| Message ID | 1346318957-5831-2-git-send-email-trenn@suse.de (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On 08/30/2012 02:29 AM, Thomas Renninger wrote: > From: "H. Peter Anvin" <hpa@linux.intel.com> > > Add a simple cpio decoder without library dependencies for the purpose > of extracting components from the initramfs blob for early kernel > uses. Intended consumers so far are microcode and ACPI override. > > Signed-off-by: H. Peter Anvin <hpa@linux.intel.com> > CC: Thomas Renninger <trenn@suse.de> > Link: http://lkml.kernel.org/r/201203261651.29640.trenn@suse.de > Signed-off-by: Thomas Renninger <trenn@suse.de> I was trying to figure out if there is a way to do what you want (support for multiple files) without the problems of the callback interface. I think it is actually fairly straightforward; we need a prefix iterator (so you can give it a string like "kernel/acpi/" rather than a full filename) and it needs to be able to accept a "last" pointer so it can resume scanning at the point it last left off. That should be a pretty trivial change. The other thing we presumably want to do -- and this is generic -- is to be able to handle multiple sources for the initramfs; at the very least there is built in vs provided from the boot loader. I had originally intended to just handle that by calling the earlycpio function once per block, but the "last left off" bit makes that a little harder. Need to think about that a little bit. I am guessing that this may not need to be something we need from the very beginning, or am I wrong? -hpa
On Tuesday, September 04, 2012 07:00:09 PM H. Peter Anvin wrote: > On 08/30/2012 02:29 AM, Thomas Renninger wrote: > > From: "H. Peter Anvin" <hpa@linux.intel.com> > > > > Add a simple cpio decoder without library dependencies for the purpose > > of extracting components from the initramfs blob for early kernel > > uses. Intended consumers so far are microcode and ACPI override. > > > > Signed-off-by: H. Peter Anvin <hpa@linux.intel.com> > > CC: Thomas Renninger <trenn@suse.de> > > Link: http://lkml.kernel.org/r/201203261651.29640.trenn@suse.de > > Signed-off-by: Thomas Renninger <trenn@suse.de> > > I was trying to figure out if there is a way to do what you want > (support for multiple files) without the problems of the callback > interface. I think it is actually fairly straightforward; we need a > prefix iterator (so you can give it a string like "kernel/acpi/" rather > than a full filename) and it needs to be able to accept a "last" pointer > so it can resume scanning at the point it last left off. That should be > a pretty trivial change. Yep, this is a good idea. > The other thing we presumably want to do -- and this is generic -- is to > be able to handle multiple sources for the initramfs; at the very least > there is built in vs provided from the boot loader. I had originally > intended to just handle that by calling the earlycpio function once per > block, but the "last left off" bit makes that a little harder. Need to > think about that a little bit. I guess I understand the first part, not sure about the "last left off" bit. "Multiple sources" means bootloader already points to multiple sources, right? This is somewhat out of scope as this would need both, bootloader and kernel adjustings. This is about "built in" which means multiple cpios concatenated together and passed via bootloader as one "file" (initrd). At least I understand it that way. The only disadvantage I run into is that once the cpios are concatenated, I couldn't figure out an easy way how to slice them into separate cpio/zip archives again. > I am guessing that this may not need to be something we need from the > very beginning, or am I wrong? I have re-worked the patches: - added an offset argument to be able to iterate over files inside a directory as suggested - removed the strlen and memcmp function duplication in earlycpio.c This is not needed for this patchset and would be confusing. This can easily be added in the context of the patches which need it I'll re-submit. Thanks, Thomas -- To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 09/21/2012 05:51 AM, Thomas Renninger wrote: > I guess I understand the first part, not sure about the "last left off" > bit. > "Multiple sources" means bootloader already points to multiple sources, > right? > This is somewhat out of scope as this would need both, bootloader and > kernel adjustings. > This is about "built in" which means multiple cpios concatenated together > and passed via bootloader as one "file" (initrd). > At least I understand it that way. > The only disadvantage I run into is that once the cpios are concatenated, > I couldn't figure out an easy way how to slice them into separate > cpio/zip archives again. No, that's not the