2019-05-29 07:17:58 -07:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-19 21:58:20 +10:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
crypto: qce - use AES fallback for small requests
Process small blocks using the fallback cipher, as a workaround for an
observed failure (DMA-related, apparently) when computing the GCM ghash
key. This brings a speed gain as well, since it avoids the latency of
using the hardware engine to process small blocks.
Using software for all 16-byte requests would be enough to make GCM
work, but to increase performance, a larger threshold would be better.
Measuring the performance of supported ciphers with openssl speed,
software matches hardware at around 768-1024 bytes.
Considering the 256-bit ciphers, software is 2-3 times faster than qce
at 256-bytes, 30% faster at 512, and about even at 768-bytes. With
128-bit keys, the break-even point would be around 1024-bytes.
This adds the 'aes_sw_max_len' parameter, to set the largest request
length processed by the software fallback. Its default is being set to
512 bytes, a little lower than the break-even point, to balance the cost
in CPU usage.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2020-02-07 12:02:26 -03:00
|
|
|
|
2026-02-23 16:57:55 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2021-02-11 15:01:22 -05:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2019-08-15 12:01:01 +03:00
|
|
|
|
2016-06-29 18:04:04 +08:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
crypto: qce - use AES fallback for small requests
Process small blocks using the fallback cipher, as a workaround for an
observed failure (DMA-related, apparently) when computing the GCM ghash
key. This brings a speed gain as well, since it avoids the latency of
using the hardware engine to process small blocks.
Using software for all 16-byte requests would be enough to make GCM
work, but to increase performance, a larger threshold would be better.
Measuring the performance of supported ciphers with openssl speed,
software matches hardware at around 768-1024 bytes.
Considering the 256-bit ciphers, software is 2-3 times faster than qce
at 256-bytes, 30% faster at 512, and about even at 768-bytes. With
128-bit keys, the break-even point would be around 1024-bytes.
This adds the 'aes_sw_max_len' parameter, to set the largest request
length processed by the software fallback. Its default is being set to
512 bytes, a little lower than the break-even point, to balance the cost
in CPU usage.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2020-02-07 12:02:26 -03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-04 15:24:55 -03:00
|
|
|
|
crypto: qce - use AES fallback for small requests
Process small blocks using the fallback cipher, as a workaround for an
observed failure (DMA-related, apparently) when computing the GCM ghash
key. This brings a speed gain as well, since it avoids the latency of
using the hardware engine to process small blocks.
Using software for all 16-byte requests would be enough to make GCM
work, but to increase performance, a larger threshold would be better.
Measuring the performance of supported ciphers with openssl speed,
software matches hardware at around 768-1024 bytes.
Considering the 256-bit ciphers, software is 2-3 times faster than qce
at 256-bytes, 30% faster at 512, and about even at 768-bytes. With
128-bit keys, the break-even point would be around 1024-bytes.
This adds the 'aes_sw_max_len' parameter, to set the largest request
length processed by the software fallback. Its default is being set to
512 bytes, a little lower than the break-even point, to balance the cost
in CPU usage.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2020-02-07 12:02:26 -03:00
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2019-12-20 16:02:16 -03:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-10-02 08:01:02 +02:00
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2019-12-20 16:02:16 -03:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-20 22:20:23 -04:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2015-10-02 08:01:02 +02:00
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2015-10-02 08:01:02 +02:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2015-11-04 21:13:36 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-02-07 12:02:25 -03:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-02-07 12:02:25 -03:00
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-20 22:20:23 -04:00
|
|
|
|
2022-08-19 08:07:51 +02:00
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2021-06-02 11:36:45 +00:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
2021-05-20 22:20:23 -04:00
|
|
|
|
2022-08-19 08:07:51 +02:00
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2021-06-02 11:36:45 +00:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-20 22:20:23 -04:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
2021-05-20 22:20:23 -04:00
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-02-11 15:01:28 -05:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-10-02 08:01:02 +02:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2015-10-02 08:01:02 +02:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2019-12-20 16:02:14 -03:00
|
|
|
|
2021-02-11 15:01:20 -05:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-02-11 15:01:20 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-08-15 12:01:01 +03:00
|
|
|
|
|
|
|
|
|
crypto: qce - use AES fallback for small requests
Process small blocks using the fallback cipher, as a workaround for an
observed failure (DMA-related, apparently) when computing the GCM ghash
key. This brings a speed gain as well, since it avoids the latency of
using the hardware engine to process small blocks.
