Re: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-09-05 Thread Paolo Bonzini
On 02/09/2016 20:10, Eduardo Habkost wrote: > On Fri, Sep 02, 2016 at 04:56:49PM +0200, Paolo Bonzini wrote: >> On 02/09/2016 16:04, Eduardo Habkost wrote: > [...] >>> Now, why exactly vhost-user-test needs accel=tcg and can't use >>> accel=qtest? >> >> Because it uses iPXE to exercise the NIC. >

Re: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-09-02 Thread Eduardo Habkost
On Fri, Sep 02, 2016 at 04:56:49PM +0200, Paolo Bonzini wrote: > On 02/09/2016 16:04, Eduardo Habkost wrote: [...] > > Now, why exactly vhost-user-test needs accel=tcg and can't use > > accel=qtest? > > Because it uses iPXE to exercise the NIC. Is it used to actually exercise the NIC, or just to

Re: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-09-02 Thread Paolo Bonzini
On 02/09/2016 16:04, Eduardo Habkost wrote: > On Fri, Sep 02, 2016 at 10:59:21AM +0200, Paolo Bonzini wrote: >> >> >> On 01/09/2016 21:34, Eduardo Habkost wrote: >>> 1) vhost_user_set_mem_table() fails because dev->mem->nregions is 0 >>> 2) dev->mem->nregions is supposed to get new entries based

Re: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-09-02 Thread Eduardo Habkost
On Fri, Sep 02, 2016 at 10:59:21AM +0200, Paolo Bonzini wrote: > > > On 01/09/2016 21:34, Eduardo Habkost wrote: > > 1) vhost_user_set_mem_table() fails because dev->mem->nregions is 0 > > 2) dev->mem->nregions is supposed to get new entries based on the > >memory listener callbacks > > 3) vh

Re: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-09-02 Thread Paolo Bonzini
On 01/09/2016 21:34, Eduardo Habkost wrote: > 1) vhost_user_set_mem_table() fails because dev->mem->nregions is 0 > 2) dev->mem->nregions is supposed to get new entries based on the >memory listener callbacks > 3) vhost_region_add() gets called properly, and calls >vhost_set_memory(), but

Re: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-09-01 Thread Markus Armbruster
Eduardo Habkost writes: > On Wed, Aug 31, 2016 at 02:47:21PM -0700, > no-re...@ec2-52-6-146-230.compute-1.amazonaws.com wrote: > [...] >> GTESTER check-qtest-x86_64 >> qemu-system-x86_64: Failed initializing vhost-user memory map, consider >> using -object memory-backend-file share=on >> qemu-s

Re: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-09-01 Thread Bandan Das
Eduardo Habkost writes: > On Thu, Sep 01, 2016 at 02:39:37PM -0300, Eduardo Habkost wrote: >> On Thu, Sep 01, 2016 at 05:41:52PM +0200, Paolo Bonzini wrote: >> > On 01/09/2016 17:10, Eduardo Habkost wrote: >> > > Ouch. It looks like the ordering requirements are messier than I >> > > thought. vho

Re: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-09-01 Thread Eduardo Habkost
On Thu, Sep 01, 2016 at 02:39:37PM -0300, Eduardo Habkost wrote: > On Thu, Sep 01, 2016 at 05:41:52PM +0200, Paolo Bonzini wrote: > > On 01/09/2016 17:10, Eduardo Habkost wrote: > > > Ouch. It looks like the ordering requirements are messier than I > > > thought. vhost-user depends on the memory ba

Re: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-09-01 Thread Eduardo Habkost
On Thu, Sep 01, 2016 at 05:41:52PM +0200, Paolo Bonzini wrote: > On 01/09/2016 17:10, Eduardo Habkost wrote: > > Ouch. It looks like the ordering requirements are messier than I > > thought. vhost-user depends on the memory backends to be already > > initialized. > > You could also look at delayin

Re: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-09-01 Thread Eduardo Habkost
On Thu, Sep 01, 2016 at 06:52:42PM +0200, Michal Privoznik wrote: > On 01.09.2016 17:10, Eduardo Habkost wrote: > > On Wed, Aug 31, 2016 at 02:47:21PM -0700, > > no-re...@ec2-52-6-146-230.compute-1.amazonaws.com wrote: > > [...] > >> GTESTER check-qtest-x86_64 > >> qemu-system-x86_64: Failed initi

Re: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-09-01 Thread Michal Privoznik
On 01.09.2016 17:10, Eduardo Habkost wrote: > On Wed, Aug 31, 2016 at 02:47:21PM -0700, > no-re...@ec2-52-6-146-230.compute-1.amazonaws.com wrote: > [...] >> GTESTER check-qtest-x86_64 >> qemu-system-x86_64: Failed initializing vhost-user memory map, consider >> using -object memory-backend-file

Re: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-09-01 Thread Paolo Bonzini
On 01/09/2016 17:10, Eduardo Habkost wrote: > Ouch. It looks like the ordering requirements are messier than I > thought. vhost-user depends on the memory backends to be already > initialized. You could also look at delaying initialization of vhost-user, not sending anything on the wire until af

Re: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-09-01 Thread Eduardo Habkost
On Wed, Aug 31, 2016 at 02:47:21PM -0700, no-re...@ec2-52-6-146-230.compute-1.amazonaws.com wrote: [...] > GTESTER check-qtest-x86_64 > qemu-system-x86_64: Failed initializing vhost-user memory map, consider using > -object memory-backend-file share=on > qemu-system-x86_64: vhost_set_mem_table fa

Re: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-08-31 Thread no-reply
Hi, Your series failed automatic build test. Please find the testing commands and their output below. If you have docker installed, you can probably reproduce it locally. Subject: [Qemu-devel] [PATCH] vl: Delay initialization of memory backends Type: series Message-id: 1472674630-18886-1-git

[Qemu-devel] [PATCH] vl: Delay initialization of memory backends

2016-08-31 Thread Eduardo Habkost
Initialization of memory backends may take a while when prealloc=yes is used, depending on their size. Initializing memory backends before chardevs may delay the creation of monitor sockets, and trigger timeouts on management software that waits until the monitor socket is created by QEMU. See, fo