The AES code uses a 'br x7' as part of a function called by
a macro, that ends up needing a BTI_J as a target. Lets
define SYN_CODE_START_LOCAL() for this and replace the
SYM_FUNC_START_LOCAL with a SYM_FUNC_CODE_LOCAL in the AES block.
Without this kernels compiled with the aes_neon_bs will OOPS
at boot.
Bad mode in Synchronous Abort handler detected on CPU1, code 0x34000003 -- BTI
CPU: 1 PID: 265 Comm: cryptomgr_test Not tainted 5.8.11-300.fc33.aarch64 #1
pstate: 20400c05 (nzCv daif +PAN -UAO BTYPE=j-)
pc : aesbs_encrypt8+0x0/0x5f0 [aes_neon_bs]
lr : aesbs_xts_encrypt+0x48/0xe0 [aes_neon_bs]
sp : ffff80001052b730
x29: ffff80001052b730 x28: 0000000000000001
x27: ffff0001ec8f4000 x26: ffff0001ec5d27b0
x25: ffff0001ec8f4000 x24: ffff80001052bc00
x23: 0000000000000018 x22: 000000000000000a
x21: ffff0001ec5d2100 x20: ffff0001ec8f4080
x19: ffff0001ec8f4000 x18: 0000000000000004
x17: 0000000000000000 x16: ffffadb4db85d310
x15: 0000000000000200 x14: 0000000000000010
x13: 0000000000000200 x12: ffff0001ec5d2100
x11: 000000000000000a x10: 0000020000200000
x9 : ffffadb4db7e7488 x8 : 0000000000000000
x7 : ffffadb4614fb110 x6 : 0000000000000000
x5 : ffff80001052bc00 x4 : 0000000000000020
x3 : 000000000000000a x2 : ffff0001ec5d2100
x1 : ffff0001ec8f4000 x0 : ffff0001ec8f4000
Kernel panic - not syncing: bad mode
CPU: 1 PID: 265 Comm: cryptomgr_test Not tainted 5.8.11-300.fc33.aarch64 #1
Call trace:
dump_backtrace+0x0/0x1c0
show_stack+0x24/0x30
dump_stack+0xc0/0x118
panic+0x144/0x358
arm64_serror_panic+0x0/0x98
el1_sync_handler+0x60/0x110
el1_sync+0xb4/0x180
aesbs_encrypt8+0x0/0x5f0 [aes_neon_bs]
__xts_crypt+0xb0/0x2dc [aes_neon_bs]
xts_encrypt+0x28/0x3c [aes_neon_bs]
crypto_skcipher_encrypt+0x50/0x84
simd_skcipher_encrypt+0xc8/0xe0
crypto_skcipher_encrypt+0x50/0x84
test_skcipher_vec_cfg+0x224/0x5f0
test_skcipher+0xbc/0x120
alg_test_skcipher+0xa0/0x1b0
alg_test+0x3dc/0x47c
cryptomgr_test+0x38/0x60
kthread+0x11c/0x120
ret_from_fork+0x10/0x18
SMP: stopping secondary CPUs
Kernel Offset: 0x2db4cb7e0000 from 0xffff800010000000
PHYS_OFFSET: 0xffffcf2440000000
CPU features: 0x7e0152,20c02020
Memory Limit: none
---[ end Kernel panic - not syncing: bad mode ]---
Fixes: commit 0e89640b640d ("crypto: arm64 - Use modern annotations for assembly functions")
Signed-off-by: Jeremy Linton <[email protected]>
---
arch/arm64/crypto/aes-neonbs-core.S | 8 ++++----
arch/arm64/include/asm/linkage.h | 6 +++++-
2 files changed, 9 insertions(+), 5 deletions(-)
