From: ming.lei-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org To: greg-U8xfFu+wG4EAvxtiuMwx3w@public.gmane.org, stern-nwvwT67g6+6dFdvTe/nMLpVzexx5G7lz@public.gmane.org Cc: linux-usb-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org, linux-omap-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Ming Lei <ming.lei-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org>, Russell King <linux-lFZ/pmaqli7XmaaqVzeoHQ@public.gmane.org> Subject: [PATCH] usb: ehci: make HC see up-to-date qh/qtd descriptor ASAP Date: Wed, 31 Aug 2011 00:03:13 +0800 [thread overview] Message-ID: <1314720193-26577-1-git-send-email-ming.lei@canonical.com> (raw) From: Ming Lei <ming.lei-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org> This patch introduces the helper of ehci_sync_mem to flush qtd/qh into memory immediately on some ARM, so that HC can see the up-to-date qtd/qh descriptor asap. This patch fixs one performance bug on ARM Cortex A9 dual core platform, which has been reported on quite a few ARM machines (OMAP4, Tegra 2, snowball...), see details from link of https://bugs.launchpad.net/bugs/709245. The patch has been tested ok on OMAP4 panda A1 board, and the performance of 'dd' over usb mass storage can be increased from 4~5MB/sec to 14~16MB/sec after applying this patch. Cc: Alan Stern <stern-nwvwT67g6+6dFdvTe/nMLpVzexx5G7lz@public.gmane.org> Cc: Russell King <linux-lFZ/pmaqli7XmaaqVzeoHQ@public.gmane.org> Signed-off-by: Ming Lei <ming.lei-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org> --- drivers/usb/host/ehci-q.c | 18 ++++++++++++++++++ drivers/usb/host/ehci.h | 17 +++++++++++++++++ 2 files changed, 35 insertions(+), 0 deletions(-) diff --git a/drivers/usb/host/ehci-q.c b/drivers/usb/host/ehci-q.c index 0917e3a..2719879 100644 --- a/drivers/usb/host/ehci-q.c +++ b/drivers/usb/host/ehci-q.c @@ -995,6 +995,12 @@ static void qh_link_async (struct ehci_hcd *ehci, struct ehci_qh *qh) head->qh_next.qh = qh; head->hw->hw_next = dma; + /* + * flush qh descriptor into memory immediately, + * see comments in qh_append_tds. + * */ + ehci_sync_mem(); + qh_get(qh); qh->xacterrs = 0; qh->qh_state = QH_STATE_LINKED; @@ -1082,6 +1088,18 @@ static struct ehci_qh *qh_append_tds ( wmb (); dummy->hw_token = token; + /* + * Writing to dma coherent buffer on ARM may + * be delayed to reach memory, so HC may not see + * hw_token of dummy qtd in time, which can cause + * the qtd transaction to be executed very late, + * and degrade performance a lot. ehci_sync_mem + * is added to flush 'token' immediatelly into + * memory, so that ehci can execute the transaction + * ASAP. + * */ + ehci_sync_mem(); + urb->hcpriv = qh_get (qh); } } diff --git a/drivers/usb/host/ehci.h b/drivers/usb/host/ehci.h index cc7d337..313d9d6 100644 --- a/drivers/usb/host/ehci.h +++ b/drivers/usb/host/ehci.h @@ -738,6 +738,23 @@ static inline u32 hc32_to_cpup (const struct ehci_hcd *ehci, const __hc32 *x) #endif +/* + * Writing to dma coherent memory on ARM may be delayed via L2 + * writing buffer, so introduce the helper which can flush L2 writing + * buffer into memory immediately, especially used to flush ehci + * descriptor to memory. + * */ +#ifdef CONFIG_ARM_DMA_MEM_BUFFERABLE +static inline void ehci_sync_mem() +{ + mb(); +} +#else +static inline void ehci_sync_mem() +{ +} +#endif + /*-------------------------------------------------------------------------*/ #ifndef DEBUG -- 1.7.4.1 -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html
