Re: kvm PCI assignment VFIO ramblings
On Mon, 2011-08-22 at 17:52 -0700, aafabbri wrote: I'm not following you. You have to enforce group/iommu domain assignment whether you have the existing uiommu API, or if you change it to your proposed ioctl(inherit_iommu) API. The only change needed to VFIO here should be to make uiommu fd assignment happen on the groups instead of on device fds. That operation fails or succeeds according to the group semantics (all-or-none assignment/same uiommu). Ok, so I missed that part where you change uiommu to operate on group fd's rather than device fd's, my apologies if you actually wrote that down :-) It might be obvious ... bare with me I just flew back from the US and I am badly jet lagged ... So I see what you mean, however... I think the question is: do we force 1:1 iommu/group mapping, or do we allow arbitrary mapping (satisfying group constraints) as we do today. I'm saying I'm an existing user who wants the arbitrary iommu/group mapping ability and definitely think the uiommu approach is cleaner than the ioctl(inherit_iommu) approach. We considered that approach before but it seemed less clean so we went with the explicit uiommu context. Possibly, the question that interest me the most is what interface will KVM end up using. I'm also not terribly fan with the (perceived) discrepancy between using uiommu to create groups but using the group fd to actually do the mappings, at least if that is still the plan. If the separate uiommu interface is kept, then anything that wants to be able to benefit from the ability to put multiple devices (or existing groups) into such a meta group would need to be explicitly modified to deal with the uiommu APIs. I tend to prefer such meta groups as being something you create statically using a configuration interface, either via sysfs, netlink or ioctl's to a control vfio device driven by a simple command line tool (which can have the configuration stored in /etc and re-apply it at boot). That way, any program capable of exploiting VFIO groups will automatically be able to exploit those meta groups (or groups of groups) as well as long as they are supported on the system. If we ever have system specific constraints as to how such groups can be created, then it can all be handled at the level of that configuration tool without impact on whatever programs know how to exploit them via the VFIO interfaces. .../... If we in singleton-group land were building our own groups which were sets of devices sharing the IOMMU domains we wanted, I suppose we could do away with uiommu fds, but it sounds like the current proposal would create 20 singleton groups (x86 iommu w/o PCI bridges = all devices are partitionable endpoints). Asking me to ioctl(inherit) them together into a blob sounds worse than the current explicit uiommu API. I'd rather have an API to create super-groups (groups of groups) statically and then you can use such groups as normal groups using the same interface. That create/management process could be done via a simple command line utility or via sysfs banging, whatever... Cheers, Ben. ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
[PATCH] SPI: fix build with CONFIG_SPI_FSL_ESPI=m
When spi_fsl_espi is chosen to be built as a module, there is a build error because we test only CONFIG_SPI_FSL_ESPI in declaration of struct mpc8xxx_spi in drivers/spi/spi_fsl_lib.h. We need to add a test for CONFIG_SPI_FSL_ESPI_MODULE too. The error looks like: drivers/spi/spi_fsl_espi.c: In function 'fsl_espi_bufs': drivers/spi/spi_fsl_espi.c:232: error: 'struct mpc8xxx_spi' has no member named 'len' ... Signed-off-by: Jiri Slaby jsl...@suse.cz --- drivers/spi/spi-fsl-lib.h |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/spi/spi-fsl-lib.h b/drivers/spi/spi-fsl-lib.h index cbe881b..97968de 100644 --- a/drivers/spi/spi-fsl-lib.h +++ b/drivers/spi/spi-fsl-lib.h @@ -28,7 +28,7 @@ struct mpc8xxx_spi { /* rx tx bufs from the spi_transfer */ const void *tx; void *rx; -#ifdef CONFIG_SPI_FSL_ESPI +#if defined(CONFIG_SPI_FSL_ESPI) || defined(CONFIG_SPI_FSL_ESPI_MODULE) int len; #endif -- 1.7.6 ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH v3] mtd/nand : workaround for Freescale FCM to support large-page Nand chip
LiuShuo a écrit : 于 2011年08月23日 00:19, Scott Wood 写道: On 08/22/2011 11:13 AM, Matthieu CASTET wrote: Scott Wood a écrit : To eliminate it we'd need to do an extra data transfer without reissuing the command, which Shuo was unable to get to work. That's weird because our controller seems quite flexible [1]. Something like that should work ? out_be32(lbc-fir, (FIR_OP_CM2 FIR_OP0_SHIFT) | (FIR_OP_CA FIR_OP1_SHIFT) | (FIR_OP_PA FIR_OP2_SHIFT) | (FIR_OP_WB FIR_OP3_SHIFT)); refill FCM buffer with next 2k data out_be32(lbc-fir, (FIR_OP_WB FIR_OP3_SHIFT) | (FIR_OP_CM3 FIR_OP4_SHIFT) | (FIR_OP_CW1 FIR_OP5_SHIFT) | (FIR_OP_RS FIR_OP6_SHIFT)); Something like that is what I originally suggested, but Shuo said it didn't work (even in theory, it requires a CE-don't-care NAND chip, since bus atomicity is broken). Shuo, what specifically did you try, and what did you see happen? -Scott First, if we want to read 4K data with once command issuing, we can't use HW_ECC. Yes, but as ivan said doesn't the cost of 2 read isn't bigger than software ecc ? Even if we use SW_ECC, we always get lots of weird '0xFF's between 1st 2k and 2nd 2k data. Did you understand where those 0xff comes (what's the size of them. Doesn't the controller try to insert spare aera ?) Could you detail the sequence you used ? Matthieu ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH v3] mtd/nand : workaround for Freescale FCM to support large-page Nand chip
于 2011年08月19日 00:25, Scott Wood 写道: On 08/17/2011 09:33 PM, b35...@freescale.com wrote: From: Liu Shuob35...@freescale.com Freescale FCM controller has a 2K size limitation of buffer RAM. In order to support the Nand flash chip whose page size is larger than 2K bytes, we divide a page into multi-2K pages for MTD layer driver. In that case, we force to set the page size to 2K bytes. We convert the page address of MTD layer driver to a real page address in flash chips and a column index in fsl_elbc driver. We can issue any column address by UA instruction of elbc controller. NOTE: Due to there is a limitation of 'Number of Partial Program Cycles in the Same Page (NOP)', the flash chip which is supported by this workaround have to meet below conditions. 1. page size is not greater than 4KB 2. 1) if main area and spare area have independent NOPs: main area NOP:=3 spare area NOP:=2? How often are the NOPs split like this? 2) if main area and spare area have a common NOP: NOP :=4 This depends on how the flash is used. If you treat it as a NOP1 flash (e.g. run ubifs rather than jffs2), then you need NOP2 for a 4K chip and NOP4 for an 8K chip. OTOH, if you would be making full use of NOP4 on a real 2K chip, you'll need NOP8 for a 4K chip. The NOP restrictions should be documented in the code itself, not just in the git changelog. Maybe print it to the console when this hack is used, along with the NOP value read from the ID. We can't read the NOP from the ID on any chip. Some chips don't give this infomation.(e.g. Micron MT29F4G08BAC) So it is hard to determine whether the probe() should fail in the code. Maybe we will always print the NOP restrictions when this hack is used, let the customers select how to use the flash on their board. -LiuShuo If it's less than 4 for 4K or 8 for 8K, also print a message saying not to use jffs2 (does yaffs2 do similar things?). If it's less than 2 for 4K or 4 for 8K, the probe should fail. -Scott ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: linux-next: boot test failure (net tree)
On Mon, 2011-08-22 at 21:02 -0700, David Miller wrote: From: Arnaud Lacombe lacom...@gmail.com Date: Mon, 22 Aug 2011 23:50:02 -0400 Are you implying we need some kind of way to migrate config ? The issue is that the dependencies for every single ethernet driver have changed. Some dependencies have been dropped (f.e. NETDEV_1 and some have been added (f.e. ETHERNET, NET_VENDOR_) So right now an automated (non-prompted, default to no on all new options) run on an existing config results in all ethernet drivers getting disabled because the new dependencies don't get enabled. This wouldn't be so bad if it was just one or two drivers, but in this case it's every single ethernet driver which will have and hit this problem. Ok, I have patch which will resolve the issue. It is the last patch in the series I am about to send out. What this patch does is set the new Kconfig options to Y, so that current defconfig's can build driver's that are currently set to build. This will fix the issue, I have confirmed this with the x86_64 defconfig. It will be nice that eventually all configs get updated so that not all the NET_VENDOR_* tags have to be enabled, but understandably this is the best way to ensure that current defconfig's will compile all expected drivers. signature.asc Description: This is a digitally signed message part ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH] SPI: fix build with CONFIG_SPI_FSL_ESPI=m
On 08/23/2011 09:59 AM, Jiri Slaby wrote: When spi_fsl_espi is chosen to be built as a module, there is a build error because we test only CONFIG_SPI_FSL_ESPI in declaration of struct mpc8xxx_spi in drivers/spi/spi_fsl_lib.h. We need to add a test for CONFIG_SPI_FSL_ESPI_MODULE too. The error looks like: drivers/spi/spi_fsl_espi.c: In function 'fsl_espi_bufs': drivers/spi/spi_fsl_espi.c:232: error: 'struct mpc8xxx_spi' has no member named 'len' ... Signed-off-by: Jiri Slaby jsl...@suse.cz --- drivers/spi/spi-fsl-lib.h |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/spi/spi-fsl-lib.h b/drivers/spi/spi-fsl-lib.h index cbe881b..97968de 100644 --- a/drivers/spi/spi-fsl-lib.h +++ b/drivers/spi/spi-fsl-lib.h @@ -28,7 +28,7 @@ struct mpc8xxx_spi { /* rx tx bufs from the spi_transfer */ const void *tx; void *rx; -#ifdef CONFIG_SPI_FSL_ESPI +#if defined(CONFIG_SPI_FSL_ESPI) || defined(CONFIG_SPI_FSL_ESPI_MODULE) int len; #endif Oh, and there are still link errors: ERROR: mpc8xxx_spi_tx_buf_u32 [drivers/spi/spi_fsl_spi.ko] undefined! ERROR: mpc8xxx_spi_rx_buf_u32 [drivers/spi/spi_fsl_spi.ko] undefined! ERROR: mpc8xxx_spi_tx_buf_u16 [drivers/spi/spi_fsl_spi.ko] undefined! ERROR: mpc8xxx_spi_rx_buf_u16 [drivers/spi/spi_fsl_spi.ko] undefined! ERROR: mpc8xxx_spi_tx_buf_u8 [drivers/spi/spi_fsl_spi.ko] undefined! ERROR: mpc8xxx_spi_rx_buf_u8 [drivers/spi/spi_fsl_spi.ko] undefined! ERROR: of_mpc8xxx_spi_probe [drivers/spi/spi_fsl_spi.ko] undefined! ERROR: mpc8xxx_spi_strmode [drivers/spi/spi_fsl_spi.ko] undefined! ERROR: mpc8xxx_spi_probe [drivers/spi/spi_fsl_spi.ko] undefined! ERROR: mpc8xxx_spi_remove [drivers/spi/spi_fsl_spi.ko] undefined! ERROR: to_of_pinfo [drivers/spi/spi_fsl_spi.ko] undefined! ERROR: mpc8xxx_spi_tx_buf_u32 [drivers/spi/spi_fsl_espi.ko] undefined! ERROR: mpc8xxx_spi_rx_buf_u32 [drivers/spi/spi_fsl_espi.ko] undefined! ERROR: of_mpc8xxx_spi_probe [drivers/spi/spi_fsl_espi.ko] undefined! ERROR: mpc8xxx_spi_probe [drivers/spi/spi_fsl_espi.ko] undefined! ERROR: mpc8xxx_spi_remove [drivers/spi/spi_fsl_espi.ko] undefined! The functions are not exported... Should I export all those or deny CONFIG_SPI_FSL_ESPI=m? thanks, -- js ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH 1/2] [hw-breakpoint] Use generic hw-breakpoint interfaces for new PPC ptrace flags
