From: Dave Hansen <dave.hansen@intel.com>
To: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>,
x86@kernel.org, platform-driver-x86@vger.kernel.org
Cc: sean.j.christopherson@intel.com, nhorman@redhat.com,
npmccallum@redhat.com, linux-sgx@vger.kernel.org,
Serge Ayoun <serge.ayoun@intel.com>,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>,
Suresh Siddha <suresh.b.siddha@intel.com>,
"open list:X86 ARCHITECTURE (32-BIT AND 64-BIT)"
<linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v13 07/13] x86/sgx: Add data structures for tracking the EPC pages
Date: Mon, 27 Aug 2018 14:07:53 -0700 [thread overview]
Message-ID: <4666cae8-c711-8dd5-cbce-3d97cc19a9e5@intel.com> (raw)
In-Reply-To: <20180827185507.17087-8-jarkko.sakkinen@linux.intel.com>
On 08/27/2018 11:53 AM, Jarkko Sakkinen wrote:
> Add data structures to track Enclave Page Cache (EPC) pages. EPC is
> divided into multiple banks (1-N) of which addresses and sizes can be
> enumerated with CPUID by the OS.
>
> On NUMA systems a node can have at most bank. A bank can be at most part of
> two nodes. SGX supports both nodes with a single memory controller and also
> sub-cluster nodes with severals memory controllers on a single die.
>
> Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
> Co-developed-by: Serge Ayoun <serge.ayoun@intel.com>
> Co-developed-by: Sean Christopherson <sean.j.christopherson@intel.com>
> Signed-off-by: Serge Ayoun <serge.ayoun@intel.com>
> Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
> ---
> arch/x86/include/asm/sgx.h | 60 ++++++++++++++++++
> arch/x86/kernel/cpu/intel_sgx.c | 106 +++++++++++++++++++++++++++++++-
> 2 files changed, 164 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/include/asm/sgx.h b/arch/x86/include/asm/sgx.h
> index 2130e639ab49..17b7b3aa66bf 100644
> --- a/arch/x86/include/asm/sgx.h
> +++ b/arch/x86/include/asm/sgx.h
> @@ -4,9 +4,69 @@
> #ifndef _ASM_X86_SGX_H
> #define _ASM_X86_SGX_H
>
> +#include <linux/bitops.h>
> +#include <linux/err.h>
> +#include <linux/rwsem.h>
> #include <linux/types.h>
> +#include <asm/sgx_arch.h>
> +#include <asm/asm.h>
> +
> +#define SGX_MAX_EPC_BANKS 8
This is _still_ missing a meaningful description of what a bank is and
whether it is a hardware or software structure.
It would also help us to determine whether your bit packing below is
really required.
> +struct sgx_epc_page {
> + unsigned long desc;
> + struct list_head list;
> +};
> +
> +struct sgx_epc_bank {
> + unsigned long pa;
> + void *va;
> + unsigned long size;
Please add units. size could be bytes or pages, or who knows what. I
can't tell you how many bugs I've tripped over in the past from simple
unit conversions
> + struct sgx_epc_page *pages_data;
> + struct sgx_epc_page **pages;
> + unsigned long free_cnt;
> + spinlock_t lock;
> +};
>
> extern bool sgx_enabled;
> extern bool sgx_lc_enabled;
> +extern struct sgx_epc_bank sgx_epc_banks[SGX_MAX_EPC_BANKS];
> +
> +/*
> + * enum sgx_epc_page_desc - defines bits and masks for an EPC page's desc
Why are you bothering packing these bits? This seems a rather
convoluted way to store two integers.
> +static __init int sgx_init_epc_bank(u64 addr, u64 size, unsigned long index,
> + struct sgx_epc_bank *bank)
> +{
> + unsigned long nr_pages = size >> PAGE_SHIFT;
> + struct sgx_epc_page *pages_data;
> + unsigned long i;
> + void *va;
> +
> + va = ioremap_cache(addr, size);
> + if (!va)
> + return -ENOMEM;
> +
> + pages_data = kcalloc(nr_pages, sizeof(struct sgx_epc_page), GFP_KERNEL);
> + if (!pages_data)
> + goto out_iomap;
This looks like you're roughly limited by the page allocator to a bank
size of ~1.4GB which seems kinda small. Is this really OK?
