<?xml version="1.0"?>
<?xml-stylesheet type="text/xsl" href="/rss.xsl.xml"?>
<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/">
<channel>
    <title>Changes in Makefile</title>
    <description></description>
    <language>en</language>
    <copyright>Copyright 2015</copyright>
    <generator>Java</generator><item>
        <title>72f51a4f - loongarch/crc32: expose CRC32 functions through lib</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/arch/loongarch/lib/Makefile#72f51a4f</link>
        <description>loongarch/crc32: expose CRC32 functions through libMove the loongarch CRC32 assembly code into the lib directory and wireit up to the library interface.  This allows it to be used without goingthrough the crypto API.  It remains usable via the crypto API too viathe shash algorithms that use the library interface.  Thus all thearch-specific &quot;shash&quot; code becomes unnecessary and is removed.Note: to see the diff from arch/loongarch/crypto/crc32-loongarch.c toarch/loongarch/lib/crc32-loongarch.c, view this commit with&apos;git show -M10&apos;.Reviewed-by: Ard Biesheuvel &lt;ardb@kernel.org&gt;Reviewed-by: WangYuli &lt;wangyuli@uniontech.com&gt;Link: https://lore.kernel.org/r/20241202010844.144356-7-ebiggers@kernel.orgSigned-off-by: Eric Biggers &lt;ebiggers@google.com&gt;

            List of files:
            /linux-6.15/arch/loongarch/lib/Makefile</description>
        <pubDate>Mon, 02 Dec 2024 01:08:31 +0000</pubDate>
        <dc:creator>Eric Biggers &lt;ebiggers@google.com&gt;</dc:creator>
    </item>
<item>
        <title>5125d033 - LoongArch: Select ARCH_SUPPORTS_INT128 if CC_HAS_INT128</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/arch/loongarch/lib/Makefile#5125d033</link>
        <description>LoongArch: Select ARCH_SUPPORTS_INT128 if CC_HAS_INT128This allows compiling a full 128-bit product of two 64-bit integers as amul/mulh pair, instead of a nasty long sequence of 20+ instructions.However, after selecting ARCH_SUPPORTS_INT128, when optimizing for sizethe compiler generates calls to __ashlti3, __ashrti3, and __lshrti3 forshifting __int128 values, causing a link failure:    loongarch64-unknown-linux-gnu-ld: kernel/sched/fair.o: in    function `mul_u64_u32_shr&apos;:    &lt;PATH&gt;/include/linux/math64.h:161:(.text+0x5e4): undefined    reference to `__lshrti3&apos;So provide the implementation of these functions if ARCH_SUPPORTS_INT128.Closes: https://lore.kernel.org/loongarch/CAAhV-H5EZ=7OF7CSiYyZ8_+wWuenpo=K2WT8-6mAT4CvzUC_4g@mail.gmail.com/Signed-off-by: Xi Ruoyao &lt;xry111@xry111.site&gt;Signed-off-by: Huacai Chen &lt;chenhuacai@loongson.cn&gt;

            List of files:
            /linux-6.15/arch/loongarch/lib/Makefile</description>
        <pubDate>Tue, 14 May 2024 04:24:18 +0000</pubDate>
        <dc:creator>Xi Ruoyao &lt;xry111@xry111.site&gt;</dc:creator>
    </item>
<item>
        <title>75ded18a - LoongArch: Add SIMD-optimized XOR routines</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/arch/loongarch/lib/Makefile#75ded18a</link>
        <description>LoongArch: Add SIMD-optimized XOR routinesAdd LSX and LASX implementations of xor operations, operating on 64bytes (one L1 cache line) at a time, for a balance between memoryutilization and instruction mix. Huacai confirmed that all futureLoongArch implementations by Loongson (that we care) will likely alsofeature 64-byte cache lines, and experiments show no throughputimprovement with further unrolling.Performance numbers measured during system boot on a 3A5000 @ 2.5GHz:&gt; 8regs           : 12702 MB/sec&gt; 8regs_prefetch  : 10920 MB/sec&gt; 32regs          : 12686 MB/sec&gt; 32regs_prefetch : 10918 MB/sec&gt; lsx             : 17589 MB/sec&gt; lasx            : 26116 MB/secAcked-by: Song Liu &lt;song@kernel.org&gt;Signed-off-by: WANG Xuerui &lt;git@xen0n.name&gt;Signed-off-by: Huacai Chen &lt;chenhuacai@loongson.cn&gt;

