From de933bd833be1a53bd361c9a327afd3c65413351 Mon Sep 17 00:00:00 2001 From: Uwe Kleine-König Date: Fri, 18 Feb 2011 21:47:20 +0100 Subject: kbuild: reenable section mismatch analysis MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This was disabled in commit e5f95c8 (kbuild: print only total number of section mismatces found) because there were too many warnings. Now we're down to a reasonable number again, so we start scaring people with the details. Signed-off-by: Uwe Kleine-König Acked-by: Randy Dunlap Acked-by: Mike Frysinger Signed-off-by: Michal Marek --- lib/Kconfig.debug | 4 ---- 1 file changed, 4 deletions(-) (limited to 'lib') diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 2d05adb9840..8b6a4f13946 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -102,11 +102,7 @@ config HEADERS_CHECK config DEBUG_SECTION_MISMATCH bool "Enable full Section mismatch analysis" - depends on UNDEFINED || (BLACKFIN) default y - # This option is on purpose disabled for now. - # It will be enabled when we are down to a reasonable number - # of section mismatch warnings (< 10 for an allyesconfig build) help The section mismatch analysis checks if there are illegal references from one section to another section. -- cgit v1.2.3-70-g09d2 From 437aa565e2656776a7104aaacd792fe789ea8b2d Mon Sep 17 00:00:00 2001 From: Ivan Djelic Date: Fri, 11 Mar 2011 11:05:32 +0100 Subject: lib: add shared BCH ECC library MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This is a new software BCH encoding/decoding library, similar to the shared Reed-Solomon library. Binary BCH (Bose-Chaudhuri-Hocquenghem) codes are widely used to correct errors in NAND flash devices requiring more than 1-bit ecc correction; they are generally better suited for NAND flash than RS codes because NAND bit errors do not occur in bursts. Latest SLC NAND devices typically require at least 4-bit ecc protection per 512 bytes block. This library provides software encoding/decoding, but may also be used with ASIC/SoC hardware BCH engines to perform error correction. It is being currently used for this purpose on an OMAP3630 board (4bit/8bit HW BCH). It has also been used to decode raw dumps of NAND devices with on-die BCH ecc engines (e.g. Micron 4bit ecc SLC devices). Latest NAND devices (including SLC) can exhibit high error rates (typically a dozen or more bitflips per hour during stress tests); in order to minimize the performance impact of error correction, this library implements recently developed algorithms for fast polynomial root finding (see bch.c header for details) instead of the traditional exhaustive Chien root search; a few performance figures are provided below: Platform: arm926ejs @ 468 MHz, 32 KiB icache, 16 KiB dcache BCH ecc : 4-bit per 512 bytes Encoding average throughput: 250 Mbits/s Error correction time (compared with Chien search): average worst average (Chien) worst (Chien) ---------------------------------------------------------- 1 bit 8.5 µs 11 µs 200 µs 383 µs 2 bit 9.7 µs 12.5 µs 477 µs 728 µs 3 bit 18.1 µs 20.6 µs 758 µs 1010 µs 4 bit 19.5 µs 23 µs 1028 µs 1280 µs In the above figures, "worst" is meant in terms of error pattern, not in terms of cache miss / page faults effects (not taken into account here). The library has been extensively tested on the following platforms: x86, x86_64, arm926ejs, omap3630, qemu-ppc64, qemu-mips. Signed-off-by: Ivan Djelic Signed-off-by: David Woodhouse --- include/linux/bch.h | 79 +++ lib/Kconfig | 39 ++ lib/Makefile | 1 + lib/bch.c | 1368 +++++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 1487 insertions(+) create mode 100644 include/linux/bch.h create mode 100644 lib/bch.c (limited to 'lib') diff --git a/include/linux/bch.h b/include/linux/bch.h new file mode 100644 index 00000000000..295b4ef153b --- /dev/null +++ b/include/linux/bch.h @@ -0,0 +1,79 @@ +/* + * Generic binary BCH encoding/decoding library + * + * This program is free software; you can redistribute it and/or modify it + * 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 that 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. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, write to the Free Software Foundation, Inc., 51 + * Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. + * + * Copyright © 2011 Parrot S.A. + * + * Author: Ivan Djelic + * + * Description: + * + * This library provides runtime configurable encoding/decoding of binary + * Bose-Chaudhuri-Hocquenghem (BCH) codes. +*/ +#ifndef _BCH_H +#define _BCH_H + +#include + +/** + * struct bch_control - BCH control structure + * @m: Galois field order + * @n: maximum codeword size in bits (= 2^m-1) + * @t: error correction capability in bits + * @ecc_bits: ecc exact size in bits, i.e. generator polynomial degree (<=m*t) + * @ecc_bytes: ecc max size (m*t bits) in bytes + * @a_pow_tab: Galois field GF(2^m) exponentiation lookup table + * @a_log_tab: Galois field GF(2^m) log lookup table + * @mod8_tab: remainder generator polynomial lookup tables + * @ecc_buf: ecc parity words buffer + * @ecc_buf2: ecc parity words buffer + * @xi_tab: GF(2^m) base for solving degree 2 polynomial roots + * @syn: syndrome buffer + * @cache: log-based polynomial representation buffer + * @elp: error locator polynomial + * @poly_2t: temporary polynomials of degree 2t + */ +struct bch_control { + unsigned int m; + unsigned int n; + unsigned int t; + unsigned int ecc_bits; + unsigned int ecc_bytes; +/* private: */ + uint16_t *a_pow_tab; + uint16_t *a_log_tab; + uint32_t *mod8_tab; + uint32_t *ecc_buf; + uint32_t *ecc_buf2; + unsigned int *xi_tab; + unsigned int *syn; + int *cache; + struct gf_poly *elp; + struct gf_poly *poly_2t[4]; +}; + +struct bch_control *init_bch(int m, int t, unsigned int prim_poly); + +void free_bch(struct bch_control *bch); + +void encode_bch(struct bch_control *bch, const uint8_t *data, + unsigned int len, uint8_t *ecc); + +int decode_bch(struct bch_control *bch, const uint8_t *data, unsigned int len, + const uint8_t *recv_ecc, const uint8_t *calc_ecc, + const unsigned int *syn, unsigned int *errloc); + +#endif /* _BCH_H */ diff --git a/lib/Kconfig b/lib/Kconfig index 0ee67e08ad3..b9fef78ed04 100644 --- a/lib/Kconfig +++ b/lib/Kconfig @@ -154,6 +154,45 @@ config REED_SOLOMON_ENC16 config REED_SOLOMON_DEC16 boolean +# +# BCH support is selected if needed +# +config BCH + tristate + +config BCH_CONST_PARAMS + boolean + help + Drivers may select this option to force specific constant + values for parameters 'm' (Galois field order) and 't' + (error correction capability). Those specific values must + be set by declaring default values for symbols BCH_CONST_M + and BCH_CONST_T. + Doing so will enable extra compiler optimizations, + improving encoding and decoding performance up to 2x for + usual (m,t) values (typically such that m*t < 200). + When this option is selected, the BCH library supports + only a single (m,t) configuration. This is mainly useful + for NAND flash board drivers requiring known, fixed BCH + parameters. + +config BCH_CONST_M + int + range 5 15 + help + Constant value for Galois field order 'm'. If 'k' is the + number of data bits to protect, 'm' should be chosen such + that (k + m*t) <= 2**m - 1. + Drivers should declare a default value for this symbol if + they select option BCH_CONST_PARAMS. + +config BCH_CONST_T + int + help + Constant value for error correction capability in bits 't'. + Drivers should declare a default value for this symbol if + they select option BCH_CONST_PARAMS. + # # Textsearch support is select'ed if needed # diff --git a/lib/Makefile b/lib/Makefile index cbb774f7d41..0544c814d82 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -67,6 +67,7 @@ obj-$(CONFIG_GENERIC_ALLOCATOR) += genalloc.o obj-$(CONFIG_ZLIB_INFLATE) += zlib_inflate/ obj-$(CONFIG_ZLIB_DEFLATE) += zlib_deflate/ obj-$(CONFIG_REED_SOLOMON) += reed_solomon/ +obj-$(CONFIG_BCH) += bch.o obj-$(CONFIG_LZO_COMPRESS) += lzo/ obj-$(CONFIG_LZO_DECOMPRESS) += lzo/ obj-$(CONFIG_XZ_DEC) += xz/ diff --git a/lib/bch.c b/lib/bch.c new file mode 100644 index 00000000000..bc89dfe4d1b --- /dev/null +++ b/lib/bch.c @@ -0,0 +1,1368 @@ +/* + * Generic binary BCH encoding/decoding library + * + * This program is free software; you can redistribute it and/or modify it + * 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 that 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. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, write to the Free Software Foundation, Inc., 51 + * Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. + * + * Copyright © 2011 Parrot S.A. + * + * Author: Ivan Djelic + * + * Description: + * + * This library provides runtime configurable encoding/decoding of binary + * Bose-Chaudhuri-Hocquenghem (BCH) codes. + * + * Call init_bch to get a pointer to a newly allocated bch_control structure for + * the given m (Galois field order), t (error correction capability) and + * (optional) primitive polynomial parameters. + * + * Call encode_bch to compute and store ecc parity bytes to a given buffer. + * Call decode_bch to detect and locate errors in received data. + * + * On systems supporting hw BCH features, intermediate results may be provided + * to decode_bch in order to skip certain steps. See decode_bch() documentation + * for details. + * + * Option CONFIG_BCH_CONST_PARAMS can be used to force fixed values of + * parameters m and t; thus allowing extra compiler optimizations and providing + * better (up to 2x) encoding performance. Using this option makes sense when + * (m,t) are fixed and known in advance, e.g. when using BCH error correction + * on a particular NAND flash device. + * + * Algorithmic details: + * + * Encoding is performed by processing 32 input bits in parallel, using 4 + * remainder lookup tables. + * + * The final stage of decoding involves the following internal steps: + * a. Syndrome computation + * b. Error locator polynomial computation using Berlekamp-Massey algorithm + * c. Error locator root finding (by far the most expensive step) + * + * In this implementation, step c is not performed using the usual Chien search. + * Instead, an alternative approach described in [1] is used. It consists in + * factoring the error locator polynomial using the Berlekamp Trace algorithm + * (BTA) down to a certain degree (4), after which ad hoc low-degree polynomial + * solving techniques [2] are used. The resulting algorithm, called BTZ, yields + * much better performance than Chien search for usual (m,t) values (typically + * m >= 13, t < 32, see [1]). + * + * [1] B. Biswas, V. Herbert. Efficient root finding of polynomials over fields + * of characteristic 2, in: Western European Workshop on Research in Cryptology + * - WEWoRC 2009, Graz, Austria, LNCS, Springer, July 2009, to appear. + * [2] [Zin96] V.A. Zinoviev. On the solution of equations of degree 10 over + * finite fields GF(2^q). In Rapport de recherche INRIA no 2829, 1996. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#if defined(CONFIG_BCH_CONST_PARAMS) +#define GF_M(_p) (CONFIG_BCH_CONST_M) +#define GF_T(_p) (CONFIG_BCH_CONST_T) +#define GF_N(_p) ((1 << (CONFIG_BCH_CONST_M))-1) +#else +#define GF_M(_p) ((_p)->m) +#define GF_T(_p) ((_p)->t) +#define GF_N(_p) ((_p)->n) +#endif + +#define BCH_ECC_WORDS(_p) DIV_ROUND_UP(GF_M(_p)*GF_T(_p), 32) +#define BCH_ECC_BYTES(_p) DIV_ROUND_UP(GF_M(_p)*GF_T(_p), 8) + +#ifndef dbg +#define dbg(_fmt, args...) do {} while (0) +#endif + +/* + * represent a polynomial over GF(2^m) + */ +struct gf_poly { + unsigned int deg; /* polynomial degree */ + unsigned int c[0]; /* polynomial terms */ +}; + +/* given its degree, compute a polynomial size in bytes */ +#define GF_POLY_SZ(_d) (sizeof(struct gf_poly)+((_d)+1)*sizeof(unsigned int)) + +/* polynomial of degree 1 */ +struct gf_poly_deg1 { + struct gf_poly poly; + unsigned int c[2]; +}; + +/* + * same as encode_bch(), but process input data one byte at a time + */ +static void encode_bch_unaligned(struct bch_control *bch, + const unsigned char *data, unsigned int len, + uint32_t *ecc) +{ + int i; + const uint32_t *p; + const int l = BCH_ECC_WORDS(bch)-1; + + while (len--) { + p = bch->mod8_tab + (l+1)*(((ecc[0] >> 24)^(*data++)) & 0xff); + + for (i = 0; i < l; i++) + ecc[i] = ((ecc[i] << 8)|(ecc[i+1] >> 24))^(*p++); + + ecc[l] = (ecc[l] << 8)^(*p); + } +} + +/* + * convert ecc bytes to aligned, zero-padded 32-bit ecc words + */ +static void load_ecc8(struct bch_control *bch, uint32_t *dst, + const uint8_t *src) +{ + uint8_t pad[4] = {0, 0, 0, 0}; + unsigned int i, nwords = BCH_ECC_WORDS(bch)-1; + + for (i = 0; i < nwords; i++, src += 4) + dst[i] = (src[0] << 24)|(src[1] << 16)|(src[2] << 8)|src[3]; + + memcpy(pad, src, BCH_ECC_BYTES(bch)-4*nwords); + dst[nwords] = (pad[0] << 24)|(pad[1] << 16)|(pad[2] << 8)|pad[3]; +} + +/* + * convert 32-bit ecc words to ecc bytes + */ +static void store_ecc8(struct bch_control *bch, uint8_t *dst, + const uint32_t *src) +{ + uint8_t pad[4]; + unsigned int i, nwords = BCH_ECC_WORDS(bch)-1; + + for (i = 0; i < nwords; i++) { + *dst++ = (src[i] >> 24); + *dst++ = (src[i] >> 16) & 0xff; + *dst++ = (src[i] >> 8) & 0xff; + *dst++ = (src[i] >> 0) & 0xff; + } + pad[0] = (src[nwords] >> 24); + pad[1] = (src[nwords] >> 16) & 0xff; + pad[2] = (src[nwords] >> 8) & 0xff; + pad[3] = (src[nwords] >> 0) & 0xff; + memcpy(dst, pad, BCH_ECC_BYTES(bch)-4*nwords); +} + +/** + * encode_bch - calculate BCH ecc parity of data + * @bch: BCH control structure + * @data: data to encode + * @len: data length in bytes + * @ecc: ecc parity data, must be initialized by caller + * + * The @ecc parity array is used both as input and output parameter, in order to + * allow incremental computations. It should be of the size indicated by member + * @ecc_bytes of @bch, and should be initialized to 0 before the first call. + * + * The exact number of computed ecc parity bits is given by member @ecc_bits of + * @bch; it may be less than m*t for large values of t. + */ +void encode_bch(struct bch_control *bch, const uint8_t *data, + unsigned int len, uint8_t *ecc) +{ + const unsigned int l = BCH_ECC_WORDS(bch)-1; + unsigned int i, mlen; + unsigned long m; + uint32_t w, r[l+1]; + const uint32_t * const tab0 = bch->mod8_tab; + const uint32_t * const tab1 = tab0 + 256*(l+1); + const uint32_t * const tab2 = tab1 + 256*(l+1); + const uint32_t * const tab3 = tab2 + 256*(l+1); + const uint32_t *pdata, *p0, *p1, *p2, *p3; + + if (ecc) { + /* load ecc parity bytes into internal 32-bit buffer */ + load_ecc8(bch, bch->ecc_buf, ecc); + } else { + memset(bch->ecc_buf, 0, sizeof(r)); + } + + /* process first unaligned data bytes */ + m = ((unsigned long)data) & 3; + if (m) { + mlen = (len < (4-m)) ? len : 4-m; + encode_bch_unaligned(bch, data, mlen, bch->ecc_buf); + data += mlen; + len -= mlen; + } + + /* process 32-bit aligned data words */ + pdata = (uint32_t *)data; + mlen = len/4; + data += 4*mlen; + len -= 4*mlen; + memcpy(r, bch->ecc_buf, sizeof(r)); + + /* + * split each 32-bit word into 4 polynomials of weight 8 as follows: + * + * 31 ...24 23 ...16 15 ... 8 7 ... 0 + * xxxxxxxx yyyyyyyy zzzzzzzz tttttttt + * tttttttt mod g = r0 (precomputed) + * zzzzzzzz 00000000 mod g = r1 (precomputed) + * yyyyyyyy 00000000 00000000 mod g = r2 (precomputed) + * xxxxxxxx 00000000 00000000 00000000 mod g = r3 (precomputed) + * xxxxxxxx yyyyyyyy zzzzzzzz tttttttt mod g = r0^r1^r2^r3 + */ + while (mlen--) { + /* input data is read in big-endian format */ + w = r[0]^cpu_to_be32(*pdata++); + p0 = tab0 + (l+1)*((w >> 0) & 0xff); + p1 = tab1 + (l+1)*((w >> 8) & 0xff); + p2 = tab2 + (l+1)*((w >> 16) & 0xff); + p3 = tab3 + (l+1)*((w >> 24) & 0xff); + + for (i = 0; i < l; i++) + r[i] = r[i+1]^p0[i]^p1[i]^p2[i]^p3[i]; + + r[l] = p0[l]^p1[l]^p2[l]^p3[l]; + } + memcpy(bch->ecc_buf, r, sizeof(r)); + + /* process last unaligned bytes */ + if (len) + encode_bch_unaligned(bch, data, len, bch->ecc_buf); + + /* store ecc parity bytes into original parity buffer */ + if (ecc) + store_ecc8(bch, ecc, bch->ecc_buf); +} +EXPORT_SYMBOL_GPL(encode_bch); + +static inline int modulo(struct bch_control *bch, unsigned int v) +{ + const unsigned int n = GF_N(bch); + while (v >= n) { + v -= n; + v = (v & n) + (v >> GF_M(bch)); + } + return v; +} + +/* + * shorter and faster modulo function, only works when v < 2N. + */ +static inline int mod_s(struct bch_control *bch, unsigned int v) +{ + const unsigned int n = GF_N(bch); + return (v < n) ? v : v-n; +} + +static inline int deg(unsigned int poly) +{ + /* polynomial degree is the most-significant bit index */ + return fls(poly)-1; +} + +static inline int parity(unsigned int x) +{ + /* + * public domain code snippet, lifted from + * http://www-graphics.stanford.edu/~seander/bithacks.html + */ + x ^= x >> 1; + x ^= x >> 2; + x = (x & 0x11111111U) * 0x11111111U; + return (x >> 28) & 1; +} + +/* Galois field basic operations: multiply, divide, inverse, etc. */ + +static inline unsigned int gf_mul(struct bch_control *bch, unsigned int a, + unsigned int b) +{ + return (a && b) ? bch->a_pow_tab[mod_s(bch, bch->a_log_tab[a]+ + bch->a_log_tab[b])] : 0; +} + +static inline unsigned int gf_sqr(struct bch_control *bch, unsigned int a) +{ + return a ? bch->a_pow_tab[mod_s(bch, 2*bch->a_log_tab[a])] : 0; +} + +static inline unsigned int gf_div(struct bch_control *bch, unsigned int a, + unsigned int b) +{ + return a ? bch->a_pow_tab[mod_s(bch, bch->a_log_tab[a]+ + GF_N(bch)-bch->a_log_tab[b])] : 0; +} + +static inline unsigned int gf_inv(struct bch_control *bch, unsigned int a) +{ + return bch->a_pow_tab[GF_N(bch)-bch->a_log_tab[a]]; +} + +static inline unsigned int a_pow(struct bch_control *bch, int i) +{ + return bch->a_pow_tab[modulo(bch, i)]; +} + +static inline int a_log(struct bch_control *bch, unsigned int x) +{ + return bch->a_log_tab[x]; +} + +static inline int a_ilog(struct bch_control *bch, unsigned int x) +{ + return mod_s(bch, GF_N(bch)-bch->a_log_tab[x]); +} + +/* + * compute 2t syndromes of ecc polynomial, i.e. ecc(a^j) for j=1..2t + */ +static void compute_syndromes(struct bch_control *bch, uint32_t *ecc, + unsigned int *syn) +{ + int i, j, s; + unsigned int m; + uint32_t poly; + const int t = GF_T(bch); + + s = bch->ecc_bits; + + /* make sure extra bits in last ecc word are cleared */ + m = ((unsigned int)s) & 31; + if (m) + ecc[s/32] &= ~((1u << (32-m))-1); + memset(syn, 0, 2*t*sizeof(*syn)); + + /* compute v(a^j) for j=1 .. 2t-1 */ + do { + poly = *ecc++; + s -= 32; + while (poly) { + i = deg(poly); + for (j = 0; j < 2*t; j += 2) + syn[j] ^= a_pow(bch, (j+1)*(i+s)); + + poly ^= (1 << i); + } + } while (s > 0); + + /* v(a^(2j)) = v(a^j)^2 */ + for (j = 0; j < t; j++) + syn[2*j+1] = gf_sqr(bch, syn[j]); +} + +static void gf_poly_copy(struct gf_poly *dst, struct gf_poly *src) +{ + memcpy(dst, src, GF_POLY_SZ(src->deg)); +} + +static int compute_error_locator_polynomial(struct bch_control *bch, + const unsigned int *syn) +{ + const unsigned int t = GF_T(bch); + const unsigned int n = GF_N(bch); + unsigned int i, j, tmp, l, pd = 1, d = syn[0]; + struct gf_poly *elp = bch->elp; + struct gf_poly *pelp = bch->poly_2t[0]; + struct gf_poly *elp_copy = bch->poly_2t[1]; + int k, pp = -1; + + memset(pelp, 0, GF_POLY_SZ(2*t)); + memset(elp, 0, GF_POLY_SZ(2*t)); + + pelp->deg = 0; + pelp->c[0] = 1; + elp->deg = 0; + elp->c[0] = 1; + + /* use simplified binary Berlekamp-Massey algorithm */ + for (i = 0; (i < t) && (elp->deg <= t); i++) { + if (d) { + k = 2*i-pp; + gf_poly_copy(elp_copy, elp); + /* e[i+1](X) = e[i](X)+di*dp^-1*X^2(i-p)*e[p](X) */ + tmp = a_log(bch, d)+n-a_log(bch, pd); + for (j = 0; j <= pelp->deg; j++) { + if (pelp->c[j]) { + l = a_log(bch, pelp->c[j]); + elp->c[j+k] ^= a_pow(bch, tmp+l); + } + } + /* compute l[i+1] = max(l[i]->c[l[p]+2*(i-p]) */ + tmp = pelp->deg+k; + if (tmp > elp->deg) { + elp->deg = tmp; + gf_poly_copy(pelp, elp_copy); + pd = d; + pp = 2*i; + } + } + /* di+1 = S(2i+3)+elp[i+1].1*S(2i+2)+...+elp[i+1].lS(2i+3-l) */ + if (i < t-1) { + d = syn[2*i+2]; + for (j = 1; j <= elp->deg; j++) + d ^= gf_mul(bch, elp->c[j], syn[2*i+2-j]); + } + } + dbg("elp=%s\n", gf_poly_str(elp)); + return (elp->deg > t) ? -1 : (int)elp->deg; +} + +/* + * solve a m x m linear system in GF(2) with an expected number of solutions, + * and return the number of found solutions + */ +static int solve_linear_system(struct bch_control *bch, unsigned int *rows, + unsigned int *sol, int nsol) +{ + const int m = GF_M(bch); + unsigned int tmp, mask; + int rem, c, r, p, k, param[m]; + + k = 0; + mask = 1 << m; + + /* Gaussian elimination */ + for (c = 0; c < m; c++) { + rem = 0; + p = c-k; + /* find suitable row for elimination */ + for (r = p; r < m; r++) { + if (rows[r] & mask) { + if (r != p) { + tmp = rows[r]; + rows[r] = rows[p]; + rows[p] = tmp; + } + rem = r+1; + break; + } + } + if (rem) { + /* perform elimination on remaining rows */ + tmp = rows[p]; + for (r = rem; r < m; r++) { + if (rows[r] & mask) + rows[r] ^= tmp; + } + } else { + /* elimination not needed, store defective row index */ + param[k++] = c; + } + mask >>= 1; + } + /* rewrite system, inserting fake parameter rows */ + if (k > 0) { + p = k; + for (r = m-1; r >= 0; r--) { + if ((r > m-1-k) && rows[r]) + /* system has no solution */ + return 0; + + rows[r] = (p && (r == param[p-1])) ? + p--, 1u << (m-r) : rows[r-p]; + } + } + + if (nsol != (1 << k)) + /* unexpected number of solutions */ + return 0; + + for (p = 0; p < nsol; p++) { + /* set parameters for p-th solution */ + for (c = 0; c < k; c++) + rows[param[c]] = (rows[param[c]] & ~1)|((p >> c) & 1); + + /* compute unique solution */ + tmp = 0; + for (r = m-1; r >= 0; r--) { + mask = rows[r] & (tmp|1); + tmp |= parity(mask) << (m-r); + } + sol[p] = tmp >> 1; + } + return nsol; +} + +/* + * this function builds and solves a linear system for finding roots of a degree + * 4 affine monic polynomial X^4+aX^2+bX+c over GF(2^m). + */ +static int find_affine4_roots(struct bch_control *bch, unsigned int a, + unsigned int b, unsigned int c, + unsigned int *roots) +{ + int i, j, k; + const int m = GF_M(bch); + unsigned int mask = 0xff, t, rows[16] = {0,}; + + j = a_log(bch, b); + k = a_log(bch, a); + rows[0] = c; + + /* buid linear system to solve X^4+aX^2+bX+c = 0 */ + for (i = 0; i < m; i++) { + rows[i+1] = bch->a_pow_tab[4*i]^ + (a ? bch->a_pow_tab[mod_s(bch, k)] : 0)^ + (b ? bch->a_pow_tab[mod_s(bch, j)] : 0); + j++; + k += 2; + } + /* + * transpose 16x16 matrix before passing it to linear solver + * warning: this code assumes m < 16 + */ + for (j = 8; j != 0; j >>= 1, mask ^= (mask << j)) { + for (k = 0; k < 16; k = (k+j+1) & ~j) { + t = ((rows[k] >> j)^rows[k+j]) & mask; + rows[k] ^= (t << j); + rows[k+j] ^= t; + } + } + return solve_linear_system(bch, rows, roots, 4); +} + +/* + * compute root r of a degree 1 polynomial over GF(2^m) (returned as log(1/r)) + */ +static int find_poly_deg1_roots(struct bch_control *bch, struct gf_poly *poly, + unsigned int *roots) +{ + int n = 0; + + if (poly->c[0]) + /* poly[X] = bX+c with c!=0, root=c/b */ + roots[n++] = mod_s(bch, GF_N(bch)-bch->a_log_tab[poly->c[0]]+ + bch->a_log_tab[poly->c[1]]); + return n; +} + +/* + * compute roots of a degree 2 polynomial over GF(2^m) + */ +static int find_poly_deg2_roots(struct bch_control *bch, struct gf_poly *poly, + unsigned int *roots) +{ + int n = 0, i, l0, l1, l2; + unsigned int u, v, r; + + if (poly->c[0] && poly->c[1]) { + + l0 = bch->a_log_tab[poly->c[0]]; + l1 = bch->a_log_tab[poly->c[1]]; + l2 = bch->a_log_tab[poly->c[2]]; + + /* using z=a/bX, transform aX^2+bX+c into z^2+z+u (u=ac/b^2) */ + u = a_pow(bch, l0+l2+2*(GF_N(bch)-l1)); + /* + * let u = sum(li.a^i) i=0..m-1; then compute r = sum(li.xi): + * r^2+r = sum(li.(xi^2+xi)) = sum(li.(a^i+Tr(a^i).a^k)) = + * u + sum(li.Tr(a^i).a^k) = u+a^k.Tr(sum(li.a^i)) = u+a^k.Tr(u) + * i.e. r and r+1 are roots iff Tr(u)=0 + */ + r = 0; + v = u; + while (v) { + i = deg(v); + r ^= bch->xi_tab[i]; + v ^= (1 << i); + } + /* verify root */ + if ((gf_sqr(bch, r)^r) == u) { + /* reverse z=a/bX transformation and compute log(1/r) */ + roots[n++] = modulo(bch, 2*GF_N(bch)-l1- + bch->a_log_tab[r]+l2); + roots[n++] = modulo(bch, 2*GF_N(bch)-l1- + bch->a_log_tab[r^1]+l2); + } + } + return n; +} + +/* + * compute roots of a degree 3 polynomial over GF(2^m) + */ +static int find_poly_deg3_roots(struct bch_control *bch, struct gf_poly *poly, + unsigned int *roots) +{ + int i, n = 0; + unsigned int a, b, c, a2, b2, c2, e3, tmp[4]; + + if (poly->c[0]) { + /* transform polynomial into monic X^3 + a2X^2 + b2X + c2 */ + e3 = poly->c[3]; + c2 = gf_div(bch, poly->c[0], e3); + b2 = gf_div(bch, poly->c[1], e3); + a2 = gf_div(bch, poly->c[2], e3); + + /* (X+a2)(X^3+a2X^2+b2X+c2) = X^4+aX^2+bX+c (affine) */ + c = gf_mul(bch, a2, c2); /* c = a2c2 */ + b = gf_mul(bch, a2, b2)^c2; /* b = a2b2 + c2 */ + a = gf_sqr(bch, a2)^b2; /* a = a2^2 + b2 */ + + /* find the 4 roots of this affine polynomial */ + if (find_affine4_roots(bch, a, b, c, tmp) == 4) { + /* remove a2 from final list of roots */ + for (i = 0; i < 4; i++) { + if (tmp[i] != a2) + roots[n++] = a_ilog(bch, tmp[i]); + } + } + } + return n; +} + +/* + * compute roots of a degree 4 polynomial over GF(2^m) + */ +static int find_poly_deg4_roots(struct bch_control *bch, struct gf_poly *poly, + unsigned int *roots) +{ + int i, l, n = 0; + unsigned int a, b, c, d, e = 0, f, a2, b2, c2, e4; + + if (poly->c[0] == 0) + return 0; + + /* transform polynomial into monic X^4 + aX^3 + bX^2 + cX + d */ + e4 = poly->c[4]; + d = gf_div(bch, poly->c[0], e4); + c = gf_div(bch, poly->c[1], e4); + b = gf_div(bch, poly->c[2], e4); + a = gf_div(bch, poly->c[3], e4); + + /* use Y=1/X transformation to get an affine polynomial */ + if (a) { + /* first, eliminate cX by using z=X+e with ae^2+c=0 */ + if (c) { + /* compute e such that e^2 = c/a */ + f = gf_div(bch, c, a); + l = a_log(bch, f); + l += (l & 1) ? GF_N(bch) : 0; + e = a_pow(bch, l/2); + /* + * use transformation z=X+e: + * z^4+e^4 + a(z^3+ez^2+e^2z+e^3) + b(z^2+e^2) +cz+ce+d + * z^4 + az^3 + (ae+b)z^2 + (ae^2+c)z+e^4+be^2+ae^3+ce+d + * z^4 + az^3 + (ae+b)z^2 + e^4+be^2+d + * z^4 + az^3 + b'z^2 + d' + */ + d = a_pow(bch, 2*l)^gf_mul(bch, b, f)^d; + b = gf_mul(bch, a, e)^b; + } + /* now, use Y=1/X to get Y^4 + b/dY^2 + a/dY + 1/d */ + if (d == 0) + /* assume all roots have multiplicity 1 */ + return 0; + + c2 = gf_inv(bch, d); + b2 = gf_div(bch, a, d); + a2 = gf_div(bch, b, d); + } else { + /* polynomial is already affine */ + c2 = d; + b2 = c; + a2 = b; + } + /* find the 4 roots of this affine polynomial */ + if (find_affine4_roots(bch, a2, b2, c2, roots) == 4) { + for (i = 0; i < 4; i++) { + /* post-process roots (reverse transformations) */ + f = a ? gf_inv(bch, roots[i]) : roots[i]; + roots[i] = a_ilog(bch, f^e); + } + n = 4; + } + return n; +} + +/* + * build monic, log-based representation of a polynomial + */ +static void gf_poly_logrep(struct bch_control *bch, + const struct gf_poly *a, int *rep) +{ + int i, d = a->deg, l = GF_N(bch)-a_log(bch, a->c[a->deg]); + + /* represent 0 values with -1; warning, rep[d] is not set to 1 */ + for (i = 0; i < d; i++) + rep[i] = a->c[i] ? mod_s(bch, a_log(bch, a->c[i])+l) : -1; +} + +/* + * compute polynomial Euclidean division remainder in GF(2^m)[X] + */ +static void gf_poly_mod(struct bch_control *bch, struct gf_poly *a, + const struct gf_poly *b, int *rep) +{ + int la, p, m; + unsigned int i, j, *c = a->c; + const unsigned int d = b->deg; + + if (a->deg < d) + return; + + /* reuse or compute log representation of denominator */ + if (!rep) { + rep = bch->cache; + gf_poly_logrep(bch, b, rep); + } + + for (j = a->deg; j >= d; j--) { + if (c[j]) { + la = a_log(bch, c[j]); + p = j-d; + for (i = 0; i < d; i++, p++) { + m = rep[i]; + if (m >= 0) + c[p] ^= bch->a_pow_tab[mod_s(bch, + m+la)]; + } + } + } + a->deg = d-1; + while (!c[a->deg] && a->deg) + a->deg--; +} + +/* + * compute polynomial Euclidean division quotient in GF(2^m)[X] + */ +static void gf_poly_div(struct bch_control *bch, struct gf_poly *a, + const struct gf_poly *b, struct gf_poly *q) +{ + if (a->deg >= b->deg) { + q->deg = a->deg-b->deg; + /* compute a mod b (modifies a) */ + gf_poly_mod(bch, a, b, NULL); + /* quotient is stored in upper part of polynomial a */ + memcpy(q->c, &a->c[b->deg], (1+q->deg)*sizeof(unsigned int)); + } else { + q->deg = 0; + q->c[0] = 0; + } +} + +/* + * compute polynomial GCD (Greatest Common Divisor) in GF(2^m)[X] + */ +static struct gf_poly *gf_poly_gcd(struct bch_control *bch, struct gf_poly *a, + struct gf_poly *b) +{ + struct gf_poly *tmp; + + dbg("gcd(%s,%s)=", gf_poly_str(a), gf_poly_str(b)); + + if (a->deg < b->deg) { + tmp = b; + b = a; + a = tmp; + } + + while (b->deg > 0) { + gf_poly_mod(bch, a, b, NULL); + tmp = b; + b = a; + a = tmp; + } + + dbg("%s\n", gf_poly_str(a)); + + return a; +} + +/* + * Given a polynomial f and an integer k, compute Tr(a^kX) mod f + * This is used in Berlekamp Trace algorithm for splitting polynomials + */ +static void compute_trace_bk_mod(struct bch_control *bch, int k, + const struct gf_poly *f, struct gf_poly *z, + struct gf_poly *out) +{ + const int m = GF_M(bch); + int i, j; + + /* z contains z^2j mod f */ + z->deg = 1; + z->c[0] = 0; + z->c[1] = bch->a_pow_tab[k]; + + out->deg = 0; + memset(out, 0, GF_POLY_SZ(f->deg)); + + /* compute f log representation only once */ + gf_poly_logrep(bch, f, bch->cache); + + for (i = 0; i < m; i++) { + /* add a^(k*2^i)(z^(2^i) mod f) and compute (z^(2^i) mod f)^2 */ + for (j = z->deg; j >= 0; j--) { + out->c[j] ^= z->c[j]; + z->c[2*j] = gf_sqr(bch, z->c[j]); + z->c[2*j+1] = 0; + } + if (z->deg > out->deg) + out->deg = z->deg; + + if (i < m-1) { + z->deg *= 2; + /* z^(2(i+1)) mod f = (z^(2^i) mod f)^2 mod f */ + gf_poly_mod(bch, z, f, bch->cache); + } + } + while (!out->c[out->deg] && out->deg) + out->deg--; + + dbg("Tr(a^%d.X) mod f = %s\n", k, gf_poly_str(out)); +} + +/* + * factor a polynomial using Berlekamp Trace algorithm (BTA) + */ +static void factor_polynomial(struct bch_control *bch, int k, struct gf_poly *f, + struct gf_poly **g, struct gf_poly **h) +{ + struct gf_poly *f2 = bch->poly_2t[0]; + struct gf_poly *q = bch->poly_2t[1]; + struct gf_poly *tk = bch->poly_2t[2]; + struct gf_poly *z = bch->poly_2t[3]; + struct gf_poly *gcd; + + dbg("factoring %s...\n", gf_poly_str(f)); + + *g = f; + *h = NULL; + + /* tk = Tr(a^k.X) mod f */ + compute_trace_bk_mod(bch, k, f, z, tk); + + if (tk->deg > 0) { + /* compute g = gcd(f, tk) (destructive operation) */ + gf_poly_copy(f2, f); + gcd = gf_poly_gcd(bch, f2, tk); + if (gcd->deg < f->deg) { + /* compute h=f/gcd(f,tk); this will modify f and q */ + gf_poly_div(bch, f, gcd, q); + /* store g and h in-place (clobbering f) */ + *h = &((struct gf_poly_deg1 *)f)[gcd->deg].poly; + gf_poly_copy(*g, gcd); + gf_poly_copy(*h, q); + } + } +} + +/* + * find roots of a polynomial, using BTZ algorithm; see the beginning of this + * file for details + */ +static int find_poly_roots(struct bch_control *bch, unsigned int k, + struct gf_poly *poly, unsigned int *roots) +{ + int cnt; + struct gf_poly *f1, *f2; + + switch (poly->deg) { + /* handle low degree polynomials with ad hoc techniques */ + case 1: + cnt = find_poly_deg1_roots(bch, poly, roots); + break; + case 2: + cnt = find_poly_deg2_roots(bch, poly, roots); + break; + case 3: + cnt = find_poly_deg3_roots(bch, poly, roots); + break; + case 4: + cnt = find_poly_deg4_roots(bch, poly, roots); + break; + default: + /* factor polynomial using Berlekamp Trace Algorithm (BTA) */ + cnt = 0; + if (poly->deg && (k <= GF_M(bch))) { + factor_polynomial(bch, k, poly, &f1, &f2); + if (f1) + cnt += find_poly_roots(bch, k+1, f1, roots); + if (f2) + cnt += find_poly_roots(bch, k+1, f2, roots+cnt); + } + break; + } + return cnt; +} + +#if defined(USE_CHIEN_SEARCH) +/* + * exhaustive root search (Chien) implementation - not used, included only for + * reference/comparison tests + */ +static int chien_search(struct bch_control *bch, unsigned int len, + struct gf_poly *p, unsigned int *roots) +{ + int m; + unsigned int i, j, syn, syn0, count = 0; + const unsigned int k = 8*len+bch->ecc_bits; + + /* use a log-based representation of polynomial */ + gf_poly_logrep(bch, p, bch->cache); + bch->cache[p->deg] = 0; + syn0 = gf_div(bch, p->c[0], p->c[p->deg]); + + for (i = GF_N(bch)-k+1; i <= GF_N(bch); i++) { + /* compute elp(a^i) */ + for (j = 1, syn = syn0; j <= p->deg; j++) { + m = bch->cache[j]; + if (m >= 0) + syn ^= a_pow(bch, m+j*i); + } + if (syn == 0) { + roots[count++] = GF_N(bch)-i; + if (count == p->deg) + break; + } + } + return (count == p->deg) ? count : 0; +} +#define find_poly_roots(_p, _k, _elp, _loc) chien_search(_p, len, _elp, _loc) +#endif /* USE_CHIEN_SEARCH */ + +/** + * decode_bch - decode received codeword and find bit error locations + * @bch: BCH control structure + * @data: received data, ignored if @calc_ecc is provided + * @len: data length in bytes, must always be provided + * @recv_ecc: received ecc, if NULL then assume it was XORed in @calc_ecc + * @calc_ecc: calculated ecc, if NULL then calc_ecc is computed from @data + * @syn: hw computed syndrome data (if NULL, syndrome is calculated) + * @errloc: output array of error locations + * + * Returns: + * The number of errors found, or -EBADMSG if decoding failed, or -EINVAL if + * invalid parameters were provided + * + * Depending on the available hw BCH support and the need to compute @calc_ecc + * separately (using encode_bch()), this function should be called with one of + * the following parameter configurations - + * + * by providing @data and @recv_ecc only: + * decode_bch(@bch, @data, @len, @recv_ecc, NULL, NULL, @errloc) + * + * by providing @recv_ecc and @calc_ecc: + * decode_bch(@bch, NULL, @len, @recv_ecc, @calc_ecc, NULL, @errloc) + * + * by providing ecc = recv_ecc XOR calc_ecc: + * decode_bch(@bch, NULL, @len, NULL, ecc, NULL, @errloc) + * + * by providing syndrome results @syn: + * decode_bch(@bch, NULL, @len, NULL, NULL, @syn, @errloc) + * + * Once decode_bch() has successfully returned with a positive value, error + * locations returned in array @errloc should be interpreted as follows - + * + * if (errloc[n] >= 8*len), then n-th error is located in ecc (no need for + * data correction) + * + * if (errloc[n] < 8*len), then n-th error is located in data and can be + * corrected with statement data[errloc[n]/8] ^= 1 << (errloc[n] % 8); + * + * Note that this function does not perform any data correction by itself, it + * merely indicates error locations. + */ +int decode_bch(struct bch_control *bch, const uint8_t *data, unsigned int len, + const uint8_t *recv_ecc, const uint8_t *calc_ecc, + const unsigned int *syn, unsigned int *errloc) +{ + const unsigned int ecc_words = BCH_ECC_WORDS(bch); + unsigned int nbits; + int i, err, nroots; + uint32_t sum; + + /* sanity check: make sure data length can be handled */ + if (8*len > (bch->n-bch->ecc_bits)) + return -EINVAL; + + /* if caller does not provide syndromes, compute them */ + if (!syn) { + if (!calc_ecc) { + /* compute received data ecc into an internal buffer */ + if (!data || !recv_ecc) + return -EINVAL; + encode_bch(bch, data, len, NULL); + } else { + /* load provided calculated ecc */ + load_ecc8(bch, bch->ecc_buf, calc_ecc); + } + /* load received ecc or assume it was XORed in calc_ecc */ + if (recv_ecc) { + load_ecc8(bch, bch->ecc_buf2, recv_ecc); + /* XOR received and calculated ecc */ + for (i = 0, sum = 0; i < (int)ecc_words; i++) { + bch->ecc_buf[i] ^= bch->ecc_buf2[i]; + sum |= bch->ecc_buf[i]; + } + if (!sum) + /* no error found */ + return 0; + } + compute_syndromes(bch, bch->ecc_buf, bch->syn); + syn = bch->syn; + } + + err = compute_error_locator_polynomial(bch, syn); + if (err > 0) { + nroots = find_poly_roots(bch, 1, bch->elp, errloc); + if (err != nroots) + err = -1; + } + if (err > 0) { + /* post-process raw error locations for easier correction */ + nbits = (len*8)+bch->ecc_bits; + for (i = 0; i < err; i++) { + if (errloc[i] >= nbits) { + err = -1; + break; + } + errloc[i] = nbits-1-errloc[i]; + errloc[i] = (errloc[i] & ~7)|(7-(errloc[i] & 7)); + } + } + return (err >= 0) ? err : -EBADMSG; +} +EXPORT_SYMBOL_GPL(decode_bch); + +/* + * generate Galois field lookup tables + */ +static int build_gf_tables(struct bch_control *bch, unsigned int poly) +{ + unsigned int i, x = 1; + const unsigned int k = 1 << deg(poly); + + /* primitive polynomial must be of degree m */ + if (k != (1u << GF_M(bch))) + return -1; + + for (i = 0; i < GF_N(bch); i++) { + bch->a_pow_tab[i] = x; + bch->a_log_tab[x] = i; + if (i && (x == 1)) + /* polynomial is not primitive (a^i=1 with 0a_pow_tab[GF_N(bch)] = 1; + bch->a_log_tab[0] = 0; + + return 0; +} + +/* + * compute generator polynomial remainder tables for fast encoding + */ +static void build_mod8_tables(struct bch_control *bch, const uint32_t *g) +{ + int i, j, b, d; + uint32_t data, hi, lo, *tab; + const int l = BCH_ECC_WORDS(bch); + const int plen = DIV_ROUND_UP(bch->ecc_bits+1, 32); + const int ecclen = DIV_ROUND_UP(bch->ecc_bits, 32); + + memset(bch->mod8_tab, 0, 4*256*l*sizeof(*bch->mod8_tab)); + + for (i = 0; i < 256; i++) { + /* p(X)=i is a small polynomial of weight <= 8 */ + for (b = 0; b < 4; b++) { + /* we want to compute (p(X).X^(8*b+deg(g))) mod g(X) */ + tab = bch->mod8_tab + (b*256+i)*l; + data = i << (8*b); + while (data) { + d = deg(data); + /* subtract X^d.g(X) from p(X).X^(8*b+deg(g)) */ + data ^= g[0] >> (31-d); + for (j = 0; j < ecclen; j++) { + hi = (d < 31) ? g[j] << (d+1) : 0; + lo = (j+1 < plen) ? + g[j+1] >> (31-d) : 0; + tab[j] ^= hi|lo; + } + } + } + } +} + +/* + * build a base for factoring degree 2 polynomials + */ +static int build_deg2_base(struct bch_control *bch) +{ + const int m = GF_M(bch); + int i, j, r; + unsigned int sum, x, y, remaining, ak = 0, xi[m]; + + /* find k s.t. Tr(a^k) = 1 and 0 <= k < m */ + for (i = 0; i < m; i++) { + for (j = 0, sum = 0; j < m; j++) + sum ^= a_pow(bch, i*(1 << j)); + + if (sum) { + ak = bch->a_pow_tab[i]; + break; + } + } + /* find xi, i=0..m-1 such that xi^2+xi = a^i+Tr(a^i).a^k */ + remaining = m; + memset(xi, 0, sizeof(xi)); + + for (x = 0; (x <= GF_N(bch)) && remaining; x++) { + y = gf_sqr(bch, x)^x; + for (i = 0; i < 2; i++) { + r = a_log(bch, y); + if (y && (r < m) && !xi[r]) { + bch->xi_tab[r] = x; + xi[r] = 1; + remaining--; + dbg("x%d = %x\n", r, x); + break; + } + y ^= ak; + } + } + /* should not happen but check anyway */ + return remaining ? -1 : 0; +} + +static void *bch_alloc(size_t size, int *err) +{ + void *ptr; + + ptr = kmalloc(size, GFP_KERNEL); + if (ptr == NULL) + *err = 1; + return ptr; +} + +/* + * compute generator polynomial for given (m,t) parameters. + */ +static uint32_t *compute_generator_polynomial(struct bch_control *bch) +{ + const unsigned int m = GF_M(bch); + const unsigned int t = GF_T(bch); + int n, err = 0; + unsigned int i, j, nbits, r, word, *roots; + struct gf_poly *g; + uint32_t *genpoly; + + g = bch_alloc(GF_POLY_SZ(m*t), &err); + roots = bch_alloc((bch->n+1)*sizeof(*roots), &err); + genpoly = bch_alloc(DIV_ROUND_UP(m*t+1, 32)*sizeof(*genpoly), &err); + + if (err) { + kfree(genpoly); + genpoly = NULL; + goto finish; + } + + /* enumerate all roots of g(X) */ + memset(roots , 0, (bch->n+1)*sizeof(*roots)); + for (i = 0; i < t; i++) { + for (j = 0, r = 2*i+1; j < m; j++) { + roots[r] = 1; + r = mod_s(bch, 2*r); + } + } + /* build generator polynomial g(X) */ + g->deg = 0; + g->c[0] = 1; + for (i = 0; i < GF_N(bch); i++) { + if (roots[i]) { + /* multiply g(X) by (X+root) */ + r = bch->a_pow_tab[i]; + g->c[g->deg+1] = 1; + for (j = g->deg; j > 0; j--) + g->c[j] = gf_mul(bch, g->c[j], r)^g->c[j-1]; + + g->c[0] = gf_mul(bch, g->c[0], r); + g->deg++; + } + } + /* store left-justified binary representation of g(X) */ + n = g->deg+1; + i = 0; + + while (n > 0) { + nbits = (n > 32) ? 32 : n; + for (j = 0, word = 0; j < nbits; j++) { + if (g->c[n-1-j]) + word |= 1u << (31-j); + } + genpoly[i++] = word; + n -= nbits; + } + bch->ecc_bits = g->deg; + +finish: + kfree(g); + kfree(roots); + + return genpoly; +} + +/** + * init_bch - initialize a BCH encoder/decoder + * @m: Galois field order, should be in the range 5-15 + * @t: maximum error correction capability, in bits + * @prim_poly: user-provided primitive polynomial (or 0 to use default) + * + * Returns: + * a newly allocated BCH control structure if successful, NULL otherwise + * + * This initialization can take some time, as lookup tables are built for fast + * encoding/decoding; make sure not to call this function from a time critical + * path. Usually, init_bch() should be called on module/driver init and + * free_bch() should be called to release memory on exit. + * + * You may provide your own primitive polynomial of degree @m in argument + * @prim_poly, or let init_bch() use its default polynomial. + * + * Once init_bch() has successfully returned a pointer to a newly allocated + * BCH control structure, ecc length in bytes is given by member @ecc_bytes of + * the structure. + */ +struct bch_control *init_bch(int m, int t, unsigned int prim_poly) +{ + int err = 0; + unsigned int i, words; + uint32_t *genpoly; + struct bch_control *bch = NULL; + + const int min_m = 5; + const int max_m = 15; + + /* default primitive polynomials */ + static const unsigned int prim_poly_tab[] = { + 0x25, 0x43, 0x83, 0x11d, 0x211, 0x409, 0x805, 0x1053, 0x201b, + 0x402b, 0x8003, + }; + +#if defined(CONFIG_BCH_CONST_PARAMS) + if ((m != (CONFIG_BCH_CONST_M)) || (t != (CONFIG_BCH_CONST_T))) { + printk(KERN_ERR "bch encoder/decoder was configured to support " + "parameters m=%d, t=%d only!\n", + CONFIG_BCH_CONST_M, CONFIG_BCH_CONST_T); + goto fail; + } +#endif + if ((m < min_m) || (m > max_m)) + /* + * values of m greater than 15 are not currently supported; + * supporting m > 15 would require changing table base type + * (uint16_t) and a small patch in matrix transposition + */ + goto fail; + + /* sanity checks */ + if ((t < 1) || (m*t >= ((1 << m)-1))) + /* invalid t value */ + goto fail; + + /* select a primitive polynomial for generating GF(2^m) */ + if (prim_poly == 0) + prim_poly = prim_poly_tab[m-min_m]; + + bch = kzalloc(sizeof(*bch), GFP_KERNEL); + if (bch == NULL) + goto fail; + + bch->m = m; + bch->t = t; + bch->n = (1 << m)-1; + words = DIV_ROUND_UP(m*t, 32); + bch->ecc_bytes = DIV_ROUND_UP(m*t, 8); + bch->a_pow_tab = bch_alloc((1+bch->n)*sizeof(*bch->a_pow_tab), &err); + bch->a_log_tab = bch_alloc((1+bch->n)*sizeof(*bch->a_log_tab), &err); + bch->mod8_tab = bch_alloc(words*1024*sizeof(*bch->mod8_tab), &err); + bch->ecc_buf = bch_alloc(words*sizeof(*bch->ecc_buf), &err); + bch->ecc_buf2 = bch_alloc(words*sizeof(*bch->ecc_buf2), &err); + bch->xi_tab = bch_alloc(m*sizeof(*bch->xi_tab), &err); + bch->syn = bch_alloc(2*t*sizeof(*bch->syn), &err); + bch->cache = bch_alloc(2*t*sizeof(*bch->cache), &err); + bch->elp = bch_alloc((t+1)*sizeof(struct gf_poly_deg1), &err); + + for (i = 0; i < ARRAY_SIZE(bch->poly_2t); i++) + bch->poly_2t[i] = bch_alloc(GF_POLY_SZ(2*t), &err); + + if (err) + goto fail; + + err = build_gf_tables(bch, prim_poly); + if (err) + goto fail; + + /* use generator polynomial for computing encoding tables */ + genpoly = compute_generator_polynomial(bch); + if (genpoly == NULL) + goto fail; + + build_mod8_tables(bch, genpoly); + kfree(genpoly); + + err = build_deg2_base(bch); + if (err) + goto fail; + + return bch; + +fail: + free_bch(bch); + return NULL; +} +EXPORT_SYMBOL_GPL(init_bch); + +/** + * free_bch - free the BCH control structure + * @bch: BCH control structure to release + */ +void free_bch(struct bch_control *bch) +{ + unsigned int i; + + if (bch) { + kfree(bch->a_pow_tab); + kfree(bch->a_log_tab); + kfree(bch->mod8_tab); + kfree(bch->ecc_buf); + kfree(bch->ecc_buf2); + kfree(bch->xi_tab); + kfree(bch->syn); + kfree(bch->cache); + kfree(bch->elp); + + for (i = 0; i < ARRAY_SIZE(bch->poly_2t); i++) + kfree(bch->poly_2t[i]); + + kfree(bch); + } +} +EXPORT_SYMBOL_GPL(free_bch); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Ivan Djelic "); +MODULE_DESCRIPTION("Binary BCH encoder/decoder"); -- cgit v1.2.3-70-g09d2 From f2c23f65f63fe0dd83fc94efdfae0364c74458b8 Mon Sep 17 00:00:00 2001 From: Michal Marek Date: Mon, 21 Mar 2011 10:44:30 +0100 Subject: kbuild: Make DEBUG_SECTION_MISMATCH selectable, but not on by default CONFIG_DEBUG_SECTION_MISMATCH has also runtime effects due to the -fno-inline-functions-called-once compiler flag, so forcing it on everyone is not a good idea. Signed-off-by: Michal Marek --- lib/Kconfig.debug | 1 - 1 file changed, 1 deletion(-) (limited to 'lib') diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 8b6a4f13946..5a24805d432 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -102,7 +102,6 @@ config HEADERS_CHECK config DEBUG_SECTION_MISMATCH bool "Enable full Section mismatch analysis" - default y help The section mismatch analysis checks if there are illegal references from one section to another section. -- cgit v1.2.3-70-g09d2 From ddd588b5dd55f14320379961e47683db4e4c1d90 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Tue, 22 Mar 2011 16:30:46 -0700 Subject: oom: suppress nodes that are not allowed from meminfo on oom kill The oom killer is extremely verbose for machines with a large number of cpus and/or nodes. This verbosity can often be harmful if it causes other important messages to be scrolled from the kernel log and incurs a signicant time delay, specifically for kernels with CONFIG_NODES_SHIFT > 8. This patch causes only memory information to be displayed for nodes that are allowed by current's cpuset when dumping the VM state. Information for all other nodes is irrelevant to the oom condition; we don't care if there's an abundance of memory elsewhere if we can't access it. This only affects the behavior of dumping memory information when an oom is triggered. Other dumps, such as for sysrq+m, still display the unfiltered form when using the existing show_mem() interface. Additionally, the per-cpu pageset statistics are extremely verbose in oom killer output, so it is now suppressed. This removes nodes_weight(current->mems_allowed) * (1 + nr_cpus) lines from the oom killer output. Callers may use __show_mem(SHOW_MEM_FILTER_NODES) to filter disallowed nodes. Signed-off-by: David Rientjes Cc: Mel Gorman Cc: KAMEZAWA Hiroyuki Cc: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 8 ++++++++ lib/show_mem.c | 9 +++++++-- mm/oom_kill.c | 2 +- mm/page_alloc.c | 34 +++++++++++++++++++++++++++++++++- 4 files changed, 49 insertions(+), 4 deletions(-) (limited to 'lib') diff --git a/include/linux/mm.h b/include/linux/mm.h index 581703d86fb..1f82adc8535 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -859,7 +859,14 @@ extern void pagefault_out_of_memory(void); #define offset_in_page(p) ((unsigned long)(p) & ~PAGE_MASK) +/* + * Flags passed to __show_mem() and __show_free_areas() to suppress output in + * various contexts. + */ +#define SHOW_MEM_FILTER_NODES (0x0001u) /* filter disallowed nodes */ + extern void show_free_areas(void); +extern void __show_free_areas(unsigned int flags); int shmem_lock(struct file *file, int lock, struct user_struct *user); struct file *shmem_file_setup(const char *name, loff_t size, unsigned long flags); @@ -1348,6 +1355,7 @@ extern void calculate_zone_inactive_ratio(struct zone *zone); extern void mem_init(void); extern void __init mmap_init(void); extern void show_mem(void); +extern void __show_mem(unsigned int flags); extern void si_meminfo(struct sysinfo * val); extern void si_meminfo_node(struct sysinfo *val, int nid); extern int after_bootmem; diff --git a/lib/show_mem.c b/lib/show_mem.c index fdc77c82f92..d8d602b58c3 100644 --- a/lib/show_mem.c +++ b/lib/show_mem.c @@ -9,14 +9,14 @@ #include #include -void show_mem(void) +void __show_mem(unsigned int filter) { pg_data_t *pgdat; unsigned long total = 0, reserved = 0, shared = 0, nonshared = 0, highmem = 0; printk("Mem-Info:\n"); - show_free_areas(); + __show_free_areas(filter); for_each_online_pgdat(pgdat) { unsigned long i, flags; @@ -61,3 +61,8 @@ void show_mem(void) quicklist_total_size()); #endif } + +void show_mem(void) +{ + __show_mem(0); +} diff --git a/mm/oom_kill.c b/mm/oom_kill.c index 33b58615072..3100bc57036 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -406,7 +406,7 @@ static void dump_header(struct task_struct *p, gfp_t gfp_mask, int order, task_unlock(current); dump_stack(); mem_cgroup_print_oom_info(mem, p); - show_mem(); + __show_mem(SHOW_MEM_FILTER_NODES); if (sysctl_oom_dump_tasks) dump_tasks(mem, nodemask); } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 7945247b1e5..36be3ba4bbe 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2411,19 +2411,42 @@ void si_meminfo_node(struct sysinfo *val, int nid) } #endif +/* + * Determine whether the zone's node should be displayed or not, depending on + * whether SHOW_MEM_FILTER_NODES was passed to __show_free_areas(). + */ +static bool skip_free_areas_zone(unsigned int flags, const struct zone *zone) +{ + bool ret = false; + + if (!(flags & SHOW_MEM_FILTER_NODES)) + goto out; + + get_mems_allowed(); + ret = !node_isset(zone->zone_pgdat->node_id, + cpuset_current_mems_allowed); + put_mems_allowed(); +out: + return ret; +} + #define K(x) ((x) << (PAGE_SHIFT-10)) /* * Show free area list (used inside shift_scroll-lock stuff) * We also calculate the percentage fragmentation. We do this by counting the * memory on each free list with the exception of the first item on the list. + * Suppresses nodes that are not allowed by current's cpuset if + * SHOW_MEM_FILTER_NODES is passed. */ -void show_free_areas(void) +void __show_free_areas(unsigned int filter) { int cpu; struct zone *zone; for_each_populated_zone(zone) { + if (skip_free_areas_zone(filter, zone)) + continue; show_node(zone); printk("%s per-cpu:\n", zone->name); @@ -2465,6 +2488,8 @@ void show_free_areas(void) for_each_populated_zone(zone) { int i; + if (skip_free_areas_zone(filter, zone)) + continue; show_node(zone); printk("%s" " free:%lukB" @@ -2532,6 +2557,8 @@ void show_free_areas(void) for_each_populated_zone(zone) { unsigned long nr[MAX_ORDER], flags, order, total = 0; + if (skip_free_areas_zone(filter, zone)) + continue; show_node(zone); printk("%s: ", zone->name); @@ -2551,6 +2578,11 @@ void show_free_areas(void) show_swap_cache_info(); } +void show_free_areas(void) +{ + __show_free_areas(0); +} + static void zoneref_set_zone(struct zone *zone, struct zoneref *zoneref) { zoneref->zone = zone; -- cgit v1.2.3-70-g09d2 From fef2c9bc1b54c0261324a96e948c0b849796e896 Mon Sep 17 00:00:00 2001 From: Don Zickus Date: Tue, 22 Mar 2011 16:34:16 -0700 Subject: kernel/watchdog.c: allow hardlockup to panic by default When a cpu is considered stuck, instead of limping along and just printing a warning, it is sometimes preferred to just panic, let kdump capture the vmcore and reboot. This gets the machine back into a stable state quickly while saving the info that got it into a stuck state to begin with. Add a Kconfig option to allow users to set the hardlockup to panic by default. Also add in a 'nmi_watchdog=nopanic' to override this. [akpm@linux-foundation.org: fix strncmp length] Signed-off-by: Don Zickus Acked-by: Peter Zijlstra Reviewed-by: WANG Cong Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 5 +++-- kernel/watchdog.c | 5 ++++- lib/Kconfig.debug | 17 +++++++++++++++++ 3 files changed, 24 insertions(+), 3 deletions(-) (limited to 'lib') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index a3b688287a8..e9261e938f6 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1597,11 +1597,12 @@ bytes respectively. Such letter suffixes can also be entirely omitted. Format: [state][,regs][,debounce][,die] nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels - Format: [panic,][num] + Format: [panic,][nopanic,][num] Valid num: 0 0 - turn nmi_watchdog off When panic is specified, panic when an NMI watchdog - timeout occurs. + timeout occurs (or 'nopanic' to override the opposite + default). This is useful when you use a panic=... timeout and need the box quickly up again. diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 18bb15776c5..054a67cca9d 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -48,12 +48,15 @@ static DEFINE_PER_CPU(struct perf_event *, watchdog_ev); * Should we panic when a soft-lockup or hard-lockup occurs: */ #ifdef CONFIG_HARDLOCKUP_DETECTOR -static int hardlockup_panic; +static int hardlockup_panic = + CONFIG_BOOTPARAM_HARDLOCKUP_PANIC_VALUE; static int __init hardlockup_panic_setup(char *str) { if (!strncmp(str, "panic", 5)) hardlockup_panic = 1; + else if (!strncmp(str, "nopanic", 7)) + hardlockup_panic = 0; else if (!strncmp(str, "0", 1)) watchdog_enabled = 0; return 1; diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 191c5c4c89f..fb0afeff943 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -171,6 +171,23 @@ config HARDLOCKUP_DETECTOR def_bool LOCKUP_DETECTOR && PERF_EVENTS && HAVE_PERF_EVENTS_NMI && \ !ARCH_HAS_NMI_WATCHDOG +config BOOTPARAM_HARDLOCKUP_PANIC + bool "Panic (Reboot) On Hard Lockups" + depends on LOCKUP_DETECTOR + help + Say Y here to enable the kernel to panic on "hard lockups", + which are bugs that cause the kernel to loop in kernel + mode with interrupts disabled for more than 60 seconds. + + Say N if unsure. + +config BOOTPARAM_HARDLOCKUP_PANIC_VALUE + int + depends on LOCKUP_DETECTOR + range 0 1 + default 0 if !BOOTPARAM_HARDLOCKUP_PANIC + default 1 if BOOTPARAM_HARDLOCKUP_PANIC + config BOOTPARAM_SOFTLOCKUP_PANIC bool "Panic (Reboot) On Soft Lockups" depends on LOCKUP_DETECTOR -- cgit v1.2.3-70-g09d2 From 26297607e09ca6c7e6f2a6b86a8bee2f23503bb8 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Tue, 22 Mar 2011 16:34:19 -0700 Subject: vsprintf: neaten %pK kptr_restrict, save a bit of code space If kptr restrictions are on, just set the passed pointer to NULL. $ size lib/vsprintf.o.