From: Zeng Zhaoxiu <zhaoxiu.zeng@gmail.com> To: Sam Ravnborg <sam@ravnborg.org>, zengzhaoxiu@163.com Cc: davem@davemloft.net, wim.coekaerts@oracle.com, linux@roeck-us.net, julian.calaby@gmail.com, sparclinux@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH v2 08/30] Add sparc-specific parity functions Date: Thu, 7 Apr 2016 11:56:08 +0800 [thread overview] Message-ID: <5705DA58.3040207@gmail.com> (raw) In-Reply-To: <20160406184447.GA5963@ravnborg.org> 在 2016年04月07日 02:44, Sam Ravnborg 写道: > Hi Zeng. > >> Use runtime patching for sparc64, lifted from hweight > No errors found in patch - but a few comments. > In general patch looks good. Thanks. Sparc, powerpc, and x86 are all new to me. >> +++ b/arch/sparc/include/asm/bitops_64.h >> @@ -47,6 +47,24 @@ unsigned int __arch_hweight16(unsigned int w); >> unsigned int __arch_hweight8(unsigned int w); >> >> #include <asm-generic/bitops/const_hweight.h> >> + >> +/* >> + * parityN: returns the parity of a N-bit word, >> + * i.e. the number of 1-bits in w modulo 2. >> + */ >> + >> +static inline unsigned int __arch_parity4(unsigned int w) >> +{ >> + w &= 0xf; >> + return (0x6996 >> w) & 1; >> +} > As Josef already said - this constant should have a name. > PARITY_BIT > ? > Maybe PARITY_MAGIC? >> +++ b/arch/sparc/kernel/sparc_ksyms_64.c >> @@ -45,6 +45,12 @@ EXPORT_SYMBOL(__arch_hweight16); >> EXPORT_SYMBOL(__arch_hweight32); >> EXPORT_SYMBOL(__arch_hweight64); >> >> +/* from parity.S */ >> +EXPORT_SYMBOL(__arch_parity8); >> +EXPORT_SYMBOL(__arch_parity16); >> +EXPORT_SYMBOL(__arch_parity32); >> +EXPORT_SYMBOL(__arch_parity64); > Did you compile this? > I wonder if bitops_64.h is indirectly included. Yes. >> index 0000000..b1945e3 >> --- /dev/null >> +++ b/arch/sparc/lib/parity.S >> @@ -0,0 +1,93 @@ >> +#include <linux/linkage.h> >> + >> + .text >> + .align 32 >> + >> +ENTRY(__arch_parity8) >> + srl %o0, 4, %g1 >> + xor %o0, %g1, %o0 >> + and %o0, 0xf, %o0 >> + sethi %hi(0x6996), %g1 >> + or %g1, %lo(0x6996), %g1 >> + srl %g1, %o0, %o0 >> + retl >> + and %o0, 1, %o0 >> +ENDPROC(__arch_parity8) > I know the level of comments in hweight is equal to none. > But please do not follow this bad example. > At least for each function a one-liner of the C code. > And in the top of the file maybe two lines that the functions > are patched at run-time if the processor has popc available. > > > Sam OK. I will try, but my english is very poor! :-)
WARNING: multiple messages have this Message-ID (diff)
From: Zeng Zhaoxiu <zhaoxiu.zeng@gmail.com> To: Sam Ravnborg <sam@ravnborg.org>, zengzhaoxiu@163.com Cc: davem@davemloft.net, wim.coekaerts@oracle.com, linux@roeck-us.net, julian.calaby@gmail.com, sparclinux@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH v2 08/30] Add sparc-specific parity functions Date: Thu, 07 Apr 2016 03:56:08 +0000 [thread overview] Message-ID: <5705DA58.3040207@gmail.com> (raw) In-Reply-To: <20160406184447.GA5963@ravnborg.org> 在 2016年04月07日 02:44, Sam Ravnborg 写道: > Hi Zeng. > >> Use runtime patching for sparc64, lifted from hweight > No errors found in patch - but a few comments. > In general patch looks good. Thanks. Sparc, powerpc, and x86 are all new to me. >> +++ b/arch/sparc/include/asm/bitops_64.h >> @@ -47,6 +47,24 @@ unsigned int __arch_hweight16(unsigned int w); >> unsigned int __arch_hweight8(unsigned int w); >> >> #include <asm-generic/bitops/const_hweight.h> >> + >> +/* >> + * parityN: returns the parity of a N-bit word, >> + * i.e. the number of 1-bits in w modulo 