            List of files:
            /linux-6.15/arch/loongarch/lib/Makefile</description>
        <pubDate>Wed, 06 Sep 2023 14:53:55 +0000</pubDate>
        <dc:creator>WANG Xuerui &lt;git@xen0n.name&gt;</dc:creator>
    </item>
<item>
        <title>8b5ee2c6 - LoongArch: Add support for function error injection</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/arch/loongarch/lib/Makefile#8b5ee2c6</link>
        <description>LoongArch: Add support for function error injectionInspired by the commit 42d038c4fb00f (&quot;arm64: Add support for functionerror injection&quot;) and the commit ee55ff803b383 (&quot;riscv: Add support forfunction error injection&quot;), this patch supports function error injectionfor LoongArch.Mainly implement two functions:(1) regs_set_return_value() which is used to overwrite the return value,(2) override_function_with_return() which is used to override the probedfunction returning and jump to its caller.Here is a simple test under CONFIG_FUNCTION_ERROR_INJECTION andCONFIG_FAIL_FUNCTION:  # echo sys_clone &gt; /sys/kernel/debug/fail_function/inject  # echo 100 &gt; /sys/kernel/debug/fail_function/probability  # dmesg  bash: fork: Invalid argument  # dmesg  ...  FAULT_INJECTION: forcing a failure.  name fail_function, interval 1, probability 100, space 0, times 1  ...  Call Trace:  [&lt;90000000002238f4&gt;] show_stack+0x5c/0x180  [&lt;90000000012e384c&gt;] dump_stack_lvl+0x60/0x88  [&lt;9000000000b1879c&gt;] should_fail_ex+0x1b0/0x1f4  [&lt;900000000032ead4&gt;] fei_kprobe_handler+0x28/0x6c  [&lt;9000000000230970&gt;] kprobe_breakpoint_handler+0xf0/0x118  [&lt;90000000012e3e60&gt;] do_bp+0x2c4/0x358  [&lt;9000000002241924&gt;] exception_handlers+0x1924/0x10000  [&lt;900000000023b7d0&gt;] sys_clone+0x0/0x4  [&lt;90000000012e4744&gt;] do_syscall+0x7c/0x94  [&lt;9000000000221e44&gt;] handle_syscall+0xc4/0x160Tested-by: Hengqi Chen &lt;hengqi.chen@gmail.com&gt;Acked-by: Masami Hiramatsu (Google) &lt;mhiramat@kernel.org&gt;Signed-off-by: Tiezhu Yang &lt;yangtiezhu@loongson.cn&gt;Signed-off-by: Huacai Chen &lt;chenhuacai@loongson.cn&gt;

            List of files:
            /linux-6.15/arch/loongarch/lib/Makefile</description>
        <pubDate>Mon, 01 May 2023 09:19:52 +0000</pubDate>
        <dc:creator>Tiezhu Yang &lt;yangtiezhu@loongson.cn&gt;</dc:creator>
    </item>
<item>
        <title>69e3a6aa - LoongArch: Add checksum optimization for 64-bit system</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/arch/loongarch/lib/Makefile#69e3a6aa</link>
        <description>LoongArch: Add checksum optimization for 64-bit systemLoongArch platform is 64-bit system, which supports 8-bytes memoryaccessing, but generic checksum functions use 4-byte memory access.So add 8-bytes memory access optimization for checksum functions onLoongArch. And the code comes from arm64 system.When network hw checksum is disabled, iperf performance improves about10% with this patch.Signed-off-by: Bibo Mao &lt;maobibo@loongson.cn&gt;Signed-off-by: Huacai Chen &lt;chenhuacai@loongson.cn&gt;

