| Message ID | 20180928064841.14117-7-takahiro.akashi@linaro.org |
|---|---|
| State | New |
| Headers | show |
| Series | None | expand |
+David Gibson On Fri, Sep 28, 2018 at 1:48 AM AKASHI Takahiro <takahiro.akashi@linaro.org> wrote: > > These functions will be used later to handle kexec-specific properties > in arm64's kexec_file implementation. > > Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org> > Cc: Rob Herring <robh+dt@kernel.org> > Cc: Frank Rowand <frowand.list@gmail.com> > Cc: devicetree@vger.kernel.org > --- > drivers/of/fdt.c | 56 ++++++++++++++++++++++++++++++++++++++++++ > include/linux/of_fdt.h | 4 +++ > 2 files changed, 60 insertions(+) > > diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c > index 800ad252cf9c..c65c31562ccb 100644 > --- a/drivers/of/fdt.c > +++ b/drivers/of/fdt.c > @@ -25,6 +25,7 @@ > #include <linux/debugfs.h> > #include <linux/serial_core.h> > #include <linux/sysfs.h> > +#include <linux/types.h> > > #include <asm/setup.h> /* for COMMAND_LINE_SIZE */ > #include <asm/page.h> > @@ -1323,3 +1324,58 @@ late_initcall(of_fdt_raw_init); > #endif > > #endif /* CONFIG_OF_EARLY_FLATTREE */ > + > +#define FDT_ALIGN(x, a) (((x) + (a) - 1) & ~((a) - 1)) > +#define FDT_TAGALIGN(x) (FDT_ALIGN((x), FDT_TAGSIZE)) > + > +int fdt_prop_len(const char *prop_name, int len) > +{ > + return (strlen(prop_name) + 1) + > + sizeof(struct fdt_property) + > + FDT_TAGALIGN(len); Looks like you are using this to calculate how much space you need to allocate in addition to the current DTB for a couple of new or replaced properties. I'm not sure that this calculation is completely accurate. And it is strange there doesn't seem to be any libfdt function for this already. It would be simpler to just add some fixed additional amount. Maybe David G has comments on this? > +} > + The rest of this should go in drivers/of/fdt_address.c. Ultimately, it should go into libfdt, but I'm fine with having it in the kernel for now. > +static void fill_property(void *buf, u64 val64, int cells) > +{ > + __be32 val32; > + > + while (cells) { > + val32 = cpu_to_fdt32((val64 >> (32 * (--cells))) & U32_MAX); > + memcpy(buf, &val32, sizeof(val32)); > + buf += sizeof(val32); This is kind of hard to read. I would copy u-boot's fdt_pack_reg function. BTW, for purposes of moving to libfdt, we'll need the authors' (Masahiro Yamada and Hans de Goede) permission to dual license. > + } > +} > + > +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, > + u64 addr, u64 size) > +{ > + int addr_cells, size_cells; > + char buf[sizeof(__be32) * 2 * 2]; > + /* assume dt_root_[addr|size]_cells <= 2 */ > + void *prop; > + size_t buf_size; > + > + addr_cells = fdt_address_cells(fdt, 0); > + if (addr_cells < 0) > + return addr_cells; > + size_cells = fdt_size_cells(fdt, 0); > + if (size_cells < 0) > + return size_cells; > + > + /* if *_cells >= 2, cells can hold 64-bit values anyway */ > + if ((addr_cells == 1) && (addr > U32_MAX)) > + return -FDT_ERR_BADVALUE; > + > + if ((size_cells == 1) && (size > U32_MAX)) > + return -FDT_ERR_BADVALUE; > + > + buf_size = (addr_cells + size_cells) * sizeof(u32); > + prop = buf; > + > + fill_property(prop, addr, addr_cells); > + prop += addr_cells * sizeof(u32); > + > + fill_property(prop, size, size_cells); > + > + return fdt_setprop(fdt, nodeoffset, name, buf, buf_size); > +} > diff --git a/include/linux/of_fdt.h b/include/linux/of_fdt.h > index b9cd9ebdf9b9..842af6ea92ea 100644 > --- a/include/linux/of_fdt.h > +++ b/include/linux/of_fdt.h > @@ -108,5 +108,9 @@ static inline void unflatten_device_tree(void) {} > static inline void unflatten_and_copy_device_tree(void) {} > #endif /* CONFIG_OF_EARLY_FLATTREE */ > > +int fdt_prop_len(const char *prop_name, int len); > +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, > + u64 addr, u64 size); > + > #endif /* __ASSEMBLY__ */ > #endif /* _LINUX_OF_FDT_H */ > -- > 2.19.0 >
On 09/28/18 06:44, Rob Herring wrote: > +David Gibson > > On Fri, Sep 28, 2018 at 1:48 AM AKASHI Takahiro > <takahiro.akashi@linaro.org> wrote: >> >> These functions will be used later to handle kexec-specific properties >> in arm64's kexec_file implementation. As I requested in version 14: The intent of the helper functions is related to properties whose values are tuples of the same format as the "reg" property of the "/memory" nodes. For example, the "linux,usable-memory-range" and "linux,elfcoredhr" properties of the "/chosen" node. The patch header and the function names should be updated to reflect this intent. >> >> Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org> >> Cc: Rob Herring <robh+dt@kernel.org> >> Cc: Frank Rowand <frowand.list@gmail.com> >> Cc: devicetree@vger.kernel.org >> --- Missing list of changes since version 14. >> drivers/of/fdt.c | 56 ++++++++++++++++++++++++++++++++++++++++++ >> include/linux/of_fdt.h | 4 +++ >> 2 files changed, 60 insertions(+) >> >> diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c >> index 800ad252cf9c..c65c31562ccb 100644 >> --- a/drivers/of/fdt.c >> +++ b/drivers/of/fdt.c >> @@ -25,6 +25,7 @@ >> #include <linux/debugfs.h> >> #include <linux/serial_core.h> >> #include <linux/sysfs.h> >> +#include <linux/types.h> >> >> #include <asm/setup.h> /* for COMMAND_LINE_SIZE */ >> #include <asm/page.h> >> @@ -1323,3 +1324,58 @@ late_initcall(of_fdt_raw_init); >> #endif >> >> #endif /* CONFIG_OF_EARLY_FLATTREE */ In v14 I requested: Please add comment: /* helper functions for arm64 kexec */ >> + >> +#define FDT_ALIGN(x, a) (((x) + (a) - 1) & ~((a) - 1)) >> +#define FDT_TAGALIGN(x) (FDT_ALIGN((x), FDT_TAGSIZE)) >> + >> +int fdt_prop_len(const char *prop_name, int len) In v14, I requested: Please rename as fdt_len_added_prop() I'm not really happy with my suggested name, but do not have a better one yet. As Rob notes, maybe David G will have a helpful comment. >> +{ >> + return (strlen(prop_name) + 1) + >> + sizeof(struct fdt_property) + >> + FDT_TAGALIGN(len); > > Looks like you are using this to calculate how much space you need to > allocate in addition to the current DTB for a couple of new or > replaced properties. I'm not sure that this calculation is completely > accurate. And it is strange there doesn't seem to be any libfdt > function for this already. It would be simpler to just add some fixed > additional amount. > > Maybe David G has comments on this? > >> +} >> + > > The rest of this should go in drivers/of/fdt_address.c. Ultimately, it > should go into