public inbox for devel@edk2.groups.io
 help / color / mirror / Atom feed
From: "Michael D Kinney" <michael.d.kinney@intel.com>
To: Chao Li <lichao@loongson.cn>,
	"devel@edk2.groups.io" <devel@edk2.groups.io>,
	"Kinney, Michael D" <michael.d.kinney@intel.com>
Cc: "Gao, Liming" <gaoliming@byosoft.com.cn>,
	"Liu, Zhiguang" <zhiguang.liu@intel.com>,
	Baoqi Zhang <zhangbaoqi@loongson.cn>
Subject: Re: [PATCH v2 28/34] MdePkg/BaseSynchronizationLib: LoongArch cache related code.
Date: Fri, 23 Sep 2022 15:57:04 +0000	[thread overview]
Message-ID: <CO1PR11MB492913D9F007DBA1D5EE4548D2519@CO1PR11MB4929.namprd11.prod.outlook.com> (raw)
In-Reply-To: <20220914094144.3697283-1-lichao@loongson.cn>

Comments below.

Recommend moving inline assembly to .S files too.

> -----Original Message-----
> From: Chao Li <lichao@loongson.cn>
> Sent: Wednesday, September 14, 2022 2:42 AM
> To: devel@edk2.groups.io
> Cc: Kinney, Michael D <michael.d.kinney@intel.com>; Gao, Liming <gaoliming@byosoft.com.cn>; Liu, Zhiguang <zhiguang.liu@intel.com>;
> Baoqi Zhang <zhangbaoqi@loongson.cn>
> Subject: [PATCH v2 28/34] MdePkg/BaseSynchronizationLib: LoongArch cache related code.
> 
> REF: https://bugzilla.tianocore.org/show_bug.cgi?id=4053
> 
> Support LoongArch cache related functions.
> 
> Cc: Michael D Kinney <michael.d.kinney@intel.com>
> Cc: Liming Gao <gaoliming@byosoft.com.cn>
> Cc: Zhiguang Liu <zhiguang.liu@intel.com>
> 
> Signed-off-by: Chao Li <lichao@loongson.cn>
> Co-authored-by: Baoqi Zhang <zhangbaoqi@loongson.cn>
> ---
>  .../BaseSynchronizationLib.inf                |   5 +
>  .../LoongArch64/Synchronization.c             | 246 ++++++++++++++++++
>  2 files changed, 251 insertions(+)
>  create mode 100644 MdePkg/Library/BaseSynchronizationLib/LoongArch64/Synchronization.c
> 
> diff --git a/MdePkg/Library/BaseSynchronizationLib/BaseSynchronizationLib.inf
> b/MdePkg/Library/BaseSynchronizationLib/BaseSynchronizationLib.inf
> index 02ba12961a..10021f3352 100755
> --- a/MdePkg/Library/BaseSynchronizationLib/BaseSynchronizationLib.inf
> +++ b/MdePkg/Library/BaseSynchronizationLib/BaseSynchronizationLib.inf
> @@ -4,6 +4,7 @@
>  #  Copyright (c) 2007 - 2018, Intel Corporation. All rights reserved.<BR>
> 
>  #  Portions copyright (c) 2008 - 2009, Apple Inc. All rights reserved.<BR>
> 
>  #  Copyright (c) 2020, Hewlett Packard Enterprise Development LP. All rights reserved.<BR>
> 
> +#  Copyright (c) 2022, Loongson Technology Corporation Limited. All rights reserved.<BR>
> 
>  #
> 
>  #  SPDX-License-Identifier: BSD-2-Clause-Patent
> 
>  #
> 
> @@ -82,6 +83,10 @@
>    Synchronization.c
> 
>    RiscV64/Synchronization.S
> 
> 
> 
> +[Sources.LOONGARCH64]
> 
> +  Synchronization.c
> 
> +  LoongArch64/Synchronization.c
> 
> +
> 
>  [Packages]
> 
>    MdePkg/MdePkg.dec
> 
> 
> 
> diff --git a/MdePkg/Library/BaseSynchronizationLib/LoongArch64/Synchronization.c
> b/MdePkg/Library/BaseSynchronizationLib/LoongArch64/Synchronization.c
> new file mode 100644
> index 0000000000..b7789f3212
> --- /dev/null
> +++ b/MdePkg/Library/BaseSynchronizationLib/LoongArch64/Synchronization.c
> @@ -0,0 +1,246 @@
> +/** @file
> 
> +  LoongArch synchronization functions.
> 
> +
> 
> +  Copyright (c) 2022, Loongson Technology Corporation Limited. All rights reserved.<BR>
> 
> +
> 
> +  SPDX-License-Identifier: BSD-2-Clause-Patent
> 
> +
> 
> +**/
> 
> +
> 
> +#include <Library/DebugLib.h>
> 
> +
> 
> +/**
> 
> +  Performs an atomic compare exchange operation on a 16-bit
> 
> +  unsigned integer.
> 
> +
> 
> +  Performs an atomic compare exchange operation on the 16-bit
> 
> +  unsigned integer specified by Value.  If Value is equal to
> 
> +  CompareValue, then Value is set to ExchangeValue and
> 
> +  CompareValue is returned.  If Value is not equal to
> 
> +  CompareValue, then Value is returned. The compare exchange
> 
> +  operation must be performed using MP safe mechanisms.
> 
> +
> 
> +  @param[in]  Value         A pointer to the 16-bit value for the
> 
> +                        compare exchange operation.
> 
> +  @param[in]  CompareValue  16-bit value used in compare operation.
> 
> +  @param[in]  ExchangeValue 16-bit value used in exchange operation.
> 
> +
> 
> +  @return The original *Value before exchange.
> 
> +
> 
> +**/
> 
> +UINT16
> 
> +EFIAPI
> 
> +InternalSyncCompareExchange16 (
> 
> +  IN      volatile UINT16  *Value,
> 
> +  IN      UINT16           CompareValue,
> 
> +  IN      UINT16           ExchangeValue
> 
> +  )
> 
> +{
> 
> +  UINT32           RetValue;
> 
> +  UINT32           Temp;
> 
> +  UINT32           Shift;
> 
> +  UINT64           Mask;
> 
> +  UINT64           LocalCompareValue;
> 
> +  UINT64           LocalExchangeValue;
> 
> +  volatile UINT32  *Ptr32;
> 
> +
> 
> +  /* Check that ptr is naturally aligned */
> 
> +  ASSERT (!((UINT64)Value & (sizeof (Value) - 1)));