            List of files:
            /linux-6.15/arch/loongarch/lib/Makefile</description>
        <pubDate>Mon, 01 May 2023 09:19:43 +0000</pubDate>
        <dc:creator>Bibo Mao &lt;maobibo@loongson.cn&gt;</dc:creator>
    </item>
<item>
        <title>a275a82d - LoongArch: Use alternative to optimize libraries</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/arch/loongarch/lib/Makefile#a275a82d</link>
        <description>LoongArch: Use alternative to optimize librariesUse the alternative to optimize common libraries according whether CPUhas UAL (hardware unaligned access support) feature, including memset(),memcopy(), memmove(), copy_user() and clear_user().We have tested UnixBench on a Loongson-3A5000 quad-core machine (1.6GHz):1, One copy, before patch:System Benchmarks Index Values               BASELINE       RESULT    INDEXDhrystone 2 using register variables         116700.0    9566582.0    819.8Double-Precision Whetstone                       55.0       2805.3    510.1Execl Throughput                                 43.0       2120.0    493.0File Copy 1024 bufsize 2000 maxblocks          3960.0     209833.0    529.9File Copy 256 bufsize 500 maxblocks            1655.0      89400.0    540.2File Copy 4096 bufsize 8000 maxblocks          5800.0     320036.0    551.8Pipe Throughput                               12440.0     340624.0    273.8Pipe-based Context Switching                   4000.0     109939.1    274.8Process Creation                                126.0       4728.7    375.3Shell Scripts (1 concurrent)                     42.4       2223.1    524.3Shell Scripts (8 concurrent)                      6.0        883.1   1471.9System Call Overhead                          15000.0     518639.1    345.8                                                                   ========System Benchmarks Index Score                                         500.22, One copy, after patch:System Benchmarks Index Values               BASELINE       RESULT    INDEXDhrystone 2 using register variables         116700.0    9567674.7    819.9Double-Precision Whetstone                       55.0       2805.5    510.1Execl Throughput                                 43.0       2392.7    556.4File Copy 1024 bufsize 2000 maxblocks          3960.0     417804.0   1055.1File Copy 256 bufsize 500 maxblocks            1655.0     112909.5    682.2File Copy 4096 bufsize 8000 maxblocks          5800.0    1255207.4   2164.2Pipe Throughput                               12440.0     555712.0    446.7Pipe-based Context Switching                   4000.0      99964.5    249.9Process Creation                                126.0       5192.5    412.1Shell Scripts (1 concurrent)                     42.4       2302.4    543.0Shell Scripts (8 concurrent)                      6.0        919.6   1532.6System Call Overhead                          15000.0     511159.3    340.8                                                                   ========System Benchmarks Index Score                                         640.13, Four copies, before patch:System Benchmarks Index Values               BASELINE       RESULT    INDEXDhrystone 2 using register variables         116700.0   38268610.5   3279.2Double-Precision Whetstone                       55.0      11222.2   2040.4Execl Throughput                                 43.0       7892.0   1835.3File Copy 1024 bufsize 2000 maxblocks          3960.0     235149.6    593.8File Copy 256 bufsize 500 maxblocks            1655.0      74959.6    452.9File Copy 4096 bufsize 8000 maxblocks          5800.0     545048.5    939.7Pipe Throughput                               12440.0    1337359.0   1075.0Pipe-based Context Switching                   4000.0     473663.9   1184.2Process Creation                                126.0      17491.2   1388.2Shell Scripts (1 concurrent)                     42.4       6865.7   1619.3Shell Scripts (8 concurrent)                      6.0       1015.9   1693.1System Call Overhead                          15000.0    1899535.2   1266.4                                                                   ========System Benchmarks Index Score                                        1278.34, Four copies, after patch:System Benchmarks Index Values               BASELINE       RESULT    INDEXDhrystone 2 using register variables         116700.0   38272815.5   3279.6Double-Precision Whetstone                       55.0      11222.8   2040.5Execl Throughput                                 43.0       8839.2   2055.6File Copy 1024 bufsize 2000 maxblocks          3960.0     313912.9    792.7File Copy 256 bufsize 500 maxblocks            1655.0      80976.1    489.3File Copy 4096 bufsize 8000 maxblocks          5800.0    1176594.3   2028.6Pipe Throughput                               12440.0    2100941.9   1688.9Pipe-based Context Switching                   4000.0     476696.4   1191.7Process Creation                                126.0      18394.7   1459.9Shell Scripts (1 concurrent)                     42.4       7172.2   1691.6Shell Scripts (8 concurrent)                      6.0       1058.3   1763.9System Call Overhead                          15000.0    1874714.7   1249.8                                                                   ========System Benchmarks Index Score                                        1488.8Signed-off-by: Jun Yi &lt;yijun@loongson.cn&gt;Signed-off-by: Huacai Chen &lt;chenhuacai@loongson.cn&gt;

            List of files:
            /linux-6.15/arch/loongarch/lib/Makefile</description>
        <pubDate>Sat, 10 Dec 2022 14:39:59 +0000</pubDate>
        <dc:creator>Huacai Chen &lt;chenhuacai@loongson.cn&gt;</dc:creator>
    </item>
<item>
        <title>61a6fccc - LoongArch: Add unaligned access support</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/arch/loongarch/lib/Makefile#61a6fccc</link>
        <description>LoongArch: Add unaligned access supportLoongson-2 series (Loongson-2K500, Loongson-2K1000) don&apos;t supportunaligned access in hardware, while Loongson-3 series (Loongson-3A5000,Loongson-3C5000) are configurable whether support unaligned access inhardware. This patch add unaligned access emulation for those LoongArchprocessors without hardware support.Signed-off-by: Huacai Chen &lt;chenhuacai@loongson.cn&gt;

            List of files:
            /linux-6.15/arch/loongarch/lib/Makefile</description>
        <pubDate>Sat, 10 Dec 2022 14:39:59 +0000</pubDate>
        <dc:creator>Huacai Chen &lt;chenhuacai@loongson.cn&gt;</dc:creator>
    </item>
<item>
        <title>fa96b57c - LoongArch: Add build infrastructure</title>
        <link>http://172.16.0.5:8080/history/linux-6.15/arch/loongarch/lib/Makefile#fa96b57c</link>
        <description>LoongArch: Add build infrastructureAdd Kbuild, Makefile, Kconfig and link script for LoongArch buildinfrastructure.Reviewed-by: Guo Ren &lt;guoren@kernel.org&gt;Reviewed-by: WANG Xuerui &lt;git@xen0n.name&gt;Reviewed-by: Jiaxun Yang &lt;jiaxun.yang@flygoat.com&gt;Signed-off-by: Huacai Chen &lt;chenhuacai@loongson.cn&gt;

            List of files:
            /linux-6.15/arch/loongarch/lib/Makefile</description>
        <pubDate>Tue, 31 May 2022 10:04:11 +0000</pubDate>
        <dc:creator>Huacai Chen &lt;chenhuacai@loongson.cn&gt;</dc:creator>
    </item>
</channel>
</rss>
