From: James Bottomley <James.Bottomley@HansenPartnership.com> To: Pavel Machek <pavel@ucw.cz> Cc: Catalin Marinas <catalin.marinas@arm.com>, FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>, benh@kernel.crashing.org, linux@arm.linux.org.uk, mdharm-kernel@one-eyed-alien.net, linux-usb@vger.kernel.org, x0082077@ti.com, sshtylyov@ru.mvista.com, tom.leiming@gmail.com, bigeasy@linutronix.de, oliver@neukum.org, linux-kernel@vger.kernel.org, santosh.shilimkar@ti.com, greg@kroah.com, linux-arm-kernel@lists.infradead.org Subject: Re: USB mass storage and ARM cache coherency Date: Thu, 04 Mar 2010 19:51:52 +0530 [thread overview] Message-ID: <1267712512.31654.176.camel@mulgrave.site> (raw) In-Reply-To: <20100304135128.GA12191@atrey.karlin.mff.cuni.cz> On Thu, 2010-03-04 at 14:51 +0100, Pavel Machek wrote: > > On Wed, 2010-03-03 at 21:54 +0000, Pavel Machek wrote: > > > > With some drivers (those doing PIO) or subsystems (SCSI mass storage > > > > over USB HCD), there is no call to flush_dcache_page() for page cache > > > > pages, hence the ARM implementation of update_mmu_cache() doesn't flush > > > > the D-cache (and only invalidating the I-cache doesn't help). > > > > > > > > The viable solutions so far: > > > > > > > > 1. Implement a PIO mapping API similar to the DMA API which takes > > > > care of the D-cache flushing. This means that PIO drivers would > > > > need to be modified to use an API like pio_kmap()/pio_kunmap() > > > > before writing to a page cache page. > > > > 2. Invert the meaning of PG_arch_1 to denote a clean page. This > > > > means that by default newly allocated page cache pages are > > > > considered dirty and even if there isn't a call to > > > > flush_dcache_page(), update_mmu_cache() would flush the D-cache. > > > > This is the PowerPC approach. > > > > > > What about option > > > > > > 3. Forget about PG_arch_1 and always do the flush? > > > > > > How big is the performance impact? Note that current code does not > > > even *work* so working, 10% slower code will be an improvement. > > > > The driver fix is as simple as calling a flush_dcache_page() and I've > > been carrying such patches in my tree for some time now. The question is > > whether we need to do it in the driver or not (would need to update > > Documentation/cachetlb.txt as well). > > > > The reason I'm not in favour always doing the flush is that we penalise > > DMA drivers where there is no need for extra D-cache flushing (already > > handled by the DMA API; option 1 above is similar, just that it is meant > > for PIO usage). An ARM patch I proposed for inverting the meaning of > > PG_arch_1 also marks a page as clean in the dma_map_* functions. > > But you are not fixing driver bug, are you? Technically, he is. In the old days, most VI architectures were high end enough not to require PIO transfers. The only exception was an IDE driver used by sparc, which lead to the arch specific ide in/out string instructions, in which sparc actually did all the necessary flushing. So no other drivers than old IDE grew up with cache flushing in the