> + bank->pages = kcalloc(nr_pages, sizeof(struct sgx_epc_page *),
> + GFP_KERNEL);
> + if (!bank->pages)
> + goto out_pdata;
> +
> + for (i = 0; i < nr_pages; i++) {
> + bank->pages[i] = &pages_data[i];
> + bank->pages[i]->desc = (addr + (i << PAGE_SHIFT)) | index;
> + }
> +
> + bank->pa = addr;
> + bank->size = size;
> + bank->va = va;
> + bank->free_cnt = nr_pages;
> + bank->pages_data = pages_data;
> + spin_lock_init(&bank->lock);
> + return 0;
> +out_pdata:
> + kfree(pages_data);
> +out_iomap:
> + iounmap(va);
> + return -ENOMEM;
> +}
> +
> +static __init void sgx_page_cache_teardown(void)
> +{
> + struct sgx_epc_bank *bank;
> + int i;
> +
> + for (i = 0; i < sgx_nr_epc_banks; i++) {
> + bank = &sgx_epc_banks[i];
> + iounmap((void *)bank->va);
> + kfree(bank->pages);
> + kfree(bank->pages_data);
> + }
> +}
> +
> +static inline u64 sgx_combine_bank_regs(u64 low, u64 high)
> +{
> + return (low & 0xFFFFF000) + ((high & 0xFFFFF) << 32);
> +}
-ENOCOMMENT for a rather weird looking calculation
> +static __init int sgx_page_cache_init(void)
> +{
> + u32 eax, ebx, ecx, edx;
> + u64 pa, size;
> + int ret;
> + int i;
> +
> + for (i = 0; i < SGX_MAX_EPC_BANKS; i++) {
> + cpuid_count(SGX_CPUID, 2 + i, &eax, &ebx, &ecx, &edx);
> + if (!(eax & 0xF))
> + break;
So, we have random data coming out of a random CPUID leaf being called
'eax' and then being tested against a random hard-coded mask. This
seems rather unfortunate for someone trying to understand the code. Can
we do better?
> + pa = sgx_combine_bank_regs(eax, ebx);
> + size = sgx_combine_bank_regs(ecx, edx);
> + pr_info("EPC bank 0x%llx-0x%llx\n", pa, pa + size - 1);
> + ret = sgx_init_epc_bank(pa, size, i, &sgx_epc_banks[i]);
> + if (ret) {
> + sgx_page_cache_teardown();
> + return ret;
> + }
So if one bank fails, we tear down all banks, yet leave sgx_nr_epc_banks
incremented? That sounds troublesome.
> + sgx_nr_epc_banks++;
> + }
> +
> + if (!sgx_nr_epc_banks) {
> + pr_err("There are zero EPC banks.\n");
> + return -ENODEV;
> + }
> +
> + return 0;
> +}
Does this support hot-addition of a bank? If not, why not?
next prev parent reply other threads:[~2018-08-27 21:07 UTC|newest]
Thread overview: 103+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-08-27 18:53 [PATCH v13 00/13] Intel SGX1 support Jarkko Sakkinen
2018-08-27 18:53 ` [PATCH v13 01/13] x86/sgx: Update MAINTAINERS Jarkko Sakkinen
2018-09-03 12:56 ` Andy Shevchenko
2018-09-03 19:10 ` Jarkko Sakkinen
2018-08-27 18:53 ` [PATCH v13 02/13] x86/cpufeature: Add SGX and SGX_LC CPU features Jarkko Sakkinen
2018-08-28 0:07 ` Huang, Kai
2018-08-28 7:17 ` Jarkko Sakkinen
2018-08-29 7:36 ` Huang, Kai
2018-08-31 12:19 ` Jarkko Sakkinen
2018-08-31 16:18 ` Dr. Greg
2018-08-27 18:53 ` [PATCH v13 03/13] x86/cpufeatures: Add Intel-defined SGX leaf CPUID_12_EAX Jarkko Sakkinen
2018-08-27 19:39 ` Dave Hansen
2018-08-28 7:23 ` Jarkko Sakkinen
2018-08-28 10:21 ` Borislav Petkov
2018-08-28 10:38 ` Jarkko Sakkinen
2018-08-27 18:53 ` [PATCH v13 04/13] x86/sgx: Architectural structures Jarkko Sakkinen
2018-08-27 19:41 ` Dave Hansen
2018-08-28 8:08 ` Jarkko Sakkinen
2018-09-03 13:16 ` Andy Shevchenko
2018-09-03 19:17 ` Jarkko Sakkinen
2018-09-04 16:04 ` Dave Hansen
2018-09-04 16:06 ` Andy Shevchenko
2018-09-05 17:32 ` Jarkko Sakkinen
2018-08-27 18:53 ` [PATCH v13 05/13] x86/msr: Add SGX definitions to msr-index.h Jarkko Sakkinen
2018-08-27 19:42 ` Dave Hansen
2018-08-28 8:11 ` Jarkko Sakkinen
2018-08-27 18:53 ` [PATCH v13 06/13] x86/sgx: Detect Intel SGX Jarkko Sakkinen
2018-08-27 19:53 ` Dave Hansen
2018-08-28 8:28 ` Jarkko Sakkinen