diff --git a/arch/arm64/crypto/aes-neonbs-core.S b/arch/arm64/crypto/aes-neonbs-core.S
index b357164379f6..76de87ee80e1 100644
--- a/arch/arm64/crypto/aes-neonbs-core.S
+++ b/arch/arm64/crypto/aes-neonbs-core.S
@@ -428,7 +428,7 @@ SYM_FUNC_START(aesbs_convert_key)
SYM_FUNC_END(aesbs_convert_key)
.align 4
-SYM_FUNC_START_LOCAL(aesbs_encrypt8)
+SYM_CODE_START_LOCAL(aesbs_encrypt8)
ldr q9, [bskey], #16 // round 0 key
ldr q8, M0SR
ldr q24, SR
@@ -488,10 +488,10 @@ SYM_FUNC_START_LOCAL(aesbs_encrypt8)
eor v2.16b, v2.16b, v12.16b
eor v5.16b, v5.16b, v12.16b
ret
-SYM_FUNC_END(aesbs_encrypt8)
+SYM_END(aesbs_encrypt8)
.align 4
-SYM_FUNC_START_LOCAL(aesbs_decrypt8)
+SYM_CODE_START_LOCAL(aesbs_decrypt8)
lsl x9, rounds, #7
add bskey, bskey, x9
@@ -553,7 +553,7 @@ SYM_FUNC_START_LOCAL(aesbs_decrypt8)
eor v3.16b, v3.16b, v12.16b
eor v5.16b, v5.16b, v12.16b
ret
-SYM_FUNC_END(aesbs_decrypt8)
+SYM_END(aesbs_decrypt8)
/*
* aesbs_ecb_encrypt(u8 out[], u8 const in[], u8 const rk[], int rounds,
diff --git a/arch/arm64/include/asm/linkage.h b/arch/arm64/include/asm/linkage.h
index ba89a9af820a..92c81ae8ac1e 100644
--- a/arch/arm64/include/asm/linkage.h
+++ b/arch/arm64/include/asm/linkage.h
@@ -11,7 +11,8 @@
* set the architecture version to v8.5 we use the hint instruction
* instead.
*/
-#define BTI_C hint 34 ;
+#define BTI_C hint 34 ;
+#define BTI_JC hint 38 ;
/*
* When using in-kernel BTI we need to ensure that PCS-conformant assembly
@@ -42,6 +43,9 @@
SYM_START(name, SYM_L_WEAK, SYM_A_NONE) \
BTI_C
+#define SYM_CODE_START_LOCAL(name) \
+ SYM_START(name, SYM_L_LOCAL, SYM_A_ALIGN) \
+ BTI_JC
#endif
/*
--
2.25.4
Hi,
On 10/5/20 1:54 PM, Ard Biesheuvel wrote:
> On Mon, 5 Oct 2020 at 20:18, Jeremy Linton <[email protected]> wrote:
>>
>> The AES code uses a 'br x7' as part of a function called by
>> a macro, that ends up needing a BTI_J as a target.
>
> Could we instead just drop the tail call, i.e, replace it with a ret
> and do a 'bl' after it returns? The indirect call does not really
> serve a purpose here anyway
Yes, that is an option, it adds an extra ret. Which probably doesn't
mean much in most cases. I assumed this code was optimized this way
because it mattered somewhere.
>
>
>> Lets
>> define SYN_CODE_START_LOCAL() for this and replace the
>> SYM_FUNC_START_LOCAL with a SYM_FUNC_CODE_LOCAL in the AES block.
^ That should have read
SYM_CODE_START_LOCAL.
>>
>> Without this kernels compiled with the aes_neon_bs will OOPS
>> at boot.