* text data bss dec hex filename 8247 4 2 8253 203d lib/vsprintf.o.new 8282 4 2 8288 2060 lib/vsprintf.o.old Signed-off-by: Joe Perches Cc: Dan Rosenberg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- lib/vsprintf.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) (limited to 'lib') diff --git a/lib/vsprintf.c b/lib/vsprintf.c index d3023df8477..070d134eef7 100644 --- a/lib/vsprintf.c +++ b/lib/vsprintf.c @@ -1047,16 +1047,12 @@ char *pointer(const char *fmt, char *buf, char *end, void *ptr, if (spec.field_width == -1) spec.field_width = 2 * sizeof(void *); return string(buf, end, "pK-error", spec); - } else if ((kptr_restrict == 0) || - (kptr_restrict == 1 && - has_capability_noaudit(current, CAP_SYSLOG))) - break; - - if (spec.field_width == -1) { - spec.field_width = 2 * sizeof(void *); - spec.flags |= ZEROPAD; } - return number(buf, end, 0, spec); + if (!((kptr_restrict == 0) || + (kptr_restrict == 1 && + has_capability_noaudit(current, CAP_SYSLOG)))) + ptr = NULL; + break; } spec.flags |= SMALL; if (spec.field_width == -1) { -- cgit v1.2.3-70-g09d2 From 9f36e2c448007b54851e7e4fa48da97d1477a175 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Tue, 22 Mar 2011 16:34:22 -0700 Subject: printk: use %pK for /proc/kallsyms and /proc/modules In an effort to reduce kernel address leaks that might be used to help target kernel privilege escalation exploits, this patch uses %pK when displaying addresses in /proc/kallsyms, /proc/modules, and /sys/module/*/sections/*. Note that this changes %x to %p, so some legitimately 0 values in /proc/kallsyms would have changed from 00000000 to "(null)". To avoid this, "(null)" is not used when using the "K" format. Anything that was already successfully parsing "(null)" in addition to full hex digits should have no problem with this change. (Thanks to Joe Perches for the suggestion.) Due to the %x to %p, "void *" casts are needed since these addresses are already "unsigned long" everywhere internally, due to their starting life as ELF section offsets. Signed-off-by: Kees Cook Cc: Eugene Teo Cc: Dan Rosenberg Cc: Rusty Russell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/kallsyms.c | 10 ++++------ kernel/module.c | 4 ++-- lib/vsprintf.c | 2 +- 3 files changed, 7 insertions(+), 9 deletions(-) (limited to 'lib') diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c index 6f6d091b575..75dcca37d61 100644 --- a/kernel/kallsyms.c +++ b/kernel/kallsyms.c @@ -477,13 +477,11 @@ static int s_show(struct seq_file *m, void *p) */ type = iter->exported ? toupper(iter->type) : tolower(iter->type); - seq_printf(m, "%0*lx %c %s\t[%s]\n", - (int)(2 * sizeof(void *)), - iter->value, type, iter->name, iter->module_name); + seq_printf(m, "%pK %c %s\t[%s]\n", (void *)iter->value, + type, iter->name, iter->module_name); } else - seq_printf(m, "%0*lx %c %s\n", - (int)(2 * sizeof(void *)), - iter->value, iter->type, iter->name); + seq_printf(m, "%pK %c %s\n", (void *)iter->value, + iter->type, iter->name); return 0; } diff --git a/kernel/module.c b/kernel/module.c index efa290ea94b..1f9f7bc56ca 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1168,7 +1168,7 @@ static ssize_t module_sect_show(struct module_attribute *mattr, { struct module_sect_attr *sattr = container_of(mattr, struct module_sect_attr, mattr); - return sprintf(buf, "0x%lx\n", sattr->address); + return sprintf(buf, "0x%pK\n", (void *)sattr->address); } static void free_sect_attrs(struct module_sect_attrs *sect_attrs) @@ -3224,7 +3224,7 @@ static int m_show(struct seq_file *m, void *p) mod->state == MODULE_STATE_COMING ? "Loading": "Live"); /* Used by oprofile and other similar tools. */ - seq_printf(m, " 0x%p", mod->module_core); + seq_printf(m, " 0x%pK", mod->module_core); /* Taints info */ if (mod->taints) diff --git a/lib/vsprintf.c b/lib/vsprintf.c index 070d134eef7..ac444ff0165 100644 --- a/lib/vsprintf.c +++ b/lib/vsprintf.c @@ -991,7 +991,7 @@ static noinline_for_stack char *pointer(const char *fmt, char *buf, char *end, void *ptr, struct printf_spec spec) { - if (!ptr) { + if (!ptr && *fmt != 'K') { /* * Print (null) with the same width as a pointer so it makes * tabular output look nice. -- cgit v1.2.3-70-g09d2 From 5af5bcb8d37f99ba415a1adc6da71051b84f93a5 Mon Sep 17 00:00:00 2001 From: Mandeep Singh Baines Date: Tue, 22 Mar 2011 16:34:23 -0700 Subject: printk: allow setting DEFAULT_MESSAGE_LEVEL via Kconfig We've been burned by regressions/bugs which we later realized could have been triaged quicker if only we'd paid closer attention to dmesg. To make it easier to audit dmesg, we'd like to make DEFAULT_MESSAGE_LEVEL Kconfig-settable. That way we can set it to KERN_NOTICE and audit any messages <= KERN_WARNING. Signed-off-by: Mandeep Singh Baines Cc: Ingo Molnar Cc: Joe Perches Cc: Olof Johansson Cc: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/printk.c | 2 +- lib/Kconfig.debug | 11 +++++++++++ 2 files changed, 12 insertions(+), 1 deletion(-) (limited to 'lib') diff --git a/kernel/printk.c b/kernel/printk.c index a53607eea6d..da8ca817eae 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -53,7 +53,7 @@ void asmlinkage __attribute__((weak)) early_printk(const char *fmt, ...) #define __LOG_BUF_LEN (1 << CONFIG_LOG_BUF_SHIFT) /* printk's without a loglevel use this.. */ -#define DEFAULT_MESSAGE_LOGLEVEL 4 /* KERN_WARNING */ +#define DEFAULT_MESSAGE_LOGLEVEL CONFIG_DEFAULT_MESSAGE_LOGLEVEL /* We show everything that is MORE important than this.. */ #define MINIMUM_CONSOLE_LOGLEVEL 1 /* Minimum loglevel we let people use */ diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index fb0afeff943..bef5faab660 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -9,6 +9,17 @@ config PRINTK_TIME operations. This is useful for identifying long delays in kernel startup. +config DEFAULT_MESSAGE_LOGLEVEL + int "Default message log level (1-7)" + range 1 7 + default "4" + help + Default log level for printk statements with no specified priority. + + This was hard-coded to KERN_WARNING since at least 2.6.10 but folks + that are auditing their logs closely may want to set it to a lower + priority. + config ENABLE_WARN_DEPRECATED bool "Enable __deprecated logic" default y -- cgit v1.2.3-70-g09d2 From 33ee3b2e2eb9b4b6c64dcf9ed66e2ac3124e748c Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 22 Mar 2011 16:34:40 -0700 Subject: kstrto*: converting strings to integers done (hopefully) right 1. simple_strto*() do not contain overflow checks and crufty, libc way to indicate failure. 2. strict_strto*() also do not have overflow checks but the name and comments pretend they do. 3. Both families have only "long long" and "long" variants, but users want strtou8() 4. Both "simple" and "strict" prefixes are wrong: Simple doesn't exactly say what's so simple, strict should not exist because conversion should be strict by default. The solution is to use "k" prefix and add convertors for more types. Enter kstrtoull() kstrtoll() kstrtoul() kstrtol() kstrtouint() kstrtoint() kstrtou64() kstrtos64() kstrtou32() kstrtos32() kstrtou16() kstrtos16() kstrtou8() kstrtos8() Include runtime testsuite (somewhat incomplete) as well. strict_strto*() become deprecated, stubbed to kstrto*() and eventually will be removed altogether. Use kstrto*() in code today! Note: on some archs _kstrtoul() and _kstrtol() are left in tree, even if they'll be unused at runtime. This is temporarily solution, because I don't want to hardcode list of archs where these functions aren't needed. Current solution with sizeof() and __alignof__ at least always works. Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/via/viafbdev.h | 3 - include/linux/kernel.h | 70 +++- lib/Kconfig.debug | 3 + lib/Makefile | 2 + lib/kstrtox.c | 227 +++++++++++++ lib/test-kstrtox.c | 739 +++++++++++++++++++++++++++++++++++++++++++ lib/vsprintf.c | 141 --------- scripts/checkpatch.pl | 4 +- 8 files changed, 1039 insertions(+), 150 deletions(-) create mode 100644 lib/kstrtox.c create mode 100644 lib/test-kstrtox.c (limited to 'lib') diff --git a/drivers/video/via/viafbdev.h b/drivers/video/via/viafbdev.h index d66f963e930..137996dc547 100644 --- a/drivers/video/via/viafbdev.h +++ b/drivers/video/via/viafbdev.h @@ -94,9 +94,6 @@ extern int viafb_LCD_ON; extern int viafb_DVI_ON; extern int viafb_hotplug; -extern int strict_strtoul(const char *cp, unsigned int base, - unsigned long *res); - u8 viafb_gpio_i2c_read_lvds(struct lvds_setting_information *plvds_setting_info, struct lvds_chip_information *plvds_chip_info, u8 index); diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 2fe6e84894a..00cec4dc0ae 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -187,14 +187,76 @@ NORET_TYPE void do_exit(long error_code) ATTRIB_NORET; NORET_TYPE void complete_and_exit(struct completion *, long) ATTRIB_NORET; + +/* Internal, do not use. */ +int __must_check _kstrtoul(const char *s, unsigned int base, unsigned long *res); +int __must_check _kstrtol(const char *s, unsigned int base, long *res); + +int __must_check kstrtoull(const char *s, unsigned int base, unsigned long long *res); +int __must_check kstrtoll(const char *s, unsigned int base, long long *res); +static inline int __must_check kstrtoul(const char *s, unsigned int base, unsigned long *res) +{ + /* + * We want to shortcut function call, but + * __builtin_types_compatible_p(unsigned long, unsigned long long) = 0. + */ + if (sizeof(unsigned long) == sizeof(unsigned long long) && + __alignof__(unsigned long) == __alignof__(unsigned long long)) + return kstrtoull(s, base, (unsigned long long *)res); + else + return _kstrtoul(s, base, res); +} + +static inline int __must_check kstrtol(const char *s, unsigned int base, long *res) +{ + /* + * We want to shortcut function call, but + * __builtin_types_compatible_p(long, long long) = 0. + */ + if (sizeof(long) == sizeof(long long) && + __alignof__(long) == __alignof__(long long)) + return kstrtoll(s, base, (long long *)res); + else + return _kstrtol(s, base, res); +} + +int __must_check kstrtouint(const char *s, unsigned int base, unsigned int *res); +int __must_check kstrtoint(const char *s, unsigned int base, int *res); + +static inline int __must_check kstrtou64(const char *s, unsigned int base, u64 *res) +{ + return kstrtoull(s, base, res); +} + +static inline int __must_check kstrtos64(const char *s, unsigned int base, s64 *res) +{ + return kstrtoll(s, base, res); +} + +static inline int __must_check kstrtou32(const char *s, unsigned int base, u32 *res) +{ + return kstrtouint(s, base, res); +} + +static inline int __must_check kstrtos32(const char *s, unsigned int base, s32 *res) +{ + return kstrtoint(s, base, res); +} + +int __must_check kstrtou16(const char *s, unsigned int base, u16 *res); +int __must_check kstrtos16(const char *s, unsigned int base, s16 *res); +int __must_check kstrtou8(const char *s, unsigned int base, u8 *res); +int __must_check kstrtos8(const char *s, unsigned int base, s8 *res); + extern unsigned long simple_strtoul(const char *,char **,unsigned int); extern long simple_strtol(const char *,char **,unsigned int); extern unsigned long long simple_strtoull(const char *,char **,unsigned int); extern long long simple_strtoll(const char *,char **,unsigned int); -extern int __must_check strict_strtoul(const char *, unsigned int, unsigned long *); -extern int __must_check strict_strtol(const char *, unsigned int, long *); -extern int __must_check strict_strtoull(const char *, unsigned int, unsigned long long *); -extern int __must_check strict_strtoll(const char *, unsigned int, long long *); +#define strict_strtoul kstrtoul +#define strict_strtol kstrtol +#define strict_strtoull kstrtoull +#define strict_strtoll kstrtoll + extern int sprintf(char * buf, const char * fmt, ...) __attribute__ ((format (printf, 2, 3))); extern int vsprintf(char *buf, const char *, va_list) diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index bef5faab660..df9234c5f9d 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1250,3 +1250,6 @@ source "samples/Kconfig" source "lib/Kconfig.kgdb" source "lib/Kconfig.kmemcheck" + +config TEST_KSTRTOX + tristate "Test kstrto*() family of functions at runtime" diff --git a/lib/Makefile b/lib/Makefile index ef7ed71a6ff..8c9de027ebb 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -22,6 +22,8 @@ lib-y += kobject.o kref.o klist.o obj-y += bcd.o div64.o sort.o parser.o halfmd4.o debug_locks.o random32.o \ bust_spinlocks.o hexdump.o kasprintf.o bitmap.o scatterlist.o \ string_helpers.o gcd.o lcm.o list_sort.o uuid.o flex_array.o +obj-y += kstrtox.o +obj-$(CONFIG_TEST_KSTRTOX) += test-kstrtox.o ifeq ($(CONFIG_DEBUG_KOBJECT),y) CFLAGS_kobject.o += -DDEBUG diff --git a/lib/kstrtox.c b/lib/kstrtox.c new file mode 100644 index 00000000000..05672e819f8 --- /dev/null +++ b/lib/kstrtox.c @@ -0,0 +1,227 @@ +/* + * Convert integer string representation to an integer. + * If an integer doesn't fit into specified type, -E is returned. + * + * Integer starts with optional sign. + * kstrtou*() functions do not accept sign "-". + * + * Radix 0 means autodetection: leading "0x" implies radix 16, + * leading "0" implies radix 8, otherwise radix is 10. + * Autodetection hints work after optional sign, but not before. + * + * If -E is returned, result is not touched. + */ +#include +#include +#include +#include +#include +#include + +static inline char _tolower(const char c) +{ + return c | 0x20; +} + +static int _kstrtoull(const char *s, unsigned int base, unsigned long long *res) +{ + unsigned long long acc; + int ok; + + if (base == 0) { + if (s[0] == '0') { + if (_tolower(s[1]) == 'x' && isxdigit(s[2])) + base = 16; + else + base = 8; + } else + base = 10; + } + if (base == 16 && s[0] == '0' && _tolower(s[1]) == 'x') + s += 2; + + acc = 0; + ok = 0; + while (*s) { + unsigned int val; + + if ('0' <= *s && *s <= '9') + val = *s - '0'; + else if ('a' <= _tolower(*s) && _tolower(*s) <= 'f') + val = _tolower(*s) - 'a' + 10; + else if (*s == '\n') { + if (*(s + 1) == '\0') + break; + else + return -EINVAL; + } else + return -EINVAL; + + if (val >= base) + return -EINVAL; + if (acc > div_u64(ULLONG_MAX - val, base)) + return -ERANGE; + acc = acc * base + val; + ok = 1; + + s++; + } + if (!ok) + return -EINVAL; + *res = acc; + return 0; +} + +int kstrtoull(const char *s, unsigned int base, unsigned long long *res) +{ + if (s[0] == '+') + s++; + return _kstrtoull(s, base, res); +} +EXPORT_SYMBOL(kstrtoull); + +int kstrtoll(const char *s, unsigned int base, long long *res) +{ + unsigned long long tmp; + int rv; + + if (s[0] == '-') { + rv = _kstrtoull(s + 1, base, &tmp); + if (rv < 0) + return rv; + if ((long long)(-tmp) >= 0) + return -ERANGE; + *res = -tmp; + } else { + rv = kstrtoull(s, base, &tmp); + if (rv < 0) + return rv; + if ((long long)tmp < 0) + return -ERANGE; + *res = tmp; + } + return 0; +} +EXPORT_SYMBOL(kstrtoll); + +/* Internal, do not use. */ +int _kstrtoul(const char *s, unsigned int base, unsigned long *res) +{ + unsigned long long tmp; + int rv; + + rv = kstrtoull(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (unsigned long long)(unsigned long)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(_kstrtoul); + +/* Internal, do not use. */ +int _kstrtol(const char *s, unsigned int base, long *res) +{ + long long tmp; + int rv; + + rv = kstrtoll(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (long long)(long)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(_kstrtol); + +int kstrtouint(const char *s, unsigned int base, unsigned int *res) +{ + unsigned long long tmp; + int rv; + + rv = kstrtoull(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (unsigned long long)(unsigned int)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(kstrtouint); + +int kstrtoint(const char *s, unsigned int base, int *res) +{ + long long tmp; + int rv; + + rv = kstrtoll(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (long long)(int)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(kstrtoint); + +int kstrtou16(const char *s, unsigned int base, u16 *res) +{ + unsigned long long tmp; + int rv; + + rv = kstrtoull(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (unsigned long long)(u16)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(kstrtou16); + +int kstrtos16(const char *s, unsigned int base, s16 *res) +{ + long long tmp; + int rv; + + rv = kstrtoll(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (long long)(s16)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(kstrtos16); + +int kstrtou8(const char *s, unsigned int base, u8 *res) +{ + unsigned long long tmp; + int rv; + + rv = kstrtoull(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (unsigned long long)(u8)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(kstrtou8); + +int kstrtos8(const char *s, unsigned int base, s8 *res) +{ + long long tmp; + int rv; + + rv = kstrtoll(s, base, &tmp); + if (rv < 0) + return rv; + if (tmp != (long long)(s8)tmp) + return -ERANGE; + *res = tmp; + return 0; +} +EXPORT_SYMBOL(kstrtos8); diff --git a/lib/test-kstrtox.c b/lib/test-kstrtox.c new file mode 100644 index 00000000000..325c2f9eceb --- /dev/null +++ b/lib/test-kstrtox.c @@ -0,0 +1,739 @@ +#include +#include +#include + +#define