If it''s important then we need to clean this up. > > Since people have started submitting patches to qemu (aka > ioemu-remote) separately from xen-unstable, and similar things are > getting committed to both, it''s not really practical now for me to do > a bulk merge from xen-unstable. > > Ian. > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel_______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Ian Jackson
2008-Jul-22 10:55 UTC
Re: [Xen-devel] [PATCH] ioemu: alternative patch fixing pt_chk_bar_overlap
Yuji Shimada writes ("Re: [Xen-devel] [PATCH] ioemu: alternative patch fixing pt_chk_bar_overlap"):> I''ve created new patch for ioemu-remote. If you apply it to ioemu-remote, > the implementation of passthrough device will become almost the same for > both ioemu trees.Thanks for that.> The patch contains following patches which have been applied to Xen > Unstable Changeset 17959-18076.This patch introduces a couple of new warnings: /u/iwj/work/qemu-iwj.git/hw/pass-through.c: In function ''pt_iomem_map'': /u/iwj/work/qemu-iwj.git/hw/pass-through.c:821: warning: format ''%lx'' expects type ''long unsigned int'', but argument 5 has type ''uint64_t'' and In file included from /u/iwj/work/qemu-iwj.git/hw/pt-msi.h:5, from /u/iwj/work/qemu-iwj.git/hw/pt-msi.c:22: /u/iwj/work/qemu-iwj.git/hw/pass-through.h:57:1: warning: "PCI_MSI_FLAGS_MASK_BIT" redefined In file included from /u/iwj/work/qemu-iwj.git/hw/pass-through.h:23, from /u/iwj/work/qemu-iwj.git/hw/pt-msi.h:5, from /u/iwj/work/qemu-iwj.git/hw/pt-msi.c:22: /usr/include/pci/header.h:274:1: warning: this is the location of the previous definition The former was easy to fix. The latter is not critical so I have committed your patch anyway, but it would be nice to get rid of it. Who do you think owns this part of the namespace ? Should we sprinkle some #ifdefs in pass-through.h ? On my system (Debian etch) /usr/include/pci/header.h is part of pciutils-dev 1:2.2.4~pre4-1. The definition is identical in effect. Ian. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Yuji Shimada
2008-Jul-24 03:00 UTC
Re: [Xen-devel] [PATCH] ioemu: alternative patch fixing pt_chk_bar_overlap
On Tue, 22 Jul 2008 11:55:21 +0100 Ian Jackson <Ian.Jackson@eu.citrix.com> wrote:> Who do you think owns this part of the namespace ? Should we sprinkle > some #ifdefs in pass-through.h ? On my system (Debian etch) > /usr/include/pci/header.h is part of pciutils-dev 1:2.2.4~pre4-1. The > definition is identical in effect.I think libpci(pciutils-devel) owns this part of the namespace. But on my system (Cent OS 5.1 with pciutils-devel-2.2.3-4), some constants are not defined. I left the definition of the constants in pass-through.h and avoided conflicts using #ifndef/#endif. I attached the patch. Thanks Signed-off-by: Yuji Shimada <shimada-yxb@necst.nec.co.jp> -- Yuji Shimada _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel