All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
To: Arnd Bergmann <arnd@arndb.de>
Cc: Kees Cook <keescook@chromium.org>,
	Herbert Xu <herbert@gondor.apana.org.au>,
	Eric Biggers <ebiggers@google.com>,
	Giovanni Cabiddu <giovanni.cabiddu@intel.com>,
	"Alasdair G. Kergon" <agk@redhat.com>,
	Mike Snitzer <snitzer@redhat.com>,
	Tudor-Dan Ambarus <tudor.ambarus@microchip.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Geert Uytterhoeven <geert@linux-m68k.org>,
	Will Deacon <will.deacon@arm.com>,
	Rasmus Villemoes <linux@rasmusvillemoes.dk>,
	David Woodhouse <dwmw@amazon.co.uk>,
	Matthew Wilcox <willy@infradead.org>,
	"David S. Miller" <davem@davemloft.net>,
	"Gustavo A. R. Silva" <gustavo@embeddedor.com>,
	"open list:HARDWARE RANDOM NUMBER GENERATOR CORE"
	<linux-crypto@vger.kernel.org>,
	device-mapper deve
Subject: Re: [PATCH v8 7/9] crypto: qat: Remove VLA usage
Date: Wed, 26 Sep 2018 10:44:41 +0200	[thread overview]
Message-ID: <CAKv+Gu-dGbWsQu0NgOWVFy7Sca3yBD0pLzEMwmkG3mg9LFPpkA@mail.gmail.com> (raw)
In-Reply-To: <CAK8P3a3kiH-a+JG8FqUw7nc_hk+9z-FdsGFT57SxR83W32zKzQ@mail.gmail.com>

On Tue, 25 Sep 2018 at 18:12, Arnd Bergmann <arnd@arndb.de> wrote:
>
> On Tue, Aug 7, 2018 at 11:18 PM Kees Cook <keescook@chromium.org> wrote:
> >
> > In the quest to remove all stack VLA usage from the kernel[1], this uses
> > the new upper bound for the stack buffer. Also adds a sanity check.
> >
> > [1] https://lkml.kernel.org/r/CA+55aFzCG-zNmZwX4A2FQpadafLfEzK6CC=qPXydAacU1RqZWA@mail.gmail.com
> >
> > Signed-off-by: Kees Cook <keescook@chromium.org>
>
> After rebasing to linux-next, I now get a warning about this file:
>
> drivers/crypto/qat/qat_common/qat_algs.c: In function 'qat_alg_do_precomputes':
> drivers/crypto/qat/qat_common/qat_algs.c:257:1: error: the frame size
> of 1112 bytes is larger than 1024 bytes [-Werror=frame-larger-than=]
>
> I assume it was already possible to get into that state with the VLA,
> but it seems bad enough that I think we need to do something
> about it.
>
> The large stack variables add up to 1084 bytes, which fully explains
> how we got here:
>
>         SHASH_DESC_ON_STACK(shash, ctx->hash_tfm); /* 360 */
>         struct sha1_state sha1;  /* 92 */
>         struct sha256_state sha256; /* 104 */
>         struct sha512_state sha512; /* 208 */
>         char ipad[MAX_ALGAPI_BLOCKSIZE]; /* 160 */
>         char opad[MAX_ALGAPI_BLOCKSIZE]; /* 160 */
>
> The question is what we can do about it. One simple step I've tried
> is to move the sha1/sha256/sha512 into a union, which saves around
> 200 bytes and should bring us (slightly) below the warning
> limit, but I suspect we can do better than that. Any ideas?
>

All the processing takes place in the context of a setkey() operation,
which means only one such operation should be in flight per tfm at any
given time. So we could move all these pieces into the tfm context
struct instead. Something like the below [untested] (whitespace
mangling courtesy of Gmail)

diff --git a/drivers/crypto/qat/qat_common/qat_algs.c
b/drivers/crypto/qat/qat_common/qat_algs.c
index 1138e41d6805..c14f1906bdf4 100644
--- a/drivers/crypto/qat/qat_common/qat_algs.c
+++ b/drivers/crypto/qat/qat_common/qat_algs.c
@@ -113,6 +113,13 @@ struct qat_alg_aead_ctx {
  struct crypto_shash *hash_tfm;
  enum icp_qat_hw_auth_algo qat_hash_alg;
  struct qat_crypto_instance *inst;
+ union {
+ struct sha1_state sha1;
+ struct sha256_state sha256;
+ struct sha512_state sha512;
+ };
+ char ipad[MAX_ALGAPI_BLOCKSIZE];
+ char opad[MAX_ALGAPI_BLOCKSIZE];
 };