for_each_test(i, test) \ + for (i = 0; i < sizeof(test) / sizeof(test[0]); i++) + +struct test_fail { + const char *str; + unsigned int base; +}; + +#define DEFINE_TEST_FAIL(test) \ + const struct test_fail test[] __initdata + +#define DECLARE_TEST_OK(type, test_type) \ + test_type { \ + const char *str; \ + unsigned int base; \ + type expected_res; \ + } + +#define DEFINE_TEST_OK(type, test) \ + const type test[] __initdata + +#define TEST_FAIL(fn, type, fmt, test) \ +{ \ + unsigned int i; \ + \ + for_each_test(i, test) { \ + const struct test_fail *t = &test[i]; \ + type tmp; \ + int rv; \ + \ + tmp = 0; \ + rv = fn(t->str, t->base, &tmp); \ + if (rv >= 0) { \ + WARN(1, "str '%s', base %u, expected -E, got %d/" fmt "\n", \ + t->str, t->base, rv, tmp); \ + continue; \ + } \ + } \ +} + +#define TEST_OK(fn, type, fmt, test) \ +{ \ + unsigned int i; \ + \ + for_each_test(i, test) { \ + const typeof(test[0]) *t = &test[i]; \ + type res; \ + int rv; \ + \ + rv = fn(t->str, t->base, &res); \ + if (rv != 0) { \ + WARN(1, "str '%s', base %u, expected 0/" fmt ", got %d\n", \ + t->str, t->base, t->expected_res, rv); \ + continue; \ + } \ + if (res != t->expected_res) { \ + WARN(1, "str '%s', base %u, expected " fmt ", got " fmt "\n", \ + t->str, t->base, t->expected_res, res); \ + continue; \ + } \ + } \ +} + +static void __init test_kstrtoull_ok(void) +{ + DECLARE_TEST_OK(unsigned long long, struct test_ull); + static DEFINE_TEST_OK(struct test_ull, test_ull_ok) = { + {"0", 10, 0ULL}, + {"1", 10, 1ULL}, + {"127", 10, 127ULL}, + {"128", 10, 128ULL}, + {"129", 10, 129ULL}, + {"255", 10, 255ULL}, + {"256", 10, 256ULL}, + {"257", 10, 257ULL}, + {"32767", 10, 32767ULL}, + {"32768", 10, 32768ULL}, + {"32769", 10, 32769ULL}, + {"65535", 10, 65535ULL}, + {"65536", 10, 65536ULL}, + {"65537", 10, 65537ULL}, + {"2147483647", 10, 2147483647ULL}, + {"2147483648", 10, 2147483648ULL}, + {"2147483649", 10, 2147483649ULL}, + {"4294967295", 10, 4294967295ULL}, + {"4294967296", 10, 4294967296ULL}, + {"4294967297", 10, 4294967297ULL}, + {"9223372036854775807", 10, 9223372036854775807ULL}, + {"9223372036854775808", 10, 9223372036854775808ULL}, + {"9223372036854775809", 10, 9223372036854775809ULL}, + {"18446744073709551614", 10, 18446744073709551614ULL}, + {"18446744073709551615", 10, 18446744073709551615ULL}, + + {"00", 8, 00ULL}, + {"01", 8, 01ULL}, + {"0177", 8, 0177ULL}, + {"0200", 8, 0200ULL}, + {"0201", 8, 0201ULL}, + {"0377", 8, 0377ULL}, + {"0400", 8, 0400ULL}, + {"0401", 8, 0401ULL}, + {"077777", 8, 077777ULL}, + {"0100000", 8, 0100000ULL}, + {"0100001", 8, 0100001ULL}, + {"0177777", 8, 0177777ULL}, + {"0200000", 8, 0200000ULL}, + {"0200001", 8, 0200001ULL}, + {"017777777777", 8, 017777777777ULL}, + {"020000000000", 8, 020000000000ULL}, + {"020000000001", 8, 020000000001ULL}, + {"037777777777", 8, 037777777777ULL}, + {"040000000000", 8, 040000000000ULL}, + {"040000000001", 8, 040000000001ULL}, + {"0777777777777777777777", 8, 0777777777777777777777ULL}, + {"01000000000000000000000", 8, 01000000000000000000000ULL}, + {"01000000000000000000001", 8, 01000000000000000000001ULL}, + {"01777777777777777777776", 8, 01777777777777777777776ULL}, + {"01777777777777777777777", 8, 01777777777777777777777ULL}, + + {"0x0", 16, 0x0ULL}, + {"0x1", 16, 0x1ULL}, + {"0x7f", 16, 0x7fULL}, + {"0x80", 16, 0x80ULL}, + {"0x81", 16, 0x81ULL}, + {"0xff", 16, 0xffULL}, + {"0x100", 16, 0x100ULL}, + {"0x101", 16, 0x101ULL}, + {"0x7fff", 16, 0x7fffULL}, + {"0x8000", 16, 0x8000ULL}, + {"0x8001", 16, 0x8001ULL}, + {"0xffff", 16, 0xffffULL}, + {"0x10000", 16, 0x10000ULL}, + {"0x10001", 16, 0x10001ULL}, + {"0x7fffffff", 16, 0x7fffffffULL}, + {"0x80000000", 16, 0x80000000ULL}, + {"0x80000001", 16, 0x80000001ULL}, + {"0xffffffff", 16, 0xffffffffULL}, + {"0x100000000", 16, 0x100000000ULL}, + {"0x100000001", 16, 0x100000001ULL}, + {"0x7fffffffffffffff", 16, 0x7fffffffffffffffULL}, + {"0x8000000000000000", 16, 0x8000000000000000ULL}, + {"0x8000000000000001", 16, 0x8000000000000001ULL}, + {"0xfffffffffffffffe", 16, 0xfffffffffffffffeULL}, + {"0xffffffffffffffff", 16, 0xffffffffffffffffULL}, + + {"0\n", 0, 0ULL}, + }; + TEST_OK(kstrtoull, unsigned long long, "%llu", test_ull_ok); +} + +static void __init test_kstrtoull_fail(void) +{ + static DEFINE_TEST_FAIL(test_ull_fail) = { + {"", 0}, + {"", 8}, + {"", 10}, + {"", 16}, + {"\n", 0}, + {"\n", 8}, + {"\n", 10}, + {"\n", 16}, + {"\n0", 0}, + {"\n0", 8}, + {"\n0", 10}, + {"\n0", 16}, + {"+", 0}, + {"+", 8}, + {"+", 10}, + {"+", 16}, + {"-", 0}, + {"-", 8}, + {"-", 10}, + {"-", 16}, + {"0x", 0}, + {"0x", 16}, + {"0X", 0}, + {"0X", 16}, + {"0 ", 0}, + {"1+", 0}, + {"1-", 0}, + {" 2", 0}, + /* base autodetection */ + {"0x0z", 0}, + {"0z", 0}, + {"a", 0}, + /* digit >= base */ + {"2", 2}, + {"8", 8}, + {"a", 10}, + {"A", 10}, + {"g", 16}, + {"G", 16}, + /* overflow */ + {"10000000000000000000000000000000000000000000000000000000000000000", 2}, + {"2000000000000000000000", 8}, + {"18446744073709551616", 10}, + {"10000000000000000", 16}, + /* negative */ + {"-0", 0}, + {"-0", 8}, + {"-0", 10}, + {"-0", 16}, + {"-1", 0}, + {"-1", 8}, + {"-1", 10}, + {"-1", 16}, + /* sign is first character if any */ + {"-+1", 0}, + {"-+1", 8}, + {"-+1", 10}, + {"-+1", 16}, + /* nothing after \n */ + {"0\n0", 0}, + {"0\n0", 8}, + {"0\n0", 10}, + {"0\n0", 16}, + {"0\n+", 0}, + {"0\n+", 8}, + {"0\n+", 10}, + {"0\n+", 16}, + {"0\n-", 0}, + {"0\n-", 8}, + {"0\n-", 10}, + {"0\n-", 16}, + {"0\n ", 0}, + {"0\n ", 8}, + {"0\n ", 10}, + {"0\n ", 16}, + }; + TEST_FAIL(kstrtoull, unsigned long long, "%llu", test_ull_fail); +} + +static void __init test_kstrtoll_ok(void) +{ + DECLARE_TEST_OK(long long, struct test_ll); + static DEFINE_TEST_OK(struct test_ll, test_ll_ok) = { + {"0", 10, 0LL}, + {"1", 10, 1LL}, + {"127", 10, 127LL}, + {"128", 10, 128LL}, + {"129", 10, 129LL}, + {"255", 10, 255LL}, + {"256", 10, 256LL}, + {"257", 10, 257LL}, + {"32767", 10, 32767LL}, + {"32768", 10, 32768LL}, + {"32769", 10, 32769LL}, + {"65535", 10, 65535LL}, + {"65536", 10, 65536LL}, + {"65537", 10, 65537LL}, + {"2147483647", 10, 2147483647LL}, + {"2147483648", 10, 2147483648LL}, + {"2147483649", 10, 2147483649LL}, + {"4294967295", 10, 4294967295LL}, + {"4294967296", 10, 4294967296LL}, + {"4294967297", 10, 4294967297LL}, + {"9223372036854775807", 10, 9223372036854775807LL}, + + {"-1", 10, -1LL}, + {"-2", 10, -2LL}, + {"-9223372036854775808", 10, LLONG_MIN}, + }; + TEST_OK(kstrtoll, long long, "%lld", test_ll_ok); +} + +static void __init test_kstrtoll_fail(void) +{ + static DEFINE_TEST_FAIL(test_ll_fail) = { + {"9223372036854775808", 10}, + {"9223372036854775809", 10}, + {"18446744073709551614", 10}, + {"18446744073709551615", 10}, + {"-9223372036854775809", 10}, + {"-18446744073709551614", 10}, + {"-18446744073709551615", 10}, + /* negative zero isn't an integer in Linux */ + {"-0", 0}, + {"-0", 8}, + {"-0", 10}, + {"-0", 16}, + /* sign is first character if any */ + {"-+1", 0}, + {"-+1", 8}, + {"-+1", 10}, + {"-+1", 16}, + }; + TEST_FAIL(kstrtoll, long long, "%lld", test_ll_fail); +} + +static void __init test_kstrtou64_ok(void) +{ + DECLARE_TEST_OK(u64, struct test_u64); + static DEFINE_TEST_OK(struct test_u64, test_u64_ok) = { + {"0", 10, 0}, + {"1", 10, 1}, + {"126", 10, 126}, + {"127", 10, 127}, + {"128", 10, 128}, + {"129", 10, 129}, + {"254", 10, 254}, + {"255", 10, 255}, + {"256", 10, 256}, + {"257", 10, 257}, + {"32766", 10, 32766}, + {"32767", 10, 32767}, + {"32768", 10, 32768}, + {"32769", 10, 32769}, + {"65534", 10, 65534}, + {"65535", 10, 65535}, + {"65536", 10, 65536}, + {"65537", 10, 65537}, + {"2147483646", 10, 2147483646}, + {"2147483647", 10, 2147483647}, + {"2147483648", 10, 2147483648}, + {"2147483649", 10, 2147483649}, + {"4294967294", 10, 4294967294}, + {"4294967295", 10, 4294967295}, + {"4294967296", 10, 4294967296}, + {"4294967297", 10, 4294967297}, + {"9223372036854775806", 10, 9223372036854775806ULL}, + {"9223372036854775807", 10, 9223372036854775807ULL}, + {"9223372036854775808", 10, 9223372036854775808ULL}, + {"9223372036854775809", 10, 9223372036854775809ULL}, + {"18446744073709551614", 10, 18446744073709551614ULL}, + {"18446744073709551615", 10, 18446744073709551615ULL}, + }; + TEST_OK(kstrtou64, u64, "%llu", test_u64_ok); +} + +static void __init test_kstrtou64_fail(void) +{ + static DEFINE_TEST_FAIL(test_u64_fail) = { + {"-2", 10}, + {"-1", 10}, + {"18446744073709551616", 10}, + {"18446744073709551617", 10}, + }; + TEST_FAIL(kstrtou64, u64, "%llu", test_u64_fail); +} + +static void __init test_kstrtos64_ok(void) +{ + DECLARE_TEST_OK(s64, struct test_s64); + static DEFINE_TEST_OK(struct test_s64, test_s64_ok) = { + {"-128", 10, -128}, + {"-127", 10, -127}, + {"-1", 10, -1}, + {"0", 10, 0}, + {"1", 10, 1}, + {"126", 10, 126}, + {"127", 10, 127}, + {"128", 10, 128}, + {"129", 10, 129}, + {"254", 10, 254}, + {"255", 10, 255}, + {"256", 10, 256}, + {"257", 10, 257}, + {"32766", 10, 32766}, + {"32767", 10, 32767}, + {"32768", 10, 32768}, + {"32769", 10, 32769}, + {"65534", 10, 65534}, + {"65535", 10, 65535}, + {"65536", 10, 65536}, + {"65537", 10, 65537}, + {"2147483646", 10, 2147483646}, + {"2147483647", 10, 2147483647}, + {"2147483648", 10, 2147483648}, + {"2147483649", 10, 2147483649}, + {"4294967294", 10, 4294967294}, + {"4294967295", 10, 4294967295}, + {"4294967296", 10, 4294967296}, + {"4294967297", 10, 4294967297}, + {"9223372036854775806", 10, 9223372036854775806LL}, + {"9223372036854775807", 10, 9223372036854775807LL}, + }; + TEST_OK(kstrtos64, s64, "%lld", test_s64_ok); +} + +static void __init test_kstrtos64_fail(void) +{ + static DEFINE_TEST_FAIL(test_s64_fail) = { + {"9223372036854775808", 10}, + {"9223372036854775809", 10}, + {"18446744073709551614", 10}, + {"18446744073709551615", 10}, + {"18446744073709551616", 10}, + {"18446744073709551617", 10}, + }; + TEST_FAIL(kstrtos64, s64, "%lld", test_s64_fail); +} + +static void __init test_kstrtou32_ok(void) +{ + DECLARE_TEST_OK(u32, struct test_u32); + static DEFINE_TEST_OK(struct test_u32, test_u32_ok) = { + {"0", 10, 0}, + {"1", 10, 1}, + {"126", 10, 126}, + {"127", 10, 127}, + {"128", 10, 128}, + {"129", 10, 129}, + {"254", 10, 254}, + {"255", 10, 255}, + {"256", 10, 256}, + {"257", 10, 257}, + {"32766", 10, 32766}, + {"32767", 10, 32767}, + {"32768", 10, 32768}, + {"32769", 10, 32769}, + {"65534", 10, 65534}, + {"65535", 10, 65535}, + {"65536", 10, 65536}, + {"65537", 10, 65537}, + {"2147483646", 10, 2147483646}, + {"2147483647", 10, 2147483647}, + {"2147483648", 10, 2147483648}, + {"2147483649", 10, 2147483649}, + {"4294967294", 10, 4294967294}, + {"4294967295", 10, 4294967295}, + }; + TEST_OK(kstrtou32, u32, "%u", test_u32_ok); +} + +static void __init test_kstrtou32_fail(void) +{ + static DEFINE_TEST_FAIL(test_u32_fail) = { + {"-2", 10}, + {"-1", 10}, + {"4294967296", 10}, + {"4294967297", 10}, + {"9223372036854775806", 10}, + {"9223372036854775807", 10}, + {"9223372036854775808", 10}, + {"9223372036854775809", 10}, + {"18446744073709551614", 10}, + {"18446744073709551615", 10}, + {"18446744073709551616", 10}, + {"18446744073709551617", 10}, + }; + TEST_FAIL(kstrtou32, u32, "%u", test_u32_fail); +} + +static void __init test_kstrtos32_ok(void) +{ + DECLARE_TEST_OK(s32, struct test_s32); + static DEFINE_TEST_OK(struct test_s32, test_s32_ok) = { + {"-128", 10, -128}, + {"-127", 10, -127}, + {"-1", 10, -1}, + {"0", 10, 0}, + {"1", 10, 1}, + {"126", 10, 126}, + {"127", 10, 127}, + {"128", 10, 128}, + {"129", 10, 129}, + {"254", 10, 254}, + {"255", 10, 255}, + {"256", 10, 256}, + {"257", 10, 257}, + {"32766", 10, 32766}, + {"32767", 10, 32767}, + {"32768", 10, 32768}, + {"32769", 10, 32769}, + {"65534", 10, 65534}, + {"65535", 10, 65535}, + {"65536", 10, 65536}, + {"65537", 10, 65537}, + {"2147483646", 10, 2147483646}, + {"2147483647", 10, 2147483647}, + }; + TEST_OK(kstrtos32, s32, "%d", test_s32_ok); +} + +static void __init test_kstrtos32_fail(void) +{ + static DEFINE_TEST_FAIL(test_s32_fail) = { + {"2147483648", 10}, + {"2147483649", 10}, + {"4294967294", 10}, + {"4294967295", 10}, + {"4294967296", 10}, + {"4294967297", 10}, + {"9223372036854775806", 10}, + {"9223372036854775807", 10}, + {"9223372036854775808", 10}, + {"9223372036854775809", 10}, + {"18446744073709551614", 10}, + {"18446744073709551615", 10}, + {"18446744073709551616", 10}, + {"18446744073709551617", 10}, + }; + TEST_FAIL(kstrtos32, s32, "%d", test_s32_fail); +} + +static void __init test_kstrtou16_ok(void) +{ + DECLARE_TEST_OK(u16, struct test_u16); + static DEFINE_TEST_OK(struct test_u16, test_u16_ok) = { + {"0", 10, 0}, + {"1", 10, 1}, + {"126", 10, 126}, + {"127", 10, 127}, + {"128", 10, 128}, + {"129", 10, 129}, + {"254", 10, 254}, + {"255", 10, 255}, + {"256", 10, 256}, + {"257", 10, 257}, + {"32766", 10, 32766}, + {"32767", 10, 32767}, + {"32768", 10, 32768}, + {"32769", 10, 32769}, + {"65534", 10, 65534}, + {"65535", 10, 65535}, + }; + TEST_OK(kstrtou16, u16, "%hu", test_u16_ok); +} + +static void __init test_kstrtou16_fail(void) +{ + static DEFINE_TEST_FAIL(test_u16_fail) = { + {"-2", 10}, + {"-1", 10}, + {"65536", 10}, + {"65537", 10}, + {"2147483646", 10}, + {"2147483647", 10}, + {"2147483648", 10}, + {"2147483649", 10}, + {"4294967294", 10}, + {"4294967295", 10}, + {"4294967296", 10}, + {"4294967297", 10}, + {"9223372036854775806", 10}, + {"9223372036854775807", 10}, + {"9223372036854775808", 10}, + {"9223372036854775809", 10}, + {"18446744073709551614", 10}, + {"18446744073709551615", 10}, + {"18446744073709551616", 10}, + {"18446744073709551617", 10}, + }; + TEST_FAIL(kstrtou16, u16, "%hu", test_u16_fail); +} + +static void __init test_kstrtos16_ok(void) +{ + DECLARE_TEST_OK(s16, struct test_s16); + static DEFINE_TEST_OK(struct test_s16, test_s16_ok) = { + {"-130", 10, -130}, + {"-129", 10, -129}, + {"-128", 10, -128}, + {"-127", 10, -127}, + {"-1", 10, -1}, + {"0", 10, 0}, + {"1", 10, 1}, + {"126", 10, 126}, + {"127", 10, 127}, + {"128", 10, 128}, + {"129", 10, 129}, + {"254", 10, 254}, + {"255", 10, 255}, + {"256", 10, 256}, + {"257", 10, 257}, + {"32766", 10, 32766}, + {"32767", 10, 32767}, + }; + TEST_OK(kstrtos16, s16, "%hd", test_s16_ok); +} + +static void __init test_kstrtos16_fail(void) +{ + static DEFINE_TEST_FAIL(test_s16_fail) = { + {"32768", 10}, + {"32769", 10}, + {"65534", 10}, + {"65535", 10}, + {"65536", 10}, + {"65537", 10}, + {"2147483646", 10}, + {"2147483647", 10}, + {"2147483648", 10}, + {"2147483649", 10}, + {"4294967294", 10}, + {"4294967295", 10}, + {"4294967296", 10}, + {"4294967297", 10}, + {"9223372036854775806", 10}, + {"9223372036854775807", 10}, + {"9223372036854775808", 10}, + {"9223372036854775809", 10}, + {"18446744073709551614", 10}, + {"18446744073709551615", 10}, + {"18446744073709551616", 10}, + {"18446744073709551617", 10}, + }; + TEST_FAIL(kstrtos16, s16, "%hd", test_s16_fail); +} + +static void __init test_kstrtou8_ok(void) +{ + DECLARE_TEST_OK(u8, struct test_u8); + static DEFINE_TEST_OK(struct test_u8, test_u8_ok) = { + {"0", 10, 0}, + {"1", 10, 1}, + {"126", 10, 126}, + {"127", 10, 127}, + {"128", 10, 128}, + {"129", 10, 129}, + {"254", 10, 254}, + {"255", 10, 255}, + }; + TEST_OK(kstrtou8, u8, "%hhu", test_u8_ok); +} + +static void __init test_kstrtou8_fail(void) +{ + static DEFINE_TEST_FAIL(test_u8_fail) = { + {"-2", 10}, + {"-1", 10}, + {"256", 10}, + {"257", 10}, + {"32766", 10}, + {"32767", 10}, + {"32768", 10}, + {"32769", 10}, + {"65534", 10}, + {"65535", 10}, + {"65536", 10}, + {"65537", 10}, + {"2147483646", 10}, + {"2147483647", 10}, + {"2147483648", 10}, + {"2147483649", 10}, + {"4294967294", 10}, + {"4294967295", 10}, + {"4294967296", 10}, + {"4294967297", 10}, + {"9223372036854775806", 10}, + {"9223372036854775807", 10}, + {"9223372036854775808", 10}, + {"9223372036854775809", 10}, + {"18446744073709551614", 10}, + {"18446744073709551615", 10}, + {"18446744073709551616", 10}, + {"18446744073709551617", 10}, + }; + TEST_FAIL(kstrtou8, u8, "%hhu", test_u8_fail); +} + +static void __init test_kstrtos8_ok(void) +{ + DECLARE_TEST_OK(s8, struct test_s8); + static DEFINE_TEST_OK(struct test_s8, test_s8_ok) = { + {"-128", 10, -128}, + {"-127", 10, -127}, + {"-1", 10, -1}, + {"0", 10, 0}, + {"1", 10, 1}, + {"126", 10, 126}, + {"127", 10, 127}, + }; + TEST_OK(kstrtos8, s8, "%hhd", test_s8_ok); +} + +static void __init test_kstrtos8_fail(void) +{ + static DEFINE_TEST_FAIL(test_s8_fail) = { + {"-130", 10}, + {"-129", 10}, + {"128", 10}, + {"129", 10}, + {"254", 10}, + {"255", 10}, + {"256", 10}, + {"257", 10}, + {"32766", 10}, + {"32767", 10}, + {"32768", 10}, + {"32769", 10}, + {"65534", 10}, + {"65535", 10}, + {"65536", 10}, + {"65537", 10}, + {"2147483646", 10}, + {"2147483647", 10}, + {"2147483648", 10}, + {"2147483649", 10}, + {"4294967294", 10}, + {"4294967295", 10}, + {"4294967296", 10}, + {"4294967297", 10}, + {"9223372036854775806", 10}, + {"9223372036854775807", 10}, + {"9223372036854775808", 10}, + {"9223372036854775809", 10}, + {"18446744073709551614", 10}, + {"18446744073709551615", 10}, + {"18446744073709551616", 10}, + {"18446744073709551617", 10}, + }; + TEST_FAIL(kstrtos8, s8, "%hhd", test_s8_fail); +} + +static int __init test_kstrtox_init(void) +{ + test_kstrtoull_ok(); + test_kstrtoull_fail(); + test_kstrtoll_ok(); + test_kstrtoll_fail(); + + test_kstrtou64_ok(); + test_kstrtou64_fail(); + test_kstrtos64_ok(); + test_kstrtos64_fail(); + + test_kstrtou32_ok(); + test_kstrtou32_fail(); + test_kstrtos32_ok(); + test_kstrtos32_fail(); + + test_kstrtou16_ok(); + test_kstrtou16_fail(); + test_kstrtos16_ok(); + test_kstrtos16_fail(); + + test_kstrtou8_ok(); + test_kstrtou8_fail(); + test_kstrtos8_ok(); + test_kstrtos8_fail(); + return -EINVAL; +} +module_init(test_kstrtox_init); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/lib/vsprintf.c b/lib/vsprintf.c index ac444ff0165..02bcdd5feac 100644 --- a/lib/vsprintf.c +++ b/lib/vsprintf.c @@ -120,147 +120,6 @@ long long simple_strtoll(const char *cp, char **endp, unsigned int base) } EXPORT_SYMBOL(simple_strtoll); -/** - * strict_strtoul - convert a string to an unsigned long strictly - * @cp: The string to be converted - * @base: The number base to use - * @res: The converted result value - * - * strict_strtoul converts a string to an unsigned long only if the - * string is really an unsigned long string, any string containing - * any invalid char at the tail will be rejected and -EINVAL is returned, - * only a newline char at the tail is acceptible because people generally - * change a module parameter in the following way: - * - * echo 1024 > /sys/module/e1000/parameters/copybreak - * - * echo will append a newline to the tail. - * - * It returns 0 if conversion is successful and *res is set to the converted - * value, otherwise it returns -EINVAL and *res is set to 0. - * - * simple_strtoul just ignores the successive invalid characters and - * return the converted value of prefix part of the string. - */ -int strict_strtoul(const char *cp, unsigned int base, unsigned long *res) -{ - char *tail; - unsigned long val; - - *res = 0; - if (!