lkml.org 
[lkml]   [2015]   [Jul]   [16]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH v4 2/2] vhost: add max_mem_regions module parameter
On Thu,  2 Jul 2015 15:08:11 +0200
Igor Mammedov <imammedo@redhat.com> wrote:

> it became possible to use a bigger amount of memory
> slots, which is used by memory hotplug for
> registering hotplugged memory.
> However QEMU crashes if it's used with more than ~60
> pc-dimm devices and vhost-net enabled since host kernel
> in module vhost-net refuses to accept more than 64
> memory regions.
>
> Allow to tweak limit via max_mem_regions module paramemter
> with default value set to 64 slots.
Michael,

what was the reason not to rise default?
As much as I think I can't come up with one.

Making it as module option doesn't make much sense
since old userspace will crash on new kernels anyway
until admins learn that there is a module option
to rise limit.

Rising default limit on par with kvm's will help
to avoid those crashes and would allow to drop
not necessary option to reduce user confusion.

>
> Signed-off-by: Igor Mammedov <imammedo@redhat.com>
> ---
> drivers/vhost/vhost.c | 8 ++++++--
> 1 file changed, 6 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
> index 6488011..9a68e2e 100644
> --- a/drivers/vhost/vhost.c
> +++ b/drivers/vhost/vhost.c
> @@ -29,8 +29,12 @@
>
> #include "vhost.h"
>
> +static ushort max_mem_regions = 64;
> +module_param(max_mem_regions, ushort, 0444);
> +MODULE_PARM_DESC(max_mem_regions,
> + "Maximum number of memory regions in memory map. (default: 64)");
> +
> enum {
> - VHOST_MEMORY_MAX_NREGIONS = 64,
> VHOST_MEMORY_F_LOG = 0x1,
> };
>
> @@ -696,7 +700,7 @@ static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m)
> return -EFAULT;
> if (mem.padding)
> return -EOPNOTSUPP;
> - if (mem.nregions > VHOST_MEMORY_MAX_NREGIONS)
> + if (mem.nregions > max_mem_regions)
> return -E2BIG;
> newmem = vhost_kvzalloc(size + mem.nregions * sizeof(*m->regions));
> if (!newmem)



\
 
 \ /
  Last update: 2015-07-16 13:21    [W:0.118 / U:0.304 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site