[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH v4 0/8] Allow dynamic allocation of software IO TLB bounce buffers
- To: Petr Tesarik <petrtesarik@xxxxxxxxxxxxxxx>
- From: Christoph Hellwig <hch@xxxxxx>
- Date: Thu, 20 Jul 2023 08:52:16 +0200
- Cc: Stefano Stabellini <sstabellini@xxxxxxxxxx>, Russell King <linux@xxxxxxxxxxxxxxx>, Thomas Bogendoerfer <tsbogend@xxxxxxxxxxxxxxxx>, Thomas Gleixner <tglx@xxxxxxxxxxxxx>, Ingo Molnar <mingo@xxxxxxxxxx>, Borislav Petkov <bp@xxxxxxxxx>, Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>, "maintainer:X86 ARCHITECTURE (32-BIT AND 64-BIT)" <x86@xxxxxxxxxx>, "H. Peter Anvin" <hpa@xxxxxxxxx>, Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>, "Rafael J. Wysocki" <rafael@xxxxxxxxxx>, Juergen Gross <jgross@xxxxxxxx>, Oleksandr Tyshchenko <oleksandr_tyshchenko@xxxxxxxx>, Christoph Hellwig <hch@xxxxxx>, Marek Szyprowski <m.szyprowski@xxxxxxxxxxx>, Robin Murphy <robin.murphy@xxxxxxx>, Petr Tesarik <petr.tesarik.ext@xxxxxxxxxx>, Jonathan Corbet <corbet@xxxxxxx>, Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx>, Hans de Goede <hdegoede@xxxxxxxxxx>, James Seo <james@xxxxxxxxxx>, James Clark <james.clark@xxxxxxx>, Kees Cook <keescook@xxxxxxxxxxxx>, "moderated list:XEN HYPERVISOR ARM" <xen-devel@xxxxxxxxxxxxxxxxxxxx>, "moderated list:ARM PORT" <linux-arm-kernel@xxxxxxxxxxxxxxxxxxx>, open list <linux-kernel@xxxxxxxxxxxxxxx>, "open list:MIPS" <linux-mips@xxxxxxxxxxxxxxx>, "open list:XEN SWIOTLB SUBSYSTEM" <iommu@xxxxxxxxxxxxxxx>, Roberto Sassu <roberto.sassu@xxxxxxxxxxxxxxx>, Kefeng Wang <wangkefeng.wang@xxxxxxxxxx>, petr@xxxxxxxxxxx
- Delivery-date: Thu, 20 Jul 2023 06:52:24 +0000
- List-id: Xen developer discussion <xen-devel.lists.xenproject.org>
Just to add a highlevel comment here after I feel like I need a little
more time to review the guts.
I'm still pretty concerned about the extra list that needs to be
consulted in is_swiotlb_buffer, but I can't really think of
anything better. Maybe an xarray has better cache characteristics,
but that one requires even more allocations in the low-level dma map
path.
One thing I'd like to see for the next version is to make the
new growing code a config option at least for now. It is a pretty
big change of the existing swiotlb behavior, and I want people to opt
into it conciously. Maybe we can drop the option again after a few
years once everything has settled.
|