libfdt, but I'm fine with having it in the kernel for > now. > >> +static void fill_property(void *buf, u64 val64, int cells) In v14 I requested: Please rename as cpu64_to_fdt_cells() >> +{ >> + __be32 val32; >> + >> + while (cells) { >> + val32 = cpu_to_fdt32((val64 >> (32 * (--cells))) & U32_MAX); >> + memcpy(buf, &val32, sizeof(val32)); >> + buf += sizeof(val32); > > This is kind of hard to read. I would copy u-boot's fdt_pack_reg function. > > BTW, for purposes of moving to libfdt, we'll need the authors' > (Masahiro Yamada and Hans de Goede) permission to dual license. > >> + } >> +} >> + >> +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, >> + u64 addr, u64 size) >> +{ >> + int addr_cells, size_cells; unsigned >> + char buf[sizeof(__be32) * 2 * 2]; >> + /* assume dt_root_[addr|size]_cells <= 2 */ >> + void *prop; >> + size_t buf_size; >> + >> + addr_cells = fdt_address_cells(fdt, 0); >> + if (addr_cells < 0) >> + return addr_cells; >> + size_cells = fdt_size_cells(fdt, 0); >> + if (size_cells < 0) >> + return size_cells; >> + >> + /* if *_cells >= 2, cells can hold 64-bit values anyway */ >> + if ((addr_cells == 1) && (addr > U32_MAX)) >> + return -FDT_ERR_BADVALUE; >> + >> + if ((size_cells == 1) && (size > U32_MAX)) >> + return -FDT_ERR_BADVALUE; In v14 I requested: Should also check that base + size does not wrap around. >> + >> + buf_size = (addr_cells + size_cells) * sizeof(u32); >> + prop = buf; >> + >> + fill_property(prop, addr, addr_cells); >> + prop += addr_cells * sizeof(u32); >> + >> + fill_property(prop, size, size_cells); >> + >> + return fdt_setprop(fdt, nodeoffset, name, buf, buf_size); >> +} >> diff --git a/include/linux/of_fdt.h b/include/linux/of_fdt.h >> index b9cd9ebdf9b9..842af6ea92ea 100644 >> --- a/include/linux/of_fdt.h >> +++ b/include/linux/of_fdt.h >> @@ -108,5 +108,9 @@ static inline void unflatten_device_tree(void) {} >> static inline void unflatten_and_copy_device_tree(void) {} >> #endif /* CONFIG_OF_EARLY_FLATTREE */ >> >> +int fdt_prop_len(const char *prop_name, int len); >> +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, >> + u64 addr, u64 size); >> + >> #endif /* __ASSEMBLY__ */ >> #endif /* _LINUX_OF_FDT_H */ >> -- >> 2.19.0 >> >
On Fri, Sep 28, 2018 at 08:44:42AM -0500, Rob Herring wrote: > +David Gibson > > On Fri, Sep 28, 2018 at 1:48 AM AKASHI Takahiro > <takahiro.akashi@linaro.org> wrote: > > > > These functions will be used later to handle kexec-specific properties > > in arm64's kexec_file implementation. > > > > Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org> > > Cc: Rob Herring <robh+dt@kernel.org> > > Cc: Frank Rowand <frowand.list@gmail.com> > > Cc: devicetree@vger.kernel.org > > --- > > drivers/of/fdt.c | 56 ++++++++++++++++++++++++++++++++++++++++++ > > include/linux/of_fdt.h | 4 +++ > > 2 files changed, 60 insertions(+) > > > > diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c > > index 800ad252cf9c..c65c31562ccb 100644 > > --- a/drivers/of/fdt.c > > +++ b/drivers/of/fdt.c > > @@ -25,6 +25,7 @@ > > #include <linux/debugfs.h> > > #include <linux/serial_core.h> > > #include <linux/sysfs.h> > > +#include <linux/types.h> > > > > #include <asm/setup.h> /* for COMMAND_LINE_SIZE */ > > #include <asm/page.h> > > @@ -1323,3 +1324,58 @@ late_initcall(of_fdt_raw_init); > > #endif > > > > #endif /* CONFIG_OF_EARLY_FLATTREE */ > > + > > +#define FDT_ALIGN(x, a) (((x) + (a) - 1) & ~((a) - 1)) > > +#define FDT_TAGALIGN(x) (FDT_ALIGN((x), FDT_TAGSIZE)) > > + > > +int fdt_prop_len(const char *prop_name, int len) > > +{ > > + return (strlen(prop_name) + 1) + > > + sizeof(struct fdt_property) + > > + FDT_TAGALIGN(len); > > Looks like you are using this to calculate how much space you need to > allocate in addition to the current DTB for a couple of new or > replaced properties. I'm not sure that this calculation is completely > accurate. And it is strange there doesn't seem to be any libfdt > function for this already. It would be simpler to just add some fixed > additional amount. > > Maybe David G has comments on this? Hrm. Yeah, the approach I'd suggest would be to just add a fixed amount, and if your functions return -FDT_ERR_NOSPACE, add some more. That's pretty much the expected way of working with fdts in write mode. If that's really not suitable, we could look at adding a helper to estimate this in libfdt. Or, actually, I suspect taking an existing libfdt internal helper, polishing and exporting it. > > > +} > > + > > The rest of this should go in drivers/of/fdt_address.c. Ultimately, it > should go into libfdt, but I'm fine with having it in the kernel for > now. > > > +static void fill_property(void *buf, u64 val64, int cells) > > +{ > > + __be32 val32; > > + > > + while (cells) { > > + val32 = cpu_to_fdt32((val64 >> (32 * (--cells))) & U32_MAX); > > + memcpy(buf, &val32, sizeof(val32)); > > + buf += sizeof(val32); > > This is kind of hard to read. I would copy u-boot's fdt_pack_reg function. > > BTW, for purposes of moving to libfdt, we'll need the authors' > (Masahiro Yamada and Hans de Goede) permission to dual license. > > > + } > > +} > > + > > +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, > > + u64 addr, u64 size) > > +{ > > + int addr_cells, size_cells; > > + char buf[sizeof(__be32) * 2 * 2]; > > + /* assume dt_root_[addr|size]_cells <= 2 */ > > + void *prop; > > + size_t buf_size; > > + > > + addr_cells = fdt_address_cells(fdt, 0); > > + if (addr_cells < 0) > > + return addr_cells; > > + size_cells = fdt_size_cells(fdt, 0); > > + if (size_cells < 0) > > + return size_cells; > > + > > + /* if *_cells >= 2, cells can hold 64-bit values anyway */ > > + if ((addr_cells == 1) && (addr > U32_MAX)) > > + return -FDT_ERR_BADVALUE; > > + > > + if ((size_cells == 1) && (size > U32_MAX)) > > + return -FDT_ERR_BADVALUE; > > + > > + buf_size = (addr_cells + size_cells) * sizeof(u32); > > + prop = buf; > > + > > + fill_property(prop, addr, addr_cells); > > + prop += addr_cells * sizeof(u32); > > + > > + fill_property(prop, size, size_cells); > > + > > + return fdt_setprop(fdt, nodeoffset, name, buf, buf_size); > > +} > > diff --git a/include/linux/of_fdt.h b/include/linux/of_fdt.h > > index b9cd9ebdf9b9..842af6ea92ea 100644 > > --- a/include/linux/of_fdt.h > > +++ b/include/linux/of_fdt.h > > @@ -108,5 +108,9 @@ static inline void unflatten_device_tree(void) {} > > static inline void unflatten_and_copy_device_tree(void) {} > > #endif /* CONFIG_OF_EARLY_FLATTREE */ > > > > +int fdt_prop_len(const char *prop_name, int len); > > +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, > > + u64 addr, u64 size); > > + > > #endif /* __ASSEMBLY__ */ > > #endif /* _LINUX_OF_FDT_H */ > > > -- 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