Recommend using macros in Base.h t check alignment.
> 
> +
> 
> +  /* Mask inputs to the correct size. */
> 
> +  Mask               = (((~0UL) - (1UL << (0)) + 1) & (~0UL >> (64 - 1 - ((sizeof (UINT16) * 8) - 1))));

This statement is hard to read. Especially with ~0 and << 0.  Base.h has MAX_UINT64 which may be better than ~0.

> 
> +  LocalCompareValue  = ((UINT64)CompareValue) & Mask;
> 
> +  LocalExchangeValue = ((UINT64)ExchangeValue) & Mask;
> 
> +
> 
> +  /*
> 
> +   * Calculate a shift & mask that correspond to the value we wish to
> 
> +   * compare & exchange within the naturally aligned 4 byte integer
> 
> +   * that includes it.
> 
> +   */
> 
> +  Shift                = (UINT64)Value & 0x3;
> 
> +  Shift               *= 8; /* BITS_PER_BYTE */
> 
> +  LocalCompareValue  <<= Shift;
> 
> +  LocalExchangeValue <<= Shift;
> 
> +  Mask               <<= Shift;
> 
> +
> 
> +  /*
> 
> +   * Calculate a pointer to the naturally aligned 4 byte integer that
> 
> +   * includes our byte of interest, and load its value.
> 
> +   */
> 
> +  Ptr32 = (UINT32 *)((UINT64)Value & ~0x3);
> 
> +
> 
> +  __asm__ __volatile__ (
> 
> +    "1:               \n"
> 
> +    "ll.w  %0, %3     \n"
> 
> +    "and   %1, %0, %4 \n"
> 
> +    "bne   %1, %5, 2f \n"
> 
> +    "andn  %1, %0, %4 \n"
> 
> +    "or    %1, %1, %6 \n"
> 
> +    "sc.w  %1, %2     \n"
> 
> +    "beqz  %1, 1b     \n"
> 
> +    "b     3f         \n"
> 
> +    "2:               \n"
> 
> +    "dbar  0          \n"
> 
> +    "3:               \n"
> 
> +    : "=&r" (RetValue), "=&r" (Temp), "=" "ZC" (*Ptr32)
> 
> +    : "ZC" (*Ptr32), "Jr" (Mask), "Jr" (LocalCompareValue), "Jr" (LocalExchangeValue)
> 
> +    : "memory"
> 
> +  );