>>
>> Bad mode in Synchronous Abort handler detected on CPU1, code 0x34000003 -- BTI
>> CPU: 1 PID: 265 Comm: cryptomgr_test Not tainted 5.8.11-300.fc33.aarch64 #1
>> pstate: 20400c05 (nzCv daif +PAN -UAO BTYPE=j-)
>> pc : aesbs_encrypt8+0x0/0x5f0 [aes_neon_bs]
>> lr : aesbs_xts_encrypt+0x48/0xe0 [aes_neon_bs]
>> sp : ffff80001052b730
>> x29: ffff80001052b730 x28: 0000000000000001
>> x27: ffff0001ec8f4000 x26: ffff0001ec5d27b0
>> x25: ffff0001ec8f4000 x24: ffff80001052bc00
>> x23: 0000000000000018 x22: 000000000000000a
>> x21: ffff0001ec5d2100 x20: ffff0001ec8f4080
>> x19: ffff0001ec8f4000 x18: 0000000000000004
>> x17: 0000000000000000 x16: ffffadb4db85d310
>> x15: 0000000000000200 x14: 0000000000000010
>> x13: 0000000000000200 x12: ffff0001ec5d2100
>> x11: 000000000000000a x10: 0000020000200000
>> x9 : ffffadb4db7e7488 x8 : 0000000000000000
>> x7 : ffffadb4614fb110 x6 : 0000000000000000
>> x5 : ffff80001052bc00 x4 : 0000000000000020
>> x3 : 000000000000000a x2 : ffff0001ec5d2100
>> x1 : ffff0001ec8f4000 x0 : ffff0001ec8f4000
>> Kernel panic - not syncing: bad mode
>> CPU: 1 PID: 265 Comm: cryptomgr_test Not tainted 5.8.11-300.fc33.aarch64 #1
>> Call trace:
>> dump_backtrace+0x0/0x1c0
>> show_stack+0x24/0x30
>> dump_stack+0xc0/0x118
>> panic+0x144/0x358
>> arm64_serror_panic+0x0/0x98
>> el1_sync_handler+0x60/0x110
>> el1_sync+0xb4/0x180
>> aesbs_encrypt8+0x0/0x5f0 [aes_neon_bs]
>> __xts_crypt+0xb0/0x2dc [aes_neon_bs]
>> xts_encrypt+0x28/0x3c [aes_neon_bs]
>> crypto_skcipher_encrypt+0x50/0x84
>> simd_skcipher_encrypt+0xc8/0xe0
>> crypto_skcipher_encrypt+0x50/0x84
>> test_skcipher_vec_cfg+0x224/0x5f0
>> test_skcipher+0xbc/0x120
>> alg_test_skcipher+0xa0/0x1b0
>> alg_test+0x3dc/0x47c
>> cryptomgr_test+0x38/0x60
>> kthread+0x11c/0x120
>> ret_from_fork+0x10/0x18
>> SMP: stopping secondary CPUs
>> Kernel Offset: 0x2db4cb7e0000 from 0xffff800010000000
>> PHYS_OFFSET: 0xffffcf2440000000
>> CPU features: 0x7e0152,20c02020
>> Memory Limit: none
>> ---[ end Kernel panic - not syncing: bad mode ]---
>>
>> Fixes: commit 0e89640b640d ("crypto: arm64 - Use modern annotations for assembly functions")
>> Signed-off-by: Jeremy Linton <[email protected]>
>> ---
>> arch/arm64/crypto/aes-neonbs-core.S | 8 ++++----
>> arch/arm64/include/asm/linkage.h | 6 +++++-
>> 2 files changed, 9 insertions(+), 5 deletions(-)
>>
>> diff --git a/arch/arm64/crypto/aes-neonbs-core.S b/arch/arm64/crypto/aes-neonbs-core.S
>> index b357164379f6..76de87ee80e1 100644
>> --- a/arch/arm64/crypto/aes-neonbs-core.S
>> +++ b/arch/arm64/crypto/aes-neonbs-core.S
>> @@ -428,7 +428,7 @@ SYM_FUNC_START(aesbs_convert_key)
>> SYM_FUNC_END(aesbs_convert_key)
>>
>> .align 4
>> -SYM_FUNC_START_LOCAL(aesbs_encrypt8)
>> +SYM_CODE_START_LOCAL(aesbs_encrypt8)
>> ldr q9, [bskey], #16 // round 0 key
>> ldr q8, M0SR
>> ldr q24, SR
>> @@ -488,10 +488,10 @@ SYM_FUNC_START_LOCAL(aesbs_encrypt8)
>> eor v2.16b, v2.16b, v12.16b
>> eor v5.16b, v5.16b, v12.16b
>> ret
>> -SYM_FUNC_END(aesbs_encrypt8)
>> +SYM_END(aesbs_encrypt8)
>>
>> .align 4
>> -SYM_FUNC_START_LOCAL(aesbs_decrypt8)
>> +SYM_CODE_START_LOCAL(aesbs_decrypt8)
>> lsl x9, rounds, #7
>> add bskey, bskey, x9
>>
>> @@ -553,7 +553,7 @@ SYM_FUNC_START_LOCAL(aesbs_decrypt8)
>> eor v3.16b, v3.16b, v12.16b
>> eor v5.16b, v5.16b, v12.16b
>> ret
>> -SYM_FUNC_END(aesbs_decrypt8)
>> +SYM_END(aesbs_decrypt8)
>>
>> /*
>> * aesbs_ecb_encrypt(u8 out[], u8 const in[], u8 const rk[], int rounds,
>> diff --git a/arch/arm64/include/asm/linkage.h b/arch/arm64/include/asm/linkage.h
>> index ba89a9af820a..92c81ae8ac1e 100644
>> --- a/arch/arm64/include/asm/linkage.h
>> +++ b/arch/arm64/include/asm/linkage.h
>> @@ -11,7 +11,8 @@
>> * set the architecture version to v8.5 we use the hint instruction
>> * instead.
