Re: [Qemu-devel] [PATCH v2 4/7] block: take lock around bdrv_read implementations

2011-11-07 Thread Avi Kivity
On 11/08/2011 12:26 AM, Avi Kivity wrote: > On 11/06/2011 07:25 PM, Paolo Bonzini wrote: > > > > Perhaps the failure is only reproduced 80-90% of the time and this > > screws up the bisection. > > Correct, bad bisect. It actually reproduces reliably, when the bisecter > is reliable. > > The new ca

Re: [Qemu-devel] [PATCH v2 4/7] block: take lock around bdrv_read implementations

2011-11-07 Thread Avi Kivity
On 11/06/2011 07:25 PM, Paolo Bonzini wrote: > > Perhaps the failure is only reproduced 80-90% of the time and this > screws up the bisection. Correct, bad bisect. It actually reproduces reliably, when the bisecter is reliable. The new candidate (disclaimer: unreliable bisecter) is: commit a5c5

Re: [Qemu-devel] [PATCH v2 4/7] block: take lock around bdrv_read implementations

2011-11-07 Thread Kevin Wolf
Am 06.11.2011 15:27, schrieb Avi Kivity: > On 10/20/2011 01:16 PM, Paolo Bonzini wrote: >> This does the first part of the conversion to coroutines, by >> wrapping bdrv_read implementations to take the read side of the >> rwlock. >> >> Drivers that implement bdrv_read rather than bdrv_co_readv can

Re: [Qemu-devel] [PATCH v2 4/7] block: take lock around bdrv_read implementations

2011-11-06 Thread Avi Kivity
On 11/06/2011 07:25 PM, Paolo Bonzini wrote: > On 11/06/2011 03:27 PM, Avi Kivity wrote: >> On 10/20/2011 01:16 PM, Paolo Bonzini wrote: >>> This does the first part of the conversion to coroutines, by >>> wrapping bdrv_read implementations to take the read side of the >>> rwlock. >>> >>> Drivers t

Re: [Qemu-devel] [PATCH v2 4/7] block: take lock around bdrv_read implementations

2011-11-06 Thread Paolo Bonzini
On 11/06/2011 03:27 PM, Avi Kivity wrote: On 10/20/2011 01:16 PM, Paolo Bonzini wrote: This does the first part of the conversion to coroutines, by wrapping bdrv_read implementations to take the read side of the rwlock. Drivers that implement bdrv_read rather than bdrv_co_readv can then benefit

Re: [Qemu-devel] [PATCH v2 4/7] block: take lock around bdrv_read implementations

2011-11-06 Thread Avi Kivity
On 10/20/2011 01:16 PM, Paolo Bonzini wrote: > This does the first part of the conversion to coroutines, by > wrapping bdrv_read implementations to take the read side of the > rwlock. > > Drivers that implement bdrv_read rather than bdrv_co_readv can > then benefit from asynchronous operation (at l

[Qemu-devel] [PATCH v2 4/7] block: take lock around bdrv_read implementations

2011-10-20 Thread Paolo Bonzini
This does the first part of the conversion to coroutines, by wrapping bdrv_read implementations to take the read side of the rwlock. Drivers that implement bdrv_read rather than bdrv_co_readv can then benefit from asynchronous operation (at least if the underlying protocol supports it, which is no