WARNING: multiple messages have this Message-ID (diff)
From: ming.lei@canonical.com (ming.lei at canonical.com) To: linux-arm-kernel@lists.infradead.org Subject: [PATCH] usb: ehci: make HC see up-to-date qh/qtd descriptor ASAP Date: Wed, 31 Aug 2011 00:03:13 +0800 [thread overview] Message-ID: <1314720193-26577-1-git-send-email-ming.lei@canonical.com> (raw) From: Ming Lei <ming.lei@canonical.com> This patch introduces the helper of ehci_sync_mem to flush qtd/qh into memory immediately on some ARM, so that HC can see the up-to-date qtd/qh descriptor asap. This patch fixs one performance bug on ARM Cortex A9 dual core platform, which has been reported on quite a few ARM machines (OMAP4, Tegra 2, snowball...), see details from link of https://bugs.launchpad.net/bugs/709245. The patch has been tested ok on OMAP4 panda A1 board, and the performance of 'dd' over usb mass storage can be increased from 4~5MB/sec to 14~16MB/sec after applying this patch. Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Russell King <linux@arm.linux.org.uk> Signed-off-by: Ming Lei <ming.lei@canonical.com> --- drivers/usb/host/ehci-q.c | 18 ++++++++++++++++++ drivers/usb/host/ehci.h | 17 +++++++++++++++++ 2 files changed, 35 insertions(+), 0 deletions(-) diff --git a/drivers/usb/host/ehci-q.c b/drivers/usb/host/ehci-q.c index 0917e3a..2719879 100644 --- a/drivers/usb/host/ehci-q.c +++ b/drivers/usb/host/ehci-q.c @@ -995,6 +995,12 @@ static void qh_link_async (struct ehci_hcd *ehci, struct ehci_qh *qh) head->qh_next.qh = qh; head->hw->hw_next = dma; + /* + * flush qh descriptor into memory immediately, + * see comments in qh_append_tds. + * */ + ehci_sync_mem(); + qh_get(qh); qh->xacterrs = 0; qh->qh_state = QH_STATE_LINKED; @@ -1082,6 +1088,18 @@ static struct ehci_qh *qh_append_tds ( wmb (); dummy->hw_token = token; + /* + * Writing to dma coherent buffer on ARM may + * be delayed to reach memory, so HC may not see + * hw_token of dummy qtd in time, which can cause + * the qtd transaction to be executed very late, + * and degrade performance a lot. ehci_sync_mem + * is added to flush 'token' immediatelly into + * memory, so that ehci can execute the transaction + * ASAP. + * */ + ehci_sync_mem(); + urb->hcpriv = qh_get (qh); } } diff --git a/drivers/usb/host/ehci.h b/drivers/usb/host/ehci.h index cc7d337..313d9d6 100644 --- a/drivers/usb/host/ehci.h +++ b/drivers/usb/host/ehci.h @@ -738,6 +738,23 @@ static inline u32 hc32_to_cpup (const struct ehci_hcd *ehci, const __hc32 *x) #endif +/* + * Writing to dma coherent memory on ARM may be delayed via L2 + * writing buffer, so introduce the helper which can flush L2 writing + * buffer into memory immediately, especially used to flush ehci + * descriptor to memory. + * */ +#ifdef CONFIG_ARM_DMA_MEM_BUFFERABLE +static inline void ehci_sync_mem() +{ + mb(); +} +#else +static inline void ehci_sync_mem() +{ +} +#endif + /*-------------------------------------------------------------------------*/ #ifndef DEBUG -- 1.7.4.1
