From: Thiago Jung Bauermann <bauerman@linux.vnet.ibm.com>
To: Ram Pai <linuxram@us.ibm.com>
Cc: linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org,
linux-arch@vger.kernel.org, linux-mm@kvack.org, x86@kernel.org,
linux-doc@vger.kernel.org, linux-kselftest@vger.kernel.org,
arnd@arndb.de, corbet@lwn.net, mhocko@kernel.org,
dave.hansen@intel.com, mingo@redhat.com, paulus@samba.org,
aneesh.kumar@linux.vnet.ibm.com, akpm@linux-foundation.org,
khandual@linux.vnet.ibm.com
Subject: Re: [RFC v6 21/62] powerpc: introduce execute-only pkey
Date: Mon, 31 Jul 2017 13:19:40 -0300 [thread overview]
Message-ID: <87efsw60kj.fsf@linux.vnet.ibm.com> (raw)
In-Reply-To: <20170730005137.GK5664@ram.oc3035372033.ibm.com>
Ram Pai <linuxram@us.ibm.com> writes:
> On Fri, Jul 28, 2017 at 07:17:13PM -0300, Thiago Jung Bauermann wrote:
>>
>> Ram Pai <linuxram@us.ibm.com> writes:
>> > --- a/arch/powerpc/mm/pkeys.c
>> > +++ b/arch/powerpc/mm/pkeys.c
>> > @@ -97,3 +97,60 @@ int __arch_set_user_pkey_access(struct task_struct *tsk, int pkey,
>> > init_iamr(pkey, new_iamr_bits);
>> > return 0;
>> > }
>> > +
>> > +static inline bool pkey_allows_readwrite(int pkey)
>> > +{
>> > + int pkey_shift = pkeyshift(pkey);
>> > +
>> > + if (!(read_uamor() & (0x3UL << pkey_shift)))
>> > + return true;
>> > +
>> > + return !(read_amr() & ((AMR_RD_BIT|AMR_WR_BIT) << pkey_shift));
>> > +}
>> > +
>> > +int __execute_only_pkey(struct mm_struct *mm)
>> > +{
>> > + bool need_to_set_mm_pkey = false;
>> > + int execute_only_pkey = mm->context.execute_only_pkey;
>> > + int ret;
>> > +
>> > + /* Do we need to assign a pkey for mm's execute-only maps? */
>> > + if (execute_only_pkey == -1) {
>> > + /* Go allocate one to use, which might fail */
>> > + execute_only_pkey = mm_pkey_alloc(mm);
>> > + if (execute_only_pkey < 0)
>> > + return -1;
>> > + need_to_set_mm_pkey = true;
>> > + }
>> > +
>> > + /*
>> > + * We do not want to go through the relatively costly
>> > + * dance to set AMR if we do not need to. Check it
>> > + * first and assume that if the execute-only pkey is
>> > + * readwrite-disabled than we do not have to set it
>> > + * ourselves.
>> > + */
>> > + if (!need_to_set_mm_pkey &&
>> > + !pkey_allows_readwrite(execute_only_pkey))
> ^^^^^
> Here uamor and amr is read once each.
You are right. What confused me was that the call to mm_pkey_alloc above
also reads uamor and amr (and also iamr, and writes to all of those) but
if that function is called, then need_to_set_mm_pkey is true and
pkey_allows_readwrite won't be called.
>> > + return execute_only_pkey;
>> > +
>> > + /*
>> > + * Set up AMR so that it denies access for everything
>> > + * other than execution.
>> > + */
>> > + ret = __arch_set_user_pkey_access(current, execute_only_pkey,
>> > + (PKEY_DISABLE_ACCESS | PKEY_DISABLE_WRITE));
> ^^^^^^^
> here amr and iamr are written once each if the
> the function returns successfully.
__arch_set_user_pkey_access also reads uamor for the second time in its
call to is_pkey_enabled, and reads amr for the second time as well in
its calls to init_amr. The first reads are in either
pkey_allows_readwrite or pkey_status_change (called from
__arch_activate_pkey).
If need_to_set_mm_pkey is true, then the iamr read in init_iamr is the
2nd one during __execute_only_pkey's execution. In this case the writes
to amr and iamr will be the 2nd ones as well. The first reads and writes
are in pkey_status_change.