On Tue, Aug 23, 2011 at 03:08:50PM +1000, David Gibson wrote: On Fri, Aug 19, 2011 at 01:21:36PM +0530, K.Prasad wrote: PPC_PTRACE_GETHWDBGINFO, PPC_PTRACE_SETHWDEBUG and PPC_PTRACE_DELHWDEBUG are PowerPC specific ptrace flags that use the watchpoint register. While they are targeted primarily towards BookE users, user-space applications such as GDB have started using them for BookS too. This patch enables the use of generic hardware breakpoint interfaces for these new flags. The version number of the associated data structures ppc_hw_breakpoint and ppc_debug_info is incremented to denote new semantics. So, the structure itself doesn't seem to have been extended. I don't understand what the semantic difference is - your patch comment needs to explain this clearly. We had a request to extend the structure but thought it was dangerous to do so. For instance if the user-space used version1 of the structure, while kernel did a copy_to_user() pertaining to version2, then we'd run into problems. Unfortunately the ptrace flags weren't designed to accept a version number as input from the user through the PPC_PTRACE_GETHWDBGINFO flag (which would have solved this issue). I'll add a comment w.r.t change in semantics - such as the ability to accept 'range' breakpoints in BookS. Apart from the usual benefits of using generic hw-breakpoint interfaces, these changes allow debuggers (such as GDB) to use a common set of ptrace flags for their watchpoint needs and allow more precise breakpoint specification (length of the variable can be specified). What is the mechanism for implementing the range breakpoint on book3s? The hw-breakpoint interface, accepts length as an argument in BookS (any value = 8 Bytes) and would filter out extraneous interrupts arising out of accesses outside the range comprising addr, addr + len inside hw_breakpoint_handler function. We put that ability to use here. [Edjunior: Identified an issue in the patch with the sanity check for version numbers] Tested-by: Edjunior Barbosa Machado emach...@linux.vnet.ibm.com Signed-off-by: K.Prasad pra...@linux.vnet.ibm.com --- Documentation/powerpc/ptrace.txt | 16 ++ arch/powerpc/kernel/ptrace.c | 104 +++--- 2 files changed, 112 insertions(+), 8 deletions(-) diff --git a/Documentation/powerpc/ptrace.txt b/Documentation/powerpc/ptrace.txt index f4a5499..97301ae 100644 --- a/Documentation/powerpc/ptrace.txt +++ b/Documentation/powerpc/ptrace.txt @@ -127,6 +127,22 @@ Some examples of using the structure to: p.addr2 = (uint64_t) end_range; p.condition_value = 0; +- set a watchpoint in server processors (BookS) using version 2 + + p.version = 2; + p.trigger_type= PPC_BREAKPOINT_TRIGGER_RW; + p.addr_mode = PPC_BREAKPOINT_MODE_RANGE_INCLUSIVE; + or + p.addr_mode = PPC_BREAKPOINT_MODE_RANGE_EXACT; + + p.condition_mode = PPC_BREAKPOINT_CONDITION_NONE; + p.addr= (uint64_t) begin_range; + /* For PPC_BREAKPOINT_MODE_RANGE_INCLUSIVE addr2 needs to be specified, where + * addr2 - addr = 8 Bytes. + */ + p.addr2 = (uint64_t) end_range; + p.condition_value = 0; + 3. PTRACE_DELHWDEBUG Takes an integer which identifies an existing breakpoint or watchpoint diff --git a/arch/powerpc/kernel/ptrace.c b/arch/powerpc/kernel/ptrace.c index 05b7dd2..18d28b6 100644 --- a/arch/powerpc/kernel/ptrace.c +++ b/arch/powerpc/kernel/ptrace.c @@ -1339,11 +1339,17 @@ static int set_dac_range(struct task_struct *child, static long ppc_set_hwdebug(struct task_struct *child, struct ppc_hw_breakpoint *bp_info) { +#ifdef CONFIG_HAVE_HW_BREAKPOINT + int ret, len = 0; + struct thread_struct *thread = (child-thread); + struct perf_event *bp; + struct perf_event_attr attr; +#endif /* CONFIG_HAVE_HW_BREAKPOINT */ I'm confused. This compiled before on book3s, and I don't see any changes to Makefile or Kconfig in the patch that will result in this code compiling when it previously didn't Why are these new guards added? The code is guarded using the CONFIG_ flags for two reasons. a) We don't want the code to be included for BookE and other architectures. b) In BookS, we're now adding a new ability based on whether CONFIG_HAVE_HW_BREAKPOINT is defined. Presently this config option is kept on by default, however there are plans to make this a config-time option. #ifndef CONFIG_PPC_ADV_DEBUG_REGS unsigned long dabr; #endif - if (bp_info-version != 1) + if ((bp_info-version != 1) (bp_info-version != 2)) return -ENOTSUPP; #ifdef CONFIG_PPC_ADV_DEBUG_REGS /* @@ -1382,13 +1388,9 @@ static long ppc_set_hwdebug(struct task_struct *child, */ if ((bp_info-trigger_type PPC_BREAKPOINT_TRIGGER_RW) == 0 ||
Re: [PATCH 2/2] [PowerPC Book3E] Introduce new ptrace debug feature flag
On Tue, Aug 23, 2011 at 03:09:31PM +1000, David Gibson wrote: On Fri, Aug 19, 2011 at 01:23:38PM +0530, K.Prasad wrote: While PPC_PTRACE_SETHWDEBUG ptrace flag in PowerPC accepts PPC_BREAKPOINT_MODE_EXACT mode of breakpoint, the same is not intimated to the user-space debuggers (like GDB) who may want to use it. Hence we introduce a new PPC_DEBUG_FEATURE_DATA_BP_EXACT flag which will be populated on the features member of struct ppc_debug_info to advertise support for the same on Book3E PowerPC processors. I thought the idea was that the BP_EXACT mode was the default - if the new interface was supported at all, then BP_EXACT was always supported. So, why do you need a new flag? Yes, BP_EXACT was always supported but not advertised through PPC_PTRACE_GETHWDBGINFO. We're now doing that. Thanks, K.Prasad ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH v3] mtd/nand : workaround for Freescale FCM to support large-page Nand chip
于 2011年08月23日 16:14, Matthieu CASTET 写道: LiuShuo a écrit : 于 2011年08月23日 00:19, Scott Wood 写道: On 08/22/2011 11:13 AM, Matthieu CASTET wrote: Scott Wood a écrit : To eliminate it we'd need to do an extra data transfer without reissuing the command, which Shuo was unable to get to work. That's weird because our controller seems quite flexible [1]. Something like that should work ? out_be32(lbc-fir, (FIR_OP_CM2 FIR_OP0_SHIFT) | (FIR_OP_CA FIR_OP1_SHIFT) | (FIR_OP_PA FIR_OP2_SHIFT) | (FIR_OP_WB FIR_OP3_SHIFT)); refill FCM buffer with next 2k data out_be32(lbc-fir, (FIR_OP_WB FIR_OP3_SHIFT) | (FIR_OP_CM3 FIR_OP4_SHIFT) | (FIR_OP_CW1 FIR_OP5_SHIFT) | (FIR_OP_RS FIR_OP6_SHIFT)); Something like that is what I originally suggested, but Shuo said it didn't work (even in theory, it requires a CE-don't-care NAND chip, since bus atomicity is broken). Shuo, what specifically did you try, and what did you see happen? -Scott First, if we want to read 4K data with once command issuing, we can't use HW_ECC. Yes, but as ivan said doesn't the cost of 2 read isn't bigger than software ecc ? Even if we use SW_ECC, we always get lots of weird '0xFF's between 1st 2k and 2nd 2k data. Did you understand where those 0xff comes (what's the size of them. Doesn't the controller try to insert spare aera ?) I don't understand. I set FBCR to 2048, the controller will read the main area without spare area. But the size of them is nearly spare area size( more or less a few bytes). I can't guess the behavior of the controller then, so I select another way. Could you try to do it and explain how those 0xff comes ? Could you detail the sequence you used ? First half : out_be32(lbc-fbcr, 2048); out_be32(lbc-fir, (FIR_OP_CM0 FIR_OP0_SHIFT) | (FIR_OP_CA FIR_OP1_SHIFT) | (FIR_OP_PA FIR_OP2_SHIFT) | (FIR_OP_CM1 FIR_OP3_SHIFT) | (FIR_OP_RBW FIR_OP4_SHIFT)); Sencond half : out_be32(lbc-fbcr, 2048); out_be32(lbc-fir, (FIR_OP_RB FIR_OP0_SHIFT) | (FIR_OP_RBW FIR_OP1_SHIFT)); -Liu Shuo Matthieu ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH v3] mtd/nand : workaround for Freescale FCM to support large-page Nand chip
LiuShuo a écrit : 于 2011年08月19日 00:25, Scott Wood 写道: On 08/17/2011 09:33 PM, b35...@freescale.com wrote: From: Liu Shuob35...@freescale.com Freescale FCM controller has a 2K size limitation of buffer RAM. In order to support the Nand flash chip whose page size is larger than 2K bytes, we divide a page into multi-2K pages for MTD layer driver. In that case, we force to set the page size to 2K bytes. We convert the page address of MTD layer driver to a real page address in flash chips and a column index in fsl_elbc driver. We can issue any column address by UA instruction of elbc controller. NOTE: Due to there is a limitation of 'Number of Partial Program Cycles in the Same Page (NOP)', the flash chip which is supported by this workaround have to meet below conditions. 1. page size is not greater than 4KB 2. 1) if main area and spare area have independent NOPs: main area NOP:=3 spare area NOP:=2? How often are the NOPs split like this? 2) if main area and spare area have a common NOP: NOP :=4 This depends on how the flash is used. If you treat it as a NOP1 flash (e.g. run ubifs rather than jffs2), then you need NOP2 for a 4K chip and NOP4 for an 8K chip. OTOH, if you would be making full use of NOP4 on a real 2K chip, you'll need NOP8 for a 4K chip. The NOP restrictions should be documented in the code itself, not just in the git changelog. Maybe print it to the console when this hack is used, along with the NOP value read from the ID. We can't read the NOP from the ID on any chip. Some chips don't give this infomation.(e.g. Micron MT29F4G08BAC) Doesn't the micron chip provide it with onfi info ? Matthieu ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH v3] mtd/nand : workaround for Freescale FCM to support large-page Nand chip
LiuShuo a écrit : 于 2011年08月23日 16:14, Matthieu CASTET 写道: LiuShuo a écrit : 于 2011年08月23日 00:19, Scott Wood 写道: On 08/22/2011 11:13 AM, Matthieu CASTET wrote: Scott Wood a écrit : To eliminate it we'd need to do an extra data transfer without reissuing the command, which Shuo was unable to get to work. That's weird because our controller seems quite flexible [1]. Something like that should work ? out_be32(lbc-fir, (FIR_OP_CM2 FIR_OP0_SHIFT) | (FIR_OP_CA FIR_OP1_SHIFT) | (FIR_OP_PA FIR_OP2_SHIFT) | (FIR_OP_WB FIR_OP3_SHIFT)); refill FCM buffer with next 2k data out_be32(lbc-fir, (FIR_OP_WB FIR_OP3_SHIFT) | (FIR_OP_CM3 FIR_OP4_SHIFT) | (FIR_OP_CW1 FIR_OP5_SHIFT) | (FIR_OP_RS FIR_OP6_SHIFT)); Something like that is what I originally suggested, but Shuo said it didn't work (even in theory, it requires a CE-don't-care NAND chip, since bus atomicity is broken). Shuo, what specifically did you try, and what did you see happen? -Scott First, if we want to read 4K data with once command issuing, we can't use HW_ECC. Yes, but as ivan said doesn't the cost of 2 read isn't bigger than software ecc ? Even if we use SW_ECC, we always get lots of weird '0xFF's between 1st 2k and 2nd 2k data. Did you understand where those 0xff comes (what's the size of them. Doesn't the controller try to insert spare aera ?) I don't understand. I set FBCR to 2048, the controller will read the main area without spare area. But the size of them is nearly spare area size( more or less a few bytes).. I can't guess the behavior of the controller then, so I select another way. Could you try to do it and explain how those 0xff comes ? Could you detail the sequence you used ? First half : out_be32(lbc-fbcr, 2048); shouldn't you read 2k+64 here ? At the end you want 4k plus spare aera (128) ? out_be32(lbc-fir, (FIR_OP_CM0 FIR_OP0_SHIFT) | (FIR_OP_CA FIR_OP1_SHIFT) | (FIR_OP_PA FIR_OP2_SHIFT) | (FIR_OP_CM1 FIR_OP3_SHIFT) | (FIR_OP_RBW FIR_OP4_SHIFT)); Sencond half : out_be32(lbc-fbcr, 2048); out_be32(lbc-fir, (FIR_OP_RB FIR_OP0_SHIFT) | (FIR_OP_RBW FIR_OP1_SHIFT)); Why do you do FIR_OP_RBW ? FIR_OP_RB already fetch the data. Matthieu ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
On Mon, Aug 22, 2011 at 08:52:18PM -0400, aafabbri wrote: You have to enforce group/iommu domain assignment whether you have the existing uiommu API, or if you change it to your proposed ioctl(inherit_iommu) API. The only change needed to VFIO here should be to make uiommu fd assignment happen on the groups instead of on device fds. That operation fails or succeeds according to the group semantics (all-or-none assignment/same uiommu). That is makes uiommu basically the same as the meta-groups, right? Joerg -- AMD Operating System Research Center Advanced Micro Devices GmbH Einsteinring 24 85609 Dornach General Managers: Alberto Bozzo, Andrew Bowd Registration: Dornach, Landkr. Muenchen; Registerger. Muenchen, HRB Nr. 43632 ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
On Tue, Aug 23, 2011 at 02:54:43AM -0400, Benjamin Herrenschmidt wrote: Possibly, the question that interest me the most is what interface will KVM end up using. I'm also not terribly fan with the (perceived) discrepancy between using uiommu to create groups but using the group fd to actually do the mappings, at least if that is still the plan. If the separate uiommu interface is kept, then anything that wants to be able to benefit from the ability to put multiple devices (or existing groups) into such a meta group would need to be explicitly modified to deal with the uiommu APIs. I tend to prefer such meta groups as being something you create statically using a configuration interface, either via sysfs, netlink or ioctl's to a control vfio device driven by a simple command line tool (which can have the configuration stored in /etc and re-apply it at boot). Hmm, I don't think that these groups are static for the systems run-time. They only exist for the lifetime of a guest per default, at least on x86. Thats why I prefer to do this grouping using VFIO and not some sysfs interface (which would be the third interface beside the ioctls and netlink a VFIO user needs to be aware of). Doing this in the ioctl interface just makes things easier. Joerg -- AMD Operating System Research Center Advanced Micro Devices GmbH Einsteinring 24 85609 Dornach General Managers: Alberto Bozzo, Andrew Bowd Registration: Dornach, Landkr. Muenchen; Registerger. Muenchen, HRB Nr. 43632 ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
[PATCH] powerpc: fixup QE_General4 errata
QE_General4 should only round up the divisor iff divisor is 3. Rounding up lower divisors makes the error too big, causing USB on MPC832x to fail. Signed-off-by: Joakim Tjernlund joakim.tjernl...@transmode.se --- arch/powerpc/sysdev/qe_lib/qe.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/arch/powerpc/sysdev/qe_lib/qe.c b/arch/powerpc/sysdev/qe_lib/qe.c index 093e0ae..5399316 100644 --- a/arch/powerpc/sysdev/qe_lib/qe.c +++ b/arch/powerpc/sysdev/qe_lib/qe.c @@ -216,7 +216,7 @@ int qe_setbrg(enum qe_clock brg, unsigned int rate, unsigned int multiplier) /* Errata QE_General4, which affects some MPC832x and MPC836x SOCs, says that the BRG divisor must be even if you're not using divide-by-16 mode. */ - if (!div16 (divisor 1)) + if (!div16 (divisor 1) (divisor 3)) divisor++; tempval = ((divisor - 1) QE_BRGC_DIVISOR_SHIFT) | -- 1.7.3.4 ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
[PATCH] usb: Allocate pram dynamically.
