From: Ingo Molnar <mingo@elte.hu>
To: Cliff Wickman <cpw@sgi.com>
Cc: linux-kernel@vger.kernel.org, hpa@zytor.com,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [PATCH] x86: UV BAU activation descriptor init
Date: Wed, 20 May 2009 10:33:50 +0200 [thread overview]
Message-ID: <20090520083350.GE6736@elte.hu> (raw)
In-Reply-To: <E1M6YCG-0002eM-1Z@eag09.americas.sgi.com>
* Cliff Wickman <cpw@sgi.com> wrote:
> From: Cliff Wickman <cpw@sgi.com>
>
> The UV tlb shootdown code has a serious initialization error.
>
> An array of structures [32*8] is initialized as if it were [32].
> The array is indexed by (cpu number on the blade)*8, so the short
> initialization works for up to 4 cpus on a blade.
> But above that, we provide an invalid opcode to the hub's
> broadcast assist unit.
>
> This patch changes the allocation of the array to use its symbolic
> dimensions for better clarity. And initializes all 32*8 entries.
>
> Tested on the UV simulator.
>
> Diffed against 2.6.30-rc6
>
> Signed-off-by: Cliff Wickman <cpw@sgi.com>
>
> ---
> arch/x86/kernel/tlb_uv.c | 18 ++++++++++++++++--
> 1 file changed, 16 insertions(+), 2 deletions(-)
>
> Index: linux/arch/x86/kernel/tlb_uv.c
> ===================================================================
> --- linux.orig/arch/x86/kernel/tlb_uv.c
> +++ linux/arch/x86/kernel/tlb_uv.c
> @@ -715,7 +715,14 @@ uv_activation_descriptor_init(int node,
> struct bau_desc *adp;
> struct bau_desc *ad2;
>
> - adp = (struct bau_desc *)kmalloc_node(16384, GFP_KERNEL, node);
> + /*
> + * each bau_desc is 64 bytes; there are 8 (UV_ITEMS_PER_DESCRIPTOR)
> + * per cpu; and up to 32 (UV_ACTIVATION_DESCRIPTOR_SIZE) cpu's
> + * per blade
> + */
> + adp = (struct bau_desc *)kmalloc_node(sizeof(struct bau_desc)*
> + UV_ACTIVATION_DESCRIPTOR_SIZE*UV_ITEMS_PER_DESCRIPTOR,
> + GFP_KERNEL, node);
> BUG_ON(!adp);
>
> pa = uv_gpa(adp); /* need the real nasid*/
> @@ -729,7 +736,14 @@ uv_activation_descriptor_init(int node,
> (n << UV_DESC_BASE_PNODE_SHIFT | m));
> }
>
> - for (i = 0, ad2 = adp; i < UV_ACTIVATION_DESCRIPTOR_SIZE; i++, ad2++) {
> + /*
> + * initializing all 8 (UV_ITEMS_PER_DESCRIPTOR) descriptors for each
> + * cpu even though we only use the first one; one descriptor can
> + * describe a broadcast to 256 nodes.
> + */
> + for (i = 0, ad2 = adp;
> + i < (UV_ACTIVATION_DESCRIPTOR_SIZE*UV_ITEMS_PER_DESCRIPTOR);
> + i++, ad2++) {
> memset(ad2, 0, sizeof(struct bau_desc));
> ad2->header.sw_ack_flag = 1;
> /*
looks good but could you please define a shortcut for:
UV_ACTIVATION_DESCRIPTOR_SIZE*UV_ITEMS_PER_DESCRIPTOR
because its length causes a lot of ugly linebreaks. Something like
UV_ADP_SIZE should do the trick?
Ingo
prev parent reply other threads:[~2009-05-20 8:34 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-05-19 22:54 [PATCH] x86: UV BAU activation descriptor init Cliff Wickman
2009-05-20 8:33 ` Ingo Molnar [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20090520083350.GE6736@elte.hu \
--to=mingo@elte.hu \
--cc=cpw@sgi.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=tglx@linutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).