2018-09-03 14:26 ` Andy Shevchenko
2018-09-04 9:56 ` Jarkko Sakkinen
2018-08-27 18:53 ` [PATCH v13 07/13] x86/sgx: Add data structures for tracking the EPC pages Jarkko Sakkinen
2018-08-27 21:07 ` Dave Hansen [this message]
2018-08-28 10:30 ` Jarkko Sakkinen
2018-08-28 16:53 ` Dave Hansen
2018-08-28 21:34 ` Sean Christopherson
2018-08-31 11:13 ` Jarkko Sakkinen
2018-08-31 11:10 ` Jarkko Sakkinen
2018-09-03 14:41 ` Andy Shevchenko
2018-09-04 9:59 ` Jarkko Sakkinen
2018-09-04 17:49 ` Sean Christopherson
2018-09-04 18:01 ` Andy Shevchenko
2018-09-04 18:17 ` Sean Christopherson
2018-09-05 17:36 ` Jarkko Sakkinen
2018-08-27 18:53 ` [PATCH v13 08/13] x86/sgx: Add wrappers for ENCLS leaf functions Jarkko Sakkinen
2018-09-03 15:01 ` Andy Shevchenko
2018-09-04 11:09 ` Jarkko Sakkinen
2018-08-27 18:53 ` [PATCH v13 09/13] x86/sgx: Enclave Page Cache (EPC) memory manager Jarkko Sakkinen
2018-08-27 21:14 ` Dave Hansen
2018-08-28 8:36 ` Jarkko Sakkinen
2018-08-27 21:15 ` Dave Hansen
2018-08-28 8:35 ` Jarkko Sakkinen
2018-08-28 14:07 ` Dave Hansen
2018-08-28 21:22 ` Sean Christopherson
2018-08-28 21:26 ` Dave Hansen
2018-08-28 21:52 ` Sean Christopherson
2018-08-31 11:22 ` Jarkko Sakkinen
2018-09-03 19:02 ` Andy Shevchenko
2018-09-04 15:38 ` Jarkko Sakkinen
2018-09-04 15:45 ` Sean Christopherson
2018-09-11 15:04 ` Sean Christopherson
2018-09-16 11:40 ` Jarkko Sakkinen
2018-08-27 18:53 ` [PATCH v13 10/13] x86/sgx: Add sgx_einit() for initializing enclaves Jarkko Sakkinen
2018-08-27 21:41 ` Huang, Kai
2018-08-28 7:01 ` Jarkko Sakkinen
2018-08-29 7:33 ` Huang, Kai
2018-08-29 20:33 ` Sean Christopherson
2018-08-29 20:58 ` Huang, Kai
2018-08-29 21:09 ` Sean Christopherson
2018-08-30 1:45 ` Huang, Kai
2018-08-31 17:43 ` Sean Christopherson
2018-08-31 21:34 ` Dr. Greg
2018-09-03 19:27 ` Jarkko Sakkinen
2018-09-03 18:15 ` Jarkko Sakkinen
2018-08-31 12:17 ` Jarkko Sakkinen
2018-08-31 18:15 ` Sean Christopherson
2018-09-03 19:19 ` Jarkko Sakkinen
2018-09-03 23:45 ` Huang, Kai
2018-09-04 14:54 ` Sean Christopherson
2018-09-04 15:30 ` Jarkko Sakkinen
2018-09-04 16:35 ` Sean Christopherson
2018-09-04 22:13 ` Huang, Kai
2018-09-05 17:39 ` Jarkko Sakkinen
2018-09-04 15:26 ` Jarkko Sakkinen
2018-09-03 13:53 ` Jann Horn
2018-09-04 9:55 ` Jarkko Sakkinen
2018-09-04 16:05 ` Andy Shevchenko
2018-08-27 18:53 ` [PATCH v13 11/13] platform/x86: Intel SGX driver Jarkko Sakkinen
2018-09-04 17:59 ` Andy Shevchenko
2018-09-05 17:33 ` Jarkko Sakkinen
2018-09-05 17:36 ` Andy Shevchenko
2018-09-06 9:21 ` Jarkko Sakkinen
2018-09-06 17:35 ` Miguel Ojeda
2018-09-07 0:50 ` Joe Perches
2018-09-07 17:02 ` Sean Christopherson
2018-09-10 18:37 ` Jarkko Sakkinen
2018-09-10 21:22 ` Joe Perches
2018-09-10 18:33 ` Jarkko Sakkinen
2018-08-27 18:53 ` [PATCH v13 12/13] platform/x86: ptrace() support for the " Jarkko Sakkinen
2018-08-27 18:53 ` [PATCH v13 13/13] x86/sgx: Driver documentation Jarkko Sakkinen
2018-08-27 19:40 ` Randy Dunlap
2018-08-28 7:58 ` Jarkko Sakkinen
2018-08-28 8:03 ` Jarkko Sakkinen
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=4666cae8-c711-8dd5-cbce-3d97cc19a9e5@intel.com \
--to=dave.hansen@intel.com \
--cc=hpa@zytor.com \
--cc=jarkko.sakkinen@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-sgx@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=nhorman@redhat.com \
--cc=npmccallum@redhat.com \
--cc=platform-driver-x86@vger.kernel.org \
--cc=sean.j.christopherson@intel.com \
--cc=serge.ayoun@intel.com \
--cc=suresh.b.siddha@intel.com \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
/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).