>> > + /*
>> > + * If the AMR-set operation failed somehow, just return
>> > + * 0 and effectively disable execute-only support.
>> > + */
>> > + if (ret) {
>> > + mm_set_pkey_free(mm, execute_only_pkey);
> ^^^
> here only if __arch_set_user_pkey_access() fails
> amr and iamr and uamor will be written once each.
I assume the error case isn't perfomance sensitive and didn't account
for mm_set_pkey_free in my analysis.
>> > + return -1;
>> > + }
>> > +
>> > + /* We got one, store it and use it from here on out */
>> > + if (need_to_set_mm_pkey)
>> > + mm->context.execute_only_pkey = execute_only_pkey;
>> > + return execute_only_pkey;
>> > +}
>>
>> If you follow the code flow in __execute_only_pkey, the AMR and UAMOR
>> are read 3 times in total, and AMR is written twice. IAMR is read and
>> written twice. Since they are SPRs and access to them is slow (or isn't
>> it?), is it worth it to read them once in __execute_only_pkey and pass
>> down their values to the callees, and then write them once at the end of
>> the function?
>
> If my calculations are right:
> uamor may be read once and may be written once.
> amr may be read once and is written once.
> iamr is written once.
> So not that bad, i think.
If I'm following the code correctly:
if need_to_set_mm_pkey = true:
uamor is read twice and written once.
amr is read twice and written twice.
iamr is read twice and written twice.
if need_to_set_mm_pkey = false:
uamor is read twice.
amr is read once or twice (depending on the value of uamor) and written once.
iamr is read once and written once.
--
Thiago Jung Bauermann
IBM Linux Technology Center
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2017-07-31 16:19 UTC|newest]
Thread overview: 103+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-07-16 3:56 [RFC v6 00/62] powerpc: Memory Protection Keys Ram Pai
2017-07-16 3:56 ` [RFC v6 01/62] powerpc: Free up four 64K PTE bits in 4K backed HPTE pages Ram Pai
2017-07-20 5:51 ` Aneesh Kumar K.V
2017-07-20 22:03 ` Ram Pai
2017-07-16 3:56 ` [RFC v6 02/62] powerpc: Free up four 64K PTE bits in 64K " Ram Pai
2017-07-20 5:53 ` Aneesh Kumar K.V
2017-07-16 3:56 ` [RFC v6 03/62] powerpc: introduce pte_set_hash_slot() helper Ram Pai
2017-07-20 5:56 ` Aneesh Kumar K.V
2017-07-16 3:56 ` [RFC v6 04/62] powerpc: introduce pte_get_hash_gslot() helper Ram Pai
2017-07-20 5:57 ` Aneesh Kumar K.V
2017-07-16 3:56 ` [RFC v6 05/62] powerpc: capture the PTE format changes in the dump pte report Ram Pai
2017-07-20 5:56 ` Aneesh Kumar K.V
2017-07-16 3:56 ` [RFC v6 06/62] powerpc: use helper functions in __hash_page_64K() for 64K PTE Ram Pai
2017-07-20 5:58 ` Aneesh Kumar K.V
2017-07-16 3:56 ` [RFC v6 07/62] powerpc: use helper functions in __hash_page_huge() " Ram Pai
2017-07-20 5:58 ` Aneesh Kumar K.V
2017-07-16 3:56 ` [RFC v6 08/62] powerpc: use helper functions in __hash_page_4K() " Ram Pai
2017-07-16 3:56 ` [RFC v6 09/62] powerpc: use helper functions in __hash_page_4K() for 4K PTE Ram Pai
2017-07-16 3:56 ` [RFC v6 10/62] powerpc: use helper functions in flush_hash_page() Ram Pai
2017-07-16 3:56 ` [RFC v6 11/62] powerpc: initial pkey plumbing Ram Pai
2017-07-20 6:04 ` Aneesh Kumar K.V
2017-07-20 22:11 ` Ram Pai
