From patchwork Mon Dec 2 01:08:36 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Eric Biggers X-Patchwork-Id: 13889808 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from bombadil.infradead.org (bombadil.infradead.org [198.137.202.133]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 920D6D49792 for ; Mon, 2 Dec 2024 02:28:53 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20210309; h=Sender:List-Subscribe:List-Help :List-Post:List-Archive:List-Unsubscribe:List-Id:Content-Transfer-Encoding: MIME-Version:References:In-Reply-To:Message-ID:Date:Subject:Cc:To:From: Reply-To:Content-Type:Content-ID:Content-Description:Resent-Date:Resent-From: Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Owner; bh=98aAhony+oG7dxlmfuBhSzcm5NZAJPMsHXlN//Enqt0=; b=ox95jg56dOImysqZZ2UZlhMafQ V+yVstlhNenA9NA65CYcbe5KJ4cKKprKMPBUh0PFznPOdpCEJJ5W9n91kHT4jCaCB0p7+Zqftvmgl 0MYNH/CvN2rOFbiP7zPB8ZcaZC8zRY3ykoCUTx3NhF6me5p4tqeXnaftKgfc07/TGx7lg3d0noZOF W7Wl66cR7Yu+mc0VwGrklZeLhzbxBuctwMMxzx2YH5vfVHRwTjbfBml084mt2CPYYUJGCal7HTQSn hlVHOwRTH0axtyGOqaDtfmp/rnxZtkVpGoeSEzAlYa7ZMFAXbeqBB/7ptfoYw5mo8SgIyuT2Ary7G rOPP+4Dg==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.98 #2 (Red Hat Linux)) id 1tHwB2-00000004nZQ-2ZCg; Mon, 02 Dec 2024 02:28:44 +0000 Received: from nyc.source.kernel.org ([2604:1380:45d1:ec00::3]) by bombadil.infradead.org with esmtps (Exim 4.98 #2 (Red Hat Linux)) id 1tHuwS-00000004dii-2Orh; Mon, 02 Dec 2024 01:09:40 +0000 Received: from smtp.kernel.org (transwarp.subspace.kernel.org [100.75.92.58]) by nyc.source.kernel.org (Postfix) with ESMTP id 574C2A40C1E; Mon, 2 Dec 2024 01:07:43 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 10744C4CEDD; Mon, 2 Dec 2024 01:09:35 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1733101775; bh=l+CLrwqJVSvfyOKygspLh9Mjv7UjhIJ0MHuliD+HqG4=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=sT/vap6RPh46beKww4TVJmC7hJ1nlRZguIaqKyI2L2sjnF0Q3JAc3a+fcPACVF2/X iRozDD804ndJyynJIWjWRZIzKOt9C/DJBDJx47gyLFh90zPxPqnbPJuvp4mXi8JyMc BnK15oWiSL6nTQLCyZJQcQ3FV2xAd1EmuWEWf2HlJlTXcKxbYF7eaMn8um5tGMH8EZ ZgDWFM8jRXt9wVjpOLdPPWx5LIGYaF9jllwYz6HJ9TxsD81OfHbxgUGSBIybwHoCVH u+ngQdT+SVxwSV3UYqtb1szMPTdg3TkLoyq2xjTRlHV0B9k0c/4X5zXiQUI5Tl1+g8 kCprWDCmk/mKA== From: Eric Biggers To: linux-kernel@vger.kernel.org Cc: linux-arch@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-crypto@vger.kernel.org, linux-ext4@vger.kernel.org, linux-f2fs-devel@lists.sourceforge.net, linux-mips@vger.kernel.org, linux-riscv@lists.infradead.org, linux-s390@vger.kernel.org, linux-scsi@vger.kernel.org, linuxppc-dev@lists.ozlabs.org, loongarch@lists.linux.dev, sparclinux@vger.kernel.org, x86@kernel.org, Ard Biesheuvel Subject: [PATCH v4 11/19] x86/crc32: update prototype for crc_pcl() Date: Sun, 1 Dec 2024 17:08:36 -0800 Message-ID: <20241202010844.144356-12-ebiggers@kernel.org> X-Mailer: git-send-email 2.47.1 In-Reply-To: <20241202010844.144356-1-ebiggers@kernel.org> References: <20241202010844.144356-1-ebiggers@kernel.org> MIME-Version: 1.0 X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20241201_170936_790992_00BACEE3 X-CRM114-Status: GOOD ( 14.61 ) X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+linux-arm-kernel=archiver.kernel.org@lists.infradead.org From: Eric Biggers - Change the len parameter from unsigned int to size_t, so that the library function which takes a size_t can safely use this code. - Rename to crc32c_x86_3way() which is much clearer. - Move the crc parameter to the front, as this is