Re: [PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-06 Thread Chris Wright
* Joerg Roedel (j...@8bytes.org) wrote: On Sun, Apr 04, 2010 at 02:44:36AM -0700, Eric W. Biederman wrote: Joerg Roedel j...@8bytes.org writes: On Sun, Apr 04, 2010 at 09:24:30AM +0200, Bernhard Walle wrote: Am 03.04.10 19:49, schrieb Eric W. Biederman: Not a problem. We require a

Re: [PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-06 Thread Vivek Goyal
On Tue, Apr 06, 2010 at 04:39:56PM -0400, Vivek Goyal wrote: On Tue, Apr 06, 2010 at 07:51:06PM +0200, Joerg Roedel wrote: On Tue, Apr 06, 2010 at 10:42:57AM -0700, Chris Wright wrote: * Joerg Roedel (j...@8bytes.org) wrote: On Sun, Apr 04, 2010 at 02:44:36AM -0700, Eric W. Biederman

Re: [PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-06 Thread Yinghai Lu
On Tue, Apr 6, 2010 at 2:13 PM, Vivek Goyal vgo...@redhat.com wrote: On Tue, Apr 06, 2010 at 04:39:56PM -0400, Vivek Goyal wrote: On Tue, Apr 06, 2010 at 07:51:06PM +0200, Joerg Roedel wrote: On Tue, Apr 06, 2010 at 10:42:57AM -0700, Chris Wright wrote: * Joerg Roedel (j...@8bytes.org)

Re: [PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-04 Thread Bernhard Walle
Am 03.04.10 19:49, schrieb Eric W. Biederman: Not a problem. We require a lot of things of the kdump kernel, and it is immediately apparent in a basic sanity test. Also, in most cases (for example: distribution kernels), the kdump kernel nowadays is identical to the running kernel. So, if the

Re: [PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-04 Thread Joerg Roedel
On Sat, Apr 03, 2010 at 10:44:22AM -0700, Eric W. Biederman wrote: Joerg Roedel j...@8bytes.org writes: Hmm, I think for this we need to change the gart code too and disable the gart before its initialization runs to not re-introduce issues fixed in commit

Re: [PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-04 Thread Joerg Roedel
On Sun, Apr 04, 2010 at 09:24:30AM +0200, Bernhard Walle wrote: Am 03.04.10 19:49, schrieb Eric W. Biederman: Not a problem. We require a lot of things of the kdump kernel, and it is immediately apparent in a basic sanity test. Also, in most cases (for example: distribution kernels), the

Re: [PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-04 Thread Eric W. Biederman
Joerg Roedel j...@8bytes.org writes: On Sat, Apr 03, 2010 at 10:44:22AM -0700, Eric W. Biederman wrote: Joerg Roedel j...@8bytes.org writes: Hmm, I think for this we need to change the gart code too and disable the gart before its initialization runs to not re-introduce issues fixed in

Re: [PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-04 Thread Eric W. Biederman
I guess I should add the place where I have serious test results for with a gart is on 2.6.29. Where there isn't that iommu shutdown and I don't have problems with it. So my experience is that touching the gart in machine_crash_shutdown is unnecessary. Eric

Re: [PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-04 Thread Joerg Roedel
On Sun, Apr 04, 2010 at 02:44:36AM -0700, Eric W. Biederman wrote: Joerg Roedel j...@8bytes.org writes: On Sun, Apr 04, 2010 at 09:24:30AM +0200, Bernhard Walle wrote: Am 03.04.10 19:49, schrieb Eric W. Biederman: Not a problem. We require a lot of things of the kdump kernel, and it

Re: [PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-04 Thread David Woodhouse
On Sat, 2010-04-03 at 19:41 +0200, Joerg Roedel wrote: Another problem: This also breaks if the kdump kernel has no iommu-support. The kdump kernel has to support the hardware it's running on. Film at 11. -- David WoodhouseOpen Source Technology Centre

Re: [PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-03 Thread Joerg Roedel
On Fri, Apr 02, 2010 at 06:27:54PM -0700, Chris Wright wrote: This effectively reverts commit 61d047be99757fd9b0af900d7abce9a13a337488. Disabling the IOMMU can potetially allow DMA transactions to complete without being translated. Leave it enabled, and allow crash kernel to do the IOMMU

Re: [PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-03 Thread Joerg Roedel
On Fri, Apr 02, 2010 at 06:27:54PM -0700, Chris Wright wrote: This effectively reverts commit 61d047be99757fd9b0af900d7abce9a13a337488. Disabling the IOMMU can potetially allow DMA transactions to complete without being translated. Leave it enabled, and allow crash kernel to do the IOMMU

Re: [PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-03 Thread Eric W. Biederman
Joerg Roedel j...@8bytes.org writes: On Fri, Apr 02, 2010 at 06:27:54PM -0700, Chris Wright wrote: This effectively reverts commit 61d047be99757fd9b0af900d7abce9a13a337488. Disabling the IOMMU can potetially allow DMA transactions to complete without being translated. Leave it enabled, and

Re: [PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-03 Thread Eric W. Biederman
Joerg Roedel j...@8bytes.org writes: On Sat, Apr 03, 2010 at 10:49:09AM -0700, Eric W. Biederman wrote: Joerg Roedel j...@8bytes.org writes: Another problem: This also breaks if the kdump kernel has no iommu-support. Not a problem. We require a lot of things of the kdump kernel, and

[PATCH 3/4] Revert x86: disable IOMMUs on kernel crash

2010-04-02 Thread Chris Wright
This effectively reverts commit 61d047be99757fd9b0af900d7abce9a13a337488. Disabling the IOMMU can potetially allow DMA transactions to complete without being translated. Leave it enabled, and allow crash kernel to do the IOMMU reinitialization properly. Cc: Joerg Roedel joerg.roe...@amd.com Cc: