Re: [Qemu-devel] [PATCH] exec: separate sections and nodes per address space

2013-12-11 Thread Marcel Apfelbaum
On Tue, 2013-12-10 at 18:48 +0200, Michael S. Tsirkin wrote: > On Tue, Dec 10, 2013 at 02:50:32PM +0200, Marcel Apfelbaum wrote: > > On Tue, 2013-12-10 at 13:38 +0100, Paolo Bonzini wrote: > > > Il 10/12/2013 13:37, Marcel Apfelbaum ha scritto: > > > > > > > > Beside this it looks OK, the branch d

Re: [Qemu-devel] [PATCH] exec: separate sections and nodes per address space

2013-12-10 Thread Michael S. Tsirkin
On Tue, Dec 10, 2013 at 02:50:32PM +0200, Marcel Apfelbaum wrote: > On Tue, 2013-12-10 at 13:38 +0100, Paolo Bonzini wrote: > > Il 10/12/2013 13:37, Marcel Apfelbaum ha scritto: > > > > > > Beside this it looks OK, the branch does not compile and I couldn't look > > > into it more... > > > Can yo

Re: [Qemu-devel] [PATCH] exec: separate sections and nodes per address space

2013-12-10 Thread Marcel Apfelbaum
On Tue, 2013-12-10 at 13:38 +0100, Paolo Bonzini wrote: > Il 10/12/2013 13:37, Marcel Apfelbaum ha scritto: > > > > Beside this it looks OK, the branch does not compile and I couldn't look > > into it more... > > Can you please also merge my other patch > > "memory.c: bugfix - ref counting mismat

Re: [Qemu-devel] [PATCH] exec: separate sections and nodes per address space

2013-12-10 Thread Paolo Bonzini
Il 10/12/2013 13:37, Marcel Apfelbaum ha scritto: > > Beside this it looks OK, the branch does not compile and I couldn't look into > it more... > Can you please also merge my other patch > "memory.c: bugfix - ref counting mismatch in memory_region_find" ? Regarding the compilation issue, I sugg

Re: [Qemu-devel] [PATCH] exec: separate sections and nodes per address space

2013-12-10 Thread Marcel Apfelbaum
On Tue, 2013-12-10 at 13:49 +0200, Michael S. Tsirkin wrote: > On Sun, Dec 01, 2013 at 02:02:23PM +0200, Marcel Apfelbaum wrote: > > Every address space has its own nodes and sections, but > > it uses the same global arrays of nodes/section. > > > > This limits the number of devices that can be at

Re: [Qemu-devel] [PATCH] exec: separate sections and nodes per address space

2013-12-10 Thread Michael S. Tsirkin
On Sun, Dec 01, 2013 at 02:02:23PM +0200, Marcel Apfelbaum wrote: > Every address space has its own nodes and sections, but > it uses the same global arrays of nodes/section. > > This limits the number of devices that can be attached > to the guest to 20-30 devices. It happens because: > - The se

Re: [Qemu-devel] [PATCH] exec: separate sections and nodes per address space

2013-12-09 Thread Paolo Bonzini
Il 01/12/2013 13:02, Marcel Apfelbaum ha scritto: > Every address space has its own nodes and sections, but > it uses the same global arrays of nodes/section. > > This limits the number of devices that can be attached > to the guest to 20-30 devices. It happens because: > - The sections array is

Re: [Qemu-devel] [PATCH] exec: separate sections and nodes per address space

2013-12-02 Thread Michael S. Tsirkin
On Sun, Dec 01, 2013 at 02:02:23PM +0200, Marcel Apfelbaum wrote: > Every address space has its own nodes and sections, but > it uses the same global arrays of nodes/section. > > This limits the number of devices that can be attached > to the guest to 20-30 devices. It happens because: > - The se

[Qemu-devel] [PATCH] exec: separate sections and nodes per address space

2013-12-01 Thread Marcel Apfelbaum
Every address space has its own nodes and sections, but it uses the same global arrays of nodes/section. This limits the number of devices that can be attached to the guest to 20-30 devices. It happens because: - The sections array is limited to 2^12 entries. - The main memory has at least 100 s