public inbox for [email protected]
help / color / mirror / Atom feedFrom: Nathan Bossart <[email protected]>
To: Noah Misch <[email protected]>
Cc: Tom Lane <[email protected]>
Cc: Matthias van de Meent <[email protected]>
Cc: Amonson, Paul D <[email protected]>
Cc: [email protected] <[email protected]>
Cc: Shankaran, Akash <[email protected]>
Subject: Re: Popcount optimization using AVX512
Date: Tue, 7 Nov 2023 14:14:41 -0600
Message-ID: <20231107201441.GA898662@nathanxps13> (raw)
In-Reply-To: <[email protected]>
References: <BL1PR11MB5304097DF7EA81D04C33F3D1DCA6A@BL1PR11MB5304.namprd11.prod.outlook.com>
<CAEze2WjaFLhp7=Eo-mbSaHsoeq7ZEs00yZ2+FpSpijH+KN_hbA@mail.gmail.com>
<20231107022240.GA729644@nathanxps13>
<[email protected]>
<[email protected]>
<20231107035926.GD729644@nathanxps13>
<[email protected]>
On Mon, Nov 06, 2023 at 09:53:15PM -0800, Noah Misch wrote:
> On Mon, Nov 06, 2023 at 09:59:26PM -0600, Nathan Bossart wrote:
>> On Mon, Nov 06, 2023 at 07:15:01PM -0800, Noah Misch wrote:
>> > The glibc/gcc "ifunc" mechanism was designed to solve this problem of choosing
>> > a function implementation based on the runtime CPU, without incurring function
>> > pointer overhead. I would not attempt to use AVX512 on non-glibc systems, and
>> > I would use ifunc to select the desired popcount implementation on glibc:
>> > https://gcc.gnu.org/onlinedocs/gcc-4.8.5/gcc/Function-Attributes.html
>>
>> Thanks, that seems promising for the function pointer cases. I'll plan on
>> trying to convert one of the existing ones to use it. BTW it looks like
>> LLVM has something similar [0].
>>
>> IIUC this unfortunately wouldn't help for cases where we wanted to keep
>> stuff inlined, such as is_valid_ascii() and the functions in pg_lfind.h,
>> unless we applied it to the calling functions, but that doesn't Ń•ound
>> particularly maintainable.
>
> Agreed, it doesn't solve inline cases. If the gains are big enough, we should
> move toward packages containing N CPU-specialized copies of the postgres
> binary, with bin/postgres just exec'ing the right one.
I performed a quick test with ifunc on my x86 machine that ordinarily uses
the runtime checks for the CRC32C code, and I actually see a consistent
3.5% regression for pg_waldump -z on 100M 65-byte records. I've attached
the patch used for testing.
The multiple-copies-of-the-postgres-binary idea seems interesting. That's
probably not something that could be enabled by default, but perhaps we
could add support for a build option.
--
Nathan Bossart
Amazon Web Services: https://aws.amazon.com
Attachments:
[text/x-diff] ifunc_test.patch (1.6K, ../20231107201441.GA898662@nathanxps13/2-ifunc_test.patch)
download | inline diff:
diff --git a/src/include/port/pg_crc32c.h b/src/include/port/pg_crc32c.h
index d085f1dc00..6db411ee29 100644
--- a/src/include/port/pg_crc32c.h
+++ b/src/include/port/pg_crc32c.h
@@ -78,7 +78,7 @@ extern pg_crc32c pg_comp_crc32c_loongarch(pg_crc32c crc, const void *data, size_
#define FIN_CRC32C(crc) ((crc) ^= 0xFFFFFFFF)
extern pg_crc32c pg_comp_crc32c_sb8(pg_crc32c crc, const void *data, size_t len);
-extern pg_crc32c (*pg_comp_crc32c) (pg_crc32c crc, const void *data, size_t len);
+extern pg_crc32c pg_comp_crc32c(pg_crc32c crc, const void *data, size_t len);
#ifdef USE_SSE42_CRC32C_WITH_RUNTIME_CHECK
extern pg_crc32c pg_comp_crc32c_sse42(pg_crc32c crc, const void *data, size_t len);
diff --git a/src/port/pg_crc32c_sse42_choose.c b/src/port/pg_crc32c_sse42_choose.c
index 41ff4a35ad..62bb981ee8 100644
--- a/src/port/pg_crc32c_sse42_choose.c
+++ b/src/port/pg_crc32c_sse42_choose.c
@@ -51,14 +51,14 @@ pg_crc32c_sse42_available(void)
* so that subsequent calls are routed directly to the chosen implementation.
*/
static pg_crc32c
-pg_comp_crc32c_choose(pg_crc32c crc, const void *data, size_t len)
+(*pg_comp_crc32c_choose (void))(pg_crc32c crc, const void *data, size_t len)
{
if (pg_crc32c_sse42_available())
- pg_comp_crc32c = pg_comp_crc32c_sse42;
+ return pg_comp_crc32c_sse42;
else
- pg_comp_crc32c = pg_comp_crc32c_sb8;
-
- return pg_comp_crc32c(crc, data, len);
+ return pg_comp_crc32c_sb8;
}
-pg_crc32c (*pg_comp_crc32c) (pg_crc32c crc, const void *data, size_t len) = pg_comp_crc32c_choose;
+pg_crc32c
+pg_comp_crc32c(pg_crc32c crc, const void *data, size_t len)
+ __attribute__ ((ifunc ("pg_comp_crc32c_choose")));
view thread (11+ messages) latest in thread
reply
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: [email protected]
Cc: [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected]
Subject: Re: Popcount optimization using AVX512
In-Reply-To: <20231107201441.GA898662@nathanxps13>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox