diff for duplicates of <1368137583.654.15@snotra> diff --git a/a/1.txt b/N1/1.txt index e9d0d3f..1de6f36 100644 --- a/a/1.txt +++ b/N1/1.txt @@ -3,7 +3,7 @@ On 05/09/2013 05:07:00 PM, Benjamin Herrenschmidt wrote: > > On 05/09/2013 07:37:42 AM, Benjamin Herrenschmidt wrote: > > > On Thu, 2013-05-09 at 17:44 +0800, tiejun.chen wrote: > > > > -> > > > Actually in the case GS=3D1 even if EE=3D0, EXT/DEC/DBELL still =20 +> > > > Actually in the case GS=1 even if EE=0, EXT/DEC/DBELL still > occur > > > as I > > > > recall. @@ -11,15 +11,15 @@ On 05/09/2013 05:07:00 PM, Benjamin Herrenschmidt wrote: > > > Only if directed to the hypervisor. > > > > This is always the case with KVM, right? At least on booke... ->=20 -> Hrm, on A2 we could choose iirc. Well not DEC but EXT at least, I =20 +> +> Hrm, on A2 we could choose iirc. Well not DEC but EXT at least, I > don't > remember about DBELL. -As for as the hardware goes we can choose on e500mc as well -- we just =20 -don't support it in KVM, because we can't distinguish between guest and =20 -host interrupts without routing the latter to critical, which is way =20 -too much of a pain for Linux (we did it in our standalone hypervisor, =20 +As for as the hardware goes we can choose on e500mc as well -- we just +don't support it in KVM, because we can't distinguish between guest and +host interrupts without routing the latter to critical, which is way +too much of a pain for Linux (we did it in our standalone hypervisor, though). --Scott= +-Scott diff --git a/a/content_digest b/N1/content_digest index 628c06b..fbe9d16 100644 --- a/a/content_digest +++ b/N1/content_digest @@ -19,7 +19,7 @@ "> > On 05/09/2013 07:37:42 AM, Benjamin Herrenschmidt wrote:\n" "> > > On Thu, 2013-05-09 at 17:44 +0800, tiejun.chen wrote:\n" "> > > >\n" - "> > > > Actually in the case GS=3D1 even if EE=3D0, EXT/DEC/DBELL still =20\n" + "> > > > Actually in the case GS=1 even if EE=0, EXT/DEC/DBELL still \n" "> occur\n" "> > > as I\n" "> > > > recall.\n" @@ -27,17 +27,17 @@ "> > > Only if directed to the hypervisor.\n" "> >\n" "> > This is always the case with KVM, right? At least on booke...\n" - ">=20\n" - "> Hrm, on A2 we could choose iirc. Well not DEC but EXT at least, I =20\n" + "> \n" + "> Hrm, on A2 we could choose iirc. Well not DEC but EXT at least, I \n" "> don't\n" "> remember about DBELL.\n" "\n" - "As for as the hardware goes we can choose on e500mc as well -- we just =20\n" - "don't support it in KVM, because we can't distinguish between guest and =20\n" - "host interrupts without routing the latter to critical, which is way =20\n" - "too much of a pain for Linux (we did it in our standalone hypervisor, =20\n" + "As for as the hardware goes we can choose on e500mc as well -- we just \n" + "don't support it in KVM, because we can't distinguish between guest and \n" + "host interrupts without routing the latter to critical, which is way \n" + "too much of a pain for Linux (we did it in our standalone hypervisor, \n" "though).\n" "\n" - -Scott= + -Scott -89214a87c590bd2b59651c2a99bae15708a0103f4eba4bcab7c6fffd86f5ce3f +ab035543c3221518a7ef1f0610a8c117e8981fb39bd9f6751049190792cda582
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.