Re: [Qemu-devel] fixing qemu busy wait

Il 18/04/2013 20:47, Orr Dvory ha scritto: > is this better? Yes -- but I don't remember if this was fixed elsewhere. Can you reproduce the problem in 1.5.0? If so, please send the patch according to the guidelines at http://wiki.qemu.org/Contribute/SubmitAPatch. Paolo > diff -uprN qemu-1.4.1/

Re: [Qemu-devel] fixing qemu busy wait

is this better? diff -uprN qemu-1.4.1/gdbstub.c qemu-1.4.1-fix/gdbstub.c --- qemu-1.4.1/gdbstub.c2013-04-15 23:25:18.0 +0300 +++ qemu-1.4.1-fix/gdbstub.c2013-04-18 21:35:00.0 +0300 @@ -379,17 +379,22 @@ static void put_buffer(GDBState *s, cons { #ifdef CONFIG_USER_ONLY

Re: [Qemu-devel] fixing qemu busy wait

Il 20/02/2013 20:39, Orr Dvory ha scritto: > if (errno != EINTR && errno != EAGAIN) > +/* there's no need to restore the > +O_NONBLOCK option. */ > return; > + But it's cleaner to do it anyway. :) Paolo

Re: [Qemu-devel] fixing qemu busy wait

Am 20.02.2013 20:39, schrieb Orr Dvory: > the new patch is attached... Please see http://wiki.qemu.org/Contribute/SubmitAPatch for multiple reasons why your patch won't get applied or get much review as is. Andreas > > > On Sun, Feb 17, 2013 at 6:32 PM, Orr Dvory > wr

Re: [Qemu-devel] fixing qemu busy wait

the new patch is attached... On Sun, Feb 17, 2013 at 6:32 PM, Orr Dvory wrote: > Yes it would :) > Should I do that? or that you already fixed the issue? > > > On Sun, Feb 17, 2013 at 12:06 AM, Richard Henderson wrote: > >> On 2013-02-15 03:12, Orr Dvory wrote: >> >>> when debugging with qemu(u

Re: [Qemu-devel] fixing qemu busy wait

On 2013-02-15 03:12, Orr Dvory wrote: when debugging with qemu(user mode), qemu waits in infinite loop to read a signal from gdb (when it waits on breakpoint for example). I added sleeps to reduce the cpu usage from 100% to about ~0%. Wouldn't it be better to toggle the O_NONBLOCK state of the