>> */
>> -#define BTI_C hint 34 ;
>> +#define BTI_C hint 34 ;
>> +#define BTI_JC hint 38 ;
>>
>> /*
>> * When using in-kernel BTI we need to ensure that PCS-conformant assembly
>> @@ -42,6 +43,9 @@
>> SYM_START(name, SYM_L_WEAK, SYM_A_NONE) \
>> BTI_C
>>
>> +#define SYM_CODE_START_LOCAL(name) \
>> + SYM_START(name, SYM_L_LOCAL, SYM_A_ALIGN) \
>> + BTI_JC
>> #endif
>>
>> /*
>> --
>> 2.25.4
>>
On Mon, Oct 05, 2020 at 01:18:04PM -0500, Jeremy Linton wrote:
> The AES code uses a 'br x7' as part of a function called by
> a macro, that ends up needing a BTI_J as a target. Lets
> define SYN_CODE_START_LOCAL() for this and replace the
> SYM_FUNC_START_LOCAL with a SYM_FUNC_CODE_LOCAL in the AES block.
Really what the subject here should say is that this code is not a
standard function and therefore should not be annotated as such - it's
wrong with or without BTI, BTI just makes it very apparent. It'd also
be better to split the change in linkage.h out into a separate patch,
that'd make things clearer for review.
> CPU: 1 PID: 265 Comm: cryptomgr_test Not tainted 5.8.11-300.fc33.aarch64 #1
> pstate: 20400c05 (nzCv daif +PAN -UAO BTYPE=j-)
> pc : aesbs_encrypt8+0x0/0x5f0 [aes_neon_bs]
> lr : aesbs_xts_encrypt+0x48/0xe0 [aes_neon_bs]
> sp : ffff80001052b730
> x29: ffff80001052b730 x28: 0000000000000001
> x27: ffff0001ec8f4000 x26: ffff0001ec5d27b0
Please think hard before including complete backtraces in upstream
reports, they are very large and contain almost no useful information
relative to their size so often obscure the relevant content in your
message. If part of the backtrace is usefully illustrative (it often is
for search engines if nothing else) then it's usually better to pull out
the relevant sections.
> -SYM_FUNC_START_LOCAL(aesbs_encrypt8)
> +SYM_CODE_START_LOCAL(aesbs_encrypt8)
> ldr q9, [bskey], #16 // round 0 key
> ldr q8, M0SR
> ldr q24, SR
> @@ -488,10 +488,10 @@ SYM_FUNC_START_LOCAL(aesbs_encrypt8)
> eor v2.16b, v2.16b, v12.16b
> eor v5.16b, v5.16b, v12.16b
> ret
> -SYM_FUNC_END(aesbs_encrypt8)
> +SYM_END(aesbs_encrypt8)
This should be SYM_CODE_END() to match the opening. However...
> * When using in-kernel BTI we need to ensure that PCS-conformant assembly
> @@ -42,6 +43,9 @@
> SYM_START(name, SYM_L_WEAK, SYM_A_NONE) \
> BTI_C
>
> +#define SYM_CODE_START_LOCAL(name) \
> + SYM_START(name, SYM_L_LOCAL, SYM_A_ALIGN) \
> + BTI_JC
...this is going to cause problems, SYM_CODE means that we should
assemble *exactly* what was written since it's some non-standard thing -
we use it for the vectors table for example. Looking at the code it's
not 100% clear that the best approach here isn't just to change the call
to a regular function call, this isn't a fast path or anything as far as
I can see so it's unclear to me why we need to tail call.
Failing that I think we need an annotation for tail called functions,
that'd need to be a new thing as I am not seeing anything appropriate in
the current generic annotations.