next reply other threads:[~2011-08-30 16:03 UTC|newest] Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top 2011-08-30 16:03 ming.lei-Z7WLFzj8eWMS+FvcfC7Uqw [this message] 2011-08-30 16:03 ` [PATCH] usb: ehci: make HC see up-to-date qh/qtd descriptor ASAP ming.lei at canonical.com 2011-08-30 16:15 ` Alan Stern 2011-08-30 16:15 ` Alan Stern 2011-08-30 16:38 ` Mark Salter 2011-08-30 16:38 ` Mark Salter 2011-08-30 17:15 ` Alan Stern 2011-08-30 17:15 ` Alan Stern 2011-08-30 18:45 ` Mark Salter 2011-08-30 18:45 ` Mark Salter 2011-08-30 17:26 ` Will Deacon 2011-08-30 17:26 ` Will Deacon [not found] ` <20110830172642.GE3464-SGELLbQ0bobZROr8t4l/smS4ubULX0JqMm0uRHvK7Nw@public.gmane.org> 2011-08-30 17:48 ` Greg KH 2011-08-30 17:48 ` Greg KH 2011-08-30 17:54 ` Will Deacon 2011-08-30 17:54 ` Will Deacon [not found] ` <20110830175432.GG3464-SGELLbQ0bobZROr8t4l/smS4ubULX0JqMm0uRHvK7Nw@public.gmane.org> 2011-08-31 0:23 ` Chen Peter-B29397 2011-08-31 0:23 ` Chen Peter-B29397 2011-08-31 8:49 ` Will Deacon 2011-08-31 8:49 ` Will Deacon 2011-08-31 12:33 ` Chen Peter-B29397 2011-08-31 12:33 ` Chen Peter-B29397 2011-08-31 13:43 ` Mark Salter 2011-08-31 13:43 ` Mark Salter 2011-08-31 15:21 ` Will Deacon 2011-08-31 15:21 ` Will Deacon 2011-08-31 15:27 ` Mark Salter 2011-08-31 15:27 ` Mark Salter 2011-08-31 16:12 ` Marc Zyngier 2011-08-31 16:12 ` Marc Zyngier 2011-08-31 16:55 ` Marc Dietrich 2011-08-31 16:55 ` Marc Dietrich 2011-09-01 10:34 ` Marc Zyngier 2011-09-01 10:34 ` Marc Zyngier [not found] ` <4E5F5FA9.3010305-5wv7dgnIgG8@public.gmane.org> 2011-09-01 11:13 ` Marc Dietich 2011-09-01 11:13 ` Marc Dietich 2011-09-01 19:08 ` Stephen Warren 2011-09-01 19:08 ` Stephen Warren 2011-09-02 9:50 ` Marc Zyngier 2011-09-02 9:50 ` Marc Zyngier 2011-09-02 17:07 ` Stephen Warren 2011-09-02 17:07 ` Stephen Warren [not found] ` <74CDBE0F657A3D45AFBB94109FB122FF04B327A383-C7FfzLzN0UxDw2glCA4ptUEOCMrvLtNR@public.gmane.org> 2011-09-02 11:13 ` Marc Dietich 2011-09-02 11:13 ` Marc Dietich 2011-08-31 17:46 ` Nicolas Pitre 2011-08-31 17:46 ` Nicolas Pitre 2011-08-31 17:51 ` Will Deacon 2011-08-31 17:51 ` Will Deacon [not found] ` <20110831175147.GI8777-SGELLbQ0bobZROr8t4l/smS4ubULX0JqMm0uRHvK7Nw@public.gmane.org> 2011-08-31 18:19 ` Rob Herring 2011-08-31 18:19 ` Rob Herring 2011-08-31 18:35 ` Mark Salter 2011-08-31 18:35 ` Mark Salter 2011-08-31 18:49 ` Rob Herring 2011-08-31 18:49 ` Rob Herring [not found] ` <4E5E8230.9060307-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> 2011-08-31 18:58 ` Mark Salter 2011-08-31 18:58 ` Mark Salter 2011-08-31 19:35 ` Will Deacon 2011-08-31 19:35 ` Will Deacon 2011-09-08 22:41 ` Mark Salter 2011-09-08 22:41 ` Mark Salter [not found] ` <1315521779.2313.29.camel-PDpCo7skNiwAicBL8TP8PQ@public.gmane.org> 2011-10-31 6:49 ` Pandita, Vikram 2011-10-31 6:49 ` Pandita, Vikram 2011-08-31 0:56 ` Ming Lei 2011-08-31 0:56 ` Ming Lei 2011-09-01 23:16 ` Grant Grundler
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=1314720193-26577-1-git-send-email-ming.lei@canonical.com \ --to=ming.lei-z7wlfzj8ewms+fvcfc7uqw@public.gmane.org \ --cc=greg-U8xfFu+wG4EAvxtiuMwx3w@public.gmane.org \ --cc=linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org \ --cc=linux-lFZ/pmaqli7XmaaqVzeoHQ@public.gmane.org \ --cc=linux-omap-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \ --cc=linux-usb-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \ --cc=stern-nwvwT67g6+6dFdvTe/nMLpVzexx5G7lz@public.gmane.org \ /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: linkBe 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.