MPC832x does not have enough MURAM to do fixed MURAM allocation. Change to dynamic allocation. Signed-off-by: Joakim Tjernlund joakim.tjernl...@transmode.se --- drivers/usb/host/fhci-hcd.c |5 - 1 files changed, 4 insertions(+), 1 deletions(-) diff --git a/drivers/usb/host/fhci-hcd.c b/drivers/usb/host/fhci-hcd.c index c7c8392..98adbe8 100644 --- a/drivers/usb/host/fhci-hcd.c +++ b/drivers/usb/host/fhci-hcd.c @@ -622,12 +622,15 @@ static int __devinit of_fhci_probe(struct of_device *ofdev, goto err_pram; } - pram_addr = cpm_muram_alloc_fixed(iprop[2], FHCI_PRAM_SIZE); + pram_addr = cpm_muram_alloc(FHCI_PRAM_SIZE, 64); if (IS_ERR_VALUE(pram_addr)) { dev_err(dev, failed to allocate usb pram\n); ret = -ENOMEM; goto err_pram; } + + qe_issue_cmd(QE_ASSIGN_PAGE_TO_DEVICE, QE_CR_SUBBLOCK_USB, +QE_CR_PROTOCOL_UNSPECIFIED, pram_addr); fhci-pram = cpm_muram_addr(pram_addr); /* GPIOs and pins */ -- 1.7.3.4 ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
[PATCH] perf: fix build for PowerPC with uclibc toolchains
libio.h is not provided by uClibc, in order to be able to test the definition of __UCLIBC__ we need to include stdlib.h, which also includes stddef.h, providing the definition of 'NULL'. Signed-off-by: Florian Fainelli flor...@openwrt.org --- FYI, I submitted the exact same patch for ARM: https://patchwork.kernel.org/patch/1049152/ diff --git a/tools/perf/arch/powerpc/util/dwarf-regs.c b/tools/perf/arch/powerpc/util/dwarf-regs.c index 48ae0c5..7cdd61d 100644 --- a/tools/perf/arch/powerpc/util/dwarf-regs.c +++ b/tools/perf/arch/powerpc/util/dwarf-regs.c @@ -9,7 +9,10 @@ * 2 of the License, or (at your option) any later version. */ +#include stdlib.h +#ifndef __UCLIBC__ #include libio.h +#endif #include dwarf-regs.h -- 1.7.4.1 ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
RE: [PATCH] RapidIO: Fix use of non-compatible registers
Andrew Morton a...@linux-foundation.org wrote: On Tue, 26 Jul 2011 14:07:26 -0400 Alexandre Bounine alexandre.boun...@idt.com wrote: Replace/remove use of RIO v.1.2 registers/bits that are not forward- compatible with newer versions of RapidIO specification. RapidIO specification v. 1.3 removed Write Port CSR, Doorbell CSR, Mailbox CSR and Mailbox and Doorbell bits of the PEF CAR. . You did a cc:stable but provided no reason (that I can understand) for backporting the patch. Please explain why the problem is sufficiently serious to warrant this action. My reason for this is that use of removed (since RIO v.1.3) register bits affects users of currently available 1.3 and 2.x compliant devices who may use not so recent kernel versions. Removing checks for unsupported bits makes corresponding routines compatible with all versions of RapidIO specification. Therefore, backporting makes stable kernel versions compliant with RIO v.1.3 and later as well. ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH] usb: Allocate pram dynamically.
On Tue, Aug 23, 2011 at 02:38:41PM +0200, Joakim Tjernlund wrote: MPC832x does not have enough MURAM to do fixed MURAM allocation. Change to dynamic allocation. Signed-off-by: Joakim Tjernlund joakim.tjernl...@transmode.se Acked-by: Anton Vorontsov cbouatmai...@gmail.com Thanks! p.s. You probably want to send this to Greg KH, + Cc linux-usb mailing list. -- Anton Vorontsov Email: cbouatmai...@gmail.com ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
On Mon, Aug 22, 2011 at 05:03:53PM -0400, Benjamin Herrenschmidt wrote: I am in favour of /dev/vfio/$GROUP. If multiple devices should be assigned to a guest, there can also be an ioctl to bind a group to an address-space of another group (certainly needs some care to not allow that both groups belong to different processes). Btw, a problem we havn't talked about yet entirely is driver-deassignment. User space can decide to de-assign the device from vfio while a fd is open on it. With PCI there is no way to let this fail (the .release function returns void last time i checked). Is this a problem, and yes, how we handle that? We can treat it as a hard unplug (like a cardbus gone away). IE. Dispose of the direct mappings (switch to MMIO emulation) and return all ff's from reads ( ignore writes). Then send an unplug event via whatever mechanism the platform provides (ACPI hotplug controller on x86 for example, we haven't quite sorted out what to do on power for hotplug yet). Hmm, good idea. But as far as I know the hotplug-event needs to be in the guest _before_ the device is actually unplugged (so that the guest can unbind its driver first). That somehow brings back the sleep-idea and the timeout in the .release function. Joerg -- AMD Operating System Research Center Advanced Micro Devices GmbH Einsteinring 24 85609 Dornach General Managers: Alberto Bozzo, Andrew Bowd Registration: Dornach, Landkr. Muenchen; Registerger. Muenchen, HRB Nr. 43632 ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
On Mon, Aug 22, 2011 at 03:17:00PM -0400, Alex Williamson wrote: On Mon, 2011-08-22 at 19:25 +0200, Joerg Roedel wrote: I am in favour of /dev/vfio/$GROUP. If multiple devices should be assigned to a guest, there can also be an ioctl to bind a group to an address-space of another group (certainly needs some care to not allow that both groups belong to different processes). That's an interesting idea. Maybe an interface similar to the current uiommu interface, where you open() the 2nd group fd and pass the fd via ioctl to the primary group. IOMMUs that don't support this would fail the attach device callback, which would fail the ioctl to bind them. It will need to be designed so any group can be removed from the super-set and the remaining group(s) still works. This feels like something that can be added after we get an initial implementation. Handling it through fds is a good idea. This makes sure that everything belongs to one process. I am not really sure yet if we go the way to just bind plain groups together or if we create meta-groups. The meta-groups thing seems somewhat cleaner, though. Btw, a problem we havn't talked about yet entirely is driver-deassignment. User space can decide to de-assign the device from vfio while a fd is open on it. With PCI there is no way to let this fail (the .release function returns void last time i checked). Is this a problem, and yes, how we handle that? The current vfio has the same problem, we can't unbind a device from vfio while it's attached to a guest. I think we'd use the same solution too; send out a netlink packet for a device removal and have the .remove call sleep on a wait_event(, refcnt == 0). We could also set a timeout and SIGBUS the PIDs holding the device if they don't return it willingly. Thanks, Putting the process to sleep (which would be uninterruptible) seems bad. The process would sleep until the guest releases the device-group, which can take days or months. The best thing (and the most intrusive :-) ) is to change PCI core to allow unbindings to fail, I think. But this probably further complicates the way to upstream VFIO... Joerg -- AMD Operating System Research Center Advanced Micro Devices GmbH Einsteinring 24 85609 Dornach General Managers: Alberto Bozzo, Andrew Bowd Registration: Dornach, Landkr. Muenchen; Registerger. Muenchen, HRB Nr. 43632 ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH] usb: Allocate pram dynamically.
Anton Vorontsov cbouatmai...@gmail.com wrote on 2011/08/23 15:02:53: From: Anton Vorontsov cbouatmai...@gmail.com To: Joakim Tjernlund joakim.tjernl...@transmode.se Cc: linuxppc-dev@lists.ozlabs.org Date: 2011/08/23 15:02 Subject: Re: [PATCH] usb: Allocate pram dynamically. On Tue, Aug 23, 2011 at 02:38:41PM +0200, Joakim Tjernlund wrote: MPC832x does not have enough MURAM to do fixed MURAM allocation. Change to dynamic allocation. Signed-off-by: Joakim Tjernlund joakim.tjernl...@transmode.se Acked-by: Anton Vorontsov cbouatmai...@gmail.com Thanks! p.s. You probably want to send this to Greg KH, + Cc linux-usb mailing list. Added linux-usb and Greg KH per Antons suggestion. Jocke From 587137e365ac1ba7e333a09962b3e4b68c587808 Mon Sep 17 00:00:00 2001 From: Joakim Tjernlund joakim.tjernl...@transmode.se Date: Tue, 23 Aug 2011 11:04:24 +0200 Subject: [PATCH] usb: Allocate pram dynamically. MPC832x does not have enough MURAM to do fixed MURAM allocation. Change to dynamic allocation. Signed-off-by: Joakim Tjernlund joakim.tjernl...@transmode.se --- drivers/usb/host/fhci-hcd.c |5 - 1 files changed, 4 insertions(+), 1 deletions(-) diff --git a/drivers/usb/host/fhci-hcd.c b/drivers/usb/host/fhci-hcd.c index c7c8392..98adbe8 100644 --- a/drivers/usb/host/fhci-hcd.c +++ b/drivers/usb/host/fhci-hcd.c @@ -622,12 +622,15 @@ static int __devinit of_fhci_probe(struct of_device *ofdev, goto err_pram; } - pram_addr = cpm_muram_alloc_fixed(iprop[2], FHCI_PRAM_SIZE); + pram_addr = cpm_muram_alloc(FHCI_PRAM_SIZE, 64); if (IS_ERR_VALUE(pram_addr)) { dev_err(dev, failed to allocate usb pram\n); ret = -ENOMEM; goto err_pram; } + + qe_issue_cmd(QE_ASSIGN_PAGE_TO_DEVICE, QE_CR_SUBBLOCK_USB, +QE_CR_PROTOCOL_UNSPECIFIED, pram_addr); fhci-pram = cpm_muram_addr(pram_addr); /* GPIOs and pins */ -- 1.7.3.4 ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
MPC5200 + BestComm support in QEMU
Hi, I'm Steve, an embedded software developper for Bombardier Transportation Canada. We use the MPC5200 for most of our onboard computers inside train control systems. To enhance our SW engineering process, we would like the emulate the MPC5200 processor using QEMU, an open source software CPU emulator. This software supports the MPC5200 CPU emulation. However, the network interface is handled with the BestComm DMA engine and it seems very difficult to simulate this co-processor with our current knowledge level. In that sense, I would like to know if someone was able to emulate correctly the MPC5200 with the BestComm DMA using QEMU software? Regards, Steve Bélanger, ing. / Eng. Développeur logiciel embarqué / Embedded Software Developper Bombardier Transportation Canada Inc. Train Control and Management System Saint-Bruno: 450-441-2020 ext.6148 Please consider the environment before you print / Merci de penser à l'environnement avant d'imprimer ___ This e-mail communication (and any attachment/s) may contain confidential or privileged information and is intended only for the individual(s) or entity named above and to others who have been specifically authorized to receive it. If you are not the intended recipient, please do not read, copy, use or disclose the contents of this communication to others. Please notify the sender that you have received this e-mail in error by reply e-mail, and delete the e-mail subsequently. Please note that in order to protect the security of our information systems an AntiSPAM solution is in use and will browse through incoming emails. Thank you. _ Ce message (ainsi que le(s) fichier(s)), transmis par courriel, peut contenir des renseignements confidentiels ou protégés et est destiné à l?usage exclusif du destinataire ci-dessus. Toute autre personne est, par les présentes, avisée qu?il est strictement interdit de le diffuser, le distribuer ou le reproduire. Si vous l?avez reçu par inadvertance, veuillez nous en aviser et détruire ce message. Veuillez prendre note qu'une solution antipollupostage (AntiSPAM) est utilisée afin d'assurer la sécurité de nos systèmes d'information et qu'elle furètera les courriels entrants. Merci. _ ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
MPC5200 + BestComm support in QEMU
Hi, I'm Steve, an embedded software developper for Bombardier Transportation Canada. We use the MPC5200 for most of our onboard computers inside train control systems. To enhance our SW engineering process, we would like the emulate the MPC5200 processor using QEMU, an open source software CPU emulator. This software supports the MPC5200 CPU emulation. However, the network interface is handled with the BestComm DMA engine and it seems very difficult to simulate this co-processor with our current knowledge level. In that sense, I would like to know if someone was able to emulate correctly the MPC5200 with the BestComm DMA using QEMU software? Steve Bélanger, ing. / Eng. Développeur logiciel embarqué / Embedded Software Developper Bombardier Transportation Canada Inc. Train Control and Management System Saint-Bruno: 450-441-2020 ext.6148 Please consider the environment before you print / Merci de penser à l'environnement avant d'imprimer ___ This e-mail communication (and any attachment/s) may contain confidential or privileged information and is intended only for the individual(s) or entity named above and to others who have been specifically authorized to receive it. If you are not the intended recipient, please do not read, copy, use or disclose the contents of this communication to others. Please notify the sender that you have received this e-mail in error by reply e-mail, and delete the e-mail subsequently. Please note that in order to protect the security of our information systems an AntiSPAM solution is in use and will browse through incoming emails. Thank you. _ Ce message (ainsi que le(s) fichier(s)), transmis par courriel, peut contenir des renseignements confidentiels ou protégés et est destiné à l?usage exclusif du destinataire ci-dessus. Toute autre personne est, par les présentes, avisée qu?il est strictement interdit de le diffuser, le distribuer ou le reproduire. Si vous l?avez reçu par inadvertance, veuillez nous en aviser et détruire ce message. Veuillez prendre note qu'une solution antipollupostage (AntiSPAM) est utilisée afin d'assurer la sécurité de nos systèmes d'information et qu'elle furètera les courriels entrants. Merci. _ ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH v3] mtd/nand : workaround for Freescale FCM to support large-page Nand chip