Recommend removing inline assembly

> 
> +
> 
> +  return (RetValue & Mask) >> Shift;
> 
> +}
> 
> +
> 
> +/**
> 
> +  Performs an atomic compare exchange operation on a 32-bit
> 
> +  unsigned integer.
> 
> +
> 
> +  Performs an atomic compare exchange operation on the 32-bit
> 
> +  unsigned integer specified by Value.  If Value is equal to
> 
> +  CompareValue, then Value is set to ExchangeValue and
> 
> +  CompareValue is returned.  If Value is not equal to
> 
> +  CompareValue, then Value is returned. The compare exchange
> 
> +  operation must be performed using MP safe mechanisms.
> 
> +
> 
> +  @param[in]  Value         A pointer to the 32-bit value for the
> 
> +                        compare exchange operation.
> 
> +  @param[in]  CompareValue  32-bit value used in compare operation.
> 
> +  @param[in]  ExchangeValue 32-bit value used in exchange operation.
> 
> +
> 
> +  @return The original *Value before exchange.
> 
> +
> 
> +**/
> 
> +UINT32
> 
> +EFIAPI
> 
> +InternalSyncCompareExchange32 (
> 
> +  IN      volatile UINT32  *Value,
> 
> +  IN      UINT32           CompareValue,
> 
> +  IN      UINT32           ExchangeValue
> 
> +  )
> 
> +{
> 
> +  UINT32  RetValue;
> 
> +
> 
> +  __asm__ __volatile__ (
> 
> +    "1:              \n"
> 
> +    "ll.w %0, %2     \n"
> 
> +    "bne  %0, %3, 2f \n"
> 
> +    "move %0, %4     \n"
> 
> +    "sc.w %0, %1     \n"
> 
> +    "beqz %0, 1b     \n"
> 
> +    "b    3f         \n"
> 
> +    "2:              \n"
> 
> +    "dbar 0          \n"
> 
> +    "3:              \n"
> 
> +    : "=&r" (RetValue), "=" "ZC" (*Value)
> 
> +    : "ZC" (*Value), "Jr" (CompareValue), "Jr" (ExchangeValue)
> 
> +    : "memory"
> 
> +  );