2017-07-16 3:56 ` [RFC v6 12/62] mm: introduce an additional vma bit for powerpc pkey Ram Pai
2017-07-16 3:56 ` [RFC v6 13/62] powerpc: track allocation status of all pkeys Ram Pai
2017-07-27 14:01 ` Thiago Jung Bauermann
2017-07-29 22:43 ` Ram Pai
2017-07-31 18:15 ` Thiago Jung Bauermann
2017-07-16 3:56 ` [RFC v6 14/62] powerpc: helper function to read,write AMR,IAMR,UAMOR registers Ram Pai
2017-07-16 3:56 ` [RFC v6 15/62] powerpc: helper functions to initialize AMR, IAMR and UMOR registers Ram Pai
2017-07-27 20:40 ` Thiago Jung Bauermann
2017-07-30 0:38 ` Ram Pai
2017-07-16 3:56 ` [RFC v6 16/62] powerpc: cleaup AMR,iAMR when a key is allocated or freed Ram Pai
2017-07-16 3:56 ` [RFC v6 17/62] powerpc: implementation for arch_set_user_pkey_access() Ram Pai
2017-07-27 14:15 ` Thiago Jung Bauermann
2017-07-29 22:59 ` Ram Pai
2017-07-16 3:56 ` [RFC v6 18/62] powerpc: sys_pkey_alloc() and sys_pkey_free() system calls Ram Pai
2017-07-16 3:56 ` [RFC v6 19/62] powerpc: ability to create execute-disabled pkeys Ram Pai
2017-07-27 14:54 ` Thiago Jung Bauermann
2017-07-27 15:34 ` Thiago Jung Bauermann
2017-07-29 23:24 ` Ram Pai
2017-07-31 12:59 ` Michael Ellerman
2017-07-16 3:56 ` [RFC v6 20/62] powerpc: store and restore the pkey state across context switches Ram Pai
2017-07-27 17:32 ` Thiago Jung Bauermann
2017-07-29 23:31 ` Ram Pai
2017-07-31 13:00 ` Michael Ellerman
2017-07-16 3:56 ` [RFC v6 21/62] powerpc: introduce execute-only pkey Ram Pai
2017-07-28 22:17 ` Thiago Jung Bauermann
2017-07-30 0:51 ` Ram Pai
2017-07-31 16:19 ` Thiago Jung Bauermann [this message]
2017-08-01 6:46 ` Michael Ellerman
2017-08-01 16:14 ` Thiago Jung Bauermann
2017-08-02 9:40 ` Michael Ellerman
[not found] ` <20170817233555.GC5427@ram.oc3035372033.ibm.com>
2017-08-17 23:42 ` Ram Pai
2017-07-16 3:56 ` [RFC v6 22/62] powerpc: ability to associate pkey to a vma Ram Pai
2017-07-16 3:56 ` [RFC v6 23/62] powerpc: implementation for arch_override_mprotect_pkey() Ram Pai
2017-07-16 3:56 ` [RFC v6 24/62] powerpc: map vma key-protection bits to pte key bits Ram Pai
2017-07-16 3:56 ` [RFC v6 25/62] powerpc: sys_pkey_mprotect() system call Ram Pai
2017-07-16 3:56 ` [RFC v6 26/62] powerpc: Program HPTE key protection bits Ram Pai
2017-07-20 6:28 ` Aneesh Kumar K.V
2017-07-16 3:56 ` [RFC v6 27/62] powerpc: helper to validate key-access permissions of a pte Ram Pai
2017-07-20 6:42 ` Aneesh Kumar K.V
2017-07-20 22:15 ` Ram Pai
2017-07-21 6:51 ` Aneesh Kumar K.V
2017-07-21 16:42 ` Ram Pai
2017-07-28 21:00 ` Thiago Jung Bauermann
2017-07-30 0:39 ` Ram Pai
2017-07-16 3:56 ` [RFC v6 28/62] powerpc: check key protection for user page access Ram Pai
2017-07-16 3:56 ` [RFC v6 29/62] powerpc: Macro the mask used for checking DSI exception Ram Pai
2017-07-16 3:56 ` [RFC v6 30/62] powerpc: implementation for arch_vma_access_permitted() Ram Pai
2017-07-16 3:56 ` [RFC v6 31/62] powerpc: Handle exceptions caused by pkey violation Ram Pai
2017-07-16 3:56 ` [RFC v6 32/62] powerpc: capture AMR register content on " Ram Pai
2017-07-16 3:56 ` [RFC v6 33/62] powerpc: introduce get_pte_pkey() helper Ram Pai
2017-07-16 3:56 ` [RFC v6 34/62] powerpc: capture the violated protection key on fault Ram Pai