 struct qat_alg_ablkcipher_ctx {
@@ -148,37 +155,32 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,
    unsigned int auth_keylen)
 {
  SHASH_DESC_ON_STACK(shash, ctx->hash_tfm);
- struct sha1_state sha1;
- struct sha256_state sha256;
- struct sha512_state sha512;
  int block_size = crypto_shash_blocksize(ctx->hash_tfm);
  int digest_size = crypto_shash_digestsize(ctx->hash_tfm);
- char ipad[block_size];
- char opad[block_size];
  __be32 *hash_state_out;
  __be64 *hash512_state_out;
  int i, offset;

- memset(ipad, 0, block_size);
- memset(opad, 0, block_size);
+ memset(ctx->ipad, 0, block_size);
+ memset(ctx->opad, 0, block_size);
  shash->tfm = ctx->hash_tfm;
  shash->flags = 0x0;

  if (auth_keylen > block_size) {
  int ret = crypto_shash_digest(shash, auth_key,
-       auth_keylen, ipad);
+       auth_keylen, ctx->ipad);
  if (ret)
  return ret;

- memcpy(opad, ipad, digest_size);
+ memcpy(ctx->opad, ctx->ipad, digest_size);
  } else {
- memcpy(ipad, auth_key, auth_keylen);
- memcpy(opad, auth_key, auth_keylen);
+ memcpy(ctx->ipad, auth_key, auth_keylen);
+ memcpy(ctx->opad, auth_key, auth_keylen);
  }

  for (i = 0; i < block_size; i++) {
- char *ipad_ptr = ipad + i;
- char *opad_ptr = opad + i;
+ char *ipad_ptr = ctx->ipad + i;
+ char *opad_ptr = ctx->opad + i;
  *ipad_ptr ^= HMAC_IPAD_VALUE;
  *opad_ptr ^= HMAC_OPAD_VALUE;
  }
@@ -186,7 +188,7 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,
  if (crypto_shash_init(shash))
  return -EFAULT;

- if (crypto_shash_update(shash, ipad, block_size))
+ if (crypto_shash_update(shash, ctx->ipad, block_size))
  return -EFAULT;

  hash_state_out = (__be32 *)hash->sha.state1;
@@ -194,22 +196,22 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,

  switch (ctx->qat_hash_alg) {
  case ICP_QAT_HW_AUTH_ALGO_SHA1:
- if (crypto_shash_export(shash, &sha1))
+ if (crypto_shash_export(shash, &ctx->sha1))
  return -EFAULT;
  for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
- *hash_state_out = cpu_to_be32(*(sha1.state + i));
+ *hash_state_out = cpu_to_be32(*(ctx->sha1.state + i));
  break;
  case ICP_QAT_HW_AUTH_ALGO_SHA256:
- if (crypto_shash_export(shash, &sha256))
+ if (crypto_shash_export(shash, &ctx->sha256))
  return -EFAULT;
  for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
- *hash_state_out = cpu_to_be32(*(sha256.state + i));
+ *hash_state_out = cpu_to_be32(*(ctx->sha256.state + i));
  break;
  case ICP_QAT_HW_AUTH_ALGO_SHA512:
- if (crypto_shash_export(shash, &sha512))
+ if (crypto_shash_export(shash, &ctx->sha512))
  return -EFAULT;
  for (i = 0; i < digest_size >> 3; i++, hash512_state_out++)
- *hash512_state_out = cpu_to_be64(*(sha512.state + i));
+ *hash512_state_out = cpu_to_be64(*(ctx->sha512.state + i));
  break;
  default:
  return -EFAULT;
@@ -218,7 +220,7 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,
  if (crypto_shash_init(shash))
  return -EFAULT;