Recommend removing inline assembly

> 
> +  return RetValue;
> 
> +}
> 
> +
> 
> +/**
> 
> +  Performs an atomic compare exchange operation on a 64-bit unsigned integer.
> 
> +
> 
> +  Performs an atomic compare exchange operation on the 64-bit unsigned integer specified
> 
> +  by Value.  If Value is equal to CompareValue, then Value is set to ExchangeValue and
> 
> +  CompareValue is returned.  If Value is not equal to CompareValue, then Value is returned.
> 
> +  The compare exchange operation must be performed using MP safe mechanisms.
> 
> +
> 
> +  @param[in]  Value         A pointer to the 64-bit value for the compare exchange
> 
> +                        operation.
> 
> +  @param[in]  CompareValue  64-bit value used in compare operation.
> 
> +  @param[in]  ExchangeValue 64-bit value used in exchange operation.
> 
> +
> 
> +  @return The original *Value before exchange.
> 
> +
> 
> +**/
> 
> +UINT64
> 
> +EFIAPI
> 
> +InternalSyncCompareExchange64 (
> 
> +  IN      volatile UINT64  *Value,
> 
> +  IN      UINT64           CompareValue,
> 
> +  IN      UINT64           ExchangeValue
> 
> +  )
> 
> +{
> 
> +  UINT64  RetValue;
> 
> +
> 
> +  __asm__ __volatile__ (
> 
> +    "1:              \n"
> 
> +    "ll.d %0, %2     \n"
> 
> +    "bne  %0, %3, 2f \n"
> 
> +    "move %0, %4     \n"
> 
> +    "sc.d %0, %1     \n"
> 
> +    "beqz %0, 1b     \n"
> 
> +    "b    3f         \n"
> 
> +    "2:              \n"
> 
> +    "dbar 0          \n"
> 
> +    "3:              \n"
> 
> +    : "=&r" (RetValue), "=" "ZC" (*Value)
> 
> +    : "ZC" (*Value), "Jr" (CompareValue), "Jr" (ExchangeValue)
> 
> +    : "memory"
> 
> +  );
> 
> +  return RetValue;
> 
> +}
> 
> +
> 
> +/**
> 
> +  Performs an atomic increment of an 32-bit unsigned integer.
> 
> +
> 
> +  Performs an atomic increment of the 32-bit unsigned integer specified by
> 
> +  Value and returns the incremented value. The increment operation must be
> 
> +  performed using MP safe mechanisms. The state of the return value is not
> 
> +  guaranteed to be MP safe.
> 
> +
> 
> +  @param[in]  Value A pointer to the 32-bit value to increment.
> 
> +
> 
> +  @return The incremented value.
> 
> +
> 
> +**/
> 
> +UINT32
> 
> +EFIAPI
> 
> +InternalSyncIncrement (
> 
> +  IN      volatile UINT32  *Value
> 
> +  )
> 
> +{
> 
> +  UINT32  Temp;
> 
> +
> 
> +  Temp = *Value;
> 
> +  __asm__ __volatile__ (
> 
> +     "dbar    0          \n"
> 
> +     "amadd.w %1, %2, %0 \n"
> 
> +     : "+ZB" (*Value), "=&r" (Temp)
> 
> +     : "r" (1)
> 
> +     : "memory"
> 
> +  );
> 
> +  return *Value;
> 
> +}
> 
> +
> 
> +/**
> 
> +  Performs an atomic decrement of an 32-bit unsigned integer.
> 
> +
> 
> +  Performs an atomic decrement of the 32-bit unsigned integer specified by
> 
> +  Value and returns the decrement value. The decrement operation must be
> 
> +  performed using MP safe mechanisms. The state of the return value is not
> 
> +  guaranteed to be MP safe.
> 
> +
> 
> +  @param[in]  Value A pointer to the 32-bit value to decrement.
> 
> +
> 
> +  @return The decrement value.
> 
> +
> 
> +**/
> 
> +UINT32
> 
> +EFIAPI
> 
> +InternalSyncDecrement (
> 
> +  IN      volatile UINT32  *Value
> 
> +  )
> 
> +{
> 
> +  UINT32  Temp;
> 
> +
> 
> +  Temp = *Value;
> 
> +  __asm__ __volatile__ (
> 
> +     "dbar    0          \n"
> 
> +     "amadd.w %1, %2, %0 \n"
> 
> +     : "+ZB" (*Value), "=&r" (Temp)
> 
> +     : "r" (-1)
> 
> +     : "memory"
> 
> +  );
> 
> +  return *Value;
> 
> +}
> 
> --
> 2.27.0


  parent reply	other threads:[~2022-09-23 15:57 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-09-14  9:41 [PATCH v2 28/34] MdePkg/BaseSynchronizationLib: LoongArch cache related code Chao Li
2022-09-23  7:21 ` Chao Li
2022-09-23 15:57 ` Michael D Kinney [this message]
2022-09-24  1:57   ` Chao Li

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-list 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=CO1PR11MB492913D9F007DBA1D5EE4548D2519@CO1PR11MB4929.namprd11.prod.outlook.com \
    --to=devel@edk2.groups.io \
    /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