On Tue, Oct 02, 2018 at 02:47:10PM +1000, David Gibson wrote: > On Fri, Sep 28, 2018 at 08:44:42AM -0500, Rob Herring wrote: > > +David Gibson > > > > On Fri, Sep 28, 2018 at 1:48 AM AKASHI Takahiro > > <takahiro.akashi@linaro.org> wrote: > > > > > > These functions will be used later to handle kexec-specific properties > > > in arm64's kexec_file implementation. > > > > > > Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org> > > > Cc: Rob Herring <robh+dt@kernel.org> > > > Cc: Frank Rowand <frowand.list@gmail.com> > > > Cc: devicetree@vger.kernel.org > > > --- > > > drivers/of/fdt.c | 56 ++++++++++++++++++++++++++++++++++++++++++ > > > include/linux/of_fdt.h | 4 +++ > > > 2 files changed, 60 insertions(+) > > > > > > diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c > > > index 800ad252cf9c..c65c31562ccb 100644 > > > --- a/drivers/of/fdt.c > > > +++ b/drivers/of/fdt.c > > > @@ -25,6 +25,7 @@ > > > #include <linux/debugfs.h> > > > #include <linux/serial_core.h> > > > #include <linux/sysfs.h> > > > +#include <linux/types.h> > > > > > > #include <asm/setup.h> /* for COMMAND_LINE_SIZE */ > > > #include <asm/page.h> > > > @@ -1323,3 +1324,58 @@ late_initcall(of_fdt_raw_init); > > > #endif > > > > > > #endif /* CONFIG_OF_EARLY_FLATTREE */ > > > + > > > +#define FDT_ALIGN(x, a) (((x) + (a) - 1) & ~((a) - 1)) > > > +#define FDT_TAGALIGN(x) (FDT_ALIGN((x), FDT_TAGSIZE)) > > > + > > > +int fdt_prop_len(const char *prop_name, int len) > > > +{ > > > + return (strlen(prop_name) + 1) + > > > + sizeof(struct fdt_property) + > > > + FDT_TAGALIGN(len); > > > > Looks like you are using this to calculate how much space you need to > > allocate in addition to the current DTB for a couple of new or > > replaced properties. I'm not sure that this calculation is completely > > accurate. And it is strange there doesn't seem to be any libfdt > > function for this already. It would be simpler to just add some fixed > > additional amount. > > > > Maybe David G has comments on this? > > Hrm. Yeah, the approach I'd suggest would be to just add a fixed > amount, and if your functions return -FDT_ERR_NOSPACE, add some more. > That's pretty much the expected way of working with fdts in write > mode. I will try to follow this approach. Thanks, -Takahiro Akashi > If that's really not suitable, we could look at adding a helper to > estimate this in libfdt. Or, actually, I suspect taking an existing > libfdt internal helper, polishing and exporting it. > > > > > > +} > > > + > > > > The rest of this should go in drivers/of/fdt_address.c. Ultimately, it > > should go into libfdt, but I'm fine with having it in the kernel for > > now. > > > > > +static void fill_property(void *buf, u64 val64, int cells) > > > +{ > > > + __be32 val32; > > > + > > > + while (cells) { > > > + val32 = cpu_to_fdt32((val64 >> (32 * (--cells))) & U32_MAX); > > > + memcpy(buf, &val32, sizeof(val32)); > > > + buf += sizeof(val32); > > > > This is kind of hard to read. I would copy u-boot's fdt_pack_reg function. > > > > BTW, for purposes of moving to libfdt, we'll need the authors' > > (Masahiro Yamada and Hans de Goede) permission to dual license. > > > > > + } > > > +} > > > + > > > +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, > > > + u64 addr, u64 size) > > > +{ > > > + int addr_cells, size_cells; > > > + char buf[sizeof(__be32) * 2 * 2]; > > > + /* assume dt_root_[addr|size]_cells <= 2 */ > > > + void *prop; > > > + size_t buf_size; > > > + > > > + addr_cells = fdt_address_cells(fdt, 0); > > > + if (addr_cells < 0) > > > + return addr_cells; > > > + size_cells = fdt_size_cells(fdt, 0); > > > + if (size_cells < 0) > > > + return size_cells; > > > + > > > + /* if *_cells >= 2, cells can hold 64-bit values anyway */ > > > + if ((addr_cells == 1) && (addr > U32_MAX)) > > > + return -FDT_ERR_BADVALUE; > > > + > > > + if ((size_cells == 1) && (size > U32_MAX)) > > > + return -FDT_ERR_BADVALUE; > > > + > > > + buf_size = (addr_cells + size_cells) * sizeof(u32); > > > + prop = buf; > > > + > > > + fill_property(prop, addr, addr_cells); > > > + prop += addr_cells * sizeof(u32); > > > + > > > + fill_property(prop, size, size_cells); > > > + > > > + return fdt_setprop(fdt, nodeoffset, name, buf, buf_size); > > > +} > > > diff --git a/include/linux/of_fdt.h b/include/linux/of_fdt.h > > > index b9cd9ebdf9b9..842af6ea92ea 100644 > > > --- a/include/linux/of_fdt.h > > > +++ b/include/linux/of_fdt.h > > > @@ -108,5 +108,9 @@ static inline void unflatten_device_tree(void) {} > > > static inline void unflatten_and_copy_device_tree(void) {} > > > #endif /* CONFIG_OF_EARLY_FLATTREE */ > > > > > > +int fdt_prop_len(const char *prop_name, int len); > > > +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, > > > + u64 addr, u64 size); > > > + > > > #endif /* __ASSEMBLY__ */ > > > #endif /* _LINUX_OF_FDT_H */ > > > > > > > -- > 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
Frank, # I haven't reply to your comments. On Fri, Sep 28, 2018 at 02:13:58PM -0700, Frank Rowand wrote: > On 09/28/18 06:44, Rob Herring wrote: > > +David Gibson > > > > On Fri, Sep 28, 2018 at 1:48 AM AKASHI Takahiro > > <takahiro.akashi@linaro.org> wrote: > >> > >> These functions will be used later to handle kexec-specific properties > >> in arm64's kexec_file implementation. > > As I requested in version 14: > > The intent of the helper functions is related to properties whose values are > tuples of the same format as the "reg" property of the "/memory" nodes. For > example, the "linux,usable-memory-range" and "linux,elfcoredhr" properties of > the "/chosen" node. > > The patch header and the function names should be updated to reflect this intent. I agree regarding the patch header. > > >> > >> Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org> > >> Cc: Rob Herring <robh+dt@kernel.org> > >> Cc: Frank Rowand <frowand.list@gmail.com> > >> Cc: devicetree@vger.kernel.org > >> --- > > Missing list of changes since version 14. Sorry for the inconvenience, but a whole change list goes into the cover letter, not individual patches. > > >> drivers/of/fdt.c | 56 ++++++++++++++++++++++++++++++++++++++++++ > >> include/linux/of_fdt.h | 4 +++ > >> 2 files changed, 60 