On 08/23/2011 05:02 AM, Matthieu CASTET wrote: LiuShuo a écrit : We can't read the NOP from the ID on any chip. Some chips don't give this infomation.(e.g. Micron MT29F4G08BAC) Are there any 4K+ chips (especially ones with insufficient NOP) that don't have the info? This chip is 2K and NOP8. Is there an easy way (without needing to have every datasheet for every chip ever made) to determine at runtime which chips supply this information? Doesn't the micron chip provide it with onfi info ? This chip doesn't appear to be ONFI. -Scott ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
On Tue, 2011-08-23 at 12:38 +1000, David Gibson wrote: On Mon, Aug 22, 2011 at 09:45:48AM -0600, Alex Williamson wrote: On Mon, 2011-08-22 at 15:55 +1000, David Gibson wrote: On Sat, Aug 20, 2011 at 09:51:39AM -0700, Alex Williamson wrote: We had an extremely productive VFIO BoF on Monday. Here's my attempt to capture the plan that I think we agreed to: We need to address both the description and enforcement of device groups. Groups are formed any time the iommu does not have resolution between a set of devices. On x86, this typically happens when a PCI-to-PCI bridge exists between the set of devices and the iommu. For Power, partitionable endpoints define a group. Grouping information needs to be exposed for both userspace and kernel internal usage. This will be a sysfs attribute setup by the iommu drivers. Perhaps: # cat /sys/devices/pci:00/:00:19.0/iommu_group 42 (I use a PCI example here, but attribute should not be PCI specific) Ok. Am I correct in thinking these group IDs are representing the minimum granularity, and are therefore always static, defined only by the connected hardware, not by configuration? Yes, that's the idea. An open question I have towards the configuration side is whether we might add iommu driver specific options to the groups. For instance on x86 where we typically have B:D.F granularity, should we have an option not to trust multi-function devices and use a B:D granularity for grouping? Right. And likewise I can see a place for configuration parameters like the present 'allow_unsafe_irqs'. But these would be more-or-less global options which affected the overall granularity, rather than detailed configuration such as explicitly binding some devices into a group, yes? Yes, currently the interrupt remapping support is a global iommu capability. I suppose it's possible that this could be an iommu option, where the iommu driver would not advertise a group if the interrupt remapping constraint isn't met. From there we have a few options. In the BoF we discussed a model where binding a device to vfio creates a /dev/vfio$GROUP character device file. This group fd provides provides dma mapping ioctls as well as ioctls to enumerate and return a device fd for each attached member of the group (similar to KVM_CREATE_VCPU). We enforce grouping by returning an error on open() of the group fd if there are members of the group not bound to the vfio driver. Each device fd would then support a similar set of ioctls and mapping (mmio/pio/config) interface as current vfio, except for the obvious domain and dma ioctls superseded by the group fd. It seems a slightly strange distinction that the group device appears when any device in the group is bound to vfio, but only becomes usable when all devices are bound. Another valid model might be that /dev/vfio/$GROUP is created for all groups when the vfio module is loaded. The group fd would allow open() and some set of iommu querying and device enumeration ioctls, but would error on dma mapping and retrieving device fds until all of the group devices are bound to the vfio driver. Which is why I marginally prefer this model, although it's not a big deal. Right, we can also combine models. Binding a device to vfio creates /dev/vfio$GROUP, which only allows a subset of ioctls and no device access until all the group devices are also bound. I think the /dev/vfio/$GROUP might help provide an enumeration interface as well though, which could be useful. I'm not entirely sure what you mean here. But, that's now several weak votes in favour of the always-present group devices, and none in favour of the created-when-first-device-bound model, so I suggest we take the /dev/vfio/$GROUP as our tentative approach. Yep In either case, the uiommu interface is removed entirely since dma mapping is done via the group fd. As necessary in the future, we can define a more high performance dma mapping interface for streaming dma via the group fd. I expect we'll also include architecture specific group ioctls to describe features and capabilities of the iommu. The group fd will need to prevent concurrent open()s to maintain a 1:1 group to userspace process ownership model. A 1:1 group-process correspondance seems wrong to me. But there are many ways you could legitimately write the userspace side of the code, many of them involving some sort of concurrency. Implementing that concurrency as multiple processes (using explicit shared memory and/or other IPC mechanisms to co-ordinate) seems a valid choice that we shouldn't arbitrarily prohibit. Obviously, only one UID may be permitted to have the group open at a time, and I think that's enough to prevent them doing any
Re: kvm PCI assignment VFIO ramblings
On 8/23/11 4:04 AM, Joerg Roedel joerg.roe...@amd.com wrote: On Mon, Aug 22, 2011 at 08:52:18PM -0400, aafabbri wrote: You have to enforce group/iommu domain assignment whether you have the existing uiommu API, or if you change it to your proposed ioctl(inherit_iommu) API. The only change needed to VFIO here should be to make uiommu fd assignment happen on the groups instead of on device fds. That operation fails or succeeds according to the group semantics (all-or-none assignment/same uiommu). That is makes uiommu basically the same as the meta-groups, right? Yes, functionality seems the same, thus my suggestion to keep uiommu explicit. Is there some need for group-groups besides defining sets of groups which share IOMMU resources? I do all this stuff (bringing up sets of devices which may share IOMMU domain) dynamically from C applications. I don't really want some static (boot-time or sysfs fiddling) supergroup config unless there is a good reason KVM/power needs it. As you say in your next email, doing it all from ioctls is very easy, programmatically. -Aaron Fabbri ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
On Tue, 2011-08-23 at 16:54 +1000, Benjamin Herrenschmidt wrote: On Mon, 2011-08-22 at 17:52 -0700, aafabbri wrote: I'm not following you. You have to enforce group/iommu domain assignment whether you have the existing uiommu API, or if you change it to your proposed ioctl(inherit_iommu) API. The only change needed to VFIO here should be to make uiommu fd assignment happen on the groups instead of on device fds. That operation fails or succeeds according to the group semantics (all-or-none assignment/same uiommu). Ok, so I missed that part where you change uiommu to operate on group fd's rather than device fd's, my apologies if you actually wrote that down :-) It might be obvious ... bare with me I just flew back from the US and I am badly jet lagged ... I missed it too, the model I'm proposing entirely removes the uiommu concept. So I see what you mean, however... I think the question is: do we force 1:1 iommu/group mapping, or do we allow arbitrary mapping (satisfying group constraints) as we do today. I'm saying I'm an existing user who wants the arbitrary iommu/group mapping ability and definitely think the uiommu approach is cleaner than the ioctl(inherit_iommu) approach. We considered that approach before but it seemed less clean so we went with the explicit uiommu context. Possibly, the question that interest me the most is what interface will KVM end up using. I'm also not terribly fan with the (perceived) discrepancy between using uiommu to create groups but using the group fd to actually do the mappings, at least if that is still the plan. Current code: uiommu creates the domain, we bind a vfio device to that domain via a SET_UIOMMU_DOMAIN ioctl on the vfio device, then do mappings via MAP_DMA on the vfio device (affecting all the vfio devices bound to the domain) My current proposal: groups are predefined. groups ~= iommu domain. The iommu domain would probably be allocated when the first device is bound to vfio. As each device is bound, it gets attached to the group. DMAs are done via an ioctl on the group. I think group + uiommu leads to effectively reliving most of the problems with the current code. The only benefit is the group assignment to enforce hardware restrictions. We still have the problem that uiommu open() = iommu_domain_alloc(), whose properties are meaningless without attached devices (groups). Which I think leads to the same awkward model of attaching groups to define the domain, then we end up doing mappings via the group to enforce ordering. If the separate uiommu interface is kept, then anything that wants to be able to benefit from the ability to put multiple devices (or existing groups) into such a meta group would need to be explicitly modified to deal with the uiommu APIs. I tend to prefer such meta groups as being something you create statically using a configuration interface, either via sysfs, netlink or ioctl's to a control vfio device driven by a simple command line tool (which can have the configuration stored in /etc and re-apply it at boot). I cringe anytime there's a mention of static. IMHO, we have to support hotplug. That means meta groups change dynamically. Maybe this supports the idea that we should be able to retrieve a new fd from the group to do mappings. Any groups bound together will return the same fd and the fd will persist so long as any member of the group is open. That way, any program capable of exploiting VFIO groups will automatically be able to exploit those meta groups (or groups of groups) as well as long as they are supported on the system. If we ever have system specific constraints as to how such groups can be created, then it can all be handled at the level of that configuration tool without impact on whatever programs know how to exploit them via the VFIO interfaces. I'd prefer to have the constraints be represented in the ioctl to bind groups. It works or not and the platform gets to define what it considers compatible. Thanks, Alex ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
On Tue, 2011-08-23 at 15:14 +0200, Roedel, Joerg wrote: On Mon, Aug 22, 2011 at 03:17:00PM -0400, Alex Williamson wrote: On Mon, 2011-08-22 at 19:25 +0200, Joerg Roedel wrote: I am in favour of /dev/vfio/$GROUP. If multiple devices should be assigned to a guest, there can also be an ioctl to bind a group to an address-space of another group (certainly needs some care to not allow that both groups belong to different processes). That's an interesting idea. Maybe an interface similar to the current uiommu interface, where you open() the 2nd group fd and pass the fd via ioctl to the primary group. IOMMUs that don't support this would fail the attach device callback, which would fail the ioctl to bind them. It will need to be designed so any group can be removed from the super-set and the remaining group(s) still works. This feels like something that can be added after we get an initial implementation. Handling it through fds is a good idea. This makes sure that everything belongs to one process. I am not really sure yet if we go the way to just bind plain groups together or if we create meta-groups. The meta-groups thing seems somewhat cleaner, though. I'm leaning towards binding because we need to make it dynamic, but I don't really have a good picture of the lifecycle of a meta-group. Btw, a problem we havn't talked about yet entirely is driver-deassignment. User space can decide to de-assign the device from vfio while a fd is open on it. With PCI there is no way to let this fail (the .release function returns void last time i checked). Is this a problem, and yes, how we handle that? The current vfio has the same problem, we can't unbind a device from vfio while it's attached to a guest. I think we'd use the same solution too; send out a netlink packet for a device removal and have the .remove call sleep on a wait_event(, refcnt == 0). We could also set a timeout and SIGBUS the PIDs holding the device if they don't return it willingly. Thanks, Putting the process to sleep (which would be uninterruptible) seems bad. The process would sleep until the guest releases the device-group, which can take days or months. The best thing (and the most intrusive :-) ) is to change PCI core to allow unbindings to fail, I think. But this probably further complicates the way to upstream VFIO... Yes, it's not ideal but I think it's sufficient for now and if we later get support for returning an error from release, we can set a timeout after notifying the user to make use of that. Thanks, Alex ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