the usual convention. Reviewed-by: Ard Biesheuvel Signed-off-by: Eric Biggers --- arch/x86/crypto/crc32c-intel_glue.c | 7 ++- arch/x86/crypto/crc32c-pcl-intel-asm_64.S | 63 ++++++++++++----------- 2 files changed, 35 insertions(+), 35 deletions(-) diff --git a/arch/x86/crypto/crc32c-intel_glue.c b/arch/x86/crypto/crc32c-intel_glue.c index 52c5d47ef5a1..603d159de400 100644 --- a/arch/x86/crypto/crc32c-intel_glue.c +++ b/arch/x86/crypto/crc32c-intel_glue.c @@ -39,12 +39,11 @@ * size is >= 512 to account * for fpu state save/restore overhead. */ #define CRC32C_PCL_BREAKEVEN 512 -asmlinkage unsigned int crc_pcl(const u8 *buffer, unsigned int len, - unsigned int crc_init); +asmlinkage u32 crc32c_x86_3way(u32 crc, const u8 *buffer, size_t len); #endif /* CONFIG_X86_64 */ static u32 crc32c_intel_le_hw_byte(u32 crc, unsigned char const *data, size_t length) { while (length--) { @@ -157,11 +156,11 @@ static int crc32c_pcl_intel_update(struct shash_desc *desc, const u8 *data, * use faster PCL version if datasize is large enough to * overcome kernel fpu state save/restore overhead */ if (len >= CRC32C_PCL_BREAKEVEN && crypto_simd_usable()) { kernel_fpu_begin(); - *crcp = crc_pcl(data, len, *crcp); + *crcp = crc32c_x86_3way(*crcp, data, len); kernel_fpu_end(); } else *crcp = crc32c_intel_le_hw(*crcp, data, len); return 0; } @@ -169,11 +168,11 @@ static int crc32c_pcl_intel_update(struct shash_desc *desc, const u8 *data, static int __crc32c_pcl_intel_finup(u32 *crcp, const u8 *data, unsigned int len, u8 *out) { if (len >= CRC32C_PCL_BREAKEVEN && crypto_simd_usable()) { kernel_fpu_begin(); - *(__le32 *)out = ~cpu_to_le32(crc_pcl(data, len, *crcp)); + *(__le32 *)out = ~cpu_to_le32(crc32c_x86_3way(*crcp, data, len)); kernel_fpu_end(); } else *(__le32 *)out = ~cpu_to_le32(crc32c_intel_le_hw(*crcp, data, len)); return 0; diff --git a/arch/x86/crypto/crc32c-pcl-intel-asm_64.S b/arch/x86/crypto/crc32c-pcl-intel-asm_64.S index 752812bc4991..9b8770503bbc 100644 --- a/arch/x86/crypto/crc32c-pcl-intel-asm_64.S +++ b/arch/x86/crypto/crc32c-pcl-intel-asm_64.S @@ -50,19 +50,20 @@ # Define threshold below which buffers are considered "small" and routed to # regular CRC code that does not interleave the CRC instructions. #define SMALL_SIZE 200 -# unsigned int crc_pcl(const u8 *buffer, unsigned int len, unsigned int crc_init); +# u32 crc32c_x86_3way(u32 crc, const u8 *buffer, size_t len); .text -SYM_FUNC_START(crc_pcl) -#define bufp %rdi -#define bufp_d %edi -#define len %esi -#define crc_init %edx -#define crc_init_q %rdx +SYM_FUNC_START(crc32c_x86_3way) +#define crc0 %edi +#define crc0_q %rdi +#define bufp %rsi +#define bufp_d %esi +#define len %rdx +#define len_dw %edx #define n_misaligned %ecx /* overlaps chunk_bytes! */ #define n_misaligned_q %rcx #define chunk_bytes %ecx /* overlaps n_misaligned! */ #define chunk_bytes_q %rcx #define crc1 %r8 @@ -83,13 +84,13 @@ SYM_FUNC_START(crc_pcl) # Process 1 <= n_misaligned <= 7 bytes individually in order to align # the remaining data to an 8-byte boundary. .Ldo_align: movq (bufp), %rax add n_misaligned_q, bufp - sub n_misaligned, len + sub n_misaligned_q, len .Lalign_loop: - crc32b %al, crc_init # compute crc32 of 1-byte + crc32b %al, crc0 # compute crc32 of 1-byte shr $8, %rax # get next byte dec n_misaligned jne .Lalign_loop .Laligned: @@ -100,11 +101,11 @@ SYM_FUNC_START(crc_pcl) cmp $128*24, len jae .Lfull_block .Lpartial_block: # Compute floor(len / 24) to get num qwords to process from each lane. - imul $2731, len, %eax # 2731 = ceil(2^16 / 24) + imul $2731, len_dw, %eax # 2731 = ceil(2^16 / 24) shr $16, %eax jmp .Lcrc_3lanes .Lfull_block: # Processing 128 qwords from each lane. @@ -123,20 +124,20 @@ SYM_FUNC_START(crc_pcl) jl .Lcrc_3lanes_4x_done # Unroll the loop by a factor of 4 to reduce the overhead of the loop # bookkeeping instructions, which can compete with crc32q for the ALUs. .Lcrc_3lanes_4x_loop: - crc32q (bufp), crc_init_q + crc32q (bufp), crc0_q crc32q (bufp,chunk_bytes_q), crc1 crc32q (bufp,chunk_bytes_q,2), crc2 - crc32q 8(bufp), crc_init_q + crc32q 8(bufp), crc0_q crc32q 8(bufp,chunk_bytes_q), crc1 crc32q 8(bufp,chunk_bytes_q,2), crc2 - crc32q 16(bufp), crc_init_q + crc32q 16(bufp), crc0_q crc32q 16(bufp,chunk_bytes_q), crc1 crc32q 16(bufp,chunk_bytes_q,2), crc2 - crc32q 24(bufp), crc_init_q + crc32q 24(bufp), crc0_q crc32q 24(bufp,chunk_bytes_q), crc1 crc32q 24(bufp,chunk_bytes_q,2), crc2 add $32, bufp sub $4, %eax jge .Lcrc_3lanes_4x_loop @@ -144,42 +145,42 @@ SYM_FUNC_START(crc_pcl) .Lcrc_3lanes_4x_done: add $4, %eax jz .Lcrc_3lanes_last_qword .Lcrc_3lanes_1x_loop: - crc32q (bufp), crc_init_q + crc32q (bufp), crc0_q crc32q (bufp,chunk_bytes_q), crc1 crc32q (bufp,chunk_bytes_q,2), crc2 add $8, bufp dec %eax jnz .Lcrc_3lanes_1x_loop .Lcrc_3lanes_last_qword: - crc32q (bufp), crc_init_q + crc32q (bufp), crc0_q crc32q (bufp,chunk_bytes_q), crc1 # SKIP crc32q (bufp,chunk_bytes_q,2), crc2 ; Don't do this one yet ################################################################ ## 4) Combine three results: ################################################################ lea (K_table-8)(%rip), %rax # first entry is for idx 1 pmovzxdq (%rax,chunk_bytes_q), %xmm0 # 2 consts: K1:K2 lea (chunk_bytes,chunk_bytes,2), %eax # chunk_bytes * 3 - sub %eax, len # len -= chunk_bytes * 3 + sub %rax, len # len -= chunk_bytes * 3 - movq crc_init_q, %xmm1 # CRC for block 1 + movq crc0_q, %xmm1 # CRC for block 1 pclmulqdq $0x00, %xmm0, %xmm1 # Multiply by K2 movq crc1, %xmm2 # CRC for block 2 pclmulqdq $0x10, %xmm0, %xmm2 # Multiply by K1 pxor %xmm2,%xmm1 movq %xmm1, %rax xor (bufp,chunk_bytes_q,2), %rax - mov crc2, crc_init_q - crc32 %rax, crc_init_q + mov crc2, crc0_q + crc32 %rax, crc0_q lea 8(bufp,chunk_bytes_q,2), bufp ################################################################ ## 5) If more blocks remain, goto (2): ################################################################ @@ -191,38 +192,38 @@ SYM_FUNC_START(crc_pcl) ####################################################################### ## 6) Process any remainder without interleaving: ####################################################################### .Lsmall: - test len, len + test len_dw, len_dw jz .Ldone - mov len, %eax + mov len_dw, %eax shr $3, %eax jz .Ldo_dword .Ldo_qwords: - crc32q (bufp), crc_init_q + crc32q (bufp), crc0_q add $8, bufp dec %eax jnz .Ldo_qwords .Ldo_dword: - test $4, len + test $4, len_dw jz .Ldo_word - crc32l (bufp), crc_init + crc32l (bufp), crc0 add $4, bufp .Ldo_word: - test $2, len + test $2, len_dw jz .Ldo_byte - crc32w (bufp), crc_init + crc32w (bufp), crc0 add $2, bufp .Ldo_byte: - test $1, len + test $1, len_dw jz .Ldone - crc32b (bufp), crc_init + crc32b (bufp), crc0 .Ldone: - mov crc_init, %eax + mov crc0, %eax RET -SYM_FUNC_END(crc_pcl) +SYM_FUNC_END(crc32c_x86_3way) .section .rodata, "a", @progbits ################################################################ ## PCLMULQDQ tables ## Table is 128 entries x 2 words (8 bytes) each