Using software for all 16-byte requests would be enough to make GCM
work, but to increase performance, a larger threshold would be better.
Measuring the performance of supported ciphers with openssl speed,
software matches hardware at around 768-1024 bytes.
Considering the 256-bit ciphers, software is 2-3 times faster than qce
at 256-bytes, 30% faster at 512, and about even at 768-bytes. With
128-bit keys, the break-even point would be around 1024-bytes.
This adds the 'aes_sw_max_len' parameter, to set the largest request
length processed by the software fallback. Its default is being set to
512 bytes, a little lower than the break-even point, to balance the cost
in CPU usage.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2020-02-07 12:02:26 -03:00
|
|
|
|
2019-08-15 12:01:01 +03:00
|
|
|
|
2021-02-11 15:01:20 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
crypto: qce - permit asynchronous skcipher as fallback
Even though the qce driver implements asynchronous versions of ecb(aes),
cbc(aes)and xts(aes), the fallbacks it allocates are required to be
synchronous. Given that SIMD based software implementations are usually
asynchronous as well, even though they rarely complete asynchronously
(this typically only happens in cases where the request was made from
softirq context, while SIMD was already in use in the task context that
it interrupted), these implementations are disregarded, and either the
generic C version or another table based version implemented in assembler
is selected instead.
Since falling back to synchronous AES is not only a performance issue, but
potentially a security issue as well (due to the fact that table based AES
is not time invariant), let's fix this, by allocating an ordinary skcipher
as the fallback, and invoke it with the completion routine that was given
to the outer request.
While at it, remove the pointless memset() from qce_skcipher_init(), and
remove the call to it qce_skcipher_init_fallback().
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2020-07-07 09:32:01 +03:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-08-15 12:01:01 +03:00
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2019-08-15 12:01:01 +03:00
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2019-08-15 12:01:01 +03:00
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2019-08-15 12:01:01 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2019-04-11 16:51:16 +08:00
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2021-02-11 15:01:21 -05:00
|
|
|
|
2019-04-11 16:51:16 +08:00
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2019-08-15 12:01:01 +03:00
|
|
|
|
2019-04-11 16:51:16 +08:00
|
|
|
|
|
|
|
|
|
2021-02-11 15:01:21 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-04-11 16:51:16 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2021-02-11 15:01:23 -05:00
|
|
|
|
2019-12-20 16:02:14 -03:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-12-20 16:02:14 -03:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2021-02-11 15:01:22 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-02-11 15:01:23 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-02-11 15:01:25 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-02-07 12:02:27 -03:00
|
|
|
|
crypto: qce - use AES fallback for small requests
Process small blocks using the fallback cipher, as a workaround for an
observed failure (DMA-related, apparently) when computing the GCM ghash
key. This brings a speed gain as well, since it avoids the latency of
using the hardware engine to process small blocks.
Using software for all 16-byte requests would be enough to make GCM
work, but to increase performance, a larger threshold would be better.
Measuring the performance of supported ciphers with openssl speed,
software matches hardware at around 768-1024 bytes.
Considering the 256-bit ciphers, software is 2-3 times faster than qce
at 256-bytes, 30% faster at 512, and about even at 768-bytes. With
128-bit keys, the break-even point would be around 1024-bytes.
This adds the 'aes_sw_max_len' parameter, to set the largest request
length processed by the software fallback. Its default is being set to
512 bytes, a little lower than the break-even point, to balance the cost
in CPU usage.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2020-02-07 12:02:26 -03:00
|
|
|
|
2021-02-11 15:01:25 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
crypto: qce - permit asynchronous skcipher as fallback
Even though the qce driver implements asynchronous versions of ecb(aes),
cbc(aes)and xts(aes), the fallbacks it allocates are required to be
synchronous. Given that SIMD based software implementations are usually
asynchronous as well, even though they rarely complete asynchronously
(this typically only happens in cases where the request was made from
softirq context, while SIMD was already in use in the task context that
it interrupted), these implementations are disregarded, and either the
generic C version or another table based version implemented in assembler
is selected instead.