*cp) - return -EINVAL; - - val = simple_strtoul(cp, &tail, base); - if (tail == cp) - return -EINVAL; - - if ((tail[0] == '\0') || (tail[0] == '\n' && tail[1] == '\0')) { - *res = val; - return 0; - } - - return -EINVAL; -} -EXPORT_SYMBOL(strict_strtoul); - -/** - * strict_strtol - convert a string to a long strictly - * @cp: The string to be converted - * @base: The number base to use - * @res: The converted result value - * - * strict_strtol is similiar to strict_strtoul, but it allows the first - * character of a string is '-'. - * - * It returns 0 if conversion is successful and *res is set to the converted - * value, otherwise it returns -EINVAL and *res is set to 0. - */ -int strict_strtol(const char *cp, unsigned int base, long *res) -{ - int ret; - if (*cp == '-') { - ret = strict_strtoul(cp + 1, base, (unsigned long *)res); - if (!ret) - *res = -(*res); - } else { - ret = strict_strtoul(cp, base, (unsigned long *)res); - } - - return ret; -} -EXPORT_SYMBOL(strict_strtol); - -/** - * strict_strtoull - convert a string to an unsigned long long strictly - * @cp: The string to be converted - * @base: The number base to use - * @res: The converted result value - * - * strict_strtoull converts a string to an unsigned long long only if the - * string is really an unsigned long long string, any string containing - * any invalid char at the tail will be rejected and -EINVAL is returned, - * only a newline char at the tail is acceptible because people generally - * change a module parameter in the following way: - * - * echo 1024 > /sys/module/e1000/parameters/copybreak - * - * echo will append a newline to the tail of the string. - * - * It returns 0 if conversion is successful and *res is set to the converted - * value, otherwise it returns -EINVAL and *res is set to 0. - * - * simple_strtoull just ignores the successive invalid characters and - * return the converted value of prefix part of the string. - */ -int strict_strtoull(const char *cp, unsigned int base, unsigned long long *res) -{ - char *tail; - unsigned long long val; - - *res = 0; - if (!*cp) - return -EINVAL; - - val = simple_strtoull(cp, &tail, base); - if (tail == cp) - return -EINVAL; - if ((tail[0] == '\0') || (tail[0] == '\n' && tail[1] == '\0')) { - *res = val; - return 0; - } - - return -EINVAL; -} -EXPORT_SYMBOL(strict_strtoull); - -/** - * strict_strtoll - convert a string to a long long strictly - * @cp: The string to be converted - * @base: The number base to use - * @res: The converted result value - * - * strict_strtoll is similiar to strict_strtoull, but it allows the first - * character of a string is '-'. - * - * It returns 0 if conversion is successful and *res is set to the converted - * value, otherwise it returns -EINVAL and *res is set to 0. - */ -int strict_strtoll(const char *cp, unsigned int base, long long *res) -{ - int ret; - if (*cp == '-') { - ret = strict_strtoull(cp + 1, base, (unsigned long long *)res); - if (!ret) - *res = -(*res); - } else { - ret = strict_strtoull(cp, base, (unsigned long long *)res); - } - - return ret; -} -EXPORT_SYMBOL(strict_strtoll); - static noinline_for_stack int skip_atoi(const char **s) { diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 58848e3e392..dfd9b310a1c 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -2804,9 +2804,9 @@ sub process { WARN("consider using a completion\n" . $herecurr); } -# recommend strict_strto* over simple_strto* +# recommend kstrto* over simple_strto* if ($line =~ /\bsimple_(strto.*?)\s*\(/) { - WARN("consider using strict_$1 in preference to simple_$1\n" . $herecurr); + WARN("consider using kstrto* in preference to simple_$1\n" . $herecurr); } # check for __initcall(), use device_initcall() explicitly please if ($line =~ /^.\s*__initcall\s*\(/) { -- cgit v1.2.3-70-g09d2 From 565d76cb7d5fd7cb010fd690602280a69ab116ef Mon Sep 17 00:00:00 2001 From: Jim Keniston Date: Tue, 22 Mar 2011 16:35:12 -0700 Subject: zlib: slim down zlib_deflate() workspace when possible Instead of always creating a huge (268K) deflate_workspace with the maximum compression parameters (windowBits=15, memLevel=8), allow the caller to obtain a smaller workspace by specifying smaller parameter values. For example, when capturing oops and panic reports to a medium with limited capacity, such as NVRAM, compression may be the only way to capture the whole report. In this case, a small workspace (24K works fine) is a win, whether you allocate the workspace when you need it (i.e., during an oops or panic) or at boot time. I've verified that this patch works with all accepted values of windowBits (positive and negative), memLevel, and compression level. Signed-off-by: Jim Keniston Cc: Herbert Xu Cc: David Miller Cc: Chris Mason Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- crypto/deflate.c | 3 ++- crypto/zlib.c | 18 +++++++++++------- drivers/net/ppp_deflate.c | 2 +- fs/btrfs/zlib.c | 3 ++- fs/jffs2/compr_zlib.c | 7 ++++--- fs/logfs/compr.c | 2 +- include/linux/zlib.h | 11 ++++++++--- lib/zlib_deflate/deflate.c | 31 +++++++++++++++++++++++++++++-- lib/zlib_deflate/defutil.h | 17 +++++++++++++---- 9 files changed, 71 insertions(+), 23 deletions(-) (limited to 'lib') diff --git a/crypto/deflate.c b/crypto/deflate.c index cbc7a33a960..b5ccae29be7 100644 --- a/crypto/deflate.c +++ b/crypto/deflate.c @@ -48,7 +48,8 @@ static int deflate_comp_init(struct deflate_ctx *ctx) int ret = 0; struct z_stream_s *stream = &ctx->comp_stream; - stream->workspace = vzalloc(zlib_deflate_workspacesize()); + stream->workspace = vzalloc(zlib_deflate_workspacesize( + -DEFLATE_DEF_WINBITS, DEFLATE_DEF_MEMLEVEL)); if (!stream->workspace) { ret = -ENOMEM; goto out; diff --git a/crypto/zlib.c b/crypto/zlib.c index 739b8fca4ce..d11d761a5e4 100644 --- a/crypto/zlib.c +++ b/crypto/zlib.c @@ -85,6 +85,7 @@ static int zlib_compress_setup(struct crypto_pcomp *tfm, void *params, struct zlib_ctx *ctx = crypto_tfm_ctx(crypto_pcomp_tfm(tfm)); struct z_stream_s *stream = &ctx->comp_stream; struct nlattr *tb[ZLIB_COMP_MAX + 1]; + int window_bits, mem_level; size_t workspacesize; int ret; @@ -94,7 +95,14 @@ static int zlib_compress_setup(struct crypto_pcomp *tfm, void *params, zlib_comp_exit(ctx); - workspacesize = zlib_deflate_workspacesize(); + window_bits = tb[ZLIB_COMP_WINDOWBITS] + ? nla_get_u32(tb[ZLIB_COMP_WINDOWBITS]) + : MAX_WBITS; + mem_level = tb[ZLIB_COMP_MEMLEVEL] + ? nla_get_u32(tb[ZLIB_COMP_MEMLEVEL]) + : DEF_MEM_LEVEL; + + workspacesize = zlib_deflate_workspacesize(window_bits, mem_level); stream->workspace = vzalloc(workspacesize); if (!stream->workspace) return -ENOMEM; @@ -106,12 +114,8 @@ static int zlib_compress_setup(struct crypto_pcomp *tfm, void *params, tb[ZLIB_COMP_METHOD] ? nla_get_u32(tb[ZLIB_COMP_METHOD]) : Z_DEFLATED, - tb[ZLIB_COMP_WINDOWBITS] - ? nla_get_u32(tb[ZLIB_COMP_WINDOWBITS]) - : MAX_WBITS, - tb[ZLIB_COMP_MEMLEVEL] - ? nla_get_u32(tb[ZLIB_COMP_MEMLEVEL]) - : DEF_MEM_LEVEL, + window_bits, + mem_level, tb[ZLIB_COMP_STRATEGY] ? nla_get_u32(tb[ZLIB_COMP_STRATEGY]) : Z_DEFAULT_STRATEGY); diff --git a/drivers/net/ppp_deflate.c b/drivers/net/ppp_deflate.c index 43583309a65..31e9407a073 100644 --- a/drivers/net/ppp_deflate.c +++ b/drivers/net/ppp_deflate.c @@ -129,7 +129,7 @@ static void *z_comp_alloc(unsigned char *options, int opt_len) state->strm.next_in = NULL; state->w_size = w_size; - state->strm.workspace = vmalloc(zlib_deflate_workspacesize()); + state->strm.workspace = vmalloc(zlib_deflate_workspacesize(-w_size, 8)); if (state->strm.workspace == NULL) goto out_free; diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c index f5ec2d44150..faccd47c6c4 100644 --- a/fs/btrfs/zlib.c +++ b/fs/btrfs/zlib.c @@ -57,7 +57,8 @@ static struct list_head *zlib_alloc_workspace(void) if (!workspace) return ERR_PTR(-ENOMEM); - workspace->def_strm.workspace = vmalloc(zlib_deflate_workspacesize()); + workspace->def_strm.workspace = vmalloc(zlib_deflate_workspacesize( + MAX_WBITS, MAX_MEM_LEVEL)); workspace->inf_strm.workspace = vmalloc(zlib_inflate_workspacesize()); workspace->buf = kmalloc(PAGE_CACHE_SIZE, GFP_NOFS); if (!workspace->def_strm.workspace || diff --git a/fs/jffs2/compr_zlib.c b/fs/jffs2/compr_zlib.c index fd05a0b9431..5a001020c54 100644 --- a/fs/jffs2/compr_zlib.c +++ b/fs/jffs2/compr_zlib.c @@ -40,12 +40,13 @@ static z_stream inf_strm, def_strm; static int __init alloc_workspaces(void) { - def_strm.workspace = vmalloc(zlib_deflate_workspacesize()); + def_strm.workspace = vmalloc(zlib_deflate_workspacesize(MAX_WBITS, + MAX_MEM_LEVEL)); if (!def_strm.workspace) { - printk(KERN_WARNING "Failed to allocate %d bytes for deflate workspace\n", zlib_deflate_workspacesize()); + printk(KERN_WARNING "Failed to allocate %d bytes for deflate workspace\n", zlib_deflate_workspacesize(MAX_WBITS, MAX_MEM_LEVEL)); return -ENOMEM; } - D1(printk(KERN_DEBUG "Allocated %d bytes for deflate workspace\n", zlib_deflate_workspacesize())); + D1(printk(KERN_DEBUG "Allocated %d bytes for deflate workspace\n", zlib_deflate_workspacesize(MAX_WBITS, MAX_MEM_LEVEL))); inf_strm.workspace = vmalloc(zlib_inflate_workspacesize()); if (!inf_strm.workspace) { printk(KERN_WARNING "Failed to allocate %d bytes for inflate workspace\n", zlib_inflate_workspacesize()); diff --git a/fs/logfs/compr.c b/fs/logfs/compr.c index 44bbfd249ab..961f02b86d9 100644 --- a/fs/logfs/compr.c +++ b/fs/logfs/compr.c @@ -81,7 +81,7 @@ error: int __init logfs_compr_init(void) { - size_t size = max(zlib_deflate_workspacesize(), + size_t size = max(zlib_deflate_workspacesize(MAX_WBITS, MAX_MEM_LEVEL), zlib_inflate_workspacesize()); stream.workspace = vmalloc(size); if (!stream.workspace) diff --git a/include/linux/zlib.h b/include/linux/zlib.h index 40c49cb3eb5..9c5a6b4de0a 100644 --- a/include/linux/zlib.h +++ b/include/linux/zlib.h @@ -179,11 +179,16 @@ typedef z_stream *z_streamp; /* basic functions */ -extern int zlib_deflate_workspacesize (void); +extern int zlib_deflate_workspacesize (int windowBits, int memLevel); /* Returns the number of bytes that needs to be allocated for a per- - stream workspace. A pointer to this number of bytes should be - returned in stream->workspace before calling zlib_deflateInit(). + stream workspace with the specified parameters. A pointer to this + number of bytes should be returned in stream->workspace before + you call zlib_deflateInit() or zlib_deflateInit2(). If you call + zlib_deflateInit(), specify windowBits = MAX_WBITS and memLevel = + MAX_MEM_LEVEL here. If you call zlib_deflateInit2(), the windowBits + and memLevel parameters passed to zlib_deflateInit2() must not + exceed those passed here. */ /* diff --git a/lib/zlib_deflate/deflate.c b/lib/zlib_deflate/deflate.c index 46a31e5f49c..d63381e8e33 100644 --- a/lib/zlib_deflate/deflate.c +++ b/lib/zlib_deflate/deflate.c @@ -176,6 +176,7 @@ int zlib_deflateInit2( deflate_state *s; int noheader = 0; deflate_workspace *mem; + char *next; ush *overlay; /* We overlay pending_buf and d_buf+l_buf. This works since the average @@ -199,6 +200,21 @@ int zlib_deflateInit2( strategy < 0 || strategy > Z_HUFFMAN_ONLY) { return Z_STREAM_ERROR; } + + /* + * Direct the workspace's pointers to the chunks that were allocated + * along with the deflate_workspace struct. + */ + next = (char *) mem; + next += sizeof(*mem); + mem->window_memory = (Byte *) next; + next += zlib_deflate_window_memsize(windowBits); + mem->prev_memory = (Pos *) next; + next += zlib_deflate_prev_memsize(windowBits); + mem->head_memory = (Pos *) next; + next += zlib_deflate_head_memsize(memLevel); + mem->overlay_memory = next; + s = (deflate_state *) &(mem->deflate_memory); strm->state = (struct internal_state *)s; s->strm = strm; @@ -1247,7 +1263,18 @@ static block_state deflate_slow( return flush == Z_FINISH ? finish_done : block_done; } -int zlib_deflate_workspacesize(void) +int zlib_deflate_workspacesize(int windowBits, int memLevel) { - return sizeof(deflate_workspace); + if (windowBits < 0) /* undocumented feature: suppress zlib header */ + windowBits = -windowBits; + + /* Since the return value is typically passed to vmalloc() unchecked... */ + BUG_ON(memLevel < 1 || memLevel > MAX_MEM_LEVEL || windowBits < 9 || + windowBits > 15); + + return sizeof(deflate_workspace) + + zlib_deflate_window_memsize(windowBits) + + zlib_deflate_prev_memsize(windowBits) + + zlib_deflate_head_memsize(memLevel) + + zlib_deflate_overlay_memsize(memLevel); } diff --git a/lib/zlib_deflate/defutil.h b/lib/zlib_deflate/defutil.h index 6b15a909ca3..b640b6402e9 100644 --- a/lib/zlib_deflate/defutil.h +++ b/lib/zlib_deflate/defutil.h @@ -241,12 +241,21 @@ typedef struct deflate_state { typedef struct deflate_workspace { /* State memory for the deflator */ deflate_state deflate_memory; - Byte window_memory[2 * (1 << MAX_WBITS)]; - Pos prev_memory[1 << MAX_WBITS]; - Pos head_memory[1 << (MAX_MEM_LEVEL + 7)]; - char overlay_memory[(1 << (MAX_MEM_LEVEL + 6)) * (sizeof(ush)+2)]; + Byte *window_memory; + Pos *prev_memory; + Pos *head_memory; + char *overlay_memory; } deflate_workspace; +#define zlib_deflate_window_memsize(windowBits) \ + (2 * (1 << (windowBits)) * sizeof(Byte)) +#define zlib_deflate_prev_memsize(windowBits) \ + ((1 << (windowBits)) * sizeof(Pos)) +#define zlib_deflate_head_memsize(memLevel) \ + ((1 << ((memLevel)+7)) * sizeof(Pos)) +#define zlib_deflate_overlay_memsize(memLevel) \ + ((1 << ((memLevel)+6)) * (sizeof(ush)+2)) + /* Output a byte on the stream. * IN assertion: there is enough room in pending_buf. */ -- cgit v1.2.3-70-g09d2 From c4945b9ed472e8796e352f10df9dbc2841ba7b61 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Wed, 23 Mar 2011 16:41:47 -0700 Subject: asm-generic: rename generic little-endian bitops functions As a preparation for providing little-endian bitops for all architectures, This renames generic implementation of little-endian bitops. (remove "generic_" prefix and postfix "_le") s/generic_find_next_le_bit/find_next_bit_le/ s/generic_find_next_zero_le_bit/find_next_zero_bit_le/ s/generic_find_first_zero_le_bit/find_first_zero_bit_le/ s/generic___test_and_set_le_bit/__test_and_set_bit_le/ s/generic___test_and_clear_le_bit/__test_and_clear_bit_le/ s/generic_test_le_bit/test_bit_le/ s/generic___set_le_bit/__set_bit_le/ s/generic___clear_le_bit/__clear_bit_le/ s/generic_test_and_set_le_bit/test_and_set_bit_le/ s/generic_test_and_clear_le_bit/test_and_clear_bit_le/ Signed-off-by: Akinobu Mita Acked-by: Arnd Bergmann Acked-by: Hans-Christian Egtvedt Cc: Geert Uytterhoeven Cc: Roman Zippel Cc: Andreas Schwab Cc: Greg Ungerer Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/avr32/kernel/avr32_ksyms.c | 4 ++-- arch/avr32/lib/findbit.S | 4 ++-- arch/m68k/include/asm/bitops_mm.h | 8 ++++---- arch/m68k/include/asm/bitops_no.h | 2 +- arch/powerpc/include/asm/bitops.h | 15 ++++++++------- include/asm-generic/bitops/ext2-non-atomic.h | 12 ++++++------ include/asm-generic/bitops/le.h | 28 ++++++++++++++-------------- include/asm-generic/bitops/minix-le.h | 10 +++++----- lib/find_next_bit.c | 9 ++++----- 9 files changed, 46 insertions(+), 46 deletions(-) (limited to 'lib') diff --git a/arch/avr32/kernel/avr32_ksyms.c b/arch/avr32/kernel/avr32_ksyms.c index 11e310c567a..d93ead02dae 100644 --- a/arch/avr32/kernel/avr32_ksyms.c +++ b/arch/avr32/kernel/avr32_ksyms.c @@ -58,8 +58,8 @@ EXPORT_SYMBOL(find_first_zero_bit); EXPORT_SYMBOL(find_next_zero_bit); EXPORT_SYMBOL(find_first_bit); EXPORT_SYMBOL(find_next_bit); -EXPORT_SYMBOL(generic_find_next_le_bit); -EXPORT_SYMBOL(generic_find_next_zero_le_bit); +EXPORT_SYMBOL(find_next_bit_le); +EXPORT_SYMBOL(find_next_zero_bit_le); /* I/O primitives (lib/io-*.S) */ EXPORT_SYMBOL(__raw_readsb); diff --git a/arch/avr32/lib/findbit.S b/arch/avr32/lib/findbit.S index 997b33b2288..b93586460be 100644 --- a/arch/avr32/lib/findbit.S +++ b/arch/avr32/lib/findbit.S @@ -123,7 +123,7 @@ ENTRY(find_next_bit) brgt 1b retal r11 -ENTRY(generic_find_next_le_bit) +ENTRY(find_next_bit_le) lsr r8, r10, 5 sub r9, r11, r10 retle r11 @@ -153,7 +153,7 @@ ENTRY(generic_find_next_le_bit) brgt 1b retal r11 -ENTRY(generic_find_next_zero_le_bit) +ENTRY(find_next_zero_bit_le) lsr r8, r10, 5 sub r9, r11, r10 retle r11 diff --git a/arch/m68k/include/asm/bitops_mm.h b/arch/m68k/include/asm/bitops_mm.h index b4ecdaada52..31a16d47d20 100644 --- a/arch/m68k/include/asm/bitops_mm.h +++ b/arch/m68k/include/asm/bitops_mm.h @@ -366,9 +366,9 @@ static inline int minix_test_bit(int nr, const void *vaddr) #define ext2_clear_bit(nr, addr) __test_and_clear_bit((nr) ^ 24, (unsigned long *)(addr)) #define