Hi,
On 10/5/20 2:59 PM, Mark Brown wrote:
> On Mon, Oct 05, 2020 at 01:18:04PM -0500, Jeremy Linton wrote:
>
>> The AES code uses a 'br x7' as part of a function called by
>> a macro, that ends up needing a BTI_J as a target. Lets
>> define SYN_CODE_START_LOCAL() for this and replace the
>> SYM_FUNC_START_LOCAL with a SYM_FUNC_CODE_LOCAL in the AES block.
>
> Really what the subject here should say is that this code is not a
> standard function and therefore should not be annotated as such - it's
> wrong with or without BTI, BTI just makes it very apparent. It'd also
> be better to split the change in linkage.h out into a separate patch,
> that'd make things clearer for review.
>
>> CPU: 1 PID: 265 Comm: cryptomgr_test Not tainted 5.8.11-300.fc33.aarch64 #1
>> pstate: 20400c05 (nzCv daif +PAN -UAO BTYPE=j-)
>> pc : aesbs_encrypt8+0x0/0x5f0 [aes_neon_bs]
>> lr : aesbs_xts_encrypt+0x48/0xe0 [aes_neon_bs]
>> sp : ffff80001052b730
>> x29: ffff80001052b730 x28: 0000000000000001
>> x27: ffff0001ec8f4000 x26: ffff0001ec5d27b0
>
> Please think hard before including complete backtraces in upstream
> reports, they are very large and contain almost no useful information
> relative to their size so often obscure the relevant content in your
> message. If part of the backtrace is usefully illustrative (it often is
> for search engines if nothing else) then it's usually better to pull out
> the relevant sections.
>
>> -SYM_FUNC_START_LOCAL(aesbs_encrypt8)
>> +SYM_CODE_START_LOCAL(aesbs_encrypt8)
>> ldr q9, [bskey], #16 // round 0 key
>> ldr q8, M0SR
>> ldr q24, SR
>> @@ -488,10 +488,10 @@ SYM_FUNC_START_LOCAL(aesbs_encrypt8)
>> eor v2.16b, v2.16b, v12.16b
>> eor v5.16b, v5.16b, v12.16b
>> ret
>> -SYM_FUNC_END(aesbs_encrypt8)
>> +SYM_END(aesbs_encrypt8)
>
> This should be SYM_CODE_END() to match the opening. However...
>
>> * When using in-kernel BTI we need to ensure that PCS-conformant assembly
>> @@ -42,6 +43,9 @@
>> SYM_START(name, SYM_L_WEAK, SYM_A_NONE) \
>> BTI_C
>>
>> +#define SYM_CODE_START_LOCAL(name) \
>> + SYM_START(name, SYM_L_LOCAL, SYM_A_ALIGN) \
>> + BTI_JC
>
> ...this is going to cause problems, SYM_CODE means that we should
> assemble *exactly* what was written since it's some non-standard thing -
> we use it for the vectors table for example. Looking at the code it's
> not 100% clear that the best approach here isn't just to change the call
> to a regular function call, this isn't a fast path or anything as far as
> I can see so it's unclear to me why we need to tail call.
Well for some workloads its could be AFAIK. OTOH, Ard mentioned dumping
the tail call too, and I think that is pretty reasonable. So it looks
like that is a better plan since it also avoids all this SYM_ flailing.
>
> Failing that I think we need an annotation for tail called functions,
> that'd need to be a new thing as I am not seeing anything appropriate in
> the current generic annotations.
>
On Mon, Oct 05, 2020 at 03:10:42PM -0500, Jeremy Linton wrote:
> On 10/5/20 2:59 PM, Mark Brown wrote:
> > ...this is going to cause problems, SYM_CODE means that we should
> > assemble *exactly* what was written since it's some non-standard thing -
> > we use it for the vectors table for example. Looking at the code it's
> > not 100% clear that the best approach here isn't just to change the call
> > to a regular function call, this isn't a fast path or anything as far as
> > I can see so it's unclear to me why we need to tail call.
> Well for some workloads its could be AFAIK. OTOH, Ard mentioned dumping the
> tail call too, and I think that is pretty reasonable. So it looks like that
> is a better plan since it also avoids all this SYM_ flailing.
Yeah, I think that's the easiest thing all round.