insertions(+) > >> > >> diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c > >> index 800ad252cf9c..c65c31562ccb 100644 > >> --- a/drivers/of/fdt.c > >> +++ b/drivers/of/fdt.c > >> @@ -25,6 +25,7 @@ > >> #include <linux/debugfs.h> > >> #include <linux/serial_core.h> > >> #include <linux/sysfs.h> > >> +#include <linux/types.h> > >> > >> #include <asm/setup.h> /* for COMMAND_LINE_SIZE */ > >> #include <asm/page.h> > >> @@ -1323,3 +1324,58 @@ late_initcall(of_fdt_raw_init); > >> #endif > >> > >> #endif /* CONFIG_OF_EARLY_FLATTREE */ > > In v14 I requested: > > Please add comment: > > /* helper functions for arm64 kexec */ Okay. > > >> + > >> +#define FDT_ALIGN(x, a) (((x) + (a) - 1) & ~((a) - 1)) > >> +#define FDT_TAGALIGN(x) (FDT_ALIGN((x), FDT_TAGSIZE)) > >> + > >> +int fdt_prop_len(const char *prop_name, int len) > > In v14, I requested: > > Please rename as fdt_len_added_prop() Anyhow, I will drop this function, preferring to new fdt_[address|size]_cells(). > I'm not really happy with my suggested name, but do not have a > better one yet. As Rob notes, maybe David G will have a helpful > comment. > > >> +{ > >> + return (strlen(prop_name) + 1) + > >> + sizeof(struct fdt_property) + > >> + FDT_TAGALIGN(len); > > > > Looks like you are using this to calculate how much space you need to > > allocate in addition to the current DTB for a couple of new or > > replaced properties. I'm not sure that this calculation is completely > > accurate. And it is strange there doesn't seem to be any libfdt > > function for this already. It would be simpler to just add some fixed > > additional amount. > > > > Maybe David G has comments on this? I'm not quit sure why it's not that accurate, but as I said in a reply to David's comment, I will take your suggestion. > >> +} > >> + > > > > The rest of this should go in drivers/of/fdt_address.c. Ultimately, it > > should go into libfdt, but I'm fine with having it in the kernel for > > now. > > > >> +static void fill_property(void *buf, u64 val64, int cells) > > In v14 I requested: > > Please rename as cpu64_to_fdt_cells() I don't mind, but this function may be dropped if Rob sticks to u-boot's fdt_pack_reg() over my fdt_setprop_reg(). > > >> +{ > >> + __be32 val32; > >> + > >> + while (cells) { > >> + val32 = cpu_to_fdt32((val64 >> (32 * (--cells))) & U32_MAX); > >> + memcpy(buf, &val32, sizeof(val32)); > >> + buf += sizeof(val32); > > > > This is kind of hard to read. I would copy u-boot's fdt_pack_reg function. > > > > BTW, for purposes of moving to libfdt, we'll need the authors' > > (Masahiro Yamada and Hans de Goede) permission to dual license. > > > >> + } > >> +} > >> + > >> +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, > >> + u64 addr, u64 size) > >> +{ > >> + int addr_cells, size_cells; > > unsigned fdt_[address|size]_cell() returns an int. > > >> + char buf[sizeof(__be32) * 2 * 2]; > >> + /* assume dt_root_[addr|size]_cells <= 2 */ > >> + void *prop; > >> + size_t buf_size; > >> + > >> + addr_cells = fdt_address_cells(fdt, 0); > >> + if (addr_cells < 0) > >> + return addr_cells; > >> + size_cells = fdt_size_cells(fdt, 0); > >> + if (size_cells < 0) > >> + return size_cells; > >> + > >> + /* if *_cells >= 2, cells can hold 64-bit values anyway */ > >> + if ((addr_cells == 1) && (addr > U32_MAX)) > >> + return -FDT_ERR_BADVALUE; > >> + > >> + if ((size_cells == 1) && (size > U32_MAX)) > >> + return -FDT_ERR_BADVALUE; > > In v14 I requested: > > Should also check that base + size does not wrap around. Okay, I will start discussion, as you have suggested, in devicetree-spec ML. Thanks, -Takahiro Akashi > > >> + > >> + buf_size = (addr_cells + size_cells) * sizeof(u32); > >> + prop = buf; > >> + > >> + fill_property(prop, addr, addr_cells); > >> + prop += addr_cells * sizeof(u32); > >> + > >> + fill_property(prop, size, size_cells); > >> + > >> + return fdt_setprop(fdt, nodeoffset, name, buf, buf_size); > >> +} > >> diff --git a/include/linux/of_fdt.h b/include/linux/of_fdt.h > >> index b9cd9ebdf9b9..842af6ea92ea 100644 > >> --- a/include/linux/of_fdt.h > >> +++ b/include/linux/of_fdt.h > >> @@ -108,5 +108,9 @@ static inline void unflatten_device_tree(void) {} > >> static inline void unflatten_and_copy_device_tree(void) {} > >> #endif /* CONFIG_OF_EARLY_FLATTREE */ > >> > >> +int fdt_prop_len(const char *prop_name, int len); > >> +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, > >> + u64 addr, u64 size); > >> + > >> #endif /* __ASSEMBLY__ */ > >> #endif /* _LINUX_OF_FDT_H */ > >> -- > >> 2.19.0 > >> > > >
On Fri, Oct 05, 2018 at 02:06:29PM +0900, AKASHI, Takahiro wrote: [snip] > > >> +{ > > >> + return (strlen(prop_name) + 1) + > > >> + sizeof(struct fdt_property) + > > >> + FDT_TAGALIGN(len); > > > > > > Looks like you are using this to calculate how much space you need to > > > allocate in addition to the current DTB for a couple of new or > > > replaced properties. I'm not sure that this calculation is completely > > > accurate. And it is strange there doesn't seem to be any libfdt > > > function for this already. It would be simpler to just add some fixed > > > additional amount. > > > > > > Maybe David G has comments on this? > > I'm not quit sure why it's not that accurate, but as I said in a reply to > David's comment, I will take your suggestion. I think it is accurate, but it is not robust. It might not be accurate if we ever make further tweaks to the libfdt format. libfdt is supposed to abstract those changes away, but if you make assumptions like this, you break that abstraction. -- 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