- if (crypto_shash_update(shash, opad, block_size))
+ if (crypto_shash_update(shash, ctx->opad, block_size))
  return -EFAULT;

  offset = round_up(qat_get_inter_state_size(ctx->qat_hash_alg), 8);
@@ -227,28 +229,28 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,

  switch (ctx->qat_hash_alg) {
  case ICP_QAT_HW_AUTH_ALGO_SHA1:
- if (crypto_shash_export(shash, &sha1))
+ if (crypto_shash_export(shash, &ctx->sha1))
  return -EFAULT;
  for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
- *hash_state_out = cpu_to_be32(*(sha1.state + i));
+ *hash_state_out = cpu_to_be32(*(ctx->sha1.state + i));
  break;
  case ICP_QAT_HW_AUTH_ALGO_SHA256:
- if (crypto_shash_export(shash, &sha256))
+ if (crypto_shash_export(shash, &ctx->sha256))
  return -EFAULT;
  for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
- *hash_state_out = cpu_to_be32(*(sha256.state + i));
+ *hash_state_out = cpu_to_be32(*(ctx->sha256.state + i));
  break;
  case ICP_QAT_HW_AUTH_ALGO_SHA512:
- if (crypto_shash_export(shash, &sha512))
+ if (crypto_shash_export(shash, &ctx->sha512))
  return -EFAULT;
  for (i = 0; i < digest_size >> 3; i++, hash512_state_out++)
- *hash512_state_out = cpu_to_be64(*(sha512.state + i));
+ *hash512_state_out = cpu_to_be64(*(ctx->sha512.state + i));
  break;
  default:
  return -EFAULT;
  }
- memzero_explicit(ipad, block_size);
- memzero_explicit(opad, block_size);
+ memzero_explicit(ctx->ipad, block_size);
+ memzero_explicit(ctx->opad, block_size);
  return 0;
 }

WARNING: multiple messages have this Message-ID (diff)
From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
To: Arnd Bergmann <arnd@arndb.de>
Cc: Kees Cook <keescook@chromium.org>,
	Herbert Xu <herbert@gondor.apana.org.au>,
	Eric Biggers <ebiggers@google.com>,
	Giovanni Cabiddu <giovanni.cabiddu@intel.com>,
	"Alasdair G. Kergon" <agk@redhat.com>,
	Mike Snitzer <snitzer@redhat.com>,
	Tudor-Dan Ambarus <tudor.ambarus@microchip.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Geert Uytterhoeven <geert@linux-m68k.org>,
	Will Deacon <will.deacon@arm.com>,
	Rasmus Villemoes <linux@rasmusvillemoes.dk>,
	David Woodhouse <dwmw@amazon.co.uk>,
	Matthew Wilcox <willy@infradead.org>,
	"David S. Miller" <davem@davemloft.net>,
	"Gustavo A. R. Silva" <gustavo@embeddedor.com>,
	"open list:HARDWARE RANDOM NUMBER GENERATOR CORE" 
	<linux-crypto@vger.kernel.org>,
	device-mapper development <dm-devel@redhat.com>,
	qat-linux@intel.com,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v8 7/9] crypto: qat: Remove VLA usage
Date: Wed, 26 Sep 2018 10:44:41 +0200	[thread overview]
Message-ID: <CAKv+Gu-dGbWsQu0NgOWVFy7Sca3yBD0pLzEMwmkG3mg9LFPpkA@mail.gmail.com> (raw)
In-Reply-To: <CAK8P3a3kiH-a+JG8FqUw7nc_hk+9z-FdsGFT57SxR83W32zKzQ@mail.gmail.com>