PIO case (and almost no high end VI hardware had an IDE interface, so they rarely got implemented in the arch layer). However, recently, with the transition from old IDE to libata and the prevalence of ARM with more commodity hardware, the deficiency is becoming exposed. Even the PA8000 workstations now come with an IDE CD, which means we're starting to have problems with them as well. > Seems like ARM has requirement other architectures do not, that is > a) not documented anywhere > b) causes problems > > You could argue that performance improvement (how big is it, anyway?) > is worth it, but this should be agreed to by wider community... Performance is always worth it provided we don't sacrifice correctness. The thing which was discovered in this thread is basically that ARM is handling deferred flushing (for D/I coherency) in a slightly different way from everyone else ... once that's fixed, ARM will likely not have the D/I problem, but we'll still have the libata (and other PIO systems) D flushing issue. James
WARNING: multiple messages have this Message-ID (diff)
From: James.Bottomley@HansenPartnership.com (James Bottomley) To: linux-arm-kernel@lists.infradead.org Subject: USB mass storage and ARM cache coherency Date: Thu, 04 Mar 2010 19:51:52 +0530 [thread overview] Message-ID: <1267712512.31654.176.camel@mulgrave.site> (raw) In-Reply-To: <20100304135128.GA12191@atrey.karlin.mff.cuni.cz> On Thu, 2010-03-04 at 14:51 +0100, Pavel Machek wrote: > > On Wed, 2010-03-03 at 21:54 +0000, Pavel Machek wrote: > > > > With some drivers (those doing PIO) or subsystems (SCSI mass storage > > > > over USB HCD), there is no call to flush_dcache_page() for page cache > > > > pages, hence the ARM implementation of update_mmu_cache() doesn't flush > > > > the D-cache (and only invalidating the I-cache doesn't help). > > > > > > > > The viable solutions so far: > > > > > > > > 1. Implement a PIO mapping API similar to the DMA API which takes > > > > care of the D-cache flushing. This means that PIO drivers would > > > > need to be modified to use an API like pio_kmap()/pio_kunmap() > > > > before writing to a page cache page. > > > > 2. Invert the meaning of PG_arch_1 to denote a clean page. This > > > > means that by default newly allocated page cache pages are > > > > considered dirty and even if there isn't a call to > > > > flush_dcache_page(), update_mmu_cache() would flush the D-cache. > > > > This is the PowerPC approach. > > > > > > What about option > > > > > > 3. Forget about PG_arch_1 and always do the flush? > > > > > > How big is the performance impact? Note that current code does not > > > even *work* so working, 10% slower code will be an improvement. > > > > The driver fix is as simple as calling a flush_dcache_page() and I've > > been carrying such patches in my tree for some time now. The question is > > whether we need to do it in the driver or not (would need to update > > Documentation/cachetlb.txt as well). > > > > The reason I'm not in favour always doing the flush is that we penalise > > DMA drivers where there is no