Hi Jeremy,
Thank you for the patch! Yet something to improve:
[auto build test ERROR on arm64/for-next/core]
[also build test ERROR on soc/for-next arm/for-next kvmarm/next v5.9-rc8 next-20201002]
[cannot apply to xlnx/master]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]
url: https://github.com/0day-ci/linux/commits/Jeremy-Linton/arm64-bti-fix-BTI-to-handle-local-indirect-branches/20201006-021958
base: https://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux.git for-next/core
config: arm64-allyesconfig (attached as .config)
compiler: aarch64-linux-gcc (GCC) 9.3.0
reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# https://github.com/0day-ci/linux/commit/37211f8bd05c9ebcede89bb5c371e34920355d4f
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Jeremy-Linton/arm64-bti-fix-BTI-to-handle-local-indirect-branches/20201006-021958
git checkout 37211f8bd05c9ebcede89bb5c371e34920355d4f
# save the attached .config to linux build tree
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-9.3.0 make.cross ARCH=arm64
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <[email protected]>
All errors (new ones prefixed by >>):
>> arch/arm64/crypto/aes-neonbs-core.S:491:23: error: macro "SYM_END" requires 2 arguments, but only 1 given
491 | SYM_END(aesbs_encrypt8)
| ^
In file included from arch/arm64/crypto/aes-neonbs-core.S:17:
include/linux/linkage.h:166: note: macro "SYM_END" defined here
166 | #define SYM_END(name, sym_type) \
|
arch/arm64/crypto/aes-neonbs-core.S:556:23: error: macro "SYM_END" requires 2 arguments, but only 1 given
556 | SYM_END(aesbs_decrypt8)
| ^
In file included from arch/arm64/crypto/aes-neonbs-core.S:17:
include/linux/linkage.h:166: note: macro "SYM_END" defined here
166 | #define SYM_END(name, sym_type) \
|
vim +/SYM_END +491 arch/arm64/crypto/aes-neonbs-core.S
429
430 .align 4
431 SYM_CODE_START_LOCAL(aesbs_encrypt8)
432 ldr q9, [bskey], #16 // round 0 key
433 ldr q8, M0SR
434 ldr q24, SR
435
436 eor v10.16b, v0.16b, v9.16b // xor with round0 key
437 eor v11.16b, v1.16b, v9.16b
438 tbl v0.16b, {v10.16b}, v8.16b
439 eor v12.16b, v2.16b, v9.16b
440 tbl v1.16b, {v11.16b}, v8.16b
441 eor v13.16b, v3.16b, v9.16b
442 tbl v2.16b, {v12.16b}, v8.16b
443 eor v14.16b, v4.16b, v9.16b
444 tbl v3.16b, {v13.16b}, v8.16b
445 eor v15.16b, v5.16b, v9.16b
446 tbl v4.16b, {v14.16b}, v8.16b
447 eor v10.16b, v6.16b, v9.16b
448 tbl v5.16b, {v15.16b}, v8.16b
449 eor v11.16b, v7.16b, v9.16b
450 tbl v6.16b, {v10.16b}, v8.16b
451 tbl v7.16b, {v11.16b}, v8.16b
452
453 bitslice v0, v1, v2, v3, v4, v5, v6, v7, v8, v9, v10, v11
454
455 sub rounds, rounds, #1
456 b .Lenc_sbox
457
458 .Lenc_loop:
459 shift_rows v0, v1, v2, v3, v4, v5, v6, v7, v24
460 .Lenc_sbox:
461 sbox v0, v1, v2, v3, v4, v5, v6, v7, v8, v9, v10, v11, v12, \
462 v13, v14, v15
463 subs rounds, rounds, #1
464 b.cc .Lenc_done
465
466 enc_next_rk
467
468 mix_cols v0, v1, v4, v6, v3, v7, v2, v5, v8, v9, v10, v11, v12, \
469 v13, v14, v15
470
471 add_round_key v0, v1, v2, v3, v4, v5, v6, v7
472
473 b.ne .Lenc_loop
474 ldr q24, SRM0
475 b .Lenc_loop
476
477 .Lenc_done:
478 ldr q12, [bskey] // last round key
479
480 bitslice v0, v1, v4, v6, v3, v7, v2, v5, v8, v9, v10, v11
481
482 eor v0.16b, v0.16b, v12.16b
483 eor v1.16b, v1.16b, v12.16b
484 eor v4.16b, v4.16b, v12.16b
485 eor v6.16b, v6.16b, v12.16b
486 eor v3.16b, v3.16b, v12.16b
487 eor v7.16b, v7.16b, v12.16b
488 eor v2.16b, v2.16b, v12.16b
489 eor v5.16b, v5.16b, v12.16b
490 ret