On Tue, 25 Sep 2018 at 18:12, Arnd Bergmann <arnd@arndb.de> wrote:
>
> On Tue, Aug 7, 2018 at 11:18 PM Kees Cook <keescook@chromium.org> wrote:
> >
> > In the quest to remove all stack VLA usage from the kernel[1], this uses
> > the new upper bound for the stack buffer. Also adds a sanity check.
> >
> > [1] https://lkml.kernel.org/r/CA+55aFzCG-zNmZwX4A2FQpadafLfEzK6CC=qPXydAacU1RqZWA@mail.gmail.com
> >
> > Signed-off-by: Kees Cook <keescook@chromium.org>
>
> After rebasing to linux-next, I now get a warning about this file:
>
> drivers/crypto/qat/qat_common/qat_algs.c: In function 'qat_alg_do_precomputes':
> drivers/crypto/qat/qat_common/qat_algs.c:257:1: error: the frame size
> of 1112 bytes is larger than 1024 bytes [-Werror=frame-larger-than=]
>
> I assume it was already possible to get into that state with the VLA,
> but it seems bad enough that I think we need to do something
> about it.
>
> The large stack variables add up to 1084 bytes, which fully explains
> how we got here:
>
>         SHASH_DESC_ON_STACK(shash, ctx->hash_tfm); /* 360 */
>         struct sha1_state sha1;  /* 92 */
>         struct sha256_state sha256; /* 104 */
>         struct sha512_state sha512; /* 208 */
>         char ipad[MAX_ALGAPI_BLOCKSIZE]; /* 160 */
>         char opad[MAX_ALGAPI_BLOCKSIZE]; /* 160 */
>
> The question is what we can do about it. One simple step I've tried
> is to move the sha1/sha256/sha512 into a union, which saves around
> 200 bytes and should bring us (slightly) below the warning
> limit, but I suspect we can do better than that. Any ideas?
>

All the processing takes place in the context of a setkey() operation,
which means only one such operation should be in flight per tfm at any
given time. So we could move all these pieces into the tfm context
struct instead. Something like the below [untested] (whitespace
mangling courtesy of Gmail)

diff --git a/drivers/crypto/qat/qat_common/qat_algs.c
b/drivers/crypto/qat/qat_common/qat_algs.c
index 1138e41d6805..c14f1906bdf4 100644
--- a/drivers/crypto/qat/qat_common/qat_algs.c
+++ b/drivers/crypto/qat/qat_common/qat_algs.c
@@ -113,6 +113,13 @@ struct qat_alg_aead_ctx {
  struct crypto_shash *hash_tfm;
  enum icp_qat_hw_auth_algo qat_hash_alg;
  struct qat_crypto_instance *inst;
+ union {
+ struct sha1_state sha1;
+ struct sha256_state sha256;
+ struct sha512_state sha512;
+ };
+ char ipad[MAX_ALGAPI_BLOCKSIZE];
+ char opad[MAX_ALGAPI_BLOCKSIZE];
 };

 struct qat_alg_ablkcipher_ctx {
@@ -148,37 +155,32 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,
    unsigned int auth_keylen)
 {
  SHASH_DESC_ON_STACK(shash, ctx->hash_tfm);
- struct sha1_state sha1;
- struct sha256_state sha256;
- struct sha512_state sha512;
  int block_size = crypto_shash_blocksize(ctx->hash_tfm);
  int digest_size = crypto_shash_digestsize(ctx->hash_tfm);
- char ipad[block_size];
- char opad[block_size];
  __be32 *hash_state_out;
  __be64 *hash512_state_out;
  int i, offset;

- memset(ipad, 0, block_size);
- memset(opad, 0, block_size);
+ memset(ctx->ipad, 0, block_size);
+ memset(ctx->opad, 0, block_size);
  shash->tfm = ctx->hash_tfm;
  shash->flags = 0x0;

  if (auth_keylen > block_size) {
  int ret = crypto_shash_digest(shash, auth_key,
-       auth_keylen, ipad);
+       auth_keylen, ctx->ipad);
  if (ret)
  return ret;

- memcpy(opad, ipad, digest_size);
+ memcpy(ctx->opad, ctx->ipad, digest_size);
  } else {
- memcpy(ipad, auth_key, auth_keylen);
- memcpy(opad, auth_key, auth_keylen);
+ memcpy(ctx->ipad, auth_key, auth_keylen);
+ memcpy(ctx->opad, auth_key, auth_keylen);
  }

  for (i = 0; i < block_size; i++) {
- char *ipad_ptr = ipad + i;
- char *opad_ptr = opad + i;
+ char *ipad_ptr = ctx->ipad + i;
+ char *opad_ptr = ctx->opad + i;
  *ipad_ptr ^= HMAC_IPAD_VALUE;
  *opad_ptr ^= HMAC_OPAD_VALUE;
  }
@@ -186,7 +188,7 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,
  if (crypto_shash_init(shash))
  return -EFAULT;