need for extra D-cache flushing (already > > handled by the DMA API; option 1 above is similar, just that it is meant > > for PIO usage). An ARM patch I proposed for inverting the meaning of > > PG_arch_1 also marks a page as clean in the dma_map_* functions. > > But you are not fixing driver bug, are you? Technically, he is. In the old days, most VI architectures were high end enough not to require PIO transfers. The only exception was an IDE driver used by sparc, which lead to the arch specific ide in/out string instructions, in which sparc actually did all the necessary flushing. So no other drivers than old IDE grew up with cache flushing in the PIO case (and almost no high end VI hardware had an IDE interface, so they rarely got implemented in the arch layer). However, recently, with the transition from old IDE to libata and the prevalence of ARM with more commodity hardware, the deficiency is becoming exposed. Even the PA8000 workstations now come with an IDE CD, which means we're starting to have problems with them as well. > Seems like ARM has requirement other architectures do not, that is > a) not documented anywhere > b) causes problems > > You could argue that performance improvement (how big is it, anyway?) > is worth it, but this should be agreed to by wider community... Performance is always worth it provided we don't sacrifice correctness. The thing which was discovered in this thread is basically that ARM is handling deferred flushing (for D/I coherency) in a slightly different way from everyone else ... once that's fixed, ARM will likely not have the D/I problem, but we'll still have the libata (and other PIO systems) D flushing issue. James
next prev parent reply other threads:[~2010-03-04 14:22 UTC|newest] Thread overview: 352+ messages / expand[flat|nested] mbox.gz Atom feed top 2010-01-29 14:34 USB mass storage and ARM cache coherency Catalin Marinas 2010-01-29 16:10 ` Oliver Neukum 2010-01-29 16:23 ` Ming Lei 2010-01-29 16:34 ` Catalin Marinas 2010-01-29 16:41 ` Oliver Neukum 2010-01-29 17:14 ` Catalin Marinas 2010-01-29 17:51 ` Sergei Shtylyov 2010-01-29 18:54 ` Matthew Dharm 2010-01-29 19:35 ` Greg KH 2010-02-01 13:49 ` Catalin Marinas 2010-02-01 17:29 ` Catalin Marinas 2010-02-01 20:14 ` Alan Stern 2010-02-02 4:24 ` Paul Mundt 2010-02-02 9:58 ` Catalin Marinas 2010-02-01 22:30 ` Andreas Mohr 2010-02-02 6:58 ` Oliver Neukum 2010-02-02 9:31 ` Florian Fainelli 2010-02-02 6:39 ` Paul Mundt 2010-02-02 11:05 ` Catalin Marinas 2010-02-02 11:15 ` Paul Mundt 2010-02-02 9:11 ` Sebastian Andrzej Siewior 2010-02-02 11:09 ` Catalin Marinas 2010-02-02 11:48 ` Oliver Neukum 2010-02-02 12:01 ` Catalin Marinas 2010-02-02 12:07 ` Oliver Neukum 2010-02-02 12:11 ` Andreas Mohr 2010-02-02 14:42 ` Clemens Ladisch 2010-02-02 14:52 ` Oliver Neukum 2010-02-02 15:10 ` Andreas Mohr 2010-02-02 15:34 ` Catalin Marinas 2010-02-02 20:38 ` Andreas Mohr 2010-02-02 12:39 ` Catalin Marinas 2010-02-02 13:08 ` Oliver Neukum 2010-02-02 14:34 ` Catalin Marinas 2010-02-02 17:11 ` Alan Stern 2010-02-02 17:20 ` Catalin Marinas 2010-02-02 21:52 ` Andreas Mohr 2010-02-03 15:15 ` Alan Stern 2010-02-08 6:55 ` Pavel Machek 2010-02-02 13:36 ` Ming Lei 2010-02-02 14:35 ` Catalin Marinas 2010-02-08 6:55 ` Pavel Machek 2010-02-08 6:55 ` Pavel Machek 2010-02-08 7:33 ` Andreas Mohr 2010-02-08 7:33 ` Andreas Mohr 2010-02-08 10:19 ` Catalin Marinas 2010-02-08 10:19 ` Catalin Marinas 2010-02-08 9:51 ` Catalin Marinas 2010-02-08 9:51 ` Catalin Marinas 2010-02-08 10:03 ` Andy Green 2010-02-08 10:03 ` Andy Green 2010-02-17 9:50 ` Sascha Hauer 2010-02-17 9:50 ` Sascha Hauer 2010-02-17 9:57 ` Andy Green 2010-02-17 9:57 ` Andy Green 2010-02-08 10:52 ` Pavel Machek 2010-02-08 10:52 ` Pavel Machek 2010-02-08 11:28 ` Catalin Marinas 2010-02-08 11:28 ` Catalin Marinas 2010-02-16 7:57 ` Shilimkar, Santosh 2010-02-16 7:57 ` Shilimkar, Santosh 2010-02-16 8:22 ` Oliver Neukum 2010-02-16 8:22 ` Oliver Neukum 2010-02-16 8:55 ` Shilimkar, Santosh 2010-02-16 8:55 ` Shilimkar, Santosh 2010-02-16 9:07 ` Oliver Neukum 2010-02-16 9:07 ` Oliver Neukum 2010-02-16 9:39 ` Russell King - ARM Linux 2010-02-16 9:39 ` Russell King - ARM Linux 2010-02-16 13:32 ` Oliver Neukum 2010-02-16 13:32 ` Oliver Neukum 2010-02-16 13:40 ` Shilimkar, Santosh 2010-02-16 13:40 ` Shilimkar, Santosh 2010-02-16 13:46 ` Oliver Neukum 2010-02-16 13:46 ` Oliver Neukum 2010-02-16 14:12 ` Shilimkar, Santosh 2010-02-16 14:12 ` Shilimkar, Santosh 2010-02-16 14:22 ` Oliver Neukum 2010-02-16 14:22 ` Oliver Neukum 2010-02-16 14:45 ` Shilimkar, Santosh 2010-02-16 14:45 ` Shilimkar, Santosh 2010-02-16 15:44 ` Alan Stern 2010-02-16 15:44 ` Alan Stern 2010-02-17 8:55 ` Shilimkar, Santosh 2010-02-17 8:55 ` Shilimkar, Santosh 2010-02-17 9:10 ` Oliver Neukum 2010-02-17 9:10 ` Oliver Neukum 2010-02-17 9:17 ` Shilimkar, Santosh 2010-02-17 9:17 ` Shilimkar, Santosh 2010-02-17 17:02 ` Alan Stern 2010-02-17 17:02 ` Alan Stern 2010-02-17 20:26 ` Russell King - ARM Linux 2010-02-17 20:26 ` Russell King - ARM Linux 2010-02-17 20:30 ` Gadiyar, Anand 2010-02-17 20:30 ` Gadiyar, Anand 2010-02-18 6:56 ` Oliver Neukum 2010-02-18 6:56 ` Oliver Neukum 2010-02-18 7:14 ` Gadiyar, Anand 2010-02-18 7:14 ` Gadiyar, Anand 2010-02-17 12:29 ` Jamie Lokier 2010-02-17 12:29 ` Jamie Lokier 2010-02-17 3:21 ` Ming Lei 2010-02-17 3:21 ` Ming Lei 2010-02-17 9:05 ` Benjamin Herrenschmidt 2010-02-17 9:05 ` Benjamin Herrenschmidt 2010-02-17 9:15 ` Oliver Neukum 2010-02-17 9:15 ` Oliver Neukum 2010-02-17 9:40 ` Benjamin Herrenschmidt 2010-02-17 9:40 ` Benjamin Herrenschmidt 2010-02-17 10:09 ` Oliver Neukum 2010-02-17 10:09 ` Oliver Neukum 2010-02-17 10:18 ` Benjamin Herrenschmidt 2010-02-17 10:18 ` Benjamin Herrenschmidt 2010-02-17 10:23 ` Oliver Neukum 2010-02-17 10:23 ` Oliver Neukum 2010-02-17 12:15 ` Benjamin Herrenschmidt 2010-02-17 12:15 ` Benjamin Herrenschmidt 2010-02-17 9:55 ` Russell King - ARM Linux 2010-02-17 9:55 ` Russell King - ARM Linux 2010-02-17 10:05 ` Benjamin Herrenschmidt 2010-02-17 10:05 ` Benjamin Herrenschmidt 2010-02-17 15:27 ` Catalin Marinas 2010-02-17 15:27 ` Catalin Marinas 2010-02-17 20:37 ` Benjamin Herrenschmidt 2010-02-17 20:37 ` Benjamin