2. >> + */ >> + >> +static inline unsigned int __arch_parity4(unsigned int w) >> +{ >> + w &= 0xf; >> + return (0x6996 >> w) & 1; >> +} > As Josef already said - this constant should have a name. > PARITY_BIT > ? > Maybe PARITY_MAGIC? >> +++ b/arch/sparc/kernel/sparc_ksyms_64.c >> @@ -45,6 +45,12 @@ EXPORT_SYMBOL(__arch_hweight16); >> EXPORT_SYMBOL(__arch_hweight32); >> EXPORT_SYMBOL(__arch_hweight64); >> >> +/* from parity.S */ >> +EXPORT_SYMBOL(__arch_parity8); >> +EXPORT_SYMBOL(__arch_parity16); >> +EXPORT_SYMBOL(__arch_parity32); >> +EXPORT_SYMBOL(__arch_parity64); > Did you compile this? > I wonder if bitops_64.h is indirectly included. Yes. >> index 0000000..b1945e3 >> --- /dev/null >> +++ b/arch/sparc/lib/parity.S >> @@ -0,0 +1,93 @@ >> +#include <linux/linkage.h> >> + >> + .text >> + .align 32 >> + >> +ENTRY(__arch_parity8) >> + srl %o0, 4, %g1 >> + xor %o0, %g1, %o0 >> + and %o0, 0xf, %o0 >> + sethi %hi(0x6996), %g1 >> + or %g1, %lo(0x6996), %g1 >> + srl %g1, %o0, %o0 >> + retl >> + and %o0, 1, %o0 >> +ENDPROC(__arch_parity8) > I know the level of comments in hweight is equal to none. > But please do not follow this bad example. > At least for each function a one-liner of the C code. > And in the top of the file maybe two lines that the functions > are patched at run-time if the processor has popc available. > > > Sam OK. I will try, but my english is very poor! :-)
next prev parent reply other threads:[~2016-04-07 3:56 UTC|newest] Thread overview: 104+ messages / expand[flat|nested] mbox.gz Atom feed top 2016-04-05 2:06 [PATCH V2 01/30] bitops: add parity functions Zeng Zhaoxiu 2016-04-05 4:23 ` [PATCH V2 02/30] Include generic parity.h in some architectures' bitops.h Zeng Zhaoxiu 2016-04-05 4:23 ` Zeng Zhaoxiu 2016-04-05 4:23 ` Zeng Zhaoxiu 2016-04-05 4:23 ` Zeng Zhaoxiu 2016-04-05 4:23 ` Zeng Zhaoxiu 2016-04-05 4:23 ` Zeng Zhaoxiu 2016-04-05 4:23 ` Zeng Zhaoxiu 2016-04-06 8:41 ` [PATCH v2 " zengzhaoxiu 2016-04-06 8:41 ` zengzhaoxiu 2016-04-06 8:41 ` zengzhaoxiu at 163.com 2016-04-06 8:41 ` zengzhaoxiu 2016-04-06 8:41 ` zengzhaoxiu 2016-04-06 8:41 ` zengzhaoxiu 2016-04-11 17:31 ` Alexey Brodkin 2016-04-11 17:31 ` Alexey Brodkin 2016-04-11 17:31 ` Alexey Brodkin 2016-04-11 17:31 ` Alexey Brodkin 2016-04-05 19:04 ` [PATCH V2 01/30] bitops: add parity functions Sam Ravnborg 2016-04-06 5:33 ` Zeng Zhaoxiu 2016-04-06 8:24 ` Sam Ravnborg 2016-04-06 8:22 ` [PATCH v2 " zengzhaoxiu 2016-04-06 8:46 ` [PATCH v2 03/30] Add alpha-specific " zengzhaoxiu 2016-04-06 8:53 ` [PATCH v2 04/30] Add blackfin-specific " zengzhaoxiu 2016-04-06 8:57 ` [PATCH v2 05/30] Add ia64-specific " zengzhaoxiu 2016-04-06 8:57 ` zengzhaoxiu 2016-04-06 8:59 ` [PATCH v2 06/30] Add mips-specific " zengzhaoxiu 2016-04-06 10:23 ` zengzhaoxiu 2016-04-06 9:03 ` [PATCH v2 07/30] Add powerpc-specific " zengzhaoxiu 2016-04-06 9:07 ` [PATCH v2 08/30] Add sparc-specific " zengzhaoxiu 2016-04-06 9:07 ` zengzhaoxiu 2016-04-06 16:37 ` Josip Rodin 2016-04-06 18:44 ` Sam Ravnborg 2016-04-06 18:44 ` Sam Ravnborg 2016-04-07 3:56 ` Zeng Zhaoxiu [this message] 2016-04-07 3:56 ` Zeng Zhaoxiu 2016-04-06 9:08 ` [PATCH v2 09/30] Add tile-specific " zengzhaoxiu 2016-04-06 13:27 ` Chris Metcalf 2016-04-07 3:55 ` Zeng Zhaoxiu 2016-04-06 9:14 ` [PATCH v2 10/30] Add x86-specific " zengzhaoxiu 2016-04-06 10:13 ` Borislav Petkov 2016-04-06 10:37 ` One Thousand Gnomes 2016-04-06 10:53 ` Borislav Petkov 2016-04-07 3:55 ` Zeng Zhaoxiu 2016-04-07 9:39 ` Borislav Petkov 2016-04-11 2:43 ` Zeng