- if (crypto_shash_update(shash, ipad, block_size))
+ if (crypto_shash_update(shash, ctx->ipad, block_size))
  return -EFAULT;

  hash_state_out = (__be32 *)hash->sha.state1;
@@ -194,22 +196,22 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,

  switch (ctx->qat_hash_alg) {
  case ICP_QAT_HW_AUTH_ALGO_SHA1:
- if (crypto_shash_export(shash, &sha1))
+ if (crypto_shash_export(shash, &ctx->sha1))
  return -EFAULT;
  for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
- *hash_state_out = cpu_to_be32(*(sha1.state + i));
+ *hash_state_out = cpu_to_be32(*(ctx->sha1.state + i));
  break;
  case ICP_QAT_HW_AUTH_ALGO_SHA256:
- if (crypto_shash_export(shash, &sha256))
+ if (crypto_shash_export(shash, &ctx->sha256))
  return -EFAULT;
  for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
- *hash_state_out = cpu_to_be32(*(sha256.state + i));
+ *hash_state_out = cpu_to_be32(*(ctx->sha256.state + i));
  break;
  case ICP_QAT_HW_AUTH_ALGO_SHA512:
- if (crypto_shash_export(shash, &sha512))
+ if (crypto_shash_export(shash, &ctx->sha512))
  return -EFAULT;
  for (i = 0; i < digest_size >> 3; i++, hash512_state_out++)
- *hash512_state_out = cpu_to_be64(*(sha512.state + i));
+ *hash512_state_out = cpu_to_be64(*(ctx->sha512.state + i));
  break;
  default:
  return -EFAULT;
@@ -218,7 +220,7 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,
  if (crypto_shash_init(shash))
  return -EFAULT;

- if (crypto_shash_update(shash, opad, block_size))
+ if (crypto_shash_update(shash, ctx->opad, block_size))
  return -EFAULT;

  offset = round_up(qat_get_inter_state_size(ctx->qat_hash_alg), 8);
@@ -227,28 +229,28 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,

  switch (ctx->qat_hash_alg) {
  case ICP_QAT_HW_AUTH_ALGO_SHA1:
- if (crypto_shash_export(shash, &sha1))
+ if (crypto_shash_export(shash, &ctx->sha1))
  return -EFAULT;
  for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
- *hash_state_out = cpu_to_be32(*(sha1.state + i));
+ *hash_state_out = cpu_to_be32(*(ctx->sha1.state + i));
  break;
  case ICP_QAT_HW_AUTH_ALGO_SHA256:
- if (crypto_shash_export(shash, &sha256))
+ if (crypto_shash_export(shash, &ctx->sha256))
  return -EFAULT;
  for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
- *hash_state_out = cpu_to_be32(*(sha256.state + i));
+ *hash_state_out = cpu_to_be32(*(ctx->sha256.state + i));
  break;
  case ICP_QAT_HW_AUTH_ALGO_SHA512:
- if (crypto_shash_export(shash, &sha512))
+ if (crypto_shash_export(shash, &ctx->sha512))
  return -EFAULT;
  for (i = 0; i < digest_size >> 3; i++, hash512_state_out++)
- *hash512_state_out = cpu_to_be64(*(sha512.state + i));
+ *hash512_state_out = cpu_to_be64(*(ctx->sha512.state + i));
  break;
  default:
  return -EFAULT;
  }
- memzero_explicit(ipad, block_size);
- memzero_explicit(opad, block_size);
+ memzero_explicit(ctx->ipad, block_size);
+ memzero_explicit(ctx->opad, block_size);
  return 0;
 }

WARNING: multiple messages have this Message-ID (diff)
From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
To: Arnd Bergmann <arnd@arndb.de>
Cc: Kees Cook <keescook@chromium.org>,
	Herbert Xu <herbert@gondor.apana.org.au>,
	Eric Biggers <ebiggers@google.com>,
	Giovanni Cabiddu <giovanni.cabiddu@intel.com>,
	"Alasdair G. Kergon" <agk@redhat.com>,
	Mike Snitzer <snitzer@redhat.com>,
	Tudor-Dan Ambarus <tudor.ambarus@microchip.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Geert Uytterhoeven <geert@linux-m68k.org>,
	Will Deacon <will.deacon@arm.com>,
	Rasmus Villemoes <linux@rasmusvillemoes.dk>,
	David Woodhouse <dwmw@amazon.co.uk>,
	Matthew Wilcox <willy@infradead.org>,
	"David S. Miller" <davem@davemloft.net>,
	"Gustavo A. R. Silva" <gustavo@embeddedor.com>,
	"open list:HARDWARE RANDOM NUMBER GENERATOR CORE"
	<linux-crypto@vger.kernel.org>,
	device-mapper
Subject: Re: [PATCH v8 7/9] crypto: qat: Remove VLA usage
Date: Wed, 26 Sep 2018 10:44:41 +0200	[thread overview]
Message-ID: <CAKv+Gu-dGbWsQu0NgOWVFy7Sca3yBD0pLzEMwmkG3mg9LFPpkA@mail.gmail.com> (raw)
In-Reply-To: <CAK8P3a3kiH-a+JG8FqUw7nc_hk+9z-FdsGFT57SxR83W32zKzQ@mail.gmail.com>