2017-07-16 3:56 ` [RFC v6 35/62] powerpc: Deliver SEGV signal on pkey violation Ram Pai
2017-08-19 19:09 ` Eric W. Biederman
2017-08-22 18:06 ` Ram Pai
2017-07-16 3:56 ` [RFC v6 36/62] mm: introduce arch_pkeys_enabled() Ram Pai
2017-07-16 3:56 ` [RFC v6 37/62] x86: implementation for arch_pkeys_enabled() Ram Pai
2017-07-16 3:56 ` [RFC v6 38/62] powerpc: " Ram Pai
2017-07-16 3:56 ` [RFC v6 39/62] mm: display pkey in smaps if arch_pkeys_enabled() is true Ram Pai
2017-07-16 3:56 ` [RFC v6 40/62] x86: delete arch_show_smap() Ram Pai
2017-07-16 3:56 ` [RFC v6 41/62] selftest/x86: Move protecton key selftest to arch neutral directory Ram Pai
2017-07-16 3:56 ` [RFC v6 42/62] selftest/vm: rename all references to pkru to a generic name Ram Pai
2017-07-16 3:56 ` [RFC v6 43/62] selftest/vm: move generic definitions to header file Ram Pai
2017-07-16 3:56 ` [RFC v6 44/62] selftest/vm: typecast the pkey register Ram Pai
2017-07-16 3:56 ` [RFC v6 45/62] selftest/vm: generics function to handle shadow key register Ram Pai
2017-07-16 3:56 ` [RFC v6 46/62] selftest/vm: fix the wrong assert in pkey_disable_set() Ram Pai
2017-07-16 3:56 ` [RFC v6 47/62] selftest/vm: fixed bugs in pkey_disable_clear() Ram Pai
2017-07-16 3:56 ` [RFC v6 48/62] selftest/vm: clear the bits in shadow reg when a pkey is freed Ram Pai
2017-07-16 3:56 ` [RFC v6 49/62] selftest/vm: fix alloc_random_pkey() to make it really random Ram Pai
2017-07-16 3:56 ` [RFC v6 50/62] selftest/vm: introduce two arch independent abstraction Ram Pai
2017-07-16 3:56 ` [RFC v6 51/62] selftest/vm: pkey register should match shadow pkey Ram Pai
2017-07-16 3:56 ` [RFC v6 52/62] selftest/vm: generic cleanup Ram Pai
2017-07-16 3:56 ` [RFC v6 53/62] selftest/vm: powerpc implementation for generic abstraction Ram Pai
2017-07-16 3:56 ` [RFC v6 54/62] selftest/vm: fix an assertion in test_pkey_alloc_exhaust() Ram Pai
2017-07-16 3:56 ` [RFC v6 55/62] selftest/vm: associate key on a mapped page and detect access violation Ram Pai
2017-07-16 3:56 ` [RFC v6 56/62] selftest/vm: detect no key violation on a freed key Ram Pai
2017-07-16 3:56 ` [RFC v6 57/62] selftest/vm: associate key on a mapped page and detect write violation Ram Pai
2017-07-16 3:57 ` [RFC v6 58/62] selftest/vm: detect no write key-violation on a freed key Ram Pai
2017-07-16 3:57 ` [RFC v6 59/62] selftest/vm: detect write violation on a mapped access-denied-key page Ram Pai
2017-07-16 3:57 ` [RFC v6 60/62] selftest/vm: sub-page allocator Ram Pai
2017-07-16 3:57 ` [RFC v6 61/62] Documentation/x86: Move protecton key documentation to arch neutral directory Ram Pai
2017-07-16 3:57 ` [RFC v6 62/62] Documentation/vm: PowerPC specific updates to memory protection keys Ram Pai
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=87efsw60kj.fsf@linux.vnet.ibm.com \
--to=bauerman@linux.vnet.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=aneesh.kumar@linux.vnet.ibm.com \
--cc=arnd@arndb.de \
--cc=corbet@lwn.net \
--cc=dave.hansen@intel.com \
--cc=khandual@linux.vnet.ibm.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=linuxram@us.ibm.com \
--cc=mhocko@kernel.org \
--cc=mingo@redhat.com \
--cc=paulus@samba.org \
--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).