On 8/23/11 10:01 AM, Alex Williamson alex.william...@redhat.com wrote: On Tue, 2011-08-23 at 16:54 +1000, Benjamin Herrenschmidt wrote: On Mon, 2011-08-22 at 17:52 -0700, aafabbri wrote: I'm not following you. You have to enforce group/iommu domain assignment whether you have the existing uiommu API, or if you change it to your proposed ioctl(inherit_iommu) API. The only change needed to VFIO here should be to make uiommu fd assignment happen on the groups instead of on device fds. That operation fails or succeeds according to the group semantics (all-or-none assignment/same uiommu). Ok, so I missed that part where you change uiommu to operate on group fd's rather than device fd's, my apologies if you actually wrote that down :-) It might be obvious ... bare with me I just flew back from the US and I am badly jet lagged ... I missed it too, the model I'm proposing entirely removes the uiommu concept. So I see what you mean, however... I think the question is: do we force 1:1 iommu/group mapping, or do we allow arbitrary mapping (satisfying group constraints) as we do today. I'm saying I'm an existing user who wants the arbitrary iommu/group mapping ability and definitely think the uiommu approach is cleaner than the ioctl(inherit_iommu) approach. We considered that approach before but it seemed less clean so we went with the explicit uiommu context. Possibly, the question that interest me the most is what interface will KVM end up using. I'm also not terribly fan with the (perceived) discrepancy between using uiommu to create groups but using the group fd to actually do the mappings, at least if that is still the plan. Current code: uiommu creates the domain, we bind a vfio device to that domain via a SET_UIOMMU_DOMAIN ioctl on the vfio device, then do mappings via MAP_DMA on the vfio device (affecting all the vfio devices bound to the domain) My current proposal: groups are predefined. groups ~= iommu domain. This is my main objection. I'd rather not lose the ability to have multiple devices (which are all predefined as singleton groups on x86 w/o PCI bridges) share IOMMU resources. Otherwise, 20 devices sharing buffers would require 20x the IOMMU/ioTLB resources. KVM doesn't care about this case? The iommu domain would probably be allocated when the first device is bound to vfio. As each device is bound, it gets attached to the group. DMAs are done via an ioctl on the group. I think group + uiommu leads to effectively reliving most of the problems with the current code. The only benefit is the group assignment to enforce hardware restrictions. We still have the problem that uiommu open() = iommu_domain_alloc(), whose properties are meaningless without attached devices (groups). Which I think leads to the same awkward model of attaching groups to define the domain, then we end up doing mappings via the group to enforce ordering. Is there a better way to allow groups to share an IOMMU domain? Maybe, instead of having an ioctl to allow a group A to inherit the same iommu domain as group B, we could have an ioctl to fully merge two groups (could be what Ben was thinking): A.ioctl(MERGE_TO_GROUP, B) The group A now goes away and its devices join group B. If A ever had an iommu domain assigned (and buffers mapped?) we fail. Groups cannot get smaller (they are defined as minimum granularity of an IOMMU, initially). They can get bigger if you want to share IOMMU resources, though. Any downsides to this approach? -AF If the separate uiommu interface is kept, then anything that wants to be able to benefit from the ability to put multiple devices (or existing groups) into such a meta group would need to be explicitly modified to deal with the uiommu APIs. I tend to prefer such meta groups as being something you create statically using a configuration interface, either via sysfs, netlink or ioctl's to a control vfio device driven by a simple command line tool (which can have the configuration stored in /etc and re-apply it at boot). I cringe anytime there's a mention of static. IMHO, we have to support hotplug. That means meta groups change dynamically. Maybe this supports the idea that we should be able to retrieve a new fd from the group to do mappings. Any groups bound together will return the same fd and the fd will persist so long as any member of the group is open. That way, any program capable of exploiting VFIO groups will automatically be able to exploit those meta groups (or groups of groups) as well as long as they are supported on the system. If we ever have system specific constraints as to how such groups can be created, then it can all be handled at the level of that configuration tool without impact on whatever programs know how to exploit them via the VFIO interfaces. I'd prefer to have the constraints be represented in the ioctl to bind groups. It works or not and
Re: [PATCH] powerpc: fixup QE_General4 errata
Joakim Tjernlund wrote: QE_General4 should only round up the divisor iff divisor is 3. Rounding up lower divisors makes the error too big, causing USB on MPC832x to fail. Signed-off-by: Joakim Tjernlund joakim.tjernl...@transmode.se Acked-by: Timur Tabi ti...@freescale.com -- Timur Tabi Linux kernel developer at Freescale ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
On Tue, 2011-08-23 at 10:33 -0700, Aaron Fabbri wrote: On 8/23/11 10:01 AM, Alex Williamson alex.william...@redhat.com wrote: On Tue, 2011-08-23 at 16:54 +1000, Benjamin Herrenschmidt wrote: On Mon, 2011-08-22 at 17:52 -0700, aafabbri wrote: I'm not following you. You have to enforce group/iommu domain assignment whether you have the existing uiommu API, or if you change it to your proposed ioctl(inherit_iommu) API. The only change needed to VFIO here should be to make uiommu fd assignment happen on the groups instead of on device fds. That operation fails or succeeds according to the group semantics (all-or-none assignment/same uiommu). Ok, so I missed that part where you change uiommu to operate on group fd's rather than device fd's, my apologies if you actually wrote that down :-) It might be obvious ... bare with me I just flew back from the US and I am badly jet lagged ... I missed it too, the model I'm proposing entirely removes the uiommu concept. So I see what you mean, however... I think the question is: do we force 1:1 iommu/group mapping, or do we allow arbitrary mapping (satisfying group constraints) as we do today. I'm saying I'm an existing user who wants the arbitrary iommu/group mapping ability and definitely think the uiommu approach is cleaner than the ioctl(inherit_iommu) approach. We considered that approach before but it seemed less clean so we went with the explicit uiommu context. Possibly, the question that interest me the most is what interface will KVM end up using. I'm also not terribly fan with the (perceived) discrepancy between using uiommu to create groups but using the group fd to actually do the mappings, at least if that is still the plan. Current code: uiommu creates the domain, we bind a vfio device to that domain via a SET_UIOMMU_DOMAIN ioctl on the vfio device, then do mappings via MAP_DMA on the vfio device (affecting all the vfio devices bound to the domain) My current proposal: groups are predefined. groups ~= iommu domain. This is my main objection. I'd rather not lose the ability to have multiple devices (which are all predefined as singleton groups on x86 w/o PCI bridges) share IOMMU resources. Otherwise, 20 devices sharing buffers would require 20x the IOMMU/ioTLB resources. KVM doesn't care about this case? We do care, I just wasn't prioritizing it as heavily since I think the typical model is probably closer to 1 device per guest. The iommu domain would probably be allocated when the first device is bound to vfio. As each device is bound, it gets attached to the group. DMAs are done via an ioctl on the group. I think group + uiommu leads to effectively reliving most of the problems with the current code. The only benefit is the group assignment to enforce hardware restrictions. We still have the problem that uiommu open() = iommu_domain_alloc(), whose properties are meaningless without attached devices (groups). Which I think leads to the same awkward model of attaching groups to define the domain, then we end up doing mappings via the group to enforce ordering. Is there a better way to allow groups to share an IOMMU domain? Maybe, instead of having an ioctl to allow a group A to inherit the same iommu domain as group B, we could have an ioctl to fully merge two groups (could be what Ben was thinking): A.ioctl(MERGE_TO_GROUP, B) The group A now goes away and its devices join group B. If A ever had an iommu domain assigned (and buffers mapped?) we fail. Groups cannot get smaller (they are defined as minimum granularity of an IOMMU, initially). They can get bigger if you want to share IOMMU resources, though. Any downsides to this approach? That's sort of the way I'm picturing it. When groups are bound together, they effectively form a pool, where all the groups are peers. When the MERGE/BIND ioctl is called on group A and passed the group B fd, A can check compatibility of the domain associated with B, unbind devices from the B domain and attach them to the A domain. The B domain would then be freed and it would bump the refcnt on the A domain. If we need to remove A from the pool, we call UNMERGE/UNBIND on B with the A fd, it will remove the A devices from the shared object, disassociate A with the shared object, re-alloc a domain for A and rebind A devices to that domain. This is where it seems like it might be helpful to make a GET_IOMMU_FD ioctl so that an iommu object is ubiquitous and persistent across the pool. Operations on any group fd work on the pool as a whole. Thanks, Alex ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH 8/9] arch/powerpc/sysdev/ehv_pic.c: add missing kfree
Julia Lawall wrote: At this point, ehv_pic has been allocated but not stored anywhere, so it should be freed before leaving the function. Acked-by: Timur Tabi ti...@freescale.com FYI, Ashish is no longer with Freescale, so I've taken over maintainership of ehv_pic. -- Timur Tabi Linux kernel developer at Freescale ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH 8/9] arch/powerpc/sysdev/ehv_pic.c: add missing kfree
Ben, Kumar, can one of you take a look at my question and help me out? wrote: On Mon, Aug 8, 2011 at 7:18 AM, Julia Lawall ju...@diku.dk wrote: diff --git a/arch/powerpc/sysdev/ehv_pic.c b/arch/powerpc/sysdev/ehv_pic.c index af1a5df..b6731e4 100644 --- a/arch/powerpc/sysdev/ehv_pic.c +++ b/arch/powerpc/sysdev/ehv_pic.c @@ -280,6 +280,7 @@ void __init ehv_pic_init(void) if (!ehv_pic-irqhost) { of_node_put(np); + kfree(ehv_pic); return; } Although the fix is correct, I think there is another bug in this function. 'np' is not released when the function finishes successfully. I've looked at other functions that use irq_alloc_host(), and most of them do the same thing: they don't call of_node_put() on the device node pointer. The only exception I've found is mpc5121_ads_cpld_pic_init(). Ben, Kumar: am I missing something? irq_alloc_host() calls of_node_get(): host-of_node = of_node_get(of_node); so doesn't that mean that the caller of irq_alloc_host() should release the device node pointer? -- Timur Tabi Linux kernel developer at Freescale ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
On Tue, 2011-08-23 at 07:01 +1000, Benjamin Herrenschmidt wrote: On Mon, 2011-08-22 at 09:45 -0600, Alex Williamson wrote: Yes, that's the idea. An open question I have towards the configuration side is whether we might add iommu driver specific options to the groups. For instance on x86 where we typically have B:D.F granularity, should we have an option not to trust multi-function devices and use a B:D granularity for grouping? Or even B or range of busses... if you want to enforce strict isolation you really can't trust anything below a bus level :-) Right, we can also combine models. Binding a device to vfio creates /dev/vfio$GROUP, which only allows a subset of ioctls and no device access until all the group devices are also bound. I think the /dev/vfio/$GROUP might help provide an enumeration interface as well though, which could be useful. Could be tho in what form ? returning sysfs pathes ? I'm at a loss there, please suggest. I think we need an ioctl that returns some kind of array of devices within the group and another that maybe takes an index from that array and returns an fd for that device. A sysfs path string might be a reasonable array element, but it sounds like a pain to work with. 1:1 group-process is probably too strong. Not allowing concurrent open()s on the group file enforces a single userspace entity is responsible for that group. Device fds can be passed to other processes, but only retrieved via the group fd. I suppose we could even branch off the dma interface into a different fd, but it seems like we would logically want to serialize dma mappings at each iommu group anyway. I'm open to alternatives, this just seemed an easy way to do it. Restricting on UID implies that we require isolated qemu instances to run as different UIDs. I know that's a goal, but I don't know if we want to make it an assumption in the group security model. 1:1 process has the advantage of linking to an -mm which makes the whole mmu notifier business doable. How do you want to track down mappings and do the second level translation in the case of explicit map/unmap (like on power) if you are not tied to an mm_struct ? Right, I threw away the mmu notifier code that was originally part of vfio because we can't do anything useful with it yet on x86. I definitely don't want to prevent it where it makes sense though. Maybe we just record current-mm on open and restrict subsequent opens to the same. Yes. I'm not sure there's a good ROI to prioritize that model. We have to assume 1 device per guest is a typical model and that the iotlb is large enough that we might improve thrashing to see both a resource and performance benefit from it. I'm open to suggestions for how we could include it though. Sharing may or may not be possible depending on setups so yes, it's a bit tricky. My preference is to have a static interface (and that's actually where your pet netlink might make some sense :-) to create synthetic groups made of other groups if the arch allows it. But that might not be the best approach. In another email I also proposed an option for a group to capture another one... I already made some comments on this in a different thread, so I won't repeat here. If that's not what you're saying, how would the domains - now made up of a user's selection of groups, rather than individual devices - be configured? Hope that captures it, feel free to jump in with corrections and suggestions. Thanks, Another aspect I don't see discussed is how we represent these things to the guest. On Power for example, I have a requirement that a given iommu domain is represented by a single dma window property in the device-tree. What that means is that that property needs to be either in the node of the device itself if there's only one device in the group or in a parent node (ie a bridge or host bridge) if there are multiple devices. Now I do -not- want to go down the path of simulating P2P bridges, besides we'll quickly run out of bus numbers if we go there. For us the most simple and logical approach (which is also what pHyp uses and what Linux handles well) is really to expose a given PCI host bridge per group to the guest. Believe it or not, it makes things easier :-) I'm all for easier. Why does exposing the bridge use less bus numbers than emulating a bridge? On x86, I want to maintain that our default assignment is at the device level. A user should be able to pick single or multiple devices from across several groups and have them all show up as individual, hotpluggable devices on bus 0 in the guest. Not surprisingly, we've also seen cases where users try to attach a bridge to the guest, assuming they'll get all the devices below the bridge, so I'd be in favor of making this just work if possible too, though we may have to prevent hotplug of those. Given the device
