[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <f0be0592-2839-c87e-0f44-bd10956f7704@suse.cz>
Date: Thu, 19 Apr 2018 20:28:30 +0200
From: Vlastimil Babka <vbabka@...e.cz>
To: Mikulas Patocka <mpatocka@...hat.com>,
David Miller <davem@...emloft.net>,
Andrew Morton <akpm@...ux-foundation.org>, linux-mm@...ck.org
Cc: eric.dumazet@...il.com, edumazet@...gle.com,
bhutchings@...arflare.com, netdev@...r.kernel.org,
linux-kernel@...r.kernel.org, mst@...hat.com, jasowang@...hat.com,
virtualization@...ts.linux-foundation.org, dm-devel@...hat.com,
Laura Abbott <labbott@...hat.com>
Subject: Re: [PATCH] kvmalloc: always use vmalloc if CONFIG_DEBUG_VM
On 04/19/2018 06:12 PM, Mikulas Patocka wrote:
> From: Mikulas Patocka <mpatocka@...hat.com>
> Subject: [PATCH] kvmalloc: always use vmalloc if CONFIG_DEBUG_VM
>
> The kvmalloc function tries to use kmalloc and falls back to vmalloc if
> kmalloc fails.
>
> Unfortunatelly, some kernel code has bugs - it uses kvmalloc and then
> uses DMA-API on the returned memory or frees it with kfree. Such bugs were
> found in the virtio-net driver, dm-integrity or RHEL7 powerpc-specific
> code.
>
> These bugs are hard to reproduce because vmalloc falls back to kmalloc
> only if memory is fragmented.
>
> In order to detect these bugs reliably I submit this patch that changes
> kvmalloc to always use vmalloc if CONFIG_DEBUG_VM is turned on.
>
> Signed-off-by: Mikulas Patocka <mpatocka@...hat.com>
Hmm AFAIK Fedora uses CONFIG_DEBUG_VM in their kernels. Sure you want to
impose this on all users? Seems too much for DEBUG_VM to me. Maybe it
should be hidden under some error injection config?
> ---
> mm/util.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> Index: linux-2.6/mm/util.c
> ===================================================================
> --- linux-2.6.orig/mm/util.c 2018-04-18 15:46:23.000000000 +0200
> +++ linux-2.6/mm/util.c 2018-04-18 16:00:43.000000000 +0200
> @@ -395,6 +395,7 @@ EXPORT_SYMBOL(vm_mmap);
> */
> void *kvmalloc_node(size_t size, gfp_t flags, int node)
> {
> +#ifndef CONFIG_DEBUG_VM
> gfp_t kmalloc_flags = flags;
> void *ret;
>
> @@ -426,6 +427,7 @@ void *kvmalloc_node(size_t size, gfp_t f
> */
> if (ret || size <= PAGE_SIZE)
> return ret;
> +#endif
Did you verify that vmalloc does the right thing for sub-page sizes?
Shouldn't those be exempted?
> return __vmalloc_node_flags_caller(size, node, flags,
> __builtin_return_address(0));
>
Powered by blists - more mailing lists