On Thu, Oct 4, 2018 at 10:07 PM AKASHI, Takahiro <takahiro.akashi@linaro.org> wrote: > > Rob, > > # I haven't replied to this comment yet. > > On Fri, Sep 28, 2018 at 08:44:42AM -0500, Rob Herring wrote: > > +David Gibson > > > > On Fri, Sep 28, 2018 at 1:48 AM AKASHI Takahiro > > <takahiro.akashi@linaro.org> wrote: > > > > > > These functions will be used later to handle kexec-specific properties > > > in arm64's kexec_file implementation. > > > > > > Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org> > > > Cc: Rob Herring <robh+dt@kernel.org> > > > Cc: Frank Rowand <frowand.list@gmail.com> > > > Cc: devicetree@vger.kernel.org > > > --- > > > drivers/of/fdt.c | 56 ++++++++++++++++++++++++++++++++++++++++++ > > > include/linux/of_fdt.h | 4 +++ > > > 2 files changed, 60 insertions(+) > > > > > > diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c > > > index 800ad252cf9c..c65c31562ccb 100644 > > > --- a/drivers/of/fdt.c > > > +++ b/drivers/of/fdt.c > > > @@ -25,6 +25,7 @@ > > > #include <linux/debugfs.h> > > > #include <linux/serial_core.h> > > > #include <linux/sysfs.h> > > > +#include <linux/types.h> > > > > > > #include <asm/setup.h> /* for COMMAND_LINE_SIZE */ > > > #include <asm/page.h> > > > @@ -1323,3 +1324,58 @@ late_initcall(of_fdt_raw_init); > > > #endif > > > > > > #endif /* CONFIG_OF_EARLY_FLATTREE */ > > > + > > > +#define FDT_ALIGN(x, a) (((x) + (a) - 1) & ~((a) - 1)) > > > +#define FDT_TAGALIGN(x) (FDT_ALIGN((x), FDT_TAGSIZE)) > > > + > > > +int fdt_prop_len(const char *prop_name, int len) > > > +{ > > > + return (strlen(prop_name) + 1) + > > > + sizeof(struct fdt_property) + > > > + FDT_TAGALIGN(len); > > > > Looks like you are using this to calculate how much space you need to > > allocate in addition to the current DTB for a couple of new or > > replaced properties. I'm not sure that this calculation is completely > > accurate. And it is strange there doesn't seem to be any libfdt > > function for this already. It would be simpler to just add some fixed > > additional amount. > > > > Maybe David G has comments on this? > > > > > +} > > > + > > > > The rest of this should go in drivers/of/fdt_address.c. Ultimately, it > > should go into libfdt, but I'm fine with having it in the kernel for > > now. > > I'd like to have this function in the kernel for now. > > > > +static void fill_property(void *buf, u64 val64, int cells) > > > +{ > > > + __be32 val32; > > > + > > > + while (cells) { > > > + val32 = cpu_to_fdt32((val64 >> (32 * (--cells))) & U32_MAX); > > > + memcpy(buf, &val32, sizeof(val32)); > > > + buf += sizeof(val32); > > > > This is kind of hard to read. I would copy u-boot's fdt_pack_reg function. > > Are you sure? > I originally implemented this function in a similar way that fdt_pack_reg() > was, but, you suggested, in your past comment[1], that we'd be better to > have of_read_number()-like implementation. > > [1] http://lists.infradead.org/pipermail/linux-arm-kernel/2018-May/579118.html Yeah, you're right. Plus, I'm not sure the u-boot one would work for unaligned accesses with armv5 and earlier h/w. My only comment then is I think you can drop the U32_MAX masking. Rob
On Fri, Oct 05, 2018 at 08:23:57AM -0500, Rob Herring wrote: > On Thu, Oct 4, 2018 at 10:07 PM AKASHI, Takahiro > <takahiro.akashi@linaro.org> wrote: > > > > Rob, > > > > # I haven't replied to this comment yet. > > > > On Fri, Sep 28, 2018 at 08:44:42AM -0500, Rob Herring wrote: > > > +David Gibson > > > > > > On Fri, Sep 28, 2018 at 1:48 AM AKASHI Takahiro > > > <takahiro.akashi@linaro.org> wrote: > > > > > > > > These functions will be used later to handle kexec-specific properties > > > > in arm64's kexec_file implementation. > > > > > > > > Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org> > > > > Cc: Rob Herring <robh+dt@kernel.org> > > > > Cc: Frank Rowand <frowand.list@gmail.com> > > > > Cc: devicetree@vger.kernel.org > > > > --- > > > > drivers/of/fdt.c | 56 ++++++++++++++++++++++++++++++++++++++++++ > > > > include/linux/of_fdt.h | 4 +++ > > > > 2 files changed, 60 insertions(+) > > > > > > > > diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c > > > > index 800ad252cf9c..c65c31562ccb 100644 > > > > --- a/drivers/of/fdt.c > > > > +++ b/drivers/of/fdt.c > > > > @@ -25,6 +25,7 @@ > > > > #include <linux/debugfs.h> > > > > #include <linux/serial_core.h> > > > > #include <linux/sysfs.h> > > > > +#include <linux/types.h> > > > > > > > > #include <asm/setup.h> /* for COMMAND_LINE_SIZE */ > > > > #include <asm/page.h> > > > > @@ -1323,3 +1324,58 @@ late_initcall(of_fdt_raw_init); > > > > #endif > > > > > > > > #endif /* CONFIG_OF_EARLY_FLATTREE */ > > > > + > > > > +#define FDT_ALIGN(x, a) (((x) + (a) - 1) & ~((a) - 1)) > > > > +#define FDT_TAGALIGN(x) (FDT_ALIGN((x), FDT_TAGSIZE)) > > > > + > > > > +int fdt_prop_len(const char *prop_name, int len) > > > > +{ > > > > + return (strlen(prop_name) + 1) + > > > > + sizeof(struct fdt_property) + > > > > + FDT_TAGALIGN(len); > > > > > > Looks like you are using this to calculate how much space you need to > > > allocate in addition to the current DTB for a couple of new or > > > replaced properties. I'm not sure that this calculation is completely > > > accurate. And it is strange there doesn't seem to be any libfdt > > > function for this already. It would be simpler to just add some fixed > > > additional amount. > > > > > > Maybe David G has comments on this? > > > > > > > +} > > > > + > > > > > > The rest of this should go in drivers/of/fdt_address.c. Ultimately, it > > > should go into libfdt, but I'm fine with having it in the kernel for > > > now. > > > > I'd like to have this function in the kernel for now. > > > > > > +static void fill_property(void *buf, u64 val64, int cells) > > > > +{ > > > > + __be32 val32; > > > > + > > > > + while (cells) { > > > > + val32 = cpu_to_fdt32((val64 >> (32 * (--cells))) & U32_MAX); > > > > + memcpy(buf, &val32, sizeof(val32)); > > > > + buf += sizeof(val32); > > > > > > This is kind of hard to read. I would copy u-boot's fdt_pack_reg function. > > > > Are you sure? > > I originally implemented this function in a similar way that fdt_pack_reg() > > was, but, you suggested, in your past comment[1], that we'd be better to > > have of_read_number()-like implementation. > > > > [1] http://lists.infradead.org/pipermail/linux-arm-kernel/2018-May/579118.html > > Yeah, you're right. Plus, I'm not sure the u-boot one would work for > unaligned accesses with armv5 and earlier h/w. > > My only comment then is I think you can drop the U32_MAX masking. Okay, then I will leave this function, yet renaming it to cpu64_to_fdt_cells() after Frank's comment. Thanks, -Takahiro Akashi > > Rob
On Tue, Oct 09, 2018 at 11:02:29AM -0700, Frank Rowand wrote: > On 10/04/18 22:06, AKASHI, Takahiro wrote: [snip] > >>>> +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, > >>>> + u64 addr, u64 size) > >>>> +{ > >>>> + int addr_cells, size_cells; > >> > >> unsigned > > > > fdt_[address|size]_cell() returns an int. > > I stand corrected. They take advantage of the fact that valid values > are in the range 0..4 and return a negative value for error. I'm not sure if the spec formally defines that these values will fit in a positive 'int'. But if addresses are 8GiB each, you're going to have much bigger problems putting it in a device tree than this function. -- 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