Re: [PATCH part1 v2 1/9] Add udbg driver using the PS3 gelic Ethernet device
Hi, We had some questions as to why we have this totally separate driver from the gelic driver, so I think it worthwhile to have an explanation of why in the commit log. Otherwise, the code looks OK. -Geoff From: Hector Martin hec...@marcansoft.com Signed-off-by: Hector Martin hec...@marcansoft.com [a.heider: Various cleanups to make checkpatch.pl happy] Signed-off-by: Andre Heider a.hei...@gmail.com --- arch/powerpc/Kconfig.debug |8 + arch/powerpc/include/asm/udbg.h |1 + arch/powerpc/kernel/udbg.c |2 + arch/powerpc/platforms/ps3/Kconfig | 12 ++ arch/powerpc/platforms/ps3/Makefile |1 + arch/powerpc/platforms/ps3/gelic_udbg.c | 273 +++ drivers/net/ps3_gelic_net.c |3 + drivers/net/ps3_gelic_net.h |6 + 8 files changed, 306 insertions(+), 0 deletions(-) create mode 100644 arch/powerpc/platforms/ps3/gelic_udbg.c diff --git a/arch/powerpc/Kconfig.debug b/arch/powerpc/Kconfig.debug index 067cb84..ab2335f 100644 --- a/arch/powerpc/Kconfig.debug +++ b/arch/powerpc/Kconfig.debug @@ -258,6 +258,14 @@ config PPC_EARLY_DEBUG_WSP depends on PPC_WSP select PPC_UDBG_16550 +config PPC_EARLY_DEBUG_PS3GELIC + bool Early debugging through the PS3 Ethernet port + depends on PPC_PS3 + select PS3GELIC_UDBG + help + Select this to enable early debugging for the PlayStation3 via + UDP broadcasts sent out through the Ethernet port. + endchoice config PPC_EARLY_DEBUG_HVSI_VTERMNO diff --git a/arch/powerpc/include/asm/udbg.h b/arch/powerpc/include/asm/udbg.h index 93e05d1..7cf796f 100644 --- a/arch/powerpc/include/asm/udbg.h +++ b/arch/powerpc/include/asm/udbg.h @@ -54,6 +54,7 @@ extern void __init udbg_init_40x_realmode(void); extern void __init udbg_init_cpm(void); extern void __init udbg_init_usbgecko(void); extern void __init udbg_init_wsp(void); +extern void __init udbg_init_ps3gelic(void); #endif /* __KERNEL__ */ #endif /* _ASM_POWERPC_UDBG_H */ diff --git a/arch/powerpc/kernel/udbg.c b/arch/powerpc/kernel/udbg.c index faa82c1..5b3e98e 100644 --- a/arch/powerpc/kernel/udbg.c +++ b/arch/powerpc/kernel/udbg.c @@ -67,6 +67,8 @@ void __init udbg_early_init(void) udbg_init_usbgecko(); #elif defined(CONFIG_PPC_EARLY_DEBUG_WSP) udbg_init_wsp(); +#elif defined(CONFIG_PPC_EARLY_DEBUG_PS3GELIC) + udbg_init_ps3gelic(); #endif #ifdef CONFIG_PPC_EARLY_DEBUG diff --git a/arch/powerpc/platforms/ps3/Kconfig b/arch/powerpc/platforms/ps3/Kconfig index dfe316b..476d9d9 100644 --- a/arch/powerpc/platforms/ps3/Kconfig +++ b/arch/powerpc/platforms/ps3/Kconfig @@ -148,4 +148,16 @@ config PS3_LPM profiling support of the Cell processor with programs like oprofile and perfmon2, then say Y or M, otherwise say N. +config PS3GELIC_UDBG + bool PS3 udbg output via UDP broadcasts on Ethernet + depends on PPC_PS3 + help + Enables udbg early debugging output by sending broadcast UDP + via the Ethernet port (UDP port number 18194). + + This driver uses a trivial implementation and is independent + from the main network driver. + + If in doubt, say N here. + endmenu diff --git a/arch/powerpc/platforms/ps3/Makefile b/arch/powerpc/platforms/ps3/Makefile index ac1bdf8..02b9e63 100644 --- a/arch/powerpc/platforms/ps3/Makefile +++ b/arch/powerpc/platforms/ps3/Makefile @@ -2,6 +2,7 @@ obj-y += setup.o mm.o time.o hvcall.o htab.o repository.o obj-y += interrupt.o exports.o os-area.o obj-y += system-bus.o +obj-$(CONFIG_PS3GELIC_UDBG) += gelic_udbg.o obj-$(CONFIG_SMP) += smp.o obj-$(CONFIG_SPU_BASE) += spu.o obj-y += device-init.o diff --git a/arch/powerpc/platforms/ps3/gelic_udbg.c b/arch/powerpc/platforms/ps3/gelic_udbg.c new file mode 100644 index 000..20b46a1 --- /dev/null +++ b/arch/powerpc/platforms/ps3/gelic_udbg.c @@ -0,0 +1,273 @@ +/* + * udbg debug output routine via GELIC UDP broadcasts + * + * Copyright (C) 2007 Sony Computer Entertainment Inc. + * Copyright 2006, 2007 Sony Corporation + * Copyright (C) 2010 Hector Martin hec...@marcansoft.com + * Copyright (C) 2011 Andre Heider a.hei...@gmail.com + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + */ + +#include asm/io.h +#include asm/udbg.h +#include asm/lv1call.h + +#define GELIC_BUS_ID 1 +#define GELIC_DEVICE_ID 0 +#define GELIC_DEBUG_PORT 18194 +#define GELIC_MAX_MESSAGE_SIZE 1000 + +#define GELIC_LV1_GET_MAC_ADDRESS 1 +#define GELIC_LV1_GET_VLAN_ID 4 +#define GELIC_LV1_VLAN_TX_ETHERNET_0 2 + +#define GELIC_DESCR_DMA_STAT_MASK 0xf000 +#define GELIC_DESCR_DMA_CARDOWNED
Re: [PATCH part1 v2 2/9] ps3: Add helper functions to read highmem info from the repository
On 08/11/2011 12:31 PM, Andre Heider wrote: An earlier step in the boot chain can preallocate the highmem region. A boot loader doing so will place the region infos in the repository. Provide helper functions to read the required nodes. Signed-off-by: Andre Heider a.hei...@gmail.com --- arch/powerpc/platforms/ps3/platform.h |3 ++ arch/powerpc/platforms/ps3/repository.c | 36 +++ 2 files changed, 39 insertions(+), 0 deletions(-) diff --git a/arch/powerpc/platforms/ps3/platform.h b/arch/powerpc/platforms/ps3/platform.h index 9a196a8..d9b4ec0 100644 --- a/arch/powerpc/platforms/ps3/platform.h +++ b/arch/powerpc/platforms/ps3/platform.h @@ -187,6 +187,9 @@ int ps3_repository_read_rm_size(unsigned int ppe_id, u64 *rm_size); int ps3_repository_read_region_total(u64 *region_total); int ps3_repository_read_mm_info(u64 *rm_base, u64 *rm_size, u64 *region_total); +int ps3_repository_read_highmem_base(u64 *highmem_base); +int ps3_repository_read_highmem_size(u64 *highmem_size); +int ps3_repository_read_highmem_info(u64 *highmem_base, u64 *highmem_size); In the general case we could have multiple regions. If we add a region arg here we can handle that if needed. region_index would be {1..}. ps3_repository_read_highmem_info could hold how many regions, so: int ps3_repository_read_highmem_base(unsigned int region_index, u64 *highmem_base); int ps3_repository_read_highmem_size(unsigned int region_index, u64 *highmem_size); int ps3_repository_read_highmem_region(unsigned int region_index, u64 *highmem_base, u64 *highmem_size); /* repository pme info */ diff --git a/arch/powerpc/platforms/ps3/repository.c b/arch/powerpc/platforms/ps3/repository.c index 5e304c2..9908d61 100644 --- a/arch/powerpc/platforms/ps3/repository.c +++ b/arch/powerpc/platforms/ps3/repository.c @@ -778,6 +778,42 @@ int ps3_repository_read_mm_info(u64 *rm_base, u64 *rm_size, u64 *region_total) : ps3_repository_read_region_total(region_total); } +int ps3_repository_read_highmem_base(u64 *highmem_base) +{ + return read_node(PS3_LPAR_ID_CURRENT, + make_first_field(bi, 0), + make_field(highmem, 0), + make_field(base, 0), + 0, + highmem_base, NULL); +} I think something like this seems better: int ps3_repository_read_highmem_base(unsigned int region_index, u64 *highmem_base) { return read_node(PS3_LPAR_ID_CURRENT, make_first_field(highmem, 0), make_field(region, region_index), make_field(base, 0), 0, highmem_base, NULL); } + +int ps3_repository_read_highmem_size(u64 *highmem_size) +{ + return read_node(PS3_LPAR_ID_CURRENT, + make_first_field(bi, 0), + make_field(highmem, 0), + make_field(size, 0), + 0, + highmem_size, NULL); +} + +/** + * ps3_repository_read_highmem_info - Read high memory info + * @highmem_base: High memory base address. + * @highmem_size: High mode memory size. + */ + +int ps3_repository_read_highmem_info(u64 *highmem_base, u64 *highmem_size) +{ + int result; + + *highmem_base = 0; + result = ps3_repository_read_highmem_base(highmem_base); + return result ? result + : ps3_repository_read_highmem_size(highmem_size); ps3_repository_read_highmem_base(1, highmem_base); ... +} + /** * ps3_repository_read_num_spu_reserved - Number of physical spus reserved. * @num_spu: Number of physical spus. -Geoff ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH part1 v2 3/9] ps3: Get lv1 high memory region from the repository
On 08/11/2011 12:31 PM, Andre Heider wrote: This lets the bootloader preallocate the high lv1 region and pass its location to the kernel through the repository. Thus, it can be used to hold the initrd. If the region info doesn't exist, the kernel retains the old behavior and attempts to allocate the region itself. Based on the patch [PS3] Get lv1 high memory region from devtree from Hector Martin hec...@marcansoft.com Signed-off-by: Andre Heider a.hei...@gmail.com --- arch/powerpc/platforms/ps3/mm.c | 46 -- 1 files changed, 43 insertions(+), 3 deletions(-) diff --git a/arch/powerpc/platforms/ps3/mm.c b/arch/powerpc/platforms/ps3/mm.c index c204588..983b719 100644 --- a/arch/powerpc/platforms/ps3/mm.c +++ b/arch/powerpc/platforms/ps3/mm.c @@ -78,12 +78,14 @@ enum { * @base: base address * @size: size in bytes * @offset: difference between base and rm.size + * @destroy: flag if region should be destroyed upon shutdown */ struct mem_region { u64 base; u64 size; unsigned long offset; + int destroy; }; /** @@ -261,6 +263,7 @@ static int ps3_mm_region_create(struct mem_region *r, unsigned long size) goto zero_region; } + r-destroy = 1; r-offset = r-base - map.rm.size; return result; @@ -279,6 +282,12 @@ static void ps3_mm_region_destroy(struct mem_region *r) int result; DBG(%s:%d: r-base = %llxh\n, __func__, __LINE__, r-base); + + if (!r-destroy) { + DBG(%s:%d: not destroying region\n, __func__, __LINE__); + return; + } + if (r-base) { result = lv1_release_memory(r-base); BUG_ON(result); @@ -287,6 +296,29 @@ static void ps3_mm_region_destroy(struct mem_region *r) } } +static int ps3_mm_get_repository_highmem(struct mem_region *r) +{ + int result = ps3_repository_read_highmem_info(r-base, r-size); + + if (result) + goto zero_region; + + if (!r-base || !r-size) { + result = -1; + goto zero_region; + } + + r-offset = r-base - map.rm.size; + DBG(%s:%d got high region from repository: %llxh %llxh\n, + __func__, __LINE__, r-base, r-size); + return 0; + +zero_region: + DBG(%s:%d no high region in repository...\n, __func__, __LINE__); Three dots implies something more is on its way. I don't think we need them. + r-size = r-base = r-offset = 0; + return result; +} + /** * ps3_mm_add_memory - hot add memory */ @@ -303,6 +335,12 @@ static int __init ps3_mm_add_memory(void) BUG_ON(!mem_init_done); + if (!map.r1.size) { + DBG(%s:%d: no region 1, not adding memory\n, + __func__, __LINE__); + return 0; + } + start_addr = map.rm.size; start_pfn = start_addr PAGE_SHIFT; nr_pages = (map.r1.size + PAGE_SIZE - 1) PAGE_SHIFT; @@ -1217,9 +1255,11 @@ void __init ps3_mm_init(void) BUG_ON(map.rm.base); BUG_ON(!map.rm.size); - - /* arrange to do this in ps3_mm_add_memory */ - ps3_mm_region_create(map.r1, map.total - map.rm.size); + /* check if we got the highmem region from an earlier boot step */ + if (ps3_mm_get_repository_highmem(map.r1)) { + /* arrange to do this in ps3_mm_add_memory */ + ps3_mm_region_create(map.r1, map.total - map.rm.size); + } /* correct map.total for the real total amount of memory we use */ map.total = map.rm.size + map.r1.size; ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH part1 v2 4/9] Add region 1 memory early
