From: Marek Vasut <marex@denx.de>
To: u-boot@lists.denx.de
Subject: [U-Boot] [PATCH 2/6] usb:udc:samsung: Remove redundant cache operation from Samsung UDC driver
Date: Sat, 1 Feb 2014 23:49:26 +0100 [thread overview]
Message-ID: <201402012349.26505.marex@denx.de> (raw)
In-Reply-To: <20140201105627.4d6c06b3@jawa>
On Saturday, February 01, 2014 at 10:56:27 AM, Lukasz Majewski wrote:
> On Sat, 1 Feb 2014 03:50:26 +0100
>
> Marek Vasut <marex@denx.de> wrote:
> > On Friday, January 31, 2014 at 01:16:25 PM, Lukasz Majewski wrote:
> > > A set of cache operations (both invalidation and flush) were
> > > redundant in the S3C HS OTG Samsung driver.
> > >
> > > Test condition
> > > - test HW + measurement: Trats - Exynos4210 rev.1
> > > - test HW Trats2 - Exynos4412 rev.1
> > > 400 MiB compressed rootfs image download with `thor 0 mmc 0`
> >
> > The commit message is missing a proper explanation _WHY_ were they
> > redundant. I do not understand why they were redundant ... and no,
> > the test you performed does not justify removal of cache management
> > calls.
>
> The s3c UDC driver is in u-boot since 2011. It has been added when
> at Samsung boards (s5p_goni) we didn't have cache enabled.
>
> Then there was a transition, after which L1 was enabled. Since UDC is
> co-working with gadgets on the beginning it was easier to perform the
> cache management inside the UDC driver.
>
> That is why we had to copy the buffers (since e.g. device descriptors
> tends to be unaligned) - which also degraded performance.
>
> Now, all gadget code seems to be memalign'ed and ready for direct buffer
> passing (despite the two overlooked kmallocs in the mass storage
> gadget - which I fix in this patch set).
>
> To sum up:
>
> 1. s3c_udc_ep0_zlp - EP0 ZLP packets don't need to invalidate the cache
> (since it is zero length transmission)
>
> 2. s3c_udc_pre_setup - cache invalidation is not needed when I setup
> buffer for OUT EP0 transmission.
>
> The above two invalidation calls had been added by me, and are mine
> mistakes. Those don't contribute to transmission speed up (and shall be
> regarded as a cosmetic changes)
>
> 3. setdma_rx - here I invalidate parts of the s3c UDC driver's internal
> buffer. This call is not needed anymore since we reuse the buffers
> passed from gadgets.
And you do correct cache management on those in the UDC driver or in the gadget
driver ?
> This is the key speed improvement here.
This should be in the commit message really ;-)
Best regards,
Marek Vasut
next prev parent reply other threads:[~2014-02-01 22:49 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-01-31 12:16 [U-Boot] [PATCH 0/6] usb:samsung: Exynos4 SoC USB code improvements Lukasz Majewski
2014-01-31 12:16 ` [U-Boot] [PATCH 1/6] usb:gadget:ums: Replace malloc calls with memalign to fix cache buffer alignment Lukasz Majewski
2014-02-01 2:48 ` Marek Vasut
2014-02-01 9:10 ` Lukasz Majewski
2014-01-31 12:16 ` [U-Boot] [PATCH 2/6] usb:udc:samsung: Remove redundant cache operation from Samsung UDC driver Lukasz Majewski
2014-02-01 2:50 ` Marek Vasut
2014-02-01 9:56 ` Lukasz Majewski
2014-02-01 22:49 ` Marek Vasut [this message]
2014-02-03 8:05 ` Lukasz Majewski
2014-02-03 18:06 ` Marek Vasut
2014-02-04 6:23 ` Lukasz Majewski
2014-01-31 12:16 ` [U-Boot] [PATCH 3/6] usb:udc:samsung: Allow burst transfers for non EP0 endpints Lukasz Majewski
2014-01-31 12:16 ` [U-Boot] [PATCH 4/6] usb:udc:samsung: Zero copy approach for data passed to Samsung's UDC driver Lukasz Majewski
2014-02-01 2:55 ` Marek Vasut
2014-02-01 11:05 ` Lukasz Majewski
2014-02-01 22:55 ` Marek Vasut
2014-02-03 11:06 ` Lukasz Majewski
2014-02-03 18:11 ` Marek Vasut
2014-02-04 7:29 ` Lukasz Majewski
2014-02-04 20:21 ` Marek Vasut
2014-02-04 21:49 ` Lukasz Majewski
2014-02-05 2:35 ` Marek Vasut
2014-01-31 12:16 ` [U-Boot] [PATCH 5/6] usb:gadget:f_thor: Allocate request up to THOR_PACKET_SIZE not ep->maxpacket Lukasz Majewski
2014-01-31 12:16 ` [U-Boot] [PATCH 6/6] usb:gadget:f_thor: cosmetic: Remove debug memset Lukasz Majewski
2014-02-05 9:10 ` [U-Boot] [PATCH v2 0/6] usb:samsung: Exynos4 SoC USB code improvements Lukasz Majewski
2014-02-05 9:10 ` [U-Boot] [PATCH v2 1/6] usb:gadget:ums: Replace malloc calls with memalign to fix cache buffer alignment Lukasz Majewski
2014-02-06 1:20 ` Marek Vasut
2014-02-06 6:33 ` Lukasz Majewski
2014-02-06 6:41 ` Marek Vasut
2014-02-06 8:16 ` Lukasz Majewski
2014-02-05 9:10 ` [U-Boot] [PATCH v2 2/6] usb:udc:samsung: Remove redundant cache operation from Samsung UDC driver Lukasz Majewski
2014-02-05 9:10 ` [U-Boot] [PATCH v2 3/6] usb:udc:samsung: Allow burst transfers for non EP0 endpints Lukasz Majewski
2014-02-05 9:10 ` [U-Boot] [PATCH v2 4/6] usb:udc:samsung: Zero copy approach for data passed to Samsung's UDC driver Lukasz Majewski
2014-02-05 9:10 ` [U-Boot] [PATCH v2 5/6] usb:gadget:f_thor: Allocate request up to THOR_PACKET_SIZE not ep->maxpacket Lukasz Majewski
2014-02-05 9:10 ` [U-Boot] [PATCH v2 6/6] usb:gadget:f_thor: cosmetic: Remove debug memset Lukasz Majewski
2014-02-06 1:24 ` [U-Boot] [PATCH v2 0/6] usb:samsung: Exynos4 SoC USB code improvements Marek Vasut
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=201402012349.26505.marex@denx.de \
--to=marex@denx.de \
--cc=u-boot@lists.denx.de \
/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.