Frank, On Tue, Oct 09, 2018 at 11:02:29AM -0700, Frank Rowand wrote: > On 10/04/18 22:06, AKASHI, Takahiro wrote: > > Frank, > > > > # I haven't reply to your comments. > > > > On Fri, Sep 28, 2018 at 02:13:58PM -0700, Frank Rowand wrote: > >> On 09/28/18 06:44, Rob Herring wrote: > >>> +David Gibson > >>> > >>> On Fri, Sep 28, 2018 at 1:48 AM AKASHI Takahiro > >>> <takahiro.akashi@linaro.org> wrote: > >>>> > >>>> These functions will be used later to handle kexec-specific properties > >>>> in arm64's kexec_file implementation. > >> > >> As I requested in version 14: > >> > >> The intent of the helper functions is related to properties whose values are > >> tuples of the same format as the "reg" property of the "/memory" nodes. For > >> example, the "linux,usable-memory-range" and "linux,elfcoredhr" properties of > >> the "/chosen" node. > >> > >> The patch header and the function names should be updated to reflect this intent. > > > > I agree regarding the patch header. > > > >> > >>>> > >>>> Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org> > >>>> Cc: Rob Herring <robh+dt@kernel.org> > >>>> Cc: Frank Rowand <frowand.list@gmail.com> > >>>> Cc: devicetree@vger.kernel.org > >>>> --- > >> > >> Missing list of changes since version 14. > > > > Sorry for the inconvenience, but a whole change list goes into > > the cover letter, not individual patches. > > That works for me, but please add us to the distro list for the cover letter. Sure, I will try to do so. > > >>>> drivers/of/fdt.c | 56 ++++++++++++++++++++++++++++++++++++++++++ > >>>> include/linux/of_fdt.h | 4 +++ > >>>> 2 files changed, 60 insertions(+) > >>>> > >>>> diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c > >>>> index 800ad252cf9c..c65c31562ccb 100644 > >>>> --- a/drivers/of/fdt.c > >>>> +++ b/drivers/of/fdt.c > >>>> @@ -25,6 +25,7 @@ > >>>> #include <linux/debugfs.h> > >>>> #include <linux/serial_core.h> > >>>> #include <linux/sysfs.h> > >>>> +#include <linux/types.h> > >>>> > >>>> #include <asm/setup.h> /* for COMMAND_LINE_SIZE */ > >>>> #include <asm/page.h> > >>>> @@ -1323,3 +1324,58 @@ late_initcall(of_fdt_raw_init); > >>>> #endif > >>>> > >>>> #endif /* CONFIG_OF_EARLY_FLATTREE */ > >> > >> In v14 I requested: > >> > >> Please add comment: > >> > >> /* helper functions for arm64 kexec */ > > > > Okay. > > > >> > >>>> + > >>>> +#define FDT_ALIGN(x, a) (((x) + (a) - 1) & ~((a) - 1)) > >>>> +#define FDT_TAGALIGN(x) (FDT_ALIGN((x), FDT_TAGSIZE)) > >>>> + > >>>> +int fdt_prop_len(const char *prop_name, int len) > >> > >> In v14, I requested: > >> > >> Please rename as fdt_len_added_prop() > > > > Anyhow, I will drop this function, preferring to new > > fdt_[address|size]_cells(). > > > >> I'm not really happy with my suggested name, but do not have a > >> better one yet. As Rob notes, maybe David G will have a helpful > >> comment. > >> > >>>> +{ > >>>> + return (strlen(prop_name) + 1) + > >>>> + sizeof(struct fdt_property) + > >>>> + FDT_TAGALIGN(len); > >>> > >>> Looks like you are using this to calculate how much space you need to > >>> allocate in addition to the current DTB for a couple of new or > >>> replaced properties. I'm not sure that this calculation is completely > >>> accurate. And it is strange there doesn't seem to be any libfdt > >>> function for this already. It would be simpler to just add some fixed > >>> additional amount. > >>> > >>> Maybe David G has comments on this? > > > > I'm not quit sure why it's not that accurate, but as I said in a reply to > > David's comment, I will take your suggestion. > > > >>>> +} > >>>> + > >>> > >>> The rest of this should go in drivers/of/fdt_address.c. Ultimately, it > >>> should go into libfdt, but I'm fine with having it in the kernel for > >>> now. > >>> > >>>> +static void fill_property(void *buf, u64 val64, int cells) > >> > >> In v14 I requested: > >> > >> Please rename as cpu64_to_fdt_cells() > > > > I don't mind, but this function may be dropped if Rob sticks to > > u-boot's fdt_pack_reg() over my fdt_setprop_reg(). > > I have another comment a couple of minutes ago in the longer thread > about this function. Okay. -Takahiro Akashi > > >> > >>>> +{ > >>>> + __be32 val32; > >>>> + > >>>> + while (cells) { > >>>> + val32 = cpu_to_fdt32((val64 >> (32 * (--cells))) & U32_MAX); > >>>> + memcpy(buf, &val32, sizeof(val32)); > >>>> + buf += sizeof(val32); > >>> > >>> This is kind of hard to read. I would copy u-boot's fdt_pack_reg function. > >>> > >>> BTW, for purposes of moving to libfdt, we'll need the authors' > >>> (Masahiro Yamada and Hans de Goede) permission to dual license. > >>> > >>>> + } > >>>> +} > >>>> + > >>>> +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, > >>>> + u64 addr, u64 size) > >>>> +{ > >>>> + int addr_cells, size_cells; > >> > >> unsigned > > > > fdt_[address|size]_cell() returns an int. > > I stand corrected. They take advantage of the fact that valid values > are in the range 0..4 and return a negative value for error. > > > >>>> + char buf[sizeof(__be32) * 2 * 2]; > >>>> + /* assume dt_root_[addr|size]_cells <= 2 */ > >>>> + void *prop; > >>>> + size_t buf_size; > >>>> + > >>>> + addr_cells = fdt_address_cells(fdt, 0); > >>>> + if (addr_cells < 0) > >>>> + return addr_cells; > >>>> + size_cells = fdt_size_cells(fdt, 0); > >>>> + if (size_cells < 0) > >>>> + return size_cells; > >>>> + > >>>> + /* if *_cells >= 2, cells can hold 64-bit values anyway */ > >>>> + if ((addr_cells == 1) && (addr > U32_MAX)) > >>>> + return -FDT_ERR_BADVALUE; > >>>> + > >>>> + if ((size_cells == 1) && (size > U32_MAX)) > >>>> + return -FDT_ERR_BADVALUE; > >> > >> In v14 I requested: > >> > >> Should also check that base + size does not wrap around. > > > > Okay, I will start discussion, as you have suggested, in devicetree-spec ML. > > > > Thanks, > > -Takahiro Akashi > > > >> > >>>> + > >>>> + buf_size = (addr_cells + size_cells) * sizeof(u32); > >>>> + prop = buf; > >>>> + > >>>> + fill_property(prop, addr, addr_cells); > >>>> + prop += addr_cells * sizeof(u32); > >>>> + > >>>> + fill_property(prop, size, size_cells); > >>>> + > >>>> + return fdt_setprop(fdt, nodeoffset, name, buf, buf_size); > >>>> +} > >>>> diff --git a/include/linux/of_fdt.h b/include/linux/of_fdt.h > >>>> index b9cd9ebdf9b9..842af6ea92ea 100644 > >>>> --- a/include/linux/of_fdt.h > >>>> +++ b/include/linux/of_fdt.h > >>>> @@ -108,5 +108,9 @@ static inline void unflatten_device_tree(void) {} > >>>> static inline void unflatten_and_copy_device_tree(void) {} > >>>> #endif /* CONFIG_OF_EARLY_FLATTREE */ > >>>> > >>>> +int fdt_prop_len(const char *prop_name, int len); > >>>> +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, > >>>> + u64 addr, u64 size); > >>>> + > >>>> #endif /* __ASSEMBLY__ */ > >>>> #endif /* _LINUX_OF_FDT_H */ > >>>> -- > >>>> 2.19.0 > >>>> > >>> > >> > > >