On Tue, 25 Sep 2018 at 18:12, Arnd Bergmann <arnd@arndb.de> wrote:
>
> On Tue, Aug 7, 2018 at 11:18 PM Kees Cook <keescook@chromium.org> wrote:
> >
> > In the quest to remove all stack VLA usage from the kernel[1], this uses
> > the new upper bound for the stack buffer. Also adds a sanity check.
> >
> > [1] https://lkml.kernel.org/r/CA+55aFzCG-zNmZwX4A2FQpadafLfEzK6CC=qPXydAacU1RqZWA@mail.gmail.com
> >
> > Signed-off-by: Kees Cook <keescook@chromium.org>
>
> After rebasing to linux-next, I now get a warning about this file:
>
> drivers/crypto/qat/qat_common/qat_algs.c: In function 'qat_alg_do_precomputes':
> drivers/crypto/qat/qat_common/qat_algs.c:257:1: error: the frame size
> of 1112 bytes is larger than 1024 bytes [-Werror=frame-larger-than=]
>
> I assume it was already possible to get into that state with the VLA,
> but it seems bad enough that I think we need to do something
> about it.
>
> The large stack variables add up to 1084 bytes, which fully explains
> how we got here:
>
>         SHASH_DESC_ON_STACK(shash, ctx->hash_tfm); /* 360 */
>         struct sha1_state sha1;  /* 92 */
>         struct sha256_state sha256; /* 104 */
>         struct sha512_state sha512; /* 208 */
>         char ipad[MAX_ALGAPI_BLOCKSIZE]; /* 160 */
>         char opad[MAX_ALGAPI_BLOCKSIZE]; /* 160 */
>
> The question is what we can do about it. One simple step I've tried
> is to move the sha1/sha256/sha512 into a union, which saves around
> 200 bytes and should bring us (slightly) below the warning
> limit, but I suspect we can do better than that. Any ideas?
>

All the processing takes place in the context of a setkey() operation,
which means only one such operation should be in flight per tfm at any
given time. So we could move all these pieces into the tfm context
struct instead. Something like the below [untested] (whitespace
mangling courtesy of Gmail)

diff --git a/drivers/crypto/qat/qat_common/qat_algs.c
b/drivers/crypto/qat/qat_common/qat_algs.c
index 1138e41d6805..c14f1906bdf4 100644
--- a/drivers/crypto/qat/qat_common/qat_algs.c
+++ b/drivers/crypto/qat/qat_common/qat_algs.c
@@ -113,6 +113,13 @@ struct qat_alg_aead_ctx {
  struct crypto_shash *hash_tfm;
  enum icp_qat_hw_auth_algo qat_hash_alg;
  struct qat_crypto_instance *inst;
+ union {
+ struct sha1_state sha1;
+ struct sha256_state sha256;
+ struct sha512_state sha512;
+ };
+ char ipad[MAX_ALGAPI_BLOCKSIZE];
+ char opad[MAX_ALGAPI_BLOCKSIZE];
 };

 struct qat_alg_ablkcipher_ctx {
@@ -148,37 +155,32 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,
    unsigned int auth_keylen)
 {
  SHASH_DESC_ON_STACK(shash, ctx->hash_tfm);
- struct sha1_state sha1;
- struct sha256_state sha256;
- struct sha512_state sha512;
  int block_size = crypto_shash_blocksize(ctx->hash_tfm);
  int digest_size = crypto_shash_digestsize(ctx->hash_tfm);
- char ipad[block_size];
- char opad[block_size];
  __be32 *hash_state_out;
  __be64 *hash512_state_out;
  int i, offset;