ext2_clear_bit_atomic(lock, nr, addr) test_and_clear_bit((nr) ^ 24, (unsigned long *)(addr)) #define ext2_find_next_zero_bit(addr, size, offset) \ - generic_find_next_zero_le_bit((unsigned long *)addr, size, offset) + find_next_zero_bit_le((unsigned long *)addr, size, offset) #define ext2_find_next_bit(addr, size, offset) \ - generic_find_next_le_bit((unsigned long *)addr, size, offset) + find_next_bit_le((unsigned long *)addr, size, offset) static inline int ext2_test_bit(int nr, const void *vaddr) { @@ -398,7 +398,7 @@ static inline int ext2_find_first_zero_bit(const void *vaddr, unsigned size) return (p - addr) * 32 + res; } -static inline unsigned long generic_find_next_zero_le_bit(const unsigned long *addr, +static inline unsigned long find_next_zero_bit_le(const unsigned long *addr, unsigned long size, unsigned long offset) { const unsigned long *p = addr + (offset >> 5); @@ -440,7 +440,7 @@ static inline int ext2_find_first_bit(const void *vaddr, unsigned size) return (p - addr) * 32 + res; } -static inline unsigned long generic_find_next_le_bit(const unsigned long *addr, +static inline unsigned long find_next_bit_le(const unsigned long *addr, unsigned long size, unsigned long offset) { const unsigned long *p = addr + (offset >> 5); diff --git a/arch/m68k/include/asm/bitops_no.h b/arch/m68k/include/asm/bitops_no.h index 9d3cbe5fad1..58c67aaf91e 100644 --- a/arch/m68k/include/asm/bitops_no.h +++ b/arch/m68k/include/asm/bitops_no.h @@ -325,7 +325,7 @@ found_middle: } #define ext2_find_next_bit(addr, size, off) \ - generic_find_next_le_bit((unsigned long *)(addr), (size), (off)) + find_next_bit_le((unsigned long *)(addr), (size), (off)) #include #endif /* __KERNEL__ */ diff --git a/arch/powerpc/include/asm/bitops.h b/arch/powerpc/include/asm/bitops.h index 8a7e9314c68..97aacbdd752 100644 --- a/arch/powerpc/include/asm/bitops.h +++ b/arch/powerpc/include/asm/bitops.h @@ -303,11 +303,12 @@ static __inline__ int test_le_bit(unsigned long nr, #define __test_and_clear_le_bit(nr, addr) \ __test_and_clear_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) -#define find_first_zero_le_bit(addr, size) generic_find_next_zero_le_bit((addr), (size), 0) -unsigned long generic_find_next_zero_le_bit(const unsigned long *addr, +#define find_first_zero_bit_le(addr, size) \ + find_next_zero_bit_le((addr), (size), 0) +unsigned long find_next_zero_bit_le(const unsigned long *addr, unsigned long size, unsigned long offset); -unsigned long generic_find_next_le_bit(const unsigned long *addr, +unsigned long find_next_bit_le(const unsigned long *addr, unsigned long size, unsigned long offset); /* Bitmap functions for the ext2 filesystem */ @@ -324,12 +325,12 @@ unsigned long generic_find_next_le_bit(const unsigned long *addr, #define ext2_test_bit(nr, addr) test_le_bit((nr),(unsigned long*)addr) #define ext2_find_first_zero_bit(addr, size) \ - find_first_zero_le_bit((unsigned long*)addr, size) + find_first_zero_bit_le((unsigned long*)addr, size) #define ext2_find_next_zero_bit(addr, size, off) \ - generic_find_next_zero_le_bit((unsigned long*)addr, size, off) + find_next_zero_bit_le((unsigned long *)addr, size, off) #define ext2_find_next_bit(addr, size, off) \ - generic_find_next_le_bit((unsigned long *)addr, size, off) + find_next_bit_le((unsigned long *)addr, size, off) /* Bitmap functions for the minix filesystem. */ #define minix_test_and_set_bit(nr,addr) \ @@ -342,7 +343,7 @@ unsigned long generic_find_next_le_bit(const unsigned long *addr, test_le_bit(nr, (unsigned long *)addr) #define minix_find_first_zero_bit(addr,size) \ - find_first_zero_le_bit((unsigned long *)addr, size) + find_first_zero_bit_le((unsigned long *)addr, size) #include diff --git a/include/asm-generic/bitops/ext2-non-atomic.h b/include/asm-generic/bitops/ext2-non-atomic.h index 63cf822431a..ba27bbe755a 100644 --- a/include/asm-generic/bitops/ext2-non-atomic.h +++ b/include/asm-generic/bitops/ext2-non-atomic.h @@ -4,17 +4,17 @@ #include #define ext2_set_bit(nr,addr) \ - generic___test_and_set_le_bit((nr),(unsigned long *)(addr)) + __test_and_set_bit_le((nr), (unsigned long *)(addr)) #define ext2_clear_bit(nr,addr) \ - generic___test_and_clear_le_bit((nr),(unsigned long *)(addr)) + __test_and_clear_bit_le((nr), (unsigned long *)(addr)) #define ext2_test_bit(nr,addr) \ - generic_test_le_bit((nr),(unsigned long *)(addr)) + test_bit_le((nr), (unsigned long *)(addr)) #define ext2_find_first_zero_bit(addr, size) \ - generic_find_first_zero_le_bit((unsigned long *)(addr), (size)) + find_first_zero_bit_le((unsigned long *)(addr), (size)) #define ext2_find_next_zero_bit(addr, size, off) \ - generic_find_next_zero_le_bit((unsigned long *)(addr), (size), (off)) + find_next_zero_bit_le((unsigned long *)(addr), (size), (off)) #define ext2_find_next_bit(addr, size, off) \ - generic_find_next_le_bit((unsigned long *)(addr), (size), (off)) + find_next_bit_le((unsigned long *)(addr), (size), (off)) #endif /* _ASM_GENERIC_BITOPS_EXT2_NON_ATOMIC_H_ */ diff --git a/include/asm-generic/bitops/le.h b/include/asm-generic/bitops/le.h index 9085429ff3c..394ac3ebcbd 100644 --- a/include/asm-generic/bitops/le.h +++ b/include/asm-generic/bitops/le.h @@ -8,44 +8,44 @@ #define BITOP_LE_SWIZZLE 0 -#define generic_find_next_zero_le_bit(addr, size, offset) \ +#define find_next_zero_bit_le(addr, size, offset) \ find_next_zero_bit(addr, size, offset) -#define generic_find_next_le_bit(addr, size, offset) \ +#define find_next_bit_le(addr, size, offset) \ find_next_bit(addr, size, offset) -#define generic_find_first_zero_le_bit(addr, size) \ +#define find_first_zero_bit_le(addr, size) \ find_first_zero_bit(addr, size) #elif defined(__BIG_ENDIAN) #define BITOP_LE_SWIZZLE ((BITS_PER_LONG-1) & ~0x7) -extern unsigned long generic_find_next_zero_le_bit(const unsigned long *addr, +extern unsigned long find_next_zero_bit_le(const unsigned long *addr, unsigned long size, unsigned long offset); -extern unsigned long generic_find_next_le_bit(const unsigned long *addr, +extern unsigned long find_next_bit_le(const unsigned long *addr, unsigned long size, unsigned long offset); -#define generic_find_first_zero_le_bit(addr, size) \ - generic_find_next_zero_le_bit((addr), (size), 0) +#define find_first_zero_bit_le(addr, size) \ + find_next_zero_bit_le((addr), (size), 0) #else #error "Please fix " #endif -#define generic_test_le_bit(nr, addr) \ +#define test_bit_le(nr, addr) \ test_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) -#define generic___set_le_bit(nr, addr) \ +#define __set_bit_le(nr, addr) \ __set_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) -#define generic___clear_le_bit(nr, addr) \ +#define __clear_bit_le(nr, addr) \ __clear_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) -#define generic_test_and_set_le_bit(nr, addr) \ +#define test_and_set_bit_le(nr, addr) \ test_and_set_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) -#define generic_test_and_clear_le_bit(nr, addr) \ +#define test_and_clear_bit_le(nr, addr) \ test_and_clear_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) -#define generic___test_and_set_le_bit(nr, addr) \ +#define __test_and_set_bit_le(nr, addr) \ __test_and_set_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) -#define generic___test_and_clear_le_bit(nr, addr) \ +#define __test_and_clear_bit_le(nr, addr) \ __test_and_clear_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) #endif /* _ASM_GENERIC_BITOPS_LE_H_ */ diff --git a/include/asm-generic/bitops/minix-le.h b/include/asm-generic/bitops/minix-le.h index 4a981c1bb1a..854e9489f62 100644 --- a/include/asm-generic/bitops/minix-le.h +++ b/include/asm-generic/bitops/minix-le.h @@ -4,14 +4,14 @@ #include #define minix_test_and_set_bit(nr,addr) \ - generic___test_and_set_le_bit((nr),(unsigned long *)(addr)) + __test_and_set_bit_le((nr), (unsigned long *)(addr)) #define minix_set_bit(nr,addr) \ - generic___set_le_bit((nr),(unsigned long *)(addr)) + __set_bit_le((nr), (unsigned long *)(addr)) #define minix_test_and_clear_bit(nr,addr) \ - generic___test_and_clear_le_bit((nr),(unsigned long *)(addr)) + __test_and_clear_bit_le((nr), (unsigned long *)(addr)) #define minix_test_bit(nr,addr) \ - generic_test_le_bit((nr),(unsigned long *)(addr)) + test_bit_le((nr), (unsigned long *)(addr)) #define minix_find_first_zero_bit(addr,size) \ - generic_find_first_zero_le_bit((unsigned long *)(addr),(size)) + find_first_zero_bit_le((unsigned long *)(addr), (size)) #endif /* _ASM_GENERIC_BITOPS_MINIX_LE_H_ */ diff --git a/lib/find_next_bit.c b/lib/find_next_bit.c index 24c59ded47a..c093ba98800 100644 --- a/lib/find_next_bit.c +++ b/lib/find_next_bit.c @@ -185,7 +185,7 @@ static inline unsigned long ext2_swab(const unsigned long y) #endif } -unsigned long generic_find_next_zero_le_bit(const unsigned long *addr, unsigned +unsigned long find_next_zero_bit_le(const unsigned long *addr, unsigned long size, unsigned long offset) { const unsigned long *p = addr + BITOP_WORD(offset); @@ -226,10 +226,9 @@ found_middle: found_middle_swap: return result + ffz(ext2_swab(tmp)); } +EXPORT_SYMBOL(find_next_zero_bit_le); -EXPORT_SYMBOL(generic_find_next_zero_le_bit); - -unsigned long generic_find_next_le_bit(const unsigned long *addr, unsigned +unsigned long find_next_bit_le(const unsigned long *addr, unsigned long size, unsigned long offset) { const unsigned long *p = addr + BITOP_WORD(offset); @@ -271,5 +270,5 @@ found_middle: found_middle_swap: return result + __ffs(ext2_swab(tmp)); } -EXPORT_SYMBOL(generic_find_next_le_bit); +EXPORT_SYMBOL(find_next_bit_le); #endif /* __BIG_ENDIAN */ -- cgit v1.2.3-70-g09d2 From a56560b3b233238e85205d4e8d7bded904ac2306 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Wed, 23 Mar 2011 16:41:50 -0700 Subject: asm-generic: change little-endian bitops to take any pointer types This makes the little-endian bitops take any pointer types by changing the prototypes and adding casts in the preprocessor macros. That would seem to at least make all the filesystem code happier, and they can continue to do just something like #define ext2_set_bit __test_and_set_bit_le (or whatever the exact sequence ends up being). Signed-off-by: Akinobu Mita Cc: Arnd Bergmann Cc: Richard Henderson Cc: Ivan Kokshaysky Cc: Mikael Starvik Cc: David Howells Cc: Yoshinori Sato Cc: "Luck, Tony" Cc: Ralf Baechle Cc: Kyle McMartin Cc: Matthew Wilcox Cc: Grant Grundler Cc: Paul Mundt Cc: Kazumoto Kojima Cc: Hirokazu Takata Cc: "David S. Miller" Cc: Chris Zankel Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Hans-Christian Egtvedt Cc: "H. Peter Anvin" Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/include/asm/bitops.h | 4 +- include/asm-generic/bitops/le.h | 77 +++++++++++++++++++++++++++------------ lib/find_next_bit.c | 10 +++-- 3 files changed, 61 insertions(+), 30 deletions(-) (limited to 'lib') diff --git a/arch/powerpc/include/asm/bitops.h b/arch/powerpc/include/asm/bitops.h index 97aacbdd752..db567ed9212 100644 --- a/arch/powerpc/include/asm/bitops.h +++ b/arch/powerpc/include/asm/bitops.h @@ -305,10 +305,10 @@ static __inline__ int test_le_bit(unsigned long nr, #define find_first_zero_bit_le(addr, size) \ find_next_zero_bit_le((addr), (size), 0) -unsigned long find_next_zero_bit_le(const unsigned long *addr, +unsigned long find_next_zero_bit_le(const void *addr, unsigned long size, unsigned long offset); -unsigned long find_next_bit_le(const unsigned long *addr, +unsigned long find_next_bit_le(const void *addr, unsigned long size, unsigned long offset); /* Bitmap functions for the ext2 filesystem */ diff --git a/include/asm-generic/bitops/le.h b/include/asm-generic/bitops/le.h index 394ac3ebcbd..946a21b1b5d 100644 --- a/include/asm-generic/bitops/le.h +++ b/include/asm-generic/bitops/le.h @@ -8,20 +8,31 @@ #define BITOP_LE_SWIZZLE 0 -#define find_next_zero_bit_le(addr, size, offset) \ - find_next_zero_bit(addr, size, offset) -#define find_next_bit_le(addr, size, offset) \ - find_next_bit(addr, size, offset) -#define find_first_zero_bit_le(addr, size) \ - find_first_zero_bit(addr, size) +static inline unsigned long find_next_zero_bit_le(const void *addr, + unsigned long size, unsigned long offset) +{ + return find_next_zero_bit(addr, size, offset); +} + +static inline unsigned long find_next_bit_le(const void *addr, + unsigned long size, unsigned long offset) +{ + return find_next_bit(addr, size, offset); +} + +static inline unsigned long find_first_zero_bit_le(const void *addr, + unsigned long size) +{ + return find_first_zero_bit(addr, size); +} #elif defined(__BIG_ENDIAN) #define BITOP_LE_SWIZZLE ((BITS_PER_LONG-1) & ~0x7) -extern unsigned long find_next_zero_bit_le(const unsigned long *addr, +extern unsigned long find_next_zero_bit_le(const void *addr, unsigned long size, unsigned long offset); -extern unsigned long find_next_bit_le(const unsigned long *addr, +extern unsigned long find_next_bit_le(const void *addr, unsigned long size, unsigned long offset); #define find_first_zero_bit_le(addr, size) \ @@ -31,21 +42,39 @@ extern unsigned long find_next_bit_le(const unsigned long *addr, #error "Please fix " #endif -#define test_bit_le(nr, addr) \ - test_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) -#define __set_bit_le(nr, addr) \ - __set_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) -#define __clear_bit_le(nr, addr) \ - __clear_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) - -#define test_and_set_bit_le(nr, addr) \ - test_and_set_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) -#define test_and_clear_bit_le(nr, addr) \ - test_and_clear_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) - -#define __test_and_set_bit_le(nr, addr) \ - __test_and_set_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) -#define __test_and_clear_bit_le(nr, addr) \ - __test_and_clear_bit((nr) ^ BITOP_LE_SWIZZLE, (addr)) +static inline int test_bit_le(int nr, const void *addr) +{ + return test_bit(nr ^ BITOP_LE_SWIZZLE, addr); +} + +static inline void __set_bit_le(int nr, void *addr) +{ + __set_bit(nr ^ BITOP_LE_SWIZZLE, addr); +} + +static inline void __clear_bit_le(int nr, void *addr) +{ + __clear_bit(nr ^ BITOP_LE_SWIZZLE, addr); +} + +static inline int test_and_set_bit_le(int nr, void *addr) +{ + return test_and_set_bit(nr ^ BITOP_LE_SWIZZLE, addr); +} + +static inline int test_and_clear_bit_le(int nr, void *addr) +{ + return test_and_clear_bit(nr ^ BITOP_LE_SWIZZLE, addr); +} + +static inline int __test_and_set_bit_le(int nr, void *addr) +{ + return __test_and_set_bit(nr ^ BITOP_LE_SWIZZLE, addr); +} + +static inline int __test_and_clear_bit_le(int nr, void *addr) +{ + return __test_and_clear_bit(nr ^ BITOP_LE_SWIZZLE, addr); +} #endif /* _ASM_GENERIC_BITOPS_LE_H_ */ diff --git a/lib/find_next_bit.c b/lib/find_next_bit.c index c093ba98800..7667c3d907d 100644 --- a/lib/find_next_bit.c +++ b/lib/find_next_bit.c @@ -185,15 +185,16 @@ static inline unsigned long ext2_swab(const unsigned long y) #endif } -unsigned long find_next_zero_bit_le(const unsigned long *addr, unsigned +unsigned long find_next_zero_bit_le(const void *addr, unsigned long size, unsigned long offset) { - const unsigned long *p = addr + BITOP_WORD(offset); + const unsigned long *p = addr; unsigned long result = offset & ~(BITS_PER_LONG - 1); unsigned long tmp; if (offset >= size) return size; + p += BITOP_WORD(offset); size -= result; offset &= (BITS_PER_LONG - 1UL); if (offset) { @@ -228,15 +229,16 @@ found_middle_swap: } EXPORT_SYMBOL(find_next_zero_bit_le); -unsigned long find_next_bit_le(const unsigned long *addr, unsigned +unsigned long find_next_bit_le(const void *addr, unsigned long size, unsigned long offset) { - const unsigned long *p = addr + BITOP_WORD(offset); + const unsigned long *p = addr; unsigned long result = offset & ~(BITS_PER_LONG - 1); unsigned long tmp; if (offset >= size) return size; + p += BITOP_WORD(offset); size -= result; offset &= (BITS_PER_LONG - 1UL); if (offset) { -- cgit v1.2.3-70-g09d2 From 0664996b7c2fdb1b7f90954469cc242274abd7db Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Wed, 23 Mar 2011 16:41:59 -0700 Subject: bitops: introduce CONFIG_GENERIC_FIND_BIT_LE This introduces CONFIG_GENERIC_FIND_BIT_LE to tell whether to use generic implementation of find_*_bit_le() in lib/find_next_bit.c or not. For now we select CONFIG_GENERIC_FIND_BIT_LE for all architectures which enable CONFIG_GENERIC_FIND_NEXT_BIT. But m68knommu wants to define own faster find_next_zero_bit_le() and continues using generic find_next_{,zero_}bit(). (CONFIG_GENERIC_FIND_NEXT_BIT and !CONFIG_GENERIC_FIND_BIT_LE) Signed-off-by: Akinobu Mita Cc: Greg Ungerer Cc: Arnd Bergmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/frv/Kconfig | 4 ++++ arch/h8300/Kconfig | 4 ++++ arch/m32r/Kconfig | 4 ++++ arch/m68knommu/Kconfig | 4 ++++ arch/microblaze/Kconfig | 3 +++ arch/mips/Kconfig | 4 ++++ arch/parisc/Kconfig | 4 ++++ arch/powerpc/Kconfig | 4 ++++ arch/sh/Kconfig | 3 +++ arch/sparc/Kconfig | 4 ++++ arch/xtensa/Kconfig | 3 +++ lib/Kconfig | 3 +++ lib/Makefile | 1 + lib/find_next_bit.c | 3 +++ 14 files changed, 48 insertions(+) (limited to 'lib') diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig index 747499a1b31..f6037b2da25 100644 --- a/arch/frv/Kconfig +++ b/arch/frv/Kconfig @@ -22,6 +22,10 @@ config GENERIC_FIND_NEXT_BIT bool default y +config GENERIC_FIND_BIT_LE + bool + default y + config GENERIC_HWEIGHT bool default y diff --git a/arch/h8300/Kconfig b/arch/h8300/Kconfig index 6df692d1475..9624db193e3 100644 --- a/arch/h8300/Kconfig +++ b/arch/h8300/Kconfig @@ -45,6 +45,10 @@ config GENERIC_FIND_NEXT_BIT bool default y +config GENERIC_FIND_BIT_LE + bool + default y + config GENERIC_HWEIGHT bool default y diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig index ef4c1e442be..62afe23c9a4 100644 --- a/arch/m32r/Kconfig +++ b/arch/m32r/Kconfig @@ -260,6 +260,10 @@ config GENERIC_FIND_NEXT_BIT bool default y +config GENERIC_FIND_BIT_LE + bool + default y + config GENERIC_HWEIGHT bool default y diff --git a/arch/m68knommu/Kconfig b/arch/m68knommu/Kconfig index b5424cf948e..5795f58728c 100644 --- a/arch/m68knommu/Kconfig +++ b/arch/m68knommu/Kconfig @@ -42,6 +42,10 @@ config GENERIC_FIND_NEXT_BIT bool default y +config GENERIC_FIND_BIT_LE + bool + default y + config GENERIC_GPIO bool default n diff --git a/arch/microblaze/Kconfig b/arch/microblaze/Kconfig index 922c4194c7b..5f0cf0e3265 100644 --- a/arch/microblaze/Kconfig +++ b/arch/microblaze/Kconfig @@ -37,6 +37,9 @@ config ARCH_HAS_ILOG2_U64 config GENERIC_FIND_NEXT_BIT def_bool y +config GENERIC_FIND_BIT_LE + def_bool y + config GENERIC_HWEIGHT def_bool y diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index d88983516e2..6e9811d40c9 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -777,6 +777,10 @@ config GENERIC_FIND_NEXT_BIT bool default y +config GENERIC_FIND_BIT_LE + bool + default y + config GENERIC_HWEIGHT bool default y diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig index fafdf30bea9..9b1f427cdc3 100644 --- a/arch/parisc/Kconfig +++ b/arch/parisc/Kconfig @@ -52,6 +52,10 @@ config GENERIC_FIND_NEXT_BIT bool default y +config GENERIC_FIND_BIT_LE + bool + default y + config GENERIC_BUG bool default y diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 71ba04721be..ce9ff55e102 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -95,6 +95,10 @@ config GENERIC_FIND_NEXT_BIT bool default y +config GENERIC_FIND_BIT_LE + bool + default y + config GENERIC_GPIO bool help diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig index 2d264fa8495..1fbf0c7583d 100644 --- a/arch/sh/Kconfig +++ b/arch/sh/Kconfig @@ -75,6 +75,9 @@ config GENERIC_CSUM config GENERIC_FIND_NEXT_BIT def_bool y +config GENERIC_FIND_BIT_LE + def_bool y + config GENERIC_HWEIGHT def_bool y diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index e48f471be54..f766e6bf370 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -192,6 +192,10 @@ config GENERIC_FIND_NEXT_BIT bool default y +config GENERIC_FIND_BIT_LE + bool + default y + config GENERIC_HWEIGHT bool default y if !ULTRA_HAS_POPULATION_COUNT diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig index d373d159e75..72839190f50 100644 --- a/arch/xtensa/Kconfig +++ b/arch/xtensa/Kconfig @@ -21,6 +21,9 @@ config RWSEM_XCHGADD_ALGORITHM config GENERIC_FIND_NEXT_BIT def_bool y +config GENERIC_FIND_BIT_LE + def_bool y + config GENERIC_HWEIGHT def_bool y diff --git a/lib/Kconfig b/lib/Kconfig index 3a55a43c43e..23fa7a359db 100644 --- a/lib/Kconfig +++ b/lib/Kconfig @@ -22,6 +22,9 @@ config GENERIC_FIND_FIRST_BIT config GENERIC_FIND_NEXT_BIT bool +config GENERIC_FIND_BIT_LE + bool + config GENERIC_FIND_LAST_BIT bool default y diff --git a/lib/Makefile b/lib/Makefile index 8c9de027ebb..d7872b5c4c1 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -40,6 +40,7 @@ lib-$(CONFIG_RWSEM_GENERIC_SPINLOCK) += rwsem-spinlock.o lib-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += rwsem.o lib-$(CONFIG_GENERIC_FIND_FIRST_BIT) += find_next_bit.o lib-$(CONFIG_GENERIC_FIND_NEXT_BIT) += find_next_bit.o +lib-$(CONFIG_GENERIC_FIND_BIT_LE) += find_next_bit.o obj-$(CONFIG_GENERIC_FIND_LAST_BIT) += find_last_bit.o CFLAGS_hweight.o = $(subst $(quote),,$(CONFIG_ARCH_HWEIGHT_CFLAGS)) diff --git a/lib/find_next_bit.c b/lib/find_next_bit.c index 7667c3d907d..b0a8767282b 100644 --- a/lib/find_next_bit.c +++ b/lib/find_next_bit.c @@ -160,6 +160,7 @@ EXPORT_SYMBOL(find_first_zero_bit); #endif /* CONFIG_GENERIC_FIND_FIRST_BIT */ #ifdef __BIG_ENDIAN +#ifdef CONFIG_GENERIC_FIND_BIT_LE /* include/linux/byteorder does not support "unsigned long" type */ static inline unsigned long ext2_swabp(const unsigned long * x) @@ -273,4 +274,6 @@ found_middle_swap: return result + __ffs(ext2_swab(tmp)); } EXPORT_SYMBOL(find_next_bit_le); + +#endif /* CONFIG_GENERIC_FIND_BIT_LE */ #endif /* __BIG_ENDIAN */ -- cgit v1.2.3-70-g09d2 From 0f77a8d378254f27df4a114a5da67223af1fe93f Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Thu, 24 Mar 2011 11:42:29 +0900 Subject: vsprintf: Introduce %pB format specifier The %pB format specifier is for stack backtrace. Its handler sprint_backtrace() does symbol lookup using (address-1) to ensure the address will not point outside of the function. If there is a tail-call to the function marked "noreturn", gcc optimized out the code after the call then causes saved return address points outside of the function (i.e. the start of the next function), so pollutes call trace somewhat. This patch adds the %pB printk mechanism that allows architecture call-trace printout functions to improve backtrace printouts. Signed-off-by: Namhyung Kim Acked-by: Steven Rostedt Acked-by: Frederic Weisbecker Cc: Linus Torvalds Cc: Andrew Morton Cc: linux-arch@vger.kernel.org LKML-Reference: <1300934550-21394-1-git-send-email-namhyung@gmail.com> Signed-off-by: Ingo Molnar --- include/linux/kallsyms.h | 7 +++++++ kernel/kallsyms.c | 44 +++++++++++++++++++++++++++++++++++++++++--- lib/vsprintf.c | 7 ++++++- 3 files changed, 54 insertions(+), 4 deletions(-) (limited to 'lib') diff --git a/include/linux/kallsyms.h b/include/linux/kallsyms.h index d8e9b3d1c23..0df513b7a9f 100644 --- a/include/linux/kallsyms.h +++ b/include/linux/kallsyms.h @@ -36,6 +36,7 @@ const char *kallsyms_lookup(unsigned long addr, /* Look up a kernel symbol and return it in a text buffer. */ extern int sprint_symbol(char *buffer, unsigned long address); +extern int sprint_backtrace(char *buffer, unsigned long address); /* Look up a kernel symbol and print it to the kernel messages. */ extern void __print_symbol(const char *fmt, unsigned long address); @@ -79,6 +80,12 @@ static inline int sprint_symbol(char *buffer, unsigned long addr) return 0; } +static inline int sprint_backtrace(char *buffer, unsigned long addr) +{ + *buffer = '\0'; + return 0; +} + static inline int lookup_symbol_name(unsigned long addr, char *symname) { return -ERANGE; diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c index 6f6d091b575..59e879929b1 100644 --- a/kernel/kallsyms.c +++ b/kernel/kallsyms.c @@ -342,13 +342,15 @@ int lookup_symbol_attrs(unsigned long addr, unsigned long *size, } /* Look up a kernel symbol and return it in a text buffer. */ -int sprint_symbol(char *buffer, unsigned long address) +static int __sprint_symbol(char *buffer, unsigned long address, + int symbol_offset) { char *modname; const char *name; unsigned long offset, size; int len; + address += symbol_offset; name = kallsyms_lookup(address, &size, &offset, &modname, buffer); if (!name) return sprintf(buffer, "0x%lx", address); @@ -357,17 +359,53 @@ int sprint_symbol(char *buffer, unsigned long address) strcpy(buffer, name); len = strlen(buffer); buffer += len; + offset -= symbol_offset; if (modname) - len += sprintf(buffer, "+%#lx/%#lx [%s]", - offset, size, modname); + len += sprintf(buffer, "+%#lx/%#lx [%s]", offset, size, modname); else len += sprintf(buffer, "+%#lx/%#lx", offset, size); return len; } + +/** + * sprint_symbol - Look up a kernel symbol and return it in a text buffer + * @buffer: buffer to be stored + * @address: address to lookup + * + * This function looks up a kernel symbol with @address and stores its name, + * offset, size and module name to @buffer if possible. If no symbol was found, + * just saves its @address as is. + * + * This function returns the number of bytes stored in @buffer. + */ +int sprint_symbol(char *buffer, unsigned long address) +{ + return __sprint_symbol(buffer, address, 0); +} + EXPORT_SYMBOL_GPL(sprint_symbol); +/** + * sprint_backtrace - Look up a backtrace symbol and return it in a text buffer + * @buffer: buffer to be stored + * @address: address to lookup + * + * This function is for stack backtrace and does the same thing as + * sprint_symbol() but with modified/decreased @address. If there is a + * tail-call to the function marked "noreturn", gcc optimized out code after + * the call so that the stack-saved return address could point outside of the + * caller. This function ensures that kallsyms will find the original caller + * by decreasing @address. + * + * This function returns the number of bytes stored in @buffer. + */ +int sprint_backtrace(char *buffer, unsigned long address) +{ + return __sprint_symbol(buffer, address, -1); +} + /* Look up a kernel symbol and print it to the kernel messages. */ void __print_symbol(const char *fmt, unsigned long address) { diff --git a/lib/vsprintf.c b/lib/vsprintf.c index d3023df8477..d9e01fc3168 100644 --- a/lib/vsprintf.c +++ b/lib/vsprintf.c @@ -574,7 +574,9 @@ char *symbol_string(char *buf, char *end, void *ptr, unsigned long value = (unsigned long) ptr; #ifdef CONFIG_KALLSYMS char sym[KSYM_SYMBOL_LEN]; - if (ext != 'f' && ext != 's') + if (ext == 'B') + sprint_backtrace(sym, value); + else if (ext != 'f' && ext != 's') sprint_symbol(sym, value); else kallsyms_lookup(value, NULL, NULL, NULL, sym); @@ -949,6 +951,7 @@ int kptr_restrict = 1; * - 'f' For simple symbolic function names without offset * - 'S' For symbolic direct pointers with offset * - 's' For symbolic direct pointers without offset + * - 'B' For backtraced symbolic direct pointers with offset * - 'R' For decoded struct resource, e.g., [mem 0x0-0x1f 64bit pref] * - 'r' For raw struct resource, e.g., [mem 0x0-0x1f flags 0x201] * - 'M' For a 6-byte MAC address, it prints the address in the @@ -1008,6 +1011,7 @@ char *pointer(const char *fmt, char *buf, char *end, void *ptr, /* Fallthrough */ case 'S': case 's': + case 'B': return symbol_string(buf, end, ptr, spec, *fmt); case 'R': case 'r': @@ -1279,6 +1283,7 @@ qualifier: * %ps output the name of a text symbol without offset * %pF output the name of a function pointer with its offset * %pf output the name of a function pointer without its offset + * %pB output the name of a backtrace symbol with its offset * %pR output the address range in a struct resource with decoded flags * %pr output the address range in a struct resource with raw flags * %pM output a 6-byte MAC address with colons -- cgit v1.2.3-70-g09d2 From b2b755b5f10eb32fbdc73a9907c07006b17f714b Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Thu, 24 Mar 2011 15:18:15 -0700 Subject: lib, arch: add filter argument to show_mem and fix private implementations Commit ddd588b5dd55 ("oom: suppress nodes that are not allowed from meminfo on oom kill") moved lib/show_mem.o out of lib/lib.a, which resulted in build warnings on all architectures that implement their own versions of show_mem(): lib/lib.a(show_mem.o): In function `show_mem': show_mem.c:(.text+0x1f4): multiple definition of `show_mem' arch/sparc/mm/built-in.o:(.text+0xd70): first defined here The fix is to remove __show_mem() and add its argument to show_mem() in all implementations to prevent this breakage. Architectures that implement their own show_mem() actually don't do anything with the argument yet, but they could be made to filter nodes that aren't allowed in the current context in the future just like the generic implementation. Reported-by: Stephen Rothwell Reported-by: James Bottomley Suggested-by: Andrew Morton Signed-off-by: David Rientjes Signed-off-by: Linus Torvalds --- arch/arm/mm/init.c | 2 +- arch/ia64/mm/contig.c | 2 +- arch/ia64/mm/discontig.c | 2 +- arch/parisc/mm/init.c | 2 +- arch/powerpc/xmon/xmon.c | 2 +- arch/sparc/mm/init_32.c | 2 +- arch/tile/mm/pgtable.c | 2 +- arch/unicore32/mm/init.c | 2 +- drivers/tty/sysrq.c | 2 +- drivers/tty/vt/keyboard.c | 2 +- include/linux/mm.h | 5 ++--- lib/show_mem.c | 7 +------ mm/oom_kill.c | 2 +- mm/page_alloc.c | 2 +- 14 files changed, 15 insertions(+), 21 deletions(-) (limited to 'lib') diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c index b3b0f0f5053..e5f6fc42834 100644 --- a/arch/arm/mm/init.c +++ b/arch/arm/mm/init.c @@ -78,7 +78,7 @@ __tagtable(ATAG_INITRD2, parse_tag_initrd2); */ struct meminfo meminfo; -void show_mem(void) +void show_mem(unsigned int filter) { int free = 0, total = 0, reserved = 0; int shared = 0, cached = 0, slab = 0, i; diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c index 54bf5405981..9a018cde5d8 100644 --- a/arch/ia64/mm/contig.c +++ b/arch/ia64/mm/contig.c @@ -36,7 +36,7 @@ static unsigned long max_gap; * Shows a simple page count of reserved and used pages in the system. * For discontig machines, it does this on a per-pgdat basis. */ -void show_mem(void) +void show_mem(unsigned int filter) { int i, total_reserved = 0; int total_shared = 0, total_cached = 0; diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index 61620323bb6..82ab1bc6afb 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -614,7 +614,7 @@ void __cpuinit *per_cpu_init(void) * Shows a simple page count of reserved and used pages in the system. * For discontig machines, it does this on a per-pgdat basis. */ -void show_mem(void) +void show_mem(unsigned int filter) { int i, total_reserved = 0; int total_shared = 0, total_cached = 0; diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c index f4f4d700833..b7ed8d7a9b3 100644 --- a/arch/parisc/mm/init.c +++ b/arch/parisc/mm/init.c @@ -544,7 +544,7 @@ void __init mem_init(void) unsigned long *empty_zero_page __read_mostly; EXPORT_SYMBOL(empty_zero_page); -void show_mem(void) +void show_mem(unsigned int filter) { int i,free = 0,total = 0,reserved = 0; int shared = 0, cached = 0; diff --git a/arch/powerpc/xmon/xmon.c b/arch/powerpc/xmon/xmon.c index d17d04cfb2c..33794c1d92c 100644 --- a/arch/powerpc/xmon/xmon.c +++ b/arch/powerpc/xmon/xmon.c @@ -821,7 +821,7 @@ cmds(struct pt_regs *excp) memzcan(); break; case 'i': - show_mem(); + show_mem(0); break; default: termch = cmd; diff --git a/arch/sparc/mm/init_32.c b/arch/sparc/mm/init_32.c index 6d0e02c4fe0..4c31e2b6e71 100644 --- a/arch/sparc/mm/init_32.c +++ b/arch/sparc/mm/init_32.c @@ -75,7 +75,7 @@ void __init kmap_init(void) kmap_prot = __pgprot(SRMMU_ET_PTE | SRMMU_PRIV | SRMMU_CACHE); } -void show_mem(void) +void show_mem(unsigned int filter) { printk("Mem-info:\n"); show_free_areas(); diff --git a/arch/tile/mm/pgtable.c b/arch/tile/mm/pgtable.c index 1a2b36f8866..de7d8e21e01 100644 --- a/arch/tile/mm/pgtable.c +++ b/arch/tile/mm/pgtable.c @@ -41,7 +41,7 @@ * The normal show_free_areas() is too verbose on Tile, with dozens * of processors and often four NUMA zones each with high and lowmem. */ -void show_mem(void) +void show_mem(unsigned int filter) { struct zone *zone; diff --git a/arch/unicore32/mm/init.c b/arch/unicore32/mm/init.c index 3dbe3709b69..1fc02633f70 100644 --- a/arch/unicore32/mm/init.c +++ b/arch/unicore32/mm/init.c @@ -55,7 +55,7 @@ early_param("initrd", early_initrd); */ struct meminfo meminfo; -void show_mem(void) +void show_mem(unsigned int filter) { int free = 0, total = 0, reserved = 0; int shared = 0, cached = 0, slab = 0, i; diff --git a/drivers/tty/sysrq.c b/drivers/tty/sysrq.c index 81f13958e75..43db715f150 100644 --- a/drivers/tty/sysrq.c +++ b/drivers/tty/sysrq.c @@ -306,7 +306,7 @@ static struct sysrq_key_op sysrq_ftrace_dump_op = { static void sysrq_handle_showmem(int key) { - show_mem(); + show_mem(0); } static struct sysrq_key_op sysrq_showmem_op = { .handler = sysrq_handle_showmem, diff --git a/drivers/tty/vt/keyboard.c b/drivers/tty/vt/keyboard.c index 6dd3c68c13a..d6b342b5b42 100644 --- a/drivers/tty/vt/keyboard.c +++ b/drivers/tty/vt/keyboard.c @@ -600,7 +600,7 @@ static void fn_scroll_back(struct vc_data *vc) static void fn_show_mem(struct vc_data *vc) { - show_mem(); + show_mem(0); } static void fn_show_state(struct vc_data *vc) diff --git a/include/linux/mm.h b/include/linux/mm.h index f9535b2c955..7606d7db96c 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -861,7 +861,7 @@ extern void pagefault_out_of_memory(void); #define offset_in_page(p) ((unsigned long)(p) & ~PAGE_MASK) /* - * Flags passed to __show_mem() and __show_free_areas() to suppress output in + * Flags passed to show_mem() and __show_free_areas() to suppress output in * various contexts. */ #define SHOW_MEM_FILTER_NODES (0x0001u) /* filter disallowed nodes */ @@ -1360,8 +1360,7 @@ extern void setup_per_zone_wmarks(void); extern void calculate_zone_inactive_ratio(struct zone *zone); extern void mem_init(void); extern void __init mmap_init(void); -extern void show_mem(void); -extern void __show_mem(unsigned int flags); +extern void show_mem(unsigned int flags); extern void si_meminfo(struct sysinfo * val); extern void si_meminfo_node(struct sysinfo *val, int nid); extern int after_bootmem; diff --git a/lib/show_mem.c b/lib/show_mem.c index d8d602b58c3..90cbe4bb596 100644 --- a/lib/show_mem.c +++ b/lib/show_mem.c @@ -9,7 +9,7 @@ #include #include -void __show_mem(unsigned int filter) +void show_mem(unsigned int filter) { pg_data_t *pgdat; unsigned long total = 0, reserved = 0, shared = 0, @@ -61,8 +61,3 @@ void __show_mem(unsigned int filter) quicklist_total_size()); #endif } - -void show_mem(void) -{ - __show_mem(0); -} diff --git a/mm/oom_kill.c b/mm/oom_kill.c index 62a5cec08a1..6a819d1b2c7 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -406,7 +406,7 @@ static void dump_header(struct task_struct *p, gfp_t gfp_mask, int order, task_unlock(current); dump_stack(); mem_cgroup_print_oom_info(mem, p); - __show_mem(SHOW_MEM_FILTER_NODES); + show_mem(SHOW_MEM_FILTER_NODES); if (sysctl_oom_dump_tasks) dump_tasks(mem, nodemask); } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 8e5726ab0d8..d6e7ba7373b 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2195,7 +2195,7 @@ nopage: current->comm, order, gfp_mask); dump_stack(); if (!should_suppress_show_mem()) - __show_mem(filter); + show_mem(filter); } return page; got_pg: -- cgit v1.2.3-70-g09d2