> 491 SYM_END(aesbs_encrypt8)
492
493 .align 4
494 SYM_CODE_START_LOCAL(aesbs_decrypt8)
495 lsl x9, rounds, #7
496 add bskey, bskey, x9
497
498 ldr q9, [bskey, #-112]! // round 0 key
499 ldr q8, M0ISR
500 ldr q24, ISR
501
502 eor v10.16b, v0.16b, v9.16b // xor with round0 key
503 eor v11.16b, v1.16b, v9.16b
504 tbl v0.16b, {v10.16b}, v8.16b
505 eor v12.16b, v2.16b, v9.16b
506 tbl v1.16b, {v11.16b}, v8.16b
507 eor v13.16b, v3.16b, v9.16b
508 tbl v2.16b, {v12.16b}, v8.16b
509 eor v14.16b, v4.16b, v9.16b
510 tbl v3.16b, {v13.16b}, v8.16b
511 eor v15.16b, v5.16b, v9.16b
512 tbl v4.16b, {v14.16b}, v8.16b
513 eor v10.16b, v6.16b, v9.16b
514 tbl v5.16b, {v15.16b}, v8.16b
515 eor v11.16b, v7.16b, v9.16b
516 tbl v6.16b, {v10.16b}, v8.16b
517 tbl v7.16b, {v11.16b}, v8.16b
518
519 bitslice v0, v1, v2, v3, v4, v5, v6, v7, v8, v9, v10, v11
520
521 sub rounds, rounds, #1
522 b .Ldec_sbox
523
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/[email protected]
On Mon, Oct 05, 2020 at 02:24:47PM -0500, Jeremy Linton wrote:
> Hi,
>
> On 10/5/20 1:54 PM, Ard Biesheuvel wrote:
> >On Mon, 5 Oct 2020 at 20:18, Jeremy Linton <[email protected]> wrote:
> >>
> >>The AES code uses a 'br x7' as part of a function called by
> >>a macro, that ends up needing a BTI_J as a target.
> >
> >Could we instead just drop the tail call, i.e, replace it with a ret
> >and do a 'bl' after it returns? The indirect call does not really
> >serve a purpose here anyway
>
> Yes, that is an option, it adds an extra ret. Which probably doesn't mean
> much in most cases. I assumed this code was optimized this way because it
> mattered somewhere.
Since this really does seem to be a tail-call and since x16 and x17
appear to be otherwise unused here, can we not just use x16 or x17
instead of x7?
This relies on there being no other calls to veneered functions in the
mix, but this code is all in a single section so that shouldn't be a
concern.
Due to the magic status of x16 and x17 in br instructions, the resulting
jump should be compatible with BTI c. I think this matches how the
compiler should typically compile tail-calls.
Cheers
---Dave
On Tue, 6 Oct 2020 at 11:38, Dave Martin <[email protected]> wrote:
>
> On Mon, Oct 05, 2020 at 02:24:47PM -0500, Jeremy Linton wrote:
> > Hi,
> >
> > On 10/5/20 1:54 PM, Ard Biesheuvel wrote:
> > >On Mon, 5 Oct 2020 at 20:18, Jeremy Linton <[email protected]> wrote:
> > >>
> > >>The AES code uses a 'br x7' as part of a function called by
> > >>a macro, that ends up needing a BTI_J as a target.
> > >
> > >Could we instead just drop the tail call, i.e, replace it with a ret
> > >and do a 'bl' after it returns? The indirect call does not really
> > >serve a purpose here anyway
> >
> > Yes, that is an option, it adds an extra ret. Which probably doesn't mean
> > much in most cases. I assumed this code was optimized this way because it
> > mattered somewhere.
>
> Since this really does seem to be a tail-call and since x16 and x17
> appear to be otherwise unused here, can we not just use x16 or x17
> instead of x7?
>
> This relies on there being no other calls to veneered functions in the
> mix, but this code is all in a single section so that shouldn't be a
> concern.
>
> Due to the magic status of x16 and x17 in br instructions, the resulting
> jump should be compatible with BTI c. I think this matches how the
> compiler should typically compile tail-calls.
>
Ah, excellent point. That is definitely the cleanest fix.