Herrenschmidt 2010-02-17 20:44 ` Russell King - ARM Linux 2010-02-17 20:44 ` Russell King - ARM Linux 2010-02-17 22:31 ` Benjamin Herrenschmidt 2010-02-17 22:31 ` Benjamin Herrenschmidt 2010-02-19 17:15 ` Catalin Marinas 2010-02-19 17:15 ` Catalin Marinas 2010-02-19 17:36 ` Catalin Marinas 2010-02-19 17:36 ` Catalin Marinas 2010-02-19 20:53 ` Oliver Neukum 2010-02-19 20:53 ` Oliver Neukum 2010-02-24 2:48 ` Benjamin Herrenschmidt 2010-02-24 2:48 ` Benjamin Herrenschmidt 2010-02-24 7:16 ` Oliver Neukum 2010-02-24 7:16 ` Oliver Neukum 2010-02-24 21:12 ` Benjamin Herrenschmidt 2010-02-24 21:12 ` Benjamin Herrenschmidt 2010-02-25 3:48 ` Oliver Neukum 2010-02-25 3:48 ` Oliver Neukum 2010-02-26 0:22 ` Benjamin Herrenschmidt 2010-02-26 0:22 ` Benjamin Herrenschmidt 2010-02-25 12:36 ` James Bottomley 2010-02-25 12:36 ` James Bottomley 2010-02-24 2:47 ` Benjamin Herrenschmidt 2010-02-24 2:47 ` Benjamin Herrenschmidt 2010-02-24 16:19 ` Alan Stern 2010-02-24 16:19 ` Alan Stern 2010-02-24 21:13 ` Benjamin Herrenschmidt 2010-02-24 21:13 ` Benjamin Herrenschmidt 2010-02-24 21:50 ` Alan Stern 2010-02-24 21:50 ` Alan Stern 2010-02-25 20:52 ` Benjamin Herrenschmidt 2010-02-25 20:52 ` Benjamin Herrenschmidt 2010-02-26 16:00 ` Catalin Marinas 2010-02-26 16:00 ` Catalin Marinas 2010-02-26 21:36 ` Benjamin Herrenschmidt 2010-02-26 21:36 ` Benjamin Herrenschmidt 2010-02-26 16:25 ` Catalin Marinas 2010-02-26 16:25 ` Catalin Marinas 2010-02-26 16:52 ` Alan Stern 2010-02-26 16:52 ` Alan Stern 2010-02-26 21:51 ` Benjamin Herrenschmidt 2010-02-26 21:51 ` Benjamin Herrenschmidt 2010-02-26 21:00 ` Russell King - ARM Linux 2010-02-26 21:00 ` Russell King - ARM Linux 2010-02-28 0:14 ` Benjamin Herrenschmidt 2010-02-28 0:14 ` Benjamin Herrenschmidt 2010-02-28 5:01 ` James Bottomley 2010-02-28 5:01 ` James Bottomley 2010-03-01 10:39 ` Catalin Marinas 2010-03-01 10:39 ` Catalin Marinas 2010-03-01 11:06 ` Russell King - ARM Linux 2010-03-01 11:06 ` Russell King - ARM Linux 2010-03-02 12:11 ` FUJITA Tomonori 2010-03-02 12:11 ` FUJITA Tomonori 2010-03-02 17:05 ` Catalin Marinas 2010-03-02 17:05 ` Catalin Marinas 2010-03-02 17:47 ` Catalin Marinas 2010-03-02 17:47 ` Catalin Marinas 2010-03-02 23:33 ` Benjamin Herrenschmidt 2010-03-02 23:33 ` Benjamin Herrenschmidt 2010-03-03 10:21 ` Catalin Marinas 2010-03-03 10:21 ` Catalin Marinas 2010-03-02 23:29 ` Benjamin Herrenschmidt 2010-03-02 23:29 ` Benjamin Herrenschmidt 2010-03-03 3:47 ` FUJITA Tomonori 2010-03-03 3:47 ` FUJITA Tomonori 2010-03-03 5:10 ` Benjamin Herrenschmidt 2010-03-03 5:10 ` Benjamin Herrenschmidt 2010-03-03 5:40 ` James Bottomley 2010-03-03 5:40 ` James Bottomley 2010-03-03 9:36 ` Russell King - ARM Linux 2010-03-03 9:36 ` Russell King - ARM Linux 2010-03-03 10:24 ` James Bottomley 2010-03-03 10:24 ` James Bottomley 2010-03-03 19:41 ` Russell King - ARM Linux 2010-03-03 19:41 ` Russell King - ARM Linux 2010-03-04 2:00 ` Benjamin Herrenschmidt 2010-03-04 2:00 ` Benjamin Herrenschmidt 2010-03-04 8:26 ` James Bottomley 2010-03-04 8:26 ` James Bottomley 2010-03-04 21:25 ` Benjamin Herrenschmidt 2010-03-04 21:25 ` Benjamin Herrenschmidt 2010-03-03 6:35 ` FUJITA Tomonori 2010-03-03 6:35 ` FUJITA Tomonori 2010-03-03 10:43 ` Catalin Marinas 2010-03-03 10:43 ` Catalin Marinas 2010-03-03 10:40 ` Catalin Marinas 2010-03-03 10:40 ` Catalin Marinas 2010-03-03 21:54 ` Pavel Machek 2010-03-03 21:54 ` Pavel Machek 2010-03-04 6:54 ` Wolfgang Mües 2010-03-04 9:31 ` Russell King - ARM Linux 2010-03-06 10:56 ` Wolfgang Mües 2010-03-06 11:05 ` Oliver Neukum 2010-03-06 19:44 ` Russell King - ARM Linux 2010-03-04 13:47 ` Catalin Marinas 2010-03-04 13:35 ` Catalin Marinas 2010-03-04 13:35 ` Catalin Marinas 2010-03-04 13:51 ` Pavel Machek 2010-03-04 13:51 ` Pavel Machek 2010-03-04 14:21 ` James Bottomley [this message] 2010-03-04 14:21 ` James Bottomley 2010-03-04 14:27 ` Russell King - ARM Linux 2010-03-04 14:27 ` Russell King - ARM Linux 2010-03-04 15:25 ` Catalin Marinas 2010-03-04 15:25 ` Catalin Marinas 2010-03-04 15:34 ` Russell King - ARM Linux 2010-03-04 15:34 ` Russell King - ARM Linux 2010-03-04 21:31 ` Benjamin Herrenschmidt 2010-03-04 21:31 ` Benjamin Herrenschmidt 2010-03-06 10:47 ` James Bottomley 2010-03-06 10:47 ` James Bottomley 2010-03-06 19:36 ` Russell King - ARM Linux 2010-03-06 19:36 ` Russell King - ARM Linux 2010-03-06 21:07 ` Benjamin Herrenschmidt 2010-03-06 21:07 ` Benjamin Herrenschmidt 2010-03-07 5:54 ` James Bottomley 2010-03-07 5:54 ` James Bottomley 2010-03-08 11:17 ` Catalin Marinas 2010-03-08 11:17 ` Catalin Marinas 2010-03-06 21:03 ` Benjamin Herrenschmidt 2010-03-06 21:03 ` Benjamin Herrenschmidt 2010-03-07 3:37 ` James Bottomley 2010-03-07 3:37 ` James Bottomley 2010-03-08 8:46 ` FUJITA Tomonori 2010-03-08 8:46 ` FUJITA Tomonori 2010-03-09 2:25 ` Benjamin Herrenschmidt 2010-03-09 2:25 ` Benjamin Herrenschmidt 2010-03-04 15:29 ` Catalin Marinas 2010-03-04 15:29 ` Catalin Marinas 2010-03-04 15:41 ` Paul Mundt 2010-03-04 15:41 ` Paul Mundt 2010-03-04 16:30 ` Russell King - ARM Linux 2010-03-04 16:30 ` Russell King - ARM Linux 2010-03-04 17:34 ` Catalin Marinas 2010-03-04 17:34 ` Catalin Marinas 2010-03-04 17:54 ` Russell King - ARM Linux 2010-03-04 17:54 ` Russell King - ARM Linux 2010-03-04 22:27 ` Andreas Mohr 2010-03-04 18:07 ` Catalin Marinas 2010-03-04 18:07 ` Catalin Marinas 2010-03-04 21:37 ` Benjamin Herrenschmidt 2010-03-04 21:37 ` Benjamin Herrenschmidt 2010-03-04 22:11 ` Catalin Marinas 2010-03-04 22:11 ` Catalin Marinas 2010-03-05 4:34 ` Benjamin Herrenschmidt 2010-03-05 4:34 ` Benjamin Herrenschmidt 2010-03-05 9:27 ` Catalin Marinas 2010-03-05 9:27 ` Catalin Marinas 2010-03-05 1:17 ` Paul Mundt 2010-03-05 1:17 ` Paul Mundt 2010-03-05 1:17 ` Paul Mundt 2010-03-05 4:44 ` Benjamin Herrenschmidt 2010-03-05 4:44 ` Benjamin Herrenschmidt 2010-03-05 4:44 ` Benjamin Herrenschmidt 2010-03-10 3:52 ` Paul Mundt 2010-03-10 3:52 ` Paul Mundt 2010-03-10 3:52 ` Paul Mundt 2010-03-11 21:44 ` Benjamin Herrenschmidt 2010-03-11 21:44 ` Benjamin Herrenschmidt 2010-03-11 21:44 ` Benjamin Herrenschmidt 2010-03-04 21:34 ` Benjamin Herrenschmidt 2010-03-04 21:34 ` Benjamin Herrenschmidt 2010-03-04 21:28 ` Benjamin Herrenschmidt 