- memset(ipad, 0, block_size);
- memset(opad, 0, block_size);
+ memset(ctx->ipad, 0, block_size);
+ memset(ctx->opad, 0, block_size);
  shash->tfm = ctx->hash_tfm;
  shash->flags = 0x0;

  if (auth_keylen > block_size) {
  int ret = crypto_shash_digest(shash, auth_key,
-       auth_keylen, ipad);
+       auth_keylen, ctx->ipad);
  if (ret)
  return ret;

- memcpy(opad, ipad, digest_size);
+ memcpy(ctx->opad, ctx->ipad, digest_size);
  } else {
- memcpy(ipad, auth_key, auth_keylen);
- memcpy(opad, auth_key, auth_keylen);
+ memcpy(ctx->ipad, auth_key, auth_keylen);
+ memcpy(ctx->opad, auth_key, auth_keylen);
  }

  for (i = 0; i < block_size; i++) {
- char *ipad_ptr = ipad + i;
- char *opad_ptr = opad + i;
+ char *ipad_ptr = ctx->ipad + i;
+ char *opad_ptr = ctx->opad + i;
  *ipad_ptr ^= HMAC_IPAD_VALUE;
  *opad_ptr ^= HMAC_OPAD_VALUE;
  }
@@ -186,7 +188,7 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,
  if (crypto_shash_init(shash))
  return -EFAULT;

- if (crypto_shash_update(shash, ipad, block_size))
+ if (crypto_shash_update(shash, ctx->ipad, block_size))
  return -EFAULT;

  hash_state_out = (__be32 *)hash->sha.state1;
@@ -194,22 +196,22 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,

  switch (ctx->qat_hash_alg) {
  case ICP_QAT_HW_AUTH_ALGO_SHA1:
- if (crypto_shash_export(shash, &sha1))
+ if (crypto_shash_export(shash, &ctx->sha1))
  return -EFAULT;
  for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
- *hash_state_out = cpu_to_be32(*(sha1.state + i));
+ *hash_state_out = cpu_to_be32(*(ctx->sha1.state + i));
  break;
  case ICP_QAT_HW_AUTH_ALGO_SHA256:
- if (crypto_shash_export(shash, &sha256))
+ if (crypto_shash_export(shash, &ctx->sha256))
  return -EFAULT;
  for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
- *hash_state_out = cpu_to_be32(*(sha256.state + i));
+ *hash_state_out = cpu_to_be32(*(ctx->sha256.state + i));
  break;
  case ICP_QAT_HW_AUTH_ALGO_SHA512:
- if (crypto_shash_export(shash, &sha512))
+ if (crypto_shash_export(shash, &ctx->sha512))
  return -EFAULT;
  for (i = 0; i < digest_size >> 3; i++, hash512_state_out++)
- *hash512_state_out = cpu_to_be64(*(sha512.state + i));
+ *hash512_state_out = cpu_to_be64(*(ctx->sha512.state + i));
  break;
  default:
  return -EFAULT;
@@ -218,7 +220,7 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,
  if (crypto_shash_init(shash))
  return -EFAULT;

- if (crypto_shash_update(shash, opad, block_size))
+ if (crypto_shash_update(shash, ctx->opad, block_size))
  return -EFAULT;

  offset = round_up(qat_get_inter_state_size(ctx->qat_hash_alg), 8);
@@ -227,28 +229,28 @@ static int qat_alg_do_precomputes(struct
icp_qat_hw_auth_algo_blk *hash,