Zhaoxiu 2016-04-15 2:11 ` Borislav Petkov 2016-04-07 3:55 ` Zeng Zhaoxiu 2016-04-07 9:41 ` Borislav Petkov 2016-04-06 19:45 ` Andi Kleen 2016-04-07 3:56 ` Zeng Zhaoxiu 2016-04-07 6:31 ` Dmitry Vyukov 2016-04-07 9:43 ` Borislav Petkov 2016-05-04 18:46 ` [RFC PATCH] x86/hweight: Get rid of the special calling convention Borislav Petkov 2016-05-04 19:31 ` Brian Gerst 2016-05-04 19:33 ` H. Peter Anvin 2016-05-04 19:41 ` Borislav Petkov 2016-05-04 19:49 ` H. Peter Anvin 2016-05-04 20:22 ` Borislav Petkov 2016-05-04 20:51 ` H. Peter Anvin 2016-05-04 21:09 ` Andi Kleen 2016-05-05 13:02 ` Denys Vlasenko 2016-05-05 14:04 ` Borislav Petkov 2016-05-10 16:53 ` [PATCH -v2] " Borislav Petkov 2016-05-10 17:23 ` Peter Zijlstra 2016-05-10 19:02 ` Borislav Petkov 2016-05-10 19:03 ` H. Peter Anvin 2016-05-10 19:10 ` Borislav Petkov 2016-05-10 22:30 ` H. Peter Anvin 2016-05-11 4:11 ` Borislav Petkov 2016-05-11 11:15 ` Brian Gerst 2016-05-11 11:24 ` Peter Zijlstra 2016-05-11 12:47 ` Borislav Petkov 2016-05-12 4:54 ` H. Peter Anvin 2016-05-12 11:57 ` Borislav Petkov 2016-05-12 12:14 ` Peter Zijlstra 2016-05-12 13:09 ` Borislav Petkov 2016-05-18 10:38 ` Borislav Petkov 2016-04-07 14:10 ` [PATCH v2 10/30] Add x86-specific parity functions One Thousand Gnomes 2016-04-06 9:27 ` [PATCH v2 11/30] sunrpc: use parity8 zengzhaoxiu 2016-04-06 9:30 ` [PATCH v2 12/30] mips: use parity functions in cerr-sb1.c zengzhaoxiu 2016-04-06 9:36 ` [PATCH v2 13/30] bch: use parity32 zengzhaoxiu 2016-04-06 9:39 ` [PATCH v2 14/30] media: use parity8 in vivid-vbi-gen.c zengzhaoxiu 2016-04-06 9:41 ` [PATCH v2 15/30] media: use parity functions in saa7115 zengzhaoxiu 2016-04-06 9:43 ` [PATCH v2 16/30] input: use parity32 in grip_mp zengzhaoxiu 2016-04-06 9:44 ` [PATCH v2 17/30] input: use parity64 in sidewinder zengzhaoxiu 2016-04-06 9:45 ` [PATCH v2 18/30] input: use parity16 in ams_delta_serio zengzhaoxiu 2016-04-06 9:47 ` [PATCH v2 19/30] scsi: use parity32 in isci's phy zengzhaoxiu 2016-04-06 9:52 ` [PATCH v2 20/30] mtd: use parity16 in ssfdc zengzhaoxiu 2016-04-06 9:53 ` [PATCH v2 21/30] mtd: use parity functions in inftlcore zengzhaoxiu 2016-04-06 9:58 ` [PATCH v2 22/30] crypto: use parity functions in qat_hal zengzhaoxiu 2016-04-06 10:05 ` [PATCH v2 23/30] mtd: use parity16 in sm_ftl zengzhaoxiu 2016-04-06 10:11 ` [PATCH v2 24/30] ethernet: use parity8 in sun/niu.c zengzhaoxiu 2016-04-06 10:14 ` [PATCH v2 25/30] input: use parity8 in pcips2 zengzhaoxiu 2016-04-06 10:15 ` [PATCH v2 26/30] input: use parity8 in sa1111ps2 zengzhaoxiu 2016-04-06 10:16 ` [PATCH v2 27/30] iio: use parity32 in adxrs450 zengzhaoxiu 2016-04-10 14:37 ` Jonathan Cameron 2016-04-10 14:41 ` Lars-Peter Clausen 2016-04-10 15:13 ` Jonathan Cameron 2016-04-10 15:14 ` Jonathan Cameron 2016-04-06 10:18 ` [PATCH v2 28/30] serial: use parity32 in max3100 zengzhaoxiu 2016-04-06 10:25 ` Greg KH 2016-04-06 10:20 ` [PATCH v2 29/30] input: use parity8 in elantech zengzhaoxiu 2016-04-06 10:21 ` [PATCH v2 30/30] ethernet: use parity8 in broadcom/tg3.c zengzhaoxiu
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=5705DA58.3040207@gmail.com \ --to=zhaoxiu.zeng@gmail.com \ --cc=davem@davemloft.net \ --cc=julian.calaby@gmail.com \ --cc=linux-kernel@vger.kernel.org \ --cc=linux@roeck-us.net \ --cc=sam@ravnborg.org \ --cc=sparclinux@vger.kernel.org \ --cc=wim.coekaerts@oracle.com \ --cc=zengzhaoxiu@163.com \ /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: linkBe 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.