2010-03-04 21:28 ` Benjamin Herrenschmidt 2010-03-04 21:40 ` Russell King - ARM Linux 2010-03-04 21:40 ` Russell King - ARM Linux 2010-03-05 4:31 ` Benjamin Herrenschmidt 2010-03-05 4:31 ` Benjamin Herrenschmidt 2010-03-04 15:35 ` Catalin Marinas 2010-03-04 15:35 ` Catalin Marinas 2010-03-07 8:23 ` Pavel Machek 2010-03-07 8:23 ` Pavel Machek 2010-03-08 10:57 ` Catalin Marinas 2010-03-08 10:57 ` Catalin Marinas 2010-03-02 23:26 ` Benjamin Herrenschmidt 2010-03-02 23:26 ` Benjamin Herrenschmidt 2010-03-01 10:42 ` Catalin Marinas 2010-03-01 10:42 ` Catalin Marinas 2010-03-03 20:24 ` Jamie Lokier 2010-03-03 20:24 ` Jamie Lokier 2010-02-26 21:40 ` Benjamin Herrenschmidt 2010-02-26 21:40 ` Benjamin Herrenschmidt 2010-02-26 21:49 ` Russell King - ARM Linux 2010-02-26 21:49 ` Russell King - ARM Linux 2010-02-28 0:24 ` Benjamin Herrenschmidt 2010-02-28 0:24 ` Benjamin Herrenschmidt 2010-02-28 19:17 ` Pavel Machek 2010-02-28 19:17 ` Pavel Machek 2010-03-01 11:10 ` Catalin Marinas 2010-03-01 11:10 ` Catalin Marinas 2010-03-02 4:11 ` Benjamin Herrenschmidt 2010-03-02 4:11 ` Benjamin Herrenschmidt 2010-02-24 2:39 ` Benjamin Herrenschmidt 2010-02-24 2:39 ` Benjamin Herrenschmidt 2010-02-26 16:44 ` Catalin Marinas 2010-02-26 16:44 ` Catalin Marinas 2010-02-26 21:49 ` Benjamin Herrenschmidt 2010-02-26 21:49 ` Benjamin Herrenschmidt 2010-02-26 22:03 ` Russell King - ARM Linux 2010-02-26 22:03 ` Russell King - ARM Linux 2010-02-28 0:29 ` Benjamin Herrenschmidt 2010-02-28 0:29 ` Benjamin Herrenschmidt 2010-02-28 23:20 ` Catalin Marinas 2010-02-28 23:20 ` Catalin Marinas 2010-02-28 23:17 ` Catalin Marinas 2010-02-28 23:17 ` Catalin Marinas 2010-02-17 15:27 ` Catalin Marinas 2010-02-17 15:27 ` Catalin Marinas 2010-02-17 15:39 ` Catalin Marinas 2010-02-17 15:39 ` Catalin Marinas 2010-02-17 15:40 ` Catalin Marinas 2010-02-17 15:40 ` Catalin Marinas 2010-02-17 15:40 ` Catalin Marinas 2010-02-17 15:40 ` Catalin Marinas 2010-02-17 16:19 ` Catalin Marinas 2010-02-17 16:19 ` Catalin Marinas 2010-02-17 16:19 ` Re: " Catalin Marinas 2010-02-17 16:19 ` Catalin Marinas 2010-02-16 8:44 ` Russell King - ARM Linux 2010-02-16 8:44 ` Russell King - ARM Linux 2010-02-16 8:51 ` Gadiyar, Anand 2010-02-16 8:51 ` Gadiyar, Anand 2010-02-20 7:21 ` Pete Zaitcev 2010-02-20 7:21 ` Pete Zaitcev 2010-02-03 23:56 George Spelvin 2010-02-04 4:39 ` Paul Mundt
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=1267712512.31654.176.camel@mulgrave.site \ --to=james.bottomley@hansenpartnership.com \ --cc=benh@kernel.crashing.org \ --cc=bigeasy@linutronix.de \ --cc=catalin.marinas@arm.com \ --cc=fujita.tomonori@lab.ntt.co.jp \ --cc=greg@kroah.com \ --cc=linux-arm-kernel@lists.infradead.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-usb@vger.kernel.org \ --cc=linux@arm.linux.org.uk \ --cc=mdharm-kernel@one-eyed-alien.net \ --cc=oliver@neukum.org \ --cc=pavel@ucw.cz \ --cc=santosh.shilimkar@ti.com \ --cc=sshtylyov@ru.mvista.com \ --cc=tom.leiming@gmail.com \ --cc=x0082077@ti.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: 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.