Frank, Rob, On Tue, Oct 09, 2018 at 10:47:15AM -0700, Frank Rowand wrote: > On 10/08/18 17:37, AKASHI, Takahiro wrote: > > On Fri, Oct 05, 2018 at 08:23:57AM -0500, Rob Herring wrote: > >> On Thu, Oct 4, 2018 at 10:07 PM AKASHI, Takahiro > >> <takahiro.akashi@linaro.org> wrote: > >>> > >>> Rob, > >>> > >>> # I haven't replied to this comment yet. > >>> > >>> On Fri, Sep 28, 2018 at 08:44:42AM -0500, Rob Herring wrote: > >>>> +David Gibson > >>>> > >>>> On Fri, Sep 28, 2018 at 1:48 AM AKASHI Takahiro > >>>> <takahiro.akashi@linaro.org> wrote: > >>>>> > >>>>> These functions will be used later to handle kexec-specific properties > >>>>> in arm64's kexec_file implementation. > >>>>> > >>>>> Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org> > >>>>> Cc: Rob Herring <robh+dt@kernel.org> > >>>>> Cc: Frank Rowand <frowand.list@gmail.com> > >>>>> Cc: devicetree@vger.kernel.org > >>>>> --- > >>>>> drivers/of/fdt.c | 56 ++++++++++++++++++++++++++++++++++++++++++ > >>>>> include/linux/of_fdt.h | 4 +++ > >>>>> 2 files changed, 60 insertions(+) > >>>>> > >>>>> diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c > >>>>> index 800ad252cf9c..c65c31562ccb 100644 > >>>>> --- a/drivers/of/fdt.c > >>>>> +++ b/drivers/of/fdt.c > >>>>> @@ -25,6 +25,7 @@ > >>>>> #include <linux/debugfs.h> > >>>>> #include <linux/serial_core.h> > >>>>> #include <linux/sysfs.h> > >>>>> +#include <linux/types.h> > >>>>> > >>>>> #include <asm/setup.h> /* for COMMAND_LINE_SIZE */ > >>>>> #include <asm/page.h> > >>>>> @@ -1323,3 +1324,58 @@ late_initcall(of_fdt_raw_init); > >>>>> #endif > >>>>> > >>>>> #endif /* CONFIG_OF_EARLY_FLATTREE */ > >>>>> + > >>>>> +#define FDT_ALIGN(x, a) (((x) + (a) - 1) & ~((a) - 1)) > >>>>> +#define FDT_TAGALIGN(x) (FDT_ALIGN((x), FDT_TAGSIZE)) > >>>>> + > >>>>> +int fdt_prop_len(const char *prop_name, int len) > >>>>> +{ > >>>>> + return (strlen(prop_name) + 1) + > >>>>> + sizeof(struct fdt_property) + > >>>>> + FDT_TAGALIGN(len); > >>>> > >>>> Looks like you are using this to calculate how much space you need to > >>>> allocate in addition to the current DTB for a couple of new or > >>>> replaced properties. I'm not sure that this calculation is completely > >>>> accurate. And it is strange there doesn't seem to be any libfdt > >>>> function for this already. It would be simpler to just add some fixed > >>>> additional amount. > >>>> > >>>> Maybe David G has comments on this? > >>>> > >>>>> +} > >>>>> + > >>>> > >>>> The rest of this should go in drivers/of/fdt_address.c. Ultimately, it > >>>> should go into libfdt, but I'm fine with having it in the kernel for > >>>> now. > >>> > >>> I'd like to have this function in the kernel for now. > >>> > >>>>> +static void fill_property(void *buf, u64 val64, int cells) > >>>>> +{ > >>>>> + __be32 val32; > >>>>> + > >>>>> + while (cells) { > >>>>> + val32 = cpu_to_fdt32((val64 >> (32 * (--cells))) & U32_MAX); > >>>>> + memcpy(buf, &val32, sizeof(val32)); > >>>>> + buf += sizeof(val32); > >>>> > >>>> This is kind of hard to read. I would copy u-boot's fdt_pack_reg function. > >>> > >>> Are you sure? > >>> I originally implemented this function in a similar way that fdt_pack_reg() > >>> was, but, you suggested, in your past comment[1], that we'd be better to > >>> have of_read_number()-like implementation. > >>> > >>> [1] http://lists.infradead.org/pipermail/linux-arm-kernel/2018-May/579118.html > >> > >> Yeah, you're right. Plus, I'm not sure the u-boot one would work for > >> unaligned accesses with armv5 and earlier h/w. > >> > >> My only comment then is I think you can drop the U32_MAX masking. > > > > Okay, then I will leave this function, yet renaming it to > > cpu64_to_fdt_cells() after Frank's comment. > > I have second guessed myself and do not like the name I suggested > because what the function really does is either cpu32 to be32 or > cpu64 to be64. Okay. > I agree with Rob that readability is important here. Instead of > having a fill_property() function, how about having inline code, > something like (untested even for thinkos): > > prop = buf; > > if (addr_cells == 1) { > *(__be32 *)prop = cpu32_to_be32(addr); > prop += 4; > } else { > *(__be64 *)prop = cpu64_to_be64(addr); > prop += 8; > } > > if (size_cells == 1) > *(__be32 *)prop = cpu32_to_be32(size); > else > *(__be64 *)prop = cpu64_to_be64(size); > > You might want to also give Rob a chance to bike shed on this > suggestion. Basically, I don't care either way, but what Rob suggested is that some architecture(s) might not handle correctly unaligned memory access here. I just want to stay tuned with Rob. Thanks, -Takahiro Akashi > -Frank > > > > > Thanks, > > -Takahiro Akashi > >> > >> Rob > > >
On 10/09/18 18:14, AKASHI, Takahiro wrote: > Frank, Rob, > > On Tue, Oct 09, 2018 at 10:47:15AM -0700, Frank Rowand wrote: >> On 10/08/18 17:37, AKASHI, Takahiro wrote: >>> On Fri, Oct 05, 2018 at 08:23:57AM -0500, Rob Herring wrote: >>>> On Thu, Oct 4, 2018 at 10:07 PM AKASHI, Takahiro >>>> <takahiro.akashi@linaro.org> wrote: >>>>> >>>>> Rob, >>>>> >>>>> # I haven't replied to this comment yet. >>>>> >>>>> On Fri, Sep 28, 2018 at 08:44:42AM -0500, Rob Herring wrote: >>>>>> +David Gibson >>>>>> >>>>>> On Fri, Sep 28, 2018 at 1:48 AM AKASHI Takahiro >>>>>> <takahiro.akashi@linaro.org> wrote: >>>>>>> >>>>>>> These functions will be used later to handle kexec-specific properties >>>>>>> in arm64's kexec_file implementation. >>>>>>> >>>>>>> Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org> >>>>>>> Cc: Rob Herring <robh+dt@kernel.org> >>>>>>> Cc: Frank Rowand <frowand.list@gmail.com> >>>>>>> Cc: devicetree@vger.kernel.org >>>>>>> --- >>>>>>> drivers/of/fdt.c | 56 ++++++++++++++++++++++++++++++++++++++++++ >>>>>>> include/linux/of_fdt.h | 4 +++ >>>>>>> 2 files changed, 60 insertions(+) >>>>>>> >>>>>>> diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c >>>>>>> index 800ad252cf9c..c65c31562ccb 100644 >>>>>>> --- a/drivers/of/fdt.c >>>>>>> +++ b/drivers/of/fdt.c >>>>>>> @@ -25,6 +25,7 @@ >>>>>>> #include <linux/debugfs.h> >>>>>>> #include <linux/serial_core.h> >>>>>>> #include <linux/sysfs.h> >>>>>>> +#include <linux/types.h> >>>>>>> >>>>>>> #include <asm/setup.h> /* for COMMAND_LINE_SIZE */ >>>>>>> #include <asm/page.h> >>>>>>> @@ -1323,3 +1324,58 @@ late_initcall(of_fdt_raw_init); >>>>>>> #endif >>>>>>> >>>>>>> #endif /* CONFIG_OF_EARLY_FLATTREE */ >>>>>>> + >>>>>>> +#define FDT_ALIGN(x, a) (((x) + (a) - 1) & ~((a) - 1)) >>>>>>> +#define FDT_TAGALIGN(x) (FDT_ALIGN((x), FDT_TAGSIZE)) >>>>>>> + >>>>>>> +int fdt_prop_len(const char *prop_name, int len) >>>>>>> +{ >>>>>>> + return (strlen(prop_name) + 1) + >>>>>>> + sizeof(struct fdt_property) + >>>>>>> + FDT_TAGALIGN(len); >>>>>> >>>>>> Looks like you are using this to calculate how much space you need to >>>>>> allocate in addition to the current DTB for a couple of new or >>>>>> replaced properties. I'm not sure that this calculation is completely >>>>>> accurate. And it is strange there doesn't seem to be any libfdt >>>>>> function for this already. It would be simpler to just add some fixed >>>>>> additional amount. >>>>>> >>>>>> Maybe David G has comments on this? >>>>>> >>>>>>> +} >>>>>>> + >>>>>> >>>>>> The rest of this should go in drivers/of/fdt_address.c. Ultimately, it >>>>>> should go into libfdt, but I'm fine with having it in the kernel for >>>>>> now. >>>>> >>>>> I'd like to have this function in the kernel for now. >>>>> >>>>>>> +static void fill_property(void *buf, u64 val64, int cells) >>>>>>> +{ >>>>>>> + __be32 val32; >>>>>>> + >>>>>>> + while (cells) { >>>>>>> + val32 = cpu_to_fdt32((val64 >> (32 * (--cells))) & U32_MAX); >>>>>>> + memcpy(buf, &val32, sizeof(val32)); >>>>>>> + buf += sizeof(val32); >>>>>> >>>>>> This is kind of hard to read. I would copy u-boot's fdt_pack_reg function. >>>>> >>>>> Are you sure? >>>>> I originally implemented this function in a similar way that fdt_pack_reg() >>>>> was, but, you suggested, in your past comment[1], that we'd be better to >>>>> have of_read_number()-like implementation. >>>>> >>>>> [1] http://lists.infradead.org/pipermail/linux-arm-kernel/2018-May/579118.html >>>> >>>> Yeah, you're right. Plus, I'm not sure the u-boot one would work for >>>> unaligned accesses with armv5 and earlier h/w. >>>> >>>> My only comment then is I think you can drop the U32_MAX masking. >>> >>> Okay, then I will leave this function, yet renaming it to >>> cpu64_to_fdt_cells() after Frank's comment. >> >> I have second guessed myself and do not like the name I suggested >> because what the function really does is either cpu32 to be32 or >> cpu64 to be64. > > Okay. > >> I agree with Rob that readability is important here. Instead of >> having a fill_property() function, how about having inline code, >> something like (untested even for thinkos): >> >> prop = buf; >> >> if (addr_cells == 1) { >> *(__be32 *)prop = cpu32_to_be32(addr); >> prop += 4; >> } else { >> *(__be64 *)prop = cpu64_to_be64(addr); >> prop += 8; >> } >> >> if (size_cells == 1) >> *(__be32 *)prop = cpu32_to_be32(size); >> else >> *(__be64 *)prop = cpu64_to_be64(size); >> >> You might want to also give Rob a chance to bike shed on this >> suggestion. > > Basically, I don't care either way, but what Rob suggested > is that some architecture(s) might not handle correctly > unaligned memory access here. Good point about unaligned memory access. That rules out my suggestion above. But if using the fill_property() implementation, either come up with a name that better describes what the function does or add a function header comment that gives the reader a clue as to what the function accomplishes. -Frank > I just want to stay tuned with Rob. > > Thanks, > -Takahiro Akashi > >> -Frank >> >>> >>> Thanks, >>> -Takahiro Akashi >>>> >>>> Rob >>> >> >
diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c index 800ad252cf9c..c65c31562ccb 100644 --- a/drivers/of/fdt.c +++ b/drivers/of/fdt.c @@ -25,6 +25,7 @@ #include <linux/debugfs.h> #include <linux/serial_core.h> #include <linux/sysfs.h> +#include <linux/types.h> #include <asm/setup.h> /* for COMMAND_LINE_SIZE */ #include <asm/page.h> @@ -1323,3 +1324,58 @@ late_initcall(of_fdt_raw_init); #endif #endif /* CONFIG_OF_EARLY_FLATTREE */ + +#define FDT_ALIGN(x, a) (((x) + (a) - 1) & ~((a) - 1)) +#define FDT_TAGALIGN(x) (FDT_ALIGN((x), FDT_TAGSIZE)) + +int fdt_prop_len(const char *prop_name, int len) +{ + return (strlen(prop_name) + 1) + + sizeof(struct fdt_property) + + FDT_TAGALIGN(len); +} + +static void fill_property(void *buf, u64 val64, int cells) +{ + __be32 val32; + + while (cells) { + val32 = cpu_to_fdt32((val64 >> (32 * (--cells))) & U32_MAX); + memcpy(buf, &val32, sizeof(val32)); + buf += sizeof(val32); + } +} + +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, + u64 addr, u64 size) +{ + int addr_cells, size_cells; + char buf[sizeof(__be32) * 2 * 2]; + /* assume dt_root_[addr|size]_cells <= 2 */ + void *prop; + size_t buf_size; + + addr_cells = fdt_address_cells(fdt, 0); + if (addr_cells < 0) + return addr_cells; + size_cells = fdt_size_cells(fdt, 0); + if (size_cells < 0) + return size_cells; + + /* if *_cells >= 2, cells can hold 64-bit values anyway */ + if ((addr_cells == 1) && (addr > U32_MAX)) + return -FDT_ERR_BADVALUE; + + if ((size_cells == 1) && (size > U32_MAX)) + return -FDT_ERR_BADVALUE; + + buf_size = (addr_cells + size_cells) * sizeof(u32); + prop = buf; + + fill_property(prop, addr, addr_cells); + prop += addr_cells * sizeof(u32); + + fill_property(prop, size, size_cells); + + return fdt_setprop(fdt, nodeoffset, name, buf, buf_size); +} diff --git a/include/linux/of_fdt.h b/include/linux/of_fdt.h index b9cd9ebdf9b9..842af6ea92ea 100644 --- a/include/linux/of_fdt.h +++ b/include/linux/of_fdt.h @@ -108,5 +108,9 @@ static inline void unflatten_device_tree(void) {} static inline void unflatten_and_copy_device_tree(void) {} #endif /* CONFIG_OF_EARLY_FLATTREE */ +int fdt_prop_len(const char *prop_name, int len); +int fdt_setprop_reg(void *fdt, int nodeoffset, const char *name, + u64 addr, u64 size); + #endif /* __ASSEMBLY__ */ #endif /* _LINUX_OF_FDT_H */
These functions will be used later to handle kexec-specific properties in arm64's kexec_file implementation. Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org> Cc: Rob Herring <robh+dt@kernel.org> Cc: Frank Rowand <frowand.list@gmail.com> Cc: devicetree@vger.kernel.org --- drivers/of/fdt.c | 56 ++++++++++++++++++++++++++++++++++++++++++ include/linux/of_fdt.h | 4 +++ 2 files changed, 60 insertions(+) -- 2.19.0