Re: [PATCH v2 2/2] kallsyms: handle special absolute symbols

2014-03-16 Thread Rusty Russell
Kees Cook writes: > On Wed, Mar 12, 2014 at 8:40 PM, Rusty Russell wrote: >> Kees Cook writes: >>> Why not just do this with 0-base-address detection like my v2? That >>> would mean we don't need to remember to add this flag in the future to >>> imagined new architectures that might want this 0-

Re: [PATCH v2 2/2] kallsyms: handle special absolute symbols

2014-03-12 Thread Kees Cook
On Wed, Mar 12, 2014 at 8:40 PM, Rusty Russell wrote: > Kees Cook writes: >> Why not just do this with 0-base-address detection like my v2? That >> would mean we don't need to remember to add this flag in the future to >> imagined new architectures that might want this 0-based per_cpu >> feature.

Re: [PATCH v2 2/2] kallsyms: handle special absolute symbols

2014-03-12 Thread Rusty Russell
Kees Cook writes: > Why not just do this with 0-base-address detection like my v2? That > would mean we don't need to remember to add this flag in the future to > imagined new architectures that might want this 0-based per_cpu > feature. Because future architectures will get this right and emit a

Re: [PATCH v2 2/2] kallsyms: handle special absolute symbols

2014-03-12 Thread Kees Cook
On Wed, Mar 12, 2014 at 5:30 PM, Rusty Russell wrote: > Kees Cook writes: >> On Fri, Mar 7, 2014 at 5:00 PM, Kees Cook wrote: >>> This forces the entire per_cpu range to be reported as absolute >>> without losing their linker symbol types, when the per_cpu area is >>> 0-based. Without this, the

Re: [PATCH v2 2/2] kallsyms: handle special absolute symbols

2014-03-12 Thread Rusty Russell
Kees Cook writes: > On Fri, Mar 7, 2014 at 5:00 PM, Kees Cook wrote: >> This forces the entire per_cpu range to be reported as absolute >> without losing their linker symbol types, when the per_cpu area is >> 0-based. Without this, the variables are incorrectly shown as relocated >> under kASLR o

Re: [PATCH v2 2/2] kallsyms: handle special absolute symbols

2014-03-11 Thread Kees Cook
On Fri, Mar 7, 2014 at 5:00 PM, Kees Cook wrote: > This forces the entire per_cpu range to be reported as absolute > without losing their linker symbol types, when the per_cpu area is > 0-based. Without this, the variables are incorrectly shown as relocated > under kASLR on x86_64. > > Several kal

[PATCH v2 2/2] kallsyms: handle special absolute symbols

2014-03-07 Thread Kees Cook
This forces the entire per_cpu range to be reported as absolute without losing their linker symbol types, when the per_cpu area is 0-based. Without this, the variables are incorrectly shown as relocated under kASLR on x86_64. Several kallsyms output in different boot states for comparison of vario