Re: [PATCH] x86: Find offset for crashkernel reservation automatically

Bernhard Walle <[EMAIL PROTECTED]> writes: > But such checks belong into kexec itself, not into the kernel. Exactly. All we should do in the kernel is bias the search for a low address. As that is more usable. Eric ___ kexec mailing list kexec@lists.

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

On Mon, Jul 14, 2008 at 11:55 AM, Eric W. Biederman <[EMAIL PROTECTED]> wrote: > Bernhard Walle <[EMAIL PROTECTED]> writes: > >> But such checks belong into kexec itself, not into the kernel. > > Exactly. > > All we should do in the kernel is bias the search for a low address. > As that is more usa

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

* Eric W. Biederman [2008-07-14 11:41]: > > "Yinghai Lu" <[EMAIL PROTECTED]> writes: > > > On Mon, Jul 14, 2008 at 10:30 AM, Eric W. Biederman > > <[EMAIL PROTECTED]> wrote: > >> "Yinghai Lu" <[EMAIL PROTECTED]> writes: > >> > >>> for 64bit, kdump can start from address above 4g with bzImage? > >>

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

"Yinghai Lu" <[EMAIL PROTECTED]> writes: > On Mon, Jul 14, 2008 at 10:30 AM, Eric W. Biederman > <[EMAIL PROTECTED]> wrote: >> "Yinghai Lu" <[EMAIL PROTECTED]> writes: >> >>> for 64bit, kdump can start from address above 4g with bzImage? >> >> I think the boot protocol was properly updated for tha

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

On Mon, Jul 14, 2008 at 10:30 AM, Eric W. Biederman <[EMAIL PROTECTED]> wrote: > "Yinghai Lu" <[EMAIL PROTECTED]> writes: > >> for 64bit, kdump can start from address above 4g with bzImage? > > I think the boot protocol was properly updated for that, it was in the > original drafts. A relocatable

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

"Yinghai Lu" <[EMAIL PROTECTED]> writes: > for 64bit, kdump can start from address above 4g with bzImage? I think the boot protocol was properly updated for that, it was in the original drafts. A relocatable kernel with a 64bit entry point. I do know that the 64bit code is ready to do that. Er

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

On Mon, Jul 14, 2008 at 2:44 AM, Eric W. Biederman <[EMAIL PROTECTED]> wrote: > Bernhard Walle <[EMAIL PROTECTED]> writes: > >> * Yinghai Lu [2008-07-14 00:11]: >>> >>> should use min_t(u64, 1ULL<<32, max_low_pfn<> >> Shouldn't we use min_t(u64, ULLONG_MAX, max_low_pfn<> should we really limit the

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

Bernhard Walle <[EMAIL PROTECTED]> writes: > * Yinghai Lu [2008-07-14 00:11]: >> >> should use min_t(u64, 1ULL<<32, max_low_pfn< > Shouldn't we use min_t(u64, ULLONG_MAX, max_low_pfn< should we really limit the crashkernel to a 32 bit address on a 64 bit > system? We should use the lowest physic

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

* Yinghai Lu [2008-07-14 00:11]: > > should use min_t(u64, 1ULL<<32, max_low_pfn

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

On Fri, Jun 27, 2008 at 6:32 AM, Vivek Goyal <[EMAIL PROTECTED]> wrote: > On Thu, Jun 26, 2008 at 09:54:08PM +0200, Bernhard Walle wrote: >> This patch removes the need of the [EMAIL PROTECTED] parameter to define >> a fixed offset for crashkernel reservation. That feature can be used together >> w

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

* Bernhard Walle <[EMAIL PROTECTED]> wrote: > > > But in general policy should go in userspace (if possible), so I > > > agree with you that kexec-tools can handle that. > > > > At a quick skim the patch looks good. I thought I had initially > > implemented the code to work this way but appar

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

* [EMAIL PROTECTED] (Eric W. Biederman) [2008-06-27 11:00]: > Bernhard Walle <[EMAIL PROTECTED]> writes: > > > Ah, that's true. Only on x86, right? (That would be an alternative for > > ia64, too ...) > > > > But in general policy should go in userspace (if possible), so I agree > > with you that

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

Bernhard Walle <[EMAIL PROTECTED]> writes: > Ah, that's true. Only on x86, right? (That would be an alternative for > ia64, too ...) > > But in general policy should go in userspace (if possible), so I agree > with you that kexec-tools can handle that. At a quick skim the patch looks good. I th

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

* Vivek Goyal [2008-06-27 10:19]: > On Fri, Jun 27, 2008 at 04:06:56PM +0200, Bernhard Walle wrote: > > * Vivek Goyal [2008-06-27 09:42]: > > > > > > Thinking more about. Let me step back. I think it is not good idea to > > > take this kernel take decision about the capability of kernel being > >

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

On Fri, Jun 27, 2008 at 04:06:56PM +0200, Bernhard Walle wrote: > * Vivek Goyal [2008-06-27 09:42]: > > > > Thinking more about. Let me step back. I think it is not good idea to > > take this kernel take decision about the capability of kernel being > > loaded. There is no way we can find out now

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

* Vivek Goyal [2008-06-27 09:42]: > > Thinking more about. Let me step back. I think it is not good idea to > take this kernel take decision about the capability of kernel being > loaded. There is no way we can find out now that if a kernel is capable > of running from this memory location or not.

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

On Fri, Jun 27, 2008 at 09:32:56AM -0400, Vivek Goyal wrote: > On Thu, Jun 26, 2008 at 09:54:08PM +0200, Bernhard Walle wrote: > > This patch removes the need of the [EMAIL PROTECTED] parameter to define > > a fixed offset for crashkernel reservation. That feature can be used > > together > > with

Re: [PATCH] x86: Find offset for crashkernel reservation automatically

On Thu, Jun 26, 2008 at 09:54:08PM +0200, Bernhard Walle wrote: > This patch removes the need of the [EMAIL PROTECTED] parameter to define > a fixed offset for crashkernel reservation. That feature can be used together > with a relocatable kernel where the kexec-tools relocate the kernel and > get

[PATCH] x86: Find offset for crashkernel reservation automatically

This patch removes the need of the [EMAIL PROTECTED] parameter to define a fixed offset for crashkernel reservation. That feature can be used together with a relocatable kernel where the kexec-tools relocate the kernel and get the actual offset from /proc/iomem. The use case is a kernel where the