On 08/11/2011 12:31 PM, Andre Heider wrote: From: Hector Martin hec...@marcansoft.com Real mode memory can be limited and runs out quickly as memory is allocated during kernel startup. Having region1 available sooner fixes this. Signed-off-by: Hector Martin hec...@marcansoft.com [a.heider: Various cleanups to make checkpatch.pl happy] Signed-off-by: Andre Heider a.hei...@gmail.com --- arch/powerpc/platforms/ps3/mm.c | 75 +++ 1 files changed, 13 insertions(+), 62 deletions(-) diff --git a/arch/powerpc/platforms/ps3/mm.c b/arch/powerpc/platforms/ps3/mm.c index 983b719..68b3879 100644 --- a/arch/powerpc/platforms/ps3/mm.c +++ b/arch/powerpc/platforms/ps3/mm.c @@ -20,7 +20,6 @@ #include linux/kernel.h #include linux/module.h -#include linux/memory_hotplug.h #include linux/memblock.h #include linux/slab.h @@ -94,10 +93,8 @@ struct mem_region { * @vas_id - HV virtual address space id * @htab_size: htab size in bytes * - * The HV virtual address space (vas) allows for hotplug memory regions. - * Memory regions can be created and destroyed in the vas at runtime. This is still true, so we should keep these comments. We are only changing the way we use the feature. * @rm: real mode (bootmem) region - * @r1: hotplug memory region(s) + * @r1: high memory region high memory region(s) * * ps3 addresses * virt_addr: a cpu 'translated' effective address @@ -223,10 +220,6 @@ void ps3_mm_vas_destroy(void) } } -/**/ -/* memory hotplug routines */ -/**/ - /** * ps3_mm_region_create - create a memory region in the vas * @r: pointer to a struct mem_region to accept initialized values @@ -319,57 +312,6 @@ zero_region: return result; } -/** - * ps3_mm_add_memory - hot add memory - */ - -static int __init ps3_mm_add_memory(void) -{ - int result; - unsigned long start_addr; - unsigned long start_pfn; - unsigned long nr_pages; - - if (!firmware_has_feature(FW_FEATURE_PS3_LV1)) - return -ENODEV; - - BUG_ON(!mem_init_done); - - if (!map.r1.size) { - DBG(%s:%d: no region 1, not adding memory\n, - __func__, __LINE__); - return 0; - } - - start_addr = map.rm.size; - start_pfn = start_addr PAGE_SHIFT; - nr_pages = (map.r1.size + PAGE_SIZE - 1) PAGE_SHIFT; - - DBG(%s:%d: start_addr %lxh, start_pfn %lxh, nr_pages %lxh\n, - __func__, __LINE__, start_addr, start_pfn, nr_pages); - - result = add_memory(0, start_addr, map.r1.size); - - if (result) { - pr_err(%s:%d: add_memory failed: (%d)\n, - __func__, __LINE__, result); - return result; - } - - memblock_add(start_addr, map.r1.size); - memblock_analyze(); - - result = online_pages(start_pfn, nr_pages); - - if (result) - pr_err(%s:%d: online_pages failed: (%d)\n, - __func__, __LINE__, result); - - return result; -} - -device_initcall(ps3_mm_add_memory); - /**/ /* dma routines */ /**/ @@ -1256,14 +1198,23 @@ void __init ps3_mm_init(void) BUG_ON(!map.rm.size); /* check if we got the highmem region from an earlier boot step */ - if (ps3_mm_get_repository_highmem(map.r1)) { - /* arrange to do this in ps3_mm_add_memory */ + if (ps3_mm_get_repository_highmem(map.r1)) ps3_mm_region_create(map.r1, map.total - map.rm.size); - } This should be folded into patch #3. /* correct map.total for the real total amount of memory we use */ map.total = map.rm.size + map.r1.size; + if (!map.r1.size) { + DBG(%s:%d: no region 1, not adding memory\n, + __func__, __LINE__); + } else { Remove brackets around a single line conditional. + DBG(%s:%d: adding memory: start %llxh, size %llxh\n, + __func__, __LINE__, map.rm.size, map.r1.size); + + memblock_add(map.rm.size, map.r1.size); + memblock_analyze(); + } + DBG( - %s:%d\n, __func__, __LINE__); } ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH part1 v2 5/9] ps3: MEMORY_HOTPLUG is not a requirement anymore
On 08/11/2011 12:31 PM, Andre Heider wrote: Signed-off-by: Andre Heider a.hei...@gmail.com --- arch/powerpc/platforms/ps3/Kconfig |1 - 1 files changed, 0 insertions(+), 1 deletions(-) diff --git a/arch/powerpc/platforms/ps3/Kconfig b/arch/powerpc/platforms/ps3/Kconfig index 476d9d9..84df5c8 100644 --- a/arch/powerpc/platforms/ps3/Kconfig +++ b/arch/powerpc/platforms/ps3/Kconfig @@ -7,7 +7,6 @@ config PPC_PS3 select USB_OHCI_BIG_ENDIAN_MMIO select USB_ARCH_HAS_EHCI select USB_EHCI_BIG_ENDIAN_MMIO - select MEMORY_HOTPLUG This one is OK, once the others are fixed up. -Geoff ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH part1 v2 6/9] ps3: Detect the current lpar
Hi, On 08/11/2011 12:31 PM, Andre Heider wrote: Detect it by reading the ss laid repository node, and make it accessible via ps3_get_ss_laid(). I'm wondering now if we even need this. It is mainly used by your later patch 8/9 that modifies ps3flash_init() to test if we should call ps3_system_bus_driver_register(). If we don't use ps3_get_ss_laid() and just allow ps3_system_bus_driver_register() to be called, would the device probe fail and have the same result as the test? I would prefer to not have ps3_get_ss_laid() if possible. -Geoff ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH part1 v2 8/9] ps3flash: Refuse to work in lpars other than OtherOS
On 08/11/2011 12:31 PM, Andre Heider wrote: The driver implements a character and misc device, meant for the axed OtherOS to exchange various settings with GameOS. Since Firmware 3.21 there is no support anymore to write these settings, so test if we're running in OtherOS, and refuse to load if that is not the case. Please see my comments to the v1 patch regarding this text. Signed-off-by: Andre Heider a.hei...@gmail.com --- arch/powerpc/platforms/ps3/Kconfig |2 +- drivers/char/ps3flash.c|7 +++ 2 files changed, 8 insertions(+), 1 deletions(-) diff --git a/arch/powerpc/platforms/ps3/Kconfig b/arch/powerpc/platforms/ps3/Kconfig index 84df5c8..72fdecd 100644 --- a/arch/powerpc/platforms/ps3/Kconfig +++ b/arch/powerpc/platforms/ps3/Kconfig @@ -121,7 +121,7 @@ config PS3_FLASH This support is required to access the PS3 FLASH ROM, which contains the boot loader and some boot options. - In general, all users will say Y or M. + In general, all PS3 OtherOS users will say Y or M. As this driver needs a fixed buffer of 256 KiB of memory, it can be disabled on the kernel command line using ps3flash=off, to diff --git a/drivers/char/ps3flash.c b/drivers/char/ps3flash.c index d0c57c2..fc6d867 100644 --- a/drivers/char/ps3flash.c +++ b/drivers/char/ps3flash.c @@ -25,6 +25,7 @@ #include asm/lv1call.h #include asm/ps3stor.h +#include asm/firmware.h #define DEVICE_NAME ps3flash @@ -464,6 +465,12 @@ static struct ps3_system_bus_driver ps3flash = { static int __init ps3flash_init(void) { + if (!firmware_has_feature(FW_FEATURE_PS3_LV1)) + return -ENODEV; + + if (ps3_get_ss_laid() != PS3_SS_LAID_OTHEROS) + return -ENODEV; + return ps3_system_bus_driver_register(ps3flash); } ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [Cbe-oss-dev] [PATCH part1 v2 4/9] Add region 1 memory early
On Tue, 23 Aug 2011 13:53:46 -0700 Geoff Levand ge...@infradead.org wrote: [...] + if (!map.r1.size) { + DBG(%s:%d: no region 1, not adding memory\n, + __func__, __LINE__); + } else { Remove brackets around a single line conditional. + DBG(%s:%d: adding memory: start %llxh, size %llxh\n, + __func__, __LINE__, map.rm.size, map.r1.size); + + memblock_add(map.rm.size, map.r1.size); + memblock_analyze(); + } + In Documentation/CodingStyle I read that if [only] one branch is a single statement then the parenthesis are OK (and even recommended) for both branches, I guess this is for style consistency. See Chapter 3, around line 169 on my copy. I guess the wording on that paragraph can be made more explicit, I'll try to fix that up. Regards, Antonio -- Antonio Ospite http://ao2.it PGP public key ID: 0x4553B001 A: Because it messes up the order in which people normally read text. See http://en.wikipedia.org/wiki/Posting_style Q: Why is top-posting such a bad thing? pgpC8bdEgALbX.pgp Description: PGP signature ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
On Tue, 2011-08-23 at 15:18 +0200, Roedel, Joerg wrote: On Mon, Aug 22, 2011 at 05:03:53PM -0400, Benjamin Herrenschmidt wrote: I am in favour of /dev/vfio/$GROUP. If multiple devices should be assigned to a guest, there can also be an ioctl to bind a group to an address-space of another group (certainly needs some care to not allow that both groups belong to different processes). Btw, a problem we havn't talked about yet entirely is driver-deassignment. User space can decide to de-assign the device from vfio while a fd is open on it. With PCI there is no way to let this fail (the .release function returns void last time i checked). Is this a problem, and yes, how we handle that? We can treat it as a hard unplug (like a cardbus gone away). IE. Dispose of the direct mappings (switch to MMIO emulation) and return all ff's from reads ( ignore writes). Then send an unplug event via whatever mechanism the platform provides (ACPI hotplug controller on x86 for example, we haven't quite sorted out what to do on power for hotplug yet). Hmm, good idea. But as far as I know the hotplug-event needs to be in the guest _before_ the device is actually unplugged (so that the guest can unbind its driver first). That somehow brings back the sleep-idea and the timeout in the .release function. That's for normal assisted hotplug, but don't we support hard hotplug ? I mean, things like cardbus, thunderbolt (if we ever support that) etc... will need it and some platforms do support hard hotplug of PCIe devices. (That's why drivers should never spin on MMIO waiting for a 1 bit to clear without a timeout :-) Cheers, Ben. ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
On Tue, 2011-08-23 at 10:23 -0600, Alex Williamson wrote: Yeah. Joerg's idea of binding groups internally (pass the fd of one group to another via ioctl) is one option. The tricky part will be implementing it to support hot unplug of any group from the supergroup. I believe Ben had a suggestion that supergroups could be created in sysfs, but I don't know what the mechanism to do that looks like. It would also be an extra management step to dynamically bind and unbind groups to the supergroup around hotplug. Thanks, I don't really care that much what the method for creating them is, to be honest, I just prefer this concept of meta groups or super groups or synthetic groups (whatever you want to name them) to having a separate uiommu file descriptor. The one reason I have a slight preference for creating them statically using some kind of separate interface (again, I don't care whether it's sysfs, netlink, etc...) is that it means things like qemu don't have to care about them. In general, apps that want to use vfio can just get passed the path to such a group or the /dev/ path or the group number (whatever we chose as the way to identify a group), and don't need to know anything about super groups, how to manipulate them, create them, possible constraints etc... Now, libvirt might want to know about that other API in order to provide control on the creation of these things, but that's a different issue. By static I mean they persist, they aren't tied to the lifetime of an fd. Now that's purely a preference on my side because I believe it will make life easier for actual programs wanting to use vfio to not have to care about those super-groups, but as I said earlier, I don't actually care that much :-) Cheers, Ben. ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
For us the most simple and logical approach (which is also what pHyp uses and what Linux handles well) is really to expose a given PCI host bridge per group to the guest. Believe it or not, it makes things easier :-) I'm all for easier. Why does exposing the bridge use less bus numbers than emulating a bridge? Because a host bridge doesn't look like a PCI to PCI bridge at all for us. It's an entire separate domain with it's own bus number space (unlike most x86 setups). In fact we have some problems afaik in qemu today with the concept of PCI domains, for example, I think qemu has assumptions about a single shared IO space domain which isn't true for us (each PCI host bridge provides a distinct IO space domain starting at 0). We'll have to fix that, but it's not a huge deal. So for each group we'd expose in the guest an entire separate PCI domain space with its own IO, MMIO etc... spaces, handed off from a single device-tree host bridge which doesn't itself appear in the config space, doesn't need any emulation of any config space etc... On x86, I want to maintain that our default assignment is at the device level. A user should be able to pick single or multiple devices from across several groups and have them all show up as individual, hotpluggable devices on bus 0 in the guest. Not surprisingly, we've also seen cases where users try to attach a bridge to the guest, assuming they'll get all the devices below the bridge, so I'd be in favor of making this just work if possible too, though we may have to prevent hotplug of those. Given the device requirement on x86 and since everything is a PCI device on x86, I'd like to keep a qemu command line something like -device vfio,host=00:19.0. I assume that some of the iommu properties, such as dma window size/address, will be query-able through an architecture specific (or general if possible) ioctl on the vfio group fd. I hope that will help the specification, but I don't fully understand what all remains. Thanks, Well, for iommu there's a couple of different issues here but yes, basically on one side we'll have some kind of ioctl to know what segment of the device(s) DMA address space is assigned to the group and we'll need to represent that to the guest via a device-tree property in some kind of parent node of all the devices in that group. We -might- be able to implement some kind of hotplug of individual devices of a group under such a PHB (PCI Host Bridge), I don't know for sure yet, some of that PAPR stuff is pretty arcane, but basically, for all intend and purpose, we really want a group to be represented as a PHB in the guest. We cannot arbitrary have individual devices of separate groups be represented in the guest as siblings on a single simulated PCI bus. Cheers, Ben. ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [Cbe-oss-dev] [PATCH part1 v2 4/9] Add region 1 memory early