issue. The issue is that there isn't just one initramfs blob, there are (currently) *TWO*... one loaded by the bootloader and one compiled into the kernel. -hpa -- To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/include/linux/earlycpio.h b/include/linux/earlycpio.h new file mode 100644 index 0000000..06db026 --- /dev/null +++ b/include/linux/earlycpio.h @@ -0,0 +1,13 @@ +#ifndef _LINUX_EARLYCPIO_H +#define _LINUX_EARLYCPIO_H + +#include <linux/types.h> + +struct cpio_data { + void *data; + size_t size; +}; + +struct cpio_data find_cpio_data(const char *name, const void *data, size_t len); + +#endif /* _LINUX_EARLYCPIO_H */ diff --git a/lib/Makefile b/lib/Makefile index 42d283e..0924041 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -12,7 +12,7 @@ lib-y := ctype.o string.o vsprintf.o cmdline.o \ idr.o int_sqrt.o extable.o prio_tree.o \ sha1.o md5.o irq_regs.o reciprocal_div.o argv_split.o \ proportions.o flex_proportions.o prio_heap.o ratelimit.o show_mem.o \ - is_single_threaded.o plist.o decompress.o + is_single_threaded.o plist.o decompress.o earlycpio.o lib-$(CONFIG_MMU) += ioremap.o lib-$(CONFIG_SMP) += cpumask.o diff --git a/lib/earlycpio.c b/lib/earlycpio.c new file mode 100644 index 0000000..b16b80b --- /dev/null +++ b/lib/earlycpio.c @@ -0,0 +1,173 @@ +/* ----------------------------------------------------------------------- * + * + * Copyright 2012 Intel Corporation; author H. Peter Anvin + * + * This file is part of the Linux kernel, and is made available + * under the terms of the GNU General Public License version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * ----------------------------------------------------------------------- */ + +/* + * earlycpio.c + * + * Find a specific cpio member; must precede any compressed content. + * This is used to locate data items in the initramfs used by the + * kernel itself during early boot (before the main initramfs is + * decompressed.) It is the responsibility of the initramfs creator + * to ensure that these items are uncompressed at the head of the + * blob. Depending on the boot loader or package tool that may be a + * separate file or part of the same file. + * + * For some architectures, e.g. i386, this file must compile to have + * no relocations and no library dependencies, so it can be called from + * a nonstandard environment. Therefore some normal library functions + * are inlined in this file. + */ + +#include <linux/earlycpio.h> +#include <linux/kernel.h> + +enum cpio_fields { + C_MAGIC, + C_INO, + C_MODE, + C_UID, + C_GID, + C_NLINK, + C_MTIME, + C_FILESIZE, + C_MAJ, + C_MIN, + C_RMAJ, + C_RMIN, + C_NAMESIZE, + C_CHKSUM, + C_NFIELDS +}; + +#ifdef CONFIG_X86 +static inline size_t strlen(const char *name) +{ + size_t n = -1; + + asm("repne; scasb" + : "+D" (name), "+c" (n) + : "a" (0)); + + return -2 - n; +} + +static inline int memcmp(const void *p1, const void *p2, size_t n) +{ + unsigned char rv; + + asm("repe; cmpsb; setne %0" + : "=r" (rv), "+S" (p1), "+D" (p2), "+c" (n)); + + return rv; +} +#else +static inline size_t strlen(const char *name) +{ + size_t n = 0; + + while (*name++) + n++; + + return n; +} + +static inline int memcmp(const void *p1, const void *p2, size_t n) +{ + const unsigned char *u1 = p1; + const unsigned char *u2 = p2; + int d; + + while (n--) { + d = *u2++ - *u1++; + if (d) + return d; + } + return 0; +} +#endif + +struct cpio_data __cpuinit find_cpio_data(const char *name, + const void *data, size_t len) +{ + const size_t cpio_header_len = 8*C_NFIELDS - 2; + struct cpio_data cd = { NULL, 0 }; + const char *p, *dptr, *nptr; + unsigned int ch[C_NFIELDS], *chp, v; + unsigned char c, x; + size_t mynamesize = strlen(name) + 1; + int i, j; + + p = data; + + while (len > cpio_header_len) { + if (!*p) { + /* All cpio headers need to be 4-byte aligned */ + p += 4; + len -= 4; + continue; + } + + j = 6; /* The magic field is only 6 characters */ + chp = ch; + for (i = C_NFIELDS; i; i--) { + v = 0; + while (j--) { + v <<= 4; + c = *p++; + + x = c - '0'; + if (x < 10) { + v += x; + continue; + } + + x = (c | 0x20) - 'a'; + if (x < 6) { + v += x + 10; + continue; + } + + goto quit; /* Invalid hexadecimal */ + } + *chp++ = v; + j = 8; /* All other fields are 8 characters */ + } + + if ((ch[C_MAGIC] - 0x070701) > 1) + goto quit; /* Invalid magic */ + + len -= cpio_header_len; + + dptr = PTR_ALIGN(p + ch[C_NAMESIZE], 4); + nptr = PTR_ALIGN(dptr + ch[C_FILESIZE], 4); + + if (nptr > p + len || dptr < p || nptr < dptr) + goto quit; /* Buffer overrun */ + + if ((ch[C_MODE] & 0170000) == 0100000 && + ch[C_NAMESIZE] == mynamesize && + !memcmp(p, name, mynamesize)) { + cd.data = (void *)dptr; + cd.size = ch[C_FILESIZE]; + return cd; /* Found it! */ + } + + len -= (nptr - p); + p = nptr; + } + +quit: + return cd; +}