  switch (ctx->qat_hash_alg) {
  case ICP_QAT_HW_AUTH_ALGO_SHA1:
- if (crypto_shash_export(shash, &sha1))
+ if (crypto_shash_export(shash, &ctx->sha1))
  return -EFAULT;
  for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
- *hash_state_out = cpu_to_be32(*(sha1.state + i));
+ *hash_state_out = cpu_to_be32(*(ctx->sha1.state + i));
  break;
  case ICP_QAT_HW_AUTH_ALGO_SHA256:
- if (crypto_shash_export(shash, &sha256))
+ if (crypto_shash_export(shash, &ctx->sha256))
  return -EFAULT;
  for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
- *hash_state_out = cpu_to_be32(*(sha256.state + i));
+ *hash_state_out = cpu_to_be32(*(ctx->sha256.state + i));
  break;
  case ICP_QAT_HW_AUTH_ALGO_SHA512:
- if (crypto_shash_export(shash, &sha512))
+ if (crypto_shash_export(shash, &ctx->sha512))
  return -EFAULT;
  for (i = 0; i < digest_size >> 3; i++, hash512_state_out++)
- *hash512_state_out = cpu_to_be64(*(sha512.state + i));
+ *hash512_state_out = cpu_to_be64(*(ctx->sha512.state + i));
  break;
  default:
  return -EFAULT;
  }
- memzero_explicit(ipad, block_size);
- memzero_explicit(opad, block_size);
+ memzero_explicit(ctx->ipad, block_size);
+ memzero_explicit(ctx->opad, block_size);
  return 0;
 }

  reply	other threads:[~2018-09-26  8:44 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-08-07 21:18 [PATCH v8 0/9] crypto: Remove VLA usage Kees Cook
2018-08-07 21:18 ` Kees Cook
2018-08-07 21:18 ` Kees Cook
2018-08-07 21:18 ` [PATCH v8 1/9] crypto: xcbc: " Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-08-07 21:18 ` [PATCH v8 2/9] crypto: cbc: " Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-08-07 21:18 ` [PATCH v8 3/9] crypto: ccm: " Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-08-07 21:18 ` [PATCH v8 4/9] crypto: hash: " Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-08-07 21:18 ` [PATCH v8 5/9] dm: Remove VLA usage from hashes Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-09-04  3:13   ` Herbert Xu
2018-09-13 16:41     ` Kees Cook
2018-09-13 16:41       ` Kees Cook
2018-09-13 17:54   ` Mike Snitzer
2018-09-14  6:10     ` Herbert Xu
2018-08-07 21:18 ` [PATCH v8 6/9] crypto alg: Introduce generic max blocksize and alignmask Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-08-07 21:18 ` [PATCH v8 7/9] crypto: qat: Remove VLA usage Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-09-25 16:12   ` Arnd Bergmann
2018-09-25 16:12     ` Arnd Bergmann
2018-09-26  8:44     ` Ard Biesheuvel [this message]
2018-09-26  8:44       ` Ard Biesheuvel
2018-09-26  8:44       ` Ard Biesheuvel
2018-09-26  8:53       ` Arnd Bergmann
2018-09-26  8:53         ` Arnd Bergmann
2018-09-26  8:55         ` Ard Biesheuvel
2018-09-26  8:55           ` Ard Biesheuvel
2018-09-26  8:55           ` Ard Biesheuvel
2018-08-07 21:18 ` [PATCH v8 8/9] crypto: shash: Remove VLA usage in unaligned hashing Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-08-07 21:18 ` [PATCH v8 9/9] crypto: skcipher: Remove VLA usage for SKCIPHER_REQUEST_ON_STACK Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-08-07 21:18   ` Kees Cook
2018-09-04  3:15   ` Herbert Xu
2018-09-04  5:19 ` [PATCH v8 0/9] crypto: Remove VLA usage Herbert Xu
2018-09-04  5:50   ` Kees Cook
2018-09-04  5:50     ` Kees Cook

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=CAKv+Gu-dGbWsQu0NgOWVFy7Sca3yBD0pLzEMwmkG3mg9LFPpkA@mail.gmail.com \
    --to=ard.biesheuvel@linaro.org \
    --cc=agk@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=arnd@arndb.de \
    --cc=davem@davemloft.net \
    --cc=dwmw@amazon.co.uk \
    --cc=ebiggers@google.com \
    --cc=geert@linux-m68k.org \
    --cc=giovanni.cabiddu@intel.com \
    --cc=gustavo@embeddedor.com \
    --cc=herbert@gondor.apana.org.au \
    --cc=keescook@chromium.org \
    --cc=linux-crypto@vger.kernel.org \
    --cc=linux@rasmusvillemoes.dk \
    --cc=snitzer@redhat.com \
    --cc=tglx@linutronix.de \
    --cc=tudor.ambarus@microchip.com \
    --cc=will.deacon@arm.com \
    --cc=willy@infradead.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.