Since falling back to synchronous AES is not only a performance issue, but
potentially a security issue as well (due to the fact that table based AES
is not time invariant), let's fix this, by allocating an ordinary skcipher
as the fallback, and invoke it with the completion routine that was given
to the outer request.
While at it, remove the pointless memset() from qce_skcipher_init(), and
remove the call to it qce_skcipher_init_fallback().
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2020-07-07 09:32:01 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
crypto: qce - permit asynchronous skcipher as fallback
Even though the qce driver implements asynchronous versions of ecb(aes),
cbc(aes)and xts(aes), the fallbacks it allocates are required to be
synchronous. Given that SIMD based software implementations are usually
asynchronous as well, even though they rarely complete asynchronously
(this typically only happens in cases where the request was made from
softirq context, while SIMD was already in use in the task context that
it interrupted), these implementations are disregarded, and either the
generic C version or another table based version implemented in assembler
is selected instead.
Since falling back to synchronous AES is not only a performance issue, but
potentially a security issue as well (due to the fact that table based AES
is not time invariant), let's fix this, by allocating an ordinary skcipher
as the fallback, and invoke it with the completion routine that was given
to the outer request.
While at it, remove the pointless memset() from qce_skcipher_init(), and
remove the call to it qce_skcipher_init_fallback().
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2020-07-07 09:32:01 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-12-20 16:02:17 -03:00
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2019-12-20 16:02:17 -03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
crypto: qce - permit asynchronous skcipher as fallback
Even though the qce driver implements asynchronous versions of ecb(aes),
cbc(aes)and xts(aes), the fallbacks it allocates are required to be
synchronous. Given that SIMD based software implementations are usually
asynchronous as well, even though they rarely complete asynchronously
(this typically only happens in cases where the request was made from
softirq context, while SIMD was already in use in the task context that
it interrupted), these implementations are disregarded, and either the
generic C version or another table based version implemented in assembler
is selected instead.
Since falling back to synchronous AES is not only a performance issue, but
potentially a security issue as well (due to the fact that table based AES
is not time invariant), let's fix this, by allocating an ordinary skcipher
as the fallback, and invoke it with the completion routine that was given
to the outer request.
While at it, remove the pointless memset() from qce_skcipher_init(), and
remove the call to it qce_skcipher_init_fallback().
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2020-07-07 09:32:01 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
crypto: qce - permit asynchronous skcipher as fallback
Even though the qce driver implements asynchronous versions of ecb(aes),
cbc(aes)and xts(aes), the fallbacks it allocates are required to be
synchronous. Given that SIMD based software implementations are usually
asynchronous as well, even though they rarely complete asynchronously
(this typically only happens in cases where the request was made from
softirq context, while SIMD was already in use in the task context that
it interrupted), these implementations are disregarded, and either the
generic C version or another table based version implemented in assembler
is selected instead.
Since falling back to synchronous AES is not only a performance issue, but
potentially a security issue as well (due to the fact that table based AES
is not time invariant), let's fix this, by allocating an ordinary skcipher
as the fallback, and invoke it with the completion routine that was given
to the outer request.
While at it, remove the pointless memset() from qce_skcipher_init(), and
remove the call to it qce_skcipher_init_fallback().
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2020-07-07 09:32:01 +03:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-12-20 16:02:13 -03:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-02-11 15:01:24 -05:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-12-20 16:02:13 -03:00
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-12-20 16:02:14 -03:00
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
2026-02-21 16:37:42 -08:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2026-02-23 16:57:55 +01:00
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2019-12-20 16:02:13 -03:00
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2024-12-03 10:05:53 -08:00
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2020-07-09 23:20:41 -07:00
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-12-20 16:02:17 -03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2021-11-04 06:46:42 -07:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-09 18:09:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-06-25 19:28:57 +03:00
|
|
|
|