Hi Antonio, On 08/23/2011 03:37 PM, Antonio Ospite wrote: Geoff Levand ge...@infradead.org wrote: + if (!map.r1.size) { + DBG(%s:%d: no region 1, not adding memory\n, + __func__, __LINE__); + } else { Remove brackets around a single line conditional. + DBG(%s:%d: adding memory: start %llxh, size %llxh\n, + __func__, __LINE__, map.rm.size, map.r1.size); + + memblock_add(map.rm.size, map.r1.size); + memblock_analyze(); + } + In Documentation/CodingStyle I read that if [only] one branch is a single statement then the parenthesis are OK (and even recommended) for both branches, I guess this is for style consistency. See Chapter 3, around line 169 on my copy. I guess the wording on that paragraph can be made more explicit, I'll try to fix that up. Thanks for the comments. I don't think its such an important change, mainly for consistency of style within the PS3 files. -Geoff ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH v3] mtd/nand : workaround for Freescale FCM to support large-page Nand chip
于 2011年08月23日 18:02, Matthieu CASTET 写道: LiuShuo a écrit : 于 2011年08月19日 00:25, Scott Wood 写道: On 08/17/2011 09:33 PM, b35...@freescale.com wrote: From: Liu Shuob35...@freescale.com Freescale FCM controller has a 2K size limitation of buffer RAM. In order to support the Nand flash chip whose page size is larger than 2K bytes, we divide a page into multi-2K pages for MTD layer driver. In that case, we force to set the page size to 2K bytes. We convert the page address of MTD layer driver to a real page address in flash chips and a column index in fsl_elbc driver. We can issue any column address by UA instruction of elbc controller. NOTE: Due to there is a limitation of 'Number of Partial Program Cycles in the Same Page (NOP)', the flash chip which is supported by this workaround have to meet below conditions. 1. page size is not greater than 4KB 2. 1) if main area and spare area have independent NOPs: main area NOP:=3 spare area NOP:=2? How often are the NOPs split like this? 2) if main area and spare area have a common NOP: NOP :=4 This depends on how the flash is used. If you treat it as a NOP1 flash (e.g. run ubifs rather than jffs2), then you need NOP2 for a 4K chip and NOP4 for an 8K chip. OTOH, if you would be making full use of NOP4 on a real 2K chip, you'll need NOP8 for a 4K chip. The NOP restrictions should be documented in the code itself, not just in the git changelog. Maybe print it to the console when this hack is used, along with the NOP value read from the ID. We can't read the NOP from the ID on any chip. Some chips don't give this infomation.(e.g. Micron MT29F4G08BAC) Doesn't the micron chip provide it with onfi info ? Sorry, there is something wrong with my expression. We can get the NOP info from datasheet, but can't get it by READID command in code. -LiuShuo Matthieu ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
On 23.08.2011, at 18:41, Benjamin Herrenschmidt wrote: On Tue, 2011-08-23 at 10:23 -0600, Alex Williamson wrote: Yeah. Joerg's idea of binding groups internally (pass the fd of one group to another via ioctl) is one option. The tricky part will be implementing it to support hot unplug of any group from the supergroup. I believe Ben had a suggestion that supergroups could be created in sysfs, but I don't know what the mechanism to do that looks like. It would also be an extra management step to dynamically bind and unbind groups to the supergroup around hotplug. Thanks, I don't really care that much what the method for creating them is, to be honest, I just prefer this concept of meta groups or super groups or synthetic groups (whatever you want to name them) to having a separate uiommu file descriptor. The one reason I have a slight preference for creating them statically using some kind of separate interface (again, I don't care whether it's sysfs, netlink, etc...) is that it means things like qemu don't have to care about them. In general, apps that want to use vfio can just get passed the path to such a group or the /dev/ path or the group number (whatever we chose as the way to identify a group), and don't need to know anything about super groups, how to manipulate them, create them, possible constraints etc... Now, libvirt might want to know about that other API in order to provide control on the creation of these things, but that's a different issue. By static I mean they persist, they aren't tied to the lifetime of an fd. Now that's purely a preference on my side because I believe it will make life easier for actual programs wanting to use vfio to not have to care about those super-groups, but as I said earlier, I don't actually care that much :-) Oh I think it's one of the building blocks we need for a sane user space device exposure API. If I want to pass user X a few devices that are all behind a single IOMMU, I just chown that device node to user X and be done with it. The user space tool actually using the VFIO interface wouldn't be in configuration business then - and it really shouldn't. That's what system configuration is there for :). But I'm fairly sure we managed to persuade Alex that this is the right path on the BOF :) Alex ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: kvm PCI assignment VFIO ramblings
On 23.08.2011, at 18:51, Benjamin Herrenschmidt wrote: For us the most simple and logical approach (which is also what pHyp uses and what Linux handles well) is really to expose a given PCI host bridge per group to the guest. Believe it or not, it makes things easier :-) I'm all for easier. Why does exposing the bridge use less bus numbers than emulating a bridge? Because a host bridge doesn't look like a PCI to PCI bridge at all for us. It's an entire separate domain with it's own bus number space (unlike most x86 setups). In fact we have some problems afaik in qemu today with the concept of PCI domains, for example, I think qemu has assumptions about a single shared IO space domain which isn't true for us (each PCI host bridge provides a distinct IO space domain starting at 0). We'll have to fix that, but it's not a huge deal. So for each group we'd expose in the guest an entire separate PCI domain space with its own IO, MMIO etc... spaces, handed off from a single device-tree host bridge which doesn't itself appear in the config space, doesn't need any emulation of any config space etc... On x86, I want to maintain that our default assignment is at the device level. A user should be able to pick single or multiple devices from across several groups and have them all show up as individual, hotpluggable devices on bus 0 in the guest. Not surprisingly, we've also seen cases where users try to attach a bridge to the guest, assuming they'll get all the devices below the bridge, so I'd be in favor of making this just work if possible too, though we may have to prevent hotplug of those. Given the device requirement on x86 and since everything is a PCI device on x86, I'd like to keep a qemu command line something like -device vfio,host=00:19.0. I assume that some of the iommu properties, such as dma window size/address, will be query-able through an architecture specific (or general if possible) ioctl on the vfio group fd. I hope that will help the specification, but I don't fully understand what all remains. Thanks, Well, for iommu there's a couple of different issues here but yes, basically on one side we'll have some kind of ioctl to know what segment of the device(s) DMA address space is assigned to the group and we'll need to represent that to the guest via a device-tree property in some kind of parent node of all the devices in that group. We -might- be able to implement some kind of hotplug of individual devices of a group under such a PHB (PCI Host Bridge), I don't know for sure yet, some of that PAPR stuff is pretty arcane, but basically, for all intend and purpose, we really want a group to be represented as a PHB in the guest. We cannot arbitrary have individual devices of separate groups be represented in the guest as siblings on a single simulated PCI bus. So would it make sense for you to go the same route that we need to go on embedded power, with a separate VFIO style interface that simply exports memory ranges and irq bindings, but doesn't know anything about PCI? For e500, we'll be using something like that to pass through a full PCI bus into the system. Alex ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH 2/2] [PowerPC Book3E] Introduce new ptrace debug feature flag
On Tue, Aug 23, 2011 at 02:57:56PM +0530, K.Prasad wrote: On Tue, Aug 23, 2011 at 03:09:31PM +1000, David Gibson wrote: On Fri, Aug 19, 2011 at 01:23:38PM +0530, K.Prasad wrote: While PPC_PTRACE_SETHWDEBUG ptrace flag in PowerPC accepts PPC_BREAKPOINT_MODE_EXACT mode of breakpoint, the same is not intimated to the user-space debuggers (like GDB) who may want to use it. Hence we introduce a new PPC_DEBUG_FEATURE_DATA_BP_EXACT flag which will be populated on the features member of struct ppc_debug_info to advertise support for the same on Book3E PowerPC processors. I thought the idea was that the BP_EXACT mode was the default - if the new interface was supported at all, then BP_EXACT was always supported. So, why do you need a new flag? Yes, BP_EXACT was always supported but not advertised through PPC_PTRACE_GETHWDBGINFO. We're now doing that. I can see that. But you haven't answered why. -- David Gibson| I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_ | _way_ _around_! http://www.ozlabs.org/~dgibson ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
Re: [PATCH 1/2] [hw-breakpoint] Use generic hw-breakpoint interfaces for new PPC ptrace flags
On Tue, Aug 23, 2011 at 02:55:13PM +0530, K.Prasad wrote: On Tue, Aug 23, 2011 at 03:08:50PM +1000, David Gibson wrote: On Fri, Aug 19, 2011 at 01:21:36PM +0530, K.Prasad wrote: PPC_PTRACE_GETHWDBGINFO, PPC_PTRACE_SETHWDEBUG and PPC_PTRACE_DELHWDEBUG are PowerPC specific ptrace flags that use the watchpoint register. While they are targeted primarily towards BookE users, user-space applications such as GDB have started using them for BookS too. This patch enables the use of generic hardware breakpoint interfaces for these new flags. The version number of the associated data structures ppc_hw_breakpoint and ppc_debug_info is incremented to denote new semantics. So, the structure itself doesn't seem to have been extended. I don't understand what the semantic difference is - your patch comment needs to explain this clearly. We had a request to extend the structure but thought it was dangerous to do so. For instance if the user-space used version1 of the structure, while kernel did a copy_to_user() pertaining to version2, then we'd run into problems. Unfortunately the ptrace flags weren't designed to accept a version number as input from the user through the PPC_PTRACE_GETHWDBGINFO flag (which would have solved this issue). I still don't follow you. I'll add a comment w.r.t change in semantics - such as the ability to accept 'range' breakpoints in BookS. Apart from the usual benefits of using generic hw-breakpoint interfaces, these changes allow debuggers (such as GDB) to use a common set of ptrace flags for their watchpoint needs and allow more precise breakpoint specification (length of the variable can be specified). What is the mechanism for implementing the range breakpoint on book3s? The hw-breakpoint interface, accepts length as an argument in BookS (any value = 8 Bytes) and would filter out extraneous interrupts arising out of accesses outside the range comprising addr, addr + len inside hw_breakpoint_handler function. We put that ability to use here. Ah, so in hardware the breakpoints are always 8 bytes long, but you filter out false hits on a shorter range? Of course, the utility of range breakpoints is questionable when length =8, but the start must be aligned on an 8-byte boundary. [snip] if ((unsigned long)bp_info-addr = TASK_SIZE) return -EIO; @@ -1398,15 +1400,86 @@ static long ppc_set_hwdebug(struct task_struct *child, dabr |= DABR_DATA_READ; if (bp_info-trigger_type PPC_BREAKPOINT_TRIGGER_WRITE) dabr |= DABR_DATA_WRITE; +#ifdef CONFIG_HAVE_HW_BREAKPOINT + if (bp_info-version == 1) + goto version_one; There are several legitimate uses of goto in the kernel, but this is definitely not one of them. You're essentially using it to put the old and new versions of the same function in one block. Nasty. Maybe it's the label that's causing bother here. It might look elegant if it was called something like exit_* or error_* :-) The goto here helps reduce code, is similar to the error exits we use everywhere. Rubbish, it is not an exception exit at all, it is two separate code paths for the different versions which would be much clearer as two different functions. + if (ptrace_get_breakpoints(child) 0) + return -ESRCH; - child-thread.dabr = dabr; + bp = thread-ptrace_bps[0]; + if (!bp_info-addr) { + if (bp) { + unregister_hw_breakpoint(bp); + thread-ptrace_bps[0] = NULL; + } + ptrace_put_breakpoints(child); + return 0; Why are you making setting a 0 watchpoint remove the existing one (I think that's what this does). I thought there was an explicit del breakpoint operation instead. We had to define the semantics for what writing a 0 to DABR could mean, and I think it is intuitive to consider it as deletion request...couldn't think of a case where DABR with addr=0 and RW=1 would be required. When a user space program maps pages at virtual address 0, which it can do. + } + /* + * Check if the request is for 'range' breakpoints. We can + * support it if range 8 bytes. + */ + if (bp_info-addr_mode == PPC_BREAKPOINT_MODE_RANGE_INCLUSIVE) + len = bp_info-addr2 - bp_info-addr; So you compute the length here, but I don't see you ever test if it is 8 and return an error. The hw-breakpoint interfaces would fail if the length was 8. Ok. + else if (bp_info-addr_mode != PPC_BREAKPOINT_MODE_EXACT) { + ptrace_put_breakpoints(child); + return -EINVAL; + } + if (bp) { + attr = bp-attr; + attr.bp_addr = (unsigned long)bp_info-addr ~HW_BREAKPOINT_ALIGN; + arch_bp_generic_fields(dabr + (DABR_DATA_WRITE |