Re: [SeaBIOS] [RFC 2/3] Transitions to and from 64 bits

2015-09-21 Thread Gerd Hoffmann
Hi, > +/ > + * Page table > + / > +void *gen_identity_page_table(u64 max_addr) > +{ > +/* Map directly all the addresses */ > +u32 pt_entries = (max_addr +

Re: [SeaBIOS] [RFC 2/3] Transitions to and from 64 bits

2015-09-21 Thread Kevin O'Connor
On Mon, Sep 21, 2015 at 12:03:29PM -0400, Kevin O'Connor wrote: > According to my Intel docs, PAE supports the same 52bit physical > address range that x86_64 supports. Thinking about this further, I could see changing SeaBIOS' 32bit code to always use PAE mode (when available). Doing so would

Re: [SeaBIOS] [RFC 2/3] Transitions to and from 64 bits

2015-09-21 Thread Kevin O'Connor
On Mon, Sep 21, 2015 at 04:38:00PM +0200, Marc Marí wrote: > On Mon, 21 Sep 2015 15:57:56 +0200 > Gerd Hoffmann wrote: > > > +/ > > > + * Page table > > > +

Re: [SeaBIOS] [RFC 2/3] Transitions to and from 64 bits

2015-09-21 Thread Marc Marí
On Mon, 21 Sep 2015 17:00:51 +0200 Paolo Bonzini wrote: > > > On 21/09/2015 16:38, Marc Marí wrote: > > True. Tried with 2M. The memory used went down from 8M to 24K more > > or less, and the time for the copying went down by 4ms (from 15ms to > > 11ms). The other option

Re: [SeaBIOS] [RFC 2/3] Transitions to and from 64 bits

2015-09-21 Thread Marc Marí
On Mon, 21 Sep 2015 12:03:29 -0400 "Kevin O'Connor" wrote: > On Mon, Sep 21, 2015 at 04:38:00PM +0200, Marc Marí wrote: > > On Mon, 21 Sep 2015 15:57:56 +0200 > > Gerd Hoffmann wrote: > > > >

Re: [SeaBIOS] [RFC 2/3] Transitions to and from 64 bits

2015-09-21 Thread Kevin O'Connor
On Mon, Sep 21, 2015 at 06:23:54PM +0200, Marc Marí wrote: > On Mon, 21 Sep 2015 12:03:29 -0400 > "Kevin O'Connor" wrote: > > Also, your code seems to run regular 32bit code when in "long mode" - > > is that valid? > > It doesn't crash. And I think it's valid. When

Re: [SeaBIOS] [RFC 2/3] Transitions to and from 64 bits

2015-09-21 Thread Marc Marí
On Mon, 21 Sep 2015 15:57:56 +0200 Gerd Hoffmann wrote: > Hi, > > > +/ > > + * Page table > > + / > > +void *gen_identity_page_table(u64 max_addr)

Re: [SeaBIOS] [RFC 2/3] Transitions to and from 64 bits

2015-09-21 Thread Paolo Bonzini
On 21/09/2015 16:38, Marc Marí wrote: > True. Tried with 2M. The memory used went down from 8M to 24K more or > less, and the time for the copying went down by 4ms (from 15ms to > 11ms). The other option is 1GB. I'll test later if it's enabled in QEMU > CPUs. Only in some AMD CPUs (plus "-cpu