From: "Jan Beulich" <jbeulich@novell.com>
To: "Bryan O'Sullivan" <bos@pathscale.com>, "Andi Kleen" <ak@suse.de>
Cc: "Roland Dreier" <rolandd@cisco.com>,
<linux-kernel@vger.kernel.org>, <patches@x86-64.org>
Subject: Re: [patches] [PATCH] [8/35] x86_64: a memcpy that tries to reducecache pressure
Date: Mon, 30 Apr 2007 08:26:05 +0100 [thread overview]
Message-ID: <4635B62D.76E4.0078.0@novell.com> (raw)
In-Reply-To: <20070428175232.5A688151CA@wotan.suse.de>
>+ testb $8, %dxl /* rdx is 3,5,6,7,9..15 */
Could you use the more conventional name %dl (or whatever is being meant)
here?
>+.L42:
>+ prefetchnta 128(%rsi)
I think I commented similarly on a previous version of the patch: This will
- result in still bringing the first 128 bytes into the cache
- prevent caching of memory beyond the buffer being dealt with
>+ sfence
Similarly, I continue to believe this is wrong here (again I commented before
that this is likely unnecessary, or if at all, it should be an lfence).
>+ orl %edx, %edx
Operations of this type generally should use 'test', not 'or' or 'and'.
Jan
next prev parent reply other threads:[~2007-04-30 7:25 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-04-28 17:52 [PATCH] [0/35] Some x86 2.6.22 candidate patches for review Andi Kleen
2007-04-28 17:52 ` [PATCH] [1/35] i386: revert i386-fix-the-verify_quirk_intel_irqbalance Andi Kleen
2007-04-28 17:52 ` [PATCH] [2/35] x86_64: revert x86_64-mm-add-genapic_force Andi Kleen
2007-04-28 17:52 ` [PATCH] [3/35] x86: revert x86_64-mm-fix-the-irqbalance-quirk-for-e7320-e7520-e7525 Andi Kleen
2007-04-28 17:52 ` [PATCH] [4/35] x86_64: optimize & fix APIC mode setup Andi Kleen
2007-04-28 17:52 ` [PATCH] [5/35] x86_64: always use physical delivery mode on > 8 CPUs Andi Kleen
2007-04-28 17:52 ` [PATCH] [6/35] x86_64: remove clustered APIC mode Andi Kleen
2007-04-28 17:52 ` [PATCH] [7/35] x86: default to physical mode on hotplug CPU kernels Andi Kleen
2007-04-28 17:52 ` [PATCH] [8/35] x86_64: a memcpy that tries to reduce cache pressure Andi Kleen
2007-04-30 7:26 ` Jan Beulich [this message]
2007-04-30 9:04 ` [patches] [PATCH] [8/35] x86_64: a memcpy that tries to reducecache pressure Andi Kleen
2007-04-30 16:14 ` Bryan O'Sullivan
2007-04-30 16:25 ` Andi Kleen
2007-04-28 17:52 ` [PATCH] [9/35] i386: adjustments to page table dump during oops (v4) Andi Kleen
2007-04-28 17:52 ` [PATCH] [10/35] x86: adjust inclusion of asm/fixmap.h Andi Kleen
2007-04-28 17:52 ` [PATCH] [11/35] x86_64: adjust inclusion of asm/vsyscall32.h Andi Kleen
2007-04-28 17:52 ` [PATCH] [12/35] x86: consolidate smp_send_stop() Andi Kleen
2007-04-28 17:52 ` [PATCH] [13/35] i386: No need to use -traditional for processing asm in i386/kernel/ Andi Kleen
2007-04-28 17:52 ` [PATCH] [14/35] i386: mtrr range check correction Andi Kleen
2007-04-28 17:52 ` [PATCH] [15/35] i386: pit_latch_buggy has no effect Andi Kleen
2007-04-28 17:52 ` [PATCH] [16/35] i386: Add an option for the VIA C7 which sets appropriate L1 cache Andi Kleen
2007-04-28 18:08 ` Jeff Garzik
2007-04-28 18:28 ` Dave Jones
2007-04-28 17:52 ` [PATCH] [17/35] i386: probe_roms() cleanup Andi Kleen
2007-04-28 17:52 ` [PATCH] [18/35] x86_64: a few missing entry.S annotations Andi Kleen
2007-04-28 17:52 ` [PATCH] [19/35] i386: Add dwarf2 annotations to *_user and checksum functions Andi Kleen
2007-04-28 17:52 ` [PATCH] [20/35] x86: Fix i386 and x86_64 fault information pollution Andi Kleen
2007-04-28 17:52 ` [PATCH] [21/35] x86_64: Some cleanup in time.c Andi Kleen
2007-04-28 17:52 ` [PATCH] [22/35] i386: i386 make NMI use PERFCTR1 for architectural perfmon (take 2) Andi Kleen
2007-04-28 17:52 ` [PATCH] [23/35] x86_64: x86_64 " Andi Kleen
2007-04-28 17:52 ` [PATCH] [24/35] i386: Add __init to probe_bigsmp Andi Kleen
2007-04-28 17:52 ` [PATCH] [25/35] i386: Change sysenter_setup to __cpuinit & improve __INIT, __INITDATA Andi Kleen
2007-04-28 17:52 ` [PATCH] [26/35] x86_64: Correct max number of CPUs in Kconfig Andi Kleen
2007-04-28 17:52 ` [PATCH] [27/35] i386: Support Oprofile for AMD Family 10 CPUs Andi Kleen
2007-04-28 17:52 ` [PATCH] [28/35] x86: Drop cc-options call for all options supported in gcc 3.2+ Andi Kleen
2007-04-28 17:52 ` [PATCH] [29/35] i386: Update __copy_to_user_inatomic linuxdoc description Andi Kleen
2007-04-28 17:52 ` [PATCH] [30/35] i386: clean up mach_reboot_fixups Andi Kleen
2007-04-28 17:52 ` [PATCH] [31/35] i386: Fix usage of -mtune when X86_GENERIC=y or CONFIG_MCORE2=y Andi Kleen
2007-04-28 17:52 ` [PATCH] [32/35] x86_64: Remove unused set_seg_base Andi Kleen
2007-04-28 17:52 ` [PATCH] [33/35] x86_64: Remove duplicated code for reading control registers Andi Kleen
2007-04-28 17:53 ` [PATCH] [34/35] x86_64: fix cpu MHz reporting on constant_tsc cpus Andi Kleen
2007-04-28 17:53 ` [PATCH] [35/35] i386: Simplify smp_call_function*() by using common implementation Andi Kleen
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=4635B62D.76E4.0078.0@novell.com \
--to=jbeulich@novell.com \
--cc=ak@suse.de \
--cc=bos@pathscale.com \
--cc=linux-kernel@vger.kernel.org \
--cc=patches@x86-64.org \
--cc=rolandd@cisco.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: link
Be 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.